2944 files changed, 86240 insertions, 45001 deletions

diff --git a/.clang-format b/.clang-format
index fe1aa1a30d40..48405c54ef27 100644
--- a/.clang-format
+++ b/.clang-format
@@ -92,6 +92,7 @@ ForEachMacros:
   - '__rq_for_each_bio'
   - '__shost_for_each_device'
   - '__sym_for_each'
+  - '_for_each_counter'
   - 'apei_estatus_for_each_section'
   - 'ata_for_each_dev'
   - 'ata_for_each_link'
@@ -141,11 +142,14 @@ ForEachMacros:
   - 'damon_for_each_target_safe'
   - 'damos_for_each_filter'
   - 'damos_for_each_filter_safe'
+  - 'damos_for_each_ops_filter'
+  - 'damos_for_each_ops_filter_safe'
   - 'damos_for_each_quota_goal'
   - 'damos_for_each_quota_goal_safe'
   - 'data__for_each_file'
   - 'data__for_each_file_new'
   - 'data__for_each_file_start'
+  - 'def_for_each_cpu'
   - 'device_for_each_child_node'
   - 'device_for_each_child_node_scoped'
   - 'dma_fence_array_for_each'
@@ -176,6 +180,7 @@ ForEachMacros:
   - 'drm_for_each_privobj'
   - 'drm_gem_for_each_gpuvm_bo'
   - 'drm_gem_for_each_gpuvm_bo_safe'
+  - 'drm_gpusvm_for_each_range'
   - 'drm_gpuva_for_each_op'
   - 'drm_gpuva_for_each_op_from_reverse'
   - 'drm_gpuva_for_each_op_reverse'
@@ -216,8 +221,10 @@ ForEachMacros:
   - 'for_each_active_dev_scope'
   - 'for_each_active_drhd_unit'
   - 'for_each_active_iommu'
+  - 'for_each_active_irq'
   - 'for_each_active_route'
   - 'for_each_aggr_pgid'
+  - 'for_each_alloc_capable_rdt_resource'
   - 'for_each_and_bit'
   - 'for_each_andnot_bit'
   - 'for_each_available_child_of_node'
@@ -228,6 +235,7 @@ ForEachMacros:
   - 'for_each_btf_ext_rec'
   - 'for_each_btf_ext_sec'
   - 'for_each_bvec'
+  - 'for_each_capable_rdt_resource'
   - 'for_each_card_auxs'
   - 'for_each_card_auxs_safe'
   - 'for_each_card_components'
@@ -241,6 +249,7 @@ ForEachMacros:
   - 'for_each_cgroup_storage_type'
   - 'for_each_child_of_node'
   - 'for_each_child_of_node_scoped'
+  - 'for_each_child_of_node_with_prefix'
   - 'for_each_clear_bit'
   - 'for_each_clear_bit_from'
   - 'for_each_clear_bitrange'
@@ -296,6 +305,7 @@ ForEachMacros:
   - 'for_each_group_member_head'
   - 'for_each_hstate'
   - 'for_each_hwgpio'
+  - 'for_each_hwgpio_in_range'
   - 'for_each_if'
   - 'for_each_inject_fn'
   - 'for_each_insn'
@@ -304,6 +314,7 @@ ForEachMacros:
   - 'for_each_intid'
   - 'for_each_iommu'
   - 'for_each_ip_tunnel_rcu'
+  - 'for_each_irq_desc'
   - 'for_each_irq_nr'
   - 'for_each_lang'
   - 'for_each_link_ch_maps'
@@ -324,6 +335,8 @@ ForEachMacros:
   - 'for_each_missing_reg'
   - 'for_each_mle_subelement'
   - 'for_each_mod_mem_type'
+  - 'for_each_mon_capable_rdt_resource'
+  - 'for_each_mp_bvec'
   - 'for_each_net'
   - 'for_each_net_continue_reverse'
   - 'for_each_net_rcu'
@@ -351,6 +364,7 @@ ForEachMacros:
   - 'for_each_node_by_name'
   - 'for_each_node_by_type'
   - 'for_each_node_mask'
+  - 'for_each_node_numadist'
   - 'for_each_node_state'
   - 'for_each_node_with_cpus'
   - 'for_each_node_with_property'
@@ -359,6 +373,8 @@ ForEachMacros:
   - 'for_each_of_allnodes'
   - 'for_each_of_allnodes_from'
   - 'for_each_of_cpu_node'
+  - 'for_each_of_graph_port'
+  - 'for_each_of_graph_port_endpoint'
   - 'for_each_of_pci_range'
   - 'for_each_old_connector_in_state'
   - 'for_each_old_crtc_in_state'
@@ -372,9 +388,11 @@ ForEachMacros:
   - 'for_each_oldnew_plane_in_state_reverse'
   - 'for_each_oldnew_private_obj_in_state'
   - 'for_each_online_cpu'
+  - 'for_each_online_cpu_wrap'
   - 'for_each_online_node'
   - 'for_each_online_pgdat'
   - 'for_each_or_bit'
+  - 'for_each_page_ext'
   - 'for_each_path'
   - 'for_each_pci_bridge'
   - 'for_each_pci_dev'
@@ -382,8 +400,10 @@ ForEachMacros:
   - 'for_each_physmem_range'
   - 'for_each_populated_zone'
   - 'for_each_possible_cpu'
+  - 'for_each_possible_cpu_wrap'
   - 'for_each_present_blessed_reg'
   - 'for_each_present_cpu'
+  - 'for_each_present_section_nr'
   - 'for_each_prime_number'
   - 'for_each_prime_number_from'
   - 'for_each_probe_cache_entry'
@@ -396,6 +416,7 @@ ForEachMacros:
   - 'for_each_prop_dlc_cpus'
   - 'for_each_prop_dlc_platforms'
   - 'for_each_property_of_node'
+  - 'for_each_rdt_resource'
   - 'for_each_reg'
   - 'for_each_reg_filtered'
   - 'for_each_reloc'
@@ -434,10 +455,10 @@ ForEachMacros:
   - 'for_each_subelement_id'
   - 'for_each_sublist'
   - 'for_each_subsystem'
+  - 'for_each_suite'
   - 'for_each_supported_activate_fn'
   - 'for_each_supported_inject_fn'
   - 'for_each_sym'
-  - 'for_each_test'
   - 'for_each_thread'
   - 'for_each_token'
   - 'for_each_unicast_dest_pgid'
@@ -499,8 +520,10 @@ ForEachMacros:
   - 'idr_for_each_entry_continue'
   - 'idr_for_each_entry_continue_ul'
   - 'idr_for_each_entry_ul'
+  - 'iio_for_each_active_channel'
   - 'in_dev_for_each_ifa_rcu'
   - 'in_dev_for_each_ifa_rtnl'
+  - 'in_dev_for_each_ifa_rtnl_net'
   - 'inet_bind_bucket_for_each'
   - 'interval_tree_for_each_span'
   - 'intlist__for_each_entry'
@@ -542,7 +565,6 @@ ForEachMacros:
   - 'list_for_each_prev'
   - 'list_for_each_prev_safe'
   - 'list_for_each_rcu'
-  - 'list_for_each_reverse'
   - 'list_for_each_safe'
   - 'llist_for_each'
   - 'llist_for_each_entry'
@@ -552,6 +574,7 @@ ForEachMacros:
   - 'map__for_each_symbol'
   - 'map__for_each_symbol_by_name'
   - 'mas_for_each'
+  - 'mas_for_each_rev'
   - 'mci_for_each_dimm'
   - 'media_device_for_each_entity'
   - 'media_device_for_each_intf'
@@ -561,10 +584,15 @@ ForEachMacros:
   - 'media_pipeline_for_each_entity'
   - 'media_pipeline_for_each_pad'
   - 'mlx5_lag_for_each_peer_mdev'
+  - 'mptcp_for_each_subflow'
   - 'msi_domain_for_each_desc'
   - 'msi_for_each_desc'
   - 'mt_for_each'
+  - 'nanddev_io_for_each_block'
   - 'nanddev_io_for_each_page'
+  - 'neigh_for_each_in_bucket'
+  - 'neigh_for_each_in_bucket_rcu'
+  - 'neigh_for_each_in_bucket_safe'
   - 'netdev_for_each_lower_dev'
   - 'netdev_for_each_lower_private'
   - 'netdev_for_each_lower_private_rcu'
@@ -604,11 +632,11 @@ ForEachMacros:
   - 'perf_evlist__for_each_entry_safe'
   - 'perf_evlist__for_each_evsel'
   - 'perf_evlist__for_each_mmap'
+  - 'perf_evsel_for_each_per_thread_period_safe'
   - 'perf_hpp_list__for_each_format'
   - 'perf_hpp_list__for_each_format_safe'
   - 'perf_hpp_list__for_each_sort_list'
   - 'perf_hpp_list__for_each_sort_list_safe'
-  - 'perf_tool_event__for_each_event'
   - 'plist_for_each'
   - 'plist_for_each_continue'
   - 'plist_for_each_entry'
@@ -627,7 +655,6 @@ ForEachMacros:
   - 'rdma_for_each_block'
   - 'rdma_for_each_port'
   - 'rdma_umem_for_each_dma_block'
-  - 'resort_rb__for_each_entry'
   - 'resource_list_for_each_entry'
   - 'resource_list_for_each_entry_safe'
   - 'rhl_for_each_entry_rcu'
@@ -658,6 +685,7 @@ ForEachMacros:
   - 'shost_for_each_device'
   - 'sk_for_each'
   - 'sk_for_each_bound'
+  - 'sk_for_each_bound_safe'
   - 'sk_for_each_entry_offset_rcu'
   - 'sk_for_each_from'
   - 'sk_for_each_rcu'
@@ -680,7 +708,11 @@ ForEachMacros:
   - 'tb_property_for_each'
   - 'tcf_act_for_each_action'
   - 'tcf_exts_for_each_action'
+  - 'test_suite__for_each_test_case'
+  - 'tool_pmu__for_each_event'
+  - 'ttm_bo_lru_for_each_reserved_guarded'
   - 'ttm_resource_manager_for_each_res'
+  - 'udp_lrpa_for_each_entry_rcu'
   - 'udp_portaddr_for_each_entry'
   - 'udp_portaddr_for_each_entry_rcu'
   - 'usb_hub_for_each_child'
@@ -690,7 +722,15 @@ ForEachMacros:
   - 'v4l2_m2m_for_each_src_buf'
   - 'v4l2_m2m_for_each_src_buf_safe'
   - 'virtio_device_for_each_vq'
+  - 'vkms_config_for_each_connector'
+  - 'vkms_config_for_each_crtc'
+  - 'vkms_config_for_each_encoder'
+  - 'vkms_config_for_each_plane'
+  - 'vkms_config_connector_for_each_possible_encoder'
+  - 'vkms_config_encoder_for_each_possible_crtc'
+  - 'vkms_config_plane_for_each_possible_crtc'
   - 'while_for_each_ftrace_op'
+  - 'workloads__for_each'
   - 'xa_for_each'
   - 'xa_for_each_marked'
   - 'xa_for_each_range'
diff --git a/.mailmap b/.mailmap
index 4f7cd8e23177..9afde79e1936 100644
--- a/.mailmap
+++ b/.mailmap
@@ -322,6 +322,7 @@ Jayachandran C <c.jayachandran@gmail.com> <jchandra@broadcom.com>
 Jayachandran C <c.jayachandran@gmail.com> <jchandra@digeo.com>
 Jayachandran C <c.jayachandran@gmail.com> <jnair@caviumnetworks.com>
 <jean-philippe@linaro.org> <jean-philippe.brucker@arm.com>
+Jean-Michel Hautbois <jeanmichel.hautbois@yoseli.org> <jeanmichel.hautbois@ideasonboard.com>
 Jean Tourrilhes <jt@hpl.hp.com>
 Jeevan Shriram <quic_jshriram@quicinc.com> <jshriram@codeaurora.org>
 Jeff Garzik <jgarzik@pretzel.yyz.us>
@@ -438,6 +439,8 @@ Linus Lüssing <linus.luessing@c0d3.blue> <ll@simonwunderlich.de>
 Li Yang <leoyang.li@nxp.com> <leoli@freescale.com>
 Li Yang <leoyang.li@nxp.com> <leo@zh-kernel.org>
 Lior David <quic_liord@quicinc.com> <liord@codeaurora.org>
+Loic Poulain <loic.poulain@oss.qualcomm.com> <loic.poulain@linaro.org>
+Loic Poulain <loic.poulain@oss.qualcomm.com> <loic.poulain@intel.com>
 Lorenzo Pieralisi <lpieralisi@kernel.org> <lorenzo.pieralisi@arm.com>
 Lorenzo Stoakes <lorenzo.stoakes@oracle.com> <lstoakes@gmail.com>
 Luca Ceresoli <luca.ceresoli@bootlin.com> <luca@lucaceresoli.net>
@@ -685,6 +688,8 @@ Simon Wunderlich <sw@simonwunderlich.de> <simon.wunderlich@saxnet.de>
 Simon Wunderlich <sw@simonwunderlich.de> <simon@open-mesh.com>
 Simon Wunderlich <sw@simonwunderlich.de> <siwu@hrz.tu-chemnitz.de>
 Sricharan Ramabadhran <quic_srichara@quicinc.com> <sricharan@codeaurora.org>
+Srinivas Kandagatla <srini@kernel.org> <srinivas.kandagatla@st.com>
+Srinivas Kandagatla <srini@kernel.org> <srinivas.kandagatla@linaro.org>
 Srinivas Ramana <quic_sramana@quicinc.com> <sramana@codeaurora.org>
 Sriram R <quic_srirrama@quicinc.com> <srirrama@codeaurora.org>
 Sriram Yagnaraman <sriram.yagnaraman@ericsson.com> <sriram.yagnaraman@est.tech>
diff --git a/CREDITS b/CREDITS
index 1b77fba6c27e..f74d230992d6 100644
--- a/CREDITS
+++ b/CREDITS
@@ -2071,6 +2071,10 @@ S: 660 Harvard Ave. #7
 S: Santa Clara, CA 95051
 S: USA
 
+N: Joonsoo Kim
+E: iamjoonsoo.kim@lge.com
+D: Slab allocators
+
 N: Kukjin Kim
 E: kgene@kernel.org
 D: Samsung S3C, S5P and Exynos ARM architectures
diff --git a/Documentation/ABI/stable/sysfs-block b/Documentation/ABI/stable/sysfs-block
index 3879963f0f01..11545c9e2e93 100644
--- a/Documentation/ABI/stable/sysfs-block
+++ b/Documentation/ABI/stable/sysfs-block
@@ -77,7 +77,7 @@ Description:
 
 What:		/sys/block/<disk>/diskseq
 Date:		February 2021
-Contact:	Matteo Croce <mcroce@microsoft.com>
+Contact:	Matteo Croce <teknoraver@meta.com>
 Description:
 		The /sys/block/<disk>/diskseq files reports the disk
 		sequence number, which is a monotonically increasing
diff --git a/Documentation/ABI/stable/sysfs-class-backlight b/Documentation/ABI/stable/sysfs-class-backlight
index 6102d6bebdf9..40b8c46b95b2 100644
--- a/Documentation/ABI/stable/sysfs-class-backlight
+++ b/Documentation/ABI/stable/sysfs-class-backlight
@@ -26,7 +26,12 @@ Date:		March 2006
 KernelVersion:	2.6.17
 Contact:	Richard Purdie <rpurdie@rpsys.net>
 Description:
-		Show the actual brightness by querying the hardware.
+		Show the actual brightness by querying the hardware. Due
+		to implementation differences in hardware this may not
+		match the value in 'brightness'. For example some hardware
+		may treat blanking differently or have custom power saving
+		features. Userspace should generally use the values in
+		'brightness' to make decisions.
 Users:		HAL
 
 What:		/sys/class/backlight/<backlight>/max_brightness
diff --git a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon
index 9bce281314df..4ca917ac6382 100644
--- a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon
+++ b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon
@@ -111,7 +111,7 @@ Description:	RO. Package current voltage in millivolt.
 
 What:		/sys/bus/pci/drivers/xe/.../hwmon/hwmon<i>/temp2_input
 Date:		March 2025
-KernelVersion:	6.14
+KernelVersion:	6.15
 Contact:	intel-xe@lists.freedesktop.org
 Description:	RO. Package temperature in millidegree Celsius.
 
@@ -119,8 +119,32 @@ Description:	RO. Package temperature in millidegree Celsius.
 
 What:		/sys/bus/pci/drivers/xe/.../hwmon/hwmon<i>/temp3_input
 Date:		March 2025
-KernelVersion:	6.14
+KernelVersion:	6.15
 Contact:	intel-xe@lists.freedesktop.org
 Description:	RO. VRAM temperature in millidegree Celsius.
 
 		Only supported for particular Intel Xe graphics platforms.
+
+What:		/sys/bus/pci/drivers/xe/.../hwmon/hwmon<i>/fan1_input
+Date:		March 2025
+KernelVersion:	6.16
+Contact:	intel-xe@lists.freedesktop.org
+Description:	RO. Fan 1 speed in RPM.
+
+		Only supported for particular Intel Xe graphics platforms.
+
+What:		/sys/bus/pci/drivers/xe/.../hwmon/hwmon<i>/fan2_input
+Date:		March 2025
+KernelVersion:	6.16
+Contact:	intel-xe@lists.freedesktop.org
+Description:	RO. Fan 2 speed in RPM.
+
+		Only supported for particular Intel Xe graphics platforms.
+
+What:		/sys/bus/pci/drivers/xe/.../hwmon/hwmon<i>/fan3_input
+Date:		March 2025
+KernelVersion:	6.16
+Contact:	intel-xe@lists.freedesktop.org
+Description:	RO. Fan 3 speed in RPM.
+
+		Only supported for particular Intel Xe graphics platforms.
diff --git a/Documentation/ABI/testing/sysfs-driver-ufs b/Documentation/ABI/testing/sysfs-driver-ufs
index ae0191295d29..e36d2de16cbd 100644
--- a/Documentation/ABI/testing/sysfs-driver-ufs
+++ b/Documentation/ABI/testing/sysfs-driver-ufs
@@ -1604,3 +1604,35 @@ Description:
 		prevent the UFS from frequently performing clock gating/ungating.
 
 		The attribute is read/write.
+
+What:		/sys/bus/platform/drivers/ufshcd/*/device_lvl_exception_count
+What:		/sys/bus/platform/devices/*.ufs/device_lvl_exception_count
+Date:		March 2025
+Contact:	Bao D. Nguyen <quic_nguyenb@quicinc.com>
+Description:
+		This attribute is applicable to ufs devices compliant to the
+		JEDEC specifications version 4.1 or later. The
+		device_lvl_exception_count is a counter indicating the number of
+		times the device level exceptions have occurred since the last
+		time this variable is reset.  Writing a 0 value to this
+		attribute will reset the device_lvl_exception_count.  If the
+		device_lvl_exception_count reads a positive value, the user
+		application should read the device_lvl_exception_id attribute to
+		know more information about the exception.
+
+		The attribute is read/write.
+
+What:		/sys/bus/platform/drivers/ufshcd/*/device_lvl_exception_id
+What:		/sys/bus/platform/devices/*.ufs/device_lvl_exception_id
+Date:		March 2025
+Contact:	Bao D. Nguyen <quic_nguyenb@quicinc.com>
+Description:
+		Reading the device_lvl_exception_id returns the
+		qDeviceLevelExceptionID attribute of the ufs device JEDEC
+		specification version 4.1. The definition of the
+		qDeviceLevelExceptionID is the ufs device vendor specific
+		implementation.  Refer to the device manufacturer datasheet for
+		more information on the meaning of the qDeviceLevelExceptionID
+		attribute value.
+
+		The attribute is read only.
diff --git a/Documentation/ABI/testing/sysfs-kernel-reboot b/Documentation/ABI/testing/sysfs-kernel-reboot
index e117aba46be0..52571fd5ddba 100644
--- a/Documentation/ABI/testing/sysfs-kernel-reboot
+++ b/Documentation/ABI/testing/sysfs-kernel-reboot
@@ -1,7 +1,7 @@
 What:		/sys/kernel/reboot
 Date:		November 2020
 KernelVersion:	5.11
-Contact:	Matteo Croce <mcroce@microsoft.com>
+Contact:	Matteo Croce <teknoraver@meta.com>
 Description:	Interface to set the kernel reboot behavior, similarly to
 		what can be done via the reboot= cmdline option.
 		(see Documentation/admin-guide/kernel-parameters.txt)
@@ -9,25 +9,25 @@ Description:	Interface to set the kernel reboot behavior, similarly to
 What:		/sys/kernel/reboot/mode
 Date:		November 2020
 KernelVersion:	5.11
-Contact:	Matteo Croce <mcroce@microsoft.com>
+Contact:	Matteo Croce <teknoraver@meta.com>
 Description:	Reboot mode. Valid values are: cold warm hard soft gpio
 
 What:		/sys/kernel/reboot/type
 Date:		November 2020
 KernelVersion:	5.11
-Contact:	Matteo Croce <mcroce@microsoft.com>
+Contact:	Matteo Croce <teknoraver@meta.com>
 Description:	Reboot type. Valid values are: bios acpi kbd triple efi pci
 
 What:		/sys/kernel/reboot/cpu
 Date:		November 2020
 KernelVersion:	5.11
-Contact:	Matteo Croce <mcroce@microsoft.com>
+Contact:	Matteo Croce <teknoraver@meta.com>
 Description:	CPU number to use to reboot.
 
 What:		/sys/kernel/reboot/force
 Date:		November 2020
 KernelVersion:	5.11
-Contact:	Matteo Croce <mcroce@microsoft.com>
+Contact:	Matteo Croce <teknoraver@meta.com>
 Description:	Don't wait for any other CPUs on reboot and
 		avoid anything that could hang.
 
diff --git a/Documentation/admin-guide/hw-vuln/index.rst b/Documentation/admin-guide/hw-vuln/index.rst
index ff0b440ef2dc..451874b8135d 100644
--- a/Documentation/admin-guide/hw-vuln/index.rst
+++ b/Documentation/admin-guide/hw-vuln/index.rst
@@ -22,3 +22,4 @@ are configurable at compile, boot or run time.
    srso
    gather_data_sampling
    reg-file-data-sampling
+   rsb
diff --git a/Documentation/admin-guide/hw-vuln/rsb.rst b/Documentation/admin-guide/hw-vuln/rsb.rst
new file mode 100644
index 000000000000..21dbf9cf25f8
--- /dev/null
+++ b/Documentation/admin-guide/hw-vuln/rsb.rst
@@ -0,0 +1,268 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=======================
+RSB-related mitigations
+=======================
+
+.. warning::
+   Please keep this document up-to-date, otherwise you will be
+   volunteered to update it and convert it to a very long comment in
+   bugs.c!
+
+Since 2018 there have been many Spectre CVEs related to the Return Stack
+Buffer (RSB) (sometimes referred to as the Return Address Stack (RAS) or
+Return Address Predictor (RAP) on AMD).
+
+Information about these CVEs and how to mitigate them is scattered
+amongst a myriad of microarchitecture-specific documents.
+
+This document attempts to consolidate all the relevant information in
+once place and clarify the reasoning behind the current RSB-related
+mitigations.  It's meant to be as concise as possible, focused only on
+the current kernel mitigations: what are the RSB-related attack vectors
+and how are they currently being mitigated?
+
+It's *not* meant to describe how the RSB mechanism operates or how the
+exploits work.  More details about those can be found in the references
+below.
+
+Rather, this is basically a glorified comment, but too long to actually
+be one.  So when the next CVE comes along, a kernel developer can
+quickly refer to this as a refresher to see what we're actually doing
+and why.
+
+At a high level, there are two classes of RSB attacks: RSB poisoning
+(Intel and AMD) and RSB underflow (Intel only).  They must each be
+considered individually for each attack vector (and microarchitecture
+where applicable).
+
+----
+
+RSB poisoning (Intel and AMD)
+=============================
+
+SpectreRSB
+~~~~~~~~~~
+
+RSB poisoning is a technique used by SpectreRSB [#spectre-rsb]_ where
+an attacker poisons an RSB entry to cause a victim's return instruction
+to speculate to an attacker-controlled address.  This can happen when
+there are unbalanced CALLs/RETs after a context switch or VMEXIT.
+
+* All attack vectors can potentially be mitigated by flushing out any
+  poisoned RSB entries using an RSB filling sequence
+  [#intel-rsb-filling]_ [#amd-rsb-filling]_ when transitioning between
+  untrusted and trusted domains.  But this has a performance impact and
+  should be avoided whenever possible.
+
+  .. DANGER::
+     **FIXME**: Currently we're flushing 32 entries.  However, some CPU
+     models have more than 32 entries.  The loop count needs to be
+     increased for those.  More detailed information is needed about RSB
+     sizes.
+
+* On context switch, the user->user mitigation requires ensuring the
+  RSB gets filled or cleared whenever IBPB gets written [#cond-ibpb]_
+  during a context switch:
+
+  * AMD:
+	On Zen 4+, IBPB (or SBPB [#amd-sbpb]_ if used) clears the RSB.
+	This is indicated by IBPB_RET in CPUID [#amd-ibpb-rsb]_.
+
+	On Zen < 4, the RSB filling sequence [#amd-rsb-filling]_ must be
+	always be done in addition to IBPB [#amd-ibpb-no-rsb]_.  This is
+	indicated by X86_BUG_IBPB_NO_RET.
+
+  * Intel:
+	IBPB always clears the RSB:
+
+	"Software that executed before the IBPB command cannot control
+	the predicted targets of indirect branches executed after the
+	command on the same logical processor. The term indirect branch
+	in this context includes near return instructions, so these
+	predicted targets may come from the RSB." [#intel-ibpb-rsb]_
+
+* On context switch, user->kernel attacks are prevented by SMEP.  User
+  space can only insert user space addresses into the RSB.  Even
+  non-canonical addresses can't be inserted due to the page gap at the
+  end of the user canonical address space reserved by TASK_SIZE_MAX.
+  A SMEP #PF at instruction fetch prevents the kernel from speculatively
+  executing user space.
+
+  * AMD:
+	"Finally, branches that are predicted as 'ret' instructions get
+	their predicted targets from the Return Address Predictor (RAP).
+	AMD recommends software use a RAP stuffing sequence (mitigation
+	V2-3 in [2]) and/or Supervisor Mode Execution Protection (SMEP)
+	to ensure that the addresses in the RAP are safe for
+	speculation. Collectively, we refer to these mitigations as "RAP
+	Protection"." [#amd-smep-rsb]_
+
+  * Intel:
+	"On processors with enhanced IBRS, an RSB overwrite sequence may
+	not suffice to prevent the predicted target of a near return
+	from using an RSB entry created in a less privileged predictor
+	mode.  Software can prevent this by enabling SMEP (for
+	transitions from user mode to supervisor mode) and by having
+	IA32_SPEC_CTRL.IBRS set during VM exits." [#intel-smep-rsb]_
+
+* On VMEXIT, guest->host attacks are mitigated by eIBRS (and PBRSB
+  mitigation if needed):
+
+  * AMD:
+	"When Automatic IBRS is enabled, the internal return address
+	stack used for return address predictions is cleared on VMEXIT."
+	[#amd-eibrs-vmexit]_
+
+  * Intel:
+	"On processors with enhanced IBRS, an RSB overwrite sequence may
+	not suffice to prevent the predicted target of a near return
+	from using an RSB entry created in a less privileged predictor
+	mode.  Software can prevent this by enabling SMEP (for
+	transitions from user mode to supervisor mode) and by having
+	IA32_SPEC_CTRL.IBRS set during VM exits. Processors with
+	enhanced IBRS still support the usage model where IBRS is set
+	only in the OS/VMM for OSes that enable SMEP. To do this, such
+	processors will ensure that guest behavior cannot control the
+	RSB after a VM exit once IBRS is set, even if IBRS was not set
+	at the time of the VM exit." [#intel-eibrs-vmexit]_
+
+    Note that some Intel CPUs are susceptible to Post-barrier Return
+    Stack Buffer Predictions (PBRSB) [#intel-pbrsb]_, where the last
+    CALL from the guest can be used to predict the first unbalanced RET.
+    In this case the PBRSB mitigation is needed in addition to eIBRS.
+
+AMD RETBleed / SRSO / Branch Type Confusion
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+On AMD, poisoned RSB entries can also be created by the AMD RETBleed
+variant [#retbleed-paper]_ [#amd-btc]_ or by Speculative Return Stack
+Overflow [#amd-srso]_ (Inception [#inception-paper]_).  The kernel
+protects itself by replacing every RET in the kernel with a branch to a
+single safe RET.
+
+----
+
+RSB underflow (Intel only)
+==========================
+
+RSB Alternate (RSBA) ("Intel Retbleed")
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some Intel Skylake-generation CPUs are susceptible to the Intel variant
+of RETBleed [#retbleed-paper]_ (Return Stack Buffer Underflow
+[#intel-rsbu]_).  If a RET is executed when the RSB buffer is empty due
+to mismatched CALLs/RETs or returning from a deep call stack, the branch
+predictor can fall back to using the Branch Target Buffer (BTB).  If a
+user forces a BTB collision then the RET can speculatively branch to a
+user-controlled address.
+
+* Note that RSB filling doesn't fully mitigate this issue.  If there
+  are enough unbalanced RETs, the RSB may still underflow and fall back
+  to using a poisoned BTB entry.
+
+* On context switch, user->user underflow attacks are mitigated by the
+  conditional IBPB [#cond-ibpb]_ on context switch which effectively
+  clears the BTB:
+
+  * "The indirect branch predictor barrier (IBPB) is an indirect branch
+    control mechanism that establishes a barrier, preventing software
+    that executed before the barrier from controlling the predicted
+    targets of indirect branches executed after the barrier on the same
+    logical processor." [#intel-ibpb-btb]_
+
+* On context switch and VMEXIT, user->kernel and guest->host RSB
+  underflows are mitigated by IBRS or eIBRS:
+
+  * "Enabling IBRS (including enhanced IBRS) will mitigate the "RSBU"
+    attack demonstrated by the researchers. As previously documented,
+    Intel recommends the use of enhanced IBRS, where supported. This
+    includes any processor that enumerates RRSBA but not RRSBA_DIS_S."
+    [#intel-rsbu]_
+
+  However, note that eIBRS and IBRS do not mitigate intra-mode attacks.
+  Like RRSBA below, this is mitigated by clearing the BHB on kernel
+  entry.
+
+  As an alternative to classic IBRS, call depth tracking (combined with
+  retpolines) can be used to track kernel returns and fill the RSB when
+  it gets close to being empty.
+
+Restricted RSB Alternate (RRSBA)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some newer Intel CPUs have Restricted RSB Alternate (RRSBA) behavior,
+which, similar to RSBA described above, also falls back to using the BTB
+on RSB underflow.  The only difference is that the predicted targets are
+restricted to the current domain when eIBRS is enabled:
+
+* "Restricted RSB Alternate (RRSBA) behavior allows alternate branch
+  predictors to be used by near RET instructions when the RSB is
+  empty.  When eIBRS is enabled, the predicted targets of these
+  alternate predictors are restricted to those belonging to the
+  indirect branch predictor entries of the current prediction domain.
+  [#intel-eibrs-rrsba]_
+
+When a CPU with RRSBA is vulnerable to Branch History Injection
+[#bhi-paper]_ [#intel-bhi]_, an RSB underflow could be used for an
+intra-mode BTI attack.  This is mitigated by clearing the BHB on
+kernel entry.
+
+However if the kernel uses retpolines instead of eIBRS, it needs to
+disable RRSBA:
+
+* "Where software is using retpoline as a mitigation for BHI or
+  intra-mode BTI, and the processor both enumerates RRSBA and
+  enumerates RRSBA_DIS controls, it should disable this behavior."
+  [#intel-retpoline-rrsba]_
+
+----
+
+References
+==========
+
+.. [#spectre-rsb] `Spectre Returns! Speculation Attacks using the Return Stack Buffer <https://arxiv.org/pdf/1807.07940.pdf>`_
+
+.. [#intel-rsb-filling] "Empty RSB Mitigation on Skylake-generation" in `Retpoline: A Branch Target Injection Mitigation <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/retpoline-branch-target-injection-mitigation.html#inpage-nav-5-1>`_
+
+.. [#amd-rsb-filling] "Mitigation V2-3" in `Software Techniques for Managing Speculation <https://www.amd.com/content/dam/amd/en/documents/processor-tech-docs/programmer-references/software-techniques-for-managing-speculation.pdf>`_
+
+.. [#cond-ibpb] Whether IBPB is written depends on whether the prev and/or next task is protected from Spectre attacks.  It typically requires opting in per task or system-wide.  For more details see the documentation for the ``spectre_v2_user`` cmdline option in Documentation/admin-guide/kernel-parameters.txt.
+
+.. [#amd-sbpb] IBPB without flushing of branch type predictions.  Only exists for AMD.
+
+.. [#amd-ibpb-rsb] "Function 8000_0008h -- Processor Capacity Parameters and Extended Feature Identification" in `AMD64 Architecture Programmer's Manual Volume 3: General-Purpose and System Instructions <https://www.amd.com/content/dam/amd/en/documents/processor-tech-docs/programmer-references/24594.pdf>`_.  SBPB behaves the same way according to `this email <https://lore.kernel.org/5175b163a3736ca5fd01cedf406735636c99a>`_.
+
+.. [#amd-ibpb-no-rsb] `Spectre Attacks: Exploiting Speculative Execution <https://comsec.ethz.ch/wp-content/files/ibpb_sp25.pdf>`_
+
+.. [#intel-ibpb-rsb] "Introduction" in `Post-barrier Return Stack Buffer Predictions / CVE-2022-26373 / INTEL-SA-00706 <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/advisory-guidance/post-barrier-return-stack-buffer-predictions.html>`_
+
+.. [#amd-smep-rsb] "Existing Mitigations" in `Technical Guidance for Mitigating Branch Type Confusion <https://www.amd.com/content/dam/amd/en/documents/resources/technical-guidance-for-mitigating-branch-type-confusion.pdf>`_
+
+.. [#intel-smep-rsb] "Enhanced IBRS" in `Indirect Branch Restricted Speculation <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/indirect-branch-restricted-speculation.html>`_
+
+.. [#amd-eibrs-vmexit] "Extended Feature Enable Register (EFER)" in `AMD64 Architecture Programmer's Manual Volume 2: System Programming <https://www.amd.com/content/dam/amd/en/documents/processor-tech-docs/programmer-references/24593.pdf>`_
+
+.. [#intel-eibrs-vmexit] "Enhanced IBRS" in `Indirect Branch Restricted Speculation <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/indirect-branch-restricted-speculation.html>`_
+
+.. [#intel-pbrsb] `Post-barrier Return Stack Buffer Predictions / CVE-2022-26373 / INTEL-SA-00706 <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/advisory-guidance/post-barrier-return-stack-buffer-predictions.html>`_
+
+.. [#retbleed-paper] `RETBleed: Arbitrary Speculative Code Execution with Return Instruction <https://comsec.ethz.ch/wp-content/files/retbleed_sec22.pdf>`_
+
+.. [#amd-btc] `Technical Guidance for Mitigating Branch Type Confusion <https://www.amd.com/content/dam/amd/en/documents/resources/technical-guidance-for-mitigating-branch-type-confusion.pdf>`_
+
+.. [#amd-srso] `Technical Update Regarding Speculative Return Stack Overflow <https://www.amd.com/content/dam/amd/en/documents/corporate/cr/speculative-return-stack-overflow-whitepaper.pdf>`_
+
+.. [#inception-paper] `Inception: Exposing New Attack Surfaces with Training in Transient Execution <https://comsec.ethz.ch/wp-content/files/inception_sec23.pdf>`_
+
+.. [#intel-rsbu] `Return Stack Buffer Underflow / Return Stack Buffer Underflow / CVE-2022-29901, CVE-2022-28693 / INTEL-SA-00702 <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/advisory-guidance/return-stack-buffer-underflow.html>`_
+
+.. [#intel-ibpb-btb] `Indirect Branch Predictor Barrier' <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/indirect-branch-predictor-barrier.html>`_
+
+.. [#intel-eibrs-rrsba] "Guidance for RSBU" in `Return Stack Buffer Underflow / Return Stack Buffer Underflow / CVE-2022-29901, CVE-2022-28693 / INTEL-SA-00702 <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/advisory-guidance/return-stack-buffer-underflow.html>`_
+
+.. [#bhi-paper] `Branch History Injection: On the Effectiveness of Hardware Mitigations Against Cross-Privilege Spectre-v2 Attacks <http://download.vusec.net/papers/bhi-spectre-bhb_sec22.pdf>`_
+
+.. [#intel-bhi] `Branch History Injection and Intra-mode Branch Target Injection / CVE-2022-0001, CVE-2022-0002 / INTEL-SA-00598 <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/branch-history-injection.html>`_
+
+.. [#intel-retpoline-rrsba] "Retpoline" in `Branch History Injection and Intra-mode Branch Target Injection / CVE-2022-0001, CVE-2022-0002 / INTEL-SA-00598 <https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/branch-history-injection.html>`_
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 76e538c77e31..d9fd26b95b34 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -1407,18 +1407,15 @@
 			earlyprintk=serial[,0x...[,baudrate]]
 			earlyprintk=ttySn[,baudrate]
 			earlyprintk=dbgp[debugController#]
+			earlyprintk=mmio32,membase[,{nocfg|baudrate}]
 			earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}]
 			earlyprintk=xdbc[xhciController#]
 			earlyprintk=bios
-			earlyprintk=mmio,membase[,{nocfg|baudrate}]
 
 			earlyprintk is useful when the kernel crashes before
 			the normal console is initialized. It is not enabled by
 			default because it has some cosmetic problems.
 
-			Only 32-bit memory addresses are supported for "mmio"
-			and "pciserial" devices.
-
 			Use "nocfg" to skip UART configuration, assume
 			BIOS/firmware has configured UART correctly.
 
diff --git a/Documentation/admin-guide/xfs.rst b/Documentation/admin-guide/xfs.rst
index b67772cf36d6..5becb441c3cb 100644
--- a/Documentation/admin-guide/xfs.rst
+++ b/Documentation/admin-guide/xfs.rst
@@ -124,6 +124,14 @@ When mounting an XFS filesystem, the following options are accepted.
 	controls the size of each buffer and so is also relevant to
 	this case.
 
+  lifetime (default) or nolifetime
+	Enable data placement based on write life time hints provided
+	by the user. This turns on co-allocation of data of similar
+	life times when statistically favorable to reduce garbage
+	collection cost.
+
+	These options are only available for zoned rt file systems.
+
   logbsize=value
 	Set the size of each in-memory log buffer.  The size may be
 	specified in bytes, or in kilobytes with a "k" suffix.
@@ -143,6 +151,14 @@ When mounting an XFS filesystem, the following options are accepted.
 	optional, and the log section can be separate from the data
 	section or contained within it.
 
+  max_open_zones=value
+	Specify the max number of zones to keep open for writing on a
+	zoned rt device. Many open zones aids file data separation
+	but may impact performance on HDDs.
+
+	If ``max_open_zones`` is not specified, the value is determined
+	by the capabilities and the size of the zoned rt device.
+
   noalign
 	Data allocations will not be aligned at stripe unit
 	boundaries. This is only relevant to filesystems created
@@ -542,3 +558,24 @@ The interesting knobs for XFS workqueues are as follows:
   nice           Relative priority of scheduling the threads.  These are the
                  same nice levels that can be applied to userspace processes.
 ============     ===========
+
+Zoned Filesystems
+=================
+
+For zoned file systems, the following attributes are exposed in:
+
+  /sys/fs/xfs/<dev>/zoned/
+
+  max_open_zones		(Min:  1  Default:  Varies  Max:  UINTMAX)
+	This read-only attribute exposes the maximum number of open zones
+	available for data placement. The value is determined at mount time and
+	is limited by the capabilities of the backing zoned device, file system
+	size and the max_open_zones mount option.
+
+  zonegc_low_space		(Min:  0  Default:  0  Max:  100)
+	Define a percentage for how much of the unused space that GC should keep
+	available for writing. A high value will reclaim more of the space
+	occupied by unused blocks, creating a larger buffer against write
+	bursts at the cost of increased write amplification.  Regardless
+	of this value, garbage collection will always aim to free a minimum
+	amount of blocks to keep max_open_zones open for data placement purposes.
diff --git a/Documentation/arch/openrisc/openrisc_port.rst b/Documentation/arch/openrisc/openrisc_port.rst
index 1565b9546e38..a8f307a3b499 100644
--- a/Documentation/arch/openrisc/openrisc_port.rst
+++ b/Documentation/arch/openrisc/openrisc_port.rst
@@ -7,10 +7,10 @@ target architecture, specifically, is the 32-bit OpenRISC 1000 family (or1k).
 
 For information about OpenRISC processors and ongoing development:
 
-	=======		=============================
+	=======		==============================
 	website		https://openrisc.io
-	email		openrisc@lists.librecores.org
-	=======		=============================
+	email		linux-openrisc@vger.kernel.org
+	=======		==============================
 
 ---------------------------------------------------------------------
 
@@ -27,11 +27,11 @@ Toolchain binaries can be obtained from openrisc.io or our github releases page.
 Instructions for building the different toolchains can be found on openrisc.io
 or Stafford's toolchain build and release scripts.
 
-	==========	=================================================
-	binaries	https://github.com/openrisc/or1k-gcc/releases
+	==========	==========================================================
+	binaries	https://github.com/stffrdhrn/or1k-toolchain-build/releases
 	toolchains	https://openrisc.io/software
 	building	https://github.com/stffrdhrn/or1k-toolchain-build
-	==========	=================================================
+	==========	==========================================================
 
 2) Building
 
diff --git a/Documentation/arch/riscv/hwprobe.rst b/Documentation/arch/riscv/hwprobe.rst
index 53607d962653..f60bf5991755 100644
--- a/Documentation/arch/riscv/hwprobe.rst
+++ b/Documentation/arch/riscv/hwprobe.rst
@@ -51,7 +51,7 @@ The following keys are defined:
 * :c:macro:`RISCV_HWPROBE_KEY_MARCHID`: Contains the value of ``marchid``, as
   defined by the RISC-V privileged architecture specification.
 
-* :c:macro:`RISCV_HWPROBE_KEY_MIMPLID`: Contains the value of ``mimplid``, as
+* :c:macro:`RISCV_HWPROBE_KEY_MIMPID`: Contains the value of ``mimpid``, as
   defined by the RISC-V privileged architecture specification.
 
 * :c:macro:`RISCV_HWPROBE_KEY_BASE_BEHAVIOR`: A bitmask containing the base
diff --git a/Documentation/arch/x86/cpuinfo.rst b/Documentation/arch/x86/cpuinfo.rst
index 6ef426a52cdc..f80e2a558d2a 100644
--- a/Documentation/arch/x86/cpuinfo.rst
+++ b/Documentation/arch/x86/cpuinfo.rst
@@ -79,8 +79,9 @@ feature flags.
 How are feature flags created?
 ==============================
 
-a: Feature flags can be derived from the contents of CPUID leaves.
-------------------------------------------------------------------
+Feature flags can be derived from the contents of CPUID leaves
+--------------------------------------------------------------
+
 These feature definitions are organized mirroring the layout of CPUID
 leaves and grouped in words with offsets as mapped in enum cpuid_leafs
 in cpufeatures.h (see arch/x86/include/asm/cpufeatures.h for details).
@@ -89,8 +90,9 @@ cpufeatures.h, and if it is detected at run time, the flags will be
 displayed accordingly in /proc/cpuinfo. For example, the flag "avx2"
 comes from X86_FEATURE_AVX2 in cpufeatures.h.
 
-b: Flags can be from scattered CPUID-based features.
-----------------------------------------------------
+Flags can be from scattered CPUID-based features
+------------------------------------------------
+
 Hardware features enumerated in sparsely populated CPUID leaves get
 software-defined values. Still, CPUID needs to be queried to determine
 if a given feature is present. This is done in init_scattered_cpuid_features().
@@ -104,8 +106,9 @@ has only one feature and would waste 31 bits of space in the x86_capability[]
 array. Since there is a struct cpuinfo_x86 for each possible CPU, the wasted
 memory is not trivial.
 
-c: Flags can be created synthetically under certain conditions for hardware features.
--------------------------------------------------------------------------------------
+Flags can be created synthetically under certain conditions for hardware features
+---------------------------------------------------------------------------------
+
 Examples of conditions include whether certain features are present in
 MSR_IA32_CORE_CAPS or specific CPU models are identified. If the needed
 conditions are met, the features are enabled by the set_cpu_cap or
@@ -114,8 +117,8 @@ the feature X86_FEATURE_SPLIT_LOCK_DETECT will be enabled and
 "split_lock_detect" will be displayed. The flag "ring3mwait" will be
 displayed only when running on INTEL_XEON_PHI_[KNL|KNM] processors.
 
-d: Flags can represent purely software features.
-------------------------------------------------
+Flags can represent purely software features
+--------------------------------------------
 These flags do not represent hardware features. Instead, they represent a
 software feature implemented in the kernel. For example, Kernel Page Table
 Isolation is purely software feature and its feature flag X86_FEATURE_PTI is
@@ -130,14 +133,18 @@ x86_cap/bug_flags[] arrays in kernel/cpu/capflags.c. The names in the
 resulting x86_cap/bug_flags[] are used to populate /proc/cpuinfo. The naming
 of flags in the x86_cap/bug_flags[] are as follows:
 
-a: The name of the flag is from the string in X86_FEATURE_<name> by default.
-----------------------------------------------------------------------------
-By default, the flag <name> in /proc/cpuinfo is extracted from the respective
-X86_FEATURE_<name> in cpufeatures.h. For example, the flag "avx2" is from
-X86_FEATURE_AVX2.
+Flags do not appear by default in /proc/cpuinfo
+-----------------------------------------------
+
+Feature flags are omitted by default from /proc/cpuinfo as it does not make
+sense for the feature to be exposed to userspace in most cases. For example,
+X86_FEATURE_ALWAYS is defined in cpufeatures.h but that flag is an internal
+kernel feature used in the alternative runtime patching functionality. So the
+flag does not appear in /proc/cpuinfo.
+
+Specify a flag name if absolutely needed
+----------------------------------------
 
-b: The naming can be overridden.
---------------------------------
 If the comment on the line for the #define X86_FEATURE_* starts with a
 double-quote character (""), the string inside the double-quote characters
 will be the name of the flags. For example, the flag "sse4_1" comes from
@@ -148,36 +155,31 @@ needed. For instance, /proc/cpuinfo is a userspace interface and must remain
 constant. If, for some reason, the naming of X86_FEATURE_<name> changes, one
 shall override the new naming with the name already used in /proc/cpuinfo.
 
-c: The naming override can be "", which means it will not appear in /proc/cpuinfo.
-----------------------------------------------------------------------------------
-The feature shall be omitted from /proc/cpuinfo if it does not make sense for
-the feature to be exposed to userspace. For example, X86_FEATURE_ALWAYS is
-defined in cpufeatures.h but that flag is an internal kernel feature used
-in the alternative runtime patching functionality. So, its name is overridden
-with "". Its flag will not appear in /proc/cpuinfo.
-
 Flags are missing when one or more of these happen
 ==================================================
 
-a: The hardware does not enumerate support for it.
---------------------------------------------------
+The hardware does not enumerate support for it
+----------------------------------------------
+
 For example, when a new kernel is running on old hardware or the feature is
 not enabled by boot firmware. Even if the hardware is new, there might be a
 problem enabling the feature at run time, the flag will not be displayed.
 
-b: The kernel does not know about the flag.
--------------------------------------------
+The kernel does not know about the flag
+---------------------------------------
+
 For example, when an old kernel is running on new hardware.
 
-c: The kernel disabled support for it at compile-time.
-------------------------------------------------------
+The kernel disabled support for it at compile-time
+--------------------------------------------------
+
 For example, if 5-level-paging is not enabled when building (i.e.,
 CONFIG_X86_5LEVEL is not selected) the flag "la57" will not show up [#f1]_.
 Even though the feature will still be detected via CPUID, the kernel disables
 it by clearing via setup_clear_cpu_cap(X86_FEATURE_LA57).
 
-d: The feature is disabled at boot-time.
-----------------------------------------
+The feature is disabled at boot-time
+------------------------------------
 A feature can be disabled either using a command-line parameter or because
 it failed to be enabled. The command-line parameter clearcpuid= can be used
 to disable features using the feature number as defined in
@@ -190,8 +192,9 @@ disable specific features. The list of parameters includes, but is not limited
 to, nofsgsbase, nosgx, noxsave, etc. 5-level paging can also be disabled using
 "no5lvl".
 
-e: The feature was known to be non-functional.
-----------------------------------------------
+The feature was known to be non-functional
+------------------------------------------
+
 The feature was known to be non-functional because a dependency was
 missing at runtime. For example, AVX flags will not show up if XSAVE feature
 is disabled since they depend on XSAVE feature. Another example would be broken
diff --git a/Documentation/bpf/bpf_devel_QA.rst b/Documentation/bpf/bpf_devel_QA.rst
index de27e1620821..0acb4c9b8d90 100644
--- a/Documentation/bpf/bpf_devel_QA.rst
+++ b/Documentation/bpf/bpf_devel_QA.rst
@@ -382,6 +382,14 @@ In case of new BPF instructions, once the changes have been accepted
 into the Linux kernel, please implement support into LLVM's BPF back
 end. See LLVM_ section below for further information.
 
+Q: What "BPF_INTERNAL" symbol namespace is for?
+-----------------------------------------------
+A: Symbols exported as BPF_INTERNAL can only be used by BPF infrastructure
+like preload kernel modules with light skeleton. Most symbols outside
+of BPF_INTERNAL are not expected to be used by code outside of BPF either.
+Symbols may lack the designation because they predate the namespaces,
+or due to an oversight.
+
 Stable submission
 =================
 
diff --git a/Documentation/core-api/printk-formats.rst b/Documentation/core-api/printk-formats.rst
index 4bdc394e86af..f531873bb3c9 100644
--- a/Documentation/core-api/printk-formats.rst
+++ b/Documentation/core-api/printk-formats.rst
@@ -648,6 +648,38 @@ Examples::
 	%p4cc	Y10  little-endian (0x20303159)
 	%p4cc	NV12 big-endian (0xb231564e)
 
+Generic FourCC code
+-------------------
+
+::
+	%p4c[h[R]lb]	gP00 (0x67503030)
+
+Print a generic FourCC code, as both ASCII characters and its numerical
+value as hexadecimal.
+
+The generic FourCC code is always printed in the big-endian format,
+the most significant byte first. This is the opposite of V4L/DRM FourCCs.
+
+The additional ``h``, ``hR``, ``l``, and ``b`` specifiers define what
+endianness is used to load the stored bytes. The data might be interpreted
+using the host, reversed host byte order, little-endian, or big-endian.
+
+Passed by reference.
+
+Examples for a little-endian machine, given &(u32)0x67503030::
+
+	%p4ch	gP00 (0x67503030)
+	%p4chR	00Pg (0x30305067)
+	%p4cl	gP00 (0x67503030)
+	%p4cb	00Pg (0x30305067)
+
+Examples for a big-endian machine, given &(u32)0x67503030::
+
+	%p4ch	gP00 (0x67503030)
+	%p4chR	00Pg (0x30305067)
+	%p4cl	00Pg (0x30305067)
+	%p4cb	gP00 (0x67503030)
+
 Rust
 ----
 
diff --git a/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml b/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml
index 6ad78429dc74..c92341888a28 100644
--- a/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml
+++ b/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml
@@ -7,7 +7,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Ceva AHCI SATA Controller
 
 maintainers:
-  - Mubin Sayyed <mubin.sayyed@amd.com>
   - Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
 
 description: |
diff --git a/Documentation/devicetree/bindings/display/bridge/nwl-dsi.yaml b/Documentation/devicetree/bindings/display/bridge/nwl-dsi.yaml
index 350fb8f400f0..5952e6448ed4 100644
--- a/Documentation/devicetree/bindings/display/bridge/nwl-dsi.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/nwl-dsi.yaml
@@ -111,11 +111,27 @@ properties:
         unevaluatedProperties: false
 
       port@1:
-        $ref: /schemas/graph.yaml#/properties/port
+        $ref: /schemas/graph.yaml#/$defs/port-base
+        unevaluatedProperties: false
         description:
           DSI output port node to the panel or the next bridge
           in the chain
 
+        properties:
+          endpoint:
+            $ref: /schemas/media/video-interfaces.yaml#
+            unevaluatedProperties: false
+
+            properties:
+              data-lanes:
+                description: array of physical DSI data lane indexes.
+                minItems: 1
+                items:
+                  - const: 1
+                  - const: 2
+                  - const: 3
+                  - const: 4
+
     required:
       - port@0
       - port@1
diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,mt8195-hdmi-ddc.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,mt8195-hdmi-ddc.yaml
new file mode 100644
index 000000000000..bde4dc556d4f
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,mt8195-hdmi-ddc.yaml
@@ -0,0 +1,41 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/mediatek/mediatek,mt8195-hdmi-ddc.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: MediaTek HDMI MT8195 series HDMI Display Data Channel (DDC)
+
+maintainers:
+  - AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
+  - CK Hu <ck.hu@mediatek.com>
+
+properties:
+  compatible:
+    oneOf:
+      - const: mediatek,mt8195-hdmi-ddc
+      - items:
+          - const: mediatek,mt8188-hdmi-ddc
+          - const: mediatek,mt8195-hdmi-ddc
+
+  clocks:
+    maxItems: 1
+
+  power-domains:
+    maxItems: 1
+
+required:
+  - compatible
+  - clocks
+
+additionalProperties: false
+
+examples:
+  - |
+    hdmi {
+        hdmi_ddc: i2c {
+            compatible = "mediatek,mt8195-hdmi-ddc";
+            clocks = <&clk26m>;
+        };
+    };
+...
diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,mt8195-hdmi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,mt8195-hdmi.yaml
new file mode 100644
index 000000000000..1b382f99d3ce
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,mt8195-hdmi.yaml
@@ -0,0 +1,151 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/mediatek/mediatek,mt8195-hdmi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: MediaTek MT8195 series HDMI-TX Encoder
+
+maintainers:
+  - AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
+  - CK Hu <ck.hu@mediatek.com>
+
+description:
+  The MediaTek HDMI-TX v2 encoder can generate HDMI format data based on
+  the HDMI Specification 2.0b.
+
+properties:
+  compatible:
+    enum:
+      - mediatek,mt8188-hdmi-tx
+      - mediatek,mt8195-hdmi-tx
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    items:
+      - description: HDMI Peripheral Bus (APB) clock
+      - description: HDCP and HDMI_TOP clock
+      - description: HDCP, HDMI_TOP and HDMI Audio reference clock
+      - description: VPP HDMI Split clock
+
+  clock-names:
+    items:
+      - const: bus
+      - const: hdcp
+      - const: hdcp24m
+      - const: hdmi-split
+
+  i2c:
+    type: object
+    $ref: /schemas/display/mediatek/mediatek,mt8195-hdmi-ddc.yaml
+    unevaluatedProperties: false
+    description: HDMI DDC I2C controller
+
+  phys:
+    maxItems: 1
+    description: PHY providing clocking TMDS and pixel to controller
+
+  phy-names:
+    items:
+      - const: hdmi
+
+  power-domains:
+    maxItems: 1
+
+  '#sound-dai-cells':
+    const: 1
+
+  ports:
+    $ref: /schemas/graph.yaml#/properties/ports
+
+    properties:
+      port@0:
+        $ref: /schemas/graph.yaml#/properties/port
+        description:
+          Input port, usually connected to the output port of a DPI
+
+      port@1:
+        $ref: /schemas/graph.yaml#/properties/port
+        description:
+          Output port that must be connected either to the input port of
+          a HDMI connector node containing a ddc-i2c-bus, or to the input
+          port of an attached bridge chip, such as a SlimPort transmitter.
+
+    required:
+      - port@0
+      - port@1
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - clocks
+  - clock-names
+  - power-domains
+  - phys
+  - phy-names
+  - ports
+
+allOf:
+  - $ref: /schemas/sound/dai-common.yaml#
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/mt8195-clk.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/mt8195-power.h>
+
+    soc {
+        #address-cells = <2>;
+        #size-cells = <2>;
+
+        hdmi@1c300000 {
+            compatible = "mediatek,mt8195-hdmi-tx";
+            reg = <0 0x1c300000 0 0x1000>;
+            clocks = <&topckgen CLK_TOP_HDMI_APB>,
+                     <&topckgen CLK_TOP_HDCP>,
+                     <&topckgen CLK_TOP_HDCP_24M>,
+                     <&vppsys1 CLK_VPP1_VPP_SPLIT_HDMI>;
+            clock-names = "bus", "hdcp", "hdcp24m", "hdmi-split";
+            interrupts = <GIC_SPI 677 IRQ_TYPE_LEVEL_HIGH 0>;
+            phys = <&hdmi_phy>;
+            phy-names = "hdmi";
+            power-domains = <&spm MT8195_POWER_DOMAIN_HDMI_TX>;
+            pinctrl-names = "default";
+            pinctrl-0 = <&hdmi_pins>;
+            #sound-dai-cells = <1>;
+
+            hdmitx_ddc: i2c {
+                compatible = "mediatek,mt8195-hdmi-ddc";
+                clocks = <&clk26m>;
+            };
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+
+                    hdmi_in: endpoint {
+                        remote-endpoint = <&dpi1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+
+                    hdmi_out: endpoint {
+                        remote-endpoint = <&hdmi_connector_in>;
+                    };
+                };
+            };
+        };
+    };
diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index e00b88332f2f..246bbb509bea 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -31,6 +31,7 @@ properties:
           - qcom,sm8650-dp
       - items:
           - enum:
+              - qcom,sar2130p-dp
               - qcom,sm6350-dp
               - qcom,sm8150-dp
               - qcom,sm8250-dp
diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
index 2aab33cd0017..82fe95a6d959 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
@@ -23,6 +23,8 @@ properties:
               - qcom,msm8996-dsi-ctrl
               - qcom,msm8998-dsi-ctrl
               - qcom,qcm2290-dsi-ctrl
+              - qcom,sa8775p-dsi-ctrl
+              - qcom,sar2130p-dsi-ctrl
               - qcom,sc7180-dsi-ctrl
               - qcom,sc7280-dsi-ctrl
               - qcom,sdm660-dsi-ctrl
@@ -314,6 +316,8 @@ allOf:
           contains:
             enum:
               - qcom,msm8998-dsi-ctrl
+              - qcom,sa8775p-dsi-ctrl
+              - qcom,sar2130p-dsi-ctrl
               - qcom,sc7180-dsi-ctrl
               - qcom,sc7280-dsi-ctrl
               - qcom,sdm845-dsi-ctrl
diff --git a/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml b/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml
index 321470435e65..3c75ff42999a 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml
@@ -17,6 +17,8 @@ properties:
     enum:
       - qcom,dsi-phy-7nm
       - qcom,dsi-phy-7nm-8150
+      - qcom,sa8775p-dsi-phy-5nm
+      - qcom,sar2130p-dsi-phy-5nm
       - qcom,sc7280-dsi-phy-7nm
       - qcom,sm6375-dsi-phy-7nm
       - qcom,sm8350-dsi-phy-5nm
diff --git a/Documentation/devicetree/bindings/display/msm/hdmi.yaml b/Documentation/devicetree/bindings/display/msm/hdmi.yaml
index d4a2033afea8..dfec6c3480f3 100644
--- a/Documentation/devicetree/bindings/display/msm/hdmi.yaml
+++ b/Documentation/devicetree/bindings/display/msm/hdmi.yaml
@@ -66,21 +66,6 @@ properties:
     maxItems: 1
     description: hpd pin
 
-  qcom,hdmi-tx-mux-en-gpios:
-    maxItems: 1
-    deprecated: true
-    description: HDMI mux enable pin
-
-  qcom,hdmi-tx-mux-sel-gpios:
-    maxItems: 1
-    deprecated: true
-    description: HDMI mux select pin
-
-  qcom,hdmi-tx-mux-lpm-gpios:
-    maxItems: 1
-    deprecated: true
-    description: HDMI mux lpm pin
-
   '#sound-dai-cells':
     const: 1
 
@@ -89,12 +74,12 @@ properties:
     $ref: /schemas/graph.yaml#/properties/ports
     properties:
       port@0:
-        $ref: /schemas/graph.yaml#/$defs/port-base
+        $ref: /schemas/graph.yaml#/properties/port
         description: |
           Input endpoints of the controller.
 
       port@1:
-        $ref: /schemas/graph.yaml#/$defs/port-base
+        $ref: /schemas/graph.yaml#/properties/port
         description: |
           Output endpoints of the controller.
 
diff --git a/Documentation/devicetree/bindings/display/msm/mdp4.yaml b/Documentation/devicetree/bindings/display/msm/mdp4.yaml
index 35204a287579..03ee09faa335 100644
--- a/Documentation/devicetree/bindings/display/msm/mdp4.yaml
+++ b/Documentation/devicetree/bindings/display/msm/mdp4.yaml
@@ -18,9 +18,10 @@ properties:
 
   clocks:
     minItems: 6
-    maxItems: 6
+    maxItems: 8
 
   clock-names:
+    minItems: 6
     items:
       - const: core_clk
       - const: iface_clk
@@ -28,6 +29,12 @@ properties:
       - const: lut_clk
       - const: hdmi_clk
       - const: tv_clk
+      - const: lcdc_clk
+      - const: pxo
+        description: XO used to drive the internal LVDS PLL
+
+  '#clock-cells':
+    const: 0
 
   reg:
     maxItems: 1
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml
index 7c6462caa442..db9c43b20e2a 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml
@@ -84,6 +84,18 @@ properties:
     items:
       - description: MDSS_CORE reset
 
+  interconnects:
+    minItems: 1
+    items:
+      - description: Interconnect path from mdp0 (or a single mdp) port to the data bus
+      - description: Interconnect path from CPU to the reg bus
+
+  interconnect-names:
+    minItems: 1
+    items:
+      - const: mdp0-mem
+      - const: cpu-cfg
+
 required:
   - compatible
   - reg
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml
index 5fac3e266703..1053b3bc4908 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml
@@ -52,12 +52,23 @@ patternProperties:
         items:
           - const: qcom,sa8775p-dp
 
+  "^dsi@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+    properties:
+      compatible:
+        contains:
+          const: qcom,sa8775p-dsi-ctrl
+
   "^phy@[0-9a-f]+$":
     type: object
     additionalProperties: true
     properties:
       compatible:
-        const: qcom,sa8775p-edp-phy
+        contains:
+          enum:
+            - qcom,sa8775p-dsi-phy-5nm
+            - qcom,sa8775p-edp-phy
 
 required:
   - compatible
@@ -139,6 +150,20 @@ examples:
                         remote-endpoint = <&mdss0_dp0_in>;
                     };
                 };
+
+                port@1 {
+                    reg = <1>;
+                    dpu_intf1_out: endpoint {
+                        remote-endpoint = <&mdss0_dsi0_in>;
+                    };
+                };
+
+                port@2 {
+                    reg = <2>;
+                    dpu_intf2_out: endpoint {
+                        remote-endpoint = <&mdss0_dsi1_in>;
+                    };
+                };
             };
 
             mdss0_mdp_opp_table: opp-table {
@@ -186,6 +211,160 @@ examples:
             vdda-pll-supply = <&vreg_l4a>;
         };
 
+        dsi@ae94000 {
+            compatible = "qcom,sa8775p-dsi-ctrl", "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae94000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispc_byte_clk>,
+                     <&dispcc_intf_clk>,
+                     <&dispcc_pclk>,
+                     <&dispcc_esc_clk>,
+                     <&dispcc_ahb_clk>,
+                     <&gcc_bus_clk>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+            assigned-clocks = <&dispcc_byte_clk>,
+                              <&dispcc_pclk>;
+            assigned-clock-parents = <&mdss0_dsi0_phy 0>, <&mdss0_dsi0_phy 1>;
+            phys = <&mdss0_dsi0_phy>;
+
+            operating-points-v2 = <&dsi0_opp_table>;
+            power-domains = <&rpmhpd SA8775P_MMCX>;
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    mdss0_dsi0_in: endpoint {
+                        remote-endpoint = <&dpu_intf1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    mdss0_dsi0_out: endpoint { };
+                };
+            };
+
+            dsi0_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-358000000 {
+                    opp-hz = /bits/ 64 <358000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+            };
+        };
+
+        mdss0_dsi0_phy: phy@ae94400 {
+            compatible = "qcom,sa8775p-dsi-phy-5nm";
+            reg = <0x0ae94400 0x200>,
+                  <0x0ae94600 0x280>,
+                  <0x0ae94900 0x27c>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc_iface_clk>,
+                     <&rpmhcc_ref_clk>;
+            clock-names = "iface", "ref";
+
+            vdds-supply = <&vreg_dsi_supply>;
+        };
+
+        dsi@ae96000 {
+            compatible = "qcom,sa8775p-dsi-ctrl", "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae96000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispc_byte_clk>,
+                     <&dispcc_intf_clk>,
+                     <&dispcc_pclk>,
+                     <&dispcc_esc_clk>,
+                     <&dispcc_ahb_clk>,
+                     <&gcc_bus_clk>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+            assigned-clocks = <&dispcc_byte_clk>,
+                              <&dispcc_pclk>;
+            assigned-clock-parents = <&mdss0_dsi1_phy 0>, <&mdss0_dsi1_phy 1>;
+            phys = <&mdss0_dsi1_phy>;
+
+            operating-points-v2 = <&dsi1_opp_table>;
+            power-domains = <&rpmhpd SA8775P_MMCX>;
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    mdss0_dsi1_in: endpoint {
+                        remote-endpoint = <&dpu_intf2_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    mdss0_dsi1_out: endpoint { };
+                };
+            };
+
+            dsi1_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-358000000 {
+                    opp-hz = /bits/ 64 <358000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+            };
+        };
+
+        mdss0_dsi1_phy: phy@ae96400 {
+            compatible = "qcom,sa8775p-dsi-phy-5nm";
+            reg = <0x0ae96400 0x200>,
+                  <0x0ae96600 0x280>,
+                  <0x0ae96900 0x27c>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc_iface_clk>,
+                     <&rpmhcc_ref_clk>;
+            clock-names = "iface", "ref";
+
+            vdds-supply = <&vreg_dsi_supply>;
+        };
+
         displayport-controller@af54000 {
             compatible = "qcom,sa8775p-dp";
 
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml
new file mode 100644
index 000000000000..870144b53cec
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml
@@ -0,0 +1,439 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sar2130p-mdss.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SAR2130P Display MDSS
+
+maintainers:
+  - Dmitry Baryshkov <lumag@kernel.org>
+
+description:
+  SAR2310P MSM Mobile Display Subsystem(MDSS), which encapsulates sub-blocks like
+  DPU display controller, DSI and DP interfaces etc.
+
+$ref: /schemas/display/msm/mdss-common.yaml#
+
+properties:
+  compatible:
+    const: qcom,sar2130p-mdss
+
+  clocks:
+    items:
+      - description: Display MDSS AHB
+      - description: Display AHB
+      - description: Display hf AXI
+      - description: Display core
+
+  iommus:
+    maxItems: 1
+
+  interconnects:
+    items:
+      - description: Interconnect path from mdp0 port to the data bus
+      - description: Interconnect path from CPU to the reg bus
+
+  interconnect-names:
+    items:
+      - const: mdp0-mem
+      - const: cpu-cfg
+
+patternProperties:
+  "^display-controller@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+    properties:
+      compatible:
+        const: qcom,sar2130p-dpu
+
+  "^displayport-controller@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+    properties:
+      compatible:
+        contains:
+          const: qcom,sar2130p-dp
+
+  "^dsi@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+    properties:
+      compatible:
+        contains:
+          const: qcom,sar2130p-dsi-ctrl
+
+  "^phy@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+    properties:
+      compatible:
+        const: qcom,sar2130p-dsi-phy-5nm
+
+required:
+  - compatible
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/qcom,rpmhpd.h>
+    #include <dt-bindings/phy/phy-qcom-qmp.h>
+
+    display-subsystem@ae00000 {
+        compatible = "qcom,sar2130p-mdss";
+        reg = <0x0ae00000 0x1000>;
+        reg-names = "mdss";
+
+        interconnects = <&mmss_noc_master_mdp &mc_virt_slave_ebi1>,
+                        <&gem_noc_master_appss_proc &config_noc_slave_display_cfg>;
+        interconnect-names = "mdp0-mem", "cpu-cfg";
+
+        resets = <&dispcc_disp_cc_mdss_core_bcr>;
+
+        power-domains = <&dispcc_mdss_gdsc>;
+
+        clocks = <&dispcc_disp_cc_mdss_ahb_clk>,
+                 <&gcc_gcc_disp_ahb_clk>,
+                 <&gcc_gcc_disp_hf_axi_clk>,
+                 <&dispcc_disp_cc_mdss_mdp_clk>;
+        clock-names = "iface", "bus", "nrt_bus", "core";
+
+        interrupts = <GIC_SPI 83 IRQ_TYPE_LEVEL_HIGH>;
+        interrupt-controller;
+        #interrupt-cells = <1>;
+
+        iommus = <&apps_smmu 0x1c00 0x2>;
+
+        #address-cells = <1>;
+        #size-cells = <1>;
+        ranges;
+
+        display-controller@ae01000 {
+            compatible = "qcom,sar2130p-dpu";
+            reg = <0x0ae01000 0x8f000>,
+                  <0x0aeb0000 0x2008>;
+            reg-names = "mdp", "vbif";
+
+            clocks = <&gcc_gcc_disp_ahb_clk>,
+                     <&gcc_gcc_disp_hf_axi_clk>,
+                     <&dispcc_disp_cc_mdss_ahb_clk>,
+                     <&dispcc_disp_cc_mdss_mdp_lut_clk>,
+                     <&dispcc_disp_cc_mdss_mdp_clk>,
+                     <&dispcc_disp_cc_mdss_vsync_clk>;
+            clock-names = "bus",
+                          "nrt_bus",
+                          "iface",
+                          "lut",
+                          "core",
+                          "vsync";
+
+            assigned-clocks = <&dispcc_disp_cc_mdss_vsync_clk>;
+            assigned-clock-rates = <19200000>;
+
+            operating-points-v2 = <&mdp_opp_table>;
+            power-domains = <&rpmhpd RPMHPD_MMCX>;
+
+            interrupt-parent = <&mdss>;
+            interrupts = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+
+                    dpu_intf0_out: endpoint {
+                        remote-endpoint = <&mdss_dp0_in>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+
+                    dpu_intf1_out: endpoint {
+                        remote-endpoint = <&mdss_dsi0_in>;
+                    };
+                };
+
+                port@2 {
+                    reg = <2>;
+
+                    dpu_intf2_out: endpoint {
+                        remote-endpoint = <&mdss_dsi1_in>;
+                    };
+                };
+            };
+
+            mdp_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-200000000 {
+                    opp-hz = /bits/ 64 <200000000>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-325000000 {
+                    opp-hz = /bits/ 64 <325000000>;
+                    required-opps = <&rpmhpd_opp_svs>;
+                };
+
+                opp-375000000 {
+                    opp-hz = /bits/ 64 <375000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+
+                opp-514000000 {
+                    opp-hz = /bits/ 64 <514000000>;
+                    required-opps = <&rpmhpd_opp_nom>;
+                };
+            };
+        };
+
+        displayport-controller@ae90000 {
+            compatible = "qcom,sar2130p-dp",
+                         "qcom,sm8350-dp";
+            reg = <0xae90000 0x200>,
+                  <0xae90200 0x200>,
+                  <0xae90400 0xc00>,
+                  <0xae91000 0x400>,
+                  <0xae91400 0x400>;
+
+            interrupt-parent = <&mdss>;
+            interrupts = <12>;
+            clocks = <&dispcc_disp_cc_mdss_ahb_clk>,
+                     <&dispcc_disp_cc_mdss_dptx0_aux_clk>,
+                     <&dispcc_disp_cc_mdss_dptx0_link_clk>,
+                     <&dispcc_disp_cc_mdss_dptx0_link_intf_clk>,
+                     <&dispcc_disp_cc_mdss_dptx0_pixel0_clk>;
+            clock-names = "core_iface",
+                          "core_aux",
+                          "ctrl_link",
+                          "ctrl_link_iface",
+                          "stream_pixel";
+
+            assigned-clocks = <&dispcc_disp_cc_mdss_dptx0_link_clk_src>,
+                              <&dispcc_disp_cc_mdss_dptx0_pixel0_clk_src>;
+            assigned-clock-parents = <&usb_dp_qmpphy_QMP_USB43DP_DP_LINK_CLK>,
+                                     <&usb_dp_qmpphy_QMP_USB43DP_DP_VCO_DIV_CLK>;
+
+            phys = <&usb_dp_qmpphy QMP_USB43DP_DP_PHY>;
+            phy-names = "dp";
+
+            #sound-dai-cells = <0>;
+
+            operating-points-v2 = <&dp_opp_table>;
+            power-domains = <&rpmhpd RPMHPD_MMCX>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    mdss_dp0_in: endpoint {
+                        remote-endpoint = <&dpu_intf0_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    mdss_dp0_out: endpoint {
+                        remote-endpoint = <&usb_dp_qmpphy_dp_in>;
+                    };
+                };
+        };
+
+        dp_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-162000000 {
+                    opp-hz = /bits/ 64 <162000000>;
+                    required-opps = <&rpmhpd_opp_low_svs_d1>;
+                };
+
+                opp-270000000 {
+                    opp-hz = /bits/ 64 <270000000>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-540000000 {
+                    opp-hz = /bits/ 64 <540000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+
+                opp-810000000 {
+                    opp-hz = /bits/ 64 <810000000>;
+                    required-opps = <&rpmhpd_opp_nom>;
+                };
+            };
+        };
+
+        dsi@ae94000 {
+            compatible = "qcom,sar2130p-dsi-ctrl",
+                         "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae94000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispcc_disp_cc_mdss_byte0_clk>,
+                     <&dispcc_disp_cc_mdss_byte0_intf_clk>,
+                     <&dispcc_disp_cc_mdss_pclk0_clk>,
+                     <&dispcc_disp_cc_mdss_esc0_clk>,
+                     <&dispcc_disp_cc_mdss_ahb_clk>,
+                     <&gcc_gcc_disp_hf_axi_clk>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc_disp_cc_mdss_byte0_clk_src>,
+                              <&dispcc_disp_cc_mdss_pclk0_clk_src>;
+            assigned-clock-parents = <&mdss_dsi0_phy 0>, <&mdss_dsi0_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd RPMHPD_MMCX>;
+
+            phys = <&mdss_dsi0_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+
+                    mdss_dsi0_in: endpoint {
+                        remote-endpoint = <&dpu_intf1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+
+                    mdss_dsi0_out: endpoint {
+                    };
+                };
+            };
+
+            dsi_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-187500000 {
+                    opp-hz = /bits/ 64 <187500000>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-300000000 {
+                    opp-hz = /bits/ 64 <300000000>;
+                    required-opps = <&rpmhpd_opp_svs>;
+                };
+
+                opp-358000000 {
+                    opp-hz = /bits/ 64 <358000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+            };
+        };
+
+        mdss_dsi0_phy: phy@ae94400 {
+            compatible = "qcom,sar2130p-dsi-phy-5nm";
+            reg = <0x0ae95000 0x200>,
+                  <0x0ae95200 0x280>,
+                  <0x0ae95500 0x400>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc_disp_cc_mdss_ahb_clk>,
+                     <&rpmhcc_rpmh_cxo_clk>;
+            clock-names = "iface", "ref";
+        };
+
+        dsi@ae96000 {
+            compatible = "qcom,sar2130p-dsi-ctrl",
+                         "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae96000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <5>;
+
+            clocks = <&dispcc_disp_cc_mdss_byte1_clk>,
+                     <&dispcc_disp_cc_mdss_byte1_intf_clk>,
+                     <&dispcc_disp_cc_mdss_pclk1_clk>,
+                     <&dispcc_disp_cc_mdss_esc1_clk>,
+                     <&dispcc_disp_cc_mdss_ahb_clk>,
+                     <&gcc_gcc_disp_hf_axi_clk>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc_disp_cc_mdss_byte1_clk_src>,
+                              <&dispcc_disp_cc_mdss_pclk1_clk_src>;
+            assigned-clock-parents = <&mdss_dsi1_phy 0>, <&mdss_dsi1_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd RPMHPD_MMCX>;
+
+            phys = <&mdss_dsi1_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+
+                    mdss_dsi1_in: endpoint {
+                        remote-endpoint = <&dpu_intf2_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+
+                    mdss_dsi1_out: endpoint {
+                    };
+                };
+            };
+        };
+
+        mdss_dsi1_phy: phy@ae97000 {
+            compatible = "qcom,sar2130p-dsi-phy-5nm";
+            reg = <0x0ae97000 0x200>,
+                  <0x0ae97200 0x280>,
+                  <0x0ae97500 0x400>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc_disp_cc_mdss_ahb_clk>,
+                     <&rpmhcc_rpmh_cxo_clk>;
+            clock-names = "iface", "ref";
+        };
+    };
+...
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sc7280-dpu.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sc7280-dpu.yaml
index 6902795b4e2c..df9ec15ad6c3 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sc7280-dpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sc7280-dpu.yaml
@@ -17,6 +17,7 @@ $ref: /schemas/display/msm/dpu-common.yaml#
 properties:
   compatible:
     enum:
+      - qcom,sar2130p-dpu
       - qcom,sc7280-dpu
       - qcom,sc8280xp-dpu
       - qcom,sm8350-dpu
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml
index 163fc83c1e80..68176de854b3 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml
@@ -38,12 +38,16 @@ properties:
     maxItems: 1
 
   interconnects:
-    maxItems: 2
+    items:
+      - description: Interconnect path from the MDP0 port to the data bus
+      - description: Interconnect path from the MDP1 port to the data bus
+      - description: Interconnect path from the CPU to the reg bus
 
   interconnect-names:
     items:
       - const: mdp0-mem
       - const: mdp1-mem
+      - const: cpu-cfg
 
 patternProperties:
   "^display-controller@[0-9a-f]+$":
@@ -88,6 +92,7 @@ examples:
     #include <dt-bindings/clock/qcom,gcc-sm8350.h>
     #include <dt-bindings/clock/qcom,rpmh.h>
     #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/interconnect/qcom,icc.h>
     #include <dt-bindings/interconnect/qcom,sm8350.h>
     #include <dt-bindings/power/qcom,rpmhpd.h>
 
@@ -97,8 +102,10 @@ examples:
         reg-names = "mdss";
 
         interconnects = <&mmss_noc MASTER_MDP0 0 &mc_virt SLAVE_EBI1 0>,
-                        <&mmss_noc MASTER_MDP1 0 &mc_virt SLAVE_EBI1 0>;
-        interconnect-names = "mdp0-mem", "mdp1-mem";
+                        <&mmss_noc MASTER_MDP1 0 &mc_virt SLAVE_EBI1 0>,
+                        <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
+                         &config_noc SLAVE_DISPLAY_CFG QCOM_ICC_TAG_ACTIVE_ONLY>;
+        interconnect-names = "mdp0-mem", "mdp1-mem", "cpu-cfg";
 
         power-domains = <&dispcc MDSS_GDSC>;
         resets = <&dispcc DISP_CC_MDSS_CORE_BCR>;
diff --git a/Documentation/devicetree/bindings/display/panel/boe,td4320.yaml b/Documentation/devicetree/bindings/display/panel/boe,td4320.yaml
new file mode 100644
index 000000000000..c6bff0ece360
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/boe,td4320.yaml
@@ -0,0 +1,65 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/boe,td4320.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: BOE TD4320 MIPI-DSI panels
+
+maintainers:
+  - Barnabas Czeman <barnabas.czeman@mainlining.org>
+
+description:
+  BOE TD4320 6.3" 1080x2340 panel found in Xiaomi Redmi Note 7 smartphone.
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    items:
+      - const: boe,td4320
+
+  reg:
+    maxItems: 1
+
+  iovcc-supply:
+    description: I/O voltage rail
+
+  vsn-supply:
+    description: Negative source voltage rail
+
+  vsp-supply:
+    description: Positive source voltage rail
+
+required:
+  - compatible
+  - reg
+  - reset-gpios
+  - port
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "boe,td4320";
+            reg = <0>;
+            backlight = <&backlight>;
+            reset-gpios = <&tlmm 45 GPIO_ACTIVE_LOW>;
+
+            port {
+                panel_in: endpoint {
+                    remote-endpoint = <&dsi_out>;
+                };
+            };
+        };
+    };
+
+...
diff --git a/Documentation/devicetree/bindings/display/panel/himax,hx8279.yaml b/Documentation/devicetree/bindings/display/panel/himax,hx8279.yaml
new file mode 100644
index 000000000000..f619aea82bdf
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/himax,hx8279.yaml
@@ -0,0 +1,75 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/himax,hx8279.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Himax HX8279/HX8279-D based MIPI-DSI panels
+
+maintainers:
+  - AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
+
+description:
+  The Himax HX8279 is a 1803 channel outputs source driver with MIPI
+  TCON, which generates the horizontal and vertical control timing to
+  the source and gate drivers.
+  This DriverIC is most suitable for 1200x1920, 1080x1920, 1200x1600,
+  and 600x1024 panels and outputs full RGB888 over two or four lanes,
+  single or dual, MIPI-DSI video interface.
+
+allOf:
+  - $ref: panel-common-dual.yaml#
+
+properties:
+  compatible:
+    items:
+      - enum:
+          - aoly,sl101pm1794fog-v15
+          - startek,kd070fhfid078
+      - const: himax,hx8279
+
+  reg:
+    maxItems: 1
+
+  iovcc-supply:
+    description: I/O voltage supply
+
+  vdd-supply:
+    description: Panel power supply
+
+required:
+  - compatible
+  - reg
+  - backlight
+  - reset-gpios
+  - iovcc-supply
+  - vdd-supply
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "startek,kd070fhfid078", "himax,hx8279";
+            reg = <0>;
+            backlight = <&backlight>;
+            enable-gpios = <&pio 25 GPIO_ACTIVE_HIGH>;
+            reset-gpios = <&pio 45 GPIO_ACTIVE_HIGH>;
+            iovcc-supply = <&vreg_lcm_vio>;
+            vdd-supply = <&vreg_lcm_vdd>;
+
+            port {
+                panel_in: endpoint {
+                    remote-endpoint = <&dsi_out>;
+                };
+            };
+        };
+    };
+
+...
diff --git a/Documentation/devicetree/bindings/display/panel/leadtek,ltk050h3146w.yaml b/Documentation/devicetree/bindings/display/panel/leadtek,ltk050h3146w.yaml
index e2a2dd4ef5fa..5fcea62fd58f 100644
--- a/Documentation/devicetree/bindings/display/panel/leadtek,ltk050h3146w.yaml
+++ b/Documentation/devicetree/bindings/display/panel/leadtek,ltk050h3146w.yaml
@@ -23,6 +23,7 @@ properties:
     maxItems: 1
 
   backlight: true
+  port: true
   reset-gpios: true
   iovcc-supply:
     description: regulator that supplies the iovcc voltage
diff --git a/Documentation/devicetree/bindings/display/panel/leadtek,ltk500hd1829.yaml b/Documentation/devicetree/bindings/display/panel/leadtek,ltk500hd1829.yaml
index af9e0ea0e72f..b0e2c82232d3 100644
--- a/Documentation/devicetree/bindings/display/panel/leadtek,ltk500hd1829.yaml
+++ b/Documentation/devicetree/bindings/display/panel/leadtek,ltk500hd1829.yaml
@@ -22,6 +22,7 @@ properties:
     maxItems: 1
 
   backlight: true
+  port: true
   reset-gpios: true
   iovcc-supply:
     description: regulator that supplies the iovcc voltage
diff --git a/Documentation/devicetree/bindings/display/panel/novatek,nt37801.yaml b/Documentation/devicetree/bindings/display/panel/novatek,nt37801.yaml
new file mode 100644
index 000000000000..1b38c1d0af68
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/novatek,nt37801.yaml
@@ -0,0 +1,69 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/novatek,nt37801.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Novatek NT37801 AMOLED DSI Panel
+
+maintainers:
+  - Krzysztof Kozlowski <krzk@kernel.org>
+
+description:
+  Naming is inconclusive and different sources claim this is either Novatek
+  NT37801 or NT37810 AMOLED DSI Panel.
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    const: novatek,nt37801
+
+  reg:
+    maxItems: 1
+    description: DSI virtual channel
+
+  vci-supply: true
+  vdd-supply: true
+  vddio-supply: true
+  port: true
+  reset-gpios: true
+
+required:
+  - compatible
+  - reg
+  - vci-supply
+  - vdd-supply
+  - vddio-supply
+  - port
+  - reset-gpios
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "novatek,nt37801";
+            reg = <0>;
+
+            vci-supply = <&vreg_l13b_3p0>;
+            vdd-supply = <&vreg_l11b_1p2>;
+            vddio-supply = <&vreg_l12b_1p8>;
+
+            reset-gpios = <&tlmm 98 GPIO_ACTIVE_LOW>;
+
+            port {
+                endpoint {
+                    remote-endpoint = <&dsi0_out>;
+                };
+            };
+        };
+    };
+...
diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index b0de4fd6f3d4..5542c9229d54 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -226,6 +226,8 @@ properties:
       - netron-dy,e231732
         # Newhaven Display International 480 x 272 TFT LCD panel
       - newhaven,nhd-4.3-480272ef-atxl
+        # NLT Technologies, Ltd. 15.6" WXGA (1366×768) LVDS TFT LCD panel
+      - nlt,nl13676bc25-03f
         # New Vision Display 7.0" 800 RGB x 480 TFT LCD panel
       - nvd,9128
         # OKAYA Electric America, Inc. RS800480T-7X0GP 7" WVGA LCD panel
@@ -246,6 +248,8 @@ properties:
       - osddisplays,osd070t1718-19ts
         # One Stop Displays OSD101T2045-53TS 10.1" 1920x1200 panel
       - osddisplays,osd101t2045-53ts
+        # POWERTIP PH128800T004-ZZA01 10.1" WXGA TFT LCD panel
+      - powertip,ph128800t004-zza01
         # POWERTIP PH128800T006-ZHC01 10.1" WXGA TFT LCD panel
       - powertip,ph128800t006-zhc01
         # POWERTIP PH800480T013-IDF2 7.0" WVGA TFT LCD panel
@@ -284,6 +288,8 @@ properties:
       - startek,kd070wvfpa
         # Team Source Display Technology TST043015CMHX 4.3" WQVGA TFT LCD panel
       - team-source-display,tst043015cmhx
+        # Tianma Micro-electronics P0700WXF1MBAA 7.0" WXGA (1280x800) LVDS TFT LCD panel
+      - tianma,p0700wxf1mbaa
         # Tianma Micro-electronics TM070JDHG30 7.0" WXGA TFT LCD panel
       - tianma,tm070jdhg30
         # Tianma Micro-electronics TM070JDHG34-00 7.0" WXGA (1280x800) LVDS TFT LCD panel
diff --git a/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml b/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
index 684c2896d238..31f0c0f038e4 100644
--- a/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
+++ b/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
@@ -19,6 +19,8 @@ properties:
       - const: samsung,atna33xc20
       - items:
           - enum:
+              # Samsung 14" WQXGA+ (2880×1800 pixels) eDP AMOLED panel
+              - samsung,atna40yk20
               # Samsung 14.5" WQXGA+ (2880x1800 pixels) eDP AMOLED panel
               - samsung,atna45af01
               # Samsung 14.5" 3K (2944x1840 pixels) eDP AMOLED panel
diff --git a/Documentation/devicetree/bindings/display/panel/truly,nt35597-2K-display.yaml b/Documentation/devicetree/bindings/display/panel/truly,nt35597-2K-display.yaml
new file mode 100644
index 000000000000..36be09c900f2
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/truly,nt35597-2K-display.yaml
@@ -0,0 +1,97 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/truly,nt35597-2K-display.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Truly NT35597 DSI 2K display
+
+maintainers:
+  - Neil Armstrong <neil.armstrong@linaro.org>
+
+description: |
+  Truly NT35597 DSI 2K display is used on the Qualcomm SDM845 MTP board.
+
+allOf:
+  - $ref: panel-common-dual.yaml#
+
+properties:
+  compatible:
+    const: truly,nt35597-2K-display
+
+  reg:
+    maxItems: 1
+
+  vdda-supply:
+    description: regulator that provides the supply voltage Power IC supply
+
+  vdispp-supply:
+    description: regulator that provides the supply voltage for positive LCD bias
+
+  vdispn-supply:
+    description: regulator that provides the supply voltage for negative LCD bias
+
+  reset-gpios: true
+
+  mode-gpios:
+    description:
+      Gpio for choosing the mode of the display for single DSI or Dual DSI.
+      This should be low for dual DSI and high for single DSI mode.
+
+  ports:
+    required:
+      - port@0
+      - port@1
+
+required:
+  - compatible
+  - reg
+  - vdda-supply
+  - reset-gpios
+  - mode-gpios
+  - ports
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "truly,nt35597-2K-display";
+            reg = <0>;
+
+            vdda-supply = <&pm8998_l14>;
+            vdispp-supply = <&lab_regulator>;
+            vdispn-supply = <&ibb_regulator>;
+
+            reset-gpios = <&tlmm 6 GPIO_ACTIVE_LOW>;
+            mode-gpios = <&tlmm 52 GPIO_ACTIVE_HIGH>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+
+                    panel0_in: endpoint {
+                        remote-endpoint = <&dsi0_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+
+                    panel1_in: endpoint {
+                        remote-endpoint = <&dsi1_out>;
+                    };
+                };
+            };
+        };
+    };
+...
diff --git a/Documentation/devicetree/bindings/display/panel/visionox,g2647fb105.yaml b/Documentation/devicetree/bindings/display/panel/visionox,g2647fb105.yaml
new file mode 100644
index 000000000000..49dcd9b8f670
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/visionox,g2647fb105.yaml
@@ -0,0 +1,79 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/visionox,g2647fb105.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Visionox G2647FB105 6.47" 1080x2340 MIPI-DSI Panel
+
+maintainers:
+  - Alexander Baransky <sanyapilot496@gmail.com>
+
+description:
+  The Visionox G2647FB105 is a 6.47 inch 1080x2340 MIPI-DSI CMD mode OLED panel.
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    const: visionox,g2647fb105
+
+  reg:
+    maxItems: 1
+
+  vdd3p3-supply:
+    description: 3.3V source voltage rail
+
+  vddio-supply:
+    description: I/O source voltage rail
+
+  vsn-supply:
+    description: Negative source voltage rail
+
+  vsp-supply:
+    description: Positive source voltage rail
+
+  reset-gpios: true
+  port: true
+
+required:
+  - compatible
+  - reg
+  - vdd3p3-supply
+  - vddio-supply
+  - vsn-supply
+  - vsp-supply
+  - reset-gpios
+  - port
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "visionox,g2647fb105";
+            reg = <0>;
+
+            vdd3p3-supply = <&vreg_l7c_3p0>;
+            vddio-supply = <&vreg_l13a_1p8>;
+            vsn-supply = <&vreg_ibb>;
+            vsp-supply = <&vreg_lab>;
+
+            reset-gpios = <&pm6150l_gpios 9 GPIO_ACTIVE_LOW>;
+
+            port {
+                panel_in: endpoint {
+                    remote-endpoint = <&mdss_dsi0_out>;
+                };
+            };
+        };
+    };
+
+...
diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,analogix-dp.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,analogix-dp.yaml
index 60dedf9b2be7..d99b23b88cc5 100644
--- a/Documentation/devicetree/bindings/display/rockchip/rockchip,analogix-dp.yaml
+++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,analogix-dp.yaml
@@ -15,6 +15,7 @@ properties:
     enum:
       - rockchip,rk3288-dp
       - rockchip,rk3399-edp
+      - rockchip,rk3588-edp
 
   clocks:
     minItems: 2
@@ -31,16 +32,23 @@ properties:
     maxItems: 1
 
   resets:
-    maxItems: 1
+    minItems: 1
+    maxItems: 2
 
   reset-names:
-    const: dp
+    minItems: 1
+    items:
+      - const: dp
+      - const: apb
 
   rockchip,grf:
     $ref: /schemas/types.yaml#/definitions/phandle
     description:
       This SoC makes use of GRF regs.
 
+  aux-bus:
+    $ref: /schemas/display/dp-aux-bus.yaml#
+
 required:
   - compatible
   - clocks
@@ -52,6 +60,19 @@ required:
 allOf:
   - $ref: /schemas/display/bridge/analogix,dp.yaml#
 
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - rockchip,rk3588-edp
+    then:
+      properties:
+        resets:
+          minItems: 2
+        reset-names:
+          minItems: 2
+
 unevaluatedProperties: false
 
 examples:
diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,inno-hdmi.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,inno-hdmi.yaml
index 5b87b0f1963e..290376bec079 100644
--- a/Documentation/devicetree/bindings/display/rockchip/rockchip,inno-hdmi.yaml
+++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,inno-hdmi.yaml
@@ -23,13 +23,11 @@ properties:
     maxItems: 1
 
   clocks:
-    minItems: 1
     items:
       - description: The HDMI controller main clock
       - description: The HDMI PHY reference clock
 
   clock-names:
-    minItems: 1
     items:
       - const: pclk
       - const: ref
@@ -58,6 +56,12 @@ properties:
       - port@0
       - port@1
 
+  rockchip,grf:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description:
+      Phandle to GRF used for control the polarity of hsync/vsync of rk3036
+      HDMI.
+
 required:
   - compatible
   - reg
@@ -77,6 +81,8 @@ allOf:
             const: rockchip,rk3036-inno-hdmi
 
     then:
+      required:
+        - rockchip,grf
       properties:
         power-domains: false
 
@@ -87,11 +93,6 @@ allOf:
             const: rockchip,rk3128-inno-hdmi
 
     then:
-      properties:
-        clocks:
-          minItems: 2
-        clock-names:
-          minItems: 2
       required:
         - power-domains
 
@@ -106,10 +107,11 @@ examples:
       compatible = "rockchip,rk3036-inno-hdmi";
       reg = <0x20034000 0x4000>;
       interrupts = <GIC_SPI 45 IRQ_TYPE_LEVEL_HIGH>;
-      clocks = <&cru  PCLK_HDMI>;
-      clock-names = "pclk";
+      clocks = <&cru PCLK_HDMI>, <&cru SCLK_LCDC>;
+      clock-names = "pclk", "ref";
       pinctrl-names = "default";
       pinctrl-0 = <&hdmi_ctl>;
+      rockchip,grf = <&grf>;
       #sound-dai-cells = <0>;
 
       ports {
diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip-vop.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip-vop.yaml
index b339b7e708c6..8b5f58103dda 100644
--- a/Documentation/devicetree/bindings/display/rockchip/rockchip-vop.yaml
+++ b/Documentation/devicetree/bindings/display/rockchip/rockchip-vop.yaml
@@ -73,12 +73,6 @@ properties:
   port:
     $ref: /schemas/graph.yaml#/properties/port
 
-  assigned-clocks:
-    maxItems: 2
-
-  assigned-clock-rates:
-    maxItems: 2
-
   iommus:
     maxItems: 1
 
diff --git a/Documentation/devicetree/bindings/display/sitronix,st7571.yaml b/Documentation/devicetree/bindings/display/sitronix,st7571.yaml
new file mode 100644
index 000000000000..4fea782fccd7
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/sitronix,st7571.yaml
@@ -0,0 +1,73 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/sitronix,st7571.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Sitronix ST7571 Display Controller
+
+maintainers:
+  - Marcus Folkesson <marcus.folkesson@gmail.com>
+
+description:
+  Sitronix ST7571 is a driver and controller for 4-level gray
+  scale and monochrome dot matrix LCD panels.
+
+allOf:
+  - $ref: panel/panel-common.yaml#
+
+properties:
+  compatible:
+    const: sitronix,st7571
+
+  reg:
+    maxItems: 1
+
+  sitronix,grayscale:
+    type: boolean
+    description:
+      Display supports 4-level grayscale.
+
+  reset-gpios: true
+  width-mm: true
+  height-mm: true
+  panel-timing: true
+
+required:
+  - compatible
+  - reg
+  - reset-gpios
+  - width-mm
+  - height-mm
+  - panel-timing
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    i2c {
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      display@3f {
+        compatible = "sitronix,st7571";
+        reg = <0x3f>;
+        reset-gpios = <&gpio0 3 GPIO_ACTIVE_LOW>;
+        width-mm = <37>;
+        height-mm = <27>;
+
+        panel-timing {
+          hactive = <128>;
+          vactive = <96>;
+          hback-porch = <0>;
+          vback-porch = <0>;
+          clock-frequency = <0>;
+          hfront-porch = <0>;
+          hsync-len = <0>;
+          vfront-porch = <0>;
+          vsync-len = <0>;
+        };
+      };
+    };
diff --git a/Documentation/devicetree/bindings/display/truly,nt35597.txt b/Documentation/devicetree/bindings/display/truly,nt35597.txt
deleted file mode 100644
index f39c77ee36ea..000000000000
--- a/Documentation/devicetree/bindings/display/truly,nt35597.txt
+++ /dev/null
@@ -1,59 +0,0 @@
-Truly model NT35597 DSI display driver
-
-The Truly NT35597 is a generic display driver, currently only configured
-for use in the 2K display on the Qualcomm SDM845 MTP board.
-
-Required properties:
-- compatible: should be "truly,nt35597-2K-display"
-- vdda-supply: phandle of the regulator that provides the supply voltage
-  Power IC supply
-- vdispp-supply: phandle of the regulator that provides the supply voltage
-  for positive LCD bias
-- vdispn-supply: phandle of the regulator that provides the supply voltage
-  for negative LCD bias
-- reset-gpios: phandle of gpio for reset line
-  This should be 8mA, gpio can be configured using mux, pinctrl, pinctrl-names
-  (active low)
-- mode-gpios: phandle of the gpio for choosing the mode of the display
-  for single DSI or Dual DSI
-  This should be low for dual DSI and high for single DSI mode
-- ports: This device has two video ports driven by two DSIs. Their connections
-  are modeled using the OF graph bindings specified in
-  Documentation/devicetree/bindings/graph.txt.
-  - port@0: DSI input port driven by master DSI
-  - port@1: DSI input port driven by secondary DSI
-
-Example:
-
-	dsi@ae94000 {
-		panel@0 {
-			compatible = "truly,nt35597-2K-display";
-			reg = <0>;
-			vdda-supply = <&pm8998_l14>;
-			vdispp-supply = <&lab_regulator>;
-			vdispn-supply = <&ibb_regulator>;
-			pinctrl-names = "default", "suspend";
-			pinctrl-0 = <&dpu_dsi_active>;
-			pinctrl-1 = <&dpu_dsi_suspend>;
-
-			reset-gpios = <&tlmm 6 GPIO_ACTIVE_LOW>;
-			mode-gpios = <&tlmm 52 GPIO_ACTIVE_HIGH>;
-			ports {
-				#address-cells = <1>;
-				#size-cells = <0>;
-				port@0 {
-					reg = <0>;
-					panel0_in: endpoint {
-						remote-endpoint = <&dsi0_out>;
-					};
-				};
-
-				port@1 {
-					reg = <1>;
-					panel1_in: endpoint {
-						remote-endpoint = <&dsi1_out>;
-					};
-				};
-			};
-		};
-	};
diff --git a/Documentation/devicetree/bindings/gpio/xlnx,zynqmp-gpio-modepin.yaml b/Documentation/devicetree/bindings/gpio/xlnx,zynqmp-gpio-modepin.yaml
index bb93baa88879..e13e9d6dd148 100644
--- a/Documentation/devicetree/bindings/gpio/xlnx,zynqmp-gpio-modepin.yaml
+++ b/Documentation/devicetree/bindings/gpio/xlnx,zynqmp-gpio-modepin.yaml
@@ -12,7 +12,6 @@ description:
   PS_MODE). Every pin can be configured as input/output.
 
 maintainers:
-  - Mubin Sayyed <mubin.sayyed@amd.com>
   - Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
 
 properties:
diff --git a/Documentation/devicetree/bindings/gpu/brcm,bcm-v3d.yaml b/Documentation/devicetree/bindings/gpu/brcm,bcm-v3d.yaml
index dc078ceeca9a..43c6d2d72456 100644
--- a/Documentation/devicetree/bindings/gpu/brcm,bcm-v3d.yaml
+++ b/Documentation/devicetree/bindings/gpu/brcm,bcm-v3d.yaml
@@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Broadcom V3D GPU
 
 maintainers:
-  - Eric Anholt <eric@anholt.net>
+  - Maíra Canal <mcanal@igalia.com>
   - Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
 
 properties:
@@ -22,20 +22,12 @@ properties:
       - brcm,7278-v3d
 
   reg:
-    items:
-      - description: hub register (required)
-      - description: core0 register (required)
-      - description: GCA cache controller register (if GCA controller present)
-      - description: bridge register (if no external reset controller)
     minItems: 2
+    maxItems: 4
 
   reg-names:
-    items:
-      - const: hub
-      - const: core0
-      - enum: [ bridge, gca ]
-      - enum: [ bridge, gca ]
     minItems: 2
+    maxItems: 4
 
   interrupts:
     items:
@@ -58,6 +50,76 @@ required:
   - reg-names
   - interrupts
 
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: brcm,2711-v3d
+    then:
+      properties:
+        reg:
+          items:
+            - description: hub register
+            - description: core0 register
+        reg-names:
+          items:
+            - const: hub
+            - const: core0
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: brcm,2712-v3d
+    then:
+      properties:
+        reg:
+          items:
+            - description: hub register
+            - description: core0 register
+            - description: SMS state manager register
+        reg-names:
+          items:
+            - const: hub
+            - const: core0
+            - const: sms
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: brcm,7268-v3d
+    then:
+      properties:
+        reg:
+          items:
+            - description: hub register
+            - description: core0 register
+            - description: GCA cache controller register
+            - description: bridge register
+        reg-names:
+          items:
+            - const: hub
+            - const: core0
+            - const: gca
+            - const: bridge
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: brcm,7278-v3d
+    then:
+      properties:
+        reg:
+          items:
+            - description: hub register
+            - description: core0 register
+            - description: bridge register
+        reg-names:
+          items:
+            - const: hub
+            - const: core0
+            - const: bridge
+
 additionalProperties: false
 
 examples:
@@ -66,9 +128,9 @@ examples:
       compatible = "brcm,7268-v3d";
       reg = <0xf1200000 0x4000>,
             <0xf1208000 0x4000>,
-            <0xf1204000 0x100>,
-            <0xf1204100 0x100>;
-      reg-names = "hub", "core0", "bridge", "gca";
+            <0xf1204100 0x100>,
+            <0xf1204000 0x100>;
+      reg-names = "hub", "core0", "gca", "bridge";
       interrupts = <0 78 4>,
                    <0 77 4>;
     };
diff --git a/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml b/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
index 256e252f8087..4450e2e73b3c 100644
--- a/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
+++ b/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
@@ -12,10 +12,28 @@ maintainers:
 
 properties:
   compatible:
-    items:
-      - enum:
-          - ti,am62-gpu
-      - const: img,img-axe # IMG AXE GPU model/revision is fully discoverable
+    oneOf:
+      - items:
+          - enum:
+              - ti,am62-gpu
+          - const: img,img-axe-1-16m
+          # This deprecated element must be kept around to allow old kernels to
+          # work with newer dts.
+          - const: img,img-axe
+          - const: img,img-rogue
+      - items:
+          - enum:
+              - ti,j721s2-gpu
+          - const: img,img-bxs-4-64
+          - const: img,img-rogue
+
+      # This legacy combination of compatible strings was introduced early on
+      # before the more specific GPU identifiers were used.
+      - items:
+          - enum:
+              - ti,am62-gpu
+          - const: img,img-axe
+        deprecated: true
 
   reg:
     maxItems: 1
@@ -35,6 +53,18 @@ properties:
     maxItems: 1
 
   power-domains:
+    minItems: 1
+    maxItems: 2
+
+  power-domain-names:
+    items:
+      - const: a
+      - const: b
+    minItems: 1
+
+  dma-coherent: true
+
+  resets:
     maxItems: 1
 
 required:
@@ -47,11 +77,49 @@ required:
 additionalProperties: false
 
 allOf:
+  # Constraints added alongside the new compatible strings that would otherwise
+  # create an ABI break.
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: img,img-rogue
+    then:
+      required:
+        - power-domains
+        - power-domain-names
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: img,img-axe-1-16m
+    then:
+      properties:
+        power-domains:
+          maxItems: 1
+        power-domain-names:
+          maxItems: 1
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: img,img-bxs-4-64
+    then:
+      properties:
+        power-domains:
+          minItems: 2
+        power-domain-names:
+          minItems: 2
+
   - if:
       properties:
         compatible:
           contains:
-            const: ti,am62-gpu
+            enum:
+              - ti,am62-gpu
+              - ti,j721s2-gpu
     then:
       properties:
         clocks:
@@ -64,10 +132,12 @@ examples:
     #include <dt-bindings/soc/ti,sci_pm_domain.h>
 
     gpu@fd00000 {
-        compatible = "ti,am62-gpu", "img,img-axe";
+        compatible = "ti,am62-gpu", "img,img-axe-1-16m", "img,img-axe",
+                     "img,img-rogue";
         reg = <0x0fd00000 0x20000>;
         clocks = <&k3_clks 187 0>;
         clock-names = "core";
         interrupts = <GIC_SPI 86 IRQ_TYPE_LEVEL_HIGH>;
         power-domains = <&k3_pds 187 TI_SCI_PD_EXCLUSIVE>;
+        power-domain-names = "a";
     };
diff --git a/Documentation/devicetree/bindings/interrupt-controller/fsl,irqsteer.yaml b/Documentation/devicetree/bindings/interrupt-controller/fsl,irqsteer.yaml
index 6076ddf56bb5..c49688be1058 100644
--- a/Documentation/devicetree/bindings/interrupt-controller/fsl,irqsteer.yaml
+++ b/Documentation/devicetree/bindings/interrupt-controller/fsl,irqsteer.yaml
@@ -19,6 +19,7 @@ properties:
               - fsl,imx8mp-irqsteer
               - fsl,imx8qm-irqsteer
               - fsl,imx8qxp-irqsteer
+              - fsl,imx94-irqsteer
           - const: fsl,imx-irqsteer
 
   reg:
diff --git a/Documentation/devicetree/bindings/nvmem/layouts/fixed-cell.yaml b/Documentation/devicetree/bindings/nvmem/layouts/fixed-cell.yaml
index 8b3826243ddd..38e3ad50ff4f 100644
--- a/Documentation/devicetree/bindings/nvmem/layouts/fixed-cell.yaml
+++ b/Documentation/devicetree/bindings/nvmem/layouts/fixed-cell.yaml
@@ -27,7 +27,7 @@ properties:
     $ref: /schemas/types.yaml#/definitions/uint32-array
     items:
       - minimum: 0
-        maximum: 7
+        maximum: 31
         description:
           Offset in bit within the address range specified by reg.
       - minimum: 1
diff --git a/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml b/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml
index 39c209249c9c..3f6dc6a3a9f1 100644
--- a/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml
+++ b/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml
@@ -19,6 +19,7 @@ properties:
       - enum:
           - qcom,apq8064-qfprom
           - qcom,apq8084-qfprom
+          - qcom,ipq5018-qfprom
           - qcom,ipq5332-qfprom
           - qcom,ipq5424-qfprom
           - qcom,ipq6018-qfprom
@@ -28,6 +29,8 @@ properties:
           - qcom,msm8226-qfprom
           - qcom,msm8916-qfprom
           - qcom,msm8917-qfprom
+          - qcom,msm8937-qfprom
+          - qcom,msm8960-qfprom
           - qcom,msm8974-qfprom
           - qcom,msm8976-qfprom
           - qcom,msm8996-qfprom
@@ -51,6 +54,7 @@ properties:
           - qcom,sm8450-qfprom
           - qcom,sm8550-qfprom
           - qcom,sm8650-qfprom
+          - qcom,x1e80100-qfprom
       - const: qcom,qfprom
 
   reg:
diff --git a/Documentation/devicetree/bindings/nvmem/rockchip,otp.yaml b/Documentation/devicetree/bindings/nvmem/rockchip,otp.yaml
index a44d44b32809..dc89020b0950 100644
--- a/Documentation/devicetree/bindings/nvmem/rockchip,otp.yaml
+++ b/Documentation/devicetree/bindings/nvmem/rockchip,otp.yaml
@@ -14,6 +14,7 @@ properties:
     enum:
       - rockchip,px30-otp
       - rockchip,rk3308-otp
+      - rockchip,rk3576-otp
       - rockchip,rk3588-otp
 
   reg:
@@ -62,6 +63,8 @@ allOf:
       properties:
         clocks:
           maxItems: 3
+        clock-names:
+          maxItems: 3
         resets:
           maxItems: 1
         reset-names:
@@ -73,11 +76,33 @@ allOf:
         compatible:
           contains:
             enum:
+              - rockchip,rk3576-otp
+    then:
+      properties:
+        clocks:
+          maxItems: 3
+        clock-names:
+          maxItems: 3
+        resets:
+          minItems: 2
+          maxItems: 2
+        reset-names:
+          items:
+            - const: otp
+            - const: apb
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
               - rockchip,rk3588-otp
     then:
       properties:
         clocks:
           minItems: 4
+        clock-names:
+          minItems: 4
         resets:
           minItems: 3
         reset-names:
diff --git a/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml b/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml
new file mode 100644
index 000000000000..a27ba7b663d4
--- /dev/null
+++ b/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml
@@ -0,0 +1,96 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/opp/opp-v2-qcom-adreno.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Adreno compatible OPP supply
+
+description:
+  Adreno GPUs present in Qualcomm's Snapdragon chipsets uses an OPP specific
+  ACD related information tailored for the specific chipset. This binding
+  provides the information needed to describe such a hardware value.
+
+maintainers:
+  - Rob Clark <robdclark@gmail.com>
+
+allOf:
+  - $ref: opp-v2-base.yaml#
+
+properties:
+  compatible:
+    contains:
+      const: operating-points-v2-adreno
+
+patternProperties:
+  '^opp-[0-9]+$':
+    type: object
+    additionalProperties: false
+
+    properties:
+      opp-hz: true
+
+      opp-level: true
+
+      opp-peak-kBps: true
+
+      opp-supported-hw: true
+
+      qcom,opp-acd-level:
+        description: |
+          A positive value representing the ACD (Adaptive Clock Distribution,
+          a fancy name for clk throttling during voltage droop) level associated
+          with this OPP node. This value is shared to a co-processor inside GPU
+          (called Graphics Management Unit a.k.a GMU) during wake up. It may not
+          be present for some OPPs and GMU will disable ACD while transitioning
+          to that OPP. This value encodes a voltage threshold, delay cycles &
+          calibration margins which are identified by characterization of the
+          SoC. So, it doesn't have any unit. This data is passed to GMU firmware
+          via 'HFI_H2F_MSG_ACD' packet.
+        $ref: /schemas/types.yaml#/definitions/uint32
+
+    required:
+      - opp-hz
+      - opp-level
+
+required:
+  - compatible
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    gpu_opp_table: opp-table {
+        compatible = "operating-points-v2-adreno", "operating-points-v2";
+
+        opp-687000000 {
+            opp-hz = /bits/ 64 <687000000>;
+            opp-level = <RPMH_REGULATOR_LEVEL_SVS_L1>;
+            opp-peak-kBps = <8171875>;
+            qcom,opp-acd-level = <0x882e5ffd>;
+        };
+
+        opp-550000000 {
+            opp-hz = /bits/ 64 <550000000>;
+            opp-level = <RPMH_REGULATOR_LEVEL_SVS>;
+            opp-peak-kBps = <6074219>;
+            qcom,opp-acd-level = <0xc0285ffd>;
+        };
+
+        opp-390000000 {
+            opp-hz = /bits/ 64 <390000000>;
+            opp-level = <RPMH_REGULATOR_LEVEL_LOW_SVS>;
+            opp-peak-kBps = <3000000>;
+            qcom,opp-acd-level = <0xc0285ffd>;
+        };
+
+        opp-300000000 {
+            opp-hz = /bits/ 64 <300000000>;
+            opp-level = <RPMH_REGULATOR_LEVEL_LOW_SVS_D1>;
+            opp-peak-kBps = <2136719>;
+            /* Intentionally left out qcom,opp-acd-level property here */
+        };
+
+    };
diff --git a/Documentation/devicetree/bindings/pwm/renesas,tpu-pwm.yaml b/Documentation/devicetree/bindings/pwm/renesas,tpu-pwm.yaml
index a4dfa09344dd..f85ee5d20ccb 100644
--- a/Documentation/devicetree/bindings/pwm/renesas,tpu-pwm.yaml
+++ b/Documentation/devicetree/bindings/pwm/renesas,tpu-pwm.yaml
@@ -9,15 +9,6 @@ title: Renesas R-Car Timer Pulse Unit PWM Controller
 maintainers:
   - Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
 
-select:
-  properties:
-    compatible:
-      contains:
-        const: renesas,tpu
-  required:
-    - compatible
-    - '#pwm-cells'
-
 properties:
   compatible:
     items:
diff --git a/Documentation/devicetree/bindings/reset/xlnx,zynqmp-reset.yaml b/Documentation/devicetree/bindings/reset/xlnx,zynqmp-reset.yaml
index 1f1b42dde94d..1db85fc9966f 100644
--- a/Documentation/devicetree/bindings/reset/xlnx,zynqmp-reset.yaml
+++ b/Documentation/devicetree/bindings/reset/xlnx,zynqmp-reset.yaml
@@ -7,7 +7,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Zynq UltraScale+ MPSoC and Versal reset
 
 maintainers:
-  - Mubin Sayyed <mubin.sayyed@amd.com>
   - Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
 
 description: |
diff --git a/Documentation/devicetree/bindings/soc/fsl/fsl,ls1028a-reset.yaml b/Documentation/devicetree/bindings/soc/fsl/fsl,ls1028a-reset.yaml
index 31295be91013..234089b5954d 100644
--- a/Documentation/devicetree/bindings/soc/fsl/fsl,ls1028a-reset.yaml
+++ b/Documentation/devicetree/bindings/soc/fsl/fsl,ls1028a-reset.yaml
@@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Freescale Layerscape Reset Registers Module
 
 maintainers:
-  - Frank Li
+  - Frank Li <Frank.Li@nxp.com>
 
 description:
   Reset Module includes chip reset, service processor control and Reset Control
diff --git a/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml b/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml
index bccd00a1ddd0..53d00ca643b3 100644
--- a/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml
+++ b/Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml
@@ -56,19 +56,18 @@ properties:
         enum:
           - snps,dw-apb-ssi
           - snps,dwc-ssi-1.01a
-      - description: Microsemi Ocelot/Jaguar2 SoC SPI Controller
-        items:
-          - enum:
-              - mscc,ocelot-spi
-              - mscc,jaguar2-spi
-          - const: snps,dw-apb-ssi
       - description: Microchip Sparx5 SoC SPI Controller
         const: microchip,sparx5-spi
       - description: Amazon Alpine SPI Controller
         const: amazon,alpine-dw-apb-ssi
-      - description: Renesas RZ/N1 SPI Controller
+      - description: Vendor controllers which use snps,dw-apb-ssi as fallback
         items:
-          - const: renesas,rzn1-spi
+          - enum:
+              - mscc,ocelot-spi
+              - mscc,jaguar2-spi
+              - renesas,rzn1-spi
+              - sophgo,sg2042-spi
+              - thead,th1520-spi
           - const: snps,dw-apb-ssi
       - description: Intel Keem Bay SPI Controller
         const: intel,keembay-ssi
@@ -88,10 +87,6 @@ properties:
               - renesas,r9a06g032-spi # RZ/N1D
               - renesas,r9a06g033-spi # RZ/N1S
           - const: renesas,rzn1-spi   # RZ/N1
-      - description: T-HEAD TH1520 SoC SPI Controller
-        items:
-          - const: thead,th1520-spi
-          - const: snps,dw-apb-ssi
 
   reg:
     minItems: 1
diff --git a/Documentation/devicetree/bindings/timer/nxp,sysctr-timer.yaml b/Documentation/devicetree/bindings/timer/nxp,sysctr-timer.yaml
index 891cca009528..6b80b060672e 100644
--- a/Documentation/devicetree/bindings/timer/nxp,sysctr-timer.yaml
+++ b/Documentation/devicetree/bindings/timer/nxp,sysctr-timer.yaml
@@ -18,9 +18,14 @@ description: |
 
 properties:
   compatible:
-    enum:
-      - nxp,imx95-sysctr-timer
-      - nxp,sysctr-timer
+    oneOf:
+      - enum:
+          - nxp,imx95-sysctr-timer
+          - nxp,sysctr-timer
+      - items:
+          - enum:
+              - nxp,imx94-sysctr-timer
+          - const: nxp,imx95-sysctr-timer
 
   reg:
     maxItems: 1
diff --git a/Documentation/devicetree/bindings/timer/renesas,tpu.yaml b/Documentation/devicetree/bindings/timer/renesas,tpu.yaml
deleted file mode 100644
index 7a473b302775..000000000000
--- a/Documentation/devicetree/bindings/timer/renesas,tpu.yaml
+++ /dev/null
@@ -1,56 +0,0 @@
-# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
-%YAML 1.2
----
-$id: http://devicetree.org/schemas/timer/renesas,tpu.yaml#
-$schema: http://devicetree.org/meta-schemas/core.yaml#
-
-title: Renesas H8/300 Timer Pulse Unit
-
-maintainers:
-  - Yoshinori Sato <ysato@users.sourceforge.jp>
-
-description:
-  The TPU is a 16bit timer/counter with configurable clock inputs and
-  programmable compare match.
-  This implementation supports only cascade mode.
-
-select:
-  properties:
-    compatible:
-      contains:
-        const: renesas,tpu
-    '#pwm-cells': false
-  required:
-    - compatible
-
-properties:
-  compatible:
-    const: renesas,tpu
-
-  reg:
-    items:
-      - description: First channel
-      - description: Second channel
-
-  clocks:
-    maxItems: 1
-
-  clock-names:
-    const: fck
-
-required:
-  - compatible
-  - reg
-  - clocks
-  - clock-names
-
-additionalProperties: false
-
-examples:
-  - |
-    tpu: tpu@ffffe0 {
-        compatible = "renesas,tpu";
-        reg = <0xffffe0 16>, <0xfffff0 12>;
-        clocks = <&pclk>;
-        clock-names = "fck";
-    };
diff --git a/Documentation/devicetree/bindings/usb/dwc3-xilinx.yaml b/Documentation/devicetree/bindings/usb/dwc3-xilinx.yaml
index b5843f4d17d8..379dacacb526 100644
--- a/Documentation/devicetree/bindings/usb/dwc3-xilinx.yaml
+++ b/Documentation/devicetree/bindings/usb/dwc3-xilinx.yaml
@@ -7,7 +7,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Xilinx SuperSpeed DWC3 USB SoC controller
 
 maintainers:
-  - Mubin Sayyed <mubin.sayyed@amd.com>
   - Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
 
 properties:
diff --git a/Documentation/devicetree/bindings/usb/microchip,usb5744.yaml b/Documentation/devicetree/bindings/usb/microchip,usb5744.yaml
index e2a72deae776..c68c04da3399 100644
--- a/Documentation/devicetree/bindings/usb/microchip,usb5744.yaml
+++ b/Documentation/devicetree/bindings/usb/microchip,usb5744.yaml
@@ -17,7 +17,6 @@ description:
 
 maintainers:
   - Michal Simek <michal.simek@amd.com>
-  - Mubin Sayyed <mubin.sayyed@amd.com>
   - Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
 
 properties:
diff --git a/Documentation/devicetree/bindings/usb/xlnx,usb2.yaml b/Documentation/devicetree/bindings/usb/xlnx,usb2.yaml
index a7f75fe36665..f295aa9d9ee7 100644
--- a/Documentation/devicetree/bindings/usb/xlnx,usb2.yaml
+++ b/Documentation/devicetree/bindings/usb/xlnx,usb2.yaml
@@ -7,7 +7,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 title: Xilinx udc controller
 
 maintainers:
-  - Mubin Sayyed <mubin.sayyed@amd.com>
   - Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
 
 properties:
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml
index 86f6a19b28ae..6136c8832a66 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.yaml
+++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml
@@ -129,6 +129,8 @@ patternProperties:
     description: Andes Technology Corporation
   "^anvo,.*":
     description: Anvo-Systems Dresden GmbH
+  "^aoly,.*":
+    description: Shenzhen Aoly Technology Co., Ltd.
   "^aosong,.*":
     description: Guangzhou Aosong Electronic Co., Ltd.
   "^apm,.*":
diff --git a/Documentation/filesystems/ext4/super.rst b/Documentation/filesystems/ext4/super.rst
index a1eb4a11a1d0..1b240661bfa3 100644
--- a/Documentation/filesystems/ext4/super.rst
+++ b/Documentation/filesystems/ext4/super.rst
@@ -328,9 +328,13 @@ The ext4 superblock is laid out as follows in
      - s_checksum_type
      - Metadata checksum algorithm type. The only valid value is 1 (crc32c).
    * - 0x176
-     - __le16
-     - s_reserved_pad
-     -
+     - \_\_u8
+     - s\_encryption\_level
+     - Versioning level for encryption.
+   * - 0x177
+     - \_\_u8
+     - s\_reserved\_pad
+     - Padding to next 32bits.
    * - 0x178
      - __le64
      - s_kbytes_written
@@ -466,9 +470,13 @@ The ext4 superblock is laid out as follows in
      - s_last_error_time_hi
      - Upper 8 bits of the s_last_error_time field.
    * - 0x27A
-     - __u8
-     - s_pad[2]
-     - Zero padding.
+     - \_\_u8
+     - s\_first\_error\_errcode
+     -
+   * - 0x27B
+     - \_\_u8
+     - s\_last\_error\_errcode
+     -
    * - 0x27C
      - __le16
      - s_encoding
diff --git a/Documentation/gpu/amdgpu/amd-hardware-list-info.rst b/Documentation/gpu/amdgpu/amd-hardware-list-info.rst
new file mode 100644
index 000000000000..1786544fe7c1
--- /dev/null
+++ b/Documentation/gpu/amdgpu/amd-hardware-list-info.rst
@@ -0,0 +1,23 @@
+=================================================
+ AMD Hardware Components Information per Product
+=================================================
+
+On this page, you can find the AMD product name and which component version is
+part of it.
+
+Accelerated Processing Units (APU) Info
+---------------------------------------
+
+.. csv-table::
+   :header-rows: 1
+   :widths: 3, 2, 2, 1, 1, 1, 1
+   :file: ./apu-asic-info-table.csv
+
+Discrete GPU Info
+-----------------
+
+.. csv-table::
+   :header-rows: 1
+   :widths: 3, 2, 2, 1, 1, 1
+   :file: ./dgpu-asic-info-table.csv
+
diff --git a/Documentation/gpu/amdgpu/amdgpu-glossary.rst b/Documentation/gpu/amdgpu/amdgpu-glossary.rst
index 1e9283e076ba..30812d9d53c6 100644
--- a/Documentation/gpu/amdgpu/amdgpu-glossary.rst
+++ b/Documentation/gpu/amdgpu/amdgpu-glossary.rst
@@ -12,18 +12,39 @@ we have a dedicated glossary for Display Core at
       The number of CUs that are active on the system.  The number of active
       CUs may be less than SE * SH * CU depending on the board configuration.
 
+    BACO
+      Bus Alive, Chip Off
+
+    BOCO
+      Bus Off, Chip Off
+
     CE
       Constant Engine
 
+    CIK
+      Sea Islands
+
+    CB
+      Color Buffer
+
     CP
       Command Processor
 
     CPLIB
       Content Protection Library
 
+    CS
+      Command Submission
+
+    CSB
+      Clear State Indirect Buffer
+
     CU
       Compute Unit
 
+    DB
+      Depth Buffer
+
     DFS
       Digital Frequency Synthesizer
 
@@ -33,6 +54,9 @@ we have a dedicated glossary for Display Core at
     EOP
       End Of Pipe/Pipeline
 
+    FLR
+      Function Level Reset
+
     GART
       Graphics Address Remapping Table.  This is the name we use for the GPUVM
       page table used by the GPU kernel driver.  It remaps system resources
@@ -45,6 +69,12 @@ we have a dedicated glossary for Display Core at
     GC
       Graphics and Compute
 
+    GDS
+      Global Data Share
+
+    GE
+      Geometry Engine
+
     GMC
       Graphic Memory Controller
 
@@ -80,6 +110,9 @@ we have a dedicated glossary for Display Core at
     KCQ
       Kernel Compute Queue
 
+    KFD
+      Kernel Fusion Driver
+
     KGQ
       Kernel Graphics Queue
 
@@ -89,6 +122,9 @@ we have a dedicated glossary for Display Core at
     MC
       Memory Controller
 
+    MCBP
+      Mid Command Buffer Preemption
+
     ME
       MicroEngine (Graphics)
 
@@ -104,6 +140,9 @@ we have a dedicated glossary for Display Core at
     MQD
       Memory Queue Descriptor
 
+    PA
+      Primitive Assembler / Physical Address
+
     PFP
       Pre-Fetch Parser (Graphics)
 
@@ -113,24 +152,39 @@ we have a dedicated glossary for Display Core at
     PSP
         Platform Security Processor
 
+    RB
+      Render Backends. Some people called it ROPs.
+
     RLC
       RunList Controller. This name is a remnant of past ages and doesn't have
       much meaning today. It's a group of general-purpose helper engines for
       the GFX block. It's involved in GFX power management and SR-IOV, among
       other things.
 
+    SC
+      Scan Converter
+
     SDMA
       System DMA
 
     SE
       Shader Engine
 
+    SGPR
+      Scalar General-Purpose Registers
+
     SH
       SHader array
 
+    SI
+      Southern Islands
+
     SMU/SMC
       System Management Unit / System Management Controller
 
+    SPI (AMDGPU)
+      Shader Processor Input
+
     SRLC
       Save/Restore List Control
 
@@ -143,12 +197,21 @@ we have a dedicated glossary for Display Core at
     SS
       Spread Spectrum
 
+    SX
+      Shader Export
+
     TA
       Trusted Application
 
+    TC
+      Texture Cache
+
     TOC
       Table of Contents
 
+    UMSCH
+      User Mode Scheduler
+
     UVD
       Unified Video Decoder
 
@@ -158,5 +221,17 @@ we have a dedicated glossary for Display Core at
     VCN
       Video Codec Next
 
+    VGPR
+      Vector General-Purpose Registers
+
+    VMID
+      Virtual Memory ID
+
     VPE
       Video Processing Engine
+
+    XCC
+      Accelerator Core Complex
+
+    XCP
+      Accelerator Core Partition
diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
index 5dd4b8762d19..1d50b539677f 100644
--- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
@@ -13,3 +13,5 @@ Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8
 Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11
 Ryzen 8x40 series, Hawk Point, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11
 Ryzen AI 300 series, Strix Point, 3.5.0, 11.5.0, 4.0.5, 6.1.0, 14.0.0
+Ryzen AI 350 series, Krackan Point, 3.5.0, 11.5.2, 4.0.5, 6.1.2, 14.0.4
+Ryzen AI Max 300 series, Strix Halo, 3.5.1, 11.5.1, 4.0.6, 6.1.1, 14.0.1
diff --git a/Documentation/gpu/amdgpu/debugfs.rst b/Documentation/gpu/amdgpu/debugfs.rst
new file mode 100644
index 000000000000..5150d0a95658
--- /dev/null
+++ b/Documentation/gpu/amdgpu/debugfs.rst
@@ -0,0 +1,210 @@
+==============
+AMDGPU DebugFS
+==============
+
+The amdgpu driver provides a number of debugfs files to aid in debugging
+issues in the driver.  These are usually found in
+/sys/kernel/debug/dri/<num>.
+
+DebugFS Files
+=============
+
+amdgpu_benchmark
+----------------
+
+Run benchmarks using the DMA engine the driver uses for GPU memory paging.
+Write a number to the file to run the test.  The results are written to the
+kernel log.  VRAM is on device memory (dGPUs) or carve out (APUs) and GTT
+(Graphics Translation Tables) is system memory that is accessible by the GPU.
+The following tests are available:
+
+- 1: simple test, VRAM to GTT and GTT to VRAM
+- 2: simple test, VRAM to VRAM
+- 3: GTT to VRAM, buffer size sweep, powers of 2
+- 4: VRAM to GTT, buffer size sweep, powers of 2
+- 5: VRAM to VRAM, buffer size sweep, powers of 2
+- 6: GTT to VRAM, buffer size sweep, common display sizes
+- 7: VRAM to GTT, buffer size sweep, common display sizes
+- 8: VRAM to VRAM, buffer size sweep, common display sizes
+
+amdgpu_test_ib
+--------------
+
+Read this file to run simple IB (Indirect Buffer) tests on all kernel managed
+rings.  IBs are command buffers usually generated by userspace applications
+which are submitted to the kernel for execution on an particular GPU engine.
+This just runs the simple IB tests included in the kernel.  These tests
+are engine specific and verify that IB submission works.
+
+amdgpu_discovery
+----------------
+
+Provides raw access to the IP discovery binary provided by the GPU.  Read this
+file to access the raw binary.  This is useful for verifying the contents of
+the IP discovery table.  It is chip specific.
+
+amdgpu_vbios
+------------
+
+Provides raw access to the ROM binary image from the GPU.  Read this file to
+access the raw binary.  This is useful for verifying the contents of the
+video BIOS ROM.  It is board specific.
+
+amdgpu_evict_gtt
+----------------
+
+Evict all buffers from the GTT memory pool.  Read this file to evict all
+buffers from this pool.
+
+amdgpu_evict_vram
+-----------------
+
+Evict all buffers from the VRAM memory pool.  Read this file to evict all
+buffers from this pool.
+
+amdgpu_gpu_recover
+------------------
+
+Trigger a GPU reset.  Read this file to trigger reset the entire GPU.
+All work currently running  on the GPU will be lost.
+
+amdgpu_ring_<name>
+------------------
+
+Provides read access to the kernel managed ring buffers for each ring <name>.
+These are useful for debugging problems on a particular ring.  The ring buffer
+is how the CPU sends commands to the GPU.  The CPU writes commands into the
+buffer and then asks the GPU engine to process it.  This is the raw binary
+contents of the ring buffer.  Use a tool like UMR to decode the rings into human
+readable form.
+
+amdgpu_mqd_<name>
+-----------------
+
+Provides read access to the kernel managed MQD (Memory Queue Descriptor) for
+ring <name> managed by the kernel driver.  MQDs define the features of the ring
+and are used to store the ring's state when it is not connected to hardware.
+The driver writes the requested ring features and metadata (GPU addresses of
+the ring itself and associated buffers) to the MQD and the firmware uses the MQD
+to populate the hardware when the ring is mapped to a hardware slot.  Only
+available on engines which use MQDs.  This provides access to the raw MQD
+binary.
+
+amdgpu_error_<name>
+-------------------
+
+Provides an interface to set an error code on the dma fences associated with
+ring <name>.  The error code specified is propogated to all fences associated
+with the ring.  Use this to inject a fence error into a ring.
+
+amdgpu_pm_info
+--------------
+
+Provides human readable information about the power management features
+and state of the GPU.  This includes current GFX clock, Memory clock,
+voltages, average SoC power, temperature, GFX load, Memory load, SMU
+feature mask, VCN power state, clock and power gating features.
+
+amdgpu_firmware_info
+--------------------
+
+Lists the firmware versions for all firmwares used by the GPU.  Only
+entries with a non-0 version are valid.  If the version is 0, the firmware
+is not valid for the GPU.
+
+amdgpu_fence_info
+-----------------
+
+Shows the last signalled and emitted fence sequence numbers for each
+kernel driver managed ring.  Fences are associated with submissions
+to the engine.  Emitted fences have been submitted to the ring
+and signalled fences have been signalled by the GPU.  Rings with a
+larger emitted fence value have outstanding work that is still being
+processed by the engine that owns that ring.  When the emitted and
+signalled fence values are equal, the ring is idle.
+
+amdgpu_gem_info
+---------------
+
+Lists all of the PIDs using the GPU and the GPU buffers that they have
+allocated.  This lists the buffer size, pool (VRAM, GTT, etc.), and buffer
+attributes (CPU access required, CPU cache attributes, etc.).
+
+amdgpu_vm_info
+--------------
+
+Lists all of the PIDs using the GPU and the GPU buffers that they have
+allocated as well as the status of those buffers relative to that process'
+GPU virtual address space (e.g., evicted, idle, invalidated, etc.).
+
+amdgpu_sa_info
+--------------
+
+Prints out all of the suballocations (sa) by the suballocation manager in the
+kernel driver.  Prints the GPU address, size, and fence info associated
+with each suballocation.  The suballocations are used internally within
+the kernel driver for various things.
+
+amdgpu_<pool>_mm
+----------------
+
+Prints TTM information about the memory pool <pool>.
+
+amdgpu_vram
+-----------
+
+Provides direct access to VRAM.  Used by tools like UMR to inspect
+objects in VRAM.
+
+amdgpu_iomem
+------------
+
+Provides direct access to GTT memory.  Used by tools like UMR to inspect
+GTT memory.
+
+amdgpu_regs_*
+-------------
+
+Provides direct access to various register aperatures on the GPU.  Used
+by tools like UMR to access GPU registers.
+
+amdgpu_regs2
+------------
+
+Provides an IOCTL interface used by UMR for interacting with GPU registers.
+
+
+amdgpu_sensors
+--------------
+
+Provides an interface to query GPU power metrics (temperature, average
+power, etc.).  Used by tools like UMR to query GPU power metrics.
+
+
+amdgpu_gca_config
+-----------------
+
+Provides an interface to query GPU details (Graphics/Compute Array config,
+PCI config, GPU family, etc.).  Used by tools like UMR to query GPU details.
+
+amdgpu_wave
+-----------
+
+Used to query GFX/compute wave information from the hardware.  Used by tools
+like UMR to query GFX/compute wave information.
+
+amdgpu_gpr
+----------
+
+Used to query GFX/compute GPR (General Purpose Register) information from the
+hardware.  Used by tools like UMR to query GPRs when debugging shaders.
+
+amdgpu_gprwave
+--------------
+
+Provides an IOCTL interface used by UMR for interacting with shader waves.
+
+amdgpu_fw_attestation
+---------------------
+
+Provides an interface for reading back firmware attestation records.
diff --git a/Documentation/gpu/amdgpu/debugging.rst b/Documentation/gpu/amdgpu/debugging.rst
index e75f97d0e4ea..7cbfea0606e1 100644
--- a/Documentation/gpu/amdgpu/debugging.rst
+++ b/Documentation/gpu/amdgpu/debugging.rst
@@ -2,6 +2,13 @@
  GPU Debugging
 ===============
 
+General Debugging Options
+=========================
+
+The DebugFS section provides documentation on a number files to aid in debugging
+issues on the GPU.
+
+
 GPUVM Debugging
 ===============
 
diff --git a/Documentation/gpu/amdgpu/display/dc-debug.rst b/Documentation/gpu/amdgpu/display/dc-debug.rst
index 013f63b271f3..605dca21f4ae 100644
--- a/Documentation/gpu/amdgpu/display/dc-debug.rst
+++ b/Documentation/gpu/amdgpu/display/dc-debug.rst
@@ -154,7 +154,7 @@ of the display parameters, but the userspace might also cause this issue. One
 way to identify the source of the problem is to take a screenshot or make a
 desktop video capture when the problem happens; after checking the
 screenshot/video recording, if you don't see any of the artifacts, it means
-that the issue is likely on the the driver side. If you can still see the
+that the issue is likely on the driver side. If you can still see the
 problem in the data collected, it is an issue that probably happened during
 rendering, and the display code just got the framebuffer already corrupted.
 
diff --git a/Documentation/gpu/amdgpu/driver-core.rst b/Documentation/gpu/amdgpu/driver-core.rst
index 32723a925377..81256318e93c 100644
--- a/Documentation/gpu/amdgpu/driver-core.rst
+++ b/Documentation/gpu/amdgpu/driver-core.rst
@@ -67,36 +67,66 @@ GC (Graphics and Compute)
     This is the graphics and compute engine, i.e., the block that
     encompasses the 3D pipeline and and shader blocks.  This is by far the
     largest block on the GPU.  The 3D pipeline has tons of sub-blocks.  In
-    addition to that, it also contains the CP microcontrollers (ME, PFP,
-    CE, MEC) and the RLC microcontroller.  It's exposed to userspace for
-    user mode drivers (OpenGL, Vulkan, OpenCL, etc.)
+    addition to that, it also contains the CP microcontrollers (ME, PFP, CE,
+    MEC) and the RLC microcontroller.  It's exposed to userspace for user mode
+    drivers (OpenGL, Vulkan, OpenCL, etc.). More details in :ref:`Graphics (GFX)
+    and Compute <amdgpu-gc>`.
 
 VCN (Video Core Next)
     This is the multi-media engine.  It handles video and image encode and
     decode.  It's exposed to userspace for user mode drivers (VA-API,
     OpenMAX, etc.)
 
-Graphics and Compute Microcontrollers
--------------------------------------
-
-CP (Command Processor)
-    The name for the hardware block that encompasses the front end of the
-    GFX/Compute pipeline.  Consists mainly of a bunch of microcontrollers
-    (PFP, ME, CE, MEC).  The firmware that runs on these microcontrollers
-    provides the driver interface to interact with the GFX/Compute engine.
-
-    MEC (MicroEngine Compute)
-        This is the microcontroller that controls the compute queues on the
-        GFX/compute engine.
-
-    MES (MicroEngine Scheduler)
-        This is a new engine for managing queues.  This is currently unused.
-
-RLC (RunList Controller)
-    This is another microcontroller in the GFX/Compute engine.  It handles
-    power management related functionality within the GFX/Compute engine.
-    The name is a vestige of old hardware where it was originally added
-    and doesn't really have much relation to what the engine does now.
+.. _pipes-and-queues-description:
+
+GFX, Compute, and SDMA Overall Behavior
+=======================================
+
+.. note:: For simplicity, whenever the term block is used in this section, it
+   means GFX, Compute, and SDMA.
+
+GFX, Compute and SDMA share a similar form of operation that can be abstracted
+to facilitate understanding of the behavior of these blocks. See the figure
+below illustrating the common components of these blocks:
+
+.. kernel-figure:: pipe_and_queue_abstraction.svg
+
+In the central part of this figure, you can see two hardware elements, one called
+**Pipes** and another called **Queues**; it is important to highlight that Queues
+must be associated with a Pipe and vice-versa. Every specific hardware IP may have
+a different number of Pipes and, in turn, a different number of Queues; for
+example, GFX 11 has two Pipes and two Queues per Pipe for the GFX front end.
+
+Pipe is the hardware that processes the instructions available in the Queues;
+in other words, it is a thread executing the operations inserted in the Queue.
+One crucial characteristic of Pipes is that they can only execute one Queue at
+a time; no matter if the hardware has multiple Queues in the Pipe, it only runs
+one Queue per Pipe.
+
+Pipes have the mechanics of swapping between queues at the hardware level.
+Nonetheless, they only make use of Queues that are considered mapped. Pipes can
+switch between queues based on any of the following inputs:
+
+1. Command Stream;
+2. Packet by Packet;
+3. Other hardware requests the change (e.g., MES).
+
+Queues within Pipes are defined by the Hardware Queue Descriptors (HQD).
+Associated with the HQD concept, we have the Memory Queue Descriptor (MQD),
+which is responsible for storing information about the state of each of the
+available Queues in the memory. The state of a Queue contains information such
+as the GPU virtual address of the queue itself, save areas, doorbell, etc. The
+MQD also stores the HQD registers, which are vital for activating or
+deactivating a given Queue.  The scheduling firmware (e.g., MES) is responsible
+for loading HQDs from MQDs and vice versa.
+
+The Queue-switching process can also happen with the firmware requesting the
+preemption or unmapping of a Queue. The firmware waits for the HQD_ACTIVE bit
+to change to low before saving the state into the MQD. To make a different
+Queue become active, the firmware copies the MQD state into the HQD registers
+and loads any additional state. Finally, it sets the HQD_ACTIVE bit to high to
+indicate that the queue is active.  The Pipe will then execute work from active
+Queues.
 
 Driver Structure
 ================
@@ -110,7 +140,8 @@ Some useful constructs:
 KIQ (Kernel Interface Queue)
     This is a control queue used by the kernel driver to manage other gfx
     and compute queues on the GFX/compute engine.  You can use it to
-    map/unmap additional queues, etc.
+    map/unmap additional queues, etc.  This is replaced by MES on
+    GFX 11 and newer hardware.
 
 IB (Indirect Buffer)
     A command buffer for a particular engine.  Rather than writing
diff --git a/Documentation/gpu/amdgpu/driver-misc.rst b/Documentation/gpu/amdgpu/driver-misc.rst
index e40e15f89fd3..25b0c857816e 100644
--- a/Documentation/gpu/amdgpu/driver-misc.rst
+++ b/Documentation/gpu/amdgpu/driver-misc.rst
@@ -50,23 +50,6 @@ board_info
 .. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
    :doc: board_info
 
-Accelerated Processing Units (APU) Info
----------------------------------------
-
-.. csv-table::
-   :header-rows: 1
-   :widths: 3, 2, 2, 1, 1, 1, 1
-   :file: ./apu-asic-info-table.csv
-
-Discrete GPU Info
------------------
-
-.. csv-table::
-   :header-rows: 1
-   :widths: 3, 2, 2, 1, 1, 1
-   :file: ./dgpu-asic-info-table.csv
-
-
 GPU Memory Usage Information
 ============================
 
diff --git a/Documentation/gpu/amdgpu/gc/index.rst b/Documentation/gpu/amdgpu/gc/index.rst
new file mode 100644
index 000000000000..ff6e9ef5cbee
--- /dev/null
+++ b/Documentation/gpu/amdgpu/gc/index.rst
@@ -0,0 +1,52 @@
+.. _amdgpu-gc:
+
+========================================
+ drm/amdgpu - Graphics and Compute (GC)
+========================================
+
+The relationship between the CPU and GPU can be described as the
+producer-consumer problem, where the CPU fills out a buffer with operations
+(producer) to be executed by the GPU (consumer). The requested operations in
+the buffer are called Command Packets, which can be summarized as a compressed
+way of transmitting command information to the graphics controller.
+
+The component that acts as the front end between the CPU and the GPU is called
+the Command Processor (CP). This component is responsible for providing greater
+flexibility to the GC since CP makes it possible to program various aspects of
+the GPU pipeline. CP also coordinates the communication between the CPU and GPU
+via a mechanism named **Ring Buffers**, where the CPU appends information to
+the buffer while the GPU removes operations. It is relevant to highlight that a
+CPU can add a pointer to the Ring Buffer that points to another region of
+memory outside the Ring Buffer, and CP can handle it; this mechanism is called
+**Indirect Buffer (IB)**. CP receives and parses the Command Streams (CS), and
+writes the operations to the correct hardware blocks.
+
+Graphics (GFX) and Compute Microcontrollers
+-------------------------------------------
+
+GC is a large block, and as a result, it has multiple firmware associated with
+it. Some of them are:
+
+CP (Command Processor)
+    The name for the hardware block that encompasses the front end of the
+    GFX/Compute pipeline. Consists mainly of a bunch of microcontrollers
+    (PFP, ME, CE, MEC). The firmware that runs on these microcontrollers
+    provides the driver interface to interact with the GFX/Compute engine.
+
+    MEC (MicroEngine Compute)
+        This is the microcontroller that controls the compute queues on the
+        GFX/compute engine.
+
+    MES (MicroEngine Scheduler)
+        This is the engine for managing queues. For more details check
+        :ref:`MicroEngine Scheduler (MES) <amdgpu-mes>`.
+
+RLC (RunList Controller)
+    This is another microcontroller in the GFX/Compute engine. It handles
+    power management related functionality within the GFX/Compute engine.
+    The name is a vestige of old hardware where it was originally added
+    and doesn't really have much relation to what the engine does now.
+
+.. toctree::
+
+   mes.rst
diff --git a/Documentation/gpu/amdgpu/gc/mes.rst b/Documentation/gpu/amdgpu/gc/mes.rst
new file mode 100644
index 000000000000..b99eb211b179
--- /dev/null
+++ b/Documentation/gpu/amdgpu/gc/mes.rst
@@ -0,0 +1,38 @@
+.. _amdgpu-mes:
+
+=============================
+ MicroEngine Scheduler (MES)
+=============================
+
+.. note::
+   Queue and ring buffer are used as a synonymous.
+
+.. note::
+   This section assumes that you are familiar with the concept of Pipes, Queues, and GC.
+   If not, check :ref:`GFX, Compute, and SDMA Overall Behavior<pipes-and-queues-description>`
+   and :ref:`drm/amdgpu - Graphics and Compute (GC) <amdgpu-gc>`.
+
+Every GFX has a pipe component with one or more hardware queues. Pipes can
+switch between queues depending on certain conditions, and one of the
+components that can request a queue switch to a pipe is the MicroEngine
+Scheduler (MES). Whenever the driver is initialized, it creates one MQD per
+hardware queue, and then the MQDs are handed to the MES firmware for mapping
+to:
+
+1. Kernel Queues (legacy): This queue is statically mapped to HQDs and never
+   preempted. Even though this is a legacy feature, it is the current default, and
+   most existing hardware supports it. When an application submits work to the
+   kernel driver, it submits all of the application command buffers to the kernel
+   queues. The CS IOCTL takes the command buffer from the applications and
+   schedules them on the kernel queue.
+
+2. User Queues: These queues are dynamically mapped to the HQDs. Regarding the
+   utilization of User Queues, the userspace application will create its user
+   queues and submit work directly to its user queues with no need to IOCTL for
+   each submission and no need to share a single kernel queue.
+
+In terms of User Queues, MES can dynamically map them to the HQD. If there are
+more MQDs than HQDs, the MES firmware will preempt other user queues to make
+sure each queues get a time slice; in other words, MES is a microcontroller
+that handles the mapping and unmapping of MQDs into HQDs, as well as the
+priorities and oversubscription of MQDs.
diff --git a/Documentation/gpu/amdgpu/index.rst b/Documentation/gpu/amdgpu/index.rst
index 302d039928ee..bb2894b5edaf 100644
--- a/Documentation/gpu/amdgpu/index.rst
+++ b/Documentation/gpu/amdgpu/index.rst
@@ -7,8 +7,10 @@ Next (GCN), Radeon DNA (RDNA), and Compute DNA (CDNA) architectures.
 
 .. toctree::
 
-   module-parameters
    driver-core
+   amd-hardware-list-info
+   module-parameters
+   gc/index
    display/index
    flashing
    xgmi
@@ -16,5 +18,6 @@ Next (GCN), Radeon DNA (RDNA), and Compute DNA (CDNA) architectures.
    thermal
    driver-misc
    debugging
+   debugfs
    process-isolation
    amdgpu-glossary
diff --git a/Documentation/gpu/amdgpu/pipe_and_queue_abstraction.svg b/Documentation/gpu/amdgpu/pipe_and_queue_abstraction.svg
new file mode 100644
index 000000000000..0df3c6b3000b
--- /dev/null
+++ b/Documentation/gpu/amdgpu/pipe_and_queue_abstraction.svg
@@ -0,0 +1,1279 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!-- Created with Inkscape (http://www.inkscape.org/) -->
+
+<svg
+   width="395.47891mm"
+   height="234.73715mm"
+   viewBox="0 0 395.47891 234.73714"
+   version="1.1"
+   id="svg1"
+   inkscape:version="1.4 (e7c3feb100, 2024-10-09)"
+   sodipodi:docname="pipe_and_queue_abstraction.svg"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg">
+  <sodipodi:namedview
+     id="namedview1"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     inkscape:document-units="mm"
+     inkscape:zoom="1.6489689"
+     inkscape:cx="713.17296"
+     inkscape:cy="466.65527"
+     inkscape:window-width="3840"
+     inkscape:window-height="2083"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1"
+     inkscape:current-layer="layer1" />
+  <defs
+     id="defs1">
+    <marker
+       style="overflow:visible"
+       id="Dot"
+       refX="0"
+       refY="0"
+       orient="auto"
+       inkscape:stockid="Dot"
+       markerWidth="1"
+       markerHeight="1"
+       viewBox="0 0 1 1"
+       inkscape:isstock="true"
+       inkscape:collect="always"
+       preserveAspectRatio="xMidYMid">
+      <path
+         transform="scale(0.5)"
+         style="fill:context-stroke;fill-rule:evenodd;stroke:none"
+         d="M 5,0 C 5,2.76 2.76,5 0,5 -2.76,5 -5,2.76 -5,0 c 0,-2.76 2.3,-5 5,-5 2.76,0 5,2.24 5,5 z"
+         sodipodi:nodetypes="sssss"
+         id="path110" />
+    </marker>
+    <marker
+       style="overflow:visible"
+       id="marker109"
+       refX="0"
+       refY="0"
+       orient="auto-start-reverse"
+       inkscape:stockid="Stylized triangle arrow"
+       markerWidth="1"
+       markerHeight="1"
+       viewBox="0 0 1 1"
+       inkscape:isstock="true"
+       inkscape:collect="always"
+       preserveAspectRatio="xMidYMid">
+      <path
+         transform="scale(0.5)"
+         style="fill:context-stroke;fill-rule:evenodd;stroke:context-stroke;stroke-width:1pt"
+         d="m 6,0 c -3,1 -7,3 -9,5 0,0 0,-4 2,-5 -2,-1 -2,-5 -2,-5 2,2 6,4 9,5 z"
+         id="path109" />
+    </marker>
+    <marker
+       style="overflow:visible"
+       id="ArrowTriangleStylized"
+       refX="0"
+       refY="0"
+       orient="auto-start-reverse"
+       inkscape:stockid="Stylized triangle arrow"
+       markerWidth="1"
+       markerHeight="1"
+       viewBox="0 0 1 1"
+       inkscape:isstock="true"
+       inkscape:collect="always"
+       preserveAspectRatio="xMidYMid">
+      <path
+         transform="scale(0.5)"
+         style="fill:context-stroke;fill-rule:evenodd;stroke:context-stroke;stroke-width:1pt"
+         d="m 6,0 c -3,1 -7,3 -9,5 0,0 0,-4 2,-5 -2,-1 -2,-5 -2,-5 2,2 6,4 9,5 z"
+         id="path108" />
+    </marker>
+    <marker
+       style="overflow:visible"
+       id="ArrowWide"
+       refX="0"
+       refY="0"
+       orient="auto-start-reverse"
+       inkscape:stockid="Wide arrow"
+       markerWidth="1"
+       markerHeight="1"
+       viewBox="0 0 1 1"
+       inkscape:isstock="true"
+       inkscape:collect="always"
+       preserveAspectRatio="xMidYMid">
+      <path
+         style="fill:none;stroke:context-stroke;stroke-width:1;stroke-linecap:butt"
+         d="M 3,-3 0,0 3,3"
+         transform="rotate(180,0.125,0)"
+         sodipodi:nodetypes="ccc"
+         id="path1" />
+    </marker>
+    <marker
+       style="overflow:visible"
+       id="Triangle"
+       refX="0"
+       refY="0"
+       orient="auto-start-reverse"
+       inkscape:stockid="Triangle arrow"
+       markerWidth="1"
+       markerHeight="1"
+       viewBox="0 0 1 1"
+       inkscape:isstock="true"
+       inkscape:collect="always"
+       preserveAspectRatio="xMidYMid">
+      <path
+         transform="scale(0.5)"
+         style="fill:context-stroke;fill-rule:evenodd;stroke:context-stroke;stroke-width:1pt"
+         d="M 5.77,0 -2.88,5 V -5 Z"
+         id="path135" />
+    </marker>
+    <marker
+       style="overflow:visible"
+       id="ArrowWideHeavy"
+       refX="0"
+       refY="0"
+       orient="auto-start-reverse"
+       inkscape:stockid="Wide, heavy arrow"
+       markerWidth="1"
+       markerHeight="1"
+       viewBox="0 0 1 1"
+       inkscape:isstock="true"
+       inkscape:collect="always"
+       preserveAspectRatio="xMidYMid">
+      <path
+         style="fill:context-stroke;fill-rule:evenodd;stroke:none"
+         d="m 1,0 -3,3 h -2 l 3,-3 -3,-3 h 2 z"
+         id="path71" />
+    </marker>
+  </defs>
+  <g
+     inkscape:label="Layer 1"
+     inkscape:groupmode="layer"
+     id="layer1"
+     transform="translate(149.03517,55.110629)">
+    <circle
+       style="fill:#ffeeaa;fill-opacity:1;stroke:#1a1a1a;stroke-width:0.733436;stroke-dasharray:none;stroke-dashoffset:0"
+       id="path98"
+       cx="-35.757576"
+       cy="-10.495151"
+       r="44.24876" />
+    <rect
+       style="fill:none;stroke:#000000;stroke-width:0.878057;stroke-dasharray:none"
+       id="rect1"
+       width="167.79619"
+       height="24.831829"
+       x="14.21942"
+       y="57.862854" />
+    <g
+       id="g11"
+       transform="translate(24.021362,-46.545299)">
+      <rect
+         style="fill:none;stroke:#006680;stroke-width:0.459999;stroke-dasharray:none"
+         id="rect2"
+         width="131.09708"
+         height="5.8163381"
+         x="23.245802"
+         y="107.16314" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 37.884707,107.26337 v 5.53565"
+         id="path2"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 52.466233,107.26337 v 5.53565"
+         id="path3"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 67.047758,107.26337 v 5.53565"
+         id="path4"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 81.629283,107.26337 v 5.53565"
+         id="path5"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 96.210809,107.26337 v 5.53565"
+         id="path6"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 110.79233,107.26337 v 5.53565"
+         id="path7"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 125.37386,107.26337 v 5.53565"
+         id="path8"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 139.95538,107.26337 v 5.53565"
+         id="path9"
+         sodipodi:nodetypes="cc" />
+      <text
+         xml:space="preserve"
+         style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         x="86.713425"
+         y="110.1963"
+         id="text11"><tspan
+           sodipodi:role="line"
+           id="tspan11"
+           style="stroke:#006680;stroke-width:0.5"
+           x="86.713425"
+           y="110.1963">. . .</tspan></text>
+    </g>
+    <g
+       id="g18"
+       transform="translate(24.021362,-32.25779)">
+      <rect
+         style="fill:none;stroke:#006680;stroke-width:0.459999;stroke-dasharray:none"
+         id="rect11"
+         width="131.09708"
+         height="5.8163381"
+         x="23.245802"
+         y="107.16314" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 37.884707,107.26337 v 5.53565"
+         id="path11"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 52.466233,107.26337 v 5.53565"
+         id="path12"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 67.047758,107.26337 v 5.53565"
+         id="path13"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 81.629283,107.26337 v 5.53565"
+         id="path14"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 96.210809,107.26337 v 5.53565"
+         id="path15"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 110.79233,107.26337 v 5.53565"
+         id="path16"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 125.37386,107.26337 v 5.53565"
+         id="path17"
+         sodipodi:nodetypes="cc" />
+      <path
+         style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         d="m 139.95538,107.26337 v 5.53565"
+         id="path18"
+         sodipodi:nodetypes="cc" />
+      <text
+         xml:space="preserve"
+         style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         x="86.713425"
+         y="110.1963"
+         id="text18"><tspan
+           sodipodi:role="line"
+           id="tspan18"
+           style="stroke:#006680;stroke-width:0.5"
+           x="86.713425"
+           y="110.1963">. . .</tspan></text>
+    </g>
+    <text
+       xml:space="preserve"
+       style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+       x="113.37768"
+       y="68.463142"
+       id="text19"><tspan
+         sodipodi:role="line"
+         id="tspan19"
+         style="writing-mode:tb-rl;stroke-width:0.5"
+         x="113.37768"
+         y="68.463142">. . .</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="25.005701"
+       y="55.308445"
+       id="text844-2-9"><tspan
+         sodipodi:role="line"
+         x="25.005701"
+         y="55.308445"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan2868">Pipe[0]</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="63.075123"
+       id="text844-2-9-4"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="63.075123"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan2868-7">MQD</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="30.264952"
+       y="65.490654"
+       id="text20"><tspan
+         sodipodi:role="line"
+         x="30.264952"
+         y="65.490654"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan20">Queue[0]</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="30.264952"
+       y="79.249001"
+       id="text21"><tspan
+         sodipodi:role="line"
+         x="30.264952"
+         y="79.249001"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan21">Queue[n]</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="30.264952"
+       y="71.84066"
+       id="text22"><tspan
+         sodipodi:role="line"
+         x="30.264952"
+         y="71.84066"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan22">...</tspan></text>
+    <g
+       id="g71"
+       transform="translate(-23.283342)">
+      <rect
+         style="fill:#ffffff;fill-opacity:0;stroke:#000000;stroke-width:0.5;stroke-dasharray:0.5, 0.5;stroke-dashoffset:0"
+         id="rect67"
+         width="18.533583"
+         height="114.96632"
+         x="250.9435"
+         y="54.754276"
+         ry="6.0427966" />
+    </g>
+    <rect
+       style="fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:0.499999;stroke-dasharray:none;stroke-dashoffset:0"
+       id="rect68"
+       width="188.21231"
+       height="139.5948"
+       x="4.0113592"
+       y="37.597778"
+       ry="0" />
+    <g
+       id="g43"
+       transform="translate(0,40.745853)">
+      <rect
+         style="fill:none;stroke:#000000;stroke-width:0.878057;stroke-dasharray:none"
+         id="rect22"
+         width="167.79619"
+         height="24.831829"
+         x="14.21942"
+         y="57.862854" />
+      <g
+         id="g30"
+         transform="translate(24.021362,-46.545299)">
+        <rect
+           style="fill:none;stroke:#006680;stroke-width:0.459999;stroke-dasharray:none"
+           id="rect23"
+           width="131.09708"
+           height="5.8163381"
+           x="23.245802"
+           y="107.16314" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 37.884707,107.26337 v 5.53565"
+           id="path23"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 52.466233,107.26337 v 5.53565"
+           id="path24"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 67.047758,107.26337 v 5.53565"
+           id="path25"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 81.629283,107.26337 v 5.53565"
+           id="path26"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 96.210809,107.26337 v 5.53565"
+           id="path27"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 110.79233,107.26337 v 5.53565"
+           id="path28"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 125.37386,107.26337 v 5.53565"
+           id="path29"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 139.95538,107.26337 v 5.53565"
+           id="path30"
+           sodipodi:nodetypes="cc" />
+        <text
+           xml:space="preserve"
+           style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           x="86.713425"
+           y="110.1963"
+           id="text30"><tspan
+             sodipodi:role="line"
+             id="tspan30"
+             style="stroke:#006680;stroke-width:0.5"
+             x="86.713425"
+             y="110.1963">. . .</tspan></text>
+      </g>
+      <g
+         id="g38"
+         transform="translate(24.021362,-32.25779)">
+        <rect
+           style="fill:none;stroke:#006680;stroke-width:0.459999;stroke-dasharray:none"
+           id="rect30"
+           width="131.09708"
+           height="5.8163381"
+           x="23.245802"
+           y="107.16314" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 37.884707,107.26337 v 5.53565"
+           id="path31"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 52.466233,107.26337 v 5.53565"
+           id="path32"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 67.047758,107.26337 v 5.53565"
+           id="path33"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 81.629283,107.26337 v 5.53565"
+           id="path34"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 96.210809,107.26337 v 5.53565"
+           id="path35"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 110.79233,107.26337 v 5.53565"
+           id="path36"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 125.37386,107.26337 v 5.53565"
+           id="path37"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 139.95538,107.26337 v 5.53565"
+           id="path38"
+           sodipodi:nodetypes="cc" />
+        <text
+           xml:space="preserve"
+           style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           x="86.713425"
+           y="110.1963"
+           id="text38"><tspan
+             sodipodi:role="line"
+             id="tspan38"
+             style="stroke:#006680;stroke-width:0.5"
+             x="86.713425"
+             y="110.1963">. . .</tspan></text>
+      </g>
+      <text
+         xml:space="preserve"
+         style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         x="113.37768"
+         y="68.463142"
+         id="text39"><tspan
+           sodipodi:role="line"
+           id="tspan39"
+           style="writing-mode:tb-rl;stroke-width:0.5"
+           x="113.37768"
+           y="68.463142">. . .</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="25.005701"
+         y="55.308445"
+         id="text40"><tspan
+           sodipodi:role="line"
+           x="25.005701"
+           y="55.308445"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan40">Pipe[1]</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="30.264952"
+         y="65.490654"
+         id="text41"><tspan
+           sodipodi:role="line"
+           x="30.264952"
+           y="65.490654"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan41">Queue[0]</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="30.264952"
+         y="79.249001"
+         id="text42"><tspan
+           sodipodi:role="line"
+           x="30.264952"
+           y="79.249001"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan42">Queue[n]</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="30.264952"
+         y="71.84066"
+         id="text43"><tspan
+           sodipodi:role="line"
+           x="30.264952"
+           y="71.84066"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan43">...</tspan></text>
+    </g>
+    <g
+       id="g64"
+       transform="translate(0,85.195881)">
+      <rect
+         style="fill:none;stroke:#000000;stroke-width:0.878057;stroke-dasharray:none"
+         id="rect43"
+         width="167.79619"
+         height="24.831829"
+         x="14.21942"
+         y="57.862854" />
+      <g
+         id="g51"
+         transform="translate(24.021362,-46.545299)">
+        <rect
+           style="fill:none;stroke:#006680;stroke-width:0.459999;stroke-dasharray:none"
+           id="rect44"
+           width="131.09708"
+           height="5.8163381"
+           x="23.245802"
+           y="107.16314" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 37.884707,107.26337 v 5.53565"
+           id="path44"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 52.466233,107.26337 v 5.53565"
+           id="path45"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 67.047758,107.26337 v 5.53565"
+           id="path46"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 81.629283,107.26337 v 5.53565"
+           id="path47"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 96.210809,107.26337 v 5.53565"
+           id="path48"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 110.79233,107.26337 v 5.53565"
+           id="path49"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 125.37386,107.26337 v 5.53565"
+           id="path50"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 139.95538,107.26337 v 5.53565"
+           id="path51"
+           sodipodi:nodetypes="cc" />
+        <text
+           xml:space="preserve"
+           style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           x="86.713425"
+           y="110.1963"
+           id="text51"><tspan
+             sodipodi:role="line"
+             id="tspan51"
+             style="stroke:#006680;stroke-width:0.5"
+             x="86.713425"
+             y="110.1963">. . .</tspan></text>
+      </g>
+      <g
+         id="g59"
+         transform="translate(24.021362,-32.25779)">
+        <rect
+           style="fill:none;stroke:#006680;stroke-width:0.459999;stroke-dasharray:none"
+           id="rect51"
+           width="131.09708"
+           height="5.8163381"
+           x="23.245802"
+           y="107.16314" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 37.884707,107.26337 v 5.53565"
+           id="path52"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 52.466233,107.26337 v 5.53565"
+           id="path53"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 67.047758,107.26337 v 5.53565"
+           id="path54"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 81.629283,107.26337 v 5.53565"
+           id="path55"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 96.210809,107.26337 v 5.53565"
+           id="path56"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 110.79233,107.26337 v 5.53565"
+           id="path57"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 125.37386,107.26337 v 5.53565"
+           id="path58"
+           sodipodi:nodetypes="cc" />
+        <path
+           style="fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           d="m 139.95538,107.26337 v 5.53565"
+           id="path59"
+           sodipodi:nodetypes="cc" />
+        <text
+           xml:space="preserve"
+           style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+           x="86.713425"
+           y="110.1963"
+           id="text59"><tspan
+             sodipodi:role="line"
+             id="tspan59"
+             style="stroke:#006680;stroke-width:0.5"
+             x="86.713425"
+             y="110.1963">. . .</tspan></text>
+      </g>
+      <text
+         xml:space="preserve"
+         style="font-size:3.175px;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:none;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+         x="113.37768"
+         y="68.463142"
+         id="text60"><tspan
+           sodipodi:role="line"
+           id="tspan60"
+           style="writing-mode:tb-rl;stroke-width:0.5"
+           x="113.37768"
+           y="68.463142">. . .</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="25.005701"
+         y="55.308445"
+         id="text61"><tspan
+           sodipodi:role="line"
+           x="25.005701"
+           y="55.308445"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan61">Pipe[n]</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="30.264952"
+         y="65.490654"
+         id="text62"><tspan
+           sodipodi:role="line"
+           x="30.264952"
+           y="65.490654"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan62">Queue[0]</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="30.264952"
+         y="79.249001"
+         id="text63"><tspan
+           sodipodi:role="line"
+           x="30.264952"
+           y="79.249001"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan63">Queue[n]</tspan></text>
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="30.264952"
+         y="71.84066"
+         id="text64"><tspan
+           sodipodi:role="line"
+           x="30.264952"
+           y="71.84066"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+           id="tspan64">...</tspan></text>
+    </g>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:7.76111px;font-family:Serif;-inkscape-font-specification:Serif;text-align:start;writing-mode:lr-tb;direction:ltr;text-anchor:start;fill:#000000;fill-opacity:0;stroke:#006680;stroke-width:0.499999;stroke-dasharray:none"
+       x="92.18071"
+       y="128.21965"
+       id="text65"><tspan
+         sodipodi:role="line"
+         id="tspan65"
+         style="font-size:7.76111px;writing-mode:tb-rl;fill:#1a1a1a;stroke:#000000;stroke-width:0.5"
+         x="92.18071"
+         y="128.21965">...</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:7.7611px;font-family:Serif;-inkscape-font-specification:Serif;text-align:start;writing-mode:tb-rl;direction:ltr;text-anchor:start;fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:0.499999;stroke-dasharray:none"
+       x="198.72205"
+       y="80.708267"
+       id="text66"><tspan
+         sodipodi:role="line"
+         id="tspan66"
+         style="stroke-width:0.5"
+         x="198.72205"
+         y="80.708267" /></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="97.905846"
+       y="44.725101"
+       id="text68"><tspan
+         sodipodi:role="line"
+         x="97.905846"
+         y="44.725101"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan68">Hardware Block</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;writing-mode:vertical-lr;text-orientation:upright;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="236.36934"
+       y="112.10503"
+       id="text68-7"><tspan
+         sodipodi:role="line"
+         x="236.36934"
+         y="112.10503"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;writing-mode:vertical-lr;text-orientation:upright;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan68-4">EXECUTION</tspan></text>
+    <rect
+       style="fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:0.940575;stroke-dasharray:7.5246, 0.940575;stroke-dashoffset:0"
+       id="rect68-1"
+       width="68.749969"
+       height="141.2751"
+       x="-129.49162"
+       y="37.881134"
+       ry="0" />
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-114.4223"
+       y="44.966106"
+       id="text68-2"><tspan
+         sodipodi:role="line"
+         x="-114.4223"
+         y="44.966106"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan68-9">Memory</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:12.7px;font-family:Serif;-inkscape-font-specification:Serif;text-align:start;writing-mode:tb-rl;direction:ltr;text-orientation:upright;text-anchor:start;fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:0.499999;stroke-dasharray:none;stroke-dashoffset:0"
+       x="212.6013"
+       y="64.823341"
+       id="text69"><tspan
+         sodipodi:role="line"
+         id="tspan69"
+         style="font-size:12.7px;stroke-width:0.5"
+         x="212.6013"
+         y="64.823341" /></text>
+    <g
+       id="g72">
+      <path
+         style="font-weight:bold;font-size:16.9333px;line-height:1.25;-inkscape-font-specification:'sans-serif Bold';text-align:center;letter-spacing:0px;word-spacing:0px;text-anchor:middle;stroke-width:0.0690111"
+         d="m 185.24734,83.96512 h 1.28693 c 1.20226,0.01693 2.09973,-1.015998 2.0828,-2.370662 v -9.393759 c -0.0169,-0.711199 0.33866,-1.066798 1.10066,-1.066798 h 0.2032 0.0677 v -1.557864 c -1.0668,0.01693 -1.38853,-0.270933 -1.37159,-1.236131 V 58.96308 c 0.0339,-1.354664 -0.88054,-2.387595 -2.0828,-2.370662 h -1.28693 v 1.676397 h 0.62653 c 0.57573,0 0.77893,0.321733 0.762,1.100664 v 9.122827 c 0,1.219198 0.44027,1.710263 1.64253,1.862663 -1.20226,0.135466 -1.64253,0.626532 -1.64253,1.84573 v 9.122826 c 0.0169,0.660399 -0.23707,0.965198 -0.762,0.965198 h -0.62653 z"
+         id="text70"
+         aria-label="}"
+         sodipodi:nodetypes="ccccsccccccccscscscscc" />
+      <path
+         style="fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:1.25;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#ArrowWideHeavy)"
+         d="M 190.40199,70.278769 H 224.6663"
+         id="path70" />
+    </g>
+    <path
+       style="font-weight:bold;font-size:16.9333px;line-height:1.25;-inkscape-font-specification:'sans-serif Bold';text-align:center;letter-spacing:0px;word-spacing:0px;text-anchor:middle;stroke-width:0.0690111"
+       d="m -88.44114,74.337891 h 1.28693 c 1.20226,0.01693 2.09973,-1.015998 2.0828,-2.370662 V 62.57347 c -0.0169,-0.711199 0.33866,-1.066798 1.10066,-1.066798 h 0.2032 0.0677 v -1.557864 c -1.0668,0.01693 -1.38853,-0.270933 -1.37159,-1.236131 v -9.376826 c 0.0339,-1.354664 -0.88054,-2.387595 -2.0828,-2.370662 h -1.28693 v 1.676397 h 0.62653 c 0.57573,0 0.77893,0.321733 0.762,1.100664 v 9.122827 c 0,1.219198 0.44027,1.710263 1.64253,1.862663 -1.20226,0.135466 -1.64253,0.626532 -1.64253,1.84573 v 9.122826 c 0.0169,0.660399 -0.23707,0.965198 -0.762,0.965198 h -0.62653 z"
+       id="text70-1"
+       aria-label="}"
+       sodipodi:nodetypes="ccccsccccccccscscscscc" />
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-variant:normal;font-weight:normal;font-stretch:normal;font-size:16.9333px;font-family:Serif;-inkscape-font-specification:Serif;text-align:start;writing-mode:tb-rl;direction:ltr;text-orientation:upright;text-anchor:start;fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:1.25;stroke-dasharray:none;stroke-dashoffset:0"
+       x="200.88817"
+       y="38.990276"
+       id="text72"><tspan
+         sodipodi:role="line"
+         id="tspan72"
+         style="stroke-width:1.25"
+         x="200.88817"
+         y="38.990276" /></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:6.35px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="205.12828"
+       y="60.090775"
+       id="text73"><tspan
+         sodipodi:role="line"
+         x="205.12828"
+         y="60.090775"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.35px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan73">e.g.,:</tspan><tspan
+         sodipodi:role="line"
+         x="205.12828"
+         y="68.028275"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.35px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan74">queue[0]</tspan></text>
+    <g
+       id="g75"
+       transform="translate(0,40.745853)">
+      <path
+         style="font-weight:bold;font-size:16.9333px;line-height:1.25;-inkscape-font-specification:'sans-serif Bold';text-align:center;letter-spacing:0px;word-spacing:0px;text-anchor:middle;stroke-width:0.0690111"
+         d="m 185.24734,83.96512 h 1.28693 c 1.20226,0.01693 2.09973,-1.015998 2.0828,-2.370662 v -9.393759 c -0.0169,-0.711199 0.33866,-1.066798 1.10066,-1.066798 h 0.2032 0.0677 v -1.557864 c -1.0668,0.01693 -1.38853,-0.270933 -1.37159,-1.236131 V 58.96308 c 0.0339,-1.354664 -0.88054,-2.387595 -2.0828,-2.370662 h -1.28693 v 1.676397 h 0.62653 c 0.57573,0 0.77893,0.321733 0.762,1.100664 v 9.122827 c 0,1.219198 0.44027,1.710263 1.64253,1.862663 -1.20226,0.135466 -1.64253,0.626532 -1.64253,1.84573 v 9.122826 c 0.0169,0.660399 -0.23707,0.965198 -0.762,0.965198 h -0.62653 z"
+         id="path74"
+         aria-label="}"
+         sodipodi:nodetypes="ccccsccccccccscscscscc" />
+      <path
+         style="fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:1.25;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#ArrowWideHeavy)"
+         d="M 190.40199,70.278769 H 224.6663"
+         id="path75" />
+    </g>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:6.35px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="205.12828"
+       y="100.83664"
+       id="text76"><tspan
+         sodipodi:role="line"
+         x="205.12828"
+         y="100.83664"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.35px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan75">e.g.,:</tspan><tspan
+         sodipodi:role="line"
+         x="205.12828"
+         y="108.77414"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.35px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan76">queue[4]</tspan></text>
+    <g
+       id="g77"
+       transform="translate(0,85.725048)">
+      <path
+         style="font-weight:bold;font-size:16.9333px;line-height:1.25;-inkscape-font-specification:'sans-serif Bold';text-align:center;letter-spacing:0px;word-spacing:0px;text-anchor:middle;stroke-width:0.0690111"
+         d="m 185.24734,83.96512 h 1.28693 c 1.20226,0.01693 2.09973,-1.015998 2.0828,-2.370662 v -9.393759 c -0.0169,-0.711199 0.33866,-1.066798 1.10066,-1.066798 h 0.2032 0.0677 v -1.557864 c -1.0668,0.01693 -1.38853,-0.270933 -1.37159,-1.236131 V 58.96308 c 0.0339,-1.354664 -0.88054,-2.387595 -2.0828,-2.370662 h -1.28693 v 1.676397 h 0.62653 c 0.57573,0 0.77893,0.321733 0.762,1.100664 v 9.122827 c 0,1.219198 0.44027,1.710263 1.64253,1.862663 -1.20226,0.135466 -1.64253,0.626532 -1.64253,1.84573 v 9.122826 c 0.0169,0.660399 -0.23707,0.965198 -0.762,0.965198 h -0.62653 z"
+         id="path76"
+         aria-label="}"
+         sodipodi:nodetypes="ccccsccccccccscscscscc" />
+      <path
+         style="fill:#1a1a1a;fill-opacity:0;stroke:#000000;stroke-width:1.25;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#ArrowWideHeavy)"
+         d="M 190.40199,70.278769 H 224.6663"
+         id="path77" />
+    </g>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:6.35px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="205.12828"
+       y="145.81558"
+       id="text78"><tspan
+         sodipodi:role="line"
+         x="205.12828"
+         y="145.81558"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.35px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan77">e.g.,:</tspan><tspan
+         sodipodi:role="line"
+         x="205.12828"
+         y="153.75308"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.35px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan78">queue[n]</tspan></text>
+    <g
+       id="g81">
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-18.502264"
+         y="65.642387"
+         id="text79"><tspan
+           sodipodi:role="line"
+           x="-18.502264"
+           y="65.642387"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;fill:#800000;stroke-width:0.0690111"
+           id="tspan79">HQD</tspan></text>
+      <ellipse
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0"
+         id="path79"
+         cx="-18.407015"
+         cy="63.2188"
+         rx="13.317666"
+         ry="4.4124799" />
+      <path
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Triangle)"
+         d="M -4.6361224,63.2188 H 10.606988"
+         id="path80" />
+    </g>
+    <g
+       id="g82"
+       transform="translate(0,14.287503)">
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-18.502264"
+         y="65.642387"
+         id="text81"><tspan
+           sodipodi:role="line"
+           x="-18.502264"
+           y="65.642387"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;fill:#800000;stroke-width:0.0690111"
+           id="tspan81">HQD</tspan></text>
+      <ellipse
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0"
+         id="ellipse81"
+         cx="-18.407015"
+         cy="63.2188"
+         rx="13.317666"
+         ry="4.4124799" />
+      <path
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Triangle)"
+         d="M -4.6361224,63.2188 H 10.606988"
+         id="path81" />
+    </g>
+    <g
+       id="g83"
+       transform="translate(0,40.745853)">
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-18.502264"
+         y="65.642387"
+         id="text82"><tspan
+           sodipodi:role="line"
+           x="-18.502264"
+           y="65.642387"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;fill:#800000;stroke-width:0.0690111"
+           id="tspan82">HQD</tspan></text>
+      <ellipse
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0"
+         id="ellipse82"
+         cx="-18.407015"
+         cy="63.2188"
+         rx="13.317666"
+         ry="4.4124799" />
+      <path
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Triangle)"
+         d="M -4.6361224,63.2188 H 10.606988"
+         id="path82" />
+    </g>
+    <g
+       id="g84"
+       transform="translate(0,55.033362)">
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-18.502264"
+         y="65.642387"
+         id="text83"><tspan
+           sodipodi:role="line"
+           x="-18.502264"
+           y="65.642387"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;fill:#800000;stroke-width:0.0690111"
+           id="tspan83">HQD</tspan></text>
+      <ellipse
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0"
+         id="ellipse83"
+         cx="-18.407015"
+         cy="63.2188"
+         rx="13.317666"
+         ry="4.4124799" />
+      <path
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Triangle)"
+         d="M -4.6361224,63.2188 H 10.606988"
+         id="path83" />
+    </g>
+    <g
+       id="g85"
+       transform="translate(0,85.195881)">
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-18.502264"
+         y="65.642387"
+         id="text84"><tspan
+           sodipodi:role="line"
+           x="-18.502264"
+           y="65.642387"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;fill:#800000;stroke-width:0.0690111"
+           id="tspan84">HQD</tspan></text>
+      <ellipse
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0"
+         id="ellipse84"
+         cx="-18.407015"
+         cy="63.2188"
+         rx="13.317666"
+         ry="4.4124799" />
+      <path
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Triangle)"
+         d="M -4.6361224,63.2188 H 10.606988"
+         id="path84" />
+    </g>
+    <g
+       id="g86"
+       transform="translate(0,99.48339)">
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-18.502264"
+         y="65.642387"
+         id="text85"><tspan
+           sodipodi:role="line"
+           x="-18.502264"
+           y="65.642387"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;fill:#800000;stroke-width:0.0690111"
+           id="tspan85">HQD</tspan></text>
+      <ellipse
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0"
+         id="ellipse85"
+         cx="-18.407015"
+         cy="63.2188"
+         rx="13.317666"
+         ry="4.4124799" />
+      <path
+         style="fill:#800000;fill-opacity:0;stroke:#800000;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Triangle)"
+         d="M -4.6361224,63.2188 H 10.606988"
+         id="path85" />
+    </g>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:5.64444px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-35.2131"
+       y="54.673237"
+       id="text86"><tspan
+         sodipodi:role="line"
+         x="-35.2131"
+         y="54.673237"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.64444px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan86">Registers</tspan></text>
+    <path
+       style="fill:#800000;fill-opacity:0;stroke:#1a1a1a;stroke-width:1;stroke-dasharray:1, 1;stroke-dashoffset:0;marker-end:url(#ArrowWide)"
+       d="m -45.247972,57.442462 v 5.888987 h 11.344412"
+       id="path86"
+       sodipodi:nodetypes="ccc" />
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="82.441582"
+       id="text87"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="82.441582"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan87">MQD</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="99.240776"
+       id="text88"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="99.240776"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan88">MQD</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="116.03998"
+       id="text89"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="116.03998"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan89">MQD</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="132.83917"
+       id="text90"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="132.83917"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan90">MQD</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:7.05556px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="168.3002"
+       id="text91"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="168.3002"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:7.05556px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan91">MQD</tspan></text>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:10.5833px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-74.441521"
+       y="148.1461"
+       id="text92"><tspan
+         sodipodi:role="line"
+         x="-74.441521"
+         y="148.1461"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:10.5833px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;writing-mode:tb-rl;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan92">...</tspan></text>
+    <g
+       id="g97"
+       transform="translate(-5.8208336)">
+      <rect
+         style="fill:#aaffcc;fill-opacity:1;stroke:#1a1a1a;stroke-width:0.5;stroke-dasharray:none;stroke-dashoffset:0"
+         id="rect96"
+         width="58.726093"
+         height="27.598055"
+         x="-142.96434"
+         y="46.852512" />
+      <text
+         xml:space="preserve"
+         style="font-style:normal;font-weight:normal;font-size:4.23333px;line-height:1.25;font-family:sans-serif;text-align:start;letter-spacing:0px;word-spacing:0px;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+         x="-139.74622"
+         y="52.62756"
+         id="text93"><tspan
+           sodipodi:role="line"
+           x="-139.74622"
+           y="52.62756"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+           id="tspan93">HQD Registers</tspan><tspan
+           sodipodi:role="line"
+           x="-139.74622"
+           y="57.91922"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+           id="tspan95">Queue Address in the GPU</tspan><tspan
+           sodipodi:role="line"
+           x="-139.74622"
+           y="63.210884"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+           id="tspan96">Doorbell</tspan><tspan
+           sodipodi:role="line"
+           x="-139.74622"
+           y="68.502548"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+           id="tspan97">...</tspan><tspan
+           sodipodi:role="line"
+           x="-139.74622"
+           y="73.794212"
+           style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+           id="tspan94" /></text>
+    </g>
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:4.23333px;line-height:1.25;font-family:sans-serif;text-align:start;letter-spacing:0px;word-spacing:0px;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-61.293022"
+       y="-19.380915"
+       id="text93-1"><tspan
+         sodipodi:role="line"
+         x="-61.293022"
+         y="-19.380915"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan104">SWITCH QUEUE:</tspan><tspan
+         sodipodi:role="line"
+         x="-61.293022"
+         y="-14.089252"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+         id="tspan101">WAIT FOR HQD_ACTIVE = 0</tspan><tspan
+         sodipodi:role="line"
+         x="-61.293022"
+         y="-8.7975903"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+         id="tspan105">SAVE QUEUE STATE TO THE MQD</tspan><tspan
+         sodipodi:role="line"
+         x="-61.293022"
+         y="-3.505928"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+         id="tspan102">COPY NEW MQD STATE</tspan><tspan
+         sodipodi:role="line"
+         x="-61.293022"
+         y="1.7857342"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:4.23333px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:start;text-anchor:start;stroke-width:0.0690111"
+         id="tspan103">SET HQD_ACTIVE = 1</tspan></text>
+    <circle
+       style="fill:#ffeeaa;fill-opacity:1;stroke:#1a1a1a;stroke-width:1.88976;stroke-dasharray:none;stroke-dashoffset:0"
+       id="path97"
+       cx="0"
+       cy="0"
+       r="0"
+       transform="matrix(0.26458333,0,0,0.26458333,-149.03517,37.347779)" />
+    <text
+       xml:space="preserve"
+       style="font-style:normal;font-weight:normal;font-size:5.64444px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.0690111"
+       x="-36.659206"
+       y="-44.828983"
+       id="text106"><tspan
+         sodipodi:role="line"
+         x="-36.659206"
+         y="-44.828983"
+         style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.64444px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';text-align:center;text-anchor:middle;stroke-width:0.0690111"
+         id="tspan106">Firmware</tspan></text>
+    <path
+       style="fill:none;fill-opacity:1;stroke:#1a1a1a;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-start:url(#marker109);marker-end:url(#ArrowTriangleStylized)"
+       d="M -84.242601,-9.1838245 H -98.041629 V 33.17598"
+       id="path106" />
+    <path
+       style="fill:none;fill-opacity:1;stroke:#1a1a1a;stroke-width:1;stroke-dasharray:none;stroke-dashoffset:0;marker-end:url(#Dot)"
+       d="M 9.9745536,-9.3442784 H 29.549918 V 37.170287"
+       id="path107"
+       sodipodi:nodetypes="ccc" />
+  </g>
+</svg>
diff --git a/Documentation/gpu/automated_testing.rst b/Documentation/gpu/automated_testing.rst
index 6d7c6086034d..62aa3ede02a5 100644
--- a/Documentation/gpu/automated_testing.rst
+++ b/Documentation/gpu/automated_testing.rst
@@ -115,6 +115,10 @@ created (eg. https://gitlab.freedesktop.org/janedoe/linux/-/pipelines)
 5. The various jobs will be run and when the pipeline is finished, all jobs
 should be green unless a regression has been found.
 
+6. Warnings in the pipeline indicate that lockdep
+(see Documentation/locking/lockdep-design.rst) issues have been detected
+during the tests.
+
 
 How to update test expectations
 ===============================
diff --git a/Documentation/gpu/driver-uapi.rst b/Documentation/gpu/driver-uapi.rst
index 971cdb4816fc..1f15a8ca1265 100644
--- a/Documentation/gpu/driver-uapi.rst
+++ b/Documentation/gpu/driver-uapi.rst
@@ -27,3 +27,8 @@ drm/xe uAPI
 ===========
 
 .. kernel-doc:: include/uapi/drm/xe_drm.h
+
+drm/asahi uAPI
+================
+
+.. kernel-doc:: include/uapi/drm/asahi_drm.h
diff --git a/Documentation/gpu/drm-kms-helpers.rst b/Documentation/gpu/drm-kms-helpers.rst
index b4ee25af1702..5139705089f2 100644
--- a/Documentation/gpu/drm-kms-helpers.rst
+++ b/Documentation/gpu/drm-kms-helpers.rst
@@ -233,6 +233,21 @@ Panel Self Refresh Helper Reference
 .. kernel-doc:: drivers/gpu/drm/drm_self_refresh_helper.c
    :export:
 
+HDMI Atomic State Helpers
+=========================
+
+Overview
+--------
+
+.. kernel-doc:: drivers/gpu/drm/display/drm_hdmi_state_helper.c
+   :doc: hdmi helpers
+
+Functions Reference
+-------------------
+
+.. kernel-doc:: drivers/gpu/drm/display/drm_hdmi_state_helper.c
+   :export:
+
 HDCP Helper Functions Reference
 ===============================
 
diff --git a/Documentation/gpu/nouveau.rst b/Documentation/gpu/nouveau.rst
index 0f34131ccc27..b8c801e0068c 100644
--- a/Documentation/gpu/nouveau.rst
+++ b/Documentation/gpu/nouveau.rst
@@ -27,3 +27,6 @@ GSP Support
 
 .. kernel-doc:: drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c
    :doc: GSP message queue element
+
+.. kernel-doc:: drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
+   :doc: GSP message handling policy
diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst
index 256d0d1cb216..c57777a24e03 100644
--- a/Documentation/gpu/todo.rst
+++ b/Documentation/gpu/todo.rst
@@ -441,14 +441,15 @@ Contact: Thomas Zimmermann <tzimmermann@suse.de>
 
 Level: Intermediate
 
-Request memory regions in all drivers
--------------------------------------
+Request memory regions in all fbdev drivers
+--------------------------------------------
 
-Go through all drivers and add code to request the memory regions that the
-driver uses. This requires adding calls to request_mem_region(),
+Old/ancient fbdev drivers do not request their memory properly.
+Go through these drivers and add code to request the memory regions
+that the driver uses. This requires adding calls to request_mem_region(),
 pci_request_region() or similar functions. Use helpers for managed cleanup
-where possible.
-
+where possible. Problematic areas include hardware that has exclusive ranges
+like VGA. VGA16fb does not request the range as it is expected.
 Drivers are pretty bad at doing this and there used to be conflicts among
 DRM and fbdev drivers. Still, it's the correct thing to do.
 
diff --git a/Documentation/gpu/vgaarbiter.rst b/Documentation/gpu/vgaarbiter.rst
index bde3c0afb059..d1e953712cc2 100644
--- a/Documentation/gpu/vgaarbiter.rst
+++ b/Documentation/gpu/vgaarbiter.rst
@@ -11,9 +11,9 @@ Section 7, Legacy Devices.
 
 The Resource Access Control (RAC) module inside the X server [0] existed for
 the legacy VGA arbitration task (besides other bus management tasks) when more
-than one legacy device co-exists on the same machine. But the problem happens
+than one legacy device co-exist on the same machine. But the problem happens
 when these devices are trying to be accessed by different userspace clients
-(e.g. two server in parallel). Their address assignments conflict. Moreover,
+(e.g. two servers in parallel). Their address assignments conflict. Moreover,
 ideally, being a userspace application, it is not the role of the X server to
 control bus resources. Therefore an arbitration scheme outside of the X server
 is needed to control the sharing of these resources. This document introduces
@@ -106,7 +106,7 @@ In-kernel interface
 libpciaccess
 ------------
 
-To use the vga arbiter char device it was implemented an API inside the
+To use the vga arbiter char device, an API was implemented inside the
 libpciaccess library. One field was added to struct pci_device (each device
 on the system)::
 
diff --git a/Documentation/gpu/xe/index.rst b/Documentation/gpu/xe/index.rst
index 92cfb25e64d3..b2369561f24e 100644
--- a/Documentation/gpu/xe/index.rst
+++ b/Documentation/gpu/xe/index.rst
@@ -25,3 +25,4 @@ DG2, etc is provided to prototype the driver.
    xe_debugging
    xe_devcoredump
    xe-drm-usage-stats.rst
+   xe_configfs
diff --git a/Documentation/gpu/xe/xe_configfs.rst b/Documentation/gpu/xe/xe_configfs.rst
new file mode 100644
index 000000000000..9b9d941eb20e
--- /dev/null
+++ b/Documentation/gpu/xe/xe_configfs.rst
@@ -0,0 +1,10 @@
+.. SPDX-License-Identifier: GPL-2.0+
+
+.. _xe_configfs:
+
+============
+Xe Configfs
+============
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_configfs.c
+   :doc: Xe Configfs
diff --git a/Documentation/gpu/xe/xe_firmware.rst b/Documentation/gpu/xe/xe_firmware.rst
index afcb561cd37d..5d23e9f27391 100644
--- a/Documentation/gpu/xe/xe_firmware.rst
+++ b/Documentation/gpu/xe/xe_firmware.rst
@@ -31,6 +31,12 @@ GuC Power Conservation (PC)
 .. kernel-doc:: drivers/gpu/drm/xe/xe_guc_pc.c
    :doc: GuC Power Conservation (PC)
 
+PCIe Gen5 Limitations
+=====================
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_device_sysfs.c
+   :doc: PCIe Gen5 Limitations
+
 Internal API
 ============
 
diff --git a/Documentation/gpu/xe/xe_pcode.rst b/Documentation/gpu/xe/xe_pcode.rst
index d2e22cc45061..5937ef3599b0 100644
--- a/Documentation/gpu/xe/xe_pcode.rst
+++ b/Documentation/gpu/xe/xe_pcode.rst
@@ -12,3 +12,10 @@ Internal API
 
 .. kernel-doc:: drivers/gpu/drm/xe/xe_pcode.c
    :internal:
+
+==================
+Boot Survivability
+==================
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_survivability_mode.c
+   :doc: Xe Boot Survivability
diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 655d8d10fe24..c650cd3dcb80 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -89,8 +89,10 @@ definitions:
           doc: Group of short_detected states
   -
     name: phy-upstream-type
-    enum-name:
+    enum-name: phy-upstream
+    header: linux/ethtool.h
     type: enum
+    name-prefix: phy-upstream
     entries: [ mac, phy ]
   -
     name: tcp-data-split
diff --git a/Documentation/netlink/specs/ovs_vport.yaml b/Documentation/netlink/specs/ovs_vport.yaml
index 86ba9ac2a521..b538bb99ee9b 100644
--- a/Documentation/netlink/specs/ovs_vport.yaml
+++ b/Documentation/netlink/specs/ovs_vport.yaml
@@ -123,12 +123,12 @@ attribute-sets:
 
 operations:
   name-prefix: ovs-vport-cmd-
+  fixed-header: ovs-header
   list:
     -
       name: new
       doc: Create a new OVS vport
       attribute-set: vport
-      fixed-header: ovs-header
       do:
         request:
           attributes:
@@ -141,7 +141,6 @@ operations:
       name: del
       doc: Delete existing OVS vport from a data path
       attribute-set: vport
-      fixed-header: ovs-header
       do:
         request:
           attributes:
@@ -152,7 +151,6 @@ operations:
       name: get
       doc: Get / dump OVS vport configuration and state
       attribute-set: vport
-      fixed-header: ovs-header
       do: &vport-get-op
         request:
           attributes:
diff --git a/Documentation/netlink/specs/rt_link.yaml b/Documentation/netlink/specs/rt_link.yaml
index 31238455f8e9..6b9d5ee87d93 100644
--- a/Documentation/netlink/specs/rt_link.yaml
+++ b/Documentation/netlink/specs/rt_link.yaml
@@ -1113,11 +1113,10 @@ attribute-sets:
       -
         name: prop-list
         type: nest
-        nested-attributes: link-attrs
+        nested-attributes: prop-list-link-attrs
       -
         name: alt-ifname
         type: string
-        multi-attr: true
       -
         name: perm-address
         type: binary
@@ -1164,6 +1163,13 @@ attribute-sets:
         name: netns-immutable
         type: u8
   -
+    name: prop-list-link-attrs
+    subset-of: link-attrs
+    attributes:
+      -
+        name: alt-ifname
+        multi-attr: true
+  -
     name: af-spec-attrs
     attributes:
       -
@@ -1585,7 +1591,7 @@ attribute-sets:
         name: nf-call-iptables
         type: u8
       -
-        name: nf-call-ip6-tables
+        name: nf-call-ip6tables
         type: u8
       -
         name: nf-call-arptables
@@ -2077,7 +2083,7 @@ attribute-sets:
         name: id
         type: u16
       -
-        name: flag
+        name: flags
         type: binary
         struct: ifla-vlan-flags
       -
@@ -2165,7 +2171,7 @@ attribute-sets:
         type: binary
         struct: ifla-cacheinfo
       -
-        name: icmp6-stats
+        name: icmp6stats
         type: binary
         struct: ifla-icmp6-stats
       -
@@ -2179,9 +2185,10 @@ attribute-sets:
         type: u32
   -
     name: mctp-attrs
+    name-prefix: ifla-mctp-
     attributes:
       -
-        name: mctp-net
+        name: net
         type: u32
       -
         name: phys-binding
@@ -2453,7 +2460,6 @@ operations:
             - min-mtu
             - max-mtu
             - prop-list
-            - alt-ifname
             - perm-address
             - proto-down-reason
             - parent-dev-name
diff --git a/Documentation/netlink/specs/rt_neigh.yaml b/Documentation/netlink/specs/rt_neigh.yaml
index e670b6dc07be..a843caa72259 100644
--- a/Documentation/netlink/specs/rt_neigh.yaml
+++ b/Documentation/netlink/specs/rt_neigh.yaml
@@ -13,25 +13,25 @@ definitions:
     type: struct
     members:
       -
-        name: family
+        name: ndm-family
         type: u8
       -
-        name: pad
+        name: ndm-pad
         type: pad
         len: 3
       -
-        name: ifindex
+        name: ndm-ifindex
         type: s32
       -
-        name: state
+        name: ndm-state
         type: u16
         enum: nud-state
       -
-        name: flags
+        name: ndm-flags
         type: u8
         enum: ntf-flags
       -
-        name: type
+        name: ndm-type
         type: u8
         enum: rtm-type
   -
@@ -189,7 +189,7 @@ attribute-sets:
         type: binary
         display-hint: ipv4
       -
-        name: lladr
+        name: lladdr
         type: binary
         display-hint: mac
       -
diff --git a/Documentation/networking/netdevices.rst b/Documentation/networking/netdevices.rst
index 6c2d8945f597..eab601ab2db0 100644
--- a/Documentation/networking/netdevices.rst
+++ b/Documentation/networking/netdevices.rst
@@ -338,10 +338,11 @@ operations directly under the netdev instance lock.
 Devices drivers are encouraged to rely on the instance lock where possible.
 
 For the (mostly software) drivers that need to interact with the core stack,
-there are two sets of interfaces: ``dev_xxx`` and ``netif_xxx`` (e.g.,
-``dev_set_mtu`` and ``netif_set_mtu``). The ``dev_xxx`` functions handle
-acquiring the instance lock themselves, while the ``netif_xxx`` functions
-assume that the driver has already acquired the instance lock.
+there are two sets of interfaces: ``dev_xxx``/``netdev_xxx`` and ``netif_xxx``
+(e.g., ``dev_set_mtu`` and ``netif_set_mtu``). The ``dev_xxx``/``netdev_xxx``
+functions handle acquiring the instance lock themselves, while the
+``netif_xxx`` functions assume that the driver has already acquired
+the instance lock.
 
 Notifiers and netdev instance lock
 ==================================
@@ -354,6 +355,7 @@ For devices with locked ops, currently only the following notifiers are
 running under the lock:
 * ``NETDEV_REGISTER``
 * ``NETDEV_UP``
+* ``NETDEV_CHANGE``
 
 The following notifiers are running without the lock:
 * ``NETDEV_UNREGISTER``
diff --git a/Documentation/power/runtime_pm.rst b/Documentation/power/runtime_pm.rst
index 12f429359a82..63344bea8393 100644
--- a/Documentation/power/runtime_pm.rst
+++ b/Documentation/power/runtime_pm.rst
@@ -154,7 +154,7 @@ suspending the device are satisfied) and to queue up a suspend request for the
 device in that case.  If there is no idle callback, or if the callback returns
 0, then the PM core will attempt to carry out a runtime suspend of the device,
 also respecting devices configured for autosuspend.  In essence this means a
-call to __pm_runtime_autosuspend() (do note that drivers needs to update the
+call to pm_runtime_autosuspend() (do note that drivers needs to update the
 device last busy mark, pm_runtime_mark_last_busy(), to control the delay under
 this circumstance).  To prevent this (for example, if the callback routine has
 started a delayed suspend), the routine must return a non-zero value.  Negative
diff --git a/Documentation/translations/zh_CN/arch/openrisc/openrisc_port.rst b/Documentation/translations/zh_CN/arch/openrisc/openrisc_port.rst
index cadc580fa23b..d728e4db0b85 100644
--- a/Documentation/translations/zh_CN/arch/openrisc/openrisc_port.rst
+++ b/Documentation/translations/zh_CN/arch/openrisc/openrisc_port.rst
@@ -17,10 +17,10 @@ OpenRISC 1000系列（或1k）。
 
 关于OpenRISC处理器和正在进行中的开发的信息:
 
-	=======		=============================
+	=======		==============================
 	网站		https://openrisc.io
-	邮箱		openrisc@lists.librecores.org
-	=======		=============================
+	邮箱		linux-openrisc@vger.kernel.org
+	=======		==============================
 
 ---------------------------------------------------------------------
 
@@ -36,11 +36,11 @@ OpenRISC工具链和Linux的构建指南
 工具链的构建指南可以在openrisc.io或Stafford的工具链构建和发布脚本
 中找到。
 
-	======      =================================================
-	二进制      https://github.com/openrisc/or1k-gcc/releases
+	======      ==========================================================
+	二进制      https://github.com/stffrdhrn/or1k-toolchain-build/releases
 	工具链      https://openrisc.io/software
 	构建        https://github.com/stffrdhrn/or1k-toolchain-build
-	======      =================================================
+	======      ==========================================================
 
 2) 构建
 
diff --git a/Documentation/translations/zh_TW/arch/openrisc/openrisc_port.rst b/Documentation/translations/zh_TW/arch/openrisc/openrisc_port.rst
index 422fe9f7a3f2..a1e4517dc601 100644
--- a/Documentation/translations/zh_TW/arch/openrisc/openrisc_port.rst
+++ b/Documentation/translations/zh_TW/arch/openrisc/openrisc_port.rst
@@ -17,10 +17,10 @@ OpenRISC 1000系列（或1k）。
 
 關於OpenRISC處理器和正在進行中的開發的信息:
 
-	=======		=============================
+	=======		==============================
 	網站		https://openrisc.io
-	郵箱		openrisc@lists.librecores.org
-	=======		=============================
+	郵箱		linux-openrisc@vger.kernel.org
+	=======		==============================
 
 ---------------------------------------------------------------------
 
@@ -36,11 +36,11 @@ OpenRISC工具鏈和Linux的構建指南
 工具鏈的構建指南可以在openrisc.io或Stafford的工具鏈構建和發佈腳本
 中找到。
 
-	======      =================================================
-	二進制      https://github.com/openrisc/or1k-gcc/releases
+	======      ==========================================================
+	二進制      https://github.com/stffrdhrn/or1k-toolchain-build/releases
 	工具鏈      https://openrisc.io/software
 	構建        https://github.com/stffrdhrn/or1k-toolchain-build
-	======      =================================================
+	======      ==========================================================
 
 2) 構建
 
diff --git a/Documentation/userspace-api/mseal.rst b/Documentation/userspace-api/mseal.rst
index 1dabfc29be0d..7195a7f91107 100644
--- a/Documentation/userspace-api/mseal.rst
+++ b/Documentation/userspace-api/mseal.rst
@@ -27,7 +27,7 @@ SYSCALL
 =======
 mseal syscall signature
 -----------------------
-   ``int mseal(void \* addr, size_t len, unsigned long flags)``
+   ``int mseal(void *addr, size_t len, unsigned long flags)``
 
    **addr**/**len**: virtual memory address range.
       The address range set by **addr**/**len** must meet:
diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
index 1f8625b7646a..47c7c3f92314 100644
--- a/Documentation/virt/kvm/api.rst
+++ b/Documentation/virt/kvm/api.rst
@@ -7447,6 +7447,75 @@ Unused bitfields in the bitarrays must be set to zero.
 
 This capability connects the vcpu to an in-kernel XIVE device.
 
+6.76 KVM_CAP_HYPERV_SYNIC
+-------------------------
+
+:Architectures: x86
+:Target: vcpu
+
+This capability, if KVM_CHECK_EXTENSION indicates that it is
+available, means that the kernel has an implementation of the
+Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
+used to support Windows Hyper-V based guest paravirt drivers(VMBus).
+
+In order to use SynIC, it has to be activated by setting this
+capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
+will disable the use of APIC hardware virtualization even if supported
+by the CPU, as it's incompatible with SynIC auto-EOI behavior.
+
+6.77 KVM_CAP_HYPERV_SYNIC2
+--------------------------
+
+:Architectures: x86
+:Target: vcpu
+
+This capability enables a newer version of Hyper-V Synthetic interrupt
+controller (SynIC).  The only difference with KVM_CAP_HYPERV_SYNIC is that KVM
+doesn't clear SynIC message and event flags pages when they are enabled by
+writing to the respective MSRs.
+
+6.78 KVM_CAP_HYPERV_DIRECT_TLBFLUSH
+-----------------------------------
+
+:Architectures: x86
+:Target: vcpu
+
+This capability indicates that KVM running on top of Hyper-V hypervisor
+enables Direct TLB flush for its guests meaning that TLB flush
+hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM.
+Due to the different ABI for hypercall parameters between Hyper-V and
+KVM, enabling this capability effectively disables all hypercall
+handling by KVM (as some KVM hypercall may be mistakenly treated as TLB
+flush hypercalls by Hyper-V) so userspace should disable KVM identification
+in CPUID and only exposes Hyper-V identification. In this case, guest
+thinks it's running on Hyper-V and only use Hyper-V hypercalls.
+
+6.79 KVM_CAP_HYPERV_ENFORCE_CPUID
+---------------------------------
+
+:Architectures: x86
+:Target: vcpu
+
+When enabled, KVM will disable emulated Hyper-V features provided to the
+guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all
+currently implemented Hyper-V features are provided unconditionally when
+Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001)
+leaf.
+
+6.80 KVM_CAP_ENFORCE_PV_FEATURE_CPUID
+-------------------------------------
+
+:Architectures: x86
+:Target: vcpu
+
+When enabled, KVM will disable paravirtual features provided to the
+guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf
+(0x40000001). Otherwise, a guest may use the paravirtual features
+regardless of what has actually been exposed through the CPUID leaf.
+
+.. _KVM_CAP_DIRTY_LOG_RING:
+
+
 .. _cap_enable_vm:
 
 7. Capabilities that can be enabled on VMs
@@ -7927,10 +7996,10 @@ by POWER10 processor.
 7.24 KVM_CAP_VM_COPY_ENC_CONTEXT_FROM
 -------------------------------------
 
-Architectures: x86 SEV enabled
-Type: vm
-Parameters: args[0] is the fd of the source vm
-Returns: 0 on success; ENOTTY on error
+:Architectures: x86 SEV enabled
+:Type: vm
+:Parameters: args[0] is the fd of the source vm
+:Returns: 0 on success; ENOTTY on error
 
 This capability enables userspace to copy encryption context from the vm
 indicated by the fd to the vm this is called on.
@@ -7963,24 +8032,6 @@ default.
 
 See Documentation/arch/x86/sgx.rst for more details.
 
-7.26 KVM_CAP_PPC_RPT_INVALIDATE
--------------------------------
-
-:Capability: KVM_CAP_PPC_RPT_INVALIDATE
-:Architectures: ppc
-:Type: vm
-
-This capability indicates that the kernel is capable of handling
-H_RPT_INVALIDATE hcall.
-
-In order to enable the use of H_RPT_INVALIDATE in the guest,
-user space might have to advertise it for the guest. For example,
-IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is
-present in the "ibm,hypertas-functions" device-tree property.
-
-This capability is enabled for hypervisors on platforms like POWER9
-that support radix MMU.
-
 7.27 KVM_CAP_EXIT_ON_EMULATION_FAILURE
 --------------------------------------
 
@@ -8038,24 +8089,9 @@ indicated by the fd to the VM this is called on.
 This is intended to support intra-host migration of VMs between userspace VMMs,
 upgrading the VMM process without interrupting the guest.
 
-7.30 KVM_CAP_PPC_AIL_MODE_3
--------------------------------
-
-:Capability: KVM_CAP_PPC_AIL_MODE_3
-:Architectures: ppc
-:Type: vm
-
-This capability indicates that the kernel supports the mode 3 setting for the
-"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location"
-resource that is controlled with the H_SET_MODE hypercall.
-
-This capability allows a guest kernel to use a better-performance mode for
-handling interrupts and system calls.
-
 7.31 KVM_CAP_DISABLE_QUIRKS2
 ----------------------------
 
-:Capability: KVM_CAP_DISABLE_QUIRKS2
 :Parameters: args[0] - set of KVM quirks to disable
 :Architectures: x86
 :Type: vm
@@ -8210,27 +8246,6 @@ This capability is aimed to mitigate the threat that malicious VMs can
 cause CPU stuck (due to event windows don't open up) and make the CPU
 unavailable to host or other VMs.
 
-7.34 KVM_CAP_MEMORY_FAULT_INFO
-------------------------------
-
-:Architectures: x86
-:Returns: Informational only, -EINVAL on direct KVM_ENABLE_CAP.
-
-The presence of this capability indicates that KVM_RUN will fill
-kvm_run.memory_fault if KVM cannot resolve a guest page fault VM-Exit, e.g. if
-there is a valid memslot but no backing VMA for the corresponding host virtual
-address.
-
-The information in kvm_run.memory_fault is valid if and only if KVM_RUN returns
-an error with errno=EFAULT or errno=EHWPOISON *and* kvm_run.exit_reason is set
-to KVM_EXIT_MEMORY_FAULT.
-
-Note: Userspaces which attempt to resolve memory faults so that they can retry
-KVM_RUN are encouraged to guard against repeatedly receiving the same
-error/annotated fault.
-
-See KVM_EXIT_MEMORY_FAULT for more information.
-
 7.35 KVM_CAP_X86_APIC_BUS_CYCLES_NS
 -----------------------------------
 
@@ -8248,19 +8263,220 @@ by KVM_CHECK_EXTENSION.
 Note: Userspace is responsible for correctly configuring CPUID 0x15, a.k.a. the
 core crystal clock frequency, if a non-zero CPUID 0x15 is exposed to the guest.
 
-7.36 KVM_CAP_X86_GUEST_MODE
-------------------------------
+7.36 KVM_CAP_DIRTY_LOG_RING/KVM_CAP_DIRTY_LOG_RING_ACQ_REL
+----------------------------------------------------------
+
+:Architectures: x86, arm64
+:Type: vm
+:Parameters: args[0] - size of the dirty log ring
+
+KVM is capable of tracking dirty memory using ring buffers that are
+mmapped into userspace; there is one dirty ring per vcpu.
+
+The dirty ring is available to userspace as an array of
+``struct kvm_dirty_gfn``.  Each dirty entry is defined as::
+
+  struct kvm_dirty_gfn {
+          __u32 flags;
+          __u32 slot; /* as_id | slot_id */
+          __u64 offset;
+  };
+
+The following values are defined for the flags field to define the
+current state of the entry::
+
+  #define KVM_DIRTY_GFN_F_DIRTY           BIT(0)
+  #define KVM_DIRTY_GFN_F_RESET           BIT(1)
+  #define KVM_DIRTY_GFN_F_MASK            0x3
+
+Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM
+ioctl to enable this capability for the new guest and set the size of
+the rings.  Enabling the capability is only allowed before creating any
+vCPU, and the size of the ring must be a power of two.  The larger the
+ring buffer, the less likely the ring is full and the VM is forced to
+exit to userspace. The optimal size depends on the workload, but it is
+recommended that it be at least 64 KiB (4096 entries).
+
+Just like for dirty page bitmaps, the buffer tracks writes to
+all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was
+set in KVM_SET_USER_MEMORY_REGION.  Once a memory region is registered
+with the flag set, userspace can start harvesting dirty pages from the
+ring buffer.
+
+An entry in the ring buffer can be unused (flag bits ``00``),
+dirty (flag bits ``01``) or harvested (flag bits ``1X``).  The
+state machine for the entry is as follows::
+
+          dirtied         harvested        reset
+     00 -----------> 01 -------------> 1X -------+
+      ^                                          |
+      |                                          |
+      +------------------------------------------+
+
+To harvest the dirty pages, userspace accesses the mmapped ring buffer
+to read the dirty GFNs.  If the flags has the DIRTY bit set (at this stage
+the RESET bit must be cleared), then it means this GFN is a dirty GFN.
+The userspace should harvest this GFN and mark the flags from state
+``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set
+to show that this GFN is harvested and waiting for a reset), and move
+on to the next GFN.  The userspace should continue to do this until the
+flags of a GFN have the DIRTY bit cleared, meaning that it has harvested
+all the dirty GFNs that were available.
+
+Note that on weakly ordered architectures, userspace accesses to the
+ring buffer (and more specifically the 'flags' field) must be ordered,
+using load-acquire/store-release accessors when available, or any
+other memory barrier that will ensure this ordering.
+
+It's not necessary for userspace to harvest the all dirty GFNs at once.
+However it must collect the dirty GFNs in sequence, i.e., the userspace
+program cannot skip one dirty GFN to collect the one next to it.
+
+After processing one or more entries in the ring buffer, userspace
+calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about
+it, so that the kernel will reprotect those collected GFNs.
+Therefore, the ioctl must be called *before* reading the content of
+the dirty pages.
+
+The dirty ring can get full.  When it happens, the KVM_RUN of the
+vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL.
+
+The dirty ring interface has a major difference comparing to the
+KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from
+userspace, it's still possible that the kernel has not yet flushed the
+processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the
+flushing is done by the KVM_GET_DIRTY_LOG ioctl).  To achieve that, one
+needs to kick the vcpu out of KVM_RUN using a signal.  The resulting
+vmexit ensures that all dirty GFNs are flushed to the dirty rings.
+
+NOTE: KVM_CAP_DIRTY_LOG_RING_ACQ_REL is the only capability that
+should be exposed by weakly ordered architecture, in order to indicate
+the additional memory ordering requirements imposed on userspace when
+reading the state of an entry and mutating it from DIRTY to HARVESTED.
+Architecture with TSO-like ordering (such as x86) are allowed to
+expose both KVM_CAP_DIRTY_LOG_RING and KVM_CAP_DIRTY_LOG_RING_ACQ_REL
+to userspace.
+
+After enabling the dirty rings, the userspace needs to detect the
+capability of KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP to see whether the
+ring structures can be backed by per-slot bitmaps. With this capability
+advertised, it means the architecture can dirty guest pages without
+vcpu/ring context, so that some of the dirty information will still be
+maintained in the bitmap structure. KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP
+can't be enabled if the capability of KVM_CAP_DIRTY_LOG_RING_ACQ_REL
+hasn't been enabled, or any memslot has been existing.
+
+Note that the bitmap here is only a backup of the ring structure. The
+use of the ring and bitmap combination is only beneficial if there is
+only a very small amount of memory that is dirtied out of vcpu/ring
+context. Otherwise, the stand-alone per-slot bitmap mechanism needs to
+be considered.
+
+To collect dirty bits in the backup bitmap, userspace can use the same
+KVM_GET_DIRTY_LOG ioctl. KVM_CLEAR_DIRTY_LOG isn't needed as long as all
+the generation of the dirty bits is done in a single pass. Collecting
+the dirty bitmap should be the very last thing that the VMM does before
+considering the state as complete. VMM needs to ensure that the dirty
+state is final and avoid missing dirty pages from another ioctl ordered
+after the bitmap collection.
+
+NOTE: Multiple examples of using the backup bitmap: (1) save vgic/its
+tables through command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} on
+KVM device "kvm-arm-vgic-its". (2) restore vgic/its tables through
+command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} on KVM device
+"kvm-arm-vgic-its". VGICv3 LPI pending status is restored. (3) save
+vgic3 pending table through KVM_DEV_ARM_VGIC_{GRP_CTRL, SAVE_PENDING_TABLES}
+command on KVM device "kvm-arm-vgic-v3".
+
+7.37 KVM_CAP_PMU_CAPABILITY
+---------------------------
 
 :Architectures: x86
-:Returns: Informational only, -EINVAL on direct KVM_ENABLE_CAP.
+:Type: vm
+:Parameters: arg[0] is bitmask of PMU virtualization capabilities.
+:Returns: 0 on success, -EINVAL when arg[0] contains invalid bits
 
-The presence of this capability indicates that KVM_RUN will update the
-KVM_RUN_X86_GUEST_MODE bit in kvm_run.flags to indicate whether the
-vCPU was executing nested guest code when it exited.
+This capability alters PMU virtualization in KVM.
 
-KVM exits with the register state of either the L1 or L2 guest
-depending on which executed at the time of an exit. Userspace must
-take care to differentiate between these cases.
+Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of
+PMU virtualization capabilities that can be adjusted on a VM.
+
+The argument to KVM_ENABLE_CAP is also a bitmask and selects specific
+PMU virtualization capabilities to be applied to the VM.  This can
+only be invoked on a VM prior to the creation of VCPUs.
+
+At this time, KVM_PMU_CAP_DISABLE is the only capability.  Setting
+this capability will disable PMU virtualization for that VM.  Usermode
+should adjust CPUID leaf 0xA to reflect that the PMU is disabled.
+
+7.38 KVM_CAP_VM_DISABLE_NX_HUGE_PAGES
+-------------------------------------
+
+:Architectures: x86
+:Type: vm
+:Parameters: arg[0] must be 0.
+:Returns: 0 on success, -EPERM if the userspace process does not
+          have CAP_SYS_BOOT, -EINVAL if args[0] is not 0 or any vCPUs have been
+          created.
+
+This capability disables the NX huge pages mitigation for iTLB MULTIHIT.
+
+The capability has no effect if the nx_huge_pages module parameter is not set.
+
+This capability may only be set before any vCPUs are created.
+
+7.39 KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE
+---------------------------------------
+
+:Architectures: arm64
+:Type: vm
+:Parameters: arg[0] is the new split chunk size.
+:Returns: 0 on success, -EINVAL if any memslot was already created.
+
+This capability sets the chunk size used in Eager Page Splitting.
+
+Eager Page Splitting improves the performance of dirty-logging (used
+in live migrations) when guest memory is backed by huge-pages.  It
+avoids splitting huge-pages (into PAGE_SIZE pages) on fault, by doing
+it eagerly when enabling dirty logging (with the
+KVM_MEM_LOG_DIRTY_PAGES flag for a memory region), or when using
+KVM_CLEAR_DIRTY_LOG.
+
+The chunk size specifies how many pages to break at a time, using a
+single allocation for each chunk. Bigger the chunk size, more pages
+need to be allocated ahead of time.
+
+The chunk size needs to be a valid block size. The list of acceptable
+block sizes is exposed in KVM_CAP_ARM_SUPPORTED_BLOCK_SIZES as a
+64-bit bitmap (each bit describing a block size). The default value is
+0, to disable the eager page splitting.
+
+7.40 KVM_CAP_EXIT_HYPERCALL
+---------------------------
+
+:Architectures: x86
+:Type: vm
+
+This capability, if enabled, will cause KVM to exit to userspace
+with KVM_EXIT_HYPERCALL exit reason to process some hypercalls.
+
+Calling KVM_CHECK_EXTENSION for this capability will return a bitmask
+of hypercalls that can be configured to exit to userspace.
+Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE.
+
+The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset
+of the result of KVM_CHECK_EXTENSION.  KVM will forward to userspace
+the hypercalls whose corresponding bit is in the argument, and return
+ENOSYS for the others.
+
+7.41 KVM_CAP_ARM_SYSTEM_SUSPEND
+-------------------------------
+
+:Architectures: arm64
+:Type: vm
+
+When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of
+type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request.
 
 7.37 KVM_CAP_ARM_WRITABLE_IMP_ID_REGS
 -------------------------------------
@@ -8297,21 +8513,6 @@ H_RANDOM hypercall backed by a hardware random-number generator.
 If present, the kernel H_RANDOM handler can be enabled for guest use
 with the KVM_CAP_PPC_ENABLE_HCALL capability.
 
-8.2 KVM_CAP_HYPERV_SYNIC
-------------------------
-
-:Architectures: x86
-
-This capability, if KVM_CHECK_EXTENSION indicates that it is
-available, means that the kernel has an implementation of the
-Hyper-V Synthetic interrupt controller(SynIC). Hyper-V SynIC is
-used to support Windows Hyper-V based guest paravirt drivers(VMBus).
-
-In order to use SynIC, it has to be activated by setting this
-capability via KVM_ENABLE_CAP ioctl on the vcpu fd. Note that this
-will disable the use of APIC hardware virtualization even if supported
-by the CPU, as it's incompatible with SynIC auto-EOI behavior.
-
 8.3 KVM_CAP_PPC_MMU_RADIX
 -------------------------
 
@@ -8362,20 +8563,6 @@ may be incompatible with the MIPS VZ ASE.
     virtualization, including standard guest virtual memory segments.
 ==  ==========================================================================
 
-8.6 KVM_CAP_MIPS_TE
--------------------
-
-:Architectures: mips
-
-This capability, if KVM_CHECK_EXTENSION on the main kvm handle indicates that
-it is available, means that the trap & emulate implementation is available to
-run guest code in user mode, even if KVM_CAP_MIPS_VZ indicates that hardware
-assisted virtualisation is also available. KVM_VM_MIPS_TE (0) must be passed
-to KVM_CREATE_VM to create a VM which utilises it.
-
-If KVM_CHECK_EXTENSION on a kvm VM handle indicates that this capability is
-available, it means that the VM is using trap & emulate.
-
 8.7 KVM_CAP_MIPS_64BIT
 ----------------------
 
@@ -8457,16 +8644,6 @@ virtual SMT modes that can be set using KVM_CAP_PPC_SMT.  If bit N
 (counting from the right) is set, then a virtual SMT mode of 2^N is
 available.
 
-8.11 KVM_CAP_HYPERV_SYNIC2
---------------------------
-
-:Architectures: x86
-
-This capability enables a newer version of Hyper-V Synthetic interrupt
-controller (SynIC).  The only difference with KVM_CAP_HYPERV_SYNIC is that KVM
-doesn't clear SynIC message and event flags pages when they are enabled by
-writing to the respective MSRs.
-
 8.12 KVM_CAP_HYPERV_VP_INDEX
 ----------------------------
 
@@ -8481,7 +8658,6 @@ capability is absent, userspace can still query this msr's value.
 -------------------------------
 
 :Architectures: s390
-:Parameters: none
 
 This capability indicates if the flic device will be able to get/set the
 AIS states for migration via the KVM_DEV_FLIC_AISM_ALL attribute and allows
@@ -8555,21 +8731,6 @@ This capability indicates that KVM supports paravirtualized Hyper-V IPI send
 hypercalls:
 HvCallSendSyntheticClusterIpi, HvCallSendSyntheticClusterIpiEx.
 
-8.21 KVM_CAP_HYPERV_DIRECT_TLBFLUSH
------------------------------------
-
-:Architectures: x86
-
-This capability indicates that KVM running on top of Hyper-V hypervisor
-enables Direct TLB flush for its guests meaning that TLB flush
-hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM.
-Due to the different ABI for hypercall parameters between Hyper-V and
-KVM, enabling this capability effectively disables all hypercall
-handling by KVM (as some KVM hypercall may be mistakenly treated as TLB
-flush hypercalls by Hyper-V) so userspace should disable KVM identification
-in CPUID and only exposes Hyper-V identification. In this case, guest
-thinks it's running on Hyper-V and only use Hyper-V hypercalls.
-
 8.22 KVM_CAP_S390_VCPU_RESETS
 -----------------------------
 
@@ -8647,142 +8808,6 @@ In combination with KVM_CAP_X86_USER_SPACE_MSR, this allows user space to
 trap and emulate MSRs that are outside of the scope of KVM as well as
 limit the attack surface on KVM's MSR emulation code.
 
-8.28 KVM_CAP_ENFORCE_PV_FEATURE_CPUID
--------------------------------------
-
-Architectures: x86
-
-When enabled, KVM will disable paravirtual features provided to the
-guest according to the bits in the KVM_CPUID_FEATURES CPUID leaf
-(0x40000001). Otherwise, a guest may use the paravirtual features
-regardless of what has actually been exposed through the CPUID leaf.
-
-.. _KVM_CAP_DIRTY_LOG_RING:
-
-8.29 KVM_CAP_DIRTY_LOG_RING/KVM_CAP_DIRTY_LOG_RING_ACQ_REL
-----------------------------------------------------------
-
-:Architectures: x86, arm64
-:Parameters: args[0] - size of the dirty log ring
-
-KVM is capable of tracking dirty memory using ring buffers that are
-mmapped into userspace; there is one dirty ring per vcpu.
-
-The dirty ring is available to userspace as an array of
-``struct kvm_dirty_gfn``.  Each dirty entry is defined as::
-
-  struct kvm_dirty_gfn {
-          __u32 flags;
-          __u32 slot; /* as_id | slot_id */
-          __u64 offset;
-  };
-
-The following values are defined for the flags field to define the
-current state of the entry::
-
-  #define KVM_DIRTY_GFN_F_DIRTY           BIT(0)
-  #define KVM_DIRTY_GFN_F_RESET           BIT(1)
-  #define KVM_DIRTY_GFN_F_MASK            0x3
-
-Userspace should call KVM_ENABLE_CAP ioctl right after KVM_CREATE_VM
-ioctl to enable this capability for the new guest and set the size of
-the rings.  Enabling the capability is only allowed before creating any
-vCPU, and the size of the ring must be a power of two.  The larger the
-ring buffer, the less likely the ring is full and the VM is forced to
-exit to userspace. The optimal size depends on the workload, but it is
-recommended that it be at least 64 KiB (4096 entries).
-
-Just like for dirty page bitmaps, the buffer tracks writes to
-all user memory regions for which the KVM_MEM_LOG_DIRTY_PAGES flag was
-set in KVM_SET_USER_MEMORY_REGION.  Once a memory region is registered
-with the flag set, userspace can start harvesting dirty pages from the
-ring buffer.
-
-An entry in the ring buffer can be unused (flag bits ``00``),
-dirty (flag bits ``01``) or harvested (flag bits ``1X``).  The
-state machine for the entry is as follows::
-
-          dirtied         harvested        reset
-     00 -----------> 01 -------------> 1X -------+
-      ^                                          |
-      |                                          |
-      +------------------------------------------+
-
-To harvest the dirty pages, userspace accesses the mmapped ring buffer
-to read the dirty GFNs.  If the flags has the DIRTY bit set (at this stage
-the RESET bit must be cleared), then it means this GFN is a dirty GFN.
-The userspace should harvest this GFN and mark the flags from state
-``01b`` to ``1Xb`` (bit 0 will be ignored by KVM, but bit 1 must be set
-to show that this GFN is harvested and waiting for a reset), and move
-on to the next GFN.  The userspace should continue to do this until the
-flags of a GFN have the DIRTY bit cleared, meaning that it has harvested
-all the dirty GFNs that were available.
-
-Note that on weakly ordered architectures, userspace accesses to the
-ring buffer (and more specifically the 'flags' field) must be ordered,
-using load-acquire/store-release accessors when available, or any
-other memory barrier that will ensure this ordering.
-
-It's not necessary for userspace to harvest the all dirty GFNs at once.
-However it must collect the dirty GFNs in sequence, i.e., the userspace
-program cannot skip one dirty GFN to collect the one next to it.
-
-After processing one or more entries in the ring buffer, userspace
-calls the VM ioctl KVM_RESET_DIRTY_RINGS to notify the kernel about
-it, so that the kernel will reprotect those collected GFNs.
-Therefore, the ioctl must be called *before* reading the content of
-the dirty pages.
-
-The dirty ring can get full.  When it happens, the KVM_RUN of the
-vcpu will return with exit reason KVM_EXIT_DIRTY_LOG_FULL.
-
-The dirty ring interface has a major difference comparing to the
-KVM_GET_DIRTY_LOG interface in that, when reading the dirty ring from
-userspace, it's still possible that the kernel has not yet flushed the
-processor's dirty page buffers into the kernel buffer (with dirty bitmaps, the
-flushing is done by the KVM_GET_DIRTY_LOG ioctl).  To achieve that, one
-needs to kick the vcpu out of KVM_RUN using a signal.  The resulting
-vmexit ensures that all dirty GFNs are flushed to the dirty rings.
-
-NOTE: KVM_CAP_DIRTY_LOG_RING_ACQ_REL is the only capability that
-should be exposed by weakly ordered architecture, in order to indicate
-the additional memory ordering requirements imposed on userspace when
-reading the state of an entry and mutating it from DIRTY to HARVESTED.
-Architecture with TSO-like ordering (such as x86) are allowed to
-expose both KVM_CAP_DIRTY_LOG_RING and KVM_CAP_DIRTY_LOG_RING_ACQ_REL
-to userspace.
-
-After enabling the dirty rings, the userspace needs to detect the
-capability of KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP to see whether the
-ring structures can be backed by per-slot bitmaps. With this capability
-advertised, it means the architecture can dirty guest pages without
-vcpu/ring context, so that some of the dirty information will still be
-maintained in the bitmap structure. KVM_CAP_DIRTY_LOG_RING_WITH_BITMAP
-can't be enabled if the capability of KVM_CAP_DIRTY_LOG_RING_ACQ_REL
-hasn't been enabled, or any memslot has been existing.
-
-Note that the bitmap here is only a backup of the ring structure. The
-use of the ring and bitmap combination is only beneficial if there is
-only a very small amount of memory that is dirtied out of vcpu/ring
-context. Otherwise, the stand-alone per-slot bitmap mechanism needs to
-be considered.
-
-To collect dirty bits in the backup bitmap, userspace can use the same
-KVM_GET_DIRTY_LOG ioctl. KVM_CLEAR_DIRTY_LOG isn't needed as long as all
-the generation of the dirty bits is done in a single pass. Collecting
-the dirty bitmap should be the very last thing that the VMM does before
-considering the state as complete. VMM needs to ensure that the dirty
-state is final and avoid missing dirty pages from another ioctl ordered
-after the bitmap collection.
-
-NOTE: Multiple examples of using the backup bitmap: (1) save vgic/its
-tables through command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_SAVE_TABLES} on
-KVM device "kvm-arm-vgic-its". (2) restore vgic/its tables through
-command KVM_DEV_ARM_{VGIC_GRP_CTRL, ITS_RESTORE_TABLES} on KVM device
-"kvm-arm-vgic-its". VGICv3 LPI pending status is restored. (3) save
-vgic3 pending table through KVM_DEV_ARM_VGIC_{GRP_CTRL, SAVE_PENDING_TABLES}
-command on KVM device "kvm-arm-vgic-v3".
-
 8.30 KVM_CAP_XEN_HVM
 --------------------
 
@@ -8847,10 +8872,9 @@ clearing the PVCLOCK_TSC_STABLE_BIT flag in Xen pvclock sources. This will be
 done when the KVM_CAP_XEN_HVM ioctl sets the
 KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE flag.
 
-8.31 KVM_CAP_PPC_MULTITCE
--------------------------
+8.31 KVM_CAP_SPAPR_MULTITCE
+---------------------------
 
-:Capability: KVM_CAP_PPC_MULTITCE
 :Architectures: ppc
 :Type: vm
 
@@ -8882,72 +8906,9 @@ This capability indicates that the KVM virtual PTP service is
 supported in the host. A VMM can check whether the service is
 available to the guest on migration.
 
-8.33 KVM_CAP_HYPERV_ENFORCE_CPUID
----------------------------------
-
-Architectures: x86
-
-When enabled, KVM will disable emulated Hyper-V features provided to the
-guest according to the bits Hyper-V CPUID feature leaves. Otherwise, all
-currently implemented Hyper-V features are provided unconditionally when
-Hyper-V identification is set in the HYPERV_CPUID_INTERFACE (0x40000001)
-leaf.
-
-8.34 KVM_CAP_EXIT_HYPERCALL
----------------------------
-
-:Capability: KVM_CAP_EXIT_HYPERCALL
-:Architectures: x86
-:Type: vm
-
-This capability, if enabled, will cause KVM to exit to userspace
-with KVM_EXIT_HYPERCALL exit reason to process some hypercalls.
-
-Calling KVM_CHECK_EXTENSION for this capability will return a bitmask
-of hypercalls that can be configured to exit to userspace.
-Right now, the only such hypercall is KVM_HC_MAP_GPA_RANGE.
-
-The argument to KVM_ENABLE_CAP is also a bitmask, and must be a subset
-of the result of KVM_CHECK_EXTENSION.  KVM will forward to userspace
-the hypercalls whose corresponding bit is in the argument, and return
-ENOSYS for the others.
-
-8.35 KVM_CAP_PMU_CAPABILITY
----------------------------
-
-:Capability: KVM_CAP_PMU_CAPABILITY
-:Architectures: x86
-:Type: vm
-:Parameters: arg[0] is bitmask of PMU virtualization capabilities.
-:Returns: 0 on success, -EINVAL when arg[0] contains invalid bits
-
-This capability alters PMU virtualization in KVM.
-
-Calling KVM_CHECK_EXTENSION for this capability returns a bitmask of
-PMU virtualization capabilities that can be adjusted on a VM.
-
-The argument to KVM_ENABLE_CAP is also a bitmask and selects specific
-PMU virtualization capabilities to be applied to the VM.  This can
-only be invoked on a VM prior to the creation of VCPUs.
-
-At this time, KVM_PMU_CAP_DISABLE is the only capability.  Setting
-this capability will disable PMU virtualization for that VM.  Usermode
-should adjust CPUID leaf 0xA to reflect that the PMU is disabled.
-
-8.36 KVM_CAP_ARM_SYSTEM_SUSPEND
--------------------------------
-
-:Capability: KVM_CAP_ARM_SYSTEM_SUSPEND
-:Architectures: arm64
-:Type: vm
-
-When enabled, KVM will exit to userspace with KVM_EXIT_SYSTEM_EVENT of
-type KVM_SYSTEM_EVENT_SUSPEND to process the guest suspend request.
-
 8.37 KVM_CAP_S390_PROTECTED_DUMP
 --------------------------------
 
-:Capability: KVM_CAP_S390_PROTECTED_DUMP
 :Architectures: s390
 :Type: vm
 
@@ -8957,27 +8918,9 @@ PV guests. The `KVM_PV_DUMP` command is available for the
 dump related UV data. Also the vcpu ioctl `KVM_S390_PV_CPU_COMMAND` is
 available and supports the `KVM_PV_DUMP_CPU` subcommand.
 
-8.38 KVM_CAP_VM_DISABLE_NX_HUGE_PAGES
--------------------------------------
-
-:Capability: KVM_CAP_VM_DISABLE_NX_HUGE_PAGES
-:Architectures: x86
-:Type: vm
-:Parameters: arg[0] must be 0.
-:Returns: 0 on success, -EPERM if the userspace process does not
-          have CAP_SYS_BOOT, -EINVAL if args[0] is not 0 or any vCPUs have been
-          created.
-
-This capability disables the NX huge pages mitigation for iTLB MULTIHIT.
-
-The capability has no effect if the nx_huge_pages module parameter is not set.
-
-This capability may only be set before any vCPUs are created.
-
 8.39 KVM_CAP_S390_CPU_TOPOLOGY
 ------------------------------
 
-:Capability: KVM_CAP_S390_CPU_TOPOLOGY
 :Architectures: s390
 :Type: vm
 
@@ -8999,37 +8942,9 @@ structure.
 When getting the Modified Change Topology Report value, the attr->addr
 must point to a byte where the value will be stored or retrieved from.
 
-8.40 KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE
----------------------------------------
-
-:Capability: KVM_CAP_ARM_EAGER_SPLIT_CHUNK_SIZE
-:Architectures: arm64
-:Type: vm
-:Parameters: arg[0] is the new split chunk size.
-:Returns: 0 on success, -EINVAL if any memslot was already created.
-
-This capability sets the chunk size used in Eager Page Splitting.
-
-Eager Page Splitting improves the performance of dirty-logging (used
-in live migrations) when guest memory is backed by huge-pages.  It
-avoids splitting huge-pages (into PAGE_SIZE pages) on fault, by doing
-it eagerly when enabling dirty logging (with the
-KVM_MEM_LOG_DIRTY_PAGES flag for a memory region), or when using
-KVM_CLEAR_DIRTY_LOG.
-
-The chunk size specifies how many pages to break at a time, using a
-single allocation for each chunk. Bigger the chunk size, more pages
-need to be allocated ahead of time.
-
-The chunk size needs to be a valid block size. The list of acceptable
-block sizes is exposed in KVM_CAP_ARM_SUPPORTED_BLOCK_SIZES as a
-64-bit bitmap (each bit describing a block size). The default value is
-0, to disable the eager page splitting.
-
 8.41 KVM_CAP_VM_TYPES
 ---------------------
 
-:Capability: KVM_CAP_MEMORY_ATTRIBUTES
 :Architectures: x86
 :Type: system ioctl
 
@@ -9046,6 +8961,67 @@ Do not use KVM_X86_SW_PROTECTED_VM for "real" VMs, and especially not in
 production.  The behavior and effective ABI for software-protected VMs is
 unstable.
 
+8.42 KVM_CAP_PPC_RPT_INVALIDATE
+-------------------------------
+
+:Architectures: ppc
+
+This capability indicates that the kernel is capable of handling
+H_RPT_INVALIDATE hcall.
+
+In order to enable the use of H_RPT_INVALIDATE in the guest,
+user space might have to advertise it for the guest. For example,
+IBM pSeries (sPAPR) guest starts using it if "hcall-rpt-invalidate" is
+present in the "ibm,hypertas-functions" device-tree property.
+
+This capability is enabled for hypervisors on platforms like POWER9
+that support radix MMU.
+
+8.43 KVM_CAP_PPC_AIL_MODE_3
+---------------------------
+
+:Architectures: ppc
+
+This capability indicates that the kernel supports the mode 3 setting for the
+"Address Translation Mode on Interrupt" aka "Alternate Interrupt Location"
+resource that is controlled with the H_SET_MODE hypercall.
+
+This capability allows a guest kernel to use a better-performance mode for
+handling interrupts and system calls.
+
+8.44 KVM_CAP_MEMORY_FAULT_INFO
+------------------------------
+
+:Architectures: x86
+
+The presence of this capability indicates that KVM_RUN will fill
+kvm_run.memory_fault if KVM cannot resolve a guest page fault VM-Exit, e.g. if
+there is a valid memslot but no backing VMA for the corresponding host virtual
+address.
+
+The information in kvm_run.memory_fault is valid if and only if KVM_RUN returns
+an error with errno=EFAULT or errno=EHWPOISON *and* kvm_run.exit_reason is set
+to KVM_EXIT_MEMORY_FAULT.
+
+Note: Userspaces which attempt to resolve memory faults so that they can retry
+KVM_RUN are encouraged to guard against repeatedly receiving the same
+error/annotated fault.
+
+See KVM_EXIT_MEMORY_FAULT for more information.
+
+8.45 KVM_CAP_X86_GUEST_MODE
+---------------------------
+
+:Architectures: x86
+
+The presence of this capability indicates that KVM_RUN will update the
+KVM_RUN_X86_GUEST_MODE bit in kvm_run.flags to indicate whether the
+vCPU was executing nested guest code when it exited.
+
+KVM exits with the register state of either the L1 or L2 guest
+depending on which executed at the time of an exit. Userspace must
+take care to differentiate between these cases.
+
 9. Known KVM API problems
 =========================
 
@@ -9076,9 +9052,10 @@ the local APIC.
 
 The same is true for the ``KVM_FEATURE_PV_UNHALT`` paravirtualized feature.
 
-CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by ``KVM_GET_SUPPORTED_CPUID``.
-It can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER`` is present and the kernel
-has enabled in-kernel emulation of the local APIC.
+On older versions of Linux, CPU[EAX=1]:ECX[24] (TSC_DEADLINE) is not reported by
+``KVM_GET_SUPPORTED_CPUID``, but it can be enabled if ``KVM_CAP_TSC_DEADLINE_TIMER``
+is present and the kernel has enabled in-kernel emulation of the local APIC.
+On newer versions, ``KVM_GET_SUPPORTED_CPUID`` does report the bit as available.
 
 CPU topology
 ~~~~~~~~~~~~
diff --git a/Documentation/wmi/devices/msi-wmi-platform.rst b/Documentation/wmi/devices/msi-wmi-platform.rst
index 31a136942892..73197b31926a 100644
--- a/Documentation/wmi/devices/msi-wmi-platform.rst
+++ b/Documentation/wmi/devices/msi-wmi-platform.rst
@@ -138,6 +138,10 @@ input data, the meaning of which depends on the subfeature being accessed.
 The output buffer contains a single byte which signals success or failure (``0x00`` on failure)
 and 31 bytes of output data, the meaning if which depends on the subfeature being accessed.
 
+.. note::
+   The ACPI control method responsible for handling the WMI method calls is not thread-safe.
+   This is a firmware bug that needs to be handled inside the driver itself.
+
 WMI method Get_EC()
 -------------------
 
diff --git a/MAINTAINERS b/MAINTAINERS
index f7cffb0a5a77..d063d711e092 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2306,6 +2306,7 @@ F:	drivers/watchdog/apple_wdt.c
 F:	include/dt-bindings/interrupt-controller/apple-aic.h
 F:	include/dt-bindings/pinctrl/apple.h
 F:	include/linux/soc/apple/*
+F:	include/uapi/drm/asahi_drm.h
 
 ARM/ARTPEC MACHINE SUPPORT
 M:	Jesper Nilsson <jesper.nilsson@axis.com>
@@ -3191,6 +3192,12 @@ M:	Dinh Nguyen <dinguyen@kernel.org>
 S:	Maintained
 F:	drivers/clk/socfpga/
 
+ARM/SOCFPGA DWMAC GLUE LAYER
+M:	Maxime Chevallier <maxime.chevallier@bootlin.com>
+S:	Maintained
+F:	Documentation/devicetree/bindings/net/socfpga-dwmac.txt
+F:	drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
+
 ARM/SOCFPGA EDAC BINDINGS
 M:	Matthew Gerlach <matthew.gerlach@altera.com>
 S:	Maintained
@@ -3867,8 +3874,9 @@ AUXILIARY BUS DRIVER
 M:	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 R:	Dave Ertman <david.m.ertman@intel.com>
 R:	Ira Weiny <ira.weiny@intel.com>
+R:	Leon Romanovsky <leon@kernel.org>
 S:	Supported
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core.git
 F:	Documentation/driver-api/auxiliary_bus.rst
 F:	drivers/base/auxiliary.c
 F:	include/linux/auxiliary_bus.h
@@ -6338,6 +6346,7 @@ F:	Documentation/process/cve.rst
 
 CW1200 WLAN driver
 S:	Orphan
+L:	linux-wireless@vger.kernel.org
 F:	drivers/net/wireless/st/
 F:	include/linux/platform_data/net-cw1200.h
 
@@ -7023,6 +7032,7 @@ L:	rust-for-linux@vger.kernel.org
 S:	Supported
 W:	https://rust-for-linux.com
 T:	git https://github.com/Rust-for-Linux/linux.git alloc-next
+F:	rust/helpers/dma.c
 F:	rust/kernel/dma.rs
 F:	samples/rust/rust_dma.rs
 
@@ -7228,7 +7238,7 @@ M:	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 M:	"Rafael J. Wysocki" <rafael@kernel.org>
 M:	Danilo Krummrich <dakr@kernel.org>
 S:	Supported
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core.git
 F:	Documentation/core-api/kobject.rst
 F:	drivers/base/
 F:	fs/debugfs/
@@ -7373,8 +7383,7 @@ M:	Javier Martinez Canillas <javierm@redhat.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Maintained
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
-F:	drivers/gpu/drm/tiny/ofdrm.c
-F:	drivers/gpu/drm/tiny/simpledrm.c
+F:	drivers/gpu/drm/sysfb/
 F:	drivers/video/aperture.c
 F:	drivers/video/nomodeset.c
 F:	include/linux/aperture.h
@@ -7515,6 +7524,7 @@ S:	Maintained
 B:	https://gitlab.freedesktop.org/drm/msm/-/issues
 T:	git https://gitlab.freedesktop.org/drm/msm.git
 F:	Documentation/devicetree/bindings/display/msm/gpu.yaml
+F:	Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml
 F:	drivers/gpu/drm/msm/adreno/
 F:	drivers/gpu/drm/msm/msm_gpu.*
 F:	drivers/gpu/drm/msm/msm_gpu_devfreq.*
@@ -7568,6 +7578,12 @@ T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/devicetree/bindings/display/panel/novatek,nt36672a.yaml
 F:	drivers/gpu/drm/panel/panel-novatek-nt36672a.c
 
+DRM DRIVER FOR NOVATEK NT37801 PANELS
+M:	Krzysztof Kozlowski <krzk@kernel.org>
+S:	Maintained
+F:	Documentation/devicetree/bindings/display/panel/novatek,nt37801.yaml
+F:	drivers/gpu/drm/panel/panel-novatek-nt37801.c
+
 DRM DRIVER FOR NVIDIA GEFORCE/QUADRO GPUS
 M:	Lyude Paul <lyude@redhat.com>
 M:	Danilo Krummrich <dakr@kernel.org>
@@ -7695,7 +7711,13 @@ M:	David Lechner <david@lechnology.com>
 S:	Maintained
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/devicetree/bindings/display/sitronix,st7586.txt
-F:	drivers/gpu/drm/tiny/st7586.c
+F:	drivers/gpu/drm/sitronix/st7586.c
+
+DRM DRIVER FOR SITRONIX ST7571 PANELS
+M:	Marcus Folkesson <marcus.folkesson@gmail.com>
+S:	Maintained
+F:	Documentation/devicetree/bindings/display/sitronix,st7571.yaml
+F:	drivers/gpu/drm/sitronix/st7571-i2c.c
 
 DRM DRIVER FOR SITRONIX ST7701 PANELS
 M:	Jagan Teki <jagan@amarulasolutions.com>
@@ -7716,7 +7738,7 @@ M:	David Lechner <david@lechnology.com>
 S:	Maintained
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/devicetree/bindings/display/sitronix,st7735r.yaml
-F:	drivers/gpu/drm/tiny/st7735r.c
+F:	drivers/gpu/drm/sitronix/st7735r.c
 
 DRM DRIVER FOR SOLOMON SSD130X OLED DISPLAYS
 M:	Javier Martinez Canillas <javierm@redhat.com>
@@ -7870,8 +7892,8 @@ F:	drivers/gpu/drm/ci/xfails/meson*
 F:	drivers/gpu/drm/meson/
 
 DRM DRIVERS FOR ATMEL HLCDC
-M:	Sam Ravnborg <sam@ravnborg.org>
-M:	Boris Brezillon <bbrezillon@kernel.org>
+M:	Manikandan Muralidharan <manikandan.m@microchip.com>
+M:	Dharma Balasubiramani <dharma.b@microchip.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Supported
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
@@ -8199,7 +8221,8 @@ F:	drivers/gpu/drm/ttm/
 F:	include/drm/ttm/
 
 DRM AUTOMATED TESTING
-M:	Helen Koike <helen.koike@collabora.com>
+M:	Helen Koike <helen.fornazier@gmail.com>
+M:	Vignesh Raman <vignesh.raman@collabora.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Maintained
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
@@ -10168,6 +10191,8 @@ F:	include/linux/gpio.h
 F:	include/linux/gpio/
 F:	include/linux/of_gpio.h
 K:	(devm_)?gpio_(request|free|direction|get|set)
+K:	GPIOD_FLAGS_BIT_NONEXCLUSIVE
+K:	devm_gpiod_unhinge
 
 GPIO UAPI
 M:	Bartosz Golaszewski <brgl@bgdev.pl>
@@ -10470,14 +10495,20 @@ S:	Supported
 F:	drivers/infiniband/hw/hfi1
 
 HFS FILESYSTEM
+M:	Viacheslav Dubeyko <slava@dubeyko.com>
+M:	John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
+M:	Yangtao Li <frank.li@vivo.com>
 L:	linux-fsdevel@vger.kernel.org
-S:	Orphan
+S:	Maintained
 F:	Documentation/filesystems/hfs.rst
 F:	fs/hfs/
 
 HFSPLUS FILESYSTEM
+M:	Viacheslav Dubeyko <slava@dubeyko.com>
+M:	John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
+M:	Yangtao Li <frank.li@vivo.com>
 L:	linux-fsdevel@vger.kernel.org
-S:	Orphan
+S:	Maintained
 F:	Documentation/filesystems/hfsplus.rst
 F:	fs/hfsplus/
 
@@ -10971,6 +11002,7 @@ F:	include/linux/platform_data/huawei-gaokun-ec.h
 
 HUGETLB SUBSYSTEM
 M:	Muchun Song <muchun.song@linux.dev>
+R:	Oscar Salvador <osalvador@suse.de>
 L:	linux-mm@kvack.org
 S:	Maintained
 F:	Documentation/ABI/testing/sysfs-kernel-mm-hugepages
@@ -11934,13 +11966,10 @@ F:	drivers/gpio/gpio-tangier.c
 F:	drivers/gpio/gpio-tangier.h
 
 INTEL GVT-g DRIVERS (Intel GPU Virtualization)
-M:	Zhenyu Wang <zhenyuw.linux@gmail.com>
-M:	Zhi Wang <zhi.wang.linux@gmail.com>
-L:	intel-gvt-dev@lists.freedesktop.org
-L:	intel-gfx@lists.freedesktop.org
-S:	Supported
+R:	Zhenyu Wang <zhenyuw.linux@gmail.com>
+R:	Zhi Wang <zhi.wang.linux@gmail.com>
+S:	Odd Fixes
 W:	https://github.com/intel/gvt-linux/wiki
-T:	git https://github.com/intel/gvt-linux.git
 F:	drivers/gpu/drm/i915/gvt/
 
 INTEL HID EVENT DRIVER
@@ -12827,6 +12856,7 @@ F:	lib/Kconfig.kcsan
 F:	scripts/Makefile.kcsan
 
 KDUMP
+M:	Andrew Morton <akpm@linux-foundation.org>
 M:	Baoquan He <bhe@redhat.com>
 R:	Vivek Goyal <vgoyal@redhat.com>
 R:	Dave Young <dyoung@redhat.com>
@@ -13123,11 +13153,13 @@ KERNFS
 M:	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 M:	Tejun Heo <tj@kernel.org>
 S:	Supported
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core.git
 F:	fs/kernfs/
 F:	include/linux/kernfs.h
 
 KEXEC
+M:	Andrew Morton <akpm@linux-foundation.org>
+M:	Baoquan He <bhe@redhat.com>
 L:	kexec@lists.infradead.org
 W:	http://kernel.org/pub/linux/utils/kernel/kexec/
 F:	include/linux/kexec.h
@@ -14300,6 +14332,7 @@ S:	Odd fixes
 F:	drivers/net/ethernet/marvell/sk*
 
 MARVELL LIBERTAS WIRELESS DRIVER
+L:	linux-wireless@vger.kernel.org
 L:	libertas-dev@lists.infradead.org
 S:	Orphan
 F:	drivers/net/wireless/marvell/libertas/
@@ -15525,6 +15558,21 @@ F:	mm/numa.c
 F:	mm/numa_emulation.c
 F:	mm/numa_memblks.c
 
+MEMORY MANAGEMENT - PAGE ALLOCATOR
+M:	Andrew Morton <akpm@linux-foundation.org>
+R:	Vlastimil Babka <vbabka@suse.cz>
+R:	Suren Baghdasaryan <surenb@google.com>
+R:	Michal Hocko <mhocko@suse.com>
+R:	Brendan Jackman <jackmanb@google.com>
+R:	Johannes Weiner <hannes@cmpxchg.org>
+R:	Zi Yan <ziy@nvidia.com>
+L:	linux-mm@kvack.org
+S:	Maintained
+F:	mm/compaction.c
+F:	mm/page_alloc.c
+F:	include/linux/gfp.h
+F:	include/linux/compaction.h
+
 MEMORY MANAGEMENT - SECRETMEM
 M:	Andrew Morton <akpm@linux-foundation.org>
 M:	Mike Rapoport <rppt@kernel.org>
@@ -15552,10 +15600,12 @@ M:	Liam R. Howlett <Liam.Howlett@oracle.com>
 M:	Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
 R:	Vlastimil Babka <vbabka@suse.cz>
 R:	Jann Horn <jannh@google.com>
+R:	Pedro Falcato <pfalcato@suse.de>
 L:	linux-mm@kvack.org
 S:	Maintained
 W:	http://www.linux-mm.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
+F:	include/trace/events/mmap.h
 F:	mm/mlock.c
 F:	mm/mmap.c
 F:	mm/mprotect.c
@@ -15566,6 +15616,36 @@ F:	mm/vma.h
 F:	mm/vma_internal.h
 F:	tools/testing/vma/
 
+MEMORY MAPPING - LOCKING
+M:	Andrew Morton <akpm@linux-foundation.org>
+M:	Suren Baghdasaryan <surenb@google.com>
+M:	Liam R. Howlett <Liam.Howlett@oracle.com>
+M:	Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
+R:	Vlastimil Babka <vbabka@suse.cz>
+R:	Shakeel Butt <shakeel.butt@linux.dev>
+L:	linux-mm@kvack.org
+S:	Maintained
+W:	http://www.linux-mm.org
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
+F:	Documentation/mm/process_addrs.rst
+F:	include/linux/mmap_lock.h
+F:	include/trace/events/mmap_lock.h
+F:	mm/mmap_lock.c
+
+MEMORY MAPPING - MADVISE (MEMORY ADVICE)
+M:	Andrew Morton <akpm@linux-foundation.org>
+M:	Liam R. Howlett <Liam.Howlett@oracle.com>
+M:	Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
+M:	David Hildenbrand <david@redhat.com>
+R:	Vlastimil Babka <vbabka@suse.cz>
+R:	Jann Horn <jannh@google.com>
+L:	linux-mm@kvack.org
+S:	Maintained
+W:	http://www.linux-mm.org
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
+F:	include/uapi/asm-generic/mman-common.h
+F:	mm/madvise.c
+
 MEMORY TECHNOLOGY DEVICES (MTD)
 M:	Miquel Raynal <miquel.raynal@bootlin.com>
 M:	Richard Weinberger <richard@nod.at>
@@ -16773,6 +16853,7 @@ F:	Documentation/networking/net_cachelines/net_device.rst
 F:	drivers/connector/
 F:	drivers/net/
 F:	drivers/ptp/
+F:	drivers/s390/net/
 F:	include/dt-bindings/net/
 F:	include/linux/cn_proc.h
 F:	include/linux/etherdevice.h
@@ -16782,6 +16863,7 @@ F:	include/linux/fddidevice.h
 F:	include/linux/hippidevice.h
 F:	include/linux/if_*
 F:	include/linux/inetdevice.h
+F:	include/linux/ism.h
 F:	include/linux/netdev*
 F:	include/linux/platform_data/wiznet.h
 F:	include/uapi/linux/cn_proc.h
@@ -17385,7 +17467,7 @@ T:	git git://git.infradead.org/nvme.git
 F:	drivers/nvme/target/
 
 NVMEM FRAMEWORK
-M:	Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
+M:	Srinivas Kandagatla <srini@kernel.org>
 S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/srini/nvmem.git
 F:	Documentation/ABI/stable/sysfs-bus-nvmem
@@ -18650,7 +18732,7 @@ F:	drivers/pci/controller/pci-xgene-msi.c
 PCI NATIVE HOST BRIDGE AND ENDPOINT DRIVERS
 M:	Lorenzo Pieralisi <lpieralisi@kernel.org>
 M:	Krzysztof Wilczyński <kw@linux.com>
-R:	Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
+M:	Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
 R:	Rob Herring <robh@kernel.org>
 L:	linux-pci@vger.kernel.org
 S:	Supported
@@ -18703,6 +18785,16 @@ F:	include/asm-generic/pci*
 F:	include/linux/of_pci.h
 F:	include/linux/pci*
 F:	include/uapi/linux/pci*
+
+PCI SUBSYSTEM [RUST]
+M:	Danilo Krummrich <dakr@kernel.org>
+R:	Bjorn Helgaas <bhelgaas@google.com>
+R:	Krzysztof Wilczyński <kwilczynski@kernel.org>
+L:	linux-pci@vger.kernel.org
+S:	Maintained
+C:	irc://irc.oftc.net/linux-pci
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci.git
+F:	rust/helpers/pci.c
 F:	rust/kernel/pci.rs
 F:	samples/rust/rust_driver_pci.rs
 
@@ -19590,7 +19682,7 @@ S:	Supported
 F:	drivers/crypto/intel/qat/
 
 QCOM AUDIO (ASoC) DRIVERS
-M:	Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
+M:	Srinivas Kandagatla <srini@kernel.org>
 L:	linux-sound@vger.kernel.org
 L:	linux-arm-msm@vger.kernel.org
 S:	Supported
@@ -19763,6 +19855,7 @@ F:	drivers/media/tuners/qt1010*
 
 QUALCOMM ATH12K WIRELESS DRIVER
 M:	Jeff Johnson <jjohnson@kernel.org>
+L:	linux-wireless@vger.kernel.org
 L:	ath12k@lists.infradead.org
 S:	Supported
 W:	https://wireless.wiki.kernel.org/en/users/Drivers/ath12k
@@ -19772,6 +19865,7 @@ N:	ath12k
 
 QUALCOMM ATHEROS ATH10K WIRELESS DRIVER
 M:	Jeff Johnson <jjohnson@kernel.org>
+L:	linux-wireless@vger.kernel.org
 L:	ath10k@lists.infradead.org
 S:	Supported
 W:	https://wireless.wiki.kernel.org/en/users/Drivers/ath10k
@@ -19781,6 +19875,7 @@ N:	ath10k
 
 QUALCOMM ATHEROS ATH11K WIRELESS DRIVER
 M:	Jeff Johnson <jjohnson@kernel.org>
+L:	linux-wireless@vger.kernel.org
 L:	ath11k@lists.infradead.org
 S:	Supported
 W:	https://wireless.wiki.kernel.org/en/users/Drivers/ath11k
@@ -19890,7 +19985,7 @@ F:	Documentation/devicetree/bindings/net/qcom,ethqos.yaml
 F:	drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
 
 QUALCOMM FASTRPC DRIVER
-M:	Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
+M:	Srinivas Kandagatla <srini@kernel.org>
 M:	Amol Maheshwari <amahesh@qti.qualcomm.com>
 L:	linux-arm-msm@vger.kernel.org
 L:	dri-devel@lists.freedesktop.org
@@ -21270,6 +21365,7 @@ L:	linux-s390@vger.kernel.org
 L:	netdev@vger.kernel.org
 S:	Supported
 F:	drivers/s390/net/
+F:	include/linux/ism.h
 
 S390 PCI SUBSYSTEM
 M:	Niklas Schnelle <schnelle@linux.ibm.com>
@@ -21938,7 +22034,7 @@ S:	Maintained
 F:	drivers/media/rc/serial_ir.c
 
 SERIAL LOW-POWER INTER-CHIP MEDIA BUS (SLIMbus)
-M:	Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
+M:	Srinivas Kandagatla <srini@kernel.org>
 L:	linux-sound@vger.kernel.org
 S:	Maintained
 F:	Documentation/devicetree/bindings/slimbus/
@@ -22154,6 +22250,7 @@ F:	drivers/platform/x86/touchscreen_dmi.c
 
 SILICON LABS WIRELESS DRIVERS (for WFxxx series)
 M:	Jérôme Pouiller <jerome.pouiller@silabs.com>
+L:	linux-wireless@vger.kernel.org
 S:	Supported
 F:	Documentation/devicetree/bindings/net/wireless/silabs,wfx.yaml
 F:	drivers/net/wireless/silabs/
@@ -22254,9 +22351,7 @@ F:	drivers/nvmem/layouts/sl28vpd.c
 
 SLAB ALLOCATOR
 M:	Christoph Lameter <cl@linux.com>
-M:	Pekka Enberg <penberg@kernel.org>
 M:	David Rientjes <rientjes@google.com>
-M:	Joonsoo Kim <iamjoonsoo.kim@lge.com>
 M:	Andrew Morton <akpm@linux-foundation.org>
 M:	Vlastimil Babka <vbabka@suse.cz>
 R:	Roman Gushchin <roman.gushchin@linux.dev>
@@ -22671,9 +22766,15 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git
 F:	Documentation/devicetree/bindings/sound/
 F:	Documentation/sound/soc/
 F:	include/dt-bindings/sound/
+F:	include/sound/cs-amp-lib.h
+F:	include/sound/cs35l*
+F:	include/sound/cs4271.h
+F:	include/sound/cs42l*
+F:	include/sound/madera-pdata.h
 F:	include/sound/soc*
 F:	include/sound/sof.h
 F:	include/sound/sof/
+F:	include/sound/wm*.h
 F:	include/trace/events/sof*.h
 F:	include/uapi/sound/asoc.h
 F:	sound/soc/
@@ -25143,9 +25244,13 @@ S:	Maintained
 F:	drivers/usb/typec/mux/pi3usb30532.c
 
 USB TYPEC PORT CONTROLLER DRIVERS
+M:	Badhri Jagan Sridharan <badhri@google.com>
 L:	linux-usb@vger.kernel.org
-S:	Orphan
-F:	drivers/usb/typec/tcpm/
+S:	Maintained
+F:	drivers/usb/typec/tcpm/tcpci.c
+F:	drivers/usb/typec/tcpm/tcpm.c
+F:	include/linux/usb/tcpci.h
+F:	include/linux/usb/tcpm.h
 
 USB TYPEC TUSB1046 MUX DRIVER
 M:	Romain Gantois <romain.gantois@bootlin.com>
@@ -25601,6 +25706,7 @@ F:	include/uapi/linux/virtio_gpio.h
 VIRTIO GPU DRIVER
 M:	David Airlie <airlied@redhat.com>
 M:	Gerd Hoffmann <kraxel@redhat.com>
+M:	Dmitry Osipenko <dmitry.osipenko@collabora.com>
 R:	Gurchetan Singh <gurchetansingh@chromium.org>
 R:	Chia-I Wu <olvaffe@gmail.com>
 L:	dri-devel@lists.freedesktop.org
diff --git a/Makefile b/Makefile
index 38689a0c3605..b29cc321ffd9 100644
--- a/Makefile
+++ b/Makefile
@@ -2,7 +2,7 @@
 VERSION = 6
 PATCHLEVEL = 15
 SUBLEVEL = 0
-EXTRAVERSION = -rc1
+EXTRAVERSION = -rc5
 NAME = Baby Opossum Posse
 
 # *DOCUMENTATION*
@@ -477,7 +477,6 @@ export rust_common_flags := --edition=2021 \
 			    -Wclippy::ignored_unit_patterns \
 			    -Wclippy::mut_mut \
 			    -Wclippy::needless_bitwise_bool \
-			    -Wclippy::needless_continue \
 			    -Aclippy::needless_lifetimes \
 			    -Wclippy::no_mangle_with_rust_abi \
 			    -Wclippy::undocumented_unsafe_blocks \
@@ -1053,10 +1052,6 @@ NOSTDINC_FLAGS += -nostdinc
 # perform bounds checking.
 KBUILD_CFLAGS += $(call cc-option, -fstrict-flex-arrays=3)
 
-#Currently, disable -Wstringop-overflow for GCC 11, globally.
-KBUILD_CFLAGS-$(CONFIG_CC_NO_STRINGOP_OVERFLOW) += $(call cc-option, -Wno-stringop-overflow)
-KBUILD_CFLAGS-$(CONFIG_CC_STRINGOP_OVERFLOW) += $(call cc-option, -Wstringop-overflow)
-
 # disable invalid "can't wrap" optimizations for signed / pointers
 KBUILD_CFLAGS	+= -fno-strict-overflow
 
@@ -1068,6 +1063,9 @@ ifdef CONFIG_CC_IS_GCC
 KBUILD_CFLAGS   += -fconserve-stack
 endif
 
+# Ensure compilers do not transform certain loops into calls to wcslen()
+KBUILD_CFLAGS += -fno-builtin-wcslen
+
 # change __FILE__ to the relative path to the source directory
 ifdef building_out_of_srctree
 KBUILD_CPPFLAGS += $(call cc-option,-ffile-prefix-map=$(srcroot)/=)
diff --git a/arch/arm/configs/at91_dt_defconfig b/arch/arm/configs/at91_dt_defconfig
index f2596a1b2f7d..ff13e1ecf4bb 100644
--- a/arch/arm/configs/at91_dt_defconfig
+++ b/arch/arm/configs/at91_dt_defconfig
@@ -232,7 +232,6 @@ CONFIG_CRYPTO_USER_API_SKCIPHER=m
 CONFIG_CRYPTO_DEV_ATMEL_AES=y
 CONFIG_CRYPTO_DEV_ATMEL_TDES=y
 CONFIG_CRYPTO_DEV_ATMEL_SHA=y
-CONFIG_CRC_CCITT=y
 CONFIG_FONTS=y
 CONFIG_FONT_8x8=y
 CONFIG_FONT_ACORN_8x8=y
diff --git a/arch/arm/configs/collie_defconfig b/arch/arm/configs/collie_defconfig
index 42cb1c854118..578c6a4af620 100644
--- a/arch/arm/configs/collie_defconfig
+++ b/arch/arm/configs/collie_defconfig
@@ -78,7 +78,6 @@ CONFIG_ROMFS_FS=y
 CONFIG_NLS_DEFAULT="cp437"
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_CCITT=y
 CONFIG_FONTS=y
 CONFIG_FONT_MINI_4x6=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
diff --git a/arch/arm/configs/davinci_all_defconfig b/arch/arm/configs/davinci_all_defconfig
index 3474e475373a..70b8c78386f4 100644
--- a/arch/arm/configs/davinci_all_defconfig
+++ b/arch/arm/configs/davinci_all_defconfig
@@ -249,7 +249,6 @@ CONFIG_NLS_ASCII=m
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_UTF8=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_T10DIF=m
 CONFIG_DMA_CMA=y
 CONFIG_DEBUG_FS=y
 CONFIG_DEBUG_RT_MUTEXES=y
diff --git a/arch/arm/configs/dove_defconfig b/arch/arm/configs/dove_defconfig
index b382a2e175fb..d76eb12d29a7 100644
--- a/arch/arm/configs/dove_defconfig
+++ b/arch/arm/configs/dove_defconfig
@@ -128,7 +128,6 @@ CONFIG_CRYPTO_DEFLATE=y
 CONFIG_CRYPTO_LZO=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 CONFIG_CRYPTO_DEV_MARVELL_CESA=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/arm/configs/exynos_defconfig b/arch/arm/configs/exynos_defconfig
index 7ad48fdda1da..e81a5d6c1c20 100644
--- a/arch/arm/configs/exynos_defconfig
+++ b/arch/arm/configs/exynos_defconfig
@@ -370,7 +370,6 @@ CONFIG_CRYPTO_AES_ARM_BS=m
 CONFIG_CRYPTO_CHACHA20_NEON=m
 CONFIG_CRYPTO_DEV_EXYNOS_RNG=y
 CONFIG_CRYPTO_DEV_S5P=y
-CONFIG_CRC_CCITT=y
 CONFIG_DMA_CMA=y
 CONFIG_CMA_SIZE_MBYTES=96
 CONFIG_FONTS=y
diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig
index 297c6a7b978a..062c1eb8dd60 100644
--- a/arch/arm/configs/imx_v6_v7_defconfig
+++ b/arch/arm/configs/imx_v6_v7_defconfig
@@ -481,8 +481,6 @@ CONFIG_SECURITYFS=y
 CONFIG_CRYPTO_DEV_FSL_CAAM=y
 CONFIG_CRYPTO_DEV_SAHARA=y
 CONFIG_CRYPTO_DEV_MXS_DCP=y
-CONFIG_CRC_CCITT=m
-CONFIG_CRC_T10DIF=y
 CONFIG_CMA_SIZE_MBYTES=64
 CONFIG_FONTS=y
 CONFIG_FONT_8x8=y
diff --git a/arch/arm/configs/lpc18xx_defconfig b/arch/arm/configs/lpc18xx_defconfig
index 2aa2ac8c6507..2d489186e945 100644
--- a/arch/arm/configs/lpc18xx_defconfig
+++ b/arch/arm/configs/lpc18xx_defconfig
@@ -147,7 +147,6 @@ CONFIG_EXT2_FS=y
 # CONFIG_INOTIFY_USER is not set
 CONFIG_JFFS2_FS=y
 # CONFIG_NETWORK_FILESYSTEMS is not set
-CONFIG_CRC_ITU_T=y
 CONFIG_PRINTK_TIME=y
 # CONFIG_ENABLE_MUST_CHECK is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
diff --git a/arch/arm/configs/lpc32xx_defconfig b/arch/arm/configs/lpc32xx_defconfig
index 98e267213b21..9afccd76446b 100644
--- a/arch/arm/configs/lpc32xx_defconfig
+++ b/arch/arm/configs/lpc32xx_defconfig
@@ -179,7 +179,6 @@ CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_UTF8=y
 CONFIG_CRYPTO_ANSI_CPRNG=y
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DYNAMIC_DEBUG=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/arm/configs/milbeaut_m10v_defconfig b/arch/arm/configs/milbeaut_m10v_defconfig
index acd16204f8d7..275ddf7a3a14 100644
--- a/arch/arm/configs/milbeaut_m10v_defconfig
+++ b/arch/arm/configs/milbeaut_m10v_defconfig
@@ -108,8 +108,6 @@ CONFIG_CRYPTO_AES_ARM_BS=m
 CONFIG_CRYPTO_AES_ARM_CE=m
 CONFIG_CRYPTO_CHACHA20_NEON=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=m
-CONFIG_CRC_ITU_T=m
 CONFIG_DMA_CMA=y
 CONFIG_CMA_SIZE_MBYTES=64
 CONFIG_PRINTK_TIME=y
diff --git a/arch/arm/configs/mmp2_defconfig b/arch/arm/configs/mmp2_defconfig
index f6f9e135353e..842a989baa27 100644
--- a/arch/arm/configs/mmp2_defconfig
+++ b/arch/arm/configs/mmp2_defconfig
@@ -67,7 +67,6 @@ CONFIG_NFS_V3=y
 CONFIG_NFS_V3_ACL=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/arm/configs/multi_v4t_defconfig b/arch/arm/configs/multi_v4t_defconfig
index 27d650635d9b..1a86dc305523 100644
--- a/arch/arm/configs/multi_v4t_defconfig
+++ b/arch/arm/configs/multi_v4t_defconfig
@@ -91,6 +91,5 @@ CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_CRAMFS=y
 CONFIG_MINIX_FS=y
-CONFIG_CRC_CCITT=y
 # CONFIG_FTRACE is not set
 CONFIG_DEBUG_USER=y
diff --git a/arch/arm/configs/multi_v5_defconfig b/arch/arm/configs/multi_v5_defconfig
index db81862bdb93..cf6180b4296e 100644
--- a/arch/arm/configs/multi_v5_defconfig
+++ b/arch/arm/configs/multi_v5_defconfig
@@ -289,7 +289,6 @@ CONFIG_NLS_UTF8=y
 CONFIG_CRYPTO_CBC=m
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_DEV_MARVELL_CESA=y
-CONFIG_CRC_CCITT=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/arm/configs/mvebu_v5_defconfig b/arch/arm/configs/mvebu_v5_defconfig
index a518d4a2581e..23dbb80fcc2e 100644
--- a/arch/arm/configs/mvebu_v5_defconfig
+++ b/arch/arm/configs/mvebu_v5_defconfig
@@ -187,7 +187,6 @@ CONFIG_NLS_UTF8=y
 CONFIG_CRYPTO_CBC=m
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_DEV_MARVELL_CESA=y
-CONFIG_CRC_CCITT=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/arm/configs/mxs_defconfig b/arch/arm/configs/mxs_defconfig
index d8a6e43c401e..c76d66135abb 100644
--- a/arch/arm/configs/mxs_defconfig
+++ b/arch/arm/configs/mxs_defconfig
@@ -160,7 +160,6 @@ CONFIG_NLS_CODEPAGE_850=y
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_ISO8859_15=y
 CONFIG_CRYPTO_DEV_MXS_DCP=y
-CONFIG_CRC_ITU_T=m
 CONFIG_FONTS=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_KERNEL=y
diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
index 113d6dfe5243..75b326bc7830 100644
--- a/arch/arm/configs/omap2plus_defconfig
+++ b/arch/arm/configs/omap2plus_defconfig
@@ -706,9 +706,6 @@ CONFIG_CRYPTO_DEV_OMAP=m
 CONFIG_CRYPTO_DEV_OMAP_SHAM=m
 CONFIG_CRYPTO_DEV_OMAP_AES=m
 CONFIG_CRYPTO_DEV_OMAP_DES=m
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_T10DIF=y
-CONFIG_CRC_ITU_T=y
 CONFIG_DMA_CMA=y
 CONFIG_FONTS=y
 CONFIG_FONT_8x8=y
diff --git a/arch/arm/configs/orion5x_defconfig b/arch/arm/configs/orion5x_defconfig
index 0629b088a584..62b9c6102789 100644
--- a/arch/arm/configs/orion5x_defconfig
+++ b/arch/arm/configs/orion5x_defconfig
@@ -136,7 +136,6 @@ CONFIG_CRYPTO_CBC=m
 CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_DEV_MARVELL_CESA=y
-CONFIG_CRC_T10DIF=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/arm/configs/pxa168_defconfig b/arch/arm/configs/pxa168_defconfig
index ce10fe2104bf..4748c7d33cb8 100644
--- a/arch/arm/configs/pxa168_defconfig
+++ b/arch/arm/configs/pxa168_defconfig
@@ -41,7 +41,6 @@ CONFIG_NFS_V3=y
 CONFIG_NFS_V3_ACL=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/arm/configs/pxa910_defconfig b/arch/arm/configs/pxa910_defconfig
index 1f28aea86014..49b59c600ae1 100644
--- a/arch/arm/configs/pxa910_defconfig
+++ b/arch/arm/configs/pxa910_defconfig
@@ -50,7 +50,6 @@ CONFIG_NFS_V3=y
 CONFIG_NFS_V3_ACL=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/arm/configs/pxa_defconfig b/arch/arm/configs/pxa_defconfig
index de0ac8f521d7..24fca8608554 100644
--- a/arch/arm/configs/pxa_defconfig
+++ b/arch/arm/configs/pxa_defconfig
@@ -663,8 +663,6 @@ CONFIG_CRYPTO_SHA1_ARM=m
 CONFIG_CRYPTO_SHA256_ARM=m
 CONFIG_CRYPTO_SHA512_ARM=m
 CONFIG_CRYPTO_AES_ARM=m
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_T10DIF=m
 CONFIG_FONTS=y
 CONFIG_FONT_8x8=y
 CONFIG_FONT_8x16=y
diff --git a/arch/arm/configs/s5pv210_defconfig b/arch/arm/configs/s5pv210_defconfig
index 5dbe85c263de..02121eec3658 100644
--- a/arch/arm/configs/s5pv210_defconfig
+++ b/arch/arm/configs/s5pv210_defconfig
@@ -113,7 +113,6 @@ CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_UTF8=y
-CONFIG_CRC_CCITT=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/arm/configs/sama7_defconfig b/arch/arm/configs/sama7_defconfig
index ea7ddf640ba7..e14720a9a5ac 100644
--- a/arch/arm/configs/sama7_defconfig
+++ b/arch/arm/configs/sama7_defconfig
@@ -227,8 +227,6 @@ CONFIG_CRYPTO_USER_API_SKCIPHER=m
 CONFIG_CRYPTO_DEV_ATMEL_AES=y
 CONFIG_CRYPTO_DEV_ATMEL_TDES=y
 CONFIG_CRYPTO_DEV_ATMEL_SHA=y
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_ITU_T=y
 CONFIG_DMA_CMA=y
 CONFIG_CMA_SIZE_MBYTES=32
 CONFIG_CMA_ALIGNMENT=9
diff --git a/arch/arm/configs/spitz_defconfig b/arch/arm/configs/spitz_defconfig
index ac5b7a5aaff6..ffec59e3f49c 100644
--- a/arch/arm/configs/spitz_defconfig
+++ b/arch/arm/configs/spitz_defconfig
@@ -234,7 +234,6 @@ CONFIG_CRYPTO_MD4=m
 CONFIG_CRYPTO_MICHAEL_MIC=m
 CONFIG_CRYPTO_SHA512=m
 CONFIG_CRYPTO_WP512=m
-CONFIG_CRC_CCITT=y
 CONFIG_FONTS=y
 CONFIG_FONT_8x8=y
 CONFIG_FONT_8x16=y
diff --git a/arch/arm/configs/stm32_defconfig b/arch/arm/configs/stm32_defconfig
index 423bb41c4225..dcd9c316072e 100644
--- a/arch/arm/configs/stm32_defconfig
+++ b/arch/arm/configs/stm32_defconfig
@@ -74,7 +74,6 @@ CONFIG_EXT3_FS=y
 # CONFIG_DNOTIFY is not set
 # CONFIG_INOTIFY_USER is not set
 CONFIG_NLS=y
-CONFIG_CRC_ITU_T=y
 CONFIG_PRINTK_TIME=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/arm/configs/wpcm450_defconfig b/arch/arm/configs/wpcm450_defconfig
index 5e4397f7f828..cd4b3e70ff68 100644
--- a/arch/arm/configs/wpcm450_defconfig
+++ b/arch/arm/configs/wpcm450_defconfig
@@ -191,8 +191,6 @@ CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=y
 CONFIG_X509_CERTIFICATE_PARSER=y
 CONFIG_PKCS7_MESSAGE_PARSER=y
 CONFIG_SYSTEM_TRUSTED_KEYRING=y
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_ITU_T=m
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/arm64/boot/dts/qcom/x1e80100.dtsi b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
index 4936fa5b98ff..8eddf0c96098 100644
--- a/arch/arm64/boot/dts/qcom/x1e80100.dtsi
+++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
@@ -3752,60 +3752,83 @@
 			};
 
 			gpu_opp_table: opp-table {
-				compatible = "operating-points-v2";
+				compatible = "operating-points-v2-adreno", "operating-points-v2";
+
+				opp-1250000000 {
+					opp-hz = /bits/ 64 <1250000000>;
+					opp-level = <RPMH_REGULATOR_LEVEL_TURBO_L3>;
+					opp-peak-kBps = <16500000>;
+					qcom,opp-acd-level = <0xa82a5ffd>;
+				};
+
+				opp-1175000000 {
+					opp-hz = /bits/ 64 <1175000000>;
+					opp-level = <RPMH_REGULATOR_LEVEL_TURBO_L2>;
+					opp-peak-kBps = <14398438>;
+					qcom,opp-acd-level = <0xa82a5ffd>;
+				};
 
 				opp-1100000000 {
 					opp-hz = /bits/ 64 <1100000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_TURBO_L1>;
-					opp-peak-kBps = <16500000>;
+					opp-peak-kBps = <14398438>;
+					qcom,opp-acd-level = <0xa82a5ffd>;
 				};
 
 				opp-1000000000 {
 					opp-hz = /bits/ 64 <1000000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_TURBO>;
 					opp-peak-kBps = <14398438>;
+					qcom,opp-acd-level = <0xa82b5ffd>;
 				};
 
 				opp-925000000 {
 					opp-hz = /bits/ 64 <925000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_NOM_L1>;
 					opp-peak-kBps = <14398438>;
+					qcom,opp-acd-level = <0xa82b5ffd>;
 				};
 
 				opp-800000000 {
 					opp-hz = /bits/ 64 <800000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_NOM>;
 					opp-peak-kBps = <12449219>;
+					qcom,opp-acd-level = <0xa82c5ffd>;
 				};
 
 				opp-744000000 {
 					opp-hz = /bits/ 64 <744000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_SVS_L2>;
 					opp-peak-kBps = <10687500>;
+					qcom,opp-acd-level = <0x882e5ffd>;
 				};
 
 				opp-687000000 {
 					opp-hz = /bits/ 64 <687000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_SVS_L1>;
 					opp-peak-kBps = <8171875>;
+					qcom,opp-acd-level = <0x882e5ffd>;
 				};
 
 				opp-550000000 {
 					opp-hz = /bits/ 64 <550000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_SVS>;
 					opp-peak-kBps = <6074219>;
+					qcom,opp-acd-level = <0xc0285ffd>;
 				};
 
 				opp-390000000 {
 					opp-hz = /bits/ 64 <390000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_LOW_SVS>;
 					opp-peak-kBps = <3000000>;
+					qcom,opp-acd-level = <0xc0285ffd>;
 				};
 
 				opp-300000000 {
 					opp-hz = /bits/ 64 <300000000>;
 					opp-level = <RPMH_REGULATOR_LEVEL_LOW_SVS_D1>;
 					opp-peak-kBps = <2136719>;
+					qcom,opp-acd-level = <0xc02b5ffd>;
 				};
 			};
 		};
diff --git a/arch/arm64/include/asm/esr.h b/arch/arm64/include/asm/esr.h
index d1b1a33f9a8b..e4f77757937e 100644
--- a/arch/arm64/include/asm/esr.h
+++ b/arch/arm64/include/asm/esr.h
@@ -121,6 +121,15 @@
 #define ESR_ELx_FSC_SEA_TTW(n)	(0x14 + (n))
 #define ESR_ELx_FSC_SECC	(0x18)
 #define ESR_ELx_FSC_SECC_TTW(n)	(0x1c + (n))
+#define ESR_ELx_FSC_ADDRSZ	(0x00)
+
+/*
+ * Annoyingly, the negative levels for Address size faults aren't laid out
+ * contiguously (or in the desired order)
+ */
+#define ESR_ELx_FSC_ADDRSZ_nL(n)	((n) == -1 ? 0x25 : 0x2C)
+#define ESR_ELx_FSC_ADDRSZ_L(n)		((n) < 0 ? ESR_ELx_FSC_ADDRSZ_nL(n) : \
+						   (ESR_ELx_FSC_ADDRSZ + (n)))
 
 /* Status codes for individual page table levels */
 #define ESR_ELx_FSC_ACCESS_L(n)	(ESR_ELx_FSC_ACCESS + (n))
@@ -161,8 +170,6 @@
 #define ESR_ELx_Xs_MASK		(GENMASK_ULL(4, 0))
 
 /* ISS field definitions for exceptions taken in to Hyp */
-#define ESR_ELx_FSC_ADDRSZ	(0x00)
-#define ESR_ELx_FSC_ADDRSZ_L(n)	(ESR_ELx_FSC_ADDRSZ + (n))
 #define ESR_ELx_CV		(UL(1) << 24)
 #define ESR_ELx_COND_SHIFT	(20)
 #define ESR_ELx_COND_MASK	(UL(0xF) << ESR_ELx_COND_SHIFT)
@@ -464,6 +471,39 @@ static inline bool esr_fsc_is_access_flag_fault(unsigned long esr)
 	       (esr == ESR_ELx_FSC_ACCESS_L(0));
 }
 
+static inline bool esr_fsc_is_addr_sz_fault(unsigned long esr)
+{
+	esr &= ESR_ELx_FSC;
+
+	return (esr == ESR_ELx_FSC_ADDRSZ_L(3))	||
+	       (esr == ESR_ELx_FSC_ADDRSZ_L(2))	||
+	       (esr == ESR_ELx_FSC_ADDRSZ_L(1)) ||
+	       (esr == ESR_ELx_FSC_ADDRSZ_L(0))	||
+	       (esr == ESR_ELx_FSC_ADDRSZ_L(-1));
+}
+
+static inline bool esr_fsc_is_sea_ttw(unsigned long esr)
+{
+	esr = esr & ESR_ELx_FSC;
+
+	return (esr == ESR_ELx_FSC_SEA_TTW(3)) ||
+	       (esr == ESR_ELx_FSC_SEA_TTW(2)) ||
+	       (esr == ESR_ELx_FSC_SEA_TTW(1)) ||
+	       (esr == ESR_ELx_FSC_SEA_TTW(0)) ||
+	       (esr == ESR_ELx_FSC_SEA_TTW(-1));
+}
+
+static inline bool esr_fsc_is_secc_ttw(unsigned long esr)
+{
+	esr = esr & ESR_ELx_FSC;
+
+	return (esr == ESR_ELx_FSC_SECC_TTW(3)) ||
+	       (esr == ESR_ELx_FSC_SECC_TTW(2)) ||
+	       (esr == ESR_ELx_FSC_SECC_TTW(1)) ||
+	       (esr == ESR_ELx_FSC_SECC_TTW(0)) ||
+	       (esr == ESR_ELx_FSC_SECC_TTW(-1));
+}
+
 /* Indicate whether ESR.EC==0x1A is for an ERETAx instruction */
 static inline bool esr_iss_is_eretax(unsigned long esr)
 {
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index d7cf66573aca..bd020fc28aa9 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -305,7 +305,12 @@ static __always_inline unsigned long kvm_vcpu_get_hfar(const struct kvm_vcpu *vc
 
 static __always_inline phys_addr_t kvm_vcpu_get_fault_ipa(const struct kvm_vcpu *vcpu)
 {
-	return ((phys_addr_t)vcpu->arch.fault.hpfar_el2 & HPFAR_MASK) << 8;
+	u64 hpfar = vcpu->arch.fault.hpfar_el2;
+
+	if (unlikely(!(hpfar & HPFAR_EL2_NS)))
+		return INVALID_GPA;
+
+	return FIELD_GET(HPFAR_EL2_FIPA, hpfar) << 12;
 }
 
 static inline u64 kvm_vcpu_get_disr(const struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index e98cfe7855a6..08ba91e6fb03 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -1588,4 +1588,9 @@ void kvm_set_vm_id_reg(struct kvm *kvm, u32 reg, u64 val);
 #define kvm_has_s1poe(k)				\
 	(kvm_has_feat((k), ID_AA64MMFR3_EL1, S1POE, IMP))
 
+static inline bool kvm_arch_has_irq_bypass(void)
+{
+	return true;
+}
+
 #endif /* __ARM64_KVM_HOST_H__ */
diff --git a/arch/arm64/include/asm/kvm_ras.h b/arch/arm64/include/asm/kvm_ras.h
index 87e10d9a635b..9398ade632aa 100644
--- a/arch/arm64/include/asm/kvm_ras.h
+++ b/arch/arm64/include/asm/kvm_ras.h
@@ -14,7 +14,7 @@
  * Was this synchronous external abort a RAS notification?
  * Returns '0' for errors handled by some RAS subsystem, or -ENOENT.
  */
-static inline int kvm_handle_guest_sea(phys_addr_t addr, u64 esr)
+static inline int kvm_handle_guest_sea(void)
 {
 	/* apei_claim_sea(NULL) expects to mask interrupts itself */
 	lockdep_assert_irqs_enabled();
diff --git a/arch/arm64/include/asm/mmu.h b/arch/arm64/include/asm/mmu.h
index 30a29e88994b..6e8aa8e72601 100644
--- a/arch/arm64/include/asm/mmu.h
+++ b/arch/arm64/include/asm/mmu.h
@@ -94,17 +94,6 @@ static inline bool kaslr_requires_kpti(void)
 			return false;
 	}
 
-	/*
-	 * Systems affected by Cavium erratum 24756 are incompatible
-	 * with KPTI.
-	 */
-	if (IS_ENABLED(CONFIG_CAVIUM_ERRATUM_27456)) {
-		extern const struct midr_range cavium_erratum_27456_cpus[];
-
-		if (is_midr_in_range_list(cavium_erratum_27456_cpus))
-			return false;
-	}
-
 	return true;
 }
 
diff --git a/arch/arm64/include/asm/rqspinlock.h b/arch/arm64/include/asm/rqspinlock.h
index 5b80785324b6..9ea0a74e5892 100644
--- a/arch/arm64/include/asm/rqspinlock.h
+++ b/arch/arm64/include/asm/rqspinlock.h
@@ -86,7 +86,7 @@
 
 #endif
 
-#define res_smp_cond_load_acquire_timewait(v, c) smp_cond_load_acquire_timewait(v, c, 0, 1)
+#define res_smp_cond_load_acquire(v, c) smp_cond_load_acquire_timewait(v, c, 0, 1)
 
 #include <asm-generic/rqspinlock.h>
 
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index b55f5f705750..6b0ad5070d3e 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -335,7 +335,7 @@ static const struct midr_range cavium_erratum_23154_cpus[] = {
 #endif
 
 #ifdef CONFIG_CAVIUM_ERRATUM_27456
-const struct midr_range cavium_erratum_27456_cpus[] = {
+static const struct midr_range cavium_erratum_27456_cpus[] = {
 	/* Cavium ThunderX, T88 pass 1.x - 2.1 */
 	MIDR_RANGE(MIDR_THUNDERX, 0, 0, 1, 1),
 	/* Cavium ThunderX, T81 pass 1.0 */
diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h
index 5e3c4b58f279..2004b4f41ade 100644
--- a/arch/arm64/kernel/image-vars.h
+++ b/arch/arm64/kernel/image-vars.h
@@ -47,10 +47,6 @@ PROVIDE(__pi_id_aa64smfr0_override	= id_aa64smfr0_override);
 PROVIDE(__pi_id_aa64zfr0_override	= id_aa64zfr0_override);
 PROVIDE(__pi_arm64_sw_feature_override	= arm64_sw_feature_override);
 PROVIDE(__pi_arm64_use_ng_mappings	= arm64_use_ng_mappings);
-#ifdef CONFIG_CAVIUM_ERRATUM_27456
-PROVIDE(__pi_cavium_erratum_27456_cpus	= cavium_erratum_27456_cpus);
-PROVIDE(__pi_is_midr_in_range_list	= is_midr_in_range_list);
-#endif
 PROVIDE(__pi__ctype			= _ctype);
 PROVIDE(__pi_memstart_offset_seed	= memstart_offset_seed);
 
diff --git a/arch/arm64/kernel/pi/map_kernel.c b/arch/arm64/kernel/pi/map_kernel.c
index e57b043f324b..c6650cfe706c 100644
--- a/arch/arm64/kernel/pi/map_kernel.c
+++ b/arch/arm64/kernel/pi/map_kernel.c
@@ -207,6 +207,29 @@ static void __init map_fdt(u64 fdt)
 	dsb(ishst);
 }
 
+/*
+ * PI version of the Cavium Eratum 27456 detection, which makes it
+ * impossible to use non-global mappings.
+ */
+static bool __init ng_mappings_allowed(void)
+{
+	static const struct midr_range cavium_erratum_27456_cpus[] __initconst = {
+		/* Cavium ThunderX, T88 pass 1.x - 2.1 */
+		MIDR_RANGE(MIDR_THUNDERX, 0, 0, 1, 1),
+		/* Cavium ThunderX, T81 pass 1.0 */
+		MIDR_REV(MIDR_THUNDERX_81XX, 0, 0),
+		{},
+	};
+
+	for (const struct midr_range *r = cavium_erratum_27456_cpus; r->model; r++) {
+		if (midr_is_cpu_model_range(read_cpuid_id(), r->model,
+					    r->rv_min, r->rv_max))
+			return false;
+	}
+
+	return true;
+}
+
 asmlinkage void __init early_map_kernel(u64 boot_status, void *fdt)
 {
 	static char const chosen_str[] __initconst = "/chosen";
@@ -246,7 +269,7 @@ asmlinkage void __init early_map_kernel(u64 boot_status, void *fdt)
 		u64 kaslr_seed = kaslr_early_init(fdt, chosen);
 
 		if (kaslr_seed && kaslr_requires_kpti())
-			arm64_use_ng_mappings = true;
+			arm64_use_ng_mappings = ng_mappings_allowed();
 
 		kaslr_offset |= kaslr_seed & ~(MIN_KIMG_ALIGN - 1);
 	}
diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c
index b198dde79e59..b607f6dfc5e6 100644
--- a/arch/arm64/kernel/proton-pack.c
+++ b/arch/arm64/kernel/proton-pack.c
@@ -879,10 +879,12 @@ static u8 spectre_bhb_loop_affected(void)
 	static const struct midr_range spectre_bhb_k132_list[] = {
 		MIDR_ALL_VERSIONS(MIDR_CORTEX_X3),
 		MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V2),
+		{},
 	};
 	static const struct midr_range spectre_bhb_k38_list[] = {
 		MIDR_ALL_VERSIONS(MIDR_CORTEX_A715),
 		MIDR_ALL_VERSIONS(MIDR_CORTEX_A720),
+		{},
 	};
 	static const struct midr_range spectre_bhb_k32_list[] = {
 		MIDR_ALL_VERSIONS(MIDR_CORTEX_A78),
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 68fec8c95fee..19ca57def629 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -2743,11 +2743,6 @@ bool kvm_arch_irqchip_in_kernel(struct kvm *kvm)
 	return irqchip_in_kernel(kvm);
 }
 
-bool kvm_arch_has_irq_bypass(void)
-{
-	return true;
-}
-
 int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons,
 				      struct irq_bypass_producer *prod)
 {
diff --git a/arch/arm64/kvm/hyp/include/hyp/fault.h b/arch/arm64/kvm/hyp/include/hyp/fault.h
index 17df94570f03..fc573fc767b0 100644
--- a/arch/arm64/kvm/hyp/include/hyp/fault.h
+++ b/arch/arm64/kvm/hyp/include/hyp/fault.h
@@ -12,6 +12,16 @@
 #include <asm/kvm_hyp.h>
 #include <asm/kvm_mmu.h>
 
+static inline bool __fault_safe_to_translate(u64 esr)
+{
+	u64 fsc = esr & ESR_ELx_FSC;
+
+	if (esr_fsc_is_sea_ttw(esr) || esr_fsc_is_secc_ttw(esr))
+		return false;
+
+	return !(fsc == ESR_ELx_FSC_EXTABT && (esr & ESR_ELx_FnV));
+}
+
 static inline bool __translate_far_to_hpfar(u64 far, u64 *hpfar)
 {
 	int ret;
@@ -44,34 +54,50 @@ static inline bool __translate_far_to_hpfar(u64 far, u64 *hpfar)
 	return true;
 }
 
-static inline bool __get_fault_info(u64 esr, struct kvm_vcpu_fault_info *fault)
+/*
+ * Checks for the conditions when HPFAR_EL2 is written, per ARM ARM R_FKLWR.
+ */
+static inline bool __hpfar_valid(u64 esr)
 {
-	u64 hpfar, far;
-
-	far = read_sysreg_el2(SYS_FAR);
-
 	/*
-	 * The HPFAR can be invalid if the stage 2 fault did not
-	 * happen during a stage 1 page table walk (the ESR_EL2.S1PTW
-	 * bit is clear) and one of the two following cases are true:
-	 *   1. The fault was due to a permission fault
-	 *   2. The processor carries errata 834220
+	 * CPUs affected by ARM erratum #834220 may incorrectly report a
+	 * stage-2 translation fault when a stage-1 permission fault occurs.
 	 *
-	 * Therefore, for all non S1PTW faults where we either have a
-	 * permission fault or the errata workaround is enabled, we
-	 * resolve the IPA using the AT instruction.
+	 * Re-walk the page tables to determine if a stage-1 fault actually
+	 * occurred.
 	 */
-	if (!(esr & ESR_ELx_S1PTW) &&
-	    (cpus_have_final_cap(ARM64_WORKAROUND_834220) ||
-	     esr_fsc_is_permission_fault(esr))) {
-		if (!__translate_far_to_hpfar(far, &hpfar))
-			return false;
-	} else {
+	if (cpus_have_final_cap(ARM64_WORKAROUND_834220) &&
+	    esr_fsc_is_translation_fault(esr))
+		return false;
+
+	if (esr_fsc_is_translation_fault(esr) || esr_fsc_is_access_flag_fault(esr))
+		return true;
+
+	if ((esr & ESR_ELx_S1PTW) && esr_fsc_is_permission_fault(esr))
+		return true;
+
+	return esr_fsc_is_addr_sz_fault(esr);
+}
+
+static inline bool __get_fault_info(u64 esr, struct kvm_vcpu_fault_info *fault)
+{
+	u64 hpfar;
+
+	fault->far_el2		= read_sysreg_el2(SYS_FAR);
+	fault->hpfar_el2	= 0;
+
+	if (__hpfar_valid(esr))
 		hpfar = read_sysreg(hpfar_el2);
-	}
+	else if (unlikely(!__fault_safe_to_translate(esr)))
+		return true;
+	else if (!__translate_far_to_hpfar(fault->far_el2, &hpfar))
+		return false;
 
-	fault->far_el2 = far;
-	fault->hpfar_el2 = hpfar;
+	/*
+	 * Hijack HPFAR_EL2.NS (RES0 in Non-secure) to indicate a valid
+	 * HPFAR value.
+	 */
+	fault->hpfar_el2 = hpfar | HPFAR_EL2_NS;
 	return true;
 }
 
diff --git a/arch/arm64/kvm/hyp/nvhe/ffa.c b/arch/arm64/kvm/hyp/nvhe/ffa.c
index e433dfab882a..3369dd0c4009 100644
--- a/arch/arm64/kvm/hyp/nvhe/ffa.c
+++ b/arch/arm64/kvm/hyp/nvhe/ffa.c
@@ -730,10 +730,10 @@ static void do_ffa_version(struct arm_smccc_res *res,
 		hyp_ffa_version = ffa_req_version;
 	}
 
-	if (hyp_ffa_post_init())
+	if (hyp_ffa_post_init()) {
 		res->a0 = FFA_RET_NOT_SUPPORTED;
-	else {
-		has_version_negotiated = true;
+	} else {
+		smp_store_release(&has_version_negotiated, true);
 		res->a0 = hyp_ffa_version;
 	}
 unlock:
@@ -809,7 +809,8 @@ bool kvm_host_ffa_handler(struct kvm_cpu_context *host_ctxt, u32 func_id)
 	if (!is_ffa_call(func_id))
 		return false;
 
-	if (!has_version_negotiated && func_id != FFA_VERSION) {
+	if (func_id != FFA_VERSION &&
+	    !smp_load_acquire(&has_version_negotiated)) {
 		ffa_to_smccc_error(&res, FFA_RET_INVALID_PARAMETERS);
 		goto out_handled;
 	}
diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
index f34f11c720d7..2a5284f749b4 100644
--- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c
+++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
@@ -578,7 +578,14 @@ void handle_host_mem_abort(struct kvm_cpu_context *host_ctxt)
 		return;
 	}
 
-	addr = (fault.hpfar_el2 & HPFAR_MASK) << 8;
+
+	/*
+	 * Yikes, we couldn't resolve the fault IPA. This should reinject an
+	 * abort into the host when we figure out how to do that.
+	 */
+	BUG_ON(!(fault.hpfar_el2 & HPFAR_EL2_NS));
+	addr = FIELD_GET(HPFAR_EL2_FIPA, fault.hpfar_el2) << 12;
+
 	ret = host_stage2_idmap(addr);
 	BUG_ON(ret && ret != -EAGAIN);
 }
diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
index 2feb6c6b63af..754f2fe0cc67 100644
--- a/arch/arm64/kvm/mmu.c
+++ b/arch/arm64/kvm/mmu.c
@@ -1794,9 +1794,28 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu)
 	gfn_t gfn;
 	int ret, idx;
 
+	/* Synchronous External Abort? */
+	if (kvm_vcpu_abt_issea(vcpu)) {
+		/*
+		 * For RAS the host kernel may handle this abort.
+		 * There is no need to pass the error into the guest.
+		 */
+		if (kvm_handle_guest_sea())
+			kvm_inject_vabt(vcpu);
+
+		return 1;
+	}
+
 	esr = kvm_vcpu_get_esr(vcpu);
 
+	/*
+	 * The fault IPA should be reliable at this point as we're not dealing
+	 * with an SEA.
+	 */
 	ipa = fault_ipa = kvm_vcpu_get_fault_ipa(vcpu);
+	if (KVM_BUG_ON(ipa == INVALID_GPA, vcpu->kvm))
+		return -EFAULT;
+
 	is_iabt = kvm_vcpu_trap_is_iabt(vcpu);
 
 	if (esr_fsc_is_translation_fault(esr)) {
@@ -1818,18 +1837,6 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu)
 		}
 	}
 
-	/* Synchronous External Abort? */
-	if (kvm_vcpu_abt_issea(vcpu)) {
-		/*
-		 * For RAS the host kernel may handle this abort.
-		 * There is no need to pass the error into the guest.
-		 */
-		if (kvm_handle_guest_sea(fault_ipa, kvm_vcpu_get_esr(vcpu)))
-			kvm_inject_vabt(vcpu);
-
-		return 1;
-	}
-
 	trace_kvm_guest_fault(*vcpu_pc(vcpu), kvm_vcpu_get_esr(vcpu),
 			      kvm_vcpu_get_hfar(vcpu), fault_ipa);
 
diff --git a/arch/arm64/tools/sysreg b/arch/arm64/tools/sysreg
index f9476848a2ed..bdf044c5d11b 100644
--- a/arch/arm64/tools/sysreg
+++ b/arch/arm64/tools/sysreg
@@ -3536,3 +3536,10 @@ Field	5	F
 Field	4	P
 Field	3:0	Align
 EndSysreg
+
+Sysreg	HPFAR_EL2	3	4	6	0	4
+Field	63	NS
+Res0	62:48
+Field	47:4	FIPA
+Res0	3:0
+EndSysreg
diff --git a/arch/hexagon/configs/comet_defconfig b/arch/hexagon/configs/comet_defconfig
index 469c025297c6..c6108f000288 100644
--- a/arch/hexagon/configs/comet_defconfig
+++ b/arch/hexagon/configs/comet_defconfig
@@ -72,9 +72,6 @@ CONFIG_INET=y
 CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=y
-CONFIG_CRC16=y
-CONFIG_CRC_T10DIF=y
 CONFIG_FRAME_WARN=0
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_FS=y
diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig
index 067c0b994648..1a2cf012b8f2 100644
--- a/arch/loongarch/Kconfig
+++ b/arch/loongarch/Kconfig
@@ -73,6 +73,7 @@ config LOONGARCH
 	select ARCH_SUPPORTS_RT
 	select ARCH_USE_BUILTIN_BSWAP
 	select ARCH_USE_CMPXCHG_LOCKREF
+	select ARCH_USE_MEMTEST
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS
 	select ARCH_WANT_DEFAULT_BPF_JIT
diff --git a/arch/loongarch/include/asm/fpu.h b/arch/loongarch/include/asm/fpu.h
index 3177674228f8..45514f314664 100644
--- a/arch/loongarch/include/asm/fpu.h
+++ b/arch/loongarch/include/asm/fpu.h
@@ -22,22 +22,29 @@
 struct sigcontext;
 
 #define kernel_fpu_available() cpu_has_fpu
-extern void kernel_fpu_begin(void);
-extern void kernel_fpu_end(void);
-
-extern void _init_fpu(unsigned int);
-extern void _save_fp(struct loongarch_fpu *);
-extern void _restore_fp(struct loongarch_fpu *);
-
-extern void _save_lsx(struct loongarch_fpu *fpu);
-extern void _restore_lsx(struct loongarch_fpu *fpu);
-extern void _init_lsx_upper(void);
-extern void _restore_lsx_upper(struct loongarch_fpu *fpu);
-
-extern void _save_lasx(struct loongarch_fpu *fpu);
-extern void _restore_lasx(struct loongarch_fpu *fpu);
-extern void _init_lasx_upper(void);
-extern void _restore_lasx_upper(struct loongarch_fpu *fpu);
+
+void kernel_fpu_begin(void);
+void kernel_fpu_end(void);
+
+asmlinkage void _init_fpu(unsigned int);
+asmlinkage void _save_fp(struct loongarch_fpu *);
+asmlinkage void _restore_fp(struct loongarch_fpu *);
+asmlinkage int _save_fp_context(void __user *fpregs, void __user *fcc, void __user *csr);
+asmlinkage int _restore_fp_context(void __user *fpregs, void __user *fcc, void __user *csr);
+
+asmlinkage void _save_lsx(struct loongarch_fpu *fpu);
+asmlinkage void _restore_lsx(struct loongarch_fpu *fpu);
+asmlinkage void _init_lsx_upper(void);
+asmlinkage void _restore_lsx_upper(struct loongarch_fpu *fpu);
+asmlinkage int _save_lsx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
+asmlinkage int _restore_lsx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
+
+asmlinkage void _save_lasx(struct loongarch_fpu *fpu);
+asmlinkage void _restore_lasx(struct loongarch_fpu *fpu);
+asmlinkage void _init_lasx_upper(void);
+asmlinkage void _restore_lasx_upper(struct loongarch_fpu *fpu);
+asmlinkage int _save_lasx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
+asmlinkage int _restore_lasx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
 
 static inline void enable_lsx(void);
 static inline void disable_lsx(void);
diff --git a/arch/loongarch/include/asm/lbt.h b/arch/loongarch/include/asm/lbt.h
index e671978bf552..38566574e562 100644
--- a/arch/loongarch/include/asm/lbt.h
+++ b/arch/loongarch/include/asm/lbt.h
@@ -12,9 +12,13 @@
 #include <asm/loongarch.h>
 #include <asm/processor.h>
 
-extern void _init_lbt(void);
-extern void _save_lbt(struct loongarch_lbt *);
-extern void _restore_lbt(struct loongarch_lbt *);
+asmlinkage void _init_lbt(void);
+asmlinkage void _save_lbt(struct loongarch_lbt *);
+asmlinkage void _restore_lbt(struct loongarch_lbt *);
+asmlinkage int _save_lbt_context(void __user *regs, void __user *eflags);
+asmlinkage int _restore_lbt_context(void __user *regs, void __user *eflags);
+asmlinkage int _save_ftop_context(void __user *ftop);
+asmlinkage int _restore_ftop_context(void __user *ftop);
 
 static inline int is_lbt_enabled(void)
 {
diff --git a/arch/loongarch/include/asm/ptrace.h b/arch/loongarch/include/asm/ptrace.h
index f3ddaed9ef7f..a5b63c84f854 100644
--- a/arch/loongarch/include/asm/ptrace.h
+++ b/arch/loongarch/include/asm/ptrace.h
@@ -33,9 +33,9 @@ struct pt_regs {
 	unsigned long __last[];
 } __aligned(8);
 
-static inline int regs_irqs_disabled(struct pt_regs *regs)
+static __always_inline bool regs_irqs_disabled(struct pt_regs *regs)
 {
-	return arch_irqs_disabled_flags(regs->csr_prmd);
+	return !(regs->csr_prmd & CSR_PRMD_PIE);
 }
 
 static inline unsigned long kernel_stack_pointer(struct pt_regs *regs)
diff --git a/arch/loongarch/kernel/Makefile b/arch/loongarch/kernel/Makefile
index 4853e8b04c6f..f9dcaa60033d 100644
--- a/arch/loongarch/kernel/Makefile
+++ b/arch/loongarch/kernel/Makefile
@@ -21,10 +21,10 @@ obj-$(CONFIG_CPU_HAS_LBT)	+= lbt.o
 
 obj-$(CONFIG_ARCH_STRICT_ALIGN)	+= unaligned.o
 
-CFLAGS_module.o		+= $(call cc-option,-Wno-override-init,)
-CFLAGS_syscall.o	+= $(call cc-option,-Wno-override-init,)
-CFLAGS_traps.o		+= $(call cc-option,-Wno-override-init,)
-CFLAGS_perf_event.o	+= $(call cc-option,-Wno-override-init,)
+CFLAGS_module.o		+= $(call cc-disable-warning, override-init)
+CFLAGS_syscall.o	+= $(call cc-disable-warning, override-init)
+CFLAGS_traps.o		+= $(call cc-disable-warning, override-init)
+CFLAGS_perf_event.o	+= $(call cc-disable-warning, override-init)
 
 ifdef CONFIG_FUNCTION_TRACER
   ifndef CONFIG_DYNAMIC_FTRACE
diff --git a/arch/loongarch/kernel/fpu.S b/arch/loongarch/kernel/fpu.S
index 6ab640101457..28caf416ae36 100644
--- a/arch/loongarch/kernel/fpu.S
+++ b/arch/loongarch/kernel/fpu.S
@@ -458,6 +458,7 @@ SYM_FUNC_START(_save_fp_context)
 	li.w		a0, 0				# success
 	jr		ra
 SYM_FUNC_END(_save_fp_context)
+EXPORT_SYMBOL_GPL(_save_fp_context)
 
 /*
  * a0: fpregs
@@ -471,6 +472,7 @@ SYM_FUNC_START(_restore_fp_context)
 	li.w		a0, 0				# success
 	jr		ra
 SYM_FUNC_END(_restore_fp_context)
+EXPORT_SYMBOL_GPL(_restore_fp_context)
 
 /*
  * a0: fpregs
@@ -484,6 +486,7 @@ SYM_FUNC_START(_save_lsx_context)
 	li.w	a0, 0					# success
 	jr	ra
 SYM_FUNC_END(_save_lsx_context)
+EXPORT_SYMBOL_GPL(_save_lsx_context)
 
 /*
  * a0: fpregs
@@ -497,6 +500,7 @@ SYM_FUNC_START(_restore_lsx_context)
 	li.w	a0, 0					# success
 	jr	ra
 SYM_FUNC_END(_restore_lsx_context)
+EXPORT_SYMBOL_GPL(_restore_lsx_context)
 
 /*
  * a0: fpregs
@@ -510,6 +514,7 @@ SYM_FUNC_START(_save_lasx_context)
 	li.w	a0, 0					# success
 	jr	ra
 SYM_FUNC_END(_save_lasx_context)
+EXPORT_SYMBOL_GPL(_save_lasx_context)
 
 /*
  * a0: fpregs
@@ -523,6 +528,7 @@ SYM_FUNC_START(_restore_lasx_context)
 	li.w	a0, 0					# success
 	jr	ra
 SYM_FUNC_END(_restore_lasx_context)
+EXPORT_SYMBOL_GPL(_restore_lasx_context)
 
 .L_fpu_fault:
 	li.w	a0, -EFAULT				# failure
diff --git a/arch/loongarch/kernel/lbt.S b/arch/loongarch/kernel/lbt.S
index 001f061d226a..71678912d24c 100644
--- a/arch/loongarch/kernel/lbt.S
+++ b/arch/loongarch/kernel/lbt.S
@@ -90,6 +90,7 @@ SYM_FUNC_START(_save_lbt_context)
 	li.w		a0, 0			# success
 	jr		ra
 SYM_FUNC_END(_save_lbt_context)
+EXPORT_SYMBOL_GPL(_save_lbt_context)
 
 /*
  * a0: scr
@@ -110,6 +111,7 @@ SYM_FUNC_START(_restore_lbt_context)
 	li.w		a0, 0			# success
 	jr		ra
 SYM_FUNC_END(_restore_lbt_context)
+EXPORT_SYMBOL_GPL(_restore_lbt_context)
 
 /*
  * a0: ftop
@@ -120,6 +122,7 @@ SYM_FUNC_START(_save_ftop_context)
 	li.w		a0, 0			# success
 	jr		ra
 SYM_FUNC_END(_save_ftop_context)
+EXPORT_SYMBOL_GPL(_save_ftop_context)
 
 /*
  * a0: ftop
@@ -150,6 +153,7 @@ SYM_FUNC_START(_restore_ftop_context)
 	li.w		a0, 0			# success
 	jr		ra
 SYM_FUNC_END(_restore_ftop_context)
+EXPORT_SYMBOL_GPL(_restore_ftop_context)
 
 .L_lbt_fault:
 	li.w		a0, -EFAULT		# failure
diff --git a/arch/loongarch/kernel/signal.c b/arch/loongarch/kernel/signal.c
index 7a555b600171..4740cb5b2388 100644
--- a/arch/loongarch/kernel/signal.c
+++ b/arch/loongarch/kernel/signal.c
@@ -51,27 +51,6 @@
 #define lock_lbt_owner()	({ preempt_disable(); pagefault_disable(); })
 #define unlock_lbt_owner()	({ pagefault_enable(); preempt_enable(); })
 
-/* Assembly functions to move context to/from the FPU */
-extern asmlinkage int
-_save_fp_context(void __user *fpregs, void __user *fcc, void __user *csr);
-extern asmlinkage int
-_restore_fp_context(void __user *fpregs, void __user *fcc, void __user *csr);
-extern asmlinkage int
-_save_lsx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
-extern asmlinkage int
-_restore_lsx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
-extern asmlinkage int
-_save_lasx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
-extern asmlinkage int
-_restore_lasx_context(void __user *fpregs, void __user *fcc, void __user *fcsr);
-
-#ifdef CONFIG_CPU_HAS_LBT
-extern asmlinkage int _save_lbt_context(void __user *regs, void __user *eflags);
-extern asmlinkage int _restore_lbt_context(void __user *regs, void __user *eflags);
-extern asmlinkage int _save_ftop_context(void __user *ftop);
-extern asmlinkage int _restore_ftop_context(void __user *ftop);
-#endif
-
 struct rt_sigframe {
 	struct siginfo rs_info;
 	struct ucontext rs_uctx;
diff --git a/arch/loongarch/kernel/traps.c b/arch/loongarch/kernel/traps.c
index 2ec3106c0da3..47fc2de6d150 100644
--- a/arch/loongarch/kernel/traps.c
+++ b/arch/loongarch/kernel/traps.c
@@ -553,9 +553,10 @@ asmlinkage void noinstr do_ale(struct pt_regs *regs)
 	die_if_kernel("Kernel ale access", regs);
 	force_sig_fault(SIGBUS, BUS_ADRALN, (void __user *)regs->csr_badvaddr);
 #else
+	bool pie = regs_irqs_disabled(regs);
 	unsigned int *pc;
 
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_enable();
 
 	perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS, 1, regs, regs->csr_badvaddr);
@@ -582,7 +583,7 @@ sigbus:
 	die_if_kernel("Kernel ale access", regs);
 	force_sig_fault(SIGBUS, BUS_ADRALN, (void __user *)regs->csr_badvaddr);
 out:
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_disable();
 #endif
 	irqentry_exit(regs, state);
@@ -621,12 +622,13 @@ static void bug_handler(struct pt_regs *regs)
 asmlinkage void noinstr do_bce(struct pt_regs *regs)
 {
 	bool user = user_mode(regs);
+	bool pie = regs_irqs_disabled(regs);
 	unsigned long era = exception_era(regs);
 	u64 badv = 0, lower = 0, upper = ULONG_MAX;
 	union loongarch_instruction insn;
 	irqentry_state_t state = irqentry_enter(regs);
 
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_enable();
 
 	current->thread.trap_nr = read_csr_excode();
@@ -692,7 +694,7 @@ asmlinkage void noinstr do_bce(struct pt_regs *regs)
 	force_sig_bnderr((void __user *)badv, (void __user *)lower, (void __user *)upper);
 
 out:
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_disable();
 
 	irqentry_exit(regs, state);
@@ -710,11 +712,12 @@ bad_era:
 asmlinkage void noinstr do_bp(struct pt_regs *regs)
 {
 	bool user = user_mode(regs);
+	bool pie = regs_irqs_disabled(regs);
 	unsigned int opcode, bcode;
 	unsigned long era = exception_era(regs);
 	irqentry_state_t state = irqentry_enter(regs);
 
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_enable();
 
 	if (__get_inst(&opcode, (u32 *)era, user))
@@ -780,7 +783,7 @@ asmlinkage void noinstr do_bp(struct pt_regs *regs)
 	}
 
 out:
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_disable();
 
 	irqentry_exit(regs, state);
@@ -1015,6 +1018,7 @@ static void init_restore_lbt(void)
 
 asmlinkage void noinstr do_lbt(struct pt_regs *regs)
 {
+	bool pie = regs_irqs_disabled(regs);
 	irqentry_state_t state = irqentry_enter(regs);
 
 	/*
@@ -1024,7 +1028,7 @@ asmlinkage void noinstr do_lbt(struct pt_regs *regs)
 	 * (including the user using 'MOVGR2GCSR' to turn on TM, which
 	 * will not trigger the BTE), we need to check PRMD first.
 	 */
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_enable();
 
 	if (!cpu_has_lbt) {
@@ -1038,7 +1042,7 @@ asmlinkage void noinstr do_lbt(struct pt_regs *regs)
 	preempt_enable();
 
 out:
-	if (regs->csr_prmd & CSR_PRMD_PIE)
+	if (!pie)
 		local_irq_disable();
 
 	irqentry_exit(regs, state);
diff --git a/arch/loongarch/kvm/Makefile b/arch/loongarch/kvm/Makefile
index f4c8e35c216a..cb41d9265662 100644
--- a/arch/loongarch/kvm/Makefile
+++ b/arch/loongarch/kvm/Makefile
@@ -21,4 +21,4 @@ kvm-y += intc/eiointc.o
 kvm-y += intc/pch_pic.o
 kvm-y += irqfd.o
 
-CFLAGS_exit.o	+= $(call cc-option,-Wno-override-init,)
+CFLAGS_exit.o	+= $(call cc-disable-warning, override-init)
diff --git a/arch/loongarch/kvm/intc/ipi.c b/arch/loongarch/kvm/intc/ipi.c
index 93f4acd44523..fe734dc062ed 100644
--- a/arch/loongarch/kvm/intc/ipi.c
+++ b/arch/loongarch/kvm/intc/ipi.c
@@ -111,7 +111,7 @@ static int send_ipi_data(struct kvm_vcpu *vcpu, gpa_t addr, uint64_t data)
 		ret = kvm_io_bus_read(vcpu, KVM_IOCSR_BUS, addr, sizeof(val), &val);
 		srcu_read_unlock(&vcpu->kvm->srcu, idx);
 		if (unlikely(ret)) {
-			kvm_err("%s: : read date from addr %llx failed\n", __func__, addr);
+			kvm_err("%s: : read data from addr %llx failed\n", __func__, addr);
 			return ret;
 		}
 		/* Construct the mask by scanning the bit 27-30 */
@@ -127,7 +127,7 @@ static int send_ipi_data(struct kvm_vcpu *vcpu, gpa_t addr, uint64_t data)
 	ret = kvm_io_bus_write(vcpu, KVM_IOCSR_BUS, addr, sizeof(val), &val);
 	srcu_read_unlock(&vcpu->kvm->srcu, idx);
 	if (unlikely(ret))
-		kvm_err("%s: : write date to addr %llx failed\n", __func__, addr);
+		kvm_err("%s: : write data to addr %llx failed\n", __func__, addr);
 
 	return ret;
 }
diff --git a/arch/loongarch/kvm/main.c b/arch/loongarch/kvm/main.c
index d165cd38c6bb..80ea63d465b8 100644
--- a/arch/loongarch/kvm/main.c
+++ b/arch/loongarch/kvm/main.c
@@ -296,10 +296,10 @@ int kvm_arch_enable_virtualization_cpu(void)
 	/*
 	 * Enable virtualization features granting guest direct control of
 	 * certain features:
-	 * GCI=2:       Trap on init or unimplement cache instruction.
+	 * GCI=2:       Trap on init or unimplemented cache instruction.
 	 * TORU=0:      Trap on Root Unimplement.
 	 * CACTRL=1:    Root control cache.
-	 * TOP=0:       Trap on Previlege.
+	 * TOP=0:       Trap on Privilege.
 	 * TOE=0:       Trap on Exception.
 	 * TIT=0:       Trap on Timer.
 	 */
diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c
index 8e427b379661..5af32ec62cb1 100644
--- a/arch/loongarch/kvm/vcpu.c
+++ b/arch/loongarch/kvm/vcpu.c
@@ -294,6 +294,7 @@ static int kvm_pre_enter_guest(struct kvm_vcpu *vcpu)
 		vcpu->arch.aux_inuse &= ~KVM_LARCH_SWCSR_LATEST;
 
 		if (kvm_request_pending(vcpu) || xfer_to_guest_mode_work_pending()) {
+			kvm_lose_pmu(vcpu);
 			/* make sure the vcpu mode has been written */
 			smp_store_mb(vcpu->mode, OUTSIDE_GUEST_MODE);
 			local_irq_enable();
@@ -902,6 +903,13 @@ static int kvm_set_one_reg(struct kvm_vcpu *vcpu,
 			vcpu->arch.st.guest_addr = 0;
 			memset(&vcpu->arch.irq_pending, 0, sizeof(vcpu->arch.irq_pending));
 			memset(&vcpu->arch.irq_clear, 0, sizeof(vcpu->arch.irq_clear));
+
+			/*
+			 * When vCPU reset, clear the ESTAT and GINTC registers
+			 * Other CSR registers are cleared with function _kvm_setcsr().
+			 */
+			kvm_write_sw_gcsr(vcpu->arch.csr, LOONGARCH_CSR_GINTC, 0);
+			kvm_write_sw_gcsr(vcpu->arch.csr, LOONGARCH_CSR_ESTAT, 0);
 			break;
 		default:
 			ret = -EINVAL;
diff --git a/arch/loongarch/mm/hugetlbpage.c b/arch/loongarch/mm/hugetlbpage.c
index e4068906143b..cea84d7f2b91 100644
--- a/arch/loongarch/mm/hugetlbpage.c
+++ b/arch/loongarch/mm/hugetlbpage.c
@@ -47,7 +47,7 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr,
 				pmd = pmd_offset(pud, addr);
 		}
 	}
-	return (pte_t *) pmd;
+	return pmd_none(pmdp_get(pmd)) ? NULL : (pte_t *) pmd;
 }
 
 uint64_t pmd_to_entrylo(unsigned long pmd_val)
diff --git a/arch/loongarch/mm/init.c b/arch/loongarch/mm/init.c
index fdb7f73ad160..06f11d9e4ec1 100644
--- a/arch/loongarch/mm/init.c
+++ b/arch/loongarch/mm/init.c
@@ -65,9 +65,6 @@ void __init paging_init(void)
 {
 	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
-#ifdef CONFIG_ZONE_DMA
-	max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN;
-#endif
 #ifdef CONFIG_ZONE_DMA32
 	max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN;
 #endif
diff --git a/arch/m68k/configs/amcore_defconfig b/arch/m68k/configs/amcore_defconfig
index 67a0d157122d..110279a64aa4 100644
--- a/arch/m68k/configs/amcore_defconfig
+++ b/arch/m68k/configs/amcore_defconfig
@@ -89,4 +89,3 @@ CONFIG_PANIC_ON_OOPS=y
 # CONFIG_CRYPTO_ECHAINIV is not set
 CONFIG_CRYPTO_ANSI_CPRNG=y
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC16=y
diff --git a/arch/mips/configs/ath79_defconfig b/arch/mips/configs/ath79_defconfig
index 8caa03a41327..cba0b85c6707 100644
--- a/arch/mips/configs/ath79_defconfig
+++ b/arch/mips/configs/ath79_defconfig
@@ -82,7 +82,6 @@ CONFIG_LEDS_GPIO=y
 # CONFIG_IOMMU_SUPPORT is not set
 # CONFIG_DNOTIFY is not set
 # CONFIG_PROC_PAGE_MONITOR is not set
-CONFIG_CRC_ITU_T=m
 CONFIG_STRIP_ASM_SYMS=y
 CONFIG_DEBUG_FS=y
 # CONFIG_SCHED_DEBUG is not set
diff --git a/arch/mips/configs/bigsur_defconfig b/arch/mips/configs/bigsur_defconfig
index fe282630b51c..8f7c36868204 100644
--- a/arch/mips/configs/bigsur_defconfig
+++ b/arch/mips/configs/bigsur_defconfig
@@ -238,7 +238,6 @@ CONFIG_CRYPTO_SERPENT=m
 CONFIG_CRYPTO_TEA=m
 CONFIG_CRYPTO_TWOFISH=m
 CONFIG_CRYPTO_LZO=m
-CONFIG_CRC_T10DIF=m
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_MEMORY_INIT=y
 CONFIG_DETECT_HUNG_TASK=y
diff --git a/arch/mips/configs/fuloong2e_defconfig b/arch/mips/configs/fuloong2e_defconfig
index 5ab149cd3178..114fcd67898d 100644
--- a/arch/mips/configs/fuloong2e_defconfig
+++ b/arch/mips/configs/fuloong2e_defconfig
@@ -218,4 +218,3 @@ CONFIG_CRYPTO_SEED=m
 CONFIG_CRYPTO_DEFLATE=m
 CONFIG_CRYPTO_LZO=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=y
diff --git a/arch/mips/configs/ip22_defconfig b/arch/mips/configs/ip22_defconfig
index 31ca93d3acc5..f1a8ccf2c459 100644
--- a/arch/mips/configs/ip22_defconfig
+++ b/arch/mips/configs/ip22_defconfig
@@ -326,5 +326,4 @@ CONFIG_CRYPTO_TEA=m
 CONFIG_CRYPTO_TWOFISH=m
 CONFIG_CRYPTO_LZO=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_T10DIF=m
 CONFIG_DEBUG_MEMORY_INIT=y
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index b8907b3d7a33..5d079941fd20 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -317,4 +317,3 @@ CONFIG_CRYPTO_SERPENT=m
 CONFIG_CRYPTO_TEA=m
 CONFIG_CRYPTO_TWOFISH=m
 CONFIG_CRYPTO_LZO=m
-CONFIG_CRC_T10DIF=m
diff --git a/arch/mips/configs/ip30_defconfig b/arch/mips/configs/ip30_defconfig
index 270181a7320a..a4524e785469 100644
--- a/arch/mips/configs/ip30_defconfig
+++ b/arch/mips/configs/ip30_defconfig
@@ -179,4 +179,3 @@ CONFIG_CRYPTO_RMD160=m
 CONFIG_CRYPTO_WP512=m
 CONFIG_CRYPTO_XCBC=m
 CONFIG_CRYPTO_LZO=m
-CONFIG_CRC_T10DIF=m
diff --git a/arch/mips/configs/ip32_defconfig b/arch/mips/configs/ip32_defconfig
index 121e7e48fa77..d8ac11427f69 100644
--- a/arch/mips/configs/ip32_defconfig
+++ b/arch/mips/configs/ip32_defconfig
@@ -177,7 +177,6 @@ CONFIG_CRYPTO_SERPENT=y
 CONFIG_CRYPTO_TEA=y
 CONFIG_CRYPTO_TWOFISH=y
 CONFIG_CRYPTO_DEFLATE=y
-CONFIG_CRC_T10DIF=y
 CONFIG_FONTS=y
 CONFIG_FONT_8x8=y
 CONFIG_FONT_8x16=y
diff --git a/arch/mips/configs/omega2p_defconfig b/arch/mips/configs/omega2p_defconfig
index 128f9abab7fc..e2bcdfd290a1 100644
--- a/arch/mips/configs/omega2p_defconfig
+++ b/arch/mips/configs/omega2p_defconfig
@@ -111,7 +111,6 @@ CONFIG_NLS_KOI8_U=y
 CONFIG_NLS_UTF8=y
 CONFIG_CRYPTO_DEFLATE=y
 CONFIG_CRYPTO_LZO=y
-CONFIG_CRC16=y
 CONFIG_XZ_DEC=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/mips/configs/rb532_defconfig b/arch/mips/configs/rb532_defconfig
index 0261969a6e45..42b161d587c7 100644
--- a/arch/mips/configs/rb532_defconfig
+++ b/arch/mips/configs/rb532_defconfig
@@ -155,5 +155,4 @@ CONFIG_JFFS2_COMPRESSION_OPTIONS=y
 CONFIG_SQUASHFS=y
 CONFIG_CRYPTO_TEST=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC16=m
 CONFIG_STRIP_ASM_SYMS=y
diff --git a/arch/mips/configs/rt305x_defconfig b/arch/mips/configs/rt305x_defconfig
index 8404e0a9d8b2..8f9701efef19 100644
--- a/arch/mips/configs/rt305x_defconfig
+++ b/arch/mips/configs/rt305x_defconfig
@@ -128,7 +128,6 @@ CONFIG_SQUASHFS=y
 # CONFIG_SQUASHFS_ZLIB is not set
 CONFIG_SQUASHFS_XZ=y
 CONFIG_CRYPTO_ARC4=m
-CONFIG_CRC_ITU_T=m
 # CONFIG_XZ_DEC_X86 is not set
 # CONFIG_XZ_DEC_POWERPC is not set
 # CONFIG_XZ_DEC_IA64 is not set
diff --git a/arch/mips/configs/sb1250_swarm_defconfig b/arch/mips/configs/sb1250_swarm_defconfig
index ce855b644bb0..ae2afff00e01 100644
--- a/arch/mips/configs/sb1250_swarm_defconfig
+++ b/arch/mips/configs/sb1250_swarm_defconfig
@@ -99,4 +99,3 @@ CONFIG_CRYPTO_TWOFISH=m
 CONFIG_CRYPTO_DEFLATE=m
 CONFIG_CRYPTO_LZO=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC16=m
diff --git a/arch/mips/configs/vocore2_defconfig b/arch/mips/configs/vocore2_defconfig
index 917967fed45f..2a9a9b12847d 100644
--- a/arch/mips/configs/vocore2_defconfig
+++ b/arch/mips/configs/vocore2_defconfig
@@ -111,7 +111,6 @@ CONFIG_NLS_KOI8_U=y
 CONFIG_NLS_UTF8=y
 CONFIG_CRYPTO_DEFLATE=y
 CONFIG_CRYPTO_LZO=y
-CONFIG_CRC16=y
 CONFIG_XZ_DEC=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
diff --git a/arch/mips/configs/xway_defconfig b/arch/mips/configs/xway_defconfig
index 7b91edfe3e07..aae8497b6872 100644
--- a/arch/mips/configs/xway_defconfig
+++ b/arch/mips/configs/xway_defconfig
@@ -140,7 +140,6 @@ CONFIG_SQUASHFS=y
 # CONFIG_SQUASHFS_ZLIB is not set
 CONFIG_SQUASHFS_XZ=y
 CONFIG_CRYPTO_ARC4=m
-CONFIG_CRC_ITU_T=m
 CONFIG_PRINTK_TIME=y
 CONFIG_STRIP_ASM_SYMS=y
 CONFIG_DEBUG_FS=y
diff --git a/arch/openrisc/include/asm/cacheflush.h b/arch/openrisc/include/asm/cacheflush.h
index 984c331ff5f4..0e60af486ec1 100644
--- a/arch/openrisc/include/asm/cacheflush.h
+++ b/arch/openrisc/include/asm/cacheflush.h
@@ -23,6 +23,9 @@
  */
 extern void local_dcache_page_flush(struct page *page);
 extern void local_icache_page_inv(struct page *page);
+extern void local_dcache_range_flush(unsigned long start, unsigned long end);
+extern void local_dcache_range_inv(unsigned long start, unsigned long end);
+extern void local_icache_range_inv(unsigned long start, unsigned long end);
 
 /*
  * Data cache flushing always happen on the local cpu. Instruction cache
@@ -39,6 +42,20 @@ extern void smp_icache_page_inv(struct page *page);
 #endif /* CONFIG_SMP */
 
 /*
+ * Even if the actual block size is larger than L1_CACHE_BYTES, paddr
+ * can be incremented by L1_CACHE_BYTES. When paddr is written to the
+ * invalidate register, the entire cache line encompassing this address
+ * is invalidated. Each subsequent reference to the same cache line will
+ * not affect the invalidation process.
+ */
+#define local_dcache_block_flush(addr) \
+	local_dcache_range_flush(addr, addr + L1_CACHE_BYTES)
+#define local_dcache_block_inv(addr) \
+	local_dcache_range_inv(addr, addr + L1_CACHE_BYTES)
+#define local_icache_block_inv(addr) \
+	local_icache_range_inv(addr, addr + L1_CACHE_BYTES)
+
+/*
  * Synchronizes caches. Whenever a cpu writes executable code to memory, this
  * should be called to make sure the processor sees the newly written code.
  */
diff --git a/arch/openrisc/include/asm/cpuinfo.h b/arch/openrisc/include/asm/cpuinfo.h
index 5e4744153d0e..3cfc4cf0b019 100644
--- a/arch/openrisc/include/asm/cpuinfo.h
+++ b/arch/openrisc/include/asm/cpuinfo.h
@@ -15,16 +15,21 @@
 #ifndef __ASM_OPENRISC_CPUINFO_H
 #define __ASM_OPENRISC_CPUINFO_H
 
+#include <asm/spr.h>
+#include <asm/spr_defs.h>
+
+struct cache_desc {
+	u32 size;
+	u32 sets;
+	u32 block_size;
+	u32 ways;
+};
+
 struct cpuinfo_or1k {
 	u32 clock_frequency;
 
-	u32 icache_size;
-	u32 icache_block_size;
-	u32 icache_ways;
-
-	u32 dcache_size;
-	u32 dcache_block_size;
-	u32 dcache_ways;
+	struct cache_desc icache;
+	struct cache_desc dcache;
 
 	u16 coreid;
 };
@@ -32,4 +37,9 @@ struct cpuinfo_or1k {
 extern struct cpuinfo_or1k cpuinfo_or1k[NR_CPUS];
 extern void setup_cpuinfo(void);
 
+/*
+ * Check if the cache component exists.
+ */
+extern bool cpu_cache_is_present(const unsigned int cache_type);
+
 #endif /* __ASM_OPENRISC_CPUINFO_H */
diff --git a/arch/openrisc/kernel/Makefile b/arch/openrisc/kernel/Makefile
index 79129161f3e0..e4c7d9bdd598 100644
--- a/arch/openrisc/kernel/Makefile
+++ b/arch/openrisc/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y	:= vmlinux.lds
 
 obj-y	:= head.o setup.o or32_ksyms.o process.o dma.o \
 	   traps.o time.o irq.o entry.o ptrace.o signal.o \
-	   sys_call_table.o unwinder.o
+	   sys_call_table.o unwinder.o cacheinfo.o
 
 obj-$(CONFIG_SMP)		+= smp.o sync-timer.o
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
diff --git a/arch/openrisc/kernel/cacheinfo.c b/arch/openrisc/kernel/cacheinfo.c
new file mode 100644
index 000000000000..61230545e4ff
--- /dev/null
+++ b/arch/openrisc/kernel/cacheinfo.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * OpenRISC cacheinfo support
+ *
+ * Based on work done for MIPS and LoongArch. All original copyrights
+ * apply as per the original source declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2025 Sahil Siddiq <sahilcdq@proton.me>
+ */
+
+#include <linux/cacheinfo.h>
+#include <asm/cpuinfo.h>
+#include <asm/spr.h>
+#include <asm/spr_defs.h>
+
+static inline void ci_leaf_init(struct cacheinfo *this_leaf, enum cache_type type,
+				unsigned int level, struct cache_desc *cache, int cpu)
+{
+	this_leaf->type = type;
+	this_leaf->level = level;
+	this_leaf->coherency_line_size = cache->block_size;
+	this_leaf->number_of_sets = cache->sets;
+	this_leaf->ways_of_associativity = cache->ways;
+	this_leaf->size = cache->size;
+	cpumask_set_cpu(cpu, &this_leaf->shared_cpu_map);
+}
+
+int init_cache_level(unsigned int cpu)
+{
+	struct cpuinfo_or1k *cpuinfo = &cpuinfo_or1k[smp_processor_id()];
+	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
+	int leaves = 0, levels = 0;
+	unsigned long upr = mfspr(SPR_UPR);
+	unsigned long iccfgr, dccfgr;
+
+	if (!(upr & SPR_UPR_UP)) {
+		printk(KERN_INFO
+		       "-- no UPR register... unable to detect configuration\n");
+		return -ENOENT;
+	}
+
+	if (cpu_cache_is_present(SPR_UPR_DCP)) {
+		dccfgr = mfspr(SPR_DCCFGR);
+		cpuinfo->dcache.ways = 1 << (dccfgr & SPR_DCCFGR_NCW);
+		cpuinfo->dcache.sets = 1 << ((dccfgr & SPR_DCCFGR_NCS) >> 3);
+		cpuinfo->dcache.block_size = 16 << ((dccfgr & SPR_DCCFGR_CBS) >> 7);
+		cpuinfo->dcache.size =
+		    cpuinfo->dcache.sets * cpuinfo->dcache.ways * cpuinfo->dcache.block_size;
+		leaves += 1;
+		printk(KERN_INFO
+		       "-- dcache: %d bytes total, %d bytes/line, %d set(s), %d way(s)\n",
+		       cpuinfo->dcache.size, cpuinfo->dcache.block_size,
+		       cpuinfo->dcache.sets, cpuinfo->dcache.ways);
+	} else
+		printk(KERN_INFO "-- dcache disabled\n");
+
+	if (cpu_cache_is_present(SPR_UPR_ICP)) {
+		iccfgr = mfspr(SPR_ICCFGR);
+		cpuinfo->icache.ways = 1 << (iccfgr & SPR_ICCFGR_NCW);
+		cpuinfo->icache.sets = 1 << ((iccfgr & SPR_ICCFGR_NCS) >> 3);
+		cpuinfo->icache.block_size = 16 << ((iccfgr & SPR_ICCFGR_CBS) >> 7);
+		cpuinfo->icache.size =
+		    cpuinfo->icache.sets * cpuinfo->icache.ways * cpuinfo->icache.block_size;
+		leaves += 1;
+		printk(KERN_INFO
+		       "-- icache: %d bytes total, %d bytes/line, %d set(s), %d way(s)\n",
+		       cpuinfo->icache.size, cpuinfo->icache.block_size,
+		       cpuinfo->icache.sets, cpuinfo->icache.ways);
+	} else
+		printk(KERN_INFO "-- icache disabled\n");
+
+	if (!leaves)
+		return -ENOENT;
+
+	levels = 1;
+
+	this_cpu_ci->num_leaves = leaves;
+	this_cpu_ci->num_levels = levels;
+
+	return 0;
+}
+
+int populate_cache_leaves(unsigned int cpu)
+{
+	struct cpuinfo_or1k *cpuinfo = &cpuinfo_or1k[smp_processor_id()];
+	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
+	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
+	int level = 1;
+
+	if (cpu_cache_is_present(SPR_UPR_DCP)) {
+		ci_leaf_init(this_leaf, CACHE_TYPE_DATA, level, &cpuinfo->dcache, cpu);
+		this_leaf->attributes = ((mfspr(SPR_DCCFGR) & SPR_DCCFGR_CWS) >> 8) ?
+					CACHE_WRITE_BACK : CACHE_WRITE_THROUGH;
+		this_leaf++;
+	}
+
+	if (cpu_cache_is_present(SPR_UPR_ICP))
+		ci_leaf_init(this_leaf, CACHE_TYPE_INST, level, &cpuinfo->icache, cpu);
+
+	this_cpu_ci->cpu_map_populated = true;
+
+	return 0;
+}
diff --git a/arch/openrisc/kernel/dma.c b/arch/openrisc/kernel/dma.c
index b3edbb33b621..3a7b5baaa450 100644
--- a/arch/openrisc/kernel/dma.c
+++ b/arch/openrisc/kernel/dma.c
@@ -17,6 +17,7 @@
 #include <linux/pagewalk.h>
 
 #include <asm/cpuinfo.h>
+#include <asm/cacheflush.h>
 #include <asm/spr_defs.h>
 #include <asm/tlbflush.h>
 
@@ -24,9 +25,6 @@ static int
 page_set_nocache(pte_t *pte, unsigned long addr,
 		 unsigned long next, struct mm_walk *walk)
 {
-	unsigned long cl;
-	struct cpuinfo_or1k *cpuinfo = &cpuinfo_or1k[smp_processor_id()];
-
 	pte_val(*pte) |= _PAGE_CI;
 
 	/*
@@ -36,8 +34,7 @@ page_set_nocache(pte_t *pte, unsigned long addr,
 	flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
 
 	/* Flush page out of dcache */
-	for (cl = __pa(addr); cl < __pa(next); cl += cpuinfo->dcache_block_size)
-		mtspr(SPR_DCBFR, cl);
+	local_dcache_range_flush(__pa(addr), __pa(next));
 
 	return 0;
 }
@@ -98,21 +95,14 @@ void arch_dma_clear_uncached(void *cpu_addr, size_t size)
 void arch_sync_dma_for_device(phys_addr_t addr, size_t size,
 		enum dma_data_direction dir)
 {
-	unsigned long cl;
-	struct cpuinfo_or1k *cpuinfo = &cpuinfo_or1k[smp_processor_id()];
-
 	switch (dir) {
 	case DMA_TO_DEVICE:
 		/* Flush the dcache for the requested range */
-		for (cl = addr; cl < addr + size;
-		     cl += cpuinfo->dcache_block_size)
-			mtspr(SPR_DCBFR, cl);
+		local_dcache_range_flush(addr, addr + size);
 		break;
 	case DMA_FROM_DEVICE:
 		/* Invalidate the dcache for the requested range */
-		for (cl = addr; cl < addr + size;
-		     cl += cpuinfo->dcache_block_size)
-			mtspr(SPR_DCBIR, cl);
+		local_dcache_range_inv(addr, addr + size);
 		break;
 	default:
 		/*
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setup.c
index be56eaafc8b9..a9fb9cc6779e 100644
--- a/arch/openrisc/kernel/setup.c
+++ b/arch/openrisc/kernel/setup.c
@@ -113,21 +113,6 @@ static void print_cpuinfo(void)
 		return;
 	}
 
-	if (upr & SPR_UPR_DCP)
-		printk(KERN_INFO
-		       "-- dcache: %4d bytes total, %2d bytes/line, %d way(s)\n",
-		       cpuinfo->dcache_size, cpuinfo->dcache_block_size,
-		       cpuinfo->dcache_ways);
-	else
-		printk(KERN_INFO "-- dcache disabled\n");
-	if (upr & SPR_UPR_ICP)
-		printk(KERN_INFO
-		       "-- icache: %4d bytes total, %2d bytes/line, %d way(s)\n",
-		       cpuinfo->icache_size, cpuinfo->icache_block_size,
-		       cpuinfo->icache_ways);
-	else
-		printk(KERN_INFO "-- icache disabled\n");
-
 	if (upr & SPR_UPR_DMP)
 		printk(KERN_INFO "-- dmmu: %4d entries, %lu way(s)\n",
 		       1 << ((mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTS) >> 2),
@@ -155,8 +140,6 @@ static void print_cpuinfo(void)
 void __init setup_cpuinfo(void)
 {
 	struct device_node *cpu;
-	unsigned long iccfgr, dccfgr;
-	unsigned long cache_set_size;
 	int cpu_id = smp_processor_id();
 	struct cpuinfo_or1k *cpuinfo = &cpuinfo_or1k[cpu_id];
 
@@ -164,20 +147,6 @@ void __init setup_cpuinfo(void)
 	if (!cpu)
 		panic("Couldn't find CPU%d in device tree...\n", cpu_id);
 
-	iccfgr = mfspr(SPR_ICCFGR);
-	cpuinfo->icache_ways = 1 << (iccfgr & SPR_ICCFGR_NCW);
-	cache_set_size = 1 << ((iccfgr & SPR_ICCFGR_NCS) >> 3);
-	cpuinfo->icache_block_size = 16 << ((iccfgr & SPR_ICCFGR_CBS) >> 7);
-	cpuinfo->icache_size =
-	    cache_set_size * cpuinfo->icache_ways * cpuinfo->icache_block_size;
-
-	dccfgr = mfspr(SPR_DCCFGR);
-	cpuinfo->dcache_ways = 1 << (dccfgr & SPR_DCCFGR_NCW);
-	cache_set_size = 1 << ((dccfgr & SPR_DCCFGR_NCS) >> 3);
-	cpuinfo->dcache_block_size = 16 << ((dccfgr & SPR_DCCFGR_CBS) >> 7);
-	cpuinfo->dcache_size =
-	    cache_set_size * cpuinfo->dcache_ways * cpuinfo->dcache_block_size;
-
 	if (of_property_read_u32(cpu, "clock-frequency",
 				 &cpuinfo->clock_frequency)) {
 		printk(KERN_WARNING
@@ -294,14 +263,14 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	unsigned int vr, cpucfgr;
 	unsigned int avr;
 	unsigned int version;
+#ifdef CONFIG_SMP
 	struct cpuinfo_or1k *cpuinfo = v;
+	seq_printf(m, "processor\t\t: %d\n", cpuinfo->coreid);
+#endif
 
 	vr = mfspr(SPR_VR);
 	cpucfgr = mfspr(SPR_CPUCFGR);
 
-#ifdef CONFIG_SMP
-	seq_printf(m, "processor\t\t: %d\n", cpuinfo->coreid);
-#endif
 	if (vr & SPR_VR_UVRP) {
 		vr = mfspr(SPR_VR2);
 		version = vr & SPR_VR2_VER;
@@ -320,14 +289,6 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		seq_printf(m, "revision\t\t: %d\n", vr & SPR_VR_REV);
 	}
 	seq_printf(m, "frequency\t\t: %ld\n", loops_per_jiffy * HZ);
-	seq_printf(m, "dcache size\t\t: %d bytes\n", cpuinfo->dcache_size);
-	seq_printf(m, "dcache block size\t: %d bytes\n",
-		   cpuinfo->dcache_block_size);
-	seq_printf(m, "dcache ways\t\t: %d\n", cpuinfo->dcache_ways);
-	seq_printf(m, "icache size\t\t: %d bytes\n", cpuinfo->icache_size);
-	seq_printf(m, "icache block size\t: %d bytes\n",
-		   cpuinfo->icache_block_size);
-	seq_printf(m, "icache ways\t\t: %d\n", cpuinfo->icache_ways);
 	seq_printf(m, "immu\t\t\t: %d entries, %lu ways\n",
 		   1 << ((mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTS) >> 2),
 		   1 + (mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTW));
diff --git a/arch/openrisc/mm/cache.c b/arch/openrisc/mm/cache.c
index eb43b73f3855..0f265b8e73ec 100644
--- a/arch/openrisc/mm/cache.c
+++ b/arch/openrisc/mm/cache.c
@@ -14,31 +14,70 @@
 #include <asm/spr_defs.h>
 #include <asm/cache.h>
 #include <asm/cacheflush.h>
+#include <asm/cpuinfo.h>
 #include <asm/tlbflush.h>
 
-static __always_inline void cache_loop(struct page *page, const unsigned int reg)
+/*
+ * Check if the cache component exists.
+ */
+bool cpu_cache_is_present(const unsigned int cache_type)
 {
-	unsigned long paddr = page_to_pfn(page) << PAGE_SHIFT;
-	unsigned long line = paddr & ~(L1_CACHE_BYTES - 1);
+	unsigned long upr = mfspr(SPR_UPR);
+	unsigned long mask = SPR_UPR_UP | cache_type;
+
+	return !((upr & mask) ^ mask);
+}
+
+static __always_inline void cache_loop(unsigned long paddr, unsigned long end,
+				       const unsigned short reg, const unsigned int cache_type)
+{
+	if (!cpu_cache_is_present(cache_type))
+		return;
 
-	while (line < paddr + PAGE_SIZE) {
-		mtspr(reg, line);
-		line += L1_CACHE_BYTES;
+	while (paddr < end) {
+		mtspr(reg, paddr);
+		paddr += L1_CACHE_BYTES;
 	}
 }
 
+static __always_inline void cache_loop_page(struct page *page, const unsigned short reg,
+					    const unsigned int cache_type)
+{
+	unsigned long paddr = page_to_pfn(page) << PAGE_SHIFT;
+	unsigned long end = paddr + PAGE_SIZE;
+
+	paddr &= ~(L1_CACHE_BYTES - 1);
+
+	cache_loop(paddr, end, reg, cache_type);
+}
+
 void local_dcache_page_flush(struct page *page)
 {
-	cache_loop(page, SPR_DCBFR);
+	cache_loop_page(page, SPR_DCBFR, SPR_UPR_DCP);
 }
 EXPORT_SYMBOL(local_dcache_page_flush);
 
 void local_icache_page_inv(struct page *page)
 {
-	cache_loop(page, SPR_ICBIR);
+	cache_loop_page(page, SPR_ICBIR, SPR_UPR_ICP);
 }
 EXPORT_SYMBOL(local_icache_page_inv);
 
+void local_dcache_range_flush(unsigned long start, unsigned long end)
+{
+	cache_loop(start, end, SPR_DCBFR, SPR_UPR_DCP);
+}
+
+void local_dcache_range_inv(unsigned long start, unsigned long end)
+{
+	cache_loop(start, end, SPR_DCBIR, SPR_UPR_DCP);
+}
+
+void local_icache_range_inv(unsigned long start, unsigned long end)
+{
+	cache_loop(start, end, SPR_ICBIR, SPR_UPR_ICP);
+}
+
 void update_cache(struct vm_area_struct *vma, unsigned long address,
 	pte_t *pte)
 {
@@ -58,4 +97,3 @@ void update_cache(struct vm_area_struct *vma, unsigned long address,
 			sync_icache_dcache(folio_page(folio, nr));
 	}
 }
-
diff --git a/arch/openrisc/mm/init.c b/arch/openrisc/mm/init.c
index be1c2eb8bb94..e4904ca6f0a0 100644
--- a/arch/openrisc/mm/init.c
+++ b/arch/openrisc/mm/init.c
@@ -35,6 +35,7 @@
 #include <asm/fixmap.h>
 #include <asm/tlbflush.h>
 #include <asm/sections.h>
+#include <asm/cacheflush.h>
 
 int mem_init_done;
 
@@ -176,8 +177,8 @@ void __init paging_init(void)
 	barrier();
 
 	/* Invalidate instruction caches after code modification */
-	mtspr(SPR_ICBIR, 0x900);
-	mtspr(SPR_ICBIR, 0xa00);
+	local_icache_block_inv(0x900);
+	local_icache_block_inv(0xa00);
 
 	/* New TLB miss handlers and kernel page tables are in now place.
 	 * Make sure that page flags get updated for all pages in TLB by
diff --git a/arch/parisc/configs/generic-32bit_defconfig b/arch/parisc/configs/generic-32bit_defconfig
index f5fffc24c3bc..5b65c9859613 100644
--- a/arch/parisc/configs/generic-32bit_defconfig
+++ b/arch/parisc/configs/generic-32bit_defconfig
@@ -264,8 +264,6 @@ CONFIG_CRYPTO_MICHAEL_MIC=m
 CONFIG_CRYPTO_SHA1=y
 CONFIG_CRYPTO_WP512=m
 CONFIG_CRYPTO_DEFLATE=y
-CONFIG_CRC_CCITT=m
-CONFIG_CRC_T10DIF=y
 CONFIG_FONTS=y
 CONFIG_PRINTK_TIME=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/parisc/configs/generic-64bit_defconfig b/arch/parisc/configs/generic-64bit_defconfig
index 2487765b7be3..ecc9ffcc11cd 100644
--- a/arch/parisc/configs/generic-64bit_defconfig
+++ b/arch/parisc/configs/generic-64bit_defconfig
@@ -292,7 +292,6 @@ CONFIG_CRYPTO_MD5=y
 CONFIG_CRYPTO_MICHAEL_MIC=m
 CONFIG_CRYPTO_DEFLATE=m
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=m
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_STRIP_ASM_SYMS=y
diff --git a/arch/parisc/math-emu/driver.c b/arch/parisc/math-emu/driver.c
index 34495446e051..71829cb7bc81 100644
--- a/arch/parisc/math-emu/driver.c
+++ b/arch/parisc/math-emu/driver.c
@@ -97,9 +97,19 @@ handle_fpe(struct pt_regs *regs)
 
 	memcpy(regs->fr, frcopy, sizeof regs->fr);
 	if (signalcode != 0) {
-	    force_sig_fault(signalcode >> 24, signalcode & 0xffffff,
-			    (void __user *) regs->iaoq[0]);
-	    return -1;
+		int sig = signalcode >> 24;
+
+		if (sig == SIGFPE) {
+			/*
+			 * Clear floating point trap bit to avoid trapping
+			 * again on the first floating-point instruction in
+			 * the userspace signal handler.
+			 */
+			regs->fr[0] &= ~(1ULL << 38);
+		}
+		force_sig_fault(sig, signalcode & 0xffffff,
+				(void __user *) regs->iaoq[0]);
+		return -1;
 	}
 
 	return signalcode ? -1 : 0;
diff --git a/arch/powerpc/boot/wrapper b/arch/powerpc/boot/wrapper
index 1db60fe13802..3d8dc822282a 100755
--- a/arch/powerpc/boot/wrapper
+++ b/arch/powerpc/boot/wrapper
@@ -234,10 +234,8 @@ fi
 
 # suppress some warnings in recent ld versions
 nowarn="-z noexecstack"
-if ! ld_is_lld; then
-	if [ "$LD_VERSION" -ge "$(echo 2.39 | ld_version)" ]; then
-		nowarn="$nowarn --no-warn-rwx-segments"
-	fi
+if "${CROSS}ld" -v --no-warn-rwx-segments >/dev/null 2>&1; then
+	nowarn="$nowarn --no-warn-rwx-segments"
 fi
 
 platformo=$object/"$platform".o
diff --git a/arch/powerpc/configs/44x/sam440ep_defconfig b/arch/powerpc/configs/44x/sam440ep_defconfig
index 2479ab62d12f..98221bda380d 100644
--- a/arch/powerpc/configs/44x/sam440ep_defconfig
+++ b/arch/powerpc/configs/44x/sam440ep_defconfig
@@ -91,5 +91,4 @@ CONFIG_AFFS_FS=m
 # CONFIG_NETWORK_FILESYSTEMS is not set
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/powerpc/configs/44x/warp_defconfig b/arch/powerpc/configs/44x/warp_defconfig
index 20891c413149..5757625469c4 100644
--- a/arch/powerpc/configs/44x/warp_defconfig
+++ b/arch/powerpc/configs/44x/warp_defconfig
@@ -85,8 +85,6 @@ CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_ISO8859_15=y
 CONFIG_NLS_UTF8=y
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_T10DIF=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_DEBUG_FS=y
diff --git a/arch/powerpc/configs/83xx/mpc832x_rdb_defconfig b/arch/powerpc/configs/83xx/mpc832x_rdb_defconfig
index 1715ff547442..b99caba8724a 100644
--- a/arch/powerpc/configs/83xx/mpc832x_rdb_defconfig
+++ b/arch/powerpc/configs/83xx/mpc832x_rdb_defconfig
@@ -73,6 +73,5 @@ CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_CODEPAGE_932=y
 CONFIG_NLS_ISO8859_8=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
diff --git a/arch/powerpc/configs/83xx/mpc834x_itx_defconfig b/arch/powerpc/configs/83xx/mpc834x_itx_defconfig
index e65c0057147f..11163052fdba 100644
--- a/arch/powerpc/configs/83xx/mpc834x_itx_defconfig
+++ b/arch/powerpc/configs/83xx/mpc834x_itx_defconfig
@@ -80,5 +80,4 @@ CONFIG_TMPFS=y
 CONFIG_NFS_FS=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_PCBC=m
diff --git a/arch/powerpc/configs/83xx/mpc834x_itxgp_defconfig b/arch/powerpc/configs/83xx/mpc834x_itxgp_defconfig
index 17714bf0ed40..312d39e4242c 100644
--- a/arch/powerpc/configs/83xx/mpc834x_itxgp_defconfig
+++ b/arch/powerpc/configs/83xx/mpc834x_itxgp_defconfig
@@ -72,5 +72,4 @@ CONFIG_TMPFS=y
 CONFIG_NFS_FS=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_PCBC=m
diff --git a/arch/powerpc/configs/83xx/mpc837x_rdb_defconfig b/arch/powerpc/configs/83xx/mpc837x_rdb_defconfig
index 58fae5131fa7..ac27f99faab8 100644
--- a/arch/powerpc/configs/83xx/mpc837x_rdb_defconfig
+++ b/arch/powerpc/configs/83xx/mpc837x_rdb_defconfig
@@ -75,6 +75,5 @@ CONFIG_TMPFS=y
 CONFIG_NFS_FS=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
diff --git a/arch/powerpc/configs/85xx/ge_imp3a_defconfig b/arch/powerpc/configs/85xx/ge_imp3a_defconfig
index 6f58ee1edf1f..7beb36a41d45 100644
--- a/arch/powerpc/configs/85xx/ge_imp3a_defconfig
+++ b/arch/powerpc/configs/85xx/ge_imp3a_defconfig
@@ -221,8 +221,6 @@ CONFIG_NLS_ISO8859_15=y
 CONFIG_NLS_KOI8_R=m
 CONFIG_NLS_KOI8_U=m
 CONFIG_NLS_UTF8=y
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_CRYPTO_CBC=y
 CONFIG_CRYPTO_MD5=y
diff --git a/arch/powerpc/configs/85xx/stx_gp3_defconfig b/arch/powerpc/configs/85xx/stx_gp3_defconfig
index e7080497048d..0a42072fa23c 100644
--- a/arch/powerpc/configs/85xx/stx_gp3_defconfig
+++ b/arch/powerpc/configs/85xx/stx_gp3_defconfig
@@ -60,8 +60,6 @@ CONFIG_CRAMFS=m
 CONFIG_NFS_FS=y
 CONFIG_ROOT_NFS=y
 CONFIG_NLS=y
-CONFIG_CRC_CCITT=y
-CONFIG_CRC_T10DIF=m
 CONFIG_DETECT_HUNG_TASK=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
 CONFIG_BDI_SWITCH=y
diff --git a/arch/powerpc/configs/85xx/xes_mpc85xx_defconfig b/arch/powerpc/configs/85xx/xes_mpc85xx_defconfig
index 3a6381aa9fdc..488d03ae6d6c 100644
--- a/arch/powerpc/configs/85xx/xes_mpc85xx_defconfig
+++ b/arch/powerpc/configs/85xx/xes_mpc85xx_defconfig
@@ -132,7 +132,6 @@ CONFIG_ROOT_NFS=y
 CONFIG_NFSD=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_T10DIF=y
 CONFIG_DETECT_HUNG_TASK=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
 CONFIG_CRYPTO_HMAC=y
diff --git a/arch/powerpc/configs/86xx-hw.config b/arch/powerpc/configs/86xx-hw.config
index 0cb24b33c88e..e7bd265fae5a 100644
--- a/arch/powerpc/configs/86xx-hw.config
+++ b/arch/powerpc/configs/86xx-hw.config
@@ -5,7 +5,6 @@ CONFIG_BROADCOM_PHY=y
 # CONFIG_CARDBUS is not set
 CONFIG_CHR_DEV_SG=y
 CONFIG_CHR_DEV_ST=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_HMAC=y
 CONFIG_DS1682=y
 CONFIG_EEPROM_LEGACY=y
diff --git a/arch/powerpc/configs/amigaone_defconfig b/arch/powerpc/configs/amigaone_defconfig
index 200bb1ecb560..69ef3dc31c4b 100644
--- a/arch/powerpc/configs/amigaone_defconfig
+++ b/arch/powerpc/configs/amigaone_defconfig
@@ -106,7 +106,6 @@ CONFIG_TMPFS=y
 CONFIG_AFFS_FS=m
 CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=m
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_MUTEXES=y
diff --git a/arch/powerpc/configs/chrp32_defconfig b/arch/powerpc/configs/chrp32_defconfig
index fb314f75ad4b..b799c95480ae 100644
--- a/arch/powerpc/configs/chrp32_defconfig
+++ b/arch/powerpc/configs/chrp32_defconfig
@@ -110,7 +110,6 @@ CONFIG_PROC_KCORE=y
 CONFIG_TMPFS=y
 CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=m
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_MUTEXES=y
diff --git a/arch/powerpc/configs/fsl-emb-nonhw.config b/arch/powerpc/configs/fsl-emb-nonhw.config
index d6d2a458847b..2f81bc2d819e 100644
--- a/arch/powerpc/configs/fsl-emb-nonhw.config
+++ b/arch/powerpc/configs/fsl-emb-nonhw.config
@@ -15,7 +15,6 @@ CONFIG_CGROUP_CPUACCT=y
 CONFIG_CGROUP_SCHED=y
 CONFIG_CGROUPS=y
 # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
-CONFIG_CRC_T10DIF=y
 CONFIG_CPUSETS=y
 CONFIG_CRAMFS=y
 CONFIG_CRYPTO_MD4=y
diff --git a/arch/powerpc/configs/g5_defconfig b/arch/powerpc/configs/g5_defconfig
index 9215bed53291..7e58f3e6c987 100644
--- a/arch/powerpc/configs/g5_defconfig
+++ b/arch/powerpc/configs/g5_defconfig
@@ -231,7 +231,6 @@ CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_ISO8859_15=y
 CONFIG_NLS_UTF8=y
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_MUTEXES=y
diff --git a/arch/powerpc/configs/gamecube_defconfig b/arch/powerpc/configs/gamecube_defconfig
index d77eeb525366..cdd99657b71b 100644
--- a/arch/powerpc/configs/gamecube_defconfig
+++ b/arch/powerpc/configs/gamecube_defconfig
@@ -82,7 +82,6 @@ CONFIG_ROOT_NFS=y
 CONFIG_CIFS=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_SPINLOCK=y
 CONFIG_DEBUG_MUTEXES=y
diff --git a/arch/powerpc/configs/linkstation_defconfig b/arch/powerpc/configs/linkstation_defconfig
index fa707de761be..b564f9e33a0d 100644
--- a/arch/powerpc/configs/linkstation_defconfig
+++ b/arch/powerpc/configs/linkstation_defconfig
@@ -125,8 +125,6 @@ CONFIG_NLS_CODEPAGE_437=m
 CONFIG_NLS_CODEPAGE_932=m
 CONFIG_NLS_ISO8859_1=m
 CONFIG_NLS_UTF8=m
-CONFIG_CRC_CCITT=m
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DETECT_HUNG_TASK=y
diff --git a/arch/powerpc/configs/mpc83xx_defconfig b/arch/powerpc/configs/mpc83xx_defconfig
index 83c4710017e9..a815d9e5e3e8 100644
--- a/arch/powerpc/configs/mpc83xx_defconfig
+++ b/arch/powerpc/configs/mpc83xx_defconfig
@@ -97,7 +97,6 @@ CONFIG_TMPFS=y
 CONFIG_NFS_FS=y
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_SHA512=y
diff --git a/arch/powerpc/configs/mpc866_ads_defconfig b/arch/powerpc/configs/mpc866_ads_defconfig
index a0d27c59ea78..dfbdd5e8e108 100644
--- a/arch/powerpc/configs/mpc866_ads_defconfig
+++ b/arch/powerpc/configs/mpc866_ads_defconfig
@@ -38,4 +38,3 @@ CONFIG_TMPFS=y
 CONFIG_CRAMFS=y
 CONFIG_NFS_FS=y
 CONFIG_ROOT_NFS=y
-CONFIG_CRC_CCITT=y
diff --git a/arch/powerpc/configs/mvme5100_defconfig b/arch/powerpc/configs/mvme5100_defconfig
index d1c7fd5bf34b..fa2b3b9c5945 100644
--- a/arch/powerpc/configs/mvme5100_defconfig
+++ b/arch/powerpc/configs/mvme5100_defconfig
@@ -107,8 +107,6 @@ CONFIG_NLS_CODEPAGE_437=m
 CONFIG_NLS_CODEPAGE_932=m
 CONFIG_NLS_ISO8859_1=m
 CONFIG_NLS_UTF8=m
-CONFIG_CRC_CCITT=m
-CONFIG_CRC_T10DIF=y
 CONFIG_XZ_DEC=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
diff --git a/arch/powerpc/configs/pasemi_defconfig b/arch/powerpc/configs/pasemi_defconfig
index 61993944db40..8bbf51b38480 100644
--- a/arch/powerpc/configs/pasemi_defconfig
+++ b/arch/powerpc/configs/pasemi_defconfig
@@ -159,7 +159,6 @@ CONFIG_NFSD=y
 CONFIG_NFSD_V4=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
diff --git a/arch/powerpc/configs/pmac32_defconfig b/arch/powerpc/configs/pmac32_defconfig
index e8b3f67bf3f5..1bc3466bc909 100644
--- a/arch/powerpc/configs/pmac32_defconfig
+++ b/arch/powerpc/configs/pmac32_defconfig
@@ -276,7 +276,6 @@ CONFIG_NFSD_V3_ACL=y
 CONFIG_NFSD_V4=y
 CONFIG_NLS_CODEPAGE_437=m
 CONFIG_NLS_ISO8859_1=m
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DETECT_HUNG_TASK=y
diff --git a/arch/powerpc/configs/ppc44x_defconfig b/arch/powerpc/configs/ppc44x_defconfig
index 8b595f67068c..41c930f74ed4 100644
--- a/arch/powerpc/configs/ppc44x_defconfig
+++ b/arch/powerpc/configs/ppc44x_defconfig
@@ -90,7 +90,6 @@ CONFIG_NFS_FS=y
 CONFIG_ROOT_NFS=y
 CONFIG_NLS_CODEPAGE_437=m
 CONFIG_NLS_ISO8859_1=m
-CONFIG_CRC_T10DIF=m
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DETECT_HUNG_TASK=y
 CONFIG_CRYPTO_ECB=y
diff --git a/arch/powerpc/configs/ppc64e_defconfig b/arch/powerpc/configs/ppc64e_defconfig
index 4c05f4e4d505..d2e659a2d8cb 100644
--- a/arch/powerpc/configs/ppc64e_defconfig
+++ b/arch/powerpc/configs/ppc64e_defconfig
@@ -207,7 +207,6 @@ CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_UTF8=y
-CONFIG_CRC_T10DIF=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_STACK_USAGE=y
diff --git a/arch/powerpc/configs/ps3_defconfig b/arch/powerpc/configs/ps3_defconfig
index 2b175ddf82f0..0b48d2b776c4 100644
--- a/arch/powerpc/configs/ps3_defconfig
+++ b/arch/powerpc/configs/ps3_defconfig
@@ -148,8 +148,6 @@ CONFIG_NLS_ISO8859_1=y
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_MICHAEL_MIC=m
 CONFIG_CRYPTO_LZO=m
-CONFIG_CRC_CCITT=m
-CONFIG_CRC_T10DIF=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_MAGIC_SYSRQ=y
diff --git a/arch/powerpc/configs/skiroot_defconfig b/arch/powerpc/configs/skiroot_defconfig
index 3086c4a12d6d..2b71a6dc399e 100644
--- a/arch/powerpc/configs/skiroot_defconfig
+++ b/arch/powerpc/configs/skiroot_defconfig
@@ -278,8 +278,6 @@ CONFIG_LOCK_DOWN_KERNEL_FORCE_INTEGRITY=y
 # CONFIG_INTEGRITY is not set
 CONFIG_LSM="yama,loadpin,safesetid,integrity"
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC16=y
-CONFIG_CRC_ITU_T=y
 # CONFIG_XZ_DEC_X86 is not set
 # CONFIG_XZ_DEC_IA64 is not set
 # CONFIG_XZ_DEC_ARM is not set
diff --git a/arch/powerpc/configs/storcenter_defconfig b/arch/powerpc/configs/storcenter_defconfig
index 7a978d396991..e415222bd839 100644
--- a/arch/powerpc/configs/storcenter_defconfig
+++ b/arch/powerpc/configs/storcenter_defconfig
@@ -75,4 +75,3 @@ CONFIG_NLS_DEFAULT="utf8"
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
 CONFIG_NLS_UTF8=y
-CONFIG_CRC_T10DIF=y
diff --git a/arch/powerpc/configs/wii_defconfig b/arch/powerpc/configs/wii_defconfig
index 5017a697b67b..7c714a19221e 100644
--- a/arch/powerpc/configs/wii_defconfig
+++ b/arch/powerpc/configs/wii_defconfig
@@ -114,7 +114,6 @@ CONFIG_ROOT_NFS=y
 CONFIG_CIFS=m
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
-CONFIG_CRC_CCITT=y
 CONFIG_PRINTK_TIME=y
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_SPINLOCK=y
diff --git a/arch/powerpc/kernel/module_64.c b/arch/powerpc/kernel/module_64.c
index 34a5aec4908f..126bf3b06ab7 100644
--- a/arch/powerpc/kernel/module_64.c
+++ b/arch/powerpc/kernel/module_64.c
@@ -258,10 +258,6 @@ static unsigned long get_stubs_size(const Elf64_Ehdr *hdr,
 			break;
 		}
 	}
-	if (i == hdr->e_shnum) {
-		pr_err("%s: doesn't contain __patchable_function_entries.\n", me->name);
-		return -ENOEXEC;
-	}
 #endif
 
 	pr_debug("Looks like a total of %lu stubs, max\n", relocs);
diff --git a/arch/powerpc/mm/book3s64/radix_pgtable.c b/arch/powerpc/mm/book3s64/radix_pgtable.c
index 311e2112d782..9f764bc42b8c 100644
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c
+++ b/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -976,7 +976,7 @@ int __meminit radix__vmemmap_create_mapping(unsigned long start,
 	return 0;
 }
 
-
+#ifdef CONFIG_ARCH_WANT_OPTIMIZE_DAX_VMEMMAP
 bool vmemmap_can_optimize(struct vmem_altmap *altmap, struct dev_pagemap *pgmap)
 {
 	if (radix_enabled())
@@ -984,6 +984,7 @@ bool vmemmap_can_optimize(struct vmem_altmap *altmap, struct dev_pagemap *pgmap)
 
 	return false;
 }
+#endif
 
 int __meminit vmemmap_check_pmd(pmd_t *pmdp, int node,
 				unsigned long addr, unsigned long next)
@@ -1120,6 +1121,19 @@ int __meminit radix__vmemmap_populate(unsigned long start, unsigned long end, in
 	pmd_t *pmd;
 	pte_t *pte;
 
+	/*
+	 * Make sure we align the start vmemmap addr so that we calculate
+	 * the correct start_pfn in altmap boundary check to decided whether
+	 * we should use altmap or RAM based backing memory allocation. Also
+	 * the address need to be aligned for set_pte operation.
+
+	 * If the start addr is already PMD_SIZE aligned we will try to use
+	 * a pmd mapping. We don't want to be too aggressive here beacause
+	 * that will cause more allocations in RAM. So only if the namespace
+	 * vmemmap start addr is PMD_SIZE aligned we will use PMD mapping.
+	 */
+
+	start = ALIGN_DOWN(start, PAGE_SIZE);
 	for (addr = start; addr < end; addr = next) {
 		next = pmd_addr_end(addr, end);
 
@@ -1145,8 +1159,8 @@ int __meminit radix__vmemmap_populate(unsigned long start, unsigned long end, in
 			 * in altmap block allocation failures, in which case
 			 * we fallback to RAM for vmemmap allocation.
 			 */
-			if (altmap && (!IS_ALIGNED(addr, PMD_SIZE) ||
-				       altmap_cross_boundary(altmap, addr, PMD_SIZE))) {
+			if (!IS_ALIGNED(addr, PMD_SIZE) || (altmap &&
+			    altmap_cross_boundary(altmap, addr, PMD_SIZE))) {
 				/*
 				 * make sure we don't create altmap mappings
 				 * covering things outside the device.
diff --git a/arch/powerpc/platforms/powernv/Kconfig b/arch/powerpc/platforms/powernv/Kconfig
index 3fbe0295ce14..95d7ba73d43d 100644
--- a/arch/powerpc/platforms/powernv/Kconfig
+++ b/arch/powerpc/platforms/powernv/Kconfig
@@ -17,7 +17,7 @@ config PPC_POWERNV
 	select MMU_NOTIFIER
 	select FORCE_SMP
 	select ARCH_SUPPORTS_PER_VMA_LOCK
-	select PPC_RADIX_BROADCAST_TLBIE
+	select PPC_RADIX_BROADCAST_TLBIE if PPC_RADIX_MMU
 	default y
 
 config OPAL_PRD
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index a934c2a262f6..fa3c2fff082a 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -23,7 +23,7 @@ config PPC_PSERIES
 	select FORCE_SMP
 	select SWIOTLB
 	select ARCH_SUPPORTS_PER_VMA_LOCK
-	select PPC_RADIX_BROADCAST_TLBIE
+	select PPC_RADIX_BROADCAST_TLBIE if PPC_RADIX_MMU
 	default y
 
 config PARAVIRT
diff --git a/arch/riscv/include/asm/alternative-macros.h b/arch/riscv/include/asm/alternative-macros.h
index 721ec275ce57..231d777d936c 100644
--- a/arch/riscv/include/asm/alternative-macros.h
+++ b/arch/riscv/include/asm/alternative-macros.h
@@ -115,24 +115,19 @@
 	\old_c
 .endm
 
-#define _ALTERNATIVE_CFG(old_c, ...)	\
-	ALTERNATIVE_CFG old_c
-
-#define _ALTERNATIVE_CFG_2(old_c, ...)	\
-	ALTERNATIVE_CFG old_c
+#define __ALTERNATIVE_CFG(old_c, ...)		ALTERNATIVE_CFG old_c
+#define __ALTERNATIVE_CFG_2(old_c, ...)		ALTERNATIVE_CFG old_c
 
 #else /* !__ASSEMBLY__ */
 
-#define __ALTERNATIVE_CFG(old_c)	\
-	old_c "\n"
+#define __ALTERNATIVE_CFG(old_c, ...)		old_c "\n"
+#define __ALTERNATIVE_CFG_2(old_c, ...)		old_c "\n"
 
-#define _ALTERNATIVE_CFG(old_c, ...)	\
-	__ALTERNATIVE_CFG(old_c)
+#endif /* __ASSEMBLY__ */
 
-#define _ALTERNATIVE_CFG_2(old_c, ...)	\
-	__ALTERNATIVE_CFG(old_c)
+#define _ALTERNATIVE_CFG(old_c, ...)		__ALTERNATIVE_CFG(old_c)
+#define _ALTERNATIVE_CFG_2(old_c, ...)		__ALTERNATIVE_CFG_2(old_c)
 
-#endif /* __ASSEMBLY__ */
 #endif /* CONFIG_RISCV_ALTERNATIVE */
 
 /*
diff --git a/arch/riscv/include/asm/cacheflush.h b/arch/riscv/include/asm/cacheflush.h
index 8de73f91bfa3..b59ffeb668d6 100644
--- a/arch/riscv/include/asm/cacheflush.h
+++ b/arch/riscv/include/asm/cacheflush.h
@@ -34,11 +34,6 @@ static inline void flush_dcache_page(struct page *page)
 	flush_dcache_folio(page_folio(page));
 }
 
-/*
- * RISC-V doesn't have an instruction to flush parts of the instruction cache,
- * so instead we just flush the whole thing.
- */
-#define flush_icache_range(start, end) flush_icache_all()
 #define flush_icache_user_page(vma, pg, addr, len)	\
 do {							\
 	if (vma->vm_flags & VM_EXEC)			\
@@ -78,6 +73,16 @@ void flush_icache_mm(struct mm_struct *mm, bool local);
 
 #endif /* CONFIG_SMP */
 
+/*
+ * RISC-V doesn't have an instruction to flush parts of the instruction cache,
+ * so instead we just flush the whole thing.
+ */
+#define flush_icache_range flush_icache_range
+static inline void flush_icache_range(unsigned long start, unsigned long end)
+{
+	flush_icache_all();
+}
+
 extern unsigned int riscv_cbom_block_size;
 extern unsigned int riscv_cboz_block_size;
 void riscv_init_cbo_blocksizes(void);
diff --git a/arch/riscv/include/asm/kgdb.h b/arch/riscv/include/asm/kgdb.h
index 46677daf708b..cc11c4544cff 100644
--- a/arch/riscv/include/asm/kgdb.h
+++ b/arch/riscv/include/asm/kgdb.h
@@ -19,16 +19,9 @@
 
 #ifndef	__ASSEMBLY__
 
+void arch_kgdb_breakpoint(void);
 extern unsigned long kgdb_compiled_break;
 
-static inline void arch_kgdb_breakpoint(void)
-{
-	asm(".global kgdb_compiled_break\n"
-	    ".option norvc\n"
-	    "kgdb_compiled_break: ebreak\n"
-	    ".option rvc\n");
-}
-
 #endif /* !__ASSEMBLY__ */
 
 #define DBG_REG_ZERO "zero"
diff --git a/arch/riscv/include/asm/syscall.h b/arch/riscv/include/asm/syscall.h
index 121fff429dce..eceabf59ae48 100644
--- a/arch/riscv/include/asm/syscall.h
+++ b/arch/riscv/include/asm/syscall.h
@@ -62,8 +62,11 @@ static inline void syscall_get_arguments(struct task_struct *task,
 					 unsigned long *args)
 {
 	args[0] = regs->orig_a0;
-	args++;
-	memcpy(args, &regs->a1, 5 * sizeof(args[0]));
+	args[1] = regs->a1;
+	args[2] = regs->a2;
+	args[3] = regs->a3;
+	args[4] = regs->a4;
+	args[5] = regs->a5;
 }
 
 static inline int syscall_get_arch(struct task_struct *task)
diff --git a/arch/riscv/kernel/Makefile b/arch/riscv/kernel/Makefile
index 8d186bfced45..f7480c9c6f8d 100644
--- a/arch/riscv/kernel/Makefile
+++ b/arch/riscv/kernel/Makefile
@@ -9,8 +9,8 @@ CFLAGS_REMOVE_patch.o	= $(CC_FLAGS_FTRACE)
 CFLAGS_REMOVE_sbi.o	= $(CC_FLAGS_FTRACE)
 CFLAGS_REMOVE_return_address.o	= $(CC_FLAGS_FTRACE)
 endif
-CFLAGS_syscall_table.o	+= $(call cc-option,-Wno-override-init,)
-CFLAGS_compat_syscall_table.o += $(call cc-option,-Wno-override-init,)
+CFLAGS_syscall_table.o	+= $(call cc-disable-warning, override-init)
+CFLAGS_compat_syscall_table.o += $(call cc-disable-warning, override-init)
 
 ifdef CONFIG_KEXEC_CORE
 AFLAGS_kexec_relocate.o := -mcmodel=medany $(call cc-option,-mno-relax)
diff --git a/arch/riscv/kernel/kgdb.c b/arch/riscv/kernel/kgdb.c
index 2e0266ae6bd7..9f3db3503dab 100644
--- a/arch/riscv/kernel/kgdb.c
+++ b/arch/riscv/kernel/kgdb.c
@@ -254,6 +254,12 @@ void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
 	regs->epc = pc;
 }
 
+noinline void arch_kgdb_breakpoint(void)
+{
+	asm(".global kgdb_compiled_break\n"
+	    "kgdb_compiled_break: ebreak\n");
+}
+
 void kgdb_arch_handle_qxfer_pkt(char *remcom_in_buffer,
 				char *remcom_out_buffer)
 {
diff --git a/arch/riscv/kernel/module-sections.c b/arch/riscv/kernel/module-sections.c
index e264e59e596e..91d0b355ceef 100644
--- a/arch/riscv/kernel/module-sections.c
+++ b/arch/riscv/kernel/module-sections.c
@@ -73,16 +73,17 @@ static bool duplicate_rela(const Elf_Rela *rela, int idx)
 static void count_max_entries(Elf_Rela *relas, int num,
 			      unsigned int *plts, unsigned int *gots)
 {
-	unsigned int type, i;
-
-	for (i = 0; i < num; i++) {
-		type = ELF_RISCV_R_TYPE(relas[i].r_info);
-		if (type == R_RISCV_CALL_PLT) {
+	for (int i = 0; i < num; i++) {
+		switch (ELF_R_TYPE(relas[i].r_info)) {
+		case R_RISCV_CALL_PLT:
+		case R_RISCV_PLT32:
 			if (!duplicate_rela(relas, i))
 				(*plts)++;
-		} else if (type == R_RISCV_GOT_HI20) {
+			break;
+		case R_RISCV_GOT_HI20:
 			if (!duplicate_rela(relas, i))
 				(*gots)++;
+			break;
 		}
 	}
 }
diff --git a/arch/riscv/kernel/module.c b/arch/riscv/kernel/module.c
index 47d0ebeec93c..7f6147c18033 100644
--- a/arch/riscv/kernel/module.c
+++ b/arch/riscv/kernel/module.c
@@ -648,7 +648,7 @@ process_accumulated_relocations(struct module *me,
 		kfree(bucket_iter);
 	}
 
-	kfree(*relocation_hashtable);
+	kvfree(*relocation_hashtable);
 }
 
 static int add_relocation_to_accumulate(struct module *me, int type,
@@ -752,9 +752,10 @@ initialize_relocation_hashtable(unsigned int num_relocations,
 
 	hashtable_size <<= should_double_size;
 
-	*relocation_hashtable = kmalloc_array(hashtable_size,
-					      sizeof(**relocation_hashtable),
-					      GFP_KERNEL);
+	/* Number of relocations may be large, so kvmalloc it */
+	*relocation_hashtable = kvmalloc_array(hashtable_size,
+					       sizeof(**relocation_hashtable),
+					       GFP_KERNEL);
 	if (!*relocation_hashtable)
 		return 0;
 
@@ -859,7 +860,7 @@ int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab,
 				}
 
 				j++;
-				if (j > sechdrs[relsec].sh_size / sizeof(*rel))
+				if (j == num_relocations)
 					j = 0;
 
 			} while (j_idx != j);
diff --git a/arch/riscv/kernel/probes/uprobes.c b/arch/riscv/kernel/probes/uprobes.c
index 4b3dc8beaf77..cc15f7ca6cc1 100644
--- a/arch/riscv/kernel/probes/uprobes.c
+++ b/arch/riscv/kernel/probes/uprobes.c
@@ -167,6 +167,7 @@ void arch_uprobe_copy_ixol(struct page *page, unsigned long vaddr,
 	/* Initialize the slot */
 	void *kaddr = kmap_atomic(page);
 	void *dst = kaddr + (vaddr & ~PAGE_MASK);
+	unsigned long start = (unsigned long)dst;
 
 	memcpy(dst, src, len);
 
@@ -176,13 +177,6 @@ void arch_uprobe_copy_ixol(struct page *page, unsigned long vaddr,
 		*(uprobe_opcode_t *)dst = __BUG_INSN_32;
 	}
 
+	flush_icache_range(start, start + len);
 	kunmap_atomic(kaddr);
-
-	/*
-	 * We probably need flush_icache_user_page() but it needs vma.
-	 * This should work on most of architectures by default. If
-	 * architecture needs to do something different it can define
-	 * its own version of the function.
-	 */
-	flush_dcache_page(page);
 }
diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
index c174544eefc8..f7c9a1caa83e 100644
--- a/arch/riscv/kernel/setup.c
+++ b/arch/riscv/kernel/setup.c
@@ -66,6 +66,9 @@ static struct resource bss_res = { .name = "Kernel bss", };
 static struct resource elfcorehdr_res = { .name = "ELF Core hdr", };
 #endif
 
+static int num_standard_resources;
+static struct resource *standard_resources;
+
 static int __init add_resource(struct resource *parent,
 				struct resource *res)
 {
@@ -139,7 +142,7 @@ static void __init init_resources(void)
 	struct resource *res = NULL;
 	struct resource *mem_res = NULL;
 	size_t mem_res_sz = 0;
-	int num_resources = 0, res_idx = 0;
+	int num_resources = 0, res_idx = 0, non_resv_res = 0;
 	int ret = 0;
 
 	/* + 1 as memblock_alloc() might increase memblock.reserved.cnt */
@@ -193,6 +196,7 @@ static void __init init_resources(void)
 	/* Add /memory regions to the resource tree */
 	for_each_mem_region(region) {
 		res = &mem_res[res_idx--];
+		non_resv_res++;
 
 		if (unlikely(memblock_is_nomap(region))) {
 			res->name = "Reserved";
@@ -210,6 +214,9 @@ static void __init init_resources(void)
 			goto error;
 	}
 
+	num_standard_resources = non_resv_res;
+	standard_resources = &mem_res[res_idx + 1];
+
 	/* Clean-up any unused pre-allocated resources */
 	if (res_idx >= 0)
 		memblock_free(mem_res, (res_idx + 1) * sizeof(*mem_res));
@@ -221,6 +228,33 @@ static void __init init_resources(void)
 	memblock_free(mem_res, mem_res_sz);
 }
 
+static int __init reserve_memblock_reserved_regions(void)
+{
+	u64 i, j;
+
+	for (i = 0; i < num_standard_resources; i++) {
+		struct resource *mem = &standard_resources[i];
+		phys_addr_t r_start, r_end, mem_size = resource_size(mem);
+
+		if (!memblock_is_region_reserved(mem->start, mem_size))
+			continue;
+
+		for_each_reserved_mem_range(j, &r_start, &r_end) {
+			resource_size_t start, end;
+
+			start = max(PFN_PHYS(PFN_DOWN(r_start)), mem->start);
+			end = min(PFN_PHYS(PFN_UP(r_end)) - 1, mem->end);
+
+			if (start > mem->end || end < mem->start)
+				continue;
+
+			reserve_region_with_split(mem, start, end, "Reserved");
+		}
+	}
+
+	return 0;
+}
+arch_initcall(reserve_memblock_reserved_regions);
 
 static void __init parse_dtb(void)
 {
diff --git a/arch/riscv/kernel/unaligned_access_speed.c b/arch/riscv/kernel/unaligned_access_speed.c
index 585d2dcf2dab..b8ba13819d05 100644
--- a/arch/riscv/kernel/unaligned_access_speed.c
+++ b/arch/riscv/kernel/unaligned_access_speed.c
@@ -439,29 +439,36 @@ static int __init check_unaligned_access_all_cpus(void)
 {
 	int cpu;
 
-	if (unaligned_scalar_speed_param == RISCV_HWPROBE_MISALIGNED_SCALAR_UNKNOWN &&
-	    !check_unaligned_access_emulated_all_cpus()) {
-		check_unaligned_access_speed_all_cpus();
-	} else {
-		pr_info("scalar unaligned access speed set to '%s' by command line\n",
-			speed_str[unaligned_scalar_speed_param]);
+	if (unaligned_scalar_speed_param != RISCV_HWPROBE_MISALIGNED_SCALAR_UNKNOWN) {
+		pr_info("scalar unaligned access speed set to '%s' (%lu) by command line\n",
+			speed_str[unaligned_scalar_speed_param], unaligned_scalar_speed_param);
 		for_each_online_cpu(cpu)
 			per_cpu(misaligned_access_speed, cpu) = unaligned_scalar_speed_param;
+	} else if (!check_unaligned_access_emulated_all_cpus()) {
+		check_unaligned_access_speed_all_cpus();
+	}
+
+	if (unaligned_vector_speed_param != RISCV_HWPROBE_MISALIGNED_VECTOR_UNKNOWN) {
+		if (!has_vector() &&
+		    unaligned_vector_speed_param != RISCV_HWPROBE_MISALIGNED_VECTOR_UNSUPPORTED) {
+			pr_warn("vector support is not available, ignoring unaligned_vector_speed=%s\n",
+				speed_str[unaligned_vector_speed_param]);
+		} else {
+			pr_info("vector unaligned access speed set to '%s' (%lu) by command line\n",
+				speed_str[unaligned_vector_speed_param], unaligned_vector_speed_param);
+		}
 	}
 
 	if (!has_vector())
 		unaligned_vector_speed_param = RISCV_HWPROBE_MISALIGNED_VECTOR_UNSUPPORTED;
 
-	if (unaligned_vector_speed_param == RISCV_HWPROBE_MISALIGNED_VECTOR_UNKNOWN &&
-	    !check_vector_unaligned_access_emulated_all_cpus() &&
-	    IS_ENABLED(CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS)) {
-		kthread_run(vec_check_unaligned_access_speed_all_cpus,
-			    NULL, "vec_check_unaligned_access_speed_all_cpus");
-	} else {
-		pr_info("vector unaligned access speed set to '%s' by command line\n",
-			speed_str[unaligned_vector_speed_param]);
+	if (unaligned_vector_speed_param != RISCV_HWPROBE_MISALIGNED_VECTOR_UNKNOWN) {
 		for_each_online_cpu(cpu)
 			per_cpu(vector_misaligned_access, cpu) = unaligned_vector_speed_param;
+	} else if (!check_vector_unaligned_access_emulated_all_cpus() &&
+		   IS_ENABLED(CONFIG_RISCV_PROBE_VECTOR_UNALIGNED_ACCESS)) {
+		kthread_run(vec_check_unaligned_access_speed_all_cpus,
+			    NULL, "vec_check_unaligned_access_speed_all_cpus");
 	}
 
 	/*
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index db8161ebb43c..99fb986fca6e 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -332,6 +332,10 @@ config HAVE_MARCH_Z16_FEATURES
 	def_bool n
 	select HAVE_MARCH_Z15_FEATURES
 
+config HAVE_MARCH_Z17_FEATURES
+	def_bool n
+	select HAVE_MARCH_Z16_FEATURES
+
 choice
 	prompt "Processor type"
 	default MARCH_Z196
@@ -397,6 +401,14 @@ config MARCH_Z16
 	  Select this to enable optimizations for IBM z16 (3931 and
 	  3932 series).
 
+config MARCH_Z17
+	bool "IBM z17"
+	select HAVE_MARCH_Z17_FEATURES
+	depends on $(cc-option,-march=z17)
+	help
+	  Select this to enable optimizations for IBM z17 (9175 and
+	  9176 series).
+
 endchoice
 
 config MARCH_Z10_TUNE
@@ -420,6 +432,9 @@ config MARCH_Z15_TUNE
 config MARCH_Z16_TUNE
 	def_bool TUNE_Z16 || MARCH_Z16 && TUNE_DEFAULT
 
+config MARCH_Z17_TUNE
+	def_bool TUNE_Z17 || MARCH_Z17 && TUNE_DEFAULT
+
 choice
 	prompt "Tune code generation"
 	default TUNE_DEFAULT
@@ -464,6 +479,10 @@ config TUNE_Z16
 	bool "IBM z16"
 	depends on $(cc-option,-mtune=z16)
 
+config TUNE_Z17
+	bool "IBM z17"
+	depends on $(cc-option,-mtune=z17)
+
 endchoice
 
 config 64BIT
diff --git a/arch/s390/Makefile b/arch/s390/Makefile
index b06dc53bfed5..7679bc16b692 100644
--- a/arch/s390/Makefile
+++ b/arch/s390/Makefile
@@ -48,6 +48,7 @@ mflags-$(CONFIG_MARCH_Z13)    := -march=z13
 mflags-$(CONFIG_MARCH_Z14)    := -march=z14
 mflags-$(CONFIG_MARCH_Z15)    := -march=z15
 mflags-$(CONFIG_MARCH_Z16)    := -march=z16
+mflags-$(CONFIG_MARCH_Z17)    := -march=z17
 
 export CC_FLAGS_MARCH := $(mflags-y)
 
@@ -61,6 +62,7 @@ cflags-$(CONFIG_MARCH_Z13_TUNE)		+= -mtune=z13
 cflags-$(CONFIG_MARCH_Z14_TUNE)		+= -mtune=z14
 cflags-$(CONFIG_MARCH_Z15_TUNE)		+= -mtune=z15
 cflags-$(CONFIG_MARCH_Z16_TUNE)		+= -mtune=z16
+cflags-$(CONFIG_MARCH_Z17_TUNE)		+= -mtune=z17
 
 cflags-y += -Wa,-I$(srctree)/arch/$(ARCH)/include
 
diff --git a/arch/s390/include/asm/march.h b/arch/s390/include/asm/march.h
index fd9eef3be44c..11a71bd14954 100644
--- a/arch/s390/include/asm/march.h
+++ b/arch/s390/include/asm/march.h
@@ -33,6 +33,10 @@
 #define MARCH_HAS_Z16_FEATURES 1
 #endif
 
+#ifdef CONFIG_HAVE_MARCH_Z17_FEATURES
+#define MARCH_HAS_Z17_FEATURES 1
+#endif
+
 #endif /* __DECOMPRESSOR */
 
 #endif /* __ASM_S390_MARCH_H */
diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c
index 33205dd410e4..e657fad7e376 100644
--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -442,7 +442,7 @@ static void cpum_cf_make_setsize(enum cpumf_ctr_set ctrset)
 			ctrset_size = 48;
 		else if (cpumf_ctr_info.csvn >= 3 && cpumf_ctr_info.csvn <= 5)
 			ctrset_size = 128;
-		else if (cpumf_ctr_info.csvn == 6 || cpumf_ctr_info.csvn == 7)
+		else if (cpumf_ctr_info.csvn >= 6 && cpumf_ctr_info.csvn <= 8)
 			ctrset_size = 160;
 		break;
 	case CPUMF_CTR_SET_MT_DIAG:
@@ -858,18 +858,13 @@ static int cpumf_pmu_event_type(struct perf_event *event)
 static int cpumf_pmu_event_init(struct perf_event *event)
 {
 	unsigned int type = event->attr.type;
-	int err;
+	int err = -ENOENT;
 
 	if (type == PERF_TYPE_HARDWARE || type == PERF_TYPE_RAW)
 		err = __hw_perf_event_init(event, type);
 	else if (event->pmu->type == type)
 		/* Registered as unknown PMU */
 		err = __hw_perf_event_init(event, cpumf_pmu_event_type(event));
-	else
-		return -ENOENT;
-
-	if (unlikely(err) && event->destroy)
-		event->destroy(event);
 
 	return err;
 }
@@ -1819,8 +1814,6 @@ static int cfdiag_event_init(struct perf_event *event)
 	event->destroy = hw_perf_event_destroy;
 
 	err = cfdiag_event_init2(event);
-	if (unlikely(err))
-		event->destroy(event);
 out:
 	return err;
 }
diff --git a/arch/s390/kernel/perf_cpum_cf_events.c b/arch/s390/kernel/perf_cpum_cf_events.c
index e4a6bfc91080..690a293eb10d 100644
--- a/arch/s390/kernel/perf_cpum_cf_events.c
+++ b/arch/s390/kernel/perf_cpum_cf_events.c
@@ -237,7 +237,6 @@ CPUMF_EVENT_ATTR(cf_z14, TX_C_TABORT_NO_SPECIAL, 0x00f4);
 CPUMF_EVENT_ATTR(cf_z14, TX_C_TABORT_SPECIAL, 0x00f5);
 CPUMF_EVENT_ATTR(cf_z14, MT_DIAG_CYCLES_ONE_THR_ACTIVE, 0x01c0);
 CPUMF_EVENT_ATTR(cf_z14, MT_DIAG_CYCLES_TWO_THR_ACTIVE, 0x01c1);
-
 CPUMF_EVENT_ATTR(cf_z15, L1D_RO_EXCL_WRITES, 0x0080);
 CPUMF_EVENT_ATTR(cf_z15, DTLB2_WRITES, 0x0081);
 CPUMF_EVENT_ATTR(cf_z15, DTLB2_MISSES, 0x0082);
@@ -365,6 +364,83 @@ CPUMF_EVENT_ATTR(cf_z16, NNPA_WAIT_LOCK, 0x010d);
 CPUMF_EVENT_ATTR(cf_z16, NNPA_HOLD_LOCK, 0x010e);
 CPUMF_EVENT_ATTR(cf_z16, MT_DIAG_CYCLES_ONE_THR_ACTIVE, 0x01c0);
 CPUMF_EVENT_ATTR(cf_z16, MT_DIAG_CYCLES_TWO_THR_ACTIVE, 0x01c1);
+CPUMF_EVENT_ATTR(cf_z17, L1D_RO_EXCL_WRITES, 0x0080);
+CPUMF_EVENT_ATTR(cf_z17, DTLB2_WRITES, 0x0081);
+CPUMF_EVENT_ATTR(cf_z17, DTLB2_MISSES, 0x0082);
+CPUMF_EVENT_ATTR(cf_z17, CRSTE_1MB_WRITES, 0x0083);
+CPUMF_EVENT_ATTR(cf_z17, DTLB2_GPAGE_WRITES, 0x0084);
+CPUMF_EVENT_ATTR(cf_z17, ITLB2_WRITES, 0x0086);
+CPUMF_EVENT_ATTR(cf_z17, ITLB2_MISSES, 0x0087);
+CPUMF_EVENT_ATTR(cf_z17, TLB2_PTE_WRITES, 0x0089);
+CPUMF_EVENT_ATTR(cf_z17, TLB2_CRSTE_WRITES, 0x008a);
+CPUMF_EVENT_ATTR(cf_z17, TLB2_ENGINES_BUSY, 0x008b);
+CPUMF_EVENT_ATTR(cf_z17, TX_C_TEND, 0x008c);
+CPUMF_EVENT_ATTR(cf_z17, TX_NC_TEND, 0x008d);
+CPUMF_EVENT_ATTR(cf_z17, L1C_TLB2_MISSES, 0x008f);
+CPUMF_EVENT_ATTR(cf_z17, DCW_REQ, 0x0091);
+CPUMF_EVENT_ATTR(cf_z17, DCW_REQ_IV, 0x0092);
+CPUMF_EVENT_ATTR(cf_z17, DCW_REQ_CHIP_HIT, 0x0093);
+CPUMF_EVENT_ATTR(cf_z17, DCW_REQ_DRAWER_HIT, 0x0094);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_CHIP, 0x0095);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_CHIP_IV, 0x0096);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_CHIP_CHIP_HIT, 0x0097);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_CHIP_DRAWER_HIT, 0x0098);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_MODULE, 0x0099);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_DRAWER, 0x009a);
+CPUMF_EVENT_ATTR(cf_z17, DCW_OFF_DRAWER, 0x009b);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_CHIP_MEMORY, 0x009c);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_MODULE_MEMORY, 0x009d);
+CPUMF_EVENT_ATTR(cf_z17, DCW_ON_DRAWER_MEMORY, 0x009e);
+CPUMF_EVENT_ATTR(cf_z17, DCW_OFF_DRAWER_MEMORY, 0x009f);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_ON_MODULE_IV, 0x00a0);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_ON_MODULE_CHIP_HIT, 0x00a1);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_ON_MODULE_DRAWER_HIT, 0x00a2);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_ON_DRAWER_IV, 0x00a3);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_ON_DRAWER_CHIP_HIT, 0x00a4);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_ON_DRAWER_DRAWER_HIT, 0x00a5);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_OFF_DRAWER_IV, 0x00a6);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_OFF_DRAWER_CHIP_HIT, 0x00a7);
+CPUMF_EVENT_ATTR(cf_z17, IDCW_OFF_DRAWER_DRAWER_HIT, 0x00a8);
+CPUMF_EVENT_ATTR(cf_z17, ICW_REQ, 0x00a9);
+CPUMF_EVENT_ATTR(cf_z17, ICW_REQ_IV, 0x00aa);
+CPUMF_EVENT_ATTR(cf_z17, ICW_REQ_CHIP_HIT, 0x00ab);
+CPUMF_EVENT_ATTR(cf_z17, ICW_REQ_DRAWER_HIT, 0x00ac);
+CPUMF_EVENT_ATTR(cf_z17, ICW_ON_CHIP, 0x00ad);
+CPUMF_EVENT_ATTR(cf_z17, ICW_ON_CHIP_IV, 0x00ae);
+CPUMF_EVENT_ATTR(cf_z17, ICW_ON_CHIP_CHIP_HIT, 0x00af);
+CPUMF_EVENT_ATTR(cf_z17, ICW_ON_CHIP_DRAWER_HIT, 0x00b0);
+CPUMF_EVENT_ATTR(cf_z17, ICW_ON_MODULE, 0x00b1);
+CPUMF_EVENT_ATTR(cf_z17, ICW_ON_DRAWER, 0x00b2);
+CPUMF_EVENT_ATTR(cf_z17, ICW_OFF_DRAWER, 0x00b3);
+CPUMF_EVENT_ATTR(cf_z17, CYCLES_SAMETHRD, 0x00ca);
+CPUMF_EVENT_ATTR(cf_z17, CYCLES_DIFFTHRD, 0x00cb);
+CPUMF_EVENT_ATTR(cf_z17, INST_SAMETHRD, 0x00cc);
+CPUMF_EVENT_ATTR(cf_z17, INST_DIFFTHRD, 0x00cd);
+CPUMF_EVENT_ATTR(cf_z17, WRONG_BRANCH_PREDICTION, 0x00ce);
+CPUMF_EVENT_ATTR(cf_z17, VX_BCD_EXECUTION_SLOTS, 0x00e1);
+CPUMF_EVENT_ATTR(cf_z17, DECIMAL_INSTRUCTIONS, 0x00e2);
+CPUMF_EVENT_ATTR(cf_z17, LAST_HOST_TRANSLATIONS, 0x00e8);
+CPUMF_EVENT_ATTR(cf_z17, TX_NC_TABORT, 0x00f4);
+CPUMF_EVENT_ATTR(cf_z17, TX_C_TABORT_NO_SPECIAL, 0x00f5);
+CPUMF_EVENT_ATTR(cf_z17, TX_C_TABORT_SPECIAL, 0x00f6);
+CPUMF_EVENT_ATTR(cf_z17, DFLT_ACCESS, 0x00f8);
+CPUMF_EVENT_ATTR(cf_z17, DFLT_CYCLES, 0x00fd);
+CPUMF_EVENT_ATTR(cf_z17, SORTL, 0x0100);
+CPUMF_EVENT_ATTR(cf_z17, DFLT_CC, 0x0109);
+CPUMF_EVENT_ATTR(cf_z17, DFLT_CCFINISH, 0x010a);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_INVOCATIONS, 0x010b);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_COMPLETIONS, 0x010c);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_WAIT_LOCK, 0x010d);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_HOLD_LOCK, 0x010e);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_INST_ONCHIP, 0x0110);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_INST_OFFCHIP, 0x0111);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_INST_DIFF, 0x0112);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_4K_PREFETCH, 0x0114);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_COMPL_LOCK, 0x0115);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_RETRY_LOCK, 0x0116);
+CPUMF_EVENT_ATTR(cf_z17, NNPA_RETRY_LOCK_WITH_PLO, 0x0117);
+CPUMF_EVENT_ATTR(cf_z17, MT_DIAG_CYCLES_ONE_THR_ACTIVE, 0x01c0);
+CPUMF_EVENT_ATTR(cf_z17, MT_DIAG_CYCLES_TWO_THR_ACTIVE, 0x01c1);
 
 static struct attribute *cpumcf_fvn1_pmu_event_attr[] __initdata = {
 	CPUMF_EVENT_PTR(cf_fvn1, CPU_CYCLES),
@@ -414,7 +490,7 @@ static struct attribute *cpumcf_svn_12345_pmu_event_attr[] __initdata = {
 	NULL,
 };
 
-static struct attribute *cpumcf_svn_67_pmu_event_attr[] __initdata = {
+static struct attribute *cpumcf_svn_678_pmu_event_attr[] __initdata = {
 	CPUMF_EVENT_PTR(cf_svn_12345, PRNG_FUNCTIONS),
 	CPUMF_EVENT_PTR(cf_svn_12345, PRNG_CYCLES),
 	CPUMF_EVENT_PTR(cf_svn_12345, PRNG_BLOCKED_FUNCTIONS),
@@ -779,6 +855,87 @@ static struct attribute *cpumcf_z16_pmu_event_attr[] __initdata = {
 	NULL,
 };
 
+static struct attribute *cpumcf_z17_pmu_event_attr[] __initdata = {
+	CPUMF_EVENT_PTR(cf_z17, L1D_RO_EXCL_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, DTLB2_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, DTLB2_MISSES),
+	CPUMF_EVENT_PTR(cf_z17, CRSTE_1MB_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, DTLB2_GPAGE_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, ITLB2_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, ITLB2_MISSES),
+	CPUMF_EVENT_PTR(cf_z17, TLB2_PTE_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, TLB2_CRSTE_WRITES),
+	CPUMF_EVENT_PTR(cf_z17, TLB2_ENGINES_BUSY),
+	CPUMF_EVENT_PTR(cf_z17, TX_C_TEND),
+	CPUMF_EVENT_PTR(cf_z17, TX_NC_TEND),
+	CPUMF_EVENT_PTR(cf_z17, L1C_TLB2_MISSES),
+	CPUMF_EVENT_PTR(cf_z17, DCW_REQ),
+	CPUMF_EVENT_PTR(cf_z17, DCW_REQ_IV),
+	CPUMF_EVENT_PTR(cf_z17, DCW_REQ_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, DCW_REQ_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_CHIP),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_CHIP_IV),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_CHIP_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_CHIP_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_MODULE),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_DRAWER),
+	CPUMF_EVENT_PTR(cf_z17, DCW_OFF_DRAWER),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_CHIP_MEMORY),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_MODULE_MEMORY),
+	CPUMF_EVENT_PTR(cf_z17, DCW_ON_DRAWER_MEMORY),
+	CPUMF_EVENT_PTR(cf_z17, DCW_OFF_DRAWER_MEMORY),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_ON_MODULE_IV),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_ON_MODULE_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_ON_MODULE_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_ON_DRAWER_IV),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_ON_DRAWER_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_ON_DRAWER_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_OFF_DRAWER_IV),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_OFF_DRAWER_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, IDCW_OFF_DRAWER_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, ICW_REQ),
+	CPUMF_EVENT_PTR(cf_z17, ICW_REQ_IV),
+	CPUMF_EVENT_PTR(cf_z17, ICW_REQ_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, ICW_REQ_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, ICW_ON_CHIP),
+	CPUMF_EVENT_PTR(cf_z17, ICW_ON_CHIP_IV),
+	CPUMF_EVENT_PTR(cf_z17, ICW_ON_CHIP_CHIP_HIT),
+	CPUMF_EVENT_PTR(cf_z17, ICW_ON_CHIP_DRAWER_HIT),
+	CPUMF_EVENT_PTR(cf_z17, ICW_ON_MODULE),
+	CPUMF_EVENT_PTR(cf_z17, ICW_ON_DRAWER),
+	CPUMF_EVENT_PTR(cf_z17, ICW_OFF_DRAWER),
+	CPUMF_EVENT_PTR(cf_z17, CYCLES_SAMETHRD),
+	CPUMF_EVENT_PTR(cf_z17, CYCLES_DIFFTHRD),
+	CPUMF_EVENT_PTR(cf_z17, INST_SAMETHRD),
+	CPUMF_EVENT_PTR(cf_z17, INST_DIFFTHRD),
+	CPUMF_EVENT_PTR(cf_z17, WRONG_BRANCH_PREDICTION),
+	CPUMF_EVENT_PTR(cf_z17, VX_BCD_EXECUTION_SLOTS),
+	CPUMF_EVENT_PTR(cf_z17, DECIMAL_INSTRUCTIONS),
+	CPUMF_EVENT_PTR(cf_z17, LAST_HOST_TRANSLATIONS),
+	CPUMF_EVENT_PTR(cf_z17, TX_NC_TABORT),
+	CPUMF_EVENT_PTR(cf_z17, TX_C_TABORT_NO_SPECIAL),
+	CPUMF_EVENT_PTR(cf_z17, TX_C_TABORT_SPECIAL),
+	CPUMF_EVENT_PTR(cf_z17, DFLT_ACCESS),
+	CPUMF_EVENT_PTR(cf_z17, DFLT_CYCLES),
+	CPUMF_EVENT_PTR(cf_z17, SORTL),
+	CPUMF_EVENT_PTR(cf_z17, DFLT_CC),
+	CPUMF_EVENT_PTR(cf_z17, DFLT_CCFINISH),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_INVOCATIONS),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_COMPLETIONS),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_WAIT_LOCK),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_HOLD_LOCK),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_INST_ONCHIP),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_INST_OFFCHIP),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_INST_DIFF),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_4K_PREFETCH),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_COMPL_LOCK),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_RETRY_LOCK),
+	CPUMF_EVENT_PTR(cf_z17, NNPA_RETRY_LOCK_WITH_PLO),
+	CPUMF_EVENT_PTR(cf_z17, MT_DIAG_CYCLES_ONE_THR_ACTIVE),
+	CPUMF_EVENT_PTR(cf_z17, MT_DIAG_CYCLES_TWO_THR_ACTIVE),
+	NULL,
+};
+
 /* END: CPUM_CF COUNTER DEFINITIONS ===================================== */
 
 static struct attribute_group cpumcf_pmu_events_group = {
@@ -859,7 +1016,7 @@ __init const struct attribute_group **cpumf_cf_event_group(void)
 	if (ci.csvn >= 1 && ci.csvn <= 5)
 		csvn = cpumcf_svn_12345_pmu_event_attr;
 	else if (ci.csvn >= 6)
-		csvn = cpumcf_svn_67_pmu_event_attr;
+		csvn = cpumcf_svn_678_pmu_event_attr;
 
 	/* Determine model-specific counter set(s) */
 	get_cpu_id(&cpu_id);
@@ -892,6 +1049,10 @@ __init const struct attribute_group **cpumf_cf_event_group(void)
 	case 0x3932:
 		model = cpumcf_z16_pmu_event_attr;
 		break;
+	case 0x9175:
+	case 0x9176:
+		model = cpumcf_z17_pmu_event_attr;
+		break;
 	default:
 		model = none;
 		break;
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index 5f60248cb468..ad22799d8a7d 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -885,9 +885,6 @@ static int cpumsf_pmu_event_init(struct perf_event *event)
 		event->attr.exclude_idle = 0;
 
 	err = __hw_perf_event_init(event);
-	if (unlikely(err))
-		if (event->destroy)
-			event->destroy(event);
 	return err;
 }
 
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index 54e281436a28..80b1f7a29f11 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -294,6 +294,10 @@ static int __init setup_elf_platform(void)
 	case 0x3932:
 		strcpy(elf_platform, "z16");
 		break;
+	case 0x9175:
+	case 0x9176:
+		strcpy(elf_platform, "z17");
+		break;
 	}
 	return 0;
 }
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index 610dd44a948b..a06a000f196c 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -95,7 +95,7 @@ static int handle_validity(struct kvm_vcpu *vcpu)
 
 	vcpu->stat.exit_validity++;
 	trace_kvm_s390_intercept_validity(vcpu, viwhy);
-	KVM_EVENT(3, "validity intercept 0x%x for pid %u (kvm 0x%pK)", viwhy,
+	KVM_EVENT(3, "validity intercept 0x%x for pid %u (kvm 0x%p)", viwhy,
 		  current->pid, vcpu->kvm);
 
 	/* do not warn on invalid runtime instrumentation mode */
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 2811a6c093b8..60c360c18690 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -3161,7 +3161,7 @@ void kvm_s390_gisa_clear(struct kvm *kvm)
 	if (!gi->origin)
 		return;
 	gisa_clear_ipm(gi->origin);
-	VM_EVENT(kvm, 3, "gisa 0x%pK cleared", gi->origin);
+	VM_EVENT(kvm, 3, "gisa 0x%p cleared", gi->origin);
 }
 
 void kvm_s390_gisa_init(struct kvm *kvm)
@@ -3177,7 +3177,7 @@ void kvm_s390_gisa_init(struct kvm *kvm)
 	hrtimer_setup(&gi->timer, gisa_vcpu_kicker, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	memset(gi->origin, 0, sizeof(struct kvm_s390_gisa));
 	gi->origin->next_alert = (u32)virt_to_phys(gi->origin);
-	VM_EVENT(kvm, 3, "gisa 0x%pK initialized", gi->origin);
+	VM_EVENT(kvm, 3, "gisa 0x%p initialized", gi->origin);
 }
 
 void kvm_s390_gisa_enable(struct kvm *kvm)
@@ -3218,7 +3218,7 @@ void kvm_s390_gisa_destroy(struct kvm *kvm)
 		process_gib_alert_list();
 	hrtimer_cancel(&gi->timer);
 	gi->origin = NULL;
-	VM_EVENT(kvm, 3, "gisa 0x%pK destroyed", gisa);
+	VM_EVENT(kvm, 3, "gisa 0x%p destroyed", gisa);
 }
 
 void kvm_s390_gisa_disable(struct kvm *kvm)
@@ -3467,7 +3467,7 @@ int __init kvm_s390_gib_init(u8 nisc)
 		}
 	}
 
-	KVM_EVENT(3, "gib 0x%pK (nisc=%d) initialized", gib, gib->nisc);
+	KVM_EVENT(3, "gib 0x%p (nisc=%d) initialized", gib, gib->nisc);
 	goto out;
 
 out_unreg_gal:
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index fff863734975..3f3175193fd7 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -1022,7 +1022,7 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 		}
 		mutex_unlock(&kvm->lock);
 		VM_EVENT(kvm, 3, "SET: max guest address: %lu", new_limit);
-		VM_EVENT(kvm, 3, "New guest asce: 0x%pK",
+		VM_EVENT(kvm, 3, "New guest asce: 0x%p",
 			 (void *) kvm->arch.gmap->asce);
 		break;
 	}
@@ -3466,7 +3466,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 		kvm_s390_gisa_init(kvm);
 	INIT_LIST_HEAD(&kvm->arch.pv.need_cleanup);
 	kvm->arch.pv.set_aside = NULL;
-	KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
+	KVM_EVENT(3, "vm 0x%p created by pid %u", kvm, current->pid);
 
 	return 0;
 out_err:
@@ -3529,7 +3529,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kvm_s390_destroy_adapters(kvm);
 	kvm_s390_clear_float_irqs(kvm);
 	kvm_s390_vsie_destroy(kvm);
-	KVM_EVENT(3, "vm 0x%pK destroyed", kvm);
+	KVM_EVENT(3, "vm 0x%p destroyed", kvm);
 }
 
 /* Section: vcpu related */
@@ -3650,7 +3650,7 @@ static int sca_switch_to_extended(struct kvm *kvm)
 
 	free_page((unsigned long)old_sca);
 
-	VM_EVENT(kvm, 2, "Switched to ESCA (0x%pK -> 0x%pK)",
+	VM_EVENT(kvm, 2, "Switched to ESCA (0x%p -> 0x%p)",
 		 old_sca, kvm->arch.sca);
 	return 0;
 }
@@ -4027,7 +4027,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 			goto out_free_sie_block;
 	}
 
-	VM_EVENT(vcpu->kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK",
+	VM_EVENT(vcpu->kvm, 3, "create cpu %d at 0x%p, sie block at 0x%p",
 		 vcpu->vcpu_id, vcpu, vcpu->arch.sie_block);
 	trace_kvm_s390_create_vcpu(vcpu->vcpu_id, vcpu, vcpu->arch.sie_block);
 
diff --git a/arch/s390/kvm/trace-s390.h b/arch/s390/kvm/trace-s390.h
index 9ac92dbf680d..9e28f165c114 100644
--- a/arch/s390/kvm/trace-s390.h
+++ b/arch/s390/kvm/trace-s390.h
@@ -56,7 +56,7 @@ TRACE_EVENT(kvm_s390_create_vcpu,
 		    __entry->sie_block = sie_block;
 		    ),
 
-	    TP_printk("create cpu %d at 0x%pK, sie block at 0x%pK",
+	    TP_printk("create cpu %d at 0x%p, sie block at 0x%p",
 		      __entry->id, __entry->vcpu, __entry->sie_block)
 	);
 
@@ -255,7 +255,7 @@ TRACE_EVENT(kvm_s390_enable_css,
 		    __entry->kvm = kvm;
 		    ),
 
-	    TP_printk("enabling channel I/O support (kvm @ %pK)\n",
+	    TP_printk("enabling channel I/O support (kvm @ %p)\n",
 		      __entry->kvm)
 	);
 
diff --git a/arch/s390/tools/gen_facilities.c b/arch/s390/tools/gen_facilities.c
index 855f818deb98..d5c68ade71ab 100644
--- a/arch/s390/tools/gen_facilities.c
+++ b/arch/s390/tools/gen_facilities.c
@@ -54,6 +54,9 @@ static struct facility_def facility_defs[] = {
 #ifdef CONFIG_HAVE_MARCH_Z15_FEATURES
 			61, /* miscellaneous-instruction-extension 3 */
 #endif
+#ifdef CONFIG_HAVE_MARCH_Z17_FEATURES
+			84, /* miscellaneous-instruction-extension 4 */
+#endif
 			-1 /* END */
 		}
 	},
diff --git a/arch/sh/configs/ap325rxa_defconfig b/arch/sh/configs/ap325rxa_defconfig
index 4464a2ad42ed..b6f36c938f1d 100644
--- a/arch/sh/configs/ap325rxa_defconfig
+++ b/arch/sh/configs/ap325rxa_defconfig
@@ -99,4 +99,3 @@ CONFIG_NLS_ISO8859_1=y
 CONFIG_CRYPTO=y
 CONFIG_CRYPTO_CBC=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/ecovec24_defconfig b/arch/sh/configs/ecovec24_defconfig
index ee1b36682155..e76694aace25 100644
--- a/arch/sh/configs/ecovec24_defconfig
+++ b/arch/sh/configs/ecovec24_defconfig
@@ -128,4 +128,3 @@ CONFIG_DEBUG_FS=y
 CONFIG_CRYPTO=y
 CONFIG_CRYPTO_CBC=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/edosk7705_defconfig b/arch/sh/configs/edosk7705_defconfig
index 296ed768cbbb..ee3f6db7d8da 100644
--- a/arch/sh/configs/edosk7705_defconfig
+++ b/arch/sh/configs/edosk7705_defconfig
@@ -33,4 +33,3 @@ CONFIG_CMDLINE_FROM_BOOTLOADER=y
 # CONFIG_PROC_FS is not set
 # CONFIG_SYSFS is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
-# CONFIG_CRC32 is not set
diff --git a/arch/sh/configs/espt_defconfig b/arch/sh/configs/espt_defconfig
index 67716a44463e..da176f100e00 100644
--- a/arch/sh/configs/espt_defconfig
+++ b/arch/sh/configs/espt_defconfig
@@ -110,4 +110,3 @@ CONFIG_NLS_UTF8=y
 # CONFIG_ENABLE_MUST_CHECK is not set
 CONFIG_DEBUG_FS=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/hp6xx_defconfig b/arch/sh/configs/hp6xx_defconfig
index 77e3185f63e4..3582af15ad86 100644
--- a/arch/sh/configs/hp6xx_defconfig
+++ b/arch/sh/configs/hp6xx_defconfig
@@ -56,5 +56,3 @@ CONFIG_CRYPTO_PCBC=y
 CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC16=y
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/kfr2r09-romimage_defconfig b/arch/sh/configs/kfr2r09-romimage_defconfig
index 42bf34181a3e..88fbb65cb9f9 100644
--- a/arch/sh/configs/kfr2r09-romimage_defconfig
+++ b/arch/sh/configs/kfr2r09-romimage_defconfig
@@ -49,4 +49,3 @@ CONFIG_TMPFS=y
 # CONFIG_NETWORK_FILESYSTEMS is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
 CONFIG_DEBUG_FS=y
-# CONFIG_CRC32 is not set
diff --git a/arch/sh/configs/landisk_defconfig b/arch/sh/configs/landisk_defconfig
index d871623955c5..924bb3233b0b 100644
--- a/arch/sh/configs/landisk_defconfig
+++ b/arch/sh/configs/landisk_defconfig
@@ -111,4 +111,3 @@ CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_CODEPAGE_932=y
 CONFIG_SH_STANDARD_BIOS=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/lboxre2_defconfig b/arch/sh/configs/lboxre2_defconfig
index 6a234761bfd7..0307bb2be79f 100644
--- a/arch/sh/configs/lboxre2_defconfig
+++ b/arch/sh/configs/lboxre2_defconfig
@@ -58,4 +58,3 @@ CONFIG_ROMFS_FS=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_SH_STANDARD_BIOS=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/magicpanelr2_defconfig b/arch/sh/configs/magicpanelr2_defconfig
index 8d443749550e..93b9aa32dc7c 100644
--- a/arch/sh/configs/magicpanelr2_defconfig
+++ b/arch/sh/configs/magicpanelr2_defconfig
@@ -86,5 +86,3 @@ CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_KOBJECT=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_FRAME_POINTER=y
-CONFIG_CRC_CCITT=m
-CONFIG_CRC16=m
diff --git a/arch/sh/configs/migor_defconfig b/arch/sh/configs/migor_defconfig
index 2d1e65cad239..fc2010c241fb 100644
--- a/arch/sh/configs/migor_defconfig
+++ b/arch/sh/configs/migor_defconfig
@@ -90,4 +90,3 @@ CONFIG_DEBUG_FS=y
 CONFIG_CRYPTO_MANAGER=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/r7780mp_defconfig b/arch/sh/configs/r7780mp_defconfig
index 6bd6c0ae85d7..f28b8c4181c2 100644
--- a/arch/sh/configs/r7780mp_defconfig
+++ b/arch/sh/configs/r7780mp_defconfig
@@ -105,4 +105,3 @@ CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_HMAC=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/r7785rp_defconfig b/arch/sh/configs/r7785rp_defconfig
index cde668569cc1..3a4239f20ff1 100644
--- a/arch/sh/configs/r7785rp_defconfig
+++ b/arch/sh/configs/r7785rp_defconfig
@@ -103,4 +103,3 @@ CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
 CONFIG_CRYPTO_HMAC=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/rts7751r2d1_defconfig b/arch/sh/configs/rts7751r2d1_defconfig
index c863a11c7592..69568cc40396 100644
--- a/arch/sh/configs/rts7751r2d1_defconfig
+++ b/arch/sh/configs/rts7751r2d1_defconfig
@@ -87,4 +87,3 @@ CONFIG_MINIX_FS=y
 CONFIG_NLS_CODEPAGE_932=y
 CONFIG_DEBUG_FS=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/rts7751r2dplus_defconfig b/arch/sh/configs/rts7751r2dplus_defconfig
index 7e4f710d46c7..ecb4bdb5bb58 100644
--- a/arch/sh/configs/rts7751r2dplus_defconfig
+++ b/arch/sh/configs/rts7751r2dplus_defconfig
@@ -92,4 +92,3 @@ CONFIG_MINIX_FS=y
 CONFIG_NLS_CODEPAGE_932=y
 CONFIG_DEBUG_FS=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/sdk7780_defconfig b/arch/sh/configs/sdk7780_defconfig
index cd24cf08210e..9870d16d9711 100644
--- a/arch/sh/configs/sdk7780_defconfig
+++ b/arch/sh/configs/sdk7780_defconfig
@@ -136,4 +136,3 @@ CONFIG_SH_STANDARD_BIOS=y
 CONFIG_CRYPTO_MD5=y
 CONFIG_CRYPTO_DES=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/se7206_defconfig b/arch/sh/configs/se7206_defconfig
index 472fdf365cad..64f9308ee586 100644
--- a/arch/sh/configs/se7206_defconfig
+++ b/arch/sh/configs/se7206_defconfig
@@ -101,6 +101,3 @@ CONFIG_CRYPTO_DEFLATE=y
 CONFIG_CRYPTO_LZO=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=y
-CONFIG_CRC16=y
-CONFIG_CRC_ITU_T=y
diff --git a/arch/sh/configs/se7712_defconfig b/arch/sh/configs/se7712_defconfig
index 49a4961889de..8770a72e6a63 100644
--- a/arch/sh/configs/se7712_defconfig
+++ b/arch/sh/configs/se7712_defconfig
@@ -96,4 +96,3 @@ CONFIG_FRAME_POINTER=y
 CONFIG_CRYPTO_ECB=m
 CONFIG_CRYPTO_PCBC=m
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_CCITT=y
diff --git a/arch/sh/configs/se7721_defconfig b/arch/sh/configs/se7721_defconfig
index de293792db84..b15c6406a0e8 100644
--- a/arch/sh/configs/se7721_defconfig
+++ b/arch/sh/configs/se7721_defconfig
@@ -122,4 +122,3 @@ CONFIG_DEBUG_KERNEL=y
 CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
 CONFIG_FRAME_POINTER=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_CCITT=y
diff --git a/arch/sh/configs/se7724_defconfig b/arch/sh/configs/se7724_defconfig
index 96521271758c..9501e69eb886 100644
--- a/arch/sh/configs/se7724_defconfig
+++ b/arch/sh/configs/se7724_defconfig
@@ -128,4 +128,3 @@ CONFIG_NLS_ISO8859_1=y
 CONFIG_CRYPTO=y
 CONFIG_CRYPTO_CBC=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/sh03_defconfig b/arch/sh/configs/sh03_defconfig
index 48f38ec236b6..4d75c92cac10 100644
--- a/arch/sh/configs/sh03_defconfig
+++ b/arch/sh/configs/sh03_defconfig
@@ -120,6 +120,5 @@ CONFIG_CRYPTO_HMAC=y
 CONFIG_CRYPTO_SHA1=y
 CONFIG_CRYPTO_DEFLATE=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_CCITT=y
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_GENERIC=y
diff --git a/arch/sh/configs/sh2007_defconfig b/arch/sh/configs/sh2007_defconfig
index 1b1174a07e36..cc6292b3235a 100644
--- a/arch/sh/configs/sh2007_defconfig
+++ b/arch/sh/configs/sh2007_defconfig
@@ -193,5 +193,3 @@ CONFIG_CRYPTO_DEFLATE=y
 CONFIG_CRYPTO_LZO=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_CCITT=y
-CONFIG_CRC16=y
diff --git a/arch/sh/configs/sh7724_generic_defconfig b/arch/sh/configs/sh7724_generic_defconfig
index 5440bd0ca4ed..e6298f22623a 100644
--- a/arch/sh/configs/sh7724_generic_defconfig
+++ b/arch/sh/configs/sh7724_generic_defconfig
@@ -39,4 +39,3 @@ CONFIG_UIO_PDRV_GENIRQ=y
 # CONFIG_SYSFS is not set
 # CONFIG_MISC_FILESYSTEMS is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
-# CONFIG_CRC32 is not set
diff --git a/arch/sh/configs/sh7763rdp_defconfig b/arch/sh/configs/sh7763rdp_defconfig
index 57923c3296cc..b77b3313157e 100644
--- a/arch/sh/configs/sh7763rdp_defconfig
+++ b/arch/sh/configs/sh7763rdp_defconfig
@@ -112,4 +112,3 @@ CONFIG_NLS_UTF8=y
 # CONFIG_ENABLE_MUST_CHECK is not set
 CONFIG_DEBUG_FS=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC_T10DIF=y
diff --git a/arch/sh/configs/sh7770_generic_defconfig b/arch/sh/configs/sh7770_generic_defconfig
index 4338af8d02d0..2e2b46980b58 100644
--- a/arch/sh/configs/sh7770_generic_defconfig
+++ b/arch/sh/configs/sh7770_generic_defconfig
@@ -41,4 +41,3 @@ CONFIG_UIO_PDRV_GENIRQ=y
 # CONFIG_SYSFS is not set
 # CONFIG_MISC_FILESYSTEMS is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
-# CONFIG_CRC32 is not set
diff --git a/arch/sh/configs/titan_defconfig b/arch/sh/configs/titan_defconfig
index 8e85f205d8f5..f022ada363b5 100644
--- a/arch/sh/configs/titan_defconfig
+++ b/arch/sh/configs/titan_defconfig
@@ -264,4 +264,3 @@ CONFIG_CRYPTO_SERPENT=m
 CONFIG_CRYPTO_TEA=m
 CONFIG_CRYPTO_TWOFISH=m
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC16=m
diff --git a/arch/sparc/configs/sparc64_defconfig b/arch/sparc/configs/sparc64_defconfig
index 01b2bdfbf9a8..f1ba0fefe1f9 100644
--- a/arch/sparc/configs/sparc64_defconfig
+++ b/arch/sparc/configs/sparc64_defconfig
@@ -229,7 +229,6 @@ CONFIG_CRYPTO_SERPENT=m
 CONFIG_CRYPTO_TEA=m
 CONFIG_CRYPTO_TWOFISH=m
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
-CONFIG_CRC16=m
 CONFIG_VCC=m
 CONFIG_PATA_CMD64X=y
 CONFIG_IP_PNP=y
diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile
index 81f55da81967..640fcac3af74 100644
--- a/arch/x86/boot/Makefile
+++ b/arch/x86/boot/Makefile
@@ -59,7 +59,7 @@ KBUILD_CFLAGS	+= $(CONFIG_CC_IMPLICIT_FALLTHROUGH)
 $(obj)/bzImage: asflags-y  := $(SVGA_MODE)
 
 quiet_cmd_image = BUILD   $@
-      cmd_image = cp $< $@; truncate -s %4K $@; cat $(obj)/vmlinux.bin >>$@
+      cmd_image = (dd if=$< bs=4k conv=sync status=none; cat $(filter-out $<,$(real-prereqs))) >$@
 
 $(obj)/bzImage: $(obj)/setup.bin $(obj)/vmlinux.bin FORCE
 	$(call if_changed,image)
diff --git a/arch/x86/boot/compressed/mem.c b/arch/x86/boot/compressed/mem.c
index dbba332e4a12..0e9f84ab4bdc 100644
--- a/arch/x86/boot/compressed/mem.c
+++ b/arch/x86/boot/compressed/mem.c
@@ -38,7 +38,7 @@ void arch_accept_memory(phys_addr_t start, phys_addr_t end)
 	if (early_is_tdx_guest()) {
 		if (!tdx_accept_memory(start, end))
 			panic("TDX: Failed to accept memory\n");
-	} else if (sev_snp_enabled()) {
+	} else if (early_is_sevsnp_guest()) {
 		snp_accept_memory(start, end);
 	} else {
 		error("Cannot accept memory: unknown platform\n");
diff --git a/arch/x86/boot/compressed/sev.c b/arch/x86/boot/compressed/sev.c
index bb55934c1cee..0003e4416efd 100644
--- a/arch/x86/boot/compressed/sev.c
+++ b/arch/x86/boot/compressed/sev.c
@@ -164,10 +164,7 @@ bool sev_snp_enabled(void)
 
 static void __page_state_change(unsigned long paddr, enum psc_op op)
 {
-	u64 val;
-
-	if (!sev_snp_enabled())
-		return;
+	u64 val, msr;
 
 	/*
 	 * If private -> shared then invalidate the page before requesting the
@@ -176,6 +173,9 @@ static void __page_state_change(unsigned long paddr, enum psc_op op)
 	if (op == SNP_PAGE_STATE_SHARED)
 		pvalidate_4k_page(paddr, paddr, false);
 
+	/* Save the current GHCB MSR value */
+	msr = sev_es_rd_ghcb_msr();
+
 	/* Issue VMGEXIT to change the page state in RMP table. */
 	sev_es_wr_ghcb_msr(GHCB_MSR_PSC_REQ_GFN(paddr >> PAGE_SHIFT, op));
 	VMGEXIT();
@@ -185,6 +185,9 @@ static void __page_state_change(unsigned long paddr, enum psc_op op)
 	if ((GHCB_RESP_CODE(val) != GHCB_MSR_PSC_RESP) || GHCB_MSR_PSC_RESP_VAL(val))
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PSC);
 
+	/* Restore the GHCB MSR value */
+	sev_es_wr_ghcb_msr(msr);
+
 	/*
 	 * Now that page state is changed in the RMP table, validate it so that it is
 	 * consistent with the RMP entry.
@@ -195,11 +198,17 @@ static void __page_state_change(unsigned long paddr, enum psc_op op)
 
 void snp_set_page_private(unsigned long paddr)
 {
+	if (!sev_snp_enabled())
+		return;
+
 	__page_state_change(paddr, SNP_PAGE_STATE_PRIVATE);
 }
 
 void snp_set_page_shared(unsigned long paddr)
 {
+	if (!sev_snp_enabled())
+		return;
+
 	__page_state_change(paddr, SNP_PAGE_STATE_SHARED);
 }
 
@@ -223,56 +232,10 @@ static bool early_setup_ghcb(void)
 	return true;
 }
 
-static phys_addr_t __snp_accept_memory(struct snp_psc_desc *desc,
-				       phys_addr_t pa, phys_addr_t pa_end)
-{
-	struct psc_hdr *hdr;
-	struct psc_entry *e;
-	unsigned int i;
-
-	hdr = &desc->hdr;
-	memset(hdr, 0, sizeof(*hdr));
-
-	e = desc->entries;
-
-	i = 0;
-	while (pa < pa_end && i < VMGEXIT_PSC_MAX_ENTRY) {
-		hdr->end_entry = i;
-
-		e->gfn = pa >> PAGE_SHIFT;
-		e->operation = SNP_PAGE_STATE_PRIVATE;
-		if (IS_ALIGNED(pa, PMD_SIZE) && (pa_end - pa) >= PMD_SIZE) {
-			e->pagesize = RMP_PG_SIZE_2M;
-			pa += PMD_SIZE;
-		} else {
-			e->pagesize = RMP_PG_SIZE_4K;
-			pa += PAGE_SIZE;
-		}
-
-		e++;
-		i++;
-	}
-
-	if (vmgexit_psc(boot_ghcb, desc))
-		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PSC);
-
-	pvalidate_pages(desc);
-
-	return pa;
-}
-
 void snp_accept_memory(phys_addr_t start, phys_addr_t end)
 {
-	struct snp_psc_desc desc = {};
-	unsigned int i;
-	phys_addr_t pa;
-
-	if (!boot_ghcb && !early_setup_ghcb())
-		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PSC);
-
-	pa = start;
-	while (pa < end)
-		pa = __snp_accept_memory(&desc, pa, end);
+	for (phys_addr_t pa = start; pa < end; pa += PAGE_SIZE)
+		__page_state_change(pa, SNP_PAGE_STATE_PRIVATE);
 }
 
 void sev_es_shutdown_ghcb(void)
@@ -682,3 +645,43 @@ void sev_prep_identity_maps(unsigned long top_level_pgt)
 
 	sev_verify_cbit(top_level_pgt);
 }
+
+bool early_is_sevsnp_guest(void)
+{
+	static bool sevsnp;
+
+	if (sevsnp)
+		return true;
+
+	if (!(sev_get_status() & MSR_AMD64_SEV_SNP_ENABLED))
+		return false;
+
+	sevsnp = true;
+
+	if (!snp_vmpl) {
+		unsigned int eax, ebx, ecx, edx;
+
+		/*
+		 * CPUID Fn8000_001F_EAX[28] - SVSM support
+		 */
+		eax = 0x8000001f;
+		ecx = 0;
+		native_cpuid(&eax, &ebx, &ecx, &edx);
+		if (eax & BIT(28)) {
+			struct msr m;
+
+			/* Obtain the address of the calling area to use */
+			boot_rdmsr(MSR_SVSM_CAA, &m);
+			boot_svsm_caa = (void *)m.q;
+			boot_svsm_caa_pa = m.q;
+
+			/*
+			 * The real VMPL level cannot be discovered, but the
+			 * memory acceptance routines make no use of that so
+			 * any non-zero value suffices here.
+			 */
+			snp_vmpl = U8_MAX;
+		}
+	}
+	return true;
+}
diff --git a/arch/x86/boot/compressed/sev.h b/arch/x86/boot/compressed/sev.h
index fc725a981b09..d3900384b8ab 100644
--- a/arch/x86/boot/compressed/sev.h
+++ b/arch/x86/boot/compressed/sev.h
@@ -12,11 +12,15 @@
 
 bool sev_snp_enabled(void);
 void snp_accept_memory(phys_addr_t start, phys_addr_t end);
+u64 sev_get_status(void);
+bool early_is_sevsnp_guest(void);
 
 #else
 
 static inline bool sev_snp_enabled(void) { return false; }
 static inline void snp_accept_memory(phys_addr_t start, phys_addr_t end) { }
+static inline u64 sev_get_status(void) { return 0; }
+static inline bool early_is_sevsnp_guest(void) { return false; }
 
 #endif
 
diff --git a/arch/x86/entry/entry.S b/arch/x86/entry/entry.S
index d3caa31240ed..175958b02f2b 100644
--- a/arch/x86/entry/entry.S
+++ b/arch/x86/entry/entry.S
@@ -17,19 +17,20 @@
 
 .pushsection .noinstr.text, "ax"
 
-SYM_FUNC_START(entry_ibpb)
+/* Clobbers AX, CX, DX */
+SYM_FUNC_START(write_ibpb)
 	ANNOTATE_NOENDBR
 	movl	$MSR_IA32_PRED_CMD, %ecx
-	movl	$PRED_CMD_IBPB, %eax
+	movl	_ASM_RIP(x86_pred_cmd), %eax
 	xorl	%edx, %edx
 	wrmsr
 
 	/* Make sure IBPB clears return stack preductions too. */
 	FILL_RETURN_BUFFER %rax, RSB_CLEAR_LOOPS, X86_BUG_IBPB_NO_RET
 	RET
-SYM_FUNC_END(entry_ibpb)
+SYM_FUNC_END(write_ibpb)
 /* For KVM */
-EXPORT_SYMBOL_GPL(entry_ibpb);
+EXPORT_SYMBOL_GPL(write_ibpb);
 
 .popsection
 
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 6866cc5acb0b..139ad80d1df3 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -629,7 +629,7 @@ int x86_pmu_hw_config(struct perf_event *event)
 	if (event->attr.type == event->pmu->type)
 		event->hw.config |= x86_pmu_get_event_config(event);
 
-	if (!event->attr.freq && x86_pmu.limit_period) {
+	if (is_sampling_event(event) && !event->attr.freq && x86_pmu.limit_period) {
 		s64 left = event->attr.sample_period;
 		x86_pmu.limit_period(event, &left);
 		if (left > event->attr.sample_period)
@@ -754,7 +754,7 @@ void x86_pmu_enable_all(int added)
 	}
 }
 
-static inline int is_x86_event(struct perf_event *event)
+int is_x86_event(struct perf_event *event)
 {
 	int i;
 
diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index 09d2d66c9f21..c5f385413392 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -3049,7 +3049,6 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 	int bit;
 	int handled = 0;
-	u64 intel_ctrl = hybrid(cpuc->pmu, intel_ctrl);
 
 	inc_irq_stat(apic_perf_irqs);
 
@@ -3093,7 +3092,6 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
 		handled++;
 		x86_pmu_handle_guest_pebs(regs, &data);
 		static_call(x86_pmu_drain_pebs)(regs, &data);
-		status &= intel_ctrl | GLOBAL_STATUS_TRACE_TOPAPMI;
 
 		/*
 		 * PMI throttle may be triggered, which stops the PEBS event.
@@ -3104,6 +3102,15 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
 		 */
 		if (pebs_enabled != cpuc->pebs_enabled)
 			wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
+
+		/*
+		 * Above PEBS handler (PEBS counters snapshotting) has updated fixed
+		 * counter 3 and perf metrics counts if they are in counter group,
+		 * unnecessary to update again.
+		 */
+		if (cpuc->events[INTEL_PMC_IDX_FIXED_SLOTS] &&
+		    is_pebs_counter_event_group(cpuc->events[INTEL_PMC_IDX_FIXED_SLOTS]))
+			status &= ~GLOBAL_STATUS_PERF_METRICS_OVF_BIT;
 	}
 
 	/*
@@ -3123,6 +3130,8 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
 		static_call(intel_pmu_update_topdown_event)(NULL, NULL);
 	}
 
+	status &= hybrid(cpuc->pmu, intel_ctrl);
+
 	/*
 	 * Checkpointed counters can lead to 'spurious' PMIs because the
 	 * rollback caused by the PMI will have cleared the overflow status
@@ -4386,7 +4395,7 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 	arr[pebs_enable] = (struct perf_guest_switch_msr){
 		.msr = MSR_IA32_PEBS_ENABLE,
 		.host = cpuc->pebs_enabled & ~cpuc->intel_ctrl_guest_mask,
-		.guest = pebs_mask & ~cpuc->intel_ctrl_host_mask,
+		.guest = pebs_mask & ~cpuc->intel_ctrl_host_mask & kvm_pmu->pebs_enable,
 	};
 
 	if (arr[pebs_enable].host) {
@@ -7305,8 +7314,17 @@ __init int intel_pmu_init(void)
 		name = "meteorlake_hybrid";
 		break;
 
+	case INTEL_PANTHERLAKE_L:
+		pr_cont("Pantherlake Hybrid events, ");
+		name = "pantherlake_hybrid";
+		goto lnl_common;
+
 	case INTEL_LUNARLAKE_M:
 	case INTEL_ARROWLAKE:
+		pr_cont("Lunarlake Hybrid events, ");
+		name = "lunarlake_hybrid";
+
+	lnl_common:
 		intel_pmu_init_hybrid(hybrid_big_small);
 
 		x86_pmu.pebs_latency_data = lnl_latency_data;
@@ -7328,8 +7346,6 @@ __init int intel_pmu_init(void)
 		intel_pmu_init_skt(&pmu->pmu);
 
 		intel_pmu_pebs_data_source_lnl();
-		pr_cont("Lunarlake Hybrid events, ");
-		name = "lunarlake_hybrid";
 		break;
 
 	case INTEL_ARROWLAKE_H:
diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index 1f7e1a692a7a..9b20acc0e932 100644
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -1399,8 +1399,10 @@ static u64 pebs_update_adaptive_cfg(struct perf_event *event)
 	 * + precise_ip < 2 for the non event IP
 	 * + For RTM TSX weight we need GPRs for the abort code.
 	 */
-	gprs = (sample_type & PERF_SAMPLE_REGS_INTR) &&
-	       (attr->sample_regs_intr & PEBS_GP_REGS);
+	gprs = ((sample_type & PERF_SAMPLE_REGS_INTR) &&
+		(attr->sample_regs_intr & PEBS_GP_REGS)) ||
+	       ((sample_type & PERF_SAMPLE_REGS_USER) &&
+		(attr->sample_regs_user & PEBS_GP_REGS));
 
 	tsx_weight = (sample_type & PERF_SAMPLE_WEIGHT_TYPE) &&
 		     ((attr->config & INTEL_ARCH_EVENT_MASK) ==
@@ -2123,7 +2125,7 @@ static void setup_pebs_adaptive_sample_data(struct perf_event *event,
 			regs->flags &= ~PERF_EFLAGS_EXACT;
 		}
 
-		if (sample_type & PERF_SAMPLE_REGS_INTR)
+		if (sample_type & (PERF_SAMPLE_REGS_INTR | PERF_SAMPLE_REGS_USER))
 			adaptive_pebs_save_regs(regs, gprs);
 	}
 
@@ -2377,8 +2379,25 @@ __intel_pmu_pebs_last_event(struct perf_event *event,
 			 */
 			intel_pmu_save_and_restart_reload(event, count);
 		}
-	} else
-		intel_pmu_save_and_restart(event);
+	} else {
+		/*
+		 * For a non-precise event, it's possible the
+		 * counters-snapshotting records a positive value for the
+		 * overflowed event. Then the HW auto-reload mechanism
+		 * reset the counter to 0 immediately, because the
+		 * pebs_event_reset is cleared if the PERF_X86_EVENT_AUTO_RELOAD
+		 * is not set. The counter backwards may be observed in a
+		 * PMI handler.
+		 *
+		 * Since the event value has been updated when processing the
+		 * counters-snapshotting record, only needs to set the new
+		 * period for the counter.
+		 */
+		if (is_pebs_counter_event_group(event))
+			static_call(x86_pmu_set_period)(event);
+		else
+			intel_pmu_save_and_restart(event);
+	}
 }
 
 static __always_inline void
diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c
index 60973c209c0e..76d96df1475a 100644
--- a/arch/x86/events/intel/uncore_snbep.c
+++ b/arch/x86/events/intel/uncore_snbep.c
@@ -4891,28 +4891,28 @@ static struct uncore_event_desc snr_uncore_iio_freerunning_events[] = {
 	INTEL_UNCORE_EVENT_DESC(ioclk,			"event=0xff,umask=0x10"),
 	/* Free-Running IIO BANDWIDTH IN Counters */
 	INTEL_UNCORE_EVENT_DESC(bw_in_port0,		"event=0xff,umask=0x20"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port1,		"event=0xff,umask=0x21"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port2,		"event=0xff,umask=0x22"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port3,		"event=0xff,umask=0x23"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port4,		"event=0xff,umask=0x24"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port5,		"event=0xff,umask=0x25"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port6,		"event=0xff,umask=0x26"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit,	"MiB"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port7,		"event=0xff,umask=0x27"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale,	"3.0517578125e-5"),
 	INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit,	"MiB"),
 	{ /* end: all zeroes */ },
 };
@@ -5485,37 +5485,6 @@ static struct freerunning_counters icx_iio_freerunning[] = {
 	[ICX_IIO_MSR_BW_IN]	= { 0xaa0, 0x1, 0x10, 8, 48, icx_iio_bw_freerunning_box_offsets },
 };
 
-static struct uncore_event_desc icx_uncore_iio_freerunning_events[] = {
-	/* Free-Running IIO CLOCKS Counter */
-	INTEL_UNCORE_EVENT_DESC(ioclk,			"event=0xff,umask=0x10"),
-	/* Free-Running IIO BANDWIDTH IN Counters */
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0,		"event=0xff,umask=0x20"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1,		"event=0xff,umask=0x21"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2,		"event=0xff,umask=0x22"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3,		"event=0xff,umask=0x23"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4,		"event=0xff,umask=0x24"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5,		"event=0xff,umask=0x25"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6,		"event=0xff,umask=0x26"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7,		"event=0xff,umask=0x27"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit,	"MiB"),
-	{ /* end: all zeroes */ },
-};
-
 static struct intel_uncore_type icx_uncore_iio_free_running = {
 	.name			= "iio_free_running",
 	.num_counters		= 9,
@@ -5523,7 +5492,7 @@ static struct intel_uncore_type icx_uncore_iio_free_running = {
 	.num_freerunning_types	= ICX_IIO_FREERUNNING_TYPE_MAX,
 	.freerunning		= icx_iio_freerunning,
 	.ops			= &skx_uncore_iio_freerunning_ops,
-	.event_descs		= icx_uncore_iio_freerunning_events,
+	.event_descs		= snr_uncore_iio_freerunning_events,
 	.format_group		= &skx_uncore_iio_freerunning_format_group,
 };
 
@@ -6320,69 +6289,13 @@ static struct freerunning_counters spr_iio_freerunning[] = {
 	[SPR_IIO_MSR_BW_OUT]	= { 0x3808, 0x1, 0x10, 8, 48 },
 };
 
-static struct uncore_event_desc spr_uncore_iio_freerunning_events[] = {
-	/* Free-Running IIO CLOCKS Counter */
-	INTEL_UNCORE_EVENT_DESC(ioclk,			"event=0xff,umask=0x10"),
-	/* Free-Running IIO BANDWIDTH IN Counters */
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0,		"event=0xff,umask=0x20"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1,		"event=0xff,umask=0x21"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2,		"event=0xff,umask=0x22"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3,		"event=0xff,umask=0x23"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4,		"event=0xff,umask=0x24"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5,		"event=0xff,umask=0x25"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6,		"event=0xff,umask=0x26"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7,		"event=0xff,umask=0x27"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit,	"MiB"),
-	/* Free-Running IIO BANDWIDTH OUT Counters */
-	INTEL_UNCORE_EVENT_DESC(bw_out_port0,		"event=0xff,umask=0x30"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port0.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port0.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port1,		"event=0xff,umask=0x31"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port1.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port1.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port2,		"event=0xff,umask=0x32"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port2.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port2.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port3,		"event=0xff,umask=0x33"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port3.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port3.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port4,		"event=0xff,umask=0x34"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port4.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port4.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port5,		"event=0xff,umask=0x35"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port5.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port5.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port6,		"event=0xff,umask=0x36"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port6.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port6.unit,	"MiB"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port7,		"event=0xff,umask=0x37"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port7.scale,	"3.814697266e-6"),
-	INTEL_UNCORE_EVENT_DESC(bw_out_port7.unit,	"MiB"),
-	{ /* end: all zeroes */ },
-};
-
 static struct intel_uncore_type spr_uncore_iio_free_running = {
 	.name			= "iio_free_running",
 	.num_counters		= 17,
 	.num_freerunning_types	= SPR_IIO_FREERUNNING_TYPE_MAX,
 	.freerunning		= spr_iio_freerunning,
 	.ops			= &skx_uncore_iio_freerunning_ops,
-	.event_descs		= spr_uncore_iio_freerunning_events,
+	.event_descs		= snr_uncore_iio_freerunning_events,
 	.format_group		= &skx_uncore_iio_freerunning_format_group,
 };
 
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index 2c0ce0e9545e..46d120597bab 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -110,14 +110,21 @@ static inline bool is_topdown_event(struct perf_event *event)
 	return is_metric_event(event) || is_slots_event(event);
 }
 
+int is_x86_event(struct perf_event *event);
+
+static inline bool check_leader_group(struct perf_event *leader, int flags)
+{
+	return is_x86_event(leader) ? !!(leader->hw.flags & flags) : false;
+}
+
 static inline bool is_branch_counters_group(struct perf_event *event)
 {
-	return event->group_leader->hw.flags & PERF_X86_EVENT_BRANCH_COUNTERS;
+	return check_leader_group(event->group_leader, PERF_X86_EVENT_BRANCH_COUNTERS);
 }
 
 static inline bool is_pebs_counter_event_group(struct perf_event *event)
 {
-	return event->group_leader->hw.flags & PERF_X86_EVENT_PEBS_CNTR;
+	return check_leader_group(event->group_leader, PERF_X86_EVENT_PEBS_CNTR);
 }
 
 struct amd_nb {
diff --git a/arch/x86/include/asm/intel-family.h b/arch/x86/include/asm/intel-family.h
index 3a97a7eefb51..be10c188614f 100644
--- a/arch/x86/include/asm/intel-family.h
+++ b/arch/x86/include/asm/intel-family.h
@@ -126,6 +126,8 @@
 #define INTEL_GRANITERAPIDS_X		IFM(6, 0xAD) /* Redwood Cove */
 #define INTEL_GRANITERAPIDS_D		IFM(6, 0xAE)
 
+#define INTEL_BARTLETTLAKE		IFM(6, 0xD7) /* Raptor Cove */
+
 /* "Hybrid" Processors (P-Core/E-Core) */
 
 #define INTEL_LAKEFIELD			IFM(6, 0x8A) /* Sunny Cove / Tremont */
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index a884ab544335..7bc174a1f1cb 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -35,6 +35,7 @@
 #include <asm/mtrr.h>
 #include <asm/msr-index.h>
 #include <asm/asm.h>
+#include <asm/irq_remapping.h>
 #include <asm/kvm_page_track.h>
 #include <asm/kvm_vcpu_regs.h>
 #include <asm/reboot.h>
@@ -1472,8 +1473,13 @@ struct kvm_arch {
 	struct once nx_once;
 
 #ifdef CONFIG_X86_64
-	/* The number of TDP MMU pages across all roots. */
+#ifdef CONFIG_KVM_PROVE_MMU
+	/*
+	 * The number of TDP MMU pages across all roots.  Used only to sanity
+	 * check that KVM isn't leaking TDP MMU pages.
+	 */
 	atomic64_t tdp_mmu_pages;
+#endif
 
 	/*
 	 * List of struct kvm_mmu_pages being used as roots.
@@ -2418,4 +2424,9 @@ int memslot_rmap_alloc(struct kvm_memory_slot *slot, unsigned long npages);
  */
 #define KVM_EXIT_HYPERCALL_MBZ		GENMASK_ULL(31, 1)
 
+static inline bool kvm_arch_has_irq_bypass(void)
+{
+	return enable_apicv && irq_remapping_cap(IRQ_POSTING_CAP);
+}
+
 #endif /* _ASM_X86_KVM_HOST_H */
diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h
index 8a5cc8e70439..5c43f145454d 100644
--- a/arch/x86/include/asm/nospec-branch.h
+++ b/arch/x86/include/asm/nospec-branch.h
@@ -269,7 +269,7 @@
  * typically has NO_MELTDOWN).
  *
  * While retbleed_untrain_ret() doesn't clobber anything but requires stack,
- * entry_ibpb() will clobber AX, CX, DX.
+ * write_ibpb() will clobber AX, CX, DX.
  *
  * As such, this must be placed after every *SWITCH_TO_KERNEL_CR3 at a point
  * where we have a stack but before any RET instruction.
@@ -279,7 +279,7 @@
 	VALIDATE_UNRET_END
 	CALL_UNTRAIN_RET
 	ALTERNATIVE_2 "",						\
-		      "call entry_ibpb", \ibpb_feature,			\
+		      "call write_ibpb", \ibpb_feature,			\
 		     __stringify(\call_depth_insns), X86_FEATURE_CALL_DEPTH
 #endif
 .endm
@@ -368,7 +368,7 @@ extern void srso_return_thunk(void);
 extern void srso_alias_return_thunk(void);
 
 extern void entry_untrain_ret(void);
-extern void entry_ibpb(void);
+extern void write_ibpb(void);
 
 #ifdef CONFIG_X86_64
 extern void clear_bhb_loop(void);
@@ -514,11 +514,11 @@ void alternative_msr_write(unsigned int msr, u64 val, unsigned int feature)
 		: "memory");
 }
 
-extern u64 x86_pred_cmd;
-
 static inline void indirect_branch_prediction_barrier(void)
 {
-	alternative_msr_write(MSR_IA32_PRED_CMD, x86_pred_cmd, X86_FEATURE_IBPB);
+	asm_inline volatile(ALTERNATIVE("", "call write_ibpb", X86_FEATURE_IBPB)
+			    : ASM_CALL_CONSTRAINT
+			    :: "rax", "rcx", "rdx", "memory");
 }
 
 /* The Intel SPEC CTRL MSR base value cache */
diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h
index a33147520044..c88691b15f3c 100644
--- a/arch/x86/include/asm/pgalloc.h
+++ b/arch/x86/include/asm/pgalloc.h
@@ -6,6 +6,8 @@
 #include <linux/mm.h>		/* for struct page */
 #include <linux/pagemap.h>
 
+#include <asm/cpufeature.h>
+
 #define __HAVE_ARCH_PTE_ALLOC_ONE
 #define __HAVE_ARCH_PGD_FREE
 #include <asm-generic/pgalloc.h>
@@ -29,16 +31,17 @@ static inline void paravirt_release_pud(unsigned long pfn) {}
 static inline void paravirt_release_p4d(unsigned long pfn) {}
 #endif
 
-#ifdef CONFIG_MITIGATION_PAGE_TABLE_ISOLATION
 /*
- * Instead of one PGD, we acquire two PGDs.  Being order-1, it is
- * both 8k in size and 8k-aligned.  That lets us just flip bit 12
- * in a pointer to swap between the two 4k halves.
+ * In case of Page Table Isolation active, we acquire two PGDs instead of one.
+ * Being order-1, it is both 8k in size and 8k-aligned.  That lets us just
+ * flip bit 12 in a pointer to swap between the two 4k halves.
  */
-#define PGD_ALLOCATION_ORDER 1
-#else
-#define PGD_ALLOCATION_ORDER 0
-#endif
+static inline unsigned int pgd_allocation_order(void)
+{
+	if (cpu_feature_enabled(X86_FEATURE_PTI))
+		return 1;
+	return 0;
+}
 
 /*
  * Allocate and free page tables.
diff --git a/arch/x86/include/asm/smap.h b/arch/x86/include/asm/smap.h
index 55a5e656e4b9..4f84d421d1cf 100644
--- a/arch/x86/include/asm/smap.h
+++ b/arch/x86/include/asm/smap.h
@@ -16,23 +16,23 @@
 #ifdef __ASSEMBLER__
 
 #define ASM_CLAC \
-	ALTERNATIVE __stringify(ANNOTATE_IGNORE_ALTERNATIVE), "clac", X86_FEATURE_SMAP
+	ALTERNATIVE "", "clac", X86_FEATURE_SMAP
 
 #define ASM_STAC \
-	ALTERNATIVE __stringify(ANNOTATE_IGNORE_ALTERNATIVE), "stac", X86_FEATURE_SMAP
+	ALTERNATIVE "", "stac", X86_FEATURE_SMAP
 
 #else /* __ASSEMBLER__ */
 
 static __always_inline void clac(void)
 {
 	/* Note: a barrier is implicit in alternative() */
-	alternative(ANNOTATE_IGNORE_ALTERNATIVE "", "clac", X86_FEATURE_SMAP);
+	alternative("", "clac", X86_FEATURE_SMAP);
 }
 
 static __always_inline void stac(void)
 {
 	/* Note: a barrier is implicit in alternative() */
-	alternative(ANNOTATE_IGNORE_ALTERNATIVE "", "stac", X86_FEATURE_SMAP);
+	alternative("", "stac", X86_FEATURE_SMAP);
 }
 
 static __always_inline unsigned long smap_save(void)
@@ -59,9 +59,9 @@ static __always_inline void smap_restore(unsigned long flags)
 
 /* These macros can be used in asm() statements */
 #define ASM_CLAC \
-	ALTERNATIVE(ANNOTATE_IGNORE_ALTERNATIVE "", "clac", X86_FEATURE_SMAP)
+	ALTERNATIVE("", "clac", X86_FEATURE_SMAP)
 #define ASM_STAC \
-	ALTERNATIVE(ANNOTATE_IGNORE_ALTERNATIVE "", "stac", X86_FEATURE_SMAP)
+	ALTERNATIVE("", "stac", X86_FEATURE_SMAP)
 
 #define ASM_CLAC_UNSAFE \
 	ALTERNATIVE("", ANNOTATE_IGNORE_ALTERNATIVE "clac", X86_FEATURE_SMAP)
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index dae6a73be40e..9fa321a95eb3 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -23,6 +23,8 @@
 #include <linux/serial_core.h>
 #include <linux/pgtable.h>
 
+#include <xen/xen.h>
+
 #include <asm/e820/api.h>
 #include <asm/irqdomain.h>
 #include <asm/pci_x86.h>
@@ -1729,6 +1731,15 @@ int __init acpi_mps_check(void)
 {
 #if defined(CONFIG_X86_LOCAL_APIC) && !defined(CONFIG_X86_MPPARSE)
 /* mptable code is not built-in*/
+
+	/*
+	 * Xen disables ACPI in PV DomU guests but it still emulates APIC and
+	 * supports SMP. Returning early here ensures that APIC is not disabled
+	 * unnecessarily and the guest is not limited to a single vCPU.
+	 */
+	if (xen_pv_domain() && !xen_initial_domain())
+		return 0;
+
 	if (acpi_disabled || acpi_noirq) {
 		pr_warn("MPS support code is not built-in, using acpi=off or acpi=noirq or pci=noacpi may have problem\n");
 		return 1;
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 79569f72b8ee..2b36379ff675 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -805,6 +805,7 @@ static void init_amd_bd(struct cpuinfo_x86 *c)
 static const struct x86_cpu_id erratum_1386_microcode[] = {
 	X86_MATCH_VFM_STEPS(VFM_MAKE(X86_VENDOR_AMD, 0x17, 0x01), 0x2, 0x2, 0x0800126e),
 	X86_MATCH_VFM_STEPS(VFM_MAKE(X86_VENDOR_AMD, 0x17, 0x31), 0x0, 0x0, 0x08301052),
+	{}
 };
 
 static void fix_erratum_1386(struct cpuinfo_x86 *c)
@@ -868,6 +869,16 @@ static void init_amd_zen1(struct cpuinfo_x86 *c)
 
 	pr_notice_once("AMD Zen1 DIV0 bug detected. Disable SMT for full protection.\n");
 	setup_force_cpu_bug(X86_BUG_DIV0);
+
+	/*
+	 * Turn off the Instructions Retired free counter on machines that are
+	 * susceptible to erratum #1054 "Instructions Retired Performance
+	 * Counter May Be Inaccurate".
+	 */
+	if (c->x86_model < 0x30) {
+		msr_clear_bit(MSR_K7_HWCR, MSR_K7_HWCR_IRPERF_EN_BIT);
+		clear_cpu_cap(c, X86_FEATURE_IRPERF);
+	}
 }
 
 static bool cpu_has_zenbleed_microcode(void)
@@ -1051,13 +1062,8 @@ static void init_amd(struct cpuinfo_x86 *c)
 	if (!cpu_feature_enabled(X86_FEATURE_XENPV))
 		set_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS);
 
-	/*
-	 * Turn on the Instructions Retired free counter on machines not
-	 * susceptible to erratum #1054 "Instructions Retired Performance
-	 * Counter May Be Inaccurate".
-	 */
-	if (cpu_has(c, X86_FEATURE_IRPERF) &&
-	    (boot_cpu_has(X86_FEATURE_ZEN1) && c->x86_model > 0x2f))
+	/* Enable the Instructions Retired free counter */
+	if (cpu_has(c, X86_FEATURE_IRPERF))
 		msr_set_bit(MSR_K7_HWCR, MSR_K7_HWCR_IRPERF_EN_BIT);
 
 	check_null_seg_clears_base(c);
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index 4386aa6c69e1..362602b705cc 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -59,7 +59,6 @@ DEFINE_PER_CPU(u64, x86_spec_ctrl_current);
 EXPORT_PER_CPU_SYMBOL_GPL(x86_spec_ctrl_current);
 
 u64 x86_pred_cmd __ro_after_init = PRED_CMD_IBPB;
-EXPORT_SYMBOL_GPL(x86_pred_cmd);
 
 static u64 __ro_after_init x86_arch_cap_msr;
 
@@ -1142,7 +1141,7 @@ do_cmd_auto:
 		setup_clear_cpu_cap(X86_FEATURE_RETHUNK);
 
 		/*
-		 * There is no need for RSB filling: entry_ibpb() ensures
+		 * There is no need for RSB filling: write_ibpb() ensures
 		 * all predictions, including the RSB, are invalidated,
 		 * regardless of IBPB implementation.
 		 */
@@ -1592,51 +1591,54 @@ static void __init spec_ctrl_disable_kernel_rrsba(void)
 	rrsba_disabled = true;
 }
 
-static void __init spectre_v2_determine_rsb_fill_type_at_vmexit(enum spectre_v2_mitigation mode)
+static void __init spectre_v2_select_rsb_mitigation(enum spectre_v2_mitigation mode)
 {
 	/*
-	 * Similar to context switches, there are two types of RSB attacks
-	 * after VM exit:
+	 * WARNING! There are many subtleties to consider when changing *any*
+	 * code related to RSB-related mitigations.  Before doing so, carefully
+	 * read the following document, and update if necessary:
 	 *
-	 * 1) RSB underflow
+	 *   Documentation/admin-guide/hw-vuln/rsb.rst
 	 *
-	 * 2) Poisoned RSB entry
+	 * In an overly simplified nutshell:
 	 *
-	 * When retpoline is enabled, both are mitigated by filling/clearing
-	 * the RSB.
+	 *   - User->user RSB attacks are conditionally mitigated during
+	 *     context switches by cond_mitigation -> write_ibpb().
 	 *
-	 * When IBRS is enabled, while #1 would be mitigated by the IBRS branch
-	 * prediction isolation protections, RSB still needs to be cleared
-	 * because of #2.  Note that SMEP provides no protection here, unlike
-	 * user-space-poisoned RSB entries.
+	 *   - User->kernel and guest->host attacks are mitigated by eIBRS or
+	 *     RSB filling.
 	 *
-	 * eIBRS should protect against RSB poisoning, but if the EIBRS_PBRSB
-	 * bug is present then a LITE version of RSB protection is required,
-	 * just a single call needs to retire before a RET is executed.
+	 *     Though, depending on config, note that other alternative
+	 *     mitigations may end up getting used instead, e.g., IBPB on
+	 *     entry/vmexit, call depth tracking, or return thunks.
 	 */
+
 	switch (mode) {
 	case SPECTRE_V2_NONE:
-		return;
+		break;
 
-	case SPECTRE_V2_EIBRS_LFENCE:
 	case SPECTRE_V2_EIBRS:
+	case SPECTRE_V2_EIBRS_LFENCE:
+	case SPECTRE_V2_EIBRS_RETPOLINE:
 		if (boot_cpu_has_bug(X86_BUG_EIBRS_PBRSB)) {
-			setup_force_cpu_cap(X86_FEATURE_RSB_VMEXIT_LITE);
 			pr_info("Spectre v2 / PBRSB-eIBRS: Retire a single CALL on VMEXIT\n");
+			setup_force_cpu_cap(X86_FEATURE_RSB_VMEXIT_LITE);
 		}
-		return;
+		break;
 
-	case SPECTRE_V2_EIBRS_RETPOLINE:
 	case SPECTRE_V2_RETPOLINE:
 	case SPECTRE_V2_LFENCE:
 	case SPECTRE_V2_IBRS:
+		pr_info("Spectre v2 / SpectreRSB: Filling RSB on context switch and VMEXIT\n");
+		setup_force_cpu_cap(X86_FEATURE_RSB_CTXSW);
 		setup_force_cpu_cap(X86_FEATURE_RSB_VMEXIT);
-		pr_info("Spectre v2 / SpectreRSB : Filling RSB on VMEXIT\n");
-		return;
-	}
+		break;
 
-	pr_warn_once("Unknown Spectre v2 mode, disabling RSB mitigation at VM exit");
-	dump_stack();
+	default:
+		pr_warn_once("Unknown Spectre v2 mode, disabling RSB mitigation\n");
+		dump_stack();
+		break;
+	}
 }
 
 /*
@@ -1830,48 +1832,7 @@ static void __init spectre_v2_select_mitigation(void)
 	spectre_v2_enabled = mode;
 	pr_info("%s\n", spectre_v2_strings[mode]);
 
-	/*
-	 * If Spectre v2 protection has been enabled, fill the RSB during a
-	 * context switch.  In general there are two types of RSB attacks
-	 * across context switches, for which the CALLs/RETs may be unbalanced.
-	 *
-	 * 1) RSB underflow
-	 *
-	 *    Some Intel parts have "bottomless RSB".  When the RSB is empty,
-	 *    speculated return targets may come from the branch predictor,
-	 *    which could have a user-poisoned BTB or BHB entry.
-	 *
-	 *    AMD has it even worse: *all* returns are speculated from the BTB,
-	 *    regardless of the state of the RSB.
-	 *
-	 *    When IBRS or eIBRS is enabled, the "user -> kernel" attack
-	 *    scenario is mitigated by the IBRS branch prediction isolation
-	 *    properties, so the RSB buffer filling wouldn't be necessary to
-	 *    protect against this type of attack.
-	 *
-	 *    The "user -> user" attack scenario is mitigated by RSB filling.
-	 *
-	 * 2) Poisoned RSB entry
-	 *
-	 *    If the 'next' in-kernel return stack is shorter than 'prev',
-	 *    'next' could be tricked into speculating with a user-poisoned RSB
-	 *    entry.
-	 *
-	 *    The "user -> kernel" attack scenario is mitigated by SMEP and
-	 *    eIBRS.
-	 *
-	 *    The "user -> user" scenario, also known as SpectreBHB, requires
-	 *    RSB clearing.
-	 *
-	 * So to mitigate all cases, unconditionally fill RSB on context
-	 * switches.
-	 *
-	 * FIXME: Is this pointless for retbleed-affected AMD?
-	 */
-	setup_force_cpu_cap(X86_FEATURE_RSB_CTXSW);
-	pr_info("Spectre v2 / SpectreRSB mitigation: Filling RSB on context switch\n");
-
-	spectre_v2_determine_rsb_fill_type_at_vmexit(mode);
+	spectre_v2_select_rsb_mitigation(mode);
 
 	/*
 	 * Retpoline protects the kernel, but doesn't protect firmware.  IBRS
@@ -2676,7 +2637,7 @@ static void __init srso_select_mitigation(void)
 				setup_clear_cpu_cap(X86_FEATURE_RETHUNK);
 
 				/*
-				 * There is no need for RSB filling: entry_ibpb() ensures
+				 * There is no need for RSB filling: write_ibpb() ensures
 				 * all predictions, including the RSB, are invalidated,
 				 * regardless of IBPB implementation.
 				 */
@@ -2701,7 +2662,7 @@ ibpb_on_vmexit:
 				srso_mitigation = SRSO_MITIGATION_IBPB_ON_VMEXIT;
 
 				/*
-				 * There is no need for RSB filling: entry_ibpb() ensures
+				 * There is no need for RSB filling: write_ibpb() ensures
 				 * all predictions, including the RSB, are invalidated,
 				 * regardless of IBPB implementation.
 				 */
diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c
index b61028cf5c8a..4a10d35e70aa 100644
--- a/arch/x86/kernel/cpu/microcode/amd.c
+++ b/arch/x86/kernel/cpu/microcode/amd.c
@@ -199,6 +199,12 @@ static bool need_sha_check(u32 cur_rev)
 	case 0xa70c0: return cur_rev <= 0xa70C009; break;
 	case 0xaa001: return cur_rev <= 0xaa00116; break;
 	case 0xaa002: return cur_rev <= 0xaa00218; break;
+	case 0xb0021: return cur_rev <= 0xb002146; break;
+	case 0xb1010: return cur_rev <= 0xb101046; break;
+	case 0xb2040: return cur_rev <= 0xb204031; break;
+	case 0xb4040: return cur_rev <= 0xb404031; break;
+	case 0xb6000: return cur_rev <= 0xb600031; break;
+	case 0xb7000: return cur_rev <= 0xb700031; break;
 	default: break;
 	}
 
@@ -214,8 +220,7 @@ static bool verify_sha256_digest(u32 patch_id, u32 cur_rev, const u8 *data, unsi
 	struct sha256_state s;
 	int i;
 
-	if (x86_family(bsp_cpuid_1_eax) < 0x17 ||
-	    x86_family(bsp_cpuid_1_eax) > 0x19)
+	if (x86_family(bsp_cpuid_1_eax) < 0x17)
 		return true;
 
 	if (!need_sha_check(cur_rev))
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 93ec829015f1..cc4a54145c83 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -3553,6 +3553,22 @@ static void mkdir_rdt_prepare_rmid_free(struct rdtgroup *rgrp)
 		free_rmid(rgrp->closid, rgrp->mon.rmid);
 }
 
+/*
+ * We allow creating mon groups only with in a directory called "mon_groups"
+ * which is present in every ctrl_mon group. Check if this is a valid
+ * "mon_groups" directory.
+ *
+ * 1. The directory should be named "mon_groups".
+ * 2. The mon group itself should "not" be named "mon_groups".
+ *   This makes sure "mon_groups" directory always has a ctrl_mon group
+ *   as parent.
+ */
+static bool is_mon_groups(struct kernfs_node *kn, const char *name)
+{
+	return (!strcmp(rdt_kn_name(kn), "mon_groups") &&
+		strcmp(name, "mon_groups"));
+}
+
 static int mkdir_rdt_prepare(struct kernfs_node *parent_kn,
 			     const char *name, umode_t mode,
 			     enum rdt_group_type rtype, struct rdtgroup **r)
@@ -3568,6 +3584,15 @@ static int mkdir_rdt_prepare(struct kernfs_node *parent_kn,
 		goto out_unlock;
 	}
 
+	/*
+	 * Check that the parent directory for a monitor group is a "mon_groups"
+	 * directory.
+	 */
+	if (rtype == RDTMON_GROUP && !is_mon_groups(parent_kn, name)) {
+		ret = -EPERM;
+		goto out_unlock;
+	}
+
 	if (rtype == RDTMON_GROUP &&
 	    (prdtgrp->mode == RDT_MODE_PSEUDO_LOCKSETUP ||
 	     prdtgrp->mode == RDT_MODE_PSEUDO_LOCKED)) {
@@ -3751,22 +3776,6 @@ out_unlock:
 	return ret;
 }
 
-/*
- * We allow creating mon groups only with in a directory called "mon_groups"
- * which is present in every ctrl_mon group. Check if this is a valid
- * "mon_groups" directory.
- *
- * 1. The directory should be named "mon_groups".
- * 2. The mon group itself should "not" be named "mon_groups".
- *   This makes sure "mon_groups" directory always has a ctrl_mon group
- *   as parent.
- */
-static bool is_mon_groups(struct kernfs_node *kn, const char *name)
-{
-	return (!strcmp(rdt_kn_name(kn), "mon_groups") &&
-		strcmp(name, "mon_groups"));
-}
-
 static int rdtgroup_mkdir(struct kernfs_node *parent_kn, const char *name,
 			  umode_t mode)
 {
@@ -3782,11 +3791,8 @@ static int rdtgroup_mkdir(struct kernfs_node *parent_kn, const char *name,
 	if (resctrl_arch_alloc_capable() && parent_kn == rdtgroup_default.kn)
 		return rdtgroup_mkdir_ctrl_mon(parent_kn, name, mode);
 
-	/*
-	 * If RDT monitoring is supported and the parent directory is a valid
-	 * "mon_groups" directory, add a monitoring subdirectory.
-	 */
-	if (resctrl_arch_mon_capable() && is_mon_groups(parent_kn, name))
+	/* Else, attempt to add a monitoring subdirectory. */
+	if (resctrl_arch_mon_capable())
 		return rdtgroup_mkdir_mon(parent_kn, name, mode);
 
 	return -EPERM;
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 57120f0749cc..9920122018a0 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -753,22 +753,21 @@ void __init e820__memory_setup_extended(u64 phys_addr, u32 data_len)
 void __init e820__register_nosave_regions(unsigned long limit_pfn)
 {
 	int i;
-	unsigned long pfn = 0;
+	u64 last_addr = 0;
 
 	for (i = 0; i < e820_table->nr_entries; i++) {
 		struct e820_entry *entry = &e820_table->entries[i];
 
-		if (pfn < PFN_UP(entry->addr))
-			register_nosave_region(pfn, PFN_UP(entry->addr));
-
-		pfn = PFN_DOWN(entry->addr + entry->size);
-
 		if (entry->type != E820_TYPE_RAM)
-			register_nosave_region(PFN_UP(entry->addr), pfn);
+			continue;
 
-		if (pfn >= limit_pfn)
-			break;
+		if (last_addr < entry->addr)
+			register_nosave_region(PFN_DOWN(last_addr), PFN_UP(entry->addr));
+
+		last_addr = entry->addr + entry->size;
 	}
+
+	register_nosave_region(PFN_DOWN(last_addr), limit_pfn);
 }
 
 #ifdef CONFIG_ACPI
@@ -1300,6 +1299,14 @@ void __init e820__memblock_setup(void)
 		memblock_add(entry->addr, entry->size);
 	}
 
+	/*
+	 * 32-bit systems are limited to 4BG of memory even with HIGHMEM and
+	 * to even less without it.
+	 * Discard memory after max_pfn - the actual limit detected at runtime.
+	 */
+	if (IS_ENABLED(CONFIG_X86_32))
+		memblock_remove(PFN_PHYS(max_pfn), -1);
+
 	/* Throw away partial pages: */
 	memblock_trim_memory(PAGE_SIZE);
 
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index 611f27e3890c..3aad78bfcb26 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -389,10 +389,10 @@ static int __init setup_early_printk(char *buf)
 	keep = (strstr(buf, "keep") != NULL);
 
 	while (*buf != '\0') {
-		if (!strncmp(buf, "mmio", 4)) {
-			early_mmio_serial_init(buf + 4);
+		if (!strncmp(buf, "mmio32", 6)) {
+			buf += 6;
+			early_mmio_serial_init(buf);
 			early_console_register(&early_serial_console, keep);
-			buf += 4;
 		}
 		if (!strncmp(buf, "serial", 6)) {
 			buf += 6;
@@ -407,9 +407,9 @@ static int __init setup_early_printk(char *buf)
 		}
 #ifdef CONFIG_PCI
 		if (!strncmp(buf, "pciserial", 9)) {
-			early_pci_serial_init(buf + 9);
+			buf += 9; /* Keep from match the above "pciserial" */
+			early_pci_serial_init(buf);
 			early_console_register(&early_serial_console, keep);
-			buf += 9; /* Keep from match the above "serial" */
 		}
 #endif
 		if (!strncmp(buf, "vga", 3) &&
diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index 80e262bb627f..cb9852ad6098 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -46,7 +46,8 @@ bool __init pit_timer_init(void)
 		 * VMMs otherwise steal CPU time just to pointlessly waggle
 		 * the (masked) IRQ.
 		 */
-		clockevent_i8253_disable();
+		scoped_guard(irq)
+			clockevent_i8253_disable();
 		return false;
 	}
 	clockevent_i8253_init(true);
diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 80265162aeff..1f325304c4a8 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -42,7 +42,7 @@ static void load_segments(void)
 
 static void machine_kexec_free_page_tables(struct kimage *image)
 {
-	free_pages((unsigned long)image->arch.pgd, PGD_ALLOCATION_ORDER);
+	free_pages((unsigned long)image->arch.pgd, pgd_allocation_order());
 	image->arch.pgd = NULL;
 #ifdef CONFIG_X86_PAE
 	free_page((unsigned long)image->arch.pmd0);
@@ -59,7 +59,7 @@ static void machine_kexec_free_page_tables(struct kimage *image)
 static int machine_kexec_alloc_page_tables(struct kimage *image)
 {
 	image->arch.pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
-						    PGD_ALLOCATION_ORDER);
+						    pgd_allocation_order());
 #ifdef CONFIG_X86_PAE
 	image->arch.pmd0 = (pmd_t *)get_zeroed_page(GFP_KERNEL);
 	image->arch.pmd1 = (pmd_t *)get_zeroed_page(GFP_KERNEL);
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 5e4d4934c0d3..571c906ffcbf 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -1427,8 +1427,8 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 		}
 		break;
 	case 0xa: { /* Architectural Performance Monitoring */
-		union cpuid10_eax eax;
-		union cpuid10_edx edx;
+		union cpuid10_eax eax = { };
+		union cpuid10_edx edx = { };
 
 		if (!enable_pmu || !static_cpu_has(X86_FEATURE_ARCH_PERFMON)) {
 			entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
@@ -1444,8 +1444,6 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 
 		if (kvm_pmu_cap.version)
 			edx.split.anythread_deprecated = 1;
-		edx.split.reserved1 = 0;
-		edx.split.reserved2 = 0;
 
 		entry->eax = eax.full;
 		entry->ebx = kvm_pmu_cap.events_mask;
@@ -1763,7 +1761,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 		break;
 	/* AMD Extended Performance Monitoring and Debug */
 	case 0x80000022: {
-		union cpuid_0x80000022_ebx ebx;
+		union cpuid_0x80000022_ebx ebx = { };
 
 		entry->ecx = entry->edx = 0;
 		if (!enable_pmu || !kvm_cpu_cap_has(X86_FEATURE_PERFMON_V2)) {
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index 7cc0564f5f97..21a3b8166242 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -40,7 +40,9 @@ void kvm_mmu_uninit_tdp_mmu(struct kvm *kvm)
 	kvm_tdp_mmu_invalidate_roots(kvm, KVM_VALID_ROOTS);
 	kvm_tdp_mmu_zap_invalidated_roots(kvm, false);
 
-	WARN_ON(atomic64_read(&kvm->arch.tdp_mmu_pages));
+#ifdef CONFIG_KVM_PROVE_MMU
+	KVM_MMU_WARN_ON(atomic64_read(&kvm->arch.tdp_mmu_pages));
+#endif
 	WARN_ON(!list_empty(&kvm->arch.tdp_mmu_roots));
 
 	/*
@@ -325,13 +327,17 @@ static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn,
 static void tdp_account_mmu_page(struct kvm *kvm, struct kvm_mmu_page *sp)
 {
 	kvm_account_pgtable_pages((void *)sp->spt, +1);
+#ifdef CONFIG_KVM_PROVE_MMU
 	atomic64_inc(&kvm->arch.tdp_mmu_pages);
+#endif
 }
 
 static void tdp_unaccount_mmu_page(struct kvm *kvm, struct kvm_mmu_page *sp)
 {
 	kvm_account_pgtable_pages((void *)sp->spt, -1);
+#ifdef CONFIG_KVM_PROVE_MMU
 	atomic64_dec(&kvm->arch.tdp_mmu_pages);
+#endif
 }
 
 /**
diff --git a/arch/x86/kvm/svm/avic.c b/arch/x86/kvm/svm/avic.c
index 65fd245a9953..7338879d1c0c 100644
--- a/arch/x86/kvm/svm/avic.c
+++ b/arch/x86/kvm/svm/avic.c
@@ -796,12 +796,15 @@ static int svm_ir_list_add(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi)
 	struct amd_svm_iommu_ir *ir;
 	u64 entry;
 
+	if (WARN_ON_ONCE(!pi->ir_data))
+		return -EINVAL;
+
 	/**
 	 * In some cases, the existing irte is updated and re-set,
 	 * so we need to check here if it's already been * added
 	 * to the ir_list.
 	 */
-	if (pi->ir_data && (pi->prev_ga_tag != 0)) {
+	if (pi->prev_ga_tag) {
 		struct kvm *kvm = svm->vcpu.kvm;
 		u32 vcpu_id = AVIC_GATAG_TO_VCPUID(pi->prev_ga_tag);
 		struct kvm_vcpu *prev_vcpu = kvm_get_vcpu_by_id(kvm, vcpu_id);
@@ -820,7 +823,7 @@ static int svm_ir_list_add(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi)
 	 * Allocating new amd_iommu_pi_data, which will get
 	 * add to the per-vcpu ir_list.
 	 */
-	ir = kzalloc(sizeof(struct amd_svm_iommu_ir), GFP_KERNEL_ACCOUNT);
+	ir = kzalloc(sizeof(struct amd_svm_iommu_ir), GFP_ATOMIC | __GFP_ACCOUNT);
 	if (!ir) {
 		ret = -ENOMEM;
 		goto out;
@@ -896,10 +899,10 @@ int avic_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 {
 	struct kvm_kernel_irq_routing_entry *e;
 	struct kvm_irq_routing_table *irq_rt;
+	bool enable_remapped_mode = true;
 	int idx, ret = 0;
 
-	if (!kvm_arch_has_assigned_device(kvm) ||
-	    !irq_remapping_cap(IRQ_POSTING_CAP))
+	if (!kvm_arch_has_assigned_device(kvm) || !kvm_arch_has_irq_bypass())
 		return 0;
 
 	pr_debug("SVM: %s: host_irq=%#x, guest_irq=%#x, set=%#x\n",
@@ -933,6 +936,8 @@ int avic_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 		    kvm_vcpu_apicv_active(&svm->vcpu)) {
 			struct amd_iommu_pi_data pi;
 
+			enable_remapped_mode = false;
+
 			/* Try to enable guest_mode in IRTE */
 			pi.base = __sme_set(page_to_phys(svm->avic_backing_page) &
 					    AVIC_HPA_MASK);
@@ -951,33 +956,6 @@ int avic_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 			 */
 			if (!ret && pi.is_guest_mode)
 				svm_ir_list_add(svm, &pi);
-		} else {
-			/* Use legacy mode in IRTE */
-			struct amd_iommu_pi_data pi;
-
-			/**
-			 * Here, pi is used to:
-			 * - Tell IOMMU to use legacy mode for this interrupt.
-			 * - Retrieve ga_tag of prior interrupt remapping data.
-			 */
-			pi.prev_ga_tag = 0;
-			pi.is_guest_mode = false;
-			ret = irq_set_vcpu_affinity(host_irq, &pi);
-
-			/**
-			 * Check if the posted interrupt was previously
-			 * setup with the guest_mode by checking if the ga_tag
-			 * was cached. If so, we need to clean up the per-vcpu
-			 * ir_list.
-			 */
-			if (!ret && pi.prev_ga_tag) {
-				int id = AVIC_GATAG_TO_VCPUID(pi.prev_ga_tag);
-				struct kvm_vcpu *vcpu;
-
-				vcpu = kvm_get_vcpu_by_id(kvm, id);
-				if (vcpu)
-					svm_ir_list_del(to_svm(vcpu), &pi);
-			}
 		}
 
 		if (!ret && svm) {
@@ -993,6 +971,34 @@ int avic_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 	}
 
 	ret = 0;
+	if (enable_remapped_mode) {
+		/* Use legacy mode in IRTE */
+		struct amd_iommu_pi_data pi;
+
+		/**
+		 * Here, pi is used to:
+		 * - Tell IOMMU to use legacy mode for this interrupt.
+		 * - Retrieve ga_tag of prior interrupt remapping data.
+		 */
+		pi.prev_ga_tag = 0;
+		pi.is_guest_mode = false;
+		ret = irq_set_vcpu_affinity(host_irq, &pi);
+
+		/**
+		 * Check if the posted interrupt was previously
+		 * setup with the guest_mode by checking if the ga_tag
+		 * was cached. If so, we need to clean up the per-vcpu
+		 * ir_list.
+		 */
+		if (!ret && pi.prev_ga_tag) {
+			int id = AVIC_GATAG_TO_VCPUID(pi.prev_ga_tag);
+			struct kvm_vcpu *vcpu;
+
+			vcpu = kvm_get_vcpu_by_id(kvm, id);
+			if (vcpu)
+				svm_ir_list_del(to_svm(vcpu), &pi);
+		}
+	}
 out:
 	srcu_read_unlock(&kvm->irq_srcu, idx);
 	return ret;
diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
index ccda95e53f62..ba736cbb0587 100644
--- a/arch/x86/kvm/trace.h
+++ b/arch/x86/kvm/trace.h
@@ -11,6 +11,13 @@
 #undef TRACE_SYSTEM
 #define TRACE_SYSTEM kvm
 
+#ifdef CREATE_TRACE_POINTS
+#define tracing_kvm_rip_read(vcpu) ({					\
+	typeof(vcpu) __vcpu = vcpu;					\
+	__vcpu->arch.guest_state_protected ? 0 : kvm_rip_read(__vcpu);	\
+	})
+#endif
+
 /*
  * Tracepoint for guest mode entry.
  */
@@ -28,7 +35,7 @@ TRACE_EVENT(kvm_entry,
 
 	TP_fast_assign(
 		__entry->vcpu_id        = vcpu->vcpu_id;
-		__entry->rip		= kvm_rip_read(vcpu);
+		__entry->rip		= tracing_kvm_rip_read(vcpu);
 		__entry->immediate_exit	= force_immediate_exit;
 
 		kvm_x86_call(get_entry_info)(vcpu, &__entry->intr_info,
@@ -319,7 +326,7 @@ TRACE_EVENT(name,							     \
 	),								     \
 									     \
 	TP_fast_assign(							     \
-		__entry->guest_rip	= kvm_rip_read(vcpu);		     \
+		__entry->guest_rip	= tracing_kvm_rip_read(vcpu);		     \
 		__entry->isa            = isa;				     \
 		__entry->vcpu_id        = vcpu->vcpu_id;		     \
 		__entry->requests       = READ_ONCE(vcpu->requests);	     \
@@ -423,7 +430,7 @@ TRACE_EVENT(kvm_page_fault,
 
 	TP_fast_assign(
 		__entry->vcpu_id	= vcpu->vcpu_id;
-		__entry->guest_rip	= kvm_rip_read(vcpu);
+		__entry->guest_rip	= tracing_kvm_rip_read(vcpu);
 		__entry->fault_address	= fault_address;
 		__entry->error_code	= error_code;
 	),
diff --git a/arch/x86/kvm/vmx/posted_intr.c b/arch/x86/kvm/vmx/posted_intr.c
index ec08fa3caf43..d70e5b90087d 100644
--- a/arch/x86/kvm/vmx/posted_intr.c
+++ b/arch/x86/kvm/vmx/posted_intr.c
@@ -31,6 +31,8 @@ static DEFINE_PER_CPU(struct list_head, wakeup_vcpus_on_cpu);
  */
 static DEFINE_PER_CPU(raw_spinlock_t, wakeup_vcpus_on_cpu_lock);
 
+#define PI_LOCK_SCHED_OUT SINGLE_DEPTH_NESTING
+
 static inline struct pi_desc *vcpu_to_pi_desc(struct kvm_vcpu *vcpu)
 {
 	return &(to_vmx(vcpu)->pi_desc);
@@ -89,9 +91,20 @@ void vmx_vcpu_pi_load(struct kvm_vcpu *vcpu, int cpu)
 	 * current pCPU if the task was migrated.
 	 */
 	if (pi_desc->nv == POSTED_INTR_WAKEUP_VECTOR) {
-		raw_spin_lock(&per_cpu(wakeup_vcpus_on_cpu_lock, vcpu->cpu));
+		raw_spinlock_t *spinlock = &per_cpu(wakeup_vcpus_on_cpu_lock, vcpu->cpu);
+
+		/*
+		 * In addition to taking the wakeup lock for the regular/IRQ
+		 * context, tell lockdep it is being taken for the "sched out"
+		 * context as well.  vCPU loads happens in task context, and
+		 * this is taking the lock of the *previous* CPU, i.e. can race
+		 * with both the scheduler and the wakeup handler.
+		 */
+		raw_spin_lock(spinlock);
+		spin_acquire(&spinlock->dep_map, PI_LOCK_SCHED_OUT, 0, _RET_IP_);
 		list_del(&vmx->pi_wakeup_list);
-		raw_spin_unlock(&per_cpu(wakeup_vcpus_on_cpu_lock, vcpu->cpu));
+		spin_release(&spinlock->dep_map, _RET_IP_);
+		raw_spin_unlock(spinlock);
 	}
 
 	dest = cpu_physical_id(cpu);
@@ -148,11 +161,23 @@ static void pi_enable_wakeup_handler(struct kvm_vcpu *vcpu)
 	struct pi_desc *pi_desc = vcpu_to_pi_desc(vcpu);
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 	struct pi_desc old, new;
-	unsigned long flags;
 
-	local_irq_save(flags);
+	lockdep_assert_irqs_disabled();
 
-	raw_spin_lock(&per_cpu(wakeup_vcpus_on_cpu_lock, vcpu->cpu));
+	/*
+	 * Acquire the wakeup lock using the "sched out" context to workaround
+	 * a lockdep false positive.  When this is called, schedule() holds
+	 * various per-CPU scheduler locks.  When the wakeup handler runs, it
+	 * holds this CPU's wakeup lock while calling try_to_wake_up(), which
+	 * can eventually take the aforementioned scheduler locks, which causes
+	 * lockdep to assume there is deadlock.
+	 *
+	 * Deadlock can't actually occur because IRQs are disabled for the
+	 * entirety of the sched_out critical section, i.e. the wakeup handler
+	 * can't run while the scheduler locks are held.
+	 */
+	raw_spin_lock_nested(&per_cpu(wakeup_vcpus_on_cpu_lock, vcpu->cpu),
+			     PI_LOCK_SCHED_OUT);
 	list_add_tail(&vmx->pi_wakeup_list,
 		      &per_cpu(wakeup_vcpus_on_cpu, vcpu->cpu));
 	raw_spin_unlock(&per_cpu(wakeup_vcpus_on_cpu_lock, vcpu->cpu));
@@ -176,8 +201,6 @@ static void pi_enable_wakeup_handler(struct kvm_vcpu *vcpu)
 	 */
 	if (pi_test_on(&new))
 		__apic_send_IPI_self(POSTED_INTR_WAKEUP_VECTOR);
-
-	local_irq_restore(flags);
 }
 
 static bool vmx_needs_pi_wakeup(struct kvm_vcpu *vcpu)
@@ -274,6 +297,7 @@ int vmx_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 {
 	struct kvm_kernel_irq_routing_entry *e;
 	struct kvm_irq_routing_table *irq_rt;
+	bool enable_remapped_mode = true;
 	struct kvm_lapic_irq irq;
 	struct kvm_vcpu *vcpu;
 	struct vcpu_data vcpu_info;
@@ -312,21 +336,8 @@ int vmx_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 
 		kvm_set_msi_irq(kvm, e, &irq);
 		if (!kvm_intr_is_single_vcpu(kvm, &irq, &vcpu) ||
-		    !kvm_irq_is_postable(&irq)) {
-			/*
-			 * Make sure the IRTE is in remapped mode if
-			 * we don't handle it in posted mode.
-			 */
-			ret = irq_set_vcpu_affinity(host_irq, NULL);
-			if (ret < 0) {
-				printk(KERN_INFO
-				   "failed to back to remapped mode, irq: %u\n",
-				   host_irq);
-				goto out;
-			}
-
+		    !kvm_irq_is_postable(&irq))
 			continue;
-		}
 
 		vcpu_info.pi_desc_addr = __pa(vcpu_to_pi_desc(vcpu));
 		vcpu_info.vector = irq.vector;
@@ -334,11 +345,12 @@ int vmx_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 		trace_kvm_pi_irte_update(host_irq, vcpu->vcpu_id, e->gsi,
 				vcpu_info.vector, vcpu_info.pi_desc_addr, set);
 
-		if (set)
-			ret = irq_set_vcpu_affinity(host_irq, &vcpu_info);
-		else
-			ret = irq_set_vcpu_affinity(host_irq, NULL);
+		if (!set)
+			continue;
+
+		enable_remapped_mode = false;
 
+		ret = irq_set_vcpu_affinity(host_irq, &vcpu_info);
 		if (ret < 0) {
 			printk(KERN_INFO "%s: failed to update PI IRTE\n",
 					__func__);
@@ -346,6 +358,9 @@ int vmx_pi_update_irte(struct kvm *kvm, unsigned int host_irq,
 		}
 	}
 
+	if (enable_remapped_mode)
+		ret = irq_set_vcpu_affinity(host_irq, NULL);
+
 	ret = 0;
 out:
 	srcu_read_unlock(&kvm->irq_srcu, idx);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c841817a914a..df5b99ea1f18 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -11098,7 +11098,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 	/*
 	 * Profile KVM exit RIPs:
 	 */
-	if (unlikely(prof_on == KVM_PROFILING)) {
+	if (unlikely(prof_on == KVM_PROFILING &&
+		     !vcpu->arch.guest_state_protected)) {
 		unsigned long rip = kvm_rip_read(vcpu);
 		profile_hit(KVM_PROFILING, (void *)rip);
 	}
@@ -11786,6 +11787,8 @@ int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 	if (kvm_mpx_supported())
 		kvm_load_guest_fpu(vcpu);
 
+	kvm_vcpu_srcu_read_lock(vcpu);
+
 	r = kvm_apic_accept_events(vcpu);
 	if (r < 0)
 		goto out;
@@ -11799,6 +11802,8 @@ int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 		mp_state->mp_state = vcpu->arch.mp_state;
 
 out:
+	kvm_vcpu_srcu_read_unlock(vcpu);
+
 	if (kvm_mpx_supported())
 		kvm_put_guest_fpu(vcpu);
 	vcpu_put(vcpu);
@@ -13552,25 +13557,27 @@ bool kvm_arch_has_noncoherent_dma(struct kvm *kvm)
 }
 EXPORT_SYMBOL_GPL(kvm_arch_has_noncoherent_dma);
 
-bool kvm_arch_has_irq_bypass(void)
-{
-	return enable_apicv && irq_remapping_cap(IRQ_POSTING_CAP);
-}
-
 int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *cons,
 				      struct irq_bypass_producer *prod)
 {
 	struct kvm_kernel_irqfd *irqfd =
 		container_of(cons, struct kvm_kernel_irqfd, consumer);
+	struct kvm *kvm = irqfd->kvm;
 	int ret;
 
-	irqfd->producer = prod;
 	kvm_arch_start_assignment(irqfd->kvm);
+
+	spin_lock_irq(&kvm->irqfds.lock);
+	irqfd->producer = prod;
+
 	ret = kvm_x86_call(pi_update_irte)(irqfd->kvm,
 					   prod->irq, irqfd->gsi, 1);
 	if (ret)
 		kvm_arch_end_assignment(irqfd->kvm);
 
+	spin_unlock_irq(&kvm->irqfds.lock);
+
+
 	return ret;
 }
 
@@ -13580,9 +13587,9 @@ void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons,
 	int ret;
 	struct kvm_kernel_irqfd *irqfd =
 		container_of(cons, struct kvm_kernel_irqfd, consumer);
+	struct kvm *kvm = irqfd->kvm;
 
 	WARN_ON(irqfd->producer != prod);
-	irqfd->producer = NULL;
 
 	/*
 	 * When producer of consumer is unregistered, we change back to
@@ -13590,12 +13597,18 @@ void kvm_arch_irq_bypass_del_producer(struct irq_bypass_consumer *cons,
 	 * when the irq is masked/disabled or the consumer side (KVM
 	 * int this case doesn't want to receive the interrupts.
 	*/
+	spin_lock_irq(&kvm->irqfds.lock);
+	irqfd->producer = NULL;
+
 	ret = kvm_x86_call(pi_update_irte)(irqfd->kvm,
 					   prod->irq, irqfd->gsi, 0);
 	if (ret)
 		printk(KERN_INFO "irq bypass consumer (token %p) unregistration"
 		       " fails: %d\n", irqfd->consumer.token, ret);
 
+	spin_unlock_irq(&kvm->irqfds.lock);
+
+
 	kvm_arch_end_assignment(irqfd->kvm);
 }
 
@@ -13608,7 +13621,8 @@ int kvm_arch_update_irqfd_routing(struct kvm *kvm, unsigned int host_irq,
 bool kvm_arch_irqfd_route_changed(struct kvm_kernel_irq_routing_entry *old,
 				  struct kvm_kernel_irq_routing_entry *new)
 {
-	if (new->type != KVM_IRQ_ROUTING_MSI)
+	if (old->type != KVM_IRQ_ROUTING_MSI ||
+	    new->type != KVM_IRQ_ROUTING_MSI)
 		return true;
 
 	return !!memcmp(&old->msi, &new->msi, sizeof(new->msi));
diff --git a/arch/x86/lib/x86-opcode-map.txt b/arch/x86/lib/x86-opcode-map.txt
index caedb3ef6688..f5dd84eb55dc 100644
--- a/arch/x86/lib/x86-opcode-map.txt
+++ b/arch/x86/lib/x86-opcode-map.txt
@@ -996,8 +996,8 @@ AVXcode: 4
 83: Grp1 Ev,Ib (1A),(es)
 # CTESTSCC instructions are: CTESTB, CTESTBE, CTESTF, CTESTL, CTESTLE, CTESTNB, CTESTNBE, CTESTNL,
 #			     CTESTNLE, CTESTNO, CTESTNS, CTESTNZ, CTESTO, CTESTS, CTESTT, CTESTZ
-84: CTESTSCC (ev)
-85: CTESTSCC (es) | CTESTSCC (66),(es)
+84: CTESTSCC Eb,Gb (ev)
+85: CTESTSCC Ev,Gv (es) | CTESTSCC Ev,Gv (66),(es)
 88: POPCNT Gv,Ev (es) | POPCNT Gv,Ev (66),(es)
 8f: POP2 Bq,Rq (000),(11B),(ev)
 a5: SHLD Ev,Gv,CL (es) | SHLD Ev,Gv,CL (66),(es)
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index a05fcddfc811..f7ae44d3dd9e 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -360,7 +360,7 @@ static inline pgd_t *_pgd_alloc(struct mm_struct *mm)
 	 * We allocate one page for pgd.
 	 */
 	if (!SHARED_KERNEL_PMD)
-		return __pgd_alloc(mm, PGD_ALLOCATION_ORDER);
+		return __pgd_alloc(mm, pgd_allocation_order());
 
 	/*
 	 * Now PAE kernel is not running as a Xen domain. We can allocate
@@ -380,7 +380,7 @@ static inline void _pgd_free(struct mm_struct *mm, pgd_t *pgd)
 
 static inline pgd_t *_pgd_alloc(struct mm_struct *mm)
 {
-	return __pgd_alloc(mm, PGD_ALLOCATION_ORDER);
+	return __pgd_alloc(mm, pgd_allocation_order());
 }
 
 static inline void _pgd_free(struct mm_struct *mm, pgd_t *pgd)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index e459d97ef397..eb83348f9305 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -667,9 +667,9 @@ static void cond_mitigation(struct task_struct *next)
 	prev_mm = this_cpu_read(cpu_tlbstate.last_user_mm_spec);
 
 	/*
-	 * Avoid user/user BTB poisoning by flushing the branch predictor
-	 * when switching between processes. This stops one process from
-	 * doing Spectre-v2 attacks on another.
+	 * Avoid user->user BTB/RSB poisoning by flushing them when switching
+	 * between processes. This stops one process from doing Spectre-v2
+	 * attacks on another.
 	 *
 	 * Both, the conditional and the always IBPB mode use the mm
 	 * pointer to avoid the IBPB when switching between tasks of the
diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c
index ac57259a432b..a4b4ebd41b8f 100644
--- a/arch/x86/platform/efi/efi_64.c
+++ b/arch/x86/platform/efi/efi_64.c
@@ -73,7 +73,7 @@ int __init efi_alloc_page_tables(void)
 	gfp_t gfp_mask;
 
 	gfp_mask = GFP_KERNEL | __GFP_ZERO;
-	efi_pgd = (pgd_t *)__get_free_pages(gfp_mask, PGD_ALLOCATION_ORDER);
+	efi_pgd = (pgd_t *)__get_free_pages(gfp_mask, pgd_allocation_order());
 	if (!efi_pgd)
 		goto fail;
 
@@ -96,7 +96,7 @@ free_p4d:
 	if (pgtable_l5_enabled())
 		free_page((unsigned long)pgd_page_vaddr(*pgd));
 free_pgd:
-	free_pages((unsigned long)efi_pgd, PGD_ALLOCATION_ORDER);
+	free_pages((unsigned long)efi_pgd, pgd_allocation_order());
 fail:
 	return -ENOMEM;
 }
diff --git a/arch/x86/power/hibernate_asm_64.S b/arch/x86/power/hibernate_asm_64.S
index 8c534c36adfa..66f066b8feda 100644
--- a/arch/x86/power/hibernate_asm_64.S
+++ b/arch/x86/power/hibernate_asm_64.S
@@ -26,7 +26,7 @@
 	 /* code below belongs to the image kernel */
 	.align PAGE_SIZE
 SYM_FUNC_START(restore_registers)
-	ANNOTATE_NOENDBR
+	ENDBR
 	/* go back to the original page tables */
 	movq    %r9, %cr3
 
@@ -120,7 +120,7 @@ SYM_FUNC_END(restore_image)
 
 	/* code below has been relocated to a safe page */
 SYM_FUNC_START(core_restore_code)
-	ANNOTATE_NOENDBR
+	ENDBR
 	/* switch to temporary page tables */
 	movq	%rax, %cr3
 	/* flush TLB */
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 43dcd8c7badc..53282dc7d5ac 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -70,6 +70,9 @@ EXPORT_SYMBOL(xen_start_flags);
  */
 struct shared_info *HYPERVISOR_shared_info = &xen_dummy_shared_info;
 
+/* Number of pages released from the initial allocation. */
+unsigned long xen_released_pages;
+
 static __ref void xen_get_vendor(void)
 {
 	init_cpu_devs();
@@ -100,10 +103,6 @@ noinstr void *__xen_hypercall_setfunc(void)
 	void (*func)(void);
 
 	/*
-	 * Xen is supported only on CPUs with CPUID, so testing for
-	 * X86_FEATURE_CPUID is a test for early_cpu_init() having been
-	 * run.
-	 *
 	 * Note that __xen_hypercall_setfunc() is noinstr only due to a nasty
 	 * dependency chain: it is being called via the xen_hypercall static
 	 * call when running as a PVH or HVM guest. Hypercalls need to be
@@ -115,8 +114,7 @@ noinstr void *__xen_hypercall_setfunc(void)
 	 */
 	instrumentation_begin();
 
-	if (!boot_cpu_has(X86_FEATURE_CPUID))
-		xen_get_vendor();
+	xen_get_vendor();
 
 	if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD ||
 	     boot_cpu_data.x86_vendor == X86_VENDOR_HYGON))
@@ -466,6 +464,13 @@ int __init arch_xen_unpopulated_init(struct resource **res)
 			xen_free_unpopulated_pages(1, &pg);
 		}
 
+		/*
+		 * Account for the region being in the physmap but unpopulated.
+		 * The value in xen_released_pages is used by the balloon
+		 * driver to know how much of the physmap is unpopulated and
+		 * set an accurate initial memory target.
+		 */
+		xen_released_pages += xen_extra_mem[i].n_pfns;
 		/* Zero so region is not also added to the balloon driver. */
 		xen_extra_mem[i].n_pfns = 0;
 	}
diff --git a/arch/x86/xen/enlighten_pvh.c b/arch/x86/xen/enlighten_pvh.c
index 0e3d930bcb89..9d25d9373945 100644
--- a/arch/x86/xen/enlighten_pvh.c
+++ b/arch/x86/xen/enlighten_pvh.c
@@ -1,5 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/acpi.h>
+#include <linux/cpufreq.h>
+#include <linux/cpuidle.h>
 #include <linux/export.h>
 #include <linux/mm.h>
 
@@ -123,8 +125,23 @@ static void __init pvh_arch_setup(void)
 {
 	pvh_reserve_extra_memory();
 
-	if (xen_initial_domain())
+	if (xen_initial_domain()) {
 		xen_add_preferred_consoles();
+
+		/*
+		 * Disable usage of CPU idle and frequency drivers: when
+		 * running as hardware domain the exposed native ACPI tables
+		 * causes idle and/or frequency drivers to attach and
+		 * malfunction.  It's Xen the entity that controls the idle and
+		 * frequency states.
+		 *
+		 * For unprivileged domains the exposed ACPI tables are
+		 * fabricated and don't contain such data.
+		 */
+		disable_cpuidle();
+		disable_cpufreq();
+		WARN_ON(xen_set_default_idle());
+	}
 }
 
 void __init xen_pvh_init(struct boot_params *boot_params)
diff --git a/arch/x86/xen/multicalls.c b/arch/x86/xen/multicalls.c
index 10c660fae8b3..7237d56a9d3f 100644
--- a/arch/x86/xen/multicalls.c
+++ b/arch/x86/xen/multicalls.c
@@ -54,14 +54,20 @@ struct mc_debug_data {
 
 static DEFINE_PER_CPU(struct mc_buffer, mc_buffer);
 static struct mc_debug_data mc_debug_data_early __initdata;
-static DEFINE_PER_CPU(struct mc_debug_data *, mc_debug_data) =
-	&mc_debug_data_early;
 static struct mc_debug_data __percpu *mc_debug_data_ptr;
 DEFINE_PER_CPU(unsigned long, xen_mc_irq_flags);
 
 static struct static_key mc_debug __ro_after_init;
 static bool mc_debug_enabled __initdata;
 
+static struct mc_debug_data * __ref get_mc_debug(void)
+{
+	if (!mc_debug_data_ptr)
+		return &mc_debug_data_early;
+
+	return this_cpu_ptr(mc_debug_data_ptr);
+}
+
 static int __init xen_parse_mc_debug(char *arg)
 {
 	mc_debug_enabled = true;
@@ -71,20 +77,16 @@ static int __init xen_parse_mc_debug(char *arg)
 }
 early_param("xen_mc_debug", xen_parse_mc_debug);
 
-void mc_percpu_init(unsigned int cpu)
-{
-	per_cpu(mc_debug_data, cpu) = per_cpu_ptr(mc_debug_data_ptr, cpu);
-}
-
 static int __init mc_debug_enable(void)
 {
 	unsigned long flags;
+	struct mc_debug_data __percpu *mcdb;
 
 	if (!mc_debug_enabled)
 		return 0;
 
-	mc_debug_data_ptr = alloc_percpu(struct mc_debug_data);
-	if (!mc_debug_data_ptr) {
+	mcdb = alloc_percpu(struct mc_debug_data);
+	if (!mcdb) {
 		pr_err("xen_mc_debug inactive\n");
 		static_key_slow_dec(&mc_debug);
 		return -ENOMEM;
@@ -93,7 +95,7 @@ static int __init mc_debug_enable(void)
 	/* Be careful when switching to percpu debug data. */
 	local_irq_save(flags);
 	xen_mc_flush();
-	mc_percpu_init(0);
+	mc_debug_data_ptr = mcdb;
 	local_irq_restore(flags);
 
 	pr_info("xen_mc_debug active\n");
@@ -155,7 +157,7 @@ void xen_mc_flush(void)
 	trace_xen_mc_flush(b->mcidx, b->argidx, b->cbidx);
 
 	if (static_key_false(&mc_debug)) {
-		mcdb = __this_cpu_read(mc_debug_data);
+		mcdb = get_mc_debug();
 		memcpy(mcdb->entries, b->entries,
 		       b->mcidx * sizeof(struct multicall_entry));
 	}
@@ -235,7 +237,7 @@ struct multicall_space __xen_mc_entry(size_t args)
 
 	ret.mc = &b->entries[b->mcidx];
 	if (static_key_false(&mc_debug)) {
-		struct mc_debug_data *mcdb = __this_cpu_read(mc_debug_data);
+		struct mc_debug_data *mcdb = get_mc_debug();
 
 		mcdb->caller[b->mcidx] = __builtin_return_address(0);
 		mcdb->argsz[b->mcidx] = args;
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index c3db71d96c43..3823e52aef52 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -37,9 +37,6 @@
 
 #define GB(x) ((uint64_t)(x) * 1024 * 1024 * 1024)
 
-/* Number of pages released from the initial allocation. */
-unsigned long xen_released_pages;
-
 /* Memory map would allow PCI passthrough. */
 bool xen_pv_pci_possible;
 
diff --git a/arch/x86/xen/smp_pv.c b/arch/x86/xen/smp_pv.c
index 688ff59318ae..9bb8ff8bff30 100644
--- a/arch/x86/xen/smp_pv.c
+++ b/arch/x86/xen/smp_pv.c
@@ -305,7 +305,6 @@ static int xen_pv_kick_ap(unsigned int cpu, struct task_struct *idle)
 		return rc;
 
 	xen_pmu_init(cpu);
-	mc_percpu_init(cpu);
 
 	/*
 	 * Why is this a BUG? If the hypercall fails then everything can be
diff --git a/arch/x86/xen/xen-asm.S b/arch/x86/xen/xen-asm.S
index 109af12f7647..461bb1526502 100644
--- a/arch/x86/xen/xen-asm.S
+++ b/arch/x86/xen/xen-asm.S
@@ -226,9 +226,7 @@ SYM_CODE_END(xen_early_idt_handler_array)
 	push %rax
 	mov  $__HYPERVISOR_iret, %eax
 	syscall		/* Do the IRET. */
-#ifdef CONFIG_MITIGATION_SLS
-	int3
-#endif
+	ud2		/* The SYSCALL should never return. */
 .endm
 
 SYM_CODE_START(xen_iret)
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index 63c13a2ccf55..25e318ef27d6 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -261,9 +261,6 @@ void xen_mc_callback(void (*fn)(void *), void *data);
  */
 struct multicall_space xen_mc_extend_args(unsigned long op, size_t arg_size);
 
-/* Do percpu data initialization for multicalls. */
-void mc_percpu_init(unsigned int cpu);
-
 extern bool is_xen_pmu;
 
 irqreturn_t xen_pmu_irq_handler(int irq, void *dev_id);
diff --git a/block/bdev.c b/block/bdev.c
index 4844d1e27b6f..889ec6e002d7 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -152,27 +152,65 @@ static void set_init_blocksize(struct block_device *bdev)
 				    get_order(bsize));
 }
 
-int set_blocksize(struct file *file, int size)
+/**
+ * bdev_validate_blocksize - check that this block size is acceptable
+ * @bdev:	blockdevice to check
+ * @block_size:	block size to check
+ *
+ * For block device users that do not use buffer heads or the block device
+ * page cache, make sure that this block size can be used with the device.
+ *
+ * Return: On success zero is returned, negative error code on failure.
+ */
+int bdev_validate_blocksize(struct block_device *bdev, int block_size)
 {
-	struct inode *inode = file->f_mapping->host;
-	struct block_device *bdev = I_BDEV(inode);
-
-	if (blk_validate_block_size(size))
+	if (blk_validate_block_size(block_size))
 		return -EINVAL;
 
 	/* Size cannot be smaller than the size supported by the device */
-	if (size < bdev_logical_block_size(bdev))
+	if (block_size < bdev_logical_block_size(bdev))
 		return -EINVAL;
 
+	return 0;
+}
+EXPORT_SYMBOL_GPL(bdev_validate_blocksize);
+
+int set_blocksize(struct file *file, int size)
+{
+	struct inode *inode = file->f_mapping->host;
+	struct block_device *bdev = I_BDEV(inode);
+	int ret;
+
+	ret = bdev_validate_blocksize(bdev, size);
+	if (ret)
+		return ret;
+
 	if (!file->private_data)
 		return -EINVAL;
 
 	/* Don't change the size if it is same as current */
 	if (inode->i_blkbits != blksize_bits(size)) {
+		/*
+		 * Flush and truncate the pagecache before we reconfigure the
+		 * mapping geometry because folio sizes are variable now.  If a
+		 * reader has already allocated a folio whose size is smaller
+		 * than the new min_order but invokes readahead after the new
+		 * min_order becomes visible, readahead will think there are
+		 * "zero" blocks per folio and crash.  Take the inode and
+		 * invalidation locks to avoid racing with
+		 * read/write/fallocate.
+		 */
+		inode_lock(inode);
+		filemap_invalidate_lock(inode->i_mapping);
+
 		sync_blockdev(bdev);
+		kill_bdev(bdev);
+
 		inode->i_blkbits = blksize_bits(size);
 		mapping_set_folio_min_order(inode->i_mapping, get_order(size));
 		kill_bdev(bdev);
+		filemap_invalidate_unlock(inode->i_mapping);
+		inode_unlock(inode);
 	}
 	return 0;
 }
@@ -777,13 +815,13 @@ static void blkdev_put_part(struct block_device *part)
 	blkdev_put_whole(whole);
 }
 
-struct block_device *blkdev_get_no_open(dev_t dev)
+struct block_device *blkdev_get_no_open(dev_t dev, bool autoload)
 {
 	struct block_device *bdev;
 	struct inode *inode;
 
 	inode = ilookup(blockdev_superblock, dev);
-	if (!inode && IS_ENABLED(CONFIG_BLOCK_LEGACY_AUTOLOAD)) {
+	if (!inode && autoload && IS_ENABLED(CONFIG_BLOCK_LEGACY_AUTOLOAD)) {
 		blk_request_module(dev);
 		inode = ilookup(blockdev_superblock, dev);
 		if (inode)
@@ -1005,7 +1043,7 @@ struct file *bdev_file_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 	if (ret)
 		return ERR_PTR(ret);
 
-	bdev = blkdev_get_no_open(dev);
+	bdev = blkdev_get_no_open(dev, true);
 	if (!bdev)
 		return ERR_PTR(-ENXIO);
 
@@ -1272,21 +1310,17 @@ void sync_bdevs(bool wait)
 /*
  * Handle STATX_{DIOALIGN, WRITE_ATOMIC} for block devices.
  */
-void bdev_statx(struct path *path, struct kstat *stat,
-		u32 request_mask)
+void bdev_statx(const struct path *path, struct kstat *stat, u32 request_mask)
 {
-	struct inode *backing_inode;
 	struct block_device *bdev;
 
-	backing_inode = d_backing_inode(path->dentry);
-
 	/*
-	 * Note that backing_inode is the inode of a block device node file,
-	 * not the block device's internal inode.  Therefore it is *not* valid
-	 * to use I_BDEV() here; the block device has to be looked up by i_rdev
+	 * Note that d_backing_inode() returns the block device node inode, not
+	 * the block device's internal inode.  Therefore it is *not* valid to
+	 * use I_BDEV() here; the block device has to be looked up by i_rdev
 	 * instead.
 	 */
-	bdev = blkdev_get_no_open(backing_inode->i_rdev);
+	bdev = blkdev_get_no_open(d_backing_inode(path->dentry)->i_rdev, false);
 	if (!bdev)
 		return;
 
diff --git a/block/bio-integrity.c b/block/bio-integrity.c
index 608594a154a5..43ef6bd06c85 100644
--- a/block/bio-integrity.c
+++ b/block/bio-integrity.c
@@ -66,16 +66,12 @@ struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio,
 }
 EXPORT_SYMBOL(bio_integrity_alloc);
 
-static void bio_integrity_unpin_bvec(struct bio_vec *bv, int nr_vecs,
-				     bool dirty)
+static void bio_integrity_unpin_bvec(struct bio_vec *bv, int nr_vecs)
 {
 	int i;
 
-	for (i = 0; i < nr_vecs; i++) {
-		if (dirty && !PageCompound(bv[i].bv_page))
-			set_page_dirty_lock(bv[i].bv_page);
+	for (i = 0; i < nr_vecs; i++)
 		unpin_user_page(bv[i].bv_page);
-	}
 }
 
 static void bio_integrity_uncopy_user(struct bio_integrity_payload *bip)
@@ -91,7 +87,7 @@ static void bio_integrity_uncopy_user(struct bio_integrity_payload *bip)
 	ret = copy_to_iter(bvec_virt(bounce_bvec), bytes, &orig_iter);
 	WARN_ON_ONCE(ret != bytes);
 
-	bio_integrity_unpin_bvec(orig_bvecs, orig_nr_vecs, true);
+	bio_integrity_unpin_bvec(orig_bvecs, orig_nr_vecs);
 }
 
 /**
@@ -111,8 +107,7 @@ void bio_integrity_unmap_user(struct bio *bio)
 		return;
 	}
 
-	bio_integrity_unpin_bvec(bip->bip_vec, bip->bip_max_vcnt,
-			bio_data_dir(bio) == READ);
+	bio_integrity_unpin_bvec(bip->bip_vec, bip->bip_max_vcnt);
 }
 
 /**
@@ -198,7 +193,7 @@ static int bio_integrity_copy_user(struct bio *bio, struct bio_vec *bvec,
 	}
 
 	if (write)
-		bio_integrity_unpin_bvec(bvec, nr_vecs, false);
+		bio_integrity_unpin_bvec(bvec, nr_vecs);
 	else
 		memcpy(&bip->bip_vec[1], bvec, nr_vecs * sizeof(*bvec));
 
@@ -319,7 +314,7 @@ int bio_integrity_map_user(struct bio *bio, struct iov_iter *iter)
 	return 0;
 
 release_pages:
-	bio_integrity_unpin_bvec(bvec, nr_bvecs, false);
+	bio_integrity_unpin_bvec(bvec, nr_bvecs);
 free_bvec:
 	if (bvec != stack_vec)
 		kfree(bvec);
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 5905f277057b..ce93706555c5 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -797,7 +797,7 @@ int blkg_conf_open_bdev(struct blkg_conf_ctx *ctx)
 		return -EINVAL;
 	input = skip_spaces(input);
 
-	bdev = blkdev_get_no_open(MKDEV(major, minor));
+	bdev = blkdev_get_no_open(MKDEV(major, minor), false);
 	if (!bdev)
 		return -ENODEV;
 	if (bdev_is_partition(bdev)) {
diff --git a/block/blk-settings.c b/block/blk-settings.c
index 6b2dbe645d23..4817e7ca03f8 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -61,8 +61,14 @@ void blk_apply_bdi_limits(struct backing_dev_info *bdi,
 	/*
 	 * For read-ahead of large files to be effective, we need to read ahead
 	 * at least twice the optimal I/O size.
+	 *
+	 * There is no hardware limitation for the read-ahead size and the user
+	 * might have increased the read-ahead size through sysfs, so don't ever
+	 * decrease it.
 	 */
-	bdi->ra_pages = max(lim->io_opt * 2 / PAGE_SIZE, VM_READAHEAD_PAGES);
+	bdi->ra_pages = max3(bdi->ra_pages,
+				lim->io_opt * 2 / PAGE_SIZE,
+				VM_READAHEAD_PAGES);
 	bdi->io_pages = lim->max_sectors >> PAGE_SECTORS_SHIFT;
 }
 
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index a2882751f0d2..1f9b45b0b9ee 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -909,6 +909,8 @@ out_unregister_ia_ranges:
 out_debugfs_remove:
 	blk_debugfs_remove(disk);
 	mutex_unlock(&q->sysfs_lock);
+	if (queue_is_mq(q))
+		blk_mq_sysfs_unregister(disk);
 out_put_queue_kobj:
 	kobject_put(&disk->queue_kobj);
 	return ret;
diff --git a/block/blk-throttle.h b/block/blk-throttle.h
index 7964cc041e06..f9f8666891ab 100644
--- a/block/blk-throttle.h
+++ b/block/blk-throttle.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef BLK_THROTTLE_H
 #define BLK_THROTTLE_H
 
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 0c77244a35c9..8f15d1aa6eb8 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -343,6 +343,7 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode,
 		op = REQ_OP_ZONE_RESET;
 
 		/* Invalidate the page cache, including dirty pages. */
+		inode_lock(bdev->bd_mapping->host);
 		filemap_invalidate_lock(bdev->bd_mapping);
 		ret = blkdev_truncate_zone_range(bdev, mode, &zrange);
 		if (ret)
@@ -364,8 +365,10 @@ int blkdev_zone_mgmt_ioctl(struct block_device *bdev, blk_mode_t mode,
 	ret = blkdev_zone_mgmt(bdev, op, zrange.sector, zrange.nr_sectors);
 
 fail:
-	if (cmd == BLKRESETZONE)
+	if (cmd == BLKRESETZONE) {
 		filemap_invalidate_unlock(bdev->bd_mapping);
+		inode_unlock(bdev->bd_mapping->host);
+	}
 
 	return ret;
 }
diff --git a/block/blk.h b/block/blk.h
index 006e3be433d2..328075787814 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -94,6 +94,9 @@ static inline void blk_wait_io(struct completion *done)
 		wait_for_completion_io(done);
 }
 
+struct block_device *blkdev_get_no_open(dev_t dev, bool autoload);
+void blkdev_put_no_open(struct block_device *bdev);
+
 #define BIO_INLINE_VECS 4
 struct bio_vec *bvec_alloc(mempool_t *pool, unsigned short *nr_vecs,
 		gfp_t gfp_mask);
diff --git a/block/fops.c b/block/fops.c
index be9f1dbea9ce..82b672d15ea4 100644
--- a/block/fops.c
+++ b/block/fops.c
@@ -642,7 +642,7 @@ static int blkdev_open(struct inode *inode, struct file *filp)
 	if (ret)
 		return ret;
 
-	bdev = blkdev_get_no_open(inode->i_rdev);
+	bdev = blkdev_get_no_open(inode->i_rdev, true);
 	if (!bdev)
 		return -ENXIO;
 
@@ -746,7 +746,14 @@ static ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from)
 			ret = direct_write_fallback(iocb, from, ret,
 					blkdev_buffered_write(iocb, from));
 	} else {
+		/*
+		 * Take i_rwsem and invalidate_lock to avoid racing with
+		 * set_blocksize changing i_blkbits/folio order and punching
+		 * out the pagecache.
+		 */
+		inode_lock_shared(bd_inode);
 		ret = blkdev_buffered_write(iocb, from);
+		inode_unlock_shared(bd_inode);
 	}
 
 	if (ret > 0)
@@ -757,6 +764,7 @@ static ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from)
 
 static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
+	struct inode *bd_inode = bdev_file_inode(iocb->ki_filp);
 	struct block_device *bdev = I_BDEV(iocb->ki_filp->f_mapping->host);
 	loff_t size = bdev_nr_bytes(bdev);
 	loff_t pos = iocb->ki_pos;
@@ -793,7 +801,13 @@ static ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to)
 			goto reexpand;
 	}
 
+	/*
+	 * Take i_rwsem and invalidate_lock to avoid racing with set_blocksize
+	 * changing i_blkbits/folio order and punching out the pagecache.
+	 */
+	inode_lock_shared(bd_inode);
 	ret = filemap_read(iocb, to, ret);
+	inode_unlock_shared(bd_inode);
 
 reexpand:
 	if (unlikely(shorted))
@@ -836,6 +850,7 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start,
 	if ((start | len) & (bdev_logical_block_size(bdev) - 1))
 		return -EINVAL;
 
+	inode_lock(inode);
 	filemap_invalidate_lock(inode->i_mapping);
 
 	/*
@@ -868,6 +883,7 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start,
 
  fail:
 	filemap_invalidate_unlock(inode->i_mapping);
+	inode_unlock(inode);
 	return error;
 }
 
diff --git a/block/ioctl.c b/block/ioctl.c
index faa40f383e27..e472cc1030c6 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -142,6 +142,7 @@ static int blk_ioctl_discard(struct block_device *bdev, blk_mode_t mode,
 	if (err)
 		return err;
 
+	inode_lock(bdev->bd_mapping->host);
 	filemap_invalidate_lock(bdev->bd_mapping);
 	err = truncate_bdev_range(bdev, mode, start, start + len - 1);
 	if (err)
@@ -174,6 +175,7 @@ out_unplug:
 	blk_finish_plug(&plug);
 fail:
 	filemap_invalidate_unlock(bdev->bd_mapping);
+	inode_unlock(bdev->bd_mapping->host);
 	return err;
 }
 
@@ -199,12 +201,14 @@ static int blk_ioctl_secure_erase(struct block_device *bdev, blk_mode_t mode,
 	    end > bdev_nr_bytes(bdev))
 		return -EINVAL;
 
+	inode_lock(bdev->bd_mapping->host);
 	filemap_invalidate_lock(bdev->bd_mapping);
 	err = truncate_bdev_range(bdev, mode, start, end - 1);
 	if (!err)
 		err = blkdev_issue_secure_erase(bdev, start >> 9, len >> 9,
 						GFP_KERNEL);
 	filemap_invalidate_unlock(bdev->bd_mapping);
+	inode_unlock(bdev->bd_mapping->host);
 	return err;
 }
 
@@ -236,6 +240,7 @@ static int blk_ioctl_zeroout(struct block_device *bdev, blk_mode_t mode,
 		return -EINVAL;
 
 	/* Invalidate the page cache, including dirty pages */
+	inode_lock(bdev->bd_mapping->host);
 	filemap_invalidate_lock(bdev->bd_mapping);
 	err = truncate_bdev_range(bdev, mode, start, end);
 	if (err)
@@ -246,6 +251,7 @@ static int blk_ioctl_zeroout(struct block_device *bdev, blk_mode_t mode,
 
 fail:
 	filemap_invalidate_unlock(bdev->bd_mapping);
+	inode_unlock(bdev->bd_mapping->host);
 	return err;
 }
 
diff --git a/crypto/ahash.c b/crypto/ahash.c
index 9f57b925b116..2d9eec2b2b1c 100644
--- a/crypto/ahash.c
+++ b/crypto/ahash.c
@@ -315,16 +315,7 @@ EXPORT_SYMBOL_GPL(crypto_ahash_setkey);
 
 static bool ahash_request_hasvirt(struct ahash_request *req)
 {
-	struct ahash_request *r2;
-
-	if (ahash_request_isvirt(req))
-		return true;
-
-	list_for_each_entry(r2, &req->base.list, base.list)
-		if (ahash_request_isvirt(r2))
-			return true;
-
-	return false;
+	return ahash_request_isvirt(req);
 }
 
 static int ahash_reqchain_virt(struct ahash_save_req_state *state,
@@ -472,7 +463,6 @@ static int ahash_do_req_chain(struct ahash_request *req,
 	bool update = op == crypto_ahash_alg(tfm)->update;
 	struct ahash_save_req_state *state;
 	struct ahash_save_req_state state0;
-	struct ahash_request *r2;
 	u8 *page = NULL;
 	int err;
 
@@ -509,7 +499,6 @@ static int ahash_do_req_chain(struct ahash_request *req,
 	state->offset = 0;
 	state->nbytes = 0;
 	INIT_LIST_HEAD(&state->head);
-	list_splice_init(&req->base.list, &state->head);
 
 	if (page)
 		sg_init_one(&state->sg, page, PAGE_SIZE);
@@ -540,9 +529,6 @@ out_free_page:
 
 out_set_chain:
 	req->base.err = err;
-	list_for_each_entry(r2, &req->base.list, base.list)
-		r2->base.err = err;
-
 	return err;
 }
 
@@ -551,19 +537,10 @@ int crypto_ahash_init(struct ahash_request *req)
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
 
 	if (likely(tfm->using_shash)) {
-		struct ahash_request *r2;
 		int err;
 
 		err = crypto_shash_init(prepare_shash_desc(req, tfm));
 		req->base.err = err;
-
-		list_for_each_entry(r2, &req->base.list, base.list) {
-			struct shash_desc *desc;
-
-			desc = prepare_shash_desc(r2, tfm);
-			r2->base.err = crypto_shash_init(desc);
-		}
-
 		return err;
 	}
 
@@ -620,19 +597,10 @@ int crypto_ahash_update(struct ahash_request *req)
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
 
 	if (likely(tfm->using_shash)) {
-		struct ahash_request *r2;
 		int err;
 
 		err = shash_ahash_update(req, ahash_request_ctx(req));
 		req->base.err = err;
-
-		list_for_each_entry(r2, &req->base.list, base.list) {
-			struct shash_desc *desc;
-
-			desc = ahash_request_ctx(r2);
-			r2->base.err = shash_ahash_update(r2, desc);
-		}
-
 		return err;
 	}
 
@@ -645,19 +613,10 @@ int crypto_ahash_final(struct ahash_request *req)
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
 
 	if (likely(tfm->using_shash)) {
-		struct ahash_request *r2;
 		int err;
 
 		err = crypto_shash_final(ahash_request_ctx(req), req->result);
 		req->base.err = err;
-
-		list_for_each_entry(r2, &req->base.list, base.list) {
-			struct shash_desc *desc;
-
-			desc = ahash_request_ctx(r2);
-			r2->base.err = crypto_shash_final(desc, r2->result);
-		}
-
 		return err;
 	}
 
@@ -670,19 +629,10 @@ int crypto_ahash_finup(struct ahash_request *req)
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
 
 	if (likely(tfm->using_shash)) {
-		struct ahash_request *r2;
 		int err;
 
 		err = shash_ahash_finup(req, ahash_request_ctx(req));
 		req->base.err = err;
-
-		list_for_each_entry(r2, &req->base.list, base.list) {
-			struct shash_desc *desc;
-
-			desc = ahash_request_ctx(r2);
-			r2->base.err = shash_ahash_finup(r2, desc);
-		}
-
 		return err;
 	}
 
@@ -757,19 +707,10 @@ int crypto_ahash_digest(struct ahash_request *req)
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
 
 	if (likely(tfm->using_shash)) {
-		struct ahash_request *r2;
 		int err;
 
 		err = shash_ahash_digest(req, prepare_shash_desc(req, tfm));
 		req->base.err = err;
-
-		list_for_each_entry(r2, &req->base.list, base.list) {
-			struct shash_desc *desc;
-
-			desc = prepare_shash_desc(r2, tfm);
-			r2->base.err = shash_ahash_digest(r2, desc);
-		}
-
 		return err;
 	}
 
@@ -1133,20 +1074,5 @@ int ahash_register_instance(struct crypto_template *tmpl,
 }
 EXPORT_SYMBOL_GPL(ahash_register_instance);
 
-void ahash_request_free(struct ahash_request *req)
-{
-	struct ahash_request *tmp;
-	struct ahash_request *r2;
-
-	if (unlikely(!req))
-		return;
-
-	list_for_each_entry_safe(r2, tmp, &req->base.list, base.list)
-		kfree_sensitive(r2);
-
-	kfree_sensitive(req);
-}
-EXPORT_SYMBOL_GPL(ahash_request_free);
-
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Asynchronous cryptographic hash type");
diff --git a/crypto/scompress.c b/crypto/scompress.c
index d435d4b24469..ffeedcf20b0f 100644
--- a/crypto/scompress.c
+++ b/crypto/scompress.c
@@ -111,10 +111,14 @@ static void scomp_free_streams(struct scomp_alg *alg)
 	struct crypto_acomp_stream __percpu *stream = alg->stream;
 	int i;
 
+	alg->stream = NULL;
+	if (!stream)
+		return;
+
 	for_each_possible_cpu(i) {
 		struct crypto_acomp_stream *ps = per_cpu_ptr(stream, i);
 
-		if (!ps->ctx)
+		if (IS_ERR_OR_NULL(ps->ctx))
 			break;
 
 		alg->free_ctx(ps->ctx);
@@ -132,6 +136,8 @@ static int scomp_alloc_streams(struct scomp_alg *alg)
 	if (!stream)
 		return -ENOMEM;
 
+	alg->stream = stream;
+
 	for_each_possible_cpu(i) {
 		struct crypto_acomp_stream *ps = per_cpu_ptr(stream, i);
 
@@ -143,8 +149,6 @@ static int scomp_alloc_streams(struct scomp_alg *alg)
 
 		spin_lock_init(&ps->lock);
 	}
-
-	alg->stream = stream;
 	return 0;
 }
 
@@ -159,11 +163,10 @@ static int crypto_scomp_init_tfm(struct crypto_tfm *tfm)
 		if (ret)
 			goto unlock;
 	}
-	if (!scomp_scratch_users) {
+	if (!scomp_scratch_users++) {
 		ret = crypto_scomp_alloc_scratches();
 		if (ret)
-			goto unlock;
-		scomp_scratch_users++;
+			scomp_scratch_users--;
 	}
 unlock:
 	mutex_unlock(&scomp_lock);
@@ -211,8 +214,8 @@ static int scomp_acomp_comp_decomp(struct acomp_req *req, int dir)
 			spage = nth_page(spage, soff / PAGE_SIZE);
 			soff = offset_in_page(soff);
 
-			n = slen / PAGE_SIZE;
-			n += (offset_in_page(slen) + soff - 1) / PAGE_SIZE;
+			n = (slen - 1) / PAGE_SIZE;
+			n += (offset_in_page(slen - 1) + soff) / PAGE_SIZE;
 			if (PageHighMem(nth_page(spage, n)) &&
 			    size_add(soff, slen) > PAGE_SIZE)
 				break;
@@ -239,9 +242,9 @@ static int scomp_acomp_comp_decomp(struct acomp_req *req, int dir)
 			dpage = nth_page(dpage, doff / PAGE_SIZE);
 			doff = offset_in_page(doff);
 
-			n = dlen / PAGE_SIZE;
-			n += (offset_in_page(dlen) + doff - 1) / PAGE_SIZE;
-			if (PageHighMem(dpage + n) &&
+			n = (dlen - 1) / PAGE_SIZE;
+			n += (offset_in_page(dlen - 1) + doff) / PAGE_SIZE;
+			if (PageHighMem(nth_page(dpage, n)) &&
 			    size_add(doff, dlen) > PAGE_SIZE)
 				break;
 			dst = kmap_local_page(dpage) + doff;
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index abd609d4c8ef..82977ea25db3 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -58,9 +58,6 @@ module_param(fuzz_iterations, uint, 0644);
 MODULE_PARM_DESC(fuzz_iterations, "number of fuzz test iterations");
 #endif
 
-/* Multibuffer is unlimited.  Set arbitrary limit for testing. */
-#define MAX_MB_MSGS	16
-
 #ifdef CONFIG_CRYPTO_MANAGER_DISABLE_TESTS
 
 /* a perfect nop */
@@ -3329,48 +3326,27 @@ static int test_acomp(struct crypto_acomp *tfm,
 		      int ctcount, int dtcount)
 {
 	const char *algo = crypto_tfm_alg_driver_name(crypto_acomp_tfm(tfm));
-	struct scatterlist *src = NULL, *dst = NULL;
-	struct acomp_req *reqs[MAX_MB_MSGS] = {};
-	char *decomp_out[MAX_MB_MSGS] = {};
-	char *output[MAX_MB_MSGS] = {};
-	struct crypto_wait wait;
-	struct acomp_req *req;
-	int ret = -ENOMEM;
 	unsigned int i;
+	char *output, *decomp_out;
+	int ret;
+	struct scatterlist src, dst;
+	struct acomp_req *req;
+	struct crypto_wait wait;
 
-	src = kmalloc_array(MAX_MB_MSGS, sizeof(*src), GFP_KERNEL);
-	if (!src)
-		goto out;
-	dst = kmalloc_array(MAX_MB_MSGS, sizeof(*dst), GFP_KERNEL);
-	if (!dst)
-		goto out;
-
-	for (i = 0; i < MAX_MB_MSGS; i++) {
-		reqs[i] = acomp_request_alloc(tfm);
-		if (!reqs[i])
-			goto out;
-
-		acomp_request_set_callback(reqs[i],
-					   CRYPTO_TFM_REQ_MAY_SLEEP |
-					   CRYPTO_TFM_REQ_MAY_BACKLOG,
-					   crypto_req_done, &wait);
-		if (i)
-			acomp_request_chain(reqs[i], reqs[0]);
-
-		output[i] = kmalloc(COMP_BUF_SIZE, GFP_KERNEL);
-		if (!output[i])
-			goto out;
+	output = kmalloc(COMP_BUF_SIZE, GFP_KERNEL);
+	if (!output)
+		return -ENOMEM;
 
-		decomp_out[i] = kmalloc(COMP_BUF_SIZE, GFP_KERNEL);
-		if (!decomp_out[i])
-			goto out;
+	decomp_out = kmalloc(COMP_BUF_SIZE, GFP_KERNEL);
+	if (!decomp_out) {
+		kfree(output);
+		return -ENOMEM;
 	}
 
 	for (i = 0; i < ctcount; i++) {
 		unsigned int dlen = COMP_BUF_SIZE;
 		int ilen = ctemplate[i].inlen;
 		void *input_vec;
-		int j;
 
 		input_vec = kmemdup(ctemplate[i].input, ilen, GFP_KERNEL);
 		if (!input_vec) {
@@ -3378,61 +3354,70 @@ static int test_acomp(struct crypto_acomp *tfm,
 			goto out;
 		}
 
+		memset(output, 0, dlen);
 		crypto_init_wait(&wait);
-		sg_init_one(src, input_vec, ilen);
+		sg_init_one(&src, input_vec, ilen);
+		sg_init_one(&dst, output, dlen);
 
-		for (j = 0; j < MAX_MB_MSGS; j++) {
-			sg_init_one(dst + j, output[j], dlen);
-			acomp_request_set_params(reqs[j], src, dst + j, ilen, dlen);
+		req = acomp_request_alloc(tfm);
+		if (!req) {
+			pr_err("alg: acomp: request alloc failed for %s\n",
+			       algo);
+			kfree(input_vec);
+			ret = -ENOMEM;
+			goto out;
 		}
 
-		req = reqs[0];
+		acomp_request_set_params(req, &src, &dst, ilen, dlen);
+		acomp_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG,
+					   crypto_req_done, &wait);
+
 		ret = crypto_wait_req(crypto_acomp_compress(req), &wait);
 		if (ret) {
 			pr_err("alg: acomp: compression failed on test %d for %s: ret=%d\n",
 			       i + 1, algo, -ret);
 			kfree(input_vec);
+			acomp_request_free(req);
 			goto out;
 		}
 
 		ilen = req->dlen;
 		dlen = COMP_BUF_SIZE;
+		sg_init_one(&src, output, ilen);
+		sg_init_one(&dst, decomp_out, dlen);
 		crypto_init_wait(&wait);
-		for (j = 0; j < MAX_MB_MSGS; j++) {
-			sg_init_one(src + j, output[j], ilen);
-			sg_init_one(dst + j, decomp_out[j], dlen);
-			acomp_request_set_params(reqs[j], src + j, dst + j, ilen, dlen);
-		}
-
-		crypto_wait_req(crypto_acomp_decompress(req), &wait);
-		for (j = 0; j < MAX_MB_MSGS; j++) {
-			ret = reqs[j]->base.err;
-			if (ret) {
-				pr_err("alg: acomp: compression failed on test %d (%d) for %s: ret=%d\n",
-				       i + 1, j, algo, -ret);
-				kfree(input_vec);
-				goto out;
-			}
+		acomp_request_set_params(req, &src, &dst, ilen, dlen);
 
-			if (reqs[j]->dlen != ctemplate[i].inlen) {
-				pr_err("alg: acomp: Compression test %d (%d) failed for %s: output len = %d\n",
-				       i + 1, j, algo, reqs[j]->dlen);
-				ret = -EINVAL;
-				kfree(input_vec);
-				goto out;
-			}
+		ret = crypto_wait_req(crypto_acomp_decompress(req), &wait);
+		if (ret) {
+			pr_err("alg: acomp: compression failed on test %d for %s: ret=%d\n",
+			       i + 1, algo, -ret);
+			kfree(input_vec);
+			acomp_request_free(req);
+			goto out;
+		}
 
-			if (memcmp(input_vec, decomp_out[j], reqs[j]->dlen)) {
-				pr_err("alg: acomp: Compression test %d (%d) failed for %s\n",
-				       i + 1, j, algo);
-				hexdump(output[j], reqs[j]->dlen);
-				ret = -EINVAL;
-				kfree(input_vec);
-				goto out;
-			}
+		if (req->dlen != ctemplate[i].inlen) {
+			pr_err("alg: acomp: Compression test %d failed for %s: output len = %d\n",
+			       i + 1, algo, req->dlen);
+			ret = -EINVAL;
+			kfree(input_vec);
+			acomp_request_free(req);
+			goto out;
+		}
+
+		if (memcmp(input_vec, decomp_out, req->dlen)) {
+			pr_err("alg: acomp: Compression test %d failed for %s\n",
+			       i + 1, algo);
+			hexdump(output, req->dlen);
+			ret = -EINVAL;
+			kfree(input_vec);
+			acomp_request_free(req);
+			goto out;
 		}
 
 		kfree(input_vec);
+		acomp_request_free(req);
 	}
 
 	for (i = 0; i < dtcount; i++) {
@@ -3446,9 +3431,10 @@ static int test_acomp(struct crypto_acomp *tfm,
 			goto out;
 		}
 
+		memset(output, 0, dlen);
 		crypto_init_wait(&wait);
-		sg_init_one(src, input_vec, ilen);
-		sg_init_one(dst, output[0], dlen);
+		sg_init_one(&src, input_vec, ilen);
+		sg_init_one(&dst, output, dlen);
 
 		req = acomp_request_alloc(tfm);
 		if (!req) {
@@ -3459,7 +3445,7 @@ static int test_acomp(struct crypto_acomp *tfm,
 			goto out;
 		}
 
-		acomp_request_set_params(req, src, dst, ilen, dlen);
+		acomp_request_set_params(req, &src, &dst, ilen, dlen);
 		acomp_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG,
 					   crypto_req_done, &wait);
 
@@ -3481,10 +3467,10 @@ static int test_acomp(struct crypto_acomp *tfm,
 			goto out;
 		}
 
-		if (memcmp(output[0], dtemplate[i].output, req->dlen)) {
+		if (memcmp(output, dtemplate[i].output, req->dlen)) {
 			pr_err("alg: acomp: Decompression test %d failed for %s\n",
 			       i + 1, algo);
-			hexdump(output[0], req->dlen);
+			hexdump(output, req->dlen);
 			ret = -EINVAL;
 			kfree(input_vec);
 			acomp_request_free(req);
@@ -3498,13 +3484,8 @@ static int test_acomp(struct crypto_acomp *tfm,
 	ret = 0;
 
 out:
-	acomp_request_free(reqs[0]);
-	for (i = 0; i < MAX_MB_MSGS; i++) {
-		kfree(output[i]);
-		kfree(decomp_out[i]);
-	}
-	kfree(dst);
-	kfree(src);
+	kfree(decomp_out);
+	kfree(output);
 	return ret;
 }
 
diff --git a/drivers/accel/amdxdna/TODO b/drivers/accel/amdxdna/TODO
index 5119bccd1917..ad8ac6e315b6 100644
--- a/drivers/accel/amdxdna/TODO
+++ b/drivers/accel/amdxdna/TODO
@@ -1,3 +1,2 @@
-- Add import and export BO support
 - Add debugfs support
 - Add debug BO support
diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index 00d215ac866e..e04549f64d69 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -758,27 +758,42 @@ int aie2_hwctx_config(struct amdxdna_hwctx *hwctx, u32 type, u64 value, void *bu
 static int aie2_populate_range(struct amdxdna_gem_obj *abo)
 {
 	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
-	struct mm_struct *mm = abo->mem.notifier.mm;
-	struct hmm_range range = { 0 };
+	struct amdxdna_umap *mapp;
 	unsigned long timeout;
+	struct mm_struct *mm;
+	bool found;
 	int ret;
 
-	XDNA_INFO_ONCE(xdna, "populate memory range %llx size %lx",
-		       abo->mem.userptr, abo->mem.size);
-	range.notifier = &abo->mem.notifier;
-	range.start = abo->mem.userptr;
-	range.end = abo->mem.userptr + abo->mem.size;
-	range.hmm_pfns = abo->mem.pfns;
-	range.default_flags = HMM_PFN_REQ_FAULT;
+	timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
+again:
+	found = false;
+	down_write(&xdna->notifier_lock);
+	list_for_each_entry(mapp, &abo->mem.umap_list, node) {
+		if (mapp->invalid) {
+			found = true;
+			break;
+		}
+	}
 
-	if (!mmget_not_zero(mm))
+	if (!found) {
+		abo->mem.map_invalid = false;
+		up_write(&xdna->notifier_lock);
+		return 0;
+	}
+	kref_get(&mapp->refcnt);
+	up_write(&xdna->notifier_lock);
+
+	XDNA_DBG(xdna, "populate memory range %lx %lx",
+		 mapp->vma->vm_start, mapp->vma->vm_end);
+	mm = mapp->notifier.mm;
+	if (!mmget_not_zero(mm)) {
+		amdxdna_umap_put(mapp);
 		return -EFAULT;
+	}
 
-	timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
-again:
-	range.notifier_seq = mmu_interval_read_begin(&abo->mem.notifier);
+	mapp->range.notifier_seq = mmu_interval_read_begin(&mapp->notifier);
 	mmap_read_lock(mm);
-	ret = hmm_range_fault(&range);
+	ret = hmm_range_fault(&mapp->range);
 	mmap_read_unlock(mm);
 	if (ret) {
 		if (time_after(jiffies, timeout)) {
@@ -786,21 +801,27 @@ again:
 			goto put_mm;
 		}
 
-		if (ret == -EBUSY)
+		if (ret == -EBUSY) {
+			amdxdna_umap_put(mapp);
 			goto again;
+		}
 
 		goto put_mm;
 	}
 
-	down_read(&xdna->notifier_lock);
-	if (mmu_interval_read_retry(&abo->mem.notifier, range.notifier_seq)) {
-		up_read(&xdna->notifier_lock);
+	down_write(&xdna->notifier_lock);
+	if (mmu_interval_read_retry(&mapp->notifier, mapp->range.notifier_seq)) {
+		up_write(&xdna->notifier_lock);
+		amdxdna_umap_put(mapp);
 		goto again;
 	}
-	abo->mem.map_invalid = false;
-	up_read(&xdna->notifier_lock);
+	mapp->invalid = false;
+	up_write(&xdna->notifier_lock);
+	amdxdna_umap_put(mapp);
+	goto again;
 
 put_mm:
+	amdxdna_umap_put(mapp);
 	mmput(mm);
 	return ret;
 }
@@ -908,10 +929,6 @@ void aie2_hmm_invalidate(struct amdxdna_gem_obj *abo,
 	struct drm_gem_object *gobj = to_gobj(abo);
 	long ret;
 
-	down_write(&xdna->notifier_lock);
-	abo->mem.map_invalid = true;
-	mmu_interval_set_seq(&abo->mem.notifier, cur_seq);
-	up_write(&xdna->notifier_lock);
 	ret = dma_resv_wait_timeout(gobj->resv, DMA_RESV_USAGE_BOOKKEEP,
 				    true, MAX_SCHEDULE_TIMEOUT);
 	if (!ret || ret == -ERESTARTSYS)
diff --git a/drivers/accel/amdxdna/aie2_message.c b/drivers/accel/amdxdna/aie2_message.c
index bf4219e32cc1..82412eec9a4b 100644
--- a/drivers/accel/amdxdna/aie2_message.c
+++ b/drivers/accel/amdxdna/aie2_message.c
@@ -525,7 +525,7 @@ aie2_cmdlist_fill_one_slot_cf(void *cmd_buf, u32 offset,
 	if (!payload)
 		return -EINVAL;
 
-	if (!slot_cf_has_space(offset, payload_len))
+	if (!slot_has_space(*buf, offset, payload_len))
 		return -ENOSPC;
 
 	buf->cu_idx = cu_idx;
@@ -558,7 +558,7 @@ aie2_cmdlist_fill_one_slot_dpu(void *cmd_buf, u32 offset,
 	if (payload_len < sizeof(*sn) || arg_sz > MAX_DPU_ARGS_SIZE)
 		return -EINVAL;
 
-	if (!slot_dpu_has_space(offset, arg_sz))
+	if (!slot_has_space(*buf, offset, arg_sz))
 		return -ENOSPC;
 
 	buf->inst_buf_addr = sn->buffer;
@@ -569,7 +569,7 @@ aie2_cmdlist_fill_one_slot_dpu(void *cmd_buf, u32 offset,
 	memcpy(buf->args, sn->prop_args, arg_sz);
 
 	/* Accurate buf size to hint firmware to do necessary copy */
-	*size += sizeof(*buf) + arg_sz;
+	*size = sizeof(*buf) + arg_sz;
 	return 0;
 }
 
diff --git a/drivers/accel/amdxdna/aie2_msg_priv.h b/drivers/accel/amdxdna/aie2_msg_priv.h
index 4e02e744b470..6df9065b13f6 100644
--- a/drivers/accel/amdxdna/aie2_msg_priv.h
+++ b/drivers/accel/amdxdna/aie2_msg_priv.h
@@ -319,18 +319,16 @@ struct async_event_msg_resp {
 } __packed;
 
 #define MAX_CHAIN_CMDBUF_SIZE SZ_4K
-#define slot_cf_has_space(offset, payload_size) \
-	(MAX_CHAIN_CMDBUF_SIZE - ((offset) + (payload_size)) > \
-	 offsetof(struct cmd_chain_slot_execbuf_cf, args[0]))
+#define slot_has_space(slot, offset, payload_size)		\
+	(MAX_CHAIN_CMDBUF_SIZE >= (offset) + (payload_size) +	\
+	 sizeof(typeof(slot)))
+
 struct cmd_chain_slot_execbuf_cf {
 	__u32 cu_idx;
 	__u32 arg_cnt;
 	__u32 args[] __counted_by(arg_cnt);
 };
 
-#define slot_dpu_has_space(offset, payload_size) \
-	(MAX_CHAIN_CMDBUF_SIZE - ((offset) + (payload_size)) > \
-	 offsetof(struct cmd_chain_slot_dpu, args[0]))
 struct cmd_chain_slot_dpu {
 	__u64 inst_buf_addr;
 	__u32 inst_size;
diff --git a/drivers/accel/amdxdna/amdxdna_ctx.c b/drivers/accel/amdxdna/amdxdna_ctx.c
index 43442b9e273b..be073224bd69 100644
--- a/drivers/accel/amdxdna/amdxdna_ctx.c
+++ b/drivers/accel/amdxdna/amdxdna_ctx.c
@@ -496,11 +496,11 @@ static int amdxdna_drm_submit_execbuf(struct amdxdna_client *client,
 				      struct amdxdna_drm_exec_cmd *args)
 {
 	struct amdxdna_dev *xdna = client->xdna;
-	u32 *arg_bo_hdls;
+	u32 *arg_bo_hdls = NULL;
 	u32 cmd_bo_hdl;
 	int ret;
 
-	if (!args->arg_count || args->arg_count > MAX_ARG_COUNT) {
+	if (args->arg_count > MAX_ARG_COUNT) {
 		XDNA_ERR(xdna, "Invalid arg bo count %d", args->arg_count);
 		return -EINVAL;
 	}
@@ -512,14 +512,16 @@ static int amdxdna_drm_submit_execbuf(struct amdxdna_client *client,
 	}
 
 	cmd_bo_hdl = (u32)args->cmd_handles;
-	arg_bo_hdls = kcalloc(args->arg_count, sizeof(u32), GFP_KERNEL);
-	if (!arg_bo_hdls)
-		return -ENOMEM;
-	ret = copy_from_user(arg_bo_hdls, u64_to_user_ptr(args->args),
-			     args->arg_count * sizeof(u32));
-	if (ret) {
-		ret = -EFAULT;
-		goto free_cmd_bo_hdls;
+	if (args->arg_count) {
+		arg_bo_hdls = kcalloc(args->arg_count, sizeof(u32), GFP_KERNEL);
+		if (!arg_bo_hdls)
+			return -ENOMEM;
+		ret = copy_from_user(arg_bo_hdls, u64_to_user_ptr(args->args),
+				     args->arg_count * sizeof(u32));
+		if (ret) {
+			ret = -EFAULT;
+			goto free_cmd_bo_hdls;
+		}
 	}
 
 	ret = amdxdna_cmd_submit(client, cmd_bo_hdl, arg_bo_hdls,
diff --git a/drivers/accel/amdxdna/amdxdna_gem.c b/drivers/accel/amdxdna/amdxdna_gem.c
index 606433d73236..26831ec69f89 100644
--- a/drivers/accel/amdxdna/amdxdna_gem.c
+++ b/drivers/accel/amdxdna/amdxdna_gem.c
@@ -9,7 +9,10 @@
 #include <drm/drm_gem.h>
 #include <drm/drm_gem_shmem_helper.h>
 #include <drm/gpu_scheduler.h>
+#include <linux/dma-buf.h>
+#include <linux/dma-direct.h>
 #include <linux/iosys-map.h>
+#include <linux/pagemap.h>
 #include <linux/vmalloc.h>
 
 #include "amdxdna_ctx.h"
@@ -18,6 +21,8 @@
 
 #define XDNA_MAX_CMD_BO_SIZE	SZ_32K
 
+MODULE_IMPORT_NS("DMA_BUF");
+
 static int
 amdxdna_gem_insert_node_locked(struct amdxdna_gem_obj *abo, bool use_vmap)
 {
@@ -55,57 +60,38 @@ amdxdna_gem_insert_node_locked(struct amdxdna_gem_obj *abo, bool use_vmap)
 	return 0;
 }
 
-static void amdxdna_gem_obj_free(struct drm_gem_object *gobj)
-{
-	struct amdxdna_dev *xdna = to_xdna_dev(gobj->dev);
-	struct amdxdna_gem_obj *abo = to_xdna_obj(gobj);
-	struct iosys_map map = IOSYS_MAP_INIT_VADDR(abo->mem.kva);
-
-	XDNA_DBG(xdna, "BO type %d xdna_addr 0x%llx", abo->type, abo->mem.dev_addr);
-	if (abo->pinned)
-		amdxdna_gem_unpin(abo);
-
-	if (abo->type == AMDXDNA_BO_DEV) {
-		mutex_lock(&abo->client->mm_lock);
-		drm_mm_remove_node(&abo->mm_node);
-		mutex_unlock(&abo->client->mm_lock);
-
-		vunmap(abo->mem.kva);
-		drm_gem_object_put(to_gobj(abo->dev_heap));
-		drm_gem_object_release(gobj);
-		mutex_destroy(&abo->lock);
-		kfree(abo);
-		return;
-	}
-
-	if (abo->type == AMDXDNA_BO_DEV_HEAP)
-		drm_mm_takedown(&abo->mm);
-
-	drm_gem_vunmap_unlocked(gobj, &map);
-	mutex_destroy(&abo->lock);
-	drm_gem_shmem_free(&abo->base);
-}
-
-static const struct drm_gem_object_funcs amdxdna_gem_dev_obj_funcs = {
-	.free = amdxdna_gem_obj_free,
-};
-
 static bool amdxdna_hmm_invalidate(struct mmu_interval_notifier *mni,
 				   const struct mmu_notifier_range *range,
 				   unsigned long cur_seq)
 {
-	struct amdxdna_gem_obj *abo = container_of(mni, struct amdxdna_gem_obj,
-						   mem.notifier);
-	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
+	struct amdxdna_umap *mapp = container_of(mni, struct amdxdna_umap, notifier);
+	struct amdxdna_gem_obj *abo = mapp->abo;
+	struct amdxdna_dev *xdna;
 
-	XDNA_DBG(xdna, "Invalid range 0x%llx, 0x%lx, type %d",
-		 abo->mem.userptr, abo->mem.size, abo->type);
+	xdna = to_xdna_dev(to_gobj(abo)->dev);
+	XDNA_DBG(xdna, "Invalidating range 0x%lx, 0x%lx, type %d",
+		 mapp->vma->vm_start, mapp->vma->vm_end, abo->type);
 
 	if (!mmu_notifier_range_blockable(range))
 		return false;
 
+	down_write(&xdna->notifier_lock);
+	abo->mem.map_invalid = true;
+	mapp->invalid = true;
+	mmu_interval_set_seq(&mapp->notifier, cur_seq);
+	up_write(&xdna->notifier_lock);
+
 	xdna->dev_info->ops->hmm_invalidate(abo, cur_seq);
 
+	if (range->event == MMU_NOTIFY_UNMAP) {
+		down_write(&xdna->notifier_lock);
+		if (!mapp->unmapped) {
+			queue_work(xdna->notifier_wq, &mapp->hmm_unreg_work);
+			mapp->unmapped = true;
+		}
+		up_write(&xdna->notifier_lock);
+	}
+
 	return true;
 }
 
@@ -113,102 +99,310 @@ static const struct mmu_interval_notifier_ops amdxdna_hmm_ops = {
 	.invalidate = amdxdna_hmm_invalidate,
 };
 
-static void amdxdna_hmm_unregister(struct amdxdna_gem_obj *abo)
+static void amdxdna_hmm_unregister(struct amdxdna_gem_obj *abo,
+				   struct vm_area_struct *vma)
 {
 	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
+	struct amdxdna_umap *mapp;
+
+	down_read(&xdna->notifier_lock);
+	list_for_each_entry(mapp, &abo->mem.umap_list, node) {
+		if (!vma || mapp->vma == vma) {
+			if (!mapp->unmapped) {
+				queue_work(xdna->notifier_wq, &mapp->hmm_unreg_work);
+				mapp->unmapped = true;
+			}
+			if (vma)
+				break;
+		}
+	}
+	up_read(&xdna->notifier_lock);
+}
 
-	if (!xdna->dev_info->ops->hmm_invalidate)
-		return;
+static void amdxdna_umap_release(struct kref *ref)
+{
+	struct amdxdna_umap *mapp = container_of(ref, struct amdxdna_umap, refcnt);
+	struct vm_area_struct *vma = mapp->vma;
+	struct amdxdna_dev *xdna;
+
+	mmu_interval_notifier_remove(&mapp->notifier);
+	if (is_import_bo(mapp->abo) && vma->vm_file && vma->vm_file->f_mapping)
+		mapping_clear_unevictable(vma->vm_file->f_mapping);
+
+	xdna = to_xdna_dev(to_gobj(mapp->abo)->dev);
+	down_write(&xdna->notifier_lock);
+	list_del(&mapp->node);
+	up_write(&xdna->notifier_lock);
+
+	kvfree(mapp->range.hmm_pfns);
+	kfree(mapp);
+}
+
+void amdxdna_umap_put(struct amdxdna_umap *mapp)
+{
+	kref_put(&mapp->refcnt, amdxdna_umap_release);
+}
+
+static void amdxdna_hmm_unreg_work(struct work_struct *work)
+{
+	struct amdxdna_umap *mapp = container_of(work, struct amdxdna_umap,
+						 hmm_unreg_work);
 
-	mmu_interval_notifier_remove(&abo->mem.notifier);
-	kvfree(abo->mem.pfns);
-	abo->mem.pfns = NULL;
+	amdxdna_umap_put(mapp);
 }
 
-static int amdxdna_hmm_register(struct amdxdna_gem_obj *abo, unsigned long addr,
-				size_t len)
+static int amdxdna_hmm_register(struct amdxdna_gem_obj *abo,
+				struct vm_area_struct *vma)
 {
 	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
+	unsigned long len = vma->vm_end - vma->vm_start;
+	unsigned long addr = vma->vm_start;
+	struct amdxdna_umap *mapp;
 	u32 nr_pages;
 	int ret;
 
 	if (!xdna->dev_info->ops->hmm_invalidate)
 		return 0;
 
-	if (abo->mem.pfns)
-		return -EEXIST;
+	mapp = kzalloc(sizeof(*mapp), GFP_KERNEL);
+	if (!mapp)
+		return -ENOMEM;
 
 	nr_pages = (PAGE_ALIGN(addr + len) - (addr & PAGE_MASK)) >> PAGE_SHIFT;
-	abo->mem.pfns = kvcalloc(nr_pages, sizeof(*abo->mem.pfns),
-				 GFP_KERNEL);
-	if (!abo->mem.pfns)
-		return -ENOMEM;
+	mapp->range.hmm_pfns = kvcalloc(nr_pages, sizeof(*mapp->range.hmm_pfns),
+					GFP_KERNEL);
+	if (!mapp->range.hmm_pfns) {
+		ret = -ENOMEM;
+		goto free_map;
+	}
 
-	ret = mmu_interval_notifier_insert_locked(&abo->mem.notifier,
+	ret = mmu_interval_notifier_insert_locked(&mapp->notifier,
 						  current->mm,
 						  addr,
 						  len,
 						  &amdxdna_hmm_ops);
 	if (ret) {
 		XDNA_ERR(xdna, "Insert mmu notifier failed, ret %d", ret);
-		kvfree(abo->mem.pfns);
+		goto free_pfns;
 	}
-	abo->mem.userptr = addr;
 
+	mapp->range.notifier = &mapp->notifier;
+	mapp->range.start = vma->vm_start;
+	mapp->range.end = vma->vm_end;
+	mapp->range.default_flags = HMM_PFN_REQ_FAULT;
+	mapp->vma = vma;
+	mapp->abo = abo;
+	kref_init(&mapp->refcnt);
+
+	if (abo->mem.userptr == AMDXDNA_INVALID_ADDR)
+		abo->mem.userptr = addr;
+	INIT_WORK(&mapp->hmm_unreg_work, amdxdna_hmm_unreg_work);
+	if (is_import_bo(abo) && vma->vm_file && vma->vm_file->f_mapping)
+		mapping_set_unevictable(vma->vm_file->f_mapping);
+
+	down_write(&xdna->notifier_lock);
+	list_add_tail(&mapp->node, &abo->mem.umap_list);
+	up_write(&xdna->notifier_lock);
+
+	return 0;
+
+free_pfns:
+	kvfree(mapp->range.hmm_pfns);
+free_map:
+	kfree(mapp);
 	return ret;
 }
 
+static int amdxdna_insert_pages(struct amdxdna_gem_obj *abo,
+				struct vm_area_struct *vma)
+{
+	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
+	unsigned long num_pages = vma_pages(vma);
+	unsigned long offset = 0;
+	int ret;
+
+	if (!is_import_bo(abo)) {
+		ret = drm_gem_shmem_mmap(&abo->base, vma);
+		if (ret) {
+			XDNA_ERR(xdna, "Failed shmem mmap %d", ret);
+			return ret;
+		}
+
+		/* The buffer is based on memory pages. Fix the flag. */
+		vm_flags_mod(vma, VM_MIXEDMAP, VM_PFNMAP);
+		ret = vm_insert_pages(vma, vma->vm_start, abo->base.pages,
+				      &num_pages);
+		if (ret) {
+			XDNA_ERR(xdna, "Failed insert pages %d", ret);
+			vma->vm_ops->close(vma);
+			return ret;
+		}
+
+		return 0;
+	}
+
+	vma->vm_private_data = NULL;
+	vma->vm_ops = NULL;
+	ret = dma_buf_mmap(to_gobj(abo)->dma_buf, vma, 0);
+	if (ret) {
+		XDNA_ERR(xdna, "Failed to mmap dma buf %d", ret);
+		return ret;
+	}
+
+	do {
+		vm_fault_t fault_ret;
+
+		fault_ret = handle_mm_fault(vma, vma->vm_start + offset,
+					    FAULT_FLAG_WRITE, NULL);
+		if (fault_ret & VM_FAULT_ERROR) {
+			vma->vm_ops->close(vma);
+			XDNA_ERR(xdna, "Fault in page failed");
+			return -EFAULT;
+		}
+
+		offset += PAGE_SIZE;
+	} while (--num_pages);
+
+	/* Drop the reference drm_gem_mmap_obj() acquired.*/
+	drm_gem_object_put(to_gobj(abo));
+
+	return 0;
+}
+
 static int amdxdna_gem_obj_mmap(struct drm_gem_object *gobj,
 				struct vm_area_struct *vma)
 {
+	struct amdxdna_dev *xdna = to_xdna_dev(gobj->dev);
 	struct amdxdna_gem_obj *abo = to_xdna_obj(gobj);
-	unsigned long num_pages;
 	int ret;
 
-	ret = amdxdna_hmm_register(abo, vma->vm_start, gobj->size);
+	ret = amdxdna_hmm_register(abo, vma);
 	if (ret)
 		return ret;
 
+	ret = amdxdna_insert_pages(abo, vma);
+	if (ret) {
+		XDNA_ERR(xdna, "Failed insert pages, ret %d", ret);
+		goto hmm_unreg;
+	}
+
+	XDNA_DBG(xdna, "BO map_offset 0x%llx type %d userptr 0x%lx size 0x%lx",
+		 drm_vma_node_offset_addr(&gobj->vma_node), abo->type,
+		 vma->vm_start, gobj->size);
+	return 0;
+
+hmm_unreg:
+	amdxdna_hmm_unregister(abo, vma);
+	return ret;
+}
+
+static int amdxdna_gem_dmabuf_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma)
+{
+	struct drm_gem_object *gobj = dma_buf->priv;
+	struct amdxdna_gem_obj *abo = to_xdna_obj(gobj);
+	unsigned long num_pages = vma_pages(vma);
+	int ret;
+
+	vma->vm_ops = &drm_gem_shmem_vm_ops;
+	vma->vm_private_data = gobj;
+
+	drm_gem_object_get(gobj);
 	ret = drm_gem_shmem_mmap(&abo->base, vma);
 	if (ret)
-		goto hmm_unreg;
+		goto put_obj;
 
-	num_pages = gobj->size >> PAGE_SHIFT;
-	/* Try to insert the pages */
+	/* The buffer is based on memory pages. Fix the flag. */
 	vm_flags_mod(vma, VM_MIXEDMAP, VM_PFNMAP);
-	ret = vm_insert_pages(vma, vma->vm_start, abo->base.pages, &num_pages);
+	ret = vm_insert_pages(vma, vma->vm_start, abo->base.pages,
+			      &num_pages);
 	if (ret)
-		XDNA_ERR(abo->client->xdna, "Failed insert pages, ret %d", ret);
+		goto close_vma;
 
 	return 0;
 
-hmm_unreg:
-	amdxdna_hmm_unregister(abo);
+close_vma:
+	vma->vm_ops->close(vma);
+put_obj:
+	drm_gem_object_put(gobj);
 	return ret;
 }
 
-static vm_fault_t amdxdna_gem_vm_fault(struct vm_fault *vmf)
+static const struct dma_buf_ops amdxdna_dmabuf_ops = {
+	.attach = drm_gem_map_attach,
+	.detach = drm_gem_map_detach,
+	.map_dma_buf = drm_gem_map_dma_buf,
+	.unmap_dma_buf = drm_gem_unmap_dma_buf,
+	.release = drm_gem_dmabuf_release,
+	.mmap = amdxdna_gem_dmabuf_mmap,
+	.vmap = drm_gem_dmabuf_vmap,
+	.vunmap = drm_gem_dmabuf_vunmap,
+};
+
+static struct dma_buf *amdxdna_gem_prime_export(struct drm_gem_object *gobj, int flags)
 {
-	return drm_gem_shmem_vm_ops.fault(vmf);
+	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
+
+	exp_info.ops = &amdxdna_dmabuf_ops;
+	exp_info.size = gobj->size;
+	exp_info.flags = flags;
+	exp_info.priv = gobj;
+	exp_info.resv = gobj->resv;
+
+	return drm_gem_dmabuf_export(gobj->dev, &exp_info);
 }
 
-static void amdxdna_gem_vm_open(struct vm_area_struct *vma)
+static void amdxdna_imported_obj_free(struct amdxdna_gem_obj *abo)
 {
-	drm_gem_shmem_vm_ops.open(vma);
+	dma_buf_unmap_attachment_unlocked(abo->attach, abo->base.sgt, DMA_BIDIRECTIONAL);
+	dma_buf_detach(abo->dma_buf, abo->attach);
+	dma_buf_put(abo->dma_buf);
+	drm_gem_object_release(to_gobj(abo));
+	kfree(abo);
 }
 
-static void amdxdna_gem_vm_close(struct vm_area_struct *vma)
+static void amdxdna_gem_obj_free(struct drm_gem_object *gobj)
 {
-	struct drm_gem_object *gobj = vma->vm_private_data;
+	struct amdxdna_dev *xdna = to_xdna_dev(gobj->dev);
+	struct amdxdna_gem_obj *abo = to_xdna_obj(gobj);
+	struct iosys_map map = IOSYS_MAP_INIT_VADDR(abo->mem.kva);
+
+	XDNA_DBG(xdna, "BO type %d xdna_addr 0x%llx", abo->type, abo->mem.dev_addr);
+
+	amdxdna_hmm_unregister(abo, NULL);
+	flush_workqueue(xdna->notifier_wq);
+
+	if (abo->pinned)
+		amdxdna_gem_unpin(abo);
 
-	amdxdna_hmm_unregister(to_xdna_obj(gobj));
-	drm_gem_shmem_vm_ops.close(vma);
+	if (abo->type == AMDXDNA_BO_DEV) {
+		mutex_lock(&abo->client->mm_lock);
+		drm_mm_remove_node(&abo->mm_node);
+		mutex_unlock(&abo->client->mm_lock);
+
+		vunmap(abo->mem.kva);
+		drm_gem_object_put(to_gobj(abo->dev_heap));
+		drm_gem_object_release(gobj);
+		mutex_destroy(&abo->lock);
+		kfree(abo);
+		return;
+	}
+
+	if (abo->type == AMDXDNA_BO_DEV_HEAP)
+		drm_mm_takedown(&abo->mm);
+
+	drm_gem_vunmap(gobj, &map);
+	mutex_destroy(&abo->lock);
+
+	if (is_import_bo(abo)) {
+		amdxdna_imported_obj_free(abo);
+		return;
+	}
+
+	drm_gem_shmem_free(&abo->base);
 }
 
-static const struct vm_operations_struct amdxdna_gem_vm_ops = {
-	.fault = amdxdna_gem_vm_fault,
-	.open = amdxdna_gem_vm_open,
-	.close = amdxdna_gem_vm_close,
+static const struct drm_gem_object_funcs amdxdna_gem_dev_obj_funcs = {
+	.free = amdxdna_gem_obj_free,
 };
 
 static const struct drm_gem_object_funcs amdxdna_gem_shmem_funcs = {
@@ -220,7 +414,8 @@ static const struct drm_gem_object_funcs amdxdna_gem_shmem_funcs = {
 	.vmap = drm_gem_shmem_object_vmap,
 	.vunmap = drm_gem_shmem_object_vunmap,
 	.mmap = amdxdna_gem_obj_mmap,
-	.vm_ops = &amdxdna_gem_vm_ops,
+	.vm_ops = &drm_gem_shmem_vm_ops,
+	.export = amdxdna_gem_prime_export,
 };
 
 static struct amdxdna_gem_obj *
@@ -239,6 +434,7 @@ amdxdna_gem_create_obj(struct drm_device *dev, size_t size)
 	abo->mem.userptr = AMDXDNA_INVALID_ADDR;
 	abo->mem.dev_addr = AMDXDNA_INVALID_ADDR;
 	abo->mem.size = size;
+	INIT_LIST_HEAD(&abo->mem.umap_list);
 
 	return abo;
 }
@@ -258,6 +454,51 @@ amdxdna_gem_create_object_cb(struct drm_device *dev, size_t size)
 	return to_gobj(abo);
 }
 
+struct drm_gem_object *
+amdxdna_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf)
+{
+	struct dma_buf_attachment *attach;
+	struct amdxdna_gem_obj *abo;
+	struct drm_gem_object *gobj;
+	struct sg_table *sgt;
+	int ret;
+
+	get_dma_buf(dma_buf);
+
+	attach = dma_buf_attach(dma_buf, dev->dev);
+	if (IS_ERR(attach)) {
+		ret = PTR_ERR(attach);
+		goto put_buf;
+	}
+
+	sgt = dma_buf_map_attachment_unlocked(attach, DMA_BIDIRECTIONAL);
+	if (IS_ERR(sgt)) {
+		ret = PTR_ERR(sgt);
+		goto fail_detach;
+	}
+
+	gobj = drm_gem_shmem_prime_import_sg_table(dev, attach, sgt);
+	if (IS_ERR(gobj)) {
+		ret = PTR_ERR(gobj);
+		goto fail_unmap;
+	}
+
+	abo = to_xdna_obj(gobj);
+	abo->attach = attach;
+	abo->dma_buf = dma_buf;
+
+	return gobj;
+
+fail_unmap:
+	dma_buf_unmap_attachment_unlocked(attach, sgt, DMA_BIDIRECTIONAL);
+fail_detach:
+	dma_buf_detach(dma_buf, attach);
+put_buf:
+	dma_buf_put(dma_buf);
+
+	return ERR_PTR(ret);
+}
+
 static struct amdxdna_gem_obj *
 amdxdna_drm_alloc_shmem(struct drm_device *dev,
 			struct amdxdna_drm_create_bo *args,
@@ -417,7 +658,7 @@ amdxdna_drm_create_cmd_bo(struct drm_device *dev,
 	abo->type = AMDXDNA_BO_CMD;
 	abo->client = filp->driver_priv;
 
-	ret = drm_gem_vmap_unlocked(to_gobj(abo), &map);
+	ret = drm_gem_vmap(to_gobj(abo), &map);
 	if (ret) {
 		XDNA_ERR(xdna, "Vmap cmd bo failed, ret %d", ret);
 		goto release_obj;
@@ -483,6 +724,9 @@ int amdxdna_gem_pin_nolock(struct amdxdna_gem_obj *abo)
 	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
 	int ret;
 
+	if (is_import_bo(abo))
+		return 0;
+
 	switch (abo->type) {
 	case AMDXDNA_BO_SHMEM:
 	case AMDXDNA_BO_DEV_HEAP:
@@ -515,6 +759,9 @@ int amdxdna_gem_pin(struct amdxdna_gem_obj *abo)
 
 void amdxdna_gem_unpin(struct amdxdna_gem_obj *abo)
 {
+	if (is_import_bo(abo))
+		return;
+
 	if (abo->type == AMDXDNA_BO_DEV)
 		abo = abo->dev_heap;
 
@@ -606,7 +853,9 @@ int amdxdna_drm_sync_bo_ioctl(struct drm_device *dev,
 		goto put_obj;
 	}
 
-	if (abo->type == AMDXDNA_BO_DEV)
+	if (is_import_bo(abo))
+		drm_clflush_sg(abo->base.sgt);
+	else if (abo->type == AMDXDNA_BO_DEV)
 		drm_clflush_pages(abo->mem.pages, abo->mem.nr_pages);
 	else
 		drm_clflush_pages(abo->base.pages, gobj->size >> PAGE_SHIFT);
diff --git a/drivers/accel/amdxdna/amdxdna_gem.h b/drivers/accel/amdxdna/amdxdna_gem.h
index 8ccc0375dd9d..aee97e971d6d 100644
--- a/drivers/accel/amdxdna/amdxdna_gem.h
+++ b/drivers/accel/amdxdna/amdxdna_gem.h
@@ -6,6 +6,20 @@
 #ifndef _AMDXDNA_GEM_H_
 #define _AMDXDNA_GEM_H_
 
+#include <linux/hmm.h>
+
+struct amdxdna_umap {
+	struct vm_area_struct		*vma;
+	struct mmu_interval_notifier	notifier;
+	struct hmm_range		range;
+	struct work_struct		hmm_unreg_work;
+	struct amdxdna_gem_obj		*abo;
+	struct list_head		node;
+	struct kref			refcnt;
+	bool				invalid;
+	bool				unmapped;
+};
+
 struct amdxdna_mem {
 	u64				userptr;
 	void				*kva;
@@ -13,8 +27,7 @@ struct amdxdna_mem {
 	size_t				size;
 	struct page			**pages;
 	u32				nr_pages;
-	struct mmu_interval_notifier	notifier;
-	unsigned long			*pfns;
+	struct list_head		umap_list;
 	bool				map_invalid;
 };
 
@@ -31,9 +44,12 @@ struct amdxdna_gem_obj {
 	struct amdxdna_gem_obj		*dev_heap; /* For AMDXDNA_BO_DEV */
 	struct drm_mm_node		mm_node; /* For AMDXDNA_BO_DEV */
 	u32				assigned_hwctx;
+	struct dma_buf			*dma_buf;
+	struct dma_buf_attachment	*attach;
 };
 
 #define to_gobj(obj)    (&(obj)->base.base)
+#define is_import_bo(obj) ((obj)->attach)
 
 static inline struct amdxdna_gem_obj *to_xdna_obj(struct drm_gem_object *gobj)
 {
@@ -47,8 +63,12 @@ static inline void amdxdna_gem_put_obj(struct amdxdna_gem_obj *abo)
 	drm_gem_object_put(to_gobj(abo));
 }
 
+void amdxdna_umap_put(struct amdxdna_umap *mapp);
+
 struct drm_gem_object *
 amdxdna_gem_create_object_cb(struct drm_device *dev, size_t size);
+struct drm_gem_object *
+amdxdna_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf);
 struct amdxdna_gem_obj *
 amdxdna_drm_alloc_dev_bo(struct drm_device *dev,
 			 struct amdxdna_drm_create_bo *args,
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.c b/drivers/accel/amdxdna/amdxdna_pci_drv.c
index f5b8497cf5ad..f2bf1d374cc7 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.c
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.c
@@ -226,6 +226,7 @@ const struct drm_driver amdxdna_drm_drv = {
 	.num_ioctls = ARRAY_SIZE(amdxdna_drm_ioctls),
 
 	.gem_create_object = amdxdna_gem_create_object_cb,
+	.gem_prime_import = amdxdna_gem_prime_import,
 };
 
 static const struct amdxdna_dev_info *
@@ -266,12 +267,16 @@ static int amdxdna_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		fs_reclaim_release(GFP_KERNEL);
 	}
 
+	xdna->notifier_wq = alloc_ordered_workqueue("notifier_wq", 0);
+	if (!xdna->notifier_wq)
+		return -ENOMEM;
+
 	mutex_lock(&xdna->dev_lock);
 	ret = xdna->dev_info->ops->init(xdna);
 	mutex_unlock(&xdna->dev_lock);
 	if (ret) {
 		XDNA_ERR(xdna, "Hardware init failed, ret %d", ret);
-		return ret;
+		goto destroy_notifier_wq;
 	}
 
 	ret = amdxdna_sysfs_init(xdna);
@@ -301,6 +306,8 @@ failed_dev_fini:
 	mutex_lock(&xdna->dev_lock);
 	xdna->dev_info->ops->fini(xdna);
 	mutex_unlock(&xdna->dev_lock);
+destroy_notifier_wq:
+	destroy_workqueue(xdna->notifier_wq);
 	return ret;
 }
 
@@ -310,6 +317,8 @@ static void amdxdna_remove(struct pci_dev *pdev)
 	struct device *dev = &pdev->dev;
 	struct amdxdna_client *client;
 
+	destroy_workqueue(xdna->notifier_wq);
+
 	pm_runtime_get_noresume(dev);
 	pm_runtime_forbid(dev);
 
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.h b/drivers/accel/amdxdna/amdxdna_pci_drv.h
index 37848a8d8031..ab79600911aa 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.h
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.h
@@ -6,6 +6,7 @@
 #ifndef _AMDXDNA_PCI_DRV_H_
 #define _AMDXDNA_PCI_DRV_H_
 
+#include <linux/workqueue.h>
 #include <linux/xarray.h>
 
 #define XDNA_INFO(xdna, fmt, args...)	drm_info(&(xdna)->ddev, fmt, ##args)
@@ -98,6 +99,7 @@ struct amdxdna_dev {
 	struct list_head		client_list;
 	struct amdxdna_fw_ver		fw_ver;
 	struct rw_semaphore		notifier_lock; /* for mmu notifier*/
+	struct workqueue_struct		*notifier_wq;
 };
 
 /*
diff --git a/drivers/accel/ivpu/ivpu_debugfs.c b/drivers/accel/ivpu/ivpu_debugfs.c
index 0825851656a2..f0dad0c9ce33 100644
--- a/drivers/accel/ivpu/ivpu_debugfs.c
+++ b/drivers/accel/ivpu/ivpu_debugfs.c
@@ -332,7 +332,7 @@ ivpu_force_recovery_fn(struct file *file, const char __user *user_buf, size_t si
 		return -EINVAL;
 
 	ret = ivpu_rpm_get(vdev);
-	if (ret)
+	if (ret < 0)
 		return ret;
 
 	ivpu_pm_trigger_recovery(vdev, "debugfs");
@@ -383,7 +383,7 @@ static int dct_active_set(void *data, u64 active_percent)
 		return -EINVAL;
 
 	ret = ivpu_rpm_get(vdev);
-	if (ret)
+	if (ret < 0)
 		return ret;
 
 	if (active_percent)
diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c
index 4fa73189502e..0e7748c5e117 100644
--- a/drivers/accel/ivpu/ivpu_drv.c
+++ b/drivers/accel/ivpu/ivpu_drv.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * Copyright (C) 2020-2024 Intel Corporation
+ * Copyright (C) 2020-2025 Intel Corporation
  */
 
 #include <linux/firmware.h>
@@ -164,7 +164,7 @@ static int ivpu_get_param_ioctl(struct drm_device *dev, void *data, struct drm_f
 		args->value = vdev->platform;
 		break;
 	case DRM_IVPU_PARAM_CORE_CLOCK_RATE:
-		args->value = ivpu_hw_ratio_to_freq(vdev, vdev->hw->pll.max_ratio);
+		args->value = ivpu_hw_dpu_max_freq_get(vdev);
 		break;
 	case DRM_IVPU_PARAM_NUM_CONTEXTS:
 		args->value = ivpu_get_context_count(vdev);
@@ -374,6 +374,9 @@ int ivpu_boot(struct ivpu_device *vdev)
 {
 	int ret;
 
+	drm_WARN_ON(&vdev->drm, atomic_read(&vdev->job_timeout_counter));
+	drm_WARN_ON(&vdev->drm, !xa_empty(&vdev->submitted_jobs_xa));
+
 	/* Update boot params located at first 4KB of FW memory */
 	ivpu_fw_boot_params_setup(vdev, ivpu_bo_vaddr(vdev->fw->mem));
 
@@ -421,9 +424,9 @@ void ivpu_prepare_for_reset(struct ivpu_device *vdev)
 {
 	ivpu_hw_irq_disable(vdev);
 	disable_irq(vdev->irq);
-	cancel_work_sync(&vdev->irq_ipc_work);
-	cancel_work_sync(&vdev->irq_dct_work);
-	cancel_work_sync(&vdev->context_abort_work);
+	flush_work(&vdev->irq_ipc_work);
+	flush_work(&vdev->irq_dct_work);
+	flush_work(&vdev->context_abort_work);
 	ivpu_ipc_disable(vdev);
 	ivpu_mmu_disable(vdev);
 }
@@ -573,6 +576,7 @@ static int ivpu_dev_init(struct ivpu_device *vdev)
 	vdev->context_xa_limit.min = IVPU_USER_CONTEXT_MIN_SSID;
 	vdev->context_xa_limit.max = IVPU_USER_CONTEXT_MAX_SSID;
 	atomic64_set(&vdev->unique_id_counter, 0);
+	atomic_set(&vdev->job_timeout_counter, 0);
 	xa_init_flags(&vdev->context_xa, XA_FLAGS_ALLOC | XA_FLAGS_LOCK_IRQ);
 	xa_init_flags(&vdev->submitted_jobs_xa, XA_FLAGS_ALLOC1);
 	xa_init_flags(&vdev->db_xa, XA_FLAGS_ALLOC1);
diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h
index 92753effb1c9..5497e7030e91 100644
--- a/drivers/accel/ivpu/ivpu_drv.h
+++ b/drivers/accel/ivpu/ivpu_drv.h
@@ -154,6 +154,7 @@ struct ivpu_device {
 	struct mutex submitted_jobs_lock; /* Protects submitted_jobs */
 	struct xarray submitted_jobs_xa;
 	struct ivpu_ipc_consumer job_done_consumer;
+	atomic_t job_timeout_counter;
 
 	atomic64_t unique_id_counter;
 
diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c
index 7a1bb92d8c81..ccaaf6c100c0 100644
--- a/drivers/accel/ivpu/ivpu_fw.c
+++ b/drivers/accel/ivpu/ivpu_fw.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * Copyright (C) 2020-2024 Intel Corporation
+ * Copyright (C) 2020-2025 Intel Corporation
  */
 
 #include <linux/firmware.h>
@@ -233,10 +233,20 @@ static int ivpu_fw_parse(struct ivpu_device *vdev)
 	fw->dvfs_mode = 0;
 
 	fw->sched_mode = ivpu_fw_sched_mode_select(vdev, fw_hdr);
-	fw->primary_preempt_buf_size = fw_hdr->preemption_buffer_1_size;
-	fw->secondary_preempt_buf_size = fw_hdr->preemption_buffer_2_size;
 	ivpu_info(vdev, "Scheduler mode: %s\n", fw->sched_mode ? "HW" : "OS");
 
+	if (fw_hdr->preemption_buffer_1_max_size)
+		fw->primary_preempt_buf_size = fw_hdr->preemption_buffer_1_max_size;
+	else
+		fw->primary_preempt_buf_size = fw_hdr->preemption_buffer_1_size;
+
+	if (fw_hdr->preemption_buffer_2_max_size)
+		fw->secondary_preempt_buf_size = fw_hdr->preemption_buffer_2_max_size;
+	else
+		fw->secondary_preempt_buf_size = fw_hdr->preemption_buffer_2_size;
+	ivpu_dbg(vdev, FW_BOOT, "Preemption buffer sizes: primary %u, secondary %u\n",
+		 fw->primary_preempt_buf_size, fw->secondary_preempt_buf_size);
+
 	if (fw_hdr->ro_section_start_address && !is_within_range(fw_hdr->ro_section_start_address,
 								 fw_hdr->ro_section_size,
 								 fw_hdr->image_load_address,
@@ -534,7 +544,7 @@ static void ivpu_fw_boot_params_print(struct ivpu_device *vdev, struct vpu_boot_
 		 boot_params->d0i3_entry_vpu_ts);
 	ivpu_dbg(vdev, FW_BOOT, "boot_params.system_time_us = %llu\n",
 		 boot_params->system_time_us);
-	ivpu_dbg(vdev, FW_BOOT, "boot_params.power_profile = %u\n",
+	ivpu_dbg(vdev, FW_BOOT, "boot_params.power_profile = 0x%x\n",
 		 boot_params->power_profile);
 }
 
@@ -566,7 +576,6 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params
 
 	boot_params->magic = VPU_BOOT_PARAMS_MAGIC;
 	boot_params->vpu_id = to_pci_dev(vdev->drm.dev)->bus->number;
-	boot_params->frequency = ivpu_hw_pll_freq_get(vdev);
 
 	/*
 	 * This param is a debug firmware feature.  It switches default clock
@@ -637,7 +646,7 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params
 	boot_params->d0i3_residency_time_us = 0;
 	boot_params->d0i3_entry_vpu_ts = 0;
 	if (IVPU_WA(disable_d0i2))
-		boot_params->power_profile = 1;
+		boot_params->power_profile |= BIT(1);
 
 	boot_params->system_time_us = ktime_to_us(ktime_get_real());
 	wmb(); /* Flush WC buffers after writing bootparams */
diff --git a/drivers/accel/ivpu/ivpu_fw.h b/drivers/accel/ivpu/ivpu_fw.h
index 1d0b2bd9d65c..9a3935be1c05 100644
--- a/drivers/accel/ivpu/ivpu_fw.h
+++ b/drivers/accel/ivpu/ivpu_fw.h
@@ -39,6 +39,7 @@ struct ivpu_fw_info {
 	u64 read_only_addr;
 	u32 read_only_size;
 	u32 sched_mode;
+	u64 last_heartbeat;
 };
 
 int ivpu_fw_init(struct ivpu_device *vdev);
diff --git a/drivers/accel/ivpu/ivpu_gem.c b/drivers/accel/ivpu/ivpu_gem.c
index 8741c73b92ce..e0d242d9f3e5 100644
--- a/drivers/accel/ivpu/ivpu_gem.c
+++ b/drivers/accel/ivpu/ivpu_gem.c
@@ -30,7 +30,7 @@ static inline void ivpu_dbg_bo(struct ivpu_device *vdev, struct ivpu_bo *bo, con
 		 "%6s: bo %8p vpu_addr %9llx size %8zu ctx %d has_pages %d dma_mapped %d mmu_mapped %d wc %d imported %d\n",
 		 action, bo, bo->vpu_addr, ivpu_bo_size(bo), bo->ctx ? bo->ctx->id : 0,
 		 (bool)bo->base.pages, (bool)bo->base.sgt, bo->mmu_mapped, bo->base.map_wc,
-		 (bool)bo->base.base.import_attach);
+		 (bool)drm_gem_is_imported(&bo->base.base));
 }
 
 /*
@@ -122,7 +122,7 @@ static void ivpu_bo_unbind_locked(struct ivpu_bo *bo)
 		bo->ctx = NULL;
 	}
 
-	if (bo->base.base.import_attach)
+	if (drm_gem_is_imported(&bo->base.base))
 		return;
 
 	dma_resv_lock(bo->base.base.resv, NULL);
@@ -282,7 +282,7 @@ static void ivpu_gem_bo_free(struct drm_gem_object *obj)
 	ivpu_bo_unbind_locked(bo);
 	mutex_destroy(&bo->lock);
 
-	drm_WARN_ON(obj->dev, bo->base.pages_use_count > 1);
+	drm_WARN_ON(obj->dev, refcount_read(&bo->base.pages_use_count) > 1);
 	drm_gem_shmem_free(&bo->base);
 }
 
@@ -362,7 +362,7 @@ ivpu_bo_create(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx,
 
 	if (flags & DRM_IVPU_BO_MAPPABLE) {
 		dma_resv_lock(bo->base.base.resv, NULL);
-		ret = drm_gem_shmem_vmap(&bo->base, &map);
+		ret = drm_gem_shmem_vmap_locked(&bo->base, &map);
 		dma_resv_unlock(bo->base.base.resv);
 
 		if (ret)
@@ -387,7 +387,7 @@ void ivpu_bo_free(struct ivpu_bo *bo)
 
 	if (bo->flags & DRM_IVPU_BO_MAPPABLE) {
 		dma_resv_lock(bo->base.base.resv, NULL);
-		drm_gem_shmem_vunmap(&bo->base, &map);
+		drm_gem_shmem_vunmap_locked(&bo->base, &map);
 		dma_resv_unlock(bo->base.base.resv);
 	}
 
@@ -461,7 +461,7 @@ static void ivpu_bo_print_info(struct ivpu_bo *bo, struct drm_printer *p)
 	if (bo->mmu_mapped)
 		drm_printf(p, " mmu_mapped");
 
-	if (bo->base.base.import_attach)
+	if (drm_gem_is_imported(&bo->base.base))
 		drm_printf(p, " imported");
 
 	drm_printf(p, "\n");
diff --git a/drivers/accel/ivpu/ivpu_hw.h b/drivers/accel/ivpu/ivpu_hw.h
index 16435f2756d0..d79668fe1609 100644
--- a/drivers/accel/ivpu/ivpu_hw.h
+++ b/drivers/accel/ivpu/ivpu_hw.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * Copyright (C) 2020-2024 Intel Corporation
+ * Copyright (C) 2020-2025 Intel Corporation
  */
 
 #ifndef __IVPU_HW_H__
@@ -82,19 +82,19 @@ static inline u64 ivpu_hw_range_size(const struct ivpu_addr_range *range)
 	return range->end - range->start;
 }
 
-static inline u32 ivpu_hw_ratio_to_freq(struct ivpu_device *vdev, u32 ratio)
+static inline u32 ivpu_hw_dpu_max_freq_get(struct ivpu_device *vdev)
 {
-	return ivpu_hw_btrs_ratio_to_freq(vdev, ratio);
+	return ivpu_hw_btrs_dpu_max_freq_get(vdev);
 }
 
-static inline void ivpu_hw_irq_clear(struct ivpu_device *vdev)
+static inline u32 ivpu_hw_dpu_freq_get(struct ivpu_device *vdev)
 {
-	ivpu_hw_ip_irq_clear(vdev);
+	return ivpu_hw_btrs_dpu_freq_get(vdev);
 }
 
-static inline u32 ivpu_hw_pll_freq_get(struct ivpu_device *vdev)
+static inline void ivpu_hw_irq_clear(struct ivpu_device *vdev)
 {
-	return ivpu_hw_btrs_pll_freq_get(vdev);
+	ivpu_hw_ip_irq_clear(vdev);
 }
 
 static inline u32 ivpu_hw_profiling_freq_get(struct ivpu_device *vdev)
diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c
index 56c56012b980..b236c7234daa 100644
--- a/drivers/accel/ivpu/ivpu_hw_btrs.c
+++ b/drivers/accel/ivpu/ivpu_hw_btrs.c
@@ -1,8 +1,10 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * Copyright (C) 2020-2024 Intel Corporation
+ * Copyright (C) 2020-2025 Intel Corporation
  */
 
+#include <linux/units.h>
+
 #include "ivpu_drv.h"
 #include "ivpu_hw.h"
 #include "ivpu_hw_btrs.h"
@@ -28,17 +30,13 @@
 
 #define BTRS_LNL_ALL_IRQ_MASK ((u32)-1)
 
-#define BTRS_MTL_WP_CONFIG_1_TILE_5_3_RATIO WP_CONFIG(MTL_CONFIG_1_TILE, MTL_PLL_RATIO_5_3)
-#define BTRS_MTL_WP_CONFIG_1_TILE_4_3_RATIO WP_CONFIG(MTL_CONFIG_1_TILE, MTL_PLL_RATIO_4_3)
-#define BTRS_MTL_WP_CONFIG_2_TILE_5_3_RATIO WP_CONFIG(MTL_CONFIG_2_TILE, MTL_PLL_RATIO_5_3)
-#define BTRS_MTL_WP_CONFIG_2_TILE_4_3_RATIO WP_CONFIG(MTL_CONFIG_2_TILE, MTL_PLL_RATIO_4_3)
-#define BTRS_MTL_WP_CONFIG_0_TILE_PLL_OFF   WP_CONFIG(0, 0)
 
 #define PLL_CDYN_DEFAULT               0x80
 #define PLL_EPP_DEFAULT                0x80
 #define PLL_CONFIG_DEFAULT             0x0
-#define PLL_SIMULATION_FREQ            10000000
-#define PLL_REF_CLK_FREQ               50000000
+#define PLL_REF_CLK_FREQ               50000000ull
+#define PLL_RATIO_TO_FREQ(x)           ((x) * PLL_REF_CLK_FREQ)
+
 #define PLL_TIMEOUT_US		       (1500 * USEC_PER_MSEC)
 #define IDLE_TIMEOUT_US		       (5 * USEC_PER_MSEC)
 #define TIMEOUT_US                     (150 * USEC_PER_MSEC)
@@ -62,6 +60,8 @@
 #define DCT_ENABLE                     0x1
 #define DCT_DISABLE                    0x0
 
+static u32 pll_ratio_to_dpu_freq(struct ivpu_device *vdev, u32 ratio);
+
 int ivpu_hw_btrs_irqs_clear_with_0_mtl(struct ivpu_device *vdev)
 {
 	REGB_WR32(VPU_HW_BTRS_MTL_INTERRUPT_STAT, BTRS_MTL_ALL_IRQ_MASK);
@@ -156,7 +156,7 @@ static int info_init_mtl(struct ivpu_device *vdev)
 
 	hw->tile_fuse = BTRS_MTL_TILE_FUSE_ENABLE_BOTH;
 	hw->sku = BTRS_MTL_TILE_SKU_BOTH;
-	hw->config = BTRS_MTL_WP_CONFIG_2_TILE_4_3_RATIO;
+	hw->config = WP_CONFIG(MTL_CONFIG_2_TILE, MTL_PLL_RATIO_4_3);
 
 	return 0;
 }
@@ -334,8 +334,8 @@ int ivpu_hw_btrs_wp_drive(struct ivpu_device *vdev, bool enable)
 
 	prepare_wp_request(vdev, &wp, enable);
 
-	ivpu_dbg(vdev, PM, "PLL workpoint request: %u Hz, config: 0x%x, epp: 0x%x, cdyn: 0x%x\n",
-		 PLL_RATIO_TO_FREQ(wp.target), wp.cfg, wp.epp, wp.cdyn);
+	ivpu_dbg(vdev, PM, "PLL workpoint request: %lu MHz, config: 0x%x, epp: 0x%x, cdyn: 0x%x\n",
+		 pll_ratio_to_dpu_freq(vdev, wp.target) / HZ_PER_MHZ, wp.cfg, wp.epp, wp.cdyn);
 
 	ret = wp_request_send(vdev, &wp);
 	if (ret) {
@@ -573,6 +573,47 @@ int ivpu_hw_btrs_wait_for_idle(struct ivpu_device *vdev)
 		return REGB_POLL_FLD(VPU_HW_BTRS_LNL_VPU_STATUS, IDLE, 0x1, IDLE_TIMEOUT_US);
 }
 
+static u32 pll_config_get_mtl(struct ivpu_device *vdev)
+{
+	return REGB_RD32(VPU_HW_BTRS_MTL_CURRENT_PLL);
+}
+
+static u32 pll_config_get_lnl(struct ivpu_device *vdev)
+{
+	return REGB_RD32(VPU_HW_BTRS_LNL_PLL_FREQ);
+}
+
+static u32 pll_ratio_to_dpu_freq_mtl(u16 ratio)
+{
+	return (PLL_RATIO_TO_FREQ(ratio) * 2) / 3;
+}
+
+static u32 pll_ratio_to_dpu_freq_lnl(u16 ratio)
+{
+	return PLL_RATIO_TO_FREQ(ratio) / 2;
+}
+
+static u32 pll_ratio_to_dpu_freq(struct ivpu_device *vdev, u32 ratio)
+{
+	if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL)
+		return pll_ratio_to_dpu_freq_mtl(ratio);
+	else
+		return pll_ratio_to_dpu_freq_lnl(ratio);
+}
+
+u32 ivpu_hw_btrs_dpu_max_freq_get(struct ivpu_device *vdev)
+{
+	return pll_ratio_to_dpu_freq(vdev, vdev->hw->pll.max_ratio);
+}
+
+u32 ivpu_hw_btrs_dpu_freq_get(struct ivpu_device *vdev)
+{
+	if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL)
+		return pll_ratio_to_dpu_freq_mtl(pll_config_get_mtl(vdev));
+	else
+		return pll_ratio_to_dpu_freq_lnl(pll_config_get_lnl(vdev));
+}
+
 /* Handler for IRQs from Buttress core (irqB) */
 bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq)
 {
@@ -582,9 +623,12 @@ bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq)
 	if (!status)
 		return false;
 
-	if (REG_TEST_FLD(VPU_HW_BTRS_MTL_INTERRUPT_STAT, FREQ_CHANGE, status))
-		ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq: %08x",
-			 REGB_RD32(VPU_HW_BTRS_MTL_CURRENT_PLL));
+	if (REG_TEST_FLD(VPU_HW_BTRS_MTL_INTERRUPT_STAT, FREQ_CHANGE, status)) {
+		u32 pll = pll_config_get_mtl(vdev);
+
+		ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq, wp %08x, %lu MHz",
+			 pll, pll_ratio_to_dpu_freq_mtl(pll) / HZ_PER_MHZ);
+	}
 
 	if (REG_TEST_FLD(VPU_HW_BTRS_MTL_INTERRUPT_STAT, ATS_ERR, status)) {
 		ivpu_err(vdev, "ATS_ERR irq 0x%016llx", REGB_RD64(VPU_HW_BTRS_MTL_ATS_ERR_LOG_0));
@@ -633,8 +677,12 @@ bool ivpu_hw_btrs_irq_handler_lnl(struct ivpu_device *vdev, int irq)
 		queue_work(system_wq, &vdev->irq_dct_work);
 	}
 
-	if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, FREQ_CHANGE, status))
-		ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq: %08x", REGB_RD32(VPU_HW_BTRS_LNL_PLL_FREQ));
+	if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, FREQ_CHANGE, status)) {
+		u32 pll = pll_config_get_lnl(vdev);
+
+		ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq, wp %08x, %lu MHz",
+			 pll, pll_ratio_to_dpu_freq_lnl(pll) / HZ_PER_MHZ);
+	}
 
 	if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, ATS_ERR, status)) {
 		ivpu_err(vdev, "ATS_ERR LOG1 0x%08x ATS_ERR_LOG2 0x%08x\n",
@@ -717,60 +765,6 @@ void ivpu_hw_btrs_dct_set_status(struct ivpu_device *vdev, bool enable, u32 acti
 	REGB_WR32(VPU_HW_BTRS_LNL_PCODE_MAILBOX_STATUS, val);
 }
 
-static u32 pll_ratio_to_freq_mtl(u32 ratio, u32 config)
-{
-	u32 pll_clock = PLL_REF_CLK_FREQ * ratio;
-	u32 cpu_clock;
-
-	if ((config & 0xff) == MTL_PLL_RATIO_4_3)
-		cpu_clock = pll_clock * 2 / 4;
-	else
-		cpu_clock = pll_clock * 2 / 5;
-
-	return cpu_clock;
-}
-
-u32 ivpu_hw_btrs_ratio_to_freq(struct ivpu_device *vdev, u32 ratio)
-{
-	struct ivpu_hw_info *hw = vdev->hw;
-
-	if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL)
-		return pll_ratio_to_freq_mtl(ratio, hw->config);
-	else
-		return PLL_RATIO_TO_FREQ(ratio);
-}
-
-static u32 pll_freq_get_mtl(struct ivpu_device *vdev)
-{
-	u32 pll_curr_ratio;
-
-	pll_curr_ratio = REGB_RD32(VPU_HW_BTRS_MTL_CURRENT_PLL);
-	pll_curr_ratio &= VPU_HW_BTRS_MTL_CURRENT_PLL_RATIO_MASK;
-
-	if (!ivpu_is_silicon(vdev))
-		return PLL_SIMULATION_FREQ;
-
-	return pll_ratio_to_freq_mtl(pll_curr_ratio, vdev->hw->config);
-}
-
-static u32 pll_freq_get_lnl(struct ivpu_device *vdev)
-{
-	u32 pll_curr_ratio;
-
-	pll_curr_ratio = REGB_RD32(VPU_HW_BTRS_LNL_PLL_FREQ);
-	pll_curr_ratio &= VPU_HW_BTRS_LNL_PLL_FREQ_RATIO_MASK;
-
-	return PLL_RATIO_TO_FREQ(pll_curr_ratio);
-}
-
-u32 ivpu_hw_btrs_pll_freq_get(struct ivpu_device *vdev)
-{
-	if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL)
-		return pll_freq_get_mtl(vdev);
-	else
-		return pll_freq_get_lnl(vdev);
-}
-
 u32 ivpu_hw_btrs_telemetry_offset_get(struct ivpu_device *vdev)
 {
 	if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL)
diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.h b/drivers/accel/ivpu/ivpu_hw_btrs.h
index 1fd71b4d4ab0..d2d82651976d 100644
--- a/drivers/accel/ivpu/ivpu_hw_btrs.h
+++ b/drivers/accel/ivpu/ivpu_hw_btrs.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * Copyright (C) 2020-2024 Intel Corporation
+ * Copyright (C) 2020-2025 Intel Corporation
  */
 
 #ifndef __IVPU_HW_BTRS_H__
@@ -13,9 +13,8 @@
 
 #define PLL_PROFILING_FREQ_DEFAULT   38400000
 #define PLL_PROFILING_FREQ_HIGH      400000000
-#define PLL_RATIO_TO_FREQ(x)         ((x) * PLL_REF_CLK_FREQ)
 
-#define DCT_DEFAULT_ACTIVE_PERCENT 15u
+#define DCT_DEFAULT_ACTIVE_PERCENT 30u
 #define DCT_PERIOD_US		   35300u
 
 int ivpu_hw_btrs_info_init(struct ivpu_device *vdev);
@@ -32,12 +31,12 @@ int ivpu_hw_btrs_ip_reset(struct ivpu_device *vdev);
 void ivpu_hw_btrs_profiling_freq_reg_set_lnl(struct ivpu_device *vdev);
 void ivpu_hw_btrs_ats_print_lnl(struct ivpu_device *vdev);
 void ivpu_hw_btrs_clock_relinquish_disable_lnl(struct ivpu_device *vdev);
+u32 ivpu_hw_btrs_dpu_max_freq_get(struct ivpu_device *vdev);
+u32 ivpu_hw_btrs_dpu_freq_get(struct ivpu_device *vdev);
 bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq);
 bool ivpu_hw_btrs_irq_handler_lnl(struct ivpu_device *vdev, int irq);
 int ivpu_hw_btrs_dct_get_request(struct ivpu_device *vdev, bool *enable);
 void ivpu_hw_btrs_dct_set_status(struct ivpu_device *vdev, bool enable, u32 dct_percent);
-u32 ivpu_hw_btrs_pll_freq_get(struct ivpu_device *vdev);
-u32 ivpu_hw_btrs_ratio_to_freq(struct ivpu_device *vdev, u32 ratio);
 u32 ivpu_hw_btrs_telemetry_offset_get(struct ivpu_device *vdev);
 u32 ivpu_hw_btrs_telemetry_size_get(struct ivpu_device *vdev);
 u32 ivpu_hw_btrs_telemetry_enable_get(struct ivpu_device *vdev);
diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c
index 0e096fd9b95d..39f83225c181 100644
--- a/drivers/accel/ivpu/ivpu_ipc.c
+++ b/drivers/accel/ivpu/ivpu_ipc.c
@@ -302,7 +302,8 @@ ivpu_ipc_send_receive_internal(struct ivpu_device *vdev, struct vpu_jsm_msg *req
 	struct ivpu_ipc_consumer cons;
 	int ret;
 
-	drm_WARN_ON(&vdev->drm, pm_runtime_status_suspended(vdev->drm.dev));
+	drm_WARN_ON(&vdev->drm, pm_runtime_status_suspended(vdev->drm.dev) &&
+		    pm_runtime_enabled(vdev->drm.dev));
 
 	ivpu_ipc_consumer_add(vdev, &cons, channel, NULL);
 
diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c
index 004059e4f1e8..863e3cd6ace5 100644
--- a/drivers/accel/ivpu/ivpu_job.c
+++ b/drivers/accel/ivpu/ivpu_job.c
@@ -470,8 +470,8 @@ static void ivpu_job_destroy(struct ivpu_job *job)
 	struct ivpu_device *vdev = job->vdev;
 	u32 i;
 
-	ivpu_dbg(vdev, JOB, "Job destroyed: id %3u ctx %2d engine %d",
-		 job->job_id, job->file_priv->ctx.id, job->engine_idx);
+	ivpu_dbg(vdev, JOB, "Job destroyed: id %3u ctx %2d cmdq_id %u engine %d",
+		 job->job_id, job->file_priv->ctx.id, job->cmdq_id, job->engine_idx);
 
 	for (i = 0; i < job->bo_count; i++)
 		if (job->bos[i])
@@ -564,8 +564,8 @@ static int ivpu_job_signal_and_destroy(struct ivpu_device *vdev, u32 job_id, u32
 	dma_fence_signal(job->done_fence);
 
 	trace_job("done", job);
-	ivpu_dbg(vdev, JOB, "Job complete:  id %3u ctx %2d engine %d status 0x%x\n",
-		 job->job_id, job->file_priv->ctx.id, job->engine_idx, job_status);
+	ivpu_dbg(vdev, JOB, "Job complete:  id %3u ctx %2d cmdq_id %u engine %d status 0x%x\n",
+		 job->job_id, job->file_priv->ctx.id, job->cmdq_id, job->engine_idx, job_status);
 
 	ivpu_job_destroy(job);
 	ivpu_stop_job_timeout_detection(vdev);
@@ -664,8 +664,8 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id)
 	}
 
 	trace_job("submit", job);
-	ivpu_dbg(vdev, JOB, "Job submitted: id %3u ctx %2d engine %d prio %d addr 0x%llx next %d\n",
-		 job->job_id, file_priv->ctx.id, job->engine_idx, cmdq->priority,
+	ivpu_dbg(vdev, JOB, "Job submitted: id %3u ctx %2d cmdq_id %u engine %d prio %d addr 0x%llx next %d\n",
+		 job->job_id, file_priv->ctx.id, cmdq->id, job->engine_idx, cmdq->priority,
 		 job->cmd_buf_vpu_addr, cmdq->jobq->header.tail);
 
 	mutex_unlock(&file_priv->lock);
@@ -777,7 +777,8 @@ static int ivpu_submit(struct drm_file *file, struct ivpu_file_priv *file_priv,
 		goto err_free_handles;
 	}
 
-	ivpu_dbg(vdev, JOB, "Submit ioctl: ctx %u buf_count %u\n", file_priv->ctx.id, buffer_count);
+	ivpu_dbg(vdev, JOB, "Submit ioctl: ctx %u cmdq_id %u buf_count %u\n",
+		 file_priv->ctx.id, cmdq_id, buffer_count);
 
 	job = ivpu_job_create(file_priv, engine, buffer_count);
 	if (!job) {
diff --git a/drivers/accel/ivpu/ivpu_ms.c b/drivers/accel/ivpu/ivpu_ms.c
index ffe7b10f8a76..2a043baf10ca 100644
--- a/drivers/accel/ivpu/ivpu_ms.c
+++ b/drivers/accel/ivpu/ivpu_ms.c
@@ -4,6 +4,7 @@
  */
 
 #include <drm/drm_file.h>
+#include <linux/pm_runtime.h>
 
 #include "ivpu_drv.h"
 #include "ivpu_gem.h"
@@ -44,6 +45,10 @@ int ivpu_ms_start_ioctl(struct drm_device *dev, void *data, struct drm_file *fil
 	    args->sampling_period_ns < MS_MIN_SAMPLE_PERIOD_NS)
 		return -EINVAL;
 
+	ret = ivpu_rpm_get(vdev);
+	if (ret < 0)
+		return ret;
+
 	mutex_lock(&file_priv->ms_lock);
 
 	if (get_instance_by_mask(file_priv, args->metric_group_mask)) {
@@ -96,6 +101,8 @@ err_free_ms:
 	kfree(ms);
 unlock:
 	mutex_unlock(&file_priv->ms_lock);
+
+	ivpu_rpm_put(vdev);
 	return ret;
 }
 
@@ -160,6 +167,10 @@ int ivpu_ms_get_data_ioctl(struct drm_device *dev, void *data, struct drm_file *
 	if (!args->metric_group_mask)
 		return -EINVAL;
 
+	ret = ivpu_rpm_get(vdev);
+	if (ret < 0)
+		return ret;
+
 	mutex_lock(&file_priv->ms_lock);
 
 	ms = get_instance_by_mask(file_priv, args->metric_group_mask);
@@ -187,6 +198,7 @@ int ivpu_ms_get_data_ioctl(struct drm_device *dev, void *data, struct drm_file *
 unlock:
 	mutex_unlock(&file_priv->ms_lock);
 
+	ivpu_rpm_put(vdev);
 	return ret;
 }
 
@@ -204,11 +216,17 @@ int ivpu_ms_stop_ioctl(struct drm_device *dev, void *data, struct drm_file *file
 {
 	struct ivpu_file_priv *file_priv = file->driver_priv;
 	struct drm_ivpu_metric_streamer_stop *args = data;
+	struct ivpu_device *vdev = file_priv->vdev;
 	struct ivpu_ms_instance *ms;
+	int ret;
 
 	if (!args->metric_group_mask)
 		return -EINVAL;
 
+	ret = ivpu_rpm_get(vdev);
+	if (ret < 0)
+		return ret;
+
 	mutex_lock(&file_priv->ms_lock);
 
 	ms = get_instance_by_mask(file_priv, args->metric_group_mask);
@@ -217,6 +235,7 @@ int ivpu_ms_stop_ioctl(struct drm_device *dev, void *data, struct drm_file *file
 
 	mutex_unlock(&file_priv->ms_lock);
 
+	ivpu_rpm_put(vdev);
 	return ms ? 0 : -EINVAL;
 }
 
@@ -281,6 +300,9 @@ unlock:
 void ivpu_ms_cleanup(struct ivpu_file_priv *file_priv)
 {
 	struct ivpu_ms_instance *ms, *tmp;
+	struct ivpu_device *vdev = file_priv->vdev;
+
+	pm_runtime_get_sync(vdev->drm.dev);
 
 	mutex_lock(&file_priv->ms_lock);
 
@@ -293,6 +315,8 @@ void ivpu_ms_cleanup(struct ivpu_file_priv *file_priv)
 		free_instance(file_priv, ms);
 
 	mutex_unlock(&file_priv->ms_lock);
+
+	pm_runtime_put_autosuspend(vdev->drm.dev);
 }
 
 void ivpu_ms_cleanup_all(struct ivpu_device *vdev)
diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c
index b5891e91f7ab..ea30db181cd7 100644
--- a/drivers/accel/ivpu/ivpu_pm.c
+++ b/drivers/accel/ivpu/ivpu_pm.c
@@ -34,6 +34,7 @@ module_param_named(tdr_timeout_ms, ivpu_tdr_timeout_ms, ulong, 0644);
 MODULE_PARM_DESC(tdr_timeout_ms, "Timeout for device hang detection, in milliseconds, 0 - default");
 
 #define PM_RESCHEDULE_LIMIT     5
+#define PM_TDR_HEARTBEAT_LIMIT  30
 
 static void ivpu_pm_prepare_cold_boot(struct ivpu_device *vdev)
 {
@@ -44,6 +45,7 @@ static void ivpu_pm_prepare_cold_boot(struct ivpu_device *vdev)
 	ivpu_fw_log_reset(vdev);
 	ivpu_fw_load(vdev);
 	fw->entry_point = fw->cold_boot_entry_point;
+	fw->last_heartbeat = 0;
 }
 
 static void ivpu_pm_prepare_warm_boot(struct ivpu_device *vdev)
@@ -189,7 +191,24 @@ static void ivpu_job_timeout_work(struct work_struct *work)
 {
 	struct ivpu_pm_info *pm = container_of(work, struct ivpu_pm_info, job_timeout_work.work);
 	struct ivpu_device *vdev = pm->vdev;
+	u64 heartbeat;
 
+	if (ivpu_jsm_get_heartbeat(vdev, 0, &heartbeat) || heartbeat <= vdev->fw->last_heartbeat) {
+		ivpu_err(vdev, "Job timeout detected, heartbeat not progressed\n");
+		goto recovery;
+	}
+
+	if (atomic_fetch_inc(&vdev->job_timeout_counter) > PM_TDR_HEARTBEAT_LIMIT) {
+		ivpu_err(vdev, "Job timeout detected, heartbeat limit exceeded\n");
+		goto recovery;
+	}
+
+	vdev->fw->last_heartbeat = heartbeat;
+	ivpu_start_job_timeout_detection(vdev);
+	return;
+
+recovery:
+	atomic_set(&vdev->job_timeout_counter, 0);
 	ivpu_pm_trigger_recovery(vdev, "TDR");
 }
 
@@ -204,6 +223,7 @@ void ivpu_start_job_timeout_detection(struct ivpu_device *vdev)
 void ivpu_stop_job_timeout_detection(struct ivpu_device *vdev)
 {
 	cancel_delayed_work_sync(&vdev->pm->job_timeout_work);
+	atomic_set(&vdev->job_timeout_counter, 0);
 }
 
 int ivpu_pm_suspend_cb(struct device *dev)
@@ -428,16 +448,17 @@ int ivpu_pm_dct_enable(struct ivpu_device *vdev, u8 active_percent)
 	active_us = (DCT_PERIOD_US * active_percent) / 100;
 	inactive_us = DCT_PERIOD_US - active_us;
 
+	vdev->pm->dct_active_percent = active_percent;
+
+	ivpu_dbg(vdev, PM, "DCT requested %u%% (D0: %uus, D0i2: %uus)\n",
+		 active_percent, active_us, inactive_us);
+
 	ret = ivpu_jsm_dct_enable(vdev, active_us, inactive_us);
 	if (ret) {
 		ivpu_err_ratelimited(vdev, "Failed to enable DCT: %d\n", ret);
 		return ret;
 	}
 
-	vdev->pm->dct_active_percent = active_percent;
-
-	ivpu_dbg(vdev, PM, "DCT set to %u%% (D0: %uus, D0i2: %uus)\n",
-		 active_percent, active_us, inactive_us);
 	return 0;
 }
 
@@ -445,15 +466,16 @@ int ivpu_pm_dct_disable(struct ivpu_device *vdev)
 {
 	int ret;
 
+	vdev->pm->dct_active_percent = 0;
+
+	ivpu_dbg(vdev, PM, "DCT requested to be disabled\n");
+
 	ret = ivpu_jsm_dct_disable(vdev);
 	if (ret) {
 		ivpu_err_ratelimited(vdev, "Failed to disable DCT: %d\n", ret);
 		return ret;
 	}
 
-	vdev->pm->dct_active_percent = 0;
-
-	ivpu_dbg(vdev, PM, "DCT disabled\n");
 	return 0;
 }
 
@@ -466,7 +488,7 @@ void ivpu_pm_irq_dct_work_fn(struct work_struct *work)
 	if (ivpu_hw_btrs_dct_get_request(vdev, &enable))
 		return;
 
-	if (vdev->pm->dct_active_percent)
+	if (enable)
 		ret = ivpu_pm_dct_enable(vdev, DCT_DEFAULT_ACTIVE_PERCENT);
 	else
 		ret = ivpu_pm_dct_disable(vdev);
diff --git a/drivers/accel/ivpu/ivpu_sysfs.c b/drivers/accel/ivpu/ivpu_sysfs.c
index 97102feaf8dd..268ab7744a8b 100644
--- a/drivers/accel/ivpu/ivpu_sysfs.c
+++ b/drivers/accel/ivpu/ivpu_sysfs.c
@@ -1,10 +1,12 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * Copyright (C) 2024 Intel Corporation
+ * Copyright (C) 2024-2025 Intel Corporation
  */
 
 #include <linux/device.h>
 #include <linux/err.h>
+#include <linux/pm_runtime.h>
+#include <linux/units.h>
 
 #include "ivpu_drv.h"
 #include "ivpu_gem.h"
@@ -90,10 +92,55 @@ sched_mode_show(struct device *dev, struct device_attribute *attr, char *buf)
 
 static DEVICE_ATTR_RO(sched_mode);
 
+/**
+ * DOC: npu_max_frequency
+ *
+ * The npu_max_frequency shows maximum frequency in MHz of the NPU's data
+ * processing unit
+ */
+static ssize_t
+npu_max_frequency_mhz_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct drm_device *drm = dev_get_drvdata(dev);
+	struct ivpu_device *vdev = to_ivpu_device(drm);
+	u32 freq = ivpu_hw_dpu_max_freq_get(vdev);
+
+	return sysfs_emit(buf, "%lu\n", freq / HZ_PER_MHZ);
+}
+
+static DEVICE_ATTR_RO(npu_max_frequency_mhz);
+
+/**
+ * DOC: npu_current_frequency_mhz
+ *
+ * The npu_current_frequency_mhz shows current frequency in MHz of the NPU's
+ * data processing unit
+ */
+static ssize_t
+npu_current_frequency_mhz_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct drm_device *drm = dev_get_drvdata(dev);
+	struct ivpu_device *vdev = to_ivpu_device(drm);
+	u32 freq = 0;
+
+	/* Read frequency only if device is active, otherwise frequency is 0 */
+	if (pm_runtime_get_if_active(vdev->drm.dev) > 0) {
+		freq = ivpu_hw_dpu_freq_get(vdev);
+
+		pm_runtime_put_autosuspend(vdev->drm.dev);
+	}
+
+	return sysfs_emit(buf, "%lu\n", freq / HZ_PER_MHZ);
+}
+
+static DEVICE_ATTR_RO(npu_current_frequency_mhz);
+
 static struct attribute *ivpu_dev_attrs[] = {
 	&dev_attr_npu_busy_time_us.attr,
 	&dev_attr_npu_memory_utilization.attr,
 	&dev_attr_sched_mode.attr,
+	&dev_attr_npu_max_frequency_mhz.attr,
+	&dev_attr_npu_current_frequency_mhz.attr,
 	NULL,
 };
 
diff --git a/drivers/accel/ivpu/vpu_boot_api.h b/drivers/accel/ivpu/vpu_boot_api.h
index 908e68ea1c39..218468bbbcad 100644
--- a/drivers/accel/ivpu/vpu_boot_api.h
+++ b/drivers/accel/ivpu/vpu_boot_api.h
@@ -26,7 +26,7 @@
  * Minor version changes when API backward compatibility is preserved.
  * Resets to 0 if Major version is incremented.
  */
-#define VPU_BOOT_API_VER_MINOR 26
+#define VPU_BOOT_API_VER_MINOR 28
 
 /*
  * API header changed (field names, documentation, formatting) but API itself has not been changed
@@ -76,8 +76,15 @@ struct vpu_firmware_header {
 	 * submission queue size and device capabilities.
 	 */
 	u32 preemption_buffer_2_size;
+	/*
+	 * Maximum preemption buffer size that the FW can use: no need for the host
+	 * driver to allocate more space than that specified by these fields.
+	 * A value of 0 means no declared limit.
+	 */
+	u32 preemption_buffer_1_max_size;
+	u32 preemption_buffer_2_max_size;
 	/* Space reserved for future preemption-related fields. */
-	u32 preemption_reserved[6];
+	u32 preemption_reserved[4];
 	/* FW image read only section start address, 4KB aligned */
 	u64 ro_section_start_address;
 	/* FW image read only section size, 4KB aligned */
@@ -134,7 +141,7 @@ enum vpu_trace_destination {
 /*
  * Processor bit shifts (for loggable HW components).
  */
-#define VPU_TRACE_PROC_BIT_ARM	     0
+#define VPU_TRACE_PROC_BIT_RESERVED  0
 #define VPU_TRACE_PROC_BIT_LRT	     1
 #define VPU_TRACE_PROC_BIT_LNN	     2
 #define VPU_TRACE_PROC_BIT_SHV_0     3
diff --git a/drivers/accel/ivpu/vpu_jsm_api.h b/drivers/accel/ivpu/vpu_jsm_api.h
index 7215c144158c..4b6b2b3d2583 100644
--- a/drivers/accel/ivpu/vpu_jsm_api.h
+++ b/drivers/accel/ivpu/vpu_jsm_api.h
@@ -22,7 +22,7 @@
 /*
  * Minor version changes when API backward compatibility is preserved.
  */
-#define VPU_JSM_API_VER_MINOR 25
+#define VPU_JSM_API_VER_MINOR 29
 
 /*
  * API header changed (field names, documentation, formatting) but API itself has not been changed
@@ -53,8 +53,7 @@
  * Engine indexes.
  */
 #define VPU_ENGINE_COMPUTE 0
-#define VPU_ENGINE_COPY	   1
-#define VPU_ENGINE_NB	   2
+#define VPU_ENGINE_NB	   1
 
 /*
  * VPU status values.
@@ -126,11 +125,13 @@ enum {
 	 * When set, indicates that job queue uses native fences (as inline commands
 	 * in job queue). Such queues may also use legacy fences (as commands in batch buffers).
 	 * When cleared, indicates the job queue only uses legacy fences.
-	 * NOTE: For queues using native fences, VPU expects that all jobs in the queue
-	 * are immediately followed by an inline command object. This object is expected
-	 * to be a fence signal command in most cases, but can also be a NOP in case the host
-	 * does not need per-job fence signalling. Other inline commands objects can be
-	 * inserted between "job and inline command" pairs.
+	 * NOTES:
+	 *   1. For queues using native fences, VPU expects that all jobs in the queue
+	 *      are immediately followed by an inline command object. This object is expected
+	 *      to be a fence signal command in most cases, but can also be a NOP in case the host
+	 *      does not need per-job fence signalling. Other inline commands objects can be
+	 *      inserted between "job and inline command" pairs.
+	 *  2. Native fence queues are only supported on VPU 40xx onwards.
 	 */
 	VPU_JOB_QUEUE_FLAGS_USE_NATIVE_FENCE_MASK = (1 << 1U),
 
@@ -275,6 +276,8 @@ struct vpu_inline_cmd {
 			u64 value;
 			/* User VA of the log buffer in which to add log entry on completion. */
 			u64 log_buffer_va;
+			/* NPU private data. */
+			u64 npu_private_data;
 		} fence;
 		/* Other commands do not have a payload. */
 		/* Payload definition for future inline commands can be inserted here. */
@@ -791,12 +794,22 @@ struct vpu_jsm_metric_streamer_update {
 	/** Metric group mask that identifies metric streamer instance. */
 	u64 metric_group_mask;
 	/**
-	 * Address and size of the buffer where the VPU will write metric data. If
-	 * the buffer address is 0 or same as the currently used buffer the VPU will
-	 * continue writing metric data to the current buffer. In this case the
-	 * buffer size is ignored and the size of the current buffer is unchanged.
-	 * If the address is non-zero and differs from the current buffer address the
-	 * VPU will immediately switch data collection to the new buffer.
+	 * Address and size of the buffer where the VPU will write metric data.
+	 * This member dictates how the update operation should perform:
+	 * 1. client needs information about the number of collected samples and the
+	 *   amount of data written to the current buffer
+	 * 2. client wants to switch to a new buffer
+	 *
+	 * Case 1. is identified by the buffer address being 0 or the same as the
+	 * currently used buffer address. In this case the buffer size is ignored and
+	 * the size of the current buffer is unchanged. The VPU will return an update
+	 * in the vpu_jsm_metric_streamer_done structure. The internal writing position
+	 * into the buffer is not changed.
+	 *
+	 * Case 2. is identified by the address being non-zero and differs from the
+	 * current buffer address. The VPU will immediately switch data collection to
+	 * the new buffer. Then the VPU will return an update in the
+	 * vpu_jsm_metric_streamer_done structure.
 	 */
 	u64 buffer_addr;
 	u64 buffer_size;
@@ -934,6 +947,7 @@ struct vpu_ipc_msg_payload_hws_priority_band_setup {
 	/*
 	 * Default quantum in 100ns units for scheduling across processes
 	 * within a priority band
+	 * Minimum value supported by NPU is 1ms (10000 in 100ns units).
 	 */
 	u32 process_quantum[VPU_HWS_NUM_PRIORITY_BANDS];
 	/*
@@ -946,8 +960,10 @@ struct vpu_ipc_msg_payload_hws_priority_band_setup {
 	 * in situations when it's starved by the focus band.
 	 */
 	u32 normal_band_percentage;
-	/* Reserved */
-	u32 reserved_0;
+	/*
+	 * TDR timeout value in milliseconds. Default value of 0 meaning no timeout.
+	 */
+	u32 tdr_timeout;
 };
 
 /*
@@ -1024,7 +1040,10 @@ struct vpu_ipc_msg_payload_hws_set_context_sched_properties {
 	s32 in_process_priority;
 	/* Zero padding / Reserved */
 	u32 reserved_1;
-	/* Context quantum relative to other contexts of same priority in the same process */
+	/*
+	 * Context quantum relative to other contexts of same priority in the same process
+	 * Minimum value supported by NPU is 1ms (10000 in 100ns units).
+	 */
 	u64 context_quantum;
 	/* Grace period when preempting context of the same priority within the same process */
 	u64 grace_period_same_priority;
diff --git a/drivers/accel/qaic/qaic_data.c b/drivers/accel/qaic/qaic_data.c
index 43aba57b48f0..1bce1af7c72c 100644
--- a/drivers/accel/qaic/qaic_data.c
+++ b/drivers/accel/qaic/qaic_data.c
@@ -609,7 +609,7 @@ static int qaic_gem_object_mmap(struct drm_gem_object *obj, struct vm_area_struc
 	struct scatterlist *sg;
 	int ret = 0;
 
-	if (obj->import_attach)
+	if (drm_gem_is_imported(obj))
 		return -EINVAL;
 
 	for (sg = bo->sgt->sgl; sg; sg = sg_next(sg)) {
@@ -630,7 +630,7 @@ static void qaic_free_object(struct drm_gem_object *obj)
 {
 	struct qaic_bo *bo = to_qaic_bo(obj);
 
-	if (obj->import_attach) {
+	if (drm_gem_is_imported(obj)) {
 		/* DMABUF/PRIME Path */
 		drm_prime_gem_destroy(obj, NULL);
 	} else {
@@ -870,7 +870,7 @@ static int qaic_prepare_bo(struct qaic_device *qdev, struct qaic_bo *bo,
 {
 	int ret;
 
-	if (bo->base.import_attach)
+	if (drm_gem_is_imported(&bo->base))
 		ret = qaic_prepare_import_bo(bo, hdr);
 	else
 		ret = qaic_prepare_export_bo(qdev, bo, hdr);
@@ -894,7 +894,7 @@ static void qaic_unprepare_export_bo(struct qaic_device *qdev, struct qaic_bo *b
 
 static void qaic_unprepare_bo(struct qaic_device *qdev, struct qaic_bo *bo)
 {
-	if (bo->base.import_attach)
+	if (drm_gem_is_imported(&bo->base))
 		qaic_unprepare_import_bo(bo);
 	else
 		qaic_unprepare_export_bo(qdev, bo);
diff --git a/drivers/accel/qaic/qaic_debugfs.c b/drivers/accel/qaic/qaic_debugfs.c
index ba0cf2f94732..a991b8198dc4 100644
--- a/drivers/accel/qaic/qaic_debugfs.c
+++ b/drivers/accel/qaic/qaic_debugfs.c
@@ -240,7 +240,6 @@ static int qaic_bootlog_mhi_probe(struct mhi_device *mhi_dev, const struct mhi_d
 mhi_unprepare:
 	mhi_unprepare_from_transfer(mhi_dev);
 destroy_workqueue:
-	flush_workqueue(qdev->bootlog_wq);
 	destroy_workqueue(qdev->bootlog_wq);
 out:
 	return ret;
@@ -253,7 +252,6 @@ static void qaic_bootlog_mhi_remove(struct mhi_device *mhi_dev)
 	qdev = dev_get_drvdata(&mhi_dev->dev);
 
 	mhi_unprepare_from_transfer(qdev->bootlog_ch);
-	flush_workqueue(qdev->bootlog_wq);
 	destroy_workqueue(qdev->bootlog_wq);
 	qdev->bootlog_ch = NULL;
 }
diff --git a/drivers/acpi/acpica/aclocal.h b/drivers/acpi/acpica/aclocal.h
index 6f4fe47c955b..6481c48c22bb 100644
--- a/drivers/acpi/acpica/aclocal.h
+++ b/drivers/acpi/acpica/aclocal.h
@@ -293,7 +293,7 @@ acpi_status (*acpi_internal_method) (struct acpi_walk_state * walk_state);
  * expected_return_btypes - Allowed type(s) for the return value
  */
 struct acpi_name_info {
-	char name[ACPI_NAMESEG_SIZE];
+	char name[ACPI_NAMESEG_SIZE] __nonstring;
 	u16 argument_list;
 	u8 expected_btypes;
 };
@@ -370,7 +370,7 @@ typedef acpi_status (*acpi_object_converter) (struct acpi_namespace_node *
 					      converted_object);
 
 struct acpi_simple_repair_info {
-	char name[ACPI_NAMESEG_SIZE];
+	char name[ACPI_NAMESEG_SIZE] __nonstring;
 	u32 unexpected_btypes;
 	u32 package_index;
 	acpi_object_converter object_converter;
diff --git a/drivers/acpi/acpica/nsrepair2.c b/drivers/acpi/acpica/nsrepair2.c
index 1bb7b71f07f1..330b5e4711da 100644
--- a/drivers/acpi/acpica/nsrepair2.c
+++ b/drivers/acpi/acpica/nsrepair2.c
@@ -25,7 +25,7 @@ acpi_status (*acpi_repair_function) (struct acpi_evaluate_info * info,
 				     return_object_ptr);
 
 typedef struct acpi_repair_info {
-	char name[ACPI_NAMESEG_SIZE];
+	char name[ACPI_NAMESEG_SIZE] __nonstring;
 	acpi_repair_function repair_function;
 
 } acpi_repair_info;
diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c
index 90b09840536d..0a7026040188 100644
--- a/drivers/acpi/button.c
+++ b/drivers/acpi/button.c
@@ -458,7 +458,7 @@ static void acpi_button_notify(acpi_handle handle, u32 event, void *data)
 	acpi_pm_wakeup_event(&device->dev);
 
 	button = acpi_driver_data(device);
-	if (button->suspended)
+	if (button->suspended || event == ACPI_BUTTON_NOTIFY_WAKE)
 		return;
 
 	input = button->input;
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
index 8db09d81918f..3c5f34892734 100644
--- a/drivers/acpi/ec.c
+++ b/drivers/acpi/ec.c
@@ -2301,6 +2301,34 @@ static const struct dmi_system_id acpi_ec_no_wakeup[] = {
 			DMI_MATCH(DMI_PRODUCT_FAMILY, "103C_5336AN HP ZHAN 66 Pro"),
 		},
 	},
+	/*
+	 * Lenovo Legion Go S; touchscreen blocks HW sleep when woken up from EC
+	 * https://gitlab.freedesktop.org/drm/amd/-/issues/3929
+	 */
+	{
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "83L3"),
+		}
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "83N6"),
+		}
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "83Q2"),
+		}
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "83Q3"),
+		}
+	},
 	{ },
 };
 
diff --git a/drivers/acpi/pptt.c b/drivers/acpi/pptt.c
index a35dd0e41c27..f73ce6e13065 100644
--- a/drivers/acpi/pptt.c
+++ b/drivers/acpi/pptt.c
@@ -229,7 +229,7 @@ static int acpi_pptt_leaf_node(struct acpi_table_header *table_hdr,
 	node_entry = ACPI_PTR_DIFF(node, table_hdr);
 	entry = ACPI_ADD_PTR(struct acpi_subtable_header, table_hdr,
 			     sizeof(struct acpi_table_pptt));
-	proc_sz = sizeof(struct acpi_pptt_processor *);
+	proc_sz = sizeof(struct acpi_pptt_processor);
 
 	while ((unsigned long)entry + proc_sz < table_end) {
 		cpu_node = (struct acpi_pptt_processor *)entry;
@@ -270,7 +270,7 @@ static struct acpi_pptt_processor *acpi_find_processor_node(struct acpi_table_he
 	table_end = (unsigned long)table_hdr + table_hdr->length;
 	entry = ACPI_ADD_PTR(struct acpi_subtable_header, table_hdr,
 			     sizeof(struct acpi_table_pptt));
-	proc_sz = sizeof(struct acpi_pptt_processor *);
+	proc_sz = sizeof(struct acpi_pptt_processor);
 
 	/* find the processor structure associated with this cpuid */
 	while ((unsigned long)entry + proc_sz < table_end) {
diff --git a/drivers/android/binder.c b/drivers/android/binder.c
index 76052006bd87..5fc2c8ee61b1 100644
--- a/drivers/android/binder.c
+++ b/drivers/android/binder.c
@@ -6373,7 +6373,7 @@ static void print_binder_transaction_ilocked(struct seq_file *m,
 		seq_printf(m, " node %d", buffer->target_node->debug_id);
 	seq_printf(m, " size %zd:%zd offset %lx\n",
 		   buffer->data_size, buffer->offsets_size,
-		   proc->alloc.vm_start - buffer->user_data);
+		   buffer->user_data - proc->alloc.vm_start);
 }
 
 static void print_binder_work_ilocked(struct seq_file *m,
diff --git a/drivers/ata/libata-sata.c b/drivers/ata/libata-sata.c
index ba300cc0a3a3..2e4463d3a356 100644
--- a/drivers/ata/libata-sata.c
+++ b/drivers/ata/libata-sata.c
@@ -1510,6 +1510,8 @@ int ata_eh_get_ncq_success_sense(struct ata_link *link)
 	unsigned int err_mask, tag;
 	u8 *sense, sk = 0, asc = 0, ascq = 0;
 	u64 sense_valid, val;
+	u16 extended_sense;
+	bool aux_icc_valid;
 	int ret = 0;
 
 	err_mask = ata_read_log_page(dev, ATA_LOG_SENSE_NCQ, 0, buf, 2);
@@ -1529,6 +1531,8 @@ int ata_eh_get_ncq_success_sense(struct ata_link *link)
 
 	sense_valid = (u64)buf[8] | ((u64)buf[9] << 8) |
 		((u64)buf[10] << 16) | ((u64)buf[11] << 24);
+	extended_sense = get_unaligned_le16(&buf[14]);
+	aux_icc_valid = extended_sense & BIT(15);
 
 	ata_qc_for_each_raw(ap, qc, tag) {
 		if (!(qc->flags & ATA_QCFLAG_EH) ||
@@ -1556,6 +1560,17 @@ int ata_eh_get_ncq_success_sense(struct ata_link *link)
 			continue;
 		}
 
+		qc->result_tf.nsect = sense[6];
+		qc->result_tf.hob_nsect = sense[7];
+		qc->result_tf.lbal = sense[8];
+		qc->result_tf.lbam = sense[9];
+		qc->result_tf.lbah = sense[10];
+		qc->result_tf.hob_lbal = sense[11];
+		qc->result_tf.hob_lbam = sense[12];
+		qc->result_tf.hob_lbah = sense[13];
+		if (aux_icc_valid)
+			qc->result_tf.auxiliary = get_unaligned_le32(&sense[16]);
+
 		/* Set sense without also setting scsicmd->result */
 		scsi_build_sense_buffer(dev->flags & ATA_DFLAG_D_SENSE,
 					qc->scsicmd->sense_buffer, sk,
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 2796c0da8257..c0eb8c67a9ff 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -2453,8 +2453,8 @@ static unsigned int ata_msense_control_ata_feature(struct ata_device *dev,
 	 */
 	put_unaligned_be16(ATA_FEATURE_SUB_MPAGE_LEN - 4, &buf[2]);
 
-	if (dev->flags & ATA_DFLAG_CDL)
-		buf[4] = 0x02; /* Support T2A and T2B pages */
+	if (dev->flags & ATA_DFLAG_CDL_ENABLED)
+		buf[4] = 0x02; /* T2A and T2B pages enabled */
 	else
 		buf[4] = 0;
 
@@ -3886,12 +3886,11 @@ static int ata_mselect_control_spg0(struct ata_queued_cmd *qc,
 }
 
 /*
- * Translate MODE SELECT control mode page, sub-pages f2h (ATA feature mode
+ * Translate MODE SELECT control mode page, sub-page f2h (ATA feature mode
  * page) into a SET FEATURES command.
  */
-static unsigned int ata_mselect_control_ata_feature(struct ata_queued_cmd *qc,
-						    const u8 *buf, int len,
-						    u16 *fp)
+static int ata_mselect_control_ata_feature(struct ata_queued_cmd *qc,
+					   const u8 *buf, int len, u16 *fp)
 {
 	struct ata_device *dev = qc->dev;
 	struct ata_taskfile *tf = &qc->tf;
@@ -3909,17 +3908,27 @@ static unsigned int ata_mselect_control_ata_feature(struct ata_queued_cmd *qc,
 	/* Check cdl_ctrl */
 	switch (buf[0] & 0x03) {
 	case 0:
-		/* Disable CDL */
+		/* Disable CDL if it is enabled */
+		if (!(dev->flags & ATA_DFLAG_CDL_ENABLED))
+			return 0;
+		ata_dev_dbg(dev, "Disabling CDL\n");
 		cdl_action = 0;
 		dev->flags &= ~ATA_DFLAG_CDL_ENABLED;
 		break;
 	case 0x02:
-		/* Enable CDL T2A/T2B: NCQ priority must be disabled */
+		/*
+		 * Enable CDL if not already enabled. Since this is mutually
+		 * exclusive with NCQ priority, allow this only if NCQ priority
+		 * is disabled.
+		 */
+		if (dev->flags & ATA_DFLAG_CDL_ENABLED)
+			return 0;
 		if (dev->flags & ATA_DFLAG_NCQ_PRIO_ENABLED) {
 			ata_dev_err(dev,
 				"NCQ priority must be disabled to enable CDL\n");
 			return -EINVAL;
 		}
+		ata_dev_dbg(dev, "Enabling CDL\n");
 		cdl_action = 1;
 		dev->flags |= ATA_DFLAG_CDL_ENABLED;
 		break;
diff --git a/drivers/ata/pata_pxa.c b/drivers/ata/pata_pxa.c
index 434f380114af..03dbaf4a13a7 100644
--- a/drivers/ata/pata_pxa.c
+++ b/drivers/ata/pata_pxa.c
@@ -223,10 +223,16 @@ static int pxa_ata_probe(struct platform_device *pdev)
 
 	ap->ioaddr.cmd_addr	= devm_ioremap(&pdev->dev, cmd_res->start,
 						resource_size(cmd_res));
+	if (!ap->ioaddr.cmd_addr)
+		return -ENOMEM;
 	ap->ioaddr.ctl_addr	= devm_ioremap(&pdev->dev, ctl_res->start,
 						resource_size(ctl_res));
+	if (!ap->ioaddr.ctl_addr)
+		return -ENOMEM;
 	ap->ioaddr.bmdma_addr	= devm_ioremap(&pdev->dev, dma_res->start,
 						resource_size(dma_res));
+	if (!ap->ioaddr.bmdma_addr)
+		return -ENOMEM;
 
 	/*
 	 * Adjust register offsets
diff --git a/drivers/ata/sata_sx4.c b/drivers/ata/sata_sx4.c
index a482741eb181..c3042eca6332 100644
--- a/drivers/ata/sata_sx4.c
+++ b/drivers/ata/sata_sx4.c
@@ -1117,9 +1117,14 @@ static int pdc20621_prog_dimm0(struct ata_host *host)
 	mmio += PDC_CHIP0_OFS;
 
 	for (i = 0; i < ARRAY_SIZE(pdc_i2c_read_data); i++)
-		pdc20621_i2c_read(host, PDC_DIMM0_SPD_DEV_ADDRESS,
-				  pdc_i2c_read_data[i].reg,
-				  &spd0[pdc_i2c_read_data[i].ofs]);
+		if (!pdc20621_i2c_read(host, PDC_DIMM0_SPD_DEV_ADDRESS,
+				       pdc_i2c_read_data[i].reg,
+				       &spd0[pdc_i2c_read_data[i].ofs])) {
+			dev_err(host->dev,
+				"Failed in i2c read at index %d: device=%#x, reg=%#x\n",
+				i, PDC_DIMM0_SPD_DEV_ADDRESS, pdc_i2c_read_data[i].reg);
+			return -EIO;
+		}
 
 	data |= (spd0[4] - 8) | ((spd0[21] != 0) << 3) | ((spd0[3]-11) << 4);
 	data |= ((spd0[17] / 4) << 6) | ((spd0[5] / 2) << 7) |
@@ -1284,6 +1289,8 @@ static unsigned int pdc20621_dimm_init(struct ata_host *host)
 
 	/* Programming DIMM0 Module Control Register (index_CID0:80h) */
 	size = pdc20621_prog_dimm0(host);
+	if (size < 0)
+		return size;
 	dev_dbg(host->dev, "Local DIMM Size = %dMB\n", size);
 
 	/* Programming DIMM Module Global Control Register (index_CID0:88h) */
diff --git a/drivers/base/auxiliary.c b/drivers/base/auxiliary.c
index afa4df4c5a3f..95717d509ca9 100644
--- a/drivers/base/auxiliary.c
+++ b/drivers/base/auxiliary.c
@@ -156,6 +156,16 @@
  *		},
  *		.ops = my_custom_ops,
  *	};
+ *
+ * Please note that such custom ops approach is valid, but it is hard to implement
+ * it right without global locks per-device to protect from auxiliary_drv removal
+ * during call to that ops. In addition, this implementation lacks proper module
+ * dependency, which causes to load/unload races between auxiliary parent and devices
+ * modules.
+ *
+ * The most easiest way to provide these ops reliably without needing to
+ * have a lock is to EXPORT_SYMBOL*() them and rely on already existing
+ * modules infrastructure for validity and correct dependencies chains.
  */
 
 static const struct auxiliary_device_id *auxiliary_match_id(const struct auxiliary_device_id *id,
diff --git a/drivers/base/base.h b/drivers/base/base.h
index 0042e4774b0c..123031a757d9 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -73,6 +73,7 @@ static inline void subsys_put(struct subsys_private *sp)
 		kset_put(&sp->subsys);
 }
 
+struct subsys_private *bus_to_subsys(const struct bus_type *bus);
 struct subsys_private *class_to_subsys(const struct class *class);
 
 struct driver_private {
@@ -180,6 +181,22 @@ int driver_add_groups(const struct device_driver *drv, const struct attribute_gr
 void driver_remove_groups(const struct device_driver *drv, const struct attribute_group **groups);
 void device_driver_detach(struct device *dev);
 
+static inline void device_set_driver(struct device *dev, const struct device_driver *drv)
+{
+	/*
+	 * Majority (all?) read accesses to dev->driver happens either
+	 * while holding device lock or in bus/driver code that is only
+	 * invoked when the device is bound to a driver and there is no
+	 * concern of the pointer being changed while it is being read.
+	 * However when reading device's uevent file we read driver pointer
+	 * without taking device lock (so we do not block there for
+	 * arbitrary amount of time). We use WRITE_ONCE() here to prevent
+	 * tearing so that READ_ONCE() can safely be used in uevent code.
+	 */
+	// FIXME - this cast should not be needed "soon"
+	WRITE_ONCE(dev->driver, (struct device_driver *)drv);
+}
+
 int devres_release_all(struct device *dev);
 void device_block_probing(void);
 void device_unblock_probing(void);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 5ea3b03af9ba..5e75e1bce551 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -57,7 +57,7 @@ static int __must_check bus_rescan_devices_helper(struct device *dev,
  * NULL.  A call to subsys_put() must be done when finished with the pointer in
  * order for it to be properly freed.
  */
-static struct subsys_private *bus_to_subsys(const struct bus_type *bus)
+struct subsys_private *bus_to_subsys(const struct bus_type *bus)
 {
 	struct subsys_private *sp = NULL;
 	struct kobject *kobj;
diff --git a/drivers/base/core.c b/drivers/base/core.c
index d2f9d3a59d6b..cbc0099d8ef2 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -2624,6 +2624,35 @@ static const char *dev_uevent_name(const struct kobject *kobj)
 	return NULL;
 }
 
+/*
+ * Try filling "DRIVER=<name>" uevent variable for a device. Because this
+ * function may race with binding and unbinding the device from a driver,
+ * we need to be careful. Binding is generally safe, at worst we miss the
+ * fact that the device is already bound to a driver (but the driver
+ * information that is delivered through uevents is best-effort, it may
+ * become obsolete as soon as it is generated anyways). Unbinding is more
+ * risky as driver pointer is transitioning to NULL, so READ_ONCE() should
+ * be used to make sure we are dealing with the same pointer, and to
+ * ensure that driver structure is not going to disappear from under us
+ * we take bus' drivers klist lock. The assumption that only registered
+ * driver can be bound to a device, and to unregister a driver bus code
+ * will take the same lock.
+ */
+static void dev_driver_uevent(const struct device *dev, struct kobj_uevent_env *env)
+{
+	struct subsys_private *sp = bus_to_subsys(dev->bus);
+
+	if (sp) {
+		scoped_guard(spinlock, &sp->klist_drivers.k_lock) {
+			struct device_driver *drv = READ_ONCE(dev->driver);
+			if (drv)
+				add_uevent_var(env, "DRIVER=%s", drv->name);
+		}
+
+		subsys_put(sp);
+	}
+}
+
 static int dev_uevent(const struct kobject *kobj, struct kobj_uevent_env *env)
 {
 	const struct device *dev = kobj_to_dev(kobj);
@@ -2655,8 +2684,8 @@ static int dev_uevent(const struct kobject *kobj, struct kobj_uevent_env *env)
 	if (dev->type && dev->type->name)
 		add_uevent_var(env, "DEVTYPE=%s", dev->type->name);
 
-	if (dev->driver)
-		add_uevent_var(env, "DRIVER=%s", dev->driver->name);
+	/* Add "DRIVER=%s" variable if the device is bound to a driver */
+	dev_driver_uevent(dev, env);
 
 	/* Add common DT information about the device */
 	of_device_uevent(dev, env);
@@ -2726,11 +2755,8 @@ static ssize_t uevent_show(struct device *dev, struct device_attribute *attr,
 	if (!env)
 		return -ENOMEM;
 
-	/* Synchronize with really_probe() */
-	device_lock(dev);
 	/* let the kset specific function add its keys */
 	retval = kset->uevent_ops->uevent(&dev->kobj, env);
-	device_unlock(dev);
 	if (retval)
 		goto out;
 
@@ -3700,7 +3726,7 @@ done:
 	device_pm_remove(dev);
 	dpm_sysfs_remove(dev);
  DPMError:
-	dev->driver = NULL;
+	device_set_driver(dev, NULL);
 	bus_remove_device(dev);
  BusError:
 	device_remove_attrs(dev);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index f0e4b4aba885..b526e0e0f52d 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -550,7 +550,7 @@ static void device_unbind_cleanup(struct device *dev)
 	arch_teardown_dma_ops(dev);
 	kfree(dev->dma_range_map);
 	dev->dma_range_map = NULL;
-	dev->driver = NULL;
+	device_set_driver(dev, NULL);
 	dev_set_drvdata(dev, NULL);
 	if (dev->pm_domain && dev->pm_domain->dismiss)
 		dev->pm_domain->dismiss(dev);
@@ -629,8 +629,7 @@ static int really_probe(struct device *dev, const struct device_driver *drv)
 	}
 
 re_probe:
-	// FIXME - this cast should not be needed "soon"
-	dev->driver = (struct device_driver *)drv;
+	device_set_driver(dev, drv);
 
 	/* If using pinctrl, bind pins now before probing */
 	ret = pinctrl_bind_pins(dev);
@@ -1014,7 +1013,7 @@ static int __device_attach(struct device *dev, bool allow_async)
 		if (ret == 0)
 			ret = 1;
 		else {
-			dev->driver = NULL;
+			device_set_driver(dev, NULL);
 			ret = 0;
 		}
 	} else {
diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c
index 6dd1a8860f1c..31bfb3194b4c 100644
--- a/drivers/base/devtmpfs.c
+++ b/drivers/base/devtmpfs.c
@@ -296,7 +296,7 @@ static int delete_path(const char *nodepath)
 	return err;
 }
 
-static int dev_mynode(struct device *dev, struct inode *inode, struct kstat *stat)
+static int dev_mynode(struct device *dev, struct inode *inode)
 {
 	/* did we create it */
 	if (inode->i_private != &thread)
@@ -304,13 +304,13 @@ static int dev_mynode(struct device *dev, struct inode *inode, struct kstat *sta
 
 	/* does the dev_t match */
 	if (is_blockdev(dev)) {
-		if (!S_ISBLK(stat->mode))
+		if (!S_ISBLK(inode->i_mode))
 			return 0;
 	} else {
-		if (!S_ISCHR(stat->mode))
+		if (!S_ISCHR(inode->i_mode))
 			return 0;
 	}
-	if (stat->rdev != dev->devt)
+	if (inode->i_rdev != dev->devt)
 		return 0;
 
 	/* ours */
@@ -321,20 +321,16 @@ static int handle_remove(const char *nodename, struct device *dev)
 {
 	struct path parent;
 	struct dentry *dentry;
-	struct kstat stat;
-	struct path p;
+	struct inode *inode;
 	int deleted = 0;
-	int err;
+	int err = 0;
 
 	dentry = kern_path_locked(nodename, &parent);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	p.mnt = parent.mnt;
-	p.dentry = dentry;
-	err = vfs_getattr(&p, &stat, STATX_TYPE | STATX_MODE,
-			  AT_STATX_SYNC_AS_STAT);
-	if (!err && dev_mynode(dev, d_inode(dentry), &stat)) {
+	inode = d_inode(dentry);
+	if (dev_mynode(dev, inode)) {
 		struct iattr newattrs;
 		/*
 		 * before unlinking this node, reset permissions
@@ -342,7 +338,7 @@ static int handle_remove(const char *nodename, struct device *dev)
 		 */
 		newattrs.ia_uid = GLOBAL_ROOT_UID;
 		newattrs.ia_gid = GLOBAL_ROOT_GID;
-		newattrs.ia_mode = stat.mode & ~0777;
+		newattrs.ia_mode = inode->i_mode & ~0777;
 		newattrs.ia_valid =
 			ATTR_UID|ATTR_GID|ATTR_MODE;
 		inode_lock(d_inode(dentry));
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 8f3a41d9bfaa..19469e7f88c2 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -816,21 +816,6 @@ static int add_memory_block(unsigned long block_id, unsigned long state,
 	return 0;
 }
 
-static int __init add_boot_memory_block(unsigned long base_section_nr)
-{
-	unsigned long nr;
-
-	for_each_present_section_nr(base_section_nr, nr) {
-		if (nr >= (base_section_nr + sections_per_block))
-			break;
-
-		return add_memory_block(memory_block_id(base_section_nr),
-					MEM_ONLINE, NULL, NULL);
-	}
-
-	return 0;
-}
-
 static int add_hotplug_memory_block(unsigned long block_id,
 				    struct vmem_altmap *altmap,
 				    struct memory_group *group)
@@ -957,7 +942,7 @@ static const struct attribute_group *memory_root_attr_groups[] = {
 void __init memory_dev_init(void)
 {
 	int ret;
-	unsigned long block_sz, nr;
+	unsigned long block_sz, block_id, nr;
 
 	/* Validate the configured memory block size */
 	block_sz = memory_block_size_bytes();
@@ -970,15 +955,23 @@ void __init memory_dev_init(void)
 		panic("%s() failed to register subsystem: %d\n", __func__, ret);
 
 	/*
-	 * Create entries for memory sections that were found
-	 * during boot and have been initialized
+	 * Create entries for memory sections that were found during boot
+	 * and have been initialized. Use @block_id to track the last
+	 * handled block and initialize it to an invalid value (ULONG_MAX)
+	 * to bypass the block ID matching check for the first present
+	 * block so that it can be covered.
 	 */
-	for (nr = 0; nr <= __highest_present_section_nr;
-	     nr += sections_per_block) {
-		ret = add_boot_memory_block(nr);
-		if (ret)
-			panic("%s() failed to add memory block: %d\n", __func__,
-			      ret);
+	block_id = ULONG_MAX;
+	for_each_present_section_nr(0, nr) {
+		if (block_id != ULONG_MAX && memory_block_id(nr) == block_id)
+			continue;
+
+		block_id = memory_block_id(nr);
+		ret = add_memory_block(block_id, MEM_ONLINE, NULL, NULL);
+		if (ret) {
+			panic("%s() failed to add memory block: %d\n",
+			      __func__, ret);
+		}
 	}
 }
 
diff --git a/drivers/base/module.c b/drivers/base/module.c
index 5bc71bea883a..218aaa096455 100644
--- a/drivers/base/module.c
+++ b/drivers/base/module.c
@@ -42,16 +42,13 @@ int module_add_driver(struct module *mod, const struct device_driver *drv)
 	if (mod)
 		mk = &mod->mkobj;
 	else if (drv->mod_name) {
-		struct kobject *mkobj;
-
-		/* Lookup built-in module entry in /sys/modules */
-		mkobj = kset_find_obj(module_kset, drv->mod_name);
-		if (mkobj) {
-			mk = container_of(mkobj, struct module_kobject, kobj);
+		/* Lookup or create built-in module entry in /sys/modules */
+		mk = lookup_or_create_module_kobject(drv->mod_name);
+		if (mk) {
 			/* remember our module structure */
 			drv->p->mkobj = mk;
-			/* kset_find_obj took a reference */
-			kobject_put(mkobj);
+			/* lookup_or_create_module_kobject took a reference */
+			kobject_put(&mk->kobj);
 		}
 	}
 
diff --git a/drivers/base/swnode.c b/drivers/base/swnode.c
index b1726a3515f6..5c78fa6ae772 100644
--- a/drivers/base/swnode.c
+++ b/drivers/base/swnode.c
@@ -1080,6 +1080,7 @@ void software_node_notify(struct device *dev)
 	if (!swnode)
 		return;
 
+	kobject_get(&swnode->kobj);
 	ret = sysfs_create_link(&dev->kobj, &swnode->kobj, "software_node");
 	if (ret)
 		return;
@@ -1089,8 +1090,6 @@ void software_node_notify(struct device *dev)
 		sysfs_remove_link(&dev->kobj, "software_node");
 		return;
 	}
-
-	kobject_get(&swnode->kobj);
 }
 
 void software_node_notify_remove(struct device *dev)
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index a97f2c40c640..e48b24be45ee 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -367,7 +367,7 @@ config BLK_DEV_RBD
 	tristate "Rados block device (RBD)"
 	depends on INET && BLOCK
 	select CEPH_LIB
-	select LIBCRC32C
+	select CRC32
 	select CRYPTO_AES
 	select CRYPTO
 	help
@@ -388,12 +388,6 @@ config BLK_DEV_UBLK
 	  definition isn't finalized yet, and might change according to future
 	  requirement, so mark is as experimental now.
 
-	  Say Y if you want to get better performance because task_work_add()
-	  can be used in IO path for replacing io_uring cmd, which will become
-	  shared between IO tasks and ubq daemon, meantime task_work_add() can
-	  can handle batch more effectively, but task_work_add() isn't exported
-	  for module, so ublk has to be built to kernel.
-
 config BLKDEV_UBLK_LEGACY_OPCODES
 	bool "Support legacy command opcode"
 	depends on BLK_DEV_UBLK
diff --git a/drivers/block/drbd/Kconfig b/drivers/block/drbd/Kconfig
index 6fb4e38fca88..495a72da04c6 100644
--- a/drivers/block/drbd/Kconfig
+++ b/drivers/block/drbd/Kconfig
@@ -10,7 +10,7 @@ config BLK_DEV_DRBD
 	tristate "DRBD Distributed Replicated Block Device support"
 	depends on PROC_FS && INET
 	select LRU_CACHE
-	select LIBCRC32C
+	select CRC32
 	help
 
 	  NOTE: In order to authenticate connections you have to select
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 674527d770dc..46cba261075f 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -211,72 +211,6 @@ static void loop_set_size(struct loop_device *lo, loff_t size)
 		kobject_uevent(&disk_to_dev(lo->lo_disk)->kobj, KOBJ_CHANGE);
 }
 
-static int lo_write_bvec(struct file *file, struct bio_vec *bvec, loff_t *ppos)
-{
-	struct iov_iter i;
-	ssize_t bw;
-
-	iov_iter_bvec(&i, ITER_SOURCE, bvec, 1, bvec->bv_len);
-
-	bw = vfs_iter_write(file, &i, ppos, 0);
-
-	if (likely(bw ==  bvec->bv_len))
-		return 0;
-
-	printk_ratelimited(KERN_ERR
-		"loop: Write error at byte offset %llu, length %i.\n",
-		(unsigned long long)*ppos, bvec->bv_len);
-	if (bw >= 0)
-		bw = -EIO;
-	return bw;
-}
-
-static int lo_write_simple(struct loop_device *lo, struct request *rq,
-		loff_t pos)
-{
-	struct bio_vec bvec;
-	struct req_iterator iter;
-	int ret = 0;
-
-	rq_for_each_segment(bvec, rq, iter) {
-		ret = lo_write_bvec(lo->lo_backing_file, &bvec, &pos);
-		if (ret < 0)
-			break;
-		cond_resched();
-	}
-
-	return ret;
-}
-
-static int lo_read_simple(struct loop_device *lo, struct request *rq,
-		loff_t pos)
-{
-	struct bio_vec bvec;
-	struct req_iterator iter;
-	struct iov_iter i;
-	ssize_t len;
-
-	rq_for_each_segment(bvec, rq, iter) {
-		iov_iter_bvec(&i, ITER_DEST, &bvec, 1, bvec.bv_len);
-		len = vfs_iter_read(lo->lo_backing_file, &i, &pos, 0);
-		if (len < 0)
-			return len;
-
-		flush_dcache_page(bvec.bv_page);
-
-		if (len != bvec.bv_len) {
-			struct bio *bio;
-
-			__rq_for_each_bio(bio, rq)
-				zero_fill_bio(bio);
-			break;
-		}
-		cond_resched();
-	}
-
-	return 0;
-}
-
 static void loop_clear_limits(struct loop_device *lo, int mode)
 {
 	struct queue_limits lim = queue_limits_start_update(lo->lo_queue);
@@ -342,7 +276,7 @@ static void lo_complete_rq(struct request *rq)
 	struct loop_cmd *cmd = blk_mq_rq_to_pdu(rq);
 	blk_status_t ret = BLK_STS_OK;
 
-	if (!cmd->use_aio || cmd->ret < 0 || cmd->ret == blk_rq_bytes(rq) ||
+	if (cmd->ret < 0 || cmd->ret == blk_rq_bytes(rq) ||
 	    req_op(rq) != REQ_OP_READ) {
 		if (cmd->ret < 0)
 			ret = errno_to_blk_status(cmd->ret);
@@ -358,14 +292,13 @@ static void lo_complete_rq(struct request *rq)
 		cmd->ret = 0;
 		blk_mq_requeue_request(rq, true);
 	} else {
-		if (cmd->use_aio) {
-			struct bio *bio = rq->bio;
+		struct bio *bio = rq->bio;
 
-			while (bio) {
-				zero_fill_bio(bio);
-				bio = bio->bi_next;
-			}
+		while (bio) {
+			zero_fill_bio(bio);
+			bio = bio->bi_next;
 		}
+
 		ret = BLK_STS_IOERR;
 end_io:
 		blk_mq_end_request(rq, ret);
@@ -445,9 +378,14 @@ static int lo_rw_aio(struct loop_device *lo, struct loop_cmd *cmd,
 
 	cmd->iocb.ki_pos = pos;
 	cmd->iocb.ki_filp = file;
-	cmd->iocb.ki_complete = lo_rw_aio_complete;
-	cmd->iocb.ki_flags = IOCB_DIRECT;
-	cmd->iocb.ki_ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_NONE, 0);
+	cmd->iocb.ki_ioprio = req_get_ioprio(rq);
+	if (cmd->use_aio) {
+		cmd->iocb.ki_complete = lo_rw_aio_complete;
+		cmd->iocb.ki_flags = IOCB_DIRECT;
+	} else {
+		cmd->iocb.ki_complete = NULL;
+		cmd->iocb.ki_flags = 0;
+	}
 
 	if (rw == ITER_SOURCE)
 		ret = file->f_op->write_iter(&cmd->iocb, &iter);
@@ -458,7 +396,7 @@ static int lo_rw_aio(struct loop_device *lo, struct loop_cmd *cmd,
 
 	if (ret != -EIOCBQUEUED)
 		lo_rw_aio_complete(&cmd->iocb, ret);
-	return 0;
+	return -EIOCBQUEUED;
 }
 
 static int do_req_filebacked(struct loop_device *lo, struct request *rq)
@@ -466,15 +404,6 @@ static int do_req_filebacked(struct loop_device *lo, struct request *rq)
 	struct loop_cmd *cmd = blk_mq_rq_to_pdu(rq);
 	loff_t pos = ((loff_t) blk_rq_pos(rq) << 9) + lo->lo_offset;
 
-	/*
-	 * lo_write_simple and lo_read_simple should have been covered
-	 * by io submit style function like lo_rw_aio(), one blocker
-	 * is that lo_read_simple() need to call flush_dcache_page after
-	 * the page is written from kernel, and it isn't easy to handle
-	 * this in io submit style function which submits all segments
-	 * of the req at one time. And direct read IO doesn't need to
-	 * run flush_dcache_page().
-	 */
 	switch (req_op(rq)) {
 	case REQ_OP_FLUSH:
 		return lo_req_flush(lo, rq);
@@ -490,15 +419,9 @@ static int do_req_filebacked(struct loop_device *lo, struct request *rq)
 	case REQ_OP_DISCARD:
 		return lo_fallocate(lo, rq, pos, FALLOC_FL_PUNCH_HOLE);
 	case REQ_OP_WRITE:
-		if (cmd->use_aio)
-			return lo_rw_aio(lo, cmd, pos, ITER_SOURCE);
-		else
-			return lo_write_simple(lo, rq, pos);
+		return lo_rw_aio(lo, cmd, pos, ITER_SOURCE);
 	case REQ_OP_READ:
-		if (cmd->use_aio)
-			return lo_rw_aio(lo, cmd, pos, ITER_DEST);
-		else
-			return lo_read_simple(lo, rq, pos);
+		return lo_rw_aio(lo, cmd, pos, ITER_DEST);
 	default:
 		WARN_ON_ONCE(1);
 		return -EIO;
@@ -662,19 +585,20 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
 	 * dependency.
 	 */
 	fput(old_file);
+	dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 0);
 	if (partscan)
 		loop_reread_partitions(lo);
 
 	error = 0;
 done:
-	/* enable and uncork uevent now that we are done */
-	dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 0);
+	kobject_uevent(&disk_to_dev(lo->lo_disk)->kobj, KOBJ_CHANGE);
 	return error;
 
 out_err:
 	loop_global_unlock(lo, is_loop);
 out_putf:
 	fput(file);
+	dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 0);
 	goto done;
 }
 
@@ -1129,8 +1053,8 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode,
 	if (partscan)
 		clear_bit(GD_SUPPRESS_PART_SCAN, &lo->lo_disk->state);
 
-	/* enable and uncork uevent now that we are done */
 	dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 0);
+	kobject_uevent(&disk_to_dev(lo->lo_disk)->kobj, KOBJ_CHANGE);
 
 	loop_global_unlock(lo, is_loop);
 	if (partscan)
@@ -1921,7 +1845,6 @@ static void loop_handle_cmd(struct loop_cmd *cmd)
 	struct loop_device *lo = rq->q->queuedata;
 	int ret = 0;
 	struct mem_cgroup *old_memcg = NULL;
-	const bool use_aio = cmd->use_aio;
 
 	if (write && (lo->lo_flags & LO_FLAGS_READ_ONLY)) {
 		ret = -EIO;
@@ -1951,7 +1874,7 @@ static void loop_handle_cmd(struct loop_cmd *cmd)
 	}
  failed:
 	/* complete non-aio request */
-	if (!use_aio || ret) {
+	if (ret != -EIOCBQUEUED) {
 		if (ret == -EOPNOTSUPP)
 			cmd->ret = ret;
 		else
diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index 3bb9cee0a9b5..aa163ae9b2aa 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -2031,7 +2031,7 @@ static int null_add_dev(struct nullb_device *dev)
 	nullb->disk->minors = 1;
 	nullb->disk->fops = &null_ops;
 	nullb->disk->private_data = nullb;
-	strscpy_pad(nullb->disk->disk_name, nullb->disk_name, DISK_NAME_LEN);
+	strscpy(nullb->disk->disk_name, nullb->disk_name);
 
 	if (nullb->dev->zoned) {
 		rv = null_register_zoned_dev(nullb);
diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 2fd05c1bd30b..f9032076bc06 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -123,15 +123,6 @@ struct ublk_uring_cmd_pdu {
 #define UBLK_IO_FLAG_OWNED_BY_SRV 0x02
 
 /*
- * IO command is aborted, so this flag is set in case of
- * !UBLK_IO_FLAG_ACTIVE.
- *
- * After this flag is observed, any pending or new incoming request
- * associated with this io command will be failed immediately
- */
-#define UBLK_IO_FLAG_ABORTED 0x04
-
-/*
  * UBLK_IO_FLAG_NEED_GET_DATA is set because IO command requires
  * get data buffer address from ublksrv.
  *
@@ -199,8 +190,6 @@ struct ublk_device {
 	struct completion	completion;
 	unsigned int		nr_queues_ready;
 	unsigned int		nr_privileged_daemon;
-
-	struct work_struct	nosrv_work;
 };
 
 /* header of ublk_params */
@@ -209,18 +198,13 @@ struct ublk_params_header {
 	__u32	types;
 };
 
-static bool ublk_abort_requests(struct ublk_device *ub, struct ublk_queue *ubq);
-
+static void ublk_stop_dev_unlocked(struct ublk_device *ub);
+static void ublk_abort_queue(struct ublk_device *ub, struct ublk_queue *ubq);
 static inline struct request *__ublk_check_and_get_req(struct ublk_device *ub,
-		struct ublk_queue *ubq, int tag, size_t offset);
+		const struct ublk_queue *ubq, int tag, size_t offset);
 static inline unsigned int ublk_req_build_flags(struct request *req);
 static inline struct ublksrv_io_desc *ublk_get_iod(struct ublk_queue *ubq,
 						   int tag);
-static inline bool ublk_dev_is_user_copy(const struct ublk_device *ub)
-{
-	return ub->dev_info.flags & (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY);
-}
-
 static inline bool ublk_dev_is_zoned(const struct ublk_device *ub)
 {
 	return ub->dev_info.flags & UBLK_F_ZONED;
@@ -620,14 +604,19 @@ static void ublk_apply_params(struct ublk_device *ub)
 		ublk_dev_param_zoned_apply(ub);
 }
 
+static inline bool ublk_support_zero_copy(const struct ublk_queue *ubq)
+{
+	return ubq->flags & UBLK_F_SUPPORT_ZERO_COPY;
+}
+
 static inline bool ublk_support_user_copy(const struct ublk_queue *ubq)
 {
-	return ubq->flags & (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY);
+	return ubq->flags & UBLK_F_USER_COPY;
 }
 
 static inline bool ublk_need_map_io(const struct ublk_queue *ubq)
 {
-	return !ublk_support_user_copy(ubq);
+	return !ublk_support_user_copy(ubq) && !ublk_support_zero_copy(ubq);
 }
 
 static inline bool ublk_need_req_ref(const struct ublk_queue *ubq)
@@ -635,8 +624,11 @@ static inline bool ublk_need_req_ref(const struct ublk_queue *ubq)
 	/*
 	 * read()/write() is involved in user copy, so request reference
 	 * has to be grabbed
+	 *
+	 * for zero copy, request buffer need to be registered to io_uring
+	 * buffer table, so reference is needed
 	 */
-	return ublk_support_user_copy(ubq);
+	return ublk_support_user_copy(ubq) || ublk_support_zero_copy(ubq);
 }
 
 static inline void ublk_init_req_ref(const struct ublk_queue *ubq,
@@ -1074,7 +1066,7 @@ static inline struct ublk_uring_cmd_pdu *ublk_get_uring_cmd_pdu(
 
 static inline bool ubq_daemon_is_dying(struct ublk_queue *ubq)
 {
-	return ubq->ubq_daemon->flags & PF_EXITING;
+	return !ubq->ubq_daemon || ubq->ubq_daemon->flags & PF_EXITING;
 }
 
 /* todo: handle partial completion */
@@ -1085,12 +1077,6 @@ static inline void __ublk_complete_rq(struct request *req)
 	unsigned int unmapped_bytes;
 	blk_status_t res = BLK_STS_OK;
 
-	/* called from ublk_abort_queue() code path */
-	if (io->flags & UBLK_IO_FLAG_ABORTED) {
-		res = BLK_STS_IOERR;
-		goto exit;
-	}
-
 	/* failed read IO if nothing is read */
 	if (!io->res && req_op(req) == REQ_OP_READ)
 		io->res = -EIO;
@@ -1140,25 +1126,6 @@ static void ublk_complete_rq(struct kref *ref)
 	__ublk_complete_rq(req);
 }
 
-/*
- * Since ublk_rq_task_work_cb always fails requests immediately during
- * exiting, __ublk_fail_req() is only called from abort context during
- * exiting. So lock is unnecessary.
- *
- * Also aborting may not be started yet, keep in mind that one failed
- * request may be issued by block layer again.
- */
-static void __ublk_fail_req(struct ublk_queue *ubq, struct ublk_io *io,
-		struct request *req)
-{
-	WARN_ON_ONCE(io->flags & UBLK_IO_FLAG_ACTIVE);
-
-	if (ublk_nosrv_should_reissue_outstanding(ubq->dev))
-		blk_mq_requeue_request(req, false);
-	else
-		ublk_put_req_ref(ubq, req);
-}
-
 static void ubq_complete_io_cmd(struct ublk_io *io, int res,
 				unsigned issue_flags)
 {
@@ -1314,8 +1281,6 @@ static void ublk_queue_cmd_list(struct ublk_queue *ubq, struct rq_list *l)
 static enum blk_eh_timer_return ublk_timeout(struct request *rq)
 {
 	struct ublk_queue *ubq = rq->mq_hctx->driver_data;
-	unsigned int nr_inflight = 0;
-	int i;
 
 	if (ubq->flags & UBLK_F_UNPRIVILEGED_DEV) {
 		if (!ubq->timeout) {
@@ -1326,30 +1291,11 @@ static enum blk_eh_timer_return ublk_timeout(struct request *rq)
 		return BLK_EH_DONE;
 	}
 
-	if (!ubq_daemon_is_dying(ubq))
-		return BLK_EH_RESET_TIMER;
-
-	for (i = 0; i < ubq->q_depth; i++) {
-		struct ublk_io *io = &ubq->ios[i];
-
-		if (!(io->flags & UBLK_IO_FLAG_ACTIVE))
-			nr_inflight++;
-	}
-
-	/* cancelable uring_cmd can't help us if all commands are in-flight */
-	if (nr_inflight == ubq->q_depth) {
-		struct ublk_device *ub = ubq->dev;
-
-		if (ublk_abort_requests(ub, ubq)) {
-			schedule_work(&ub->nosrv_work);
-		}
-		return BLK_EH_DONE;
-	}
-
 	return BLK_EH_RESET_TIMER;
 }
 
-static blk_status_t ublk_prep_req(struct ublk_queue *ubq, struct request *rq)
+static blk_status_t ublk_prep_req(struct ublk_queue *ubq, struct request *rq,
+				  bool check_cancel)
 {
 	blk_status_t res;
 
@@ -1368,7 +1314,7 @@ static blk_status_t ublk_prep_req(struct ublk_queue *ubq, struct request *rq)
 	if (ublk_nosrv_should_queue_io(ubq) && unlikely(ubq->force_abort))
 		return BLK_STS_IOERR;
 
-	if (unlikely(ubq->canceling))
+	if (check_cancel && unlikely(ubq->canceling))
 		return BLK_STS_IOERR;
 
 	/* fill iod to slot in io cmd buffer */
@@ -1387,7 +1333,7 @@ static blk_status_t ublk_queue_rq(struct blk_mq_hw_ctx *hctx,
 	struct request *rq = bd->rq;
 	blk_status_t res;
 
-	res = ublk_prep_req(ubq, rq);
+	res = ublk_prep_req(ubq, rq, false);
 	if (res != BLK_STS_OK)
 		return res;
 
@@ -1419,7 +1365,7 @@ static void ublk_queue_rqs(struct rq_list *rqlist)
 			ublk_queue_cmd_list(ubq, &submit_list);
 		ubq = this_q;
 
-		if (ublk_prep_req(ubq, req) == BLK_STS_OK)
+		if (ublk_prep_req(ubq, req, true) == BLK_STS_OK)
 			rq_list_add_tail(&submit_list, req);
 		else
 			rq_list_add_tail(&requeue_list, req);
@@ -1447,6 +1393,37 @@ static const struct blk_mq_ops ublk_mq_ops = {
 	.timeout	= ublk_timeout,
 };
 
+static void ublk_queue_reinit(struct ublk_device *ub, struct ublk_queue *ubq)
+{
+	int i;
+
+	/* All old ioucmds have to be completed */
+	ubq->nr_io_ready = 0;
+
+	/*
+	 * old daemon is PF_EXITING, put it now
+	 *
+	 * It could be NULL in case of closing one quisced device.
+	 */
+	if (ubq->ubq_daemon)
+		put_task_struct(ubq->ubq_daemon);
+	/* We have to reset it to NULL, otherwise ub won't accept new FETCH_REQ */
+	ubq->ubq_daemon = NULL;
+	ubq->timeout = false;
+
+	for (i = 0; i < ubq->q_depth; i++) {
+		struct ublk_io *io = &ubq->ios[i];
+
+		/*
+		 * UBLK_IO_FLAG_CANCELED is kept for avoiding to touch
+		 * io->cmd
+		 */
+		io->flags &= UBLK_IO_FLAG_CANCELED;
+		io->cmd = NULL;
+		io->addr = 0;
+	}
+}
+
 static int ublk_ch_open(struct inode *inode, struct file *filp)
 {
 	struct ublk_device *ub = container_of(inode->i_cdev,
@@ -1458,10 +1435,119 @@ static int ublk_ch_open(struct inode *inode, struct file *filp)
 	return 0;
 }
 
+static void ublk_reset_ch_dev(struct ublk_device *ub)
+{
+	int i;
+
+	for (i = 0; i < ub->dev_info.nr_hw_queues; i++)
+		ublk_queue_reinit(ub, ublk_get_queue(ub, i));
+
+	/* set to NULL, otherwise new ubq_daemon cannot mmap the io_cmd_buf */
+	ub->mm = NULL;
+	ub->nr_queues_ready = 0;
+	ub->nr_privileged_daemon = 0;
+}
+
+static struct gendisk *ublk_get_disk(struct ublk_device *ub)
+{
+	struct gendisk *disk;
+
+	spin_lock(&ub->lock);
+	disk = ub->ub_disk;
+	if (disk)
+		get_device(disk_to_dev(disk));
+	spin_unlock(&ub->lock);
+
+	return disk;
+}
+
+static void ublk_put_disk(struct gendisk *disk)
+{
+	if (disk)
+		put_device(disk_to_dev(disk));
+}
+
 static int ublk_ch_release(struct inode *inode, struct file *filp)
 {
 	struct ublk_device *ub = filp->private_data;
+	struct gendisk *disk;
+	int i;
+
+	/*
+	 * disk isn't attached yet, either device isn't live, or it has
+	 * been removed already, so we needn't to do anything
+	 */
+	disk = ublk_get_disk(ub);
+	if (!disk)
+		goto out;
+
+	/*
+	 * All uring_cmd are done now, so abort any request outstanding to
+	 * the ublk server
+	 *
+	 * This can be done in lockless way because ublk server has been
+	 * gone
+	 *
+	 * More importantly, we have to provide forward progress guarantee
+	 * without holding ub->mutex, otherwise control task grabbing
+	 * ub->mutex triggers deadlock
+	 *
+	 * All requests may be inflight, so ->canceling may not be set, set
+	 * it now.
+	 */
+	for (i = 0; i < ub->dev_info.nr_hw_queues; i++) {
+		struct ublk_queue *ubq = ublk_get_queue(ub, i);
+
+		ubq->canceling = true;
+		ublk_abort_queue(ub, ubq);
+	}
+	blk_mq_kick_requeue_list(disk->queue);
+
+	/*
+	 * All infligh requests have been completed or requeued and any new
+	 * request will be failed or requeued via `->canceling` now, so it is
+	 * fine to grab ub->mutex now.
+	 */
+	mutex_lock(&ub->mutex);
+
+	/* double check after grabbing lock */
+	if (!ub->ub_disk)
+		goto unlock;
+
+	/*
+	 * Transition the device to the nosrv state. What exactly this
+	 * means depends on the recovery flags
+	 */
+	blk_mq_quiesce_queue(disk->queue);
+	if (ublk_nosrv_should_stop_dev(ub)) {
+		/*
+		 * Allow any pending/future I/O to pass through quickly
+		 * with an error. This is needed because del_gendisk
+		 * waits for all pending I/O to complete
+		 */
+		for (i = 0; i < ub->dev_info.nr_hw_queues; i++)
+			ublk_get_queue(ub, i)->force_abort = true;
+		blk_mq_unquiesce_queue(disk->queue);
 
+		ublk_stop_dev_unlocked(ub);
+	} else {
+		if (ublk_nosrv_dev_should_queue_io(ub)) {
+			/* ->canceling is set and all requests are aborted */
+			ub->dev_info.state = UBLK_S_DEV_QUIESCED;
+		} else {
+			ub->dev_info.state = UBLK_S_DEV_FAIL_IO;
+			for (i = 0; i < ub->dev_info.nr_hw_queues; i++)
+				ublk_get_queue(ub, i)->fail_io = true;
+		}
+		blk_mq_unquiesce_queue(disk->queue);
+	}
+unlock:
+	mutex_unlock(&ub->mutex);
+	ublk_put_disk(disk);
+
+	/* all uring_cmd has been done now, reset device & ubq */
+	ublk_reset_ch_dev(ub);
+out:
 	clear_bit(UB_STATE_OPEN, &ub->state);
 	return 0;
 }
@@ -1528,10 +1614,26 @@ static void ublk_commit_completion(struct ublk_device *ub,
 		ublk_put_req_ref(ubq, req);
 }
 
+static void __ublk_fail_req(struct ublk_queue *ubq, struct ublk_io *io,
+		struct request *req)
+{
+	WARN_ON_ONCE(io->flags & UBLK_IO_FLAG_ACTIVE);
+
+	if (ublk_nosrv_should_reissue_outstanding(ubq->dev))
+		blk_mq_requeue_request(req, false);
+	else {
+		io->res = -EIO;
+		__ublk_complete_rq(req);
+	}
+}
+
 /*
- * Called from ubq_daemon context via cancel fn, meantime quiesce ublk
- * blk-mq queue, so we are called exclusively with blk-mq and ubq_daemon
- * context, so everything is serialized.
+ * Called from ublk char device release handler, when any uring_cmd is
+ * done, meantime request queue is "quiesced" since all inflight requests
+ * can't be completed because ublk server is dead.
+ *
+ * So no one can hold our request IO reference any more, simply ignore the
+ * reference, and complete the request immediately
  */
 static void ublk_abort_queue(struct ublk_device *ub, struct ublk_queue *ubq)
 {
@@ -1548,46 +1650,29 @@ static void ublk_abort_queue(struct ublk_device *ub, struct ublk_queue *ubq)
 			 * will do it
 			 */
 			rq = blk_mq_tag_to_rq(ub->tag_set.tags[ubq->q_id], i);
-			if (rq && blk_mq_request_started(rq)) {
-				io->flags |= UBLK_IO_FLAG_ABORTED;
+			if (rq && blk_mq_request_started(rq))
 				__ublk_fail_req(ubq, io, rq);
-			}
 		}
 	}
 }
 
 /* Must be called when queue is frozen */
-static bool ublk_mark_queue_canceling(struct ublk_queue *ubq)
+static void ublk_mark_queue_canceling(struct ublk_queue *ubq)
 {
-	bool canceled;
-
 	spin_lock(&ubq->cancel_lock);
-	canceled = ubq->canceling;
-	if (!canceled)
+	if (!ubq->canceling)
 		ubq->canceling = true;
 	spin_unlock(&ubq->cancel_lock);
-
-	return canceled;
 }
 
-static bool ublk_abort_requests(struct ublk_device *ub, struct ublk_queue *ubq)
+static void ublk_start_cancel(struct ublk_queue *ubq)
 {
-	bool was_canceled = ubq->canceling;
-	struct gendisk *disk;
-
-	if (was_canceled)
-		return false;
-
-	spin_lock(&ub->lock);
-	disk = ub->ub_disk;
-	if (disk)
-		get_device(disk_to_dev(disk));
-	spin_unlock(&ub->lock);
+	struct ublk_device *ub = ubq->dev;
+	struct gendisk *disk = ublk_get_disk(ub);
 
 	/* Our disk has been dead */
 	if (!disk)
-		return false;
-
+		return;
 	/*
 	 * Now we are serialized with ublk_queue_rq()
 	 *
@@ -1596,25 +1681,36 @@ static bool ublk_abort_requests(struct ublk_device *ub, struct ublk_queue *ubq)
 	 * touch completed uring_cmd
 	 */
 	blk_mq_quiesce_queue(disk->queue);
-	was_canceled = ublk_mark_queue_canceling(ubq);
-	if (!was_canceled) {
-		/* abort queue is for making forward progress */
-		ublk_abort_queue(ub, ubq);
-	}
+	ublk_mark_queue_canceling(ubq);
 	blk_mq_unquiesce_queue(disk->queue);
-	put_device(disk_to_dev(disk));
-
-	return !was_canceled;
+	ublk_put_disk(disk);
 }
 
-static void ublk_cancel_cmd(struct ublk_queue *ubq, struct ublk_io *io,
+static void ublk_cancel_cmd(struct ublk_queue *ubq, unsigned tag,
 		unsigned int issue_flags)
 {
+	struct ublk_io *io = &ubq->ios[tag];
+	struct ublk_device *ub = ubq->dev;
+	struct request *req;
 	bool done;
 
 	if (!(io->flags & UBLK_IO_FLAG_ACTIVE))
 		return;
 
+	/*
+	 * Don't try to cancel this command if the request is started for
+	 * avoiding race between io_uring_cmd_done() and
+	 * io_uring_cmd_complete_in_task().
+	 *
+	 * Either the started request will be aborted via __ublk_abort_rq(),
+	 * then this uring_cmd is canceled next time, or it will be done in
+	 * task work function ublk_dispatch_req() because io_uring guarantees
+	 * that ublk_dispatch_req() is always called
+	 */
+	req = blk_mq_tag_to_rq(ub->tag_set.tags[ubq->q_id], tag);
+	if (req && blk_mq_request_started(req))
+		return;
+
 	spin_lock(&ubq->cancel_lock);
 	done = !!(io->flags & UBLK_IO_FLAG_CANCELED);
 	if (!done)
@@ -1628,6 +1724,17 @@ static void ublk_cancel_cmd(struct ublk_queue *ubq, struct ublk_io *io,
 /*
  * The ublk char device won't be closed when calling cancel fn, so both
  * ublk device and queue are guaranteed to be live
+ *
+ * Two-stage cancel:
+ *
+ * - make every active uring_cmd done in ->cancel_fn()
+ *
+ * - aborting inflight ublk IO requests in ublk char device release handler,
+ *   which depends on 1st stage because device can only be closed iff all
+ *   uring_cmd are done
+ *
+ * Do _not_ try to acquire ub->mutex before all inflight requests are
+ * aborted, otherwise deadlock may be caused.
  */
 static void ublk_uring_cmd_cancel_fn(struct io_uring_cmd *cmd,
 		unsigned int issue_flags)
@@ -1635,9 +1742,6 @@ static void ublk_uring_cmd_cancel_fn(struct io_uring_cmd *cmd,
 	struct ublk_uring_cmd_pdu *pdu = ublk_get_uring_cmd_pdu(cmd);
 	struct ublk_queue *ubq = pdu->ubq;
 	struct task_struct *task;
-	struct ublk_device *ub;
-	bool need_schedule;
-	struct ublk_io *io;
 
 	if (WARN_ON_ONCE(!ubq))
 		return;
@@ -1649,16 +1753,11 @@ static void ublk_uring_cmd_cancel_fn(struct io_uring_cmd *cmd,
 	if (WARN_ON_ONCE(task && task != ubq->ubq_daemon))
 		return;
 
-	ub = ubq->dev;
-	need_schedule = ublk_abort_requests(ub, ubq);
+	if (!ubq->canceling)
+		ublk_start_cancel(ubq);
 
-	io = &ubq->ios[pdu->tag];
-	WARN_ON_ONCE(io->cmd != cmd);
-	ublk_cancel_cmd(ubq, io, issue_flags);
-
-	if (need_schedule) {
-		schedule_work(&ub->nosrv_work);
-	}
+	WARN_ON_ONCE(ubq->ios[pdu->tag].cmd != cmd);
+	ublk_cancel_cmd(ubq, pdu->tag, issue_flags);
 }
 
 static inline bool ublk_queue_ready(struct ublk_queue *ubq)
@@ -1671,7 +1770,7 @@ static void ublk_cancel_queue(struct ublk_queue *ubq)
 	int i;
 
 	for (i = 0; i < ubq->q_depth; i++)
-		ublk_cancel_cmd(ubq, &ubq->ios[i], IO_URING_F_UNLOCKED);
+		ublk_cancel_cmd(ubq, i, IO_URING_F_UNLOCKED);
 }
 
 /* Cancel all pending commands, must be called after del_gendisk() returns */
@@ -1709,33 +1808,20 @@ static void ublk_wait_tagset_rqs_idle(struct ublk_device *ub)
 	}
 }
 
-static void __ublk_quiesce_dev(struct ublk_device *ub)
-{
-	pr_devel("%s: quiesce ub: dev_id %d state %s\n",
-			__func__, ub->dev_info.dev_id,
-			ub->dev_info.state == UBLK_S_DEV_LIVE ?
-			"LIVE" : "QUIESCED");
-	blk_mq_quiesce_queue(ub->ub_disk->queue);
-	ublk_wait_tagset_rqs_idle(ub);
-	ub->dev_info.state = UBLK_S_DEV_QUIESCED;
-}
-
-static void ublk_unquiesce_dev(struct ublk_device *ub)
+static void ublk_force_abort_dev(struct ublk_device *ub)
 {
 	int i;
 
-	pr_devel("%s: unquiesce ub: dev_id %d state %s\n",
+	pr_devel("%s: force abort ub: dev_id %d state %s\n",
 			__func__, ub->dev_info.dev_id,
 			ub->dev_info.state == UBLK_S_DEV_LIVE ?
 			"LIVE" : "QUIESCED");
-	/* quiesce_work has run. We let requeued rqs be aborted
-	 * before running fallback_wq. "force_abort" must be seen
-	 * after request queue is unqiuesced. Then del_gendisk()
-	 * can move on.
-	 */
+	blk_mq_quiesce_queue(ub->ub_disk->queue);
+	if (ub->dev_info.state == UBLK_S_DEV_LIVE)
+		ublk_wait_tagset_rqs_idle(ub);
+
 	for (i = 0; i < ub->dev_info.nr_hw_queues; i++)
 		ublk_get_queue(ub, i)->force_abort = true;
-
 	blk_mq_unquiesce_queue(ub->ub_disk->queue);
 	/* We may have requeued some rqs in ublk_quiesce_queue() */
 	blk_mq_kick_requeue_list(ub->ub_disk->queue);
@@ -1756,61 +1842,51 @@ static struct gendisk *ublk_detach_disk(struct ublk_device *ub)
 	return disk;
 }
 
-static void ublk_stop_dev(struct ublk_device *ub)
+static void ublk_stop_dev_unlocked(struct ublk_device *ub)
+	__must_hold(&ub->mutex)
 {
 	struct gendisk *disk;
 
-	mutex_lock(&ub->mutex);
 	if (ub->dev_info.state == UBLK_S_DEV_DEAD)
-		goto unlock;
-	if (ublk_nosrv_dev_should_queue_io(ub)) {
-		if (ub->dev_info.state == UBLK_S_DEV_LIVE)
-			__ublk_quiesce_dev(ub);
-		ublk_unquiesce_dev(ub);
-	}
+		return;
+
+	if (ublk_nosrv_dev_should_queue_io(ub))
+		ublk_force_abort_dev(ub);
 	del_gendisk(ub->ub_disk);
 	disk = ublk_detach_disk(ub);
 	put_disk(disk);
- unlock:
+}
+
+static void ublk_stop_dev(struct ublk_device *ub)
+{
+	mutex_lock(&ub->mutex);
+	ublk_stop_dev_unlocked(ub);
 	mutex_unlock(&ub->mutex);
 	ublk_cancel_dev(ub);
 }
 
-static void ublk_nosrv_work(struct work_struct *work)
+/* reset ublk io_uring queue & io flags */
+static void ublk_reset_io_flags(struct ublk_device *ub)
 {
-	struct ublk_device *ub =
-		container_of(work, struct ublk_device, nosrv_work);
-	int i;
-
-	if (ublk_nosrv_should_stop_dev(ub)) {
-		ublk_stop_dev(ub);
-		return;
-	}
+	int i, j;
 
-	mutex_lock(&ub->mutex);
-	if (ub->dev_info.state != UBLK_S_DEV_LIVE)
-		goto unlock;
+	for (i = 0; i < ub->dev_info.nr_hw_queues; i++) {
+		struct ublk_queue *ubq = ublk_get_queue(ub, i);
 
-	if (ublk_nosrv_dev_should_queue_io(ub)) {
-		__ublk_quiesce_dev(ub);
-	} else {
-		blk_mq_quiesce_queue(ub->ub_disk->queue);
-		ub->dev_info.state = UBLK_S_DEV_FAIL_IO;
-		for (i = 0; i < ub->dev_info.nr_hw_queues; i++) {
-			ublk_get_queue(ub, i)->fail_io = true;
-		}
-		blk_mq_unquiesce_queue(ub->ub_disk->queue);
+		/* UBLK_IO_FLAG_CANCELED can be cleared now */
+		spin_lock(&ubq->cancel_lock);
+		for (j = 0; j < ubq->q_depth; j++)
+			ubq->ios[j].flags &= ~UBLK_IO_FLAG_CANCELED;
+		spin_unlock(&ubq->cancel_lock);
+		ubq->canceling = false;
+		ubq->fail_io = false;
 	}
-
- unlock:
-	mutex_unlock(&ub->mutex);
-	ublk_cancel_dev(ub);
 }
 
 /* device can only be started after all IOs are ready */
 static void ublk_mark_io_ready(struct ublk_device *ub, struct ublk_queue *ubq)
+	__must_hold(&ub->mutex)
 {
-	mutex_lock(&ub->mutex);
 	ubq->nr_io_ready++;
 	if (ublk_queue_ready(ubq)) {
 		ubq->ubq_daemon = current;
@@ -1820,18 +1896,12 @@ static void ublk_mark_io_ready(struct ublk_device *ub, struct ublk_queue *ubq)
 		if (capable(CAP_SYS_ADMIN))
 			ub->nr_privileged_daemon++;
 	}
-	if (ub->nr_queues_ready == ub->dev_info.nr_hw_queues)
-		complete_all(&ub->completion);
-	mutex_unlock(&ub->mutex);
-}
-
-static void ublk_handle_need_get_data(struct ublk_device *ub, int q_id,
-		int tag)
-{
-	struct ublk_queue *ubq = ublk_get_queue(ub, q_id);
-	struct request *req = blk_mq_tag_to_rq(ub->tag_set.tags[q_id], tag);
 
-	ublk_queue_cmd(ubq, req);
+	if (ub->nr_queues_ready == ub->dev_info.nr_hw_queues) {
+		/* now we are ready for handling ublk io request */
+		ublk_reset_io_flags(ub);
+		complete_all(&ub->completion);
+	}
 }
 
 static inline int ublk_check_cmd_op(u32 cmd_op)
@@ -1879,13 +1949,20 @@ static void ublk_io_release(void *priv)
 }
 
 static int ublk_register_io_buf(struct io_uring_cmd *cmd,
-				struct ublk_queue *ubq, unsigned int tag,
+				const struct ublk_queue *ubq, unsigned int tag,
 				unsigned int index, unsigned int issue_flags)
 {
 	struct ublk_device *ub = cmd->file->private_data;
+	const struct ublk_io *io = &ubq->ios[tag];
 	struct request *req;
 	int ret;
 
+	if (!ublk_support_zero_copy(ubq))
+		return -EINVAL;
+
+	if (!(io->flags & UBLK_IO_FLAG_OWNED_BY_SRV))
+		return -EINVAL;
+
 	req = __ublk_check_and_get_req(ub, ubq, tag, 0);
 	if (!req)
 		return -EINVAL;
@@ -1901,11 +1978,66 @@ static int ublk_register_io_buf(struct io_uring_cmd *cmd,
 }
 
 static int ublk_unregister_io_buf(struct io_uring_cmd *cmd,
+				  const struct ublk_queue *ubq, unsigned int tag,
 				  unsigned int index, unsigned int issue_flags)
 {
+	const struct ublk_io *io = &ubq->ios[tag];
+
+	if (!ublk_support_zero_copy(ubq))
+		return -EINVAL;
+
+	if (!(io->flags & UBLK_IO_FLAG_OWNED_BY_SRV))
+		return -EINVAL;
+
 	return io_buffer_unregister_bvec(cmd, index, issue_flags);
 }
 
+static int ublk_fetch(struct io_uring_cmd *cmd, struct ublk_queue *ubq,
+		      struct ublk_io *io, __u64 buf_addr)
+{
+	struct ublk_device *ub = ubq->dev;
+	int ret = 0;
+
+	/*
+	 * When handling FETCH command for setting up ublk uring queue,
+	 * ub->mutex is the innermost lock, and we won't block for handling
+	 * FETCH, so it is fine even for IO_URING_F_NONBLOCK.
+	 */
+	mutex_lock(&ub->mutex);
+	/* UBLK_IO_FETCH_REQ is only allowed before queue is setup */
+	if (ublk_queue_ready(ubq)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	/* allow each command to be FETCHed at most once */
+	if (io->flags & UBLK_IO_FLAG_ACTIVE) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	WARN_ON_ONCE(io->flags & UBLK_IO_FLAG_OWNED_BY_SRV);
+
+	if (ublk_need_map_io(ubq)) {
+		/*
+		 * FETCH_RQ has to provide IO buffer if NEED GET
+		 * DATA is not enabled
+		 */
+		if (!buf_addr && !ublk_need_get_data(ubq))
+			goto out;
+	} else if (buf_addr) {
+		/* User copy requires addr to be unset */
+		ret = -EINVAL;
+		goto out;
+	}
+
+	ublk_fill_io_cmd(io, cmd, buf_addr);
+	ublk_mark_io_ready(ub, ubq);
+out:
+	mutex_unlock(&ub->mutex);
+	return ret;
+}
+
 static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 			       unsigned int issue_flags,
 			       const struct ublksrv_io_cmd *ub_cmd)
@@ -1960,35 +2092,11 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 	case UBLK_IO_REGISTER_IO_BUF:
 		return ublk_register_io_buf(cmd, ubq, tag, ub_cmd->addr, issue_flags);
 	case UBLK_IO_UNREGISTER_IO_BUF:
-		return ublk_unregister_io_buf(cmd, ub_cmd->addr, issue_flags);
+		return ublk_unregister_io_buf(cmd, ubq, tag, ub_cmd->addr, issue_flags);
 	case UBLK_IO_FETCH_REQ:
-		/* UBLK_IO_FETCH_REQ is only allowed before queue is setup */
-		if (ublk_queue_ready(ubq)) {
-			ret = -EBUSY;
-			goto out;
-		}
-		/*
-		 * The io is being handled by server, so COMMIT_RQ is expected
-		 * instead of FETCH_REQ
-		 */
-		if (io->flags & UBLK_IO_FLAG_OWNED_BY_SRV)
-			goto out;
-
-		if (ublk_need_map_io(ubq)) {
-			/*
-			 * FETCH_RQ has to provide IO buffer if NEED GET
-			 * DATA is not enabled
-			 */
-			if (!ub_cmd->addr && !ublk_need_get_data(ubq))
-				goto out;
-		} else if (ub_cmd->addr) {
-			/* User copy requires addr to be unset */
-			ret = -EINVAL;
+		ret = ublk_fetch(cmd, ubq, io, ub_cmd->addr);
+		if (ret)
 			goto out;
-		}
-
-		ublk_fill_io_cmd(io, cmd, ub_cmd->addr);
-		ublk_mark_io_ready(ub, ubq);
 		break;
 	case UBLK_IO_COMMIT_AND_FETCH_REQ:
 		req = blk_mq_tag_to_rq(ub->tag_set.tags[ub_cmd->q_id], tag);
@@ -2020,8 +2128,9 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 		if (!(io->flags & UBLK_IO_FLAG_OWNED_BY_SRV))
 			goto out;
 		ublk_fill_io_cmd(io, cmd, ub_cmd->addr);
-		ublk_handle_need_get_data(ub, ub_cmd->q_id, ub_cmd->tag);
-		break;
+		req = blk_mq_tag_to_rq(ub->tag_set.tags[ub_cmd->q_id], tag);
+		ublk_dispatch_req(ubq, req, issue_flags);
+		return -EIOCBQUEUED;
 	default:
 		goto out;
 	}
@@ -2035,13 +2144,10 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 }
 
 static inline struct request *__ublk_check_and_get_req(struct ublk_device *ub,
-		struct ublk_queue *ubq, int tag, size_t offset)
+		const struct ublk_queue *ubq, int tag, size_t offset)
 {
 	struct request *req;
 
-	if (!ublk_need_req_ref(ubq))
-		return NULL;
-
 	req = blk_mq_tag_to_rq(ub->tag_set.tags[ubq->q_id], tag);
 	if (!req)
 		return NULL;
@@ -2155,6 +2261,9 @@ static struct request *ublk_check_and_get_req(struct kiocb *iocb,
 	if (!ubq)
 		return ERR_PTR(-EINVAL);
 
+	if (!ublk_support_user_copy(ubq))
+		return ERR_PTR(-EACCES);
+
 	if (tag >= ubq->q_depth)
 		return ERR_PTR(-EINVAL);
 
@@ -2388,7 +2497,6 @@ static void ublk_remove(struct ublk_device *ub)
 	bool unprivileged;
 
 	ublk_stop_dev(ub);
-	cancel_work_sync(&ub->nosrv_work);
 	cdev_device_del(&ub->cdev, &ub->cdev_dev);
 	unprivileged = ub->dev_info.flags & UBLK_F_UNPRIVILEGED_DEV;
 	ublk_put_device(ub);
@@ -2413,9 +2521,9 @@ static struct ublk_device *ublk_get_device_from_id(int idx)
 	return ub;
 }
 
-static int ublk_ctrl_start_dev(struct ublk_device *ub, struct io_uring_cmd *cmd)
+static int ublk_ctrl_start_dev(struct ublk_device *ub,
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	const struct ublk_param_basic *p = &ub->params.basic;
 	int ublksrv_pid = (int)header->data[0];
 	struct queue_limits lim = {
@@ -2534,9 +2642,8 @@ out_unlock:
 }
 
 static int ublk_ctrl_get_queue_affinity(struct ublk_device *ub,
-		struct io_uring_cmd *cmd)
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
 	cpumask_var_t cpumask;
 	unsigned long queue;
@@ -2585,9 +2692,8 @@ static inline void ublk_dump_dev_info(struct ublksrv_ctrl_dev_info *info)
 			info->nr_hw_queues, info->queue_depth);
 }
 
-static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd)
+static int ublk_ctrl_add_dev(const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
 	struct ublksrv_ctrl_dev_info info;
 	struct ublk_device *ub;
@@ -2675,7 +2781,6 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd)
 		goto out_unlock;
 	mutex_init(&ub->mutex);
 	spin_lock_init(&ub->lock);
-	INIT_WORK(&ub->nosrv_work, ublk_nosrv_work);
 
 	ret = ublk_alloc_dev_number(ub, header->dev_id);
 	if (ret < 0)
@@ -2697,13 +2802,18 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd)
 	ub->dev_info.flags |= UBLK_F_CMD_IOCTL_ENCODE |
 		UBLK_F_URING_CMD_COMP_IN_TASK;
 
-	/* GET_DATA isn't needed any more with USER_COPY */
-	if (ublk_dev_is_user_copy(ub))
+	/* GET_DATA isn't needed any more with USER_COPY or ZERO COPY */
+	if (ub->dev_info.flags & (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY))
 		ub->dev_info.flags &= ~UBLK_F_NEED_GET_DATA;
 
-	/* Zoned storage support requires user copy feature */
+	/*
+	 * Zoned storage support requires reuse `ublksrv_io_cmd->addr` for
+	 * returning write_append_lba, which is only allowed in case of
+	 * user copy or zero copy
+	 */
 	if (ublk_dev_is_zoned(ub) &&
-	    (!IS_ENABLED(CONFIG_BLK_DEV_ZONED) || !ublk_dev_is_user_copy(ub))) {
+	    (!IS_ENABLED(CONFIG_BLK_DEV_ZONED) || !(ub->dev_info.flags &
+	     (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY)))) {
 		ret = -EINVAL;
 		goto out_free_dev_number;
 	}
@@ -2807,14 +2917,12 @@ static inline void ublk_ctrl_cmd_dump(struct io_uring_cmd *cmd)
 static int ublk_ctrl_stop_dev(struct ublk_device *ub)
 {
 	ublk_stop_dev(ub);
-	cancel_work_sync(&ub->nosrv_work);
 	return 0;
 }
 
 static int ublk_ctrl_get_dev_info(struct ublk_device *ub,
-		struct io_uring_cmd *cmd)
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
 
 	if (header->len < sizeof(struct ublksrv_ctrl_dev_info) || !header->addr)
@@ -2843,9 +2951,8 @@ static void ublk_ctrl_fill_params_devt(struct ublk_device *ub)
 }
 
 static int ublk_ctrl_get_params(struct ublk_device *ub,
-		struct io_uring_cmd *cmd)
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
 	struct ublk_params_header ph;
 	int ret;
@@ -2874,9 +2981,8 @@ static int ublk_ctrl_get_params(struct ublk_device *ub,
 }
 
 static int ublk_ctrl_set_params(struct ublk_device *ub,
-		struct io_uring_cmd *cmd)
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
 	struct ublk_params_header ph;
 	int ret = -EFAULT;
@@ -2914,43 +3020,14 @@ static int ublk_ctrl_set_params(struct ublk_device *ub,
 	return ret;
 }
 
-static void ublk_queue_reinit(struct ublk_device *ub, struct ublk_queue *ubq)
-{
-	int i;
-
-	WARN_ON_ONCE(!(ubq->ubq_daemon && ubq_daemon_is_dying(ubq)));
-
-	/* All old ioucmds have to be completed */
-	ubq->nr_io_ready = 0;
-	/* old daemon is PF_EXITING, put it now */
-	put_task_struct(ubq->ubq_daemon);
-	/* We have to reset it to NULL, otherwise ub won't accept new FETCH_REQ */
-	ubq->ubq_daemon = NULL;
-	ubq->timeout = false;
-	ubq->canceling = false;
-
-	for (i = 0; i < ubq->q_depth; i++) {
-		struct ublk_io *io = &ubq->ios[i];
-
-		/* forget everything now and be ready for new FETCH_REQ */
-		io->flags = 0;
-		io->cmd = NULL;
-		io->addr = 0;
-	}
-}
-
 static int ublk_ctrl_start_recovery(struct ublk_device *ub,
-		struct io_uring_cmd *cmd)
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	int ret = -EINVAL;
-	int i;
 
 	mutex_lock(&ub->mutex);
 	if (ublk_nosrv_should_stop_dev(ub))
 		goto out_unlock;
-	if (!ub->nr_queues_ready)
-		goto out_unlock;
 	/*
 	 * START_RECOVERY is only allowd after:
 	 *
@@ -2974,12 +3051,6 @@ static int ublk_ctrl_start_recovery(struct ublk_device *ub,
 		goto out_unlock;
 	}
 	pr_devel("%s: start recovery for dev id %d.\n", __func__, header->dev_id);
-	for (i = 0; i < ub->dev_info.nr_hw_queues; i++)
-		ublk_queue_reinit(ub, ublk_get_queue(ub, i));
-	/* set to NULL, otherwise new ubq_daemon cannot mmap the io_cmd_buf */
-	ub->mm = NULL;
-	ub->nr_queues_ready = 0;
-	ub->nr_privileged_daemon = 0;
 	init_completion(&ub->completion);
 	ret = 0;
  out_unlock:
@@ -2988,12 +3059,10 @@ static int ublk_ctrl_start_recovery(struct ublk_device *ub,
 }
 
 static int ublk_ctrl_end_recovery(struct ublk_device *ub,
-		struct io_uring_cmd *cmd)
+		const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	int ublksrv_pid = (int)header->data[0];
 	int ret = -EINVAL;
-	int i;
 
 	pr_devel("%s: Waiting for new ubq_daemons(nr: %d) are ready, dev id %d...\n",
 			__func__, ub->dev_info.nr_hw_queues, header->dev_id);
@@ -3013,33 +3082,18 @@ static int ublk_ctrl_end_recovery(struct ublk_device *ub,
 		goto out_unlock;
 	}
 	ub->dev_info.ublksrv_pid = ublksrv_pid;
+	ub->dev_info.state = UBLK_S_DEV_LIVE;
 	pr_devel("%s: new ublksrv_pid %d, dev id %d\n",
 			__func__, ublksrv_pid, header->dev_id);
-
-	if (ublk_nosrv_dev_should_queue_io(ub)) {
-		ub->dev_info.state = UBLK_S_DEV_LIVE;
-		blk_mq_unquiesce_queue(ub->ub_disk->queue);
-		pr_devel("%s: queue unquiesced, dev id %d.\n",
-				__func__, header->dev_id);
-		blk_mq_kick_requeue_list(ub->ub_disk->queue);
-	} else {
-		blk_mq_quiesce_queue(ub->ub_disk->queue);
-		ub->dev_info.state = UBLK_S_DEV_LIVE;
-		for (i = 0; i < ub->dev_info.nr_hw_queues; i++) {
-			ublk_get_queue(ub, i)->fail_io = false;
-		}
-		blk_mq_unquiesce_queue(ub->ub_disk->queue);
-	}
-
+	blk_mq_kick_requeue_list(ub->ub_disk->queue);
 	ret = 0;
  out_unlock:
 	mutex_unlock(&ub->mutex);
 	return ret;
 }
 
-static int ublk_ctrl_get_features(struct io_uring_cmd *cmd)
+static int ublk_ctrl_get_features(const struct ublksrv_ctrl_cmd *header)
 {
-	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
 	u64 features = UBLK_F_ALL;
 
@@ -3178,7 +3232,7 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd,
 		goto out;
 
 	if (cmd_op == UBLK_U_CMD_GET_FEATURES) {
-		ret = ublk_ctrl_get_features(cmd);
+		ret = ublk_ctrl_get_features(header);
 		goto out;
 	}
 
@@ -3195,17 +3249,17 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd,
 
 	switch (_IOC_NR(cmd_op)) {
 	case UBLK_CMD_START_DEV:
-		ret = ublk_ctrl_start_dev(ub, cmd);
+		ret = ublk_ctrl_start_dev(ub, header);
 		break;
 	case UBLK_CMD_STOP_DEV:
 		ret = ublk_ctrl_stop_dev(ub);
 		break;
 	case UBLK_CMD_GET_DEV_INFO:
 	case UBLK_CMD_GET_DEV_INFO2:
-		ret = ublk_ctrl_get_dev_info(ub, cmd);
+		ret = ublk_ctrl_get_dev_info(ub, header);
 		break;
 	case UBLK_CMD_ADD_DEV:
-		ret = ublk_ctrl_add_dev(cmd);
+		ret = ublk_ctrl_add_dev(header);
 		break;
 	case UBLK_CMD_DEL_DEV:
 		ret = ublk_ctrl_del_dev(&ub, true);
@@ -3214,19 +3268,19 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd,
 		ret = ublk_ctrl_del_dev(&ub, false);
 		break;
 	case UBLK_CMD_GET_QUEUE_AFFINITY:
-		ret = ublk_ctrl_get_queue_affinity(ub, cmd);
+		ret = ublk_ctrl_get_queue_affinity(ub, header);
 		break;
 	case UBLK_CMD_GET_PARAMS:
-		ret = ublk_ctrl_get_params(ub, cmd);
+		ret = ublk_ctrl_get_params(ub, header);
 		break;
 	case UBLK_CMD_SET_PARAMS:
-		ret = ublk_ctrl_set_params(ub, cmd);
+		ret = ublk_ctrl_set_params(ub, header);
 		break;
 	case UBLK_CMD_START_USER_RECOVERY:
-		ret = ublk_ctrl_start_recovery(ub, cmd);
+		ret = ublk_ctrl_start_recovery(ub, header);
 		break;
 	case UBLK_CMD_END_USER_RECOVERY:
-		ret = ublk_ctrl_end_recovery(ub, cmd);
+		ret = ublk_ctrl_end_recovery(ub, header);
 		break;
 	default:
 		ret = -EOPNOTSUPP;
diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c
index c1e69fcc9c4f..0a759ea26fd3 100644
--- a/drivers/bluetooth/btintel_pcie.c
+++ b/drivers/bluetooth/btintel_pcie.c
@@ -957,8 +957,10 @@ static int btintel_pcie_recv_event(struct hci_dev *hdev, struct sk_buff *skb)
 		/* This is a debug event that comes from IML and OP image when it
 		 * starts execution. There is no need pass this event to stack.
 		 */
-		if (skb->data[2] == 0x97)
+		if (skb->data[2] == 0x97) {
+			hci_recv_diag(hdev, skb);
 			return 0;
+		}
 	}
 
 	return hci_recv_frame(hdev, skb);
@@ -974,7 +976,6 @@ static int btintel_pcie_recv_frame(struct btintel_pcie_data *data,
 	u8 pkt_type;
 	u16 plen;
 	u32 pcie_pkt_type;
-	struct sk_buff *new_skb;
 	void *pdata;
 	struct hci_dev *hdev = data->hdev;
 
@@ -1051,24 +1052,20 @@ static int btintel_pcie_recv_frame(struct btintel_pcie_data *data,
 
 	bt_dev_dbg(hdev, "pkt_type: 0x%2.2x len: %u", pkt_type, plen);
 
-	new_skb = bt_skb_alloc(plen, GFP_ATOMIC);
-	if (!new_skb) {
-		bt_dev_err(hdev, "Failed to allocate memory for skb of len: %u",
-			   skb->len);
-		ret = -ENOMEM;
-		goto exit_error;
-	}
-
-	hci_skb_pkt_type(new_skb) = pkt_type;
-	skb_put_data(new_skb, skb->data, plen);
+	hci_skb_pkt_type(skb) = pkt_type;
 	hdev->stat.byte_rx += plen;
+	skb_trim(skb, plen);
 
 	if (pcie_pkt_type == BTINTEL_PCIE_HCI_EVT_PKT)
-		ret = btintel_pcie_recv_event(hdev, new_skb);
+		ret = btintel_pcie_recv_event(hdev, skb);
 	else
-		ret = hci_recv_frame(hdev, new_skb);
+		ret = hci_recv_frame(hdev, skb);
+	skb = NULL; /* skb is freed in the callee  */
 
 exit_error:
+	if (skb)
+		kfree_skb(skb);
+
 	if (ret)
 		hdev->stat.err_rx++;
 
@@ -1202,8 +1199,6 @@ static void btintel_pcie_rx_work(struct work_struct *work)
 	struct btintel_pcie_data *data = container_of(work,
 					struct btintel_pcie_data, rx_work);
 	struct sk_buff *skb;
-	int err;
-	struct hci_dev *hdev = data->hdev;
 
 	if (test_bit(BTINTEL_PCIE_HWEXP_INPROGRESS, &data->flags)) {
 		/* Unlike usb products, controller will not send hardware
@@ -1224,11 +1219,7 @@ static void btintel_pcie_rx_work(struct work_struct *work)
 
 	/* Process the sk_buf in queue and send to the HCI layer */
 	while ((skb = skb_dequeue(&data->rx_skb_q))) {
-		err = btintel_pcie_recv_frame(data, skb);
-		if (err)
-			bt_dev_err(hdev, "Failed to send received frame: %d",
-				   err);
-		kfree_skb(skb);
+		btintel_pcie_recv_frame(data, skb);
 	}
 }
 
@@ -1281,10 +1272,8 @@ static void btintel_pcie_msix_rx_handle(struct btintel_pcie_data *data)
 	bt_dev_dbg(hdev, "RXQ: cr_hia: %u  cr_tia: %u", cr_hia, cr_tia);
 
 	/* Check CR_TIA and CR_HIA for change */
-	if (cr_tia == cr_hia) {
-		bt_dev_warn(hdev, "RXQ: no new CD found");
+	if (cr_tia == cr_hia)
 		return;
-	}
 
 	rxq = &data->rxq;
 
@@ -1320,6 +1309,16 @@ static irqreturn_t btintel_pcie_msix_isr(int irq, void *data)
 	return IRQ_WAKE_THREAD;
 }
 
+static inline bool btintel_pcie_is_rxq_empty(struct btintel_pcie_data *data)
+{
+	return data->ia.cr_hia[BTINTEL_PCIE_RXQ_NUM] == data->ia.cr_tia[BTINTEL_PCIE_RXQ_NUM];
+}
+
+static inline bool btintel_pcie_is_txackq_empty(struct btintel_pcie_data *data)
+{
+	return data->ia.cr_tia[BTINTEL_PCIE_TXQ_NUM] == data->ia.cr_hia[BTINTEL_PCIE_TXQ_NUM];
+}
+
 static irqreturn_t btintel_pcie_irq_msix_handler(int irq, void *dev_id)
 {
 	struct msix_entry *entry = dev_id;
@@ -1351,12 +1350,18 @@ static irqreturn_t btintel_pcie_irq_msix_handler(int irq, void *dev_id)
 		btintel_pcie_msix_gp0_handler(data);
 
 	/* For TX */
-	if (intr_fh & BTINTEL_PCIE_MSIX_FH_INT_CAUSES_0)
+	if (intr_fh & BTINTEL_PCIE_MSIX_FH_INT_CAUSES_0) {
 		btintel_pcie_msix_tx_handle(data);
+		if (!btintel_pcie_is_rxq_empty(data))
+			btintel_pcie_msix_rx_handle(data);
+	}
 
 	/* For RX */
-	if (intr_fh & BTINTEL_PCIE_MSIX_FH_INT_CAUSES_1)
+	if (intr_fh & BTINTEL_PCIE_MSIX_FH_INT_CAUSES_1) {
 		btintel_pcie_msix_rx_handle(data);
+		if (!btintel_pcie_is_txackq_empty(data))
+			btintel_pcie_msix_tx_handle(data);
+	}
 
 	/*
 	 * Before sending the interrupt the HW disables it to prevent a nested
diff --git a/drivers/bluetooth/btmtksdio.c b/drivers/bluetooth/btmtksdio.c
index edd5eead1e93..1d26207b2ba7 100644
--- a/drivers/bluetooth/btmtksdio.c
+++ b/drivers/bluetooth/btmtksdio.c
@@ -723,6 +723,10 @@ static int btmtksdio_close(struct hci_dev *hdev)
 {
 	struct btmtksdio_dev *bdev = hci_get_drvdata(hdev);
 
+	/* Skip btmtksdio_close if BTMTKSDIO_FUNC_ENABLED isn't set */
+	if (!test_bit(BTMTKSDIO_FUNC_ENABLED, &bdev->tx_state))
+		return 0;
+
 	sdio_claim_host(bdev->func);
 
 	/* Disable interrupt */
@@ -1443,11 +1447,15 @@ static void btmtksdio_remove(struct sdio_func *func)
 	if (!bdev)
 		return;
 
+	hdev = bdev->hdev;
+
+	/* Make sure to call btmtksdio_close before removing sdio card */
+	if (test_bit(BTMTKSDIO_FUNC_ENABLED, &bdev->tx_state))
+		btmtksdio_close(hdev);
+
 	/* Be consistent the state in btmtksdio_probe */
 	pm_runtime_get_noresume(bdev->dev);
 
-	hdev = bdev->hdev;
-
 	sdio_set_drvdata(func, NULL);
 	hci_unregister_dev(hdev);
 	hci_free_dev(hdev);
diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 5091dea762a0..604ab2bba231 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -1286,7 +1286,9 @@ static void nxp_coredump(struct hci_dev *hdev)
 	u8 pcmd = 2;
 
 	skb = nxp_drv_send_cmd(hdev, HCI_NXP_TRIGGER_DUMP, 1, &pcmd);
-	if (!IS_ERR(skb))
+	if (IS_ERR(skb))
+		bt_dev_err(hdev, "Failed to trigger FW Dump. (%ld)", PTR_ERR(skb));
+	else
 		kfree_skb(skb);
 }
 
@@ -1445,9 +1447,6 @@ static int nxp_shutdown(struct hci_dev *hdev)
 		/* HCI_NXP_IND_RESET command may not returns any response */
 		if (!IS_ERR(skb))
 			kfree_skb(skb);
-	} else if (nxpdev->current_baudrate != nxpdev->fw_init_baudrate) {
-		nxpdev->new_baudrate = nxpdev->fw_init_baudrate;
-		nxp_set_baudrate_cmd(hdev, NULL);
 	}
 
 	return 0;
@@ -1799,13 +1798,15 @@ static void nxp_serdev_remove(struct serdev_device *serdev)
 		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
 		wake_up_interruptible(&nxpdev->check_boot_sign_wait_q);
 		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
-	}
-
-	if (test_bit(HCI_RUNNING, &hdev->flags)) {
-		/* Ensure shutdown callback is executed before unregistering, so
-		 * that baudrate is reset to initial value.
+	} else {
+		/* Restore FW baudrate to fw_init_baudrate if changed.
+		 * This will ensure FW baudrate is in sync with
+		 * driver baudrate in case this driver is re-inserted.
 		 */
-		nxp_shutdown(hdev);
+		if (nxpdev->current_baudrate != nxpdev->fw_init_baudrate) {
+			nxpdev->new_baudrate = nxpdev->fw_init_baudrate;
+			nxp_set_baudrate_cmd(hdev, NULL);
+		}
 	}
 
 	ps_cleanup(nxpdev);
diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c
index 3d6778b95e00..edefb9dc76aa 100644
--- a/drivers/bluetooth/btqca.c
+++ b/drivers/bluetooth/btqca.c
@@ -889,7 +889,7 @@ int qca_uart_setup(struct hci_dev *hdev, uint8_t baudrate,
 			if (le32_to_cpu(ver.soc_id) == QCA_WCN3950_SOC_ID_T)
 				variant = "t";
 			else if (le32_to_cpu(ver.soc_id) == QCA_WCN3950_SOC_ID_S)
-				variant = "u";
+				variant = "s";
 
 			snprintf(config.fwname, sizeof(config.fwname),
 				 "qca/cmnv%02x%s.bin", rom_ver, variant);
diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index d3eba0d4a57d..7838c89e529e 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -1215,6 +1215,8 @@ next:
 			rtl_dev_err(hdev, "mandatory config file %s not found",
 				    btrtl_dev->ic_info->cfg_name);
 			ret = btrtl_dev->cfg_len;
+			if (!ret)
+				ret = -EINVAL;
 			goto err_free;
 		}
 	}
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 5012b5ff92c8..a42dedb78e0a 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3010,22 +3010,16 @@ static void btusb_coredump_qca(struct hci_dev *hdev)
 		bt_dev_err(hdev, "%s: triggle crash failed (%d)", __func__, err);
 }
 
-/*
- * ==0: not a dump pkt.
- * < 0: fails to handle a dump pkt
- * > 0: otherwise.
- */
+/* Return: 0 on success, negative errno on failure. */
 static int handle_dump_pkt_qca(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	int ret = 1;
+	int ret = 0;
 	u8 pkt_type;
 	u8 *sk_ptr;
 	unsigned int sk_len;
 	u16 seqno;
 	u32 dump_size;
 
-	struct hci_event_hdr *event_hdr;
-	struct hci_acl_hdr *acl_hdr;
 	struct qca_dump_hdr *dump_hdr;
 	struct btusb_data *btdata = hci_get_drvdata(hdev);
 	struct usb_device *udev = btdata->udev;
@@ -3035,30 +3029,14 @@ static int handle_dump_pkt_qca(struct hci_dev *hdev, struct sk_buff *skb)
 	sk_len = skb->len;
 
 	if (pkt_type == HCI_ACLDATA_PKT) {
-		acl_hdr = hci_acl_hdr(skb);
-		if (le16_to_cpu(acl_hdr->handle) != QCA_MEMDUMP_ACL_HANDLE)
-			return 0;
 		sk_ptr += HCI_ACL_HDR_SIZE;
 		sk_len -= HCI_ACL_HDR_SIZE;
-		event_hdr = (struct hci_event_hdr *)sk_ptr;
-	} else {
-		event_hdr = hci_event_hdr(skb);
 	}
 
-	if ((event_hdr->evt != HCI_VENDOR_PKT)
-		|| (event_hdr->plen != (sk_len - HCI_EVENT_HDR_SIZE)))
-		return 0;
-
 	sk_ptr += HCI_EVENT_HDR_SIZE;
 	sk_len -= HCI_EVENT_HDR_SIZE;
 
 	dump_hdr = (struct qca_dump_hdr *)sk_ptr;
-	if ((sk_len < offsetof(struct qca_dump_hdr, data))
-		|| (dump_hdr->vse_class != QCA_MEMDUMP_VSE_CLASS)
-	    || (dump_hdr->msg_type != QCA_MEMDUMP_MSG_TYPE))
-		return 0;
-
-	/*it is dump pkt now*/
 	seqno = le16_to_cpu(dump_hdr->seqno);
 	if (seqno == 0) {
 		set_bit(BTUSB_HW_SSR_ACTIVE, &btdata->flags);
@@ -3132,17 +3110,84 @@ out:
 	return ret;
 }
 
+/* Return: true if the ACL packet is a dump packet, false otherwise. */
+static bool acl_pkt_is_dump_qca(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	u8 *sk_ptr;
+	unsigned int sk_len;
+
+	struct hci_event_hdr *event_hdr;
+	struct hci_acl_hdr *acl_hdr;
+	struct qca_dump_hdr *dump_hdr;
+
+	sk_ptr = skb->data;
+	sk_len = skb->len;
+
+	acl_hdr = hci_acl_hdr(skb);
+	if (le16_to_cpu(acl_hdr->handle) != QCA_MEMDUMP_ACL_HANDLE)
+		return false;
+
+	sk_ptr += HCI_ACL_HDR_SIZE;
+	sk_len -= HCI_ACL_HDR_SIZE;
+	event_hdr = (struct hci_event_hdr *)sk_ptr;
+
+	if ((event_hdr->evt != HCI_VENDOR_PKT) ||
+	    (event_hdr->plen != (sk_len - HCI_EVENT_HDR_SIZE)))
+		return false;
+
+	sk_ptr += HCI_EVENT_HDR_SIZE;
+	sk_len -= HCI_EVENT_HDR_SIZE;
+
+	dump_hdr = (struct qca_dump_hdr *)sk_ptr;
+	if ((sk_len < offsetof(struct qca_dump_hdr, data)) ||
+	    (dump_hdr->vse_class != QCA_MEMDUMP_VSE_CLASS) ||
+	    (dump_hdr->msg_type != QCA_MEMDUMP_MSG_TYPE))
+		return false;
+
+	return true;
+}
+
+/* Return: true if the event packet is a dump packet, false otherwise. */
+static bool evt_pkt_is_dump_qca(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	u8 *sk_ptr;
+	unsigned int sk_len;
+
+	struct hci_event_hdr *event_hdr;
+	struct qca_dump_hdr *dump_hdr;
+
+	sk_ptr = skb->data;
+	sk_len = skb->len;
+
+	event_hdr = hci_event_hdr(skb);
+
+	if ((event_hdr->evt != HCI_VENDOR_PKT)
+	    || (event_hdr->plen != (sk_len - HCI_EVENT_HDR_SIZE)))
+		return false;
+
+	sk_ptr += HCI_EVENT_HDR_SIZE;
+	sk_len -= HCI_EVENT_HDR_SIZE;
+
+	dump_hdr = (struct qca_dump_hdr *)sk_ptr;
+	if ((sk_len < offsetof(struct qca_dump_hdr, data)) ||
+	    (dump_hdr->vse_class != QCA_MEMDUMP_VSE_CLASS) ||
+	    (dump_hdr->msg_type != QCA_MEMDUMP_MSG_TYPE))
+		return false;
+
+	return true;
+}
+
 static int btusb_recv_acl_qca(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	if (handle_dump_pkt_qca(hdev, skb))
-		return 0;
+	if (acl_pkt_is_dump_qca(hdev, skb))
+		return handle_dump_pkt_qca(hdev, skb);
 	return hci_recv_frame(hdev, skb);
 }
 
 static int btusb_recv_evt_qca(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	if (handle_dump_pkt_qca(hdev, skb))
-		return 0;
+	if (evt_pkt_is_dump_qca(hdev, skb))
+		return handle_dump_pkt_qca(hdev, skb);
 	return hci_recv_frame(hdev, skb);
 }
 
diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c
index a51935d37e5d..59f4d7bdffdc 100644
--- a/drivers/bluetooth/hci_vhci.c
+++ b/drivers/bluetooth/hci_vhci.c
@@ -289,18 +289,18 @@ static void vhci_coredump(struct hci_dev *hdev)
 
 static void vhci_coredump_hdr(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	char buf[80];
+	const char *buf;
 
-	snprintf(buf, sizeof(buf), "Controller Name: vhci_ctrl\n");
+	buf = "Controller Name: vhci_ctrl\n";
 	skb_put_data(skb, buf, strlen(buf));
 
-	snprintf(buf, sizeof(buf), "Firmware Version: vhci_fw\n");
+	buf = "Firmware Version: vhci_fw\n";
 	skb_put_data(skb, buf, strlen(buf));
 
-	snprintf(buf, sizeof(buf), "Driver: vhci_drv\n");
+	buf = "Driver: vhci_drv\n";
 	skb_put_data(skb, buf, strlen(buf));
 
-	snprintf(buf, sizeof(buf), "Vendor: vhci\n");
+	buf = "Vendor: vhci\n";
 	skb_put_data(skb, buf, strlen(buf));
 }
 
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index ef30445527a2..bcc26785175d 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -53,6 +53,7 @@ struct intel_gtt_driver {
 	 * of the mmio register file, that's done in the generic code. */
 	void (*cleanup)(void);
 	void (*write_entry)(dma_addr_t addr, unsigned int entry, unsigned int flags);
+	dma_addr_t (*read_entry)(unsigned int entry, bool *is_present, bool *is_local);
 	/* Flags is a more or less chipset specific opaque value.
 	 * For chipsets that need to support old ums (non-gem) code, this
 	 * needs to be identical to the various supported agp memory types! */
@@ -336,6 +337,19 @@ static void i810_write_entry(dma_addr_t addr, unsigned int entry,
 	writel_relaxed(addr | pte_flags, intel_private.gtt + entry);
 }
 
+static dma_addr_t i810_read_entry(unsigned int entry,
+				  bool *is_present, bool *is_local)
+{
+	u32 val;
+
+	val = readl(intel_private.gtt + entry);
+
+	*is_present = val & I810_PTE_VALID;
+	*is_local = val & I810_PTE_LOCAL;
+
+	return val & ~0xfff;
+}
+
 static resource_size_t intel_gtt_stolen_size(void)
 {
 	u16 gmch_ctrl;
@@ -741,6 +755,19 @@ static void i830_write_entry(dma_addr_t addr, unsigned int entry,
 	writel_relaxed(addr | pte_flags, intel_private.gtt + entry);
 }
 
+static dma_addr_t i830_read_entry(unsigned int entry,
+				  bool *is_present, bool *is_local)
+{
+	u32 val;
+
+	val = readl(intel_private.gtt + entry);
+
+	*is_present = val & I810_PTE_VALID;
+	*is_local = false;
+
+	return val & ~0xfff;
+}
+
 bool intel_gmch_enable_gtt(void)
 {
 	u8 __iomem *reg;
@@ -878,6 +905,13 @@ void intel_gmch_gtt_insert_sg_entries(struct sg_table *st,
 }
 EXPORT_SYMBOL(intel_gmch_gtt_insert_sg_entries);
 
+dma_addr_t intel_gmch_gtt_read_entry(unsigned int pg,
+				     bool *is_present, bool *is_local)
+{
+	return intel_private.driver->read_entry(pg, is_present, is_local);
+}
+EXPORT_SYMBOL(intel_gmch_gtt_read_entry);
+
 #if IS_ENABLED(CONFIG_AGP_INTEL)
 static void intel_gmch_gtt_insert_pages(unsigned int first_entry,
 					unsigned int num_entries,
@@ -1126,6 +1160,19 @@ static void i965_write_entry(dma_addr_t addr,
 	writel_relaxed(addr | pte_flags, intel_private.gtt + entry);
 }
 
+static dma_addr_t i965_read_entry(unsigned int entry,
+				  bool *is_present, bool *is_local)
+{
+	u64 val;
+
+	val = readl(intel_private.gtt + entry);
+
+	*is_present = val & I810_PTE_VALID;
+	*is_local = false;
+
+	return ((val & 0xf0) << 28) | (val & ~0xfff);
+}
+
 static int i9xx_setup(void)
 {
 	phys_addr_t reg_addr;
@@ -1187,6 +1234,7 @@ static const struct intel_gtt_driver i81x_gtt_driver = {
 	.cleanup = i810_cleanup,
 	.check_flags = i830_check_flags,
 	.write_entry = i810_write_entry,
+	.read_entry = i810_read_entry,
 };
 static const struct intel_gtt_driver i8xx_gtt_driver = {
 	.gen = 2,
@@ -1194,6 +1242,7 @@ static const struct intel_gtt_driver i8xx_gtt_driver = {
 	.setup = i830_setup,
 	.cleanup = i830_cleanup,
 	.write_entry = i830_write_entry,
+	.read_entry = i830_read_entry,
 	.dma_mask_size = 32,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i830_chipset_flush,
@@ -1205,6 +1254,7 @@ static const struct intel_gtt_driver i915_gtt_driver = {
 	.cleanup = i9xx_cleanup,
 	/* i945 is the last gpu to need phys mem (for overlay and cursors). */
 	.write_entry = i830_write_entry,
+	.read_entry = i830_read_entry,
 	.dma_mask_size = 32,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i9xx_chipset_flush,
@@ -1215,6 +1265,7 @@ static const struct intel_gtt_driver g33_gtt_driver = {
 	.setup = i9xx_setup,
 	.cleanup = i9xx_cleanup,
 	.write_entry = i965_write_entry,
+	.read_entry = i965_read_entry,
 	.dma_mask_size = 36,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i9xx_chipset_flush,
@@ -1225,6 +1276,7 @@ static const struct intel_gtt_driver pineview_gtt_driver = {
 	.setup = i9xx_setup,
 	.cleanup = i9xx_cleanup,
 	.write_entry = i965_write_entry,
+	.read_entry = i965_read_entry,
 	.dma_mask_size = 36,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i9xx_chipset_flush,
@@ -1235,6 +1287,7 @@ static const struct intel_gtt_driver i965_gtt_driver = {
 	.setup = i9xx_setup,
 	.cleanup = i9xx_cleanup,
 	.write_entry = i965_write_entry,
+	.read_entry = i965_read_entry,
 	.dma_mask_size = 36,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i9xx_chipset_flush,
@@ -1244,6 +1297,7 @@ static const struct intel_gtt_driver g4x_gtt_driver = {
 	.setup = i9xx_setup,
 	.cleanup = i9xx_cleanup,
 	.write_entry = i965_write_entry,
+	.read_entry = i965_read_entry,
 	.dma_mask_size = 36,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i9xx_chipset_flush,
@@ -1254,6 +1308,7 @@ static const struct intel_gtt_driver ironlake_gtt_driver = {
 	.setup = i9xx_setup,
 	.cleanup = i9xx_cleanup,
 	.write_entry = i965_write_entry,
+	.read_entry = i965_read_entry,
 	.dma_mask_size = 36,
 	.check_flags = i830_check_flags,
 	.chipset_flush = i9xx_chipset_flush,
diff --git a/drivers/char/misc.c b/drivers/char/misc.c
index f7dd455dd0dd..dda466f9181a 100644
--- a/drivers/char/misc.c
+++ b/drivers/char/misc.c
@@ -315,7 +315,7 @@ static int __init misc_init(void)
 		goto fail_remove;
 
 	err = -EIO;
-	if (register_chrdev(MISC_MAJOR, "misc", &misc_fops))
+	if (__register_chrdev(MISC_MAJOR, 0, MINORMASK + 1, "misc", &misc_fops))
 		goto fail_printk;
 	return 0;
 
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index 5f04951d0dd4..088182e54deb 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -1576,8 +1576,8 @@ static void handle_control_message(struct virtio_device *vdev,
 		break;
 	case VIRTIO_CONSOLE_RESIZE: {
 		struct {
-			__u16 rows;
-			__u16 cols;
+			__virtio16 cols;
+			__virtio16 rows;
 		} size;
 
 		if (!is_console_port(port))
@@ -1585,7 +1585,8 @@ static void handle_control_message(struct virtio_device *vdev,
 
 		memcpy(&size, buf->buf + buf->offset + sizeof(*cpkt),
 		       sizeof(size));
-		set_console_size(port, size.rows, size.cols);
+		set_console_size(port, virtio16_to_cpu(vdev, size.rows),
+				 virtio16_to_cpu(vdev, size.cols));
 
 		port->cons.hvc->irq_requested = 1;
 		resize_console(port);
diff --git a/drivers/comedi/drivers/jr3_pci.c b/drivers/comedi/drivers/jr3_pci.c
index cdc842b32bab..75dce1ff2419 100644
--- a/drivers/comedi/drivers/jr3_pci.c
+++ b/drivers/comedi/drivers/jr3_pci.c
@@ -758,7 +758,7 @@ static void jr3_pci_detach(struct comedi_device *dev)
 	struct jr3_pci_dev_private *devpriv = dev->private;
 
 	if (devpriv)
-		timer_delete_sync(&devpriv->timer);
+		timer_shutdown_sync(&devpriv->timer);
 
 	comedi_pci_detach(dev);
 }
diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
index 4f9cb943d945..0d46402e3094 100644
--- a/drivers/cpufreq/Kconfig.arm
+++ b/drivers/cpufreq/Kconfig.arm
@@ -76,7 +76,7 @@ config ARM_VEXPRESS_SPC_CPUFREQ
 config ARM_BRCMSTB_AVS_CPUFREQ
 	tristate "Broadcom STB AVS CPUfreq driver"
 	depends on (ARCH_BRCMSTB && !ARM_SCMI_CPUFREQ) || COMPILE_TEST
-	default y
+	default y if ARCH_BRCMSTB && !ARM_SCMI_CPUFREQ
 	help
 	  Some Broadcom STB SoCs use a co-processor running proprietary firmware
 	  ("AVS") to handle voltage and frequency scaling. This driver provides
@@ -88,7 +88,7 @@ config ARM_HIGHBANK_CPUFREQ
 	tristate "Calxeda Highbank-based"
 	depends on ARCH_HIGHBANK || COMPILE_TEST
 	depends on CPUFREQ_DT && REGULATOR && PL320_MBOX
-	default m
+	default m if ARCH_HIGHBANK
 	help
 	  This adds the CPUFreq driver for Calxeda Highbank SoC
 	  based boards.
@@ -133,7 +133,7 @@ config ARM_MEDIATEK_CPUFREQ
 config ARM_MEDIATEK_CPUFREQ_HW
 	tristate "MediaTek CPUFreq HW driver"
 	depends on ARCH_MEDIATEK || COMPILE_TEST
-	default m
+	default m if ARCH_MEDIATEK
 	help
 	  Support for the CPUFreq HW driver.
 	  Some MediaTek chipsets have a HW engine to offload the steps
@@ -181,7 +181,7 @@ config ARM_RASPBERRYPI_CPUFREQ
 config ARM_S3C64XX_CPUFREQ
 	bool "Samsung S3C64XX"
 	depends on CPU_S3C6410 || COMPILE_TEST
-	default y
+	default CPU_S3C6410
 	help
 	  This adds the CPUFreq driver for Samsung S3C6410 SoC.
 
@@ -190,7 +190,7 @@ config ARM_S3C64XX_CPUFREQ
 config ARM_S5PV210_CPUFREQ
 	bool "Samsung S5PV210 and S5PC110"
 	depends on CPU_S5PV210 || COMPILE_TEST
-	default y
+	default CPU_S5PV210
 	help
 	  This adds the CPUFreq driver for Samsung S5PV210 and
 	  S5PC110 SoCs.
@@ -214,7 +214,7 @@ config ARM_SCMI_CPUFREQ
 config ARM_SPEAR_CPUFREQ
 	bool "SPEAr CPUFreq support"
 	depends on PLAT_SPEAR || COMPILE_TEST
-	default y
+	default PLAT_SPEAR
 	help
 	  This adds the CPUFreq driver support for SPEAr SOCs.
 
@@ -233,7 +233,7 @@ config ARM_TEGRA20_CPUFREQ
 	tristate "Tegra20/30 CPUFreq support"
 	depends on ARCH_TEGRA || COMPILE_TEST
 	depends on CPUFREQ_DT
-	default y
+	default ARCH_TEGRA
 	help
 	  This adds the CPUFreq driver support for Tegra20/30 SOCs.
 
@@ -241,7 +241,7 @@ config ARM_TEGRA124_CPUFREQ
 	bool "Tegra124 CPUFreq support"
 	depends on ARCH_TEGRA || COMPILE_TEST
 	depends on CPUFREQ_DT
-	default y
+	default ARCH_TEGRA
 	help
 	  This adds the CPUFreq driver support for Tegra124 SOCs.
 
@@ -256,14 +256,14 @@ config ARM_TEGRA194_CPUFREQ
 	tristate "Tegra194 CPUFreq support"
 	depends on ARCH_TEGRA_194_SOC || ARCH_TEGRA_234_SOC || (64BIT && COMPILE_TEST)
 	depends on TEGRA_BPMP
-	default y
+	default ARCH_TEGRA_194_SOC || ARCH_TEGRA_234_SOC
 	help
 	  This adds CPU frequency driver support for Tegra194 SOCs.
 
 config ARM_TI_CPUFREQ
 	bool "Texas Instruments CPUFreq support"
 	depends on ARCH_OMAP2PLUS || ARCH_K3 || COMPILE_TEST
-	default y
+	default ARCH_OMAP2PLUS || ARCH_K3
 	help
 	  This driver enables valid OPPs on the running platform based on
 	  values contained within the SoC in use. Enable this in order to
diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c
index 924314cdeebc..d26b610e4f24 100644
--- a/drivers/cpufreq/acpi-cpufreq.c
+++ b/drivers/cpufreq/acpi-cpufreq.c
@@ -909,8 +909,19 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	if (perf->states[0].core_frequency * 1000 != freq_table[0].frequency)
 		pr_warn(FW_WARN "P-state 0 is not max freq\n");
 
-	if (acpi_cpufreq_driver.set_boost)
-		policy->boost_supported = true;
+	if (acpi_cpufreq_driver.set_boost) {
+		if (policy->boost_supported) {
+			/*
+			 * The firmware may have altered boost state while the
+			 * CPU was offline (for example during a suspend-resume
+			 * cycle).
+			 */
+			if (policy->boost_enabled != boost_state(cpu))
+				set_boost(policy, policy->boost_enabled);
+		} else {
+			policy->boost_supported = true;
+		}
+	}
 
 	return result;
 
diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
index 6789eed1bb5b..b961f3a3b580 100644
--- a/drivers/cpufreq/amd-pstate.c
+++ b/drivers/cpufreq/amd-pstate.c
@@ -607,13 +607,16 @@ static void amd_pstate_update_min_max_limit(struct cpufreq_policy *policy)
 	union perf_cached perf = READ_ONCE(cpudata->perf);
 
 	perf.max_limit_perf = freq_to_perf(perf, cpudata->nominal_freq, policy->max);
-	perf.min_limit_perf = freq_to_perf(perf, cpudata->nominal_freq, policy->min);
+	WRITE_ONCE(cpudata->max_limit_freq, policy->max);
 
-	if (cpudata->policy == CPUFREQ_POLICY_PERFORMANCE)
+	if (cpudata->policy == CPUFREQ_POLICY_PERFORMANCE) {
 		perf.min_limit_perf = min(perf.nominal_perf, perf.max_limit_perf);
+		WRITE_ONCE(cpudata->min_limit_freq, min(cpudata->nominal_freq, cpudata->max_limit_freq));
+	} else {
+		perf.min_limit_perf = freq_to_perf(perf, cpudata->nominal_freq, policy->min);
+		WRITE_ONCE(cpudata->min_limit_freq, policy->min);
+	}
 
-	WRITE_ONCE(cpudata->max_limit_freq, policy->max);
-	WRITE_ONCE(cpudata->min_limit_freq, policy->min);
 	WRITE_ONCE(cpudata->perf, perf);
 }
 
@@ -791,16 +794,6 @@ static void amd_perf_ctl_reset(unsigned int cpu)
 	wrmsrl_on_cpu(cpu, MSR_AMD_PERF_CTL, 0);
 }
 
-/*
- * Set amd-pstate preferred core enable can't be done directly from cpufreq callbacks
- * due to locking, so queue the work for later.
- */
-static void amd_pstste_sched_prefcore_workfn(struct work_struct *work)
-{
-	sched_set_itmt_support();
-}
-static DECLARE_WORK(sched_prefcore_work, amd_pstste_sched_prefcore_workfn);
-
 #define CPPC_MAX_PERF	U8_MAX
 
 static void amd_pstate_init_prefcore(struct amd_cpudata *cpudata)
@@ -811,14 +804,8 @@ static void amd_pstate_init_prefcore(struct amd_cpudata *cpudata)
 
 	cpudata->hw_prefcore = true;
 
-	/*
-	 * The priorities can be set regardless of whether or not
-	 * sched_set_itmt_support(true) has been called and it is valid to
-	 * update them at any time after it has been called.
-	 */
+	/* Priorities must be initialized before ITMT support can be toggled on. */
 	sched_set_itmt_core_prio((int)READ_ONCE(cpudata->prefcore_ranking), cpudata->cpu);
-
-	schedule_work(&sched_prefcore_work);
 }
 
 static void amd_pstate_update_limits(unsigned int cpu)
@@ -1193,6 +1180,9 @@ static ssize_t show_energy_performance_preference(
 
 static void amd_pstate_driver_cleanup(void)
 {
+	if (amd_pstate_prefcore)
+		sched_clear_itmt_support();
+
 	cppc_state = AMD_PSTATE_DISABLE;
 	current_pstate_driver = NULL;
 }
@@ -1235,6 +1225,10 @@ static int amd_pstate_register_driver(int mode)
 		return ret;
 	}
 
+	/* Enable ITMT support once all CPUs have initialized their asym priorities. */
+	if (amd_pstate_prefcore)
+		sched_set_itmt_support();
+
 	return 0;
 }
 
diff --git a/drivers/cpufreq/apple-soc-cpufreq.c b/drivers/cpufreq/apple-soc-cpufreq.c
index 4994c86feb57..b1d29b7af232 100644
--- a/drivers/cpufreq/apple-soc-cpufreq.c
+++ b/drivers/cpufreq/apple-soc-cpufreq.c
@@ -134,11 +134,17 @@ static const struct of_device_id apple_soc_cpufreq_of_match[] __maybe_unused = {
 
 static unsigned int apple_soc_cpufreq_get_rate(unsigned int cpu)
 {
-	struct cpufreq_policy *policy = cpufreq_cpu_get_raw(cpu);
-	struct apple_cpu_priv *priv = policy->driver_data;
+	struct cpufreq_policy *policy;
+	struct apple_cpu_priv *priv;
 	struct cpufreq_frequency_table *p;
 	unsigned int pstate;
 
+	policy = cpufreq_cpu_get_raw(cpu);
+	if (unlikely(!policy))
+		return 0;
+
+	priv = policy->driver_data;
+
 	if (priv->info->cur_pstate_mask) {
 		u32 reg = readl_relaxed(priv->reg_base + APPLE_DVFS_STATUS);
 
diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
index b3d74f9adcf0..cb93f00bafdb 100644
--- a/drivers/cpufreq/cppc_cpufreq.c
+++ b/drivers/cpufreq/cppc_cpufreq.c
@@ -747,7 +747,7 @@ static unsigned int cppc_cpufreq_get_rate(unsigned int cpu)
 	int ret;
 
 	if (!policy)
-		return -ENODEV;
+		return 0;
 
 	cpu_data = policy->driver_data;
 
diff --git a/drivers/cpufreq/cpufreq-dt-platdev.c b/drivers/cpufreq/cpufreq-dt-platdev.c
index 2aa00769cf09..a010da0f6337 100644
--- a/drivers/cpufreq/cpufreq-dt-platdev.c
+++ b/drivers/cpufreq/cpufreq-dt-platdev.c
@@ -175,6 +175,7 @@ static const struct of_device_id blocklist[] __initconst = {
 	{ .compatible = "qcom,sm8350", },
 	{ .compatible = "qcom,sm8450", },
 	{ .compatible = "qcom,sm8550", },
+	{ .compatible = "qcom,sm8650", },
 
 	{ .compatible = "st,stih407", },
 	{ .compatible = "st,stih410", },
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 3841c9da6cac..f45ded62b0e0 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -536,16 +536,18 @@ void cpufreq_disable_fast_switch(struct cpufreq_policy *policy)
 EXPORT_SYMBOL_GPL(cpufreq_disable_fast_switch);
 
 static unsigned int __resolve_freq(struct cpufreq_policy *policy,
-		unsigned int target_freq, unsigned int relation)
+				   unsigned int target_freq,
+				   unsigned int min, unsigned int max,
+				   unsigned int relation)
 {
 	unsigned int idx;
 
-	target_freq = clamp_val(target_freq, policy->min, policy->max);
+	target_freq = clamp_val(target_freq, min, max);
 
 	if (!policy->freq_table)
 		return target_freq;
 
-	idx = cpufreq_frequency_table_target(policy, target_freq, relation);
+	idx = cpufreq_frequency_table_target(policy, target_freq, min, max, relation);
 	policy->cached_resolved_idx = idx;
 	policy->cached_target_freq = target_freq;
 	return policy->freq_table[idx].frequency;
@@ -565,7 +567,21 @@ static unsigned int __resolve_freq(struct cpufreq_policy *policy,
 unsigned int cpufreq_driver_resolve_freq(struct cpufreq_policy *policy,
 					 unsigned int target_freq)
 {
-	return __resolve_freq(policy, target_freq, CPUFREQ_RELATION_LE);
+	unsigned int min = READ_ONCE(policy->min);
+	unsigned int max = READ_ONCE(policy->max);
+
+	/*
+	 * If this function runs in parallel with cpufreq_set_policy(), it may
+	 * read policy->min before the update and policy->max after the update
+	 * or the other way around, so there is no ordering guarantee.
+	 *
+	 * Resolve this by always honoring the max (in case it comes from
+	 * thermal throttling or similar).
+	 */
+	if (unlikely(min > max))
+		min = max;
+
+	return __resolve_freq(policy, target_freq, min, max, CPUFREQ_RELATION_LE);
 }
 EXPORT_SYMBOL_GPL(cpufreq_driver_resolve_freq);
 
@@ -2384,7 +2400,8 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
 	if (cpufreq_disabled())
 		return -ENODEV;
 
-	target_freq = __resolve_freq(policy, target_freq, relation);
+	target_freq = __resolve_freq(policy, target_freq, policy->min,
+				     policy->max, relation);
 
 	pr_debug("target for CPU %u: %u kHz, relation %u, requested %u kHz\n",
 		 policy->cpu, target_freq, relation, old_target_freq);
@@ -2708,11 +2725,18 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
 	 * Resolve policy min/max to available frequencies. It ensures
 	 * no frequency resolution will neither overshoot the requested maximum
 	 * nor undershoot the requested minimum.
+	 *
+	 * Avoid storing intermediate values in policy->max or policy->min and
+	 * compiler optimizations around them because they may be accessed
+	 * concurrently by cpufreq_driver_resolve_freq() during the update.
 	 */
-	policy->min = new_data.min;
-	policy->max = new_data.max;
-	policy->min = __resolve_freq(policy, policy->min, CPUFREQ_RELATION_L);
-	policy->max = __resolve_freq(policy, policy->max, CPUFREQ_RELATION_H);
+	WRITE_ONCE(policy->max, __resolve_freq(policy, new_data.max,
+					       new_data.min, new_data.max,
+					       CPUFREQ_RELATION_H));
+	new_data.min = __resolve_freq(policy, new_data.min, new_data.min,
+				      new_data.max, CPUFREQ_RELATION_L);
+	WRITE_ONCE(policy->min, new_data.min > policy->max ? policy->max : new_data.min);
+
 	trace_cpu_frequency_limits(policy);
 
 	cpufreq_update_pressure(policy);
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index a7c38b8b3e78..0e65d37c9231 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -76,7 +76,8 @@ static unsigned int generic_powersave_bias_target(struct cpufreq_policy *policy,
 		return freq_next;
 	}
 
-	index = cpufreq_frequency_table_target(policy, freq_next, relation);
+	index = cpufreq_frequency_table_target(policy, freq_next, policy->min,
+					       policy->max, relation);
 	freq_req = freq_table[index].frequency;
 	freq_reduc = freq_req * od_tuners->powersave_bias / 1000;
 	freq_avg = freq_req - freq_reduc;
diff --git a/drivers/cpufreq/freq_table.c b/drivers/cpufreq/freq_table.c
index c03a91502f84..35de513af6c9 100644
--- a/drivers/cpufreq/freq_table.c
+++ b/drivers/cpufreq/freq_table.c
@@ -115,8 +115,8 @@ int cpufreq_generic_frequency_table_verify(struct cpufreq_policy_data *policy)
 EXPORT_SYMBOL_GPL(cpufreq_generic_frequency_table_verify);
 
 int cpufreq_table_index_unsorted(struct cpufreq_policy *policy,
-				 unsigned int target_freq,
-				 unsigned int relation)
+				 unsigned int target_freq, unsigned int min,
+				 unsigned int max, unsigned int relation)
 {
 	struct cpufreq_frequency_table optimal = {
 		.driver_data = ~0,
@@ -147,7 +147,7 @@ int cpufreq_table_index_unsorted(struct cpufreq_policy *policy,
 	cpufreq_for_each_valid_entry_idx(pos, table, i) {
 		freq = pos->frequency;
 
-		if ((freq < policy->min) || (freq > policy->max))
+		if (freq < min || freq > max)
 			continue;
 		if (freq == target_freq) {
 			optimal.driver_data = i;
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index 4aad79d26c64..ba9bf06f1c77 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -598,6 +598,9 @@ static bool turbo_is_disabled(void)
 {
 	u64 misc_en;
 
+	if (!cpu_feature_enabled(X86_FEATURE_IDA))
+		return true;
+
 	rdmsrl(MSR_IA32_MISC_ENABLE, misc_en);
 
 	return !!(misc_en & MSR_IA32_MISC_ENABLE_TURBO_DISABLE);
@@ -2209,7 +2212,7 @@ static int knl_get_turbo_pstate(int cpu)
 static int hwp_get_cpu_scaling(int cpu)
 {
 	if (hybrid_scaling_factor) {
-		struct cpuinfo_x86 *c = &cpu_data(smp_processor_id());
+		struct cpuinfo_x86 *c = &cpu_data(cpu);
 		u8 cpu_type = c->topo.intel_type;
 
 		/*
diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c
index c310aeebc8f3..944e899eb1be 100644
--- a/drivers/cpufreq/scmi-cpufreq.c
+++ b/drivers/cpufreq/scmi-cpufreq.c
@@ -37,11 +37,17 @@ static struct cpufreq_driver scmi_cpufreq_driver;
 
 static unsigned int scmi_cpufreq_get_rate(unsigned int cpu)
 {
-	struct cpufreq_policy *policy = cpufreq_cpu_get_raw(cpu);
-	struct scmi_data *priv = policy->driver_data;
+	struct cpufreq_policy *policy;
+	struct scmi_data *priv;
 	unsigned long rate;
 	int ret;
 
+	policy = cpufreq_cpu_get_raw(cpu);
+	if (unlikely(!policy))
+		return 0;
+
+	priv = policy->driver_data;
+
 	ret = perf_ops->freq_get(ph, priv->domain_id, &rate, false);
 	if (ret)
 		return 0;
diff --git a/drivers/cpufreq/scpi-cpufreq.c b/drivers/cpufreq/scpi-cpufreq.c
index 17cda84f00df..dcbb0ae7dd47 100644
--- a/drivers/cpufreq/scpi-cpufreq.c
+++ b/drivers/cpufreq/scpi-cpufreq.c
@@ -29,9 +29,16 @@ static struct scpi_ops *scpi_ops;
 
 static unsigned int scpi_cpufreq_get_rate(unsigned int cpu)
 {
-	struct cpufreq_policy *policy = cpufreq_cpu_get_raw(cpu);
-	struct scpi_data *priv = policy->driver_data;
-	unsigned long rate = clk_get_rate(priv->clk);
+	struct cpufreq_policy *policy;
+	struct scpi_data *priv;
+	unsigned long rate;
+
+	policy = cpufreq_cpu_get_raw(cpu);
+	if (unlikely(!policy))
+		return 0;
+
+	priv = policy->driver_data;
+	rate = clk_get_rate(priv->clk);
 
 	return rate / 1000;
 }
diff --git a/drivers/cpufreq/sun50i-cpufreq-nvmem.c b/drivers/cpufreq/sun50i-cpufreq-nvmem.c
index 47d6840b3489..744312a44279 100644
--- a/drivers/cpufreq/sun50i-cpufreq-nvmem.c
+++ b/drivers/cpufreq/sun50i-cpufreq-nvmem.c
@@ -194,7 +194,9 @@ static int sun50i_cpufreq_get_efuse(void)
 	struct nvmem_cell *speedbin_nvmem;
 	const struct of_device_id *match;
 	struct device *cpu_dev;
-	u32 *speedbin;
+	void *speedbin_ptr;
+	u32 speedbin = 0;
+	size_t len;
 	int ret;
 
 	cpu_dev = get_cpu_device(0);
@@ -217,14 +219,18 @@ static int sun50i_cpufreq_get_efuse(void)
 		return dev_err_probe(cpu_dev, PTR_ERR(speedbin_nvmem),
 				     "Could not get nvmem cell\n");
 
-	speedbin = nvmem_cell_read(speedbin_nvmem, NULL);
+	speedbin_ptr = nvmem_cell_read(speedbin_nvmem, &len);
 	nvmem_cell_put(speedbin_nvmem);
-	if (IS_ERR(speedbin))
-		return PTR_ERR(speedbin);
+	if (IS_ERR(speedbin_ptr))
+		return PTR_ERR(speedbin_ptr);
 
-	ret = opp_data->efuse_xlate(*speedbin);
+	if (len <= 4)
+		memcpy(&speedbin, speedbin_ptr, len);
+	speedbin = le32_to_cpu(speedbin);
 
-	kfree(speedbin);
+	ret = opp_data->efuse_xlate(speedbin);
+
+	kfree(speedbin_ptr);
 
 	return ret;
 };
diff --git a/drivers/crypto/atmel-sha204a.c b/drivers/crypto/atmel-sha204a.c
index 75bebec2c757..0fcf4a39de27 100644
--- a/drivers/crypto/atmel-sha204a.c
+++ b/drivers/crypto/atmel-sha204a.c
@@ -163,6 +163,12 @@ static int atmel_sha204a_probe(struct i2c_client *client)
 	i2c_priv->hwrng.name = dev_name(&client->dev);
 	i2c_priv->hwrng.read = atmel_sha204a_rng_read;
 
+	/*
+	 * According to review by Bill Cox [1], this HWRNG has very low entropy.
+	 * [1] https://www.metzdowd.com/pipermail/cryptography/2014-December/023858.html
+	 */
+	i2c_priv->hwrng.quality = 1;
+
 	ret = devm_hwrng_register(&client->dev, &i2c_priv->hwrng);
 	if (ret)
 		dev_warn(&client->dev, "failed to register RNG (%d)\n", ret);
diff --git a/drivers/crypto/caam/qi.c b/drivers/crypto/caam/qi.c
index 7701d00bcb3a..b6e7c0b29d4e 100644
--- a/drivers/crypto/caam/qi.c
+++ b/drivers/crypto/caam/qi.c
@@ -122,12 +122,12 @@ int caam_qi_enqueue(struct device *qidev, struct caam_drv_req *req)
 	qm_fd_addr_set64(&fd, addr);
 
 	do {
+		refcount_inc(&req->drv_ctx->refcnt);
 		ret = qman_enqueue(req->drv_ctx->req_fq, &fd);
-		if (likely(!ret)) {
-			refcount_inc(&req->drv_ctx->refcnt);
+		if (likely(!ret))
 			return 0;
-		}
 
+		refcount_dec(&req->drv_ctx->refcnt);
 		if (ret != -EBUSY)
 			break;
 		num_retries++;
diff --git a/drivers/crypto/tegra/tegra-se-aes.c b/drivers/crypto/tegra/tegra-se-aes.c
index ca9d0cca1f74..0e07d0523291 100644
--- a/drivers/crypto/tegra/tegra-se-aes.c
+++ b/drivers/crypto/tegra/tegra-se-aes.c
@@ -269,7 +269,7 @@ static int tegra_aes_do_one_req(struct crypto_engine *engine, void *areq)
 	unsigned int cmdlen, key1_id, key2_id;
 	int ret;
 
-	rctx->iv = (u32 *)req->iv;
+	rctx->iv = (ctx->alg == SE_ALG_ECB) ? NULL : (u32 *)req->iv;
 	rctx->len = req->cryptlen;
 	key1_id = ctx->key1_id;
 	key2_id = ctx->key2_id;
@@ -498,9 +498,6 @@ static int tegra_aes_crypt(struct skcipher_request *req, bool encrypt)
 	if (!req->cryptlen)
 		return 0;
 
-	if (ctx->alg == SE_ALG_ECB)
-		req->iv = NULL;
-
 	rctx->encrypt = encrypt;
 
 	return crypto_transfer_skcipher_request_to_engine(ctx->se->engine, req);
diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
index 15699299dc11..17b692eb3257 100644
--- a/drivers/cxl/core/core.h
+++ b/drivers/cxl/core/core.h
@@ -119,7 +119,7 @@ int cxl_port_get_switch_dport_bandwidth(struct cxl_port *port,
 
 int cxl_ras_init(void);
 void cxl_ras_exit(void);
-int cxl_gpf_port_setup(struct device *dport_dev, struct cxl_port *port);
+int cxl_gpf_port_setup(struct cxl_dport *dport);
 int cxl_acpi_get_extended_linear_cache_size(struct resource *backing_res,
 					    int nid, resource_size_t *size);
 
diff --git a/drivers/cxl/core/features.c b/drivers/cxl/core/features.c
index f4daefe3180e..1498e2369c37 100644
--- a/drivers/cxl/core/features.c
+++ b/drivers/cxl/core/features.c
@@ -528,13 +528,13 @@ static void *cxlctl_set_feature(struct cxl_features_state *cxlfs,
 	rc = cxl_set_feature(cxl_mbox, &feat_in->uuid,
 			     feat_in->version, feat_in->feat_data,
 			     data_size, flags, offset, &return_code);
+	*out_len = sizeof(*rpc_out);
 	if (rc) {
 		rpc_out->retval = return_code;
 		return no_free_ptr(rpc_out);
 	}
 
 	rpc_out->retval = CXL_MBOX_CMD_RC_SUCCESS;
-	*out_len = sizeof(*rpc_out);
 
 	return no_free_ptr(rpc_out);
 }
@@ -677,7 +677,7 @@ static void free_memdev_fwctl(void *_fwctl_dev)
 	fwctl_put(fwctl_dev);
 }
 
-int devm_cxl_setup_fwctl(struct cxl_memdev *cxlmd)
+int devm_cxl_setup_fwctl(struct device *host, struct cxl_memdev *cxlmd)
 {
 	struct cxl_dev_state *cxlds = cxlmd->cxlds;
 	struct cxl_features_state *cxlfs;
@@ -700,7 +700,7 @@ int devm_cxl_setup_fwctl(struct cxl_memdev *cxlmd)
 	if (rc)
 		return rc;
 
-	return devm_add_action_or_reset(&cxlmd->dev, free_memdev_fwctl,
+	return devm_add_action_or_reset(host, free_memdev_fwctl,
 					no_free_ptr(fwctl_dev));
 }
 EXPORT_SYMBOL_NS_GPL(devm_cxl_setup_fwctl, "CXL");
diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
index 96fecb799cbc..3b80e9a76ba8 100644
--- a/drivers/cxl/core/pci.c
+++ b/drivers/cxl/core/pci.c
@@ -1072,14 +1072,20 @@ int cxl_pci_get_bandwidth(struct pci_dev *pdev, struct access_coordinate *c)
 #define GPF_TIMEOUT_BASE_MAX 2
 #define GPF_TIMEOUT_SCALE_MAX 7 /* 10 seconds */
 
-u16 cxl_gpf_get_dvsec(struct device *dev, bool is_port)
+u16 cxl_gpf_get_dvsec(struct device *dev)
 {
+	struct pci_dev *pdev;
+	bool is_port = true;
 	u16 dvsec;
 
 	if (!dev_is_pci(dev))
 		return 0;
 
-	dvsec = pci_find_dvsec_capability(to_pci_dev(dev), PCI_VENDOR_ID_CXL,
+	pdev = to_pci_dev(dev);
+	if (pci_pcie_type(pdev) == PCI_EXP_TYPE_ENDPOINT)
+		is_port = false;
+
+	dvsec = pci_find_dvsec_capability(pdev, PCI_VENDOR_ID_CXL,
 			is_port ? CXL_DVSEC_PORT_GPF : CXL_DVSEC_DEVICE_GPF);
 	if (!dvsec)
 		dev_warn(dev, "%s GPF DVSEC not present\n",
@@ -1128,26 +1134,24 @@ static int update_gpf_port_dvsec(struct pci_dev *pdev, int dvsec, int phase)
 	return rc;
 }
 
-int cxl_gpf_port_setup(struct device *dport_dev, struct cxl_port *port)
+int cxl_gpf_port_setup(struct cxl_dport *dport)
 {
-	struct pci_dev *pdev;
-
-	if (!port)
+	if (!dport)
 		return -EINVAL;
 
-	if (!port->gpf_dvsec) {
+	if (!dport->gpf_dvsec) {
+		struct pci_dev *pdev;
 		int dvsec;
 
-		dvsec = cxl_gpf_get_dvsec(dport_dev, true);
+		dvsec = cxl_gpf_get_dvsec(dport->dport_dev);
 		if (!dvsec)
 			return -EINVAL;
 
-		port->gpf_dvsec = dvsec;
+		dport->gpf_dvsec = dvsec;
+		pdev = to_pci_dev(dport->dport_dev);
+		update_gpf_port_dvsec(pdev, dport->gpf_dvsec, 1);
+		update_gpf_port_dvsec(pdev, dport->gpf_dvsec, 2);
 	}
 
-	pdev = to_pci_dev(dport_dev);
-	update_gpf_port_dvsec(pdev, port->gpf_dvsec, 1);
-	update_gpf_port_dvsec(pdev, port->gpf_dvsec, 2);
-
 	return 0;
 }
diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
index 0fd6646c1a2e..726bd4a7de27 100644
--- a/drivers/cxl/core/port.c
+++ b/drivers/cxl/core/port.c
@@ -1678,7 +1678,7 @@ retry:
 			if (rc && rc != -EBUSY)
 				return rc;
 
-			cxl_gpf_port_setup(dport_dev, port);
+			cxl_gpf_port_setup(dport);
 
 			/* Any more ports to add between this one and the root? */
 			if (!dev_is_cxl_root_child(&port->dev))
diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c
index 117c2e94c761..5ca7b0eed568 100644
--- a/drivers/cxl/core/regs.c
+++ b/drivers/cxl/core/regs.c
@@ -581,7 +581,6 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
 	resource_size_t rcrb = ri->base;
 	void __iomem *addr;
 	u32 bar0, bar1;
-	u16 cmd;
 	u32 id;
 
 	if (which == CXL_RCRB_UPSTREAM)
@@ -603,7 +602,6 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
 	}
 
 	id = readl(addr + PCI_VENDOR_ID);
-	cmd = readw(addr + PCI_COMMAND);
 	bar0 = readl(addr + PCI_BASE_ADDRESS_0);
 	bar1 = readl(addr + PCI_BASE_ADDRESS_1);
 	iounmap(addr);
@@ -618,8 +616,6 @@ resource_size_t __rcrb_to_component(struct device *dev, struct cxl_rcrb_info *ri
 			dev_err(dev, "Failed to access Downstream Port RCRB\n");
 		return CXL_RESOURCE_NONE;
 	}
-	if (!(cmd & PCI_COMMAND_MEMORY))
-		return CXL_RESOURCE_NONE;
 	/* The RCRB is a Memory Window, and the MEM_TYPE_1M bit is obsolete */
 	if (bar0 & (PCI_BASE_ADDRESS_MEM_TYPE_1M | PCI_BASE_ADDRESS_SPACE_IO))
 		return CXL_RESOURCE_NONE;
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index be8a7dc77719..a9ab46eb0610 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -592,7 +592,6 @@ struct cxl_dax_region {
  * @cdat: Cached CDAT data
  * @cdat_available: Should a CDAT attribute be available in sysfs
  * @pci_latency: Upstream latency in picoseconds
- * @gpf_dvsec: Cached GPF port DVSEC
  */
 struct cxl_port {
 	struct device dev;
@@ -616,7 +615,6 @@ struct cxl_port {
 	} cdat;
 	bool cdat_available;
 	long pci_latency;
-	int gpf_dvsec;
 };
 
 /**
@@ -664,6 +662,7 @@ struct cxl_rcrb_info {
  * @regs: Dport parsed register blocks
  * @coord: access coordinates (bandwidth and latency performance attributes)
  * @link_latency: calculated PCIe downstream latency
+ * @gpf_dvsec: Cached GPF port DVSEC
  */
 struct cxl_dport {
 	struct device *dport_dev;
@@ -675,6 +674,7 @@ struct cxl_dport {
 	struct cxl_regs regs;
 	struct access_coordinate coord[ACCESS_COORDINATE_MAX];
 	long link_latency;
+	int gpf_dvsec;
 };
 
 /**
@@ -910,6 +910,6 @@ bool cxl_endpoint_decoder_reset_detected(struct cxl_port *port);
 #define __mock static
 #endif
 
-u16 cxl_gpf_get_dvsec(struct device *dev, bool is_port);
+u16 cxl_gpf_get_dvsec(struct device *dev);
 
 #endif /* __CXL_H__ */
diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
index 7b14a154463c..785aa2af5eaa 100644
--- a/drivers/cxl/pci.c
+++ b/drivers/cxl/pci.c
@@ -1018,7 +1018,7 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (rc)
 		return rc;
 
-	rc = devm_cxl_setup_fwctl(cxlmd);
+	rc = devm_cxl_setup_fwctl(&pdev->dev, cxlmd);
 	if (rc)
 		dev_dbg(&pdev->dev, "No CXL FWCTL setup\n");
 
diff --git a/drivers/cxl/pmem.c b/drivers/cxl/pmem.c
index d061fe3d2b86..e197883690ef 100644
--- a/drivers/cxl/pmem.c
+++ b/drivers/cxl/pmem.c
@@ -108,7 +108,7 @@ static void cxl_nvdimm_arm_dirty_shutdown_tracking(struct cxl_nvdimm *cxl_nvd)
 		return;
 	}
 
-	if (!cxl_gpf_get_dvsec(cxlds->dev, false))
+	if (!cxl_gpf_get_dvsec(cxlds->dev))
 		return;
 
 	if (cxl_get_dirty_count(mds, &count)) {
diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
index 5baa83b85515..0c48d41dd5eb 100644
--- a/drivers/dma-buf/dma-buf.c
+++ b/drivers/dma-buf/dma-buf.c
@@ -636,10 +636,6 @@ struct dma_buf *dma_buf_export(const struct dma_buf_export_info *exp_info)
 		    || !exp_info->ops->release))
 		return ERR_PTR(-EINVAL);
 
-	if (WARN_ON(exp_info->ops->cache_sgt_mapping &&
-		    (exp_info->ops->pin || exp_info->ops->unpin)))
-		return ERR_PTR(-EINVAL);
-
 	if (WARN_ON(!exp_info->ops->pin != !exp_info->ops->unpin))
 		return ERR_PTR(-EINVAL);
 
@@ -782,7 +778,7 @@ static void mangle_sg_table(struct sg_table *sg_table)
 
 	/* To catch abuse of the underlying struct page by importers mix
 	 * up the bits, but take care to preserve the low SG_ bits to
-	 * not corrupt the sgt. The mixing is undone in __unmap_dma_buf
+	 * not corrupt the sgt. The mixing is undone on unmap
 	 * before passing the sgt back to the exporter.
 	 */
 	for_each_sgtable_sg(sg_table, sg, i)
@@ -790,29 +786,19 @@ static void mangle_sg_table(struct sg_table *sg_table)
 #endif
 
 }
-static struct sg_table *__map_dma_buf(struct dma_buf_attachment *attach,
-				       enum dma_data_direction direction)
-{
-	struct sg_table *sg_table;
-	signed long ret;
-
-	sg_table = attach->dmabuf->ops->map_dma_buf(attach, direction);
-	if (IS_ERR_OR_NULL(sg_table))
-		return sg_table;
 
-	if (!dma_buf_attachment_is_dynamic(attach)) {
-		ret = dma_resv_wait_timeout(attach->dmabuf->resv,
-					    DMA_RESV_USAGE_KERNEL, true,
-					    MAX_SCHEDULE_TIMEOUT);
-		if (ret < 0) {
-			attach->dmabuf->ops->unmap_dma_buf(attach, sg_table,
-							   direction);
-			return ERR_PTR(ret);
-		}
-	}
+static inline bool
+dma_buf_attachment_is_dynamic(struct dma_buf_attachment *attach)
+{
+	return !!attach->importer_ops;
+}
 
-	mangle_sg_table(sg_table);
-	return sg_table;
+static bool
+dma_buf_pin_on_map(struct dma_buf_attachment *attach)
+{
+	return attach->dmabuf->ops->pin &&
+		(!dma_buf_attachment_is_dynamic(attach) ||
+		 !IS_ENABLED(CONFIG_DMABUF_MOVE_NOTIFY));
 }
 
 /**
@@ -935,48 +921,11 @@ dma_buf_dynamic_attach(struct dma_buf *dmabuf, struct device *dev,
 	list_add(&attach->node, &dmabuf->attachments);
 	dma_resv_unlock(dmabuf->resv);
 
-	/* When either the importer or the exporter can't handle dynamic
-	 * mappings we cache the mapping here to avoid issues with the
-	 * reservation object lock.
-	 */
-	if (dma_buf_attachment_is_dynamic(attach) !=
-	    dma_buf_is_dynamic(dmabuf)) {
-		struct sg_table *sgt;
-
-		dma_resv_lock(attach->dmabuf->resv, NULL);
-		if (dma_buf_is_dynamic(attach->dmabuf)) {
-			ret = dmabuf->ops->pin(attach);
-			if (ret)
-				goto err_unlock;
-		}
-
-		sgt = __map_dma_buf(attach, DMA_BIDIRECTIONAL);
-		if (!sgt)
-			sgt = ERR_PTR(-ENOMEM);
-		if (IS_ERR(sgt)) {
-			ret = PTR_ERR(sgt);
-			goto err_unpin;
-		}
-		dma_resv_unlock(attach->dmabuf->resv);
-		attach->sgt = sgt;
-		attach->dir = DMA_BIDIRECTIONAL;
-	}
-
 	return attach;
 
 err_attach:
 	kfree(attach);
 	return ERR_PTR(ret);
-
-err_unpin:
-	if (dma_buf_is_dynamic(attach->dmabuf))
-		dmabuf->ops->unpin(attach);
-
-err_unlock:
-	dma_resv_unlock(attach->dmabuf->resv);
-
-	dma_buf_detach(dmabuf, attach);
-	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL_NS_GPL(dma_buf_dynamic_attach, "DMA_BUF");
 
@@ -995,16 +944,6 @@ struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
 }
 EXPORT_SYMBOL_NS_GPL(dma_buf_attach, "DMA_BUF");
 
-static void __unmap_dma_buf(struct dma_buf_attachment *attach,
-			    struct sg_table *sg_table,
-			    enum dma_data_direction direction)
-{
-	/* uses XOR, hence this unmangles */
-	mangle_sg_table(sg_table);
-
-	attach->dmabuf->ops->unmap_dma_buf(attach, sg_table, direction);
-}
-
 /**
  * dma_buf_detach - Remove the given attachment from dmabuf's attachments list
  * @dmabuf:	[in]	buffer to detach from.
@@ -1020,16 +959,7 @@ void dma_buf_detach(struct dma_buf *dmabuf, struct dma_buf_attachment *attach)
 		return;
 
 	dma_resv_lock(dmabuf->resv, NULL);
-
-	if (attach->sgt) {
-
-		__unmap_dma_buf(attach, attach->sgt, attach->dir);
-
-		if (dma_buf_is_dynamic(attach->dmabuf))
-			dmabuf->ops->unpin(attach);
-	}
 	list_del(&attach->node);
-
 	dma_resv_unlock(dmabuf->resv);
 
 	if (dmabuf->ops->detach)
@@ -1058,7 +988,7 @@ int dma_buf_pin(struct dma_buf_attachment *attach)
 	struct dma_buf *dmabuf = attach->dmabuf;
 	int ret = 0;
 
-	WARN_ON(!dma_buf_attachment_is_dynamic(attach));
+	WARN_ON(!attach->importer_ops);
 
 	dma_resv_assert_held(dmabuf->resv);
 
@@ -1081,7 +1011,7 @@ void dma_buf_unpin(struct dma_buf_attachment *attach)
 {
 	struct dma_buf *dmabuf = attach->dmabuf;
 
-	WARN_ON(!dma_buf_attachment_is_dynamic(attach));
+	WARN_ON(!attach->importer_ops);
 
 	dma_resv_assert_held(dmabuf->resv);
 
@@ -1115,7 +1045,7 @@ struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach,
 					enum dma_data_direction direction)
 {
 	struct sg_table *sg_table;
-	int r;
+	signed long ret;
 
 	might_sleep();
 
@@ -1124,41 +1054,37 @@ struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach,
 
 	dma_resv_assert_held(attach->dmabuf->resv);
 
-	if (attach->sgt) {
+	if (dma_buf_pin_on_map(attach)) {
+		ret = attach->dmabuf->ops->pin(attach);
 		/*
-		 * Two mappings with different directions for the same
-		 * attachment are not allowed.
+		 * Catch exporters making buffers inaccessible even when
+		 * attachments preventing that exist.
 		 */
-		if (attach->dir != direction &&
-		    attach->dir != DMA_BIDIRECTIONAL)
-			return ERR_PTR(-EBUSY);
-
-		return attach->sgt;
-	}
-
-	if (dma_buf_is_dynamic(attach->dmabuf)) {
-		if (!IS_ENABLED(CONFIG_DMABUF_MOVE_NOTIFY)) {
-			r = attach->dmabuf->ops->pin(attach);
-			if (r)
-				return ERR_PTR(r);
-		}
+		WARN_ON_ONCE(ret == EBUSY);
+		if (ret)
+			return ERR_PTR(ret);
 	}
 
-	sg_table = __map_dma_buf(attach, direction);
+	sg_table = attach->dmabuf->ops->map_dma_buf(attach, direction);
 	if (!sg_table)
 		sg_table = ERR_PTR(-ENOMEM);
+	if (IS_ERR(sg_table))
+		goto error_unpin;
 
-	if (IS_ERR(sg_table) && dma_buf_is_dynamic(attach->dmabuf) &&
-	     !IS_ENABLED(CONFIG_DMABUF_MOVE_NOTIFY))
-		attach->dmabuf->ops->unpin(attach);
-
-	if (!IS_ERR(sg_table) && attach->dmabuf->ops->cache_sgt_mapping) {
-		attach->sgt = sg_table;
-		attach->dir = direction;
+	/*
+	 * Importers with static attachments don't wait for fences.
+	 */
+	if (!dma_buf_attachment_is_dynamic(attach)) {
+		ret = dma_resv_wait_timeout(attach->dmabuf->resv,
+					    DMA_RESV_USAGE_KERNEL, true,
+					    MAX_SCHEDULE_TIMEOUT);
+		if (ret < 0)
+			goto error_unmap;
 	}
+	mangle_sg_table(sg_table);
 
 #ifdef CONFIG_DMA_API_DEBUG
-	if (!IS_ERR(sg_table)) {
+	{
 		struct scatterlist *sg;
 		u64 addr;
 		int len;
@@ -1175,6 +1101,16 @@ struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach,
 	}
 #endif /* CONFIG_DMA_API_DEBUG */
 	return sg_table;
+
+error_unmap:
+	attach->dmabuf->ops->unmap_dma_buf(attach, sg_table, direction);
+	sg_table = ERR_PTR(ret);
+
+error_unpin:
+	if (dma_buf_pin_on_map(attach))
+		attach->dmabuf->ops->unpin(attach);
+
+	return sg_table;
 }
 EXPORT_SYMBOL_NS_GPL(dma_buf_map_attachment, "DMA_BUF");
 
@@ -1227,14 +1163,11 @@ void dma_buf_unmap_attachment(struct dma_buf_attachment *attach,
 
 	dma_resv_assert_held(attach->dmabuf->resv);
 
-	if (attach->sgt == sg_table)
-		return;
-
-	__unmap_dma_buf(attach, sg_table, direction);
+	mangle_sg_table(sg_table);
+	attach->dmabuf->ops->unmap_dma_buf(attach, sg_table, direction);
 
-	if (dma_buf_is_dynamic(attach->dmabuf) &&
-	    !IS_ENABLED(CONFIG_DMABUF_MOVE_NOTIFY))
-		dma_buf_unpin(attach);
+	if (dma_buf_pin_on_map(attach))
+		attach->dmabuf->ops->unpin(attach);
 }
 EXPORT_SYMBOL_NS_GPL(dma_buf_unmap_attachment, "DMA_BUF");
 
diff --git a/drivers/dma-buf/dma-fence-unwrap.c b/drivers/dma-buf/dma-fence-unwrap.c
index 2a059ac0ed27..a495d8a6c2e3 100644
--- a/drivers/dma-buf/dma-fence-unwrap.c
+++ b/drivers/dma-buf/dma-fence-unwrap.c
@@ -79,6 +79,41 @@ static int fence_cmp(const void *_a, const void *_b)
 	return 0;
 }
 
+/**
+ * dma_fence_dedup_array - Sort and deduplicate an array of dma_fence pointers
+ * @fences:     Array of dma_fence pointers to be deduplicated
+ * @num_fences: Number of entries in the @fences array
+ *
+ * Sorts the input array by context, then removes duplicate
+ * fences with the same context, keeping only the most recent one.
+ *
+ * The array is modified in-place and unreferenced duplicate fences are released
+ * via dma_fence_put(). The function returns the new number of fences after
+ * deduplication.
+ *
+ * Return: Number of unique fences remaining in the array.
+ */
+int dma_fence_dedup_array(struct dma_fence **fences, int num_fences)
+{
+	int i, j;
+
+	sort(fences, num_fences, sizeof(*fences), fence_cmp, NULL);
+
+	/*
+	 * Only keep the most recent fence for each context.
+	 */
+	j = 0;
+	for (i = 1; i < num_fences; i++) {
+		if (fences[i]->context == fences[j]->context)
+			dma_fence_put(fences[i]);
+		else
+			fences[++j] = fences[i];
+	}
+
+	return ++j;
+}
+EXPORT_SYMBOL_GPL(dma_fence_dedup_array);
+
 /* Implementation for the dma_fence_merge() marco, don't use directly */
 struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences,
 					   struct dma_fence **fences,
@@ -87,7 +122,7 @@ struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences,
 	struct dma_fence *tmp, *unsignaled = NULL, **array;
 	struct dma_fence_array *result;
 	ktime_t timestamp;
-	int i, j, count;
+	int i, count;
 
 	count = 0;
 	timestamp = ns_to_ktime(0);
@@ -141,19 +176,7 @@ struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences,
 	if (count == 0 || count == 1)
 		goto return_fastpath;
 
-	sort(array, count, sizeof(*array), fence_cmp, NULL);
-
-	/*
-	 * Only keep the most recent fence for each context.
-	 */
-	j = 0;
-	for (i = 1; i < count; i++) {
-		if (array[i]->context == array[j]->context)
-			dma_fence_put(array[i]);
-		else
-			array[++j] = array[i];
-	}
-	count = ++j;
+	count = dma_fence_dedup_array(array, count);
 
 	if (count > 1) {
 		result = dma_fence_array_create(count, array,
diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c
index 26d5dc89ea16..82b1b714300d 100644
--- a/drivers/dma-buf/heaps/system_heap.c
+++ b/drivers/dma-buf/heaps/system_heap.c
@@ -21,8 +21,6 @@
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
 
-static struct dma_heap *sys_heap;
-
 struct system_heap_buffer {
 	struct dma_heap *heap;
 	struct list_head attachments;
@@ -424,6 +422,7 @@ static const struct dma_heap_ops system_heap_ops = {
 static int __init system_heap_create(void)
 {
 	struct dma_heap_export_info exp_info;
+	struct dma_heap *sys_heap;
 
 	exp_info.name = "system";
 	exp_info.ops = &system_heap_ops;
diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c
index f5905d67dedb..4f27ee93a00c 100644
--- a/drivers/dma-buf/sw_sync.c
+++ b/drivers/dma-buf/sw_sync.c
@@ -173,20 +173,6 @@ static bool timeline_fence_signaled(struct dma_fence *fence)
 	return !__dma_fence_is_later(fence->seqno, parent->value, fence->ops);
 }
 
-static void timeline_fence_value_str(struct dma_fence *fence,
-				    char *str, int size)
-{
-	snprintf(str, size, "%lld", fence->seqno);
-}
-
-static void timeline_fence_timeline_value_str(struct dma_fence *fence,
-					     char *str, int size)
-{
-	struct sync_timeline *parent = dma_fence_parent(fence);
-
-	snprintf(str, size, "%d", parent->value);
-}
-
 static void timeline_fence_set_deadline(struct dma_fence *fence, ktime_t deadline)
 {
 	struct sync_pt *pt = dma_fence_to_sync_pt(fence);
@@ -208,8 +194,6 @@ static const struct dma_fence_ops timeline_fence_ops = {
 	.get_timeline_name = timeline_fence_get_timeline_name,
 	.signaled = timeline_fence_signaled,
 	.release = timeline_fence_release,
-	.fence_value_str = timeline_fence_value_str,
-	.timeline_value_str = timeline_fence_timeline_value_str,
 	.set_deadline = timeline_fence_set_deadline,
 };
 
@@ -438,15 +422,17 @@ static int sw_sync_ioctl_get_deadline(struct sync_timeline *obj, unsigned long a
 		return -EINVAL;
 
 	pt = dma_fence_to_sync_pt(fence);
-	if (!pt)
-		return -EINVAL;
+	if (!pt) {
+		ret = -EINVAL;
+		goto put_fence;
+	}
 
 	spin_lock_irqsave(fence->lock, flags);
-	if (test_bit(SW_SYNC_HAS_DEADLINE_BIT, &fence->flags)) {
-		data.deadline_ns = ktime_to_ns(pt->deadline);
-	} else {
+	if (!test_bit(SW_SYNC_HAS_DEADLINE_BIT, &fence->flags)) {
 		ret = -ENOENT;
+		goto unlock;
 	}
+	data.deadline_ns = ktime_to_ns(pt->deadline);
 	spin_unlock_irqrestore(fence->lock, flags);
 
 	dma_fence_put(fence);
@@ -458,6 +444,13 @@ static int sw_sync_ioctl_get_deadline(struct sync_timeline *obj, unsigned long a
 		return -EFAULT;
 
 	return 0;
+
+unlock:
+	spin_unlock_irqrestore(fence->lock, flags);
+put_fence:
+	dma_fence_put(fence);
+
+	return ret;
 }
 
 static long sw_sync_ioctl(struct file *file, unsigned int cmd,
diff --git a/drivers/dma-buf/sync_debug.c b/drivers/dma-buf/sync_debug.c
index 237bce21d1e7..67cd69551e42 100644
--- a/drivers/dma-buf/sync_debug.c
+++ b/drivers/dma-buf/sync_debug.c
@@ -12,8 +12,6 @@ static struct dentry *dbgfs;
 
 static LIST_HEAD(sync_timeline_list_head);
 static DEFINE_SPINLOCK(sync_timeline_list_lock);
-static LIST_HEAD(sync_file_list_head);
-static DEFINE_SPINLOCK(sync_file_list_lock);
 
 void sync_timeline_debug_add(struct sync_timeline *obj)
 {
@@ -33,24 +31,6 @@ void sync_timeline_debug_remove(struct sync_timeline *obj)
 	spin_unlock_irqrestore(&sync_timeline_list_lock, flags);
 }
 
-void sync_file_debug_add(struct sync_file *sync_file)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&sync_file_list_lock, flags);
-	list_add_tail(&sync_file->sync_file_list, &sync_file_list_head);
-	spin_unlock_irqrestore(&sync_file_list_lock, flags);
-}
-
-void sync_file_debug_remove(struct sync_file *sync_file)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&sync_file_list_lock, flags);
-	list_del(&sync_file->sync_file_list);
-	spin_unlock_irqrestore(&sync_file_list_lock, flags);
-}
-
 static const char *sync_status_str(int status)
 {
 	if (status < 0)
@@ -82,25 +62,8 @@ static void sync_print_fence(struct seq_file *s,
 		seq_printf(s, "@%lld.%09ld", (s64)ts64.tv_sec, ts64.tv_nsec);
 	}
 
-	if (fence->ops->timeline_value_str &&
-		fence->ops->fence_value_str) {
-		char value[64];
-		bool success;
-
-		fence->ops->fence_value_str(fence, value, sizeof(value));
-		success = strlen(value);
-
-		if (success) {
-			seq_printf(s, ": %s", value);
-
-			fence->ops->timeline_value_str(fence, value,
-						       sizeof(value));
-
-			if (strlen(value))
-				seq_printf(s, " / %s", value);
-		}
-	}
-
+	seq_printf(s, ": %lld", fence->seqno);
+	seq_printf(s, " / %d", parent->value);
 	seq_putc(s, '\n');
 }
 
@@ -118,26 +81,6 @@ static void sync_print_obj(struct seq_file *s, struct sync_timeline *obj)
 	spin_unlock(&obj->lock);
 }
 
-static void sync_print_sync_file(struct seq_file *s,
-				  struct sync_file *sync_file)
-{
-	char buf[128];
-	int i;
-
-	seq_printf(s, "[%p] %s: %s\n", sync_file,
-		   sync_file_get_name(sync_file, buf, sizeof(buf)),
-		   sync_status_str(dma_fence_get_status(sync_file->fence)));
-
-	if (dma_fence_is_array(sync_file->fence)) {
-		struct dma_fence_array *array = to_dma_fence_array(sync_file->fence);
-
-		for (i = 0; i < array->num_fences; ++i)
-			sync_print_fence(s, array->fences[i], true);
-	} else {
-		sync_print_fence(s, sync_file->fence, true);
-	}
-}
-
 static int sync_info_debugfs_show(struct seq_file *s, void *unused)
 {
 	struct list_head *pos;
@@ -157,15 +100,6 @@ static int sync_info_debugfs_show(struct seq_file *s, void *unused)
 
 	seq_puts(s, "fences:\n--------------\n");
 
-	spin_lock_irq(&sync_file_list_lock);
-	list_for_each(pos, &sync_file_list_head) {
-		struct sync_file *sync_file =
-			container_of(pos, struct sync_file, sync_file_list);
-
-		sync_print_sync_file(s, sync_file);
-		seq_putc(s, '\n');
-	}
-	spin_unlock_irq(&sync_file_list_lock);
 	return 0;
 }
 
diff --git a/drivers/dma-buf/sync_debug.h b/drivers/dma-buf/sync_debug.h
index a1bdd62efccd..02af347293d0 100644
--- a/drivers/dma-buf/sync_debug.h
+++ b/drivers/dma-buf/sync_debug.h
@@ -68,7 +68,5 @@ extern const struct file_operations sw_sync_debugfs_fops;
 
 void sync_timeline_debug_add(struct sync_timeline *obj);
 void sync_timeline_debug_remove(struct sync_timeline *obj);
-void sync_file_debug_add(struct sync_file *fence);
-void sync_file_debug_remove(struct sync_file *fence);
 
 #endif /* _LINUX_SYNC_H */
diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c
index cc7398cc17d6..7eee3eb47a8e 100644
--- a/drivers/dma-buf/udmabuf.c
+++ b/drivers/dma-buf/udmabuf.c
@@ -285,7 +285,6 @@ static int end_cpu_udmabuf(struct dma_buf *buf,
 }
 
 static const struct dma_buf_ops udmabuf_ops = {
-	.cache_sgt_mapping = true,
 	.map_dma_buf	   = map_udmabuf,
 	.unmap_dma_buf	   = unmap_udmabuf,
 	.release	   = release_udmabuf,
@@ -393,7 +392,7 @@ static long udmabuf_create(struct miscdevice *device,
 	if (!ubuf)
 		return -ENOMEM;
 
-	pglimit = (size_limit_mb * 1024 * 1024) >> PAGE_SHIFT;
+	pglimit = ((u64)size_limit_mb * 1024 * 1024) >> PAGE_SHIFT;
 	for (i = 0; i < head->count; i++) {
 		pgoff_t subpgcnt;
 
diff --git a/drivers/edac/altera_edac.c b/drivers/edac/altera_edac.c
index 3e971f902363..dcd7008fe06b 100644
--- a/drivers/edac/altera_edac.c
+++ b/drivers/edac/altera_edac.c
@@ -99,7 +99,7 @@ static irqreturn_t altr_sdram_mc_err_handler(int irq, void *dev_id)
 	if (status & priv->ecc_stat_ce_mask) {
 		regmap_read(drvdata->mc_vbase, priv->ecc_saddr_offset,
 			    &err_addr);
-		if (priv->ecc_uecnt_offset)
+		if (priv->ecc_cecnt_offset)
 			regmap_read(drvdata->mc_vbase,  priv->ecc_cecnt_offset,
 				    &err_count);
 		edac_mc_handle_error(HW_EVENT_ERR_CORRECTED, mci, err_count,
@@ -1005,9 +1005,6 @@ altr_init_a10_ecc_block(struct device_node *np, u32 irq_mask,
 		}
 	}
 
-	/* Interrupt mode set to every SBERR */
-	regmap_write(ecc_mgr_map, ALTR_A10_ECC_INTMODE_OFST,
-		     ALTR_A10_ECC_INTMODE);
 	/* Enable ECC */
 	ecc_set_bits(ecc_ctrl_en_mask, (ecc_block_base +
 					ALTR_A10_ECC_CTRL_OFST));
@@ -2127,6 +2124,10 @@ static int altr_edac_a10_probe(struct platform_device *pdev)
 		return PTR_ERR(edac->ecc_mgr_map);
 	}
 
+	/* Set irq mask for DDR SBE to avoid any pending irq before registration */
+	regmap_write(edac->ecc_mgr_map, A10_SYSMGR_ECC_INTMASK_SET_OFST,
+		     (A10_SYSMGR_ECC_INTMASK_SDMMCB | A10_SYSMGR_ECC_INTMASK_DDR0));
+
 	edac->irq_chip.name = pdev->dev.of_node->name;
 	edac->irq_chip.irq_mask = a10_eccmgr_irq_mask;
 	edac->irq_chip.irq_unmask = a10_eccmgr_irq_unmask;
diff --git a/drivers/edac/altera_edac.h b/drivers/edac/altera_edac.h
index 3727e72c8c2e..7248d24c4908 100644
--- a/drivers/edac/altera_edac.h
+++ b/drivers/edac/altera_edac.h
@@ -249,6 +249,8 @@ struct altr_sdram_mc_data {
 #define A10_SYSMGR_ECC_INTMASK_SET_OFST   0x94
 #define A10_SYSMGR_ECC_INTMASK_CLR_OFST   0x98
 #define A10_SYSMGR_ECC_INTMASK_OCRAM      BIT(1)
+#define A10_SYSMGR_ECC_INTMASK_SDMMCB     BIT(16)
+#define A10_SYSMGR_ECC_INTMASK_DDR0       BIT(17)
 
 #define A10_SYSMGR_ECC_INTSTAT_SERR_OFST  0x9C
 #define A10_SYSMGR_ECC_INTSTAT_DERR_OFST  0xA0
diff --git a/drivers/firmware/cirrus/Kconfig b/drivers/firmware/cirrus/Kconfig
index 0a883091259a..e3c2e38b746d 100644
--- a/drivers/firmware/cirrus/Kconfig
+++ b/drivers/firmware/cirrus/Kconfig
@@ -6,14 +6,11 @@ config FW_CS_DSP
 
 config FW_CS_DSP_KUNIT_TEST_UTILS
 	tristate
-	depends on KUNIT && REGMAP
-	select FW_CS_DSP
 
 config FW_CS_DSP_KUNIT_TEST
 	tristate "KUnit tests for Cirrus Logic cs_dsp" if !KUNIT_ALL_TESTS
-	depends on KUNIT && REGMAP
+	depends on KUNIT && REGMAP && FW_CS_DSP
 	default KUNIT_ALL_TESTS
-	select FW_CS_DSP
 	select FW_CS_DSP_KUNIT_TEST_UTILS
 	help
 	  This builds KUnit tests for cs_dsp.
diff --git a/drivers/firmware/cirrus/test/cs_dsp_mock_mem_maps.c b/drivers/firmware/cirrus/test/cs_dsp_mock_mem_maps.c
index 161272e47bda..73412bcef50c 100644
--- a/drivers/firmware/cirrus/test/cs_dsp_mock_mem_maps.c
+++ b/drivers/firmware/cirrus/test/cs_dsp_mock_mem_maps.c
@@ -462,36 +462,6 @@ unsigned int cs_dsp_mock_xm_header_get_alg_base_in_words(struct cs_dsp_test *pri
 EXPORT_SYMBOL_NS_GPL(cs_dsp_mock_xm_header_get_alg_base_in_words, "FW_CS_DSP_KUNIT_TEST_UTILS");
 
 /**
- * cs_dsp_mock_xm_header_get_fw_version_from_regmap() - Firmware version.
- *
- * @priv:	Pointer to struct cs_dsp_test.
- *
- * Return: Firmware version word value.
- */
-unsigned int cs_dsp_mock_xm_header_get_fw_version_from_regmap(struct cs_dsp_test *priv)
-{
-	unsigned int xm = cs_dsp_mock_base_addr_for_mem(priv, WMFW_ADSP2_XM);
-	union {
-		struct wmfw_id_hdr adsp2;
-		struct wmfw_v3_id_hdr halo;
-	} hdr;
-
-	switch (priv->dsp->type) {
-	case WMFW_ADSP2:
-		regmap_raw_read(priv->dsp->regmap, xm, &hdr.adsp2, sizeof(hdr.adsp2));
-		return be32_to_cpu(hdr.adsp2.ver);
-	case WMFW_HALO:
-		regmap_raw_read(priv->dsp->regmap, xm, &hdr.halo, sizeof(hdr.halo));
-		return be32_to_cpu(hdr.halo.ver);
-	default:
-		KUNIT_FAIL(priv->test, NULL);
-		return 0;
-	}
-}
-EXPORT_SYMBOL_NS_GPL(cs_dsp_mock_xm_header_get_fw_version_from_regmap,
-		     "FW_CS_DSP_KUNIT_TEST_UTILS");
-
-/**
  * cs_dsp_mock_xm_header_get_fw_version() - Firmware version.
  *
  * @header:	Pointer to struct cs_dsp_mock_xm_header.
diff --git a/drivers/firmware/cirrus/test/cs_dsp_test_bin.c b/drivers/firmware/cirrus/test/cs_dsp_test_bin.c
index 1e161bbc5b4a..163b7faecff4 100644
--- a/drivers/firmware/cirrus/test/cs_dsp_test_bin.c
+++ b/drivers/firmware/cirrus/test/cs_dsp_test_bin.c
@@ -2198,7 +2198,7 @@ static int cs_dsp_bin_test_common_init(struct kunit *test, struct cs_dsp *dsp)
 
 	priv->local->bin_builder =
 		cs_dsp_mock_bin_init(priv, 1,
-				     cs_dsp_mock_xm_header_get_fw_version_from_regmap(priv));
+				     cs_dsp_mock_xm_header_get_fw_version(xm_hdr));
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->local->bin_builder);
 
 	/* We must provide a dummy wmfw to load */
diff --git a/drivers/firmware/cirrus/test/cs_dsp_test_bin_error.c b/drivers/firmware/cirrus/test/cs_dsp_test_bin_error.c
index 8748874f0552..a7ec956d2724 100644
--- a/drivers/firmware/cirrus/test/cs_dsp_test_bin_error.c
+++ b/drivers/firmware/cirrus/test/cs_dsp_test_bin_error.c
@@ -451,7 +451,7 @@ static int cs_dsp_bin_err_test_common_init(struct kunit *test, struct cs_dsp *ds
 
 	local->bin_builder =
 		cs_dsp_mock_bin_init(priv, 1,
-				     cs_dsp_mock_xm_header_get_fw_version_from_regmap(priv));
+				     cs_dsp_mock_xm_header_get_fw_version(local->xm_header));
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, local->bin_builder);
 
 	/* Init cs_dsp */
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index 7309394b8fc9..e57bff702b5f 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -558,6 +558,7 @@ int __efi_mem_desc_lookup(u64 phys_addr, efi_memory_desc_t *out_md)
 
 extern int efi_mem_desc_lookup(u64 phys_addr, efi_memory_desc_t *out_md)
 	__weak __alias(__efi_mem_desc_lookup);
+EXPORT_SYMBOL_GPL(efi_mem_desc_lookup);
 
 /*
  * Calculate the highest address of an efi memory descriptor.
diff --git a/drivers/firmware/smccc/kvm_guest.c b/drivers/firmware/smccc/kvm_guest.c
index 5767aed25cdc..a123c05cbc9e 100644
--- a/drivers/firmware/smccc/kvm_guest.c
+++ b/drivers/firmware/smccc/kvm_guest.c
@@ -95,7 +95,7 @@ void  __init kvm_arm_target_impl_cpu_init(void)
 
 	for (i = 0; i < max_cpus; i++) {
 		arm_smccc_1_1_invoke(ARM_SMCCC_VENDOR_HYP_KVM_DISCOVER_IMPL_CPUS_FUNC_ID,
-				     i, &res);
+				     i, 0, 0, &res);
 		if (res.a0 != SMCCC_RET_SUCCESS) {
 			pr_warn("Discovering target implementation CPUs failed\n");
 			goto mem_free;
@@ -103,7 +103,7 @@ void  __init kvm_arm_target_impl_cpu_init(void)
 		target[i].midr = res.a1;
 		target[i].revidr = res.a2;
 		target[i].aidr = res.a3;
-	};
+	}
 
 	if (!cpu_errata_set_target_impl(max_cpus, target)) {
 		pr_warn("Failed to set target implementation CPUs\n");
diff --git a/drivers/firmware/stratix10-svc.c b/drivers/firmware/stratix10-svc.c
index 3c52cb73237a..e3f990d888d7 100644
--- a/drivers/firmware/stratix10-svc.c
+++ b/drivers/firmware/stratix10-svc.c
@@ -1224,22 +1224,28 @@ static int stratix10_svc_drv_probe(struct platform_device *pdev)
 	if (!svc->intel_svc_fcs) {
 		dev_err(dev, "failed to allocate %s device\n", INTEL_FCS);
 		ret = -ENOMEM;
-		goto err_unregister_dev;
+		goto err_unregister_rsu_dev;
 	}
 
 	ret = platform_device_add(svc->intel_svc_fcs);
 	if (ret) {
 		platform_device_put(svc->intel_svc_fcs);
-		goto err_unregister_dev;
+		goto err_unregister_rsu_dev;
 	}
 
+	ret = of_platform_default_populate(dev_of_node(dev), NULL, dev);
+	if (ret)
+		goto err_unregister_fcs_dev;
+
 	dev_set_drvdata(dev, svc);
 
 	pr_info("Intel Service Layer Driver Initialized\n");
 
 	return 0;
 
-err_unregister_dev:
+err_unregister_fcs_dev:
+	platform_device_unregister(svc->intel_svc_fcs);
+err_unregister_rsu_dev:
 	platform_device_unregister(svc->stratix10_svc_rsu);
 err_free_kfifo:
 	kfifo_free(&controller->svc_fifo);
@@ -1253,6 +1259,8 @@ static void stratix10_svc_drv_remove(struct platform_device *pdev)
 	struct stratix10_svc *svc = dev_get_drvdata(&pdev->dev);
 	struct stratix10_svc_controller *ctrl = platform_get_drvdata(pdev);
 
+	of_platform_depopulate(ctrl->dev);
+
 	platform_device_unregister(svc->intel_svc_fcs);
 	platform_device_unregister(svc->stratix10_svc_rsu);
 
diff --git a/drivers/firmware/sysfb_simplefb.c b/drivers/firmware/sysfb_simplefb.c
index 75a186bf8f8e..592d8a644619 100644
--- a/drivers/firmware/sysfb_simplefb.c
+++ b/drivers/firmware/sysfb_simplefb.c
@@ -35,36 +35,7 @@ __init bool sysfb_parse_mode(const struct screen_info *si,
 	if (type != VIDEO_TYPE_VLFB && type != VIDEO_TYPE_EFI)
 		return false;
 
-	/*
-	 * The meaning of depth and bpp for direct-color formats is
-	 * inconsistent:
-	 *
-	 *  - DRM format info specifies depth as the number of color
-	 *    bits; including alpha, but not including filler bits.
-	 *  - Linux' EFI platform code computes lfb_depth from the
-	 *    individual color channels, including the reserved bits.
-	 *  - VBE 1.1 defines lfb_depth for XRGB1555 as 16, but later
-	 *    versions use 15.
-	 *  - On the kernel command line, 'bpp' of 32 is usually
-	 *    XRGB8888 including the filler bits, but 15 is XRGB1555
-	 *    not including the filler bit.
-	 *
-	 * It's not easily possible to fix this in struct screen_info,
-	 * as this could break UAPI. The best solution is to compute
-	 * bits_per_pixel from the color bits, reserved bits and
-	 * reported lfb_depth, whichever is highest.  In the loop below,
-	 * ignore simplefb formats with alpha bits, as EFI and VESA
-	 * don't specify alpha channels.
-	 */
-	if (si->lfb_depth > 8) {
-		bits_per_pixel = max(max3(si->red_size + si->red_pos,
-					  si->green_size + si->green_pos,
-					  si->blue_size + si->blue_pos),
-				     si->rsvd_size + si->rsvd_pos);
-		bits_per_pixel = max_t(u32, bits_per_pixel, si->lfb_depth);
-	} else {
-		bits_per_pixel = si->lfb_depth;
-	}
+	bits_per_pixel = __screen_info_lfb_bits_per_pixel(si);
 
 	for (i = 0; i < ARRAY_SIZE(formats); ++i) {
 		const struct simplefb_format *f = &formats[i];
diff --git a/drivers/fpga/tests/fpga-bridge-test.c b/drivers/fpga/tests/fpga-bridge-test.c
index b9ab29809e96..124ba40e32b1 100644
--- a/drivers/fpga/tests/fpga-bridge-test.c
+++ b/drivers/fpga/tests/fpga-bridge-test.c
@@ -170,4 +170,5 @@ static struct kunit_suite fpga_bridge_suite = {
 
 kunit_test_suite(fpga_bridge_suite);
 
+MODULE_DESCRIPTION("KUnit test for the FPGA Bridge");
 MODULE_LICENSE("GPL");
diff --git a/drivers/fpga/tests/fpga-mgr-test.c b/drivers/fpga/tests/fpga-mgr-test.c
index 9cb37aefbac4..8748babb0504 100644
--- a/drivers/fpga/tests/fpga-mgr-test.c
+++ b/drivers/fpga/tests/fpga-mgr-test.c
@@ -330,4 +330,5 @@ static struct kunit_suite fpga_mgr_suite = {
 
 kunit_test_suite(fpga_mgr_suite);
 
+MODULE_DESCRIPTION("KUnit test for the FPGA Manager");
 MODULE_LICENSE("GPL");
diff --git a/drivers/fpga/tests/fpga-region-test.c b/drivers/fpga/tests/fpga-region-test.c
index 6a108cafded8..020ceac48509 100644
--- a/drivers/fpga/tests/fpga-region-test.c
+++ b/drivers/fpga/tests/fpga-region-test.c
@@ -214,4 +214,5 @@ static struct kunit_suite fpga_region_suite = {
 
 kunit_test_suite(fpga_region_suite);
 
+MODULE_DESCRIPTION("KUnit test for the FPGA Region");
 MODULE_LICENSE("GPL");
diff --git a/drivers/fwctl/main.c b/drivers/fwctl/main.c
index cb1ac9c40239..bc6378506296 100644
--- a/drivers/fwctl/main.c
+++ b/drivers/fwctl/main.c
@@ -105,7 +105,7 @@ static int fwctl_cmd_rpc(struct fwctl_ucmd *ucmd)
 		if (!test_and_set_bit(0, &fwctl_tainted)) {
 			dev_warn(
 				&fwctl->dev,
-				"%s(%d): has requested full access to the physical device device",
+				"%s(%d): has requested full access to the physical device",
 				current->comm, task_pid_nr(current));
 			add_taint(TAINT_FWCTL, LOCKDEP_STILL_OK);
 		}
diff --git a/drivers/fwctl/pds/main.c b/drivers/fwctl/pds/main.c
index 284c4165fdd4..9b9d1f6b5556 100644
--- a/drivers/fwctl/pds/main.c
+++ b/drivers/fwctl/pds/main.c
@@ -105,12 +105,14 @@ static int pdsfc_identify(struct pdsfc_dev *pdsfc)
 static void pdsfc_free_endpoints(struct pdsfc_dev *pdsfc)
 {
 	struct device *dev = &pdsfc->fwctl.dev;
+	u32 num_endpoints;
 	int i;
 
 	if (!pdsfc->endpoints)
 		return;
 
-	for (i = 0; pdsfc->endpoint_info && i < pdsfc->endpoints->num_entries; i++)
+	num_endpoints = le32_to_cpu(pdsfc->endpoints->num_entries);
+	for (i = 0; pdsfc->endpoint_info && i < num_endpoints; i++)
 		mutex_destroy(&pdsfc->endpoint_info[i].lock);
 	vfree(pdsfc->endpoint_info);
 	pdsfc->endpoint_info = NULL;
@@ -199,7 +201,7 @@ static int pdsfc_init_endpoints(struct pdsfc_dev *pdsfc)
 	ep_entry = (struct pds_fwctl_query_data_endpoint *)pdsfc->endpoints->entries;
 	for (i = 0; i < num_endpoints; i++) {
 		mutex_init(&pdsfc->endpoint_info[i].lock);
-		pdsfc->endpoint_info[i].endpoint = ep_entry[i].id;
+		pdsfc->endpoint_info[i].endpoint = le32_to_cpu(ep_entry[i].id);
 	}
 
 	return 0;
@@ -214,6 +216,7 @@ static struct pds_fwctl_query_data *pdsfc_get_operations(struct pdsfc_dev *pdsfc
 	struct pds_fwctl_query_data *data;
 	union pds_core_adminq_cmd cmd;
 	dma_addr_t data_pa;
+	u32 num_entries;
 	int err;
 	int i;
 
@@ -246,8 +249,9 @@ static struct pds_fwctl_query_data *pdsfc_get_operations(struct pdsfc_dev *pdsfc
 	*pa = data_pa;
 
 	entries = (struct pds_fwctl_query_data_operation *)data->entries;
-	dev_dbg(dev, "num_entries %d\n", data->num_entries);
-	for (i = 0; i < data->num_entries; i++) {
+	num_entries = le32_to_cpu(data->num_entries);
+	dev_dbg(dev, "num_entries %d\n", num_entries);
+	for (i = 0; i < num_entries; i++) {
 
 		/* Translate FW command attribute to fwctl scope */
 		switch (entries[i].scope) {
@@ -267,7 +271,7 @@ static struct pds_fwctl_query_data *pdsfc_get_operations(struct pdsfc_dev *pdsfc
 			break;
 		}
 		dev_dbg(dev, "endpoint %d operation: id %x scope %d\n",
-			ep, entries[i].id, entries[i].scope);
+			ep, le32_to_cpu(entries[i].id), entries[i].scope);
 	}
 
 	return data;
@@ -280,24 +284,26 @@ static int pdsfc_validate_rpc(struct pdsfc_dev *pdsfc,
 	struct pds_fwctl_query_data_operation *op_entry;
 	struct pdsfc_rpc_endpoint_info *ep_info = NULL;
 	struct device *dev = &pdsfc->fwctl.dev;
+	u32 num_entries;
 	int i;
 
 	/* validate rpc in_len & out_len based
 	 * on ident.max_req_sz & max_resp_sz
 	 */
-	if (rpc->in.len > pdsfc->ident.max_req_sz) {
+	if (rpc->in.len > le32_to_cpu(pdsfc->ident.max_req_sz)) {
 		dev_dbg(dev, "Invalid request size %u, max %u\n",
-			rpc->in.len, pdsfc->ident.max_req_sz);
+			rpc->in.len, le32_to_cpu(pdsfc->ident.max_req_sz));
 		return -EINVAL;
 	}
 
-	if (rpc->out.len > pdsfc->ident.max_resp_sz) {
+	if (rpc->out.len > le32_to_cpu(pdsfc->ident.max_resp_sz)) {
 		dev_dbg(dev, "Invalid response size %u, max %u\n",
-			rpc->out.len, pdsfc->ident.max_resp_sz);
+			rpc->out.len, le32_to_cpu(pdsfc->ident.max_resp_sz));
 		return -EINVAL;
 	}
 
-	for (i = 0; i < pdsfc->endpoints->num_entries; i++) {
+	num_entries = le32_to_cpu(pdsfc->endpoints->num_entries);
+	for (i = 0; i < num_entries; i++) {
 		if (pdsfc->endpoint_info[i].endpoint == rpc->in.ep) {
 			ep_info = &pdsfc->endpoint_info[i];
 			break;
@@ -326,8 +332,9 @@ static int pdsfc_validate_rpc(struct pdsfc_dev *pdsfc,
 
 	/* reject unsupported and/or out of scope commands */
 	op_entry = (struct pds_fwctl_query_data_operation *)ep_info->operations->entries;
-	for (i = 0; i < ep_info->operations->num_entries; i++) {
-		if (PDS_FWCTL_RPC_OPCODE_CMP(rpc->in.op, op_entry[i].id)) {
+	num_entries = le32_to_cpu(ep_info->operations->num_entries);
+	for (i = 0; i < num_entries; i++) {
+		if (PDS_FWCTL_RPC_OPCODE_CMP(rpc->in.op, le32_to_cpu(op_entry[i].id))) {
 			if (scope < op_entry[i].scope)
 				return -EPERM;
 			return 0;
@@ -402,7 +409,7 @@ static void *pdsfc_fw_rpc(struct fwctl_uctx *uctx, enum fwctl_rpc_scope scope,
 	cmd = (union pds_core_adminq_cmd) {
 		.fwctl_rpc = {
 			.opcode = PDS_FWCTL_CMD_RPC,
-			.flags = PDS_FWCTL_RPC_IND_REQ | PDS_FWCTL_RPC_IND_RESP,
+			.flags = cpu_to_le16(PDS_FWCTL_RPC_IND_REQ | PDS_FWCTL_RPC_IND_RESP),
 			.ep = cpu_to_le32(rpc->in.ep),
 			.op = cpu_to_le32(rpc->in.op),
 			.req_pa = cpu_to_le64(in_payload_dma_addr),
diff --git a/drivers/gpio/TODO b/drivers/gpio/TODO
index b5f0a7a2e1bf..4b70cbaa1caa 100644
--- a/drivers/gpio/TODO
+++ b/drivers/gpio/TODO
@@ -186,3 +186,37 @@ their hardware offsets within the chip.
 
 Encourage users to switch to using them and eventually remove the existing
 global export/unexport attribues.
+
+-------------------------------------------------------------------------------
+
+Remove GPIOD_FLAGS_BIT_NONEXCLUSIVE
+
+GPIOs in the linux kernel are meant to be an exclusive resource. This means
+that the GPIO descriptors (the software representation of the hardware concept)
+are not reference counted and - in general - only one user at a time can
+request a GPIO line and control its settings. The consumer API is designed
+around full control of the line's state as evidenced by the fact that, for
+instance, gpiod_set_value() does indeed drive the line as requested, instead
+of bumping an enable counter of some sort.
+
+A problematic use-case for GPIOs is when two consumers want to use the same
+descriptor independently. An example of such a user is the regulator subsystem
+which may instantiate several struct regulator_dev instances containing
+a struct device but using the same enable GPIO line.
+
+A workaround was introduced in the form of the GPIOD_FLAGS_BIT_NONEXCLUSIVE
+flag but its implementation is problematic: it does not provide any
+synchronization of usage nor did it introduce any enable count meaning the
+non-exclusive users of the same descriptor will in fact "fight" for the
+control over it. This flag should be removed and replaced with a better
+solution, possibly based on the new power sequencing subsystem.
+
+-------------------------------------------------------------------------------
+
+Remove devm_gpiod_unhinge()
+
+devm_gpiod_unhinge() is provided as a way to transfer the ownership of managed
+enable GPIOs to the regulator core. Rather than doing that however, we should
+make it possible for the regulator subsystem to deal with GPIO resources the
+lifetime of which it doesn't control as logically, a GPIO obtained by a caller
+should also be freed by it.
diff --git a/drivers/gpio/gpio-mpc8xxx.c b/drivers/gpio/gpio-mpc8xxx.c
index 0cd4c36ae8aa..541517536489 100644
--- a/drivers/gpio/gpio-mpc8xxx.c
+++ b/drivers/gpio/gpio-mpc8xxx.c
@@ -410,7 +410,9 @@ static int mpc8xxx_probe(struct platform_device *pdev)
 		goto err;
 	}
 
-	device_init_wakeup(dev, true);
+	ret = devm_device_init_wakeup(dev);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to init wakeup\n");
 
 	return 0;
 err:
diff --git a/drivers/gpio/gpio-tegra186.c b/drivers/gpio/gpio-tegra186.c
index 6895b65c86af..d27bfac6c9f5 100644
--- a/drivers/gpio/gpio-tegra186.c
+++ b/drivers/gpio/gpio-tegra186.c
@@ -823,6 +823,7 @@ static int tegra186_gpio_probe(struct platform_device *pdev)
 	struct gpio_irq_chip *irq;
 	struct tegra_gpio *gpio;
 	struct device_node *np;
+	struct resource *res;
 	char **names;
 	int err;
 
@@ -842,19 +843,19 @@ static int tegra186_gpio_probe(struct platform_device *pdev)
 	gpio->num_banks++;
 
 	/* get register apertures */
-	gpio->secure = devm_platform_ioremap_resource_byname(pdev, "security");
-	if (IS_ERR(gpio->secure)) {
-		gpio->secure = devm_platform_ioremap_resource(pdev, 0);
-		if (IS_ERR(gpio->secure))
-			return PTR_ERR(gpio->secure);
-	}
-
-	gpio->base = devm_platform_ioremap_resource_byname(pdev, "gpio");
-	if (IS_ERR(gpio->base)) {
-		gpio->base = devm_platform_ioremap_resource(pdev, 1);
-		if (IS_ERR(gpio->base))
-			return PTR_ERR(gpio->base);
-	}
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "security");
+	if (!res)
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	gpio->secure = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(gpio->secure))
+		return PTR_ERR(gpio->secure);
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "gpio");
+	if (!res)
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	gpio->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(gpio->base))
+		return PTR_ERR(gpio->base);
 
 	err = platform_irq_count(pdev);
 	if (err < 0)
diff --git a/drivers/gpio/gpio-zynq.c b/drivers/gpio/gpio-zynq.c
index be81fa2b17ab..3dae63f3ea21 100644
--- a/drivers/gpio/gpio-zynq.c
+++ b/drivers/gpio/gpio-zynq.c
@@ -1011,6 +1011,7 @@ static void zynq_gpio_remove(struct platform_device *pdev)
 	ret = pm_runtime_get_sync(&pdev->dev);
 	if (ret < 0)
 		dev_warn(&pdev->dev, "pm_runtime_get_sync() Failed\n");
+	device_init_wakeup(&pdev->dev, 0);
 	gpiochip_remove(&gpio->chip);
 	device_set_wakeup_capable(&pdev->dev, 0);
 	pm_runtime_disable(&pdev->dev);
diff --git a/drivers/gpio/gpiolib-devres.c b/drivers/gpio/gpiolib-devres.c
index 08205f355ceb..120d1ec5af3b 100644
--- a/drivers/gpio/gpiolib-devres.c
+++ b/drivers/gpio/gpiolib-devres.c
@@ -317,11 +317,15 @@ EXPORT_SYMBOL_GPL(devm_gpiod_put);
  * @dev:	GPIO consumer
  * @desc:	GPIO descriptor to remove resource management from
  *
+ * *DEPRECATED*
+ * This function should not be used. It's been provided as a workaround for
+ * resource ownership issues in the regulator framework and should be replaced
+ * with a better solution.
+ *
  * Remove resource management from a GPIO descriptor. This is needed when
  * you want to hand over lifecycle management of a descriptor to another
  * mechanism.
  */
-
 void devm_gpiod_unhinge(struct device *dev, struct gpio_desc *desc)
 {
 	int ret;
diff --git a/drivers/gpio/gpiolib-of.c b/drivers/gpio/gpiolib-of.c
index eb667f8f1ead..65f6a7177b78 100644
--- a/drivers/gpio/gpiolib-of.c
+++ b/drivers/gpio/gpiolib-of.c
@@ -193,6 +193,8 @@ static void of_gpio_try_fixup_polarity(const struct device_node *np,
 		 */
 		{ "himax,hx8357",	"gpios-reset",	false },
 		{ "himax,hx8369",	"gpios-reset",	false },
+#endif
+#if IS_ENABLED(CONFIG_MTD_NAND_JZ4780)
 		/*
 		 * The rb-gpios semantics was undocumented and qi,lb60 (along with
 		 * the ingenic driver) got it wrong. The active state encodes the
@@ -266,6 +268,9 @@ static void of_gpio_set_polarity_by_property(const struct device_node *np,
 		{ "fsl,imx8qm-fec",  "phy-reset-gpios", "phy-reset-active-high" },
 		{ "fsl,s32v234-fec", "phy-reset-gpios", "phy-reset-active-high" },
 #endif
+#if IS_ENABLED(CONFIG_MMC_ATMELMCI)
+		{ "atmel,hsmci",       "cd-gpios",     "cd-inverted" },
+#endif
 #if IS_ENABLED(CONFIG_PCI_IMX6)
 		{ "fsl,imx6q-pcie",  "reset-gpio", "reset-gpio-active-high" },
 		{ "fsl,imx6sx-pcie", "reset-gpio", "reset-gpio-active-high" },
@@ -292,9 +297,6 @@ static void of_gpio_set_polarity_by_property(const struct device_node *np,
 		{ "regulator-gpio",    "enable-gpio",  "enable-active-high" },
 		{ "regulator-gpio",    "enable-gpios", "enable-active-high" },
 #endif
-#if IS_ENABLED(CONFIG_MMC_ATMELMCI)
-		{ "atmel,hsmci",       "cd-gpios",     "cd-inverted" },
-#endif
 	};
 	unsigned int i;
 	bool active_high;
diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index b8197502a5ac..cd4fecbb41f2 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -2879,7 +2879,7 @@ static int gpiod_direction_output_raw_commit(struct gpio_desc *desc, int value)
 	 * output-only, but if there is then not even a .set() operation it
 	 * is pretty tricky to drive the output line.
 	 */
-	if (!guard.gc->set && !guard.gc->direction_output) {
+	if (!guard.gc->set && !guard.gc->set_rv && !guard.gc->direction_output) {
 		gpiod_warn(desc,
 			   "%s: missing set() and direction_output() operations\n",
 			   __func__);
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 90cdaa0068bd..f094797f3b2b 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -26,6 +26,11 @@ menuconfig DRM
 	  details.  You should also select and configure AGP
 	  (/dev/agpgart) support if it is available for your platform.
 
+menu "DRM debugging options"
+depends on DRM
+source "drivers/gpu/drm/Kconfig.debug"
+endmenu
+
 if DRM
 
 config DRM_MIPI_DBI
@@ -37,65 +42,6 @@ config DRM_MIPI_DSI
 	bool
 	depends on DRM
 
-config DRM_DEBUG_MM
-	bool "Insert extra checks and debug info into the DRM range managers"
-	default n
-	depends on DRM
-	depends on STACKTRACE_SUPPORT
-	select STACKDEPOT
-	help
-	  Enable allocation tracking of memory manager and leak detection on
-	  shutdown.
-
-	  Recommended for driver developers only.
-
-	  If in doubt, say "N".
-
-config DRM_USE_DYNAMIC_DEBUG
-	bool "use dynamic debug to implement drm.debug"
-	default n
-	depends on BROKEN
-	depends on DRM
-	depends on DYNAMIC_DEBUG || DYNAMIC_DEBUG_CORE
-	depends on JUMP_LABEL
-	help
-	  Use dynamic-debug to avoid drm_debug_enabled() runtime overheads.
-	  Due to callsite counts in DRM drivers (~4k in amdgpu) and 56
-	  bytes per callsite, the .data costs can be substantial, and
-	  are therefore configurable.
-
-config DRM_KUNIT_TEST_HELPERS
-	tristate
-	depends on DRM && KUNIT
-	select DRM_KMS_HELPER
-	help
-	  KUnit Helpers for KMS drivers.
-
-config DRM_KUNIT_TEST
-	tristate "KUnit tests for DRM" if !KUNIT_ALL_TESTS
-	depends on DRM && KUNIT && MMU
-	select DRM_BUDDY
-	select DRM_DISPLAY_DP_HELPER
-	select DRM_DISPLAY_HDMI_STATE_HELPER
-	select DRM_DISPLAY_HELPER
-	select DRM_EXEC
-	select DRM_EXPORT_FOR_TESTS if m
-	select DRM_GEM_SHMEM_HELPER
-	select DRM_KUNIT_TEST_HELPERS
-	select DRM_LIB_RANDOM
-	select PRIME_NUMBERS
-	default KUNIT_ALL_TESTS
-	help
-	  This builds unit tests for DRM. This option is not useful for
-	  distributions or general kernels, but only for kernel
-	  developers working on DRM and associated drivers.
-
-	  For more information on KUnit and unit tests in general,
-	  please refer to the KUnit documentation in
-	  Documentation/dev-tools/kunit/.
-
-	  If in doubt, say "N".
-
 config DRM_KMS_HELPER
 	tristate
 	depends on DRM
@@ -188,7 +134,7 @@ config DRM_DEBUG_DP_MST_TOPOLOGY_REFS
         bool "Enable refcount backtrace history in the DP MST helpers"
 	depends on STACKTRACE_SUPPORT
         select STACKDEPOT
-        depends on DRM_KMS_HELPER
+        select DRM_KMS_HELPER
         depends on DEBUG_KERNEL
         depends on EXPERT
         help
@@ -247,23 +193,6 @@ config DRM_TTM
 	  GPU memory types. Will be enabled automatically if a device driver
 	  uses it.
 
-config DRM_TTM_KUNIT_TEST
-        tristate "KUnit tests for TTM" if !KUNIT_ALL_TESTS
-        default n
-        depends on DRM && KUNIT && MMU && (UML || COMPILE_TEST)
-        select DRM_TTM
-        select DRM_BUDDY
-        select DRM_EXPORT_FOR_TESTS if m
-        select DRM_KUNIT_TEST_HELPERS
-        default KUNIT_ALL_TESTS
-        help
-          Enables unit tests for TTM, a GPU memory manager subsystem used
-          to manage memory buffers. This option is mostly useful for kernel
-          developers. It depends on (UML || COMPILE_TEST) since no other driver
-          which uses TTM can be loaded while running the tests.
-
-          If in doubt, say "N".
-
 config DRM_EXEC
 	tristate
 	depends on DRM
@@ -335,6 +264,8 @@ config DRM_SCHED
 	tristate
 	depends on DRM
 
+source "drivers/gpu/drm/sysfb/Kconfig"
+
 source "drivers/gpu/drm/arm/Kconfig"
 
 source "drivers/gpu/drm/radeon/Kconfig"
@@ -456,6 +387,8 @@ source "drivers/gpu/drm/xlnx/Kconfig"
 
 source "drivers/gpu/drm/gud/Kconfig"
 
+source "drivers/gpu/drm/sitronix/Kconfig"
+
 source "drivers/gpu/drm/solomon/Kconfig"
 
 source "drivers/gpu/drm/sprd/Kconfig"
@@ -476,9 +409,6 @@ config DRM_HYPERV
 
 	 If M is selected the module will be called hyperv_drm.
 
-config DRM_EXPORT_FOR_TESTS
-	bool
-
 # Separate option as not all DRM drivers use it
 config DRM_PANEL_BACKLIGHT_QUIRKS
 	tristate
@@ -491,31 +421,6 @@ config DRM_PRIVACY_SCREEN
 	bool
 	default n
 
-config DRM_WERROR
-	bool "Compile the drm subsystem with warnings as errors"
-	depends on DRM && EXPERT
-	depends on !WERROR
-	default n
-	help
-	  A kernel build should not cause any compiler warnings, and this
-	  enables the '-Werror' flag to enforce that rule in the drm subsystem.
-
-	  The drm subsystem enables more warnings than the kernel default, so
-	  this config option is disabled by default.
-
-	  If in doubt, say N.
-
-config DRM_HEADER_TEST
-	bool "Ensure DRM headers are self-contained and pass kernel-doc"
-	depends on DRM && EXPERT && BROKEN
-	default n
-	help
-	  Ensure the DRM subsystem headers both under drivers/gpu/drm and
-	  include/drm compile, are self-contained, have header guards, and have
-	  no kernel-doc warnings.
-
-	  If in doubt, say N.
-
 endif
 
 # Separate option because drm_panel_orientation_quirks.c is shared with fbdev
diff --git a/drivers/gpu/drm/Kconfig.debug b/drivers/gpu/drm/Kconfig.debug
new file mode 100644
index 000000000000..fa6ee76f4d3c
--- /dev/null
+++ b/drivers/gpu/drm/Kconfig.debug
@@ -0,0 +1,116 @@
+config DRM_USE_DYNAMIC_DEBUG
+	bool "use dynamic debug to implement drm.debug"
+	default n
+	depends on BROKEN
+	depends on DRM
+	depends on DYNAMIC_DEBUG || DYNAMIC_DEBUG_CORE
+	depends on JUMP_LABEL
+	help
+	 Use dynamic-debug to avoid drm_debug_enabled() runtime overheads.
+	 Due to callsite counts in DRM drivers (~4k in amdgpu) and 56
+	 bytes per callsite, the .data costs can be substantial, and
+	 are therefore configurable.
+
+config DRM_WERROR
+	bool "Compile the drm subsystem with warnings as errors"
+	depends on DRM && EXPERT
+	depends on !WERROR
+	default n
+	help
+	  A kernel build should not cause any compiler warnings, and this
+	  enables the '-Werror' flag to enforce that rule in the drm subsystem.
+
+	  The drm subsystem enables more warnings than the kernel default, so
+	  this config option is disabled by default.
+
+	  If in doubt, say N.
+
+config DRM_HEADER_TEST
+	bool "Ensure DRM headers are self-contained and pass kernel-doc"
+	depends on DRM && EXPERT && BROKEN
+	default n
+	help
+	  Ensure the DRM subsystem headers both under drivers/gpu/drm and
+	  include/drm compile, are self-contained, have header guards, and have
+	  no kernel-doc warnings.
+
+	  If in doubt, say N.
+
+config DRM_DEBUG_MM
+	bool "Insert extra checks and debug info into the DRM range managers"
+	default n
+	depends on DRM
+	depends on STACKTRACE_SUPPORT
+	select STACKDEPOT
+	help
+	  Enable allocation tracking of memory manager and leak detection on
+	  shutdown.
+
+	  Recommended for driver developers only.
+
+	  If in doubt, say "N".
+
+config DRM_KUNIT_TEST_HELPERS
+	tristate
+	depends on DRM && KUNIT
+	select DRM_KMS_HELPER
+	help
+	  KUnit Helpers for KMS drivers.
+
+config DRM_KUNIT_TEST
+	tristate "KUnit tests for DRM" if !KUNIT_ALL_TESTS
+	depends on DRM && KUNIT && MMU
+	select DRM_BRIDGE_CONNECTOR
+	select DRM_BUDDY
+	select DRM_DISPLAY_DP_HELPER
+	select DRM_DISPLAY_HDMI_STATE_HELPER
+	select DRM_DISPLAY_HELPER
+	select DRM_EXEC
+	select DRM_EXPORT_FOR_TESTS if m
+	select DRM_GEM_SHMEM_HELPER
+	select DRM_KUNIT_TEST_HELPERS
+	select DRM_LIB_RANDOM
+	select PRIME_NUMBERS
+	default KUNIT_ALL_TESTS
+	help
+	  This builds unit tests for DRM. This option is not useful for
+	  distributions or general kernels, but only for kernel
+	  developers working on DRM and associated drivers.
+
+	  For more information on KUnit and unit tests in general,
+	  please refer to the KUnit documentation in
+	  Documentation/dev-tools/kunit/.
+
+	  If in doubt, say "N".
+
+config DRM_TTM_KUNIT_TEST
+	tristate "KUnit tests for TTM" if !KUNIT_ALL_TESTS
+	default n
+	depends on DRM && KUNIT && MMU && (UML || COMPILE_TEST)
+	select DRM_TTM
+	select DRM_BUDDY
+	select DRM_EXPORT_FOR_TESTS if m
+	select DRM_KUNIT_TEST_HELPERS
+	default KUNIT_ALL_TESTS
+	help
+	  Enables unit tests for TTM, a GPU memory manager subsystem used
+	  to manage memory buffers. This option is mostly useful for kernel
+	  developers. It depends on (UML || COMPILE_TEST) since no other driver
+	  which uses TTM can be loaded while running the tests.
+
+	  If in doubt, say "N".
+
+config DRM_SCHED_KUNIT_TEST
+	tristate "KUnit tests for the DRM scheduler" if !KUNIT_ALL_TESTS
+	select DRM_SCHED
+	depends on DRM && KUNIT
+	default KUNIT_ALL_TESTS
+	help
+	  Choose this option to build unit tests for the DRM scheduler.
+
+	  Recommended for driver developers only.
+
+	  If in doubt, say "N".
+
+config DRM_EXPORT_FOR_TESTS
+	bool
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 947ed126135b..4199715670b1 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -134,6 +134,7 @@ obj-$(CONFIG_DRM_TTM_HELPER) += drm_ttm_helper.o
 drm_kms_helper-y := \
 	drm_atomic_helper.o \
 	drm_atomic_state_helper.o \
+	drm_bridge_helper.o \
 	drm_crtc_helper.o \
 	drm_damage_helper.o \
 	drm_flip_work.o \
@@ -205,6 +206,7 @@ obj-$(CONFIG_DRM_FSL_DCU) += fsl-dcu/
 obj-$(CONFIG_DRM_ETNAVIV) += etnaviv/
 obj-y			+= hisilicon/
 obj-y			+= mxsfb/
+obj-y			+= sysfb/
 obj-y			+= tiny/
 obj-$(CONFIG_DRM_PL111) += pl111/
 obj-$(CONFIG_DRM_TVE200) += tve200/
@@ -220,6 +222,7 @@ obj-$(CONFIG_DRM_TIDSS) += tidss/
 obj-y			+= xlnx/
 obj-y			+= gud/
 obj-$(CONFIG_DRM_HYPERV) += hyperv/
+obj-y			+= sitronix/
 obj-y			+= solomon/
 obj-$(CONFIG_DRM_SPRD) += sprd/
 obj-$(CONFIG_DRM_LOONGSON) += loongson/
diff --git a/drivers/gpu/drm/adp/adp-mipi.c b/drivers/gpu/drm/adp/adp-mipi.c
index ad80542b60ed..2b60128e2c69 100644
--- a/drivers/gpu/drm/adp/adp-mipi.c
+++ b/drivers/gpu/drm/adp/adp-mipi.c
@@ -212,12 +212,13 @@ static const struct mipi_dsi_host_ops adp_dsi_host_ops = {
 };
 
 static int adp_dsi_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct adp_mipi_drv_private *adp =
 		container_of(bridge, struct adp_mipi_drv_private, bridge);
 
-	return drm_bridge_attach(bridge->encoder, adp->next_bridge, bridge, flags);
+	return drm_bridge_attach(encoder, adp->next_bridge, bridge, flags);
 }
 
 static const struct drm_bridge_funcs adp_dsi_bridge_funcs = {
diff --git a/drivers/gpu/drm/adp/adp_drv.c b/drivers/gpu/drm/adp/adp_drv.c
index c98c647f981d..54cde090c3f4 100644
--- a/drivers/gpu/drm/adp/adp_drv.c
+++ b/drivers/gpu/drm/adp/adp_drv.c
@@ -121,7 +121,6 @@ struct adp_drv_private {
 	dma_addr_t mask_iova;
 	int be_irq;
 	int fe_irq;
-	spinlock_t irq_lock;
 	struct drm_pending_vblank_event *event;
 };
 
@@ -288,6 +287,7 @@ static void adp_crtc_atomic_enable(struct drm_crtc *crtc,
 	writel(BIT(0), adp->be + ADBE_BLEND_EN3);
 	writel(BIT(0), adp->be + ADBE_BLEND_BYPASS);
 	writel(BIT(0), adp->be + ADBE_BLEND_EN4);
+	drm_crtc_vblank_on(crtc);
 }
 
 static void adp_crtc_atomic_disable(struct drm_crtc *crtc,
@@ -310,6 +310,7 @@ static void adp_crtc_atomic_flush(struct drm_crtc *crtc,
 				  struct drm_atomic_state *state)
 {
 	u32 frame_num = 1;
+	unsigned long flags;
 	struct adp_drv_private *adp = crtc_to_adp(crtc);
 	struct drm_crtc_state *new_state = drm_atomic_get_new_crtc_state(state, crtc);
 	u64 new_size = ALIGN(new_state->mode.hdisplay *
@@ -330,13 +331,19 @@ static void adp_crtc_atomic_flush(struct drm_crtc *crtc,
 	}
 	writel(ADBE_FIFO_SYNC | frame_num, adp->be + ADBE_FIFO);
 	//FIXME: use adbe flush interrupt
-	spin_lock_irq(&crtc->dev->event_lock);
 	if (crtc->state->event) {
-		drm_crtc_vblank_get(crtc);
-		adp->event = crtc->state->event;
+		struct drm_pending_vblank_event *event = crtc->state->event;
+
+		crtc->state->event = NULL;
+		spin_lock_irqsave(&crtc->dev->event_lock, flags);
+
+		if (drm_crtc_vblank_get(crtc) != 0)
+			drm_crtc_send_vblank_event(crtc, event);
+		else
+			adp->event = event;
+
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
 	}
-	crtc->state->event = NULL;
-	spin_unlock_irq(&crtc->dev->event_lock);
 }
 
 static const struct drm_crtc_funcs adp_crtc_funcs = {
@@ -482,8 +489,6 @@ static irqreturn_t adp_fe_irq(int irq, void *arg)
 	u32 int_status;
 	u32 int_ctl;
 
-	spin_lock(&adp->irq_lock);
-
 	int_status = readl(adp->fe + ADP_INT_STATUS);
 	if (int_status & ADP_INT_STATUS_VBLANK) {
 		drm_crtc_handle_vblank(&adp->crtc);
@@ -501,7 +506,6 @@ static irqreturn_t adp_fe_irq(int irq, void *arg)
 
 	writel(int_status, adp->fe + ADP_INT_STATUS);
 
-	spin_unlock(&adp->irq_lock);
 
 	return IRQ_HANDLED;
 }
@@ -512,8 +516,7 @@ static int adp_drm_bind(struct device *dev)
 	struct adp_drv_private *adp = to_adp(drm);
 	int err;
 
-	adp_disable_vblank(adp);
-	writel(ADP_CTRL_FIFO_ON | ADP_CTRL_VBLANK_ON, adp->fe + ADP_CTRL);
+	writel(ADP_CTRL_FIFO_ON, adp->fe + ADP_CTRL);
 
 	adp->next_bridge = drmm_of_get_bridge(&adp->drm, dev->of_node, 0, 0);
 	if (IS_ERR(adp->next_bridge)) {
@@ -567,8 +570,6 @@ static int adp_probe(struct platform_device *pdev)
 	if (IS_ERR(adp))
 		return PTR_ERR(adp);
 
-	spin_lock_init(&adp->irq_lock);
-
 	dev_set_drvdata(&pdev->dev, &adp->drm);
 
 	err = adp_parse_of(pdev, adp);
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index aacc810cabb3..87080c06e5fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -66,7 +66,7 @@ amdgpu-y += amdgpu_device.o amdgpu_doorbell_mgr.o amdgpu_kms.o \
 	amdgpu_fw_attestation.o amdgpu_securedisplay.o \
 	amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o \
 	amdgpu_ring_mux.o amdgpu_xcp.o amdgpu_seq64.o amdgpu_aca.o amdgpu_dev_coredump.o \
-	amdgpu_cper.o
+	amdgpu_cper.o amdgpu_userq_fence.o amdgpu_eviction_fence.o
 
 amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o
 
@@ -174,7 +174,10 @@ amdgpu-y += \
 amdgpu-y += \
 	amdgpu_mes.o \
 	mes_v11_0.o \
-	mes_v12_0.o
+	mes_v12_0.o \
+
+# add GFX userqueue support
+amdgpu-y += mes_userqueue.o
 
 # add UVD block
 amdgpu-y += \
@@ -253,6 +256,8 @@ amdgpu-y += \
 # add amdkfd interfaces
 amdgpu-y += amdgpu_amdkfd.o
 
+# add gfx usermode queue
+amdgpu-y += amdgpu_userq.o
 
 ifneq ($(CONFIG_HSA_AMD),)
 AMDKFD_PATH := ../amdkfd
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6d83ccfa42ee..836ea081088a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -113,6 +113,8 @@
 #include "amdgpu_xcp.h"
 #include "amdgpu_seq64.h"
 #include "amdgpu_reg_state.h"
+#include "amdgpu_userq.h"
+#include "amdgpu_eviction_fence.h"
 #if defined(CONFIG_DRM_AMD_ISP)
 #include "amdgpu_isp.h"
 #endif
@@ -228,7 +230,7 @@ extern int amdgpu_force_asic_type;
 extern int amdgpu_smartshift_bias;
 extern int amdgpu_use_xgmi_p2p;
 extern int amdgpu_mtype_local;
-extern bool enforce_isolation;
+extern int amdgpu_enforce_isolation;
 #ifdef CONFIG_HSA_AMD
 extern int sched_policy;
 extern bool debug_evictions;
@@ -266,8 +268,10 @@ extern int amdgpu_umsch_mm_fwlog;
 
 extern int amdgpu_user_partt_mode;
 extern int amdgpu_agp;
+extern int amdgpu_rebar;
 
 extern int amdgpu_wbrf;
+extern int amdgpu_user_queue;
 
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
@@ -353,7 +357,6 @@ enum amdgpu_kiq_irq {
 	AMDGPU_CP_KIQ_IRQ_DRIVER0 = 0,
 	AMDGPU_CP_KIQ_IRQ_LAST
 };
-#define SRIOV_USEC_TIMEOUT  1200000 /* wait 12 * 100ms for SRIOV */
 #define MAX_KIQ_REG_WAIT       5000 /* in usecs, 5ms */
 #define MAX_KIQ_REG_BAILOUT_INTERVAL   5 /* in msecs, 5ms */
 #define MAX_KIQ_REG_TRY 1000
@@ -489,7 +492,6 @@ struct amdgpu_flip_work {
 	bool				async;
 };
 
-
 /*
  * file private structure
  */
@@ -502,6 +504,11 @@ struct amdgpu_fpriv {
 	struct mutex		bo_list_lock;
 	struct idr		bo_list_handles;
 	struct amdgpu_ctx_mgr	ctx_mgr;
+	struct amdgpu_userq_mgr	userq_mgr;
+
+	/* Eviction fence infra */
+	struct amdgpu_eviction_fence_mgr evf_mgr;
+
 	/** GPU partition selection */
 	uint32_t		xcp_id;
 };
@@ -513,12 +520,62 @@ int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv);
  */
 #define AMDGPU_MAX_WB 1024	/* Reserve at most 1024 WB slots for amdgpu-owned rings. */
 
+/**
+ * amdgpu_wb - This struct is used for small GPU memory allocation.
+ *
+ * This struct is used to allocate a small amount of GPU memory that can be
+ * used to shadow certain states into the memory. This is especially useful for
+ * providing easy CPU access to some states without requiring register access
+ * (e.g., if some block is power gated, reading register may be problematic).
+ *
+ * Note: the term writeback was initially used because many of the amdgpu
+ * components had some level of writeback memory, and this struct initially
+ * described those components.
+ */
 struct amdgpu_wb {
+
+	/**
+	 * @wb_obj:
+	 *
+	 * Buffer Object used for the writeback memory.
+	 */
 	struct amdgpu_bo	*wb_obj;
+
+	/**
+	 * @wb:
+	 *
+	 * Pointer to the first writeback slot. In terms of CPU address
+	 * this value can be accessed directly by using the offset as an index.
+	 * For the GPU address, it is necessary to use gpu_addr and the offset.
+	 */
 	volatile uint32_t	*wb;
+
+	/**
+	 * @gpu_addr:
+	 *
+	 * Writeback base address in the GPU.
+	 */
 	uint64_t		gpu_addr;
-	u32			num_wb;	/* Number of wb slots actually reserved for amdgpu. */
+
+	/**
+	 * @num_wb:
+	 *
+	 * Number of writeback slots reserved for amdgpu.
+	 */
+	u32			num_wb;
+
+	/**
+	 * @used:
+	 *
+	 * Track the writeback slot already used.
+	 */
 	unsigned long		used[DIV_ROUND_UP(AMDGPU_MAX_WB, BITS_PER_LONG)];
+
+	/**
+	 * @lock:
+	 *
+	 * Protects read and write of the used field array.
+	 */
 	spinlock_t		lock;
 };
 
@@ -552,6 +609,7 @@ struct amdgpu_allowed_register_entry {
  *                   are reset depends on the ASIC. Notably doesn't reset IPs
  *                   shared with the CPU on APUs or the memory controllers (so
  *                   VRAM is not lost). Not available on all ASICs.
+ * @AMD_RESET_LINK: Triggers SW-UP link reset on other GPUs
  * @AMD_RESET_BACO: BACO (Bus Alive, Chip Off) method powers off and on the card
  *                  but without powering off the PCI bus. Suitable only for
  *                  discrete GPUs.
@@ -569,6 +627,7 @@ enum amd_reset_method {
 	AMD_RESET_METHOD_MODE0,
 	AMD_RESET_METHOD_MODE1,
 	AMD_RESET_METHOD_MODE2,
+	AMD_RESET_METHOD_LINK,
 	AMD_RESET_METHOD_BACO,
 	AMD_RESET_METHOD_PCI,
 	AMD_RESET_METHOD_ON_INIT,
@@ -822,6 +881,11 @@ struct amdgpu_mqd_prop {
 	uint32_t hqd_queue_priority;
 	bool allow_tunneling;
 	bool hqd_active;
+	uint64_t shadow_addr;
+	uint64_t gds_bkup_addr;
+	uint64_t csa_addr;
+	uint64_t fence_address;
+	bool tmz_queue;
 };
 
 struct amdgpu_mqd {
@@ -830,6 +894,12 @@ struct amdgpu_mqd {
 			struct amdgpu_mqd_prop *p);
 };
 
+struct amdgpu_pcie_reset_ctx {
+	bool in_link_reset;
+	bool occurs_dpc;
+	bool audio_suspended;
+};
+
 /*
  * Custom Init levels could be defined for different situations where a full
  * initialization of all hardware blocks are not expected. Sample cases are
@@ -854,6 +924,14 @@ struct amdgpu_init_level {
 struct amdgpu_reset_domain;
 struct amdgpu_fru_info;
 
+enum amdgpu_enforce_isolation_mode {
+	AMDGPU_ENFORCE_ISOLATION_DISABLE = 0,
+	AMDGPU_ENFORCE_ISOLATION_ENABLE = 1,
+	AMDGPU_ENFORCE_ISOLATION_ENABLE_LEGACY = 2,
+	AMDGPU_ENFORCE_ISOLATION_NO_CLEANER_SHADER = 3,
+};
+
+
 /*
  * Non-zero (true) if the GPU has VRAM. Zero (false) otherwise.
  */
@@ -1082,6 +1160,13 @@ struct amdgpu_device {
 	bool                            enable_uni_mes;
 	struct amdgpu_mes               mes;
 	struct amdgpu_mqd               mqds[AMDGPU_HW_IP_NUM];
+	const struct amdgpu_userq_funcs *userq_funcs[AMDGPU_HW_IP_NUM];
+
+	/* xarray used to retrieve the user queue fence driver reference
+	 * in the EOP interrupt handler to signal the particular user
+	 * queue fence.
+	 */
+	struct xarray			userq_xa;
 
 	/* df */
 	struct amdgpu_df                df;
@@ -1124,6 +1209,7 @@ struct amdgpu_device {
 	bool				in_s3;
 	bool				in_s4;
 	bool				in_s0ix;
+	suspend_state_t			last_suspend_state;
 
 	enum pp_mp1_state               mp1_state;
 	struct amdgpu_doorbell_index doorbell_index;
@@ -1160,6 +1246,8 @@ struct amdgpu_device {
 	struct pci_saved_state          *pci_state;
 	pci_channel_state_t		pci_channel_state;
 
+	struct amdgpu_pcie_reset_ctx	pcie_reset_ctx;
+
 	/* Track auto wait count on s_barrier settings */
 	bool				barrier_has_auto_waitcnt;
 
@@ -1193,10 +1281,11 @@ struct amdgpu_device {
 	bool                            debug_enable_ras_aca;
 	bool                            debug_exp_resets;
 	bool                            debug_disable_gpu_ring_reset;
+	bool                            debug_vm_userptr;
 
 	/* Protection for the following isolation structure */
 	struct mutex                    enforce_isolation_mutex;
-	bool				enforce_isolation[MAX_XCP];
+	enum amdgpu_enforce_isolation_mode	enforce_isolation[MAX_XCP];
 	struct amdgpu_isolation {
 		void			*owner;
 		struct dma_fence	*spearhead;
@@ -1210,6 +1299,10 @@ struct amdgpu_device {
 	 * in KFD: VRAM or GTT.
 	 */
 	bool                            apu_prefer_gtt;
+
+	struct list_head		userq_mgr_list;
+	struct mutex                    userq_mutex;
+	bool                            userq_halt_for_enforce_isolation;
 };
 
 static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev,
@@ -1464,6 +1557,7 @@ void amdgpu_device_program_register_sequence(struct amdgpu_device *adev,
 					     const u32 array_size);
 
 int amdgpu_device_mode1_reset(struct amdgpu_device *adev);
+int amdgpu_device_link_reset(struct amdgpu_device *adev);
 bool amdgpu_device_supports_atpx(struct drm_device *dev);
 bool amdgpu_device_supports_px(struct drm_device *dev);
 bool amdgpu_device_supports_boco(struct drm_device *dev);
@@ -1614,11 +1708,9 @@ static inline void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_cap
 #if defined(CONFIG_ACPI) && defined(CONFIG_SUSPEND)
 bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev);
 bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev);
-void amdgpu_choose_low_power_state(struct amdgpu_device *adev);
 #else
 static inline bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) { return false; }
 static inline bool amdgpu_acpi_is_s3_active(struct amdgpu_device *adev) { return false; }
-static inline void amdgpu_choose_low_power_state(struct amdgpu_device *adev) { }
 #endif
 
 void amdgpu_register_gpu_instance(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
index b4ad163f42a7..3835f2592914 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
@@ -120,6 +120,9 @@ static void aca_smu_bank_dump(struct amdgpu_device *adev, int idx, int total, st
 	for (i = 0; i < ARRAY_SIZE(aca_regs); i++)
 		RAS_EVENT_LOG(adev, event_id, HW_ERR "ACA[%02d/%02d].%s=0x%016llx\n",
 			      idx + 1, total, aca_regs[i].name, bank->regs[aca_regs[i].reg_idx]);
+
+	if (ACA_REG__STATUS__SCRUB(bank->regs[ACA_REG_IDX_STATUS]))
+		RAS_EVENT_LOG(adev, event_id, HW_ERR "hardware error logged by the scrubber\n");
 }
 
 static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_type type,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index b7f8f2ff143d..707e131f89d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1533,22 +1533,4 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev)
 #endif /* CONFIG_AMD_PMC */
 }
 
-/**
- * amdgpu_choose_low_power_state
- *
- * @adev: amdgpu_device_pointer
- *
- * Choose the target low power state for the GPU
- */
-void amdgpu_choose_low_power_state(struct amdgpu_device *adev)
-{
-	if (adev->in_runpm)
-		return;
-
-	if (amdgpu_acpi_is_s0ix_active(adev))
-		adev->in_s0ix = true;
-	else if (amdgpu_acpi_is_s3_active(adev))
-		adev->in_s3 = true;
-}
-
 #endif /* CONFIG_SUSPEND */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index d2ec4130a316..260165bbe373 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -2559,6 +2559,18 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info,
 			if (ret != -EFAULT)
 				return ret;
 
+			/* If applications unmap memory before destroying the userptr
+			 * from the KFD, trigger a segmentation fault in VM debug mode.
+			 */
+			if (amdgpu_ttm_adev(bo->tbo.bdev)->debug_vm_userptr) {
+				pr_err("Pid %d unmapped memory before destroying userptr at GPU addr 0x%llx\n",
+								pid_nr(process_info->pid), mem->va);
+
+				// Send GPU VM fault to user space
+				kfd_signal_vm_fault_event_with_userptr(kfd_lookup_process_by_pid(process_info->pid),
+								mem->va);
+			}
+
 			ret = 0;
 		}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index eb015bdda8a7..c7d32fb216e4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -281,6 +281,9 @@ static int convert_atom_mem_type_to_vram_type(struct amdgpu_device *adev,
 		case ATOM_DGPU_VRAM_TYPE_GDDR6:
 			vram_type = AMDGPU_VRAM_TYPE_GDDR6;
 			break;
+		case ATOM_DGPU_VRAM_TYPE_HBM3E:
+			vram_type = AMDGPU_VRAM_TYPE_HBM3E;
+			break;
 		default:
 			vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
 			break;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
index 68bce6a6d09d..004a6a9d6b9f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
@@ -252,83 +252,22 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
 
 		if (!adev->pm.fw) {
 			switch (adev->asic_type) {
-			case CHIP_TAHITI:
-				strcpy(fw_name, "radeon/tahiti_smc.bin");
-				break;
-			case CHIP_PITCAIRN:
-				if ((adev->pdev->revision == 0x81) &&
-				    ((adev->pdev->device == 0x6810) ||
-				    (adev->pdev->device == 0x6811))) {
-					info->is_kicker = true;
-					strcpy(fw_name, "radeon/pitcairn_k_smc.bin");
-				} else {
-					strcpy(fw_name, "radeon/pitcairn_smc.bin");
-				}
-				break;
-			case CHIP_VERDE:
-				if (((adev->pdev->device == 0x6820) &&
-					((adev->pdev->revision == 0x81) ||
-					(adev->pdev->revision == 0x83))) ||
-				    ((adev->pdev->device == 0x6821) &&
-					((adev->pdev->revision == 0x83) ||
-					(adev->pdev->revision == 0x87))) ||
-				    ((adev->pdev->revision == 0x87) &&
-					((adev->pdev->device == 0x6823) ||
-					(adev->pdev->device == 0x682b)))) {
-					info->is_kicker = true;
-					strcpy(fw_name, "radeon/verde_k_smc.bin");
-				} else {
-					strcpy(fw_name, "radeon/verde_smc.bin");
-				}
-				break;
-			case CHIP_OLAND:
-				if (((adev->pdev->revision == 0x81) &&
-					((adev->pdev->device == 0x6600) ||
-					(adev->pdev->device == 0x6604) ||
-					(adev->pdev->device == 0x6605) ||
-					(adev->pdev->device == 0x6610))) ||
-				    ((adev->pdev->revision == 0x83) &&
-					(adev->pdev->device == 0x6610))) {
-					info->is_kicker = true;
-					strcpy(fw_name, "radeon/oland_k_smc.bin");
-				} else {
-					strcpy(fw_name, "radeon/oland_smc.bin");
-				}
-				break;
-			case CHIP_HAINAN:
-				if (((adev->pdev->revision == 0x81) &&
-					(adev->pdev->device == 0x6660)) ||
-				    ((adev->pdev->revision == 0x83) &&
-					((adev->pdev->device == 0x6660) ||
-					(adev->pdev->device == 0x6663) ||
-					(adev->pdev->device == 0x6665) ||
-					 (adev->pdev->device == 0x6667)))) {
-					info->is_kicker = true;
-					strcpy(fw_name, "radeon/hainan_k_smc.bin");
-				} else if ((adev->pdev->revision == 0xc3) &&
-					 (adev->pdev->device == 0x6665)) {
-					info->is_kicker = true;
-					strcpy(fw_name, "radeon/banks_k_2_smc.bin");
-				} else {
-					strcpy(fw_name, "radeon/hainan_smc.bin");
-				}
-				break;
 			case CHIP_BONAIRE:
 				if ((adev->pdev->revision == 0x80) ||
 					(adev->pdev->revision == 0x81) ||
 					(adev->pdev->device == 0x665f)) {
 					info->is_kicker = true;
-					strcpy(fw_name, "amdgpu/bonaire_k_smc.bin");
+					strscpy(fw_name, "amdgpu/bonaire_k_smc.bin");
 				} else {
-					strcpy(fw_name, "amdgpu/bonaire_smc.bin");
+					strscpy(fw_name, "amdgpu/bonaire_smc.bin");
 				}
 				break;
 			case CHIP_HAWAII:
 				if (adev->pdev->revision == 0x80) {
 					info->is_kicker = true;
-					strcpy(fw_name, "amdgpu/hawaii_k_smc.bin");
+					strscpy(fw_name, "amdgpu/hawaii_k_smc.bin");
 				} else {
-					strcpy(fw_name, "amdgpu/hawaii_smc.bin");
+					strscpy(fw_name, "amdgpu/hawaii_smc.bin");
 				}
 				break;
 			case CHIP_TOPAZ:
@@ -338,76 +277,76 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
 				    ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0xD1)) ||
 				    ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0xD3))) {
 					info->is_kicker = true;
-					strcpy(fw_name, "amdgpu/topaz_k_smc.bin");
+					strscpy(fw_name, "amdgpu/topaz_k_smc.bin");
 				} else
-					strcpy(fw_name, "amdgpu/topaz_smc.bin");
+					strscpy(fw_name, "amdgpu/topaz_smc.bin");
 				break;
 			case CHIP_TONGA:
 				if (((adev->pdev->device == 0x6939) && (adev->pdev->revision == 0xf1)) ||
 				    ((adev->pdev->device == 0x6938) && (adev->pdev->revision == 0xf1))) {
 					info->is_kicker = true;
-					strcpy(fw_name, "amdgpu/tonga_k_smc.bin");
+					strscpy(fw_name, "amdgpu/tonga_k_smc.bin");
 				} else
-					strcpy(fw_name, "amdgpu/tonga_smc.bin");
+					strscpy(fw_name, "amdgpu/tonga_smc.bin");
 				break;
 			case CHIP_FIJI:
-				strcpy(fw_name, "amdgpu/fiji_smc.bin");
+				strscpy(fw_name, "amdgpu/fiji_smc.bin");
 				break;
 			case CHIP_POLARIS11:
 				if (type == CGS_UCODE_ID_SMU) {
 					if (ASICID_IS_P21(adev->pdev->device, adev->pdev->revision)) {
 						info->is_kicker = true;
-						strcpy(fw_name, "amdgpu/polaris11_k_smc.bin");
+						strscpy(fw_name, "amdgpu/polaris11_k_smc.bin");
 					} else if (ASICID_IS_P31(adev->pdev->device, adev->pdev->revision)) {
 						info->is_kicker = true;
-						strcpy(fw_name, "amdgpu/polaris11_k2_smc.bin");
+						strscpy(fw_name, "amdgpu/polaris11_k2_smc.bin");
 					} else {
-						strcpy(fw_name, "amdgpu/polaris11_smc.bin");
+						strscpy(fw_name, "amdgpu/polaris11_smc.bin");
 					}
 				} else if (type == CGS_UCODE_ID_SMU_SK) {
-					strcpy(fw_name, "amdgpu/polaris11_smc_sk.bin");
+					strscpy(fw_name, "amdgpu/polaris11_smc_sk.bin");
 				}
 				break;
 			case CHIP_POLARIS10:
 				if (type == CGS_UCODE_ID_SMU) {
 					if (ASICID_IS_P20(adev->pdev->device, adev->pdev->revision)) {
 						info->is_kicker = true;
-						strcpy(fw_name, "amdgpu/polaris10_k_smc.bin");
+						strscpy(fw_name, "amdgpu/polaris10_k_smc.bin");
 					} else if (ASICID_IS_P30(adev->pdev->device, adev->pdev->revision)) {
 						info->is_kicker = true;
-						strcpy(fw_name, "amdgpu/polaris10_k2_smc.bin");
+						strscpy(fw_name, "amdgpu/polaris10_k2_smc.bin");
 					} else {
-						strcpy(fw_name, "amdgpu/polaris10_smc.bin");
+						strscpy(fw_name, "amdgpu/polaris10_smc.bin");
 					}
 				} else if (type == CGS_UCODE_ID_SMU_SK) {
-					strcpy(fw_name, "amdgpu/polaris10_smc_sk.bin");
+					strscpy(fw_name, "amdgpu/polaris10_smc_sk.bin");
 				}
 				break;
 			case CHIP_POLARIS12:
 				if (ASICID_IS_P23(adev->pdev->device, adev->pdev->revision)) {
 					info->is_kicker = true;
-					strcpy(fw_name, "amdgpu/polaris12_k_smc.bin");
+					strscpy(fw_name, "amdgpu/polaris12_k_smc.bin");
 				} else {
-					strcpy(fw_name, "amdgpu/polaris12_smc.bin");
+					strscpy(fw_name, "amdgpu/polaris12_smc.bin");
 				}
 				break;
 			case CHIP_VEGAM:
-				strcpy(fw_name, "amdgpu/vegam_smc.bin");
+				strscpy(fw_name, "amdgpu/vegam_smc.bin");
 				break;
 			case CHIP_VEGA10:
 				if ((adev->pdev->device == 0x687f) &&
 					((adev->pdev->revision == 0xc0) ||
 					(adev->pdev->revision == 0xc1) ||
 					(adev->pdev->revision == 0xc3)))
-					strcpy(fw_name, "amdgpu/vega10_acg_smc.bin");
+					strscpy(fw_name, "amdgpu/vega10_acg_smc.bin");
 				else
-					strcpy(fw_name, "amdgpu/vega10_smc.bin");
+					strscpy(fw_name, "amdgpu/vega10_smc.bin");
 				break;
 			case CHIP_VEGA12:
-				strcpy(fw_name, "amdgpu/vega12_smc.bin");
+				strscpy(fw_name, "amdgpu/vega12_smc.bin");
 				break;
 			case CHIP_VEGA20:
-				strcpy(fw_name, "amdgpu/vega20_smc.bin");
+				strscpy(fw_name, "amdgpu/vega20_smc.bin");
 				break;
 			default:
 				DRM_ERROR("SMC firmware not supported\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
index 360e07a5c7c1..5a234eadae8b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
@@ -549,7 +549,7 @@ int amdgpu_cper_init(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (!amdgpu_aca_is_enabled(adev))
+	if (!amdgpu_aca_is_enabled(adev) && !amdgpu_sriov_ras_cper_en(adev))
 		return 0;
 
 	r = amdgpu_cper_ring_init(adev);
@@ -568,7 +568,7 @@ int amdgpu_cper_init(struct amdgpu_device *adev)
 
 int amdgpu_cper_fini(struct amdgpu_device *adev)
 {
-	if (!amdgpu_aca_is_enabled(adev))
+	if (!amdgpu_aca_is_enabled(adev) && !amdgpu_sriov_ras_cper_en(adev))
 		return 0;
 
 	adev->cper.enabled = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 82df06a72ee0..9ea0d9b71f48 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -296,7 +296,25 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 				       num_ibs[i], &p->jobs[i]);
 		if (ret)
 			goto free_all_kdata;
-		p->jobs[i]->enforce_isolation = p->adev->enforce_isolation[fpriv->xcp_id];
+		switch (p->adev->enforce_isolation[fpriv->xcp_id]) {
+		case AMDGPU_ENFORCE_ISOLATION_DISABLE:
+		default:
+			p->jobs[i]->enforce_isolation = false;
+			p->jobs[i]->run_cleaner_shader = false;
+			break;
+		case AMDGPU_ENFORCE_ISOLATION_ENABLE:
+			p->jobs[i]->enforce_isolation = true;
+			p->jobs[i]->run_cleaner_shader = true;
+			break;
+		case AMDGPU_ENFORCE_ISOLATION_ENABLE_LEGACY:
+			p->jobs[i]->enforce_isolation = true;
+			p->jobs[i]->run_cleaner_shader = false;
+			break;
+		case AMDGPU_ENFORCE_ISOLATION_NO_CLEANER_SHADER:
+			p->jobs[i]->enforce_isolation = true;
+			p->jobs[i]->run_cleaner_shader = false;
+			break;
+		}
 	}
 	p->gang_leader = p->jobs[p->gang_leader_idx];
 
@@ -349,6 +367,10 @@ static int amdgpu_cs_p2_ib(struct amdgpu_cs_parser *p,
 	ring = amdgpu_job_ring(job);
 	ib = &job->ibs[job->num_ibs++];
 
+	/* submissions to kernel queues are disabled */
+	if (ring->no_user_submission)
+		return -EINVAL;
+
 	/* MM engine doesn't support user fences */
 	if (p->uf_bo && ring->funcs->no_user_fence)
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
index cfdf558b48b6..02138aa55793 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -109,7 +109,7 @@ int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	struct drm_exec exec;
 	int r;
 
-	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
+	drm_exec_init(&exec, 0, 0);
 	drm_exec_until_all_locked(&exec) {
 		r = amdgpu_vm_lock_pd(vm, &exec, 0);
 		if (likely(!r))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index a1450f13d963..8e626f50b362 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -2105,6 +2105,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 	amdgpu_rap_debugfs_init(adev);
 	amdgpu_securedisplay_debugfs_init(adev);
 	amdgpu_fw_attestation_debugfs_init(adev);
+	amdgpu_psp_debugfs_init(adev);
 
 	debugfs_create_file("amdgpu_evict_vram", 0400, root, adev,
 			    &amdgpu_evict_vram_fops);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index a30111d2c3ea..4d1b54f58495 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -85,6 +85,7 @@
 
 #if IS_ENABLED(CONFIG_X86)
 #include <asm/intel-family.h>
+#include <asm/cpu_device_id.h>
 #endif
 
 MODULE_FIRMWARE("amdgpu/vega10_gpu_info.bin");
@@ -1680,6 +1681,9 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev)
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
+	if (!amdgpu_rebar)
+		return 0;
+
 	/* resizing on Dell G5 SE platforms causes problems with runtime pm */
 	if ((amdgpu_runtime_pm != 0) &&
 	    adev->pdev->vendor == PCI_VENDOR_ID_ATI &&
@@ -1870,6 +1874,35 @@ static bool amdgpu_device_pcie_dynamic_switching_supported(struct amdgpu_device
 	return true;
 }
 
+static bool amdgpu_device_aspm_support_quirk(struct amdgpu_device *adev)
+{
+#if IS_ENABLED(CONFIG_X86)
+	struct cpuinfo_x86 *c = &cpu_data(0);
+
+	if (!(amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 0) ||
+		  amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(12, 0, 1)))
+		return false;
+
+	if (c->x86 == 6 &&
+		adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) {
+		switch (c->x86_model) {
+		case VFM_MODEL(INTEL_ALDERLAKE):
+		case VFM_MODEL(INTEL_ALDERLAKE_L):
+		case VFM_MODEL(INTEL_RAPTORLAKE):
+		case VFM_MODEL(INTEL_RAPTORLAKE_P):
+		case VFM_MODEL(INTEL_RAPTORLAKE_S):
+			return true;
+		default:
+			return false;
+		}
+	} else {
+		return false;
+	}
+#else
+	return false;
+#endif
+}
+
 /**
  * amdgpu_device_should_use_aspm - check if the device should program ASPM
  *
@@ -1894,7 +1927,7 @@ bool amdgpu_device_should_use_aspm(struct amdgpu_device *adev)
 	}
 	if (adev->flags & AMD_IS_APU)
 		return false;
-	if (!(adev->pm.pp_feature & PP_PCIE_DPM_MASK))
+	if (amdgpu_device_aspm_support_quirk(adev))
 		return false;
 	return pcie_aspm_enabled(adev->pdev);
 }
@@ -2112,8 +2145,31 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
 
 	adev->firmware.load_type = amdgpu_ucode_get_load_type(adev, amdgpu_fw_load_type);
 
-	for (i = 0; i < MAX_XCP; i++)
-		adev->enforce_isolation[i] = !!enforce_isolation;
+	for (i = 0; i < MAX_XCP; i++) {
+		switch (amdgpu_enforce_isolation) {
+		case -1:
+		case 0:
+		default:
+			/* disable */
+			adev->enforce_isolation[i] = AMDGPU_ENFORCE_ISOLATION_DISABLE;
+			break;
+		case 1:
+			/* enable */
+			adev->enforce_isolation[i] =
+				AMDGPU_ENFORCE_ISOLATION_ENABLE;
+			break;
+		case 2:
+			/* enable legacy mode */
+			adev->enforce_isolation[i] =
+				AMDGPU_ENFORCE_ISOLATION_ENABLE_LEGACY;
+			break;
+		case 3:
+			/* enable only process isolation without submitting cleaner shader */
+			adev->enforce_isolation[i] =
+				AMDGPU_ENFORCE_ISOLATION_NO_CLEANER_SHADER;
+			break;
+		}
+	}
 
 	return 0;
 }
@@ -2689,6 +2745,13 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 		break;
 	}
 
+	/* Check for IP version 9.4.3 with A0 hardware */
+	if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) &&
+	    !amdgpu_device_get_rev_id(adev)) {
+		dev_err(adev->dev, "Unsupported A0 hardware\n");
+		return -ENODEV;	/* device unsupported - no device error */
+	}
+
 	if (amdgpu_has_atpx() &&
 	    (amdgpu_is_atpx_hybrid() ||
 	     amdgpu_has_atpx_dgpu_power_cntl()) &&
@@ -2701,7 +2764,6 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 		adev->has_pr3 = parent ? pci_pr3_present(parent) : false;
 	}
 
-
 	adev->pm.pp_feature = amdgpu_pp_feature_mask;
 	if (amdgpu_sriov_vf(adev) || sched_policy == KFD_SCHED_POLICY_NO_HWS)
 		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
@@ -3172,6 +3234,7 @@ static bool amdgpu_device_check_vram_lost(struct amdgpu_device *adev)
 	 * always assumed to be lost.
 	 */
 	switch (amdgpu_asic_reset_method(adev)) {
+	case AMD_RESET_METHOD_LINK:
 	case AMD_RESET_METHOD_BACO:
 	case AMD_RESET_METHOD_MODE1:
 		return true;
@@ -3455,6 +3518,7 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev)
 	amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE);
 
 	amdgpu_amdkfd_suspend(adev, false);
+	amdgpu_userq_suspend(adev);
 
 	/* Workaround for ASICs need to disable SMC first */
 	amdgpu_device_smu_fini_early(adev);
@@ -3510,6 +3574,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 			amdgpu_device_mem_scratch_fini(adev);
 			amdgpu_ib_pool_fini(adev);
 			amdgpu_seq64_fini(adev);
+			amdgpu_doorbell_fini(adev);
 		}
 		if (adev->ip_blocks[i].version->funcs->sw_fini) {
 			r = adev->ip_blocks[i].version->funcs->sw_fini(&adev->ip_blocks[i]);
@@ -3643,6 +3708,13 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev)
 			    adev, adev->ip_blocks[i].version->type))
 			continue;
 
+		/* Since we skip suspend for S0i3, we need to cancel the delayed
+		 * idle work here as the suspend callback never gets called.
+		 */
+		if (adev->in_s0ix &&
+		    adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX &&
+		    amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(10, 0, 0))
+			cancel_delayed_work_sync(&adev->gfx.idle_work);
 		/* skip suspend of gfx/mes and psp for S0ix
 		 * gfx is in gfxoff state, so on resume it will exit gfxoff just
 		 * like at runtime. PSP is also part of the always on hardware
@@ -4299,9 +4371,10 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		amdgpu_sync_create(&adev->isolation[i].active);
 		amdgpu_sync_create(&adev->isolation[i].prev);
 	}
-	mutex_init(&adev->gfx.kfd_sch_mutex);
+	mutex_init(&adev->gfx.userq_sch_mutex);
 	mutex_init(&adev->gfx.workload_profile_mutex);
 	mutex_init(&adev->vcn.workload_profile_mutex);
+	mutex_init(&adev->userq_mutex);
 
 	amdgpu_device_init_apu_flags(adev);
 
@@ -4321,12 +4394,16 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	spin_lock_init(&adev->virt.rlcg_reg_lock);
 	spin_lock_init(&adev->wb.lock);
 
+	xa_init_flags(&adev->userq_xa, XA_FLAGS_LOCK_IRQ);
+
 	INIT_LIST_HEAD(&adev->reset_list);
 
 	INIT_LIST_HEAD(&adev->ras_list);
 
 	INIT_LIST_HEAD(&adev->pm.od_kobj_list);
 
+	INIT_LIST_HEAD(&adev->userq_mgr_list);
+
 	INIT_DELAYED_WORK(&adev->delayed_init_work,
 			  amdgpu_device_delayed_init_work_handler);
 	INIT_DELAYED_WORK(&adev->gfx.gfx_off_delay_work,
@@ -4851,7 +4928,6 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev)
 
 		iounmap(adev->rmmio);
 		adev->rmmio = NULL;
-		amdgpu_doorbell_fini(adev);
 		drm_dev_exit(idx);
 	}
 
@@ -4900,28 +4976,20 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev)
  * @data: data
  *
  * This function is called when the system is about to suspend or hibernate.
- * It is used to evict resources from the device before the system goes to
- * sleep while there is still access to swap.
+ * It is used to set the appropriate flags so that eviction can be optimized
+ * in the pm prepare callback.
  */
 static int amdgpu_device_pm_notifier(struct notifier_block *nb, unsigned long mode,
 				     void *data)
 {
 	struct amdgpu_device *adev = container_of(nb, struct amdgpu_device, pm_nb);
-	int r;
 
 	switch (mode) {
 	case PM_HIBERNATION_PREPARE:
 		adev->in_s4 = true;
-		fallthrough;
-	case PM_SUSPEND_PREPARE:
-		r = amdgpu_device_evict_resources(adev);
-		/*
-		 * This is considered non-fatal at this time because
-		 * amdgpu_device_prepare() will also fatally evict resources.
-		 * See https://gitlab.freedesktop.org/drm/amd/-/issues/3781
-		 */
-		if (r)
-			drm_warn(adev_to_drm(adev), "Failed to evict resources, freeze active processes if problems occur: %d\n", r);
+		break;
+	case PM_POST_HIBERNATION:
+		adev->in_s4 = false;
 		break;
 	}
 
@@ -4942,15 +5010,13 @@ int amdgpu_device_prepare(struct drm_device *dev)
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	int i, r;
 
-	amdgpu_choose_low_power_state(adev);
-
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
 	/* Evict the majority of BOs before starting suspend sequence */
 	r = amdgpu_device_evict_resources(adev);
 	if (r)
-		goto unprepare;
+		return r;
 
 	flush_delayed_work(&adev->gfx.gfx_off_delay_work);
 
@@ -4961,15 +5027,10 @@ int amdgpu_device_prepare(struct drm_device *dev)
 			continue;
 		r = adev->ip_blocks[i].version->funcs->prepare_suspend(&adev->ip_blocks[i]);
 		if (r)
-			goto unprepare;
+			return r;
 	}
 
 	return 0;
-
-unprepare:
-	adev->in_s0ix = adev->in_s3 = adev->in_s4 = false;
-
-	return r;
 }
 
 /**
@@ -5011,8 +5072,10 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 
 	amdgpu_device_ip_suspend_phase1(adev);
 
-	if (!adev->in_s0ix)
+	if (!adev->in_s0ix) {
 		amdgpu_amdkfd_suspend(adev, adev->in_runpm);
+		amdgpu_userq_suspend(adev);
+	}
 
 	r = amdgpu_device_evict_resources(adev);
 	if (r)
@@ -5079,6 +5142,10 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
 		r = amdgpu_amdkfd_resume(adev, adev->in_runpm);
 		if (r)
 			goto exit;
+
+		r = amdgpu_userq_resume(adev);
+		if (r)
+			goto exit;
 	}
 
 	r = amdgpu_device_ip_late_init(adev);
@@ -5127,9 +5194,6 @@ exit:
 	}
 	adev->in_suspend = false;
 
-	if (adev->enable_mes)
-		amdgpu_mes_self_test(adev);
-
 	if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D0))
 		DRM_WARN("smart shift update failed\n");
 
@@ -5510,6 +5574,29 @@ mode1_reset_failed:
 	return ret;
 }
 
+int amdgpu_device_link_reset(struct amdgpu_device *adev)
+{
+	int ret = 0;
+
+	dev_info(adev->dev, "GPU link reset\n");
+
+	if (!adev->pcie_reset_ctx.occurs_dpc)
+		ret = amdgpu_dpm_link_reset(adev);
+
+	if (ret)
+		goto link_reset_failed;
+
+	ret = amdgpu_psp_wait_for_bootloader(adev);
+	if (ret)
+		goto link_reset_failed;
+
+	return 0;
+
+link_reset_failed:
+	dev_err(adev->dev, "GPU link reset failed\n");
+	return ret;
+}
+
 int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 				 struct amdgpu_reset_context *reset_context)
 {
@@ -5814,6 +5901,7 @@ static void amdgpu_device_set_mp1_state(struct amdgpu_device *adev)
 
 	switch (amdgpu_asic_reset_method(adev)) {
 	case AMD_RESET_METHOD_MODE1:
+	case AMD_RESET_METHOD_LINK:
 		adev->mp1_state = PP_MP1_STATE_SHUTDOWN;
 		break;
 	case AMD_RESET_METHOD_MODE2:
@@ -5930,94 +6018,42 @@ static int amdgpu_device_health_check(struct list_head *device_list_handle)
 	return ret;
 }
 
-/**
- * amdgpu_device_gpu_recover - reset the asic and recover scheduler
- *
- * @adev: amdgpu_device pointer
- * @job: which job trigger hang
- * @reset_context: amdgpu reset context pointer
- *
- * Attempt to reset the GPU if it has hung (all asics).
- * Attempt to do soft-reset or full-reset and reinitialize Asic
- * Returns 0 for success or an error on failure.
- */
-
-int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
+static int amdgpu_device_halt_activities(struct amdgpu_device *adev,
 			      struct amdgpu_job *job,
-			      struct amdgpu_reset_context *reset_context)
+			      struct amdgpu_reset_context *reset_context,
+			      struct list_head *device_list,
+			      struct amdgpu_hive_info *hive,
+			      bool need_emergency_restart)
 {
-	struct list_head device_list, *device_list_handle =  NULL;
-	bool job_signaled = false;
-	struct amdgpu_hive_info *hive = NULL;
+	struct list_head *device_list_handle =  NULL;
 	struct amdgpu_device *tmp_adev = NULL;
 	int i, r = 0;
-	bool need_emergency_restart = false;
-	bool audio_suspended = false;
-	int retry_limit = AMDGPU_MAX_RETRY_LIMIT;
-
-	/*
-	 * If it reaches here because of hang/timeout and a RAS error is
-	 * detected at the same time, let RAS recovery take care of it.
-	 */
-	if (amdgpu_ras_is_err_state(adev, AMDGPU_RAS_BLOCK__ANY) &&
-	    !amdgpu_sriov_vf(adev) &&
-	    reset_context->src != AMDGPU_RESET_SRC_RAS) {
-		dev_dbg(adev->dev,
-			"Gpu recovery from source: %d yielding to RAS error recovery handling",
-			reset_context->src);
-		return 0;
-	}
-	/*
-	 * Special case: RAS triggered and full reset isn't supported
-	 */
-	need_emergency_restart = amdgpu_ras_need_emergency_restart(adev);
-
-	/*
-	 * Flush RAM to disk so that after reboot
-	 * the user can read log and see why the system rebooted.
-	 */
-	if (need_emergency_restart && amdgpu_ras_get_context(adev) &&
-		amdgpu_ras_get_context(adev)->reboot) {
-		DRM_WARN("Emergency reboot.");
-
-		ksys_sync_helper();
-		emergency_restart();
-	}
-
-	dev_info(adev->dev, "GPU %s begin!\n",
-		need_emergency_restart ? "jobs stop":"reset");
-
-	if (!amdgpu_sriov_vf(adev))
-		hive = amdgpu_get_xgmi_hive(adev);
-	if (hive)
-		mutex_lock(&hive->hive_lock);
 
-	reset_context->job = job;
-	reset_context->hive = hive;
 	/*
 	 * Build list of devices to reset.
 	 * In case we are in XGMI hive mode, resort the device list
 	 * to put adev in the 1st position.
 	 */
-	INIT_LIST_HEAD(&device_list);
 	if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1) && hive) {
 		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
-			list_add_tail(&tmp_adev->reset_list, &device_list);
+			list_add_tail(&tmp_adev->reset_list, device_list);
 			if (adev->shutdown)
 				tmp_adev->shutdown = true;
+			if (adev->pcie_reset_ctx.occurs_dpc)
+				tmp_adev->pcie_reset_ctx.in_link_reset = true;
 		}
-		if (!list_is_first(&adev->reset_list, &device_list))
-			list_rotate_to_front(&adev->reset_list, &device_list);
-		device_list_handle = &device_list;
+		if (!list_is_first(&adev->reset_list, device_list))
+			list_rotate_to_front(&adev->reset_list, device_list);
+		device_list_handle = device_list;
 	} else {
-		list_add_tail(&adev->reset_list, &device_list);
-		device_list_handle = &device_list;
+		list_add_tail(&adev->reset_list, device_list);
+		device_list_handle = device_list;
 	}
 
-	if (!amdgpu_sriov_vf(adev)) {
+	if (!amdgpu_sriov_vf(adev) && (!adev->pcie_reset_ctx.occurs_dpc)) {
 		r = amdgpu_device_health_check(device_list_handle);
 		if (r)
-			goto end_reset;
+			return r;
 	}
 
 	/* We need to lock reset domain only once both for XGMI and single device */
@@ -6041,7 +6077,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 		 * some audio codec errors.
 		 */
 		if (!amdgpu_device_suspend_display_audio(tmp_adev))
-			audio_suspended = true;
+			tmp_adev->pcie_reset_ctx.audio_suspended = true;
 
 		amdgpu_ras_set_error_query_ready(tmp_adev, false);
 
@@ -6059,6 +6095,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 
 		/* disable ras on ALL IPs */
 		if (!need_emergency_restart &&
+		      (!adev->pcie_reset_ctx.occurs_dpc) &&
 		      amdgpu_device_ip_need_full_reset(tmp_adev))
 			amdgpu_ras_suspend(tmp_adev);
 
@@ -6076,24 +6113,24 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 		atomic_inc(&tmp_adev->gpu_reset_counter);
 	}
 
-	if (need_emergency_restart)
-		goto skip_sched_resume;
+	return r;
+}
 
-	/*
-	 * Must check guilty signal here since after this point all old
-	 * HW fences are force signaled.
-	 *
-	 * job->base holds a reference to parent fence
-	 */
-	if (job && dma_fence_is_signaled(&job->hw_fence)) {
-		job_signaled = true;
-		dev_info(adev->dev, "Guilty job already signaled, skipping HW reset");
-		goto skip_hw_reset;
-	}
+static int amdgpu_device_asic_reset(struct amdgpu_device *adev,
+			      struct list_head *device_list,
+			      struct amdgpu_reset_context *reset_context)
+{
+	struct amdgpu_device *tmp_adev = NULL;
+	int retry_limit = AMDGPU_MAX_RETRY_LIMIT;
+	int r = 0;
 
 retry:	/* Rest of adevs pre asic reset from XGMI hive. */
-	list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
+	list_for_each_entry(tmp_adev, device_list, reset_list) {
+		if (adev->pcie_reset_ctx.occurs_dpc)
+			tmp_adev->no_hw_access = true;
 		r = amdgpu_device_pre_asic_reset(tmp_adev, reset_context);
+		if (adev->pcie_reset_ctx.occurs_dpc)
+			tmp_adev->no_hw_access = false;
 		/*TODO Should we stop ?*/
 		if (r) {
 			dev_err(tmp_adev->dev, "GPU pre asic reset failed with err, %d for drm dev, %s ",
@@ -6105,6 +6142,11 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 	/* Actual ASIC resets if needed.*/
 	/* Host driver will handle XGMI hive reset for SRIOV */
 	if (amdgpu_sriov_vf(adev)) {
+
+		/* Bail out of reset early */
+		if (amdgpu_ras_is_rma(adev))
+			return -ENODEV;
+
 		if (amdgpu_ras_get_fed_status(adev) || amdgpu_virt_rcvd_ras_interrupt(adev)) {
 			dev_dbg(adev->dev, "Detected RAS error, wait for FLR completion\n");
 			amdgpu_ras_set_fed(adev, true);
@@ -6119,12 +6161,12 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 		if (r)
 			adev->asic_reset_res = r;
 	} else {
-		r = amdgpu_do_asic_reset(device_list_handle, reset_context);
+		r = amdgpu_do_asic_reset(device_list, reset_context);
 		if (r && r == -EAGAIN)
 			goto retry;
 	}
 
-	list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
+	list_for_each_entry(tmp_adev, device_list, reset_list) {
 		/*
 		 * Drop any pending non scheduler resets queued before reset is done.
 		 * Any reset scheduled after this point would be valid. Scheduler resets
@@ -6134,10 +6176,18 @@ retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 		amdgpu_device_stop_pending_resets(tmp_adev);
 	}
 
-skip_hw_reset:
+	return r;
+}
+
+static int amdgpu_device_sched_resume(struct list_head *device_list,
+			      struct amdgpu_reset_context *reset_context,
+			      bool   job_signaled)
+{
+	struct amdgpu_device *tmp_adev = NULL;
+	int i, r = 0;
 
 	/* Post ASIC reset for all devs .*/
-	list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
+	list_for_each_entry(tmp_adev, device_list, reset_list) {
 
 		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
 			struct amdgpu_ring *ring = tmp_adev->rings[i];
@@ -6173,8 +6223,16 @@ skip_hw_reset:
 		}
 	}
 
-skip_sched_resume:
-	list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
+	return r;
+}
+
+static void amdgpu_device_gpu_resume(struct amdgpu_device *adev,
+			      struct list_head *device_list,
+			      bool   need_emergency_restart)
+{
+	struct amdgpu_device *tmp_adev = NULL;
+
+	list_for_each_entry(tmp_adev, device_list, reset_list) {
 		/* unlock kfd: SRIOV would do it separately */
 		if (!need_emergency_restart && !amdgpu_sriov_vf(tmp_adev))
 			amdgpu_amdkfd_post_reset(tmp_adev);
@@ -6185,18 +6243,114 @@ skip_sched_resume:
 		if (!adev->kfd.init_complete)
 			amdgpu_amdkfd_device_init(adev);
 
-		if (audio_suspended)
+		if (tmp_adev->pcie_reset_ctx.audio_suspended)
 			amdgpu_device_resume_display_audio(tmp_adev);
 
 		amdgpu_device_unset_mp1_state(tmp_adev);
 
 		amdgpu_ras_set_error_query_ready(tmp_adev, true);
+
 	}
 
-	tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device,
+	tmp_adev = list_first_entry(device_list, struct amdgpu_device,
 					    reset_list);
 	amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain);
 
+}
+
+
+/**
+ * amdgpu_device_gpu_recover - reset the asic and recover scheduler
+ *
+ * @adev: amdgpu_device pointer
+ * @job: which job trigger hang
+ * @reset_context: amdgpu reset context pointer
+ *
+ * Attempt to reset the GPU if it has hung (all asics).
+ * Attempt to do soft-reset or full-reset and reinitialize Asic
+ * Returns 0 for success or an error on failure.
+ */
+
+int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
+			      struct amdgpu_job *job,
+			      struct amdgpu_reset_context *reset_context)
+{
+	struct list_head device_list;
+	bool job_signaled = false;
+	struct amdgpu_hive_info *hive = NULL;
+	int r = 0;
+	bool need_emergency_restart = false;
+
+	/*
+	 * If it reaches here because of hang/timeout and a RAS error is
+	 * detected at the same time, let RAS recovery take care of it.
+	 */
+	if (amdgpu_ras_is_err_state(adev, AMDGPU_RAS_BLOCK__ANY) &&
+	    !amdgpu_sriov_vf(adev) &&
+	    reset_context->src != AMDGPU_RESET_SRC_RAS) {
+		dev_dbg(adev->dev,
+			"Gpu recovery from source: %d yielding to RAS error recovery handling",
+			reset_context->src);
+		return 0;
+	}
+
+	/*
+	 * Special case: RAS triggered and full reset isn't supported
+	 */
+	need_emergency_restart = amdgpu_ras_need_emergency_restart(adev);
+
+	/*
+	 * Flush RAM to disk so that after reboot
+	 * the user can read log and see why the system rebooted.
+	 */
+	if (need_emergency_restart && amdgpu_ras_get_context(adev) &&
+		amdgpu_ras_get_context(adev)->reboot) {
+		DRM_WARN("Emergency reboot.");
+
+		ksys_sync_helper();
+		emergency_restart();
+	}
+
+	dev_info(adev->dev, "GPU %s begin!\n",
+		need_emergency_restart ? "jobs stop":"reset");
+
+	if (!amdgpu_sriov_vf(adev))
+		hive = amdgpu_get_xgmi_hive(adev);
+	if (hive)
+		mutex_lock(&hive->hive_lock);
+
+	reset_context->job = job;
+	reset_context->hive = hive;
+	INIT_LIST_HEAD(&device_list);
+
+	r = amdgpu_device_halt_activities(adev, job, reset_context, &device_list,
+					 hive, need_emergency_restart);
+	if (r)
+		goto end_reset;
+
+	if (need_emergency_restart)
+		goto skip_sched_resume;
+	/*
+	 * Must check guilty signal here since after this point all old
+	 * HW fences are force signaled.
+	 *
+	 * job->base holds a reference to parent fence
+	 */
+	if (job && dma_fence_is_signaled(&job->hw_fence)) {
+		job_signaled = true;
+		dev_info(adev->dev, "Guilty job already signaled, skipping HW reset");
+		goto skip_hw_reset;
+	}
+
+	r = amdgpu_device_asic_reset(adev, &device_list, reset_context);
+	if (r)
+		goto end_reset;
+skip_hw_reset:
+	r = amdgpu_device_sched_resume(&device_list, reset_context, job_signaled);
+	if (r)
+		goto end_reset;
+skip_sched_resume:
+	amdgpu_device_gpu_resume(adev, &device_list, need_emergency_restart);
 end_reset:
 	if (hive) {
 		mutex_unlock(&hive->hive_lock);
@@ -6580,12 +6734,15 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct amdgpu_device *adev = drm_to_adev(dev);
-	int i;
+	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
+	struct amdgpu_reset_context reset_context;
+	struct list_head device_list;
+	int r = 0;
 
-	DRM_INFO("PCI error: detected callback, state(%d)!!\n", state);
+	dev_info(adev->dev, "PCI error: detected callback!!\n");
 
-	if (adev->gmc.xgmi.num_physical_nodes > 1) {
-		DRM_WARN("No support for XGMI hive yet...");
+	if (!amdgpu_dpm_is_link_reset_supported(adev)) {
+		dev_warn(adev->dev, "No support for XGMI hive yet...\n");
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
 
@@ -6593,32 +6750,30 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 
 	switch (state) {
 	case pci_channel_io_normal:
+		dev_info(adev->dev, "pci_channel_io_normal: state(%d)!!\n", state);
 		return PCI_ERS_RESULT_CAN_RECOVER;
-	/* Fatal error, prepare for slot reset */
 	case pci_channel_io_frozen:
-		/*
-		 * Locking adev->reset_domain->sem will prevent any external access
-		 * to GPU during PCI error recovery
-		 */
-		amdgpu_device_lock_reset_domain(adev->reset_domain);
-		amdgpu_device_set_mp1_state(adev);
-
-		/*
-		 * Block any work scheduling as we do for regular GPU reset
-		 * for the duration of the recovery
-		 */
-		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-			struct amdgpu_ring *ring = adev->rings[i];
-
-			if (!amdgpu_ring_sched_ready(ring))
-				continue;
-
-			drm_sched_stop(&ring->sched, NULL);
+		/* Fatal error, prepare for slot reset */
+		dev_info(adev->dev, "pci_channel_io_frozen: state(%d)!!\n", state);
+
+		if (hive)
+			mutex_lock(&hive->hive_lock);
+		adev->pcie_reset_ctx.occurs_dpc = true;
+		memset(&reset_context, 0, sizeof(reset_context));
+		INIT_LIST_HEAD(&device_list);
+
+		r = amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list,
+					 hive, false);
+		if (hive) {
+			mutex_unlock(&hive->hive_lock);
+			amdgpu_put_xgmi_hive(hive);
 		}
-		atomic_inc(&adev->gpu_reset_counter);
+		if (r)
+			return PCI_ERS_RESULT_DISCONNECT;
 		return PCI_ERS_RESULT_NEED_RESET;
 	case pci_channel_io_perm_failure:
 		/* Permanent error, prepare for device removal */
+		dev_info(adev->dev, "pci_channel_io_perm_failure: state(%d)!!\n", state);
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
 
@@ -6631,8 +6786,10 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
  */
 pci_ers_result_t amdgpu_pci_mmio_enabled(struct pci_dev *pdev)
 {
+	struct drm_device *dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
-	DRM_INFO("PCI error: mmio enabled callback!!\n");
+	dev_info(adev->dev, "PCI error: mmio enabled callback!!\n");
 
 	/* TODO - dump whatever for debugging purposes */
 
@@ -6656,10 +6813,12 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct amdgpu_device *adev = drm_to_adev(dev);
-	int r, i;
 	struct amdgpu_reset_context reset_context;
-	u32 memsize;
+	struct amdgpu_device *tmp_adev;
+	struct amdgpu_hive_info *hive;
 	struct list_head device_list;
+	int r = 0, i;
+	u32 memsize;
 
 	/* PCI error slot reset should be skipped During RAS recovery */
 	if ((amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) ||
@@ -6667,15 +6826,12 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
 	    amdgpu_ras_in_recovery(adev))
 		return PCI_ERS_RESULT_RECOVERED;
 
-	DRM_INFO("PCI error: slot reset callback!!\n");
+	dev_info(adev->dev, "PCI error: slot reset callback!!\n");
 
 	memset(&reset_context, 0, sizeof(reset_context));
 
-	INIT_LIST_HEAD(&device_list);
-	list_add_tail(&adev->reset_list, &device_list);
-
 	/* wait for asic to come out of reset */
-	msleep(500);
+	msleep(700);
 
 	/* Restore PCI confspace */
 	amdgpu_device_load_pci_state(pdev);
@@ -6696,26 +6852,40 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
 	reset_context.method = AMD_RESET_METHOD_NONE;
 	reset_context.reset_req_dev = adev;
 	set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags);
-	set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags);
-
-	adev->no_hw_access = true;
-	r = amdgpu_device_pre_asic_reset(adev, &reset_context);
-	adev->no_hw_access = false;
-	if (r)
-		goto out;
+	set_bit(AMDGPU_SKIP_COREDUMP, &reset_context.flags);
+	INIT_LIST_HEAD(&device_list);
 
-	r = amdgpu_do_asic_reset(&device_list, &reset_context);
+	hive = amdgpu_get_xgmi_hive(adev);
+	if (hive) {
+		mutex_lock(&hive->hive_lock);
+		reset_context.hive = hive;
+		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+			tmp_adev->pcie_reset_ctx.in_link_reset = true;
+			list_add_tail(&tmp_adev->reset_list, &device_list);
+		}
+	} else {
+		set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags);
+		list_add_tail(&adev->reset_list, &device_list);
+	}
 
+	r = amdgpu_device_asic_reset(adev, &device_list, &reset_context);
 out:
 	if (!r) {
 		if (amdgpu_device_cache_pci_state(adev->pdev))
 			pci_restore_state(adev->pdev);
-
-		DRM_INFO("PCIe error recovery succeeded\n");
+		dev_info(adev->dev, "PCIe error recovery succeeded\n");
 	} else {
-		DRM_ERROR("PCIe error recovery failed, err:%d", r);
-		amdgpu_device_unset_mp1_state(adev);
-		amdgpu_device_unlock_reset_domain(adev->reset_domain);
+		dev_err(adev->dev, "PCIe error recovery failed, err:%d\n", r);
+		if (hive) {
+			list_for_each_entry(tmp_adev, &device_list, reset_list)
+				amdgpu_device_unset_mp1_state(tmp_adev);
+			amdgpu_device_unlock_reset_domain(adev->reset_domain);
+		}
+	}
+
+	if (hive) {
+		mutex_unlock(&hive->hive_lock);
+		amdgpu_put_xgmi_hive(hive);
 	}
 
 	return r ? PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_RECOVERED;
@@ -6732,26 +6902,36 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct amdgpu_device *adev = drm_to_adev(dev);
-	int i;
-
+	struct list_head device_list;
+	struct amdgpu_hive_info *hive = NULL;
+	struct amdgpu_device *tmp_adev = NULL;
 
-	DRM_INFO("PCI error: resume callback!!\n");
+	dev_info(adev->dev, "PCI error: resume callback!!\n");
 
 	/* Only continue execution for the case of pci_channel_io_frozen */
 	if (adev->pci_channel_state != pci_channel_io_frozen)
 		return;
 
-	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		struct amdgpu_ring *ring = adev->rings[i];
+	INIT_LIST_HEAD(&device_list);
 
-		if (!amdgpu_ring_sched_ready(ring))
-			continue;
+	hive = amdgpu_get_xgmi_hive(adev);
+	if (hive) {
+		mutex_lock(&hive->hive_lock);
+		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+			tmp_adev->pcie_reset_ctx.in_link_reset = false;
+			list_add_tail(&tmp_adev->reset_list, &device_list);
+		}
+	} else
+		list_add_tail(&adev->reset_list, &device_list);
 
-		drm_sched_start(&ring->sched, 0);
-	}
+	amdgpu_device_sched_resume(&device_list, NULL, NULL);
+	amdgpu_device_gpu_resume(adev, &device_list, false);
+	adev->pcie_reset_ctx.occurs_dpc = false;
 
-	amdgpu_device_unset_mp1_state(adev);
-	amdgpu_device_unlock_reset_domain(adev->reset_domain);
+	if (hive) {
+		mutex_unlock(&hive->hive_lock);
+		amdgpu_put_xgmi_hive(hive);
+	}
 }
 
 bool amdgpu_device_cache_pci_state(struct pci_dev *pdev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index dc2713ec95a5..9e738fae2b74 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -120,6 +120,8 @@ MODULE_FIRMWARE("amdgpu/vega20_ip_discovery.bin");
 MODULE_FIRMWARE("amdgpu/raven_ip_discovery.bin");
 MODULE_FIRMWARE("amdgpu/raven2_ip_discovery.bin");
 MODULE_FIRMWARE("amdgpu/picasso_ip_discovery.bin");
+MODULE_FIRMWARE("amdgpu/arcturus_ip_discovery.bin");
+MODULE_FIRMWARE("amdgpu/aldebaran_ip_discovery.bin");
 
 #define mmIP_DISCOVERY_VERSION  0x16A00
 #define mmRCC_CONFIG_MEMSIZE	0xde3
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index 9f627caedc3f..44e120f9f764 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -43,6 +43,29 @@
 #include <linux/dma-fence-array.h>
 #include <linux/pci-p2pdma.h>
 
+static const struct dma_buf_attach_ops amdgpu_dma_buf_attach_ops;
+
+/**
+ * dma_buf_attach_adev - Helper to get adev of an attachment
+ *
+ * @attach: attachment
+ *
+ * Returns:
+ * A struct amdgpu_device * if the attaching device is an amdgpu device or
+ * partition, NULL otherwise.
+ */
+static struct amdgpu_device *dma_buf_attach_adev(struct dma_buf_attachment *attach)
+{
+	if (attach->importer_ops == &amdgpu_dma_buf_attach_ops) {
+		struct drm_gem_object *obj = attach->importer_priv;
+		struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+
+		return amdgpu_ttm_adev(bo->tbo.bdev);
+	}
+
+	return NULL;
+}
+
 /**
  * amdgpu_dma_buf_attach - &dma_buf_ops.attach implementation
  *
@@ -54,11 +77,13 @@
 static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf,
 				 struct dma_buf_attachment *attach)
 {
+	struct amdgpu_device *attach_adev = dma_buf_attach_adev(attach);
 	struct drm_gem_object *obj = dmabuf->priv;
 	struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 
-	if (pci_p2pdma_distance(adev->pdev, attach->dev, false) < 0)
+	if (!amdgpu_dmabuf_is_xgmi_accessible(attach_adev, bo) &&
+	    pci_p2pdma_distance(adev->pdev, attach->dev, false) < 0)
 		attach->peer2peer = false;
 
 	amdgpu_vm_bo_update_shared(bo);
@@ -75,11 +100,35 @@ static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf,
  */
 static int amdgpu_dma_buf_pin(struct dma_buf_attachment *attach)
 {
-	struct drm_gem_object *obj = attach->dmabuf->priv;
-	struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+	struct dma_buf *dmabuf = attach->dmabuf;
+	struct amdgpu_bo *bo = gem_to_amdgpu_bo(dmabuf->priv);
+	u32 domains = bo->allowed_domains;
+
+	dma_resv_assert_held(dmabuf->resv);
+
+	/* Try pinning into VRAM to allow P2P with RDMA NICs without ODP
+	 * support if all attachments can do P2P. If any attachment can't do
+	 * P2P just pin into GTT instead.
+	 *
+	 * To avoid with conflicting pinnings between GPUs and RDMA when move
+	 * notifiers are disabled, only allow pinning in VRAM when move
+	 * notiers are enabled.
+	 */
+	if (!IS_ENABLED(CONFIG_DMABUF_MOVE_NOTIFY)) {
+		domains &= ~AMDGPU_GEM_DOMAIN_VRAM;
+	} else {
+		list_for_each_entry(attach, &dmabuf->attachments, node)
+			if (!attach->peer2peer)
+				domains &= ~AMDGPU_GEM_DOMAIN_VRAM;
+	}
 
-	/* pin buffer into GTT */
-	return amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
+	if (domains & AMDGPU_GEM_DOMAIN_VRAM)
+		bo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
+
+	if (WARN_ON(!domains))
+		return -EINVAL;
+
+	return amdgpu_bo_pin(bo, domains);
 }
 
 /**
@@ -134,9 +183,6 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach,
 		r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 		if (r)
 			return ERR_PTR(r);
-
-	} else if (bo->tbo.resource->mem_type != TTM_PL_TT) {
-		return ERR_PTR(-EBUSY);
 	}
 
 	switch (bo->tbo.resource->mem_type) {
@@ -153,6 +199,11 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach,
 		break;
 
 	case TTM_PL_VRAM:
+		/* XGMI-accessible memory should never be DMA-mapped */
+		if (WARN_ON(amdgpu_dmabuf_is_xgmi_accessible(
+				dma_buf_attach_adev(attach), bo)))
+			return ERR_PTR(-EINVAL);
+
 		r = amdgpu_vram_mgr_alloc_sgt(adev, bo->tbo.resource, 0,
 					      bo->tbo.base.size, attach->dev,
 					      dir, &sgt);
@@ -184,7 +235,7 @@ static void amdgpu_dma_buf_unmap(struct dma_buf_attachment *attach,
 				 struct sg_table *sgt,
 				 enum dma_data_direction dir)
 {
-	if (sgt->sgl->page_link) {
+	if (sg_page(sgt->sgl)) {
 		dma_unmap_sgtable(attach->dev, sgt, dir, 0);
 		sg_free_table(sgt);
 		kfree(sgt);
@@ -459,6 +510,9 @@ bool amdgpu_dmabuf_is_xgmi_accessible(struct amdgpu_device *adev,
 	struct drm_gem_object *obj = &bo->tbo.base;
 	struct drm_gem_object *gobj;
 
+	if (!adev)
+		return false;
+
 	if (obj->import_attach) {
 		struct dma_buf *dma_buf = obj->import_attach->dmabuf;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 26bf896f1444..4ddd08ce8885 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -51,6 +51,8 @@
 #include "amdgpu_reset.h"
 #include "amdgpu_sched.h"
 #include "amdgpu_xgmi.h"
+#include "amdgpu_userq.h"
+#include "amdgpu_userq_fence.h"
 #include "../amdxcp/amdgpu_xcp_drv.h"
 
 /*
@@ -123,9 +125,10 @@
  * - 3.61.0 - Contains fix for RV/PCO compute queues
  * - 3.62.0 - Add AMDGPU_IDS_FLAGS_MODE_PF, AMDGPU_IDS_FLAGS_MODE_VF & AMDGPU_IDS_FLAGS_MODE_PT
  * - 3.63.0 - GFX12 display DCC supports 256B max compressed block size
+ * - 3.64.0 - Userq IP support query
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	63
+#define KMS_DRIVER_MINOR	64
 #define KMS_DRIVER_PATCHLEVEL	0
 
 /*
@@ -140,6 +143,7 @@ enum AMDGPU_DEBUG_MASK {
 	AMDGPU_DEBUG_ENABLE_EXP_RESETS = BIT(5),
 	AMDGPU_DEBUG_DISABLE_GPU_RING_RESET = BIT(6),
 	AMDGPU_DEBUG_SMU_POOL = BIT(7),
+	AMDGPU_DEBUG_VM_USERPTR = BIT(8),
 };
 
 unsigned int amdgpu_vram_limit = UINT_MAX;
@@ -176,7 +180,7 @@ uint amdgpu_pg_mask = 0xffffffff;
 uint amdgpu_sdma_phase_quantum = 32;
 char *amdgpu_disable_cu;
 char *amdgpu_virtual_display;
-bool enforce_isolation;
+int amdgpu_enforce_isolation = -1;
 int amdgpu_modeset = -1;
 
 /* Specifies the default granularity for SVM, used in buffer
@@ -238,6 +242,8 @@ int amdgpu_agp = -1; /* auto */
 int amdgpu_wbrf = -1;
 int amdgpu_damage_clips = -1; /* auto */
 int amdgpu_umsch_mm_fwlog;
+int amdgpu_rebar = -1; /* auto */
+int amdgpu_user_queue = -1;
 
 DECLARE_DYNDBG_CLASSMAP(drm_debug_classes, DD_CLASS_TYPE_DISJOINT_BITS, 0,
 			"DRM_UT_CORE",
@@ -1033,11 +1039,13 @@ module_param_named(user_partt_mode, amdgpu_user_partt_mode, uint, 0444);
 
 
 /**
- * DOC: enforce_isolation (bool)
- * enforce process isolation between graphics and compute via using the same reserved vmid.
+ * DOC: enforce_isolation (int)
+ * enforce process isolation between graphics and compute.
+ * (-1 = auto, 0 = disable, 1 = enable, 2 = enable legacy mode, 3 = enable without cleaner shader)
  */
-module_param(enforce_isolation, bool, 0444);
-MODULE_PARM_DESC(enforce_isolation, "enforce process isolation between graphics and compute . enforce_isolation = on");
+module_param_named(enforce_isolation, amdgpu_enforce_isolation, int, 0444);
+MODULE_PARM_DESC(enforce_isolation,
+"enforce process isolation between graphics and compute. (-1 = auto, 0 = disable, 1 = enable, 2 = enable legacy mode, 3 = enable without cleaner shader)");
 
 /**
  * DOC: modeset (int)
@@ -1096,6 +1104,28 @@ MODULE_PARM_DESC(wbrf,
 	"Enable Wifi RFI interference mitigation (0 = disabled, 1 = enabled, -1 = auto(default)");
 module_param_named(wbrf, amdgpu_wbrf, int, 0444);
 
+/**
+ * DOC: rebar (int)
+ * Allow BAR resizing.  Disable this to prevent the driver from attempting
+ * to resize the BAR if the GPU supports it and there is available MMIO space.
+ * Note that this just prevents the driver from resizing the BAR.  The BIOS
+ * may have already resized the BAR at boot time.
+ */
+MODULE_PARM_DESC(rebar, "Resizable BAR (-1 = auto (default), 0 = disable, 1 = enable)");
+module_param_named(rebar, amdgpu_rebar, int, 0444);
+
+/**
+ * DOC: user_queue (int)
+ * Enable user queues on systems that support user queues. Possible values:
+ *
+ * - -1 = auto (ASIC specific default)
+ * -  0 = user queues disabled
+ * -  1 = user queues enabled and kernel queues enabled (if supported)
+ * -  2 = user queues enabled and kernel queues disabled
+ */
+MODULE_PARM_DESC(user_queue, "Enable user queues (-1 = auto (default), 0 = disable, 1 = enable, 2 = enable UQs and disable KQs)");
+module_param_named(user_queue, amdgpu_user_queue, int, 0444);
+
 /* These devices are not supported by amdgpu.
  * They are supported by the mach64, r128, radeon drivers
  */
@@ -2244,6 +2274,10 @@ static void amdgpu_init_debug_options(struct amdgpu_device *adev)
 		pr_info("debug: use vram for smu pool\n");
 		adev->pm.smu_debug_mask |= SMU_DEBUG_POOL_USE_VRAM;
 	}
+	if (amdgpu_debug_mask & AMDGPU_DEBUG_VM_USERPTR) {
+		pr_info("debug: VM mode debug for userptr is enabled\n");
+		adev->debug_vm_userptr = true;
+	}
 }
 
 static unsigned long amdgpu_fix_asic_type(struct pci_dev *pdev, unsigned long flags)
@@ -2548,8 +2582,20 @@ static int amdgpu_pmops_suspend(struct device *dev)
 		adev->in_s0ix = true;
 	else if (amdgpu_acpi_is_s3_active(adev))
 		adev->in_s3 = true;
-	if (!adev->in_s0ix && !adev->in_s3)
+	if (!adev->in_s0ix && !adev->in_s3) {
+		/* don't allow going deep first time followed by s2idle the next time */
+		if (adev->last_suspend_state != PM_SUSPEND_ON &&
+		    adev->last_suspend_state != pm_suspend_target_state) {
+			drm_err_once(drm_dev, "Unsupported suspend state %d\n",
+				     pm_suspend_target_state);
+			return -EINVAL;
+		}
 		return 0;
+	}
+
+	/* cache the state last used for suspend */
+	adev->last_suspend_state = pm_suspend_target_state;
+
 	return amdgpu_device_suspend(drm_dev, true);
 }
 
@@ -2603,13 +2649,8 @@ static int amdgpu_pmops_freeze(struct device *dev)
 static int amdgpu_pmops_thaw(struct device *dev)
 {
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = drm_to_adev(drm_dev);
-	int r;
-
-	r = amdgpu_device_resume(drm_dev, true);
-	adev->in_s4 = false;
 
-	return r;
+	return amdgpu_device_resume(drm_dev, true);
 }
 
 static int amdgpu_pmops_poweroff(struct device *dev)
@@ -2622,9 +2663,6 @@ static int amdgpu_pmops_poweroff(struct device *dev)
 static int amdgpu_pmops_restore(struct device *dev)
 {
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = drm_to_adev(drm_dev);
-
-	adev->in_s4 = false;
 
 	return amdgpu_device_resume(drm_dev, true);
 }
@@ -2696,6 +2734,29 @@ static int amdgpu_runtime_idle_check_display(struct device *dev)
 	return 0;
 }
 
+static int amdgpu_runtime_idle_check_userq(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct drm_device *drm_dev = pci_get_drvdata(pdev);
+	struct amdgpu_device *adev = drm_to_adev(drm_dev);
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	int queue_id;
+	int ret = 0;
+
+	mutex_lock(&adev->userq_mutex);
+	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
+			ret = -EBUSY;
+			goto done;
+		}
+	}
+done:
+	mutex_unlock(&adev->userq_mutex);
+
+	return ret;
+}
+
 static int amdgpu_pmops_runtime_suspend(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
@@ -2711,6 +2772,9 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev)
 	ret = amdgpu_runtime_idle_check_display(dev);
 	if (ret)
 		return ret;
+	ret = amdgpu_runtime_idle_check_userq(dev);
+	if (ret)
+		return ret;
 
 	/* wait for all rings to drain before suspending */
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
@@ -2832,12 +2896,30 @@ static int amdgpu_pmops_runtime_idle(struct device *dev)
 	}
 
 	ret = amdgpu_runtime_idle_check_display(dev);
+	if (ret)
+		goto done;
 
+	ret = amdgpu_runtime_idle_check_userq(dev);
+done:
 	pm_runtime_mark_last_busy(dev);
 	pm_runtime_autosuspend(dev);
 	return ret;
 }
 
+static int amdgpu_drm_release(struct inode *inode, struct file *filp)
+{
+	struct drm_file *file_priv = filp->private_data;
+	struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
+
+	if (fpriv) {
+		fpriv->evf_mgr.fd_closing = true;
+		amdgpu_userq_mgr_fini(&fpriv->userq_mgr);
+		amdgpu_eviction_fence_destroy(&fpriv->evf_mgr);
+	}
+
+	return drm_release(inode, filp);
+}
+
 long amdgpu_drm_ioctl(struct file *filp,
 		      unsigned int cmd, unsigned long arg)
 {
@@ -2889,7 +2971,7 @@ static const struct file_operations amdgpu_driver_kms_fops = {
 	.owner = THIS_MODULE,
 	.open = drm_open,
 	.flush = amdgpu_flush,
-	.release = drm_release,
+	.release = amdgpu_drm_release,
 	.unlocked_ioctl = amdgpu_drm_ioctl,
 	.mmap = drm_gem_mmap,
 	.poll = drm_poll,
@@ -2936,6 +3018,9 @@ const struct drm_ioctl_desc amdgpu_ioctls_kms[] = {
 	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_VA, amdgpu_gem_va_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_OP, amdgpu_gem_op_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_USERPTR, amdgpu_gem_userptr_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_USERQ, amdgpu_userq_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_USERQ_SIGNAL, amdgpu_userq_signal_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_USERQ_WAIT, amdgpu_userq_wait_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 
 static const struct drm_driver amdgpu_kms_driver = {
@@ -3026,6 +3111,10 @@ static int __init amdgpu_init(void)
 	if (r)
 		goto error_fence;
 
+	r = amdgpu_userq_fence_slab_init();
+	if (r)
+		goto error_fence;
+
 	DRM_INFO("amdgpu kernel modesetting enabled.\n");
 	amdgpu_register_atpx_handler();
 	amdgpu_acpi_detect();
@@ -3057,6 +3146,7 @@ static void __exit amdgpu_exit(void)
 	amdgpu_acpi_release();
 	amdgpu_sync_fini();
 	amdgpu_fence_slab_fini();
+	amdgpu_userq_fence_slab_fini();
 	mmu_notifier_synchronize();
 	amdgpu_xcp_drv_release();
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c
new file mode 100644
index 000000000000..73b629b5f56f
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c
@@ -0,0 +1,233 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2024 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include <linux/sched.h>
+#include <drm/drm_exec.h>
+#include "amdgpu.h"
+
+#define work_to_evf_mgr(w, name) container_of(w, struct amdgpu_eviction_fence_mgr, name)
+#define evf_mgr_to_fpriv(e) container_of(e, struct amdgpu_fpriv, evf_mgr)
+
+static const char *
+amdgpu_eviction_fence_get_driver_name(struct dma_fence *fence)
+{
+	return "amdgpu_eviction_fence";
+}
+
+static const char *
+amdgpu_eviction_fence_get_timeline_name(struct dma_fence *f)
+{
+	struct amdgpu_eviction_fence *ef;
+
+	ef = container_of(f, struct amdgpu_eviction_fence, base);
+	return ef->timeline_name;
+}
+
+int
+amdgpu_eviction_fence_replace_fence(struct amdgpu_eviction_fence_mgr *evf_mgr,
+				    struct drm_exec *exec)
+{
+	struct amdgpu_eviction_fence *old_ef, *new_ef;
+	struct drm_gem_object *obj;
+	unsigned long index;
+	int ret;
+
+	if (evf_mgr->ev_fence &&
+	    !dma_fence_is_signaled(&evf_mgr->ev_fence->base))
+		return 0;
+	/*
+	 * Steps to replace eviction fence:
+	 * * lock all objects in exec (caller)
+	 * * create a new eviction fence
+	 * * update new eviction fence in evf_mgr
+	 * * attach the new eviction fence to BOs
+	 * * release the old fence
+	 * * unlock the objects (caller)
+	 */
+	new_ef = amdgpu_eviction_fence_create(evf_mgr);
+	if (!new_ef) {
+		DRM_ERROR("Failed to create new eviction fence\n");
+		return -ENOMEM;
+	}
+
+	/* Update the eviction fence now */
+	spin_lock(&evf_mgr->ev_fence_lock);
+	old_ef = evf_mgr->ev_fence;
+	evf_mgr->ev_fence = new_ef;
+	spin_unlock(&evf_mgr->ev_fence_lock);
+
+	/* Attach the new fence */
+	drm_exec_for_each_locked_object(exec, index, obj) {
+		struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+
+		if (!bo)
+			continue;
+		ret = amdgpu_eviction_fence_attach(evf_mgr, bo);
+		if (ret) {
+			DRM_ERROR("Failed to attch new eviction fence\n");
+			goto free_err;
+		}
+	}
+
+	/* Free old fence */
+	if (old_ef)
+		dma_fence_put(&old_ef->base);
+	return 0;
+
+free_err:
+	kfree(new_ef);
+	return ret;
+}
+
+static void
+amdgpu_eviction_fence_suspend_worker(struct work_struct *work)
+{
+	struct amdgpu_eviction_fence_mgr *evf_mgr = work_to_evf_mgr(work, suspend_work.work);
+	struct amdgpu_fpriv *fpriv = evf_mgr_to_fpriv(evf_mgr);
+	struct amdgpu_userq_mgr *uq_mgr = &fpriv->userq_mgr;
+	struct amdgpu_eviction_fence *ev_fence;
+
+	mutex_lock(&uq_mgr->userq_mutex);
+	ev_fence = evf_mgr->ev_fence;
+	if (!ev_fence)
+		goto unlock;
+
+	amdgpu_userq_evict(uq_mgr, ev_fence);
+
+unlock:
+	mutex_unlock(&uq_mgr->userq_mutex);
+}
+
+static bool amdgpu_eviction_fence_enable_signaling(struct dma_fence *f)
+{
+	struct amdgpu_eviction_fence_mgr *evf_mgr;
+	struct amdgpu_eviction_fence *ev_fence;
+
+	if (!f)
+		return true;
+
+	ev_fence = to_ev_fence(f);
+	evf_mgr = ev_fence->evf_mgr;
+
+	schedule_delayed_work(&evf_mgr->suspend_work, 0);
+	return true;
+}
+
+static const struct dma_fence_ops amdgpu_eviction_fence_ops = {
+	.use_64bit_seqno = true,
+	.get_driver_name = amdgpu_eviction_fence_get_driver_name,
+	.get_timeline_name = amdgpu_eviction_fence_get_timeline_name,
+	.enable_signaling = amdgpu_eviction_fence_enable_signaling,
+};
+
+void amdgpu_eviction_fence_signal(struct amdgpu_eviction_fence_mgr *evf_mgr,
+				  struct amdgpu_eviction_fence *ev_fence)
+{
+	spin_lock(&evf_mgr->ev_fence_lock);
+	dma_fence_signal(&ev_fence->base);
+	spin_unlock(&evf_mgr->ev_fence_lock);
+}
+
+struct amdgpu_eviction_fence *
+amdgpu_eviction_fence_create(struct amdgpu_eviction_fence_mgr *evf_mgr)
+{
+	struct amdgpu_eviction_fence *ev_fence;
+
+	ev_fence = kzalloc(sizeof(*ev_fence), GFP_KERNEL);
+	if (!ev_fence)
+		return NULL;
+
+	ev_fence->evf_mgr = evf_mgr;
+	get_task_comm(ev_fence->timeline_name, current);
+	spin_lock_init(&ev_fence->lock);
+	dma_fence_init(&ev_fence->base, &amdgpu_eviction_fence_ops,
+		       &ev_fence->lock, evf_mgr->ev_fence_ctx,
+		       atomic_inc_return(&evf_mgr->ev_fence_seq));
+	return ev_fence;
+}
+
+void amdgpu_eviction_fence_destroy(struct amdgpu_eviction_fence_mgr *evf_mgr)
+{
+	struct amdgpu_eviction_fence *ev_fence;
+
+	/* Wait for any pending work to execute */
+	flush_delayed_work(&evf_mgr->suspend_work);
+
+	spin_lock(&evf_mgr->ev_fence_lock);
+	ev_fence = evf_mgr->ev_fence;
+	spin_unlock(&evf_mgr->ev_fence_lock);
+
+	if (!ev_fence)
+		return;
+
+	dma_fence_wait(&ev_fence->base, false);
+
+	/* Last unref of ev_fence */
+	dma_fence_put(&ev_fence->base);
+}
+
+int amdgpu_eviction_fence_attach(struct amdgpu_eviction_fence_mgr *evf_mgr,
+				 struct amdgpu_bo *bo)
+{
+	struct amdgpu_eviction_fence *ev_fence;
+	struct dma_resv *resv = bo->tbo.base.resv;
+	int ret;
+
+	if (!resv)
+		return 0;
+
+	ret = dma_resv_reserve_fences(resv, 1);
+	if (ret) {
+		DRM_DEBUG_DRIVER("Failed to resv fence space\n");
+		return ret;
+	}
+
+	spin_lock(&evf_mgr->ev_fence_lock);
+	ev_fence = evf_mgr->ev_fence;
+	if (ev_fence)
+		dma_resv_add_fence(resv, &ev_fence->base, DMA_RESV_USAGE_BOOKKEEP);
+	spin_unlock(&evf_mgr->ev_fence_lock);
+
+	return 0;
+}
+
+void amdgpu_eviction_fence_detach(struct amdgpu_eviction_fence_mgr *evf_mgr,
+				  struct amdgpu_bo *bo)
+{
+	struct dma_fence *stub = dma_fence_get_stub();
+
+	dma_resv_replace_fences(bo->tbo.base.resv, evf_mgr->ev_fence_ctx,
+				stub, DMA_RESV_USAGE_BOOKKEEP);
+	dma_fence_put(stub);
+}
+
+int amdgpu_eviction_fence_init(struct amdgpu_eviction_fence_mgr *evf_mgr)
+{
+	/* This needs to be done one time per open */
+	atomic_set(&evf_mgr->ev_fence_seq, 0);
+	evf_mgr->ev_fence_ctx = dma_fence_context_alloc(1);
+	spin_lock_init(&evf_mgr->ev_fence_lock);
+
+	INIT_DELAYED_WORK(&evf_mgr->suspend_work, amdgpu_eviction_fence_suspend_worker);
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.h
new file mode 100644
index 000000000000..fcd867b7147d
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.h
@@ -0,0 +1,69 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef AMDGPU_EV_FENCE_H_
+#define AMDGPU_EV_FENCE_H_
+
+struct amdgpu_eviction_fence {
+	struct dma_fence base;
+	spinlock_t	 lock;
+	char		 timeline_name[TASK_COMM_LEN];
+	struct amdgpu_eviction_fence_mgr *evf_mgr;
+};
+
+struct amdgpu_eviction_fence_mgr {
+	u64			ev_fence_ctx;
+	atomic_t		ev_fence_seq;
+	spinlock_t		ev_fence_lock;
+	struct amdgpu_eviction_fence *ev_fence;
+	struct delayed_work	suspend_work;
+	uint8_t fd_closing;
+};
+
+/* Eviction fence helper functions */
+struct amdgpu_eviction_fence *
+amdgpu_eviction_fence_create(struct amdgpu_eviction_fence_mgr *evf_mgr);
+
+void
+amdgpu_eviction_fence_destroy(struct amdgpu_eviction_fence_mgr *evf_mgr);
+
+int
+amdgpu_eviction_fence_attach(struct amdgpu_eviction_fence_mgr *evf_mgr,
+			     struct amdgpu_bo *bo);
+
+void
+amdgpu_eviction_fence_detach(struct amdgpu_eviction_fence_mgr *evf_mgr,
+			     struct amdgpu_bo *bo);
+
+int
+amdgpu_eviction_fence_init(struct amdgpu_eviction_fence_mgr *evf_mgr);
+
+void
+amdgpu_eviction_fence_signal(struct amdgpu_eviction_fence_mgr *evf_mgr,
+			     struct amdgpu_eviction_fence *ev_fence);
+
+int
+amdgpu_eviction_fence_replace_fence(struct amdgpu_eviction_fence_mgr *evf_mgr,
+				    struct drm_exec *exec);
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 69429df09477..2c68118fe9fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -36,6 +36,7 @@
 #include <drm/drm_exec.h>
 #include <drm/drm_gem_ttm_helper.h>
 #include <drm/ttm/ttm_tt.h>
+#include <drm/drm_syncobj.h>
 
 #include "amdgpu.h"
 #include "amdgpu_display.h"
@@ -44,6 +45,114 @@
 #include "amdgpu_xgmi.h"
 #include "amdgpu_vm.h"
 
+static int
+amdgpu_gem_add_input_fence(struct drm_file *filp,
+			   uint64_t syncobj_handles_array,
+			   uint32_t num_syncobj_handles)
+{
+	struct dma_fence *fence;
+	uint32_t *syncobj_handles;
+	int ret, i;
+
+	if (!num_syncobj_handles)
+		return 0;
+
+	syncobj_handles = memdup_user(u64_to_user_ptr(syncobj_handles_array),
+				      sizeof(uint32_t) * num_syncobj_handles);
+	if (IS_ERR(syncobj_handles))
+		return PTR_ERR(syncobj_handles);
+
+	for (i = 0; i < num_syncobj_handles; i++) {
+
+		if (!syncobj_handles[i]) {
+			ret = -EINVAL;
+			goto free_memdup;
+		}
+
+		ret = drm_syncobj_find_fence(filp, syncobj_handles[i], 0, 0, &fence);
+		if (ret)
+			goto free_memdup;
+
+		dma_fence_wait(fence, false);
+
+		/* TODO: optimize async handling */
+		dma_fence_put(fence);
+	}
+
+free_memdup:
+	kfree(syncobj_handles);
+	return ret;
+}
+
+static int
+amdgpu_gem_update_timeline_node(struct drm_file *filp,
+				uint32_t syncobj_handle,
+				uint64_t point,
+				struct drm_syncobj **syncobj,
+				struct dma_fence_chain **chain)
+{
+	if (!syncobj_handle)
+		return 0;
+
+	/* Find the sync object */
+	*syncobj = drm_syncobj_find(filp, syncobj_handle);
+	if (!*syncobj)
+		return -ENOENT;
+
+	if (!point)
+		return 0;
+
+	/* Allocate the chain node */
+	*chain = dma_fence_chain_alloc();
+	if (!*chain) {
+		drm_syncobj_put(*syncobj);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void
+amdgpu_gem_update_bo_mapping(struct drm_file *filp,
+			     struct amdgpu_bo_va *bo_va,
+			     uint32_t operation,
+			     uint64_t point,
+			     struct dma_fence *fence,
+			     struct drm_syncobj *syncobj,
+			     struct dma_fence_chain *chain)
+{
+	struct amdgpu_bo *bo = bo_va ? bo_va->base.bo : NULL;
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_vm *vm = &fpriv->vm;
+	struct dma_fence *last_update;
+
+	if (!syncobj)
+		return;
+
+	/* Find the last update fence */
+	switch (operation) {
+	case AMDGPU_VA_OP_MAP:
+	case AMDGPU_VA_OP_REPLACE:
+		if (bo && (bo->tbo.base.resv == vm->root.bo->tbo.base.resv))
+			last_update = vm->last_update;
+		else
+			last_update = bo_va->last_pt_update;
+		break;
+	case AMDGPU_VA_OP_UNMAP:
+	case AMDGPU_VA_OP_CLEAR:
+		last_update = fence;
+		break;
+	default:
+		return;
+	}
+
+	/* Add fence to timeline */
+	if (!point)
+		drm_syncobj_replace_fence(syncobj, last_update);
+	else
+		drm_syncobj_add_point(syncobj, chain, last_update, point);
+}
+
 static vm_fault_t amdgpu_gem_fault(struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *bo = vmf->vma->vm_private_data;
@@ -184,6 +293,15 @@ static int amdgpu_gem_object_open(struct drm_gem_object *obj,
 		bo_va = amdgpu_vm_bo_add(adev, vm, abo);
 	else
 		++bo_va->ref_count;
+
+	/* attach gfx eviction fence */
+	r = amdgpu_eviction_fence_attach(&fpriv->evf_mgr, abo);
+	if (r) {
+		DRM_DEBUG_DRIVER("Failed to attach eviction fence to BO\n");
+		amdgpu_bo_unreserve(abo);
+		return r;
+	}
+
 	amdgpu_bo_unreserve(abo);
 
 	/* Validate and add eviction fence to DMABuf imports with dynamic
@@ -247,6 +365,9 @@ static void amdgpu_gem_object_close(struct drm_gem_object *obj,
 			goto out_unlock;
 	}
 
+	if (!amdgpu_vm_is_bo_always_valid(vm, bo))
+		amdgpu_eviction_fence_detach(&fpriv->evf_mgr, bo);
+
 	bo_va = amdgpu_vm_bo_find(vm, bo);
 	if (!bo_va || --bo_va->ref_count)
 		goto out_unlock;
@@ -321,10 +442,6 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	uint32_t handle, initial_domain;
 	int r;
 
-	/* reject DOORBELLs until userspace code to use it is available */
-	if (args->in.domains & AMDGPU_GEM_DOMAIN_DOORBELL)
-		return -EINVAL;
-
 	/* reject invalid gem flags */
 	if (flags & ~(AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
 		      AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
@@ -638,18 +755,23 @@ out:
  *
  * Update the bo_va directly after setting its address. Errors are not
  * vital here, so they are not reported back to userspace.
+ *
+ * Returns resulting fence if freed BO(s) got cleared from the PT.
+ * otherwise stub fence in case of error.
  */
-static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
-				    struct amdgpu_vm *vm,
-				    struct amdgpu_bo_va *bo_va,
-				    uint32_t operation)
+static struct dma_fence *
+amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
+			struct amdgpu_vm *vm,
+			struct amdgpu_bo_va *bo_va,
+			uint32_t operation)
 {
+	struct dma_fence *fence = dma_fence_get_stub();
 	int r;
 
 	if (!amdgpu_vm_ready(vm))
-		return;
+		return fence;
 
-	r = amdgpu_vm_clear_freed(adev, vm, NULL);
+	r = amdgpu_vm_clear_freed(adev, vm, &fence);
 	if (r)
 		goto error;
 
@@ -665,6 +787,8 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
 error:
 	if (r && r != -ERESTARTSYS)
 		DRM_ERROR("Couldn't update BO_VA (%d)\n", r);
+
+	return fence;
 }
 
 /**
@@ -713,6 +837,9 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	struct amdgpu_bo *abo;
 	struct amdgpu_bo_va *bo_va;
+	struct drm_syncobj *timeline_syncobj = NULL;
+	struct dma_fence_chain *timeline_chain = NULL;
+	struct dma_fence *fence;
 	struct drm_exec exec;
 	uint64_t va_flags;
 	uint64_t vm_size;
@@ -774,6 +901,12 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 		abo = NULL;
 	}
 
+	r = amdgpu_gem_add_input_fence(filp,
+				       args->input_fence_syncobj_handles,
+				       args->num_syncobj_handles);
+	if (r)
+		goto error_put_gobj;
+
 	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT |
 		      DRM_EXEC_IGNORE_DUPLICATES, 0);
 	drm_exec_until_all_locked(&exec) {
@@ -802,6 +935,14 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 		bo_va = NULL;
 	}
 
+	r = amdgpu_gem_update_timeline_node(filp,
+					    args->vm_timeline_syncobj_out,
+					    args->vm_timeline_point,
+					    &timeline_syncobj,
+					    &timeline_chain);
+	if (r)
+		goto error;
+
 	switch (args->operation) {
 	case AMDGPU_VA_OP_MAP:
 		va_flags = amdgpu_gem_va_map_flags(adev, args->flags);
@@ -827,12 +968,24 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	default:
 		break;
 	}
-	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE) && !adev->debug_vm)
-		amdgpu_gem_va_update_vm(adev, &fpriv->vm, bo_va,
-					args->operation);
+	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE) && !adev->debug_vm) {
+		fence = amdgpu_gem_va_update_vm(adev, &fpriv->vm, bo_va,
+						args->operation);
+
+		if (timeline_syncobj)
+			amdgpu_gem_update_bo_mapping(filp, bo_va,
+					     args->operation,
+					     args->vm_timeline_point,
+					     fence, timeline_syncobj,
+					     timeline_chain);
+		else
+			dma_fence_put(fence);
+
+	}
 
 error:
 	drm_exec_fini(&exec);
+error_put_gobj:
 	drm_gem_object_put(gobj);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 72af5e5a894a..1db1e6ec0184 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -33,6 +33,7 @@
 #include "amdgpu_reset.h"
 #include "amdgpu_xcp.h"
 #include "amdgpu_xgmi.h"
+#include "nvd.h"
 
 /* delay 0.1 second to enable gfx off feature */
 #define GFX_OFF_DELAY_ENABLE         msecs_to_jiffies(100)
@@ -74,14 +75,15 @@ bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev,
 			adev->gfx.mec_bitmap[xcc_id].queue_bitmap);
 }
 
-int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev,
-			       int me, int pipe, int queue)
+static int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev,
+				      int me, int pipe, int queue)
 {
+	int num_queue_per_pipe = 1; /* we only enable 1 KGQ per pipe */
 	int bit = 0;
 
 	bit += me * adev->gfx.me.num_pipe_per_me
-		* adev->gfx.me.num_queue_per_pipe;
-	bit += pipe * adev->gfx.me.num_queue_per_pipe;
+		* num_queue_per_pipe;
+	bit += pipe * num_queue_per_pipe;
 	bit += queue;
 
 	return bit;
@@ -238,8 +240,8 @@ void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
 {
 	int i, queue, pipe;
 	bool multipipe_policy = amdgpu_gfx_is_graphics_multipipe_capable(adev);
-	int max_queues_per_me = adev->gfx.me.num_pipe_per_me *
-					adev->gfx.me.num_queue_per_pipe;
+	int num_queue_per_pipe = 1; /* we only enable 1 KGQ per pipe */
+	int max_queues_per_me = adev->gfx.me.num_pipe_per_me * num_queue_per_pipe;
 
 	if (multipipe_policy) {
 		/* policy: amdgpu owns the first queue per pipe at this stage
@@ -247,9 +249,9 @@ void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
 		for (i = 0; i < max_queues_per_me; i++) {
 			pipe = i % adev->gfx.me.num_pipe_per_me;
 			queue = (i / adev->gfx.me.num_pipe_per_me) %
-				adev->gfx.me.num_queue_per_pipe;
+				num_queue_per_pipe;
 
-			set_bit(pipe * adev->gfx.me.num_queue_per_pipe + queue,
+			set_bit(pipe * num_queue_per_pipe + queue,
 				adev->gfx.me.queue_bitmap);
 		}
 	} else {
@@ -258,8 +260,9 @@ void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
 	}
 
 	/* update the number of active graphics rings */
-	adev->gfx.num_gfx_rings =
-		bitmap_weight(adev->gfx.me.queue_bitmap, AMDGPU_MAX_GFX_QUEUES);
+	if (adev->gfx.num_gfx_rings)
+		adev->gfx.num_gfx_rings =
+			bitmap_weight(adev->gfx.me.queue_bitmap, AMDGPU_MAX_GFX_QUEUES);
 }
 
 static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev,
@@ -1351,6 +1354,10 @@ static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
 	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int mode;
 
+	/* Only minimal precaution taken to reject requests while in reset.*/
+	if (amdgpu_in_reset(adev))
+		return -EPERM;
+
 	mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
 					       AMDGPU_XCP_FL_NONE);
 
@@ -1394,8 +1401,14 @@ static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 		return -EINVAL;
 	}
 
+	/* Don't allow a switch while under reset */
+	if (!down_read_trylock(&adev->reset_domain->sem))
+		return -EPERM;
+
 	ret = amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, mode);
 
+	up_read(&adev->reset_domain->sem);
+
 	if (ret)
 		return ret;
 
@@ -1438,9 +1451,11 @@ static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring)
 	struct amdgpu_device *adev = ring->adev;
 	struct drm_gpu_scheduler *sched = &ring->sched;
 	struct drm_sched_entity entity;
+	static atomic_t counter;
 	struct dma_fence *f;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
+	void *owner;
 	int i, r;
 
 	/* Initialize the scheduler entity */
@@ -1451,13 +1466,21 @@ static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring)
 		goto err;
 	}
 
-	r = amdgpu_job_alloc_with_ib(ring->adev, &entity, NULL,
-				     64, 0,
-				     &job);
+	/*
+	 * Use some unique dummy value as the owner to make sure we execute
+	 * the cleaner shader on each submission. The value just need to change
+	 * for each submission and is otherwise meaningless.
+	 */
+	owner = (void *)(unsigned long)atomic_inc_return(&counter);
+
+	r = amdgpu_job_alloc_with_ib(ring->adev, &entity, owner,
+				     64, 0, &job);
 	if (r)
 		goto err;
 
 	job->enforce_isolation = true;
+	/* always run the cleaner shader */
+	job->run_cleaner_shader = true;
 
 	ib = &job->ibs[0];
 	for (i = 0; i <= ring->funcs->align_mask; ++i)
@@ -1544,6 +1567,9 @@ static ssize_t amdgpu_gfx_set_run_cleaner_shader(struct device *dev,
 	if (adev->in_suspend && !adev->in_runpm)
 		return -EPERM;
 
+	if (adev->gfx.disable_kq)
+		return -EPERM;
+
 	ret = kstrtol(buf, 0, &value);
 
 	if (ret)
@@ -1586,7 +1612,8 @@ static ssize_t amdgpu_gfx_set_run_cleaner_shader(struct device *dev,
  * Provides the sysfs read interface to get the current settings of the 'enforce_isolation'
  * feature for each GPU partition. Reading from the 'enforce_isolation'
  * sysfs file returns the isolation settings for all partitions, where '0'
- * indicates disabled and '1' indicates enabled.
+ * indicates disabled, '1' indicates enabled, and '2' indicates enabled in legacy mode,
+ * and '3' indicates enabled without cleaner shader.
  *
  * Return: The number of bytes read from the sysfs file.
  */
@@ -1621,9 +1648,12 @@ static ssize_t amdgpu_gfx_get_enforce_isolation(struct device *dev,
  * @count: The size of the input data
  *
  * This function allows control over the 'enforce_isolation' feature, which
- * serializes access to the graphics engine. Writing '1' or '0' to the
- * 'enforce_isolation' sysfs file enables or disables process isolation for
- * each partition. The input should specify the setting for all partitions.
+ * serializes access to the graphics engine. Writing '0' to disable, '1' to
+ * enable isolation with cleaner shader, '2' to enable legacy isolation without
+ * cleaner shader, or '3' to enable process isolation without submitting the
+ * cleaner shader to the 'enforce_isolation' sysfs file sets the isolation mode
+ * for each partition. The input should specify the setting for all
+ * partitions.
  *
  * Return: The number of bytes written to the sysfs file.
  */
@@ -1660,13 +1690,34 @@ static ssize_t amdgpu_gfx_set_enforce_isolation(struct device *dev,
 		return -EINVAL;
 
 	for (i = 0; i < num_partitions; i++) {
-		if (partition_values[i] != 0 && partition_values[i] != 1)
+		if (partition_values[i] != 0 &&
+		    partition_values[i] != 1 &&
+		    partition_values[i] != 2 &&
+		    partition_values[i] != 3)
 			return -EINVAL;
 	}
 
 	mutex_lock(&adev->enforce_isolation_mutex);
-	for (i = 0; i < num_partitions; i++)
-		adev->enforce_isolation[i] = partition_values[i];
+	for (i = 0; i < num_partitions; i++) {
+		switch (partition_values[i]) {
+		case 0:
+		default:
+			adev->enforce_isolation[i] = AMDGPU_ENFORCE_ISOLATION_DISABLE;
+			break;
+		case 1:
+			adev->enforce_isolation[i] =
+				AMDGPU_ENFORCE_ISOLATION_ENABLE;
+			break;
+		case 2:
+			adev->enforce_isolation[i] =
+				AMDGPU_ENFORCE_ISOLATION_ENABLE_LEGACY;
+			break;
+		case 3:
+			adev->enforce_isolation[i] =
+				AMDGPU_ENFORCE_ISOLATION_NO_CLEANER_SHADER;
+			break;
+		}
+	}
 	mutex_unlock(&adev->enforce_isolation_mutex);
 
 	amdgpu_mes_update_enforce_isolation(adev);
@@ -1915,39 +1966,41 @@ void amdgpu_gfx_cleaner_shader_init(struct amdgpu_device *adev,
 static void amdgpu_gfx_kfd_sch_ctrl(struct amdgpu_device *adev, u32 idx,
 				    bool enable)
 {
-	mutex_lock(&adev->gfx.kfd_sch_mutex);
+	mutex_lock(&adev->gfx.userq_sch_mutex);
 
 	if (enable) {
 		/* If the count is already 0, it means there's an imbalance bug somewhere.
 		 * Note that the bug may be in a different caller than the one which triggers the
 		 * WARN_ON_ONCE.
 		 */
-		if (WARN_ON_ONCE(adev->gfx.kfd_sch_req_count[idx] == 0)) {
+		if (WARN_ON_ONCE(adev->gfx.userq_sch_req_count[idx] == 0)) {
 			dev_err(adev->dev, "Attempted to enable KFD scheduler when reference count is already zero\n");
 			goto unlock;
 		}
 
-		adev->gfx.kfd_sch_req_count[idx]--;
+		adev->gfx.userq_sch_req_count[idx]--;
 
-		if (adev->gfx.kfd_sch_req_count[idx] == 0 &&
-		    adev->gfx.kfd_sch_inactive[idx]) {
+		if (adev->gfx.userq_sch_req_count[idx] == 0 &&
+		    adev->gfx.userq_sch_inactive[idx]) {
 			schedule_delayed_work(&adev->gfx.enforce_isolation[idx].work,
 					      msecs_to_jiffies(adev->gfx.enforce_isolation_time[idx]));
 		}
 	} else {
-		if (adev->gfx.kfd_sch_req_count[idx] == 0) {
+		if (adev->gfx.userq_sch_req_count[idx] == 0) {
 			cancel_delayed_work_sync(&adev->gfx.enforce_isolation[idx].work);
-			if (!adev->gfx.kfd_sch_inactive[idx]) {
-				amdgpu_amdkfd_stop_sched(adev, idx);
-				adev->gfx.kfd_sch_inactive[idx] = true;
+			if (!adev->gfx.userq_sch_inactive[idx]) {
+				amdgpu_userq_stop_sched_for_enforce_isolation(adev, idx);
+				if (adev->kfd.init_complete)
+					amdgpu_amdkfd_stop_sched(adev, idx);
+				adev->gfx.userq_sch_inactive[idx] = true;
 			}
 		}
 
-		adev->gfx.kfd_sch_req_count[idx]++;
+		adev->gfx.userq_sch_req_count[idx]++;
 	}
 
 unlock:
-	mutex_unlock(&adev->gfx.kfd_sch_mutex);
+	mutex_unlock(&adev->gfx.userq_sch_mutex);
 }
 
 /**
@@ -1992,12 +2045,13 @@ void amdgpu_gfx_enforce_isolation_handler(struct work_struct *work)
 				      msecs_to_jiffies(1));
 	} else {
 		/* Tell KFD to resume the runqueue */
-		if (adev->kfd.init_complete) {
-			WARN_ON_ONCE(!adev->gfx.kfd_sch_inactive[idx]);
-			WARN_ON_ONCE(adev->gfx.kfd_sch_req_count[idx]);
+		WARN_ON_ONCE(!adev->gfx.userq_sch_inactive[idx]);
+		WARN_ON_ONCE(adev->gfx.userq_sch_req_count[idx]);
+
+		amdgpu_userq_start_sched_for_enforce_isolation(adev, idx);
+		if (adev->kfd.init_complete)
 			amdgpu_amdkfd_start_sched(adev, idx);
-			adev->gfx.kfd_sch_inactive[idx] = false;
-		}
+		adev->gfx.userq_sch_inactive[idx] = false;
 	}
 	mutex_unlock(&adev->enforce_isolation_mutex);
 }
@@ -2021,7 +2075,7 @@ amdgpu_gfx_enforce_isolation_wait_for_kfd(struct amdgpu_device *adev,
 	bool wait = false;
 
 	mutex_lock(&adev->enforce_isolation_mutex);
-	if (adev->enforce_isolation[idx]) {
+	if (adev->enforce_isolation[idx] == AMDGPU_ENFORCE_ISOLATION_ENABLE) {
 		/* set the initial values if nothing is set */
 		if (!adev->gfx.enforce_isolation_jiffies[idx]) {
 			adev->gfx.enforce_isolation_jiffies[idx] = jiffies;
@@ -2088,7 +2142,7 @@ void amdgpu_gfx_enforce_isolation_ring_begin_use(struct amdgpu_ring *ring)
 	amdgpu_gfx_enforce_isolation_wait_for_kfd(adev, idx);
 
 	mutex_lock(&adev->enforce_isolation_mutex);
-	if (adev->enforce_isolation[idx]) {
+	if (adev->enforce_isolation[idx] == AMDGPU_ENFORCE_ISOLATION_ENABLE) {
 		if (adev->kfd.init_complete)
 			sched_work = true;
 	}
@@ -2125,7 +2179,7 @@ void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring)
 		return;
 
 	mutex_lock(&adev->enforce_isolation_mutex);
-	if (adev->enforce_isolation[idx]) {
+	if (adev->enforce_isolation[idx] == AMDGPU_ENFORCE_ISOLATION_ENABLE) {
 		if (adev->kfd.init_complete)
 			sched_work = true;
 	}
@@ -2209,6 +2263,74 @@ void amdgpu_gfx_profile_ring_end_use(struct amdgpu_ring *ring)
 	schedule_delayed_work(&ring->adev->gfx.idle_work, GFX_PROFILE_IDLE_TIMEOUT);
 }
 
+/**
+ * amdgpu_gfx_csb_preamble_start - Set CSB preamble start
+ *
+ * @buffer: This is an output variable that gets the PACKET3 preamble setup.
+ *
+ * Return:
+ * return the latest index.
+ */
+u32 amdgpu_gfx_csb_preamble_start(volatile u32 *buffer)
+{
+	u32 count = 0;
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
+	buffer[count++] = cpu_to_le32(0x80000000);
+	buffer[count++] = cpu_to_le32(0x80000000);
+
+	return count;
+}
+
+/**
+ * amdgpu_gfx_csb_data_parser - Parser CS data
+ *
+ * @adev: amdgpu_device pointer used to get the CS data and other gfx info.
+ * @buffer: This is an output variable that gets the PACKET3 preamble end.
+ * @count: Index to start set the preemble end.
+ *
+ * Return:
+ * return the latest index.
+ */
+u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, volatile u32 *buffer, u32 count)
+{
+	const struct cs_section_def *sect = NULL;
+	const struct cs_extent_def *ext = NULL;
+	u32 i;
+
+	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
+		for (ext = sect->section; ext->extent != NULL; ++ext) {
+			if (sect->id == SECT_CONTEXT) {
+				buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
+				buffer[count++] = cpu_to_le32(ext->reg_index - PACKET3_SET_CONTEXT_REG_START);
+
+				for (i = 0; i < ext->reg_count; i++)
+					buffer[count++] = cpu_to_le32(ext->extent[i]);
+			}
+		}
+	}
+
+	return count;
+}
+
+/**
+ * amdgpu_gfx_csb_preamble_end - Set CSB preamble end
+ *
+ * @buffer: This is an output variable that gets the PACKET3 preamble end.
+ * @count: Index to start set the preemble end.
+ */
+void amdgpu_gfx_csb_preamble_end(volatile u32 *buffer, u32 count)
+{
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
+	buffer[count++] = cpu_to_le32(0);
+}
+
 /*
  * debugfs for to enable/disable gfx job submission to specific core.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 87e862188766..08f268dab8f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -170,10 +170,46 @@ struct amdgpu_kiq {
 #define AMDGPU_GFX_MAX_SE 4
 #define AMDGPU_GFX_MAX_SH_PER_SE 2
 
+/**
+ * amdgpu_rb_config - Configure a single Render Backend (RB)
+ *
+ * Bad RBs are fused off and there is a harvest register the driver reads to
+ * determine which RB(s) are fused off so that the driver can configure the
+ * hardware state so that nothing gets sent to them. There are also user
+ * harvest registers that the driver can program to disable additional RBs,
+ * etc., for testing purposes.
+ */
 struct amdgpu_rb_config {
+	/**
+	 * @rb_backend_disable:
+	 *
+	 * The value captured from register RB_BACKEND_DISABLE indicates if the
+	 * RB backend is disabled or not.
+	 */
 	uint32_t rb_backend_disable;
+
+	/**
+	 * @user_rb_backend_disable:
+	 *
+	 * The value captured from register USER_RB_BACKEND_DISABLE indicates
+	 * if the User RB backend is disabled or not.
+	 */
 	uint32_t user_rb_backend_disable;
+
+	/**
+	 * @raster_config:
+	 *
+	 * To set up all of the states, it is necessary to have two registers
+	 * to keep all of the states. This field holds the first register.
+	 */
 	uint32_t raster_config;
+
+	/**
+	 * @raster_config_1:
+	 *
+	 * To set up all of the states, it is necessary to have two registers
+	 * to keep all of the states. This field holds the second register.
+	 */
 	uint32_t raster_config_1;
 };
 
@@ -221,6 +257,13 @@ struct amdgpu_gfx_config {
 	uint32_t macrotile_mode_array[16];
 
 	struct gb_addr_config gb_addr_config_fields;
+
+	/**
+	 * @rb_config:
+	 *
+	 * Matrix that keeps all the Render Backend (color and depth buffer
+	 * handling) configuration on the 3D engine.
+	 */
 	struct amdgpu_rb_config rb_config[AMDGPU_GFX_MAX_SE][AMDGPU_GFX_MAX_SH_PER_SE];
 
 	/* gfx configure feature */
@@ -305,7 +348,8 @@ struct amdgpu_gfx_funcs {
 	void (*init_spm_golden)(struct amdgpu_device *adev);
 	void (*update_perfmon_mgcg)(struct amdgpu_device *adev, bool enable);
 	int (*get_gfx_shadow_info)(struct amdgpu_device *adev,
-				   struct amdgpu_gfx_shadow_info *shadow_info);
+				   struct amdgpu_gfx_shadow_info *shadow_info,
+				   bool skip_check);
 	enum amdgpu_gfx_partition
 			(*query_partition_mode)(struct amdgpu_device *adev);
 	int (*switch_partition_mode)(struct amdgpu_device *adev,
@@ -474,9 +518,9 @@ struct amdgpu_gfx {
 	bool				enable_cleaner_shader;
 	struct amdgpu_isolation_work	enforce_isolation[MAX_XCP];
 	/* Mutex for synchronizing KFD scheduler operations */
-	struct mutex                    kfd_sch_mutex;
-	u64				kfd_sch_req_count[MAX_XCP];
-	bool				kfd_sch_inactive[MAX_XCP];
+	struct mutex                    userq_sch_mutex;
+	u64				userq_sch_req_count[MAX_XCP];
+	bool				userq_sch_inactive[MAX_XCP];
 	unsigned long			enforce_isolation_jiffies[MAX_XCP];
 	unsigned long			enforce_isolation_time[MAX_XCP];
 
@@ -484,6 +528,9 @@ struct amdgpu_gfx {
 	struct delayed_work		idle_work;
 	bool				workload_profile_active;
 	struct mutex                    workload_profile_mutex;
+
+	bool				disable_kq;
+	bool				disable_uq;
 };
 
 struct amdgpu_gfx_ras_reg_entry {
@@ -503,7 +550,7 @@ struct amdgpu_gfx_ras_mem_id_entry {
 #define amdgpu_gfx_select_se_sh(adev, se, sh, instance, xcc_id) ((adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance), (xcc_id)))
 #define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid, xcc_id) ((adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid), (xcc_id)))
 #define amdgpu_gfx_init_spm_golden(adev) (adev)->gfx.funcs->init_spm_golden((adev))
-#define amdgpu_gfx_get_gfx_shadow_info(adev, si) ((adev)->gfx.funcs->get_gfx_shadow_info((adev), (si)))
+#define amdgpu_gfx_get_gfx_shadow_info(adev, si) ((adev)->gfx.funcs->get_gfx_shadow_info((adev), (si), false))
 
 /**
  * amdgpu_gfx_create_bitmask - create a bitmask
@@ -550,8 +597,6 @@ bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev,
 					       struct amdgpu_ring *ring);
 bool amdgpu_gfx_is_high_priority_graphics_queue(struct amdgpu_device *adev,
 						struct amdgpu_ring *ring);
-int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev, int me,
-			       int pipe, int queue);
 bool amdgpu_gfx_is_me_queue_enabled(struct amdgpu_device *adev, int me,
 				    int pipe, int queue);
 void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable);
@@ -597,6 +642,9 @@ void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring);
 void amdgpu_gfx_profile_idle_work_handler(struct work_struct *work);
 void amdgpu_gfx_profile_ring_begin_use(struct amdgpu_ring *ring);
 void amdgpu_gfx_profile_ring_end_use(struct amdgpu_ring *ring);
+u32 amdgpu_gfx_csb_preamble_start(volatile u32 *buffer);
+u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, volatile u32 *buffer, u32 count);
+void amdgpu_gfx_csb_preamble_end(volatile u32 *buffer, u32 count);
 
 void amdgpu_debugfs_gfx_sched_mask_init(struct amdgpu_device *adev);
 void amdgpu_debugfs_compute_sched_mask_init(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 464625282872..6b0fbbb91e57 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -699,12 +699,10 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid,
 				   uint32_t flush_type, bool all_hub,
 				   uint32_t inst)
 {
-	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT :
-		adev->usec_timeout;
 	struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst];
 	unsigned int ndw;
-	int r;
+	int r, cnt = 0;
 	uint32_t seq;
 
 	/*
@@ -761,10 +759,21 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid,
 
 		amdgpu_ring_commit(ring);
 		spin_unlock(&adev->gfx.kiq[inst].ring_lock);
-		if (amdgpu_fence_wait_polling(ring, seq, usec_timeout) < 1) {
+
+		r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
+
+		might_sleep();
+		while (r < 1 && cnt++ < MAX_KIQ_REG_TRY &&
+		       !amdgpu_reset_pending(adev->reset_domain)) {
+			msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
+			r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
+		}
+
+		if (cnt > MAX_KIQ_REG_TRY) {
 			dev_err(adev->dev, "timeout waiting for kiq fence\n");
 			r = -ETIME;
-		}
+		} else
+			r = 0;
 	}
 
 error_unlock_reset:
@@ -1221,6 +1230,10 @@ static ssize_t current_memory_partition_show(
 	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amdgpu_memory_partition mode;
 
+	/* Only minimal precaution taken to reject requests while in reset */
+	if (amdgpu_in_reset(adev))
+		return -EPERM;
+
 	mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
 	if ((mode >= ARRAY_SIZE(nps_desc)) ||
 	    (BIT(mode) & AMDGPU_ALL_NPS_MASK) != BIT(mode))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index bd7fc123b8f9..80fa29c26e9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -62,6 +62,9 @@
  */
 #define AMDGPU_GMC_FAULT_TIMEOUT	5000ULL
 
+/* XNACK flags */
+#define AMDGPU_GMC_XNACK_FLAG_CHAIN BIT(0)
+
 struct firmware;
 
 enum amdgpu_memory_partition {
@@ -301,6 +304,7 @@ struct amdgpu_gmc {
 	struct amdgpu_xgmi xgmi;
 	struct amdgpu_irq_src	ecc_irq;
 	int noretry;
+	uint32_t xnack_flags;
 
 	uint32_t	vmid0_page_table_block_size;
 	uint32_t	vmid0_page_table_depth;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c
index b6cf801939aa..6e02fb9ac2f6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c
@@ -22,6 +22,7 @@
  */
 #include "amdgpu.h"
 #include "amdgpu_ras.h"
+#include <uapi/linux/kfd_ioctl.h>
 
 int amdgpu_hdp_ras_sw_init(struct amdgpu_device *adev)
 {
@@ -46,3 +47,22 @@ int amdgpu_hdp_ras_sw_init(struct amdgpu_device *adev)
 	/* hdp ras follows amdgpu_ras_block_late_init_default for late init */
 	return 0;
 }
+
+void amdgpu_hdp_generic_flush(struct amdgpu_device *adev,
+			      struct amdgpu_ring *ring)
+{
+	if (!ring || !ring->funcs->emit_wreg) {
+		WREG32((adev->rmmio_remap.reg_offset +
+			KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >>
+			       2,
+		       0);
+		if (adev->nbio.funcs->get_memsize)
+			adev->nbio.funcs->get_memsize(adev);
+	} else {
+		amdgpu_ring_emit_wreg(ring,
+				      (adev->rmmio_remap.reg_offset +
+				       KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >>
+					      2,
+				      0);
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h
index 7b8a6152dc8d..4cfd932b7e91 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h
@@ -44,4 +44,6 @@ struct amdgpu_hdp {
 };
 
 int amdgpu_hdp_ras_sw_init(struct amdgpu_device *adev);
+void amdgpu_hdp_generic_flush(struct amdgpu_device *adev,
+			      struct amdgpu_ring *ring);
 #endif /* __AMDGPU_HDP_H__ */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 2ea98ec60220..802743efa3b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -163,12 +163,12 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs,
 		init_shadow = false;
 	}
 
-	if (!ring->sched.ready && !ring->is_mes_queue) {
+	if (!ring->sched.ready) {
 		dev_err(adev->dev, "couldn't schedule ib on ring <%s>\n", ring->name);
 		return -EINVAL;
 	}
 
-	if (vm && !job->vmid && !ring->is_mes_queue) {
+	if (vm && !job->vmid) {
 		dev_err(adev->dev, "VM IB without ID\n");
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index 4c4e087230ac..5dd78a9cb12d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
@@ -576,8 +576,16 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 		INIT_LIST_HEAD(&id_mgr->ids_lru);
 		id_mgr->reserved_use_count = 0;
 
-		/* manage only VMIDs not used by KFD */
-		id_mgr->num_ids = adev->vm_manager.first_kfd_vmid;
+		/* for GC <10, SDMA uses MMHUB so use first_kfd_vmid for both GC and MM */
+		if (amdgpu_ip_version(adev, GC_HWIP, 0) < IP_VERSION(10, 0, 0))
+			/* manage only VMIDs not used by KFD */
+			id_mgr->num_ids = adev->vm_manager.first_kfd_vmid;
+		else if (AMDGPU_IS_MMHUB0(i) ||
+			 AMDGPU_IS_MMHUB1(i))
+			id_mgr->num_ids = 16;
+		else
+			/* manage only VMIDs not used by KFD */
+			id_mgr->num_ids = adev->vm_manager.first_kfd_vmid;
 
 		/* skip over VMID 0, since it is the system VM */
 		for (j = 1; j < id_mgr->num_ids; ++j) {
@@ -588,7 +596,7 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 	}
 	/* alloc a default reserved vmid to enforce isolation */
 	for (i = 0; i < (adev->xcp_mgr ? adev->xcp_mgr->num_xcps : 1); i++) {
-		if (adev->enforce_isolation[i])
+		if (adev->enforce_isolation[i] != AMDGPU_ENFORCE_ISOLATION_DISABLE)
 			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(i));
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index 901f8b12c672..30f16968b578 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -25,6 +25,7 @@
 
 #include "amdgpu.h"
 #include "amdgpu_ih.h"
+#include "amdgpu_reset.h"
 
 /**
  * amdgpu_ih_ring_init - initialize the IH state
@@ -227,13 +228,23 @@ restart_ih:
 		ih->rptr &= ih->ptr_mask;
 	}
 
-	amdgpu_ih_set_rptr(adev, ih);
+	if (!ih->overflow)
+		amdgpu_ih_set_rptr(adev, ih);
+
 	wake_up_all(&ih->wait_process);
 
 	/* make sure wptr hasn't changed while processing */
 	wptr = amdgpu_ih_get_wptr(adev, ih);
 	if (wptr != ih->rptr)
-		goto restart_ih;
+		if (!ih->overflow)
+			goto restart_ih;
+
+	if (ih->overflow)
+		if (amdgpu_sriov_runtime(adev))
+			WARN_ONCE(!amdgpu_reset_domain_schedule(adev->reset_domain,
+				   &adev->virt.flr_work),
+				  "Failed to queue work! at %s",
+				  __func__);
 
 	return IRQ_HANDLED;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
index b0a88f92cd82..7f7ea046e209 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
@@ -72,6 +72,7 @@ struct amdgpu_ih_ring {
 	/* For waiting on IH processing at checkpoint. */
 	wait_queue_head_t wait_process;
 	uint64_t		processed_timestamp;
+	bool overflow;
 };
 
 /* return true if time stamp t2 is after t1 with 48bit wrap around */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 19ce4da285e8..5517451fc75d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -619,6 +619,10 @@ int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 		   unsigned int type)
 {
+	/* When the threshold is reached,the interrupt source may not be enabled.return -EINVAL */
+	if (amdgpu_ras_is_rma(adev))
+		return -EINVAL;
+
 	if (!adev->irq.installed)
 		return -ENOENT;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index ce6b9ba967ff..f2c049129661 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -78,6 +78,7 @@ struct amdgpu_job {
 
 	/* enforce isolation */
 	bool			enforce_isolation;
+	bool			run_cleaner_shader;
 
 	uint32_t		num_ibs;
 	struct amdgpu_ib	ibs[];
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 27bfe9c8af06..9fbb04aee97b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -45,6 +45,7 @@
 #include "amdgpu_ras.h"
 #include "amdgpu_reset.h"
 #include "amd_pcie.h"
+#include "amdgpu_userq.h"
 
 void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev)
 {
@@ -370,6 +371,26 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info,
 	return 0;
 }
 
+static int amdgpu_userq_metadata_info_gfx(struct amdgpu_device *adev,
+					  struct drm_amdgpu_info *info,
+					  struct drm_amdgpu_info_uq_metadata_gfx *meta)
+{
+	int ret = -EOPNOTSUPP;
+
+	if (adev->gfx.funcs->get_gfx_shadow_info) {
+		struct amdgpu_gfx_shadow_info shadow = {};
+
+		adev->gfx.funcs->get_gfx_shadow_info(adev, &shadow, true);
+		meta->shadow_size = shadow.shadow_size;
+		meta->shadow_alignment = shadow.shadow_alignment;
+		meta->csa_size = shadow.csa_size;
+		meta->csa_alignment = shadow.csa_alignment;
+		ret = 0;
+	}
+
+	return ret;
+}
+
 static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 			     struct drm_amdgpu_info *info,
 			     struct drm_amdgpu_info_hw_ip *result)
@@ -387,7 +408,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 	case AMDGPU_HW_IP_GFX:
 		type = AMD_IP_BLOCK_TYPE_GFX;
 		for (i = 0; i < adev->gfx.num_gfx_rings; i++)
-			if (adev->gfx.gfx_ring[i].sched.ready)
+			if (adev->gfx.gfx_ring[i].sched.ready &&
+			    !adev->gfx.gfx_ring[i].no_user_submission)
 				++num_rings;
 		ib_start_alignment = 32;
 		ib_size_alignment = 32;
@@ -395,7 +417,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 	case AMDGPU_HW_IP_COMPUTE:
 		type = AMD_IP_BLOCK_TYPE_GFX;
 		for (i = 0; i < adev->gfx.num_compute_rings; i++)
-			if (adev->gfx.compute_ring[i].sched.ready)
+			if (adev->gfx.compute_ring[i].sched.ready &&
+			    !adev->gfx.compute_ring[i].no_user_submission)
 				++num_rings;
 		ib_start_alignment = 32;
 		ib_size_alignment = 32;
@@ -403,7 +426,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 	case AMDGPU_HW_IP_DMA:
 		type = AMD_IP_BLOCK_TYPE_SDMA;
 		for (i = 0; i < adev->sdma.num_instances; i++)
-			if (adev->sdma.instance[i].ring.sched.ready)
+			if (adev->sdma.instance[i].ring.sched.ready &&
+			    !adev->sdma.instance[i].ring.no_user_submission)
 				++num_rings;
 		ib_start_alignment = 256;
 		ib_size_alignment = 4;
@@ -414,7 +438,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 			if (adev->uvd.harvest_config & (1 << i))
 				continue;
 
-			if (adev->uvd.inst[i].ring.sched.ready)
+			if (adev->uvd.inst[i].ring.sched.ready &&
+			    !adev->uvd.inst[i].ring.no_user_submission)
 				++num_rings;
 		}
 		ib_start_alignment = 256;
@@ -423,7 +448,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 	case AMDGPU_HW_IP_VCE:
 		type = AMD_IP_BLOCK_TYPE_VCE;
 		for (i = 0; i < adev->vce.num_rings; i++)
-			if (adev->vce.ring[i].sched.ready)
+			if (adev->vce.ring[i].sched.ready &&
+			    !adev->vce.ring[i].no_user_submission)
 				++num_rings;
 		ib_start_alignment = 256;
 		ib_size_alignment = 4;
@@ -435,7 +461,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 				continue;
 
 			for (j = 0; j < adev->uvd.num_enc_rings; j++)
-				if (adev->uvd.inst[i].ring_enc[j].sched.ready)
+				if (adev->uvd.inst[i].ring_enc[j].sched.ready &&
+				    !adev->uvd.inst[i].ring_enc[j].no_user_submission)
 					++num_rings;
 		}
 		ib_start_alignment = 256;
@@ -447,7 +474,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 			if (adev->vcn.harvest_config & (1 << i))
 				continue;
 
-			if (adev->vcn.inst[i].ring_dec.sched.ready)
+			if (adev->vcn.inst[i].ring_dec.sched.ready &&
+			    !adev->vcn.inst[i].ring_dec.no_user_submission)
 				++num_rings;
 		}
 		ib_start_alignment = 256;
@@ -460,7 +488,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 				continue;
 
 			for (j = 0; j < adev->vcn.inst[i].num_enc_rings; j++)
-				if (adev->vcn.inst[i].ring_enc[j].sched.ready)
+				if (adev->vcn.inst[i].ring_enc[j].sched.ready &&
+				    !adev->vcn.inst[i].ring_enc[j].no_user_submission)
 					++num_rings;
 		}
 		ib_start_alignment = 256;
@@ -475,7 +504,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 				continue;
 
 			for (j = 0; j < adev->jpeg.num_jpeg_rings; j++)
-				if (adev->jpeg.inst[i].ring_dec[j].sched.ready)
+				if (adev->jpeg.inst[i].ring_dec[j].sched.ready &&
+				    !adev->jpeg.inst[i].ring_dec[j].no_user_submission)
 					++num_rings;
 		}
 		ib_start_alignment = 256;
@@ -483,7 +513,8 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 		break;
 	case AMDGPU_HW_IP_VPE:
 		type = AMD_IP_BLOCK_TYPE_VPE;
-		if (adev->vpe.ring.sched.ready)
+		if (adev->vpe.ring.sched.ready &&
+		    !adev->vpe.ring.no_user_submission)
 			++num_rings;
 		ib_start_alignment = 256;
 		ib_size_alignment = 4;
@@ -978,6 +1009,8 @@ out:
 			}
 		}
 
+		dev_info->userq_ip_mask = amdgpu_userq_get_supported_ip_mask(adev);
+
 		ret = copy_to_user(out, dev_info,
 				   min((size_t)size, sizeof(*dev_info))) ? -EFAULT : 0;
 		kfree(dev_info);
@@ -1293,6 +1326,22 @@ out:
 		return copy_to_user(out, &gpuvm_fault,
 				    min((size_t)size, sizeof(gpuvm_fault))) ? -EFAULT : 0;
 	}
+	case AMDGPU_INFO_UQ_FW_AREAS: {
+		struct drm_amdgpu_info_uq_metadata meta_info = {};
+
+		switch (info->query_hw_ip.type) {
+		case AMDGPU_HW_IP_GFX:
+			ret = amdgpu_userq_metadata_info_gfx(adev, info, &meta_info.gfx);
+			if (ret)
+				return ret;
+
+			ret = copy_to_user(out, &meta_info,
+						min((size_t)size, sizeof(meta_info))) ? -EFAULT : 0;
+			return 0;
+		default:
+			return -EINVAL;
+		}
+	}
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->query);
 		return -EINVAL;
@@ -1376,6 +1425,14 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 	mutex_init(&fpriv->bo_list_lock);
 	idr_init_base(&fpriv->bo_list_handles, 1);
 
+	r = amdgpu_userq_mgr_init(&fpriv->userq_mgr, file_priv, adev);
+	if (r)
+		DRM_WARN("Can't setup usermode queues, use legacy workload submission only\n");
+
+	r = amdgpu_eviction_fence_init(&fpriv->evf_mgr);
+	if (r)
+		goto error_vm;
+
 	amdgpu_ctx_mgr_init(&fpriv->ctx_mgr, adev);
 
 	file_priv->driver_priv = fpriv;
@@ -1445,6 +1502,11 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
 		amdgpu_bo_unreserve(pd);
 	}
 
+	if (!fpriv->evf_mgr.fd_closing) {
+		fpriv->evf_mgr.fd_closing = true;
+		amdgpu_userq_mgr_fini(&fpriv->userq_mgr);
+		amdgpu_eviction_fence_destroy(&fpriv->evf_mgr);
+	}
 	amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr);
 	amdgpu_vm_fini(adev, &fpriv->vm);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index fb212f0a1136..2febb63ab232 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -39,42 +39,6 @@ int amdgpu_mes_doorbell_process_slice(struct amdgpu_device *adev)
 		       PAGE_SIZE);
 }
 
-static int amdgpu_mes_kernel_doorbell_get(struct amdgpu_device *adev,
-					 int ip_type, uint64_t *doorbell_index)
-{
-	unsigned int offset, found;
-	struct amdgpu_mes *mes = &adev->mes;
-
-	if (ip_type == AMDGPU_RING_TYPE_SDMA)
-		offset = adev->doorbell_index.sdma_engine[0];
-	else
-		offset = 0;
-
-	found = find_next_zero_bit(mes->doorbell_bitmap, mes->num_mes_dbs, offset);
-	if (found >= mes->num_mes_dbs) {
-		DRM_WARN("No doorbell available\n");
-		return -ENOSPC;
-	}
-
-	set_bit(found, mes->doorbell_bitmap);
-
-	/* Get the absolute doorbell index on BAR */
-	*doorbell_index = mes->db_start_dw_offset + found * 2;
-	return 0;
-}
-
-static void amdgpu_mes_kernel_doorbell_free(struct amdgpu_device *adev,
-					   uint32_t doorbell_index)
-{
-	unsigned int old, rel_index;
-	struct amdgpu_mes *mes = &adev->mes;
-
-	/* Find the relative index of the doorbell in this object */
-	rel_index = (doorbell_index - mes->db_start_dw_offset) / 2;
-	old = test_and_clear_bit(rel_index, mes->doorbell_bitmap);
-	WARN_ON(!old);
-}
-
 static int amdgpu_mes_doorbell_init(struct amdgpu_device *adev)
 {
 	int i;
@@ -126,7 +90,7 @@ static void amdgpu_mes_doorbell_free(struct amdgpu_device *adev)
 
 int amdgpu_mes_init(struct amdgpu_device *adev)
 {
-	int i, r;
+	int i, r, num_pipes;
 
 	adev->mes.adev = adev;
 
@@ -142,19 +106,52 @@ int amdgpu_mes_init(struct amdgpu_device *adev)
 
 	adev->mes.total_max_queue = AMDGPU_FENCE_MES_QUEUE_ID_MASK;
 	adev->mes.vmid_mask_mmhub = 0xffffff00;
-	adev->mes.vmid_mask_gfxhub = 0xffffff00;
+	adev->mes.vmid_mask_gfxhub = adev->gfx.disable_kq ? 0xfffffffe : 0xffffff00;
+
+	num_pipes = adev->gfx.me.num_pipe_per_me * adev->gfx.me.num_me;
+	if (num_pipes > AMDGPU_MES_MAX_GFX_PIPES)
+		dev_warn(adev->dev, "more gfx pipes than supported by MES! (%d vs %d)\n",
+			 num_pipes, AMDGPU_MES_MAX_GFX_PIPES);
+
+	for (i = 0; i < AMDGPU_MES_MAX_GFX_PIPES; i++) {
+		if (i >= num_pipes)
+			break;
+		if (amdgpu_ip_version(adev, GC_HWIP, 0) >=
+		    IP_VERSION(12, 0, 0))
+			/*
+			 * GFX V12 has only one GFX pipe, but 8 queues in it.
+			 * GFX pipe 0 queue 0 is being used by Kernel queue.
+			 * Set GFX pipe 0 queue 1-7 for MES scheduling
+			 * mask = 1111 1110b
+			 */
+			adev->mes.gfx_hqd_mask[i] = adev->gfx.disable_kq ? 0xFF : 0xFE;
+		else
+			/*
+			 * GFX pipe 0 queue 0 is being used by Kernel queue.
+			 * Set GFX pipe 0 queue 1 for MES scheduling
+			 * mask = 10b
+			 */
+			adev->mes.gfx_hqd_mask[i] = adev->gfx.disable_kq ? 0x3 : 0x2;
+	}
+
+	num_pipes = adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_mec;
+	if (num_pipes > AMDGPU_MES_MAX_COMPUTE_PIPES)
+		dev_warn(adev->dev, "more compute pipes than supported by MES! (%d vs %d)\n",
+			 num_pipes, AMDGPU_MES_MAX_COMPUTE_PIPES);
 
 	for (i = 0; i < AMDGPU_MES_MAX_COMPUTE_PIPES; i++) {
-		if (i >= (adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_mec))
+		if (i >= num_pipes)
 			break;
-		adev->mes.compute_hqd_mask[i] = 0xc;
+		adev->mes.compute_hqd_mask[i] = adev->gfx.disable_kq ? 0xF : 0xC;
 	}
 
-	for (i = 0; i < AMDGPU_MES_MAX_GFX_PIPES; i++)
-		adev->mes.gfx_hqd_mask[i] = i ? 0 : 0xfffffffe;
+	num_pipes = adev->sdma.num_instances;
+	if (num_pipes > AMDGPU_MES_MAX_SDMA_PIPES)
+		dev_warn(adev->dev, "more SDMA pipes than supported by MES! (%d vs %d)\n",
+			 num_pipes, AMDGPU_MES_MAX_SDMA_PIPES);
 
 	for (i = 0; i < AMDGPU_MES_MAX_SDMA_PIPES; i++) {
-		if (i >= adev->sdma.num_instances)
+		if (i >= num_pipes)
 			break;
 		adev->mes.sdma_hqd_mask[i] = 0xfc;
 	}
@@ -240,244 +237,6 @@ void amdgpu_mes_fini(struct amdgpu_device *adev)
 	mutex_destroy(&adev->mes.mutex_hidden);
 }
 
-static void amdgpu_mes_queue_free_mqd(struct amdgpu_mes_queue *q)
-{
-	amdgpu_bo_free_kernel(&q->mqd_obj,
-			      &q->mqd_gpu_addr,
-			      &q->mqd_cpu_ptr);
-}
-
-int amdgpu_mes_create_process(struct amdgpu_device *adev, int pasid,
-			      struct amdgpu_vm *vm)
-{
-	struct amdgpu_mes_process *process;
-	int r;
-
-	/* allocate the mes process buffer */
-	process = kzalloc(sizeof(struct amdgpu_mes_process), GFP_KERNEL);
-	if (!process) {
-		DRM_ERROR("no more memory to create mes process\n");
-		return -ENOMEM;
-	}
-
-	/* allocate the process context bo and map it */
-	r = amdgpu_bo_create_kernel(adev, AMDGPU_MES_PROC_CTX_SIZE, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_GTT,
-				    &process->proc_ctx_bo,
-				    &process->proc_ctx_gpu_addr,
-				    &process->proc_ctx_cpu_ptr);
-	if (r) {
-		DRM_ERROR("failed to allocate process context bo\n");
-		goto clean_up_memory;
-	}
-	memset(process->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE);
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	/* add the mes process to idr list */
-	r = idr_alloc(&adev->mes.pasid_idr, process, pasid, pasid + 1,
-		      GFP_KERNEL);
-	if (r < 0) {
-		DRM_ERROR("failed to lock pasid=%d\n", pasid);
-		goto clean_up_ctx;
-	}
-
-	INIT_LIST_HEAD(&process->gang_list);
-	process->vm = vm;
-	process->pasid = pasid;
-	process->process_quantum = adev->mes.default_process_quantum;
-	process->pd_gpu_addr = amdgpu_bo_gpu_offset(vm->root.bo);
-
-	amdgpu_mes_unlock(&adev->mes);
-	return 0;
-
-clean_up_ctx:
-	amdgpu_mes_unlock(&adev->mes);
-	amdgpu_bo_free_kernel(&process->proc_ctx_bo,
-			      &process->proc_ctx_gpu_addr,
-			      &process->proc_ctx_cpu_ptr);
-clean_up_memory:
-	kfree(process);
-	return r;
-}
-
-void amdgpu_mes_destroy_process(struct amdgpu_device *adev, int pasid)
-{
-	struct amdgpu_mes_process *process;
-	struct amdgpu_mes_gang *gang, *tmp1;
-	struct amdgpu_mes_queue *queue, *tmp2;
-	struct mes_remove_queue_input queue_input;
-	unsigned long flags;
-	int r;
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	process = idr_find(&adev->mes.pasid_idr, pasid);
-	if (!process) {
-		DRM_WARN("pasid %d doesn't exist\n", pasid);
-		amdgpu_mes_unlock(&adev->mes);
-		return;
-	}
-
-	/* Remove all queues from hardware */
-	list_for_each_entry_safe(gang, tmp1, &process->gang_list, list) {
-		list_for_each_entry_safe(queue, tmp2, &gang->queue_list, list) {
-			spin_lock_irqsave(&adev->mes.queue_id_lock, flags);
-			idr_remove(&adev->mes.queue_id_idr, queue->queue_id);
-			spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-
-			queue_input.doorbell_offset = queue->doorbell_off;
-			queue_input.gang_context_addr = gang->gang_ctx_gpu_addr;
-
-			r = adev->mes.funcs->remove_hw_queue(&adev->mes,
-							     &queue_input);
-			if (r)
-				DRM_WARN("failed to remove hardware queue\n");
-		}
-
-		idr_remove(&adev->mes.gang_id_idr, gang->gang_id);
-	}
-
-	idr_remove(&adev->mes.pasid_idr, pasid);
-	amdgpu_mes_unlock(&adev->mes);
-
-	/* free all memory allocated by the process */
-	list_for_each_entry_safe(gang, tmp1, &process->gang_list, list) {
-		/* free all queues in the gang */
-		list_for_each_entry_safe(queue, tmp2, &gang->queue_list, list) {
-			amdgpu_mes_queue_free_mqd(queue);
-			list_del(&queue->list);
-			kfree(queue);
-		}
-		amdgpu_bo_free_kernel(&gang->gang_ctx_bo,
-				      &gang->gang_ctx_gpu_addr,
-				      &gang->gang_ctx_cpu_ptr);
-		list_del(&gang->list);
-		kfree(gang);
-
-	}
-	amdgpu_bo_free_kernel(&process->proc_ctx_bo,
-			      &process->proc_ctx_gpu_addr,
-			      &process->proc_ctx_cpu_ptr);
-	kfree(process);
-}
-
-int amdgpu_mes_add_gang(struct amdgpu_device *adev, int pasid,
-			struct amdgpu_mes_gang_properties *gprops,
-			int *gang_id)
-{
-	struct amdgpu_mes_process *process;
-	struct amdgpu_mes_gang *gang;
-	int r;
-
-	/* allocate the mes gang buffer */
-	gang = kzalloc(sizeof(struct amdgpu_mes_gang), GFP_KERNEL);
-	if (!gang) {
-		return -ENOMEM;
-	}
-
-	/* allocate the gang context bo and map it to cpu space */
-	r = amdgpu_bo_create_kernel(adev, AMDGPU_MES_GANG_CTX_SIZE, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_GTT,
-				    &gang->gang_ctx_bo,
-				    &gang->gang_ctx_gpu_addr,
-				    &gang->gang_ctx_cpu_ptr);
-	if (r) {
-		DRM_ERROR("failed to allocate process context bo\n");
-		goto clean_up_mem;
-	}
-	memset(gang->gang_ctx_cpu_ptr, 0, AMDGPU_MES_GANG_CTX_SIZE);
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	process = idr_find(&adev->mes.pasid_idr, pasid);
-	if (!process) {
-		DRM_ERROR("pasid %d doesn't exist\n", pasid);
-		r = -EINVAL;
-		goto clean_up_ctx;
-	}
-
-	/* add the mes gang to idr list */
-	r = idr_alloc(&adev->mes.gang_id_idr, gang, 1, 0,
-		      GFP_KERNEL);
-	if (r < 0) {
-		DRM_ERROR("failed to allocate idr for gang\n");
-		goto clean_up_ctx;
-	}
-
-	gang->gang_id = r;
-	*gang_id = r;
-
-	INIT_LIST_HEAD(&gang->queue_list);
-	gang->process = process;
-	gang->priority = gprops->priority;
-	gang->gang_quantum = gprops->gang_quantum ?
-		gprops->gang_quantum : adev->mes.default_gang_quantum;
-	gang->global_priority_level = gprops->global_priority_level;
-	gang->inprocess_gang_priority = gprops->inprocess_gang_priority;
-	list_add_tail(&gang->list, &process->gang_list);
-
-	amdgpu_mes_unlock(&adev->mes);
-	return 0;
-
-clean_up_ctx:
-	amdgpu_mes_unlock(&adev->mes);
-	amdgpu_bo_free_kernel(&gang->gang_ctx_bo,
-			      &gang->gang_ctx_gpu_addr,
-			      &gang->gang_ctx_cpu_ptr);
-clean_up_mem:
-	kfree(gang);
-	return r;
-}
-
-int amdgpu_mes_remove_gang(struct amdgpu_device *adev, int gang_id)
-{
-	struct amdgpu_mes_gang *gang;
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	gang = idr_find(&adev->mes.gang_id_idr, gang_id);
-	if (!gang) {
-		DRM_ERROR("gang id %d doesn't exist\n", gang_id);
-		amdgpu_mes_unlock(&adev->mes);
-		return -EINVAL;
-	}
-
-	if (!list_empty(&gang->queue_list)) {
-		DRM_ERROR("queue list is not empty\n");
-		amdgpu_mes_unlock(&adev->mes);
-		return -EBUSY;
-	}
-
-	idr_remove(&adev->mes.gang_id_idr, gang->gang_id);
-	list_del(&gang->list);
-	amdgpu_mes_unlock(&adev->mes);
-
-	amdgpu_bo_free_kernel(&gang->gang_ctx_bo,
-			      &gang->gang_ctx_gpu_addr,
-			      &gang->gang_ctx_cpu_ptr);
-
-	kfree(gang);
-
-	return 0;
-}
-
 int amdgpu_mes_suspend(struct amdgpu_device *adev)
 {
 	struct mes_suspend_gang_input input;
@@ -526,304 +285,6 @@ int amdgpu_mes_resume(struct amdgpu_device *adev)
 	return r;
 }
 
-static int amdgpu_mes_queue_alloc_mqd(struct amdgpu_device *adev,
-				     struct amdgpu_mes_queue *q,
-				     struct amdgpu_mes_queue_properties *p)
-{
-	struct amdgpu_mqd *mqd_mgr = &adev->mqds[p->queue_type];
-	u32 mqd_size = mqd_mgr->mqd_size;
-	int r;
-
-	r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_GTT,
-				    &q->mqd_obj,
-				    &q->mqd_gpu_addr, &q->mqd_cpu_ptr);
-	if (r) {
-		dev_warn(adev->dev, "failed to create queue mqd bo (%d)", r);
-		return r;
-	}
-	memset(q->mqd_cpu_ptr, 0, mqd_size);
-
-	r = amdgpu_bo_reserve(q->mqd_obj, false);
-	if (unlikely(r != 0))
-		goto clean_up;
-
-	return 0;
-
-clean_up:
-	amdgpu_bo_free_kernel(&q->mqd_obj,
-			      &q->mqd_gpu_addr,
-			      &q->mqd_cpu_ptr);
-	return r;
-}
-
-static void amdgpu_mes_queue_init_mqd(struct amdgpu_device *adev,
-				     struct amdgpu_mes_queue *q,
-				     struct amdgpu_mes_queue_properties *p)
-{
-	struct amdgpu_mqd *mqd_mgr = &adev->mqds[p->queue_type];
-	struct amdgpu_mqd_prop mqd_prop = {0};
-
-	mqd_prop.mqd_gpu_addr = q->mqd_gpu_addr;
-	mqd_prop.hqd_base_gpu_addr = p->hqd_base_gpu_addr;
-	mqd_prop.rptr_gpu_addr = p->rptr_gpu_addr;
-	mqd_prop.wptr_gpu_addr = p->wptr_gpu_addr;
-	mqd_prop.queue_size = p->queue_size;
-	mqd_prop.use_doorbell = true;
-	mqd_prop.doorbell_index = p->doorbell_off;
-	mqd_prop.eop_gpu_addr = p->eop_gpu_addr;
-	mqd_prop.hqd_pipe_priority = p->hqd_pipe_priority;
-	mqd_prop.hqd_queue_priority = p->hqd_queue_priority;
-	mqd_prop.hqd_active = false;
-
-	if (p->queue_type == AMDGPU_RING_TYPE_GFX ||
-	    p->queue_type == AMDGPU_RING_TYPE_COMPUTE) {
-		mutex_lock(&adev->srbm_mutex);
-		amdgpu_gfx_select_me_pipe_q(adev, p->ring->me, p->ring->pipe, 0, 0, 0);
-	}
-
-	mqd_mgr->init_mqd(adev, q->mqd_cpu_ptr, &mqd_prop);
-
-	if (p->queue_type == AMDGPU_RING_TYPE_GFX ||
-	    p->queue_type == AMDGPU_RING_TYPE_COMPUTE) {
-		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0, 0);
-		mutex_unlock(&adev->srbm_mutex);
-	}
-
-	amdgpu_bo_unreserve(q->mqd_obj);
-}
-
-int amdgpu_mes_add_hw_queue(struct amdgpu_device *adev, int gang_id,
-			    struct amdgpu_mes_queue_properties *qprops,
-			    int *queue_id)
-{
-	struct amdgpu_mes_queue *queue;
-	struct amdgpu_mes_gang *gang;
-	struct mes_add_queue_input queue_input;
-	unsigned long flags;
-	int r;
-
-	memset(&queue_input, 0, sizeof(struct mes_add_queue_input));
-
-	/* allocate the mes queue buffer */
-	queue = kzalloc(sizeof(struct amdgpu_mes_queue), GFP_KERNEL);
-	if (!queue) {
-		DRM_ERROR("Failed to allocate memory for queue\n");
-		return -ENOMEM;
-	}
-
-	/* Allocate the queue mqd */
-	r = amdgpu_mes_queue_alloc_mqd(adev, queue, qprops);
-	if (r)
-		goto clean_up_memory;
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	gang = idr_find(&adev->mes.gang_id_idr, gang_id);
-	if (!gang) {
-		DRM_ERROR("gang id %d doesn't exist\n", gang_id);
-		r = -EINVAL;
-		goto clean_up_mqd;
-	}
-
-	/* add the mes gang to idr list */
-	spin_lock_irqsave(&adev->mes.queue_id_lock, flags);
-	r = idr_alloc(&adev->mes.queue_id_idr, queue, 1, 0,
-		      GFP_ATOMIC);
-	if (r < 0) {
-		spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-		goto clean_up_mqd;
-	}
-	spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-	*queue_id = queue->queue_id = r;
-
-	/* allocate a doorbell index for the queue */
-	r = amdgpu_mes_kernel_doorbell_get(adev,
-					  qprops->queue_type,
-					  &qprops->doorbell_off);
-	if (r)
-		goto clean_up_queue_id;
-
-	/* initialize the queue mqd */
-	amdgpu_mes_queue_init_mqd(adev, queue, qprops);
-
-	/* add hw queue to mes */
-	queue_input.process_id = gang->process->pasid;
-
-	queue_input.page_table_base_addr =
-		adev->vm_manager.vram_base_offset + gang->process->pd_gpu_addr -
-		adev->gmc.vram_start;
-
-	queue_input.process_va_start = 0;
-	queue_input.process_va_end =
-		(adev->vm_manager.max_pfn - 1) << AMDGPU_GPU_PAGE_SHIFT;
-	queue_input.process_quantum = gang->process->process_quantum;
-	queue_input.process_context_addr = gang->process->proc_ctx_gpu_addr;
-	queue_input.gang_quantum = gang->gang_quantum;
-	queue_input.gang_context_addr = gang->gang_ctx_gpu_addr;
-	queue_input.inprocess_gang_priority = gang->inprocess_gang_priority;
-	queue_input.gang_global_priority_level = gang->global_priority_level;
-	queue_input.doorbell_offset = qprops->doorbell_off;
-	queue_input.mqd_addr = queue->mqd_gpu_addr;
-	queue_input.wptr_addr = qprops->wptr_gpu_addr;
-	queue_input.wptr_mc_addr = qprops->wptr_mc_addr;
-	queue_input.queue_type = qprops->queue_type;
-	queue_input.paging = qprops->paging;
-	queue_input.is_kfd_process = 0;
-
-	r = adev->mes.funcs->add_hw_queue(&adev->mes, &queue_input);
-	if (r) {
-		DRM_ERROR("failed to add hardware queue to MES, doorbell=0x%llx\n",
-			  qprops->doorbell_off);
-		goto clean_up_doorbell;
-	}
-
-	DRM_DEBUG("MES hw queue was added, pasid=%d, gang id=%d, "
-		  "queue type=%d, doorbell=0x%llx\n",
-		  gang->process->pasid, gang_id, qprops->queue_type,
-		  qprops->doorbell_off);
-
-	queue->ring = qprops->ring;
-	queue->doorbell_off = qprops->doorbell_off;
-	queue->wptr_gpu_addr = qprops->wptr_gpu_addr;
-	queue->queue_type = qprops->queue_type;
-	queue->paging = qprops->paging;
-	queue->gang = gang;
-	queue->ring->mqd_ptr = queue->mqd_cpu_ptr;
-	list_add_tail(&queue->list, &gang->queue_list);
-
-	amdgpu_mes_unlock(&adev->mes);
-	return 0;
-
-clean_up_doorbell:
-	amdgpu_mes_kernel_doorbell_free(adev, qprops->doorbell_off);
-clean_up_queue_id:
-	spin_lock_irqsave(&adev->mes.queue_id_lock, flags);
-	idr_remove(&adev->mes.queue_id_idr, queue->queue_id);
-	spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-clean_up_mqd:
-	amdgpu_mes_unlock(&adev->mes);
-	amdgpu_mes_queue_free_mqd(queue);
-clean_up_memory:
-	kfree(queue);
-	return r;
-}
-
-int amdgpu_mes_remove_hw_queue(struct amdgpu_device *adev, int queue_id)
-{
-	unsigned long flags;
-	struct amdgpu_mes_queue *queue;
-	struct amdgpu_mes_gang *gang;
-	struct mes_remove_queue_input queue_input;
-	int r;
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	/* remove the mes gang from idr list */
-	spin_lock_irqsave(&adev->mes.queue_id_lock, flags);
-
-	queue = idr_find(&adev->mes.queue_id_idr, queue_id);
-	if (!queue) {
-		spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-		amdgpu_mes_unlock(&adev->mes);
-		DRM_ERROR("queue id %d doesn't exist\n", queue_id);
-		return -EINVAL;
-	}
-
-	idr_remove(&adev->mes.queue_id_idr, queue_id);
-	spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-
-	DRM_DEBUG("try to remove queue, doorbell off = 0x%llx\n",
-		  queue->doorbell_off);
-
-	gang = queue->gang;
-	queue_input.doorbell_offset = queue->doorbell_off;
-	queue_input.gang_context_addr = gang->gang_ctx_gpu_addr;
-
-	r = adev->mes.funcs->remove_hw_queue(&adev->mes, &queue_input);
-	if (r)
-		DRM_ERROR("failed to remove hardware queue, queue id = %d\n",
-			  queue_id);
-
-	list_del(&queue->list);
-	amdgpu_mes_kernel_doorbell_free(adev, queue->doorbell_off);
-	amdgpu_mes_unlock(&adev->mes);
-
-	amdgpu_mes_queue_free_mqd(queue);
-	kfree(queue);
-	return 0;
-}
-
-int amdgpu_mes_reset_hw_queue(struct amdgpu_device *adev, int queue_id)
-{
-	unsigned long flags;
-	struct amdgpu_mes_queue *queue;
-	struct amdgpu_mes_gang *gang;
-	struct mes_reset_queue_input queue_input;
-	int r;
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-
-	/* remove the mes gang from idr list */
-	spin_lock_irqsave(&adev->mes.queue_id_lock, flags);
-
-	queue = idr_find(&adev->mes.queue_id_idr, queue_id);
-	if (!queue) {
-		spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-		amdgpu_mes_unlock(&adev->mes);
-		DRM_ERROR("queue id %d doesn't exist\n", queue_id);
-		return -EINVAL;
-	}
-	spin_unlock_irqrestore(&adev->mes.queue_id_lock, flags);
-
-	DRM_DEBUG("try to reset queue, doorbell off = 0x%llx\n",
-		  queue->doorbell_off);
-
-	gang = queue->gang;
-	queue_input.doorbell_offset = queue->doorbell_off;
-	queue_input.gang_context_addr = gang->gang_ctx_gpu_addr;
-
-	r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input);
-	if (r)
-		DRM_ERROR("failed to reset hardware queue, queue id = %d\n",
-			  queue_id);
-
-	amdgpu_mes_unlock(&adev->mes);
-
-	return 0;
-}
-
-int amdgpu_mes_reset_hw_queue_mmio(struct amdgpu_device *adev, int queue_type,
-				   int me_id, int pipe_id, int queue_id, int vmid)
-{
-	struct mes_reset_queue_input queue_input;
-	int r;
-
-	queue_input.queue_type = queue_type;
-	queue_input.use_mmio = true;
-	queue_input.me_id = me_id;
-	queue_input.pipe_id = pipe_id;
-	queue_input.queue_id = queue_id;
-	queue_input.vmid = vmid;
-	r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input);
-	if (r)
-		DRM_ERROR("failed to reset hardware queue by mmio, queue id = %d\n",
-			  queue_id);
-	return r;
-}
-
 int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring)
 {
@@ -874,7 +335,7 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
 				  unsigned int vmid,
 				  bool use_mmio)
 {
-	struct mes_reset_legacy_queue_input queue_input;
+	struct mes_reset_queue_input queue_input;
 	int r;
 
 	memset(&queue_input, 0, sizeof(queue_input));
@@ -888,8 +349,11 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
 	queue_input.wptr_addr = ring->wptr_gpu_addr;
 	queue_input.vmid = vmid;
 	queue_input.use_mmio = use_mmio;
+	queue_input.is_kq = true;
+	if (ring->funcs->type == AMDGPU_RING_TYPE_GFX)
+		queue_input.legacy_gfx = true;
 
-	r = adev->mes.funcs->reset_legacy_queue(&adev->mes, &queue_input);
+	r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input);
 	if (r)
 		DRM_ERROR("failed to reset legacy queue\n");
 
@@ -905,7 +369,7 @@ uint32_t amdgpu_mes_rreg(struct amdgpu_device *adev, uint32_t reg)
 	uint32_t *read_val_ptr;
 
 	if (amdgpu_device_wb_get(adev, &addr_offset)) {
-		DRM_ERROR("critical bug! too many mes readers\n");
+		dev_err(adev->dev, "critical bug! too many mes readers\n");
 		goto error;
 	}
 	read_val_gpu_addr = adev->wb.gpu_addr + (addr_offset * 4);
@@ -915,13 +379,13 @@ uint32_t amdgpu_mes_rreg(struct amdgpu_device *adev, uint32_t reg)
 	op_input.read_reg.buffer_addr = read_val_gpu_addr;
 
 	if (!adev->mes.funcs->misc_op) {
-		DRM_ERROR("mes rreg is not supported!\n");
+		dev_err(adev->dev, "mes rreg is not supported!\n");
 		goto error;
 	}
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
 	if (r)
-		DRM_ERROR("failed to read reg (0x%x)\n", reg);
+		dev_err(adev->dev, "failed to read reg (0x%x)\n", reg);
 	else
 		val = *(read_val_ptr);
 
@@ -942,14 +406,14 @@ int amdgpu_mes_wreg(struct amdgpu_device *adev,
 	op_input.write_reg.reg_value = val;
 
 	if (!adev->mes.funcs->misc_op) {
-		DRM_ERROR("mes wreg is not supported!\n");
+		dev_err(adev->dev, "mes wreg is not supported!\n");
 		r = -EINVAL;
 		goto error;
 	}
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
 	if (r)
-		DRM_ERROR("failed to write reg (0x%x)\n", reg);
+		dev_err(adev->dev, "failed to write reg (0x%x)\n", reg);
 
 error:
 	return r;
@@ -969,14 +433,14 @@ int amdgpu_mes_reg_write_reg_wait(struct amdgpu_device *adev,
 	op_input.wrm_reg.mask = mask;
 
 	if (!adev->mes.funcs->misc_op) {
-		DRM_ERROR("mes reg_write_reg_wait is not supported!\n");
+		dev_err(adev->dev, "mes reg_write_reg_wait is not supported!\n");
 		r = -EINVAL;
 		goto error;
 	}
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
 	if (r)
-		DRM_ERROR("failed to reg_write_reg_wait\n");
+		dev_err(adev->dev, "failed to reg_write_reg_wait\n");
 
 error:
 	return r;
@@ -994,14 +458,14 @@ int amdgpu_mes_reg_wait(struct amdgpu_device *adev, uint32_t reg,
 	op_input.wrm_reg.mask = mask;
 
 	if (!adev->mes.funcs->misc_op) {
-		DRM_ERROR("mes reg wait is not supported!\n");
+		dev_err(adev->dev, "mes reg wait is not supported!\n");
 		r = -EINVAL;
 		goto error;
 	}
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
 	if (r)
-		DRM_ERROR("failed to reg_write_reg_wait\n");
+		dev_err(adev->dev, "failed to reg_write_reg_wait\n");
 
 error:
 	return r;
@@ -1075,25 +539,6 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev,
 	return r;
 }
 
-static void
-amdgpu_mes_ring_to_queue_props(struct amdgpu_device *adev,
-			       struct amdgpu_ring *ring,
-			       struct amdgpu_mes_queue_properties *props)
-{
-	props->queue_type = ring->funcs->type;
-	props->hqd_base_gpu_addr = ring->gpu_addr;
-	props->rptr_gpu_addr = ring->rptr_gpu_addr;
-	props->wptr_gpu_addr = ring->wptr_gpu_addr;
-	props->wptr_mc_addr =
-		ring->mes_ctx->meta_data_mc_addr + ring->wptr_offs;
-	props->queue_size = ring->ring_size;
-	props->eop_gpu_addr = ring->eop_gpu_addr;
-	props->hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_NORMAL;
-	props->hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MINIMUM;
-	props->paging = false;
-	props->ring = ring;
-}
-
 #define DEFINE_AMDGPU_MES_CTX_GET_OFFS_ENG(_eng)			\
 do {									\
        if (id_offs < AMDGPU_MES_CTX_MAX_OFFS)				\
@@ -1130,453 +575,12 @@ int amdgpu_mes_ctx_get_offs(struct amdgpu_ring *ring, unsigned int id_offs)
 	return -EINVAL;
 }
 
-int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id,
-			int queue_type, int idx,
-			struct amdgpu_mes_ctx_data *ctx_data,
-			struct amdgpu_ring **out)
-{
-	struct amdgpu_ring *ring;
-	struct amdgpu_mes_gang *gang;
-	struct amdgpu_mes_queue_properties qprops = {0};
-	int r, queue_id, pasid;
-
-	/*
-	 * Avoid taking any other locks under MES lock to avoid circular
-	 * lock dependencies.
-	 */
-	amdgpu_mes_lock(&adev->mes);
-	gang = idr_find(&adev->mes.gang_id_idr, gang_id);
-	if (!gang) {
-		DRM_ERROR("gang id %d doesn't exist\n", gang_id);
-		amdgpu_mes_unlock(&adev->mes);
-		return -EINVAL;
-	}
-	pasid = gang->process->pasid;
-
-	ring = kzalloc(sizeof(struct amdgpu_ring), GFP_KERNEL);
-	if (!ring) {
-		amdgpu_mes_unlock(&adev->mes);
-		return -ENOMEM;
-	}
-
-	ring->ring_obj = NULL;
-	ring->use_doorbell = true;
-	ring->is_mes_queue = true;
-	ring->mes_ctx = ctx_data;
-	ring->idx = idx;
-	ring->no_scheduler = true;
-
-	if (queue_type == AMDGPU_RING_TYPE_COMPUTE) {
-		int offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				      compute[ring->idx].mec_hpd);
-		ring->eop_gpu_addr =
-			amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-	}
-
-	switch (queue_type) {
-	case AMDGPU_RING_TYPE_GFX:
-		ring->funcs = adev->gfx.gfx_ring[0].funcs;
-		ring->me = adev->gfx.gfx_ring[0].me;
-		ring->pipe = adev->gfx.gfx_ring[0].pipe;
-		break;
-	case AMDGPU_RING_TYPE_COMPUTE:
-		ring->funcs = adev->gfx.compute_ring[0].funcs;
-		ring->me = adev->gfx.compute_ring[0].me;
-		ring->pipe = adev->gfx.compute_ring[0].pipe;
-		break;
-	case AMDGPU_RING_TYPE_SDMA:
-		ring->funcs = adev->sdma.instance[0].ring.funcs;
-		break;
-	default:
-		BUG();
-	}
-
-	r = amdgpu_ring_init(adev, ring, 1024, NULL, 0,
-			     AMDGPU_RING_PRIO_DEFAULT, NULL);
-	if (r) {
-		amdgpu_mes_unlock(&adev->mes);
-		goto clean_up_memory;
-	}
-
-	amdgpu_mes_ring_to_queue_props(adev, ring, &qprops);
-
-	dma_fence_wait(gang->process->vm->last_update, false);
-	dma_fence_wait(ctx_data->meta_data_va->last_pt_update, false);
-	amdgpu_mes_unlock(&adev->mes);
-
-	r = amdgpu_mes_add_hw_queue(adev, gang_id, &qprops, &queue_id);
-	if (r)
-		goto clean_up_ring;
-
-	ring->hw_queue_id = queue_id;
-	ring->doorbell_index = qprops.doorbell_off;
-
-	if (queue_type == AMDGPU_RING_TYPE_GFX)
-		sprintf(ring->name, "gfx_%d.%d.%d", pasid, gang_id, queue_id);
-	else if (queue_type == AMDGPU_RING_TYPE_COMPUTE)
-		sprintf(ring->name, "compute_%d.%d.%d", pasid, gang_id,
-			queue_id);
-	else if (queue_type == AMDGPU_RING_TYPE_SDMA)
-		sprintf(ring->name, "sdma_%d.%d.%d", pasid, gang_id,
-			queue_id);
-	else
-		BUG();
-
-	*out = ring;
-	return 0;
-
-clean_up_ring:
-	amdgpu_ring_fini(ring);
-clean_up_memory:
-	kfree(ring);
-	return r;
-}
-
-void amdgpu_mes_remove_ring(struct amdgpu_device *adev,
-			    struct amdgpu_ring *ring)
-{
-	if (!ring)
-		return;
-
-	amdgpu_mes_remove_hw_queue(adev, ring->hw_queue_id);
-	timer_delete_sync(&ring->fence_drv.fallback_timer);
-	amdgpu_ring_fini(ring);
-	kfree(ring);
-}
-
 uint32_t amdgpu_mes_get_aggregated_doorbell_index(struct amdgpu_device *adev,
 						   enum amdgpu_mes_priority_level prio)
 {
 	return adev->mes.aggregated_doorbells[prio];
 }
 
-int amdgpu_mes_ctx_alloc_meta_data(struct amdgpu_device *adev,
-				   struct amdgpu_mes_ctx_data *ctx_data)
-{
-	int r;
-
-	r = amdgpu_bo_create_kernel(adev,
-			    sizeof(struct amdgpu_mes_ctx_meta_data),
-			    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT,
-			    &ctx_data->meta_data_obj,
-			    &ctx_data->meta_data_mc_addr,
-			    &ctx_data->meta_data_ptr);
-	if (r) {
-		dev_warn(adev->dev, "(%d) create CTX bo failed\n", r);
-		return r;
-	}
-
-	if (!ctx_data->meta_data_obj)
-		return -ENOMEM;
-
-	memset(ctx_data->meta_data_ptr, 0,
-	       sizeof(struct amdgpu_mes_ctx_meta_data));
-
-	return 0;
-}
-
-void amdgpu_mes_ctx_free_meta_data(struct amdgpu_mes_ctx_data *ctx_data)
-{
-	if (ctx_data->meta_data_obj)
-		amdgpu_bo_free_kernel(&ctx_data->meta_data_obj,
-				      &ctx_data->meta_data_mc_addr,
-				      &ctx_data->meta_data_ptr);
-}
-
-int amdgpu_mes_ctx_map_meta_data(struct amdgpu_device *adev,
-				 struct amdgpu_vm *vm,
-				 struct amdgpu_mes_ctx_data *ctx_data)
-{
-	struct amdgpu_bo_va *bo_va;
-	struct amdgpu_sync sync;
-	struct drm_exec exec;
-	int r;
-
-	amdgpu_sync_create(&sync);
-
-	drm_exec_init(&exec, 0, 0);
-	drm_exec_until_all_locked(&exec) {
-		r = drm_exec_lock_obj(&exec,
-				      &ctx_data->meta_data_obj->tbo.base);
-		drm_exec_retry_on_contention(&exec);
-		if (unlikely(r))
-			goto error_fini_exec;
-
-		r = amdgpu_vm_lock_pd(vm, &exec, 0);
-		drm_exec_retry_on_contention(&exec);
-		if (unlikely(r))
-			goto error_fini_exec;
-	}
-
-	bo_va = amdgpu_vm_bo_add(adev, vm, ctx_data->meta_data_obj);
-	if (!bo_va) {
-		DRM_ERROR("failed to create bo_va for meta data BO\n");
-		r = -ENOMEM;
-		goto error_fini_exec;
-	}
-
-	r = amdgpu_vm_bo_map(adev, bo_va, ctx_data->meta_data_gpu_addr, 0,
-			     sizeof(struct amdgpu_mes_ctx_meta_data),
-			     AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
-			     AMDGPU_PTE_EXECUTABLE);
-
-	if (r) {
-		DRM_ERROR("failed to do bo_map on meta data, err=%d\n", r);
-		goto error_del_bo_va;
-	}
-
-	r = amdgpu_vm_bo_update(adev, bo_va, false);
-	if (r) {
-		DRM_ERROR("failed to do vm_bo_update on meta data\n");
-		goto error_del_bo_va;
-	}
-	amdgpu_sync_fence(&sync, bo_va->last_pt_update, GFP_KERNEL);
-
-	r = amdgpu_vm_update_pdes(adev, vm, false);
-	if (r) {
-		DRM_ERROR("failed to update pdes on meta data\n");
-		goto error_del_bo_va;
-	}
-	amdgpu_sync_fence(&sync, vm->last_update, GFP_KERNEL);
-
-	amdgpu_sync_wait(&sync, false);
-	drm_exec_fini(&exec);
-
-	amdgpu_sync_free(&sync);
-	ctx_data->meta_data_va = bo_va;
-	return 0;
-
-error_del_bo_va:
-	amdgpu_vm_bo_del(adev, bo_va);
-
-error_fini_exec:
-	drm_exec_fini(&exec);
-	amdgpu_sync_free(&sync);
-	return r;
-}
-
-int amdgpu_mes_ctx_unmap_meta_data(struct amdgpu_device *adev,
-				   struct amdgpu_mes_ctx_data *ctx_data)
-{
-	struct amdgpu_bo_va *bo_va = ctx_data->meta_data_va;
-	struct amdgpu_bo *bo = ctx_data->meta_data_obj;
-	struct amdgpu_vm *vm = bo_va->base.vm;
-	struct dma_fence *fence;
-	struct drm_exec exec;
-	long r;
-
-	drm_exec_init(&exec, 0, 0);
-	drm_exec_until_all_locked(&exec) {
-		r = drm_exec_lock_obj(&exec,
-				      &ctx_data->meta_data_obj->tbo.base);
-		drm_exec_retry_on_contention(&exec);
-		if (unlikely(r))
-			goto out_unlock;
-
-		r = amdgpu_vm_lock_pd(vm, &exec, 0);
-		drm_exec_retry_on_contention(&exec);
-		if (unlikely(r))
-			goto out_unlock;
-	}
-
-	amdgpu_vm_bo_del(adev, bo_va);
-	if (!amdgpu_vm_ready(vm))
-		goto out_unlock;
-
-	r = dma_resv_get_singleton(bo->tbo.base.resv, DMA_RESV_USAGE_BOOKKEEP,
-				   &fence);
-	if (r)
-		goto out_unlock;
-	if (fence) {
-		amdgpu_bo_fence(bo, fence, true);
-		fence = NULL;
-	}
-
-	r = amdgpu_vm_clear_freed(adev, vm, &fence);
-	if (r || !fence)
-		goto out_unlock;
-
-	dma_fence_wait(fence, false);
-	amdgpu_bo_fence(bo, fence, true);
-	dma_fence_put(fence);
-
-out_unlock:
-	if (unlikely(r < 0))
-		dev_err(adev->dev, "failed to clear page tables (%ld)\n", r);
-	drm_exec_fini(&exec);
-
-	return r;
-}
-
-static int amdgpu_mes_test_create_gang_and_queues(struct amdgpu_device *adev,
-					  int pasid, int *gang_id,
-					  int queue_type, int num_queue,
-					  struct amdgpu_ring **added_rings,
-					  struct amdgpu_mes_ctx_data *ctx_data)
-{
-	struct amdgpu_ring *ring;
-	struct amdgpu_mes_gang_properties gprops = {0};
-	int r, j;
-
-	/* create a gang for the process */
-	gprops.priority = AMDGPU_MES_PRIORITY_LEVEL_NORMAL;
-	gprops.gang_quantum = adev->mes.default_gang_quantum;
-	gprops.inprocess_gang_priority = AMDGPU_MES_PRIORITY_LEVEL_NORMAL;
-	gprops.priority_level = AMDGPU_MES_PRIORITY_LEVEL_NORMAL;
-	gprops.global_priority_level = AMDGPU_MES_PRIORITY_LEVEL_NORMAL;
-
-	r = amdgpu_mes_add_gang(adev, pasid, &gprops, gang_id);
-	if (r) {
-		DRM_ERROR("failed to add gang\n");
-		return r;
-	}
-
-	/* create queues for the gang */
-	for (j = 0; j < num_queue; j++) {
-		r = amdgpu_mes_add_ring(adev, *gang_id, queue_type, j,
-					ctx_data, &ring);
-		if (r) {
-			DRM_ERROR("failed to add ring\n");
-			break;
-		}
-
-		DRM_INFO("ring %s was added\n", ring->name);
-		added_rings[j] = ring;
-	}
-
-	return 0;
-}
-
-static int amdgpu_mes_test_queues(struct amdgpu_ring **added_rings)
-{
-	struct amdgpu_ring *ring;
-	int i, r;
-
-	for (i = 0; i < AMDGPU_MES_CTX_MAX_RINGS; i++) {
-		ring = added_rings[i];
-		if (!ring)
-			continue;
-
-		r = amdgpu_ring_test_helper(ring);
-		if (r)
-			return r;
-
-		r = amdgpu_ring_test_ib(ring, 1000 * 10);
-		if (r) {
-			DRM_DEV_ERROR(ring->adev->dev,
-				      "ring %s ib test failed (%d)\n",
-				      ring->name, r);
-			return r;
-		} else
-			DRM_INFO("ring %s ib test pass\n", ring->name);
-	}
-
-	return 0;
-}
-
-int amdgpu_mes_self_test(struct amdgpu_device *adev)
-{
-	struct amdgpu_vm *vm = NULL;
-	struct amdgpu_mes_ctx_data ctx_data = {0};
-	struct amdgpu_ring *added_rings[AMDGPU_MES_CTX_MAX_RINGS] = { NULL };
-	int gang_ids[3] = {0};
-	int queue_types[][2] = { { AMDGPU_RING_TYPE_GFX, 1 },
-				 { AMDGPU_RING_TYPE_COMPUTE, 1 },
-				 { AMDGPU_RING_TYPE_SDMA, 1} };
-	int i, r, pasid, k = 0;
-
-	pasid = amdgpu_pasid_alloc(16);
-	if (pasid < 0) {
-		dev_warn(adev->dev, "No more PASIDs available!");
-		pasid = 0;
-	}
-
-	vm = kzalloc(sizeof(*vm), GFP_KERNEL);
-	if (!vm) {
-		r = -ENOMEM;
-		goto error_pasid;
-	}
-
-	r = amdgpu_vm_init(adev, vm, -1);
-	if (r) {
-		DRM_ERROR("failed to initialize vm\n");
-		goto error_pasid;
-	}
-
-	r = amdgpu_mes_ctx_alloc_meta_data(adev, &ctx_data);
-	if (r) {
-		DRM_ERROR("failed to alloc ctx meta data\n");
-		goto error_fini;
-	}
-
-	ctx_data.meta_data_gpu_addr = AMDGPU_VA_RESERVED_BOTTOM;
-	r = amdgpu_mes_ctx_map_meta_data(adev, vm, &ctx_data);
-	if (r) {
-		DRM_ERROR("failed to map ctx meta data\n");
-		goto error_vm;
-	}
-
-	r = amdgpu_mes_create_process(adev, pasid, vm);
-	if (r) {
-		DRM_ERROR("failed to create MES process\n");
-		goto error_vm;
-	}
-
-	for (i = 0; i < ARRAY_SIZE(queue_types); i++) {
-		/* On GFX v10.3, fw hasn't supported to map sdma queue. */
-		if (amdgpu_ip_version(adev, GC_HWIP, 0) >=
-			    IP_VERSION(10, 3, 0) &&
-		    amdgpu_ip_version(adev, GC_HWIP, 0) <
-			    IP_VERSION(11, 0, 0) &&
-		    queue_types[i][0] == AMDGPU_RING_TYPE_SDMA)
-			continue;
-
-		r = amdgpu_mes_test_create_gang_and_queues(adev, pasid,
-							   &gang_ids[i],
-							   queue_types[i][0],
-							   queue_types[i][1],
-							   &added_rings[k],
-							   &ctx_data);
-		if (r)
-			goto error_queues;
-
-		k += queue_types[i][1];
-	}
-
-	/* start ring test and ib test for MES queues */
-	amdgpu_mes_test_queues(added_rings);
-
-error_queues:
-	/* remove all queues */
-	for (i = 0; i < ARRAY_SIZE(added_rings); i++) {
-		if (!added_rings[i])
-			continue;
-		amdgpu_mes_remove_ring(adev, added_rings[i]);
-	}
-
-	for (i = 0; i < ARRAY_SIZE(gang_ids); i++) {
-		if (!gang_ids[i])
-			continue;
-		amdgpu_mes_remove_gang(adev, gang_ids[i]);
-	}
-
-	amdgpu_mes_destroy_process(adev, pasid);
-
-error_vm:
-	amdgpu_mes_ctx_unmap_meta_data(adev, &ctx_data);
-
-error_fini:
-	amdgpu_vm_fini(adev, vm);
-
-error_pasid:
-	if (pasid)
-		amdgpu_pasid_free(pasid);
-
-	amdgpu_mes_ctx_free_meta_data(&ctx_data);
-	kfree(vm);
-	return 0;
-}
-
 int amdgpu_mes_init_microcode(struct amdgpu_device *adev, int pipe)
 {
 	const struct mes_firmware_header_v1_0 *mes_hdr;
@@ -1705,7 +709,7 @@ int amdgpu_mes_update_enforce_isolation(struct amdgpu_device *adev)
 	if (adev->enable_mes && adev->gfx.enable_cleaner_shader) {
 		mutex_lock(&adev->enforce_isolation_mutex);
 		for (i = 0; i < (adev->xcp_mgr ? adev->xcp_mgr->num_xcps : 1); i++) {
-			if (adev->enforce_isolation[i])
+			if (adev->enforce_isolation[i] == AMDGPU_ENFORCE_ISOLATION_ENABLE)
 				r |= amdgpu_mes_set_enforce_isolation(adev, i, true);
 			else
 				r |= amdgpu_mes_set_enforce_isolation(adev, i, false);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index da2c9a8cb3e0..a41f65b4f733 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -111,8 +111,8 @@ struct amdgpu_mes {
 
 	uint32_t                        vmid_mask_gfxhub;
 	uint32_t                        vmid_mask_mmhub;
-	uint32_t                        compute_hqd_mask[AMDGPU_MES_MAX_COMPUTE_PIPES];
 	uint32_t                        gfx_hqd_mask[AMDGPU_MES_MAX_GFX_PIPES];
+	uint32_t                        compute_hqd_mask[AMDGPU_MES_MAX_COMPUTE_PIPES];
 	uint32_t                        sdma_hqd_mask[AMDGPU_MES_MAX_SDMA_PIPES];
 	uint32_t                        aggregated_doorbells[AMDGPU_MES_PRIORITY_NUM_LEVELS];
 	uint32_t                        sch_ctx_offs[AMDGPU_MAX_MES_PIPES];
@@ -149,19 +149,6 @@ struct amdgpu_mes {
 
 };
 
-struct amdgpu_mes_process {
-	int			pasid;
-	struct			amdgpu_vm *vm;
-	uint64_t		pd_gpu_addr;
-	struct amdgpu_bo 	*proc_ctx_bo;
-	uint64_t 		proc_ctx_gpu_addr;
-	void 			*proc_ctx_cpu_ptr;
-	uint64_t 		process_quantum;
-	struct 			list_head gang_list;
-	uint32_t 		doorbell_index;
-	struct mutex		doorbell_lock;
-};
-
 struct amdgpu_mes_gang {
 	int 				gang_id;
 	int 				priority;
@@ -248,18 +235,6 @@ struct mes_remove_queue_input {
 	uint64_t	gang_context_addr;
 };
 
-struct mes_reset_queue_input {
-	uint32_t	doorbell_offset;
-	uint64_t	gang_context_addr;
-	bool		use_mmio;
-	uint32_t	queue_type;
-	uint32_t	me_id;
-	uint32_t	pipe_id;
-	uint32_t	queue_id;
-	uint32_t	xcc_id;
-	uint32_t	vmid;
-};
-
 struct mes_map_legacy_queue_input {
 	uint32_t                           queue_type;
 	uint32_t                           doorbell_offset;
@@ -291,7 +266,7 @@ struct mes_resume_gang_input {
 	uint64_t	gang_context_addr;
 };
 
-struct mes_reset_legacy_queue_input {
+struct mes_reset_queue_input {
 	uint32_t                           queue_type;
 	uint32_t                           doorbell_offset;
 	bool                               use_mmio;
@@ -301,6 +276,8 @@ struct mes_reset_legacy_queue_input {
 	uint64_t                           mqd_addr;
 	uint64_t                           wptr_addr;
 	uint32_t                           vmid;
+	bool                               legacy_gfx;
+	bool                               is_kq;
 };
 
 enum mes_misc_opcode {
@@ -388,9 +365,6 @@ struct amdgpu_mes_funcs {
 	int (*misc_op)(struct amdgpu_mes *mes,
 		       struct mes_misc_op_input *input);
 
-	int (*reset_legacy_queue)(struct amdgpu_mes *mes,
-				  struct mes_reset_legacy_queue_input *input);
-
 	int (*reset_hw_queue)(struct amdgpu_mes *mes,
 			      struct mes_reset_queue_input *input);
 };
@@ -404,26 +378,9 @@ int amdgpu_mes_init_microcode(struct amdgpu_device *adev, int pipe);
 int amdgpu_mes_init(struct amdgpu_device *adev);
 void amdgpu_mes_fini(struct amdgpu_device *adev);
 
-int amdgpu_mes_create_process(struct amdgpu_device *adev, int pasid,
-			      struct amdgpu_vm *vm);
-void amdgpu_mes_destroy_process(struct amdgpu_device *adev, int pasid);
-
-int amdgpu_mes_add_gang(struct amdgpu_device *adev, int pasid,
-			struct amdgpu_mes_gang_properties *gprops,
-			int *gang_id);
-int amdgpu_mes_remove_gang(struct amdgpu_device *adev, int gang_id);
-
 int amdgpu_mes_suspend(struct amdgpu_device *adev);
 int amdgpu_mes_resume(struct amdgpu_device *adev);
 
-int amdgpu_mes_add_hw_queue(struct amdgpu_device *adev, int gang_id,
-			    struct amdgpu_mes_queue_properties *qprops,
-			    int *queue_id);
-int amdgpu_mes_remove_hw_queue(struct amdgpu_device *adev, int queue_id);
-int amdgpu_mes_reset_hw_queue(struct amdgpu_device *adev, int queue_id);
-int amdgpu_mes_reset_hw_queue_mmio(struct amdgpu_device *adev, int queue_type,
-				   int me_id, int pipe_id, int queue_id, int vmid);
-
 int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev,
 				struct amdgpu_ring *ring);
 int amdgpu_mes_unmap_legacy_queue(struct amdgpu_device *adev,
@@ -451,27 +408,10 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 				bool trap_en);
 int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev,
 				uint64_t process_context_addr);
-int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id,
-			int queue_type, int idx,
-			struct amdgpu_mes_ctx_data *ctx_data,
-			struct amdgpu_ring **out);
-void amdgpu_mes_remove_ring(struct amdgpu_device *adev,
-			    struct amdgpu_ring *ring);
 
 uint32_t amdgpu_mes_get_aggregated_doorbell_index(struct amdgpu_device *adev,
 						   enum amdgpu_mes_priority_level prio);
 
-int amdgpu_mes_ctx_alloc_meta_data(struct amdgpu_device *adev,
-				   struct amdgpu_mes_ctx_data *ctx_data);
-void amdgpu_mes_ctx_free_meta_data(struct amdgpu_mes_ctx_data *ctx_data);
-int amdgpu_mes_ctx_map_meta_data(struct amdgpu_device *adev,
-				 struct amdgpu_vm *vm,
-				 struct amdgpu_mes_ctx_data *ctx_data);
-int amdgpu_mes_ctx_unmap_meta_data(struct amdgpu_device *adev,
-				   struct amdgpu_mes_ctx_data *ctx_data);
-
-int amdgpu_mes_self_test(struct amdgpu_device *adev);
-
 int amdgpu_mes_doorbell_process_slice(struct amdgpu_device *adev);
 
 /*
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 80cd6f5273db..73403744331a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -163,8 +163,8 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 		 * When GTT is just an alternative to VRAM make sure that we
 		 * only use it as fallback and still try to fill up VRAM first.
 		 */
-		if (domain & abo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM &&
-		    !(adev->flags & AMD_IS_APU))
+		if (abo->tbo.resource && !(adev->flags & AMD_IS_APU) &&
+		    domain & abo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM)
 			places[c].flags |= TTM_PL_FLAG_FALLBACK;
 		c++;
 	}
@@ -1044,7 +1044,8 @@ static const char * const amdgpu_vram_names[] = {
 	"GDDR6",
 	"DDR5",
 	"LPDDR4",
-	"LPDDR5"
+	"LPDDR5",
+	"HBM3E"
 };
 
 /**
@@ -1644,7 +1645,11 @@ u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m)
 	amdgpu_bo_print_flag(m, bo, VRAM_CONTIGUOUS);
 	amdgpu_bo_print_flag(m, bo, VM_ALWAYS_VALID);
 	amdgpu_bo_print_flag(m, bo, EXPLICIT_SYNC);
-
+	/* Add the gem obj resv fence dump*/
+	if (dma_resv_trylock(bo->tbo.base.resv)) {
+		dma_resv_describe(bo->tbo.base.resv, m);
+		dma_resv_unlock(bo->tbo.base.resv);
+	}
 	seq_puts(m, "\n");
 
 	return size;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index df5d5dbd7f0f..e6f0b035e20b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -2214,7 +2214,8 @@ static int psp_securedisplay_initialize(struct psp_context *psp)
 
 	if (!psp->securedisplay_context.context.bin_desc.size_bytes ||
 	    !psp->securedisplay_context.context.bin_desc.start_addr) {
-		dev_info(psp->adev->dev, "SECUREDISPLAY: securedisplay ta ucode is not available\n");
+		dev_info(psp->adev->dev,
+			 "SECUREDISPLAY: optional securedisplay ta ucode is not available\n");
 		return 0;
 	}
 
@@ -4185,6 +4186,110 @@ const struct attribute_group amdgpu_flash_attr_group = {
 	.is_visible = amdgpu_flash_attr_is_visible,
 };
 
+#if defined(CONFIG_DEBUG_FS)
+static int psp_read_spirom_debugfs_open(struct inode *inode, struct file *filp)
+{
+	struct amdgpu_device *adev = filp->f_inode->i_private;
+	struct spirom_bo *bo_triplet;
+	int ret;
+
+	/* serialize the open() file calling */
+	if (!mutex_trylock(&adev->psp.mutex))
+		return -EBUSY;
+
+	/*
+	 * make sure only one userpace process is alive for dumping so that
+	 * only one memory buffer of AMD_VBIOS_FILE_MAX_SIZE * 2 is consumed.
+	 * let's say the case where one process try opening the file while
+	 * another one has proceeded to read or release. In this way, eliminate
+	 * the use of mutex for read() or release() callback as well.
+	 */
+	if (adev->psp.spirom_dump_trip) {
+		mutex_unlock(&adev->psp.mutex);
+		return -EBUSY;
+	}
+
+	bo_triplet = kzalloc(sizeof(struct spirom_bo), GFP_KERNEL);
+	if (!bo_triplet) {
+		mutex_unlock(&adev->psp.mutex);
+		return -ENOMEM;
+	}
+
+	ret = amdgpu_bo_create_kernel(adev, AMD_VBIOS_FILE_MAX_SIZE_B * 2,
+				      AMDGPU_GPU_PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_GTT,
+				      &bo_triplet->bo,
+				      &bo_triplet->mc_addr,
+				      &bo_triplet->cpu_addr);
+	if (ret)
+		goto rel_trip;
+
+	ret = psp_dump_spirom(&adev->psp, bo_triplet->mc_addr);
+	if (ret)
+		goto rel_bo;
+
+	adev->psp.spirom_dump_trip = bo_triplet;
+	mutex_unlock(&adev->psp.mutex);
+	return 0;
+rel_bo:
+	amdgpu_bo_free_kernel(&bo_triplet->bo, &bo_triplet->mc_addr,
+			      &bo_triplet->cpu_addr);
+rel_trip:
+	kfree(bo_triplet);
+	mutex_unlock(&adev->psp.mutex);
+	dev_err(adev->dev, "Trying IFWI dump fails, err = %d\n", ret);
+	return ret;
+}
+
+static ssize_t psp_read_spirom_debugfs_read(struct file *filp, char __user *buf, size_t size,
+					    loff_t *pos)
+{
+	struct amdgpu_device *adev = filp->f_inode->i_private;
+	struct spirom_bo *bo_triplet = adev->psp.spirom_dump_trip;
+
+	if (!bo_triplet)
+		return -EINVAL;
+
+	return simple_read_from_buffer(buf,
+				       size,
+				       pos, bo_triplet->cpu_addr,
+				       AMD_VBIOS_FILE_MAX_SIZE_B * 2);
+}
+
+static int psp_read_spirom_debugfs_release(struct inode *inode, struct file *filp)
+{
+	struct amdgpu_device *adev = filp->f_inode->i_private;
+	struct spirom_bo *bo_triplet = adev->psp.spirom_dump_trip;
+
+	if (bo_triplet) {
+		amdgpu_bo_free_kernel(&bo_triplet->bo, &bo_triplet->mc_addr,
+				      &bo_triplet->cpu_addr);
+		kfree(bo_triplet);
+	}
+
+	adev->psp.spirom_dump_trip = NULL;
+	return 0;
+}
+
+static const struct file_operations psp_dump_spirom_debugfs_ops = {
+	.owner = THIS_MODULE,
+	.open = psp_read_spirom_debugfs_open,
+	.read = psp_read_spirom_debugfs_read,
+	.release = psp_read_spirom_debugfs_release,
+	.llseek = default_llseek,
+};
+#endif
+
+void amdgpu_psp_debugfs_init(struct amdgpu_device *adev)
+{
+#if defined(CONFIG_DEBUG_FS)
+	struct drm_minor *minor = adev_to_drm(adev)->primary;
+
+	debugfs_create_file_size("psp_spirom_dump", 0444, minor->debugfs_root,
+				 adev, &psp_dump_spirom_debugfs_ops, AMD_VBIOS_FILE_MAX_SIZE_B * 2);
+#endif
+}
+
 const struct amd_ip_funcs psp_ip_funcs = {
 	.name = "psp",
 	.early_init = psp_early_init,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 8d5acc415d38..428adc7f741d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -39,6 +39,18 @@
 #define PSP_TMR_ALIGNMENT	0x100000
 #define PSP_FW_NAME_LEN		0x24
 
+/* VBIOS gfl defines */
+#define MBOX_READY_MASK 0x80000000
+#define MBOX_STATUS_MASK 0x0000FFFF
+#define MBOX_COMMAND_MASK 0x00FF0000
+#define MBOX_READY_FLAG 0x80000000
+#define C2PMSG_CMD_SPI_UPDATE_ROM_IMAGE_ADDR_LO 0x2
+#define C2PMSG_CMD_SPI_UPDATE_ROM_IMAGE_ADDR_HI 0x3
+#define C2PMSG_CMD_SPI_UPDATE_FLASH_IMAGE 0x4
+#define C2PMSG_CMD_SPI_GET_ROM_IMAGE_ADDR_LO 0xf
+#define C2PMSG_CMD_SPI_GET_ROM_IMAGE_ADDR_HI 0x10
+#define C2PMSG_CMD_SPI_GET_FLASH_IMAGE 0x11
+
 extern const struct attribute_group amdgpu_flash_attr_group;
 
 enum psp_shared_mem_size {
@@ -107,6 +119,7 @@ enum psp_reg_prog_id {
 	PSP_REG_IH_RB_CNTL        = 0,  /* register IH_RB_CNTL */
 	PSP_REG_IH_RB_CNTL_RING1  = 1,  /* register IH_RB_CNTL_RING1 */
 	PSP_REG_IH_RB_CNTL_RING2  = 2,  /* register IH_RB_CNTL_RING2 */
+	PSP_REG_MMHUB_L1_TLB_CNTL = 25,
 	PSP_REG_LAST
 };
 
@@ -137,11 +150,14 @@ struct psp_funcs {
 	int (*load_usbc_pd_fw)(struct psp_context *psp, uint64_t fw_pri_mc_addr);
 	int (*read_usbc_pd_fw)(struct psp_context *psp, uint32_t *fw_ver);
 	int (*update_spirom)(struct psp_context *psp, uint64_t fw_pri_mc_addr);
+	int (*dump_spirom)(struct psp_context *psp, uint64_t fw_pri_mc_addr);
 	int (*vbflash_stat)(struct psp_context *psp);
 	int (*fatal_error_recovery_quirk)(struct psp_context *psp);
 	bool (*get_ras_capability)(struct psp_context *psp);
 	bool (*is_aux_sos_load_required)(struct psp_context *psp);
 	bool (*is_reload_needed)(struct psp_context *psp);
+	int (*reg_program_no_ring)(struct psp_context *psp, uint32_t val,
+				   enum psp_reg_prog_id id);
 };
 
 struct ta_funcs {
@@ -319,6 +335,14 @@ struct psp_runtime_scpm_entry {
 	enum psp_runtime_scpm_authentication scpm_status;
 };
 
+#if defined(CONFIG_DEBUG_FS)
+struct spirom_bo {
+	struct amdgpu_bo *bo;
+	uint64_t mc_addr;
+	void *cpu_addr;
+};
+#endif
+
 struct psp_context {
 	struct amdgpu_device		*adev;
 	struct psp_ring			km_ring;
@@ -406,6 +430,9 @@ struct psp_context {
 	char				*vbflash_tmp_buf;
 	size_t				vbflash_image_size;
 	bool				vbflash_done;
+#if defined(CONFIG_DEBUG_FS)
+	struct spirom_bo *spirom_dump_trip;
+#endif
 };
 
 struct amdgpu_psp_funcs {
@@ -464,6 +491,10 @@ struct amdgpu_psp_funcs {
 	((psp)->funcs->update_spirom ? \
 	(psp)->funcs->update_spirom((psp), fw_pri_mc_addr) : -EINVAL)
 
+#define psp_dump_spirom(psp, fw_pri_mc_addr) \
+	((psp)->funcs->dump_spirom ? \
+	(psp)->funcs->dump_spirom((psp), fw_pri_mc_addr) : -EINVAL)
+
 #define psp_vbflash_status(psp) \
 	((psp)->funcs->vbflash_stat ? \
 	(psp)->funcs->vbflash_stat((psp)) : -EINVAL)
@@ -475,6 +506,10 @@ struct amdgpu_psp_funcs {
 #define psp_is_aux_sos_load_required(psp) \
 	((psp)->funcs->is_aux_sos_load_required ? (psp)->funcs->is_aux_sos_load_required((psp)) : 0)
 
+#define psp_reg_program_no_ring(psp, val, id) \
+	((psp)->funcs->reg_program_no_ring ? \
+	(psp)->funcs->reg_program_no_ring((psp), val, id) : -EINVAL)
+
 extern const struct amd_ip_funcs psp_ip_funcs;
 
 extern const struct amdgpu_ip_block_version psp_v3_1_ip_block;
@@ -569,5 +604,9 @@ bool amdgpu_psp_get_ras_capability(struct psp_context *psp);
 int psp_config_sq_perfmon(struct psp_context *psp, uint32_t xcp_id,
 	bool core_override_enable, bool reg_override_enable, bool perfmon_override_enable);
 bool amdgpu_psp_tos_reload_needed(struct amdgpu_device *adev);
+int amdgpu_psp_reg_program_no_ring(struct psp_context *psp, uint32_t val,
+				   enum psp_reg_prog_id id);
+void amdgpu_psp_debugfs_init(struct amdgpu_device *adev);
+
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 443409d4f4b0..dc07936d2fcb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1498,6 +1498,9 @@ int amdgpu_ras_reset_error_count(struct amdgpu_device *adev,
 	    !amdgpu_ras_get_aca_debug_mode(adev))
 		return -EOPNOTSUPP;
 
+	if (amdgpu_sriov_vf(adev))
+		return -EOPNOTSUPP;
+
 	/* skip ras error reset in gpu reset */
 	if ((amdgpu_in_reset(adev) || amdgpu_ras_in_recovery(adev)) &&
 	    ((smu_funcs && smu_funcs->set_debug_mode) ||
@@ -2161,7 +2164,7 @@ void amdgpu_ras_interrupt_fatal_error_handler(struct amdgpu_device *adev)
 	/* Fatal error events are handled on host side */
 	if (amdgpu_sriov_vf(adev))
 		return;
-	/**
+	/*
 	 * If the current interrupt is caused by a non-fatal RAS error, skip
 	 * check for fatal error. For fatal errors, FED status of all devices
 	 * in XGMI hive gets set when the first device gets fatal error
@@ -2886,6 +2889,7 @@ static int __amdgpu_ras_convert_rec_from_rom(struct amdgpu_device *adev,
 		if (amdgpu_ras_mca2pa_by_idx(adev, bps, err_data))
 			return -EINVAL;
 	}
+
 	return __amdgpu_ras_restore_bad_pages(adev, err_data->err_addr,
 									adev->umc.retire_unit);
 }
@@ -2900,7 +2904,7 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
 			&adev->psp.ras_context.ras->eeprom_control;
 	enum amdgpu_memory_partition nps = AMDGPU_NPS1_PARTITION_MODE;
 	int ret = 0;
-	uint32_t i;
+	uint32_t i = 0;
 
 	if (!con || !con->eh_data || !bps || pages <= 0)
 		return 0;
@@ -2921,34 +2925,36 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
 	mutex_lock(&con->recovery_lock);
 
 	if (from_rom) {
-		for (i = 0; i < pages; i++) {
-			if (control->ras_num_recs - i >= adev->umc.retire_unit) {
-				if ((bps[i].address == bps[i + 1].address) &&
-				    (bps[i].mem_channel == bps[i + 1].mem_channel)) {
-					//deal with retire_unit records a time
-					ret = __amdgpu_ras_convert_rec_array_from_rom(adev,
-									&bps[i], &err_data, nps);
-					if (ret)
-						goto free;
-					i += (adev->umc.retire_unit - 1);
+		/* there is no pa recs in V3, so skip pa recs processing */
+		if (control->tbl_hdr.version < RAS_TABLE_VER_V3) {
+			for (i = 0; i < pages; i++) {
+				if (control->ras_num_recs - i >= adev->umc.retire_unit) {
+					if ((bps[i].address == bps[i + 1].address) &&
+						(bps[i].mem_channel == bps[i + 1].mem_channel)) {
+						/* deal with retire_unit records a time */
+						ret = __amdgpu_ras_convert_rec_array_from_rom(adev,
+										&bps[i], &err_data, nps);
+						if (ret)
+							control->ras_num_bad_pages -= adev->umc.retire_unit;
+						i += (adev->umc.retire_unit - 1);
+					} else {
+						break;
+					}
 				} else {
 					break;
 				}
-			} else {
-				break;
 			}
 		}
 		for (; i < pages; i++) {
 			ret = __amdgpu_ras_convert_rec_from_rom(adev,
 				&bps[i], &err_data, nps);
 			if (ret)
-				goto free;
+				control->ras_num_bad_pages -= adev->umc.retire_unit;
 		}
 	} else {
 		ret = __amdgpu_ras_restore_bad_pages(adev, bps, pages);
 	}
 
-free:
 	if (from_rom)
 		kfree(err_data.err_addr);
 	mutex_unlock(&con->recovery_lock);
@@ -3037,21 +3043,28 @@ static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev)
 		dev_err(adev->dev, "Failed to load EEPROM table records!");
 	} else {
 		if (adev->umc.ras && adev->umc.ras->convert_ras_err_addr) {
-			for (i = 0; i < control->ras_num_recs; i++) {
-				if ((control->ras_num_recs - i) >= adev->umc.retire_unit) {
-					if ((bps[i].address == bps[i + 1].address) &&
-						(bps[i].mem_channel == bps[i + 1].mem_channel)) {
-						control->ras_num_pa_recs += adev->umc.retire_unit;
-						i += (adev->umc.retire_unit - 1);
+			/*In V3, there is no pa recs, and some cases(when address==0) may be parsed
+			as pa recs, so add verion check to avoid it.
+			*/
+			if (control->tbl_hdr.version < RAS_TABLE_VER_V3) {
+				for (i = 0; i < control->ras_num_recs; i++) {
+					if ((control->ras_num_recs - i) >= adev->umc.retire_unit) {
+						if ((bps[i].address == bps[i + 1].address) &&
+							(bps[i].mem_channel == bps[i + 1].mem_channel)) {
+							control->ras_num_pa_recs += adev->umc.retire_unit;
+							i += (adev->umc.retire_unit - 1);
+						} else {
+							control->ras_num_mca_recs +=
+										(control->ras_num_recs - i);
+							break;
+						}
 					} else {
-						control->ras_num_mca_recs +=
-									(control->ras_num_recs - i);
+						control->ras_num_mca_recs += (control->ras_num_recs - i);
 						break;
 					}
-				} else {
-					control->ras_num_mca_recs += (control->ras_num_recs - i);
-					break;
 				}
+			} else {
+				control->ras_num_mca_recs = control->ras_num_recs;
 			}
 		}
 
@@ -3460,6 +3473,10 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev)
 	if (!adev->umc.ras || !adev->umc.ras->convert_ras_err_addr)
 		control->ras_num_pa_recs = control->ras_num_recs;
 
+	if (adev->umc.ras &&
+	    adev->umc.ras->get_retire_flip_bits)
+		adev->umc.ras->get_retire_flip_bits(adev);
+
 	if (control->ras_num_recs) {
 		ret = amdgpu_ras_load_bad_pages(adev);
 		if (ret)
@@ -3793,10 +3810,12 @@ init_ras_enabled_flag:
 		adev->ras_hw_enabled & amdgpu_ras_mask;
 
 	/* aca is disabled by default except for psp v13_0_6/v13_0_12/v13_0_14 */
-	adev->aca.is_enabled =
-		(amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 6) ||
-		 amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 12) ||
-		 amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 14));
+	if (!amdgpu_sriov_vf(adev)) {
+		adev->aca.is_enabled =
+			(amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 6) ||
+			amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 12) ||
+			amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 14));
+	}
 
 	/* bad page feature is not applicable to specific app platform */
 	if (adev->gmc.is_app_apu &&
@@ -4479,8 +4498,11 @@ void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev)
 		enum ras_event_type type = RAS_EVENT_TYPE_FATAL;
 		u64 event_id;
 
-		if (amdgpu_ras_mark_ras_event(adev, type))
+		if (amdgpu_ras_mark_ras_event(adev, type)) {
+			dev_err(adev->dev,
+				"uncorrectable hardware error (ERREVENT_ATHUB_INTERRUPT) detected!\n");
 			return;
+		}
 
 		event_id = amdgpu_ras_acquire_event_id(adev, type);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index 0ea7cfaf3587..2c58e09e56f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -418,6 +418,7 @@ static void amdgpu_ras_set_eeprom_table_version(struct amdgpu_ras_eeprom_control
 		hdr->version = RAS_TABLE_VER_V2_1;
 		return;
 	case IP_VERSION(12, 0, 0):
+	case IP_VERSION(12, 5, 0):
 		hdr->version = RAS_TABLE_VER_V3;
 		return;
 	default:
@@ -1392,17 +1393,39 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control)
 
 	__decode_table_header_from_buf(hdr, buf);
 
-	if (hdr->version >= RAS_TABLE_VER_V2_1) {
+	if (hdr->header != RAS_TABLE_HDR_VAL &&
+	    hdr->header != RAS_TABLE_HDR_BAD) {
+		dev_info(adev->dev, "Creating a new EEPROM table");
+		return amdgpu_ras_eeprom_reset_table(control);
+	}
+
+	switch (hdr->version) {
+	case RAS_TABLE_VER_V2_1:
+	case RAS_TABLE_VER_V3:
 		control->ras_num_recs = RAS_NUM_RECS_V2_1(hdr);
 		control->ras_record_offset = RAS_RECORD_START_V2_1;
 		control->ras_max_record_count = RAS_MAX_RECORD_COUNT_V2_1;
-	} else {
+		break;
+	case RAS_TABLE_VER_V1:
 		control->ras_num_recs = RAS_NUM_RECS(hdr);
 		control->ras_record_offset = RAS_RECORD_START;
 		control->ras_max_record_count = RAS_MAX_RECORD_COUNT;
+		break;
+	default:
+		dev_err(adev->dev,
+			"RAS header invalid, unsupported version: %u",
+			hdr->version);
+		return -EINVAL;
 	}
-	control->ras_fri = RAS_OFFSET_TO_INDEX(control, hdr->first_rec_offset);
 
+	if (control->ras_num_recs > control->ras_max_record_count) {
+		dev_err(adev->dev,
+			"RAS header invalid, records in header: %u max allowed :%u",
+			control->ras_num_recs, control->ras_max_record_count);
+		return -EINVAL;
+	}
+
+	control->ras_fri = RAS_OFFSET_TO_INDEX(control, hdr->first_rec_offset);
 	control->ras_num_mca_recs = 0;
 	control->ras_num_pa_recs = 0;
 	return 0;
@@ -1413,7 +1436,7 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control)
 	struct amdgpu_device *adev = to_amdgpu_device(control);
 	struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr;
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
-	int res;
+	int res = 0;
 
 	if (!__is_ras_eeprom_supported(adev))
 		return 0;
@@ -1494,10 +1517,6 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control)
 					 "User defined threshold is set, runtime service will be halt when threshold is reached\n");
 			}
 		}
-	} else {
-		DRM_INFO("Creating a new EEPROM table");
-
-		res = amdgpu_ras_eeprom_reset_table(control);
 	}
 
 	return res < 0 ? res : 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 59acdbfe28d8..426834806fbf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -187,14 +187,10 @@ void amdgpu_ring_undo(struct amdgpu_ring *ring)
 }
 
 #define amdgpu_ring_get_gpu_addr(ring, offset)				\
-	(ring->is_mes_queue ?						\
-	 (ring->mes_ctx->meta_data_gpu_addr + offset) :			\
-	 (ring->adev->wb.gpu_addr + offset * 4))
+	 (ring->adev->wb.gpu_addr + offset * 4)
 
 #define amdgpu_ring_get_cpu_addr(ring, offset)				\
-	(ring->is_mes_queue ?						\
-	 (void *)((uint8_t *)(ring->mes_ctx->meta_data_ptr) + offset) : \
-	 (&ring->adev->wb.wb[offset]))
+	 (&ring->adev->wb.wb[offset])
 
 /**
  * amdgpu_ring_init - init driver ring struct.
@@ -243,57 +239,42 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		ring->sched_score = sched_score;
 		ring->vmid_wait = dma_fence_get_stub();
 
-		if (!ring->is_mes_queue) {
-			ring->idx = adev->num_rings++;
-			adev->rings[ring->idx] = ring;
-		}
+		ring->idx = adev->num_rings++;
+		adev->rings[ring->idx] = ring;
 
 		r = amdgpu_fence_driver_init_ring(ring);
 		if (r)
 			return r;
 	}
 
-	if (ring->is_mes_queue) {
-		ring->rptr_offs = amdgpu_mes_ctx_get_offs(ring,
-				AMDGPU_MES_CTX_RPTR_OFFS);
-		ring->wptr_offs = amdgpu_mes_ctx_get_offs(ring,
-				AMDGPU_MES_CTX_WPTR_OFFS);
-		ring->fence_offs = amdgpu_mes_ctx_get_offs(ring,
-				AMDGPU_MES_CTX_FENCE_OFFS);
-		ring->trail_fence_offs = amdgpu_mes_ctx_get_offs(ring,
-				AMDGPU_MES_CTX_TRAIL_FENCE_OFFS);
-		ring->cond_exe_offs = amdgpu_mes_ctx_get_offs(ring,
-				AMDGPU_MES_CTX_COND_EXE_OFFS);
-	} else {
-		r = amdgpu_device_wb_get(adev, &ring->rptr_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &ring->rptr_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r);
+		return r;
+	}
 
-		r = amdgpu_device_wb_get(adev, &ring->wptr_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring wptr_offs wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &ring->wptr_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring wptr_offs wb alloc failed\n", r);
+		return r;
+	}
 
-		r = amdgpu_device_wb_get(adev, &ring->fence_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring fence_offs wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &ring->fence_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring fence_offs wb alloc failed\n", r);
+		return r;
+	}
 
-		r = amdgpu_device_wb_get(adev, &ring->trail_fence_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring trail_fence_offs wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &ring->trail_fence_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring trail_fence_offs wb alloc failed\n", r);
+		return r;
+	}
 
-		r = amdgpu_device_wb_get(adev, &ring->cond_exe_offs);
-		if (r) {
-			dev_err(adev->dev, "(%d) ring cond_exec_polling wb alloc failed\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &ring->cond_exe_offs);
+	if (r) {
+		dev_err(adev->dev, "(%d) ring cond_exec_polling wb alloc failed\n", r);
+		return r;
 	}
 
 	ring->fence_gpu_addr =
@@ -353,18 +334,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 	ring->cached_rptr = 0;
 
 	/* Allocate ring buffer */
-	if (ring->is_mes_queue) {
-		int offset = 0;
-
-		BUG_ON(ring->ring_size > PAGE_SIZE*4);
-
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_RING_OFFS);
-		ring->gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ring->ring = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		amdgpu_ring_clear_ring(ring);
-
-	} else if (ring->ring_obj == NULL) {
+	if (ring->ring_obj == NULL) {
 		r = amdgpu_bo_create_kernel(adev, ring->ring_size + ring->funcs->extra_dw, PAGE_SIZE,
 					    AMDGPU_GEM_DOMAIN_GTT,
 					    &ring->ring_obj,
@@ -401,32 +371,26 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 {
 
 	/* Not to finish a ring which is not initialized */
-	if (!(ring->adev) ||
-	    (!ring->is_mes_queue && !(ring->adev->rings[ring->idx])))
+	if (!(ring->adev) || !(ring->adev->rings[ring->idx]))
 		return;
 
 	ring->sched.ready = false;
 
-	if (!ring->is_mes_queue) {
-		amdgpu_device_wb_free(ring->adev, ring->rptr_offs);
-		amdgpu_device_wb_free(ring->adev, ring->wptr_offs);
+	amdgpu_device_wb_free(ring->adev, ring->rptr_offs);
+	amdgpu_device_wb_free(ring->adev, ring->wptr_offs);
 
-		amdgpu_device_wb_free(ring->adev, ring->cond_exe_offs);
-		amdgpu_device_wb_free(ring->adev, ring->fence_offs);
+	amdgpu_device_wb_free(ring->adev, ring->cond_exe_offs);
+	amdgpu_device_wb_free(ring->adev, ring->fence_offs);
 
-		amdgpu_bo_free_kernel(&ring->ring_obj,
-				      &ring->gpu_addr,
-				      (void **)&ring->ring);
-	} else {
-		kfree(ring->fence_drv.fences);
-	}
+	amdgpu_bo_free_kernel(&ring->ring_obj,
+			      &ring->gpu_addr,
+			      (void **)&ring->ring);
 
 	dma_fence_put(ring->vmid_wait);
 	ring->vmid_wait = NULL;
 	ring->me = 0;
 
-	if (!ring->is_mes_queue)
-		ring->adev->rings[ring->idx] = NULL;
+	ring->adev->rings[ring->idx] = NULL;
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index bb2b66385223..b95b47110769 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -164,8 +164,24 @@ void amdgpu_fence_update_start_timestamp(struct amdgpu_ring *ring, uint32_t seq,
 
 /* provided by hw blocks that expose a ring buffer for commands */
 struct amdgpu_ring_funcs {
+	/**
+	 * @type:
+	 *
+	 * GFX, Compute, SDMA, UVD, VCE, VCN, VPE, KIQ, MES, UMSCH, and CPER
+	 * use ring buffers. The type field just identifies which component the
+	 * ring buffer is associated with.
+	 */
 	enum amdgpu_ring_type	type;
 	uint32_t		align_mask;
+
+	/**
+	 * @nop:
+	 *
+	 * Every block in the amdgpu has no-op instructions (e.g., GFX 10
+	 * uses PACKET3(PACKET3_NOP, 0x3FFF), VCN 5 uses VCN_ENC_CMD_NO_OP,
+	 * etc). This field receives the specific no-op for the component
+	 * that initializes the ring.
+	 */
 	u32			nop;
 	bool			support_64bit_ptrs;
 	bool			no_user_fence;
@@ -241,6 +257,9 @@ struct amdgpu_ring_funcs {
 	bool (*is_guilty)(struct amdgpu_ring *ring);
 };
 
+/**
+ * amdgpu_ring - Holds ring information
+ */
 struct amdgpu_ring {
 	struct amdgpu_device		*adev;
 	const struct amdgpu_ring_funcs	*funcs;
@@ -252,13 +271,61 @@ struct amdgpu_ring {
 	unsigned		rptr_offs;
 	u64			rptr_gpu_addr;
 	volatile u32		*rptr_cpu_addr;
+
+	/**
+	 * @wptr:
+	 *
+	 * This is part of the Ring buffer implementation and represents the
+	 * write pointer. The wptr determines where the host has written.
+	 */
 	u64			wptr;
+
+	/**
+	 * @wptr_old:
+	 *
+	 * Before update wptr with the new value, usually the old value is
+	 * stored in the wptr_old.
+	 */
 	u64			wptr_old;
 	unsigned		ring_size;
+
+	/**
+	 * @max_dw:
+	 *
+	 * Maximum number of DWords for ring allocation. This information is
+	 * provided at the ring initialization time, and each IP block can
+	 * specify a specific value. Check places that invoke
+	 * amdgpu_ring_init() to see the maximum size per block.
+	 */
 	unsigned		max_dw;
+
+	/**
+	 * @count_dw:
+	 *
+	 * This value starts with the maximum amount of DWords supported by the
+	 * ring. This value is updated based on the ring manipulation.
+	 */
 	int			count_dw;
 	uint64_t		gpu_addr;
+
+	/**
+	 * @ptr_mask:
+	 *
+	 * Some IPs provide support for 64-bit pointers and others for 32-bit
+	 * only; this behavior is component-specific and defined by the field
+	 * support_64bit_ptr. If the IP block supports 64-bits, the mask
+	 * 0xffffffffffffffff is set; otherwise, this value assumes buf_mask.
+	 * Notice that this field is used to keep wptr under a valid range.
+	 */
 	uint64_t		ptr_mask;
+
+	/**
+	 * @buf_mask:
+	 *
+	 * Buffer mask is a value used to keep wptr count under its
+	 * thresholding. Buffer mask initialized during the ring buffer
+	 * initialization time, and it is defined as (ring_size / 4) -1.
+	 */
 	uint32_t		buf_mask;
 	u32			idx;
 	u32			xcc_id;
@@ -276,6 +343,13 @@ struct amdgpu_ring {
 	bool			use_pollmem;
 	unsigned		wptr_offs;
 	u64			wptr_gpu_addr;
+
+	/**
+	 * @wptr_cpu_addr:
+	 *
+	 * This is the CPU address pointer in the writeback slot. This is used
+	 * to commit changes to the GPU.
+	 */
 	volatile u32		*wptr_cpu_addr;
 	unsigned		fence_offs;
 	u64			fence_gpu_addr;
@@ -297,20 +371,15 @@ struct amdgpu_ring {
 	struct dma_fence	*vmid_wait;
 	bool			has_compute_vm_bug;
 	bool			no_scheduler;
+	bool			no_user_submission;
 	int			hw_prio;
 	unsigned 		num_hw_submission;
 	atomic_t		*sched_score;
 
-	/* used for mes */
-	bool			is_mes_queue;
-	uint32_t		hw_queue_id;
-	struct amdgpu_mes_ctx_data *mes_ctx;
-
 	bool            is_sw_ring;
 	unsigned int    entry_index;
 	/* store the cached rptr to restore after reset */
 	uint64_t cached_rptr;
-
 };
 
 #define amdgpu_ring_parse_cs(r, p, job, ib) ((r)->funcs->parse_cs((p), (job), (ib)))
@@ -435,15 +504,6 @@ static inline void amdgpu_ring_patch_cond_exec(struct amdgpu_ring *ring,
 	ring->ring[offset] = cur - offset;
 }
 
-#define amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset)			\
-	(ring->is_mes_queue && ring->mes_ctx ?				\
-	 (ring->mes_ctx->meta_data_gpu_addr + offset) : 0)
-
-#define amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset)			\
-	(ring->is_mes_queue && ring->mes_ctx ?				\
-	 (void *)((uint8_t *)(ring->mes_ctx->meta_data_ptr) + offset) : \
-	 NULL)
-
 int amdgpu_ring_test_helper(struct amdgpu_ring *ring);
 
 void amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
index fce22d3f816b..c210625be220 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
@@ -237,6 +237,20 @@ struct amdgpu_rlc_funcs {
 	void (*unset_safe_mode)(struct amdgpu_device *adev, int xcc_id);
 	int  (*init)(struct amdgpu_device *adev);
 	u32  (*get_csb_size)(struct amdgpu_device *adev);
+
+	/**
+	 * @get_csb_buffer: Get the clear state to be put into the hardware.
+	 *
+	 * The parameter adev is used to get the CS data and other gfx info,
+	 * and buffer is the RLC CS pointer
+	 *
+	 * Sometimes, the user space puts a request to clear the state in the
+	 * command buffer; this function provides the clear state that gets put
+	 * into the hardware. Note that the driver programs Clear State
+	 * Indirect Buffer (CSB) explicitly when it sets up the kernel rings,
+	 * and it also provides a pointer to it which is used by the firmware
+	 * to load the clear state in some cases.
+	 */
 	void (*get_csb_buffer)(struct amdgpu_device *adev, volatile u32 *buffer);
 	int  (*get_cp_table_num)(struct amdgpu_device *adev);
 	int  (*resume)(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
index 529c9696c2f3..6716ac281c49 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
@@ -26,6 +26,8 @@
 #include "amdgpu_sdma.h"
 #include "amdgpu_ras.h"
 #include "amdgpu_reset.h"
+#include "gc/gc_10_1_0_offset.h"
+#include "gc/gc_10_3_0_sh_mask.h"
 
 #define AMDGPU_CSA_SDMA_SIZE 64
 /* SDMA CSA reside in the 3rd page of CSA */
@@ -76,22 +78,14 @@ uint64_t amdgpu_sdma_get_csa_mc_addr(struct amdgpu_ring *ring,
 	if (amdgpu_sriov_vf(adev) || vmid == 0 || !adev->gfx.mcbp)
 		return 0;
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
+	r = amdgpu_sdma_get_index_from_ring(ring, &index);
 
-		offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				  sdma[ring->idx].sdma_meta_data);
-		csa_mc_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-	} else {
-		r = amdgpu_sdma_get_index_from_ring(ring, &index);
-
-		if (r || index > 31)
-			csa_mc_addr = 0;
-		else
-			csa_mc_addr = amdgpu_csa_vaddr(adev) +
-				AMDGPU_CSA_SDMA_OFFSET +
-				index * AMDGPU_CSA_SDMA_SIZE;
-	}
+	if (r || index > 31)
+		csa_mc_addr = 0;
+	else
+		csa_mc_addr = amdgpu_csa_vaddr(adev) +
+			AMDGPU_CSA_SDMA_OFFSET +
+			index * AMDGPU_CSA_SDMA_SIZE;
 
 	return csa_mc_addr;
 }
@@ -537,28 +531,38 @@ bool amdgpu_sdma_is_shared_inv_eng(struct amdgpu_device *adev, struct amdgpu_rin
 		return false;
 }
 
-/**
- * amdgpu_sdma_register_on_reset_callbacks - Register SDMA reset callbacks
- * @funcs: Pointer to the callback structure containing pre_reset and post_reset functions
- *
- * This function allows KFD and AMDGPU to register their own callbacks for handling
- * pre-reset and post-reset operations for engine reset. These are needed because engine
- * reset will stop all queues on that engine.
- */
-void amdgpu_sdma_register_on_reset_callbacks(struct amdgpu_device *adev, struct sdma_on_reset_funcs *funcs)
+static int amdgpu_sdma_soft_reset(struct amdgpu_device *adev, u32 instance_id)
 {
-	if (!funcs)
-		return;
-
-	/* Ensure the reset_callback_list is initialized */
-	if (!adev->sdma.reset_callback_list.next) {
-		INIT_LIST_HEAD(&adev->sdma.reset_callback_list);
+	struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id];
+	int r = -EOPNOTSUPP;
+
+	switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) {
+	case IP_VERSION(4, 4, 2):
+	case IP_VERSION(4, 4, 4):
+	case IP_VERSION(4, 4, 5):
+		/* For SDMA 4.x, use the existing DPM interface for backward compatibility */
+		r = amdgpu_dpm_reset_sdma(adev, 1 << instance_id);
+		break;
+	case IP_VERSION(5, 0, 0):
+	case IP_VERSION(5, 0, 1):
+	case IP_VERSION(5, 0, 2):
+	case IP_VERSION(5, 0, 5):
+	case IP_VERSION(5, 2, 0):
+	case IP_VERSION(5, 2, 2):
+	case IP_VERSION(5, 2, 4):
+	case IP_VERSION(5, 2, 5):
+	case IP_VERSION(5, 2, 6):
+	case IP_VERSION(5, 2, 3):
+	case IP_VERSION(5, 2, 1):
+	case IP_VERSION(5, 2, 7):
+		if (sdma_instance->funcs->soft_reset_kernel_queue)
+			r = sdma_instance->funcs->soft_reset_kernel_queue(adev, instance_id);
+		break;
+	default:
+		break;
 	}
-	/* Initialize the list node in the callback structure */
-	INIT_LIST_HEAD(&funcs->list);
 
-	/* Add the callback structure to the global list */
-	list_add_tail(&funcs->list, &adev->sdma.reset_callback_list);
+	return r;
 }
 
 /**
@@ -566,16 +570,10 @@ void amdgpu_sdma_register_on_reset_callbacks(struct amdgpu_device *adev, struct
  * @adev: Pointer to the AMDGPU device
  * @instance_id: ID of the SDMA engine instance to reset
  *
- * This function performs the following steps:
- * 1. Calls all registered pre_reset callbacks to allow KFD and AMDGPU to save their state.
- * 2. Resets the specified SDMA engine instance.
- * 3. Calls all registered post_reset callbacks to allow KFD and AMDGPU to restore their state.
- *
  * Returns: 0 on success, or a negative error code on failure.
  */
 int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id)
 {
-	struct sdma_on_reset_funcs *funcs;
 	int ret = 0;
 	struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id];
 	struct amdgpu_ring *gfx_ring = &sdma_instance->ring;
@@ -597,38 +595,18 @@ int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id)
 		page_sched_stopped = true;
 	}
 
-	/* Invoke all registered pre_reset callbacks */
-	list_for_each_entry(funcs, &adev->sdma.reset_callback_list, list) {
-		if (funcs->pre_reset) {
-			ret = funcs->pre_reset(adev, instance_id);
-			if (ret) {
-				dev_err(adev->dev,
-				"beforeReset callback failed for instance %u: %d\n",
-					instance_id, ret);
-				goto exit;
-			}
-		}
-	}
+	if (sdma_instance->funcs->stop_kernel_queue)
+		sdma_instance->funcs->stop_kernel_queue(gfx_ring);
 
 	/* Perform the SDMA reset for the specified instance */
-	ret = amdgpu_dpm_reset_sdma(adev, 1 << instance_id);
+	ret = amdgpu_sdma_soft_reset(adev, instance_id);
 	if (ret) {
 		dev_err(adev->dev, "Failed to reset SDMA instance %u\n", instance_id);
 		goto exit;
 	}
 
-	/* Invoke all registered post_reset callbacks */
-	list_for_each_entry(funcs, &adev->sdma.reset_callback_list, list) {
-		if (funcs->post_reset) {
-			ret = funcs->post_reset(adev, instance_id);
-			if (ret) {
-				dev_err(adev->dev,
-				"afterReset callback failed for instance %u: %d\n",
-					instance_id, ret);
-				goto exit;
-			}
-		}
-	}
+	if (sdma_instance->funcs->start_kernel_queue)
+		sdma_instance->funcs->start_kernel_queue(gfx_ring);
 
 exit:
 	/* Restart the scheduler's work queue for the GFX and page rings
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index 47d56fd0589f..5605921212f0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -50,6 +50,12 @@ enum amdgpu_sdma_irq {
 
 #define NUM_SDMA(x) hweight32(x)
 
+struct amdgpu_sdma_funcs {
+	int (*stop_kernel_queue)(struct amdgpu_ring *ring);
+	int (*start_kernel_queue)(struct amdgpu_ring *ring);
+	int (*soft_reset_kernel_queue)(struct amdgpu_device *adev, u32 instance_id);
+};
+
 struct amdgpu_sdma_instance {
 	/* SDMA firmware */
 	const struct firmware	*fw;
@@ -68,7 +74,7 @@ struct amdgpu_sdma_instance {
 	/* track guilty state of GFX and PAGE queues */
 	bool			gfx_guilty;
 	bool			page_guilty;
-
+	const struct amdgpu_sdma_funcs   *funcs;
 };
 
 enum amdgpu_sdma_ras_memory_id {
@@ -103,13 +109,6 @@ struct amdgpu_sdma_ras {
 	struct amdgpu_ras_block_object ras_block;
 };
 
-struct sdma_on_reset_funcs {
-	int (*pre_reset)(struct amdgpu_device *adev, uint32_t instance_id);
-	int (*post_reset)(struct amdgpu_device *adev, uint32_t instance_id);
-	/* Linked list node to store this structure in a list; */
-	struct list_head list;
-};
-
 struct amdgpu_sdma {
 	struct amdgpu_sdma_instance instance[AMDGPU_MAX_SDMA_INSTANCES];
 	struct amdgpu_irq_src	trap_irq;
@@ -131,6 +130,8 @@ struct amdgpu_sdma {
 	uint32_t		*ip_dump;
 	uint32_t 		supported_reset;
 	struct list_head	reset_callback_list;
+	bool			no_user_submission;
+	bool			disable_uq;
 };
 
 /*
@@ -170,7 +171,6 @@ struct amdgpu_buffer_funcs {
 				 uint32_t byte_count);
 };
 
-void amdgpu_sdma_register_on_reset_callbacks(struct amdgpu_device *adev, struct sdma_on_reset_funcs *funcs);
 int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id);
 
 #define amdgpu_emit_copy_buffer(adev, ib, s, d, b, t) (adev)->mman.buffer_funcs->emit_copy_buffer((ib),  (s), (d), (b), (t))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c
index e22cb2b5cd92..3939761be31c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c
@@ -45,7 +45,11 @@
  */
 static inline u64 amdgpu_seq64_get_va_base(struct amdgpu_device *adev)
 {
-	return AMDGPU_VA_RESERVED_SEQ64_START(adev);
+	u64 addr = AMDGPU_VA_RESERVED_SEQ64_START(adev);
+
+	addr = amdgpu_gmc_sign_extend(addr);
+
+	return addr;
 }
 
 /**
@@ -63,9 +67,9 @@ static inline u64 amdgpu_seq64_get_va_base(struct amdgpu_device *adev)
 int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		     struct amdgpu_bo_va **bo_va)
 {
+	u64 seq64_addr, va_flags;
 	struct amdgpu_bo *bo;
 	struct drm_exec exec;
-	u64 seq64_addr;
 	int r;
 
 	bo = adev->seq64.sbo;
@@ -88,9 +92,11 @@ int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		goto error;
 	}
 
-	seq64_addr = amdgpu_seq64_get_va_base(adev);
+	seq64_addr = amdgpu_seq64_get_va_base(adev) & AMDGPU_GMC_HOLE_MASK;
+
+	va_flags = amdgpu_gem_va_map_flags(adev, AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_MTYPE_UC);
 	r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0, AMDGPU_VA_RESERVED_SEQ64_SIZE,
-			     AMDGPU_PTE_READABLE);
+			     va_flags);
 	if (r) {
 		DRM_ERROR("failed to do bo_map on userq sem, err=%d\n", r);
 		amdgpu_vm_bo_del(adev, *bo_va);
@@ -156,6 +162,7 @@ error:
  *
  * @adev: amdgpu_device pointer
  * @va: VA to access the seq in process address space
+ * @gpu_addr: GPU address to access the seq
  * @cpu_addr: CPU address to access the seq
  *
  * Alloc a 64 bit memory from seq64 pool.
@@ -163,7 +170,8 @@ error:
  * Returns:
  * 0 on success or a negative error code on failure
  */
-int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *va, u64 **cpu_addr)
+int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *va,
+		       u64 *gpu_addr, u64 **cpu_addr)
 {
 	unsigned long bit_pos;
 
@@ -172,7 +180,12 @@ int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *va, u64 **cpu_addr)
 		return -ENOSPC;
 
 	__set_bit(bit_pos, adev->seq64.used);
+
 	*va = bit_pos * sizeof(u64) + amdgpu_seq64_get_va_base(adev);
+
+	if (gpu_addr)
+		*gpu_addr = bit_pos * sizeof(u64) + adev->seq64.gpu_addr;
+
 	*cpu_addr = bit_pos + adev->seq64.cpu_base_addr;
 
 	return 0;
@@ -233,7 +246,7 @@ int amdgpu_seq64_init(struct amdgpu_device *adev)
 	 */
 	r = amdgpu_bo_create_kernel(adev, AMDGPU_VA_RESERVED_SEQ64_SIZE,
 				    PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT,
-				    &adev->seq64.sbo, NULL,
+				    &adev->seq64.sbo, &adev->seq64.gpu_addr,
 				    (void **)&adev->seq64.cpu_base_addr);
 	if (r) {
 		dev_warn(adev->dev, "(%d) create seq64 failed\n", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h
index 4203b2ab318d..26a249aaaee1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h
@@ -32,13 +32,14 @@
 struct amdgpu_seq64 {
 	struct amdgpu_bo *sbo;
 	u32 num_sem;
+	u64 gpu_addr;
 	u64 *cpu_base_addr;
 	DECLARE_BITMAP(used, AMDGPU_MAX_SEQ64_SLOTS);
 };
 
 void amdgpu_seq64_fini(struct amdgpu_device *adev);
 int amdgpu_seq64_init(struct amdgpu_device *adev);
-int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *gpu_addr, u64 **cpu_addr);
+int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *va, u64 *gpu_addr, u64 **cpu_addr);
 void amdgpu_seq64_free(struct amdgpu_device *adev, u64 gpu_addr);
 int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		     struct amdgpu_bo_va **bo_va);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index 5576ed0b508f..d6ae9974c952 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -249,9 +249,8 @@ int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 
 	if (resv == NULL)
 		return -EINVAL;
-
-	/* TODO: Use DMA_RESV_USAGE_READ here */
-	dma_resv_for_each_fence(&cursor, resv, DMA_RESV_USAGE_BOOKKEEP, f) {
+	/* Implicitly sync only to KERNEL, WRITE and READ */
+	dma_resv_for_each_fence(&cursor, resv, DMA_RESV_USAGE_READ, f) {
 		dma_fence_chain_for_each(f, f) {
 			struct dma_fence *tmp = dma_fence_chain_contained(f);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 53b71e9d8076..9c5df35f05b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -2081,6 +2081,8 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	amdgpu_vram_mgr_fini(adev);
 	amdgpu_gtt_mgr_fini(adev);
 	amdgpu_preempt_mgr_fini(adev);
+	amdgpu_doorbell_fini(adev);
+
 	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_GDS);
 	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_GWS);
 	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_OA);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
index 3d9e9fdc10b4..4a72c2bbd49e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
@@ -767,6 +767,7 @@ FW_VERSION_ATTR(vcn_fw_version, 0444, vcn.fw_version);
 FW_VERSION_ATTR(dmcu_fw_version, 0444, dm.dmcu_fw_version);
 FW_VERSION_ATTR(mes_fw_version, 0444, mes.sched_version & AMDGPU_MES_VERSION_MASK);
 FW_VERSION_ATTR(mes_kiq_fw_version, 0444, mes.kiq_version & AMDGPU_MES_VERSION_MASK);
+FW_VERSION_ATTR(pldm_fw_version, 0444, firmware.pldm_version);
 
 static struct attribute *fw_attrs[] = {
 	&dev_attr_vce_fw_version.attr, &dev_attr_uvd_fw_version.attr,
@@ -781,7 +782,7 @@ static struct attribute *fw_attrs[] = {
 	&dev_attr_sdma2_fw_version.attr, &dev_attr_vcn_fw_version.attr,
 	&dev_attr_dmcu_fw_version.attr, &dev_attr_imu_fw_version.attr,
 	&dev_attr_mes_fw_version.attr, &dev_attr_mes_kiq_fw_version.attr,
-	NULL
+	&dev_attr_pldm_fw_version.attr, NULL
 };
 
 #define to_dev_attr(x) container_of(x, struct device_attribute, attr)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
index 4eedd92f000b..9e89c3487be5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
@@ -25,6 +25,8 @@
 
 #include "amdgpu_socbb.h"
 
+#define RS64_FW_UC_START_ADDR_LO 0x3000
+
 struct common_firmware_header {
 	uint32_t size_bytes; /* size of the entire header+image(s) in bytes */
 	uint32_t header_size_bytes; /* size of just the header in bytes */
@@ -600,6 +602,7 @@ struct amdgpu_firmware {
 
 	void *fw_buf_ptr;
 	uint64_t fw_buf_mc;
+	uint32_t pldm_version;
 };
 
 void amdgpu_ucode_print_mc_hdr(const struct common_firmware_header *hdr);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
index 0a1ef95b2866..8c6e55b5b967 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
@@ -529,6 +529,7 @@ int amdgpu_umc_lookup_bad_pages_in_a_row(struct amdgpu_device *adev,
 		pfns[i] = err_data.err_addr[i].retired_page;
 	}
 	ret = i;
+	adev->umc.err_addr_cnt = err_data.err_addr_cnt;
 
 out:
 	kfree(err_data.err_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
index 857693bcd8d4..29ce6b1d214a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
@@ -78,6 +78,18 @@
 #define UMC_NPS_SHIFT 40
 #define UMC_NPS_MASK 0xffULL
 
+/* three column bits and one row bit in MCA address flip
+ * in bad page retirement
+ */
+#define RETIRE_FLIP_BITS_NUM 4
+
+struct amdgpu_umc_flip_bits {
+	uint32_t flip_bits_in_pa[RETIRE_FLIP_BITS_NUM];
+	uint32_t flip_row_bit;
+	uint32_t r13_in_pa;
+	uint32_t bit_num;
+};
+
 typedef int (*umc_func)(struct amdgpu_device *adev, uint32_t node_inst,
 			uint32_t umc_inst, uint32_t ch_inst, void *data);
 
@@ -100,6 +112,7 @@ struct amdgpu_umc_ras {
 			bool dump_addr);
 	uint32_t (*get_die_id_from_pa)(struct amdgpu_device *adev,
 			uint64_t mca_addr, uint64_t retired_page);
+	void (*get_retire_flip_bits)(struct amdgpu_device *adev);
 };
 
 struct amdgpu_umc_funcs {
@@ -130,6 +143,10 @@ struct amdgpu_umc {
 
 	/* active mask for umc node instance */
 	unsigned long active_mask;
+
+	struct amdgpu_umc_flip_bits flip_bits;
+
+	unsigned long err_addr_cnt;
 };
 
 int amdgpu_umc_ras_sw_init(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
new file mode 100644
index 000000000000..295e7186e156
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -0,0 +1,924 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include <drm/drm_auth.h>
+#include <drm/drm_exec.h>
+#include <linux/pm_runtime.h>
+
+#include "amdgpu.h"
+#include "amdgpu_vm.h"
+#include "amdgpu_userq.h"
+#include "amdgpu_userq_fence.h"
+
+u32 amdgpu_userq_get_supported_ip_mask(struct amdgpu_device *adev)
+{
+	int i;
+	u32 userq_ip_mask = 0;
+
+	for (i = 0; i < AMDGPU_HW_IP_NUM; i++) {
+		if (adev->userq_funcs[i])
+			userq_ip_mask |= (1 << i);
+	}
+
+	return userq_ip_mask;
+}
+
+static int
+amdgpu_userq_unmap_helper(struct amdgpu_userq_mgr *uq_mgr,
+			  struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	const struct amdgpu_userq_funcs *userq_funcs =
+		adev->userq_funcs[queue->queue_type];
+	int r = 0;
+
+	if (queue->state == AMDGPU_USERQ_STATE_MAPPED) {
+		r = userq_funcs->unmap(uq_mgr, queue);
+		if (r)
+			queue->state = AMDGPU_USERQ_STATE_HUNG;
+		else
+			queue->state = AMDGPU_USERQ_STATE_UNMAPPED;
+	}
+	return r;
+}
+
+static int
+amdgpu_userq_map_helper(struct amdgpu_userq_mgr *uq_mgr,
+			struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	const struct amdgpu_userq_funcs *userq_funcs =
+		adev->userq_funcs[queue->queue_type];
+	int r = 0;
+
+	if (queue->state == AMDGPU_USERQ_STATE_UNMAPPED) {
+		r = userq_funcs->map(uq_mgr, queue);
+		if (r) {
+			queue->state = AMDGPU_USERQ_STATE_HUNG;
+		} else {
+			queue->state = AMDGPU_USERQ_STATE_MAPPED;
+		}
+	}
+	return r;
+}
+
+static void
+amdgpu_userq_wait_for_last_fence(struct amdgpu_userq_mgr *uq_mgr,
+				 struct amdgpu_usermode_queue *queue)
+{
+	struct dma_fence *f = queue->last_fence;
+	int ret;
+
+	if (f && !dma_fence_is_signaled(f)) {
+		ret = dma_fence_wait_timeout(f, true, msecs_to_jiffies(100));
+		if (ret <= 0)
+			drm_file_err(uq_mgr->file, "Timed out waiting for fence=%llu:%llu\n",
+				     f->context, f->seqno);
+	}
+}
+
+static void
+amdgpu_userq_cleanup(struct amdgpu_userq_mgr *uq_mgr,
+		     struct amdgpu_usermode_queue *queue,
+		     int queue_id)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	const struct amdgpu_userq_funcs *uq_funcs = adev->userq_funcs[queue->queue_type];
+
+	uq_funcs->mqd_destroy(uq_mgr, queue);
+	amdgpu_userq_fence_driver_free(queue);
+	idr_remove(&uq_mgr->userq_idr, queue_id);
+	kfree(queue);
+}
+
+int
+amdgpu_userq_active(struct amdgpu_userq_mgr *uq_mgr)
+{
+	struct amdgpu_usermode_queue *queue;
+	int queue_id;
+	int ret = 0;
+
+	mutex_lock(&uq_mgr->userq_mutex);
+	/* Resume all the queues for this process */
+	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id)
+		ret += queue->state == AMDGPU_USERQ_STATE_MAPPED;
+
+	mutex_unlock(&uq_mgr->userq_mutex);
+	return ret;
+}
+
+static struct amdgpu_usermode_queue *
+amdgpu_userq_find(struct amdgpu_userq_mgr *uq_mgr, int qid)
+{
+	return idr_find(&uq_mgr->userq_idr, qid);
+}
+
+void
+amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *uq_mgr,
+			     struct amdgpu_eviction_fence_mgr *evf_mgr)
+{
+	struct amdgpu_eviction_fence *ev_fence;
+
+retry:
+	/* Flush any pending resume work to create ev_fence */
+	flush_delayed_work(&uq_mgr->resume_work);
+
+	mutex_lock(&uq_mgr->userq_mutex);
+	spin_lock(&evf_mgr->ev_fence_lock);
+	ev_fence = evf_mgr->ev_fence;
+	spin_unlock(&evf_mgr->ev_fence_lock);
+	if (!ev_fence || dma_fence_is_signaled(&ev_fence->base)) {
+		mutex_unlock(&uq_mgr->userq_mutex);
+		/*
+		 * Looks like there was no pending resume work,
+		 * add one now to create a valid eviction fence
+		 */
+		schedule_delayed_work(&uq_mgr->resume_work, 0);
+		goto retry;
+	}
+}
+
+int amdgpu_userq_create_object(struct amdgpu_userq_mgr *uq_mgr,
+			       struct amdgpu_userq_obj *userq_obj,
+			       int size)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct amdgpu_bo_param bp;
+	int r;
+
+	memset(&bp, 0, sizeof(bp));
+	bp.byte_align = PAGE_SIZE;
+	bp.domain = AMDGPU_GEM_DOMAIN_GTT;
+	bp.flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
+		   AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
+	bp.type = ttm_bo_type_kernel;
+	bp.size = size;
+	bp.resv = NULL;
+	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
+
+	r = amdgpu_bo_create(adev, &bp, &userq_obj->obj);
+	if (r) {
+		drm_file_err(uq_mgr->file, "Failed to allocate BO for userqueue (%d)", r);
+		return r;
+	}
+
+	r = amdgpu_bo_reserve(userq_obj->obj, true);
+	if (r) {
+		drm_file_err(uq_mgr->file, "Failed to reserve BO to map (%d)", r);
+		goto free_obj;
+	}
+
+	r = amdgpu_ttm_alloc_gart(&(userq_obj->obj)->tbo);
+	if (r) {
+		drm_file_err(uq_mgr->file, "Failed to alloc GART for userqueue object (%d)", r);
+		goto unresv;
+	}
+
+	r = amdgpu_bo_kmap(userq_obj->obj, &userq_obj->cpu_ptr);
+	if (r) {
+		drm_file_err(uq_mgr->file, "Failed to map BO for userqueue (%d)", r);
+		goto unresv;
+	}
+
+	userq_obj->gpu_addr = amdgpu_bo_gpu_offset(userq_obj->obj);
+	amdgpu_bo_unreserve(userq_obj->obj);
+	memset(userq_obj->cpu_ptr, 0, size);
+	return 0;
+
+unresv:
+	amdgpu_bo_unreserve(userq_obj->obj);
+
+free_obj:
+	amdgpu_bo_unref(&userq_obj->obj);
+	return r;
+}
+
+void amdgpu_userq_destroy_object(struct amdgpu_userq_mgr *uq_mgr,
+				 struct amdgpu_userq_obj *userq_obj)
+{
+	amdgpu_bo_kunmap(userq_obj->obj);
+	amdgpu_bo_unref(&userq_obj->obj);
+}
+
+uint64_t
+amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr,
+				struct amdgpu_db_info *db_info,
+				struct drm_file *filp)
+{
+	uint64_t index;
+	struct drm_gem_object *gobj;
+	struct amdgpu_userq_obj *db_obj = db_info->db_obj;
+	int r, db_size;
+
+	gobj = drm_gem_object_lookup(filp, db_info->doorbell_handle);
+	if (gobj == NULL) {
+		drm_file_err(uq_mgr->file, "Can't find GEM object for doorbell\n");
+		return -EINVAL;
+	}
+
+	db_obj->obj = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj));
+	drm_gem_object_put(gobj);
+
+	r = amdgpu_bo_reserve(db_obj->obj, true);
+	if (r) {
+		drm_file_err(uq_mgr->file, "[Usermode queues] Failed to pin doorbell object\n");
+		goto unref_bo;
+	}
+
+	/* Pin the BO before generating the index, unpin in queue destroy */
+	r = amdgpu_bo_pin(db_obj->obj, AMDGPU_GEM_DOMAIN_DOORBELL);
+	if (r) {
+		drm_file_err(uq_mgr->file, "[Usermode queues] Failed to pin doorbell object\n");
+		goto unresv_bo;
+	}
+
+	switch (db_info->queue_type) {
+	case AMDGPU_HW_IP_GFX:
+	case AMDGPU_HW_IP_COMPUTE:
+	case AMDGPU_HW_IP_DMA:
+		db_size = sizeof(u64);
+		break;
+
+	case AMDGPU_HW_IP_VCN_ENC:
+		db_size = sizeof(u32);
+		db_info->doorbell_offset += AMDGPU_NAVI10_DOORBELL64_VCN0_1 << 1;
+		break;
+
+	case AMDGPU_HW_IP_VPE:
+		db_size = sizeof(u32);
+		db_info->doorbell_offset += AMDGPU_NAVI10_DOORBELL64_VPE << 1;
+		break;
+
+	default:
+		drm_file_err(uq_mgr->file, "[Usermode queues] IP %d not support\n",
+			     db_info->queue_type);
+		r = -EINVAL;
+		goto unpin_bo;
+	}
+
+	index = amdgpu_doorbell_index_on_bar(uq_mgr->adev, db_obj->obj,
+					     db_info->doorbell_offset, db_size);
+	drm_dbg_driver(adev_to_drm(uq_mgr->adev),
+		       "[Usermode queues] doorbell index=%lld\n", index);
+	amdgpu_bo_unreserve(db_obj->obj);
+	return index;
+
+unpin_bo:
+	amdgpu_bo_unpin(db_obj->obj);
+unresv_bo:
+	amdgpu_bo_unreserve(db_obj->obj);
+unref_bo:
+	amdgpu_bo_unref(&db_obj->obj);
+	return r;
+}
+
+static int
+amdgpu_userq_destroy(struct drm_file *filp, int queue_id)
+{
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_userq_mgr *uq_mgr = &fpriv->userq_mgr;
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct amdgpu_usermode_queue *queue;
+	int r = 0;
+
+	cancel_delayed_work_sync(&uq_mgr->resume_work);
+	mutex_lock(&uq_mgr->userq_mutex);
+
+	queue = amdgpu_userq_find(uq_mgr, queue_id);
+	if (!queue) {
+		drm_dbg_driver(adev_to_drm(uq_mgr->adev), "Invalid queue id to destroy\n");
+		mutex_unlock(&uq_mgr->userq_mutex);
+		return -EINVAL;
+	}
+	amdgpu_userq_wait_for_last_fence(uq_mgr, queue);
+	r = amdgpu_bo_reserve(queue->db_obj.obj, true);
+	if (!r) {
+		amdgpu_bo_unpin(queue->db_obj.obj);
+		amdgpu_bo_unreserve(queue->db_obj.obj);
+	}
+	amdgpu_bo_unref(&queue->db_obj.obj);
+	r = amdgpu_userq_unmap_helper(uq_mgr, queue);
+	amdgpu_userq_cleanup(uq_mgr, queue, queue_id);
+	mutex_unlock(&uq_mgr->userq_mutex);
+
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
+
+	return r;
+}
+
+static int amdgpu_userq_priority_permit(struct drm_file *filp,
+					int priority)
+{
+	if (priority < AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_HIGH)
+		return 0;
+
+	if (capable(CAP_SYS_NICE))
+		return 0;
+
+	if (drm_is_current_master(filp))
+		return 0;
+
+	return -EACCES;
+}
+
+static int
+amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args)
+{
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_userq_mgr *uq_mgr = &fpriv->userq_mgr;
+	struct amdgpu_device *adev = uq_mgr->adev;
+	const struct amdgpu_userq_funcs *uq_funcs;
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_db_info db_info;
+	bool skip_map_queue;
+	uint64_t index;
+	int qid, r = 0;
+	int priority =
+		(args->in.flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_MASK) >>
+		AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_SHIFT;
+
+	/* Usermode queues are only supported for GFX IP as of now */
+	if (args->in.ip_type != AMDGPU_HW_IP_GFX &&
+	    args->in.ip_type != AMDGPU_HW_IP_DMA &&
+	    args->in.ip_type != AMDGPU_HW_IP_COMPUTE) {
+		drm_file_err(uq_mgr->file, "Usermode queue doesn't support IP type %u\n",
+			     args->in.ip_type);
+		return -EINVAL;
+	}
+
+	r = amdgpu_userq_priority_permit(filp, priority);
+	if (r)
+		return r;
+
+	if ((args->in.flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE) &&
+	    (args->in.ip_type != AMDGPU_HW_IP_GFX) &&
+	    (args->in.ip_type != AMDGPU_HW_IP_COMPUTE) &&
+	    !amdgpu_is_tmz(adev)) {
+		drm_file_err(uq_mgr->file, "Secure only supported on GFX/Compute queues\n");
+		return -EINVAL;
+	}
+
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
+	if (r < 0) {
+		drm_file_err(uq_mgr->file, "pm_runtime_get_sync() failed for userqueue create\n");
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
+		return r;
+	}
+
+	/*
+	 * There could be a situation that we are creating a new queue while
+	 * the other queues under this UQ_mgr are suspended. So if there is any
+	 * resume work pending, wait for it to get done.
+	 *
+	 * This will also make sure we have a valid eviction fence ready to be used.
+	 */
+	mutex_lock(&adev->userq_mutex);
+	amdgpu_userq_ensure_ev_fence(&fpriv->userq_mgr, &fpriv->evf_mgr);
+
+	uq_funcs = adev->userq_funcs[args->in.ip_type];
+	if (!uq_funcs) {
+		drm_file_err(uq_mgr->file, "Usermode queue is not supported for this IP (%u)\n",
+			     args->in.ip_type);
+		r = -EINVAL;
+		goto unlock;
+	}
+
+	queue = kzalloc(sizeof(struct amdgpu_usermode_queue), GFP_KERNEL);
+	if (!queue) {
+		drm_file_err(uq_mgr->file, "Failed to allocate memory for queue\n");
+		r = -ENOMEM;
+		goto unlock;
+	}
+	queue->doorbell_handle = args->in.doorbell_handle;
+	queue->queue_type = args->in.ip_type;
+	queue->vm = &fpriv->vm;
+	queue->priority = priority;
+
+	db_info.queue_type = queue->queue_type;
+	db_info.doorbell_handle = queue->doorbell_handle;
+	db_info.db_obj = &queue->db_obj;
+	db_info.doorbell_offset = args->in.doorbell_offset;
+
+	/* Convert relative doorbell offset into absolute doorbell index */
+	index = amdgpu_userq_get_doorbell_index(uq_mgr, &db_info, filp);
+	if (index == (uint64_t)-EINVAL) {
+		drm_file_err(uq_mgr->file, "Failed to get doorbell for queue\n");
+		kfree(queue);
+		goto unlock;
+	}
+
+	queue->doorbell_index = index;
+	xa_init_flags(&queue->fence_drv_xa, XA_FLAGS_ALLOC);
+	r = amdgpu_userq_fence_driver_alloc(adev, queue);
+	if (r) {
+		drm_file_err(uq_mgr->file, "Failed to alloc fence driver\n");
+		goto unlock;
+	}
+
+	r = uq_funcs->mqd_create(uq_mgr, &args->in, queue);
+	if (r) {
+		drm_file_err(uq_mgr->file, "Failed to create Queue\n");
+		amdgpu_userq_fence_driver_free(queue);
+		kfree(queue);
+		goto unlock;
+	}
+
+
+	qid = idr_alloc(&uq_mgr->userq_idr, queue, 1, AMDGPU_MAX_USERQ_COUNT, GFP_KERNEL);
+	if (qid < 0) {
+		drm_file_err(uq_mgr->file, "Failed to allocate a queue id\n");
+		amdgpu_userq_fence_driver_free(queue);
+		uq_funcs->mqd_destroy(uq_mgr, queue);
+		kfree(queue);
+		r = -ENOMEM;
+		goto unlock;
+	}
+
+	/* don't map the queue if scheduling is halted */
+	if (adev->userq_halt_for_enforce_isolation &&
+	    ((queue->queue_type == AMDGPU_HW_IP_GFX) ||
+	     (queue->queue_type == AMDGPU_HW_IP_COMPUTE)))
+		skip_map_queue = true;
+	else
+		skip_map_queue = false;
+	if (!skip_map_queue) {
+		r = amdgpu_userq_map_helper(uq_mgr, queue);
+		if (r) {
+			drm_file_err(uq_mgr->file, "Failed to map Queue\n");
+			idr_remove(&uq_mgr->userq_idr, qid);
+			amdgpu_userq_fence_driver_free(queue);
+			uq_funcs->mqd_destroy(uq_mgr, queue);
+			kfree(queue);
+			goto unlock;
+		}
+	}
+
+
+	args->out.queue_id = qid;
+
+unlock:
+	mutex_unlock(&uq_mgr->userq_mutex);
+	mutex_unlock(&adev->userq_mutex);
+
+	return r;
+}
+
+int amdgpu_userq_ioctl(struct drm_device *dev, void *data,
+		       struct drm_file *filp)
+{
+	union drm_amdgpu_userq *args = data;
+	int r;
+
+	switch (args->in.op) {
+	case AMDGPU_USERQ_OP_CREATE:
+		if (args->in.flags & ~(AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_MASK |
+				       AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE))
+			return -EINVAL;
+		r = amdgpu_userq_create(filp, args);
+		if (r)
+			drm_file_err(filp, "Failed to create usermode queue\n");
+		break;
+
+	case AMDGPU_USERQ_OP_FREE:
+		if (args->in.ip_type ||
+		    args->in.doorbell_handle ||
+		    args->in.doorbell_offset ||
+		    args->in.flags ||
+		    args->in.queue_va ||
+		    args->in.queue_size ||
+		    args->in.rptr_va ||
+		    args->in.wptr_va ||
+		    args->in.wptr_va ||
+		    args->in.mqd ||
+		    args->in.mqd_size)
+			return -EINVAL;
+		r = amdgpu_userq_destroy(filp, args->in.queue_id);
+		if (r)
+			drm_file_err(filp, "Failed to destroy usermode queue\n");
+		break;
+
+	default:
+		drm_dbg_driver(dev, "Invalid user queue op specified: %d\n", args->in.op);
+		return -EINVAL;
+	}
+
+	return r;
+}
+
+static int
+amdgpu_userq_restore_all(struct amdgpu_userq_mgr *uq_mgr)
+{
+	struct amdgpu_usermode_queue *queue;
+	int queue_id;
+	int ret = 0, r;
+
+	/* Resume all the queues for this process */
+	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id) {
+		r = amdgpu_userq_map_helper(uq_mgr, queue);
+		if (r)
+			ret = r;
+	}
+
+	if (ret)
+		drm_file_err(uq_mgr->file, "Failed to map all the queues\n");
+	return ret;
+}
+
+static int
+amdgpu_userq_validate_vm_bo(void *_unused, struct amdgpu_bo *bo)
+{
+	struct ttm_operation_ctx ctx = { false, false };
+	int ret;
+
+	amdgpu_bo_placement_from_domain(bo, bo->allowed_domains);
+
+	ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
+	if (ret)
+		DRM_ERROR("Fail to validate\n");
+
+	return ret;
+}
+
+static int
+amdgpu_userq_validate_bos(struct amdgpu_userq_mgr *uq_mgr)
+{
+	struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr);
+	struct amdgpu_vm *vm = &fpriv->vm;
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct amdgpu_bo_va *bo_va;
+	struct ww_acquire_ctx *ticket;
+	struct drm_exec exec;
+	struct amdgpu_bo *bo;
+	struct dma_resv *resv;
+	bool clear, unlock;
+	int ret = 0;
+
+	drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES, 0);
+	drm_exec_until_all_locked(&exec) {
+		ret = amdgpu_vm_lock_pd(vm, &exec, 2);
+		drm_exec_retry_on_contention(&exec);
+		if (unlikely(ret)) {
+			drm_file_err(uq_mgr->file, "Failed to lock PD\n");
+			goto unlock_all;
+		}
+
+		/* Lock the done list */
+		list_for_each_entry(bo_va, &vm->done, base.vm_status) {
+			bo = bo_va->base.bo;
+			if (!bo)
+				continue;
+
+			ret = drm_exec_lock_obj(&exec, &bo->tbo.base);
+			drm_exec_retry_on_contention(&exec);
+			if (unlikely(ret))
+				goto unlock_all;
+		}
+	}
+
+	spin_lock(&vm->status_lock);
+	while (!list_empty(&vm->moved)) {
+		bo_va = list_first_entry(&vm->moved, struct amdgpu_bo_va,
+					 base.vm_status);
+		spin_unlock(&vm->status_lock);
+
+		/* Per VM BOs never need to bo cleared in the page tables */
+		ret = amdgpu_vm_bo_update(adev, bo_va, false);
+		if (ret)
+			goto unlock_all;
+		spin_lock(&vm->status_lock);
+	}
+
+	ticket = &exec.ticket;
+	while (!list_empty(&vm->invalidated)) {
+		bo_va = list_first_entry(&vm->invalidated, struct amdgpu_bo_va,
+					 base.vm_status);
+		resv = bo_va->base.bo->tbo.base.resv;
+		spin_unlock(&vm->status_lock);
+
+		bo = bo_va->base.bo;
+		ret = amdgpu_userq_validate_vm_bo(NULL, bo);
+		if (ret) {
+			drm_file_err(uq_mgr->file, "Failed to validate BO\n");
+			goto unlock_all;
+		}
+
+		/* Try to reserve the BO to avoid clearing its ptes */
+		if (!adev->debug_vm && dma_resv_trylock(resv)) {
+			clear = false;
+			unlock = true;
+		/* The caller is already holding the reservation lock */
+		} else if (dma_resv_locking_ctx(resv) == ticket) {
+			clear = false;
+			unlock = false;
+		/* Somebody else is using the BO right now */
+		} else {
+			clear = true;
+			unlock = false;
+		}
+
+		ret = amdgpu_vm_bo_update(adev, bo_va, clear);
+
+		if (unlock)
+			dma_resv_unlock(resv);
+		if (ret)
+			goto unlock_all;
+
+		spin_lock(&vm->status_lock);
+	}
+	spin_unlock(&vm->status_lock);
+
+	ret = amdgpu_eviction_fence_replace_fence(&fpriv->evf_mgr, &exec);
+	if (ret)
+		drm_file_err(uq_mgr->file, "Failed to replace eviction fence\n");
+
+unlock_all:
+	drm_exec_fini(&exec);
+	return ret;
+}
+
+static void amdgpu_userq_restore_worker(struct work_struct *work)
+{
+	struct amdgpu_userq_mgr *uq_mgr = work_to_uq_mgr(work, resume_work.work);
+	struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr);
+	int ret;
+
+	flush_work(&fpriv->evf_mgr.suspend_work.work);
+
+	mutex_lock(&uq_mgr->userq_mutex);
+
+	ret = amdgpu_userq_validate_bos(uq_mgr);
+	if (ret) {
+		drm_file_err(uq_mgr->file, "Failed to validate BOs to restore\n");
+		goto unlock;
+	}
+
+	ret = amdgpu_userq_restore_all(uq_mgr);
+	if (ret) {
+		drm_file_err(uq_mgr->file, "Failed to restore all queues\n");
+		goto unlock;
+	}
+
+unlock:
+	mutex_unlock(&uq_mgr->userq_mutex);
+}
+
+static int
+amdgpu_userq_evict_all(struct amdgpu_userq_mgr *uq_mgr)
+{
+	struct amdgpu_usermode_queue *queue;
+	int queue_id;
+	int ret = 0, r;
+
+	/* Try to unmap all the queues in this process ctx */
+	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id) {
+		r = amdgpu_userq_unmap_helper(uq_mgr, queue);
+		if (r)
+			ret = r;
+	}
+
+	if (ret)
+		drm_file_err(uq_mgr->file, "Couldn't unmap all the queues\n");
+	return ret;
+}
+
+static int
+amdgpu_userq_wait_for_signal(struct amdgpu_userq_mgr *uq_mgr)
+{
+	struct amdgpu_usermode_queue *queue;
+	int queue_id, ret;
+
+	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id) {
+		struct dma_fence *f = queue->last_fence;
+
+		if (!f || dma_fence_is_signaled(f))
+			continue;
+		ret = dma_fence_wait_timeout(f, true, msecs_to_jiffies(100));
+		if (ret <= 0) {
+			drm_file_err(uq_mgr->file, "Timed out waiting for fence=%llu:%llu\n",
+				     f->context, f->seqno);
+			return -ETIMEDOUT;
+		}
+	}
+
+	return 0;
+}
+
+void
+amdgpu_userq_evict(struct amdgpu_userq_mgr *uq_mgr,
+		   struct amdgpu_eviction_fence *ev_fence)
+{
+	int ret;
+	struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr);
+	struct amdgpu_eviction_fence_mgr *evf_mgr = &fpriv->evf_mgr;
+
+	/* Wait for any pending userqueue fence work to finish */
+	ret = amdgpu_userq_wait_for_signal(uq_mgr);
+	if (ret) {
+		drm_file_err(uq_mgr->file, "Not evicting userqueue, timeout waiting for work\n");
+		return;
+	}
+
+	ret = amdgpu_userq_evict_all(uq_mgr);
+	if (ret) {
+		drm_file_err(uq_mgr->file, "Failed to evict userqueue\n");
+		return;
+	}
+
+	/* Signal current eviction fence */
+	amdgpu_eviction_fence_signal(evf_mgr, ev_fence);
+
+	if (evf_mgr->fd_closing) {
+		cancel_delayed_work_sync(&uq_mgr->resume_work);
+		return;
+	}
+
+	/* Schedule a resume work */
+	schedule_delayed_work(&uq_mgr->resume_work, 0);
+}
+
+int amdgpu_userq_mgr_init(struct amdgpu_userq_mgr *userq_mgr, struct drm_file *file_priv,
+			  struct amdgpu_device *adev)
+{
+	mutex_init(&userq_mgr->userq_mutex);
+	idr_init_base(&userq_mgr->userq_idr, 1);
+	userq_mgr->adev = adev;
+	userq_mgr->file = file_priv;
+
+	mutex_lock(&adev->userq_mutex);
+	list_add(&userq_mgr->list, &adev->userq_mgr_list);
+	mutex_unlock(&adev->userq_mutex);
+
+	INIT_DELAYED_WORK(&userq_mgr->resume_work, amdgpu_userq_restore_worker);
+	return 0;
+}
+
+void amdgpu_userq_mgr_fini(struct amdgpu_userq_mgr *userq_mgr)
+{
+	struct amdgpu_device *adev = userq_mgr->adev;
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	uint32_t queue_id;
+
+	cancel_delayed_work_sync(&userq_mgr->resume_work);
+
+	mutex_lock(&adev->userq_mutex);
+	mutex_lock(&userq_mgr->userq_mutex);
+	idr_for_each_entry(&userq_mgr->userq_idr, queue, queue_id) {
+		amdgpu_userq_wait_for_last_fence(userq_mgr, queue);
+		amdgpu_userq_unmap_helper(userq_mgr, queue);
+		amdgpu_userq_cleanup(userq_mgr, queue, queue_id);
+	}
+
+	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+		if (uqm == userq_mgr) {
+			list_del(&uqm->list);
+			break;
+		}
+	}
+	idr_destroy(&userq_mgr->userq_idr);
+	mutex_unlock(&userq_mgr->userq_mutex);
+	mutex_unlock(&adev->userq_mutex);
+	mutex_destroy(&userq_mgr->userq_mutex);
+}
+
+int amdgpu_userq_suspend(struct amdgpu_device *adev)
+{
+	u32 ip_mask = amdgpu_userq_get_supported_ip_mask(adev);
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	int queue_id;
+	int ret = 0, r;
+
+	if (!ip_mask)
+		return 0;
+
+	mutex_lock(&adev->userq_mutex);
+	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+		cancel_delayed_work_sync(&uqm->resume_work);
+		mutex_lock(&uqm->userq_mutex);
+		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
+			r = amdgpu_userq_unmap_helper(uqm, queue);
+			if (r)
+				ret = r;
+		}
+		mutex_unlock(&uqm->userq_mutex);
+	}
+	mutex_unlock(&adev->userq_mutex);
+	return ret;
+}
+
+int amdgpu_userq_resume(struct amdgpu_device *adev)
+{
+	u32 ip_mask = amdgpu_userq_get_supported_ip_mask(adev);
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	int queue_id;
+	int ret = 0, r;
+
+	if (!ip_mask)
+		return 0;
+
+	mutex_lock(&adev->userq_mutex);
+	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+		mutex_lock(&uqm->userq_mutex);
+		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
+			r = amdgpu_userq_map_helper(uqm, queue);
+			if (r)
+				ret = r;
+		}
+		mutex_unlock(&uqm->userq_mutex);
+	}
+	mutex_unlock(&adev->userq_mutex);
+	return ret;
+}
+
+int amdgpu_userq_stop_sched_for_enforce_isolation(struct amdgpu_device *adev,
+						  u32 idx)
+{
+	u32 ip_mask = amdgpu_userq_get_supported_ip_mask(adev);
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	int queue_id;
+	int ret = 0, r;
+
+	/* only need to stop gfx/compute */
+	if (!(ip_mask & ((1 << AMDGPU_HW_IP_GFX) | (1 << AMDGPU_HW_IP_COMPUTE))))
+		return 0;
+
+	mutex_lock(&adev->userq_mutex);
+	if (adev->userq_halt_for_enforce_isolation)
+		dev_warn(adev->dev, "userq scheduling already stopped!\n");
+	adev->userq_halt_for_enforce_isolation = true;
+	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+		cancel_delayed_work_sync(&uqm->resume_work);
+		mutex_lock(&uqm->userq_mutex);
+		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
+			if (((queue->queue_type == AMDGPU_HW_IP_GFX) ||
+			     (queue->queue_type == AMDGPU_HW_IP_COMPUTE)) &&
+			    (queue->xcp_id == idx)) {
+				r = amdgpu_userq_unmap_helper(uqm, queue);
+				if (r)
+					ret = r;
+			}
+		}
+		mutex_unlock(&uqm->userq_mutex);
+	}
+	mutex_unlock(&adev->userq_mutex);
+	return ret;
+}
+
+int amdgpu_userq_start_sched_for_enforce_isolation(struct amdgpu_device *adev,
+						   u32 idx)
+{
+	u32 ip_mask = amdgpu_userq_get_supported_ip_mask(adev);
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	int queue_id;
+	int ret = 0, r;
+
+	/* only need to stop gfx/compute */
+	if (!(ip_mask & ((1 << AMDGPU_HW_IP_GFX) | (1 << AMDGPU_HW_IP_COMPUTE))))
+		return 0;
+
+	mutex_lock(&adev->userq_mutex);
+	if (!adev->userq_halt_for_enforce_isolation)
+		dev_warn(adev->dev, "userq scheduling already started!\n");
+	adev->userq_halt_for_enforce_isolation = false;
+	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+		mutex_lock(&uqm->userq_mutex);
+		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
+			if (((queue->queue_type == AMDGPU_HW_IP_GFX) ||
+			     (queue->queue_type == AMDGPU_HW_IP_COMPUTE)) &&
+			    (queue->xcp_id == idx)) {
+				r = amdgpu_userq_map_helper(uqm, queue);
+				if (r)
+					ret = r;
+			}
+		}
+		mutex_unlock(&uqm->userq_mutex);
+	}
+	mutex_unlock(&adev->userq_mutex);
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
new file mode 100644
index 000000000000..ec040c2fd6c9
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
@@ -0,0 +1,135 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef AMDGPU_USERQ_H_
+#define AMDGPU_USERQ_H_
+#include "amdgpu_eviction_fence.h"
+
+#define AMDGPU_MAX_USERQ_COUNT 512
+
+#define to_ev_fence(f) container_of(f, struct amdgpu_eviction_fence, base)
+#define uq_mgr_to_fpriv(u) container_of(u, struct amdgpu_fpriv, userq_mgr)
+#define work_to_uq_mgr(w, name) container_of(w, struct amdgpu_userq_mgr, name)
+
+enum amdgpu_userq_state {
+	AMDGPU_USERQ_STATE_UNMAPPED = 0,
+	AMDGPU_USERQ_STATE_MAPPED,
+	AMDGPU_USERQ_STATE_PREEMPTED,
+	AMDGPU_USERQ_STATE_HUNG,
+};
+
+struct amdgpu_mqd_prop;
+
+struct amdgpu_userq_obj {
+	void		 *cpu_ptr;
+	uint64_t	 gpu_addr;
+	struct amdgpu_bo *obj;
+};
+
+struct amdgpu_usermode_queue {
+	int			queue_type;
+	enum amdgpu_userq_state state;
+	uint64_t		doorbell_handle;
+	uint64_t		doorbell_index;
+	uint64_t		flags;
+	struct amdgpu_mqd_prop	*userq_prop;
+	struct amdgpu_userq_mgr *userq_mgr;
+	struct amdgpu_vm	*vm;
+	struct amdgpu_userq_obj mqd;
+	struct amdgpu_userq_obj	db_obj;
+	struct amdgpu_userq_obj fw_obj;
+	struct amdgpu_userq_obj wptr_obj;
+	struct xarray		fence_drv_xa;
+	struct amdgpu_userq_fence_driver *fence_drv;
+	struct dma_fence	*last_fence;
+	u32			xcp_id;
+	int			priority;
+};
+
+struct amdgpu_userq_funcs {
+	int (*mqd_create)(struct amdgpu_userq_mgr *uq_mgr,
+			  struct drm_amdgpu_userq_in *args,
+			  struct amdgpu_usermode_queue *queue);
+	void (*mqd_destroy)(struct amdgpu_userq_mgr *uq_mgr,
+			    struct amdgpu_usermode_queue *uq);
+	int (*unmap)(struct amdgpu_userq_mgr *uq_mgr,
+		     struct amdgpu_usermode_queue *queue);
+	int (*map)(struct amdgpu_userq_mgr *uq_mgr,
+		   struct amdgpu_usermode_queue *queue);
+};
+
+/* Usermode queues for gfx */
+struct amdgpu_userq_mgr {
+	struct idr			userq_idr;
+	struct mutex			userq_mutex;
+	struct amdgpu_device		*adev;
+	struct delayed_work		resume_work;
+	struct list_head		list;
+	struct drm_file			*file;
+};
+
+struct amdgpu_db_info {
+	uint64_t doorbell_handle;
+	uint32_t queue_type;
+	uint32_t doorbell_offset;
+	struct amdgpu_userq_obj	*db_obj;
+};
+
+int amdgpu_userq_ioctl(struct drm_device *dev, void *data, struct drm_file *filp);
+
+int amdgpu_userq_mgr_init(struct amdgpu_userq_mgr *userq_mgr, struct drm_file *file_priv,
+			  struct amdgpu_device *adev);
+
+void amdgpu_userq_mgr_fini(struct amdgpu_userq_mgr *userq_mgr);
+
+int amdgpu_userq_create_object(struct amdgpu_userq_mgr *uq_mgr,
+			       struct amdgpu_userq_obj *userq_obj,
+			       int size);
+
+void amdgpu_userq_destroy_object(struct amdgpu_userq_mgr *uq_mgr,
+				 struct amdgpu_userq_obj *userq_obj);
+
+void amdgpu_userq_evict(struct amdgpu_userq_mgr *uq_mgr,
+			struct amdgpu_eviction_fence *ev_fence);
+
+int amdgpu_userq_active(struct amdgpu_userq_mgr *uq_mgr);
+
+void amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *userq_mgr,
+				  struct amdgpu_eviction_fence_mgr *evf_mgr);
+
+uint64_t amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr,
+					 struct amdgpu_db_info *db_info,
+					     struct drm_file *filp);
+
+u32 amdgpu_userq_get_supported_ip_mask(struct amdgpu_device *adev);
+
+int amdgpu_userq_suspend(struct amdgpu_device *adev);
+int amdgpu_userq_resume(struct amdgpu_device *adev);
+
+int amdgpu_userq_stop_sched_for_enforce_isolation(struct amdgpu_device *adev,
+						  u32 idx);
+int amdgpu_userq_start_sched_for_enforce_isolation(struct amdgpu_device *adev,
+						   u32 idx);
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
new file mode 100644
index 000000000000..fc4d0d42e223
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
@@ -0,0 +1,968 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include <linux/kref.h>
+#include <linux/slab.h>
+#include <linux/dma-fence-unwrap.h>
+
+#include <drm/drm_exec.h>
+#include <drm/drm_syncobj.h>
+
+#include "amdgpu.h"
+#include "amdgpu_userq_fence.h"
+
+static const struct dma_fence_ops amdgpu_userq_fence_ops;
+static struct kmem_cache *amdgpu_userq_fence_slab;
+
+int amdgpu_userq_fence_slab_init(void)
+{
+	amdgpu_userq_fence_slab = kmem_cache_create("amdgpu_userq_fence",
+						    sizeof(struct amdgpu_userq_fence),
+						    0,
+						    SLAB_HWCACHE_ALIGN,
+						    NULL);
+	if (!amdgpu_userq_fence_slab)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void amdgpu_userq_fence_slab_fini(void)
+{
+	rcu_barrier();
+	kmem_cache_destroy(amdgpu_userq_fence_slab);
+}
+
+static inline struct amdgpu_userq_fence *to_amdgpu_userq_fence(struct dma_fence *f)
+{
+	if (!f || f->ops != &amdgpu_userq_fence_ops)
+		return NULL;
+
+	return container_of(f, struct amdgpu_userq_fence, base);
+}
+
+static u64 amdgpu_userq_fence_read(struct amdgpu_userq_fence_driver *fence_drv)
+{
+	return le64_to_cpu(*fence_drv->cpu_addr);
+}
+
+int amdgpu_userq_fence_driver_alloc(struct amdgpu_device *adev,
+				    struct amdgpu_usermode_queue *userq)
+{
+	struct amdgpu_userq_fence_driver *fence_drv;
+	unsigned long flags;
+	int r;
+
+	fence_drv = kzalloc(sizeof(*fence_drv), GFP_KERNEL);
+	if (!fence_drv)
+		return -ENOMEM;
+
+	/* Acquire seq64 memory */
+	r = amdgpu_seq64_alloc(adev, &fence_drv->va, &fence_drv->gpu_addr,
+			       &fence_drv->cpu_addr);
+	if (r)
+		goto free_fence_drv;
+
+	memset(fence_drv->cpu_addr, 0, sizeof(u64));
+
+	kref_init(&fence_drv->refcount);
+	INIT_LIST_HEAD(&fence_drv->fences);
+	spin_lock_init(&fence_drv->fence_list_lock);
+
+	fence_drv->adev = adev;
+	fence_drv->context = dma_fence_context_alloc(1);
+	get_task_comm(fence_drv->timeline_name, current);
+
+	xa_lock_irqsave(&adev->userq_xa, flags);
+	r = xa_err(__xa_store(&adev->userq_xa, userq->doorbell_index,
+			      fence_drv, GFP_KERNEL));
+	xa_unlock_irqrestore(&adev->userq_xa, flags);
+	if (r)
+		goto free_seq64;
+
+	userq->fence_drv = fence_drv;
+
+	return 0;
+
+free_seq64:
+	amdgpu_seq64_free(adev, fence_drv->va);
+free_fence_drv:
+	kfree(fence_drv);
+
+	return r;
+}
+
+static void amdgpu_userq_walk_and_drop_fence_drv(struct xarray *xa)
+{
+	struct amdgpu_userq_fence_driver *fence_drv;
+	unsigned long index;
+
+	if (xa_empty(xa))
+		return;
+
+	xa_lock(xa);
+	xa_for_each(xa, index, fence_drv) {
+		__xa_erase(xa, index);
+		amdgpu_userq_fence_driver_put(fence_drv);
+	}
+
+	xa_unlock(xa);
+}
+
+void
+amdgpu_userq_fence_driver_free(struct amdgpu_usermode_queue *userq)
+{
+	amdgpu_userq_walk_and_drop_fence_drv(&userq->fence_drv_xa);
+	xa_destroy(&userq->fence_drv_xa);
+	/* Drop the fence_drv reference held by user queue */
+	amdgpu_userq_fence_driver_put(userq->fence_drv);
+}
+
+void amdgpu_userq_fence_driver_process(struct amdgpu_userq_fence_driver *fence_drv)
+{
+	struct amdgpu_userq_fence *userq_fence, *tmp;
+	struct dma_fence *fence;
+	u64 rptr;
+	int i;
+
+	if (!fence_drv)
+		return;
+
+	rptr = amdgpu_userq_fence_read(fence_drv);
+
+	spin_lock(&fence_drv->fence_list_lock);
+	list_for_each_entry_safe(userq_fence, tmp, &fence_drv->fences, link) {
+		fence = &userq_fence->base;
+
+		if (rptr < fence->seqno)
+			break;
+
+		dma_fence_signal(fence);
+
+		for (i = 0; i < userq_fence->fence_drv_array_count; i++)
+			amdgpu_userq_fence_driver_put(userq_fence->fence_drv_array[i]);
+
+		list_del(&userq_fence->link);
+		dma_fence_put(fence);
+	}
+	spin_unlock(&fence_drv->fence_list_lock);
+}
+
+void amdgpu_userq_fence_driver_destroy(struct kref *ref)
+{
+	struct amdgpu_userq_fence_driver *fence_drv = container_of(ref,
+					 struct amdgpu_userq_fence_driver,
+					 refcount);
+	struct amdgpu_userq_fence_driver *xa_fence_drv;
+	struct amdgpu_device *adev = fence_drv->adev;
+	struct amdgpu_userq_fence *fence, *tmp;
+	struct xarray *xa = &adev->userq_xa;
+	unsigned long index, flags;
+	struct dma_fence *f;
+
+	spin_lock_irqsave(&fence_drv->fence_list_lock, flags);
+	list_for_each_entry_safe(fence, tmp, &fence_drv->fences, link) {
+		f = &fence->base;
+
+		if (!dma_fence_is_signaled(f)) {
+			dma_fence_set_error(f, -ECANCELED);
+			dma_fence_signal(f);
+		}
+
+		list_del(&fence->link);
+		dma_fence_put(f);
+	}
+	spin_unlock_irqrestore(&fence_drv->fence_list_lock, flags);
+
+	xa_lock_irqsave(xa, flags);
+	xa_for_each(xa, index, xa_fence_drv)
+		if (xa_fence_drv == fence_drv)
+			__xa_erase(xa, index);
+	xa_unlock_irqrestore(xa, flags);
+
+	/* Free seq64 memory */
+	amdgpu_seq64_free(adev, fence_drv->va);
+	kfree(fence_drv);
+}
+
+void amdgpu_userq_fence_driver_get(struct amdgpu_userq_fence_driver *fence_drv)
+{
+	kref_get(&fence_drv->refcount);
+}
+
+void amdgpu_userq_fence_driver_put(struct amdgpu_userq_fence_driver *fence_drv)
+{
+	kref_put(&fence_drv->refcount, amdgpu_userq_fence_driver_destroy);
+}
+
+static int amdgpu_userq_fence_alloc(struct amdgpu_userq_fence **userq_fence)
+{
+	*userq_fence = kmem_cache_alloc(amdgpu_userq_fence_slab, GFP_ATOMIC);
+	return *userq_fence ? 0 : -ENOMEM;
+}
+
+static int amdgpu_userq_fence_create(struct amdgpu_usermode_queue *userq,
+				     struct amdgpu_userq_fence *userq_fence,
+				     u64 seq, struct dma_fence **f)
+{
+	struct amdgpu_userq_fence_driver *fence_drv;
+	struct dma_fence *fence;
+	unsigned long flags;
+
+	fence_drv = userq->fence_drv;
+	if (!fence_drv)
+		return -EINVAL;
+
+	spin_lock_init(&userq_fence->lock);
+	INIT_LIST_HEAD(&userq_fence->link);
+	fence = &userq_fence->base;
+	userq_fence->fence_drv = fence_drv;
+
+	dma_fence_init(fence, &amdgpu_userq_fence_ops, &userq_fence->lock,
+		       fence_drv->context, seq);
+
+	amdgpu_userq_fence_driver_get(fence_drv);
+	dma_fence_get(fence);
+
+	if (!xa_empty(&userq->fence_drv_xa)) {
+		struct amdgpu_userq_fence_driver *stored_fence_drv;
+		unsigned long index, count = 0;
+		int i = 0;
+
+		xa_lock(&userq->fence_drv_xa);
+		xa_for_each(&userq->fence_drv_xa, index, stored_fence_drv)
+			count++;
+
+		userq_fence->fence_drv_array =
+			kvmalloc_array(count,
+				       sizeof(struct amdgpu_userq_fence_driver *),
+				       GFP_ATOMIC);
+
+		if (userq_fence->fence_drv_array) {
+			xa_for_each(&userq->fence_drv_xa, index, stored_fence_drv) {
+				userq_fence->fence_drv_array[i] = stored_fence_drv;
+				__xa_erase(&userq->fence_drv_xa, index);
+				i++;
+			}
+		}
+
+		userq_fence->fence_drv_array_count = i;
+		xa_unlock(&userq->fence_drv_xa);
+	} else {
+		userq_fence->fence_drv_array = NULL;
+		userq_fence->fence_drv_array_count = 0;
+	}
+
+	/* Check if hardware has already processed the job */
+	spin_lock_irqsave(&fence_drv->fence_list_lock, flags);
+	if (!dma_fence_is_signaled_locked(fence))
+		list_add_tail(&userq_fence->link, &fence_drv->fences);
+	else
+		dma_fence_put(fence);
+
+	spin_unlock_irqrestore(&fence_drv->fence_list_lock, flags);
+
+	*f = fence;
+
+	return 0;
+}
+
+static const char *amdgpu_userq_fence_get_driver_name(struct dma_fence *f)
+{
+	return "amdgpu_userq_fence";
+}
+
+static const char *amdgpu_userq_fence_get_timeline_name(struct dma_fence *f)
+{
+	struct amdgpu_userq_fence *fence = to_amdgpu_userq_fence(f);
+
+	return fence->fence_drv->timeline_name;
+}
+
+static bool amdgpu_userq_fence_signaled(struct dma_fence *f)
+{
+	struct amdgpu_userq_fence *fence = to_amdgpu_userq_fence(f);
+	struct amdgpu_userq_fence_driver *fence_drv = fence->fence_drv;
+	u64 rptr, wptr;
+
+	rptr = amdgpu_userq_fence_read(fence_drv);
+	wptr = fence->base.seqno;
+
+	if (rptr >= wptr)
+		return true;
+
+	return false;
+}
+
+static void amdgpu_userq_fence_free(struct rcu_head *rcu)
+{
+	struct dma_fence *fence = container_of(rcu, struct dma_fence, rcu);
+	struct amdgpu_userq_fence *userq_fence = to_amdgpu_userq_fence(fence);
+	struct amdgpu_userq_fence_driver *fence_drv = userq_fence->fence_drv;
+
+	/* Release the fence driver reference */
+	amdgpu_userq_fence_driver_put(fence_drv);
+
+	kvfree(userq_fence->fence_drv_array);
+	kmem_cache_free(amdgpu_userq_fence_slab, userq_fence);
+}
+
+static void amdgpu_userq_fence_release(struct dma_fence *f)
+{
+	call_rcu(&f->rcu, amdgpu_userq_fence_free);
+}
+
+static const struct dma_fence_ops amdgpu_userq_fence_ops = {
+	.use_64bit_seqno = true,
+	.get_driver_name = amdgpu_userq_fence_get_driver_name,
+	.get_timeline_name = amdgpu_userq_fence_get_timeline_name,
+	.signaled = amdgpu_userq_fence_signaled,
+	.release = amdgpu_userq_fence_release,
+};
+
+/**
+ * amdgpu_userq_fence_read_wptr - Read the userq wptr value
+ *
+ * @queue: user mode queue structure pointer
+ * @wptr: write pointer value
+ *
+ * Read the wptr value from userq's MQD. The userq signal IOCTL
+ * creates a dma_fence for the shared buffers that expects the
+ * RPTR value written to seq64 memory >= WPTR.
+ *
+ * Returns wptr value on success, error on failure.
+ */
+static int amdgpu_userq_fence_read_wptr(struct amdgpu_usermode_queue *queue,
+					u64 *wptr)
+{
+	struct amdgpu_bo_va_mapping *mapping;
+	struct amdgpu_bo *bo;
+	u64 addr, *ptr;
+	int r;
+
+	r = amdgpu_bo_reserve(queue->vm->root.bo, false);
+	if (r)
+		return r;
+
+	addr = queue->userq_prop->wptr_gpu_addr;
+	addr &= AMDGPU_GMC_HOLE_MASK;
+
+	mapping = amdgpu_vm_bo_lookup_mapping(queue->vm, addr >> PAGE_SHIFT);
+	if (!mapping) {
+		amdgpu_bo_unreserve(queue->vm->root.bo);
+		DRM_ERROR("Failed to lookup amdgpu_bo_va_mapping\n");
+		return -EINVAL;
+	}
+
+	bo = amdgpu_bo_ref(mapping->bo_va->base.bo);
+	amdgpu_bo_unreserve(queue->vm->root.bo);
+	r = amdgpu_bo_reserve(bo, true);
+	if (r) {
+		DRM_ERROR("Failed to reserve userqueue wptr bo");
+		return r;
+	}
+
+	r = amdgpu_bo_kmap(bo, (void **)&ptr);
+	if (r) {
+		DRM_ERROR("Failed mapping the userqueue wptr bo");
+		goto map_error;
+	}
+
+	*wptr = le64_to_cpu(*ptr);
+
+	amdgpu_bo_kunmap(bo);
+	amdgpu_bo_unreserve(bo);
+	amdgpu_bo_unref(&bo);
+
+	return 0;
+
+map_error:
+	amdgpu_bo_unreserve(bo);
+	amdgpu_bo_unref(&bo);
+
+	return r;
+}
+
+static void amdgpu_userq_fence_cleanup(struct dma_fence *fence)
+{
+	dma_fence_put(fence);
+}
+
+int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *filp)
+{
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_userq_mgr *userq_mgr = &fpriv->userq_mgr;
+	struct drm_amdgpu_userq_signal *args = data;
+	struct drm_gem_object **gobj_write = NULL;
+	struct drm_gem_object **gobj_read = NULL;
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_fence *userq_fence;
+	struct drm_syncobj **syncobj = NULL;
+	u32 *bo_handles_write, num_write_bo_handles;
+	u32 *syncobj_handles, num_syncobj_handles;
+	u32 *bo_handles_read, num_read_bo_handles;
+	int r, i, entry, rentry, wentry;
+	struct dma_fence *fence;
+	struct drm_exec exec;
+	u64 wptr;
+
+	num_syncobj_handles = args->num_syncobj_handles;
+	syncobj_handles = memdup_user(u64_to_user_ptr(args->syncobj_handles),
+				      sizeof(u32) * num_syncobj_handles);
+	if (IS_ERR(syncobj_handles))
+		return PTR_ERR(syncobj_handles);
+
+	/* Array of pointers to the looked up syncobjs */
+	syncobj = kmalloc_array(num_syncobj_handles, sizeof(*syncobj), GFP_KERNEL);
+	if (!syncobj) {
+		r = -ENOMEM;
+		goto free_syncobj_handles;
+	}
+
+	for (entry = 0; entry < num_syncobj_handles; entry++) {
+		syncobj[entry] = drm_syncobj_find(filp, syncobj_handles[entry]);
+		if (!syncobj[entry]) {
+			r = -ENOENT;
+			goto free_syncobj;
+		}
+	}
+
+	num_read_bo_handles = args->num_bo_read_handles;
+	bo_handles_read = memdup_user(u64_to_user_ptr(args->bo_read_handles),
+				      sizeof(u32) * num_read_bo_handles);
+	if (IS_ERR(bo_handles_read)) {
+		r = PTR_ERR(bo_handles_read);
+		goto free_syncobj;
+	}
+
+	/* Array of pointers to the GEM read objects */
+	gobj_read = kmalloc_array(num_read_bo_handles, sizeof(*gobj_read), GFP_KERNEL);
+	if (!gobj_read) {
+		r = -ENOMEM;
+		goto free_bo_handles_read;
+	}
+
+	for (rentry = 0; rentry < num_read_bo_handles; rentry++) {
+		gobj_read[rentry] = drm_gem_object_lookup(filp, bo_handles_read[rentry]);
+		if (!gobj_read[rentry]) {
+			r = -ENOENT;
+			goto put_gobj_read;
+		}
+	}
+
+	num_write_bo_handles = args->num_bo_write_handles;
+	bo_handles_write = memdup_user(u64_to_user_ptr(args->bo_write_handles),
+				       sizeof(u32) * num_write_bo_handles);
+	if (IS_ERR(bo_handles_write)) {
+		r = PTR_ERR(bo_handles_write);
+		goto put_gobj_read;
+	}
+
+	/* Array of pointers to the GEM write objects */
+	gobj_write = kmalloc_array(num_write_bo_handles, sizeof(*gobj_write), GFP_KERNEL);
+	if (!gobj_write) {
+		r = -ENOMEM;
+		goto free_bo_handles_write;
+	}
+
+	for (wentry = 0; wentry < num_write_bo_handles; wentry++) {
+		gobj_write[wentry] = drm_gem_object_lookup(filp, bo_handles_write[wentry]);
+		if (!gobj_write[wentry]) {
+			r = -ENOENT;
+			goto put_gobj_write;
+		}
+	}
+
+	/* Retrieve the user queue */
+	queue = idr_find(&userq_mgr->userq_idr, args->queue_id);
+	if (!queue) {
+		r = -ENOENT;
+		goto put_gobj_write;
+	}
+
+	r = amdgpu_userq_fence_read_wptr(queue, &wptr);
+	if (r)
+		goto put_gobj_write;
+
+	r = amdgpu_userq_fence_alloc(&userq_fence);
+	if (r)
+		goto put_gobj_write;
+
+	/* We are here means UQ is active, make sure the eviction fence is valid */
+	amdgpu_userq_ensure_ev_fence(&fpriv->userq_mgr, &fpriv->evf_mgr);
+
+	/* Create a new fence */
+	r = amdgpu_userq_fence_create(queue, userq_fence, wptr, &fence);
+	if (r) {
+		mutex_unlock(&userq_mgr->userq_mutex);
+		kmem_cache_free(amdgpu_userq_fence_slab, userq_fence);
+		goto put_gobj_write;
+	}
+
+	dma_fence_put(queue->last_fence);
+	queue->last_fence = dma_fence_get(fence);
+	mutex_unlock(&userq_mgr->userq_mutex);
+
+	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT,
+		      (num_read_bo_handles + num_write_bo_handles));
+
+	/* Lock all BOs with retry handling */
+	drm_exec_until_all_locked(&exec) {
+		r = drm_exec_prepare_array(&exec, gobj_read, num_read_bo_handles, 1);
+		drm_exec_retry_on_contention(&exec);
+		if (r) {
+			amdgpu_userq_fence_cleanup(fence);
+			goto exec_fini;
+		}
+
+		r = drm_exec_prepare_array(&exec, gobj_write, num_write_bo_handles, 1);
+		drm_exec_retry_on_contention(&exec);
+		if (r) {
+			amdgpu_userq_fence_cleanup(fence);
+			goto exec_fini;
+		}
+	}
+
+	for (i = 0; i < num_read_bo_handles; i++) {
+		if (!gobj_read || !gobj_read[i]->resv)
+			continue;
+
+		dma_resv_add_fence(gobj_read[i]->resv, fence,
+				   DMA_RESV_USAGE_READ);
+	}
+
+	for (i = 0; i < num_write_bo_handles; i++) {
+		if (!gobj_write || !gobj_write[i]->resv)
+			continue;
+
+		dma_resv_add_fence(gobj_write[i]->resv, fence,
+				   DMA_RESV_USAGE_WRITE);
+	}
+
+	/* Add the created fence to syncobj/BO's */
+	for (i = 0; i < num_syncobj_handles; i++)
+		drm_syncobj_replace_fence(syncobj[i], fence);
+
+	/* drop the reference acquired in fence creation function */
+	dma_fence_put(fence);
+
+exec_fini:
+	drm_exec_fini(&exec);
+put_gobj_write:
+	while (wentry-- > 0)
+		drm_gem_object_put(gobj_write[wentry]);
+	kfree(gobj_write);
+free_bo_handles_write:
+	kfree(bo_handles_write);
+put_gobj_read:
+	while (rentry-- > 0)
+		drm_gem_object_put(gobj_read[rentry]);
+	kfree(gobj_read);
+free_bo_handles_read:
+	kfree(bo_handles_read);
+free_syncobj:
+	while (entry-- > 0)
+		if (syncobj[entry])
+			drm_syncobj_put(syncobj[entry]);
+	kfree(syncobj);
+free_syncobj_handles:
+	kfree(syncobj_handles);
+
+	return r;
+}
+
+int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *filp)
+{
+	u32 *syncobj_handles, *timeline_points, *timeline_handles, *bo_handles_read, *bo_handles_write;
+	u32 num_syncobj, num_read_bo_handles, num_write_bo_handles;
+	struct drm_amdgpu_userq_fence_info *fence_info = NULL;
+	struct drm_amdgpu_userq_wait *wait_info = data;
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_userq_mgr *userq_mgr = &fpriv->userq_mgr;
+	struct amdgpu_usermode_queue *waitq;
+	struct drm_gem_object **gobj_write;
+	struct drm_gem_object **gobj_read;
+	struct dma_fence **fences = NULL;
+	u16 num_points, num_fences = 0;
+	int r, i, rentry, wentry, cnt;
+	struct drm_exec exec;
+
+	num_read_bo_handles = wait_info->num_bo_read_handles;
+	bo_handles_read = memdup_user(u64_to_user_ptr(wait_info->bo_read_handles),
+				      sizeof(u32) * num_read_bo_handles);
+	if (IS_ERR(bo_handles_read))
+		return PTR_ERR(bo_handles_read);
+
+	num_write_bo_handles = wait_info->num_bo_write_handles;
+	bo_handles_write = memdup_user(u64_to_user_ptr(wait_info->bo_write_handles),
+				       sizeof(u32) * num_write_bo_handles);
+	if (IS_ERR(bo_handles_write)) {
+		r = PTR_ERR(bo_handles_write);
+		goto free_bo_handles_read;
+	}
+
+	num_syncobj = wait_info->num_syncobj_handles;
+	syncobj_handles = memdup_user(u64_to_user_ptr(wait_info->syncobj_handles),
+				      sizeof(u32) * num_syncobj);
+	if (IS_ERR(syncobj_handles)) {
+		r = PTR_ERR(syncobj_handles);
+		goto free_bo_handles_write;
+	}
+
+	num_points = wait_info->num_syncobj_timeline_handles;
+	timeline_handles = memdup_user(u64_to_user_ptr(wait_info->syncobj_timeline_handles),
+				       sizeof(u32) * num_points);
+	if (IS_ERR(timeline_handles)) {
+		r = PTR_ERR(timeline_handles);
+		goto free_syncobj_handles;
+	}
+
+	timeline_points = memdup_user(u64_to_user_ptr(wait_info->syncobj_timeline_points),
+				      sizeof(u32) * num_points);
+	if (IS_ERR(timeline_points)) {
+		r = PTR_ERR(timeline_points);
+		goto free_timeline_handles;
+	}
+
+	gobj_read = kmalloc_array(num_read_bo_handles, sizeof(*gobj_read), GFP_KERNEL);
+	if (!gobj_read) {
+		r = -ENOMEM;
+		goto free_timeline_points;
+	}
+
+	for (rentry = 0; rentry < num_read_bo_handles; rentry++) {
+		gobj_read[rentry] = drm_gem_object_lookup(filp, bo_handles_read[rentry]);
+		if (!gobj_read[rentry]) {
+			r = -ENOENT;
+			goto put_gobj_read;
+		}
+	}
+
+	gobj_write = kmalloc_array(num_write_bo_handles, sizeof(*gobj_write), GFP_KERNEL);
+	if (!gobj_write) {
+		r = -ENOMEM;
+		goto put_gobj_read;
+	}
+
+	for (wentry = 0; wentry < num_write_bo_handles; wentry++) {
+		gobj_write[wentry] = drm_gem_object_lookup(filp, bo_handles_write[wentry]);
+		if (!gobj_write[wentry]) {
+			r = -ENOENT;
+			goto put_gobj_write;
+		}
+	}
+
+	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT,
+		      (num_read_bo_handles + num_write_bo_handles));
+
+	/* Lock all BOs with retry handling */
+	drm_exec_until_all_locked(&exec) {
+		r = drm_exec_prepare_array(&exec, gobj_read, num_read_bo_handles, 1);
+		drm_exec_retry_on_contention(&exec);
+		if (r) {
+			drm_exec_fini(&exec);
+			goto put_gobj_write;
+		}
+
+		r = drm_exec_prepare_array(&exec, gobj_write, num_write_bo_handles, 1);
+		drm_exec_retry_on_contention(&exec);
+		if (r) {
+			drm_exec_fini(&exec);
+			goto put_gobj_write;
+		}
+	}
+
+	if (!wait_info->num_fences) {
+		if (num_points) {
+			struct dma_fence_unwrap iter;
+			struct dma_fence *fence;
+			struct dma_fence *f;
+
+			for (i = 0; i < num_points; i++) {
+				r = drm_syncobj_find_fence(filp, timeline_handles[i],
+							   timeline_points[i],
+							   DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT,
+							   &fence);
+				if (r)
+					goto exec_fini;
+
+				dma_fence_unwrap_for_each(f, &iter, fence)
+					num_fences++;
+
+				dma_fence_put(fence);
+			}
+		}
+
+		/* Count syncobj's fence */
+		for (i = 0; i < num_syncobj; i++) {
+			struct dma_fence *fence;
+
+			r = drm_syncobj_find_fence(filp, syncobj_handles[i],
+						   0,
+						   DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT,
+						   &fence);
+			if (r)
+				goto exec_fini;
+
+			num_fences++;
+			dma_fence_put(fence);
+		}
+
+		/* Count GEM objects fence */
+		for (i = 0; i < num_read_bo_handles; i++) {
+			struct dma_resv_iter resv_cursor;
+			struct dma_fence *fence;
+
+			dma_resv_for_each_fence(&resv_cursor, gobj_read[i]->resv,
+						DMA_RESV_USAGE_READ, fence)
+				num_fences++;
+		}
+
+		for (i = 0; i < num_write_bo_handles; i++) {
+			struct dma_resv_iter resv_cursor;
+			struct dma_fence *fence;
+
+			dma_resv_for_each_fence(&resv_cursor, gobj_write[i]->resv,
+						DMA_RESV_USAGE_WRITE, fence)
+				num_fences++;
+		}
+
+		/*
+		 * Passing num_fences = 0 means that userspace doesn't want to
+		 * retrieve userq_fence_info. If num_fences = 0 we skip filling
+		 * userq_fence_info and return the actual number of fences on
+		 * args->num_fences.
+		 */
+		wait_info->num_fences = num_fences;
+	} else {
+		/* Array of fence info */
+		fence_info = kmalloc_array(wait_info->num_fences, sizeof(*fence_info), GFP_KERNEL);
+		if (!fence_info) {
+			r = -ENOMEM;
+			goto exec_fini;
+		}
+
+		/* Array of fences */
+		fences = kmalloc_array(wait_info->num_fences, sizeof(*fences), GFP_KERNEL);
+		if (!fences) {
+			r = -ENOMEM;
+			goto free_fence_info;
+		}
+
+		/* Retrieve GEM read objects fence */
+		for (i = 0; i < num_read_bo_handles; i++) {
+			struct dma_resv_iter resv_cursor;
+			struct dma_fence *fence;
+
+			dma_resv_for_each_fence(&resv_cursor, gobj_read[i]->resv,
+						DMA_RESV_USAGE_READ, fence) {
+				if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) {
+					r = -EINVAL;
+					goto free_fences;
+				}
+
+				fences[num_fences++] = fence;
+				dma_fence_get(fence);
+			}
+		}
+
+		/* Retrieve GEM write objects fence */
+		for (i = 0; i < num_write_bo_handles; i++) {
+			struct dma_resv_iter resv_cursor;
+			struct dma_fence *fence;
+
+			dma_resv_for_each_fence(&resv_cursor, gobj_write[i]->resv,
+						DMA_RESV_USAGE_WRITE, fence) {
+				if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) {
+					r = -EINVAL;
+					goto free_fences;
+				}
+
+				fences[num_fences++] = fence;
+				dma_fence_get(fence);
+			}
+		}
+
+		if (num_points) {
+			struct dma_fence_unwrap iter;
+			struct dma_fence *fence;
+			struct dma_fence *f;
+
+			for (i = 0; i < num_points; i++) {
+				r = drm_syncobj_find_fence(filp, timeline_handles[i],
+							   timeline_points[i],
+							   DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT,
+							   &fence);
+				if (r)
+					goto free_fences;
+
+				dma_fence_unwrap_for_each(f, &iter, fence) {
+					if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) {
+						r = -EINVAL;
+						goto free_fences;
+					}
+
+					dma_fence_get(f);
+					fences[num_fences++] = f;
+				}
+
+				dma_fence_put(fence);
+			}
+		}
+
+		/* Retrieve syncobj's fence */
+		for (i = 0; i < num_syncobj; i++) {
+			struct dma_fence *fence;
+
+			r = drm_syncobj_find_fence(filp, syncobj_handles[i],
+						   0,
+						   DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT,
+						   &fence);
+			if (r)
+				goto free_fences;
+
+			if (WARN_ON_ONCE(num_fences >= wait_info->num_fences)) {
+				r = -EINVAL;
+				goto free_fences;
+			}
+
+			fences[num_fences++] = fence;
+		}
+
+		/*
+		 * Keep only the latest fences to reduce the number of values
+		 * given back to userspace.
+		 */
+		num_fences = dma_fence_dedup_array(fences, num_fences);
+
+		waitq = idr_find(&userq_mgr->userq_idr, wait_info->waitq_id);
+		if (!waitq) {
+			r = -EINVAL;
+			goto free_fences;
+		}
+
+		for (i = 0, cnt = 0; i < num_fences; i++) {
+			struct amdgpu_userq_fence_driver *fence_drv;
+			struct amdgpu_userq_fence *userq_fence;
+			u32 index;
+
+			userq_fence = to_amdgpu_userq_fence(fences[i]);
+			if (!userq_fence) {
+				/*
+				 * Just waiting on other driver fences should
+				 * be good for now
+				 */
+				r = dma_fence_wait(fences[i], true);
+				if (r) {
+					dma_fence_put(fences[i]);
+					goto free_fences;
+				}
+
+				dma_fence_put(fences[i]);
+				continue;
+			}
+
+			fence_drv = userq_fence->fence_drv;
+			/*
+			 * We need to make sure the user queue release their reference
+			 * to the fence drivers at some point before queue destruction.
+			 * Otherwise, we would gather those references until we don't
+			 * have any more space left and crash.
+			 */
+			r = xa_alloc(&waitq->fence_drv_xa, &index, fence_drv,
+				     xa_limit_32b, GFP_KERNEL);
+			if (r)
+				goto free_fences;
+
+			amdgpu_userq_fence_driver_get(fence_drv);
+
+			/* Store drm syncobj's gpu va address and value */
+			fence_info[cnt].va = fence_drv->va;
+			fence_info[cnt].value = fences[i]->seqno;
+
+			dma_fence_put(fences[i]);
+			/* Increment the actual userq fence count */
+			cnt++;
+		}
+
+		wait_info->num_fences = cnt;
+		/* Copy userq fence info to user space */
+		if (copy_to_user(u64_to_user_ptr(wait_info->out_fences),
+				 fence_info, wait_info->num_fences * sizeof(*fence_info))) {
+			r = -EFAULT;
+			goto free_fences;
+		}
+
+		kfree(fences);
+		kfree(fence_info);
+	}
+
+	drm_exec_fini(&exec);
+	for (i = 0; i < num_read_bo_handles; i++)
+		drm_gem_object_put(gobj_read[i]);
+	kfree(gobj_read);
+
+	for (i = 0; i < num_write_bo_handles; i++)
+		drm_gem_object_put(gobj_write[i]);
+	kfree(gobj_write);
+
+	kfree(timeline_points);
+	kfree(timeline_handles);
+	kfree(syncobj_handles);
+	kfree(bo_handles_write);
+	kfree(bo_handles_read);
+
+	return 0;
+
+free_fences:
+	while (num_fences-- > 0)
+		dma_fence_put(fences[num_fences]);
+	kfree(fences);
+free_fence_info:
+	kfree(fence_info);
+exec_fini:
+	drm_exec_fini(&exec);
+put_gobj_write:
+	while (wentry-- > 0)
+		drm_gem_object_put(gobj_write[wentry]);
+	kfree(gobj_write);
+put_gobj_read:
+	while (rentry-- > 0)
+		drm_gem_object_put(gobj_read[rentry]);
+	kfree(gobj_read);
+free_timeline_points:
+	kfree(timeline_points);
+free_timeline_handles:
+	kfree(timeline_handles);
+free_syncobj_handles:
+	kfree(syncobj_handles);
+free_bo_handles_write:
+	kfree(bo_handles_write);
+free_bo_handles_read:
+	kfree(bo_handles_read);
+
+	return r;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h
new file mode 100644
index 000000000000..97a125ab8a78
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h
@@ -0,0 +1,76 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __AMDGPU_USERQ_FENCE_H__
+#define __AMDGPU_USERQ_FENCE_H__
+
+#include <linux/types.h>
+
+#include "amdgpu_userq.h"
+
+struct amdgpu_userq_fence {
+	struct dma_fence base;
+	/*
+	 * This lock is necessary to synchronize the
+	 * userqueue dma fence operations.
+	 */
+	spinlock_t lock;
+	struct list_head link;
+	unsigned long fence_drv_array_count;
+	struct amdgpu_userq_fence_driver *fence_drv;
+	struct amdgpu_userq_fence_driver **fence_drv_array;
+};
+
+struct amdgpu_userq_fence_driver {
+	struct kref refcount;
+	u64 va;
+	u64 gpu_addr;
+	u64 *cpu_addr;
+	u64 context;
+	/*
+	 * This lock is necesaary to synchronize the access
+	 * to the fences list by the fence driver.
+	 */
+	spinlock_t fence_list_lock;
+	struct list_head fences;
+	struct amdgpu_device *adev;
+	char timeline_name[TASK_COMM_LEN];
+};
+
+int amdgpu_userq_fence_slab_init(void);
+void amdgpu_userq_fence_slab_fini(void);
+
+void amdgpu_userq_fence_driver_get(struct amdgpu_userq_fence_driver *fence_drv);
+void amdgpu_userq_fence_driver_put(struct amdgpu_userq_fence_driver *fence_drv);
+int amdgpu_userq_fence_driver_alloc(struct amdgpu_device *adev,
+				    struct amdgpu_usermode_queue *userq);
+void amdgpu_userq_fence_driver_free(struct amdgpu_usermode_queue *userq);
+void amdgpu_userq_fence_driver_process(struct amdgpu_userq_fence_driver *fence_drv);
+void amdgpu_userq_fence_driver_destroy(struct kref *ref);
+int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *filp);
+int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *filp);
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 1991dd3d1056..c8885c3d54b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -353,9 +353,9 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev, int i)
 
 	cancel_delayed_work_sync(&adev->vcn.inst[i].idle_work);
 
-	/* err_event_athub will corrupt VCPU buffer, so we need to
+	/* err_event_athub and dpc recovery will corrupt VCPU buffer, so we need to
 	 * restore fw data and clear buffer in amdgpu_vcn_resume() */
-	if (in_ras_intr)
+	if (in_ras_intr || adev->pcie_reset_ctx.in_link_reset)
 		return 0;
 
 	return amdgpu_vcn_save_vcpu_bo_inst(adev, i);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index cdcdae7f71ce..83adf81defc7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -66,7 +66,6 @@
 #define VCN_ENC_CMD_REG_WAIT		0x0000000c
 
 #define VCN_AON_SOC_ADDRESS_2_0 	0x1f800
-#define VCN1_AON_SOC_ADDRESS_3_0 	0x48000
 #define VCN_VID_IP_ADDRESS_2_0		0x0
 #define VCN_AON_IP_ADDRESS_2_0		0x30000
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 0bb8cbe0dcc0..13f0cdeb59c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -1323,6 +1323,9 @@ static int amdgpu_virt_req_ras_err_count_internal(struct amdgpu_device *adev, bo
 {
 	struct amdgpu_virt *virt = &adev->virt;
 
+	if (!virt->ops || !virt->ops->req_ras_err_count)
+		return -EOPNOTSUPP;
+
 	/* Host allows 15 ras telemetry requests per 60 seconds. Afterwhich, the Host
 	 * will ignore incoming guest messages. Ratelimit the guest messages to
 	 * prevent guest self DOS.
@@ -1378,14 +1381,16 @@ amdgpu_virt_write_cpers_to_ring(struct amdgpu_device *adev,
 	used_size = host_telemetry->header.used_size;
 
 	if (used_size > (AMD_SRIOV_RAS_TELEMETRY_SIZE_KB << 10))
-		return 0;
+		return -EINVAL;
 
 	cper_dump = kmemdup(&host_telemetry->body.cper_dump, used_size, GFP_KERNEL);
 	if (!cper_dump)
 		return -ENOMEM;
 
-	if (checksum != amd_sriov_msg_checksum(cper_dump, used_size, 0, 0))
+	if (checksum != amd_sriov_msg_checksum(cper_dump, used_size, 0, 0)) {
+		ret = -EINVAL;
 		goto out;
+	}
 
 	*more = cper_dump->more;
 
@@ -1425,7 +1430,7 @@ static int amdgpu_virt_req_ras_cper_dump_internal(struct amdgpu_device *adev)
 	int ret = 0;
 	uint32_t more = 0;
 
-	if (!amdgpu_sriov_ras_cper_en(adev))
+	if (!virt->ops || !virt->ops->req_ras_cper_dump)
 		return -EOPNOTSUPP;
 
 	do {
@@ -1434,7 +1439,7 @@ static int amdgpu_virt_req_ras_cper_dump_internal(struct amdgpu_device *adev)
 				adev, virt->fw_reserve.ras_telemetry, &more);
 		else
 			ret = 0;
-	} while (more);
+	} while (more && !ret);
 
 	return ret;
 }
@@ -1444,6 +1449,9 @@ int amdgpu_virt_req_ras_cper_dump(struct amdgpu_device *adev, bool force_update)
 	struct amdgpu_virt *virt = &adev->virt;
 	int ret = 0;
 
+	if (!amdgpu_sriov_ras_cper_en(adev))
+		return -EOPNOTSUPP;
+
 	if ((__ratelimit(&virt->ras.ras_cper_dump_rs) || force_update) &&
 	    down_read_trylock(&adev->reset_domain->sem)) {
 		mutex_lock(&virt->ras.ras_telemetry_mutex);
@@ -1480,3 +1488,16 @@ bool amdgpu_virt_ras_telemetry_block_en(struct amdgpu_device *adev,
 
 	return true;
 }
+
+/*
+ * amdgpu_virt_request_bad_pages() - request bad pages
+ * @adev: amdgpu device.
+ * Send command to GPU hypervisor to write new bad pages into the shared PF2VF region
+ */
+void amdgpu_virt_request_bad_pages(struct amdgpu_device *adev)
+{
+	struct amdgpu_virt *virt = &adev->virt;
+
+	if (virt->ops && virt->ops->req_bad_pages)
+		virt->ops->req_bad_pages(adev);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index df03dba67ab8..577c6194db78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -97,6 +97,7 @@ struct amdgpu_virt_ops {
 	bool (*rcvd_ras_intr)(struct amdgpu_device *adev);
 	int (*req_ras_err_count)(struct amdgpu_device *adev);
 	int (*req_ras_cper_dump)(struct amdgpu_device *adev, u64 vf_rptr);
+	int (*req_bad_pages)(struct amdgpu_device *adev);
 };
 
 /*
@@ -146,11 +147,13 @@ enum AMDGIM_FEATURE_FLAG {
 
 enum AMDGIM_REG_ACCESS_FLAG {
 	/* Use PSP to program IH_RB_CNTL */
-	AMDGIM_FEATURE_IH_REG_PSP_EN     = (1 << 0),
+	AMDGIM_FEATURE_IH_REG_PSP_EN      = (1 << 0),
 	/* Use RLC to program MMHUB regs */
-	AMDGIM_FEATURE_MMHUB_REG_RLC_EN  = (1 << 1),
+	AMDGIM_FEATURE_MMHUB_REG_RLC_EN   = (1 << 1),
 	/* Use RLC to program GC regs */
-	AMDGIM_FEATURE_GC_REG_RLC_EN     = (1 << 2),
+	AMDGIM_FEATURE_GC_REG_RLC_EN      = (1 << 2),
+	/* Use PSP to program L1_TLB_CNTL*/
+	AMDGIM_FEATURE_L1_TLB_CNTL_PSP_EN = (1 << 3),
 };
 
 struct amdgim_pf2vf_info_v1 {
@@ -260,7 +263,10 @@ struct amdgpu_virt {
 	uint32_t			reg_val_offs;
 	struct amdgpu_irq_src		ack_irq;
 	struct amdgpu_irq_src		rcv_irq;
+
 	struct work_struct		flr_work;
+	struct work_struct		bad_pages_work;
+
 	struct amdgpu_mm_table		mm_table;
 	const struct amdgpu_virt_ops	*ops;
 	struct amdgpu_vf_error_buffer	vf_errors;
@@ -330,6 +336,10 @@ struct amdgpu_video_codec_info;
 (amdgpu_sriov_vf((adev)) && \
 	((adev)->virt.reg_access & (AMDGIM_FEATURE_GC_REG_RLC_EN)))
 
+#define amdgpu_sriov_reg_indirect_l1_tlb_cntl(adev) \
+(amdgpu_sriov_vf((adev)) && \
+	((adev)->virt.reg_access & (AMDGIM_FEATURE_L1_TLB_CNTL_PSP_EN)))
+
 #define amdgpu_sriov_rlcg_error_report_enabled(adev) \
         (amdgpu_sriov_reg_indirect_mmhub(adev) || amdgpu_sriov_reg_indirect_gc(adev))
 
@@ -423,4 +433,5 @@ int amdgpu_virt_req_ras_cper_dump(struct amdgpu_device *adev, bool force_update)
 int amdgpu_virt_ras_telemetry_post_reset(struct amdgpu_device *adev);
 bool amdgpu_virt_ras_telemetry_block_en(struct amdgpu_device *adev,
 					enum amdgpu_ras_block block);
+void amdgpu_virt_request_bad_pages(struct amdgpu_device *adev);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index ce52b4d75e94..3911c78f8282 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -787,7 +787,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
 	pasid_mapping_needed &= adev->gmc.gmc_funcs->emit_pasid_mapping &&
 		ring->funcs->emit_wreg;
 
-	cleaner_shader_needed = adev->gfx.enable_cleaner_shader &&
+	cleaner_shader_needed = job->run_cleaner_shader &&
+		adev->gfx.enable_cleaner_shader &&
 		ring->funcs->emit_cleaner_shader && job->base.s_fence &&
 		&job->base.s_fence->scheduled == isolation->spearhead;
 
@@ -817,7 +818,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
 	if (spm_update_needed && adev->gfx.rlc.funcs->update_spm_vmid)
 		adev->gfx.rlc.funcs->update_spm_vmid(adev, ring, job->vmid);
 
-	if (!ring->is_mes_queue && ring->funcs->emit_gds_switch &&
+	if (ring->funcs->emit_gds_switch &&
 	    gds_switch_needed) {
 		amdgpu_ring_emit_gds_switch(ring, job->vmid, job->gds_base,
 					    job->gds_size, job->gws_base,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 6da8994e0469..2d7f82e98df9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -24,6 +24,7 @@
 
 #include <linux/dma-mapping.h>
 #include <drm/ttm/ttm_range_manager.h>
+#include <drm/drm_drv.h>
 
 #include "amdgpu.h"
 #include "amdgpu_vm.h"
@@ -907,6 +908,9 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev)
 	struct ttm_resource_manager *man = &mgr->manager;
 	int err;
 
+	man->cg = drmm_cgroup_register_region(adev_to_drm(adev), "vram", adev->gmc.real_vram_size);
+	if (IS_ERR(man->cg))
+		return PTR_ERR(man->cg);
 	ttm_resource_manager_init(man, &adev->mman.bdev,
 				  adev->gmc.real_vram_size);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 23b6f7a4aa4a..b03c3895897b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -709,10 +709,10 @@ void amdgpu_xcp_cfg_sysfs_fini(struct amdgpu_device *adev)
 	struct amdgpu_xcp_cfg *xcp_cfg;
 	int i;
 
-	if (!adev->xcp_mgr)
+	if (!adev->xcp_mgr || !adev->xcp_mgr->xcp_cfg)
 		return;
 
-	xcp_cfg =  adev->xcp_mgr->xcp_cfg;
+	xcp_cfg = adev->xcp_mgr->xcp_cfg;
 	for (i = 0; i < xcp_cfg->num_res; i++) {
 		xcp_res = &xcp_cfg->xcp_res[i];
 		kobject_put(&xcp_res->kobj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 477424472bbe..f51ef4cf16e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -296,15 +296,27 @@ static const struct amdgpu_pcs_ras_field xgmi3x16_pcs_ras_fields[] = {
 
 static u32 xgmi_v6_4_get_link_status(struct amdgpu_device *adev, int global_link_num)
 {
-	const u32 smnpcs_xgmi3x16_pcs_state_hist1 = 0x11a00070;
-	const int xgmi_inst = 2;
-	u32 link_inst;
+	const u32 smn_xgmi_6_4_pcs_state_hist1[2] = { 0x11a00070, 0x11b00070 };
+	const u32 smn_xgmi_6_4_1_pcs_state_hist1[2] = { 0x12100070,
+							0x11b00070 };
+	u32 i, n;
 	u64 addr;
 
-	link_inst = global_link_num % xgmi_inst;
+	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
+	case IP_VERSION(6, 4, 0):
+		n = ARRAY_SIZE(smn_xgmi_6_4_pcs_state_hist1);
+		addr = smn_xgmi_6_4_pcs_state_hist1[global_link_num % n];
+		break;
+	case IP_VERSION(6, 4, 1):
+		n = ARRAY_SIZE(smn_xgmi_6_4_1_pcs_state_hist1);
+		addr = smn_xgmi_6_4_1_pcs_state_hist1[global_link_num % n];
+		break;
+	default:
+		return U32_MAX;
+	}
 
-	addr = (smnpcs_xgmi3x16_pcs_state_hist1 | (link_inst << 20)) +
-		adev->asic_funcs->encode_ext_smn_addressing(global_link_num / xgmi_inst);
+	i = global_link_num / n;
+	addr += adev->asic_funcs->encode_ext_smn_addressing(i);
 
 	return RREG32_PCIE_EXT(addr);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
index d6ac2652f0ac..92ca13097aaa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
@@ -109,10 +109,11 @@ union amd_sriov_msg_feature_flags {
 
 union amd_sriov_reg_access_flags {
 	struct {
-		uint32_t vf_reg_access_ih	: 1;
-		uint32_t vf_reg_access_mmhub	: 1;
-		uint32_t vf_reg_access_gc	: 1;
-		uint32_t reserved		: 29;
+		uint32_t vf_reg_access_ih		: 1;
+		uint32_t vf_reg_access_mmhub		: 1;
+		uint32_t vf_reg_access_gc		: 1;
+		uint32_t vf_reg_access_l1_tlb_cntl	: 1;
+		uint32_t reserved			: 28;
 	} flags;
 	uint32_t all;
 };
@@ -330,6 +331,7 @@ enum amd_sriov_mailbox_request_message {
 	MB_REQ_MSG_RAS_POISON = 202,
 	MB_REQ_RAS_ERROR_COUNT = 203,
 	MB_REQ_RAS_CPER_DUMP = 204,
+	MB_REQ_RAS_BAD_PAGES = 205,
 };
 
 /* mailbox message send from host to guest  */
@@ -347,6 +349,9 @@ enum amd_sriov_mailbox_response_message {
 	MB_RES_MSG_GPU_RMA			= 10,
 	MB_RES_MSG_RAS_ERROR_COUNT_READY	= 11,
 	MB_REQ_RAS_CPER_DUMP_READY		= 14,
+	MB_RES_MSG_RAS_BAD_PAGES_READY		= 15,
+	MB_RES_MSG_RAS_BAD_PAGES_NOTIFICATION	= 16,
+	MB_RES_MSG_UNRECOV_ERR_NOTIFICATION	= 17,
 	MB_RES_MSG_TEXT_MESSAGE			= 255
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
index ae071985f26e..1c083304ae77 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
@@ -448,53 +448,71 @@ static int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int x
 	return 0;
 }
 
-static int aqua_vanjaram_get_xcp_res_info(struct amdgpu_xcp_mgr *xcp_mgr,
-					  int mode,
-					  struct amdgpu_xcp_cfg *xcp_cfg)
+static int __aqua_vanjaram_get_px_mode_info(struct amdgpu_xcp_mgr *xcp_mgr,
+					    int px_mode, int *num_xcp,
+					    uint16_t *nps_modes)
 {
 	struct amdgpu_device *adev = xcp_mgr->adev;
-	int max_res[AMDGPU_XCP_RES_MAX] = {};
-	bool res_lt_xcp;
-	int num_xcp, i;
-	u16 nps_modes;
 
-	if (!(xcp_mgr->supp_xcp_modes & BIT(mode)))
+	if (!num_xcp || !nps_modes || !(xcp_mgr->supp_xcp_modes & BIT(px_mode)))
 		return -EINVAL;
 
-	max_res[AMDGPU_XCP_RES_XCC] = NUM_XCC(adev->gfx.xcc_mask);
-	max_res[AMDGPU_XCP_RES_DMA] = adev->sdma.num_instances;
-	max_res[AMDGPU_XCP_RES_DEC] = adev->vcn.num_vcn_inst;
-	max_res[AMDGPU_XCP_RES_JPEG] = adev->jpeg.num_jpeg_inst;
-
-	switch (mode) {
+	switch (px_mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
-		num_xcp = 1;
-		nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE);
+		*num_xcp = 1;
+		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE);
 		break;
 	case AMDGPU_DPX_PARTITION_MODE:
-		num_xcp = 2;
-		nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
-			    BIT(AMDGPU_NPS2_PARTITION_MODE);
+		*num_xcp = 2;
+		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
+			     BIT(AMDGPU_NPS2_PARTITION_MODE);
 		break;
 	case AMDGPU_TPX_PARTITION_MODE:
-		num_xcp = 3;
-		nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
-			    BIT(AMDGPU_NPS4_PARTITION_MODE);
+		*num_xcp = 3;
+		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
+			     BIT(AMDGPU_NPS4_PARTITION_MODE);
 		break;
 	case AMDGPU_QPX_PARTITION_MODE:
-		num_xcp = 4;
-		nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
-			    BIT(AMDGPU_NPS4_PARTITION_MODE);
+		*num_xcp = 4;
+		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
+			     BIT(AMDGPU_NPS4_PARTITION_MODE);
 		break;
 	case AMDGPU_CPX_PARTITION_MODE:
-		num_xcp = NUM_XCC(adev->gfx.xcc_mask);
-		nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
-			    BIT(AMDGPU_NPS4_PARTITION_MODE);
+		*num_xcp = NUM_XCC(adev->gfx.xcc_mask);
+		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
+			     BIT(AMDGPU_NPS4_PARTITION_MODE);
+		if (amdgpu_sriov_vf(adev))
+			*nps_modes |= BIT(AMDGPU_NPS2_PARTITION_MODE);
 		break;
 	default:
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+static int aqua_vanjaram_get_xcp_res_info(struct amdgpu_xcp_mgr *xcp_mgr,
+					  int mode,
+					  struct amdgpu_xcp_cfg *xcp_cfg)
+{
+	struct amdgpu_device *adev = xcp_mgr->adev;
+	int max_res[AMDGPU_XCP_RES_MAX] = {};
+	bool res_lt_xcp;
+	int num_xcp, i, r;
+	u16 nps_modes;
+
+	if (!(xcp_mgr->supp_xcp_modes & BIT(mode)))
+		return -EINVAL;
+
+	max_res[AMDGPU_XCP_RES_XCC] = NUM_XCC(adev->gfx.xcc_mask);
+	max_res[AMDGPU_XCP_RES_DMA] = adev->sdma.num_instances;
+	max_res[AMDGPU_XCP_RES_DEC] = adev->vcn.num_vcn_inst;
+	max_res[AMDGPU_XCP_RES_JPEG] = adev->jpeg.num_jpeg_inst;
+
+	r = __aqua_vanjaram_get_px_mode_info(xcp_mgr, mode, &num_xcp, &nps_modes);
+	if (r)
+		return r;
+
 	xcp_cfg->compatible_nps_modes =
 		(adev->gmc.supported_nps_modes & nps_modes);
 	xcp_cfg->num_res = ARRAY_SIZE(max_res);
@@ -543,30 +561,31 @@ static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 					  enum amdgpu_gfx_partition mode)
 {
 	struct amdgpu_device *adev = xcp_mgr->adev;
-	int num_xcc, num_xccs_per_xcp;
+	int num_xcc, num_xccs_per_xcp, r;
+	int num_xcp, nps_mode;
+	u16 supp_nps_modes;
+	bool comp_mode;
+
+	nps_mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
+	r = __aqua_vanjaram_get_px_mode_info(xcp_mgr, mode, &num_xcp,
+					       &supp_nps_modes);
+	if (r)
+		return false;
 
+	comp_mode = !!(BIT(nps_mode) & supp_nps_modes);
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
-		return adev->gmc.num_mem_partitions == 1 && num_xcc > 0;
+		return comp_mode && num_xcc > 0;
 	case AMDGPU_DPX_PARTITION_MODE:
-		return adev->gmc.num_mem_partitions <= 2 && (num_xcc % 4) == 0;
+		return comp_mode && (num_xcc % 4) == 0;
 	case AMDGPU_TPX_PARTITION_MODE:
-		return (adev->gmc.num_mem_partitions == 1 ||
-			adev->gmc.num_mem_partitions == 3) &&
-		       ((num_xcc % 3) == 0);
+		return comp_mode && ((num_xcc % 3) == 0);
 	case AMDGPU_QPX_PARTITION_MODE:
 		num_xccs_per_xcp = num_xcc / 4;
-		return (adev->gmc.num_mem_partitions == 1 ||
-			adev->gmc.num_mem_partitions == 4) &&
-		       (num_xccs_per_xcp >= 2);
+		return comp_mode && (num_xccs_per_xcp >= 2);
 	case AMDGPU_CPX_PARTITION_MODE:
-		/* (num_xcc > 1) because 1 XCC is considered SPX, not CPX.
-		 * (num_xcc % adev->gmc.num_mem_partitions) == 0 because
-		 * num_compute_partitions can't be less than num_mem_partitions
-		 */
-		return ((num_xcc > 1) &&
-		       (num_xcc % adev->gmc.num_mem_partitions) == 0);
+		return comp_mode && (num_xcc > 1);
 	default:
 		return false;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index 81d195d366ce..427b073de2fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -1444,6 +1444,7 @@ static void atom_get_vbios_pn(struct atom_context *ctx)
 		if (vbios_str == NULL)
 			vbios_str += sizeof(BIOS_ATOM_PREFIX) - 1;
 	}
+	OPTIMIZER_HIDE_VAR(vbios_str);
 	if (vbios_str != NULL && *vbios_str == 0)
 		vbios_str++;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
index 521b9faab180..492813ab1b54 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
@@ -458,8 +458,8 @@ bool amdgpu_atombios_dp_needs_link_train(struct amdgpu_connector *amdgpu_connect
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	struct amdgpu_connector_atom_dig *dig = amdgpu_connector->con_priv;
 
-	if (drm_dp_dpcd_read_link_status(&amdgpu_connector->ddc_bus->aux, link_status)
-	    <= 0)
+	if (drm_dp_dpcd_read_link_status(&amdgpu_connector->ddc_bus->aux,
+					 link_status) < 0)
 		return false;
 	if (drm_dp_channel_eq_ok(link_status, dig->dp_lane_count))
 		return false;
@@ -616,7 +616,7 @@ amdgpu_atombios_dp_link_train_cr(struct amdgpu_atombios_dp_link_train_info *dp_i
 		drm_dp_link_train_clock_recovery_delay(dp_info->aux, dp_info->dpcd);
 
 		if (drm_dp_dpcd_read_link_status(dp_info->aux,
-						 dp_info->link_status) <= 0) {
+						 dp_info->link_status) < 0) {
 			DRM_ERROR("displayport link status failed\n");
 			break;
 		}
@@ -681,7 +681,7 @@ amdgpu_atombios_dp_link_train_ce(struct amdgpu_atombios_dp_link_train_info *dp_i
 		drm_dp_link_train_channel_eq_delay(dp_info->aux, dp_info->dpcd);
 
 		if (drm_dp_dpcd_read_link_status(dp_info->aux,
-						 dp_info->link_status) <= 0) {
+						 dp_info->link_status) < 0) {
 			DRM_ERROR("displayport link status failed\n");
 			break;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 508cea965983..9e8715b4739d 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -56,6 +56,8 @@ static void cik_sdma_set_buffer_funcs(struct amdgpu_device *adev);
 static void cik_sdma_set_vm_pte_funcs(struct amdgpu_device *adev);
 static int cik_sdma_soft_reset(struct amdgpu_ip_block *ip_block);
 
+u32 amdgpu_cik_gpu_check_soft_reset(struct amdgpu_device *adev);
+
 MODULE_FIRMWARE("amdgpu/bonaire_sdma.bin");
 MODULE_FIRMWARE("amdgpu/bonaire_sdma1.bin");
 MODULE_FIRMWARE("amdgpu/hawaii_sdma.bin");
@@ -67,9 +69,6 @@ MODULE_FIRMWARE("amdgpu/kabini_sdma1.bin");
 MODULE_FIRMWARE("amdgpu/mullins_sdma.bin");
 MODULE_FIRMWARE("amdgpu/mullins_sdma1.bin");
 
-u32 amdgpu_cik_gpu_check_soft_reset(struct amdgpu_device *adev);
-
-
 static void cik_sdma_free_microcode(struct amdgpu_device *adev)
 {
 	int i;
@@ -993,14 +992,9 @@ static int cik_sdma_sw_fini(struct amdgpu_ip_block *ip_block)
 
 static int cik_sdma_hw_init(struct amdgpu_ip_block *ip_block)
 {
-	int r;
 	struct amdgpu_device *adev = ip_block->adev;
 
-	r = cik_sdma_start(adev);
-	if (r)
-		return r;
-
-	return r;
+	return cik_sdma_start(adev);
 }
 
 static int cik_sdma_hw_fini(struct amdgpu_ip_block *ip_block)
@@ -1040,14 +1034,10 @@ static bool cik_sdma_is_idle(struct amdgpu_ip_block *ip_block)
 static int cik_sdma_wait_for_idle(struct amdgpu_ip_block *ip_block)
 {
 	unsigned i;
-	u32 tmp;
 	struct amdgpu_device *adev = ip_block->adev;
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(mmSRBM_STATUS2) & (SRBM_STATUS2__SDMA_BUSY_MASK |
-				SRBM_STATUS2__SDMA1_BUSY_MASK);
-
-		if (!tmp)
+		if (cik_sdma_is_idle(ip_block))
 			return 0;
 		udelay(1);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/cikd.h b/drivers/gpu/drm/amd/amdgpu/cikd.h
index 279288365940..8aca4f2734f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/cikd.h
+++ b/drivers/gpu/drm/amd/amdgpu/cikd.h
@@ -60,9 +60,6 @@
 #define AUD5_REGISTER_OFFSET                 (0x179d - 0x1780)
 #define AUD6_REGISTER_OFFSET                 (0x17a4 - 0x1780)
 
-#define BONAIRE_GB_ADDR_CONFIG_GOLDEN        0x12010001
-#define HAWAII_GB_ADDR_CONFIG_GOLDEN         0x12011003
-
 #define		PIPEID(x)					((x) << 0)
 #define		MEID(x)						((x) << 2)
 #define		VMID(x)						((x) << 4)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index df401aded662..bf7c22f81cda 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -3075,7 +3075,7 @@ static int dce_v10_0_set_hpd_irq_state(struct amdgpu_device *adev,
 	u32 tmp;
 
 	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", hpd);
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return 0;
 	}
 
@@ -3227,7 +3227,7 @@ static void dce_v10_0_hpd_int_ack(struct amdgpu_device *adev,
 	u32 tmp;
 
 	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", hpd);
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 80f01c3989cd..47e05783c4a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -3206,7 +3206,7 @@ static int dce_v11_0_set_hpd_irq_state(struct amdgpu_device *adev,
 	u32 tmp;
 
 	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", hpd);
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return 0;
 	}
 
@@ -3358,7 +3358,7 @@ static void dce_v11_0_hpd_int_ack(struct amdgpu_device *adev,
 	u32 tmp;
 
 	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", hpd);
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return;
 	}
 
@@ -3488,8 +3488,7 @@ static const struct amd_ip_funcs dce_v11_0_ip_funcs = {
 	.set_powergating_state = dce_v11_0_set_powergating_state,
 };
 
-static void
-dce_v11_0_encoder_mode_set(struct drm_encoder *encoder,
+static void dce_v11_0_encoder_mode_set(struct drm_encoder *encoder,
 			  struct drm_display_mode *mode,
 			  struct drm_display_mode *adjusted_mode)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index 255c70959343..276c025c4c03 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -287,7 +287,7 @@ static void dce_v6_0_hpd_int_ack(struct amdgpu_device *adev,
 	u32 tmp;
 
 	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", hpd);
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return;
 	}
 
@@ -412,7 +412,7 @@ static void dce_v6_0_set_vga_render_state(struct amdgpu_device *adev,
 {
 	if (!render)
 		WREG32(mmVGA_RENDER_CONTROL,
-			RREG32(mmVGA_RENDER_CONTROL) & VGA_VSTATUS_CNTL);
+		       RREG32(mmVGA_RENDER_CONTROL) & ~VGA_RENDER_CONTROL__VGA_VSTATUS_CNTL_MASK);
 }
 
 static int dce_v6_0_get_num_crtc(struct amdgpu_device *adev)
@@ -1011,16 +1011,16 @@ static void dce_v6_0_program_watermarks(struct amdgpu_device *adev,
 	/* select wm A */
 	arb_control3 = RREG32(mmDPG_PIPE_ARBITRATION_CONTROL3 + amdgpu_crtc->crtc_offset);
 	tmp = arb_control3;
-	tmp &= ~LATENCY_WATERMARK_MASK(3);
-	tmp |= LATENCY_WATERMARK_MASK(1);
+	tmp &= ~(3 << DPG_PIPE_ARBITRATION_CONTROL3__URGENCY_WATERMARK_MASK__SHIFT);
+	tmp |= (1 << DPG_PIPE_ARBITRATION_CONTROL3__URGENCY_WATERMARK_MASK__SHIFT);
 	WREG32(mmDPG_PIPE_ARBITRATION_CONTROL3 + amdgpu_crtc->crtc_offset, tmp);
 	WREG32(mmDPG_PIPE_URGENCY_CONTROL + amdgpu_crtc->crtc_offset,
 	       ((latency_watermark_a << DPG_PIPE_URGENCY_CONTROL__URGENCY_LOW_WATERMARK__SHIFT)  |
 		(line_time << DPG_PIPE_URGENCY_CONTROL__URGENCY_HIGH_WATERMARK__SHIFT)));
 	/* select wm B */
 	tmp = RREG32(mmDPG_PIPE_ARBITRATION_CONTROL3 + amdgpu_crtc->crtc_offset);
-	tmp &= ~LATENCY_WATERMARK_MASK(3);
-	tmp |= LATENCY_WATERMARK_MASK(2);
+	tmp &= ~(3 << DPG_PIPE_ARBITRATION_CONTROL3__URGENCY_WATERMARK_MASK__SHIFT);
+	tmp |= (2 << DPG_PIPE_ARBITRATION_CONTROL3__URGENCY_WATERMARK_MASK__SHIFT);
 	WREG32(mmDPG_PIPE_ARBITRATION_CONTROL3 + amdgpu_crtc->crtc_offset, tmp);
 	WREG32(mmDPG_PIPE_URGENCY_CONTROL + amdgpu_crtc->crtc_offset,
 	       ((latency_watermark_b << DPG_PIPE_URGENCY_CONTROL__URGENCY_LOW_WATERMARK__SHIFT) |
@@ -1089,7 +1089,7 @@ static u32 dce_v6_0_line_buffer_adjust(struct amdgpu_device *adev,
 	}
 
 	WREG32(mmDC_LB_MEMORY_SPLIT + amdgpu_crtc->crtc_offset,
-	       DC_LB_MEMORY_CONFIG(tmp));
+	       (tmp << DC_LB_MEMORY_SPLIT__DC_LB_MEMORY_CONFIG__SHIFT));
 
 	WREG32(mmPIPE0_DMIF_BUFFER_CONTROL + pipe_offset,
 	       (buffer_alloc << PIPE0_DMIF_BUFFER_CONTROL__DMIF_BUFFERS_ALLOCATED__SHIFT));
@@ -1306,6 +1306,7 @@ static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
+	u32 offset;
 	struct drm_connector *connector;
 	struct drm_connector_list_iter iter;
 	struct amdgpu_connector *amdgpu_connector = NULL;
@@ -1327,6 +1328,11 @@ static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
 		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
 	};
 
+	if (!dig || !dig->afmt || !dig->afmt->pin)
+		return;
+
+	offset = dig->afmt->pin->offset;
+
 	drm_connector_list_iter_begin(dev, &iter);
 	drm_for_each_connector_iter(connector, &iter) {
 		if (connector->encoder == encoder) {
@@ -1348,7 +1354,7 @@ static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
 		return;
 
 	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
-		u32 tmp = 0;
+		u32 value = 0;
 		u8 stereo_freqs = 0;
 		int max_channels = -1;
 		int j;
@@ -1358,12 +1364,12 @@ static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
 
 			if (sad->format == eld_reg_to_type[i][1]) {
 				if (sad->channels > max_channels) {
-					tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-							MAX_CHANNELS, sad->channels);
-					tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-							DESCRIPTOR_BYTE_2, sad->byte2);
-					tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-							SUPPORTED_FREQUENCIES, sad->freq);
+					value = (sad->channels <<
+						AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__MAX_CHANNELS__SHIFT) |
+					       (sad->byte2 <<
+						AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__DESCRIPTOR_BYTE_2__SHIFT) |
+					       (sad->freq <<
+						AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__SUPPORTED_FREQUENCIES__SHIFT);
 					max_channels = sad->channels;
 				}
 
@@ -1374,13 +1380,13 @@ static void dce_v6_0_audio_write_sad_regs(struct drm_encoder *encoder)
 			}
 		}
 
-		tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-				SUPPORTED_FREQUENCIES_STEREO, stereo_freqs);
-		WREG32_AUDIO_ENDPT(dig->afmt->pin->offset, eld_reg_to_type[i][0], tmp);
+		value |= (stereo_freqs <<
+			AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__SUPPORTED_FREQUENCIES_STEREO__SHIFT);
+
+		WREG32_AUDIO_ENDPT(offset, eld_reg_to_type[i][0], value);
 	}
 
 	kfree(sads);
-
 }
 
 static void dce_v6_0_audio_enable(struct amdgpu_device *adev,
@@ -1886,7 +1892,7 @@ static int dce_v6_0_crtc_do_set_base(struct drm_crtc *crtc,
 	struct amdgpu_bo *abo;
 	uint64_t fb_location, tiling_flags;
 	uint32_t fb_format, fb_pitch_pixels, pipe_config;
-	u32 fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_NONE);
+	u32 fb_swap = (GRPH_ENDIAN_NONE << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 	u32 viewport_w, viewport_h;
 	int r;
 	bool bypass_lut = false;
@@ -1926,76 +1932,76 @@ static int dce_v6_0_crtc_do_set_base(struct drm_crtc *crtc,
 
 	switch (target_fb->format->format) {
 	case DRM_FORMAT_C8:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_8BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_INDEXED));
+		fb_format = ((GRPH_DEPTH_8BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_INDEXED << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 		break;
 	case DRM_FORMAT_XRGB4444:
 	case DRM_FORMAT_ARGB4444:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_16BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_ARGB4444));
+		fb_format = ((GRPH_DEPTH_16BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_ARGB4444 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN16);
+		fb_swap = (GRPH_ENDIAN_8IN16 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		break;
 	case DRM_FORMAT_XRGB1555:
 	case DRM_FORMAT_ARGB1555:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_16BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_ARGB1555));
+		fb_format = ((GRPH_DEPTH_16BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_ARGB1555 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN16);
+		fb_swap = (GRPH_ENDIAN_8IN16 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		break;
 	case DRM_FORMAT_BGRX5551:
 	case DRM_FORMAT_BGRA5551:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_16BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_BGRA5551));
+		fb_format = ((GRPH_DEPTH_16BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_BGRA5551 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN16);
+		fb_swap = (GRPH_ENDIAN_8IN16 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		break;
 	case DRM_FORMAT_RGB565:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_16BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_ARGB565));
+		fb_format = ((GRPH_DEPTH_16BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_ARGB565 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN16);
+		fb_swap = (GRPH_ENDIAN_8IN16 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		break;
 	case DRM_FORMAT_XRGB8888:
 	case DRM_FORMAT_ARGB8888:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_32BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_ARGB8888));
+		fb_format = ((GRPH_DEPTH_32BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_ARGB8888 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN32);
+		fb_swap = (GRPH_ENDIAN_8IN32 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		break;
 	case DRM_FORMAT_XRGB2101010:
 	case DRM_FORMAT_ARGB2101010:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_32BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_ARGB2101010));
+		fb_format = ((GRPH_DEPTH_32BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_ARGB2101010 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN32);
+		fb_swap = (GRPH_ENDIAN_8IN32 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
 		bypass_lut = true;
 		break;
 	case DRM_FORMAT_BGRX1010102:
 	case DRM_FORMAT_BGRA1010102:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_32BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_BGRA1010102));
+		fb_format = ((GRPH_DEPTH_32BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_BGRA1010102 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap = GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN32);
+		fb_swap = (GRPH_ENDIAN_8IN32 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
 		bypass_lut = true;
 		break;
 	case DRM_FORMAT_XBGR8888:
 	case DRM_FORMAT_ABGR8888:
-		fb_format = (GRPH_DEPTH(GRPH_DEPTH_32BPP) |
-			     GRPH_FORMAT(GRPH_FORMAT_ARGB8888));
-		fb_swap = (GRPH_RED_CROSSBAR(GRPH_RED_SEL_B) |
-			   GRPH_BLUE_CROSSBAR(GRPH_BLUE_SEL_R));
+		fb_format = ((GRPH_DEPTH_32BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) |
+			     (GRPH_FORMAT_ARGB8888 << GRPH_CONTROL__GRPH_FORMAT__SHIFT));
+		fb_swap = ((GRPH_RED_SEL_B << GRPH_SWAP_CNTL__GRPH_RED_CROSSBAR__SHIFT) |
+			   (GRPH_BLUE_SEL_R << GRPH_SWAP_CNTL__GRPH_BLUE_CROSSBAR__SHIFT));
 #ifdef __BIG_ENDIAN
-		fb_swap |= GRPH_ENDIAN_SWAP(GRPH_ENDIAN_8IN32);
+		fb_swap |= (GRPH_ENDIAN_8IN32 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT);
 #endif
 		break;
 	default:
@@ -2013,18 +2019,18 @@ static int dce_v6_0_crtc_do_set_base(struct drm_crtc *crtc,
 		tile_split = AMDGPU_TILING_GET(tiling_flags, TILE_SPLIT);
 		num_banks = AMDGPU_TILING_GET(tiling_flags, NUM_BANKS);
 
-		fb_format |= GRPH_NUM_BANKS(num_banks);
-		fb_format |= GRPH_ARRAY_MODE(GRPH_ARRAY_2D_TILED_THIN1);
-		fb_format |= GRPH_TILE_SPLIT(tile_split);
-		fb_format |= GRPH_BANK_WIDTH(bankw);
-		fb_format |= GRPH_BANK_HEIGHT(bankh);
-		fb_format |= GRPH_MACRO_TILE_ASPECT(mtaspect);
+		fb_format |= (num_banks << GRPH_CONTROL__GRPH_NUM_BANKS__SHIFT);
+		fb_format |= (GRPH_ARRAY_2D_TILED_THIN1 << GRPH_CONTROL__GRPH_ARRAY_MODE__SHIFT);
+		fb_format |= (tile_split << GRPH_CONTROL__GRPH_TILE_SPLIT__SHIFT);
+		fb_format |= (bankw << GRPH_CONTROL__GRPH_BANK_WIDTH__SHIFT);
+		fb_format |= (bankh << GRPH_CONTROL__GRPH_BANK_HEIGHT__SHIFT);
+		fb_format |= (mtaspect << GRPH_CONTROL__GRPH_MACRO_TILE_ASPECT__SHIFT);
 	} else if (AMDGPU_TILING_GET(tiling_flags, ARRAY_MODE) == ARRAY_1D_TILED_THIN1) {
-		fb_format |= GRPH_ARRAY_MODE(GRPH_ARRAY_1D_TILED_THIN1);
+		fb_format |= (GRPH_ARRAY_1D_TILED_THIN1 << GRPH_CONTROL__GRPH_ARRAY_MODE__SHIFT);
 	}
 
 	pipe_config = AMDGPU_TILING_GET(tiling_flags, PIPE_CONFIG);
-	fb_format |= GRPH_PIPE_CONFIG(pipe_config);
+	fb_format |= (pipe_config << GRPH_CONTROL__GRPH_PIPE_CONFIG__SHIFT);
 
 	dce_v6_0_vga_enable(crtc, false);
 
@@ -2040,7 +2046,7 @@ static int dce_v6_0_crtc_do_set_base(struct drm_crtc *crtc,
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
 	       (u32)fb_location & GRPH_PRIMARY_SURFACE_ADDRESS__GRPH_PRIMARY_SURFACE_ADDRESS_MASK);
 	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       (u32) fb_location & GRPH_PRIMARY_SURFACE_ADDRESS__GRPH_PRIMARY_SURFACE_ADDRESS_MASK);
+	       (u32) fb_location & GRPH_SECONDARY_SURFACE_ADDRESS__GRPH_SECONDARY_SURFACE_ADDRESS_MASK);
 	WREG32(mmGRPH_CONTROL + amdgpu_crtc->crtc_offset, fb_format);
 	WREG32(mmGRPH_SWAP_CNTL + amdgpu_crtc->crtc_offset, fb_swap);
 
@@ -2108,14 +2114,13 @@ static void dce_v6_0_set_interleave(struct drm_crtc *crtc,
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
 		WREG32(mmDATA_FORMAT + amdgpu_crtc->crtc_offset,
-		       INTERLEAVE_EN);
+			DATA_FORMAT__INTERLEAVE_EN_MASK);
 	else
 		WREG32(mmDATA_FORMAT + amdgpu_crtc->crtc_offset, 0);
 }
 
 static void dce_v6_0_crtc_load_lut(struct drm_crtc *crtc)
 {
-
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct amdgpu_device *adev = drm_to_adev(dev);
@@ -2125,15 +2130,15 @@ static void dce_v6_0_crtc_load_lut(struct drm_crtc *crtc)
 	DRM_DEBUG_KMS("%d\n", amdgpu_crtc->crtc_id);
 
 	WREG32(mmINPUT_CSC_CONTROL + amdgpu_crtc->crtc_offset,
-	       ((0 << INPUT_CSC_CONTROL__INPUT_CSC_GRPH_MODE__SHIFT) |
-		(0 << INPUT_CSC_CONTROL__INPUT_CSC_OVL_MODE__SHIFT)));
+	       ((INPUT_CSC_BYPASS << INPUT_CSC_CONTROL__INPUT_CSC_GRPH_MODE__SHIFT) |
+		(INPUT_CSC_BYPASS << INPUT_CSC_CONTROL__INPUT_CSC_OVL_MODE__SHIFT)));
 	WREG32(mmPRESCALE_GRPH_CONTROL + amdgpu_crtc->crtc_offset,
 	       PRESCALE_GRPH_CONTROL__GRPH_PRESCALE_BYPASS_MASK);
 	WREG32(mmPRESCALE_OVL_CONTROL + amdgpu_crtc->crtc_offset,
 	       PRESCALE_OVL_CONTROL__OVL_PRESCALE_BYPASS_MASK);
 	WREG32(mmINPUT_GAMMA_CONTROL + amdgpu_crtc->crtc_offset,
-	       ((0 << INPUT_GAMMA_CONTROL__GRPH_INPUT_GAMMA_MODE__SHIFT) |
-		(0 << INPUT_GAMMA_CONTROL__OVL_INPUT_GAMMA_MODE__SHIFT)));
+	       ((INPUT_GAMMA_USE_LUT << INPUT_GAMMA_CONTROL__GRPH_INPUT_GAMMA_MODE__SHIFT) |
+		(INPUT_GAMMA_USE_LUT << INPUT_GAMMA_CONTROL__OVL_INPUT_GAMMA_MODE__SHIFT)));
 
 	WREG32(mmDC_LUT_CONTROL + amdgpu_crtc->crtc_offset, 0);
 
@@ -2160,19 +2165,19 @@ static void dce_v6_0_crtc_load_lut(struct drm_crtc *crtc)
 	}
 
 	WREG32(mmDEGAMMA_CONTROL + amdgpu_crtc->crtc_offset,
-	       ((0 << DEGAMMA_CONTROL__GRPH_DEGAMMA_MODE__SHIFT) |
-		(0 << DEGAMMA_CONTROL__OVL_DEGAMMA_MODE__SHIFT) |
-		ICON_DEGAMMA_MODE(0) |
-		(0 << DEGAMMA_CONTROL__CURSOR_DEGAMMA_MODE__SHIFT)));
+	       ((DEGAMMA_BYPASS << DEGAMMA_CONTROL__GRPH_DEGAMMA_MODE__SHIFT) |
+		(DEGAMMA_BYPASS << DEGAMMA_CONTROL__OVL_DEGAMMA_MODE__SHIFT) |
+		(DEGAMMA_BYPASS << DEGAMMA_CONTROL__ICON_DEGAMMA_MODE__SHIFT) |
+		(DEGAMMA_BYPASS << DEGAMMA_CONTROL__CURSOR_DEGAMMA_MODE__SHIFT)));
 	WREG32(mmGAMUT_REMAP_CONTROL + amdgpu_crtc->crtc_offset,
-	       ((0 << GAMUT_REMAP_CONTROL__GRPH_GAMUT_REMAP_MODE__SHIFT) |
-		(0 << GAMUT_REMAP_CONTROL__OVL_GAMUT_REMAP_MODE__SHIFT)));
+	       ((GAMUT_REMAP_BYPASS << GAMUT_REMAP_CONTROL__GRPH_GAMUT_REMAP_MODE__SHIFT) |
+		(GAMUT_REMAP_BYPASS << GAMUT_REMAP_CONTROL__OVL_GAMUT_REMAP_MODE__SHIFT)));
 	WREG32(mmREGAMMA_CONTROL + amdgpu_crtc->crtc_offset,
-	       ((0 << REGAMMA_CONTROL__GRPH_REGAMMA_MODE__SHIFT) |
-		(0 << REGAMMA_CONTROL__OVL_REGAMMA_MODE__SHIFT)));
+	       ((REGAMMA_BYPASS << REGAMMA_CONTROL__GRPH_REGAMMA_MODE__SHIFT) |
+		(REGAMMA_BYPASS << REGAMMA_CONTROL__OVL_REGAMMA_MODE__SHIFT)));
 	WREG32(mmOUTPUT_CSC_CONTROL + amdgpu_crtc->crtc_offset,
-	       ((0 << OUTPUT_CSC_CONTROL__OUTPUT_CSC_GRPH_MODE__SHIFT) |
-		(0 << OUTPUT_CSC_CONTROL__OUTPUT_CSC_OVL_MODE__SHIFT)));
+	       ((OUTPUT_CSC_BYPASS << OUTPUT_CSC_CONTROL__OUTPUT_CSC_GRPH_MODE__SHIFT) |
+		(OUTPUT_CSC_BYPASS << OUTPUT_CSC_CONTROL__OUTPUT_CSC_OVL_MODE__SHIFT)));
 	/* XXX match this to the depth of the crtc fmt block, move to modeset? */
 	WREG32(0x1a50 + amdgpu_crtc->crtc_offset, 0);
 
@@ -2267,8 +2272,6 @@ static void dce_v6_0_hide_cursor(struct drm_crtc *crtc)
 	WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset,
 	       (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) |
 	       (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT));
-
-
 }
 
 static void dce_v6_0_show_cursor(struct drm_crtc *crtc)
@@ -2285,7 +2288,6 @@ static void dce_v6_0_show_cursor(struct drm_crtc *crtc)
 	       CUR_CONTROL__CURSOR_EN_MASK |
 	       (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) |
 	       (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT));
-
 }
 
 static int dce_v6_0_cursor_move_locked(struct drm_crtc *crtc,
@@ -2596,7 +2598,6 @@ static bool dce_v6_0_crtc_mode_fixup(struct drm_crtc *crtc,
 				     const struct drm_display_mode *mode,
 				     struct drm_display_mode *adjusted_mode)
 {
-
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct drm_encoder *encoder;
@@ -2669,7 +2670,7 @@ static void dce_v6_0_panic_flush(struct drm_plane *plane)
 
 	/* Disable DC tiling */
 	fb_format = RREG32(mmGRPH_CONTROL + amdgpu_crtc->crtc_offset);
-	fb_format &= ~GRPH_ARRAY_MODE(0x7);
+	fb_format &= ~GRPH_CONTROL__GRPH_ARRAY_MODE_MASK;
 	WREG32(mmGRPH_CONTROL + amdgpu_crtc->crtc_offset, fb_format);
 
 }
@@ -2745,7 +2746,6 @@ static int dce_v6_0_early_init(struct amdgpu_ip_block *ip_block)
 static int dce_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 {
 	int r, i;
-	bool ret;
 	struct amdgpu_device *adev = ip_block->adev;
 
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
@@ -2789,8 +2789,7 @@ static int dce_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	ret = amdgpu_atombios_get_connector_info_from_object_table(adev);
-	if (ret)
+	if (amdgpu_atombios_get_connector_info_from_object_table(adev))
 		amdgpu_display_print_display_setup(adev_to_drm(adev));
 	else
 		return -EINVAL;
@@ -2986,12 +2985,12 @@ static void dce_v6_0_set_crtc_vblank_interrupt_state(struct amdgpu_device *adev,
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 		interrupt_mask = RREG32(mmINT_MASK + reg_block);
-		interrupt_mask &= ~VBLANK_INT_MASK;
+		interrupt_mask &= ~INT_MASK__VBLANK_INT_MASK;
 		WREG32(mmINT_MASK + reg_block, interrupt_mask);
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
 		interrupt_mask = RREG32(mmINT_MASK + reg_block);
-		interrupt_mask |= VBLANK_INT_MASK;
+		interrupt_mask |= INT_MASK__VBLANK_INT_MASK;
 		WREG32(mmINT_MASK + reg_block, interrupt_mask);
 		break;
 	default:
@@ -3006,28 +3005,28 @@ static void dce_v6_0_set_crtc_vline_interrupt_state(struct amdgpu_device *adev,
 
 }
 
-static int dce_v6_0_set_hpd_interrupt_state(struct amdgpu_device *adev,
+static int dce_v6_0_set_hpd_irq_state(struct amdgpu_device *adev,
 					    struct amdgpu_irq_src *src,
-					    unsigned type,
+					    unsigned hpd,
 					    enum amdgpu_interrupt_state state)
 {
 	u32 dc_hpd_int_cntl;
 
-	if (type >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", type);
+	if (hpd >= adev->mode_info.num_hpd) {
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return 0;
 	}
 
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
-		dc_hpd_int_cntl = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[type]);
-		dc_hpd_int_cntl &= ~DC_HPDx_INT_EN;
-		WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[type], dc_hpd_int_cntl);
+		dc_hpd_int_cntl = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd]);
+		dc_hpd_int_cntl &= ~DC_HPD1_INT_CONTROL__DC_HPD1_INT_EN_MASK;
+		WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], dc_hpd_int_cntl);
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
-		dc_hpd_int_cntl = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[type]);
-		dc_hpd_int_cntl |= DC_HPDx_INT_EN;
-		WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[type], dc_hpd_int_cntl);
+		dc_hpd_int_cntl = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd]);
+		dc_hpd_int_cntl |= DC_HPD1_INT_CONTROL__DC_HPD1_INT_EN_MASK;
+		WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], dc_hpd_int_cntl);
 		break;
 	default:
 		break;
@@ -3036,7 +3035,7 @@ static int dce_v6_0_set_hpd_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int dce_v6_0_set_crtc_interrupt_state(struct amdgpu_device *adev,
+static int dce_v6_0_set_crtc_irq_state(struct amdgpu_device *adev,
 					     struct amdgpu_irq_src *src,
 					     unsigned type,
 					     enum amdgpu_interrupt_state state)
@@ -3096,7 +3095,7 @@ static int dce_v6_0_crtc_irq(struct amdgpu_device *adev,
 	switch (entry->src_data[0]) {
 	case 0: /* vblank */
 		if (disp_int & interrupt_status_offsets[crtc].vblank)
-			WREG32(mmVBLANK_STATUS + crtc_offsets[crtc], VBLANK_ACK);
+			WREG32(mmVBLANK_STATUS + crtc_offsets[crtc], VBLANK_STATUS__VBLANK_ACK_MASK);
 		else
 			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
 
@@ -3107,7 +3106,7 @@ static int dce_v6_0_crtc_irq(struct amdgpu_device *adev,
 		break;
 	case 1: /* vline */
 		if (disp_int & interrupt_status_offsets[crtc].vline)
-			WREG32(mmVLINE_STATUS + crtc_offsets[crtc], VLINE_ACK);
+			WREG32(mmVLINE_STATUS + crtc_offsets[crtc], VLINE_STATUS__VLINE_ACK_MASK);
 		else
 			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
 
@@ -3121,7 +3120,7 @@ static int dce_v6_0_crtc_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int dce_v6_0_set_pageflip_interrupt_state(struct amdgpu_device *adev,
+static int dce_v6_0_set_pageflip_irq_state(struct amdgpu_device *adev,
 						 struct amdgpu_irq_src *src,
 						 unsigned type,
 						 enum amdgpu_interrupt_state state)
@@ -3172,7 +3171,7 @@ static int dce_v6_0_pageflip_irq(struct amdgpu_device *adev,
 
 	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 	works = amdgpu_crtc->pflip_works;
-	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){
+	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED) {
 		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
 						"AMDGPU_FLIP_SUBMITTED(%d)\n",
 						amdgpu_crtc->pflip_status,
@@ -3249,12 +3248,10 @@ static const struct amd_ip_funcs dce_v6_0_ip_funcs = {
 	.set_powergating_state = dce_v6_0_set_powergating_state,
 };
 
-static void
-dce_v6_0_encoder_mode_set(struct drm_encoder *encoder,
+static void dce_v6_0_encoder_mode_set(struct drm_encoder *encoder,
 			  struct drm_display_mode *mode,
 			  struct drm_display_mode *adjusted_mode)
 {
-
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	int em = amdgpu_atombios_encoder_get_encoder_mode(encoder);
 
@@ -3274,7 +3271,6 @@ dce_v6_0_encoder_mode_set(struct drm_encoder *encoder,
 
 static void dce_v6_0_encoder_prepare(struct drm_encoder *encoder)
 {
-
 	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
@@ -3314,7 +3310,6 @@ static void dce_v6_0_encoder_prepare(struct drm_encoder *encoder)
 
 static void dce_v6_0_encoder_commit(struct drm_encoder *encoder)
 {
-
 	struct drm_device *dev = encoder->dev;
 	struct amdgpu_device *adev = drm_to_adev(dev);
 
@@ -3325,7 +3320,6 @@ static void dce_v6_0_encoder_commit(struct drm_encoder *encoder)
 
 static void dce_v6_0_encoder_disable(struct drm_encoder *encoder)
 {
-
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct amdgpu_encoder_atom_dig *dig;
 	int em = amdgpu_atombios_encoder_get_encoder_mode(encoder);
@@ -3541,17 +3535,17 @@ static void dce_v6_0_set_display_funcs(struct amdgpu_device *adev)
 }
 
 static const struct amdgpu_irq_src_funcs dce_v6_0_crtc_irq_funcs = {
-	.set = dce_v6_0_set_crtc_interrupt_state,
+	.set = dce_v6_0_set_crtc_irq_state,
 	.process = dce_v6_0_crtc_irq,
 };
 
 static const struct amdgpu_irq_src_funcs dce_v6_0_pageflip_irq_funcs = {
-	.set = dce_v6_0_set_pageflip_interrupt_state,
+	.set = dce_v6_0_set_pageflip_irq_state,
 	.process = dce_v6_0_pageflip_irq,
 };
 
 static const struct amdgpu_irq_src_funcs dce_v6_0_hpd_irq_funcs = {
-	.set = dce_v6_0_set_hpd_interrupt_state,
+	.set = dce_v6_0_set_hpd_irq_state,
 	.process = dce_v6_0_hpd_irq,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 07358546581f..e62ccf9eb73d 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -271,7 +271,7 @@ static void dce_v8_0_hpd_int_ack(struct amdgpu_device *adev,
 	u32 tmp;
 
 	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", hpd);
+		DRM_DEBUG("invalid hpd %d\n", hpd);
 		return;
 	}
 
@@ -3021,7 +3021,7 @@ static void dce_v8_0_set_crtc_vline_interrupt_state(struct amdgpu_device *adev,
 	}
 }
 
-static int dce_v8_0_set_hpd_interrupt_state(struct amdgpu_device *adev,
+static int dce_v8_0_set_hpd_irq_state(struct amdgpu_device *adev,
 					    struct amdgpu_irq_src *src,
 					    unsigned type,
 					    enum amdgpu_interrupt_state state)
@@ -3029,7 +3029,7 @@ static int dce_v8_0_set_hpd_interrupt_state(struct amdgpu_device *adev,
 	u32 dc_hpd_int_cntl;
 
 	if (type >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hdp %d\n", type);
+		DRM_DEBUG("invalid hpd %d\n", type);
 		return 0;
 	}
 
@@ -3051,7 +3051,7 @@ static int dce_v8_0_set_hpd_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int dce_v8_0_set_crtc_interrupt_state(struct amdgpu_device *adev,
+static int dce_v8_0_set_crtc_irq_state(struct amdgpu_device *adev,
 					     struct amdgpu_irq_src *src,
 					     unsigned type,
 					     enum amdgpu_interrupt_state state)
@@ -3136,7 +3136,7 @@ static int dce_v8_0_crtc_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int dce_v8_0_set_pageflip_interrupt_state(struct amdgpu_device *adev,
+static int dce_v8_0_set_pageflip_irq_state(struct amdgpu_device *adev,
 						 struct amdgpu_irq_src *src,
 						 unsigned type,
 						 enum amdgpu_interrupt_state state)
@@ -3547,17 +3547,17 @@ static void dce_v8_0_set_display_funcs(struct amdgpu_device *adev)
 }
 
 static const struct amdgpu_irq_src_funcs dce_v8_0_crtc_irq_funcs = {
-	.set = dce_v8_0_set_crtc_interrupt_state,
+	.set = dce_v8_0_set_crtc_irq_state,
 	.process = dce_v8_0_crtc_irq,
 };
 
 static const struct amdgpu_irq_src_funcs dce_v8_0_pageflip_irq_funcs = {
-	.set = dce_v8_0_set_pageflip_interrupt_state,
+	.set = dce_v8_0_set_pageflip_irq_state,
 	.process = dce_v8_0_pageflip_irq,
 };
 
 static const struct amdgpu_irq_src_funcs dce_v8_0_hpd_irq_funcs = {
-	.set = dce_v8_0_set_hpd_interrupt_state,
+	.set = dce_v8_0_set_hpd_irq_state,
 	.process = dce_v8_0_hpd_irq,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index a63ce747863f..75ea071744eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -368,11 +368,6 @@ static const struct amdgpu_hwip_reg_entry gc_reg_list_10_1[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, mmRLC_GPM_DEBUG_INST_ADDR),
 	SOC15_REG_ENTRY_STR(GC, 0, mmRLC_LX6_CORE_PDEBUG_INST),
 	/* cp header registers */
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME2_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MES_HEADER_DUMP),
 	/* SE status registers */
 	SOC15_REG_ENTRY_STR(GC, 0, mmGRBM_STATUS_SE0),
@@ -421,7 +416,16 @@ static const struct amdgpu_hwip_reg_entry gc_cp_reg_list_10[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_SUSPEND_CNTL_STACK_OFFSET),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_SUSPEND_CNTL_STACK_DW_CNT),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_SUSPEND_WG_STATE_OFFSET),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_DEQUEUE_STATUS)
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_DEQUEUE_STATUS),
+	/* cp header registers */
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
 };
 
 static const struct amdgpu_hwip_reg_entry gc_gfx_queue_reg_list_10[] = {
@@ -448,7 +452,32 @@ static const struct amdgpu_hwip_reg_entry gc_gfx_queue_reg_list_10[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_GFX_MQD_BASE_ADDR),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_GFX_MQD_BASE_ADDR_HI),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI)
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI),
+	/* gfx header registers */
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
 };
 
 static const struct soc15_reg_golden golden_settings_gc_10_1[] = {
@@ -4296,9 +4325,7 @@ static u32 gfx_v10_0_get_csb_size(struct amdgpu_device *adev)
 static void gfx_v10_0_get_csb_buffer(struct amdgpu_device *adev,
 				    volatile u32 *buffer)
 {
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
+	u32 count = 0;
 	int ctx_reg_offset;
 
 	if (adev->gfx.rlc.cs_data == NULL)
@@ -4306,39 +4333,15 @@ static void gfx_v10_0_get_csb_buffer(struct amdgpu_device *adev,
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
+	count = amdgpu_gfx_csb_preamble_start(buffer);
+	count = amdgpu_gfx_csb_data_parser(adev, buffer, count);
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index -
-						PACKET3_SET_CONTEXT_REG_START);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
-
-	ctx_reg_offset =
-		SOC15_REG_OFFSET(GC, 0, mmPA_SC_TILE_STEERING_OVERRIDE) - PACKET3_SET_CONTEXT_REG_START;
+	ctx_reg_offset = SOC15_REG_OFFSET(GC, 0, mmPA_SC_TILE_STEERING_OVERRIDE) - PACKET3_SET_CONTEXT_REG_START;
 	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 1));
 	buffer[count++] = cpu_to_le32(ctx_reg_offset);
 	buffer[count++] = cpu_to_le32(adev->gfx.config.pa_sc_tile_steering_override);
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
+	amdgpu_gfx_csb_preamble_end(buffer, count);
 }
 
 static void gfx_v10_0_rlc_fini(struct amdgpu_device *adev)
@@ -4752,6 +4755,7 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 	int i, j, k, r, ring_id = 0;
 	int xcc_id = 0;
 	struct amdgpu_device *adev = ip_block->adev;
+	int num_queue_per_pipe = 1; /* we only enable 1 KGQ per pipe */
 
 	INIT_DELAYED_WORK(&adev->gfx.idle_work, amdgpu_gfx_profile_idle_work_handler);
 
@@ -4763,7 +4767,7 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 	case IP_VERSION(10, 1, 4):
 		adev->gfx.me.num_me = 1;
 		adev->gfx.me.num_pipe_per_me = 1;
-		adev->gfx.me.num_queue_per_pipe = 1;
+		adev->gfx.me.num_queue_per_pipe = 8;
 		adev->gfx.mec.num_mec = 2;
 		adev->gfx.mec.num_pipe_per_mec = 4;
 		adev->gfx.mec.num_queue_per_pipe = 8;
@@ -4778,7 +4782,7 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 	case IP_VERSION(10, 3, 7):
 		adev->gfx.me.num_me = 1;
 		adev->gfx.me.num_pipe_per_me = 2;
-		adev->gfx.me.num_queue_per_pipe = 1;
+		adev->gfx.me.num_queue_per_pipe = 2;
 		adev->gfx.mec.num_mec = 2;
 		adev->gfx.mec.num_pipe_per_mec = 4;
 		adev->gfx.mec.num_queue_per_pipe = 4;
@@ -4800,7 +4804,7 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 		adev->gfx.cleaner_shader_size = sizeof(gfx_10_1_10_cleaner_shader_hex);
 		if (adev->gfx.me_fw_version >= 101 &&
 		    adev->gfx.pfp_fw_version  >= 158 &&
-		    adev->gfx.mec_fw_version >= 152) {
+		    adev->gfx.mec_fw_version >= 151) {
 			adev->gfx.enable_cleaner_shader = true;
 			r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size);
 			if (r) {
@@ -4810,7 +4814,9 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 		}
 		break;
 	case IP_VERSION(10, 3, 0):
+	case IP_VERSION(10, 3, 1):
 	case IP_VERSION(10, 3, 2):
+	case IP_VERSION(10, 3, 3):
 	case IP_VERSION(10, 3, 4):
 	case IP_VERSION(10, 3, 5):
 		adev->gfx.cleaner_shader_ptr = gfx_10_3_0_cleaner_shader_hex;
@@ -4826,6 +4832,34 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 			}
 		}
 		break;
+	case IP_VERSION(10, 3, 6):
+		adev->gfx.cleaner_shader_ptr = gfx_10_3_0_cleaner_shader_hex;
+		adev->gfx.cleaner_shader_size = sizeof(gfx_10_3_0_cleaner_shader_hex);
+		if (adev->gfx.me_fw_version >= 14 &&
+		    adev->gfx.pfp_fw_version >= 17 &&
+		    adev->gfx.mec_fw_version >= 24) {
+			adev->gfx.enable_cleaner_shader = true;
+			r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size);
+			if (r) {
+				adev->gfx.enable_cleaner_shader = false;
+				dev_err(adev->dev, "Failed to initialize cleaner shader\n");
+			}
+		}
+		break;
+	case IP_VERSION(10, 3, 7):
+		adev->gfx.cleaner_shader_ptr = gfx_10_3_0_cleaner_shader_hex;
+		adev->gfx.cleaner_shader_size = sizeof(gfx_10_3_0_cleaner_shader_hex);
+		if (adev->gfx.me_fw_version >= 4 &&
+		    adev->gfx.pfp_fw_version >= 9 &&
+		    adev->gfx.mec_fw_version >= 12) {
+			adev->gfx.enable_cleaner_shader = true;
+			r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size);
+			if (r) {
+				adev->gfx.enable_cleaner_shader = false;
+				dev_err(adev->dev, "Failed to initialize cleaner shader\n");
+			}
+		}
+		break;
 	default:
 		adev->gfx.enable_cleaner_shader = false;
 		break;
@@ -4886,7 +4920,7 @@ static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block)
 
 	/* set up the gfx ring */
 	for (i = 0; i < adev->gfx.me.num_me; i++) {
-		for (j = 0; j < adev->gfx.me.num_queue_per_pipe; j++) {
+		for (j = 0; j < num_queue_per_pipe; j++) {
 			for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
 				if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
 					continue;
@@ -6114,7 +6148,7 @@ static int gfx_v10_0_cp_gfx_load_pfp_microcode(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, mmCP_PFP_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_PFP_IC_BASE_CNTL, VMID, 0);
@@ -6192,7 +6226,7 @@ static int gfx_v10_0_cp_gfx_load_ce_microcode(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, mmCP_CE_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_CE_IC_BASE_CNTL, VMID, 0);
@@ -6269,7 +6303,7 @@ static int gfx_v10_0_cp_gfx_load_me_microcode(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, mmCP_ME_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_ME_IC_BASE_CNTL, VMID, 0);
@@ -6644,7 +6678,7 @@ static int gfx_v10_0_cp_compute_load_microcode(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, mmCP_CPC_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0);
@@ -9645,9 +9679,14 @@ static void gfx_v10_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printe
 			for (k = 0; k < adev->gfx.mec.num_queue_per_pipe; k++) {
 				drm_printf(p, "\nmec %d, pipe %d, queue %d\n", i, j, k);
 				for (reg = 0; reg < reg_count; reg++) {
-					drm_printf(p, "%-50s \t 0x%08x\n",
-						   gc_cp_reg_list_10[reg].reg_name,
-						   adev->gfx.ip_dump_compute_queues[index + reg]);
+					if (i && gc_cp_reg_list_10[reg].reg_offset == mmCP_MEC_ME1_HEADER_DUMP)
+						drm_printf(p, "%-50s \t 0x%08x\n",
+							   "mmCP_MEC_ME2_HEADER_DUMP",
+							   adev->gfx.ip_dump_compute_queues[index + reg]);
+					else
+						drm_printf(p, "%-50s \t 0x%08x\n",
+							   gc_cp_reg_list_10[reg].reg_name,
+							   adev->gfx.ip_dump_compute_queues[index + reg]);
 				}
 				index += reg_count;
 			}
@@ -9708,9 +9747,13 @@ static void gfx_v10_ip_dump(struct amdgpu_ip_block *ip_block)
 				nv_grbm_select(adev, adev->gfx.me.num_me + i, j, k, 0);
 
 				for (reg = 0; reg < reg_count; reg++) {
-					adev->gfx.ip_dump_compute_queues[index + reg] =
-						RREG32(SOC15_REG_ENTRY_OFFSET(
-							gc_cp_reg_list_10[reg]));
+					if (i && gc_cp_reg_list_10[reg].reg_offset == mmCP_MEC_ME1_HEADER_DUMP)
+						adev->gfx.ip_dump_compute_queues[index + reg] =
+							RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME2_HEADER_DUMP));
+					else
+						adev->gfx.ip_dump_compute_queues[index + reg] =
+							RREG32(SOC15_REG_ENTRY_OFFSET(
+								       gc_cp_reg_list_10[reg]));
 				}
 				index += reg_count;
 			}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index d57db42f9536..afd6d59164bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -48,6 +48,8 @@
 #include "gfx_v11_0_3.h"
 #include "nbio_v4_3.h"
 #include "mes_v11_0.h"
+#include "mes_userqueue.h"
+#include "amdgpu_userq_fence.h"
 
 #define GFX11_NUM_GFX_RINGS		1
 #define GFX11_MEC_HPD_SIZE	2048
@@ -177,9 +179,13 @@ static const struct amdgpu_hwip_reg_entry gc_reg_list_11_0[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_INSTR_PNTR),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_CPC_STATUS),
 	/* cp header registers */
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
 	/* SE status registers */
 	SOC15_REG_ENTRY_STR(GC, 0, regGRBM_STATUS_SE0),
@@ -230,7 +236,16 @@ static const struct amdgpu_hwip_reg_entry gc_cp_reg_list_11[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_SUSPEND_CNTL_STACK_OFFSET),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_SUSPEND_CNTL_STACK_DW_CNT),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_SUSPEND_WG_STATE_OFFSET),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_DEQUEUE_STATUS)
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_DEQUEUE_STATUS),
+	/* cp header registers */
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
 };
 
 static const struct amdgpu_hwip_reg_entry gc_gfx_queue_reg_list_11[] = {
@@ -259,7 +274,24 @@ static const struct amdgpu_hwip_reg_entry gc_gfx_queue_reg_list_11[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BASE_LO),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BASE_HI),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_CMD_BUFSZ),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BUFSZ)
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BUFSZ),
+	/* cp header registers */
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
 };
 
 static const struct soc15_reg_golden golden_settings_gc_11_0[] = {
@@ -580,33 +612,18 @@ static int gfx_v11_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t padding, offset;
-
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		padding = amdgpu_mes_ctx_get_offs(ring,
-						  AMDGPU_MES_CTX_PADDING_OFFS);
-
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, padding);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, padding);
-		*cpu_ptr = cpu_to_le32(0xCAFEDEAD);
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r)
-			return r;
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r)
+		return r;
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
-		cpu_ptr = &adev->wb.wb[index];
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
+	cpu_ptr = &adev->wb.wb[index];
 
-		r = amdgpu_ib_get(adev, NULL, 20, AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
-			goto err1;
-		}
+	r = amdgpu_ib_get(adev, NULL, 20, AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+		goto err1;
 	}
 
 	ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3);
@@ -633,12 +650,10 @@ static int gfx_v11_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	else
 		r = -EINVAL;
 err2:
-	if (!ring->is_mes_queue)
-		amdgpu_ib_free(&ib, NULL);
+	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err1:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -833,9 +848,7 @@ static u32 gfx_v11_0_get_csb_size(struct amdgpu_device *adev)
 static void gfx_v11_0_get_csb_buffer(struct amdgpu_device *adev,
 				    volatile u32 *buffer)
 {
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
+	u32 count = 0;
 	int ctx_reg_offset;
 
 	if (adev->gfx.rlc.cs_data == NULL)
@@ -843,39 +856,15 @@ static void gfx_v11_0_get_csb_buffer(struct amdgpu_device *adev,
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index -
-						PACKET3_SET_CONTEXT_REG_START);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
+	count = amdgpu_gfx_csb_preamble_start(buffer);
+	count = amdgpu_gfx_csb_data_parser(adev, buffer, count);
 
-	ctx_reg_offset =
-		SOC15_REG_OFFSET(GC, 0, regPA_SC_TILE_STEERING_OVERRIDE) - PACKET3_SET_CONTEXT_REG_START;
+	ctx_reg_offset = SOC15_REG_OFFSET(GC, 0, regPA_SC_TILE_STEERING_OVERRIDE) - PACKET3_SET_CONTEXT_REG_START;
 	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 1));
 	buffer[count++] = cpu_to_le32(ctx_reg_offset);
 	buffer[count++] = cpu_to_le32(adev->gfx.config.pa_sc_tile_steering_override);
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
+	amdgpu_gfx_csb_preamble_end(buffer, count);
 }
 
 static void gfx_v11_0_rlc_fini(struct amdgpu_device *adev)
@@ -1056,14 +1045,21 @@ static void gfx_v11_0_select_me_pipe_q(struct amdgpu_device *adev,
 #define MQD_FWWORKAREA_SIZE       484
 #define MQD_FWWORKAREA_ALIGNMENT  256
 
-static int gfx_v11_0_get_gfx_shadow_info(struct amdgpu_device *adev,
+static void gfx_v11_0_get_gfx_shadow_info_nocheck(struct amdgpu_device *adev,
 					 struct amdgpu_gfx_shadow_info *shadow_info)
 {
-	if (adev->gfx.cp_gfx_shadow) {
-		shadow_info->shadow_size = MQD_SHADOW_BASE_SIZE;
-		shadow_info->shadow_alignment = MQD_SHADOW_BASE_ALIGNMENT;
-		shadow_info->csa_size = MQD_FWWORKAREA_SIZE;
-		shadow_info->csa_alignment = MQD_FWWORKAREA_ALIGNMENT;
+	shadow_info->shadow_size = MQD_SHADOW_BASE_SIZE;
+	shadow_info->shadow_alignment = MQD_SHADOW_BASE_ALIGNMENT;
+	shadow_info->csa_size = MQD_FWWORKAREA_SIZE;
+	shadow_info->csa_alignment = MQD_FWWORKAREA_ALIGNMENT;
+}
+
+static int gfx_v11_0_get_gfx_shadow_info(struct amdgpu_device *adev,
+					 struct amdgpu_gfx_shadow_info *shadow_info,
+					 bool skip_check)
+{
+	if (adev->gfx.cp_gfx_shadow || skip_check) {
+		gfx_v11_0_get_gfx_shadow_info_nocheck(adev, shadow_info);
 		return 0;
 	} else {
 		memset(shadow_info, 0, sizeof(struct amdgpu_gfx_shadow_info));
@@ -1136,6 +1132,10 @@ static int gfx_v11_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id,
 
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
+	if (adev->gfx.disable_kq) {
+		ring->no_scheduler = true;
+		ring->no_user_submission = true;
+	}
 
 	if (!ring_id)
 		ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1;
@@ -1568,24 +1568,18 @@ static void gfx_v11_0_alloc_ip_dump(struct amdgpu_device *adev)
 
 static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 {
-	int i, j, k, r, ring_id = 0;
+	int i, j, k, r, ring_id;
 	int xcc_id = 0;
 	struct amdgpu_device *adev = ip_block->adev;
+	int num_queue_per_pipe = 1; /* we only enable 1 KGQ per pipe */
 
 	INIT_DELAYED_WORK(&adev->gfx.idle_work, amdgpu_gfx_profile_idle_work_handler);
 
 	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
 	case IP_VERSION(11, 0, 0):
+	case IP_VERSION(11, 0, 1):
 	case IP_VERSION(11, 0, 2):
 	case IP_VERSION(11, 0, 3):
-		adev->gfx.me.num_me = 1;
-		adev->gfx.me.num_pipe_per_me = 1;
-		adev->gfx.me.num_queue_per_pipe = 1;
-		adev->gfx.mec.num_mec = 1;
-		adev->gfx.mec.num_pipe_per_mec = 4;
-		adev->gfx.mec.num_queue_per_pipe = 4;
-		break;
-	case IP_VERSION(11, 0, 1):
 	case IP_VERSION(11, 0, 4):
 	case IP_VERSION(11, 5, 0):
 	case IP_VERSION(11, 5, 1):
@@ -1593,7 +1587,7 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 	case IP_VERSION(11, 5, 3):
 		adev->gfx.me.num_me = 1;
 		adev->gfx.me.num_pipe_per_me = 1;
-		adev->gfx.me.num_queue_per_pipe = 1;
+		adev->gfx.me.num_queue_per_pipe = 2;
 		adev->gfx.mec.num_mec = 1;
 		adev->gfx.mec.num_pipe_per_mec = 4;
 		adev->gfx.mec.num_queue_per_pipe = 4;
@@ -1612,6 +1606,35 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 	case IP_VERSION(11, 0, 0):
 	case IP_VERSION(11, 0, 2):
 	case IP_VERSION(11, 0, 3):
+		if (!adev->gfx.disable_uq &&
+		    adev->gfx.me_fw_version  >= 2390 &&
+		    adev->gfx.pfp_fw_version >= 2530 &&
+		    adev->gfx.mec_fw_version >= 2600 &&
+		    adev->mes.fw_version[0] >= 120) {
+			adev->userq_funcs[AMDGPU_HW_IP_GFX] = &userq_mes_funcs;
+			adev->userq_funcs[AMDGPU_HW_IP_COMPUTE] = &userq_mes_funcs;
+		}
+		break;
+	case IP_VERSION(11, 0, 1):
+	case IP_VERSION(11, 0, 4):
+	case IP_VERSION(11, 5, 0):
+	case IP_VERSION(11, 5, 1):
+	case IP_VERSION(11, 5, 2):
+	case IP_VERSION(11, 5, 3):
+		/* add firmware version checks here */
+		if (0 && !adev->gfx.disable_uq) {
+			adev->userq_funcs[AMDGPU_HW_IP_GFX] = &userq_mes_funcs;
+			adev->userq_funcs[AMDGPU_HW_IP_COMPUTE] = &userq_mes_funcs;
+		}
+		break;
+	default:
+		break;
+	}
+
+	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
+	case IP_VERSION(11, 0, 0):
+	case IP_VERSION(11, 0, 2):
+	case IP_VERSION(11, 0, 3):
 		adev->gfx.cleaner_shader_ptr = gfx_11_0_3_cleaner_shader_hex;
 		adev->gfx.cleaner_shader_size = sizeof(gfx_11_0_3_cleaner_shader_hex);
 		if (adev->gfx.me_fw_version  >= 2280 &&
@@ -1640,6 +1663,34 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 			}
 		}
 		break;
+	case IP_VERSION(11, 5, 2):
+		adev->gfx.cleaner_shader_ptr = gfx_11_0_3_cleaner_shader_hex;
+		adev->gfx.cleaner_shader_size = sizeof(gfx_11_0_3_cleaner_shader_hex);
+		if (adev->gfx.me_fw_version  >= 12 &&
+		    adev->gfx.pfp_fw_version >= 15 &&
+		    adev->gfx.mec_fw_version >= 15) {
+			adev->gfx.enable_cleaner_shader = true;
+			r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size);
+			if (r) {
+				adev->gfx.enable_cleaner_shader = false;
+				dev_err(adev->dev, "Failed to initialize cleaner shader\n");
+			}
+		}
+		break;
+	case IP_VERSION(11, 5, 3):
+		adev->gfx.cleaner_shader_ptr = gfx_11_0_3_cleaner_shader_hex;
+		adev->gfx.cleaner_shader_size = sizeof(gfx_11_0_3_cleaner_shader_hex);
+		if (adev->gfx.me_fw_version  >= 7 &&
+		    adev->gfx.pfp_fw_version >= 8 &&
+		    adev->gfx.mec_fw_version >= 8) {
+			adev->gfx.enable_cleaner_shader = true;
+			r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size);
+			if (r) {
+				adev->gfx.enable_cleaner_shader = false;
+				dev_err(adev->dev, "Failed to initialize cleaner shader\n");
+			}
+		}
+		break;
 	default:
 		adev->gfx.enable_cleaner_shader = false;
 		break;
@@ -1701,37 +1752,42 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 		return r;
 	}
 
-	/* set up the gfx ring */
-	for (i = 0; i < adev->gfx.me.num_me; i++) {
-		for (j = 0; j < adev->gfx.me.num_queue_per_pipe; j++) {
-			for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
-				if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
-					continue;
-
-				r = gfx_v11_0_gfx_ring_init(adev, ring_id,
-							    i, k, j);
-				if (r)
-					return r;
-				ring_id++;
+	if (adev->gfx.num_gfx_rings) {
+		ring_id = 0;
+		/* set up the gfx ring */
+		for (i = 0; i < adev->gfx.me.num_me; i++) {
+			for (j = 0; j < num_queue_per_pipe; j++) {
+				for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
+					if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
+						continue;
+
+					r = gfx_v11_0_gfx_ring_init(adev, ring_id,
+								    i, k, j);
+					if (r)
+						return r;
+					ring_id++;
+				}
 			}
 		}
 	}
 
-	ring_id = 0;
-	/* set up the compute queues - allocate horizontally across pipes */
-	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
-		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
-			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
-								     k, j))
-					continue;
+	if (adev->gfx.num_compute_rings) {
+		ring_id = 0;
+		/* set up the compute queues - allocate horizontally across pipes */
+		for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
+			for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
+				for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
+					if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+									     k, j))
+						continue;
 
-				r = gfx_v11_0_compute_ring_init(adev, ring_id,
-								i, k, j);
-				if (r)
-					return r;
+					r = gfx_v11_0_compute_ring_init(adev, ring_id,
+									i, k, j);
+					if (r)
+						return r;
 
-				ring_id++;
+					ring_id++;
+				}
 			}
 		}
 	}
@@ -2428,7 +2484,7 @@ static int gfx_v11_0_config_me_cache(struct amdgpu_device *adev, uint64_t addr)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, regCP_ME_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_ME_IC_BASE_CNTL, VMID, 0);
@@ -2472,7 +2528,7 @@ static int gfx_v11_0_config_pfp_cache(struct amdgpu_device *adev, uint64_t addr)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, regCP_PFP_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_PFP_IC_BASE_CNTL, VMID, 0);
@@ -2517,7 +2573,7 @@ static int gfx_v11_0_config_mec_cache(struct amdgpu_device *adev, uint64_t addr)
 	}
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	tmp = RREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_CNTL);
 	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0);
@@ -3153,7 +3209,7 @@ static int gfx_v11_0_cp_gfx_load_pfp_microcode_rs64(struct amdgpu_device *adev)
 	amdgpu_bo_unreserve(adev->gfx.pfp.pfp_fw_data_obj);
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	WREG32_SOC15(GC, 0, regCP_PFP_IC_BASE_LO,
 		lower_32_bits(adev->gfx.pfp.pfp_fw_gpu_addr));
@@ -3371,7 +3427,7 @@ static int gfx_v11_0_cp_gfx_load_me_microcode_rs64(struct amdgpu_device *adev)
 	amdgpu_bo_unreserve(adev->gfx.me.me_fw_data_obj);
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	WREG32_SOC15(GC, 0, regCP_ME_IC_BASE_LO,
 		lower_32_bits(adev->gfx.me.me_fw_gpu_addr));
@@ -4061,6 +4117,8 @@ static int gfx_v11_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 #ifdef __BIG_ENDIAN
 	tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, BUF_SWAP, 1);
 #endif
+	if (prop->tmz_queue)
+		tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, TMZ_MATCH, 1);
 	mqd->cp_gfx_hqd_cntl = tmp;
 
 	/* set up cp_doorbell_control */
@@ -4081,6 +4139,16 @@ static int gfx_v11_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 	/* active the queue */
 	mqd->cp_gfx_hqd_active = 1;
 
+	/* set gfx UQ items */
+	mqd->shadow_base_lo = lower_32_bits(prop->shadow_addr);
+	mqd->shadow_base_hi = upper_32_bits(prop->shadow_addr);
+	mqd->gds_bkup_base_lo = lower_32_bits(prop->gds_bkup_addr);
+	mqd->gds_bkup_base_hi = upper_32_bits(prop->gds_bkup_addr);
+	mqd->fw_work_area_base_lo = lower_32_bits(prop->csa_addr);
+	mqd->fw_work_area_base_hi = upper_32_bits(prop->csa_addr);
+	mqd->fence_address_lo = lower_32_bits(prop->fence_address);
+	mqd->fence_address_hi = upper_32_bits(prop->fence_address);
+
 	return 0;
 }
 
@@ -4205,6 +4273,8 @@ static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 			    prop->allow_tunneling);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	if (prop->tmz_queue)
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TMZ, 1);
 	mqd->cp_hqd_pq_control = tmp;
 
 	/* set the wb address whether it's enabled or not */
@@ -4256,6 +4326,10 @@ static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 
 	mqd->cp_hqd_active = prop->hqd_active;
 
+	/* set UQ fenceaddress */
+	mqd->fence_address_lo = lower_32_bits(prop->fence_address);
+	mqd->fence_address_hi = upper_32_bits(prop->fence_address);
+
 	return 0;
 }
 
@@ -4509,11 +4583,23 @@ static int gfx_v11_0_cp_resume(struct amdgpu_device *adev)
 			return r;
 	}
 
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-		ring = &adev->gfx.gfx_ring[i];
-		r = amdgpu_ring_test_helper(ring);
-		if (r)
-			return r;
+	if (adev->gfx.disable_kq) {
+		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+			ring = &adev->gfx.gfx_ring[i];
+			/* we don't want to set ring->ready */
+			r = amdgpu_ring_test_ring(ring);
+			if (r)
+				return r;
+		}
+		if (amdgpu_async_gfx_ring)
+			amdgpu_gfx_disable_kgq(adev, 0);
+	} else {
+		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+			ring = &adev->gfx.gfx_ring[i];
+			r = amdgpu_ring_test_helper(ring);
+			if (r)
+				return r;
+		}
 	}
 
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
@@ -4541,7 +4627,7 @@ static int gfx_v11_0_gfxhub_enable(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
 		false : true;
@@ -4722,6 +4808,49 @@ static int gfx_v11_0_hw_init(struct amdgpu_ip_block *ip_block)
 	return r;
 }
 
+static int gfx_v11_0_set_userq_eop_interrupts(struct amdgpu_device *adev,
+					      bool enable)
+{
+	unsigned int irq_type;
+	int m, p, r;
+
+	if (adev->userq_funcs[AMDGPU_HW_IP_GFX]) {
+		for (m = 0; m < adev->gfx.me.num_me; m++) {
+			for (p = 0; p < adev->gfx.me.num_pipe_per_me; p++) {
+				irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + p;
+				if (enable)
+					r = amdgpu_irq_get(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				else
+					r = amdgpu_irq_put(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				if (r)
+					return r;
+			}
+		}
+	}
+
+	if (adev->userq_funcs[AMDGPU_HW_IP_COMPUTE]) {
+		for (m = 0; m < adev->gfx.mec.num_mec; ++m) {
+			for (p = 0; p < adev->gfx.mec.num_pipe_per_mec; p++) {
+				irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
+					+ (m * adev->gfx.mec.num_pipe_per_mec)
+					+ p;
+				if (enable)
+					r = amdgpu_irq_get(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				else
+					r = amdgpu_irq_put(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				if (r)
+					return r;
+			}
+		}
+	}
+
+	return 0;
+}
+
 static int gfx_v11_0_hw_fini(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -4731,9 +4860,11 @@ static int gfx_v11_0_hw_fini(struct amdgpu_ip_block *ip_block)
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.bad_op_irq, 0);
+	gfx_v11_0_set_userq_eop_interrupts(adev, false);
 
 	if (!adev->no_hw_access) {
-		if (amdgpu_async_gfx_ring) {
+		if (amdgpu_async_gfx_ring &&
+		    !adev->gfx.disable_kq) {
 			if (amdgpu_gfx_disable_kgq(adev, 0))
 				DRM_ERROR("KGQ disable failed\n");
 		}
@@ -5059,11 +5190,36 @@ static int gfx_v11_0_early_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 
+	switch (amdgpu_user_queue) {
+	case -1:
+	case 0:
+	default:
+		adev->gfx.disable_kq = false;
+		adev->gfx.disable_uq = true;
+		break;
+	case 1:
+		adev->gfx.disable_kq = false;
+		adev->gfx.disable_uq = false;
+		break;
+	case 2:
+		adev->gfx.disable_kq = true;
+		adev->gfx.disable_uq = false;
+		break;
+	}
+
 	adev->gfx.funcs = &gfx_v11_0_gfx_funcs;
 
-	adev->gfx.num_gfx_rings = GFX11_NUM_GFX_RINGS;
-	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
-					  AMDGPU_MAX_COMPUTE_RINGS);
+	if (adev->gfx.disable_kq) {
+		/* We need one GFX ring temporarily to set up
+		 * the clear state.
+		 */
+		adev->gfx.num_gfx_rings = 1;
+		adev->gfx.num_compute_rings = 0;
+	} else {
+		adev->gfx.num_gfx_rings = GFX11_NUM_GFX_RINGS;
+		adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
+						  AMDGPU_MAX_COMPUTE_RINGS);
+	}
 
 	gfx_v11_0_set_kiq_pm4_funcs(adev);
 	gfx_v11_0_set_ring_funcs(adev);
@@ -5094,6 +5250,11 @@ static int gfx_v11_0_late_init(struct amdgpu_ip_block *ip_block)
 	r = amdgpu_irq_get(adev, &adev->gfx.bad_op_irq, 0);
 	if (r)
 		return r;
+
+	r = gfx_v11_0_set_userq_eop_interrupts(adev, true);
+	if (r)
+		return r;
+
 	return 0;
 }
 
@@ -5691,10 +5852,6 @@ static void gfx_v11_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 				    (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false);
 	}
 
-	if (ring->is_mes_queue)
-		/* inherit vmid from mqd */
-		control |= 0x400000;
-
 	amdgpu_ring_write(ring, header);
 	BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
 	amdgpu_ring_write(ring,
@@ -5714,10 +5871,6 @@ static void gfx_v11_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
 
-	if (ring->is_mes_queue)
-		/* inherit vmid from mqd */
-		control |= 0x40000000;
-
 	/* Currently, there is a high possibility to get wave ID mismatch
 	 * between ME and GDS, leading to a hw deadlock, because ME generates
 	 * different wave IDs than the GDS expects. This situation happens
@@ -5775,8 +5928,7 @@ static void gfx_v11_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
 	amdgpu_ring_write(ring, upper_32_bits(addr));
 	amdgpu_ring_write(ring, lower_32_bits(seq));
 	amdgpu_ring_write(ring, upper_32_bits(seq));
-	amdgpu_ring_write(ring, ring->is_mes_queue ?
-			 (ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0);
+	amdgpu_ring_write(ring, 0);
 }
 
 static void gfx_v11_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
@@ -5804,10 +5956,7 @@ static void gfx_v11_0_ring_invalidate_tlbs(struct amdgpu_ring *ring,
 static void gfx_v11_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 					 unsigned vmid, uint64_t pd_addr)
 {
-	if (ring->is_mes_queue)
-		gfx_v11_0_ring_invalidate_tlbs(ring, 0, 0, false, 0);
-	else
-		amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
+	amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
 
 	/* compute doesn't have PFP */
 	if (ring->funcs->type == AMDGPU_RING_TYPE_GFX) {
@@ -6036,28 +6185,13 @@ static void gfx_v11_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
 	void *de_payload_cpu_addr;
 	int cnt;
 
-	if (ring->is_mes_queue) {
-		offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				  gfx[0].gfx_meta_data) +
-			offsetof(struct v10_gfx_meta_data, de_payload);
-		de_payload_gpu_addr =
-			amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		de_payload_cpu_addr =
-			amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				  gfx[0].gds_backup) +
-			offsetof(struct v10_gfx_meta_data, de_payload);
-		gds_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-	} else {
-		offset = offsetof(struct v10_gfx_meta_data, de_payload);
-		de_payload_gpu_addr = amdgpu_csa_vaddr(ring->adev) + offset;
-		de_payload_cpu_addr = adev->virt.csa_cpu_addr + offset;
+	offset = offsetof(struct v10_gfx_meta_data, de_payload);
+	de_payload_gpu_addr = amdgpu_csa_vaddr(ring->adev) + offset;
+	de_payload_cpu_addr = adev->virt.csa_cpu_addr + offset;
 
-		gds_addr = ALIGN(amdgpu_csa_vaddr(ring->adev) +
-				 AMDGPU_CSA_SIZE - adev->gds.gds_size,
-				 PAGE_SIZE);
-	}
+	gds_addr = ALIGN(amdgpu_csa_vaddr(ring->adev) +
+			 AMDGPU_CSA_SIZE - adev->gds.gds_size,
+			 PAGE_SIZE);
 
 	de_payload.gds_backup_addrlo = lower_32_bits(gds_addr);
 	de_payload.gds_backup_addrhi = upper_32_bits(gds_addr);
@@ -6296,25 +6430,23 @@ static int gfx_v11_0_eop_irq(struct amdgpu_device *adev,
 			     struct amdgpu_irq_src *source,
 			     struct amdgpu_iv_entry *entry)
 {
-	int i;
+	u32 doorbell_offset = entry->src_data[0];
 	u8 me_id, pipe_id, queue_id;
 	struct amdgpu_ring *ring;
-	uint32_t mes_queue_id = entry->src_data[0];
+	int i;
 
 	DRM_DEBUG("IH: CP EOP\n");
 
-	if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) {
-		struct amdgpu_mes_queue *queue;
-
-		mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK;
+	if (adev->enable_mes && doorbell_offset) {
+		struct amdgpu_userq_fence_driver *fence_drv = NULL;
+		struct xarray *xa = &adev->userq_xa;
+		unsigned long flags;
 
-		spin_lock(&adev->mes.queue_id_lock);
-		queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id);
-		if (queue) {
-			DRM_DEBUG("process mes queue id = %d\n", mes_queue_id);
-			amdgpu_fence_process(queue->ring);
-		}
-		spin_unlock(&adev->mes.queue_id_lock);
+		xa_lock_irqsave(xa, flags);
+		fence_drv = xa_load(xa, doorbell_offset);
+		if (fence_drv)
+			amdgpu_userq_fence_driver_process(fence_drv);
+		xa_unlock_irqrestore(xa, flags);
 	} else {
 		me_id = (entry->ring_id & 0x0c) >> 2;
 		pipe_id = (entry->ring_id & 0x03) >> 0;
@@ -6481,27 +6613,29 @@ static void gfx_v11_0_handle_priv_fault(struct amdgpu_device *adev,
 	pipe_id = (entry->ring_id & 0x03) >> 0;
 	queue_id = (entry->ring_id & 0x70) >> 4;
 
-	switch (me_id) {
-	case 0:
-		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-			ring = &adev->gfx.gfx_ring[i];
-			if (ring->me == me_id && ring->pipe == pipe_id &&
-			    ring->queue == queue_id)
-				drm_sched_fault(&ring->sched);
-		}
-		break;
-	case 1:
-	case 2:
-		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			ring = &adev->gfx.compute_ring[i];
-			if (ring->me == me_id && ring->pipe == pipe_id &&
-			    ring->queue == queue_id)
-				drm_sched_fault(&ring->sched);
+	if (!adev->gfx.disable_kq) {
+		switch (me_id) {
+		case 0:
+			for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+				ring = &adev->gfx.gfx_ring[i];
+				if (ring->me == me_id && ring->pipe == pipe_id &&
+				    ring->queue == queue_id)
+					drm_sched_fault(&ring->sched);
+			}
+			break;
+		case 1:
+		case 2:
+			for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+				ring = &adev->gfx.compute_ring[i];
+				if (ring->me == me_id && ring->pipe == pipe_id &&
+				    ring->queue == queue_id)
+					drm_sched_fault(&ring->sched);
+			}
+			break;
+		default:
+			BUG();
+			break;
 		}
-		break;
-	default:
-		BUG();
-		break;
 	}
 }
 
@@ -6609,6 +6743,69 @@ static void gfx_v11_0_emit_mem_sync(struct amdgpu_ring *ring)
 	amdgpu_ring_write(ring, gcr_cntl); /* GCR_CNTL */
 }
 
+static bool gfx_v11_pipe_reset_support(struct amdgpu_device *adev)
+{
+	/* Disable the pipe reset until the CPFW fully support it.*/
+	dev_warn_once(adev->dev, "The CPFW hasn't support pipe reset yet.\n");
+	return false;
+}
+
+
+static int gfx_v11_reset_gfx_pipe(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t reset_pipe = 0, clean_pipe = 0;
+	int r;
+
+	if (!gfx_v11_pipe_reset_support(adev))
+		return -EOPNOTSUPP;
+
+	gfx_v11_0_set_safe_mode(adev, 0);
+	mutex_lock(&adev->srbm_mutex);
+	soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+
+	switch (ring->pipe) {
+	case 0:
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   PFP_PIPE0_RESET, 1);
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   ME_PIPE0_RESET, 1);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   PFP_PIPE0_RESET, 0);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   ME_PIPE0_RESET, 0);
+		break;
+	case 1:
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   PFP_PIPE1_RESET, 1);
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   ME_PIPE1_RESET, 1);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   PFP_PIPE1_RESET, 0);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   ME_PIPE1_RESET, 0);
+		break;
+	default:
+		break;
+	}
+
+	WREG32_SOC15(GC, 0, regCP_ME_CNTL, reset_pipe);
+	WREG32_SOC15(GC, 0, regCP_ME_CNTL, clean_pipe);
+
+	r = (RREG32(SOC15_REG_OFFSET(GC, 0, regCP_GFX_RS64_INSTR_PNTR1)) << 2) -
+						RS64_FW_UC_START_ADDR_LO;
+	soc21_grbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+	gfx_v11_0_unset_safe_mode(adev, 0);
+
+	dev_info(adev->dev, "The ring %s pipe reset to the ME firmware start PC: %s\n", ring->name,
+			r == 0 ? "successfully" : "failed");
+	/* FIXME: Sometimes driver can't cache the ME firmware start PC correctly,
+	 * so the pipe reset status relies on the later gfx ring test result.
+	 */
+	return 0;
+}
+
 static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -6618,8 +6815,13 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid)
 		return -EINVAL;
 
 	r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false);
-	if (r)
-		return r;
+	if (r) {
+
+		dev_warn(adev->dev, "reset via MES failed and try pipe reset %d\n", r);
+		r = gfx_v11_reset_gfx_pipe(ring);
+		if (r)
+			return r;
+	}
 
 	r = gfx_v11_0_kgq_init_queue(ring, true);
 	if (r) {
@@ -6636,6 +6838,136 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid)
 	return amdgpu_ring_test_ring(ring);
 }
 
+static int gfx_v11_0_reset_compute_pipe(struct amdgpu_ring *ring)
+{
+
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t reset_pipe = 0, clean_pipe = 0;
+	int r;
+
+	if (!gfx_v11_pipe_reset_support(adev))
+		return -EOPNOTSUPP;
+
+	gfx_v11_0_set_safe_mode(adev, 0);
+	mutex_lock(&adev->srbm_mutex);
+	soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+
+	reset_pipe = RREG32_SOC15(GC, 0, regCP_MEC_RS64_CNTL);
+	clean_pipe = reset_pipe;
+
+	if (adev->gfx.rs64_enable) {
+
+		switch (ring->pipe) {
+		case 0:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE0_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE0_RESET, 0);
+			break;
+		case 1:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE1_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE1_RESET, 0);
+			break;
+		case 2:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE2_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE2_RESET, 0);
+			break;
+		case 3:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE3_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE3_RESET, 0);
+			break;
+		default:
+			break;
+		}
+		WREG32_SOC15(GC, 0, regCP_MEC_RS64_CNTL, reset_pipe);
+		WREG32_SOC15(GC, 0, regCP_MEC_RS64_CNTL, clean_pipe);
+		r = (RREG32_SOC15(GC, 0, regCP_MEC_RS64_INSTR_PNTR) << 2) -
+					RS64_FW_UC_START_ADDR_LO;
+	} else {
+		if (ring->me == 1) {
+			switch (ring->pipe) {
+			case 0:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE0_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE0_RESET, 0);
+				break;
+			case 1:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE1_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE1_RESET, 0);
+				break;
+			case 2:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE2_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE2_RESET, 0);
+				break;
+			case 3:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE3_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE3_RESET, 0);
+				break;
+			default:
+				break;
+			}
+			/* mec1 fw pc: CP_MEC1_INSTR_PNTR */
+		} else {
+			switch (ring->pipe) {
+			case 0:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE0_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE0_RESET, 0);
+				break;
+			case 1:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE1_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE1_RESET, 0);
+				break;
+			case 2:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE2_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE2_RESET, 0);
+				break;
+			case 3:
+				reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE3_RESET, 1);
+				clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME2_PIPE3_RESET, 0);
+				break;
+			default:
+				break;
+			}
+			/* mec2 fw pc: CP:CP_MEC2_INSTR_PNTR */
+		}
+		WREG32_SOC15(GC, 0, regCP_MEC_CNTL, reset_pipe);
+		WREG32_SOC15(GC, 0, regCP_MEC_CNTL, clean_pipe);
+		r = RREG32(SOC15_REG_OFFSET(GC, 0, regCP_MEC1_INSTR_PNTR));
+	}
+
+	soc21_grbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+	gfx_v11_0_unset_safe_mode(adev, 0);
+
+	dev_info(adev->dev, "The ring %s pipe resets to MEC FW start PC: %s\n", ring->name,
+			r == 0 ? "successfully" : "failed");
+	/*FIXME:Sometimes driver can't cache the MEC firmware start PC correctly, so the pipe
+	 * reset status relies on the compute ring test result.
+	 */
+	return 0;
+}
+
 static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -6646,8 +6978,10 @@ static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid)
 
 	r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, true);
 	if (r) {
-		dev_err(adev->dev, "reset via MMIO failed %d\n", r);
-		return r;
+		dev_warn(adev->dev, "fail(%d) to reset kcq and try pipe reset\n", r);
+		r = gfx_v11_0_reset_compute_pipe(ring);
+		if (r)
+			return r;
 	}
 
 	r = gfx_v11_0_kcq_init_queue(ring, true);
@@ -6693,9 +7027,14 @@ static void gfx_v11_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printe
 			for (k = 0; k < adev->gfx.mec.num_queue_per_pipe; k++) {
 				drm_printf(p, "\nmec %d, pipe %d, queue %d\n", i, j, k);
 				for (reg = 0; reg < reg_count; reg++) {
-					drm_printf(p, "%-50s \t 0x%08x\n",
-						   gc_cp_reg_list_11[reg].reg_name,
-						   adev->gfx.ip_dump_compute_queues[index + reg]);
+					if (i && gc_cp_reg_list_11[reg].reg_offset == regCP_MEC_ME1_HEADER_DUMP)
+						drm_printf(p, "%-50s \t 0x%08x\n",
+							   "regCP_MEC_ME2_HEADER_DUMP",
+							   adev->gfx.ip_dump_compute_queues[index + reg]);
+					else
+						drm_printf(p, "%-50s \t 0x%08x\n",
+							   gc_cp_reg_list_11[reg].reg_name,
+							   adev->gfx.ip_dump_compute_queues[index + reg]);
 				}
 				index += reg_count;
 			}
@@ -6755,9 +7094,16 @@ static void gfx_v11_ip_dump(struct amdgpu_ip_block *ip_block)
 				/* ME0 is for GFX so start from 1 for CP */
 				soc21_grbm_select(adev, adev->gfx.me.num_me + i, j, k, 0);
 				for (reg = 0; reg < reg_count; reg++) {
-					adev->gfx.ip_dump_compute_queues[index + reg] =
-						RREG32(SOC15_REG_ENTRY_OFFSET(
-							gc_cp_reg_list_11[reg]));
+					if (i &&
+					    gc_cp_reg_list_11[reg].reg_offset ==
+						    regCP_MEC_ME1_HEADER_DUMP)
+						adev->gfx.ip_dump_compute_queues[index + reg] =
+							RREG32(SOC15_REG_OFFSET(GC, 0,
+							       regCP_MEC_ME2_HEADER_DUMP));
+					else
+						adev->gfx.ip_dump_compute_queues[index + reg] =
+							RREG32(SOC15_REG_ENTRY_OFFSET(
+								       gc_cp_reg_list_11[reg]));
 				}
 				index += reg_count;
 			}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index e7b58e470292..f09d96bfee16 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -44,6 +44,8 @@
 #include "gfx_v12_0.h"
 #include "nbif_v6_3_1.h"
 #include "mes_v12_0.h"
+#include "mes_userqueue.h"
+#include "amdgpu_userq_fence.h"
 
 #define GFX12_NUM_GFX_RINGS	1
 #define GFX12_MEC_HPD_SIZE	2048
@@ -133,11 +135,14 @@ static const struct amdgpu_hwip_reg_entry gc_reg_list_12_0[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_GFX_RS64_INSTR_PNTR0),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_GFX_RS64_INSTR_PNTR1),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_RS64_INSTR_PNTR),
-
 	/* cp header registers */
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_MES_HEADER_DUMP),
 	/* SE status registers */
 	SOC15_REG_ENTRY_STR(GC, 0, regGRBM_STATUS_SE0),
@@ -186,7 +191,16 @@ static const struct amdgpu_hwip_reg_entry gc_cp_reg_list_12[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_SUSPEND_CNTL_STACK_OFFSET),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_SUSPEND_CNTL_STACK_DW_CNT),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_SUSPEND_WG_STATE_OFFSET),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_DEQUEUE_STATUS)
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_DEQUEUE_STATUS),
+	/* cp header registers */
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
 };
 
 static const struct amdgpu_hwip_reg_entry gc_gfx_queue_reg_list_12[] = {
@@ -215,7 +229,24 @@ static const struct amdgpu_hwip_reg_entry gc_gfx_queue_reg_list_12[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BASE_LO),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BASE_HI),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_CMD_BUFSZ),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BUFSZ)
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_IB1_BUFSZ),
+	/* cp header registers */
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_ME_HEADER_DUMP),
 };
 
 static const struct soc15_reg_golden golden_settings_gc_12_0_rev0[] = {
@@ -475,33 +506,18 @@ static int gfx_v12_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t padding, offset;
-
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		padding = amdgpu_mes_ctx_get_offs(ring,
-						  AMDGPU_MES_CTX_PADDING_OFFS);
-
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, padding);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, padding);
-		*cpu_ptr = cpu_to_le32(0xCAFEDEAD);
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r)
-			return r;
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r)
+		return r;
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
-		cpu_ptr = &adev->wb.wb[index];
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
+	cpu_ptr = &adev->wb.wb[index];
 
-		r = amdgpu_ib_get(adev, NULL, 16, AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			dev_err(adev->dev, "amdgpu: failed to get ib (%ld).\n", r);
-			goto err1;
-		}
+	r = amdgpu_ib_get(adev, NULL, 16, AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		dev_err(adev->dev, "amdgpu: failed to get ib (%ld).\n", r);
+		goto err1;
 	}
 
 	ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3);
@@ -528,12 +544,10 @@ static int gfx_v12_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	else
 		r = -EINVAL;
 err2:
-	if (!ring->is_mes_queue)
-		amdgpu_ib_free(&ib, NULL);
+	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err1:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -881,6 +895,34 @@ static void gfx_v12_0_select_me_pipe_q(struct amdgpu_device *adev,
 	soc24_grbm_select(adev, me, pipe, q, vm);
 }
 
+/* all sizes are in bytes */
+#define MQD_SHADOW_BASE_SIZE      73728
+#define MQD_SHADOW_BASE_ALIGNMENT 256
+#define MQD_FWWORKAREA_SIZE       484
+#define MQD_FWWORKAREA_ALIGNMENT  256
+
+static void gfx_v12_0_get_gfx_shadow_info_nocheck(struct amdgpu_device *adev,
+						  struct amdgpu_gfx_shadow_info *shadow_info)
+{
+	shadow_info->shadow_size = MQD_SHADOW_BASE_SIZE;
+	shadow_info->shadow_alignment = MQD_SHADOW_BASE_ALIGNMENT;
+	shadow_info->csa_size = MQD_FWWORKAREA_SIZE;
+	shadow_info->csa_alignment = MQD_FWWORKAREA_ALIGNMENT;
+}
+
+static int gfx_v12_0_get_gfx_shadow_info(struct amdgpu_device *adev,
+					 struct amdgpu_gfx_shadow_info *shadow_info,
+					 bool skip_check)
+{
+	if (adev->gfx.cp_gfx_shadow || skip_check) {
+		gfx_v12_0_get_gfx_shadow_info_nocheck(adev, shadow_info);
+		return 0;
+	}
+
+	memset(shadow_info, 0, sizeof(struct amdgpu_gfx_shadow_info));
+	return -EINVAL;
+}
+
 static const struct amdgpu_gfx_funcs gfx_v12_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v12_0_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v12_0_select_se_sh,
@@ -889,6 +931,7 @@ static const struct amdgpu_gfx_funcs gfx_v12_0_gfx_funcs = {
 	.read_wave_vgprs = &gfx_v12_0_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v12_0_select_me_pipe_q,
 	.update_perfmon_mgcg = &gfx_v12_0_update_perf_clk,
+	.get_gfx_shadow_info = &gfx_v12_0_get_gfx_shadow_info,
 };
 
 static int gfx_v12_0_gpu_early_init(struct amdgpu_device *adev)
@@ -1346,6 +1389,7 @@ static int gfx_v12_0_sw_init(struct amdgpu_ip_block *ip_block)
 	unsigned num_compute_rings;
 	int xcc_id = 0;
 	struct amdgpu_device *adev = ip_block->adev;
+	int num_queue_per_pipe = 1; /* we only enable 1 KGQ per pipe */
 
 	INIT_DELAYED_WORK(&adev->gfx.idle_work, amdgpu_gfx_profile_idle_work_handler);
 
@@ -1354,7 +1398,7 @@ static int gfx_v12_0_sw_init(struct amdgpu_ip_block *ip_block)
 	case IP_VERSION(12, 0, 1):
 		adev->gfx.me.num_me = 1;
 		adev->gfx.me.num_pipe_per_me = 1;
-		adev->gfx.me.num_queue_per_pipe = 1;
+		adev->gfx.me.num_queue_per_pipe = 8;
 		adev->gfx.mec.num_mec = 1;
 		adev->gfx.mec.num_pipe_per_mec = 2;
 		adev->gfx.mec.num_queue_per_pipe = 4;
@@ -1372,6 +1416,22 @@ static int gfx_v12_0_sw_init(struct amdgpu_ip_block *ip_block)
 	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
 	case IP_VERSION(12, 0, 0):
 	case IP_VERSION(12, 0, 1):
+		if (!adev->gfx.disable_uq &&
+		    adev->gfx.me_fw_version  >= 2780 &&
+		    adev->gfx.pfp_fw_version >= 2840 &&
+		    adev->gfx.mec_fw_version >= 3050 &&
+		    adev->mes.fw_version[0] >= 123) {
+			adev->userq_funcs[AMDGPU_HW_IP_GFX] = &userq_mes_funcs;
+			adev->userq_funcs[AMDGPU_HW_IP_COMPUTE] = &userq_mes_funcs;
+		}
+		break;
+	default:
+		break;
+	}
+
+	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
+	case IP_VERSION(12, 0, 0):
+	case IP_VERSION(12, 0, 1):
 		if (adev->gfx.me_fw_version  >= 2480 &&
 		    adev->gfx.pfp_fw_version >= 2530 &&
 		    adev->gfx.mec_fw_version >= 2680 &&
@@ -1383,11 +1443,13 @@ static int gfx_v12_0_sw_init(struct amdgpu_ip_block *ip_block)
 		break;
 	}
 
-	/* recalculate compute rings to use based on hardware configuration */
-	num_compute_rings = (adev->gfx.mec.num_pipe_per_mec *
-			     adev->gfx.mec.num_queue_per_pipe) / 2;
-	adev->gfx.num_compute_rings = min(adev->gfx.num_compute_rings,
-					  num_compute_rings);
+	if (adev->gfx.num_compute_rings) {
+		/* recalculate compute rings to use based on hardware configuration */
+		num_compute_rings = (adev->gfx.mec.num_pipe_per_mec *
+				     adev->gfx.mec.num_queue_per_pipe) / 2;
+		adev->gfx.num_compute_rings = min(adev->gfx.num_compute_rings,
+						  num_compute_rings);
+	}
 
 	/* EOP Event */
 	r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_GRBM_CP,
@@ -1433,37 +1495,41 @@ static int gfx_v12_0_sw_init(struct amdgpu_ip_block *ip_block)
 		return r;
 	}
 
-	/* set up the gfx ring */
-	for (i = 0; i < adev->gfx.me.num_me; i++) {
-		for (j = 0; j < adev->gfx.me.num_queue_per_pipe; j++) {
-			for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
-				if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
-					continue;
-
-				r = gfx_v12_0_gfx_ring_init(adev, ring_id,
-							    i, k, j);
-				if (r)
-					return r;
-				ring_id++;
+	if (adev->gfx.num_gfx_rings) {
+		/* set up the gfx ring */
+		for (i = 0; i < adev->gfx.me.num_me; i++) {
+			for (j = 0; j < num_queue_per_pipe; j++) {
+				for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
+					if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
+						continue;
+
+					r = gfx_v12_0_gfx_ring_init(adev, ring_id,
+								    i, k, j);
+					if (r)
+						return r;
+					ring_id++;
+				}
 			}
 		}
 	}
 
-	ring_id = 0;
-	/* set up the compute queues - allocate horizontally across pipes */
-	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
-		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
-			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev,
-								0, i, k, j))
-					continue;
+	if (adev->gfx.num_compute_rings) {
+		ring_id = 0;
+		/* set up the compute queues - allocate horizontally across pipes */
+		for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
+			for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
+				for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
+					if (!amdgpu_gfx_is_mec_queue_enabled(adev,
+									     0, i, k, j))
+						continue;
 
-				r = gfx_v12_0_compute_ring_init(adev, ring_id,
-								i, k, j);
-				if (r)
-					return r;
+					r = gfx_v12_0_compute_ring_init(adev, ring_id,
+									i, k, j);
+					if (r)
+						return r;
 
-				ring_id++;
+					ring_id++;
+				}
 			}
 		}
 	}
@@ -2324,7 +2390,7 @@ static int gfx_v12_0_cp_gfx_load_pfp_microcode_rs64(struct amdgpu_device *adev)
 	amdgpu_bo_unreserve(adev->gfx.pfp.pfp_fw_data_obj);
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	WREG32_SOC15(GC, 0, regCP_PFP_IC_BASE_LO,
 		lower_32_bits(adev->gfx.pfp.pfp_fw_gpu_addr));
@@ -2468,7 +2534,7 @@ static int gfx_v12_0_cp_gfx_load_me_microcode_rs64(struct amdgpu_device *adev)
 	amdgpu_bo_unreserve(adev->gfx.me.me_fw_data_obj);
 
 	if (amdgpu_emu_mode == 1)
-		adev->hdp.funcs->flush_hdp(adev, NULL);
+		amdgpu_device_flush_hdp(adev, NULL);
 
 	WREG32_SOC15(GC, 0, regCP_ME_IC_BASE_LO,
 		lower_32_bits(adev->gfx.me.me_fw_gpu_addr));
@@ -2948,6 +3014,8 @@ static int gfx_v12_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 #ifdef __BIG_ENDIAN
 	tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, BUF_SWAP, 1);
 #endif
+	if (prop->tmz_queue)
+		tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, TMZ_MATCH, 1);
 	mqd->cp_gfx_hqd_cntl = tmp;
 
 	/* set up cp_doorbell_control */
@@ -2968,6 +3036,14 @@ static int gfx_v12_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 	/* active the queue */
 	mqd->cp_gfx_hqd_active = 1;
 
+	/* set gfx UQ items */
+	mqd->shadow_base_lo = lower_32_bits(prop->shadow_addr);
+	mqd->shadow_base_hi = upper_32_bits(prop->shadow_addr);
+	mqd->fw_work_area_base_lo = lower_32_bits(prop->csa_addr);
+	mqd->fw_work_area_base_hi = upper_32_bits(prop->csa_addr);
+	mqd->fence_address_lo = lower_32_bits(prop->fence_address);
+	mqd->fence_address_hi = upper_32_bits(prop->fence_address);
+
 	return 0;
 }
 
@@ -3091,6 +3167,8 @@ static int gfx_v12_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, 0);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	if (prop->tmz_queue)
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TMZ, 1);
 	mqd->cp_hqd_pq_control = tmp;
 
 	/* set the wb address whether it's enabled or not */
@@ -3142,6 +3220,10 @@ static int gfx_v12_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 
 	mqd->cp_hqd_active = prop->hqd_active;
 
+	/* set UQ fenceaddress */
+	mqd->fence_address_lo = lower_32_bits(prop->fence_address);
+	mqd->fence_address_hi = upper_32_bits(prop->fence_address);
+
 	return 0;
 }
 
@@ -3426,7 +3508,7 @@ static int gfx_v12_0_gfxhub_enable(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
 		false : true;
@@ -3600,6 +3682,49 @@ static int gfx_v12_0_hw_init(struct amdgpu_ip_block *ip_block)
 	return r;
 }
 
+static int gfx_v12_0_set_userq_eop_interrupts(struct amdgpu_device *adev,
+					      bool enable)
+{
+	unsigned int irq_type;
+	int m, p, r;
+
+	if (adev->userq_funcs[AMDGPU_HW_IP_GFX]) {
+		for (m = 0; m < adev->gfx.me.num_me; m++) {
+			for (p = 0; p < adev->gfx.me.num_pipe_per_me; p++) {
+				irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + p;
+				if (enable)
+					r = amdgpu_irq_get(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				else
+					r = amdgpu_irq_put(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				if (r)
+					return r;
+			}
+		}
+	}
+
+	if (adev->userq_funcs[AMDGPU_HW_IP_COMPUTE]) {
+		for (m = 0; m < adev->gfx.mec.num_mec; ++m) {
+			for (p = 0; p < adev->gfx.mec.num_pipe_per_mec; p++) {
+				irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
+					+ (m * adev->gfx.mec.num_pipe_per_mec)
+					+ p;
+				if (enable)
+					r = amdgpu_irq_get(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				else
+					r = amdgpu_irq_put(adev, &adev->gfx.eop_irq,
+							   irq_type);
+				if (r)
+					return r;
+			}
+		}
+	}
+
+	return 0;
+}
+
 static int gfx_v12_0_hw_fini(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -3610,6 +3735,7 @@ static int gfx_v12_0_hw_fini(struct amdgpu_ip_block *ip_block)
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.bad_op_irq, 0);
+	gfx_v12_0_set_userq_eop_interrupts(adev, false);
 
 	if (!adev->no_hw_access) {
 		if (amdgpu_async_gfx_ring) {
@@ -3698,11 +3824,33 @@ static int gfx_v12_0_early_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 
+	switch (amdgpu_user_queue) {
+	case -1:
+	case 0:
+	default:
+		adev->gfx.disable_kq = false;
+		adev->gfx.disable_uq = true;
+		break;
+	case 1:
+		adev->gfx.disable_kq = false;
+		adev->gfx.disable_uq = false;
+		break;
+	case 2:
+		adev->gfx.disable_kq = true;
+		adev->gfx.disable_uq = false;
+		break;
+	}
+
 	adev->gfx.funcs = &gfx_v12_0_gfx_funcs;
 
-	adev->gfx.num_gfx_rings = GFX12_NUM_GFX_RINGS;
-	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
-					  AMDGPU_MAX_COMPUTE_RINGS);
+	if (adev->gfx.disable_kq) {
+		adev->gfx.num_gfx_rings = 0;
+		adev->gfx.num_compute_rings = 0;
+	} else {
+		adev->gfx.num_gfx_rings = GFX12_NUM_GFX_RINGS;
+		adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
+						  AMDGPU_MAX_COMPUTE_RINGS);
+	}
 
 	gfx_v12_0_set_kiq_pm4_funcs(adev);
 	gfx_v12_0_set_ring_funcs(adev);
@@ -3733,6 +3881,10 @@ static int gfx_v12_0_late_init(struct amdgpu_ip_block *ip_block)
 	if (r)
 		return r;
 
+	r = gfx_v12_0_set_userq_eop_interrupts(adev, true);
+	if (r)
+		return r;
+
 	return 0;
 }
 
@@ -4172,45 +4324,17 @@ static u64 gfx_v12_0_ring_get_wptr_gfx(struct amdgpu_ring *ring)
 static void gfx_v12_0_ring_set_wptr_gfx(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	uint32_t *wptr_saved;
-	uint32_t *is_queue_unmap;
-	uint64_t aggregated_db_index;
-	uint32_t mqd_size = adev->mqds[AMDGPU_HW_IP_GFX].mqd_size;
-	uint64_t wptr_tmp;
-
-	if (ring->is_mes_queue) {
-		wptr_saved = (uint32_t *)(ring->mqd_ptr + mqd_size);
-		is_queue_unmap = (uint32_t *)(ring->mqd_ptr + mqd_size +
-					      sizeof(uint32_t));
-		aggregated_db_index =
-			amdgpu_mes_get_aggregated_doorbell_index(adev,
-								 ring->hw_prio);
-
-		wptr_tmp = ring->wptr & ring->buf_mask;
-		atomic64_set((atomic64_t *)ring->wptr_cpu_addr, wptr_tmp);
-		*wptr_saved = wptr_tmp;
-		/* assume doorbell always being used by mes mapped queue */
-		if (*is_queue_unmap) {
-			WDOORBELL64(aggregated_db_index, wptr_tmp);
-			WDOORBELL64(ring->doorbell_index, wptr_tmp);
-		} else {
-			WDOORBELL64(ring->doorbell_index, wptr_tmp);
 
-			if (*is_queue_unmap)
-				WDOORBELL64(aggregated_db_index, wptr_tmp);
-		}
+	if (ring->use_doorbell) {
+		/* XXX check if swapping is necessary on BE */
+		atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
+			     ring->wptr);
+		WDOORBELL64(ring->doorbell_index, ring->wptr);
 	} else {
-		if (ring->use_doorbell) {
-			/* XXX check if swapping is necessary on BE */
-			atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
-				     ring->wptr);
-			WDOORBELL64(ring->doorbell_index, ring->wptr);
-		} else {
-			WREG32_SOC15(GC, 0, regCP_RB0_WPTR,
-				     lower_32_bits(ring->wptr));
-			WREG32_SOC15(GC, 0, regCP_RB0_WPTR_HI,
-				     upper_32_bits(ring->wptr));
-		}
+		WREG32_SOC15(GC, 0, regCP_RB0_WPTR,
+			     lower_32_bits(ring->wptr));
+		WREG32_SOC15(GC, 0, regCP_RB0_WPTR_HI,
+			     upper_32_bits(ring->wptr));
 	}
 }
 
@@ -4235,42 +4359,14 @@ static u64 gfx_v12_0_ring_get_wptr_compute(struct amdgpu_ring *ring)
 static void gfx_v12_0_ring_set_wptr_compute(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	uint32_t *wptr_saved;
-	uint32_t *is_queue_unmap;
-	uint64_t aggregated_db_index;
-	uint32_t mqd_size = adev->mqds[AMDGPU_HW_IP_COMPUTE].mqd_size;
-	uint64_t wptr_tmp;
-
-	if (ring->is_mes_queue) {
-		wptr_saved = (uint32_t *)(ring->mqd_ptr + mqd_size);
-		is_queue_unmap = (uint32_t *)(ring->mqd_ptr + mqd_size +
-					      sizeof(uint32_t));
-		aggregated_db_index =
-			amdgpu_mes_get_aggregated_doorbell_index(adev,
-								 ring->hw_prio);
-
-		wptr_tmp = ring->wptr & ring->buf_mask;
-		atomic64_set((atomic64_t *)ring->wptr_cpu_addr, wptr_tmp);
-		*wptr_saved = wptr_tmp;
-		/* assume doorbell always used by mes mapped queue */
-		if (*is_queue_unmap) {
-			WDOORBELL64(aggregated_db_index, wptr_tmp);
-			WDOORBELL64(ring->doorbell_index, wptr_tmp);
-		} else {
-			WDOORBELL64(ring->doorbell_index, wptr_tmp);
 
-			if (*is_queue_unmap)
-				WDOORBELL64(aggregated_db_index, wptr_tmp);
-		}
+	/* XXX check if swapping is necessary on BE */
+	if (ring->use_doorbell) {
+		atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
+			     ring->wptr);
+		WDOORBELL64(ring->doorbell_index, ring->wptr);
 	} else {
-		/* XXX check if swapping is necessary on BE */
-		if (ring->use_doorbell) {
-			atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
-				     ring->wptr);
-			WDOORBELL64(ring->doorbell_index, ring->wptr);
-		} else {
-			BUG(); /* only DOORBELL method supported on gfx12 now */
-		}
+		BUG(); /* only DOORBELL method supported on gfx12 now */
 	}
 }
 
@@ -4317,10 +4413,6 @@ static void gfx_v12_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 
 	control |= ib->length_dw | (vmid << 24);
 
-	if (ring->is_mes_queue)
-		/* inherit vmid from mqd */
-		control |= 0x400000;
-
 	amdgpu_ring_write(ring, header);
 	BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
 	amdgpu_ring_write(ring,
@@ -4340,10 +4432,6 @@ static void gfx_v12_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
 	u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
 
-	if (ring->is_mes_queue)
-		/* inherit vmid from mqd */
-		control |= 0x40000000;
-
 	amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
 	BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
 	amdgpu_ring_write(ring,
@@ -4383,8 +4471,7 @@ static void gfx_v12_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
 	amdgpu_ring_write(ring, upper_32_bits(addr));
 	amdgpu_ring_write(ring, lower_32_bits(seq));
 	amdgpu_ring_write(ring, upper_32_bits(seq));
-	amdgpu_ring_write(ring, ring->is_mes_queue ?
-			 (ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0);
+	amdgpu_ring_write(ring, 0);
 }
 
 static void gfx_v12_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
@@ -4412,10 +4499,7 @@ static void gfx_v12_0_ring_invalidate_tlbs(struct amdgpu_ring *ring,
 static void gfx_v12_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 					 unsigned vmid, uint64_t pd_addr)
 {
-	if (ring->is_mes_queue)
-		gfx_v12_0_ring_invalidate_tlbs(ring, 0, 0, false, 0);
-	else
-		amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
+	amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
 
 	/* compute doesn't have PFP */
 	if (ring->funcs->type == AMDGPU_RING_TYPE_GFX) {
@@ -4749,25 +4833,23 @@ static int gfx_v12_0_eop_irq(struct amdgpu_device *adev,
 			     struct amdgpu_irq_src *source,
 			     struct amdgpu_iv_entry *entry)
 {
-	int i;
+	u32 doorbell_offset = entry->src_data[0];
 	u8 me_id, pipe_id, queue_id;
 	struct amdgpu_ring *ring;
-	uint32_t mes_queue_id = entry->src_data[0];
+	int i;
 
 	DRM_DEBUG("IH: CP EOP\n");
 
-	if (adev->enable_mes && (mes_queue_id & AMDGPU_FENCE_MES_QUEUE_FLAG)) {
-		struct amdgpu_mes_queue *queue;
+	if (adev->enable_mes && doorbell_offset) {
+		struct amdgpu_userq_fence_driver *fence_drv = NULL;
+		struct xarray *xa = &adev->userq_xa;
+		unsigned long flags;
 
-		mes_queue_id &= AMDGPU_FENCE_MES_QUEUE_ID_MASK;
-
-		spin_lock(&adev->mes.queue_id_lock);
-		queue = idr_find(&adev->mes.queue_id_idr, mes_queue_id);
-		if (queue) {
-			DRM_DEBUG("process mes queue id = %d\n", mes_queue_id);
-			amdgpu_fence_process(queue->ring);
-		}
-		spin_unlock(&adev->mes.queue_id_lock);
+		xa_lock_irqsave(xa, flags);
+		fence_drv = xa_load(xa, doorbell_offset);
+		if (fence_drv)
+			amdgpu_userq_fence_driver_process(fence_drv);
+		xa_unlock_irqrestore(xa, flags);
 	} else {
 		me_id = (entry->ring_id & 0x0c) >> 2;
 		pipe_id = (entry->ring_id & 0x03) >> 0;
@@ -4934,27 +5016,29 @@ static void gfx_v12_0_handle_priv_fault(struct amdgpu_device *adev,
 	pipe_id = (entry->ring_id & 0x03) >> 0;
 	queue_id = (entry->ring_id & 0x70) >> 4;
 
-	switch (me_id) {
-	case 0:
-		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-			ring = &adev->gfx.gfx_ring[i];
-			if (ring->me == me_id && ring->pipe == pipe_id &&
-			    ring->queue == queue_id)
-				drm_sched_fault(&ring->sched);
-		}
-		break;
-	case 1:
-	case 2:
-		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			ring = &adev->gfx.compute_ring[i];
-			if (ring->me == me_id && ring->pipe == pipe_id &&
-			    ring->queue == queue_id)
-				drm_sched_fault(&ring->sched);
+	if (!adev->gfx.disable_kq) {
+		switch (me_id) {
+		case 0:
+			for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+				ring = &adev->gfx.gfx_ring[i];
+				if (ring->me == me_id && ring->pipe == pipe_id &&
+				    ring->queue == queue_id)
+					drm_sched_fault(&ring->sched);
+			}
+			break;
+		case 1:
+		case 2:
+			for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+				ring = &adev->gfx.compute_ring[i];
+				if (ring->me == me_id && ring->pipe == pipe_id &&
+				    ring->queue == queue_id)
+					drm_sched_fault(&ring->sched);
+			}
+			break;
+		default:
+			BUG();
+			break;
 		}
-		break;
-	default:
-		BUG();
-		break;
 	}
 }
 
@@ -5160,6 +5244,69 @@ static void gfx_v12_ip_dump(struct amdgpu_ip_block *ip_block)
 	amdgpu_gfx_off_ctrl(adev, true);
 }
 
+static bool gfx_v12_pipe_reset_support(struct amdgpu_device *adev)
+{
+	/* Disable the pipe reset until the CPFW fully support it.*/
+	dev_warn_once(adev->dev, "The CPFW hasn't support pipe reset yet.\n");
+	return false;
+}
+
+static int gfx_v12_reset_gfx_pipe(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t reset_pipe = 0, clean_pipe = 0;
+	int r;
+
+	if (!gfx_v12_pipe_reset_support(adev))
+		return -EOPNOTSUPP;
+
+	gfx_v12_0_set_safe_mode(adev, 0);
+	mutex_lock(&adev->srbm_mutex);
+	soc24_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+
+	switch (ring->pipe) {
+	case 0:
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   PFP_PIPE0_RESET, 1);
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   ME_PIPE0_RESET, 1);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   PFP_PIPE0_RESET, 0);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   ME_PIPE0_RESET, 0);
+		break;
+	case 1:
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   PFP_PIPE1_RESET, 1);
+		reset_pipe = REG_SET_FIELD(reset_pipe, CP_ME_CNTL,
+					   ME_PIPE1_RESET, 1);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   PFP_PIPE1_RESET, 0);
+		clean_pipe = REG_SET_FIELD(clean_pipe, CP_ME_CNTL,
+					   ME_PIPE1_RESET, 0);
+		break;
+	default:
+		break;
+	}
+
+	WREG32_SOC15(GC, 0, regCP_ME_CNTL, reset_pipe);
+	WREG32_SOC15(GC, 0, regCP_ME_CNTL, clean_pipe);
+
+	r = (RREG32(SOC15_REG_OFFSET(GC, 0, regCP_GFX_RS64_INSTR_PNTR1)) << 2) -
+					RS64_FW_UC_START_ADDR_LO;
+	soc24_grbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+	gfx_v12_0_unset_safe_mode(adev, 0);
+
+	dev_info(adev->dev, "The ring %s pipe reset: %s\n", ring->name,
+			r == 0 ? "successfully" : "failed");
+	/* Sometimes the ME start pc counter can't cache correctly, so the
+	 * PC check only as a reference and pipe reset result rely on the
+	 * later ring test.
+	 */
+	return 0;
+}
+
 static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -5170,8 +5317,10 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid)
 
 	r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false);
 	if (r) {
-		dev_err(adev->dev, "reset via MES failed %d\n", r);
-		return r;
+		dev_warn(adev->dev, "reset via MES failed and try pipe reset %d\n", r);
+		r = gfx_v12_reset_gfx_pipe(ring);
+		if (r)
+			return r;
 	}
 
 	r = gfx_v12_0_kgq_init_queue(ring, true);
@@ -5189,6 +5338,89 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid)
 	return amdgpu_ring_test_ring(ring);
 }
 
+static int gfx_v12_0_reset_compute_pipe(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t reset_pipe = 0, clean_pipe = 0;
+	int r = 0;
+
+	if (!gfx_v12_pipe_reset_support(adev))
+		return -EOPNOTSUPP;
+
+	gfx_v12_0_set_safe_mode(adev, 0);
+	mutex_lock(&adev->srbm_mutex);
+	soc24_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+
+	reset_pipe = RREG32_SOC15(GC, 0, regCP_MEC_RS64_CNTL);
+	clean_pipe = reset_pipe;
+
+	if (adev->gfx.rs64_enable) {
+		switch (ring->pipe) {
+		case 0:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE0_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE0_RESET, 0);
+			break;
+		case 1:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE1_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE1_RESET, 0);
+			break;
+		case 2:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE2_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE2_RESET, 0);
+			break;
+		case 3:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE3_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_RS64_CNTL,
+						   MEC_PIPE3_RESET, 0);
+			break;
+		default:
+			break;
+		}
+		WREG32_SOC15(GC, 0, regCP_MEC_RS64_CNTL, reset_pipe);
+		WREG32_SOC15(GC, 0, regCP_MEC_RS64_CNTL, clean_pipe);
+		r = (RREG32_SOC15(GC, 0, regCP_MEC_RS64_INSTR_PNTR) << 2) -
+				RS64_FW_UC_START_ADDR_LO;
+	} else {
+		switch (ring->pipe) {
+		case 0:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE0_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE0_RESET, 0);
+			break;
+		case 1:
+			reset_pipe = REG_SET_FIELD(reset_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE1_RESET, 1);
+			clean_pipe = REG_SET_FIELD(clean_pipe, CP_MEC_CNTL,
+							   MEC_ME1_PIPE1_RESET, 0);
+			break;
+		default:
+		break;
+		}
+		WREG32_SOC15(GC, 0, regCP_MEC_CNTL, reset_pipe);
+		WREG32_SOC15(GC, 0, regCP_MEC_CNTL, clean_pipe);
+		/* Doesn't find the F32 MEC instruction pointer register, and suppose
+		 * the driver won't run into the F32 mode.
+		 */
+	}
+
+	soc24_grbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+	gfx_v12_0_unset_safe_mode(adev, 0);
+
+	dev_info(adev->dev, "The ring %s pipe resets: %s\n", ring->name,
+			r == 0 ? "successfully" : "failed");
+	/* Need the ring test to verify the pipe reset result.*/
+	return 0;
+}
+
 static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -5199,8 +5431,10 @@ static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid)
 
 	r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, true);
 	if (r) {
-		dev_err(adev->dev, "reset via MMIO failed %d\n", r);
-		return r;
+		dev_warn(adev->dev, "fail(%d) to reset kcq  and try pipe reset\n", r);
+		r = gfx_v12_0_reset_compute_pipe(ring);
+		if (r)
+			return r;
 	}
 
 	r = gfx_v12_0_kcq_init_queue(ring, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 13fbee46417a..70d7a1f434c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -53,6 +53,9 @@
 #define VERDE_GB_ADDR_CONFIG_GOLDEN         0x12010002
 #define HAINAN_GB_ADDR_CONFIG_GOLDEN        0x02010001
 
+#define GFX6_NUM_GFX_RINGS     1
+#define GFX6_NUM_COMPUTE_RINGS 2
+
 static void gfx_v6_0_set_ring_funcs(struct amdgpu_device *adev);
 static void gfx_v6_0_set_irq_funcs(struct amdgpu_device *adev);
 static void gfx_v6_0_get_cu_info(struct amdgpu_device *adev);
@@ -1732,10 +1735,14 @@ static void gfx_v6_0_constants_init(struct amdgpu_device *adev)
 	gfx_v6_0_get_cu_info(adev);
 	gfx_v6_0_config_init(adev);
 
-	WREG32(mmCP_QUEUE_THRESHOLDS, ((0x16 << CP_QUEUE_THRESHOLDS__ROQ_IB1_START__SHIFT) |
-				       (0x2b << CP_QUEUE_THRESHOLDS__ROQ_IB2_START__SHIFT)));
-	WREG32(mmCP_MEQ_THRESHOLDS, (0x30 << CP_MEQ_THRESHOLDS__MEQ1_START__SHIFT) |
-				    (0x60 << CP_MEQ_THRESHOLDS__MEQ2_START__SHIFT));
+	WREG32(mmCP_QUEUE_THRESHOLDS,
+		((0x16 << CP_QUEUE_THRESHOLDS__ROQ_IB1_START__SHIFT) |
+		(0x2b << CP_QUEUE_THRESHOLDS__ROQ_IB2_START__SHIFT)));
+
+	/* set HW defaults for 3D engine */
+	WREG32(mmCP_MEQ_THRESHOLDS,
+		(0x30 << CP_MEQ_THRESHOLDS__MEQ1_START__SHIFT) |
+		(0x60 << CP_MEQ_THRESHOLDS__MEQ2_START__SHIFT));
 
 	sx_debug_1 = RREG32(mmSX_DEBUG_1);
 	WREG32(mmSX_DEBUG_1, sx_debug_1);
@@ -2851,44 +2858,21 @@ static u32 gfx_v6_0_get_csb_size(struct amdgpu_device *adev)
 static void gfx_v6_0_get_csb_buffer(struct amdgpu_device *adev,
 				    volatile u32 *buffer)
 {
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
+	u32 count = 0;
 
 	if (adev->gfx.rlc.cs_data == NULL)
 		return;
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index - 0xa000);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
+	count = amdgpu_gfx_csb_preamble_start(buffer);
+	count = amdgpu_gfx_csb_data_parser(adev, buffer, count);
 
 	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 1));
 	buffer[count++] = cpu_to_le32(mmPA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START);
 	buffer[count++] = cpu_to_le32(adev->gfx.config.rb_config[0][0].raster_config);
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
+	amdgpu_gfx_csb_preamble_end(buffer, count);
 }
 
 static void gfx_v6_0_init_pg(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 8181bd0e4f18..da0534ff1271 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -55,6 +55,9 @@
 #define GFX7_NUM_GFX_RINGS     1
 #define GFX7_MEC_HPD_SIZE      2048
 
+#define BONAIRE_GB_ADDR_CONFIG_GOLDEN        0x12010001
+#define HAWAII_GB_ADDR_CONFIG_GOLDEN         0x12011003
+
 static void gfx_v7_0_set_ring_funcs(struct amdgpu_device *adev);
 static void gfx_v7_0_set_irq_funcs(struct amdgpu_device *adev);
 static void gfx_v7_0_set_gds_init(struct amdgpu_device *adev);
@@ -3882,67 +3885,22 @@ static u32 gfx_v7_0_get_csb_size(struct amdgpu_device *adev)
 static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev,
 				    volatile u32 *buffer)
 {
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
+	u32 count = 0;
 
 	if (adev->gfx.rlc.cs_data == NULL)
 		return;
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index - PACKET3_SET_CONTEXT_REG_START);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
+	count = amdgpu_gfx_csb_preamble_start(buffer);
+	count = amdgpu_gfx_csb_data_parser(adev, buffer, count);
 
 	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 2));
 	buffer[count++] = cpu_to_le32(mmPA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START);
-	switch (adev->asic_type) {
-	case CHIP_BONAIRE:
-		buffer[count++] = cpu_to_le32(0x16000012);
-		buffer[count++] = cpu_to_le32(0x00000000);
-		break;
-	case CHIP_KAVERI:
-		buffer[count++] = cpu_to_le32(0x00000000); /* XXX */
-		buffer[count++] = cpu_to_le32(0x00000000);
-		break;
-	case CHIP_KABINI:
-	case CHIP_MULLINS:
-		buffer[count++] = cpu_to_le32(0x00000000); /* XXX */
-		buffer[count++] = cpu_to_le32(0x00000000);
-		break;
-	case CHIP_HAWAII:
-		buffer[count++] = cpu_to_le32(0x3a00161a);
-		buffer[count++] = cpu_to_le32(0x0000002e);
-		break;
-	default:
-		buffer[count++] = cpu_to_le32(0x00000000);
-		buffer[count++] = cpu_to_le32(0x00000000);
-		break;
-	}
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
+	buffer[count++] = cpu_to_le32(adev->gfx.config.rb_config[0][0].raster_config);
+	buffer[count++] = cpu_to_le32(adev->gfx.config.rb_config[0][0].raster_config_1);
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
+	amdgpu_gfx_csb_preamble_end(buffer, count);
 }
 
 static void gfx_v7_0_init_pg(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index bfedd487efc5..5ee2237d8ee8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1223,48 +1223,22 @@ out:
 static void gfx_v8_0_get_csb_buffer(struct amdgpu_device *adev,
 				    volatile u32 *buffer)
 {
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
+	u32 count = 0;
 
 	if (adev->gfx.rlc.cs_data == NULL)
 		return;
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index -
-						PACKET3_SET_CONTEXT_REG_START);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
+	count = amdgpu_gfx_csb_preamble_start(buffer);
+	count = amdgpu_gfx_csb_data_parser(adev, buffer, count);
 
 	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	buffer[count++] = cpu_to_le32(mmPA_SC_RASTER_CONFIG -
-			PACKET3_SET_CONTEXT_REG_START);
+	buffer[count++] = cpu_to_le32(mmPA_SC_RASTER_CONFIG - PACKET3_SET_CONTEXT_REG_START);
 	buffer[count++] = cpu_to_le32(adev->gfx.config.rb_config[0][0].raster_config);
 	buffer[count++] = cpu_to_le32(adev->gfx.config.rb_config[0][0].raster_config_1);
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
+	amdgpu_gfx_csb_preamble_end(buffer, count);
 }
 
 static int gfx_v8_0_cp_jump_table_num(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index d7db4cb907ae..d377a7c57d5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -225,17 +225,36 @@ static const struct amdgpu_hwip_reg_entry gc_reg_list_9[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, mmRLC_SMU_SAFE_MODE),
 	SOC15_REG_ENTRY_STR(GC, 0, mmRLC_INT_STAT),
 	SOC15_REG_ENTRY_STR(GC, 0, mmRLC_GPM_GENERAL_6),
-	/* cp header registers */
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME2_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
 	/* SE status registers */
 	SOC15_REG_ENTRY_STR(GC, 0, mmGRBM_STATUS_SE0),
 	SOC15_REG_ENTRY_STR(GC, 0, mmGRBM_STATUS_SE1),
 	SOC15_REG_ENTRY_STR(GC, 0, mmGRBM_STATUS_SE2),
-	SOC15_REG_ENTRY_STR(GC, 0, mmGRBM_STATUS_SE3)
+	SOC15_REG_ENTRY_STR(GC, 0, mmGRBM_STATUS_SE3),
+	/* packet headers */
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_CE_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_PFP_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_ME_HEADER_DUMP)
 };
 
 static const struct amdgpu_hwip_reg_entry gc_cp_reg_list_9[] = {
@@ -277,6 +296,14 @@ static const struct amdgpu_hwip_reg_entry gc_cp_reg_list_9[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_PQ_WPTR_LO),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_PQ_WPTR_HI),
 	SOC15_REG_ENTRY_STR(GC, 0, mmCP_HQD_GFX_STATUS),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, mmCP_MEC_ME1_HEADER_DUMP)
 };
 
 enum ta_ras_gfx_subblock {
@@ -1624,42 +1651,16 @@ static u32 gfx_v9_0_get_csb_size(struct amdgpu_device *adev)
 static void gfx_v9_0_get_csb_buffer(struct amdgpu_device *adev,
 				    volatile u32 *buffer)
 {
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
+	u32 count = 0;
 
 	if (adev->gfx.rlc.cs_data == NULL)
 		return;
 	if (buffer == NULL)
 		return;
 
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index -
-						PACKET3_SET_CONTEXT_REG_START);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
+	count = amdgpu_gfx_csb_preamble_start(buffer);
+	count = amdgpu_gfx_csb_data_parser(adev, buffer, count);
+	amdgpu_gfx_csb_preamble_end(buffer, count);
 }
 
 static void gfx_v9_0_init_always_on_cu_mask(struct amdgpu_device *adev)
@@ -5441,16 +5442,8 @@ static void gfx_v9_0_ring_patch_ce_meta(struct amdgpu_ring *ring,
 
 	payload_size = sizeof(struct v9_ce_ib_state);
 
-	if (ring->is_mes_queue) {
-		payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-					  gfx[0].gfx_meta_data) +
-			offsetof(struct v9_gfx_meta_data, ce_payload);
-		ce_payload_cpu_addr =
-			amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset);
-	} else {
-		payload_offset = offsetof(struct v9_gfx_meta_data, ce_payload);
-		ce_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
-	}
+	payload_offset = offsetof(struct v9_gfx_meta_data, ce_payload);
+	ce_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
 
 	if (offset + (payload_size >> 2) <= ring->buf_mask + 1) {
 		memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr, payload_size);
@@ -5473,16 +5466,8 @@ static void gfx_v9_0_ring_patch_de_meta(struct amdgpu_ring *ring,
 
 	payload_size = sizeof(struct v9_de_ib_state);
 
-	if (ring->is_mes_queue) {
-		payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-					  gfx[0].gfx_meta_data) +
-			offsetof(struct v9_gfx_meta_data, de_payload);
-		de_payload_cpu_addr =
-			amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset);
-	} else {
-		payload_offset = offsetof(struct v9_gfx_meta_data, de_payload);
-		de_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
-	}
+	payload_offset = offsetof(struct v9_gfx_meta_data, de_payload);
+	de_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
 
 	((struct v9_de_ib_state *)de_payload_cpu_addr)->ib_completion_status =
 		IB_COMPLETION_STATUS_PREEMPTED;
@@ -5672,19 +5657,9 @@ static void gfx_v9_0_ring_emit_ce_meta(struct amdgpu_ring *ring, bool resume)
 
 	cnt = (sizeof(ce_payload) >> 2) + 4 - 2;
 
-	if (ring->is_mes_queue) {
-		offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				  gfx[0].gfx_meta_data) +
-			offsetof(struct v9_gfx_meta_data, ce_payload);
-		ce_payload_gpu_addr =
-			amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ce_payload_cpu_addr =
-			amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-	} else {
-		offset = offsetof(struct v9_gfx_meta_data, ce_payload);
-		ce_payload_gpu_addr = amdgpu_csa_vaddr(ring->adev) + offset;
-		ce_payload_cpu_addr = adev->virt.csa_cpu_addr + offset;
-	}
+	offset = offsetof(struct v9_gfx_meta_data, ce_payload);
+	ce_payload_gpu_addr = amdgpu_csa_vaddr(ring->adev) + offset;
+	ce_payload_cpu_addr = adev->virt.csa_cpu_addr + offset;
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, cnt));
 	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(2) |
@@ -5770,28 +5745,13 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bo
 	void *de_payload_cpu_addr;
 	int cnt;
 
-	if (ring->is_mes_queue) {
-		offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				  gfx[0].gfx_meta_data) +
-			offsetof(struct v9_gfx_meta_data, de_payload);
-		de_payload_gpu_addr =
-			amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		de_payload_cpu_addr =
-			amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = offsetof(struct amdgpu_mes_ctx_meta_data,
-				  gfx[0].gds_backup) +
-			offsetof(struct v9_gfx_meta_data, de_payload);
-		gds_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-	} else {
-		offset = offsetof(struct v9_gfx_meta_data, de_payload);
-		de_payload_gpu_addr = amdgpu_csa_vaddr(ring->adev) + offset;
-		de_payload_cpu_addr = adev->virt.csa_cpu_addr + offset;
+	offset = offsetof(struct v9_gfx_meta_data, de_payload);
+	de_payload_gpu_addr = amdgpu_csa_vaddr(ring->adev) + offset;
+	de_payload_cpu_addr = adev->virt.csa_cpu_addr + offset;
 
-		gds_addr = ALIGN(amdgpu_csa_vaddr(ring->adev) +
-				 AMDGPU_CSA_SIZE - adev->gds.gds_size,
-				 PAGE_SIZE);
-	}
+	gds_addr = ALIGN(amdgpu_csa_vaddr(ring->adev) +
+			 AMDGPU_CSA_SIZE - adev->gds.gds_size,
+			 PAGE_SIZE);
 
 	if (usegds) {
 		de_payload.gds_backup_addrlo = lower_32_bits(gds_addr);
@@ -7339,9 +7299,14 @@ static void gfx_v9_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer
 			for (k = 0; k < adev->gfx.mec.num_queue_per_pipe; k++) {
 				drm_printf(p, "\nmec %d, pipe %d, queue %d\n", i, j, k);
 				for (reg = 0; reg < reg_count; reg++) {
-					drm_printf(p, "%-50s \t 0x%08x\n",
-						   gc_cp_reg_list_9[reg].reg_name,
-						   adev->gfx.ip_dump_compute_queues[index + reg]);
+					if (i && gc_cp_reg_list_9[reg].reg_offset == mmCP_MEC_ME1_HEADER_DUMP)
+						drm_printf(p, "%-50s \t 0x%08x\n",
+							   "mmCP_MEC_ME2_HEADER_DUMP",
+							   adev->gfx.ip_dump_compute_queues[index + reg]);
+					else
+						drm_printf(p, "%-50s \t 0x%08x\n",
+							   gc_cp_reg_list_9[reg].reg_name,
+							   adev->gfx.ip_dump_compute_queues[index + reg]);
 				}
 				index += reg_count;
 			}
@@ -7378,9 +7343,13 @@ static void gfx_v9_ip_dump(struct amdgpu_ip_block *ip_block)
 				soc15_grbm_select(adev, 1 + i, j, k, 0, 0);
 
 				for (reg = 0; reg < reg_count; reg++) {
-					adev->gfx.ip_dump_compute_queues[index + reg] =
-						RREG32(SOC15_REG_ENTRY_OFFSET(
-							gc_cp_reg_list_9[reg]));
+					if (i && gc_cp_reg_list_9[reg].reg_offset == mmCP_MEC_ME1_HEADER_DUMP)
+						adev->gfx.ip_dump_compute_queues[index + reg] =
+							RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME2_HEADER_DUMP));
+					else
+						adev->gfx.ip_dump_compute_queues[index + reg] =
+							RREG32(SOC15_REG_ENTRY_OFFSET(
+								       gc_cp_reg_list_9[reg]));
 				}
 				index += reg_count;
 			}
@@ -7394,8 +7363,14 @@ static void gfx_v9_ip_dump(struct amdgpu_ip_block *ip_block)
 
 static void gfx_v9_0_ring_emit_cleaner_shader(struct amdgpu_ring *ring)
 {
+	struct amdgpu_device *adev = ring->adev;
+
 	/* Emit the cleaner shader */
-	amdgpu_ring_write(ring, PACKET3(PACKET3_RUN_CLEANER_SHADER, 0));
+	if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2))
+		amdgpu_ring_write(ring, PACKET3(PACKET3_RUN_CLEANER_SHADER, 0));
+	else
+		amdgpu_ring_write(ring, PACKET3(PACKET3_RUN_CLEANER_SHADER_9_0, 0));
+
 	amdgpu_ring_write(ring, 0);  /* RESERVED field, programmed to zero */
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index d81449f9d822..c48cd47b531f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -1547,7 +1547,7 @@ static void gfx_v9_4_2_log_utc_edc_count(struct amdgpu_device *adev,
 {
 	uint32_t bank, way, mem;
 	static const char * const vml2_way_str[] = { "BIGK", "4K" };
-	static const char * const utcl2_rounter_str[] = { "VMC", "APT" };
+	static const char * const utcl2_router_str[] = { "VMC", "APT" };
 
 	mem = instance % blk->num_mem_blocks;
 	way = (instance / blk->num_mem_blocks) % blk->num_ways;
@@ -1568,7 +1568,7 @@ static void gfx_v9_4_2_log_utc_edc_count(struct amdgpu_device *adev,
 		dev_info(
 			adev->dev,
 			"GFX SubBlock UTCL2_ROUTER_IFIF%d_GROUP0_%s, SED %d, DED %d\n",
-			bank, utcl2_rounter_str[mem], sec_cnt, ded_cnt);
+			bank, utcl2_router_str[mem], sec_cnt, ded_cnt);
 		break;
 	case ATC_L2_CACHE_2M:
 		dev_info(
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 53fbf6ca7cdb..c233edf60569 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -105,9 +105,6 @@ static const struct amdgpu_hwip_reg_entry gc_reg_list_9_4_3[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regRLC_SMU_SAFE_MODE),
 	SOC15_REG_ENTRY_STR(GC, 0, regRLC_INT_STAT),
 	SOC15_REG_ENTRY_STR(GC, 0, regRLC_GPM_GENERAL_6),
-	/* cp header registers */
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
-	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME2_HEADER_DUMP),
 	/* SE status registers */
 	SOC15_REG_ENTRY_STR(GC, 0, regGRBM_STATUS_SE0),
 	SOC15_REG_ENTRY_STR(GC, 0, regGRBM_STATUS_SE1),
@@ -154,6 +151,14 @@ static const struct amdgpu_hwip_reg_entry gc_cp_reg_list_9_4_3[] = {
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_PQ_WPTR_LO),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_PQ_WPTR_HI),
 	SOC15_REG_ENTRY_STR(GC, 0, regCP_HQD_GFX_STATUS),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
+	SOC15_REG_ENTRY_STR(GC, 0, regCP_MEC_ME1_HEADER_DUMP),
 };
 
 struct amdgpu_gfx_ras gfx_v9_4_3_ras;
@@ -1148,6 +1153,12 @@ static int gfx_v9_4_3_sw_init(struct amdgpu_ip_block *ip_block)
 			adev->gfx.compute_supported_reset |= AMDGPU_RESET_TYPE_PER_PIPE;
 		}
 		break;
+	case IP_VERSION(9, 5, 0):
+		if (adev->gfx.mec_fw_version >= 21) {
+			adev->gfx.compute_supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
+			adev->gfx.compute_supported_reset |= AMDGPU_RESET_TYPE_PER_PIPE;
+		}
+		break;
 	default:
 		break;
 	}
@@ -1262,6 +1273,22 @@ static void gfx_v9_4_3_xcc_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
+/* For ASICs that needs xnack chain and MEC version supports, set SG_CONFIG1
+ * DISABLE_XNACK_CHECK_IN_RETRY_DISABLE bit and inform KFD to set xnack_chain
+ * bit in SET_RESOURCES
+ */
+static void gfx_v9_4_3_xcc_init_sq(struct amdgpu_device *adev, int xcc_id)
+{
+	uint32_t data;
+
+	if (!(adev->gmc.xnack_flags & AMDGPU_GMC_XNACK_FLAG_CHAIN))
+		return;
+
+	data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_CONFIG1);
+	data = REG_SET_FIELD(data, SQ_CONFIG1, DISABLE_XNACK_CHECK_IN_RETRY_DISABLE, 1);
+	WREG32_SOC15(GC, xcc_id, regSQ_CONFIG1, data);
+}
+
 static void gfx_v9_4_3_xcc_constants_init(struct amdgpu_device *adev,
 					  int xcc_id)
 {
@@ -1306,6 +1333,7 @@ static void gfx_v9_4_3_xcc_constants_init(struct amdgpu_device *adev,
 
 	gfx_v9_4_3_xcc_init_compute_vmid(adev, xcc_id);
 	gfx_v9_4_3_xcc_init_gds_vmid(adev, xcc_id);
+	gfx_v9_4_3_xcc_init_sq(adev, xcc_id);
 }
 
 static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
@@ -1318,6 +1346,20 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	adev->gfx.config.db_debug2 =
 		RREG32_SOC15(GC, GET_INST(GC, 0), regDB_DEBUG2);
 
+	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
+	/* ToDo: GC 9.4.4 */
+	case IP_VERSION(9, 4, 3):
+		if (adev->gfx.mec_fw_version >= 184)
+			adev->gmc.xnack_flags |= AMDGPU_GMC_XNACK_FLAG_CHAIN;
+		break;
+	case IP_VERSION(9, 5, 0):
+		if (adev->gfx.mec_fw_version >= 23)
+			adev->gmc.xnack_flags |= AMDGPU_GMC_XNACK_FLAG_CHAIN;
+		break;
+	default:
+		break;
+	}
+
 	for (i = 0; i < num_xcc; i++)
 		gfx_v9_4_3_xcc_constants_init(adev, i);
 }
@@ -3447,9 +3489,7 @@ static int gfx_v9_4_3_unmap_done(struct amdgpu_device *adev, uint32_t me,
 
 static bool gfx_v9_4_3_pipe_reset_support(struct amdgpu_device *adev)
 {
-	/*TODO: Need check gfx9.4.4 mec fw whether supports pipe reset as well.*/
-	if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) &&
-			adev->gfx.mec_fw_version >= 0x0000009b)
+	if (!!(adev->gfx.compute_supported_reset & AMDGPU_RESET_TYPE_PER_PIPE))
 		return true;
 	else
 		dev_warn_once(adev->dev, "Please use the latest MEC version to see whether support pipe reset\n");
@@ -4558,12 +4598,21 @@ static void gfx_v9_4_3_ip_print(struct amdgpu_ip_block *ip_block, struct drm_pri
 						   "\nxcc:%d mec:%d, pipe:%d, queue:%d\n",
 						    xcc_id, i, j, k);
 					for (reg = 0; reg < reg_count; reg++) {
-						drm_printf(p,
-							   "%-50s \t 0x%08x\n",
-							   gc_cp_reg_list_9_4_3[reg].reg_name,
-							   adev->gfx.ip_dump_compute_queues
-								[xcc_offset + inst_offset +
-								reg]);
+						if (i && gc_cp_reg_list_9_4_3[reg].reg_offset ==
+						    regCP_MEC_ME1_HEADER_DUMP)
+							drm_printf(p,
+								   "%-50s \t 0x%08x\n",
+								   "regCP_MEC_ME2_HEADER_DUMP",
+								   adev->gfx.ip_dump_compute_queues
+								   [xcc_offset + inst_offset +
+								    reg]);
+						else
+							drm_printf(p,
+								   "%-50s \t 0x%08x\n",
+								   gc_cp_reg_list_9_4_3[reg].reg_name,
+								   adev->gfx.ip_dump_compute_queues
+								   [xcc_offset + inst_offset +
+								    reg]);
 					}
 					inst_offset += reg_count;
 				}
@@ -4612,12 +4661,20 @@ static void gfx_v9_4_3_ip_dump(struct amdgpu_ip_block *ip_block)
 							  GET_INST(GC, xcc_id));
 
 					for (reg = 0; reg < reg_count; reg++) {
-						adev->gfx.ip_dump_compute_queues
-							[xcc_offset +
-							 inst_offset + reg] =
-							RREG32(SOC15_REG_ENTRY_OFFSET_INST(
-								gc_cp_reg_list_9_4_3[reg],
-								GET_INST(GC, xcc_id)));
+						if (i && gc_cp_reg_list_9_4_3[reg].reg_offset ==
+						    regCP_MEC_ME1_HEADER_DUMP)
+							adev->gfx.ip_dump_compute_queues
+								[xcc_offset +
+								 inst_offset + reg] =
+								RREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id),
+											regCP_MEC_ME2_HEADER_DUMP));
+						else
+							adev->gfx.ip_dump_compute_queues
+								[xcc_offset +
+								 inst_offset + reg] =
+								RREG32(SOC15_REG_ENTRY_OFFSET_INST(
+									       gc_cp_reg_list_9_4_3[reg],
+									       GET_INST(GC, xcc_id)));
 					}
 					inst_offset += reg_count;
 				}
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 95d894a231fc..a3e2787501f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -268,7 +268,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng;
 
 	/* flush hdp cache */
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	/* This is necessary for SRIOV as well as for GFXOFF to function
 	 * properly under bare metal
@@ -428,10 +428,6 @@ static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t reg;
 
-	/* MES fw manages IH_VMID_x_LUT updating */
-	if (ring->is_mes_queue)
-		return;
-
 	if (ring->vm_hub == AMDGPU_GFXHUB(0))
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid;
 	else
@@ -969,7 +965,7 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev)
 	adev->hdp.funcs->init_registers(adev);
 
 	/* Flush HDP after it is initialized */
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
 		false : true;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index ad099f136f84..72211409227b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -229,7 +229,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng;
 
 	/* flush hdp cache */
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	/* This is necessary for SRIOV as well as for GFXOFF to function
 	 * properly under bare metal
@@ -393,10 +393,6 @@ static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t reg;
 
-	/* MES fw manages IH_VMID_x_LUT updating */
-	if (ring->is_mes_queue)
-		return;
-
 	if (ring->vm_hub == AMDGPU_GFXHUB(0))
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid;
 	else
@@ -752,6 +748,18 @@ static int gmc_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 	adev->gmc.vram_type = vram_type;
 	adev->gmc.vram_vendor = vram_vendor;
 
+	/* The mall_size is already calculated as mall_size_per_umc * num_umc.
+	 * However, for gfx1151, which features a 2-to-1 UMC mapping,
+	 * the result must be multiplied by 2 to determine the actual mall size.
+	 */
+	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
+	case IP_VERSION(11, 5, 1):
+		adev->gmc.mall_size *= 2;
+		break;
+	default:
+		break;
+	}
+
 	switch (amdgpu_ip_version(adev, GC_HWIP, 0)) {
 	case IP_VERSION(11, 0, 0):
 	case IP_VERSION(11, 0, 1):
@@ -832,7 +840,7 @@ static int gmc_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 	 * amdgpu graphics/compute will use VMIDs 1-7
 	 * amdkfd will use VMIDs 8-15
 	 */
-	adev->vm_manager.first_kfd_vmid = 8;
+	adev->vm_manager.first_kfd_vmid = adev->gfx.disable_kq ? 1 : 8;
 
 	amdgpu_vm_manager_init(adev);
 
@@ -899,7 +907,7 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev)
 		return r;
 
 	/* Flush HDP after it is initialized */
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
 		false : true;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
index 05c026d0b0d9..b645d3e6a6c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
@@ -297,7 +297,7 @@ static void gmc_v12_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 		return;
 
 	/* flush hdp cache */
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	/* This is necessary for SRIOV as well as for GFXOFF to function
 	 * properly under bare metal
@@ -413,10 +413,6 @@ static void gmc_v12_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t reg;
 
-	/* MES fw manages IH_VMID_x_LUT updating */
-	if (ring->is_mes_queue)
-		return;
-
 	if (ring->vm_hub == AMDGPU_GFXHUB(0))
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid;
 	else
@@ -820,7 +816,7 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block)
 	 * amdgpu graphics/compute will use VMIDs 1-7
 	 * amdkfd will use VMIDs 8-15
 	 */
-	adev->vm_manager.first_kfd_vmid = 8;
+	adev->vm_manager.first_kfd_vmid = adev->gfx.disable_kq ? 1 : 8;
 
 	amdgpu_vm_manager_init(adev);
 
@@ -881,7 +877,7 @@ static int gmc_v12_0_gart_enable(struct amdgpu_device *adev)
 		return r;
 
 	/* Flush HDP after it is initialized */
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
 		false : true;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index a992e79d9581..8030fcd64210 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -249,7 +249,7 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev)
 
 		/* disable VGA render */
 		tmp = RREG32(mmVGA_RENDER_CONTROL);
-		tmp &= ~VGA_VSTATUS_CNTL;
+		tmp &= VGA_RENDER_CONTROL__VGA_VSTATUS_CNTL_MASK;
 		WREG32(mmVGA_RENDER_CONTROL, tmp);
 	}
 	/* Update configuration */
@@ -627,17 +627,16 @@ static void gmc_v6_0_vm_decode_fault(struct amdgpu_device *adev,
 	       "write" : "read", block, mc_client, mc_id);
 }
 
-/*
 static const u32 mc_cg_registers[] = {
-	MC_HUB_MISC_HUB_CG,
-	MC_HUB_MISC_SIP_CG,
-	MC_HUB_MISC_VM_CG,
-	MC_XPB_CLK_GAT,
-	ATC_MISC_CG,
-	MC_CITF_MISC_WR_CG,
-	MC_CITF_MISC_RD_CG,
-	MC_CITF_MISC_VM_CG,
-	VM_L2_CG,
+	mmMC_HUB_MISC_HUB_CG,
+	mmMC_HUB_MISC_SIP_CG,
+	mmMC_HUB_MISC_VM_CG,
+	mmMC_XPB_CLK_GAT,
+	mmATC_MISC_CG,
+	mmMC_CITF_MISC_WR_CG,
+	mmMC_CITF_MISC_RD_CG,
+	mmMC_CITF_MISC_VM_CG,
+	mmVM_L2_CG,
 };
 
 static const u32 mc_cg_ls_en[] = {
@@ -672,7 +671,7 @@ static void gmc_v6_0_enable_mc_ls(struct amdgpu_device *adev,
 
 	for (i = 0; i < ARRAY_SIZE(mc_cg_registers); i++) {
 		orig = data = RREG32(mc_cg_registers[i]);
-		if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_MC_LS))
+		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_LS))
 			data |= mc_cg_ls_en[i];
 		else
 			data &= ~mc_cg_ls_en[i];
@@ -689,7 +688,7 @@ static void gmc_v6_0_enable_mc_mgcg(struct amdgpu_device *adev,
 
 	for (i = 0; i < ARRAY_SIZE(mc_cg_registers); i++) {
 		orig = data = RREG32(mc_cg_registers[i]);
-		if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_MC_MGCG))
+		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG))
 			data |= mc_cg_en[i];
 		else
 			data &= ~mc_cg_en[i];
@@ -705,7 +704,7 @@ static void gmc_v6_0_enable_bif_mgls(struct amdgpu_device *adev,
 
 	orig = data = RREG32_PCIE(ixPCIE_CNTL2);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_BIF_LS)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) {
 		data = REG_SET_FIELD(data, PCIE_CNTL2, SLV_MEM_LS_EN, 1);
 		data = REG_SET_FIELD(data, PCIE_CNTL2, MST_MEM_LS_EN, 1);
 		data = REG_SET_FIELD(data, PCIE_CNTL2, REPLAY_MEM_LS_EN, 1);
@@ -728,7 +727,7 @@ static void gmc_v6_0_enable_hdp_mgcg(struct amdgpu_device *adev,
 
 	orig = data = RREG32(mmHDP_HOST_PATH_CNTL);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_HDP_MGCG))
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_MGCG))
 		data = REG_SET_FIELD(data, HDP_HOST_PATH_CNTL, CLOCK_GATING_DIS, 0);
 	else
 		data = REG_SET_FIELD(data, HDP_HOST_PATH_CNTL, CLOCK_GATING_DIS, 1);
@@ -744,7 +743,7 @@ static void gmc_v6_0_enable_hdp_ls(struct amdgpu_device *adev,
 
 	orig = data = RREG32(mmHDP_MEM_POWER_LS);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_HDP_LS))
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS))
 		data = REG_SET_FIELD(data, HDP_MEM_POWER_LS, LS_ENABLE, 1);
 	else
 		data = REG_SET_FIELD(data, HDP_MEM_POWER_LS, LS_ENABLE, 0);
@@ -752,7 +751,6 @@ static void gmc_v6_0_enable_hdp_ls(struct amdgpu_device *adev,
 	if (orig != data)
 		WREG32(mmHDP_MEM_POWER_LS, data);
 }
-*/
 
 static int gmc_v6_0_convert_vram_type(int mc_seq_vram_type)
 {
@@ -1098,6 +1096,20 @@ static int gmc_v6_0_process_interrupt(struct amdgpu_device *adev,
 static int gmc_v6_0_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 					  enum amd_clockgating_state state)
 {
+	struct amdgpu_device *adev = ip_block->adev;
+	bool gate = false;
+
+	if (state == AMD_CG_STATE_GATE)
+		gate = true;
+
+	if (!(adev->flags & AMD_IS_APU)) {
+		gmc_v6_0_enable_mc_mgcg(adev, gate);
+		gmc_v6_0_enable_mc_ls(adev, gate);
+	}
+	gmc_v6_0_enable_bif_mgls(adev, gate);
+	gmc_v6_0_enable_hdp_mgcg(adev, gate);
+	gmc_v6_0_enable_hdp_ls(adev, gate);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 83e39f16044a..a8d5795084fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -1157,17 +1157,10 @@ static bool gmc_v7_0_is_idle(struct amdgpu_ip_block *ip_block)
 static int gmc_v7_0_wait_for_idle(struct amdgpu_ip_block *ip_block)
 {
 	unsigned int i;
-	u32 tmp;
 	struct amdgpu_device *adev = ip_block->adev;
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		/* read MC_STATUS */
-		tmp = RREG32(mmSRBM_STATUS) & (SRBM_STATUS__MCB_BUSY_MASK |
-					       SRBM_STATUS__MCB_NON_DISPLAY_BUSY_MASK |
-					       SRBM_STATUS__MCC_BUSY_MASK |
-					       SRBM_STATUS__MCD_BUSY_MASK |
-					       SRBM_STATUS__VMC_BUSY_MASK);
-		if (!tmp)
+		if (gmc_v7_0_is_idle(ip_block))
 			return 0;
 		udelay(1);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 783e0c3b86b4..282197f4ffb1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1213,10 +1213,7 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		if (uncached) {
 			mtype = MTYPE_UC;
 		} else if (ext_coherent) {
-			if (gc_ip_version == IP_VERSION(9, 5, 0) || adev->rev_id)
-				mtype = is_local ? MTYPE_CC : MTYPE_UC;
-			else
-				mtype = MTYPE_UC;
+			mtype = is_local ? MTYPE_CC : MTYPE_UC;
 		} else if (adev->flags & AMD_IS_APU) {
 			mtype = is_local ? mtype_local : MTYPE_NC;
 		} else {
@@ -1336,7 +1333,7 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
 				mtype_local = MTYPE_CC;
 
 			*flags = AMDGPU_PTE_MTYPE_VG10(*flags, mtype_local);
-		} else if (adev->rev_id) {
+		} else {
 			/* MTYPE_UC case */
 			*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC);
 		}
@@ -1505,7 +1502,6 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
 		adev->umc.umc_inst_num = UMC_V12_0_UMC_INSTANCE_NUM;
 		adev->umc.node_inst_num /= UMC_V12_0_UMC_INSTANCE_NUM;
 		adev->umc.channel_offs = UMC_V12_0_PER_CHANNEL_OFFSET;
-		adev->umc.retire_unit = UMC_V12_0_BAD_PAGE_NUM_PER_CHANNEL;
 		if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu)
 			adev->umc.ras = &umc_v12_0_ras;
 		break;
@@ -2075,6 +2071,9 @@ static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev)
 {
 	adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM;
 	adev->gmc.vram_width = 128 * 64;
+
+	if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 5, 0))
+		adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM3E;
 }
 
 static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block)
@@ -2411,13 +2410,6 @@ static int gmc_v9_0_hw_init(struct amdgpu_ip_block *ip_block)
 	adev->gmc.flush_tlb_needs_extra_type_2 =
 		amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 0) &&
 		adev->gmc.xgmi.num_physical_nodes;
-	/*
-	 * TODO: This workaround is badly documented and had a buggy
-	 * implementation. We should probably verify what we do here.
-	 */
-	adev->gmc.flush_tlb_needs_extra_type_0 =
-		amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) &&
-		adev->rev_id == 0;
 
 	/* The sequence of these two function calls matters.*/
 	gmc_v9_0_init_golden_registers(adev);
@@ -2435,7 +2427,7 @@ static int gmc_v9_0_hw_init(struct amdgpu_ip_block *ip_block)
 	adev->hdp.funcs->init_registers(adev);
 
 	/* After HDP is initialized, flush HDP.*/
-	adev->hdp.funcs->flush_hdp(adev, NULL);
+	amdgpu_device_flush_hdp(adev, NULL);
 
 	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
 		value = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
index f1dc13b3ab38..e6c0d86d3486 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
@@ -36,17 +36,6 @@
 #define HDP_MEM_POWER_CTRL__RC_MEM_POWER_LS_EN_MASK     0x00020000L
 #define mmHDP_MEM_POWER_CTRL_BASE_IDX   0
 
-static void hdp_v4_0_flush_hdp(struct amdgpu_device *adev,
-				struct amdgpu_ring *ring)
-{
-	if (!ring || !ring->funcs->emit_wreg) {
-		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
-	} else {
-		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	}
-}
-
 static void hdp_v4_0_invalidate_hdp(struct amdgpu_device *adev,
 				    struct amdgpu_ring *ring)
 {
@@ -180,7 +169,7 @@ struct amdgpu_hdp_ras hdp_v4_0_ras = {
 };
 
 const struct amdgpu_hdp_funcs hdp_v4_0_funcs = {
-	.flush_hdp = hdp_v4_0_flush_hdp,
+	.flush_hdp = amdgpu_hdp_generic_flush,
 	.invalidate_hdp = hdp_v4_0_invalidate_hdp,
 	.update_clock_gating = hdp_v4_0_update_clock_gating,
 	.get_clock_gating_state = hdp_v4_0_get_clockgating_state,
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c
index 43195c079748..8bc001dc9f63 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_0.c
@@ -27,17 +27,6 @@
 #include "hdp/hdp_5_0_0_sh_mask.h"
 #include <uapi/linux/kfd_ioctl.h>
 
-static void hdp_v5_0_flush_hdp(struct amdgpu_device *adev,
-				struct amdgpu_ring *ring)
-{
-	if (!ring || !ring->funcs->emit_wreg) {
-		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
-	} else {
-		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	}
-}
-
 static void hdp_v5_0_invalidate_hdp(struct amdgpu_device *adev,
 				    struct amdgpu_ring *ring)
 {
@@ -217,7 +206,7 @@ static void hdp_v5_0_init_registers(struct amdgpu_device *adev)
 }
 
 const struct amdgpu_hdp_funcs hdp_v5_0_funcs = {
-	.flush_hdp = hdp_v5_0_flush_hdp,
+	.flush_hdp = amdgpu_hdp_generic_flush,
 	.invalidate_hdp = hdp_v5_0_invalidate_hdp,
 	.update_clock_gating = hdp_v5_0_update_clock_gating,
 	.get_clock_gating_state = hdp_v5_0_get_clockgating_state,
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
index fcb8dd2876bc..40940b4ab400 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v5_2.c
@@ -33,7 +33,17 @@ static void hdp_v5_2_flush_hdp(struct amdgpu_device *adev,
 	if (!ring || !ring->funcs->emit_wreg) {
 		WREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2,
 			0);
-		RREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+		if (amdgpu_sriov_vf(adev)) {
+			/* this is fine because SR_IOV doesn't remap the register */
+			RREG32_NO_KIQ((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
+		} else {
+			/* We just need to read back a register to post the write.
+			 * Reading back the remapped register causes problems on
+			 * some platforms so just read back the memory size register.
+			 */
+			if (adev->nbio.funcs->get_memsize)
+				adev->nbio.funcs->get_memsize(adev);
+		}
 	} else {
 		amdgpu_ring_emit_wreg(ring,
 			(adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2,
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c
index a88d25a06c29..ec20daf4272c 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v6_0.c
@@ -30,17 +30,6 @@
 #define regHDP_CLK_CNTL_V6_1	0xd5
 #define regHDP_CLK_CNTL_V6_1_BASE_IDX 0
 
-static void hdp_v6_0_flush_hdp(struct amdgpu_device *adev,
-				struct amdgpu_ring *ring)
-{
-	if (!ring || !ring->funcs->emit_wreg) {
-		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
-	} else {
-		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	}
-}
-
 static void hdp_v6_0_update_clock_gating(struct amdgpu_device *adev,
 					 bool enable)
 {
@@ -149,7 +138,7 @@ static void hdp_v6_0_get_clockgating_state(struct amdgpu_device *adev,
 }
 
 const struct amdgpu_hdp_funcs hdp_v6_0_funcs = {
-	.flush_hdp = hdp_v6_0_flush_hdp,
+	.flush_hdp = amdgpu_hdp_generic_flush,
 	.update_clock_gating = hdp_v6_0_update_clock_gating,
 	.get_clock_gating_state = hdp_v6_0_get_clockgating_state,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c
index 49f7eb4fbd11..ed1debc03507 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v7_0.c
@@ -27,17 +27,6 @@
 #include "hdp/hdp_7_0_0_sh_mask.h"
 #include <uapi/linux/kfd_ioctl.h>
 
-static void hdp_v7_0_flush_hdp(struct amdgpu_device *adev,
-				struct amdgpu_ring *ring)
-{
-	if (!ring || !ring->funcs->emit_wreg) {
-		WREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-		RREG32((adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2);
-	} else {
-		amdgpu_ring_emit_wreg(ring, (adev->rmmio_remap.reg_offset + KFD_MMIO_REMAP_HDP_MEM_FLUSH_CNTL) >> 2, 0);
-	}
-}
-
 static void hdp_v7_0_update_clock_gating(struct amdgpu_device *adev,
 					 bool enable)
 {
@@ -137,7 +126,7 @@ static void hdp_v7_0_get_clockgating_state(struct amdgpu_device *adev,
 }
 
 const struct amdgpu_hdp_funcs hdp_v7_0_funcs = {
-	.flush_hdp = hdp_v7_0_flush_hdp,
+	.flush_hdp = amdgpu_hdp_generic_flush,
 	.update_clock_gating = hdp_v7_0_update_clock_gating,
 	.get_clock_gating_state = hdp_v7_0_get_clockgating_state,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
index eb4185dcbd1d..5900b560b7de 100644
--- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
@@ -349,6 +349,7 @@ static int ih_v6_0_irq_init(struct amdgpu_device *adev)
 			if (ret)
 				return ret;
 		}
+		ih[i]->overflow = false;
 	}
 
 	/* update doorbell range for ih ring 0 */
@@ -446,7 +447,10 @@ static u32 ih_v6_0_get_wptr(struct amdgpu_device *adev,
 	wptr = RREG32_NO_KIQ(ih_regs->ih_rb_wptr);
 	if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW))
 		goto out;
-	wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
+	if (!amdgpu_sriov_vf(adev))
+		wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
+	else
+		ih->overflow = true;
 
 	/* When a ring buffer overflow happen start parsing interrupt
 	 * from the last not overwritten vector (wptr + 32). Hopefully
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
index 218e16b68f1d..cb94bd71300f 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -28,11 +28,13 @@
 #include "soc15d.h"
 #include "jpeg_v4_0_3.h"
 #include "jpeg_v5_0_1.h"
+#include "mmsch_v5_0.h"
 
 #include "vcn/vcn_5_0_0_offset.h"
 #include "vcn/vcn_5_0_0_sh_mask.h"
 #include "ivsrcid/vcn/irqsrcs_vcn_5_0.h"
 
+static int jpeg_v5_0_1_start_sriov(struct amdgpu_device *adev);
 static void jpeg_v5_0_1_set_dec_ring_funcs(struct amdgpu_device *adev);
 static void jpeg_v5_0_1_set_irq_funcs(struct amdgpu_device *adev);
 static int jpeg_v5_0_1_set_powergating_state(struct amdgpu_ip_block *ip_block,
@@ -156,21 +158,16 @@ static int jpeg_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			ring = &adev->jpeg.inst[i].ring_dec[j];
-			ring->use_doorbell = false;
+			ring->use_doorbell = true;
 			ring->vm_hub = AMDGPU_MMHUB0(adev->jpeg.inst[i].aid_id);
 			if (!amdgpu_sriov_vf(adev)) {
 				ring->doorbell_index =
 					(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
 					1 + j + 11 * jpeg_inst;
 			} else {
-				if (j < 4)
-					ring->doorbell_index =
-						(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
-						4 + j + 32 * jpeg_inst;
-				else
-					ring->doorbell_index =
-						(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
-						8 + j + 32 * jpeg_inst;
+				ring->doorbell_index =
+					(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+					2 + j + 32 * jpeg_inst;
 			}
 			sprintf(ring->name, "jpeg_dec_%d.%d", adev->jpeg.inst[i].aid_id, j);
 			r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
@@ -237,7 +234,10 @@ static int jpeg_v5_0_1_hw_init(struct amdgpu_ip_block *ip_block)
 	int i, j, r, jpeg_inst;
 
 	if (amdgpu_sriov_vf(adev)) {
-		/* jpeg_v5_0_1_start_sriov(adev); */
+		r = jpeg_v5_0_1_start_sriov(adev);
+		if (r)
+			return r;
+
 		for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
 			for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 				ring = &adev->jpeg.inst[i].ring_dec[j];
@@ -264,7 +264,7 @@ static int jpeg_v5_0_1_hw_init(struct amdgpu_ip_block *ip_block)
 			ring = &adev->jpeg.inst[i].ring_dec[j];
 			if (ring->use_doorbell)
 				WREG32_SOC15_OFFSET(VCN, GET_INST(VCN, i), regVCN_JPEG_DB_CTRL,
-						    (ring->pipe ? (ring->pipe - 0x15) : 0),
+						    ring->pipe,
 						    ring->doorbell_index <<
 						    VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
 						    VCN_JPEG_DB_CTRL__EN_MASK);
@@ -291,8 +291,10 @@ static int jpeg_v5_0_1_hw_fini(struct amdgpu_ip_block *ip_block)
 
 	cancel_delayed_work_sync(&adev->jpeg.idle_work);
 
-	if (adev->jpeg.cur_state != AMD_PG_STATE_GATE)
-		ret = jpeg_v5_0_1_set_powergating_state(ip_block, AMD_PG_STATE_GATE);
+	if (!amdgpu_sriov_vf(adev)) {
+		if (adev->jpeg.cur_state != AMD_PG_STATE_GATE)
+			ret = jpeg_v5_0_1_set_powergating_state(ip_block, AMD_PG_STATE_GATE);
+	}
 
 	return ret;
 }
@@ -422,6 +424,119 @@ static void jpeg_v5_0_1_init_jrbc(struct amdgpu_ring *ring)
 					 reg_offset);
 }
 
+static int jpeg_v5_0_1_start_sriov(struct amdgpu_device *adev)
+{
+	struct amdgpu_ring *ring;
+	uint64_t ctx_addr;
+	uint32_t param, resp, expected;
+	uint32_t tmp, timeout;
+
+	struct amdgpu_mm_table *table = &adev->virt.mm_table;
+	uint32_t *table_loc;
+	uint32_t table_size;
+	uint32_t size, size_dw, item_offset;
+	uint32_t init_status;
+	int i, j, jpeg_inst;
+
+	struct mmsch_v5_0_cmd_direct_write
+		direct_wt = { {0} };
+	struct mmsch_v5_0_cmd_end end = { {0} };
+	struct mmsch_v5_0_init_header header;
+
+	direct_wt.cmd_header.command_type =
+		MMSCH_COMMAND__DIRECT_REG_WRITE;
+	end.cmd_header.command_type =
+		MMSCH_COMMAND__END;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) {
+		jpeg_inst = GET_INST(JPEG, i);
+
+		memset(&header, 0, sizeof(struct mmsch_v5_0_init_header));
+		header.version = MMSCH_VERSION;
+		header.total_size = sizeof(struct mmsch_v5_0_init_header) >> 2;
+
+		table_loc = (uint32_t *)table->cpu_addr;
+		table_loc += header.total_size;
+
+		item_offset = header.total_size;
+
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; j++) {
+			ring = &adev->jpeg.inst[i].ring_dec[j];
+			table_size = 0;
+
+			tmp = SOC15_REG_OFFSET(JPEG, 0, regUVD_LMI_JRBC_RB_64BIT_BAR_LOW);
+			MMSCH_V5_0_INSERT_DIRECT_WT(tmp, lower_32_bits(ring->gpu_addr));
+			tmp = SOC15_REG_OFFSET(JPEG, 0, regUVD_LMI_JRBC_RB_64BIT_BAR_HIGH);
+			MMSCH_V5_0_INSERT_DIRECT_WT(tmp, upper_32_bits(ring->gpu_addr));
+			tmp = SOC15_REG_OFFSET(JPEG, 0, regUVD_JRBC_RB_SIZE);
+			MMSCH_V5_0_INSERT_DIRECT_WT(tmp, ring->ring_size / 4);
+
+			if (j < 5) {
+				header.mjpegdec0[j].table_offset = item_offset;
+				header.mjpegdec0[j].init_status = 0;
+				header.mjpegdec0[j].table_size = table_size;
+			} else {
+				header.mjpegdec1[j - 5].table_offset = item_offset;
+				header.mjpegdec1[j - 5].init_status = 0;
+				header.mjpegdec1[j - 5].table_size = table_size;
+			}
+			header.total_size += table_size;
+			item_offset += table_size;
+		}
+
+		MMSCH_V5_0_INSERT_END();
+
+		/* send init table to MMSCH */
+		size = sizeof(struct mmsch_v5_0_init_header);
+		table_loc = (uint32_t *)table->cpu_addr;
+		memcpy((void *)table_loc, &header, size);
+
+		ctx_addr = table->gpu_addr;
+		WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_CTX_ADDR_LO, lower_32_bits(ctx_addr));
+		WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_CTX_ADDR_HI, upper_32_bits(ctx_addr));
+
+		tmp = RREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_VMID);
+		tmp &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK;
+		tmp |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT);
+		WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_VMID, tmp);
+
+		size = header.total_size;
+		WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_CTX_SIZE, size);
+
+		WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_MAILBOX_RESP, 0);
+
+		param = 0x00000001;
+		WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_MAILBOX_HOST, param);
+		tmp = 0;
+		timeout = 1000;
+		resp = 0;
+		expected = MMSCH_VF_MAILBOX_RESP__OK;
+		init_status =
+			((struct mmsch_v5_0_init_header *)(table_loc))->mjpegdec0[i].init_status;
+		while (resp != expected) {
+			resp = RREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_MAILBOX_RESP);
+
+			if (resp != 0)
+				break;
+			udelay(10);
+			tmp = tmp + 10;
+			if (tmp >= timeout) {
+				DRM_ERROR("failed to init MMSCH. TIME-OUT after %d usec"\
+					" waiting for regMMSCH_VF_MAILBOX_RESP "\
+					"(expected=0x%08x, readback=0x%08x)\n",
+					tmp, expected, resp);
+				return -EBUSY;
+			}
+		}
+		if (resp != expected && resp != MMSCH_VF_MAILBOX_RESP__INCOMPLETE &&
+				init_status != MMSCH_VF_ENGINE_STATUS__PASS)
+			DRM_ERROR("MMSCH init status is incorrect! readback=0x%08x, header init status for jpeg: %x\n",
+					resp, init_status);
+
+	}
+	return 0;
+}
+
 /**
  * jpeg_v5_0_1_start - start JPEG block
  *
@@ -581,6 +696,11 @@ static int jpeg_v5_0_1_set_powergating_state(struct amdgpu_ip_block *ip_block,
 	struct amdgpu_device *adev = ip_block->adev;
 	int ret;
 
+	if (amdgpu_sriov_vf(adev)) {
+		adev->jpeg.cur_state = AMD_PG_STATE_UNGATE;
+		return 0;
+	}
+
 	if (state == adev->jpeg.cur_state)
 		return 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
new file mode 100644
index 000000000000..d6f50b13e2ba
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
@@ -0,0 +1,355 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2024 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "amdgpu_gfx.h"
+#include "mes_userqueue.h"
+#include "amdgpu_userq_fence.h"
+
+#define AMDGPU_USERQ_PROC_CTX_SZ PAGE_SIZE
+#define AMDGPU_USERQ_GANG_CTX_SZ PAGE_SIZE
+
+static int
+mes_userq_map_gtt_bo_to_gart(struct amdgpu_bo *bo)
+{
+	int ret;
+
+	ret = amdgpu_bo_reserve(bo, true);
+	if (ret) {
+		DRM_ERROR("Failed to reserve bo. ret %d\n", ret);
+		goto err_reserve_bo_failed;
+	}
+
+	ret = amdgpu_ttm_alloc_gart(&bo->tbo);
+	if (ret) {
+		DRM_ERROR("Failed to bind bo to GART. ret %d\n", ret);
+		goto err_map_bo_gart_failed;
+	}
+
+	amdgpu_bo_unreserve(bo);
+	bo = amdgpu_bo_ref(bo);
+
+	return 0;
+
+err_map_bo_gart_failed:
+	amdgpu_bo_unreserve(bo);
+err_reserve_bo_failed:
+	return ret;
+}
+
+static int
+mes_userq_create_wptr_mapping(struct amdgpu_userq_mgr *uq_mgr,
+			      struct amdgpu_usermode_queue *queue,
+			      uint64_t wptr)
+{
+	struct amdgpu_bo_va_mapping *wptr_mapping;
+	struct amdgpu_vm *wptr_vm;
+	struct amdgpu_userq_obj *wptr_obj = &queue->wptr_obj;
+	int ret;
+
+	wptr_vm = queue->vm;
+	ret = amdgpu_bo_reserve(wptr_vm->root.bo, false);
+	if (ret)
+		return ret;
+
+	wptr &= AMDGPU_GMC_HOLE_MASK;
+	wptr_mapping = amdgpu_vm_bo_lookup_mapping(wptr_vm, wptr >> PAGE_SHIFT);
+	amdgpu_bo_unreserve(wptr_vm->root.bo);
+	if (!wptr_mapping) {
+		DRM_ERROR("Failed to lookup wptr bo\n");
+		return -EINVAL;
+	}
+
+	wptr_obj->obj = wptr_mapping->bo_va->base.bo;
+	if (wptr_obj->obj->tbo.base.size > PAGE_SIZE) {
+		DRM_ERROR("Requested GART mapping for wptr bo larger than one page\n");
+		return -EINVAL;
+	}
+
+	ret = mes_userq_map_gtt_bo_to_gart(wptr_obj->obj);
+	if (ret) {
+		DRM_ERROR("Failed to map wptr bo to GART\n");
+		return ret;
+	}
+
+	queue->wptr_obj.gpu_addr = amdgpu_bo_gpu_offset_no_check(wptr_obj->obj);
+	return 0;
+}
+
+static int convert_to_mes_priority(int priority)
+{
+	switch (priority) {
+	case AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_NORMAL_LOW:
+	default:
+		return AMDGPU_MES_PRIORITY_LEVEL_NORMAL;
+	case AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_LOW:
+		return AMDGPU_MES_PRIORITY_LEVEL_LOW;
+	case AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_NORMAL_HIGH:
+		return AMDGPU_MES_PRIORITY_LEVEL_MEDIUM;
+	case AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_HIGH:
+		return AMDGPU_MES_PRIORITY_LEVEL_HIGH;
+	}
+}
+
+static int mes_userq_map(struct amdgpu_userq_mgr *uq_mgr,
+			 struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct amdgpu_userq_obj *ctx = &queue->fw_obj;
+	struct amdgpu_mqd_prop *userq_props = queue->userq_prop;
+	struct mes_add_queue_input queue_input;
+	int r;
+
+	memset(&queue_input, 0x0, sizeof(struct mes_add_queue_input));
+
+	queue_input.process_va_start = 0;
+	queue_input.process_va_end = adev->vm_manager.max_pfn - 1;
+
+	/* set process quantum to 10 ms and gang quantum to 1 ms as default */
+	queue_input.process_quantum = 100000;
+	queue_input.gang_quantum = 10000;
+	queue_input.paging = false;
+
+	queue_input.process_context_addr = ctx->gpu_addr;
+	queue_input.gang_context_addr = ctx->gpu_addr + AMDGPU_USERQ_PROC_CTX_SZ;
+	queue_input.inprocess_gang_priority = AMDGPU_MES_PRIORITY_LEVEL_NORMAL;
+	queue_input.gang_global_priority_level = convert_to_mes_priority(queue->priority);
+
+	queue_input.process_id = queue->vm->pasid;
+	queue_input.queue_type = queue->queue_type;
+	queue_input.mqd_addr = queue->mqd.gpu_addr;
+	queue_input.wptr_addr = userq_props->wptr_gpu_addr;
+	queue_input.queue_size = userq_props->queue_size >> 2;
+	queue_input.doorbell_offset = userq_props->doorbell_index;
+	queue_input.page_table_base_addr = amdgpu_gmc_pd_addr(queue->vm->root.bo);
+	queue_input.wptr_mc_addr = queue->wptr_obj.gpu_addr;
+
+	amdgpu_mes_lock(&adev->mes);
+	r = adev->mes.funcs->add_hw_queue(&adev->mes, &queue_input);
+	amdgpu_mes_unlock(&adev->mes);
+	if (r) {
+		DRM_ERROR("Failed to map queue in HW, err (%d)\n", r);
+		return r;
+	}
+
+	DRM_DEBUG_DRIVER("Queue (doorbell:%d) mapped successfully\n", userq_props->doorbell_index);
+	return 0;
+}
+
+static int mes_userq_unmap(struct amdgpu_userq_mgr *uq_mgr,
+			   struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct mes_remove_queue_input queue_input;
+	struct amdgpu_userq_obj *ctx = &queue->fw_obj;
+	int r;
+
+	memset(&queue_input, 0x0, sizeof(struct mes_remove_queue_input));
+	queue_input.doorbell_offset = queue->doorbell_index;
+	queue_input.gang_context_addr = ctx->gpu_addr + AMDGPU_USERQ_PROC_CTX_SZ;
+
+	amdgpu_mes_lock(&adev->mes);
+	r = adev->mes.funcs->remove_hw_queue(&adev->mes, &queue_input);
+	amdgpu_mes_unlock(&adev->mes);
+	if (r)
+		DRM_ERROR("Failed to unmap queue in HW, err (%d)\n", r);
+	return r;
+}
+
+static int mes_userq_create_ctx_space(struct amdgpu_userq_mgr *uq_mgr,
+				      struct amdgpu_usermode_queue *queue,
+				      struct drm_amdgpu_userq_in *mqd_user)
+{
+	struct amdgpu_userq_obj *ctx = &queue->fw_obj;
+	int r, size;
+
+	/*
+	 * The FW expects at least one page space allocated for
+	 * process ctx and gang ctx each. Create an object
+	 * for the same.
+	 */
+	size = AMDGPU_USERQ_PROC_CTX_SZ + AMDGPU_USERQ_GANG_CTX_SZ;
+	r = amdgpu_userq_create_object(uq_mgr, ctx, size);
+	if (r) {
+		DRM_ERROR("Failed to allocate ctx space bo for userqueue, err:%d\n", r);
+		return r;
+	}
+
+	return 0;
+}
+
+static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
+				struct drm_amdgpu_userq_in *args_in,
+				struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct amdgpu_mqd *mqd_hw_default = &adev->mqds[queue->queue_type];
+	struct drm_amdgpu_userq_in *mqd_user = args_in;
+	struct amdgpu_mqd_prop *userq_props;
+	int r;
+
+	/* Structure to initialize MQD for userqueue using generic MQD init function */
+	userq_props = kzalloc(sizeof(struct amdgpu_mqd_prop), GFP_KERNEL);
+	if (!userq_props) {
+		DRM_ERROR("Failed to allocate memory for userq_props\n");
+		return -ENOMEM;
+	}
+
+	if (!mqd_user->wptr_va || !mqd_user->rptr_va ||
+	    !mqd_user->queue_va || mqd_user->queue_size == 0) {
+		DRM_ERROR("Invalid MQD parameters for userqueue\n");
+		r = -EINVAL;
+		goto free_props;
+	}
+
+	r = amdgpu_userq_create_object(uq_mgr, &queue->mqd, mqd_hw_default->mqd_size);
+	if (r) {
+		DRM_ERROR("Failed to create MQD object for userqueue\n");
+		goto free_props;
+	}
+
+	/* Initialize the MQD BO with user given values */
+	userq_props->wptr_gpu_addr = mqd_user->wptr_va;
+	userq_props->rptr_gpu_addr = mqd_user->rptr_va;
+	userq_props->queue_size = mqd_user->queue_size;
+	userq_props->hqd_base_gpu_addr = mqd_user->queue_va;
+	userq_props->mqd_gpu_addr = queue->mqd.gpu_addr;
+	userq_props->use_doorbell = true;
+	userq_props->doorbell_index = queue->doorbell_index;
+	userq_props->fence_address = queue->fence_drv->gpu_addr;
+
+	if (queue->queue_type == AMDGPU_HW_IP_COMPUTE) {
+		struct drm_amdgpu_userq_mqd_compute_gfx11 *compute_mqd;
+
+		if (mqd_user->mqd_size != sizeof(*compute_mqd)) {
+			DRM_ERROR("Invalid compute IP MQD size\n");
+			r = -EINVAL;
+			goto free_mqd;
+		}
+
+		compute_mqd = memdup_user(u64_to_user_ptr(mqd_user->mqd), mqd_user->mqd_size);
+		if (IS_ERR(compute_mqd)) {
+			DRM_ERROR("Failed to read user MQD\n");
+			r = -ENOMEM;
+			goto free_mqd;
+		}
+
+		userq_props->eop_gpu_addr = compute_mqd->eop_va;
+		userq_props->hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_NORMAL;
+		userq_props->hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MINIMUM;
+		userq_props->hqd_active = false;
+		userq_props->tmz_queue =
+			mqd_user->flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE;
+		kfree(compute_mqd);
+	} else if (queue->queue_type == AMDGPU_HW_IP_GFX) {
+		struct drm_amdgpu_userq_mqd_gfx11 *mqd_gfx_v11;
+
+		if (mqd_user->mqd_size != sizeof(*mqd_gfx_v11) || !mqd_user->mqd) {
+			DRM_ERROR("Invalid GFX MQD\n");
+			r = -EINVAL;
+			goto free_mqd;
+		}
+
+		mqd_gfx_v11 = memdup_user(u64_to_user_ptr(mqd_user->mqd), mqd_user->mqd_size);
+		if (IS_ERR(mqd_gfx_v11)) {
+			DRM_ERROR("Failed to read user MQD\n");
+			r = -ENOMEM;
+			goto free_mqd;
+		}
+
+		userq_props->shadow_addr = mqd_gfx_v11->shadow_va;
+		userq_props->csa_addr = mqd_gfx_v11->csa_va;
+		userq_props->tmz_queue =
+			mqd_user->flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE;
+		kfree(mqd_gfx_v11);
+	} else if (queue->queue_type == AMDGPU_HW_IP_DMA) {
+		struct drm_amdgpu_userq_mqd_sdma_gfx11 *mqd_sdma_v11;
+
+		if (mqd_user->mqd_size != sizeof(*mqd_sdma_v11) || !mqd_user->mqd) {
+			DRM_ERROR("Invalid SDMA MQD\n");
+			r = -EINVAL;
+			goto free_mqd;
+		}
+
+		mqd_sdma_v11 = memdup_user(u64_to_user_ptr(mqd_user->mqd), mqd_user->mqd_size);
+		if (IS_ERR(mqd_sdma_v11)) {
+			DRM_ERROR("Failed to read sdma user MQD\n");
+			r = -ENOMEM;
+			goto free_mqd;
+		}
+
+		userq_props->csa_addr = mqd_sdma_v11->csa_va;
+		kfree(mqd_sdma_v11);
+	}
+
+	queue->userq_prop = userq_props;
+
+	r = mqd_hw_default->init_mqd(adev, (void *)queue->mqd.cpu_ptr, userq_props);
+	if (r) {
+		DRM_ERROR("Failed to initialize MQD for userqueue\n");
+		goto free_mqd;
+	}
+
+	/* Create BO for FW operations */
+	r = mes_userq_create_ctx_space(uq_mgr, queue, mqd_user);
+	if (r) {
+		DRM_ERROR("Failed to allocate BO for userqueue (%d)", r);
+		goto free_mqd;
+	}
+
+	/* FW expects WPTR BOs to be mapped into GART */
+	r = mes_userq_create_wptr_mapping(uq_mgr, queue, userq_props->wptr_gpu_addr);
+	if (r) {
+		DRM_ERROR("Failed to create WPTR mapping\n");
+		goto free_ctx;
+	}
+
+	return 0;
+
+free_ctx:
+	amdgpu_userq_destroy_object(uq_mgr, &queue->fw_obj);
+
+free_mqd:
+	amdgpu_userq_destroy_object(uq_mgr, &queue->mqd);
+
+free_props:
+	kfree(userq_props);
+
+	return r;
+}
+
+static void
+mes_userq_mqd_destroy(struct amdgpu_userq_mgr *uq_mgr,
+		      struct amdgpu_usermode_queue *queue)
+{
+	amdgpu_userq_destroy_object(uq_mgr, &queue->fw_obj);
+	kfree(queue->userq_prop);
+	amdgpu_userq_destroy_object(uq_mgr, &queue->mqd);
+}
+
+const struct amdgpu_userq_funcs userq_mes_funcs = {
+	.mqd_create = mes_userq_mqd_create,
+	.mqd_destroy = mes_userq_mqd_destroy,
+	.unmap = mes_userq_unmap,
+	.map = mes_userq_map,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ad102.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.h
index 7ac87ef26aec..090ae8897770 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ad102.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.h
@@ -1,5 +1,6 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * Copyright 2023 Red Hat Inc.
+ * Copyright 2024 Advanced Micro Devices, Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -18,27 +19,12 @@
  * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
+ *
  */
-#include "priv.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ad102_ofa = {
-	.sclass = {
-		{ -1, -1, NVC9FA_VIDEO_OFA },
-		{}
-	}
-};
 
-int
-ad102_ofa_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-	      struct nvkm_engine **pengine)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_ofa_new(&ad102_ofa, device, type, inst, pengine);
+#ifndef MES_USERQ_H
+#define MES_USERQ_H
+#include "amdgpu_userq.h"
 
-	return -ENODEV;
-}
+extern const struct amdgpu_userq_funcs userq_mes_funcs;
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index e65916ada23b..c9eba537de09 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -287,6 +287,23 @@ static int convert_to_mes_queue_type(int queue_type)
 	return -1;
 }
 
+static int convert_to_mes_priority_level(int priority_level)
+{
+	switch (priority_level) {
+	case AMDGPU_MES_PRIORITY_LEVEL_LOW:
+		return AMD_PRIORITY_LEVEL_LOW;
+	case AMDGPU_MES_PRIORITY_LEVEL_NORMAL:
+	default:
+		return AMD_PRIORITY_LEVEL_NORMAL;
+	case AMDGPU_MES_PRIORITY_LEVEL_MEDIUM:
+		return AMD_PRIORITY_LEVEL_MEDIUM;
+	case AMDGPU_MES_PRIORITY_LEVEL_HIGH:
+		return AMD_PRIORITY_LEVEL_HIGH;
+	case AMDGPU_MES_PRIORITY_LEVEL_REALTIME:
+		return AMD_PRIORITY_LEVEL_REALTIME;
+	}
+}
+
 static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes,
 				  struct mes_add_queue_input *input)
 {
@@ -310,9 +327,9 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes,
 	mes_add_queue_pkt.gang_quantum = input->gang_quantum;
 	mes_add_queue_pkt.gang_context_addr = input->gang_context_addr;
 	mes_add_queue_pkt.inprocess_gang_priority =
-		input->inprocess_gang_priority;
+		convert_to_mes_priority_level(input->inprocess_gang_priority);
 	mes_add_queue_pkt.gang_global_priority_level =
-		input->gang_global_priority_level;
+		convert_to_mes_priority_level(input->gang_global_priority_level);
 	mes_add_queue_pkt.doorbell_offset = input->doorbell_offset;
 	mes_add_queue_pkt.mqd_addr = input->mqd_addr;
 
@@ -458,31 +475,6 @@ static int mes_v11_0_reset_queue_mmio(struct amdgpu_mes *mes, uint32_t queue_typ
 	return r;
 }
 
-static int mes_v11_0_reset_hw_queue(struct amdgpu_mes *mes,
-				    struct mes_reset_queue_input *input)
-{
-	if (input->use_mmio)
-		return mes_v11_0_reset_queue_mmio(mes, input->queue_type,
-						  input->me_id, input->pipe_id,
-						  input->queue_id, input->vmid);
-
-	union MESAPI__RESET mes_reset_queue_pkt;
-
-	memset(&mes_reset_queue_pkt, 0, sizeof(mes_reset_queue_pkt));
-
-	mes_reset_queue_pkt.header.type = MES_API_TYPE_SCHEDULER;
-	mes_reset_queue_pkt.header.opcode = MES_SCH_API_RESET;
-	mes_reset_queue_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
-
-	mes_reset_queue_pkt.doorbell_offset = input->doorbell_offset;
-	mes_reset_queue_pkt.gang_context_addr = input->gang_context_addr;
-	/*mes_reset_queue_pkt.reset_queue_only = 1;*/
-
-	return mes_v11_0_submit_pkt_and_poll_completion(mes,
-			&mes_reset_queue_pkt, sizeof(mes_reset_queue_pkt),
-			offsetof(union MESAPI__REMOVE_QUEUE, api_status));
-}
-
 static int mes_v11_0_map_legacy_queue(struct amdgpu_mes *mes,
 				      struct mes_map_legacy_queue_input *input)
 {
@@ -649,7 +641,7 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes,
 		break;
 	case MES_MISC_OP_CHANGE_CONFIG:
 		if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) < 0x63) {
-			dev_err(mes->adev->dev, "MES FW versoin must be larger than 0x63 to support limit single process feature.\n");
+			dev_err(mes->adev->dev, "MES FW version must be larger than 0x63 to support limit single process feature.\n");
 			return -EINVAL;
 		}
 		misc_pkt.opcode = MESAPI_MISC__CHANGE_CONFIG;
@@ -694,7 +686,8 @@ static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes)
 			mes->compute_hqd_mask[i];
 
 	for (i = 0; i < MAX_GFX_PIPES; i++)
-		mes_set_hw_res_pkt.gfx_hqd_mask[i] = mes->gfx_hqd_mask[i];
+		mes_set_hw_res_pkt.gfx_hqd_mask[i] =
+			mes->gfx_hqd_mask[i];
 
 	for (i = 0; i < MAX_SDMA_PIPES; i++)
 		mes_set_hw_res_pkt.sdma_hqd_mask[i] = mes->sdma_hqd_mask[i];
@@ -723,7 +716,7 @@ static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes)
 					mes->event_log_gpu_addr;
 	}
 
-	if (enforce_isolation)
+	if (adev->enforce_isolation[0] == AMDGPU_ENFORCE_ISOLATION_ENABLE)
 		mes_set_hw_res_pkt.limit_single_process = 1;
 
 	return mes_v11_0_submit_pkt_and_poll_completion(mes,
@@ -753,8 +746,8 @@ static int mes_v11_0_set_hw_resources_1(struct amdgpu_mes *mes)
 			offsetof(union MESAPI_SET_HW_RESOURCES_1, api_status));
 }
 
-static int mes_v11_0_reset_legacy_queue(struct amdgpu_mes *mes,
-					struct mes_reset_legacy_queue_input *input)
+static int mes_v11_0_reset_hw_queue(struct amdgpu_mes *mes,
+				    struct mes_reset_queue_input *input)
 {
 	union MESAPI__RESET mes_reset_queue_pkt;
 
@@ -772,7 +765,7 @@ static int mes_v11_0_reset_legacy_queue(struct amdgpu_mes *mes,
 	mes_reset_queue_pkt.queue_type =
 		convert_to_mes_queue_type(input->queue_type);
 
-	if (mes_reset_queue_pkt.queue_type == MES_QUEUE_TYPE_GFX) {
+	if (input->legacy_gfx) {
 		mes_reset_queue_pkt.reset_legacy_gfx = 1;
 		mes_reset_queue_pkt.pipe_id_lp = input->pipe_id;
 		mes_reset_queue_pkt.queue_id_lp = input->queue_id;
@@ -798,7 +791,6 @@ static const struct amdgpu_mes_funcs mes_v11_0_funcs = {
 	.suspend_gang = mes_v11_0_suspend_gang,
 	.resume_gang = mes_v11_0_resume_gang,
 	.misc_op = mes_v11_0_misc_op,
-	.reset_legacy_queue = mes_v11_0_reset_legacy_queue,
 	.reset_hw_queue = mes_v11_0_reset_hw_queue,
 };
 
@@ -894,6 +886,10 @@ static void mes_v11_0_get_fw_version(struct amdgpu_device *adev)
 {
 	int pipe;
 
+	/* return early if we have already fetched these */
+	if (adev->mes.sched_version && adev->mes.kiq_version)
+		return;
+
 	/* get MES scheduler/KIQ versions */
 	mutex_lock(&adev->srbm_mutex);
 
@@ -1697,22 +1693,10 @@ static int mes_v11_0_early_init(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
-static int mes_v11_0_late_init(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-
-	/* it's only intended for use in mes_self_test case, not for s0ix and reset */
-	if (!amdgpu_in_reset(adev) && !adev->in_s0ix && !adev->in_suspend &&
-	    (amdgpu_ip_version(adev, GC_HWIP, 0) != IP_VERSION(11, 0, 3)))
-		amdgpu_mes_self_test(adev);
-
-	return 0;
-}
-
 static const struct amd_ip_funcs mes_v11_0_ip_funcs = {
 	.name = "mes_v11_0",
 	.early_init = mes_v11_0_early_init,
-	.late_init = mes_v11_0_late_init,
+	.late_init = NULL,
 	.sw_init = mes_v11_0_sw_init,
 	.sw_fini = mes_v11_0_sw_fini,
 	.hw_init = mes_v11_0_hw_init,
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index 183dd3346da5..b4f17332d466 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -274,6 +274,23 @@ static int convert_to_mes_queue_type(int queue_type)
 	return -1;
 }
 
+static int convert_to_mes_priority_level(int priority_level)
+{
+	switch (priority_level) {
+	case AMDGPU_MES_PRIORITY_LEVEL_LOW:
+		return AMD_PRIORITY_LEVEL_LOW;
+	case AMDGPU_MES_PRIORITY_LEVEL_NORMAL:
+	default:
+		return AMD_PRIORITY_LEVEL_NORMAL;
+	case AMDGPU_MES_PRIORITY_LEVEL_MEDIUM:
+		return AMD_PRIORITY_LEVEL_MEDIUM;
+	case AMDGPU_MES_PRIORITY_LEVEL_HIGH:
+		return AMD_PRIORITY_LEVEL_HIGH;
+	case AMDGPU_MES_PRIORITY_LEVEL_REALTIME:
+		return AMD_PRIORITY_LEVEL_REALTIME;
+	}
+}
+
 static int mes_v12_0_add_hw_queue(struct amdgpu_mes *mes,
 				  struct mes_add_queue_input *input)
 {
@@ -297,9 +314,9 @@ static int mes_v12_0_add_hw_queue(struct amdgpu_mes *mes,
 	mes_add_queue_pkt.gang_quantum = input->gang_quantum;
 	mes_add_queue_pkt.gang_context_addr = input->gang_context_addr;
 	mes_add_queue_pkt.inprocess_gang_priority =
-		input->inprocess_gang_priority;
+		convert_to_mes_priority_level(input->inprocess_gang_priority);
 	mes_add_queue_pkt.gang_global_priority_level =
-		input->gang_global_priority_level;
+		convert_to_mes_priority_level(input->gang_global_priority_level);
 	mes_add_queue_pkt.doorbell_offset = input->doorbell_offset;
 	mes_add_queue_pkt.mqd_addr = input->mqd_addr;
 
@@ -477,32 +494,6 @@ static int mes_v12_0_reset_queue_mmio(struct amdgpu_mes *mes, uint32_t queue_typ
 	return r;
 }
 
-static int mes_v12_0_reset_hw_queue(struct amdgpu_mes *mes,
-				    struct mes_reset_queue_input *input)
-{
-	union MESAPI__RESET mes_reset_queue_pkt;
-	int pipe;
-
-	memset(&mes_reset_queue_pkt, 0, sizeof(mes_reset_queue_pkt));
-
-	mes_reset_queue_pkt.header.type = MES_API_TYPE_SCHEDULER;
-	mes_reset_queue_pkt.header.opcode = MES_SCH_API_RESET;
-	mes_reset_queue_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
-
-	mes_reset_queue_pkt.doorbell_offset = input->doorbell_offset;
-	mes_reset_queue_pkt.gang_context_addr = input->gang_context_addr;
-	/*mes_reset_queue_pkt.reset_queue_only = 1;*/
-
-	if (mes->adev->enable_uni_mes)
-		pipe = AMDGPU_MES_KIQ_PIPE;
-	else
-		pipe = AMDGPU_MES_SCHED_PIPE;
-
-	return mes_v12_0_submit_pkt_and_poll_completion(mes, pipe,
-			&mes_reset_queue_pkt, sizeof(mes_reset_queue_pkt),
-			offsetof(union MESAPI__REMOVE_QUEUE, api_status));
-}
-
 static int mes_v12_0_map_legacy_queue(struct amdgpu_mes *mes,
 				      struct mes_map_legacy_queue_input *input)
 {
@@ -762,7 +753,7 @@ static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe)
 				pipe * (AMDGPU_MES_LOG_BUFFER_SIZE + AMDGPU_MES_MSCRATCH_SIZE);
 	}
 
-	if (enforce_isolation)
+	if (adev->enforce_isolation[0] == AMDGPU_ENFORCE_ISOLATION_ENABLE)
 		mes_set_hw_res_pkt.limit_single_process = 1;
 
 	return mes_v12_0_submit_pkt_and_poll_completion(mes, pipe,
@@ -845,8 +836,8 @@ static void mes_v12_0_enable_unmapped_doorbell_handling(
 	WREG32_SOC15(GC, 0, regCP_UNMAPPED_DOORBELL, data);
 }
 
-static int mes_v12_0_reset_legacy_queue(struct amdgpu_mes *mes,
-					struct mes_reset_legacy_queue_input *input)
+static int mes_v12_0_reset_hw_queue(struct amdgpu_mes *mes,
+				    struct mes_reset_queue_input *input)
 {
 	union MESAPI__RESET mes_reset_queue_pkt;
 	int pipe;
@@ -865,7 +856,7 @@ static int mes_v12_0_reset_legacy_queue(struct amdgpu_mes *mes,
 	mes_reset_queue_pkt.queue_type =
 		convert_to_mes_queue_type(input->queue_type);
 
-	if (mes_reset_queue_pkt.queue_type == MES_QUEUE_TYPE_GFX) {
+	if (input->legacy_gfx) {
 		mes_reset_queue_pkt.reset_legacy_gfx = 1;
 		mes_reset_queue_pkt.pipe_id_lp = input->pipe_id;
 		mes_reset_queue_pkt.queue_id_lp = input->queue_id;
@@ -878,7 +869,7 @@ static int mes_v12_0_reset_legacy_queue(struct amdgpu_mes *mes,
 		mes_reset_queue_pkt.doorbell_offset = input->doorbell_offset;
 	}
 
-	if (mes->adev->enable_uni_mes)
+	if (input->is_kq)
 		pipe = AMDGPU_MES_KIQ_PIPE;
 	else
 		pipe = AMDGPU_MES_SCHED_PIPE;
@@ -896,7 +887,6 @@ static const struct amdgpu_mes_funcs mes_v12_0_funcs = {
 	.suspend_gang = mes_v12_0_suspend_gang,
 	.resume_gang = mes_v12_0_resume_gang,
 	.misc_op = mes_v12_0_misc_op,
-	.reset_legacy_queue = mes_v12_0_reset_legacy_queue,
 	.reset_hw_queue = mes_v12_0_reset_hw_queue,
 };
 
@@ -1392,17 +1382,20 @@ static int mes_v12_0_queue_init(struct amdgpu_device *adev,
 		mes_v12_0_queue_init_register(ring);
 	}
 
-	/* get MES scheduler/KIQ versions */
-	mutex_lock(&adev->srbm_mutex);
-	soc21_grbm_select(adev, 3, pipe, 0, 0);
+	if (((pipe == AMDGPU_MES_SCHED_PIPE) && !adev->mes.sched_version) ||
+	    ((pipe == AMDGPU_MES_KIQ_PIPE) && !adev->mes.kiq_version)) {
+		/* get MES scheduler/KIQ versions */
+		mutex_lock(&adev->srbm_mutex);
+		soc21_grbm_select(adev, 3, pipe, 0, 0);
 
-	if (pipe == AMDGPU_MES_SCHED_PIPE)
-		adev->mes.sched_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO);
-	else if (pipe == AMDGPU_MES_KIQ_PIPE && adev->enable_mes_kiq)
-		adev->mes.kiq_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO);
+		if (pipe == AMDGPU_MES_SCHED_PIPE)
+			adev->mes.sched_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO);
+		else if (pipe == AMDGPU_MES_KIQ_PIPE && adev->enable_mes_kiq)
+			adev->mes.kiq_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO);
 
-	soc21_grbm_select(adev, 0, 0, 0, 0);
-	mutex_unlock(&adev->srbm_mutex);
+		soc21_grbm_select(adev, 0, 0, 0, 0);
+		mutex_unlock(&adev->srbm_mutex);
+	}
 
 	return 0;
 }
@@ -1808,21 +1801,10 @@ static int mes_v12_0_early_init(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
-static int mes_v12_0_late_init(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-
-	/* it's only intended for use in mes_self_test case, not for s0ix and reset */
-	if (!amdgpu_in_reset(adev) && !adev->in_s0ix && !adev->in_suspend)
-		amdgpu_mes_self_test(adev);
-
-	return 0;
-}
-
 static const struct amd_ip_funcs mes_v12_0_ip_funcs = {
 	.name = "mes_v12_0",
 	.early_init = mes_v12_0_early_init,
-	.late_init = mes_v12_0_late_init,
+	.late_init = NULL,
 	.sw_init = mes_v12_0_sw_init,
 	.sw_fini = mes_v12_0_sw_fini,
 	.hw_init = mes_v12_0_hw_init,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 84cde1239ee4..76167fadb292 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -30,6 +30,7 @@
 #include "soc15_common.h"
 #include "soc15.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_psp.h"
 
 #define regVM_L2_CNTL3_DEFAULT	0x80100007
 #define regVM_L2_CNTL4_DEFAULT	0x000000c1
@@ -192,10 +193,8 @@ static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev)
 	uint32_t tmp, inst_mask;
 	int i;
 
-	/* Setup TLB control */
-	inst_mask = adev->aid_mask;
-	for_each_inst(i, inst_mask) {
-		tmp = RREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL);
+	if (amdgpu_sriov_reg_indirect_l1_tlb_cntl(adev)) {
+		tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL);
 
 		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
 				    1);
@@ -209,7 +208,26 @@ static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev)
 				    MTYPE, MTYPE_UC);/* XXX for emulation. */
 		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
 
-		WREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL, tmp);
+		psp_reg_program_no_ring(&adev->psp, tmp, PSP_REG_MMHUB_L1_TLB_CNTL);
+	} else {
+		inst_mask = adev->aid_mask;
+		for_each_inst(i, inst_mask) {
+			tmp = RREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL);
+
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
+					    1);
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+					    SYSTEM_ACCESS_MODE, 3);
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+					    ENABLE_ADVANCED_DRIVER_MODEL, 1);
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+					    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+					    MTYPE, MTYPE_UC);/* XXX for emulation. */
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
+
+			WREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL, tmp);
+		}
 	}
 }
 
@@ -221,6 +239,9 @@ static void mmhub_v1_8_init_snoop_override_regs(struct amdgpu_device *adev)
 	uint32_t distance = regDAGB1_WRCLI_GPU_SNOOP_OVERRIDE -
 			    regDAGB0_WRCLI_GPU_SNOOP_OVERRIDE;
 
+	if (amdgpu_sriov_vf(adev))
+		return;
+
 	inst_mask = adev->aid_mask;
 	for_each_inst(i, inst_mask) {
 		for (j = 0; j < 5; j++) { /* DAGB instances */
@@ -454,6 +475,30 @@ static int mmhub_v1_8_gart_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
+static void mmhub_v1_8_disable_l1_tlb(struct amdgpu_device *adev)
+{
+	u32 tmp;
+	u32 i, inst_mask;
+
+	if (amdgpu_sriov_reg_indirect_l1_tlb_cntl(adev)) {
+		tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    ENABLE_ADVANCED_DRIVER_MODEL, 0);
+		psp_reg_program_no_ring(&adev->psp, tmp, PSP_REG_MMHUB_L1_TLB_CNTL);
+	} else {
+		inst_mask = adev->aid_mask;
+		for_each_inst(i, inst_mask) {
+			tmp = RREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL);
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
+					    0);
+			tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+					    ENABLE_ADVANCED_DRIVER_MODEL, 0);
+			WREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL, tmp);
+		}
+	}
+}
+
 static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
@@ -467,15 +512,6 @@ static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 		for (i = 0; i < 16; i++)
 			WREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT0_CNTL,
 					    i * hub->ctx_distance, 0);
-
-		/* Setup TLB control */
-		tmp = RREG32_SOC15(MMHUB, j, regMC_VM_MX_L1_TLB_CNTL);
-		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
-				    0);
-		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-				    ENABLE_ADVANCED_DRIVER_MODEL, 0);
-		WREG32_SOC15(MMHUB, j, regMC_VM_MX_L1_TLB_CNTL, tmp);
-
 		if (!amdgpu_sriov_vf(adev)) {
 			/* Setup L2 cache */
 			tmp = RREG32_SOC15(MMHUB, j, regVM_L2_CNTL);
@@ -485,6 +521,8 @@ static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 			WREG32_SOC15(MMHUB, j, regVM_L2_CNTL3, 0);
 		}
 	}
+
+	mmhub_v1_8_disable_l1_tlb(adev);
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v5_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v5_0.h
new file mode 100644
index 000000000000..6f749814929f
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v5_0.h
@@ -0,0 +1,144 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __MMSCH_V5_0_H__
+#define __MMSCH_V5_0_H__
+
+#include "amdgpu_vcn.h"
+
+#define MMSCH_VERSION_MAJOR	5
+#define MMSCH_VERSION_MINOR	0
+#define MMSCH_VERSION	(MMSCH_VERSION_MAJOR << 16 | MMSCH_VERSION_MINOR)
+
+#define RB_ENABLED (1 << 0)
+#define RB4_ENABLED (1 << 1)
+
+#define MMSCH_VF_ENGINE_STATUS__PASS 0x1
+
+#define MMSCH_VF_MAILBOX_RESP__OK                      0x1
+#define MMSCH_VF_MAILBOX_RESP__INCOMPLETE              0x2
+#define MMSCH_VF_MAILBOX_RESP__FAILED                  0x3
+#define MMSCH_VF_MAILBOX_RESP__FAILED_SMALL_CTX_SIZE   0x4
+#define MMSCH_VF_MAILBOX_RESP__UNKNOWN_CMD             0x5
+
+enum mmsch_v5_0_command_type {
+	MMSCH_COMMAND__DIRECT_REG_WRITE = 0,
+	MMSCH_COMMAND__DIRECT_REG_POLLING = 2,
+	MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE = 3,
+	MMSCH_COMMAND__INDIRECT_REG_WRITE = 8,
+	MMSCH_COMMAND__END = 0xf
+};
+
+struct mmsch_v5_0_table_info {
+	uint32_t init_status;
+	uint32_t table_offset;
+	uint32_t table_size;
+};
+
+struct mmsch_v5_0_init_header {
+	uint32_t version;
+	uint32_t total_size;
+	struct mmsch_v5_0_table_info vcn0;
+	struct mmsch_v5_0_table_info mjpegdec0[5];
+	struct mmsch_v5_0_table_info mjpegdec1[5];
+};
+
+struct mmsch_v5_0_cmd_direct_reg_header {
+	uint32_t reg_offset   : 28;
+	uint32_t command_type : 4;
+};
+
+struct mmsch_v5_0_cmd_indirect_reg_header {
+	uint32_t reg_offset    : 20;
+	uint32_t reg_idx_space : 8;
+	uint32_t command_type  : 4;
+};
+
+struct mmsch_v5_0_cmd_direct_write {
+	struct mmsch_v5_0_cmd_direct_reg_header cmd_header;
+	uint32_t reg_value;
+};
+
+struct mmsch_v5_0_cmd_direct_read_modify_write {
+	struct mmsch_v5_0_cmd_direct_reg_header cmd_header;
+	uint32_t write_data;
+	uint32_t mask_value;
+};
+
+struct mmsch_v5_0_cmd_direct_polling {
+	struct mmsch_v5_0_cmd_direct_reg_header cmd_header;
+	uint32_t mask_value;
+	uint32_t wait_value;
+};
+
+struct mmsch_v5_0_cmd_end {
+	struct mmsch_v5_0_cmd_direct_reg_header cmd_header;
+};
+
+struct mmsch_v5_0_cmd_indirect_write {
+	struct mmsch_v5_0_cmd_indirect_reg_header cmd_header;
+	uint32_t reg_value;
+};
+
+#define MMSCH_V5_0_INSERT_DIRECT_RD_MOD_WT(reg, mask, data) { \
+	size = sizeof(struct mmsch_v5_0_cmd_direct_read_modify_write); \
+	size_dw = size / 4; \
+	direct_rd_mod_wt.cmd_header.reg_offset = reg; \
+	direct_rd_mod_wt.mask_value = mask; \
+	direct_rd_mod_wt.write_data = data; \
+	memcpy((void *)table_loc, &direct_rd_mod_wt, size); \
+	table_loc += size_dw; \
+	table_size += size_dw; \
+}
+
+#define MMSCH_V5_0_INSERT_DIRECT_WT(reg, value) { \
+	size = sizeof(struct mmsch_v5_0_cmd_direct_write); \
+	size_dw = size / 4; \
+	direct_wt.cmd_header.reg_offset = reg; \
+	direct_wt.reg_value = value; \
+	memcpy((void *)table_loc, &direct_wt, size); \
+	table_loc += size_dw; \
+	table_size += size_dw; \
+}
+
+#define MMSCH_V5_0_INSERT_DIRECT_POLL(reg, mask, wait) { \
+	size = sizeof(struct mmsch_v5_0_cmd_direct_polling); \
+	size_dw = size / 4; \
+	direct_poll.cmd_header.reg_offset = reg; \
+	direct_poll.mask_value = mask; \
+	direct_poll.wait_value = wait; \
+	memcpy((void *)table_loc, &direct_poll, size); \
+	table_loc += size_dw; \
+	table_size += size_dw; \
+}
+
+#define MMSCH_V5_0_INSERT_END() { \
+	size = sizeof(struct mmsch_v5_0_cmd_end); \
+	size_dw = size / 4; \
+	memcpy((void *)table_loc, &end, size); \
+	table_loc += size_dw; \
+	table_size += size_dw; \
+}
+
+#endif
+
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
index f5411b798e11..48101a34e049 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
@@ -274,6 +274,7 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work)
 {
 	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work);
 	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+	struct amdgpu_reset_context reset_context = { 0 };
 
 	amdgpu_virt_fini_data_exchange(adev);
 
@@ -281,8 +282,6 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work)
 	if (amdgpu_device_should_recover_gpu(adev)
 		&& (!amdgpu_device_has_job_running(adev) ||
 			adev->sdma_timeout == MAX_SCHEDULE_TIMEOUT)) {
-		struct amdgpu_reset_context reset_context;
-		memset(&reset_context, 0, sizeof(reset_context));
 
 		reset_context.method = AMD_RESET_METHOD_NONE;
 		reset_context.reset_req_dev = adev;
@@ -293,6 +292,19 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work)
 	}
 }
 
+static void xgpu_ai_mailbox_bad_pages_work(struct work_struct *work)
+{
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work);
+	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+
+	if (down_read_trylock(&adev->reset_domain->sem)) {
+		amdgpu_virt_fini_data_exchange(adev);
+		amdgpu_virt_request_bad_pages(adev);
+		amdgpu_virt_init_data_exchange(adev);
+		up_read(&adev->reset_domain->sem);
+	}
+}
+
 static int xgpu_ai_set_mailbox_rcv_irq(struct amdgpu_device *adev,
 				       struct amdgpu_irq_src *src,
 				       unsigned type,
@@ -312,26 +324,42 @@ static int xgpu_ai_mailbox_rcv_irq(struct amdgpu_device *adev,
 				   struct amdgpu_iv_entry *entry)
 {
 	enum idh_event event = xgpu_ai_mailbox_peek_msg(adev);
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	switch (event) {
-		case IDH_FLR_NOTIFICATION:
+	case IDH_RAS_BAD_PAGES_NOTIFICATION:
+		xgpu_ai_mailbox_send_ack(adev);
+		if (amdgpu_sriov_runtime(adev))
+			schedule_work(&adev->virt.bad_pages_work);
+		break;
+	case IDH_UNRECOV_ERR_NOTIFICATION:
+		xgpu_ai_mailbox_send_ack(adev);
+		ras->is_rma = true;
+		dev_err(adev->dev, "VF is in an unrecoverable state. Runtime Services are halted.\n");
 		if (amdgpu_sriov_runtime(adev))
 			WARN_ONCE(!amdgpu_reset_domain_schedule(adev->reset_domain,
-								&adev->virt.flr_work),
-				  "Failed to queue work! at %s",
-				  __func__);
+					&adev->virt.flr_work),
+					"Failed to queue work! at %s",
+					__func__);
 		break;
-		case IDH_QUERY_ALIVE:
-			xgpu_ai_mailbox_send_ack(adev);
-			break;
-		/* READY_TO_ACCESS_GPU is fetched by kernel polling, IRQ can ignore
-		 * it byfar since that polling thread will handle it,
-		 * other msg like flr complete is not handled here.
-		 */
-		case IDH_CLR_MSG_BUF:
-		case IDH_FLR_NOTIFICATION_CMPL:
-		case IDH_READY_TO_ACCESS_GPU:
-		default:
+	case IDH_FLR_NOTIFICATION:
+		if (amdgpu_sriov_runtime(adev))
+			WARN_ONCE(!amdgpu_reset_domain_schedule(adev->reset_domain,
+						&adev->virt.flr_work),
+					"Failed to queue work! at %s",
+					__func__);
+		break;
+	case IDH_QUERY_ALIVE:
+		xgpu_ai_mailbox_send_ack(adev);
+		break;
+	/* READY_TO_ACCESS_GPU is fetched by kernel polling, IRQ can ignore
+	 * it byfar since that polling thread will handle it,
+	 * other msg like flr complete is not handled here.
+	 */
+	case IDH_CLR_MSG_BUF:
+	case IDH_FLR_NOTIFICATION_CMPL:
+	case IDH_READY_TO_ACCESS_GPU:
+	default:
 		break;
 	}
 
@@ -387,6 +415,7 @@ int xgpu_ai_mailbox_get_irq(struct amdgpu_device *adev)
 	}
 
 	INIT_WORK(&adev->virt.flr_work, xgpu_ai_mailbox_flr_work);
+	INIT_WORK(&adev->virt.bad_pages_work, xgpu_ai_mailbox_bad_pages_work);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
index ed57cbc150af..874b9f8f9804 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h
@@ -40,6 +40,7 @@ enum idh_request {
 	IDH_LOG_VF_ERROR       = 200,
 	IDH_READY_TO_RESET 	= 201,
 	IDH_RAS_POISON  = 202,
+	IDH_REQ_RAS_BAD_PAGES = 205,
 };
 
 enum idh_event {
@@ -54,6 +55,9 @@ enum idh_event {
 	IDH_RAS_POISON_READY,
 	IDH_PF_SOFT_FLR_NOTIFICATION,
 	IDH_RAS_ERROR_DETECTED,
+	IDH_RAS_BAD_PAGES_READY = 15,
+	IDH_RAS_BAD_PAGES_NOTIFICATION = 16,
+	IDH_UNRECOV_ERR_NOTIFICATION = 17,
 	IDH_TEXT_MESSAGE = 255,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
index 5aadf24cb202..f6d8597452ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
@@ -67,6 +67,8 @@ static int xgpu_nv_mailbox_rcv_msg(struct amdgpu_device *adev,
 	reg = RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW0);
 	if (reg == IDH_FAIL)
 		r = -EINVAL;
+	if (reg == IDH_UNRECOV_ERR_NOTIFICATION)
+		r = -ENODEV;
 	else if (reg != event)
 		return -ENOENT;
 
@@ -103,6 +105,7 @@ static int xgpu_nv_poll_msg(struct amdgpu_device *adev, enum idh_event event)
 {
 	int r;
 	uint64_t timeout, now;
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	now = (uint64_t)ktime_to_ms(ktime_get());
 	timeout = now + NV_MAILBOX_POLL_MSG_TIMEDOUT;
@@ -110,8 +113,16 @@ static int xgpu_nv_poll_msg(struct amdgpu_device *adev, enum idh_event event)
 	do {
 		r = xgpu_nv_mailbox_rcv_msg(adev, event);
 		if (!r) {
-			dev_dbg(adev->dev, "rcv_msg 0x%x after %llu ms\n", event, NV_MAILBOX_POLL_MSG_TIMEDOUT - timeout + now);
+			dev_dbg(adev->dev, "rcv_msg 0x%x after %llu ms\n",
+					event, NV_MAILBOX_POLL_MSG_TIMEDOUT - timeout + now);
 			return 0;
+		} else if (r == -ENODEV) {
+			if (!amdgpu_ras_is_rma(adev)) {
+				ras->is_rma = true;
+				dev_err(adev->dev, "VF is in an unrecoverable state. "
+						"Runtime Services are halted.\n");
+			}
+			return r;
 		}
 
 		msleep(10);
@@ -166,6 +177,10 @@ static int xgpu_nv_send_access_requests_with_param(struct amdgpu_device *adev,
 	enum idh_event event = -1;
 
 send_request:
+
+	if (amdgpu_ras_is_rma(adev))
+		return -ENODEV;
+
 	xgpu_nv_mailbox_trans_msg(adev, req, data1, data2, data3);
 
 	switch (req) {
@@ -187,6 +202,9 @@ send_request:
 	case IDH_REQ_RAS_CPER_DUMP:
 		event = IDH_RAS_CPER_DUMP_READY;
 		break;
+	case IDH_REQ_RAS_BAD_PAGES:
+		event = IDH_RAS_BAD_PAGES_READY;
+		break;
 	default:
 		break;
 	}
@@ -320,6 +338,7 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 {
 	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, flr_work);
 	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+	struct amdgpu_reset_context reset_context = { 0 };
 
 	amdgpu_virt_fini_data_exchange(adev);
 
@@ -330,8 +349,6 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 		adev->gfx_timeout == MAX_SCHEDULE_TIMEOUT ||
 		adev->compute_timeout == MAX_SCHEDULE_TIMEOUT ||
 		adev->video_timeout == MAX_SCHEDULE_TIMEOUT)) {
-		struct amdgpu_reset_context reset_context;
-		memset(&reset_context, 0, sizeof(reset_context));
 
 		reset_context.method = AMD_RESET_METHOD_NONE;
 		reset_context.reset_req_dev = adev;
@@ -342,6 +359,19 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 	}
 }
 
+static void xgpu_nv_mailbox_bad_pages_work(struct work_struct *work)
+{
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work);
+	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+
+	if (down_read_trylock(&adev->reset_domain->sem)) {
+		amdgpu_virt_fini_data_exchange(adev);
+		amdgpu_virt_request_bad_pages(adev);
+		amdgpu_virt_init_data_exchange(adev);
+		up_read(&adev->reset_domain->sem);
+	}
+}
+
 static int xgpu_nv_set_mailbox_rcv_irq(struct amdgpu_device *adev,
 				       struct amdgpu_irq_src *src,
 				       unsigned type,
@@ -364,8 +394,27 @@ static int xgpu_nv_mailbox_rcv_irq(struct amdgpu_device *adev,
 				   struct amdgpu_iv_entry *entry)
 {
 	enum idh_event event = xgpu_nv_mailbox_peek_msg(adev);
+	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	switch (event) {
+	case IDH_RAS_BAD_PAGES_NOTIFICATION:
+		xgpu_nv_mailbox_send_ack(adev);
+		if (amdgpu_sriov_runtime(adev))
+			schedule_work(&adev->virt.bad_pages_work);
+		break;
+	case IDH_UNRECOV_ERR_NOTIFICATION:
+		xgpu_nv_mailbox_send_ack(adev);
+		if (!amdgpu_ras_is_rma(adev)) {
+			ras->is_rma = true;
+			dev_err(adev->dev, "VF is in an unrecoverable state. Runtime Services are halted.\n");
+		}
+
+		if (amdgpu_sriov_runtime(adev))
+			WARN_ONCE(!amdgpu_reset_domain_schedule(adev->reset_domain,
+						&adev->virt.flr_work),
+					"Failed to queue work! at %s",
+					__func__);
+		break;
 	case IDH_FLR_NOTIFICATION:
 		if (amdgpu_sriov_runtime(adev))
 			WARN_ONCE(!amdgpu_reset_domain_schedule(adev->reset_domain,
@@ -436,6 +485,7 @@ int xgpu_nv_mailbox_get_irq(struct amdgpu_device *adev)
 	}
 
 	INIT_WORK(&adev->virt.flr_work, xgpu_nv_mailbox_flr_work);
+	INIT_WORK(&adev->virt.bad_pages_work, xgpu_nv_mailbox_bad_pages_work);
 
 	return 0;
 }
@@ -480,6 +530,11 @@ static int xgpu_nv_req_ras_cper_dump(struct amdgpu_device *adev, u64 vf_rptr)
 		adev, IDH_REQ_RAS_CPER_DUMP, vf_rptr_hi, vf_rptr_lo, 0);
 }
 
+static int xgpu_nv_req_ras_bad_pages(struct amdgpu_device *adev)
+{
+	return xgpu_nv_send_access_requests(adev, IDH_REQ_RAS_BAD_PAGES);
+}
+
 const struct amdgpu_virt_ops xgpu_nv_virt_ops = {
 	.req_full_gpu	= xgpu_nv_request_full_gpu_access,
 	.rel_full_gpu	= xgpu_nv_release_full_gpu_access,
@@ -492,4 +547,5 @@ const struct amdgpu_virt_ops xgpu_nv_virt_ops = {
 	.rcvd_ras_intr = xgpu_nv_rcvd_ras_intr,
 	.req_ras_err_count = xgpu_nv_req_ras_err_count,
 	.req_ras_cper_dump = xgpu_nv_req_ras_cper_dump,
+	.req_bad_pages = xgpu_nv_req_ras_bad_pages,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h
index 72c9fceb9d79..5808689562cc 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h
@@ -42,6 +42,7 @@ enum idh_request {
 	IDH_RAS_POISON	= 202,
 	IDH_REQ_RAS_ERROR_COUNT = 203,
 	IDH_REQ_RAS_CPER_DUMP = 204,
+	IDH_REQ_RAS_BAD_PAGES = 205,
 };
 
 enum idh_event {
@@ -58,6 +59,9 @@ enum idh_event {
 	IDH_RAS_ERROR_DETECTED,
 	IDH_RAS_ERROR_COUNT_READY = 11,
 	IDH_RAS_CPER_DUMP_READY = 14,
+	IDH_RAS_BAD_PAGES_READY = 15,
+	IDH_RAS_BAD_PAGES_NOTIFICATION = 16,
+	IDH_UNRECOV_ERR_NOTIFICATION = 17,
 
 	IDH_TEXT_MESSAGE = 255,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_11.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_11.c
index 2ece3ae75ec1..bed5ef4d8788 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_11.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_11.c
@@ -360,7 +360,7 @@ static void nbio_v7_11_get_clockgating_state(struct amdgpu_device *adev,
 		*flags |= AMD_CG_SUPPORT_BIF_LS;
 }
 
-#define MMIO_REG_HOLE_OFFSET (0x80000 - PAGE_SIZE)
+#define MMIO_REG_HOLE_OFFSET 0x44000
 
 static void nbio_v7_11_set_reg_remap(struct amdgpu_device *adev)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index f23cb79110d6..a376f072700d 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -177,8 +177,12 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 {
 	u32 doorbell_range = 0, doorbell_ctrl = 0;
 	u32 aid_id = instance;
+	u32 range_size;
 
 	if (use_doorbell) {
+		range_size = (amdgpu_ip_version(adev, GC_HWIP, 0) ==
+						IP_VERSION(9, 5, 0)) ?
+						0xb : 0x9;
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
 				BIF_DOORBELL0_RANGE_OFFSET_ENTRY,
@@ -186,7 +190,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
 				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
-				0x9);
+				range_size);
 		if (aid_id)
 			doorbell_range = REG_SET_FIELD(doorbell_range,
 					DOORBELL0_CTRL_ENTRY_0,
@@ -204,7 +208,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 				S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x4);
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
-				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x9);
+				S2A_DOORBELL_PORT1_RANGE_SIZE, range_size);
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4);
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index bb5dfc410a66..215543575f47 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -533,7 +533,7 @@ static int psp_v11_0_memory_training(struct psp_context *psp, uint32_t ops)
 			}
 
 			memcpy_toio(adev->mman.aper_base_kaddr, buf, sz);
-			adev->hdp.funcs->flush_hdp(adev, NULL);
+			amdgpu_device_flush_hdp(adev, NULL);
 			vfree(buf);
 			drm_dev_exit(idx);
 		} else {
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
index fcd708eae75c..80153f837470 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c
@@ -34,9 +34,6 @@
 #include "sdma0/sdma0_4_0_offset.h"
 #include "nbio/nbio_7_4_offset.h"
 
-#include "oss/osssys_4_0_offset.h"
-#include "oss/osssys_4_0_sh_mask.h"
-
 MODULE_FIRMWARE("amdgpu/renoir_asd.bin");
 MODULE_FIRMWARE("amdgpu/renoir_ta.bin");
 MODULE_FIRMWARE("amdgpu/green_sardine_asd.bin");
@@ -99,9 +96,6 @@ static int psp_v12_0_bootloader_load_sysdrv(struct psp_context *psp)
 	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35,
 	       psp_gfxdrv_command_reg);
 
-	/* there might be handshake issue with hardware which needs delay */
-	mdelay(20);
-
 	ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35),
 			   0x80000000, 0x80000000, false);
 
@@ -138,8 +132,6 @@ static int psp_v12_0_bootloader_load_sos(struct psp_context *psp)
 	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35,
 	       psp_gfxdrv_command_reg);
 
-	/* there might be handshake issue with hardware which needs delay */
-	mdelay(20);
 	ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_81),
 			   RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81),
 			   0, true);
@@ -147,37 +139,6 @@ static int psp_v12_0_bootloader_load_sos(struct psp_context *psp)
 	return ret;
 }
 
-static void psp_v12_0_reroute_ih(struct psp_context *psp)
-{
-	struct amdgpu_device *adev = psp->adev;
-	uint32_t tmp;
-
-	/* Change IH ring for VMC */
-	tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1244b);
-	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, CLIENT_TYPE, 1);
-	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1);
-
-	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 3);
-	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp);
-	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET);
-
-	mdelay(20);
-	psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
-		     0x80000000, 0x8000FFFF, false);
-
-	/* Change IH ring for UMC */
-	tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1216b);
-	tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1);
-
-	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, 4);
-	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, tmp);
-	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_GBR_IH_SET);
-
-	mdelay(20);
-	psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
-		     0x80000000, 0x8000FFFF, false);
-}
-
 static int psp_v12_0_ring_create(struct psp_context *psp,
 				enum psp_ring_type ring_type)
 {
@@ -186,49 +147,23 @@ static int psp_v12_0_ring_create(struct psp_context *psp,
 	struct psp_ring *ring = &psp->km_ring;
 	struct amdgpu_device *adev = psp->adev;
 
-	psp_v12_0_reroute_ih(psp);
-
-	if (amdgpu_sriov_vf(psp->adev)) {
-		/* Write low address of the ring to C2PMSG_102 */
-		psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr);
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, psp_ring_reg);
-		/* Write high address of the ring to C2PMSG_103 */
-		psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr);
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_103, psp_ring_reg);
-
-		/* Write the ring initialization command to C2PMSG_101 */
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101,
-					     GFX_CTRL_CMD_ID_INIT_GPCOM_RING);
-
-		/* there might be handshake issue with hardware which needs delay */
-		mdelay(20);
-
-		/* Wait for response flag (bit 31) in C2PMSG_101 */
-		ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101),
-				   0x80000000, 0x8000FFFF, false);
-
-	} else {
-		/* Write low address of the ring to C2PMSG_69 */
-		psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr);
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, psp_ring_reg);
-		/* Write high address of the ring to C2PMSG_70 */
-		psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr);
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, psp_ring_reg);
-		/* Write size of ring to C2PMSG_71 */
-		psp_ring_reg = ring->ring_size;
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_71, psp_ring_reg);
-		/* Write the ring initialization command to C2PMSG_64 */
-		psp_ring_reg = ring_type;
-		psp_ring_reg = psp_ring_reg << 16;
-		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg);
-
-		/* there might be handshake issue with hardware which needs delay */
-		mdelay(20);
-
-		/* Wait for response flag (bit 31) in C2PMSG_64 */
-		ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
-				   0x80000000, 0x8000FFFF, false);
-	}
+	/* Write low address of the ring to C2PMSG_69 */
+	psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, psp_ring_reg);
+	/* Write high address of the ring to C2PMSG_70 */
+	psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr);
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, psp_ring_reg);
+	/* Write size of ring to C2PMSG_71 */
+	psp_ring_reg = ring->ring_size;
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_71, psp_ring_reg);
+	/* Write the ring initialization command to C2PMSG_64 */
+	psp_ring_reg = ring_type;
+	psp_ring_reg = psp_ring_reg << 16;
+	WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg);
+
+	/* Wait for response flag (bit 31) in C2PMSG_64 */
+	ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64),
+			   0x80000000, 0x8000FFFF, false);
 
 	return ret;
 }
@@ -247,9 +182,6 @@ static int psp_v12_0_ring_stop(struct psp_context *psp,
 		WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64,
 				     GFX_CTRL_CMD_ID_DESTROY_RINGS);
 
-	/* there might be handshake issue with hardware which needs delay */
-	mdelay(20);
-
 	/* Wait for response flag (bit 31) */
 	if (amdgpu_sriov_vf(adev))
 		ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101),
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
index cc621064610f..df612fd9cc50 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
@@ -71,20 +71,13 @@ MODULE_FIRMWARE("amdgpu/psp_14_0_4_ta.bin");
 /* Retry times for vmbx ready wait */
 #define PSP_VMBX_POLLING_LIMIT 3000
 
-/* VBIOS gfl defines */
-#define MBOX_READY_MASK 0x80000000
-#define MBOX_STATUS_MASK 0x0000FFFF
-#define MBOX_COMMAND_MASK 0x00FF0000
-#define MBOX_READY_FLAG 0x80000000
-#define C2PMSG_CMD_SPI_UPDATE_ROM_IMAGE_ADDR_LO 0x2
-#define C2PMSG_CMD_SPI_UPDATE_ROM_IMAGE_ADDR_HI 0x3
-#define C2PMSG_CMD_SPI_UPDATE_FLASH_IMAGE 0x4
-
 /* memory training timeout define */
 #define MEM_TRAIN_SEND_MSG_TIMEOUT_US	3000000
 
 #define regMP1_PUB_SCRATCH0	0x3b10090
 
+#define PSP13_BL_STATUS_SIZE 100
+
 static int psp_v13_0_init_microcode(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -151,6 +144,32 @@ static bool psp_v13_0_is_sos_alive(struct psp_context *psp)
 	return sol_reg != 0x0;
 }
 
+static void psp_v13_0_bootloader_print_status(struct psp_context *psp,
+					      const char *msg)
+{
+	struct amdgpu_device *adev = psp->adev;
+	u32 bl_status_reg;
+	char bl_status_msg[PSP13_BL_STATUS_SIZE];
+	int i, at;
+
+	if (amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 6) ||
+	    amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 12) ||
+	    amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 14)) {
+		at = 0;
+		for_each_inst(i, adev->aid_mask) {
+			bl_status_reg =
+				(SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_92)
+				 << 2) +
+				adev->asic_funcs->encode_ext_smn_addressing(i);
+			at += snprintf(bl_status_msg + at,
+				       PSP13_BL_STATUS_SIZE - at,
+				       " status(%02i): 0x%08x", i,
+				       RREG32_PCIE_EXT(bl_status_reg));
+		}
+		dev_info(adev->dev, "%s - %s", msg, bl_status_msg);
+	}
+}
+
 static int psp_v13_0_wait_for_vmbx_ready(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -196,6 +215,9 @@ static int psp_v13_0_wait_for_bootloader(struct psp_context *psp)
 
 		if (ret == 0)
 			return 0;
+		if (retry_loop && !(retry_loop % 10))
+			psp_v13_0_bootloader_print_status(
+				psp, "Waiting for bootloader completion");
 	}
 
 	return ret;
@@ -610,7 +632,7 @@ static int psp_v13_0_memory_training(struct psp_context *psp, uint32_t ops)
 			}
 
 			memcpy_toio(adev->mman.aper_base_kaddr, buf, sz);
-			adev->hdp.funcs->flush_hdp(adev, NULL);
+			amdgpu_device_flush_hdp(adev, NULL);
 			vfree(buf);
 			drm_dev_exit(idx);
 		} else {
@@ -710,7 +732,8 @@ static int psp_v13_0_exec_spi_cmd(struct psp_context *psp, int cmd)
 	/* Ring the doorbell */
 	WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_73, 1);
 
-	if (cmd == C2PMSG_CMD_SPI_UPDATE_FLASH_IMAGE)
+	if (cmd == C2PMSG_CMD_SPI_UPDATE_FLASH_IMAGE ||
+	    cmd == C2PMSG_CMD_SPI_GET_FLASH_IMAGE)
 		ret = psp_wait_for_spirom_update(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115),
 						 MBOX_READY_FLAG, MBOX_READY_MASK, PSP_SPIROM_UPDATE_TIMEOUT);
 	else
@@ -766,6 +789,37 @@ static int psp_v13_0_update_spirom(struct psp_context *psp,
 	return 0;
 }
 
+static int psp_v13_0_dump_spirom(struct psp_context *psp,
+				 uint64_t fw_pri_mc_addr)
+{
+	struct amdgpu_device *adev = psp->adev;
+	int ret;
+
+	/* Confirm PSP is ready to start */
+	ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115),
+			   MBOX_READY_FLAG, MBOX_READY_MASK, false);
+	if (ret) {
+		dev_err(adev->dev, "PSP Not ready to start processing, ret = %d", ret);
+		return ret;
+	}
+
+	WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_116, lower_32_bits(fw_pri_mc_addr));
+
+	ret = psp_v13_0_exec_spi_cmd(psp, C2PMSG_CMD_SPI_GET_ROM_IMAGE_ADDR_LO);
+	if (ret)
+		return ret;
+
+	WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_116, upper_32_bits(fw_pri_mc_addr));
+
+	ret = psp_v13_0_exec_spi_cmd(psp, C2PMSG_CMD_SPI_GET_ROM_IMAGE_ADDR_HI);
+	if (ret)
+		return ret;
+
+	ret = psp_v13_0_exec_spi_cmd(psp, C2PMSG_CMD_SPI_GET_FLASH_IMAGE);
+
+	return ret;
+}
+
 static int psp_v13_0_vbflash_status(struct psp_context *psp)
 {
 	struct amdgpu_device *adev = psp->adev;
@@ -858,6 +912,25 @@ static bool psp_v13_0_is_reload_needed(struct psp_context *psp)
 	return false;
 }
 
+static int psp_v13_0_reg_program_no_ring(struct psp_context *psp, uint32_t val,
+					 enum psp_reg_prog_id id)
+{
+	struct amdgpu_device *adev = psp->adev;
+	int ret = -EOPNOTSUPP;
+
+	/* PSP will broadcast the value to all instances */
+	if (amdgpu_sriov_vf(adev)) {
+		WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_GBR_IH_SET);
+		WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_102, id);
+		WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_103, val);
+
+		ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101),
+				   0x80000000, 0x80000000, false);
+	}
+
+	return ret;
+}
+
 static const struct psp_funcs psp_v13_0_funcs = {
 	.init_microcode = psp_v13_0_init_microcode,
 	.wait_for_bootloader = psp_v13_0_wait_for_bootloader_steady_state,
@@ -879,11 +952,13 @@ static const struct psp_funcs psp_v13_0_funcs = {
 	.load_usbc_pd_fw = psp_v13_0_load_usbc_pd_fw,
 	.read_usbc_pd_fw = psp_v13_0_read_usbc_pd_fw,
 	.update_spirom = psp_v13_0_update_spirom,
+	.dump_spirom = psp_v13_0_dump_spirom,
 	.vbflash_stat = psp_v13_0_vbflash_status,
 	.fatal_error_recovery_quirk = psp_v13_0_fatal_error_recovery_quirk,
 	.get_ras_capability = psp_v13_0_get_ras_capability,
 	.is_aux_sos_load_required = psp_v13_0_is_aux_sos_load_required,
 	.is_reload_needed = psp_v13_0_is_reload_needed,
+	.reg_program_no_ring = psp_v13_0_reg_program_no_ring,
 };
 
 void psp_v13_0_set_psp_funcs(struct psp_context *psp)
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c
index 7c49c3f3c388..256288c6cd78 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c
@@ -498,7 +498,7 @@ static int psp_v14_0_memory_training(struct psp_context *psp, uint32_t ops)
 			}
 
 			memcpy_toio(adev->mman.aper_base_kaddr, buf, sz);
-			adev->hdp.funcs->flush_hdp(adev, NULL);
+			amdgpu_device_flush_hdp(adev, NULL);
 			vfree(buf);
 			drm_dev_exit(idx);
 		} else {
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 688a720bbbbd..9c169112a5e7 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -106,8 +106,9 @@ static void sdma_v4_4_2_set_buffer_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_set_vm_pte_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_set_irq_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev);
-static void sdma_v4_4_2_set_engine_reset_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_update_reset_mask(struct amdgpu_device *adev);
+static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring);
+static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring);
 
 static u32 sdma_v4_4_2_get_reg_offset(struct amdgpu_device *adev,
 		u32 instance, u32 offset)
@@ -1333,6 +1334,11 @@ static bool sdma_v4_4_2_fw_support_paging_queue(struct amdgpu_device *adev)
 	}
 }
 
+static const struct amdgpu_sdma_funcs sdma_v4_4_2_sdma_funcs = {
+	.stop_kernel_queue = &sdma_v4_4_2_stop_queue,
+	.start_kernel_queue = &sdma_v4_4_2_restore_queue,
+};
+
 static int sdma_v4_4_2_early_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -1351,8 +1357,6 @@ static int sdma_v4_4_2_early_init(struct amdgpu_ip_block *ip_block)
 	sdma_v4_4_2_set_vm_pte_funcs(adev);
 	sdma_v4_4_2_set_irq_funcs(adev);
 	sdma_v4_4_2_set_ras_funcs(adev);
-	sdma_v4_4_2_set_engine_reset_funcs(adev);
-
 	return 0;
 }
 
@@ -1447,6 +1451,7 @@ static int sdma_v4_4_2_sw_init(struct amdgpu_ip_block *ip_block)
 		/* Initialize guilty flags for GFX and PAGE queues */
 		adev->sdma.instance[i].gfx_guilty = false;
 		adev->sdma.instance[i].page_guilty = false;
+		adev->sdma.instance[i].funcs = &sdma_v4_4_2_sdma_funcs;
 
 		ring = &adev->sdma.instance[i].ring;
 		ring->ring_obj = NULL;
@@ -1678,11 +1683,12 @@ static int sdma_v4_4_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 	return r;
 }
 
-static int sdma_v4_4_2_stop_queue(struct amdgpu_device *adev, uint32_t instance_id)
+static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring)
 {
+	struct amdgpu_device *adev = ring->adev;
+	u32 instance_id = GET_INST(SDMA0, ring->me);
 	u32 inst_mask;
 	uint64_t rptr;
-	struct amdgpu_ring *ring = &adev->sdma.instance[instance_id].ring;
 
 	if (amdgpu_sriov_vf(adev))
 		return -EINVAL;
@@ -1715,11 +1721,11 @@ static int sdma_v4_4_2_stop_queue(struct amdgpu_device *adev, uint32_t instance_
 	return 0;
 }
 
-static int sdma_v4_4_2_restore_queue(struct amdgpu_device *adev, uint32_t instance_id)
+static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring)
 {
-	int i;
+	struct amdgpu_device *adev = ring->adev;
 	u32 inst_mask;
-	struct amdgpu_ring *ring = &adev->sdma.instance[instance_id].ring;
+	int i;
 
 	inst_mask = 1 << ring->me;
 	udelay(50);
@@ -1739,16 +1745,6 @@ static int sdma_v4_4_2_restore_queue(struct amdgpu_device *adev, uint32_t instan
 	return sdma_v4_4_2_inst_start(adev, inst_mask, true);
 }
 
-static struct sdma_on_reset_funcs sdma_v4_4_2_engine_reset_funcs = {
-	.pre_reset = sdma_v4_4_2_stop_queue,
-	.post_reset = sdma_v4_4_2_restore_queue,
-};
-
-static void sdma_v4_4_2_set_engine_reset_funcs(struct amdgpu_device *adev)
-{
-	amdgpu_sdma_register_on_reset_callbacks(adev, &sdma_v4_4_2_engine_reset_funcs);
-}
-
 static int sdma_v4_4_2_set_trap_irq_state(struct amdgpu_device *adev,
 					struct amdgpu_irq_src *source,
 					unsigned type,
@@ -2373,7 +2369,9 @@ static void sdma_v4_4_2_update_reset_mask(struct amdgpu_device *adev)
 			adev->sdma.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
 		break;
 	case IP_VERSION(9, 5, 0):
-		/*TODO: enable the queue reset flag until fw supported */
+		if ((adev->gfx.mec_fw_version >= 0xf) && amdgpu_dpm_reset_sdma_is_supported(adev))
+			adev->sdma.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index 0dce59f4f6e2..9505ae96fbec 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -112,6 +112,8 @@ static void sdma_v5_0_set_ring_funcs(struct amdgpu_device *adev);
 static void sdma_v5_0_set_buffer_funcs(struct amdgpu_device *adev);
 static void sdma_v5_0_set_vm_pte_funcs(struct amdgpu_device *adev);
 static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev);
+static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring);
+static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring);
 
 static const struct soc15_reg_golden golden_settings_sdma_5[] = {
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_CHICKEN_BITS, 0xffbf1f0f, 0x03ab0107),
@@ -369,67 +371,36 @@ static uint64_t sdma_v5_0_ring_get_wptr(struct amdgpu_ring *ring)
 static void sdma_v5_0_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	uint32_t *wptr_saved;
-	uint32_t *is_queue_unmap;
-	uint64_t aggregated_db_index;
-	uint32_t mqd_size = adev->mqds[AMDGPU_HW_IP_DMA].mqd_size;
 
 	DRM_DEBUG("Setting write pointer\n");
-	if (ring->is_mes_queue) {
-		wptr_saved = (uint32_t *)(ring->mqd_ptr + mqd_size);
-		is_queue_unmap = (uint32_t *)(ring->mqd_ptr + mqd_size +
-					      sizeof(uint32_t));
-		aggregated_db_index =
-			amdgpu_mes_get_aggregated_doorbell_index(adev,
-			AMDGPU_MES_PRIORITY_LEVEL_NORMAL);
-
+	if (ring->use_doorbell) {
+		DRM_DEBUG("Using doorbell -- "
+			  "wptr_offs == 0x%08x "
+			  "lower_32_bits(ring->wptr) << 2 == 0x%08x "
+			  "upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
+			  ring->wptr_offs,
+			  lower_32_bits(ring->wptr << 2),
+			  upper_32_bits(ring->wptr << 2));
+		/* XXX check if swapping is necessary on BE */
 		atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
 			     ring->wptr << 2);
-		*wptr_saved = ring->wptr << 2;
-		if (*is_queue_unmap) {
-			WDOORBELL64(aggregated_db_index, ring->wptr << 2);
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-					ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		} else {
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-					ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-
-			if (*is_queue_unmap)
-				WDOORBELL64(aggregated_db_index,
-					    ring->wptr << 2);
-		}
+		DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
+			  ring->doorbell_index, ring->wptr << 2);
+		WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
 	} else {
-		if (ring->use_doorbell) {
-			DRM_DEBUG("Using doorbell -- "
-				  "wptr_offs == 0x%08x "
-				  "lower_32_bits(ring->wptr) << 2 == 0x%08x "
-				  "upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
-				  ring->wptr_offs,
-				  lower_32_bits(ring->wptr << 2),
-				  upper_32_bits(ring->wptr << 2));
-			/* XXX check if swapping is necessary on BE */
-			atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
-				     ring->wptr << 2);
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-				  ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		} else {
-			DRM_DEBUG("Not using doorbell -- "
-				  "mmSDMA%i_GFX_RB_WPTR == 0x%08x "
-				  "mmSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
-				  ring->me,
-				  lower_32_bits(ring->wptr << 2),
-				  ring->me,
-				  upper_32_bits(ring->wptr << 2));
-			WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
-					     ring->me, mmSDMA0_GFX_RB_WPTR),
-					lower_32_bits(ring->wptr << 2));
-			WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
-					     ring->me, mmSDMA0_GFX_RB_WPTR_HI),
-					upper_32_bits(ring->wptr << 2));
-		}
+		DRM_DEBUG("Not using doorbell -- "
+			  "mmSDMA%i_GFX_RB_WPTR == 0x%08x "
+			  "mmSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
+			  ring->me,
+			  lower_32_bits(ring->wptr << 2),
+			  ring->me,
+			  upper_32_bits(ring->wptr << 2));
+		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
+							     ring->me, mmSDMA0_GFX_RB_WPTR),
+				lower_32_bits(ring->wptr << 2));
+		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev,
+							     ring->me, mmSDMA0_GFX_RB_WPTR_HI),
+				upper_32_bits(ring->wptr << 2));
 	}
 }
 
@@ -575,11 +546,9 @@ static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
 	}
 
 	if (flags & AMDGPU_FENCE_FLAG_INT) {
-		uint32_t ctx = ring->is_mes_queue ?
-			(ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0;
 		/* generate an interrupt */
 		amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_TRAP));
-		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(ctx));
+		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0));
 	}
 }
 
@@ -588,15 +557,15 @@ static void sdma_v5_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
  * sdma_v5_0_gfx_stop - stop the gfx async dma engines
  *
  * @adev: amdgpu_device pointer
- *
+ * @inst_mask: mask of dma engine instances to be disabled
  * Stop the gfx async dma ring buffers (NAVI10).
  */
-static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev)
+static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev, uint32_t inst_mask)
 {
 	u32 rb_cntl, ib_cntl;
 	int i;
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
 		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
@@ -688,9 +657,11 @@ static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable)
 {
 	u32 f32_cntl;
 	int i;
+	uint32_t inst_mask;
 
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
 	if (!enable) {
-		sdma_v5_0_gfx_stop(adev);
+		sdma_v5_0_gfx_stop(adev, 1 << inst_mask);
 		sdma_v5_0_rlc_stop(adev);
 	}
 
@@ -1046,33 +1017,22 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 	int r;
 	u32 tmp;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
-			return r;
-		}
-
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+		return r;
 	}
 
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
+
 	r = amdgpu_ring_alloc(ring, 20);
 	if (r) {
 		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
-		if (!ring->is_mes_queue)
-			amdgpu_device_wb_free(adev, index);
+		amdgpu_device_wb_free(adev, index);
 		return r;
 	}
 
@@ -1085,10 +1045,7 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (ring->is_mes_queue)
-			tmp = le32_to_cpu(*cpu_ptr);
-		else
-			tmp = le32_to_cpu(adev->wb.wb[index]);
+		tmp = le32_to_cpu(adev->wb.wb[index]);
 		if (tmp == 0xDEADBEEF)
 			break;
 		if (amdgpu_emu_mode == 1)
@@ -1100,8 +1057,7 @@ static int sdma_v5_0_ring_test_ring(struct amdgpu_ring *ring)
 	if (i >= adev->usec_timeout)
 		r = -ETIMEDOUT;
 
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 
 	return r;
 }
@@ -1124,38 +1080,24 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	long r;
 	u32 tmp = 0;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = (void *)amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+		return r;
+	}
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
 
-		r = amdgpu_ib_get(adev, NULL, 256,
-					AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
-			goto err0;
-		}
+	r = amdgpu_ib_get(adev, NULL, 256,
+			  AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+		goto err0;
 	}
 
 	ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
@@ -1183,10 +1125,7 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		goto err1;
 	}
 
-	if (ring->is_mes_queue)
-		tmp = le32_to_cpu(*cpu_ptr);
-	else
-		tmp = le32_to_cpu(adev->wb.wb[index]);
+	tmp = le32_to_cpu(adev->wb.wb[index]);
 
 	if (tmp == 0xDEADBEEF)
 		r = 0;
@@ -1197,8 +1136,7 @@ err1:
 	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err0:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -1389,6 +1327,36 @@ static void sdma_v5_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
 	amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask);
 }
 
+static int sdma_v5_0_soft_reset_engine(struct amdgpu_device *adev, u32 instance_id)
+{
+	u32 grbm_soft_reset;
+	u32 tmp;
+
+	grbm_soft_reset = REG_SET_FIELD(0,
+					GRBM_SOFT_RESET, SOFT_RESET_SDMA0,
+					1);
+	grbm_soft_reset <<= instance_id;
+
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	tmp |= grbm_soft_reset;
+	DRM_DEBUG("GRBM_SOFT_RESET=0x%08X\n", tmp);
+	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+
+	udelay(50);
+
+	tmp &= ~grbm_soft_reset;
+	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	return 0;
+}
+
+static const struct amdgpu_sdma_funcs sdma_v5_0_sdma_funcs = {
+	.stop_kernel_queue = &sdma_v5_0_stop_queue,
+	.start_kernel_queue = &sdma_v5_0_restore_queue,
+	.soft_reset_kernel_queue = &sdma_v5_0_soft_reset_engine,
+};
+
 static int sdma_v5_0_early_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -1431,6 +1399,7 @@ static int sdma_v5_0_sw_init(struct amdgpu_ip_block *ip_block)
 		return r;
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
+		adev->sdma.instance[i].funcs = &sdma_v5_0_sdma_funcs;
 		ring = &adev->sdma.instance[i].ring;
 		ring->ring_obj = NULL;
 		ring->use_doorbell = true;
@@ -1572,32 +1541,25 @@ static int sdma_v5_0_soft_reset(struct amdgpu_ip_block *ip_block)
 static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
-	int i, j, r;
-	u32 rb_cntl, ib_cntl, f32_cntl, freeze, cntl, preempt, soft_reset, stat1_reg;
+	u32 inst_id = ring->me;
 
-	if (amdgpu_sriov_vf(adev))
-		return -EINVAL;
+	return amdgpu_sdma_reset_engine(adev, inst_id);
+}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		if (ring == &adev->sdma.instance[i].ring)
-			break;
-	}
+static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring)
+{
+	u32 f32_cntl, freeze, cntl, stat1_reg;
+	struct amdgpu_device *adev = ring->adev;
+	int i, j, r = 0;
 
-	if (i == adev->sdma.num_instances) {
-		DRM_ERROR("sdma instance not found\n");
+	if (amdgpu_sriov_vf(adev))
 		return -EINVAL;
-	}
 
+	i = ring->me;
 	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	/* stop queue */
-	ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
-	ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
-
-	rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
-	rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
+	sdma_v5_0_gfx_stop(adev, 1 << i);
 
 	/* engine stop SDMA1_F32_CNTL.HALT to 1 and SDMAx_FREEZE freeze bit to 1 */
 	freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE));
@@ -1628,30 +1590,25 @@ static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 	cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL));
 	cntl = REG_SET_FIELD(cntl, SDMA0_CNTL, UTC_L1_ENABLE, 0);
 	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), cntl);
+err0:
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
+	return r;
+}
 
-	/* soft reset SDMA_GFX_PREEMPT.IB_PREEMPT = 0 mmGRBM_SOFT_RESET.SOFT_RESET_SDMA0/1 = 1 */
-	preempt = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT));
-	preempt = REG_SET_FIELD(preempt, SDMA0_GFX_PREEMPT, IB_PREEMPT, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT), preempt);
-
-	soft_reset = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-	soft_reset |= 1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i;
-
-	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset);
-
-	udelay(50);
-
-	soft_reset &= ~(1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i);
-	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset);
+static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	u32 inst_id = ring->me;
+	u32 freeze;
+	int r;
 
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	/* unfreeze*/
-	freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE));
+	freeze = RREG32(sdma_v5_0_get_reg_offset(adev, inst_id, mmSDMA0_FREEZE));
 	freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 0);
-	WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze);
+	WREG32(sdma_v5_0_get_reg_offset(adev, inst_id, mmSDMA0_FREEZE), freeze);
 
-	r = sdma_v5_0_gfx_resume_instance(adev, i, true);
-
-err0:
+	r = sdma_v5_0_gfx_resume_instance(adev, inst_id, true);
 	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
index 2b39a03ff0c1..a6e612b4a892 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
@@ -113,6 +113,8 @@ static void sdma_v5_2_set_ring_funcs(struct amdgpu_device *adev);
 static void sdma_v5_2_set_buffer_funcs(struct amdgpu_device *adev);
 static void sdma_v5_2_set_vm_pte_funcs(struct amdgpu_device *adev);
 static void sdma_v5_2_set_irq_funcs(struct amdgpu_device *adev);
+static int sdma_v5_2_stop_queue(struct amdgpu_ring *ring);
+static int sdma_v5_2_restore_queue(struct amdgpu_ring *ring);
 
 static u32 sdma_v5_2_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 internal_offset)
 {
@@ -394,11 +396,9 @@ static void sdma_v5_2_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
 	}
 
 	if ((flags & AMDGPU_FENCE_FLAG_INT)) {
-		uint32_t ctx = ring->is_mes_queue ?
-			(ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0;
 		/* generate an interrupt */
 		amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_TRAP));
-		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(ctx));
+		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0));
 	}
 }
 
@@ -407,15 +407,15 @@ static void sdma_v5_2_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
  * sdma_v5_2_gfx_stop - stop the gfx async dma engines
  *
  * @adev: amdgpu_device pointer
- *
+ * @inst_mask: mask of dma engine instances to be disabled
  * Stop the gfx async dma ring buffers.
  */
-static void sdma_v5_2_gfx_stop(struct amdgpu_device *adev)
+static void sdma_v5_2_gfx_stop(struct amdgpu_device *adev,  uint32_t inst_mask)
 {
 	u32 rb_cntl, ib_cntl;
 	int i;
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
 		WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
@@ -506,9 +506,11 @@ static void sdma_v5_2_enable(struct amdgpu_device *adev, bool enable)
 {
 	u32 f32_cntl;
 	int i;
+	uint32_t inst_mask;
 
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
 	if (!enable) {
-		sdma_v5_2_gfx_stop(adev);
+		sdma_v5_2_gfx_stop(adev, inst_mask);
 		sdma_v5_2_rlc_stop(adev);
 	}
 
@@ -761,37 +763,49 @@ static int sdma_v5_2_load_microcode(struct amdgpu_device *adev)
 	return 0;
 }
 
-static int sdma_v5_2_soft_reset(struct amdgpu_ip_block *ip_block)
+static int sdma_v5_2_soft_reset_engine(struct amdgpu_device *adev, u32 instance_id)
 {
-	struct amdgpu_device *adev = ip_block->adev;
 	u32 grbm_soft_reset;
 	u32 tmp;
-	int i;
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		grbm_soft_reset = REG_SET_FIELD(0,
-						GRBM_SOFT_RESET, SOFT_RESET_SDMA0,
-						1);
-		grbm_soft_reset <<= i;
+	grbm_soft_reset = REG_SET_FIELD(0,
+					GRBM_SOFT_RESET, SOFT_RESET_SDMA0,
+					1);
+	grbm_soft_reset <<= instance_id;
 
-		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-		tmp |= grbm_soft_reset;
-		DRM_DEBUG("GRBM_SOFT_RESET=0x%08X\n", tmp);
-		WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
-		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	tmp |= grbm_soft_reset;
+	DRM_DEBUG("GRBM_SOFT_RESET=0x%08X\n", tmp);
+	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
 
-		udelay(50);
+	udelay(50);
+
+	tmp &= ~grbm_soft_reset;
+	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+	tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+	return 0;
+}
 
-		tmp &= ~grbm_soft_reset;
-		WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
-		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+static int sdma_v5_2_soft_reset(struct amdgpu_ip_block *ip_block)
+{
+	struct amdgpu_device *adev = ip_block->adev;
+	int i;
 
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		sdma_v5_2_soft_reset_engine(adev, i);
 		udelay(50);
 	}
 
 	return 0;
 }
 
+static const struct amdgpu_sdma_funcs sdma_v5_2_sdma_funcs = {
+	.stop_kernel_queue = &sdma_v5_2_stop_queue,
+	.start_kernel_queue = &sdma_v5_2_restore_queue,
+	.soft_reset_kernel_queue = &sdma_v5_2_soft_reset_engine,
+};
+
 /**
  * sdma_v5_2_start - setup and start the async dma engines
  *
@@ -903,33 +917,22 @@ static int sdma_v5_2_ring_test_ring(struct amdgpu_ring *ring)
 	int r;
 	u32 tmp;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
-			return r;
-		}
-
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+		return r;
 	}
 
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
+
 	r = amdgpu_ring_alloc(ring, 20);
 	if (r) {
 		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
-		if (!ring->is_mes_queue)
-			amdgpu_device_wb_free(adev, index);
+		amdgpu_device_wb_free(adev, index);
 		return r;
 	}
 
@@ -942,10 +945,7 @@ static int sdma_v5_2_ring_test_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (ring->is_mes_queue)
-			tmp = le32_to_cpu(*cpu_ptr);
-		else
-			tmp = le32_to_cpu(adev->wb.wb[index]);
+		tmp = le32_to_cpu(adev->wb.wb[index]);
 		if (tmp == 0xDEADBEEF)
 			break;
 		if (amdgpu_emu_mode == 1)
@@ -957,8 +957,7 @@ static int sdma_v5_2_ring_test_ring(struct amdgpu_ring *ring)
 	if (i >= adev->usec_timeout)
 		r = -ETIMEDOUT;
 
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 
 	return r;
 }
@@ -981,37 +980,23 @@ static int sdma_v5_2_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	long r;
 	u32 tmp = 0;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = (void *)amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+		return r;
+	}
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
 
-		r = amdgpu_ib_get(adev, NULL, 256, AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
-			goto err0;
-		}
+	r = amdgpu_ib_get(adev, NULL, 256, AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+		goto err0;
 	}
 
 	ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
@@ -1039,10 +1024,7 @@ static int sdma_v5_2_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		goto err1;
 	}
 
-	if (ring->is_mes_queue)
-		tmp = le32_to_cpu(*cpu_ptr);
-	else
-		tmp = le32_to_cpu(adev->wb.wb[index]);
+	tmp = le32_to_cpu(adev->wb.wb[index]);
 
 	if (tmp == 0xDEADBEEF)
 		r = 0;
@@ -1053,8 +1035,7 @@ err1:
 	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err0:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -1337,6 +1318,7 @@ static int sdma_v5_2_sw_init(struct amdgpu_ip_block *ip_block)
 	}
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
+		adev->sdma.instance[i].funcs = &sdma_v5_2_sdma_funcs;
 		ring = &adev->sdma.instance[i].ring;
 		ring->ring_obj = NULL;
 		ring->use_doorbell = true;
@@ -1472,32 +1454,25 @@ static int sdma_v5_2_wait_for_idle(struct amdgpu_ip_block *ip_block)
 static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
-	int i, j, r;
-	u32 rb_cntl, ib_cntl, f32_cntl, freeze, cntl, preempt, soft_reset, stat1_reg;
+	u32 inst_id = ring->me;
 
-	if (amdgpu_sriov_vf(adev))
-		return -EINVAL;
+	return amdgpu_sdma_reset_engine(adev, inst_id);
+}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		if (ring == &adev->sdma.instance[i].ring)
-			break;
-	}
+static int sdma_v5_2_stop_queue(struct amdgpu_ring *ring)
+{
+	u32 f32_cntl, freeze, cntl, stat1_reg;
+	struct amdgpu_device *adev = ring->adev;
+	int i, j, r = 0;
 
-	if (i == adev->sdma.num_instances) {
-		DRM_ERROR("sdma instance not found\n");
+	if (amdgpu_sriov_vf(adev))
 		return -EINVAL;
-	}
 
+	i = ring->me;
 	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	/* stop queue */
-	ib_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
-	ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
-	WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
-
-	rb_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
-	rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
-	WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
+	sdma_v5_2_gfx_stop(adev, 1 << i);
 
 	/*engine stop SDMA1_F32_CNTL.HALT to 1 and SDMAx_FREEZE freeze bit to 1 */
 	freeze = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE));
@@ -1530,31 +1505,26 @@ static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 	cntl = REG_SET_FIELD(cntl, SDMA0_CNTL, UTC_L1_ENABLE, 0);
 	WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), cntl);
 
-	/* soft reset SDMA_GFX_PREEMPT.IB_PREEMPT = 0 mmGRBM_SOFT_RESET.SOFT_RESET_SDMA0/1 = 1 */
-	preempt = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT));
-	preempt = REG_SET_FIELD(preempt, SDMA0_GFX_PREEMPT, IB_PREEMPT, 0);
-	WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT), preempt);
-
-	soft_reset = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-	soft_reset |= 1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i;
-
-
-	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset);
-
-	udelay(50);
-
-	soft_reset &= ~(1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i);
+err0:
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
+	return r;
+}
 
-	WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset);
+static int sdma_v5_2_restore_queue(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	u32 inst_id = ring->me;
+	u32 freeze;
+	int r;
 
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	/* unfreeze and unhalt */
-	freeze = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE));
+	freeze = RREG32(sdma_v5_2_get_reg_offset(adev, inst_id, mmSDMA0_FREEZE));
 	freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 0);
-	WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze);
+	WREG32(sdma_v5_2_get_reg_offset(adev, inst_id, mmSDMA0_FREEZE), freeze);
 
-	r = sdma_v5_2_gfx_resume_instance(adev, i, true);
+	r = sdma_v5_2_gfx_resume_instance(adev, inst_id, true);
 
-err0:
 	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index c214c3d2149b..da5b5d64f137 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -43,6 +43,7 @@
 #include "sdma_common.h"
 #include "sdma_v6_0.h"
 #include "v11_structs.h"
+#include "mes_userqueue.h"
 
 MODULE_FIRMWARE("amdgpu/sdma_6_0_0.bin");
 MODULE_FIRMWARE("amdgpu/sdma_6_0_1.bin");
@@ -376,11 +377,9 @@ static void sdma_v6_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
 	}
 
 	if (flags & AMDGPU_FENCE_FLAG_INT) {
-		uint32_t ctx = ring->is_mes_queue ?
-			(ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0;
 		/* generate an interrupt */
 		amdgpu_ring_write(ring, SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_TRAP));
-		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(ctx));
+		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0));
 	}
 }
 
@@ -891,6 +890,9 @@ static int sdma_v6_0_mqd_init(struct amdgpu_device *adev, void *mqd,
 	m->sdmax_rlcx_rb_aql_cntl = regSDMA0_QUEUE0_RB_AQL_CNTL_DEFAULT;
 	m->sdmax_rlcx_dummy_reg = regSDMA0_QUEUE0_DUMMY_REG_DEFAULT;
 
+	m->sdmax_rlcx_csa_addr_lo = lower_32_bits(prop->csa_addr);
+	m->sdmax_rlcx_csa_addr_hi = upper_32_bits(prop->csa_addr);
+
 	return 0;
 }
 
@@ -917,33 +919,22 @@ static int sdma_v6_0_ring_test_ring(struct amdgpu_ring *ring)
 	int r;
 	u32 tmp;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
-			return r;
-		}
-
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+		return r;
 	}
 
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
+
 	r = amdgpu_ring_alloc(ring, 5);
 	if (r) {
 		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
-		if (!ring->is_mes_queue)
-			amdgpu_device_wb_free(adev, index);
+		amdgpu_device_wb_free(adev, index);
 		return r;
 	}
 
@@ -956,10 +947,7 @@ static int sdma_v6_0_ring_test_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (ring->is_mes_queue)
-			tmp = le32_to_cpu(*cpu_ptr);
-		else
-			tmp = le32_to_cpu(adev->wb.wb[index]);
+		tmp = le32_to_cpu(adev->wb.wb[index]);
 		if (tmp == 0xDEADBEEF)
 			break;
 		if (amdgpu_emu_mode == 1)
@@ -971,8 +959,7 @@ static int sdma_v6_0_ring_test_ring(struct amdgpu_ring *ring)
 	if (i >= adev->usec_timeout)
 		r = -ETIMEDOUT;
 
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 
 	return r;
 }
@@ -995,37 +982,23 @@ static int sdma_v6_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	long r;
 	u32 tmp = 0;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = (void *)amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+		return r;
+	}
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
 
-		r = amdgpu_ib_get(adev, NULL, 256, AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
-			goto err0;
-		}
+	r = amdgpu_ib_get(adev, NULL, 256, AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+		goto err0;
 	}
 
 	ib.ptr[0] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_WRITE) |
@@ -1053,10 +1026,7 @@ static int sdma_v6_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		goto err1;
 	}
 
-	if (ring->is_mes_queue)
-		tmp = le32_to_cpu(*cpu_ptr);
-	else
-		tmp = le32_to_cpu(adev->wb.wb[index]);
+	tmp = le32_to_cpu(adev->wb.wb[index]);
 
 	if (tmp == 0xDEADBEEF)
 		r = 0;
@@ -1067,8 +1037,7 @@ err1:
 	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err0:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -1300,6 +1269,23 @@ static int sdma_v6_0_early_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	int r;
 
+	switch (amdgpu_user_queue) {
+	case -1:
+	case 0:
+	default:
+		adev->sdma.no_user_submission = false;
+		adev->sdma.disable_uq = true;
+		break;
+	case 1:
+		adev->sdma.no_user_submission = false;
+		adev->sdma.disable_uq = false;
+		break;
+	case 2:
+		adev->sdma.no_user_submission = true;
+		adev->sdma.disable_uq = false;
+		break;
+	}
+
 	r = amdgpu_sdma_init_microcode(adev, 0, true);
 	if (r)
 		return r;
@@ -1334,6 +1320,7 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 		ring->ring_obj = NULL;
 		ring->use_doorbell = true;
 		ring->me = i;
+		ring->no_user_submission = adev->sdma.no_user_submission;
 
 		DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i,
 				ring->use_doorbell?"true":"false");
@@ -1376,6 +1363,10 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 	else
 		DRM_ERROR("Failed to allocated memory for SDMA IP Dump\n");
 
+	/* add firmware version checks here */
+	if (0 && !adev->sdma.disable_uq)
+		adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
+
 	r = amdgpu_sdma_sysfs_reset_mask_init(adev);
 	if (r)
 		return r;
@@ -1399,11 +1390,39 @@ static int sdma_v6_0_sw_fini(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
+static int sdma_v6_0_set_userq_trap_interrupts(struct amdgpu_device *adev,
+					       bool enable)
+{
+	unsigned int irq_type;
+	int i, r;
+
+	if (adev->userq_funcs[AMDGPU_HW_IP_DMA]) {
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			irq_type = AMDGPU_SDMA_IRQ_INSTANCE0 + i;
+			if (enable)
+				r = amdgpu_irq_get(adev, &adev->sdma.trap_irq,
+						   irq_type);
+			else
+				r = amdgpu_irq_put(adev, &adev->sdma.trap_irq,
+						   irq_type);
+			if (r)
+				return r;
+		}
+	}
+
+	return 0;
+}
+
 static int sdma_v6_0_hw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
+	int r;
 
-	return sdma_v6_0_start(adev);
+	r = sdma_v6_0_start(adev);
+	if (r)
+		return r;
+
+	return sdma_v6_0_set_userq_trap_interrupts(adev, true);
 }
 
 static int sdma_v6_0_hw_fini(struct amdgpu_ip_block *ip_block)
@@ -1415,6 +1434,7 @@ static int sdma_v6_0_hw_fini(struct amdgpu_ip_block *ip_block)
 
 	sdma_v6_0_ctxempty_int_enable(adev, false);
 	sdma_v6_0_enable(adev, false);
+	sdma_v6_0_set_userq_trap_interrupts(adev, false);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
index b2706221df99..befe013b11a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
@@ -42,6 +42,7 @@
 #include "sdma_common.h"
 #include "sdma_v7_0.h"
 #include "v12_structs.h"
+#include "mes_userqueue.h"
 
 MODULE_FIRMWARE("amdgpu/sdma_7_0_0.bin");
 MODULE_FIRMWARE("amdgpu/sdma_7_0_1.bin");
@@ -204,66 +205,39 @@ static uint64_t sdma_v7_0_ring_get_wptr(struct amdgpu_ring *ring)
 static void sdma_v7_0_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	uint32_t *wptr_saved;
-	uint32_t *is_queue_unmap;
-	uint64_t aggregated_db_index;
-	uint32_t mqd_size = adev->mqds[AMDGPU_HW_IP_DMA].mqd_size;
 
 	DRM_DEBUG("Setting write pointer\n");
 
-	if (ring->is_mes_queue) {
-		wptr_saved = (uint32_t *)(ring->mqd_ptr + mqd_size);
-		is_queue_unmap = (uint32_t *)(ring->mqd_ptr + mqd_size +
-					      sizeof(uint32_t));
-		aggregated_db_index =
-			amdgpu_mes_get_aggregated_doorbell_index(adev,
-							 ring->hw_prio);
-
+	if (ring->use_doorbell) {
+		DRM_DEBUG("Using doorbell -- "
+			  "wptr_offs == 0x%08x "
+			  "lower_32_bits(ring->wptr) << 2 == 0x%08x "
+			  "upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
+			  ring->wptr_offs,
+			  lower_32_bits(ring->wptr << 2),
+			  upper_32_bits(ring->wptr << 2));
+		/* XXX check if swapping is necessary on BE */
 		atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
 			     ring->wptr << 2);
-		*wptr_saved = ring->wptr << 2;
-		if (*is_queue_unmap) {
-			WDOORBELL64(aggregated_db_index, ring->wptr << 2);
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-					ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		} else {
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-					ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		}
+		DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
+			  ring->doorbell_index, ring->wptr << 2);
+		WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
 	} else {
-		if (ring->use_doorbell) {
-			DRM_DEBUG("Using doorbell -- "
-				  "wptr_offs == 0x%08x "
-				  "lower_32_bits(ring->wptr) << 2 == 0x%08x "
-				  "upper_32_bits(ring->wptr) << 2 == 0x%08x\n",
-				  ring->wptr_offs,
-				  lower_32_bits(ring->wptr << 2),
-				  upper_32_bits(ring->wptr << 2));
-			/* XXX check if swapping is necessary on BE */
-			atomic64_set((atomic64_t *)ring->wptr_cpu_addr,
-				     ring->wptr << 2);
-			DRM_DEBUG("calling WDOORBELL64(0x%08x, 0x%016llx)\n",
-				  ring->doorbell_index, ring->wptr << 2);
-			WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
-		} else {
-			DRM_DEBUG("Not using doorbell -- "
-				  "regSDMA%i_GFX_RB_WPTR == 0x%08x "
-				  "regSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
-				  ring->me,
-				  lower_32_bits(ring->wptr << 2),
-				  ring->me,
-				  upper_32_bits(ring->wptr << 2));
-			WREG32_SOC15_IP(GC, sdma_v7_0_get_reg_offset(adev,
-								     ring->me,
-								     regSDMA0_QUEUE0_RB_WPTR),
-					lower_32_bits(ring->wptr << 2));
-			WREG32_SOC15_IP(GC, sdma_v7_0_get_reg_offset(adev,
-								     ring->me,
-								     regSDMA0_QUEUE0_RB_WPTR_HI),
-					upper_32_bits(ring->wptr << 2));
-		}
+		DRM_DEBUG("Not using doorbell -- "
+			  "regSDMA%i_GFX_RB_WPTR == 0x%08x "
+			  "regSDMA%i_GFX_RB_WPTR_HI == 0x%08x\n",
+			  ring->me,
+			  lower_32_bits(ring->wptr << 2),
+			  ring->me,
+			  upper_32_bits(ring->wptr << 2));
+		WREG32_SOC15_IP(GC, sdma_v7_0_get_reg_offset(adev,
+							     ring->me,
+							     regSDMA0_QUEUE0_RB_WPTR),
+				lower_32_bits(ring->wptr << 2));
+		WREG32_SOC15_IP(GC, sdma_v7_0_get_reg_offset(adev,
+							     ring->me,
+							     regSDMA0_QUEUE0_RB_WPTR_HI),
+				upper_32_bits(ring->wptr << 2));
 	}
 }
 
@@ -407,11 +381,9 @@ static void sdma_v7_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se
 	}
 
 	if (flags & AMDGPU_FENCE_FLAG_INT) {
-		uint32_t ctx = ring->is_mes_queue ?
-			(ring->hw_queue_id | AMDGPU_FENCE_MES_QUEUE_FLAG) : 0;
 		/* generate an interrupt */
 		amdgpu_ring_write(ring, SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_TRAP));
-		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(ctx));
+		amdgpu_ring_write(ring, SDMA_PKT_TRAP_INT_CONTEXT_INT_CONTEXT(0));
 	}
 }
 
@@ -935,6 +907,9 @@ static int sdma_v7_0_mqd_init(struct amdgpu_device *adev, void *mqd,
 	m->sdmax_rlcx_rb_aql_cntl = 0x4000;	//regSDMA0_QUEUE0_RB_AQL_CNTL_DEFAULT;
 	m->sdmax_rlcx_dummy_reg = 0xf;	//regSDMA0_QUEUE0_DUMMY_REG_DEFAULT;
 
+	m->sdmax_rlcx_csa_addr_lo = lower_32_bits(prop->csa_addr);
+	m->sdmax_rlcx_csa_addr_hi = upper_32_bits(prop->csa_addr);
+
 	return 0;
 }
 
@@ -961,33 +936,22 @@ static int sdma_v7_0_ring_test_ring(struct amdgpu_ring *ring)
 	int r;
 	u32 tmp;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
-			return r;
-		}
-
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+		return r;
 	}
 
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
+
 	r = amdgpu_ring_alloc(ring, 5);
 	if (r) {
 		DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
-		if (!ring->is_mes_queue)
-			amdgpu_device_wb_free(adev, index);
+		amdgpu_device_wb_free(adev, index);
 		return r;
 	}
 
@@ -1000,10 +964,7 @@ static int sdma_v7_0_ring_test_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (ring->is_mes_queue)
-			tmp = le32_to_cpu(*cpu_ptr);
-		else
-			tmp = le32_to_cpu(adev->wb.wb[index]);
+		tmp = le32_to_cpu(adev->wb.wb[index]);
 		if (tmp == 0xDEADBEEF)
 			break;
 		if (amdgpu_emu_mode == 1)
@@ -1015,8 +976,7 @@ static int sdma_v7_0_ring_test_ring(struct amdgpu_ring *ring)
 	if (i >= adev->usec_timeout)
 		r = -ETIMEDOUT;
 
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 
 	return r;
 }
@@ -1039,37 +999,23 @@ static int sdma_v7_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	long r;
 	u32 tmp = 0;
 	u64 gpu_addr;
-	volatile uint32_t *cpu_ptr = NULL;
 
 	tmp = 0xCAFEDEAD;
 	memset(&ib, 0, sizeof(ib));
 
-	if (ring->is_mes_queue) {
-		uint32_t offset = 0;
-		offset = amdgpu_mes_ctx_get_offs(ring, AMDGPU_MES_CTX_IB_OFFS);
-		ib.gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		ib.ptr = (void *)amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-
-		offset = amdgpu_mes_ctx_get_offs(ring,
-					 AMDGPU_MES_CTX_PADDING_OFFS);
-		gpu_addr = amdgpu_mes_ctx_get_offs_gpu_addr(ring, offset);
-		cpu_ptr = amdgpu_mes_ctx_get_offs_cpu_addr(ring, offset);
-		*cpu_ptr = tmp;
-	} else {
-		r = amdgpu_device_wb_get(adev, &index);
-		if (r) {
-			dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
-			return r;
-		}
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r) {
+		dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+		return r;
+	}
 
-		gpu_addr = adev->wb.gpu_addr + (index * 4);
-		adev->wb.wb[index] = cpu_to_le32(tmp);
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(tmp);
 
-		r = amdgpu_ib_get(adev, NULL, 256, AMDGPU_IB_POOL_DIRECT, &ib);
-		if (r) {
-			DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
-			goto err0;
-		}
+	r = amdgpu_ib_get(adev, NULL, 256, AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r) {
+		DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+		goto err0;
 	}
 
 	ib.ptr[0] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_WRITE) |
@@ -1097,10 +1043,7 @@ static int sdma_v7_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 		goto err1;
 	}
 
-	if (ring->is_mes_queue)
-		tmp = le32_to_cpu(*cpu_ptr);
-	else
-		tmp = le32_to_cpu(adev->wb.wb[index]);
+	tmp = le32_to_cpu(adev->wb.wb[index]);
 
 	if (tmp == 0xDEADBEEF)
 		r = 0;
@@ -1111,8 +1054,7 @@ err1:
 	amdgpu_ib_free(&ib, NULL);
 	dma_fence_put(f);
 err0:
-	if (!ring->is_mes_queue)
-		amdgpu_device_wb_free(adev, index);
+	amdgpu_device_wb_free(adev, index);
 	return r;
 }
 
@@ -1312,6 +1254,23 @@ static int sdma_v7_0_early_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	int r;
 
+	switch (amdgpu_user_queue) {
+	case -1:
+	case 0:
+	default:
+		adev->sdma.no_user_submission = false;
+		adev->sdma.disable_uq = true;
+		break;
+	case 1:
+		adev->sdma.no_user_submission = false;
+		adev->sdma.disable_uq = false;
+		break;
+	case 2:
+		adev->sdma.no_user_submission = true;
+		adev->sdma.disable_uq = false;
+		break;
+	}
+
 	r = amdgpu_sdma_init_microcode(adev, 0, true);
 	if (r) {
 		DRM_ERROR("Failed to init sdma firmware!\n");
@@ -1347,6 +1306,7 @@ static int sdma_v7_0_sw_init(struct amdgpu_ip_block *ip_block)
 		ring->ring_obj = NULL;
 		ring->use_doorbell = true;
 		ring->me = i;
+		ring->no_user_submission = adev->sdma.no_user_submission;
 
 		DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i,
 				ring->use_doorbell?"true":"false");
@@ -1378,6 +1338,10 @@ static int sdma_v7_0_sw_init(struct amdgpu_ip_block *ip_block)
 	else
 		DRM_ERROR("Failed to allocated memory for SDMA IP Dump\n");
 
+	/* add firmware version checks here */
+	if (0 && !adev->sdma.disable_uq)
+		adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
+
 	return r;
 }
 
@@ -1400,11 +1364,39 @@ static int sdma_v7_0_sw_fini(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
+static int sdma_v7_0_set_userq_trap_interrupts(struct amdgpu_device *adev,
+					       bool enable)
+{
+	unsigned int irq_type;
+	int i, r;
+
+	if (adev->userq_funcs[AMDGPU_HW_IP_DMA]) {
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			irq_type = AMDGPU_SDMA_IRQ_INSTANCE0 + i;
+			if (enable)
+				r = amdgpu_irq_get(adev, &adev->sdma.trap_irq,
+						   irq_type);
+			else
+				r = amdgpu_irq_put(adev, &adev->sdma.trap_irq,
+						   irq_type);
+			if (r)
+				return r;
+		}
+	}
+
+	return 0;
+}
+
 static int sdma_v7_0_hw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
+	int r;
 
-	return sdma_v7_0_start(adev);
+	r = sdma_v7_0_start(adev);
+	if (r)
+		return r;
+
+	return sdma_v7_0_set_userq_trap_interrupts(adev, true);
 }
 
 static int sdma_v7_0_hw_fini(struct amdgpu_ip_block *ip_block)
@@ -1416,6 +1408,7 @@ static int sdma_v7_0_hw_fini(struct amdgpu_ip_block *ip_block)
 
 	sdma_v7_0_ctx_switch_enable(adev, false);
 	sdma_v7_0_enable(adev, false);
+	sdma_v7_0_set_userq_trap_interrupts(adev, false);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c
index 2247f6a94858..e0f139de7991 100644
--- a/drivers/gpu/drm/amd/amdgpu/si.c
+++ b/drivers/gpu/drm/amd/amdgpu/si.c
@@ -35,6 +35,7 @@
 #include "amdgpu_vce.h"
 #include "atom.h"
 #include "amd_pcie.h"
+
 #include "si_dpm.h"
 #include "sid.h"
 #include "si_ih.h"
@@ -44,17 +45,30 @@
 #include "dce_v6_0.h"
 #include "si.h"
 #include "uvd_v3_1.h"
-#include "amdgpu_vkms.h"
+
+#include "uvd/uvd_4_0_d.h"
+
+#include "smu/smu_6_0_d.h"
+#include "smu/smu_6_0_sh_mask.h"
+
 #include "gca/gfx_6_0_d.h"
+#include "gca/gfx_6_0_sh_mask.h"
+
 #include "oss/oss_1_0_d.h"
 #include "oss/oss_1_0_sh_mask.h"
+
 #include "gmc/gmc_6_0_d.h"
+#include"gmc/gmc_6_0_sh_mask.h"
+
 #include "dce/dce_6_0_d.h"
-#include "uvd/uvd_4_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
 #include "bif/bif_3_0_d.h"
 #include "bif/bif_3_0_sh_mask.h"
+#include "si_enums.h"
 
 #include "amdgpu_dm.h"
+#include "amdgpu_vkms.h"
 
 static const u32 tahiti_golden_registers[] =
 {
@@ -1071,8 +1085,8 @@ static u32 si_smc_rreg(struct amdgpu_device *adev, u32 reg)
 	u32 r;
 
 	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	WREG32(SMC_IND_INDEX_0, (reg));
-	r = RREG32(SMC_IND_DATA_0);
+	WREG32(mmSMC_IND_INDEX_0, (reg));
+	r = RREG32(mmSMC_IND_DATA_0);
 	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
 	return r;
 }
@@ -1082,8 +1096,8 @@ static void si_smc_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
 	unsigned long flags;
 
 	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	WREG32(SMC_IND_INDEX_0, (reg));
-	WREG32(SMC_IND_DATA_0, (v));
+	WREG32(mmSMC_IND_INDEX_0, (reg));
+	WREG32(mmSMC_IND_DATA_0, (v));
 	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
 }
 
@@ -1110,20 +1124,20 @@ static void si_uvd_ctx_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
 }
 
 static struct amdgpu_allowed_register_entry si_allowed_read_registers[] = {
-	{GRBM_STATUS},
+	{mmGRBM_STATUS},
 	{mmGRBM_STATUS2},
 	{mmGRBM_STATUS_SE0},
 	{mmGRBM_STATUS_SE1},
 	{mmSRBM_STATUS},
 	{mmSRBM_STATUS2},
-	{DMA_STATUS_REG + DMA0_REGISTER_OFFSET},
-	{DMA_STATUS_REG + DMA1_REGISTER_OFFSET},
+	{mmDMA_STATUS_REG + DMA0_REGISTER_OFFSET},
+	{mmDMA_STATUS_REG + DMA1_REGISTER_OFFSET},
 	{mmCP_STAT},
 	{mmCP_STALLED_STAT1},
 	{mmCP_STALLED_STAT2},
 	{mmCP_STALLED_STAT3},
-	{GB_ADDR_CONFIG},
-	{MC_ARB_RAMCFG},
+	{mmGB_ADDR_CONFIG},
+	{mmMC_ARB_RAMCFG},
 	{mmGB_TILE_MODE0},
 	{mmGB_TILE_MODE1},
 	{mmGB_TILE_MODE2},
@@ -1156,7 +1170,7 @@ static struct amdgpu_allowed_register_entry si_allowed_read_registers[] = {
 	{mmGB_TILE_MODE29},
 	{mmGB_TILE_MODE30},
 	{mmGB_TILE_MODE31},
-	{CC_RB_BACKEND_DISABLE, true},
+	{mmCC_RB_BACKEND_DISABLE, true},
 	{mmGC_USER_RB_BACKEND_DISABLE, true},
 	{mmPA_SC_RASTER_CONFIG, true},
 };
@@ -1264,37 +1278,37 @@ static bool si_read_disabled_bios(struct amdgpu_device *adev)
 	u32 rom_cntl;
 	bool r;
 
-	bus_cntl = RREG32(R600_BUS_CNTL);
+	bus_cntl = RREG32(mmBUS_CNTL);
 	if (adev->mode_info.num_crtc) {
-		d1vga_control = RREG32(AVIVO_D1VGA_CONTROL);
-		d2vga_control = RREG32(AVIVO_D2VGA_CONTROL);
-		vga_render_control = RREG32(VGA_RENDER_CONTROL);
+		d1vga_control = RREG32(mmD1VGA_CONTROL);
+		d2vga_control = RREG32(mmD2VGA_CONTROL);
+		vga_render_control = RREG32(mmVGA_RENDER_CONTROL);
 	}
 	rom_cntl = RREG32(R600_ROM_CNTL);
 
 	/* enable the rom */
-	WREG32(R600_BUS_CNTL, (bus_cntl & ~R600_BIOS_ROM_DIS));
+	WREG32(mmBUS_CNTL, (bus_cntl & ~BUS_CNTL__BIOS_ROM_DIS_MASK));
 	if (adev->mode_info.num_crtc) {
 		/* Disable VGA mode */
-		WREG32(AVIVO_D1VGA_CONTROL,
-		       (d1vga_control & ~(AVIVO_DVGA_CONTROL_MODE_ENABLE |
-					  AVIVO_DVGA_CONTROL_TIMING_SELECT)));
-		WREG32(AVIVO_D2VGA_CONTROL,
-		       (d2vga_control & ~(AVIVO_DVGA_CONTROL_MODE_ENABLE |
-					  AVIVO_DVGA_CONTROL_TIMING_SELECT)));
-		WREG32(VGA_RENDER_CONTROL,
-		       (vga_render_control & C_000300_VGA_VSTATUS_CNTL));
+		WREG32(mmD1VGA_CONTROL,
+		       (d1vga_control & ~(D1VGA_CONTROL__D1VGA_MODE_ENABLE_MASK |
+					  D1VGA_CONTROL__D1VGA_TIMING_SELECT_MASK)));
+		WREG32(mmD2VGA_CONTROL,
+		       (d2vga_control & ~(D1VGA_CONTROL__D1VGA_MODE_ENABLE_MASK |
+					  D1VGA_CONTROL__D1VGA_TIMING_SELECT_MASK)));
+		WREG32(mmVGA_RENDER_CONTROL,
+		       (vga_render_control & ~VGA_RENDER_CONTROL__VGA_VSTATUS_CNTL_MASK));
 	}
 	WREG32(R600_ROM_CNTL, rom_cntl | R600_SCK_OVERWRITE);
 
 	r = amdgpu_read_bios(adev);
 
 	/* restore regs */
-	WREG32(R600_BUS_CNTL, bus_cntl);
+	WREG32(mmBUS_CNTL, bus_cntl);
 	if (adev->mode_info.num_crtc) {
-		WREG32(AVIVO_D1VGA_CONTROL, d1vga_control);
-		WREG32(AVIVO_D2VGA_CONTROL, d2vga_control);
-		WREG32(VGA_RENDER_CONTROL, vga_render_control);
+		WREG32(mmD1VGA_CONTROL, d1vga_control);
+		WREG32(mmD2VGA_CONTROL, d2vga_control);
+		WREG32(mmVGA_RENDER_CONTROL, vga_render_control);
 	}
 	WREG32(R600_ROM_CNTL, rom_cntl);
 	return r;
@@ -1331,23 +1345,24 @@ static void si_set_clk_bypass_mode(struct amdgpu_device *adev)
 {
 	u32 tmp, i;
 
-	tmp = RREG32(CG_SPLL_FUNC_CNTL);
-	tmp |= SPLL_BYPASS_EN;
-	WREG32(CG_SPLL_FUNC_CNTL, tmp);
+	tmp = RREG32(mmCG_SPLL_FUNC_CNTL);
+	tmp |= CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK;
+	WREG32(mmCG_SPLL_FUNC_CNTL, tmp);
 
-	tmp = RREG32(CG_SPLL_FUNC_CNTL_2);
-	tmp |= SPLL_CTLREQ_CHG;
-	WREG32(CG_SPLL_FUNC_CNTL_2, tmp);
+	tmp = RREG32(mmCG_SPLL_FUNC_CNTL_2);
+	tmp |= CG_SPLL_FUNC_CNTL_2__SPLL_CTLREQ_CHG_MASK;
+	WREG32(mmCG_SPLL_FUNC_CNTL_2, tmp);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32(SPLL_STATUS) & SPLL_CHG_STATUS)
+		if (RREG32(mmCG_SPLL_STATUS) & CG_SPLL_STATUS__SPLL_CHG_STATUS_MASK)
 			break;
 		udelay(1);
 	}
 
-	tmp = RREG32(CG_SPLL_FUNC_CNTL_2);
-	tmp &= ~(SPLL_CTLREQ_CHG | SCLK_MUX_UPDATE);
-	WREG32(CG_SPLL_FUNC_CNTL_2, tmp);
+	tmp = RREG32(mmCG_SPLL_FUNC_CNTL_2);
+	tmp &= ~(CG_SPLL_FUNC_CNTL_2__SPLL_CTLREQ_CHG_MASK |
+		CG_SPLL_FUNC_CNTL_2__SCLK_MUX_UPDATE_MASK);
+	WREG32(mmCG_SPLL_FUNC_CNTL_2, tmp);
 
 	tmp = RREG32(MPLL_CNTL_MODE);
 	tmp &= ~MPLL_MCLK_SEL;
@@ -1358,21 +1373,21 @@ static void si_spll_powerdown(struct amdgpu_device *adev)
 {
 	u32 tmp;
 
-	tmp = RREG32(SPLL_CNTL_MODE);
-	tmp |= SPLL_SW_DIR_CONTROL;
-	WREG32(SPLL_CNTL_MODE, tmp);
+	tmp = RREG32(mmSPLL_CNTL_MODE);
+	tmp |= SPLL_CNTL_MODE__SPLL_SW_DIR_CONTROL_MASK;
+	WREG32(mmSPLL_CNTL_MODE, tmp);
 
-	tmp = RREG32(CG_SPLL_FUNC_CNTL);
-	tmp |= SPLL_RESET;
-	WREG32(CG_SPLL_FUNC_CNTL, tmp);
+	tmp = RREG32(mmCG_SPLL_FUNC_CNTL);
+	tmp |= CG_SPLL_FUNC_CNTL__SPLL_RESET_MASK;
+	WREG32(mmCG_SPLL_FUNC_CNTL, tmp);
 
-	tmp = RREG32(CG_SPLL_FUNC_CNTL);
-	tmp |= SPLL_SLEEP;
-	WREG32(CG_SPLL_FUNC_CNTL, tmp);
+	tmp = RREG32(mmCG_SPLL_FUNC_CNTL);
+	tmp |= CG_SPLL_FUNC_CNTL__SPLL_SLEEP_MASK;
+	WREG32(mmCG_SPLL_FUNC_CNTL, tmp);
 
-	tmp = RREG32(SPLL_CNTL_MODE);
-	tmp &= ~SPLL_SW_DIR_CONTROL;
-	WREG32(SPLL_CNTL_MODE, tmp);
+	tmp = RREG32(mmSPLL_CNTL_MODE);
+	tmp &= ~SPLL_CNTL_MODE__SPLL_SW_DIR_CONTROL_MASK;
+	WREG32(mmSPLL_CNTL_MODE, tmp);
 }
 
 static int si_gpu_pci_config_reset(struct amdgpu_device *adev)
@@ -1454,14 +1469,14 @@ static void si_vga_set_state(struct amdgpu_device *adev, bool state)
 {
 	uint32_t temp;
 
-	temp = RREG32(CONFIG_CNTL);
+	temp = RREG32(mmCONFIG_CNTL);
 	if (!state) {
 		temp &= ~(1<<0);
 		temp |= (1<<1);
 	} else {
 		temp &= ~(1<<1);
 	}
-	WREG32(CONFIG_CNTL, temp);
+	WREG32(mmCONFIG_CNTL, temp);
 }
 
 static u32 si_get_xclk(struct amdgpu_device *adev)
@@ -1469,12 +1484,12 @@ static u32 si_get_xclk(struct amdgpu_device *adev)
 	u32 reference_clock = adev->clock.spll.reference_freq;
 	u32 tmp;
 
-	tmp = RREG32(CG_CLKPIN_CNTL_2);
-	if (tmp & MUX_TCLK_TO_XCLK)
+	tmp = RREG32(mmCG_CLKPIN_CNTL_2);
+	if (tmp & CG_CLKPIN_CNTL_2__MUX_TCLK_TO_XCLK_MASK)
 		return TCLK;
 
-	tmp = RREG32(CG_CLKPIN_CNTL);
-	if (tmp & XTALIN_DIVIDE)
+	tmp = RREG32(mmCG_CLKPIN_CNTL);
+	if (tmp & CG_CLKPIN_CNTL__XTALIN_DIVIDE_MASK)
 		return reference_clock / 4;
 
 	return reference_clock;
@@ -1519,9 +1534,9 @@ static int si_get_pcie_lanes(struct amdgpu_device *adev)
 	if (adev->flags & AMD_IS_APU)
 		return 0;
 
-	link_width_cntl = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL);
+	link_width_cntl = RREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL);
 
-	switch ((link_width_cntl & LC_LINK_WIDTH_RD_MASK) >> LC_LINK_WIDTH_RD_SHIFT) {
+	switch ((link_width_cntl & PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK) >> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT) {
 	case LC_LINK_WIDTH_X1:
 		return 1;
 	case LC_LINK_WIDTH_X2:
@@ -1568,13 +1583,13 @@ static void si_set_pcie_lanes(struct amdgpu_device *adev, int lanes)
 		return;
 	}
 
-	link_width_cntl = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL);
-	link_width_cntl &= ~LC_LINK_WIDTH_MASK;
-	link_width_cntl |= mask << LC_LINK_WIDTH_SHIFT;
-	link_width_cntl |= (LC_RECONFIG_NOW |
-			    LC_RECONFIG_ARC_MISSING_ESCAPE);
+	link_width_cntl = RREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL);
+	link_width_cntl &= ~PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_MASK;
+	link_width_cntl |= mask << PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH__SHIFT;
+	link_width_cntl |= (PCIE_LC_LINK_WIDTH_CNTL__LC_RECONFIG_NOW_MASK |
+		PCIE_LC_LINK_WIDTH_CNTL__LC_RECONFIG_ARC_MISSING_ESCAPE_MASK);
 
-	WREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL, link_width_cntl);
+	WREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL, link_width_cntl);
 }
 
 static void si_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0,
@@ -2018,7 +2033,7 @@ static const struct amdgpu_asic_funcs si_asic_funcs =
 
 static uint32_t si_get_rev_id(struct amdgpu_device *adev)
 {
-	return (RREG32(CC_DRM_ID_STRAPS) & CC_DRM_ID_STRAPS__ATI_REV_ID_MASK)
+	return (RREG32(mmCC_DRM_ID_STRAPS) & CC_DRM_ID_STRAPS__ATI_REV_ID_MASK)
 		>> CC_DRM_ID_STRAPS__ATI_REV_ID__SHIFT;
 }
 
@@ -2239,9 +2254,9 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev)
 					CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)))
 		return;
 
-	speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
-	current_data_rate = (speed_cntl & LC_CURRENT_DATA_RATE_MASK) >>
-		LC_CURRENT_DATA_RATE_SHIFT;
+	speed_cntl = RREG32_PCIE_PORT(ixPCIE_LC_SPEED_CNTL);
+	current_data_rate = (speed_cntl & PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK) >>
+		PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
 	if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) {
 		if (current_data_rate == 2) {
 			DRM_INFO("PCIE gen 3 link speeds already enabled\n");
@@ -2268,17 +2283,17 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev)
 			pcie_capability_set_word(root, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD);
 			pcie_capability_set_word(adev->pdev, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD);
 
-			tmp = RREG32_PCIE(PCIE_LC_STATUS1);
-			max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT;
-			current_lw = (tmp & LC_OPERATING_LINK_WIDTH_MASK) >> LC_OPERATING_LINK_WIDTH_SHIFT;
+			tmp = RREG32_PCIE(ixPCIE_LC_STATUS1);
+			max_lw = (tmp & PCIE_LC_STATUS1__LC_DETECTED_LINK_WIDTH_MASK) >> PCIE_LC_STATUS1__LC_DETECTED_LINK_WIDTH__SHIFT;
+			current_lw = (tmp & PCIE_LC_STATUS1__LC_OPERATING_LINK_WIDTH_MASK) >> PCIE_LC_STATUS1__LC_OPERATING_LINK_WIDTH__SHIFT;
 
 			if (current_lw < max_lw) {
-				tmp = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL);
-				if (tmp & LC_RENEGOTIATION_SUPPORT) {
-					tmp &= ~(LC_LINK_WIDTH_MASK | LC_UPCONFIGURE_DIS);
-					tmp |= (max_lw << LC_LINK_WIDTH_SHIFT);
-					tmp |= LC_UPCONFIGURE_SUPPORT | LC_RENEGOTIATE_EN | LC_RECONFIG_NOW;
-					WREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL, tmp);
+				tmp = RREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL);
+				if (tmp & PCIE_LC_LINK_WIDTH_CNTL__LC_RENEGOTIATION_SUPPORT_MASK) {
+					tmp &= ~(PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_MASK | PCIE_LC_LINK_WIDTH_CNTL__LC_UPCONFIGURE_DIS_MASK);
+					tmp |= (max_lw << PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH__SHIFT);
+					tmp |= PCIE_LC_LINK_WIDTH_CNTL__LC_UPCONFIGURE_SUPPORT_MASK | PCIE_LC_LINK_WIDTH_CNTL__LC_RENEGOTIATE_EN_MASK | PCIE_LC_LINK_WIDTH_CNTL__LC_RECONFIG_NOW_MASK;
+					WREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL, tmp);
 				}
 			}
 
@@ -2301,13 +2316,13 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev)
 							  PCI_EXP_LNKCTL2,
 							  &gpu_cfg2);
 
-				tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4);
-				tmp |= LC_SET_QUIESCE;
-				WREG32_PCIE_PORT(PCIE_LC_CNTL4, tmp);
+				tmp = RREG32_PCIE_PORT(ixPCIE_LC_CNTL4);
+				tmp |= PCIE_LC_CNTL4__LC_SET_QUIESCE_MASK;
+				WREG32_PCIE_PORT(ixPCIE_LC_CNTL4, tmp);
 
-				tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4);
-				tmp |= LC_REDO_EQ;
-				WREG32_PCIE_PORT(PCIE_LC_CNTL4, tmp);
+				tmp = RREG32_PCIE_PORT(ixPCIE_LC_CNTL4);
+				tmp |= PCIE_LC_CNTL4__LC_REDO_EQ_MASK;
+				WREG32_PCIE_PORT(ixPCIE_LC_CNTL4, tmp);
 
 				mdelay(100);
 
@@ -2333,16 +2348,16 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev)
 								   (PCI_EXP_LNKCTL2_ENTER_COMP |
 								    PCI_EXP_LNKCTL2_TX_MARGIN));
 
-				tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4);
-				tmp &= ~LC_SET_QUIESCE;
-				WREG32_PCIE_PORT(PCIE_LC_CNTL4, tmp);
+				tmp = RREG32_PCIE_PORT(ixPCIE_LC_CNTL4);
+				tmp &= ~PCIE_LC_CNTL4__LC_SET_QUIESCE_MASK;
+				WREG32_PCIE_PORT(ixPCIE_LC_CNTL4, tmp);
 			}
 		}
 	}
 
-	speed_cntl |= LC_FORCE_EN_SW_SPEED_CHANGE | LC_FORCE_DIS_HW_SPEED_CHANGE;
-	speed_cntl &= ~LC_FORCE_DIS_SW_SPEED_CHANGE;
-	WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl);
+	speed_cntl |= PCIE_LC_SPEED_CNTL__LC_FORCE_EN_SW_SPEED_CHANGE_MASK | PCIE_LC_SPEED_CNTL__LC_FORCE_DIS_HW_SPEED_CHANGE_MASK;
+	speed_cntl &= ~PCIE_LC_SPEED_CNTL__LC_FORCE_DIS_SW_SPEED_CHANGE_MASK;
+	WREG32_PCIE_PORT(ixPCIE_LC_SPEED_CNTL, speed_cntl);
 
 	tmp16 = 0;
 	if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
@@ -2354,13 +2369,13 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev)
 	pcie_capability_clear_and_set_word(adev->pdev, PCI_EXP_LNKCTL2,
 					   PCI_EXP_LNKCTL2_TLS, tmp16);
 
-	speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
-	speed_cntl |= LC_INITIATE_LINK_SPEED_CHANGE;
-	WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl);
+	speed_cntl = RREG32_PCIE_PORT(ixPCIE_LC_SPEED_CNTL);
+	speed_cntl |= PCIE_LC_SPEED_CNTL__LC_INITIATE_LINK_SPEED_CHANGE_MASK;
+	WREG32_PCIE_PORT(ixPCIE_LC_SPEED_CNTL, speed_cntl);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
-		if ((speed_cntl & LC_INITIATE_LINK_SPEED_CHANGE) == 0)
+		speed_cntl = RREG32_PCIE_PORT(ixPCIE_LC_SPEED_CNTL);
+		if ((speed_cntl & PCIE_LC_SPEED_CNTL__LC_INITIATE_LINK_SPEED_CHANGE_MASK) == 0)
 			break;
 		udelay(1);
 	}
@@ -2418,121 +2433,121 @@ static void si_program_aspm(struct amdgpu_device *adev)
 	if (!amdgpu_device_should_use_aspm(adev))
 		return;
 
-	orig = data = RREG32_PCIE_PORT(PCIE_LC_N_FTS_CNTL);
-	data &= ~LC_XMIT_N_FTS_MASK;
-	data |= LC_XMIT_N_FTS(0x24) | LC_XMIT_N_FTS_OVERRIDE_EN;
+	orig = data = RREG32_PCIE_PORT(ixPCIE_LC_N_FTS_CNTL);
+	data &= ~PCIE_LC_N_FTS_CNTL__LC_XMIT_N_FTS_MASK;
+	data |= (0x24 << PCIE_LC_N_FTS_CNTL__LC_XMIT_N_FTS__SHIFT) | PCIE_LC_N_FTS_CNTL__LC_XMIT_N_FTS_OVERRIDE_EN_MASK;
 	if (orig != data)
-		WREG32_PCIE_PORT(PCIE_LC_N_FTS_CNTL, data);
+		WREG32_PCIE_PORT(ixPCIE_LC_N_FTS_CNTL, data);
 
-	orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL3);
-	data |= LC_GO_TO_RECOVERY;
+	orig = data = RREG32_PCIE_PORT(ixPCIE_LC_CNTL3);
+	data |= PCIE_LC_CNTL3__LC_GO_TO_RECOVERY_MASK;
 	if (orig != data)
-		WREG32_PCIE_PORT(PCIE_LC_CNTL3, data);
+		WREG32_PCIE_PORT(ixPCIE_LC_CNTL3, data);
 
-	orig = data = RREG32_PCIE(PCIE_P_CNTL);
-	data |= P_IGNORE_EDB_ERR;
+	orig = data = RREG32_PCIE(ixPCIE_P_CNTL);
+	data |= PCIE_P_CNTL__P_IGNORE_EDB_ERR_MASK;
 	if (orig != data)
-		WREG32_PCIE(PCIE_P_CNTL, data);
+		WREG32_PCIE(ixPCIE_P_CNTL, data);
 
-	orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL);
-	data &= ~(LC_L0S_INACTIVITY_MASK | LC_L1_INACTIVITY_MASK);
-	data |= LC_PMI_TO_L1_DIS;
+	orig = data = RREG32_PCIE_PORT(ixPCIE_LC_CNTL);
+	data &= ~(PCIE_LC_CNTL__LC_L0S_INACTIVITY_MASK | PCIE_LC_CNTL__LC_L1_INACTIVITY_MASK);
+	data |= PCIE_LC_CNTL__LC_PMI_TO_L1_DIS_MASK;
 	if (!disable_l0s)
-		data |= LC_L0S_INACTIVITY(7);
+		data |= (7 << PCIE_LC_CNTL__LC_L0S_INACTIVITY__SHIFT);
 
 	if (!disable_l1) {
-		data |= LC_L1_INACTIVITY(7);
-		data &= ~LC_PMI_TO_L1_DIS;
+		data |= (7 << PCIE_LC_CNTL__LC_L1_INACTIVITY__SHIFT);
+		data &= ~PCIE_LC_CNTL__LC_PMI_TO_L1_DIS_MASK;
 		if (orig != data)
-			WREG32_PCIE_PORT(PCIE_LC_CNTL, data);
+			WREG32_PCIE_PORT(ixPCIE_LC_CNTL, data);
 
 		if (!disable_plloff_in_l1) {
 			bool clk_req_support;
 
-			orig = data = si_pif_phy0_rreg(adev,PB0_PIF_PWRDOWN_0);
-			data &= ~(PLL_POWER_STATE_IN_OFF_0_MASK | PLL_POWER_STATE_IN_TXS2_0_MASK);
-			data |= PLL_POWER_STATE_IN_OFF_0(7) | PLL_POWER_STATE_IN_TXS2_0(7);
+			orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_PWRDOWN_0);
+			data &= ~(PB0_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_OFF_0_MASK | PB0_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_TXS2_0_MASK);
+			data |= (7 << PB0_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_OFF_0__SHIFT) | (7 << PB0_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_TXS2_0__SHIFT);
 			if (orig != data)
-				si_pif_phy0_wreg(adev,PB0_PIF_PWRDOWN_0, data);
+				si_pif_phy0_wreg(adev,ixPB0_PIF_PWRDOWN_0, data);
 
-			orig = data = si_pif_phy0_rreg(adev,PB0_PIF_PWRDOWN_1);
-			data &= ~(PLL_POWER_STATE_IN_OFF_1_MASK | PLL_POWER_STATE_IN_TXS2_1_MASK);
-			data |= PLL_POWER_STATE_IN_OFF_1(7) | PLL_POWER_STATE_IN_TXS2_1(7);
+			orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_PWRDOWN_1);
+			data &= ~(PB0_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_OFF_1_MASK | PB0_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_TXS2_1_MASK);
+			data |= (7 << PB0_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_OFF_1__SHIFT) | (7 << PB0_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_TXS2_1__SHIFT);
 			if (orig != data)
-				si_pif_phy0_wreg(adev,PB0_PIF_PWRDOWN_1, data);
+				si_pif_phy0_wreg(adev,ixPB0_PIF_PWRDOWN_1, data);
 
-			orig = data = si_pif_phy1_rreg(adev,PB1_PIF_PWRDOWN_0);
-			data &= ~(PLL_POWER_STATE_IN_OFF_0_MASK | PLL_POWER_STATE_IN_TXS2_0_MASK);
-			data |= PLL_POWER_STATE_IN_OFF_0(7) | PLL_POWER_STATE_IN_TXS2_0(7);
+			orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_PWRDOWN_0);
+			data &= ~(PB1_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_OFF_0_MASK | PB1_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_TXS2_0_MASK);
+			data |= (7 << PB1_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_OFF_0__SHIFT) | (7 << PB1_PIF_PWRDOWN_0__PLL_POWER_STATE_IN_TXS2_0__SHIFT);
 			if (orig != data)
-				si_pif_phy1_wreg(adev,PB1_PIF_PWRDOWN_0, data);
+				si_pif_phy1_wreg(adev,ixPB1_PIF_PWRDOWN_0, data);
 
-			orig = data = si_pif_phy1_rreg(adev,PB1_PIF_PWRDOWN_1);
-			data &= ~(PLL_POWER_STATE_IN_OFF_1_MASK | PLL_POWER_STATE_IN_TXS2_1_MASK);
-			data |= PLL_POWER_STATE_IN_OFF_1(7) | PLL_POWER_STATE_IN_TXS2_1(7);
+			orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_PWRDOWN_1);
+			data &= ~(PB1_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_OFF_1_MASK | PB1_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_TXS2_1_MASK);
+			data |= (7 << PB1_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_OFF_1__SHIFT) | (7 << PB1_PIF_PWRDOWN_1__PLL_POWER_STATE_IN_TXS2_1__SHIFT);
 			if (orig != data)
-				si_pif_phy1_wreg(adev,PB1_PIF_PWRDOWN_1, data);
+				si_pif_phy1_wreg(adev,ixPB1_PIF_PWRDOWN_1, data);
 
 			if ((adev->asic_type != CHIP_OLAND) && (adev->asic_type != CHIP_HAINAN)) {
-				orig = data = si_pif_phy0_rreg(adev,PB0_PIF_PWRDOWN_0);
-				data &= ~PLL_RAMP_UP_TIME_0_MASK;
+				orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_PWRDOWN_0);
+				data &= ~PB0_PIF_PWRDOWN_0__PLL_RAMP_UP_TIME_0_MASK;
 				if (orig != data)
-					si_pif_phy0_wreg(adev,PB0_PIF_PWRDOWN_0, data);
+					si_pif_phy0_wreg(adev,ixPB0_PIF_PWRDOWN_0, data);
 
-				orig = data = si_pif_phy0_rreg(adev,PB0_PIF_PWRDOWN_1);
-				data &= ~PLL_RAMP_UP_TIME_1_MASK;
+				orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_PWRDOWN_1);
+				data &= ~PB0_PIF_PWRDOWN_1__PLL_RAMP_UP_TIME_1_MASK;
 				if (orig != data)
-					si_pif_phy0_wreg(adev,PB0_PIF_PWRDOWN_1, data);
+					si_pif_phy0_wreg(adev,ixPB0_PIF_PWRDOWN_1, data);
 
-				orig = data = si_pif_phy0_rreg(adev,PB0_PIF_PWRDOWN_2);
-				data &= ~PLL_RAMP_UP_TIME_2_MASK;
+				orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_PWRDOWN_2);
+				data &= ~PB0_PIF_PWRDOWN_2__PLL_RAMP_UP_TIME_2_MASK;
 				if (orig != data)
-					si_pif_phy0_wreg(adev,PB0_PIF_PWRDOWN_2, data);
+					si_pif_phy0_wreg(adev,ixPB0_PIF_PWRDOWN_2, data);
 
-				orig = data = si_pif_phy0_rreg(adev,PB0_PIF_PWRDOWN_3);
-				data &= ~PLL_RAMP_UP_TIME_3_MASK;
+				orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_PWRDOWN_3);
+				data &= ~PB0_PIF_PWRDOWN_3__PLL_RAMP_UP_TIME_3_MASK;
 				if (orig != data)
-					si_pif_phy0_wreg(adev,PB0_PIF_PWRDOWN_3, data);
+					si_pif_phy0_wreg(adev,ixPB0_PIF_PWRDOWN_3, data);
 
-				orig = data = si_pif_phy1_rreg(adev,PB1_PIF_PWRDOWN_0);
-				data &= ~PLL_RAMP_UP_TIME_0_MASK;
+				orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_PWRDOWN_0);
+				data &= ~PB1_PIF_PWRDOWN_0__PLL_RAMP_UP_TIME_0_MASK;
 				if (orig != data)
-					si_pif_phy1_wreg(adev,PB1_PIF_PWRDOWN_0, data);
+					si_pif_phy1_wreg(adev,ixPB1_PIF_PWRDOWN_0, data);
 
-				orig = data = si_pif_phy1_rreg(adev,PB1_PIF_PWRDOWN_1);
-				data &= ~PLL_RAMP_UP_TIME_1_MASK;
+				orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_PWRDOWN_1);
+				data &= ~PB1_PIF_PWRDOWN_1__PLL_RAMP_UP_TIME_1_MASK;
 				if (orig != data)
-					si_pif_phy1_wreg(adev,PB1_PIF_PWRDOWN_1, data);
+					si_pif_phy1_wreg(adev,ixPB1_PIF_PWRDOWN_1, data);
 
-				orig = data = si_pif_phy1_rreg(adev,PB1_PIF_PWRDOWN_2);
-				data &= ~PLL_RAMP_UP_TIME_2_MASK;
+				orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_PWRDOWN_2);
+				data &= ~PB1_PIF_PWRDOWN_2__PLL_RAMP_UP_TIME_2_MASK;
 				if (orig != data)
-					si_pif_phy1_wreg(adev,PB1_PIF_PWRDOWN_2, data);
+					si_pif_phy1_wreg(adev,ixPB1_PIF_PWRDOWN_2, data);
 
-				orig = data = si_pif_phy1_rreg(adev,PB1_PIF_PWRDOWN_3);
-				data &= ~PLL_RAMP_UP_TIME_3_MASK;
+				orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_PWRDOWN_3);
+				data &= ~PB1_PIF_PWRDOWN_3__PLL_RAMP_UP_TIME_3_MASK;
 				if (orig != data)
-					si_pif_phy1_wreg(adev,PB1_PIF_PWRDOWN_3, data);
+					si_pif_phy1_wreg(adev,ixPB1_PIF_PWRDOWN_3, data);
 			}
-			orig = data = RREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL);
-			data &= ~LC_DYN_LANES_PWR_STATE_MASK;
-			data |= LC_DYN_LANES_PWR_STATE(3);
+			orig = data = RREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL);
+			data &= ~PCIE_LC_LINK_WIDTH_CNTL__LC_DYN_LANES_PWR_STATE_MASK;
+			data |= (3 << PCIE_LC_LINK_WIDTH_CNTL__LC_DYN_LANES_PWR_STATE__SHIFT);
 			if (orig != data)
-				WREG32_PCIE_PORT(PCIE_LC_LINK_WIDTH_CNTL, data);
+				WREG32_PCIE_PORT(ixPCIE_LC_LINK_WIDTH_CNTL, data);
 
-			orig = data = si_pif_phy0_rreg(adev,PB0_PIF_CNTL);
-			data &= ~LS2_EXIT_TIME_MASK;
+			orig = data = si_pif_phy0_rreg(adev,ixPB0_PIF_CNTL);
+			data &= ~PB0_PIF_CNTL__LS2_EXIT_TIME_MASK;
 			if ((adev->asic_type == CHIP_OLAND) || (adev->asic_type == CHIP_HAINAN))
-				data |= LS2_EXIT_TIME(5);
+				data |= (5 << PB0_PIF_CNTL__LS2_EXIT_TIME__SHIFT);
 			if (orig != data)
-				si_pif_phy0_wreg(adev,PB0_PIF_CNTL, data);
+				si_pif_phy0_wreg(adev,ixPB0_PIF_CNTL, data);
 
-			orig = data = si_pif_phy1_rreg(adev,PB1_PIF_CNTL);
-			data &= ~LS2_EXIT_TIME_MASK;
+			orig = data = si_pif_phy1_rreg(adev,ixPB1_PIF_CNTL);
+			data &= ~PB1_PIF_CNTL__LS2_EXIT_TIME_MASK;
 			if ((adev->asic_type == CHIP_OLAND) || (adev->asic_type == CHIP_HAINAN))
-				data |= LS2_EXIT_TIME(5);
+				data |= (5 << PB1_PIF_CNTL__LS2_EXIT_TIME__SHIFT);
 			if (orig != data)
-				si_pif_phy1_wreg(adev,PB1_PIF_CNTL, data);
+				si_pif_phy1_wreg(adev,ixPB1_PIF_CNTL, data);
 
 			if (!disable_clkreq &&
 			    !pci_is_root_bus(adev->pdev->bus)) {
@@ -2548,64 +2563,64 @@ static void si_program_aspm(struct amdgpu_device *adev)
 			}
 
 			if (clk_req_support) {
-				orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL2);
-				data |= LC_ALLOW_PDWN_IN_L1 | LC_ALLOW_PDWN_IN_L23;
+				orig = data = RREG32_PCIE_PORT(ixPCIE_LC_CNTL2);
+				data |= PCIE_LC_CNTL2__LC_ALLOW_PDWN_IN_L1_MASK | PCIE_LC_CNTL2__LC_ALLOW_PDWN_IN_L23_MASK;
 				if (orig != data)
-					WREG32_PCIE_PORT(PCIE_LC_CNTL2, data);
+					WREG32_PCIE_PORT(ixPCIE_LC_CNTL2, data);
 
-				orig = data = RREG32(THM_CLK_CNTL);
-				data &= ~(CMON_CLK_SEL_MASK | TMON_CLK_SEL_MASK);
-				data |= CMON_CLK_SEL(1) | TMON_CLK_SEL(1);
+				orig = data = RREG32(mmTHM_CLK_CNTL);
+				data &= ~(THM_CLK_CNTL__CMON_CLK_SEL_MASK | THM_CLK_CNTL__TMON_CLK_SEL_MASK);
+				data |= (1 << THM_CLK_CNTL__CMON_CLK_SEL__SHIFT) | (1 << THM_CLK_CNTL__TMON_CLK_SEL__SHIFT);
 				if (orig != data)
-					WREG32(THM_CLK_CNTL, data);
+					WREG32(mmTHM_CLK_CNTL, data);
 
-				orig = data = RREG32(MISC_CLK_CNTL);
-				data &= ~(DEEP_SLEEP_CLK_SEL_MASK | ZCLK_SEL_MASK);
-				data |= DEEP_SLEEP_CLK_SEL(1) | ZCLK_SEL(1);
+				orig = data = RREG32(mmMISC_CLK_CNTL);
+				data &= ~(MISC_CLK_CNTL__DEEP_SLEEP_CLK_SEL_MASK | MISC_CLK_CNTL__ZCLK_SEL_MASK);
+				data |= (1 << MISC_CLK_CNTL__DEEP_SLEEP_CLK_SEL__SHIFT) | (1 << MISC_CLK_CNTL__ZCLK_SEL__SHIFT);
 				if (orig != data)
-					WREG32(MISC_CLK_CNTL, data);
+					WREG32(mmMISC_CLK_CNTL, data);
 
-				orig = data = RREG32(CG_CLKPIN_CNTL);
-				data &= ~BCLK_AS_XCLK;
+				orig = data = RREG32(mmCG_CLKPIN_CNTL);
+				data &= ~CG_CLKPIN_CNTL__BCLK_AS_XCLK_MASK;
 				if (orig != data)
-					WREG32(CG_CLKPIN_CNTL, data);
+					WREG32(mmCG_CLKPIN_CNTL, data);
 
-				orig = data = RREG32(CG_CLKPIN_CNTL_2);
-				data &= ~FORCE_BIF_REFCLK_EN;
+				orig = data = RREG32(mmCG_CLKPIN_CNTL_2);
+				data &= ~CG_CLKPIN_CNTL_2__FORCE_BIF_REFCLK_EN_MASK;
 				if (orig != data)
-					WREG32(CG_CLKPIN_CNTL_2, data);
+					WREG32(mmCG_CLKPIN_CNTL_2, data);
 
-				orig = data = RREG32(MPLL_BYPASSCLK_SEL);
-				data &= ~MPLL_CLKOUT_SEL_MASK;
-				data |= MPLL_CLKOUT_SEL(4);
+				orig = data = RREG32(mmMPLL_BYPASSCLK_SEL);
+				data &= ~MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL_MASK;
+				data |= 4 << MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL__SHIFT;
 				if (orig != data)
-					WREG32(MPLL_BYPASSCLK_SEL, data);
+					WREG32(mmMPLL_BYPASSCLK_SEL, data);
 
-				orig = data = RREG32(SPLL_CNTL_MODE);
-				data &= ~SPLL_REFCLK_SEL_MASK;
+				orig = data = RREG32(mmSPLL_CNTL_MODE);
+				data &= ~SPLL_CNTL_MODE__SPLL_REFCLK_SEL_MASK;
 				if (orig != data)
-					WREG32(SPLL_CNTL_MODE, data);
+					WREG32(mmSPLL_CNTL_MODE, data);
 			}
 		}
 	} else {
 		if (orig != data)
-			WREG32_PCIE_PORT(PCIE_LC_CNTL, data);
+			WREG32_PCIE_PORT(ixPCIE_LC_CNTL, data);
 	}
 
-	orig = data = RREG32_PCIE(PCIE_CNTL2);
-	data |= SLV_MEM_LS_EN | MST_MEM_LS_EN | REPLAY_MEM_LS_EN;
+	orig = data = RREG32_PCIE(ixPCIE_CNTL2);
+	data |= PCIE_CNTL2__SLV_MEM_LS_EN_MASK | PCIE_CNTL2__MST_MEM_LS_EN_MASK | PCIE_CNTL2__REPLAY_MEM_LS_EN_MASK;
 	if (orig != data)
-		WREG32_PCIE(PCIE_CNTL2, data);
+		WREG32_PCIE(ixPCIE_CNTL2, data);
 
 	if (!disable_l0s) {
-		data = RREG32_PCIE_PORT(PCIE_LC_N_FTS_CNTL);
-		if((data & LC_N_FTS_MASK) == LC_N_FTS_MASK) {
-			data = RREG32_PCIE(PCIE_LC_STATUS1);
-			if ((data & LC_REVERSE_XMIT) && (data & LC_REVERSE_RCVR)) {
-				orig = data = RREG32_PCIE_PORT(PCIE_LC_CNTL);
-				data &= ~LC_L0S_INACTIVITY_MASK;
+		data = RREG32_PCIE_PORT(ixPCIE_LC_N_FTS_CNTL);
+		if((data & PCIE_LC_N_FTS_CNTL__LC_N_FTS_MASK) == PCIE_LC_N_FTS_CNTL__LC_N_FTS_MASK) {
+			data = RREG32_PCIE(ixPCIE_LC_STATUS1);
+			if ((data & PCIE_LC_STATUS1__LC_REVERSE_XMIT_MASK) && (data & PCIE_LC_STATUS1__LC_REVERSE_RCVR_MASK)) {
+				orig = data = RREG32_PCIE_PORT(ixPCIE_LC_CNTL);
+				data &= ~PCIE_LC_CNTL__LC_L0S_INACTIVITY_MASK;
 				if (orig != data)
-					WREG32_PCIE_PORT(PCIE_LC_CNTL, data);
+					WREG32_PCIE_PORT(ixPCIE_LC_CNTL, data);
 			}
 		}
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index e2089c8da71b..7f18e4875287 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -27,6 +27,8 @@
 #include "si.h"
 #include "sid.h"
 
+#include "oss/oss_1_0_d.h"
+#include "oss/oss_1_0_sh_mask.h"
 const u32 sdma_offsets[SDMA_MAX_INSTANCE] =
 {
 	DMA0_REGISTER_OFFSET,
@@ -38,17 +40,31 @@ static void si_dma_set_buffer_funcs(struct amdgpu_device *adev);
 static void si_dma_set_vm_pte_funcs(struct amdgpu_device *adev);
 static void si_dma_set_irq_funcs(struct amdgpu_device *adev);
 
+/**
+ * si_dma_ring_get_rptr - get the current read pointer
+ *
+ * @ring: amdgpu ring pointer
+ *
+ * Get the current rptr from the hardware (SI).
+ */
 static uint64_t si_dma_ring_get_rptr(struct amdgpu_ring *ring)
 {
 	return *ring->rptr_cpu_addr;
 }
 
+/**
+ * si_dma_ring_get_wptr - get the current write pointer
+ *
+ * @ring: amdgpu ring pointer
+ *
+ * Get the current wptr from the hardware (SI).
+ */
 static uint64_t si_dma_ring_get_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	u32 me = (ring == &adev->sdma.instance[0].ring) ? 0 : 1;
 
-	return (RREG32(DMA_RB_WPTR + sdma_offsets[me]) & 0x3fffc) >> 2;
+	return (RREG32(mmDMA_GFX_RB_WPTR + sdma_offsets[me]) & 0x3fffc) >> 2;
 }
 
 static void si_dma_ring_set_wptr(struct amdgpu_ring *ring)
@@ -56,7 +72,7 @@ static void si_dma_ring_set_wptr(struct amdgpu_ring *ring)
 	struct amdgpu_device *adev = ring->adev;
 	u32 me = (ring == &adev->sdma.instance[0].ring) ? 0 : 1;
 
-	WREG32(DMA_RB_WPTR + sdma_offsets[me], (ring->wptr << 2) & 0x3fffc);
+	WREG32(mmDMA_GFX_RB_WPTR + sdma_offsets[me], (ring->wptr << 2) & 0x3fffc);
 }
 
 static void si_dma_ring_emit_ib(struct amdgpu_ring *ring,
@@ -117,9 +133,9 @@ static void si_dma_stop(struct amdgpu_device *adev)
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
 		/* dma0 */
-		rb_cntl = RREG32(DMA_RB_CNTL + sdma_offsets[i]);
-		rb_cntl &= ~DMA_RB_ENABLE;
-		WREG32(DMA_RB_CNTL + sdma_offsets[i], rb_cntl);
+		rb_cntl = RREG32(mmDMA_GFX_RB_CNTL + sdma_offsets[i]);
+		rb_cntl &= ~DMA_GFX_RB_CNTL__RB_ENABLE_MASK;
+		WREG32(mmDMA_GFX_RB_CNTL + sdma_offsets[i], rb_cntl);
 	}
 }
 
@@ -133,44 +149,44 @@ static int si_dma_start(struct amdgpu_device *adev)
 	for (i = 0; i < adev->sdma.num_instances; i++) {
 		ring = &adev->sdma.instance[i].ring;
 
-		WREG32(DMA_SEM_INCOMPLETE_TIMER_CNTL + sdma_offsets[i], 0);
-		WREG32(DMA_SEM_WAIT_FAIL_TIMER_CNTL + sdma_offsets[i], 0);
+		WREG32(mmDMA_SEM_INCOMPLETE_TIMER_CNTL + sdma_offsets[i], 0);
+		WREG32(mmDMA_SEM_WAIT_FAIL_TIMER_CNTL + sdma_offsets[i], 0);
 
 		/* Set ring buffer size in dwords */
 		rb_bufsz = order_base_2(ring->ring_size / 4);
 		rb_cntl = rb_bufsz << 1;
 #ifdef __BIG_ENDIAN
-		rb_cntl |= DMA_RB_SWAP_ENABLE | DMA_RPTR_WRITEBACK_SWAP_ENABLE;
+		rb_cntl |= DMA_GFX_RB_CNTL__RB_SWAP_ENABLE_MASK | DMA_GFX_RB_CNTL__RPTR_WRITEBACK_SWAP_ENABLE_MASK;
 #endif
-		WREG32(DMA_RB_CNTL + sdma_offsets[i], rb_cntl);
+		WREG32(mmDMA_GFX_RB_CNTL + sdma_offsets[i], rb_cntl);
 
 		/* Initialize the ring buffer's read and write pointers */
-		WREG32(DMA_RB_RPTR + sdma_offsets[i], 0);
-		WREG32(DMA_RB_WPTR + sdma_offsets[i], 0);
+		WREG32(mmDMA_GFX_RB_RPTR + sdma_offsets[i], 0);
+		WREG32(mmDMA_GFX_RB_WPTR + sdma_offsets[i], 0);
 
 		rptr_addr = ring->rptr_gpu_addr;
 
-		WREG32(DMA_RB_RPTR_ADDR_LO + sdma_offsets[i], lower_32_bits(rptr_addr));
-		WREG32(DMA_RB_RPTR_ADDR_HI + sdma_offsets[i], upper_32_bits(rptr_addr) & 0xFF);
+		WREG32(mmDMA_GFX_RB_RPTR_ADDR_LO + sdma_offsets[i], lower_32_bits(rptr_addr));
+		WREG32(mmDMA_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i], upper_32_bits(rptr_addr) & 0xFF);
 
-		rb_cntl |= DMA_RPTR_WRITEBACK_ENABLE;
+		rb_cntl |= DMA_GFX_RB_CNTL__RPTR_WRITEBACK_ENABLE_MASK;
 
-		WREG32(DMA_RB_BASE + sdma_offsets[i], ring->gpu_addr >> 8);
+		WREG32(mmDMA_GFX_RB_BASE + sdma_offsets[i], ring->gpu_addr >> 8);
 
 		/* enable DMA IBs */
-		ib_cntl = DMA_IB_ENABLE | CMD_VMID_FORCE;
+		ib_cntl = DMA_GFX_IB_CNTL__IB_ENABLE_MASK | DMA_GFX_IB_CNTL__CMD_VMID_FORCE_MASK;
 #ifdef __BIG_ENDIAN
-		ib_cntl |= DMA_IB_SWAP_ENABLE;
+		ib_cntl |= DMA_GFX_IB_CNTL__IB_SWAP_ENABLE_MASK;
 #endif
-		WREG32(DMA_IB_CNTL + sdma_offsets[i], ib_cntl);
+		WREG32(mmDMA_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
 
-		dma_cntl = RREG32(DMA_CNTL + sdma_offsets[i]);
-		dma_cntl &= ~CTXEMPTY_INT_ENABLE;
-		WREG32(DMA_CNTL + sdma_offsets[i], dma_cntl);
+		dma_cntl = RREG32(mmDMA_CNTL + sdma_offsets[i]);
+		dma_cntl &= ~DMA_CNTL__CTXEMPTY_INT_ENABLE_MASK;
+		WREG32(mmDMA_CNTL + sdma_offsets[i], dma_cntl);
 
 		ring->wptr = 0;
-		WREG32(DMA_RB_WPTR + sdma_offsets[i], ring->wptr << 2);
-		WREG32(DMA_RB_CNTL + sdma_offsets[i], rb_cntl | DMA_RB_ENABLE);
+		WREG32(mmDMA_GFX_RB_WPTR + sdma_offsets[i], ring->wptr << 2);
+		WREG32(mmDMA_GFX_RB_CNTL + sdma_offsets[i], rb_cntl | DMA_GFX_RB_CNTL__RB_ENABLE_MASK);
 
 		r = amdgpu_ring_test_helper(ring);
 		if (r)
@@ -461,7 +477,7 @@ static int si_dma_early_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 
-	adev->sdma.num_instances = 2;
+	adev->sdma.num_instances = SDMA_MAX_INSTANCE;
 
 	si_dma_set_ring_funcs(adev);
 	si_dma_set_buffer_funcs(adev);
@@ -545,9 +561,9 @@ static bool si_dma_is_idle(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 
-	u32 tmp = RREG32(SRBM_STATUS2);
+	u32 tmp = RREG32(mmSRBM_STATUS2);
 
-	if (tmp & (DMA_BUSY_MASK | DMA1_BUSY_MASK))
+	if (tmp & (SRBM_STATUS2__DMA_BUSY_MASK | SRBM_STATUS2__DMA1_BUSY_MASK))
 	    return false;
 
 	return true;
@@ -583,14 +599,14 @@ static int si_dma_set_trap_irq_state(struct amdgpu_device *adev,
 	case AMDGPU_SDMA_IRQ_INSTANCE0:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
-			sdma_cntl = RREG32(DMA_CNTL + DMA0_REGISTER_OFFSET);
-			sdma_cntl &= ~TRAP_ENABLE;
-			WREG32(DMA_CNTL + DMA0_REGISTER_OFFSET, sdma_cntl);
+			sdma_cntl = RREG32(mmDMA_CNTL + DMA0_REGISTER_OFFSET);
+			sdma_cntl &= ~DMA_CNTL__TRAP_ENABLE_MASK;
+			WREG32(mmDMA_CNTL + DMA0_REGISTER_OFFSET, sdma_cntl);
 			break;
 		case AMDGPU_IRQ_STATE_ENABLE:
-			sdma_cntl = RREG32(DMA_CNTL + DMA0_REGISTER_OFFSET);
-			sdma_cntl |= TRAP_ENABLE;
-			WREG32(DMA_CNTL + DMA0_REGISTER_OFFSET, sdma_cntl);
+			sdma_cntl = RREG32(mmDMA_CNTL + DMA0_REGISTER_OFFSET);
+			sdma_cntl |= DMA_CNTL__TRAP_ENABLE_MASK;
+			WREG32(mmDMA_CNTL + DMA0_REGISTER_OFFSET, sdma_cntl);
 			break;
 		default:
 			break;
@@ -599,14 +615,14 @@ static int si_dma_set_trap_irq_state(struct amdgpu_device *adev,
 	case AMDGPU_SDMA_IRQ_INSTANCE1:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
-			sdma_cntl = RREG32(DMA_CNTL + DMA1_REGISTER_OFFSET);
-			sdma_cntl &= ~TRAP_ENABLE;
-			WREG32(DMA_CNTL + DMA1_REGISTER_OFFSET, sdma_cntl);
+			sdma_cntl = RREG32(mmDMA_CNTL + DMA1_REGISTER_OFFSET);
+			sdma_cntl &= ~DMA_CNTL__TRAP_ENABLE_MASK;
+			WREG32(mmDMA_CNTL + DMA1_REGISTER_OFFSET, sdma_cntl);
 			break;
 		case AMDGPU_IRQ_STATE_ENABLE:
-			sdma_cntl = RREG32(DMA_CNTL + DMA1_REGISTER_OFFSET);
-			sdma_cntl |= TRAP_ENABLE;
-			WREG32(DMA_CNTL + DMA1_REGISTER_OFFSET, sdma_cntl);
+			sdma_cntl = RREG32(mmDMA_CNTL + DMA1_REGISTER_OFFSET);
+			sdma_cntl |= DMA_CNTL__TRAP_ENABLE_MASK;
+			WREG32(mmDMA_CNTL + DMA1_REGISTER_OFFSET, sdma_cntl);
 			break;
 		default:
 			break;
@@ -645,11 +661,11 @@ static int si_dma_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 				offset = DMA0_REGISTER_OFFSET;
 			else
 				offset = DMA1_REGISTER_OFFSET;
-			orig = data = RREG32(DMA_POWER_CNTL + offset);
-			data &= ~MEM_POWER_OVERRIDE;
+			orig = data = RREG32(mmDMA_POWER_CNTL + offset);
+			data &= ~DMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
 			if (data != orig)
-				WREG32(DMA_POWER_CNTL + offset, data);
-			WREG32(DMA_CLK_CTRL + offset, 0x00000100);
+				WREG32(mmDMA_POWER_CNTL + offset, data);
+			WREG32(mmDMA_CLK_CTRL + offset, 0x00000100);
 		}
 	} else {
 		for (i = 0; i < adev->sdma.num_instances; i++) {
@@ -657,15 +673,15 @@ static int si_dma_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 				offset = DMA0_REGISTER_OFFSET;
 			else
 				offset = DMA1_REGISTER_OFFSET;
-			orig = data = RREG32(DMA_POWER_CNTL + offset);
-			data |= MEM_POWER_OVERRIDE;
+			orig = data = RREG32(mmDMA_POWER_CNTL + offset);
+			data |= DMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
 			if (data != orig)
-				WREG32(DMA_POWER_CNTL + offset, data);
+				WREG32(mmDMA_POWER_CNTL + offset, data);
 
-			orig = data = RREG32(DMA_CLK_CTRL + offset);
+			orig = data = RREG32(mmDMA_CLK_CTRL + offset);
 			data = 0xff000000;
 			if (data != orig)
-				WREG32(DMA_CLK_CTRL + offset, data);
+				WREG32(mmDMA_CLK_CTRL + offset, data);
 		}
 	}
 
@@ -679,11 +695,11 @@ static int si_dma_set_powergating_state(struct amdgpu_ip_block *ip_block,
 
 	struct amdgpu_device *adev = ip_block->adev;
 
-	WREG32(DMA_PGFSM_WRITE,  0x00002000);
-	WREG32(DMA_PGFSM_CONFIG, 0x100010ff);
+	WREG32(mmDMA_PGFSM_WRITE,  0x00002000);
+	WREG32(mmDMA_PGFSM_CONFIG, 0x100010ff);
 
 	for (tmp = 0; tmp < 5; tmp++)
-		WREG32(DMA_PGFSM_WRITE, 0);
+		WREG32(mmDMA_PGFSM_WRITE, 0);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/si_enums.h b/drivers/gpu/drm/amd/amdgpu/si_enums.h
index d656ef1fa6e1..6da65778292b 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_enums.h
+++ b/drivers/gpu/drm/amd/amdgpu/si_enums.h
@@ -23,115 +23,15 @@
 #ifndef SI_ENUMS_H
 #define SI_ENUMS_H
 
-#define VBLANK_INT_MASK                (1 << 0)
-#define DC_HPDx_INT_EN                 (1 << 16)
-#define VBLANK_ACK                     (1 << 4)
-#define VLINE_ACK                      (1 << 4)
-
-#define CURSOR_WIDTH 64
-#define CURSOR_HEIGHT 64
-
-#define VGA_VSTATUS_CNTL               0xFFFCFFFF
 #define PRIORITY_MARK_MASK             0x7fff
 #define PRIORITY_OFF                   (1 << 16)
 #define PRIORITY_ALWAYS_ON             (1 << 20)
-#define INTERLEAVE_EN                  (1 << 0)
-
-#define LATENCY_WATERMARK_MASK(x)      ((x) << 16)
-#define DC_LB_MEMORY_CONFIG(x)         ((x) << 20)
-#define ICON_DEGAMMA_MODE(x)           (((x) & 0x3) << 8)
-
-#define GRPH_ENDIAN_SWAP(x)            (((x) & 0x3) << 0)
-#define GRPH_ENDIAN_NONE               0
-#define GRPH_ENDIAN_8IN16              1
-#define GRPH_ENDIAN_8IN32              2
-#define GRPH_ENDIAN_8IN64              3
-#define GRPH_RED_CROSSBAR(x)           (((x) & 0x3) << 4)
-#define GRPH_RED_SEL_R                 0
-#define GRPH_RED_SEL_G                 1
-#define GRPH_RED_SEL_B                 2
-#define GRPH_RED_SEL_A                 3
-#define GRPH_GREEN_CROSSBAR(x)         (((x) & 0x3) << 6)
-#define GRPH_GREEN_SEL_G               0
-#define GRPH_GREEN_SEL_B               1
-#define GRPH_GREEN_SEL_A               2
-#define GRPH_GREEN_SEL_R               3
-#define GRPH_BLUE_CROSSBAR(x)          (((x) & 0x3) << 8)
-#define GRPH_BLUE_SEL_B                0
-#define GRPH_BLUE_SEL_A                1
-#define GRPH_BLUE_SEL_R                2
-#define GRPH_BLUE_SEL_G                3
-#define GRPH_ALPHA_CROSSBAR(x)         (((x) & 0x3) << 10)
-#define GRPH_ALPHA_SEL_A               0
-#define GRPH_ALPHA_SEL_R               1
-#define GRPH_ALPHA_SEL_G               2
-#define GRPH_ALPHA_SEL_B               3
-
-#define GRPH_DEPTH(x)                  (((x) & 0x3) << 0)
-#define GRPH_DEPTH_8BPP                0
-#define GRPH_DEPTH_16BPP               1
-#define GRPH_DEPTH_32BPP               2
-
-#define GRPH_FORMAT(x)                 (((x) & 0x7) << 8)
-#define GRPH_FORMAT_INDEXED            0
-#define GRPH_FORMAT_ARGB1555           0
-#define GRPH_FORMAT_ARGB565            1
-#define GRPH_FORMAT_ARGB4444           2
-#define GRPH_FORMAT_AI88               3
-#define GRPH_FORMAT_MONO16             4
-#define GRPH_FORMAT_BGRA5551           5
-#define GRPH_FORMAT_ARGB8888           0
-#define GRPH_FORMAT_ARGB2101010        1
-#define GRPH_FORMAT_32BPP_DIG          2
-#define GRPH_FORMAT_8B_ARGB2101010     3
-#define GRPH_FORMAT_BGRA1010102        4
-#define GRPH_FORMAT_8B_BGRA1010102     5
-#define GRPH_FORMAT_RGB111110          6
-#define GRPH_FORMAT_BGR101111          7
-
-#define GRPH_NUM_BANKS(x)              (((x) & 0x3) << 2)
-#define GRPH_ARRAY_MODE(x)             (((x) & 0x7) << 20)
-#define GRPH_ARRAY_LINEAR_GENERAL      0
-#define GRPH_ARRAY_LINEAR_ALIGNED      1
-#define GRPH_ARRAY_1D_TILED_THIN1      2
-#define GRPH_ARRAY_2D_TILED_THIN1      4
-#define GRPH_TILE_SPLIT(x)             (((x) & 0x7) << 13)
-#define GRPH_BANK_WIDTH(x)             (((x) & 0x3) << 6)
-#define GRPH_BANK_HEIGHT(x)            (((x) & 0x3) << 11)
-#define GRPH_MACRO_TILE_ASPECT(x)      (((x) & 0x3) << 18)
-#define GRPH_ARRAY_MODE(x)             (((x) & 0x7) << 20)
-#define GRPH_PIPE_CONFIG(x)                   (((x) & 0x1f) << 24)
-
-#define CURSOR_EN                      (1 << 0)
-#define CURSOR_MODE(x)                 (((x) & 0x3) << 8)
-#define CURSOR_MONO                    0
-#define CURSOR_24_1                    1
-#define CURSOR_24_8_PRE_MULT           2
-#define CURSOR_24_8_UNPRE_MULT         3
-#define CURSOR_2X_MAGNIFY              (1 << 16)
-#define CURSOR_FORCE_MC_ON             (1 << 20)
-#define CURSOR_URGENT_CONTROL(x)       (((x) & 0x7) << 24)
-#define CURSOR_URGENT_ALWAYS           0
-#define CURSOR_URGENT_1_8              1
-#define CURSOR_URGENT_1_4              2
-#define CURSOR_URGENT_3_8              3
-#define CURSOR_URGENT_1_2              4
-#define CURSOR_UPDATE_PENDING          (1 << 0)
-#define CURSOR_UPDATE_TAKEN            (1 << 1)
-#define CURSOR_UPDATE_LOCK             (1 << 16)
-#define CURSOR_DISABLE_MULTIPLE_UPDATE (1 << 24)
-
-
-#define ES_AND_GS_AUTO       3
-#define RADEON_PACKET_TYPE3  3
-#define CE_PARTITION_BASE    3
-#define BUF_SWAP_32BIT       (2 << 16)
 
 #define GFX_POWER_STATUS                           (1 << 1)
 #define GFX_CLOCK_STATUS                           (1 << 2)
 #define GFX_LS_STATUS                              (1 << 3)
-#define RLC_BUSY_STATUS                            (1 << 0)
 
+#define RLC_BUSY_STATUS                            (1 << 0)
 #define RLC_PUD(x)                               ((x) << 0)
 #define RLC_PUD_MASK                             (0xff << 0)
 #define RLC_PDD(x)                               ((x) << 8)
@@ -140,140 +40,8 @@
 #define RLC_TTPD_MASK                            (0xff << 16)
 #define RLC_MSD(x)                               ((x) << 24)
 #define RLC_MSD_MASK                             (0xff << 24)
-#define WRITE_DATA_ENGINE_SEL(x) ((x) << 30)
-#define WRITE_DATA_DST_SEL(x) ((x) << 8)
-#define EVENT_TYPE(x) ((x) << 0)
-#define EVENT_INDEX(x) ((x) << 8)
-#define WAIT_REG_MEM_MEM_SPACE(x)               ((x) << 4)
-#define WAIT_REG_MEM_FUNCTION(x)                ((x) << 0)
-#define WAIT_REG_MEM_ENGINE(x)                  ((x) << 8)
 
-#define GFX6_NUM_GFX_RINGS     1
-#define GFX6_NUM_COMPUTE_RINGS 2
 #define RLC_SAVE_AND_RESTORE_STARTING_OFFSET 0x90
 #define RLC_CLEAR_STATE_DESCRIPTOR_OFFSET    0x3D
 
-#define PACKET3(op, n)  ((RADEON_PACKET_TYPE3 << 30) |                  \
-                         (((op) & 0xFF) << 8) |                         \
-                         ((n) & 0x3FFF) << 16)
-#define PACKET3_COMPUTE(op, n) (PACKET3(op, n) | 1 << 1)
-#define	PACKET3_NOP					0x10
-#define	PACKET3_SET_BASE				0x11
-#define		PACKET3_BASE_INDEX(x)                  ((x) << 0)
-#define	PACKET3_CLEAR_STATE				0x12
-#define	PACKET3_INDEX_BUFFER_SIZE			0x13
-#define	PACKET3_DISPATCH_DIRECT				0x15
-#define	PACKET3_DISPATCH_INDIRECT			0x16
-#define	PACKET3_ALLOC_GDS				0x1B
-#define	PACKET3_WRITE_GDS_RAM				0x1C
-#define	PACKET3_ATOMIC_GDS				0x1D
-#define	PACKET3_ATOMIC					0x1E
-#define	PACKET3_OCCLUSION_QUERY				0x1F
-#define	PACKET3_SET_PREDICATION				0x20
-#define	PACKET3_REG_RMW					0x21
-#define	PACKET3_COND_EXEC				0x22
-#define	PACKET3_PRED_EXEC				0x23
-#define	PACKET3_DRAW_INDIRECT				0x24
-#define	PACKET3_DRAW_INDEX_INDIRECT			0x25
-#define	PACKET3_INDEX_BASE				0x26
-#define	PACKET3_DRAW_INDEX_2				0x27
-#define	PACKET3_CONTEXT_CONTROL				0x28
-#define	PACKET3_INDEX_TYPE				0x2A
-#define	PACKET3_DRAW_INDIRECT_MULTI			0x2C
-#define	PACKET3_DRAW_INDEX_AUTO				0x2D
-#define	PACKET3_DRAW_INDEX_IMMD				0x2E
-#define	PACKET3_NUM_INSTANCES				0x2F
-#define	PACKET3_DRAW_INDEX_MULTI_AUTO			0x30
-#define	PACKET3_INDIRECT_BUFFER_CONST			0x31
-#define	PACKET3_INDIRECT_BUFFER				0x3F
-#define	PACKET3_STRMOUT_BUFFER_UPDATE			0x34
-#define	PACKET3_DRAW_INDEX_OFFSET_2			0x35
-#define	PACKET3_DRAW_INDEX_MULTI_ELEMENT		0x36
-#define	PACKET3_WRITE_DATA				0x37
-#define	PACKET3_DRAW_INDEX_INDIRECT_MULTI		0x38
-#define	PACKET3_MEM_SEMAPHORE				0x39
-#define	PACKET3_MPEG_INDEX				0x3A
-#define	PACKET3_COPY_DW					0x3B
-#define	PACKET3_WAIT_REG_MEM				0x3C
-#define	PACKET3_MEM_WRITE				0x3D
-#define	PACKET3_COPY_DATA				0x40
-#define	PACKET3_CP_DMA					0x41
-#              define PACKET3_CP_DMA_DST_SEL(x)    ((x) << 20)
-#              define PACKET3_CP_DMA_ENGINE(x)     ((x) << 27)
-#              define PACKET3_CP_DMA_SRC_SEL(x)    ((x) << 29)
-#              define PACKET3_CP_DMA_CP_SYNC       (1 << 31)
-#              define PACKET3_CP_DMA_DIS_WC        (1 << 21)
-#              define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 22)
-#              define PACKET3_CP_DMA_CMD_DST_SWAP(x) ((x) << 24)
-#              define PACKET3_CP_DMA_CMD_SAS       (1 << 26)
-#              define PACKET3_CP_DMA_CMD_DAS       (1 << 27)
-#              define PACKET3_CP_DMA_CMD_SAIC      (1 << 28)
-#              define PACKET3_CP_DMA_CMD_DAIC      (1 << 29)
-#              define PACKET3_CP_DMA_CMD_RAW_WAIT  (1 << 30)
-#define	PACKET3_PFP_SYNC_ME				0x42
-#define	PACKET3_SURFACE_SYNC				0x43
-#              define PACKET3_DEST_BASE_0_ENA      (1 << 0)
-#              define PACKET3_DEST_BASE_1_ENA      (1 << 1)
-#              define PACKET3_CB0_DEST_BASE_ENA    (1 << 6)
-#              define PACKET3_CB1_DEST_BASE_ENA    (1 << 7)
-#              define PACKET3_CB2_DEST_BASE_ENA    (1 << 8)
-#              define PACKET3_CB3_DEST_BASE_ENA    (1 << 9)
-#              define PACKET3_CB4_DEST_BASE_ENA    (1 << 10)
-#              define PACKET3_CB5_DEST_BASE_ENA    (1 << 11)
-#              define PACKET3_CB6_DEST_BASE_ENA    (1 << 12)
-#              define PACKET3_CB7_DEST_BASE_ENA    (1 << 13)
-#              define PACKET3_DB_DEST_BASE_ENA     (1 << 14)
-#              define PACKET3_DEST_BASE_2_ENA      (1 << 19)
-#              define PACKET3_DEST_BASE_3_ENA      (1 << 21)
-#              define PACKET3_TCL1_ACTION_ENA      (1 << 22)
-#              define PACKET3_TC_ACTION_ENA        (1 << 23)
-#              define PACKET3_CB_ACTION_ENA        (1 << 25)
-#              define PACKET3_DB_ACTION_ENA        (1 << 26)
-#              define PACKET3_SH_KCACHE_ACTION_ENA (1 << 27)
-#              define PACKET3_SH_ICACHE_ACTION_ENA (1 << 29)
-#define	PACKET3_ME_INITIALIZE				0x44
-#define		PACKET3_ME_INITIALIZE_DEVICE_ID(x) ((x) << 16)
-#define	PACKET3_COND_WRITE				0x45
-#define	PACKET3_EVENT_WRITE				0x46
-#define	PACKET3_EVENT_WRITE_EOP				0x47
-#define	PACKET3_EVENT_WRITE_EOS				0x48
-#define	PACKET3_PREAMBLE_CNTL				0x4A
-#              define PACKET3_PREAMBLE_BEGIN_CLEAR_STATE     (2 << 28)
-#              define PACKET3_PREAMBLE_END_CLEAR_STATE       (3 << 28)
-#define	PACKET3_ONE_REG_WRITE				0x57
-#define	PACKET3_LOAD_CONFIG_REG				0x5F
-#define	PACKET3_LOAD_CONTEXT_REG			0x60
-#define	PACKET3_LOAD_SH_REG				0x61
-#define	PACKET3_SET_CONFIG_REG				0x68
-#define		PACKET3_SET_CONFIG_REG_START			0x00002000
-#define		PACKET3_SET_CONFIG_REG_END			0x00002c00
-#define	PACKET3_SET_CONTEXT_REG				0x69
-#define		PACKET3_SET_CONTEXT_REG_START			0x000a000
-#define		PACKET3_SET_CONTEXT_REG_END			0x000a400
-#define	PACKET3_SET_CONTEXT_REG_INDIRECT		0x73
-#define	PACKET3_SET_RESOURCE_INDIRECT			0x74
-#define	PACKET3_SET_SH_REG				0x76
-#define		PACKET3_SET_SH_REG_START			0x00002c00
-#define		PACKET3_SET_SH_REG_END				0x00003000
-#define	PACKET3_SET_SH_REG_OFFSET			0x77
-#define	PACKET3_ME_WRITE				0x7A
-#define	PACKET3_SCRATCH_RAM_WRITE			0x7D
-#define	PACKET3_SCRATCH_RAM_READ			0x7E
-#define	PACKET3_CE_WRITE				0x7F
-#define	PACKET3_LOAD_CONST_RAM				0x80
-#define	PACKET3_WRITE_CONST_RAM				0x81
-#define	PACKET3_WRITE_CONST_RAM_OFFSET			0x82
-#define	PACKET3_DUMP_CONST_RAM				0x83
-#define	PACKET3_INCREMENT_CE_COUNTER			0x84
-#define	PACKET3_INCREMENT_DE_COUNTER			0x85
-#define	PACKET3_WAIT_ON_CE_COUNTER			0x86
-#define	PACKET3_WAIT_ON_DE_COUNTER			0x87
-#define	PACKET3_WAIT_ON_DE_COUNTER_DIFF			0x88
-#define	PACKET3_SET_CE_DE_COUNTERS			0x89
-#define	PACKET3_WAIT_ON_AVAIL_BUFFER			0x8A
-#define	PACKET3_SWITCH_BUFFER				0x8B
-#define PACKET3_SEM_WAIT_ON_SIGNAL    (0x1 << 12)
-#define PACKET3_SEM_SEL_SIGNAL	    (0x6 << 29)
-#define PACKET3_SEM_SEL_WAIT	    (0x7 << 29)
-
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/si_ih.c b/drivers/gpu/drm/amd/amdgpu/si_ih.c
index 5c38e1fb1dca..1df00f8a2406 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_ih.c
@@ -27,6 +27,7 @@
 #include "amdgpu_ih.h"
 #include "sid.h"
 #include "si_ih.h"
+
 #include "oss/oss_1_0_d.h"
 #include "oss/oss_1_0_sh_mask.h"
 
@@ -213,7 +214,7 @@ static int si_ih_resume(struct amdgpu_ip_block *ip_block)
 static bool si_ih_is_idle(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
-	u32 tmp = RREG32(SRBM_STATUS);
+	u32 tmp = RREG32(mmSRBM_STATUS);
 
 	if (tmp & SRBM_STATUS__IH_BUSY_MASK)
 		return false;
@@ -239,23 +240,23 @@ static int si_ih_soft_reset(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 
 	u32 srbm_soft_reset = 0;
-	u32 tmp = RREG32(SRBM_STATUS);
+	u32 tmp = RREG32(mmSRBM_STATUS);
 
 	if (tmp & SRBM_STATUS__IH_BUSY_MASK)
 		srbm_soft_reset |= SRBM_SOFT_RESET__SOFT_RESET_IH_MASK;
 
 	if (srbm_soft_reset) {
-		tmp = RREG32(SRBM_SOFT_RESET);
+		tmp = RREG32(mmSRBM_SOFT_RESET);
 		tmp |= srbm_soft_reset;
-		dev_info(adev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp);
-		WREG32(SRBM_SOFT_RESET, tmp);
-		tmp = RREG32(SRBM_SOFT_RESET);
+		dev_info(adev->dev, "mmSRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32(mmSRBM_SOFT_RESET, tmp);
+		tmp = RREG32(mmSRBM_SOFT_RESET);
 
 		udelay(50);
 
 		tmp &= ~srbm_soft_reset;
-		WREG32(SRBM_SOFT_RESET, tmp);
-		tmp = RREG32(SRBM_SOFT_RESET);
+		WREG32(mmSRBM_SOFT_RESET, tmp);
+		tmp = RREG32(mmSRBM_SOFT_RESET);
 
 		udelay(50);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/sid.h b/drivers/gpu/drm/amd/amdgpu/sid.h
index cbf232f5235b..cbd4f8951cfa 100644
--- a/drivers/gpu/drm/amd/amdgpu/sid.h
+++ b/drivers/gpu/drm/amd/amdgpu/sid.h
@@ -24,43 +24,12 @@
 #ifndef SI_H
 #define SI_H
 
-#define TAHITI_RB_BITMAP_WIDTH_PER_SH  2
-
-#define SI_MAX_SH_GPRS		 	256
-#define SI_MAX_TEMP_GPRS         	16
-#define SI_MAX_SH_THREADS        	256
-#define SI_MAX_SH_STACK_ENTRIES  	4096
-#define SI_MAX_FRC_EOV_CNT       	16384
-#define SI_MAX_BACKENDS          	8
-#define SI_MAX_BACKENDS_MASK     	0xFF
-#define SI_MAX_BACKENDS_PER_SE_MASK     0x0F
-#define SI_MAX_SIMDS             	12
-#define SI_MAX_SIMDS_MASK        	0x0FFF
-#define SI_MAX_SIMDS_PER_SE_MASK        0x00FF
-#define SI_MAX_PIPES            	8
-#define SI_MAX_PIPES_MASK        	0xFF
-#define SI_MAX_PIPES_PER_SIMD_MASK      0x3F
-#define SI_MAX_LDS_NUM           	0xFFFF
-#define SI_MAX_TCC               	16
-#define SI_MAX_TCC_MASK          	0xFFFF
 #define SI_MAX_CTLACKS_ASSERTION_WAIT   100
 
-/* SMC IND accessor regs */
-#define SMC_IND_INDEX_0                              0x80
-#define SMC_IND_DATA_0                               0x81
-
-#define SMC_IND_ACCESS_CNTL                          0x8A
-#       define AUTO_INCREMENT_IND_0                  (1 << 0)
-#define SMC_MESSAGE_0                                0x8B
-#define SMC_RESP_0                                   0x8C
-
 /* CG IND registers are accessed via SMC indirect space + SMC_CG_IND_START */
 #define SMC_CG_IND_START                    0xc0030000
 #define SMC_CG_IND_END                      0xc0040000
 
-#define	CG_CGTT_LOCAL_0				0x400
-#define	CG_CGTT_LOCAL_1				0x401
-
 /* SMC IND registers */
 #define	SMC_SYSCON_RESET_CNTL				0x80000000
 #       define RST_REG                                  (1 << 0)
@@ -68,9 +37,6 @@
 #       define CK_DISABLE                               (1 << 0)
 #       define CKEN                                     (1 << 24)
 
-#define VGA_HDP_CONTROL  				0xCA
-#define		VGA_MEMORY_DISABLE				(1 << 4)
-
 #define DCCG_DISP_SLOW_SELECT_REG                       0x13F
 #define		DCCG_DISP1_SLOW_SELECT(x)		((x) << 0)
 #define		DCCG_DISP1_SLOW_SELECT_MASK		(7 << 0)
@@ -79,47 +45,6 @@
 #define		DCCG_DISP2_SLOW_SELECT_MASK		(7 << 4)
 #define		DCCG_DISP2_SLOW_SELECT_SHIFT		4
 
-#define	CG_SPLL_FUNC_CNTL				0x180
-#define		SPLL_RESET				(1 << 0)
-#define		SPLL_SLEEP				(1 << 1)
-#define		SPLL_BYPASS_EN				(1 << 3)
-#define		SPLL_REF_DIV(x)				((x) << 4)
-#define		SPLL_REF_DIV_MASK			(0x3f << 4)
-#define		SPLL_PDIV_A(x)				((x) << 20)
-#define		SPLL_PDIV_A_MASK			(0x7f << 20)
-#define		SPLL_PDIV_A_SHIFT			20
-#define	CG_SPLL_FUNC_CNTL_2				0x181
-#define		SCLK_MUX_SEL(x)				((x) << 0)
-#define		SCLK_MUX_SEL_MASK			(0x1ff << 0)
-#define		SPLL_CTLREQ_CHG				(1 << 23)
-#define		SCLK_MUX_UPDATE				(1 << 26)
-#define	CG_SPLL_FUNC_CNTL_3				0x182
-#define		SPLL_FB_DIV(x)				((x) << 0)
-#define		SPLL_FB_DIV_MASK			(0x3ffffff << 0)
-#define		SPLL_FB_DIV_SHIFT			0
-#define		SPLL_DITHEN				(1 << 28)
-#define	CG_SPLL_FUNC_CNTL_4				0x183
-
-#define	SPLL_STATUS					0x185
-#define		SPLL_CHG_STATUS				(1 << 1)
-#define	SPLL_CNTL_MODE					0x186
-#define		SPLL_SW_DIR_CONTROL			(1 << 0)
-#	define SPLL_REFCLK_SEL(x)			((x) << 26)
-#	define SPLL_REFCLK_SEL_MASK			(3 << 26)
-
-#define	CG_SPLL_SPREAD_SPECTRUM				0x188
-#define		SSEN					(1 << 0)
-#define		CLK_S(x)				((x) << 4)
-#define		CLK_S_MASK				(0xfff << 4)
-#define		CLK_S_SHIFT				4
-#define	CG_SPLL_SPREAD_SPECTRUM_2			0x189
-#define		CLK_V(x)				((x) << 0)
-#define		CLK_V_MASK				(0x3ffffff << 0)
-#define		CLK_V_SHIFT				0
-
-#define	CG_SPLL_AUTOSCALE_CNTL				0x18b
-#       define AUTOSCALE_ON_SS_CLEAR                    (1 << 9)
-
 /* discrete uvd clocks */
 #define	CG_UPLL_FUNC_CNTL				0x18d
 #	define UPLL_RESET_MASK				0x00000001
@@ -149,317 +74,13 @@
 #define	CG_UPLL_SPREAD_SPECTRUM				0x194
 #	define SSEN_MASK				0x00000001
 
-#define	MPLL_BYPASSCLK_SEL				0x197
-#	define MPLL_CLKOUT_SEL(x)			((x) << 8)
-#	define MPLL_CLKOUT_SEL_MASK			0xFF00
-
-#define CG_CLKPIN_CNTL                                    0x198
-#       define XTALIN_DIVIDE                              (1 << 1)
-#       define BCLK_AS_XCLK                               (1 << 2)
-#define CG_CLKPIN_CNTL_2                                  0x199
-#       define FORCE_BIF_REFCLK_EN                        (1 << 3)
-#       define MUX_TCLK_TO_XCLK                           (1 << 8)
-
-#define	THM_CLK_CNTL					0x19b
-#	define CMON_CLK_SEL(x)				((x) << 0)
-#	define CMON_CLK_SEL_MASK			0xFF
-#	define TMON_CLK_SEL(x)				((x) << 8)
-#	define TMON_CLK_SEL_MASK			0xFF00
-#define	MISC_CLK_CNTL					0x19c
-#	define DEEP_SLEEP_CLK_SEL(x)			((x) << 0)
-#	define DEEP_SLEEP_CLK_SEL_MASK			0xFF
-#	define ZCLK_SEL(x)				((x) << 8)
-#	define ZCLK_SEL_MASK				0xFF00
-
-#define	CG_THERMAL_CTRL					0x1c0
-#define 	DPM_EVENT_SRC(x)			((x) << 0)
-#define 	DPM_EVENT_SRC_MASK			(7 << 0)
-#define		DIG_THERM_DPM(x)			((x) << 14)
-#define		DIG_THERM_DPM_MASK			0x003FC000
-#define		DIG_THERM_DPM_SHIFT			14
-#define	CG_THERMAL_STATUS				0x1c1
-#define		FDO_PWM_DUTY(x)				((x) << 9)
-#define		FDO_PWM_DUTY_MASK			(0xff << 9)
-#define		FDO_PWM_DUTY_SHIFT			9
-#define	CG_THERMAL_INT					0x1c2
-#define		DIG_THERM_INTH(x)			((x) << 8)
-#define		DIG_THERM_INTH_MASK			0x0000FF00
-#define		DIG_THERM_INTH_SHIFT			8
-#define		DIG_THERM_INTL(x)			((x) << 16)
-#define		DIG_THERM_INTL_MASK			0x00FF0000
-#define		DIG_THERM_INTL_SHIFT			16
-#define 	THERM_INT_MASK_HIGH			(1 << 24)
-#define 	THERM_INT_MASK_LOW			(1 << 25)
-
-#define	CG_MULT_THERMAL_CTRL					0x1c4
-#define		TEMP_SEL(x)					((x) << 20)
-#define		TEMP_SEL_MASK					(0xff << 20)
-#define		TEMP_SEL_SHIFT					20
-#define	CG_MULT_THERMAL_STATUS					0x1c5
-#define		ASIC_MAX_TEMP(x)				((x) << 0)
-#define		ASIC_MAX_TEMP_MASK				0x000001ff
-#define		ASIC_MAX_TEMP_SHIFT				0
-#define		CTF_TEMP(x)					((x) << 9)
-#define		CTF_TEMP_MASK					0x0003fe00
-#define		CTF_TEMP_SHIFT					9
-
-#define	CG_FDO_CTRL0					0x1d5
-#define		FDO_STATIC_DUTY(x)			((x) << 0)
-#define		FDO_STATIC_DUTY_MASK			0x000000FF
-#define		FDO_STATIC_DUTY_SHIFT			0
-#define	CG_FDO_CTRL1					0x1d6
-#define		FMAX_DUTY100(x)				((x) << 0)
-#define		FMAX_DUTY100_MASK			0x000000FF
-#define		FMAX_DUTY100_SHIFT			0
-#define	CG_FDO_CTRL2					0x1d7
-#define		TMIN(x)					((x) << 0)
-#define		TMIN_MASK				0x000000FF
-#define		TMIN_SHIFT				0
-#define		FDO_PWM_MODE(x)				((x) << 11)
-#define		FDO_PWM_MODE_MASK			(7 << 11)
-#define		FDO_PWM_MODE_SHIFT			11
-#define		TACH_PWM_RESP_RATE(x)			((x) << 25)
-#define		TACH_PWM_RESP_RATE_MASK			(0x7f << 25)
-#define		TACH_PWM_RESP_RATE_SHIFT		25
-
-#define CG_TACH_CTRL                                    0x1dc
-#       define EDGE_PER_REV(x)                          ((x) << 0)
-#       define EDGE_PER_REV_MASK                        (0x7 << 0)
-#       define EDGE_PER_REV_SHIFT                       0
-#       define TARGET_PERIOD(x)                         ((x) << 3)
-#       define TARGET_PERIOD_MASK                       0xfffffff8
-#       define TARGET_PERIOD_SHIFT                      3
-#define CG_TACH_STATUS                                  0x1dd
-#       define TACH_PERIOD(x)                           ((x) << 0)
-#       define TACH_PERIOD_MASK                         0xffffffff
-#       define TACH_PERIOD_SHIFT                        0
-
-#define GENERAL_PWRMGT                                  0x1e0
-#       define GLOBAL_PWRMGT_EN                         (1 << 0)
-#       define STATIC_PM_EN                             (1 << 1)
-#       define THERMAL_PROTECTION_DIS                   (1 << 2)
-#       define THERMAL_PROTECTION_TYPE                  (1 << 3)
-#       define SW_SMIO_INDEX(x)                         ((x) << 6)
-#       define SW_SMIO_INDEX_MASK                       (1 << 6)
-#       define SW_SMIO_INDEX_SHIFT                      6
-#       define VOLT_PWRMGT_EN                           (1 << 10)
-#       define DYN_SPREAD_SPECTRUM_EN                   (1 << 23)
-#define CG_TPC                                            0x1e1
-#define SCLK_PWRMGT_CNTL                                  0x1e2
-#       define SCLK_PWRMGT_OFF                            (1 << 0)
-#       define SCLK_LOW_D1                                (1 << 1)
-#       define FIR_RESET                                  (1 << 4)
-#       define FIR_FORCE_TREND_SEL                        (1 << 5)
-#       define FIR_TREND_MODE                             (1 << 6)
-#       define DYN_GFX_CLK_OFF_EN                         (1 << 7)
-#       define GFX_CLK_FORCE_ON                           (1 << 8)
-#       define GFX_CLK_REQUEST_OFF                        (1 << 9)
-#       define GFX_CLK_FORCE_OFF                          (1 << 10)
-#       define GFX_CLK_OFF_ACPI_D1                        (1 << 11)
-#       define GFX_CLK_OFF_ACPI_D2                        (1 << 12)
-#       define GFX_CLK_OFF_ACPI_D3                        (1 << 13)
-#       define DYN_LIGHT_SLEEP_EN                         (1 << 14)
-
-#define TARGET_AND_CURRENT_PROFILE_INDEX                  0x1e6
-#       define CURRENT_STATE_INDEX_MASK                   (0xf << 4)
-#       define CURRENT_STATE_INDEX_SHIFT                  4
-
-#define CG_FTV                                            0x1ef
-
-#define CG_FFCT_0                                         0x1f0
-#       define UTC_0(x)                                   ((x) << 0)
-#       define UTC_0_MASK                                 (0x3ff << 0)
-#       define DTC_0(x)                                   ((x) << 10)
-#       define DTC_0_MASK                                 (0x3ff << 10)
-
-#define CG_BSP                                          0x1ff
-#       define BSP(x)					((x) << 0)
-#       define BSP_MASK					(0xffff << 0)
-#       define BSU(x)					((x) << 16)
-#       define BSU_MASK					(0xf << 16)
-#define CG_AT                                           0x200
-#       define CG_R(x)					((x) << 0)
-#       define CG_R_MASK				(0xffff << 0)
-#       define CG_L(x)					((x) << 16)
-#       define CG_L_MASK				(0xffff << 16)
-
-#define CG_GIT                                          0x201
-#       define CG_GICST(x)                              ((x) << 0)
-#       define CG_GICST_MASK                            (0xffff << 0)
-#       define CG_GIPOT(x)                              ((x) << 16)
-#       define CG_GIPOT_MASK                            (0xffff << 16)
-
-#define CG_SSP                                            0x203
-#       define SST(x)                                     ((x) << 0)
-#       define SST_MASK                                   (0xffff << 0)
-#       define SSTU(x)                                    ((x) << 16)
-#       define SSTU_MASK                                  (0xf << 16)
-
-#define CG_DISPLAY_GAP_CNTL                               0x20a
-#       define DISP1_GAP(x)                               ((x) << 0)
-#       define DISP1_GAP_MASK                             (3 << 0)
-#       define DISP2_GAP(x)                               ((x) << 2)
-#       define DISP2_GAP_MASK                             (3 << 2)
-#       define VBI_TIMER_COUNT(x)                         ((x) << 4)
-#       define VBI_TIMER_COUNT_MASK                       (0x3fff << 4)
-#       define VBI_TIMER_UNIT(x)                          ((x) << 20)
-#       define VBI_TIMER_UNIT_MASK                        (7 << 20)
-#       define DISP1_GAP_MCHG(x)                          ((x) << 24)
-#       define DISP1_GAP_MCHG_MASK                        (3 << 24)
-#       define DISP2_GAP_MCHG(x)                          ((x) << 26)
-#       define DISP2_GAP_MCHG_MASK                        (3 << 26)
-
-#define	CG_ULV_CONTROL					0x21e
-#define	CG_ULV_PARAMETER				0x21f
-
-#define	SMC_SCRATCH0					0x221
-
-#define	CG_CAC_CTRL					0x22e
-#	define CAC_WINDOW(x)				((x) << 0)
-#	define CAC_WINDOW_MASK				0x00ffffff
-
-#define DMIF_ADDR_CONFIG  				0x2F5
-
-#define DMIF_ADDR_CALC  				0x300
-
-#define	PIPE0_DMIF_BUFFER_CONTROL			  0x0328
-#       define DMIF_BUFFERS_ALLOCATED(x)                  ((x) << 0)
-#       define DMIF_BUFFERS_ALLOCATED_COMPLETED           (1 << 4)
-
-#define	SRBM_STATUS				        0x394
-#define		GRBM_RQ_PENDING 			(1 << 5)
-#define		VMC_BUSY 				(1 << 8)
-#define		MCB_BUSY 				(1 << 9)
-#define		MCB_NON_DISPLAY_BUSY 			(1 << 10)
-#define		MCC_BUSY 				(1 << 11)
-#define		MCD_BUSY 				(1 << 12)
-#define		SEM_BUSY 				(1 << 14)
-#define		IH_BUSY 				(1 << 17)
-
-#define	SRBM_SOFT_RESET				        0x398
-#define		SOFT_RESET_BIF				(1 << 1)
-#define		SOFT_RESET_DC				(1 << 5)
-#define		SOFT_RESET_DMA1				(1 << 6)
-#define		SOFT_RESET_GRBM				(1 << 8)
-#define		SOFT_RESET_HDP				(1 << 9)
-#define		SOFT_RESET_IH				(1 << 10)
-#define		SOFT_RESET_MC				(1 << 11)
-#define		SOFT_RESET_ROM				(1 << 14)
-#define		SOFT_RESET_SEM				(1 << 15)
-#define		SOFT_RESET_VMC				(1 << 17)
-#define		SOFT_RESET_DMA				(1 << 20)
-#define		SOFT_RESET_TST				(1 << 21)
-#define		SOFT_RESET_REGBB			(1 << 22)
-#define		SOFT_RESET_ORB				(1 << 23)
-
-#define	CC_SYS_RB_BACKEND_DISABLE			0x3A0
-#define	GC_USER_SYS_RB_BACKEND_DISABLE			0x3A1
-
-#define SRBM_READ_ERROR					0x3A6
-#define SRBM_INT_CNTL					0x3A8
-#define SRBM_INT_ACK					0x3AA
-
-#define	SRBM_STATUS2				        0x3B1
-#define		DMA_BUSY 				(1 << 5)
-#define		DMA1_BUSY 				(1 << 6)
-
-#define VM_L2_CNTL					0x500
-#define		ENABLE_L2_CACHE					(1 << 0)
-#define		ENABLE_L2_FRAGMENT_PROCESSING			(1 << 1)
-#define		L2_CACHE_PTE_ENDIAN_SWAP_MODE(x)		((x) << 2)
-#define		L2_CACHE_PDE_ENDIAN_SWAP_MODE(x)		((x) << 4)
-#define		ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE		(1 << 9)
-#define		ENABLE_L2_PDE0_CACHE_LRU_UPDATE_BY_WRITE	(1 << 10)
-#define		EFFECTIVE_L2_QUEUE_SIZE(x)			(((x) & 7) << 15)
-#define		CONTEXT1_IDENTITY_ACCESS_MODE(x)		(((x) & 3) << 19)
-#define VM_L2_CNTL2					0x501
-#define		INVALIDATE_ALL_L1_TLBS				(1 << 0)
-#define		INVALIDATE_L2_CACHE				(1 << 1)
-#define		INVALIDATE_CACHE_MODE(x)			((x) << 26)
-#define			INVALIDATE_PTE_AND_PDE_CACHES		0
-#define			INVALIDATE_ONLY_PTE_CACHES		1
-#define			INVALIDATE_ONLY_PDE_CACHES		2
-#define VM_L2_CNTL3					0x502
-#define		BANK_SELECT(x)					((x) << 0)
-#define		L2_CACHE_UPDATE_MODE(x)				((x) << 6)
-#define		L2_CACHE_BIGK_FRAGMENT_SIZE(x)			((x) << 15)
-#define		L2_CACHE_BIGK_ASSOCIATIVITY			(1 << 20)
-#define	VM_L2_STATUS					0x503
-#define		L2_BUSY						(1 << 0)
-#define VM_CONTEXT0_CNTL				0x504
-#define		ENABLE_CONTEXT					(1 << 0)
-#define		PAGE_TABLE_DEPTH(x)				(((x) & 3) << 1)
-#define		RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 3)
-#define		RANGE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 4)
-#define		DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT	(1 << 6)
-#define		DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT	(1 << 7)
-#define		PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 9)
-#define		PDE0_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 10)
-#define		VALID_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 12)
-#define		VALID_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 13)
-#define		READ_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 15)
-#define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
-#define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
-#define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
-#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
-#define VM_CONTEXT1_CNTL				0x505
-#define VM_CONTEXT0_CNTL2				0x50C
-#define VM_CONTEXT1_CNTL2				0x50D
-#define	VM_CONTEXT8_PAGE_TABLE_BASE_ADDR		0x50E
-#define	VM_CONTEXT9_PAGE_TABLE_BASE_ADDR		0x50F
-#define	VM_CONTEXT10_PAGE_TABLE_BASE_ADDR		0x510
-#define	VM_CONTEXT11_PAGE_TABLE_BASE_ADDR		0x511
-#define	VM_CONTEXT12_PAGE_TABLE_BASE_ADDR		0x512
-#define	VM_CONTEXT13_PAGE_TABLE_BASE_ADDR		0x513
-#define	VM_CONTEXT14_PAGE_TABLE_BASE_ADDR		0x514
-#define	VM_CONTEXT15_PAGE_TABLE_BASE_ADDR		0x515
-
-#define	VM_CONTEXT1_PROTECTION_FAULT_ADDR		0x53f
-#define	VM_CONTEXT1_PROTECTION_FAULT_STATUS		0x537
-#define		PROTECTIONS_MASK			(0xf << 0)
-#define		PROTECTIONS_SHIFT			0
-		/* bit 0: range
-		 * bit 1: pde0
-		 * bit 2: valid
-		 * bit 3: read
-		 * bit 4: write
-		 */
-#define		MEMORY_CLIENT_ID_MASK			(0xff << 12)
-#define		MEMORY_CLIENT_ID_SHIFT			12
-#define		MEMORY_CLIENT_RW_MASK			(1 << 24)
-#define		MEMORY_CLIENT_RW_SHIFT			24
-#define		FAULT_VMID_MASK				(0xf << 25)
-#define		FAULT_VMID_SHIFT			25
-
 #define VM_INVALIDATE_REQUEST				0x51E
 #define VM_INVALIDATE_RESPONSE				0x51F
 
-#define VM_CONTEXT0_PROTECTION_FAULT_DEFAULT_ADDR	0x546
-#define VM_CONTEXT1_PROTECTION_FAULT_DEFAULT_ADDR	0x547
-
-#define	VM_CONTEXT0_PAGE_TABLE_BASE_ADDR		0x54F
-#define	VM_CONTEXT1_PAGE_TABLE_BASE_ADDR		0x550
-#define	VM_CONTEXT2_PAGE_TABLE_BASE_ADDR		0x551
-#define	VM_CONTEXT3_PAGE_TABLE_BASE_ADDR		0x552
-#define	VM_CONTEXT4_PAGE_TABLE_BASE_ADDR		0x553
-#define	VM_CONTEXT5_PAGE_TABLE_BASE_ADDR		0x554
-#define	VM_CONTEXT6_PAGE_TABLE_BASE_ADDR		0x555
-#define	VM_CONTEXT7_PAGE_TABLE_BASE_ADDR		0x556
-#define	VM_CONTEXT0_PAGE_TABLE_START_ADDR		0x557
-#define	VM_CONTEXT1_PAGE_TABLE_START_ADDR		0x558
-
-#define	VM_CONTEXT0_PAGE_TABLE_END_ADDR			0x55F
-#define	VM_CONTEXT1_PAGE_TABLE_END_ADDR			0x560
-
 #define VM_L2_CG           				0x570
 #define		MC_CG_ENABLE				(1 << 18)
 #define		MC_LS_ENABLE				(1 << 19)
 
-#define MC_SHARED_CHMAP						0x801
-#define		NOOFCHAN_SHIFT					12
-#define		NOOFCHAN_MASK					0x0000f000
-#define MC_SHARED_CHREMAP					0x802
-
 #define	MC_VM_FB_LOCATION				0x809
 #define	MC_VM_AGP_TOP					0x80A
 #define	MC_VM_AGP_BOT					0x80B
@@ -491,21 +112,6 @@
 #define MC_CITF_MISC_WR_CG           			0x993
 #define MC_CITF_MISC_VM_CG           			0x994
 
-#define	MC_ARB_RAMCFG					0x9D8
-#define		NOOFBANK_SHIFT					0
-#define		NOOFBANK_MASK					0x00000003
-#define		NOOFRANK_SHIFT					2
-#define		NOOFRANK_MASK					0x00000004
-#define		NOOFROWS_SHIFT					3
-#define		NOOFROWS_MASK					0x00000038
-#define		NOOFCOLS_SHIFT					6
-#define		NOOFCOLS_MASK					0x000000C0
-#define		CHANSIZE_SHIFT					8
-#define		CHANSIZE_MASK					0x00000100
-#define		CHANSIZE_OVERRIDE				(1 << 11)
-#define		NOOFGROUPS_SHIFT				12
-#define		NOOFGROUPS_MASK					0x00001000
-
 #define	MC_ARB_DRAM_TIMING				0x9DD
 #define	MC_ARB_DRAM_TIMING2				0x9DE
 
@@ -631,20 +237,6 @@
 #define		CLKS(x)					((x) << 0)
 #define		CLKS_MASK				(0xfff << 0)
 
-#define	HDP_HOST_PATH_CNTL				0xB00
-#define 	CLOCK_GATING_DIS			(1 << 23)
-#define	HDP_NONSURFACE_BASE				0xB01
-#define	HDP_NONSURFACE_INFO				0xB02
-#define	HDP_NONSURFACE_SIZE				0xB03
-
-#define HDP_DEBUG0  					0xBCC
-
-#define HDP_ADDR_CONFIG  				0xBD2
-#define HDP_MISC_CNTL					0xBD3
-#define 	HDP_FLUSH_INVALIDATE_CACHE			(1 << 0)
-#define HDP_MEM_POWER_LS				0xBD4
-#define 	HDP_LS_ENABLE				(1 << 0)
-
 #define ATC_MISC_CG           				0xCD4
 
 #define IH_RB_CNTL                                        0xF80
@@ -674,8 +266,6 @@
 #       define MC_WR_CLEAN_CNT(x)                         ((x) << 20)
 #       define MC_VMID(x)                                 ((x) << 25)
 
-#define	CONFIG_MEMSIZE					0x150A
-
 #define INTERRUPT_CNTL                                    0x151A
 #       define IH_DUMMY_RD_OVERRIDE                       (1 << 0)
 #       define IH_DUMMY_RD_EN                             (1 << 1)
@@ -683,486 +273,22 @@
 #       define GEN_IH_INT_EN                              (1 << 8)
 #define INTERRUPT_CNTL2                                   0x151B
 
-#define HDP_MEM_COHERENCY_FLUSH_CNTL			0x1520
-
-#define	BIF_FB_EN						0x1524
-#define		FB_READ_EN					(1 << 0)
-#define		FB_WRITE_EN					(1 << 1)
-
-#define HDP_REG_COHERENCY_FLUSH_CNTL			0x1528
-
-/* DCE6 ELD audio interface */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0        0x28 /* LPCM */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR1        0x29 /* AC3 */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR2        0x2A /* MPEG1 */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR3        0x2B /* MP3 */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR4        0x2C /* MPEG2 */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR5        0x2D /* AAC */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR6        0x2E /* DTS */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR7        0x2F /* ATRAC */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR8        0x30 /* one bit audio - leave at 0 (default) */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR9        0x31 /* Dolby Digital */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR10       0x32 /* DTS-HD */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR11       0x33 /* MAT-MLP */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR12       0x34 /* DTS */
-#define AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13       0x35 /* WMA Pro */
-#       define MAX_CHANNELS(x)                            (((x) & 0x7) << 0)
-/* max channels minus one.  7 = 8 channels */
-#       define SUPPORTED_FREQUENCIES(x)                   (((x) & 0xff) << 8)
-#       define DESCRIPTOR_BYTE_2(x)                       (((x) & 0xff) << 16)
-#       define SUPPORTED_FREQUENCIES_STEREO(x)            (((x) & 0xff) << 24) /* LPCM only */
-/* SUPPORTED_FREQUENCIES, SUPPORTED_FREQUENCIES_STEREO
- * bit0 = 32 kHz
- * bit1 = 44.1 kHz
- * bit2 = 48 kHz
- * bit3 = 88.2 kHz
- * bit4 = 96 kHz
- * bit5 = 176.4 kHz
- * bit6 = 192 kHz
- */
-
-#define AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC         0x37
-#       define VIDEO_LIPSYNC(x)                           (((x) & 0xff) << 0)
-#       define AUDIO_LIPSYNC(x)                           (((x) & 0xff) << 8)
-/* VIDEO_LIPSYNC, AUDIO_LIPSYNC
- * 0   = invalid
- * x   = legal delay value
- * 255 = sync not supported
- */
-#define AZ_F0_CODEC_PIN_CONTROL_RESPONSE_HBR             0x38
-#       define HBR_CAPABLE                                (1 << 0) /* enabled by default */
-
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO0               0x3a
-#       define MANUFACTURER_ID(x)                        (((x) & 0xffff) << 0)
-#       define PRODUCT_ID(x)                             (((x) & 0xffff) << 16)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO1               0x3b
-#       define SINK_DESCRIPTION_LEN(x)                   (((x) & 0xff) << 0)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO2               0x3c
-#       define PORT_ID0(x)                               (((x) & 0xffffffff) << 0)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO3               0x3d
-#       define PORT_ID1(x)                               (((x) & 0xffffffff) << 0)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO4               0x3e
-#       define DESCRIPTION0(x)                           (((x) & 0xff) << 0)
-#       define DESCRIPTION1(x)                           (((x) & 0xff) << 8)
-#       define DESCRIPTION2(x)                           (((x) & 0xff) << 16)
-#       define DESCRIPTION3(x)                           (((x) & 0xff) << 24)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO5               0x3f
-#       define DESCRIPTION4(x)                           (((x) & 0xff) << 0)
-#       define DESCRIPTION5(x)                           (((x) & 0xff) << 8)
-#       define DESCRIPTION6(x)                           (((x) & 0xff) << 16)
-#       define DESCRIPTION7(x)                           (((x) & 0xff) << 24)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO6               0x40
-#       define DESCRIPTION8(x)                           (((x) & 0xff) << 0)
-#       define DESCRIPTION9(x)                           (((x) & 0xff) << 8)
-#       define DESCRIPTION10(x)                          (((x) & 0xff) << 16)
-#       define DESCRIPTION11(x)                          (((x) & 0xff) << 24)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO7               0x41
-#       define DESCRIPTION12(x)                          (((x) & 0xff) << 0)
-#       define DESCRIPTION13(x)                          (((x) & 0xff) << 8)
-#       define DESCRIPTION14(x)                          (((x) & 0xff) << 16)
-#       define DESCRIPTION15(x)                          (((x) & 0xff) << 24)
-#define AZ_F0_CODEC_PIN_CONTROL_SINK_INFO8               0x42
-#       define DESCRIPTION16(x)                          (((x) & 0xff) << 0)
-#       define DESCRIPTION17(x)                          (((x) & 0xff) << 8)
-
-#define AZ_F0_CODEC_PIN_CONTROL_HOT_PLUG_CONTROL         0x54
-#       define AUDIO_ENABLED                             (1 << 31)
-
-#define AZ_F0_CODEC_PIN_CONTROL_RESPONSE_CONFIGURATION_DEFAULT  0x56
-#define		PORT_CONNECTIVITY_MASK				(3 << 30)
-#define		PORT_CONNECTIVITY_SHIFT				30
-
-#define	DC_LB_MEMORY_SPLIT					0x1AC3
-#define		DC_LB_MEMORY_CONFIG(x)				((x) << 20)
-
-#define	PRIORITY_A_CNT						0x1AC6
-#define		PRIORITY_MARK_MASK				0x7fff
-#define		PRIORITY_OFF					(1 << 16)
-#define		PRIORITY_ALWAYS_ON				(1 << 20)
-#define	PRIORITY_B_CNT						0x1AC7
-
-#define	DPG_PIPE_ARBITRATION_CONTROL3				0x1B32
-#       define LATENCY_WATERMARK_MASK(x)			((x) << 16)
-#define	DPG_PIPE_LATENCY_CONTROL				0x1B33
-#       define LATENCY_LOW_WATERMARK(x)				((x) << 0)
-#       define LATENCY_HIGH_WATERMARK(x)			((x) << 16)
-
-/* 0x6bb8, 0x77b8, 0x103b8, 0x10fb8, 0x11bb8, 0x127b8 */
-#define VLINE_STATUS                                    0x1AEE
-#       define VLINE_OCCURRED                           (1 << 0)
-#       define VLINE_ACK                                (1 << 4)
-#       define VLINE_STAT                               (1 << 12)
-#       define VLINE_INTERRUPT                          (1 << 16)
-#       define VLINE_INTERRUPT_TYPE                     (1 << 17)
-/* 0x6bbc, 0x77bc, 0x103bc, 0x10fbc, 0x11bbc, 0x127bc */
-#define VBLANK_STATUS                                   0x1AEF
-#       define VBLANK_OCCURRED                          (1 << 0)
-#       define VBLANK_ACK                               (1 << 4)
-#       define VBLANK_STAT                              (1 << 12)
-#       define VBLANK_INTERRUPT                         (1 << 16)
-#       define VBLANK_INTERRUPT_TYPE                    (1 << 17)
-
-/* 0x6b40, 0x7740, 0x10340, 0x10f40, 0x11b40, 0x12740 */
-#define INT_MASK                                        0x1AD0
-#       define VBLANK_INT_MASK                          (1 << 0)
-#       define VLINE_INT_MASK                           (1 << 4)
-
-#define DISP_INTERRUPT_STATUS                           0x183D
-#       define LB_D1_VLINE_INTERRUPT                    (1 << 2)
-#       define LB_D1_VBLANK_INTERRUPT                   (1 << 3)
-#       define DC_HPD1_INTERRUPT                        (1 << 17)
-#       define DC_HPD1_RX_INTERRUPT                     (1 << 18)
-#       define DACA_AUTODETECT_INTERRUPT                (1 << 22)
-#       define DACB_AUTODETECT_INTERRUPT                (1 << 23)
-#       define DC_I2C_SW_DONE_INTERRUPT                 (1 << 24)
-#       define DC_I2C_HW_DONE_INTERRUPT                 (1 << 25)
-#define DISP_INTERRUPT_STATUS_CONTINUE                  0x183E
-#       define LB_D2_VLINE_INTERRUPT                    (1 << 2)
-#       define LB_D2_VBLANK_INTERRUPT                   (1 << 3)
-#       define DC_HPD2_INTERRUPT                        (1 << 17)
-#       define DC_HPD2_RX_INTERRUPT                     (1 << 18)
-#       define DISP_TIMER_INTERRUPT                     (1 << 24)
-#define DISP_INTERRUPT_STATUS_CONTINUE2                 0x183F
-#       define LB_D3_VLINE_INTERRUPT                    (1 << 2)
-#       define LB_D3_VBLANK_INTERRUPT                   (1 << 3)
-#       define DC_HPD3_INTERRUPT                        (1 << 17)
-#       define DC_HPD3_RX_INTERRUPT                     (1 << 18)
-#define DISP_INTERRUPT_STATUS_CONTINUE3                 0x1840
-#       define LB_D4_VLINE_INTERRUPT                    (1 << 2)
-#       define LB_D4_VBLANK_INTERRUPT                   (1 << 3)
-#       define DC_HPD4_INTERRUPT                        (1 << 17)
-#       define DC_HPD4_RX_INTERRUPT                     (1 << 18)
-#define DISP_INTERRUPT_STATUS_CONTINUE4                 0x1853
-#       define LB_D5_VLINE_INTERRUPT                    (1 << 2)
-#       define LB_D5_VBLANK_INTERRUPT                   (1 << 3)
-#       define DC_HPD5_INTERRUPT                        (1 << 17)
-#       define DC_HPD5_RX_INTERRUPT                     (1 << 18)
-#define DISP_INTERRUPT_STATUS_CONTINUE5                 0x1854
-#       define LB_D6_VLINE_INTERRUPT                    (1 << 2)
-#       define LB_D6_VBLANK_INTERRUPT                   (1 << 3)
-#       define DC_HPD6_INTERRUPT                        (1 << 17)
-#       define DC_HPD6_RX_INTERRUPT                     (1 << 18)
-
-/* 0x6858, 0x7458, 0x10058, 0x10c58, 0x11858, 0x12458 */
-#define GRPH_INT_STATUS                                 0x1A16
-#       define GRPH_PFLIP_INT_OCCURRED                  (1 << 0)
-#       define GRPH_PFLIP_INT_CLEAR                     (1 << 8)
-/* 0x685c, 0x745c, 0x1005c, 0x10c5c, 0x1185c, 0x1245c */
-#define	GRPH_INT_CONTROL			        0x1A17
-#       define GRPH_PFLIP_INT_MASK                      (1 << 0)
-#       define GRPH_PFLIP_INT_TYPE                      (1 << 8)
-
-#define	DAC_AUTODETECT_INT_CONTROL			0x19F2
-
-#define DC_HPD1_INT_STATUS                              0x1807
-#define DC_HPD2_INT_STATUS                              0x180A
-#define DC_HPD3_INT_STATUS                              0x180D
-#define DC_HPD4_INT_STATUS                              0x1810
-#define DC_HPD5_INT_STATUS                              0x1813
-#define DC_HPD6_INT_STATUS                              0x1816
-#       define DC_HPDx_INT_STATUS                       (1 << 0)
-#       define DC_HPDx_SENSE                            (1 << 1)
-#       define DC_HPDx_RX_INT_STATUS                    (1 << 8)
-
-#define DC_HPD1_INT_CONTROL                             0x1808
-#define DC_HPD2_INT_CONTROL                             0x180B
-#define DC_HPD3_INT_CONTROL                             0x180E
-#define DC_HPD4_INT_CONTROL                             0x1811
-#define DC_HPD5_INT_CONTROL                             0x1814
-#define DC_HPD6_INT_CONTROL                             0x1817
-#       define DC_HPDx_INT_ACK                          (1 << 0)
-#       define DC_HPDx_INT_POLARITY                     (1 << 8)
-#       define DC_HPDx_INT_EN                           (1 << 16)
-#       define DC_HPDx_RX_INT_ACK                       (1 << 20)
-#       define DC_HPDx_RX_INT_EN                        (1 << 24)
-
-#define DC_HPD1_CONTROL                                   0x1809
-#define DC_HPD2_CONTROL                                   0x180C
-#define DC_HPD3_CONTROL                                   0x180F
-#define DC_HPD4_CONTROL                                   0x1812
-#define DC_HPD5_CONTROL                                   0x1815
-#define DC_HPD6_CONTROL                                   0x1818
-#       define DC_HPDx_CONNECTION_TIMER(x)                ((x) << 0)
-#       define DC_HPDx_RX_INT_TIMER(x)                    ((x) << 16)
-#       define DC_HPDx_EN                                 (1 << 28)
-
-#define DPG_PIPE_STUTTER_CONTROL                          0x1B35
-#       define STUTTER_ENABLE                             (1 << 0)
-
-/* 0x6e98, 0x7a98, 0x10698, 0x11298, 0x11e98, 0x12a98 */
-#define CRTC_STATUS_FRAME_COUNT                         0x1BA6
-
-/* Audio clocks */
-#define DCCG_AUDIO_DTO0_PHASE                           0x05b0
-#define DCCG_AUDIO_DTO0_MODULE                          0x05b4
-#define DCCG_AUDIO_DTO1_PHASE                           0x05c0
-#define DCCG_AUDIO_DTO1_MODULE                          0x05c4
-
-#define	GRBM_CNTL					0x2000
-#define		GRBM_READ_TIMEOUT(x)				((x) << 0)
-
-#define	GRBM_STATUS2					0x2002
-#define		RLC_RQ_PENDING 					(1 << 0)
-#define		RLC_BUSY 					(1 << 8)
-#define		TC_BUSY 					(1 << 9)
-
-#define	GRBM_STATUS					0x2004
-#define		CMDFIFO_AVAIL_MASK				0x0000000F
-#define		RING2_RQ_PENDING				(1 << 4)
-#define		SRBM_RQ_PENDING					(1 << 5)
-#define		RING1_RQ_PENDING				(1 << 6)
-#define		CF_RQ_PENDING					(1 << 7)
-#define		PF_RQ_PENDING					(1 << 8)
-#define		GDS_DMA_RQ_PENDING				(1 << 9)
-#define		GRBM_EE_BUSY					(1 << 10)
-#define		DB_CLEAN					(1 << 12)
-#define		CB_CLEAN					(1 << 13)
-#define		TA_BUSY 					(1 << 14)
-#define		GDS_BUSY 					(1 << 15)
-#define		VGT_BUSY					(1 << 17)
-#define		IA_BUSY_NO_DMA					(1 << 18)
-#define		IA_BUSY						(1 << 19)
-#define		SX_BUSY 					(1 << 20)
-#define		SPI_BUSY					(1 << 22)
-#define		BCI_BUSY					(1 << 23)
-#define		SC_BUSY 					(1 << 24)
-#define		PA_BUSY 					(1 << 25)
-#define		DB_BUSY 					(1 << 26)
-#define		CP_COHERENCY_BUSY      				(1 << 28)
-#define		CP_BUSY 					(1 << 29)
-#define		CB_BUSY 					(1 << 30)
-#define		GUI_ACTIVE					(1 << 31)
-#define	GRBM_STATUS_SE0					0x2005
-#define	GRBM_STATUS_SE1					0x2006
-#define		SE_DB_CLEAN					(1 << 1)
-#define		SE_CB_CLEAN					(1 << 2)
-#define		SE_BCI_BUSY					(1 << 22)
-#define		SE_VGT_BUSY					(1 << 23)
-#define		SE_PA_BUSY					(1 << 24)
-#define		SE_TA_BUSY					(1 << 25)
-#define		SE_SX_BUSY					(1 << 26)
-#define		SE_SPI_BUSY					(1 << 27)
-#define		SE_SC_BUSY					(1 << 29)
-#define		SE_DB_BUSY					(1 << 30)
-#define		SE_CB_BUSY					(1 << 31)
-
-#define GRBM_INT_CNTL                                   0x2018
-#       define RDERR_INT_ENABLE                         (1 << 0)
-#       define GUI_IDLE_INT_ENABLE                      (1 << 19)
-
-#define	CP_STRMOUT_CNTL					0x213F
-#define	SCRATCH_REG0					0x2140
-#define	SCRATCH_REG1					0x2141
-#define	SCRATCH_REG2					0x2142
-#define	SCRATCH_REG3					0x2143
-#define	SCRATCH_REG4					0x2144
-#define	SCRATCH_REG5					0x2145
-#define	SCRATCH_REG6					0x2146
-#define	SCRATCH_REG7					0x2147
-
-#define	SCRATCH_UMSK					0x2150
-#define	SCRATCH_ADDR					0x2151
-
-#define	CP_SEM_WAIT_TIMER				0x216F
-
-#define	CP_SEM_INCOMPLETE_TIMER_CNTL			0x2172
-
-#define CP_ME_CNTL					0x21B6
-#define		CP_CE_HALT					(1 << 24)
-#define		CP_PFP_HALT					(1 << 26)
-#define		CP_ME_HALT					(1 << 28)
-
-#define	CP_COHER_CNTL2					0x217A
-
-#define	CP_RB2_RPTR					0x21BE
-#define	CP_RB1_RPTR					0x21BF
-#define	CP_RB0_RPTR					0x21C0
-#define	CP_RB_WPTR_DELAY				0x21C1
-
-#define	CP_QUEUE_THRESHOLDS				0x21D8
-#define		ROQ_IB1_START(x)				((x) << 0)
-#define		ROQ_IB2_START(x)				((x) << 8)
-#define CP_MEQ_THRESHOLDS				0x21D9
-#define		MEQ1_START(x)				((x) << 0)
-#define		MEQ2_START(x)				((x) << 8)
-
-#define	CP_PERFMON_CNTL					0x21FF
-
 #define	VGT_VTX_VECT_EJECT_REG				0x222C
-
 #define	VGT_ESGS_RING_SIZE				0x2232
 #define	VGT_GSVS_RING_SIZE				0x2233
-
 #define	VGT_GS_VERTEX_REUSE				0x2235
-
 #define	VGT_PRIMITIVE_TYPE				0x2256
 #define	VGT_INDEX_TYPE					0x2257
-
 #define	VGT_NUM_INDICES					0x225C
 #define	VGT_NUM_INSTANCES				0x225D
-
 #define	VGT_TF_RING_SIZE				0x2262
-
 #define	VGT_HS_OFFCHIP_PARAM				0x226C
-
 #define	VGT_TF_MEMORY_BASE				0x226E
 
-#define	PA_CL_ENHANCE					0x2285
-#define		CLIP_VTX_REORDER_ENA				(1 << 0)
-#define		NUM_CLIP_SEQ(x)					((x) << 1)
-
-#define	PA_SU_LINE_STIPPLE_VALUE			0x2298
-
-#define	PA_SC_LINE_STIPPLE_STATE			0x22C4
-
-#define	PA_SC_FORCE_EOV_MAX_CNTS			0x22C9
-#define		FORCE_EOV_MAX_CLK_CNT(x)			((x) << 0)
-#define		FORCE_EOV_MAX_REZ_CNT(x)			((x) << 16)
-
-#define	PA_SC_FIFO_SIZE					0x22F3
-#define		SC_FRONTEND_PRIM_FIFO_SIZE(x)			((x) << 0)
-#define		SC_BACKEND_PRIM_FIFO_SIZE(x)			((x) << 6)
-#define		SC_HIZ_TILE_FIFO_SIZE(x)			((x) << 15)
-#define		SC_EARLYZ_TILE_FIFO_SIZE(x)			((x) << 23)
-
 #define	PA_SC_ENHANCE					0x22FC
 
-#define	SQ_CONFIG					0x2300
-
-#define	SQC_CACHES					0x2302
-
-#define SQ_POWER_THROTTLE                               0x2396
-#define		MIN_POWER(x)				((x) << 0)
-#define		MIN_POWER_MASK				(0x3fff << 0)
-#define		MIN_POWER_SHIFT				0
-#define		MAX_POWER(x)				((x) << 16)
-#define		MAX_POWER_MASK				(0x3fff << 16)
-#define		MAX_POWER_SHIFT				0
-#define SQ_POWER_THROTTLE2                              0x2397
-#define		MAX_POWER_DELTA(x)			((x) << 0)
-#define		MAX_POWER_DELTA_MASK			(0x3fff << 0)
-#define		MAX_POWER_DELTA_SHIFT			0
-#define		STI_SIZE(x)				((x) << 16)
-#define		STI_SIZE_MASK				(0x3ff << 16)
-#define		STI_SIZE_SHIFT				16
-#define		LTI_RATIO(x)				((x) << 27)
-#define		LTI_RATIO_MASK				(0xf << 27)
-#define		LTI_RATIO_SHIFT				27
-
-#define	SX_DEBUG_1					0x2418
-
-#define	SPI_STATIC_THREAD_MGMT_1			0x2438
-#define	SPI_STATIC_THREAD_MGMT_2			0x2439
-#define	SPI_STATIC_THREAD_MGMT_3			0x243A
-#define	SPI_PS_MAX_WAVE_ID				0x243B
-
-#define	SPI_CONFIG_CNTL					0x2440
-
-#define	SPI_CONFIG_CNTL_1				0x244F
-#define		VTX_DONE_DELAY(x)				((x) << 0)
-#define		INTERP_ONE_PRIM_PER_ROW				(1 << 4)
-
-#define	CGTS_TCC_DISABLE				0x2452
-#define	CGTS_USER_TCC_DISABLE				0x2453
-#define		TCC_DISABLE_MASK				0xFFFF0000
-#define		TCC_DISABLE_SHIFT				16
-#define	CGTS_SM_CTRL_REG				0x2454
-#define		OVERRIDE				(1 << 21)
-#define		LS_OVERRIDE				(1 << 22)
-
-#define	SPI_LB_CU_MASK					0x24D5
-
 #define	TA_CNTL_AUX					0x2542
 
-#define CC_RB_BACKEND_DISABLE				0x263D
-#define		BACKEND_DISABLE(x)     			((x) << 16)
-#define GB_ADDR_CONFIG  				0x263E
-#define		NUM_PIPES(x)				((x) << 0)
-#define		NUM_PIPES_MASK				0x00000007
-#define		NUM_PIPES_SHIFT				0
-#define		PIPE_INTERLEAVE_SIZE(x)			((x) << 4)
-#define		PIPE_INTERLEAVE_SIZE_MASK		0x00000070
-#define		PIPE_INTERLEAVE_SIZE_SHIFT		4
-#define		NUM_SHADER_ENGINES(x)			((x) << 12)
-#define		NUM_SHADER_ENGINES_MASK			0x00003000
-#define		NUM_SHADER_ENGINES_SHIFT		12
-#define		SHADER_ENGINE_TILE_SIZE(x)     		((x) << 16)
-#define		SHADER_ENGINE_TILE_SIZE_MASK		0x00070000
-#define		SHADER_ENGINE_TILE_SIZE_SHIFT		16
-#define		NUM_GPUS(x)     			((x) << 20)
-#define		NUM_GPUS_MASK				0x00700000
-#define		NUM_GPUS_SHIFT				20
-#define		MULTI_GPU_TILE_SIZE(x)     		((x) << 24)
-#define		MULTI_GPU_TILE_SIZE_MASK		0x03000000
-#define		MULTI_GPU_TILE_SIZE_SHIFT		24
-#define		ROW_SIZE(x)             		((x) << 28)
-#define		ROW_SIZE_MASK				0x30000000
-#define		ROW_SIZE_SHIFT				28
-
-#define	CB_PERFCOUNTER0_SELECT0				0x2688
-#define	CB_PERFCOUNTER0_SELECT1				0x2689
-#define	CB_PERFCOUNTER1_SELECT0				0x268A
-#define	CB_PERFCOUNTER1_SELECT1				0x268B
-#define	CB_PERFCOUNTER2_SELECT0				0x268C
-#define	CB_PERFCOUNTER2_SELECT1				0x268D
-#define	CB_PERFCOUNTER3_SELECT0				0x268E
-#define	CB_PERFCOUNTER3_SELECT1				0x268F
-
-#define	CB_CGTT_SCLK_CTRL				0x2698
-
-#define	TCP_CHAN_STEER_LO				0x2B03
-#define	TCP_CHAN_STEER_HI				0x2B94
-
-#define	CP_RB0_BASE					0x3040
-#define	CP_RB0_CNTL					0x3041
-#define		RB_BUFSZ(x)					((x) << 0)
-#define		RB_BLKSZ(x)					((x) << 8)
-#define		BUF_SWAP_32BIT					(2 << 16)
-#define		RB_NO_UPDATE					(1 << 27)
-#define		RB_RPTR_WR_ENA					(1 << 31)
-
-#define	CP_RB0_RPTR_ADDR				0x3043
-#define	CP_RB0_RPTR_ADDR_HI				0x3044
-#define	CP_RB0_WPTR					0x3045
-
-#define	CP_PFP_UCODE_ADDR				0x3054
-#define	CP_PFP_UCODE_DATA				0x3055
-#define	CP_ME_RAM_RADDR					0x3056
-#define	CP_ME_RAM_WADDR					0x3057
-#define	CP_ME_RAM_DATA					0x3058
-
-#define	CP_CE_UCODE_ADDR				0x305A
-#define	CP_CE_UCODE_DATA				0x305B
-
-#define	CP_RB1_BASE					0x3060
-#define	CP_RB1_CNTL					0x3061
-#define	CP_RB1_RPTR_ADDR				0x3062
-#define	CP_RB1_RPTR_ADDR_HI				0x3063
-#define	CP_RB1_WPTR					0x3064
-#define	CP_RB2_BASE					0x3065
-#define	CP_RB2_CNTL					0x3066
-#define	CP_RB2_RPTR_ADDR				0x3067
-#define	CP_RB2_RPTR_ADDR_HI				0x3068
-#define	CP_RB2_WPTR					0x3069
-#define CP_INT_CNTL_RING0                               0x306A
-#define CP_INT_CNTL_RING1                               0x306B
-#define CP_INT_CNTL_RING2                               0x306C
-#       define CNTX_BUSY_INT_ENABLE                     (1 << 19)
-#       define CNTX_EMPTY_INT_ENABLE                    (1 << 20)
-#       define WAIT_MEM_SEM_INT_ENABLE                  (1 << 21)
-#       define TIME_STAMP_INT_ENABLE                    (1 << 26)
-#       define CP_RINGID2_INT_ENABLE                    (1 << 29)
-#       define CP_RINGID1_INT_ENABLE                    (1 << 30)
-#       define CP_RINGID0_INT_ENABLE                    (1 << 31)
-#define CP_INT_STATUS_RING0                             0x306D
-#define CP_INT_STATUS_RING1                             0x306E
-#define CP_INT_STATUS_RING2                             0x306F
-#       define WAIT_MEM_SEM_INT_STAT                    (1 << 21)
-#       define TIME_STAMP_INT_STAT                      (1 << 26)
-#       define CP_RINGID2_INT_STAT                      (1 << 29)
-#       define CP_RINGID1_INT_STAT                      (1 << 30)
-#       define CP_RINGID0_INT_STAT                      (1 << 31)
-
 // #define PA_SC_RASTER_CONFIG                             0xA0D4
 #	define RB_XSEL2(x)				((x) << 4)
 #	define RB_XSEL2_MASK				(0x3 << 4)
@@ -1185,171 +311,14 @@
 #	define SE_YSEL(x)				((x) << 28)
 #	define SE_YSEL_MASK				(0x3 << 28)
 
-/* PIF PHY0 registers idx/data 0x8/0xc */
-#define PB0_PIF_CNTL                                      0x10
-#       define LS2_EXIT_TIME(x)                           ((x) << 17)
-#       define LS2_EXIT_TIME_MASK                         (0x7 << 17)
-#       define LS2_EXIT_TIME_SHIFT                        17
-#define PB0_PIF_PAIRING                                   0x11
-#       define MULTI_PIF                                  (1 << 25)
-#define PB0_PIF_PWRDOWN_0                                 0x12
-#       define PLL_POWER_STATE_IN_TXS2_0(x)               ((x) << 7)
-#       define PLL_POWER_STATE_IN_TXS2_0_MASK             (0x7 << 7)
-#       define PLL_POWER_STATE_IN_TXS2_0_SHIFT            7
-#       define PLL_POWER_STATE_IN_OFF_0(x)                ((x) << 10)
-#       define PLL_POWER_STATE_IN_OFF_0_MASK              (0x7 << 10)
-#       define PLL_POWER_STATE_IN_OFF_0_SHIFT             10
-#       define PLL_RAMP_UP_TIME_0(x)                      ((x) << 24)
-#       define PLL_RAMP_UP_TIME_0_MASK                    (0x7 << 24)
-#       define PLL_RAMP_UP_TIME_0_SHIFT                   24
-#define PB0_PIF_PWRDOWN_1                                 0x13
-#       define PLL_POWER_STATE_IN_TXS2_1(x)               ((x) << 7)
-#       define PLL_POWER_STATE_IN_TXS2_1_MASK             (0x7 << 7)
-#       define PLL_POWER_STATE_IN_TXS2_1_SHIFT            7
-#       define PLL_POWER_STATE_IN_OFF_1(x)                ((x) << 10)
-#       define PLL_POWER_STATE_IN_OFF_1_MASK              (0x7 << 10)
-#       define PLL_POWER_STATE_IN_OFF_1_SHIFT             10
-#       define PLL_RAMP_UP_TIME_1(x)                      ((x) << 24)
-#       define PLL_RAMP_UP_TIME_1_MASK                    (0x7 << 24)
-#       define PLL_RAMP_UP_TIME_1_SHIFT                   24
-
-#define PB0_PIF_PWRDOWN_2                                 0x17
-#       define PLL_POWER_STATE_IN_TXS2_2(x)               ((x) << 7)
-#       define PLL_POWER_STATE_IN_TXS2_2_MASK             (0x7 << 7)
-#       define PLL_POWER_STATE_IN_TXS2_2_SHIFT            7
-#       define PLL_POWER_STATE_IN_OFF_2(x)                ((x) << 10)
-#       define PLL_POWER_STATE_IN_OFF_2_MASK              (0x7 << 10)
-#       define PLL_POWER_STATE_IN_OFF_2_SHIFT             10
-#       define PLL_RAMP_UP_TIME_2(x)                      ((x) << 24)
-#       define PLL_RAMP_UP_TIME_2_MASK                    (0x7 << 24)
-#       define PLL_RAMP_UP_TIME_2_SHIFT                   24
-#define PB0_PIF_PWRDOWN_3                                 0x18
-#       define PLL_POWER_STATE_IN_TXS2_3(x)               ((x) << 7)
-#       define PLL_POWER_STATE_IN_TXS2_3_MASK             (0x7 << 7)
-#       define PLL_POWER_STATE_IN_TXS2_3_SHIFT            7
-#       define PLL_POWER_STATE_IN_OFF_3(x)                ((x) << 10)
-#       define PLL_POWER_STATE_IN_OFF_3_MASK              (0x7 << 10)
-#       define PLL_POWER_STATE_IN_OFF_3_SHIFT             10
-#       define PLL_RAMP_UP_TIME_3(x)                      ((x) << 24)
-#       define PLL_RAMP_UP_TIME_3_MASK                    (0x7 << 24)
-#       define PLL_RAMP_UP_TIME_3_SHIFT                   24
-/* PIF PHY1 registers idx/data 0x10/0x14 */
-#define PB1_PIF_CNTL                                      0x10
-#define PB1_PIF_PAIRING                                   0x11
-#define PB1_PIF_PWRDOWN_0                                 0x12
-#define PB1_PIF_PWRDOWN_1                                 0x13
-
-#define PB1_PIF_PWRDOWN_2                                 0x17
-#define PB1_PIF_PWRDOWN_3                                 0x18
-/* PCIE registers idx/data 0x30/0x34 */
-#define PCIE_CNTL2                                        0x1c /* PCIE */
-#       define SLV_MEM_LS_EN                              (1 << 16)
-#       define SLV_MEM_AGGRESSIVE_LS_EN                   (1 << 17)
-#       define MST_MEM_LS_EN                              (1 << 18)
-#       define REPLAY_MEM_LS_EN                           (1 << 19)
-#define PCIE_LC_STATUS1                                   0x28 /* PCIE */
-#       define LC_REVERSE_RCVR                            (1 << 0)
-#       define LC_REVERSE_XMIT                            (1 << 1)
-#       define LC_OPERATING_LINK_WIDTH_MASK               (0x7 << 2)
-#       define LC_OPERATING_LINK_WIDTH_SHIFT              2
-#       define LC_DETECTED_LINK_WIDTH_MASK                (0x7 << 5)
-#       define LC_DETECTED_LINK_WIDTH_SHIFT               5
-
-#define PCIE_P_CNTL                                       0x40 /* PCIE */
-#       define P_IGNORE_EDB_ERR                           (1 << 6)
-
 /* PCIE PORT registers idx/data 0x38/0x3c */
-#define PCIE_LC_CNTL                                      0xa0
-#       define LC_L0S_INACTIVITY(x)                       ((x) << 8)
-#       define LC_L0S_INACTIVITY_MASK                     (0xf << 8)
-#       define LC_L0S_INACTIVITY_SHIFT                    8
-#       define LC_L1_INACTIVITY(x)                        ((x) << 12)
-#       define LC_L1_INACTIVITY_MASK                      (0xf << 12)
-#       define LC_L1_INACTIVITY_SHIFT                     12
-#       define LC_PMI_TO_L1_DIS                           (1 << 16)
-#       define LC_ASPM_TO_L1_DIS                          (1 << 24)
-#define PCIE_LC_LINK_WIDTH_CNTL                           0xa2 /* PCIE_P */
-#       define LC_LINK_WIDTH_SHIFT                        0
-#       define LC_LINK_WIDTH_MASK                         0x7
+// #define PCIE_LC_LINK_WIDTH_CNTL                           0xa2 /* PCIE_P */
 #       define LC_LINK_WIDTH_X0                           0
 #       define LC_LINK_WIDTH_X1                           1
 #       define LC_LINK_WIDTH_X2                           2
 #       define LC_LINK_WIDTH_X4                           3
 #       define LC_LINK_WIDTH_X8                           4
 #       define LC_LINK_WIDTH_X16                          6
-#       define LC_LINK_WIDTH_RD_SHIFT                     4
-#       define LC_LINK_WIDTH_RD_MASK                      0x70
-#       define LC_RECONFIG_ARC_MISSING_ESCAPE             (1 << 7)
-#       define LC_RECONFIG_NOW                            (1 << 8)
-#       define LC_RENEGOTIATION_SUPPORT                   (1 << 9)
-#       define LC_RENEGOTIATE_EN                          (1 << 10)
-#       define LC_SHORT_RECONFIG_EN                       (1 << 11)
-#       define LC_UPCONFIGURE_SUPPORT                     (1 << 12)
-#       define LC_UPCONFIGURE_DIS                         (1 << 13)
-#       define LC_DYN_LANES_PWR_STATE(x)                  ((x) << 21)
-#       define LC_DYN_LANES_PWR_STATE_MASK                (0x3 << 21)
-#       define LC_DYN_LANES_PWR_STATE_SHIFT               21
-#define PCIE_LC_N_FTS_CNTL                                0xa3 /* PCIE_P */
-#       define LC_XMIT_N_FTS(x)                           ((x) << 0)
-#       define LC_XMIT_N_FTS_MASK                         (0xff << 0)
-#       define LC_XMIT_N_FTS_SHIFT                        0
-#       define LC_XMIT_N_FTS_OVERRIDE_EN                  (1 << 8)
-#       define LC_N_FTS_MASK                              (0xff << 24)
-#define PCIE_LC_SPEED_CNTL                                0xa4 /* PCIE_P */
-#       define LC_GEN2_EN_STRAP                           (1 << 0)
-#       define LC_GEN3_EN_STRAP                           (1 << 1)
-#       define LC_TARGET_LINK_SPEED_OVERRIDE_EN           (1 << 2)
-#       define LC_TARGET_LINK_SPEED_OVERRIDE_MASK         (0x3 << 3)
-#       define LC_TARGET_LINK_SPEED_OVERRIDE_SHIFT        3
-#       define LC_FORCE_EN_SW_SPEED_CHANGE                (1 << 5)
-#       define LC_FORCE_DIS_SW_SPEED_CHANGE               (1 << 6)
-#       define LC_FORCE_EN_HW_SPEED_CHANGE                (1 << 7)
-#       define LC_FORCE_DIS_HW_SPEED_CHANGE               (1 << 8)
-#       define LC_INITIATE_LINK_SPEED_CHANGE              (1 << 9)
-#       define LC_SPEED_CHANGE_ATTEMPTS_ALLOWED_MASK      (0x3 << 10)
-#       define LC_SPEED_CHANGE_ATTEMPTS_ALLOWED_SHIFT     10
-#       define LC_CURRENT_DATA_RATE_MASK                  (0x3 << 13) /* 0/1/2 = gen1/2/3 */
-#       define LC_CURRENT_DATA_RATE_SHIFT                 13
-#       define LC_CLR_FAILED_SPD_CHANGE_CNT               (1 << 16)
-#       define LC_OTHER_SIDE_EVER_SENT_GEN2               (1 << 18)
-#       define LC_OTHER_SIDE_SUPPORTS_GEN2                (1 << 19)
-#       define LC_OTHER_SIDE_EVER_SENT_GEN3               (1 << 20)
-#       define LC_OTHER_SIDE_SUPPORTS_GEN3                (1 << 21)
-
-#define PCIE_LC_CNTL2                                     0xb1
-#       define LC_ALLOW_PDWN_IN_L1                        (1 << 17)
-#       define LC_ALLOW_PDWN_IN_L23                       (1 << 18)
-
-#define PCIE_LC_CNTL3                                     0xb5 /* PCIE_P */
-#       define LC_GO_TO_RECOVERY                          (1 << 30)
-#define PCIE_LC_CNTL4                                     0xb6 /* PCIE_P */
-#       define LC_REDO_EQ                                 (1 << 5)
-#       define LC_SET_QUIESCE                             (1 << 13)
-
-/*
- * UVD
- */
-#define UVD_UDEC_ADDR_CONFIG				0x3bd3
-#define UVD_UDEC_DB_ADDR_CONFIG				0x3bd4
-#define UVD_UDEC_DBW_ADDR_CONFIG			0x3bd5
-#define UVD_RBC_RB_RPTR					0x3da4
-#define UVD_RBC_RB_WPTR					0x3da5
-#define UVD_STATUS					0x3daf
-
-#define	UVD_CGC_CTRL					0x3dc2
-#	define DCM					(1 << 0)
-#	define CG_DT(x)					((x) << 2)
-#	define CG_DT_MASK				(0xf << 2)
-#	define CLK_OD(x)				((x) << 6)
-#	define CLK_OD_MASK				(0x1f << 6)
-
- /* UVD CTX indirect */
-#define	UVD_CGC_MEM_CTRL				0xC0
-#define	UVD_CGC_CTRL2					0xC1
-#	define DYN_OR_EN				(1 << 0)
-#	define DYN_RR_EN				(1 << 1)
-#	define G_DIV_ID(x)				((x) << 2)
-#	define G_DIV_ID_MASK				(0x7 << 2)
 
 /*
  * PM4
@@ -1583,45 +552,7 @@
 /* ASYNC DMA - first instance at 0xd000, second at 0xd800 */
 #define DMA0_REGISTER_OFFSET                              0x0 /* not a register */
 #define DMA1_REGISTER_OFFSET                              0x200 /* not a register */
-
-#define DMA_RB_CNTL                                       0x3400
-#       define DMA_RB_ENABLE                              (1 << 0)
-#       define DMA_RB_SIZE(x)                             ((x) << 1) /* log2 */
-#       define DMA_RB_SWAP_ENABLE                         (1 << 9) /* 8IN32 */
-#       define DMA_RPTR_WRITEBACK_ENABLE                  (1 << 12)
-#       define DMA_RPTR_WRITEBACK_SWAP_ENABLE             (1 << 13)  /* 8IN32 */
-#       define DMA_RPTR_WRITEBACK_TIMER(x)                ((x) << 16) /* log2 */
-#define DMA_RB_BASE                                       0x3401
-#define DMA_RB_RPTR                                       0x3402
-#define DMA_RB_WPTR                                       0x3403
-
-#define DMA_RB_RPTR_ADDR_HI                               0x3407
-#define DMA_RB_RPTR_ADDR_LO                               0x3408
-
-#define DMA_IB_CNTL                                       0x3409
-#       define DMA_IB_ENABLE                              (1 << 0)
-#       define DMA_IB_SWAP_ENABLE                         (1 << 4)
-#       define CMD_VMID_FORCE                             (1 << 31)
-#define DMA_IB_RPTR                                       0x340a
-#define DMA_CNTL                                          0x340b
-#       define TRAP_ENABLE                                (1 << 0)
-#       define SEM_INCOMPLETE_INT_ENABLE                  (1 << 1)
-#       define SEM_WAIT_INT_ENABLE                        (1 << 2)
-#       define DATA_SWAP_ENABLE                           (1 << 3)
-#       define FENCE_SWAP_ENABLE                          (1 << 4)
-#       define CTXEMPTY_INT_ENABLE                        (1 << 28)
-#define DMA_STATUS_REG                                    0x340d
-#       define DMA_IDLE                                   (1 << 0)
-#define DMA_TILING_CONFIG  				  0x342e
-
-#define	DMA_POWER_CNTL					0x342f
-#       define MEM_POWER_OVERRIDE                       (1 << 8)
-#define	DMA_CLK_CTRL					0x3430
-
-#define	DMA_PG						0x3435
-#	define PG_CNTL_ENABLE				(1 << 0)
-#define	DMA_PGFSM_CONFIG				0x3436
-#define	DMA_PGFSM_WRITE					0x3437
+#define SDMA_MAX_INSTANCE 2
 
 #define DMA_PACKET(cmd, b, t, s, n)	((((cmd) & 0xF) << 28) |	\
 					 (((b) & 0x1) << 26) |		\
@@ -1650,6 +581,7 @@
 #define	DMA_PACKET_POLL_REG_MEM				  0xe
 #define	DMA_PACKET_NOP					  0xf
 
+/* VCE */
 #define VCE_STATUS					0x20004
 #define VCE_VCPU_CNTL					0x20014
 #define		VCE_CLK_EN				(1 << 0)
@@ -1726,378 +658,118 @@
 
 #define CURSOR_WIDTH 64
 #define CURSOR_HEIGHT 64
-#define AMDGPU_MM_INDEX		        0x0000
-#define AMDGPU_MM_DATA		        0x0001
-
-#define VERDE_NUM_CRTC 6
-#define	BLACKOUT_MODE_MASK			0x00000007
-#define	VGA_RENDER_CONTROL			0xC0
-#define R_000300_VGA_RENDER_CONTROL             0xC0
-#define C_000300_VGA_VSTATUS_CNTL               0xFFFCFFFF
-#define EVERGREEN_CRTC_STATUS                   0x1BA3
-#define EVERGREEN_CRTC_V_BLANK                  (1 << 0)
-#define EVERGREEN_CRTC_STATUS_POSITION          0x1BA4
-/* CRTC blocks at 0x6df0, 0x79f0, 0x105f0, 0x111f0, 0x11df0, 0x129f0 */
-#define EVERGREEN_CRTC_V_BLANK_START_END                0x1b8d
-#define EVERGREEN_CRTC_CONTROL                          0x1b9c
-#define EVERGREEN_CRTC_MASTER_EN                 (1 << 0)
-#define EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE (1 << 24)
-#define EVERGREEN_CRTC_BLANK_CONTROL                    0x1b9d
-#define EVERGREEN_CRTC_BLANK_DATA_EN             (1 << 8)
-#define EVERGREEN_CRTC_V_BLANK                   (1 << 0)
-#define EVERGREEN_CRTC_STATUS_HV_COUNT                  0x1ba8
-#define EVERGREEN_CRTC_UPDATE_LOCK                      0x1bb5
-#define EVERGREEN_MASTER_UPDATE_LOCK                    0x1bbd
-#define EVERGREEN_MASTER_UPDATE_MODE                    0x1bbe
-#define EVERGREEN_GRPH_UPDATE_LOCK               (1 << 16)
-#define EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH     0x1a07
-#define EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH   0x1a08
-#define EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS          0x1a04
-#define EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS        0x1a05
-#define EVERGREEN_GRPH_UPDATE                           0x1a11
-#define EVERGREEN_VGA_MEMORY_BASE_ADDRESS               0xc4
-#define EVERGREEN_VGA_MEMORY_BASE_ADDRESS_HIGH          0xc9
-#define EVERGREEN_GRPH_SURFACE_UPDATE_PENDING    (1 << 2)
-
-#define EVERGREEN_DATA_FORMAT                           0x1ac0
-#       define EVERGREEN_INTERLEAVE_EN                  (1 << 0)
-
-#define R600_D1GRPH_ARRAY_MODE_LINEAR_GENERAL            (0 << 20)
-#define R600_D1GRPH_ARRAY_MODE_LINEAR_ALIGNED            (1 << 20)
-#define R600_D1GRPH_ARRAY_MODE_1D_TILED_THIN1            (2 << 20)
-#define R600_D1GRPH_ARRAY_MODE_2D_TILED_THIN1            (4 << 20)
-
-#define R700_D1GRPH_PRIMARY_SURFACE_ADDRESS_HIGH                0x1a45
-#define R700_D2GRPH_PRIMARY_SURFACE_ADDRESS_HIGH                0x1845
-
-#define R700_D2GRPH_SECONDARY_SURFACE_ADDRESS_HIGH              0x1847
-#define R700_D1GRPH_SECONDARY_SURFACE_ADDRESS_HIGH              0x1a47
-
-#define R600_D1GRPH_SWAP_CONTROL                               0x1843
-#define R600_D1GRPH_SWAP_ENDIAN_NONE                    (0 << 0)
-#define R600_D1GRPH_SWAP_ENDIAN_16BIT                   (1 << 0)
-#define R600_D1GRPH_SWAP_ENDIAN_32BIT                   (2 << 0)
-#define R600_D1GRPH_SWAP_ENDIAN_64BIT                   (3 << 0)
-
-#define AVIVO_D1VGA_CONTROL					0x00cc
-#       define AVIVO_DVGA_CONTROL_MODE_ENABLE            (1 << 0)
-#       define AVIVO_DVGA_CONTROL_TIMING_SELECT          (1 << 8)
-#       define AVIVO_DVGA_CONTROL_SYNC_POLARITY_SELECT   (1 << 9)
-#       define AVIVO_DVGA_CONTROL_OVERSCAN_TIMING_SELECT (1 << 10)
-#       define AVIVO_DVGA_CONTROL_OVERSCAN_COLOR_EN      (1 << 16)
-#       define AVIVO_DVGA_CONTROL_ROTATE                 (1 << 24)
-#define AVIVO_D2VGA_CONTROL					0x00ce
-
-#define R600_BUS_CNTL                                           0x1508
-#       define R600_BIOS_ROM_DIS                                (1 << 1)
+
 
 #define R600_ROM_CNTL                              0x580
 #       define R600_SCK_OVERWRITE                  (1 << 1)
 #       define R600_SCK_PRESCALE_CRYSTAL_CLK_SHIFT 28
 #       define R600_SCK_PRESCALE_CRYSTAL_CLK_MASK  (0xf << 28)
 
-#define FMT_BIT_DEPTH_CONTROL                0x1bf2
-#define FMT_TRUNCATE_EN               (1 << 0)
-#define FMT_TRUNCATE_DEPTH            (1 << 4)
-#define FMT_SPATIAL_DITHER_EN         (1 << 8)
-#define FMT_SPATIAL_DITHER_MODE(x)    ((x) << 9)
-#define FMT_SPATIAL_DITHER_DEPTH      (1 << 12)
-#define FMT_FRAME_RANDOM_ENABLE       (1 << 13)
-#define FMT_RGB_RANDOM_ENABLE         (1 << 14)
-#define FMT_HIGHPASS_RANDOM_ENABLE    (1 << 15)
-#define FMT_TEMPORAL_DITHER_EN        (1 << 16)
-#define FMT_TEMPORAL_DITHER_DEPTH     (1 << 20)
-#define FMT_TEMPORAL_DITHER_OFFSET(x) ((x) << 21)
-#define FMT_TEMPORAL_LEVEL            (1 << 24)
-#define FMT_TEMPORAL_DITHER_RESET     (1 << 25)
-#define FMT_25FRC_SEL(x)              ((x) << 26)
-#define FMT_50FRC_SEL(x)              ((x) << 28)
-#define FMT_75FRC_SEL(x)              ((x) << 30)
-
-#define EVERGREEN_DC_LUT_CONTROL                        0x1a80
-#define EVERGREEN_DC_LUT_BLACK_OFFSET_BLUE              0x1a81
-#define EVERGREEN_DC_LUT_BLACK_OFFSET_GREEN             0x1a82
-#define EVERGREEN_DC_LUT_BLACK_OFFSET_RED               0x1a83
-#define EVERGREEN_DC_LUT_WHITE_OFFSET_BLUE              0x1a84
-#define EVERGREEN_DC_LUT_WHITE_OFFSET_GREEN             0x1a85
-#define EVERGREEN_DC_LUT_WHITE_OFFSET_RED               0x1a86
-#define EVERGREEN_DC_LUT_30_COLOR                       0x1a7c
-#define EVERGREEN_DC_LUT_RW_INDEX                       0x1a79
-#define EVERGREEN_DC_LUT_WRITE_EN_MASK                  0x1a7e
-#define EVERGREEN_DC_LUT_RW_MODE                        0x1a78
-
-#define EVERGREEN_GRPH_ENABLE                           0x1a00
-#define EVERGREEN_GRPH_CONTROL                          0x1a01
-#define EVERGREEN_GRPH_DEPTH(x)                  (((x) & 0x3) << 0)
-#define EVERGREEN_GRPH_DEPTH_8BPP                0
-#define EVERGREEN_GRPH_DEPTH_16BPP               1
-#define EVERGREEN_GRPH_DEPTH_32BPP               2
-#define EVERGREEN_GRPH_NUM_BANKS(x)              (((x) & 0x3) << 2)
-#define EVERGREEN_ADDR_SURF_2_BANK               0
-#define EVERGREEN_ADDR_SURF_4_BANK               1
-#define EVERGREEN_ADDR_SURF_8_BANK               2
-#define EVERGREEN_ADDR_SURF_16_BANK              3
-#define EVERGREEN_GRPH_Z(x)                      (((x) & 0x3) << 4)
-#define EVERGREEN_GRPH_BANK_WIDTH(x)             (((x) & 0x3) << 6)
-#define EVERGREEN_ADDR_SURF_BANK_WIDTH_1         0
-#define EVERGREEN_ADDR_SURF_BANK_WIDTH_2         1
-#define EVERGREEN_ADDR_SURF_BANK_WIDTH_4         2
-#define EVERGREEN_ADDR_SURF_BANK_WIDTH_8         3
-#define EVERGREEN_GRPH_FORMAT(x)                 (((x) & 0x7) << 8)
-
-#define EVERGREEN_GRPH_FORMAT_INDEXED            0
-#define EVERGREEN_GRPH_FORMAT_ARGB1555           0
-#define EVERGREEN_GRPH_FORMAT_ARGB565            1
-#define EVERGREEN_GRPH_FORMAT_ARGB4444           2
-#define EVERGREEN_GRPH_FORMAT_AI88               3
-#define EVERGREEN_GRPH_FORMAT_MONO16             4
-#define EVERGREEN_GRPH_FORMAT_BGRA5551           5
+#define GRPH_ARRAY_LINEAR_GENERAL      0
+#define GRPH_ARRAY_LINEAR_ALIGNED      1
+#define GRPH_ARRAY_1D_TILED_THIN1      2
+#define GRPH_ARRAY_2D_TILED_THIN1      4
+
+#define ES_AND_GS_AUTO       3
+#define BUF_SWAP_32BIT       (2 << 16)
+
+#define GRPH_DEPTH_8BPP                0
+#define GRPH_DEPTH_16BPP               1
+#define GRPH_DEPTH_32BPP               2
+
+/* 8 BPP */
+#define GRPH_FORMAT_INDEXED            0
+
+/* 16 BPP */
+#define GRPH_FORMAT_ARGB1555           0
+#define GRPH_FORMAT_ARGB565            1
+#define GRPH_FORMAT_ARGB4444           2
+#define GRPH_FORMAT_AI88               3
+#define GRPH_FORMAT_MONO16             4
+#define GRPH_FORMAT_BGRA5551           5
 
 /* 32 BPP */
-#define EVERGREEN_GRPH_FORMAT_ARGB8888           0
-#define EVERGREEN_GRPH_FORMAT_ARGB2101010        1
-#define EVERGREEN_GRPH_FORMAT_32BPP_DIG          2
-#define EVERGREEN_GRPH_FORMAT_8B_ARGB2101010     3
-#define EVERGREEN_GRPH_FORMAT_BGRA1010102        4
-#define EVERGREEN_GRPH_FORMAT_8B_BGRA1010102     5
-#define EVERGREEN_GRPH_FORMAT_RGB111110          6
-#define EVERGREEN_GRPH_FORMAT_BGR101111          7
-#define EVERGREEN_GRPH_BANK_HEIGHT(x)            (((x) & 0x3) << 11)
-#define EVERGREEN_ADDR_SURF_BANK_HEIGHT_1        0
-#define EVERGREEN_ADDR_SURF_BANK_HEIGHT_2        1
-#define EVERGREEN_ADDR_SURF_BANK_HEIGHT_4        2
-#define EVERGREEN_ADDR_SURF_BANK_HEIGHT_8        3
-#define EVERGREEN_GRPH_TILE_SPLIT(x)             (((x) & 0x7) << 13)
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_64B       0
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_128B      1
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_256B      2
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_512B      3
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_1KB       4
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_2KB       5
-#define EVERGREEN_ADDR_SURF_TILE_SPLIT_4KB       6
-#define EVERGREEN_GRPH_MACRO_TILE_ASPECT(x)      (((x) & 0x3) << 18)
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_1  0
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_2  1
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_4  2
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_8  3
-#define EVERGREEN_GRPH_ARRAY_MODE(x)             (((x) & 0x7) << 20)
-#define EVERGREEN_GRPH_ARRAY_LINEAR_GENERAL      0
-#define EVERGREEN_GRPH_ARRAY_LINEAR_ALIGNED      1
-#define EVERGREEN_GRPH_ARRAY_1D_TILED_THIN1      2
-#define EVERGREEN_GRPH_ARRAY_2D_TILED_THIN1      4
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_1  0
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_2  1
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_4  2
-#define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_8  3
-
-#define EVERGREEN_GRPH_SWAP_CONTROL                     0x1a03
-#define EVERGREEN_GRPH_ENDIAN_SWAP(x)            (((x) & 0x3) << 0)
-#       define EVERGREEN_GRPH_ENDIAN_NONE               0
-#       define EVERGREEN_GRPH_ENDIAN_8IN16              1
-#       define EVERGREEN_GRPH_ENDIAN_8IN32              2
-#       define EVERGREEN_GRPH_ENDIAN_8IN64              3
-#define EVERGREEN_GRPH_RED_CROSSBAR(x)           (((x) & 0x3) << 4)
-#       define EVERGREEN_GRPH_RED_SEL_R                 0
-#       define EVERGREEN_GRPH_RED_SEL_G                 1
-#       define EVERGREEN_GRPH_RED_SEL_B                 2
-#       define EVERGREEN_GRPH_RED_SEL_A                 3
-#define EVERGREEN_GRPH_GREEN_CROSSBAR(x)         (((x) & 0x3) << 6)
-#       define EVERGREEN_GRPH_GREEN_SEL_G               0
-#       define EVERGREEN_GRPH_GREEN_SEL_B               1
-#       define EVERGREEN_GRPH_GREEN_SEL_A               2
-#       define EVERGREEN_GRPH_GREEN_SEL_R               3
-#define EVERGREEN_GRPH_BLUE_CROSSBAR(x)          (((x) & 0x3) << 8)
-#       define EVERGREEN_GRPH_BLUE_SEL_B                0
-#       define EVERGREEN_GRPH_BLUE_SEL_A                1
-#       define EVERGREEN_GRPH_BLUE_SEL_R                2
-#       define EVERGREEN_GRPH_BLUE_SEL_G                3
-#define EVERGREEN_GRPH_ALPHA_CROSSBAR(x)         (((x) & 0x3) << 10)
-#       define EVERGREEN_GRPH_ALPHA_SEL_A               0
-#       define EVERGREEN_GRPH_ALPHA_SEL_R               1
-#       define EVERGREEN_GRPH_ALPHA_SEL_G               2
-#       define EVERGREEN_GRPH_ALPHA_SEL_B               3
-
-#define EVERGREEN_D3VGA_CONTROL                         0xf8
-#define EVERGREEN_D4VGA_CONTROL                         0xf9
-#define EVERGREEN_D5VGA_CONTROL                         0xfa
-#define EVERGREEN_D6VGA_CONTROL                         0xfb
-
-#define EVERGREEN_GRPH_SURFACE_ADDRESS_MASK      0xffffff00
-
-#define EVERGREEN_GRPH_LUT_10BIT_BYPASS_CONTROL         0x1a02
-#define EVERGREEN_LUT_10BIT_BYPASS_EN            (1 << 8)
-
-#define EVERGREEN_GRPH_PITCH                            0x1a06
-#define EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH     0x1a07
-#define EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH   0x1a08
-#define EVERGREEN_GRPH_SURFACE_OFFSET_X                 0x1a09
-#define EVERGREEN_GRPH_SURFACE_OFFSET_Y                 0x1a0a
-#define EVERGREEN_GRPH_X_START                          0x1a0b
-#define EVERGREEN_GRPH_Y_START                          0x1a0c
-#define EVERGREEN_GRPH_X_END                            0x1a0d
-#define EVERGREEN_GRPH_Y_END                            0x1a0e
-#define EVERGREEN_GRPH_UPDATE                           0x1a11
-#define EVERGREEN_GRPH_SURFACE_UPDATE_PENDING    (1 << 2)
-#define EVERGREEN_GRPH_UPDATE_LOCK               (1 << 16)
-#define EVERGREEN_GRPH_FLIP_CONTROL                     0x1a12
-#define EVERGREEN_GRPH_SURFACE_UPDATE_H_RETRACE_EN (1 << 0)
-
-#define EVERGREEN_VIEWPORT_START                        0x1b5c
-#define EVERGREEN_VIEWPORT_SIZE                         0x1b5d
-#define EVERGREEN_DESKTOP_HEIGHT                        0x1ac1
-
-/* CUR blocks at 0x6998, 0x7598, 0x10198, 0x10d98, 0x11998, 0x12598 */
-#define EVERGREEN_CUR_CONTROL                           0x1a66
-#       define EVERGREEN_CURSOR_EN                      (1 << 0)
-#       define EVERGREEN_CURSOR_MODE(x)                 (((x) & 0x3) << 8)
-#       define EVERGREEN_CURSOR_MONO                    0
-#       define EVERGREEN_CURSOR_24_1                    1
-#       define EVERGREEN_CURSOR_24_8_PRE_MULT           2
-#       define EVERGREEN_CURSOR_24_8_UNPRE_MULT         3
-#       define EVERGREEN_CURSOR_2X_MAGNIFY              (1 << 16)
-#       define EVERGREEN_CURSOR_FORCE_MC_ON             (1 << 20)
-#       define EVERGREEN_CURSOR_URGENT_CONTROL(x)       (((x) & 0x7) << 24)
-#       define EVERGREEN_CURSOR_URGENT_ALWAYS           0
-#       define EVERGREEN_CURSOR_URGENT_1_8              1
-#       define EVERGREEN_CURSOR_URGENT_1_4              2
-#       define EVERGREEN_CURSOR_URGENT_3_8              3
-#       define EVERGREEN_CURSOR_URGENT_1_2              4
-#define EVERGREEN_CUR_SURFACE_ADDRESS                   0x1a67
-#       define EVERGREEN_CUR_SURFACE_ADDRESS_MASK       0xfffff000
-#define EVERGREEN_CUR_SIZE                              0x1a68
-#define EVERGREEN_CUR_SURFACE_ADDRESS_HIGH              0x1a69
-#define EVERGREEN_CUR_POSITION                          0x1a6a
-#define EVERGREEN_CUR_HOT_SPOT                          0x1a6b
-#define EVERGREEN_CUR_COLOR1                            0x1a6c
-#define EVERGREEN_CUR_COLOR2                            0x1a6d
-#define EVERGREEN_CUR_UPDATE                            0x1a6e
-#       define EVERGREEN_CURSOR_UPDATE_PENDING          (1 << 0)
-#       define EVERGREEN_CURSOR_UPDATE_TAKEN            (1 << 1)
-#       define EVERGREEN_CURSOR_UPDATE_LOCK             (1 << 16)
-#       define EVERGREEN_CURSOR_DISABLE_MULTIPLE_UPDATE (1 << 24)
-
-
-#define NI_INPUT_CSC_CONTROL                           0x1a35
-#       define NI_INPUT_CSC_GRPH_MODE(x)               (((x) & 0x3) << 0)
-#       define NI_INPUT_CSC_BYPASS                     0
-#       define NI_INPUT_CSC_PROG_COEFF                 1
-#       define NI_INPUT_CSC_PROG_SHARED_MATRIXA        2
-#       define NI_INPUT_CSC_OVL_MODE(x)                (((x) & 0x3) << 4)
-
-#define NI_OUTPUT_CSC_CONTROL                          0x1a3c
-#       define NI_OUTPUT_CSC_GRPH_MODE(x)              (((x) & 0x7) << 0)
-#       define NI_OUTPUT_CSC_BYPASS                    0
-#       define NI_OUTPUT_CSC_TV_RGB                    1
-#       define NI_OUTPUT_CSC_YCBCR_601                 2
-#       define NI_OUTPUT_CSC_YCBCR_709                 3
-#       define NI_OUTPUT_CSC_PROG_COEFF                4
-#       define NI_OUTPUT_CSC_PROG_SHARED_MATRIXB       5
-#       define NI_OUTPUT_CSC_OVL_MODE(x)               (((x) & 0x7) << 4)
-
-#define NI_DEGAMMA_CONTROL                             0x1a58
-#       define NI_GRPH_DEGAMMA_MODE(x)                 (((x) & 0x3) << 0)
-#       define NI_DEGAMMA_BYPASS                       0
-#       define NI_DEGAMMA_SRGB_24                      1
-#       define NI_DEGAMMA_XVYCC_222                    2
-#       define NI_OVL_DEGAMMA_MODE(x)                  (((x) & 0x3) << 4)
-#       define NI_ICON_DEGAMMA_MODE(x)                 (((x) & 0x3) << 8)
-#       define NI_CURSOR_DEGAMMA_MODE(x)               (((x) & 0x3) << 12)
-
-#define NI_GAMUT_REMAP_CONTROL                         0x1a59
-#       define NI_GRPH_GAMUT_REMAP_MODE(x)             (((x) & 0x3) << 0)
-#       define NI_GAMUT_REMAP_BYPASS                   0
-#       define NI_GAMUT_REMAP_PROG_COEFF               1
-#       define NI_GAMUT_REMAP_PROG_SHARED_MATRIXA      2
-#       define NI_GAMUT_REMAP_PROG_SHARED_MATRIXB      3
-#       define NI_OVL_GAMUT_REMAP_MODE(x)              (((x) & 0x3) << 4)
-
-#define NI_REGAMMA_CONTROL                             0x1aa0
-#       define NI_GRPH_REGAMMA_MODE(x)                 (((x) & 0x7) << 0)
-#       define NI_REGAMMA_BYPASS                       0
-#       define NI_REGAMMA_SRGB_24                      1
-#       define NI_REGAMMA_XVYCC_222                    2
-#       define NI_REGAMMA_PROG_A                       3
-#       define NI_REGAMMA_PROG_B                       4
-#       define NI_OVL_REGAMMA_MODE(x)                  (((x) & 0x7) << 4)
-
-
-#define NI_PRESCALE_GRPH_CONTROL                       0x1a2d
-#       define NI_GRPH_PRESCALE_BYPASS                 (1 << 4)
-
-#define NI_PRESCALE_OVL_CONTROL                        0x1a31
-#       define NI_OVL_PRESCALE_BYPASS                  (1 << 4)
-
-#define NI_INPUT_GAMMA_CONTROL                         0x1a10
-#       define NI_GRPH_INPUT_GAMMA_MODE(x)             (((x) & 0x3) << 0)
-#       define NI_INPUT_GAMMA_USE_LUT                  0
-#       define NI_INPUT_GAMMA_BYPASS                   1
-#       define NI_INPUT_GAMMA_SRGB_24                  2
-#       define NI_INPUT_GAMMA_XVYCC_222                3
-#       define NI_OVL_INPUT_GAMMA_MODE(x)              (((x) & 0x3) << 4)
-
-#define	BLACKOUT_MODE_MASK			0x00000007
-#define	VGA_RENDER_CONTROL			0xC0
-#define R_000300_VGA_RENDER_CONTROL             0xC0
-#define C_000300_VGA_VSTATUS_CNTL               0xFFFCFFFF
-#define EVERGREEN_CRTC_STATUS                   0x1BA3
-#define EVERGREEN_CRTC_V_BLANK                  (1 << 0)
-#define EVERGREEN_CRTC_STATUS_POSITION          0x1BA4
-/* CRTC blocks at 0x6df0, 0x79f0, 0x105f0, 0x111f0, 0x11df0, 0x129f0 */
-#define EVERGREEN_CRTC_V_BLANK_START_END                0x1b8d
-#define EVERGREEN_CRTC_CONTROL                          0x1b9c
-#       define EVERGREEN_CRTC_MASTER_EN                 (1 << 0)
-#       define EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE (1 << 24)
-#define EVERGREEN_CRTC_BLANK_CONTROL                    0x1b9d
-#       define EVERGREEN_CRTC_BLANK_DATA_EN             (1 << 8)
-#       define EVERGREEN_CRTC_V_BLANK                   (1 << 0)
-#define EVERGREEN_CRTC_STATUS_HV_COUNT                  0x1ba8
-#define EVERGREEN_CRTC_UPDATE_LOCK                      0x1bb5
-#define EVERGREEN_MASTER_UPDATE_LOCK                    0x1bbd
-#define EVERGREEN_MASTER_UPDATE_MODE                    0x1bbe
-#define EVERGREEN_GRPH_UPDATE_LOCK               (1 << 16)
-#define EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH     0x1a07
-#define EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH   0x1a08
-#define EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS          0x1a04
-#define EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS        0x1a05
-#define EVERGREEN_GRPH_UPDATE                           0x1a11
-#define EVERGREEN_VGA_MEMORY_BASE_ADDRESS               0xc4
-#define EVERGREEN_VGA_MEMORY_BASE_ADDRESS_HIGH          0xc9
-#define EVERGREEN_GRPH_SURFACE_UPDATE_PENDING    (1 << 2)
-
-#define mmVM_CONTEXT1_CNTL__xxRANGE_PROTECTION_FAULT_ENABLE_DEFAULT_MASK 0x10
-#define mmVM_CONTEXT1_CNTL__xxRANGE_PROTECTION_FAULT_ENABLE_DEFAULT__SHIFT 0x4
-#define mmVM_CONTEXT1_CNTL__xxDUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT_MASK 0x80
-#define mmVM_CONTEXT1_CNTL__xxDUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT__SHIFT 0x7
-#define mmVM_CONTEXT1_CNTL__xxPDE0_PROTECTION_FAULT_ENABLE_DEFAULT_MASK 0x400
-#define mmVM_CONTEXT1_CNTL__xxPDE0_PROTECTION_FAULT_ENABLE_DEFAULT__SHIFT 0xa
-#define mmVM_CONTEXT1_CNTL__xxVALID_PROTECTION_FAULT_ENABLE_DEFAULT_MASK 0x2000
-#define mmVM_CONTEXT1_CNTL__xxVALID_PROTECTION_FAULT_ENABLE_DEFAULT__SHIFT 0xd
-#define mmVM_CONTEXT1_CNTL__xxREAD_PROTECTION_FAULT_ENABLE_DEFAULT_MASK 0x10000
-#define mmVM_CONTEXT1_CNTL__xxREAD_PROTECTION_FAULT_ENABLE_DEFAULT__SHIFT 0x10
-#define mmVM_CONTEXT1_CNTL__xxWRITE_PROTECTION_FAULT_ENABLE_DEFAULT_MASK 0x80000
-#define mmVM_CONTEXT1_CNTL__xxWRITE_PROTECTION_FAULT_ENABLE_DEFAULT__SHIFT 0x13
-
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxVMID_MASK 0x1e000000
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxVMID__SHIFT 0x19
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxPROTECTIONS_MASK 0xff
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxPROTECTIONS__SHIFT 0x0
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxMEMORY_CLIENT_ID_MASK 0xff000
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxMEMORY_CLIENT_ID__SHIFT 0xc
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxMEMORY_CLIENT_RW_MASK 0x1000000
-#define mmVM_CONTEXT1_PROTECTION_FAULT_STATUS__xxMEMORY_CLIENT_RW__SHIFT 0x18
-
-#define mmMC_SHARED_BLACKOUT_CNTL__xxBLACKOUT_MODE_MASK 0x7
-#define mmMC_SHARED_BLACKOUT_CNTL__xxBLACKOUT_MODE__SHIFT 0x0
-
-#define mmBIF_FB_EN__xxFB_READ_EN_MASK 0x1
-#define mmBIF_FB_EN__xxFB_READ_EN__SHIFT 0x0
-#define mmBIF_FB_EN__xxFB_WRITE_EN_MASK 0x2
-#define mmBIF_FB_EN__xxFB_WRITE_EN__SHIFT 0x1
-
-#define mmSRBM_SOFT_RESET__xxSOFT_RESET_VMC_MASK 0x20000
-#define mmSRBM_SOFT_RESET__xxSOFT_RESET_VMC__SHIFT 0x11
-#define mmSRBM_SOFT_RESET__xxSOFT_RESET_MC_MASK 0x800
-#define mmSRBM_SOFT_RESET__xxSOFT_RESET_MC__SHIFT 0xb
+#define GRPH_FORMAT_ARGB8888           0
+#define GRPH_FORMAT_ARGB2101010        1
+#define GRPH_FORMAT_32BPP_DIG          2
+#define GRPH_FORMAT_8B_ARGB2101010     3
+#define GRPH_FORMAT_BGRA1010102        4
+#define GRPH_FORMAT_8B_BGRA1010102     5
+#define GRPH_FORMAT_RGB111110          6
+#define GRPH_FORMAT_BGR101111          7
+
+#define GRPH_ENDIAN_NONE               0
+#define GRPH_ENDIAN_8IN16              1
+#define GRPH_ENDIAN_8IN32              2
+#define GRPH_ENDIAN_8IN64              3
+#define GRPH_RED_SEL_R                 0
+#define GRPH_RED_SEL_G                 1
+#define GRPH_RED_SEL_B                 2
+#define GRPH_RED_SEL_A                 3
+
+#define GRPH_GREEN_SEL_G               0
+#define GRPH_GREEN_SEL_B               1
+#define GRPH_GREEN_SEL_A               2
+#define GRPH_GREEN_SEL_R               3
+
+#define GRPH_BLUE_SEL_B                0
+#define GRPH_BLUE_SEL_A                1
+#define GRPH_BLUE_SEL_R                2
+#define GRPH_BLUE_SEL_G                3
+
+#define GRPH_ALPHA_SEL_A               0
+#define GRPH_ALPHA_SEL_R               1
+#define GRPH_ALPHA_SEL_G               2
+#define GRPH_ALPHA_SEL_B               3
+
+/* CUR_CONTROL */
+	#define CURSOR_MONO                    0
+	#define CURSOR_24_1                    1
+	#define CURSOR_24_8_PRE_MULT           2
+	#define CURSOR_24_8_UNPRE_MULT         3
+	#define CURSOR_URGENT_ALWAYS           0
+	#define CURSOR_URGENT_1_8              1
+	#define CURSOR_URGENT_1_4              2
+	#define CURSOR_URGENT_3_8              3
+	#define CURSOR_URGENT_1_2              4
+
+/* INPUT_CSC_CONTROL */
+#       define INPUT_CSC_BYPASS                     0
+#       define INPUT_CSC_PROG_COEFF                 1
+#       define INPUT_CSC_PROG_SHARED_MATRIXA        2
+
+/* OUTPUT_CSC_CONTROL */
+#       define OUTPUT_CSC_BYPASS                    0
+#       define OUTPUT_CSC_TV_RGB                    1
+#       define OUTPUT_CSC_YCBCR_601                 2
+#       define OUTPUT_CSC_YCBCR_709                 3
+#       define OUTPUT_CSC_PROG_COEFF                4
+#       define OUTPUT_CSC_PROG_SHARED_MATRIXB       5
+
+/* DEGAMMA_CONTROL */
+#       define DEGAMMA_BYPASS                       0
+#       define DEGAMMA_SRGB_24                      1
+#       define DEGAMMA_XVYCC_222                    2
+
+/* GAMUT_REMAP_CONTROL */
+#       define GAMUT_REMAP_BYPASS                   0
+#       define GAMUT_REMAP_PROG_COEFF               1
+#       define GAMUT_REMAP_PROG_SHARED_MATRIXA      2
+#       define GAMUT_REMAP_PROG_SHARED_MATRIXB      3
+
+/* REGAMMA_CONTROL */
+#       define REGAMMA_BYPASS                       0
+#       define REGAMMA_SRGB_24                      1
+#       define REGAMMA_XVYCC_222                    2
+#       define REGAMMA_PROG_A                       3
+#       define REGAMMA_PROG_B                       4
+
+
+/* INPUT_GAMMA_CONTROL */
+#       define INPUT_GAMMA_USE_LUT                  0
+#       define INPUT_GAMMA_BYPASS                   1
+#       define INPUT_GAMMA_SRGB_24                  2
+#       define INPUT_GAMMA_XVYCC_222                3
 
 #define MC_SEQ_MISC0__MT__MASK	0xf0000000
 #define MC_SEQ_MISC0__MT__GDDR1  0x10000000
@@ -2113,20 +785,9 @@
 #define PACKET3_SEM_SEL_SIGNAL	    (0x6 << 29)
 #define PACKET3_SEM_SEL_WAIT	    (0x7 << 29)
 
-#define CONFIG_CNTL	0x1509
-#define CC_DRM_ID_STRAPS	0X1559
 #define AMDGPU_PCIE_INDEX	0xc
 #define AMDGPU_PCIE_DATA	0xd
 
-#define DMA_SEM_INCOMPLETE_TIMER_CNTL                     0x3411
-#define DMA_SEM_WAIT_FAIL_TIMER_CNTL                      0x3412
-#define DMA_MODE                                          0x342f
-#define DMA_RB_RPTR_ADDR_HI                               0x3407
-#define DMA_RB_RPTR_ADDR_LO                               0x3408
-#define DMA_BUSY_MASK 0x20
-#define DMA1_BUSY_MASK 0X40
-#define SDMA_MAX_INSTANCE 2
-
 #define PCIE_BUS_CLK    10000
 #define TCLK            (PCIE_BUS_CLK / 10)
 #define	PCIE_PORT_INDEX					0xe
@@ -2136,8 +797,6 @@
 #define EVERGREEN_PIF_PHY1_INDEX                        0x10
 #define EVERGREEN_PIF_PHY1_DATA				0x14
 
-#define	MC_VM_FB_OFFSET					0x81a
-
 /* Discrete VCE clocks */
 #define CG_VCEPLL_FUNC_CNTL                             0xc0030600
 #define    VCEPLL_RESET_MASK                            0x00000001
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 659eab9b90be..c457be3a3c56 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -584,6 +584,8 @@ soc15_asic_reset_method(struct amdgpu_device *adev)
 		 * Enable triggering of GPU reset only if specified
 		 * by module parameter.
 		 */
+		if (adev->pcie_reset_ctx.in_link_reset)
+			return AMD_RESET_METHOD_LINK;
 		if (amdgpu_gpu_recovery == 4 || amdgpu_gpu_recovery == 5)
 			return AMD_RESET_METHOD_MODE2;
 		else if (!(adev->flags & AMD_IS_APU))
@@ -640,6 +642,9 @@ asic_reset:
 	case AMD_RESET_METHOD_MODE2:
 		dev_info(adev->dev, "MODE2 reset\n");
 		return amdgpu_dpm_mode2_reset(adev);
+	case AMD_RESET_METHOD_LINK:
+		dev_info(adev->dev, "Link reset\n");
+		return amdgpu_device_link_reset(adev);
 	default:
 		dev_info(adev->dev, "MODE1 reset\n");
 		return amdgpu_device_mode1_reset(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15d.h b/drivers/gpu/drm/amd/amdgpu/soc15d.h
index a5000c171c02..cf93fa477674 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15d.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15d.h
@@ -552,6 +552,11 @@
 #              define PACKET3_QUERY_STATUS_DOORBELL_OFFSET(x)  ((x) << 2)
 #              define PACKET3_QUERY_STATUS_ENG_SEL(x)          ((x) << 25)
 
+#define PACKET3_RUN_CLEANER_SHADER_9_0                      0xD7
+/* 1. header
+ * 2. RESERVED [31:0]
+ */
+
 #define PACKET3_RUN_CLEANER_SHADER                      0xD2
 /* 1. header
  * 2. RESERVED [31:0]
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
index 0e404c074975..e590cbdd8de9 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
@@ -174,19 +174,76 @@ static void umc_v12_0_query_ras_error_count(struct amdgpu_device *adev,
 	umc_v12_0_reset_error_count(adev);
 }
 
+static void umc_v12_0_get_retire_flip_bits(struct amdgpu_device *adev)
+{
+	enum amdgpu_memory_partition nps = AMDGPU_NPS1_PARTITION_MODE;
+	uint32_t vram_type = adev->gmc.vram_type;
+	struct amdgpu_umc_flip_bits *flip_bits = &(adev->umc.flip_bits);
+
+	if (adev->gmc.gmc_funcs->query_mem_partition_mode)
+		nps = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
+
+	/* default setting */
+	flip_bits->flip_bits_in_pa[0] = UMC_V12_0_PA_C2_BIT;
+	flip_bits->flip_bits_in_pa[1] = UMC_V12_0_PA_C3_BIT;
+	flip_bits->flip_bits_in_pa[2] = UMC_V12_0_PA_C4_BIT;
+	flip_bits->flip_bits_in_pa[3] = UMC_V12_0_PA_R13_BIT;
+	flip_bits->flip_row_bit = 13;
+	flip_bits->bit_num = 4;
+	flip_bits->r13_in_pa = UMC_V12_0_PA_R13_BIT;
+
+	if (nps == AMDGPU_NPS2_PARTITION_MODE) {
+		flip_bits->flip_bits_in_pa[0] = UMC_V12_0_PA_CH5_BIT;
+		flip_bits->flip_bits_in_pa[1] = UMC_V12_0_PA_C2_BIT;
+		flip_bits->flip_bits_in_pa[2] = UMC_V12_0_PA_B1_BIT;
+		flip_bits->r13_in_pa = UMC_V12_0_PA_R12_BIT;
+	} else if (nps == AMDGPU_NPS4_PARTITION_MODE) {
+		flip_bits->flip_bits_in_pa[0] = UMC_V12_0_PA_CH4_BIT;
+		flip_bits->flip_bits_in_pa[1] = UMC_V12_0_PA_CH5_BIT;
+		flip_bits->flip_bits_in_pa[2] = UMC_V12_0_PA_B0_BIT;
+		flip_bits->r13_in_pa = UMC_V12_0_PA_R11_BIT;
+	}
+
+	switch (vram_type) {
+	case AMDGPU_VRAM_TYPE_HBM:
+		/* other nps modes are taken as nps1 */
+		if (nps == AMDGPU_NPS2_PARTITION_MODE)
+			flip_bits->flip_bits_in_pa[3] = UMC_V12_0_PA_R12_BIT;
+		else if (nps == AMDGPU_NPS4_PARTITION_MODE)
+			flip_bits->flip_bits_in_pa[3] = UMC_V12_0_PA_R11_BIT;
+
+		break;
+	case AMDGPU_VRAM_TYPE_HBM3E:
+		flip_bits->flip_bits_in_pa[3] = UMC_V12_0_PA_R12_BIT;
+		flip_bits->flip_row_bit = 12;
+
+		if (nps == AMDGPU_NPS2_PARTITION_MODE)
+			flip_bits->flip_bits_in_pa[3] = UMC_V12_0_PA_R11_BIT;
+		else if (nps == AMDGPU_NPS4_PARTITION_MODE)
+			flip_bits->flip_bits_in_pa[3] = UMC_V12_0_PA_R10_BIT;
+
+		break;
+	default:
+		dev_warn(adev->dev,
+			"Unknown HBM type, set RAS retire flip bits to the value in NPS1 mode.\n");
+		break;
+	}
+
+	adev->umc.retire_unit = 0x1 << flip_bits->bit_num;
+}
+
 static int umc_v12_0_convert_error_address(struct amdgpu_device *adev,
 					struct ras_err_data *err_data,
 					struct ta_ras_query_address_input *addr_in,
 					struct ta_ras_query_address_output *addr_out,
 					bool dump_addr)
 {
-	uint32_t col, col_lower, row, row_lower, bank;
+	uint32_t col, col_lower, row, row_lower, row_high, bank;
 	uint32_t channel_index = 0, umc_inst = 0;
-	uint32_t i, loop_bits[UMC_V12_0_RETIRE_LOOP_BITS];
+	uint32_t i, bit_num, retire_unit, *flip_bits;
 	uint64_t soc_pa, column, err_addr;
 	struct ta_ras_query_address_output addr_out_tmp;
 	struct ta_ras_query_address_output *paddr_out;
-	enum amdgpu_memory_partition nps = AMDGPU_NPS1_PARTITION_MODE;
 	int ret = 0;
 
 	if (!addr_out)
@@ -211,46 +268,46 @@ static int umc_v12_0_convert_error_address(struct amdgpu_device *adev,
 		umc_inst = addr_in->ma.umc_inst;
 	}
 
-	loop_bits[0] = UMC_V12_0_PA_C2_BIT;
-	loop_bits[1] = UMC_V12_0_PA_C3_BIT;
-	loop_bits[2] = UMC_V12_0_PA_C4_BIT;
-	loop_bits[3] = UMC_V12_0_PA_R13_BIT;
-
-	if (adev->gmc.gmc_funcs->query_mem_partition_mode)
-		nps = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
-
-	/* other nps modes are taken as nps1 */
-	if (nps == AMDGPU_NPS4_PARTITION_MODE) {
-		loop_bits[0] = UMC_V12_0_PA_CH4_BIT;
-		loop_bits[1] = UMC_V12_0_PA_CH5_BIT;
-		loop_bits[2] = UMC_V12_0_PA_B0_BIT;
-		loop_bits[3] = UMC_V12_0_PA_R11_BIT;
-	}
+	flip_bits = adev->umc.flip_bits.flip_bits_in_pa;
+	bit_num = adev->umc.flip_bits.bit_num;
+	retire_unit = adev->umc.retire_unit;
 
 	soc_pa = paddr_out->pa.pa;
 	channel_index = paddr_out->pa.channel_idx;
 	/* clear loop bits in soc physical address */
-	for (i = 0; i < UMC_V12_0_RETIRE_LOOP_BITS; i++)
-		soc_pa &= ~BIT_ULL(loop_bits[i]);
+	for (i = 0; i < bit_num; i++)
+		soc_pa &= ~BIT_ULL(flip_bits[i]);
 
 	paddr_out->pa.pa = soc_pa;
 	/* get column bit 0 and 1 in mca address */
 	col_lower = (err_addr >> 1) & 0x3ULL;
-	/* MA_R13_BIT will be handled later */
+	/* extra row bit will be handled later */
 	row_lower = (err_addr >> UMC_V12_0_MA_R0_BIT) & 0x1fffULL;
+	row_lower &= ~BIT_ULL(adev->umc.flip_bits.flip_row_bit);
+
+	if (amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(9, 5, 0)) {
+		row_high = (soc_pa >> adev->umc.flip_bits.r13_in_pa) & 0x3ULL;
+		/* it's 2.25GB in each channel, from MCA address to PA
+		 * [R14 R13] is converted if the two bits value are 0x3,
+		 * get them from PA instead of MCA address.
+		 */
+		row_lower |= (row_high << 13);
+	}
 
 	if (!err_data && !dump_addr)
 		goto out;
 
 	/* loop for all possibilities of retired bits */
-	for (column = 0; column < UMC_V12_0_BAD_PAGE_NUM_PER_CHANNEL; column++) {
+	for (column = 0; column < retire_unit; column++) {
 		soc_pa = paddr_out->pa.pa;
-		for (i = 0; i < UMC_V12_0_RETIRE_LOOP_BITS; i++)
-			soc_pa |= (((column >> i) & 0x1ULL) << loop_bits[i]);
+		for (i = 0; i < bit_num; i++)
+			soc_pa |= (((column >> i) & 0x1ULL) << flip_bits[i]);
 
 		col = ((column & 0x7) << 2) | col_lower;
-		/* add row bit 13 */
-		row = ((column >> 3) << 13) | row_lower;
+		/* handle extra row bit */
+		if (bit_num == RETIRE_FLIP_BITS_NUM)
+			row = ((column >> 3) << adev->umc.flip_bits.flip_row_bit) |
+					row_lower;
 
 		if (dump_addr)
 			dev_info(adev->dev,
@@ -428,8 +485,12 @@ static int umc_v12_0_aca_bank_parser(struct aca_handle *handle, struct aca_bank
 		bank->regs[ACA_REG_IDX_ADDR]);
 
 	ext_error_code = ACA_REG__STATUS__ERRORCODEEXT(status);
-	count = ext_error_code == 0 ?
-		ACA_REG__MISC0__ERRCNT(bank->regs[ACA_REG_IDX_MISC0]) : 1ULL;
+	if (umc_v12_0_is_deferred_error(adev, status))
+		count = ext_error_code == 0 ?
+			adev->umc.err_addr_cnt / adev->umc.retire_unit : 1ULL;
+	else
+		count = ext_error_code == 0 ?
+			ACA_REG__MISC0__ERRCNT(bank->regs[ACA_REG_IDX_MISC0]) : 1ULL;
 
 	return aca_error_cache_log_bank_error(handle, &info, err_type, count);
 }
@@ -469,8 +530,7 @@ static int umc_v12_0_update_ecc_status(struct amdgpu_device *adev,
 	uint64_t err_addr, pa_addr = 0;
 	struct ras_ecc_err *ecc_err;
 	struct ta_ras_query_address_output addr_out;
-	enum amdgpu_memory_partition nps = AMDGPU_NPS1_PARTITION_MODE;
-	uint32_t shift_bit = UMC_V12_0_PA_C4_BIT;
+	uint32_t shift_bit = adev->umc.flip_bits.flip_bits_in_pa[2];
 	int count, ret, i;
 
 	hwid = REG_GET_FIELD(ipid, MCMP1_IPIDT0, HardwareID);
@@ -515,11 +575,6 @@ static int umc_v12_0_update_ecc_status(struct amdgpu_device *adev,
 	ecc_err->pa_pfn = pa_addr >> AMDGPU_GPU_PAGE_SHIFT;
 	ecc_err->channel_idx = addr_out.pa.channel_idx;
 
-	if (adev->gmc.gmc_funcs->query_mem_partition_mode)
-		nps = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
-	if (nps == AMDGPU_NPS4_PARTITION_MODE)
-		shift_bit = UMC_V12_0_PA_B0_BIT;
-
 	/* If converted pa_pfn is 0, use pa C4 pfn. */
 	if (!ecc_err->pa_pfn)
 		ecc_err->pa_pfn = BIT_ULL(shift_bit) >> AMDGPU_GPU_PAGE_SHIFT;
@@ -665,5 +720,6 @@ struct amdgpu_umc_ras umc_v12_0_ras = {
 	.update_ecc_status = umc_v12_0_update_ecc_status,
 	.convert_ras_err_addr = umc_v12_0_convert_error_address,
 	.get_die_id_from_pa = umc_v12_0_get_die_id,
+	.get_retire_flip_bits = umc_v12_0_get_retire_flip_bits,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h
index 9298018d938f..63b7e7254526 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h
@@ -55,8 +55,6 @@
 #define UMC_V12_0_NA_MAP_PA_NUM        8
 /* R13 bit shift should be considered, double the number */
 #define UMC_V12_0_BAD_PAGE_NUM_PER_CHANNEL (UMC_V12_0_NA_MAP_PA_NUM * 2)
-/* C2, C3, C4, R13, four bits in MCA address are looped in retirement */
-#define UMC_V12_0_RETIRE_LOOP_BITS 4
 
 /* column bits in SOC physical address */
 #define UMC_V12_0_PA_C2_BIT 15
@@ -64,13 +62,16 @@
 #define UMC_V12_0_PA_C4_BIT 21
 /* row bits in SOC physical address */
 #define UMC_V12_0_PA_R0_BIT 22
+#define UMC_V12_0_PA_R10_BIT 32
 #define UMC_V12_0_PA_R11_BIT 33
+#define UMC_V12_0_PA_R12_BIT 34
 #define UMC_V12_0_PA_R13_BIT 35
 /* channel bit in SOC physical address */
 #define UMC_V12_0_PA_CH4_BIT 12
 #define UMC_V12_0_PA_CH5_BIT 13
 /* bank bit in SOC physical address */
 #define UMC_V12_0_PA_B0_BIT 19
+#define UMC_V12_0_PA_B1_BIT 20
 /* row bits in MCA address */
 #define UMC_V12_0_MA_R0_BIT 10
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 8e7a36f26e9c..b8d835c9e17e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -39,6 +39,7 @@
 
 #define VCN_VID_SOC_ADDRESS_2_0					0x1fa00
 #define VCN1_VID_SOC_ADDRESS_3_0				0x48200
+#define VCN1_AON_SOC_ADDRESS_3_0				0x48000
 
 #define mmUVD_CONTEXT_ID_INTERNAL_OFFSET			0x1fd
 #define mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET			0x503
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index d716510b8dd6..3eec1b8feaee 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -39,6 +39,7 @@
 
 #define VCN_VID_SOC_ADDRESS_2_0					0x1fa00
 #define VCN1_VID_SOC_ADDRESS_3_0				0x48200
+#define VCN1_AON_SOC_ADDRESS_3_0				0x48000
 
 #define mmUVD_CONTEXT_ID_INTERNAL_OFFSET			0x27
 #define mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET			0x0f
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 22ae1939476f..0b19f0ab4480 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -40,6 +40,7 @@
 
 #define VCN_VID_SOC_ADDRESS_2_0					0x1fa00
 #define VCN1_VID_SOC_ADDRESS_3_0				0x48200
+#define VCN1_AON_SOC_ADDRESS_3_0				0x48000
 
 #define mmUVD_CONTEXT_ID_INTERNAL_OFFSET			0x27
 #define mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET			0x0f
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index c6f6392c1c20..8fff470bce87 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -46,6 +46,7 @@
 
 #define VCN_VID_SOC_ADDRESS_2_0							0x1fb00
 #define VCN1_VID_SOC_ADDRESS_3_0						0x48300
+#define VCN1_AON_SOC_ADDRESS_3_0						0x48000
 
 #define VCN_HARVEST_MMSCH								0
 
@@ -238,9 +239,9 @@ static int vcn_v4_0_sw_init(struct amdgpu_ip_block *ip_block)
 			adev->vcn.inst[i].pause_dpg_mode = vcn_v4_0_pause_dpg_mode;
 	}
 
-	/* TODO: Add queue reset mask when FW fully supports it */
 	adev->vcn.supported_reset =
 		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
+	adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
 
 	if (amdgpu_sriov_vf(adev)) {
 		r = amdgpu_virt_alloc_mm_table(adev);
@@ -614,7 +615,8 @@ static void vcn_v4_0_mc_resume_dpg_mode(struct amdgpu_vcn_inst *vinst,
 
 	/* VCN global tiling registers */
 	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
-		VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
+			VCN, inst_idx, regUVD_GFX10_ADDR_CONFIG),
+			adev->gfx.config.gb_addr_config, 0, indirect);
 }
 
 /**
@@ -1945,6 +1947,20 @@ static int vcn_v4_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p,
 	return 0;
 }
 
+static int vcn_v4_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me];
+
+	if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE))
+		return -EOPNOTSUPP;
+
+	vcn_v4_0_stop(vinst);
+	vcn_v4_0_start(vinst);
+
+	return amdgpu_ring_test_helper(ring);
+}
+
 static struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
@@ -1974,6 +1990,7 @@ static struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = {
 	.emit_wreg = vcn_v2_0_enc_ring_emit_wreg,
 	.emit_reg_wait = vcn_v2_0_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+	.reset = vcn_v4_0_ring_reset,
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 3e176b4b7c69..712e1fba33ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -45,6 +45,7 @@
 
 #define VCN_VID_SOC_ADDRESS_2_0		0x1fb00
 #define VCN1_VID_SOC_ADDRESS_3_0	0x48300
+#define VCN1_AON_SOC_ADDRESS_3_0	0x48000
 
 static const struct amdgpu_hwip_reg_entry vcn_reg_list_4_0_3[] = {
 	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_POWER_STATUS),
@@ -287,6 +288,31 @@ static int vcn_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
+static int vcn_v4_0_3_hw_init_inst(struct amdgpu_vcn_inst *vinst)
+{
+	int vcn_inst;
+	struct amdgpu_device *adev = vinst->adev;
+	struct amdgpu_ring *ring;
+	int inst_idx = vinst->inst;
+
+	vcn_inst = GET_INST(VCN, inst_idx);
+	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
+	if (ring->use_doorbell) {
+		adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+			(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * vcn_inst,
+			adev->vcn.inst[inst_idx].aid_id);
+
+		WREG32_SOC15(VCN, vcn_inst, regVCN_RB1_DB_CTRL,
+			ring->doorbell_index << VCN_RB1_DB_CTRL__OFFSET__SHIFT |
+			VCN_RB1_DB_CTRL__EN_MASK);
+
+		/* Read DB_CTRL to flush the write DB_CTRL command. */
+		RREG32_SOC15(VCN, vcn_inst, regVCN_RB1_DB_CTRL);
+	}
+
+	return 0;
+}
+
 /**
  * vcn_v4_0_3_hw_init - start and test VCN block
  *
@@ -298,7 +324,8 @@ static int vcn_v4_0_3_hw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 	struct amdgpu_ring *ring;
-	int i, r, vcn_inst;
+	struct amdgpu_vcn_inst *vinst;
+	int i, r;
 
 	if (amdgpu_sriov_vf(adev)) {
 		r = vcn_v4_0_3_start_sriov(adev);
@@ -321,28 +348,9 @@ static int vcn_v4_0_3_hw_init(struct amdgpu_ip_block *ip_block)
 		for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
 			struct amdgpu_vcn4_fw_shared *fw_shared;
 
-			vcn_inst = GET_INST(VCN, i);
 			ring = &adev->vcn.inst[i].ring_enc[0];
-
-			if (ring->use_doorbell) {
-				adev->nbio.funcs->vcn_doorbell_range(
-					adev, ring->use_doorbell,
-					(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
-						9 * vcn_inst,
-					adev->vcn.inst[i].aid_id);
-
-				WREG32_SOC15(
-					VCN, GET_INST(VCN, ring->me),
-					regVCN_RB1_DB_CTRL,
-					ring->doorbell_index
-							<< VCN_RB1_DB_CTRL__OFFSET__SHIFT |
-						VCN_RB1_DB_CTRL__EN_MASK);
-
-				/* Read DB_CTRL to flush the write DB_CTRL command. */
-				RREG32_SOC15(
-					VCN, GET_INST(VCN, ring->me),
-					regVCN_RB1_DB_CTRL);
-			}
+			vinst = &adev->vcn.inst[i];
+			vcn_v4_0_3_hw_init_inst(vinst);
 
 			/* Re-init fw_shared when RAS fatal error occurred */
 			fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
@@ -1563,6 +1571,37 @@ static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring)
 	}
 }
 
+static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid)
+{
+	int r = 0;
+	int vcn_inst;
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me];
+
+	if (amdgpu_sriov_vf(ring->adev))
+		return -EOPNOTSUPP;
+
+	if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE))
+		return -EOPNOTSUPP;
+
+	vcn_inst = GET_INST(VCN, ring->me);
+	r = amdgpu_dpm_reset_vcn(adev, 1 << vcn_inst);
+
+	if (r) {
+		DRM_DEV_ERROR(adev->dev, "VCN reset fail : %d\n", r);
+		return r;
+	}
+
+	/* This flag is not set for VF, assumed to be disabled always */
+	if (RREG32_SOC15(VCN, GET_INST(VCN, 0), regVCN_RRMT_CNTL) & 0x100)
+		adev->vcn.caps |= AMDGPU_VCN_CAPS(RRMT_ENABLED);
+	vcn_v4_0_3_hw_init_inst(vinst);
+	vcn_v4_0_3_start_dpg_mode(vinst, adev->vcn.inst[ring->me].indirect_sram);
+	r = amdgpu_ring_test_helper(ring);
+
+	return r;
+}
+
 static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
@@ -1591,6 +1630,7 @@ static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = {
 	.emit_wreg = vcn_v4_0_3_enc_ring_emit_wreg,
 	.emit_reg_wait = vcn_v4_0_3_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+	.reset = vcn_v4_0_3_ring_reset,
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index ba603b2246e2..a09f9a2dd471 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -46,6 +46,7 @@
 
 #define VCN_VID_SOC_ADDRESS_2_0						0x1fb00
 #define VCN1_VID_SOC_ADDRESS_3_0					(0x48300 + 0x38000)
+#define VCN1_AON_SOC_ADDRESS_3_0					(0x48000 + 0x38000)
 
 #define VCN_HARVEST_MMSCH							0
 
@@ -207,6 +208,10 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block)
 		if (amdgpu_sriov_vf(adev))
 			fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_VF_RB_SETUP_FLAG);
 
+		fw_shared->present_flag_0 |= AMDGPU_FW_SHARED_FLAG_0_DRM_KEY_INJECT;
+		fw_shared->drm_key_wa.method =
+			AMDGPU_DRM_KEY_INJECT_WORKAROUND_VCNFW_ASD_HANDSHAKING;
+
 		if (amdgpu_vcnfw_log)
 			amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]);
 
@@ -214,6 +219,13 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block)
 			adev->vcn.inst[i].pause_dpg_mode = vcn_v4_0_5_pause_dpg_mode;
 	}
 
+	adev->vcn.supported_reset = amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
+	adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
+
+	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
+	if (r)
+		return r;
+
 	if (amdgpu_sriov_vf(adev)) {
 		r = amdgpu_virt_alloc_mm_table(adev);
 		if (r)
@@ -1022,6 +1034,10 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 			ring->doorbell_index << VCN_RB1_DB_CTRL__OFFSET__SHIFT |
 			VCN_RB1_DB_CTRL__EN_MASK);
 
+	/* Keeping one read-back to ensure all register writes are done, otherwise
+	 * it may introduce race conditions */
+	RREG32_SOC15(VCN, inst_idx, regVCN_RB1_DB_CTRL);
+
 	return 0;
 }
 
@@ -1204,6 +1220,10 @@ static int vcn_v4_0_5_start(struct amdgpu_vcn_inst *vinst)
 	WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
 	fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF);
 
+	/* Keeping one read-back to ensure all register writes are done, otherwise
+	 * it may introduce race conditions */
+	RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
+
 	return 0;
 }
 
@@ -1435,6 +1455,20 @@ static void vcn_v4_0_5_unified_ring_set_wptr(struct amdgpu_ring *ring)
 	}
 }
 
+static int vcn_v4_0_5_ring_reset(struct amdgpu_ring *ring, unsigned int vmid)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me];
+
+	if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE))
+		return -EOPNOTSUPP;
+
+	vcn_v4_0_5_stop(vinst);
+	vcn_v4_0_5_start(vinst);
+
+	return amdgpu_ring_test_helper(ring);
+}
+
 static struct amdgpu_ring_funcs vcn_v4_0_5_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
@@ -1462,6 +1496,7 @@ static struct amdgpu_ring_funcs vcn_v4_0_5_unified_ring_vm_funcs = {
 	.emit_wreg = vcn_v2_0_enc_ring_emit_wreg,
 	.emit_reg_wait = vcn_v2_0_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+	.reset = vcn_v4_0_5_ring_reset,
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index d99d05f42f1d..27dcc6f37a73 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -196,9 +196,9 @@ static int vcn_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block)
 			adev->vcn.inst[i].pause_dpg_mode = vcn_v5_0_0_pause_dpg_mode;
 	}
 
-	/* TODO: Add queue reset mask when FW fully supports it */
 	adev->vcn.supported_reset =
 		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
+	adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
 
 	vcn_v5_0_0_alloc_ip_dump(adev);
 
@@ -533,7 +533,8 @@ static void vcn_v5_0_0_mc_resume_dpg_mode(struct amdgpu_vcn_inst *vinst,
 
 	/* VCN global tiling registers */
 	WREG32_SOC24_DPG_MODE(inst_idx, SOC24_DPG_MODE_OFFSET(
-		VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
+		VCN, inst_idx, regUVD_GFX10_ADDR_CONFIG),
+		adev->gfx.config.gb_addr_config, 0, indirect);
 
 	return;
 }
@@ -1171,6 +1172,20 @@ static void vcn_v5_0_0_unified_ring_set_wptr(struct amdgpu_ring *ring)
 	}
 }
 
+static int vcn_v5_0_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me];
+
+	if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE))
+		return -EOPNOTSUPP;
+
+	vcn_v5_0_0_stop(vinst);
+	vcn_v5_0_0_start(vinst);
+
+	return amdgpu_ring_test_helper(ring);
+}
+
 static const struct amdgpu_ring_funcs vcn_v5_0_0_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
@@ -1198,6 +1213,7 @@ static const struct amdgpu_ring_funcs vcn_v5_0_0_unified_ring_vm_funcs = {
 	.emit_wreg = vcn_v2_0_enc_ring_emit_wreg,
 	.emit_reg_wait = vcn_v2_0_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+	.reset = vcn_v5_0_0_ring_reset,
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 581d8629b9d9..8e843011703c 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -30,6 +30,7 @@
 #include "soc15_hw_ip.h"
 #include "vcn_v2_0.h"
 #include "vcn_v4_0_3.h"
+#include "mmsch_v5_0.h"
 
 #include "vcn/vcn_5_0_0_offset.h"
 #include "vcn/vcn_5_0_0_sh_mask.h"
@@ -39,6 +40,7 @@
 
 #include <drm/drm_drv.h>
 
+static int vcn_v5_0_1_start_sriov(struct amdgpu_device *adev);
 static void vcn_v5_0_1_set_unified_ring_funcs(struct amdgpu_device *adev);
 static void vcn_v5_0_1_set_irq_funcs(struct amdgpu_device *adev);
 static int vcn_v5_0_1_set_pg_state(struct amdgpu_vcn_inst *vinst,
@@ -126,7 +128,14 @@ static int vcn_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 
 		ring = &adev->vcn.inst[i].ring_enc[0];
 		ring->use_doorbell = true;
-		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * vcn_inst;
+		if (!amdgpu_sriov_vf(adev))
+			ring->doorbell_index =
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+				11 * vcn_inst;
+		else
+			ring->doorbell_index =
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+				32 * vcn_inst;
 
 		ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id);
 		sprintf(ring->name, "vcn_unified_%d", adev->vcn.inst[i].aid_id);
@@ -143,6 +152,12 @@ static int vcn_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 	adev->vcn.supported_reset =
 		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
 
+	if (amdgpu_sriov_vf(adev)) {
+		r = amdgpu_virt_alloc_mm_table(adev);
+		if (r)
+			return r;
+	}
+
 	vcn_v5_0_0_alloc_ip_dump(adev);
 
 	return amdgpu_vcn_sysfs_reset_mask_init(adev);
@@ -172,6 +187,9 @@ static int vcn_v5_0_1_sw_fini(struct amdgpu_ip_block *ip_block)
 		drm_dev_exit(idx);
 	}
 
+	if (amdgpu_sriov_vf(adev))
+		amdgpu_virt_free_mm_table(adev);
+
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 		r = amdgpu_vcn_suspend(adev, i);
 		if (r)
@@ -204,24 +222,38 @@ static int vcn_v5_0_1_hw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_ring *ring;
 	int i, r, vcn_inst;
 
-	if (RREG32_SOC15(VCN, GET_INST(VCN, 0), regVCN_RRMT_CNTL) & 0x100)
-		adev->vcn.caps |= AMDGPU_VCN_CAPS(RRMT_ENABLED);
-	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		vcn_inst = GET_INST(VCN, i);
-		ring = &adev->vcn.inst[i].ring_enc[0];
-
-		if (ring->use_doorbell)
-			adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-				((adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
-				 9 * vcn_inst),
-				adev->vcn.inst[i].aid_id);
-
-		/* Re-init fw_shared, if required */
-		vcn_v5_0_1_fw_shared_init(adev, i);
-
-		r = amdgpu_ring_test_helper(ring);
+	if (amdgpu_sriov_vf(adev)) {
+		r = vcn_v5_0_1_start_sriov(adev);
 		if (r)
 			return r;
+
+		for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+			ring = &adev->vcn.inst[i].ring_enc[0];
+			ring->wptr = 0;
+			ring->wptr_old = 0;
+			vcn_v5_0_1_unified_ring_set_wptr(ring);
+			ring->sched.ready = true;
+		}
+	} else {
+		if (RREG32_SOC15(VCN, GET_INST(VCN, 0), regVCN_RRMT_CNTL) & 0x100)
+			adev->vcn.caps |= AMDGPU_VCN_CAPS(RRMT_ENABLED);
+		for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+			vcn_inst = GET_INST(VCN, i);
+			ring = &adev->vcn.inst[i].ring_enc[0];
+
+			if (ring->use_doorbell)
+				adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+					((adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+					11 * vcn_inst),
+					adev->vcn.inst[i].aid_id);
+
+			/* Re-init fw_shared, if required */
+			vcn_v5_0_1_fw_shared_init(adev, i);
+
+			r = amdgpu_ring_test_helper(ring);
+			if (r)
+				return r;
+		}
 	}
 
 	return 0;
@@ -503,6 +535,52 @@ static void vcn_v5_0_1_enable_clock_gating(struct amdgpu_vcn_inst *vinst)
 }
 
 /**
+ * vcn_v5_0_1_pause_dpg_mode - VCN pause with dpg mode
+ *
+ * @vinst: VCN instance
+ * @new_state: pause state
+ *
+ * Pause dpg mode for VCN block
+ */
+static int vcn_v5_0_1_pause_dpg_mode(struct amdgpu_vcn_inst *vinst,
+				     struct dpg_pause_state *new_state)
+{
+	struct amdgpu_device *adev = vinst->adev;
+	uint32_t reg_data = 0;
+	int vcn_inst;
+
+	vcn_inst = GET_INST(VCN, vinst->inst);
+
+	/* pause/unpause if state is changed */
+	if (vinst->pause_state.fw_based != new_state->fw_based) {
+		DRM_DEV_DEBUG(adev->dev, "dpg pause state changed %d -> %d %s\n",
+			vinst->pause_state.fw_based, new_state->fw_based,
+			new_state->fw_based ? "VCN_DPG_STATE__PAUSE" : "VCN_DPG_STATE__UNPAUSE");
+		reg_data = RREG32_SOC15(VCN, vcn_inst, regUVD_DPG_PAUSE) &
+			(~UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK);
+
+		if (new_state->fw_based == VCN_DPG_STATE__PAUSE) {
+			/* pause DPG */
+			reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK;
+			WREG32_SOC15(VCN, vcn_inst, regUVD_DPG_PAUSE, reg_data);
+
+			/* wait for ACK */
+			SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_DPG_PAUSE,
+					UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK,
+					UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK);
+		} else {
+			/* unpause DPG, no need to wait */
+			reg_data &= ~UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK;
+			WREG32_SOC15(VCN, vcn_inst, regUVD_DPG_PAUSE, reg_data);
+		}
+		vinst->pause_state.fw_based = new_state->fw_based;
+	}
+
+	return 0;
+}
+
+
+/**
  * vcn_v5_0_1_start_dpg_mode - VCN start with dpg mode
  *
  * @vinst: VCN instance
@@ -518,6 +596,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 	volatile struct amdgpu_vcn5_fw_shared *fw_shared =
 		adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
+	struct dpg_pause_state state = {.fw_based = VCN_DPG_STATE__PAUSE};
 	int vcn_inst;
 	uint32_t tmp;
 
@@ -582,6 +661,9 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 	if (indirect)
 		amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM);
 
+	/* Pause dpg */
+	vcn_v5_0_1_pause_dpg_mode(vinst, &state);
+
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
 	WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_LO, lower_32_bits(ring->gpu_addr));
@@ -613,6 +695,195 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 	return 0;
 }
 
+static int vcn_v5_0_1_start_sriov(struct amdgpu_device *adev)
+{
+	int i, vcn_inst;
+	struct amdgpu_ring *ring_enc;
+	uint64_t cache_addr;
+	uint64_t rb_enc_addr;
+	uint64_t ctx_addr;
+	uint32_t param, resp, expected;
+	uint32_t offset, cache_size;
+	uint32_t tmp, timeout;
+
+	struct amdgpu_mm_table *table = &adev->virt.mm_table;
+	uint32_t *table_loc;
+	uint32_t table_size;
+	uint32_t size, size_dw;
+	uint32_t init_status;
+	uint32_t enabled_vcn;
+
+	struct mmsch_v5_0_cmd_direct_write
+		direct_wt = { {0} };
+	struct mmsch_v5_0_cmd_direct_read_modify_write
+		direct_rd_mod_wt = { {0} };
+	struct mmsch_v5_0_cmd_end end = { {0} };
+	struct mmsch_v5_0_init_header header;
+
+	volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+	volatile struct amdgpu_fw_shared_rb_setup *rb_setup;
+
+	direct_wt.cmd_header.command_type =
+		MMSCH_COMMAND__DIRECT_REG_WRITE;
+	direct_rd_mod_wt.cmd_header.command_type =
+		MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE;
+	end.cmd_header.command_type = MMSCH_COMMAND__END;
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+		vcn_inst = GET_INST(VCN, i);
+
+		vcn_v5_0_1_fw_shared_init(adev, vcn_inst);
+
+		memset(&header, 0, sizeof(struct mmsch_v5_0_init_header));
+		header.version = MMSCH_VERSION;
+		header.total_size = sizeof(struct mmsch_v5_0_init_header) >> 2;
+
+		table_loc = (uint32_t *)table->cpu_addr;
+		table_loc += header.total_size;
+
+		table_size = 0;
+
+		MMSCH_V5_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(VCN, 0, regUVD_STATUS),
+			~UVD_STATUS__UVD_BUSY, UVD_STATUS__UVD_BUSY);
+
+		cache_size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.inst[i].fw->size + 4);
+
+		if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+			MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+				regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW),
+				adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_lo);
+
+			MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+				regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH),
+				adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_hi);
+
+			offset = 0;
+			MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+				regUVD_VCPU_CACHE_OFFSET0), 0);
+		} else {
+			MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+				regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW),
+				lower_32_bits(adev->vcn.inst[i].gpu_addr));
+			MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+				regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH),
+				upper_32_bits(adev->vcn.inst[i].gpu_addr));
+			offset = cache_size;
+			MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+				regUVD_VCPU_CACHE_OFFSET0),
+				AMDGPU_UVD_FIRMWARE_OFFSET >> 3);
+		}
+
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_VCPU_CACHE_SIZE0),
+			cache_size);
+
+		cache_addr = adev->vcn.inst[vcn_inst].gpu_addr + offset;
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), lower_32_bits(cache_addr));
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), upper_32_bits(cache_addr));
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_VCPU_CACHE_OFFSET1), 0);
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE);
+
+		cache_addr = adev->vcn.inst[vcn_inst].gpu_addr + offset +
+			AMDGPU_VCN_STACK_SIZE;
+
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), lower_32_bits(cache_addr));
+
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), upper_32_bits(cache_addr));
+
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_VCPU_CACHE_OFFSET2), 0);
+
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE);
+
+		fw_shared = adev->vcn.inst[vcn_inst].fw_shared.cpu_addr;
+		rb_setup = &fw_shared->rb_setup;
+
+		ring_enc = &adev->vcn.inst[vcn_inst].ring_enc[0];
+		ring_enc->wptr = 0;
+		rb_enc_addr = ring_enc->gpu_addr;
+
+		rb_setup->is_rb_enabled_flags |= RB_ENABLED;
+		rb_setup->rb_addr_lo = lower_32_bits(rb_enc_addr);
+		rb_setup->rb_addr_hi = upper_32_bits(rb_enc_addr);
+		rb_setup->rb_size = ring_enc->ring_size / 4;
+		fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_VF_RB_SETUP_FLAG);
+
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW),
+			lower_32_bits(adev->vcn.inst[vcn_inst].fw_shared.gpu_addr));
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH),
+			upper_32_bits(adev->vcn.inst[vcn_inst].fw_shared.gpu_addr));
+		MMSCH_V5_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0,
+			regUVD_VCPU_NONCACHE_SIZE0),
+			AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)));
+		MMSCH_V5_0_INSERT_END();
+
+		header.vcn0.init_status = 0;
+		header.vcn0.table_offset = header.total_size;
+		header.vcn0.table_size = table_size;
+		header.total_size += table_size;
+
+		/* Send init table to mmsch */
+		size = sizeof(struct mmsch_v5_0_init_header);
+		table_loc = (uint32_t *)table->cpu_addr;
+		memcpy((void *)table_loc, &header, size);
+
+		ctx_addr = table->gpu_addr;
+		WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_CTX_ADDR_LO, lower_32_bits(ctx_addr));
+		WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_CTX_ADDR_HI, upper_32_bits(ctx_addr));
+
+		tmp = RREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_VMID);
+		tmp &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK;
+		tmp |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT);
+		WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_VMID, tmp);
+
+		size = header.total_size;
+		WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_CTX_SIZE, size);
+
+		WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_MAILBOX_RESP, 0);
+
+		param = 0x00000001;
+		WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_MAILBOX_HOST, param);
+		tmp = 0;
+		timeout = 1000;
+		resp = 0;
+		expected = MMSCH_VF_MAILBOX_RESP__OK;
+		while (resp != expected) {
+			resp = RREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_MAILBOX_RESP);
+			if (resp != 0)
+				break;
+
+			udelay(10);
+			tmp = tmp + 10;
+			if (tmp >= timeout) {
+				DRM_ERROR("failed to init MMSCH. TIME-OUT after %d usec"\
+					" waiting for regMMSCH_VF_MAILBOX_RESP "\
+					"(expected=0x%08x, readback=0x%08x)\n",
+					tmp, expected, resp);
+				return -EBUSY;
+			}
+		}
+
+		enabled_vcn = amdgpu_vcn_is_disabled_vcn(adev, VCN_DECODE_RING, 0) ? 1 : 0;
+		init_status = ((struct mmsch_v5_0_init_header *)(table_loc))->vcn0.init_status;
+		if (resp != expected && resp != MMSCH_VF_MAILBOX_RESP__INCOMPLETE
+					&& init_status != MMSCH_VF_ENGINE_STATUS__PASS) {
+			DRM_ERROR("MMSCH init status is incorrect! readback=0x%08x, header init "\
+				"status for VCN%x: 0x%x\n", resp, enabled_vcn, init_status);
+		}
+	}
+
+	return 0;
+}
+
 /**
  * vcn_v5_0_1_start - VCN start
  *
@@ -775,9 +1046,13 @@ static void vcn_v5_0_1_stop_dpg_mode(struct amdgpu_vcn_inst *vinst)
 	int inst_idx = vinst->inst;
 	uint32_t tmp;
 	int vcn_inst;
+	struct dpg_pause_state state = {.fw_based = VCN_DPG_STATE__UNPAUSE};
 
 	vcn_inst = GET_INST(VCN, inst_idx);
 
+	/* Unpause dpg */
+	vcn_v5_0_1_pause_dpg_mode(vinst, &state);
+
 	/* Wait for power status to be 1 */
 	SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_POWER_STATUS, 1,
 		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
@@ -1049,8 +1324,18 @@ static int vcn_v5_0_1_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 static int vcn_v5_0_1_set_pg_state(struct amdgpu_vcn_inst *vinst,
 				   enum amd_powergating_state state)
 {
+	struct amdgpu_device *adev = vinst->adev;
 	int ret = 0;
 
+	/* for SRIOV, guest should not control VCN Power-gating
+	 * MMSCH FW should control Power-gating and clock-gating
+	 * guest should avoid touching CGC and PG
+	 */
+	if (amdgpu_sriov_vf(adev)) {
+		vinst->cur_state = AMD_PG_STATE_UNGATE;
+		return 0;
+	}
+
 	if (state == vinst->cur_state)
 		return 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
index faa0dd75dd6d..85846fd08ce4 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
@@ -350,6 +350,7 @@ static int vega20_ih_irq_init(struct amdgpu_device *adev)
 			if (ret)
 				return ret;
 		}
+		ih[i]->overflow = false;
 	}
 
 	if (!amdgpu_sriov_vf(adev))
@@ -437,7 +438,10 @@ static u32 vega20_ih_get_wptr(struct amdgpu_device *adev,
 	if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW))
 		goto out;
 
-	wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
+	if (!amdgpu_sriov_vf(adev))
+		wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0);
+	else
+		ih->overflow = true;
 
 	/* When a ring buffer overflow happen start parsing interrupt
 	 * from the last not overwritten vector (wptr + 32). Hopefully
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 86d8bc10d90a..9b3510e53112 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -239,6 +239,13 @@ static const struct amdgpu_video_codec_info cz_video_codecs_decode_array[] =
 		.max_pixels_per_frame = 4096 * 4096,
 		.max_level = 186,
 	},
+	{
+		.codec_type = AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG,
+		.max_width = 4096,
+		.max_height = 4096,
+		.max_pixels_per_frame = 4096 * 4096,
+		.max_level = 0,
+	},
 };
 
 static const struct amdgpu_video_codecs cz_video_codecs_decode =
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c b/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c
index 4a5a0a4e00f2..9bde2c64540f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c
@@ -27,6 +27,16 @@
 #include "kfd_priv.h"
 
 static struct dentry *debugfs_root;
+static struct dentry *debugfs_proc;
+static struct list_head procs;
+
+struct debugfs_proc_entry {
+	struct list_head list;
+	struct dentry *proc_dentry;
+	pid_t pid;
+};
+
+#define MAX_DEBUGFS_FILENAME_LEN 32
 
 static int kfd_debugfs_open(struct inode *inode, struct file *file)
 {
@@ -92,6 +102,8 @@ static const struct file_operations kfd_debugfs_hang_hws_fops = {
 void kfd_debugfs_init(void)
 {
 	debugfs_root = debugfs_create_dir("kfd", NULL);
+	debugfs_proc = debugfs_create_dir("proc", debugfs_root);
+	INIT_LIST_HEAD(&procs);
 
 	debugfs_create_file("mqds", S_IFREG | 0444, debugfs_root,
 			    kfd_debugfs_mqds_by_process, &kfd_debugfs_fops);
@@ -107,5 +119,69 @@ void kfd_debugfs_init(void)
 
 void kfd_debugfs_fini(void)
 {
+	debugfs_remove_recursive(debugfs_proc);
 	debugfs_remove_recursive(debugfs_root);
 }
+
+static ssize_t kfd_debugfs_pasid_read(struct file *file, char __user *buf,
+				      size_t count, loff_t *ppos)
+{
+	struct kfd_process_device *pdd = file_inode(file)->i_private;
+	char tmp[32];
+	int len;
+
+	len = snprintf(tmp, sizeof(tmp), "%u\n", pdd->pasid);
+
+	return simple_read_from_buffer(buf, count, ppos, tmp, len);
+}
+
+static const struct file_operations kfd_debugfs_pasid_fops = {
+	.owner = THIS_MODULE,
+	.read = kfd_debugfs_pasid_read,
+};
+
+void kfd_debugfs_add_process(struct kfd_process *p)
+{
+	int i;
+	char name[MAX_DEBUGFS_FILENAME_LEN];
+	struct debugfs_proc_entry *entry;
+
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		return;
+
+	list_add(&entry->list, &procs);
+	entry->pid = p->lead_thread->pid;
+	snprintf(name, MAX_DEBUGFS_FILENAME_LEN, "%d",
+		 (int)entry->pid);
+	entry->proc_dentry = debugfs_create_dir(name, debugfs_proc);
+
+	/* Create debugfs files for each GPU:
+	 * - proc/<pid>/pasid_<gpuid>
+	 */
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+
+		snprintf(name, MAX_DEBUGFS_FILENAME_LEN, "pasid_%u",
+			 pdd->dev->id);
+		debugfs_create_file((const char *)name, S_IFREG | 0444,
+				    entry->proc_dentry, pdd,
+				    &kfd_debugfs_pasid_fops);
+	}
+}
+
+void kfd_debugfs_remove_process(struct kfd_process *p)
+{
+	struct debugfs_proc_entry *entry, *next;
+
+	mutex_lock(&kfd_processes_mutex);
+	list_for_each_entry_safe(entry, next, &procs, list) {
+		if (entry->pid != p->lead_thread->pid)
+			continue;
+
+		debugfs_remove_recursive(entry->proc_dentry);
+		list_del(&entry->list);
+		kfree(entry);
+	}
+	mutex_unlock(&kfd_processes_mutex);
+}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index b9c82be6ce13..bf0854bd5555 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -352,11 +352,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 			f2g = &aldebaran_kfd2kgd;
 			break;
 		case IP_VERSION(9, 4, 3):
-			gfx_target_version = adev->rev_id >= 1 ? 90402
-					   : adev->flags & AMD_IS_APU ? 90400
-					   : 90401;
-			f2g = &gc_9_4_3_kfd2kgd;
-			break;
 		case IP_VERSION(9, 4, 4):
 			gfx_target_version = 90402;
 			f2g = &gc_9_4_3_kfd2kgd;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index c610e172a2b8..76359c6a3f3a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -1576,8 +1576,9 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 	int bit;
 
 	if (q->properties.type == KFD_QUEUE_TYPE_SDMA) {
-		if (bitmap_empty(dqm->sdma_bitmap, KFD_MAX_SDMA_QUEUES)) {
-			dev_err(dev, "No more SDMA queue to allocate\n");
+		if (bitmap_empty(dqm->sdma_bitmap, get_num_sdma_queues(dqm))) {
+			dev_warn(dev, "No more SDMA queue to allocate (%d total queues)\n",
+				 get_num_sdma_queues(dqm));
 			return -ENOMEM;
 		}
 
@@ -1602,8 +1603,9 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 		q->properties.sdma_queue_id = q->sdma_id /
 				kfd_get_num_sdma_engines(dqm->dev);
 	} else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) {
-		if (bitmap_empty(dqm->xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES)) {
-			dev_err(dev, "No more XGMI SDMA queue to allocate\n");
+		if (bitmap_empty(dqm->xgmi_sdma_bitmap, get_num_xgmi_sdma_queues(dqm))) {
+			dev_warn(dev, "No more XGMI SDMA queue to allocate (%d total queues)\n",
+				 get_num_xgmi_sdma_queues(dqm));
 			return -ENOMEM;
 		}
 		if (restore_sdma_id) {
@@ -1662,8 +1664,8 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 		}
 
 		if (!free_bit_found) {
-			dev_err(dev, "No more SDMA queue to allocate for target ID %i\n",
-				q->properties.sdma_engine_id);
+			dev_warn(dev, "No more SDMA queue to allocate for target ID %i (%d total queues)\n",
+				 q->properties.sdma_engine_id, num_queues);
 			return -ENOMEM;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index fecdb6794075..e54e708ed82d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -1177,6 +1177,25 @@ void kfd_signal_hw_exception_event(u32 pasid)
 	kfd_unref_process(p);
 }
 
+void kfd_signal_vm_fault_event_with_userptr(struct kfd_process *p, uint64_t gpu_va)
+{
+	struct kfd_process_device *pdd;
+	struct kfd_hsa_memory_exception_data exception_data;
+	int i;
+
+	memset(&exception_data, 0, sizeof(exception_data));
+	exception_data.va = gpu_va;
+	exception_data.failure.NotPresent = 1;
+
+	// Send VM seg fault to all kfd process device
+	for (i = 0; i < p->n_pdds; i++) {
+		pdd = p->pdds[i];
+		exception_data.gpu_id = pdd->user_gpu_id;
+		kfd_evict_process_device(pdd);
+		kfd_signal_vm_fault_event(pdd, NULL, &exception_data);
+	}
+}
+
 void kfd_signal_vm_fault_event(struct kfd_process_device *pdd,
 				struct kfd_vm_fault_info *info,
 				struct kfd_hsa_memory_exception_data *data)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
index 37b69fe0ede3..3e1ad8974797 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
@@ -168,14 +168,14 @@ static bool event_interrupt_isr_v10(struct kfd_node *dev,
 	    client_id != SOC15_IH_CLIENTID_SE3SH)
 		return false;
 
-	pr_debug("client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
-		 client_id, source_id, vmid, pasid);
-	pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
-		 data[0], data[1], data[2], data[3],
-		 data[4], data[5], data[6], data[7]);
+	dev_dbg(dev->adev->dev,
+		"client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
+		client_id, source_id, vmid, pasid);
+	dev_dbg(dev->adev->dev, "%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
+		data[0], data[1], data[2], data[3], data[4], data[5], data[6],
+		data[7]);
 
-	/* If there is no valid PASID, it's likely a bug */
-	if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt"))
+	if (pasid == 0)
 		return 0;
 
 	/* Interrupt types we care about: various signals and faults.
@@ -217,37 +217,66 @@ static void event_interrupt_wq_v10(struct kfd_node *dev,
 						SQ_INTERRUPT_WORD_WAVE_CTXID1, ENCODING);
 			switch (encoding) {
 			case SQ_INTERRUPT_WORD_ENCODING_AUTO:
-				pr_debug_ratelimited(
+				dev_dbg_ratelimited(
+					dev->adev->dev,
 					"sq_intr: auto, se %d, ttrace %d, wlt %d, ttrac_buf0_full %d, ttrac_buf1_full %d, ttrace_utc_err %d\n",
-					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_AUTO_CTXID1,
-							SE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
-							THREAD_TRACE),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
-							WLT),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
-							THREAD_TRACE_BUF0_FULL),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
-							THREAD_TRACE_BUF1_FULL),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
-							THREAD_TRACE_UTC_ERROR));
+					REG_GET_FIELD(
+						context_id1,
+						SQ_INTERRUPT_WORD_AUTO_CTXID1,
+						SE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID0,
+						THREAD_TRACE),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID0,
+						WLT),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID0,
+						THREAD_TRACE_BUF0_FULL),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID0,
+						THREAD_TRACE_BUF1_FULL),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID0,
+						THREAD_TRACE_UTC_ERROR));
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_INST:
-				pr_debug_ratelimited("sq_intr: inst, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d\n",
-					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
-							SE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							DATA),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							SA_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							PRIV),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							WAVE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							SIMD_ID),
-					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
-							WGP_ID));
+				dev_dbg_ratelimited(
+					dev->adev->dev,
+					"sq_intr: inst, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d\n",
+					REG_GET_FIELD(
+						context_id1,
+						SQ_INTERRUPT_WORD_WAVE_CTXID1,
+						SE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						DATA),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						SA_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						PRIV),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						WAVE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						SIMD_ID),
+					REG_GET_FIELD(
+						context_id1,
+						SQ_INTERRUPT_WORD_WAVE_CTXID1,
+						WGP_ID));
 				if (context_id0 & SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV_MASK) {
 					if (kfd_set_dbg_ev_from_interrupt(dev, pasid,
 							KFD_DEBUG_DOORBELL_ID(context_id0),
@@ -259,21 +288,37 @@ static void event_interrupt_wq_v10(struct kfd_node *dev,
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
 				sq_intr_err_type = REG_GET_FIELD(context_id0, KFD_CTXID0,
 								ERR_TYPE);
-				pr_warn_ratelimited("sq_intr: error, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d, err_type %d\n",
-					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
-							SE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							DATA),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							SA_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							PRIV),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							WAVE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
-							SIMD_ID),
-					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
-							WGP_ID),
+				dev_warn_ratelimited(
+					dev->adev->dev,
+					"sq_intr: error, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d, err_type %d\n",
+					REG_GET_FIELD(
+						context_id1,
+						SQ_INTERRUPT_WORD_WAVE_CTXID1,
+						SE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						DATA),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						SA_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						PRIV),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						WAVE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0,
+						SIMD_ID),
+					REG_GET_FIELD(
+						context_id1,
+						SQ_INTERRUPT_WORD_WAVE_CTXID1,
+						WGP_ID),
 					sq_intr_err_type);
 				break;
 			default:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index c5f97e6e36ff..2788a52714d1 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -148,44 +148,69 @@ enum SQ_INTERRUPT_ERROR_TYPE {
 #define KFD_CTXID0_DOORBELL_ID(ctxid0)		((ctxid0) & \
 				KFD_CTXID0_DOORBELL_ID_MASK)
 
-static void print_sq_intr_info_auto(uint32_t context_id0, uint32_t context_id1)
+static void print_sq_intr_info_auto(struct kfd_node *dev, uint32_t context_id0,
+				    uint32_t context_id1)
 {
-	pr_debug_ratelimited(
+	dev_dbg_ratelimited(
+		dev->adev->dev,
 		"sq_intr: auto, ttrace %d, wlt %d, ttrace_buf_full %d, reg_tms %d, cmd_tms %d, host_cmd_ovf %d, host_reg_ovf %d, immed_ovf %d, ttrace_utc_err %d\n",
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, THREAD_TRACE),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      THREAD_TRACE),
 		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, WLT),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, THREAD_TRACE_BUF_FULL),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, REG_TIMESTAMP),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, CMD_TIMESTAMP),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, HOST_CMD_OVERFLOW),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, HOST_REG_OVERFLOW),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, IMMED_OVERFLOW),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, THREAD_TRACE_UTC_ERROR));
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      THREAD_TRACE_BUF_FULL),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      REG_TIMESTAMP),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      CMD_TIMESTAMP),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      HOST_CMD_OVERFLOW),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      HOST_REG_OVERFLOW),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      IMMED_OVERFLOW),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+			      THREAD_TRACE_UTC_ERROR));
 }
 
-static void print_sq_intr_info_inst(uint32_t context_id0, uint32_t context_id1)
+static void print_sq_intr_info_inst(struct kfd_node *dev, uint32_t context_id0,
+				    uint32_t context_id1)
 {
-	pr_debug_ratelimited(
+	dev_dbg_ratelimited(
+		dev->adev->dev,
 		"sq_intr: inst, data 0x%08x, sh %d, priv %d, wave_id %d, simd_id %d, wgp_id %d\n",
 		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, DATA),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, SH_ID),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+			      SH_ID),
 		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, PRIV),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, WAVE_ID),
-		REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1, SIMD_ID),
-		REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1, WGP_ID));
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+			      WAVE_ID),
+		REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
+			      SIMD_ID),
+		REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
+			      WGP_ID));
 }
 
-static void print_sq_intr_info_error(uint32_t context_id0, uint32_t context_id1)
+static void print_sq_intr_info_error(struct kfd_node *dev, uint32_t context_id0,
+				     uint32_t context_id1)
 {
-	pr_warn_ratelimited(
+	dev_warn_ratelimited(
+		dev->adev->dev,
 		"sq_intr: error, detail 0x%08x, type %d, sh %d, priv %d, wave_id %d, simd_id %d, wgp_id %d\n",
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0, DETAIL),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0, TYPE),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0, SH_ID),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0, PRIV),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0, WAVE_ID),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID1, SIMD_ID),
-		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID1, WGP_ID));
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0,
+			      DETAIL),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0,
+			      TYPE),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0,
+			      SH_ID),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0,
+			      PRIV),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID0,
+			      WAVE_ID),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID1,
+			      SIMD_ID),
+		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID1,
+			      WGP_ID));
 }
 
 static void event_interrupt_poison_consumption_v11(struct kfd_node *dev,
@@ -255,14 +280,14 @@ static bool event_interrupt_isr_v11(struct kfd_node *dev,
 	    (context_id0 & AMDGPU_FENCE_MES_QUEUE_FLAG))
 		return false;
 
-	pr_debug("client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
-		 client_id, source_id, vmid, pasid);
-	pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
-		 data[0], data[1], data[2], data[3],
-		 data[4], data[5], data[6], data[7]);
+	dev_dbg(dev->adev->dev,
+		"client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
+		client_id, source_id, vmid, pasid);
+	dev_dbg(dev->adev->dev, "%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
+		data[0], data[1], data[2], data[3], data[4], data[5], data[6],
+		data[7]);
 
-	/* If there is no valid PASID, it's likely a bug */
-	if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt"))
+	if (pasid == 0)
 		return false;
 
 	/* Interrupt types we care about: various signals and faults.
@@ -353,10 +378,10 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 					SQ_INTERRUPT_WORD_WAVE_CTXID1, ENCODING);
 			switch (sq_int_enc) {
 			case SQ_INTERRUPT_WORD_ENCODING_AUTO:
-				print_sq_intr_info_auto(context_id0, context_id1);
+				print_sq_intr_info_auto(dev, context_id0, context_id1);
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_INST:
-				print_sq_intr_info_inst(context_id0, context_id1);
+				print_sq_intr_info_inst(dev, context_id0, context_id1);
 				sq_int_priv = REG_GET_FIELD(context_id0,
 						SQ_INTERRUPT_WORD_WAVE_CTXID0, PRIV);
 				if (sq_int_priv && (kfd_set_dbg_ev_from_interrupt(dev, pasid,
@@ -366,7 +391,7 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 					return;
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
-				print_sq_intr_info_error(context_id0, context_id1);
+				print_sq_intr_info_error(dev, context_id0, context_id1);
 				sq_int_errtype = REG_GET_FIELD(context_id0,
 						SQ_INTERRUPT_WORD_ERROR_CTXID0, TYPE);
 				if (sq_int_errtype != SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST &&
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index b8a91bf4ef30..4ceb251312a6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -314,11 +314,12 @@ static bool event_interrupt_isr_v9(struct kfd_node *dev,
 					& ~pasid_mask) | pasid);
 	}
 
-	pr_debug("client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
-		 client_id, source_id, vmid, pasid);
-	pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
-		 data[0], data[1], data[2], data[3],
-		 data[4], data[5], data[6], data[7]);
+	dev_dbg(dev->adev->dev,
+		"client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
+		client_id, source_id, vmid, pasid);
+	dev_dbg(dev->adev->dev, "%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
+		data[0], data[1], data[2], data[3], data[4], data[5], data[6],
+		data[7]);
 
 	/* If there is no valid PASID, it's likely a bug */
 	if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt"))
@@ -379,28 +380,82 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 			encoding = REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, ENCODING);
 			switch (encoding) {
 			case SQ_INTERRUPT_WORD_ENCODING_AUTO:
-				pr_debug_ratelimited(
+				dev_dbg_ratelimited(
+					dev->adev->dev,
 					"sq_intr: auto, se %d, ttrace %d, wlt %d, ttrac_buf_full %d, reg_tms %d, cmd_tms %d, host_cmd_ovf %d, host_reg_ovf %d, immed_ovf %d, ttrace_utc_err %d\n",
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, SE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, THREAD_TRACE),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, WLT),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, THREAD_TRACE_BUF_FULL),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, REG_TIMESTAMP),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, CMD_TIMESTAMP),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, HOST_CMD_OVERFLOW),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, HOST_REG_OVERFLOW),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, IMMED_OVERFLOW),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID, THREAD_TRACE_UTC_ERROR));
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						SE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						THREAD_TRACE),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						WLT),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						THREAD_TRACE_BUF_FULL),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						REG_TIMESTAMP),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						CMD_TIMESTAMP),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						HOST_CMD_OVERFLOW),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						HOST_REG_OVERFLOW),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						IMMED_OVERFLOW),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_AUTO_CTXID,
+						THREAD_TRACE_UTC_ERROR));
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_INST:
-				pr_debug_ratelimited("sq_intr: inst, se %d, data 0x%x, sh %d, priv %d, wave_id %d, simd_id %d, cu_id %d, intr_data 0x%x\n",
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, DATA),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SH_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, PRIV),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, WAVE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SIMD_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, CU_ID),
+				dev_dbg_ratelimited(
+					dev->adev->dev,
+					"sq_intr: inst, se %d, data 0x%x, sh %d, priv %d, wave_id %d, simd_id %d, cu_id %d, intr_data 0x%x\n",
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						SE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						DATA),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						SH_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						PRIV),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						WAVE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						SIMD_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						CU_ID),
 					sq_int_data);
 				if (context_id0 & SQ_INTERRUPT_WORD_WAVE_CTXID__PRIV_MASK) {
 					if (kfd_set_dbg_ev_from_interrupt(dev, pasid,
@@ -412,14 +467,37 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
 				sq_intr_err = REG_GET_FIELD(sq_int_data, KFD_SQ_INT_DATA, ERR_TYPE);
-				pr_warn_ratelimited("sq_intr: error, se %d, data 0x%x, sh %d, priv %d, wave_id %d, simd_id %d, cu_id %d, err_type %d\n",
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, DATA),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SH_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, PRIV),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, WAVE_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SIMD_ID),
-					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, CU_ID),
+				dev_warn_ratelimited(
+					dev->adev->dev,
+					"sq_intr: error, se %d, data 0x%x, sh %d, priv %d, wave_id %d, simd_id %d, cu_id %d, err_type %d\n",
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						SE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						DATA),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						SH_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						PRIV),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						WAVE_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						SIMD_ID),
+					REG_GET_FIELD(
+						context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID,
+						CU_ID),
 					sq_intr_err);
 				if (sq_intr_err != SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST &&
 					sq_intr_err != SQ_INTERRUPT_ERROR_TYPE_MEMVIOL) {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 80320a6c8854..97933d2a3803 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -495,6 +495,10 @@ static void update_mqd_sdma(struct mqd_manager *mm, void *mqd,
 	m->sdma_engine_id = q->sdma_engine_id;
 	m->sdma_queue_id = q->sdma_queue_id;
 	m->sdmax_rlcx_dummy_reg = SDMA_RLC_DUMMY_DEFAULT;
+	/* Allow context switch so we don't cross-process starve with a massive
+	 * command buffer of long-running SDMA commands
+	 */
+	m->sdmax_rlcx_ib_cntl |= SDMA0_GFX_IB_CNTL__SWITCH_INSIDE_IB_MASK;
 
 	q->is_active = QUEUE_IS_ACTIVE(*q);
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
index 271c567242ab..b1a6eb349bb3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
@@ -31,6 +31,7 @@
 #define OVER_SUBSCRIPTION_PROCESS_COUNT (1 << 0)
 #define OVER_SUBSCRIPTION_COMPUTE_QUEUE_COUNT (1 << 1)
 #define OVER_SUBSCRIPTION_GWS_QUEUE_COUNT (1 << 2)
+#define OVER_SUBSCRIPTION_XNACK_CONFLICT (1 << 3)
 
 static inline void inc_wptr(unsigned int *wptr, unsigned int increment_bytes,
 				unsigned int buffer_size_bytes)
@@ -44,7 +45,8 @@ static inline void inc_wptr(unsigned int *wptr, unsigned int increment_bytes,
 
 static void pm_calc_rlib_size(struct packet_manager *pm,
 				unsigned int *rlib_size,
-				int *over_subscription)
+				int *over_subscription,
+				int xnack_conflict)
 {
 	unsigned int process_count, queue_count, compute_queue_count, gws_queue_count;
 	unsigned int map_queue_size;
@@ -73,6 +75,8 @@ static void pm_calc_rlib_size(struct packet_manager *pm,
 		*over_subscription |= OVER_SUBSCRIPTION_COMPUTE_QUEUE_COUNT;
 	if (gws_queue_count > 1)
 		*over_subscription |= OVER_SUBSCRIPTION_GWS_QUEUE_COUNT;
+	if (xnack_conflict && (node->adev->gmc.xnack_flags & AMDGPU_GMC_XNACK_FLAG_CHAIN))
+		*over_subscription |= OVER_SUBSCRIPTION_XNACK_CONFLICT;
 
 	if (*over_subscription)
 		dev_dbg(dev, "Over subscribed runlist\n");
@@ -96,7 +100,8 @@ static int pm_allocate_runlist_ib(struct packet_manager *pm,
 				unsigned int **rl_buffer,
 				uint64_t *rl_gpu_buffer,
 				unsigned int *rl_buffer_size,
-				int *is_over_subscription)
+				int *is_over_subscription,
+				int xnack_conflict)
 {
 	struct kfd_node *node = pm->dqm->dev;
 	struct device *dev = node->adev->dev;
@@ -105,7 +110,8 @@ static int pm_allocate_runlist_ib(struct packet_manager *pm,
 	if (WARN_ON(pm->allocated))
 		return -EINVAL;
 
-	pm_calc_rlib_size(pm, rl_buffer_size, is_over_subscription);
+	pm_calc_rlib_size(pm, rl_buffer_size, is_over_subscription,
+				xnack_conflict);
 
 	mutex_lock(&pm->lock);
 
@@ -142,11 +148,27 @@ static int pm_create_runlist_ib(struct packet_manager *pm,
 	struct queue *q;
 	struct kernel_queue *kq;
 	int is_over_subscription;
+	int xnack_enabled = -1;
+	bool xnack_conflict = 0;
 
 	rl_wptr = retval = processes_mapped = 0;
 
+	/* Check if processes set different xnack modes */
+	list_for_each_entry(cur, queues, list) {
+		qpd = cur->qpd;
+		if (xnack_enabled < 0)
+			/* First process */
+			xnack_enabled = qpd->pqm->process->xnack_enabled;
+		else if (qpd->pqm->process->xnack_enabled != xnack_enabled) {
+			/* Found a process with a different xnack mode */
+			xnack_conflict = 1;
+			break;
+		}
+	}
+
 	retval = pm_allocate_runlist_ib(pm, &rl_buffer, rl_gpu_addr,
-				&alloc_size_bytes, &is_over_subscription);
+				&alloc_size_bytes, &is_over_subscription,
+				xnack_conflict);
 	if (retval)
 		return retval;
 
@@ -156,9 +178,13 @@ static int pm_create_runlist_ib(struct packet_manager *pm,
 	dev_dbg(dev, "Building runlist ib process count: %d queues count %d\n",
 		pm->dqm->processes_count, pm->dqm->active_queue_count);
 
+build_runlist_ib:
 	/* build the run list ib packet */
 	list_for_each_entry(cur, queues, list) {
 		qpd = cur->qpd;
+		/* group processes with the same xnack mode together */
+		if (qpd->pqm->process->xnack_enabled != xnack_enabled)
+			continue;
 		/* build map process packet */
 		if (processes_mapped >= pm->dqm->processes_count) {
 			dev_dbg(dev, "Not enough space left in runlist IB\n");
@@ -215,18 +241,26 @@ static int pm_create_runlist_ib(struct packet_manager *pm,
 				alloc_size_bytes);
 		}
 	}
+	if (xnack_conflict) {
+		/* pick up processes with the other xnack mode */
+		xnack_enabled = !xnack_enabled;
+		xnack_conflict = 0;
+		goto build_runlist_ib;
+	}
 
 	dev_dbg(dev, "Finished map process and queues to runlist\n");
 
 	if (is_over_subscription) {
 		if (!pm->is_over_subscription)
-			dev_warn(dev, "Runlist is getting oversubscribed due to%s%s%s. Expect reduced ROCm performance.\n",
-				 is_over_subscription & OVER_SUBSCRIPTION_PROCESS_COUNT ?
-				 " too many processes." : "",
-				 is_over_subscription & OVER_SUBSCRIPTION_COMPUTE_QUEUE_COUNT ?
-				 " too many queues." : "",
-				 is_over_subscription & OVER_SUBSCRIPTION_GWS_QUEUE_COUNT ?
-				 " multiple processes using cooperative launch." : "");
+			dev_warn(dev, "Runlist is getting oversubscribed due to%s%s%s%s. Expect reduced ROCm performance.\n",
+				is_over_subscription & OVER_SUBSCRIPTION_PROCESS_COUNT ?
+				" too many processes" : "",
+				is_over_subscription & OVER_SUBSCRIPTION_COMPUTE_QUEUE_COUNT ?
+				" too many queues" : "",
+				is_over_subscription & OVER_SUBSCRIPTION_GWS_QUEUE_COUNT ?
+				" multiple processes using cooperative launch" : "",
+				is_over_subscription & OVER_SUBSCRIPTION_XNACK_CONFLICT ?
+				" xnack on/off processes mixed on gfx9" : "");
 
 		retval = pm->pmf->runlist(pm, &rl_buffer[rl_wptr],
 					*rl_gpu_addr,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
index 2893fd5e5d00..8fa6489b6f5d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
@@ -43,7 +43,7 @@ static int pm_map_process_v9(struct packet_manager *pm,
 	memset(buffer, 0, sizeof(struct pm4_mes_map_process));
 	packet->header.u32All = pm_build_pm4_header(IT_MAP_PROCESS,
 					sizeof(struct pm4_mes_map_process));
-	if (adev->enforce_isolation[kfd->node_id])
+	if (adev->enforce_isolation[kfd->node_id] == AMDGPU_ENFORCE_ISOLATION_ENABLE)
 		packet->bitfields2.exec_cleaner_shader = 1;
 	packet->bitfields2.diq_enable = (qpd->is_debug) ? 1 : 0;
 	packet->bitfields2.process_quantum = 10;
@@ -102,7 +102,8 @@ static int pm_map_process_aldebaran(struct packet_manager *pm,
 	memset(buffer, 0, sizeof(struct pm4_mes_map_process_aldebaran));
 	packet->header.u32All = pm_build_pm4_header(IT_MAP_PROCESS,
 			sizeof(struct pm4_mes_map_process_aldebaran));
-	if (adev->enforce_isolation[knode->node_id])
+	if (adev->enforce_isolation[knode->node_id] ==
+	    AMDGPU_ENFORCE_ISOLATION_ENABLE)
 		packet->bitfields2.exec_cleaner_shader = 1;
 	packet->bitfields2.diq_enable = (qpd->is_debug) ? 1 : 0;
 	packet->bitfields2.process_quantum = 10;
@@ -165,9 +166,9 @@ static int pm_runlist_v9(struct packet_manager *pm, uint32_t *buffer,
 	 * hws_max_conc_proc has been done in
 	 * kgd2kfd_device_init().
 	 */
-	concurrent_proc_cnt = adev->enforce_isolation[kfd->node_id] ?
-			1 : min(pm->dqm->processes_count,
-			kfd->max_proc_per_quantum);
+	concurrent_proc_cnt = (adev->enforce_isolation[kfd->node_id] ==
+			       AMDGPU_ENFORCE_ISOLATION_ENABLE) ?
+		1 : min(pm->dqm->processes_count, kfd->max_proc_per_quantum);
 
 	packet = (struct pm4_mes_runlist *)buffer;
 
@@ -202,6 +203,8 @@ static int pm_set_resources_v9(struct packet_manager *pm, uint32_t *buffer,
 			queue_type__mes_set_resources__hsa_interface_queue_hiq;
 	packet->bitfields2.vmid_mask = res->vmid_mask;
 	packet->bitfields2.unmap_latency = KFD_UNMAP_LATENCY_MS / 100;
+	if (pm->dqm->dev->adev->gmc.xnack_flags & AMDGPU_GMC_XNACK_FLAG_CHAIN)
+		packet->bitfields2.enb_xnack_retry_disable_check = 1;
 	packet->bitfields7.oac_mask = res->oac_mask;
 	packet->bitfields8.gds_heap_base = res->gds_heap_base;
 	packet->bitfields8.gds_heap_size = res->gds_heap_size;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
index cd8611401a66..e356a207d03c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
@@ -63,7 +63,8 @@ struct pm4_mes_set_resources {
 		struct {
 			uint32_t vmid_mask:16;
 			uint32_t unmap_latency:8;
-			uint32_t reserved1:5;
+			uint32_t reserved1:4;
+			uint32_t enb_xnack_retry_disable_check:1;
 			enum mes_set_resources_queue_type_enum queue_type:3;
 		} bitfields2;
 		uint32_t ordinal2;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index f6aedf69c644..d221c58dccc3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1507,6 +1507,8 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p,
 int kfd_get_num_events(struct kfd_process *p);
 int kfd_event_destroy(struct kfd_process *p, uint32_t event_id);
 
+void kfd_signal_vm_fault_event_with_userptr(struct kfd_process *p, uint64_t gpu_va);
+
 void kfd_signal_vm_fault_event(struct kfd_process_device *pdd,
 				struct kfd_vm_fault_info *info,
 				struct kfd_hsa_memory_exception_data *data);
@@ -1581,10 +1583,15 @@ int kfd_debugfs_hang_hws(struct kfd_node *dev);
 int pm_debugfs_hang_hws(struct packet_manager *pm);
 int dqm_debugfs_hang_hws(struct device_queue_manager *dqm);
 
+void kfd_debugfs_add_process(struct kfd_process *p);
+void kfd_debugfs_remove_process(struct kfd_process *p);
+
 #else
 
 static inline void kfd_debugfs_init(void) {}
 static inline void kfd_debugfs_fini(void) {}
+static inline void kfd_debugfs_add_process(struct kfd_process *p) {}
+static inline void kfd_debugfs_remove_process(struct kfd_process *p) {}
 
 #endif
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 7c0c24732481..722ac1662bdc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -900,6 +900,8 @@ struct kfd_process *kfd_create_process(struct task_struct *thread)
 		kfd_procfs_add_sysfs_files(process);
 		kfd_procfs_add_sysfs_counters(process);
 
+		kfd_debugfs_add_process(process);
+
 		init_waitqueue_head(&process->wait_irq_drain);
 	}
 out:
@@ -1054,6 +1056,8 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
 	for (i = 0; i < p->n_pdds; i++) {
 		struct kfd_process_device *pdd = p->pdds[i];
 
+		kfd_smi_event_process(pdd, false);
+
 		pr_debug("Releasing pdd (topology id %d, for pid %d)\n",
 			pdd->dev->id, p->lead_thread->pid);
 		kfd_process_device_destroy_cwsr_dgpu(pdd);
@@ -1174,6 +1178,7 @@ static void kfd_process_wq_release(struct work_struct *work)
 		dma_fence_signal(ef);
 
 	kfd_process_remove_sysfs(p);
+	kfd_debugfs_remove_process(p);
 
 	kfd_process_kunmap_signal_bo(p);
 	kfd_process_free_outstanding_kfd_bos(p);
@@ -1715,6 +1720,8 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd,
 	pdd->pasid = avm->pasid;
 	pdd->drm_file = drm_file;
 
+	kfd_smi_event_process(pdd, true);
+
 	return 0;
 
 err_get_pasid:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 7eb370b68159..6d5fa57d4a23 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -451,8 +451,15 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 	}
 
 	if (retval != 0) {
-		pr_err("process pid %d DQM create queue type %d failed. ret %d\n",
-			pqm->process->lead_thread->pid, type, retval);
+		if ((type == KFD_QUEUE_TYPE_SDMA ||
+		    type == KFD_QUEUE_TYPE_SDMA_XGMI ||
+		    type == KFD_QUEUE_TYPE_SDMA_BY_ENG_ID) &&
+		    retval == -ENOMEM)
+			pr_warn("process pid %d DQM create queue type %d failed. ret %d\n",
+				pqm->process->lead_thread->pid, type, retval);
+		else
+			pr_err("process pid %d DQM create queue type %d failed. ret %d\n",
+				pqm->process->lead_thread->pid, type, retval);
 		goto err_create_queue;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
index 4afff7094caf..a65c67cf56ff 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
@@ -402,7 +402,7 @@ static u32 kfd_get_vgpr_size_per_cu(u32 gfxv)
 {
 	u32 vgpr_size = 0x40000;
 
-	if ((gfxv / 100 * 100) == 90400 ||	/* GFX_VERSION_AQUA_VANJARAM */
+	if (gfxv == 90402 ||			/* GFX_VERSION_AQUA_VANJARAM */
 	    gfxv == 90010 ||			/* GFX_VERSION_ALDEBARAN */
 	    gfxv == 90008 ||			/* GFX_VERSION_ARCTURUS */
 	    gfxv == 90500)
@@ -462,7 +462,7 @@ void kfd_queue_ctx_save_restore_size(struct kfd_topology_device *dev)
 
 	if (gfxv == 80002)	/* GFX_VERSION_TONGA */
 		props->eop_buffer_size = 0x8000;
-	else if ((gfxv / 100 * 100) == 90400)	/* GFX_VERSION_AQUA_VANJARAM */
+	else if (gfxv == 90402)	/* GFX_VERSION_AQUA_VANJARAM */
 		props->eop_buffer_size = 4096;
 	else if (gfxv >= 80000)
 		props->eop_buffer_size = 4096;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index 9b8169761ec5..83d9384ac815 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
@@ -163,10 +163,9 @@ static int kfd_smi_ev_release(struct inode *inode, struct file *filep)
 static bool kfd_smi_ev_enabled(pid_t pid, struct kfd_smi_client *client,
 			       unsigned int event)
 {
-	uint64_t all = KFD_SMI_EVENT_MASK_FROM_INDEX(KFD_SMI_EVENT_ALL_PROCESS);
 	uint64_t events = READ_ONCE(client->events);
 
-	if (pid && client->pid != pid && !(client->suser && (events & all)))
+	if (pid && client->pid != pid && !client->suser)
 		return false;
 
 	return events & KFD_SMI_EVENT_MASK_FROM_INDEX(event);
@@ -345,6 +344,27 @@ void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid,
 			  pid, address, last - address + 1, node->id, trigger));
 }
 
+void kfd_smi_event_process(struct kfd_process_device *pdd, bool start)
+{
+	struct amdgpu_task_info *task_info;
+	struct amdgpu_vm *avm;
+
+	if (!pdd->drm_priv)
+		return;
+
+	avm = drm_priv_to_vm(pdd->drm_priv);
+	task_info = amdgpu_vm_get_task_info_vm(avm);
+
+	if (task_info) {
+		kfd_smi_event_add(0, pdd->dev,
+				  start ? KFD_SMI_EVENT_PROCESS_START :
+				  KFD_SMI_EVENT_PROCESS_END,
+				  KFD_EVENT_FMT_PROCESS(task_info->pid,
+				  task_info->task_name));
+		amdgpu_vm_put_task_info(task_info);
+	}
+}
+
 int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd)
 {
 	struct kfd_smi_client *client;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
index 503bff13d815..bb4d72b57387 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
@@ -53,4 +53,5 @@ void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm);
 void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid,
 				  unsigned long address, unsigned long last,
 				  uint32_t trigger);
+void kfd_smi_event_process(struct kfd_process_device *pdd, bool start);
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 100717a98ec1..72be6e152e88 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1245,8 +1245,7 @@ svm_range_get_pte_flags(struct kfd_node *node,
 	case IP_VERSION(9, 4, 4):
 	case IP_VERSION(9, 5, 0):
 		if (ext_coherent)
-			mtype_local = (gc_ip_version < IP_VERSION(9, 5, 0) && !node->adev->rev_id) ?
-					AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_CC;
+			mtype_local = AMDGPU_VM_MTYPE_CC;
 		else
 			mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC :
 				amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index e477d7509646..baa2374acdeb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1267,34 +1267,41 @@ static void kfd_set_recommended_sdma_engines(struct kfd_topology_device *to_dev,
 {
 	struct kfd_node *gpu = outbound_link->gpu;
 	struct amdgpu_device *adev = gpu->adev;
-	int num_xgmi_nodes = adev->gmc.xgmi.num_physical_nodes;
+	unsigned int num_xgmi_nodes = adev->gmc.xgmi.num_physical_nodes;
+	unsigned int num_xgmi_sdma_engines = kfd_get_num_xgmi_sdma_engines(gpu);
+	unsigned int num_sdma_engines = kfd_get_num_sdma_engines(gpu);
+	uint32_t sdma_eng_id_mask = (1 << num_sdma_engines) - 1;
+	uint32_t xgmi_sdma_eng_id_mask =
+			((1 << num_xgmi_sdma_engines) - 1) << num_sdma_engines;
+
 	bool support_rec_eng = !amdgpu_sriov_vf(adev) && to_dev->gpu &&
 		adev->aid_mask && num_xgmi_nodes && gpu->kfd->num_nodes == 1 &&
-		kfd_get_num_xgmi_sdma_engines(gpu) >= 14 &&
-		(!(adev->flags & AMD_IS_APU) && num_xgmi_nodes == 8);
+		num_xgmi_sdma_engines >= 6 && (!(adev->flags & AMD_IS_APU) &&
+		num_xgmi_nodes == 8);
 
 	if (support_rec_eng) {
 		int src_socket_id = adev->gmc.xgmi.physical_node_id;
 		int dst_socket_id = to_dev->gpu->adev->gmc.xgmi.physical_node_id;
+		unsigned int reshift = num_xgmi_sdma_engines == 6 ? 1 : 0;
 
 		outbound_link->rec_sdma_eng_id_mask =
-			1 << rec_sdma_eng_map[src_socket_id][dst_socket_id];
+			1 << (rec_sdma_eng_map[src_socket_id][dst_socket_id] >> reshift);
 		inbound_link->rec_sdma_eng_id_mask =
-			1 << rec_sdma_eng_map[dst_socket_id][src_socket_id];
-	} else {
-		int num_sdma_eng = kfd_get_num_sdma_engines(gpu);
-		int i, eng_offset = 0;
+			1 << (rec_sdma_eng_map[dst_socket_id][src_socket_id] >> reshift);
 
-		if (outbound_link->iolink_type == CRAT_IOLINK_TYPE_XGMI &&
-		    kfd_get_num_xgmi_sdma_engines(gpu) && to_dev->gpu) {
-			eng_offset = num_sdma_eng;
-			num_sdma_eng = kfd_get_num_xgmi_sdma_engines(gpu);
-		}
+		/* If recommended engine is out of range, need to reset the mask */
+		if (outbound_link->rec_sdma_eng_id_mask & sdma_eng_id_mask)
+			outbound_link->rec_sdma_eng_id_mask = xgmi_sdma_eng_id_mask;
+		if (inbound_link->rec_sdma_eng_id_mask & sdma_eng_id_mask)
+			inbound_link->rec_sdma_eng_id_mask = xgmi_sdma_eng_id_mask;
 
-		for (i = 0; i < num_sdma_eng; i++) {
-			outbound_link->rec_sdma_eng_id_mask |= (1 << (i + eng_offset));
-			inbound_link->rec_sdma_eng_id_mask |= (1 << (i + eng_offset));
-		}
+	} else {
+		uint32_t engine_mask = (outbound_link->iolink_type == CRAT_IOLINK_TYPE_XGMI &&
+				num_xgmi_sdma_engines && to_dev->gpu) ? xgmi_sdma_eng_id_mask :
+				sdma_eng_id_mask;
+
+		outbound_link->rec_sdma_eng_id_mask = engine_mask;
+		inbound_link->rec_sdma_eng_id_mask = engine_mask;
 	}
 }
 
@@ -1983,9 +1990,6 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev)
 	if (kfd_dbg_has_ttmps_always_setup(dev->gpu))
 		dev->node_props.debug_prop |= HSA_DBG_DISPATCH_INFO_ALWAYS_VALID;
 
-	if (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)
-		dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED;
-
 	if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0)) {
 		if (KFD_GC_VERSION(dev->gpu) == IP_VERSION(9, 4, 3) ||
 		    KFD_GC_VERSION(dev->gpu) == IP_VERSION(9, 4, 4))
@@ -2001,7 +2005,11 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev)
 			dev->node_props.capability |=
 				HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED;
 
-		dev->node_props.capability |= HSA_CAP_PER_QUEUE_RESET_SUPPORTED;
+		if (!amdgpu_sriov_vf(dev->gpu->adev))
+			dev->node_props.capability |= HSA_CAP_PER_QUEUE_RESET_SUPPORTED;
+
+		if (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)
+			dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED;
 	} else {
 		dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 |
 					HSA_DBG_WATCH_ADDR_MASK_HI_BIT;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile b/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile
index ab2a97e354da..7329b8cc2576 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile
@@ -38,6 +38,7 @@ AMDGPUDM = \
 	amdgpu_dm_pp_smu.o \
 	amdgpu_dm_psr.o \
 	amdgpu_dm_replay.o \
+	amdgpu_dm_quirks.o \
 	amdgpu_dm_wb.o
 
 ifdef CONFIG_DRM_AMD_DC_FP
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index d0d8ad5368c3..4b4e9241619f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -80,7 +80,6 @@
 #include <linux/power_supply.h>
 #include <linux/firmware.h>
 #include <linux/component.h>
-#include <linux/dmi.h>
 #include <linux/sort.h>
 
 #include <drm/display/drm_dp_mst_helper.h>
@@ -115,6 +114,8 @@
 #include "modules/inc/mod_freesync.h"
 #include "modules/power/power_helpers.h"
 
+static_assert(AMDGPU_DMUB_NOTIFICATION_MAX == DMUB_NOTIFICATION_MAX, "AMDGPU_DMUB_NOTIFICATION_MAX mismatch");
+
 #define FIRMWARE_RENOIR_DMUB "amdgpu/renoir_dmcub.bin"
 MODULE_FIRMWARE(FIRMWARE_RENOIR_DMUB);
 #define FIRMWARE_SIENNA_CICHLID_DMUB "amdgpu/sienna_cichlid_dmcub.bin"
@@ -280,7 +281,7 @@ static u32 dm_vblank_get_counter(struct amdgpu_device *adev, int crtc)
 	acrtc = adev->mode_info.crtcs[crtc];
 
 	if (!acrtc->dm_irq_params.stream) {
-		DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n",
+		drm_err(adev_to_drm(adev), "dc_stream_state is NULL for crtc '%d'!\n",
 			  crtc);
 		return 0;
 	}
@@ -301,7 +302,7 @@ static int dm_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
 	acrtc = adev->mode_info.crtcs[crtc];
 
 	if (!acrtc->dm_irq_params.stream) {
-		DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n",
+		drm_err(adev_to_drm(adev), "dc_stream_state is NULL for crtc '%d'!\n",
 			  crtc);
 		return 0;
 	}
@@ -372,6 +373,8 @@ get_crtc_by_otg_inst(struct amdgpu_device *adev,
 static inline bool is_dc_timing_adjust_needed(struct dm_crtc_state *old_state,
 					      struct dm_crtc_state *new_state)
 {
+	if (new_state->stream->adjust.timing_adjust_pending)
+		return true;
 	if (new_state->freesync_config.state ==  VRR_STATE_ACTIVE_FIXED)
 		return true;
 	else if (amdgpu_dm_crtc_vrr_active(old_state) != amdgpu_dm_crtc_vrr_active(new_state))
@@ -673,15 +676,21 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 
 	if (acrtc->dm_irq_params.stream &&
-	    acrtc->dm_irq_params.vrr_params.supported &&
-	    acrtc->dm_irq_params.freesync_config.state ==
-		    VRR_STATE_ACTIVE_VARIABLE) {
+		acrtc->dm_irq_params.vrr_params.supported) {
+		bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled;
+		bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled;
+		bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state == VRR_STATE_ACTIVE_VARIABLE;
+
 		mod_freesync_handle_v_update(adev->dm.freesync_module,
 					     acrtc->dm_irq_params.stream,
 					     &acrtc->dm_irq_params.vrr_params);
 
-		dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc->dm_irq_params.stream,
-					   &acrtc->dm_irq_params.vrr_params.adjust);
+		/* update vmin_vmax only if freesync is enabled, or only if PSR and REPLAY are disabled */
+		if (fs_active_var_en || (!fs_active_var_en && !replay_en && !psr_en)) {
+			dc_stream_adjust_vmin_vmax(adev->dm.dc,
+					acrtc->dm_irq_params.stream,
+					&acrtc->dm_irq_params.vrr_params.adjust);
+		}
 	}
 
 	/*
@@ -749,6 +758,29 @@ static void dmub_aux_setconfig_callback(struct amdgpu_device *adev,
 		complete(&adev->dm.dmub_aux_transfer_done);
 }
 
+static void dmub_aux_fused_io_callback(struct amdgpu_device *adev,
+					struct dmub_notification *notify)
+{
+	if (!adev || !notify) {
+		ASSERT(false);
+		return;
+	}
+
+	const struct dmub_cmd_fused_request *req = &notify->fused_request;
+	const uint8_t ddc_line = req->u.aux.ddc_line;
+
+	if (ddc_line >= ARRAY_SIZE(adev->dm.fused_io)) {
+		ASSERT(false);
+		return;
+	}
+
+	struct fused_io_sync *sync = &adev->dm.fused_io[ddc_line];
+
+	static_assert(sizeof(*req) <= sizeof(sync->reply_data), "Size mismatch");
+	memcpy(sync->reply_data, req, sizeof(*req));
+	complete(&sync->replied);
+}
+
 /**
  * dmub_hpd_callback - DMUB HPD interrupt processing callback.
  * @adev: amdgpu_device pointer
@@ -772,18 +804,18 @@ static void dmub_hpd_callback(struct amdgpu_device *adev,
 		return;
 
 	if (notify == NULL) {
-		DRM_ERROR("DMUB HPD callback notification was NULL");
+		drm_err(adev_to_drm(adev), "DMUB HPD callback notification was NULL");
 		return;
 	}
 
 	if (notify->link_index > adev->dm.dc->link_count) {
-		DRM_ERROR("DMUB HPD index (%u)is abnormal", notify->link_index);
+		drm_err(adev_to_drm(adev), "DMUB HPD index (%u)is abnormal", notify->link_index);
 		return;
 	}
 
 	/* Skip DMUB HPD IRQ in suspend/resume. We will probe them later. */
 	if (notify->type == DMUB_NOTIFICATION_HPD && adev->in_suspend) {
-		DRM_INFO("Skip DMUB HPD IRQ callback in suspend/resume\n");
+		drm_info(adev_to_drm(adev), "Skip DMUB HPD IRQ callback in suspend/resume\n");
 		return;
 	}
 
@@ -800,11 +832,11 @@ static void dmub_hpd_callback(struct amdgpu_device *adev,
 		aconnector = to_amdgpu_dm_connector(connector);
 		if (link && aconnector->dc_link == link) {
 			if (notify->type == DMUB_NOTIFICATION_HPD)
-				DRM_INFO("DMUB HPD IRQ callback: link_index=%u\n", link_index);
+				drm_info(adev_to_drm(adev), "DMUB HPD IRQ callback: link_index=%u\n", link_index);
 			else if (notify->type == DMUB_NOTIFICATION_HPD_IRQ)
-				DRM_INFO("DMUB HPD RX IRQ callback: link_index=%u\n", link_index);
+				drm_info(adev_to_drm(adev), "DMUB HPD RX IRQ callback: link_index=%u\n", link_index);
 			else
-				DRM_WARN("DMUB Unknown HPD callback type %d, link_index=%u\n",
+				drm_warn(adev_to_drm(adev), "DMUB Unknown HPD callback type %d, link_index=%u\n",
 						notify->type, link_index);
 
 			hpd_aconnector = aconnector;
@@ -816,7 +848,7 @@ static void dmub_hpd_callback(struct amdgpu_device *adev,
 	if (hpd_aconnector) {
 		if (notify->type == DMUB_NOTIFICATION_HPD) {
 			if (hpd_aconnector->dc_link->hpd_status == (notify->hpd_status == DP_HPD_PLUG))
-				DRM_WARN("DMUB reported hpd status unchanged. link_index=%u\n", link_index);
+				drm_warn(adev_to_drm(adev), "DMUB reported hpd status unchanged. link_index=%u\n", link_index);
 			handle_hpd_irq_helper(hpd_aconnector);
 		} else if (notify->type == DMUB_NOTIFICATION_HPD_IRQ) {
 			handle_hpd_rx_irq(hpd_aconnector);
@@ -835,7 +867,7 @@ static void dmub_hpd_callback(struct amdgpu_device *adev,
 static void dmub_hpd_sense_callback(struct amdgpu_device *adev,
 			      struct dmub_notification *notify)
 {
-	DRM_DEBUG_DRIVER("DMUB HPD SENSE callback.\n");
+	drm_dbg_driver(adev_to_drm(adev), "DMUB HPD SENSE callback.\n");
 }
 
 /**
@@ -871,7 +903,7 @@ static void dm_handle_hpd_work(struct work_struct *work)
 	dmub_hpd_wrk = container_of(work, struct dmub_hpd_work, handle_hpd_work);
 
 	if (!dmub_hpd_wrk->dmub_notify) {
-		DRM_ERROR("dmub_hpd_wrk dmub_notify is NULL");
+		drm_err(adev_to_drm(dmub_hpd_wrk->adev), "dmub_hpd_wrk dmub_notify is NULL");
 		return;
 	}
 
@@ -885,6 +917,30 @@ static void dm_handle_hpd_work(struct work_struct *work)
 
 }
 
+static const char *dmub_notification_type_str(enum dmub_notification_type e)
+{
+	switch (e) {
+	case DMUB_NOTIFICATION_NO_DATA:
+		return "NO_DATA";
+	case DMUB_NOTIFICATION_AUX_REPLY:
+		return "AUX_REPLY";
+	case DMUB_NOTIFICATION_HPD:
+		return "HPD";
+	case DMUB_NOTIFICATION_HPD_IRQ:
+		return "HPD_IRQ";
+	case DMUB_NOTIFICATION_SET_CONFIG_REPLY:
+		return "SET_CONFIG_REPLY";
+	case DMUB_NOTIFICATION_DPIA_NOTIFICATION:
+		return "DPIA_NOTIFICATION";
+	case DMUB_NOTIFICATION_HPD_SENSE_NOTIFY:
+		return "HPD_SENSE_NOTIFY";
+	case DMUB_NOTIFICATION_FUSED_IO:
+		return "FUSED_IO";
+	default:
+		return "<unknown>";
+	}
+}
+
 #define DMUB_TRACE_MAX_READ 64
 /**
  * dm_dmub_outbox1_low_irq() - Handles Outbox interrupt
@@ -902,22 +958,13 @@ static void dm_dmub_outbox1_low_irq(void *interrupt_params)
 	struct dmcub_trace_buf_entry entry = { 0 };
 	u32 count = 0;
 	struct dmub_hpd_work *dmub_hpd_wrk;
-	static const char *const event_type[] = {
-		"NO_DATA",
-		"AUX_REPLY",
-		"HPD",
-		"HPD_IRQ",
-		"SET_CONFIGC_REPLY",
-		"DPIA_NOTIFICATION",
-		"HPD_SENSE_NOTIFY",
-	};
 
 	do {
 		if (dc_dmub_srv_get_dmub_outbox0_msg(dm->dc, &entry)) {
 			trace_amdgpu_dmub_trace_high_irq(entry.trace_code, entry.tick_count,
 							entry.param0, entry.param1);
 
-			DRM_DEBUG_DRIVER("trace_code:%u, tick_count:%u, param0:%u, param1:%u\n",
+			drm_dbg_driver(adev_to_drm(adev), "trace_code:%u, tick_count:%u, param0:%u, param1:%u\n",
 				 entry.trace_code, entry.tick_count, entry.param0, entry.param1);
 		} else
 			break;
@@ -927,7 +974,7 @@ static void dm_dmub_outbox1_low_irq(void *interrupt_params)
 	} while (count <= DMUB_TRACE_MAX_READ);
 
 	if (count > DMUB_TRACE_MAX_READ)
-		DRM_DEBUG_DRIVER("Warning : count > DMUB_TRACE_MAX_READ");
+		drm_dbg_driver(adev_to_drm(adev), "Warning : count > DMUB_TRACE_MAX_READ");
 
 	if (dc_enable_dmub_notifications(adev->dm.dc) &&
 		irq_params->irq_src == DC_IRQ_SOURCE_DMCUB_OUTBOX) {
@@ -935,25 +982,25 @@ static void dm_dmub_outbox1_low_irq(void *interrupt_params)
 		do {
 			dc_stat_get_dmub_notification(adev->dm.dc, &notify);
 			if (notify.type >= ARRAY_SIZE(dm->dmub_thread_offload)) {
-				DRM_ERROR("DM: notify type %d invalid!", notify.type);
+				drm_err(adev_to_drm(adev), "DM: notify type %d invalid!", notify.type);
 				continue;
 			}
 			if (!dm->dmub_callback[notify.type]) {
-				DRM_WARN("DMUB notification skipped due to no handler: type=%s\n",
-					event_type[notify.type]);
+				drm_warn(adev_to_drm(adev), "DMUB notification skipped due to no handler: type=%s\n",
+					dmub_notification_type_str(notify.type));
 				continue;
 			}
 			if (dm->dmub_thread_offload[notify.type] == true) {
 				dmub_hpd_wrk = kzalloc(sizeof(*dmub_hpd_wrk), GFP_ATOMIC);
 				if (!dmub_hpd_wrk) {
-					DRM_ERROR("Failed to allocate dmub_hpd_wrk");
+					drm_err(adev_to_drm(adev), "Failed to allocate dmub_hpd_wrk");
 					return;
 				}
 				dmub_hpd_wrk->dmub_notify = kmemdup(&notify, sizeof(struct dmub_notification),
 								    GFP_ATOMIC);
 				if (!dmub_hpd_wrk->dmub_notify) {
 					kfree(dmub_hpd_wrk);
-					DRM_ERROR("Failed to allocate dmub_hpd_wrk->dmub_notify");
+					drm_err(adev_to_drm(adev), "Failed to allocate dmub_hpd_wrk->dmub_notify");
 					return;
 				}
 				INIT_WORK(&dmub_hpd_wrk->handle_hpd_work, dm_handle_hpd_work);
@@ -1011,10 +1058,10 @@ static void amdgpu_dm_fbc_init(struct drm_connector *connector)
 			    &compressor->gpu_addr, &compressor->cpu_addr);
 
 		if (r)
-			DRM_ERROR("DM: Failed to initialize FBC\n");
+			drm_err(adev_to_drm(adev), "DM: Failed to initialize FBC\n");
 		else {
 			adev->dm.dc->ctx->fbc_gpu_addr = compressor->gpu_addr;
-			DRM_INFO("DM: FBC alloc %lu\n", max_size*4);
+			drm_info(adev_to_drm(adev), "DM: FBC alloc %lu\n", max_size*4);
 		}
 
 	}
@@ -1179,13 +1226,13 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 		return 0;
 
 	if (!fb_info) {
-		DRM_ERROR("No framebuffer info for DMUB service.\n");
+		drm_err(adev_to_drm(adev), "No framebuffer info for DMUB service.\n");
 		return -EINVAL;
 	}
 
 	if (!dmub_fw) {
 		/* Firmware required for DMUB support. */
-		DRM_ERROR("No firmware provided for DMUB.\n");
+		drm_err(adev_to_drm(adev), "No firmware provided for DMUB.\n");
 		return -EINVAL;
 	}
 
@@ -1195,19 +1242,19 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 
 	status = dmub_srv_has_hw_support(dmub_srv, &has_hw_support);
 	if (status != DMUB_STATUS_OK) {
-		DRM_ERROR("Error checking HW support for DMUB: %d\n", status);
+		drm_err(adev_to_drm(adev), "Error checking HW support for DMUB: %d\n", status);
 		return -EINVAL;
 	}
 
 	if (!has_hw_support) {
-		DRM_INFO("DMUB unsupported on ASIC\n");
+		drm_info(adev_to_drm(adev), "DMUB unsupported on ASIC\n");
 		return 0;
 	}
 
 	/* Reset DMCUB if it was previously running - before we overwrite its memory. */
 	status = dmub_srv_hw_reset(dmub_srv);
 	if (status != DMUB_STATUS_OK)
-		DRM_WARN("Error resetting DMUB HW: %d\n", status);
+		drm_warn(adev_to_drm(adev), "Error resetting DMUB HW: %d\n", status);
 
 	hdr = (const struct dmcub_firmware_header_v1_0 *)dmub_fw->data;
 
@@ -1290,6 +1337,7 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 	case IP_VERSION(3, 5, 1):
 	case IP_VERSION(3, 6, 0):
 		hw_params.ips_sequential_ono = adev->external_rev_id > 0x10;
+		hw_params.lower_hbr3_phy_ssc = true;
 		break;
 	default:
 		break;
@@ -1297,14 +1345,14 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 
 	status = dmub_srv_hw_init(dmub_srv, &hw_params);
 	if (status != DMUB_STATUS_OK) {
-		DRM_ERROR("Error initializing DMUB HW: %d\n", status);
+		drm_err(adev_to_drm(adev), "Error initializing DMUB HW: %d\n", status);
 		return -EINVAL;
 	}
 
 	/* Wait for firmware load to finish. */
 	status = dmub_srv_wait_for_auto_load(dmub_srv, 100000);
 	if (status != DMUB_STATUS_OK)
-		DRM_WARN("Wait for DMUB auto-load failed: %d\n", status);
+		drm_warn(adev_to_drm(adev), "Wait for DMUB auto-load failed: %d\n", status);
 
 	/* Init DMCU and ABM if available. */
 	if (dmcu && abm) {
@@ -1315,11 +1363,11 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev)
 	if (!adev->dm.dc->ctx->dmub_srv)
 		adev->dm.dc->ctx->dmub_srv = dc_dmub_srv_create(adev->dm.dc, dmub_srv);
 	if (!adev->dm.dc->ctx->dmub_srv) {
-		DRM_ERROR("Couldn't allocate DC DMUB server!\n");
+		drm_err(adev_to_drm(adev), "Couldn't allocate DC DMUB server!\n");
 		return -ENOMEM;
 	}
 
-	DRM_INFO("DMUB hardware initialized: version=0x%08X\n",
+	drm_info(adev_to_drm(adev), "DMUB hardware initialized: version=0x%08X\n",
 		 adev->dm.dmcub_fw_version);
 
 	/* Keeping sanity checks off if
@@ -1362,18 +1410,18 @@ static void dm_dmub_hw_resume(struct amdgpu_device *adev)
 
 	status = dmub_srv_is_hw_init(dmub_srv, &init);
 	if (status != DMUB_STATUS_OK)
-		DRM_WARN("DMUB hardware init check failed: %d\n", status);
+		drm_warn(adev_to_drm(adev), "DMUB hardware init check failed: %d\n", status);
 
 	if (status == DMUB_STATUS_OK && init) {
 		/* Wait for firmware load to finish. */
 		status = dmub_srv_wait_for_auto_load(dmub_srv, 100000);
 		if (status != DMUB_STATUS_OK)
-			DRM_WARN("Wait for DMUB auto-load failed: %d\n", status);
+			drm_warn(adev_to_drm(adev), "Wait for DMUB auto-load failed: %d\n", status);
 	} else {
 		/* Perform the full hardware initialization. */
 		r = dm_dmub_hw_init(adev);
 		if (r)
-			DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r);
+			drm_err(adev_to_drm(adev), "DMUB interface failed to initialize: status=%d\n", r);
 	}
 }
 
@@ -1483,18 +1531,18 @@ static void dm_handle_hpd_rx_offload_work(struct work_struct *work)
 
 	offload_work = container_of(work, struct hpd_rx_irq_offload_work, work);
 	aconnector = offload_work->offload_wq->aconnector;
+	adev = offload_work->adev;
 
 	if (!aconnector) {
-		DRM_ERROR("Can't retrieve aconnector in hpd_rx_irq_offload_work");
+		drm_err(adev_to_drm(adev), "Can't retrieve aconnector in hpd_rx_irq_offload_work");
 		goto skip;
 	}
 
-	adev = drm_to_adev(aconnector->base.dev);
 	dc_link = aconnector->dc_link;
 
 	mutex_lock(&aconnector->hpd_lock);
 	if (!dc_link_detect_connection_type(dc_link, &new_connection_type))
-		DRM_ERROR("KMS: Failed to detect connector\n");
+		drm_err(adev_to_drm(adev), "KMS: Failed to detect connector\n");
 	mutex_unlock(&aconnector->hpd_lock);
 
 	if (new_connection_type == dc_connection_none)
@@ -1563,8 +1611,9 @@ skip:
 
 }
 
-static struct hpd_rx_irq_offload_work_queue *hpd_rx_irq_create_workqueue(struct dc *dc)
+static struct hpd_rx_irq_offload_work_queue *hpd_rx_irq_create_workqueue(struct amdgpu_device *adev)
 {
+	struct dc *dc = adev->dm.dc;
 	int max_caps = dc->caps.max_links;
 	int i = 0;
 	struct hpd_rx_irq_offload_work_queue *hpd_rx_offload_wq = NULL;
@@ -1580,7 +1629,7 @@ static struct hpd_rx_irq_offload_work_queue *hpd_rx_irq_create_workqueue(struct
 				    create_singlethread_workqueue("amdgpu_dm_hpd_rx_offload_wq");
 
 		if (hpd_rx_offload_wq[i].wq == NULL) {
-			DRM_ERROR("create amdgpu_dm_hpd_rx_offload_wq fail!");
+			drm_err(adev_to_drm(adev), "create amdgpu_dm_hpd_rx_offload_wq fail!");
 			goto out_err;
 		}
 
@@ -1629,132 +1678,6 @@ static bool dm_should_disable_stutter(struct pci_dev *pdev)
 	return false;
 }
 
-struct amdgpu_dm_quirks {
-	bool aux_hpd_discon;
-	bool support_edp0_on_dp1;
-};
-
-static struct amdgpu_dm_quirks quirk_entries = {
-	.aux_hpd_discon = false,
-	.support_edp0_on_dp1 = false
-};
-
-static int edp0_on_dp1_callback(const struct dmi_system_id *id)
-{
-	quirk_entries.support_edp0_on_dp1 = true;
-	return 0;
-}
-
-static int aux_hpd_discon_callback(const struct dmi_system_id *id)
-{
-	quirk_entries.aux_hpd_discon = true;
-	return 0;
-}
-
-static const struct dmi_system_id dmi_quirk_table[] = {
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3660"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3260"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3460"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Tower Plus 7010"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Tower 7010"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex SFF Plus 7010"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex SFF 7010"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Micro Plus 7010"),
-		},
-	},
-	{
-		.callback = aux_hpd_discon_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Micro 7010"),
-		},
-	},
-	{
-		.callback = edp0_on_dp1_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "HP Elite mt645 G8 Mobile Thin Client"),
-		},
-	},
-	{
-		.callback = edp0_on_dp1_callback,
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "HP EliteBook 665 16 inch G11 Notebook PC"),
-		},
-	},
-	{}
-	/* TODO: refactor this from a fixed table to a dynamic option */
-};
-
-static void retrieve_dmi_info(struct amdgpu_display_manager *dm, struct dc_init_data *init_data)
-{
-	int dmi_id;
-	struct drm_device *dev = dm->ddev;
-
-	dm->aux_hpd_discon_quirk = false;
-	init_data->flags.support_edp0_on_dp1 = false;
-
-	dmi_id = dmi_check_system(dmi_quirk_table);
-
-	if (!dmi_id)
-		return;
-
-	if (quirk_entries.aux_hpd_discon) {
-		dm->aux_hpd_discon_quirk = true;
-		drm_info(dev, "aux_hpd_discon_quirk attached\n");
-	}
-	if (quirk_entries.support_edp0_on_dp1) {
-		init_data->flags.support_edp0_on_dp1 = true;
-		drm_info(dev, "support_edp0_on_dp1 attached\n");
-	}
-}
 
 void*
 dm_allocate_gpu_mem(
@@ -1899,26 +1822,6 @@ static enum dmub_ips_disable_type dm_get_default_ips_mode(
 	switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) {
 	case IP_VERSION(3, 5, 0):
 	case IP_VERSION(3, 6, 0):
-		/*
-		 * On DCN35 systems with Z8 enabled, it's possible for IPS2 + Z8 to
-		 * cause a hard hang. A fix exists for newer PMFW.
-		 *
-		 * As a workaround, for non-fixed PMFW, force IPS1+RCG as the deepest
-		 * IPS state in all cases, except for s0ix and all displays off (DPMS),
-		 * where IPS2 is allowed.
-		 *
-		 * When checking pmfw version, use the major and minor only.
-		 */
-		if ((adev->pm.fw_version & 0x00FFFF00) < 0x005D6300)
-			ret = DMUB_IPS_RCG_IN_ACTIVE_IPS2_IN_OFF;
-		else if (amdgpu_ip_version(adev, GC_HWIP, 0) > IP_VERSION(11, 5, 0))
-			/*
-			 * Other ASICs with DCN35 that have residency issues with
-			 * IPS2 in idle.
-			 * We want them to use IPS2 only in display off cases.
-			 */
-			ret =  DMUB_IPS_RCG_IN_ACTIVE_IPS2_IN_OFF;
-		break;
 	case IP_VERSION(3, 5, 1):
 		ret =  DMUB_IPS_RCG_IN_ACTIVE_IPS2_IN_OFF;
 		break;
@@ -1950,7 +1853,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	mutex_init(&adev->dm.audio_lock);
 
 	if (amdgpu_dm_irq_init(adev)) {
-		DRM_ERROR("amdgpu: failed to initialize DM IRQ support.\n");
+		drm_err(adev_to_drm(adev), "amdgpu: failed to initialize DM IRQ support.\n");
 		goto error;
 	}
 
@@ -2061,7 +1964,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(3, 0, 0))
 		init_data.num_virtual_links = 1;
 
-	retrieve_dmi_info(&adev->dm, &init_data);
+	retrieve_dmi_info(&adev->dm);
+	if (adev->dm.edp0_on_dp1_quirk)
+		init_data.flags.support_edp0_on_dp1 = true;
 
 	if (adev->dm.bb_from_dmub)
 		init_data.bb_from_dmub = adev->dm.bb_from_dmub;
@@ -2072,10 +1977,10 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	adev->dm.dc = dc_create(&init_data);
 
 	if (adev->dm.dc) {
-		DRM_INFO("Display Core v%s initialized on %s\n", DC_VER,
+		drm_info(adev_to_drm(adev), "Display Core v%s initialized on %s\n", DC_VER,
 			 dce_version_to_string(adev->dm.dc->ctx->dce_version));
 	} else {
-		DRM_INFO("Display Core failed to initialize with v%s!\n", DC_VER);
+		drm_info(adev_to_drm(adev), "Display Core failed to initialize with v%s!\n", DC_VER);
 		goto error;
 	}
 
@@ -2109,25 +2014,31 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		adev->dm.dc->debug.using_dml21 = true;
 	}
 
+	if (amdgpu_dc_debug_mask & DC_HDCP_LC_FORCE_FW_ENABLE)
+		adev->dm.dc->debug.hdcp_lc_force_fw_enable = true;
+
+	if (amdgpu_dc_debug_mask & DC_HDCP_LC_ENABLE_SW_FALLBACK)
+		adev->dm.dc->debug.hdcp_lc_enable_sw_fallback = true;
+
 	adev->dm.dc->debug.visual_confirm = amdgpu_dc_visual_confirm;
 
 	/* TODO: Remove after DP2 receiver gets proper support of Cable ID feature */
 	adev->dm.dc->debug.ignore_cable_id = true;
 
 	if (adev->dm.dc->caps.dp_hdmi21_pcon_support)
-		DRM_INFO("DP-HDMI FRL PCON supported\n");
+		drm_info(adev_to_drm(adev), "DP-HDMI FRL PCON supported\n");
 
 	r = dm_dmub_hw_init(adev);
 	if (r) {
-		DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r);
+		drm_err(adev_to_drm(adev), "DMUB interface failed to initialize: status=%d\n", r);
 		goto error;
 	}
 
 	dc_hardware_init(adev->dm.dc);
 
-	adev->dm.hpd_rx_offload_wq = hpd_rx_irq_create_workqueue(adev->dm.dc);
+	adev->dm.hpd_rx_offload_wq = hpd_rx_irq_create_workqueue(adev);
 	if (!adev->dm.hpd_rx_offload_wq) {
-		DRM_ERROR("amdgpu: failed to create hpd rx offload workqueue.\n");
+		drm_err(adev_to_drm(adev), "amdgpu: failed to create hpd rx offload workqueue.\n");
 		goto error;
 	}
 
@@ -2142,10 +2053,10 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 
 	adev->dm.freesync_module = mod_freesync_create(adev->dm.dc);
 	if (!adev->dm.freesync_module) {
-		DRM_ERROR(
+		drm_err(adev_to_drm(adev),
 		"amdgpu: failed to initialize freesync_module.\n");
 	} else
-		DRM_DEBUG_DRIVER("amdgpu: freesync_module init done %p.\n",
+		drm_dbg_driver(adev_to_drm(adev), "amdgpu: freesync_module init done %p.\n",
 				adev->dm.freesync_module);
 
 	amdgpu_dm_init_color_mod();
@@ -2154,7 +2065,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		adev->dm.vblank_control_workqueue =
 			create_singlethread_workqueue("dm_vblank_control_workqueue");
 		if (!adev->dm.vblank_control_workqueue)
-			DRM_ERROR("amdgpu: failed to initialize vblank_workqueue.\n");
+			drm_err(adev_to_drm(adev), "amdgpu: failed to initialize vblank_workqueue.\n");
 	}
 
 	if (adev->dm.dc->caps.ips_support &&
@@ -2165,9 +2076,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		adev->dm.hdcp_workqueue = hdcp_create_workqueue(adev, &init_params.cp_psp, adev->dm.dc);
 
 		if (!adev->dm.hdcp_workqueue)
-			DRM_ERROR("amdgpu: failed to initialize hdcp_workqueue.\n");
+			drm_err(adev_to_drm(adev), "amdgpu: failed to initialize hdcp_workqueue.\n");
 		else
-			DRM_DEBUG_DRIVER("amdgpu: hdcp_workqueue init done %p.\n", adev->dm.hdcp_workqueue);
+			drm_dbg_driver(adev_to_drm(adev), "amdgpu: hdcp_workqueue init done %p.\n", adev->dm.hdcp_workqueue);
 
 		dc_init_callbacks(adev->dm.dc, &init_params);
 	}
@@ -2175,20 +2086,29 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		init_completion(&adev->dm.dmub_aux_transfer_done);
 		adev->dm.dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_KERNEL);
 		if (!adev->dm.dmub_notify) {
-			DRM_INFO("amdgpu: fail to allocate adev->dm.dmub_notify");
+			drm_info(adev_to_drm(adev), "amdgpu: fail to allocate adev->dm.dmub_notify");
 			goto error;
 		}
 
 		adev->dm.delayed_hpd_wq = create_singlethread_workqueue("amdgpu_dm_hpd_wq");
 		if (!adev->dm.delayed_hpd_wq) {
-			DRM_ERROR("amdgpu: failed to create hpd offload workqueue.\n");
+			drm_err(adev_to_drm(adev), "amdgpu: failed to create hpd offload workqueue.\n");
 			goto error;
 		}
 
 		amdgpu_dm_outbox_init(adev);
 		if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_AUX_REPLY,
 			dmub_aux_setconfig_callback, false)) {
-			DRM_ERROR("amdgpu: fail to register dmub aux callback");
+			drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub aux callback");
+			goto error;
+		}
+
+		for (size_t i = 0; i < ARRAY_SIZE(adev->dm.fused_io); i++)
+			init_completion(&adev->dm.fused_io[i].replied);
+
+		if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_FUSED_IO,
+			dmub_aux_fused_io_callback, false)) {
+			drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub fused io callback");
 			goto error;
 		}
 		/* Enable outbox notification only after IRQ handlers are registered and DMUB is alive.
@@ -2205,7 +2125,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_dm_initialize_drm_device(adev)) {
-		DRM_ERROR(
+		drm_err(adev_to_drm(adev),
 		"amdgpu: failed to initialize sw for display support.\n");
 		goto error;
 	}
@@ -2220,7 +2140,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	adev_to_drm(adev)->mode_config.cursor_height = adev->dm.dc->caps.max_cursor_size;
 
 	if (drm_vblank_init(adev_to_drm(adev), adev->dm.display_indexes_num)) {
-		DRM_ERROR(
+		drm_err(adev_to_drm(adev),
 		"amdgpu: failed to initialize sw for display support.\n");
 		goto error;
 	}
@@ -2228,14 +2148,14 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
 	amdgpu_dm_crtc_secure_display_create_contexts(adev);
 	if (!adev->dm.secure_display_ctx.crtc_ctx)
-		DRM_ERROR("amdgpu: failed to initialize secure display contexts.\n");
+		drm_err(adev_to_drm(adev), "amdgpu: failed to initialize secure display contexts.\n");
 
 	if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(4, 0, 1))
 		adev->dm.secure_display_ctx.support_mul_roi = true;
 
 #endif
 
-	DRM_DEBUG_DRIVER("KMS initialized.\n");
+	drm_dbg_driver(adev_to_drm(adev), "KMS initialized.\n");
 
 	return 0;
 error:
@@ -2408,7 +2328,7 @@ static int load_dmcu_fw(struct amdgpu_device *adev)
 		default:
 			break;
 		}
-		DRM_ERROR("Unsupported ASIC type: 0x%X\n", adev->asic_type);
+		drm_err(adev_to_drm(adev), "Unsupported ASIC type: 0x%X\n", adev->asic_type);
 		return -EINVAL;
 	}
 
@@ -2426,7 +2346,7 @@ static int load_dmcu_fw(struct amdgpu_device *adev)
 		return 0;
 	}
 	if (r) {
-		dev_err(adev->dev, "amdgpu_dm: Can't validate firmware \"%s\"\n",
+		drm_err(adev_to_drm(adev), "amdgpu_dm: Can't validate firmware \"%s\"\n",
 			fw_name_dmcu);
 		amdgpu_ucode_release(&adev->dm.fw_dmcu);
 		return r;
@@ -2551,7 +2471,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev)
 		adev->firmware.fw_size +=
 			ALIGN(le32_to_cpu(hdr->inst_const_bytes), PAGE_SIZE);
 
-		DRM_INFO("Loading DMUB firmware via PSP: version=0x%08X\n",
+		drm_info(adev_to_drm(adev), "Loading DMUB firmware via PSP: version=0x%08X\n",
 			 adev->dm.dmcub_fw_version);
 	}
 
@@ -2560,7 +2480,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev)
 	dmub_srv = adev->dm.dmub_srv;
 
 	if (!dmub_srv) {
-		DRM_ERROR("Failed to allocate DMUB service!\n");
+		drm_err(adev_to_drm(adev), "Failed to allocate DMUB service!\n");
 		return -ENOMEM;
 	}
 
@@ -2573,7 +2493,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev)
 	/* Create the DMUB service. */
 	status = dmub_srv_create(dmub_srv, &create_params);
 	if (status != DMUB_STATUS_OK) {
-		DRM_ERROR("Error creating DMUB service: %d\n", status);
+		drm_err(adev_to_drm(adev), "Error creating DMUB service: %d\n", status);
 		return -EINVAL;
 	}
 
@@ -2598,7 +2518,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev)
 					   &region_info);
 
 	if (status != DMUB_STATUS_OK) {
-		DRM_ERROR("Error calculating DMUB region info: %d\n", status);
+		drm_err(adev_to_drm(adev), "Error calculating DMUB region info: %d\n", status);
 		return -EINVAL;
 	}
 
@@ -2627,14 +2547,14 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev)
 	fb_info = adev->dm.dmub_fb_info;
 
 	if (!fb_info) {
-		DRM_ERROR(
+		drm_err(adev_to_drm(adev),
 			"Failed to allocate framebuffer info for DMUB service!\n");
 		return -ENOMEM;
 	}
 
 	status = dmub_srv_calc_mem_info(dmub_srv, &memory_params, fb_info);
 	if (status != DMUB_STATUS_OK) {
-		DRM_ERROR("Error calculating DMUB FB info: %d\n", status);
+		drm_err(adev_to_drm(adev), "Error calculating DMUB FB info: %d\n", status);
 		return -EINVAL;
 	}
 
@@ -2651,7 +2571,7 @@ static int dm_sw_init(struct amdgpu_ip_block *ip_block)
 	adev->dm.cgs_device = amdgpu_cgs_create_device(adev);
 
 	if (!adev->dm.cgs_device) {
-		DRM_ERROR("amdgpu: failed to create cgs device.\n");
+		drm_err(adev_to_drm(adev), "amdgpu: failed to create cgs device.\n");
 		return -EINVAL;
 	}
 
@@ -2957,7 +2877,7 @@ static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev)
 
 	ret = amdgpu_dpm_write_watermarks_table(adev);
 	if (ret) {
-		DRM_ERROR("Failed to update WMTABLE!\n");
+		drm_err(adev_to_drm(adev), "Failed to update WMTABLE!\n");
 		return ret;
 	}
 
@@ -2975,13 +2895,13 @@ static int dm_oem_i2c_hw_init(struct amdgpu_device *adev)
 	if (oem_ddc_service) {
 		oem_i2c = create_i2c(oem_ddc_service, true);
 		if (!oem_i2c) {
-			dev_info(adev->dev, "Failed to create oem i2c adapter data\n");
+			drm_info(adev_to_drm(adev), "Failed to create oem i2c adapter data\n");
 			return -ENOMEM;
 		}
 
 		r = i2c_add_adapter(&oem_i2c->base);
 		if (r) {
-			dev_info(adev->dev, "Failed to register oem i2c\n");
+			drm_info(adev_to_drm(adev), "Failed to register oem i2c\n");
 			kfree(oem_i2c);
 			return r;
 		}
@@ -3024,7 +2944,7 @@ static int dm_hw_init(struct amdgpu_ip_block *ip_block)
 
 	r = dm_oem_i2c_hw_init(adev);
 	if (r)
-		dev_info(adev->dev, "Failed to add OEM i2c bus\n");
+		drm_info(adev_to_drm(adev), "Failed to add OEM i2c bus\n");
 
 	return 0;
 }
@@ -3067,7 +2987,7 @@ static void dm_gpureset_toggle_interrupts(struct amdgpu_device *adev,
 			irq_source = IRQ_TYPE_PFLIP + acrtc->otg_inst;
 			rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY;
 			if (rc)
-				DRM_WARN("Failed to %s pflip interrupts\n",
+				drm_warn(adev_to_drm(adev), "Failed to %s pflip interrupts\n",
 					 enable ? "enable" : "disable");
 
 			if (enable) {
@@ -3077,14 +2997,14 @@ static void dm_gpureset_toggle_interrupts(struct amdgpu_device *adev,
 				rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, false);
 
 			if (rc)
-				DRM_WARN("Failed to %sable vupdate interrupt\n", enable ? "en" : "dis");
+				drm_warn(adev_to_drm(adev), "Failed to %sable vupdate interrupt\n", enable ? "en" : "dis");
 
 			irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
 			/* During gpu-reset we disable and then enable vblank irq, so
 			 * don't use amdgpu_irq_get/put() to avoid refcount change.
 			 */
 			if (!dc_interrupt_set(adev->dm.dc, irq_source, enable))
-				DRM_WARN("Failed to %sable vblank interrupt\n", enable ? "en" : "dis");
+				drm_warn(adev_to_drm(adev), "Failed to %sable vblank interrupt\n", enable ? "en" : "dis");
 		}
 	}
 
@@ -3334,16 +3254,16 @@ static void dm_gpureset_commit_state(struct dc_state *dc_state,
 	for (k = 0; k < dc_state->stream_count; k++) {
 		bundle->stream_update.stream = dc_state->streams[k];
 
-		for (m = 0; m < dc_state->stream_status->plane_count; m++) {
+		for (m = 0; m < dc_state->stream_status[k].plane_count; m++) {
 			bundle->surface_updates[m].surface =
-				dc_state->stream_status->plane_states[m];
+				dc_state->stream_status[k].plane_states[m];
 			bundle->surface_updates[m].surface->force_full_update =
 				true;
 		}
 
 		update_planes_and_stream_adapter(dm->dc,
 					 UPDATE_TYPE_FULL,
-					 dc_state->stream_status->plane_count,
+					 dc_state->stream_status[k].plane_count,
 					 dc_state->streams[k],
 					 &bundle->stream_update,
 					 bundle->surface_updates);
@@ -3919,20 +3839,21 @@ static void handle_hpd_irq(void *param)
 
 }
 
-static void schedule_hpd_rx_offload_work(struct hpd_rx_irq_offload_work_queue *offload_wq,
+static void schedule_hpd_rx_offload_work(struct amdgpu_device *adev, struct hpd_rx_irq_offload_work_queue *offload_wq,
 							union hpd_irq_data hpd_irq_data)
 {
 	struct hpd_rx_irq_offload_work *offload_work =
 				kzalloc(sizeof(*offload_work), GFP_KERNEL);
 
 	if (!offload_work) {
-		DRM_ERROR("Failed to allocate hpd_rx_irq_offload_work.\n");
+		drm_err(adev_to_drm(adev), "Failed to allocate hpd_rx_irq_offload_work.\n");
 		return;
 	}
 
 	INIT_WORK(&offload_work->work, dm_handle_hpd_rx_offload_work);
 	offload_work->data = hpd_irq_data;
 	offload_work->offload_wq = offload_wq;
+	offload_work->adev = adev;
 
 	queue_work(offload_wq->wq, &offload_work->work);
 	DRM_DEBUG_KMS("queue work to handle hpd_rx offload work");
@@ -3974,7 +3895,7 @@ static void handle_hpd_rx_irq(void *param)
 		goto out;
 
 	if (hpd_irq_data.bytes.device_service_irq.bits.AUTOMATED_TEST) {
-		schedule_hpd_rx_offload_work(offload_wq, hpd_irq_data);
+		schedule_hpd_rx_offload_work(adev, offload_wq, hpd_irq_data);
 		goto out;
 	}
 
@@ -3996,7 +3917,7 @@ static void handle_hpd_rx_irq(void *param)
 			spin_unlock(&offload_wq->offload_lock);
 
 			if (!skip)
-				schedule_hpd_rx_offload_work(offload_wq, hpd_irq_data);
+				schedule_hpd_rx_offload_work(adev, offload_wq, hpd_irq_data);
 
 			goto out;
 		}
@@ -4013,7 +3934,7 @@ static void handle_hpd_rx_irq(void *param)
 			spin_unlock(&offload_wq->offload_lock);
 
 			if (!skip)
-				schedule_hpd_rx_offload_work(offload_wq, hpd_irq_data);
+				schedule_hpd_rx_offload_work(adev, offload_wq, hpd_irq_data);
 
 			goto out;
 		}
@@ -4023,7 +3944,7 @@ out:
 	if (result && !is_mst_root_connector) {
 		/* Downstream Port status changed. */
 		if (!dc_link_detect_connection_type(dc_link, &new_connection_type))
-			DRM_ERROR("KMS: Failed to detect connector\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to detect connector\n");
 
 		if (aconnector->base.force && new_connection_type == dc_connection_none) {
 			emulated_link_detect(dc_link);
@@ -4086,19 +4007,19 @@ static int register_hpd_handlers(struct amdgpu_device *adev)
 	if (dc_is_dmub_outbox_supported(adev->dm.dc)) {
 		if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD,
 			dmub_hpd_callback, true)) {
-			DRM_ERROR("amdgpu: fail to register dmub hpd callback");
+			drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd callback");
 			return -EINVAL;
 		}
 
 		if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD_IRQ,
 			dmub_hpd_callback, true)) {
-			DRM_ERROR("amdgpu: fail to register dmub hpd callback");
+			drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd callback");
 			return -EINVAL;
 		}
 
 		if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD_SENSE_NOTIFY,
 			dmub_hpd_sense_callback, true)) {
-			DRM_ERROR("amdgpu: fail to register dmub hpd sense callback");
+			drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd sense callback");
 			return -EINVAL;
 		}
 	}
@@ -4119,7 +4040,7 @@ static int register_hpd_handlers(struct amdgpu_device *adev)
 			if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 				int_params.irq_source  < DC_IRQ_SOURCE_HPD1 ||
 				int_params.irq_source  > DC_IRQ_SOURCE_HPD6) {
-				DRM_ERROR("Failed to register hpd irq!\n");
+				drm_err(adev_to_drm(adev), "Failed to register hpd irq!\n");
 				return -EINVAL;
 			}
 
@@ -4137,7 +4058,7 @@ static int register_hpd_handlers(struct amdgpu_device *adev)
 			if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 				int_params.irq_source  < DC_IRQ_SOURCE_HPD1RX ||
 				int_params.irq_source  > DC_IRQ_SOURCE_HPD6RX) {
-				DRM_ERROR("Failed to register hpd rx irq!\n");
+				drm_err(adev_to_drm(adev), "Failed to register hpd rx irq!\n");
 				return -EINVAL;
 			}
 
@@ -4179,7 +4100,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev)
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
 		r = amdgpu_irq_add_id(adev, client_id, i + 1, &adev->crtc_irq);
 		if (r) {
-			DRM_ERROR("Failed to add crtc irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add crtc irq id!\n");
 			return r;
 		}
 
@@ -4190,7 +4111,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_VBLANK1 ||
 			int_params.irq_source  > DC_IRQ_SOURCE_VBLANK6) {
-			DRM_ERROR("Failed to register vblank irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register vblank irq!\n");
 			return -EINVAL;
 		}
 
@@ -4209,7 +4130,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev)
 			i <= VISLANDS30_IV_SRCID_D6_GRPH_PFLIP; i += 2) {
 		r = amdgpu_irq_add_id(adev, client_id, i, &adev->pageflip_irq);
 		if (r) {
-			DRM_ERROR("Failed to add page flip irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add page flip irq id!\n");
 			return r;
 		}
 
@@ -4220,7 +4141,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_PFLIP_FIRST ||
 			int_params.irq_source  > DC_IRQ_SOURCE_PFLIP_LAST) {
-			DRM_ERROR("Failed to register pflip irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register pflip irq!\n");
 			return -EINVAL;
 		}
 
@@ -4238,7 +4159,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev)
 	r = amdgpu_irq_add_id(adev, client_id,
 			VISLANDS30_IV_SRCID_HOTPLUG_DETECT_A, &adev->hpd_irq);
 	if (r) {
-		DRM_ERROR("Failed to add hpd irq id!\n");
+		drm_err(adev_to_drm(adev), "Failed to add hpd irq id!\n");
 		return r;
 	}
 
@@ -4280,7 +4201,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 	for (i = VISLANDS30_IV_SRCID_D1_VERTICAL_INTERRUPT0; i <= VISLANDS30_IV_SRCID_D6_VERTICAL_INTERRUPT0; i++) {
 		r = amdgpu_irq_add_id(adev, client_id, i, &adev->crtc_irq);
 		if (r) {
-			DRM_ERROR("Failed to add crtc irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add crtc irq id!\n");
 			return r;
 		}
 
@@ -4291,7 +4212,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_VBLANK1 ||
 			int_params.irq_source  > DC_IRQ_SOURCE_VBLANK6) {
-			DRM_ERROR("Failed to register vblank irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register vblank irq!\n");
 			return -EINVAL;
 		}
 
@@ -4309,7 +4230,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 	for (i = VISLANDS30_IV_SRCID_D1_V_UPDATE_INT; i <= VISLANDS30_IV_SRCID_D6_V_UPDATE_INT; i += 2) {
 		r = amdgpu_irq_add_id(adev, client_id, i, &adev->vupdate_irq);
 		if (r) {
-			DRM_ERROR("Failed to add vupdate irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add vupdate irq id!\n");
 			return r;
 		}
 
@@ -4320,7 +4241,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_VUPDATE1 ||
 			int_params.irq_source  > DC_IRQ_SOURCE_VUPDATE6) {
-			DRM_ERROR("Failed to register vupdate irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register vupdate irq!\n");
 			return -EINVAL;
 		}
 
@@ -4339,7 +4260,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 			i <= VISLANDS30_IV_SRCID_D6_GRPH_PFLIP; i += 2) {
 		r = amdgpu_irq_add_id(adev, client_id, i, &adev->pageflip_irq);
 		if (r) {
-			DRM_ERROR("Failed to add page flip irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add page flip irq id!\n");
 			return r;
 		}
 
@@ -4350,7 +4271,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_PFLIP_FIRST ||
 			int_params.irq_source  > DC_IRQ_SOURCE_PFLIP_LAST) {
-			DRM_ERROR("Failed to register pflip irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register pflip irq!\n");
 			return -EINVAL;
 		}
 
@@ -4368,7 +4289,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev)
 	r = amdgpu_irq_add_id(adev, client_id,
 			VISLANDS30_IV_SRCID_HOTPLUG_DETECT_A, &adev->hpd_irq);
 	if (r) {
-		DRM_ERROR("Failed to add hpd irq id!\n");
+		drm_err(adev_to_drm(adev), "Failed to add hpd irq id!\n");
 		return r;
 	}
 
@@ -4418,7 +4339,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->crtc_irq);
 
 		if (r) {
-			DRM_ERROR("Failed to add crtc irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add crtc irq id!\n");
 			return r;
 		}
 
@@ -4429,7 +4350,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_VBLANK1 ||
 			int_params.irq_source  > DC_IRQ_SOURCE_VBLANK6) {
-			DRM_ERROR("Failed to register vblank irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register vblank irq!\n");
 			return -EINVAL;
 		}
 
@@ -4450,7 +4371,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 				vrtl_int_srcid[i], &adev->vline0_irq);
 
 		if (r) {
-			DRM_ERROR("Failed to add vline0 irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add vline0 irq id!\n");
 			return r;
 		}
 
@@ -4461,7 +4382,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source < DC_IRQ_SOURCE_DC1_VLINE0 ||
 			int_params.irq_source > DC_IRQ_SOURCE_DC6_VLINE0) {
-			DRM_ERROR("Failed to register vline0 irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register vline0 irq!\n");
 			return -EINVAL;
 		}
 
@@ -4489,7 +4410,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->vupdate_irq);
 
 		if (r) {
-			DRM_ERROR("Failed to add vupdate irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add vupdate irq id!\n");
 			return r;
 		}
 
@@ -4500,7 +4421,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_VUPDATE1 ||
 			int_params.irq_source  > DC_IRQ_SOURCE_VUPDATE6) {
-			DRM_ERROR("Failed to register vupdate irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register vupdate irq!\n");
 			return -EINVAL;
 		}
 
@@ -4520,7 +4441,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 			i++) {
 		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->pageflip_irq);
 		if (r) {
-			DRM_ERROR("Failed to add page flip irq id!\n");
+			drm_err(adev_to_drm(adev), "Failed to add page flip irq id!\n");
 			return r;
 		}
 
@@ -4531,7 +4452,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 		if (int_params.irq_source == DC_IRQ_SOURCE_INVALID ||
 			int_params.irq_source  < DC_IRQ_SOURCE_PFLIP_FIRST ||
 			int_params.irq_source  > DC_IRQ_SOURCE_PFLIP_LAST) {
-			DRM_ERROR("Failed to register pflip irq!\n");
+			drm_err(adev_to_drm(adev), "Failed to register pflip irq!\n");
 			return -EINVAL;
 		}
 
@@ -4549,7 +4470,7 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev)
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, DCN_1_0__SRCID__DC_HPD1_INT,
 			&adev->hpd_irq);
 	if (r) {
-		DRM_ERROR("Failed to add hpd irq id!\n");
+		drm_err(adev_to_drm(adev), "Failed to add hpd irq id!\n");
 		return r;
 	}
 
@@ -4571,7 +4492,7 @@ static int register_outbox_irq_handlers(struct amdgpu_device *adev)
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, DCN_1_0__SRCID__DMCUB_OUTBOX_LOW_PRIORITY_READY_INT,
 			&adev->dmub_outbox_irq);
 	if (r) {
-		DRM_ERROR("Failed to add outbox irq id!\n");
+		drm_err(adev_to_drm(adev), "Failed to add outbox irq id!\n");
 		return r;
 	}
 
@@ -4803,41 +4724,54 @@ static int get_brightness_range(const struct amdgpu_dm_backlight_caps *caps,
 	return 1;
 }
 
-static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *caps,
-					uint32_t brightness)
+static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *caps,
+				      uint32_t *brightness)
 {
-	unsigned int min, max;
 	u8 prev_signal = 0, prev_lum = 0;
+	int i = 0;
 
-	if (!get_brightness_range(caps, &min, &max))
-		return brightness;
-
-	for (int i = 0; i < caps->data_points; i++) {
-		u8 signal, lum;
+	if (amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)
+		return;
 
-		if (amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)
-			break;
+	if (!caps->data_points)
+		return;
 
-		signal = caps->luminance_data[i].input_signal;
-		lum = caps->luminance_data[i].luminance;
+	/* choose start to run less interpolation steps */
+	if (caps->luminance_data[caps->data_points/2].input_signal > *brightness)
+		i = caps->data_points/2;
+	do {
+		u8 signal = caps->luminance_data[i].input_signal;
+		u8 lum = caps->luminance_data[i].luminance;
 
 		/*
 		 * brightness == signal: luminance is percent numerator
 		 * brightness < signal: interpolate between previous and current luminance numerator
 		 * brightness > signal: find next data point
 		 */
-		if (brightness < signal)
-			lum = prev_lum + DIV_ROUND_CLOSEST((lum - prev_lum) *
-							   (brightness - prev_signal),
-							   signal - prev_signal);
-		else if (brightness > signal) {
+		if (*brightness > signal) {
 			prev_signal = signal;
 			prev_lum = lum;
+			i++;
 			continue;
 		}
-		brightness = DIV_ROUND_CLOSEST(lum * brightness, 101);
-		break;
-	}
+		if (*brightness < signal)
+			lum = prev_lum + DIV_ROUND_CLOSEST((lum - prev_lum) *
+							   (*brightness - prev_signal),
+							   signal - prev_signal);
+		*brightness = DIV_ROUND_CLOSEST(lum * *brightness, 101);
+		return;
+	} while (i < caps->data_points);
+}
+
+static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *caps,
+					uint32_t brightness)
+{
+	unsigned int min, max;
+
+	if (!get_brightness_range(caps, &min, &max))
+		return brightness;
+
+	convert_custom_brightness(caps, &brightness);
 
 	// Rescale 0..255 to min..max
 	return min + DIV_ROUND_CLOSEST((max - min) * brightness,
@@ -5016,10 +4950,10 @@ amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector)
 	dm->brightness[aconnector->bl_idx] = props.brightness;
 
 	if (IS_ERR(dm->backlight_dev[aconnector->bl_idx])) {
-		DRM_ERROR("DM: Backlight registration failed!\n");
+		drm_err(drm, "DM: Backlight registration failed!\n");
 		dm->backlight_dev[aconnector->bl_idx] = NULL;
 	} else
-		DRM_DEBUG_DRIVER("DM: Registered Backlight device: %s\n", bl_name);
+		drm_dbg_driver(drm, "DM: Registered Backlight device: %s\n", bl_name);
 }
 
 static int initialize_plane(struct amdgpu_display_manager *dm,
@@ -5033,7 +4967,7 @@ static int initialize_plane(struct amdgpu_display_manager *dm,
 
 	plane = kzalloc(sizeof(struct drm_plane), GFP_KERNEL);
 	if (!plane) {
-		DRM_ERROR("KMS: Failed to allocate plane\n");
+		drm_err(adev_to_drm(dm->adev), "KMS: Failed to allocate plane\n");
 		return -ENOMEM;
 	}
 	plane->type = plane_type;
@@ -5051,7 +4985,7 @@ static int initialize_plane(struct amdgpu_display_manager *dm,
 	ret = amdgpu_dm_plane_init(dm, plane, possible_crtcs, plane_cap);
 
 	if (ret) {
-		DRM_ERROR("KMS: Failed to initialize plane\n");
+		drm_err(adev_to_drm(dm->adev), "KMS: Failed to initialize plane\n");
 		kfree(plane);
 		return ret;
 	}
@@ -5120,14 +5054,14 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 
 	link_cnt = dm->dc->caps.max_links;
 	if (amdgpu_dm_mode_config_init(dm->adev)) {
-		DRM_ERROR("DM: Failed to initialize mode config\n");
+		drm_err(adev_to_drm(adev), "DM: Failed to initialize mode config\n");
 		return -EINVAL;
 	}
 
 	/* There is one primary plane per CRTC */
 	primary_planes = dm->dc->caps.max_streams;
 	if (primary_planes > AMDGPU_MAX_PLANES) {
-		DRM_ERROR("DM: Plane nums out of 6 planes\n");
+		drm_err(adev_to_drm(adev), "DM: Plane nums out of 6 planes\n");
 		return -EINVAL;
 	}
 
@@ -5140,7 +5074,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 
 		if (initialize_plane(dm, mode_info, i,
 				     DRM_PLANE_TYPE_PRIMARY, plane)) {
-			DRM_ERROR("KMS: Failed to initialize primary plane\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to initialize primary plane\n");
 			goto fail;
 		}
 	}
@@ -5172,14 +5106,14 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 
 		if (initialize_plane(dm, NULL, primary_planes + i,
 				     DRM_PLANE_TYPE_OVERLAY, plane)) {
-			DRM_ERROR("KMS: Failed to initialize overlay plane\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to initialize overlay plane\n");
 			goto fail;
 		}
 	}
 
 	for (i = 0; i < dm->dc->caps.max_streams; i++)
 		if (amdgpu_dm_crtc_init(dm, mode_info->planes[i], i)) {
-			DRM_ERROR("KMS: Failed to initialize crtc\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to initialize crtc\n");
 			goto fail;
 		}
 
@@ -5199,7 +5133,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	case IP_VERSION(3, 6, 0):
 	case IP_VERSION(4, 0, 1):
 		if (register_outbox_irq_handlers(dm->adev)) {
-			DRM_ERROR("DM: Failed to initialize IRQ\n");
+			drm_err(adev_to_drm(adev), "DM: Failed to initialize IRQ\n");
 			goto fail;
 		}
 		break;
@@ -5249,7 +5183,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	}
 
 	if (link_cnt > MAX_LINKS) {
-		DRM_ERROR(
+		drm_err(adev_to_drm(adev),
 			"KMS: Cannot support more than %d display indexes\n",
 				MAX_LINKS);
 		goto fail;
@@ -5265,12 +5199,12 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 			struct amdgpu_dm_wb_connector *wbcon = kzalloc(sizeof(*wbcon), GFP_KERNEL);
 
 			if (!wbcon) {
-				DRM_ERROR("KMS: Failed to allocate writeback connector\n");
+				drm_err(adev_to_drm(adev), "KMS: Failed to allocate writeback connector\n");
 				continue;
 			}
 
 			if (amdgpu_dm_wb_connector_init(dm, wbcon, i)) {
-				DRM_ERROR("KMS: Failed to initialize writeback connector\n");
+				drm_err(adev_to_drm(adev), "KMS: Failed to initialize writeback connector\n");
 				kfree(wbcon);
 				continue;
 			}
@@ -5290,12 +5224,12 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 			goto fail;
 
 		if (amdgpu_dm_encoder_init(dm->ddev, aencoder, i)) {
-			DRM_ERROR("KMS: Failed to initialize encoder\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to initialize encoder\n");
 			goto fail;
 		}
 
 		if (amdgpu_dm_connector_init(dm, aconnector, i, aencoder)) {
-			DRM_ERROR("KMS: Failed to initialize connector\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to initialize connector\n");
 			goto fail;
 		}
 
@@ -5304,7 +5238,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 				aconnector;
 
 		if (!dc_link_detect_connection_type(link, &new_connection_type))
-			DRM_ERROR("KMS: Failed to detect connector\n");
+			drm_err(adev_to_drm(adev), "KMS: Failed to detect connector\n");
 
 		if (aconnector->base.force && new_connection_type == dc_connection_none) {
 			emulated_link_detect(link);
@@ -5326,8 +5260,15 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 					if (amdgpu_dm_set_replay_caps(link, aconnector))
 						psr_feature_enabled = false;
 
-				if (psr_feature_enabled)
+				if (psr_feature_enabled) {
 					amdgpu_dm_set_psr_caps(link);
+					drm_info(adev_to_drm(adev), "PSR support %d, DC PSR ver %d, sink PSR ver %d DPCD caps 0x%x su_y_granularity %d\n",
+						 link->psr_settings.psr_feature_enabled,
+						 link->psr_settings.psr_version,
+						 link->dpcd_caps.psr_info.psr_version,
+						 link->dpcd_caps.psr_info.psr_dpcd_caps.raw,
+						 link->dpcd_caps.psr_info.psr2_su_y_granularity_cap);
+				}
 			}
 		}
 		amdgpu_set_panel_orientation(&aconnector->base);
@@ -5341,7 +5282,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	case CHIP_VERDE:
 	case CHIP_OLAND:
 		if (dce60_register_irq_handlers(dm->adev)) {
-			DRM_ERROR("DM: Failed to initialize IRQ\n");
+			drm_err(adev_to_drm(adev), "DM: Failed to initialize IRQ\n");
 			goto fail;
 		}
 		break;
@@ -5363,7 +5304,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 	case CHIP_VEGA12:
 	case CHIP_VEGA20:
 		if (dce110_register_irq_handlers(dm->adev)) {
-			DRM_ERROR("DM: Failed to initialize IRQ\n");
+			drm_err(adev_to_drm(adev), "DM: Failed to initialize IRQ\n");
 			goto fail;
 		}
 		break;
@@ -5391,12 +5332,12 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
 		case IP_VERSION(3, 6, 0):
 		case IP_VERSION(4, 0, 1):
 			if (dcn10_register_irq_handlers(dm->adev)) {
-				DRM_ERROR("DM: Failed to initialize IRQ\n");
+				drm_err(adev_to_drm(adev), "DM: Failed to initialize IRQ\n");
 				goto fail;
 			}
 			break;
 		default:
-			DRM_ERROR("Unsupported DCE IP versions: 0x%X\n",
+			drm_err(adev_to_drm(adev), "Unsupported DCE IP versions: 0x%X\n",
 					amdgpu_ip_version(adev, DCE_HWIP, 0));
 			goto fail;
 		}
@@ -5557,7 +5498,7 @@ static int dm_early_init(struct amdgpu_ip_block *ip_block)
 	/* if there is no object header, skip DM */
 	if (!amdgpu_atom_parse_data_header(ctx, index, NULL, NULL, NULL, &data_offset)) {
 		adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK;
-		dev_info(adev->dev, "No object header, skipping DM\n");
+		drm_info(adev_to_drm(adev), "No object header, skipping DM\n");
 		return -ENOENT;
 	}
 
@@ -5669,7 +5610,7 @@ static int dm_early_init(struct amdgpu_ip_block *ip_block)
 			adev->mode_info.num_dig = 4;
 			break;
 		default:
-			DRM_ERROR("Unsupported DCE IP versions: 0x%x\n",
+			drm_err(adev_to_drm(adev), "Unsupported DCE IP versions: 0x%x\n",
 					amdgpu_ip_version(adev, DCE_HWIP, 0));
 			return -EINVAL;
 		}
@@ -5818,7 +5759,7 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev,
 		plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616;
 		break;
 	default:
-		DRM_ERROR(
+		drm_err(adev_to_drm(adev),
 			"Unsupported screen format %p4cc\n",
 			&fb->format->format);
 		return -EINVAL;
@@ -6339,6 +6280,7 @@ static void fill_stream_properties_from_drm_display_mode(
 	struct amdgpu_dm_connector *aconnector = NULL;
 	struct hdmi_vendor_infoframe hv_frame;
 	struct hdmi_avi_infoframe avi_frame;
+	ssize_t err;
 
 	if (connector->connector_type != DRM_MODE_CONNECTOR_WRITEBACK)
 		aconnector = to_amdgpu_dm_connector(connector);
@@ -6385,9 +6327,17 @@ static void fill_stream_properties_from_drm_display_mode(
 	}
 
 	if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) {
-		drm_hdmi_avi_infoframe_from_display_mode(&avi_frame, (struct drm_connector *)connector, mode_in);
+		err = drm_hdmi_avi_infoframe_from_display_mode(&avi_frame,
+							       (struct drm_connector *)connector,
+							       mode_in);
+		if (err < 0)
+			drm_warn_once(connector->dev, "Failed to setup avi infoframe on connector %s: %zd \n", connector->name, err);
 		timing_out->vic = avi_frame.video_code;
-		drm_hdmi_vendor_infoframe_from_display_mode(&hv_frame, (struct drm_connector *)connector, mode_in);
+		err = drm_hdmi_vendor_infoframe_from_display_mode(&hv_frame,
+								  (struct drm_connector *)connector,
+								  mode_in);
+		if (err < 0)
+			drm_warn_once(connector->dev, "Failed to setup vendor infoframe on connector %s: %zd \n", connector->name, err);
 		timing_out->hdmi_vic = hv_frame.vic;
 	}
 
@@ -6500,19 +6450,19 @@ decide_crtc_timing_for_drm_display_mode(struct drm_display_mode *drm_mode,
 					const struct drm_display_mode *native_mode,
 					bool scale_enabled)
 {
-	if (scale_enabled) {
-		copy_crtc_timing_for_drm_display_mode(native_mode, drm_mode);
-	} else if (native_mode->clock == drm_mode->clock &&
-			native_mode->htotal == drm_mode->htotal &&
-			native_mode->vtotal == drm_mode->vtotal) {
-		copy_crtc_timing_for_drm_display_mode(native_mode, drm_mode);
+	if (scale_enabled || (
+	    native_mode->clock == drm_mode->clock &&
+	    native_mode->htotal == drm_mode->htotal &&
+	    native_mode->vtotal == drm_mode->vtotal)) {
+		if (native_mode->crtc_clock)
+			copy_crtc_timing_for_drm_display_mode(native_mode, drm_mode);
 	} else {
 		/* no scaling nor amdgpu inserted, no need to patch */
 	}
 }
 
 static struct dc_sink *
-create_fake_sink(struct dc_link *link)
+create_fake_sink(struct drm_device *dev, struct dc_link *link)
 {
 	struct dc_sink_init_data sink_init_data = { 0 };
 	struct dc_sink *sink = NULL;
@@ -6522,7 +6472,7 @@ create_fake_sink(struct dc_link *link)
 
 	sink = dc_sink_create(&sink_init_data);
 	if (!sink) {
-		DRM_ERROR("Failed to create sink!\n");
+		drm_err(dev, "Failed to create sink!\n");
 		return NULL;
 	}
 	sink->sink_signal = SIGNAL_TYPE_VIRTUAL;
@@ -6655,7 +6605,7 @@ get_highest_refresh_rate_mode(struct amdgpu_dm_connector *aconnector,
 		m_pref = list_first_entry_or_null(
 				&aconnector->base.modes, struct drm_display_mode, head);
 		if (!m_pref) {
-			DRM_DEBUG_DRIVER("No preferred mode found in EDID\n");
+			drm_dbg_driver(aconnector->base.dev, "No preferred mode found in EDID\n");
 			return NULL;
 		}
 	}
@@ -6830,7 +6780,7 @@ static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector,
 						dc_link_get_highest_encoding_format(aconnector->dc_link),
 						&stream->timing.dsc_cfg)) {
 				stream->timing.flags.DSC = 1;
-				DRM_DEBUG_DRIVER("%s: SST_DSC [%s] DSC is selected from SST RX\n",
+				drm_dbg_driver(drm_connector->dev, "%s: SST_DSC [%s] DSC is selected from SST RX\n",
 							__func__, drm_connector->name);
 			}
 		} else if (sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER) {
@@ -6850,7 +6800,7 @@ static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector,
 						dc_link_get_highest_encoding_format(aconnector->dc_link),
 						&stream->timing.dsc_cfg)) {
 					stream->timing.flags.DSC = 1;
-					DRM_DEBUG_DRIVER("%s: SST_DSC [%s] DSC is selected from DP-HDMI PCON\n",
+					drm_dbg_driver(drm_connector->dev, "%s: SST_DSC [%s] DSC is selected from DP-HDMI PCON\n",
 									 __func__, drm_connector->name);
 				}
 		}
@@ -6878,6 +6828,7 @@ create_stream_for_sink(struct drm_connector *connector,
 		       const struct dc_stream_state *old_stream,
 		       int requested_bpc)
 {
+	struct drm_device *dev = connector->dev;
 	struct amdgpu_dm_connector *aconnector = NULL;
 	struct drm_display_mode *preferred_mode = NULL;
 	const struct drm_connector_state *con_state = &dm_state->base;
@@ -6900,11 +6851,6 @@ create_stream_for_sink(struct drm_connector *connector,
 	drm_mode_init(&mode, drm_mode);
 	memset(&saved_mode, 0, sizeof(saved_mode));
 
-	if (connector == NULL) {
-		DRM_ERROR("connector is NULL!\n");
-		return stream;
-	}
-
 	if (connector->connector_type != DRM_MODE_CONNECTOR_WRITEBACK) {
 		aconnector = NULL;
 		aconnector = to_amdgpu_dm_connector(connector);
@@ -6919,7 +6865,7 @@ create_stream_for_sink(struct drm_connector *connector,
 	}
 
 	if (!aconnector || !aconnector->dc_sink) {
-		sink = create_fake_sink(link);
+		sink = create_fake_sink(dev, link);
 		if (!sink)
 			return stream;
 
@@ -6931,7 +6877,7 @@ create_stream_for_sink(struct drm_connector *connector,
 	stream = dc_create_stream_for_sink(sink);
 
 	if (stream == NULL) {
-		DRM_ERROR("Failed to create stream for sink!\n");
+		drm_err(dev, "Failed to create stream for sink!\n");
 		goto finish;
 	}
 
@@ -6963,7 +6909,7 @@ create_stream_for_sink(struct drm_connector *connector,
 		 * case, we call set mode ourselves to restore the previous mode
 		 * and the modelist may not be filled in time.
 		 */
-		DRM_DEBUG_DRIVER("No preferred mode found\n");
+		drm_dbg_driver(dev, "No preferred mode found\n");
 	} else if (aconnector) {
 		recalculate_timing = amdgpu_freesync_vid_mode &&
 				 is_freesync_video_mode(&mode, aconnector);
@@ -7413,6 +7359,7 @@ static void amdgpu_dm_connector_funcs_force(struct drm_connector *connector)
 	struct dc_sink *dc_em_sink = aconnector->dc_em_sink;
 	const struct drm_edid *drm_edid;
 	struct i2c_adapter *ddc;
+	struct drm_device *dev = connector->dev;
 
 	if (dc_link && dc_link->aux_mode)
 		ddc = &aconnector->dm_dp_aux.aux.ddc;
@@ -7422,7 +7369,7 @@ static void amdgpu_dm_connector_funcs_force(struct drm_connector *connector)
 	drm_edid = drm_edid_read_ddc(connector, ddc);
 	drm_edid_connector_update(connector, drm_edid);
 	if (!drm_edid) {
-		DRM_ERROR("No EDID found on connector: %s.\n", connector->name);
+		drm_err(dev, "No EDID found on connector: %s.\n", connector->name);
 		return;
 	}
 
@@ -7481,7 +7428,7 @@ static void create_eml_sink(struct amdgpu_dm_connector *aconnector)
 	drm_edid = drm_edid_read_ddc(connector, ddc);
 	drm_edid_connector_update(connector, drm_edid);
 	if (!drm_edid) {
-		DRM_ERROR("No EDID found on connector: %s.\n", connector->name);
+		drm_err(connector->dev, "No EDID found on connector: %s.\n", connector->name);
 		return;
 	}
 
@@ -7615,7 +7562,7 @@ create_validate_stream_for_sink(struct drm_connector *connector,
 						dm_state, old_stream,
 						requested_bpc);
 		if (stream == NULL) {
-			DRM_ERROR("Failed to create stream for sink!\n");
+			drm_err(adev_to_drm(adev), "Failed to create stream for sink!\n");
 			break;
 		}
 
@@ -7690,7 +7637,7 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec
 
 	if (dc_sink == NULL && aconnector->base.force != DRM_FORCE_ON_DIGITAL &&
 				aconnector->base.force != DRM_FORCE_ON) {
-		DRM_ERROR("dc_sink is NULL!\n");
+		drm_err(connector->dev, "dc_sink is NULL!\n");
 		goto fail;
 	}
 
@@ -8598,7 +8545,7 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm,
 
 	i2c = create_i2c(link->ddc, false);
 	if (!i2c) {
-		DRM_ERROR("Failed to create i2c adapter data\n");
+		drm_err(adev_to_drm(dm->adev), "Failed to create i2c adapter data\n");
 		return -ENOMEM;
 	}
 
@@ -8606,7 +8553,7 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm,
 	res = i2c_add_adapter(&i2c->base);
 
 	if (res) {
-		DRM_ERROR("Failed to register hw i2c %d\n", link->link_index);
+		drm_err(adev_to_drm(dm->adev), "Failed to register hw i2c %d\n", link->link_index);
 		goto out_free;
 	}
 
@@ -8620,7 +8567,7 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm,
 			&i2c->base);
 
 	if (res) {
-		DRM_ERROR("connector_init failed\n");
+		drm_err(adev_to_drm(dm->adev), "connector_init failed\n");
 		aconnector->connector_id = -1;
 		goto out_free;
 	}
@@ -9110,7 +9057,7 @@ static void amdgpu_dm_handle_vrr_transition(struct dm_crtc_state *old_state,
 		 */
 		WARN_ON(amdgpu_dm_crtc_set_vupdate_irq(new_state->base.crtc, true) != 0);
 		WARN_ON(drm_crtc_vblank_get(new_state->base.crtc) != 0);
-		DRM_DEBUG_DRIVER("%s: crtc=%u VRR off->on: Get vblank ref\n",
+		drm_dbg_driver(new_state->base.crtc->dev, "%s: crtc=%u VRR off->on: Get vblank ref\n",
 				 __func__, new_state->base.crtc->base.id);
 	} else if (old_vrr_active && !new_vrr_active) {
 		/* Transition VRR active -> inactive:
@@ -9118,7 +9065,7 @@ static void amdgpu_dm_handle_vrr_transition(struct dm_crtc_state *old_state,
 		 */
 		WARN_ON(amdgpu_dm_crtc_set_vupdate_irq(new_state->base.crtc, false) != 0);
 		drm_crtc_vblank_put(new_state->base.crtc);
-		DRM_DEBUG_DRIVER("%s: crtc=%u VRR on->off: Drop vblank ref\n",
+		drm_dbg_driver(new_state->base.crtc->dev, "%s: crtc=%u VRR on->off: Drop vblank ref\n",
 				 __func__, new_state->base.crtc->base.id);
 	}
 }
@@ -9205,13 +9152,13 @@ static void amdgpu_dm_update_cursor(struct drm_plane *plane,
 	if (crtc_state->stream) {
 		if (!dc_stream_set_cursor_attributes(crtc_state->stream,
 						     &attributes))
-			DRM_ERROR("DC failed to set cursor attributes\n");
+			drm_err(adev_to_drm(adev), "DC failed to set cursor attributes\n");
 
 		update->cursor_attributes = &crtc_state->stream->cursor_attributes;
 
 		if (!dc_stream_set_cursor_position(crtc_state->stream,
 						   &position))
-			DRM_ERROR("DC failed to set cursor position\n");
+			drm_err(adev_to_drm(adev), "DC failed to set cursor position\n");
 
 		update->cursor_position = &crtc_state->stream->cursor_position;
 	}
@@ -9462,7 +9409,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		bundle->surface_updates[planes_count].surface = dc_plane;
 
 		if (!bundle->surface_updates[planes_count].surface) {
-			DRM_ERROR("No surface for CRTC: id=%d\n",
+			drm_err(dev, "No surface for CRTC: id=%d\n",
 					acrtc_attach->crtc_id);
 			continue;
 		}
@@ -9978,20 +9925,20 @@ static void dm_set_writeback(struct amdgpu_display_manager *dm,
 
 	wb_info = kzalloc(sizeof(*wb_info), GFP_KERNEL);
 	if (!wb_info) {
-		DRM_ERROR("Failed to allocate wb_info\n");
+		drm_err(adev_to_drm(adev), "Failed to allocate wb_info\n");
 		return;
 	}
 
 	acrtc = to_amdgpu_crtc(wb_conn->encoder.crtc);
 	if (!acrtc) {
-		DRM_ERROR("no amdgpu_crtc found\n");
+		drm_err(adev_to_drm(adev), "no amdgpu_crtc found\n");
 		kfree(wb_info);
 		return;
 	}
 
 	afb = to_amdgpu_framebuffer(new_con_state->writeback_job->fb);
 	if (!afb) {
-		DRM_ERROR("No amdgpu_framebuffer found\n");
+		drm_err(adev_to_drm(adev), "No amdgpu_framebuffer found\n");
 		kfree(wb_info);
 		return;
 	}
@@ -10212,7 +10159,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 				new_con_state->content_protection >= DRM_MODE_CONTENT_PROTECTION_DESIRED)
 				enable_encryption = true;
 
-			DRM_INFO("[HDCP_DM] hdcp_update_display enable_encryption = %x\n", enable_encryption);
+			drm_info(adev_to_drm(adev), "[HDCP_DM] hdcp_update_display enable_encryption = %x\n", enable_encryption);
 
 			if (aconnector->dc_link)
 				hdcp_update_display(
@@ -10304,7 +10251,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		 */
 		dummy_updates = kzalloc(sizeof(struct dc_surface_update) * MAX_SURFACES, GFP_ATOMIC);
 		if (!dummy_updates) {
-			DRM_ERROR("Failed to allocate memory for dummy_updates.\n");
+			drm_err(adev_to_drm(adev), "Failed to allocate memory for dummy_updates.\n");
 			continue;
 		}
 		for (j = 0; j < status->plane_count; j++)
@@ -10512,16 +10459,20 @@ static int dm_force_atomic_commit(struct drm_connector *connector)
 	 */
 	conn_state = drm_atomic_get_connector_state(state, connector);
 
-	ret = PTR_ERR_OR_ZERO(conn_state);
-	if (ret)
+	/* Check for error in getting connector state */
+	if (IS_ERR(conn_state)) {
+		ret = PTR_ERR(conn_state);
 		goto out;
+	}
 
 	/* Attach crtc to drm_atomic_state*/
 	crtc_state = drm_atomic_get_crtc_state(state, &disconnected_acrtc->base);
 
-	ret = PTR_ERR_OR_ZERO(crtc_state);
-	if (ret)
+	/* Check for error in getting crtc state */
+	if (IS_ERR(crtc_state)) {
+		ret = PTR_ERR(crtc_state);
 		goto out;
+	}
 
 	/* force a restore */
 	crtc_state->mode_changed = true;
@@ -10529,9 +10480,11 @@ static int dm_force_atomic_commit(struct drm_connector *connector)
 	/* Attach plane to drm_atomic_state */
 	plane_state = drm_atomic_get_plane_state(state, plane);
 
-	ret = PTR_ERR_OR_ZERO(plane_state);
-	if (ret)
+	/* Check for error in getting plane state */
+	if (IS_ERR(plane_state)) {
+		ret = PTR_ERR(plane_state);
 		goto out;
+	}
 
 	/* Call commit internally with the state we just constructed */
 	ret = drm_atomic_commit(state);
@@ -10539,7 +10492,7 @@ static int dm_force_atomic_commit(struct drm_connector *connector)
 out:
 	drm_atomic_state_put(state);
 	if (ret)
-		DRM_ERROR("Restoring old state failed with %i\n", ret);
+		drm_err(ddev, "Restoring old state failed with %i\n", ret);
 
 	return ret;
 }
@@ -10623,7 +10576,7 @@ static int do_aquire_global_lock(struct drm_device *dev,
 					&commit->flip_done, 10*HZ);
 
 		if (ret == 0)
-			DRM_ERROR("[CRTC:%d:%s] hw_done or flip_done timed out\n",
+			drm_err(dev, "[CRTC:%d:%s] hw_done or flip_done timed out\n",
 				  crtc->base.id, crtc->name);
 
 		drm_crtc_commit_put(commit);
@@ -10739,6 +10692,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 	struct dm_atomic_state *dm_state = NULL;
 	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
 	struct dc_stream_state *new_stream;
+	struct amdgpu_device *adev = dm->adev;
 	int ret = 0;
 
 	/*
@@ -10768,8 +10722,8 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		drm_old_conn_state = drm_atomic_get_old_connector_state(state,
 									connector);
 
-		if (IS_ERR(drm_new_conn_state)) {
-			ret = PTR_ERR_OR_ZERO(drm_new_conn_state);
+		if (WARN_ON(!drm_new_conn_state)) {
+			ret = -EINVAL;
 			goto fail;
 		}
 
@@ -10792,7 +10746,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		 */
 
 		if (!new_stream) {
-			DRM_DEBUG_DRIVER("%s: Failed to create new stream for crtc %d\n",
+			drm_dbg_driver(adev_to_drm(adev), "%s: Failed to create new stream for crtc %d\n",
 					__func__, acrtc->base.base.id);
 			ret = -ENOMEM;
 			goto fail;
@@ -10830,7 +10784,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		    dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) &&
 		    dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream)) {
 			new_crtc_state->mode_changed = false;
-			DRM_DEBUG_DRIVER("Mode change not required, setting mode_changed to %d",
+			drm_dbg_driver(adev_to_drm(adev), "Mode change not required, setting mode_changed to %d",
 					 new_crtc_state->mode_changed);
 		}
 	}
@@ -10868,7 +10822,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		    is_timing_unchanged_for_freesync(new_crtc_state,
 						     old_crtc_state)) {
 			new_crtc_state->mode_changed = false;
-			DRM_DEBUG_DRIVER(
+			drm_dbg_driver(adev_to_drm(adev),
 				"Mode change not required for front porch change, setting mode_changed to %d",
 				new_crtc_state->mode_changed);
 
@@ -10889,7 +10843,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		if (ret)
 			goto fail;
 
-		DRM_DEBUG_DRIVER("Disabling DRM crtc: %d\n",
+		drm_dbg_driver(adev_to_drm(adev), "Disabling DRM crtc: %d\n",
 				crtc->base.id);
 
 		/* i.e. reset mode */
@@ -11022,6 +10976,9 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 	    state->allow_modeset)
 		return true;
 
+	if (amdgpu_in_reset(adev) && state->allow_modeset)
+		return true;
+
 	/* Exit early if we know that we're adding or removing the plane. */
 	if (old_plane_state->crtc != new_plane_state->crtc)
 		return true;
@@ -11739,7 +11696,7 @@ static bool amdgpu_dm_crtc_mem_type_changed(struct drm_device *dev,
 		old_plane_state = drm_atomic_get_plane_state(state, plane);
 
 		if (IS_ERR(new_plane_state) || IS_ERR(old_plane_state)) {
-			DRM_ERROR("Failed to get plane state for plane %s\n", plane->name);
+			drm_err(dev, "Failed to get plane state for plane %s\n", plane->name);
 			return false;
 		}
 
@@ -12308,7 +12265,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm,
 
 	res = dc_wake_and_execute_dmub_cmd(dm->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY);
 	if (!res) {
-		DRM_ERROR("EDID CEA parser failed\n");
+		drm_err(adev_to_drm(dm->adev), "EDID CEA parser failed\n");
 		return false;
 	}
 
@@ -12316,7 +12273,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm,
 
 	if (output->type == DMUB_CMD__EDID_CEA_ACK) {
 		if (!output->ack.success) {
-			DRM_ERROR("EDID CEA ack failed at offset %d\n",
+			drm_err(adev_to_drm(dm->adev), "EDID CEA ack failed at offset %d\n",
 					output->ack.offset);
 		}
 	} else if (output->type == DMUB_CMD__EDID_CEA_AMD_VSDB) {
@@ -12328,7 +12285,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm,
 		vsdb->min_refresh_rate_hz = output->amd_vsdb.min_frame_rate;
 		vsdb->max_refresh_rate_hz = output->amd_vsdb.max_frame_rate;
 	} else {
-		DRM_WARN("Unknown EDID CEA parser results\n");
+		drm_warn(adev_to_drm(dm->adev), "Unknown EDID CEA parser results\n");
 		return false;
 	}
 
@@ -12544,7 +12501,7 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 	enum adaptive_sync_type as_type = ADAPTIVE_SYNC_TYPE_NONE;
 
 	if (!connector->state) {
-		DRM_ERROR("%s - Connector has no state", __func__);
+		drm_err(adev_to_drm(adev), "%s - Connector has no state", __func__);
 		goto update;
 	}
 
@@ -12729,7 +12686,7 @@ int amdgpu_dm_process_dmub_aux_transfer_sync(
 	}
 
 	if (!wait_for_completion_timeout(&adev->dm.dmub_aux_transfer_done, 10 * HZ)) {
-		DRM_ERROR("wait_for_completion_timeout timeout!");
+		drm_err(adev_to_drm(adev), "wait_for_completion_timeout timeout!");
 		*operation_result = AUX_RET_ERROR_TIMEOUT;
 		goto out;
 	}
@@ -12739,31 +12696,24 @@ int amdgpu_dm_process_dmub_aux_transfer_sync(
 		 * Transient states before tunneling is enabled could
 		 * lead to this error. We can ignore this for now.
 		 */
-		if (p_notify->result != AUX_RET_ERROR_PROTOCOL_ERROR) {
-			DRM_WARN("DPIA AUX failed on 0x%x(%d), error %d\n",
+		if (p_notify->result == AUX_RET_ERROR_PROTOCOL_ERROR) {
+			drm_warn(adev_to_drm(adev), "DPIA AUX failed on 0x%x(%d), error %d\n",
 					payload->address, payload->length,
 					p_notify->result);
 		}
-		*operation_result = AUX_RET_ERROR_INVALID_REPLY;
+		*operation_result = p_notify->result;
 		goto out;
 	}
 
+	payload->reply[0] = adev->dm.dmub_notify->aux_reply.command & 0xF;
+	if (adev->dm.dmub_notify->aux_reply.command & 0xF0)
+		/* The reply is stored in the top nibble of the command. */
+		payload->reply[0] = (adev->dm.dmub_notify->aux_reply.command >> 4) & 0xF;
 
-	payload->reply[0] = adev->dm.dmub_notify->aux_reply.command;
-	if (!payload->write && p_notify->aux_reply.length &&
-			(payload->reply[0] == AUX_TRANSACTION_REPLY_AUX_ACK)) {
-
-		if (payload->length != p_notify->aux_reply.length) {
-			DRM_WARN("invalid read length %d from DPIA AUX 0x%x(%d)!\n",
-				p_notify->aux_reply.length,
-					payload->address, payload->length);
-			*operation_result = AUX_RET_ERROR_INVALID_REPLY;
-			goto out;
-		}
-
+	/*write req may receive a byte indicating partially written number as well*/
+	if (p_notify->aux_reply.length)
 		memcpy(payload->data, p_notify->aux_reply.data,
 				p_notify->aux_reply.length);
-	}
 
 	/* success */
 	ret = p_notify->aux_reply.length;
@@ -12774,6 +12724,79 @@ out:
 	return ret;
 }
 
+static void abort_fused_io(
+		struct dc_context *ctx,
+		const struct dmub_cmd_fused_request *request
+)
+{
+	union dmub_rb_cmd command = { 0 };
+	struct dmub_rb_cmd_fused_io *io = &command.fused_io;
+
+	io->header.type = DMUB_CMD__FUSED_IO;
+	io->header.sub_type = DMUB_CMD__FUSED_IO_ABORT;
+	io->header.payload_bytes = sizeof(*io) - sizeof(io->header);
+	io->request = *request;
+	dm_execute_dmub_cmd(ctx, &command, DM_DMUB_WAIT_TYPE_NO_WAIT);
+}
+
+static bool execute_fused_io(
+		struct amdgpu_device *dev,
+		struct dc_context *ctx,
+		union dmub_rb_cmd *commands,
+		uint8_t count,
+		uint32_t timeout_us
+)
+{
+	const uint8_t ddc_line = commands[0].fused_io.request.u.aux.ddc_line;
+
+	if (ddc_line >= ARRAY_SIZE(dev->dm.fused_io))
+		return false;
+
+	struct fused_io_sync *sync = &dev->dm.fused_io[ddc_line];
+	struct dmub_rb_cmd_fused_io *first = &commands[0].fused_io;
+	const bool result = dm_execute_dmub_cmd_list(ctx, count, commands, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)
+			&& first->header.ret_status
+			&& first->request.status == FUSED_REQUEST_STATUS_SUCCESS;
+
+	if (!result)
+		return false;
+
+	while (wait_for_completion_timeout(&sync->replied, usecs_to_jiffies(timeout_us))) {
+		reinit_completion(&sync->replied);
+
+		struct dmub_cmd_fused_request *reply = (struct dmub_cmd_fused_request *) sync->reply_data;
+
+		static_assert(sizeof(*reply) <= sizeof(sync->reply_data), "Size mismatch");
+
+		if (reply->identifier == first->request.identifier) {
+			first->request = *reply;
+			return true;
+		}
+	}
+
+	reinit_completion(&sync->replied);
+	first->request.status = FUSED_REQUEST_STATUS_TIMEOUT;
+	abort_fused_io(ctx, &first->request);
+	return false;
+}
+
+bool amdgpu_dm_execute_fused_io(
+		struct amdgpu_device *dev,
+		struct dc_link *link,
+		union dmub_rb_cmd *commands,
+		uint8_t count,
+		uint32_t timeout_us)
+{
+	struct amdgpu_display_manager *dm = &dev->dm;
+
+	mutex_lock(&dm->dpia_aux_lock);
+
+	const bool result = execute_fused_io(dev, link->ctx, commands, count, timeout_us);
+
+	mutex_unlock(&dm->dpia_aux_lock);
+	return result;
+}
+
 int amdgpu_dm_process_dmub_set_config_sync(
 		struct dc_context *ctx,
 		unsigned int link_index,
@@ -12792,7 +12815,7 @@ int amdgpu_dm_process_dmub_set_config_sync(
 		ret = 0;
 		*operation_result = adev->dm.dmub_notify->sc_status;
 	} else {
-		DRM_ERROR("wait_for_completion_timeout timeout!");
+		drm_err(adev_to_drm(adev), "wait_for_completion_timeout timeout!");
 		ret = -1;
 		*operation_result = SET_CONFIG_UNKNOWN_ERROR;
 	}
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 385faaca6e26..d7d92f9911e4 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -50,7 +50,7 @@
 
 #define AMDGPU_DM_MAX_NUM_EDP 2
 
-#define AMDGPU_DMUB_NOTIFICATION_MAX 7
+#define AMDGPU_DMUB_NOTIFICATION_MAX 8
 
 #define HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_IEEE_REGISTRATION_ID 0x00001A
 #define AMD_VSDB_VERSION_3_FEATURECAP_REPLAYMODE 0x40
@@ -81,6 +81,7 @@ struct amdgpu_bo;
 struct dmub_srv;
 struct dc_plane_state;
 struct dmub_notification;
+struct dmub_cmd_fused_request;
 
 struct amd_vsdb_block {
 	unsigned char ieee_id[3];
@@ -276,6 +277,10 @@ struct hpd_rx_irq_offload_work {
 	 * @offload_wq: offload work queue that this work is queued to
 	 */
 	struct hpd_rx_irq_offload_work_queue *offload_wq;
+	/**
+	 * @adev: amdgpu_device pointer
+	 */
+	struct amdgpu_device *adev;
 };
 
 /**
@@ -614,6 +619,13 @@ struct amdgpu_display_manager {
 	bool aux_hpd_discon_quirk;
 
 	/**
+	 * @edp0_on_dp1_quirk:
+	 *
+	 * quirk for platforms that put edp0 on DP1.
+	 */
+	bool edp0_on_dp1_quirk;
+
+	/**
 	 * @dpia_aux_lock:
 	 *
 	 * Guards access to DPIA AUX
@@ -633,6 +645,16 @@ struct amdgpu_display_manager {
 	 * OEM i2c bus
 	 */
 	struct amdgpu_i2c_adapter *oem_i2c;
+
+	/**
+	 * @fused_io:
+	 *
+	 * dmub fused io interface
+	 */
+	struct fused_io_sync {
+		struct completion replied;
+		char reply_data[0x40];  // Cannot include dmub_cmd here
+	} fused_io[8];
 };
 
 enum dsc_clock_force_state {
@@ -1012,6 +1034,14 @@ extern const struct drm_encoder_helper_funcs amdgpu_dm_encoder_helper_funcs;
 int amdgpu_dm_process_dmub_aux_transfer_sync(struct dc_context *ctx, unsigned int link_index,
 					struct aux_payload *payload, enum aux_return_code_type *operation_result);
 
+bool amdgpu_dm_execute_fused_io(
+		struct amdgpu_device *dev,
+		struct dc_link *link,
+		union dmub_rb_cmd *commands,
+		uint8_t count,
+		uint32_t timeout_us
+);
+
 int amdgpu_dm_process_dmub_set_config_sync(struct dc_context *ctx, unsigned int link_index,
 					struct set_config_cmd_payload *payload, enum set_config_status *operation_result);
 
@@ -1045,4 +1075,6 @@ void hdmi_cec_set_edid(struct amdgpu_dm_connector *aconnector);
 void hdmi_cec_unset_edid(struct amdgpu_dm_connector *aconnector);
 int amdgpu_dm_initialize_hdmi_connector(struct amdgpu_dm_connector *aconnector);
 
+void retrieve_dmi_info(struct amdgpu_display_manager *dm);
+
 #endif /* __AMDGPU_DM_H__ */
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index 36a830a7440f..e8bdd7f0c460 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -113,6 +113,7 @@ bool amdgpu_dm_crtc_vrr_active(const struct dm_crtc_state *dm_state)
  *
  * Panel Replay and PSR SU
  * - Enable when:
+ *      - VRR is disabled
  *      - vblank counter is disabled
  *      - entry is allowed: usermode demonstrates an adequate number of fast
  *        commits)
@@ -131,19 +132,20 @@ static void amdgpu_dm_crtc_set_panel_sr_feature(
 	bool is_sr_active = (link->replay_settings.replay_allow_active ||
 				 link->psr_settings.psr_allow_active);
 	bool is_crc_window_active = false;
+	bool vrr_active = amdgpu_dm_crtc_vrr_active_irq(vblank_work->acrtc);
 
 #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY
 	is_crc_window_active =
 		amdgpu_dm_crc_window_is_activated(&vblank_work->acrtc->base);
 #endif
 
-	if (link->replay_settings.replay_feature_enabled &&
+	if (link->replay_settings.replay_feature_enabled && !vrr_active &&
 		allow_sr_entry && !is_sr_active && !is_crc_window_active) {
 		amdgpu_dm_replay_enable(vblank_work->stream, true);
 	} else if (vblank_enabled) {
 		if (link->psr_settings.psr_version < DC_PSR_VERSION_SU_1 && is_sr_active)
 			amdgpu_dm_psr_disable(vblank_work->stream, false);
-	} else if (link->psr_settings.psr_feature_enabled &&
+	} else if (link->psr_settings.psr_feature_enabled && !vrr_active &&
 		allow_sr_entry && !is_sr_active && !is_crc_window_active) {
 
 		struct amdgpu_dm_connector *aconn =
@@ -244,6 +246,8 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work)
 	struct vblank_control_work *vblank_work =
 		container_of(work, struct vblank_control_work, work);
 	struct amdgpu_display_manager *dm = vblank_work->dm;
+	struct amdgpu_device *adev = drm_to_adev(dm->ddev);
+	int r;
 
 	mutex_lock(&dm->dc_lock);
 
@@ -271,8 +275,15 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work)
 			vblank_work->acrtc->dm_irq_params.allow_sr_entry);
 	}
 
-	if (dm->active_vblank_irq_count == 0)
+	if (dm->active_vblank_irq_count == 0) {
+		r = amdgpu_dpm_pause_power_profile(adev, true);
+		if (r)
+			dev_warn(adev->dev, "failed to set default power profile mode\n");
 		dc_allow_idle_optimizations(dm->dc, true);
+		r = amdgpu_dpm_pause_power_profile(adev, false);
+		if (r)
+			dev_warn(adev->dev, "failed to restore the power profile mode\n");
+	}
 
 	mutex_unlock(&dm->dc_lock);
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
index 5198a079b463..c16962256514 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
@@ -26,6 +26,7 @@
 #include "amdgpu_dm_hdcp.h"
 #include "amdgpu.h"
 #include "amdgpu_dm.h"
+#include "dc_fused_io.h"
 #include "dm_helpers.h"
 #include <drm/display/drm_hdcp_helper.h>
 #include "hdcp_psp.h"
@@ -76,6 +77,34 @@ lp_read_dpcd(void *handle, uint32_t address, uint8_t *data, uint32_t size)
 	return dm_helpers_dp_read_dpcd(link->ctx, link, address, data, size);
 }
 
+static bool lp_atomic_write_poll_read_i2c(
+		void *handle,
+		const struct mod_hdcp_atomic_op_i2c *write,
+		const struct mod_hdcp_atomic_op_i2c *poll,
+		struct mod_hdcp_atomic_op_i2c *read,
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+)
+{
+	struct dc_link *link = handle;
+
+	return dm_atomic_write_poll_read_i2c(link, write, poll, read, poll_timeout_us, poll_mask_msb);
+}
+
+static bool lp_atomic_write_poll_read_aux(
+		void *handle,
+		const struct mod_hdcp_atomic_op_aux *write,
+		const struct mod_hdcp_atomic_op_aux *poll,
+		struct mod_hdcp_atomic_op_aux *read,
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+)
+{
+	struct dc_link *link = handle;
+
+	return dm_atomic_write_poll_read_aux(link, write, poll, read, poll_timeout_us, poll_mask_msb);
+}
+
 static uint8_t *psp_get_srm(struct psp_context *psp, uint32_t *srm_version, uint32_t *srm_size)
 {
 	struct ta_hdcp_shared_memory *hdcp_cmd;
@@ -173,6 +202,9 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work,
 	unsigned int conn_index = aconnector->base.index;
 
 	guard(mutex)(&hdcp_w->mutex);
+	drm_connector_get(&aconnector->base);
+	if (hdcp_w->aconnector[conn_index])
+		drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
 	hdcp_w->aconnector[conn_index] = aconnector;
 
 	memset(&link_adjust, 0, sizeof(link_adjust));
@@ -220,7 +252,6 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,
 	unsigned int conn_index = aconnector->base.index;
 
 	guard(mutex)(&hdcp_w->mutex);
-	hdcp_w->aconnector[conn_index] = aconnector;
 
 	/* the removal of display will invoke auth reset -> hdcp destroy and
 	 * we'd expect the Content Protection (CP) property changed back to
@@ -236,7 +267,10 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,
 	}
 
 	mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output);
-
+	if (hdcp_w->aconnector[conn_index]) {
+		drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
+		hdcp_w->aconnector[conn_index] = NULL;
+	}
 	process_output(hdcp_w);
 }
 
@@ -254,6 +288,10 @@ void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_inde
 	for (conn_index = 0; conn_index < AMDGPU_DM_MAX_DISPLAY_INDEX; conn_index++) {
 		hdcp_w->encryption_status[conn_index] =
 			MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF;
+		if (hdcp_w->aconnector[conn_index]) {
+			drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
+			hdcp_w->aconnector[conn_index] = NULL;
+		}
 	}
 
 	process_output(hdcp_w);
@@ -488,6 +526,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 	struct hdcp_workqueue *hdcp_work = handle;
 	struct amdgpu_dm_connector *aconnector = config->dm_stream_ctx;
 	int link_index = aconnector->dc_link->link_index;
+	unsigned int conn_index = aconnector->base.index;
 	struct mod_hdcp_display *display = &hdcp_work[link_index].display;
 	struct mod_hdcp_link *link = &hdcp_work[link_index].link;
 	struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index];
@@ -544,7 +583,10 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 	guard(mutex)(&hdcp_w->mutex);
 
 	mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output);
-
+	drm_connector_get(&aconnector->base);
+	if (hdcp_w->aconnector[conn_index])
+		drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
+	hdcp_w->aconnector[conn_index] = aconnector;
 	process_output(hdcp_w);
 }
 
@@ -719,7 +761,10 @@ struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev,
 		INIT_DELAYED_WORK(&hdcp_work[i].watchdog_timer_dwork, event_watchdog_timer);
 		INIT_DELAYED_WORK(&hdcp_work[i].property_validate_dwork, event_property_validate);
 
-		hdcp_work[i].hdcp.config.psp.handle = &adev->psp;
+		struct mod_hdcp_config *config = &hdcp_work[i].hdcp.config;
+		struct mod_hdcp_ddc_funcs *ddc_funcs = &config->ddc.funcs;
+
+		config->psp.handle = &adev->psp;
 		if (dc->ctx->dce_version == DCN_VERSION_3_1 ||
 		    dc->ctx->dce_version == DCN_VERSION_3_14 ||
 		    dc->ctx->dce_version == DCN_VERSION_3_15 ||
@@ -727,12 +772,22 @@ struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev,
 		    dc->ctx->dce_version == DCN_VERSION_3_51 ||
 		    dc->ctx->dce_version == DCN_VERSION_3_6 ||
 		    dc->ctx->dce_version == DCN_VERSION_3_16)
-			hdcp_work[i].hdcp.config.psp.caps.dtm_v3_supported = 1;
-		hdcp_work[i].hdcp.config.ddc.handle = dc_get_link_at_index(dc, i);
-		hdcp_work[i].hdcp.config.ddc.funcs.write_i2c = lp_write_i2c;
-		hdcp_work[i].hdcp.config.ddc.funcs.read_i2c = lp_read_i2c;
-		hdcp_work[i].hdcp.config.ddc.funcs.write_dpcd = lp_write_dpcd;
-		hdcp_work[i].hdcp.config.ddc.funcs.read_dpcd = lp_read_dpcd;
+			config->psp.caps.dtm_v3_supported = 1;
+		config->ddc.handle = dc_get_link_at_index(dc, i);
+
+		ddc_funcs->write_i2c = lp_write_i2c;
+		ddc_funcs->read_i2c = lp_read_i2c;
+		ddc_funcs->write_dpcd = lp_write_dpcd;
+		ddc_funcs->read_dpcd = lp_read_dpcd;
+
+		config->debug.lc_enable_sw_fallback = dc->debug.hdcp_lc_enable_sw_fallback;
+		if (dc->caps.fused_io_supported || dc->debug.hdcp_lc_force_fw_enable) {
+			ddc_funcs->atomic_write_poll_read_i2c = lp_atomic_write_poll_read_i2c;
+			ddc_funcs->atomic_write_poll_read_aux = lp_atomic_write_poll_read_aux;
+		} else {
+			ddc_funcs->atomic_write_poll_read_i2c = NULL;
+			ddc_funcs->atomic_write_poll_read_aux = NULL;
+		}
 
 		memset(hdcp_work[i].aconnector, 0,
 		       sizeof(struct amdgpu_dm_connector *) *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 2cd35392e2da..d4395b92fb85 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -630,6 +630,19 @@ bool dm_helpers_submit_i2c(
 	return result;
 }
 
+bool dm_helpers_execute_fused_io(
+		struct dc_context *ctx,
+		struct dc_link *link,
+		union dmub_rb_cmd *commands,
+		uint8_t count,
+		uint32_t timeout_us
+)
+{
+	struct amdgpu_device *dev = ctx->driver_context;
+
+	return amdgpu_dm_execute_fused_io(dev, link, commands, count, timeout_us);
+}
+
 static bool execute_synaptics_rc_command(struct drm_dp_aux *aux,
 		bool is_write_cmd,
 		unsigned char cmd,
@@ -918,7 +931,7 @@ dm_helpers_probe_acpi_edid(void *data, u8 *buf, unsigned int block, size_t len)
 {
 	struct drm_connector *connector = data;
 	struct acpi_device *acpidev = ACPI_COMPANION(connector->dev->dev);
-	unsigned char start = block * EDID_LENGTH;
+	unsigned short start = block * EDID_LENGTH;
 	struct edid *edid;
 	int r;
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 7ceedf626d23..25e8befbcc47 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -51,6 +51,9 @@
 
 #define PEAK_FACTOR_X1000 1006
 
+/*
+ * This function handles both native AUX and I2C-Over-AUX transactions.
+ */
 static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 				  struct drm_dp_aux_msg *msg)
 {
@@ -59,6 +62,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 	enum aux_return_code_type operation_result;
 	struct amdgpu_device *adev;
 	struct ddc_service *ddc;
+	uint8_t copy[16];
 
 	if (WARN_ON(msg->size > 16))
 		return -E2BIG;
@@ -74,6 +78,11 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 			(msg->request & DP_AUX_I2C_WRITE_STATUS_UPDATE) != 0;
 	payload.defer_delay = 0;
 
+	if (payload.write) {
+		memcpy(copy, msg->buffer, msg->size);
+		payload.data = copy;
+	}
+
 	result = dc_link_aux_transfer_raw(TO_DM_AUX(aux)->ddc_service, &payload,
 				      &operation_result);
 
@@ -87,15 +96,25 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 	if (adev->dm.aux_hpd_discon_quirk) {
 		if (msg->address == DP_SIDEBAND_MSG_DOWN_REQ_BASE &&
 			operation_result == AUX_RET_ERROR_HPD_DISCON) {
-			result = 0;
+			result = msg->size;
 			operation_result = AUX_RET_SUCCESS;
 		}
 	}
 
-	if (payload.write && result >= 0)
-		result = msg->size;
+	/*
+	 * result equals to 0 includes the cases of AUX_DEFER/I2C_DEFER
+	 */
+	if (payload.write && result >= 0) {
+		if (result) {
+			/*one byte indicating partially written bytes*/
+			drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX partially written\n");
+			result = payload.data[0];
+		} else if (!payload.reply[0])
+			/*I2C_ACK|AUX_ACK*/
+			result = msg->size;
+	}
 
-	if (result < 0)
+	if (result < 0) {
 		switch (operation_result) {
 		case AUX_RET_SUCCESS:
 			break;
@@ -114,6 +133,13 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
 			break;
 		}
 
+		drm_dbg_dp(adev_to_drm(adev), "amdgpu: DP AUX transfer fail:%d\n", operation_result);
+	}
+
+	if (payload.reply[0])
+		drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX reply command not ACK: 0x%02x.",
+			payload.reply[0]);
+
 	return result;
 }
 
@@ -1713,16 +1739,17 @@ static bool is_dsc_common_config_possible(struct dc_stream_state *stream,
 					  struct dc_dsc_bw_range *bw_range)
 {
 	struct dc_dsc_policy dsc_policy = {0};
+	bool is_dsc_possible;
 
 	dc_dsc_get_policy_for_timing(&stream->timing, 0, &dsc_policy, dc_link_get_highest_encoding_format(stream->link));
-	dc_dsc_compute_bandwidth_range(stream->sink->ctx->dc->res_pool->dscs[0],
-				       stream->sink->ctx->dc->debug.dsc_min_slice_height_override,
-				       dsc_policy.min_target_bpp * 16,
-				       dsc_policy.max_target_bpp * 16,
-				       &stream->sink->dsc_caps.dsc_dec_caps,
-				       &stream->timing, dc_link_get_highest_encoding_format(stream->link), bw_range);
-
-	return bw_range->max_target_bpp_x16 && bw_range->min_target_bpp_x16;
+	is_dsc_possible = dc_dsc_compute_bandwidth_range(stream->sink->ctx->dc->res_pool->dscs[0],
+							 stream->sink->ctx->dc->debug.dsc_min_slice_height_override,
+							 dsc_policy.min_target_bpp * 16,
+							 dsc_policy.max_target_bpp * 16,
+							 &stream->sink->dsc_caps.dsc_dec_caps,
+							 &stream->timing, dc_link_get_highest_encoding_format(stream->link), bw_range);
+
+	return is_dsc_possible;
 }
 #endif
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
index 3e0f45f1711c..b7c6e8d13435 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
@@ -948,13 +948,13 @@ static int amdgpu_dm_plane_helper_prepare_fb(struct drm_plane *plane,
 	adev = amdgpu_ttm_adev(rbo->tbo.bdev);
 	r = amdgpu_bo_reserve(rbo, true);
 	if (r) {
-		dev_err(adev->dev, "fail to reserve bo (%d)\n", r);
+		drm_err(adev_to_drm(adev), "fail to reserve bo (%d)\n", r);
 		return r;
 	}
 
 	r = dma_resv_reserve_fences(rbo->tbo.base.resv, 1);
 	if (r) {
-		dev_err(adev->dev, "reserving fence slot failed (%d)\n", r);
+		drm_err(adev_to_drm(adev), "reserving fence slot failed (%d)\n", r);
 		goto error_unlock;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
index e140b7a04d72..f984cb0cb889 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
@@ -87,14 +87,6 @@ void amdgpu_dm_set_psr_caps(struct dc_link *link)
 
 		link->psr_settings.psr_feature_enabled = true;
 	}
-
-	DRM_INFO("PSR support %d, DC PSR ver %d, sink PSR ver %d DPCD caps 0x%x su_y_granularity %d\n",
-		link->psr_settings.psr_feature_enabled,
-		link->psr_settings.psr_version,
-		link->dpcd_caps.psr_info.psr_version,
-		link->dpcd_caps.psr_info.psr_dpcd_caps.raw,
-		link->dpcd_caps.psr_info.psr2_su_y_granularity_cap);
-
 }
 
 /*
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_quirks.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_quirks.c
new file mode 100644
index 000000000000..1da07ebf9217
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_quirks.c
@@ -0,0 +1,178 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#include <linux/dmi.h>
+
+#include "amdgpu.h"
+#include "amdgpu_dm.h"
+
+struct amdgpu_dm_quirks {
+	bool aux_hpd_discon;
+	bool support_edp0_on_dp1;
+};
+
+static struct amdgpu_dm_quirks quirk_entries = {
+	.aux_hpd_discon = false,
+	.support_edp0_on_dp1 = false
+};
+
+static int edp0_on_dp1_callback(const struct dmi_system_id *id)
+{
+	quirk_entries.support_edp0_on_dp1 = true;
+	return 0;
+}
+
+static int aux_hpd_discon_callback(const struct dmi_system_id *id)
+{
+	quirk_entries.aux_hpd_discon = true;
+	return 0;
+}
+
+static const struct dmi_system_id dmi_quirk_table[] = {
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3660"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3260"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Precision 3460"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Tower Plus 7010"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Tower 7010"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex SFF Plus 7010"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex SFF 7010"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Micro Plus 7010"),
+		},
+	},
+	{
+		.callback = aux_hpd_discon_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex Micro 7010"),
+		},
+	},
+	{
+		.callback = edp0_on_dp1_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP Elite mt645 G8 Mobile Thin Client"),
+		},
+	},
+	{
+		.callback = edp0_on_dp1_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP EliteBook 645 14 inch G11 Notebook PC"),
+		},
+	},
+	{
+		.callback = edp0_on_dp1_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP EliteBook 665 16 inch G11 Notebook PC"),
+		},
+	},
+	{
+		.callback = edp0_on_dp1_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP ProBook 445 14 inch G11 Notebook PC"),
+		},
+	},
+	{
+		.callback = edp0_on_dp1_callback,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP ProBook 465 16 inch G11 Notebook PC"),
+		},
+	},
+	{}
+	/* TODO: refactor this from a fixed table to a dynamic option */
+};
+
+void retrieve_dmi_info(struct amdgpu_display_manager *dm)
+{
+	struct drm_device *dev = dm->ddev;
+	int dmi_id;
+
+	dm->aux_hpd_discon_quirk = false;
+	dm->edp0_on_dp1_quirk = false;
+
+	dmi_id = dmi_check_system(dmi_quirk_table);
+
+	if (!dmi_id)
+		return;
+
+	if (quirk_entries.aux_hpd_discon) {
+		dm->aux_hpd_discon_quirk = true;
+		drm_info(dev, "aux_hpd_discon_quirk attached\n");
+	}
+	if (quirk_entries.support_edp0_on_dp1) {
+		dm->edp0_on_dp1_quirk = true;
+		drm_info(dev, "support_edp0_on_dp1 attached\n");
+	}
+}
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c
index 0d5fefb0f591..d9527c05fc87 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c
@@ -102,13 +102,13 @@ static int amdgpu_dm_wb_prepare_job(struct drm_writeback_connector *wb_connector
 
 	r = amdgpu_bo_reserve(rbo, true);
 	if (r) {
-		dev_err(adev->dev, "fail to reserve bo (%d)\n", r);
+		drm_err(adev_to_drm(adev), "fail to reserve bo (%d)\n", r);
 		return r;
 	}
 
 	r = dma_resv_reserve_fences(rbo->tbo.base.resv, 1);
 	if (r) {
-		dev_err(adev->dev, "reserving fence slot failed (%d)\n", r);
+		drm_err(adev_to_drm(adev), "reserving fence slot failed (%d)\n", r);
 		goto error_unlock;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile
index 3e1f5b689718..3c9ecea7eebc 100644
--- a/drivers/gpu/drm/amd/display/dc/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/Makefile
@@ -53,31 +53,30 @@ DC_LIBS += hdcp
 
 ifdef CONFIG_DRM_AMD_DC_FP
 DC_LIBS += sspl
-DC_SPL_TRANS += dc_spl_translate.o
+AMD_DISPLAY_FILES += $(addprefix $(AMDDALPATH)/dc/, dc_spl_translate.o)
 endif
 
 AMD_DC = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/dc/,$(DC_LIBS)))
 
 include $(AMD_DC)
 
-DISPLAY_CORE = dc.o dc_stat.o dc_resource.o dc_hw_sequencer.o dc_sink.o \
-dc_surface.o dc_debug.o dc_stream.o dc_link_enc_cfg.o dc_link_exports.o dc_state.o
+FILES =
+FILES += dc_dmub_srv.o
+FILES += dc_edid_parser.o
+FILES += dc_fused_io.o
+FILES += dc_helper.o
+FILES += core/dc.o
+FILES += core/dc_debug.o
+FILES += core/dc_hw_sequencer.o
+FILES += core/dc_link_enc_cfg.o
+FILES += core/dc_link_exports.o
+FILES += core/dc_resource.o
+FILES += core/dc_sink.o
+FILES += core/dc_stat.o
+FILES += core/dc_state.o
+FILES += core/dc_stream.o
+FILES += core/dc_surface.o
+FILES += core/dc_vm_helper.o
+
+AMD_DISPLAY_FILES += $(addprefix $(AMDDALPATH)/dc/, $(FILES))
 
-DISPLAY_CORE += dc_vm_helper.o
-
-AMD_DISPLAY_CORE = $(addprefix $(AMDDALPATH)/dc/core/,$(DISPLAY_CORE))
-
-AMD_DM_REG_UPDATE = $(addprefix $(AMDDALPATH)/dc/,dc_helper.o)
-
-AMD_DC_SPL_TRANS = $(addprefix $(AMDDALPATH)/dc/,$(DC_SPL_TRANS))
-
-AMD_DISPLAY_FILES += $(AMD_DISPLAY_CORE)
-AMD_DISPLAY_FILES += $(AMD_DM_REG_UPDATE)
-
-DC_DMUB += dc_dmub_srv.o
-DC_EDID += dc_edid_parser.o
-AMD_DISPLAY_DMUB = $(addprefix $(AMDDALPATH)/dc/,$(DC_DMUB))
-AMD_DISPLAY_EDID = $(addprefix $(AMDDALPATH)/dc/,$(DC_EDID))
-AMD_DISPLAY_FILES += $(AMD_DISPLAY_DMUB) $(AMD_DISPLAY_EDID)
-
-AMD_DISPLAY_FILES += $(AMD_DC_SPL_TRANS)
diff --git a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
index 88d3f9d7dd55..452206b5095e 100644
--- a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
+++ b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
@@ -51,8 +51,6 @@ static inline unsigned long long complete_integer_division_u64(
 {
 	unsigned long long result;
 
-	ASSERT(divisor);
-
 	result = div64_u64_rem(dividend, divisor, remainder);
 
 	return result;
@@ -213,9 +211,6 @@ struct fixed31_32 dc_fixpt_recip(struct fixed31_32 arg)
 	 * @note
 	 * Good idea to use Newton's method
 	 */
-
-	ASSERT(arg.value);
-
 	return dc_fixpt_from_fraction(
 		dc_fixpt_one.value,
 		arg.value);
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
index 3bacf470f7c5..67f08495b7e6 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
@@ -2384,10 +2384,10 @@ static enum bp_result get_integrated_info_v8(
 }
 
 /*
- * get_integrated_info_v8
+ * get_integrated_info_v9
  *
  * @brief
- * Get V8 integrated BIOS information
+ * Get V9 integrated BIOS information
  *
  * @param
  * bios_parser *bp - [in]BIOS parser handler to get master data table
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c
index 19897fa52e7e..d82a52319088 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c
@@ -142,17 +142,3 @@ int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_di
 
 	return actual_dispclk_set_mhz * 1000;
 }
-
-int rv1_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr)
-{
-	int actual_dprefclk_set_mhz = -1;
-
-	actual_dprefclk_set_mhz = rv1_vbios_smu_send_msg_with_param(
-			clk_mgr,
-			VBIOSSMC_MSG_SetDprefclkFreq,
-			khz_to_mhz_ceil(clk_mgr->base.dprefclk_khz));
-
-	/* TODO: add code for programing DP DTO, currently this is down by command table */
-
-	return actual_dprefclk_set_mhz * 1000;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.h
index 083cb3158859..81d7c912549c 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.h
@@ -27,6 +27,5 @@
 #define DAL_DC_DCN10_RV1_CLK_MGR_VBIOS_SMU_H_
 
 int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz);
-int rv1_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr);
 
 #endif /* DAL_DC_DCN10_RV1_CLK_MGR_VBIOS_SMU_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
index 23b390245b5d..5a633333dbb5 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
@@ -164,20 +164,6 @@ int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dis
 	return actual_dispclk_set_mhz * 1000;
 }
 
-int rn_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr)
-{
-	int actual_dprefclk_set_mhz = -1;
-
-	actual_dprefclk_set_mhz = rn_vbios_smu_send_msg_with_param(
-			clk_mgr,
-			VBIOSSMC_MSG_SetDprefclkFreq,
-			khz_to_mhz_ceil(clk_mgr->base.dprefclk_khz));
-
-	/* TODO: add code for programing DP DTO, currently this is down by command table */
-
-	return actual_dprefclk_set_mhz * 1000;
-}
-
 int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_dcfclk_khz)
 {
 	int actual_dcfclk_set_mhz = -1;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h
index 1ce19d875358..f76fad87f0e1 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.h
@@ -30,7 +30,6 @@ enum dcn_pwr_state;
 
 int rn_vbios_smu_get_smu_version(struct clk_mgr_internal *clk_mgr);
 int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dispclk_khz);
-int rn_vbios_smu_set_dprefclk(struct clk_mgr_internal *clk_mgr);
 int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_dcfclk_khz);
 int rn_vbios_smu_set_min_deep_sleep_dcfclk(struct clk_mgr_internal *clk_mgr, int requested_min_ds_dcfclk_khz);
 void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phyclk_khz);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c
index 2d14346b680e..478b4d6a3544 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c
@@ -49,12 +49,9 @@ static const struct IP_BASE MP0_BASE = { { { { 0x00016000, 0x00DC0000, 0x00E0000
 					{ { 0, 0, 0, 0, 0, 0 } },
 					{ { 0, 0, 0, 0, 0, 0 } },
 					{ { 0, 0, 0, 0, 0, 0 } } } };
-static const struct IP_BASE NBIO_BASE = { { { { 0x00000000, 0x00000014, 0x00000D20, 0x00010400, 0x0241B000, 0x04040000 } },
-					{ { 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0 } } } };
+
+#define CTX clk_mgr->base.ctx
+#define IND_REG(offset)	offset
 
 #define regBIF_BX_PF2_RSMU_INDEX                                                                        0x0000
 #define regBIF_BX_PF2_RSMU_INDEX_BASE_IDX                                                               1
@@ -67,9 +64,6 @@ static const struct IP_BASE NBIO_BASE = { { { { 0x00000000, 0x00000014, 0x00000D
 #define FN(reg_name, field) \
 	FD(reg_name##__##field)
 
-#define REG_NBIO(reg_name) \
-	(NBIO_BASE.instance[0].segment[regBIF_BX_PF2_ ## reg_name ## _BASE_IDX] + regBIF_BX_PF2_ ## reg_name)
-
 #undef DC_LOGGER
 #define DC_LOGGER \
 	CTX->logger
@@ -77,6 +71,13 @@ static const struct IP_BASE NBIO_BASE = { { { { 0x00000000, 0x00000014, 0x00000D
 
 #define mmMP1_C2PMSG_3                            0x3B1050C
 
+#define reg__MP1_C2PMSG_3_MASK					(0xFFFFFFFF)
+#define reg__MP1_C2PMSG_3__SHIFT					(0)
+
+
+#define data_reg_name__MP1_C2PMSG_3_MASK		(0xFFFFFFFF)
+#define data_reg_name__MP1_C2PMSG_3__SHIFT		(0)
+
 #define VBIOSSMC_MSG_TestMessage                  0x01 ///< To check if PMFW is alive and responding. Requirement specified by PMFW team
 #define VBIOSSMC_MSG_GetPmfwVersion               0x02 ///< Get PMFW version
 #define VBIOSSMC_MSG_Spare0                       0x03 ///< Spare0
@@ -153,12 +154,10 @@ static int dcn315_smu_send_msg_with_param(
 
 	for (i = 0; i < SMU_REGISTER_WRITE_RETRY_COUNT; i++) {
 		/* Trigger the message transaction by writing the message ID */
-		generic_write_indirect_reg(CTX,
-			REG_NBIO(RSMU_INDEX), REG_NBIO(RSMU_DATA),
-			mmMP1_C2PMSG_3, msg_id);
-		read_back_data = generic_read_indirect_reg(CTX,
-			REG_NBIO(RSMU_INDEX), REG_NBIO(RSMU_DATA),
-			mmMP1_C2PMSG_3);
+		IX_REG_SET_SYNC(mmMP1_C2PMSG_3, 0,
+			MP1_C2PMSG_3, msg_id);
+		IX_REG_GET_SYNC(mmMP1_C2PMSG_3,
+			MP1_C2PMSG_3, &read_back_data);
 		if (read_back_data == msg_id)
 			break;
 		udelay(2);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn351_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn351_clk_mgr.c
index 6a6ae618650b..4607eff07253 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn351_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn351_clk_mgr.c
@@ -65,6 +65,7 @@
 #define mmCLK1_CLK5_ALLOW_DS 0x16EB1
 
 #define mmCLK5_spll_field_8 0x1B04B
+#define mmCLK6_spll_field_8 0x1B24B
 #define mmDENTIST_DISPCLK_CNTL 0x0124
 #define regDENTIST_DISPCLK_CNTL 0x0064
 #define regDENTIST_DISPCLK_CNTL_BASE_IDX 1
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
index 142de8938d7c..bb1ac12a2b09 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
@@ -90,6 +90,7 @@
 #define mmCLK1_CLK5_ALLOW_DS 0x16EB1
 
 #define mmCLK5_spll_field_8 0x1B24B
+#define mmCLK6_spll_field_8 0x1B24B
 #define mmDENTIST_DISPCLK_CNTL 0x0124
 #define regDENTIST_DISPCLK_CNTL 0x0064
 #define regDENTIST_DISPCLK_CNTL_BASE_IDX 1
@@ -116,6 +117,7 @@
 #define DENTIST_DISPCLK_CNTL__DENTIST_DPPCLK_WDIVIDER_MASK 0x7F000000L
 
 #define CLK5_spll_field_8__spll_ssc_en_MASK 0x00002000L
+#define CLK6_spll_field_8__spll_ssc_en_MASK 0x00002000L
 
 #define SMU_VER_THRESHOLD 0x5D4A00 //93.74.0
 #undef FN
@@ -596,7 +598,11 @@ static bool dcn35_is_spll_ssc_enabled(struct clk_mgr *clk_mgr_base)
 
 	uint32_t ssc_enable;
 
-	ssc_enable = REG_READ(CLK5_spll_field_8) & CLK5_spll_field_8__spll_ssc_en_MASK;
+	if (clk_mgr_base->ctx->dce_version == DCN_VERSION_3_51) {
+		ssc_enable = REG_READ(CLK6_spll_field_8) & CLK6_spll_field_8__spll_ssc_en_MASK;
+	} else {
+		ssc_enable = REG_READ(CLK5_spll_field_8) & CLK5_spll_field_8__spll_ssc_en_MASK;
+	}
 
 	return ssc_enable != 0;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c
index f6f0e6a33001..604d256cb47a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c
@@ -84,8 +84,8 @@
 #define VBIOSSMC_MSG_AllowZstatesEntry            0x15
 #define VBIOSSMC_MSG_DisallowZstatesEntry     	  0x16
 #define VBIOSSMC_MSG_SetDtbClk                    0x17
-#define VBIOSSMC_MSG_DispPsrEntry                 0x18 ///< Display PSR entry, DMU
-#define VBIOSSMC_MSG_DispPsrExit                  0x19 ///< Display PSR exit, DMU
+#define VBIOSSMC_MSG_DispIPS2Entry                0x18 ///< Display IPS2 entry, DMU
+#define VBIOSSMC_MSG_DispIPS2Exit                 0x19 ///< Display IPS2 exit, DMU
 #define VBIOSSMC_MSG_DisableLSdma                 0x1A ///< Disable LSDMA; only sent by VBIOS
 #define VBIOSSMC_MSG_DpControllerPhyStatus        0x1B ///< Inform PMFW about the pre conditions for turning SLDO2 on/off . bit[0]==1 precondition is met, bit[1-2] are for DPPHY number
 #define VBIOSSMC_MSG_QueryIPS2Support             0x1C ///< Return 1: support; else not supported
@@ -475,7 +475,7 @@ int dcn35_smu_exit_low_power_state(struct clk_mgr_internal *clk_mgr)
 
 	retv = dcn35_smu_send_msg_with_param(
 		clk_mgr,
-		VBIOSSMC_MSG_DispPsrExit,
+		VBIOSSMC_MSG_DispIPS2Exit,
 		0);
 	smu_print("%s: smu_exit_low_power_state return = %d\n", __func__, retv);
 	return retv;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 28d1353f403d..56d011a1323c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -36,7 +36,9 @@
 #include "resource.h"
 #include "dc_state.h"
 #include "dc_state_priv.h"
+#include "dc_plane.h"
 #include "dc_plane_priv.h"
+#include "dc_stream_priv.h"
 
 #include "gpio_service_interface.h"
 #include "clk_mgr.h"
@@ -439,9 +441,12 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc,
 	 * Don't adjust DRR while there's bandwidth optimizations pending to
 	 * avoid conflicting with firmware updates.
 	 */
-	if (dc->ctx->dce_version > DCE_VERSION_MAX)
-		if (dc->optimized_required || dc->wm_optimized_required)
+	if (dc->ctx->dce_version > DCE_VERSION_MAX) {
+		if (dc->optimized_required || dc->wm_optimized_required) {
+			stream->adjust.timing_adjust_pending = true;
 			return false;
+		}
+	}
 
 	dc_exit_ips_for_hw_access(dc);
 
@@ -1192,6 +1197,12 @@ static void apply_ctx_interdependent_lock(struct dc *dc,
 
 static void dc_update_visual_confirm_color(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx)
 {
+	if (dc->debug.visual_confirm & VISUAL_CONFIRM_EXPLICIT) {
+		memcpy(&pipe_ctx->visual_confirm_color, &pipe_ctx->plane_state->visual_confirm_color,
+		sizeof(pipe_ctx->visual_confirm_color));
+		return;
+	}
+
 	if (dc->ctx->dce_version >= DCN_VERSION_1_0) {
 		memset(&pipe_ctx->visual_confirm_color, 0, sizeof(struct tg_color));
 
@@ -1225,6 +1236,51 @@ static void dc_update_visual_confirm_color(struct dc *dc, struct dc_state *conte
 	}
 }
 
+void dc_get_visual_confirm_for_stream(
+	struct dc *dc,
+	struct dc_stream_state *stream_state,
+	struct tg_color *color)
+{
+	struct dc_stream_status *stream_status = dc_stream_get_status(stream_state);
+	struct pipe_ctx *pipe_ctx;
+	int i;
+	struct dc_plane_state *plane_state = NULL;
+
+	if (!stream_status)
+		return;
+
+	switch (dc->debug.visual_confirm) {
+	case VISUAL_CONFIRM_DISABLE:
+		return;
+	case VISUAL_CONFIRM_PSR:
+	case VISUAL_CONFIRM_FAMS:
+		pipe_ctx = dc_stream_get_pipe_ctx(stream_state);
+		if (!pipe_ctx)
+			return;
+		dc_dmub_srv_get_visual_confirm_color_cmd(dc, pipe_ctx);
+		memcpy(color, &dc->ctx->dmub_srv->dmub->visual_confirm_color, sizeof(struct tg_color));
+		return;
+
+	default:
+		/* find plane with highest layer_index */
+		for (i = 0; i < stream_status->plane_count; i++) {
+			if (stream_status->plane_states[i]->visible)
+				plane_state = stream_status->plane_states[i];
+		}
+		if (!plane_state)
+			return;
+		/* find pipe that contains plane with highest layer index */
+		for (i = 0; i < MAX_PIPES; i++) {
+			struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+			if (pipe->plane_state == plane_state) {
+				memcpy(color, &pipe->visual_confirm_color, sizeof(struct tg_color));
+				return;
+			}
+		}
+	}
+}
+
 static void disable_dangling_plane(struct dc *dc, struct dc_state *context)
 {
 	int i, j;
@@ -2053,6 +2109,18 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 		dc->hwss.enable_accelerated_mode(dc, context);
 	}
 
+	if (dc->hwseq->funcs.wait_for_pipe_update_if_needed) {
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			pipe = &context->res_ctx.pipe_ctx[i];
+			//Only delay otg master for a given config
+			if (resource_is_pipe_type(pipe, OTG_MASTER)) {
+				//dc_commit_state_no_check is always a full update
+				dc->hwseq->funcs.wait_for_pipe_update_if_needed(dc, pipe, false);
+				break;
+			}
+		}
+	}
+
 	if (context->stream_count > get_seamless_boot_stream_count(context) ||
 		context->stream_count == 0)
 		dc->hwss.prepare_bandwidth(dc, context);
@@ -2117,6 +2185,14 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 	if (dc->hwss.program_front_end_for_ctx) {
 		dc->hwss.interdependent_update_lock(dc, context, true);
 		dc->hwss.program_front_end_for_ctx(dc, context);
+
+		if (dc->hwseq->funcs.set_wait_for_update_needed_for_pipe) {
+			for (i = 0; i < dc->res_pool->pipe_count; i++) {
+				pipe = &context->res_ctx.pipe_ctx[i];
+				dc->hwseq->funcs.set_wait_for_update_needed_for_pipe(dc, pipe);
+			}
+		}
+
 		dc->hwss.interdependent_update_lock(dc, context, false);
 		dc->hwss.post_unlock_program_front_end(dc, context);
 	}
@@ -2258,11 +2334,15 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_commit_streams_params
 	for (i = 0; i < params->stream_count; i++) {
 		struct dc_stream_state *stream = params->streams[i];
 		struct dc_stream_status *status = dc_stream_get_status(stream);
+		struct dc_sink *sink = stream->sink;
 
 		/* revalidate streams */
-		res = dc_validate_stream(dc, stream);
-		if (res != DC_OK)
-			return res;
+		if (!dc_is_virtual_signal(sink->sink_signal)) {
+			res = dc_validate_stream(dc, stream);
+			if (res != DC_OK)
+				return res;
+		}
+
 
 		dc_stream_log(dc, stream);
 
@@ -2815,7 +2895,7 @@ static enum surface_update_type check_update_surfaces_for_stream(
 	int i;
 	enum surface_update_type overall_type = UPDATE_TYPE_FAST;
 
-	if (dc->idle_optimizations_allowed)
+	if (dc->idle_optimizations_allowed || dc_can_clear_cursor_limit(dc))
 		overall_type = UPDATE_TYPE_FULL;
 
 	if (stream_status == NULL || stream_status->plane_count != surface_count)
@@ -3168,7 +3248,8 @@ static void copy_stream_update_to_stream(struct dc *dc,
 
 	if (update->crtc_timing_adjust) {
 		if (stream->adjust.v_total_min != update->crtc_timing_adjust->v_total_min ||
-			stream->adjust.v_total_max != update->crtc_timing_adjust->v_total_max)
+			stream->adjust.v_total_max != update->crtc_timing_adjust->v_total_max ||
+			stream->adjust.timing_adjust_pending)
 			update->crtc_timing_adjust->timing_adjust_pending = true;
 		stream->adjust = *update->crtc_timing_adjust;
 		update->crtc_timing_adjust->timing_adjust_pending = false;
@@ -3219,7 +3300,7 @@ static void copy_stream_update_to_stream(struct dc *dc,
 		if (dsc_validate_context) {
 			stream->timing.dsc_cfg = *update->dsc_config;
 			stream->timing.flags.DSC = enable_dsc;
-			if (!dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, true)) {
+			if (dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, true) != DC_OK) {
 				stream->timing.dsc_cfg = old_dsc_cfg;
 				stream->timing.flags.DSC = old_dsc_enabled;
 				update->dsc_config = NULL;
@@ -3248,7 +3329,7 @@ static void backup_planes_and_stream_state(
 		return;
 
 	for (i = 0; i < status->plane_count; i++) {
-		scratch->plane_states[i] = *status->plane_states[i];
+		dc_plane_copy_config(&scratch->plane_states[i], status->plane_states[i]);
 	}
 	scratch->stream_state = *stream;
 }
@@ -3264,10 +3345,7 @@ static void restore_planes_and_stream_state(
 		return;
 
 	for (i = 0; i < status->plane_count; i++) {
-		/* refcount will always be valid, restore everything else */
-		struct kref refcount = status->plane_states[i]->refcount;
-		*status->plane_states[i] = scratch->plane_states[i];
-		status->plane_states[i]->refcount = refcount;
+		dc_plane_copy_config(status->plane_states[i], &scratch->plane_states[i]);
 	}
 	*stream = scratch->stream_state;
 }
@@ -3444,7 +3522,7 @@ static bool update_planes_and_stream_state(struct dc *dc,
 	}
 
 	if (update_type == UPDATE_TYPE_FULL) {
-		if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) {
+		if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) {
 			BREAK_TO_DEBUGGER();
 			goto fail;
 		}
@@ -3998,6 +4076,7 @@ static void commit_planes_for_stream(struct dc *dc,
 				&context->res_ctx,
 				stream);
 	ASSERT(top_pipe_to_program != NULL);
+
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i];
 
@@ -4048,6 +4127,9 @@ static void commit_planes_for_stream(struct dc *dc,
 		dc->hwss.wait_for_dcc_meta_propagation(dc, top_pipe_to_program);
 	}
 
+	if (dc->hwseq->funcs.wait_for_pipe_update_if_needed)
+		dc->hwseq->funcs.wait_for_pipe_update_if_needed(dc, top_pipe_to_program, update_type == UPDATE_TYPE_FAST);
+
 	if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) {
 		if (dc->hwss.subvp_pipe_control_lock)
 			dc->hwss.subvp_pipe_control_lock(dc, context, true, should_lock_all_pipes, NULL, subvp_prev_use);
@@ -4168,12 +4250,6 @@ static void commit_planes_for_stream(struct dc *dc,
 			if (update_type == UPDATE_TYPE_FAST)
 				continue;
 
-			ASSERT(!pipe_ctx->plane_state->triplebuffer_flips);
-			if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) {
-				/*turn off triple buffer for full update*/
-				dc->hwss.program_triplebuffer(
-					dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips);
-			}
 			stream_status =
 				stream_get_status(context, pipe_ctx->stream);
 
@@ -4182,8 +4258,37 @@ static void commit_planes_for_stream(struct dc *dc,
 					dc, pipe_ctx->stream, stream_status->plane_count, context);
 		}
 	}
+
+	for (j = 0; j < dc->res_pool->pipe_count; j++) {
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+
+		if (!pipe_ctx->plane_state)
+			continue;
+
+		/* Full fe update*/
+		if (update_type == UPDATE_TYPE_FAST)
+			continue;
+
+		ASSERT(!pipe_ctx->plane_state->triplebuffer_flips);
+		if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) {
+			/*turn off triple buffer for full update*/
+			dc->hwss.program_triplebuffer(
+				dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips);
+		}
+	}
+
 	if (dc->hwss.program_front_end_for_ctx && update_type != UPDATE_TYPE_FAST) {
 		dc->hwss.program_front_end_for_ctx(dc, context);
+
+		//Pipe busy until some frame and line #
+		if (dc->hwseq->funcs.set_wait_for_update_needed_for_pipe && update_type == UPDATE_TYPE_FULL) {
+			for (j = 0; j < dc->res_pool->pipe_count; j++) {
+				struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+
+				dc->hwseq->funcs.set_wait_for_update_needed_for_pipe(dc, pipe_ctx);
+			}
+		}
+
 		if (dc->debug.validate_dml_output) {
 			for (i = 0; i < dc->res_pool->pipe_count; i++) {
 				struct pipe_ctx *cur_pipe = &context->res_ctx.pipe_ctx[i];
@@ -4523,7 +4628,7 @@ static struct dc_state *create_minimal_transition_state(struct dc *dc,
 
 	backup_and_set_minimal_pipe_split_policy(dc, base_context, policy);
 	/* commit minimal state */
-	if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, false)) {
+	if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, false) == DC_OK) {
 		/* prevent underflow and corruption when reconfiguring pipes */
 		force_vsync_flip_in_minimal_transition_context(minimal_transition_context);
 	} else {
@@ -4958,6 +5063,9 @@ static bool full_update_required(struct dc *dc,
 	if (dc->idle_optimizations_allowed)
 		return true;
 
+	if (dc_can_clear_cursor_limit(dc))
+		return true;
+
 	return false;
 }
 
@@ -5043,7 +5151,7 @@ static bool update_planes_and_stream_v1(struct dc *dc,
 	copy_stream_update_to_stream(dc, context, stream, stream_update);
 
 	if (update_type >= UPDATE_TYPE_FULL) {
-		if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) {
+		if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) {
 			DC_ERROR("Mode validation failed for stream update!\n");
 			dc_state_release(context);
 			return false;
@@ -6187,15 +6295,22 @@ bool dc_abm_save_restore(
 void dc_query_current_properties(struct dc *dc, struct dc_current_properties *properties)
 {
 	unsigned int i;
-	bool subvp_sw_cursor_req = false;
+	unsigned int max_cursor_size = dc->caps.max_cursor_size;
+	unsigned int stream_cursor_size;
 
-	for (i = 0; i < dc->current_state->stream_count; i++) {
-		if (check_subvp_sw_cursor_fallback_req(dc, dc->current_state->streams[i]) && !dc->current_state->streams[i]->hw_cursor_req) {
-			subvp_sw_cursor_req = true;
-			break;
+	if (dc->debug.allow_sw_cursor_fallback && dc->res_pool->funcs->get_max_hw_cursor_size) {
+		for (i = 0; i < dc->current_state->stream_count; i++) {
+			stream_cursor_size = dc->res_pool->funcs->get_max_hw_cursor_size(dc,
+					dc->current_state,
+					dc->current_state->streams[i]);
+
+			if (stream_cursor_size < max_cursor_size) {
+				max_cursor_size = stream_cursor_size;
+			}
 		}
 	}
-	properties->cursor_size_limit = subvp_sw_cursor_req ? 64 : dc->caps.max_cursor_size;
+
+	properties->cursor_size_limit = max_cursor_size;
 }
 
 /**
@@ -6261,3 +6376,27 @@ unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context)
 	else
 		return 0;
 }
+
+bool dc_is_cursor_limit_pending(struct dc *dc)
+{
+	uint32_t i;
+
+	for (i = 0; i < dc->current_state->stream_count; i++) {
+		if (dc_stream_is_cursor_limit_pending(dc, dc->current_state->streams[i]))
+			return true;
+	}
+
+	return false;
+}
+
+bool dc_can_clear_cursor_limit(struct dc *dc)
+{
+	uint32_t i;
+
+	for (i = 0; i < dc->current_state->stream_count; i++) {
+		if (dc_state_can_clear_stream_cursor_subvp_limit(dc->current_state->streams[i], dc->current_state))
+			return true;
+	}
+
+	return false;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
index 650e89825968..7551d0a3fe82 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
@@ -266,6 +266,8 @@ char *dc_status_to_str(enum dc_status status)
 		return "Fail dp payload allocation";
 	case DC_FAIL_DP_LINK_BANDWIDTH:
 		return "Insufficient DP link bandwidth";
+	case DC_FAIL_HW_CURSOR_SUPPORT:
+		return "HW Cursor not supported";
 	case DC_ERROR_UNEXPECTED:
 		return "Unexpected error";
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
index 55b32dfbfdd6..7014b8d000bb 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
@@ -697,7 +697,7 @@ void get_fams2_visual_confirm_color(
 void hwss_build_fast_sequence(struct dc *dc,
 		struct dc_dmub_cmd *dc_dmub_cmd,
 		unsigned int dmub_cmd_count,
-		struct block_sequence block_sequence[],
+		struct block_sequence block_sequence[MAX_HWSS_BLOCK_SEQUENCE_SIZE],
 		unsigned int *num_steps,
 		struct pipe_ctx *pipe_ctx,
 		struct dc_stream_status *stream_status,
@@ -896,7 +896,7 @@ void hwss_build_fast_sequence(struct dc *dc,
 }
 
 void hwss_execute_sequence(struct dc *dc,
-		struct block_sequence block_sequence[],
+		struct block_sequence block_sequence[MAX_HWSS_BLOCK_SEQUENCE_SIZE],
 		int num_steps)
 {
 	unsigned int i;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 313a32248cd7..3da25bd8b578 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -1342,32 +1342,6 @@ static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx)
 	data->viewport_c.y += src.y / vpc_div;
 }
 
-static bool is_subvp_high_refresh_candidate(struct dc_stream_state *stream)
-{
-	uint32_t refresh_rate;
-	struct dc *dc = stream->ctx->dc;
-
-	refresh_rate = (stream->timing.pix_clk_100hz * (uint64_t)100 +
-		stream->timing.v_total * stream->timing.h_total - (uint64_t)1);
-	refresh_rate = div_u64(refresh_rate, stream->timing.v_total);
-	refresh_rate = div_u64(refresh_rate, stream->timing.h_total);
-
-	/* If there's any stream that fits the SubVP high refresh criteria,
-	 * we must return true. This is because cursor updates are asynchronous
-	 * with full updates, so we could transition into a SubVP config and
-	 * remain in HW cursor mode if there's no cursor update which will
-	 * then cause corruption.
-	 */
-	if ((refresh_rate >= 120 && refresh_rate <= 175 &&
-			stream->timing.v_addressable >= 1080 &&
-			stream->timing.v_addressable <= 2160) &&
-			(dc->current_state->stream_count > 1 ||
-			(dc->current_state->stream_count == 1 && !stream->allow_freesync)))
-		return true;
-
-	return false;
-}
-
 static enum controller_dp_test_pattern convert_dp_to_controller_test_pattern(
 				enum dp_test_pattern test_pattern)
 {
@@ -3937,6 +3911,10 @@ enum dc_status resource_map_pool_resources(
 		if (!dc->link_srv->dp_decide_link_settings(stream,
 				&pipe_ctx->link_config.dp_link_settings))
 			return DC_FAIL_DP_LINK_BANDWIDTH;
+
+		dc->link_srv->dp_decide_tunnel_settings(stream,
+				&pipe_ctx->link_config.dp_tunnel_settings);
+
 		if (dc->link_srv->dp_get_encoding_format(
 				&pipe_ctx->link_config.dp_link_settings) == DP_128b_132b_ENCODING) {
 			pipe_ctx->stream_res.hpo_dp_stream_enc =
@@ -4259,6 +4237,11 @@ enum dc_status dc_validate_with_context(struct dc *dc,
 		}
 	}
 
+	/* clear subvp cursor limitations */
+	for (i = 0; i < context->stream_count; i++) {
+		dc_state_set_stream_subvp_cursor_limit(context->streams[i], context, false);
+	}
+
 	res = dc_validate_global_state(dc, context, fast_validate);
 
 	/* calculate pixel rate divider after deciding pxiel clock & odm combine  */
@@ -4385,8 +4368,7 @@ enum dc_status dc_validate_global_state(
 	result = resource_build_scaling_params_for_context(dc, new_ctx);
 
 	if (result == DC_OK)
-		if (!dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, fast_validate))
-			result = DC_FAIL_BANDWIDTH_VALIDATE;
+		result = dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, fast_validate);
 
 	return result;
 }
@@ -5538,23 +5520,17 @@ enum dc_status update_dp_encoder_resources_for_test_harness(const struct dc *dc,
 	return DC_OK;
 }
 
-bool check_subvp_sw_cursor_fallback_req(const struct dc *dc, struct dc_stream_state *stream)
+struct dscl_prog_data *resource_get_dscl_prog_data(struct pipe_ctx *pipe_ctx)
 {
-	if (!dc->debug.disable_subvp_high_refresh && is_subvp_high_refresh_candidate(stream))
-		return true;
-	if (dc->current_state->stream_count == 1 && stream->timing.v_addressable >= 2880 &&
-			((stream->timing.pix_clk_100hz * 100) / stream->timing.v_total / stream->timing.h_total) < 120)
-		return true;
-	else if (dc->current_state->stream_count > 1 && stream->timing.v_addressable >= 1080 &&
-			((stream->timing.pix_clk_100hz * 100) / stream->timing.v_total / stream->timing.h_total) < 120)
-		return true;
-
-	return false;
+	return &pipe_ctx->plane_res.scl_data.dscl_prog_data;
 }
 
-struct dscl_prog_data *resource_get_dscl_prog_data(struct pipe_ctx *pipe_ctx)
+static bool resource_allocate_mcache(struct dc_state *context, const struct  dc_mcache_params *mcache_params)
 {
-	return &pipe_ctx->plane_res.scl_data.dscl_prog_data;
+	if (context->clk_mgr->ctx->dc->res_pool->funcs->program_mcache_pipe_config)
+		context->clk_mgr->ctx->dc->res_pool->funcs->program_mcache_pipe_config(context, mcache_params);
+
+	return true;
 }
 
 void resource_init_common_dml2_callbacks(struct dc *dc, struct dml2_configuration_options *dml2_options)
@@ -5576,6 +5552,7 @@ void resource_init_common_dml2_callbacks(struct dc *dc, struct dml2_configuratio
 	dml2_options->callbacks.get_stream_status = &dc_state_get_stream_status;
 	dml2_options->callbacks.get_stream_from_id = &dc_state_get_stream_from_id;
 	dml2_options->callbacks.get_max_flickerless_instant_vtotal_increase = &dc_stream_get_max_flickerless_instant_vtotal_increase;
+	dml2_options->callbacks.allocate_mcache = &resource_allocate_mcache;
 
 	dml2_options->svp_pstate.callbacks.dc = dc;
 	dml2_options->svp_pstate.callbacks.add_phantom_plane = &dc_state_add_phantom_plane;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c
index 1b2cce127981..4db7383720fd 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c
@@ -22,6 +22,7 @@
  * Authors: AMD
  *
  */
+#include "dc_types.h"
 #include "core_types.h"
 #include "core_status.h"
 #include "dc_state.h"
@@ -812,8 +813,12 @@ enum dc_status dc_state_add_phantom_stream(const struct dc *dc,
 	if (phantom_stream_status) {
 		phantom_stream_status->mall_stream_config.type = SUBVP_PHANTOM;
 		phantom_stream_status->mall_stream_config.paired_stream = main_stream;
+		phantom_stream_status->mall_stream_config.subvp_limit_cursor_size = false;
+		phantom_stream_status->mall_stream_config.cursor_size_limit_subvp = false;
 	}
 
+	dc_state_set_stream_subvp_cursor_limit(main_stream, state, true);
+
 	return res;
 }
 
@@ -939,13 +944,20 @@ void dc_state_release_phantom_streams_and_planes(
 		const struct dc *dc,
 		struct dc_state *state)
 {
+	unsigned int phantom_count;
+	struct dc_stream_state *phantom_streams[MAX_PHANTOM_PIPES];
+	struct dc_plane_state *phantom_planes[MAX_PHANTOM_PIPES];
 	int i;
 
-	for (i = 0; i < state->phantom_stream_count; i++)
-		dc_state_release_phantom_stream(dc, state, state->phantom_streams[i]);
+	phantom_count = state->phantom_stream_count;
+	memcpy(phantom_streams, state->phantom_streams, sizeof(struct dc_stream_state *) * MAX_PHANTOM_PIPES);
+	for (i = 0; i < phantom_count; i++)
+		dc_state_release_phantom_stream(dc, state, phantom_streams[i]);
 
-	for (i = 0; i < state->phantom_plane_count; i++)
-		dc_state_release_phantom_plane(dc, state, state->phantom_planes[i]);
+	phantom_count = state->phantom_plane_count;
+	memcpy(phantom_planes, state->phantom_planes, sizeof(struct dc_plane_state *) * MAX_PHANTOM_PIPES);
+	for (i = 0; i < phantom_count; i++)
+		dc_state_release_phantom_plane(dc, state, phantom_planes[i]);
 }
 
 struct dc_stream_state *dc_state_get_stream_from_id(const struct dc_state *state, unsigned int id)
@@ -977,3 +989,94 @@ bool dc_state_is_fams2_in_use(
 
 	return is_fams2_in_use;
 }
+
+void dc_state_set_stream_subvp_cursor_limit(const struct dc_stream_state *stream,
+		struct dc_state *state,
+		bool limit)
+{
+	struct dc_stream_status *stream_status;
+
+	stream_status = dc_state_get_stream_status(state, stream);
+
+	if (stream_status) {
+		stream_status->mall_stream_config.subvp_limit_cursor_size = limit;
+	}
+}
+
+bool dc_state_get_stream_subvp_cursor_limit(const struct dc_stream_state *stream,
+		struct dc_state *state)
+{
+	bool limit = false;
+
+	struct dc_stream_status *stream_status;
+
+	stream_status = dc_state_get_stream_status(state, stream);
+
+	if (stream_status) {
+		limit = stream_status->mall_stream_config.subvp_limit_cursor_size;
+	}
+
+	return limit;
+}
+
+void dc_state_set_stream_cursor_subvp_limit(const struct dc_stream_state *stream,
+		struct dc_state *state,
+		bool limit)
+{
+	struct dc_stream_status *stream_status;
+
+	stream_status = dc_state_get_stream_status(state, stream);
+
+	if (stream_status) {
+		stream_status->mall_stream_config.cursor_size_limit_subvp = limit;
+	}
+}
+
+bool dc_state_get_stream_cursor_subvp_limit(const struct dc_stream_state *stream,
+		struct dc_state *state)
+{
+	bool limit = false;
+
+	struct dc_stream_status *stream_status;
+
+	stream_status = dc_state_get_stream_status(state, stream);
+
+	if (stream_status) {
+		limit = stream_status->mall_stream_config.cursor_size_limit_subvp;
+	}
+
+	return limit;
+}
+
+bool dc_state_can_clear_stream_cursor_subvp_limit(const struct dc_stream_state *stream,
+		struct dc_state *state)
+{
+	bool can_clear_limit = false;
+
+	struct dc_stream_status *stream_status;
+
+	stream_status = dc_state_get_stream_status(state, stream);
+
+	if (stream_status) {
+		can_clear_limit = dc_state_get_stream_cursor_subvp_limit(stream, state) &&
+				(stream_status->mall_stream_config.type == SUBVP_PHANTOM ||
+				stream->hw_cursor_req ||
+				!stream_status->mall_stream_config.subvp_limit_cursor_size ||
+				!stream->cursor_position.enable ||
+				dc_stream_check_cursor_attributes(stream, state, &stream->cursor_attributes));
+	}
+
+	return can_clear_limit;
+}
+
+bool dc_state_is_subvp_in_use(struct dc_state *state)
+{
+	uint32_t i;
+
+	for (i = 0; i < state->stream_count; i++) {
+		if (dc_state_get_stream_subvp_type(state, state->streams[i]) != SUBVP_NONE)
+			return true;
+	}
+
+	return false;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 0478dd856d8c..b883fb24fa12 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -265,13 +265,16 @@ void program_cursor_attributes(
 }
 
 /*
- * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address
+ * dc_stream_check_cursor_attributes() - Check validitity of cursor attributes and surface address
  */
-bool dc_stream_set_cursor_attributes(
-	struct dc_stream_state *stream,
+bool dc_stream_check_cursor_attributes(
+	const struct dc_stream_state *stream,
+	struct dc_state *state,
 	const struct dc_cursor_attributes *attributes)
 {
-	struct dc  *dc;
+	const struct dc *dc;
+
+	unsigned int max_cursor_size;
 
 	if (NULL == stream) {
 		dm_error("DC: dc_stream is NULL!\n");
@@ -289,24 +292,38 @@ bool dc_stream_set_cursor_attributes(
 
 	dc = stream->ctx->dc;
 
-	/* SubVP is not compatible with HW cursor larger than 64 x 64 x 4.
-	 * Therefore, if cursor is greater than 64 x 64 x 4, fallback to SW cursor in the following case:
-	 * 1. If the config is a candidate for SubVP high refresh (both single an dual display configs)
-	 * 2. If not subvp high refresh, for single display cases, if resolution is >= 5K and refresh rate < 120hz
-	 * 3. If not subvp high refresh, for multi display cases, if resolution is >= 4K and refresh rate < 120hz
+	/* SubVP is not compatible with HW cursor larger than what can fit in cursor SRAM.
+	 * Therefore, if cursor is greater than this, fallback to SW cursor.
 	 */
-	if (dc->debug.allow_sw_cursor_fallback &&
-		attributes->height * attributes->width * 4 > 16384 &&
-		!stream->hw_cursor_req) {
-		if (check_subvp_sw_cursor_fallback_req(dc, stream))
+	if (dc->debug.allow_sw_cursor_fallback && dc->res_pool->funcs->get_max_hw_cursor_size) {
+		max_cursor_size = dc->res_pool->funcs->get_max_hw_cursor_size(dc, state, stream);
+		max_cursor_size = max_cursor_size * max_cursor_size * 4;
+
+		if (attributes->height * attributes->width * 4 > max_cursor_size) {
 			return false;
+		}
 	}
 
-	stream->cursor_attributes = *attributes;
-
 	return true;
 }
 
+/*
+ * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address
+ */
+bool dc_stream_set_cursor_attributes(
+	struct dc_stream_state *stream,
+	const struct dc_cursor_attributes *attributes)
+{
+	bool result = false;
+
+	if (dc_stream_check_cursor_attributes(stream, stream->ctx->dc->current_state, attributes)) {
+		stream->cursor_attributes = *attributes;
+		result = true;
+	}
+
+	return result;
+}
+
 bool dc_stream_program_cursor_attributes(
 	struct dc_stream_state *stream,
 	const struct dc_cursor_attributes *attributes)
@@ -552,6 +569,14 @@ bool dc_stream_fc_disable_writeback(struct dc *dc,
 	return true;
 }
 
+/**
+ * dc_stream_remove_writeback() - Disables writeback and removes writeback info.
+ * @dc: Display core control structure.
+ * @stream: Display core stream state.
+ * @dwb_pipe_inst: Display writeback pipe.
+ *
+ * Return: returns true on success, false otherwise.
+ */
 bool dc_stream_remove_writeback(struct dc *dc,
 		struct dc_stream_state *stream,
 		uint32_t dwb_pipe_inst)
@@ -1109,3 +1134,26 @@ unsigned int dc_stream_get_max_flickerless_instant_vtotal_increase(struct dc_str
 
 	return dc_stream_get_max_flickerless_instant_vtotal_delta(stream, is_gaming, false);
 }
+
+bool dc_stream_is_cursor_limit_pending(struct dc *dc, struct dc_stream_state *stream)
+{
+	bool is_limit_pending = false;
+
+	if (dc->current_state)
+		is_limit_pending = dc_state_get_stream_cursor_subvp_limit(stream, dc->current_state);
+
+	return is_limit_pending;
+}
+
+bool dc_stream_can_clear_cursor_limit(struct dc *dc, struct dc_stream_state *stream)
+{
+	bool can_clear_limit = false;
+
+	if (dc->current_state)
+		can_clear_limit = dc_state_get_stream_cursor_subvp_limit(stream, dc->current_state) &&
+				(stream->hw_cursor_req ||
+				!stream->cursor_position.enable ||
+				dc_stream_check_cursor_attributes(stream, dc->current_state, &stream->cursor_attributes));
+
+	return can_clear_limit;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
index e6fcc21bb9bc..922f23557f5d 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
@@ -109,7 +109,8 @@ struct dc_plane_state *dc_create_plane_state(const struct dc *dc)
  *****************************************************************************
  */
 const struct dc_plane_status *dc_plane_get_status(
-		const struct dc_plane_state *plane_state)
+		const struct dc_plane_state *plane_state,
+		union dc_plane_status_update_flags flags)
 {
 	const struct dc_plane_status *plane_status;
 	struct dc  *dc;
@@ -136,7 +137,7 @@ const struct dc_plane_status *dc_plane_get_status(
 		if (pipe_ctx->plane_state != plane_state)
 			continue;
 
-		if (pipe_ctx->plane_state)
+		if (pipe_ctx->plane_state && flags.bits.address)
 			pipe_ctx->plane_state->status.is_flip_pending = false;
 
 		break;
@@ -151,7 +152,8 @@ const struct dc_plane_status *dc_plane_get_status(
 		if (pipe_ctx->plane_state != plane_state)
 			continue;
 
-		dc->hwss.update_pending_status(pipe_ctx);
+		if (flags.bits.address)
+			dc->hwss.update_pending_status(pipe_ctx);
 	}
 
 	return plane_status;
@@ -294,3 +296,17 @@ void dc_plane_force_dcc_and_tiling_disable(struct dc_plane_state *plane_state,
 			dc->hwss.clear_surface_dcc_and_tiling(pipe_ctx, plane_state, clear_tiling);
 	}
 }
+
+void dc_plane_copy_config(struct dc_plane_state *dst, const struct dc_plane_state *src)
+{
+	struct kref temp_refcount;
+
+	/* backup persistent info */
+	memcpy(&temp_refcount, &dst->refcount, sizeof(struct kref));
+
+	/* copy all configuration information */
+	memcpy(dst, src, sizeof(struct dc_plane_state));
+
+	/* restore persistent info */
+	memcpy(&dst->refcount, &temp_refcount, sizeof(struct kref));
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 7c2ee0526926..1d917be36fc4 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -53,7 +53,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.325"
+#define DC_VER "3.2.334"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC
@@ -249,6 +249,7 @@ struct dc_caps {
 	uint32_t i2c_speed_in_khz_hdcp;
 	uint32_t dmdata_alloc_size;
 	unsigned int max_cursor_size;
+	unsigned int max_buffered_cursor_size;
 	unsigned int max_video_width;
 	/*
 	 * max video plane width that can be safely assumed to be always
@@ -282,6 +283,7 @@ struct dc_caps {
 	bool edp_dsc_support;
 	bool vbios_lttpr_aware;
 	bool vbios_lttpr_enable;
+	bool fused_io_supported;
 	uint32_t max_otg_num;
 	uint32_t max_cab_allocation_bytes;
 	uint32_t cache_line_size;
@@ -447,6 +449,7 @@ struct dc_config {
 	bool enable_windowed_mpo_odm;
 	bool forceHBR2CP2520; // Used for switching between test patterns TPS4 and CP2520
 	uint32_t allow_edp_hotplug_detection;
+	bool skip_riommu_prefetch_wa;
 	bool clamp_min_dcfclk;
 	uint64_t vblank_alignment_dto_params;
 	uint8_t  vblank_alignment_max_frame_time_diff;
@@ -496,6 +499,7 @@ enum visual_confirm {
 	VISUAL_CONFIRM_HW_CURSOR = 20,
 	VISUAL_CONFIRM_VABC = 21,
 	VISUAL_CONFIRM_DCC = 22,
+	VISUAL_CONFIRM_EXPLICIT = 0x80000000,
 };
 
 enum dc_psr_power_opts {
@@ -902,6 +906,9 @@ struct dc_debug_options {
 	bool voltage_align_fclk;
 	bool disable_min_fclk;
 
+	bool hdcp_lc_force_fw_enable;
+	bool hdcp_lc_enable_sw_fallback;
+
 	bool disable_dfs_bypass;
 	bool disable_dpp_power_gate;
 	bool disable_hubp_power_gate;
@@ -1418,6 +1425,171 @@ struct dc_scratch_space {
 	struct dc_stream_state stream_state;
 };
 
+/*
+ * A link contains one or more sinks and their connected status.
+ * The currently active signal type (HDMI, DP-SST, DP-MST) is also reported.
+ */
+ struct dc_link {
+	struct dc_sink *remote_sinks[MAX_SINKS_PER_LINK];
+	unsigned int sink_count;
+	struct dc_sink *local_sink;
+	unsigned int link_index;
+	enum dc_connection_type type;
+	enum signal_type connector_signal;
+	enum dc_irq_source irq_source_hpd;
+	enum dc_irq_source irq_source_hpd_rx;/* aka DP Short Pulse  */
+	enum dc_irq_source irq_source_read_request;/* Read Request */
+
+	bool is_hpd_filter_disabled;
+	bool dp_ss_off;
+
+	/**
+	 * @link_state_valid:
+	 *
+	 * If there is no link and local sink, this variable should be set to
+	 * false. Otherwise, it should be set to true; usually, the function
+	 * core_link_enable_stream sets this field to true.
+	 */
+	bool link_state_valid;
+	bool aux_access_disabled;
+	bool sync_lt_in_progress;
+	bool skip_stream_reenable;
+	bool is_internal_display;
+	/** @todo Rename. Flag an endpoint as having a programmable mapping to a DIG encoder. */
+	bool is_dig_mapping_flexible;
+	bool hpd_status; /* HPD status of link without physical HPD pin. */
+	bool is_hpd_pending; /* Indicates a new received hpd */
+
+	/* USB4 DPIA links skip verifying link cap, instead performing the fallback method
+	 * for every link training. This is incompatible with DP LL compliance automation,
+	 * which expects the same link settings to be used every retry on a link loss.
+	 * This flag is used to skip the fallback when link loss occurs during automation.
+	 */
+	bool skip_fallback_on_link_loss;
+
+	bool edp_sink_present;
+
+	struct dp_trace dp_trace;
+
+	/* caps is the same as reported_link_cap. link_traing use
+	 * reported_link_cap. Will clean up.  TODO
+	 */
+	struct dc_link_settings reported_link_cap;
+	struct dc_link_settings verified_link_cap;
+	struct dc_link_settings cur_link_settings;
+	struct dc_lane_settings cur_lane_setting[LANE_COUNT_DP_MAX];
+	struct dc_link_settings preferred_link_setting;
+	/* preferred_training_settings are override values that
+	 * come from DM. DM is responsible for the memory
+	 * management of the override pointers.
+	 */
+	struct dc_link_training_overrides preferred_training_settings;
+	struct dp_audio_test_data audio_test_data;
+
+	uint8_t ddc_hw_inst;
+
+	uint8_t hpd_src;
+
+	uint8_t link_enc_hw_inst;
+	/* DIG link encoder ID. Used as index in link encoder resource pool.
+	 * For links with fixed mapping to DIG, this is not changed after dc_link
+	 * object creation.
+	 */
+	enum engine_id eng_id;
+	enum engine_id dpia_preferred_eng_id;
+
+	bool test_pattern_enabled;
+	/* Pending/Current test pattern are only used to perform and track
+	 * FIXED_VS retimer test pattern/lane adjustment override state.
+	 * Pending allows link HWSS to differentiate PHY vs non-PHY pattern,
+	 * to perform specific lane adjust overrides before setting certain
+	 * PHY test patterns. In cases when lane adjust and set test pattern
+	 * calls are not performed atomically (i.e. performing link training),
+	 * pending_test_pattern will be invalid or contain a non-PHY test pattern
+	 * and current_test_pattern will contain required context for any future
+	 * set pattern/set lane adjust to transition between override state(s).
+	 * */
+	enum dp_test_pattern current_test_pattern;
+	enum dp_test_pattern pending_test_pattern;
+
+	union compliance_test_state compliance_test_state;
+
+	void *priv;
+
+	struct ddc_service *ddc;
+
+	enum dp_panel_mode panel_mode;
+	bool aux_mode;
+
+	/* Private to DC core */
+
+	const struct dc *dc;
+
+	struct dc_context *ctx;
+
+	struct panel_cntl *panel_cntl;
+	struct link_encoder *link_enc;
+	struct graphics_object_id link_id;
+	/* Endpoint type distinguishes display endpoints which do not have entries
+	 * in the BIOS connector table from those that do. Helps when tracking link
+	 * encoder to display endpoint assignments.
+	 */
+	enum display_endpoint_type ep_type;
+	union ddi_channel_mapping ddi_channel_mapping;
+	struct connector_device_tag_info device_tag;
+	struct dpcd_caps dpcd_caps;
+	uint32_t dongle_max_pix_clk;
+	unsigned short chip_caps;
+	unsigned int dpcd_sink_count;
+	struct hdcp_caps hdcp_caps;
+	enum edp_revision edp_revision;
+	union dpcd_sink_ext_caps dpcd_sink_ext_caps;
+
+	struct psr_settings psr_settings;
+	struct replay_settings replay_settings;
+
+	/* Drive settings read from integrated info table */
+	struct dc_lane_settings bios_forced_drive_settings;
+
+	/* Vendor specific LTTPR workaround variables */
+	uint8_t vendor_specific_lttpr_link_rate_wa;
+	bool apply_vendor_specific_lttpr_link_rate_wa;
+
+	/* MST record stream using this link */
+	struct link_flags {
+		bool dp_keep_receiver_powered;
+		bool dp_skip_DID2;
+		bool dp_skip_reset_segment;
+		bool dp_skip_fs_144hz;
+		bool dp_mot_reset_segment;
+		/* Some USB4 docks do not handle turning off MST DSC once it has been enabled. */
+		bool dpia_mst_dsc_always_on;
+		/* Forced DPIA into TBT3 compatibility mode. */
+		bool dpia_forced_tbt3_mode;
+		bool dongle_mode_timing_override;
+		bool blank_stream_on_ocs_change;
+		bool read_dpcd204h_on_irq_hpd;
+		bool force_dp_ffe_preset;
+	} wa_flags;
+	union dc_dp_ffe_preset forced_dp_ffe_preset;
+	struct link_mst_stream_allocation_table mst_stream_alloc_table;
+
+	struct dc_link_status link_status;
+	struct dprx_states dprx_states;
+
+	struct gpio *hpd_gpio;
+	enum dc_link_fec_state fec_state;
+	bool link_powered_externally;	// Used to bypass hardware sequencing delays when panel is powered down forcibly
+
+	struct dc_panel_config panel_config;
+	struct phy_state phy_state;
+	uint32_t phy_transition_bitmask;
+	// BW ALLOCATON USB4 ONLY
+	struct dc_dpia_bw_alloc dpia_bw_alloc_config;
+	bool skip_implict_edp_power_control;
+	enum backlight_control_type backlight_control_type;
+};
+
 struct dc {
 	struct dc_debug_options debug;
 	struct dc_versions versions;
@@ -1485,6 +1657,7 @@ struct dc {
 		struct dc_scratch_space current_state;
 		struct dc_scratch_space new_state;
 		struct dc_stream_state temp_stream; // Used so we don't need to allocate stream on the stack
+		struct dc_link temp_link;
 		bool pipes_to_unlock_first[MAX_PIPES]; /* Any of the pipes indicated here should be unlocked first */
 	} scratch;
 
@@ -1651,170 +1824,6 @@ uint32_t dc_bandwidth_in_kbps_from_timing(
 		const enum dc_link_encoding_format link_encoding);
 
 /* Link Interfaces */
-/*
- * A link contains one or more sinks and their connected status.
- * The currently active signal type (HDMI, DP-SST, DP-MST) is also reported.
- */
-struct dc_link {
-	struct dc_sink *remote_sinks[MAX_SINKS_PER_LINK];
-	unsigned int sink_count;
-	struct dc_sink *local_sink;
-	unsigned int link_index;
-	enum dc_connection_type type;
-	enum signal_type connector_signal;
-	enum dc_irq_source irq_source_hpd;
-	enum dc_irq_source irq_source_hpd_rx;/* aka DP Short Pulse  */
-
-	bool is_hpd_filter_disabled;
-	bool dp_ss_off;
-
-	/**
-	 * @link_state_valid:
-	 *
-	 * If there is no link and local sink, this variable should be set to
-	 * false. Otherwise, it should be set to true; usually, the function
-	 * core_link_enable_stream sets this field to true.
-	 */
-	bool link_state_valid;
-	bool aux_access_disabled;
-	bool sync_lt_in_progress;
-	bool skip_stream_reenable;
-	bool is_internal_display;
-	/** @todo Rename. Flag an endpoint as having a programmable mapping to a DIG encoder. */
-	bool is_dig_mapping_flexible;
-	bool hpd_status; /* HPD status of link without physical HPD pin. */
-	bool is_hpd_pending; /* Indicates a new received hpd */
-
-	/* USB4 DPIA links skip verifying link cap, instead performing the fallback method
-	 * for every link training. This is incompatible with DP LL compliance automation,
-	 * which expects the same link settings to be used every retry on a link loss.
-	 * This flag is used to skip the fallback when link loss occurs during automation.
-	 */
-	bool skip_fallback_on_link_loss;
-
-	bool edp_sink_present;
-
-	struct dp_trace dp_trace;
-
-	/* caps is the same as reported_link_cap. link_traing use
-	 * reported_link_cap. Will clean up.  TODO
-	 */
-	struct dc_link_settings reported_link_cap;
-	struct dc_link_settings verified_link_cap;
-	struct dc_link_settings cur_link_settings;
-	struct dc_lane_settings cur_lane_setting[LANE_COUNT_DP_MAX];
-	struct dc_link_settings preferred_link_setting;
-	/* preferred_training_settings are override values that
-	 * come from DM. DM is responsible for the memory
-	 * management of the override pointers.
-	 */
-	struct dc_link_training_overrides preferred_training_settings;
-	struct dp_audio_test_data audio_test_data;
-
-	uint8_t ddc_hw_inst;
-
-	uint8_t hpd_src;
-
-	uint8_t link_enc_hw_inst;
-	/* DIG link encoder ID. Used as index in link encoder resource pool.
-	 * For links with fixed mapping to DIG, this is not changed after dc_link
-	 * object creation.
-	 */
-	enum engine_id eng_id;
-	enum engine_id dpia_preferred_eng_id;
-
-	bool test_pattern_enabled;
-	/* Pending/Current test pattern are only used to perform and track
-	 * FIXED_VS retimer test pattern/lane adjustment override state.
-	 * Pending allows link HWSS to differentiate PHY vs non-PHY pattern,
-	 * to perform specific lane adjust overrides before setting certain
-	 * PHY test patterns. In cases when lane adjust and set test pattern
-	 * calls are not performed atomically (i.e. performing link training),
-	 * pending_test_pattern will be invalid or contain a non-PHY test pattern
-	 * and current_test_pattern will contain required context for any future
-	 * set pattern/set lane adjust to transition between override state(s).
-	 * */
-	enum dp_test_pattern current_test_pattern;
-	enum dp_test_pattern pending_test_pattern;
-
-	union compliance_test_state compliance_test_state;
-
-	void *priv;
-
-	struct ddc_service *ddc;
-
-	enum dp_panel_mode panel_mode;
-	bool aux_mode;
-
-	/* Private to DC core */
-
-	const struct dc *dc;
-
-	struct dc_context *ctx;
-
-	struct panel_cntl *panel_cntl;
-	struct link_encoder *link_enc;
-	struct graphics_object_id link_id;
-	/* Endpoint type distinguishes display endpoints which do not have entries
-	 * in the BIOS connector table from those that do. Helps when tracking link
-	 * encoder to display endpoint assignments.
-	 */
-	enum display_endpoint_type ep_type;
-	union ddi_channel_mapping ddi_channel_mapping;
-	struct connector_device_tag_info device_tag;
-	struct dpcd_caps dpcd_caps;
-	uint32_t dongle_max_pix_clk;
-	unsigned short chip_caps;
-	unsigned int dpcd_sink_count;
-	struct hdcp_caps hdcp_caps;
-	enum edp_revision edp_revision;
-	union dpcd_sink_ext_caps dpcd_sink_ext_caps;
-
-	struct psr_settings psr_settings;
-	struct replay_settings replay_settings;
-
-	/* Drive settings read from integrated info table */
-	struct dc_lane_settings bios_forced_drive_settings;
-
-	/* Vendor specific LTTPR workaround variables */
-	uint8_t vendor_specific_lttpr_link_rate_wa;
-	bool apply_vendor_specific_lttpr_link_rate_wa;
-
-	/* MST record stream using this link */
-	struct link_flags {
-		bool dp_keep_receiver_powered;
-		bool dp_skip_DID2;
-		bool dp_skip_reset_segment;
-		bool dp_skip_fs_144hz;
-		bool dp_mot_reset_segment;
-		/* Some USB4 docks do not handle turning off MST DSC once it has been enabled. */
-		bool dpia_mst_dsc_always_on;
-		/* Forced DPIA into TBT3 compatibility mode. */
-		bool dpia_forced_tbt3_mode;
-		bool dongle_mode_timing_override;
-		bool blank_stream_on_ocs_change;
-		bool read_dpcd204h_on_irq_hpd;
-		bool force_dp_ffe_preset;
-	} wa_flags;
-	union dc_dp_ffe_preset forced_dp_ffe_preset;
-	struct link_mst_stream_allocation_table mst_stream_alloc_table;
-
-	struct dc_link_status link_status;
-	struct dprx_states dprx_states;
-
-	struct gpio *hpd_gpio;
-	enum dc_link_fec_state fec_state;
-	bool link_powered_externally;	// Used to bypass hardware sequencing delays when panel is powered down forcibly
-
-	struct dc_panel_config panel_config;
-	struct phy_state phy_state;
-	uint32_t phy_transition_bitmask;
-	// BW ALLOCATON USB4 ONLY
-	struct dc_dpia_bw_alloc dpia_bw_alloc_config;
-	bool skip_implict_edp_power_control;
-	enum backlight_control_type backlight_control_type;
-};
-
 /* Return an enumerated dc_link.
  * dc_link order is constant and determined at
  * boot time.  They cannot be created or destroyed.
@@ -2589,10 +2598,18 @@ unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context);
 /* DSC Interfaces */
 #include "dc_dsc.h"
 
+void dc_get_visual_confirm_for_stream(
+	struct dc *dc,
+	struct dc_stream_state *stream_state,
+	struct tg_color *color);
+
 /* Disable acc mode Interfaces */
 void dc_disable_accelerated_mode(struct dc *dc);
 
 bool dc_is_timing_changed(struct dc_stream_state *cur_stream,
 		       struct dc_stream_state *new_stream);
 
+bool dc_is_cursor_limit_pending(struct dc *dc);
+bool dc_can_clear_cursor_limit(struct dc *dc);
+
 #endif /* DC_INTERFACE_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 614e03bfd598..afbcf866520e 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -39,6 +39,7 @@
 
 #define CTX dc_dmub_srv->ctx
 #define DC_LOGGER CTX->logger
+#define GPINT_RETRY_NUM 20
 
 static void dc_dmub_srv_construct(struct dc_dmub_srv *dc_srv, struct dc *dc,
 				  struct dmub_srv *dmub)
@@ -70,20 +71,28 @@ void dc_dmub_srv_destroy(struct dc_dmub_srv **dmub_srv)
 	}
 }
 
-void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv)
+bool dc_dmub_srv_wait_for_pending(struct dc_dmub_srv *dc_dmub_srv)
 {
-	struct dmub_srv *dmub = dc_dmub_srv->dmub;
-	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
+	struct dmub_srv *dmub;
+	struct dc_context *dc_ctx;
 	enum dmub_status status;
 
+	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
+		return false;
+
+	dc_ctx = dc_dmub_srv->ctx;
+	dmub = dc_dmub_srv->dmub;
+
 	do {
-		status = dmub_srv_wait_for_idle(dmub, 100000);
+		status = dmub_srv_wait_for_pending(dmub, 100000);
 	} while (dc_dmub_srv->ctx->dc->debug.disable_timeout && status != DMUB_STATUS_OK);
 
 	if (status != DMUB_STATUS_OK) {
 		DC_ERROR("Error waiting for DMUB idle: status=%d\n", status);
 		dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
 	}
+
+	return status == DMUB_STATUS_OK;
 }
 
 void dc_dmub_srv_clear_inbox0_ack(struct dc_dmub_srv *dc_dmub_srv)
@@ -126,7 +135,49 @@ void dc_dmub_srv_send_inbox0_cmd(struct dc_dmub_srv *dc_dmub_srv,
 	}
 }
 
-bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
+static bool dc_dmub_srv_reg_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
+		unsigned int count,
+		union dmub_rb_cmd *cmd_list)
+{
+	struct dc_context *dc_ctx;
+	struct dmub_srv *dmub;
+	enum dmub_status status = DMUB_STATUS_OK;
+	int i;
+
+	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
+		return false;
+
+	dc_ctx = dc_dmub_srv->ctx;
+	dmub = dc_dmub_srv->dmub;
+
+	for (i = 0 ; i < count; i++) {
+		/* confirm no messages pending */
+		do {
+			status = dmub_srv_wait_for_idle(dmub, 100000);
+		} while (dc_dmub_srv->ctx->dc->debug.disable_timeout && status != DMUB_STATUS_OK);
+
+		/* queue command */
+		if (status == DMUB_STATUS_OK)
+			status = dmub_srv_reg_cmd_execute(dmub, &cmd_list[i]);
+
+		/* check for errors */
+		if (status != DMUB_STATUS_OK) {
+			break;
+		}
+	}
+
+	if (status != DMUB_STATUS_OK) {
+		if (status != DMUB_STATUS_POWER_STATE_D3) {
+			DC_ERROR("Error starting DMUB execution: status=%d\n", status);
+			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
+		}
+		return false;
+	}
+
+	return true;
+}
+
+static bool dc_dmub_srv_fb_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
 		unsigned int count,
 		union dmub_rb_cmd *cmd_list)
 {
@@ -143,20 +194,25 @@ bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
 
 	for (i = 0 ; i < count; i++) {
 		// Queue command
-		status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
+		if (!cmd_list[i].cmd_common.header.multi_cmd_pending ||
+				dmub_rb_num_free(&dmub->inbox1.rb) >= count - i) {
+			status = dmub_srv_fb_cmd_queue(dmub, &cmd_list[i]);
+		} else {
+			status = DMUB_STATUS_QUEUE_FULL;
+		}
 
 		if (status == DMUB_STATUS_QUEUE_FULL) {
 			/* Execute and wait for queue to become empty again. */
-			status = dmub_srv_cmd_execute(dmub);
+			status = dmub_srv_fb_cmd_execute(dmub);
 			if (status == DMUB_STATUS_POWER_STATE_D3)
 				return false;
 
 			do {
-				status = dmub_srv_wait_for_idle(dmub, 100000);
+					status = dmub_srv_wait_for_inbox_free(dmub, 100000, count - i);
 			} while (dc_dmub_srv->ctx->dc->debug.disable_timeout && status != DMUB_STATUS_OK);
 
 			/* Requeue the command. */
-			status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
+			status = dmub_srv_fb_cmd_queue(dmub, &cmd_list[i]);
 		}
 
 		if (status != DMUB_STATUS_OK) {
@@ -168,7 +224,7 @@ bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
 		}
 	}
 
-	status = dmub_srv_cmd_execute(dmub);
+	status = dmub_srv_fb_cmd_execute(dmub);
 	if (status != DMUB_STATUS_OK) {
 		if (status != DMUB_STATUS_POWER_STATE_D3) {
 			DC_ERROR("Error starting DMUB execution: status=%d\n", status);
@@ -180,6 +236,26 @@ bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
 	return true;
 }
 
+bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
+		unsigned int count,
+		union dmub_rb_cmd *cmd_list)
+{
+	bool res = false;
+
+	if (dc_dmub_srv && dc_dmub_srv->dmub) {
+		if (dc_dmub_srv->dmub->inbox_type == DMUB_CMD_INTERFACE_REG) {
+			res = dc_dmub_srv_reg_cmd_list_queue_execute(dc_dmub_srv, count, cmd_list);
+		} else {
+			res = dc_dmub_srv_fb_cmd_list_queue_execute(dc_dmub_srv, count, cmd_list);
+		}
+
+		if (res)
+			res = dmub_srv_update_inbox_status(dc_dmub_srv->dmub) == DMUB_STATUS_OK;
+	}
+
+	return res;
+}
+
 bool dc_dmub_srv_wait_for_idle(struct dc_dmub_srv *dc_dmub_srv,
 		enum dm_dmub_wait_type wait_type,
 		union dmub_rb_cmd *cmd_list)
@@ -202,7 +278,8 @@ bool dc_dmub_srv_wait_for_idle(struct dc_dmub_srv *dc_dmub_srv,
 			DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status);
 			if (!dmub->debug.timeout_info.timeout_occured) {
 				dmub->debug.timeout_info.timeout_occured = true;
-				dmub->debug.timeout_info.timeout_cmd = *cmd_list;
+				if (cmd_list)
+					dmub->debug.timeout_info.timeout_cmd = *cmd_list;
 				dmub->debug.timeout_info.timestamp = dm_get_timestamp(dc_dmub_srv->ctx);
 			}
 			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
@@ -210,8 +287,9 @@ bool dc_dmub_srv_wait_for_idle(struct dc_dmub_srv *dc_dmub_srv,
 		}
 
 		// Copy data back from ring buffer into command
-		if (wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)
-			dmub_rb_get_return_data(&dmub->inbox1_rb, cmd_list);
+		if (wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY && cmd_list) {
+			dmub_srv_cmd_get_response(dc_dmub_srv->dmub, cmd_list);
+		}
 	}
 
 	return true;
@@ -224,74 +302,10 @@ bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd
 
 bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type)
 {
-	struct dc_context *dc_ctx;
-	struct dmub_srv *dmub;
-	enum dmub_status status;
-	int i;
-
-	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
-		return false;
-
-	dc_ctx = dc_dmub_srv->ctx;
-	dmub = dc_dmub_srv->dmub;
-
-	for (i = 0 ; i < count; i++) {
-		// Queue command
-		status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
-
-		if (status == DMUB_STATUS_QUEUE_FULL) {
-			/* Execute and wait for queue to become empty again. */
-			status = dmub_srv_cmd_execute(dmub);
-			if (status == DMUB_STATUS_POWER_STATE_D3)
-				return false;
-
-			status = dmub_srv_wait_for_idle(dmub, 100000);
-			if (status != DMUB_STATUS_OK)
-				return false;
-
-			/* Requeue the command. */
-			status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
-		}
-
-		if (status != DMUB_STATUS_OK) {
-			if (status != DMUB_STATUS_POWER_STATE_D3) {
-				DC_ERROR("Error queueing DMUB command: status=%d\n", status);
-				dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
-			}
-			return false;
-		}
-	}
-
-	status = dmub_srv_cmd_execute(dmub);
-	if (status != DMUB_STATUS_OK) {
-		if (status != DMUB_STATUS_POWER_STATE_D3) {
-			DC_ERROR("Error starting DMUB execution: status=%d\n", status);
-			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
-		}
+	if (!dc_dmub_srv_cmd_list_queue_execute(dc_dmub_srv, count, cmd_list))
 		return false;
-	}
 
-	// Wait for DMUB to process command
-	if (wait_type != DM_DMUB_WAIT_TYPE_NO_WAIT) {
-		if (dc_dmub_srv->ctx->dc->debug.disable_timeout) {
-			do {
-				status = dmub_srv_wait_for_idle(dmub, 100000);
-			} while (status != DMUB_STATUS_OK);
-		} else
-			status = dmub_srv_wait_for_idle(dmub, 100000);
-
-		if (status != DMUB_STATUS_OK) {
-			DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status);
-			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
-			return false;
-		}
-
-		// Copy data back from ring buffer into command
-		if (wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)
-			dmub_rb_get_return_data(&dmub->inbox1_rb, cmd_list);
-	}
-
-	return true;
+	return dc_dmub_srv_wait_for_idle(dc_dmub_srv, wait_type, cmd_list);
 }
 
 bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv)
@@ -1243,7 +1257,7 @@ static void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle)
 			ips_fw->signals.bits.ips1_commit,
 			ips_fw->signals.bits.ips2_commit);
 
-		dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+		dc_dmub_srv_wait_for_idle(dc->ctx->dmub_srv, DM_DMUB_WAIT_TYPE_WAIT, NULL);
 
 		memset(&new_signals, 0, sizeof(new_signals));
 
@@ -1355,14 +1369,15 @@ static void dc_dmub_srv_exit_low_power_state(const struct dc *dc)
 			if (!dc->debug.optimize_ips_handshake || !ips_fw->signals.bits.ips2_commit)
 				udelay(dc->debug.ips2_eval_delay_us);
 
-			if (ips_fw->signals.bits.ips2_commit) {
-				DC_LOG_IPS(
-					"exit IPS2 #1 (ips1_commit=%u ips2_commit=%u)",
-					ips_fw->signals.bits.ips1_commit,
-					ips_fw->signals.bits.ips2_commit);
+			DC_LOG_IPS(
+				"exit IPS2 #1 (ips1_commit=%u ips2_commit=%u)",
+				ips_fw->signals.bits.ips1_commit,
+				ips_fw->signals.bits.ips2_commit);
 
-				// Tell PMFW to exit low power state
-				dc->clk_mgr->funcs->exit_low_power_state(dc->clk_mgr);
+			// Tell PMFW to exit low power state
+			dc->clk_mgr->funcs->exit_low_power_state(dc->clk_mgr);
+
+			if (ips_fw->signals.bits.ips2_commit) {
 
 				DC_LOG_IPS(
 					"wait IPS2 entry delay (ips1_commit=%u ips2_commit=%u)",
@@ -1400,7 +1415,7 @@ static void dc_dmub_srv_exit_low_power_state(const struct dc *dc)
 					ips_fw->signals.bits.ips1_commit,
 					ips_fw->signals.bits.ips2_commit);
 
-				dmub_srv_sync_inbox1(dc->ctx->dmub_srv->dmub);
+				dmub_srv_sync_inboxes(dc->ctx->dmub_srv->dmub);
 			}
 		}
 
@@ -1654,7 +1669,8 @@ void dc_dmub_srv_fams2_update_config(struct dc *dc,
 	/* fill in generic command header */
 	global_cmd->header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
 	global_cmd->header.sub_type = DMUB_CMD__FAMS2_CONFIG;
-	global_cmd->header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
+	global_cmd->header.payload_bytes =
+			sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
 
 	if (enable) {
 		/* send global configuration parameters */
@@ -1673,11 +1689,13 @@ void dc_dmub_srv_fams2_update_config(struct dc *dc,
 			/* configure command header */
 			stream_base_cmd->header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
 			stream_base_cmd->header.sub_type = DMUB_CMD__FAMS2_CONFIG;
-			stream_base_cmd->header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
+			stream_base_cmd->header.payload_bytes =
+					sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
 			stream_base_cmd->header.multi_cmd_pending = 1;
 			stream_sub_state_cmd->header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
 			stream_sub_state_cmd->header.sub_type = DMUB_CMD__FAMS2_CONFIG;
-			stream_sub_state_cmd->header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
+			stream_sub_state_cmd->header.payload_bytes =
+					sizeof(struct dmub_rb_cmd_fams2) - sizeof(struct dmub_cmd_header);
 			stream_sub_state_cmd->header.multi_cmd_pending = 1;
 			/* copy stream static base state */
 			memcpy(&stream_base_cmd->config,
@@ -1723,7 +1741,8 @@ void dc_dmub_srv_fams2_drr_update(struct dc *dc,
 	cmd.fams2_drr_update.dmub_optc_state_req.v_total_mid_frame_num = vtotal_mid_frame_num;
 	cmd.fams2_drr_update.dmub_optc_state_req.program_manual_trigger = program_manual_trigger;
 
-	cmd.fams2_drr_update.header.payload_bytes = sizeof(cmd.fams2_drr_update) - sizeof(cmd.fams2_drr_update.header);
+	cmd.fams2_drr_update.header.payload_bytes =
+			sizeof(cmd.fams2_drr_update) - sizeof(cmd.fams2_drr_update.header);
 
 	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
@@ -1759,7 +1778,8 @@ void dc_dmub_srv_fams2_passthrough_flip(
 		/* build command header */
 		cmds[num_cmds].fams2_flip.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH;
 		cmds[num_cmds].fams2_flip.header.sub_type = DMUB_CMD__FAMS2_FLIP;
-		cmds[num_cmds].fams2_flip.header.payload_bytes = sizeof(struct dmub_rb_cmd_fams2_flip);
+		cmds[num_cmds].fams2_flip.header.payload_bytes =
+				sizeof(struct dmub_rb_cmd_fams2_flip) - sizeof(struct dmub_cmd_header);
 
 		/* for chaining multiple commands, all but last command should set to 1 */
 		cmds[num_cmds].fams2_flip.header.multi_cmd_pending = 1;
@@ -1869,11 +1889,14 @@ void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struc
 	if (command_code == DMUB_GPINT__INVALID_COMMAND)
 		return;
 
-	// send gpint commands and wait for ack
-	if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_PERCENT,
-				      (uint16_t)(output->ips_mode),
-				       &output->residency_percent, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY))
-		output->residency_percent = 0;
+	for (i = 0; i < GPINT_RETRY_NUM; i++) {
+		// false could mean GPINT timeout, in which case we should retry
+		if (dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_PERCENT,
+					      (uint16_t)(output->ips_mode), &output->residency_percent,
+					      DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY))
+			break;
+		udelay(100);
+	}
 
 	if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_ENTRY_COUNTER,
 				      (uint16_t)(output->ips_mode),
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
index a636f4c3f01d..ada5c2fb2db3 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
@@ -58,7 +58,7 @@ struct dc_dmub_srv {
 	bool needs_idle_wake;
 };
 
-void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv);
+bool dc_dmub_srv_wait_for_pending(struct dc_dmub_srv *dc_dmub_srv);
 
 bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index 77c87ad57220..0bad8304ccf6 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -159,6 +159,11 @@ struct dc_link_settings {
 	uint8_t link_rate_set;
 };
 
+struct dc_tunnel_settings {
+	bool should_enable_dp_tunneling;
+	bool should_use_dp_bw_allocation;
+};
+
 union dc_dp_ffe_preset {
 	struct {
 		uint8_t level		: 4;
@@ -943,10 +948,20 @@ union dpia_info {
 	uint8_t raw;
 };
 
+/* DPCD[0xE0020] USB4_DRIVER_BW_CAPABILITY register. */
+union usb4_driver_bw_cap {
+	struct {
+		uint8_t rsvd :7;
+		uint8_t driver_bw_alloc_support :1;
+	} bits;
+	uint8_t raw;
+};
+
 /* DP Tunneling over USB4 */
 struct dpcd_usb4_dp_tunneling_info {
 	union dp_tun_cap_support dp_tun_cap;
 	union dpia_info dpia_info;
+	union usb4_driver_bw_cap driver_bw_cap;
 	uint8_t usb4_driver_id;
 	uint8_t usb4_topology_id[DPCD_USB4_TOPOLOGY_ID_LEN];
 };
@@ -1486,5 +1501,11 @@ struct dp_trace {
 # ifndef DP_TUNNELING_BW_ALLOC_CAP_CHANGED
 # define DP_TUNNELING_BW_ALLOC_CAP_CHANGED		(1 << 3)
 # endif
+# ifndef DPTX_BW_ALLOC_UNMASK_IRQ
+# define DPTX_BW_ALLOC_UNMASK_IRQ			(1 << 6)
+# endif
+# ifndef DPTX_BW_ALLOC_MODE_ENABLE
+# define DPTX_BW_ALLOC_MODE_ENABLE			(1 << 7)
+# endif
 
 #endif /* DC_DP_TYPES_H */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_fused_io.c b/drivers/gpu/drm/amd/display/dc/dc_fused_io.c
new file mode 100644
index 000000000000..fee69642fb93
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dc_fused_io.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#include "dc_fused_io.h"
+
+#include "dm_helpers.h"
+#include "gpio.h"
+
+static bool op_i2c_convert(
+		union dmub_rb_cmd *cmd,
+		const struct mod_hdcp_atomic_op_i2c *op,
+		enum dmub_cmd_fused_request_type type,
+		uint32_t ddc_line,
+		bool over_aux
+)
+{
+	struct dmub_cmd_fused_request *req = &cmd->fused_io.request;
+	struct dmub_cmd_fused_request_location_i2c *loc = &req->u.i2c;
+
+	if (!op || op->size > sizeof(req->buffer))
+		return false;
+
+	req->type = type;
+	loc->is_aux = false;
+	loc->ddc_line = ddc_line;
+	loc->over_aux = over_aux;
+	loc->address = op->address;
+	loc->offset = op->offset;
+	loc->length = op->size;
+	memcpy(req->buffer, op->data, op->size);
+
+	return true;
+}
+
+static bool op_aux_convert(
+		union dmub_rb_cmd *cmd,
+		const struct mod_hdcp_atomic_op_aux *op,
+		enum dmub_cmd_fused_request_type type,
+		uint32_t ddc_line
+)
+{
+	struct dmub_cmd_fused_request *req = &cmd->fused_io.request;
+	struct dmub_cmd_fused_request_location_aux *loc = &req->u.aux;
+
+	if (!op || op->size > sizeof(req->buffer))
+		return false;
+
+	req->type = type;
+	loc->is_aux = true;
+	loc->ddc_line = ddc_line;
+	loc->address = op->address;
+	loc->length = op->size;
+	memcpy(req->buffer, op->data, op->size);
+
+	return true;
+}
+
+static bool atomic_write_poll_read(
+		struct dc_link *link,
+		union dmub_rb_cmd commands[3],
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+)
+{
+	const uint8_t count = 3;
+	const uint32_t timeout_per_request_us = 10000;
+	const uint32_t timeout_per_aux_transaction_us = 10000;
+	uint64_t timeout_us = 0;
+
+	commands[1].fused_io.request.poll_mask_msb = poll_mask_msb;
+	commands[1].fused_io.request.timeout_us = poll_timeout_us;
+
+	for (uint8_t i = 0; i < count; i++) {
+		struct dmub_rb_cmd_fused_io *io = &commands[i].fused_io;
+
+		io->header.type = DMUB_CMD__FUSED_IO;
+		io->header.sub_type = DMUB_CMD__FUSED_IO_EXECUTE;
+		io->header.multi_cmd_pending = i != count - 1;
+		io->header.payload_bytes = sizeof(commands[i].fused_io) - sizeof(io->header);
+
+		timeout_us += timeout_per_request_us + io->request.timeout_us;
+		if (!io->request.timeout_us && io->request.u.aux.is_aux)
+			timeout_us += timeout_per_aux_transaction_us * (io->request.u.aux.length / 16);
+	}
+
+	if (!dm_helpers_execute_fused_io(link->ctx, link, commands, count, timeout_us))
+		return false;
+
+	return commands[0].fused_io.request.status == FUSED_REQUEST_STATUS_SUCCESS;
+}
+
+bool dm_atomic_write_poll_read_i2c(
+		struct dc_link *link,
+		const struct mod_hdcp_atomic_op_i2c *write,
+		const struct mod_hdcp_atomic_op_i2c *poll,
+		struct mod_hdcp_atomic_op_i2c *read,
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+)
+{
+	if (!link)
+		return false;
+
+	const bool over_aux = false;
+	const uint32_t ddc_line = link->ddc->ddc_pin->pin_data->en;
+
+	union dmub_rb_cmd commands[3] = { 0 };
+	const bool converted = op_i2c_convert(&commands[0], write, FUSED_REQUEST_WRITE, ddc_line, over_aux)
+			&& op_i2c_convert(&commands[1], poll, FUSED_REQUEST_POLL, ddc_line, over_aux)
+			&& op_i2c_convert(&commands[2], read, FUSED_REQUEST_READ, ddc_line, over_aux);
+
+	if (!converted)
+		return false;
+
+	const bool result = atomic_write_poll_read(link, commands, poll_timeout_us, poll_mask_msb);
+
+	memcpy(read->data, commands[0].fused_io.request.buffer, read->size);
+	return result;
+}
+
+bool dm_atomic_write_poll_read_aux(
+		struct dc_link *link,
+		const struct mod_hdcp_atomic_op_aux *write,
+		const struct mod_hdcp_atomic_op_aux *poll,
+		struct mod_hdcp_atomic_op_aux *read,
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+)
+{
+	if (!link)
+		return false;
+
+	const uint32_t ddc_line = link->ddc->ddc_pin->pin_data->en;
+	union dmub_rb_cmd commands[3] = { 0 };
+	const bool converted = op_aux_convert(&commands[0], write, FUSED_REQUEST_WRITE, ddc_line)
+			&& op_aux_convert(&commands[1], poll, FUSED_REQUEST_POLL, ddc_line)
+			&& op_aux_convert(&commands[2], read, FUSED_REQUEST_READ, ddc_line);
+
+	if (!converted)
+		return false;
+
+	const bool result = atomic_write_poll_read(link, commands, poll_timeout_us, poll_mask_msb);
+
+	memcpy(read->data, commands[0].fused_io.request.buffer, read->size);
+	return result;
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/dc_fused_io.h b/drivers/gpu/drm/amd/display/dc/dc_fused_io.h
new file mode 100644
index 000000000000..c74917240985
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dc_fused_io.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ */
+
+#ifndef __DC_FUSED_IO_H__
+#define __DC_FUSED_IO_H__
+
+#include "dc.h"
+#include "mod_hdcp.h"
+
+bool dm_atomic_write_poll_read_i2c(
+		struct dc_link *link,
+		const struct mod_hdcp_atomic_op_i2c *write,
+		const struct mod_hdcp_atomic_op_i2c *poll,
+		struct mod_hdcp_atomic_op_i2c *read,
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+);
+
+bool dm_atomic_write_poll_read_aux(
+		struct dc_link *link,
+		const struct mod_hdcp_atomic_op_aux *write,
+		const struct mod_hdcp_atomic_op_aux *poll,
+		struct mod_hdcp_atomic_op_aux *read,
+		uint32_t poll_timeout_us,
+		uint8_t poll_mask_msb
+);
+
+#endif  // __DC_FUSED_IO_H__
+
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index 8f077e15b4f0..7217de258851 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -682,13 +682,19 @@ void reg_sequence_wait_done(const struct dc_context *ctx)
 	if (offload &&
 	    ctx->dc->debug.dmub_offload_enabled &&
 	    !ctx->dc->debug.dmcub_emulation) {
-		dc_dmub_srv_wait_idle(ctx->dmub_srv);
+		dc_dmub_srv_wait_for_idle(ctx->dmub_srv, DM_DMUB_WAIT_TYPE_WAIT, NULL);
 	}
 }
 
 char *dce_version_to_string(const int version)
 {
 	switch (version) {
+	case DCE_VERSION_6_0:
+		return "DCE 6.0";
+	case DCE_VERSION_6_1:
+		return "DCE 6.1";
+	case DCE_VERSION_6_4:
+		return "DCE 6.4";
 	case DCE_VERSION_8_0:
 		return "DCE 8.0";
 	case DCE_VERSION_8_1:
diff --git a/drivers/gpu/drm/amd/display/dc/dc_plane.h b/drivers/gpu/drm/amd/display/dc/dc_plane.h
index e9413685ed4f..14feb843e694 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_plane.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_plane.h
@@ -28,13 +28,24 @@
 
 #include "dc_hw_types.h"
 
+union dc_plane_status_update_flags {
+	struct {
+		uint32_t address : 1;
+	} bits;
+	uint32_t raw;
+};
+
 struct dc_plane_state *dc_create_plane_state(const struct dc *dc);
 const struct dc_plane_status *dc_plane_get_status(
-		const struct dc_plane_state *plane_state);
+		const struct dc_plane_state *plane_state,
+		union dc_plane_status_update_flags flags);
 void dc_plane_state_retain(struct dc_plane_state *plane_state);
 void dc_plane_state_release(struct dc_plane_state *plane_state);
 
 void dc_plane_force_dcc_and_tiling_disable(struct dc_plane_state *plane_state,
 					   bool clear_tiling);
 
+
+void dc_plane_copy_config(struct dc_plane_state *dst, const struct dc_plane_state *src);
+
 #endif /* _DC_PLANE_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_state_priv.h b/drivers/gpu/drm/amd/display/dc/dc_state_priv.h
index 1a12ef579ff4..1d9bae56ff6a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_state_priv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_state_priv.h
@@ -105,4 +105,24 @@ bool dc_state_is_fams2_in_use(
 		const struct dc *dc,
 		const struct dc_state *state);
 
+
+void dc_state_set_stream_subvp_cursor_limit(const struct dc_stream_state *stream,
+		struct dc_state *state,
+		bool limit);
+
+bool dc_state_get_stream_subvp_cursor_limit(const struct dc_stream_state *stream,
+		struct dc_state *state);
+
+void dc_state_set_stream_cursor_subvp_limit(const struct dc_stream_state *stream,
+		struct dc_state *state,
+		bool limit);
+
+bool dc_state_get_stream_cursor_subvp_limit(const struct dc_stream_state *stream,
+		struct dc_state *state);
+
+bool dc_state_can_clear_stream_cursor_subvp_limit(const struct dc_stream_state *stream,
+		struct dc_state *state);
+
+bool dc_state_is_subvp_in_use(struct dc_state *state);
+
 #endif /* _DC_STATE_PRIV_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index e0bfddaa23e3..341d2ffb64b1 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -44,6 +44,8 @@ struct mall_stream_config {
 	 */
 	enum mall_stream_type type;
 	struct dc_stream_state *paired_stream;	// master / slave stream
+	bool subvp_limit_cursor_size; /* stream has/is using subvp limiting hw cursor support */
+	bool cursor_size_limit_subvp; /* stream is using hw cursor config preventing subvp */
 };
 
 struct dc_stream_status {
@@ -503,6 +505,11 @@ void program_cursor_position(
 	struct dc *dc,
 	struct dc_stream_state *stream);
 
+bool dc_stream_check_cursor_attributes(
+	const struct dc_stream_state *stream,
+	struct dc_state *state,
+	const struct dc_cursor_attributes *attributes);
+
 bool dc_stream_set_cursor_attributes(
 	struct dc_stream_state *stream,
 	const struct dc_cursor_attributes *attributes);
@@ -579,4 +586,8 @@ void dc_dmub_update_dirty_rect(struct dc *dc,
 			       struct dc_stream_state *stream,
 			       struct dc_surface_update *srf_updates,
 			       struct dc_state *context);
+
+bool dc_stream_is_cursor_limit_pending(struct dc *dc, struct dc_stream_state *stream);
+bool dc_stream_can_clear_cursor_limit(struct dc *dc, struct dc_stream_state *stream);
+
 #endif /* DC_STREAM_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 83ffaae9f439..a4cd0eb39a3a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -210,6 +210,7 @@ struct dc_edid_caps {
 
 	bool edid_hdmi;
 	bool hdr_supported;
+	bool rr_capable;
 
 	struct dc_panel_patch panel_patch;
 };
@@ -1089,7 +1090,8 @@ union replay_low_refresh_rate_enable_options {
 	struct {
 	//BIT[0-3]: Replay Low Hz Support control
 		unsigned int ENABLE_LOW_RR_SUPPORT          :1;
-		unsigned int RESERVED_1_3                   :3;
+		unsigned int SKIP_ASIC_CHECK                :1;
+		unsigned int RESERVED_2_3                   :2;
 	//BIT[4-15]: Replay Low Hz Enable Scenarios
 		unsigned int ENABLE_STATIC_SCREEN           :1;
 		unsigned int ENABLE_FULL_SCREEN_VIDEO       :1;
@@ -1129,6 +1131,10 @@ struct replay_config {
 	union replay_low_refresh_rate_enable_options low_rr_enable_options;
 	/* Replay coasting vtotal is within low refresh rate range. */
 	bool low_rr_activated;
+	/* Replay low refresh rate supported*/
+	bool low_rr_supported;
+	/* Replay Video Conferencing Optimization Enabled */
+	bool replay_video_conferencing_optimization_enabled;
 };
 
 /* Replay feature flags*/
@@ -1249,6 +1255,7 @@ enum dc_cm2_gpu_mem_layout {
 
 enum dc_cm2_gpu_mem_pixel_component_order {
 	DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA,
+	DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_BGRA
 };
 
 enum dc_cm2_gpu_mem_format {
@@ -1270,7 +1277,8 @@ struct dc_cm2_gpu_mem_format_parameters {
 
 enum dc_cm2_gpu_mem_size {
 	DC_CM2_GPU_MEM_SIZE_171717,
-	DC_CM2_GPU_MEM_SIZE_TRANSFORMED
+	DC_CM2_GPU_MEM_SIZE_333333,
+	DC_CM2_GPU_MEM_SIZE_TRANSFORMED,
 };
 
 struct dc_cm2_gpu_mem_parameters {
@@ -1279,6 +1287,7 @@ struct dc_cm2_gpu_mem_parameters {
 	struct dc_cm2_gpu_mem_format_parameters format_params;
 	enum dc_cm2_gpu_mem_pixel_component_order component_order;
 	enum dc_cm2_gpu_mem_size  size;
+	uint16_t bit_depth;
 };
 
 enum dc_cm2_transfer_func_source {
@@ -1302,6 +1311,10 @@ struct dc_cm2_func_luts {
 			const struct dc_3dlut *lut3d_func;
 			struct dc_cm2_gpu_mem_parameters gpu_mem_params;
 		};
+		bool rmcm_3dlut_shaper_select;
+		bool mpc_3dlut_enable;
+		bool rmcm_3dlut_enable;
+		bool mpc_mcm_post_blend;
 	} lut3d_data;
 	const struct dc_transfer_func *lut1d_func;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
index b363f5360818..58c84f555c0f 100644
--- a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
@@ -391,6 +391,7 @@ static void dccg35_set_dppclk_rcg(struct dccg *dccg,
 
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+
 	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dpp && enable)
 		return;
 
@@ -411,6 +412,8 @@ static void dccg35_set_dppclk_rcg(struct dccg *dccg,
 	BREAK_TO_DEBUGGER();
 		break;
 	}
+	//DC_LOG_DEBUG("%s: inst(%d) DPPCLK rcg_disable: %d\n", __func__, inst, enable ? 0 : 1);
+
 }
 
 static void dccg35_set_dpstreamclk_rcg(
@@ -1035,6 +1038,7 @@ static void dccg35_enable_dpp_clk_new(
 			  DPPCLK0_DTO_MODULO, 0xFF);
 }
 
+
 static void dccg35_disable_dpp_clk_new(
 	struct dccg *dccg,
 	int inst)
@@ -1112,30 +1116,24 @@ static void dcn35_set_dppclk_enable(struct dccg *dccg,
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+
 	switch (dpp_inst) {
 	case 0:
 		REG_UPDATE(DPPCLK_CTRL, DPPCLK0_EN, enable);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK0_ROOT_GATE_DISABLE, enable);
 		break;
 	case 1:
 		REG_UPDATE(DPPCLK_CTRL, DPPCLK1_EN, enable);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK1_ROOT_GATE_DISABLE, enable);
 		break;
 	case 2:
 		REG_UPDATE(DPPCLK_CTRL, DPPCLK2_EN, enable);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK2_ROOT_GATE_DISABLE, enable);
 		break;
 	case 3:
 		REG_UPDATE(DPPCLK_CTRL, DPPCLK3_EN, enable);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK3_ROOT_GATE_DISABLE, enable);
 		break;
 	default:
 		break;
 	}
+	//DC_LOG_DEBUG("%s: dpp_inst(%d) DPPCLK_EN = %d\n", __func__, dpp_inst, enable);
 
 }
 
@@ -1163,14 +1161,18 @@ static void dccg35_update_dpp_dto(struct dccg *dccg, int dpp_inst,
 			ASSERT(false);
 			phase = 0xff;
 		}
+		dccg35_set_dppclk_rcg(dccg, dpp_inst, false);
 
 		REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0,
 				DPPCLK0_DTO_PHASE, phase,
 				DPPCLK0_DTO_MODULO, modulo);
 
 		dcn35_set_dppclk_enable(dccg, dpp_inst, true);
-	} else
+	} else {
 		dcn35_set_dppclk_enable(dccg, dpp_inst, false);
+		/*we have this in hwss: disable_plane*/
+		//dccg35_set_dppclk_rcg(dccg, dpp_inst, true);
+	}
 	dccg->pipe_dppclk_khz[dpp_inst] = req_dppclk;
 }
 
@@ -1182,6 +1184,7 @@ static void dccg35_set_dppclk_root_clock_gating(struct dccg *dccg,
 	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
 		return;
 
+
 	switch (dpp_inst) {
 	case 0:
 		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK0_ROOT_GATE_DISABLE, enable);
@@ -1198,6 +1201,8 @@ static void dccg35_set_dppclk_root_clock_gating(struct dccg *dccg,
 	default:
 		break;
 	}
+	//DC_LOG_DEBUG("%s: dpp_inst(%d) rcg: %d\n", __func__, dpp_inst, enable);
+
 }
 
 static void dccg35_get_pixel_rate_div(
@@ -1521,28 +1526,30 @@ static void dccg35_set_physymclk_root_clock_gating(
 	switch (phy_inst) {
 	case 0:
 		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-				PHYASYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+				PHYASYMCLK_ROOT_GATE_DISABLE, enable ? 0 : 1);
 		break;
 	case 1:
 		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-				PHYBSYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+				PHYBSYMCLK_ROOT_GATE_DISABLE, enable ? 0 : 1);
 		break;
 	case 2:
 		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-				PHYCSYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+				PHYCSYMCLK_ROOT_GATE_DISABLE, enable ? 0 : 1);
 		break;
 	case 3:
 		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-				PHYDSYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+				PHYDSYMCLK_ROOT_GATE_DISABLE, enable ? 0 : 1);
 		break;
 	case 4:
 		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-				PHYESYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+				PHYESYMCLK_ROOT_GATE_DISABLE, enable ? 0 : 1);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	//DC_LOG_DEBUG("%s: dpp_inst(%d) PHYESYMCLK_ROOT_GATE_DISABLE:\n", __func__, phy_inst, enable ? 0 : 1);
+
 }
 
 static void dccg35_set_physymclk(
@@ -1643,6 +1650,8 @@ static void dccg35_dpp_root_clock_control(
 		return;
 
 	if (clock_on) {
+		dccg35_set_dppclk_rcg(dccg, dpp_inst, false);
+
 		/* turn off the DTO and leave phase/modulo at max */
 		dcn35_set_dppclk_enable(dccg, dpp_inst, 1);
 		REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0,
@@ -1654,6 +1663,8 @@ static void dccg35_dpp_root_clock_control(
 		REG_SET_2(DPPCLK_DTO_PARAM[dpp_inst], 0,
 			  DPPCLK0_DTO_PHASE, 0,
 			  DPPCLK0_DTO_MODULO, 1);
+		/*we have this in hwss: disable_plane*/
+		//dccg35_set_dppclk_rcg(dccg, dpp_inst, true);
 	}
 
 	dccg->dpp_clock_gated[dpp_inst] = !clock_on;
@@ -1771,36 +1782,40 @@ static void dccg35_enable_dscclk(struct dccg *dccg, int inst)
 	//Disable DTO
 	switch (inst) {
 	case 0:
+		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
+			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK0_ROOT_GATE_DISABLE, 1);
+
 		REG_UPDATE_2(DSCCLK0_DTO_PARAM,
 				DSCCLK0_DTO_PHASE, 0,
 				DSCCLK0_DTO_MODULO, 0);
 		REG_UPDATE(DSCCLK_DTO_CTRL,	DSCCLK0_EN, 1);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK0_ROOT_GATE_DISABLE, 1);
 		break;
 	case 1:
+		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
+			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK1_ROOT_GATE_DISABLE, 1);
+
 		REG_UPDATE_2(DSCCLK1_DTO_PARAM,
 				DSCCLK1_DTO_PHASE, 0,
 				DSCCLK1_DTO_MODULO, 0);
 		REG_UPDATE(DSCCLK_DTO_CTRL, DSCCLK1_EN, 1);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK1_ROOT_GATE_DISABLE, 1);
 		break;
 	case 2:
+		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
+			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK2_ROOT_GATE_DISABLE, 1);
+
 		REG_UPDATE_2(DSCCLK2_DTO_PARAM,
 				DSCCLK2_DTO_PHASE, 0,
 				DSCCLK2_DTO_MODULO, 0);
 		REG_UPDATE(DSCCLK_DTO_CTRL, DSCCLK2_EN, 1);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK2_ROOT_GATE_DISABLE, 1);
 		break;
 	case 3:
+		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
+			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK3_ROOT_GATE_DISABLE, 1);
+
 		REG_UPDATE_2(DSCCLK3_DTO_PARAM,
 				DSCCLK3_DTO_PHASE, 0,
 				DSCCLK3_DTO_MODULO, 0);
 		REG_UPDATE(DSCCLK_DTO_CTRL, DSCCLK3_EN, 1);
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK3_ROOT_GATE_DISABLE, 1);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
@@ -1813,9 +1828,6 @@ static void dccg35_disable_dscclk(struct dccg *dccg,
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
-	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-		return;
-
 	switch (inst) {
 	case 0:
 		REG_UPDATE(DSCCLK_DTO_CTRL, DSCCLK0_EN, 0);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
index 077337698e0a..b4f5b4a6331a 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
@@ -976,11 +976,12 @@ static bool dcn31_program_pix_clk(
 	struct bp_pixel_clock_parameters bp_pc_params = {0};
 	enum transmitter_color_depth bp_pc_colour_depth = TRANSMITTER_COLOR_DEPTH_24;
 
-	// Apply ssed(spread spectrum) dpref clock for edp only.
-	if (clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz != 0
-		&& pix_clk_params->signal_type == SIGNAL_TYPE_EDP
-		&& encoding == DP_8b_10b_ENCODING)
+	// Apply ssed(spread spectrum) dpref clock for edp and dp
+	if (clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz != 0 &&
+		dc_is_dp_signal(pix_clk_params->signal_type) &&
+		encoding == DP_8b_10b_ENCODING)
 		dp_dto_ref_khz = clock_source->ctx->dc->clk_mgr->dp_dto_source_clock_in_khz;
+
 	// For these signal types Driver to program DP_DTO without calling VBIOS Command table
 	if (dc_is_dp_signal(pix_clk_params->signal_type) || dc_is_virtual_signal(pix_clk_params->signal_type)) {
 		if (e) {
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h
index 0721ae895ae9..94128f7a18b1 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.h
@@ -257,7 +257,7 @@ bool dce110_clk_src_construct(
 	struct dce110_clk_src *clk_src,
 	struct dc_context *ctx,
 	struct dc_bios *bios,
-	enum clock_source_id,
+	enum clock_source_id id,
 	const struct dce110_clk_src_regs *regs,
 	const struct dce110_clk_src_shift *cs_shift,
 	const struct dce110_clk_src_mask *cs_mask);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
index ccc154b0281c..3b9011ef9b68 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
@@ -28,6 +28,8 @@
 #include "dc.h"
 #include "core_types.h"
 #include "dmub_cmd.h"
+#include "dc_dmub_srv.h"
+#include "dmub/dmub_srv.h"
 
 #define TO_DMUB_ABM(abm)\
 	container_of(abm, struct dce_abm, base)
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
index 0d7e7f3b81a1..a641ae04450c 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
@@ -240,7 +240,8 @@ bool dmub_abm_save_restore(
 	cmd.abm_save_restore.abm_init_config_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
 	cmd.abm_save_restore.abm_init_config_data.panel_mask = panel_mask;
 
-	cmd.abm_save_restore.header.payload_bytes = sizeof(struct dmub_rb_cmd_abm_save_restore);
+	cmd.abm_save_restore.header.payload_bytes =
+			sizeof(struct dmub_rb_cmd_abm_save_restore) - sizeof(struct dmub_cmd_header);
 
 	dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
index c31e4f26a305..fcd3d86ad517 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
@@ -280,7 +280,9 @@ static void dmub_replay_set_power_opt_and_coasting_vtotal(struct dmub_replay *dm
 	memset(&cmd, 0, sizeof(cmd));
 	pCmd->header.type = DMUB_CMD__REPLAY;
 	pCmd->header.sub_type = DMUB_CMD__REPLAY_SET_POWER_OPT_AND_COASTING_VTOTAL;
-	pCmd->header.payload_bytes = sizeof(struct dmub_rb_cmd_replay_set_power_opt_and_coasting_vtotal);
+	pCmd->header.payload_bytes =
+			sizeof(struct dmub_rb_cmd_replay_set_power_opt_and_coasting_vtotal) -
+			sizeof(struct dmub_cmd_header);
 	pCmd->replay_set_power_opt_data.power_opt = power_opt;
 	pCmd->replay_set_power_opt_data.panel_inst = panel_inst;
 	pCmd->replay_set_coasting_vtotal_data.coasting_vtotal = (coasting_vtotal & 0xFFFF);
@@ -319,7 +321,8 @@ static void dmub_replay_send_cmd(struct dmub_replay *dmub,
 		cmd.replay_set_timing_sync.header.sub_type =
 			DMUB_CMD__REPLAY_SET_TIMING_SYNC_SUPPORTED;
 		cmd.replay_set_timing_sync.header.payload_bytes =
-			sizeof(struct dmub_rb_cmd_replay_set_timing_sync);
+			sizeof(struct dmub_rb_cmd_replay_set_timing_sync) -
+			sizeof(struct dmub_cmd_header);
 		//Cmd Body
 		cmd.replay_set_timing_sync.replay_set_timing_sync_data.panel_inst =
 						cmd_element->sync_data.panel_inst;
@@ -331,7 +334,8 @@ static void dmub_replay_send_cmd(struct dmub_replay *dmub,
 		cmd.replay_set_frameupdate_timer.header.sub_type =
 			DMUB_CMD__REPLAY_SET_RESIDENCY_FRAMEUPDATE_TIMER;
 		cmd.replay_set_frameupdate_timer.header.payload_bytes =
-			sizeof(struct dmub_rb_cmd_replay_set_frameupdate_timer);
+			sizeof(struct dmub_rb_cmd_replay_set_frameupdate_timer) -
+			sizeof(struct dmub_cmd_header);
 		//Cmd Body
 		cmd.replay_set_frameupdate_timer.data.panel_inst =
 						cmd_element->panel_inst;
@@ -345,7 +349,8 @@ static void dmub_replay_send_cmd(struct dmub_replay *dmub,
 		cmd.replay_set_pseudo_vtotal.header.sub_type =
 			DMUB_CMD__REPLAY_SET_PSEUDO_VTOTAL;
 		cmd.replay_set_pseudo_vtotal.header.payload_bytes =
-			sizeof(struct dmub_rb_cmd_replay_set_pseudo_vtotal);
+			sizeof(struct dmub_rb_cmd_replay_set_pseudo_vtotal) -
+			sizeof(struct dmub_cmd_header);
 		//Cmd Body
 		cmd.replay_set_pseudo_vtotal.data.panel_inst =
 			cmd_element->pseudo_vtotal_data.panel_inst;
@@ -357,7 +362,8 @@ static void dmub_replay_send_cmd(struct dmub_replay *dmub,
 		cmd.replay_disabled_adaptive_sync_sdp.header.sub_type =
 			DMUB_CMD__REPLAY_DISABLED_ADAPTIVE_SYNC_SDP;
 		cmd.replay_disabled_adaptive_sync_sdp.header.payload_bytes =
-			sizeof(struct dmub_rb_cmd_replay_disabled_adaptive_sync_sdp);
+			sizeof(struct dmub_rb_cmd_replay_disabled_adaptive_sync_sdp) -
+			sizeof(struct dmub_cmd_header);
 		//Cmd Body
 		cmd.replay_disabled_adaptive_sync_sdp.data.panel_inst =
 			cmd_element->disabled_adaptive_sync_sdp_data.panel_inst;
@@ -369,7 +375,8 @@ static void dmub_replay_send_cmd(struct dmub_replay *dmub,
 		cmd.replay_set_general_cmd.header.sub_type =
 			DMUB_CMD__REPLAY_SET_GENERAL_CMD;
 		cmd.replay_set_general_cmd.header.payload_bytes =
-			sizeof(struct dmub_rb_cmd_replay_set_general_cmd);
+			sizeof(struct dmub_rb_cmd_replay_set_general_cmd) -
+			sizeof(struct dmub_cmd_header);
 		//Cmd Body
 		cmd.replay_set_general_cmd.data.panel_inst =
 			cmd_element->set_general_cmd_data.panel_inst;
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/Makefile b/drivers/gpu/drm/amd/display/dc/dce60/Makefile
index eede83ad91fa..824f73eb3326 100644
--- a/drivers/gpu/drm/amd/display/dc/dce60/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dce60/Makefile
@@ -25,8 +25,7 @@
 
 CFLAGS_$(AMDDALPATH)/dc/dce60/dce60_resource.o = -Wno-override-init
 
-DCE60 = dce60_timing_generator.o dce60_hw_sequencer.o \
-	dce60_resource.o
+DCE60 = dce60_timing_generator.o
 
 AMD_DAL_DCE60 = $(addprefix $(AMDDALPATH)/dc/dce60/,$(DCE60))
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c
index 003a9330c286..88e7a1fc9a30 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_timing_generator.c
@@ -105,7 +105,7 @@ static void program_pix_dur(struct timing_generator *tg, uint32_t pix_clk_100hz)
 	dm_write_reg(tg->ctx, addr, value);
 }
 
-static void program_timing(struct timing_generator *tg,
+static void dce80_timing_generator_program_timing(struct timing_generator *tg,
 	const struct dc_crtc_timing *timing,
 	int vready_offset,
 	int vstartup_start,
@@ -185,7 +185,7 @@ static void dce80_timing_generator_enable_advanced_request(
 
 static const struct timing_generator_funcs dce80_tg_funcs = {
 		.validate_timing = dce110_tg_validate_timing,
-		.program_timing = program_timing,
+		.program_timing = dce80_timing_generator_program_timing,
 		.enable_crtc = dce110_timing_generator_enable_crtc,
 		.disable_crtc = dce110_timing_generator_disable_crtc,
 		.is_counter_moving = dce110_timing_generator_is_counter_moving,
diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h
index 5efddd48d5c5..9d160b39e8c5 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h
@@ -153,6 +153,14 @@ bool dm_helpers_submit_i2c(
 		const struct dc_link *link,
 		struct i2c_command *cmd);
 
+bool dm_helpers_execute_fused_io(
+		struct dc_context *ctx,
+		struct dc_link *link,
+		union dmub_rb_cmd *commands,
+		uint8_t count,
+		uint32_t timeout_us
+);
+
 bool dm_helpers_dp_write_dsc_enable(
 		struct dc_context *ctx,
 		const struct dc_stream_state *stream,
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
index f1fe49401bc0..8d24763938ea 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
@@ -1002,6 +1002,7 @@ static bool CalculatePrefetchSchedule(
 
 	dst_y_prefetch_equ = VStartup - (Tsetup + dml_max(TWait + TCalc, *Tdmdl)) / LineTime
 			- (*DSTYAfterScaler + *DSTXAfterScaler / myPipe->HTotal);
+	dst_y_prefetch_equ = dml_min(dst_y_prefetch_equ, 63.75); // limit to the reg limit of U6.2 for DST_Y_PREFETCH
 
 	Lsw_oto = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC);
 	Tsw_oto = Lsw_oto * LineTime;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
index f567a9023682..ed59c77bc6f6 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
@@ -1105,6 +1105,7 @@ static bool CalculatePrefetchSchedule(
 	Tr0_oto_lines = dml_ceil(4.0 * Tr0_oto / LineTime, 1) / 4.0;
 	dst_y_prefetch_oto = Tvm_oto_lines + 2 * Tr0_oto_lines + Lsw_oto;
 	dst_y_prefetch_equ =  VStartup - (*TSetup + dml_max(TWait + TCalc, *Tdmdl)) / LineTime - (*DSTYAfterScaler + *DSTXAfterScaler / myPipe->HTotal);
+	dst_y_prefetch_equ = dml_min(dst_y_prefetch_equ, 63.75); // limit to the reg limit of U6.2 for DST_Y_PREFETCH
 	dst_y_prefetch_equ = dml_floor(4.0 * (dst_y_prefetch_equ + 0.125), 1) / 4.0;
 	Tpre_rounded = dst_y_prefetch_equ * LineTime;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
index 5865e8fa2d8e..9f3938a50240 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
@@ -1123,6 +1123,7 @@ static bool CalculatePrefetchSchedule(
 	Tr0_oto_lines = dml_ceil(4.0 * Tr0_oto / LineTime, 1) / 4.0;
 	dst_y_prefetch_oto = Tvm_oto_lines + 2 * Tr0_oto_lines + Lsw_oto;
 	dst_y_prefetch_equ =  VStartup - (*TSetup + dml_max(TWait + TCalc, *Tdmdl)) / LineTime - (*DSTYAfterScaler + *DSTXAfterScaler / myPipe->HTotal);
+	dst_y_prefetch_equ = dml_min(dst_y_prefetch_equ, 63.75); // limit to the reg limit of U6.2 for DST_Y_PREFETCH
 	dst_y_prefetch_equ = dml_floor(4.0 * (dst_y_prefetch_equ + 0.125), 1) / 4.0;
 	Tpre_rounded = dst_y_prefetch_equ * LineTime;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 56dda686e299..b0fc1fd20208 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -627,6 +627,7 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 		 */
 		if (pipe->plane_state && !pipe->top_pipe && !pipe->prev_odm_pipe && !dcn32_is_center_timing(pipe) &&
 				!pipe->stream->hw_cursor_req &&
+				!dc_state_get_stream_cursor_subvp_limit(pipe->stream, context) &&
 				!(pipe->stream->timing.pix_clk_100hz / 10000 > DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ) &&
 				(!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) &&
 				dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_NONE &&
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/Makefile b/drivers/gpu/drm/amd/display/dc/dml2/Makefile
index 21fd466dba26..157ecf008d6c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dml2/Makefile
@@ -99,7 +99,6 @@ CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_standalone_libraries/lib_float_math.
 CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml21_wrapper.o := $(dml2_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_translation_helper.o := $(dml2_ccflags)
 CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_ccflags)
-CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/inc/dml2_debug.o := $(dml2_ccflags)
 
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_rcflags)
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_rcflags)
@@ -117,11 +116,9 @@ CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_standalone_libraries/lib_floa
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml21_wrapper.o := $(dml2_rcflags)
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/dml21_translation_helper.o := $(dml2_rcflags)
 CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_rcflags)
-CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/inc/dml2_debug.o := $(dml2_rcflags)
 
 DML21 := src/dml2_top/dml2_top_interfaces.o
 DML21 += src/dml2_top/dml2_top_soc15.o
-DML21 += src/inc/dml2_debug.o
 DML21 += src/dml2_core/dml2_core_dcn4.o
 DML21 += src/dml2_core/dml2_core_factory.o
 DML21 += src/dml2_core/dml2_core_dcn4_calcs.o
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
index 0c8ec30ea672..d47cacfdb695 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
@@ -526,7 +526,8 @@ static void populate_dml21_output_config_from_stream_state(struct dml2_link_outp
 
 static void populate_dml21_stream_overrides_from_stream_state(
 		struct dml2_stream_parameters *stream_desc,
-		struct dc_stream_state *stream)
+		struct dc_stream_state *stream,
+		struct dc_stream_status *stream_status)
 {
 	switch (stream->debug.force_odm_combine_segments) {
 	case 0:
@@ -551,7 +552,9 @@ static void populate_dml21_stream_overrides_from_stream_state(
 	if (!stream->ctx->dc->debug.enable_single_display_2to1_odm_policy ||
 			stream->debug.force_odm_combine_segments > 0)
 		stream_desc->overrides.disable_dynamic_odm = true;
-	stream_desc->overrides.disable_subvp = stream->ctx->dc->debug.force_disable_subvp || stream->hw_cursor_req;
+	stream_desc->overrides.disable_subvp = stream->ctx->dc->debug.force_disable_subvp ||
+			stream->hw_cursor_req ||
+			stream_status->mall_stream_config.cursor_size_limit_subvp;
 }
 
 static enum dml2_swizzle_mode gfx_addr3_to_dml2_swizzle_mode(enum swizzle_mode_addr3_values addr3_mode)
@@ -885,6 +888,9 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm
 		case DC_CM2_GPU_MEM_SIZE_171717:
 			plane->tdlut.tdlut_width_mode = dml2_tdlut_width_17_cube;
 			break;
+		case DC_CM2_GPU_MEM_SIZE_333333:
+			plane->tdlut.tdlut_width_mode = dml2_tdlut_width_33_cube;
+			break;
 		case DC_CM2_GPU_MEM_SIZE_TRANSFORMED:
 			//plane->tdlut.tdlut_width_mode = dml2_tdlut_width_flatten; // dml2_tdlut_width_flatten undefined
 			break;
@@ -910,7 +916,7 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm
 }
 
 //TODO : Could be possibly moved to a common helper layer.
-static bool dml21_wrapper_get_plane_id(const struct dc_state *context, const struct dc_plane_state *plane, unsigned int *plane_id)
+static bool dml21_wrapper_get_plane_id(const struct dc_state *context, unsigned int stream_id, const struct dc_plane_state *plane, unsigned int *plane_id)
 {
 	int i, j;
 
@@ -918,10 +924,12 @@ static bool dml21_wrapper_get_plane_id(const struct dc_state *context, const str
 		return false;
 
 	for (i = 0; i < context->stream_count; i++) {
-		for (j = 0; j < context->stream_status[i].plane_count; j++) {
-			if (context->stream_status[i].plane_states[j] == plane) {
-				*plane_id = (i << 16) | j;
-				return true;
+		if (context->streams[i]->stream_id == stream_id) {
+			for (j = 0; j < context->stream_status[i].plane_count; j++) {
+				if (context->stream_status[i].plane_states[j] == plane) {
+					*plane_id = (i << 16) | j;
+					return true;
+				}
 			}
 		}
 	}
@@ -944,14 +952,14 @@ static unsigned int map_stream_to_dml21_display_cfg(const struct dml2_context *d
 	return location;
 }
 
-static unsigned int map_plane_to_dml21_display_cfg(const struct dml2_context *dml_ctx,
+unsigned int map_plane_to_dml21_display_cfg(const struct dml2_context *dml_ctx, unsigned int stream_id,
 		const struct dc_plane_state *plane, const struct dc_state *context)
 {
 	unsigned int plane_id;
 	int i = 0;
 	int location = -1;
 
-	if (!dml21_wrapper_get_plane_id(context, plane, &plane_id)) {
+	if (!dml21_wrapper_get_plane_id(context, stream_id, plane, &plane_id)) {
 		ASSERT(false);
 		return -1;
 	}
@@ -1021,7 +1029,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s
 		populate_dml21_timing_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, context->streams[stream_index], dml_ctx);
 		adjust_dml21_hblank_timing_config_from_pipe_ctx(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, &context->res_ctx.pipe_ctx[stream_index]);
 		populate_dml21_output_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].output, context->streams[stream_index], &context->res_ctx.pipe_ctx[stream_index]);
-		populate_dml21_stream_overrides_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location], context->streams[stream_index]);
+		populate_dml21_stream_overrides_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location], context->streams[stream_index], &context->stream_status[stream_index]);
 
 		dml_dispcfg->stream_descriptors[disp_cfg_stream_location].overrides.hw.twait_budgeting.fclk_pstate = dml2_twait_budgeting_setting_if_needed;
 		dml_dispcfg->stream_descriptors[disp_cfg_stream_location].overrides.hw.twait_budgeting.uclk_pstate = dml2_twait_budgeting_setting_if_needed;
@@ -1037,7 +1045,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s
 			dml_dispcfg->plane_descriptors[disp_cfg_plane_location].stream_index = disp_cfg_stream_location;
 		} else {
 			for (plane_index = 0; plane_index < context->stream_status[stream_index].plane_count; plane_index++) {
-				disp_cfg_plane_location = map_plane_to_dml21_display_cfg(dml_ctx, context->stream_status[stream_index].plane_states[plane_index], context);
+				disp_cfg_plane_location = map_plane_to_dml21_display_cfg(dml_ctx, context->streams[stream_index]->stream_id, context->stream_status[stream_index].plane_states[plane_index], context);
 
 				if (disp_cfg_plane_location < 0)
 					disp_cfg_plane_location = dml_dispcfg->num_planes++;
@@ -1048,7 +1056,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s
 				populate_dml21_plane_config_from_plane_state(dml_ctx, &dml_dispcfg->plane_descriptors[disp_cfg_plane_location], context->stream_status[stream_index].plane_states[plane_index], context, stream_index);
 				dml_dispcfg->plane_descriptors[disp_cfg_plane_location].stream_index = disp_cfg_stream_location;
 
-				if (dml21_wrapper_get_plane_id(context, context->stream_status[stream_index].plane_states[plane_index], &dml_ctx->v21.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id[disp_cfg_plane_location]))
+				if (dml21_wrapper_get_plane_id(context, context->streams[stream_index]->stream_id, context->stream_status[stream_index].plane_states[plane_index], &dml_ctx->v21.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id[disp_cfg_plane_location]))
 					dml_ctx->v21.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id_valid[disp_cfg_plane_location] = true;
 
 				/* apply forced pstate policy */
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h
index 069b939c672a..73a013be1e48 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h
@@ -11,6 +11,7 @@ struct dc_state;
 struct dcn_watermarks;
 union dcn_watermark_set;
 struct pipe_ctx;
+struct dc_plane_state;
 
 struct dml2_context;
 struct dml2_configuration_options;
@@ -25,4 +26,5 @@ void dml21_extract_watermark_sets(const struct dc *in_dc, union dcn_watermark_se
 void dml21_map_hw_resources(struct dml2_context *dml_ctx);
 void dml21_get_pipe_mcache_config(struct dc_state *context, struct pipe_ctx *pipe_ctx, struct dml2_per_plane_programming *pln_prog, struct dml2_pipe_configuration_descriptor *mcache_pipe_config);
 void dml21_set_dc_p_state_type(struct pipe_ctx *pipe_ctx, struct dml2_per_stream_programming *stream_programming, bool sub_vp_enabled);
+unsigned int map_plane_to_dml21_display_cfg(const struct dml2_context *dml_ctx, unsigned int stream_id, const struct dc_plane_state *plane, const struct dc_state *context);
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
index be54f0e696ce..208d3651b6ba 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
@@ -2,6 +2,7 @@
 //
 // Copyright 2024 Advanced Micro Devices, Inc.
 
+#include <linux/vmalloc.h>
 
 #include "dml2_internal_types.h"
 #include "dml_top.h"
@@ -11,13 +12,15 @@
 #include "dml21_translation_helper.h"
 #include "dml2_dc_resource_mgmt.h"
 
+#define INVALID -1
+
 static bool dml21_allocate_memory(struct dml2_context **dml_ctx)
 {
-	*dml_ctx = kzalloc(sizeof(struct dml2_context), GFP_KERNEL);
+	*dml_ctx = vzalloc(sizeof(struct dml2_context));
 	if (!(*dml_ctx))
 		return false;
 
-	(*dml_ctx)->v21.dml_init.dml2_instance = kzalloc(sizeof(struct dml2_instance), GFP_KERNEL);
+	(*dml_ctx)->v21.dml_init.dml2_instance = vzalloc(sizeof(struct dml2_instance));
 	if (!((*dml_ctx)->v21.dml_init.dml2_instance))
 		return false;
 
@@ -27,7 +30,7 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx)
 	(*dml_ctx)->v21.mode_support.display_config = &(*dml_ctx)->v21.display_config;
 	(*dml_ctx)->v21.mode_programming.display_config = (*dml_ctx)->v21.mode_support.display_config;
 
-	(*dml_ctx)->v21.mode_programming.programming = kzalloc(sizeof(struct dml2_display_cfg_programming), GFP_KERNEL);
+	(*dml_ctx)->v21.mode_programming.programming = vzalloc(sizeof(struct dml2_display_cfg_programming));
 	if (!((*dml_ctx)->v21.mode_programming.programming))
 		return false;
 
@@ -86,6 +89,8 @@ static void dml21_init(const struct dc *in_dc, struct dml2_context **dml_ctx, co
 	/* Store configuration options */
 	(*dml_ctx)->config = *config;
 
+	DC_FP_START();
+
 	/*Initialize SOCBB and DCNIP params */
 	dml21_initialize_soc_bb_params(&(*dml_ctx)->v21.dml_init, config, in_dc);
 	dml21_initialize_ip_params(&(*dml_ctx)->v21.dml_init, config, in_dc);
@@ -96,6 +101,8 @@ static void dml21_init(const struct dc *in_dc, struct dml2_context **dml_ctx, co
 
 	/*Initialize DML21 instance */
 	dml2_initialize_instance(&(*dml_ctx)->v21.dml_init);
+
+	DC_FP_END();
 }
 
 bool dml21_create(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config)
@@ -111,8 +118,8 @@ bool dml21_create(const struct dc *in_dc, struct dml2_context **dml_ctx, const s
 
 void dml21_destroy(struct dml2_context *dml2)
 {
-	kfree(dml2->v21.dml_init.dml2_instance);
-	kfree(dml2->v21.mode_programming.programming);
+	vfree(dml2->v21.dml_init.dml2_instance);
+	vfree(dml2->v21.mode_programming.programming);
 }
 
 static void dml21_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_state *context, struct resource_context *out_new_hw_state,
@@ -203,10 +210,40 @@ static void dml21_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_sta
 	}
 }
 
+static void dml21_prepare_mcache_params(struct dml2_context *dml_ctx, struct dc_state *context, struct dc_mcache_params *mcache_params)
+{
+	int dc_plane_idx = 0;
+	int dml_prog_idx, stream_idx, plane_idx;
+	struct dml2_per_plane_programming *pln_prog = NULL;
+
+	for (stream_idx = 0; stream_idx < context->stream_count; stream_idx++) {
+		for (plane_idx = 0; plane_idx < context->stream_status[stream_idx].plane_count; plane_idx++) {
+			dml_prog_idx = map_plane_to_dml21_display_cfg(dml_ctx, context->streams[stream_idx]->stream_id, context->stream_status[stream_idx].plane_states[plane_idx], context);
+			if (dml_prog_idx == INVALID) {
+				continue;
+			}
+			pln_prog = &dml_ctx->v21.mode_programming.programming->plane_programming[dml_prog_idx];
+			mcache_params[dc_plane_idx].valid = pln_prog->mcache_allocation.valid;
+			mcache_params[dc_plane_idx].num_mcaches_plane0 = pln_prog->mcache_allocation.num_mcaches_plane0;
+			mcache_params[dc_plane_idx].num_mcaches_plane1 = pln_prog->mcache_allocation.num_mcaches_plane1;
+			mcache_params[dc_plane_idx].requires_dedicated_mall_mcache = pln_prog->mcache_allocation.requires_dedicated_mall_mcache;
+			mcache_params[dc_plane_idx].last_slice_sharing.plane0_plane1 = pln_prog->mcache_allocation.last_slice_sharing.plane0_plane1;
+			memcpy(mcache_params[dc_plane_idx].mcache_x_offsets_plane0,
+				pln_prog->mcache_allocation.mcache_x_offsets_plane0,
+				sizeof(int) * (DML2_MAX_MCACHES + 1));
+			memcpy(mcache_params[dc_plane_idx].mcache_x_offsets_plane1,
+				pln_prog->mcache_allocation.mcache_x_offsets_plane1,
+				sizeof(int) * (DML2_MAX_MCACHES + 1));
+			dc_plane_idx++;
+		}
+	}
+}
+
 static bool dml21_mode_check_and_programming(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx)
 {
 	bool result = false;
 	struct dml2_build_mode_programming_in_out *mode_programming = &dml_ctx->v21.mode_programming;
+	struct dc_mcache_params mcache_params[MAX_PLANES] = {0};
 
 	memset(&dml_ctx->v21.display_config, 0, sizeof(struct dml2_display_cfg));
 	memset(&dml_ctx->v21.dml_to_dc_pipe_mapping, 0, sizeof(struct dml2_dml_to_dc_pipe_mapping));
@@ -229,7 +266,9 @@ static bool dml21_mode_check_and_programming(const struct dc *in_dc, struct dc_s
 	if (!result)
 		return false;
 
+	DC_FP_START();
 	result = dml2_build_mode_programming(mode_programming);
+	DC_FP_END();
 	if (!result)
 		return false;
 
@@ -239,6 +278,14 @@ static bool dml21_mode_check_and_programming(const struct dc *in_dc, struct dc_s
 		dml2_map_dc_pipes(dml_ctx, context, NULL, &dml_ctx->v21.dml_to_dc_pipe_mapping, in_dc->current_state);
 		/* if subvp phantoms are present, expand them into dc context */
 		dml21_handle_phantom_streams_planes(in_dc, context, dml_ctx);
+
+		if (in_dc->res_pool->funcs->program_mcache_pipe_config) {
+			//Prepare mcache params for each plane based on mcache output from DML
+			dml21_prepare_mcache_params(dml_ctx, context, mcache_params);
+
+			//populate mcache regs to each pipe
+			dml_ctx->config.callbacks.allocate_mcache(context, mcache_params);
+		}
 	}
 
 	/* Copy DML CLK, WM and REG outputs to bandwidth context */
@@ -272,7 +319,9 @@ static bool dml21_check_mode_support(const struct dc *in_dc, struct dc_state *co
 	mode_support->dml2_instance = dml_init->dml2_instance;
 	dml21_map_dc_state_into_dml_display_cfg(in_dc, context, dml_ctx);
 	dml_ctx->v21.mode_programming.dml2_instance->scratch.build_mode_programming_locals.mode_programming_params.programming = dml_ctx->v21.mode_programming.programming;
+	DC_FP_START();
 	is_supported = dml2_check_mode_supported(mode_support);
+	DC_FP_END();
 	if (!is_supported)
 		return false;
 
@@ -284,10 +333,11 @@ bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml
 	bool out = false;
 
 	/* Use dml_validate_only for fast_validate path */
-	if (fast_validate) {
+	if (fast_validate)
 		out = dml21_check_mode_support(in_dc, context, dml_ctx);
-	} else
+	else
 		out = dml21_mode_check_and_programming(in_dc, context, dml_ctx);
+
 	return out;
 }
 
@@ -426,8 +476,12 @@ void dml21_copy(struct dml2_context *dst_dml_ctx,
 
 	dst_dml_ctx->v21.mode_programming.programming = dst_dml2_programming;
 
+	DC_FP_START();
+
 	/* need to initialize copied instance for internal references to be correct */
 	dml2_initialize_instance(&dst_dml_ctx->v21.dml_init);
+
+	DC_FP_END();
 }
 
 bool dml21_create_copy(struct dml2_context **dst_dml_ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h
index b2075b8c363b..42e715024bc9 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h
@@ -8,6 +8,7 @@
 
 #include "os_types.h"
 #include "dml_top_soc_parameter_types.h"
+#include "dml_top_display_cfg_types.h"
 
 struct dc;
 struct dc_state;
@@ -65,4 +66,67 @@ struct socbb_ip_params_external {
 	struct dml2_ip_capabilities ip_params;
 	struct dml2_soc_bb soc_bb;
 };
+
+/*mcache parameters decided by dml*/
+struct dc_mcache_params {
+	bool valid;
+	/*
+	* For iMALL, dedicated mall mcaches are required (sharing of last
+	* slice possible), for legacy phantom or phantom without return
+	* the only mall mcaches need to be valid.
+	*/
+	bool requires_dedicated_mall_mcache;
+	unsigned int num_mcaches_plane0;
+	unsigned int num_mcaches_plane1;
+	/*
+	* Generally, plane0/1 slices must use a disjoint set of caches
+	* but in some cases the final segement of the two planes can
+	* use the same cache. If plane0_plane1 is set, then this is
+	* allowed.
+	*
+	* Similarly, the caches allocated to MALL prefetcher are generally
+	* disjoint, but if mall_prefetch is set, then the final segment
+	* between the main and the mall pixel requestor can use the same
+	* cache.
+	*
+	* Note that both bits may be set at the same time.
+	*/
+	struct {
+		bool mall_comb_mcache_p0;
+		bool mall_comb_mcache_p1;
+		bool plane0_plane1;
+	} last_slice_sharing;
+	/*
+	* A plane is divided into vertical slices of mcaches,
+	* which wrap on the surface width.
+	*
+	* For example, if the surface width is 7680, and split into
+	* three slices of equal width, the boundary array would contain
+	* [2560, 5120, 7680]
+	*
+	* The assignments are
+	* 0 = [0 .. 2559]
+	* 1 = [2560 .. 5119]
+	* 2 = [5120 .. 7679]
+	* 0 = [7680 .. INF]
+	* The final element implicitly is the same as the first, and
+	* at first seems invalid since it is never referenced (since)
+	* it is outside the surface. However, its useful when shifting
+	* (see below).
+	*
+	* For any given valid mcache assignment, a shifted version, wrapped
+	* on the surface width boundary is also assumed to be valid.
+	*
+	* For example, shifting [2560, 5120, 7680] by -50 results in
+	* [2510, 5170, 7630].
+	*
+	* The assignments are now:
+	* 0 = [0 .. 2509]
+	* 1 = [2510 .. 5169]
+	* 2 = [5170 .. 7629]
+	* 0 = [7630 .. INF]
+	*/
+	int mcache_x_offsets_plane0[DML2_MAX_MCACHES + 1];
+	int mcache_x_offsets_plane1[DML2_MAX_MCACHES + 1];
+};
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h
index a64ec4dcf11a..c047d56527c4 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h
@@ -43,4 +43,5 @@ bool dml2_build_mode_programming(struct dml2_build_mode_programming_in_out *in_o
  */
 bool dml2_build_mcache_programming(struct dml2_build_mcache_programming_in_out *in_out);
 
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h
index 25b607e7b726..84c90050668c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h
@@ -156,6 +156,8 @@ struct dml2_dchub_watermark_regs {
 	uint32_t urgent;
 	uint32_t sr_enter;
 	uint32_t sr_exit;
+	uint32_t sr_enter_z8;
+	uint32_t sr_exit_z8;
 	uint32_t uclk_pstate;
 	uint32_t fclk_pstate;
 	uint32_t temp_read_or_ppt;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h
index 5e1ab6d97640..255f05de362c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h
@@ -166,7 +166,7 @@ struct dml2_surface_cfg {
 	enum dml2_swizzle_mode tiling;
 
 	struct {
-		unsigned long pitch;
+		unsigned long pitch; // In elements, two pixels per element in 422 packed format
 		unsigned long width;
 		unsigned long height;
 	} plane0;
@@ -385,6 +385,7 @@ struct dml2_plane_parameters {
 		long reserved_vblank_time_ns;
 		unsigned int max_vactive_det_fill_delay_us; // 0 = no reserved time, +ve = explicit max delay
 		unsigned int gpuvm_min_page_size_kbytes;
+		unsigned int hostvm_min_page_size_kbytes;
 
 		enum dml2_svp_mode_override legacy_svp_config; //TODO remove in favor of svp_config
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c
index bb863c8c6b39..6ee37386f672 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c
@@ -456,10 +456,10 @@ bool core_dcn4_mode_support(struct dml2_core_mode_support_in_out *in_out)
 		in_out->mode_support_result.global.active.urgent_bw_dram_kbps = (unsigned long)math_ceil2((l->mode_support_ex_params.out_evaluation_info->urg_bandwidth_required_flip[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_dram] * 1000), 1.0);
 		in_out->mode_support_result.global.svp_prefetch.average_bw_dram_kbps = (unsigned long)math_ceil2((l->mode_support_ex_params.out_evaluation_info->avg_bandwidth_required[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_dram] * 1000), 1.0);
 		in_out->mode_support_result.global.svp_prefetch.urgent_bw_dram_kbps = (unsigned long)math_ceil2((l->mode_support_ex_params.out_evaluation_info->urg_bandwidth_required_flip[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_dram] * 1000), 1.0);
-		dml2_printf("DML::%s: in_out->mode_support_result.global.active.urgent_bw_sdp_kbps = %ld\n", __func__, in_out->mode_support_result.global.active.urgent_bw_sdp_kbps);
-		dml2_printf("DML::%s: in_out->mode_support_result.global.svp_prefetch.urgent_bw_sdp_kbps = %ld\n", __func__, in_out->mode_support_result.global.svp_prefetch.urgent_bw_sdp_kbps);
-		dml2_printf("DML::%s: in_out->mode_support_result.global.active.urgent_bw_dram_kbps = %ld\n", __func__, in_out->mode_support_result.global.active.urgent_bw_dram_kbps);
-		dml2_printf("DML::%s: in_out->mode_support_result.global.svp_prefetch.urgent_bw_dram_kbps = %ld\n", __func__, in_out->mode_support_result.global.svp_prefetch.urgent_bw_dram_kbps);
+		DML_LOG_VERBOSE("DML::%s: in_out->mode_support_result.global.active.urgent_bw_sdp_kbps = %ld\n", __func__, in_out->mode_support_result.global.active.urgent_bw_sdp_kbps);
+		DML_LOG_VERBOSE("DML::%s: in_out->mode_support_result.global.svp_prefetch.urgent_bw_sdp_kbps = %ld\n", __func__, in_out->mode_support_result.global.svp_prefetch.urgent_bw_sdp_kbps);
+		DML_LOG_VERBOSE("DML::%s: in_out->mode_support_result.global.active.urgent_bw_dram_kbps = %ld\n", __func__, in_out->mode_support_result.global.active.urgent_bw_dram_kbps);
+		DML_LOG_VERBOSE("DML::%s: in_out->mode_support_result.global.svp_prefetch.urgent_bw_dram_kbps = %ld\n", __func__, in_out->mode_support_result.global.svp_prefetch.urgent_bw_dram_kbps);
 
 		for (i = 0; i < l->svp_expanded_display_cfg.num_planes; i++) {
 			in_out->mode_support_result.per_plane[i].dppclk_khz = (unsigned int)(core->clean_me_up.mode_lib.ms.RequiredDPPCLK[i] * 1000);
@@ -509,7 +509,7 @@ bool core_dcn4_mode_support(struct dml2_core_mode_support_in_out *in_out)
 			stream_index = l->svp_expanded_display_cfg.plane_descriptors[i].stream_index;
 
 			in_out->mode_support_result.per_stream[stream_index].dscclk_khz = (unsigned int)core->clean_me_up.mode_lib.ms.required_dscclk_freq_mhz[i] * 1000;
-			dml2_printf("CORE_DCN4::%s: i=%d stream_index=%d, in_out->mode_support_result.per_stream[stream_index].dscclk_khz = %u\n", __func__, i, stream_index, in_out->mode_support_result.per_stream[stream_index].dscclk_khz);
+			DML_LOG_VERBOSE("CORE_DCN4::%s: i=%d stream_index=%d, in_out->mode_support_result.per_stream[stream_index].dscclk_khz = %u\n", __func__, i, stream_index, in_out->mode_support_result.per_stream[stream_index].dscclk_khz);
 
 			if (!((stream_bitmask >> stream_index) & 0x1)) {
 				in_out->mode_support_result.cfg_support_info.stream_support_info[stream_index].odms_used = odm_count;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c
index 4c504cb0e1c5..c4dad7164d31 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c
@@ -54,104 +54,104 @@ static double dml2_core_div_rem(double dividend, unsigned int divisor, unsigned
 
 static void dml2_print_mode_support_info(const struct dml2_core_internal_mode_support_info *support, bool fail_only)
 {
-	dml2_printf("DML: ===================================== \n");
-	dml2_printf("DML: DML_MODE_SUPPORT_INFO_ST\n");
+	DML_LOG_VERBOSE("DML: ===================================== \n");
+	DML_LOG_VERBOSE("DML: DML_MODE_SUPPORT_INFO_ST\n");
 	if (!fail_only || support->ScaleRatioAndTapsSupport == 0)
-		dml2_printf("DML: support: ScaleRatioAndTapsSupport = %d\n", support->ScaleRatioAndTapsSupport);
+		DML_LOG_VERBOSE("DML: support: ScaleRatioAndTapsSupport = %d\n", support->ScaleRatioAndTapsSupport);
 	if (!fail_only || support->SourceFormatPixelAndScanSupport == 0)
-		dml2_printf("DML: support: SourceFormatPixelAndScanSupport = %d\n", support->SourceFormatPixelAndScanSupport);
+		DML_LOG_VERBOSE("DML: support: SourceFormatPixelAndScanSupport = %d\n", support->SourceFormatPixelAndScanSupport);
 	if (!fail_only || support->ViewportSizeSupport == 0)
-		dml2_printf("DML: support: ViewportSizeSupport = %d\n", support->ViewportSizeSupport);
+		DML_LOG_VERBOSE("DML: support: ViewportSizeSupport = %d\n", support->ViewportSizeSupport);
 	if (!fail_only || support->LinkRateDoesNotMatchDPVersion == 1)
-		dml2_printf("DML: support: LinkRateDoesNotMatchDPVersion = %d\n", support->LinkRateDoesNotMatchDPVersion);
+		DML_LOG_VERBOSE("DML: support: LinkRateDoesNotMatchDPVersion = %d\n", support->LinkRateDoesNotMatchDPVersion);
 	if (!fail_only || support->LinkRateForMultistreamNotIndicated == 1)
-		dml2_printf("DML: support: LinkRateForMultistreamNotIndicated = %d\n", support->LinkRateForMultistreamNotIndicated);
+		DML_LOG_VERBOSE("DML: support: LinkRateForMultistreamNotIndicated = %d\n", support->LinkRateForMultistreamNotIndicated);
 	if (!fail_only || support->BPPForMultistreamNotIndicated == 1)
-		dml2_printf("DML: support: BPPForMultistreamNotIndicated = %d\n", support->BPPForMultistreamNotIndicated);
+		DML_LOG_VERBOSE("DML: support: BPPForMultistreamNotIndicated = %d\n", support->BPPForMultistreamNotIndicated);
 	if (!fail_only || support->MultistreamWithHDMIOreDP == 1)
-		dml2_printf("DML: support: MultistreamWithHDMIOreDP = %d\n", support->MultistreamWithHDMIOreDP);
+		DML_LOG_VERBOSE("DML: support: MultistreamWithHDMIOreDP = %d\n", support->MultistreamWithHDMIOreDP);
 	if (!fail_only || support->ExceededMultistreamSlots == 1)
-		dml2_printf("DML: support: ExceededMultistreamSlots = %d\n", support->ExceededMultistreamSlots);
+		DML_LOG_VERBOSE("DML: support: ExceededMultistreamSlots = %d\n", support->ExceededMultistreamSlots);
 	if (!fail_only || support->MSOOrODMSplitWithNonDPLink == 1)
-		dml2_printf("DML: support: MSOOrODMSplitWithNonDPLink = %d\n", support->MSOOrODMSplitWithNonDPLink);
+		DML_LOG_VERBOSE("DML: support: MSOOrODMSplitWithNonDPLink = %d\n", support->MSOOrODMSplitWithNonDPLink);
 	if (!fail_only || support->NotEnoughLanesForMSO == 1)
-		dml2_printf("DML: support: NotEnoughLanesForMSO = %d\n", support->NotEnoughLanesForMSO);
+		DML_LOG_VERBOSE("DML: support: NotEnoughLanesForMSO = %d\n", support->NotEnoughLanesForMSO);
 	if (!fail_only || support->P2IWith420 == 1)
-		dml2_printf("DML: support: P2IWith420 = %d\n", support->P2IWith420);
+		DML_LOG_VERBOSE("DML: support: P2IWith420 = %d\n", support->P2IWith420);
 	if (!fail_only || support->DSC422NativeNotSupported == 1)
-		dml2_printf("DML: support: DSC422NativeNotSupported = %d\n", support->DSC422NativeNotSupported);
+		DML_LOG_VERBOSE("DML: support: DSC422NativeNotSupported = %d\n", support->DSC422NativeNotSupported);
 	if (!fail_only || support->DSCSlicesODMModeSupported == 0)
-		dml2_printf("DML: support: DSCSlicesODMModeSupported = %d\n", support->DSCSlicesODMModeSupported);
+		DML_LOG_VERBOSE("DML: support: DSCSlicesODMModeSupported = %d\n", support->DSCSlicesODMModeSupported);
 	if (!fail_only || support->NotEnoughDSCUnits == 1)
-		dml2_printf("DML: support: NotEnoughDSCUnits = %d\n", support->NotEnoughDSCUnits);
+		DML_LOG_VERBOSE("DML: support: NotEnoughDSCUnits = %d\n", support->NotEnoughDSCUnits);
 	if (!fail_only || support->NotEnoughDSCSlices == 1)
-		dml2_printf("DML: support: NotEnoughDSCSlices = %d\n", support->NotEnoughDSCSlices);
+		DML_LOG_VERBOSE("DML: support: NotEnoughDSCSlices = %d\n", support->NotEnoughDSCSlices);
 	if (!fail_only || support->ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe == 1)
-		dml2_printf("DML: support: ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe = %d\n", support->ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe);
+		DML_LOG_VERBOSE("DML: support: ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe = %d\n", support->ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe);
 	if (!fail_only || support->InvalidCombinationOfMALLUseForPStateAndStaticScreen == 1)
-		dml2_printf("DML: support: InvalidCombinationOfMALLUseForPStateAndStaticScreen = %d\n", support->InvalidCombinationOfMALLUseForPStateAndStaticScreen);
+		DML_LOG_VERBOSE("DML: support: InvalidCombinationOfMALLUseForPStateAndStaticScreen = %d\n", support->InvalidCombinationOfMALLUseForPStateAndStaticScreen);
 	if (!fail_only || support->DSCCLKRequiredMoreThanSupported == 1)
-		dml2_printf("DML: support: DSCCLKRequiredMoreThanSupported = %d\n", support->DSCCLKRequiredMoreThanSupported);
+		DML_LOG_VERBOSE("DML: support: DSCCLKRequiredMoreThanSupported = %d\n", support->DSCCLKRequiredMoreThanSupported);
 	if (!fail_only || support->PixelsPerLinePerDSCUnitSupport == 0)
-		dml2_printf("DML: support: PixelsPerLinePerDSCUnitSupport = %d\n", support->PixelsPerLinePerDSCUnitSupport);
+		DML_LOG_VERBOSE("DML: support: PixelsPerLinePerDSCUnitSupport = %d\n", support->PixelsPerLinePerDSCUnitSupport);
 	if (!fail_only || support->DTBCLKRequiredMoreThanSupported == 1)
-		dml2_printf("DML: support: DTBCLKRequiredMoreThanSupported = %d\n", support->DTBCLKRequiredMoreThanSupported);
+		DML_LOG_VERBOSE("DML: support: DTBCLKRequiredMoreThanSupported = %d\n", support->DTBCLKRequiredMoreThanSupported);
 	if (!fail_only || support->InvalidCombinationOfMALLUseForPState == 1)
-		dml2_printf("DML: support: InvalidCombinationOfMALLUseForPState = %d\n", support->InvalidCombinationOfMALLUseForPState);
+		DML_LOG_VERBOSE("DML: support: InvalidCombinationOfMALLUseForPState = %d\n", support->InvalidCombinationOfMALLUseForPState);
 	if (!fail_only || support->ROBSupport == 0)
-		dml2_printf("DML: support: ROBSupport = %d\n", support->ROBSupport);
+		DML_LOG_VERBOSE("DML: support: ROBSupport = %d\n", support->ROBSupport);
 	if (!fail_only || support->OutstandingRequestsSupport == 0)
-		dml2_printf("DML: support: OutstandingRequestsSupport = %d\n", support->OutstandingRequestsSupport);
+		DML_LOG_VERBOSE("DML: support: OutstandingRequestsSupport = %d\n", support->OutstandingRequestsSupport);
 	if (!fail_only || support->OutstandingRequestsUrgencyAvoidance == 0)
-		dml2_printf("DML: support: OutstandingRequestsUrgencyAvoidance = %d\n", support->OutstandingRequestsUrgencyAvoidance);
+		DML_LOG_VERBOSE("DML: support: OutstandingRequestsUrgencyAvoidance = %d\n", support->OutstandingRequestsUrgencyAvoidance);
 	if (!fail_only || support->DISPCLK_DPPCLK_Support == 0)
-		dml2_printf("DML: support: DISPCLK_DPPCLK_Support = %d\n", support->DISPCLK_DPPCLK_Support);
+		DML_LOG_VERBOSE("DML: support: DISPCLK_DPPCLK_Support = %d\n", support->DISPCLK_DPPCLK_Support);
 	if (!fail_only || support->TotalAvailablePipesSupport == 0)
-		dml2_printf("DML: support: TotalAvailablePipesSupport = %d\n", support->TotalAvailablePipesSupport);
+		DML_LOG_VERBOSE("DML: support: TotalAvailablePipesSupport = %d\n", support->TotalAvailablePipesSupport);
 	if (!fail_only || support->NumberOfOTGSupport == 0)
-		dml2_printf("DML: support: NumberOfOTGSupport = %d\n", support->NumberOfOTGSupport);
+		DML_LOG_VERBOSE("DML: support: NumberOfOTGSupport = %d\n", support->NumberOfOTGSupport);
 	if (!fail_only || support->NumberOfHDMIFRLSupport == 0)
-		dml2_printf("DML: support: NumberOfHDMIFRLSupport = %d\n", support->NumberOfHDMIFRLSupport);
+		DML_LOG_VERBOSE("DML: support: NumberOfHDMIFRLSupport = %d\n", support->NumberOfHDMIFRLSupport);
 	if (!fail_only || support->NumberOfDP2p0Support == 0)
-		dml2_printf("DML: support: NumberOfDP2p0Support = %d\n", support->NumberOfDP2p0Support);
+		DML_LOG_VERBOSE("DML: support: NumberOfDP2p0Support = %d\n", support->NumberOfDP2p0Support);
 	if (!fail_only || support->EnoughWritebackUnits == 0)
-		dml2_printf("DML: support: EnoughWritebackUnits = %d\n", support->EnoughWritebackUnits);
+		DML_LOG_VERBOSE("DML: support: EnoughWritebackUnits = %d\n", support->EnoughWritebackUnits);
 	if (!fail_only || support->WritebackScaleRatioAndTapsSupport == 0)
-		dml2_printf("DML: support: WritebackScaleRatioAndTapsSupport = %d\n", support->WritebackScaleRatioAndTapsSupport);
+		DML_LOG_VERBOSE("DML: support: WritebackScaleRatioAndTapsSupport = %d\n", support->WritebackScaleRatioAndTapsSupport);
 	if (!fail_only || support->WritebackLatencySupport == 0)
-		dml2_printf("DML: support: WritebackLatencySupport = %d\n", support->WritebackLatencySupport);
+		DML_LOG_VERBOSE("DML: support: WritebackLatencySupport = %d\n", support->WritebackLatencySupport);
 	if (!fail_only || support->CursorSupport == 0)
-		dml2_printf("DML: support: CursorSupport = %d\n", support->CursorSupport);
+		DML_LOG_VERBOSE("DML: support: CursorSupport = %d\n", support->CursorSupport);
 	if (!fail_only || support->PitchSupport == 0)
-		dml2_printf("DML: support: PitchSupport = %d\n", support->PitchSupport);
+		DML_LOG_VERBOSE("DML: support: PitchSupport = %d\n", support->PitchSupport);
 	if (!fail_only || support->ViewportExceedsSurface == 1)
-		dml2_printf("DML: support: ViewportExceedsSurface = %d\n", support->ViewportExceedsSurface);
+		DML_LOG_VERBOSE("DML: support: ViewportExceedsSurface = %d\n", support->ViewportExceedsSurface);
 	if (!fail_only || support->PrefetchSupported == 0)
-		dml2_printf("DML: support: PrefetchSupported = %d\n", support->PrefetchSupported);
+		DML_LOG_VERBOSE("DML: support: PrefetchSupported = %d\n", support->PrefetchSupported);
 	if (!fail_only || support->EnoughUrgentLatencyHidingSupport == 0)
-		dml2_printf("DML: support: EnoughUrgentLatencyHidingSupport = %d\n", support->EnoughUrgentLatencyHidingSupport);
+		DML_LOG_VERBOSE("DML: support: EnoughUrgentLatencyHidingSupport = %d\n", support->EnoughUrgentLatencyHidingSupport);
 	if (!fail_only || support->AvgBandwidthSupport == 0)
-		dml2_printf("DML: support: AvgBandwidthSupport = %d\n", support->AvgBandwidthSupport);
+		DML_LOG_VERBOSE("DML: support: AvgBandwidthSupport = %d\n", support->AvgBandwidthSupport);
 	if (!fail_only || support->DynamicMetadataSupported == 0)
-		dml2_printf("DML: support: DynamicMetadataSupported = %d\n", support->DynamicMetadataSupported);
+		DML_LOG_VERBOSE("DML: support: DynamicMetadataSupported = %d\n", support->DynamicMetadataSupported);
 	if (!fail_only || support->VRatioInPrefetchSupported == 0)
-		dml2_printf("DML: support: VRatioInPrefetchSupported = %d\n", support->VRatioInPrefetchSupported);
+		DML_LOG_VERBOSE("DML: support: VRatioInPrefetchSupported = %d\n", support->VRatioInPrefetchSupported);
 	if (!fail_only || support->PTEBufferSizeNotExceeded == 0)
-		dml2_printf("DML: support: PTEBufferSizeNotExceeded = %d\n", support->PTEBufferSizeNotExceeded);
+		DML_LOG_VERBOSE("DML: support: PTEBufferSizeNotExceeded = %d\n", support->PTEBufferSizeNotExceeded);
 	if (!fail_only || support->DCCMetaBufferSizeNotExceeded == 0)
-		dml2_printf("DML: support: DCCMetaBufferSizeNotExceeded = %d\n", support->DCCMetaBufferSizeNotExceeded);
+		DML_LOG_VERBOSE("DML: support: DCCMetaBufferSizeNotExceeded = %d\n", support->DCCMetaBufferSizeNotExceeded);
 	if (!fail_only || support->ExceededMALLSize == 1)
-		dml2_printf("DML: support: ExceededMALLSize = %d\n", support->ExceededMALLSize);
+		DML_LOG_VERBOSE("DML: support: ExceededMALLSize = %d\n", support->ExceededMALLSize);
 	if (!fail_only || support->g6_temp_read_support == 0)
-		dml2_printf("DML: support: g6_temp_read_support = %d\n", support->g6_temp_read_support);
+		DML_LOG_VERBOSE("DML: support: g6_temp_read_support = %d\n", support->g6_temp_read_support);
 	if (!fail_only || support->ImmediateFlipSupport == 0)
-		dml2_printf("DML: support: ImmediateFlipSupport = %d\n", support->ImmediateFlipSupport);
+		DML_LOG_VERBOSE("DML: support: ImmediateFlipSupport = %d\n", support->ImmediateFlipSupport);
 	if (!fail_only || support->LinkCapacitySupport == 0)
-		dml2_printf("DML: support: LinkCapacitySupport = %d\n", support->LinkCapacitySupport);
+		DML_LOG_VERBOSE("DML: support: LinkCapacitySupport = %d\n", support->LinkCapacitySupport);
 
 	if (!fail_only || support->ModeSupport == 0)
-		dml2_printf("DML: support: ModeSupport = %d\n", support->ModeSupport);
-	dml2_printf("DML: ===================================== \n");
+		DML_LOG_VERBOSE("DML: support: ModeSupport = %d\n", support->ModeSupport);
+	DML_LOG_VERBOSE("DML: ===================================== \n");
 }
 
 static void get_stream_output_bpp(double *out_bpp, const struct dml2_display_cfg *display_cfg)
@@ -179,11 +179,9 @@ static void get_stream_output_bpp(double *out_bpp, const struct dml2_display_cfg
 		} else {
 			out_bpp[k] = 0;
 		}
-#ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%d bpc=%f\n", __func__, k, bpc);
-		dml2_printf("DML::%s: k=%d dsc.enable=%d\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.dsc.enable);
-		dml2_printf("DML::%s: k=%d out_bpp=%f\n", __func__, k, out_bpp[k]);
-#endif
+		DML_LOG_VERBOSE("DML::%s: k=%d bpc=%f\n", __func__, k, bpc);
+		DML_LOG_VERBOSE("DML::%s: k=%d dsc.enable=%d\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.dsc.enable);
+		DML_LOG_VERBOSE("DML::%s: k=%d out_bpp=%f\n", __func__, k, out_bpp[k]);
 	}
 }
 
@@ -212,9 +210,7 @@ static unsigned int dml_get_num_active_pipes(int unsigned num_planes, const stru
 		num_active_pipes = num_active_pipes + (unsigned int)cfg_support_info->plane_support_info[k].dpps_used;
 	}
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: num_active_pipes = %d\n", __func__, num_active_pipes);
-#endif
+	DML_LOG_VERBOSE("DML::%s: num_active_pipes = %d\n", __func__, num_active_pipes);
 	return num_active_pipes;
 }
 
@@ -251,7 +247,7 @@ static bool dml_get_is_phantom_pipe(const struct dml2_display_cfg *display_cfg,
 	unsigned int plane_idx = mode_lib->mp.pipe_plane[pipe_idx];
 
 	bool is_phantom = dml_is_phantom_pipe(&display_cfg->plane_descriptors[plane_idx]);
-	dml2_printf("DML::%s: pipe_idx=%d legacy_svp_config=%0d is_phantom=%d\n", __func__, pipe_idx, display_cfg->plane_descriptors[plane_idx].overrides.legacy_svp_config, is_phantom);
+	DML_LOG_VERBOSE("DML::%s: pipe_idx=%d legacy_svp_config=%0d is_phantom=%d\n", __func__, pipe_idx, display_cfg->plane_descriptors[plane_idx].overrides.legacy_svp_config, is_phantom);
 	return is_phantom;
 }
 
@@ -415,19 +411,17 @@ static void CalculateMaxDETAndMinCompressedBufferSize(
 	*nomDETInKByte = (unsigned int)(math_floor2((double)*MaxTotalDETInKByte / (double)MaxNumDPP, ConfigReturnBufferSegmentSizeInKByte));
 	*MinCompressedBufferSizeInKByte = ConfigReturnBufferSizeInKByte - *MaxTotalDETInKByte;
 
-#if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: is_mrq_present = %u\n", __func__, is_mrq_present);
-	dml2_printf("DML::%s: ConfigReturnBufferSizeInKByte = %u\n", __func__, ConfigReturnBufferSizeInKByte);
-	dml2_printf("DML::%s: ROBBufferSizeInKByte = %u\n", __func__, ROBBufferSizeInKByte);
-	dml2_printf("DML::%s: MaxNumDPP = %u\n", __func__, MaxNumDPP);
-	dml2_printf("DML::%s: MaxTotalDETInKByte = %u\n", __func__, *MaxTotalDETInKByte);
-	dml2_printf("DML::%s: nomDETInKByte = %u\n", __func__, *nomDETInKByte);
-	dml2_printf("DML::%s: MinCompressedBufferSizeInKByte = %u\n", __func__, *MinCompressedBufferSizeInKByte);
-#endif
+	DML_LOG_VERBOSE("DML::%s: is_mrq_present = %u\n", __func__, is_mrq_present);
+	DML_LOG_VERBOSE("DML::%s: ConfigReturnBufferSizeInKByte = %u\n", __func__, ConfigReturnBufferSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: ROBBufferSizeInKByte = %u\n", __func__, ROBBufferSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: MaxNumDPP = %u\n", __func__, MaxNumDPP);
+	DML_LOG_VERBOSE("DML::%s: MaxTotalDETInKByte = %u\n", __func__, *MaxTotalDETInKByte);
+	DML_LOG_VERBOSE("DML::%s: nomDETInKByte = %u\n", __func__, *nomDETInKByte);
+	DML_LOG_VERBOSE("DML::%s: MinCompressedBufferSizeInKByte = %u\n", __func__, *MinCompressedBufferSizeInKByte);
 
 	if (nomDETInKByteOverrideEnable) {
 		*nomDETInKByte = nomDETInKByteOverrideValue;
-		dml2_printf("DML::%s: nomDETInKByte = %u (overrided)\n", __func__, *nomDETInKByte);
+		DML_LOG_VERBOSE("DML::%s: nomDETInKByte = %u (overrided)\n", __func__, *nomDETInKByte);
 	}
 }
 
@@ -502,7 +496,7 @@ static bool dml_is_420(enum dml2_source_format_class source_format)
 		val = 0;
 		break;
 	default:
-		DML2_ASSERT(0);
+		DML_ASSERT(0);
 		break;
 	}
 	return val;
@@ -535,7 +529,7 @@ static unsigned int dml_get_tile_block_size_bytes(enum dml2_swizzle_mode sw_mode
 	else if (sw_mode == dml2_gfx11_sw_256kb_r_x)
 		return 262144;
 	else {
-		DML2_ASSERT(0);
+		DML_ASSERT(0);
 		return 256;
 	}
 }
@@ -570,8 +564,8 @@ static int unsigned dml_get_gfx_version(enum dml2_swizzle_mode sw_mode)
 		sw_mode == dml2_gfx11_sw_256kb_r_x) {
 		version = 11;
 	} else {
-		dml2_printf("ERROR: Invalid sw_mode setting! val=%u\n", sw_mode);
-		DML2_ASSERT(0);
+		DML_LOG_VERBOSE("ERROR: Invalid sw_mode setting! val=%u\n", sw_mode);
+		DML_ASSERT(0);
 	}
 
 	return version;
@@ -645,21 +639,19 @@ static void CalculateBytePerPixelAndBlockSizes(
 		*BytePerPixelY = 2;
 		*BytePerPixelC = 4;
 	} else {
-		dml2_printf("ERROR: DML::%s: SourcePixelFormat = %u not supported!\n", __func__, SourcePixelFormat);
-		DML2_ASSERT(0);
+		DML_LOG_VERBOSE("ERROR: DML::%s: SourcePixelFormat = %u not supported!\n", __func__, SourcePixelFormat);
+		DML_ASSERT(0);
 	}
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: SourcePixelFormat = %u\n", __func__, SourcePixelFormat);
-	dml2_printf("DML::%s: BytePerPixelDETY = %f\n", __func__, *BytePerPixelDETY);
-	dml2_printf("DML::%s: BytePerPixelDETC = %f\n", __func__, *BytePerPixelDETC);
-	dml2_printf("DML::%s: BytePerPixelY = %u\n", __func__, *BytePerPixelY);
-	dml2_printf("DML::%s: BytePerPixelC = %u\n", __func__, *BytePerPixelC);
-	dml2_printf("DML::%s: pitch_y = %u\n", __func__, pitch_y);
-	dml2_printf("DML::%s: pitch_c = %u\n", __func__, pitch_c);
-	dml2_printf("DML::%s: surf_linear128_l = %u\n", __func__, *surf_linear128_l);
-	dml2_printf("DML::%s: surf_linear128_c = %u\n", __func__, *surf_linear128_c);
-#endif
+	DML_LOG_VERBOSE("DML::%s: SourcePixelFormat = %u\n", __func__, SourcePixelFormat);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelDETY = %f\n", __func__, *BytePerPixelDETY);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelDETC = %f\n", __func__, *BytePerPixelDETC);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelY = %u\n", __func__, *BytePerPixelY);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelC = %u\n", __func__, *BytePerPixelC);
+	DML_LOG_VERBOSE("DML::%s: pitch_y = %u\n", __func__, pitch_y);
+	DML_LOG_VERBOSE("DML::%s: pitch_c = %u\n", __func__, pitch_c);
+	DML_LOG_VERBOSE("DML::%s: surf_linear128_l = %u\n", __func__, *surf_linear128_l);
+	DML_LOG_VERBOSE("DML::%s: surf_linear128_c = %u\n", __func__, *surf_linear128_c);
 
 	if (dml_get_gfx_version(SurfaceTiling) == 11) {
 		*surf_linear128_l = 0;
@@ -703,12 +695,10 @@ static void CalculateBytePerPixelAndBlockSizes(
 		*BlockWidth256BytesY = 256U / *BytePerPixelY / *BlockHeight256BytesY;
 		*BlockWidth256BytesC = 256U / *BytePerPixelC / *BlockHeight256BytesC;
 	}
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: BlockWidth256BytesY = %u\n", __func__, *BlockWidth256BytesY);
-	dml2_printf("DML::%s: BlockHeight256BytesY = %u\n", __func__, *BlockHeight256BytesY);
-	dml2_printf("DML::%s: BlockWidth256BytesC = %u\n", __func__, *BlockWidth256BytesC);
-	dml2_printf("DML::%s: BlockHeight256BytesC = %u\n", __func__, *BlockHeight256BytesC);
-#endif
+	DML_LOG_VERBOSE("DML::%s: BlockWidth256BytesY = %u\n", __func__, *BlockWidth256BytesY);
+	DML_LOG_VERBOSE("DML::%s: BlockHeight256BytesY = %u\n", __func__, *BlockHeight256BytesY);
+	DML_LOG_VERBOSE("DML::%s: BlockWidth256BytesC = %u\n", __func__, *BlockWidth256BytesC);
+	DML_LOG_VERBOSE("DML::%s: BlockHeight256BytesC = %u\n", __func__, *BlockHeight256BytesC);
 
 	if (dml_get_gfx_version(SurfaceTiling) == 11) {
 		if (SurfaceTiling == dml2_gfx11_sw_linear) {
@@ -752,8 +742,8 @@ static void CalculateBytePerPixelAndBlockSizes(
 		} else if (SurfaceTiling == dml2_sw_256kb_2d) {
 			macro_tile_scale = 32;
 		} else {
-			dml2_printf("ERROR: Invalid SurfaceTiling setting! val=%u\n", SurfaceTiling);
-			DML2_ASSERT(0);
+			DML_LOG_VERBOSE("ERROR: Invalid SurfaceTiling setting! val=%u\n", SurfaceTiling);
+			DML_ASSERT(0);
 		}
 
 		*MacroTileHeightY = macro_tile_scale * *BlockHeight256BytesY;
@@ -766,12 +756,10 @@ static void CalculateBytePerPixelAndBlockSizes(
 		}
 	}
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: MacroTileWidthY = %u\n", __func__, *MacroTileWidthY);
-	dml2_printf("DML::%s: MacroTileHeightY = %u\n", __func__, *MacroTileHeightY);
-	dml2_printf("DML::%s: MacroTileWidthC = %u\n", __func__, *MacroTileWidthC);
-	dml2_printf("DML::%s: MacroTileHeightC = %u\n", __func__, *MacroTileHeightC);
-#endif
+	DML_LOG_VERBOSE("DML::%s: MacroTileWidthY = %u\n", __func__, *MacroTileWidthY);
+	DML_LOG_VERBOSE("DML::%s: MacroTileHeightY = %u\n", __func__, *MacroTileHeightY);
+	DML_LOG_VERBOSE("DML::%s: MacroTileWidthC = %u\n", __func__, *MacroTileWidthC);
+	DML_LOG_VERBOSE("DML::%s: MacroTileHeightC = %u\n", __func__, *MacroTileHeightC);
 }
 
 static void CalculateSinglePipeDPPCLKAndSCLThroughput(
@@ -860,10 +848,8 @@ static void CalculateSwathWidth(
 	unsigned int surface_width_ub_c;
 	unsigned int surface_height_ub_c;
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: ForceSingleDPP = %u\n", __func__, ForceSingleDPP);
-	dml2_printf("DML::%s: NumberOfActiveSurfaces = %u\n", __func__, NumberOfActiveSurfaces);
-#endif
+	DML_LOG_VERBOSE("DML::%s: ForceSingleDPP = %u\n", __func__, ForceSingleDPP);
+	DML_LOG_VERBOSE("DML::%s: NumberOfActiveSurfaces = %u\n", __func__, NumberOfActiveSurfaces);
 
 	for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
 		if (!dml_is_vertical_rotation(display_cfg->plane_descriptors[k].composition.rotation_angle)) {
@@ -872,11 +858,9 @@ static void CalculateSwathWidth(
 			SwathWidthSingleDPPY[k] = (unsigned int)display_cfg->plane_descriptors[k].composition.viewport.plane0.height;
 		}
 
-#ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u ViewportWidth=%u\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.width);
-		dml2_printf("DML::%s: k=%u ViewportHeight=%u\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.height);
-		dml2_printf("DML::%s: k=%u DPPPerSurface=%u\n", __func__, k, DPPPerSurface[k]);
-#endif
+		DML_LOG_VERBOSE("DML::%s: k=%u ViewportWidth=%lu\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.width);
+		DML_LOG_VERBOSE("DML::%s: k=%u ViewportHeight=%lu\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.height);
+		DML_LOG_VERBOSE("DML::%s: k=%u DPPPerSurface=%u\n", __func__, k, DPPPerSurface[k]);
 
 		MainSurfaceODMMode = ODMMode[k];
 
@@ -899,13 +883,11 @@ static void CalculateSwathWidth(
 			}
 		}
 
-#ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u HActive=%u\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_active);
-		dml2_printf("DML::%s: k=%u HRatio=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio);
-		dml2_printf("DML::%s: k=%u MainSurfaceODMMode=%u\n", __func__, k, MainSurfaceODMMode);
-		dml2_printf("DML::%s: k=%u SwathWidthSingleDPPY=%u\n", __func__, k, SwathWidthSingleDPPY[k]);
-		dml2_printf("DML::%s: k=%u SwathWidthY=%u\n", __func__, k, SwathWidthY[k]);
-#endif
+		DML_LOG_VERBOSE("DML::%s: k=%u HActive=%lu\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_active);
+		DML_LOG_VERBOSE("DML::%s: k=%u HRatio=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u MainSurfaceODMMode=%u\n", __func__, k, MainSurfaceODMMode);
+		DML_LOG_VERBOSE("DML::%s: k=%u SwathWidthSingleDPPY=%u\n", __func__, k, SwathWidthSingleDPPY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u SwathWidthY=%u\n", __func__, k, SwathWidthY[k]);
 
 		if (dml_is_420(display_cfg->plane_descriptors[k].pixel_format)) {
 			SwathWidthC[k] = SwathWidthY[k] / 2;
@@ -934,22 +916,20 @@ static void CalculateSwathWidth(
 		surface_width_ub_c = (unsigned int)math_ceil2((double)display_cfg->plane_descriptors[k].surface.plane1.width, req_width_horz_c);
 		surface_height_ub_c = (unsigned int)math_ceil2((double)display_cfg->plane_descriptors[k].surface.plane1.height, Read256BytesBlockHeightC[k]);
 
-#ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u surface_width_ub_l=%u\n", __func__, k, surface_width_ub_l);
-		dml2_printf("DML::%s: k=%u surface_height_ub_l=%u\n", __func__, k, surface_height_ub_l);
-		dml2_printf("DML::%s: k=%u surface_width_ub_c=%u\n", __func__, k, surface_width_ub_c);
-		dml2_printf("DML::%s: k=%u surface_height_ub_c=%u\n", __func__, k, surface_height_ub_c);
-		dml2_printf("DML::%s: k=%u req_width_horz_y=%u\n", __func__, k, req_width_horz_y);
-		dml2_printf("DML::%s: k=%u req_width_horz_c=%u\n", __func__, k, req_width_horz_c);
-		dml2_printf("DML::%s: k=%u Read256BytesBlockWidthY=%u\n", __func__, k, Read256BytesBlockWidthY[k]);
-		dml2_printf("DML::%s: k=%u Read256BytesBlockHeightY=%u\n", __func__, k, Read256BytesBlockHeightY[k]);
-		dml2_printf("DML::%s: k=%u Read256BytesBlockWidthC=%u\n", __func__, k, Read256BytesBlockWidthC[k]);
-		dml2_printf("DML::%s: k=%u Read256BytesBlockHeightC=%u\n", __func__, k, Read256BytesBlockHeightC[k]);
-		dml2_printf("DML::%s: k=%u req_width_horz_y=%u\n", __func__, k, req_width_horz_y);
-		dml2_printf("DML::%s: k=%u req_width_horz_c=%u\n", __func__, k, req_width_horz_c);
-		dml2_printf("DML::%s: k=%u ViewportStationary=%u\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.stationary);
-		dml2_printf("DML::%s: k=%u DPPPerSurface=%u\n", __func__, k, DPPPerSurface[k]);
-#endif
+		DML_LOG_VERBOSE("DML::%s: k=%u surface_width_ub_l=%u\n", __func__, k, surface_width_ub_l);
+		DML_LOG_VERBOSE("DML::%s: k=%u surface_height_ub_l=%u\n", __func__, k, surface_height_ub_l);
+		DML_LOG_VERBOSE("DML::%s: k=%u surface_width_ub_c=%u\n", __func__, k, surface_width_ub_c);
+		DML_LOG_VERBOSE("DML::%s: k=%u surface_height_ub_c=%u\n", __func__, k, surface_height_ub_c);
+		DML_LOG_VERBOSE("DML::%s: k=%u req_width_horz_y=%u\n", __func__, k, req_width_horz_y);
+		DML_LOG_VERBOSE("DML::%s: k=%u req_width_horz_c=%u\n", __func__, k, req_width_horz_c);
+		DML_LOG_VERBOSE("DML::%s: k=%u Read256BytesBlockWidthY=%u\n", __func__, k, Read256BytesBlockWidthY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u Read256BytesBlockHeightY=%u\n", __func__, k, Read256BytesBlockHeightY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u Read256BytesBlockWidthC=%u\n", __func__, k, Read256BytesBlockWidthC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u Read256BytesBlockHeightC=%u\n", __func__, k, Read256BytesBlockHeightC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u req_width_horz_y=%u\n", __func__, k, req_width_horz_y);
+		DML_LOG_VERBOSE("DML::%s: k=%u req_width_horz_c=%u\n", __func__, k, req_width_horz_c);
+		DML_LOG_VERBOSE("DML::%s: k=%u ViewportStationary=%u\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.stationary);
+		DML_LOG_VERBOSE("DML::%s: k=%u DPPPerSurface=%u\n", __func__, k, DPPPerSurface[k]);
 
 		req_per_swath_ub_l[k] = 0;
 		req_per_swath_ub_c[k] = 0;
@@ -995,15 +975,12 @@ static void CalculateSwathWidth(
 			}
 		}
 
-#ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u swath_width_luma_ub=%u\n", __func__, k, swath_width_luma_ub[k]);
-		dml2_printf("DML::%s: k=%u swath_width_chroma_ub=%u\n", __func__, k, swath_width_chroma_ub[k]);
-		dml2_printf("DML::%s: k=%u MaximumSwathHeightY=%u\n", __func__, k, MaximumSwathHeightY[k]);
-		dml2_printf("DML::%s: k=%u MaximumSwathHeightC=%u\n", __func__, k, MaximumSwathHeightC[k]);
-		dml2_printf("DML::%s: k=%u req_per_swath_ub_l=%u\n", __func__, k, req_per_swath_ub_l[k]);
-		dml2_printf("DML::%s: k=%u req_per_swath_ub_c=%u\n", __func__, k, req_per_swath_ub_c[k]);
-#endif
-
+		DML_LOG_VERBOSE("DML::%s: k=%u swath_width_luma_ub=%u\n", __func__, k, swath_width_luma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u swath_width_chroma_ub=%u\n", __func__, k, swath_width_chroma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathHeightY=%u\n", __func__, k, MaximumSwathHeightY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathHeightC=%u\n", __func__, k, MaximumSwathHeightC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u req_per_swath_ub_l=%u\n", __func__, k, req_per_swath_ub_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u req_per_swath_ub_c=%u\n", __func__, k, req_per_swath_ub_c[k]);
 	}
 }
 
@@ -1018,13 +995,11 @@ static bool UnboundedRequest(bool unb_req_force_en, bool unb_req_force_val, unsi
 	if (unb_req_force_en) {
 		unb_req_en = unb_req_force_val && unb_req_ok;
 	}
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: unb_req_force_en = %u\n", __func__, unb_req_force_en);
-	dml2_printf("DML::%s: unb_req_force_val = %u\n", __func__, unb_req_force_val);
-	dml2_printf("DML::%s: unb_req_ok = %u\n", __func__, unb_req_ok);
-	dml2_printf("DML::%s: unb_req_en = %u\n", __func__, unb_req_en);
-#endif
-	return (unb_req_en);
+	DML_LOG_VERBOSE("DML::%s: unb_req_force_en = %u\n", __func__, unb_req_force_en);
+	DML_LOG_VERBOSE("DML::%s: unb_req_force_val = %u\n", __func__, unb_req_force_val);
+	DML_LOG_VERBOSE("DML::%s: unb_req_ok = %u\n", __func__, unb_req_ok);
+	DML_LOG_VERBOSE("DML::%s: unb_req_en = %u\n", __func__, unb_req_en);
+	return unb_req_en;
 }
 
 static void CalculateDETBufferSize(
@@ -1054,16 +1029,14 @@ static void CalculateDETBufferSize(
 	bool NextPotentialSurfaceToAssignDETPieceFound;
 	bool MinimizeReallocationSuccess = false;
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: ForceSingleDPP = %u\n", __func__, ForceSingleDPP);
-	dml2_printf("DML::%s: nomDETInKByte = %u\n", __func__, nomDETInKByte);
-	dml2_printf("DML::%s: NumberOfActiveSurfaces = %u\n", __func__, NumberOfActiveSurfaces);
-	dml2_printf("DML::%s: UnboundedRequestEnabled = %u\n", __func__, UnboundedRequestEnabled);
-	dml2_printf("DML::%s: MaxTotalDETInKByte = %u\n", __func__, MaxTotalDETInKByte);
-	dml2_printf("DML::%s: ConfigReturnBufferSizeInKByte = %u\n", __func__, ConfigReturnBufferSizeInKByte);
-	dml2_printf("DML::%s: MinCompressedBufferSizeInKByte = %u\n", __func__, MinCompressedBufferSizeInKByte);
-	dml2_printf("DML::%s: CompressedBufferSegmentSizeInkByte = %u\n", __func__, CompressedBufferSegmentSizeInkByte);
-#endif
+	DML_LOG_VERBOSE("DML::%s: ForceSingleDPP = %u\n", __func__, ForceSingleDPP);
+	DML_LOG_VERBOSE("DML::%s: nomDETInKByte = %u\n", __func__, nomDETInKByte);
+	DML_LOG_VERBOSE("DML::%s: NumberOfActiveSurfaces = %u\n", __func__, NumberOfActiveSurfaces);
+	DML_LOG_VERBOSE("DML::%s: UnboundedRequestEnabled = %u\n", __func__, UnboundedRequestEnabled);
+	DML_LOG_VERBOSE("DML::%s: MaxTotalDETInKByte = %u\n", __func__, MaxTotalDETInKByte);
+	DML_LOG_VERBOSE("DML::%s: ConfigReturnBufferSizeInKByte = %u\n", __func__, ConfigReturnBufferSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: MinCompressedBufferSizeInKByte = %u\n", __func__, MinCompressedBufferSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: CompressedBufferSegmentSizeInkByte = %u\n", __func__, CompressedBufferSegmentSizeInkByte);
 
 	// Note: Will use default det size if that fits 2 swaths
 	if (UnboundedRequestEnabled) {
@@ -1092,19 +1065,15 @@ static void CalculateDETBufferSize(
 				l->minDET = l->minDET + ConfigReturnBufferSegmentSizeInkByte;
 			}
 
-#ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u minDET = %u\n", __func__, k, l->minDET);
-			dml2_printf("DML::%s: k=%u max_minDET = %u\n", __func__, k, l->max_minDET);
-			dml2_printf("DML::%s: k=%u minDET_pipe = %u\n", __func__, k, l->minDET_pipe);
-			dml2_printf("DML::%s: k=%u full_swath_bytes_l = %u\n", __func__, k, full_swath_bytes_l[k]);
-			dml2_printf("DML::%s: k=%u full_swath_bytes_c = %u\n", __func__, k, full_swath_bytes_c[k]);
-#endif
+			DML_LOG_VERBOSE("DML::%s: k=%u minDET = %u\n", __func__, k, l->minDET);
+			DML_LOG_VERBOSE("DML::%s: k=%u max_minDET = %u\n", __func__, k, l->max_minDET);
+			DML_LOG_VERBOSE("DML::%s: k=%u minDET_pipe = %u\n", __func__, k, l->minDET_pipe);
+			DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_l = %u\n", __func__, k, full_swath_bytes_l[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_c = %u\n", __func__, k, full_swath_bytes_c[k]);
 
 			if (l->minDET_pipe == 0) {
 				l->minDET_pipe = (unsigned int)(math_max2(128, math_ceil2(((double)full_swath_bytes_l[k] + (double)full_swath_bytes_c[k]) / 1024.0, ConfigReturnBufferSegmentSizeInkByte)));
-#ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k=%u minDET_pipe = %u (assume each plane take half DET)\n", __func__, k, l->minDET_pipe);
-#endif
+				DML_LOG_VERBOSE("DML::%s: k=%u minDET_pipe = %u (assume each plane take half DET)\n", __func__, k, l->minDET_pipe);
 			}
 
 			if (dml_is_phantom_pipe(&display_cfg->plane_descriptors[k])) {
@@ -1117,12 +1086,10 @@ static void CalculateDETBufferSize(
 				l->DETBufferSizePoolInKByte = l->DETBufferSizePoolInKByte - (ForceSingleDPP ? 1 : DPPPerSurface[k]) * l->minDET_pipe;
 			}
 
-#ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u DPPPerSurface = %u\n", __func__, k, DPPPerSurface[k]);
-			dml2_printf("DML::%s: k=%u DETSizeOverride = %u\n", __func__, k, display_cfg->plane_descriptors[k].overrides.det_size_override_kb);
-			dml2_printf("DML::%s: k=%u DETBufferSizeInKByte = %u\n", __func__, k, DETBufferSizeInKByte[k]);
-			dml2_printf("DML::%s: DETBufferSizePoolInKByte = %u\n", __func__, l->DETBufferSizePoolInKByte);
-#endif
+			DML_LOG_VERBOSE("DML::%s: k=%u DPPPerSurface = %u\n", __func__, k, DPPPerSurface[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u DETSizeOverride = %u\n", __func__, k, display_cfg->plane_descriptors[k].overrides.det_size_override_kb);
+			DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeInKByte = %u\n", __func__, k, DETBufferSizeInKByte[k]);
+			DML_LOG_VERBOSE("DML::%s: DETBufferSizePoolInKByte = %u\n", __func__, l->DETBufferSizePoolInKByte);
 		}
 
 		if (display_cfg->minimize_det_reallocation) {
@@ -1194,14 +1161,12 @@ static void CalculateDETBufferSize(
 					l->TotalBandwidth = l->TotalBandwidth + ReadBandwidthLuma[k] + ReadBandwidthChroma[k];
 				}
 			}
-#ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: --- Before bandwidth adjustment ---\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: --- Before bandwidth adjustment ---\n", __func__);
 			for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
-				dml2_printf("DML::%s: k=%u DETBufferSizeInKByte = %u\n", __func__, k, DETBufferSizeInKByte[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeInKByte = %u\n", __func__, k, DETBufferSizeInKByte[k]);
 			}
-			dml2_printf("DML::%s: --- DET allocation with bandwidth ---\n", __func__);
-#endif
-			dml2_printf("DML::%s: TotalBandwidth = %f\n", __func__, l->TotalBandwidth);
+			DML_LOG_VERBOSE("DML::%s: --- DET allocation with bandwidth ---\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: TotalBandwidth = %f\n", __func__, l->TotalBandwidth);
 			l->BandwidthOfSurfacesNotAssignedDETPiece = l->TotalBandwidth;
 			for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
 
@@ -1213,10 +1178,8 @@ static void CalculateDETBufferSize(
 				} else {
 					DETPieceAssignedToThisSurfaceAlready[k] = false;
 				}
-#ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k=%u DETPieceAssignedToThisSurfaceAlready = %u\n", __func__, k, DETPieceAssignedToThisSurfaceAlready[k]);
-				dml2_printf("DML::%s: k=%u BandwidthOfSurfacesNotAssignedDETPiece = %f\n", __func__, k, l->BandwidthOfSurfacesNotAssignedDETPiece);
-#endif
+				DML_LOG_VERBOSE("DML::%s: k=%u DETPieceAssignedToThisSurfaceAlready = %u\n", __func__, k, DETPieceAssignedToThisSurfaceAlready[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u BandwidthOfSurfacesNotAssignedDETPiece = %f\n", __func__, k, l->BandwidthOfSurfacesNotAssignedDETPiece);
 			}
 
 			for (unsigned int j = 0; j < NumberOfActiveSurfaces; ++j) {
@@ -1224,22 +1187,18 @@ static void CalculateDETBufferSize(
 				l->NextSurfaceToAssignDETPiece = 0;
 
 				for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
-#ifdef __DML_VBA_DEBUG__
-					dml2_printf("DML::%s: j=%u k=%u, ReadBandwidthLuma[k] = %f\n", __func__, j, k, ReadBandwidthLuma[k]);
-					dml2_printf("DML::%s: j=%u k=%u, ReadBandwidthChroma[k] = %f\n", __func__, j, k, ReadBandwidthChroma[k]);
-					dml2_printf("DML::%s: j=%u k=%u, ReadBandwidthLuma[Next] = %f\n", __func__, j, k, ReadBandwidthLuma[l->NextSurfaceToAssignDETPiece]);
-					dml2_printf("DML::%s: j=%u k=%u, ReadBandwidthChroma[Next] = %f\n", __func__, j, k, ReadBandwidthChroma[l->NextSurfaceToAssignDETPiece]);
-					dml2_printf("DML::%s: j=%u k=%u, NextSurfaceToAssignDETPiece = %u\n", __func__, j, k, l->NextSurfaceToAssignDETPiece);
-#endif
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, ReadBandwidthLuma[k] = %f\n", __func__, j, k, ReadBandwidthLuma[k]);
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, ReadBandwidthChroma[k] = %f\n", __func__, j, k, ReadBandwidthChroma[k]);
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, ReadBandwidthLuma[Next] = %f\n", __func__, j, k, ReadBandwidthLuma[l->NextSurfaceToAssignDETPiece]);
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, ReadBandwidthChroma[Next] = %f\n", __func__, j, k, ReadBandwidthChroma[l->NextSurfaceToAssignDETPiece]);
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, NextSurfaceToAssignDETPiece = %u\n", __func__, j, k, l->NextSurfaceToAssignDETPiece);
 					if (!DETPieceAssignedToThisSurfaceAlready[k] && (!NextPotentialSurfaceToAssignDETPieceFound ||
 						ReadBandwidthLuma[k] + ReadBandwidthChroma[k] < ReadBandwidthLuma[l->NextSurfaceToAssignDETPiece] + ReadBandwidthChroma[l->NextSurfaceToAssignDETPiece])) {
 						l->NextSurfaceToAssignDETPiece = k;
 						NextPotentialSurfaceToAssignDETPieceFound = true;
 					}
-#ifdef __DML_VBA_DEBUG__
-					dml2_printf("DML::%s: j=%u k=%u, DETPieceAssignedToThisSurfaceAlready = %u\n", __func__, j, k, DETPieceAssignedToThisSurfaceAlready[k]);
-					dml2_printf("DML::%s: j=%u k=%u, NextPotentialSurfaceToAssignDETPieceFound = %u\n", __func__, j, k, NextPotentialSurfaceToAssignDETPieceFound);
-#endif
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, DETPieceAssignedToThisSurfaceAlready = %u\n", __func__, j, k, DETPieceAssignedToThisSurfaceAlready[k]);
+					DML_LOG_VERBOSE("DML::%s: j=%u k=%u, NextPotentialSurfaceToAssignDETPieceFound = %u\n", __func__, j, k, NextPotentialSurfaceToAssignDETPieceFound);
 				}
 
 				if (NextPotentialSurfaceToAssignDETPieceFound) {
@@ -1249,20 +1208,16 @@ static void CalculateDETBufferSize(
 						* (ForceSingleDPP ? 1 : DPPPerSurface[l->NextSurfaceToAssignDETPiece]) * ConfigReturnBufferSegmentSizeInkByte,
 						math_floor2((double)l->DETBufferSizePoolInKByte, (ForceSingleDPP ? 1 : DPPPerSurface[l->NextSurfaceToAssignDETPiece]) * ConfigReturnBufferSegmentSizeInkByte)));
 
-#ifdef __DML_VBA_DEBUG__
-					dml2_printf("DML::%s: j=%u, DETBufferSizePoolInKByte = %u\n", __func__, j, l->DETBufferSizePoolInKByte);
-					dml2_printf("DML::%s: j=%u, NextSurfaceToAssignDETPiece = %u\n", __func__, j, l->NextSurfaceToAssignDETPiece);
-					dml2_printf("DML::%s: j=%u, ReadBandwidthLuma[%u] = %f\n", __func__, j, l->NextSurfaceToAssignDETPiece, ReadBandwidthLuma[l->NextSurfaceToAssignDETPiece]);
-					dml2_printf("DML::%s: j=%u, ReadBandwidthChroma[%u] = %f\n", __func__, j, l->NextSurfaceToAssignDETPiece, ReadBandwidthChroma[l->NextSurfaceToAssignDETPiece]);
-					dml2_printf("DML::%s: j=%u, BandwidthOfSurfacesNotAssignedDETPiece = %f\n", __func__, j, l->BandwidthOfSurfacesNotAssignedDETPiece);
-					dml2_printf("DML::%s: j=%u, NextDETBufferPieceInKByte = %u\n", __func__, j, l->NextDETBufferPieceInKByte);
-					dml2_printf("DML::%s: j=%u, DETBufferSizeInKByte[%u] increases from %u ", __func__, j, l->NextSurfaceToAssignDETPiece, DETBufferSizeInKByte[l->NextSurfaceToAssignDETPiece]);
-#endif
+					DML_LOG_VERBOSE("DML::%s: j=%u, DETBufferSizePoolInKByte = %u\n", __func__, j, l->DETBufferSizePoolInKByte);
+					DML_LOG_VERBOSE("DML::%s: j=%u, NextSurfaceToAssignDETPiece = %u\n", __func__, j, l->NextSurfaceToAssignDETPiece);
+					DML_LOG_VERBOSE("DML::%s: j=%u, ReadBandwidthLuma[%u] = %f\n", __func__, j, l->NextSurfaceToAssignDETPiece, ReadBandwidthLuma[l->NextSurfaceToAssignDETPiece]);
+					DML_LOG_VERBOSE("DML::%s: j=%u, ReadBandwidthChroma[%u] = %f\n", __func__, j, l->NextSurfaceToAssignDETPiece, ReadBandwidthChroma[l->NextSurfaceToAssignDETPiece]);
+					DML_LOG_VERBOSE("DML::%s: j=%u, BandwidthOfSurfacesNotAssignedDETPiece = %f\n", __func__, j, l->BandwidthOfSurfacesNotAssignedDETPiece);
+					DML_LOG_VERBOSE("DML::%s: j=%u, NextDETBufferPieceInKByte = %u\n", __func__, j, l->NextDETBufferPieceInKByte);
+					DML_LOG_VERBOSE("DML::%s: j=%u, DETBufferSizeInKByte[%u] increases from %u ", __func__, j, l->NextSurfaceToAssignDETPiece, DETBufferSizeInKByte[l->NextSurfaceToAssignDETPiece]);
 
 					DETBufferSizeInKByte[l->NextSurfaceToAssignDETPiece] = DETBufferSizeInKByte[l->NextSurfaceToAssignDETPiece] + l->NextDETBufferPieceInKByte / (ForceSingleDPP ? 1 : DPPPerSurface[l->NextSurfaceToAssignDETPiece]);
-#ifdef __DML_VBA_DEBUG__
-					dml2_printf("to %u\n", DETBufferSizeInKByte[l->NextSurfaceToAssignDETPiece]);
-#endif
+					DML_LOG_VERBOSE("to %u\n", DETBufferSizeInKByte[l->NextSurfaceToAssignDETPiece]);
 
 					l->DETBufferSizePoolInKByte = l->DETBufferSizePoolInKByte - l->NextDETBufferPieceInKByte;
 					DETPieceAssignedToThisSurfaceAlready[l->NextSurfaceToAssignDETPiece] = true;
@@ -1274,13 +1229,11 @@ static void CalculateDETBufferSize(
 	}
 	*CompressedBufferSizeInkByte = *CompressedBufferSizeInkByte * CompressedBufferSegmentSizeInkByte / ConfigReturnBufferSegmentSizeInkByte;
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: --- After bandwidth adjustment ---\n", __func__);
-	dml2_printf("DML::%s: CompressedBufferSizeInkByte = %u\n", __func__, *CompressedBufferSizeInkByte);
+	DML_LOG_VERBOSE("DML::%s: --- After bandwidth adjustment ---\n", __func__);
+	DML_LOG_VERBOSE("DML::%s: CompressedBufferSizeInkByte = %u\n", __func__, *CompressedBufferSizeInkByte);
 	for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
-		dml2_printf("DML::%s: k=%u DETBufferSizeInKByte = %u (TotalReadBandWidth=%f)\n", __func__, k, DETBufferSizeInKByte[k], ReadBandwidthLuma[k] + ReadBandwidthChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeInKByte = %u (TotalReadBandWidth=%f)\n", __func__, k, DETBufferSizeInKByte[k], ReadBandwidthLuma[k] + ReadBandwidthChroma[k]);
 	}
-#endif
 }
 
 static double CalculateRequiredDispclk(
@@ -1510,15 +1463,13 @@ static unsigned int dscceComputeDelay(
 	//pixel delay is group_delay (converted to pixels) + pipeline, however, first group is a special case since it is processed as soon as it arrives (i.e., in 3 cycles regardless of pixel format)
 	pixels = (group_delay - 1) * cycles_per_group + 3 + pipeline_delay;
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: bpc: %u\n", __func__, bpc);
-	dml2_printf("DML::%s: BPP: %f\n", __func__, BPP);
-	dml2_printf("DML::%s: sliceWidth: %u\n", __func__, sliceWidth);
-	dml2_printf("DML::%s: numSlices: %u\n", __func__, numSlices);
-	dml2_printf("DML::%s: pixelFormat: %u\n", __func__, pixelFormat);
-	dml2_printf("DML::%s: Output: %u\n", __func__, Output);
-	dml2_printf("DML::%s: pixels: %u\n", __func__, pixels);
-#endif
+	DML_LOG_VERBOSE("DML::%s: bpc: %u\n", __func__, bpc);
+	DML_LOG_VERBOSE("DML::%s: BPP: %f\n", __func__, BPP);
+	DML_LOG_VERBOSE("DML::%s: sliceWidth: %u\n", __func__, sliceWidth);
+	DML_LOG_VERBOSE("DML::%s: numSlices: %u\n", __func__, numSlices);
+	DML_LOG_VERBOSE("DML::%s: pixelFormat: %u\n", __func__, pixelFormat);
+	DML_LOG_VERBOSE("DML::%s: Output: %u\n", __func__, Output);
+	DML_LOG_VERBOSE("DML::%s: pixels: %u\n", __func__, pixels);
 	return pixels;
 }
 
@@ -1593,10 +1544,8 @@ static unsigned int dscComputeDelay(enum dml2_output_format_class pixelFormat, e
 
 	// sft
 	Delay = Delay + 1;
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: pixelFormat = %u\n", __func__, pixelFormat);
-	dml2_printf("DML::%s: Delay = %u\n", __func__, Delay);
-#endif
+	DML_LOG_VERBOSE("DML::%s: pixelFormat = %u\n", __func__, pixelFormat);
+	DML_LOG_VERBOSE("DML::%s: Delay = %u\n", __func__, Delay);
 
 	return Delay;
 }
@@ -1667,10 +1616,8 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 	}
 
 	meta_surface_bytes = (unsigned int)(p->DCCMetaPitch * vp_height_meta_ub * p->BytePerPixel / 256.0);
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DCCMetaPitch = %u\n", __func__, p->DCCMetaPitch);
-	dml2_printf("DML::%s: meta_surface_bytes = %u\n", __func__, meta_surface_bytes);
-#endif
+	DML_LOG_VERBOSE("DML::%s: DCCMetaPitch = %u\n", __func__, p->DCCMetaPitch);
+	DML_LOG_VERBOSE("DML::%s: meta_surface_bytes = %u\n", __func__, meta_surface_bytes);
 	if (p->GPUVMEnable == true) {
 		double meta_vmpg_bytes = 4.0 * 1024.0;
 		*p->meta_pte_bytes_per_frame_ub = (unsigned int)((math_ceil2((double) (meta_surface_bytes - meta_vmpg_bytes) / (8 * meta_vmpg_bytes), 1) + 1) * 64);
@@ -1724,25 +1671,23 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 
 	vm_bytes = *p->meta_pte_bytes_per_frame_ub + extra_mpde_bytes + *p->dpde0_bytes_per_frame_ub + extra_dpde_bytes;
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DCCEnable = %u\n", __func__, p->DCCEnable);
-	dml2_printf("DML::%s: GPUVMEnable = %u\n", __func__, p->GPUVMEnable);
-	dml2_printf("DML::%s: SwModeLinear = %u\n", __func__, p->SurfaceTiling == dml2_sw_linear);
-	dml2_printf("DML::%s: BytePerPixel = %u\n", __func__, p->BytePerPixel);
-	dml2_printf("DML::%s: GPUVMMaxPageTableLevels = %u\n", __func__, p->GPUVMMaxPageTableLevels);
-	dml2_printf("DML::%s: BlockHeight256Bytes = %u\n", __func__, p->BlockHeight256Bytes);
-	dml2_printf("DML::%s: BlockWidth256Bytes = %u\n", __func__, p->BlockWidth256Bytes);
-	dml2_printf("DML::%s: MacroTileHeight = %u\n", __func__, p->MacroTileHeight);
-	dml2_printf("DML::%s: MacroTileWidth = %u\n", __func__, p->MacroTileWidth);
-	dml2_printf("DML::%s: meta_pte_bytes_per_frame_ub = %u\n", __func__, *p->meta_pte_bytes_per_frame_ub);
-	dml2_printf("DML::%s: dpde0_bytes_per_frame_ub = %u\n", __func__, *p->dpde0_bytes_per_frame_ub);
-	dml2_printf("DML::%s: extra_mpde_bytes = %u\n", __func__, extra_mpde_bytes);
-	dml2_printf("DML::%s: extra_dpde_bytes = %u\n", __func__, extra_dpde_bytes);
-	dml2_printf("DML::%s: vm_bytes = %u\n", __func__, vm_bytes);
-	dml2_printf("DML::%s: ViewportHeight = %u\n", __func__, p->ViewportHeight);
-	dml2_printf("DML::%s: SwathWidth = %u\n", __func__, p->SwathWidth);
-	dml2_printf("DML::%s: vp_height_dpte_ub = %u\n", __func__, vp_height_dpte_ub);
-#endif
+	DML_LOG_VERBOSE("DML::%s: DCCEnable = %u\n", __func__, p->DCCEnable);
+	DML_LOG_VERBOSE("DML::%s: GPUVMEnable = %u\n", __func__, p->GPUVMEnable);
+	DML_LOG_VERBOSE("DML::%s: SwModeLinear = %u\n", __func__, p->SurfaceTiling == dml2_sw_linear);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixel = %u\n", __func__, p->BytePerPixel);
+	DML_LOG_VERBOSE("DML::%s: GPUVMMaxPageTableLevels = %u\n", __func__, p->GPUVMMaxPageTableLevels);
+	DML_LOG_VERBOSE("DML::%s: BlockHeight256Bytes = %u\n", __func__, p->BlockHeight256Bytes);
+	DML_LOG_VERBOSE("DML::%s: BlockWidth256Bytes = %u\n", __func__, p->BlockWidth256Bytes);
+	DML_LOG_VERBOSE("DML::%s: MacroTileHeight = %u\n", __func__, p->MacroTileHeight);
+	DML_LOG_VERBOSE("DML::%s: MacroTileWidth = %u\n", __func__, p->MacroTileWidth);
+	DML_LOG_VERBOSE("DML::%s: meta_pte_bytes_per_frame_ub = %u\n", __func__, *p->meta_pte_bytes_per_frame_ub);
+	DML_LOG_VERBOSE("DML::%s: dpde0_bytes_per_frame_ub = %u\n", __func__, *p->dpde0_bytes_per_frame_ub);
+	DML_LOG_VERBOSE("DML::%s: extra_mpde_bytes = %u\n", __func__, extra_mpde_bytes);
+	DML_LOG_VERBOSE("DML::%s: extra_dpde_bytes = %u\n", __func__, extra_dpde_bytes);
+	DML_LOG_VERBOSE("DML::%s: vm_bytes = %u\n", __func__, vm_bytes);
+	DML_LOG_VERBOSE("DML::%s: ViewportHeight = %u\n", __func__, p->ViewportHeight);
+	DML_LOG_VERBOSE("DML::%s: SwathWidth = %u\n", __func__, p->SwathWidth);
+	DML_LOG_VERBOSE("DML::%s: vp_height_dpte_ub = %u\n", __func__, vp_height_dpte_ub);
 
 	if (p->SurfaceTiling == dml2_sw_linear) {
 		*p->PixelPTEReqHeight = 1;
@@ -1778,22 +1723,20 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 		*p->vmpg_width = 1024 * p->GPUVMMinPageSizeKBytes / (p->MacroTileHeight * p->BytePerPixel);
 
 		if (p->GPUVMEnable == true) {
-			dml2_printf("DML::%s: GPUVMMinPageSizeKBytes=%u and sw_mode=%u (tile_size=%d) not supported!\n",
+			DML_LOG_VERBOSE("DML::%s: GPUVMMinPageSizeKBytes=%u and sw_mode=%u (tile_size=%d) not supported!\n",
 				__func__, p->GPUVMMinPageSizeKBytes, p->SurfaceTiling, dml_get_tile_block_size_bytes(p->SurfaceTiling));
-			DML2_ASSERT(0);
+			DML_ASSERT(0);
 		}
 	}
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: GPUVMMinPageSizeKBytes = %u\n", __func__, p->GPUVMMinPageSizeKBytes);
-	dml2_printf("DML::%s: PixelPTEReqHeight = %u\n", __func__, *p->PixelPTEReqHeight);
-	dml2_printf("DML::%s: PixelPTEReqWidth = %u\n", __func__, *p->PixelPTEReqWidth);
-	dml2_printf("DML::%s: PixelPTEReqWidth_linear = %u\n", __func__, PixelPTEReqWidth_linear);
-	dml2_printf("DML::%s: PTERequestSize = %u\n", __func__, *p->PTERequestSize);
-	dml2_printf("DML::%s: Pitch = %u\n", __func__, p->Pitch);
-	dml2_printf("DML::%s: vmpg_width = %u\n", __func__, *p->vmpg_width);
-	dml2_printf("DML::%s: vmpg_height = %u\n", __func__, *p->vmpg_height);
-#endif
+	DML_LOG_VERBOSE("DML::%s: GPUVMMinPageSizeKBytes = %u\n", __func__, p->GPUVMMinPageSizeKBytes);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEReqHeight = %u\n", __func__, *p->PixelPTEReqHeight);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEReqWidth = %u\n", __func__, *p->PixelPTEReqWidth);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEReqWidth_linear = %u\n", __func__, PixelPTEReqWidth_linear);
+	DML_LOG_VERBOSE("DML::%s: PTERequestSize = %u\n", __func__, *p->PTERequestSize);
+	DML_LOG_VERBOSE("DML::%s: Pitch = %u\n", __func__, p->Pitch);
+	DML_LOG_VERBOSE("DML::%s: vmpg_width = %u\n", __func__, *p->vmpg_width);
+	DML_LOG_VERBOSE("DML::%s: vmpg_height = %u\n", __func__, *p->vmpg_height);
 
 	*p->dpte_row_height_one_row_per_frame = vp_height_dpte_ub;
 	*p->dpte_row_width_ub_one_row_per_frame = (unsigned int)((math_ceil2(((double)p->Pitch * (double)*p->dpte_row_height_one_row_per_frame / (double)*p->PixelPTEReqHeight - 1) / (double)*p->PixelPTEReqWidth, 1) + 1) * (double)*p->PixelPTEReqWidth);
@@ -1811,7 +1754,7 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 			*p->dpte_row_height_linear = 128;
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: dpte_row_width_ub = %u (linear)\n", __func__, *p->dpte_row_width_ub);
+		DML_LOG_VERBOSE("DML::%s: dpte_row_width_ub = %u (linear)\n", __func__, *p->dpte_row_width_ub);
 #endif
 
 	} else if (!dml_is_vertical_rotation(p->RotationAngle)) {
@@ -1825,7 +1768,7 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 			*p->dpte_row_width_ub = (unsigned int)((math_ceil2((double)(p->SwathWidth - 1) / (double)*p->PixelPTEReqWidth, 1) + 1.0) * *p->PixelPTEReqWidth);
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: dpte_row_width_ub = %u (tiled horz)\n", __func__, *p->dpte_row_width_ub);
+		DML_LOG_VERBOSE("DML::%s: dpte_row_width_ub = %u (tiled horz)\n", __func__, *p->dpte_row_width_ub);
 #endif
 
 		*p->PixelPTEBytesPerRow = *p->dpte_row_width_ub / *p->PixelPTEReqWidth * *p->PTERequestSize;
@@ -1840,7 +1783,7 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 
 		*p->PixelPTEBytesPerRow = (unsigned int)((double)*p->dpte_row_width_ub / (double)*p->PixelPTEReqHeight * *p->PTERequestSize);
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: dpte_row_width_ub = %u (tiled vert)\n", __func__, *p->dpte_row_width_ub);
+		DML_LOG_VERBOSE("DML::%s: dpte_row_width_ub = %u (tiled vert)\n", __func__, *p->dpte_row_width_ub);
 #endif
 	}
 
@@ -1852,18 +1795,18 @@ static unsigned int CalculateVMAndRowBytes(struct dml2_core_shared_calculate_vm_
 	*p->PixelPTEBytesPerRowStorage = *p->PixelPTEBytesPerRow;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: GPUVMMinPageSizeKBytes = %u\n", __func__, p->GPUVMMinPageSizeKBytes);
-	dml2_printf("DML::%s: GPUVMEnable = %u\n", __func__, p->GPUVMEnable);
-	dml2_printf("DML::%s: meta_row_height = %u\n", __func__, *p->meta_row_height);
-	dml2_printf("DML::%s: dpte_row_height = %u\n", __func__, *p->dpte_row_height);
-	dml2_printf("DML::%s: dpte_row_height_linear = %u\n", __func__, *p->dpte_row_height_linear);
-	dml2_printf("DML::%s: dpte_row_width_ub = %u\n", __func__, *p->dpte_row_width_ub);
-	dml2_printf("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, *p->PixelPTEBytesPerRow);
-	dml2_printf("DML::%s: PixelPTEBytesPerRowStorage = %u\n", __func__, *p->PixelPTEBytesPerRowStorage);
-	dml2_printf("DML::%s: PTEBufferSizeInRequests = %u\n", __func__, p->PTEBufferSizeInRequests);
-	dml2_printf("DML::%s: dpte_row_height_one_row_per_frame = %u\n", __func__, *p->dpte_row_height_one_row_per_frame);
-	dml2_printf("DML::%s: dpte_row_width_ub_one_row_per_frame = %u\n", __func__, *p->dpte_row_width_ub_one_row_per_frame);
-	dml2_printf("DML::%s: PixelPTEBytesPerRow_one_row_per_frame = %u\n", __func__, *p->PixelPTEBytesPerRow_one_row_per_frame);
+	DML_LOG_VERBOSE("DML::%s: GPUVMMinPageSizeKBytes = %u\n", __func__, p->GPUVMMinPageSizeKBytes);
+	DML_LOG_VERBOSE("DML::%s: GPUVMEnable = %u\n", __func__, p->GPUVMEnable);
+	DML_LOG_VERBOSE("DML::%s: meta_row_height = %u\n", __func__, *p->meta_row_height);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_height = %u\n", __func__, *p->dpte_row_height);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_height_linear = %u\n", __func__, *p->dpte_row_height_linear);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_width_ub = %u\n", __func__, *p->dpte_row_width_ub);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, *p->PixelPTEBytesPerRow);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEBytesPerRowStorage = %u\n", __func__, *p->PixelPTEBytesPerRowStorage);
+	DML_LOG_VERBOSE("DML::%s: PTEBufferSizeInRequests = %u\n", __func__, p->PTEBufferSizeInRequests);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_height_one_row_per_frame = %u\n", __func__, *p->dpte_row_height_one_row_per_frame);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_width_ub_one_row_per_frame = %u\n", __func__, *p->dpte_row_width_ub_one_row_per_frame);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEBytesPerRow_one_row_per_frame = %u\n", __func__, *p->PixelPTEBytesPerRow_one_row_per_frame);
 #endif
 
 	return vm_bytes;
@@ -1894,12 +1837,12 @@ static unsigned int CalculatePrefetchSourceLines(
 	double numLines = 0;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: VRatio = %f\n", __func__, VRatio);
-	dml2_printf("DML::%s: VTaps = %u\n", __func__, VTaps);
-	dml2_printf("DML::%s: ViewportXStart = %u\n", __func__, ViewportXStart);
-	dml2_printf("DML::%s: ViewportYStart = %u\n", __func__, ViewportYStart);
-	dml2_printf("DML::%s: ViewportStationary = %u\n", __func__, ViewportStationary);
-	dml2_printf("DML::%s: SwathHeight = %u\n", __func__, SwathHeight);
+	DML_LOG_VERBOSE("DML::%s: VRatio = %f\n", __func__, VRatio);
+	DML_LOG_VERBOSE("DML::%s: VTaps = %u\n", __func__, VTaps);
+	DML_LOG_VERBOSE("DML::%s: ViewportXStart = %u\n", __func__, ViewportXStart);
+	DML_LOG_VERBOSE("DML::%s: ViewportYStart = %u\n", __func__, ViewportYStart);
+	DML_LOG_VERBOSE("DML::%s: ViewportStationary = %u\n", __func__, ViewportStationary);
+	DML_LOG_VERBOSE("DML::%s: SwathHeight = %u\n", __func__, SwathHeight);
 #endif
 	if (ProgressiveToInterlaceUnitInOPP)
 		*VInitPreFill = (unsigned int)(math_floor2((VRatio + (double)VTaps + 1) / 2.0, 1));
@@ -1934,11 +1877,11 @@ static unsigned int CalculatePrefetchSourceLines(
 	numLines = *MaxNumSwath * SwathHeight + MaxPartialSwath;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: vp_start_rot = %u\n", __func__, vp_start_rot);
-	dml2_printf("DML::%s: VInitPreFill = %u\n", __func__, *VInitPreFill);
-	dml2_printf("DML::%s: MaxPartialSwath = %u\n", __func__, MaxPartialSwath);
-	dml2_printf("DML::%s: MaxNumSwath = %u\n", __func__, *MaxNumSwath);
-	dml2_printf("DML::%s: Prefetch source lines = %3.2f\n", __func__, numLines);
+	DML_LOG_VERBOSE("DML::%s: vp_start_rot = %u\n", __func__, vp_start_rot);
+	DML_LOG_VERBOSE("DML::%s: VInitPreFill = %u\n", __func__, *VInitPreFill);
+	DML_LOG_VERBOSE("DML::%s: MaxPartialSwath = %u\n", __func__, MaxPartialSwath);
+	DML_LOG_VERBOSE("DML::%s: MaxNumSwath = %u\n", __func__, *MaxNumSwath);
+	DML_LOG_VERBOSE("DML::%s: Prefetch source lines = %3.2f\n", __func__, numLines);
 #endif
 	return (unsigned int)(numLines);
 
@@ -2007,8 +1950,8 @@ static void CalculateMALLUseForStaticScreen(
 		if (is_using_mall_for_ss[k])
 			TotalSurfaceSizeInMALL = TotalSurfaceSizeInMALL + SurfaceSizeInMALL[k];
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, is_using_mall_for_ss = %u\n", __func__, k, is_using_mall_for_ss[k]);
-		dml2_printf("DML::%s: k=%u, TotalSurfaceSizeInMALL = %u\n", __func__, k, TotalSurfaceSizeInMALL);
+		DML_LOG_VERBOSE("DML::%s: k=%u, is_using_mall_for_ss = %u\n", __func__, k, is_using_mall_for_ss[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, TotalSurfaceSizeInMALL = %u\n", __func__, k, TotalSurfaceSizeInMALL);
 #endif
 	}
 
@@ -2022,7 +1965,7 @@ static void CalculateMALLUseForStaticScreen(
 				(!CanAddAnotherSurfaceToMALL || SurfaceSizeInMALL[k] < SurfaceSizeInMALL[SurfaceToAddToMALL])) {
 				CanAddAnotherSurfaceToMALL = true;
 				SurfaceToAddToMALL = k;
-				dml2_printf("DML::%s: k=%u, UseMALLForStaticScreen = %u (dis, en, optimize)\n", __func__, k, display_cfg->plane_descriptors[k].overrides.refresh_from_mall);
+				DML_LOG_VERBOSE("DML::%s: k=%u, UseMALLForStaticScreen = %u (dis, en, optimize)\n", __func__, k, display_cfg->plane_descriptors[k].overrides.refresh_from_mall);
 			}
 		}
 		if (CanAddAnotherSurfaceToMALL) {
@@ -2030,8 +1973,8 @@ static void CalculateMALLUseForStaticScreen(
 			TotalSurfaceSizeInMALL = TotalSurfaceSizeInMALL + SurfaceSizeInMALL[SurfaceToAddToMALL];
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: SurfaceToAddToMALL = %u\n", __func__, SurfaceToAddToMALL);
-			dml2_printf("DML::%s: TotalSurfaceSizeInMALL = %u\n", __func__, TotalSurfaceSizeInMALL);
+			DML_LOG_VERBOSE("DML::%s: SurfaceToAddToMALL = %u\n", __func__, SurfaceToAddToMALL);
+			DML_LOG_VERBOSE("DML::%s: TotalSurfaceSizeInMALL = %u\n", __func__, TotalSurfaceSizeInMALL);
 #endif
 		}
 	}
@@ -2203,15 +2146,15 @@ static void CalculateDCCConfiguration(
 		segment_order_vert_contiguous_chroma = 0;
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DCCEnabled = %u\n", __func__, DCCEnabled);
-	dml2_printf("DML::%s: nomDETInKByte = %u\n", __func__, nomDETInKByte);
-	dml2_printf("DML::%s: DETBufferSizeForDCC = %u\n", __func__, DETBufferSizeForDCC);
-	dml2_printf("DML::%s: req128_horz_wc_l = %u\n", __func__, req128_horz_wc_l);
-	dml2_printf("DML::%s: req128_horz_wc_c = %u\n", __func__, req128_horz_wc_c);
-	dml2_printf("DML::%s: full_swath_bytes_horz_wc_l = %u\n", __func__, full_swath_bytes_horz_wc_l);
-	dml2_printf("DML::%s: full_swath_bytes_vert_wc_c = %u\n", __func__, full_swath_bytes_vert_wc_c);
-	dml2_printf("DML::%s: segment_order_horz_contiguous_luma = %u\n", __func__, segment_order_horz_contiguous_luma);
-	dml2_printf("DML::%s: segment_order_horz_contiguous_chroma = %u\n", __func__, segment_order_horz_contiguous_chroma);
+	DML_LOG_VERBOSE("DML::%s: DCCEnabled = %u\n", __func__, DCCEnabled);
+	DML_LOG_VERBOSE("DML::%s: nomDETInKByte = %u\n", __func__, nomDETInKByte);
+	DML_LOG_VERBOSE("DML::%s: DETBufferSizeForDCC = %u\n", __func__, DETBufferSizeForDCC);
+	DML_LOG_VERBOSE("DML::%s: req128_horz_wc_l = %u\n", __func__, req128_horz_wc_l);
+	DML_LOG_VERBOSE("DML::%s: req128_horz_wc_c = %u\n", __func__, req128_horz_wc_c);
+	DML_LOG_VERBOSE("DML::%s: full_swath_bytes_horz_wc_l = %u\n", __func__, full_swath_bytes_horz_wc_l);
+	DML_LOG_VERBOSE("DML::%s: full_swath_bytes_vert_wc_c = %u\n", __func__, full_swath_bytes_vert_wc_c);
+	DML_LOG_VERBOSE("DML::%s: segment_order_horz_contiguous_luma = %u\n", __func__, segment_order_horz_contiguous_luma);
+	DML_LOG_VERBOSE("DML::%s: segment_order_horz_contiguous_chroma = %u\n", __func__, segment_order_horz_contiguous_chroma);
 #endif
 	if (DCCProgrammingAssumesScanDirectionUnknown == true) {
 		if (req128_horz_wc_l == 0 && req128_vert_wc_l == 0) {
@@ -2301,12 +2244,12 @@ static void CalculateDCCConfiguration(
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: MaxUncompressedBlockLuma = %u\n", __func__, *MaxUncompressedBlockLuma);
-	dml2_printf("DML::%s: MaxCompressedBlockLuma = %u\n", __func__, *MaxCompressedBlockLuma);
-	dml2_printf("DML::%s: IndependentBlockLuma = %u\n", __func__, *IndependentBlockLuma);
-	dml2_printf("DML::%s: MaxUncompressedBlockChroma = %u\n", __func__, *MaxUncompressedBlockChroma);
-	dml2_printf("DML::%s: MaxCompressedBlockChroma = %u\n", __func__, *MaxCompressedBlockChroma);
-	dml2_printf("DML::%s: IndependentBlockChroma = %u\n", __func__, *IndependentBlockChroma);
+	DML_LOG_VERBOSE("DML::%s: MaxUncompressedBlockLuma = %u\n", __func__, *MaxUncompressedBlockLuma);
+	DML_LOG_VERBOSE("DML::%s: MaxCompressedBlockLuma = %u\n", __func__, *MaxCompressedBlockLuma);
+	DML_LOG_VERBOSE("DML::%s: IndependentBlockLuma = %u\n", __func__, *IndependentBlockLuma);
+	DML_LOG_VERBOSE("DML::%s: MaxUncompressedBlockChroma = %u\n", __func__, *MaxUncompressedBlockChroma);
+	DML_LOG_VERBOSE("DML::%s: MaxCompressedBlockChroma = %u\n", __func__, *MaxCompressedBlockChroma);
+	DML_LOG_VERBOSE("DML::%s: IndependentBlockChroma = %u\n", __func__, *IndependentBlockChroma);
 #endif
 
 }
@@ -2326,26 +2269,26 @@ static void calculate_mcache_row_bytes(
 	unsigned int mvmpg_per_mcache;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: num_chans = %u\n", __func__, p->num_chans);
-	dml2_printf("DML::%s: mem_word_bytes = %u\n", __func__, p->mem_word_bytes);
-	dml2_printf("DML::%s: mcache_line_size_bytes = %u\n", __func__, p->mcache_line_size_bytes);
-	dml2_printf("DML::%s: mcache_size_bytes = %u\n", __func__, p->mcache_size_bytes);
-	dml2_printf("DML::%s: gpuvm_enable = %u\n", __func__, p->gpuvm_enable);
-	dml2_printf("DML::%s: gpuvm_page_size_kbytes = %u\n", __func__, p->gpuvm_page_size_kbytes);
-	dml2_printf("DML::%s: vp_stationary = %u\n", __func__, p->vp_stationary);
-	dml2_printf("DML::%s: tiling_mode = %u\n", __func__, p->tiling_mode);
-	dml2_printf("DML::%s: vp_start_x = %u\n", __func__, p->vp_start_x);
-	dml2_printf("DML::%s: vp_start_y = %u\n", __func__, p->vp_start_y);
-	dml2_printf("DML::%s: full_vp_width = %u\n", __func__, p->full_vp_width);
-	dml2_printf("DML::%s: full_vp_height = %u\n", __func__, p->full_vp_height);
-	dml2_printf("DML::%s: blk_width = %u\n", __func__, p->blk_width);
-	dml2_printf("DML::%s: blk_height = %u\n", __func__, p->blk_height);
-	dml2_printf("DML::%s: vmpg_width = %u\n", __func__, p->vmpg_width);
-	dml2_printf("DML::%s: vmpg_height = %u\n", __func__, p->vmpg_height);
-	dml2_printf("DML::%s: full_swath_bytes = %u\n", __func__, p->full_swath_bytes);
-#endif
-	DML2_ASSERT(p->mcache_line_size_bytes != 0);
-	DML2_ASSERT(p->mcache_size_bytes != 0);
+	DML_LOG_VERBOSE("DML::%s: num_chans = %u\n", __func__, p->num_chans);
+	DML_LOG_VERBOSE("DML::%s: mem_word_bytes = %u\n", __func__, p->mem_word_bytes);
+	DML_LOG_VERBOSE("DML::%s: mcache_line_size_bytes = %u\n", __func__, p->mcache_line_size_bytes);
+	DML_LOG_VERBOSE("DML::%s: mcache_size_bytes = %u\n", __func__, p->mcache_size_bytes);
+	DML_LOG_VERBOSE("DML::%s: gpuvm_enable = %u\n", __func__, p->gpuvm_enable);
+	DML_LOG_VERBOSE("DML::%s: gpuvm_page_size_kbytes = %u\n", __func__, p->gpuvm_page_size_kbytes);
+	DML_LOG_VERBOSE("DML::%s: vp_stationary = %u\n", __func__, p->vp_stationary);
+	DML_LOG_VERBOSE("DML::%s: tiling_mode = %u\n", __func__, p->tiling_mode);
+	DML_LOG_VERBOSE("DML::%s: vp_start_x = %u\n", __func__, p->vp_start_x);
+	DML_LOG_VERBOSE("DML::%s: vp_start_y = %u\n", __func__, p->vp_start_y);
+	DML_LOG_VERBOSE("DML::%s: full_vp_width = %u\n", __func__, p->full_vp_width);
+	DML_LOG_VERBOSE("DML::%s: full_vp_height = %u\n", __func__, p->full_vp_height);
+	DML_LOG_VERBOSE("DML::%s: blk_width = %u\n", __func__, p->blk_width);
+	DML_LOG_VERBOSE("DML::%s: blk_height = %u\n", __func__, p->blk_height);
+	DML_LOG_VERBOSE("DML::%s: vmpg_width = %u\n", __func__, p->vmpg_width);
+	DML_LOG_VERBOSE("DML::%s: vmpg_height = %u\n", __func__, p->vmpg_height);
+	DML_LOG_VERBOSE("DML::%s: full_swath_bytes = %u\n", __func__, p->full_swath_bytes);
+#endif
+	DML_ASSERT(p->mcache_line_size_bytes != 0);
+	DML_ASSERT(p->mcache_size_bytes != 0);
 
 	*p->mvmpg_width = 0;
 	*p->mvmpg_height = 0;
@@ -2370,8 +2313,8 @@ static void calculate_mcache_row_bytes(
 				*p->mvmpg_width = p->vmpg_width;
 				*p->mvmpg_height = p->vmpg_height;
 			} else if (!((blk_bytes == 65536) && (vmpg_bytes == 4096))) {
-				dml2_printf("ERROR: DML::%s: Tiling size and vm page size combination not supported\n", __func__);
-				DML2_ASSERT(0);
+				DML_LOG_VERBOSE("ERROR: DML::%s: Tiling size and vm page size combination not supported\n", __func__);
+				DML_ASSERT(0);
 			}
 		}
 
@@ -2439,25 +2382,25 @@ static void calculate_mcache_row_bytes(
 		*p->mvmpg_per_mcache_lb = (unsigned int)math_floor2(mvmpg_per_mcache, 1);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: gpuvm_enable = %u\n", __func__, p->gpuvm_enable);
-		dml2_printf("DML::%s: vmpg_bytes = %u\n", __func__, vmpg_bytes);
-		dml2_printf("DML::%s: blk_bytes = %u\n", __func__, blk_bytes);
-		dml2_printf("DML::%s: meta_per_mvmpg_per_channel = %f\n", __func__, meta_per_mvmpg_per_channel);
-		dml2_printf("DML::%s: mvmpg_per_row_ub = %u\n", __func__, mvmpg_per_row_ub);
-		dml2_printf("DML::%s: meta_row_width_ub = %u\n", __func__, *p->meta_row_width_ub);
-		dml2_printf("DML::%s: mvmpg_width = %u\n", __func__, *p->mvmpg_width);
-		dml2_printf("DML::%s: mvmpg_height = %u\n", __func__, *p->mvmpg_height);
-		dml2_printf("DML::%s: dcc_dram_bw_nom_overhead_factor = %f\n", __func__, *p->dcc_dram_bw_nom_overhead_factor);
-		dml2_printf("DML::%s: dcc_dram_bw_pref_overhead_factor = %f\n", __func__, *p->dcc_dram_bw_pref_overhead_factor);
+		DML_LOG_VERBOSE("DML::%s: gpuvm_enable = %u\n", __func__, p->gpuvm_enable);
+		DML_LOG_VERBOSE("DML::%s: vmpg_bytes = %u\n", __func__, vmpg_bytes);
+		DML_LOG_VERBOSE("DML::%s: blk_bytes = %u\n", __func__, blk_bytes);
+		DML_LOG_VERBOSE("DML::%s: meta_per_mvmpg_per_channel = %f\n", __func__, meta_per_mvmpg_per_channel);
+		DML_LOG_VERBOSE("DML::%s: mvmpg_per_row_ub = %u\n", __func__, mvmpg_per_row_ub);
+		DML_LOG_VERBOSE("DML::%s: meta_row_width_ub = %u\n", __func__, *p->meta_row_width_ub);
+		DML_LOG_VERBOSE("DML::%s: mvmpg_width = %u\n", __func__, *p->mvmpg_width);
+		DML_LOG_VERBOSE("DML::%s: mvmpg_height = %u\n", __func__, *p->mvmpg_height);
+		DML_LOG_VERBOSE("DML::%s: dcc_dram_bw_nom_overhead_factor = %f\n", __func__, *p->dcc_dram_bw_nom_overhead_factor);
+		DML_LOG_VERBOSE("DML::%s: dcc_dram_bw_pref_overhead_factor = %f\n", __func__, *p->dcc_dram_bw_pref_overhead_factor);
 #endif
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: mcache_row_bytes = %u\n", __func__, *p->mcache_row_bytes);
-	dml2_printf("DML::%s: mcache_row_bytes_per_channel = %u\n", __func__, *p->mcache_row_bytes_per_channel);
-	dml2_printf("DML::%s: num_mcaches = %u\n", __func__, *p->num_mcaches);
+	DML_LOG_VERBOSE("DML::%s: mcache_row_bytes = %u\n", __func__, *p->mcache_row_bytes);
+	DML_LOG_VERBOSE("DML::%s: mcache_row_bytes_per_channel = %u\n", __func__, *p->mcache_row_bytes_per_channel);
+	DML_LOG_VERBOSE("DML::%s: num_mcaches = %u\n", __func__, *p->num_mcaches);
 #endif
-	DML2_ASSERT(*p->num_mcaches > 0);
+	DML_ASSERT(*p->num_mcaches > 0);
 }
 
 static void calculate_mcache_setting(
@@ -2523,7 +2466,7 @@ static void calculate_mcache_setting(
 	l->l_p.mvmpg_per_mcache_lb = &l->mvmpg_per_mcache_lb_l;
 
 	calculate_mcache_row_bytes(scratch, &l->l_p);
-	DML2_ASSERT(*p->num_mcaches_l > 0);
+	DML_ASSERT(*p->num_mcaches_l > 0);
 
 	if (l->is_dual_plane) {
 		l->c_p.num_chans = p->num_chans;
@@ -2559,7 +2502,7 @@ static void calculate_mcache_setting(
 		l->c_p.mvmpg_per_mcache_lb = &l->mvmpg_per_mcache_lb_c;
 
 		calculate_mcache_row_bytes(scratch, &l->c_p);
-		DML2_ASSERT(*p->num_mcaches_c > 0);
+		DML_ASSERT(*p->num_mcaches_c > 0);
 	}
 
 	// Sharing for iMALL access
@@ -2598,28 +2541,28 @@ static void calculate_mcache_setting(
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: imall_enable = %u\n", __func__, p->imall_enable);
-	dml2_printf("DML::%s: is_dual_plane = %u\n", __func__, l->is_dual_plane);
-	dml2_printf("DML::%s: surf_vert = %u\n", __func__, p->surf_vert);
-	dml2_printf("DML::%s: mvmpg_width_l = %u\n", __func__, l->mvmpg_width_l);
-	dml2_printf("DML::%s: mvmpg_height_l = %u\n", __func__, l->mvmpg_height_l);
-	dml2_printf("DML::%s: mcache_remainder_l = %f\n", __func__, l->mcache_remainder_l);
-	dml2_printf("DML::%s: num_mcaches_l = %u\n", __func__, *p->num_mcaches_l);
-	dml2_printf("DML::%s: avg_mcache_element_size_l = %u\n", __func__, l->avg_mcache_element_size_l);
-	dml2_printf("DML::%s: mvmpg_access_width_l = %u\n", __func__, l->mvmpg_access_width_l);
-	dml2_printf("DML::%s: mall_comb_mcache_l = %u\n", __func__, *p->mall_comb_mcache_l);
+	DML_LOG_VERBOSE("DML::%s: imall_enable = %u\n", __func__, p->imall_enable);
+	DML_LOG_VERBOSE("DML::%s: is_dual_plane = %u\n", __func__, l->is_dual_plane);
+	DML_LOG_VERBOSE("DML::%s: surf_vert = %u\n", __func__, p->surf_vert);
+	DML_LOG_VERBOSE("DML::%s: mvmpg_width_l = %u\n", __func__, l->mvmpg_width_l);
+	DML_LOG_VERBOSE("DML::%s: mvmpg_height_l = %u\n", __func__, l->mvmpg_height_l);
+	DML_LOG_VERBOSE("DML::%s: mcache_remainder_l = %f\n", __func__, l->mcache_remainder_l);
+	DML_LOG_VERBOSE("DML::%s: num_mcaches_l = %u\n", __func__, *p->num_mcaches_l);
+	DML_LOG_VERBOSE("DML::%s: avg_mcache_element_size_l = %u\n", __func__, l->avg_mcache_element_size_l);
+	DML_LOG_VERBOSE("DML::%s: mvmpg_access_width_l = %u\n", __func__, l->mvmpg_access_width_l);
+	DML_LOG_VERBOSE("DML::%s: mall_comb_mcache_l = %u\n", __func__, *p->mall_comb_mcache_l);
 
 	if (l->is_dual_plane) {
-		dml2_printf("DML::%s: mvmpg_width_c = %u\n", __func__, l->mvmpg_width_c);
-		dml2_printf("DML::%s: mvmpg_height_c = %u\n", __func__, l->mvmpg_height_c);
-		dml2_printf("DML::%s: mcache_remainder_c = %f\n", __func__, l->mcache_remainder_c);
-		dml2_printf("DML::%s: luma_time_factor = %f\n", __func__, l->luma_time_factor);
-		dml2_printf("DML::%s: num_mcaches_c = %u\n", __func__, *p->num_mcaches_c);
-		dml2_printf("DML::%s: avg_mcache_element_size_c = %u\n", __func__, l->avg_mcache_element_size_c);
-		dml2_printf("DML::%s: mvmpg_access_width_c = %u\n", __func__, l->mvmpg_access_width_c);
-		dml2_printf("DML::%s: mall_comb_mcache_c = %u\n", __func__, *p->mall_comb_mcache_c);
-		dml2_printf("DML::%s: lc_comb_last_mcache_size = %u\n", __func__, l->lc_comb_last_mcache_size);
-		dml2_printf("DML::%s: lc_comb_mcache = %u\n", __func__, *p->lc_comb_mcache);
+		DML_LOG_VERBOSE("DML::%s: mvmpg_width_c = %u\n", __func__, l->mvmpg_width_c);
+		DML_LOG_VERBOSE("DML::%s: mvmpg_height_c = %u\n", __func__, l->mvmpg_height_c);
+		DML_LOG_VERBOSE("DML::%s: mcache_remainder_c = %f\n", __func__, l->mcache_remainder_c);
+		DML_LOG_VERBOSE("DML::%s: luma_time_factor = %f\n", __func__, l->luma_time_factor);
+		DML_LOG_VERBOSE("DML::%s: num_mcaches_c = %u\n", __func__, *p->num_mcaches_c);
+		DML_LOG_VERBOSE("DML::%s: avg_mcache_element_size_c = %u\n", __func__, l->avg_mcache_element_size_c);
+		DML_LOG_VERBOSE("DML::%s: mvmpg_access_width_c = %u\n", __func__, l->mvmpg_access_width_c);
+		DML_LOG_VERBOSE("DML::%s: mall_comb_mcache_c = %u\n", __func__, *p->mall_comb_mcache_c);
+		DML_LOG_VERBOSE("DML::%s: lc_comb_last_mcache_size = %u\n", __func__, l->lc_comb_last_mcache_size);
+		DML_LOG_VERBOSE("DML::%s: lc_comb_mcache = %u\n", __func__, *p->lc_comb_mcache);
 	}
 #endif
 	// calculate split_coordinate
@@ -2639,11 +2582,11 @@ static void calculate_mcache_setting(
 	}
 #ifdef __DML_VBA_DEBUG__
 	for (n = 0; n < *p->num_mcaches_l; n++)
-		dml2_printf("DML::%s: mcache_offsets_l[%u] = %u\n", __func__, n, p->mcache_offsets_l[n]);
+		DML_LOG_VERBOSE("DML::%s: mcache_offsets_l[%u] = %u\n", __func__, n, p->mcache_offsets_l[n]);
 
 	if (l->is_dual_plane) {
 		for (n = 0; n < *p->num_mcaches_c; n++)
-			dml2_printf("DML::%s: mcache_offsets_c[%u] = %u\n", __func__, n, p->mcache_offsets_c[n]);
+			DML_LOG_VERBOSE("DML::%s: mcache_offsets_c[%u] = %u\n", __func__, n, p->mcache_offsets_c[n]);
 	}
 #endif
 
@@ -2660,10 +2603,10 @@ static void calculate_mcache_setting(
 
 #ifdef __DML_VBA_DEBUG__
 		for (n = 0; n < *p->num_mcaches_l; n++)
-			dml2_printf("DML::%s: mcache_offsets_l[%u] = %u\n", __func__, n, p->mcache_offsets_l[n]);
+			DML_LOG_VERBOSE("DML::%s: mcache_offsets_l[%u] = %u\n", __func__, n, p->mcache_offsets_l[n]);
 
 		for (n = 0; n < *p->num_mcaches_c; n++)
-			dml2_printf("DML::%s: mcache_offsets_c[%u] = %u\n", __func__, n, p->mcache_offsets_c[n]);
+			DML_LOG_VERBOSE("DML::%s: mcache_offsets_c[%u] = %u\n", __func__, n, p->mcache_offsets_c[n]);
 #endif
 	}
 
@@ -2694,8 +2637,8 @@ static void calculate_mall_bw_overhead_factor(
 			mall_prefetch_dram_overhead_factor[k] = 2.0;
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, mall_prefetch_sdp_overhead_factor = %f\n", __func__, k, mall_prefetch_sdp_overhead_factor[k]);
-		dml2_printf("DML::%s: k=%u, mall_prefetch_dram_overhead_factor = %f\n", __func__, k, mall_prefetch_dram_overhead_factor[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, mall_prefetch_sdp_overhead_factor = %f\n", __func__, k, mall_prefetch_sdp_overhead_factor[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, mall_prefetch_dram_overhead_factor = %f\n", __func__, k, mall_prefetch_dram_overhead_factor[k]);
 #endif
 	}
 }
@@ -2772,22 +2715,20 @@ static double dml_get_return_bandwidth_available(
 	else // dml2_core_internal_bw_dram
 		return_bw_mbps = derate_dram_bandwidth;
 
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: is_avg_bw = %u\n", __func__, is_avg_bw);
-	dml2_printf("DML::%s: is_hvm_en = %u\n", __func__, is_hvm_en);
-	dml2_printf("DML::%s: is_hvm_only = %u\n", __func__, is_hvm_only);
-	dml2_printf("DML::%s: state_type = %s\n", __func__, dml2_core_internal_soc_state_type_str(state_type));
-	dml2_printf("DML::%s: bw_type = %s\n", __func__, dml2_core_internal_bw_type_str(bw_type));
-	dml2_printf("DML::%s: dcfclk_mhz = %f\n", __func__, dcfclk_mhz);
-	dml2_printf("DML::%s: fclk_mhz = %f\n", __func__, fclk_mhz);
-	dml2_printf("DML::%s: ideal_sdp_bandwidth = %f\n", __func__, ideal_sdp_bandwidth);
-	dml2_printf("DML::%s: ideal_fabric_bandwidth = %f\n", __func__, ideal_fabric_bandwidth);
-	dml2_printf("DML::%s: ideal_dram_bandwidth = %f\n", __func__, ideal_dram_bandwidth);
-	dml2_printf("DML::%s: derate_sdp_bandwidth = %f (derate %f)\n", __func__, derate_sdp_bandwidth, derate_sdp_factor);
-	dml2_printf("DML::%s: derate_fabric_bandwidth = %f (derate %f)\n", __func__, derate_fabric_bandwidth, derate_fabric_factor);
-	dml2_printf("DML::%s: derate_dram_bandwidth = %f (derate %f)\n", __func__, derate_dram_bandwidth, derate_dram_factor);
-	dml2_printf("DML::%s: return_bw_mbps = %f\n", __func__, return_bw_mbps);
-#endif
+	DML_LOG_VERBOSE("DML::%s: is_avg_bw = %u\n", __func__, is_avg_bw);
+	DML_LOG_VERBOSE("DML::%s: is_hvm_en = %u\n", __func__, is_hvm_en);
+	DML_LOG_VERBOSE("DML::%s: is_hvm_only = %u\n", __func__, is_hvm_only);
+	DML_LOG_VERBOSE("DML::%s: state_type = %s\n", __func__, dml2_core_internal_soc_state_type_str(state_type));
+	DML_LOG_VERBOSE("DML::%s: bw_type = %s\n", __func__, dml2_core_internal_bw_type_str(bw_type));
+	DML_LOG_VERBOSE("DML::%s: dcfclk_mhz = %f\n", __func__, dcfclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: fclk_mhz = %f\n", __func__, fclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: ideal_sdp_bandwidth = %f\n", __func__, ideal_sdp_bandwidth);
+	DML_LOG_VERBOSE("DML::%s: ideal_fabric_bandwidth = %f\n", __func__, ideal_fabric_bandwidth);
+	DML_LOG_VERBOSE("DML::%s: ideal_dram_bandwidth = %f\n", __func__, ideal_dram_bandwidth);
+	DML_LOG_VERBOSE("DML::%s: derate_sdp_bandwidth = %f (derate %f)\n", __func__, derate_sdp_bandwidth, derate_sdp_factor);
+	DML_LOG_VERBOSE("DML::%s: derate_fabric_bandwidth = %f (derate %f)\n", __func__, derate_fabric_bandwidth, derate_fabric_factor);
+	DML_LOG_VERBOSE("DML::%s: derate_dram_bandwidth = %f (derate %f)\n", __func__, derate_dram_bandwidth, derate_dram_factor);
+	DML_LOG_VERBOSE("DML::%s: return_bw_mbps = %f\n", __func__, return_bw_mbps);
 	return return_bw_mbps;
 }
 
@@ -2807,9 +2748,9 @@ static noinline_for_stack void calculate_bandwidth_available(
 {
 	unsigned int n, m;
 
-	dml2_printf("DML::%s: dcfclk_mhz = %f\n", __func__, dcfclk_mhz);
-	dml2_printf("DML::%s: fclk_mhz = %f\n", __func__, fclk_mhz);
-	dml2_printf("DML::%s: dram_bw_mbps = %f\n", __func__, dram_bw_mbps);
+	DML_LOG_VERBOSE("DML::%s: dcfclk_mhz = %f\n", __func__, dcfclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: fclk_mhz = %f\n", __func__, fclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: dram_bw_mbps = %f\n", __func__, dram_bw_mbps);
 
 	// Calculate all the bandwidth availabe
 	for (m = 0; m < dml2_core_internal_soc_state_max; m++) {
@@ -2828,8 +2769,8 @@ static noinline_for_stack void calculate_bandwidth_available(
 
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: avg_bandwidth_available[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), avg_bandwidth_available[m][n]);
-			dml2_printf("DML::%s: urg_bandwidth_available[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), urg_bandwidth_available[m][n]);
+			DML_LOG_VERBOSE("DML::%s: avg_bandwidth_available[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), avg_bandwidth_available[m][n]);
+			DML_LOG_VERBOSE("DML::%s: urg_bandwidth_available[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), urg_bandwidth_available[m][n]);
 #endif
 
 			// urg_bandwidth_available_vm_only is indexed by soc_state
@@ -2843,9 +2784,9 @@ static noinline_for_stack void calculate_bandwidth_available(
 		urg_bandwidth_available_min[m] = math_min2(urg_bandwidth_available[m][dml2_core_internal_bw_dram], urg_bandwidth_available[m][dml2_core_internal_bw_sdp]);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: avg_bandwidth_available_min[%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), avg_bandwidth_available_min[m]);
-		dml2_printf("DML::%s: urg_bandwidth_available_min[%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), urg_bandwidth_available_min[m]);
-		dml2_printf("DML::%s: urg_bandwidth_available_vm_only[%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), urg_bandwidth_available_vm_only[n]);
+		DML_LOG_VERBOSE("DML::%s: avg_bandwidth_available_min[%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), avg_bandwidth_available_min[m]);
+		DML_LOG_VERBOSE("DML::%s: urg_bandwidth_available_min[%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), urg_bandwidth_available_min[m]);
+		DML_LOG_VERBOSE("DML::%s: urg_bandwidth_available_vm_only[%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), urg_bandwidth_available_vm_only[n]);
 #endif
 	}
 }
@@ -2879,13 +2820,13 @@ static void calculate_avg_bandwidth_required(
 	// SysActive and SVP Prefetch AVG bandwidth Check
 	for (k = 0; k < num_active_planes; ++k) {
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: plane %0d\n", __func__, k);
-		dml2_printf("DML::%s: ReadBandwidthLuma=%f\n", __func__, ReadBandwidthLuma[k]);
-		dml2_printf("DML::%s: ReadBandwidthChroma=%f\n", __func__, ReadBandwidthChroma[k]);
-		dml2_printf("DML::%s: dcc_dram_bw_nom_overhead_factor_p0=%f\n", __func__, dcc_dram_bw_nom_overhead_factor_p0[k]);
-		dml2_printf("DML::%s: dcc_dram_bw_nom_overhead_factor_p1=%f\n", __func__, dcc_dram_bw_nom_overhead_factor_p1[k]);
-		dml2_printf("DML::%s: mall_prefetch_dram_overhead_factor=%f\n", __func__, mall_prefetch_dram_overhead_factor[k]);
-		dml2_printf("DML::%s: mall_prefetch_sdp_overhead_factor=%f\n", __func__, mall_prefetch_sdp_overhead_factor[k]);
+		DML_LOG_VERBOSE("DML::%s: plane %0d\n", __func__, k);
+		DML_LOG_VERBOSE("DML::%s: ReadBandwidthLuma=%f\n", __func__, ReadBandwidthLuma[k]);
+		DML_LOG_VERBOSE("DML::%s: ReadBandwidthChroma=%f\n", __func__, ReadBandwidthChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: dcc_dram_bw_nom_overhead_factor_p0=%f\n", __func__, dcc_dram_bw_nom_overhead_factor_p0[k]);
+		DML_LOG_VERBOSE("DML::%s: dcc_dram_bw_nom_overhead_factor_p1=%f\n", __func__, dcc_dram_bw_nom_overhead_factor_p1[k]);
+		DML_LOG_VERBOSE("DML::%s: mall_prefetch_dram_overhead_factor=%f\n", __func__, mall_prefetch_dram_overhead_factor[k]);
+		DML_LOG_VERBOSE("DML::%s: mall_prefetch_sdp_overhead_factor=%f\n", __func__, mall_prefetch_sdp_overhead_factor[k]);
 #endif
 
 		sdp_overhead_factor = mall_prefetch_sdp_overhead_factor[k];
@@ -2902,10 +2843,10 @@ static void calculate_avg_bandwidth_required(
 		avg_bandwidth_required[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_dram] += dram_overhead_factor_p0 * ReadBandwidthLuma[k] + dram_overhead_factor_p1 * ReadBandwidthChroma[k] + cursor_bw[k];
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_sys_active), dml2_core_internal_bw_type_str(dml2_core_internal_bw_sdp), avg_bandwidth_required[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_sdp]);
-		dml2_printf("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_sys_active), dml2_core_internal_bw_type_str(dml2_core_internal_bw_dram), avg_bandwidth_required[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_dram]);
-		dml2_printf("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_svp_prefetch), dml2_core_internal_bw_type_str(dml2_core_internal_bw_sdp), avg_bandwidth_required[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_sdp]);
-		dml2_printf("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_svp_prefetch), dml2_core_internal_bw_type_str(dml2_core_internal_bw_dram), avg_bandwidth_required[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_dram]);
+		DML_LOG_VERBOSE("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_sys_active), dml2_core_internal_bw_type_str(dml2_core_internal_bw_sdp), avg_bandwidth_required[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_sdp]);
+		DML_LOG_VERBOSE("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_sys_active), dml2_core_internal_bw_type_str(dml2_core_internal_bw_dram), avg_bandwidth_required[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_dram]);
+		DML_LOG_VERBOSE("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_svp_prefetch), dml2_core_internal_bw_type_str(dml2_core_internal_bw_sdp), avg_bandwidth_required[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_sdp]);
+		DML_LOG_VERBOSE("DML::%s: avg_bandwidth_required[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(dml2_core_internal_soc_state_svp_prefetch), dml2_core_internal_bw_type_str(dml2_core_internal_bw_dram), avg_bandwidth_required[dml2_core_internal_soc_state_svp_prefetch][dml2_core_internal_bw_dram]);
 #endif
 	}
 }
@@ -3080,10 +3021,10 @@ static void CalculateVMRowAndSwath(struct dml2_core_internal_scratch *scratch,
 			&p->MaxNumSwathY[k]);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, vm_bytes_l = %u (before hvm level)\n", __func__, k, s->vm_bytes_l);
-		dml2_printf("DML::%s: k=%u, vm_bytes_c = %u (before hvm level)\n", __func__, k, s->vm_bytes_c);
-		dml2_printf("DML::%s: k=%u, meta_row_bytes_per_row_ub_l = %u\n", __func__, k, s->meta_row_bytes_per_row_ub_l[k]);
-		dml2_printf("DML::%s: k=%u, meta_row_bytes_per_row_ub_c = %u\n", __func__, k, s->meta_row_bytes_per_row_ub_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, vm_bytes_l = %u (before hvm level)\n", __func__, k, s->vm_bytes_l);
+		DML_LOG_VERBOSE("DML::%s: k=%u, vm_bytes_c = %u (before hvm level)\n", __func__, k, s->vm_bytes_c);
+		DML_LOG_VERBOSE("DML::%s: k=%u, meta_row_bytes_per_row_ub_l = %u\n", __func__, k, s->meta_row_bytes_per_row_ub_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, meta_row_bytes_per_row_ub_c = %u\n", __func__, k, s->meta_row_bytes_per_row_ub_c[k]);
 #endif
 		p->vm_bytes[k] = (s->vm_bytes_l + s->vm_bytes_c) * (1 + 8 * s->HostVMDynamicLevels);
 		p->meta_row_bytes[k] = s->meta_row_bytes_per_row_ub_l[k] + s->meta_row_bytes_per_row_ub_c[k];
@@ -3091,8 +3032,8 @@ static void CalculateVMRowAndSwath(struct dml2_core_internal_scratch *scratch,
 		p->meta_row_bytes_per_row_ub_c[k] = s->meta_row_bytes_per_row_ub_c[k];
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, meta_row_bytes = %u\n", __func__, k, p->meta_row_bytes[k]);
-		dml2_printf("DML::%s: k=%u, vm_bytes = %u (after hvm level)\n", __func__, k, p->vm_bytes[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, meta_row_bytes = %u\n", __func__, k, p->meta_row_bytes[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, vm_bytes = %u (after hvm level)\n", __func__, k, p->vm_bytes[k]);
 #endif
 		if (s->PixelPTEBytesPerRowStorageY[k] <= 64 * s->PTEBufferSizeInRequestsForLuma[k] && s->PixelPTEBytesPerRowStorageC[k] <= 64 * s->PTEBufferSizeInRequestsForChroma[k]) {
 			p->PTEBufferSizeNotExceeded[k] = true;
@@ -3104,18 +3045,18 @@ static void CalculateVMRowAndSwath(struct dml2_core_internal_scratch *scratch,
 												s->PixelPTEBytesPerRowC_one_row_per_frame[k] <= 64 * 2 * s->PTEBufferSizeInRequestsForChroma[k]);
 #ifdef __DML_VBA_DEBUG__
 		if (p->PTEBufferSizeNotExceeded[k] == 0 || s->one_row_per_frame_fits_in_buffer[k] == 0) {
-			dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowY = %u (before hvm level)\n", __func__, k, s->PixelPTEBytesPerRowY[k]);
-			dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowC = %u (before hvm level)\n", __func__, k, s->PixelPTEBytesPerRowC[k]);
-			dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowStorageY = %u\n", __func__, k, s->PixelPTEBytesPerRowStorageY[k]);
-			dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowStorageC = %u\n", __func__, k, s->PixelPTEBytesPerRowStorageC[k]);
-			dml2_printf("DML::%s: k=%u, PTEBufferSizeInRequestsForLuma = %u\n", __func__, k, s->PTEBufferSizeInRequestsForLuma[k]);
-			dml2_printf("DML::%s: k=%u, PTEBufferSizeInRequestsForChroma = %u\n", __func__, k, s->PTEBufferSizeInRequestsForChroma[k]);
-			dml2_printf("DML::%s: k=%u, PTEBufferSizeNotExceeded (not one_row_per_frame) = %u\n", __func__, k, p->PTEBufferSizeNotExceeded[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowY = %u (before hvm level)\n", __func__, k, s->PixelPTEBytesPerRowY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowC = %u (before hvm level)\n", __func__, k, s->PixelPTEBytesPerRowC[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowStorageY = %u\n", __func__, k, s->PixelPTEBytesPerRowStorageY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowStorageC = %u\n", __func__, k, s->PixelPTEBytesPerRowStorageC[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PTEBufferSizeInRequestsForLuma = %u\n", __func__, k, s->PTEBufferSizeInRequestsForLuma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PTEBufferSizeInRequestsForChroma = %u\n", __func__, k, s->PTEBufferSizeInRequestsForChroma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PTEBufferSizeNotExceeded (not one_row_per_frame) = %u\n", __func__, k, p->PTEBufferSizeNotExceeded[k]);
 
-			dml2_printf("DML::%s: k=%u, HostVMDynamicLevels = %u\n", __func__, k, s->HostVMDynamicLevels);
-			dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowY_one_row_per_frame = %u\n", __func__, k, s->PixelPTEBytesPerRowY_one_row_per_frame[k]);
-			dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowC_one_row_per_frame = %u\n", __func__, k, s->PixelPTEBytesPerRowC_one_row_per_frame[k]);
-			dml2_printf("DML::%s: k=%u, one_row_per_frame_fits_in_buffer = %u\n", __func__, k, s->one_row_per_frame_fits_in_buffer[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, HostVMDynamicLevels = %u\n", __func__, k, s->HostVMDynamicLevels);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowY_one_row_per_frame = %u\n", __func__, k, s->PixelPTEBytesPerRowY_one_row_per_frame[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowC_one_row_per_frame = %u\n", __func__, k, s->PixelPTEBytesPerRowC_one_row_per_frame[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, one_row_per_frame_fits_in_buffer = %u\n", __func__, k, s->one_row_per_frame_fits_in_buffer[k]);
 		}
 #endif
 	}
@@ -3146,8 +3087,8 @@ static void CalculateVMRowAndSwath(struct dml2_core_internal_scratch *scratch,
 	for (unsigned int k = 0; k < p->NumberOfActiveSurfaces; ++k) {
 		p->DCCMetaBufferSizeNotExceeded[k] = true;
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, SurfaceSizeInMALL = %u\n", __func__, k, p->SurfaceSizeInMALL[k]);
-		dml2_printf("DML::%s: k=%u, is_using_mall_for_ss = %u\n", __func__, k, p->is_using_mall_for_ss[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, SurfaceSizeInMALL = %u\n", __func__, k, p->SurfaceSizeInMALL[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, is_using_mall_for_ss = %u\n", __func__, k, p->is_using_mall_for_ss[k]);
 #endif
 		p->use_one_row_for_frame[k] = p->myPipe[k].FORCE_ONE_ROW_FOR_FRAME || p->is_using_mall_for_ss[k] || (p->display_cfg->plane_descriptors[k].overrides.legacy_svp_config == dml2_svp_mode_override_main_pipe) ||
 			(dml_is_phantom_pipe(&p->display_cfg->plane_descriptors[k])) || (p->display_cfg->plane_descriptors[k].overrides.gpuvm_min_page_size_kbytes > 64 && dml_is_vertical_rotation(p->myPipe[k].RotationAngle));
@@ -3170,9 +3111,9 @@ static void CalculateVMRowAndSwath(struct dml2_core_internal_scratch *scratch,
 			p->DCCMetaBufferSizeNotExceeded[k] = false;
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%d, meta_row_bytes = %d\n",  __func__, k, p->meta_row_bytes[k]);
-			dml2_printf("DML::%s: k=%d, DCCMetaBufferSizeBytes = %d\n",  __func__, k, p->DCCMetaBufferSizeBytes);
-			dml2_printf("DML::%s: k=%d, DCCMetaBufferSizeNotExceeded = %d\n",  __func__, k, p->DCCMetaBufferSizeNotExceeded[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, meta_row_bytes = %d\n",  __func__, k, p->meta_row_bytes[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, DCCMetaBufferSizeBytes = %d\n",  __func__, k, p->DCCMetaBufferSizeBytes);
+			DML_LOG_VERBOSE("DML::%s: k=%d, DCCMetaBufferSizeNotExceeded = %d\n",  __func__, k, p->DCCMetaBufferSizeNotExceeded[k]);
 #endif
 		}
 
@@ -3209,20 +3150,20 @@ static void CalculateVMRowAndSwath(struct dml2_core_internal_scratch *scratch,
 			&p->dpte_row_bw[k],
 			&p->meta_row_bw[k]);
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, use_one_row_for_frame = %u\n", __func__, k, p->use_one_row_for_frame[k]);
-		dml2_printf("DML::%s: k=%u, use_one_row_for_frame_flip = %u\n", __func__, k, p->use_one_row_for_frame_flip[k]);
-		dml2_printf("DML::%s: k=%u, UseMALLForPStateChange = %u\n", __func__, k, p->display_cfg->plane_descriptors[k].overrides.legacy_svp_config);
-		dml2_printf("DML::%s: k=%u, dpte_row_height_luma = %u\n", __func__, k, p->dpte_row_height_luma[k]);
-		dml2_printf("DML::%s: k=%u, dpte_row_width_luma_ub = %u\n", __func__, k, p->dpte_row_width_luma_ub[k]);
-		dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowY = %u (after hvm level)\n", __func__, k, s->PixelPTEBytesPerRowY[k]);
-		dml2_printf("DML::%s: k=%u, dpte_row_height_chroma = %u\n", __func__, k, p->dpte_row_height_chroma[k]);
-		dml2_printf("DML::%s: k=%u, dpte_row_width_chroma_ub = %u\n", __func__, k, p->dpte_row_width_chroma_ub[k]);
-		dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRowC = %u (after hvm level)\n", __func__, k, s->PixelPTEBytesPerRowC[k]);
-		dml2_printf("DML::%s: k=%u, PixelPTEBytesPerRow = %u\n", __func__, k, p->PixelPTEBytesPerRow[k]);
-		dml2_printf("DML::%s: k=%u, PTEBufferSizeNotExceeded = %u\n", __func__, k, p->PTEBufferSizeNotExceeded[k]);
-		dml2_printf("DML::%s: k=%u, gpuvm_enable = %u\n", __func__, k, p->display_cfg->gpuvm_enable);
-		dml2_printf("DML::%s: k=%u, PTE_BUFFER_MODE = %u\n", __func__, k, p->PTE_BUFFER_MODE[k]);
-		dml2_printf("DML::%s: k=%u, BIGK_FRAGMENT_SIZE = %u\n", __func__, k, p->BIGK_FRAGMENT_SIZE[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, use_one_row_for_frame = %u\n", __func__, k, p->use_one_row_for_frame[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, use_one_row_for_frame_flip = %u\n", __func__, k, p->use_one_row_for_frame_flip[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, UseMALLForPStateChange = %u\n", __func__, k, p->display_cfg->plane_descriptors[k].overrides.legacy_svp_config);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dpte_row_height_luma = %u\n", __func__, k, p->dpte_row_height_luma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dpte_row_width_luma_ub = %u\n", __func__, k, p->dpte_row_width_luma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowY = %u (after hvm level)\n", __func__, k, s->PixelPTEBytesPerRowY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dpte_row_height_chroma = %u\n", __func__, k, p->dpte_row_height_chroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dpte_row_width_chroma_ub = %u\n", __func__, k, p->dpte_row_width_chroma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRowC = %u (after hvm level)\n", __func__, k, s->PixelPTEBytesPerRowC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEBytesPerRow = %u\n", __func__, k, p->PixelPTEBytesPerRow[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PTEBufferSizeNotExceeded = %u\n", __func__, k, p->PTEBufferSizeNotExceeded[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, gpuvm_enable = %u\n", __func__, k, p->display_cfg->gpuvm_enable);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PTE_BUFFER_MODE = %u\n", __func__, k, p->PTE_BUFFER_MODE[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, BIGK_FRAGMENT_SIZE = %u\n", __func__, k, p->BIGK_FRAGMENT_SIZE[k]);
 #endif
 	}
 }
@@ -3257,19 +3198,19 @@ static double CalculateUrgentLatency(
 	}
 #ifdef __DML_VBA_DEBUG__
 	if (qos_type == dml2_qos_param_type_dcn4x) {
-		dml2_printf("DML::%s: qos_type = %d\n", __func__, qos_type);
-		dml2_printf("DML::%s: urgent_ramp_uclk_cycles = %d\n", __func__, urgent_ramp_uclk_cycles);
-		dml2_printf("DML::%s: uclk_freq_mhz = %f\n", __func__, uclk_freq_mhz);
-		dml2_printf("DML::%s: umc_urgent_ramp_latency_margin = %f\n", __func__, umc_urgent_ramp_latency_margin);
+		DML_LOG_VERBOSE("DML::%s: qos_type = %d\n", __func__, qos_type);
+		DML_LOG_VERBOSE("DML::%s: urgent_ramp_uclk_cycles = %d\n", __func__, urgent_ramp_uclk_cycles);
+		DML_LOG_VERBOSE("DML::%s: uclk_freq_mhz = %f\n", __func__, uclk_freq_mhz);
+		DML_LOG_VERBOSE("DML::%s: umc_urgent_ramp_latency_margin = %f\n", __func__, umc_urgent_ramp_latency_margin);
 	} else {
-		dml2_printf("DML::%s: UrgentLatencyPixelDataOnly = %f\n", __func__, UrgentLatencyPixelDataOnly);
-		dml2_printf("DML::%s: UrgentLatencyPixelMixedWithVMData = %f\n", __func__, UrgentLatencyPixelMixedWithVMData);
-		dml2_printf("DML::%s: UrgentLatencyVMDataOnly = %f\n", __func__, UrgentLatencyVMDataOnly);
-		dml2_printf("DML::%s: UrgentLatencyAdjustmentFabricClockComponent = %f\n", __func__, UrgentLatencyAdjustmentFabricClockComponent);
-		dml2_printf("DML::%s: UrgentLatencyAdjustmentFabricClockReference = %f\n", __func__, UrgentLatencyAdjustmentFabricClockReference);
+		DML_LOG_VERBOSE("DML::%s: UrgentLatencyPixelDataOnly = %f\n", __func__, UrgentLatencyPixelDataOnly);
+		DML_LOG_VERBOSE("DML::%s: UrgentLatencyPixelMixedWithVMData = %f\n", __func__, UrgentLatencyPixelMixedWithVMData);
+		DML_LOG_VERBOSE("DML::%s: UrgentLatencyVMDataOnly = %f\n", __func__, UrgentLatencyVMDataOnly);
+		DML_LOG_VERBOSE("DML::%s: UrgentLatencyAdjustmentFabricClockComponent = %f\n", __func__, UrgentLatencyAdjustmentFabricClockComponent);
+		DML_LOG_VERBOSE("DML::%s: UrgentLatencyAdjustmentFabricClockReference = %f\n", __func__, UrgentLatencyAdjustmentFabricClockReference);
 	}
-	dml2_printf("DML::%s: FabricClock = %f\n", __func__, FabricClock);
-	dml2_printf("DML::%s: UrgentLatency = %f\n", __func__, urgent_latency);
+	DML_LOG_VERBOSE("DML::%s: FabricClock = %f\n", __func__, FabricClock);
+	DML_LOG_VERBOSE("DML::%s: UrgentLatency = %f\n", __func__, urgent_latency);
 #endif
 	return urgent_latency;
 }
@@ -3296,18 +3237,18 @@ static double CalculateTripToMemory(
 
 #ifdef __DML_VBA_DEBUG__
 	if (qos_type == dml2_qos_param_type_dcn4x) {
-		dml2_printf("DML::%s: qos_type = %d\n", __func__, qos_type);
-		dml2_printf("DML::%s: max_round_trip_to_furthest_cs_fclk_cycles = %d\n", __func__, max_round_trip_to_furthest_cs_fclk_cycles);
-		dml2_printf("DML::%s: mall_overhead_fclk_cycles = %d\n", __func__, mall_overhead_fclk_cycles);
-		dml2_printf("DML::%s: trip_to_memory_uclk_cycles = %d\n", __func__, trip_to_memory_uclk_cycles);
-		dml2_printf("DML::%s: uclk_freq_mhz = %f\n", __func__, uclk_freq_mhz);
-		dml2_printf("DML::%s: FabricClock = %f\n", __func__, FabricClock);
-		dml2_printf("DML::%s: fabric_max_transport_latency_margin = %f\n", __func__, fabric_max_transport_latency_margin);
-		dml2_printf("DML::%s: umc_max_latency_margin = %f\n", __func__, umc_max_latency_margin);
+		DML_LOG_VERBOSE("DML::%s: qos_type = %d\n", __func__, qos_type);
+		DML_LOG_VERBOSE("DML::%s: max_round_trip_to_furthest_cs_fclk_cycles = %d\n", __func__, max_round_trip_to_furthest_cs_fclk_cycles);
+		DML_LOG_VERBOSE("DML::%s: mall_overhead_fclk_cycles = %d\n", __func__, mall_overhead_fclk_cycles);
+		DML_LOG_VERBOSE("DML::%s: trip_to_memory_uclk_cycles = %d\n", __func__, trip_to_memory_uclk_cycles);
+		DML_LOG_VERBOSE("DML::%s: uclk_freq_mhz = %f\n", __func__, uclk_freq_mhz);
+		DML_LOG_VERBOSE("DML::%s: FabricClock = %f\n", __func__, FabricClock);
+		DML_LOG_VERBOSE("DML::%s: fabric_max_transport_latency_margin = %f\n", __func__, fabric_max_transport_latency_margin);
+		DML_LOG_VERBOSE("DML::%s: umc_max_latency_margin = %f\n", __func__, umc_max_latency_margin);
 	} else {
-		dml2_printf("DML::%s: UrgLatency = %f\n", __func__, UrgLatency);
+		DML_LOG_VERBOSE("DML::%s: UrgLatency = %f\n", __func__, UrgLatency);
 	}
-	dml2_printf("DML::%s: trip_to_memory_us = %f\n", __func__, trip_to_memory_us);
+	DML_LOG_VERBOSE("DML::%s: trip_to_memory_us = %f\n", __func__, trip_to_memory_us);
 #endif
 
 
@@ -3334,14 +3275,14 @@ static double CalculateMetaTripToMemory(
 
 #ifdef __DML_VBA_DEBUG__
 	if (qos_type == dml2_qos_param_type_dcn4x) {
-		dml2_printf("DML::%s: qos_type = %d\n", __func__, qos_type);
-		dml2_printf("DML::%s: meta_trip_to_memory_fclk_cycles = %d\n", __func__, meta_trip_to_memory_fclk_cycles);
-		dml2_printf("DML::%s: meta_trip_to_memory_uclk_cycles = %d\n", __func__, meta_trip_to_memory_uclk_cycles);
-		dml2_printf("DML::%s: uclk_freq_mhz = %f\n", __func__, uclk_freq_mhz);
+		DML_LOG_VERBOSE("DML::%s: qos_type = %d\n", __func__, qos_type);
+		DML_LOG_VERBOSE("DML::%s: meta_trip_to_memory_fclk_cycles = %d\n", __func__, meta_trip_to_memory_fclk_cycles);
+		DML_LOG_VERBOSE("DML::%s: meta_trip_to_memory_uclk_cycles = %d\n", __func__, meta_trip_to_memory_uclk_cycles);
+		DML_LOG_VERBOSE("DML::%s: uclk_freq_mhz = %f\n", __func__, uclk_freq_mhz);
 	} else {
-		dml2_printf("DML::%s: UrgLatency = %f\n", __func__, UrgLatency);
+		DML_LOG_VERBOSE("DML::%s: UrgLatency = %f\n", __func__, UrgLatency);
 	}
-	dml2_printf("DML::%s: meta_trip_to_memory_us = %f\n", __func__, meta_trip_to_memory_us);
+	DML_LOG_VERBOSE("DML::%s: meta_trip_to_memory_us = %f\n", __func__, meta_trip_to_memory_us);
 #endif
 
 
@@ -3358,7 +3299,6 @@ static void calculate_cursor_req_attributes(
 	unsigned int *cursor_bytes_per_chunk,
 	unsigned int *cursor_bytes)
 {
-	unsigned int cursor_pitch = 0;
 	unsigned int cursor_bytes_per_req = 0;
 	unsigned int cursor_width_bytes = 0;
 	unsigned int cursor_height = 0;
@@ -3366,10 +3306,6 @@ static void calculate_cursor_req_attributes(
 	//SW determines the cursor pitch to support the maximum cursor_width that will be used but the following restrictions apply.
 	//- For 2bpp, cursor_pitch = 256 pixels due to min cursor request size of 64B
 	//- For 32 or 64 bpp, cursor_pitch = 64, 128 or 256 pixels depending on the cursor width
-	if (cursor_bpp == 2)
-		cursor_pitch = 256;
-	else
-		cursor_pitch = (unsigned int)1 << (unsigned int)math_ceil2(math_log((float)cursor_width, 2), 1);
 
 	//The cursor requestor uses a cursor request size of 64B, 128B, or 256B depending on the cursor_width and cursor_bpp as follows.
 
@@ -3409,8 +3345,8 @@ static void calculate_cursor_req_attributes(
 			*cursor_lines_per_chunk = 1;
 	} else {
 		if (cursor_width > 0) {
-			dml2_printf("DML::%s: Invalid cursor_bpp = %d\n", __func__, cursor_bpp);
-			DML2_ASSERT(0);
+			DML_LOG_VERBOSE("DML::%s: Invalid cursor_bpp = %d\n", __func__, cursor_bpp);
+			DML_ASSERT(0);
 		}
 	}
 
@@ -3421,15 +3357,15 @@ static void calculate_cursor_req_attributes(
 	cursor_height = cursor_width;
 	*cursor_bytes = *cursor_bytes_per_line * cursor_height;
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: cursor_bpp = %d\n", __func__, cursor_bpp);
-	dml2_printf("DML::%s: cursor_width = %d\n", __func__, cursor_width);
-	dml2_printf("DML::%s: cursor_width_bytes = %d\n", __func__, cursor_width_bytes);
-	dml2_printf("DML::%s: cursor_bytes_per_req = %d\n", __func__, cursor_bytes_per_req);
-	dml2_printf("DML::%s: cursor_lines_per_chunk = %d\n", __func__, *cursor_lines_per_chunk);
-	dml2_printf("DML::%s: cursor_bytes_per_line = %d\n", __func__, *cursor_bytes_per_line);
-	dml2_printf("DML::%s: cursor_bytes_per_chunk = %d\n", __func__, *cursor_bytes_per_chunk);
-	dml2_printf("DML::%s: cursor_bytes = %d\n", __func__, *cursor_bytes);
-	dml2_printf("DML::%s: cursor_pitch = %d\n", __func__, cursor_pitch);
+	DML_LOG_VERBOSE("DML::%s: cursor_bpp = %d\n", __func__, cursor_bpp);
+	DML_LOG_VERBOSE("DML::%s: cursor_width = %d\n", __func__, cursor_width);
+	DML_LOG_VERBOSE("DML::%s: cursor_width_bytes = %d\n", __func__, cursor_width_bytes);
+	DML_LOG_VERBOSE("DML::%s: cursor_bytes_per_req = %d\n", __func__, cursor_bytes_per_req);
+	DML_LOG_VERBOSE("DML::%s: cursor_lines_per_chunk = %d\n", __func__, *cursor_lines_per_chunk);
+	DML_LOG_VERBOSE("DML::%s: cursor_bytes_per_line = %d\n", __func__, *cursor_bytes_per_line);
+	DML_LOG_VERBOSE("DML::%s: cursor_bytes_per_chunk = %d\n", __func__, *cursor_bytes_per_chunk);
+	DML_LOG_VERBOSE("DML::%s: cursor_bytes = %d\n", __func__, *cursor_bytes);
+	DML_LOG_VERBOSE("DML::%s: cursor_pitch = %d\n", __func__, cursor_bpp == 2 ? 256 : (unsigned int)1 << (unsigned int)math_ceil2(math_log((float)cursor_width, 2), 1));
 #endif
 }
 
@@ -3460,13 +3396,13 @@ static void calculate_cursor_urgent_burst_factor(
 		}
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: LinesInCursorBuffer = %u\n", __func__, LinesInCursorBuffer);
-		dml2_printf("DML::%s: CursorBufferSizeInTime = %f\n", __func__, CursorBufferSizeInTime);
-		dml2_printf("DML::%s: CursorBufferSize = %u (kbytes)\n", __func__, CursorBufferSize);
-		dml2_printf("DML::%s: cursor_bytes_per_chunk = %u\n", __func__, cursor_bytes_per_chunk);
-		dml2_printf("DML::%s: cursor_lines_per_chunk = %u\n", __func__, cursor_lines_per_chunk);
-		dml2_printf("DML::%s: UrgentBurstFactorCursor = %f\n", __func__, *UrgentBurstFactorCursor);
-		dml2_printf("DML::%s: NotEnoughUrgentLatencyHiding = %d\n", __func__, *NotEnoughUrgentLatencyHiding);
+		DML_LOG_VERBOSE("DML::%s: LinesInCursorBuffer = %u\n", __func__, LinesInCursorBuffer);
+		DML_LOG_VERBOSE("DML::%s: CursorBufferSizeInTime = %f\n", __func__, CursorBufferSizeInTime);
+		DML_LOG_VERBOSE("DML::%s: CursorBufferSize = %u (kbytes)\n", __func__, CursorBufferSize);
+		DML_LOG_VERBOSE("DML::%s: cursor_bytes_per_chunk = %u\n", __func__, cursor_bytes_per_chunk);
+		DML_LOG_VERBOSE("DML::%s: cursor_lines_per_chunk = %u\n", __func__, cursor_lines_per_chunk);
+		DML_LOG_VERBOSE("DML::%s: UrgentBurstFactorCursor = %f\n", __func__, *UrgentBurstFactorCursor);
+		DML_LOG_VERBOSE("DML::%s: NotEnoughUrgentLatencyHiding = %d\n", __func__, *NotEnoughUrgentLatencyHiding);
 #endif
 
 	}
@@ -3501,15 +3437,15 @@ static void CalculateUrgentBurstFactor(
 	*UrgentBurstFactorChroma = 0;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: VRatio = %f\n", __func__, VRatio);
-	dml2_printf("DML::%s: VRatioC = %f\n", __func__, VRatioC);
-	dml2_printf("DML::%s: DETBufferSizeY = %d\n", __func__, DETBufferSizeY);
-	dml2_printf("DML::%s: DETBufferSizeC = %d\n", __func__, DETBufferSizeC);
-	dml2_printf("DML::%s: BytePerPixelInDETY = %f\n", __func__, BytePerPixelInDETY);
-	dml2_printf("DML::%s: swath_width_luma_ub = %d\n", __func__, swath_width_luma_ub);
-	dml2_printf("DML::%s: LineTime = %f\n", __func__, LineTime);
+	DML_LOG_VERBOSE("DML::%s: VRatio = %f\n", __func__, VRatio);
+	DML_LOG_VERBOSE("DML::%s: VRatioC = %f\n", __func__, VRatioC);
+	DML_LOG_VERBOSE("DML::%s: DETBufferSizeY = %d\n", __func__, DETBufferSizeY);
+	DML_LOG_VERBOSE("DML::%s: DETBufferSizeC = %d\n", __func__, DETBufferSizeC);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelInDETY = %f\n", __func__, BytePerPixelInDETY);
+	DML_LOG_VERBOSE("DML::%s: swath_width_luma_ub = %d\n", __func__, swath_width_luma_ub);
+	DML_LOG_VERBOSE("DML::%s: LineTime = %f\n", __func__, LineTime);
 #endif
-	DML2_ASSERT(VRatio > 0);
+	DML_ASSERT(VRatio > 0);
 
 	LinesInDETLuma = (dml_is_phantom_pipe(plane_cfg) ? 1024 * 1024 : DETBufferSizeY) / BytePerPixelInDETY / swath_width_luma_ub;
 
@@ -3534,12 +3470,12 @@ static void CalculateUrgentBurstFactor(
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: LinesInDETLuma = %f\n", __func__, LinesInDETLuma);
-	dml2_printf("DML::%s: UrgentLatency = %f\n", __func__, UrgentLatency);
-	dml2_printf("DML::%s: DETBufferSizeInTimeLuma = %f\n", __func__, DETBufferSizeInTimeLuma);
-	dml2_printf("DML::%s: UrgentBurstFactorLuma = %f\n", __func__, *UrgentBurstFactorLuma);
-	dml2_printf("DML::%s: UrgentBurstFactorChroma = %f\n", __func__, *UrgentBurstFactorChroma);
-	dml2_printf("DML::%s: NotEnoughUrgentLatencyHiding = %d\n", __func__, *NotEnoughUrgentLatencyHiding);
+	DML_LOG_VERBOSE("DML::%s: LinesInDETLuma = %f\n", __func__, LinesInDETLuma);
+	DML_LOG_VERBOSE("DML::%s: UrgentLatency = %f\n", __func__, UrgentLatency);
+	DML_LOG_VERBOSE("DML::%s: DETBufferSizeInTimeLuma = %f\n", __func__, DETBufferSizeInTimeLuma);
+	DML_LOG_VERBOSE("DML::%s: UrgentBurstFactorLuma = %f\n", __func__, *UrgentBurstFactorLuma);
+	DML_LOG_VERBOSE("DML::%s: UrgentBurstFactorChroma = %f\n", __func__, *UrgentBurstFactorChroma);
+	DML_LOG_VERBOSE("DML::%s: NotEnoughUrgentLatencyHiding = %d\n", __func__, *NotEnoughUrgentLatencyHiding);
 #endif
 }
 
@@ -3600,10 +3536,10 @@ static void CalculateDCFCLKDeepSleepTdlut(
 		if (display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut && tdlut_bytes_to_deliver[k] > 0) {
 			double tdlut_required_deepsleep_dcfclk = (double) tdlut_bytes_to_deliver[k] / 64.0 / prefetch_swath_time_us[k];
 
-			dml2_printf("DML::%s: k=%d, DCFClkDeepSleepPerSurface = %f\n", __func__, k, DCFClkDeepSleepPerSurface[k]);
-			dml2_printf("DML::%s: k=%d, tdlut_bytes_to_deliver = %d\n", __func__, k, tdlut_bytes_to_deliver[k]);
-			dml2_printf("DML::%s: k=%d, prefetch_swath_time_us = %f\n", __func__, k, prefetch_swath_time_us[k]);
-			dml2_printf("DML::%s: k=%d, tdlut_required_deepsleep_dcfclk = %f\n", __func__, k, tdlut_required_deepsleep_dcfclk);
+			DML_LOG_VERBOSE("DML::%s: k=%d, DCFClkDeepSleepPerSurface = %f\n", __func__, k, DCFClkDeepSleepPerSurface[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, tdlut_bytes_to_deliver = %d\n", __func__, k, tdlut_bytes_to_deliver[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, prefetch_swath_time_us = %f\n", __func__, k, prefetch_swath_time_us[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, tdlut_required_deepsleep_dcfclk = %f\n", __func__, k, tdlut_required_deepsleep_dcfclk);
 
 			// increase the deepsleep dcfclk to match the original dispclk throughput rate
 			if (tdlut_required_deepsleep_dcfclk > DCFClkDeepSleepPerSurface[k]) {
@@ -3613,8 +3549,8 @@ static void CalculateDCFCLKDeepSleepTdlut(
 		}
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, PixelClock = %f\n", __func__, k, pixel_rate_mhz);
-		dml2_printf("DML::%s: k=%u, DCFClkDeepSleepPerSurface = %f\n", __func__, k, DCFClkDeepSleepPerSurface[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PixelClock = %f\n", __func__, k, pixel_rate_mhz);
+		DML_LOG_VERBOSE("DML::%s: k=%u, DCFClkDeepSleepPerSurface = %f\n", __func__, k, DCFClkDeepSleepPerSurface[k]);
 #endif
 	}
 
@@ -3625,17 +3561,17 @@ static void CalculateDCFCLKDeepSleepTdlut(
 	*DCFClkDeepSleep = math_max2(8.0, __DML2_CALCS_DCFCLK_FACTOR__ * ReadBandwidth / (double)ReturnBusWidth);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: __DML2_CALCS_DCFCLK_FACTOR__ = %f\n", __func__, __DML2_CALCS_DCFCLK_FACTOR__);
-	dml2_printf("DML::%s: ReadBandwidth = %f\n", __func__, ReadBandwidth);
-	dml2_printf("DML::%s: ReturnBusWidth = %u\n", __func__, ReturnBusWidth);
-	dml2_printf("DML::%s: DCFClkDeepSleep = %f\n", __func__, *DCFClkDeepSleep);
+	DML_LOG_VERBOSE("DML::%s: __DML2_CALCS_DCFCLK_FACTOR__ = %f\n", __func__, __DML2_CALCS_DCFCLK_FACTOR__);
+	DML_LOG_VERBOSE("DML::%s: ReadBandwidth = %f\n", __func__, ReadBandwidth);
+	DML_LOG_VERBOSE("DML::%s: ReturnBusWidth = %u\n", __func__, ReturnBusWidth);
+	DML_LOG_VERBOSE("DML::%s: DCFClkDeepSleep = %f\n", __func__, *DCFClkDeepSleep);
 #endif
 
 	for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
 		*DCFClkDeepSleep = math_max2(*DCFClkDeepSleep, DCFClkDeepSleepPerSurface[k]);
 	}
 
-	dml2_printf("DML::%s: DCFClkDeepSleep = %f (final)\n", __func__, *DCFClkDeepSleep);
+	DML_LOG_VERBOSE("DML::%s: DCFClkDeepSleep = %f (final)\n", __func__, *DCFClkDeepSleep);
 }
 
 static noinline_for_stack void CalculateDCFCLKDeepSleep(
@@ -3731,12 +3667,12 @@ static unsigned int CalculateMaxVStartup(
 	else
 		max_vstartup_lines = vblank_size - (unsigned int)math_max2(1.0, math_ceil2(write_back_delay_us / line_time_us, 1.0));
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: VBlankNom = %u\n", __func__, timing->vblank_nom);
-	dml2_printf("DML::%s: vblank_nom_default_us = %u\n", __func__, vblank_nom_default_us);
-	dml2_printf("DML::%s: line_time_us = %f\n", __func__, line_time_us);
-	dml2_printf("DML::%s: vblank_actual = %u\n", __func__, vblank_actual);
-	dml2_printf("DML::%s: vblank_avail = %u\n", __func__, vblank_avail);
-	dml2_printf("DML::%s: max_vstartup_lines = %u\n", __func__, max_vstartup_lines);
+	DML_LOG_VERBOSE("DML::%s: VBlankNom = %lu\n", __func__, timing->vblank_nom);
+	DML_LOG_VERBOSE("DML::%s: vblank_nom_default_us = %u\n", __func__, vblank_nom_default_us);
+	DML_LOG_VERBOSE("DML::%s: line_time_us = %f\n", __func__, line_time_us);
+	DML_LOG_VERBOSE("DML::%s: vblank_actual = %u\n", __func__, vblank_actual);
+	DML_LOG_VERBOSE("DML::%s: vblank_avail = %u\n", __func__, vblank_avail);
+	DML_LOG_VERBOSE("DML::%s: max_vstartup_lines = %u\n", __func__, max_vstartup_lines);
 #endif
 	max_vstartup_lines = (unsigned int)math_min2(max_vstartup_lines, DML_MAX_VSTARTUP_START);
 	return max_vstartup_lines;
@@ -3761,9 +3697,9 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 	const long MAXIMUMCOMPRESSION = 4;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: ForceSingleDPP = %u\n", __func__, p->ForceSingleDPP);
+	DML_LOG_VERBOSE("DML::%s: ForceSingleDPP = %u\n", __func__, p->ForceSingleDPP);
 	for (unsigned int k = 0; k < p->NumberOfActiveSurfaces; ++k) {
-		dml2_printf("DML::%s: DPPPerSurface[%u] = %u\n", __func__, k, p->DPPPerSurface[k]);
+		DML_LOG_VERBOSE("DML::%s: DPPPerSurface[%u] = %u\n", __func__, k, p->DPPPerSurface[k]);
 	}
 #endif
 	CalculateSwathWidth(
@@ -3797,15 +3733,15 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 		p->full_swath_bytes_l[k] = (unsigned int)(p->swath_width_luma_ub[k] * p->BytePerPixDETY[k] * MaximumSwathHeightY[k]);
 		p->full_swath_bytes_c[k] = (unsigned int)(p->swath_width_chroma_ub[k] * p->BytePerPixDETC[k] * MaximumSwathHeightC[k]);
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u DPPPerSurface = %u\n", __func__, k, p->DPPPerSurface[k]);
-		dml2_printf("DML::%s: k=%u swath_width_luma_ub = %u\n", __func__, k, p->swath_width_luma_ub[k]);
-		dml2_printf("DML::%s: k=%u BytePerPixDETY = %f\n", __func__, k, p->BytePerPixDETY[k]);
-		dml2_printf("DML::%s: k=%u MaximumSwathHeightY = %u\n", __func__, k, MaximumSwathHeightY[k]);
-		dml2_printf("DML::%s: k=%u full_swath_bytes_l = %u\n", __func__, k, p->full_swath_bytes_l[k]);
-		dml2_printf("DML::%s: k=%u swath_width_chroma_ub = %u\n", __func__, k, p->swath_width_chroma_ub[k]);
-		dml2_printf("DML::%s: k=%u BytePerPixDETC = %f\n", __func__, k, p->BytePerPixDETC[k]);
-		dml2_printf("DML::%s: k=%u MaximumSwathHeightC = %u\n", __func__, k, MaximumSwathHeightC[k]);
-		dml2_printf("DML::%s: k=%u full_swath_bytes_c = %u\n", __func__, k, p->full_swath_bytes_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u DPPPerSurface = %u\n", __func__, k, p->DPPPerSurface[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u swath_width_luma_ub = %u\n", __func__, k, p->swath_width_luma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u BytePerPixDETY = %f\n", __func__, k, p->BytePerPixDETY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathHeightY = %u\n", __func__, k, MaximumSwathHeightY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_l = %u\n", __func__, k, p->full_swath_bytes_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u swath_width_chroma_ub = %u\n", __func__, k, p->swath_width_chroma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u BytePerPixDETC = %f\n", __func__, k, p->BytePerPixDETC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathHeightC = %u\n", __func__, k, MaximumSwathHeightC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_c = %u\n", __func__, k, p->full_swath_bytes_c[k]);
 #endif
 		if (p->display_cfg->plane_descriptors[k].pixel_format == dml2_420_10) {
 			p->full_swath_bytes_l[k] = (unsigned int)(math_ceil2((double)p->full_swath_bytes_l[k], 256));
@@ -3848,11 +3784,11 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 		p->CompressedBufferSizeInkByte);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: TotalActiveDPP = %u\n", __func__, TotalActiveDPP);
-	dml2_printf("DML::%s: nomDETInKByte = %u\n", __func__, p->nomDETInKByte);
-	dml2_printf("DML::%s: ConfigReturnBufferSizeInKByte = %u\n", __func__, p->ConfigReturnBufferSizeInKByte);
-	dml2_printf("DML::%s: UnboundedRequestEnabled = %u\n", __func__, *p->UnboundedRequestEnabled);
-	dml2_printf("DML::%s: CompressedBufferSizeInkByte = %u\n", __func__, *p->CompressedBufferSizeInkByte);
+	DML_LOG_VERBOSE("DML::%s: TotalActiveDPP = %u\n", __func__, TotalActiveDPP);
+	DML_LOG_VERBOSE("DML::%s: nomDETInKByte = %u\n", __func__, p->nomDETInKByte);
+	DML_LOG_VERBOSE("DML::%s: ConfigReturnBufferSizeInKByte = %u\n", __func__, p->ConfigReturnBufferSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: UnboundedRequestEnabled = %u\n", __func__, *p->UnboundedRequestEnabled);
+	DML_LOG_VERBOSE("DML::%s: CompressedBufferSizeInkByte = %u\n", __func__, *p->CompressedBufferSizeInkByte);
 #endif
 
 	*p->ViewportSizeSupport = true;
@@ -3860,7 +3796,7 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 
 		DETBufferSizeInKByteForSwathCalculation = (dml_is_phantom_pipe(&p->display_cfg->plane_descriptors[k]) ? 1024 : p->DETBufferSizeInKByte[k]);
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u DETBufferSizeInKByteForSwathCalculation = %u\n", __func__, k, DETBufferSizeInKByteForSwathCalculation);
+		DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeInKByteForSwathCalculation = %u\n", __func__, k, DETBufferSizeInKByteForSwathCalculation);
 #endif
 		if (p->display_cfg->plane_descriptors[k].surface.tiling == dml2_sw_linear) {
 			p->SwathHeightY[k] = MaximumSwathHeightY[k];
@@ -3917,13 +3853,13 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 		if ((p->full_swath_bytes_l[k] / 2 + p->full_swath_bytes_c[k] / 2 > DETBufferSizeInKByteForSwathCalculation * 1024 / 2) ||
 			p->SwathWidth[k] > p->MaximumSwathWidthLuma[k] || (p->SwathHeightC[k] > 0 && p->SwathWidthChroma[k] > p->MaximumSwathWidthChroma[k])) {
 			*p->ViewportSizeSupport = false;
-			dml2_printf("DML::%s: k=%u full_swath_bytes_l=%u\n", __func__, k, p->full_swath_bytes_l[k]);
-			dml2_printf("DML::%s: k=%u full_swath_bytes_c=%u\n", __func__, k, p->full_swath_bytes_c[k]);
-			dml2_printf("DML::%s: k=%u DETBufferSizeInKByteForSwathCalculation=%u\n", __func__, k, DETBufferSizeInKByteForSwathCalculation);
-			dml2_printf("DML::%s: k=%u SwathWidth=%u\n", __func__, k, p->SwathWidth[k]);
-			dml2_printf("DML::%s: k=%u MaximumSwathWidthLuma=%f\n", __func__, k, p->MaximumSwathWidthLuma[k]);
-			dml2_printf("DML::%s: k=%u SwathWidthChroma=%d\n", __func__, k, p->SwathWidthChroma[k]);
-			dml2_printf("DML::%s: k=%u MaximumSwathWidthChroma=%f\n", __func__, k, p->MaximumSwathWidthChroma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_l=%u\n", __func__, k, p->full_swath_bytes_l[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_c=%u\n", __func__, k, p->full_swath_bytes_c[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeInKByteForSwathCalculation=%u\n", __func__, k, DETBufferSizeInKByteForSwathCalculation);
+			DML_LOG_VERBOSE("DML::%s: k=%u SwathWidth=%u\n", __func__, k, p->SwathWidth[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthLuma=%f\n", __func__, k, p->MaximumSwathWidthLuma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u SwathWidthChroma=%d\n", __func__, k, p->SwathWidthChroma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthChroma=%f\n", __func__, k, p->MaximumSwathWidthChroma[k]);
 			p->ViewportSizeSupportPerSurface[k] = false;
 		} else {
 			p->ViewportSizeSupportPerSurface[k] = true;
@@ -3931,35 +3867,35 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 
 		if (p->SwathHeightC[k] == 0) {
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, All DET will be used for plane0\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: k=%u, All DET will be used for plane0\n", __func__, k);
 #endif
 			p->DETBufferSizeY[k] = p->DETBufferSizeInKByte[k] * 1024;
 			p->DETBufferSizeC[k] = 0;
 		} else if (RoundedUpSwathSizeBytesY[k] <= 1.5 * RoundedUpSwathSizeBytesC[k]) {
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, Half DET will be used for plane0, and half for plane1\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: k=%u, Half DET will be used for plane0, and half for plane1\n", __func__, k);
 #endif
 			p->DETBufferSizeY[k] = p->DETBufferSizeInKByte[k] * 1024 / 2;
 			p->DETBufferSizeC[k] = p->DETBufferSizeInKByte[k] * 1024 / 2;
 		} else {
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, 2/3 DET will be used for plane0, and 1/3 for plane1\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: k=%u, 2/3 DET will be used for plane0, and 1/3 for plane1\n", __func__, k);
 #endif
 			p->DETBufferSizeY[k] = (unsigned int)(math_floor2(p->DETBufferSizeInKByte[k] * 1024 * 2 / 3, 1024));
 			p->DETBufferSizeC[k] = p->DETBufferSizeInKByte[k] * 1024 - p->DETBufferSizeY[k];
 		}
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u SwathHeightY = %u\n", __func__, k, p->SwathHeightY[k]);
-		dml2_printf("DML::%s: k=%u SwathHeightC = %u\n", __func__, k, p->SwathHeightC[k]);
-		dml2_printf("DML::%s: k=%u full_swath_bytes_l = %u\n", __func__, k, p->full_swath_bytes_l[k]);
-		dml2_printf("DML::%s: k=%u full_swath_bytes_c = %u\n", __func__, k, p->full_swath_bytes_c[k]);
-		dml2_printf("DML::%s: k=%u RoundedUpSwathSizeBytesY = %u\n", __func__, k, RoundedUpSwathSizeBytesY[k]);
-		dml2_printf("DML::%s: k=%u RoundedUpSwathSizeBytesC = %u\n", __func__, k, RoundedUpSwathSizeBytesC[k]);
-		dml2_printf("DML::%s: k=%u DETBufferSizeInKByte = %u\n", __func__, k, p->DETBufferSizeInKByte[k]);
-		dml2_printf("DML::%s: k=%u DETBufferSizeY = %u\n", __func__, k, p->DETBufferSizeY[k]);
-		dml2_printf("DML::%s: k=%u DETBufferSizeC = %u\n", __func__, k, p->DETBufferSizeC[k]);
-		dml2_printf("DML::%s: k=%u ViewportSizeSupportPerSurface = %u\n", __func__, k, p->ViewportSizeSupportPerSurface[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u SwathHeightY = %u\n", __func__, k, p->SwathHeightY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u SwathHeightC = %u\n", __func__, k, p->SwathHeightC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_l = %u\n", __func__, k, p->full_swath_bytes_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u full_swath_bytes_c = %u\n", __func__, k, p->full_swath_bytes_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u RoundedUpSwathSizeBytesY = %u\n", __func__, k, RoundedUpSwathSizeBytesY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u RoundedUpSwathSizeBytesC = %u\n", __func__, k, RoundedUpSwathSizeBytesC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeInKByte = %u\n", __func__, k, p->DETBufferSizeInKByte[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeY = %u\n", __func__, k, p->DETBufferSizeY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u DETBufferSizeC = %u\n", __func__, k, p->DETBufferSizeC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u ViewportSizeSupportPerSurface = %u\n", __func__, k, p->ViewportSizeSupportPerSurface[k]);
 #endif
 
 	}
@@ -3969,12 +3905,12 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 		*p->compbuf_reserved_space_64b = (unsigned int)math_ceil2(math_max2(*p->compbuf_reserved_space_64b,
 			(double)(p->rob_buffer_size_kbytes * 1024 / 64) - (double)(RoundedUpSwathSizeBytesY[SurfaceDoingUnboundedRequest] * TTUFIFODEPTH / (p->mrq_present ? MAXIMUMCOMPRESSION : 1) / 64)), 1.0);
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: RoundedUpSwathSizeBytesY[%d] = %u\n", __func__, SurfaceDoingUnboundedRequest, RoundedUpSwathSizeBytesY[SurfaceDoingUnboundedRequest]);
-		dml2_printf("DML::%s: rob_buffer_size_kbytes = %u\n", __func__, p->rob_buffer_size_kbytes);
+		DML_LOG_VERBOSE("DML::%s: RoundedUpSwathSizeBytesY[%d] = %u\n", __func__, SurfaceDoingUnboundedRequest, RoundedUpSwathSizeBytesY[SurfaceDoingUnboundedRequest]);
+		DML_LOG_VERBOSE("DML::%s: rob_buffer_size_kbytes = %u\n", __func__, p->rob_buffer_size_kbytes);
 #endif
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: compbuf_reserved_space_64b = %u\n", __func__, *p->compbuf_reserved_space_64b);
+	DML_LOG_VERBOSE("DML::%s: compbuf_reserved_space_64b = %u\n", __func__, *p->compbuf_reserved_space_64b);
 #endif
 
 	*p->hw_debug5 = false;
@@ -3989,12 +3925,12 @@ static void CalculateSwathAndDETConfiguration(struct dml2_core_internal_scratch
 				+ *p->CompressedBufferSizeInkByte * MAXIMUMCOMPRESSION * 1024) > TTUFIFODEPTH * (RoundedUpSwathSizeBytesY[k] + RoundedUpSwathSizeBytesC[k])))
 			*p->hw_debug5 = true;
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u UnboundedRequestEnabled = %u\n", __func__, k, *p->UnboundedRequestEnabled);
-		dml2_printf("DML::%s: k=%u MAXIMUMCOMPRESSION = %lu\n", __func__, k, MAXIMUMCOMPRESSION);
-		dml2_printf("DML::%s: k=%u TTUFIFODEPTH = %lu\n", __func__, k, TTUFIFODEPTH);
-		dml2_printf("DML::%s: k=%u CompressedBufferSizeInkByte = %u\n", __func__, k, *p->CompressedBufferSizeInkByte);
-		dml2_printf("DML::%s: k=%u RoundedUpSwathSizeBytesC = %u\n", __func__, k, RoundedUpSwathSizeBytesC[k]);
-		dml2_printf("DML::%s: k=%u hw_debug5 = %u\n", __func__, k, *p->hw_debug5);
+		DML_LOG_VERBOSE("DML::%s: k=%u UnboundedRequestEnabled = %u\n", __func__, k, *p->UnboundedRequestEnabled);
+		DML_LOG_VERBOSE("DML::%s: k=%u MAXIMUMCOMPRESSION = %lu\n", __func__, k, MAXIMUMCOMPRESSION);
+		DML_LOG_VERBOSE("DML::%s: k=%u TTUFIFODEPTH = %lu\n", __func__, k, TTUFIFODEPTH);
+		DML_LOG_VERBOSE("DML::%s: k=%u CompressedBufferSizeInkByte = %u\n", __func__, k, *p->CompressedBufferSizeInkByte);
+		DML_LOG_VERBOSE("DML::%s: k=%u RoundedUpSwathSizeBytesC = %u\n", __func__, k, RoundedUpSwathSizeBytesC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u hw_debug5 = %u\n", __func__, k, *p->hw_debug5);
 #endif
 	}
 #endif
@@ -4192,15 +4128,15 @@ static noinline_for_stack void CalculateODMMode(
 	SurfaceRequiredDISPCLKWithODMCombineThreeToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_3to1, PixelClock);
 	SurfaceRequiredDISPCLKWithODMCombineFourToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_4to1, PixelClock);
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: ODMUse = %d\n", __func__, ODMUse);
-	dml2_printf("DML::%s: Output = %d\n", __func__, Output);
-	dml2_printf("DML::%s: DSCEnable = %d\n", __func__, DSCEnable);
-	dml2_printf("DML::%s: MaxDispclk = %f\n", __func__, MaxDispclk);
-	dml2_printf("DML::%s: MaximumPixelsPerLinePerDSCUnit = %d\n", __func__, MaximumPixelsPerLinePerDSCUnit);
-	dml2_printf("DML::%s: SurfaceRequiredDISPCLKWithoutODMCombine = %f\n", __func__, SurfaceRequiredDISPCLKWithoutODMCombine);
-	dml2_printf("DML::%s: SurfaceRequiredDISPCLKWithODMCombineTwoToOne = %f\n", __func__, SurfaceRequiredDISPCLKWithODMCombineTwoToOne);
-	dml2_printf("DML::%s: SurfaceRequiredDISPCLKWithODMCombineThreeToOne = %f\n", __func__, SurfaceRequiredDISPCLKWithODMCombineThreeToOne);
-	dml2_printf("DML::%s: SurfaceRequiredDISPCLKWithODMCombineFourToOne = %f\n", __func__, SurfaceRequiredDISPCLKWithODMCombineFourToOne);
+	DML_LOG_VERBOSE("DML::%s: ODMUse = %d\n", __func__, ODMUse);
+	DML_LOG_VERBOSE("DML::%s: Output = %d\n", __func__, Output);
+	DML_LOG_VERBOSE("DML::%s: DSCEnable = %d\n", __func__, DSCEnable);
+	DML_LOG_VERBOSE("DML::%s: MaxDispclk = %f\n", __func__, MaxDispclk);
+	DML_LOG_VERBOSE("DML::%s: MaximumPixelsPerLinePerDSCUnit = %d\n", __func__, MaximumPixelsPerLinePerDSCUnit);
+	DML_LOG_VERBOSE("DML::%s: SurfaceRequiredDISPCLKWithoutODMCombine = %f\n", __func__, SurfaceRequiredDISPCLKWithoutODMCombine);
+	DML_LOG_VERBOSE("DML::%s: SurfaceRequiredDISPCLKWithODMCombineTwoToOne = %f\n", __func__, SurfaceRequiredDISPCLKWithODMCombineTwoToOne);
+	DML_LOG_VERBOSE("DML::%s: SurfaceRequiredDISPCLKWithODMCombineThreeToOne = %f\n", __func__, SurfaceRequiredDISPCLKWithODMCombineThreeToOne);
+	DML_LOG_VERBOSE("DML::%s: SurfaceRequiredDISPCLKWithODMCombineFourToOne = %f\n", __func__, SurfaceRequiredDISPCLKWithODMCombineFourToOne);
 #endif
 	if (ODMUse == dml2_odm_mode_auto)
 		DecidedODMMode = DecideODMMode(HActive,
@@ -4245,10 +4181,10 @@ static noinline_for_stack void CalculateODMMode(
 	*NumberOfDPP = NumberOfDPPRequired;
 	*RequiredDISPCLKPerSurface = success ? DISPCLKRequired : 0;
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: ODMMode = %d\n", __func__, *ODMMode);
-	dml2_printf("DML::%s: NumberOfDPP = %d\n", __func__, *NumberOfDPP);
-	dml2_printf("DML::%s: TotalAvailablePipesSupport = %d\n", __func__, *TotalAvailablePipesSupport);
-	dml2_printf("DML::%s: RequiredDISPCLKPerSurface = %f\n", __func__, *RequiredDISPCLKPerSurface);
+	DML_LOG_VERBOSE("DML::%s: ODMMode = %d\n", __func__, *ODMMode);
+	DML_LOG_VERBOSE("DML::%s: NumberOfDPP = %d\n", __func__, *NumberOfDPP);
+	DML_LOG_VERBOSE("DML::%s: TotalAvailablePipesSupport = %d\n", __func__, *TotalAvailablePipesSupport);
+	DML_LOG_VERBOSE("DML::%s: RequiredDISPCLKPerSurface = %f\n", __func__, *RequiredDISPCLKPerSurface);
 #endif
 }
 
@@ -4292,17 +4228,17 @@ static noinline_for_stack void CalculateOutputLink(
 	*OutputRate = dml2_core_internal_output_rate_unknown;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DSCEnable = %u (dis, en, en_if_necessary)\n", __func__, DSCEnable);
-	dml2_printf("DML::%s: PHYCLK = %f\n", __func__, PHYCLK);
-	dml2_printf("DML::%s: PixelClockBackEnd = %f\n", __func__, PixelClockBackEnd);
-	dml2_printf("DML::%s: AudioSampleRate = %f\n", __func__, AudioSampleRate);
-	dml2_printf("DML::%s: HActive = %u\n", __func__, HActive);
-	dml2_printf("DML::%s: HTotal = %u\n", __func__, HTotal);
-	dml2_printf("DML::%s: ODMModeNoDSC = %u\n", __func__, ODMModeNoDSC);
-	dml2_printf("DML::%s: ODMModeDSC = %u\n", __func__, ODMModeDSC);
-	dml2_printf("DML::%s: ForcedOutputLinkBPP = %f\n", __func__, ForcedOutputLinkBPP);
-	dml2_printf("DML::%s: Output (encoder) = %u\n", __func__, Output);
-	dml2_printf("DML::%s: OutputLinkDPRate = %u\n", __func__, OutputLinkDPRate);
+	DML_LOG_VERBOSE("DML::%s: DSCEnable = %u (dis, en, en_if_necessary)\n", __func__, DSCEnable);
+	DML_LOG_VERBOSE("DML::%s: PHYCLK = %f\n", __func__, PHYCLK);
+	DML_LOG_VERBOSE("DML::%s: PixelClockBackEnd = %f\n", __func__, PixelClockBackEnd);
+	DML_LOG_VERBOSE("DML::%s: AudioSampleRate = %f\n", __func__, AudioSampleRate);
+	DML_LOG_VERBOSE("DML::%s: HActive = %u\n", __func__, HActive);
+	DML_LOG_VERBOSE("DML::%s: HTotal = %u\n", __func__, HTotal);
+	DML_LOG_VERBOSE("DML::%s: ODMModeNoDSC = %u\n", __func__, ODMModeNoDSC);
+	DML_LOG_VERBOSE("DML::%s: ODMModeDSC = %u\n", __func__, ODMModeDSC);
+	DML_LOG_VERBOSE("DML::%s: ForcedOutputLinkBPP = %f\n", __func__, ForcedOutputLinkBPP);
+	DML_LOG_VERBOSE("DML::%s: Output (encoder) = %u\n", __func__, Output);
+	DML_LOG_VERBOSE("DML::%s: OutputLinkDPRate = %u\n", __func__, OutputLinkDPRate);
 #endif
 	{
 		if (Output == dml2_hdmi) {
@@ -4487,9 +4423,9 @@ static noinline_for_stack void CalculateOutputLink(
 		}
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: RequiresDSC = %u\n", __func__, *RequiresDSC);
-	dml2_printf("DML::%s: RequiresFEC = %u\n", __func__, *RequiresFEC);
-	dml2_printf("DML::%s: OutBpp = %f\n", __func__, *OutBpp);
+	DML_LOG_VERBOSE("DML::%s: RequiresDSC = %u\n", __func__, *RequiresDSC);
+	DML_LOG_VERBOSE("DML::%s: RequiresFEC = %u\n", __func__, *RequiresFEC);
+	DML_LOG_VERBOSE("DML::%s: OutBpp = %f\n", __func__, *OutBpp);
 #endif
 }
 
@@ -4571,17 +4507,17 @@ static unsigned int DSCDelayRequirement(
 		DSCDelayRequirement_val = 0;
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DSCEnabled= %u\n", __func__, DSCEnabled);
-	dml2_printf("DML::%s: ODMMode = %u\n", __func__, ODMMode);
-	dml2_printf("DML::%s: OutputBpp = %f\n", __func__, OutputBpp);
-	dml2_printf("DML::%s: HActive = %u\n", __func__, HActive);
-	dml2_printf("DML::%s: HTotal = %u\n", __func__, HTotal);
-	dml2_printf("DML::%s: PixelClock = %f\n", __func__, PixelClock);
-	dml2_printf("DML::%s: PixelClockBackEnd = %f\n", __func__, PixelClockBackEnd);
-	dml2_printf("DML::%s: OutputFormat = %u\n", __func__, OutputFormat);
-	dml2_printf("DML::%s: DSCInputBitPerComponent = %u\n", __func__, DSCInputBitPerComponent);
-	dml2_printf("DML::%s: NumberOfDSCSlices = %u\n", __func__, NumberOfDSCSlices);
-	dml2_printf("DML::%s: DSCDelayRequirement_val = %u\n", __func__, DSCDelayRequirement_val);
+	DML_LOG_VERBOSE("DML::%s: DSCEnabled= %u\n", __func__, DSCEnabled);
+	DML_LOG_VERBOSE("DML::%s: ODMMode = %u\n", __func__, ODMMode);
+	DML_LOG_VERBOSE("DML::%s: OutputBpp = %f\n", __func__, OutputBpp);
+	DML_LOG_VERBOSE("DML::%s: HActive = %u\n", __func__, HActive);
+	DML_LOG_VERBOSE("DML::%s: HTotal = %u\n", __func__, HTotal);
+	DML_LOG_VERBOSE("DML::%s: PixelClock = %f\n", __func__, PixelClock);
+	DML_LOG_VERBOSE("DML::%s: PixelClockBackEnd = %f\n", __func__, PixelClockBackEnd);
+	DML_LOG_VERBOSE("DML::%s: OutputFormat = %u\n", __func__, OutputFormat);
+	DML_LOG_VERBOSE("DML::%s: DSCInputBitPerComponent = %u\n", __func__, DSCInputBitPerComponent);
+	DML_LOG_VERBOSE("DML::%s: NumberOfDSCSlices = %u\n", __func__, NumberOfDSCSlices);
+	DML_LOG_VERBOSE("DML::%s: DSCDelayRequirement_val = %u\n", __func__, DSCDelayRequirement_val);
 #endif
 
 	return DSCDelayRequirement_val;
@@ -4654,10 +4590,10 @@ static void CalculateSurfaceSizeInMall(
 		(TotalSurfaceSizeInMALLForSubVP > MALLAllocatedForDCNInBytes);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: MALLAllocatedForDCN = %u\n", __func__, MALLAllocatedForDCN * 1024 * 1024);
-	dml2_printf("DML::%s: TotalSurfaceSizeInMALLForSubVP = %u\n", __func__, TotalSurfaceSizeInMALLForSubVP);
-	dml2_printf("DML::%s: TotalSurfaceSizeInMALLForSS = %u\n", __func__, TotalSurfaceSizeInMALLForSS);
-	dml2_printf("DML::%s: ExceededMALLSize = %u\n", __func__, *ExceededMALLSize);
+	DML_LOG_VERBOSE("DML::%s: MALLAllocatedForDCN = %u\n", __func__, MALLAllocatedForDCN * 1024 * 1024);
+	DML_LOG_VERBOSE("DML::%s: TotalSurfaceSizeInMALLForSubVP = %u\n", __func__, TotalSurfaceSizeInMALLForSubVP);
+	DML_LOG_VERBOSE("DML::%s: TotalSurfaceSizeInMALLForSS = %u\n", __func__, TotalSurfaceSizeInMALLForSS);
+	DML_LOG_VERBOSE("DML::%s: ExceededMALLSize = %u\n", __func__, *ExceededMALLSize);
 #endif
 }
 
@@ -4674,7 +4610,6 @@ static void calculate_tdlut_setting(
 	unsigned int tdlut_vmpg_per_frame;
 	unsigned int tdlut_pte_req_per_frame;
 	unsigned int tdlut_bytes_per_line;
-	unsigned int tdlut_delivery_cycles;
 	double tdlut_drain_rate;
 	unsigned int tdlut_mpc_width;
 	unsigned int tdlut_bytes_per_group_simple;
@@ -4737,13 +4672,13 @@ static void calculate_tdlut_setting(
 		*p->tdlut_bytes_per_frame = tdlut_bytes_per_line * tdlut_mpc_width * tdlut_mpc_width;
 		*p->tdlut_bytes_per_group = tdlut_bytes_per_line * tdlut_mpc_width;
 		//the delivery cycles is DispClk cycles per line * number of lines * number of slices
-		tdlut_delivery_cycles = (unsigned int)math_ceil2(tdlut_mpc_width/2.0, 1) * tdlut_mpc_width * tdlut_mpc_width;
+		//tdlut_delivery_cycles = (unsigned int)math_ceil2(tdlut_mpc_width/2.0, 1) * tdlut_mpc_width * tdlut_mpc_width;
 		tdlut_drain_rate = tdlut_bytes_per_line * p->dispclk_mhz / math_ceil2(tdlut_mpc_width/2.0, 1);
 	} else {
 		//tdlut_addressing_mode = tdlut_simple_linear, 3dlut width should be 4*1229=4916 elements
 		*p->tdlut_bytes_per_frame = (unsigned int)math_ceil2(tdlut_width * tdlut_bpe, 256);
 		*p->tdlut_bytes_per_group = tdlut_bytes_per_group_simple;
-		tdlut_delivery_cycles = (unsigned int)math_ceil2(tdlut_width/2.0, 1);
+		//tdlut_delivery_cycles = (unsigned int)math_ceil2(tdlut_width/2.0, 1);
 		tdlut_drain_rate = 2 * tdlut_bpe * p->dispclk_mhz;
 	}
 
@@ -4756,25 +4691,25 @@ static void calculate_tdlut_setting(
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: gpuvm_enable = %d\n", __func__, p->gpuvm_enable);
-	dml2_printf("DML::%s: vmpg_bytes = %d\n", __func__, vmpg_bytes);
-	dml2_printf("DML::%s: tdlut_vmpg_per_frame = %d\n", __func__, tdlut_vmpg_per_frame);
-	dml2_printf("DML::%s: tdlut_pte_req_per_frame = %d\n", __func__, tdlut_pte_req_per_frame);
+	DML_LOG_VERBOSE("DML::%s: gpuvm_enable = %d\n", __func__, p->gpuvm_enable);
+	DML_LOG_VERBOSE("DML::%s: vmpg_bytes = %d\n", __func__, vmpg_bytes);
+	DML_LOG_VERBOSE("DML::%s: tdlut_vmpg_per_frame = %d\n", __func__, tdlut_vmpg_per_frame);
+	DML_LOG_VERBOSE("DML::%s: tdlut_pte_req_per_frame = %d\n", __func__, tdlut_pte_req_per_frame);
 
-	dml2_printf("DML::%s: dispclk_mhz = %f\n", __func__, p->dispclk_mhz);
-	dml2_printf("DML::%s: tdlut_width = %u\n", __func__, tdlut_width);
-	dml2_printf("DML::%s: tdlut_addressing_mode = %s\n", __func__, (p->tdlut_addressing_mode == dml2_tdlut_sw_linear) ? "sw_linear" : "simple_linear");
-	dml2_printf("DML::%s: tdlut_pitch_bytes = %u\n", __func__, tdlut_pitch_bytes);
-	dml2_printf("DML::%s: tdlut_footprint_bytes = %u\n", __func__, tdlut_footprint_bytes);
-	dml2_printf("DML::%s: tdlut_bytes_per_frame = %u\n", __func__, *p->tdlut_bytes_per_frame);
-	dml2_printf("DML::%s: tdlut_bytes_per_line = %u\n", __func__, tdlut_bytes_per_line);
-	dml2_printf("DML::%s: tdlut_bytes_per_group = %u\n", __func__, *p->tdlut_bytes_per_group);
-	dml2_printf("DML::%s: tdlut_drain_rate = %f\n", __func__, tdlut_drain_rate);
-	dml2_printf("DML::%s: tdlut_delivery_cycles = %u\n", __func__, tdlut_delivery_cycles);
-	dml2_printf("DML::%s: tdlut_opt_time = %f\n", __func__, *p->tdlut_opt_time);
-	dml2_printf("DML::%s: tdlut_drain_time = %f\n", __func__, *p->tdlut_drain_time);
-	dml2_printf("DML::%s: tdlut_bytes_to_deliver = %d\n", __func__, *p->tdlut_bytes_to_deliver);
-	dml2_printf("DML::%s: tdlut_groups_per_2row_ub = %d\n", __func__, *p->tdlut_groups_per_2row_ub);
+	DML_LOG_VERBOSE("DML::%s: dispclk_mhz = %f\n", __func__, p->dispclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: tdlut_width = %u\n", __func__, tdlut_width);
+	DML_LOG_VERBOSE("DML::%s: tdlut_addressing_mode = %s\n", __func__, (p->tdlut_addressing_mode == dml2_tdlut_sw_linear) ? "sw_linear" : "simple_linear");
+	DML_LOG_VERBOSE("DML::%s: tdlut_pitch_bytes = %u\n", __func__, tdlut_pitch_bytes);
+	DML_LOG_VERBOSE("DML::%s: tdlut_footprint_bytes = %u\n", __func__, tdlut_footprint_bytes);
+	DML_LOG_VERBOSE("DML::%s: tdlut_bytes_per_frame = %u\n", __func__, *p->tdlut_bytes_per_frame);
+	DML_LOG_VERBOSE("DML::%s: tdlut_bytes_per_line = %u\n", __func__, tdlut_bytes_per_line);
+	DML_LOG_VERBOSE("DML::%s: tdlut_bytes_per_group = %u\n", __func__, *p->tdlut_bytes_per_group);
+	DML_LOG_VERBOSE("DML::%s: tdlut_drain_rate = %f\n", __func__, tdlut_drain_rate);
+	DML_LOG_VERBOSE("DML::%s: tdlut_delivery_cycles = %u\n", __func__, p->tdlut_addressing_mode == dml2_tdlut_sw_linear ? (unsigned int)math_ceil2(tdlut_mpc_width/2.0, 1) * tdlut_mpc_width * tdlut_mpc_width : (unsigned int)math_ceil2(tdlut_width/2.0, 1));
+	DML_LOG_VERBOSE("DML::%s: tdlut_opt_time = %f\n", __func__, *p->tdlut_opt_time);
+	DML_LOG_VERBOSE("DML::%s: tdlut_drain_time = %f\n", __func__, *p->tdlut_drain_time);
+	DML_LOG_VERBOSE("DML::%s: tdlut_bytes_to_deliver = %d\n", __func__, *p->tdlut_bytes_to_deliver);
+	DML_LOG_VERBOSE("DML::%s: tdlut_groups_per_2row_ub = %d\n", __func__, *p->tdlut_groups_per_2row_ub);
 #endif
 }
 
@@ -4820,10 +4755,10 @@ static void CalculateTarb(
 	*Tarb = extra_bytes / ReturnBW;
 	*Tarb_prefetch = extra_bytes_prefetch / ReturnBW;
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: PixelChunkSizeInKByte = %d\n", __func__, PixelChunkSizeInKByte);
-	dml2_printf("DML::%s: MetaChunkSize = %d\n", __func__, MetaChunkSize);
-	dml2_printf("DML::%s: extra_bytes = %f\n", __func__, extra_bytes);
-	dml2_printf("DML::%s: extra_bytes_prefetch = %f\n", __func__, extra_bytes_prefetch);
+	DML_LOG_VERBOSE("DML::%s: PixelChunkSizeInKByte = %d\n", __func__, PixelChunkSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: MetaChunkSize = %d\n", __func__, MetaChunkSize);
+	DML_LOG_VERBOSE("DML::%s: extra_bytes = %f\n", __func__, extra_bytes);
+	DML_LOG_VERBOSE("DML::%s: extra_bytes_prefetch = %f\n", __func__, extra_bytes_prefetch);
 #endif
 }
 
@@ -4838,10 +4773,10 @@ static double CalculateTWait(
 	TWait = math_max2(reserved_vblank_time_ns/1000.0, g6_temp_read_blackout_us) + t_urg_trip;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: reserved_vblank_time_ns = %d\n", __func__, reserved_vblank_time_ns);
-	dml2_printf("DML::%s: UrgentLatency = %f\n", __func__, UrgentLatency);
-	dml2_printf("DML::%s: Ttrip = %f\n", __func__, Ttrip);
-	dml2_printf("DML::%s: TWait = %f\n", __func__, TWait);
+	DML_LOG_VERBOSE("DML::%s: reserved_vblank_time_ns = %ld\n", __func__, reserved_vblank_time_ns);
+	DML_LOG_VERBOSE("DML::%s: UrgentLatency = %f\n", __func__, UrgentLatency);
+	DML_LOG_VERBOSE("DML::%s: Ttrip = %f\n", __func__, Ttrip);
+	DML_LOG_VERBOSE("DML::%s: TWait = %f\n", __func__, TWait);
 #endif
 	return TWait;
 }
@@ -4887,20 +4822,20 @@ static void CalculateVUpdateAndDynamicMetadataParameters(
 		*Tdmsks = *Tdmsks / 2;
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DynamicMetadataLinesBeforeActiveRequired = %u\n", __func__, DynamicMetadataLinesBeforeActiveRequired);
-	dml2_printf("DML::%s: VBlank = %u\n", __func__, VBlank);
-	dml2_printf("DML::%s: HTotal = %u\n", __func__, HTotal);
-	dml2_printf("DML::%s: PixelClock = %f\n", __func__, PixelClock);
-	dml2_printf("DML::%s: Dppclk = %f\n", __func__, Dppclk);
-	dml2_printf("DML::%s: DCFClkDeepSleep = %f\n", __func__, DCFClkDeepSleep);
-	dml2_printf("DML::%s: MaxInterDCNTileRepeaters = %u\n", __func__, MaxInterDCNTileRepeaters);
-	dml2_printf("DML::%s: TotalRepeaterDelayTime = %f\n", __func__, TotalRepeaterDelayTime);
+	DML_LOG_VERBOSE("DML::%s: DynamicMetadataLinesBeforeActiveRequired = %u\n", __func__, DynamicMetadataLinesBeforeActiveRequired);
+	DML_LOG_VERBOSE("DML::%s: VBlank = %u\n", __func__, VBlank);
+	DML_LOG_VERBOSE("DML::%s: HTotal = %u\n", __func__, HTotal);
+	DML_LOG_VERBOSE("DML::%s: PixelClock = %f\n", __func__, PixelClock);
+	DML_LOG_VERBOSE("DML::%s: Dppclk = %f\n", __func__, Dppclk);
+	DML_LOG_VERBOSE("DML::%s: DCFClkDeepSleep = %f\n", __func__, DCFClkDeepSleep);
+	DML_LOG_VERBOSE("DML::%s: MaxInterDCNTileRepeaters = %u\n", __func__, MaxInterDCNTileRepeaters);
+	DML_LOG_VERBOSE("DML::%s: TotalRepeaterDelayTime = %f\n", __func__, TotalRepeaterDelayTime);
 
-	dml2_printf("DML::%s: VUpdateWidthPix = %u\n", __func__, *VUpdateWidthPix);
-	dml2_printf("DML::%s: VReadyOffsetPix = %u\n", __func__, *VReadyOffsetPix);
-	dml2_printf("DML::%s: VUpdateOffsetPix = %u\n", __func__, *VUpdateOffsetPix);
+	DML_LOG_VERBOSE("DML::%s: VUpdateWidthPix = %u\n", __func__, *VUpdateWidthPix);
+	DML_LOG_VERBOSE("DML::%s: VReadyOffsetPix = %u\n", __func__, *VReadyOffsetPix);
+	DML_LOG_VERBOSE("DML::%s: VUpdateOffsetPix = %u\n", __func__, *VUpdateOffsetPix);
 
-	dml2_printf("DML::%s: Tdmsks = %f\n", __func__, *Tdmsks);
+	DML_LOG_VERBOSE("DML::%s: Tdmsks = %f\n", __func__, *Tdmsks);
 #endif
 }
 
@@ -4962,11 +4897,11 @@ static double get_urgent_bandwidth_required(
 		l->adj_factor_cur_pre = UrgentBurstFactorCursorPre[k];
 
 		bool is_phantom = dml_is_phantom_pipe(&display_cfg->plane_descriptors[k]);
-		bool exclude_this_plane = 0;
+		bool exclude_this_plane = false;
 
 		// Exclude phantom pipe in bw calculation for non svp prefetch state
 		if (state_type != dml2_core_internal_soc_state_svp_prefetch && is_phantom)
-			exclude_this_plane = 1;
+			exclude_this_plane = true;
 
 		// The qualified row bandwidth, qual_row_bw, accounts for the regular non-flip row bandwidth when there is no possible immediate flip or HostVM invalidation flip.
 		// The qual_row_bw is zero if HostVM is possible and only non-zero and equal to row_bw(i) if immediate flip is not allowed for that pipe.
@@ -4995,12 +4930,12 @@ static double get_urgent_bandwidth_required(
 			surface_peak_required_bw[k] = math_max2(surface_required_bw[k], surface_peak_required_bw[k]);
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%d, max1: vm_row_bw=%f\n", __func__, k, l->vm_row_bw);
-			dml2_printf("DML::%s: k=%d, max2: flip_and_active_bw=%f\n", __func__, k, l->flip_and_active_bw);
-			dml2_printf("DML::%s: k=%d, max3: flip_and_prefetch_bw=%f\n", __func__, k, l->flip_and_prefetch_bw);
-			dml2_printf("DML::%s: k=%d, max4: active_and_excess_bw=%f\n", __func__, k, l->active_and_excess_bw);
-			dml2_printf("DML::%s: k=%d, surface_required_bw=%f\n", __func__, k, surface_required_bw[k]);
-			dml2_printf("DML::%s: k=%d, surface_peak_required_bw=%f\n", __func__, k, surface_peak_required_bw[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, max1: vm_row_bw=%f\n", __func__, k, l->vm_row_bw);
+			DML_LOG_VERBOSE("DML::%s: k=%d, max2: flip_and_active_bw=%f\n", __func__, k, l->flip_and_active_bw);
+			DML_LOG_VERBOSE("DML::%s: k=%d, max3: flip_and_prefetch_bw=%f\n", __func__, k, l->flip_and_prefetch_bw);
+			DML_LOG_VERBOSE("DML::%s: k=%d, max4: active_and_excess_bw=%f\n", __func__, k, l->active_and_excess_bw);
+			DML_LOG_VERBOSE("DML::%s: k=%d, surface_required_bw=%f\n", __func__, k, surface_required_bw[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, surface_peak_required_bw=%f\n", __func__, k, surface_peak_required_bw[k]);
 #endif
 		} else {
 			surface_required_bw[k] = 0.0;
@@ -5009,34 +4944,34 @@ static double get_urgent_bandwidth_required(
 		l->required_bandwidth_mbps += surface_required_bw[k];
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%d, NumberOfDPP=%d\n", __func__, k, NumberOfDPP[k]);
-		dml2_printf("DML::%s: k=%d, use_qual_row_bw=%d\n", __func__, k, use_qual_row_bw);
-		dml2_printf("DML::%s: k=%d, immediate_flip=%d\n", __func__, k, display_cfg->plane_descriptors[k].immediate_flip);
-		dml2_printf("DML::%s: k=%d, mall_svp_prefetch_factor=%f\n", __func__, k, l->mall_svp_prefetch_factor);
-		dml2_printf("DML::%s: k=%d, adj_factor_p0=%f\n", __func__, k, l->adj_factor_p0);
-		dml2_printf("DML::%s: k=%d, adj_factor_p1=%f\n", __func__, k, l->adj_factor_p1);
-		dml2_printf("DML::%s: k=%d, adj_factor_cur=%f\n", __func__, k, l->adj_factor_cur);
+		DML_LOG_VERBOSE("DML::%s: k=%d, NumberOfDPP=%d\n", __func__, k, NumberOfDPP[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, use_qual_row_bw=%d\n", __func__, k, use_qual_row_bw);
+		DML_LOG_VERBOSE("DML::%s: k=%d, immediate_flip=%d\n", __func__, k, display_cfg->plane_descriptors[k].immediate_flip);
+		DML_LOG_VERBOSE("DML::%s: k=%d, mall_svp_prefetch_factor=%f\n", __func__, k, l->mall_svp_prefetch_factor);
+		DML_LOG_VERBOSE("DML::%s: k=%d, adj_factor_p0=%f\n", __func__, k, l->adj_factor_p0);
+		DML_LOG_VERBOSE("DML::%s: k=%d, adj_factor_p1=%f\n", __func__, k, l->adj_factor_p1);
+		DML_LOG_VERBOSE("DML::%s: k=%d, adj_factor_cur=%f\n", __func__, k, l->adj_factor_cur);
 
-		dml2_printf("DML::%s: k=%d, adj_factor_p0_pre=%f\n", __func__, k, l->adj_factor_p0_pre);
-		dml2_printf("DML::%s: k=%d, adj_factor_p1_pre=%f\n", __func__, k, l->adj_factor_p1_pre);
-		dml2_printf("DML::%s: k=%d, adj_factor_cur_pre=%f\n", __func__, k, l->adj_factor_cur_pre);
+		DML_LOG_VERBOSE("DML::%s: k=%d, adj_factor_p0_pre=%f\n", __func__, k, l->adj_factor_p0_pre);
+		DML_LOG_VERBOSE("DML::%s: k=%d, adj_factor_p1_pre=%f\n", __func__, k, l->adj_factor_p1_pre);
+		DML_LOG_VERBOSE("DML::%s: k=%d, adj_factor_cur_pre=%f\n", __func__, k, l->adj_factor_cur_pre);
 
-		dml2_printf("DML::%s: k=%d, per_plane_flip_bw=%f\n", __func__, k, l->per_plane_flip_bw[k]);
-		dml2_printf("DML::%s: k=%d, prefetch_vmrow_bw=%f\n", __func__, k, prefetch_vmrow_bw[k]);
-		dml2_printf("DML::%s: k=%d, ReadBandwidthLuma=%f\n", __func__, k, ReadBandwidthLuma[k]);
-		dml2_printf("DML::%s: k=%d, ReadBandwidthChroma=%f\n", __func__, k, ReadBandwidthChroma[k]);
-		dml2_printf("DML::%s: k=%d, excess_vactive_fill_bw_l=%f\n", __func__, k, excess_vactive_fill_bw_l[k]);
-		dml2_printf("DML::%s: k=%d, excess_vactive_fill_bw_c=%f\n", __func__, k, excess_vactive_fill_bw_c[k]);
-		dml2_printf("DML::%s: k=%d, cursor_bw=%f\n", __func__, k, cursor_bw[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, per_plane_flip_bw=%f\n", __func__, k, l->per_plane_flip_bw[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, prefetch_vmrow_bw=%f\n", __func__, k, prefetch_vmrow_bw[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, ReadBandwidthLuma=%f\n", __func__, k, ReadBandwidthLuma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, ReadBandwidthChroma=%f\n", __func__, k, ReadBandwidthChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, excess_vactive_fill_bw_l=%f\n", __func__, k, excess_vactive_fill_bw_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, excess_vactive_fill_bw_c=%f\n", __func__, k, excess_vactive_fill_bw_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, cursor_bw=%f\n", __func__, k, cursor_bw[k]);
 
-		dml2_printf("DML::%s: k=%d, meta_row_bw=%f\n", __func__, k, meta_row_bw[k]);
-		dml2_printf("DML::%s: k=%d, dpte_row_bw=%f\n", __func__, k, dpte_row_bw[k]);
-		dml2_printf("DML::%s: k=%d, PrefetchBandwidthLuma=%f\n", __func__, k, PrefetchBandwidthLuma[k]);
-		dml2_printf("DML::%s: k=%d, PrefetchBandwidthChroma=%f\n", __func__, k, PrefetchBandwidthChroma[k]);
-		dml2_printf("DML::%s: k=%d, prefetch_cursor_bw=%f\n", __func__, k, prefetch_cursor_bw[k]);
-		dml2_printf("DML::%s: k=%d, required_bandwidth_mbps=%f (total), inc_flip_bw=%d, is_phantom=%d exclude_this_plane=%d\n", __func__, k, l->required_bandwidth_mbps, inc_flip_bw, is_phantom, exclude_this_plane);
-		dml2_printf("DML::%s: k=%d, required_bandwidth_mbps=%f (total), soc_state=%s, inc_flip_bw=%d, is_phantom=%d exclude_this_plane=%d\n", __func__, k, l->required_bandwidth_mbps, dml2_core_internal_soc_state_type_str(state_type), inc_flip_bw, is_phantom, exclude_this_plane);
-		dml2_printf("DML::%s: k=%d, required_bandwidth_mbps=%f (total), inc_flip_bw=%d, is_phantom=%d exclude_this_plane=%d\n", __func__, k, l->required_bandwidth_mbps, inc_flip_bw, is_phantom, exclude_this_plane);
+		DML_LOG_VERBOSE("DML::%s: k=%d, meta_row_bw=%f\n", __func__, k, meta_row_bw[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, dpte_row_bw=%f\n", __func__, k, dpte_row_bw[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, PrefetchBandwidthLuma=%f\n", __func__, k, PrefetchBandwidthLuma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, PrefetchBandwidthChroma=%f\n", __func__, k, PrefetchBandwidthChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, prefetch_cursor_bw=%f\n", __func__, k, prefetch_cursor_bw[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, required_bandwidth_mbps=%f (total), inc_flip_bw=%d, is_phantom=%d exclude_this_plane=%d\n", __func__, k, l->required_bandwidth_mbps, inc_flip_bw, is_phantom, exclude_this_plane);
+		DML_LOG_VERBOSE("DML::%s: k=%d, required_bandwidth_mbps=%f (total), soc_state=%s, inc_flip_bw=%d, is_phantom=%d exclude_this_plane=%d\n", __func__, k, l->required_bandwidth_mbps, dml2_core_internal_soc_state_type_str(state_type), inc_flip_bw, is_phantom, exclude_this_plane);
+		DML_LOG_VERBOSE("DML::%s: k=%d, required_bandwidth_mbps=%f (total), inc_flip_bw=%d, is_phantom=%d exclude_this_plane=%d\n", __func__, k, l->required_bandwidth_mbps, inc_flip_bw, is_phantom, exclude_this_plane);
 #endif
 	}
 
@@ -5120,19 +5055,19 @@ static void CalculateExtraLatency(
 	*ExtraLatency_sr = *ExtraLatency_sr + Tarb;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: qos_type=%u\n", __func__, qos_type);
-	dml2_printf("DML::%s: hostvm_mode=%u\n", __func__, hostvm_mode);
-	dml2_printf("DML::%s: Tex_trips=%u\n", __func__, Tex_trips);
-	dml2_printf("DML::%s: max_outstanding_when_urgent_expected=%u\n", __func__, max_outstanding_when_urgent_expected);
-	dml2_printf("DML::%s: FabricClock=%f\n", __func__, FabricClock);
-	dml2_printf("DML::%s: DCFCLK=%f\n", __func__, DCFCLK);
-	dml2_printf("DML::%s: ReturnBW=%f\n", __func__, ReturnBW);
-	dml2_printf("DML::%s: RoundTripPingLatencyCycles=%u\n", __func__, RoundTripPingLatencyCycles);
-	dml2_printf("DML::%s: ReorderingBytes=%u\n", __func__, ReorderingBytes);
-	dml2_printf("DML::%s: Tarb=%f\n", __func__, Tarb);
-	dml2_printf("DML::%s: ExtraLatency=%f\n", __func__, *ExtraLatency);
-	dml2_printf("DML::%s: ExtraLatency_sr=%f\n", __func__, *ExtraLatency_sr);
-	dml2_printf("DML::%s: ExtraLatencyPrefetch=%f\n", __func__, *ExtraLatencyPrefetch);
+	DML_LOG_VERBOSE("DML::%s: qos_type=%u\n", __func__, qos_type);
+	DML_LOG_VERBOSE("DML::%s: hostvm_mode=%u\n", __func__, hostvm_mode);
+	DML_LOG_VERBOSE("DML::%s: Tex_trips=%f\n", __func__, Tex_trips);
+	DML_LOG_VERBOSE("DML::%s: max_outstanding_when_urgent_expected=%u\n", __func__, max_outstanding_when_urgent_expected);
+	DML_LOG_VERBOSE("DML::%s: FabricClock=%f\n", __func__, FabricClock);
+	DML_LOG_VERBOSE("DML::%s: DCFCLK=%f\n", __func__, DCFCLK);
+	DML_LOG_VERBOSE("DML::%s: ReturnBW=%f\n", __func__, ReturnBW);
+	DML_LOG_VERBOSE("DML::%s: RoundTripPingLatencyCycles=%u\n", __func__, RoundTripPingLatencyCycles);
+	DML_LOG_VERBOSE("DML::%s: ReorderingBytes=%u\n", __func__, ReorderingBytes);
+	DML_LOG_VERBOSE("DML::%s: Tarb=%f\n", __func__, Tarb);
+	DML_LOG_VERBOSE("DML::%s: ExtraLatency=%f\n", __func__, *ExtraLatency);
+	DML_LOG_VERBOSE("DML::%s: ExtraLatency_sr=%f\n", __func__, *ExtraLatency_sr);
+	DML_LOG_VERBOSE("DML::%s: ExtraLatencyPrefetch=%f\n", __func__, *ExtraLatencyPrefetch);
 #endif
 }
 
@@ -5199,20 +5134,20 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		s->HostVMDynamicLevelsTrips = 0;
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: dcc_enable = %u\n", __func__, p->dcc_enable);
-	dml2_printf("DML::%s: mrq_present = %u\n", __func__, p->mrq_present);
-	dml2_printf("DML::%s: dcc_mrq_enable = %u\n", __func__, dcc_mrq_enable);
-	dml2_printf("DML::%s: GPUVMEnable = %u\n", __func__, p->display_cfg->gpuvm_enable);
-	dml2_printf("DML::%s: GPUVMPageTableLevels = %u\n", __func__, p->display_cfg->gpuvm_max_page_table_levels);
-	dml2_printf("DML::%s: DCCEnable = %u\n", __func__, p->myPipe->DCCEnable);
-	dml2_printf("DML::%s: VStartup = %u\n", __func__, p->VStartup);
-	dml2_printf("DML::%s: HostVMEnable = %u\n", __func__, p->display_cfg->hostvm_enable);
-	dml2_printf("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, p->HostVMInefficiencyFactor);
-	dml2_printf("DML::%s: TWait = %f\n", __func__, p->TWait);
-	dml2_printf("DML::%s: TWait_p = %f\n", __func__, s->TWait_p);
-	dml2_printf("DML::%s: Ttrip = %f\n", __func__, p->Ttrip);
-	dml2_printf("DML::%s: myPipe->Dppclk = %f\n", __func__, p->myPipe->Dppclk);
-	dml2_printf("DML::%s: myPipe->Dispclk = %f\n", __func__, p->myPipe->Dispclk);
+	DML_LOG_VERBOSE("DML::%s: dcc_enable = %u\n", __func__, p->dcc_enable);
+	DML_LOG_VERBOSE("DML::%s: mrq_present = %u\n", __func__, p->mrq_present);
+	DML_LOG_VERBOSE("DML::%s: dcc_mrq_enable = %u\n", __func__, dcc_mrq_enable);
+	DML_LOG_VERBOSE("DML::%s: GPUVMEnable = %u\n", __func__, p->display_cfg->gpuvm_enable);
+	DML_LOG_VERBOSE("DML::%s: GPUVMPageTableLevels = %u\n", __func__, p->display_cfg->gpuvm_max_page_table_levels);
+	DML_LOG_VERBOSE("DML::%s: DCCEnable = %u\n", __func__, p->myPipe->DCCEnable);
+	DML_LOG_VERBOSE("DML::%s: VStartup = %u\n", __func__, p->VStartup);
+	DML_LOG_VERBOSE("DML::%s: HostVMEnable = %u\n", __func__, p->display_cfg->hostvm_enable);
+	DML_LOG_VERBOSE("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, p->HostVMInefficiencyFactor);
+	DML_LOG_VERBOSE("DML::%s: TWait = %f\n", __func__, p->TWait);
+	DML_LOG_VERBOSE("DML::%s: TWait_p = %f\n", __func__, s->TWait_p);
+	DML_LOG_VERBOSE("DML::%s: Ttrip = %f\n", __func__, p->Ttrip);
+	DML_LOG_VERBOSE("DML::%s: myPipe->Dppclk = %f\n", __func__, p->myPipe->Dppclk);
+	DML_LOG_VERBOSE("DML::%s: myPipe->Dispclk = %f\n", __func__, p->myPipe->Dispclk);
 #endif
 	CalculateVUpdateAndDynamicMetadataParameters(
 		p->MaxInterDCNTileRepeaters,
@@ -5258,11 +5193,11 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 	if (p->DynamicMetadataEnable == true) {
 		if (p->VStartup * s->LineTime < *p->TSetup + *p->Tdmdl + s->Tdmbf + s->Tdmec + s->Tdmsks) {
 			*p->NotEnoughTimeForDynamicMetadata = true;
-			dml2_printf("DML::%s: Not Enough Time for Dynamic Meta!\n", __func__);
-			dml2_printf("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", __func__, s->Tdmbf);
-			dml2_printf("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, s->Tdmec);
-			dml2_printf("DML::%s: Tdmsks: %fus - time before active dmd must complete transmission at dio\n", __func__, s->Tdmsks);
-			dml2_printf("DML::%s: Tdmdl: %fus - time for fabric to become ready and fetch dmd \n", __func__, *p->Tdmdl);
+			DML_LOG_VERBOSE("DML::%s: Not Enough Time for Dynamic Meta!\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", __func__, s->Tdmbf);
+			DML_LOG_VERBOSE("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, s->Tdmec);
+			DML_LOG_VERBOSE("DML::%s: Tdmsks: %fus - time before active dmd must complete transmission at dio\n", __func__, s->Tdmsks);
+			DML_LOG_VERBOSE("DML::%s: Tdmdl: %fus - time for fabric to become ready and fetch dmd \n", __func__, *p->Tdmdl);
 		} else {
 			*p->NotEnoughTimeForDynamicMetadata = false;
 		}
@@ -5288,21 +5223,21 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		((p->myPipe->ODMMode == dml2_odm_mode_mso_1to4) ? (double)p->myPipe->HActive * 3.0 / 4.0 : 0));
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DynamicMetadataVMEnabled = %u\n", __func__, p->DynamicMetadataVMEnabled);
-	dml2_printf("DML::%s: DPPCycles = %u\n", __func__, s->DPPCycles);
-	dml2_printf("DML::%s: PixelClock = %f\n", __func__, p->myPipe->PixelClock);
-	dml2_printf("DML::%s: Dppclk = %f\n", __func__, p->myPipe->Dppclk);
-	dml2_printf("DML::%s: DISPCLKCycles = %u\n", __func__, s->DISPCLKCycles);
-	dml2_printf("DML::%s: DISPCLK = %f\n", __func__, p->myPipe->Dispclk);
-	dml2_printf("DML::%s: DSCDelay = %u\n", __func__, p->DSCDelay);
-	dml2_printf("DML::%s: ODMMode = %u\n", __func__, p->myPipe->ODMMode);
-	dml2_printf("DML::%s: DPP_RECOUT_WIDTH = %u\n", __func__, p->DPP_RECOUT_WIDTH);
-	dml2_printf("DML::%s: DSTXAfterScaler = %u\n", __func__, *p->DSTXAfterScaler);
+	DML_LOG_VERBOSE("DML::%s: DynamicMetadataVMEnabled = %u\n", __func__, p->DynamicMetadataVMEnabled);
+	DML_LOG_VERBOSE("DML::%s: DPPCycles = %u\n", __func__, s->DPPCycles);
+	DML_LOG_VERBOSE("DML::%s: PixelClock = %f\n", __func__, p->myPipe->PixelClock);
+	DML_LOG_VERBOSE("DML::%s: Dppclk = %f\n", __func__, p->myPipe->Dppclk);
+	DML_LOG_VERBOSE("DML::%s: DISPCLKCycles = %u\n", __func__, s->DISPCLKCycles);
+	DML_LOG_VERBOSE("DML::%s: DISPCLK = %f\n", __func__, p->myPipe->Dispclk);
+	DML_LOG_VERBOSE("DML::%s: DSCDelay = %u\n", __func__, p->DSCDelay);
+	DML_LOG_VERBOSE("DML::%s: ODMMode = %u\n", __func__, p->myPipe->ODMMode);
+	DML_LOG_VERBOSE("DML::%s: DPP_RECOUT_WIDTH = %u\n", __func__, p->DPP_RECOUT_WIDTH);
+	DML_LOG_VERBOSE("DML::%s: DSTXAfterScaler = %u\n", __func__, *p->DSTXAfterScaler);
 
-	dml2_printf("DML::%s: setup_for_tdlut = %u\n", __func__, p->setup_for_tdlut);
-	dml2_printf("DML::%s: tdlut_opt_time = %f\n", __func__, p->tdlut_opt_time);
-	dml2_printf("DML::%s: tdlut_pte_bytes_per_frame = %u\n", __func__, p->tdlut_pte_bytes_per_frame);
-	dml2_printf("DML::%s: tdlut_drain_time = %f\n", __func__, p->tdlut_drain_time);
+	DML_LOG_VERBOSE("DML::%s: setup_for_tdlut = %u\n", __func__, p->setup_for_tdlut);
+	DML_LOG_VERBOSE("DML::%s: tdlut_opt_time = %f\n", __func__, p->tdlut_opt_time);
+	DML_LOG_VERBOSE("DML::%s: tdlut_pte_bytes_per_frame = %u\n", __func__, p->tdlut_pte_bytes_per_frame);
+	DML_LOG_VERBOSE("DML::%s: tdlut_drain_time = %f\n", __func__, p->tdlut_drain_time);
 #endif
 
 	if (p->OutputFormat == dml2_420 || (p->myPipe->InterlaceEnable && p->myPipe->ProgressiveToInterlaceUnitInOPP))
@@ -5314,17 +5249,17 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 	*p->DSTYAfterScaler = (unsigned int)(math_floor2(s->DSTTotalPixelsAfterScaler / p->myPipe->HTotal, 1));
 	*p->DSTXAfterScaler = (unsigned int)(s->DSTTotalPixelsAfterScaler - ((double)(*p->DSTYAfterScaler * p->myPipe->HTotal)));
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DSTXAfterScaler = %u (final)\n", __func__, *p->DSTXAfterScaler);
-	dml2_printf("DML::%s: DSTYAfterScaler = %u (final)\n", __func__, *p->DSTYAfterScaler);
+	DML_LOG_VERBOSE("DML::%s: DSTXAfterScaler = %u (final)\n", __func__, *p->DSTXAfterScaler);
+	DML_LOG_VERBOSE("DML::%s: DSTYAfterScaler = %u (final)\n", __func__, *p->DSTYAfterScaler);
 #endif
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: Tr0_trips = %f\n", __func__, *p->Tr0_trips);
-	dml2_printf("DML::%s: Tvm_trips = %f\n", __func__, *p->Tvm_trips);
-	dml2_printf("DML::%s: trip_to_mem = %f\n", __func__, s->trip_to_mem);
-	dml2_printf("DML::%s: ExtraLatencyPrefetch = %f\n", __func__, p->ExtraLatencyPrefetch);
-	dml2_printf("DML::%s: GPUVMPageTableLevels = %u\n", __func__, p->display_cfg->gpuvm_max_page_table_levels);
-	dml2_printf("DML::%s: HostVMDynamicLevelsTrips = %u\n", __func__, s->HostVMDynamicLevelsTrips);
+	DML_LOG_VERBOSE("DML::%s: Tr0_trips = %f\n", __func__, *p->Tr0_trips);
+	DML_LOG_VERBOSE("DML::%s: Tvm_trips = %f\n", __func__, *p->Tvm_trips);
+	DML_LOG_VERBOSE("DML::%s: trip_to_mem = %f\n", __func__, s->trip_to_mem);
+	DML_LOG_VERBOSE("DML::%s: ExtraLatencyPrefetch = %f\n", __func__, p->ExtraLatencyPrefetch);
+	DML_LOG_VERBOSE("DML::%s: GPUVMPageTableLevels = %u\n", __func__, p->display_cfg->gpuvm_max_page_table_levels);
+	DML_LOG_VERBOSE("DML::%s: HostVMDynamicLevelsTrips = %u\n", __func__, s->HostVMDynamicLevelsTrips);
 #endif
 	if (p->display_cfg->gpuvm_enable) {
 		s->Tvm_trips_rounded = math_ceil2(4.0 * *p->Tvm_trips / s->LineTime, 1.0) / 4.0 * s->LineTime;
@@ -5402,7 +5337,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 	}
 
 	/* oto prefetch bw should be always be less than total vactive bw */
-	//DML2_ASSERT(s->prefetch_bw_oto < s->per_pipe_vactive_sw_bw * p->myPipe->DPPPerSurface);
+	//DML_ASSERT(s->prefetch_bw_oto < s->per_pipe_vactive_sw_bw * p->myPipe->DPPPerSurface);
 
 	s->prefetch_bw_oto = math_max2(s->per_pipe_vactive_sw_bw, s->prefetch_bw_oto) * p->mall_prefetch_sdp_overhead_factor;
 
@@ -5421,9 +5356,9 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 	*p->RequiredPrefetchBWOTO = s->prefetch_bw_oto;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: vactive_sw_bw_l = %f\n", __func__, p->vactive_sw_bw_l);
-	dml2_printf("DML::%s: vactive_sw_bw_c = %f\n", __func__, p->vactive_sw_bw_c);
-	dml2_printf("DML::%s: per_pipe_vactive_sw_bw = %f\n", __func__, s->per_pipe_vactive_sw_bw);
+	DML_LOG_VERBOSE("DML::%s: vactive_sw_bw_l = %f\n", __func__, p->vactive_sw_bw_l);
+	DML_LOG_VERBOSE("DML::%s: vactive_sw_bw_c = %f\n", __func__, p->vactive_sw_bw_c);
+	DML_LOG_VERBOSE("DML::%s: per_pipe_vactive_sw_bw = %f\n", __func__, s->per_pipe_vactive_sw_bw);
 #endif
 
 	if (p->display_cfg->gpuvm_enable == true) {
@@ -5433,9 +5368,9 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			s->LineTime / 4.0);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: Tvm_oto max0 = %f\n", __func__, *p->Tvm_trips);
-		dml2_printf("DML::%s: Tvm_oto max1 = %f\n", __func__, *p->Tno_bw + vm_bytes * p->HostVMInefficiencyFactor / s->prefetch_bw_oto);
-		dml2_printf("DML::%s: Tvm_oto max2 = %f\n", __func__, s->LineTime / 4.0);
+		DML_LOG_VERBOSE("DML::%s: Tvm_oto max0 = %f\n", __func__, *p->Tvm_trips);
+		DML_LOG_VERBOSE("DML::%s: Tvm_oto max1 = %f\n", __func__, *p->Tno_bw + vm_bytes * p->HostVMInefficiencyFactor / s->prefetch_bw_oto);
+		DML_LOG_VERBOSE("DML::%s: Tvm_oto max2 = %f\n", __func__, s->LineTime / 4.0);
 #endif
 	} else {
 		s->Tvm_oto = s->Tvm_trips_rounded;
@@ -5447,9 +5382,9 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			(p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes) / s->prefetch_bw_oto,
 			s->LineTime / 4.0);
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: Tr0_oto max0 = %f\n", __func__, *p->Tr0_trips);
-		dml2_printf("DML::%s: Tr0_oto max1 = %f\n", __func__, (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes) / s->prefetch_bw_oto);
-		dml2_printf("DML::%s: Tr0_oto max2 = %f\n", __func__, s->LineTime / 4);
+		DML_LOG_VERBOSE("DML::%s: Tr0_oto max0 = %f\n", __func__, *p->Tr0_trips);
+		DML_LOG_VERBOSE("DML::%s: Tr0_oto max1 = %f\n", __func__, (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes) / s->prefetch_bw_oto);
+		DML_LOG_VERBOSE("DML::%s: Tr0_oto max2 = %f\n", __func__, s->LineTime / 4);
 #endif
 	} else
 		s->Tr0_oto = s->LineTime / 4.0;
@@ -5459,11 +5394,11 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 	s->dst_y_prefetch_oto = s->Tvm_oto_lines + 2 * s->Tr0_oto_lines + s->Lsw_oto;
 
 #ifdef DML_GLOBAL_PREFETCH_CHECK
-	dml2_printf("DML::%s: impacted_Tpre = %f\n", __func__, p->impacted_dst_y_pre);
+	DML_LOG_VERBOSE("DML::%s: impacted_Tpre = %f\n", __func__, p->impacted_dst_y_pre);
 	if (p->impacted_dst_y_pre > 0) {
-		dml2_printf("DML::%s: dst_y_prefetch_oto = %f\n", __func__, s->dst_y_prefetch_oto);
+		DML_LOG_VERBOSE("DML::%s: dst_y_prefetch_oto = %f\n", __func__, s->dst_y_prefetch_oto);
 		s->dst_y_prefetch_oto = math_max2(s->dst_y_prefetch_oto, p->impacted_dst_y_pre);
-		dml2_printf("DML::%s: dst_y_prefetch_oto = %f (impacted)\n", __func__, s->dst_y_prefetch_oto);
+		DML_LOG_VERBOSE("DML::%s: dst_y_prefetch_oto = %f (impacted)\n", __func__, s->dst_y_prefetch_oto);
 	}
 #endif
 	*p->Tpre_oto = s->dst_y_prefetch_oto * s->LineTime;
@@ -5492,72 +5427,71 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 	s->dst_y_prefetch_equ = math_min2(s->dst_y_prefetch_equ, 63.75); // limit to the reg limit of U6.2 for DST_Y_PREFETCH
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: HTotal = %u\n", __func__, p->myPipe->HTotal);
-	dml2_printf("DML::%s: min_Lsw_oto = %f\n", __func__, s->min_Lsw_oto);
-	dml2_printf("DML::%s: min_Lsw_equ = %f\n", __func__, s->min_Lsw_equ);
-	dml2_printf("DML::%s: Tno_bw = %f\n", __func__, *p->Tno_bw);
-	dml2_printf("DML::%s: Tno_bw_flip = %f\n", __func__, *p->Tno_bw_flip);
-	dml2_printf("DML::%s: ExtraLatencyPrefetch = %f\n", __func__, p->ExtraLatencyPrefetch);
-	dml2_printf("DML::%s: trip_to_mem = %f\n", __func__, s->trip_to_mem);
-	dml2_printf("DML::%s: mall_prefetch_sdp_overhead_factor = %f\n", __func__, p->mall_prefetch_sdp_overhead_factor);
-	dml2_printf("DML::%s: BytePerPixelY = %u\n", __func__, p->myPipe->BytePerPixelY);
-	dml2_printf("DML::%s: PrefetchSourceLinesY = %f\n", __func__, p->PrefetchSourceLinesY);
-	dml2_printf("DML::%s: swath_width_luma_ub = %u\n", __func__, p->swath_width_luma_ub);
-	dml2_printf("DML::%s: BytePerPixelC = %u\n", __func__, p->myPipe->BytePerPixelC);
-	dml2_printf("DML::%s: PrefetchSourceLinesC = %f\n", __func__, p->PrefetchSourceLinesC);
-	dml2_printf("DML::%s: swath_width_chroma_ub = %u\n", __func__, p->swath_width_chroma_ub);
-	dml2_printf("DML::%s: prefetch_sw_bytes = %f\n", __func__, *p->prefetch_sw_bytes);
-	dml2_printf("DML::%s: max_Tsw = %f\n", __func__, s->max_Tsw);
-	dml2_printf("DML::%s: bytes_pp = %f\n", __func__, s->bytes_pp);
-	dml2_printf("DML::%s: vm_bytes = %u\n", __func__, vm_bytes);
-	dml2_printf("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, p->PixelPTEBytesPerRow);
-	dml2_printf("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, p->HostVMInefficiencyFactor);
-	dml2_printf("DML::%s: Tvm_trips = %f\n", __func__, *p->Tvm_trips);
-	dml2_printf("DML::%s: Tr0_trips = %f\n", __func__, *p->Tr0_trips);
-	dml2_printf("DML::%s: Tvm_trips_flip = %f\n", __func__, *p->Tvm_trips_flip);
-	dml2_printf("DML::%s: Tr0_trips_flip = %f\n", __func__, *p->Tr0_trips_flip);
-	dml2_printf("DML::%s: prefetch_bw_pr = %f\n", __func__, s->prefetch_bw_pr);
-	dml2_printf("DML::%s: prefetch_bw_oto = %f\n", __func__, s->prefetch_bw_oto);
-	dml2_printf("DML::%s: Tr0_oto = %f\n", __func__, s->Tr0_oto);
-	dml2_printf("DML::%s: Tvm_oto = %f\n", __func__, s->Tvm_oto);
-	dml2_printf("DML::%s: Tvm_oto_lines = %f\n", __func__, s->Tvm_oto_lines);
-	dml2_printf("DML::%s: Tr0_oto_lines = %f\n", __func__, s->Tr0_oto_lines);
-	dml2_printf("DML::%s: Lsw_oto = %f\n", __func__, s->Lsw_oto);
-	dml2_printf("DML::%s: dst_y_prefetch_oto = %f\n", __func__, s->dst_y_prefetch_oto);
-	dml2_printf("DML::%s: dst_y_prefetch_equ = %f\n", __func__, s->dst_y_prefetch_equ);
-	dml2_printf("DML::%s: tdlut_row_bytes = %d\n", __func__, tdlut_row_bytes);
-	dml2_printf("DML::%s: meta_row_bytes = %d\n", __func__, p->meta_row_bytes);
-#endif
-	double Tpre = s->dst_y_prefetch_equ * s->LineTime;
+	DML_LOG_VERBOSE("DML::%s: HTotal = %u\n", __func__, p->myPipe->HTotal);
+	DML_LOG_VERBOSE("DML::%s: min_Lsw_oto = %f\n", __func__, s->min_Lsw_oto);
+	DML_LOG_VERBOSE("DML::%s: min_Lsw_equ = %f\n", __func__, s->min_Lsw_equ);
+	DML_LOG_VERBOSE("DML::%s: Tno_bw = %f\n", __func__, *p->Tno_bw);
+	DML_LOG_VERBOSE("DML::%s: Tno_bw_flip = %f\n", __func__, *p->Tno_bw_flip);
+	DML_LOG_VERBOSE("DML::%s: ExtraLatencyPrefetch = %f\n", __func__, p->ExtraLatencyPrefetch);
+	DML_LOG_VERBOSE("DML::%s: trip_to_mem = %f\n", __func__, s->trip_to_mem);
+	DML_LOG_VERBOSE("DML::%s: mall_prefetch_sdp_overhead_factor = %f\n", __func__, p->mall_prefetch_sdp_overhead_factor);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelY = %u\n", __func__, p->myPipe->BytePerPixelY);
+	DML_LOG_VERBOSE("DML::%s: PrefetchSourceLinesY = %f\n", __func__, p->PrefetchSourceLinesY);
+	DML_LOG_VERBOSE("DML::%s: swath_width_luma_ub = %u\n", __func__, p->swath_width_luma_ub);
+	DML_LOG_VERBOSE("DML::%s: BytePerPixelC = %u\n", __func__, p->myPipe->BytePerPixelC);
+	DML_LOG_VERBOSE("DML::%s: PrefetchSourceLinesC = %f\n", __func__, p->PrefetchSourceLinesC);
+	DML_LOG_VERBOSE("DML::%s: swath_width_chroma_ub = %u\n", __func__, p->swath_width_chroma_ub);
+	DML_LOG_VERBOSE("DML::%s: prefetch_sw_bytes = %f\n", __func__, *p->prefetch_sw_bytes);
+	DML_LOG_VERBOSE("DML::%s: max_Tsw = %f\n", __func__, s->max_Tsw);
+	DML_LOG_VERBOSE("DML::%s: bytes_pp = %f\n", __func__, s->bytes_pp);
+	DML_LOG_VERBOSE("DML::%s: vm_bytes = %u\n", __func__, vm_bytes);
+	DML_LOG_VERBOSE("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, p->PixelPTEBytesPerRow);
+	DML_LOG_VERBOSE("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, p->HostVMInefficiencyFactor);
+	DML_LOG_VERBOSE("DML::%s: Tvm_trips = %f\n", __func__, *p->Tvm_trips);
+	DML_LOG_VERBOSE("DML::%s: Tr0_trips = %f\n", __func__, *p->Tr0_trips);
+	DML_LOG_VERBOSE("DML::%s: Tvm_trips_flip = %f\n", __func__, *p->Tvm_trips_flip);
+	DML_LOG_VERBOSE("DML::%s: Tr0_trips_flip = %f\n", __func__, *p->Tr0_trips_flip);
+	DML_LOG_VERBOSE("DML::%s: prefetch_bw_pr = %f\n", __func__, s->prefetch_bw_pr);
+	DML_LOG_VERBOSE("DML::%s: prefetch_bw_oto = %f\n", __func__, s->prefetch_bw_oto);
+	DML_LOG_VERBOSE("DML::%s: Tr0_oto = %f\n", __func__, s->Tr0_oto);
+	DML_LOG_VERBOSE("DML::%s: Tvm_oto = %f\n", __func__, s->Tvm_oto);
+	DML_LOG_VERBOSE("DML::%s: Tvm_oto_lines = %f\n", __func__, s->Tvm_oto_lines);
+	DML_LOG_VERBOSE("DML::%s: Tr0_oto_lines = %f\n", __func__, s->Tr0_oto_lines);
+	DML_LOG_VERBOSE("DML::%s: Lsw_oto = %f\n", __func__, s->Lsw_oto);
+	DML_LOG_VERBOSE("DML::%s: dst_y_prefetch_oto = %f\n", __func__, s->dst_y_prefetch_oto);
+	DML_LOG_VERBOSE("DML::%s: dst_y_prefetch_equ = %f\n", __func__, s->dst_y_prefetch_equ);
+	DML_LOG_VERBOSE("DML::%s: tdlut_row_bytes = %d\n", __func__, tdlut_row_bytes);
+	DML_LOG_VERBOSE("DML::%s: meta_row_bytes = %d\n", __func__, p->meta_row_bytes);
+#endif
 	s->dst_y_prefetch_equ = math_floor2(4.0 * (s->dst_y_prefetch_equ + 0.125), 1) / 4.0;
 	*p->Tpre_rounded = s->dst_y_prefetch_equ * s->LineTime;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: dst_y_prefetch_equ: %f (after round)\n", __func__, s->dst_y_prefetch_equ);
-	dml2_printf("DML::%s: LineTime: %f\n", __func__, s->LineTime);
-	dml2_printf("DML::%s: VStartup: %u\n", __func__, p->VStartup);
-	dml2_printf("DML::%s: Tvstartup: %fus - time between vstartup and first pixel of active\n", __func__, p->VStartup * s->LineTime);
-	dml2_printf("DML::%s: TSetup: %fus - time from vstartup to vready\n", __func__, *p->TSetup);
-	dml2_printf("DML::%s: TCalc: %fus - time for calculations in dchub starting at vready\n", __func__, p->TCalc);
-	dml2_printf("DML::%s: TWait: %fus - time for fabric to become ready max(pstate exit,cstate enter/exit, urgent latency) after TCalc\n", __func__, p->TWait);
-	dml2_printf("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", __func__, s->Tdmbf);
-	dml2_printf("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, s->Tdmec);
-	dml2_printf("DML::%s: Tdmsks: %fus - time before active dmd must complete transmission at dio\n", __func__, s->Tdmsks);
-	dml2_printf("DML::%s: TWait = %f\n", __func__, p->TWait);
-	dml2_printf("DML::%s: TWait_p = %f\n", __func__, s->TWait_p);
-	dml2_printf("DML::%s: Ttrip = %f\n", __func__, p->Ttrip);
-	dml2_printf("DML::%s: Tex = %f\n", __func__, p->ExtraLatencyPrefetch);
-	dml2_printf("DML::%s: Tdmdl_vm: %fus - time for vm stages of dmd \n", __func__, *p->Tdmdl_vm);
-	dml2_printf("DML::%s: Tdmdl: %fus - time for fabric to become ready and fetch dmd \n", __func__, *p->Tdmdl);
-	dml2_printf("DML::%s: TWait_p: %fus\n", __func__, s->TWait_p);
-	dml2_printf("DML::%s: Ttrip: %fus\n", __func__, p->Ttrip);
-	dml2_printf("DML::%s: DSTXAfterScaler: %u pixels - number of pixel clocks pipeline and buffer delay after scaler \n", __func__, *p->DSTXAfterScaler);
-	dml2_printf("DML::%s: DSTYAfterScaler: %u lines - number of lines of pipeline and buffer delay after scaler \n", __func__, *p->DSTYAfterScaler);
-	dml2_printf("DML::%s: vm_bytes: %f (hvm inefficiency scaled)\n", __func__, vm_bytes*p->HostVMInefficiencyFactor);
-	dml2_printf("DML::%s: row_bytes: %f (hvm inefficiency scaled, 1 row)\n", __func__, p->PixelPTEBytesPerRow*p->HostVMInefficiencyFactor+p->meta_row_bytes+tdlut_row_bytes);
-	dml2_printf("DML::%s: Tno_bw: %f\n", __func__, *p->Tno_bw);
-	dml2_printf("DML::%s: Tpre=%f Tpre_rounded: %f, delta=%f\n", __func__, Tpre, *p->Tpre_rounded, (*p->Tpre_rounded - Tpre));
-	dml2_printf("DML::%s: Tvm_trips=%f Tvm_trips_rounded: %f, delta=%f\n", __func__, *p->Tvm_trips, s->Tvm_trips_rounded, (s->Tvm_trips_rounded - *p->Tvm_trips));
+	DML_LOG_VERBOSE("DML::%s: dst_y_prefetch_equ: %f (after round)\n", __func__, s->dst_y_prefetch_equ);
+	DML_LOG_VERBOSE("DML::%s: LineTime: %f\n", __func__, s->LineTime);
+	DML_LOG_VERBOSE("DML::%s: VStartup: %u\n", __func__, p->VStartup);
+	DML_LOG_VERBOSE("DML::%s: Tvstartup: %fus - time between vstartup and first pixel of active\n", __func__, p->VStartup * s->LineTime);
+	DML_LOG_VERBOSE("DML::%s: TSetup: %fus - time from vstartup to vready\n", __func__, *p->TSetup);
+	DML_LOG_VERBOSE("DML::%s: TCalc: %fus - time for calculations in dchub starting at vready\n", __func__, p->TCalc);
+	DML_LOG_VERBOSE("DML::%s: TWait: %fus - time for fabric to become ready max(pstate exit,cstate enter/exit, urgent latency) after TCalc\n", __func__, p->TWait);
+	DML_LOG_VERBOSE("DML::%s: Tdmbf: %fus - time for dmd transfer from dchub to dio output buffer\n", __func__, s->Tdmbf);
+	DML_LOG_VERBOSE("DML::%s: Tdmec: %fus - time dio takes to transfer dmd\n", __func__, s->Tdmec);
+	DML_LOG_VERBOSE("DML::%s: Tdmsks: %fus - time before active dmd must complete transmission at dio\n", __func__, s->Tdmsks);
+	DML_LOG_VERBOSE("DML::%s: TWait = %f\n", __func__, p->TWait);
+	DML_LOG_VERBOSE("DML::%s: TWait_p = %f\n", __func__, s->TWait_p);
+	DML_LOG_VERBOSE("DML::%s: Ttrip = %f\n", __func__, p->Ttrip);
+	DML_LOG_VERBOSE("DML::%s: Tex = %f\n", __func__, p->ExtraLatencyPrefetch);
+	DML_LOG_VERBOSE("DML::%s: Tdmdl_vm: %fus - time for vm stages of dmd \n", __func__, *p->Tdmdl_vm);
+	DML_LOG_VERBOSE("DML::%s: Tdmdl: %fus - time for fabric to become ready and fetch dmd \n", __func__, *p->Tdmdl);
+	DML_LOG_VERBOSE("DML::%s: TWait_p: %fus\n", __func__, s->TWait_p);
+	DML_LOG_VERBOSE("DML::%s: Ttrip: %fus\n", __func__, p->Ttrip);
+	DML_LOG_VERBOSE("DML::%s: DSTXAfterScaler: %u pixels - number of pixel clocks pipeline and buffer delay after scaler \n", __func__, *p->DSTXAfterScaler);
+	DML_LOG_VERBOSE("DML::%s: DSTYAfterScaler: %u lines - number of lines of pipeline and buffer delay after scaler \n", __func__, *p->DSTYAfterScaler);
+	DML_LOG_VERBOSE("DML::%s: vm_bytes: %f (hvm inefficiency scaled)\n", __func__, vm_bytes*p->HostVMInefficiencyFactor);
+	DML_LOG_VERBOSE("DML::%s: row_bytes: %f (hvm inefficiency scaled, 1 row)\n", __func__, p->PixelPTEBytesPerRow*p->HostVMInefficiencyFactor+p->meta_row_bytes+tdlut_row_bytes);
+	DML_LOG_VERBOSE("DML::%s: Tno_bw: %f\n", __func__, *p->Tno_bw);
+	DML_LOG_VERBOSE("DML::%s: Tpre=%f Tpre_rounded: %f, delta=%f\n", __func__, (s->dst_y_prefetch_equ * s->LineTime), *p->Tpre_rounded, (*p->Tpre_rounded - (s->dst_y_prefetch_equ * s->LineTime)));
+	DML_LOG_VERBOSE("DML::%s: Tvm_trips=%f Tvm_trips_rounded: %f, delta=%f\n", __func__, *p->Tvm_trips, s->Tvm_trips_rounded, (s->Tvm_trips_rounded - *p->Tvm_trips));
 #endif
 
 	*p->dst_y_per_vm_vblank = 0;
@@ -5596,19 +5530,19 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		} else
 			s->prefetch_bw1 = 0;
 
-		dml2_printf("DML::%s: prefetch_bw1: %f\n", __func__, s->prefetch_bw1);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw1: %f\n", __func__, s->prefetch_bw1);
 		if ((s->Tsw_est1 < s->min_Lsw_equ * s->LineTime) && (*p->Tpre_rounded - s->min_Lsw_equ * s->LineTime - 0.75 * s->LineTime - *p->Tno_bw > 0)) {
 			s->prefetch_bw1 = (vm_bytes * p->HostVMInefficiencyFactor + 2 * (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes)) /
 				(*p->Tpre_rounded - s->min_Lsw_equ * s->LineTime - 0.75 * s->LineTime - *p->Tno_bw);
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: vm and 2 rows bytes = %f\n", __func__, (vm_bytes * p->HostVMInefficiencyFactor + 2 * (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes)));
-			dml2_printf("DML::%s: Tpre_rounded = %f\n", __func__, *p->Tpre_rounded);
-			dml2_printf("DML::%s: minus term = %f\n", __func__, s->min_Lsw_equ * s->LineTime + 0.75 * s->LineTime + *p->Tno_bw);
-			dml2_printf("DML::%s: min_Lsw_equ = %f\n", __func__, s->min_Lsw_equ);
-			dml2_printf("DML::%s: LineTime = %f\n", __func__, s->LineTime);
-			dml2_printf("DML::%s: Tno_bw = %f\n", __func__, *p->Tno_bw);
-			dml2_printf("DML::%s: Time to fetch vm and 2 rows = %f\n", __func__, (*p->Tpre_rounded - s->min_Lsw_equ * s->LineTime - 0.75 * s->LineTime - *p->Tno_bw));
-			dml2_printf("DML::%s: prefetch_bw1: %f (updated)\n", __func__, s->prefetch_bw1);
+			DML_LOG_VERBOSE("DML::%s: vm and 2 rows bytes = %f\n", __func__, (vm_bytes * p->HostVMInefficiencyFactor + 2 * (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes)));
+			DML_LOG_VERBOSE("DML::%s: Tpre_rounded = %f\n", __func__, *p->Tpre_rounded);
+			DML_LOG_VERBOSE("DML::%s: minus term = %f\n", __func__, s->min_Lsw_equ * s->LineTime + 0.75 * s->LineTime + *p->Tno_bw);
+			DML_LOG_VERBOSE("DML::%s: min_Lsw_equ = %f\n", __func__, s->min_Lsw_equ);
+			DML_LOG_VERBOSE("DML::%s: LineTime = %f\n", __func__, s->LineTime);
+			DML_LOG_VERBOSE("DML::%s: Tno_bw = %f\n", __func__, *p->Tno_bw);
+			DML_LOG_VERBOSE("DML::%s: Time to fetch vm and 2 rows = %f\n", __func__, (*p->Tpre_rounded - s->min_Lsw_equ * s->LineTime - 0.75 * s->LineTime - *p->Tno_bw));
+			DML_LOG_VERBOSE("DML::%s: prefetch_bw1: %f (updated)\n", __func__, s->prefetch_bw1);
 #endif
 		}
 
@@ -5620,10 +5554,10 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		} else
 			s->prefetch_bw2 = 0;
 
-		dml2_printf("DML::%s: prefetch_bw2: %f\n", __func__, s->prefetch_bw2);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw2: %f\n", __func__, s->prefetch_bw2);
 		if ((s->Tsw_est2 < s->min_Lsw_equ * s->LineTime) && ((*p->Tpre_rounded - *p->Tno_bw - 2.0 * s->Tr0_trips_rounded - s->min_Lsw_equ * s->LineTime - 0.25 * s->LineTime) > 0)) {
 			s->prefetch_bw2 = vm_bytes * p->HostVMInefficiencyFactor / (*p->Tpre_rounded - *p->Tno_bw - 2.0 * s->Tr0_trips_rounded - s->min_Lsw_equ * s->LineTime - 0.25 * s->LineTime);
-			dml2_printf("DML::%s: prefetch_bw2: %f (updated)\n", __func__, s->prefetch_bw2);
+			DML_LOG_VERBOSE("DML::%s: prefetch_bw2: %f (updated)\n", __func__, s->prefetch_bw2);
 		}
 
 		// prefetch_bw3: 2*R0 + SW
@@ -5634,10 +5568,10 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		} else
 			s->prefetch_bw3 = 0;
 
-		dml2_printf("DML::%s: prefetch_bw3: %f\n", __func__, s->prefetch_bw3);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw3: %f\n", __func__, s->prefetch_bw3);
 		if ((s->Tsw_est3 < s->min_Lsw_equ * s->LineTime) && ((*p->Tpre_rounded - s->min_Lsw_equ * s->LineTime - 0.5 * s->LineTime - s->Tvm_trips_rounded) > 0)) {
 			s->prefetch_bw3 = (2 * (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes)) / (*p->Tpre_rounded - s->min_Lsw_equ * s->LineTime - 0.5 * s->LineTime - s->Tvm_trips_rounded);
-			dml2_printf("DML::%s: prefetch_bw3: %f (updated)\n", __func__, s->prefetch_bw3);
+			DML_LOG_VERBOSE("DML::%s: prefetch_bw3: %f (updated)\n", __func__, s->prefetch_bw3);
 		}
 
 		// prefetch_bw4: SW
@@ -5647,17 +5581,17 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			s->prefetch_bw4 = 0;
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: Tno_bw: %f\n", __func__, *p->Tno_bw);
-		dml2_printf("DML::%s: Tpre=%f Tpre_rounded: %f, delta=%f\n", __func__, Tpre, *p->Tpre_rounded, (*p->Tpre_rounded - Tpre));
-		dml2_printf("DML::%s: Tvm_trips=%f Tvm_trips_rounded: %f, delta=%f\n", __func__, *p->Tvm_trips, s->Tvm_trips_rounded, (s->Tvm_trips_rounded - *p->Tvm_trips));
-		dml2_printf("DML::%s: Tr0_trips=%f Tr0_trips_rounded: %f, delta=%f\n", __func__, *p->Tr0_trips, s->Tr0_trips_rounded, (s->Tr0_trips_rounded - *p->Tr0_trips));
-		dml2_printf("DML::%s: Tsw_est1: %f\n", __func__, s->Tsw_est1);
-		dml2_printf("DML::%s: Tsw_est2: %f\n", __func__, s->Tsw_est2);
-		dml2_printf("DML::%s: Tsw_est3: %f\n", __func__, s->Tsw_est3);
-		dml2_printf("DML::%s: prefetch_bw1: %f (final)\n", __func__, s->prefetch_bw1);
-		dml2_printf("DML::%s: prefetch_bw2: %f (final)\n", __func__, s->prefetch_bw2);
-		dml2_printf("DML::%s: prefetch_bw3: %f (final)\n", __func__, s->prefetch_bw3);
-		dml2_printf("DML::%s: prefetch_bw4: %f (final)\n", __func__, s->prefetch_bw4);
+		DML_LOG_VERBOSE("DML::%s: Tno_bw: %f\n", __func__, *p->Tno_bw);
+		DML_LOG_VERBOSE("DML::%s: Tpre=%f Tpre_rounded: %f, delta=%f\n", __func__, s->dst_y_prefetch_equ * s->LineTime, *p->Tpre_rounded, (*p->Tpre_rounded - (s->dst_y_prefetch_equ * s->LineTime)));
+		DML_LOG_VERBOSE("DML::%s: Tvm_trips=%f Tvm_trips_rounded: %f, delta=%f\n", __func__, *p->Tvm_trips, s->Tvm_trips_rounded, (s->Tvm_trips_rounded - *p->Tvm_trips));
+		DML_LOG_VERBOSE("DML::%s: Tr0_trips=%f Tr0_trips_rounded: %f, delta=%f\n", __func__, *p->Tr0_trips, s->Tr0_trips_rounded, (s->Tr0_trips_rounded - *p->Tr0_trips));
+		DML_LOG_VERBOSE("DML::%s: Tsw_est1: %f\n", __func__, s->Tsw_est1);
+		DML_LOG_VERBOSE("DML::%s: Tsw_est2: %f\n", __func__, s->Tsw_est2);
+		DML_LOG_VERBOSE("DML::%s: Tsw_est3: %f\n", __func__, s->Tsw_est3);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw1: %f (final)\n", __func__, s->prefetch_bw1);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw2: %f (final)\n", __func__, s->prefetch_bw2);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw3: %f (final)\n", __func__, s->prefetch_bw3);
+		DML_LOG_VERBOSE("DML::%s: prefetch_bw4: %f (final)\n", __func__, s->prefetch_bw4);
 #endif
 		{
 			bool Case1OK = false;
@@ -5676,14 +5610,14 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 
 			double total_row_bytes = (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes);
 
-			dml2_printf("DML::%s: Tvm_trips_rounded = %f\n", __func__, s->Tvm_trips_rounded);
-			dml2_printf("DML::%s: Tr0_trips_rounded = %f\n", __func__, s->Tr0_trips_rounded);
+			DML_LOG_VERBOSE("DML::%s: Tvm_trips_rounded = %f\n", __func__, s->Tvm_trips_rounded);
+			DML_LOG_VERBOSE("DML::%s: Tr0_trips_rounded = %f\n", __func__, s->Tr0_trips_rounded);
 
 			if (s->prefetch_bw1 > 0) {
 				double vm_transfer_time = *p->Tno_bw + vm_bytes * p->HostVMInefficiencyFactor / s->prefetch_bw1;
 				double row_transfer_time = total_row_bytes / s->prefetch_bw1;
-				dml2_printf("DML::%s: Case1: vm_transfer_time  = %f\n", __func__, vm_transfer_time);
-				dml2_printf("DML::%s: Case1: row_transfer_time = %f\n", __func__, row_transfer_time);
+				DML_LOG_VERBOSE("DML::%s: Case1: vm_transfer_time  = %f\n", __func__, vm_transfer_time);
+				DML_LOG_VERBOSE("DML::%s: Case1: row_transfer_time = %f\n", __func__, row_transfer_time);
 				if (vm_transfer_time >= s->Tvm_trips_rounded && row_transfer_time >= s->Tr0_trips_rounded) {
 					Case1OK = true;
 				}
@@ -5696,8 +5630,8 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			if (s->prefetch_bw2 > 0) {
 				double vm_transfer_time = *p->Tno_bw + vm_bytes * p->HostVMInefficiencyFactor / s->prefetch_bw2;
 				double row_transfer_time = total_row_bytes / s->prefetch_bw2;
-				dml2_printf("DML::%s: Case2: vm_transfer_time  = %f\n", __func__, vm_transfer_time);
-				dml2_printf("DML::%s: Case2: row_transfer_time = %f\n", __func__, row_transfer_time);
+				DML_LOG_VERBOSE("DML::%s: Case2: vm_transfer_time  = %f\n", __func__, vm_transfer_time);
+				DML_LOG_VERBOSE("DML::%s: Case2: row_transfer_time = %f\n", __func__, row_transfer_time);
 				if (vm_transfer_time >= s->Tvm_trips_rounded && row_transfer_time < s->Tr0_trips_rounded) {
 					Case2OK = true;
 				}
@@ -5709,8 +5643,8 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			if (s->prefetch_bw3 > 0) {
 				double vm_transfer_time = *p->Tno_bw + vm_bytes * p->HostVMInefficiencyFactor / s->prefetch_bw3;
 				double row_transfer_time = total_row_bytes / s->prefetch_bw3;
-				dml2_printf("DML::%s: Case3: vm_transfer_time  = %f\n", __func__, vm_transfer_time);
-				dml2_printf("DML::%s: Case3: row_transfer_time = %f\n", __func__, row_transfer_time);
+				DML_LOG_VERBOSE("DML::%s: Case3: vm_transfer_time  = %f\n", __func__, vm_transfer_time);
+				DML_LOG_VERBOSE("DML::%s: Case3: row_transfer_time = %f\n", __func__, row_transfer_time);
 				if (vm_transfer_time < s->Tvm_trips_rounded && row_transfer_time >= s->Tr0_trips_rounded) {
 					Case3OK = true;
 				}
@@ -5730,10 +5664,10 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 							p->vm_bytes * p->HostVMInefficiencyFactor / (31 * s->LineTime) - *p->Tno_bw,
 							(p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + p->meta_row_bytes + tdlut_row_bytes) / (15 * s->LineTime));
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: Case1OK: %u\n", __func__, Case1OK);
-			dml2_printf("DML::%s: Case2OK: %u\n", __func__, Case2OK);
-			dml2_printf("DML::%s: Case3OK: %u\n", __func__, Case3OK);
-			dml2_printf("DML::%s: prefetch_bw_equ: %f\n", __func__, s->prefetch_bw_equ);
+			DML_LOG_VERBOSE("DML::%s: Case1OK: %u\n", __func__, Case1OK);
+			DML_LOG_VERBOSE("DML::%s: Case2OK: %u\n", __func__, Case2OK);
+			DML_LOG_VERBOSE("DML::%s: Case3OK: %u\n", __func__, Case3OK);
+			DML_LOG_VERBOSE("DML::%s: prefetch_bw_equ: %f\n", __func__, s->prefetch_bw_equ);
 #endif
 
 			if (s->prefetch_bw_equ > 0) {
@@ -5753,12 +5687,12 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			} else {
 				s->Tvm_equ = 0;
 				s->Tr0_equ = 0;
-				dml2_printf("DML::%s: prefetch_bw_equ equals 0!\n", __func__);
+				DML_LOG_VERBOSE("DML::%s: prefetch_bw_equ equals 0!\n", __func__);
 			}
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: Tvm_equ = %f\n", __func__, s->Tvm_equ);
-		dml2_printf("DML::%s: Tr0_equ = %f\n", __func__, s->Tr0_equ);
+		DML_LOG_VERBOSE("DML::%s: Tvm_equ = %f\n", __func__, s->Tvm_equ);
+		DML_LOG_VERBOSE("DML::%s: Tr0_equ = %f\n", __func__, s->Tr0_equ);
 #endif
 		// Use the more stressful prefetch schedule
 		if (s->dst_y_prefetch_oto < s->dst_y_prefetch_equ) {
@@ -5769,7 +5703,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			*p->dst_y_per_vm_vblank = math_ceil2(4.0 * s->TimeForFetchingVM / s->LineTime, 1.0) / 4.0;
 			*p->dst_y_per_row_vblank = math_ceil2(4.0 * s->TimeForFetchingRowInVBlank / s->LineTime, 1.0) / 4.0;
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: Using oto scheduling for prefetch\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: Using oto scheduling for prefetch\n", __func__);
 #endif
 
 		} else {
@@ -5785,7 +5719,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		*p->dst_y_per_row_vblank = math_ceil2(4.0 * s->TimeForFetchingRowInVBlank / s->LineTime, 1.0) / 4.0;
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: Using equ bw scheduling for prefetch\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: Using equ bw scheduling for prefetch\n", __func__);
 #endif
 		}
 
@@ -5797,31 +5731,31 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		*p->prefetch_swath_time_us = (s->LinesToRequestPrefetchPixelData * s->LineTime);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: TimeForFetchingVM = %f\n", __func__, s->TimeForFetchingVM);
-		dml2_printf("DML::%s: TimeForFetchingRowInVBlank = %f\n", __func__, s->TimeForFetchingRowInVBlank);
-		dml2_printf("DML::%s: LineTime = %f\n", __func__, s->LineTime);
-		dml2_printf("DML::%s: dst_y_prefetch = %f\n", __func__, *p->dst_y_prefetch);
-		dml2_printf("DML::%s: dst_y_per_vm_vblank = %f\n", __func__, *p->dst_y_per_vm_vblank);
-		dml2_printf("DML::%s: dst_y_per_row_vblank = %f\n", __func__, *p->dst_y_per_row_vblank);
-		dml2_printf("DML::%s: LinesToRequestPrefetchPixelData = %f\n", __func__, s->LinesToRequestPrefetchPixelData);
-		dml2_printf("DML::%s: PrefetchSourceLinesY = %f\n", __func__, p->PrefetchSourceLinesY);
-		dml2_printf("DML::%s: prefetch_swath_time_us = %f\n", __func__, *p->prefetch_swath_time_us);
+		DML_LOG_VERBOSE("DML::%s: TimeForFetchingVM = %f\n", __func__, s->TimeForFetchingVM);
+		DML_LOG_VERBOSE("DML::%s: TimeForFetchingRowInVBlank = %f\n", __func__, s->TimeForFetchingRowInVBlank);
+		DML_LOG_VERBOSE("DML::%s: LineTime = %f\n", __func__, s->LineTime);
+		DML_LOG_VERBOSE("DML::%s: dst_y_prefetch = %f\n", __func__, *p->dst_y_prefetch);
+		DML_LOG_VERBOSE("DML::%s: dst_y_per_vm_vblank = %f\n", __func__, *p->dst_y_per_vm_vblank);
+		DML_LOG_VERBOSE("DML::%s: dst_y_per_row_vblank = %f\n", __func__, *p->dst_y_per_row_vblank);
+		DML_LOG_VERBOSE("DML::%s: LinesToRequestPrefetchPixelData = %f\n", __func__, s->LinesToRequestPrefetchPixelData);
+		DML_LOG_VERBOSE("DML::%s: PrefetchSourceLinesY = %f\n", __func__, p->PrefetchSourceLinesY);
+		DML_LOG_VERBOSE("DML::%s: prefetch_swath_time_us = %f\n", __func__, *p->prefetch_swath_time_us);
 
-		dml2_printf("DML::%s: cursor_bytes_per_chunk = %d\n", __func__, p->cursor_bytes_per_chunk);
-		dml2_printf("DML::%s: cursor_bytes_per_line = %d\n", __func__, p->cursor_bytes_per_line);
-		dml2_printf("DML::%s: cursor_prefetch_bytes = %d\n", __func__, s->cursor_prefetch_bytes);
-		dml2_printf("DML::%s: prefetch_cursor_bw = %f\n", __func__, *p->prefetch_cursor_bw);
+		DML_LOG_VERBOSE("DML::%s: cursor_bytes_per_chunk = %d\n", __func__, p->cursor_bytes_per_chunk);
+		DML_LOG_VERBOSE("DML::%s: cursor_bytes_per_line = %d\n", __func__, p->cursor_bytes_per_line);
+		DML_LOG_VERBOSE("DML::%s: cursor_prefetch_bytes = %d\n", __func__, s->cursor_prefetch_bytes);
+		DML_LOG_VERBOSE("DML::%s: prefetch_cursor_bw = %f\n", __func__, *p->prefetch_cursor_bw);
 #endif
-		DML2_ASSERT(*p->dst_y_prefetch < 64);
+		DML_ASSERT(*p->dst_y_prefetch < 64);
 
 		unsigned int min_lsw_required = (unsigned int)math_max2(2, p->tdlut_drain_time / s->LineTime);
 		if (s->LinesToRequestPrefetchPixelData >= min_lsw_required && s->prefetch_bw_equ > 0) {
 			*p->VRatioPrefetchY = (double)p->PrefetchSourceLinesY / s->LinesToRequestPrefetchPixelData;
 			*p->VRatioPrefetchY = math_max2(*p->VRatioPrefetchY, 1.0);
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: VRatioPrefetchY = %f\n", __func__, *p->VRatioPrefetchY);
-			dml2_printf("DML::%s: SwathHeightY = %u\n", __func__, p->SwathHeightY);
-			dml2_printf("DML::%s: VInitPreFillY = %u\n", __func__, p->VInitPreFillY);
+			DML_LOG_VERBOSE("DML::%s: VRatioPrefetchY = %f\n", __func__, *p->VRatioPrefetchY);
+			DML_LOG_VERBOSE("DML::%s: SwathHeightY = %u\n", __func__, p->SwathHeightY);
+			DML_LOG_VERBOSE("DML::%s: VInitPreFillY = %u\n", __func__, p->VInitPreFillY);
 #endif
 			if ((p->SwathHeightY > 4) && (p->VInitPreFillY > 3)) {
 				if (s->LinesToRequestPrefetchPixelData > (p->VInitPreFillY - 3.0) / 2.0) {
@@ -5829,13 +5763,13 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 						(double)p->MaxNumSwathY * p->SwathHeightY / (s->LinesToRequestPrefetchPixelData - (p->VInitPreFillY - 3.0) / 2.0));
 				} else {
 					s->NoTimeToPrefetch = true;
-					dml2_printf("DML::%s: No time to prefetch!. LinesToRequestPrefetchPixelData=%f VinitPreFillY=%u\n", __func__, s->LinesToRequestPrefetchPixelData, p->VInitPreFillY);
+					DML_LOG_VERBOSE("DML::%s: No time to prefetch!. LinesToRequestPrefetchPixelData=%f VinitPreFillY=%u\n", __func__, s->LinesToRequestPrefetchPixelData, p->VInitPreFillY);
 					*p->VRatioPrefetchY = 0;
 				}
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: VRatioPrefetchY = %f\n", __func__, *p->VRatioPrefetchY);
-				dml2_printf("DML::%s: PrefetchSourceLinesY = %f\n", __func__, p->PrefetchSourceLinesY);
-				dml2_printf("DML::%s: MaxNumSwathY = %u\n", __func__, p->MaxNumSwathY);
+				DML_LOG_VERBOSE("DML::%s: VRatioPrefetchY = %f\n", __func__, *p->VRatioPrefetchY);
+				DML_LOG_VERBOSE("DML::%s: PrefetchSourceLinesY = %f\n", __func__, p->PrefetchSourceLinesY);
+				DML_LOG_VERBOSE("DML::%s: MaxNumSwathY = %u\n", __func__, p->MaxNumSwathY);
 #endif
 			}
 
@@ -5843,22 +5777,22 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			*p->VRatioPrefetchC = math_max2(*p->VRatioPrefetchC, 1.0);
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: VRatioPrefetchC = %f\n", __func__, *p->VRatioPrefetchC);
-			dml2_printf("DML::%s: SwathHeightC = %u\n", __func__, p->SwathHeightC);
-			dml2_printf("DML::%s: VInitPreFillC = %u\n", __func__, p->VInitPreFillC);
+			DML_LOG_VERBOSE("DML::%s: VRatioPrefetchC = %f\n", __func__, *p->VRatioPrefetchC);
+			DML_LOG_VERBOSE("DML::%s: SwathHeightC = %u\n", __func__, p->SwathHeightC);
+			DML_LOG_VERBOSE("DML::%s: VInitPreFillC = %u\n", __func__, p->VInitPreFillC);
 #endif
 			if ((p->SwathHeightC > 4) && (p->VInitPreFillC > 3)) {
 				if (s->LinesToRequestPrefetchPixelData > (p->VInitPreFillC - 3.0) / 2.0) {
 					*p->VRatioPrefetchC = math_max2(*p->VRatioPrefetchC, (double)p->MaxNumSwathC * p->SwathHeightC / (s->LinesToRequestPrefetchPixelData - (p->VInitPreFillC - 3.0) / 2.0));
 				} else {
 					s->NoTimeToPrefetch = true;
-					dml2_printf("DML::%s: No time to prefetch!. LinesToRequestPrefetchPixelData=%f VInitPreFillC=%u\n", __func__, s->LinesToRequestPrefetchPixelData, p->VInitPreFillC);
+					DML_LOG_VERBOSE("DML::%s: No time to prefetch!. LinesToRequestPrefetchPixelData=%f VInitPreFillC=%u\n", __func__, s->LinesToRequestPrefetchPixelData, p->VInitPreFillC);
 					*p->VRatioPrefetchC = 0;
 				}
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: VRatioPrefetchC = %f\n", __func__, *p->VRatioPrefetchC);
-				dml2_printf("DML::%s: PrefetchSourceLinesC = %f\n", __func__, p->PrefetchSourceLinesC);
-				dml2_printf("DML::%s: MaxNumSwathC = %u\n", __func__, p->MaxNumSwathC);
+				DML_LOG_VERBOSE("DML::%s: VRatioPrefetchC = %f\n", __func__, *p->VRatioPrefetchC);
+				DML_LOG_VERBOSE("DML::%s: PrefetchSourceLinesC = %f\n", __func__, p->PrefetchSourceLinesC);
+				DML_LOG_VERBOSE("DML::%s: MaxNumSwathC = %u\n", __func__, p->MaxNumSwathC);
 #endif
 			}
 
@@ -5866,36 +5800,34 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			*p->RequiredPrefetchPixelDataBWChroma = (double)p->PrefetchSourceLinesC / s->LinesToRequestPrefetchPixelData * p->myPipe->BytePerPixelC * p->swath_width_chroma_ub / s->LineTime;
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: BytePerPixelY = %u\n", __func__, p->myPipe->BytePerPixelY);
-			dml2_printf("DML::%s: swath_width_luma_ub = %u\n", __func__, p->swath_width_luma_ub);
-			dml2_printf("DML::%s: LineTime = %f\n", __func__, s->LineTime);
-			dml2_printf("DML::%s: RequiredPrefetchPixelDataBWLuma = %f\n", __func__, *p->RequiredPrefetchPixelDataBWLuma);
-			dml2_printf("DML::%s: RequiredPrefetchPixelDataBWChroma = %f\n", __func__, *p->RequiredPrefetchPixelDataBWChroma);
+			DML_LOG_VERBOSE("DML::%s: BytePerPixelY = %u\n", __func__, p->myPipe->BytePerPixelY);
+			DML_LOG_VERBOSE("DML::%s: swath_width_luma_ub = %u\n", __func__, p->swath_width_luma_ub);
+			DML_LOG_VERBOSE("DML::%s: LineTime = %f\n", __func__, s->LineTime);
+			DML_LOG_VERBOSE("DML::%s: RequiredPrefetchPixelDataBWLuma = %f\n", __func__, *p->RequiredPrefetchPixelDataBWLuma);
+			DML_LOG_VERBOSE("DML::%s: RequiredPrefetchPixelDataBWChroma = %f\n", __func__, *p->RequiredPrefetchPixelDataBWChroma);
 #endif
 		} else {
 			s->NoTimeToPrefetch = true;
-			dml2_printf("DML::%s: No time to prefetch!, LinesToRequestPrefetchPixelData: %f, should be >= %d\n", __func__, s->LinesToRequestPrefetchPixelData, min_lsw_required);
-			dml2_printf("DML::%s: No time to prefetch!, prefetch_bw_equ: %f, should be > 0\n", __func__, s->prefetch_bw_equ);
+			DML_LOG_VERBOSE("DML::%s: No time to prefetch!, LinesToRequestPrefetchPixelData: %f, should be >= %d\n", __func__, s->LinesToRequestPrefetchPixelData, min_lsw_required);
+			DML_LOG_VERBOSE("DML::%s: No time to prefetch!, prefetch_bw_equ: %f, should be > 0\n", __func__, s->prefetch_bw_equ);
 			*p->VRatioPrefetchY = 0;
 			*p->VRatioPrefetchC = 0;
 			*p->RequiredPrefetchPixelDataBWLuma = 0;
 			*p->RequiredPrefetchPixelDataBWChroma = 0;
 		}
-		dml2_printf("DML: Tpre: %fus - sum of time to request 2 x data pte, swaths\n", (double)s->LinesToRequestPrefetchPixelData * s->LineTime + 2.0 * s->TimeForFetchingRowInVBlank + s->TimeForFetchingVM);
-		dml2_printf("DML: Tvm: %fus - time to fetch vm\n", s->TimeForFetchingVM);
-		dml2_printf("DML: Tr0: %fus - time to fetch first row of data pagetables\n", s->TimeForFetchingRowInVBlank);
-		dml2_printf("DML: Tsw: %fus = time to fetch enough pixel data and cursor data to feed the scalers init position and detile\n", (double)s->LinesToRequestPrefetchPixelData * s->LineTime);
-		dml2_printf("DML: To: %fus - time for propagation from scaler to optc\n", (*p->DSTYAfterScaler + ((double)(*p->DSTXAfterScaler) / (double)p->myPipe->HTotal)) * s->LineTime);
-		dml2_printf("DML: Tvstartup - TSetup - Tcalc - TWait - Tpre - To > 0\n");
-		dml2_printf("DML: Tslack(pre): %fus - time left over in schedule\n", p->VStartup * s->LineTime - s->TimeForFetchingVM - 2 * s->TimeForFetchingRowInVBlank - (*p->DSTYAfterScaler + ((double)(*p->DSTXAfterScaler) / (double)p->myPipe->HTotal)) * s->LineTime - p->TWait - p->TCalc - *p->TSetup);
-		dml2_printf("DML: row_bytes = dpte_row_bytes (per_pipe) = PixelPTEBytesPerRow = : %u\n", p->PixelPTEBytesPerRow);
+		DML_LOG_VERBOSE("DML: Tpre: %fus - sum of time to request 2 x data pte, swaths\n", (double)s->LinesToRequestPrefetchPixelData * s->LineTime + 2.0 * s->TimeForFetchingRowInVBlank + s->TimeForFetchingVM);
+		DML_LOG_VERBOSE("DML: Tvm: %fus - time to fetch vm\n", s->TimeForFetchingVM);
+		DML_LOG_VERBOSE("DML: Tr0: %fus - time to fetch first row of data pagetables\n", s->TimeForFetchingRowInVBlank);
+		DML_LOG_VERBOSE("DML: Tsw: %fus = time to fetch enough pixel data and cursor data to feed the scalers init position and detile\n", (double)s->LinesToRequestPrefetchPixelData * s->LineTime);
+		DML_LOG_VERBOSE("DML: To: %fus - time for propagation from scaler to optc\n", (*p->DSTYAfterScaler + ((double)(*p->DSTXAfterScaler) / (double)p->myPipe->HTotal)) * s->LineTime);
+		DML_LOG_VERBOSE("DML: Tvstartup - TSetup - Tcalc - TWait - Tpre - To > 0\n");
+		DML_LOG_VERBOSE("DML: Tslack(pre): %fus - time left over in schedule\n", p->VStartup * s->LineTime - s->TimeForFetchingVM - 2 * s->TimeForFetchingRowInVBlank - (*p->DSTYAfterScaler + ((double)(*p->DSTXAfterScaler) / (double)p->myPipe->HTotal)) * s->LineTime - p->TWait - p->TCalc - *p->TSetup);
+		DML_LOG_VERBOSE("DML: row_bytes = dpte_row_bytes (per_pipe) = PixelPTEBytesPerRow = : %u\n", p->PixelPTEBytesPerRow);
 
 	} else {
-		dml2_printf("DML::%s: No time to prefetch! dst_y_prefetch_equ = %f (should be > 1)\n", __func__, s->dst_y_prefetch_equ);
-		dml2_printf("DML::%s: No time to prefetch! min_Lsw_equ_ok = %d, Tpre_rounded (%f) should be >= Tvm_trips_rounded (%f) + 2.0*Tr0_trips_rounded (%f) + min_Tsw_equ (%f)\n",
+		DML_LOG_VERBOSE("DML::%s: No time to prefetch! dst_y_prefetch_equ = %f (should be > 1)\n", __func__, s->dst_y_prefetch_equ);
+		DML_LOG_VERBOSE("DML::%s: No time to prefetch! min_Lsw_equ_ok = %d, Tpre_rounded (%f) should be >= Tvm_trips_rounded (%f) + 2.0*Tr0_trips_rounded (%f) + min_Tsw_equ (%f)\n",
 				__func__, min_Lsw_equ_ok, *p->Tpre_rounded, s->Tvm_trips_rounded, 2.0*s->Tr0_trips_rounded, s->min_Lsw_equ*s->LineTime);
-		dml2_printf("DML::%s: No time to prefetch! min_Lsw_equ_ok = %d, Tpre_rounded+Tvm_trips_rounded+2.0*Tr0_trips_rounded+min_Tsw_equ (%f) should be > \n",
-				__func__, tpre_gt_req_latency, (s->min_Lsw_equ*s->LineTime + s->Tvm_trips_rounded + 2.0*s->Tr0_trips_rounded), p->Turg, s->trip_to_mem, p->ExtraLatencyPrefetch);
 		s->NoTimeToPrefetch = true;
 		s->TimeForFetchingVM = 0;
 		s->TimeForFetchingRowInVBlank = 0;
@@ -5916,18 +5848,18 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			prefetch_vm_bw = 0;
 		} else if (*p->dst_y_per_vm_vblank > 0) {
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, p->HostVMInefficiencyFactor);
-			dml2_printf("DML::%s: dst_y_per_vm_vblank = %f\n", __func__, *p->dst_y_per_vm_vblank);
-			dml2_printf("DML::%s: LineTime = %f\n", __func__, s->LineTime);
+			DML_LOG_VERBOSE("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, p->HostVMInefficiencyFactor);
+			DML_LOG_VERBOSE("DML::%s: dst_y_per_vm_vblank = %f\n", __func__, *p->dst_y_per_vm_vblank);
+			DML_LOG_VERBOSE("DML::%s: LineTime = %f\n", __func__, s->LineTime);
 #endif
 			prefetch_vm_bw = vm_bytes * p->HostVMInefficiencyFactor / (*p->dst_y_per_vm_vblank * s->LineTime);
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: prefetch_vm_bw = %f\n", __func__, prefetch_vm_bw);
+			DML_LOG_VERBOSE("DML::%s: prefetch_vm_bw = %f\n", __func__, prefetch_vm_bw);
 #endif
 		} else {
 			prefetch_vm_bw = 0;
 			s->NoTimeToPrefetch = true;
-			dml2_printf("DML::%s: No time to prefetch!. dst_y_per_vm_vblank=%f (should be > 0)\n", __func__, *p->dst_y_per_vm_vblank);
+			DML_LOG_VERBOSE("DML::%s: No time to prefetch!. dst_y_per_vm_vblank=%f (should be > 0)\n", __func__, *p->dst_y_per_vm_vblank);
 		}
 
 		if (p->PixelPTEBytesPerRow == 0 && tdlut_row_bytes == 0) {
@@ -5936,14 +5868,14 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 			prefetch_row_bw = (p->PixelPTEBytesPerRow * p->HostVMInefficiencyFactor + tdlut_row_bytes) / (*p->dst_y_per_row_vblank * s->LineTime);
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, p->PixelPTEBytesPerRow);
-			dml2_printf("DML::%s: dst_y_per_row_vblank = %f\n", __func__, *p->dst_y_per_row_vblank);
-			dml2_printf("DML::%s: prefetch_row_bw = %f\n", __func__, prefetch_row_bw);
+			DML_LOG_VERBOSE("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, p->PixelPTEBytesPerRow);
+			DML_LOG_VERBOSE("DML::%s: dst_y_per_row_vblank = %f\n", __func__, *p->dst_y_per_row_vblank);
+			DML_LOG_VERBOSE("DML::%s: prefetch_row_bw = %f\n", __func__, prefetch_row_bw);
 #endif
 		} else {
 			prefetch_row_bw = 0;
 			s->NoTimeToPrefetch = true;
-			dml2_printf("DML::%s: No time to prefetch!. dst_y_per_row_vblank=%f (should be > 0)\n", __func__, *p->dst_y_per_row_vblank);
+			DML_LOG_VERBOSE("DML::%s: No time to prefetch!. dst_y_per_row_vblank=%f (should be > 0)\n", __func__, *p->dst_y_per_row_vblank);
 		}
 
 		*p->prefetch_vmrow_bw = math_max2(prefetch_vm_bw, prefetch_row_bw);
@@ -5963,12 +5895,12 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch
 		*p->prefetch_vmrow_bw = 0;
 	}
 
-	dml2_printf("DML::%s: dst_y_per_vm_vblank = %f (final)\n", __func__, *p->dst_y_per_vm_vblank);
-	dml2_printf("DML::%s: dst_y_per_row_vblank = %f (final)\n", __func__, *p->dst_y_per_row_vblank);
-	dml2_printf("DML::%s: prefetch_vmrow_bw = %f (final)\n", __func__, *p->prefetch_vmrow_bw);
-	dml2_printf("DML::%s: RequiredPrefetchPixelDataBWLuma = %f (final)\n", __func__, *p->RequiredPrefetchPixelDataBWLuma);
-	dml2_printf("DML::%s: RequiredPrefetchPixelDataBWChroma = %f (final)\n", __func__, *p->RequiredPrefetchPixelDataBWChroma);
-	dml2_printf("DML::%s: NoTimeToPrefetch=%d\n", __func__, s->NoTimeToPrefetch);
+	DML_LOG_VERBOSE("DML::%s: dst_y_per_vm_vblank = %f (final)\n", __func__, *p->dst_y_per_vm_vblank);
+	DML_LOG_VERBOSE("DML::%s: dst_y_per_row_vblank = %f (final)\n", __func__, *p->dst_y_per_row_vblank);
+	DML_LOG_VERBOSE("DML::%s: prefetch_vmrow_bw = %f (final)\n", __func__, *p->prefetch_vmrow_bw);
+	DML_LOG_VERBOSE("DML::%s: RequiredPrefetchPixelDataBWLuma = %f (final)\n", __func__, *p->RequiredPrefetchPixelDataBWLuma);
+	DML_LOG_VERBOSE("DML::%s: RequiredPrefetchPixelDataBWChroma = %f (final)\n", __func__, *p->RequiredPrefetchPixelDataBWChroma);
+	DML_LOG_VERBOSE("DML::%s: NoTimeToPrefetch=%d\n", __func__, s->NoTimeToPrefetch);
 
 	return s->NoTimeToPrefetch;
 }
@@ -6005,7 +5937,7 @@ static unsigned int find_max_impact_plane(unsigned int this_plane_idx, unsigned
 		}
 	}
 	if (max_idx <= 0) {
-		DML2_ASSERT(max_idx >= 0);
+		DML_ASSERT(max_idx >= 0);
 		max_idx = this_plane_idx;
 	}
 
@@ -6037,12 +5969,12 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 	// worst case if the rob and cdb is fully hogged
 	s->max_Trpd_dcfclk_cycles = (unsigned int) math_ceil2((p->rob_buffer_size_kbytes*1024 + p->compressed_buffer_size_kbytes*DML_MAX_COMPRESSION_RATIO*1024)/64.0, 1.0);
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: num_active_planes = %d\n", __func__, p->num_active_planes);
-	dml2_printf("DML::%s: rob_buffer_size_kbytes = %d\n", __func__, p->rob_buffer_size_kbytes);
-	dml2_printf("DML::%s: compressed_buffer_size_kbytes = %d\n", __func__, p->compressed_buffer_size_kbytes);
-	dml2_printf("DML::%s: estimated_urg_bandwidth_required_mbps = %f\n", __func__, p->estimated_urg_bandwidth_required_mbps);
-	dml2_printf("DML::%s: estimated_dcfclk_mhz = %f\n", __func__, p->estimated_dcfclk_mhz);
-	dml2_printf("DML::%s: max_Trpd_dcfclk_cycles = %u\n", __func__, s->max_Trpd_dcfclk_cycles);
+	DML_LOG_VERBOSE("DML::%s: num_active_planes = %d\n", __func__, p->num_active_planes);
+	DML_LOG_VERBOSE("DML::%s: rob_buffer_size_kbytes = %d\n", __func__, p->rob_buffer_size_kbytes);
+	DML_LOG_VERBOSE("DML::%s: compressed_buffer_size_kbytes = %d\n", __func__, p->compressed_buffer_size_kbytes);
+	DML_LOG_VERBOSE("DML::%s: estimated_urg_bandwidth_required_mbps = %f\n", __func__, p->estimated_urg_bandwidth_required_mbps);
+	DML_LOG_VERBOSE("DML::%s: estimated_dcfclk_mhz = %f\n", __func__, p->estimated_dcfclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: max_Trpd_dcfclk_cycles = %u\n", __func__, s->max_Trpd_dcfclk_cycles);
 #endif
 
 	// calculate the return impact from each plane, request is 256B per dcfclk
@@ -6063,12 +5995,12 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 		s->burst_bytes_to_fill_det += (unsigned int) (math_floor2(p->lb_source_lines_l[i] / p->swath_height_l[i], 1) * s->src_swath_bytes_l[i]);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: i=%u pixel_format = %d\n", __func__, i, p->pixel_format[i]);
-		dml2_printf("DML::%s: i=%u chunk_bytes_l = %d\n", __func__, i, p->chunk_bytes_l);
-		dml2_printf("DML::%s: i=%u lb_source_lines_l = %d\n", __func__, i, p->lb_source_lines_l[i]);
-		dml2_printf("DML::%s: i=%u src_detile_buf_size_bytes_l=%d\n", __func__, i, s->src_detile_buf_size_bytes_l[i]);
-		dml2_printf("DML::%s: i=%u src_swath_bytes_l=%d\n", __func__, i, s->src_swath_bytes_l[i]);
-		dml2_printf("DML::%s: i=%u burst_bytes_to_fill_det=%d (luma)\n", __func__, i, s->burst_bytes_to_fill_det);
+		DML_LOG_VERBOSE("DML::%s: i=%u pixel_format = %d\n", __func__, i, p->pixel_format[i]);
+		DML_LOG_VERBOSE("DML::%s: i=%u chunk_bytes_l = %d\n", __func__, i, p->chunk_bytes_l);
+		DML_LOG_VERBOSE("DML::%s: i=%u lb_source_lines_l = %d\n", __func__, i, p->lb_source_lines_l[i]);
+		DML_LOG_VERBOSE("DML::%s: i=%u src_detile_buf_size_bytes_l=%d\n", __func__, i, s->src_detile_buf_size_bytes_l[i]);
+		DML_LOG_VERBOSE("DML::%s: i=%u src_swath_bytes_l=%d\n", __func__, i, s->src_swath_bytes_l[i]);
+		DML_LOG_VERBOSE("DML::%s: i=%u burst_bytes_to_fill_det=%d (luma)\n", __func__, i, s->burst_bytes_to_fill_det);
 #endif
 
 		if (s->src_swath_bytes_c[i] > 0) { // dual_plane
@@ -6079,10 +6011,10 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 			}
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: i=%u chunk_bytes_c = %d\n", __func__, i, p->chunk_bytes_c);
-			dml2_printf("DML::%s: i=%u lb_source_lines_c = %d\n", __func__, i, p->lb_source_lines_c[i]);
-			dml2_printf("DML::%s: i=%u src_detile_buf_size_bytes_c=%d\n", __func__, i, s->src_detile_buf_size_bytes_c[i]);
-			dml2_printf("DML::%s: i=%u src_swath_bytes_c=%d\n", __func__, i, s->src_swath_bytes_c[i]);
+			DML_LOG_VERBOSE("DML::%s: i=%u chunk_bytes_c = %d\n", __func__, i, p->chunk_bytes_c);
+			DML_LOG_VERBOSE("DML::%s: i=%u lb_source_lines_c = %d\n", __func__, i, p->lb_source_lines_c[i]);
+			DML_LOG_VERBOSE("DML::%s: i=%u src_detile_buf_size_bytes_c=%d\n", __func__, i, s->src_detile_buf_size_bytes_c[i]);
+			DML_LOG_VERBOSE("DML::%s: i=%u src_swath_bytes_c=%d\n", __func__, i, s->src_swath_bytes_c[i]);
 #endif
 		}
 
@@ -6090,9 +6022,9 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 		s->accumulated_return_path_dcfclk_cycles[i] = (unsigned int) math_ceil2(((DML_MAX_COMPRESSION_RATIO-1) * 64 * p->estimated_dcfclk_mhz) * s->time_to_fill_det_us / 64.0, 1.0); //for 64B per DCFClk
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: i=%u burst_bytes_to_fill_det=%d\n", __func__, i, s->burst_bytes_to_fill_det);
-		dml2_printf("DML::%s: i=%u time_to_fill_det_us=%f\n", __func__, i, s->time_to_fill_det_us);
-		dml2_printf("DML::%s: i=%u accumulated_return_path_dcfclk_cycles=%u\n", __func__, i, s->accumulated_return_path_dcfclk_cycles[i]);
+		DML_LOG_VERBOSE("DML::%s: i=%u burst_bytes_to_fill_det=%d\n", __func__, i, s->burst_bytes_to_fill_det);
+		DML_LOG_VERBOSE("DML::%s: i=%u time_to_fill_det_us=%f\n", __func__, i, s->time_to_fill_det_us);
+		DML_LOG_VERBOSE("DML::%s: i=%u accumulated_return_path_dcfclk_cycles=%u\n", __func__, i, s->accumulated_return_path_dcfclk_cycles[i]);
 #endif
 		// clamping to worst case delay which is one which occupy the full rob+cdb
 		if (s->accumulated_return_path_dcfclk_cycles[i] > s->max_Trpd_dcfclk_cycles)
@@ -6109,7 +6041,7 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 		p->impacted_dst_y_pre[i] = math_ceil2(p->impacted_dst_y_pre[i] / p->line_time[i], 0.25);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: i=%u impacted_Tpre=%f (k=%u)\n", __func__, i, p->impacted_dst_y_pre[i], k);
+		DML_LOG_VERBOSE("DML::%s: i=%u impacted_Tpre=%f (k=%u)\n", __func__, i, p->impacted_dst_y_pre[i], k);
 #endif
 	}
 
@@ -6120,8 +6052,8 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 				*p->recalc_prefetch_schedule = 1;
 			}
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: i=%u Tpre_rounded=%f\n", __func__, i, p->Tpre_rounded[i]);
-			dml2_printf("DML::%s: i=%u Tpre_oto=%f\n", __func__, i, p->Tpre_oto[i]);
+			DML_LOG_VERBOSE("DML::%s: i=%u Tpre_rounded=%f\n", __func__, i, p->Tpre_rounded[i]);
+			DML_LOG_VERBOSE("DML::%s: i=%u Tpre_oto=%f\n", __func__, i, p->Tpre_oto[i]);
 #endif
 		}
 	} else {
@@ -6131,8 +6063,8 @@ static noinline_for_stack bool CheckGlobalPrefetchAdmissibility(struct dml2_core
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: prefetch_global_check_passed=%u\n", __func__, s->prefetch_global_check_passed);
-	dml2_printf("DML::%s: recalc_prefetch_schedule=%u\n", __func__, *p->recalc_prefetch_schedule);
+	DML_LOG_VERBOSE("DML::%s: prefetch_global_check_passed=%u\n", __func__, s->prefetch_global_check_passed);
+	DML_LOG_VERBOSE("DML::%s: recalc_prefetch_schedule=%u\n", __func__, *p->recalc_prefetch_schedule);
 #endif
 
 	return s->prefetch_global_check_passed;
@@ -6150,8 +6082,8 @@ static void calculate_peak_bandwidth_required(
 	memset(l, 0, sizeof(struct dml2_core_shared_calculate_peak_bandwidth_required_locals));
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: inc_flip_bw = %d\n", __func__, p->inc_flip_bw);
-	dml2_printf("DML::%s: NumberOfActiveSurfaces = %d\n", __func__, p->num_active_planes);
+	DML_LOG_VERBOSE("DML::%s: inc_flip_bw = %d\n", __func__, p->inc_flip_bw);
+	DML_LOG_VERBOSE("DML::%s: NumberOfActiveSurfaces = %d\n", __func__, p->num_active_planes);
 #endif
 
 	for (unsigned int k = 0; k < p->num_active_planes; ++k) {
@@ -6347,12 +6279,12 @@ static void calculate_peak_bandwidth_required(
 				p->surface_peak_required_bw[m][n]);
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: urg_vactive_bandwidth_required%s[%s][%s]=%f\n", __func__, (p->inc_flip_bw ? "_flip" : ""), dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->urg_vactive_bandwidth_required[m][n]);
-			dml2_printf("DML::%s: urg_bandwidth_required%s[%s][%s]=%f\n", __func__, (p->inc_flip_bw ? "_flip" : ""), dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->urg_bandwidth_required[m][n]);
-			dml2_printf("DML::%s: urg_bandwidth_required_qual[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->urg_bandwidth_required[m][n]);
-			dml2_printf("DML::%s: non_urg_bandwidth_required%s[%s][%s]=%f\n", __func__, (p->inc_flip_bw ? "_flip" : ""), dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->non_urg_bandwidth_required[m][n]);
+			DML_LOG_VERBOSE("DML::%s: urg_vactive_bandwidth_required%s[%s][%s]=%f\n", __func__, (p->inc_flip_bw ? "_flip" : ""), dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->urg_vactive_bandwidth_required[m][n]);
+			DML_LOG_VERBOSE("DML::%s: urg_bandwidth_required%s[%s][%s]=%f\n", __func__, (p->inc_flip_bw ? "_flip" : ""), dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->urg_bandwidth_required[m][n]);
+			DML_LOG_VERBOSE("DML::%s: urg_bandwidth_required_qual[%s][%s]=%f\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->urg_bandwidth_required[m][n]);
+			DML_LOG_VERBOSE("DML::%s: non_urg_bandwidth_required%s[%s][%s]=%f\n", __func__, (p->inc_flip_bw ? "_flip" : ""), dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n), p->non_urg_bandwidth_required[m][n]);
 #endif
-			DML2_ASSERT(p->urg_bandwidth_required[m][n] >= p->non_urg_bandwidth_required[m][n]);
+			DML_ASSERT(p->urg_bandwidth_required[m][n] >= p->non_urg_bandwidth_required[m][n]);
 		}
 	}
 }
@@ -6414,18 +6346,18 @@ static void check_urgent_bandwidth_support(
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: frac_urg_bandwidth_nom_sdp = %f\n", __func__, frac_urg_bandwidth_nom_sdp);
-	dml2_printf("DML::%s: frac_urg_bandwidth_nom_dram = %f\n", __func__, frac_urg_bandwidth_nom_dram);
-	dml2_printf("DML::%s: frac_urg_bandwidth_nom = %f\n", __func__, *frac_urg_bandwidth_nom);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_nom_sdp = %f\n", __func__, frac_urg_bandwidth_nom_sdp);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_nom_dram = %f\n", __func__, frac_urg_bandwidth_nom_dram);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_nom = %f\n", __func__, *frac_urg_bandwidth_nom);
 
-	dml2_printf("DML::%s: frac_urg_bandwidth_mall_sdp = %f\n", __func__, frac_urg_bandwidth_mall_sdp);
-	dml2_printf("DML::%s: frac_urg_bandwidth_mall_dram = %f\n", __func__, frac_urg_bandwidth_mall_dram);
-	dml2_printf("DML::%s: frac_urg_bandwidth_mall = %f\n", __func__, *frac_urg_bandwidth_mall);
-	dml2_printf("DML::%s: bandwidth_support_ok = %d\n", __func__, *bandwidth_support_ok);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_mall_sdp = %f\n", __func__, frac_urg_bandwidth_mall_sdp);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_mall_dram = %f\n", __func__, frac_urg_bandwidth_mall_dram);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_mall = %f\n", __func__, *frac_urg_bandwidth_mall);
+	DML_LOG_VERBOSE("DML::%s: bandwidth_support_ok = %d\n", __func__, *bandwidth_support_ok);
 
 	for (unsigned int m = 0; m < dml2_core_internal_soc_state_max; m++) {
 		for (unsigned int n = 0; n < dml2_core_internal_bw_max; n++) {
-			dml2_printf("DML::%s: state:%s bw_type:%s urg_bandwidth_available=%f %s urg_bandwidth_required=%f\n",
+			DML_LOG_VERBOSE("DML::%s: state:%s bw_type:%s urg_bandwidth_available=%f %s urg_bandwidth_required=%f\n",
 			__func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n),
 			urg_bandwidth_available[m][n], (urg_bandwidth_available[m][n] < urg_bandwidth_required[m][n]) ? "<" : ">=", urg_bandwidth_required[m][n]);
 		}
@@ -6446,14 +6378,14 @@ static double get_bandwidth_available_for_immediate_flip(enum dml2_core_internal
 	flip_bw_available_mbps = flip_bw_available_sdp_mbps < flip_bw_available_dram_mbps ? flip_bw_available_sdp_mbps : flip_bw_available_dram_mbps;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: eval_state = %s\n", __func__, dml2_core_internal_soc_state_type_str(eval_state));
-	dml2_printf("DML::%s: urg_bandwidth_available_sdp_mbps = %f\n", __func__, urg_bandwidth_available[eval_state][dml2_core_internal_bw_sdp]);
-	dml2_printf("DML::%s: urg_bandwidth_available_dram_mbps = %f\n", __func__, urg_bandwidth_available[eval_state][dml2_core_internal_bw_dram]);
-	dml2_printf("DML::%s: urg_bandwidth_required_sdp_mbps = %f\n", __func__, urg_bandwidth_required[eval_state][dml2_core_internal_bw_sdp]);
-	dml2_printf("DML::%s: urg_bandwidth_required_dram_mbps = %f\n", __func__, urg_bandwidth_required[eval_state][dml2_core_internal_bw_dram]);
-	dml2_printf("DML::%s: flip_bw_available_sdp_mbps = %f\n", __func__, flip_bw_available_sdp_mbps);
-	dml2_printf("DML::%s: flip_bw_available_dram_mbps = %f\n", __func__, flip_bw_available_dram_mbps);
-	dml2_printf("DML::%s: flip_bw_available_mbps = %f\n", __func__, flip_bw_available_mbps);
+	DML_LOG_VERBOSE("DML::%s: eval_state = %s\n", __func__, dml2_core_internal_soc_state_type_str(eval_state));
+	DML_LOG_VERBOSE("DML::%s: urg_bandwidth_available_sdp_mbps = %f\n", __func__, urg_bandwidth_available[eval_state][dml2_core_internal_bw_sdp]);
+	DML_LOG_VERBOSE("DML::%s: urg_bandwidth_available_dram_mbps = %f\n", __func__, urg_bandwidth_available[eval_state][dml2_core_internal_bw_dram]);
+	DML_LOG_VERBOSE("DML::%s: urg_bandwidth_required_sdp_mbps = %f\n", __func__, urg_bandwidth_required[eval_state][dml2_core_internal_bw_sdp]);
+	DML_LOG_VERBOSE("DML::%s: urg_bandwidth_required_dram_mbps = %f\n", __func__, urg_bandwidth_required[eval_state][dml2_core_internal_bw_dram]);
+	DML_LOG_VERBOSE("DML::%s: flip_bw_available_sdp_mbps = %f\n", __func__, flip_bw_available_sdp_mbps);
+	DML_LOG_VERBOSE("DML::%s: flip_bw_available_dram_mbps = %f\n", __func__, flip_bw_available_dram_mbps);
+	DML_LOG_VERBOSE("DML::%s: flip_bw_available_mbps = %f\n", __func__, flip_bw_available_mbps);
 #endif
 
 	return flip_bw_available_mbps;
@@ -6478,28 +6410,28 @@ static void calculate_immediate_flip_bandwidth_support(
 		*flip_bandwidth_support_ok &= urg_bandwidth_available[eval_state][n] >= urg_bandwidth_required_flip[eval_state][n];
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: n = %s\n", __func__, dml2_core_internal_bw_type_str(n));
-		dml2_printf("DML::%s: urg_bandwidth_available = %f\n", __func__, urg_bandwidth_available[eval_state][n]);
-		dml2_printf("DML::%s: non_urg_bandwidth_required_flip = %f\n", __func__, non_urg_bandwidth_required_flip[eval_state][n]);
-		dml2_printf("DML::%s: urg_bandwidth_required_flip = %f\n", __func__, urg_bandwidth_required_flip[eval_state][n]);
-		dml2_printf("DML::%s: flip_bandwidth_support_ok = %d\n", __func__, *flip_bandwidth_support_ok);
+		DML_LOG_VERBOSE("DML::%s: n = %s\n", __func__, dml2_core_internal_bw_type_str(n));
+		DML_LOG_VERBOSE("DML::%s: urg_bandwidth_available = %f\n", __func__, urg_bandwidth_available[eval_state][n]);
+		DML_LOG_VERBOSE("DML::%s: non_urg_bandwidth_required_flip = %f\n", __func__, non_urg_bandwidth_required_flip[eval_state][n]);
+		DML_LOG_VERBOSE("DML::%s: urg_bandwidth_required_flip = %f\n", __func__, urg_bandwidth_required_flip[eval_state][n]);
+		DML_LOG_VERBOSE("DML::%s: flip_bandwidth_support_ok = %d\n", __func__, *flip_bandwidth_support_ok);
 #endif
-		DML2_ASSERT(urg_bandwidth_required_flip[eval_state][n] >= non_urg_bandwidth_required_flip[eval_state][n]);
+		DML_ASSERT(urg_bandwidth_required_flip[eval_state][n] >= non_urg_bandwidth_required_flip[eval_state][n]);
 	}
 
 	*frac_urg_bandwidth_flip = (frac_urg_bw_flip_sdp > frac_urg_bw_flip_dram) ? frac_urg_bw_flip_sdp : frac_urg_bw_flip_dram;
 	*flip_bandwidth_support_ok &= (*frac_urg_bandwidth_flip <= 1.0);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: eval_state = %s\n", __func__, dml2_core_internal_soc_state_type_str(eval_state));
-	dml2_printf("DML::%s: frac_urg_bw_flip_sdp = %f\n", __func__, frac_urg_bw_flip_sdp);
-	dml2_printf("DML::%s: frac_urg_bw_flip_dram = %f\n", __func__, frac_urg_bw_flip_dram);
-	dml2_printf("DML::%s: frac_urg_bandwidth_flip = %f\n", __func__, *frac_urg_bandwidth_flip);
-	dml2_printf("DML::%s: flip_bandwidth_support_ok = %d\n", __func__, *flip_bandwidth_support_ok);
+	DML_LOG_VERBOSE("DML::%s: eval_state = %s\n", __func__, dml2_core_internal_soc_state_type_str(eval_state));
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bw_flip_sdp = %f\n", __func__, frac_urg_bw_flip_sdp);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bw_flip_dram = %f\n", __func__, frac_urg_bw_flip_dram);
+	DML_LOG_VERBOSE("DML::%s: frac_urg_bandwidth_flip = %f\n", __func__, *frac_urg_bandwidth_flip);
+	DML_LOG_VERBOSE("DML::%s: flip_bandwidth_support_ok = %d\n", __func__, *flip_bandwidth_support_ok);
 
 	for (unsigned int m = 0; m < dml2_core_internal_soc_state_max; m++) {
 		for (unsigned int n = 0; n < dml2_core_internal_bw_max; n++) {
-			dml2_printf("DML::%s: state:%s bw_type:%s, urg_bandwidth_available=%f %s urg_bandwidth_required=%f\n",
+			DML_LOG_VERBOSE("DML::%s: state:%s bw_type:%s, urg_bandwidth_available=%f %s urg_bandwidth_required=%f\n",
 			__func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n),
 			urg_bandwidth_available[m][n], (urg_bandwidth_available[m][n] < urg_bandwidth_required_flip[m][n]) ? "<" : ">=", urg_bandwidth_required_flip[m][n]);
 		}
@@ -6549,27 +6481,27 @@ static void CalculateFlipSchedule(
 	l->dpte_row_bytes = DPTEBytesPerRow;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: GPUVMEnable = %u\n", __func__, GPUVMEnable);
-	dml2_printf("DML::%s: ip.max_flip_time_us = %d\n", __func__, max_flip_time_us);
-	dml2_printf("DML::%s: ip.max_flip_time_lines = %d\n", __func__, max_flip_time_lines);
-	dml2_printf("DML::%s: BandwidthAvailableForImmediateFlip = %f\n", __func__, BandwidthAvailableForImmediateFlip);
-	dml2_printf("DML::%s: TotImmediateFlipBytes = %u\n", __func__, TotImmediateFlipBytes);
-	dml2_printf("DML::%s: use_lb_flip_bw = %u\n", __func__, use_lb_flip_bw);
-	dml2_printf("DML::%s: iflip_enable = %u\n", __func__, iflip_enable);
-	dml2_printf("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, HostVMInefficiencyFactor);
-	dml2_printf("DML::%s: LineTime = %f\n", __func__, LineTime);
-	dml2_printf("DML::%s: Tno_bw_flip = %f\n", __func__, Tno_bw_flip);
-	dml2_printf("DML::%s: Tvm_trips_flip = %f\n", __func__, Tvm_trips_flip);
-	dml2_printf("DML::%s: Tr0_trips_flip = %f\n", __func__, Tr0_trips_flip);
-	dml2_printf("DML::%s: Tvm_trips_flip_rounded = %f\n", __func__, Tvm_trips_flip_rounded);
-	dml2_printf("DML::%s: Tr0_trips_flip_rounded = %f\n", __func__, Tr0_trips_flip_rounded);
-	dml2_printf("DML::%s: vm_bytes = %f\n", __func__, vm_bytes);
-	dml2_printf("DML::%s: DPTEBytesPerRow = %f\n", __func__, DPTEBytesPerRow);
-	dml2_printf("DML::%s: meta_row_bytes = %d\n", __func__, meta_row_bytes);
-	dml2_printf("DML::%s: dpte_row_bytes = %f\n", __func__, l->dpte_row_bytes);
-	dml2_printf("DML::%s: dpte_row_height = %d\n", __func__, dpte_row_height);
-	dml2_printf("DML::%s: meta_row_height = %d\n", __func__, meta_row_height);
-	dml2_printf("DML::%s: VRatio = %f\n", __func__, VRatio);
+	DML_LOG_VERBOSE("DML::%s: GPUVMEnable = %u\n", __func__, GPUVMEnable);
+	DML_LOG_VERBOSE("DML::%s: ip.max_flip_time_us = %d\n", __func__, max_flip_time_us);
+	DML_LOG_VERBOSE("DML::%s: ip.max_flip_time_lines = %d\n", __func__, max_flip_time_lines);
+	DML_LOG_VERBOSE("DML::%s: BandwidthAvailableForImmediateFlip = %f\n", __func__, BandwidthAvailableForImmediateFlip);
+	DML_LOG_VERBOSE("DML::%s: TotImmediateFlipBytes = %u\n", __func__, TotImmediateFlipBytes);
+	DML_LOG_VERBOSE("DML::%s: use_lb_flip_bw = %u\n", __func__, use_lb_flip_bw);
+	DML_LOG_VERBOSE("DML::%s: iflip_enable = %u\n", __func__, iflip_enable);
+	DML_LOG_VERBOSE("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, HostVMInefficiencyFactor);
+	DML_LOG_VERBOSE("DML::%s: LineTime = %f\n", __func__, LineTime);
+	DML_LOG_VERBOSE("DML::%s: Tno_bw_flip = %f\n", __func__, Tno_bw_flip);
+	DML_LOG_VERBOSE("DML::%s: Tvm_trips_flip = %f\n", __func__, Tvm_trips_flip);
+	DML_LOG_VERBOSE("DML::%s: Tr0_trips_flip = %f\n", __func__, Tr0_trips_flip);
+	DML_LOG_VERBOSE("DML::%s: Tvm_trips_flip_rounded = %f\n", __func__, Tvm_trips_flip_rounded);
+	DML_LOG_VERBOSE("DML::%s: Tr0_trips_flip_rounded = %f\n", __func__, Tr0_trips_flip_rounded);
+	DML_LOG_VERBOSE("DML::%s: vm_bytes = %f\n", __func__, vm_bytes);
+	DML_LOG_VERBOSE("DML::%s: DPTEBytesPerRow = %f\n", __func__, DPTEBytesPerRow);
+	DML_LOG_VERBOSE("DML::%s: meta_row_bytes = %d\n", __func__, meta_row_bytes);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_bytes = %f\n", __func__, l->dpte_row_bytes);
+	DML_LOG_VERBOSE("DML::%s: dpte_row_height = %d\n", __func__, dpte_row_height);
+	DML_LOG_VERBOSE("DML::%s: meta_row_height = %d\n", __func__, meta_row_height);
+	DML_LOG_VERBOSE("DML::%s: VRatio = %f\n", __func__, VRatio);
 #endif
 
 	if (TotImmediateFlipBytes > 0 && (GPUVMEnable || dcc_mrq_enable)) {
@@ -6596,9 +6528,9 @@ static void CalculateFlipSchedule(
 			l->min_row_time = l->min_row_height * LineTime / VRatio;
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: min_row_time = %f\n", __func__, l->min_row_time);
+		DML_LOG_VERBOSE("DML::%s: min_row_time = %f\n", __func__, l->min_row_time);
 #endif
-		DML2_ASSERT(l->min_row_time > 0);
+		DML_ASSERT(l->min_row_time > 0);
 
 		if (use_lb_flip_bw) {
 			// For mode check, calculation the flip bw requirement with worst case flip time
@@ -6619,20 +6551,20 @@ static void CalculateFlipSchedule(
 					l->hvm_scaled_vm_bytes / (l->max_flip_time - Tno_bw_flip - 2 * Tr0_trips_flip_rounded),
 					l->hvm_scaled_row_bytes / (l->max_flip_time - Tvm_trips_flip_rounded));
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: max_flip_time = %f\n", __func__, l->max_flip_time);
-				dml2_printf("DML::%s: total vm bytes (hvm ineff scaled) = %f\n", __func__, l->hvm_scaled_vm_bytes);
-				dml2_printf("DML::%s: total row bytes (%d row, hvm ineff scaled) = %f\n", __func__, l->num_rows, l->hvm_scaled_row_bytes);
-				dml2_printf("DML::%s: total vm+row bytes (hvm ineff scaled) = %f\n", __func__, l->hvm_scaled_vm_row_bytes);
-				dml2_printf("DML::%s: lb_flip_bw for vm and row = %f\n", __func__, l->hvm_scaled_vm_row_bytes / (l->max_flip_time - Tno_bw_flip));
-				dml2_printf("DML::%s: lb_flip_bw for vm = %f\n", __func__, l->hvm_scaled_vm_bytes / (l->max_flip_time - Tno_bw_flip - 2 * Tr0_trips_flip_rounded));
-				dml2_printf("DML::%s: lb_flip_bw for row = %f\n", __func__, l->hvm_scaled_row_bytes / (l->max_flip_time - Tvm_trips_flip_rounded));
+				DML_LOG_VERBOSE("DML::%s: max_flip_time = %f\n", __func__, l->max_flip_time);
+				DML_LOG_VERBOSE("DML::%s: total vm bytes (hvm ineff scaled) = %f\n", __func__, l->hvm_scaled_vm_bytes);
+				DML_LOG_VERBOSE("DML::%s: total row bytes (%f row, hvm ineff scaled) = %f\n", __func__, l->num_rows, l->hvm_scaled_row_bytes);
+				DML_LOG_VERBOSE("DML::%s: total vm+row bytes (hvm ineff scaled) = %f\n", __func__, l->hvm_scaled_vm_row_bytes);
+				DML_LOG_VERBOSE("DML::%s: lb_flip_bw for vm and row = %f\n", __func__, l->hvm_scaled_vm_row_bytes / (l->max_flip_time - Tno_bw_flip));
+				DML_LOG_VERBOSE("DML::%s: lb_flip_bw for vm = %f\n", __func__, l->hvm_scaled_vm_bytes / (l->max_flip_time - Tno_bw_flip - 2 * Tr0_trips_flip_rounded));
+				DML_LOG_VERBOSE("DML::%s: lb_flip_bw for row = %f\n", __func__, l->hvm_scaled_row_bytes / (l->max_flip_time - Tvm_trips_flip_rounded));
 
 				if (l->lb_flip_bw > 0) {
-					dml2_printf("DML::%s: mode_support est Tvm_flip = %f (bw-based)\n", __func__, Tno_bw_flip + l->hvm_scaled_vm_bytes / l->lb_flip_bw);
-					dml2_printf("DML::%s: mode_support est Tr0_flip = %f (bw-based)\n", __func__, l->hvm_scaled_row_bytes / l->lb_flip_bw / l->num_rows);
-					dml2_printf("DML::%s: mode_support est dst_y_per_vm_flip = %f (bw-based)\n", __func__, Tno_bw_flip + l->hvm_scaled_vm_bytes / l->lb_flip_bw / LineTime);
-					dml2_printf("DML::%s: mode_support est dst_y_per_row_flip = %f (bw-based)\n", __func__, l->hvm_scaled_row_bytes / l->lb_flip_bw / LineTime / l->num_rows);
-					dml2_printf("DML::%s: Tvm_trips_flip_rounded + 2*Tr0_trips_flip_rounded = %f\n", __func__, (Tvm_trips_flip_rounded + 2 * Tr0_trips_flip_rounded));
+					DML_LOG_VERBOSE("DML::%s: mode_support est Tvm_flip = %f (bw-based)\n", __func__, Tno_bw_flip + l->hvm_scaled_vm_bytes / l->lb_flip_bw);
+					DML_LOG_VERBOSE("DML::%s: mode_support est Tr0_flip = %f (bw-based)\n", __func__, l->hvm_scaled_row_bytes / l->lb_flip_bw / l->num_rows);
+					DML_LOG_VERBOSE("DML::%s: mode_support est dst_y_per_vm_flip = %f (bw-based)\n", __func__, Tno_bw_flip + l->hvm_scaled_vm_bytes / l->lb_flip_bw / LineTime);
+					DML_LOG_VERBOSE("DML::%s: mode_support est dst_y_per_row_flip = %f (bw-based)\n", __func__, l->hvm_scaled_row_bytes / l->lb_flip_bw / LineTime / l->num_rows);
+					DML_LOG_VERBOSE("DML::%s: Tvm_trips_flip_rounded + 2*Tr0_trips_flip_rounded = %f\n", __func__, (Tvm_trips_flip_rounded + 2 * Tr0_trips_flip_rounded));
 				}
 #endif
 				l->lb_flip_bw = math_max3(l->lb_flip_bw,
@@ -6640,8 +6572,8 @@ static void CalculateFlipSchedule(
 						(l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes) / (15 * LineTime));
 
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: lb_flip_bw for vm reg limit = %f\n", __func__, l->hvm_scaled_vm_bytes / (31 * LineTime) - Tno_bw_flip);
-				dml2_printf("DML::%s: lb_flip_bw for row reg limit = %f\n", __func__, (l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes) / (15 * LineTime));
+				DML_LOG_VERBOSE("DML::%s: lb_flip_bw for vm reg limit = %f\n", __func__, l->hvm_scaled_vm_bytes / (31 * LineTime) - Tno_bw_flip);
+				DML_LOG_VERBOSE("DML::%s: lb_flip_bw for row reg limit = %f\n", __func__, (l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes) / (15 * LineTime));
 #endif
 			}
 
@@ -6653,13 +6585,12 @@ static void CalculateFlipSchedule(
 		} else {
 			if (iflip_enable) {
 				l->ImmediateFlipBW = (double)per_pipe_flip_bytes * BandwidthAvailableForImmediateFlip / (double)TotImmediateFlipBytes; // flip_bw(i)
-				double portion = (double)per_pipe_flip_bytes / (double)TotImmediateFlipBytes;
 
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: per_pipe_flip_bytes = %d\n", __func__, per_pipe_flip_bytes);
-				dml2_printf("DML::%s: BandwidthAvailableForImmediateFlip = %f\n", __func__, BandwidthAvailableForImmediateFlip);
-				dml2_printf("DML::%s: ImmediateFlipBW = %f\n", __func__, l->ImmediateFlipBW);
-				dml2_printf("DML::%s: portion of flip bw = %f\n", __func__, portion);
+				DML_LOG_VERBOSE("DML::%s: per_pipe_flip_bytes = %d\n", __func__, per_pipe_flip_bytes);
+				DML_LOG_VERBOSE("DML::%s: BandwidthAvailableForImmediateFlip = %f\n", __func__, BandwidthAvailableForImmediateFlip);
+				DML_LOG_VERBOSE("DML::%s: ImmediateFlipBW = %f\n", __func__, l->ImmediateFlipBW);
+				DML_LOG_VERBOSE("DML::%s: portion of flip bw = %f\n", __func__, (double)per_pipe_flip_bytes / (double)TotImmediateFlipBytes);
 #endif
 				if (l->ImmediateFlipBW == 0) {
 					l->Tvm_flip = 0;
@@ -6674,11 +6605,11 @@ static void CalculateFlipSchedule(
 						LineTime / 4.0);
 				}
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: total vm bytes (hvm ineff scaled) = %f\n", __func__, vm_bytes * HostVMInefficiencyFactor);
-				dml2_printf("DML::%s: total row bytes (hvm ineff scaled, one row) = %f\n", __func__, (l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes));
+				DML_LOG_VERBOSE("DML::%s: total vm bytes (hvm ineff scaled) = %f\n", __func__, vm_bytes * HostVMInefficiencyFactor);
+				DML_LOG_VERBOSE("DML::%s: total row bytes (hvm ineff scaled, one row) = %f\n", __func__, (l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes));
 
-				dml2_printf("DML::%s: Tvm_flip = %f (bw-based), Tvm_trips_flip = %f (latency-based)\n", __func__, Tno_bw_flip + vm_bytes * HostVMInefficiencyFactor / l->ImmediateFlipBW, Tvm_trips_flip);
-				dml2_printf("DML::%s: Tr0_flip = %f (bw-based), Tr0_trips_flip = %f (latency-based)\n", __func__, (l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes) / l->ImmediateFlipBW, Tr0_trips_flip);
+				DML_LOG_VERBOSE("DML::%s: Tvm_flip = %f (bw-based), Tvm_trips_flip = %f (latency-based)\n", __func__, Tno_bw_flip + vm_bytes * HostVMInefficiencyFactor / l->ImmediateFlipBW, Tvm_trips_flip);
+				DML_LOG_VERBOSE("DML::%s: Tr0_flip = %f (bw-based), Tr0_trips_flip = %f (latency-based)\n", __func__, (l->dpte_row_bytes * HostVMInefficiencyFactor + meta_row_bytes) / l->ImmediateFlipBW, Tr0_trips_flip);
 #endif
 				*dst_y_per_vm_flip = math_ceil2(4.0 * (l->Tvm_flip / LineTime), 1.0) / 4.0;
 				*dst_y_per_row_flip = math_ceil2(4.0 * (l->Tr0_flip / LineTime), 1.0) / 4.0;
@@ -6711,14 +6642,14 @@ static void CalculateFlipSchedule(
 
 #ifdef __DML_VBA_DEBUG__
 	if (!use_lb_flip_bw) {
-		dml2_printf("DML::%s: dst_y_per_vm_flip = %f (should be < 32)\n", __func__, *dst_y_per_vm_flip);
-		dml2_printf("DML::%s: dst_y_per_row_flip = %f (should be < 16)\n", __func__, *dst_y_per_row_flip);
-		dml2_printf("DML::%s: Tvm_flip = %f (final)\n", __func__, l->Tvm_flip);
-		dml2_printf("DML::%s: Tr0_flip = %f (final)\n", __func__, l->Tr0_flip);
-		dml2_printf("DML::%s: Tvm_flip + 2*Tr0_flip = %f (should be <= min_row_time=%f)\n", __func__, l->Tvm_flip + 2 * l->Tr0_flip, l->min_row_time);
+		DML_LOG_VERBOSE("DML::%s: dst_y_per_vm_flip = %f (should be < 32)\n", __func__, *dst_y_per_vm_flip);
+		DML_LOG_VERBOSE("DML::%s: dst_y_per_row_flip = %f (should be < 16)\n", __func__, *dst_y_per_row_flip);
+		DML_LOG_VERBOSE("DML::%s: Tvm_flip = %f (final)\n", __func__, l->Tvm_flip);
+		DML_LOG_VERBOSE("DML::%s: Tr0_flip = %f (final)\n", __func__, l->Tr0_flip);
+		DML_LOG_VERBOSE("DML::%s: Tvm_flip + 2*Tr0_flip = %f (should be <= min_row_time=%f)\n", __func__, l->Tvm_flip + 2 * l->Tr0_flip, l->min_row_time);
 	}
-	dml2_printf("DML::%s: final_flip_bw = %f\n", __func__, *final_flip_bw);
-	dml2_printf("DML::%s: ImmediateFlipSupportedForPipe = %u\n", __func__, *ImmediateFlipSupportedForPipe);
+	DML_LOG_VERBOSE("DML::%s: final_flip_bw = %f\n", __func__, *final_flip_bw);
+	DML_LOG_VERBOSE("DML::%s: ImmediateFlipSupportedForPipe = %u\n", __func__, *ImmediateFlipSupportedForPipe);
 #endif
 }
 
@@ -6736,7 +6667,7 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 	p->Watermark->UrgentWatermark = p->mmSOCParameters.UrgentLatency + p->mmSOCParameters.ExtraLatency;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: UrgentWatermark = %f\n", __func__, p->Watermark->UrgentWatermark);
+	DML_LOG_VERBOSE("DML::%s: UrgentWatermark = %f\n", __func__, p->Watermark->UrgentWatermark);
 #endif
 
 	p->Watermark->USRRetrainingWatermark = p->mmSOCParameters.UrgentLatency + p->mmSOCParameters.ExtraLatency + p->mmSOCParameters.USRRetrainingLatency + p->mmSOCParameters.SMNLatency;
@@ -6755,20 +6686,20 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 	p->Watermark->temp_read_or_ppt_watermark_us = p->mmSOCParameters.g6_temp_read_blackout_us + p->Watermark->UrgentWatermark;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: UrgentLatency = %f\n", __func__, p->mmSOCParameters.UrgentLatency);
-	dml2_printf("DML::%s: ExtraLatency = %f\n", __func__, p->mmSOCParameters.ExtraLatency);
-	dml2_printf("DML::%s: DRAMClockChangeLatency = %f\n", __func__, p->mmSOCParameters.DRAMClockChangeLatency);
-	dml2_printf("DML::%s: SREnterPlusExitZ8Time = %f\n", __func__, p->mmSOCParameters.SREnterPlusExitZ8Time);
-	dml2_printf("DML::%s: SREnterPlusExitTime = %f\n", __func__, p->mmSOCParameters.SREnterPlusExitTime);
-	dml2_printf("DML::%s: UrgentWatermark = %f\n", __func__, p->Watermark->UrgentWatermark);
-	dml2_printf("DML::%s: USRRetrainingWatermark = %f\n", __func__, p->Watermark->USRRetrainingWatermark);
-	dml2_printf("DML::%s: DRAMClockChangeWatermark = %f\n", __func__, p->Watermark->DRAMClockChangeWatermark);
-	dml2_printf("DML::%s: FCLKChangeWatermark = %f\n", __func__, p->Watermark->FCLKChangeWatermark);
-	dml2_printf("DML::%s: StutterExitWatermark = %f\n", __func__, p->Watermark->StutterExitWatermark);
-	dml2_printf("DML::%s: StutterEnterPlusExitWatermark = %f\n", __func__, p->Watermark->StutterEnterPlusExitWatermark);
-	dml2_printf("DML::%s: Z8StutterExitWatermark = %f\n", __func__, p->Watermark->Z8StutterExitWatermark);
-	dml2_printf("DML::%s: Z8StutterEnterPlusExitWatermark = %f\n", __func__, p->Watermark->Z8StutterEnterPlusExitWatermark);
-	dml2_printf("DML::%s: temp_read_or_ppt_watermark_us = %f\n", __func__, p->Watermark->temp_read_or_ppt_watermark_us);
+	DML_LOG_VERBOSE("DML::%s: UrgentLatency = %f\n", __func__, p->mmSOCParameters.UrgentLatency);
+	DML_LOG_VERBOSE("DML::%s: ExtraLatency = %f\n", __func__, p->mmSOCParameters.ExtraLatency);
+	DML_LOG_VERBOSE("DML::%s: DRAMClockChangeLatency = %f\n", __func__, p->mmSOCParameters.DRAMClockChangeLatency);
+	DML_LOG_VERBOSE("DML::%s: SREnterPlusExitZ8Time = %f\n", __func__, p->mmSOCParameters.SREnterPlusExitZ8Time);
+	DML_LOG_VERBOSE("DML::%s: SREnterPlusExitTime = %f\n", __func__, p->mmSOCParameters.SREnterPlusExitTime);
+	DML_LOG_VERBOSE("DML::%s: UrgentWatermark = %f\n", __func__, p->Watermark->UrgentWatermark);
+	DML_LOG_VERBOSE("DML::%s: USRRetrainingWatermark = %f\n", __func__, p->Watermark->USRRetrainingWatermark);
+	DML_LOG_VERBOSE("DML::%s: DRAMClockChangeWatermark = %f\n", __func__, p->Watermark->DRAMClockChangeWatermark);
+	DML_LOG_VERBOSE("DML::%s: FCLKChangeWatermark = %f\n", __func__, p->Watermark->FCLKChangeWatermark);
+	DML_LOG_VERBOSE("DML::%s: StutterExitWatermark = %f\n", __func__, p->Watermark->StutterExitWatermark);
+	DML_LOG_VERBOSE("DML::%s: StutterEnterPlusExitWatermark = %f\n", __func__, p->Watermark->StutterEnterPlusExitWatermark);
+	DML_LOG_VERBOSE("DML::%s: Z8StutterExitWatermark = %f\n", __func__, p->Watermark->Z8StutterExitWatermark);
+	DML_LOG_VERBOSE("DML::%s: Z8StutterEnterPlusExitWatermark = %f\n", __func__, p->Watermark->Z8StutterEnterPlusExitWatermark);
+	DML_LOG_VERBOSE("DML::%s: temp_read_or_ppt_watermark_us = %f\n", __func__, p->Watermark->temp_read_or_ppt_watermark_us);
 #endif
 
 	s->TotalActiveWriteback = 0;
@@ -6801,11 +6732,11 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 		p->Watermark->WritebackFCLKChangeWatermark = p->Watermark->WritebackFCLKChangeWatermark + p->mmSOCParameters.USRRetrainingLatency;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: WritebackDRAMClockChangeWatermark = %f\n", __func__, p->Watermark->WritebackDRAMClockChangeWatermark);
-	dml2_printf("DML::%s: WritebackFCLKChangeWatermark = %f\n", __func__, p->Watermark->WritebackFCLKChangeWatermark);
-	dml2_printf("DML::%s: WritebackUrgentWatermark = %f\n", __func__, p->Watermark->WritebackUrgentWatermark);
-	dml2_printf("DML::%s: USRRetrainingRequired = %u\n", __func__, p->USRRetrainingRequired);
-	dml2_printf("DML::%s: USRRetrainingLatency = %f\n", __func__, p->mmSOCParameters.USRRetrainingLatency);
+	DML_LOG_VERBOSE("DML::%s: WritebackDRAMClockChangeWatermark = %f\n", __func__, p->Watermark->WritebackDRAMClockChangeWatermark);
+	DML_LOG_VERBOSE("DML::%s: WritebackFCLKChangeWatermark = %f\n", __func__, p->Watermark->WritebackFCLKChangeWatermark);
+	DML_LOG_VERBOSE("DML::%s: WritebackUrgentWatermark = %f\n", __func__, p->Watermark->WritebackUrgentWatermark);
+	DML_LOG_VERBOSE("DML::%s: USRRetrainingRequired = %u\n", __func__, p->USRRetrainingRequired);
+	DML_LOG_VERBOSE("DML::%s: USRRetrainingLatency = %f\n", __func__, p->mmSOCParameters.USRRetrainingLatency);
 #endif
 
 	s->TotalPixelBW = 0.0;
@@ -6836,11 +6767,11 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 		s->LBLatencyHidingSourceLinesC[k] = (unsigned int)(math_min2((double)p->MaxLineBufferLines, math_floor2((double)p->LineBufferSize / LBBitPerPixel / ((double)p->SwathWidthC[k] / math_max2(h_ratio_c, 1.0)), 1)) - (v_taps_c - 1));
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, MaxLineBufferLines = %u\n", __func__, k, p->MaxLineBufferLines);
-		dml2_printf("DML::%s: k=%u, LineBufferSize = %u\n", __func__, k, p->LineBufferSize);
-		dml2_printf("DML::%s: k=%u, LBBitPerPixel = %u\n", __func__, k, LBBitPerPixel);
-		dml2_printf("DML::%s: k=%u, HRatio = %f\n", __func__, k, h_ratio);
-		dml2_printf("DML::%s: k=%u, VTaps = %f\n", __func__, k, v_taps);
+		DML_LOG_VERBOSE("DML::%s: k=%u, MaxLineBufferLines = %u\n", __func__, k, p->MaxLineBufferLines);
+		DML_LOG_VERBOSE("DML::%s: k=%u, LineBufferSize = %u\n", __func__, k, p->LineBufferSize);
+		DML_LOG_VERBOSE("DML::%s: k=%u, LBBitPerPixel = %f\n", __func__, k, LBBitPerPixel);
+		DML_LOG_VERBOSE("DML::%s: k=%u, HRatio = %f\n", __func__, k, h_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u, VTaps = %f\n", __func__, k, v_taps);
 #endif
 
 		s->EffectiveLBLatencyHidingY = s->LBLatencyHidingSourceLinesY[k] / v_ratio * (h_total / pixel_clock_mhz);
@@ -6943,16 +6874,16 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 		s->sub_vp_lines_l = s->src_y_pstate_l + s->src_y_ahead_l + p->meta_row_height_l[k];
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, DETBufferSizeY = %u\n", __func__, k, p->DETBufferSizeY[k]);
-		dml2_printf("DML::%s: k=%u, BytePerPixelDETY = %f\n", __func__, k, p->BytePerPixelDETY[k]);
-		dml2_printf("DML::%s: k=%u, SwathWidthY = %u\n", __func__, k, p->SwathWidthY[k]);
-		dml2_printf("DML::%s: k=%u, SwathHeightY = %u\n", __func__, k, p->SwathHeightY[k]);
-		dml2_printf("DML::%s: k=%u, LBLatencyHidingSourceLinesY = %u\n", __func__, k, s->LBLatencyHidingSourceLinesY[k]);
-		dml2_printf("DML::%s: k=%u, dst_y_pstate = %u\n", __func__, k, s->dst_y_pstate);
-		dml2_printf("DML::%s: k=%u, src_y_pstate_l = %u\n", __func__, k, s->src_y_pstate_l);
-		dml2_printf("DML::%s: k=%u, src_y_ahead_l = %u\n", __func__, k, s->src_y_ahead_l);
-		dml2_printf("DML::%s: k=%u, meta_row_height_l = %u\n", __func__, k, p->meta_row_height_l[k]);
-		dml2_printf("DML::%s: k=%u, sub_vp_lines_l = %u\n", __func__, k, s->sub_vp_lines_l);
+		DML_LOG_VERBOSE("DML::%s: k=%u, DETBufferSizeY = %u\n", __func__, k, p->DETBufferSizeY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, BytePerPixelDETY = %f\n", __func__, k, p->BytePerPixelDETY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, SwathWidthY = %u\n", __func__, k, p->SwathWidthY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, SwathHeightY = %u\n", __func__, k, p->SwathHeightY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, LBLatencyHidingSourceLinesY = %u\n", __func__, k, s->LBLatencyHidingSourceLinesY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dst_y_pstate = %u\n", __func__, k, s->dst_y_pstate);
+		DML_LOG_VERBOSE("DML::%s: k=%u, src_y_pstate_l = %u\n", __func__, k, s->src_y_pstate_l);
+		DML_LOG_VERBOSE("DML::%s: k=%u, src_y_ahead_l = %u\n", __func__, k, s->src_y_ahead_l);
+		DML_LOG_VERBOSE("DML::%s: k=%u, meta_row_height_l = %u\n", __func__, k, p->meta_row_height_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, sub_vp_lines_l = %u\n", __func__, k, s->sub_vp_lines_l);
 #endif
 		p->SubViewportLinesNeededInMALL[k] = s->sub_vp_lines_l;
 
@@ -6967,10 +6898,10 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 				p->SubViewportLinesNeededInMALL[k] = (unsigned int)(math_max2(s->sub_vp_lines_l, s->sub_vp_lines_c));
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, meta_row_height_c = %u\n", __func__, k, p->meta_row_height_c[k]);
-			dml2_printf("DML::%s: k=%u, src_y_pstate_c = %u\n", __func__, k, s->src_y_pstate_c);
-			dml2_printf("DML::%s: k=%u, src_y_ahead_c = %u\n", __func__, k, s->src_y_ahead_c);
-			dml2_printf("DML::%s: k=%u, sub_vp_lines_c = %u\n", __func__, k, s->sub_vp_lines_c);
+			DML_LOG_VERBOSE("DML::%s: k=%u, meta_row_height_c = %u\n", __func__, k, p->meta_row_height_c[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, src_y_pstate_c = %u\n", __func__, k, s->src_y_pstate_c);
+			DML_LOG_VERBOSE("DML::%s: k=%u, src_y_ahead_c = %u\n", __func__, k, s->src_y_ahead_c);
+			DML_LOG_VERBOSE("DML::%s: k=%u, sub_vp_lines_c = %u\n", __func__, k, s->sub_vp_lines_c);
 #endif
 		}
 	}
@@ -6992,10 +6923,10 @@ static void CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DRAMClockChangeSupport = %u\n", __func__, *p->global_dram_clock_change_supported);
-	dml2_printf("DML::%s: FCLKChangeSupport = %u\n", __func__, *p->global_fclk_change_supported);
-	dml2_printf("DML::%s: MaxActiveFCLKChangeLatencySupported = %f\n", __func__, *p->MaxActiveFCLKChangeLatencySupported);
-	dml2_printf("DML::%s: USRRetrainingSupport = %u\n", __func__, *p->USRRetrainingSupport);
+	DML_LOG_VERBOSE("DML::%s: DRAMClockChangeSupport = %u\n", __func__, *p->global_dram_clock_change_supported);
+	DML_LOG_VERBOSE("DML::%s: FCLKChangeSupport = %u\n", __func__, *p->global_fclk_change_supported);
+	DML_LOG_VERBOSE("DML::%s: MaxActiveFCLKChangeLatencySupported = %f\n", __func__, *p->MaxActiveFCLKChangeLatencySupported);
+	DML_LOG_VERBOSE("DML::%s: USRRetrainingSupport = %u\n", __func__, *p->USRRetrainingSupport);
 #endif
 }
 
@@ -7141,7 +7072,7 @@ static unsigned int get_qos_param_index(unsigned long uclk_freq_khz, const struc
 	unsigned int index = 0;
 
 	for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) {
-		dml2_printf("DML::%s: per_uclk_dpm_params[%d].minimum_uclk_khz = %d\n", __func__, i, per_uclk_dpm_params[i].minimum_uclk_khz);
+		DML_LOG_VERBOSE("DML::%s: per_uclk_dpm_params[%d].minimum_uclk_khz = %ld\n", __func__, i, per_uclk_dpm_params[i].minimum_uclk_khz);
 
 		if (i == 0)
 			index = 0;
@@ -7153,32 +7084,30 @@ static unsigned int get_qos_param_index(unsigned long uclk_freq_khz, const struc
 			break;
 		}
 	}
-#if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: uclk_freq_khz = %d\n", __func__, uclk_freq_khz);
-	dml2_printf("DML::%s: index = %d\n", __func__, index);
-#endif
+	DML_LOG_VERBOSE("DML::%s: uclk_freq_khz = %ld\n", __func__, uclk_freq_khz);
+	DML_LOG_VERBOSE("DML::%s: index = %d\n", __func__, index);
 	return index;
 }
 
 static unsigned int get_active_min_uclk_dpm_index(unsigned long uclk_freq_khz, const struct dml2_soc_state_table *clk_table)
 {
 	unsigned int i;
-	bool clk_entry_found = 0;
+	bool clk_entry_found = false;
 
 	for (i = 0; i < clk_table->uclk.num_clk_values; i++) {
-		dml2_printf("DML::%s: clk_table.uclk.clk_values_khz[%d] = %d\n", __func__, i, clk_table->uclk.clk_values_khz[i]);
+		DML_LOG_VERBOSE("DML::%s: clk_table.uclk.clk_values_khz[%d] = %ld\n", __func__, i, clk_table->uclk.clk_values_khz[i]);
 
 		if (uclk_freq_khz == clk_table->uclk.clk_values_khz[i]) {
-			clk_entry_found = 1;
+			clk_entry_found = true;
 			break;
 		}
 	}
 
 	if (!clk_entry_found)
-		DML2_ASSERT(clk_entry_found);
+		DML_ASSERT(clk_entry_found);
 #if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: uclk_freq_khz = %ld\n", __func__, uclk_freq_khz);
-	dml2_printf("DML::%s: index = %d\n", __func__, i);
+	DML_LOG_VERBOSE("DML::%s: uclk_freq_khz = %ld\n", __func__, uclk_freq_khz);
+	DML_LOG_VERBOSE("DML::%s: index = %d\n", __func__, i);
 #endif
 	return i;
 }
@@ -7218,10 +7147,10 @@ static void calculate_hostvm_inefficiency_factor(
 			if ((*HostVMInefficiencyFactorPrefetch < 4) && (remote_iommu_outstanding_translations < max_outstanding_reqs))
 				*HostVMInefficiencyFactorPrefetch = 4;
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: urg_bandwidth_avail_active_pixel_and_vm = %f\n", __func__, urg_bandwidth_avail_active_pixel_and_vm);
-			dml2_printf("DML::%s: urg_bandwidth_avail_active_vm_only = %f\n", __func__, urg_bandwidth_avail_active_vm_only);
-			dml2_printf("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, *HostVMInefficiencyFactor);
-			dml2_printf("DML::%s: HostVMInefficiencyFactorPrefetch = %f\n", __func__, *HostVMInefficiencyFactorPrefetch);
+			DML_LOG_VERBOSE("DML::%s: urg_bandwidth_avail_active_pixel_and_vm = %f\n", __func__, urg_bandwidth_avail_active_pixel_and_vm);
+			DML_LOG_VERBOSE("DML::%s: urg_bandwidth_avail_active_vm_only = %f\n", __func__, urg_bandwidth_avail_active_vm_only);
+			DML_LOG_VERBOSE("DML::%s: HostVMInefficiencyFactor = %f\n", __func__, *HostVMInefficiencyFactor);
+			DML_LOG_VERBOSE("DML::%s: HostVMInefficiencyFactorPrefetch = %f\n", __func__, *HostVMInefficiencyFactorPrefetch);
 #endif
 		}
 }
@@ -7335,30 +7264,659 @@ static void calculate_pstate_keepout_dst_lines(
 	}
 }
 
+static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_internal_display_mode_lib *mode_lib,
+	const struct dml2_display_cfg *display_cfg)
+{
+	struct dml2_core_calcs_mode_support_locals *s = &mode_lib->scratch.dml_core_mode_support_locals;
+	struct dml2_core_calcs_calculate_tdlut_setting_params *calculate_tdlut_setting_params = &mode_lib->scratch.calculate_tdlut_setting_params;
+	struct dml2_core_calcs_CalculatePrefetchSchedule_params *CalculatePrefetchSchedule_params = &mode_lib->scratch.CalculatePrefetchSchedule_params;
+	struct dml2_core_calcs_calculate_peak_bandwidth_required_params *calculate_peak_bandwidth_params = &mode_lib->scratch.calculate_peak_bandwidth_params;
+#ifdef DML_GLOBAL_PREFETCH_CHECK
+	struct dml2_core_calcs_CheckGlobalPrefetchAdmissibility_params *CheckGlobalPrefetchAdmissibility_params = &mode_lib->scratch.CheckGlobalPrefetchAdmissibility_params;
+#endif
+	struct dml2_core_calcs_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport_params *CalculateWatermarks_params = &mode_lib->scratch.CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport_params;
+
+	double min_return_bw_for_latency;
+	unsigned int k;
+
+	mode_lib->ms.TimeCalc = 24 / mode_lib->ms.dcfclk_deepsleep;
+
+	calculate_hostvm_inefficiency_factor(
+			&s->HostVMInefficiencyFactor,
+			&s->HostVMInefficiencyFactorPrefetch,
+
+			display_cfg->gpuvm_enable,
+			display_cfg->hostvm_enable,
+			mode_lib->ip.remote_iommu_outstanding_translations,
+			mode_lib->soc.max_outstanding_reqs,
+			mode_lib->ms.support.urg_bandwidth_available_pixel_and_vm[dml2_core_internal_soc_state_sys_active],
+			mode_lib->ms.support.urg_bandwidth_available_vm_only[dml2_core_internal_soc_state_sys_active]);
+
+	mode_lib->ms.Total3dlutActive = 0;
+	for (k = 0; k <= mode_lib->ms.num_active_planes - 1; k++) {
+		if (display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut)
+			mode_lib->ms.Total3dlutActive = mode_lib->ms.Total3dlutActive + 1;
+
+		// Calculate tdlut schedule related terms
+		calculate_tdlut_setting_params->dispclk_mhz = mode_lib->ms.RequiredDISPCLK;
+		calculate_tdlut_setting_params->setup_for_tdlut = display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut;
+		calculate_tdlut_setting_params->tdlut_width_mode = display_cfg->plane_descriptors[k].tdlut.tdlut_width_mode;
+		calculate_tdlut_setting_params->tdlut_addressing_mode = display_cfg->plane_descriptors[k].tdlut.tdlut_addressing_mode;
+		calculate_tdlut_setting_params->cursor_buffer_size = mode_lib->ip.cursor_buffer_size;
+		calculate_tdlut_setting_params->gpuvm_enable = display_cfg->gpuvm_enable;
+		calculate_tdlut_setting_params->gpuvm_page_size_kbytes = display_cfg->plane_descriptors[k].overrides.gpuvm_min_page_size_kbytes;
+		calculate_tdlut_setting_params->tdlut_mpc_width_flag = display_cfg->plane_descriptors[k].tdlut.tdlut_mpc_width_flag;
+		calculate_tdlut_setting_params->is_gfx11 = dml_get_gfx_version(display_cfg->plane_descriptors[k].surface.tiling);
+
+		// output
+		calculate_tdlut_setting_params->tdlut_pte_bytes_per_frame = &s->tdlut_pte_bytes_per_frame[k];
+		calculate_tdlut_setting_params->tdlut_bytes_per_frame = &s->tdlut_bytes_per_frame[k];
+		calculate_tdlut_setting_params->tdlut_groups_per_2row_ub = &s->tdlut_groups_per_2row_ub[k];
+		calculate_tdlut_setting_params->tdlut_opt_time = &s->tdlut_opt_time[k];
+		calculate_tdlut_setting_params->tdlut_drain_time = &s->tdlut_drain_time[k];
+		calculate_tdlut_setting_params->tdlut_bytes_to_deliver = &s->tdlut_bytes_to_deliver[k];
+		calculate_tdlut_setting_params->tdlut_bytes_per_group = &s->tdlut_bytes_per_group[k];
+
+		calculate_tdlut_setting(&mode_lib->scratch, calculate_tdlut_setting_params);
+	}
+
+	min_return_bw_for_latency = mode_lib->ms.support.urg_bandwidth_available_min_latency[dml2_core_internal_soc_state_sys_active];
+
+	if (mode_lib->soc.qos_parameters.qos_type == dml2_qos_param_type_dcn3)
+		s->ReorderingBytes = (unsigned int)(mode_lib->soc.clk_table.dram_config.channel_count * math_max3(mode_lib->soc.qos_parameters.qos_params.dcn32x.urgent_out_of_order_return_per_channel_pixel_only_bytes,
+										mode_lib->soc.qos_parameters.qos_params.dcn32x.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes,
+										mode_lib->soc.qos_parameters.qos_params.dcn32x.urgent_out_of_order_return_per_channel_vm_only_bytes));
+
+	CalculateExtraLatency(
+		display_cfg,
+		mode_lib->ip.rob_buffer_size_kbytes,
+		mode_lib->soc.qos_parameters.qos_params.dcn32x.loaded_round_trip_latency_fclk_cycles,
+		s->ReorderingBytes,
+		mode_lib->ms.DCFCLK,
+		mode_lib->ms.FabricClock,
+		mode_lib->ip.pixel_chunk_size_kbytes,
+		min_return_bw_for_latency,
+		mode_lib->ms.num_active_planes,
+		mode_lib->ms.NoOfDPP,
+		mode_lib->ms.dpte_group_bytes,
+		s->tdlut_bytes_per_group,
+		s->HostVMInefficiencyFactor,
+		s->HostVMInefficiencyFactorPrefetch,
+		mode_lib->soc.hostvm_min_page_size_kbytes,
+		mode_lib->soc.qos_parameters.qos_type,
+		!(display_cfg->overrides.max_outstanding_when_urgent_expected_disable),
+		mode_lib->soc.max_outstanding_reqs,
+		mode_lib->ms.support.request_size_bytes_luma,
+		mode_lib->ms.support.request_size_bytes_chroma,
+		mode_lib->ip.meta_chunk_size_kbytes,
+		mode_lib->ip.dchub_arb_to_ret_delay,
+		mode_lib->ms.TripToMemory,
+		mode_lib->ip.hostvm_mode,
+
+		// output
+		&mode_lib->ms.ExtraLatency,
+		&mode_lib->ms.ExtraLatency_sr,
+		&mode_lib->ms.ExtraLatencyPrefetch);
+
+	for (k = 0; k < mode_lib->ms.num_active_planes; k++)
+		s->impacted_dst_y_pre[k] = 0;
+
+	s->recalc_prefetch_schedule = 0;
+	s->recalc_prefetch_done = 0;
+	do {
+		mode_lib->ms.support.PrefetchSupported = true;
+
+		for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+			s->line_times[k] = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
+			s->pixel_format[k] = display_cfg->plane_descriptors[k].pixel_format;
+
+			s->lb_source_lines_l[k] = get_num_lb_source_lines(mode_lib->ip.max_line_buffer_lines, mode_lib->ip.line_buffer_size_bits,
+																mode_lib->ms.NoOfDPP[k],
+																display_cfg->plane_descriptors[k].composition.viewport.plane0.width,
+																display_cfg->plane_descriptors[k].composition.viewport.plane0.height,
+																display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio,
+																display_cfg->plane_descriptors[k].composition.rotation_angle);
+
+			s->lb_source_lines_c[k] = get_num_lb_source_lines(mode_lib->ip.max_line_buffer_lines, mode_lib->ip.line_buffer_size_bits,
+																mode_lib->ms.NoOfDPP[k],
+																display_cfg->plane_descriptors[k].composition.viewport.plane1.width,
+																display_cfg->plane_descriptors[k].composition.viewport.plane1.height,
+																display_cfg->plane_descriptors[k].composition.scaler_info.plane1.h_ratio,
+																display_cfg->plane_descriptors[k].composition.rotation_angle);
+
+			struct dml2_core_internal_DmlPipe *myPipe = &s->myPipe;
+
+			mode_lib->ms.TWait[k] = CalculateTWait(
+				display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns,
+				mode_lib->ms.UrgLatency,
+				mode_lib->ms.TripToMemory,
+				!dml_is_phantom_pipe(&display_cfg->plane_descriptors[k]) && display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.drr_config.enabled ?
+				get_g6_temp_read_blackout_us(&mode_lib->soc, (unsigned int)(mode_lib->ms.uclk_freq_mhz * 1000), mode_lib->ms.state_idx) : 0.0);
+
+			myPipe->Dppclk = mode_lib->ms.RequiredDPPCLK[k];
+			myPipe->Dispclk = mode_lib->ms.RequiredDISPCLK;
+			myPipe->PixelClock = ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
+			myPipe->DCFClkDeepSleep = mode_lib->ms.dcfclk_deepsleep;
+			myPipe->DPPPerSurface = mode_lib->ms.NoOfDPP[k];
+			myPipe->ScalerEnabled = display_cfg->plane_descriptors[k].composition.scaler_info.enabled;
+			myPipe->VRatio = display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio;
+			myPipe->VRatioChroma = display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio;
+			myPipe->VTaps = display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_taps;
+			myPipe->VTapsChroma = display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_taps;
+			myPipe->RotationAngle = display_cfg->plane_descriptors[k].composition.rotation_angle;
+			myPipe->mirrored = display_cfg->plane_descriptors[k].composition.mirrored;
+			myPipe->BlockWidth256BytesY = mode_lib->ms.Read256BlockWidthY[k];
+			myPipe->BlockHeight256BytesY = mode_lib->ms.Read256BlockHeightY[k];
+			myPipe->BlockWidth256BytesC = mode_lib->ms.Read256BlockWidthC[k];
+			myPipe->BlockHeight256BytesC = mode_lib->ms.Read256BlockHeightC[k];
+			myPipe->InterlaceEnable = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.interlaced;
+			myPipe->NumberOfCursors = display_cfg->plane_descriptors[k].cursor.num_cursors;
+			myPipe->VBlank = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_total - display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_active;
+			myPipe->HTotal = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total;
+			myPipe->HActive = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_active;
+			myPipe->DCCEnable = display_cfg->plane_descriptors[k].surface.dcc.enable;
+			myPipe->ODMMode = mode_lib->ms.ODMMode[k];
+			myPipe->SourcePixelFormat = display_cfg->plane_descriptors[k].pixel_format;
+			myPipe->BytePerPixelY = mode_lib->ms.BytePerPixelY[k];
+			myPipe->BytePerPixelC = mode_lib->ms.BytePerPixelC[k];
+			myPipe->ProgressiveToInterlaceUnitInOPP = mode_lib->ip.ptoi_supported;
+
+#ifdef __DML_VBA_DEBUG__
+			DML_LOG_VERBOSE("DML::%s: Calling CalculatePrefetchSchedule for k=%u\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: MaximumVStartup = %u\n", __func__, s->MaximumVStartup[k]);
+#endif
+			CalculatePrefetchSchedule_params->display_cfg = display_cfg;
+			CalculatePrefetchSchedule_params->HostVMInefficiencyFactor = s->HostVMInefficiencyFactorPrefetch;
+			CalculatePrefetchSchedule_params->myPipe = myPipe;
+			CalculatePrefetchSchedule_params->DSCDelay = mode_lib->ms.DSCDelay[k];
+			CalculatePrefetchSchedule_params->DPPCLKDelaySubtotalPlusCNVCFormater = mode_lib->ip.dppclk_delay_subtotal + mode_lib->ip.dppclk_delay_cnvc_formatter;
+			CalculatePrefetchSchedule_params->DPPCLKDelaySCL = mode_lib->ip.dppclk_delay_scl;
+			CalculatePrefetchSchedule_params->DPPCLKDelaySCLLBOnly = mode_lib->ip.dppclk_delay_scl_lb_only;
+			CalculatePrefetchSchedule_params->DPPCLKDelayCNVCCursor = mode_lib->ip.dppclk_delay_cnvc_cursor;
+			CalculatePrefetchSchedule_params->DISPCLKDelaySubtotal = mode_lib->ip.dispclk_delay_subtotal;
+			CalculatePrefetchSchedule_params->DPP_RECOUT_WIDTH = (unsigned int)(mode_lib->ms.SwathWidthY[k] / display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio);
+			CalculatePrefetchSchedule_params->OutputFormat = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].output.output_format;
+			CalculatePrefetchSchedule_params->MaxInterDCNTileRepeaters = mode_lib->ip.max_inter_dcn_tile_repeaters;
+			CalculatePrefetchSchedule_params->VStartup = s->MaximumVStartup[k];
+			CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->soc.hostvm_min_page_size_kbytes;
+			CalculatePrefetchSchedule_params->DynamicMetadataEnable = display_cfg->plane_descriptors[k].dynamic_meta_data.enable;
+			CalculatePrefetchSchedule_params->DynamicMetadataVMEnabled = mode_lib->ip.dynamic_metadata_vm_enabled;
+			CalculatePrefetchSchedule_params->DynamicMetadataLinesBeforeActiveRequired = display_cfg->plane_descriptors[k].dynamic_meta_data.lines_before_active_required;
+			CalculatePrefetchSchedule_params->DynamicMetadataTransmittedBytes = display_cfg->plane_descriptors[k].dynamic_meta_data.transmitted_bytes;
+			CalculatePrefetchSchedule_params->UrgentLatency = mode_lib->ms.UrgLatency;
+			CalculatePrefetchSchedule_params->ExtraLatencyPrefetch = mode_lib->ms.ExtraLatencyPrefetch;
+			CalculatePrefetchSchedule_params->TCalc = mode_lib->ms.TimeCalc;
+			CalculatePrefetchSchedule_params->vm_bytes = mode_lib->ms.vm_bytes[k];
+			CalculatePrefetchSchedule_params->PixelPTEBytesPerRow = mode_lib->ms.DPTEBytesPerRow[k];
+			CalculatePrefetchSchedule_params->PrefetchSourceLinesY = mode_lib->ms.PrefetchLinesY[k];
+			CalculatePrefetchSchedule_params->VInitPreFillY = mode_lib->ms.PrefillY[k];
+			CalculatePrefetchSchedule_params->MaxNumSwathY = mode_lib->ms.MaxNumSwathY[k];
+			CalculatePrefetchSchedule_params->PrefetchSourceLinesC = mode_lib->ms.PrefetchLinesC[k];
+			CalculatePrefetchSchedule_params->VInitPreFillC = mode_lib->ms.PrefillC[k];
+			CalculatePrefetchSchedule_params->MaxNumSwathC = mode_lib->ms.MaxNumSwathC[k];
+			CalculatePrefetchSchedule_params->swath_width_luma_ub = mode_lib->ms.swath_width_luma_ub[k];
+			CalculatePrefetchSchedule_params->swath_width_chroma_ub = mode_lib->ms.swath_width_chroma_ub[k];
+			CalculatePrefetchSchedule_params->SwathHeightY = mode_lib->ms.SwathHeightY[k];
+			CalculatePrefetchSchedule_params->SwathHeightC = mode_lib->ms.SwathHeightC[k];
+			CalculatePrefetchSchedule_params->TWait = mode_lib->ms.TWait[k];
+			CalculatePrefetchSchedule_params->Ttrip = mode_lib->ms.TripToMemory;
+			CalculatePrefetchSchedule_params->Turg = mode_lib->ms.UrgLatency;
+			CalculatePrefetchSchedule_params->setup_for_tdlut = display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut;
+			CalculatePrefetchSchedule_params->tdlut_pte_bytes_per_frame = s->tdlut_pte_bytes_per_frame[k];
+			CalculatePrefetchSchedule_params->tdlut_bytes_per_frame = s->tdlut_bytes_per_frame[k];
+			CalculatePrefetchSchedule_params->tdlut_opt_time = s->tdlut_opt_time[k];
+			CalculatePrefetchSchedule_params->tdlut_drain_time = s->tdlut_drain_time[k];
+			CalculatePrefetchSchedule_params->num_cursors = (display_cfg->plane_descriptors[k].cursor.cursor_width > 0);
+			CalculatePrefetchSchedule_params->cursor_bytes_per_chunk = s->cursor_bytes_per_chunk[k];
+			CalculatePrefetchSchedule_params->cursor_bytes_per_line = s->cursor_bytes_per_line[k];
+			CalculatePrefetchSchedule_params->dcc_enable = display_cfg->plane_descriptors[k].surface.dcc.enable;
+			CalculatePrefetchSchedule_params->mrq_present = mode_lib->ip.dcn_mrq_present;
+			CalculatePrefetchSchedule_params->meta_row_bytes = mode_lib->ms.meta_row_bytes[k];
+			CalculatePrefetchSchedule_params->mall_prefetch_sdp_overhead_factor = mode_lib->ms.mall_prefetch_sdp_overhead_factor[k];
+			CalculatePrefetchSchedule_params->impacted_dst_y_pre = s->impacted_dst_y_pre[k];
+			CalculatePrefetchSchedule_params->vactive_sw_bw_l = mode_lib->ms.vactive_sw_bw_l[k];
+			CalculatePrefetchSchedule_params->vactive_sw_bw_c = mode_lib->ms.vactive_sw_bw_c[k];
+
+			// output
+			CalculatePrefetchSchedule_params->DSTXAfterScaler = &s->DSTXAfterScaler[k];
+			CalculatePrefetchSchedule_params->DSTYAfterScaler = &s->DSTYAfterScaler[k];
+			CalculatePrefetchSchedule_params->dst_y_prefetch = &mode_lib->ms.dst_y_prefetch[k];
+			CalculatePrefetchSchedule_params->dst_y_per_vm_vblank = &mode_lib->ms.LinesForVM[k];
+			CalculatePrefetchSchedule_params->dst_y_per_row_vblank = &mode_lib->ms.LinesForDPTERow[k];
+			CalculatePrefetchSchedule_params->VRatioPrefetchY = &mode_lib->ms.VRatioPreY[k];
+			CalculatePrefetchSchedule_params->VRatioPrefetchC = &mode_lib->ms.VRatioPreC[k];
+			CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWLuma = &mode_lib->ms.RequiredPrefetchPixelDataBWLuma[k]; // prefetch_sw_bw_l
+			CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWChroma = &mode_lib->ms.RequiredPrefetchPixelDataBWChroma[k]; // prefetch_sw_bw_c
+			CalculatePrefetchSchedule_params->RequiredPrefetchBWOTO = &mode_lib->ms.RequiredPrefetchBWOTO[k];
+			CalculatePrefetchSchedule_params->NotEnoughTimeForDynamicMetadata = &mode_lib->ms.NoTimeForDynamicMetadata[k];
+			CalculatePrefetchSchedule_params->Tno_bw = &mode_lib->ms.Tno_bw[k];
+			CalculatePrefetchSchedule_params->Tno_bw_flip = &mode_lib->ms.Tno_bw_flip[k];
+			CalculatePrefetchSchedule_params->prefetch_vmrow_bw = &mode_lib->ms.prefetch_vmrow_bw[k];
+			CalculatePrefetchSchedule_params->Tdmdl_vm = &s->dummy_single[0];
+			CalculatePrefetchSchedule_params->Tdmdl = &s->dummy_single[1];
+			CalculatePrefetchSchedule_params->TSetup = &s->dummy_single[2];
+			CalculatePrefetchSchedule_params->Tvm_trips = &s->Tvm_trips[k];
+			CalculatePrefetchSchedule_params->Tr0_trips = &s->Tr0_trips[k];
+			CalculatePrefetchSchedule_params->Tvm_trips_flip = &s->Tvm_trips_flip[k];
+			CalculatePrefetchSchedule_params->Tr0_trips_flip = &s->Tr0_trips_flip[k];
+			CalculatePrefetchSchedule_params->Tvm_trips_flip_rounded = &s->Tvm_trips_flip_rounded[k];
+			CalculatePrefetchSchedule_params->Tr0_trips_flip_rounded = &s->Tr0_trips_flip_rounded[k];
+			CalculatePrefetchSchedule_params->VUpdateOffsetPix = &s->dummy_integer[0];
+			CalculatePrefetchSchedule_params->VUpdateWidthPix = &s->dummy_integer[1];
+			CalculatePrefetchSchedule_params->VReadyOffsetPix = &s->dummy_integer[2];
+			CalculatePrefetchSchedule_params->prefetch_cursor_bw = &mode_lib->ms.prefetch_cursor_bw[k];
+			CalculatePrefetchSchedule_params->prefetch_sw_bytes = &s->prefetch_sw_bytes[k];
+			CalculatePrefetchSchedule_params->Tpre_rounded = &s->Tpre_rounded[k];
+			CalculatePrefetchSchedule_params->Tpre_oto = &s->Tpre_oto[k];
+			CalculatePrefetchSchedule_params->prefetch_swath_time_us = &s->prefetch_swath_time_us[k];
+
+			mode_lib->ms.NoTimeForPrefetch[k] = CalculatePrefetchSchedule(&mode_lib->scratch, CalculatePrefetchSchedule_params);
+
+			mode_lib->ms.support.PrefetchSupported &= !mode_lib->ms.NoTimeForPrefetch[k];
+			DML_LOG_VERBOSE("DML::%s: k=%d, dst_y_per_vm_vblank = %f\n", __func__, k, *CalculatePrefetchSchedule_params->dst_y_per_vm_vblank);
+			DML_LOG_VERBOSE("DML::%s: k=%d, dst_y_per_row_vblank = %f\n", __func__, k, *CalculatePrefetchSchedule_params->dst_y_per_row_vblank);
+		} // for k num_planes
+
+		CalculateDCFCLKDeepSleepTdlut(
+			display_cfg,
+			mode_lib->ms.num_active_planes,
+			mode_lib->ms.BytePerPixelY,
+			mode_lib->ms.BytePerPixelC,
+			mode_lib->ms.SwathWidthY,
+			mode_lib->ms.SwathWidthC,
+			mode_lib->ms.NoOfDPP,
+			mode_lib->ms.PSCL_FACTOR,
+			mode_lib->ms.PSCL_FACTOR_CHROMA,
+			mode_lib->ms.RequiredDPPCLK,
+			mode_lib->ms.vactive_sw_bw_l,
+			mode_lib->ms.vactive_sw_bw_c,
+			mode_lib->soc.return_bus_width_bytes,
+			mode_lib->ms.RequiredDISPCLK,
+			s->tdlut_bytes_to_deliver,
+			s->prefetch_swath_time_us,
+
+			/* Output */
+			&mode_lib->ms.dcfclk_deepsleep);
+
+		for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+			if (mode_lib->ms.dst_y_prefetch[k] < 2.0
+				|| mode_lib->ms.LinesForVM[k] >= 32.0
+				|| mode_lib->ms.LinesForDPTERow[k] >= 16.0
+				|| mode_lib->ms.NoTimeForPrefetch[k] == true
+				|| s->DSTYAfterScaler[k] > 8) {
+				mode_lib->ms.support.PrefetchSupported = false;
+				DML_LOG_VERBOSE("DML::%s: k=%d, dst_y_prefetch=%f (should not be < 2)\n", __func__, k, mode_lib->ms.dst_y_prefetch[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%d, LinesForVM=%f (should not be >= 32)\n", __func__, k, mode_lib->ms.LinesForVM[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%d, LinesForDPTERow=%f (should not be >= 16)\n", __func__, k, mode_lib->ms.LinesForDPTERow[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%d, DSTYAfterScaler=%d (should be <= 8)\n", __func__, k, s->DSTYAfterScaler[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%d, NoTimeForPrefetch=%d\n", __func__, k, mode_lib->ms.NoTimeForPrefetch[k]);
+			}
+		}
+
+		mode_lib->ms.support.DynamicMetadataSupported = true;
+		for (k = 0; k < mode_lib->ms.num_active_planes; ++k) {
+			if (mode_lib->ms.NoTimeForDynamicMetadata[k] == true) {
+				mode_lib->ms.support.DynamicMetadataSupported = false;
+			}
+		}
+
+		mode_lib->ms.support.VRatioInPrefetchSupported = true;
+		for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+			if (mode_lib->ms.VRatioPreY[k] > __DML2_CALCS_MAX_VRATIO_PRE__ ||
+				mode_lib->ms.VRatioPreC[k] > __DML2_CALCS_MAX_VRATIO_PRE__) {
+				mode_lib->ms.support.VRatioInPrefetchSupported = false;
+				DML_LOG_VERBOSE("DML::%s: k=%d VRatioPreY = %f (should be <= %f)\n", __func__, k, mode_lib->ms.VRatioPreY[k], __DML2_CALCS_MAX_VRATIO_PRE__);
+				DML_LOG_VERBOSE("DML::%s: k=%d VRatioPreC = %f (should be <= %f)\n", __func__, k, mode_lib->ms.VRatioPreC[k], __DML2_CALCS_MAX_VRATIO_PRE__);
+				DML_LOG_VERBOSE("DML::%s: VRatioInPrefetchSupported = %u\n", __func__, mode_lib->ms.support.VRatioInPrefetchSupported);
+			}
+		}
+
+		mode_lib->ms.support.PrefetchSupported &= mode_lib->ms.support.VRatioInPrefetchSupported;
+
+		// By default, do not recalc prefetch schedule
+		s->recalc_prefetch_schedule = 0;
+
+		// Only do urg vs prefetch bandwidth check, flip schedule check, power saving feature support check IF the Prefetch Schedule Check is ok
+		if (mode_lib->ms.support.PrefetchSupported) {
+			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+				// Calculate Urgent burst factor for prefetch
+#ifdef __DML_VBA_DEBUG__
+				DML_LOG_VERBOSE("DML::%s: k=%d, Calling CalculateUrgentBurstFactor (for prefetch)\n", __func__, k);
+				DML_LOG_VERBOSE("DML::%s: k=%d, VRatioPreY=%f\n", __func__, k, mode_lib->ms.VRatioPreY[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%d, VRatioPreC=%f\n", __func__, k, mode_lib->ms.VRatioPreC[k]);
+#endif
+				CalculateUrgentBurstFactor(
+					&display_cfg->plane_descriptors[k],
+					mode_lib->ms.swath_width_luma_ub[k],
+					mode_lib->ms.swath_width_chroma_ub[k],
+					mode_lib->ms.SwathHeightY[k],
+					mode_lib->ms.SwathHeightC[k],
+					s->line_times[k],
+					mode_lib->ms.UrgLatency,
+					mode_lib->ms.VRatioPreY[k],
+					mode_lib->ms.VRatioPreC[k],
+					mode_lib->ms.BytePerPixelInDETY[k],
+					mode_lib->ms.BytePerPixelInDETC[k],
+					mode_lib->ms.DETBufferSizeY[k],
+					mode_lib->ms.DETBufferSizeC[k],
+					/* Output */
+					&mode_lib->ms.UrgentBurstFactorLumaPre[k],
+					&mode_lib->ms.UrgentBurstFactorChromaPre[k],
+					&mode_lib->ms.NotEnoughUrgentLatencyHidingPre[k]);
+			}
+
+			// Calculate urgent bandwidth required, both urg and non urg peak bandwidth
+			// assume flip bw is 0 at this point
+			for (k = 0; k < mode_lib->ms.num_active_planes; k++)
+				mode_lib->ms.final_flip_bw[k] = 0;
+
+			calculate_peak_bandwidth_params->urg_vactive_bandwidth_required = mode_lib->ms.support.urg_vactive_bandwidth_required;
+			calculate_peak_bandwidth_params->urg_bandwidth_required = mode_lib->ms.support.urg_bandwidth_required;
+			calculate_peak_bandwidth_params->urg_bandwidth_required_qual = mode_lib->ms.support.urg_bandwidth_required_qual;
+			calculate_peak_bandwidth_params->non_urg_bandwidth_required = mode_lib->ms.support.non_urg_bandwidth_required;
+			calculate_peak_bandwidth_params->surface_avg_vactive_required_bw = mode_lib->ms.surface_avg_vactive_required_bw;
+			calculate_peak_bandwidth_params->surface_peak_required_bw = mode_lib->ms.surface_peak_required_bw;
+
+			calculate_peak_bandwidth_params->display_cfg = display_cfg;
+			calculate_peak_bandwidth_params->inc_flip_bw = 0;
+			calculate_peak_bandwidth_params->num_active_planes =  mode_lib->ms.num_active_planes;
+			calculate_peak_bandwidth_params->num_of_dpp = mode_lib->ms.NoOfDPP;
+			calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p0;
+			calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p1;
+			calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p0;
+			calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p1;
+			calculate_peak_bandwidth_params->mall_prefetch_sdp_overhead_factor = mode_lib->ms.mall_prefetch_sdp_overhead_factor;
+			calculate_peak_bandwidth_params->mall_prefetch_dram_overhead_factor = mode_lib->ms.mall_prefetch_dram_overhead_factor;
+
+			calculate_peak_bandwidth_params->surface_read_bandwidth_l = mode_lib->ms.vactive_sw_bw_l;
+			calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c;
+			calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma;
+			calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma;
+			calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO;
+			calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l;
+			calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c;
+			calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw;
+			calculate_peak_bandwidth_params->dpte_row_bw = mode_lib->ms.dpte_row_bw;
+			calculate_peak_bandwidth_params->meta_row_bw = mode_lib->ms.meta_row_bw;
+			calculate_peak_bandwidth_params->prefetch_cursor_bw = mode_lib->ms.prefetch_cursor_bw;
+			calculate_peak_bandwidth_params->prefetch_vmrow_bw = mode_lib->ms.prefetch_vmrow_bw;
+			calculate_peak_bandwidth_params->flip_bw = mode_lib->ms.final_flip_bw;
+			calculate_peak_bandwidth_params->urgent_burst_factor_l = mode_lib->ms.UrgentBurstFactorLuma;
+			calculate_peak_bandwidth_params->urgent_burst_factor_c = mode_lib->ms.UrgentBurstFactorChroma;
+			calculate_peak_bandwidth_params->urgent_burst_factor_cursor = mode_lib->ms.UrgentBurstFactorCursor;
+			calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_l = mode_lib->ms.UrgentBurstFactorLumaPre;
+			calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_c = mode_lib->ms.UrgentBurstFactorChromaPre;
+			calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_cursor = mode_lib->ms.UrgentBurstFactorCursorPre;
+
+			calculate_peak_bandwidth_required(
+					&mode_lib->scratch,
+					calculate_peak_bandwidth_params);
+
+			// Check urg peak bandwidth against available urg bw
+			// check at SDP and DRAM, for all soc states (SVP prefetch an Sys Active)
+			check_urgent_bandwidth_support(
+				&s->dummy_single[0], // double* frac_urg_bandwidth
+				&s->dummy_single[1], // double* frac_urg_bandwidth_mall
+				&mode_lib->ms.support.UrgVactiveBandwidthSupport,
+				&mode_lib->ms.support.PrefetchBandwidthSupported,
+
+				mode_lib->soc.mall_allocated_for_dcn_mbytes,
+				mode_lib->ms.support.non_urg_bandwidth_required,
+				mode_lib->ms.support.urg_vactive_bandwidth_required,
+				mode_lib->ms.support.urg_bandwidth_required,
+				mode_lib->ms.support.urg_bandwidth_available);
+
+			mode_lib->ms.support.PrefetchSupported &= mode_lib->ms.support.PrefetchBandwidthSupported;
+			DML_LOG_VERBOSE("DML::%s: PrefetchBandwidthSupported=%0d\n", __func__, mode_lib->ms.support.PrefetchBandwidthSupported);
+
+			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+				if (mode_lib->ms.NotEnoughUrgentLatencyHidingPre[k]) {
+					mode_lib->ms.support.PrefetchSupported = false;
+					DML_LOG_VERBOSE("DML::%s: k=%d, NotEnoughUrgentLatencyHidingPre=%d\n", __func__, k, mode_lib->ms.NotEnoughUrgentLatencyHidingPre[k]);
+				}
+			}
+
+#ifdef DML_GLOBAL_PREFETCH_CHECK
+			if (mode_lib->ms.support.PrefetchSupported && mode_lib->ms.num_active_planes > 1 && s->recalc_prefetch_done == 0) {
+				CheckGlobalPrefetchAdmissibility_params->num_active_planes =  mode_lib->ms.num_active_planes;
+				CheckGlobalPrefetchAdmissibility_params->pixel_format = s->pixel_format;
+				CheckGlobalPrefetchAdmissibility_params->chunk_bytes_l = mode_lib->ip.pixel_chunk_size_kbytes * 1024;
+				CheckGlobalPrefetchAdmissibility_params->chunk_bytes_c = mode_lib->ip.pixel_chunk_size_kbytes * 1024;
+				CheckGlobalPrefetchAdmissibility_params->lb_source_lines_l = s->lb_source_lines_l;
+				CheckGlobalPrefetchAdmissibility_params->lb_source_lines_c = s->lb_source_lines_c;
+				CheckGlobalPrefetchAdmissibility_params->swath_height_l =  mode_lib->ms.SwathHeightY;
+				CheckGlobalPrefetchAdmissibility_params->swath_height_c =  mode_lib->ms.SwathHeightC;
+				CheckGlobalPrefetchAdmissibility_params->rob_buffer_size_kbytes = mode_lib->ip.rob_buffer_size_kbytes;
+				CheckGlobalPrefetchAdmissibility_params->compressed_buffer_size_kbytes = mode_lib->ms.CompressedBufferSizeInkByte;
+				CheckGlobalPrefetchAdmissibility_params->detile_buffer_size_bytes_l = mode_lib->ms.DETBufferSizeY;
+				CheckGlobalPrefetchAdmissibility_params->detile_buffer_size_bytes_c = mode_lib->ms.DETBufferSizeC;
+				CheckGlobalPrefetchAdmissibility_params->full_swath_bytes_l = s->full_swath_bytes_l;
+				CheckGlobalPrefetchAdmissibility_params->full_swath_bytes_c = s->full_swath_bytes_c;
+				CheckGlobalPrefetchAdmissibility_params->prefetch_sw_bytes = s->prefetch_sw_bytes;
+				CheckGlobalPrefetchAdmissibility_params->Tpre_rounded = s->Tpre_rounded;
+				CheckGlobalPrefetchAdmissibility_params->Tpre_oto = s->Tpre_oto;
+				CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps = mode_lib->ms.support.urg_bandwidth_required[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_sdp];
+				CheckGlobalPrefetchAdmissibility_params->line_time = s->line_times;
+				CheckGlobalPrefetchAdmissibility_params->dst_y_prefetch = mode_lib->ms.dst_y_prefetch;
+				if (CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps < 10 * 1024)
+					CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps = 10 * 1024;
+
+				CheckGlobalPrefetchAdmissibility_params->estimated_dcfclk_mhz = (CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps / (double) mode_lib->soc.return_bus_width_bytes) /
+																				((double)mode_lib->soc.qos_parameters.derate_table.system_active_urgent.dcfclk_derate_percent / 100.0);
+
+				// if recalc_prefetch_schedule is set, recalculate the prefetch schedule with the new impacted_Tpre, prefetch should be possible
+				CheckGlobalPrefetchAdmissibility_params->recalc_prefetch_schedule = &s->recalc_prefetch_schedule;
+				CheckGlobalPrefetchAdmissibility_params->impacted_dst_y_pre = s->impacted_dst_y_pre;
+				mode_lib->ms.support.PrefetchSupported = CheckGlobalPrefetchAdmissibility(&mode_lib->scratch, CheckGlobalPrefetchAdmissibility_params);
+				s->recalc_prefetch_done = 1;
+				s->recalc_prefetch_schedule = 1;
+			}
+#endif
+		} // prefetch schedule ok, do urg bw and flip schedule
+	} while (s->recalc_prefetch_schedule);
+
+	// Flip Schedule
+	// Both prefetch schedule and BW okay
+	if (mode_lib->ms.support.PrefetchSupported == true) {
+		mode_lib->ms.BandwidthAvailableForImmediateFlip =
+			get_bandwidth_available_for_immediate_flip(
+				dml2_core_internal_soc_state_sys_active,
+				mode_lib->ms.support.urg_bandwidth_required_qual, // no flip
+				mode_lib->ms.support.urg_bandwidth_available);
+
+		mode_lib->ms.TotImmediateFlipBytes = 0;
+		for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+			if (display_cfg->plane_descriptors[k].immediate_flip) {
+				s->per_pipe_flip_bytes[k] = get_pipe_flip_bytes(
+								s->HostVMInefficiencyFactor,
+								mode_lib->ms.vm_bytes[k],
+								mode_lib->ms.DPTEBytesPerRow[k],
+								mode_lib->ms.meta_row_bytes[k]);
+			} else {
+				s->per_pipe_flip_bytes[k] = 0;
+			}
+			mode_lib->ms.TotImmediateFlipBytes += s->per_pipe_flip_bytes[k] * mode_lib->ms.NoOfDPP[k];
+
+		}
+
+		for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
+			CalculateFlipSchedule(
+				&mode_lib->scratch,
+				display_cfg->plane_descriptors[k].immediate_flip,
+				1, // use_lb_flip_bw
+				s->HostVMInefficiencyFactor,
+				s->Tvm_trips_flip[k],
+				s->Tr0_trips_flip[k],
+				s->Tvm_trips_flip_rounded[k],
+				s->Tr0_trips_flip_rounded[k],
+				display_cfg->gpuvm_enable,
+				mode_lib->ms.vm_bytes[k],
+				mode_lib->ms.DPTEBytesPerRow[k],
+				mode_lib->ms.BandwidthAvailableForImmediateFlip,
+				mode_lib->ms.TotImmediateFlipBytes,
+				display_cfg->plane_descriptors[k].pixel_format,
+				(display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)),
+				display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio,
+				display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio,
+				mode_lib->ms.Tno_bw_flip[k],
+				mode_lib->ms.dpte_row_height[k],
+				mode_lib->ms.dpte_row_height_chroma[k],
+				mode_lib->ms.use_one_row_for_frame_flip[k],
+				mode_lib->ip.max_flip_time_us,
+				mode_lib->ip.max_flip_time_lines,
+				s->per_pipe_flip_bytes[k],
+				mode_lib->ms.meta_row_bytes[k],
+				s->meta_row_height_luma[k],
+				s->meta_row_height_chroma[k],
+				mode_lib->ip.dcn_mrq_present && display_cfg->plane_descriptors[k].surface.dcc.enable,
+
+				/* Output */
+				&mode_lib->ms.dst_y_per_vm_flip[k],
+				&mode_lib->ms.dst_y_per_row_flip[k],
+				&mode_lib->ms.final_flip_bw[k],
+				&mode_lib->ms.ImmediateFlipSupportedForPipe[k]);
+		}
+
+		calculate_peak_bandwidth_params->urg_vactive_bandwidth_required = s->dummy_bw;
+		calculate_peak_bandwidth_params->urg_bandwidth_required = mode_lib->ms.support.urg_bandwidth_required_flip;
+		calculate_peak_bandwidth_params->urg_bandwidth_required_qual = s->dummy_bw;
+		calculate_peak_bandwidth_params->non_urg_bandwidth_required = mode_lib->ms.support.non_urg_bandwidth_required_flip;
+		calculate_peak_bandwidth_params->surface_avg_vactive_required_bw = s->surface_dummy_bw;
+		calculate_peak_bandwidth_params->surface_peak_required_bw = mode_lib->ms.surface_peak_required_bw;
+
+		calculate_peak_bandwidth_params->display_cfg = display_cfg;
+		calculate_peak_bandwidth_params->inc_flip_bw = 1;
+		calculate_peak_bandwidth_params->num_active_planes = mode_lib->ms.num_active_planes;
+		calculate_peak_bandwidth_params->num_of_dpp = mode_lib->ms.NoOfDPP;
+		calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p0;
+		calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p1;
+		calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p0;
+		calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p1;
+		calculate_peak_bandwidth_params->mall_prefetch_sdp_overhead_factor = mode_lib->ms.mall_prefetch_sdp_overhead_factor;
+		calculate_peak_bandwidth_params->mall_prefetch_dram_overhead_factor = mode_lib->ms.mall_prefetch_dram_overhead_factor;
+
+		calculate_peak_bandwidth_params->surface_read_bandwidth_l = mode_lib->ms.vactive_sw_bw_l;
+		calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c;
+		calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma;
+		calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma;
+		calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO;
+		calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l;
+		calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c;
+		calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw;
+		calculate_peak_bandwidth_params->dpte_row_bw = mode_lib->ms.dpte_row_bw;
+		calculate_peak_bandwidth_params->meta_row_bw = mode_lib->ms.meta_row_bw;
+		calculate_peak_bandwidth_params->prefetch_cursor_bw = mode_lib->ms.prefetch_cursor_bw;
+		calculate_peak_bandwidth_params->prefetch_vmrow_bw = mode_lib->ms.prefetch_vmrow_bw;
+		calculate_peak_bandwidth_params->flip_bw = mode_lib->ms.final_flip_bw;
+		calculate_peak_bandwidth_params->urgent_burst_factor_l = mode_lib->ms.UrgentBurstFactorLuma;
+		calculate_peak_bandwidth_params->urgent_burst_factor_c = mode_lib->ms.UrgentBurstFactorChroma;
+		calculate_peak_bandwidth_params->urgent_burst_factor_cursor = mode_lib->ms.UrgentBurstFactorCursor;
+		calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_l = mode_lib->ms.UrgentBurstFactorLumaPre;
+		calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_c = mode_lib->ms.UrgentBurstFactorChromaPre;
+		calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_cursor = mode_lib->ms.UrgentBurstFactorCursorPre;
+
+		calculate_peak_bandwidth_required(
+				&mode_lib->scratch,
+				calculate_peak_bandwidth_params);
+
+		calculate_immediate_flip_bandwidth_support(
+			&s->dummy_single[0], // double* frac_urg_bandwidth_flip
+			&mode_lib->ms.support.ImmediateFlipSupport,
+
+			dml2_core_internal_soc_state_sys_active,
+			mode_lib->ms.support.urg_bandwidth_required_flip,
+			mode_lib->ms.support.non_urg_bandwidth_required_flip,
+			mode_lib->ms.support.urg_bandwidth_available);
+
+		for (k = 0; k <= mode_lib->ms.num_active_planes - 1; k++) {
+			if (display_cfg->plane_descriptors[k].immediate_flip == true && mode_lib->ms.ImmediateFlipSupportedForPipe[k] == false)
+				mode_lib->ms.support.ImmediateFlipSupport = false;
+		}
+
+	} else { // if prefetch not support, assume iflip is not supported too
+		mode_lib->ms.support.ImmediateFlipSupport = false;
+	}
+
+	s->mSOCParameters.UrgentLatency = mode_lib->ms.UrgLatency;
+	s->mSOCParameters.ExtraLatency = mode_lib->ms.ExtraLatency;
+	s->mSOCParameters.ExtraLatency_sr = mode_lib->ms.ExtraLatency_sr;
+	s->mSOCParameters.WritebackLatency = mode_lib->soc.qos_parameters.writeback.base_latency_us;
+	s->mSOCParameters.DRAMClockChangeLatency = mode_lib->soc.power_management_parameters.dram_clk_change_blackout_us;
+	s->mSOCParameters.FCLKChangeLatency = mode_lib->soc.power_management_parameters.fclk_change_blackout_us;
+	s->mSOCParameters.SRExitTime = mode_lib->soc.power_management_parameters.stutter_exit_latency_us;
+	s->mSOCParameters.SREnterPlusExitTime = mode_lib->soc.power_management_parameters.stutter_enter_plus_exit_latency_us;
+	s->mSOCParameters.SRExitZ8Time = mode_lib->soc.power_management_parameters.z8_stutter_exit_latency_us;
+	s->mSOCParameters.SREnterPlusExitZ8Time = mode_lib->soc.power_management_parameters.z8_stutter_enter_plus_exit_latency_us;
+	s->mSOCParameters.USRRetrainingLatency = 0;
+	s->mSOCParameters.SMNLatency = 0;
+	s->mSOCParameters.g6_temp_read_blackout_us = get_g6_temp_read_blackout_us(&mode_lib->soc, (unsigned int)(mode_lib->ms.uclk_freq_mhz * 1000), mode_lib->ms.state_idx);
+	s->mSOCParameters.max_urgent_latency_us = get_max_urgent_latency_us(&mode_lib->soc.qos_parameters.qos_params.dcn4x, mode_lib->ms.uclk_freq_mhz, mode_lib->ms.FabricClock, mode_lib->ms.state_idx);
+	s->mSOCParameters.df_response_time_us = mode_lib->soc.qos_parameters.qos_params.dcn4x.df_qos_response_time_fclk_cycles / mode_lib->ms.FabricClock;
+	s->mSOCParameters.qos_type = mode_lib->soc.qos_parameters.qos_type;
+
+	CalculateWatermarks_params->display_cfg = display_cfg;
+	CalculateWatermarks_params->USRRetrainingRequired = false;
+	CalculateWatermarks_params->NumberOfActiveSurfaces = mode_lib->ms.num_active_planes;
+	CalculateWatermarks_params->MaxLineBufferLines = mode_lib->ip.max_line_buffer_lines;
+	CalculateWatermarks_params->LineBufferSize = mode_lib->ip.line_buffer_size_bits;
+	CalculateWatermarks_params->WritebackInterfaceBufferSize = mode_lib->ip.writeback_interface_buffer_size_kbytes;
+	CalculateWatermarks_params->DCFCLK = mode_lib->ms.DCFCLK;
+	CalculateWatermarks_params->SynchronizeTimings = display_cfg->overrides.synchronize_timings;
+	CalculateWatermarks_params->SynchronizeDRRDisplaysForUCLKPStateChange = display_cfg->overrides.synchronize_ddr_displays_for_uclk_pstate_change;
+	CalculateWatermarks_params->dpte_group_bytes = mode_lib->ms.dpte_group_bytes;
+	CalculateWatermarks_params->mmSOCParameters = s->mSOCParameters;
+	CalculateWatermarks_params->WritebackChunkSize = mode_lib->ip.writeback_chunk_size_kbytes;
+	CalculateWatermarks_params->SOCCLK = mode_lib->ms.SOCCLK;
+	CalculateWatermarks_params->DCFClkDeepSleep = mode_lib->ms.dcfclk_deepsleep;
+	CalculateWatermarks_params->DETBufferSizeY = mode_lib->ms.DETBufferSizeY;
+	CalculateWatermarks_params->DETBufferSizeC = mode_lib->ms.DETBufferSizeC;
+	CalculateWatermarks_params->SwathHeightY = mode_lib->ms.SwathHeightY;
+	CalculateWatermarks_params->SwathHeightC = mode_lib->ms.SwathHeightC;
+	CalculateWatermarks_params->SwathWidthY = mode_lib->ms.SwathWidthY;
+	CalculateWatermarks_params->SwathWidthC = mode_lib->ms.SwathWidthC;
+	CalculateWatermarks_params->DPPPerSurface = mode_lib->ms.NoOfDPP;
+	CalculateWatermarks_params->BytePerPixelDETY = mode_lib->ms.BytePerPixelInDETY;
+	CalculateWatermarks_params->BytePerPixelDETC = mode_lib->ms.BytePerPixelInDETC;
+	CalculateWatermarks_params->DSTXAfterScaler = s->DSTXAfterScaler;
+	CalculateWatermarks_params->DSTYAfterScaler = s->DSTYAfterScaler;
+	CalculateWatermarks_params->UnboundedRequestEnabled = mode_lib->ms.UnboundedRequestEnabled;
+	CalculateWatermarks_params->CompressedBufferSizeInkByte = mode_lib->ms.CompressedBufferSizeInkByte;
+	CalculateWatermarks_params->meta_row_height_l = s->meta_row_height_luma;
+	CalculateWatermarks_params->meta_row_height_c = s->meta_row_height_chroma;
+
+	// Output
+	CalculateWatermarks_params->Watermark = &mode_lib->ms.support.watermarks; // Watermarks *Watermark
+	CalculateWatermarks_params->DRAMClockChangeSupport = mode_lib->ms.support.DRAMClockChangeSupport;
+	CalculateWatermarks_params->global_dram_clock_change_supported = &mode_lib->ms.support.global_dram_clock_change_supported;
+	CalculateWatermarks_params->MaxActiveDRAMClockChangeLatencySupported = &s->dummy_single_array[0]; // double *MaxActiveDRAMClockChangeLatencySupported[]
+	CalculateWatermarks_params->SubViewportLinesNeededInMALL = mode_lib->ms.SubViewportLinesNeededInMALL; // unsigned int SubViewportLinesNeededInMALL[]
+	CalculateWatermarks_params->FCLKChangeSupport = mode_lib->ms.support.FCLKChangeSupport;
+	CalculateWatermarks_params->global_fclk_change_supported = &mode_lib->ms.support.global_fclk_change_supported;
+	CalculateWatermarks_params->MaxActiveFCLKChangeLatencySupported = &s->dummy_single[0]; // double *MaxActiveFCLKChangeLatencySupported
+	CalculateWatermarks_params->USRRetrainingSupport = &mode_lib->ms.support.USRRetrainingSupport;
+	CalculateWatermarks_params->g6_temp_read_support = &mode_lib->ms.support.g6_temp_read_support;
+	CalculateWatermarks_params->VActiveLatencyHidingMargin = mode_lib->ms.VActiveLatencyHidingMargin;
+	CalculateWatermarks_params->VActiveLatencyHidingUs = mode_lib->ms.VActiveLatencyHidingUs;
+
+	CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(&mode_lib->scratch, CalculateWatermarks_params);
+
+	calculate_pstate_keepout_dst_lines(display_cfg, &mode_lib->ms.support.watermarks, s->dummy_integer_array[0]);
+	DML_LOG_VERBOSE("DML::%s: Done prefetch calculation\n", __func__);
+
+}
+
 static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out_params)
 {
 	struct dml2_core_internal_display_mode_lib *mode_lib = in_out_params->mode_lib;
 	const struct dml2_display_cfg *display_cfg = in_out_params->in_display_cfg;
 	const struct dml2_mcg_min_clock_table *min_clk_table = in_out_params->min_clk_table;
 
-#if defined(__DML_VBA_DEBUG__)
-	double old_ReadBandwidthLuma;
-	double old_ReadBandwidthChroma;
-#endif
 	double outstanding_latency_us = 0;
-	double min_return_bw_for_latency;
 
 	struct dml2_core_calcs_mode_support_locals *s = &mode_lib->scratch.dml_core_mode_support_locals;
-	struct dml2_core_calcs_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport_params *CalculateWatermarks_params = &mode_lib->scratch.CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport_params;
 	struct dml2_core_calcs_CalculateVMRowAndSwath_params *CalculateVMRowAndSwath_params = &mode_lib->scratch.CalculateVMRowAndSwath_params;
 	struct dml2_core_calcs_CalculateSwathAndDETConfiguration_params *CalculateSwathAndDETConfiguration_params = &mode_lib->scratch.CalculateSwathAndDETConfiguration_params;
-	struct dml2_core_calcs_CalculatePrefetchSchedule_params *CalculatePrefetchSchedule_params = &mode_lib->scratch.CalculatePrefetchSchedule_params;
-#ifdef DML_GLOBAL_PREFETCH_CHECK
-	struct dml2_core_calcs_CheckGlobalPrefetchAdmissibility_params *CheckGlobalPrefetchAdmissibility_params = &mode_lib->scratch.CheckGlobalPrefetchAdmissibility_params;
-#endif
-	struct dml2_core_calcs_calculate_tdlut_setting_params *calculate_tdlut_setting_params = &mode_lib->scratch.calculate_tdlut_setting_params;
 	struct dml2_core_calcs_calculate_mcache_setting_params *calculate_mcache_setting_params = &mode_lib->scratch.calculate_mcache_setting_params;
-	struct dml2_core_calcs_calculate_peak_bandwidth_required_params *calculate_peak_bandwidth_params = &mode_lib->scratch.calculate_peak_bandwidth_params;
 	struct dml2_core_calcs_calculate_bytes_to_fetch_required_to_hide_latency_params *calculate_bytes_to_fetch_required_to_hide_latency_params = &mode_lib->scratch.calculate_bytes_to_fetch_required_to_hide_latency_params;
 	unsigned int k, m, n;
 
@@ -7374,9 +7932,9 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 	mode_lib->ms.FabricClock = ((double)min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].min_fclk_khz / 1000);
 	mode_lib->ms.MaxDCFCLK = (double)min_clk_table->max_clocks_khz.dcfclk / 1000;
 	mode_lib->ms.MaxFabricClock = (double)min_clk_table->max_clocks_khz.fclk / 1000;
-	mode_lib->ms.max_dispclk_freq_mhz = (double)min_clk_table->max_clocks_khz.dispclk / 1000;
+	mode_lib->ms.max_dispclk_freq_mhz = (double)min_clk_table->max_ss_clocks_khz.dispclk / 1000;
 	mode_lib->ms.max_dscclk_freq_mhz = (double)min_clk_table->max_clocks_khz.dscclk / 1000;
-	mode_lib->ms.max_dppclk_freq_mhz = (double)min_clk_table->max_clocks_khz.dppclk / 1000;
+	mode_lib->ms.max_dppclk_freq_mhz = (double)min_clk_table->max_ss_clocks_khz.dppclk / 1000;
 	mode_lib->ms.uclk_freq_mhz = dram_bw_kbps_to_uclk_mhz(min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].pre_derate_dram_bw_kbps, &mode_lib->soc.clk_table.dram_config);
 	mode_lib->ms.dram_bw_mbps = ((double)min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].pre_derate_dram_bw_kbps / 1000);
 	mode_lib->ms.max_dram_bw_mbps = ((double)min_clk_table->dram_bw_table.entries[min_clk_table->dram_bw_table.num_entries - 1].pre_derate_dram_bw_kbps / 1000);
@@ -7384,25 +7942,25 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 	mode_lib->ms.active_min_uclk_dpm_index = get_active_min_uclk_dpm_index((unsigned int) (mode_lib->ms.uclk_freq_mhz * 1000.0), &mode_lib->soc.clk_table);
 
 #if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: --- START --- \n", __func__);
-	dml2_printf("DML::%s: num_active_planes = %u\n", __func__, mode_lib->ms.num_active_planes);
-	dml2_printf("DML::%s: min_clk_index = %0d\n", __func__, in_out_params->min_clk_index);
-	dml2_printf("DML::%s: qos_param_index = %0d\n", __func__, mode_lib->ms.qos_param_index);
-	dml2_printf("DML::%s: SOCCLK = %f\n", __func__, mode_lib->ms.SOCCLK);
-	dml2_printf("DML::%s: dram_bw_mbps = %f\n", __func__, mode_lib->ms.dram_bw_mbps);
-	dml2_printf("DML::%s: uclk_freq_mhz = %f\n", __func__, mode_lib->ms.uclk_freq_mhz);
-	dml2_printf("DML::%s: DCFCLK = %f\n", __func__, mode_lib->ms.DCFCLK);
-	dml2_printf("DML::%s: FabricClock = %f\n", __func__, mode_lib->ms.FabricClock);
-	dml2_printf("DML::%s: MaxDCFCLK = %f\n", __func__, mode_lib->ms.MaxDCFCLK);
-	dml2_printf("DML::%s: max_dispclk_freq_mhz = %f\n", __func__, mode_lib->ms.max_dispclk_freq_mhz);
-	dml2_printf("DML::%s: max_dscclk_freq_mhz = %f\n", __func__, mode_lib->ms.max_dscclk_freq_mhz);
-	dml2_printf("DML::%s: max_dppclk_freq_mhz = %f\n", __func__, mode_lib->ms.max_dppclk_freq_mhz);
-	dml2_printf("DML::%s: MaxFabricClock = %f\n", __func__, mode_lib->ms.MaxFabricClock);
-	dml2_printf("DML::%s: ip.compressed_buffer_segment_size_in_kbytes = %u\n", __func__, mode_lib->ip.compressed_buffer_segment_size_in_kbytes);
-	dml2_printf("DML::%s: ip.dcn_mrq_present = %u\n", __func__, mode_lib->ip.dcn_mrq_present);
+	DML_LOG_VERBOSE("DML::%s: --- START --- \n", __func__);
+	DML_LOG_VERBOSE("DML::%s: num_active_planes = %u\n", __func__, mode_lib->ms.num_active_planes);
+	DML_LOG_VERBOSE("DML::%s: min_clk_index = %0d\n", __func__, in_out_params->min_clk_index);
+	DML_LOG_VERBOSE("DML::%s: qos_param_index = %0d\n", __func__, mode_lib->ms.qos_param_index);
+	DML_LOG_VERBOSE("DML::%s: SOCCLK = %f\n", __func__, mode_lib->ms.SOCCLK);
+	DML_LOG_VERBOSE("DML::%s: dram_bw_mbps = %f\n", __func__, mode_lib->ms.dram_bw_mbps);
+	DML_LOG_VERBOSE("DML::%s: uclk_freq_mhz = %f\n", __func__, mode_lib->ms.uclk_freq_mhz);
+	DML_LOG_VERBOSE("DML::%s: DCFCLK = %f\n", __func__, mode_lib->ms.DCFCLK);
+	DML_LOG_VERBOSE("DML::%s: FabricClock = %f\n", __func__, mode_lib->ms.FabricClock);
+	DML_LOG_VERBOSE("DML::%s: MaxDCFCLK = %f\n", __func__, mode_lib->ms.MaxDCFCLK);
+	DML_LOG_VERBOSE("DML::%s: max_dispclk_freq_mhz = %f\n", __func__, mode_lib->ms.max_dispclk_freq_mhz);
+	DML_LOG_VERBOSE("DML::%s: max_dscclk_freq_mhz = %f\n", __func__, mode_lib->ms.max_dscclk_freq_mhz);
+	DML_LOG_VERBOSE("DML::%s: max_dppclk_freq_mhz = %f\n", __func__, mode_lib->ms.max_dppclk_freq_mhz);
+	DML_LOG_VERBOSE("DML::%s: MaxFabricClock = %f\n", __func__, mode_lib->ms.MaxFabricClock);
+	DML_LOG_VERBOSE("DML::%s: ip.compressed_buffer_segment_size_in_kbytes = %u\n", __func__, mode_lib->ip.compressed_buffer_segment_size_in_kbytes);
+	DML_LOG_VERBOSE("DML::%s: ip.dcn_mrq_present = %u\n", __func__, mode_lib->ip.dcn_mrq_present);
 
 	for (k = 0; k < mode_lib->ms.num_active_planes; k++)
-		dml2_printf("DML::%s: plane_%d: reserved_vblank_time_ns = %u\n", __func__, k, display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns);
+		DML_LOG_VERBOSE("DML::%s: plane_%d: reserved_vblank_time_ns = %lu\n", __func__, k, display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns);
 #endif
 
 	CalculateMaxDETAndMinCompressedBufferSize(
@@ -7504,12 +8062,10 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			display_cfg->plane_descriptors[k].cursor.cursor_bpp / 8.0 / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000));
 
 #ifdef __DML_VBA_DEBUG__
-		old_ReadBandwidthLuma = mode_lib->ms.SwathWidthYSingleDPP[k] * math_ceil2(mode_lib->ms.BytePerPixelInDETY[k], 1.0) / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) * display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio;
-		old_ReadBandwidthChroma = mode_lib->ms.SwathWidthYSingleDPP[k] / 2 * math_ceil2(mode_lib->ms.BytePerPixelInDETC[k], 2.0) / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) * display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio / 2.0;
-		dml2_printf("DML::%s: k=%u, old_ReadBandwidthLuma = %f\n", __func__, k, old_ReadBandwidthLuma);
-		dml2_printf("DML::%s: k=%u, old_ReadBandwidthChroma = %f\n", __func__, k, old_ReadBandwidthChroma);
-		dml2_printf("DML::%s: k=%u, vactive_sw_bw_l = %f\n", __func__, k, mode_lib->ms.vactive_sw_bw_l[k]);
-		dml2_printf("DML::%s: k=%u, vactive_sw_bw_c = %f\n", __func__, k, mode_lib->ms.vactive_sw_bw_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, old_ReadBandwidthLuma = %f\n", __func__, k, mode_lib->ms.SwathWidthYSingleDPP[k] * math_ceil2(mode_lib->ms.BytePerPixelInDETY[k], 1.0) / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) * display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u, old_ReadBandwidthChroma = %f\n", __func__, k, mode_lib->ms.SwathWidthYSingleDPP[k] / 2 * math_ceil2(mode_lib->ms.BytePerPixelInDETC[k], 2.0) / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) * display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio / 2.0);
+		DML_LOG_VERBOSE("DML::%s: k=%u, vactive_sw_bw_l = %f\n", __func__, k, mode_lib->ms.vactive_sw_bw_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, vactive_sw_bw_c = %f\n", __func__, k, mode_lib->ms.vactive_sw_bw_c[k]);
 #endif
 	}
 
@@ -7629,13 +8185,13 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 		mode_lib->ms.MaximumSwathWidthLuma[k] = math_min2(s->MaximumSwathWidthSupportLuma, mode_lib->ms.MaximumSwathWidthInLineBufferLuma);
 		mode_lib->ms.MaximumSwathWidthChroma[k] = math_min2(s->MaximumSwathWidthSupportChroma, mode_lib->ms.MaximumSwathWidthInLineBufferChroma);
 
-		dml2_printf("DML::%s: k=%u MaximumSwathWidthLuma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthLuma[k]);
-		dml2_printf("DML::%s: k=%u MaximumSwathWidthSupportLuma=%u\n", __func__, k, s->MaximumSwathWidthSupportLuma);
-		dml2_printf("DML::%s: k=%u MaximumSwathWidthInLineBufferLuma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthInLineBufferLuma);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthLuma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthLuma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthSupportLuma=%u\n", __func__, k, s->MaximumSwathWidthSupportLuma);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthInLineBufferLuma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthInLineBufferLuma);
 
-		dml2_printf("DML::%s: k=%u MaximumSwathWidthChroma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthChroma[k]);
-		dml2_printf("DML::%s: k=%u MaximumSwathWidthSupportChroma=%u\n", __func__, k, s->MaximumSwathWidthSupportChroma);
-		dml2_printf("DML::%s: k=%u MaximumSwathWidthInLineBufferChroma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthInLineBufferChroma);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthChroma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthSupportChroma=%u\n", __func__, k, s->MaximumSwathWidthSupportChroma);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaximumSwathWidthInLineBufferChroma=%f\n", __func__, k, mode_lib->ms.MaximumSwathWidthInLineBufferChroma);
 	}
 
 	/* Cursor Support Check */
@@ -7672,11 +8228,11 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			mode_lib->ms.support.AlignedCPitch[k] > display_cfg->plane_descriptors[k].surface.plane1.pitch) {
 			mode_lib->ms.support.PitchSupport = false;
 #if defined(__DML_VBA_DEBUG__)
-			dml2_printf("DML::%s: k=%u AlignedYPitch = %d\n", __func__, k, mode_lib->ms.support.AlignedYPitch[k]);
-			dml2_printf("DML::%s: k=%u PitchY = %d\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane0.pitch);
-			dml2_printf("DML::%s: k=%u AlignedCPitch = %d\n", __func__, k, mode_lib->ms.support.AlignedCPitch[k]);
-			dml2_printf("DML::%s: k=%u PitchC = %d\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane1.pitch);
-			dml2_printf("DML::%s: k=%u PitchSupport = %d\n", __func__, k, mode_lib->ms.support.PitchSupport);
+			DML_LOG_VERBOSE("DML::%s: k=%u AlignedYPitch = %d\n", __func__, k, mode_lib->ms.support.AlignedYPitch[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u PitchY = %ld\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane0.pitch);
+			DML_LOG_VERBOSE("DML::%s: k=%u AlignedCPitch = %d\n", __func__, k, mode_lib->ms.support.AlignedCPitch[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u PitchC = %ld\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane1.pitch);
+			DML_LOG_VERBOSE("DML::%s: k=%u PitchSupport = %d\n", __func__, k, mode_lib->ms.support.PitchSupport);
 #endif
 		}
 
@@ -7708,11 +8264,11 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 				display_cfg->plane_descriptors[k].composition.viewport.plane0.height > display_cfg->plane_descriptors[k].surface.plane0.height) {
 				mode_lib->ms.support.ViewportExceedsSurface = true;
 #if defined(__DML_VBA_DEBUG__)
-				dml2_printf("DML::%s: k=%u ViewportWidth = %d\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.width);
-				dml2_printf("DML::%s: k=%u SurfaceWidthY = %d\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane0.width);
-				dml2_printf("DML::%s: k=%u ViewportHeight = %d\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.height);
-				dml2_printf("DML::%s: k=%u SurfaceHeightY = %d\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane0.height);
-				dml2_printf("DML::%s: k=%u ViewportExceedsSurface = %d\n", __func__, k, mode_lib->ms.support.ViewportExceedsSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u ViewportWidth = %ld\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.width);
+				DML_LOG_VERBOSE("DML::%s: k=%u SurfaceWidthY = %ld\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane0.width);
+				DML_LOG_VERBOSE("DML::%s: k=%u ViewportHeight = %ld\n", __func__, k, display_cfg->plane_descriptors[k].composition.viewport.plane0.height);
+				DML_LOG_VERBOSE("DML::%s: k=%u SurfaceHeightY = %ld\n", __func__, k, display_cfg->plane_descriptors[k].surface.plane0.height);
+				DML_LOG_VERBOSE("DML::%s: k=%u ViewportExceedsSurface = %d\n", __func__, k, mode_lib->ms.support.ViewportExceedsSurface);
 #endif
 			}
 			if (dml_is_420(display_cfg->plane_descriptors[k].pixel_format) || display_cfg->plane_descriptors[k].pixel_format == dml2_rgbe_alpha) {
@@ -7894,8 +8450,8 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			mode_lib->ms.TotalNumberOfActiveDPP = mode_lib->ms.TotalNumberOfActiveDPP + s->NumberOfDPPDSC;
 		}
 #if defined(__DML_VBA_DEBUG__)
-		dml2_printf("DML::%s: k=%d RequiresDSC = %d\n", __func__, k, mode_lib->ms.RequiresDSC[k]);
-		dml2_printf("DML::%s: k=%d ODMMode = %d\n", __func__, k, mode_lib->ms.ODMMode[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d RequiresDSC = %d\n", __func__, k, mode_lib->ms.RequiresDSC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d ODMMode = %d\n", __func__, k, mode_lib->ms.ODMMode[k]);
 #endif
 
 		// ensure the number dsc slices is integer multiple based on ODM mode
@@ -7911,9 +8467,9 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 					mode_lib->ms.support.DSCSlicesODMModeSupported = ((mode_lib->ms.support.NumberOfDSCSlices[k] % 4) == 0);
 #if defined(__DML_VBA_DEBUG__)
 				if (!mode_lib->ms.support.DSCSlicesODMModeSupported) {
-					dml2_printf("DML::%s: k=%d Invalid dsc num_slices and ODM mode setting\n", __func__, k);
-					dml2_printf("DML::%s: k=%d num_slices = %d\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.dsc.overrides.num_slices);
-					dml2_printf("DML::%s: k=%d ODMMode = %d\n", __func__, k, mode_lib->ms.ODMMode[k]);
+					DML_LOG_VERBOSE("DML::%s: k=%d Invalid dsc num_slices and ODM mode setting\n", __func__, k);
+					DML_LOG_VERBOSE("DML::%s: k=%d num_slices = %d\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.dsc.overrides.num_slices);
+					DML_LOG_VERBOSE("DML::%s: k=%d ODMMode = %d\n", __func__, k, mode_lib->ms.ODMMode[k]);
 				}
 #endif
 			} else {
@@ -7958,7 +8514,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			mode_lib->ms.MPCCombine[k] = false;
 			mode_lib->ms.NoOfDPP[k] = 1;
 			if (!mode_lib->ms.SingleDPPViewportSizeSupportPerSurface[k]) {
-				dml2_printf("WARNING: DML::%s: MPCC is override to disable but viewport is too large to be supported with single pipe!\n", __func__);
+				DML_LOG_VERBOSE("WARNING: DML::%s: MPCC is override to disable but viewport is too large to be supported with single pipe!\n", __func__);
 			}
 		} else {
 			if ((mode_lib->ms.MinDPPCLKUsingSingleDPP[k] > mode_lib->ms.max_dppclk_freq_mhz) || !mode_lib->ms.SingleDPPViewportSizeSupportPerSurface[k]) {
@@ -7968,7 +8524,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			}
 		}
 #if defined(__DML_VBA_DEBUG__)
-		dml2_printf("DML::%s: k=%d, NoOfDPP = %d\n", __func__, k, mode_lib->ms.NoOfDPP[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, NoOfDPP = %d\n", __func__, k, mode_lib->ms.NoOfDPP[k]);
 #endif
 	}
 
@@ -8138,7 +8694,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 				display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].output.audio_sample_rate,
 				display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].output.audio_sample_layout);
 
-			if (mode_lib->ms.RequiredDTBCLK[k] > ((double)min_clk_table->max_clocks_khz.dtbclk / 1000)) {
+			if (mode_lib->ms.RequiredDTBCLK[k] > ((double)min_clk_table->max_ss_clocks_khz.dtbclk / 1000)) {
 				mode_lib->ms.support.DTBCLKRequiredMoreThanSupported = true;
 			}
 		} else {
@@ -8167,7 +8723,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 				s->DSCFormatFactor = 1;
 			}
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, RequiresDSC = %u\n", __func__, k, mode_lib->ms.RequiresDSC[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, RequiresDSC = %u\n", __func__, k, mode_lib->ms.RequiresDSC[k]);
 #endif
 			if (mode_lib->ms.RequiresDSC[k] == true) {
 				s->PixelClockBackEndFactor = 3.0;
@@ -8185,10 +8741,10 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 				}
 
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k=%u, PixelClockBackEnd = %f\n", __func__, k, s->PixelClockBackEnd[k]);
-				dml2_printf("DML::%s: k=%u, required_dscclk_freq_mhz = %f\n", __func__, k, mode_lib->ms.required_dscclk_freq_mhz[k]);
-				dml2_printf("DML::%s: k=%u, DSCFormatFactor = %u\n", __func__, k, s->DSCFormatFactor);
-				dml2_printf("DML::%s: k=%u, DSCCLKRequiredMoreThanSupported = %u\n", __func__, k, mode_lib->ms.support.DSCCLKRequiredMoreThanSupported);
+				DML_LOG_VERBOSE("DML::%s: k=%u, PixelClockBackEnd = %f\n", __func__, k, s->PixelClockBackEnd[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, required_dscclk_freq_mhz = %f\n", __func__, k, mode_lib->ms.required_dscclk_freq_mhz[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, DSCFormatFactor = %u\n", __func__, k, s->DSCFormatFactor);
+				DML_LOG_VERBOSE("DML::%s: k=%u, DSCCLKRequiredMoreThanSupported = %u\n", __func__, k, mode_lib->ms.support.DSCCLKRequiredMoreThanSupported);
 #endif
 			}
 		}
@@ -8423,13 +8979,13 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			mode_lib->ms.support.DCCMetaBufferSizeNotExceeded = false;
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, PTEBufferSizeNotExceeded = %u\n", __func__, k, mode_lib->ms.PTEBufferSizeNotExceeded[k]);
-		dml2_printf("DML::%s: k=%u, DCCMetaBufferSizeNotExceeded = %u\n", __func__, k, mode_lib->ms.DCCMetaBufferSizeNotExceeded[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, PTEBufferSizeNotExceeded = %u\n", __func__, k, mode_lib->ms.PTEBufferSizeNotExceeded[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, DCCMetaBufferSizeNotExceeded = %u\n", __func__, k, mode_lib->ms.DCCMetaBufferSizeNotExceeded[k]);
 #endif
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: PTEBufferSizeNotExceeded = %u\n", __func__, mode_lib->ms.support.PTEBufferSizeNotExceeded);
-	dml2_printf("DML::%s: DCCMetaBufferSizeNotExceeded = %u\n", __func__, mode_lib->ms.support.DCCMetaBufferSizeNotExceeded);
+	DML_LOG_VERBOSE("DML::%s: PTEBufferSizeNotExceeded = %u\n", __func__, mode_lib->ms.support.PTEBufferSizeNotExceeded);
+	DML_LOG_VERBOSE("DML::%s: DCCMetaBufferSizeNotExceeded = %u\n", __func__, mode_lib->ms.support.DCCMetaBufferSizeNotExceeded);
 #endif
 
 	/* VActive bytes to fetch for UCLK P-State */
@@ -8502,7 +9058,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 
 	for (k = 0; k < mode_lib->ms.num_active_planes; ++k) {
 		double line_time_us = (double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
-		bool cursor_not_enough_urgent_latency_hiding = 0;
+		bool cursor_not_enough_urgent_latency_hiding = false;
 
 		if (display_cfg->plane_descriptors[k].cursor.num_cursors > 0) {
 			calculate_cursor_req_attributes(
@@ -8531,9 +9087,9 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 		mode_lib->ms.UrgentBurstFactorCursorPre[k] = mode_lib->ms.UrgentBurstFactorCursor[k];
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%d, Calling CalculateUrgentBurstFactor\n", __func__, k);
-		dml2_printf("DML::%s: k=%d, VRatio=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
-		dml2_printf("DML::%s: k=%d, VRatioChroma=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%d, Calling CalculateUrgentBurstFactor\n", __func__, k);
+		DML_LOG_VERBOSE("DML::%s: k=%d, VRatio=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%d, VRatioChroma=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio);
 #endif
 
 		CalculateUrgentBurstFactor(
@@ -8605,7 +9161,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: k=%u, MaximumVStartup = %u\n", __func__, k, s->MaximumVStartup[k]);
+	DML_LOG_VERBOSE("DML::%s: k=%u, MaximumVStartup = %u\n", __func__, k, s->MaximumVStartup[k]);
 #endif
 
 	/* Immediate Flip and MALL parameters */
@@ -8654,16 +9210,15 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 		(s->SubViewportMALLPStateMethod && s->FullFrameMALLPStateMethod) || s->SubViewportMALLRefreshGreaterThan120Hz;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: SubViewportMALLPStateMethod = %u\n", __func__, s->SubViewportMALLPStateMethod);
-	dml2_printf("DML::%s: PhantomPipeMALLPStateMethod = %u\n", __func__, s->PhantomPipeMALLPStateMethod);
-	dml2_printf("DML::%s: FullFrameMALLPStateMethod = %u\n", __func__, s->FullFrameMALLPStateMethod);
-	dml2_printf("DML::%s: SubViewportMALLRefreshGreaterThan120Hz = %u\n", __func__, s->SubViewportMALLRefreshGreaterThan120Hz);
-	dml2_printf("DML::%s: InvalidCombinationOfMALLUseForPState = %u\n", __func__, mode_lib->ms.support.InvalidCombinationOfMALLUseForPState);
-	dml2_printf("DML::%s: in_out_params->min_clk_index = %u\n", __func__, in_out_params->min_clk_index);
-	dml2_printf("DML::%s: mode_lib->ms.DCFCLK = %f\n", __func__, mode_lib->ms.DCFCLK);
-	dml2_printf("DML::%s: mode_lib->ms.FabricClock = %f\n", __func__, mode_lib->ms.FabricClock);
-	dml2_printf("DML::%s: mode_lib->ms.uclk_freq_mhz = %f\n", __func__, mode_lib->ms.uclk_freq_mhz);
-	dml2_printf("DML::%s: urgent latency tolarance = %f\n", __func__, ((mode_lib->ip.rob_buffer_size_kbytes - mode_lib->ip.pixel_chunk_size_kbytes) * 1024 / (mode_lib->ms.DCFCLK * mode_lib->soc.return_bus_width_bytes)));
+	DML_LOG_VERBOSE("DML::%s: SubViewportMALLPStateMethod = %u\n", __func__, s->SubViewportMALLPStateMethod);
+	DML_LOG_VERBOSE("DML::%s: PhantomPipeMALLPStateMethod = %u\n", __func__, s->PhantomPipeMALLPStateMethod);
+	DML_LOG_VERBOSE("DML::%s: FullFrameMALLPStateMethod = %u\n", __func__, s->FullFrameMALLPStateMethod);
+	DML_LOG_VERBOSE("DML::%s: SubViewportMALLRefreshGreaterThan120Hz = %u\n", __func__, s->SubViewportMALLRefreshGreaterThan120Hz);
+	DML_LOG_VERBOSE("DML::%s: InvalidCombinationOfMALLUseForPState = %u\n", __func__, mode_lib->ms.support.InvalidCombinationOfMALLUseForPState);
+	DML_LOG_VERBOSE("DML::%s: in_out_params->min_clk_index = %u\n", __func__, in_out_params->min_clk_index);
+	DML_LOG_VERBOSE("DML::%s: mode_lib->ms.DCFCLK = %f\n", __func__, mode_lib->ms.DCFCLK);
+	DML_LOG_VERBOSE("DML::%s: mode_lib->ms.FabricClock = %f\n", __func__, mode_lib->ms.FabricClock);
+	DML_LOG_VERBOSE("DML::%s: mode_lib->ms.uclk_freq_mhz = %f\n", __func__, mode_lib->ms.uclk_freq_mhz);
 #endif
 
 	mode_lib->ms.support.OutstandingRequestsSupport = true;
@@ -8703,10 +9258,10 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			}
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: avg_urgent_latency_us = %f\n", __func__, mode_lib->ms.support.avg_urgent_latency_us);
-			dml2_printf("DML::%s: avg_non_urgent_latency_us = %f\n", __func__, mode_lib->ms.support.avg_non_urgent_latency_us);
-			dml2_printf("DML::%s: k=%d, request_size_bytes_luma = %d\n", __func__, k, mode_lib->ms.support.request_size_bytes_luma[k]);
-			dml2_printf("DML::%s: k=%d, outstanding_latency_us = %f (luma)\n", __func__, k, outstanding_latency_us);
+			DML_LOG_VERBOSE("DML::%s: avg_urgent_latency_us = %f\n", __func__, mode_lib->ms.support.avg_urgent_latency_us);
+			DML_LOG_VERBOSE("DML::%s: avg_non_urgent_latency_us = %f\n", __func__, mode_lib->ms.support.avg_non_urgent_latency_us);
+			DML_LOG_VERBOSE("DML::%s: k=%d, request_size_bytes_luma = %d\n", __func__, k, mode_lib->ms.support.request_size_bytes_luma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, outstanding_latency_us = %f (luma)\n", __func__, k, outstanding_latency_us);
 #endif
 		}
 
@@ -8722,8 +9277,8 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 				mode_lib->ms.support.OutstandingRequestsUrgencyAvoidance = false;
 			}
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%d, request_size_bytes_chroma = %d\n", __func__, k, mode_lib->ms.support.request_size_bytes_chroma[k]);
-			dml2_printf("DML::%s: k=%d, outstanding_latency_us = %f (chroma)\n", __func__, k, outstanding_latency_us);
+			DML_LOG_VERBOSE("DML::%s: k=%d, request_size_bytes_chroma = %d\n", __func__, k, mode_lib->ms.support.request_size_bytes_chroma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d, outstanding_latency_us = %f (chroma)\n", __func__, k, outstanding_latency_us);
 #endif
 		}
 	}
@@ -8869,7 +9424,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 	for (k = 0; k < mode_lib->ms.num_active_planes; ++k) {
 		if (mode_lib->ms.NotEnoughUrgentLatencyHiding[k]) {
 			mode_lib->ms.support.EnoughUrgentLatencyHidingSupport = false;
-			dml2_printf("DML::%s: k=%u NotEnoughUrgentLatencyHiding set\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: k=%u NotEnoughUrgentLatencyHiding set\n", __func__, k);
 
 		}
 	}
@@ -8878,639 +9433,13 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			if (!mode_lib->ms.support.avg_bandwidth_support_ok[m][n] && (m == dml2_core_internal_soc_state_sys_active || mode_lib->soc.mall_allocated_for_dcn_mbytes > 0)) {
 				mode_lib->ms.support.AvgBandwidthSupport = false;
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: avg_bandwidth_support_ok[%s][%s] not ok\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n));
+				DML_LOG_VERBOSE("DML::%s: avg_bandwidth_support_ok[%s][%s] not ok\n", __func__, dml2_core_internal_soc_state_type_str(m), dml2_core_internal_bw_type_str(n));
 #endif
 			}
 		}
 	}
 
-	/* Prefetch Check */
-	{
-		mode_lib->ms.TimeCalc = 24 / mode_lib->ms.dcfclk_deepsleep;
-
-		calculate_hostvm_inefficiency_factor(
-				&s->HostVMInefficiencyFactor,
-				&s->HostVMInefficiencyFactorPrefetch,
-
-				display_cfg->gpuvm_enable,
-				display_cfg->hostvm_enable,
-				mode_lib->ip.remote_iommu_outstanding_translations,
-				mode_lib->soc.max_outstanding_reqs,
-				mode_lib->ms.support.urg_bandwidth_available_pixel_and_vm[dml2_core_internal_soc_state_sys_active],
-				mode_lib->ms.support.urg_bandwidth_available_vm_only[dml2_core_internal_soc_state_sys_active]);
-
-		mode_lib->ms.Total3dlutActive = 0;
-		for (k = 0; k <= mode_lib->ms.num_active_planes - 1; k++) {
-			if (display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut)
-				mode_lib->ms.Total3dlutActive = mode_lib->ms.Total3dlutActive + 1;
-
-			// Calculate tdlut schedule related terms
-			calculate_tdlut_setting_params->dispclk_mhz = mode_lib->ms.RequiredDISPCLK;
-			calculate_tdlut_setting_params->setup_for_tdlut = display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut;
-			calculate_tdlut_setting_params->tdlut_width_mode = display_cfg->plane_descriptors[k].tdlut.tdlut_width_mode;
-			calculate_tdlut_setting_params->tdlut_addressing_mode = display_cfg->plane_descriptors[k].tdlut.tdlut_addressing_mode;
-			calculate_tdlut_setting_params->cursor_buffer_size = mode_lib->ip.cursor_buffer_size;
-			calculate_tdlut_setting_params->gpuvm_enable = display_cfg->gpuvm_enable;
-			calculate_tdlut_setting_params->gpuvm_page_size_kbytes = display_cfg->plane_descriptors[k].overrides.gpuvm_min_page_size_kbytes;
-			calculate_tdlut_setting_params->tdlut_mpc_width_flag = display_cfg->plane_descriptors[k].tdlut.tdlut_mpc_width_flag;
-			calculate_tdlut_setting_params->is_gfx11 = dml_get_gfx_version(display_cfg->plane_descriptors[k].surface.tiling);
-
-			// output
-			calculate_tdlut_setting_params->tdlut_pte_bytes_per_frame = &s->tdlut_pte_bytes_per_frame[k];
-			calculate_tdlut_setting_params->tdlut_bytes_per_frame = &s->tdlut_bytes_per_frame[k];
-			calculate_tdlut_setting_params->tdlut_groups_per_2row_ub = &s->tdlut_groups_per_2row_ub[k];
-			calculate_tdlut_setting_params->tdlut_opt_time = &s->tdlut_opt_time[k];
-			calculate_tdlut_setting_params->tdlut_drain_time = &s->tdlut_drain_time[k];
-			calculate_tdlut_setting_params->tdlut_bytes_to_deliver = &s->tdlut_bytes_to_deliver[k];
-			calculate_tdlut_setting_params->tdlut_bytes_per_group = &s->tdlut_bytes_per_group[k];
-
-			calculate_tdlut_setting(&mode_lib->scratch, calculate_tdlut_setting_params);
-		}
-
-		min_return_bw_for_latency = mode_lib->ms.support.urg_bandwidth_available_min_latency[dml2_core_internal_soc_state_sys_active];
-
-		if (mode_lib->soc.qos_parameters.qos_type == dml2_qos_param_type_dcn3)
-			s->ReorderingBytes = (unsigned int)(mode_lib->soc.clk_table.dram_config.channel_count * math_max3(mode_lib->soc.qos_parameters.qos_params.dcn32x.urgent_out_of_order_return_per_channel_pixel_only_bytes,
-											mode_lib->soc.qos_parameters.qos_params.dcn32x.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes,
-											mode_lib->soc.qos_parameters.qos_params.dcn32x.urgent_out_of_order_return_per_channel_vm_only_bytes));
-
-		CalculateExtraLatency(
-			display_cfg,
-			mode_lib->ip.rob_buffer_size_kbytes,
-			mode_lib->soc.qos_parameters.qos_params.dcn32x.loaded_round_trip_latency_fclk_cycles,
-			s->ReorderingBytes,
-			mode_lib->ms.DCFCLK,
-			mode_lib->ms.FabricClock,
-			mode_lib->ip.pixel_chunk_size_kbytes,
-			min_return_bw_for_latency,
-			mode_lib->ms.num_active_planes,
-			mode_lib->ms.NoOfDPP,
-			mode_lib->ms.dpte_group_bytes,
-			s->tdlut_bytes_per_group,
-			s->HostVMInefficiencyFactor,
-			s->HostVMInefficiencyFactorPrefetch,
-			mode_lib->soc.hostvm_min_page_size_kbytes,
-			mode_lib->soc.qos_parameters.qos_type,
-			!(display_cfg->overrides.max_outstanding_when_urgent_expected_disable),
-			mode_lib->soc.max_outstanding_reqs,
-			mode_lib->ms.support.request_size_bytes_luma,
-			mode_lib->ms.support.request_size_bytes_chroma,
-			mode_lib->ip.meta_chunk_size_kbytes,
-			mode_lib->ip.dchub_arb_to_ret_delay,
-			mode_lib->ms.TripToMemory,
-			mode_lib->ip.hostvm_mode,
-
-			// output
-			&mode_lib->ms.ExtraLatency,
-			&mode_lib->ms.ExtraLatency_sr,
-			&mode_lib->ms.ExtraLatencyPrefetch);
-
-		for (k = 0; k < mode_lib->ms.num_active_planes; k++)
-			s->impacted_dst_y_pre[k] = 0;
-
-		s->recalc_prefetch_schedule = 0;
-		s->recalc_prefetch_done = 0;
-		do {
-			mode_lib->ms.support.PrefetchSupported = true;
-
-			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-				s->line_times[k] = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
-				s->pixel_format[k] = display_cfg->plane_descriptors[k].pixel_format;
-
-				s->lb_source_lines_l[k] = get_num_lb_source_lines(mode_lib->ip.max_line_buffer_lines, mode_lib->ip.line_buffer_size_bits,
-																	mode_lib->ms.NoOfDPP[k],
-																	display_cfg->plane_descriptors[k].composition.viewport.plane0.width,
-																	display_cfg->plane_descriptors[k].composition.viewport.plane0.height,
-																	display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio,
-																	display_cfg->plane_descriptors[k].composition.rotation_angle);
-
-				s->lb_source_lines_c[k] = get_num_lb_source_lines(mode_lib->ip.max_line_buffer_lines, mode_lib->ip.line_buffer_size_bits,
-																	mode_lib->ms.NoOfDPP[k],
-																	display_cfg->plane_descriptors[k].composition.viewport.plane1.width,
-																	display_cfg->plane_descriptors[k].composition.viewport.plane1.height,
-																	display_cfg->plane_descriptors[k].composition.scaler_info.plane1.h_ratio,
-																	display_cfg->plane_descriptors[k].composition.rotation_angle);
-
-				struct dml2_core_internal_DmlPipe *myPipe = &s->myPipe;
-
-				mode_lib->ms.TWait[k] = CalculateTWait(
-					display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns,
-					mode_lib->ms.UrgLatency,
-					mode_lib->ms.TripToMemory,
-					!dml_is_phantom_pipe(&display_cfg->plane_descriptors[k]) && display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.drr_config.enabled ?
-					get_g6_temp_read_blackout_us(&mode_lib->soc, (unsigned int)(mode_lib->ms.uclk_freq_mhz * 1000), in_out_params->min_clk_index) : 0.0);
-
-				myPipe->Dppclk = mode_lib->ms.RequiredDPPCLK[k];
-				myPipe->Dispclk = mode_lib->ms.RequiredDISPCLK;
-				myPipe->PixelClock = ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
-				myPipe->DCFClkDeepSleep = mode_lib->ms.dcfclk_deepsleep;
-				myPipe->DPPPerSurface = mode_lib->ms.NoOfDPP[k];
-				myPipe->ScalerEnabled = display_cfg->plane_descriptors[k].composition.scaler_info.enabled;
-				myPipe->VRatio = display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio;
-				myPipe->VRatioChroma = display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio;
-				myPipe->VTaps = display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_taps;
-				myPipe->VTapsChroma = display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_taps;
-				myPipe->RotationAngle = display_cfg->plane_descriptors[k].composition.rotation_angle;
-				myPipe->mirrored = display_cfg->plane_descriptors[k].composition.mirrored;
-				myPipe->BlockWidth256BytesY = mode_lib->ms.Read256BlockWidthY[k];
-				myPipe->BlockHeight256BytesY = mode_lib->ms.Read256BlockHeightY[k];
-				myPipe->BlockWidth256BytesC = mode_lib->ms.Read256BlockWidthC[k];
-				myPipe->BlockHeight256BytesC = mode_lib->ms.Read256BlockHeightC[k];
-				myPipe->InterlaceEnable = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.interlaced;
-				myPipe->NumberOfCursors = display_cfg->plane_descriptors[k].cursor.num_cursors;
-				myPipe->VBlank = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_total - display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_active;
-				myPipe->HTotal = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total;
-				myPipe->HActive = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_active;
-				myPipe->DCCEnable = display_cfg->plane_descriptors[k].surface.dcc.enable;
-				myPipe->ODMMode = mode_lib->ms.ODMMode[k];
-				myPipe->SourcePixelFormat = display_cfg->plane_descriptors[k].pixel_format;
-				myPipe->BytePerPixelY = mode_lib->ms.BytePerPixelY[k];
-				myPipe->BytePerPixelC = mode_lib->ms.BytePerPixelC[k];
-				myPipe->ProgressiveToInterlaceUnitInOPP = mode_lib->ip.ptoi_supported;
-
-#ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: Calling CalculatePrefetchSchedule for k=%u\n", __func__, k);
-				dml2_printf("DML::%s: MaximumVStartup = %u\n", __func__, s->MaximumVStartup[k]);
-#endif
-				CalculatePrefetchSchedule_params->display_cfg = display_cfg;
-				CalculatePrefetchSchedule_params->HostVMInefficiencyFactor = s->HostVMInefficiencyFactorPrefetch;
-				CalculatePrefetchSchedule_params->myPipe = myPipe;
-				CalculatePrefetchSchedule_params->DSCDelay = mode_lib->ms.DSCDelay[k];
-				CalculatePrefetchSchedule_params->DPPCLKDelaySubtotalPlusCNVCFormater = mode_lib->ip.dppclk_delay_subtotal + mode_lib->ip.dppclk_delay_cnvc_formatter;
-				CalculatePrefetchSchedule_params->DPPCLKDelaySCL = mode_lib->ip.dppclk_delay_scl;
-				CalculatePrefetchSchedule_params->DPPCLKDelaySCLLBOnly = mode_lib->ip.dppclk_delay_scl_lb_only;
-				CalculatePrefetchSchedule_params->DPPCLKDelayCNVCCursor = mode_lib->ip.dppclk_delay_cnvc_cursor;
-				CalculatePrefetchSchedule_params->DISPCLKDelaySubtotal = mode_lib->ip.dispclk_delay_subtotal;
-				CalculatePrefetchSchedule_params->DPP_RECOUT_WIDTH = (unsigned int)(mode_lib->ms.SwathWidthY[k] / display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio);
-				CalculatePrefetchSchedule_params->OutputFormat = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].output.output_format;
-				CalculatePrefetchSchedule_params->MaxInterDCNTileRepeaters = mode_lib->ip.max_inter_dcn_tile_repeaters;
-				CalculatePrefetchSchedule_params->VStartup = s->MaximumVStartup[k];
-				CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->soc.hostvm_min_page_size_kbytes;
-				CalculatePrefetchSchedule_params->DynamicMetadataEnable = display_cfg->plane_descriptors[k].dynamic_meta_data.enable;
-				CalculatePrefetchSchedule_params->DynamicMetadataVMEnabled = mode_lib->ip.dynamic_metadata_vm_enabled;
-				CalculatePrefetchSchedule_params->DynamicMetadataLinesBeforeActiveRequired = display_cfg->plane_descriptors[k].dynamic_meta_data.lines_before_active_required;
-				CalculatePrefetchSchedule_params->DynamicMetadataTransmittedBytes = display_cfg->plane_descriptors[k].dynamic_meta_data.transmitted_bytes;
-				CalculatePrefetchSchedule_params->UrgentLatency = mode_lib->ms.UrgLatency;
-				CalculatePrefetchSchedule_params->ExtraLatencyPrefetch = mode_lib->ms.ExtraLatencyPrefetch;
-				CalculatePrefetchSchedule_params->TCalc = mode_lib->ms.TimeCalc;
-				CalculatePrefetchSchedule_params->vm_bytes = mode_lib->ms.vm_bytes[k];
-				CalculatePrefetchSchedule_params->PixelPTEBytesPerRow = mode_lib->ms.DPTEBytesPerRow[k];
-				CalculatePrefetchSchedule_params->PrefetchSourceLinesY = mode_lib->ms.PrefetchLinesY[k];
-				CalculatePrefetchSchedule_params->VInitPreFillY = mode_lib->ms.PrefillY[k];
-				CalculatePrefetchSchedule_params->MaxNumSwathY = mode_lib->ms.MaxNumSwathY[k];
-				CalculatePrefetchSchedule_params->PrefetchSourceLinesC = mode_lib->ms.PrefetchLinesC[k];
-				CalculatePrefetchSchedule_params->VInitPreFillC = mode_lib->ms.PrefillC[k];
-				CalculatePrefetchSchedule_params->MaxNumSwathC = mode_lib->ms.MaxNumSwathC[k];
-				CalculatePrefetchSchedule_params->swath_width_luma_ub = mode_lib->ms.swath_width_luma_ub[k];
-				CalculatePrefetchSchedule_params->swath_width_chroma_ub = mode_lib->ms.swath_width_chroma_ub[k];
-				CalculatePrefetchSchedule_params->SwathHeightY = mode_lib->ms.SwathHeightY[k];
-				CalculatePrefetchSchedule_params->SwathHeightC = mode_lib->ms.SwathHeightC[k];
-				CalculatePrefetchSchedule_params->TWait = mode_lib->ms.TWait[k];
-				CalculatePrefetchSchedule_params->Ttrip = mode_lib->ms.TripToMemory;
-				CalculatePrefetchSchedule_params->Turg = mode_lib->ms.UrgLatency;
-				CalculatePrefetchSchedule_params->setup_for_tdlut = display_cfg->plane_descriptors[k].tdlut.setup_for_tdlut;
-				CalculatePrefetchSchedule_params->tdlut_pte_bytes_per_frame = s->tdlut_pte_bytes_per_frame[k];
-				CalculatePrefetchSchedule_params->tdlut_bytes_per_frame = s->tdlut_bytes_per_frame[k];
-				CalculatePrefetchSchedule_params->tdlut_opt_time = s->tdlut_opt_time[k];
-				CalculatePrefetchSchedule_params->tdlut_drain_time = s->tdlut_drain_time[k];
-				CalculatePrefetchSchedule_params->num_cursors = (display_cfg->plane_descriptors[k].cursor.cursor_width > 0);
-				CalculatePrefetchSchedule_params->cursor_bytes_per_chunk = s->cursor_bytes_per_chunk[k];
-				CalculatePrefetchSchedule_params->cursor_bytes_per_line = s->cursor_bytes_per_line[k];
-				CalculatePrefetchSchedule_params->dcc_enable = display_cfg->plane_descriptors[k].surface.dcc.enable;
-				CalculatePrefetchSchedule_params->mrq_present = mode_lib->ip.dcn_mrq_present;
-				CalculatePrefetchSchedule_params->meta_row_bytes = mode_lib->ms.meta_row_bytes[k];
-				CalculatePrefetchSchedule_params->mall_prefetch_sdp_overhead_factor = mode_lib->ms.mall_prefetch_sdp_overhead_factor[k];
-				CalculatePrefetchSchedule_params->impacted_dst_y_pre = s->impacted_dst_y_pre[k];
-				CalculatePrefetchSchedule_params->vactive_sw_bw_l = mode_lib->ms.vactive_sw_bw_l[k];
-				CalculatePrefetchSchedule_params->vactive_sw_bw_c = mode_lib->ms.vactive_sw_bw_c[k];
-
-				// output
-				CalculatePrefetchSchedule_params->DSTXAfterScaler = &s->DSTXAfterScaler[k];
-				CalculatePrefetchSchedule_params->DSTYAfterScaler = &s->DSTYAfterScaler[k];
-				CalculatePrefetchSchedule_params->dst_y_prefetch = &mode_lib->ms.dst_y_prefetch[k];
-				CalculatePrefetchSchedule_params->dst_y_per_vm_vblank = &mode_lib->ms.LinesForVM[k];
-				CalculatePrefetchSchedule_params->dst_y_per_row_vblank = &mode_lib->ms.LinesForDPTERow[k];
-				CalculatePrefetchSchedule_params->VRatioPrefetchY = &mode_lib->ms.VRatioPreY[k];
-				CalculatePrefetchSchedule_params->VRatioPrefetchC = &mode_lib->ms.VRatioPreC[k];
-				CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWLuma = &mode_lib->ms.RequiredPrefetchPixelDataBWLuma[k]; // prefetch_sw_bw_l
-				CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWChroma = &mode_lib->ms.RequiredPrefetchPixelDataBWChroma[k]; // prefetch_sw_bw_c
-				CalculatePrefetchSchedule_params->RequiredPrefetchBWOTO = &mode_lib->ms.RequiredPrefetchBWOTO[k];
-				CalculatePrefetchSchedule_params->NotEnoughTimeForDynamicMetadata = &mode_lib->ms.NoTimeForDynamicMetadata[k];
-				CalculatePrefetchSchedule_params->Tno_bw = &mode_lib->ms.Tno_bw[k];
-				CalculatePrefetchSchedule_params->Tno_bw_flip = &mode_lib->ms.Tno_bw_flip[k];
-				CalculatePrefetchSchedule_params->prefetch_vmrow_bw = &mode_lib->ms.prefetch_vmrow_bw[k];
-				CalculatePrefetchSchedule_params->Tdmdl_vm = &s->dummy_single[0];
-				CalculatePrefetchSchedule_params->Tdmdl = &s->dummy_single[1];
-				CalculatePrefetchSchedule_params->TSetup = &s->dummy_single[2];
-				CalculatePrefetchSchedule_params->Tvm_trips = &s->Tvm_trips[k];
-				CalculatePrefetchSchedule_params->Tr0_trips = &s->Tr0_trips[k];
-				CalculatePrefetchSchedule_params->Tvm_trips_flip = &s->Tvm_trips_flip[k];
-				CalculatePrefetchSchedule_params->Tr0_trips_flip = &s->Tr0_trips_flip[k];
-				CalculatePrefetchSchedule_params->Tvm_trips_flip_rounded = &s->Tvm_trips_flip_rounded[k];
-				CalculatePrefetchSchedule_params->Tr0_trips_flip_rounded = &s->Tr0_trips_flip_rounded[k];
-				CalculatePrefetchSchedule_params->VUpdateOffsetPix = &s->dummy_integer[0];
-				CalculatePrefetchSchedule_params->VUpdateWidthPix = &s->dummy_integer[1];
-				CalculatePrefetchSchedule_params->VReadyOffsetPix = &s->dummy_integer[2];
-				CalculatePrefetchSchedule_params->prefetch_cursor_bw = &mode_lib->ms.prefetch_cursor_bw[k];
-				CalculatePrefetchSchedule_params->prefetch_sw_bytes = &s->prefetch_sw_bytes[k];
-				CalculatePrefetchSchedule_params->Tpre_rounded = &s->Tpre_rounded[k];
-				CalculatePrefetchSchedule_params->Tpre_oto = &s->Tpre_oto[k];
-				CalculatePrefetchSchedule_params->prefetch_swath_time_us = &s->prefetch_swath_time_us[k];
-
-				mode_lib->ms.NoTimeForPrefetch[k] = CalculatePrefetchSchedule(&mode_lib->scratch, CalculatePrefetchSchedule_params);
-
-				mode_lib->ms.support.PrefetchSupported &= !mode_lib->ms.NoTimeForPrefetch[k];
-				dml2_printf("DML::%s: k=%d, dst_y_per_vm_vblank = %f\n", __func__, k, *CalculatePrefetchSchedule_params->dst_y_per_vm_vblank);
-				dml2_printf("DML::%s: k=%d, dst_y_per_row_vblank = %f\n", __func__, k, *CalculatePrefetchSchedule_params->dst_y_per_row_vblank);
-			} // for k num_planes
-
-			CalculateDCFCLKDeepSleepTdlut(
-				display_cfg,
-				mode_lib->ms.num_active_planes,
-				mode_lib->ms.BytePerPixelY,
-				mode_lib->ms.BytePerPixelC,
-				mode_lib->ms.SwathWidthY,
-				mode_lib->ms.SwathWidthC,
-				mode_lib->ms.NoOfDPP,
-				mode_lib->ms.PSCL_FACTOR,
-				mode_lib->ms.PSCL_FACTOR_CHROMA,
-				mode_lib->ms.RequiredDPPCLK,
-				mode_lib->ms.vactive_sw_bw_l,
-				mode_lib->ms.vactive_sw_bw_c,
-				mode_lib->soc.return_bus_width_bytes,
-				mode_lib->ms.RequiredDISPCLK,
-				s->tdlut_bytes_to_deliver,
-				s->prefetch_swath_time_us,
-
-				/* Output */
-				&mode_lib->ms.dcfclk_deepsleep);
-
-			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-				if (mode_lib->ms.dst_y_prefetch[k] < 2.0
-					|| mode_lib->ms.LinesForVM[k] >= 32.0
-					|| mode_lib->ms.LinesForDPTERow[k] >= 16.0
-					|| mode_lib->ms.NoTimeForPrefetch[k] == true
-					|| s->DSTYAfterScaler[k] > 8) {
-					mode_lib->ms.support.PrefetchSupported = false;
-					dml2_printf("DML::%s: k=%d, dst_y_prefetch=%f (should not be < 2)\n", __func__, k, mode_lib->ms.dst_y_prefetch[k]);
-					dml2_printf("DML::%s: k=%d, LinesForVM=%f (should not be >= 32)\n", __func__, k, mode_lib->ms.LinesForVM[k]);
-					dml2_printf("DML::%s: k=%d, LinesForDPTERow=%f (should not be >= 16)\n", __func__, k, mode_lib->ms.LinesForDPTERow[k]);
-					dml2_printf("DML::%s: k=%d, DSTYAfterScaler=%d (should be <= 8)\n", __func__, k, s->DSTYAfterScaler[k]);
-					dml2_printf("DML::%s: k=%d, NoTimeForPrefetch=%d\n", __func__, k, mode_lib->ms.NoTimeForPrefetch[k]);
-				}
-			}
-
-			mode_lib->ms.support.DynamicMetadataSupported = true;
-			for (k = 0; k < mode_lib->ms.num_active_planes; ++k) {
-				if (mode_lib->ms.NoTimeForDynamicMetadata[k] == true) {
-					mode_lib->ms.support.DynamicMetadataSupported = false;
-				}
-			}
-
-			mode_lib->ms.support.VRatioInPrefetchSupported = true;
-			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-				if (mode_lib->ms.VRatioPreY[k] > __DML2_CALCS_MAX_VRATIO_PRE__ ||
-					mode_lib->ms.VRatioPreC[k] > __DML2_CALCS_MAX_VRATIO_PRE__) {
-					mode_lib->ms.support.VRatioInPrefetchSupported = false;
-					dml2_printf("DML::%s: k=%d VRatioPreY = %f (should be <= %f)\n", __func__, k, mode_lib->ms.VRatioPreY[k], __DML2_CALCS_MAX_VRATIO_PRE__);
-					dml2_printf("DML::%s: k=%d VRatioPreC = %f (should be <= %f)\n", __func__, k, mode_lib->ms.VRatioPreC[k], __DML2_CALCS_MAX_VRATIO_PRE__);
-					dml2_printf("DML::%s: VRatioInPrefetchSupported = %u\n", __func__, mode_lib->ms.support.VRatioInPrefetchSupported);
-				}
-			}
-
-			mode_lib->ms.support.PrefetchSupported &= mode_lib->ms.support.VRatioInPrefetchSupported;
-
-			// By default, do not recalc prefetch schedule
-			s->recalc_prefetch_schedule = 0;
-
-			// Only do urg vs prefetch bandwidth check, flip schedule check, power saving feature support check IF the Prefetch Schedule Check is ok
-			if (mode_lib->ms.support.PrefetchSupported) {
-				for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-					// Calculate Urgent burst factor for prefetch
-#ifdef __DML_VBA_DEBUG__
-					dml2_printf("DML::%s: k=%d, Calling CalculateUrgentBurstFactor (for prefetch)\n", __func__, k);
-					dml2_printf("DML::%s: k=%d, VRatioPreY=%f\n", __func__, k, mode_lib->ms.VRatioPreY[k]);
-					dml2_printf("DML::%s: k=%d, VRatioPreC=%f\n", __func__, k, mode_lib->ms.VRatioPreC[k]);
-#endif
-					CalculateUrgentBurstFactor(
-						&display_cfg->plane_descriptors[k],
-						mode_lib->ms.swath_width_luma_ub[k],
-						mode_lib->ms.swath_width_chroma_ub[k],
-						mode_lib->ms.SwathHeightY[k],
-						mode_lib->ms.SwathHeightC[k],
-						s->line_times[k],
-						mode_lib->ms.UrgLatency,
-						mode_lib->ms.VRatioPreY[k],
-						mode_lib->ms.VRatioPreC[k],
-						mode_lib->ms.BytePerPixelInDETY[k],
-						mode_lib->ms.BytePerPixelInDETC[k],
-						mode_lib->ms.DETBufferSizeY[k],
-						mode_lib->ms.DETBufferSizeC[k],
-						/* Output */
-						&mode_lib->ms.UrgentBurstFactorLumaPre[k],
-						&mode_lib->ms.UrgentBurstFactorChromaPre[k],
-						&mode_lib->ms.NotEnoughUrgentLatencyHidingPre[k]);
-				}
-
-				// Calculate urgent bandwidth required, both urg and non urg peak bandwidth
-				// assume flip bw is 0 at this point
-				for (k = 0; k < mode_lib->ms.num_active_planes; k++)
-					mode_lib->ms.final_flip_bw[k] = 0;
-
-				calculate_peak_bandwidth_params->urg_vactive_bandwidth_required = mode_lib->ms.support.urg_vactive_bandwidth_required;
-				calculate_peak_bandwidth_params->urg_bandwidth_required = mode_lib->ms.support.urg_bandwidth_required;
-				calculate_peak_bandwidth_params->urg_bandwidth_required_qual = mode_lib->ms.support.urg_bandwidth_required_qual;
-				calculate_peak_bandwidth_params->non_urg_bandwidth_required = mode_lib->ms.support.non_urg_bandwidth_required;
-				calculate_peak_bandwidth_params->surface_avg_vactive_required_bw = mode_lib->ms.surface_avg_vactive_required_bw;
-				calculate_peak_bandwidth_params->surface_peak_required_bw = mode_lib->ms.surface_peak_required_bw;
-
-				calculate_peak_bandwidth_params->display_cfg = display_cfg;
-				calculate_peak_bandwidth_params->inc_flip_bw = 0;
-				calculate_peak_bandwidth_params->num_active_planes =  mode_lib->ms.num_active_planes;
-				calculate_peak_bandwidth_params->num_of_dpp = mode_lib->ms.NoOfDPP;
-				calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p0;
-				calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p1;
-				calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p0;
-				calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p1;
-				calculate_peak_bandwidth_params->mall_prefetch_sdp_overhead_factor = mode_lib->ms.mall_prefetch_sdp_overhead_factor;
-				calculate_peak_bandwidth_params->mall_prefetch_dram_overhead_factor = mode_lib->ms.mall_prefetch_dram_overhead_factor;
-
-				calculate_peak_bandwidth_params->surface_read_bandwidth_l = mode_lib->ms.vactive_sw_bw_l;
-				calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c;
-				calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma;
-				calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma;
-				calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO;
-				calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l;
-				calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c;
-				calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw;
-				calculate_peak_bandwidth_params->dpte_row_bw = mode_lib->ms.dpte_row_bw;
-				calculate_peak_bandwidth_params->meta_row_bw = mode_lib->ms.meta_row_bw;
-				calculate_peak_bandwidth_params->prefetch_cursor_bw = mode_lib->ms.prefetch_cursor_bw;
-				calculate_peak_bandwidth_params->prefetch_vmrow_bw = mode_lib->ms.prefetch_vmrow_bw;
-				calculate_peak_bandwidth_params->flip_bw = mode_lib->ms.final_flip_bw;
-				calculate_peak_bandwidth_params->urgent_burst_factor_l = mode_lib->ms.UrgentBurstFactorLuma;
-				calculate_peak_bandwidth_params->urgent_burst_factor_c = mode_lib->ms.UrgentBurstFactorChroma;
-				calculate_peak_bandwidth_params->urgent_burst_factor_cursor = mode_lib->ms.UrgentBurstFactorCursor;
-				calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_l = mode_lib->ms.UrgentBurstFactorLumaPre;
-				calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_c = mode_lib->ms.UrgentBurstFactorChromaPre;
-				calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_cursor = mode_lib->ms.UrgentBurstFactorCursorPre;
-
-				calculate_peak_bandwidth_required(
-						&mode_lib->scratch,
-						calculate_peak_bandwidth_params);
-
-				// Check urg peak bandwidth against available urg bw
-				// check at SDP and DRAM, for all soc states (SVP prefetch an Sys Active)
-				check_urgent_bandwidth_support(
-					&s->dummy_single[0], // double* frac_urg_bandwidth
-					&s->dummy_single[1], // double* frac_urg_bandwidth_mall
-					&mode_lib->ms.support.UrgVactiveBandwidthSupport,
-					&mode_lib->ms.support.PrefetchBandwidthSupported,
-
-					mode_lib->soc.mall_allocated_for_dcn_mbytes,
-					mode_lib->ms.support.non_urg_bandwidth_required,
-					mode_lib->ms.support.urg_vactive_bandwidth_required,
-					mode_lib->ms.support.urg_bandwidth_required,
-					mode_lib->ms.support.urg_bandwidth_available);
-
-				mode_lib->ms.support.PrefetchSupported &= mode_lib->ms.support.PrefetchBandwidthSupported;
-				dml2_printf("DML::%s: PrefetchBandwidthSupported=%0d\n", __func__, mode_lib->ms.support.PrefetchBandwidthSupported);
-
-				for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-					if (mode_lib->ms.NotEnoughUrgentLatencyHidingPre[k]) {
-						mode_lib->ms.support.PrefetchSupported = false;
-						dml2_printf("DML::%s: k=%d, NotEnoughUrgentLatencyHidingPre=%d\n", __func__, k, mode_lib->ms.NotEnoughUrgentLatencyHidingPre[k]);
-					}
-				}
-
-#ifdef DML_GLOBAL_PREFETCH_CHECK
-				if (mode_lib->ms.support.PrefetchSupported && mode_lib->ms.num_active_planes > 1 && s->recalc_prefetch_done == 0) {
-					CheckGlobalPrefetchAdmissibility_params->num_active_planes =  mode_lib->ms.num_active_planes;
-					CheckGlobalPrefetchAdmissibility_params->pixel_format = s->pixel_format;
-					CheckGlobalPrefetchAdmissibility_params->chunk_bytes_l = mode_lib->ip.pixel_chunk_size_kbytes * 1024;
-					CheckGlobalPrefetchAdmissibility_params->chunk_bytes_c = mode_lib->ip.pixel_chunk_size_kbytes * 1024;
-					CheckGlobalPrefetchAdmissibility_params->lb_source_lines_l = s->lb_source_lines_l;
-					CheckGlobalPrefetchAdmissibility_params->lb_source_lines_c = s->lb_source_lines_c;
-					CheckGlobalPrefetchAdmissibility_params->swath_height_l =  mode_lib->ms.SwathHeightY;
-					CheckGlobalPrefetchAdmissibility_params->swath_height_c =  mode_lib->ms.SwathHeightC;
-					CheckGlobalPrefetchAdmissibility_params->rob_buffer_size_kbytes = mode_lib->ip.rob_buffer_size_kbytes;
-					CheckGlobalPrefetchAdmissibility_params->compressed_buffer_size_kbytes = mode_lib->ms.CompressedBufferSizeInkByte;
-					CheckGlobalPrefetchAdmissibility_params->detile_buffer_size_bytes_l = mode_lib->ms.DETBufferSizeY;
-					CheckGlobalPrefetchAdmissibility_params->detile_buffer_size_bytes_c = mode_lib->ms.DETBufferSizeC;
-					CheckGlobalPrefetchAdmissibility_params->full_swath_bytes_l = s->full_swath_bytes_l;
-					CheckGlobalPrefetchAdmissibility_params->full_swath_bytes_c = s->full_swath_bytes_c;
-					CheckGlobalPrefetchAdmissibility_params->prefetch_sw_bytes = s->prefetch_sw_bytes;
-					CheckGlobalPrefetchAdmissibility_params->Tpre_rounded = s->Tpre_rounded;
-					CheckGlobalPrefetchAdmissibility_params->Tpre_oto = s->Tpre_oto;
-					CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps = mode_lib->ms.support.urg_bandwidth_required[dml2_core_internal_soc_state_sys_active][dml2_core_internal_bw_sdp];
-					CheckGlobalPrefetchAdmissibility_params->line_time = s->line_times;
-					CheckGlobalPrefetchAdmissibility_params->dst_y_prefetch = mode_lib->ms.dst_y_prefetch;
-					if (CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps < 10 * 1024)
-						CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps = 10 * 1024;
-
-					CheckGlobalPrefetchAdmissibility_params->estimated_dcfclk_mhz = (CheckGlobalPrefetchAdmissibility_params->estimated_urg_bandwidth_required_mbps / (double) mode_lib->soc.return_bus_width_bytes) /
-																					((double)mode_lib->soc.qos_parameters.derate_table.system_active_urgent.dcfclk_derate_percent / 100.0);
-
-					// if recalc_prefetch_schedule is set, recalculate the prefetch schedule with the new impacted_Tpre, prefetch should be possible
-					CheckGlobalPrefetchAdmissibility_params->recalc_prefetch_schedule = &s->recalc_prefetch_schedule;
-					CheckGlobalPrefetchAdmissibility_params->impacted_dst_y_pre = s->impacted_dst_y_pre;
-					mode_lib->ms.support.PrefetchSupported = CheckGlobalPrefetchAdmissibility(&mode_lib->scratch, CheckGlobalPrefetchAdmissibility_params);
-					s->recalc_prefetch_done = 1;
-					s->recalc_prefetch_schedule = 1;
-				}
-#endif
-			} // prefetch schedule ok, do urg bw and flip schedule
-		} while (s->recalc_prefetch_schedule);
-
-		// Flip Schedule
-		// Both prefetch schedule and BW okay
-		if (mode_lib->ms.support.PrefetchSupported == true) {
-			mode_lib->ms.BandwidthAvailableForImmediateFlip =
-				get_bandwidth_available_for_immediate_flip(
-					dml2_core_internal_soc_state_sys_active,
-					mode_lib->ms.support.urg_bandwidth_required_qual, // no flip
-					mode_lib->ms.support.urg_bandwidth_available);
-
-			mode_lib->ms.TotImmediateFlipBytes = 0;
-			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-				if (display_cfg->plane_descriptors[k].immediate_flip) {
-					s->per_pipe_flip_bytes[k] = get_pipe_flip_bytes(
-									s->HostVMInefficiencyFactor,
-									mode_lib->ms.vm_bytes[k],
-									mode_lib->ms.DPTEBytesPerRow[k],
-									mode_lib->ms.meta_row_bytes[k]);
-				} else {
-					s->per_pipe_flip_bytes[k] = 0;
-				}
-				mode_lib->ms.TotImmediateFlipBytes += s->per_pipe_flip_bytes[k] * mode_lib->ms.NoOfDPP[k];
-
-			}
-
-			for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
-				CalculateFlipSchedule(
-					&mode_lib->scratch,
-					display_cfg->plane_descriptors[k].immediate_flip,
-					1, // use_lb_flip_bw
-					s->HostVMInefficiencyFactor,
-					s->Tvm_trips_flip[k],
-					s->Tr0_trips_flip[k],
-					s->Tvm_trips_flip_rounded[k],
-					s->Tr0_trips_flip_rounded[k],
-					display_cfg->gpuvm_enable,
-					mode_lib->ms.vm_bytes[k],
-					mode_lib->ms.DPTEBytesPerRow[k],
-					mode_lib->ms.BandwidthAvailableForImmediateFlip,
-					mode_lib->ms.TotImmediateFlipBytes,
-					display_cfg->plane_descriptors[k].pixel_format,
-					(display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)),
-					display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio,
-					display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio,
-					mode_lib->ms.Tno_bw_flip[k],
-					mode_lib->ms.dpte_row_height[k],
-					mode_lib->ms.dpte_row_height_chroma[k],
-					mode_lib->ms.use_one_row_for_frame_flip[k],
-					mode_lib->ip.max_flip_time_us,
-					mode_lib->ip.max_flip_time_lines,
-					s->per_pipe_flip_bytes[k],
-					mode_lib->ms.meta_row_bytes[k],
-					s->meta_row_height_luma[k],
-					s->meta_row_height_chroma[k],
-					mode_lib->ip.dcn_mrq_present && display_cfg->plane_descriptors[k].surface.dcc.enable,
-
-					/* Output */
-					&mode_lib->ms.dst_y_per_vm_flip[k],
-					&mode_lib->ms.dst_y_per_row_flip[k],
-					&mode_lib->ms.final_flip_bw[k],
-					&mode_lib->ms.ImmediateFlipSupportedForPipe[k]);
-			}
-
-			calculate_peak_bandwidth_params->urg_vactive_bandwidth_required = s->dummy_bw;
-			calculate_peak_bandwidth_params->urg_bandwidth_required = mode_lib->ms.support.urg_bandwidth_required_flip;
-			calculate_peak_bandwidth_params->urg_bandwidth_required_qual = s->dummy_bw;
-			calculate_peak_bandwidth_params->non_urg_bandwidth_required = mode_lib->ms.support.non_urg_bandwidth_required_flip;
-			calculate_peak_bandwidth_params->surface_avg_vactive_required_bw = s->surface_dummy_bw;
-			calculate_peak_bandwidth_params->surface_peak_required_bw = mode_lib->ms.surface_peak_required_bw;
-
-			calculate_peak_bandwidth_params->display_cfg = display_cfg;
-			calculate_peak_bandwidth_params->inc_flip_bw = 1;
-			calculate_peak_bandwidth_params->num_active_planes = mode_lib->ms.num_active_planes;
-			calculate_peak_bandwidth_params->num_of_dpp = mode_lib->ms.NoOfDPP;
-			calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p0;
-			calculate_peak_bandwidth_params->dcc_dram_bw_nom_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_nom_overhead_factor_p1;
-			calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p0 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p0;
-			calculate_peak_bandwidth_params->dcc_dram_bw_pref_overhead_factor_p1 = mode_lib->ms.dcc_dram_bw_pref_overhead_factor_p1;
-			calculate_peak_bandwidth_params->mall_prefetch_sdp_overhead_factor = mode_lib->ms.mall_prefetch_sdp_overhead_factor;
-			calculate_peak_bandwidth_params->mall_prefetch_dram_overhead_factor = mode_lib->ms.mall_prefetch_dram_overhead_factor;
-
-			calculate_peak_bandwidth_params->surface_read_bandwidth_l = mode_lib->ms.vactive_sw_bw_l;
-			calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c;
-			calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma;
-			calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma;
-			calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO;
-			calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l;
-			calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c;
-			calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw;
-			calculate_peak_bandwidth_params->dpte_row_bw = mode_lib->ms.dpte_row_bw;
-			calculate_peak_bandwidth_params->meta_row_bw = mode_lib->ms.meta_row_bw;
-			calculate_peak_bandwidth_params->prefetch_cursor_bw = mode_lib->ms.prefetch_cursor_bw;
-			calculate_peak_bandwidth_params->prefetch_vmrow_bw = mode_lib->ms.prefetch_vmrow_bw;
-			calculate_peak_bandwidth_params->flip_bw = mode_lib->ms.final_flip_bw;
-			calculate_peak_bandwidth_params->urgent_burst_factor_l = mode_lib->ms.UrgentBurstFactorLuma;
-			calculate_peak_bandwidth_params->urgent_burst_factor_c = mode_lib->ms.UrgentBurstFactorChroma;
-			calculate_peak_bandwidth_params->urgent_burst_factor_cursor = mode_lib->ms.UrgentBurstFactorCursor;
-			calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_l = mode_lib->ms.UrgentBurstFactorLumaPre;
-			calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_c = mode_lib->ms.UrgentBurstFactorChromaPre;
-			calculate_peak_bandwidth_params->urgent_burst_factor_prefetch_cursor = mode_lib->ms.UrgentBurstFactorCursorPre;
-
-			calculate_peak_bandwidth_required(
-					&mode_lib->scratch,
-					calculate_peak_bandwidth_params);
-
-			calculate_immediate_flip_bandwidth_support(
-				&s->dummy_single[0], // double* frac_urg_bandwidth_flip
-				&mode_lib->ms.support.ImmediateFlipSupport,
-
-				dml2_core_internal_soc_state_sys_active,
-				mode_lib->ms.support.urg_bandwidth_required_flip,
-				mode_lib->ms.support.non_urg_bandwidth_required_flip,
-				mode_lib->ms.support.urg_bandwidth_available);
-
-			for (k = 0; k <= mode_lib->ms.num_active_planes - 1; k++) {
-				if (display_cfg->plane_descriptors[k].immediate_flip == true && mode_lib->ms.ImmediateFlipSupportedForPipe[k] == false)
-					mode_lib->ms.support.ImmediateFlipSupport = false;
-			}
-
-		} else { // if prefetch not support, assume iflip is not supported too
-			mode_lib->ms.support.ImmediateFlipSupport = false;
-		}
-
-		s->mSOCParameters.UrgentLatency = mode_lib->ms.UrgLatency;
-		s->mSOCParameters.ExtraLatency = mode_lib->ms.ExtraLatency;
-		s->mSOCParameters.ExtraLatency_sr = mode_lib->ms.ExtraLatency_sr;
-		s->mSOCParameters.WritebackLatency = mode_lib->soc.qos_parameters.writeback.base_latency_us;
-		s->mSOCParameters.DRAMClockChangeLatency = mode_lib->soc.power_management_parameters.dram_clk_change_blackout_us;
-		s->mSOCParameters.FCLKChangeLatency = mode_lib->soc.power_management_parameters.fclk_change_blackout_us;
-		s->mSOCParameters.SRExitTime = mode_lib->soc.power_management_parameters.stutter_exit_latency_us;
-		s->mSOCParameters.SREnterPlusExitTime = mode_lib->soc.power_management_parameters.stutter_enter_plus_exit_latency_us;
-		s->mSOCParameters.SRExitZ8Time = mode_lib->soc.power_management_parameters.z8_stutter_exit_latency_us;
-		s->mSOCParameters.SREnterPlusExitZ8Time = mode_lib->soc.power_management_parameters.z8_stutter_enter_plus_exit_latency_us;
-		s->mSOCParameters.USRRetrainingLatency = 0;
-		s->mSOCParameters.SMNLatency = 0;
-		s->mSOCParameters.g6_temp_read_blackout_us = get_g6_temp_read_blackout_us(&mode_lib->soc, (unsigned int)(mode_lib->ms.uclk_freq_mhz * 1000), in_out_params->min_clk_index);
-		s->mSOCParameters.max_urgent_latency_us = get_max_urgent_latency_us(&mode_lib->soc.qos_parameters.qos_params.dcn4x, mode_lib->ms.uclk_freq_mhz, mode_lib->ms.FabricClock, in_out_params->min_clk_index);
-		s->mSOCParameters.df_response_time_us = mode_lib->soc.qos_parameters.qos_params.dcn4x.df_qos_response_time_fclk_cycles / mode_lib->ms.FabricClock;
-		s->mSOCParameters.qos_type = mode_lib->soc.qos_parameters.qos_type;
-
-		CalculateWatermarks_params->display_cfg = display_cfg;
-		CalculateWatermarks_params->USRRetrainingRequired = false;
-		CalculateWatermarks_params->NumberOfActiveSurfaces = mode_lib->ms.num_active_planes;
-		CalculateWatermarks_params->MaxLineBufferLines = mode_lib->ip.max_line_buffer_lines;
-		CalculateWatermarks_params->LineBufferSize = mode_lib->ip.line_buffer_size_bits;
-		CalculateWatermarks_params->WritebackInterfaceBufferSize = mode_lib->ip.writeback_interface_buffer_size_kbytes;
-		CalculateWatermarks_params->DCFCLK = mode_lib->ms.DCFCLK;
-		CalculateWatermarks_params->SynchronizeTimings = display_cfg->overrides.synchronize_timings;
-		CalculateWatermarks_params->SynchronizeDRRDisplaysForUCLKPStateChange = display_cfg->overrides.synchronize_ddr_displays_for_uclk_pstate_change;
-		CalculateWatermarks_params->dpte_group_bytes = mode_lib->ms.dpte_group_bytes;
-		CalculateWatermarks_params->mmSOCParameters = s->mSOCParameters;
-		CalculateWatermarks_params->WritebackChunkSize = mode_lib->ip.writeback_chunk_size_kbytes;
-		CalculateWatermarks_params->SOCCLK = mode_lib->ms.SOCCLK;
-		CalculateWatermarks_params->DCFClkDeepSleep = mode_lib->ms.dcfclk_deepsleep;
-		CalculateWatermarks_params->DETBufferSizeY = mode_lib->ms.DETBufferSizeY;
-		CalculateWatermarks_params->DETBufferSizeC = mode_lib->ms.DETBufferSizeC;
-		CalculateWatermarks_params->SwathHeightY = mode_lib->ms.SwathHeightY;
-		CalculateWatermarks_params->SwathHeightC = mode_lib->ms.SwathHeightC;
-		CalculateWatermarks_params->SwathWidthY = mode_lib->ms.SwathWidthY;
-		CalculateWatermarks_params->SwathWidthC = mode_lib->ms.SwathWidthC;
-		CalculateWatermarks_params->DPPPerSurface = mode_lib->ms.NoOfDPP;
-		CalculateWatermarks_params->BytePerPixelDETY = mode_lib->ms.BytePerPixelInDETY;
-		CalculateWatermarks_params->BytePerPixelDETC = mode_lib->ms.BytePerPixelInDETC;
-		CalculateWatermarks_params->DSTXAfterScaler = s->DSTXAfterScaler;
-		CalculateWatermarks_params->DSTYAfterScaler = s->DSTYAfterScaler;
-		CalculateWatermarks_params->UnboundedRequestEnabled = mode_lib->ms.UnboundedRequestEnabled;
-		CalculateWatermarks_params->CompressedBufferSizeInkByte = mode_lib->ms.CompressedBufferSizeInkByte;
-		CalculateWatermarks_params->meta_row_height_l = s->meta_row_height_luma;
-		CalculateWatermarks_params->meta_row_height_c = s->meta_row_height_chroma;
-
-		// Output
-		CalculateWatermarks_params->Watermark = &mode_lib->ms.support.watermarks; // Watermarks *Watermark
-		CalculateWatermarks_params->DRAMClockChangeSupport = mode_lib->ms.support.DRAMClockChangeSupport;
-		CalculateWatermarks_params->global_dram_clock_change_supported = &mode_lib->ms.support.global_dram_clock_change_supported;
-		CalculateWatermarks_params->MaxActiveDRAMClockChangeLatencySupported = &s->dummy_single_array[0]; // double *MaxActiveDRAMClockChangeLatencySupported[]
-		CalculateWatermarks_params->SubViewportLinesNeededInMALL = mode_lib->ms.SubViewportLinesNeededInMALL; // unsigned int SubViewportLinesNeededInMALL[]
-		CalculateWatermarks_params->FCLKChangeSupport = mode_lib->ms.support.FCLKChangeSupport;
-		CalculateWatermarks_params->global_fclk_change_supported = &mode_lib->ms.support.global_fclk_change_supported;
-		CalculateWatermarks_params->MaxActiveFCLKChangeLatencySupported = &s->dummy_single[0]; // double *MaxActiveFCLKChangeLatencySupported
-		CalculateWatermarks_params->USRRetrainingSupport = &mode_lib->ms.support.USRRetrainingSupport;
-		CalculateWatermarks_params->g6_temp_read_support = &mode_lib->ms.support.g6_temp_read_support;
-		CalculateWatermarks_params->VActiveLatencyHidingMargin = mode_lib->ms.VActiveLatencyHidingMargin;
-		CalculateWatermarks_params->VActiveLatencyHidingUs = mode_lib->ms.VActiveLatencyHidingUs;
-
-		CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(&mode_lib->scratch, CalculateWatermarks_params);
-
-		calculate_pstate_keepout_dst_lines(display_cfg, &mode_lib->ms.support.watermarks, s->dummy_integer_array[0]);
-	}
-	dml2_printf("DML::%s: Done prefetch calculation\n", __func__);
-	// End of Prefetch Check
+	dml_core_ms_prefetch_check(mode_lib, display_cfg);
 
 	mode_lib->ms.support.max_urgent_latency_us = s->mSOCParameters.max_urgent_latency_us;
 
@@ -9546,8 +9475,8 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			mode_lib->ms.dram_change_vactive_det_fill_delay_us);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: max_urgent_latency_us = %f\n", __func__, s->mSOCParameters.max_urgent_latency_us);
-	dml2_printf("DML::%s: ROBSupport = %u\n", __func__, mode_lib->ms.support.ROBSupport);
+	DML_LOG_VERBOSE("DML::%s: max_urgent_latency_us = %f\n", __func__, s->mSOCParameters.max_urgent_latency_us);
+	DML_LOG_VERBOSE("DML::%s: ROBSupport = %u\n", __func__, mode_lib->ms.support.ROBSupport);
 #endif
 
 	/*Mode Support, Voltage State and SOC Configuration*/
@@ -9597,17 +9526,17 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 			&& !mode_lib->ms.support.ExceededMALLSize
 			&& mode_lib->ms.support.g6_temp_read_support
 			&& ((!display_cfg->hostvm_enable && !s->ImmediateFlipRequired) || mode_lib->ms.support.ImmediateFlipSupport)) {
-			dml2_printf("DML::%s: mode is supported\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: mode is supported\n", __func__);
 			mode_lib->ms.support.ModeSupport = true;
 		} else {
-			dml2_printf("DML::%s: mode is NOT supported\n", __func__);
+			DML_LOG_VERBOSE("DML::%s: mode is NOT supported\n", __func__);
 			mode_lib->ms.support.ModeSupport = false;
 		}
 	}
 
 	// Since now the mode_support work on 1 particular power state, so there is only 1 state idx (index 0).
-	dml2_printf("DML::%s: ModeSupport = %u\n", __func__, mode_lib->ms.support.ModeSupport);
-	dml2_printf("DML::%s: ImmediateFlipSupport = %u\n", __func__, mode_lib->ms.support.ImmediateFlipSupport);
+	DML_LOG_VERBOSE("DML::%s: ModeSupport = %u\n", __func__, mode_lib->ms.support.ModeSupport);
+	DML_LOG_VERBOSE("DML::%s: ImmediateFlipSupport = %u\n", __func__, mode_lib->ms.support.ImmediateFlipSupport);
 
 	for (k = 0; k < mode_lib->ms.num_active_planes; k++) {
 		mode_lib->ms.support.MPCCombineEnable[k] = mode_lib->ms.MPCCombine[k];
@@ -9623,8 +9552,8 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 		mode_lib->ms.support.OutputRate[k] = mode_lib->ms.OutputRate[k];
 
 #if defined(__DML_VBA_DEBUG__)
-		dml2_printf("DML::%s: k=%d, ODMMode = %u\n", __func__, k, mode_lib->ms.support.ODMMode[k]);
-		dml2_printf("DML::%s: k=%d, DSCEnabled = %u\n", __func__, k, mode_lib->ms.support.DSCEnabled[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, ODMMode = %u\n", __func__, k, mode_lib->ms.support.ODMMode[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, DSCEnabled = %u\n", __func__, k, mode_lib->ms.support.DSCEnabled[k]);
 #endif
 	}
 
@@ -9632,7 +9561,7 @@ static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out
 	if (!mode_lib->ms.support.ModeSupport)
 		dml2_print_mode_support_info(&mode_lib->ms.support, true);
 
-	dml2_printf("DML::%s: --- DONE --- \n", __func__);
+	DML_LOG_VERBOSE("DML::%s: --- DONE --- \n", __func__);
 #endif
 
 	return mode_lib->ms.support.ModeSupport;
@@ -9642,18 +9571,18 @@ unsigned int dml2_core_calcs_mode_support_ex(struct dml2_core_calcs_mode_support
 {
 	unsigned int result;
 
-	dml2_printf("DML::%s: ------------- START ----------\n", __func__);
+	DML_LOG_VERBOSE("DML::%s: ------------- START ----------\n", __func__);
 	result = dml_core_mode_support(in_out_params);
 
 	if (result)
 		*in_out_params->out_evaluation_info = in_out_params->mode_lib->ms.support;
 
-	dml2_printf("DML::%s: is_mode_support = %u (min_clk_index=%d)\n", __func__, result, in_out_params->min_clk_index);
+	DML_LOG_VERBOSE("DML::%s: is_mode_support = %u (min_clk_index=%d)\n", __func__, result, in_out_params->min_clk_index);
 
 	for (unsigned int k = 0; k < in_out_params->in_display_cfg->num_planes; k++)
-		dml2_printf("DML::%s: plane_%d: reserved_vblank_time_ns = %u\n", __func__, k, in_out_params->in_display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns);
+		DML_LOG_VERBOSE("DML::%s: plane_%d: reserved_vblank_time_ns = %lu\n", __func__, k, in_out_params->in_display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns);
 
-	dml2_printf("DML::%s: ------------- DONE ----------\n", __func__);
+	DML_LOG_VERBOSE("DML::%s: ------------- DONE ----------\n", __func__);
 
 	return result;
 }
@@ -9687,19 +9616,19 @@ static void CalculatePixelDeliveryTimes(
 		double pixel_clock_mhz = ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u : HRatio = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio);
-		dml2_printf("DML::%s: k=%u : VRatio = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
-		dml2_printf("DML::%s: k=%u : HRatioChroma = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane1.h_ratio);
-		dml2_printf("DML::%s: k=%u : VRatioChroma = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio);
-		dml2_printf("DML::%s: k=%u : VRatioPrefetchY = %f\n", __func__, k, VRatioPrefetchY[k]);
-		dml2_printf("DML::%s: k=%u : VRatioPrefetchC = %f\n", __func__, k, VRatioPrefetchC[k]);
-		dml2_printf("DML::%s: k=%u : swath_width_luma_ub = %u\n", __func__, k, swath_width_luma_ub[k]);
-		dml2_printf("DML::%s: k=%u : swath_width_chroma_ub = %u\n", __func__, k, swath_width_chroma_ub[k]);
-		dml2_printf("DML::%s: k=%u : PSCL_THROUGHPUT = %f\n", __func__, k, PSCL_THROUGHPUT[k]);
-		dml2_printf("DML::%s: k=%u : PSCL_THROUGHPUT_CHROMA = %f\n", __func__, k, PSCL_THROUGHPUT_CHROMA[k]);
-		dml2_printf("DML::%s: k=%u : DPPPerSurface = %u\n", __func__, k, cfg_support_info->plane_support_info[k].dpps_used);
-		dml2_printf("DML::%s: k=%u : pixel_clock_mhz = %f\n", __func__, k, pixel_clock_mhz);
-		dml2_printf("DML::%s: k=%u : Dppclk = %f\n", __func__, k, Dppclk[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : HRatio = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u : VRatio = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u : HRatioChroma = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane1.h_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u : VRatioChroma = %f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio);
+		DML_LOG_VERBOSE("DML::%s: k=%u : VRatioPrefetchY = %f\n", __func__, k, VRatioPrefetchY[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : VRatioPrefetchC = %f\n", __func__, k, VRatioPrefetchC[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : swath_width_luma_ub = %u\n", __func__, k, swath_width_luma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : swath_width_chroma_ub = %u\n", __func__, k, swath_width_chroma_ub[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : PSCL_THROUGHPUT = %f\n", __func__, k, PSCL_THROUGHPUT[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : PSCL_THROUGHPUT_CHROMA = %f\n", __func__, k, PSCL_THROUGHPUT_CHROMA[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DPPPerSurface = %u\n", __func__, k, cfg_support_info->plane_support_info[k].dpps_used);
+		DML_LOG_VERBOSE("DML::%s: k=%u : pixel_clock_mhz = %f\n", __func__, k, pixel_clock_mhz);
+		DML_LOG_VERBOSE("DML::%s: k=%u : Dppclk = %f\n", __func__, k, Dppclk[k]);
 #endif
 		if (display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio <= 1) {
 			DisplayPipeLineDeliveryTimeLuma[k] = swath_width_luma_ub[k] * cfg_support_info->plane_support_info[k].dpps_used / display_cfg->plane_descriptors[k].composition.scaler_info.plane0.h_ratio / pixel_clock_mhz;
@@ -9733,10 +9662,10 @@ static void CalculatePixelDeliveryTimes(
 			}
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u : DisplayPipeLineDeliveryTimeLuma = %f\n", __func__, k, DisplayPipeLineDeliveryTimeLuma[k]);
-		dml2_printf("DML::%s: k=%u : DisplayPipeLineDeliveryTimeLumaPrefetch = %f\n", __func__, k, DisplayPipeLineDeliveryTimeLumaPrefetch[k]);
-		dml2_printf("DML::%s: k=%u : DisplayPipeLineDeliveryTimeChroma = %f\n", __func__, k, DisplayPipeLineDeliveryTimeChroma[k]);
-		dml2_printf("DML::%s: k=%u : DisplayPipeLineDeliveryTimeChromaPrefetch = %f\n", __func__, k, DisplayPipeLineDeliveryTimeChromaPrefetch[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeLineDeliveryTimeLuma = %f\n", __func__, k, DisplayPipeLineDeliveryTimeLuma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeLineDeliveryTimeLumaPrefetch = %f\n", __func__, k, DisplayPipeLineDeliveryTimeLumaPrefetch[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeLineDeliveryTimeChroma = %f\n", __func__, k, DisplayPipeLineDeliveryTimeChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeLineDeliveryTimeChromaPrefetch = %f\n", __func__, k, DisplayPipeLineDeliveryTimeChromaPrefetch[k]);
 #endif
 	}
 
@@ -9752,12 +9681,12 @@ static void CalculatePixelDeliveryTimes(
 			DisplayPipeRequestDeliveryTimeChromaPrefetch[k] = DisplayPipeLineDeliveryTimeChromaPrefetch[k] / req_per_swath_ub_c[k];
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeLuma = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeLuma[k]);
-		dml2_printf("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeLumaPrefetch = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeLumaPrefetch[k]);
-		dml2_printf("DML::%s: k=%u : req_per_swath_ub_l = %d\n", __func__, k, req_per_swath_ub_l[k]);
-		dml2_printf("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeChroma = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeChroma[k]);
-		dml2_printf("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeChromaPrefetch = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeChromaPrefetch[k]);
-		dml2_printf("DML::%s: k=%u : req_per_swath_ub_c = %d\n", __func__, k, req_per_swath_ub_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeLuma = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeLuma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeLumaPrefetch = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeLumaPrefetch[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : req_per_swath_ub_l = %d\n", __func__, k, req_per_swath_ub_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeChroma = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeChroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : DisplayPipeRequestDeliveryTimeChromaPrefetch = %f\n", __func__, k, DisplayPipeRequestDeliveryTimeChromaPrefetch[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u : req_per_swath_ub_c = %d\n", __func__, k, req_per_swath_ub_c[k]);
 #endif
 	}
 }
@@ -9853,14 +9782,14 @@ static void CalculateMetaAndPTETimes(struct dml2_core_shared_CalculateMetaAndPTE
 		}
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%d, DST_Y_PER_META_ROW_NOM_L = %f\n", __func__, k, p->DST_Y_PER_META_ROW_NOM_L[k]);
-		dml2_printf("DML::%s: k=%d, DST_Y_PER_META_ROW_NOM_C = %f\n", __func__, k, p->DST_Y_PER_META_ROW_NOM_C[k]);
-		dml2_printf("DML::%s: k=%d, TimePerMetaChunkNominal		  = %f\n", __func__, k, p->TimePerMetaChunkNominal[k]);
-		dml2_printf("DML::%s: k=%d, TimePerMetaChunkVBlank		   = %f\n", __func__, k, p->TimePerMetaChunkVBlank[k]);
-		dml2_printf("DML::%s: k=%d, TimePerMetaChunkFlip			 = %f\n", __func__, k, p->TimePerMetaChunkFlip[k]);
-		dml2_printf("DML::%s: k=%d, TimePerChromaMetaChunkNominal	= %f\n", __func__, k, p->TimePerChromaMetaChunkNominal[k]);
-		dml2_printf("DML::%s: k=%d, TimePerChromaMetaChunkVBlank	 = %f\n", __func__, k, p->TimePerChromaMetaChunkVBlank[k]);
-		dml2_printf("DML::%s: k=%d, TimePerChromaMetaChunkFlip	   = %f\n", __func__, k, p->TimePerChromaMetaChunkFlip[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, DST_Y_PER_META_ROW_NOM_L = %f\n", __func__, k, p->DST_Y_PER_META_ROW_NOM_L[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, DST_Y_PER_META_ROW_NOM_C = %f\n", __func__, k, p->DST_Y_PER_META_ROW_NOM_C[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, TimePerMetaChunkNominal		  = %f\n", __func__, k, p->TimePerMetaChunkNominal[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, TimePerMetaChunkVBlank		   = %f\n", __func__, k, p->TimePerMetaChunkVBlank[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, TimePerMetaChunkFlip			 = %f\n", __func__, k, p->TimePerMetaChunkFlip[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, TimePerChromaMetaChunkNominal	= %f\n", __func__, k, p->TimePerChromaMetaChunkNominal[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, TimePerChromaMetaChunkVBlank	 = %f\n", __func__, k, p->TimePerChromaMetaChunkVBlank[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, TimePerChromaMetaChunkFlip	   = %f\n", __func__, k, p->TimePerChromaMetaChunkFlip[k]);
 #endif
 	}
 
@@ -9881,7 +9810,7 @@ static void CalculateMetaAndPTETimes(struct dml2_core_shared_CalculateMetaAndPTE
 		else
 			p->time_per_tdlut_group[k] = 0;
 
-		dml2_printf("DML::%s: k=%u, time_per_tdlut_group = %f\n", __func__, k, p->time_per_tdlut_group[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_tdlut_group = %f\n", __func__, k, p->time_per_tdlut_group[k]);
 
 		if (p->display_cfg->gpuvm_enable == true) {
 			if (!dml_is_vertical_rotation(p->display_cfg->plane_descriptors[k].composition.rotation_angle)) {
@@ -9897,14 +9826,14 @@ static void CalculateMetaAndPTETimes(struct dml2_core_shared_CalculateMetaAndPTE
 			if (dpte_groups_per_row_luma_ub <= 2) {
 				dpte_groups_per_row_luma_ub = dpte_groups_per_row_luma_ub + 1;
 			}
-			dml2_printf("DML::%s: k=%u, use_one_row_for_frame = %u\n", __func__, k, p->use_one_row_for_frame[k]);
-			dml2_printf("DML::%s: k=%u, dpte_group_bytes = %u\n", __func__, k, p->dpte_group_bytes[k]);
-			dml2_printf("DML::%s: k=%u, PTERequestSizeY = %u\n", __func__, k, p->PTERequestSizeY[k]);
-			dml2_printf("DML::%s: k=%u, PixelPTEReqWidthY = %u\n", __func__, k, p->PixelPTEReqWidthY[k]);
-			dml2_printf("DML::%s: k=%u, PixelPTEReqHeightY = %u\n", __func__, k, p->PixelPTEReqHeightY[k]);
-			dml2_printf("DML::%s: k=%u, dpte_row_width_luma_ub = %u\n", __func__, k, p->dpte_row_width_luma_ub[k]);
-			dml2_printf("DML::%s: k=%u, dpte_group_width_luma = %u\n", __func__, k, dpte_group_width_luma);
-			dml2_printf("DML::%s: k=%u, dpte_groups_per_row_luma_ub = %u\n", __func__, k, dpte_groups_per_row_luma_ub);
+			DML_LOG_VERBOSE("DML::%s: k=%u, use_one_row_for_frame = %u\n", __func__, k, p->use_one_row_for_frame[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, dpte_group_bytes = %u\n", __func__, k, p->dpte_group_bytes[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PTERequestSizeY = %u\n", __func__, k, p->PTERequestSizeY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEReqWidthY = %u\n", __func__, k, p->PixelPTEReqWidthY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, PixelPTEReqHeightY = %u\n", __func__, k, p->PixelPTEReqHeightY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, dpte_row_width_luma_ub = %u\n", __func__, k, p->dpte_row_width_luma_ub[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, dpte_group_width_luma = %u\n", __func__, k, dpte_group_width_luma);
+			DML_LOG_VERBOSE("DML::%s: k=%u, dpte_groups_per_row_luma_ub = %u\n", __func__, k, dpte_groups_per_row_luma_ub);
 
 			p->time_per_pte_group_nom_luma[k] = p->DST_Y_PER_PTE_ROW_NOM_L[k] * p->display_cfg->stream_descriptors[p->display_cfg->plane_descriptors[k].stream_index].timing.h_total / pixel_clock_mhz / dpte_groups_per_row_luma_ub;
 			p->time_per_pte_group_vblank_luma[k] = p->dst_y_per_row_vblank[k] * p->display_cfg->stream_descriptors[p->display_cfg->plane_descriptors[k].stream_index].timing.h_total / pixel_clock_mhz / dpte_groups_per_row_luma_ub;
@@ -9928,9 +9857,9 @@ static void CalculateMetaAndPTETimes(struct dml2_core_shared_CalculateMetaAndPTE
 				if (dpte_groups_per_row_chroma_ub <= 2) {
 					dpte_groups_per_row_chroma_ub = dpte_groups_per_row_chroma_ub + 1;
 				}
-				dml2_printf("DML::%s: k=%u, dpte_row_width_chroma_ub = %u\n", __func__, k, p->dpte_row_width_chroma_ub[k]);
-				dml2_printf("DML::%s: k=%u, dpte_group_width_chroma = %u\n", __func__, k, dpte_group_width_chroma);
-				dml2_printf("DML::%s: k=%u, dpte_groups_per_row_chroma_ub = %u\n", __func__, k, dpte_groups_per_row_chroma_ub);
+				DML_LOG_VERBOSE("DML::%s: k=%u, dpte_row_width_chroma_ub = %u\n", __func__, k, p->dpte_row_width_chroma_ub[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, dpte_group_width_chroma = %u\n", __func__, k, dpte_group_width_chroma);
+				DML_LOG_VERBOSE("DML::%s: k=%u, dpte_groups_per_row_chroma_ub = %u\n", __func__, k, dpte_groups_per_row_chroma_ub);
 
 				p->time_per_pte_group_nom_chroma[k] = p->DST_Y_PER_PTE_ROW_NOM_C[k] * p->display_cfg->stream_descriptors[p->display_cfg->plane_descriptors[k].stream_index].timing.h_total / pixel_clock_mhz / dpte_groups_per_row_chroma_ub;
 				p->time_per_pte_group_vblank_chroma[k] = p->dst_y_per_row_vblank[k] * p->display_cfg->stream_descriptors[p->display_cfg->plane_descriptors[k].stream_index].timing.h_total / pixel_clock_mhz / dpte_groups_per_row_chroma_ub;
@@ -9945,17 +9874,17 @@ static void CalculateMetaAndPTETimes(struct dml2_core_shared_CalculateMetaAndPTE
 			p->time_per_pte_group_flip_chroma[k] = 0;
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, dst_y_per_row_vblank = %f\n", __func__, k, p->dst_y_per_row_vblank[k]);
-		dml2_printf("DML::%s: k=%u, dst_y_per_row_flip = %f\n", __func__, k, p->dst_y_per_row_flip[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dst_y_per_row_vblank = %f\n", __func__, k, p->dst_y_per_row_vblank[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dst_y_per_row_flip = %f\n", __func__, k, p->dst_y_per_row_flip[k]);
 
-		dml2_printf("DML::%s: k=%u, DST_Y_PER_PTE_ROW_NOM_L = %f\n", __func__, k, p->DST_Y_PER_PTE_ROW_NOM_L[k]);
-		dml2_printf("DML::%s: k=%u, DST_Y_PER_PTE_ROW_NOM_C = %f\n", __func__, k, p->DST_Y_PER_PTE_ROW_NOM_C[k]);
-		dml2_printf("DML::%s: k=%u, time_per_pte_group_nom_luma = %f\n", __func__, k, p->time_per_pte_group_nom_luma[k]);
-		dml2_printf("DML::%s: k=%u, time_per_pte_group_vblank_luma = %f\n", __func__, k, p->time_per_pte_group_vblank_luma[k]);
-		dml2_printf("DML::%s: k=%u, time_per_pte_group_flip_luma = %f\n", __func__, k, p->time_per_pte_group_flip_luma[k]);
-		dml2_printf("DML::%s: k=%u, time_per_pte_group_nom_chroma = %f\n", __func__, k, p->time_per_pte_group_nom_chroma[k]);
-		dml2_printf("DML::%s: k=%u, time_per_pte_group_vblank_chroma = %f\n", __func__, k, p->time_per_pte_group_vblank_chroma[k]);
-		dml2_printf("DML::%s: k=%u, time_per_pte_group_flip_chroma = %f\n", __func__, k, p->time_per_pte_group_flip_chroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, DST_Y_PER_PTE_ROW_NOM_L = %f\n", __func__, k, p->DST_Y_PER_PTE_ROW_NOM_L[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, DST_Y_PER_PTE_ROW_NOM_C = %f\n", __func__, k, p->DST_Y_PER_PTE_ROW_NOM_C[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_pte_group_nom_luma = %f\n", __func__, k, p->time_per_pte_group_nom_luma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_pte_group_vblank_luma = %f\n", __func__, k, p->time_per_pte_group_vblank_luma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_pte_group_flip_luma = %f\n", __func__, k, p->time_per_pte_group_flip_luma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_pte_group_nom_chroma = %f\n", __func__, k, p->time_per_pte_group_nom_chroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_pte_group_vblank_chroma = %f\n", __func__, k, p->time_per_pte_group_vblank_chroma[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, time_per_pte_group_flip_chroma = %f\n", __func__, k, p->time_per_pte_group_flip_chroma[k]);
 #endif
 	}
 } // CalculateMetaAndPTETimes
@@ -9991,18 +9920,18 @@ static void CalculateVMGroupAndRequestTimes(
 	double line_time;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: NumberOfActiveSurfaces = %u\n", __func__, NumberOfActiveSurfaces);
+	DML_LOG_VERBOSE("DML::%s: NumberOfActiveSurfaces = %u\n", __func__, NumberOfActiveSurfaces);
 #endif
 	for (unsigned int k = 0; k < NumberOfActiveSurfaces; ++k) {
 		double pixel_clock_mhz = ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
 		bool dcc_mrq_enable = display_cfg->plane_descriptors[k].surface.dcc.enable && mrq_present;
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, dcc_mrq_enable = %u\n", __func__, k, dcc_mrq_enable);
-		dml2_printf("DML::%s: k=%u, vm_group_bytes = %u\n", __func__, k, vm_group_bytes[k]);
-		dml2_printf("DML::%s: k=%u, dpde0_bytes_per_frame_ub_l = %u\n", __func__, k, dpde0_bytes_per_frame_ub_l[k]);
-		dml2_printf("DML::%s: k=%u, dpde0_bytes_per_frame_ub_c = %u\n", __func__, k, dpde0_bytes_per_frame_ub_c[k]);
-		dml2_printf("DML::%s: k=%d, meta_pte_bytes_per_frame_ub_l = %d\n", __func__, k, meta_pte_bytes_per_frame_ub_l[k]);
-		dml2_printf("DML::%s: k=%d, meta_pte_bytes_per_frame_ub_c = %d\n", __func__, k, meta_pte_bytes_per_frame_ub_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dcc_mrq_enable = %u\n", __func__, k, dcc_mrq_enable);
+		DML_LOG_VERBOSE("DML::%s: k=%u, vm_group_bytes = %u\n", __func__, k, vm_group_bytes[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dpde0_bytes_per_frame_ub_l = %u\n", __func__, k, dpde0_bytes_per_frame_ub_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, dpde0_bytes_per_frame_ub_c = %u\n", __func__, k, dpde0_bytes_per_frame_ub_c[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, meta_pte_bytes_per_frame_ub_l = %d\n", __func__, k, meta_pte_bytes_per_frame_ub_l[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d, meta_pte_bytes_per_frame_ub_c = %d\n", __func__, k, meta_pte_bytes_per_frame_ub_c[k]);
 #endif
 
 		if (display_cfg->gpuvm_enable) {
@@ -10071,13 +10000,13 @@ static void CalculateVMGroupAndRequestTimes(
 			else
 				TimePerVMRequestFlip[k] = 0.0;
 
-			dml2_printf("DML::%s: k=%u, dst_y_per_vm_vblank = %f\n", __func__, k, dst_y_per_vm_vblank[k]);
-			dml2_printf("DML::%s: k=%u, dst_y_per_vm_flip = %f\n", __func__, k, dst_y_per_vm_flip[k]);
-			dml2_printf("DML::%s: k=%u, line_time = %f\n", __func__, k, line_time);
-			dml2_printf("DML::%s: k=%u, num_group_per_lower_vm_stage_pref = %f\n", __func__, k, num_group_per_lower_vm_stage_pref);
-			dml2_printf("DML::%s: k=%u, num_group_per_lower_vm_stage_flip = %f\n", __func__, k, num_group_per_lower_vm_stage_flip);
-			dml2_printf("DML::%s: k=%u, num_req_per_lower_vm_stage_pref = %f\n", __func__, k, num_req_per_lower_vm_stage_pref);
-			dml2_printf("DML::%s: k=%u, num_req_per_lower_vm_stage_flip = %f\n", __func__, k, num_req_per_lower_vm_stage_flip);
+			DML_LOG_VERBOSE("DML::%s: k=%u, dst_y_per_vm_vblank = %f\n", __func__, k, dst_y_per_vm_vblank[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, dst_y_per_vm_flip = %f\n", __func__, k, dst_y_per_vm_flip[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, line_time = %f\n", __func__, k, line_time);
+			DML_LOG_VERBOSE("DML::%s: k=%u, num_group_per_lower_vm_stage_pref = %d\n", __func__, k, num_group_per_lower_vm_stage_pref);
+			DML_LOG_VERBOSE("DML::%s: k=%u, num_group_per_lower_vm_stage_flip = %d\n", __func__, k, num_group_per_lower_vm_stage_flip);
+			DML_LOG_VERBOSE("DML::%s: k=%u, num_req_per_lower_vm_stage_pref = %d\n", __func__, k, num_req_per_lower_vm_stage_pref);
+			DML_LOG_VERBOSE("DML::%s: k=%u, num_req_per_lower_vm_stage_flip = %d\n", __func__, k, num_req_per_lower_vm_stage_flip);
 
 			if (display_cfg->gpuvm_max_page_table_levels > 2) {
 				TimePerVMGroupVBlank[k] = TimePerVMGroupVBlank[k] / 2;
@@ -10094,10 +10023,10 @@ static void CalculateVMGroupAndRequestTimes(
 		}
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u, TimePerVMGroupVBlank = %f\n", __func__, k, TimePerVMGroupVBlank[k]);
-		dml2_printf("DML::%s: k=%u, TimePerVMGroupFlip = %f\n", __func__, k, TimePerVMGroupFlip[k]);
-		dml2_printf("DML::%s: k=%u, TimePerVMRequestVBlank = %f\n", __func__, k, TimePerVMRequestVBlank[k]);
-		dml2_printf("DML::%s: k=%u, TimePerVMRequestFlip = %f\n", __func__, k, TimePerVMRequestFlip[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, TimePerVMGroupVBlank = %f\n", __func__, k, TimePerVMGroupVBlank[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, TimePerVMGroupFlip = %f\n", __func__, k, TimePerVMGroupFlip[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, TimePerVMRequestVBlank = %f\n", __func__, k, TimePerVMRequestVBlank[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u, TimePerVMRequestFlip = %f\n", __func__, k, TimePerVMRequestFlip[k]);
 #endif
 	}
 }
@@ -10113,7 +10042,6 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 	unsigned int SingleVTotal = 0;
 	bool SameTiming = true;
 	bool FoundCriticalSurface = false;
-	double LastZ8StutterPeriod = 0;
 
 	memset(l, 0, sizeof(struct dml2_core_calcs_CalculateStutterEfficiency_locals));
 
@@ -10127,9 +10055,9 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 				}
 				l->TotalCompressedReadBandwidth = l->TotalCompressedReadBandwidth + p->ReadBandwidthSurfaceLuma[k] / math_min2(p->display_cfg->plane_descriptors[k].surface.dcc.informative.dcc_rate_plane0, l->MaximumEffectiveCompressionLuma);
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k=%u, ReadBandwidthSurfaceLuma = %f\n", __func__, k, p->ReadBandwidthSurfaceLuma[k]);
-				dml2_printf("DML::%s: k=%u, NetDCCRateLuma = %f\n", __func__, k, p->display_cfg->plane_descriptors[k].surface.dcc.informative.dcc_rate_plane0);
-				dml2_printf("DML::%s: k=%u, MaximumEffectiveCompressionLuma = %f\n", __func__, k, l->MaximumEffectiveCompressionLuma);
+				DML_LOG_VERBOSE("DML::%s: k=%u, ReadBandwidthSurfaceLuma = %f\n", __func__, k, p->ReadBandwidthSurfaceLuma[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, NetDCCRateLuma = %f\n", __func__, k, p->display_cfg->plane_descriptors[k].surface.dcc.informative.dcc_rate_plane0);
+				DML_LOG_VERBOSE("DML::%s: k=%u, MaximumEffectiveCompressionLuma = %f\n", __func__, k, l->MaximumEffectiveCompressionLuma);
 #endif
 				l->TotalZeroSizeRequestReadBandwidth = l->TotalZeroSizeRequestReadBandwidth + p->ReadBandwidthSurfaceLuma[k] * p->display_cfg->plane_descriptors[k].surface.dcc.informative.fraction_of_zero_size_request_plane0;
 				l->TotalZeroSizeCompressedReadBandwidth = l->TotalZeroSizeCompressedReadBandwidth + p->ReadBandwidthSurfaceLuma[k] * p->display_cfg->plane_descriptors[k].surface.dcc.informative.fraction_of_zero_size_request_plane0 / l->MaximumEffectiveCompressionLuma;
@@ -10142,9 +10070,9 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 					}
 					l->TotalCompressedReadBandwidth = l->TotalCompressedReadBandwidth + p->ReadBandwidthSurfaceChroma[k] / math_min2(p->display_cfg->plane_descriptors[k].surface.dcc.informative.dcc_rate_plane1, l->MaximumEffectiveCompressionChroma);
 #ifdef __DML_VBA_DEBUG__
-					dml2_printf("DML::%s: k=%u, ReadBandwidthSurfaceChroma = %f\n", __func__, k, p->ReadBandwidthSurfaceChroma[k]);
-					dml2_printf("DML::%s: k=%u, NetDCCRateChroma = %f\n", __func__, k, p->display_cfg->plane_descriptors[k].surface.dcc.informative.dcc_rate_plane1);
-					dml2_printf("DML::%s: k=%u, MaximumEffectiveCompressionChroma = %f\n", __func__, k, l->MaximumEffectiveCompressionChroma);
+					DML_LOG_VERBOSE("DML::%s: k=%u, ReadBandwidthSurfaceChroma = %f\n", __func__, k, p->ReadBandwidthSurfaceChroma[k]);
+					DML_LOG_VERBOSE("DML::%s: k=%u, NetDCCRateChroma = %f\n", __func__, k, p->display_cfg->plane_descriptors[k].surface.dcc.informative.dcc_rate_plane1);
+					DML_LOG_VERBOSE("DML::%s: k=%u, MaximumEffectiveCompressionChroma = %f\n", __func__, k, l->MaximumEffectiveCompressionChroma);
 #endif
 					l->TotalZeroSizeRequestReadBandwidth = l->TotalZeroSizeRequestReadBandwidth + p->ReadBandwidthSurfaceChroma[k] * p->display_cfg->plane_descriptors[k].surface.dcc.informative.fraction_of_zero_size_request_plane1;
 					l->TotalZeroSizeCompressedReadBandwidth = l->TotalZeroSizeCompressedReadBandwidth + p->ReadBandwidthSurfaceChroma[k] * p->display_cfg->plane_descriptors[k].surface.dcc.informative.fraction_of_zero_size_request_plane1 / l->MaximumEffectiveCompressionChroma;
@@ -10160,19 +10088,19 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 	l->AverageDCCZeroSizeFraction = l->TotalZeroSizeRequestReadBandwidth / p->TotalDataReadBandwidth;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: UnboundedRequestEnabled = %u\n", __func__, p->UnboundedRequestEnabled);
-	dml2_printf("DML::%s: TotalCompressedReadBandwidth = %f\n", __func__, l->TotalCompressedReadBandwidth);
-	dml2_printf("DML::%s: TotalZeroSizeRequestReadBandwidth = %f\n", __func__, l->TotalZeroSizeRequestReadBandwidth);
-	dml2_printf("DML::%s: TotalZeroSizeCompressedReadBandwidth = %f\n", __func__, l->TotalZeroSizeCompressedReadBandwidth);
-	dml2_printf("DML::%s: MaximumEffectiveCompressionLuma = %f\n", __func__, l->MaximumEffectiveCompressionLuma);
-	dml2_printf("DML::%s: MaximumEffectiveCompressionChroma = %f\n", __func__, l->MaximumEffectiveCompressionChroma);
-	dml2_printf("DML::%s: AverageDCCCompressionRate = %f\n", __func__, l->AverageDCCCompressionRate);
-	dml2_printf("DML::%s: AverageDCCZeroSizeFraction = %f\n", __func__, l->AverageDCCZeroSizeFraction);
+	DML_LOG_VERBOSE("DML::%s: UnboundedRequestEnabled = %u\n", __func__, p->UnboundedRequestEnabled);
+	DML_LOG_VERBOSE("DML::%s: TotalCompressedReadBandwidth = %f\n", __func__, l->TotalCompressedReadBandwidth);
+	DML_LOG_VERBOSE("DML::%s: TotalZeroSizeRequestReadBandwidth = %f\n", __func__, l->TotalZeroSizeRequestReadBandwidth);
+	DML_LOG_VERBOSE("DML::%s: TotalZeroSizeCompressedReadBandwidth = %f\n", __func__, l->TotalZeroSizeCompressedReadBandwidth);
+	DML_LOG_VERBOSE("DML::%s: MaximumEffectiveCompressionLuma = %f\n", __func__, l->MaximumEffectiveCompressionLuma);
+	DML_LOG_VERBOSE("DML::%s: MaximumEffectiveCompressionChroma = %f\n", __func__, l->MaximumEffectiveCompressionChroma);
+	DML_LOG_VERBOSE("DML::%s: AverageDCCCompressionRate = %f\n", __func__, l->AverageDCCCompressionRate);
+	DML_LOG_VERBOSE("DML::%s: AverageDCCZeroSizeFraction = %f\n", __func__, l->AverageDCCZeroSizeFraction);
 
-	dml2_printf("DML::%s: CompbufReservedSpace64B = %u (%f kbytes)\n", __func__, p->CompbufReservedSpace64B, p->CompbufReservedSpace64B * 64 / 1024.0);
-	dml2_printf("DML::%s: CompbufReservedSpaceZs = %u\n", __func__, p->CompbufReservedSpaceZs);
-	dml2_printf("DML::%s: CompressedBufferSizeInkByte = %u kbytes\n", __func__, p->CompressedBufferSizeInkByte);
-	dml2_printf("DML::%s: ROBBufferSizeInKByte = %u kbytes\n", __func__, p->ROBBufferSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: CompbufReservedSpace64B = %u (%f kbytes)\n", __func__, p->CompbufReservedSpace64B, p->CompbufReservedSpace64B * 64 / 1024.0);
+	DML_LOG_VERBOSE("DML::%s: CompbufReservedSpaceZs = %u\n", __func__, p->CompbufReservedSpaceZs);
+	DML_LOG_VERBOSE("DML::%s: CompressedBufferSizeInkByte = %u kbytes\n", __func__, p->CompressedBufferSizeInkByte);
+	DML_LOG_VERBOSE("DML::%s: ROBBufferSizeInKByte = %u kbytes\n", __func__, p->ROBBufferSizeInKByte);
 #endif
 	if (l->AverageDCCZeroSizeFraction == 1) {
 		l->AverageZeroSizeCompressionRate = l->TotalZeroSizeRequestReadBandwidth / l->TotalZeroSizeCompressedReadBandwidth;
@@ -10189,10 +10117,10 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: min 1 = %f\n", __func__, p->CompressedBufferSizeInkByte * 1024 * l->AverageDCCCompressionRate);
-		dml2_printf("DML::%s: min 2 = %f\n", __func__, p->MetaFIFOSizeInKEntries * 1024 * 64 / (l->AverageDCCZeroSizeFraction / l->AverageZeroSizeCompressionRate + 1 / l->AverageDCCCompressionRate));
-		dml2_printf("DML::%s: min 3 = %d\n", __func__, (p->ROBBufferSizeInKByte * 1024 - p->CompbufReservedSpace64B * 64));
-		dml2_printf("DML::%s: min 4 = %f\n", __func__, (p->ZeroSizeBufferEntries - p->CompbufReservedSpaceZs) * 64 / (l->AverageDCCZeroSizeFraction / l->AverageZeroSizeCompressionRate));
+		DML_LOG_VERBOSE("DML::%s: min 1 = %f\n", __func__, p->CompressedBufferSizeInkByte * 1024 * l->AverageDCCCompressionRate);
+		DML_LOG_VERBOSE("DML::%s: min 2 = %f\n", __func__, p->MetaFIFOSizeInKEntries * 1024 * 64 / (l->AverageDCCZeroSizeFraction / l->AverageZeroSizeCompressionRate + 1 / l->AverageDCCCompressionRate));
+		DML_LOG_VERBOSE("DML::%s: min 3 = %d\n", __func__, (p->ROBBufferSizeInKByte * 1024 - p->CompbufReservedSpace64B * 64));
+		DML_LOG_VERBOSE("DML::%s: min 4 = %f\n", __func__, (p->ZeroSizeBufferEntries - p->CompbufReservedSpaceZs) * 64 / (l->AverageDCCZeroSizeFraction / l->AverageZeroSizeCompressionRate));
 #endif
 	} else {
 		l->EffectiveCompressedBufferSize = math_min2((double)p->CompressedBufferSizeInkByte * 1024 * l->AverageDCCCompressionRate,
@@ -10200,16 +10128,16 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 			((double)p->ROBBufferSizeInKByte * 1024 - p->CompbufReservedSpace64B * 64) * (p->rob_alloc_compressed ? l->AverageDCCCompressionRate : 1.0);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: min 1 = %f\n", __func__, p->CompressedBufferSizeInkByte * 1024 * l->AverageDCCCompressionRate);
-		dml2_printf("DML::%s: min 2 = %f\n", __func__, p->MetaFIFOSizeInKEntries * 1024 * 64 * l->AverageDCCCompressionRate);
+		DML_LOG_VERBOSE("DML::%s: min 1 = %f\n", __func__, p->CompressedBufferSizeInkByte * 1024 * l->AverageDCCCompressionRate);
+		DML_LOG_VERBOSE("DML::%s: min 2 = %f\n", __func__, p->MetaFIFOSizeInKEntries * 1024 * 64 * l->AverageDCCCompressionRate);
 #endif
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: MetaFIFOSizeInKEntries = %u\n", __func__, p->MetaFIFOSizeInKEntries);
-	dml2_printf("DML::%s: ZeroSizeBufferEntries = %u\n", __func__, p->ZeroSizeBufferEntries);
-	dml2_printf("DML::%s: AverageZeroSizeCompressionRate = %f\n", __func__, l->AverageZeroSizeCompressionRate);
-	dml2_printf("DML::%s: EffectiveCompressedBufferSize = %f (%f kbytes)\n", __func__, l->EffectiveCompressedBufferSize, l->EffectiveCompressedBufferSize / 1024.0);
+	DML_LOG_VERBOSE("DML::%s: MetaFIFOSizeInKEntries = %u\n", __func__, p->MetaFIFOSizeInKEntries);
+	DML_LOG_VERBOSE("DML::%s: ZeroSizeBufferEntries = %u\n", __func__, p->ZeroSizeBufferEntries);
+	DML_LOG_VERBOSE("DML::%s: AverageZeroSizeCompressionRate = %f\n", __func__, l->AverageZeroSizeCompressionRate);
+	DML_LOG_VERBOSE("DML::%s: EffectiveCompressedBufferSize = %f (%f kbytes)\n", __func__, l->EffectiveCompressedBufferSize, l->EffectiveCompressedBufferSize / 1024.0);
 #endif
 
 	*p->StutterPeriod = 0;
@@ -10220,15 +10148,15 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 			l->LinesInDETYRoundedDownToSwath = math_floor2(l->LinesInDETY, p->SwathHeightY[k]);
 			l->DETBufferingTimeY = l->LinesInDETYRoundedDownToSwath * ((double)p->display_cfg->stream_descriptors[p->display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)p->display_cfg->stream_descriptors[p->display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) / p->display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio;
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, DETBufferSizeY = %u (%u kbytes)\n", __func__, k, p->DETBufferSizeY[k], p->DETBufferSizeY[k] / 1024);
-			dml2_printf("DML::%s: k=%u, BytePerPixelDETY = %f\n", __func__, k, p->BytePerPixelDETY[k]);
-			dml2_printf("DML::%s: k=%u, SwathWidthY = %u\n", __func__, k, p->SwathWidthY[k]);
-			dml2_printf("DML::%s: k=%u, ReadBandwidthSurfaceLuma = %f\n", __func__, k, p->ReadBandwidthSurfaceLuma[k]);
-			dml2_printf("DML::%s: k=%u, TotalDataReadBandwidth = %f\n", __func__, k, p->TotalDataReadBandwidth);
-			dml2_printf("DML::%s: k=%u, LinesInDETY = %f\n", __func__, k, l->LinesInDETY);
-			dml2_printf("DML::%s: k=%u, LinesInDETYRoundedDownToSwath = %f\n", __func__, k, l->LinesInDETYRoundedDownToSwath);
-			dml2_printf("DML::%s: k=%u, VRatio = %f\n", __func__, k, p->display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
-			dml2_printf("DML::%s: k=%u, DETBufferingTimeY = %f\n", __func__, k, l->DETBufferingTimeY);
+			DML_LOG_VERBOSE("DML::%s: k=%u, DETBufferSizeY = %u (%u kbytes)\n", __func__, k, p->DETBufferSizeY[k], p->DETBufferSizeY[k] / 1024);
+			DML_LOG_VERBOSE("DML::%s: k=%u, BytePerPixelDETY = %f\n", __func__, k, p->BytePerPixelDETY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, SwathWidthY = %u\n", __func__, k, p->SwathWidthY[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, ReadBandwidthSurfaceLuma = %f\n", __func__, k, p->ReadBandwidthSurfaceLuma[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, TotalDataReadBandwidth = %f\n", __func__, k, p->TotalDataReadBandwidth);
+			DML_LOG_VERBOSE("DML::%s: k=%u, LinesInDETY = %f\n", __func__, k, l->LinesInDETY);
+			DML_LOG_VERBOSE("DML::%s: k=%u, LinesInDETYRoundedDownToSwath = %f\n", __func__, k, l->LinesInDETYRoundedDownToSwath);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VRatio = %f\n", __func__, k, p->display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
+			DML_LOG_VERBOSE("DML::%s: k=%u, DETBufferingTimeY = %f\n", __func__, k, l->DETBufferingTimeY);
 #endif
 
 			if (!FoundCriticalSurface || l->DETBufferingTimeY < *p->StutterPeriod) {
@@ -10248,17 +10176,17 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 				l->SinglePipeCriticalSurface = (p->DPPPerSurface[k] == 1);
 
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k=%u, FoundCriticalSurface = %u\n", __func__, k, FoundCriticalSurface);
-				dml2_printf("DML::%s: k=%u, StutterPeriod = %f\n", __func__, k, *p->StutterPeriod);
-				dml2_printf("DML::%s: k=%u, MinTTUVBlankCriticalSurface = %f\n", __func__, k, l->MinTTUVBlankCriticalSurface);
-				dml2_printf("DML::%s: k=%u, FrameTimeCriticalSurface= %f\n", __func__, k, l->FrameTimeCriticalSurface);
-				dml2_printf("DML::%s: k=%u, VActiveTimeCriticalSurface = %f\n", __func__, k, l->VActiveTimeCriticalSurface);
-				dml2_printf("DML::%s: k=%u, BytePerPixelYCriticalSurface = %u\n", __func__, k, l->BytePerPixelYCriticalSurface);
-				dml2_printf("DML::%s: k=%u, SwathWidthYCriticalSurface = %f\n", __func__, k, l->SwathWidthYCriticalSurface);
-				dml2_printf("DML::%s: k=%u, SwathHeightYCriticalSurface = %f\n", __func__, k, l->SwathHeightYCriticalSurface);
-				dml2_printf("DML::%s: k=%u, BlockWidth256BytesYCriticalSurface = %u\n", __func__, k, l->BlockWidth256BytesYCriticalSurface);
-				dml2_printf("DML::%s: k=%u, SinglePlaneCriticalSurface = %u\n", __func__, k, l->SinglePlaneCriticalSurface);
-				dml2_printf("DML::%s: k=%u, SinglePipeCriticalSurface = %u\n", __func__, k, l->SinglePipeCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, FoundCriticalSurface = %u\n", __func__, k, FoundCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, StutterPeriod = %f\n", __func__, k, *p->StutterPeriod);
+				DML_LOG_VERBOSE("DML::%s: k=%u, MinTTUVBlankCriticalSurface = %f\n", __func__, k, l->MinTTUVBlankCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, FrameTimeCriticalSurface= %f\n", __func__, k, l->FrameTimeCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, VActiveTimeCriticalSurface = %f\n", __func__, k, l->VActiveTimeCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, BytePerPixelYCriticalSurface = %u\n", __func__, k, l->BytePerPixelYCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, SwathWidthYCriticalSurface = %f\n", __func__, k, l->SwathWidthYCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, SwathHeightYCriticalSurface = %f\n", __func__, k, l->SwathHeightYCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, BlockWidth256BytesYCriticalSurface = %u\n", __func__, k, l->BlockWidth256BytesYCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, SinglePlaneCriticalSurface = %u\n", __func__, k, l->SinglePlaneCriticalSurface);
+				DML_LOG_VERBOSE("DML::%s: k=%u, SinglePipeCriticalSurface = %u\n", __func__, k, l->SinglePipeCriticalSurface);
 #endif
 			}
 		}
@@ -10276,14 +10204,14 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 
 	l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer = math_min2(*p->StutterPeriod * p->TotalDataReadBandwidth, l->EffectiveCompressedBufferSize);
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: AverageDCCCompressionRate = %f\n", __func__, l->AverageDCCCompressionRate);
-	dml2_printf("DML::%s: StutterPeriod*TotalDataReadBandwidth = %f (%f kbytes)\n", __func__, *p->StutterPeriod * p->TotalDataReadBandwidth, (*p->StutterPeriod * p->TotalDataReadBandwidth) / 1024.0);
-	dml2_printf("DML::%s: EffectiveCompressedBufferSize = %f (%f kbytes)\n", __func__, l->EffectiveCompressedBufferSize, l->EffectiveCompressedBufferSize / 1024.0);
-	dml2_printf("DML::%s: PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer = %f (%f kbytes)\n", __func__, l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer, l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer / 1024);
-	dml2_printf("DML::%s: ReturnBW = %f\n", __func__, p->ReturnBW);
-	dml2_printf("DML::%s: TotalDataReadBandwidth = %f\n", __func__, p->TotalDataReadBandwidth);
-	dml2_printf("DML::%s: TotalRowReadBandwidth = %f\n", __func__, l->TotalRowReadBandwidth);
-	dml2_printf("DML::%s: DCFCLK = %f\n", __func__, p->DCFCLK);
+	DML_LOG_VERBOSE("DML::%s: AverageDCCCompressionRate = %f\n", __func__, l->AverageDCCCompressionRate);
+	DML_LOG_VERBOSE("DML::%s: StutterPeriod*TotalDataReadBandwidth = %f (%f kbytes)\n", __func__, *p->StutterPeriod * p->TotalDataReadBandwidth, (*p->StutterPeriod * p->TotalDataReadBandwidth) / 1024.0);
+	DML_LOG_VERBOSE("DML::%s: EffectiveCompressedBufferSize = %f (%f kbytes)\n", __func__, l->EffectiveCompressedBufferSize, l->EffectiveCompressedBufferSize / 1024.0);
+	DML_LOG_VERBOSE("DML::%s: PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer = %f (%f kbytes)\n", __func__, l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer, l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer / 1024);
+	DML_LOG_VERBOSE("DML::%s: ReturnBW = %f\n", __func__, p->ReturnBW);
+	DML_LOG_VERBOSE("DML::%s: TotalDataReadBandwidth = %f\n", __func__, p->TotalDataReadBandwidth);
+	DML_LOG_VERBOSE("DML::%s: TotalRowReadBandwidth = %f\n", __func__, l->TotalRowReadBandwidth);
+	DML_LOG_VERBOSE("DML::%s: DCFCLK = %f\n", __func__, p->DCFCLK);
 #endif
 
 	l->StutterBurstTime = l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer
@@ -10292,10 +10220,10 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 		/ math_min2(p->DCFCLK * 64, p->ReturnBW * (p->hw_debug5 ? 1 : l->AverageDCCCompressionRate)) +
 		*p->StutterPeriod * l->TotalRowReadBandwidth / p->ReturnBW;
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: Part 1 = %f\n", __func__, l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer / p->ReturnBW / (p->hw_debug5 ? 1 : l->AverageDCCCompressionRate));
-	dml2_printf("DML::%s: Part 2 = %f\n", __func__, (*p->StutterPeriod * p->TotalDataReadBandwidth - l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer) / (p->DCFCLK * 64));
-	dml2_printf("DML::%s: Part 3 = %f\n", __func__, *p->StutterPeriod * l->TotalRowReadBandwidth / p->ReturnBW);
-	dml2_printf("DML::%s: StutterBurstTime = %f\n", __func__, l->StutterBurstTime);
+	DML_LOG_VERBOSE("DML::%s: Part 1 = %f\n", __func__, l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer / p->ReturnBW / (p->hw_debug5 ? 1 : l->AverageDCCCompressionRate));
+	DML_LOG_VERBOSE("DML::%s: Part 2 = %f\n", __func__, (*p->StutterPeriod * p->TotalDataReadBandwidth - l->PartOfUncompressedPixelBurstThatFitsInROBAndCompressedBuffer) / (p->DCFCLK * 64));
+	DML_LOG_VERBOSE("DML::%s: Part 3 = %f\n", __func__, *p->StutterPeriod * l->TotalRowReadBandwidth / p->ReturnBW);
+	DML_LOG_VERBOSE("DML::%s: StutterBurstTime = %f\n", __func__, l->StutterBurstTime);
 #endif
 	l->TotalActiveWriteback = 0;
 	memset(l->stream_visited, 0, DML2_MAX_PLANES * sizeof(bool));
@@ -10324,9 +10252,9 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 
 	if (l->TotalActiveWriteback == 0) {
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: SRExitTime = %f\n", __func__, p->SRExitTime);
-		dml2_printf("DML::%s: SRExitZ8Time = %f\n", __func__, p->SRExitZ8Time);
-		dml2_printf("DML::%s: StutterPeriod = %f\n", __func__, *p->StutterPeriod);
+		DML_LOG_VERBOSE("DML::%s: SRExitTime = %f\n", __func__, p->SRExitTime);
+		DML_LOG_VERBOSE("DML::%s: SRExitZ8Time = %f\n", __func__, p->SRExitZ8Time);
+		DML_LOG_VERBOSE("DML::%s: StutterPeriod = %f\n", __func__, *p->StutterPeriod);
 #endif
 		*p->StutterEfficiencyNotIncludingVBlank = math_max2(0., 1 - (p->SRExitTime + l->StutterBurstTime) / *p->StutterPeriod) * 100;
 		*p->Z8StutterEfficiencyNotIncludingVBlank = math_max2(0., 1 - (p->SRExitZ8Time + l->StutterBurstTime) / *p->StutterPeriod) * 100;
@@ -10339,11 +10267,11 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 		*p->Z8NumberOfStutterBurstsPerFrame = 0;
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: VActiveTimeCriticalSurface = %f\n", __func__, l->VActiveTimeCriticalSurface);
-	dml2_printf("DML::%s: StutterEfficiencyNotIncludingVBlank = %f\n", __func__, *p->StutterEfficiencyNotIncludingVBlank);
-	dml2_printf("DML::%s: Z8StutterEfficiencyNotIncludingVBlank = %f\n", __func__, *p->Z8StutterEfficiencyNotIncludingVBlank);
-	dml2_printf("DML::%s: NumberOfStutterBurstsPerFrame = %u\n", __func__, *p->NumberOfStutterBurstsPerFrame);
-	dml2_printf("DML::%s: Z8NumberOfStutterBurstsPerFrame = %u\n", __func__, *p->Z8NumberOfStutterBurstsPerFrame);
+	DML_LOG_VERBOSE("DML::%s: VActiveTimeCriticalSurface = %f\n", __func__, l->VActiveTimeCriticalSurface);
+	DML_LOG_VERBOSE("DML::%s: StutterEfficiencyNotIncludingVBlank = %f\n", __func__, *p->StutterEfficiencyNotIncludingVBlank);
+	DML_LOG_VERBOSE("DML::%s: Z8StutterEfficiencyNotIncludingVBlank = %f\n", __func__, *p->Z8StutterEfficiencyNotIncludingVBlank);
+	DML_LOG_VERBOSE("DML::%s: NumberOfStutterBurstsPerFrame = %u\n", __func__, *p->NumberOfStutterBurstsPerFrame);
+	DML_LOG_VERBOSE("DML::%s: Z8NumberOfStutterBurstsPerFrame = %u\n", __func__, *p->Z8NumberOfStutterBurstsPerFrame);
 #endif
 
 	if (*p->StutterEfficiencyNotIncludingVBlank > 0) {
@@ -10358,7 +10286,7 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 	}
 
 	if (*p->Z8StutterEfficiencyNotIncludingVBlank > 0) {
-		LastZ8StutterPeriod = l->VActiveTimeCriticalSurface - (*p->Z8NumberOfStutterBurstsPerFrame - 1) * *p->StutterPeriod;
+		//LastZ8StutterPeriod = l->VActiveTimeCriticalSurface - (*p->Z8NumberOfStutterBurstsPerFrame - 1) * *p->StutterPeriod;
 		if (!((p->SynchronizeTimings || TotalNumberOfActiveOTG == 1) && SameTiming)) {
 			*p->Z8StutterEfficiency = *p->Z8StutterEfficiencyNotIncludingVBlank;
 		} else {
@@ -10370,25 +10298,25 @@ static void CalculateStutterEfficiency(struct dml2_core_internal_scratch *scratc
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: TotalNumberOfActiveOTG = %u\n", __func__, TotalNumberOfActiveOTG);
-	dml2_printf("DML::%s: SameTiming = %u\n", __func__, SameTiming);
-	dml2_printf("DML::%s: SynchronizeTimings = %u\n", __func__, p->SynchronizeTimings);
-	dml2_printf("DML::%s: LastZ8StutterPeriod = %f\n", __func__, LastZ8StutterPeriod);
-	dml2_printf("DML::%s: Z8StutterEnterPlusExitWatermark = %f\n", __func__, p->Z8StutterEnterPlusExitWatermark);
-	dml2_printf("DML::%s: StutterBurstTime = %f\n", __func__, l->StutterBurstTime);
-	dml2_printf("DML::%s: StutterPeriod = %f\n", __func__, *p->StutterPeriod);
-	dml2_printf("DML::%s: StutterEfficiency = %f\n", __func__, *p->StutterEfficiency);
-	dml2_printf("DML::%s: Z8StutterEfficiency = %f\n", __func__, *p->Z8StutterEfficiency);
-	dml2_printf("DML::%s: StutterEfficiencyNotIncludingVBlank = %f\n", __func__, *p->StutterEfficiencyNotIncludingVBlank);
-	dml2_printf("DML::%s: Z8NumberOfStutterBurstsPerFrame = %u\n", __func__, *p->Z8NumberOfStutterBurstsPerFrame);
+	DML_LOG_VERBOSE("DML::%s: TotalNumberOfActiveOTG = %u\n", __func__, TotalNumberOfActiveOTG);
+	DML_LOG_VERBOSE("DML::%s: SameTiming = %u\n", __func__, SameTiming);
+	DML_LOG_VERBOSE("DML::%s: SynchronizeTimings = %u\n", __func__, p->SynchronizeTimings);
+	DML_LOG_VERBOSE("DML::%s: LastZ8StutterPeriod = %f\n", __func__, *p->Z8StutterEfficiencyNotIncludingVBlank > 0 ? l->VActiveTimeCriticalSurface - (*p->Z8NumberOfStutterBurstsPerFrame - 1) * *p->StutterPeriod : 0);
+	DML_LOG_VERBOSE("DML::%s: Z8StutterEnterPlusExitWatermark = %f\n", __func__, p->Z8StutterEnterPlusExitWatermark);
+	DML_LOG_VERBOSE("DML::%s: StutterBurstTime = %f\n", __func__, l->StutterBurstTime);
+	DML_LOG_VERBOSE("DML::%s: StutterPeriod = %f\n", __func__, *p->StutterPeriod);
+	DML_LOG_VERBOSE("DML::%s: StutterEfficiency = %f\n", __func__, *p->StutterEfficiency);
+	DML_LOG_VERBOSE("DML::%s: Z8StutterEfficiency = %f\n", __func__, *p->Z8StutterEfficiency);
+	DML_LOG_VERBOSE("DML::%s: StutterEfficiencyNotIncludingVBlank = %f\n", __func__, *p->StutterEfficiencyNotIncludingVBlank);
+	DML_LOG_VERBOSE("DML::%s: Z8NumberOfStutterBurstsPerFrame = %u\n", __func__, *p->Z8NumberOfStutterBurstsPerFrame);
 #endif
 
 	*p->DCHUBBUB_ARB_CSTATE_MAX_CAP_MODE = !(!p->UnboundedRequestEnabled && (p->NumberOfActiveSurfaces == 1) && l->SinglePlaneCriticalSurface && l->SinglePipeCriticalSurface);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: DETBufferSizeYCriticalSurface = %u\n", __func__, l->DETBufferSizeYCriticalSurface);
-	dml2_printf("DML::%s: PixelChunkSizeInKByte = %u\n", __func__, p->PixelChunkSizeInKByte);
-	dml2_printf("DML::%s: DCHUBBUB_ARB_CSTATE_MAX_CAP_MODE = %u\n", __func__, *p->DCHUBBUB_ARB_CSTATE_MAX_CAP_MODE);
+	DML_LOG_VERBOSE("DML::%s: DETBufferSizeYCriticalSurface = %u\n", __func__, l->DETBufferSizeYCriticalSurface);
+	DML_LOG_VERBOSE("DML::%s: PixelChunkSizeInKByte = %u\n", __func__, p->PixelChunkSizeInKByte);
+	DML_LOG_VERBOSE("DML::%s: DCHUBBUB_ARB_CSTATE_MAX_CAP_MODE = %u\n", __func__, *p->DCHUBBUB_ARB_CSTATE_MAX_CAP_MODE);
 #endif
 }
 
@@ -10422,7 +10350,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 	double max_uclk_mhz = 0;
 	double min_return_latency_in_DCFCLK_cycles = 0;
 
-	dml2_printf("DML::%s: --- START --- \n", __func__);
+	DML_LOG_VERBOSE("DML::%s: --- START --- \n", __func__);
 
 	memset(&mode_lib->scratch, 0, sizeof(struct dml2_core_internal_scratch));
 	memset(&mode_lib->mp, 0, sizeof(struct dml2_core_internal_mode_program));
@@ -10444,13 +10372,13 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 
 	for (k = 0; k < s->num_active_planes; ++k) {
 		unsigned int stream_index = display_cfg->plane_descriptors[k].stream_index;
-		DML2_ASSERT(cfg_support_info->stream_support_info[stream_index].odms_used <= 4);
-		DML2_ASSERT(cfg_support_info->stream_support_info[stream_index].num_odm_output_segments == 4 ||
+		DML_ASSERT(cfg_support_info->stream_support_info[stream_index].odms_used <= 4);
+		DML_ASSERT(cfg_support_info->stream_support_info[stream_index].num_odm_output_segments == 4 ||
 					cfg_support_info->stream_support_info[stream_index].num_odm_output_segments == 2 ||
 					cfg_support_info->stream_support_info[stream_index].num_odm_output_segments == 1);
 
 		if (cfg_support_info->stream_support_info[stream_index].odms_used > 1)
-			DML2_ASSERT(cfg_support_info->stream_support_info[stream_index].num_odm_output_segments == 1);
+			DML_ASSERT(cfg_support_info->stream_support_info[stream_index].num_odm_output_segments == 1);
 
 		switch (cfg_support_info->stream_support_info[stream_index].odms_used) {
 		case (4):
@@ -10476,51 +10404,51 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 	for (k = 0; k < s->num_active_planes; ++k) {
 		mode_lib->mp.NoOfDPP[k] = cfg_support_info->plane_support_info[k].dpps_used;
 		mode_lib->mp.Dppclk[k] = programming->plane_programming[k].min_clocks.dcn4x.dppclk_khz / 1000.0;
-		DML2_ASSERT(mode_lib->mp.Dppclk[k] > 0);
+		DML_ASSERT(mode_lib->mp.Dppclk[k] > 0);
 	}
 
 	for (k = 0; k < s->num_active_planes; ++k) {
 		unsigned int stream_index = display_cfg->plane_descriptors[k].stream_index;
 		mode_lib->mp.DSCCLK[k] = programming->stream_programming[stream_index].min_clocks.dcn4x.dscclk_khz / 1000.0;
-		dml2_printf("DML::%s: k=%d stream_index=%d, mode_lib->mp.DSCCLK = %f\n", __func__, k, stream_index, mode_lib->mp.DSCCLK[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d stream_index=%d, mode_lib->mp.DSCCLK = %f\n", __func__, k, stream_index, mode_lib->mp.DSCCLK[k]);
 	}
 
 	mode_lib->mp.Dispclk = programming->min_clocks.dcn4x.dispclk_khz / 1000.0;
 	mode_lib->mp.DCFCLKDeepSleep = programming->min_clocks.dcn4x.deepsleep_dcfclk_khz / 1000.0;
 
-	DML2_ASSERT(mode_lib->mp.Dcfclk > 0);
-	DML2_ASSERT(mode_lib->mp.FabricClock > 0);
-	DML2_ASSERT(mode_lib->mp.dram_bw_mbps > 0);
-	DML2_ASSERT(mode_lib->mp.uclk_freq_mhz > 0);
-	DML2_ASSERT(mode_lib->mp.GlobalDPPCLK > 0);
-	DML2_ASSERT(mode_lib->mp.Dispclk > 0);
-	DML2_ASSERT(mode_lib->mp.DCFCLKDeepSleep > 0);
-	DML2_ASSERT(s->SOCCLK > 0);
-
-#ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: num_active_planes = %u\n", __func__, s->num_active_planes);
-	dml2_printf("DML::%s: num_active_pipes = %u\n", __func__, mode_lib->mp.num_active_pipes);
-	dml2_printf("DML::%s: Dcfclk = %f\n", __func__, mode_lib->mp.Dcfclk);
-	dml2_printf("DML::%s: FabricClock = %f\n", __func__, mode_lib->mp.FabricClock);
-	dml2_printf("DML::%s: dram_bw_mbps = %f\n", __func__, mode_lib->mp.dram_bw_mbps);
-	dml2_printf("DML::%s: uclk_freq_mhz = %f\n", __func__, mode_lib->mp.uclk_freq_mhz);
-	dml2_printf("DML::%s: Dispclk = %f\n", __func__, mode_lib->mp.Dispclk);
+	DML_ASSERT(mode_lib->mp.Dcfclk > 0);
+	DML_ASSERT(mode_lib->mp.FabricClock > 0);
+	DML_ASSERT(mode_lib->mp.dram_bw_mbps > 0);
+	DML_ASSERT(mode_lib->mp.uclk_freq_mhz > 0);
+	DML_ASSERT(mode_lib->mp.GlobalDPPCLK > 0);
+	DML_ASSERT(mode_lib->mp.Dispclk > 0);
+	DML_ASSERT(mode_lib->mp.DCFCLKDeepSleep > 0);
+	DML_ASSERT(s->SOCCLK > 0);
+
+#ifdef __DML_VBA_DEBUG__
+	DML_LOG_VERBOSE("DML::%s: num_active_planes = %u\n", __func__, s->num_active_planes);
+	DML_LOG_VERBOSE("DML::%s: num_active_pipes = %u\n", __func__, mode_lib->mp.num_active_pipes);
+	DML_LOG_VERBOSE("DML::%s: Dcfclk = %f\n", __func__, mode_lib->mp.Dcfclk);
+	DML_LOG_VERBOSE("DML::%s: FabricClock = %f\n", __func__, mode_lib->mp.FabricClock);
+	DML_LOG_VERBOSE("DML::%s: dram_bw_mbps = %f\n", __func__, mode_lib->mp.dram_bw_mbps);
+	DML_LOG_VERBOSE("DML::%s: uclk_freq_mhz = %f\n", __func__, mode_lib->mp.uclk_freq_mhz);
+	DML_LOG_VERBOSE("DML::%s: Dispclk = %f\n", __func__, mode_lib->mp.Dispclk);
 	for (k = 0; k < s->num_active_planes; ++k) {
-		dml2_printf("DML::%s: Dppclk[%0d] = %f\n", __func__, k, mode_lib->mp.Dppclk[k]);
-	}
-	dml2_printf("DML::%s: GlobalDPPCLK = %f\n", __func__, mode_lib->mp.GlobalDPPCLK);
-	dml2_printf("DML::%s: DCFCLKDeepSleep = %f\n", __func__, mode_lib->mp.DCFCLKDeepSleep);
-	dml2_printf("DML::%s: SOCCLK = %f\n", __func__, s->SOCCLK);
-	dml2_printf("DML::%s: min_clk_index = %0d\n", __func__, in_out_params->min_clk_index);
-	dml2_printf("DML::%s: min_clk_table min_fclk_khz = %d\n", __func__, min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].min_fclk_khz);
-	dml2_printf("DML::%s: min_clk_table uclk_mhz = %f\n", __func__, dram_bw_kbps_to_uclk_mhz(min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].pre_derate_dram_bw_kbps, &mode_lib->soc.clk_table.dram_config));
+		DML_LOG_VERBOSE("DML::%s: Dppclk[%0d] = %f\n", __func__, k, mode_lib->mp.Dppclk[k]);
+	}
+	DML_LOG_VERBOSE("DML::%s: GlobalDPPCLK = %f\n", __func__, mode_lib->mp.GlobalDPPCLK);
+	DML_LOG_VERBOSE("DML::%s: DCFCLKDeepSleep = %f\n", __func__, mode_lib->mp.DCFCLKDeepSleep);
+	DML_LOG_VERBOSE("DML::%s: SOCCLK = %f\n", __func__, s->SOCCLK);
+	DML_LOG_VERBOSE("DML::%s: min_clk_index = %0d\n", __func__, in_out_params->min_clk_index);
+	DML_LOG_VERBOSE("DML::%s: min_clk_table min_fclk_khz = %ld\n", __func__, min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].min_fclk_khz);
+	DML_LOG_VERBOSE("DML::%s: min_clk_table uclk_mhz = %f\n", __func__, dram_bw_kbps_to_uclk_mhz(min_clk_table->dram_bw_table.entries[in_out_params->min_clk_index].pre_derate_dram_bw_kbps, &mode_lib->soc.clk_table.dram_config));
 	for (k = 0; k < mode_lib->mp.num_active_pipes; ++k) {
-		dml2_printf("DML::%s: pipe=%d is in plane=%d\n", __func__, k, mode_lib->mp.pipe_plane[k]);
-		dml2_printf("DML::%s: Per-plane DPPPerSurface[%0d] = %d\n", __func__, k, mode_lib->mp.NoOfDPP[k]);
+		DML_LOG_VERBOSE("DML::%s: pipe=%d is in plane=%d\n", __func__, k, mode_lib->mp.pipe_plane[k]);
+		DML_LOG_VERBOSE("DML::%s: Per-plane DPPPerSurface[%0d] = %d\n", __func__, k, mode_lib->mp.NoOfDPP[k]);
 	}
 
 	for (k = 0; k < s->num_active_planes; k++)
-		dml2_printf("DML::%s: plane_%d: reserved_vblank_time_ns = %u\n", __func__, k, display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns);
+		DML_LOG_VERBOSE("DML::%s: plane_%d: reserved_vblank_time_ns = %lu\n", __func__, k, display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns);
 #endif
 
 	CalculateMaxDETAndMinCompressedBufferSize(
@@ -10617,8 +10545,8 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 			((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000));
 		mode_lib->mp.vactive_sw_bw_l[k] = mode_lib->mp.SwathWidthSingleDPPY[k] * mode_lib->mp.BytePerPixelY[k] / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) * display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio;
 		mode_lib->mp.vactive_sw_bw_c[k] = mode_lib->mp.SwathWidthSingleDPPC[k] * mode_lib->mp.BytePerPixelC[k] / (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000)) * display_cfg->plane_descriptors[k].composition.scaler_info.plane1.v_ratio;
-		dml2_printf("DML::%s: vactive_sw_bw_l[%i] = %fBps\n", __func__, k, mode_lib->mp.vactive_sw_bw_l[k]);
-		dml2_printf("DML::%s: vactive_sw_bw_c[%i] = %fBps\n", __func__, k, mode_lib->mp.vactive_sw_bw_c[k]);
+		DML_LOG_VERBOSE("DML::%s: vactive_sw_bw_l[%i] = %fBps\n", __func__, k, mode_lib->mp.vactive_sw_bw_l[k]);
+		DML_LOG_VERBOSE("DML::%s: vactive_sw_bw_c[%i] = %fBps\n", __func__, k, mode_lib->mp.vactive_sw_bw_c[k]);
 	}
 
 	CalculateSwathAndDETConfiguration_params->display_cfg = display_cfg;
@@ -11097,7 +11025,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 		mode_lib->soc.qos_parameters.qos_params.dcn4x.fabric_max_transport_latency_margin);
 
 	for (k = 0; k < s->num_active_planes; ++k) {
-		bool cursor_not_enough_urgent_latency_hiding = 0;
+		bool cursor_not_enough_urgent_latency_hiding = false;
 		s->line_times[k] = display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total /
 			((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
 
@@ -11173,8 +11101,8 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 			mode_lib->mp.WritebackDelay[k]);
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%u MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
-		dml2_printf("DML::%s: k=%u WritebackDelay = %f\n", __func__, k, mode_lib->mp.WritebackDelay[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%u WritebackDelay = %f\n", __func__, k, mode_lib->mp.WritebackDelay[k]);
 #endif
 	}
 
@@ -11183,7 +11111,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 		s->immediate_flip_required = s->immediate_flip_required || display_cfg->plane_descriptors[k].immediate_flip;
 	}
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: immediate_flip_required = %u\n", __func__, s->immediate_flip_required);
+	DML_LOG_VERBOSE("DML::%s: immediate_flip_required = %u\n", __func__, s->immediate_flip_required);
 #endif
 
 	if (s->num_active_planes > 1) {
@@ -11219,12 +11147,12 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 		s->DestinationLineTimesForPrefetchLessThan2 = false;
 		s->VRatioPrefetchMoreThanMax = false;
 
-		dml2_printf("DML::%s: Start one iteration of prefetch schedule evaluation\n", __func__);
+		DML_LOG_VERBOSE("DML::%s: Start one iteration of prefetch schedule evaluation\n", __func__);
 
 		for (k = 0; k < s->num_active_planes; ++k) {
 			struct dml2_core_internal_DmlPipe *myPipe = &s->myPipe;
 
-			dml2_printf("DML::%s: k=%d MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%d MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
 			mode_lib->mp.TWait[k] = CalculateTWait(
 					display_cfg->plane_descriptors[k].overrides.reserved_vblank_time_ns,
 					mode_lib->mp.UrgentLatency,
@@ -11261,7 +11189,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 			myPipe->ProgressiveToInterlaceUnitInOPP = mode_lib->ip.ptoi_supported;
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: Calling CalculatePrefetchSchedule for k=%u\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: Calling CalculatePrefetchSchedule for k=%u\n", __func__, k);
 #endif
 			CalculatePrefetchSchedule_params->display_cfg = display_cfg;
 			CalculatePrefetchSchedule_params->HostVMInefficiencyFactor = s->HostVMInefficiencyFactorPrefetch;
@@ -11356,7 +11284,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 				mode_lib->mp.impacted_prefetch_margin_us[k] = 0;
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%0u NoTimeToPrefetch=%0d\n", __func__, k, mode_lib->mp.NoTimeToPrefetch[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%0u NoTimeToPrefetch=%0d\n", __func__, k, mode_lib->mp.NoTimeToPrefetch[k]);
 #endif
 			mode_lib->mp.VStartupMin[k] = s->MaxVStartupLines[k];
 		} // for k
@@ -11366,9 +11294,9 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 			if (mode_lib->mp.NoTimeToPrefetch[k] == true ||
 				mode_lib->mp.NotEnoughTimeForDynamicMetadata[k] ||
 				mode_lib->mp.DSTYAfterScaler[k] > 8) {
-				dml2_printf("DML::%s: k=%u, NoTimeToPrefetch = %0d\n", __func__, k, mode_lib->mp.NoTimeToPrefetch[k]);
-				dml2_printf("DML::%s: k=%u, NotEnoughTimeForDynamicMetadata=%u\n", __func__, k, mode_lib->mp.NotEnoughTimeForDynamicMetadata[k]);
-				dml2_printf("DML::%s: k=%u, DSTYAfterScaler=%u (should be <= 0)\n", __func__, k, mode_lib->mp.DSTYAfterScaler[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, NoTimeToPrefetch = %0d\n", __func__, k, mode_lib->mp.NoTimeToPrefetch[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, NotEnoughTimeForDynamicMetadata=%u\n", __func__, k, mode_lib->mp.NotEnoughTimeForDynamicMetadata[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, DSTYAfterScaler=%u (should be <= 0)\n", __func__, k, mode_lib->mp.DSTYAfterScaler[k]);
 				mode_lib->mp.PrefetchModeSupported = false;
 			}
 			if (mode_lib->mp.dst_y_prefetch[k] < 2)
@@ -11377,24 +11305,24 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 			if (mode_lib->mp.VRatioPrefetchY[k] > __DML2_CALCS_MAX_VRATIO_PRE__ ||
 				mode_lib->mp.VRatioPrefetchC[k] > __DML2_CALCS_MAX_VRATIO_PRE__) {
 				s->VRatioPrefetchMoreThanMax = true;
-				dml2_printf("DML::%s: k=%d, VRatioPrefetchY=%f (should not be < %f)\n", __func__, k, mode_lib->mp.VRatioPrefetchY[k], __DML2_CALCS_MAX_VRATIO_PRE__);
-				dml2_printf("DML::%s: k=%d, VRatioPrefetchC=%f (should not be < %f)\n", __func__, k, mode_lib->mp.VRatioPrefetchC[k], __DML2_CALCS_MAX_VRATIO_PRE__);
-				dml2_printf("DML::%s: VRatioPrefetchMoreThanMax = %u\n", __func__, s->VRatioPrefetchMoreThanMax);
+				DML_LOG_VERBOSE("DML::%s: k=%d, VRatioPrefetchY=%f (should not be < %f)\n", __func__, k, mode_lib->mp.VRatioPrefetchY[k], __DML2_CALCS_MAX_VRATIO_PRE__);
+				DML_LOG_VERBOSE("DML::%s: k=%d, VRatioPrefetchC=%f (should not be < %f)\n", __func__, k, mode_lib->mp.VRatioPrefetchC[k], __DML2_CALCS_MAX_VRATIO_PRE__);
+				DML_LOG_VERBOSE("DML::%s: VRatioPrefetchMoreThanMax = %u\n", __func__, s->VRatioPrefetchMoreThanMax);
 			}
 
 			if (mode_lib->mp.NotEnoughUrgentLatencyHiding[k]) {
-				dml2_printf("DML::%s: k=%u, NotEnoughUrgentLatencyHiding = %u\n", __func__, k, mode_lib->mp.NotEnoughUrgentLatencyHiding[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%u, NotEnoughUrgentLatencyHiding = %u\n", __func__, k, mode_lib->mp.NotEnoughUrgentLatencyHiding[k]);
 				mode_lib->mp.PrefetchModeSupported = false;
 			}
 		}
 
 		if (s->VRatioPrefetchMoreThanMax == true || s->DestinationLineTimesForPrefetchLessThan2 == true) {
-			dml2_printf("DML::%s: VRatioPrefetchMoreThanMax = %u\n", __func__, s->VRatioPrefetchMoreThanMax);
-			dml2_printf("DML::%s: DestinationLineTimesForPrefetchLessThan2 = %u\n", __func__, s->DestinationLineTimesForPrefetchLessThan2);
+			DML_LOG_VERBOSE("DML::%s: VRatioPrefetchMoreThanMax = %u\n", __func__, s->VRatioPrefetchMoreThanMax);
+			DML_LOG_VERBOSE("DML::%s: DestinationLineTimesForPrefetchLessThan2 = %u\n", __func__, s->DestinationLineTimesForPrefetchLessThan2);
 			mode_lib->mp.PrefetchModeSupported = false;
 		}
 
-		dml2_printf("DML::%s: Prefetch schedule is %sOK at vstartup = %u\n", __func__,
+		DML_LOG_VERBOSE("DML::%s: Prefetch schedule is %sOK at vstartup = %u\n", __func__,
 			mode_lib->mp.PrefetchModeSupported ? "" : "NOT ", CalculatePrefetchSchedule_params->VStartup);
 
 		// Prefetch schedule OK, now check prefetch bw
@@ -11422,24 +11350,24 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 					&mode_lib->mp.NotEnoughUrgentLatencyHidingPre[k]);
 
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k=%0u DPPPerSurface=%u\n", __func__, k, mode_lib->mp.NoOfDPP[k]);
-				dml2_printf("DML::%s: k=%0u UrgentBurstFactorLuma=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorLuma[k]);
-				dml2_printf("DML::%s: k=%0u UrgentBurstFactorChroma=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorChroma[k]);
-				dml2_printf("DML::%s: k=%0u UrgentBurstFactorLumaPre=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorLumaPre[k]);
-				dml2_printf("DML::%s: k=%0u UrgentBurstFactorChromaPre=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorChromaPre[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u DPPPerSurface=%u\n", __func__, k, mode_lib->mp.NoOfDPP[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u UrgentBurstFactorLuma=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorLuma[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u UrgentBurstFactorChroma=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorChroma[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u UrgentBurstFactorLumaPre=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorLumaPre[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u UrgentBurstFactorChromaPre=%f\n", __func__, k, mode_lib->mp.UrgentBurstFactorChromaPre[k]);
 
-				dml2_printf("DML::%s: k=%0u VRatioPrefetchY=%f\n", __func__, k, mode_lib->mp.VRatioPrefetchY[k]);
-				dml2_printf("DML::%s: k=%0u VRatioY=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
+				DML_LOG_VERBOSE("DML::%s: k=%0u VRatioPrefetchY=%f\n", __func__, k, mode_lib->mp.VRatioPrefetchY[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u VRatioY=%f\n", __func__, k, display_cfg->plane_descriptors[k].composition.scaler_info.plane0.v_ratio);
 
-				dml2_printf("DML::%s: k=%0u prefetch_vmrow_bw=%f\n", __func__, k, mode_lib->mp.prefetch_vmrow_bw[k]);
-				dml2_printf("DML::%s: k=%0u vactive_sw_bw_l=%f\n", __func__, k, mode_lib->mp.vactive_sw_bw_l[k]);
-				dml2_printf("DML::%s: k=%0u vactive_sw_bw_c=%f\n", __func__, k, mode_lib->mp.vactive_sw_bw_c[k]);
-				dml2_printf("DML::%s: k=%0u cursor_bw=%f\n", __func__, k, mode_lib->mp.cursor_bw[k]);
-				dml2_printf("DML::%s: k=%0u dpte_row_bw=%f\n", __func__, k, mode_lib->mp.dpte_row_bw[k]);
-				dml2_printf("DML::%s: k=%0u meta_row_bw=%f\n", __func__, k, mode_lib->mp.meta_row_bw[k]);
-				dml2_printf("DML::%s: k=%0u RequiredPrefetchPixelDataBWLuma=%f\n", __func__, k, mode_lib->mp.RequiredPrefetchPixelDataBWLuma[k]);
-				dml2_printf("DML::%s: k=%0u RequiredPrefetchPixelDataBWChroma=%f\n", __func__, k, mode_lib->mp.RequiredPrefetchPixelDataBWChroma[k]);
-				dml2_printf("DML::%s: k=%0u prefetch_cursor_bw=%f\n", __func__, k, mode_lib->mp.prefetch_cursor_bw[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u prefetch_vmrow_bw=%f\n", __func__, k, mode_lib->mp.prefetch_vmrow_bw[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u vactive_sw_bw_l=%f\n", __func__, k, mode_lib->mp.vactive_sw_bw_l[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u vactive_sw_bw_c=%f\n", __func__, k, mode_lib->mp.vactive_sw_bw_c[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u cursor_bw=%f\n", __func__, k, mode_lib->mp.cursor_bw[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u dpte_row_bw=%f\n", __func__, k, mode_lib->mp.dpte_row_bw[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u meta_row_bw=%f\n", __func__, k, mode_lib->mp.meta_row_bw[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u RequiredPrefetchPixelDataBWLuma=%f\n", __func__, k, mode_lib->mp.RequiredPrefetchPixelDataBWLuma[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u RequiredPrefetchPixelDataBWChroma=%f\n", __func__, k, mode_lib->mp.RequiredPrefetchPixelDataBWChroma[k]);
+				DML_LOG_VERBOSE("DML::%s: k=%0u prefetch_cursor_bw=%f\n", __func__, k, mode_lib->mp.prefetch_cursor_bw[k]);
 #endif
 			}
 
@@ -11503,11 +11431,11 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 				mode_lib->mp.urg_bandwidth_available);
 
 			if (!mode_lib->mp.PrefetchModeSupported)
-				dml2_printf("DML::%s: Bandwidth not sufficient for prefetch!\n", __func__);
+				DML_LOG_VERBOSE("DML::%s: Bandwidth not sufficient for prefetch!\n", __func__);
 
 			for (k = 0; k < s->num_active_planes; ++k) {
 				if (mode_lib->mp.NotEnoughUrgentLatencyHidingPre[k]) {
-					dml2_printf("DML::%s: k=%u, NotEnoughUrgentLatencyHidingPre = %u\n", __func__, k, mode_lib->mp.NotEnoughUrgentLatencyHidingPre[k]);
+					DML_LOG_VERBOSE("DML::%s: k=%u, NotEnoughUrgentLatencyHidingPre = %u\n", __func__, k, mode_lib->mp.NotEnoughUrgentLatencyHidingPre[k]);
 					mode_lib->mp.PrefetchModeSupported = false;
 				}
 			}
@@ -11533,12 +11461,12 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 				}
 				mode_lib->mp.TotImmediateFlipBytes += s->per_pipe_flip_bytes[k] * mode_lib->mp.NoOfDPP[k];
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML::%s: k = %u\n", __func__, k);
-				dml2_printf("DML::%s: DPPPerSurface = %u\n", __func__, mode_lib->mp.NoOfDPP[k]);
-				dml2_printf("DML::%s: vm_bytes = %u\n", __func__, mode_lib->mp.vm_bytes[k]);
-				dml2_printf("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, mode_lib->mp.PixelPTEBytesPerRow[k]);
-				dml2_printf("DML::%s: meta_row_bytes = %u\n", __func__, mode_lib->mp.meta_row_bytes[k]);
-				dml2_printf("DML::%s: TotImmediateFlipBytes = %u\n", __func__, mode_lib->mp.TotImmediateFlipBytes);
+				DML_LOG_VERBOSE("DML::%s: k = %u\n", __func__, k);
+				DML_LOG_VERBOSE("DML::%s: DPPPerSurface = %u\n", __func__, mode_lib->mp.NoOfDPP[k]);
+				DML_LOG_VERBOSE("DML::%s: vm_bytes = %u\n", __func__, mode_lib->mp.vm_bytes[k]);
+				DML_LOG_VERBOSE("DML::%s: PixelPTEBytesPerRow = %u\n", __func__, mode_lib->mp.PixelPTEBytesPerRow[k]);
+				DML_LOG_VERBOSE("DML::%s: meta_row_bytes = %u\n", __func__, mode_lib->mp.meta_row_bytes[k]);
+				DML_LOG_VERBOSE("DML::%s: TotImmediateFlipBytes = %u\n", __func__, mode_lib->mp.TotImmediateFlipBytes);
 #endif
 			}
 			for (k = 0; k < s->num_active_planes; ++k) {
@@ -11631,13 +11559,13 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 				mode_lib->mp.urg_bandwidth_available);
 
 			if (!mode_lib->mp.ImmediateFlipSupported)
-				dml2_printf("DML::%s: Bandwidth not sufficient for flip!", __func__);
+				DML_LOG_VERBOSE("DML::%s: Bandwidth not sufficient for flip!", __func__);
 
 			for (k = 0; k < s->num_active_planes; ++k) {
 				if (display_cfg->plane_descriptors[k].immediate_flip && mode_lib->mp.ImmediateFlipSupportedForPipe[k] == false) {
 					mode_lib->mp.ImmediateFlipSupported = false;
 #ifdef __DML_VBA_DEBUG__
-					dml2_printf("DML::%s: Pipe %0d not supporting iflip!\n", __func__, k);
+					DML_LOG_VERBOSE("DML::%s: Pipe %0d not supporting iflip!\n", __func__, k);
 #endif
 				}
 			}
@@ -11650,28 +11578,28 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 		mode_lib->mp.PrefetchAndImmediateFlipSupported = (mode_lib->mp.PrefetchModeSupported == true && (!must_support_iflip || mode_lib->mp.ImmediateFlipSupported));
 
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: PrefetchModeSupported = %u\n", __func__, mode_lib->mp.PrefetchModeSupported);
+		DML_LOG_VERBOSE("DML::%s: PrefetchModeSupported = %u\n", __func__, mode_lib->mp.PrefetchModeSupported);
 		for (k = 0; k < s->num_active_planes; ++k)
-			dml2_printf("DML::%s: immediate_flip_required[%u] = %u\n", __func__, k, display_cfg->plane_descriptors[k].immediate_flip);
-		dml2_printf("DML::%s: HostVMEnable = %u\n", __func__, display_cfg->hostvm_enable);
-		dml2_printf("DML::%s: ImmediateFlipSupported = %u\n", __func__, mode_lib->mp.ImmediateFlipSupported);
-		dml2_printf("DML::%s: PrefetchAndImmediateFlipSupported = %u\n", __func__, mode_lib->mp.PrefetchAndImmediateFlipSupported);
+			DML_LOG_VERBOSE("DML::%s: immediate_flip_required[%u] = %u\n", __func__, k, display_cfg->plane_descriptors[k].immediate_flip);
+		DML_LOG_VERBOSE("DML::%s: HostVMEnable = %u\n", __func__, display_cfg->hostvm_enable);
+		DML_LOG_VERBOSE("DML::%s: ImmediateFlipSupported = %u\n", __func__, mode_lib->mp.ImmediateFlipSupported);
+		DML_LOG_VERBOSE("DML::%s: PrefetchAndImmediateFlipSupported = %u\n", __func__, mode_lib->mp.PrefetchAndImmediateFlipSupported);
 #endif
-		dml2_printf("DML::%s: Done one iteration: k=%d, MaxVStartupLines=%u\n", __func__, k, s->MaxVStartupLines[k]);
+		DML_LOG_VERBOSE("DML::%s: Done one iteration: k=%d, MaxVStartupLines=%u\n", __func__, k, s->MaxVStartupLines[k]);
 	}
 
 	for (k = 0; k < s->num_active_planes; ++k)
-		dml2_printf("DML::%s: k=%d MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
 
 	if (!mode_lib->mp.PrefetchAndImmediateFlipSupported) {
-		dml2_printf("DML::%s: Bad, Prefetch and flip scheduling solution NOT found!\n", __func__);
+		DML_LOG_VERBOSE("DML::%s: Bad, Prefetch and flip scheduling solution NOT found!\n", __func__);
 	} else {
-		dml2_printf("DML::%s: Good, Prefetch and flip scheduling solution found\n", __func__);
+		DML_LOG_VERBOSE("DML::%s: Good, Prefetch and flip scheduling solution found\n", __func__);
 
 		// DCC Configuration
 		for (k = 0; k < s->num_active_planes; ++k) {
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: Calculate DCC configuration for surface k=%u\n", __func__, k);
+			DML_LOG_VERBOSE("DML::%s: Calculate DCC configuration for surface k=%u\n", __func__, k);
 #endif
 			CalculateDCCConfiguration(
 				display_cfg->plane_descriptors[k].surface.dcc.enable,
@@ -11780,8 +11708,8 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 
 		calculate_pstate_keepout_dst_lines(display_cfg, &mode_lib->mp.Watermark, mode_lib->mp.pstate_keepout_dst_lines);
 
-		dml2_printf("DML::%s: DEBUG stream_index = %0d\n", __func__, display_cfg->plane_descriptors[0].stream_index);
-		dml2_printf("DML::%s: DEBUG PixelClock = %d kHz\n", __func__, (display_cfg->stream_descriptors[display_cfg->plane_descriptors[0].stream_index].timing.pixel_clock_khz));
+		DML_LOG_VERBOSE("DML::%s: DEBUG stream_index = %0d\n", __func__, display_cfg->plane_descriptors[0].stream_index);
+		DML_LOG_VERBOSE("DML::%s: DEBUG PixelClock = %ld kHz\n", __func__, (display_cfg->stream_descriptors[display_cfg->plane_descriptors[0].stream_index].timing.pixel_clock_khz));
 
 		//Display Pipeline Delivery Time in Prefetch, Groups
 		CalculatePixelDeliveryTimes(
@@ -11893,15 +11821,15 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 				mode_lib->mp.MinTTUVBlank[k] = mode_lib->mp.TCalc + mode_lib->mp.MinTTUVBlank[k];
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, MinTTUVBlank = %f (before vstartup margin)\n", __func__, k, mode_lib->mp.MinTTUVBlank[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, MinTTUVBlank = %f (before vstartup margin)\n", __func__, k, mode_lib->mp.MinTTUVBlank[k]);
 #endif
 			s->Tvstartup_margin = (s->MaxVStartupLines[k] - mode_lib->mp.VStartupMin[k]) * display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total / ((double)display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.pixel_clock_khz / 1000);
 			mode_lib->mp.MinTTUVBlank[k] = mode_lib->mp.MinTTUVBlank[k] + s->Tvstartup_margin;
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, Tvstartup_margin = %f\n", __func__, k, s->Tvstartup_margin);
-			dml2_printf("DML::%s: k=%u, MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
-			dml2_printf("DML::%s: k=%u, MinTTUVBlank = %f\n", __func__, k, mode_lib->mp.MinTTUVBlank[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, Tvstartup_margin = %f\n", __func__, k, s->Tvstartup_margin);
+			DML_LOG_VERBOSE("DML::%s: k=%u, MaxVStartupLines = %u\n", __func__, k, s->MaxVStartupLines[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, MinTTUVBlank = %f\n", __func__, k, mode_lib->mp.MinTTUVBlank[k]);
 #endif
 
 			mode_lib->mp.Tdmdl[k] = mode_lib->mp.Tdmdl[k] + s->Tvstartup_margin;
@@ -11920,9 +11848,9 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 			s->blank_lines_remaining = (display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_total - display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_active) - mode_lib->mp.VStartup[k];
 
 			if (s->blank_lines_remaining < 0) {
-				dml2_printf("ERROR: Vstartup is larger than vblank!?\n");
+				DML_LOG_VERBOSE("ERROR: Vstartup is larger than vblank!?\n");
 				s->blank_lines_remaining = 0;
-				DML2_ASSERT(0);
+				DML_ASSERT(0);
 			}
 			mode_lib->mp.MIN_DST_Y_NEXT_START[k] = s->dlg_vblank_start + s->blank_lines_remaining + s->LSetup;
 
@@ -11936,18 +11864,18 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 				mode_lib->mp.VREADY_AT_OR_AFTER_VSYNC[k] = false;
 			}
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, VStartup = %u (max)\n", __func__, k, mode_lib->mp.VStartup[k]);
-			dml2_printf("DML::%s: k=%u, VStartupMin = %u (max)\n", __func__, k, mode_lib->mp.VStartupMin[k]);
-			dml2_printf("DML::%s: k=%u, VUpdateOffsetPix = %u\n", __func__, k, mode_lib->mp.VUpdateOffsetPix[k]);
-			dml2_printf("DML::%s: k=%u, VUpdateWidthPix = %u\n", __func__, k, mode_lib->mp.VUpdateWidthPix[k]);
-			dml2_printf("DML::%s: k=%u, VReadyOffsetPix = %u\n", __func__, k, mode_lib->mp.VReadyOffsetPix[k]);
-			dml2_printf("DML::%s: k=%u, HTotal = %u\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total);
-			dml2_printf("DML::%s: k=%u, VTotal = %u\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_total);
-			dml2_printf("DML::%s: k=%u, VActive = %u\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_active);
-			dml2_printf("DML::%s: k=%u, VFrontPorch = %u\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_front_porch);
-			dml2_printf("DML::%s: k=%u, TSetup = %f\n", __func__, k, mode_lib->mp.TSetup[k]);
-			dml2_printf("DML::%s: k=%u, MIN_DST_Y_NEXT_START = %f\n", __func__, k, mode_lib->mp.MIN_DST_Y_NEXT_START[k]);
-			dml2_printf("DML::%s: k=%u, VREADY_AT_OR_AFTER_VSYNC = %u\n", __func__, k, mode_lib->mp.VREADY_AT_OR_AFTER_VSYNC[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VStartup = %u (max)\n", __func__, k, mode_lib->mp.VStartup[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VStartupMin = %u (max)\n", __func__, k, mode_lib->mp.VStartupMin[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VUpdateOffsetPix = %u\n", __func__, k, mode_lib->mp.VUpdateOffsetPix[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VUpdateWidthPix = %u\n", __func__, k, mode_lib->mp.VUpdateWidthPix[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VReadyOffsetPix = %u\n", __func__, k, mode_lib->mp.VReadyOffsetPix[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, HTotal = %lu\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.h_total);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VTotal = %lu\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_total);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VActive = %lu\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_active);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VFrontPorch = %lu\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.v_front_porch);
+			DML_LOG_VERBOSE("DML::%s: k=%u, TSetup = %f\n", __func__, k, mode_lib->mp.TSetup[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, MIN_DST_Y_NEXT_START = %f\n", __func__, k, mode_lib->mp.MIN_DST_Y_NEXT_START[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, VREADY_AT_OR_AFTER_VSYNC = %u\n", __func__, k, mode_lib->mp.VREADY_AT_OR_AFTER_VSYNC[k]);
 #endif
 		}
 
@@ -11969,9 +11897,9 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 		for (k = 0; k < s->num_active_planes; ++k) {
 			mode_lib->mp.TotalDataReadBandwidth = mode_lib->mp.TotalDataReadBandwidth + mode_lib->mp.vactive_sw_bw_l[k] + mode_lib->mp.vactive_sw_bw_c[k];
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML::%s: k=%u, TotalDataReadBandwidth = %f\n", __func__, k, mode_lib->mp.TotalDataReadBandwidth);
-			dml2_printf("DML::%s: k=%u, vactive_sw_bw_l = %f\n", __func__, k, mode_lib->mp.vactive_sw_bw_l[k]);
-			dml2_printf("DML::%s: k=%u, vactive_sw_bw_c = %f\n", __func__, k, mode_lib->mp.vactive_sw_bw_c[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, TotalDataReadBandwidth = %f\n", __func__, k, mode_lib->mp.TotalDataReadBandwidth);
+			DML_LOG_VERBOSE("DML::%s: k=%u, vactive_sw_bw_l = %f\n", __func__, k, mode_lib->mp.vactive_sw_bw_l[k]);
+			DML_LOG_VERBOSE("DML::%s: k=%u, vactive_sw_bw_c = %f\n", __func__, k, mode_lib->mp.vactive_sw_bw_c[k]);
 #endif
 		}
 
@@ -12051,28 +11979,28 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex
 	min_return_latency_in_DCFCLK_cycles = (min_return_uclk_cycles / max_uclk_mhz + min_return_fclk_cycles / max_fclk_mhz) * hard_minimum_dcfclk_mhz;
 	mode_lib->mp.min_return_latency_in_dcfclk = (unsigned int)min_return_latency_in_DCFCLK_cycles;
 	mode_lib->mp.dcfclk_deep_sleep_hysteresis = (unsigned int)math_max2(32, (double)mode_lib->ip.pixel_chunk_size_kbytes * 1024 * 3 / 4 / 64 - min_return_latency_in_DCFCLK_cycles);
-	DML2_ASSERT(mode_lib->mp.dcfclk_deep_sleep_hysteresis < 256);
+	DML_ASSERT(mode_lib->mp.dcfclk_deep_sleep_hysteresis < 256);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: max_fclk_mhz = %f\n", __func__, max_fclk_mhz);
-	dml2_printf("DML::%s: max_uclk_mhz = %f\n", __func__, max_uclk_mhz);
-	dml2_printf("DML::%s: hard_minimum_dcfclk_mhz = %f\n", __func__, hard_minimum_dcfclk_mhz);
-	dml2_printf("DML::%s: min_return_uclk_cycles = %d\n", __func__, min_return_uclk_cycles);
-	dml2_printf("DML::%s: min_return_fclk_cycles = %d\n", __func__, min_return_fclk_cycles);
-	dml2_printf("DML::%s: min_return_latency_in_DCFCLK_cycles = %f\n", __func__, min_return_latency_in_DCFCLK_cycles);
-	dml2_printf("DML::%s: dcfclk_deep_sleep_hysteresis = %d \n", __func__, mode_lib->mp.dcfclk_deep_sleep_hysteresis);
-	dml2_printf("DML::%s: --- END --- \n", __func__);
+	DML_LOG_VERBOSE("DML::%s: max_fclk_mhz = %f\n", __func__, max_fclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: max_uclk_mhz = %f\n", __func__, max_uclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: hard_minimum_dcfclk_mhz = %f\n", __func__, hard_minimum_dcfclk_mhz);
+	DML_LOG_VERBOSE("DML::%s: min_return_uclk_cycles = %ld\n", __func__, min_return_uclk_cycles);
+	DML_LOG_VERBOSE("DML::%s: min_return_fclk_cycles = %ld\n", __func__, min_return_fclk_cycles);
+	DML_LOG_VERBOSE("DML::%s: min_return_latency_in_DCFCLK_cycles = %f\n", __func__, min_return_latency_in_DCFCLK_cycles);
+	DML_LOG_VERBOSE("DML::%s: dcfclk_deep_sleep_hysteresis = %d \n", __func__, mode_lib->mp.dcfclk_deep_sleep_hysteresis);
+	DML_LOG_VERBOSE("DML::%s: --- END --- \n", __func__);
 #endif
 	return (in_out_params->mode_lib->mp.PrefetchAndImmediateFlipSupported);
 }
 
 bool dml2_core_calcs_mode_programming_ex(struct dml2_core_calcs_mode_programming_ex *in_out_params)
 {
-	dml2_printf("DML::%s: ------------- START ----------\n", __func__);
+	DML_LOG_VERBOSE("DML::%s: ------------- START ----------\n", __func__);
 	bool result = dml_core_mode_programming(in_out_params);
 
-	dml2_printf("DML::%s: result = %0d\n", __func__, result);
-	dml2_printf("DML::%s: ------------- DONE ----------\n", __func__);
+	DML_LOG_VERBOSE("DML::%s: result = %0d\n", __func__, result);
+	DML_LOG_VERBOSE("DML::%s: ------------- DONE ----------\n", __func__);
 	return result;
 }
 
@@ -12130,16 +12058,16 @@ void dml2_core_calcs_get_dpte_row_height(
 	unsigned int MacroTileHeight		= is_plane1 ? MacroTileHeightC : MacroTileHeightY;
 	unsigned int PTEBufferSizeInRequests = is_plane1 ? mode_lib->ip.dpte_buffer_size_in_pte_reqs_chroma : mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma;
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML: %s: is_plane1 = %u\n", __func__, is_plane1);
-	dml2_printf("DML: %s: BytePerPixel = %u\n", __func__, BytePerPixel);
-	dml2_printf("DML: %s: BlockHeight256Bytes = %u\n", __func__, BlockHeight256Bytes);
-	dml2_printf("DML: %s: BlockWidth256Bytes = %u\n", __func__, BlockWidth256Bytes);
-	dml2_printf("DML: %s: MacroTileWidth = %u\n", __func__, MacroTileWidth);
-	dml2_printf("DML: %s: MacroTileHeight = %u\n", __func__, MacroTileHeight);
-	dml2_printf("DML: %s: PTEBufferSizeInRequests = %u\n", __func__, PTEBufferSizeInRequests);
-	dml2_printf("DML: %s: dpte_buffer_size_in_pte_reqs_luma = %u\n", __func__, mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma);
-	dml2_printf("DML: %s: dpte_buffer_size_in_pte_reqs_chroma = %u\n", __func__, mode_lib->ip.dpte_buffer_size_in_pte_reqs_chroma);
-	dml2_printf("DML: %s: GPUVMMinPageSizeKBytes = %u\n", __func__, GPUVMMinPageSizeKBytes);
+	DML_LOG_VERBOSE("DML: %s: is_plane1 = %u\n", __func__, is_plane1);
+	DML_LOG_VERBOSE("DML: %s: BytePerPixel = %u\n", __func__, BytePerPixel);
+	DML_LOG_VERBOSE("DML: %s: BlockHeight256Bytes = %u\n", __func__, BlockHeight256Bytes);
+	DML_LOG_VERBOSE("DML: %s: BlockWidth256Bytes = %u\n", __func__, BlockWidth256Bytes);
+	DML_LOG_VERBOSE("DML: %s: MacroTileWidth = %u\n", __func__, MacroTileWidth);
+	DML_LOG_VERBOSE("DML: %s: MacroTileHeight = %u\n", __func__, MacroTileHeight);
+	DML_LOG_VERBOSE("DML: %s: PTEBufferSizeInRequests = %u\n", __func__, PTEBufferSizeInRequests);
+	DML_LOG_VERBOSE("DML: %s: dpte_buffer_size_in_pte_reqs_luma = %u\n", __func__, mode_lib->ip.dpte_buffer_size_in_pte_reqs_luma);
+	DML_LOG_VERBOSE("DML: %s: dpte_buffer_size_in_pte_reqs_chroma = %u\n", __func__, mode_lib->ip.dpte_buffer_size_in_pte_reqs_chroma);
+	DML_LOG_VERBOSE("DML: %s: GPUVMMinPageSizeKBytes = %u\n", __func__, GPUVMMinPageSizeKBytes);
 #endif
 	unsigned int dummy_integer[21];
 
@@ -12193,16 +12121,16 @@ void dml2_core_calcs_get_dpte_row_height(
 	CalculateVMAndRowBytes(&mode_lib->scratch.calculate_vm_and_row_bytes_params);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML: %s: dpte_row_height = %u\n", __func__, *dpte_row_height);
+	DML_LOG_VERBOSE("DML: %s: dpte_row_height = %u\n", __func__, *dpte_row_height);
 #endif
 }
 
 static bool is_dual_plane(enum dml2_source_format_class source_format)
 {
-	bool ret_val = 0;
+	bool ret_val = false;
 
 	if ((source_format == dml2_420_12) || (source_format == dml2_420_8) || (source_format == dml2_420_10) || (source_format == dml2_rgbe_alpha))
-		ret_val = 1;
+		ret_val = true;
 
 	return ret_val;
 }
@@ -12220,6 +12148,8 @@ static void rq_dlg_get_wm_regs(const struct dml2_display_cfg *display_cfg, const
 	wm_regs->fclk_pstate = (int unsigned)(mode_lib->mp.Watermark.FCLKChangeWatermark * refclk_freq_in_mhz);
 	wm_regs->sr_enter = (int unsigned)(mode_lib->mp.Watermark.StutterEnterPlusExitWatermark * refclk_freq_in_mhz);
 	wm_regs->sr_exit = (int unsigned)(mode_lib->mp.Watermark.StutterExitWatermark * refclk_freq_in_mhz);
+	wm_regs->sr_enter_z8 = (int unsigned)(mode_lib->mp.Watermark.Z8StutterEnterPlusExitWatermark * refclk_freq_in_mhz);
+	wm_regs->sr_exit_z8 = (int unsigned)(mode_lib->mp.Watermark.Z8StutterExitWatermark * refclk_freq_in_mhz);
 	wm_regs->temp_read_or_ppt = (int unsigned)(mode_lib->mp.Watermark.temp_read_or_ppt_watermark_us * refclk_freq_in_mhz);
 	wm_regs->uclk_pstate = (int unsigned)(mode_lib->mp.Watermark.DRAMClockChangeWatermark * refclk_freq_in_mhz);
 	wm_regs->urgent = (int unsigned)(mode_lib->mp.Watermark.UrgentWatermark * refclk_freq_in_mhz);
@@ -12246,11 +12176,11 @@ void dml2_core_calcs_cursor_dlg_reg(struct dml2_cursor_dlg_regs *cursor_dlg_regs
 	cursor_dlg_regs->dst_x_offset = (unsigned int) ((dst_x_offset > 0) ? dst_x_offset : 0);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML_DLG::%s: cursor_x_position=%d\n", __func__, p->cursor_x_position);
-	dml2_printf("DML_DLG::%s: dlg_refclk_mhz=%f\n", __func__, p->dlg_refclk_mhz);
-	dml2_printf("DML_DLG::%s: pixel_rate_mhz=%f\n", __func__, p->pixel_rate_mhz);
-	dml2_printf("DML_DLG::%s: dst_x_offset=%d\n", __func__, dst_x_offset);
-	dml2_printf("DML_DLG::%s: dst_x_offset=%d (reg)\n", __func__, cursor_dlg_regs->dst_x_offset);
+	DML_LOG_VERBOSE("DML_DLG::%s: cursor_x_position=%d\n", __func__, p->cursor_x_position);
+	DML_LOG_VERBOSE("DML_DLG::%s: dlg_refclk_mhz=%f\n", __func__, p->dlg_refclk_mhz);
+	DML_LOG_VERBOSE("DML_DLG::%s: pixel_rate_mhz=%f\n", __func__, p->pixel_rate_mhz);
+	DML_LOG_VERBOSE("DML_DLG::%s: dst_x_offset=%d\n", __func__, dst_x_offset);
+	DML_LOG_VERBOSE("DML_DLG::%s: dst_x_offset=%d (reg)\n", __func__, cursor_dlg_regs->dst_x_offset);
 #endif
 
 	cursor_dlg_regs->chunk_hdl_adjust = 3;
@@ -12286,7 +12216,7 @@ static void rq_dlg_get_rq_reg(struct dml2_display_rq_regs *rq_regs,
 	double stored_swath_c_bytes;
 	bool is_phantom_pipe;
 
-	dml2_printf("DML_DLG::%s: Calculation for pipe[%d] start\n", __func__, pipe_idx);
+	DML_LOG_VERBOSE("DML_DLG::%s: Calculation for pipe[%d] start\n", __func__, pipe_idx);
 
 	pixel_chunk_bytes = (unsigned int)(mode_lib->ip.pixel_chunk_size_kbytes * 1024);
 	min_pixel_chunk_bytes = (unsigned int)(mode_lib->ip.min_pixel_chunk_size_bytes);
@@ -12329,19 +12259,19 @@ static void rq_dlg_get_rq_reg(struct dml2_display_rq_regs *rq_regs,
 	if (sw_mode == dml2_sw_linear && display_cfg->gpuvm_enable) {
 		unsigned int p0_pte_row_height_linear = (unsigned int)(dml_get_dpte_row_height_linear_l(mode_lib, pipe_idx));
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML_DLG: %s: p0_pte_row_height_linear = %u\n", __func__, p0_pte_row_height_linear);
+		DML_LOG_VERBOSE("DML_DLG: %s: p0_pte_row_height_linear = %u\n", __func__, p0_pte_row_height_linear);
 #endif
-		DML2_ASSERT(p0_pte_row_height_linear >= 8);
+		DML_ASSERT(p0_pte_row_height_linear >= 8);
 
 		rq_regs->rq_regs_l.pte_row_height_linear = math_log2_approx(p0_pte_row_height_linear) - 3;
 		if (dual_plane) {
 			unsigned int p1_pte_row_height_linear = (unsigned int)(dml_get_dpte_row_height_linear_c(mode_lib, pipe_idx));
 
 #ifdef __DML_VBA_DEBUG__
-			dml2_printf("DML_DLG: %s: p1_pte_row_height_linear = %u\n", __func__, p1_pte_row_height_linear);
+			DML_LOG_VERBOSE("DML_DLG: %s: p1_pte_row_height_linear = %u\n", __func__, p1_pte_row_height_linear);
 #endif
 			if (sw_mode == dml2_sw_linear) {
-				DML2_ASSERT(p1_pte_row_height_linear >= 8);
+				DML_ASSERT(p1_pte_row_height_linear >= 8);
 			}
 			rq_regs->rq_regs_c.pte_row_height_linear = math_log2_approx(p1_pte_row_height_linear) - 3;
 		}
@@ -12375,12 +12305,12 @@ static void rq_dlg_get_rq_reg(struct dml2_display_rq_regs *rq_regs,
 			if (stored_swath_l_bytes / stored_swath_c_bytes <= 1.5) {
 				detile_buf_plane1_addr = (unsigned int)(detile_buf_size_in_bytes / 2.0 / 1024.0); // half to chroma
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML_DLG: %s: detile_buf_plane1_addr = %d (1/2 to chroma)\n", __func__, detile_buf_plane1_addr);
+				DML_LOG_VERBOSE("DML_DLG: %s: detile_buf_plane1_addr = %d (1/2 to chroma)\n", __func__, detile_buf_plane1_addr);
 #endif
 			} else {
 				detile_buf_plane1_addr = (unsigned int)(dml_round_to_multiple((unsigned int)((2.0 * detile_buf_size_in_bytes) / 3.0), 1024, 0) / 1024.0); // 2/3 to luma
 #ifdef __DML_VBA_DEBUG__
-				dml2_printf("DML_DLG: %s: detile_buf_plane1_addr = %d (1/3 chroma)\n", __func__, detile_buf_plane1_addr);
+				DML_LOG_VERBOSE("DML_DLG: %s: detile_buf_plane1_addr = %d (1/3 chroma)\n", __func__, detile_buf_plane1_addr);
 #endif
 			}
 		}
@@ -12388,15 +12318,15 @@ static void rq_dlg_get_rq_reg(struct dml2_display_rq_regs *rq_regs,
 	rq_regs->plane1_base_address = detile_buf_plane1_addr;
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML_DLG: %s: is_phantom_pipe = %d\n", __func__, is_phantom_pipe);
-	dml2_printf("DML_DLG: %s: stored_swath_l_bytes = %f\n", __func__, stored_swath_l_bytes);
-	dml2_printf("DML_DLG: %s: stored_swath_c_bytes = %f\n", __func__, stored_swath_c_bytes);
-	dml2_printf("DML_DLG: %s: detile_buf_size_in_bytes = %d\n", __func__, detile_buf_size_in_bytes);
-	dml2_printf("DML_DLG: %s: detile_buf_plane1_addr = %d\n", __func__, detile_buf_plane1_addr);
-	dml2_printf("DML_DLG: %s: plane1_base_address = %d\n", __func__, rq_regs->plane1_base_address);
+	DML_LOG_VERBOSE("DML_DLG: %s: is_phantom_pipe = %d\n", __func__, is_phantom_pipe);
+	DML_LOG_VERBOSE("DML_DLG: %s: stored_swath_l_bytes = %f\n", __func__, stored_swath_l_bytes);
+	DML_LOG_VERBOSE("DML_DLG: %s: stored_swath_c_bytes = %f\n", __func__, stored_swath_c_bytes);
+	DML_LOG_VERBOSE("DML_DLG: %s: detile_buf_size_in_bytes = %d\n", __func__, detile_buf_size_in_bytes);
+	DML_LOG_VERBOSE("DML_DLG: %s: detile_buf_plane1_addr = %d\n", __func__, detile_buf_plane1_addr);
+	DML_LOG_VERBOSE("DML_DLG: %s: plane1_base_address = %d\n", __func__, rq_regs->plane1_base_address);
 #endif
-	//dml2_printf_rq_regs_st(rq_regs);
-	dml2_printf("DML_DLG::%s: Calculation for pipe[%d] done\n", __func__, pipe_idx);
+	//DML_LOG_VERBOSE_rq_regs_st(rq_regs);
+	DML_LOG_VERBOSE("DML_DLG::%s: Calculation for pipe[%d] done\n", __func__, pipe_idx);
 }
 
 static void rq_dlg_get_dlg_reg(
@@ -12411,10 +12341,10 @@ static void rq_dlg_get_dlg_reg(
 
 	memset(l, 0, sizeof(struct dml2_core_shared_rq_dlg_get_dlg_reg_locals));
 
-	dml2_printf("DML_DLG::%s: Calculation for pipe_idx=%d\n", __func__, pipe_idx);
+	DML_LOG_VERBOSE("DML_DLG::%s: Calculation for pipe_idx=%d\n", __func__, pipe_idx);
 
 	l->plane_idx = dml_get_plane_idx(mode_lib, pipe_idx);
-	DML2_ASSERT(l->plane_idx < DML2_MAX_PLANES);
+	DML_ASSERT(l->plane_idx < DML2_MAX_PLANES);
 
 	l->source_format = dml2_444_8;
 	l->odm_mode = dml2_odm_mode_bypass;
@@ -12444,18 +12374,18 @@ static void rq_dlg_get_dlg_reg(
 		l->pclk_freq_in_mhz = (double)l->timing->pixel_clock_khz / 1000;
 		l->ref_freq_to_pix_freq = l->refclk_freq_in_mhz / l->pclk_freq_in_mhz;
 
-		dml2_printf("DML_DLG::%s: plane_idx = %d\n", __func__, l->plane_idx);
-		dml2_printf("DML_DLG: %s: htotal = %d\n", __func__, l->htotal);
-		dml2_printf("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, l->refclk_freq_in_mhz);
-		dml2_printf("DML_DLG: %s: dlg_ref_clk_mhz = %3.2f\n", __func__, display_cfg->overrides.hw.dlg_ref_clk_mhz);
-		dml2_printf("DML_DLG: %s: soc.refclk_mhz = %3.2f\n", __func__, mode_lib->soc.dchub_refclk_mhz);
-		dml2_printf("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, l->pclk_freq_in_mhz);
-		dml2_printf("DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", __func__, l->ref_freq_to_pix_freq);
-		dml2_printf("DML_DLG: %s: interlaced = %d\n", __func__, l->interlaced);
+		DML_LOG_VERBOSE("DML_DLG::%s: plane_idx = %d\n", __func__, l->plane_idx);
+		DML_LOG_VERBOSE("DML_DLG: %s: htotal = %d\n", __func__, l->htotal);
+		DML_LOG_VERBOSE("DML_DLG: %s: refclk_freq_in_mhz = %3.2f\n", __func__, l->refclk_freq_in_mhz);
+		DML_LOG_VERBOSE("DML_DLG: %s: dlg_ref_clk_mhz = %3.2f\n", __func__, display_cfg->overrides.hw.dlg_ref_clk_mhz);
+		DML_LOG_VERBOSE("DML_DLG: %s: soc.refclk_mhz = %d\n", __func__, mode_lib->soc.dchub_refclk_mhz);
+		DML_LOG_VERBOSE("DML_DLG: %s: pclk_freq_in_mhz = %3.2f\n", __func__, l->pclk_freq_in_mhz);
+		DML_LOG_VERBOSE("DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", __func__, l->ref_freq_to_pix_freq);
+		DML_LOG_VERBOSE("DML_DLG: %s: interlaced = %d\n", __func__, l->interlaced);
 
-		DML2_ASSERT(l->refclk_freq_in_mhz != 0);
-		DML2_ASSERT(l->pclk_freq_in_mhz != 0);
-		DML2_ASSERT(l->ref_freq_to_pix_freq < 4.0);
+		DML_ASSERT(l->refclk_freq_in_mhz != 0);
+		DML_ASSERT(l->pclk_freq_in_mhz != 0);
+		DML_ASSERT(l->ref_freq_to_pix_freq < 4.0);
 
 		// Need to figure out which side of odm combine we're in
 		// Assume the pipe instance under the same plane is in order
@@ -12484,14 +12414,14 @@ static void rq_dlg_get_dlg_reg(
 			l->pipe_idx_in_combine = pipe_idx - l->first_pipe_idx_in_plane; // DML assumes the pipes in the same plane will have continuous indexing (i.e. plane 0 use pipe 0, 1, and plane 1 uses pipe 2, 3, etc.)
 
 			disp_dlg_regs->refcyc_h_blank_end = (unsigned int)(((double)l->hblank_end + (double)l->pipe_idx_in_combine * (double)l->hactive / (double)l->odm_combine_factor) * l->ref_freq_to_pix_freq);
-			dml2_printf("DML_DLG: %s: pipe_idx = %d\n", __func__, pipe_idx);
-			dml2_printf("DML_DLG: %s: first_pipe_idx_in_plane = %d\n", __func__, l->first_pipe_idx_in_plane);
-			dml2_printf("DML_DLG: %s: pipe_idx_in_combine = %d\n", __func__, l->pipe_idx_in_combine);
-			dml2_printf("DML_DLG: %s: odm_combine_factor = %d\n", __func__, l->odm_combine_factor);
+			DML_LOG_VERBOSE("DML_DLG: %s: pipe_idx = %d\n", __func__, pipe_idx);
+			DML_LOG_VERBOSE("DML_DLG: %s: first_pipe_idx_in_plane = %d\n", __func__, l->first_pipe_idx_in_plane);
+			DML_LOG_VERBOSE("DML_DLG: %s: pipe_idx_in_combine = %d\n", __func__, l->pipe_idx_in_combine);
+			DML_LOG_VERBOSE("DML_DLG: %s: odm_combine_factor = %d\n", __func__, l->odm_combine_factor);
 		}
-		dml2_printf("DML_DLG: %s: refcyc_h_blank_end = %d\n", __func__, disp_dlg_regs->refcyc_h_blank_end);
+		DML_LOG_VERBOSE("DML_DLG: %s: refcyc_h_blank_end = %d\n", __func__, disp_dlg_regs->refcyc_h_blank_end);
 
-		DML2_ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_dlg_regs->refcyc_h_blank_end < (unsigned int)math_pow(2, 13));
 
 		disp_dlg_regs->ref_freq_to_pix_freq = (unsigned int)(l->ref_freq_to_pix_freq * math_pow(2, 19));
 		disp_dlg_regs->refcyc_per_htotal = (unsigned int)(l->ref_freq_to_pix_freq * (double)l->htotal * math_pow(2, 8));
@@ -12500,20 +12430,20 @@ static void rq_dlg_get_dlg_reg(
 		l->min_ttu_vblank = mode_lib->mp.MinTTUVBlank[mode_lib->mp.pipe_plane[pipe_idx]];
 		l->min_dst_y_next_start = (unsigned int)(mode_lib->mp.MIN_DST_Y_NEXT_START[mode_lib->mp.pipe_plane[pipe_idx]]);
 
-		dml2_printf("DML_DLG: %s: min_ttu_vblank (us) = %3.2f\n", __func__, l->min_ttu_vblank);
-		dml2_printf("DML_DLG: %s: min_dst_y_next_start = %d\n", __func__, l->min_dst_y_next_start);
-		dml2_printf("DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", __func__, l->ref_freq_to_pix_freq);
+		DML_LOG_VERBOSE("DML_DLG: %s: min_ttu_vblank (us) = %3.2f\n", __func__, l->min_ttu_vblank);
+		DML_LOG_VERBOSE("DML_DLG: %s: min_dst_y_next_start = %d\n", __func__, l->min_dst_y_next_start);
+		DML_LOG_VERBOSE("DML_DLG: %s: ref_freq_to_pix_freq = %3.2f\n", __func__, l->ref_freq_to_pix_freq);
 
 		l->vready_after_vcount0 = (unsigned int)(mode_lib->mp.VREADY_AT_OR_AFTER_VSYNC[mode_lib->mp.pipe_plane[pipe_idx]]);
 		disp_dlg_regs->vready_after_vcount0 = l->vready_after_vcount0;
 
-		dml2_printf("DML_DLG: %s: vready_after_vcount0 = %d\n", __func__, disp_dlg_regs->vready_after_vcount0);
+		DML_LOG_VERBOSE("DML_DLG: %s: vready_after_vcount0 = %d\n", __func__, disp_dlg_regs->vready_after_vcount0);
 
 		l->dst_x_after_scaler = (unsigned int)(mode_lib->mp.DSTXAfterScaler[mode_lib->mp.pipe_plane[pipe_idx]]);
 		l->dst_y_after_scaler = (unsigned int)(mode_lib->mp.DSTYAfterScaler[mode_lib->mp.pipe_plane[pipe_idx]]);
 
-		dml2_printf("DML_DLG: %s: dst_x_after_scaler = %d\n", __func__, l->dst_x_after_scaler);
-		dml2_printf("DML_DLG: %s: dst_y_after_scaler = %d\n", __func__, l->dst_y_after_scaler);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_x_after_scaler = %d\n", __func__, l->dst_x_after_scaler);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_y_after_scaler = %d\n", __func__, l->dst_y_after_scaler);
 
 		l->dst_y_prefetch = mode_lib->mp.dst_y_prefetch[mode_lib->mp.pipe_plane[pipe_idx]];
 		l->dst_y_per_vm_vblank = mode_lib->mp.dst_y_per_vm_vblank[mode_lib->mp.pipe_plane[pipe_idx]];
@@ -12521,28 +12451,28 @@ static void rq_dlg_get_dlg_reg(
 		l->dst_y_per_vm_flip = mode_lib->mp.dst_y_per_vm_flip[mode_lib->mp.pipe_plane[pipe_idx]];
 		l->dst_y_per_row_flip = mode_lib->mp.dst_y_per_row_flip[mode_lib->mp.pipe_plane[pipe_idx]];
 
-		dml2_printf("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, l->dst_y_prefetch);
-		dml2_printf("DML_DLG: %s: dst_y_per_vm_flip = %3.2f\n", __func__, l->dst_y_per_vm_flip);
-		dml2_printf("DML_DLG: %s: dst_y_per_row_flip = %3.2f\n", __func__, l->dst_y_per_row_flip);
-		dml2_printf("DML_DLG: %s: dst_y_per_vm_vblank = %3.2f\n", __func__, l->dst_y_per_vm_vblank);
-		dml2_printf("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, l->dst_y_per_row_vblank);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_y_prefetch (after rnd) = %3.2f\n", __func__, l->dst_y_prefetch);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_y_per_vm_flip = %3.2f\n", __func__, l->dst_y_per_vm_flip);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_y_per_row_flip = %3.2f\n", __func__, l->dst_y_per_row_flip);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_y_per_vm_vblank = %3.2f\n", __func__, l->dst_y_per_vm_vblank);
+		DML_LOG_VERBOSE("DML_DLG: %s: dst_y_per_row_vblank = %3.2f\n", __func__, l->dst_y_per_row_vblank);
 
 		if (l->dst_y_prefetch > 0 && l->dst_y_per_vm_vblank > 0 && l->dst_y_per_row_vblank > 0) {
-			DML2_ASSERT(l->dst_y_prefetch > (l->dst_y_per_vm_vblank + l->dst_y_per_row_vblank));
+			DML_ASSERT(l->dst_y_prefetch > (l->dst_y_per_vm_vblank + l->dst_y_per_row_vblank));
 		}
 
 		l->vratio_pre_l = mode_lib->mp.VRatioPrefetchY[mode_lib->mp.pipe_plane[pipe_idx]];
 		l->vratio_pre_c = mode_lib->mp.VRatioPrefetchC[mode_lib->mp.pipe_plane[pipe_idx]];
 
-		dml2_printf("DML_DLG: %s: vratio_pre_l = %3.2f\n", __func__, l->vratio_pre_l);
-		dml2_printf("DML_DLG: %s: vratio_pre_c = %3.2f\n", __func__, l->vratio_pre_c);
+		DML_LOG_VERBOSE("DML_DLG: %s: vratio_pre_l = %3.2f\n", __func__, l->vratio_pre_l);
+		DML_LOG_VERBOSE("DML_DLG: %s: vratio_pre_c = %3.2f\n", __func__, l->vratio_pre_c);
 
 		// Active
 		l->refcyc_per_line_delivery_pre_l = mode_lib->mp.DisplayPipeLineDeliveryTimeLumaPrefetch[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 		l->refcyc_per_line_delivery_l = mode_lib->mp.DisplayPipeLineDeliveryTimeLuma[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 
-		dml2_printf("DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", __func__, l->refcyc_per_line_delivery_pre_l);
-		dml2_printf("DML_DLG: %s: refcyc_per_line_delivery_l = %3.2f\n", __func__, l->refcyc_per_line_delivery_l);
+		DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_line_delivery_pre_l = %3.2f\n", __func__, l->refcyc_per_line_delivery_pre_l);
+		DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_line_delivery_l = %3.2f\n", __func__, l->refcyc_per_line_delivery_l);
 
 		l->refcyc_per_line_delivery_pre_c = 0.0;
 		l->refcyc_per_line_delivery_c = 0.0;
@@ -12551,8 +12481,8 @@ static void rq_dlg_get_dlg_reg(
 			l->refcyc_per_line_delivery_pre_c = mode_lib->mp.DisplayPipeLineDeliveryTimeChromaPrefetch[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 			l->refcyc_per_line_delivery_c = mode_lib->mp.DisplayPipeLineDeliveryTimeChroma[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 
-			dml2_printf("DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n", __func__, l->refcyc_per_line_delivery_pre_c);
-			dml2_printf("DML_DLG: %s: refcyc_per_line_delivery_c = %3.2f\n", __func__, l->refcyc_per_line_delivery_c);
+			DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_line_delivery_pre_c = %3.2f\n", __func__, l->refcyc_per_line_delivery_pre_c);
+			DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_line_delivery_c = %3.2f\n", __func__, l->refcyc_per_line_delivery_c);
 		}
 
 		disp_dlg_regs->refcyc_per_vm_dmdata = (unsigned int)(mode_lib->mp.Tdmdl_vm[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz);
@@ -12561,8 +12491,8 @@ static void rq_dlg_get_dlg_reg(
 		l->refcyc_per_req_delivery_pre_l = mode_lib->mp.DisplayPipeRequestDeliveryTimeLumaPrefetch[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 		l->refcyc_per_req_delivery_l = mode_lib->mp.DisplayPipeRequestDeliveryTimeLuma[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 
-		dml2_printf("DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", __func__, l->refcyc_per_req_delivery_pre_l);
-		dml2_printf("DML_DLG: %s: refcyc_per_req_delivery_l = %3.2f\n", __func__, l->refcyc_per_req_delivery_l);
+		DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_req_delivery_pre_l = %3.2f\n", __func__, l->refcyc_per_req_delivery_pre_l);
+		DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_req_delivery_l = %3.2f\n", __func__, l->refcyc_per_req_delivery_l);
 
 		l->refcyc_per_req_delivery_pre_c = 0.0;
 		l->refcyc_per_req_delivery_c = 0.0;
@@ -12570,16 +12500,16 @@ static void rq_dlg_get_dlg_reg(
 			l->refcyc_per_req_delivery_pre_c = mode_lib->mp.DisplayPipeRequestDeliveryTimeChromaPrefetch[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 			l->refcyc_per_req_delivery_c = mode_lib->mp.DisplayPipeRequestDeliveryTimeChroma[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz;
 
-			dml2_printf("DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n", __func__, l->refcyc_per_req_delivery_pre_c);
-			dml2_printf("DML_DLG: %s: refcyc_per_req_delivery_c = %3.2f\n", __func__, l->refcyc_per_req_delivery_c);
+			DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_req_delivery_pre_c = %3.2f\n", __func__, l->refcyc_per_req_delivery_pre_c);
+			DML_LOG_VERBOSE("DML_DLG: %s: refcyc_per_req_delivery_c = %3.2f\n", __func__, l->refcyc_per_req_delivery_c);
 		}
 
 		// TTU - Cursor
-		DML2_ASSERT(display_cfg->plane_descriptors[l->plane_idx].cursor.num_cursors <= 1);
+		DML_ASSERT(display_cfg->plane_descriptors[l->plane_idx].cursor.num_cursors <= 1);
 
 		// Assign to register structures
 		disp_dlg_regs->min_dst_y_next_start = (unsigned int)((double)l->min_dst_y_next_start * math_pow(2, 2));
-		DML2_ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)math_pow(2, 18));
+		DML_ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)math_pow(2, 18));
 
 		disp_dlg_regs->dst_y_after_scaler = l->dst_y_after_scaler; // in terms of line
 		disp_dlg_regs->refcyc_x_after_scaler = (unsigned int)((double)l->dst_x_after_scaler * l->ref_freq_to_pix_freq); // in terms of refclk
@@ -12592,10 +12522,10 @@ static void rq_dlg_get_dlg_reg(
 		disp_dlg_regs->vratio_prefetch = (unsigned int)(l->vratio_pre_l * math_pow(2, 19));
 		disp_dlg_regs->vratio_prefetch_c = (unsigned int)(l->vratio_pre_c * math_pow(2, 19));
 
-		dml2_printf("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_vblank);
-		dml2_printf("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_vblank);
-		dml2_printf("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip);
-		dml2_printf("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip);
+		DML_LOG_VERBOSE("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_vblank);
+		DML_LOG_VERBOSE("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_vblank = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_vblank);
+		DML_LOG_VERBOSE("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip);
+		DML_LOG_VERBOSE("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip);
 
 		disp_dlg_regs->refcyc_per_vm_group_vblank = (unsigned int)(mode_lib->mp.TimePerVMGroupVBlank[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz);
 		disp_dlg_regs->refcyc_per_vm_group_flip = (unsigned int)(mode_lib->mp.TimePerVMGroupFlip[mode_lib->mp.pipe_plane[pipe_idx]] * l->refclk_freq_in_mhz);
@@ -12662,11 +12592,11 @@ static void rq_dlg_get_dlg_reg(
 		disp_ttu_regs->qos_ramp_disable_c = 0;
 		disp_ttu_regs->min_ttu_vblank = (unsigned int)(l->min_ttu_vblank * l->refclk_freq_in_mhz);
 
-		// CHECK for HW registers' range, DML2_ASSERT or clamp
-		DML2_ASSERT(l->refcyc_per_req_delivery_pre_l < math_pow(2, 13));
-		DML2_ASSERT(l->refcyc_per_req_delivery_l < math_pow(2, 13));
-		DML2_ASSERT(l->refcyc_per_req_delivery_pre_c < math_pow(2, 13));
-		DML2_ASSERT(l->refcyc_per_req_delivery_c < math_pow(2, 13));
+		// CHECK for HW registers' range, DML_ASSERT or clamp
+		DML_ASSERT(l->refcyc_per_req_delivery_pre_l < math_pow(2, 13));
+		DML_ASSERT(l->refcyc_per_req_delivery_l < math_pow(2, 13));
+		DML_ASSERT(l->refcyc_per_req_delivery_pre_c < math_pow(2, 13));
+		DML_ASSERT(l->refcyc_per_req_delivery_c < math_pow(2, 13));
 		if (disp_dlg_regs->refcyc_per_vm_group_vblank >= (unsigned int)math_pow(2, 23))
 			disp_dlg_regs->refcyc_per_vm_group_vblank = (unsigned int)(math_pow(2, 23) - 1);
 
@@ -12680,16 +12610,16 @@ static void rq_dlg_get_dlg_reg(
 			disp_dlg_regs->refcyc_per_vm_req_flip = (unsigned int)(math_pow(2, 23) - 1);
 
 
-		DML2_ASSERT(disp_dlg_regs->dst_y_after_scaler < (unsigned int)8);
-		DML2_ASSERT(disp_dlg_regs->refcyc_x_after_scaler < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_dlg_regs->dst_y_after_scaler < (unsigned int)8);
+		DML_ASSERT(disp_dlg_regs->refcyc_x_after_scaler < (unsigned int)math_pow(2, 13));
 
 		if (disp_dlg_regs->dst_y_per_pte_row_nom_l >= (unsigned int)math_pow(2, 17)) {
-			dml2_printf("DML_DLG: %s: Warning DST_Y_PER_PTE_ROW_NOM_L %u > register max U15.2 %u, clamp to max\n", __func__, disp_dlg_regs->dst_y_per_pte_row_nom_l, (unsigned int)math_pow(2, 17) - 1);
+			DML_LOG_VERBOSE("DML_DLG: %s: Warning DST_Y_PER_PTE_ROW_NOM_L %u > register max U15.2 %u, clamp to max\n", __func__, disp_dlg_regs->dst_y_per_pte_row_nom_l, (unsigned int)math_pow(2, 17) - 1);
 			l->dst_y_per_pte_row_nom_l = (unsigned int)math_pow(2, 17) - 1;
 		}
 		if (l->dual_plane) {
 			if (disp_dlg_regs->dst_y_per_pte_row_nom_c >= (unsigned int)math_pow(2, 17)) {
-				dml2_printf("DML_DLG: %s: Warning DST_Y_PER_PTE_ROW_NOM_C %u > register max U15.2 %u, clamp to max\n", __func__, disp_dlg_regs->dst_y_per_pte_row_nom_c, (unsigned int)math_pow(2, 17) - 1);
+				DML_LOG_VERBOSE("DML_DLG: %s: Warning DST_Y_PER_PTE_ROW_NOM_C %u > register max U15.2 %u, clamp to max\n", __func__, disp_dlg_regs->dst_y_per_pte_row_nom_c, (unsigned int)math_pow(2, 17) - 1);
 				l->dst_y_per_pte_row_nom_c = (unsigned int)math_pow(2, 17) - 1;
 			}
 		}
@@ -12700,20 +12630,20 @@ static void rq_dlg_get_dlg_reg(
 			if (disp_dlg_regs->refcyc_per_pte_group_nom_c >= (unsigned int)math_pow(2, 23))
 				disp_dlg_regs->refcyc_per_pte_group_nom_c = (unsigned int)(math_pow(2, 23) - 1);
 		}
-		DML2_ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)math_pow(2, 13));
 		if (l->dual_plane) {
-			DML2_ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c < (unsigned int)math_pow(2, 13));
+			DML_ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c < (unsigned int)math_pow(2, 13));
 		}
 
-		DML2_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int)math_pow(2, 13));
-		DML2_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_l < (unsigned int)math_pow(2, 13));
-		DML2_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int)math_pow(2, 13));
-		DML2_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_c < (unsigned int)math_pow(2, 13));
-		DML2_ASSERT(disp_ttu_regs->qos_level_low_wm < (unsigned int)math_pow(2, 14));
-		DML2_ASSERT(disp_ttu_regs->qos_level_high_wm < (unsigned int)math_pow(2, 14));
-		DML2_ASSERT(disp_ttu_regs->min_ttu_vblank < (unsigned int)math_pow(2, 24));
+		DML_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_l < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_l < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_pre_c < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_dlg_regs->refcyc_per_line_delivery_c < (unsigned int)math_pow(2, 13));
+		DML_ASSERT(disp_ttu_regs->qos_level_low_wm < (unsigned int)math_pow(2, 14));
+		DML_ASSERT(disp_ttu_regs->qos_level_high_wm < (unsigned int)math_pow(2, 14));
+		DML_ASSERT(disp_ttu_regs->min_ttu_vblank < (unsigned int)math_pow(2, 24));
 
-		dml2_printf("DML_DLG::%s: Calculation for pipe[%d] done\n", __func__, pipe_idx);
+		DML_LOG_VERBOSE("DML_DLG::%s: Calculation for pipe[%d] done\n", __func__, pipe_idx);
 
 	}
 }
@@ -12736,11 +12666,11 @@ static void rq_dlg_get_arb_params(const struct dml2_display_cfg *display_cfg, co
 	arb_param->pstate_stall_threshold = (unsigned int)(mode_lib->ip_caps.fams2.max_allow_delay_us * refclk_freq_in_mhz);
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: max_req_outstanding = %d\n", __func__, arb_param->max_req_outstanding);
-	dml2_printf("DML::%s: sdpif_request_rate_limit = %d\n", __func__, arb_param->sdpif_request_rate_limit);
-	dml2_printf("DML::%s: compbuf_reserved_space_kbytes = %d\n", __func__, arb_param->compbuf_reserved_space_kbytes);
-	dml2_printf("DML::%s: allow_sdpif_rate_limit_when_cstate_req = %d\n", __func__, arb_param->allow_sdpif_rate_limit_when_cstate_req);
-	dml2_printf("DML::%s: dcfclk_deep_sleep_hysteresis = %d\n", __func__, arb_param->dcfclk_deep_sleep_hysteresis);
+	DML_LOG_VERBOSE("DML::%s: max_req_outstanding = %d\n", __func__, arb_param->max_req_outstanding);
+	DML_LOG_VERBOSE("DML::%s: sdpif_request_rate_limit = %d\n", __func__, arb_param->sdpif_request_rate_limit);
+	DML_LOG_VERBOSE("DML::%s: compbuf_reserved_space_kbytes = %d\n", __func__, arb_param->compbuf_reserved_space_kbytes);
+	DML_LOG_VERBOSE("DML::%s: allow_sdpif_rate_limit_when_cstate_req = %d\n", __func__, arb_param->allow_sdpif_rate_limit_when_cstate_req);
+	DML_LOG_VERBOSE("DML::%s: dcfclk_deep_sleep_hysteresis = %d\n", __func__, arb_param->dcfclk_deep_sleep_hysteresis);
 #endif
 
 }
@@ -13013,10 +12943,10 @@ void dml2_core_calcs_get_stream_support_info(const struct dml2_display_cfg *disp
 
 	out->vblank_reserved_time_us = display_cfg->plane_descriptors[plane_index].overrides.reserved_vblank_time_ns / 1000;
 #if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: subvp_fw_processing_delay_us = %d\n", __func__, mode_lib->ip.subvp_fw_processing_delay_us);
-	dml2_printf("DML::%s: subvp_pstate_allow_width_us = %d\n", __func__, mode_lib->ip.subvp_pstate_allow_width_us);
-	dml2_printf("DML::%s: subvp_swath_height_margin_lines = %d\n", __func__, mode_lib->ip.subvp_swath_height_margin_lines);
-	dml2_printf("DML::%s: vblank_reserved_time_us = %f\n", __func__, out->vblank_reserved_time_us);
+	DML_LOG_VERBOSE("DML::%s: subvp_fw_processing_delay_us = %d\n", __func__, mode_lib->ip.subvp_fw_processing_delay_us);
+	DML_LOG_VERBOSE("DML::%s: subvp_pstate_allow_width_us = %d\n", __func__, mode_lib->ip.subvp_pstate_allow_width_us);
+	DML_LOG_VERBOSE("DML::%s: subvp_swath_height_margin_lines = %d\n", __func__, mode_lib->ip.subvp_swath_height_margin_lines);
+	DML_LOG_VERBOSE("DML::%s: vblank_reserved_time_us = %u\n", __func__, out->vblank_reserved_time_us);
 #endif
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h
index 4e502f0a6d20..bdee6ad7bc59 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h
@@ -1078,6 +1078,8 @@ struct dml2_core_calcs_mode_programming_locals {
 	enum dml2_source_format_class pixel_format[DML2_MAX_PLANES];
 	unsigned int lb_source_lines_l[DML2_MAX_PLANES];
 	unsigned int lb_source_lines_c[DML2_MAX_PLANES];
+	unsigned int num_dsc_slices[DML2_MAX_PLANES];
+	bool dsc_enable[DML2_MAX_PLANES];
 };
 
 struct dml2_core_calcs_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport_locals {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c
index 2504d9c2ec34..7a220c0141c2 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c
@@ -82,7 +82,7 @@ bool dml2_core_utils_is_420(enum dml2_source_format_class source_format)
 		val = 0;
 		break;
 	default:
-		DML2_ASSERT(0);
+		DML_ASSERT(0);
 		break;
 	}
 	return val;
@@ -145,7 +145,7 @@ bool dml2_core_utils_is_422_planar(enum dml2_source_format_class source_format)
 		val = 0;
 		break;
 	default:
-		DML2_ASSERT(0);
+		DML_ASSERT(0);
 		break;
 	}
 	return val;
@@ -208,7 +208,7 @@ bool dml2_core_utils_is_422_packed(enum dml2_source_format_class source_format)
 		val = 1;
 		break;
 	default:
-		DML2_ASSERT(0);
+		DML_ASSERT(0);
 		break;
 	}
 	return val;
@@ -216,104 +216,104 @@ bool dml2_core_utils_is_422_packed(enum dml2_source_format_class source_format)
 
 void dml2_core_utils_print_mode_support_info(const struct dml2_core_internal_mode_support_info *support, bool fail_only)
 {
-	dml2_printf("DML: ===================================== \n");
-	dml2_printf("DML: DML_MODE_SUPPORT_INFO_ST\n");
+	DML_LOG_VERBOSE("DML: ===================================== \n");
+	DML_LOG_VERBOSE("DML: DML_MODE_SUPPORT_INFO_ST\n");
 	if (!fail_only || support->ScaleRatioAndTapsSupport == 0)
-		dml2_printf("DML: support: ScaleRatioAndTapsSupport = %d\n", support->ScaleRatioAndTapsSupport);
+		DML_LOG_VERBOSE("DML: support: ScaleRatioAndTapsSupport = %d\n", support->ScaleRatioAndTapsSupport);
 	if (!fail_only || support->SourceFormatPixelAndScanSupport == 0)
-		dml2_printf("DML: support: SourceFormatPixelAndScanSupport = %d\n", support->SourceFormatPixelAndScanSupport);
+		DML_LOG_VERBOSE("DML: support: SourceFormatPixelAndScanSupport = %d\n", support->SourceFormatPixelAndScanSupport);
 	if (!fail_only || support->ViewportSizeSupport == 0)
-		dml2_printf("DML: support: ViewportSizeSupport = %d\n", support->ViewportSizeSupport);
+		DML_LOG_VERBOSE("DML: support: ViewportSizeSupport = %d\n", support->ViewportSizeSupport);
 	if (!fail_only || support->LinkRateDoesNotMatchDPVersion == 1)
-		dml2_printf("DML: support: LinkRateDoesNotMatchDPVersion = %d\n", support->LinkRateDoesNotMatchDPVersion);
+		DML_LOG_VERBOSE("DML: support: LinkRateDoesNotMatchDPVersion = %d\n", support->LinkRateDoesNotMatchDPVersion);
 	if (!fail_only || support->LinkRateForMultistreamNotIndicated == 1)
-		dml2_printf("DML: support: LinkRateForMultistreamNotIndicated = %d\n", support->LinkRateForMultistreamNotIndicated);
+		DML_LOG_VERBOSE("DML: support: LinkRateForMultistreamNotIndicated = %d\n", support->LinkRateForMultistreamNotIndicated);
 	if (!fail_only || support->BPPForMultistreamNotIndicated == 1)
-		dml2_printf("DML: support: BPPForMultistreamNotIndicated = %d\n", support->BPPForMultistreamNotIndicated);
+		DML_LOG_VERBOSE("DML: support: BPPForMultistreamNotIndicated = %d\n", support->BPPForMultistreamNotIndicated);
 	if (!fail_only || support->MultistreamWithHDMIOreDP == 1)
-		dml2_printf("DML: support: MultistreamWithHDMIOreDP = %d\n", support->MultistreamWithHDMIOreDP);
+		DML_LOG_VERBOSE("DML: support: MultistreamWithHDMIOreDP = %d\n", support->MultistreamWithHDMIOreDP);
 	if (!fail_only || support->ExceededMultistreamSlots == 1)
-		dml2_printf("DML: support: ExceededMultistreamSlots = %d\n", support->ExceededMultistreamSlots);
+		DML_LOG_VERBOSE("DML: support: ExceededMultistreamSlots = %d\n", support->ExceededMultistreamSlots);
 	if (!fail_only || support->MSOOrODMSplitWithNonDPLink == 1)
-		dml2_printf("DML: support: MSOOrODMSplitWithNonDPLink = %d\n", support->MSOOrODMSplitWithNonDPLink);
+		DML_LOG_VERBOSE("DML: support: MSOOrODMSplitWithNonDPLink = %d\n", support->MSOOrODMSplitWithNonDPLink);
 	if (!fail_only || support->NotEnoughLanesForMSO == 1)
-		dml2_printf("DML: support: NotEnoughLanesForMSO = %d\n", support->NotEnoughLanesForMSO);
+		DML_LOG_VERBOSE("DML: support: NotEnoughLanesForMSO = %d\n", support->NotEnoughLanesForMSO);
 	if (!fail_only || support->P2IWith420 == 1)
-		dml2_printf("DML: support: P2IWith420 = %d\n", support->P2IWith420);
+		DML_LOG_VERBOSE("DML: support: P2IWith420 = %d\n", support->P2IWith420);
 	if (!fail_only || support->DSC422NativeNotSupported == 1)
-		dml2_printf("DML: support: DSC422NativeNotSupported = %d\n", support->DSC422NativeNotSupported);
+		DML_LOG_VERBOSE("DML: support: DSC422NativeNotSupported = %d\n", support->DSC422NativeNotSupported);
 	if (!fail_only || support->DSCSlicesODMModeSupported == 0)
-		dml2_printf("DML: support: DSCSlicesODMModeSupported = %d\n", support->DSCSlicesODMModeSupported);
+		DML_LOG_VERBOSE("DML: support: DSCSlicesODMModeSupported = %d\n", support->DSCSlicesODMModeSupported);
 	if (!fail_only || support->NotEnoughDSCUnits == 1)
-		dml2_printf("DML: support: NotEnoughDSCUnits = %d\n", support->NotEnoughDSCUnits);
+		DML_LOG_VERBOSE("DML: support: NotEnoughDSCUnits = %d\n", support->NotEnoughDSCUnits);
 	if (!fail_only || support->NotEnoughDSCSlices == 1)
-		dml2_printf("DML: support: NotEnoughDSCSlices = %d\n", support->NotEnoughDSCSlices);
+		DML_LOG_VERBOSE("DML: support: NotEnoughDSCSlices = %d\n", support->NotEnoughDSCSlices);
 	if (!fail_only || support->ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe == 1)
-		dml2_printf("DML: support: ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe = %d\n", support->ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe);
+		DML_LOG_VERBOSE("DML: support: ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe = %d\n", support->ImmediateFlipOrHostVMAndPStateWithMALLFullFrameOrPhantomPipe);
 	if (!fail_only || support->InvalidCombinationOfMALLUseForPStateAndStaticScreen == 1)
-		dml2_printf("DML: support: InvalidCombinationOfMALLUseForPStateAndStaticScreen = %d\n", support->InvalidCombinationOfMALLUseForPStateAndStaticScreen);
+		DML_LOG_VERBOSE("DML: support: InvalidCombinationOfMALLUseForPStateAndStaticScreen = %d\n", support->InvalidCombinationOfMALLUseForPStateAndStaticScreen);
 	if (!fail_only || support->DSCCLKRequiredMoreThanSupported == 1)
-		dml2_printf("DML: support: DSCCLKRequiredMoreThanSupported = %d\n", support->DSCCLKRequiredMoreThanSupported);
+		DML_LOG_VERBOSE("DML: support: DSCCLKRequiredMoreThanSupported = %d\n", support->DSCCLKRequiredMoreThanSupported);
 	if (!fail_only || support->PixelsPerLinePerDSCUnitSupport == 0)
-		dml2_printf("DML: support: PixelsPerLinePerDSCUnitSupport = %d\n", support->PixelsPerLinePerDSCUnitSupport);
+		DML_LOG_VERBOSE("DML: support: PixelsPerLinePerDSCUnitSupport = %d\n", support->PixelsPerLinePerDSCUnitSupport);
 	if (!fail_only || support->DTBCLKRequiredMoreThanSupported == 1)
-		dml2_printf("DML: support: DTBCLKRequiredMoreThanSupported = %d\n", support->DTBCLKRequiredMoreThanSupported);
+		DML_LOG_VERBOSE("DML: support: DTBCLKRequiredMoreThanSupported = %d\n", support->DTBCLKRequiredMoreThanSupported);
 	if (!fail_only || support->InvalidCombinationOfMALLUseForPState == 1)
-		dml2_printf("DML: support: InvalidCombinationOfMALLUseForPState = %d\n", support->InvalidCombinationOfMALLUseForPState);
+		DML_LOG_VERBOSE("DML: support: InvalidCombinationOfMALLUseForPState = %d\n", support->InvalidCombinationOfMALLUseForPState);
 	if (!fail_only || support->ROBSupport == 0)
-		dml2_printf("DML: support: ROBSupport = %d\n", support->ROBSupport);
+		DML_LOG_VERBOSE("DML: support: ROBSupport = %d\n", support->ROBSupport);
 	if (!fail_only || support->OutstandingRequestsSupport == 0)
-		dml2_printf("DML: support: OutstandingRequestsSupport = %d\n", support->OutstandingRequestsSupport);
+		DML_LOG_VERBOSE("DML: support: OutstandingRequestsSupport = %d\n", support->OutstandingRequestsSupport);
 	if (!fail_only || support->OutstandingRequestsUrgencyAvoidance == 0)
-		dml2_printf("DML: support: OutstandingRequestsUrgencyAvoidance = %d\n", support->OutstandingRequestsUrgencyAvoidance);
+		DML_LOG_VERBOSE("DML: support: OutstandingRequestsUrgencyAvoidance = %d\n", support->OutstandingRequestsUrgencyAvoidance);
 	if (!fail_only || support->DISPCLK_DPPCLK_Support == 0)
-		dml2_printf("DML: support: DISPCLK_DPPCLK_Support = %d\n", support->DISPCLK_DPPCLK_Support);
+		DML_LOG_VERBOSE("DML: support: DISPCLK_DPPCLK_Support = %d\n", support->DISPCLK_DPPCLK_Support);
 	if (!fail_only || support->TotalAvailablePipesSupport == 0)
-		dml2_printf("DML: support: TotalAvailablePipesSupport = %d\n", support->TotalAvailablePipesSupport);
+		DML_LOG_VERBOSE("DML: support: TotalAvailablePipesSupport = %d\n", support->TotalAvailablePipesSupport);
 	if (!fail_only || support->NumberOfOTGSupport == 0)
-		dml2_printf("DML: support: NumberOfOTGSupport = %d\n", support->NumberOfOTGSupport);
+		DML_LOG_VERBOSE("DML: support: NumberOfOTGSupport = %d\n", support->NumberOfOTGSupport);
 	if (!fail_only || support->NumberOfHDMIFRLSupport == 0)
-		dml2_printf("DML: support: NumberOfHDMIFRLSupport = %d\n", support->NumberOfHDMIFRLSupport);
+		DML_LOG_VERBOSE("DML: support: NumberOfHDMIFRLSupport = %d\n", support->NumberOfHDMIFRLSupport);
 	if (!fail_only || support->NumberOfDP2p0Support == 0)
-		dml2_printf("DML: support: NumberOfDP2p0Support = %d\n", support->NumberOfDP2p0Support);
+		DML_LOG_VERBOSE("DML: support: NumberOfDP2p0Support = %d\n", support->NumberOfDP2p0Support);
 	if (!fail_only || support->EnoughWritebackUnits == 0)
-		dml2_printf("DML: support: EnoughWritebackUnits = %d\n", support->EnoughWritebackUnits);
+		DML_LOG_VERBOSE("DML: support: EnoughWritebackUnits = %d\n", support->EnoughWritebackUnits);
 	if (!fail_only || support->WritebackScaleRatioAndTapsSupport == 0)
-		dml2_printf("DML: support: WritebackScaleRatioAndTapsSupport = %d\n", support->WritebackScaleRatioAndTapsSupport);
+		DML_LOG_VERBOSE("DML: support: WritebackScaleRatioAndTapsSupport = %d\n", support->WritebackScaleRatioAndTapsSupport);
 	if (!fail_only || support->WritebackLatencySupport == 0)
-		dml2_printf("DML: support: WritebackLatencySupport = %d\n", support->WritebackLatencySupport);
+		DML_LOG_VERBOSE("DML: support: WritebackLatencySupport = %d\n", support->WritebackLatencySupport);
 	if (!fail_only || support->CursorSupport == 0)
-		dml2_printf("DML: support: CursorSupport = %d\n", support->CursorSupport);
+		DML_LOG_VERBOSE("DML: support: CursorSupport = %d\n", support->CursorSupport);
 	if (!fail_only || support->PitchSupport == 0)
-		dml2_printf("DML: support: PitchSupport = %d\n", support->PitchSupport);
+		DML_LOG_VERBOSE("DML: support: PitchSupport = %d\n", support->PitchSupport);
 	if (!fail_only || support->ViewportExceedsSurface == 1)
-		dml2_printf("DML: support: ViewportExceedsSurface = %d\n", support->ViewportExceedsSurface);
+		DML_LOG_VERBOSE("DML: support: ViewportExceedsSurface = %d\n", support->ViewportExceedsSurface);
 	if (!fail_only || support->PrefetchSupported == 0)
-		dml2_printf("DML: support: PrefetchSupported = %d\n", support->PrefetchSupported);
+		DML_LOG_VERBOSE("DML: support: PrefetchSupported = %d\n", support->PrefetchSupported);
 	if (!fail_only || support->EnoughUrgentLatencyHidingSupport == 0)
-		dml2_printf("DML: support: EnoughUrgentLatencyHidingSupport = %d\n", support->EnoughUrgentLatencyHidingSupport);
+		DML_LOG_VERBOSE("DML: support: EnoughUrgentLatencyHidingSupport = %d\n", support->EnoughUrgentLatencyHidingSupport);
 	if (!fail_only || support->AvgBandwidthSupport == 0)
-		dml2_printf("DML: support: AvgBandwidthSupport = %d\n", support->AvgBandwidthSupport);
+		DML_LOG_VERBOSE("DML: support: AvgBandwidthSupport = %d\n", support->AvgBandwidthSupport);
 	if (!fail_only || support->DynamicMetadataSupported == 0)
-		dml2_printf("DML: support: DynamicMetadataSupported = %d\n", support->DynamicMetadataSupported);
+		DML_LOG_VERBOSE("DML: support: DynamicMetadataSupported = %d\n", support->DynamicMetadataSupported);
 	if (!fail_only || support->VRatioInPrefetchSupported == 0)
-		dml2_printf("DML: support: VRatioInPrefetchSupported = %d\n", support->VRatioInPrefetchSupported);
+		DML_LOG_VERBOSE("DML: support: VRatioInPrefetchSupported = %d\n", support->VRatioInPrefetchSupported);
 	if (!fail_only || support->PTEBufferSizeNotExceeded == 0)
-		dml2_printf("DML: support: PTEBufferSizeNotExceeded = %d\n", support->PTEBufferSizeNotExceeded);
+		DML_LOG_VERBOSE("DML: support: PTEBufferSizeNotExceeded = %d\n", support->PTEBufferSizeNotExceeded);
 	if (!fail_only || support->DCCMetaBufferSizeNotExceeded == 0)
-		dml2_printf("DML: support: DCCMetaBufferSizeNotExceeded = %d\n", support->DCCMetaBufferSizeNotExceeded);
+		DML_LOG_VERBOSE("DML: support: DCCMetaBufferSizeNotExceeded = %d\n", support->DCCMetaBufferSizeNotExceeded);
 	if (!fail_only || support->ExceededMALLSize == 1)
-		dml2_printf("DML: support: ExceededMALLSize = %d\n", support->ExceededMALLSize);
+		DML_LOG_VERBOSE("DML: support: ExceededMALLSize = %d\n", support->ExceededMALLSize);
 	if (!fail_only || support->g6_temp_read_support == 0)
-		dml2_printf("DML: support: g6_temp_read_support = %d\n", support->g6_temp_read_support);
+		DML_LOG_VERBOSE("DML: support: g6_temp_read_support = %d\n", support->g6_temp_read_support);
 	if (!fail_only || support->ImmediateFlipSupport == 0)
-		dml2_printf("DML: support: ImmediateFlipSupport = %d\n", support->ImmediateFlipSupport);
+		DML_LOG_VERBOSE("DML: support: ImmediateFlipSupport = %d\n", support->ImmediateFlipSupport);
 	if (!fail_only || support->LinkCapacitySupport == 0)
-		dml2_printf("DML: support: LinkCapacitySupport = %d\n", support->LinkCapacitySupport);
+		DML_LOG_VERBOSE("DML: support: LinkCapacitySupport = %d\n", support->LinkCapacitySupport);
 
 	if (!fail_only || support->ModeSupport == 0)
-		dml2_printf("DML: support: ModeSupport = %d\n", support->ModeSupport);
-	dml2_printf("DML: ===================================== \n");
+		DML_LOG_VERBOSE("DML: support: ModeSupport = %d\n", support->ModeSupport);
+	DML_LOG_VERBOSE("DML: ===================================== \n");
 }
 
 const char *dml2_core_utils_internal_soc_state_type_str(enum dml2_core_internal_soc_state_type dml2_core_internal_soc_state_type)
@@ -358,9 +358,9 @@ void dml2_core_utils_get_stream_output_bpp(double *out_bpp, const struct dml2_di
 			out_bpp[k] = 0;
 		}
 #ifdef __DML_VBA_DEBUG__
-		dml2_printf("DML::%s: k=%d bpc=%f\n", __func__, k, bpc);
-		dml2_printf("DML::%s: k=%d dsc.enable=%d\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.dsc.enable);
-		dml2_printf("DML::%s: k=%d out_bpp=%f\n", __func__, k, out_bpp[k]);
+		DML_LOG_VERBOSE("DML::%s: k=%d bpc=%f\n", __func__, k, bpc);
+		DML_LOG_VERBOSE("DML::%s: k=%d dsc.enable=%d\n", __func__, k, display_cfg->stream_descriptors[display_cfg->plane_descriptors[k].stream_index].timing.dsc.enable);
+		DML_LOG_VERBOSE("DML::%s: k=%d out_bpp=%f\n", __func__, k, out_bpp[k]);
 #endif
 	}
 }
@@ -391,7 +391,7 @@ unsigned int dml2_core_util_get_num_active_pipes(int unsigned num_planes, const
 	}
 
 #ifdef __DML_VBA_DEBUG__
-	dml2_printf("DML::%s: num_active_pipes = %d\n", __func__, num_active_pipes);
+	DML_LOG_VERBOSE("DML::%s: num_active_pipes = %d\n", __func__, num_active_pipes);
 #endif
 	return num_active_pipes;
 }
@@ -452,7 +452,7 @@ unsigned int dml2_core_utils_get_tile_block_size_bytes(enum dml2_swizzle_mode sw
 	else if (sw_mode == dml2_gfx11_sw_256kb_r_x)
 		return 262144;
 	else {
-		DML2_ASSERT(0);
+		DML_ASSERT(0);
 		return 256;
 	};
 }
@@ -498,8 +498,8 @@ int unsigned dml2_core_utils_get_gfx_version(enum dml2_swizzle_mode sw_mode)
 		sw_mode == dml2_gfx11_sw_256kb_r_x)
 		version = 11;
 	else {
-		dml2_printf("ERROR: Invalid sw_mode setting! val=%u\n", sw_mode);
-		DML2_ASSERT(0);
+		DML_LOG_VERBOSE("ERROR: Invalid sw_mode setting! val=%u\n", sw_mode);
+		DML_ASSERT(0);
 	}
 
 	return version;
@@ -511,7 +511,7 @@ unsigned int dml2_core_utils_get_qos_param_index(unsigned long uclk_freq_khz, co
 	unsigned int index = 0;
 
 	for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) {
-		dml2_printf("DML::%s: per_uclk_dpm_params[%d].minimum_uclk_khz = %d\n", __func__, i, per_uclk_dpm_params[i].minimum_uclk_khz);
+		DML_LOG_VERBOSE("DML::%s: per_uclk_dpm_params[%d].minimum_uclk_khz = %ld\n", __func__, i, per_uclk_dpm_params[i].minimum_uclk_khz);
 
 		if (i == 0)
 			index = 0;
@@ -524,8 +524,8 @@ unsigned int dml2_core_utils_get_qos_param_index(unsigned long uclk_freq_khz, co
 		}
 	}
 #if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: uclk_freq_khz = %d\n", __func__, uclk_freq_khz);
-	dml2_printf("DML::%s: index = %d\n", __func__, index);
+	DML_LOG_VERBOSE("DML::%s: uclk_freq_khz = %ld\n", __func__, uclk_freq_khz);
+	DML_LOG_VERBOSE("DML::%s: index = %d\n", __func__, index);
 #endif
 	return index;
 }
@@ -533,32 +533,32 @@ unsigned int dml2_core_utils_get_qos_param_index(unsigned long uclk_freq_khz, co
 unsigned int dml2_core_utils_get_active_min_uclk_dpm_index(unsigned long uclk_freq_khz, const struct dml2_soc_state_table *clk_table)
 {
 	unsigned int i;
-	bool clk_entry_found = 0;
+	bool clk_entry_found = false;
 
 	for (i = 0; i < clk_table->uclk.num_clk_values; i++) {
-		dml2_printf("DML::%s: clk_table.uclk.clk_values_khz[%d] = %d\n", __func__, i, clk_table->uclk.clk_values_khz[i]);
+		DML_LOG_VERBOSE("DML::%s: clk_table.uclk.clk_values_khz[%d] = %ld\n", __func__, i, clk_table->uclk.clk_values_khz[i]);
 
 		if (uclk_freq_khz == clk_table->uclk.clk_values_khz[i]) {
-			clk_entry_found = 1;
+			clk_entry_found = true;
 			break;
 		}
 	}
 
 	if (!clk_entry_found)
-		DML2_ASSERT(clk_entry_found);
+		DML_ASSERT(clk_entry_found);
 #if defined(__DML_VBA_DEBUG__)
-	dml2_printf("DML::%s: uclk_freq_khz = %ld\n", __func__, uclk_freq_khz);
-	dml2_printf("DML::%s: index = %d\n", __func__, i);
+	DML_LOG_VERBOSE("DML::%s: uclk_freq_khz = %ld\n", __func__, uclk_freq_khz);
+	DML_LOG_VERBOSE("DML::%s: index = %d\n", __func__, i);
 #endif
 	return i;
 }
 
 bool dml2_core_utils_is_dual_plane(enum dml2_source_format_class source_format)
 {
-	bool ret_val = 0;
+	bool ret_val = false;
 
 	if (dml2_core_utils_is_420(source_format) || dml2_core_utils_is_422_planar(source_format) || (source_format == dml2_rgbe_alpha))
-		ret_val = 1;
+		ret_val = true;
 
 	return ret_val;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c
index 15507926f3a4..f486b090bbfc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c
@@ -754,6 +754,8 @@ bool dpmm_dcn4_map_watermarks(struct dml2_dpmm_map_watermarks_params_in_out *in_
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].fclk_pstate = (int unsigned)(mode_lib->mp.Watermark.FCLKChangeWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].sr_enter = (int unsigned)(mode_lib->mp.Watermark.StutterEnterPlusExitWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].sr_exit = (int unsigned)(mode_lib->mp.Watermark.StutterExitWatermark * refclk_freq_in_mhz);
+	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].sr_enter_z8 = (int unsigned)(mode_lib->mp.Watermark.Z8StutterEnterPlusExitWatermark * refclk_freq_in_mhz);
+	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].sr_exit_z8 = (int unsigned)(mode_lib->mp.Watermark.Z8StutterExitWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].temp_read_or_ppt = (int unsigned)(mode_lib->mp.Watermark.temp_read_or_ppt_watermark_us * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].uclk_pstate = (int unsigned)(mode_lib->mp.Watermark.DRAMClockChangeWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_A].urgent = (int unsigned)(mode_lib->mp.Watermark.UrgentWatermark * refclk_freq_in_mhz);
@@ -768,6 +770,8 @@ bool dpmm_dcn4_map_watermarks(struct dml2_dpmm_map_watermarks_params_in_out *in_
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].fclk_pstate = (int unsigned)(mode_lib->mp.Watermark.FCLKChangeWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].sr_enter = (int unsigned)(mode_lib->mp.Watermark.StutterEnterPlusExitWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].sr_exit = (int unsigned)(mode_lib->mp.Watermark.StutterExitWatermark * refclk_freq_in_mhz);
+	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].sr_enter_z8 = (int unsigned)(mode_lib->mp.Watermark.Z8StutterEnterPlusExitWatermark * refclk_freq_in_mhz);
+	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].sr_exit_z8 = (int unsigned)(mode_lib->mp.Watermark.Z8StutterExitWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].temp_read_or_ppt = (int unsigned)(mode_lib->mp.Watermark.temp_read_or_ppt_watermark_us * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].uclk_pstate = (int unsigned)(mode_lib->mp.Watermark.DRAMClockChangeWatermark * refclk_freq_in_mhz);
 	dchubbub_regs->wm_regs[DML2_DCHUB_WATERMARK_SET_B].urgent = (int unsigned)(mode_lib->mp.Watermark.UrgentWatermark * refclk_freq_in_mhz);
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.c
index f4b1a7d02d42..a265f254152c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.c
@@ -182,6 +182,10 @@ static bool build_min_clock_table(const struct dml2_soc_bb *soc_bb, struct dml2_
 	min_table->max_clocks_khz.dtbclk = soc_bb->clk_table.dtbclk.clk_values_khz[soc_bb->clk_table.dtbclk.num_clk_values - 1];
 	min_table->max_clocks_khz.phyclk = soc_bb->clk_table.phyclk.clk_values_khz[soc_bb->clk_table.phyclk.num_clk_values - 1];
 
+	min_table->max_ss_clocks_khz.dispclk = (unsigned int)((double)min_table->max_clocks_khz.dispclk / (1.0 + soc_bb->dcn_downspread_percent / 100.0));
+	min_table->max_ss_clocks_khz.dppclk = (unsigned int)((double)min_table->max_clocks_khz.dppclk / (1.0 + soc_bb->dcn_downspread_percent / 100.0));
+	min_table->max_ss_clocks_khz.dtbclk = (unsigned int)((double)min_table->max_clocks_khz.dtbclk / (1.0 + soc_bb->dcn_downspread_percent / 100.0));
+
 	min_table->max_clocks_khz.dcfclk = soc_bb->clk_table.dcfclk.clk_values_khz[soc_bb->clk_table.dcfclk.num_clk_values - 1];
 	min_table->max_clocks_khz.fclk = soc_bb->clk_table.fclk.clk_values_khz[soc_bb->clk_table.fclk.num_clk_values - 1];
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c
index f50662b83296..d88b3e0082dd 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c
@@ -659,7 +659,7 @@ bool pmo_dcn4_fams2_initialize(struct dml2_pmo_initialize_in_out *in_out)
 	for (i = 1; i <= PMO_DCN4_MAX_DISPLAYS; i++) {
 		switch (i) {
 		case 1:
-			DML2_ASSERT(base_strategy_list_1_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
+			DML_ASSERT(base_strategy_list_1_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
 
 			/* populate list */
 			pmo_dcn4_fams2_expand_base_pstate_strategies(
@@ -670,7 +670,7 @@ bool pmo_dcn4_fams2_initialize(struct dml2_pmo_initialize_in_out *in_out)
 					&pmo->init_data.pmo_dcn4.num_expanded_strategies_per_list[i - 1]);
 			break;
 		case 2:
-			DML2_ASSERT(base_strategy_list_2_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
+			DML_ASSERT(base_strategy_list_2_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
 
 			/* populate list */
 			pmo_dcn4_fams2_expand_base_pstate_strategies(
@@ -681,7 +681,7 @@ bool pmo_dcn4_fams2_initialize(struct dml2_pmo_initialize_in_out *in_out)
 					&pmo->init_data.pmo_dcn4.num_expanded_strategies_per_list[i - 1]);
 			break;
 		case 3:
-			DML2_ASSERT(base_strategy_list_3_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
+			DML_ASSERT(base_strategy_list_3_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
 
 			/* populate list */
 			pmo_dcn4_fams2_expand_base_pstate_strategies(
@@ -692,7 +692,7 @@ bool pmo_dcn4_fams2_initialize(struct dml2_pmo_initialize_in_out *in_out)
 					&pmo->init_data.pmo_dcn4.num_expanded_strategies_per_list[i - 1]);
 			break;
 		case 4:
-			DML2_ASSERT(base_strategy_list_4_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
+			DML_ASSERT(base_strategy_list_4_display_size <= PMO_DCN4_MAX_BASE_STRATEGIES);
 
 			/* populate list */
 			pmo_dcn4_fams2_expand_base_pstate_strategies(
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.c
index dc2ce5e77f57..4a7c4c62111e 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.c
@@ -761,7 +761,7 @@ bool dml2_top_mcache_calc_mcache_count_and_offsets(struct top_mcache_calc_mcache
 				total_mcaches_required--;
 		}
 	}
-	dml2_printf("DML_CORE_DCN3::%s: plane_%d, total_mcaches_required=%d\n", __func__, i, total_mcaches_required);
+	DML_LOG_VERBOSE("DML_CORE_DCN3::%s: plane_%d, total_mcaches_required=%d\n", __func__, i, total_mcaches_required);
 
 	if (total_mcaches_required > dml->soc_bbox.num_dcc_mcaches) {
 		result = false;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.c
deleted file mode 100644
index c506667897c4..000000000000
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.c
+++ /dev/null
@@ -1,31 +0,0 @@
-// SPDX-License-Identifier: MIT
-//
-// Copyright 2024 Advanced Micro Devices, Inc.
-
-#include "dml2_debug.h"
-
-int dml2_log_internal(const char *format, ...)
-{
-	return 0;
-}
-
-int dml2_printf(const char *format, ...)
-{
-#ifdef _DEBUG
-#ifdef _DEBUG_PRINTS
-	int result;
-	va_list args;
-	va_start(args, format);
-
-	result = vprintf(format, args);
-
-	va_end(args);
-
-	return result;
-#else
-	return 0;
-#endif
-#else
-	return 0;
-#endif
-}
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h
index bfe6f236d2e4..b226225103c3 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h
@@ -5,55 +5,62 @@
 #ifndef __DML2_DEBUG_H__
 #define __DML2_DEBUG_H__
 
-#ifndef DML2_ASSERT
-#define DML2_ASSERT(condition) ((void)0)
-#endif
+#include "os_types.h"
+#define DML_ASSERT(condition) ASSERT(condition)
+#define DML_LOG_LEVEL_DEFAULT DML_LOG_LEVEL_WARN
+#define DML_LOG_INTERNAL(fmt, ...) dm_output_to_console(fmt, ## __VA_ARGS__)
 
-/*
- * DML_LOG_FATAL - fatal errors for unrecoverable DML states until a restart.
- * DML_LOG_ERROR - unexpected but recoverable failures inside DML
- * DML_LOG_WARN - unexpected inputs or events to DML
- * DML_LOG_INFO - high level tracing of DML interfaces
- * DML_LOG_DEBUG - detailed tracing of DML internal components
- * DML_LOG_VERBOSE - detailed tracing of DML calculation procedure
- */
-#if !defined(DML_LOG_LEVEL)
-#if defined(_DEBUG) && defined(_DEBUG_PRINTS)
-/* for backward compatibility with old macros */
-#define DML_LOG_LEVEL 5
-#else
-#define DML_LOG_LEVEL 0
-#endif
-#endif
+/* ASSERT with message output */
+#define DML_ASSERT_MSG(condition, fmt, ...)								\
+	do {												\
+		if (!(condition)) {									\
+			DML_LOG_ERROR("DML ASSERT hit in %s line %d\n", __func__, __LINE__);	\
+			DML_LOG_ERROR(fmt, ## __VA_ARGS__);						\
+			DML_ASSERT(condition);								\
+		}											\
+	} while (0)
+
+/* fatal errors for unrecoverable DML states until a full reset */
+#define DML_LOG_LEVEL_FATAL 0
+/* unexpected but recoverable failures inside DML */
+#define DML_LOG_LEVEL_ERROR 1
+/* unexpected inputs or events to DML */
+#define DML_LOG_LEVEL_WARN 2
+/* high level tracing of DML interfaces */
+#define DML_LOG_LEVEL_INFO 3
+/* detailed tracing of DML internal components */
+#define DML_LOG_LEVEL_DEBUG 4
+/* detailed tracing of DML calculation procedure */
+#define DML_LOG_LEVEL_VERBOSE 5
 
-#define DML_LOG_FATAL(fmt, ...) dml2_log_internal(fmt, ## __VA_ARGS__)
-#if DML_LOG_LEVEL >= 1
-#define DML_LOG_ERROR(fmt, ...) dml2_log_internal(fmt, ## __VA_ARGS__)
+#ifndef DML_LOG_LEVEL
+#define DML_LOG_LEVEL DML_LOG_LEVEL_DEFAULT
+#endif /* #ifndef DML_LOG_LEVEL */
+
+#define DML_LOG_FATAL(fmt, ...) DML_LOG_INTERNAL("[DML FATAL] " fmt, ## __VA_ARGS__)
+#if DML_LOG_LEVEL >= DML_LOG_LEVEL_ERROR
+#define DML_LOG_ERROR(fmt, ...) DML_LOG_INTERNAL("[DML ERROR] "fmt, ## __VA_ARGS__)
 #else
 #define DML_LOG_ERROR(fmt, ...) ((void)0)
 #endif
-#if DML_LOG_LEVEL >= 2
-#define DML_LOG_WARN(fmt, ...) dml2_log_internal(fmt, ## __VA_ARGS__)
+#if DML_LOG_LEVEL >= DML_LOG_LEVEL_WARN
+#define DML_LOG_WARN(fmt, ...) DML_LOG_INTERNAL("[DML WARN] "fmt, ## __VA_ARGS__)
 #else
 #define DML_LOG_WARN(fmt, ...) ((void)0)
 #endif
-#if DML_LOG_LEVEL >= 3
-#define DML_LOG_INFO(fmt, ...) dml2_log_internal(fmt, ## __VA_ARGS__)
+#if DML_LOG_LEVEL >= DML_LOG_LEVEL_INFO
+#define DML_LOG_INFO(fmt, ...) DML_LOG_INTERNAL("[DML INFO] "fmt, ## __VA_ARGS__)
 #else
 #define DML_LOG_INFO(fmt, ...) ((void)0)
 #endif
-#if DML_LOG_LEVEL >= 4
-#define DML_LOG_DEBUG(fmt, ...) dml2_log_internal(fmt, ## __VA_ARGS__)
+#if DML_LOG_LEVEL >= DML_LOG_LEVEL_DEBUG
+#define DML_LOG_DEBUG(fmt, ...) DML_LOG_INTERNAL("[DML DEBUG] "fmt, ## __VA_ARGS__)
 #else
 #define DML_LOG_DEBUG(fmt, ...) ((void)0)
 #endif
-#if DML_LOG_LEVEL >= 5
-#define DML_LOG_VERBOSE(fmt, ...) dml2_log_internal(fmt, ## __VA_ARGS__)
+#if DML_LOG_LEVEL >= DML_LOG_LEVEL_VERBOSE
+#define DML_LOG_VERBOSE(fmt, ...) DML_LOG_INTERNAL("[DML VERBOSE] "fmt, ## __VA_ARGS__)
 #else
 #define DML_LOG_VERBOSE(fmt, ...) ((void)0)
 #endif
-
-int dml2_log_internal(const char *format, ...);
-int dml2_printf(const char *format, ...);
-
-#endif
+#endif /* __DML2_DEBUG_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h
index d8d01dceacdd..00688b9f1df4 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h
@@ -38,6 +38,12 @@ struct dml2_mcg_min_clock_table {
 	} max_clocks_khz;
 
 	struct {
+		unsigned int dispclk;
+		unsigned int dppclk;
+		unsigned int dtbclk;
+	} max_ss_clocks_khz;
+
+	struct {
 		unsigned int dprefclk;
 		unsigned int xtalclk;
 		unsigned int pcierefclk;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
index a966abd40788..5f1b49a50049 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
@@ -1082,22 +1082,22 @@ bool dml2_map_dc_pipes(struct dml2_context *ctx, struct dc_state *state, const s
 		if (stream_disp_cfg_index >= disp_cfg_index_max)
 			continue;
 
-		if (ODMMode[stream_disp_cfg_index] == dml_odm_mode_bypass) {
-			scratch.odm_info.odm_factor = 1;
-		} else if (ODMMode[stream_disp_cfg_index] == dml_odm_mode_combine_2to1) {
-			scratch.odm_info.odm_factor = 2;
-		} else if (ODMMode[stream_disp_cfg_index] == dml_odm_mode_combine_4to1) {
-			scratch.odm_info.odm_factor = 4;
-		} else {
-			ASSERT(false);
-			scratch.odm_info.odm_factor = 1;
-		}
-
+		if (ctx->architecture == dml2_architecture_20) {
+			if (ODMMode[stream_disp_cfg_index] == dml_odm_mode_bypass) {
+				scratch.odm_info.odm_factor = 1;
+			} else if (ODMMode[stream_disp_cfg_index] == dml_odm_mode_combine_2to1) {
+				scratch.odm_info.odm_factor = 2;
+			} else if (ODMMode[stream_disp_cfg_index] == dml_odm_mode_combine_4to1) {
+				scratch.odm_info.odm_factor = 4;
+			} else {
+				ASSERT(false);
+				scratch.odm_info.odm_factor = 1;
+			}
+		} else if (ctx->architecture == dml2_architecture_21) {
 		/* After DML2.1 update, ODM interpretation needs to change and is no longer same as for DML2.0.
 		 * This is not an issue with new resource management logic. This block ensure backcompat
 		 * with legacy pipe management with updated DML.
 		 * */
-		if (ctx->architecture == dml2_architecture_21) {
 			if (ODMMode[stream_disp_cfg_index] == 1) {
 				scratch.odm_info.odm_factor = 1;
 			} else if (ODMMode[stream_disp_cfg_index] == 2) {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c
index 2061d43b92e1..5de775fd8fce 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c
@@ -896,7 +896,7 @@ static void populate_dummy_dml_surface_cfg(struct dml_surface_cfg_st *out, unsig
 	out->SurfaceWidthC[location] = in->timing.h_addressable;
 	out->SurfaceHeightC[location] = in->timing.v_addressable;
 	out->PitchY[location] = ((out->SurfaceWidthY[location] + 127) / 128) * 128;
-	out->PitchC[location] = 0;
+	out->PitchC[location] = 1;
 	out->DCCEnable[location] = false;
 	out->DCCMetaPitchY[location] = 0;
 	out->DCCMetaPitchC[location] = 0;
@@ -973,7 +973,9 @@ static void populate_dml_surface_cfg_from_plane_state(enum dml_project_id dml2_p
 	}
 }
 
-static void get_scaler_data_for_plane(const struct dc_plane_state *in, struct dc_state *context, struct scaler_data *out)
+static struct scaler_data *get_scaler_data_for_plane(
+		const struct dc_plane_state *in,
+		struct dc_state *context)
 {
 	int i;
 	struct pipe_ctx *temp_pipe = &context->res_ctx.temp_pipe;
@@ -994,7 +996,7 @@ static void get_scaler_data_for_plane(const struct dc_plane_state *in, struct dc
 	}
 
 	ASSERT(i < MAX_PIPES);
-	memcpy(out, &temp_pipe->plane_res.scl_data, sizeof(*out));
+	return &temp_pipe->plane_res.scl_data;
 }
 
 static void populate_dummy_dml_plane_cfg(struct dml_plane_cfg_st *out, unsigned int location,
@@ -1057,11 +1059,7 @@ static void populate_dml_plane_cfg_from_plane_state(struct dml_plane_cfg_st *out
 						    const struct dc_plane_state *in, struct dc_state *context,
 						    const struct soc_bounding_box_st *soc)
 {
-	struct scaler_data *scaler_data = kzalloc(sizeof(*scaler_data), GFP_KERNEL);
-	if (!scaler_data)
-		return;
-
-	get_scaler_data_for_plane(in, context, scaler_data);
+	struct scaler_data *scaler_data = get_scaler_data_for_plane(in, context);
 
 	out->CursorBPP[location] = dml_cur_32bit;
 	out->CursorWidth[location] = 256;
@@ -1126,8 +1124,6 @@ static void populate_dml_plane_cfg_from_plane_state(struct dml_plane_cfg_st *out
 	out->DynamicMetadataTransmittedBytes[location] = 0;
 
 	out->NumberOfCursors[location] = 1;
-
-	kfree(scaler_data);
 }
 
 static unsigned int map_stream_to_dml_display_cfg(const struct dml2_context *dml2,
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
index 939ee0708bd2..525b7d04bf84 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
@@ -24,6 +24,8 @@
  *
  */
 
+#include <linux/vmalloc.h>
+
 #include "display_mode_core.h"
 #include "dml2_internal_types.h"
 #include "dml2_utils.h"
@@ -661,7 +663,10 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s
 		dml2_copy_clocks_to_dc_state(&out_clks, context);
 		dml2_extract_watermark_set(&context->bw_ctx.bw.dcn.watermarks.a, &dml2->v20.dml_core_ctx);
 		dml2_extract_watermark_set(&context->bw_ctx.bw.dcn.watermarks.b, &dml2->v20.dml_core_ctx);
-		memcpy(&context->bw_ctx.bw.dcn.watermarks.c, &dml2->v20.g6_temp_read_watermark_set, sizeof(context->bw_ctx.bw.dcn.watermarks.c));
+		if (context->streams[0]->sink->link->dc->caps.is_apu)
+			dml2_extract_watermark_set(&context->bw_ctx.bw.dcn.watermarks.c, &dml2->v20.dml_core_ctx);
+		else
+			memcpy(&context->bw_ctx.bw.dcn.watermarks.c, &dml2->v20.g6_temp_read_watermark_set, sizeof(context->bw_ctx.bw.dcn.watermarks.c));
 		dml2_extract_watermark_set(&context->bw_ctx.bw.dcn.watermarks.d, &dml2->v20.dml_core_ctx);
 		dml2_extract_writeback_wm(context, &dml2->v20.dml_core_ctx);
 		//copy for deciding zstate use
@@ -732,17 +737,22 @@ bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2
 		return out;
 	}
 
+	DC_FP_START();
+
 	/* Use dml_validate_only for fast_validate path */
 	if (fast_validate)
 		out = dml2_validate_only(context);
 	else
 		out = dml2_validate_and_build_resource(in_dc, context);
+
+	DC_FP_END();
+
 	return out;
 }
 
 static inline struct dml2_context *dml2_allocate_memory(void)
 {
-	return (struct dml2_context *) kzalloc(sizeof(struct dml2_context), GFP_KERNEL);
+	return (struct dml2_context *) vzalloc(sizeof(struct dml2_context));
 }
 
 static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2)
@@ -779,11 +789,15 @@ static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_op
 		break;
 	}
 
+	DC_FP_START();
+
 	initialize_dml2_ip_params(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.ip);
 
 	initialize_dml2_soc_bbox(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.soc);
 
 	initialize_dml2_soc_states(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.soc, &(*dml2)->v20.dml_core_ctx.states);
+
+	DC_FP_END();
 }
 
 bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2)
@@ -812,7 +826,7 @@ void dml2_destroy(struct dml2_context *dml2)
 
 	if (dml2->architecture == dml2_architecture_21)
 		dml21_destroy(dml2);
-	kfree(dml2);
+	vfree(dml2);
 }
 
 void dml2_extract_dram_and_fclk_change_support(struct dml2_context *dml2,
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h
index 785226945699..5100f269368e 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h
@@ -40,6 +40,7 @@ struct dc_sink;
 struct dc_stream_state;
 struct resource_context;
 struct display_stream_compressor;
+struct dc_mcache_params;
 
 // Configuration of the MALL on the SoC
 struct dml2_soc_mall_info {
@@ -107,6 +108,7 @@ struct dml2_dc_callbacks {
 	unsigned int (*get_max_flickerless_instant_vtotal_increase)(
 			struct dc_stream_state *stream,
 			bool is_gaming);
+	bool (*allocate_mcache)(struct dc_state *context, const struct dc_mcache_params *mcache_params);
 };
 
 struct dml2_dc_svp_callbacks {
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c
index abf439e743f2..2d70586cef40 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c
@@ -790,8 +790,7 @@ static bool dpp3_program_blnd_lut(struct dpp *dpp_base,
 
 	if (params == NULL) {
 		REG_SET(CM_BLNDGAM_CONTROL, 0, CM_BLNDGAM_MODE, 0);
-		if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
-			dpp3_power_on_blnd_lut(dpp_base, false);
+		dpp3_power_on_blnd_lut(dpp_base, false);
 		return false;
 	}
 
@@ -1204,8 +1203,7 @@ static bool dpp3_program_shaper(struct dpp *dpp_base,
 
 	if (params == NULL) {
 		REG_SET(CM_SHAPER_CONTROL, 0, CM_SHAPER_LUT_MODE, 0);
-		if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
-			dpp3_power_on_shaper(dpp_base, false);
+		dpp3_power_on_shaper(dpp_base, false);
 		return false;
 	}
 
@@ -1399,8 +1397,7 @@ static bool dpp3_program_3dlut(struct dpp *dpp_base,
 
 	if (params == NULL) {
 		dpp3_set_3dlut_mode(dpp_base, LUT_BYPASS, false, false);
-		if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.cm)
-			dpp3_power_on_hdr3dlut(dpp_base, false);
+		dpp3_power_on_hdr3dlut(dpp_base, false);
 		return false;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c
index 62b7012cda43..f7a373a3d70a 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c
@@ -138,7 +138,7 @@ bool dpp35_construct(
 	dpp->base.funcs = &dcn35_dpp_funcs;
 
 	// w/a for cursor memory stuck in LS by programming DISPCLK_R_GATE_DISABLE, limit w/a to some ASIC revs
-	if (dpp->base.ctx->asic_id.hw_internal_rev <= 0x10)
+	if (dpp->base.ctx->asic_id.hw_internal_rev < 0x40)
 		dpp->dispclk_r_gate_disable = true;
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c
index 1236e0f9a256..712aff7e17f7 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c
@@ -120,10 +120,11 @@ void dpp401_set_cursor_attributes(
 	enum dc_cursor_color_format color_format = cursor_attributes->color_format;
 	int cur_rom_en = 0;
 
-	// DCN4 should always do Cursor degamma for Cursor Color modes
 	if (color_format == CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA ||
 		color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) {
-		cur_rom_en = 1;
+		if (cursor_attributes->attribute_flags.bits.ENABLE_CURSOR_DEGAMMA) {
+			cur_rom_en = 1;
+		}
 	}
 
 	REG_UPDATE_3(CURSOR0_CONTROL,
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c
index 75128fd34306..bd1b9aef6d5c 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c
@@ -57,13 +57,6 @@ static const struct dsc_funcs dcn20_dsc_funcs = {
 #define DC_LOGGER \
 	dsc->ctx->logger
 
-enum dsc_bits_per_comp {
-	DSC_BPC_8 = 8,
-	DSC_BPC_10 = 10,
-	DSC_BPC_12 = 12,
-	DSC_BPC_UNKNOWN
-};
-
 /* API functions (external or via structure->function_pointer) */
 
 void dsc2_construct(struct dcn20_dsc *dsc,
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.h b/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.h
index 1fb90b52b814..a9c04fc95bd1 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.h
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.h
@@ -457,6 +457,12 @@
 	type DSCRM_DSC_DOUBLE_BUFFER_REG_UPDATE_PENDING; \
 	type DSCRM_DSC_FORWARD_EN_STATUS
 
+enum dsc_bits_per_comp {
+	DSC_BPC_8 = 8,
+	DSC_BPC_10 = 10,
+	DSC_BPC_12 = 12,
+	DSC_BPC_UNKNOWN
+};
 
 struct dcn20_dsc_registers {
 	uint32_t DSC_TOP_CONTROL;
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c
index 4893b793fec0..4222679fd4c9 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c
@@ -45,12 +45,6 @@ static const struct dsc_funcs dcn401_dsc_funcs = {
 #define DC_LOGGER \
 	dsc->ctx->logger
 
-enum dsc_bits_per_comp {
-	DSC_BPC_8 = 8,
-	DSC_BPC_10 = 10,
-	DSC_BPC_12 = 12,
-	DSC_BPC_UNKNOWN
-};
 
 /* API functions (external or via structure->function_pointer) */
 
diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c
index b099989d9364..942d9f0b6df2 100644
--- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c
+++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c
@@ -411,6 +411,20 @@ enum dc_irq_source dal_irq_get_rx_source(
 	}
 }
 
+enum dc_irq_source dal_irq_get_read_request(
+	const struct gpio *irq)
+{
+	enum gpio_id id = dal_gpio_get_id(irq);
+
+	switch (id) {
+	case GPIO_ID_HPD:
+		return (enum dc_irq_source)(DC_IRQ_SOURCE_DCI2C_RR_DDC1 +
+			dal_gpio_get_enum(irq));
+	default:
+		return DC_IRQ_SOURCE_INVALID;
+	}
+}
+
 enum gpio_result dal_irq_setup_hpd_filter(
 	struct gpio *irq,
 	struct gpio_hpd_config *config)
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn21/dcn21_hubbub.c
index 2546224b326a..e4496ad203b2 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn21/dcn21_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn21/dcn21_hubbub.c
@@ -132,9 +132,9 @@ int hubbub21_init_dchub(struct hubbub *hubbub,
 		// Init VMID 0 based on PA config
 		dcn20_vmid_setup(&hubbub1->vmid[0], &phys_config);
 	}
-
-	dcn21_dchvm_init(hubbub);
-
+	if (!hubbub1->base.ctx->dc->config.skip_riommu_prefetch_wa) {
+		dcn21_dchvm_init(hubbub);
+	}
 	return hubbub1->num_vmid;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
index 5ed195377a6c..baed31611477 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
@@ -1032,7 +1032,7 @@ static struct hubp_funcs dcn401_hubp_funcs = {
 	.hubp_program_3dlut_fl_tmz_protected = hubp401_program_3dlut_fl_tmz_protected,
 	.hubp_program_3dlut_fl_crossbar = hubp401_program_3dlut_fl_crossbar,
 	.hubp_get_3dlut_fl_done = hubp401_get_3dlut_fl_done,
-	.hubp_clear_tiling = hubp2_clear_tiling,
+	.hubp_clear_tiling = hubp401_clear_tiling,
 };
 
 bool hubp401_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/Makefile b/drivers/gpu/drm/amd/display/dc/hwss/Makefile
index 40ecebea1ba0..bee617ca0838 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/hwss/Makefile
@@ -27,6 +27,24 @@
 #  DCE
 ###############################################################################
 
+ifdef CONFIG_DRM_AMD_DC_SI
+HWSS_DCE60 = dce60_hwseq.o
+
+AMD_DAL_HWSS_DCE60 = $(addprefix $(AMDDALPATH)/dc/hwss/dce60/,$(HWSS_DCE60))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCE60)
+endif
+
+###############################################################################
+
+HWSS_DCE80 = dce80_hwseq.o
+
+AMD_DAL_HWSS_DCE80 = $(addprefix $(AMDDALPATH)/dc/hwss/dce80/,$(HWSS_DCE80))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCE80)
+
+###############################################################################
+
 HWSS_DCE = dce_hwseq.o
 
 AMD_DAL_HWSS_DCE = $(addprefix $(AMDDALPATH)/dc/hwss/dce/,$(HWSS_DCE))
@@ -65,14 +83,6 @@ AMD_DAL_HWSS_DCE120 = $(addprefix $(AMDDALPATH)/dc/hwss/dce120/,$(HWSS_DCE120))
 
 AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCE120)
 
-###############################################################################
-
-HWSS_DCE80 = dce80_hwseq.o
-
-AMD_DAL_HWSS_DCE80 = $(addprefix $(AMDDALPATH)/dc/hwss/dce80/,$(HWSS_DCE80))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_HWSS_DCE80)
-
 ifdef CONFIG_DRM_AMD_DC_FP
 ###############################################################################
 # DCN
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index 5656d10368ad..23bec5d25ed6 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -2763,12 +2763,12 @@ static void dce110_enable_per_frame_crtc_position_reset(
 
 }
 
-static void init_pipes(struct dc *dc, struct dc_state *context)
+static void dce110_init_pipes(struct dc *dc, struct dc_state *context)
 {
 	// Do nothing
 }
 
-static void init_hw(struct dc *dc)
+static void dce110_init_hw(struct dc *dc)
 {
 	int i;
 	struct dc_bios *bp;
@@ -3327,7 +3327,7 @@ void dce110_disable_link_output(struct dc_link *link,
 static const struct hw_sequencer_funcs dce110_funcs = {
 	.program_gamut_remap = program_gamut_remap,
 	.program_output_csc = program_output_csc,
-	.init_hw = init_hw,
+	.init_hw = dce110_init_hw,
 	.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
 	.apply_ctx_for_surface = dce110_apply_ctx_for_surface,
 	.post_unlock_program_front_end = dce110_post_unlock_program_front_end,
@@ -3371,7 +3371,7 @@ static const struct hw_sequencer_funcs dce110_funcs = {
 };
 
 static const struct hwseq_private_funcs dce110_private_funcs = {
-	.init_pipes = init_pipes,
+	.init_pipes = dce110_init_pipes,
 	.set_input_transfer_func = dce110_set_input_transfer_func,
 	.set_output_transfer_func = dce110_set_output_transfer_func,
 	.power_down = dce110_power_down,
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/hwss/dce60/dce60_hwseq.c
index 44b56490e152..a08e9f9eec17 100644
--- a/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce60/dce60_hwseq.c
@@ -26,7 +26,7 @@
 #include "dm_services.h"
 #include "dc.h"
 #include "core_types.h"
-#include "dce60_hw_sequencer.h"
+#include "dce60_hwseq.h"
 
 #include "dce/dce_hwseq.h"
 #include "dce110/dce110_hwseq.h"
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/dce60/dce60_hwseq.h
index f3b2d8b60d5b..f3b2d8b60d5b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce60/dce60_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce60/dce60_hwseq.h
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index 912f96323ed6..f9ee55998b6b 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -94,6 +94,128 @@ static void print_microsec(struct dc_context *dc_ctx,
 			us_x10 % frac);
 }
 
+/*
+ * Delay until we passed busy-until-point to which we can
+ * do necessary locking/programming on consecutive full updates
+ */
+void dcn10_wait_for_pipe_update_if_needed(struct dc *dc, struct pipe_ctx *pipe_ctx, bool is_surface_update_only)
+{
+	struct crtc_position position;
+	struct dc_stream_state *stream = pipe_ctx->stream;
+	unsigned int vpos, frame_count;
+	uint32_t vupdate_start, vupdate_end, vblank_start;
+	unsigned int lines_to_vupdate, us_to_vupdate;
+	unsigned int us_per_line, us_vupdate;
+
+	if (!pipe_ctx->stream ||
+		!pipe_ctx->stream_res.tg ||
+		!pipe_ctx->stream_res.stream_enc)
+		return;
+
+	if (pipe_ctx->prev_odm_pipe &&
+				pipe_ctx->stream)
+		return;
+
+	if (!pipe_ctx->wait_is_required)
+		return;
+
+	struct timing_generator *tg = pipe_ctx->stream_res.tg;
+
+	if (tg->funcs->is_tg_enabled && !tg->funcs->is_tg_enabled(tg))
+		return;
+
+	dc->hwss.calc_vupdate_position(dc, pipe_ctx, &vupdate_start,
+						&vupdate_end);
+
+	dc->hwss.get_position(&pipe_ctx, 1, &position);
+	vpos = position.vertical_count;
+
+	frame_count = tg->funcs->get_frame_count(tg);
+
+	if (frame_count - pipe_ctx->wait_frame_count > 2)
+		return;
+
+	vblank_start = pipe_ctx->pipe_dlg_param.vblank_start;
+
+	if (vpos >= vupdate_start && vupdate_start >= vblank_start)
+		lines_to_vupdate = stream->timing.v_total - vpos + vupdate_start;
+	else
+		lines_to_vupdate = vupdate_start - vpos;
+
+	us_per_line =
+		stream->timing.h_total * 10000u / stream->timing.pix_clk_100hz;
+	us_to_vupdate = lines_to_vupdate * us_per_line;
+
+	if (vupdate_end < vupdate_start)
+		vupdate_end += stream->timing.v_total;
+
+	if (lines_to_vupdate > stream->timing.v_total - vupdate_end + vupdate_start)
+		us_to_vupdate = 0;
+
+	us_vupdate = (vupdate_end - vupdate_start + 1) * us_per_line;
+
+	if (is_surface_update_only && us_to_vupdate + us_vupdate > 200) {
+		//surface updates come in at high irql
+		pipe_ctx->wait_is_required = true;
+		return;
+	}
+
+	fsleep(us_to_vupdate + us_vupdate);
+
+	//clear
+	pipe_ctx->next_vupdate = 0;
+	pipe_ctx->wait_frame_count = 0;
+	pipe_ctx->wait_is_required = false;
+}
+
+/*
+ * On pipe unlock and programming, indicate pipe will be busy
+ * until some frame and line (vupdate), this is required for consecutive
+ * full updates, need to wait for updates
+ * to latch to try and program the next update
+ */
+void dcn10_set_wait_for_update_needed_for_pipe(struct dc *dc, struct pipe_ctx *pipe_ctx)
+{
+	uint32_t vupdate_start, vupdate_end;
+	struct crtc_position position;
+	unsigned int vpos, cur_frame;
+
+	if (!pipe_ctx->stream ||
+		!pipe_ctx->stream_res.tg ||
+		!pipe_ctx->stream_res.stream_enc)
+		return;
+
+	dc->hwss.get_position(&pipe_ctx, 1, &position);
+	vpos = position.vertical_count;
+
+	dc->hwss.calc_vupdate_position(dc, pipe_ctx, &vupdate_start,
+						&vupdate_end);
+
+	struct timing_generator *tg = pipe_ctx->stream_res.tg;
+
+	struct optc *optc1 = DCN10TG_FROM_TG(tg);
+
+	ASSERT(optc1->max_frame_count != 0);
+
+	if (tg->funcs->is_tg_enabled && !tg->funcs->is_tg_enabled(tg))
+		return;
+
+	pipe_ctx->next_vupdate = vupdate_start;
+
+	cur_frame = tg->funcs->get_frame_count(tg);
+
+	if (vpos < vupdate_start) {
+		pipe_ctx->wait_frame_count = cur_frame;
+	} else {
+		if (cur_frame + 1 > optc1->max_frame_count)
+			pipe_ctx->wait_frame_count = cur_frame + 1 - optc1->max_frame_count;
+		else
+			pipe_ctx->wait_frame_count = cur_frame + 1;
+	}
+
+	pipe_ctx->wait_is_required = true;
+}
+
 void dcn10_lock_all_pipes(struct dc *dc,
 	struct dc_state *context,
 	bool lock)
@@ -2664,7 +2786,6 @@ void dcn10_update_visual_confirm_color(struct dc *dc,
 	struct mpc *mpc = dc->res_pool->mpc;
 
 	if (mpc->funcs->set_bg_color) {
-		memcpy(&pipe_ctx->plane_state->visual_confirm_color, &(pipe_ctx->visual_confirm_color), sizeof(struct tg_color));
 		mpc->funcs->set_bg_color(mpc, &(pipe_ctx->visual_confirm_color), mpcc_id);
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h
index 42ffd1e1299c..57d30ea225f2 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.h
@@ -50,6 +50,13 @@ void dcn10_optimize_bandwidth(
 void dcn10_prepare_bandwidth(
 		struct dc *dc,
 		struct dc_state *context);
+void dcn10_wait_for_pipe_update_if_needed(
+		struct dc *dc,
+		struct pipe_ctx *pipe_ctx,
+		bool is_surface_update_only);
+void dcn10_set_wait_for_update_needed_for_pipe(
+	struct dc *dc,
+	struct pipe_ctx *pipe_ctx);
 void dcn10_pipe_control_lock(
 	struct dc *dc,
 	struct pipe_ctx *pipe,
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index 846c9c51f2d9..858288c3b1ac 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -2053,7 +2053,7 @@ void dcn20_program_front_end_for_ctx(
 		for (i = 0; i < dc->res_pool->pipe_count; i++) {
 			pipe = &context->res_ctx.pipe_ctx[i];
 
-			if (!pipe->top_pipe && !pipe->prev_odm_pipe && pipe->plane_state) {
+			if (pipe->plane_state) {
 				ASSERT(!pipe->plane_state->triplebuffer_flips);
 				/*turn off triple buffer for full update*/
 				dc->hwss.program_triplebuffer(
@@ -2482,7 +2482,7 @@ bool dcn20_update_bandwidth(
 	struct dce_hwseq *hws = dc->hwseq;
 
 	/* recalculate DML parameters */
-	if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false))
+	if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK)
 		return false;
 
 	/* apply updated bandwidth parameters */
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
index be26c925fdfa..e68f21fd5f0f 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
@@ -84,6 +84,20 @@ static void update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 		struct dsc_config dsc_cfg;
 		struct dsc_optc_config dsc_optc_cfg = {0};
 		enum optc_dsc_mode optc_dsc_mode;
+		struct dcn_dsc_state dsc_state = {0};
+
+		if (!dsc) {
+			DC_LOG_DSC("DSC is NULL for tg instance %d:", pipe_ctx->stream_res.tg->inst);
+			return;
+		}
+
+		if (dsc->funcs->dsc_read_state) {
+			dsc->funcs->dsc_read_state(dsc, &dsc_state);
+			if (!dsc_state.dsc_fw_en) {
+				DC_LOG_DSC("DSC has been disabled for tg instance %d:", pipe_ctx->stream_res.tg->inst);
+				return;
+			}
+		}
 
 		/* Enable DSC hw block */
 		dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt;
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
index cd0adf72b223..a0b05b9ef660 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
@@ -1181,6 +1181,7 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	unsigned int odm_combine_factor = 0;
 	bool two_pix_per_container = false;
+	struct dce_hwseq *hws = stream->ctx->dc->hwseq;
 
 	two_pix_per_container = pipe_ctx->stream_res.tg->funcs->is_two_pixels_per_container(&stream->timing);
 	odm_combine_factor = get_odm_config(pipe_ctx, NULL);
@@ -1201,7 +1202,8 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign
 		} else {
 			*k1_div = PIXEL_RATE_DIV_BY_1;
 			*k2_div = PIXEL_RATE_DIV_BY_4;
-			if ((odm_combine_factor == 2) || dcn32_is_dp_dig_pixel_rate_div_policy(pipe_ctx))
+			if ((odm_combine_factor == 2) || (hws->funcs.is_dp_dig_pixel_rate_div_policy &&
+				hws->funcs.is_dp_dig_pixel_rate_div_policy(pipe_ctx)))
 				*k2_div = PIXEL_RATE_DIV_BY_2;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
index 922b8d71cf1a..c814d957305a 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
@@ -241,11 +241,6 @@ void dcn35_init_hw(struct dc *dc)
 			dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub,
 					!dc->res_pool->hubbub->ctx->dc->debug.disable_stutter);
 	}
-	if (res_pool->dccg->funcs->dccg_root_gate_disable_control) {
-		for (i = 0; i < res_pool->pipe_count; i++)
-			res_pool->dccg->funcs->dccg_root_gate_disable_control(res_pool->dccg, i, 0);
-	}
-
 	for (i = 0; i < res_pool->audio_count; i++) {
 		struct audio *audio = res_pool->audios[i];
 
@@ -901,12 +896,18 @@ void dcn35_init_pipes(struct dc *dc, struct dc_state *context)
 void dcn35_enable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx,
 			       struct dc_state *context)
 {
+	struct dpp *dpp = pipe_ctx->plane_res.dpp;
+	struct dccg *dccg = dc->res_pool->dccg;
+
+
 	/* enable DCFCLK current DCHUB */
 	pipe_ctx->plane_res.hubp->funcs->hubp_clk_cntl(pipe_ctx->plane_res.hubp, true);
 
 	/* initialize HUBP on power up */
 	pipe_ctx->plane_res.hubp->funcs->hubp_init(pipe_ctx->plane_res.hubp);
-
+	/*make sure DPPCLK is on*/
+	dccg->funcs->dccg_root_gate_disable_control(dccg, dpp->inst, true);
+	dpp->funcs->dpp_dppclk_control(dpp, false, true);
 	/* make sure OPP_PIPE_CLOCK_EN = 1 */
 	pipe_ctx->stream_res.opp->funcs->opp_pipe_clock_control(
 			pipe_ctx->stream_res.opp,
@@ -923,6 +924,7 @@ void dcn35_enable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx,
 		// Program system aperture settings
 		pipe_ctx->plane_res.hubp->funcs->hubp_set_vm_system_aperture_settings(pipe_ctx->plane_res.hubp, &apt);
 	}
+	//DC_LOG_DEBUG("%s: dpp_inst(%d) =\n", __func__, dpp->inst);
 
 	if (!pipe_ctx->top_pipe
 		&& pipe_ctx->plane_state
@@ -938,6 +940,8 @@ void dcn35_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx)
 {
 	struct hubp *hubp = pipe_ctx->plane_res.hubp;
 	struct dpp *dpp = pipe_ctx->plane_res.dpp;
+	struct dccg *dccg = dc->res_pool->dccg;
+
 
 	dc->hwss.wait_for_mpcc_disconnect(dc, dc->res_pool, pipe_ctx);
 
@@ -955,7 +959,8 @@ void dcn35_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	hubp->funcs->hubp_clk_cntl(hubp, false);
 
 	dpp->funcs->dpp_dppclk_control(dpp, false, false);
-/*to do, need to support both case*/
+	dccg->funcs->dccg_root_gate_disable_control(dccg, dpp->inst, false);
+
 	hubp->power_gated = true;
 
 	hubp->funcs->hubp_reset(hubp);
@@ -967,6 +972,8 @@ void dcn35_plane_atomic_disable(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	pipe_ctx->top_pipe = NULL;
 	pipe_ctx->bottom_pipe = NULL;
 	pipe_ctx->plane_state = NULL;
+	//DC_LOG_DEBUG("%s: dpp_inst(%d)=\n", __func__, dpp->inst);
+
 }
 
 void dcn35_disable_plane(struct dc *dc, struct dc_state *state, struct pipe_ctx *pipe_ctx)
@@ -1543,7 +1550,7 @@ static bool should_avoid_empty_tu(struct pipe_ctx *pipe_ctx)
 	struct dc_link_settings *link_settings = &pipe_ctx->link_config.dp_link_settings;
 	const struct dc *dc = pipe_ctx->stream->link->dc;
 
-	if (pipe_ctx->stream->link->ep_type != DISPLAY_ENDPOINT_USB4_DPIA)
+	if (pipe_ctx->link_config.dp_tunnel_settings.should_enable_dp_tunneling == false)
 		return false;
 
 	// Not necessary for MST configurations
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
index 6a82a865209c..a3ccf805bd16 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
@@ -168,6 +168,8 @@ static const struct hwseq_private_funcs dcn35_private_funcs = {
 	.dsc_pg_control = dcn35_dsc_pg_control,
 	.dsc_pg_status = dcn32_dsc_pg_status,
 	.enable_plane = dcn35_enable_plane,
+	.wait_for_pipe_update_if_needed = dcn10_wait_for_pipe_update_if_needed,
+	.set_wait_for_update_needed_for_pipe = dcn10_set_wait_for_update_needed_for_pipe,
 };
 
 void dcn35_hw_sequencer_construct(struct dc *dc)
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
index 902a96940a01..58f2be2a326b 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
@@ -158,10 +158,12 @@ static const struct hwseq_private_funcs dcn351_private_funcs = {
 	.set_mcm_luts = dcn32_set_mcm_luts,
 	.setup_hpo_hw_control = dcn35_setup_hpo_hw_control,
 	.calculate_dccg_k1_k2_values = dcn32_calculate_dccg_k1_k2_values,
-	.is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy,
+	.is_dp_dig_pixel_rate_div_policy = dcn35_is_dp_dig_pixel_rate_div_policy,
 	.dsc_pg_control = dcn35_dsc_pg_control,
 	.dsc_pg_status = dcn32_dsc_pg_status,
 	.enable_plane = dcn35_enable_plane,
+	.wait_for_pipe_update_if_needed = dcn10_wait_for_pipe_update_if_needed,
+	.set_wait_for_update_needed_for_pipe = dcn10_set_wait_for_update_needed_for_pipe,
 };
 
 void dcn351_hw_sequencer_construct(struct dc *dc)
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index 5489f3d431f6..c4177a9a662f 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -396,6 +396,249 @@ static void dcn401_get_mcm_lut_xable_from_pipe_ctx(struct dc *dc, struct pipe_ct
 	}
 }
 
+static void dcn401_set_mcm_location_post_blend(struct dc *dc, struct pipe_ctx *pipe_ctx, bool bPostBlend)
+{
+	struct mpc *mpc = dc->res_pool->mpc;
+	int mpcc_id = pipe_ctx->plane_res.hubp->inst;
+
+	if (!pipe_ctx->plane_state)
+		return;
+
+	mpc->funcs->set_movable_cm_location(mpc, MPCC_MOVABLE_CM_LOCATION_BEFORE, mpcc_id);
+	pipe_ctx->plane_state->mcm_location = (bPostBlend) ?
+											MPCC_MOVABLE_CM_LOCATION_AFTER :
+											MPCC_MOVABLE_CM_LOCATION_BEFORE;
+}
+
+static void dc_get_lut_mode(
+	enum dc_cm2_gpu_mem_layout layout,
+	enum hubp_3dlut_fl_mode *mode,
+	enum hubp_3dlut_fl_addressing_mode *addr_mode)
+{
+	switch (layout) {
+	case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_RGB:
+		*mode = hubp_3dlut_fl_mode_native_1;
+		*addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear;
+		break;
+	case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_BGR:
+		*mode = hubp_3dlut_fl_mode_native_2;
+		*addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear;
+		break;
+	case DC_CM2_GPU_MEM_LAYOUT_1D_PACKED_LINEAR:
+		*mode = hubp_3dlut_fl_mode_transform;
+		*addr_mode = hubp_3dlut_fl_addressing_mode_simple_linear;
+		break;
+	default:
+		*mode = hubp_3dlut_fl_mode_disable;
+		*addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear;
+		break;
+	}
+}
+
+static void dc_get_lut_format(
+	enum dc_cm2_gpu_mem_format dc_format,
+	enum hubp_3dlut_fl_format *format)
+{
+	switch (dc_format) {
+	case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12MSB:
+		*format = hubp_3dlut_fl_format_unorm_12msb_bitslice;
+		break;
+	case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12LSB:
+		*format = hubp_3dlut_fl_format_unorm_12lsb_bitslice;
+		break;
+	case DC_CM2_GPU_MEM_FORMAT_16161616_FLOAT_FP1_5_10:
+		*format = hubp_3dlut_fl_format_float_fp1_5_10;
+		break;
+	}
+}
+
+static void dc_get_lut_xbar(
+	enum dc_cm2_gpu_mem_pixel_component_order order,
+	enum hubp_3dlut_fl_crossbar_bit_slice *cr_r,
+	enum hubp_3dlut_fl_crossbar_bit_slice *y_g,
+	enum hubp_3dlut_fl_crossbar_bit_slice *cb_b)
+{
+	switch (order) {
+	case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA:
+		*cr_r = hubp_3dlut_fl_crossbar_bit_slice_32_47;
+		*y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31;
+		*cb_b =  hubp_3dlut_fl_crossbar_bit_slice_0_15;
+		break;
+	case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_BGRA:
+		*cr_r = hubp_3dlut_fl_crossbar_bit_slice_0_15;
+		*y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31;
+		*cb_b = hubp_3dlut_fl_crossbar_bit_slice_32_47;
+		break;
+	}
+}
+
+static void dc_get_lut_width(
+	enum dc_cm2_gpu_mem_size  size,
+	enum hubp_3dlut_fl_width *width)
+{
+	switch (size) {
+	case DC_CM2_GPU_MEM_SIZE_333333:
+		*width = hubp_3dlut_fl_width_33;
+		break;
+	case DC_CM2_GPU_MEM_SIZE_171717:
+		*width = hubp_3dlut_fl_width_17;
+		break;
+	case DC_CM2_GPU_MEM_SIZE_TRANSFORMED:
+		*width = hubp_3dlut_fl_width_transformed;
+		break;
+	}
+}
+static bool dc_is_rmcm_3dlut_supported(struct hubp *hubp, struct mpc *mpc)
+{
+	if (mpc->funcs->rmcm.update_3dlut_fast_load_select &&
+		mpc->funcs->rmcm.program_lut_read_write_control &&
+		hubp->funcs->hubp_program_3dlut_fl_addr &&
+		mpc->funcs->rmcm.program_bit_depth &&
+		hubp->funcs->hubp_program_3dlut_fl_mode &&
+		hubp->funcs->hubp_program_3dlut_fl_addressing_mode &&
+		hubp->funcs->hubp_program_3dlut_fl_format &&
+		hubp->funcs->hubp_update_3dlut_fl_bias_scale &&
+		mpc->funcs->rmcm.program_bias_scale &&
+		hubp->funcs->hubp_program_3dlut_fl_crossbar &&
+		hubp->funcs->hubp_program_3dlut_fl_width &&
+		mpc->funcs->rmcm.update_3dlut_fast_load_select &&
+		mpc->funcs->rmcm.populate_lut &&
+		mpc->funcs->rmcm.program_lut_mode &&
+		hubp->funcs->hubp_enable_3dlut_fl &&
+		mpc->funcs->rmcm.enable_3dlut_fl)
+		return true;
+
+	return false;
+}
+
+bool dcn401_program_rmcm_luts(
+	struct hubp *hubp,
+	struct pipe_ctx *pipe_ctx,
+	enum dc_cm2_transfer_func_source lut3d_src,
+	struct dc_cm2_func_luts *mcm_luts,
+	struct mpc *mpc,
+	bool lut_bank_a,
+	int mpcc_id)
+{
+	struct dpp *dpp_base = pipe_ctx->plane_res.dpp;
+	union mcm_lut_params m_lut_params;
+	enum MCM_LUT_XABLE shaper_xable, lut3d_xable = MCM_LUT_DISABLE, lut1d_xable;
+	enum hubp_3dlut_fl_mode mode;
+	enum hubp_3dlut_fl_addressing_mode addr_mode;
+	enum hubp_3dlut_fl_format format = 0;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g = 0;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b = 0;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r = 0;
+	enum hubp_3dlut_fl_width width = 0;
+	struct dc *dc = hubp->ctx->dc;
+
+	bool bypass_rmcm_3dlut  = false;
+	bool bypass_rmcm_shaper = false;
+
+	dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable);
+
+	/* 3DLUT */
+	switch (lut3d_src) {
+	case DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM:
+		memset(&m_lut_params, 0, sizeof(m_lut_params));
+		// Don't know what to do in this case.
+		//case DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM:
+		break;
+	case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM:
+		dc_get_lut_width(mcm_luts->lut3d_data.gpu_mem_params.size, &width);
+		if (!dc_is_rmcm_3dlut_supported(hubp, mpc) ||
+			!mpc->funcs->rmcm.is_config_supported(width))
+			return false;
+
+		//0. disable fl on mpc
+		mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, 0xF);
+
+		//1. power down the block
+		mpc->funcs->rmcm.power_on_shaper_3dlut(mpc, mpcc_id, false);
+
+		//2. program RMCM
+		//2a. 3dlut reg programming
+		mpc->funcs->rmcm.program_lut_read_write_control(mpc, MCM_LUT_3DLUT, lut_bank_a,
+				(!bypass_rmcm_3dlut) && lut3d_xable != MCM_LUT_DISABLE, mpcc_id);
+
+		hubp->funcs->hubp_program_3dlut_fl_addr(hubp,
+				mcm_luts->lut3d_data.gpu_mem_params.addr);
+
+		mpc->funcs->rmcm.program_bit_depth(mpc,
+				mcm_luts->lut3d_data.gpu_mem_params.bit_depth, mpcc_id);
+
+		// setting native or transformed mode,
+		dc_get_lut_mode(mcm_luts->lut3d_data.gpu_mem_params.layout, &mode, &addr_mode);
+
+		//these program the mcm 3dlut
+		hubp->funcs->hubp_program_3dlut_fl_mode(hubp, mode);
+
+		hubp->funcs->hubp_program_3dlut_fl_addressing_mode(hubp, addr_mode);
+
+		//seems to be only for the MCM
+		dc_get_lut_format(mcm_luts->lut3d_data.gpu_mem_params.format_params.format, &format);
+		hubp->funcs->hubp_program_3dlut_fl_format(hubp, format);
+
+		mpc->funcs->rmcm.program_bias_scale(mpc,
+			mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.bias,
+			mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.scale,
+			mpcc_id);
+		hubp->funcs->hubp_update_3dlut_fl_bias_scale(hubp,
+					mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.bias,
+					mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.scale);
+
+		dc_get_lut_xbar(
+			mcm_luts->lut3d_data.gpu_mem_params.component_order,
+			&crossbar_bit_slice_cr_r,
+			&crossbar_bit_slice_y_g,
+			&crossbar_bit_slice_cb_b);
+
+		hubp->funcs->hubp_program_3dlut_fl_crossbar(hubp,
+			crossbar_bit_slice_cr_r,
+			crossbar_bit_slice_y_g,
+			crossbar_bit_slice_cb_b);
+
+		mpc->funcs->rmcm.program_3dlut_size(mpc, width, mpcc_id);
+
+		mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, hubp->inst);
+
+		//2b. shaper reg programming
+		memset(&m_lut_params, 0, sizeof(m_lut_params));
+
+		if (mcm_luts->shaper->type == TF_TYPE_HWPWL) {
+			m_lut_params.pwl = &mcm_luts->shaper->pwl;
+		} else if (mcm_luts->shaper->type == TF_TYPE_DISTRIBUTED_POINTS) {
+			ASSERT(false);
+			cm_helper_translate_curve_to_hw_format(
+					dc->ctx,
+					mcm_luts->shaper,
+					&dpp_base->regamma_params, true);
+			m_lut_params.pwl = &dpp_base->regamma_params;
+		}
+		if (m_lut_params.pwl) {
+			mpc->funcs->rmcm.populate_lut(mpc, m_lut_params, lut_bank_a, mpcc_id);
+			mpc->funcs->rmcm.program_lut_mode(mpc, !bypass_rmcm_shaper, lut_bank_a, mpcc_id);
+		} else {
+			//RMCM 3dlut won't work without its shaper
+			return false;
+		}
+
+		//3. Select the hubp connected to this RMCM
+		hubp->funcs->hubp_enable_3dlut_fl(hubp, true);
+		mpc->funcs->rmcm.enable_3dlut_fl(mpc, true, mpcc_id);
+
+		//4. power on the block
+		if (m_lut_params.pwl)
+			mpc->funcs->rmcm.power_on_shaper_3dlut(mpc, mpcc_id, true);
+
+		break;
+	default:
+		return false;
+	}
+
+	return true;
+}
+
 void dcn401_populate_mcm_luts(struct dc *dc,
 		struct pipe_ctx *pipe_ctx,
 		struct dc_cm2_func_luts mcm_luts,
@@ -407,21 +650,39 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 	struct mpc *mpc = dc->res_pool->mpc;
 	union mcm_lut_params m_lut_params;
 	enum dc_cm2_transfer_func_source lut3d_src = mcm_luts.lut3d_data.lut3d_src;
-	enum hubp_3dlut_fl_format format;
+	enum hubp_3dlut_fl_format format = 0;
 	enum hubp_3dlut_fl_mode mode;
-	enum hubp_3dlut_fl_width width;
+	enum hubp_3dlut_fl_width width = 0;
 	enum hubp_3dlut_fl_addressing_mode addr_mode;
-	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g;
-	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b;
-	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g = 0;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b = 0;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r = 0;
 	enum MCM_LUT_XABLE shaper_xable = MCM_LUT_DISABLE;
 	enum MCM_LUT_XABLE lut3d_xable = MCM_LUT_DISABLE;
 	enum MCM_LUT_XABLE lut1d_xable = MCM_LUT_DISABLE;
-	bool is_17x17x17 = true;
 	bool rval;
 
 	dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable);
 
+	//MCM - setting its location (Before/After) blender
+	//set to post blend (true)
+	dcn401_set_mcm_location_post_blend(
+		dc,
+		pipe_ctx,
+		mcm_luts.lut3d_data.mpc_mcm_post_blend);
+
+	//RMCM - 3dLUT+Shaper
+	if (mcm_luts.lut3d_data.rmcm_3dlut_enable) {
+		dcn401_program_rmcm_luts(
+			hubp,
+			pipe_ctx,
+			lut3d_src,
+			&mcm_luts,
+			mpc,
+			lut_bank_a,
+			mpcc_id);
+	}
+
 	/* 1D LUT */
 	if (mcm_luts.lut1d_func) {
 		memset(&m_lut_params, 0, sizeof(m_lut_params));
@@ -442,7 +703,7 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 	}
 
 	/* Shaper */
-	if (mcm_luts.shaper) {
+	if (mcm_luts.shaper && mcm_luts.lut3d_data.mpc_3dlut_enable) {
 		memset(&m_lut_params, 0, sizeof(m_lut_params));
 		if (mcm_luts.shaper->type == TF_TYPE_HWPWL)
 			m_lut_params.pwl = &mcm_luts.shaper->pwl;
@@ -454,11 +715,11 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 			m_lut_params.pwl = rval ? &dpp_base->regamma_params : NULL;
 		}
 		if (m_lut_params.pwl) {
-			if (mpc->funcs->populate_lut)
-				mpc->funcs->populate_lut(mpc, MCM_LUT_SHAPER, m_lut_params, lut_bank_a, mpcc_id);
+			if (mpc->funcs->mcm.populate_lut)
+				mpc->funcs->mcm.populate_lut(mpc, m_lut_params, lut_bank_a, mpcc_id);
+			if (mpc->funcs->program_lut_mode)
+				mpc->funcs->program_lut_mode(mpc, MCM_LUT_SHAPER, MCM_LUT_ENABLE, lut_bank_a, mpcc_id);
 		}
-		if (mpc->funcs->program_lut_mode)
-			mpc->funcs->program_lut_mode(mpc, MCM_LUT_SHAPER, shaper_xable, lut_bank_a, mpcc_id);
 	}
 
 	/* 3DLUT */
@@ -467,6 +728,7 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 		memset(&m_lut_params, 0, sizeof(m_lut_params));
 		if (hubp->funcs->hubp_enable_3dlut_fl)
 			hubp->funcs->hubp_enable_3dlut_fl(hubp, false);
+
 		if (mcm_luts.lut3d_data.lut3d_func && mcm_luts.lut3d_data.lut3d_func->state.bits.initialized) {
 			m_lut_params.lut3d = &mcm_luts.lut3d_data.lut3d_func->lut_3d;
 			if (mpc->funcs->populate_lut)
@@ -476,16 +738,35 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 						mpcc_id);
 		}
 		break;
-	case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM:
+		case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM:
+		switch (mcm_luts.lut3d_data.gpu_mem_params.size) {
+		case DC_CM2_GPU_MEM_SIZE_333333:
+			width = hubp_3dlut_fl_width_33;
+			break;
+		case DC_CM2_GPU_MEM_SIZE_171717:
+			width = hubp_3dlut_fl_width_17;
+			break;
+		case DC_CM2_GPU_MEM_SIZE_TRANSFORMED:
+			width = hubp_3dlut_fl_width_transformed;
+			break;
+		}
+
+		//check for support
+		if (mpc->funcs->mcm.is_config_supported &&
+			!mpc->funcs->mcm.is_config_supported(width))
+			break;
 
 		if (mpc->funcs->program_lut_read_write_control)
 			mpc->funcs->program_lut_read_write_control(mpc, MCM_LUT_3DLUT, lut_bank_a, mpcc_id);
 		if (mpc->funcs->program_lut_mode)
 			mpc->funcs->program_lut_mode(mpc, MCM_LUT_3DLUT, lut3d_xable, lut_bank_a, mpcc_id);
-		if (mpc->funcs->program_3dlut_size)
-			mpc->funcs->program_3dlut_size(mpc, is_17x17x17, mpcc_id);
+
 		if (hubp->funcs->hubp_program_3dlut_fl_addr)
 			hubp->funcs->hubp_program_3dlut_fl_addr(hubp, mcm_luts.lut3d_data.gpu_mem_params.addr);
+
+		if (mpc->funcs->mcm.program_bit_depth)
+			mpc->funcs->mcm.program_bit_depth(mpc, mcm_luts.lut3d_data.gpu_mem_params.bit_depth, mpcc_id);
+
 		switch (mcm_luts.lut3d_data.gpu_mem_params.layout) {
 		case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_RGB:
 			mode = hubp_3dlut_fl_mode_native_1;
@@ -512,7 +793,6 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 
 		switch (mcm_luts.lut3d_data.gpu_mem_params.format_params.format) {
 		case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12MSB:
-		default:
 			format = hubp_3dlut_fl_format_unorm_12msb_bitslice;
 			break;
 		case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12LSB:
@@ -524,37 +804,37 @@ void dcn401_populate_mcm_luts(struct dc *dc,
 		}
 		if (hubp->funcs->hubp_program_3dlut_fl_format)
 			hubp->funcs->hubp_program_3dlut_fl_format(hubp, format);
-		if (hubp->funcs->hubp_update_3dlut_fl_bias_scale)
+		if (hubp->funcs->hubp_update_3dlut_fl_bias_scale &&
+				mpc->funcs->mcm.program_bias_scale) {
+			mpc->funcs->mcm.program_bias_scale(mpc,
+				mcm_luts.lut3d_data.gpu_mem_params.format_params.float_params.bias,
+				mcm_luts.lut3d_data.gpu_mem_params.format_params.float_params.scale,
+				mpcc_id);
 			hubp->funcs->hubp_update_3dlut_fl_bias_scale(hubp,
-					mcm_luts.lut3d_data.gpu_mem_params.format_params.float_params.bias,
-					mcm_luts.lut3d_data.gpu_mem_params.format_params.float_params.scale);
-
-		switch (mcm_luts.lut3d_data.gpu_mem_params.component_order) {
-		case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA:
-		default:
-			crossbar_bit_slice_cr_r = hubp_3dlut_fl_crossbar_bit_slice_0_15;
-			crossbar_bit_slice_y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31;
-			crossbar_bit_slice_cb_b = hubp_3dlut_fl_crossbar_bit_slice_32_47;
-			break;
+						mcm_luts.lut3d_data.gpu_mem_params.format_params.float_params.bias,
+						mcm_luts.lut3d_data.gpu_mem_params.format_params.float_params.scale);
 		}
 
+		//navi 4x has a bug and r and blue are swapped and need to be worked around here in
+		//TODO: need to make a method for get_xbar per asic OR do the workaround in program_crossbar for 4x
+		dc_get_lut_xbar(
+			mcm_luts.lut3d_data.gpu_mem_params.component_order,
+			&crossbar_bit_slice_cr_r,
+			&crossbar_bit_slice_y_g,
+			&crossbar_bit_slice_cb_b);
+
 		if (hubp->funcs->hubp_program_3dlut_fl_crossbar)
 			hubp->funcs->hubp_program_3dlut_fl_crossbar(hubp,
+					crossbar_bit_slice_cr_r,
 					crossbar_bit_slice_y_g,
-					crossbar_bit_slice_cb_b,
-					crossbar_bit_slice_cr_r);
+					crossbar_bit_slice_cb_b);
+
+		if (mpc->funcs->mcm.program_lut_read_write_control)
+			mpc->funcs->mcm.program_lut_read_write_control(mpc, MCM_LUT_3DLUT, lut_bank_a, true, mpcc_id);
+
+		if (mpc->funcs->mcm.program_3dlut_size)
+			mpc->funcs->mcm.program_3dlut_size(mpc, width, mpcc_id);
 
-		switch (mcm_luts.lut3d_data.gpu_mem_params.size) {
-		case DC_CM2_GPU_MEM_SIZE_171717:
-		default:
-			width = hubp_3dlut_fl_width_17;
-			break;
-		case DC_CM2_GPU_MEM_SIZE_TRANSFORMED:
-			width = hubp_3dlut_fl_width_transformed;
-			break;
-		}
-		if (hubp->funcs->hubp_program_3dlut_fl_width)
-			hubp->funcs->hubp_program_3dlut_fl_width(hubp, width);
 		if (mpc->funcs->update_3dlut_fast_load_select)
 			mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, hubp->inst);
 
@@ -1980,9 +2260,9 @@ void dcn401_program_pipe(
 				dc->res_pool->hubbub, pipe_ctx->plane_res.hubp->inst, pipe_ctx->hubp_regs.det_size);
 	}
 
-	if (pipe_ctx->update_flags.raw ||
-		(pipe_ctx->plane_state && pipe_ctx->plane_state->update_flags.raw) ||
-		pipe_ctx->stream->update_flags.raw)
+	if (pipe_ctx->plane_state && (pipe_ctx->update_flags.raw ||
+	    pipe_ctx->plane_state->update_flags.raw ||
+	    pipe_ctx->stream->update_flags.raw))
 		dc->hwss.update_dchubp_dpp(dc, pipe_ctx, context);
 
 	if (pipe_ctx->plane_state && (pipe_ctx->update_flags.bits.enable ||
@@ -2081,7 +2361,7 @@ void dcn401_program_front_end_for_ctx(
 		for (i = 0; i < dc->res_pool->pipe_count; i++) {
 			pipe = &context->res_ctx.pipe_ctx[i];
 
-			if (!pipe->top_pipe && !pipe->prev_odm_pipe && pipe->plane_state) {
+			if (pipe->plane_state) {
 				if (pipe->plane_state->triplebuffer_flips)
 					BREAK_TO_DEBUGGER();
 
@@ -2371,7 +2651,7 @@ bool dcn401_update_bandwidth(
 	struct dce_hwseq *hws = dc->hwseq;
 
 	/* recalculate DML parameters */
-	if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false))
+	if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK)
 		return false;
 
 	/* apply updated bandwidth parameters */
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h
index 781cf0efccc6..ce65b4f6c672 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h
@@ -109,4 +109,12 @@ void dcn401_detect_pipe_changes(
 void dcn401_plane_atomic_power_down(struct dc *dc,
 		struct dpp *dpp,
 		struct hubp *hubp);
+bool dcn401_program_rmcm_luts(
+	struct hubp *hubp,
+	struct pipe_ctx *pipe_ctx,
+	enum dc_cm2_transfer_func_source lut3d_src,
+	struct dc_cm2_func_luts *mcm_luts,
+	struct mpc *mpc,
+	bool lut_bank_a,
+	int mpcc_id);
 #endif /* __DC_HWSS_DCN401_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h
index c8b5ed834579..3a0795045bc6 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h
@@ -195,6 +195,8 @@ enum block_sequence_func {
 	DMUB_SUBVP_SAVE_SURF_ADDR,
 	HUBP_WAIT_FOR_DCC_META_PROP,
 	DMUB_FAMS2_GLOBAL_CONTROL_LOCK_FAST,
+	/* This must be the last value in this enum, add new ones above */
+	HWSS_BLOCK_SEQUENCE_FUNC_COUNT
 };
 
 struct block_sequence {
@@ -202,6 +204,8 @@ struct block_sequence {
 	enum block_sequence_func func;
 };
 
+#define MAX_HWSS_BLOCK_SEQUENCE_SIZE (HWSS_BLOCK_SEQUENCE_FUNC_COUNT * MAX_PIPES)
+
 struct hw_sequencer_funcs {
 	void (*hardware_release)(struct dc *dc);
 	/* Embedded Display Related */
@@ -534,13 +538,13 @@ void set_drr_and_clear_adjust_pending(
 		struct drr_params *params);
 
 void hwss_execute_sequence(struct dc *dc,
-		struct block_sequence block_sequence[],
+		struct block_sequence block_sequence[MAX_HWSS_BLOCK_SEQUENCE_SIZE],
 		int num_steps);
 
 void hwss_build_fast_sequence(struct dc *dc,
 		struct dc_dmub_cmd *dc_dmub_cmd,
 		unsigned int dmub_cmd_count,
-		struct block_sequence block_sequence[],
+		struct block_sequence block_sequence[MAX_HWSS_BLOCK_SEQUENCE_SIZE],
 		unsigned int *num_steps,
 		struct pipe_ctx *pipe_ctx,
 		struct dc_stream_status *stream_status,
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h
index 22a5d4a03c98..09bc65c2fa23 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h
@@ -183,6 +183,8 @@ struct hwseq_private_funcs {
 			struct dc_cm2_func_luts mcm_luts,
 			bool lut_bank_a);
 	void (*perform_3dlut_wa_unlock)(struct pipe_ctx *pipe_ctx);
+	void (*wait_for_pipe_update_if_needed)(struct dc *dc, struct pipe_ctx *pipe_ctx, bool is_surface_update_only);
+	void (*set_wait_for_update_needed_for_pipe)(struct dc *dc, struct pipe_ctx *pipe_ctx);
 };
 
 struct dce_hwseq {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h
index b5afd8c3103d..f3696143590c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h
@@ -26,6 +26,8 @@
 #ifndef _CORE_STATUS_H_
 #define _CORE_STATUS_H_
 
+#include "dc_hw_types.h"
+
 enum dc_status {
 	DC_OK = 1,
 
@@ -56,6 +58,7 @@ enum dc_status {
 	DC_NO_LINK_ENC_RESOURCE = 26,
 	DC_FAIL_DP_PAYLOAD_ALLOCATION = 27,
 	DC_FAIL_DP_LINK_BANDWIDTH = 28,
+	DC_FAIL_HW_CURSOR_SUPPORT = 29,
 	DC_ERROR_UNEXPECTED = -1
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index d0021f25f3d8..0cf349cafb3e 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -65,6 +65,7 @@ struct resource_pool;
 struct dc_state;
 struct resource_context;
 struct clk_bw_params;
+struct dc_mcache_params;
 
 struct resource_funcs {
 	enum engine_id (*get_preferred_eng_id_dpia)(unsigned int dpia_index);
@@ -78,8 +79,7 @@ struct resource_funcs {
 	/* Create a minimal link encoder object with no dc_link object
 	 * associated with it. */
 	struct link_encoder *(*link_enc_create_minimal)(struct dc_context *ctx, enum engine_id eng_id);
-
-	bool (*validate_bandwidth)(
+	enum dc_status (*validate_bandwidth)(
 					struct dc *dc,
 					struct dc_state *context,
 					bool fast_validate);
@@ -218,6 +218,11 @@ struct resource_funcs {
 	int (*get_power_profile)(const struct dc_state *context);
 	unsigned int (*get_det_buffer_size)(const struct dc_state *context);
 	unsigned int (*get_vstartup_for_pipe)(struct pipe_ctx *pipe_ctx);
+	unsigned int (*get_max_hw_cursor_size)(const struct dc *dc,
+			struct dc_state *state,
+			const struct dc_stream_state *stream);
+	bool (*program_mcache_pipe_config)(struct dc_state *context,
+		const struct dc_mcache_params *mcache_params);
 };
 
 struct audio_support{
@@ -382,7 +387,9 @@ struct link_resource {
 
 struct link_config {
 	struct dc_link_settings dp_link_settings;
+	struct dc_tunnel_settings dp_tunnel_settings;
 };
+
 union pipe_update_flags {
 	struct {
 		uint32_t enable : 1;
@@ -480,6 +487,10 @@ struct pipe_ctx {
 	struct pixel_rate_divider pixel_rate_divider;
 	/* pixels borrowed from hblank to hactive */
 	uint8_t hblank_borrow;
+	/* next vupdate */
+	uint32_t next_vupdate;
+	uint32_t wait_frame_count;
+	bool wait_is_required;
 };
 
 /* Data used for dynamic link encoder assignment.
@@ -507,7 +518,7 @@ struct resource_context {
 	unsigned int hpo_dp_link_enc_to_link_idx[MAX_HPO_DP2_LINK_ENCODERS];
 	int hpo_dp_link_enc_ref_cnts[MAX_HPO_DP2_LINK_ENCODERS];
 	bool is_mpc_3dlut_acquired[MAX_PIPES];
-	/* solely used for build scalar data in dml2 */
+	/* used to build scalar data in dml2 and for edp backlight programming */
 	struct pipe_ctx temp_pipe;
 };
 
@@ -630,7 +641,7 @@ struct dc_state {
 	 */
 	struct bw_context bw_ctx;
 
-	struct block_sequence block_sequence[100];
+	struct block_sequence block_sequence[MAX_HWSS_BLOCK_SEQUENCE_SIZE];
 	unsigned int block_sequence_steps;
 	struct dc_dmub_cmd dc_dmub_cmd[10];
 	unsigned int dmub_cmd_count;
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
index 221645c023b5..bac8febad69a 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h
@@ -199,6 +199,7 @@ enum dentist_divider_range {
 	CLK_SR_DCN35(CLK1_CLK4_ALLOW_DS), \
 	CLK_SR_DCN35(CLK1_CLK5_ALLOW_DS), \
 	CLK_SR_DCN35(CLK5_spll_field_8), \
+	CLK_SR_DCN35(CLK6_spll_field_8), \
 	SR(DENTIST_DISPCLK_CNTL), \
 
 #define CLK_COMMON_MASK_SH_LIST_DCN32(mask_sh) \
@@ -307,7 +308,7 @@ struct clk_mgr_registers {
 	uint32_t CLK1_CLK4_ALLOW_DS;
 	uint32_t CLK1_CLK5_ALLOW_DS;
 	uint32_t CLK5_spll_field_8;
-
+	uint32_t CLK6_spll_field_8;
 };
 
 struct clk_mgr_shift {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
index 3a89cc0cffc1..6e303b81bfb0 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
@@ -967,23 +967,6 @@ struct mpc_funcs {
 	*/
 
 	void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx);
-	/**
-	* @get_3dlut_fast_load_status:
-	*
-	* Get 3D LUT fast load status and reference them with done, soft_underflow and hard_underflow pointers.
-	*
-	* Parameters:
-	* - [in/out] mpc - MPC context.
-	* - [in] mpcc_id
-	* - [in/out] done
-	* - [in/out] soft_underflow
-	* - [in/out] hard_underflow
-	*
-	* Return:
-	*
-	* void
-	*/
-	void (*get_3dlut_fast_load_status)(struct mpc *mpc, int mpcc_id, uint32_t *done, uint32_t *soft_underflow, uint32_t *hard_underflow);
 
 	/**
 	* @populate_lut:
@@ -1054,6 +1037,35 @@ struct mpc_funcs {
 	* void
 	*/
 	void (*program_3dlut_size)(struct mpc *mpc, bool is_17x17x17, int mpcc_id);
+
+	struct {
+		void (*program_3dlut_size)(struct mpc *mpc, uint32_t width, int mpcc_id);
+		void (*program_bias_scale)(struct mpc *mpc, uint16_t bias, uint16_t scale, int mpcc_id);
+		void (*program_bit_depth)(struct mpc *mpc, uint16_t bit_depth, int mpcc_id);
+		bool (*is_config_supported)(uint32_t width);
+		void (*program_lut_read_write_control)(struct mpc *mpc, const enum MCM_LUT_ID id,
+			bool lut_bank_a, bool enabled, int mpcc_id);
+
+		void (*populate_lut)(struct mpc *mpc, const union mcm_lut_params params,
+			bool lut_bank_a, int mpcc_id);
+	} mcm;
+
+	struct {
+		void (*enable_3dlut_fl)(struct mpc *mpc, bool enable, int mpcc_id);
+		void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx);
+		void (*program_lut_read_write_control)(struct mpc *mpc, const enum MCM_LUT_ID id,
+			bool lut_bank_a, bool enabled, int mpcc_id);
+		void (*program_lut_mode)(struct mpc *mpc, const enum MCM_LUT_XABLE xable,
+			bool lut_bank_a, int mpcc_id);
+		void (*program_3dlut_size)(struct mpc *mpc, uint32_t width, int mpcc_id);
+		void (*program_bias_scale)(struct mpc *mpc, uint16_t bias, uint16_t scale, int mpcc_id);
+		void (*program_bit_depth)(struct mpc *mpc, uint16_t bit_depth, int mpcc_id);
+		bool (*is_config_supported)(uint32_t width);
+
+		void (*power_on_shaper_3dlut)(struct mpc *mpc, uint32_t mpcc_id, bool power_on);
+		void (*populate_lut)(struct mpc *mpc, const union mcm_lut_params params,
+			bool lut_bank_a, int mpcc_id);
+	} rmcm;
 };
 
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h
index 7f371cbb35cd..0d5a8358a778 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h
@@ -68,6 +68,7 @@ struct optc {
 	int pstate_keepout;
 	struct dc_crtc_timing orginal_patched_timing;
 	enum signal_type signal;
+	uint32_t max_frame_count;
 };
 
 void optc1_read_otg_state(struct timing_generator *optc, struct dcn_otg_state *s);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h
index 2948a696ee12..7d16351bba99 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/link.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/link.h
@@ -207,6 +207,9 @@ struct link_service {
 	bool (*dp_decide_link_settings)(
 		struct dc_stream_state *stream,
 		struct dc_link_settings *link_setting);
+	void (*dp_decide_tunnel_settings)(
+		struct dc_stream_state *stream,
+		struct dc_tunnel_settings *dp_tunnel_setting);
 	enum dp_link_encoding (*mst_decide_link_encoding_format)(
 			const struct dc_link *link);
 	bool (*edp_decide_link_settings)(struct dc_link *link,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h
index a402df225a76..26cb1459b743 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h
@@ -508,6 +508,10 @@ uint32_t generic_indirect_reg_update_ex(const struct dc_context *ctx,
 				initial_val, \
 				n, __VA_ARGS__)
 
+#define IX_REG_SET_SYNC(index, init_value, f1, v1)	\
+		IX_REG_SET_N_SYNC(index, 1, init_value, \
+				FN(reg, f1), v1)
+
 #define IX_REG_SET_2_SYNC(index, init_value, f1, v1, f2, v2)	\
 		IX_REG_SET_N_SYNC(index, 2, init_value, \
 				FN(reg, f1), v1,\
diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h
index 9458187b834d..a890f581f4e8 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/resource.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h
@@ -32,6 +32,7 @@
 
 #define MEMORY_TYPE_MULTIPLIER_CZ 4
 #define MEMORY_TYPE_HBM 2
+#define MAX_MCACHES 8
 
 
 #define IS_PIPE_SYNCD_VALID(pipe) ((((pipe)->pipe_idx_syncd) & 0x80)?1:0)
@@ -65,6 +66,13 @@ struct resource_straps {
 	uint32_t audio_stream_number;
 };
 
+struct dc_mcache_allocations {
+	int global_mcache_ids_plane0[MAX_MCACHES + 1];
+	int global_mcache_ids_plane1[MAX_MCACHES + 1];
+	int global_mcache_ids_mall_plane0[MAX_MCACHES + 1];
+	int global_mcache_ids_mall_plane1[MAX_MCACHES + 1];
+};
+
 struct resource_create_funcs {
 	void (*read_dce_straps)(
 			struct dc_context *ctx, struct resource_straps *straps);
@@ -628,8 +636,6 @@ enum dc_status update_dp_encoder_resources_for_test_harness(const struct dc *dc,
 		struct dc_state *context,
 		struct pipe_ctx *pipe_ctx);
 
-bool check_subvp_sw_cursor_fallback_req(const struct dc *dc, struct dc_stream_state *stream);
-
 /* Get hw programming parameters container from pipe context
  * @pipe_ctx: pipe context
  * @dscl_prog_data: struct to hold programmable hw reg values
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c b/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c
index 953f4a4dacad..33ce470e4c88 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce120/irq_service_dce120.c
@@ -37,36 +37,9 @@
 
 #include "ivsrcid/ivsrcid_vislands30.h"
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c b/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c
index 2c72074310c7..d777b85e70da 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce60/irq_service_dce60.c
@@ -46,36 +46,9 @@
 
 #include "dc_types.h"
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			DC_HPD1_INT_STATUS,
-			DC_HPD1_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		DC_HPD1_INT_CONTROL,
-		DC_HPD1_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd1_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
@@ -391,5 +364,3 @@ struct irq_service *dal_irq_service_dce60_create(
 	dce60_irq_construct(irq_service, init_data);
 	return irq_service;
 }
-
-
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c b/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c
index 49317934ef4f..3a9163acb49b 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dce80/irq_service_dce80.c
@@ -37,36 +37,9 @@
 
 #include "dc_types.h"
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			DC_HPD1_INT_STATUS,
-			DC_HPD1_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		DC_HPD1_INT_CONTROL,
-		DC_HPD1_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd1_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
@@ -303,5 +276,3 @@ struct irq_service *dal_irq_service_dce80_create(
 	dce80_irq_construct(irq_service, init_data);
 	return irq_service;
 }
-
-
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c
index 9ca28565a9d1..4ce9edd16344 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c
@@ -129,36 +129,9 @@ static enum dc_irq_source to_dal_irq_source_dcn10(struct irq_service *irq_servic
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c b/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c
index 916f0c974637..5847af0e66cb 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn20/irq_service_dcn20.c
@@ -130,36 +130,9 @@ static enum dc_irq_source to_dal_irq_source_dcn20(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c b/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c
index 1d61d475d36f..6417011d2246 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c
@@ -80,36 +80,9 @@ static enum dc_irq_source to_dal_irq_source_dcn201(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c
index 42cdfe6c3538..71d2f065140b 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c
@@ -132,36 +132,9 @@ static enum dc_irq_source to_dal_irq_source_dcn21(struct irq_service *irq_servic
 	return DC_IRQ_SOURCE_INVALID;
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c b/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c
index a443a8abb1ea..2a4080bdcf6b 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c
@@ -139,36 +139,9 @@ static enum dc_irq_source to_dal_irq_source_dcn30(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
@@ -447,4 +420,3 @@ struct irq_service *dal_irq_service_dcn30_create(
 	dcn30_irq_construct(irq_service, init_data);
 	return irq_service;
 }
-
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn302/irq_service_dcn302.c b/drivers/gpu/drm/amd/display/dc/irq/dcn302/irq_service_dcn302.c
index 8ffc7e2c681a..624f1ac309f8 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn302/irq_service_dcn302.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn302/irq_service_dcn302.c
@@ -126,26 +126,9 @@ static enum dc_irq_source to_dal_irq_source_dcn302(struct irq_service *irq_servi
 	}
 }
 
-static bool hpd_ack(struct irq_service *irq_service, const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status = get_reg_field_value(value, HPD0_DC_HPD_INT_STATUS, DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(value, current_status ? 0 : 1, HPD0_DC_HPD_INT_CONTROL, DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 		.set = NULL,
-		.ack = hpd_ack
+		.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn303/irq_service_dcn303.c b/drivers/gpu/drm/amd/display/dc/irq/dcn303/irq_service_dcn303.c
index 262bb8b74b15..137caffae916 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn303/irq_service_dcn303.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn303/irq_service_dcn303.c
@@ -77,26 +77,9 @@ static enum dc_irq_source to_dal_irq_source_dcn303(struct irq_service *irq_servi
 	}
 }
 
-static bool hpd_ack(struct irq_service *irq_service, const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status = get_reg_field_value(value, HPD0_DC_HPD_INT_STATUS, DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(value, current_status ? 0 : 1, HPD0_DC_HPD_INT_CONTROL, DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 		.set = NULL,
-		.ack = hpd_ack
+		.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c b/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c
index 53e78ae7eecf..921cb167d920 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c
@@ -128,36 +128,9 @@ static enum dc_irq_source to_dal_irq_source_dcn31(struct irq_service *irq_servic
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c b/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c
index e0563e880432..0118fd6e5db0 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c
@@ -130,36 +130,9 @@ static enum dc_irq_source to_dal_irq_source_dcn314(struct irq_service *irq_servi
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn315/irq_service_dcn315.c b/drivers/gpu/drm/amd/display/dc/irq/dcn315/irq_service_dcn315.c
index 2ef22299101a..adebfc888618 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn315/irq_service_dcn315.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn315/irq_service_dcn315.c
@@ -135,36 +135,9 @@ static enum dc_irq_source to_dal_irq_source_dcn315(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn32/irq_service_dcn32.c b/drivers/gpu/drm/amd/display/dc/irq/dcn32/irq_service_dcn32.c
index f0ac0aeeac51..e9e315c75d76 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn32/irq_service_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn32/irq_service_dcn32.c
@@ -129,36 +129,9 @@ static enum dc_irq_source to_dal_irq_source_dcn32(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs  = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
@@ -191,6 +164,16 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 	.ack = NULL
 };
 
+static struct irq_source_info_funcs vline1_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
+static struct irq_source_info_funcs vline2_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
 #undef BASE_INNER
 #define BASE_INNER(seg) DCN_BASE__INST0_SEG ## seg
 
@@ -259,6 +242,13 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 		.funcs = &pflip_irq_info_funcs\
 	}
 
+#define vblank_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\
+		IRQ_REG_ENTRY(OTG, reg_num,\
+			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_INT_EN,\
+			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_EVENT_CLEAR),\
+		.funcs = &vblank_irq_info_funcs\
+	}
 /* vupdate_no_lock_int_entry maps to DC_IRQ_SOURCE_VUPDATEx, to match semantic
  * of DCE's DC_IRQ_SOURCE_VUPDATEx.
  */
@@ -270,14 +260,6 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 		.funcs = &vupdate_no_lock_irq_info_funcs\
 	}
 
-#define vblank_int_entry(reg_num)\
-	[DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\
-		IRQ_REG_ENTRY(OTG, reg_num,\
-			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_INT_EN,\
-			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_EVENT_CLEAR),\
-		.funcs = &vblank_irq_info_funcs\
-}
-
 #define vline0_int_entry(reg_num)\
 	[DC_IRQ_SOURCE_DC1_VLINE0 + reg_num] = {\
 		IRQ_REG_ENTRY(OTG, reg_num,\
@@ -285,6 +267,20 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 			OTG_VERTICAL_INTERRUPT0_CONTROL, OTG_VERTICAL_INTERRUPT0_CLEAR),\
 		.funcs = &vline0_irq_info_funcs\
 	}
+#define vline1_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_DC1_VLINE1 + reg_num] = {\
+		IRQ_REG_ENTRY(OTG, reg_num,\
+			OTG_VERTICAL_INTERRUPT1_CONTROL, OTG_VERTICAL_INTERRUPT1_INT_ENABLE,\
+			OTG_VERTICAL_INTERRUPT1_CONTROL, OTG_VERTICAL_INTERRUPT1_CLEAR),\
+		.funcs = &vline1_irq_info_funcs\
+	}
+#define vline2_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_DC1_VLINE2 + reg_num] = {\
+		IRQ_REG_ENTRY(OTG, reg_num,\
+			OTG_VERTICAL_INTERRUPT2_CONTROL, OTG_VERTICAL_INTERRUPT2_INT_ENABLE,\
+			OTG_VERTICAL_INTERRUPT2_CONTROL, OTG_VERTICAL_INTERRUPT2_CLEAR),\
+		.funcs = &vline2_irq_info_funcs\
+	}
 #define dmub_outbox_int_entry()\
 	[DC_IRQ_SOURCE_DMCUB_OUTBOX] = {\
 		IRQ_REG_ENTRY_DMUB(\
@@ -387,21 +383,29 @@ irq_source_info_dcn32[DAL_IRQ_SOURCES_NUMBER] = {
 	dc_underflow_int_entry(6),
 	[DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(),
 	[DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(),
-	vupdate_no_lock_int_entry(0),
-	vupdate_no_lock_int_entry(1),
-	vupdate_no_lock_int_entry(2),
-	vupdate_no_lock_int_entry(3),
 	vblank_int_entry(0),
 	vblank_int_entry(1),
 	vblank_int_entry(2),
 	vblank_int_entry(3),
+	[DC_IRQ_SOURCE_DC5_VLINE1] = dummy_irq_entry(),
+	[DC_IRQ_SOURCE_DC6_VLINE1] = dummy_irq_entry(),
+	dmub_outbox_int_entry(),
+	vupdate_no_lock_int_entry(0),
+	vupdate_no_lock_int_entry(1),
+	vupdate_no_lock_int_entry(2),
+	vupdate_no_lock_int_entry(3),
 	vline0_int_entry(0),
 	vline0_int_entry(1),
 	vline0_int_entry(2),
 	vline0_int_entry(3),
-	[DC_IRQ_SOURCE_DC5_VLINE1] = dummy_irq_entry(),
-	[DC_IRQ_SOURCE_DC6_VLINE1] = dummy_irq_entry(),
-	dmub_outbox_int_entry(),
+	vline1_int_entry(0),
+	vline1_int_entry(1),
+	vline1_int_entry(2),
+	vline1_int_entry(3),
+	vline2_int_entry(0),
+	vline2_int_entry(1),
+	vline2_int_entry(2),
+	vline2_int_entry(3)
 };
 
 static const struct irq_service_funcs irq_service_funcs_dcn32 = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn35/irq_service_dcn35.c b/drivers/gpu/drm/amd/display/dc/irq/dcn35/irq_service_dcn35.c
index ea8c271171bc..79e5e8c137ca 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn35/irq_service_dcn35.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn35/irq_service_dcn35.c
@@ -127,36 +127,9 @@ static enum dc_irq_source to_dal_irq_source_dcn35(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn351/irq_service_dcn351.c b/drivers/gpu/drm/amd/display/dc/irq/dcn351/irq_service_dcn351.c
index 7ec8e0de2f01..163b8ee9ebf7 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn351/irq_service_dcn351.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn351/irq_service_dcn351.c
@@ -106,36 +106,9 @@ static enum dc_irq_source to_dal_irq_source_dcn351(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn36/irq_service_dcn36.c b/drivers/gpu/drm/amd/display/dc/irq/dcn36/irq_service_dcn36.c
index ea958628f8b8..f716ab0fd30e 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn36/irq_service_dcn36.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn36/irq_service_dcn36.c
@@ -105,36 +105,9 @@ static enum dc_irq_source to_dal_irq_source_dcn36(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn401/irq_service_dcn401.c b/drivers/gpu/drm/amd/display/dc/irq/dcn401/irq_service_dcn401.c
index b43c9524b0de..fd9bb1950c20 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/dcn401/irq_service_dcn401.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/dcn401/irq_service_dcn401.c
@@ -109,36 +109,9 @@ static enum dc_irq_source to_dal_irq_source_dcn401(
 	}
 }
 
-static bool hpd_ack(
-	struct irq_service *irq_service,
-	const struct irq_source_info *info)
-{
-	uint32_t addr = info->status_reg;
-	uint32_t value = dm_read_reg(irq_service->ctx, addr);
-	uint32_t current_status =
-		get_reg_field_value(
-			value,
-			HPD0_DC_HPD_INT_STATUS,
-			DC_HPD_SENSE_DELAYED);
-
-	dal_irq_service_ack_generic(irq_service, info);
-
-	value = dm_read_reg(irq_service->ctx, info->enable_reg);
-
-	set_reg_field_value(
-		value,
-		current_status ? 0 : 1,
-		HPD0_DC_HPD_INT_CONTROL,
-		DC_HPD_INT_POLARITY);
-
-	dm_write_reg(irq_service->ctx, info->enable_reg, value);
-
-	return true;
-}
-
 static struct irq_source_info_funcs hpd_irq_info_funcs = {
 	.set = NULL,
-	.ack = hpd_ack
+	.ack = hpd0_ack
 };
 
 static struct irq_source_info_funcs hpd_rx_irq_info_funcs = {
@@ -171,6 +144,16 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 	.ack = NULL
 };
 
+static struct irq_source_info_funcs vline1_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
+static struct irq_source_info_funcs vline2_irq_info_funcs = {
+	.set = NULL,
+	.ack = NULL
+};
+
 #undef BASE_INNER
 #define BASE_INNER(seg) DCN_BASE__INST0_SEG ## seg
 
@@ -239,6 +222,13 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 		.funcs = &pflip_irq_info_funcs\
 	}
 
+#define vblank_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\
+		IRQ_REG_ENTRY(OTG, reg_num,\
+			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_INT_EN,\
+			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_EVENT_CLEAR),\
+		.funcs = &vblank_irq_info_funcs\
+	}
 /* vupdate_no_lock_int_entry maps to DC_IRQ_SOURCE_VUPDATEx, to match semantic
  * of DCE's DC_IRQ_SOURCE_VUPDATEx.
  */
@@ -250,13 +240,6 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 		.funcs = &vupdate_no_lock_irq_info_funcs\
 	}
 
-#define vblank_int_entry(reg_num)\
-	[DC_IRQ_SOURCE_VBLANK1 + reg_num] = {\
-		IRQ_REG_ENTRY(OTG, reg_num,\
-			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_INT_EN,\
-			OTG_GLOBAL_SYNC_STATUS, VSTARTUP_EVENT_CLEAR),\
-		.funcs = &vblank_irq_info_funcs\
-	}
 #define vline0_int_entry(reg_num)\
 	[DC_IRQ_SOURCE_DC1_VLINE0 + reg_num] = {\
 		IRQ_REG_ENTRY(OTG, reg_num,\
@@ -264,6 +247,20 @@ static struct irq_source_info_funcs vline0_irq_info_funcs = {
 			OTG_VERTICAL_INTERRUPT0_CONTROL, OTG_VERTICAL_INTERRUPT0_CLEAR),\
 		.funcs = &vline0_irq_info_funcs\
 	}
+#define vline1_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_DC1_VLINE1 + reg_num] = {\
+		IRQ_REG_ENTRY(OTG, reg_num,\
+			OTG_VERTICAL_INTERRUPT1_CONTROL, OTG_VERTICAL_INTERRUPT1_INT_ENABLE,\
+			OTG_VERTICAL_INTERRUPT1_CONTROL, OTG_VERTICAL_INTERRUPT1_CLEAR),\
+		.funcs = &vline1_irq_info_funcs\
+	}
+#define vline2_int_entry(reg_num)\
+	[DC_IRQ_SOURCE_DC1_VLINE2 + reg_num] = {\
+		IRQ_REG_ENTRY(OTG, reg_num,\
+			OTG_VERTICAL_INTERRUPT2_CONTROL, OTG_VERTICAL_INTERRUPT2_INT_ENABLE,\
+			OTG_VERTICAL_INTERRUPT2_CONTROL, OTG_VERTICAL_INTERRUPT2_CLEAR),\
+		.funcs = &vline2_irq_info_funcs\
+	}
 #define dmub_outbox_int_entry()\
 	[DC_IRQ_SOURCE_DMCUB_OUTBOX] = {\
 		IRQ_REG_ENTRY_DMUB(\
@@ -364,21 +361,29 @@ irq_source_info_dcn401[DAL_IRQ_SOURCES_NUMBER] = {
 	dc_underflow_int_entry(6),
 	[DC_IRQ_SOURCE_DMCU_SCP] = dummy_irq_entry(),
 	[DC_IRQ_SOURCE_VBIOS_SW] = dummy_irq_entry(),
-	vupdate_no_lock_int_entry(0),
-	vupdate_no_lock_int_entry(1),
-	vupdate_no_lock_int_entry(2),
-	vupdate_no_lock_int_entry(3),
 	vblank_int_entry(0),
 	vblank_int_entry(1),
 	vblank_int_entry(2),
 	vblank_int_entry(3),
+	[DC_IRQ_SOURCE_DC5_VLINE1] = dummy_irq_entry(),
+	[DC_IRQ_SOURCE_DC6_VLINE1] = dummy_irq_entry(),
+	dmub_outbox_int_entry(),
+	vupdate_no_lock_int_entry(0),
+	vupdate_no_lock_int_entry(1),
+	vupdate_no_lock_int_entry(2),
+	vupdate_no_lock_int_entry(3),
 	vline0_int_entry(0),
 	vline0_int_entry(1),
 	vline0_int_entry(2),
 	vline0_int_entry(3),
-	[DC_IRQ_SOURCE_DC5_VLINE1] = dummy_irq_entry(),
-	[DC_IRQ_SOURCE_DC6_VLINE1] = dummy_irq_entry(),
-	dmub_outbox_int_entry(),
+	vline1_int_entry(0),
+	vline1_int_entry(1),
+	vline1_int_entry(2),
+	vline1_int_entry(3),
+	vline2_int_entry(0),
+	vline2_int_entry(1),
+	vline2_int_entry(2),
+	vline2_int_entry(3),
 };
 
 static const struct irq_service_funcs irq_service_funcs_dcn401 = {
diff --git a/drivers/gpu/drm/amd/display/dc/irq/irq_service.c b/drivers/gpu/drm/amd/display/dc/irq/irq_service.c
index eca3d7ee7e4e..b595a11c5eaf 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/irq_service.c
+++ b/drivers/gpu/drm/amd/display/dc/irq/irq_service.c
@@ -41,6 +41,16 @@
 #include "reg_helper.h"
 #include "irq_service.h"
 
+//HPD0_DC_HPD_INT_STATUS
+#define HPD0_DC_HPD_INT_STATUS__DC_HPD_SENSE_DELAYED_MASK		0x00000010L
+#define HPD0_DC_HPD_INT_CONTROL__DC_HPD_INT_POLARITY_MASK		0x00000100L
+#define HPD0_DC_HPD_INT_STATUS__DC_HPD_SENSE_DELAYED__SHIFT		0x4
+#define HPD0_DC_HPD_INT_CONTROL__DC_HPD_INT_POLARITY__SHIFT     0x8
+//HPD1_DC_HPD_INT_STATUS
+#define DC_HPD1_INT_STATUS__DC_HPD1_SENSE_DELAYED_MASK			0x10
+#define DC_HPD1_INT_STATUS__DC_HPD1_SENSE_DELAYED__SHIFT		0x4
+#define DC_HPD1_INT_CONTROL__DC_HPD1_INT_POLARITY_MASK			0x100
+#define DC_HPD1_INT_CONTROL__DC_HPD1_INT_POLARITY__SHIFT		0x8
 
 
 #define CTX \
@@ -177,3 +187,57 @@ enum dc_irq_source dal_irq_service_to_irq_source(
 		src_id,
 		ext_id);
 }
+
+bool hpd0_ack(
+	struct irq_service *irq_service,
+	const struct irq_source_info *info)
+{
+	uint32_t addr = info->status_reg;
+	uint32_t value = dm_read_reg(irq_service->ctx, addr);
+	uint32_t current_status =
+		get_reg_field_value(
+			value,
+			HPD0_DC_HPD_INT_STATUS,
+			DC_HPD_SENSE_DELAYED);
+
+	dal_irq_service_ack_generic(irq_service, info);
+
+	value = dm_read_reg(irq_service->ctx, info->enable_reg);
+
+	set_reg_field_value(
+		value,
+		current_status ? 0 : 1,
+		HPD0_DC_HPD_INT_CONTROL,
+		DC_HPD_INT_POLARITY);
+
+	dm_write_reg(irq_service->ctx, info->enable_reg, value);
+
+	return true;
+}
+
+bool hpd1_ack(
+	struct irq_service *irq_service,
+	const struct irq_source_info *info)
+{
+	uint32_t addr = info->status_reg;
+	uint32_t value = dm_read_reg(irq_service->ctx, addr);
+	uint32_t current_status =
+		get_reg_field_value(
+			value,
+			DC_HPD1_INT_STATUS,
+			DC_HPD1_SENSE_DELAYED);
+
+	dal_irq_service_ack_generic(irq_service, info);
+
+	value = dm_read_reg(irq_service->ctx, info->enable_reg);
+
+	set_reg_field_value(
+		value,
+		current_status ? 0 : 1,
+		DC_HPD1_INT_CONTROL,
+		DC_HPD1_INT_POLARITY);
+
+	dm_write_reg(irq_service->ctx, info->enable_reg, value);
+
+	return true;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/irq/irq_service.h b/drivers/gpu/drm/amd/display/dc/irq/irq_service.h
index b178f85944cd..bbcef3d2fe33 100644
--- a/drivers/gpu/drm/amd/display/dc/irq/irq_service.h
+++ b/drivers/gpu/drm/amd/display/dc/irq/irq_service.h
@@ -82,4 +82,12 @@ void dal_irq_service_set_generic(
 	const struct irq_source_info *info,
 	bool enable);
 
+bool hpd0_ack(
+	struct irq_service *irq_service,
+	const struct irq_source_info *info);
+
+bool hpd1_ack(
+	struct irq_service *irq_service,
+	const struct irq_source_info *info);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/irq_types.h b/drivers/gpu/drm/amd/display/dc/irq_types.h
index 110f656d43ae..a2f7b933bebf 100644
--- a/drivers/gpu/drm/amd/display/dc/irq_types.h
+++ b/drivers/gpu/drm/amd/display/dc/irq_types.h
@@ -161,6 +161,20 @@ enum dc_irq_source {
 	DC_IRQ_SOURCE_DPCX_TX_PHYE,
 	DC_IRQ_SOURCE_DPCX_TX_PHYF,
 
+	DC_IRQ_SOURCE_DC1_VLINE2,
+	DC_IRQ_SOURCE_DC2_VLINE2,
+	DC_IRQ_SOURCE_DC3_VLINE2,
+	DC_IRQ_SOURCE_DC4_VLINE2,
+	DC_IRQ_SOURCE_DC5_VLINE2,
+	DC_IRQ_SOURCE_DC6_VLINE2,
+
+	DC_IRQ_SOURCE_DCI2C_RR_DDC1,
+	DC_IRQ_SOURCE_DCI2C_RR_DDC2,
+	DC_IRQ_SOURCE_DCI2C_RR_DDC3,
+	DC_IRQ_SOURCE_DCI2C_RR_DDC4,
+	DC_IRQ_SOURCE_DCI2C_RR_DDC5,
+	DC_IRQ_SOURCE_DCI2C_RR_DDC6,
+
 	DAL_IRQ_SOURCES_NUMBER
 };
 
@@ -170,6 +184,8 @@ enum irq_type
 	IRQ_TYPE_VUPDATE = DC_IRQ_SOURCE_VUPDATE1,
 	IRQ_TYPE_VBLANK = DC_IRQ_SOURCE_VBLANK1,
 	IRQ_TYPE_VLINE0 = DC_IRQ_SOURCE_DC1_VLINE0,
+	IRQ_TYPE_VLINE1 = DC_IRQ_SOURCE_DC1_VLINE1,
+	IRQ_TYPE_VLINE2 = DC_IRQ_SOURCE_DC1_VLINE2,
 	IRQ_TYPE_DCUNDERFLOW = DC_IRQ_SOURCE_DC1UNDERFLOW,
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
index cc9191a5c9e6..9655e6fa53a4 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
@@ -611,6 +611,7 @@ static bool detect_dp(struct dc_link *link,
 		link->dpcd_caps.dongle_type = sink_caps->dongle_type;
 		link->dpcd_caps.is_dongle_type_one = sink_caps->is_dongle_type_one;
 		link->dpcd_caps.dpcd_rev.raw = 0;
+		link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.raw = 0;
 	}
 
 	return true;
@@ -1007,21 +1008,11 @@ static bool detect_link_and_local_sink(struct dc_link *link,
 					link->reported_link_cap.link_rate > LINK_RATE_HIGH3)
 				link->reported_link_cap.link_rate = LINK_RATE_HIGH3;
 
-			/*
-			 * If this is DP over USB4 link then we need to:
-			 * - Enable BW ALLOC support on DPtx if applicable
-			 */
-			if (dc->config.usb4_bw_alloc_support) {
-				if (link_dp_dpia_set_dptx_usb4_bw_alloc_support(link)) {
-					/* update with non reduced link cap if bw allocation mode is supported */
-					if (link->dpia_bw_alloc_config.nrd_max_link_rate &&
-						link->dpia_bw_alloc_config.nrd_max_lane_count) {
-						link->reported_link_cap.link_rate =
-							link->dpia_bw_alloc_config.nrd_max_link_rate;
-						link->reported_link_cap.lane_count =
-							link->dpia_bw_alloc_config.nrd_max_lane_count;
-					}
-				}
+			if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling
+					&& link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc
+					&& link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support) {
+				if (link_dpia_enable_usb4_dp_bw_alloc_mode(link) == false)
+					link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc = false;
 			}
 			break;
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 268626e73c54..273a3be6d593 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -148,6 +148,7 @@ void link_blank_dp_stream(struct dc_link *link, bool hw_init)
 void link_set_all_streams_dpms_off_for_link(struct dc_link *link)
 {
 	struct pipe_ctx *pipes[MAX_PIPES];
+	struct dc_stream_state *streams[MAX_PIPES];
 	struct dc_state *state = link->dc->current_state;
 	uint8_t count;
 	int i;
@@ -160,10 +161,18 @@ void link_set_all_streams_dpms_off_for_link(struct dc_link *link)
 
 	link_get_master_pipes_with_dpms_on(link, state, &count, pipes);
 
+	/* The subsequent call to dc_commit_updates_for_stream for a full update
+	 * will release the current state and swap to a new state. Releasing the
+	 * current state results in the stream pointers in the pipe_ctx structs
+	 * to be zero'd. Hence, cache all streams prior to dc_commit_updates_for_stream.
+	 */
+	for (i = 0; i < count; i++)
+		streams[i] = pipes[i]->stream;
+
 	for (i = 0; i < count; i++) {
-		stream_update.stream = pipes[i]->stream;
+		stream_update.stream = streams[i];
 		dc_commit_updates_for_stream(link->ctx->dc, NULL, 0,
-				pipes[i]->stream, &stream_update,
+				streams[i], &stream_update,
 				state);
 	}
 
@@ -2365,7 +2374,7 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx)
 	update_psp_stream_config(pipe_ctx, true);
 	dc->hwss.blank_stream(pipe_ctx);
 
-	if (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA)
+	if (pipe_ctx->link_config.dp_tunnel_settings.should_use_dp_bw_allocation)
 		deallocate_usb4_bandwidth(pipe_ctx->stream);
 
 	if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
@@ -2433,7 +2442,7 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx)
 	if (link->connector_signal == SIGNAL_TYPE_EDP && dc->debug.psp_disabled_wa) {
 		/* reset internal save state to default since eDP is  off */
 		enum dp_panel_mode panel_mode = dp_get_panel_mode(pipe_ctx->stream->link);
-		/* since current psp not loaded, we need to reset it to default*/
+		/* since current psp not loaded, we need to reset it to default */
 		link->panel_mode = panel_mode;
 	}
 }
@@ -2611,7 +2620,7 @@ void link_set_dpms_on(
 	if (dc_is_dp_signal(pipe_ctx->stream->signal))
 		dp_set_hblank_reduction_on_rx(pipe_ctx);
 
-	if (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA)
+	if (pipe_ctx->link_config.dp_tunnel_settings.should_use_dp_bw_allocation)
 		allocate_usb4_bandwidth(pipe_ctx->stream);
 
 	if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
index f6b6b19e7481..1a04f4b74585 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
@@ -156,6 +156,7 @@ static void construct_link_service_dp_capability(struct link_service *link_srv)
 	link_srv->dp_get_encoding_format = link_dp_get_encoding_format;
 	link_srv->dp_should_enable_fec = dp_should_enable_fec;
 	link_srv->dp_decide_link_settings = link_decide_link_settings;
+	link_srv->dp_decide_tunnel_settings = link_decide_dp_tunnel_settings;
 	link_srv->mst_decide_link_encoding_format =
 			mst_decide_link_encoding_format;
 	link_srv->edp_decide_link_settings = edp_decide_link_settings;
@@ -464,6 +465,7 @@ static bool construct_phy(struct dc_link *link,
 
 	link->irq_source_hpd = DC_IRQ_SOURCE_INVALID;
 	link->irq_source_hpd_rx = DC_IRQ_SOURCE_INVALID;
+	link->irq_source_read_request = DC_IRQ_SOURCE_INVALID;
 	link->link_status.dpcd_caps = &link->dpcd_caps;
 
 	link->dc = init_params->dc;
@@ -514,6 +516,9 @@ static bool construct_phy(struct dc_link *link,
 	case CONNECTOR_ID_HDMI_TYPE_A:
 		link->connector_signal = SIGNAL_TYPE_HDMI_TYPE_A;
 
+		if (link->hpd_gpio)
+			link->irq_source_read_request =
+					dal_irq_get_read_request(link->hpd_gpio);
 		break;
 	case CONNECTOR_ID_SINGLE_LINK_DVID:
 	case CONNECTOR_ID_SINGLE_LINK_DVII:
@@ -653,7 +658,7 @@ static bool construct_phy(struct dc_link *link,
 		}
 
 		/* Look for device tag that matches connector signal,
-		 * CRT for rgb, LCD for other supported signal tyes
+		 * CRT for rgb, LCD for other supported signal types
 		 */
 		if (!bp_funcs->is_device_id_supported(dc_ctx->dc_bios,
 						      link->device_tag.dev_id))
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 21ee0d96c9d4..8f79881ad9f1 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -158,6 +158,14 @@ uint8_t dp_parse_lttpr_repeater_count(uint8_t lttpr_repeater_count)
 	return 0; // invalid value
 }
 
+uint32_t dp_get_closest_lttpr_offset(uint8_t lttpr_count)
+{
+	/* Calculate offset for LTTPR closest to DPTX which is highest in the chain
+	 * Offset is 0 for single LTTPR cases as base LTTPR DPCD addresses target LTTPR 1
+	 */
+	return DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE * (lttpr_count - 1);
+}
+
 uint32_t link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw)
 {
 	switch (bw) {
@@ -2013,11 +2021,9 @@ static bool retrieve_link_cap(struct dc_link *link)
 			sizeof(link->dpcd_caps.max_uncompressed_pixel_rate_cap.raw));
 
 	/* Read DP tunneling information. */
-	if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) {
-		status = dpcd_get_tunneling_device_data(link);
-		if (status != DC_OK)
-			dm_error("%s: Read DP tunneling device data failed.\n", __func__);
-	}
+	status = dpcd_get_tunneling_device_data(link);
+	if (status != DC_OK)
+		dm_error("%s: Read DP tunneling device data failed.\n", __func__);
 
 	retrieve_cable_id(link);
 	dpcd_write_cable_id_to_dprx(link);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
index 0ce0af3ddbeb..940b147cc5d4 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
@@ -48,6 +48,9 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link);
 /* Convert PHY repeater count read from DPCD uint8_t. */
 uint8_t dp_parse_lttpr_repeater_count(uint8_t lttpr_repeater_count);
 
+/* Calculate embedded LTTPR address offset for vendor-specific behaviour */
+uint32_t dp_get_closest_lttpr_offset(uint8_t lttpr_count);
+
 bool dp_is_sink_present(struct dc_link *link);
 
 bool dp_is_lttpr_present(struct dc_link *link);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c
index 0d123e647652..22bfdced64ab 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c
@@ -62,6 +62,36 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link)
 	if (status != DC_OK)
 		goto err;
 
+	link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.raw =
+			dpcd_dp_tun_data[DP_TUNNELING_CAPABILITIES_SUPPORT - DP_TUNNELING_CAPABILITIES_SUPPORT];
+
+	if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling == false)
+		goto err;
+
+	link->dpcd_caps.usb4_dp_tun_info.dpia_info.raw =
+			dpcd_dp_tun_data[DP_IN_ADAPTER_INFO - DP_TUNNELING_CAPABILITIES_SUPPORT];
+	link->dpcd_caps.usb4_dp_tun_info.usb4_driver_id =
+			dpcd_dp_tun_data[DP_USB4_DRIVER_ID - DP_TUNNELING_CAPABILITIES_SUPPORT];
+
+	if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc) {
+		status = core_link_read_dpcd(link, USB4_DRIVER_BW_CAPABILITY,
+				dpcd_dp_tun_data, 1);
+
+		if (status != DC_OK)
+			goto err;
+
+		link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.raw = dpcd_dp_tun_data[0];
+	}
+
+	DC_LOG_DEBUG("%s: Link[%d]  DP tunneling support  (RouterId=%d  AdapterId=%d)  "
+			"DPIA_BW_Alloc_support=%d "
+			"CM_BW_Alloc_support=%d ",
+			__func__, link->link_index,
+			link->dpcd_caps.usb4_dp_tun_info.usb4_driver_id,
+			link->dpcd_caps.usb4_dp_tun_info.dpia_info.bits.dpia_num,
+			link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc,
+			link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support);
+
 	status = core_link_read_dpcd(
 			link,
 			DP_USB4_ROUTER_TOPOLOGY_ID,
@@ -71,13 +101,6 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link)
 	if (status != DC_OK)
 		goto err;
 
-	link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.raw =
-			dpcd_dp_tun_data[DP_TUNNELING_CAPABILITIES_SUPPORT - DP_TUNNELING_CAPABILITIES_SUPPORT];
-	link->dpcd_caps.usb4_dp_tun_info.dpia_info.raw =
-			dpcd_dp_tun_data[DP_IN_ADAPTER_INFO - DP_TUNNELING_CAPABILITIES_SUPPORT];
-	link->dpcd_caps.usb4_dp_tun_info.usb4_driver_id =
-			dpcd_dp_tun_data[DP_USB4_DRIVER_ID - DP_TUNNELING_CAPABILITIES_SUPPORT];
-
 	for (i = 0; i < DPCD_USB4_TOPOLOGY_ID_LEN; i++)
 		link->dpcd_caps.usb4_dp_tun_info.usb4_topology_id[i] = dpcd_topology_data[i];
 
@@ -92,6 +115,7 @@ bool dpia_query_hpd_status(struct dc_link *link)
 
 	/* prepare QUERY_HPD command */
 	cmd.query_hpd.header.type = DMUB_CMD__QUERY_HPD_STATE;
+	cmd.query_hpd.header.payload_bytes = sizeof(cmd.query_hpd.data);
 	cmd.query_hpd.data.instance = link->link_id.enum_id - ENUM_ID_1;
 	cmd.query_hpd.data.ch_type = AUX_CHANNEL_DPIA;
 
@@ -119,3 +143,20 @@ bool dpia_query_hpd_status(struct dc_link *link)
 	return link->hpd_status;
 }
 
+void link_decide_dp_tunnel_settings(struct dc_stream_state *stream,
+			struct dc_tunnel_settings *dp_tunnel_setting)
+{
+	struct dc_link *link = stream->link;
+
+	memset(dp_tunnel_setting, 0, sizeof(*dp_tunnel_setting));
+
+	if ((stream->signal == SIGNAL_TYPE_DISPLAY_PORT) || (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)) {
+		dp_tunnel_setting->should_enable_dp_tunneling =
+					link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling;
+
+		if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc
+				&& link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support)
+			dp_tunnel_setting->should_use_dp_bw_allocation = true;
+	}
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h
index 363f45a1a964..a61edfc9ca7a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h
@@ -38,4 +38,10 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link);
  * Returns true if HPD high.
  */
 bool dpia_query_hpd_status(struct dc_link *link);
+
+/* Decide the DP tunneling settings based on the DPCD capabilities
+ */
+void link_decide_dp_tunnel_settings(struct dc_stream_state *stream,
+			struct dc_tunnel_settings *dp_tunnel_setting);
+
 #endif /* __DC_LINK_DPIA_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
index a254ead2f7e8..3af7564a84f1 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
@@ -46,9 +46,10 @@
  */
 static bool link_dp_is_bw_alloc_available(struct dc_link *link)
 {
-	return (link && link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA
-		&& link->hpd_status
-		&& link->dpia_bw_alloc_config.bw_alloc_enabled);
+	return (link && link->hpd_status
+		&& link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling
+		&& link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc
+		&& link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support);
 }
 
 static void reset_bw_alloc_struct(struct dc_link *link)
@@ -141,7 +142,7 @@ static int get_non_reduced_max_lane_count(struct dc_link *link)
  * granuality, Driver_ID, CM_Group, & populate the BW allocation structs
  * for host router and dpia
  */
-static void init_usb4_bw_struct(struct dc_link *link)
+static void retrieve_usb4_dp_bw_allocation_info(struct dc_link *link)
 {
 	reset_bw_alloc_struct(link);
 
@@ -282,49 +283,26 @@ static void link_dpia_send_bw_alloc_request(struct dc_link *link, int req_bw)
 // ------------------------------------------------------------------
 // PUBLIC FUNCTIONS
 // ------------------------------------------------------------------
-bool link_dp_dpia_set_dptx_usb4_bw_alloc_support(struct dc_link *link)
+bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link)
 {
 	bool ret = false;
-	uint8_t response = 0,
-			bw_support_dpia = 0,
-			bw_support_cm = 0;
+	uint8_t val;
 
-	if (!(link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && link->hpd_status))
-		goto out;
+	if (link->hpd_status) {
+		val = DPTX_BW_ALLOC_MODE_ENABLE | DPTX_BW_ALLOC_UNMASK_IRQ;
 
-	if (core_link_read_dpcd(
-			link,
-			DP_TUNNELING_CAPABILITIES,
-			&response,
-			sizeof(uint8_t)) == DC_OK)
-		bw_support_dpia = (response >> 7) & 1;
-
-	if (core_link_read_dpcd(
-		link,
-		USB4_DRIVER_BW_CAPABILITY,
-		&response,
-		sizeof(uint8_t)) == DC_OK)
-		bw_support_cm = (response >> 7) & 1;
-
-	/* Send request acknowledgment to Turn ON DPTX support */
-	if (bw_support_cm && bw_support_dpia) {
-
-		response = 0x80;
-		if (core_link_write_dpcd(
-				link,
-				DPTX_BW_ALLOCATION_MODE_CONTROL,
-				&response,
-				sizeof(uint8_t)) != DC_OK) {
-			DC_LOG_DEBUG("%s: FAILURE Enabling DPtx BW Allocation Mode Support for link(%d)\n",
-				__func__, link->link_index);
-		} else {
-			// SUCCESS Enabled DPtx BW Allocation Mode Support
-			DC_LOG_DEBUG("%s: SUCCESS Enabling DPtx BW Allocation Mode Support for link(%d)\n",
-				__func__, link->link_index);
+		if (core_link_write_dpcd(link, DPTX_BW_ALLOCATION_MODE_CONTROL, &val, sizeof(uint8_t)) == DC_OK) {
+			DC_LOG_DEBUG("%s:  link[%d] DPTX BW allocation mode enabled", __func__, link->link_index);
+
+			retrieve_usb4_dp_bw_allocation_info(link);
+
+			if (link->dpia_bw_alloc_config.nrd_max_link_rate && link->dpia_bw_alloc_config.nrd_max_lane_count) {
+				link->reported_link_cap.link_rate = link->dpia_bw_alloc_config.nrd_max_link_rate;
+				link->reported_link_cap.lane_count = link->dpia_bw_alloc_config.nrd_max_lane_count;
+			}
 
-			ret = true;
-			init_usb4_bw_struct(link);
 			link->dpia_bw_alloc_config.bw_alloc_enabled = true;
+			ret = true;
 
 			/*
 			 * During DP tunnel creation, CM preallocates BW and reduces estimated BW of other
@@ -332,11 +310,12 @@ bool link_dp_dpia_set_dptx_usb4_bw_alloc_support(struct dc_link *link)
 			 * to make the CM to release preallocation and update estimated BW correctly for
 			 * all DPIAs per host router
 			 */
+			// TODO: Zero allocation can be removed once the MSFT CM fix has been released
 			link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0);
-		}
+		} else
+			DC_LOG_DEBUG("%s:  link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index);
 	}
 
-out:
 	return ret;
 }
 
@@ -378,7 +357,8 @@ void link_dp_dpia_handle_bw_alloc_status(struct dc_link *link, uint8_t status)
  */
 void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int peak_bw)
 {
-	if (link && link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && link->dpia_bw_alloc_config.bw_alloc_enabled) {
+	if (link && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling
+			&& link->dpia_bw_alloc_config.bw_alloc_enabled) {
 		//1. Hot Plug
 		if (link->hpd_status && peak_bw > 0) {
 			// If DP over USB4 then we need to check BW allocation
@@ -401,7 +381,7 @@ void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int r
 	if (link_dp_is_bw_alloc_available(link))
 		link_dpia_send_bw_alloc_request(link, req_bw);
 	else
-		DC_LOG_DEBUG("%s:  Not able to send the BW Allocation request", __func__);
+		DC_LOG_DEBUG("%s:  BW Allocation mode not available", __func__);
 }
 
 bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed_per_dpia, const unsigned int num_dpias)
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h
index 6df9b946b00f..801965b5f9a4 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h
@@ -43,13 +43,13 @@ enum bw_type {
 };
 
 /*
- * Enable BW Allocation Mode Support from the DP-Tx side
+ * Enable USB4 DP BW allocation mode
  *
  * @link: pointer to the dc_link struct instance
  *
  * return: SUCCESS or FAILURE
  */
-bool link_dp_dpia_set_dptx_usb4_bw_alloc_support(struct dc_link *link);
+bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link);
 
 /*
  * Allocates only what the stream needs for bw, so if:
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
index 5be00e4ce10b..693477413347 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
@@ -229,6 +229,10 @@ static void handle_hpd_irq_replay_sink(struct dc_link *link)
 
 		link->replay_settings.config.replay_error_status.raw |= replay_error_status.raw;
 
+		/* Increment desync error counter if a desync error is detected */
+		if (replay_configuration.bits.DESYNC_ERROR_STATUS)
+			link->replay_settings.replay_desync_error_fail_count++;
+
 		if (link->replay_settings.config.force_disable_desync_error_check)
 			return;
 
@@ -240,9 +244,6 @@ static void handle_hpd_irq_replay_sink(struct dc_link *link)
 			&replay_configuration.raw,
 			sizeof(replay_configuration.raw));
 
-		/* Update desync error counter */
-		link->replay_settings.replay_desync_error_fail_count++;
-
 		/* Acknowledge and clear error bits */
 		dm_helpers_dp_write_dpcd(
 			link->ctx,
@@ -351,7 +352,7 @@ enum dc_status dp_read_hpd_rx_irq_data(
 			irq_data->raw,
 			DP_SINK_STATUS - DP_SINK_COUNT + 1);
 
-		if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) {
+		if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling) {
 			retval = core_link_read_dpcd(
 					link, DP_LINK_SERVICE_IRQ_VECTOR_ESI0,
 					&irq_data->bytes.link_service_irq_esi0.raw, 1);
@@ -520,7 +521,7 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link,
 		dp_trace_link_loss_increment(link);
 	}
 
-	if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) {
+	if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling) {
 		if (hpd_irq_dpcd_data.bytes.link_service_irq_esi0.bits.DP_LINK_TUNNELING_IRQ)
 			dp_handle_tunneling_irq(link);
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
index ef358afdfb65..2dc1a660e504 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
@@ -785,7 +785,6 @@ void override_training_settings(
 		lt_settings->lttpr_mode = LTTPR_MODE_NON_LTTPR;
 
 	dp_get_lttpr_mode_override(link, &lt_settings->lttpr_mode);
-
 }
 
 enum dc_dp_training_pattern decide_cr_training_pattern(
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
index 34d2e097ca2e..66d0fb1b9b9d 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
@@ -35,6 +35,17 @@
 #define DC_LOGGER \
 	link->ctx->logger
 
+static void get_default_8b_10b_lttpr_aux_rd_interval(
+		union training_aux_rd_interval *training_rd_interval)
+{
+	/* LTTPR are required to program DPCD 0000Eh to 0x4 (16ms) upon AUX
+	 * read reply to this register. Since old sinks with DPCD rev 1.1
+	 * and earlier may not support this register, assume the mandatory
+	 * value is programmed by the LTTPR to avoid AUX timeout issues.
+	 */
+	training_rd_interval->raw = 0x4;
+}
+
 static int32_t get_cr_training_aux_rd_interval(struct dc_link *link,
 		const struct dc_link_settings *link_settings,
 		enum lttpr_mode lttpr_mode)
@@ -43,17 +54,22 @@ static int32_t get_cr_training_aux_rd_interval(struct dc_link *link,
 	uint32_t wait_in_micro_secs = 100;
 
 	memset(&training_rd_interval, 0, sizeof(training_rd_interval));
-	if (link_dp_get_encoding_format(link_settings) == DP_8b_10b_ENCODING &&
-			link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12) {
-		core_link_read_dpcd(
-				link,
-				DP_TRAINING_AUX_RD_INTERVAL,
-				(uint8_t *)&training_rd_interval,
-				sizeof(training_rd_interval));
-		if (lttpr_mode != LTTPR_MODE_NON_TRANSPARENT)
-			wait_in_micro_secs = 400;
-		if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL)
-			wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000;
+	if (link_dp_get_encoding_format(link_settings) == DP_8b_10b_ENCODING) {
+		if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12)
+			core_link_read_dpcd(
+					link,
+					DP_TRAINING_AUX_RD_INTERVAL,
+					(uint8_t *)&training_rd_interval,
+					sizeof(training_rd_interval));
+		else if (dp_is_lttpr_present(link))
+			get_default_8b_10b_lttpr_aux_rd_interval(&training_rd_interval);
+
+		if (training_rd_interval.raw != 0) {
+			if (lttpr_mode != LTTPR_MODE_NON_TRANSPARENT)
+				wait_in_micro_secs = 400;
+			if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL)
+				wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000;
+		}
 	}
 	return wait_in_micro_secs;
 }
@@ -71,13 +87,15 @@ static uint32_t get_eq_training_aux_rd_interval(
 				DP_128B132B_TRAINING_AUX_RD_INTERVAL,
 				(uint8_t *)&training_rd_interval,
 				sizeof(training_rd_interval));
-	} else if (link_dp_get_encoding_format(link_settings) == DP_8b_10b_ENCODING &&
-			link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12) {
-		core_link_read_dpcd(
-				link,
-				DP_TRAINING_AUX_RD_INTERVAL,
-				(uint8_t *)&training_rd_interval,
-				sizeof(training_rd_interval));
+	} else if (link_dp_get_encoding_format(link_settings) == DP_8b_10b_ENCODING) {
+		if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12)
+			core_link_read_dpcd(
+					link,
+					DP_TRAINING_AUX_RD_INTERVAL,
+					(uint8_t *)&training_rd_interval,
+					sizeof(training_rd_interval));
+		else if (dp_is_lttpr_present(link))
+			get_default_8b_10b_lttpr_aux_rd_interval(&training_rd_interval);
 	}
 
 	switch (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL) {
@@ -124,6 +142,14 @@ void decide_8b_10b_training_settings(
 	lt_settings->lttpr_mode = dp_decide_8b_10b_lttpr_mode(link);
 	lt_settings->cr_pattern_time = get_cr_training_aux_rd_interval(link, link_setting, lt_settings->lttpr_mode);
 	dp_hw_to_dpcd_lane_settings(lt_settings, lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings);
+
+	/* Some embedded LTTPRs rely on receiving TPS2 before LT to interop reliably with sensitive VGA dongles
+	 * This allows these LTTPRs to minimize freq/phase and skew variation during lock and deskew sequences
+	 */
+	if ((link->chip_caps & AMD_EXT_DISPLAY_PATH_CAPS__EXT_CHIP_MASK) ==
+			AMD_EXT_DISPLAY_PATH_CAPS__DP_EARLY_8B10B_TPS2) {
+		lt_settings->lttpr_early_tps2 = true;
+	}
 }
 
 enum lttpr_mode dp_decide_8b_10b_lttpr_mode(struct dc_link *link)
@@ -155,6 +181,42 @@ enum lttpr_mode dp_decide_8b_10b_lttpr_mode(struct dc_link *link)
 	return LTTPR_MODE_NON_LTTPR;
 }
 
+static void set_link_settings_and_perform_early_tps2_retimer_pre_lt_sequence(struct dc_link *link,
+	const struct link_resource *link_res,
+	struct link_training_settings *lt_settings,
+	uint32_t lttpr_count)
+{
+	/* Vendor-specific LTTPR early TPS2 sequence:
+	* 1. Output TPS2
+	* 2. Wait 400us
+	* 3. Set link settings as usual
+	* 4. Write TPS1 to DP_TRAINING_PATTERN_SET_PHY_REPEATERx targeting LTTPR closest to host
+	* 5. Wait 1ms
+	* 6. Begin link training as usual
+	* */
+
+	uint32_t closest_lttpr_address_offset = dp_get_closest_lttpr_offset(lttpr_count);
+
+	union dpcd_training_pattern dpcd_pattern = {0};
+
+	dpcd_pattern.v1_4.TRAINING_PATTERN_SET = 1;
+	dpcd_pattern.v1_4.SCRAMBLING_DISABLE = 1;
+
+	DC_LOG_HW_LINK_TRAINING("%s\n GPU sends TPS2. Wait 400us.\n", __func__);
+
+	dp_set_hw_training_pattern(link, link_res, DP_TRAINING_PATTERN_SEQUENCE_2, DPRX);
+
+	dp_set_hw_lane_settings(link, link_res, lt_settings, DPRX);
+
+	udelay(400);
+
+	dpcd_set_link_settings(link, lt_settings);
+
+	core_link_write_dpcd(link, DP_TRAINING_PATTERN_SET_PHY_REPEATER1 + closest_lttpr_address_offset, &dpcd_pattern.raw, 1);
+
+	udelay(1000);
+	}
+
 enum link_training_result perform_8b_10b_clock_recovery_sequence(
 	struct dc_link *link,
 	const struct link_resource *link_res,
@@ -365,7 +427,7 @@ enum link_training_result dp_perform_8b_10b_link_training(
 {
 	enum link_training_result status = LINK_TRAINING_SUCCESS;
 
-	uint8_t repeater_cnt;
+	uint8_t repeater_cnt = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
 	uint8_t repeater_id;
 	uint8_t lane = 0;
 
@@ -373,14 +435,16 @@ enum link_training_result dp_perform_8b_10b_link_training(
 		start_clock_recovery_pattern_early(link, link_res, lt_settings, DPRX);
 
 	/* 1. set link rate, lane count and spread. */
-	dpcd_set_link_settings(link, lt_settings);
+	if (lt_settings->lttpr_early_tps2)
+		set_link_settings_and_perform_early_tps2_retimer_pre_lt_sequence(link, link_res, lt_settings, repeater_cnt);
+	else
+		dpcd_set_link_settings(link, lt_settings);
 
 	if (lt_settings->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) {
 
 		/* 2. perform link training (set link training done
 		 *  to false is done as well)
 		 */
-		repeater_cnt = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
 
 		for (repeater_id = repeater_cnt; (repeater_id > 0 && status == LINK_TRAINING_SUCCESS);
 				repeater_id--) {
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index 1e4adbc764ea..da74c2b5854f 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -524,7 +524,7 @@ bool edp_set_backlight_level(const struct dc_link *link,
 	struct dc  *dc = link->ctx->dc;
 	uint32_t backlight_pwm_u16_16 = backlight_level_params->backlight_pwm_u16_16;
 	uint32_t frame_ramp = backlight_level_params->frame_ramp;
-	DC_LOGGER_INIT(link->ctx->logger);
+
 	DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n",
 			backlight_pwm_u16_16, backlight_pwm_u16_16);
 
@@ -1022,6 +1022,9 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream
 			&alpm_config.raw,
 			sizeof(alpm_config.raw));
 	}
+
+	link->replay_settings.config.replay_video_conferencing_optimization_enabled = false;
+
 	return true;
 }
 
@@ -1130,11 +1133,11 @@ static struct abm *get_abm_from_stream_res(const struct dc_link *link)
 	struct abm *abm = NULL;
 
 	for (i = 0; i < MAX_PIPES; i++) {
-		struct pipe_ctx pipe_ctx = dc->current_state->res_ctx.pipe_ctx[i];
-		struct dc_stream_state *stream = pipe_ctx.stream;
+		struct pipe_ctx *pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
+		struct dc_stream_state *stream = pipe_ctx->stream;
 
 		if (stream && stream->link == link) {
-			abm = pipe_ctx.stream_res.abm;
+			abm = pipe_ctx->stream_res.abm;
 			break;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c
index a0e9e9f0441a..b4cea2b8cb2a 100644
--- a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c
+++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c
@@ -370,275 +370,279 @@ void mpc32_program_shaper_luta_settings(
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].red.custom_float_y);
 
 	curve = params->arr_curve_points;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_0_1[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_2_3[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_4_5[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_6_7[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_8_9[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_10_11[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_12_13[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_14_15[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_16_17[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_18_19[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_20_21[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_22_23[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_24_25[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_26_27[mpcc_id], 0,
+	if (curve) {
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_0_1[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_28_29[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_30_31[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_32_33[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-}
-
-
-void mpc32_program_shaper_lutb_settings(
-		struct mpc *mpc,
-		const struct pwl_params *params,
-		uint32_t mpcc_id)
-{
-	const struct gamma_curve *curve;
-	struct dcn30_mpc *mpc30 = TO_DCN30_MPC(mpc);
-
-	REG_SET_2(MPCC_MCM_SHAPER_RAMB_START_CNTL_B[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].blue.custom_float_x,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
-	REG_SET_2(MPCC_MCM_SHAPER_RAMB_START_CNTL_G[mpcc_id], 0,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].green.custom_float_x,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
-	REG_SET_2(MPCC_MCM_SHAPER_RAMB_START_CNTL_R[mpcc_id], 0,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].red.custom_float_x,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
-
-	REG_SET_2(MPCC_MCM_SHAPER_RAMB_END_CNTL_B[mpcc_id], 0,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].blue.custom_float_x,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].blue.custom_float_y);
-	REG_SET_2(MPCC_MCM_SHAPER_RAMB_END_CNTL_G[mpcc_id], 0,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].green.custom_float_x,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].green.custom_float_y);
-	REG_SET_2(MPCC_MCM_SHAPER_RAMB_END_CNTL_R[mpcc_id], 0,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].red.custom_float_x,
-			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].red.custom_float_y);
-
-	curve = params->arr_curve_points;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_0_1[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_2_3[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_2_3[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_4_5[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_4_5[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_6_7[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_6_7[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_8_9[mpcc_id], 0,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
-		MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_8_9[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_10_11[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_10_11[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_12_13[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_12_13[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_14_15[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_14_15[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_16_17[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_16_17[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_18_19[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_18_19[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_20_21[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_20_21[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_22_23[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_22_23[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_24_25[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_24_25[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_26_27[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_26_27[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_28_29[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_28_29[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_30_31[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_30_31[mpcc_id], 0,
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMA_REGION_32_33[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+	}
+}
+
+
+void mpc32_program_shaper_lutb_settings(
+		struct mpc *mpc,
+		const struct pwl_params *params,
+		uint32_t mpcc_id)
+{
+	const struct gamma_curve *curve;
+	struct dcn30_mpc *mpc30 = TO_DCN30_MPC(mpc);
+
+	REG_SET_2(MPCC_MCM_SHAPER_RAMB_START_CNTL_B[mpcc_id], 0,
+		MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].blue.custom_float_x,
+		MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
+	REG_SET_2(MPCC_MCM_SHAPER_RAMB_START_CNTL_G[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].green.custom_float_x,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
+	REG_SET_2(MPCC_MCM_SHAPER_RAMB_START_CNTL_R[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_B, params->corner_points[0].red.custom_float_x,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_START_SEGMENT_B, 0);
 
-	curve += 2;
-	REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_32_33[mpcc_id], 0,
+	REG_SET_2(MPCC_MCM_SHAPER_RAMB_END_CNTL_B[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].blue.custom_float_x,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].blue.custom_float_y);
+	REG_SET_2(MPCC_MCM_SHAPER_RAMB_END_CNTL_G[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].green.custom_float_x,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].green.custom_float_y);
+	REG_SET_2(MPCC_MCM_SHAPER_RAMB_END_CNTL_R[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_B, params->corner_points[1].red.custom_float_x,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION_END_BASE_B, params->corner_points[1].red.custom_float_y);
+
+	curve = params->arr_curve_points;
+	if (curve) {
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_0_1[mpcc_id], 0,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
 			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_2_3[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_4_5[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_6_7[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_8_9[mpcc_id], 0,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+			MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_10_11[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_12_13[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_14_15[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_16_17[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_18_19[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_20_21[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_22_23[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_24_25[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_26_27[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_28_29[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_30_31[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+
+		curve += 2;
+		REG_SET_4(MPCC_MCM_SHAPER_RAMB_REGION_32_33[mpcc_id], 0,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_LUT_OFFSET, curve[0].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION0_NUM_SEGMENTS, curve[0].segments_num,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_LUT_OFFSET, curve[1].offset,
+				MPCC_MCM_SHAPER_RAMA_EXP_REGION1_NUM_SEGMENTS, curve[1].segments_num);
+	}
 }
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c
index ad67197557ca..98cf0cbd59ba 100644
--- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c
+++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c
@@ -47,16 +47,6 @@ void mpc401_update_3dlut_fast_load_select(struct mpc *mpc, int mpcc_id, int hubp
 	REG_SET(MPCC_MCM_3DLUT_FAST_LOAD_SELECT[mpcc_id], 0, MPCC_MCM_3DLUT_FL_SEL, hubp_idx);
 }
 
-void mpc401_get_3dlut_fast_load_status(struct mpc *mpc, int mpcc_id, uint32_t *done, uint32_t *soft_underflow, uint32_t *hard_underflow)
-{
-	struct dcn401_mpc *mpc401 = TO_DCN401_MPC(mpc);
-
-	REG_GET_3(MPCC_MCM_3DLUT_FAST_LOAD_STATUS[mpcc_id],
-			MPCC_MCM_3DLUT_FL_DONE, done,
-			MPCC_MCM_3DLUT_FL_SOFT_UNDERFLOW, soft_underflow,
-			MPCC_MCM_3DLUT_FL_HARD_UNDERFLOW, hard_underflow);
-}
-
 void mpc401_set_movable_cm_location(struct mpc *mpc, enum mpcc_movable_cm_location location, int mpcc_id)
 {
 	struct dcn401_mpc *mpc401 = TO_DCN401_MPC(mpc);
@@ -618,7 +608,6 @@ static const struct mpc_funcs dcn401_mpc_funcs = {
 	.set_bg_color = mpc1_set_bg_color,
 	.set_movable_cm_location = mpc401_set_movable_cm_location,
 	.update_3dlut_fast_load_select = mpc401_update_3dlut_fast_load_select,
-	.get_3dlut_fast_load_status = mpc401_get_3dlut_fast_load_status,
 	.populate_lut = mpc401_populate_lut,
 	.program_lut_read_write_control = mpc401_program_lut_read_write_control,
 	.program_lut_mode = mpc401_program_lut_mode,
diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h
index ce6fbcf14d7a..8e35ebc603a9 100644
--- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h
+++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h
@@ -241,23 +241,9 @@ void mpc401_update_3dlut_fast_load_select(
 	int mpcc_id,
 	int hubp_idx);
 
-void mpc401_get_3dlut_fast_load_status(
-	struct mpc *mpc,
-	int mpcc_id,
-	uint32_t *done,
-	uint32_t *soft_underflow,
-	uint32_t *hard_underflow);
-
 void mpc401_update_3dlut_fast_load_select(
 	struct mpc *mpc,
 	int mpcc_id,
 	int hubp_idx);
 
-void mpc401_get_3dlut_fast_load_status(
-	struct mpc *mpc,
-	int mpcc_id,
-	uint32_t *done,
-	uint32_t *soft_underflow,
-	uint32_t *hard_underflow);
-
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c
index b86fe2b094f8..4cfc6c0fa147 100644
--- a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c
@@ -507,6 +507,7 @@ void dcn35_timing_generator_init(struct optc *optc1)
 	optc1->min_v_blank_interlace = 5;
 	optc1->min_h_sync_width = 4;
 	optc1->min_v_sync_width = 1;
+	optc1->max_frame_count = 0xFFFFFF;
 
 	dcn35_timing_generator_set_fgcg(
 		optc1, CTX->dc->debug.enable_fine_grain_clock_gating.bits.optc);
diff --git a/drivers/gpu/drm/amd/display/dc/resource/Makefile b/drivers/gpu/drm/amd/display/dc/resource/Makefile
index b8cddef6b3d2..5b42da8b79c2 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/resource/Makefile
@@ -27,6 +27,24 @@
 #  DCE
 ###############################################################################
 
+ifdef CONFIG_DRM_AMD_DC_SI
+RESOURCE_DCE60 = dce60_resource.o
+
+AMD_DAL_RESOURCE_DCE60 = $(addprefix $(AMDDALPATH)/dc/resource/dce60/,$(RESOURCE_DCE60))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCE60)
+endif
+
+###############################################################################
+
+RESOURCE_DCE80 = dce80_resource.o
+
+AMD_DAL_RESOURCE_DCE80 = $(addprefix $(AMDDALPATH)/dc/resource/dce80/,$(RESOURCE_DCE80))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCE80)
+
+###############################################################################
+
 RESOURCE_DCE100 = dce100_resource.o
 
 AMD_DAL_RESOURCE_DCE100 = $(addprefix $(AMDDALPATH)/dc/resource/dce100/,$(RESOURCE_DCE100))
@@ -57,14 +75,6 @@ AMD_DAL_RESOURCE_DCE120 = $(addprefix $(AMDDALPATH)/dc/resource/dce120/,$(RESOUR
 
 AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCE120)
 
-###############################################################################
-
-RESOURCE_DCE80 = dce80_resource.o
-
-AMD_DAL_RESOURCE_DCE80 = $(addprefix $(AMDDALPATH)/dc/resource/dce80/,$(RESOURCE_DCE80))
-
-AMD_DISPLAY_FILES += $(AMD_DAL_RESOURCE_DCE80)
-
 ifdef CONFIG_DRM_AMD_DC_FP
 ###############################################################################
 # DCN
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
index e698543ec937..84f73fdb0f95 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
@@ -836,7 +836,7 @@ static enum dc_status build_mapped_resource(
 	return DC_OK;
 }
 
-static bool dce100_validate_bandwidth(
+static enum dc_status dce100_validate_bandwidth(
 	struct dc  *dc,
 	struct dc_state *context,
 	bool fast_validate)
@@ -858,7 +858,7 @@ static bool dce100_validate_bandwidth(
 		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
-	return true;
+	return DC_OK;
 }
 
 static bool dce100_validate_surface_sets(
@@ -1069,7 +1069,7 @@ static bool dce100_resource_construct(
 	pool->base.timing_generator_count = pool->base.res_cap->num_timing_generator;
 	dc->caps.max_downscale_ratio = 200;
 	dc->caps.i2c_speed_in_khz = 40;
-	dc->caps.i2c_speed_in_khz = 40;
+	dc->caps.i2c_speed_in_khz_hdcp = 40;
 	dc->caps.max_cursor_size = 128;
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dual_link_dvi = true;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c
index 035c6cfdaee5..f3d5baac11bf 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c
@@ -960,7 +960,7 @@ static enum dc_status build_mapped_resource(
 	return DC_OK;
 }
 
-static bool dce110_validate_bandwidth(
+static enum dc_status dce110_validate_bandwidth(
 	struct dc *dc,
 	struct dc_state *context,
 	bool fast_validate)
@@ -1031,7 +1031,7 @@ static bool dce110_validate_bandwidth(
 			context->bw_ctx.bw.dce.yclk_khz,
 			context->bw_ctx.bw.dce.blackout_recovery_time_us);
 	}
-	return result;
+	return result ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 static enum dc_status dce110_validate_plane(const struct dc_plane_state *plane_state,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
index 480a50967385..4225cae68c10 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
@@ -883,7 +883,7 @@ static enum dc_status build_mapped_resource(
 	return DC_OK;
 }
 
-bool dce112_validate_bandwidth(
+enum dc_status dce112_validate_bandwidth(
 	struct dc *dc,
 	struct dc_state *context,
 	bool fast_validate)
@@ -952,7 +952,7 @@ bool dce112_validate_bandwidth(
 			context->bw_ctx.bw.dce.yclk_khz,
 			context->bw_ctx.bw.dce.blackout_recovery_time_us);
 	}
-	return result;
+	return result ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 enum dc_status resource_map_phy_clock_resources(
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h
index 1f57ebc6f9b4..6221d749246d 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h
@@ -42,7 +42,7 @@ enum dc_status dce112_validate_with_context(
 		struct dc_state *context,
 		struct dc_state *old_context);
 
-bool dce112_validate_bandwidth(
+enum dc_status dce112_validate_bandwidth(
 	struct dc *dc,
 	struct dc_state *context,
 	bool fast_validate);
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
index 889f314cac65..d9ffdded5ce1 100644
--- a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
@@ -48,7 +48,7 @@
 #include "dce/dce_clock_source.h"
 #include "dce/dce_audio.h"
 #include "dce/dce_hwseq.h"
-#include "dce60/dce60_hw_sequencer.h"
+#include "dce60/dce60_hwseq.h"
 #include "dce100/dce100_resource.h"
 #include "dce/dce_panel_cntl.h"
 
@@ -863,7 +863,7 @@ static void dce60_resource_destruct(struct dce110_resource_pool *pool)
 	}
 }
 
-static bool dce60_validate_bandwidth(
+static enum dc_status dce60_validate_bandwidth(
 	struct dc *dc,
 	struct dc_state *context,
 	bool fast_validate)
@@ -885,7 +885,7 @@ static bool dce60_validate_bandwidth(
 		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
-	return true;
+	return DC_OK;
 }
 
 static bool dce60_validate_surface_sets(
diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.h
index 5d653a76b0b0..5d653a76b0b0 100644
--- a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.h
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
index 3d5113f010bb..bd5811f97531 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
@@ -869,7 +869,7 @@ static void dce80_resource_destruct(struct dce110_resource_pool *pool)
 	}
 }
 
-static bool dce80_validate_bandwidth(
+static enum dc_status dce80_validate_bandwidth(
 	struct dc *dc,
 	struct dc_state *context,
 	bool fast_validate)
@@ -891,7 +891,7 @@ static bool dce80_validate_bandwidth(
 		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
-	return true;
+	return DC_OK;
 }
 
 static bool dce80_validate_surface_sets(
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c
index e92f14d50adb..be4ade0853e9 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c
@@ -23,6 +23,7 @@
  *
  */
 
+#include "core_status.h"
 #include "dm_services.h"
 #include "dc.h"
 
@@ -1125,7 +1126,7 @@ static void dcn10_destroy_resource_pool(struct resource_pool **pool)
 	*pool = NULL;
 }
 
-static bool dcn10_validate_bandwidth(
+static enum dc_status dcn10_validate_bandwidth(
 		struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
@@ -1136,7 +1137,7 @@ static bool dcn10_validate_bandwidth(
 	voltage_supported = dcn_validate_bandwidth(dc, context, fast_validate);
 	DC_FP_END();
 
-	return voltage_supported;
+	return voltage_supported ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 static enum dc_status dcn10_validate_plane(const struct dc_plane_state *plane_state, struct dc_caps *caps)
@@ -1245,6 +1246,10 @@ struct stream_encoder *dcn10_find_first_free_match_stream_enc_for_link(
 			if (link->ep_type == DISPLAY_ENDPOINT_PHY && pool->stream_enc[i]->id ==
 					link->link_enc->preferred_engine)
 				return pool->stream_enc[i];
+
+			if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && pool->stream_enc[i]->id ==
+					link->dpia_preferred_eng_id)
+				return pool->stream_enc[i];
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
index e4eca3e32c1b..3405be07f5e3 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
@@ -2124,7 +2124,7 @@ validate_out:
 	return out;
 }
 
-bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context,
+enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context,
 		bool fast_validate)
 {
 	bool voltage_supported;
@@ -2132,14 +2132,14 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context,
 
 	pipes = kcalloc(dc->res_pool->pipe_count, sizeof(display_e2e_pipe_params_st), GFP_KERNEL);
 	if (!pipes)
-		return false;
+		return DC_FAIL_BANDWIDTH_VALIDATE;
 
 	DC_FP_START();
 	voltage_supported = dcn20_validate_bandwidth_fp(dc, context, fast_validate, pipes);
 	DC_FP_END();
 
 	kfree(pipes);
-	return voltage_supported;
+	return voltage_supported ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 struct pipe_ctx *dcn20_acquire_free_pipe_for_layer(
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h
index 4cee3fa11a7f..c0e062c7407d 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h
@@ -119,7 +119,7 @@ void dcn20_set_mcif_arb_params(
 		struct dc_state *context,
 		display_e2e_pipe_params_st *pipes,
 		int pipe_cnt);
-bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate);
+enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate);
 void dcn20_merge_pipes_for_validate(
 		struct dc *dc,
 		struct dc_state *context);
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c
index 4bd5c2278596..9ab01b65b177 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c
@@ -923,7 +923,7 @@ validate_out:
  * with DC_FP_START()/DC_FP_END(). Use the same approach as for
  * dcn20_validate_bandwidth in dcn20_resource.c.
  */
-static bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context,
+static enum dc_status dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context,
 		bool fast_validate)
 {
 	bool voltage_supported;
@@ -931,14 +931,14 @@ static bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context,
 
 	pipes = kcalloc(dc->res_pool->pipe_count, sizeof(display_e2e_pipe_params_st), GFP_KERNEL);
 	if (!pipes)
-		return false;
+		return DC_FAIL_BANDWIDTH_VALIDATE;
 
 	DC_FP_START();
 	voltage_supported = dcn21_validate_bandwidth_fp(dc, context, fast_validate, pipes);
 	DC_FP_END();
 
 	kfree(pipes);
-	return voltage_supported;
+	return voltage_supported ? DC_OK : DC_NOT_SUPPORTED;
 }
 
 static void dcn21_destroy_resource_pool(struct resource_pool **pool)
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
index f01ced015072..f631ae34e320 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
@@ -1891,8 +1891,6 @@ static int get_refresh_rate(struct dc_state *context)
 
 	/* check if refresh rate at least 120hz */
 	timing = &context->streams[0]->timing;
-	if (timing == NULL)
-		return 0;
 
 	h_v_total = timing->h_total * timing->v_total;
 	if (h_v_total == 0)
@@ -2037,7 +2035,7 @@ void dcn30_calculate_wm_and_dlg(
 	DC_FP_END();
 }
 
-bool dcn30_validate_bandwidth(struct dc *dc,
+enum dc_status dcn30_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
@@ -2094,7 +2092,7 @@ validate_out:
 
 	BW_VAL_TRACE_FINISH();
 
-	return out;
+	return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h
index 8e6b8b7368fd..689d9bdace81 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h
@@ -56,7 +56,7 @@ unsigned int dcn30_calc_max_scaled_time(
 		enum mmhubbub_wbif_mode mode,
 		unsigned int urgent_watermark);
 
-bool dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context,
+enum dc_status dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context,
 		bool fast_validate);
 bool dcn30_internal_validate_bw(
 		struct dc *dc,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
index dddddbfef85f..7e0af5297dc4 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
@@ -1758,7 +1758,7 @@ dcn31_set_mcif_arb_params(struct dc *dc,
 	DC_FP_END();
 }
 
-bool dcn31_validate_bandwidth(struct dc *dc,
+enum dc_status dcn31_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
@@ -1813,7 +1813,7 @@ validate_out:
 
 	BW_VAL_TRACE_FINISH();
 
-	return out;
+	return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 static void dcn31_get_panel_config_defaults(struct dc_panel_config *panel_config)
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
index 551ad912f7be..dd82815d7efe 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
@@ -37,7 +37,7 @@ struct dcn31_resource_pool {
 	struct resource_pool base;
 };
 
-bool dcn31_validate_bandwidth(struct dc *dc,
+enum dc_status dcn31_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate);
 void dcn31_calculate_wm_and_dlg(
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
index 26becc4cb804..d96bc6cb73ad 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
@@ -1694,7 +1694,7 @@ static void dcn314_get_panel_config_defaults(struct dc_panel_config *panel_confi
 	*panel_config = panel_config_defaults;
 }
 
-bool dcn314_validate_bandwidth(struct dc *dc,
+enum dc_status dcn314_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
@@ -1750,7 +1750,7 @@ validate_out:
 
 	BW_VAL_TRACE_FINISH();
 
-	return out;
+	return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 static struct resource_funcs dcn314_res_pool_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h
index 49ffe71018df..f8ba531d6342 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h
@@ -39,7 +39,7 @@ struct dcn314_resource_pool {
 	struct resource_pool base;
 };
 
-bool dcn314_validate_bandwidth(struct dc *dc,
+enum dc_status dcn314_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate);
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
index 2a59cc61ed8c..bb0dae0be5b8 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
@@ -24,6 +24,7 @@
  *
  */
 
+#include "dc_types.h"
 #include "dm_services.h"
 #include "dc.h"
 
@@ -1806,19 +1807,56 @@ validate_out:
 	return out;
 }
 
-bool dcn32_validate_bandwidth(struct dc *dc,
+enum dc_status dcn32_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
-	bool out = false;
+	unsigned int i;
+	enum dc_status status;
+	const struct dc_stream_state *stream;
+
+	/* reset cursor limitations on subvp */
+	for (i = 0; i < context->stream_count; i++) {
+		stream = context->streams[i];
+
+		if (dc_state_can_clear_stream_cursor_subvp_limit(stream, context)) {
+			dc_state_set_stream_cursor_subvp_limit(stream, context, false);
+		}
+	}
 
 	if (dc->debug.using_dml2)
-		out = dml2_validate(dc, context,
+		status = dml2_validate(dc, context,
 				context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
-				fast_validate);
+				fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 	else
-		out = dml1_validate(dc, context, fast_validate);
-	return out;
+		status = dml1_validate(dc, context, fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
+
+	if (!fast_validate && status == DC_OK && dc_state_is_subvp_in_use(context)) {
+		/* check new stream configuration still supports cursor if subvp used */
+		for (i = 0; i < context->stream_count; i++) {
+			stream = context->streams[i];
+
+			if (dc_state_get_stream_subvp_type(context, stream) != SUBVP_PHANTOM &&
+					stream->cursor_position.enable &&
+					!dc_stream_check_cursor_attributes(stream, context, &stream->cursor_attributes)) {
+				/* hw cursor cannot be supported with subvp active, so disable subvp for now */
+				dc_state_set_stream_cursor_subvp_limit(stream, context, true);
+				status = DC_FAIL_HW_CURSOR_SUPPORT;
+			}
+		};
+	}
+
+	if (!fast_validate && status == DC_FAIL_HW_CURSOR_SUPPORT) {
+		/* attempt to validate again with subvp disabled due to cursor */
+		if (dc->debug.using_dml2)
+			status = dml2_validate(dc, context,
+					context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
+					fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
+		else
+			status = dml1_validate(dc, context, fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
+	}
+
+	return status;
 }
 
 int dcn32_populate_dml_pipes_from_context(
@@ -2042,6 +2080,18 @@ static void dcn32_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw
 	DC_FP_END();
 }
 
+unsigned int dcn32_get_max_hw_cursor_size(const struct dc *dc,
+			struct dc_state *state,
+			const struct dc_stream_state *stream)
+{
+	bool limit_cur_to_buf;
+
+	limit_cur_to_buf = dc_state_get_stream_subvp_cursor_limit(stream, state) &&
+			!stream->hw_cursor_req;
+
+	return limit_cur_to_buf ? dc->caps.max_buffered_cursor_size : dc->caps.max_cursor_size;
+}
+
 static struct resource_funcs dcn32_res_pool_funcs = {
 	.destroy = dcn32_destroy_resource_pool,
 	.link_enc_create = dcn32_link_encoder_create,
@@ -2067,7 +2117,8 @@ static struct resource_funcs dcn32_res_pool_funcs = {
 	.add_phantom_pipes = dcn32_add_phantom_pipes,
 	.build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params,
 	.calculate_mall_ways_from_bytes = dcn32_calculate_mall_ways_from_bytes,
-	.get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe
+	.get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe,
+	.get_max_hw_cursor_size = dcn32_get_max_hw_cursor_size,
 };
 
 static uint32_t read_pipe_fuses(struct dc_context *ctx)
@@ -2114,8 +2165,6 @@ static bool dcn32_resource_construct(
 #define REG_STRUCT dccg_regs
 	dccg_regs_init();
 
-	DC_FP_START();
-
 	ctx->dc_bios->regs = &bios_regs;
 
 	pool->base.res_cap = &res_cap_dcn32;
@@ -2153,6 +2202,7 @@ static bool dcn32_resource_construct(
 	dc->caps.i2c_speed_in_khz_hdcp = 100; /*1.4 w/a applied by default*/
 	/* TODO: Bring max_cursor_size back to 256 after subvp cursor corruption is fixed*/
 	dc->caps.max_cursor_size = 64;
+	dc->caps.max_buffered_cursor_size = 64; // sqrt(16 * 1024 / 4)
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dmdata_alloc_size = 2048;
 	dc->caps.mall_size_per_mem_channel = 4;
@@ -2501,14 +2551,10 @@ static bool dcn32_resource_construct(
 	if (ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev) && (dc->config.sdpif_request_limit_words_per_umc == 0))
 		dc->config.sdpif_request_limit_words_per_umc = 16;
 
-	DC_FP_END();
-
 	return true;
 
 create_fail:
 
-	DC_FP_END();
-
 	dcn32_resource_destruct(pool);
 
 	return false;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
index 1aa4ced29291..d60ed77eda80 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
@@ -98,7 +98,7 @@ void dcn32_add_phantom_pipes(struct dc *dc,
 		unsigned int pipe_cnt,
 		unsigned int index);
 
-bool dcn32_validate_bandwidth(struct dc *dc,
+enum dc_status dcn32_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate);
 
@@ -188,6 +188,10 @@ void dcn32_override_min_req_dcfclk(struct dc *dc, struct dc_state *context);
 
 unsigned int dcn32_calculate_mall_ways_from_bytes(const struct dc *dc, unsigned int total_size_in_mall_bytes);
 
+unsigned int dcn32_get_max_hw_cursor_size(const struct dc *dc,
+			struct dc_state *state,
+			const struct dc_stream_state *stream);
+
 /* definitions for run time init of reg offsets */
 
 /* CLK SRC */
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
index 38d76434683e..7db1f7a5613f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
@@ -1624,7 +1624,8 @@ static struct resource_funcs dcn321_res_pool_funcs = {
 	.add_phantom_pipes = dcn32_add_phantom_pipes,
 	.build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params,
 	.calculate_mall_ways_from_bytes = dcn32_calculate_mall_ways_from_bytes,
-	.get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe
+	.get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe,
+	.get_max_hw_cursor_size = dcn32_get_max_hw_cursor_size,
 };
 
 static uint32_t read_pipe_fuses(struct dc_context *ctx)
@@ -1709,6 +1710,7 @@ static bool dcn321_resource_construct(
 	dc->caps.i2c_speed_in_khz_hdcp = 100; /*1.4 w/a applied by default*/
 	/* TODO: Bring max cursor size back to 256 after subvp cursor corruption is fixed*/
 	dc->caps.max_cursor_size = 64;
+	dc->caps.max_buffered_cursor_size = 64; // sqrt(16 * 1024 / 4)
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dmdata_alloc_size = 2048;
 	dc->caps.mall_size_per_mem_channel = 4;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
index ffd2b816cd02..72c6cf047db0 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
@@ -1732,7 +1732,7 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config
 }
 
 
-static bool dcn35_validate_bandwidth(struct dc *dc,
+static enum dc_status dcn35_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
@@ -1743,13 +1743,13 @@ static bool dcn35_validate_bandwidth(struct dc *dc,
 			fast_validate);
 
 	if (fast_validate)
-		return out;
+		return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 
 	DC_FP_START();
 	dcn35_decide_zstate_support(dc, context);
 	DC_FP_END();
 
-	return out;
+	return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 enum dc_status dcn35_patch_unknown_plane_state(struct dc_plane_state *plane_state)
@@ -1903,7 +1903,7 @@ static bool dcn35_resource_construct(
 	dc->caps.max_disp_clock_khz_at_vmin = 650000;
 
 	/* Sequential ONO is based on ASIC. */
-	if (dc->ctx->asic_id.hw_internal_rev > 0x10)
+	if (dc->ctx->asic_id.hw_internal_rev >= 0x40)
 		dc->caps.sequential_ono = true;
 
 	/* Use pipe context based otg sync logic */
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
index 98f5bc1b929e..989a270f7dea 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
@@ -1712,7 +1712,7 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config
 }
 
 
-static bool dcn351_validate_bandwidth(struct dc *dc,
+static enum dc_status dcn351_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
@@ -1723,13 +1723,13 @@ static bool dcn351_validate_bandwidth(struct dc *dc,
 			fast_validate);
 
 	if (fast_validate)
-		return out;
+		return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 
 	DC_FP_START();
 	dcn35_decide_zstate_support(dc, context);
 	DC_FP_END();
 
-	return out;
+	return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 static struct resource_funcs dcn351_res_pool_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
index b6468573dc33..48e1f234185f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
@@ -1713,7 +1713,7 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config
 }
 
 
-static bool dcn35_validate_bandwidth(struct dc *dc,
+static enum dc_status dcn35_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
@@ -1724,13 +1724,13 @@ static bool dcn35_validate_bandwidth(struct dc *dc,
 			fast_validate);
 
 	if (fast_validate)
-		return out;
+		return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 
 	DC_FP_START();
 	dcn35_decide_zstate_support(dc, context);
 	DC_FP_END();
 
-	return out;
+	return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
 }
 
 
@@ -1876,7 +1876,7 @@ static bool dcn36_resource_construct(
 	dc->caps.max_disp_clock_khz_at_vmin = 650000;
 
 	/* Sequential ONO is based on ASIC. */
-	if (dc->ctx->asic_id.hw_internal_rev > 0x10)
+	if (dc->ctx->asic_id.hw_internal_rev >= 0x40)
 		dc->caps.sequential_ono = true;
 
 	/* Use pipe context based otg sync logic */
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
index 7436dfbdf927..e0e32975ca34 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
@@ -1642,16 +1642,52 @@ enum dc_status dcn401_patch_unknown_plane_state(struct dc_plane_state *plane_sta
 	return DC_OK;
 }
 
-bool dcn401_validate_bandwidth(struct dc *dc,
+enum dc_status dcn401_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
 {
-	bool out = false;
+	unsigned int i;
+	enum dc_status status = DC_OK;
+	const struct dc_stream_state *stream;
+
+	/* reset cursor limitations on subvp */
+	for (i = 0; i < context->stream_count; i++) {
+		stream = context->streams[i];
+
+		if (dc_state_can_clear_stream_cursor_subvp_limit(stream, context)) {
+			dc_state_set_stream_cursor_subvp_limit(stream, context, false);
+		}
+	}
+
 	if (dc->debug.using_dml2)
-		out = dml2_validate(dc, context,
+		status = dml2_validate(dc, context,
 				context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
-				fast_validate);
-	return out;
+				fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
+
+	if (!fast_validate && status == DC_OK && dc_state_is_subvp_in_use(context)) {
+		/* check new stream configuration still supports cursor if subvp used */
+		for (i = 0; i < context->stream_count; i++) {
+			stream = context->streams[i];
+
+			if (dc_state_get_stream_subvp_type(context, stream) != SUBVP_PHANTOM &&
+					stream->cursor_position.enable &&
+					!dc_stream_check_cursor_attributes(stream, context, &stream->cursor_attributes))	{
+				/* hw cursor cannot be supported with subvp active, so disable subvp for now */
+				dc_state_set_stream_cursor_subvp_limit(stream, context, true);
+				status = DC_FAIL_HW_CURSOR_SUPPORT;
+			}
+		};
+	}
+
+	if (!fast_validate && status == DC_FAIL_HW_CURSOR_SUPPORT) {
+		/* attempt to validate again with subvp disabled due to cursor */
+		if (dc->debug.using_dml2)
+			status = dml2_validate(dc, context,
+					context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2,
+					fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE;
+	}
+
+	return status;
 }
 
 void dcn401_prepare_mcache_programming(struct dc *dc,
@@ -1770,7 +1806,8 @@ static struct resource_funcs dcn401_res_pool_funcs = {
 	.build_pipe_pix_clk_params = dcn401_build_pipe_pix_clk_params,
 	.calculate_mall_ways_from_bytes = dcn32_calculate_mall_ways_from_bytes,
 	.get_power_profile = dcn401_get_power_profile,
-	.get_vstartup_for_pipe = dcn401_get_vstartup_for_pipe
+	.get_vstartup_for_pipe = dcn401_get_vstartup_for_pipe,
+	.get_max_hw_cursor_size = dcn32_get_max_hw_cursor_size
 };
 
 static uint32_t read_pipe_fuses(struct dc_context *ctx)
@@ -1846,8 +1883,9 @@ static bool dcn401_resource_construct(
 	dc->caps.max_downscale_ratio = 600;
 	dc->caps.i2c_speed_in_khz = 95;
 	dc->caps.i2c_speed_in_khz_hdcp = 95; /*1.4 w/a applied by default*/
-	/* TODO: Bring max cursor size back to 256 after subvp cursor corruption is fixed*/
+	/* used to set cursor pitch, so must be aligned to power of 2 (HW actually supported 78x78) */
 	dc->caps.max_cursor_size = 64;
+	dc->caps.max_buffered_cursor_size = 64;
 	dc->caps.cursor_not_scaled = true;
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dmdata_alloc_size = 2048;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h
index 4c259745d519..dc52a30991af 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h
@@ -22,7 +22,7 @@ struct resource_pool *dcn401_create_resource_pool(
 
 enum dc_status dcn401_patch_unknown_plane_state(struct dc_plane_state *plane_state);
 
-bool dcn401_validate_bandwidth(struct dc *dc,
+enum dc_status dcn401_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate);
 
diff --git a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c
index 28348734d900..e0008c5f08ad 100644
--- a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c
+++ b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c
@@ -776,7 +776,7 @@ static enum scl_mode spl_get_dscl_mode(const struct spl_in *spl_in,
 	 * Do not bypass UV at 1:1 for cositing to be applied
 	 */
 	if (!enable_isharp) {
-		if (data->ratios.horz.value == one && data->ratios.vert.value == one)
+		if (data->ratios.horz.value == one && data->ratios.vert.value == one && !spl_in->basic_out.always_scale)
 			return SCL_MODE_SCALING_420_LUMA_BYPASS;
 	}
 
@@ -884,7 +884,7 @@ static bool spl_get_isharp_en(struct spl_in *spl_in,
 
 /* Calculate number of tap with adaptive scaling off */
 static void spl_get_taps_non_adaptive_scaler(
-	  struct spl_scratch *spl_scratch, const struct spl_taps *in_taps)
+	  struct spl_scratch *spl_scratch, const struct spl_taps *in_taps, bool always_scale)
 {
 	bool check_max_downscale = false;
 
@@ -944,15 +944,15 @@ static void spl_get_taps_non_adaptive_scaler(
 		spl_fixpt_from_fraction(6, 1));
 	SPL_ASSERT(check_max_downscale);
 
-	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz))
+
+	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz) && !always_scale)
 		spl_scratch->scl_data.taps.h_taps = 1;
-	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert))
+	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert) && !always_scale)
 		spl_scratch->scl_data.taps.v_taps = 1;
-	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c))
+	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !always_scale)
 		spl_scratch->scl_data.taps.h_taps_c = 1;
-	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c))
+	if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c) && !always_scale)
 		spl_scratch->scl_data.taps.v_taps_c = 1;
-
 }
 
 /* Calculate optimal number of taps */
@@ -965,13 +965,15 @@ static bool spl_get_optimal_number_of_taps(
 	unsigned int max_taps_y, max_taps_c;
 	unsigned int min_taps_y, min_taps_c;
 	enum lb_memory_config lb_config;
-	bool skip_easf = false;
+	bool skip_easf     = false;
+	bool always_scale  = spl_in->basic_out.always_scale;
 	bool is_subsampled = spl_is_subsampled_format(spl_in->basic_in.format);
 
+
 	if (spl_scratch->scl_data.viewport.width > spl_scratch->scl_data.h_active &&
 		max_downscale_src_width != 0 &&
 		spl_scratch->scl_data.viewport.width > max_downscale_src_width) {
-		spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps);
+		spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale);
 		*enable_easf_v = false;
 		*enable_easf_h = false;
 		*enable_isharp = false;
@@ -980,7 +982,7 @@ static bool spl_get_optimal_number_of_taps(
 
 	/* Disable adaptive scaler and sharpener when integer scaling is enabled */
 	if (spl_in->scaling_quality.integer_scaling) {
-		spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps);
+		spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale);
 		*enable_easf_v = false;
 		*enable_easf_h = false;
 		*enable_isharp = false;
@@ -996,7 +998,7 @@ static bool spl_get_optimal_number_of_taps(
 	 * taps = 4 for upscaling
 	 */
 	if (skip_easf)
-		spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps);
+		spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale);
 	else {
 		if (spl_is_video_format(spl_in->basic_in.format)) {
 			spl_scratch->scl_data.taps.h_taps = 6;
@@ -1297,7 +1299,7 @@ static void spl_set_easf_data(struct spl_scratch *spl_scratch, struct spl_out *s
 	if (enable_easf_v) {
 		dscl_prog_data->easf_v_en = true;
 		dscl_prog_data->easf_v_ring = 0;
-		dscl_prog_data->easf_v_sharp_factor = 0;
+		dscl_prog_data->easf_v_sharp_factor = 1;
 		dscl_prog_data->easf_v_bf1_en = 1;	// 1-bit, BF1 calculation enable, 0=disable, 1=enable
 		dscl_prog_data->easf_v_bf2_mode = 0xF;	// 4-bit, BF2 calculation mode
 		/* 2-bit, BF3 chroma mode correction calculation mode */
@@ -1461,7 +1463,7 @@ static void spl_set_easf_data(struct spl_scratch *spl_scratch, struct spl_out *s
 	if (enable_easf_h) {
 		dscl_prog_data->easf_h_en = true;
 		dscl_prog_data->easf_h_ring = 0;
-		dscl_prog_data->easf_h_sharp_factor = 0;
+		dscl_prog_data->easf_h_sharp_factor = 1;
 		dscl_prog_data->easf_h_bf1_en =
 			1;	// 1-bit, BF1 calculation enable, 0=disable, 1=enable
 		dscl_prog_data->easf_h_bf2_mode =
@@ -1898,3 +1900,4 @@ bool SPL_NAMESPACE(spl_get_number_of_taps(struct spl_in *spl_in, struct spl_out
 	spl_set_taps_data(dscl_prog_data, data);
 	return res;
 }
+
diff --git a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h
index 1c3949b24611..36a284305a70 100644
--- a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h
+++ b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h
@@ -480,6 +480,10 @@ enum sharpness_setting	{
 	SHARPNESS_ZERO,
 	SHARPNESS_CUSTOM
 };
+enum sharpness_range_source	{
+	SHARPNESS_RANGE_DCN = 0,
+	SHARPNESS_RANGE_DCN_OVERRIDE
+};
 struct spl_sharpness_range {
 	int sdr_rgb_min;
 	int sdr_rgb_max;
diff --git a/drivers/gpu/drm/amd/display/dc/sspl/spl_fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/sspl/spl_fixpt31_32.c
index 52d97918a3bd..ebf0287417e0 100644
--- a/drivers/gpu/drm/amd/display/dc/sspl/spl_fixpt31_32.c
+++ b/drivers/gpu/drm/amd/display/dc/sspl/spl_fixpt31_32.c
@@ -29,8 +29,6 @@ static inline unsigned long long spl_complete_integer_division_u64(
 {
 	unsigned long long result;
 
-	SPL_ASSERT(divisor);
-
 	result = spl_div64_u64_rem(dividend, divisor, remainder);
 
 	return result;
@@ -196,8 +194,6 @@ struct spl_fixed31_32 spl_fixpt_recip(struct spl_fixed31_32 arg)
 	 * Good idea to use Newton's method
 	 */
 
-	SPL_ASSERT(arg.value);
-
 	return spl_fixpt_from_fraction(
 		spl_fixpt_one.value,
 		arg.value);
diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 4e0efff92dca..3f3fa1b6a69e 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -51,8 +51,8 @@
  * for the cache windows.
  *
  * The call to dmub_srv_hw_init() programs the DMCUB registers to prepare
- * for command submission. Commands can be queued via dmub_srv_cmd_queue()
- * and executed via dmub_srv_cmd_execute().
+ * for command submission. Commands can be queued via dmub_srv_fb_cmd_queue()
+ * and executed via dmub_srv_fb_cmd_execute().
  *
  * If the queue is full the dmub_srv_wait_for_idle() call can be used to
  * wait until the queue has been cleared.
@@ -142,6 +142,7 @@ enum dmub_notification_type {
 	DMUB_NOTIFICATION_SET_CONFIG_REPLY,
 	DMUB_NOTIFICATION_DPIA_NOTIFICATION,
 	DMUB_NOTIFICATION_HPD_SENSE_NOTIFY,
+	DMUB_NOTIFICATION_FUSED_IO,
 	DMUB_NOTIFICATION_MAX
 };
 
@@ -170,6 +171,13 @@ enum dmub_srv_power_state_type {
 	DMUB_POWER_STATE_D3 = 8
 };
 
+/* enum dmub_inbox_cmd_interface type - defines default interface for host->dmub commands */
+enum dmub_inbox_cmd_interface_type {
+	DMUB_CMD_INTERFACE_DEFAULT = 0,
+	DMUB_CMD_INTERFACE_FB = 1,
+	DMUB_CMD_INTERFACE_REG = 2,
+};
+
 /**
  * struct dmub_region - dmub hw memory region
  * @base: base address for region, must be 256 byte aligned
@@ -349,6 +357,21 @@ struct dmub_diagnostic_data {
 	uint8_t is_cw6_enabled : 1;
 };
 
+struct dmub_srv_inbox {
+	/* generic status */
+	uint64_t num_submitted;
+	uint64_t num_reported;
+	union {
+		/* frame buffer mailbox status */
+		struct dmub_rb rb;
+		/* register mailbox status */
+		struct {
+			bool is_pending;
+			bool is_multi_pending;
+		};
+	};
+};
+
 /**
  * struct dmub_srv_base_funcs - Driver specific base callbacks
  */
@@ -422,6 +445,8 @@ struct dmub_srv_hw_funcs {
 
 	uint32_t (*emul_get_inbox1_rptr)(struct dmub_srv *dmub);
 
+	uint32_t (*emul_get_inbox1_wptr)(struct dmub_srv *dmub);
+
 	void (*emul_set_inbox1_wptr)(struct dmub_srv *dmub, uint32_t wptr_offset);
 
 	bool (*is_supported)(struct dmub_srv *dmub);
@@ -462,18 +487,21 @@ struct dmub_srv_hw_funcs {
 	void (*init_reg_offsets)(struct dmub_srv *dmub, struct dc_context *ctx);
 
 	void (*subvp_save_surf_addr)(struct dmub_srv *dmub, const struct dc_plane_address *addr, uint8_t subvp_index);
+
 	void (*send_reg_inbox0_cmd_msg)(struct dmub_srv *dmub,
 			union dmub_rb_cmd *cmd);
 	uint32_t (*read_reg_inbox0_rsp_int_status)(struct dmub_srv *dmub);
 	void (*read_reg_inbox0_cmd_rsp)(struct dmub_srv *dmub,
 			union dmub_rb_cmd *cmd);
 	void (*write_reg_inbox0_rsp_int_ack)(struct dmub_srv *dmub);
+	void (*clear_reg_inbox0_rsp_int_ack)(struct dmub_srv *dmub);
+	void (*enable_reg_inbox0_rsp_int)(struct dmub_srv *dmub, bool enable);
+
 	uint32_t (*read_reg_outbox0_rdy_int_status)(struct dmub_srv *dmub);
 	void (*write_reg_outbox0_rdy_int_ack)(struct dmub_srv *dmub);
 	void (*read_reg_outbox0_msg)(struct dmub_srv *dmub, uint32_t *msg);
 	void (*write_reg_outbox0_rsp)(struct dmub_srv *dmub, uint32_t *rsp);
 	uint32_t (*read_reg_outbox0_rsp_int_status)(struct dmub_srv *dmub);
-	void (*enable_reg_inbox0_rsp_int)(struct dmub_srv *dmub, bool enable);
 	void (*enable_reg_outbox0_rdy_int)(struct dmub_srv *dmub, bool enable);
 };
 
@@ -493,6 +521,7 @@ struct dmub_srv_create_params {
 	enum dmub_asic asic;
 	uint32_t fw_version;
 	bool is_virtual;
+	enum dmub_inbox_cmd_interface_type inbox_type;
 };
 
 /**
@@ -521,8 +550,9 @@ struct dmub_srv {
 	const struct dmub_srv_dcn401_regs *regs_dcn401;
 	struct dmub_srv_base_funcs funcs;
 	struct dmub_srv_hw_funcs hw_funcs;
-	struct dmub_rb inbox1_rb;
+	struct dmub_srv_inbox inbox1;
 	uint32_t inbox1_last_wptr;
+	struct dmub_srv_inbox reg_inbox0;
 	/**
 	 * outbox1_rb is accessed without locks (dal & dc)
 	 * and to be used only in dmub_srv_stat_get_notification()
@@ -542,6 +572,7 @@ struct dmub_srv {
 	struct dmub_fw_meta_info meta_info;
 	struct dmub_feature_caps feature_caps;
 	struct dmub_visual_confirm_color visual_confirm_color;
+	enum dmub_inbox_cmd_interface_type inbox_type;
 
 	enum dmub_srv_power_state_type power_state;
 	struct dmub_diagnostic_data debug;
@@ -566,11 +597,8 @@ struct dmub_notification {
 		struct aux_reply_data aux_reply;
 		enum dp_hpd_status hpd_status;
 		enum set_config_status sc_status;
-		/**
-		 * DPIA notification command.
-		 */
-		struct dmub_rb_cmd_dpia_notification dpia_notification;
 		struct dmub_rb_cmd_hpd_sense_notify_data hpd_sense_notify;
+		struct dmub_cmd_fused_request fused_request;
 	};
 };
 
@@ -699,19 +727,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
 enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub);
 
 /**
- * dmub_srv_sync_inbox1() - sync sw state with hw state
- * @dmub: the dmub service
- *
- * Sync sw state with hw state when resume from S0i3
- *
- * Return:
- *   DMUB_STATUS_OK - success
- *   DMUB_STATUS_INVALID - unspecified error
- */
-enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub);
-
-/**
- * dmub_srv_cmd_queue() - queues a command to the DMUB
+ * dmub_srv_fb_cmd_queue() - queues a command to the DMUB
  * @dmub: the dmub service
  * @cmd: the command to queue
  *
@@ -723,11 +739,11 @@ enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub);
  *   DMUB_STATUS_QUEUE_FULL - no remaining room in queue
  *   DMUB_STATUS_INVALID - unspecified error
  */
-enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub,
+enum dmub_status dmub_srv_fb_cmd_queue(struct dmub_srv *dmub,
 				    const union dmub_rb_cmd *cmd);
 
 /**
- * dmub_srv_cmd_execute() - Executes a queued sequence to the dmub
+ * dmub_srv_fb_cmd_execute() - Executes a queued sequence to the dmub
  * @dmub: the dmub service
  *
  * Begins execution of queued commands on the dmub.
@@ -736,7 +752,7 @@ enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub,
  *   DMUB_STATUS_OK - success
  *   DMUB_STATUS_INVALID - unspecified error
  */
-enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub);
+enum dmub_status dmub_srv_fb_cmd_execute(struct dmub_srv *dmub);
 
 /**
  * dmub_srv_wait_for_hw_pwr_up() - Waits for firmware hardware power up is completed
@@ -795,6 +811,23 @@ enum dmub_status dmub_srv_wait_for_phy_init(struct dmub_srv *dmub,
 					    uint32_t timeout_us);
 
 /**
+ * dmub_srv_wait_for_pending() - Re-entrant wait for messages currently pending
+ * @dmub: the dmub service
+ * @timeout_us: the maximum number of microseconds to wait
+ *
+ * Waits until the commands queued prior to this call are complete.
+ * If interfaces remain busy due to additional work being submitted
+ * concurrently, this function will not continue to wait.
+ *
+ * Return:
+ *   DMUB_STATUS_OK - success
+ *   DMUB_STATUS_TIMEOUT - wait for buffer to flush timed out
+ *   DMUB_STATUS_INVALID - unspecified error
+ */
+enum dmub_status dmub_srv_wait_for_pending(struct dmub_srv *dmub,
+					uint32_t timeout_us);
+
+/**
  * dmub_srv_wait_for_idle() - Waits for the DMUB to be idle
  * @dmub: the dmub service
  * @timeout_us: the maximum number of microseconds to wait
@@ -892,9 +925,6 @@ enum dmub_status dmub_srv_get_fw_boot_status(struct dmub_srv *dmub,
 enum dmub_status dmub_srv_get_fw_boot_option(struct dmub_srv *dmub,
 					     union dmub_fw_boot_options *option);
 
-enum dmub_status dmub_srv_cmd_with_reply_data(struct dmub_srv *dmub,
-					      union dmub_rb_cmd *cmd);
-
 enum dmub_status dmub_srv_set_skip_panel_power_sequence(struct dmub_srv *dmub,
 					     bool skip);
 
@@ -959,26 +989,6 @@ enum dmub_status dmub_srv_clear_inbox0_ack(struct dmub_srv *dmub);
 void dmub_srv_subvp_save_surf_addr(struct dmub_srv *dmub, const struct dc_plane_address *addr, uint8_t subvp_index);
 
 /**
- * dmub_srv_send_reg_inbox0_cmd() - send a dmub command and wait for the command
- * being processed by DMUB.
- * @dmub: The dmub service
- * @cmd: The dmub command being sent. If with_replay is true, the function will
- * update cmd with replied data.
- * @with_reply: true if DMUB reply needs to be copied back to cmd. false if the
- * cmd doesn't need to be replied.
- * @timeout_us: timeout in microseconds.
- *
- * Return:
- * DMUB_STATUS_OK - success
- * DMUB_STATUS_TIMEOUT - DMUB fails to process the command within the timeout
- * interval.
- */
-enum dmub_status dmub_srv_send_reg_inbox0_cmd(
-		struct dmub_srv *dmub,
-		union dmub_rb_cmd *cmd,
-		bool with_reply, uint32_t timeout_us);
-
-/**
  * dmub_srv_set_power_state() - Track DC power state in dmub_srv
  * @dmub: The dmub service
  * @power_state: DC power state setting
@@ -990,4 +1000,71 @@ enum dmub_status dmub_srv_send_reg_inbox0_cmd(
  */
 void dmub_srv_set_power_state(struct dmub_srv *dmub, enum dmub_srv_power_state_type dmub_srv_power_state);
 
+/**
+ * dmub_srv_reg_cmd_execute() - Executes provided command to the dmub
+ * @dmub: the dmub service
+ * @cmd: the command packet to be executed
+ *
+ * Executes a single command for the dmub.
+ *
+ * Return:
+ *   DMUB_STATUS_OK - success
+ *   DMUB_STATUS_INVALID - unspecified error
+ */
+enum dmub_status dmub_srv_reg_cmd_execute(struct dmub_srv *dmub, union dmub_rb_cmd *cmd);
+
+
+/**
+ * dmub_srv_cmd_get_response() - Copies return data for command into buffer
+ * @dmub: the dmub service
+ * @cmd_rsp: response buffer
+ *
+ * Copies return data for command into buffer
+ */
+void dmub_srv_cmd_get_response(struct dmub_srv *dmub,
+		union dmub_rb_cmd *cmd_rsp);
+
+/**
+ * dmub_srv_sync_inboxes() - Sync inbox state
+ * @dmub: the dmub service
+ *
+ * Sync inbox state
+ *
+ * Return:
+ *   DMUB_STATUS_OK - success
+ *   DMUB_STATUS_INVALID - unspecified error
+ */
+enum dmub_status dmub_srv_sync_inboxes(struct dmub_srv *dmub);
+
+/**
+ * dmub_srv_wait_for_inbox_free() - Waits for space in the DMUB inbox to free up
+ * @dmub: the dmub service
+ * @timeout_us: the maximum number of microseconds to wait
+ * @num_free_required: number of free entries required
+ *
+ * Waits until the DMUB buffer is freed to the specified number.
+ *  The maximum wait time is given in microseconds to prevent spinning
+ * forever.
+ *
+ * Return:
+ *   DMUB_STATUS_OK - success
+ *   DMUB_STATUS_TIMEOUT - wait for buffer to flush timed out
+ *   DMUB_STATUS_INVALID - unspecified error
+ */
+enum dmub_status dmub_srv_wait_for_inbox_free(struct dmub_srv *dmub,
+		uint32_t timeout_us,
+		uint32_t num_free_required);
+
+/**
+ * dmub_srv_update_inbox_status() - Updates pending status for inbox & reg inbox0
+ * @dmub: the dmub service
+ *
+ * Return:
+ *   DMUB_STATUS_OK - success
+ *   DMUB_STATUS_TIMEOUT - wait for buffer to flush timed out
+ *   DMUB_STATUS_HW_FAILURE - issue with HW programming
+ *   DMUB_STATUS_INVALID - unspecified error
+ */
+enum dmub_status dmub_srv_update_inbox_status(struct dmub_srv *dmub);
+
 #endif /* _DMUB_SRV_H_ */
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 1f5f4e3e49d4..57fa05bddb45 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -36,6 +36,9 @@
 //<DMUB_TYPES>==================================================================
 /* Basic type definitions. */
 
+#ifdef __forceinline
+#undef __forceinline
+#endif
 #define __forceinline inline
 
 /**
@@ -547,6 +550,11 @@ union replay_hw_flags {
 		 * @is_alpm_initialized: Indicates whether ALPM is initialized
 		 */
 		uint32_t is_alpm_initialized : 1;
+
+		/**
+		 * @alpm_mode: Indicates ALPM mode selected
+		 */
+		uint32_t alpm_mode : 2;
 	} bitfields;
 
 	uint32_t u32All;
@@ -739,6 +747,14 @@ enum dmub_ips_disable_type {
 	DMUB_IPS_DISABLE_IPS2_Z10 = 4,
 	DMUB_IPS_DISABLE_DYNAMIC = 5,
 	DMUB_IPS_RCG_IN_ACTIVE_IPS2_IN_OFF = 6,
+	DMUB_IPS_DISABLE_Z8_RETENTION = 7,
+};
+
+enum dmub_ips_rcg_disable_type {
+	DMUB_IPS_RCG_ENABLE = 0,
+	DMUB_IPS0_RCG_DISABLE = 1,
+	DMUB_IPS1_RCG_DISABLE = 2,
+	DMUB_IPS_RCG_DISABLE = 3
 };
 
 #define DMUB_IPS1_ALLOW_MASK 0x00000001
@@ -817,11 +833,12 @@ enum dmub_shared_state_feature_id {
  */
 union dmub_shared_state_ips_fw_signals {
 	struct {
-		uint32_t ips1_commit : 1;  /**< 1 if in IPS1 */
+		uint32_t ips1_commit : 1;  /**< 1 if in IPS1 or IPS0 RCG */
 		uint32_t ips2_commit : 1; /**< 1 if in IPS2 */
 		uint32_t in_idle : 1; /**< 1 if DMCUB is in idle */
 		uint32_t detection_required : 1; /**< 1 if detection is required */
-		uint32_t reserved_bits : 28; /**< Reversed */
+		uint32_t ips1z8_commit: 1; /**< 1 if in IPS1 Z8 Retention */
+		uint32_t reserved_bits : 27; /**< Reversed */
 	} bits;
 	uint32_t all;
 };
@@ -836,7 +853,10 @@ union dmub_shared_state_ips_driver_signals {
 		uint32_t allow_ips2 : 1; /**< 1 is IPS1 is allowed */
 		uint32_t allow_z10 : 1; /**< 1 if Z10 is allowed */
 		uint32_t allow_idle: 1; /**< 1 if driver is allowing idle */
-		uint32_t reserved_bits : 27; /**< Reversed bits */
+		uint32_t allow_ips0_rcg : 1; /**< 1 is IPS0 RCG is allowed */
+		uint32_t allow_ips1_rcg : 1; /**< 1 is IPS1 RCG is allowed */
+		uint32_t allow_ips1z8 : 1; /**< 1 is IPS1 Z8 Retention is allowed */
+		uint32_t reserved_bits : 24; /**< Reversed bits */
 	} bits;
 	uint32_t all;
 };
@@ -865,7 +885,9 @@ struct dmub_shared_state_ips_fw {
 	uint32_t ips1_exit_count; /**< Exit counter for IPS1 */
 	uint32_t ips2_entry_count; /**< Entry counter for IPS2 */
 	uint32_t ips2_exit_count; /**< Exit counter for IPS2 */
-	uint32_t reserved[55]; /**< Reversed, to be updated when adding new fields. */
+	uint32_t ips1_z8ret_entry_count; /**< Entry counter for IPS1 Z8 Retention */
+	uint32_t ips1_z8ret_exit_count; /**< Exit counter for IPS1 Z8 Retention */
+	uint32_t reserved[53]; /**< Reversed, to be updated when adding new fields. */
 }; /* 248-bytes, fixed */
 
 /**
@@ -1253,6 +1275,10 @@ enum dmub_gpint_command {
 	 * DESC: Setup debug configs.
 	 */
 	DMUB_GPINT__SETUP_DEBUG_MODE = 136,
+	/**
+	 * DESC: Initiates IPS wake sequence.
+	 */
+	DMUB_GPINT__IPS_DEBUG_WAKE = 137,
 };
 
 /**
@@ -2113,6 +2139,11 @@ union dmub_cmd_fams2_config {
 	} stream_v1; //v1
 };
 
+struct dmub_fams2_config_v2 {
+	struct dmub_cmd_fams2_global_config global;
+	struct dmub_fams2_stream_static_state_v1 stream_v1[DMUB_MAX_STREAMS]; //v1
+};
+
 /**
  * DMUB rb command definition for FAMS2 (merged SubVP, FPO, Legacy)
  */
@@ -2122,6 +2153,22 @@ struct dmub_rb_cmd_fams2 {
 };
 
 /**
+ * Indirect buffer descriptor
+ */
+struct dmub_ib_data {
+	union dmub_addr src; // location of indirect buffer in memory
+	uint16_t size; // indirect buffer size in bytes
+};
+
+/**
+ * DMUB rb command definition for commands passed over indirect buffer
+ */
+struct dmub_rb_cmd_ib {
+	struct dmub_cmd_header header;
+	struct dmub_ib_data ib_data;
+};
+
+/**
  * enum dmub_cmd_idle_opt_type - Idle optimization command type.
  */
 enum dmub_cmd_idle_opt_type {
@@ -2144,6 +2191,11 @@ enum dmub_cmd_idle_opt_type {
 	 * DCN hardware notify power state.
 	 */
 	DMUB_CMD__IDLE_OPT_SET_DC_POWER_STATE = 3,
+
+	/**
+	 * DCN notify to release HW.
+	 */
+	 DMUB_CMD__IDLE_OPT_RELEASE_HW = 4,
 };
 
 /**
@@ -2636,7 +2688,11 @@ enum dp_hpd_type {
 	/**
 	 * DP HPD short pulse
 	 */
-	DP_IRQ
+	DP_IRQ = 1,
+	/**
+	 * Failure to acquire DP HPD state
+	 */
+	DP_NONE_HPD = 2
 };
 
 /**
@@ -2901,8 +2957,9 @@ enum dmub_cmd_fams_type {
 	 */
 	DMUB_CMD__FAMS_SET_MANUAL_TRIGGER = 3,
 	DMUB_CMD__FAMS2_CONFIG = 4,
-	DMUB_CMD__FAMS2_DRR_UPDATE = 5,
-	DMUB_CMD__FAMS2_FLIP = 6,
+	DMUB_CMD__FAMS2_IB_CONFIG = 5,
+	DMUB_CMD__FAMS2_DRR_UPDATE = 6,
+	DMUB_CMD__FAMS2_FLIP = 7,
 };
 
 /**
@@ -3609,6 +3666,12 @@ struct dmub_rb_cmd_psr_set_power_opt {
 	struct dmub_cmd_psr_set_power_opt_data psr_set_power_opt_data;
 };
 
+enum dmub_alpm_mode {
+	ALPM_AUXWAKE = 0,
+	ALPM_AUXLESS = 1,
+	ALPM_UNSUPPORTED = 2,
+};
+
 /**
  * Definition of Replay Residency GPINT command.
  * Bit[0] - Residency mode for Revision 0
@@ -3742,6 +3805,15 @@ enum dmub_cmd_replay_general_subtype {
 	REPLAY_GENERAL_CMD_SET_LOW_RR_ACTIVATE,
 };
 
+struct dmub_alpm_auxless_data {
+	uint16_t lfps_setup_ns;
+	uint16_t lfps_period_ns;
+	uint16_t lfps_silence_ns;
+	uint16_t lfps_t1_t2_override_us;
+	short lfps_t1_t2_offset_us;
+	uint8_t lttpr_count;
+};
+
 /**
  * Data passed from driver to FW in a DMUB_CMD__REPLAY_COPY_SETTINGS command.
  */
@@ -3812,6 +3884,10 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * Use FSM state for Replay power up/down
 	 */
 	uint8_t use_phy_fsm;
+	/**
+	 * Use for AUX-less ALPM LFPS wake operation
+	 */
+	struct dmub_alpm_auxless_data auxless_alpm_data;
 };
 
 /**
@@ -4360,6 +4436,11 @@ enum dmub_cmd_abm_type {
 	 * Get the current ACE curve.
 	 */
 	DMUB_CMD__ABM_GET_ACE_CURVE = 10,
+
+	/**
+	 * Get current histogram data
+	 */
+	DMUB_CMD__ABM_GET_HISTOGRAM_DATA = 11,
 };
 
 struct abm_ace_curve {
@@ -4954,6 +5035,20 @@ enum dmub_abm_ace_curve_type {
 };
 
 /**
+ * enum dmub_abm_histogram_type - Histogram type.
+ */
+enum dmub_abm_histogram_type {
+	/**
+	 * ACE curve as defined by the SW layer.
+	 */
+	ABM_HISTOGRAM_TYPE__SW = 0,
+	/**
+	 * ACE curve as defined by the SW to HW translation interface layer.
+	 */
+	ABM_HISTOGRAM_TYPE__SW_IF = 1,
+};
+
+/**
  * Definition of a DMUB_CMD__ABM_GET_ACE_CURVE command.
  */
 struct dmub_rb_cmd_abm_get_ace_curve {
@@ -4989,6 +5084,41 @@ struct dmub_rb_cmd_abm_get_ace_curve {
 };
 
 /**
+ * Definition of a DMUB_CMD__ABM_GET_HISTOGRAM command.
+ */
+struct dmub_rb_cmd_abm_get_histogram {
+	/**
+	 * Command header.
+	 */
+	struct dmub_cmd_header header;
+
+	/**
+	 * Address where Histogram should be copied.
+	 */
+	union dmub_addr dest;
+
+	/**
+	 * Type of Histogram being queried.
+	 */
+	enum dmub_abm_histogram_type histogram_type;
+
+	/**
+	 * Indirect buffer length.
+	 */
+	uint16_t bytes;
+
+	/**
+	 * eDP panel instance.
+	 */
+	uint8_t panel_inst;
+
+	/**
+	 * Explicit padding to 4 byte boundary.
+	 */
+	uint8_t pad;
+};
+
+/**
  * Definition of a DMUB_CMD__ABM_SAVE_RESTORE command.
  */
 struct dmub_rb_cmd_abm_save_restore {
@@ -5389,7 +5519,8 @@ struct dmub_cmd_fused_request {
 		struct dmub_cmd_fused_request_location_i2c {
 			uint8_t is_aux : 1;  // False
 			uint8_t ddc_line : 3;
-			uint8_t _reserved0 : 4;
+			uint8_t over_aux : 1;
+			uint8_t _reserved0 : 3;
 			uint8_t address;
 			uint8_t offset;
 			uint8_t length;
@@ -5687,6 +5818,11 @@ union dmub_rb_cmd {
 	struct dmub_rb_cmd_abm_get_ace_curve abm_get_ace_curve;
 
 	/**
+	 * Definition of a DMUB_CMD__ABM_GET_HISTOGRAM command.
+	 */
+	struct dmub_rb_cmd_abm_get_histogram abm_get_histogram;
+
+	/**
 	 * Definition of a DMUB_CMD__ABM_SET_EVENT command.
 	 */
 	struct dmub_rb_cmd_abm_set_event abm_set_event;
@@ -5817,8 +5953,11 @@ union dmub_rb_cmd {
 	 * Definition of a DMUB_CMD__PSP_ASSR_ENABLE command.
 	 */
 	struct dmub_rb_cmd_assr_enable assr_enable;
+
 	struct dmub_rb_cmd_fams2 fams2_config;
 
+	struct dmub_rb_cmd_ib ib_fams2_config;
+
 	struct dmub_rb_cmd_fams2_drr_update fams2_drr_update;
 
 	struct dmub_rb_cmd_fams2_flip fams2_flip;
@@ -5934,6 +6073,9 @@ static inline uint32_t dmub_rb_num_free(struct dmub_rb *rb)
 	else
 		data_count = rb->capacity - (rb->rptr - rb->wrpt);
 
+	/* +1 because 1 entry is always unusable */
+	data_count += DMUB_RB_CMD_SIZE;
+
 	return (rb->capacity - data_count) / DMUB_RB_CMD_SIZE;
 }
 
@@ -5953,6 +6095,7 @@ static inline bool dmub_rb_full(struct dmub_rb *rb)
 	else
 		data_count = rb->capacity - (rb->rptr - rb->wrpt);
 
+	/* -1 because 1 entry is always unusable */
 	return (data_count == (rb->capacity - DMUB_RB_CMD_SIZE));
 }
 
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c
index e67f7c4784eb..2575dbc448f7 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c
@@ -66,24 +66,20 @@ void dmub_dcn401_reset(struct dmub_srv *dmub)
 	const uint32_t timeout_us = 1 * 1000 * 1000; //1s
 	const uint32_t poll_delay_us = 1; //1us
 	uint32_t i = 0;
-	uint32_t in_reset, scratch, pwait_mode;
+	uint32_t enabled, in_reset, scratch, pwait_mode;
 
-	REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset);
+	REG_GET(DMCUB_CNTL,
+			DMCUB_ENABLE, &enabled);
+	REG_GET(DMCUB_CNTL2,
+			DMCUB_SOFT_RESET, &in_reset);
 
-	if (in_reset == 0) {
+	if (enabled && in_reset == 0) {
 		cmd.bits.status = 1;
 		cmd.bits.command_code = DMUB_GPINT__STOP_FW;
 		cmd.bits.param = 0;
 
 		dmub->hw_funcs.set_gpint(dmub, cmd);
 
-		for (i = 0; i < timeout_us; i++) {
-			if (dmub->hw_funcs.is_gpint_acked(dmub, cmd))
-				break;
-
-			udelay(poll_delay_us);
-		}
-
 		for (; i < timeout_us; i++) {
 			scratch = dmub->hw_funcs.get_gpint_response(dmub);
 			if (scratch == DMUB_GPINT__STOP_FW_RESPONSE)
@@ -517,28 +513,69 @@ void dmub_dcn401_send_reg_inbox0_cmd_msg(struct dmub_srv *dmub,
 		union dmub_rb_cmd *cmd)
 {
 	uint32_t *dwords = (uint32_t *)cmd;
-
+	int32_t payload_size_bytes = cmd->cmd_common.header.payload_bytes;
+	uint32_t msg_index;
 	static_assert(sizeof(*cmd) == 64, "DMUB command size mismatch");
 
-	REG_WRITE(DMCUB_REG_INBOX0_MSG0, dwords[0]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG1, dwords[1]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG2, dwords[2]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG3, dwords[3]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG4, dwords[4]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG5, dwords[5]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG6, dwords[6]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG7, dwords[7]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG8, dwords[8]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG9, dwords[9]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG10, dwords[10]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG11, dwords[11]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG12, dwords[12]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG13, dwords[13]);
-	REG_WRITE(DMCUB_REG_INBOX0_MSG14, dwords[14]);
+	/* read remaining data based on payload size */
+	for (msg_index = 0; msg_index < 15; msg_index++) {
+		if (payload_size_bytes <= msg_index * 4) {
+			break;
+		}
+
+		switch (msg_index) {
+		case 0:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG0, dwords[msg_index + 1]);
+			break;
+		case 1:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG1, dwords[msg_index + 1]);
+			break;
+		case 2:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG2, dwords[msg_index + 1]);
+			break;
+		case 3:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG3, dwords[msg_index + 1]);
+			break;
+		case 4:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG4, dwords[msg_index + 1]);
+			break;
+		case 5:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG5, dwords[msg_index + 1]);
+			break;
+		case 6:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG6, dwords[msg_index + 1]);
+			break;
+		case 7:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG7, dwords[msg_index + 1]);
+			break;
+		case 8:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG8, dwords[msg_index + 1]);
+			break;
+		case 9:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG9, dwords[msg_index + 1]);
+			break;
+		case 10:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG10, dwords[msg_index + 1]);
+			break;
+		case 11:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG11, dwords[msg_index + 1]);
+			break;
+		case 12:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG12, dwords[msg_index + 1]);
+			break;
+		case 13:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG13, dwords[msg_index + 1]);
+			break;
+		case 14:
+			REG_WRITE(DMCUB_REG_INBOX0_MSG14, dwords[msg_index + 1]);
+			break;
+		}
+	}
+
 	/* writing to INBOX RDY register will trigger DMUB REG INBOX0 RDY
 	 * interrupt.
 	 */
-	REG_WRITE(DMCUB_REG_INBOX0_RDY, dwords[15]);
+	REG_WRITE(DMCUB_REG_INBOX0_RDY, dwords[0]);
 }
 
 uint32_t dmub_dcn401_read_reg_inbox0_rsp_int_status(struct dmub_srv *dmub)
@@ -556,30 +593,39 @@ void dmub_dcn401_read_reg_inbox0_cmd_rsp(struct dmub_srv *dmub,
 
 	static_assert(sizeof(*cmd) == 64, "DMUB command size mismatch");
 
-	dwords[0] = REG_READ(DMCUB_REG_INBOX0_MSG0);
-	dwords[1] = REG_READ(DMCUB_REG_INBOX0_MSG1);
-	dwords[2] = REG_READ(DMCUB_REG_INBOX0_MSG2);
-	dwords[3] = REG_READ(DMCUB_REG_INBOX0_MSG3);
-	dwords[4] = REG_READ(DMCUB_REG_INBOX0_MSG4);
-	dwords[5] = REG_READ(DMCUB_REG_INBOX0_MSG5);
-	dwords[6] = REG_READ(DMCUB_REG_INBOX0_MSG6);
-	dwords[7] = REG_READ(DMCUB_REG_INBOX0_MSG7);
-	dwords[8] = REG_READ(DMCUB_REG_INBOX0_MSG8);
-	dwords[9] = REG_READ(DMCUB_REG_INBOX0_MSG9);
-	dwords[10] = REG_READ(DMCUB_REG_INBOX0_MSG10);
-	dwords[11] = REG_READ(DMCUB_REG_INBOX0_MSG11);
-	dwords[12] = REG_READ(DMCUB_REG_INBOX0_MSG12);
-	dwords[13] = REG_READ(DMCUB_REG_INBOX0_MSG13);
-	dwords[14] = REG_READ(DMCUB_REG_INBOX0_MSG14);
-	dwords[15] = REG_READ(DMCUB_REG_INBOX0_RSP);
+	dwords[0] = REG_READ(DMCUB_REG_INBOX0_RSP);
+	dwords[1] = REG_READ(DMCUB_REG_INBOX0_MSG0);
+	dwords[2] = REG_READ(DMCUB_REG_INBOX0_MSG1);
+	dwords[3] = REG_READ(DMCUB_REG_INBOX0_MSG2);
+	dwords[4] = REG_READ(DMCUB_REG_INBOX0_MSG3);
+	dwords[5] = REG_READ(DMCUB_REG_INBOX0_MSG4);
+	dwords[6] = REG_READ(DMCUB_REG_INBOX0_MSG5);
+	dwords[7] = REG_READ(DMCUB_REG_INBOX0_MSG6);
+	dwords[8] = REG_READ(DMCUB_REG_INBOX0_MSG7);
+	dwords[9] = REG_READ(DMCUB_REG_INBOX0_MSG8);
+	dwords[10] = REG_READ(DMCUB_REG_INBOX0_MSG9);
+	dwords[11] = REG_READ(DMCUB_REG_INBOX0_MSG10);
+	dwords[12] = REG_READ(DMCUB_REG_INBOX0_MSG11);
+	dwords[13] = REG_READ(DMCUB_REG_INBOX0_MSG12);
+	dwords[14] = REG_READ(DMCUB_REG_INBOX0_MSG13);
+	dwords[15] = REG_READ(DMCUB_REG_INBOX0_MSG14);
 }
 
 void dmub_dcn401_write_reg_inbox0_rsp_int_ack(struct dmub_srv *dmub)
 {
 	REG_UPDATE(HOST_INTERRUPT_CSR, HOST_REG_INBOX0_RSP_INT_ACK, 1);
+}
+
+void dmub_dcn401_clear_reg_inbox0_rsp_int_ack(struct dmub_srv *dmub)
+{
 	REG_UPDATE(HOST_INTERRUPT_CSR, HOST_REG_INBOX0_RSP_INT_ACK, 0);
 }
 
+void dmub_dcn401_enable_reg_inbox0_rsp_int(struct dmub_srv *dmub, bool enable)
+{
+	REG_UPDATE(HOST_INTERRUPT_CSR, HOST_REG_INBOX0_RSP_INT_EN, enable ? 1:0);
+}
+
 void dmub_dcn401_write_reg_outbox0_rdy_int_ack(struct dmub_srv *dmub)
 {
 	REG_UPDATE(HOST_INTERRUPT_CSR, HOST_REG_OUTBOX0_RDY_INT_ACK, 1);
@@ -604,11 +650,6 @@ uint32_t dmub_dcn401_read_reg_outbox0_rsp_int_status(struct dmub_srv *dmub)
 	return status;
 }
 
-void dmub_dcn401_enable_reg_inbox0_rsp_int(struct dmub_srv *dmub, bool enable)
-{
-	REG_UPDATE(HOST_INTERRUPT_CSR, HOST_REG_INBOX0_RSP_INT_EN, enable ? 1:0);
-}
-
 void dmub_dcn401_enable_reg_outbox0_rdy_int(struct dmub_srv *dmub, bool enable)
 {
 	REG_UPDATE(HOST_INTERRUPT_CSR, HOST_REG_OUTBOX0_RDY_INT_EN, enable ? 1:0);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h
index c35be52676f6..88c3a44d67d9 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h
@@ -277,11 +277,13 @@ uint32_t dmub_dcn401_read_reg_inbox0_rsp_int_status(struct dmub_srv *dmub);
 void dmub_dcn401_read_reg_inbox0_cmd_rsp(struct dmub_srv *dmub,
 		union dmub_rb_cmd *cmd);
 void dmub_dcn401_write_reg_inbox0_rsp_int_ack(struct dmub_srv *dmub);
+void dmub_dcn401_clear_reg_inbox0_rsp_int_ack(struct dmub_srv *dmub);
+void dmub_dcn401_enable_reg_inbox0_rsp_int(struct dmub_srv *dmub, bool enable);
+
 void dmub_dcn401_write_reg_outbox0_rdy_int_ack(struct dmub_srv *dmub);
 void dmub_dcn401_read_reg_outbox0_msg(struct dmub_srv *dmub, uint32_t *msg);
 void dmub_dcn401_write_reg_outbox0_rsp(struct dmub_srv *dmub, uint32_t *msg);
 uint32_t dmub_dcn401_read_reg_outbox0_rsp_int_status(struct dmub_srv *dmub);
-void dmub_dcn401_enable_reg_inbox0_rsp_int(struct dmub_srv *dmub, bool enable);
 void dmub_dcn401_enable_reg_outbox0_rdy_int(struct dmub_srv *dmub, bool enable);
 uint32_t dmub_dcn401_read_reg_outbox0_rdy_int_status(struct dmub_srv *dmub);
 
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
index ae8133816b43..acca7943a8c8 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
@@ -157,6 +157,9 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 {
 	struct dmub_srv_hw_funcs *funcs = &dmub->hw_funcs;
 
+	/* default to specifying now inbox type */
+	enum dmub_inbox_cmd_interface_type default_inbox_type = DMUB_CMD_INTERFACE_DEFAULT;
+
 	switch (asic) {
 	case DMUB_ASIC_DCN20:
 	case DMUB_ASIC_DCN21:
@@ -395,10 +398,15 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 
 		funcs->get_current_time = dmub_dcn401_get_current_time;
 		funcs->get_diagnostic_data = dmub_dcn401_get_diagnostic_data;
+
 		funcs->send_reg_inbox0_cmd_msg = dmub_dcn401_send_reg_inbox0_cmd_msg;
 		funcs->read_reg_inbox0_rsp_int_status = dmub_dcn401_read_reg_inbox0_rsp_int_status;
 		funcs->read_reg_inbox0_cmd_rsp = dmub_dcn401_read_reg_inbox0_cmd_rsp;
 		funcs->write_reg_inbox0_rsp_int_ack = dmub_dcn401_write_reg_inbox0_rsp_int_ack;
+		funcs->clear_reg_inbox0_rsp_int_ack = dmub_dcn401_clear_reg_inbox0_rsp_int_ack;
+		funcs->enable_reg_inbox0_rsp_int = dmub_dcn401_enable_reg_inbox0_rsp_int;
+		default_inbox_type = DMUB_CMD_INTERFACE_FB; // still default to FB for now
+
 		funcs->write_reg_outbox0_rdy_int_ack = dmub_dcn401_write_reg_outbox0_rdy_int_ack;
 		funcs->read_reg_outbox0_msg = dmub_dcn401_read_reg_outbox0_msg;
 		funcs->write_reg_outbox0_rsp = dmub_dcn401_write_reg_outbox0_rsp;
@@ -411,6 +419,20 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 		return false;
 	}
 
+	/* set default inbox type if not overriden */
+	if (dmub->inbox_type == DMUB_CMD_INTERFACE_DEFAULT) {
+		if (default_inbox_type != DMUB_CMD_INTERFACE_DEFAULT) {
+			/* use default inbox type as specified by DCN rev */
+			dmub->inbox_type = default_inbox_type;
+		} else if (funcs->send_reg_inbox0_cmd_msg) {
+			/* prefer reg as default inbox type if present */
+			dmub->inbox_type = DMUB_CMD_INTERFACE_REG;
+		} else {
+			/* use fb as fallback */
+			dmub->inbox_type = DMUB_CMD_INTERFACE_FB;
+		}
+	}
+
 	return true;
 }
 
@@ -426,6 +448,7 @@ enum dmub_status dmub_srv_create(struct dmub_srv *dmub,
 	dmub->asic = params->asic;
 	dmub->fw_version = params->fw_version;
 	dmub->is_virtual = params->is_virtual;
+	dmub->inbox_type = params->inbox_type;
 
 	/* Setup asic dependent hardware funcs. */
 	if (!dmub_srv_hw_setup(dmub, params->asic)) {
@@ -695,7 +718,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
 	inbox1.base = cw4.region.base;
 	inbox1.top = cw4.region.base + DMUB_RB_SIZE;
 	outbox1.base = inbox1.top;
-	outbox1.top = cw4.region.top;
+	outbox1.top = inbox1.top + DMUB_RB_SIZE;
 
 	cw5.offset.quad_part = tracebuff_fb->gpu_addr;
 	cw5.region.base = DMUB_CW5_BASE;
@@ -737,7 +760,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
 	rb_params.ctx = dmub;
 	rb_params.base_address = mail_fb->cpu_addr;
 	rb_params.capacity = DMUB_RB_SIZE;
-	dmub_rb_init(&dmub->inbox1_rb, &rb_params);
+	dmub_rb_init(&dmub->inbox1.rb, &rb_params);
 
 	// Initialize outbox1 ring buffer
 	rb_params.ctx = dmub;
@@ -768,27 +791,6 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
 	return DMUB_STATUS_OK;
 }
 
-enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub)
-{
-	if (!dmub->sw_init)
-		return DMUB_STATUS_INVALID;
-
-	if (dmub->hw_funcs.get_inbox1_rptr && dmub->hw_funcs.get_inbox1_wptr) {
-		uint32_t rptr = dmub->hw_funcs.get_inbox1_rptr(dmub);
-		uint32_t wptr = dmub->hw_funcs.get_inbox1_wptr(dmub);
-
-		if (rptr > dmub->inbox1_rb.capacity || wptr > dmub->inbox1_rb.capacity) {
-			return DMUB_STATUS_HW_FAILURE;
-		} else {
-			dmub->inbox1_rb.rptr = rptr;
-			dmub->inbox1_rb.wrpt = wptr;
-			dmub->inbox1_last_wptr = dmub->inbox1_rb.wrpt;
-		}
-	}
-
-	return DMUB_STATUS_OK;
-}
-
 enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub)
 {
 	if (!dmub->sw_init)
@@ -799,8 +801,13 @@ enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub)
 
 	/* mailboxes have been reset in hw, so reset the sw state as well */
 	dmub->inbox1_last_wptr = 0;
-	dmub->inbox1_rb.wrpt = 0;
-	dmub->inbox1_rb.rptr = 0;
+	dmub->inbox1.rb.wrpt = 0;
+	dmub->inbox1.rb.rptr = 0;
+	dmub->inbox1.num_reported = 0;
+	dmub->inbox1.num_submitted = 0;
+	dmub->reg_inbox0.num_reported = 0;
+	dmub->reg_inbox0.num_submitted = 0;
+	dmub->reg_inbox0.is_pending = 0;
 	dmub->outbox0_rb.wrpt = 0;
 	dmub->outbox0_rb.rptr = 0;
 	dmub->outbox1_rb.wrpt = 0;
@@ -811,7 +818,7 @@ enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub)
 	return DMUB_STATUS_OK;
 }
 
-enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub,
+enum dmub_status dmub_srv_fb_cmd_queue(struct dmub_srv *dmub,
 				    const union dmub_rb_cmd *cmd)
 {
 	if (!dmub->hw_init)
@@ -820,18 +827,20 @@ enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub,
 	if (dmub->power_state != DMUB_POWER_STATE_D0)
 		return DMUB_STATUS_POWER_STATE_D3;
 
-	if (dmub->inbox1_rb.rptr > dmub->inbox1_rb.capacity ||
-	    dmub->inbox1_rb.wrpt > dmub->inbox1_rb.capacity) {
+	if (dmub->inbox1.rb.rptr > dmub->inbox1.rb.capacity ||
+	    dmub->inbox1.rb.wrpt > dmub->inbox1.rb.capacity) {
 		return DMUB_STATUS_HW_FAILURE;
 	}
 
-	if (dmub_rb_push_front(&dmub->inbox1_rb, cmd))
+	if (dmub_rb_push_front(&dmub->inbox1.rb, cmd)) {
+		dmub->inbox1.num_submitted++;
 		return DMUB_STATUS_OK;
+	}
 
 	return DMUB_STATUS_QUEUE_FULL;
 }
 
-enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub)
+enum dmub_status dmub_srv_fb_cmd_execute(struct dmub_srv *dmub)
 {
 	struct dmub_rb flush_rb;
 
@@ -846,13 +855,13 @@ enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub)
 	 * been flushed to framebuffer memory. Otherwise DMCUB might
 	 * read back stale, fully invalid or partially invalid data.
 	 */
-	flush_rb = dmub->inbox1_rb;
+	flush_rb = dmub->inbox1.rb;
 	flush_rb.rptr = dmub->inbox1_last_wptr;
 	dmub_rb_flush_pending(&flush_rb);
 
-		dmub->hw_funcs.set_inbox1_wptr(dmub, dmub->inbox1_rb.wrpt);
+		dmub->hw_funcs.set_inbox1_wptr(dmub, dmub->inbox1.rb.wrpt);
 
-	dmub->inbox1_last_wptr = dmub->inbox1_rb.wrpt;
+	dmub->inbox1_last_wptr = dmub->inbox1.rb.wrpt;
 
 	return DMUB_STATUS_OK;
 }
@@ -910,26 +919,84 @@ enum dmub_status dmub_srv_wait_for_auto_load(struct dmub_srv *dmub,
 	return DMUB_STATUS_TIMEOUT;
 }
 
+static void dmub_srv_update_reg_inbox0_status(struct dmub_srv *dmub)
+{
+	if (dmub->reg_inbox0.is_pending) {
+		dmub->reg_inbox0.is_pending = dmub->hw_funcs.read_reg_inbox0_rsp_int_status &&
+				!dmub->hw_funcs.read_reg_inbox0_rsp_int_status(dmub);
+
+		if (!dmub->reg_inbox0.is_pending) {
+			/* ack the rsp interrupt */
+			if (dmub->hw_funcs.write_reg_inbox0_rsp_int_ack)
+				dmub->hw_funcs.write_reg_inbox0_rsp_int_ack(dmub);
+
+			/* only update the reported count if commands aren't being batched */
+			if (!dmub->reg_inbox0.is_pending && !dmub->reg_inbox0.is_multi_pending) {
+				dmub->reg_inbox0.num_reported = dmub->reg_inbox0.num_submitted;
+			}
+		}
+	}
+}
+
+enum dmub_status dmub_srv_wait_for_pending(struct dmub_srv *dmub,
+					uint32_t timeout_us)
+{
+	uint32_t i;
+	const uint32_t polling_interval_us = 1;
+	struct dmub_srv_inbox scratch_reg_inbox0 = dmub->reg_inbox0;
+	struct dmub_srv_inbox scratch_inbox1 = dmub->inbox1;
+	const volatile struct dmub_srv_inbox *reg_inbox0 = &dmub->reg_inbox0;
+	const volatile struct dmub_srv_inbox *inbox1 = &dmub->inbox1;
+
+	if (!dmub->hw_init ||
+			!dmub->hw_funcs.get_inbox1_wptr)
+		return DMUB_STATUS_INVALID;
+
+	for (i = 0; i <= timeout_us; i += polling_interval_us) {
+			scratch_inbox1.rb.wrpt = dmub->hw_funcs.get_inbox1_wptr(dmub);
+			scratch_inbox1.rb.rptr = dmub->hw_funcs.get_inbox1_rptr(dmub);
+
+		scratch_reg_inbox0.is_pending = scratch_reg_inbox0.is_pending &&
+				dmub->hw_funcs.read_reg_inbox0_rsp_int_status &&
+				!dmub->hw_funcs.read_reg_inbox0_rsp_int_status(dmub);
+
+		if (scratch_inbox1.rb.rptr > dmub->inbox1.rb.capacity)
+			return DMUB_STATUS_HW_FAILURE;
+
+		/* check current HW state first, but use command submission vs reported as a fallback */
+		if ((dmub_rb_empty(&scratch_inbox1.rb) ||
+				inbox1->num_reported >= scratch_inbox1.num_submitted) &&
+				(!scratch_reg_inbox0.is_pending ||
+				reg_inbox0->num_reported >= scratch_reg_inbox0.num_submitted))
+			return DMUB_STATUS_OK;
+
+		udelay(polling_interval_us);
+	}
+
+	return DMUB_STATUS_TIMEOUT;
+}
+
 enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub,
 					uint32_t timeout_us)
 {
-	uint32_t i, rptr;
+	enum dmub_status status;
+	uint32_t i;
+	const uint32_t polling_interval_us = 1;
 
 	if (!dmub->hw_init)
 		return DMUB_STATUS_INVALID;
 
-	for (i = 0; i <= timeout_us; ++i) {
-			rptr = dmub->hw_funcs.get_inbox1_rptr(dmub);
+	for (i = 0; i < timeout_us; i += polling_interval_us) {
+		status = dmub_srv_update_inbox_status(dmub);
 
-		if (rptr > dmub->inbox1_rb.capacity)
-			return DMUB_STATUS_HW_FAILURE;
+		if (status != DMUB_STATUS_OK)
+			return status;
 
-		dmub->inbox1_rb.rptr = rptr;
-
-		if (dmub_rb_empty(&dmub->inbox1_rb))
+		/* check for idle */
+		if (dmub_rb_empty(&dmub->inbox1.rb) && !dmub->reg_inbox0.is_pending)
 			return DMUB_STATUS_OK;
 
-		udelay(1);
+		udelay(polling_interval_us);
 	}
 
 	return DMUB_STATUS_TIMEOUT;
@@ -1040,35 +1107,6 @@ enum dmub_status dmub_srv_set_skip_panel_power_sequence(struct dmub_srv *dmub,
 	return DMUB_STATUS_OK;
 }
 
-enum dmub_status dmub_srv_cmd_with_reply_data(struct dmub_srv *dmub,
-					      union dmub_rb_cmd *cmd)
-{
-	enum dmub_status status = DMUB_STATUS_OK;
-
-	// Queue command
-	status = dmub_srv_cmd_queue(dmub, cmd);
-
-	if (status != DMUB_STATUS_OK)
-		return status;
-
-	// Execute command
-	status = dmub_srv_cmd_execute(dmub);
-
-	if (status != DMUB_STATUS_OK)
-		return status;
-
-	// Wait for DMUB to process command
-	status = dmub_srv_wait_for_idle(dmub, 100000);
-
-	if (status != DMUB_STATUS_OK)
-		return status;
-
-	// Copy data back from ring buffer into command
-	dmub_rb_get_return_data(&dmub->inbox1_rb, cmd);
-
-	return status;
-}
-
 static inline bool dmub_rb_out_trace_buffer_front(struct dmub_rb *rb,
 				 void *entry)
 {
@@ -1160,47 +1198,162 @@ void dmub_srv_subvp_save_surf_addr(struct dmub_srv *dmub, const struct dc_plane_
 	}
 }
 
+void dmub_srv_set_power_state(struct dmub_srv *dmub, enum dmub_srv_power_state_type dmub_srv_power_state)
+{
+	if (!dmub || !dmub->hw_init)
+		return;
+
+	dmub->power_state = dmub_srv_power_state;
+}
 
-enum dmub_status dmub_srv_send_reg_inbox0_cmd(
-		struct dmub_srv *dmub,
-		union dmub_rb_cmd *cmd,
-		bool with_reply, uint32_t timeout_us)
+enum dmub_status dmub_srv_reg_cmd_execute(struct dmub_srv *dmub, union dmub_rb_cmd *cmd)
 {
-	uint32_t rsp_ready = 0;
-	uint32_t i;
+	uint32_t num_pending = 0;
+
+	if (!dmub->hw_init)
+		return DMUB_STATUS_INVALID;
 
+	if (dmub->power_state != DMUB_POWER_STATE_D0)
+		return DMUB_STATUS_POWER_STATE_D3;
+
+	if (!dmub->hw_funcs.send_reg_inbox0_cmd_msg ||
+			!dmub->hw_funcs.clear_reg_inbox0_rsp_int_ack)
+		return DMUB_STATUS_INVALID;
+
+	if (dmub->reg_inbox0.num_submitted >= dmub->reg_inbox0.num_reported)
+		num_pending = dmub->reg_inbox0.num_submitted - dmub->reg_inbox0.num_reported;
+	else
+		/* num_submitted wrapped */
+		num_pending = DMUB_REG_INBOX0_RB_MAX_ENTRY -
+				(dmub->reg_inbox0.num_reported - dmub->reg_inbox0.num_submitted);
+
+	if (num_pending >= DMUB_REG_INBOX0_RB_MAX_ENTRY)
+		return DMUB_STATUS_QUEUE_FULL;
+
+	/* clear last rsp ack and send message */
+	dmub->hw_funcs.clear_reg_inbox0_rsp_int_ack(dmub);
 	dmub->hw_funcs.send_reg_inbox0_cmd_msg(dmub, cmd);
 
-	for (i = 0; i < timeout_us; i++) {
-		rsp_ready = dmub->hw_funcs.read_reg_inbox0_rsp_int_status(dmub);
-		if (rsp_ready)
-			break;
-		udelay(1);
+	dmub->reg_inbox0.num_submitted++;
+	dmub->reg_inbox0.is_pending = true;
+	dmub->reg_inbox0.is_multi_pending = cmd->cmd_common.header.multi_cmd_pending;
+
+	return DMUB_STATUS_OK;
+}
+
+void dmub_srv_cmd_get_response(struct dmub_srv *dmub,
+		union dmub_rb_cmd *cmd_rsp)
+{
+	if (dmub) {
+		if (dmub->inbox_type == DMUB_CMD_INTERFACE_REG &&
+				dmub->hw_funcs.read_reg_inbox0_cmd_rsp) {
+			dmub->hw_funcs.read_reg_inbox0_cmd_rsp(dmub, cmd_rsp);
+		} else {
+			dmub_rb_get_return_data(&dmub->inbox1.rb, cmd_rsp);
+		}
 	}
-	if (rsp_ready == 0)
-		return DMUB_STATUS_TIMEOUT;
+}
+
+static enum dmub_status dmub_srv_sync_reg_inbox0(struct dmub_srv *dmub)
+{
+	if (!dmub || !dmub->sw_init)
+		return DMUB_STATUS_INVALID;
 
-	if (with_reply)
-		dmub->hw_funcs.read_reg_inbox0_cmd_rsp(dmub, cmd);
+	dmub->reg_inbox0.is_pending = 0;
+	dmub->reg_inbox0.is_multi_pending = 0;
 
-	dmub->hw_funcs.write_reg_inbox0_rsp_int_ack(dmub);
+	return DMUB_STATUS_OK;
+}
 
-	/* wait for rsp int status is cleared to initial state before exit */
-	for (; i <= timeout_us; i++) {
-		rsp_ready = dmub->hw_funcs.read_reg_inbox0_rsp_int_status(dmub);
-		if (rsp_ready == 0)
-			break;
-		udelay(1);
+static enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub)
+{
+	if (!dmub->sw_init)
+		return DMUB_STATUS_INVALID;
+
+	if (dmub->hw_funcs.get_inbox1_rptr && dmub->hw_funcs.get_inbox1_wptr) {
+		uint32_t rptr = dmub->hw_funcs.get_inbox1_rptr(dmub);
+		uint32_t wptr = dmub->hw_funcs.get_inbox1_wptr(dmub);
+
+		if (rptr > dmub->inbox1.rb.capacity || wptr > dmub->inbox1.rb.capacity) {
+			return DMUB_STATUS_HW_FAILURE;
+		} else {
+			dmub->inbox1.rb.rptr = rptr;
+			dmub->inbox1.rb.wrpt = wptr;
+			dmub->inbox1_last_wptr = dmub->inbox1.rb.wrpt;
+		}
 	}
-	ASSERT(rsp_ready == 0);
 
 	return DMUB_STATUS_OK;
 }
 
-void dmub_srv_set_power_state(struct dmub_srv *dmub, enum dmub_srv_power_state_type dmub_srv_power_state)
+enum dmub_status dmub_srv_sync_inboxes(struct dmub_srv *dmub)
 {
-	if (!dmub || !dmub->hw_init)
-		return;
+	enum dmub_status status;
 
-	dmub->power_state = dmub_srv_power_state;
+	status = dmub_srv_sync_reg_inbox0(dmub);
+	if (status != DMUB_STATUS_OK)
+		return status;
+
+	status = dmub_srv_sync_inbox1(dmub);
+	if (status != DMUB_STATUS_OK)
+		return status;
+
+	return DMUB_STATUS_OK;
+}
+
+enum dmub_status dmub_srv_wait_for_inbox_free(struct dmub_srv *dmub,
+		uint32_t timeout_us,
+		uint32_t num_free_required)
+{
+	enum dmub_status status;
+	uint32_t i;
+	const uint32_t polling_interval_us = 1;
+
+	if (!dmub->hw_init)
+		return DMUB_STATUS_INVALID;
+
+	for (i = 0; i < timeout_us; i += polling_interval_us) {
+		status = dmub_srv_update_inbox_status(dmub);
+
+		if (status != DMUB_STATUS_OK)
+			return status;
+
+		/* check for space in inbox1 */
+		if (dmub_rb_num_free(&dmub->inbox1.rb) >= num_free_required)
+			return DMUB_STATUS_OK;
+
+		udelay(polling_interval_us);
+	}
+
+	return DMUB_STATUS_TIMEOUT;
+}
+
+enum dmub_status dmub_srv_update_inbox_status(struct dmub_srv *dmub)
+{
+	uint32_t rptr;
+
+	if (!dmub->hw_init)
+		return DMUB_STATUS_INVALID;
+
+	if (dmub->power_state != DMUB_POWER_STATE_D0)
+		return DMUB_STATUS_POWER_STATE_D3;
+
+	/* update inbox1 state */
+	rptr = dmub->hw_funcs.get_inbox1_rptr(dmub);
+
+	if (rptr > dmub->inbox1.rb.capacity)
+		return DMUB_STATUS_HW_FAILURE;
+
+	if (dmub->inbox1.rb.rptr > rptr) {
+		/* rb wrapped */
+		dmub->inbox1.num_reported += (rptr + dmub->inbox1.rb.capacity - dmub->inbox1.rb.rptr) / DMUB_RB_CMD_SIZE;
+	} else {
+		dmub->inbox1.num_reported += (rptr - dmub->inbox1.rb.rptr) / DMUB_RB_CMD_SIZE;
+	}
+	dmub->inbox1.rb.rptr = rptr;
+
+	/* update reg_inbox0 */
+	dmub_srv_update_reg_inbox0_status(dmub);
+
+	return DMUB_STATUS_OK;
 }
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c
index cce887cefc01..567c5b1aeb7a 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c
@@ -95,23 +95,6 @@ enum dmub_status dmub_srv_stat_get_notification(struct dmub_srv *dmub,
 	case DMUB_OUT_CMD__DPIA_NOTIFICATION:
 		notify->type = DMUB_NOTIFICATION_DPIA_NOTIFICATION;
 		notify->link_index = cmd.dpia_notification.payload.header.instance;
-
-		if (cmd.dpia_notification.payload.header.type == DPIA_NOTIFY__BW_ALLOCATION) {
-
-			notify->dpia_notification.payload.data.dpia_bw_alloc.estimated_bw =
-					cmd.dpia_notification.payload.data.dpia_bw_alloc.estimated_bw;
-			notify->dpia_notification.payload.data.dpia_bw_alloc.allocated_bw =
-					cmd.dpia_notification.payload.data.dpia_bw_alloc.allocated_bw;
-
-			if (cmd.dpia_notification.payload.data.dpia_bw_alloc.bits.bw_request_failed)
-				notify->result = DPIA_BW_REQ_FAILED;
-			else if (cmd.dpia_notification.payload.data.dpia_bw_alloc.bits.bw_request_succeeded)
-				notify->result = DPIA_BW_REQ_SUCCESS;
-			else if (cmd.dpia_notification.payload.data.dpia_bw_alloc.bits.est_bw_changed)
-				notify->result = DPIA_EST_BW_CHANGED;
-			else if (cmd.dpia_notification.payload.data.dpia_bw_alloc.bits.bw_alloc_cap_changed)
-				notify->result = DPIA_BW_ALLOC_CAPS_CHANGED;
-		}
 		break;
 	case DMUB_OUT_CMD__HPD_SENSE_NOTIFY:
 		notify->type = DMUB_NOTIFICATION_HPD_SENSE_NOTIFY;
@@ -119,6 +102,10 @@ enum dmub_status dmub_srv_stat_get_notification(struct dmub_srv *dmub,
 			    &cmd.hpd_sense_notify.data,
 			    sizeof(cmd.hpd_sense_notify.data));
 		break;
+	case DMUB_OUT_CMD__FUSED_IO:
+		notify->type = DMUB_NOTIFICATION_FUSED_IO;
+		dmub_memcpy(&notify->fused_request, &cmd.fused_io.request, sizeof(cmd.fused_io.request));
+		break;
 	default:
 		notify->type = DMUB_NOTIFICATION_NO_DATA;
 		break;
diff --git a/drivers/gpu/drm/amd/display/include/gpio_service_interface.h b/drivers/gpu/drm/amd/display/include/gpio_service_interface.h
index 7e3240e73c1f..63813009a3a6 100644
--- a/drivers/gpu/drm/amd/display/include/gpio_service_interface.h
+++ b/drivers/gpu/drm/amd/display/include/gpio_service_interface.h
@@ -86,6 +86,9 @@ enum dc_irq_source dal_irq_get_source(
 enum dc_irq_source dal_irq_get_rx_source(
 	const struct gpio *irq);
 
+enum dc_irq_source dal_irq_get_read_request(
+	const struct gpio *irq);
+
 enum gpio_result dal_irq_setup_hpd_filter(
 	struct gpio *irq,
 	struct gpio_hpd_config *config);
diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h
index 1867aac57cf2..da74ed66c8f9 100644
--- a/drivers/gpu/drm/amd/display/include/link_service_types.h
+++ b/drivers/gpu/drm/amd/display/include/link_service_types.h
@@ -89,6 +89,8 @@ struct link_training_settings {
 	bool enhanced_framing;
 	enum lttpr_mode lttpr_mode;
 
+	bool lttpr_early_tps2;
+
 	/* disallow different lanes to have different lane settings */
 	bool disallow_per_lane_settings;
 	/* dpcd lane settings will always use the same hw lane settings
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
index 55c7d873175f..a37634942b07 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h
@@ -386,6 +386,7 @@ enum mod_hdcp_status mod_hdcp_write_repeater_auth_ack(struct mod_hdcp *hdcp);
 enum mod_hdcp_status mod_hdcp_write_stream_manage(struct mod_hdcp *hdcp);
 enum mod_hdcp_status mod_hdcp_write_content_type(struct mod_hdcp *hdcp);
 enum mod_hdcp_status mod_hdcp_clear_cp_irq_status(struct mod_hdcp *hdcp);
+enum mod_hdcp_status mod_hdcp_write_poll_read_lc_fw(struct mod_hdcp *hdcp);
 
 /* hdcp version helpers */
 static inline uint8_t is_dp_hdcp(struct mod_hdcp *hdcp)
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c
index 1d41dd58f6bc..bb8ae80b37f8 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c
@@ -452,21 +452,12 @@ out:
 	return status;
 }
 
-static enum mod_hdcp_status locality_check(struct mod_hdcp *hdcp,
+static enum mod_hdcp_status locality_check_sw(struct mod_hdcp *hdcp,
 		struct mod_hdcp_event_context *event_ctx,
 		struct mod_hdcp_transition_input_hdcp2 *input)
 {
 	enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS;
 
-	if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) {
-		event_ctx->unexpected_event = 1;
-		goto out;
-	}
-
-	if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_prepare_lc_init,
-			&input->lc_init_prepare, &status,
-			hdcp, "lc_init_prepare"))
-		goto out;
 	if (!mod_hdcp_execute_and_set(mod_hdcp_write_lc_init,
 			&input->lc_init_write, &status,
 			 hdcp, "lc_init_write"))
@@ -482,6 +473,48 @@ static enum mod_hdcp_status locality_check(struct mod_hdcp *hdcp,
 			&input->l_prime_read, &status,
 			hdcp, "l_prime_read"))
 		goto out;
+out:
+	return status;
+}
+
+static enum mod_hdcp_status locality_check_fw(struct mod_hdcp *hdcp,
+		struct mod_hdcp_event_context *event_ctx,
+		struct mod_hdcp_transition_input_hdcp2 *input)
+{
+	enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS;
+
+	if (!mod_hdcp_execute_and_set(mod_hdcp_write_poll_read_lc_fw,
+			&input->l_prime_read, &status,
+			hdcp, "l_prime_read"))
+		goto out;
+
+out:
+	return status;
+}
+
+static enum mod_hdcp_status locality_check(struct mod_hdcp *hdcp,
+		struct mod_hdcp_event_context *event_ctx,
+		struct mod_hdcp_transition_input_hdcp2 *input)
+{
+	enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS;
+	const bool use_fw = hdcp->config.ddc.funcs.atomic_write_poll_read_i2c
+			&& hdcp->config.ddc.funcs.atomic_write_poll_read_aux
+			&& !hdcp->connection.link.adjust.hdcp2.force_sw_locality_check;
+
+	if (event_ctx->event != MOD_HDCP_EVENT_CALLBACK) {
+		event_ctx->unexpected_event = 1;
+		goto out;
+	}
+
+	if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_prepare_lc_init,
+			&input->lc_init_prepare, &status,
+			hdcp, "lc_init_prepare"))
+		goto out;
+
+	status = (use_fw ? locality_check_fw : locality_check_sw)(hdcp, event_ctx, input);
+	if (status != MOD_HDCP_STATUS_SUCCESS)
+		goto out;
+
 	if (!mod_hdcp_execute_and_set(mod_hdcp_hdcp2_validate_l_prime,
 			&input->l_prime_validation, &status,
 			hdcp, "l_prime_validation"))
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c
index c5f6c11de7e5..89ffb89e1932 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_transition.c
@@ -184,17 +184,28 @@ enum mod_hdcp_status mod_hdcp_hdcp2_transition(struct mod_hdcp *hdcp,
 		callback_in_ms(0, output);
 		set_state_id(hdcp, output, H2_A2_LOCALITY_CHECK);
 		break;
-	case H2_A2_LOCALITY_CHECK:
+	case H2_A2_LOCALITY_CHECK: {
+		const bool use_fw = hdcp->config.ddc.funcs.atomic_write_poll_read_i2c
+				&& !adjust->hdcp2.force_sw_locality_check;
+
+		/*
+		 * 1A-05: consider disconnection after LC init a failure
+		 * 1A-13-1: consider invalid l' a failure
+		 * 1A-13-2: consider l' timeout a failure
+		 */
 		if (hdcp->state.stay_count > 10 ||
 				input->lc_init_prepare != PASS ||
-				input->lc_init_write != PASS ||
-				input->l_prime_available_poll != PASS ||
-				input->l_prime_read != PASS) {
-			/*
-			 * 1A-05: consider disconnection after LC init a failure
-			 * 1A-13-1: consider invalid l' a failure
-			 * 1A-13-2: consider l' timeout a failure
-			 */
+				(!use_fw && input->lc_init_write != PASS) ||
+				(!use_fw && input->l_prime_available_poll != PASS)) {
+			fail_and_restart_in_ms(0, &status, output);
+			break;
+		} else if (input->l_prime_read != PASS) {
+			if (use_fw && hdcp->config.debug.lc_enable_sw_fallback) {
+				adjust->hdcp2.force_sw_locality_check = true;
+				callback_in_ms(0, output);
+				break;
+			}
+
 			fail_and_restart_in_ms(0, &status, output);
 			break;
 		} else if (input->l_prime_validation != PASS) {
@@ -205,6 +216,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_transition(struct mod_hdcp *hdcp,
 		callback_in_ms(0, output);
 		set_state_id(hdcp, output, H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER);
 		break;
+	}
 	case H2_A3_EXCHANGE_KS_AND_TEST_FOR_REPEATER:
 		if (input->eks_prepare != PASS ||
 				input->eks_write != PASS) {
@@ -498,14 +510,25 @@ enum mod_hdcp_status mod_hdcp_hdcp2_dp_transition(struct mod_hdcp *hdcp,
 		callback_in_ms(0, output);
 		set_state_id(hdcp, output, D2_A2_LOCALITY_CHECK);
 		break;
-	case D2_A2_LOCALITY_CHECK:
+	case D2_A2_LOCALITY_CHECK: {
+		const bool use_fw = hdcp->config.ddc.funcs.atomic_write_poll_read_aux
+				&& !adjust->hdcp2.force_sw_locality_check;
+
 		if (hdcp->state.stay_count > 10 ||
 				input->lc_init_prepare != PASS ||
-				input->lc_init_write != PASS ||
-				input->l_prime_read != PASS) {
+				(!use_fw && input->lc_init_write != PASS)) {
 			/* 1A-12: consider invalid l' a failure */
 			fail_and_restart_in_ms(0, &status, output);
 			break;
+		} else if (input->l_prime_read != PASS) {
+			if (use_fw && hdcp->config.debug.lc_enable_sw_fallback) {
+				adjust->hdcp2.force_sw_locality_check = true;
+				callback_in_ms(0, output);
+				break;
+			}
+
+			fail_and_restart_in_ms(0, &status, output);
+			break;
 		} else if (input->l_prime_validation != PASS) {
 			callback_in_ms(0, output);
 			increment_stay_counter(hdcp);
@@ -514,6 +537,7 @@ enum mod_hdcp_status mod_hdcp_hdcp2_dp_transition(struct mod_hdcp *hdcp,
 		callback_in_ms(0, output);
 		set_state_id(hdcp, output, D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER);
 		break;
+	}
 	case D2_A34_EXCHANGE_KS_AND_TEST_FOR_REPEATER:
 		if (input->eks_prepare != PASS ||
 				input->eks_write != PASS) {
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c
index 6e064e6ae949..2e6408579194 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c
@@ -688,3 +688,76 @@ enum mod_hdcp_status mod_hdcp_clear_cp_irq_status(struct mod_hdcp *hdcp)
 
 	return MOD_HDCP_STATUS_INVALID_OPERATION;
 }
+
+static bool write_stall_read_lc_fw_aux(struct mod_hdcp *hdcp)
+{
+	struct mod_hdcp_message_hdcp2 *hdcp2 = &hdcp->auth.msg.hdcp2;
+
+	struct mod_hdcp_atomic_op_aux write = {
+		hdcp_dpcd_addrs[MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT],
+		hdcp2->lc_init + 1,
+		sizeof(hdcp2->lc_init) - 1,
+	};
+	struct mod_hdcp_atomic_op_aux stall = { 0, NULL, 0, };
+	struct mod_hdcp_atomic_op_aux read = {
+		hdcp_dpcd_addrs[MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME],
+		hdcp2->lc_l_prime + 1,
+		sizeof(hdcp2->lc_l_prime) - 1,
+	};
+
+	hdcp2->lc_l_prime[0] = HDCP_2_2_LC_SEND_LPRIME;
+
+	return hdcp->config.ddc.funcs.atomic_write_poll_read_aux(
+			hdcp->config.ddc.handle,
+			&write,
+			&stall,
+			&read,
+			16 * 1000,
+			0
+	);
+}
+
+static bool write_poll_read_lc_fw_i2c(struct mod_hdcp *hdcp)
+{
+	struct mod_hdcp_message_hdcp2 *hdcp2 = &hdcp->auth.msg.hdcp2;
+	uint8_t expected_rxstatus[2] = { sizeof(hdcp2->lc_l_prime) };
+
+	hdcp->buf[0] = hdcp_i2c_offsets[MOD_HDCP_MESSAGE_ID_WRITE_LC_INIT];
+	memmove(&hdcp->buf[1], hdcp2->lc_init, sizeof(hdcp2->lc_init));
+
+	struct mod_hdcp_atomic_op_i2c write = {
+		HDCP_I2C_ADDR,
+		0,
+		hdcp->buf,
+		sizeof(hdcp2->lc_init) + 1,
+	};
+	struct mod_hdcp_atomic_op_i2c poll = {
+		HDCP_I2C_ADDR,
+		hdcp_i2c_offsets[MOD_HDCP_MESSAGE_ID_READ_RXSTATUS],
+		expected_rxstatus,
+		sizeof(expected_rxstatus),
+	};
+	struct mod_hdcp_atomic_op_i2c read = {
+		HDCP_I2C_ADDR,
+		hdcp_i2c_offsets[MOD_HDCP_MESSAGE_ID_READ_LC_SEND_L_PRIME],
+		hdcp2->lc_l_prime,
+		sizeof(hdcp2->lc_l_prime),
+	};
+
+	return hdcp->config.ddc.funcs.atomic_write_poll_read_i2c(
+			hdcp->config.ddc.handle,
+			&write,
+			&poll,
+			&read,
+			20 * 1000,
+			6
+	);
+}
+
+enum mod_hdcp_status mod_hdcp_write_poll_read_lc_fw(struct mod_hdcp *hdcp)
+{
+	const bool success = (is_dp_hdcp(hdcp) ? write_stall_read_lc_fw_aux : write_poll_read_lc_fw_i2c)(hdcp);
+
+	return success ? MOD_HDCP_STATUS_SUCCESS : MOD_HDCP_STATUS_DDC_FAILURE;
+}
+
diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h
index a4d344a4db9e..c42468bb70ac 100644
--- a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h
+++ b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h
@@ -133,9 +133,22 @@ enum mod_hdcp_display_disable_option {
 	MOD_HDCP_DISPLAY_DISABLE_ENCRYPTION,
 };
 
+struct mod_hdcp_atomic_op_i2c {
+	uint8_t address;
+	uint8_t offset;
+	uint8_t *data;
+	uint32_t size;
+};
+
+struct mod_hdcp_atomic_op_aux {
+	uint32_t address;
+	uint8_t *data;
+	uint32_t size;
+};
+
 struct mod_hdcp_ddc {
 	void *handle;
-	struct {
+	struct mod_hdcp_ddc_funcs {
 		bool (*read_i2c)(void *handle,
 				uint32_t address,
 				uint8_t offset,
@@ -153,6 +166,22 @@ struct mod_hdcp_ddc {
 				uint32_t address,
 				const uint8_t *data,
 				uint32_t size);
+		bool (*atomic_write_poll_read_i2c)(
+				void *handle,
+				const struct mod_hdcp_atomic_op_i2c *write,
+				const struct mod_hdcp_atomic_op_i2c *poll,
+				struct mod_hdcp_atomic_op_i2c *read,
+				uint32_t poll_timeout_us,
+				uint8_t poll_mask_msb
+		);
+		bool (*atomic_write_poll_read_aux)(
+				void *handle,
+				const struct mod_hdcp_atomic_op_aux *write,
+				const struct mod_hdcp_atomic_op_aux *poll,
+				struct mod_hdcp_atomic_op_aux *read,
+				uint32_t poll_timeout_us,
+				uint8_t poll_mask_msb
+		);
 	} funcs;
 };
 
@@ -185,7 +214,8 @@ struct mod_hdcp_link_adjustment_hdcp2 {
 	uint8_t force_type		: 2;
 	uint8_t force_no_stored_km	: 1;
 	uint8_t increase_h_prime_timeout: 1;
-	uint8_t reserved		: 3;
+	uint8_t force_sw_locality_check : 1;
+	uint8_t reserved		: 2;
 };
 
 struct mod_hdcp_link_adjustment {
@@ -272,6 +302,10 @@ struct mod_hdcp_display_query {
 struct mod_hdcp_config {
 	struct mod_hdcp_psp psp;
 	struct mod_hdcp_ddc ddc;
+	struct {
+		uint8_t lc_enable_sw_fallback : 1;
+		uint8_t reserved : 7;
+	} debug;
 	uint8_t index;
 };
 
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 4c95b885d1d0..c8eccee9b023 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -366,7 +366,7 @@ enum DC_DEBUG_MASK {
 	DC_HDCP_LC_FORCE_FW_ENABLE = 0x80000,
 
 	/**
-	 * @DC_HDCP_LC_ENABLE_SW_FALLBACK If set, upon HDCP Locality Check FW
+	 * @DC_HDCP_LC_ENABLE_SW_FALLBACK: If set, upon HDCP Locality Check FW
 	 * path failure, retry using legacy SW path.
 	 */
 	DC_HDCP_LC_ENABLE_SW_FALLBACK = 0x100000,
diff --git a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h
index bd8085ec54ed..2d6a598a6c25 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h
@@ -5242,6 +5242,8 @@
 #define DEGAMMA_CONTROL__CURSOR_DEGAMMA_MODE__SHIFT 0x0000000c
 #define DEGAMMA_CONTROL__GRPH_DEGAMMA_MODE_MASK 0x00000003L
 #define DEGAMMA_CONTROL__GRPH_DEGAMMA_MODE__SHIFT 0x00000000
+#define DEGAMMA_CONTROL__ICON_DEGAMMA_MODE_MASK 0x00000300L
+#define DEGAMMA_CONTROL__ICON_DEGAMMA_MODE__SHIFT 0x00000008
 #define DEGAMMA_CONTROL__OVL_DEGAMMA_MODE_MASK 0x00000030L
 #define DEGAMMA_CONTROL__OVL_DEGAMMA_MODE__SHIFT 0x00000004
 #define DENORM_CONTROL__DENORM_MODE_MASK 0x00000007L
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_6_0_d.h b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_6_0_d.h
index c75aee25619e..6f44345277af 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_6_0_d.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_6_0_d.h
@@ -1779,6 +1779,8 @@
 #define mmRLC_TTOP_D                                    0x3105
 #define mmRLC_CLEAR_STATE_RESTORE_BASE                  0x30C8
 #define mmRLC_PG_AO_CU_MASK                             0x310B
+#define mmSPI_STATIC_THREAD_MGMT_1			0x2438
+#define mmSPI_STATIC_THREAD_MGMT_2			0x2439
 #define mmSPI_STATIC_THREAD_MGMT_3                      0x243A
 
 #endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_d.h b/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_d.h
index edc8a793a95d..4dd386b98748 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_d.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_d.h
@@ -234,6 +234,26 @@
 #define mmIH_RB_WPTR_ADDR_HI 0x0F84
 #define mmIH_RB_WPTR_ADDR_LO 0x0F85
 #define mmIH_STATUS 0x0F88
+
+#define mmDMA_GFX_RB_CNTL                                       0x3400
+#define mmDMA_GFX_RB_BASE                                       0x3401
+#define mmDMA_GFX_RB_RPTR                                       0x3402
+#define mmDMA_GFX_RB_WPTR                                       0x3403
+#define mmDMA_GFX_RB_RPTR_ADDR_HI                               0x3407
+#define mmDMA_GFX_RB_RPTR_ADDR_LO                               0x3408
+#define mmDMA_GFX_IB_CNTL                                       0x3409
+#define mmDMA_GFX_IB_RPTR                                       0x340a
+#define mmDMA_CNTL                                          0x340b
+#define mmDMA_STATUS_REG                                    0x340D
+#define mmDMA_TILING_CONFIG  				  0x342E
+#define mmDMA_SEM_INCOMPLETE_TIMER_CNTL                     0x3411
+#define mmDMA_SEM_WAIT_FAIL_TIMER_CNTL                      0x3412
+#define mmDMA_POWER_CNTL					0x342F
+#define mmDMA_CLK_CTRL					0x3430
+#define mmDMA_PG						0x3435
+#define mmDMA_PGFSM_CONFIG				0x3436
+#define mmDMA_PGFSM_WRITE					0x3437
+
 #define mmSEM_MAILBOX 0x0F9B
 #define mmSEM_MAILBOX_CLIENTCONFIG 0x0F9A
 #define mmSEM_MAILBOX_CONTROL 0x0F9C
@@ -269,7 +289,4 @@
 #define mmVCE_CONFIG 0x0F94
 #define mmXDMA_MSTR_MEM_OVERFLOW_CNTL 0x03F8
 
-/* from the old sid.h */
-#define mmDMA_TILING_CONFIG 0x342E
-
 #endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_sh_mask.h
index 1c540fe136cb..9f7fc2428b69 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/oss/oss_1_0_sh_mask.h
@@ -823,6 +823,43 @@
 #define LX3__RESERVED__SHIFT 0x00000000
 #define RINGOSC_MASK__MASK_MASK 0x0000ffffL
 #define RINGOSC_MASK__MASK__SHIFT 0x00000000
+
+#define DMA_CNTL__TRAP_ENABLE_MASK 0x00000001L
+#define DMA_CNTL__TRAP_ENABLE__SHIFT 0x00000000
+#define DMA_CNTL__SEM_INCOMPLETE_INT_ENABLE_MASK 0x00000002L
+#define DMA_CNTL__SEM_INCOMPLETE_INT_ENABLE__SHIFT 0x00000001
+#define DMA_CNTL__SEM_WAIT_INT_ENABLE_MASK 0x00000004L
+#define DMA_CNTL__SEM_WAIT_INT_ENABLE__SHIFT 0x00000002
+#define DMA_CNTL__DATA_SWAP_ENABLE_MASK 0x00000008L
+#define DMA_CNTL__DATA_SWAP_ENABLE__SHIFT 0x00000003
+#define DMA_CNTL__FENCE_SWAP_ENABLE_MASK 0x00000010L
+#define DMA_CNTL__FENCE_SWAP_ENABLE__SHIFT 0x00000004
+#define DMA_CNTL__CTXEMPTY_INT_ENABLE_MASK 0x10000000L
+#define DMA_CNTL__CTXEMPTY_INT_ENABLE__SHIFT 0x0000001C
+#define DMA_GFX_RB_CNTL__RB_ENABLE_MASK 0x00000001L
+#define DMA_GFX_RB_CNTL__RB_ENABLE__SHIFT 0x00000000
+#define DMA_GFX_RB_CNTL__RB_SIZE__SHIFT 0x00000001
+#define DMA_GFX_RB_CNTL__RB_SWAP_ENABLE_MASK 0x00000200L
+#define DMA_GFX_RB_CNTL__RB_SWAP_ENABLE__SHIFT 0x00000009
+#define DMA_GFX_RB_CNTL__RPTR_WRITEBACK_ENABLE_MASK 0x00001000L
+#define DMA_GFX_RB_CNTL__RPTR_WRITEBACK_ENABLE__SHIFT 0x0000000C
+#define DMA_GFX_RB_CNTL__RPTR_WRITEBACK_SWAP_ENABLE_MASK 0x00002000L
+#define DMA_GFX_RB_CNTL__RPTR_WRITEBACK_SWAP_ENABLE__SHIFT 0x0000000D
+#define DMA_GFX_RB_CNTL__RPTR_WRITEBACK_TIMER__SHIFT 0x00000010
+#define DMA_GFX_IB_CNTL__IB_ENABLE_MASK 0x00000001L
+#define DMA_GFX_IB_CNTL__IB_ENABLE__SHIFT 0x00000000
+#define DMA_GFX_IB_CNTL__IB_SWAP_ENABLE_MASK 0x00000010L
+#define DMA_GFX_IB_CNTL__IB_SWAP_ENABLE__SHIFT 0x00000004
+#define DMA_GFX_IB_CNTL__CMD_VMID_FORCE_MASK 0x80000000L
+#define DMA_GFX_IB_CNTL__CMD_VMID_FORCE__SHIFT 0x0000001F
+
+#define DMA_STATUS_REG__IDLE_MASK 0x00000001L
+#define DMA_STATUS_REG__IDLE__SHIFT 0x00000000
+#define DMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK 0x00000100L
+#define DMA_POWER_CNTL__MEM_POWER_OVERRIDE__SHIFT 0x00000008
+#define DMA_PG__PG_CNTL_ENABLE_MASK 0x00000001L
+#define DMA_PG__PG_CNTL_ENABLE__SHIFT 0x00000000
+
 #define SEM_MAILBOX_CLIENTCONFIG__CP_CLIENT0_MASK 0x00000007L
 #define SEM_MAILBOX_CLIENTCONFIG__CP_CLIENT0__SHIFT 0x00000000
 #define SEM_MAILBOX_CLIENTCONFIG__CP_CLIENT1_MASK 0x00000038L
@@ -1015,6 +1052,10 @@
 #define SRBM_STATUS2__VCE_BUSY__SHIFT 0x00000007
 #define SRBM_STATUS2__VCE_RQ_PENDING_MASK 0x00000008L
 #define SRBM_STATUS2__VCE_RQ_PENDING__SHIFT 0x00000003
+#define SRBM_STATUS2__DMA_BUSY_MASK 0x00000020L
+#define SRBM_STATUS2__DMA_BUSY__SHIFT 0x00000005
+#define SRBM_STATUS2__DMA1_BUSY_MASK 0x00000040L
+#define SRBM_STATUS2__DMA1_BUSY__SHIFT 0x00000006
 #define SRBM_STATUS2__XDMA_BUSY_MASK 0x00000100L
 #define SRBM_STATUS2__XDMA_BUSY__SHIFT 0x00000008
 #define SRBM_STATUS2__XSP_BUSY_MASK 0x00000010L
diff --git a/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_d.h b/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_d.h
index 6b10be61efc3..bdef1f743df7 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_d.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_d.h
@@ -41,7 +41,49 @@
 #define ixLCAC_MC5_CNTL 0x012B
 #define ixLCAC_MC5_OVR_SEL 0x012C
 #define ixLCAC_MC5_OVR_VAL 0x012D
+
+#define mmCG_SPLL_FUNC_CNTL				0x0180
+#define mmCG_SPLL_FUNC_CNTL_2				0x0181
+#define mmCG_SPLL_FUNC_CNTL_3				0x0182
+#define mmCG_SPLL_FUNC_CNTL_4				0x0183
+#define mmCG_SPLL_STATUS					0x0185
+#define mmSPLL_CNTL_MODE					0x0186
+#define mmCG_SPLL_SPREAD_SPECTRUM				0x0188
+#define mmCG_SPLL_SPREAD_SPECTRUM_2			0x0189
+#define mmCG_SPLL_AUTOSCALE_CNTL				0x018B
+#define mmMPLL_BYPASSCLK_SEL				0x0197
+#define mmCG_CLKPIN_CNTL                                    0x0198
+#define mmCG_CLKPIN_CNTL_2                                  0x0199
+#define mmTHM_CLK_CNTL					0x019B
+#define mmMISC_CLK_CNTL					0x019C
+#define mmCG_THERMAL_CTRL					0x01C0
+#define mmCG_THERMAL_STATUS				0x01C1
+#define mmCG_THERMAL_INT					0x01C2
+#define mmCG_MULT_THERMAL_CTRL					0x01C4
+#define mmCG_MULT_THERMAL_STATUS					0x01C5
+#define mmCG_FDO_CTRL0					0x01D5
+#define mmCG_FDO_CTRL1					0x01D6
+#define mmCG_FDO_CTRL2					0x01D7
+#define mmCG_TACH_CTRL                                    0x01DC
+#define mmCG_TACH_STATUS                                  0x01DD
+#define mmGENERAL_PWRMGT                                  0x1E0
+#define mmCG_TPC                                            0x1E1
+#define mmSCLK_PWRMGT_CNTL                                  0x1E2
+#define mmTARGET_AND_CURRENT_PROFILE_INDEX                  0x01E6
+#define mmCG_FTV                                            0x01EF
+#define mmCG_FFCT_0                                         0x01F0
+#define mmCG_BSP                                          0x01FF
+#define mmCG_AT                                           0x0200
+#define mmCG_GIT                                          0x0201
+#define mmCG_SSP                                            0x0203
+#define mmCG_DISPLAY_GAP_CNTL                               0x020A
+#define mmCG_ULV_CONTROL					0x021E
+#define mmCG_ULV_PARAMETER				0x021F
+#define mmSMC_SCRATCH0					0x0221
+#define mmCG_CAC_CTRL					0x022E
+
 #define ixSMC_PC_C 0x80000370
+
 #define ixTHM_TMON0_DEBUG 0x03F0
 #define ixTHM_TMON0_INT_DATA 0x0380
 #define ixTHM_TMON0_RDIL0_DATA 0x0300
@@ -110,6 +152,7 @@
 #define ixTHM_TMON1_RDIR7_DATA 0x0337
 #define ixTHM_TMON1_RDIR8_DATA 0x0338
 #define ixTHM_TMON1_RDIR9_DATA 0x0339
+
 #define mmGPIOPAD_A 0x05E7
 #define mmGPIOPAD_EN 0x05E8
 #define mmGPIOPAD_EXTERN_TRIG_CNTL 0x05F1
@@ -127,6 +170,7 @@
 #define mmGPIOPAD_STRENGTH 0x05E5
 #define mmGPIOPAD_SW_INT_STAT 0x05E4
 #define mmGPIOPAD_Y 0x05E9
+
 #define mmSMC_IND_ACCESS_CNTL 0x008A
 #define mmSMC_IND_DATA_0 0x0081
 #define mmSMC_IND_DATA 0x0081
diff --git a/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_sh_mask.h
index 7d3925b7266e..67d3c7e13a48 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/smu/smu_6_0_sh_mask.h
@@ -23,10 +23,142 @@
 #ifndef SMU_6_0_SH_MASK_H
 #define SMU_6_0_SH_MASK_H
 
-#define CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV_MASK 0x03ffffffL
-#define CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV__SHIFT 0x00000000
-#define CG_SPLL_FUNC_CNTL__SPLL_REF_DIV_MASK 0x000003f0L
+#define CG_AT__CG_R_MASK 0x0000FFFFL
+#define CG_AT__CG_R__SHIFT 0x00000000
+#define CG_AT__CG_L_MASK 0xFFFF0000L
+#define CG_AT__CG_L__SHIFT 0x00000010
+
+#define CG_BSP__BSP_MASK 0x0000FFFFL
+#define CG_BSP__BSP__SHIFT 0x00000000
+#define CG_BSP__BSU_MASK 0x000F0000L
+#define CG_BSP__BSU__SHIFT 0x00000010
+
+#define CG_CAC_CTRL__CAC_WINDOW_MASK 0x00FFFFFFL
+#define CG_CAC_CTRL__CAC_WINDOW__SHIFT 0x00000000
+
+#define CG_CLKPIN_CNTL__XTALIN_DIVIDE_MASK 0x00000002L
+#define CG_CLKPIN_CNTL__XTALIN_DIVIDE__SHIFT 0x00000001
+#define CG_CLKPIN_CNTL__BCLK_AS_XCLK_MASK 0x00000004L
+#define CG_CLKPIN_CNTL__BCLK_AS_XCLK__SHIFT 0x00000002
+#define CG_CLKPIN_CNTL_2__FORCE_BIF_REFCLK_EN_MASK 0x00000008L
+#define CG_CLKPIN_CNTL_2__FORCE_BIF_REFCLK_EN__SHIFT 0x00000003
+#define CG_CLKPIN_CNTL_2__MUX_TCLK_TO_XCLK_MASK 0x00000100L
+#define CG_CLKPIN_CNTL_2__MUX_TCLK_TO_XCLK__SHIFT 0x00000008
+
+#define CG_DISPLAY_GAP_CNTL__DISP1_GAP_MASK 0x00000003L
+#define CG_DISPLAY_GAP_CNTL__DISP1_GAP__SHIFT 0x00000000
+#define CG_DISPLAY_GAP_CNTL__DISP2_GAP_MASK 0x0000000CL
+#define CG_DISPLAY_GAP_CNTL__DISP2_GAP__SHIFT 0x00000002
+#define CG_DISPLAY_GAP_CNTL__VBI_TIMER_COUNT_MASK 0x0003FFF0L
+#define CG_DISPLAY_GAP_CNTL__VBI_TIMER_COUNT__SHIFT 0x00000004
+#define CG_DISPLAY_GAP_CNTL__VBI_TIMER_UNIT_MASK 0x00700000
+#define CG_DISPLAY_GAP_CNTL__VBI_TIMER_UNIT__SHIFT 0x00000014
+#define CG_DISPLAY_GAP_CNTL__DISP1_GAP_MCHG_MASK 0x03000000L
+#define CG_DISPLAY_GAP_CNTL__DISP1_GAP_MCHG__SHIFT 0x00000018
+#define CG_DISPLAY_GAP_CNTL__DISP2_GAP_MCHG_MASK 0x0C000000L
+#define CG_DISPLAY_GAP_CNTL__DISP2_GAP_MCHG__SHIFT 0x0000001A
+
+#define CG_FFCT_0__UTC_0_MASK 0x000003FFL
+#define CG_FFCT_0__UTC_0__SHIFT 0x00000000
+#define CG_FFCT_0__DTC_0_MASK 0x000FFC00L
+#define CG_FFCT_0__DTC_0__SHIFT 0x0000000A
+
+#define CG_GIT__CG_GICST_MASK 0x0000FFFFL
+#define CG_GIT__CG_GICST__SHIFT 0x00000000
+#define CG_GIT__CG_GIPOT_MASK 0xFFFF0000L
+#define CG_GIT__CG_GIPOT__SHIFT 0x00000010
+
+#define CG_SPLL_FUNC_CNTL__SPLL_RESET_MASK 0x00000001L
+#define CG_SPLL_FUNC_CNTL__SPLL_RESET__SHIFT 0x00000000
+#define CG_SPLL_FUNC_CNTL__SPLL_SLEEP_MASK 0x00000002L
+#define CG_SPLL_FUNC_CNTL__SPLL_SLEEP__SHIFT 0x00000001
+#define CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK 0x00000008L
+#define CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN__SHIFT 0x00000003
+#define CG_SPLL_FUNC_CNTL__SPLL_REF_DIV_MASK 0x000003F0L
 #define CG_SPLL_FUNC_CNTL__SPLL_REF_DIV__SHIFT 0x00000004
+#define CG_SPLL_FUNC_CNTL__SPLL_PDIV_A_MASK 0x007F00000
+#define CG_SPLL_FUNC_CNTL__SPLL_PDIV_A__SHIFT 0x00000014
+#define CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL_MASK 0x0000001FF
+#define CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL__SHIFT 0x00000000
+#define CG_SPLL_FUNC_CNTL_2__SPLL_CTLREQ_CHG_MASK 0x00800000
+#define CG_SPLL_FUNC_CNTL_2__SPLL_CTLREQ_CHG__SHIFT 0x00000017
+#define CG_SPLL_FUNC_CNTL_2__SCLK_MUX_UPDATE_MASK 0x04000000
+#define CG_SPLL_FUNC_CNTL_2__SCLK_MUX_UPDATE__SHIFT 0x0000001A
+#define CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV_MASK 0x03FFFFFFL
+#define CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV__SHIFT 0x00000000
+#define CG_SPLL_FUNC_CNTL_3__SPLL_DITHEN_MASK 0x10000000L
+#define CG_SPLL_FUNC_CNTL_3__SPLL_DITHEN__SHIFT 0x0000001C
+#define CG_SPLL_STATUS__SPLL_CHG_STATUS_MASK 0x00000002L
+#define CG_SPLL_STATUS__SPLL_CHG_STATUS__SHIFT 0x00000001
+#define CG_SPLL_SPREAD_SPECTRUM__SSEN_MASK 0x00000001L
+#define CG_SPLL_SPREAD_SPECTRUM__SSEN__SHIFT 0x00000000
+#define CG_SPLL_SPREAD_SPECTRUM__CLK_S_MASK 0x0000FFF0L
+#define CG_SPLL_SPREAD_SPECTRUM__CLK_S__SHIFT 0x00000004
+#define CG_SPLL_SPREAD_SPECTRUM_2__CLK_V_MASK 0x00000200L
+#define CG_SPLL_SPREAD_SPECTRUM_2__CLK_V__SHIFT 0x00000000
+#define CG_SPLL_AUTOSCALE_CNTL__AUTOSCALE_ON_SS_CLEAR_MASK 0x03FFFFFFL
+#define CG_SPLL_AUTOSCALE_CNTL__AUTOSCALE_ON_SS_CLEAR__SHIFT 0x00000009
+
+#define CG_SSP__SST_MASK 0x0000FFFFL
+#define CG_SSP__SST__SHIFT 0x00000000
+#define CG_SSP__SSTU_MASK 0x000F0000L
+#define CG_SSP__SSTU__SHIFT 0x00000010
+
+#define CG_THERMAL_CTRL__DPM_EVENT_SRC_MASK 0x00000007L
+#define CG_THERMAL_CTRL__DPM_EVENT_SRC__SHIFT 0x00000000
+#define CG_THERMAL_CTRL__DIG_THERM_DPM_MASK 0x003FC000
+#define CG_THERMAL_CTRL__DIG_THERM_DPM__SHIFT 0x0000000E
+#define CG_THERMAL_STATUS__FDO_PWM_DUTY_MASK 0x0001FE00L
+#define CG_THERMAL_STATUS__FDO_PWM_DUTY__SHIFT 0x00000009
+#define CG_THERMAL_INT__DIG_THERM_INTH_MASK 0x0000FF00L
+#define CG_THERMAL_INT__DIG_THERM_INTH__SHIFT 0x00000008
+#define CG_THERMAL_INT__DIG_THERM_INTL_MASK 0x00FF0000L
+#define CG_THERMAL_INT__DIG_THERM_INTL__SHIFT 0x00000010
+#define CG_THERMAL_INT__THERM_INT_MASK_HIGH_MASK 0x01000000L
+#define CG_THERMAL_INT__THERM_INT_MASK_HIGH__SHIFT 0x00000018
+#define CG_THERMAL_INT__THERM_INT_MASK_LOW_MASK	0x02000000
+#define CG_THERMAL_INT__THERM_INT_MASK_LOW__SHIFT 0x00000019
+
+#define CG_MULT_THERMAL_CTRL__TEMP_SEL_MASK	0x0FF00000L
+#define CG_MULT_THERMAL_CTRL__TEMP_SEL__SHIFT 0x00000014
+#define CG_MULT_THERMAL_STATUS__ASIC_MAX_TEMP_MASK	0x000001FFL
+#define CG_MULT_THERMAL_STATUS__ASIC_MAX_TEMP__SHIFT 0x00000000
+#define CG_MULT_THERMAL_STATUS__CTF_TEMP_MASK	0x0003fe00L
+#define CG_MULT_THERMAL_STATUS__CTF_TEMP__SHIFT 0x00000009
+
+#define CG_FDO_CTRL0__FDO_STATIC_DUTY_MASK	0x000000FFL
+#define CG_FDO_CTRL0__FDO_STATIC_DUTY__SHIFT 0x00000000
+#define CG_FDO_CTRL1__FMAX_DUTY100_MASK	0x000000FFL
+#define CG_FDO_CTRL1__FMAX_DUTY100__SHIFT 0x00000000
+#define CG_FDO_CTRL2__TMIN_MASK	0x000000FFL
+#define CG_FDO_CTRL2__TMIN__SHIFT 0x00000000
+#define CG_FDO_CTRL2__FDO_PWM_MODE_MASK	0x00003800L
+#define CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT 0x0000000B
+#define CG_FDO_CTRL2__TACH_PWM_RESP_RATE_MASK	0xFE000000L
+#define CG_FDO_CTRL2__TACH_PWM_RESP_RATE__SHIFT 0x00000019
+
+#define CG_TACH_CTRL__EDGE_PER_REV_MASK	0x00000007L
+#define CG_TACH_CTRL__EDGE_PER_REV__SHIFT 0x00000000
+#define CG_TACH_CTRL__TARGET_PERIOD_MASK	0xFFFFFFF8L
+#define CG_TACH_CTRL__TARGET_PERIOD__SHIFT 0x00000003
+#define CG_TACH_STATUS__TACH_PERIOD_MASK	0xFFFFFFFFL
+#define CG_TACH_STATUS__TACH_PERIOD__SHIFT 0x00000000
+
+#define GENERAL_PWRMGT__GLOBAL_PWRMGT_EN_MASK	0x00000001L
+#define GENERAL_PWRMGT__GLOBAL_PWRMGT_EN__SHIFT 0x00000000
+#define GENERAL_PWRMGT__STATIC_PM_EN_MASK	0x00000002L
+#define GENERAL_PWRMGT__STATIC_PM_EN__SHIFT 0x00000001
+#define GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK	0x00000004L
+#define GENERAL_PWRMGT__THERMAL_PROTECTION_DIS__SHIFT 0x00000002
+#define GENERAL_PWRMGT__THERMAL_PROTECTION_TYPE_MASK	0x00000008L
+#define GENERAL_PWRMGT__THERMAL_PROTECTION_TYPE__SHIFT 0x00000003
+#define GENERAL_PWRMGT__SW_SMIO_INDEX_MASK	0x00000040L
+#define GENERAL_PWRMGT__SW_SMIO_INDEX__SHIFT 0x00000006
+#define GENERAL_PWRMGT__VOLT_PWRMGT_EN_MASK	0x00000400L
+#define GENERAL_PWRMGT__VOLT_PWRMGT_EN__SHIFT 0x0000000A
+#define GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN_MASK	0x00800000L
+#define GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN__SHIFT 0x00000017
+
 #define GPIOPAD_A__GPIO_A_MASK 0x7fffffffL
 #define GPIOPAD_A__GPIO_A__SHIFT 0x00000000
 #define GPIOPAD_EN__GPIO_EN_MASK 0x7fffffffL
@@ -195,6 +327,7 @@
 #define GPIOPAD_SW_INT_STAT__SW_INT_STAT__SHIFT 0x00000000
 #define GPIOPAD_Y__GPIO_Y_MASK 0x7fffffffL
 #define GPIOPAD_Y__GPIO_Y__SHIFT 0x00000000
+
 #define LCAC_MC0_CNTL__MC0_ENABLE_MASK 0x00000001L
 #define LCAC_MC0_CNTL__MC0_ENABLE__SHIFT 0x00000000
 #define LCAC_MC0_CNTL__MC0_THRESHOLD_MASK 0x0001fffeL
@@ -243,6 +376,37 @@
 #define LCAC_MC5_OVR_SEL__MC5_OVR_SEL__SHIFT 0x00000000
 #define LCAC_MC5_OVR_VAL__MC5_OVR_VAL_MASK 0xffffffffL
 #define LCAC_MC5_OVR_VAL__MC5_OVR_VAL__SHIFT 0x00000000
+
+#define MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL_MASK 0x0000FF00L
+#define MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL__SHIFT 0x00000008
+
+#define SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF_MASK	0x00000001L
+#define SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF__SHIFT 0x00000000
+#define SCLK_PWRMGT_CNTL__SCLK_LOW_D1_MASK	0x00000002L
+#define SCLK_PWRMGT_CNTL__SCLK_LOW_D1__SHIFT 0x00000001
+#define SCLK_PWRMGT_CNTL__FIR_RESET_MASK	0x00000010L
+#define SCLK_PWRMGT_CNTL__FIR_RESET__SHIFT 0x00000004
+#define SCLK_PWRMGT_CNTL__FIR_FORCE_TREND_SEL_MASK	0x00000020L
+#define SCLK_PWRMGT_CNTL__FIR_FORCE_TREND_SEL__SHIFT 0x00000005
+#define SCLK_PWRMGT_CNTL__FIR_TREND_MODE_MASK	0x00000040L
+#define SCLK_PWRMGT_CNTL__FIR_TREND_MODE__SHIFT 0x00000006
+#define SCLK_PWRMGT_CNTL__DYN_GFX_CLK_OFF_EN_MASK	0x00000080L
+#define SCLK_PWRMGT_CNTL__DYN_GFX_CLK_OFF_EN__SHIFT 0x00000007
+#define SCLK_PWRMGT_CNTL__GFX_CLK_FORCE_ON_MASK	0x00000100L
+#define SCLK_PWRMGT_CNTL__GFX_CLK_FORCE_ON__SHIFT 0x00000008
+#define SCLK_PWRMGT_CNTL__GFX_CLK_REQUEST_OFF_MASK	0x00000200L
+#define SCLK_PWRMGT_CNTL__GFX_CLK_REQUEST_OFF__SHIFT 0x00000009
+#define SCLK_PWRMGT_CNTL__GFX_CLK_FORCE_OFF_MASK	0x00000400L
+#define SCLK_PWRMGT_CNTL__GFX_CLK_FORCE_OFF__SHIFT 0x0000000A
+#define SCLK_PWRMGT_CNTL__GFX_CLK_OFF_ACPI_D1_MASK	0x00000800L
+#define SCLK_PWRMGT_CNTL__GFX_CLK_OFF_ACPI_D1__SHIFT 0x0000000B
+#define SCLK_PWRMGT_CNTL__GFX_CLK_OFF_ACPI_D2_MASK	0x00001000L
+#define SCLK_PWRMGT_CNTL__GFX_CLK_OFF_ACPI_D2__SHIFT 0x0000000C
+#define SCLK_PWRMGT_CNTL__GFX_CLK_OFF_ACPI_D3_MASK	0x00002000L
+#define SCLK_PWRMGT_CNTL__GFX_CLK_OFF_ACPI_D3__SHIFT 0x0000000D
+#define SCLK_PWRMGT_CNTL__DYN_LIGHT_SLEEP_EN_MASK	0x00004000L
+#define SCLK_PWRMGT_CNTL__DYN_LIGHT_SLEEP_EN__SHIFT 0x0000000E
+
 #define SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK 0x00000001L
 #define SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0__SHIFT 0x00000000
 #define SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_1_MASK 0x00000100L
@@ -285,6 +449,7 @@
 #define SMC_RESP_1__SMC_RESP__SHIFT 0x00000000
 #define SMC_RESP_2__SMC_RESP_MASK 0xffffffffL
 #define SMC_RESP_2__SMC_RESP__SHIFT 0x00000000
+
 #define SPLL_CNTL_MODE__SPLL_CTLREQ_DLY_CNT_MASK 0x000ff000L
 #define SPLL_CNTL_MODE__SPLL_CTLREQ_DLY_CNT__SHIFT 0x0000000c
 #define SPLL_CNTL_MODE__SPLL_ENSAT_MASK 0x00000010L
@@ -293,6 +458,8 @@
 #define SPLL_CNTL_MODE__SPLL_FASTEN__SHIFT 0x00000003
 #define SPLL_CNTL_MODE__SPLL_LEGACY_PDIV_MASK 0x00000002L
 #define SPLL_CNTL_MODE__SPLL_LEGACY_PDIV__SHIFT 0x00000001
+#define SPLL_CNTL_MODE__SPLL_REFCLK_SEL_MASK 0x0C000000L
+#define SPLL_CNTL_MODE__SPLL_REFCLK_SEL__SHIFT 0x0000001A
 #define SPLL_CNTL_MODE__SPLL_RESET_EN_MASK 0x10000000L
 #define SPLL_CNTL_MODE__SPLL_RESET_EN__SHIFT 0x0000001c
 #define SPLL_CNTL_MODE__SPLL_SW_DIR_CONTROL_MASK 0x00000001L
@@ -303,10 +470,25 @@
 #define SPLL_CNTL_MODE__SPLL_TEST__SHIFT 0x00000002
 #define SPLL_CNTL_MODE__SPLL_VCO_MODE_MASK 0x60000000L
 #define SPLL_CNTL_MODE__SPLL_VCO_MODE__SHIFT 0x0000001d
+
 #define TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX_MASK 0x0f000000L
 #define TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX__SHIFT 0x00000018
 #define TARGET_AND_CURRENT_PROFILE_INDEX_1__TARG_PCIE_INDEX_MASK 0xf0000000L
 #define TARGET_AND_CURRENT_PROFILE_INDEX_1__TARG_PCIE_INDEX__SHIFT 0x0000001c
+
+#define TARGET_AND_CURRENT_PROFILE_INDEX__CURRENT_STATE_INDEX_MASK 0x000000F0L
+#define TARGET_AND_CURRENT_PROFILE_INDEX__CURRENT_STATE_INDEX__SHIFT 0x00000004
+
+#define THM_CLK_CNTL__CMON_CLK_SEL_MASK 0x000000FFL
+#define THM_CLK_CNTL__CMON_CLK_SEL__SHIFT 0x00000000
+#define THM_CLK_CNTL__TMON_CLK_SEL_MASK 0x0000FF00L
+#define THM_CLK_CNTL__TMON_CLK_SEL__SHIFT 0x00000008
+
+#define MISC_CLK_CNTL__DEEP_SLEEP_CLK_SEL_MASK 0x000000FFL
+#define MISC_CLK_CNTL__DEEP_SLEEP_CLK_SEL__SHIFT 0x00000000
+#define MISC_CLK_CNTL__ZCLK_SEL_MASK 0x0000FF00L
+#define MISC_CLK_CNTL__ZCLK_SEL__SHIFT 0x00000008
+
 #define THM_TMON0_DEBUG__DEBUG_RDI_MASK 0x0000001fL
 #define THM_TMON0_DEBUG__DEBUG_RDI__SHIFT 0x00000000
 #define THM_TMON0_DEBUG__DEBUG_Z_MASK 0x0000ffe0L
diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_offset.h
index 14574112c469..c4aaa86a95e2 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_offset.h
@@ -1147,6 +1147,22 @@
 #define regUVD_DPG_LMA_CTL2_BASE_IDX                                                                    1
 
 
+// addressBlock: uvd_mmsch_dec
+// base address: 0x20d2c
+#define regMMSCH_VF_VMID                                                                                0x054b
+#define regMMSCH_VF_VMID_BASE_IDX                                                                       1
+#define regMMSCH_VF_CTX_ADDR_LO                                                                         0x054c
+#define regMMSCH_VF_CTX_ADDR_LO_BASE_IDX                                                                1
+#define regMMSCH_VF_CTX_ADDR_HI                                                                         0x054d
+#define regMMSCH_VF_CTX_ADDR_HI_BASE_IDX                                                                1
+#define regMMSCH_VF_CTX_SIZE                                                                            0x054e
+#define regMMSCH_VF_CTX_SIZE_BASE_IDX                                                                   1
+#define regMMSCH_VF_MAILBOX_HOST                                                                        0x0552
+#define regMMSCH_VF_MAILBOX_HOST_BASE_IDX                                                               1
+#define regMMSCH_VF_MAILBOX_RESP                                                                        0x0553
+#define regMMSCH_VF_MAILBOX_RESP_BASE_IDX                                                               1
+
+
 // addressBlock: uvd_vcn_umsch_dec
 // base address: 0x21500
 #define regVCN_UMSCH_MES_CNTL                                                                           0x0740
diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_sh_mask.h
index 5c119a6b87fb..bd7242e4e9c6 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_5_0_0_sh_mask.h
@@ -5929,6 +5929,29 @@
 #define UVD_DPG_LMA_CTL2__JPEG_WRITE_PTR_MASK                                                                 0x0000FE00L
 
 
+// addressBlock: uvd_mmsch_dec
+//MMSCH_VF_VMID
+#define MMSCH_VF_VMID__VF_CTX_VMID__SHIFT                                                                     0x0
+#define MMSCH_VF_VMID__VF_GPCOM_VMID__SHIFT                                                                   0x5
+#define MMSCH_VF_VMID__VF_CTX_VMID_MASK                                                                       0x0000001FL
+#define MMSCH_VF_VMID__VF_GPCOM_VMID_MASK                                                                     0x000003E0L
+//MMSCH_VF_CTX_ADDR_LO
+#define MMSCH_VF_CTX_ADDR_LO__VF_CTX_ADDR_LO__SHIFT                                                           0x6
+#define MMSCH_VF_CTX_ADDR_LO__VF_CTX_ADDR_LO_MASK                                                             0xFFFFFFC0L
+//MMSCH_VF_CTX_ADDR_HI
+#define MMSCH_VF_CTX_ADDR_HI__VF_CTX_ADDR_HI__SHIFT                                                           0x0
+#define MMSCH_VF_CTX_ADDR_HI__VF_CTX_ADDR_HI_MASK                                                             0xFFFFFFFFL
+//MMSCH_VF_CTX_SIZE
+#define MMSCH_VF_CTX_SIZE__VF_CTX_SIZE__SHIFT                                                                 0x0
+#define MMSCH_VF_CTX_SIZE__VF_CTX_SIZE_MASK                                                                   0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_HOST
+#define MMSCH_VF_MAILBOX_HOST__DATA__SHIFT                                                                    0x0
+#define MMSCH_VF_MAILBOX_HOST__DATA_MASK                                                                      0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_RESP
+#define MMSCH_VF_MAILBOX_RESP__RESP__SHIFT                                                                    0x0
+#define MMSCH_VF_MAILBOX_RESP__RESP_MASK                                                                      0xFFFFFFFFL
+
+
 // addressBlock: uvd_vcn_umsch_dec
 //VCN_UMSCH_MES_CNTL
 #define VCN_UMSCH_MES_CNTL__PIPE_ID__SHIFT                                                                    0x0
diff --git a/drivers/gpu/drm/amd/include/atombios.h b/drivers/gpu/drm/amd/include/atombios.h
index b78360a71bc9..52bac19fb404 100644
--- a/drivers/gpu/drm/amd/include/atombios.h
+++ b/drivers/gpu/drm/amd/include/atombios.h
@@ -4308,7 +4308,7 @@ typedef struct _ATOM_DPCD_INFO
 // note2: From RV770, the memory is more than 32bit addressable, so we will change
 //        ucTableFormatRevision=1,ucTableContentRevision=4, the strcuture remains
 //        exactly same as 1.1 and 1.2 (1.3 is never in use), but ulStartAddrUsedByFirmware
-//        (in offset to start of memory address) is KB aligned instead of byte aligend.
+//        (in offset to start of memory address) is KB aligned instead of byte aligned.
 // Note3:
 /* If we change usReserved to "usFBUsedbyDrvInKB", then to VBIOS this usFBUsedbyDrvInKB is a predefined, unchanged
 constant across VGA or non VGA adapter,
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h
index 0160d65f3f5e..2d1135bdc4b9 100644
--- a/drivers/gpu/drm/amd/include/atomfirmware.h
+++ b/drivers/gpu/drm/amd/include/atomfirmware.h
@@ -183,6 +183,7 @@ enum atom_dgpu_vram_type {
   ATOM_DGPU_VRAM_TYPE_HBM2E = 0x61,
   ATOM_DGPU_VRAM_TYPE_GDDR6 = 0x70,
   ATOM_DGPU_VRAM_TYPE_HBM3 = 0x80,
+	ATOM_DGPU_VRAM_TYPE_HBM3E = 0x81,
 };
 
 enum atom_dp_vs_preemph_def{
diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
index 2a9606118d89..0f7542d7074b 100644
--- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
@@ -128,6 +128,7 @@ enum amd_pp_sensors {
 	AMDGPU_PP_SENSOR_CPU_CLK,
 	AMDGPU_PP_SENSOR_VDDNB,
 	AMDGPU_PP_SENSOR_VDDGFX,
+	AMDGPU_PP_SENSOR_VDDBOARD,
 	AMDGPU_PP_SENSOR_UVD_VCLK,
 	AMDGPU_PP_SENSOR_UVD_DCLK,
 	AMDGPU_PP_SENSOR_VCE_ECCLK,
@@ -429,6 +430,7 @@ struct amd_pm_funcs {
 	int (*set_pp_table)(void *handle, const char *buf, size_t size);
 	void (*debugfs_print_current_performance_level)(void *handle, struct seq_file *m);
 	int (*switch_power_profile)(void *handle, enum PP_SMC_POWER_PROFILE type, bool en);
+	int (*pause_power_profile)(void *handle, bool pause);
 /* export to amdgpu */
 	struct amd_vce_state *(*get_vce_clock_state)(void *handle, u32 idx);
 	int (*dispatch_tasks)(void *handle, enum amd_pp_task task_id,
diff --git a/drivers/gpu/drm/amd/include/v11_structs.h b/drivers/gpu/drm/amd/include/v11_structs.h
index f8008270f813..3728389fc3be 100644
--- a/drivers/gpu/drm/amd/include/v11_structs.h
+++ b/drivers/gpu/drm/amd/include/v11_structs.h
@@ -535,8 +535,8 @@ struct v11_gfx_mqd {
 	uint32_t reserved_507; // offset: 507  (0x1FB)
 	uint32_t reserved_508; // offset: 508  (0x1FC)
 	uint32_t reserved_509; // offset: 509  (0x1FD)
-	uint32_t reserved_510; // offset: 510  (0x1FE)
-	uint32_t reserved_511; // offset: 511  (0x1FF)
+	uint32_t fence_address_lo; // offset: 510  (0x1FE)
+	uint32_t fence_address_hi; // offset: 511  (0x1FF)
 };
 
 struct v11_sdma_mqd {
@@ -1118,8 +1118,8 @@ struct v11_compute_mqd {
 	uint32_t reserved_443; // offset: 443  (0x1BB)
 	uint32_t reserved_444; // offset: 444  (0x1BC)
 	uint32_t reserved_445; // offset: 445  (0x1BD)
-	uint32_t reserved_446; // offset: 446  (0x1BE)
-	uint32_t reserved_447; // offset: 447  (0x1BF)
+	uint32_t fence_address_lo; // offset: 446  (0x1BE)
+	uint32_t fence_address_hi; // offset: 447  (0x1BF)
 	uint32_t gws_0_val; // offset: 448  (0x1C0)
 	uint32_t gws_1_val; // offset: 449  (0x1C1)
 	uint32_t gws_2_val; // offset: 450  (0x1C2)
diff --git a/drivers/gpu/drm/amd/include/v12_structs.h b/drivers/gpu/drm/amd/include/v12_structs.h
index 5eabab611b02..03a35f8a65b0 100644
--- a/drivers/gpu/drm/amd/include/v12_structs.h
+++ b/drivers/gpu/drm/amd/include/v12_structs.h
@@ -535,8 +535,8 @@ struct v12_gfx_mqd {
     uint32_t reserved_507; // offset: 507  (0x1FB)
     uint32_t reserved_508; // offset: 508  (0x1FC)
     uint32_t reserved_509; // offset: 509  (0x1FD)
-    uint32_t reserved_510; // offset: 510  (0x1FE)
-    uint32_t reserved_511; // offset: 511  (0x1FF)
+    uint32_t fence_address_lo; // offset: 510  (0x1FE)
+    uint32_t fence_address_hi; // offset: 511  (0x1FF)
 };
 
 struct v12_sdma_mqd {
@@ -1118,8 +1118,8 @@ struct v12_compute_mqd {
     uint32_t reserved_443; // offset: 443  (0x1BB)
     uint32_t reserved_444; // offset: 444  (0x1BC)
     uint32_t reserved_445; // offset: 445  (0x1BD)
-    uint32_t reserved_446; // offset: 446  (0x1BE)
-    uint32_t reserved_447; // offset: 447  (0x1BF)
+    uint32_t fence_address_lo; // offset: 446  (0x1BE)
+    uint32_t fence_address_hi; // offset: 447  (0x1BF)
     uint32_t gws_0_val; // offset: 448  (0x1C0)
     uint32_t gws_1_val; // offset: 449  (0x1C1)
     uint32_t gws_2_val; // offset: 450  (0x1C2)
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 81e9b443ca0a..2148c8db5a59 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -329,6 +329,34 @@ int amdgpu_dpm_mode1_reset(struct amdgpu_device *adev)
 	return ret;
 }
 
+bool amdgpu_dpm_is_link_reset_supported(struct amdgpu_device *adev)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	bool support_link_reset = false;
+
+	if (is_support_sw_smu(adev)) {
+		mutex_lock(&adev->pm.mutex);
+		support_link_reset = smu_link_reset_is_support(smu);
+		mutex_unlock(&adev->pm.mutex);
+	}
+
+	return support_link_reset;
+}
+
+int amdgpu_dpm_link_reset(struct amdgpu_device *adev)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	int ret = -EOPNOTSUPP;
+
+	if (is_support_sw_smu(adev)) {
+		mutex_lock(&adev->pm.mutex);
+		ret = smu_link_reset(smu);
+		mutex_unlock(&adev->pm.mutex);
+	}
+
+	return ret;
+}
+
 int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev,
 				    enum PP_SMC_POWER_PROFILE type,
 				    bool en)
@@ -349,6 +377,25 @@ int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev,
 	return ret;
 }
 
+int amdgpu_dpm_pause_power_profile(struct amdgpu_device *adev,
+				   bool pause)
+{
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+	int ret = 0;
+
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
+	if (pp_funcs && pp_funcs->pause_power_profile) {
+		mutex_lock(&adev->pm.mutex);
+		ret = pp_funcs->pause_power_profile(
+			adev->powerplay.pp_handle, pause);
+		mutex_unlock(&adev->pm.mutex);
+	}
+
+	return ret;
+}
+
 int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev,
 			       uint32_t pstate)
 {
@@ -761,6 +808,21 @@ int amdgpu_dpm_reset_sdma(struct amdgpu_device *adev, uint32_t inst_mask)
 	return ret;
 }
 
+int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	int ret;
+
+	if (!is_support_sw_smu(adev))
+		return -EOPNOTSUPP;
+
+	mutex_lock(&adev->pm.mutex);
+	ret = smu_reset_vcn(smu, inst_mask);
+	mutex_unlock(&adev->pm.mutex);
+
+	return ret;
+}
+
 int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev,
 				  enum pp_clock_type type,
 				  uint32_t *min,
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 922def51685b..5537dcf23b5c 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -2944,6 +2944,23 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev,
 	return sysfs_emit(buf, "%d\n", vddgfx);
 }
 
+static ssize_t amdgpu_hwmon_show_vddboard(struct device *dev,
+					  struct device_attribute *attr,
+					  char *buf)
+{
+	struct amdgpu_device *adev = dev_get_drvdata(dev);
+	u32 vddboard;
+	int r;
+
+	/* get the voltage */
+	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDBOARD,
+					    (void *)&vddboard);
+	if (r)
+		return r;
+
+	return sysfs_emit(buf, "%d\n", vddboard);
+}
+
 static ssize_t amdgpu_hwmon_show_vddgfx_label(struct device *dev,
 					      struct device_attribute *attr,
 					      char *buf)
@@ -2951,6 +2968,12 @@ static ssize_t amdgpu_hwmon_show_vddgfx_label(struct device *dev,
 	return sysfs_emit(buf, "vddgfx\n");
 }
 
+static ssize_t amdgpu_hwmon_show_vddboard_label(struct device *dev,
+						struct device_attribute *attr,
+						char *buf)
+{
+	return sysfs_emit(buf, "vddboard\n");
+}
 static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev,
 				       struct device_attribute *attr,
 				       char *buf)
@@ -3294,6 +3317,8 @@ static SENSOR_DEVICE_ATTR(in0_input, S_IRUGO, amdgpu_hwmon_show_vddgfx, NULL, 0)
 static SENSOR_DEVICE_ATTR(in0_label, S_IRUGO, amdgpu_hwmon_show_vddgfx_label, NULL, 0);
 static SENSOR_DEVICE_ATTR(in1_input, S_IRUGO, amdgpu_hwmon_show_vddnb, NULL, 0);
 static SENSOR_DEVICE_ATTR(in1_label, S_IRUGO, amdgpu_hwmon_show_vddnb_label, NULL, 0);
+static SENSOR_DEVICE_ATTR(in2_input, S_IRUGO, amdgpu_hwmon_show_vddboard, NULL, 0);
+static SENSOR_DEVICE_ATTR(in2_label, S_IRUGO, amdgpu_hwmon_show_vddboard_label, NULL, 0);
 static SENSOR_DEVICE_ATTR(power1_average, S_IRUGO, amdgpu_hwmon_show_power_avg, NULL, 0);
 static SENSOR_DEVICE_ATTR(power1_input, S_IRUGO, amdgpu_hwmon_show_power_input, NULL, 0);
 static SENSOR_DEVICE_ATTR(power1_cap_max, S_IRUGO, amdgpu_hwmon_show_power_cap_max, NULL, 0);
@@ -3341,6 +3366,8 @@ static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_in0_label.dev_attr.attr,
 	&sensor_dev_attr_in1_input.dev_attr.attr,
 	&sensor_dev_attr_in1_label.dev_attr.attr,
+	&sensor_dev_attr_in2_input.dev_attr.attr,
+	&sensor_dev_attr_in2_label.dev_attr.attr,
 	&sensor_dev_attr_power1_average.dev_attr.attr,
 	&sensor_dev_attr_power1_input.dev_attr.attr,
 	&sensor_dev_attr_power1_cap_max.dev_attr.attr,
@@ -3492,6 +3519,13 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_in1_label.dev_attr.attr))
 		return 0;
 
+	/* only few boards support vddboard */
+	if ((attr == &sensor_dev_attr_in2_input.dev_attr.attr ||
+	     attr == &sensor_dev_attr_in2_label.dev_attr.attr) &&
+	     amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDBOARD,
+					     (void *)&tmp) == -EOPNOTSUPP)
+		return 0;
+
 	/* no mclk on APUs other than gc 9,4,3*/
 	if (((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(9, 4, 3))) &&
 	    (attr == &sensor_dev_attr_freq2_input.dev_attr.attr ||
diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
index f93d287dbf13..2c3c97587dd5 100644
--- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
@@ -410,15 +410,19 @@ int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev,
 int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev,
 				    enum PP_SMC_POWER_PROFILE type,
 				    bool en);
+int amdgpu_dpm_pause_power_profile(struct amdgpu_device *adev,
+				   bool pause);
 
 int amdgpu_dpm_baco_reset(struct amdgpu_device *adev);
 
 int amdgpu_dpm_mode2_reset(struct amdgpu_device *adev);
+int amdgpu_dpm_link_reset(struct amdgpu_device *adev);
 int amdgpu_dpm_enable_gfx_features(struct amdgpu_device *adev);
 
 int amdgpu_dpm_is_baco_supported(struct amdgpu_device *adev);
 
 bool amdgpu_dpm_is_mode1_reset_supported(struct amdgpu_device *adev);
+bool amdgpu_dpm_is_link_reset_supported(struct amdgpu_device *adev);
 int amdgpu_dpm_mode1_reset(struct amdgpu_device *adev);
 
 int amdgpu_dpm_set_mp1_state(struct amdgpu_device *adev,
@@ -605,5 +609,6 @@ ssize_t amdgpu_dpm_get_pm_policy_info(struct amdgpu_device *adev,
 				      enum pp_pm_policy p_type, char *buf);
 int amdgpu_dpm_reset_sdma(struct amdgpu_device *adev, uint32_t inst_mask);
 bool amdgpu_dpm_reset_sdma_is_supported(struct amdgpu_device *adev);
+int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask);
 
 #endif
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index 1c25f3023e93..4c0e976004ba 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -30,16 +30,32 @@
 #include "amdgpu_atombios.h"
 #include "amdgpu_dpm_internal.h"
 #include "amd_pcie.h"
-#include "sid.h"
+#include "atom.h"
+#include "gfx_v6_0.h"
 #include "r600_dpm.h"
+#include "sid.h"
 #include "si_dpm.h"
-#include "atom.h"
 #include "../include/pptable.h"
 #include <linux/math64.h>
 #include <linux/seq_file.h>
 #include <linux/firmware.h>
 #include <legacy_dpm.h>
 
+#include "bif/bif_3_0_d.h"
+#include "bif/bif_3_0_sh_mask.h"
+
+#include "dce/dce_6_0_d.h"
+#include "dce/dce_6_0_sh_mask.h"
+
+#include "gca/gfx_6_0_d.h"
+#include "gca/gfx_6_0_sh_mask.h"
+
+#include"gmc/gmc_6_0_d.h"
+#include"gmc/gmc_6_0_sh_mask.h"
+
+#include "smu/smu_6_0_d.h"
+#include "smu/smu_6_0_sh_mask.h"
+
 #define MC_CG_ARB_FREQ_F0           0x0a
 #define MC_CG_ARB_FREQ_F1           0x0b
 #define MC_CG_ARB_FREQ_F2           0x0c
@@ -2193,7 +2209,7 @@ static u32 si_calculate_cac_wintime(struct amdgpu_device *adev)
 	if (xclk == 0)
 		return 0;
 
-	cac_window = RREG32(CG_CAC_CTRL) & CAC_WINDOW_MASK;
+	cac_window = RREG32(mmCG_CAC_CTRL) & CG_CAC_CTRL__CAC_WINDOW_MASK;
 	cac_window_size = ((cac_window & 0xFFFF0000) >> 16) * (cac_window & 0x0000FFFF);
 
 	wintime = (cac_window_size * 100) / xclk;
@@ -2489,19 +2505,19 @@ static int si_populate_sq_ramping_values(struct amdgpu_device *adev,
 	if (adev->pm.dpm.sq_ramping_threshold == 0)
 		return -EINVAL;
 
-	if (SISLANDS_DPM2_SQ_RAMP_MAX_POWER > (MAX_POWER_MASK >> MAX_POWER_SHIFT))
+	if (SISLANDS_DPM2_SQ_RAMP_MAX_POWER > (SQ_POWER_THROTTLE__MAX_POWER_MASK >> SQ_POWER_THROTTLE__MAX_POWER__SHIFT))
 		enable_sq_ramping = false;
 
-	if (SISLANDS_DPM2_SQ_RAMP_MIN_POWER > (MIN_POWER_MASK >> MIN_POWER_SHIFT))
+	if (SISLANDS_DPM2_SQ_RAMP_MIN_POWER > (SQ_POWER_THROTTLE__MIN_POWER_MASK >> SQ_POWER_THROTTLE__MIN_POWER__SHIFT))
 		enable_sq_ramping = false;
 
-	if (SISLANDS_DPM2_SQ_RAMP_MAX_POWER_DELTA > (MAX_POWER_DELTA_MASK >> MAX_POWER_DELTA_SHIFT))
+	if (SISLANDS_DPM2_SQ_RAMP_MAX_POWER_DELTA > (SQ_POWER_THROTTLE2__MAX_POWER_DELTA_MASK >> SQ_POWER_THROTTLE2__MAX_POWER_DELTA__SHIFT))
 		enable_sq_ramping = false;
 
-	if (SISLANDS_DPM2_SQ_RAMP_STI_SIZE > (STI_SIZE_MASK >> STI_SIZE_SHIFT))
+	if (SISLANDS_DPM2_SQ_RAMP_STI_SIZE > (SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE_MASK >> SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE__SHIFT))
 		enable_sq_ramping = false;
 
-	if (SISLANDS_DPM2_SQ_RAMP_LTI_RATIO > (LTI_RATIO_MASK >> LTI_RATIO_SHIFT))
+	if (SISLANDS_DPM2_SQ_RAMP_LTI_RATIO > (SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO_MASK >> SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO__SHIFT))
 		enable_sq_ramping = false;
 
 	for (i = 0; i < state->performance_level_count; i++) {
@@ -2510,14 +2526,17 @@ static int si_populate_sq_ramping_values(struct amdgpu_device *adev,
 
 		if ((state->performance_levels[i].sclk >= adev->pm.dpm.sq_ramping_threshold) &&
 		    enable_sq_ramping) {
-			sq_power_throttle |= MAX_POWER(SISLANDS_DPM2_SQ_RAMP_MAX_POWER);
-			sq_power_throttle |= MIN_POWER(SISLANDS_DPM2_SQ_RAMP_MIN_POWER);
-			sq_power_throttle2 |= MAX_POWER_DELTA(SISLANDS_DPM2_SQ_RAMP_MAX_POWER_DELTA);
-			sq_power_throttle2 |= STI_SIZE(SISLANDS_DPM2_SQ_RAMP_STI_SIZE);
-			sq_power_throttle2 |= LTI_RATIO(SISLANDS_DPM2_SQ_RAMP_LTI_RATIO);
+			sq_power_throttle |= SISLANDS_DPM2_SQ_RAMP_MAX_POWER << SQ_POWER_THROTTLE__MAX_POWER__SHIFT;
+			sq_power_throttle |= SISLANDS_DPM2_SQ_RAMP_MIN_POWER << SQ_POWER_THROTTLE__MIN_POWER__SHIFT;
+			sq_power_throttle2 |= SISLANDS_DPM2_SQ_RAMP_MAX_POWER_DELTA << SQ_POWER_THROTTLE2__MAX_POWER_DELTA__SHIFT;
+			sq_power_throttle2 |= SISLANDS_DPM2_SQ_RAMP_STI_SIZE << SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE__SHIFT;
+			sq_power_throttle2 |= SISLANDS_DPM2_SQ_RAMP_LTI_RATIO << SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO__SHIFT;
 		} else {
-			sq_power_throttle |= MAX_POWER_MASK | MIN_POWER_MASK;
-			sq_power_throttle2 |= MAX_POWER_DELTA_MASK | STI_SIZE_MASK | LTI_RATIO_MASK;
+			sq_power_throttle |= SQ_POWER_THROTTLE__MAX_POWER_MASK |
+								SQ_POWER_THROTTLE__MIN_POWER_MASK;
+			sq_power_throttle2 |= SQ_POWER_THROTTLE2__MAX_POWER_DELTA_MASK |
+								SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE_MASK |
+								SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO_MASK;
 		}
 
 		smc_state->levels[i].SQPowerThrottle = cpu_to_be32(sq_power_throttle);
@@ -2761,9 +2780,9 @@ static int si_initialize_smc_cac_tables(struct amdgpu_device *adev)
 	if (!cac_tables)
 		return -ENOMEM;
 
-	reg = RREG32(CG_CAC_CTRL) & ~CAC_WINDOW_MASK;
-	reg |= CAC_WINDOW(si_pi->powertune_data->cac_window);
-	WREG32(CG_CAC_CTRL, reg);
+	reg = RREG32(mmCG_CAC_CTRL) & ~CG_CAC_CTRL__CAC_WINDOW_MASK;
+	reg |= (si_pi->powertune_data->cac_window << CG_CAC_CTRL__CAC_WINDOW__SHIFT);
+	WREG32(mmCG_CAC_CTRL, reg);
 
 	si_pi->dyn_powertune_data.cac_leakage = adev->pm.dpm.cac_leakage;
 	si_pi->dyn_powertune_data.dc_pwr_value =
@@ -2962,10 +2981,10 @@ static int si_init_smc_spll_table(struct amdgpu_device *adev)
 		ret = si_calculate_sclk_params(adev, sclk, &sclk_params);
 		if (ret)
 			break;
-		p_div = (sclk_params.vCG_SPLL_FUNC_CNTL & SPLL_PDIV_A_MASK) >> SPLL_PDIV_A_SHIFT;
-		fb_div = (sclk_params.vCG_SPLL_FUNC_CNTL_3 & SPLL_FB_DIV_MASK) >> SPLL_FB_DIV_SHIFT;
-		clk_s = (sclk_params.vCG_SPLL_SPREAD_SPECTRUM & CLK_S_MASK) >> CLK_S_SHIFT;
-		clk_v = (sclk_params.vCG_SPLL_SPREAD_SPECTRUM_2 & CLK_V_MASK) >> CLK_V_SHIFT;
+		p_div = (sclk_params.vCG_SPLL_FUNC_CNTL & CG_SPLL_FUNC_CNTL__SPLL_PDIV_A_MASK) >> CG_SPLL_FUNC_CNTL__SPLL_PDIV_A__SHIFT;
+		fb_div = (sclk_params.vCG_SPLL_FUNC_CNTL_3 & CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV_MASK) >> CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV__SHIFT;
+		clk_s = (sclk_params.vCG_SPLL_SPREAD_SPECTRUM & CG_SPLL_SPREAD_SPECTRUM__CLK_S_MASK) >> CG_SPLL_SPREAD_SPECTRUM__CLK_S__SHIFT;
+		clk_v = (sclk_params.vCG_SPLL_SPREAD_SPECTRUM_2 & CG_SPLL_SPREAD_SPECTRUM_2__CLK_V_MASK) >> CG_SPLL_SPREAD_SPECTRUM_2__CLK_V__SHIFT;
 
 		fb_div &= ~0x00001FFF;
 		fb_div >>= 1;
@@ -3669,10 +3688,10 @@ static bool si_is_special_1gb_platform(struct amdgpu_device *adev)
 	WREG32(MC_SEQ_IO_DEBUG_INDEX, 0xb);
 	width = ((RREG32(MC_SEQ_IO_DEBUG_DATA) >> 1) & 1) ? 16 : 32;
 
-	tmp = RREG32(MC_ARB_RAMCFG);
-	row = ((tmp & NOOFROWS_MASK) >> NOOFROWS_SHIFT) + 10;
-	column = ((tmp & NOOFCOLS_MASK) >> NOOFCOLS_SHIFT) + 8;
-	bank = ((tmp & NOOFBANK_MASK) >> NOOFBANK_SHIFT) + 2;
+	tmp = RREG32(mmMC_ARB_RAMCFG);
+	row = ((tmp & MC_ARB_RAMCFG__NOOFROWS_MASK) >> MC_ARB_RAMCFG__NOOFROWS__SHIFT) + 10;
+	column = ((tmp & MC_ARB_RAMCFG__NOOFCOLS_MASK) >> MC_ARB_RAMCFG__NOOFCOLS__SHIFT) + 8;
+	bank = ((tmp & MC_ARB_RAMCFG__NOOFBANK_MASK) >> MC_ARB_RAMCFG__NOOFBANK__SHIFT) + 2;
 
 	density = (1 << (row + column - 20 + bank)) * width;
 
@@ -3756,11 +3775,11 @@ static void si_set_dpm_event_sources(struct amdgpu_device *adev, u32 sources)
 	}
 
 	if (want_thermal_protection) {
-		WREG32_P(CG_THERMAL_CTRL, DPM_EVENT_SRC(dpm_event_src), ~DPM_EVENT_SRC_MASK);
+		WREG32_P(mmCG_THERMAL_CTRL, dpm_event_src << CG_THERMAL_CTRL__DPM_EVENT_SRC__SHIFT, ~CG_THERMAL_CTRL__DPM_EVENT_SRC_MASK);
 		if (pi->thermal_protection)
-			WREG32_P(GENERAL_PWRMGT, 0, ~THERMAL_PROTECTION_DIS);
+			WREG32_P(mmGENERAL_PWRMGT, 0, ~GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK);
 	} else {
-		WREG32_P(GENERAL_PWRMGT, THERMAL_PROTECTION_DIS, ~THERMAL_PROTECTION_DIS);
+		WREG32_P(mmGENERAL_PWRMGT, GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK, ~GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK);
 	}
 }
 
@@ -3785,20 +3804,20 @@ static void si_enable_auto_throttle_source(struct amdgpu_device *adev,
 
 static void si_start_dpm(struct amdgpu_device *adev)
 {
-	WREG32_P(GENERAL_PWRMGT, GLOBAL_PWRMGT_EN, ~GLOBAL_PWRMGT_EN);
+	WREG32_P(mmGENERAL_PWRMGT, GENERAL_PWRMGT__GLOBAL_PWRMGT_EN_MASK, ~GENERAL_PWRMGT__GLOBAL_PWRMGT_EN_MASK);
 }
 
 static void si_stop_dpm(struct amdgpu_device *adev)
 {
-	WREG32_P(GENERAL_PWRMGT, 0, ~GLOBAL_PWRMGT_EN);
+	WREG32_P(mmGENERAL_PWRMGT, 0, ~GENERAL_PWRMGT__GLOBAL_PWRMGT_EN_MASK);
 }
 
 static void si_enable_sclk_control(struct amdgpu_device *adev, bool enable)
 {
 	if (enable)
-		WREG32_P(SCLK_PWRMGT_CNTL, 0, ~SCLK_PWRMGT_OFF);
+		WREG32_P(mmSCLK_PWRMGT_CNTL, 0, ~SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF_MASK);
 	else
-		WREG32_P(SCLK_PWRMGT_CNTL, SCLK_PWRMGT_OFF, ~SCLK_PWRMGT_OFF);
+		WREG32_P(mmSCLK_PWRMGT_CNTL, SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF_MASK, ~SCLK_PWRMGT_CNTL__SCLK_PWRMGT_OFF_MASK);
 
 }
 
@@ -3838,7 +3857,7 @@ static int si_notify_hw_of_powersource(struct amdgpu_device *adev, bool ac_power
 static PPSMC_Result si_send_msg_to_smc_with_parameter(struct amdgpu_device *adev,
 						      PPSMC_Msg msg, u32 parameter)
 {
-	WREG32(SMC_SCRATCH0, parameter);
+	WREG32(mmSMC_SCRATCH0, parameter);
 	return amdgpu_si_send_msg_to_smc(adev, msg);
 }
 
@@ -4023,12 +4042,12 @@ static void si_read_clock_registers(struct amdgpu_device *adev)
 {
 	struct si_power_info *si_pi = si_get_pi(adev);
 
-	si_pi->clock_registers.cg_spll_func_cntl = RREG32(CG_SPLL_FUNC_CNTL);
-	si_pi->clock_registers.cg_spll_func_cntl_2 = RREG32(CG_SPLL_FUNC_CNTL_2);
-	si_pi->clock_registers.cg_spll_func_cntl_3 = RREG32(CG_SPLL_FUNC_CNTL_3);
-	si_pi->clock_registers.cg_spll_func_cntl_4 = RREG32(CG_SPLL_FUNC_CNTL_4);
-	si_pi->clock_registers.cg_spll_spread_spectrum = RREG32(CG_SPLL_SPREAD_SPECTRUM);
-	si_pi->clock_registers.cg_spll_spread_spectrum_2 = RREG32(CG_SPLL_SPREAD_SPECTRUM_2);
+	si_pi->clock_registers.cg_spll_func_cntl = RREG32(mmCG_SPLL_FUNC_CNTL);
+	si_pi->clock_registers.cg_spll_func_cntl_2 = RREG32(mmCG_SPLL_FUNC_CNTL_2);
+	si_pi->clock_registers.cg_spll_func_cntl_3 = RREG32(mmCG_SPLL_FUNC_CNTL_3);
+	si_pi->clock_registers.cg_spll_func_cntl_4 = RREG32(mmCG_SPLL_FUNC_CNTL_4);
+	si_pi->clock_registers.cg_spll_spread_spectrum = RREG32(mmCG_SPLL_SPREAD_SPECTRUM);
+	si_pi->clock_registers.cg_spll_spread_spectrum_2 = RREG32(mmCG_SPLL_SPREAD_SPECTRUM_2);
 	si_pi->clock_registers.dll_cntl = RREG32(DLL_CNTL);
 	si_pi->clock_registers.mclk_pwrmgt_cntl = RREG32(MCLK_PWRMGT_CNTL);
 	si_pi->clock_registers.mpll_ad_func_cntl = RREG32(MPLL_AD_FUNC_CNTL);
@@ -4044,14 +4063,14 @@ static void si_enable_thermal_protection(struct amdgpu_device *adev,
 					  bool enable)
 {
 	if (enable)
-		WREG32_P(GENERAL_PWRMGT, 0, ~THERMAL_PROTECTION_DIS);
+		WREG32_P(mmGENERAL_PWRMGT, 0, ~GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK);
 	else
-		WREG32_P(GENERAL_PWRMGT, THERMAL_PROTECTION_DIS, ~THERMAL_PROTECTION_DIS);
+		WREG32_P(mmGENERAL_PWRMGT, GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK, ~GENERAL_PWRMGT__THERMAL_PROTECTION_DIS_MASK);
 }
 
 static void si_enable_acpi_power_management(struct amdgpu_device *adev)
 {
-	WREG32_P(GENERAL_PWRMGT, STATIC_PM_EN, ~STATIC_PM_EN);
+	WREG32_P(mmGENERAL_PWRMGT, GENERAL_PWRMGT__STATIC_PM_EN_MASK, ~GENERAL_PWRMGT__STATIC_PM_EN_MASK);
 }
 
 #if 0
@@ -4132,9 +4151,9 @@ static void si_program_ds_registers(struct amdgpu_device *adev)
 		tmp = 0x1;
 
 	if (eg_pi->sclk_deep_sleep) {
-		WREG32_P(MISC_CLK_CNTL, DEEP_SLEEP_CLK_SEL(tmp), ~DEEP_SLEEP_CLK_SEL_MASK);
-		WREG32_P(CG_SPLL_AUTOSCALE_CNTL, AUTOSCALE_ON_SS_CLEAR,
-			 ~AUTOSCALE_ON_SS_CLEAR);
+		WREG32_P(mmMISC_CLK_CNTL, (tmp << MISC_CLK_CNTL__DEEP_SLEEP_CLK_SEL__SHIFT), ~MISC_CLK_CNTL__DEEP_SLEEP_CLK_SEL_MASK);
+		WREG32_P(mmCG_SPLL_AUTOSCALE_CNTL, CG_SPLL_AUTOSCALE_CNTL__AUTOSCALE_ON_SS_CLEAR_MASK,
+			 ~CG_SPLL_AUTOSCALE_CNTL__AUTOSCALE_ON_SS_CLEAR_MASK);
 	}
 }
 
@@ -4143,18 +4162,18 @@ static void si_program_display_gap(struct amdgpu_device *adev)
 	u32 tmp, pipe;
 	int i;
 
-	tmp = RREG32(CG_DISPLAY_GAP_CNTL) & ~(DISP1_GAP_MASK | DISP2_GAP_MASK);
+	tmp = RREG32(mmCG_DISPLAY_GAP_CNTL) & ~(CG_DISPLAY_GAP_CNTL__DISP1_GAP_MASK | CG_DISPLAY_GAP_CNTL__DISP2_GAP_MASK);
 	if (adev->pm.dpm.new_active_crtc_count > 0)
-		tmp |= DISP1_GAP(R600_PM_DISPLAY_GAP_VBLANK_OR_WM);
+		tmp |= R600_PM_DISPLAY_GAP_VBLANK_OR_WM << CG_DISPLAY_GAP_CNTL__DISP1_GAP__SHIFT;
 	else
-		tmp |= DISP1_GAP(R600_PM_DISPLAY_GAP_IGNORE);
+		tmp |= R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP1_GAP__SHIFT;
 
 	if (adev->pm.dpm.new_active_crtc_count > 1)
-		tmp |= DISP2_GAP(R600_PM_DISPLAY_GAP_VBLANK_OR_WM);
+		tmp |= R600_PM_DISPLAY_GAP_VBLANK_OR_WM << CG_DISPLAY_GAP_CNTL__DISP2_GAP__SHIFT;
 	else
-		tmp |= DISP2_GAP(R600_PM_DISPLAY_GAP_IGNORE);
+		tmp |= R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP2_GAP__SHIFT;
 
-	WREG32(CG_DISPLAY_GAP_CNTL, tmp);
+	WREG32(mmCG_DISPLAY_GAP_CNTL, tmp);
 
 	tmp = RREG32(DCCG_DISP_SLOW_SELECT_REG);
 	pipe = (tmp & DCCG_DISP1_SLOW_SELECT_MASK) >> DCCG_DISP1_SLOW_SELECT_SHIFT;
@@ -4189,10 +4208,10 @@ static void si_enable_spread_spectrum(struct amdgpu_device *adev, bool enable)
 
 	if (enable) {
 		if (pi->sclk_ss)
-			WREG32_P(GENERAL_PWRMGT, DYN_SPREAD_SPECTRUM_EN, ~DYN_SPREAD_SPECTRUM_EN);
+			WREG32_P(mmGENERAL_PWRMGT, GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN_MASK, ~GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN_MASK);
 	} else {
-		WREG32_P(CG_SPLL_SPREAD_SPECTRUM, 0, ~SSEN);
-		WREG32_P(GENERAL_PWRMGT, 0, ~DYN_SPREAD_SPECTRUM_EN);
+		WREG32_P(mmCG_SPLL_SPREAD_SPECTRUM, 0, ~CG_SPLL_SPREAD_SPECTRUM__SSEN_MASK);
+		WREG32_P(mmGENERAL_PWRMGT, 0, ~GENERAL_PWRMGT__DYN_SPREAD_SPECTRUM_EN_MASK);
 	}
 }
 
@@ -4214,15 +4233,15 @@ static void si_setup_bsp(struct amdgpu_device *adev)
 			       &pi->pbsu);
 
 
-        pi->dsp = BSP(pi->bsp) | BSU(pi->bsu);
-	pi->psp = BSP(pi->pbsp) | BSU(pi->pbsu);
+        pi->dsp = (pi->bsp << CG_BSP__BSP__SHIFT) | (pi->bsu << CG_BSP__BSU__SHIFT);
+	pi->psp = (pi->pbsp << CG_BSP__BSP__SHIFT) | (pi->pbsu << CG_BSP__BSU__SHIFT);
 
-	WREG32(CG_BSP, pi->dsp);
+	WREG32(mmCG_BSP, pi->dsp);
 }
 
 static void si_program_git(struct amdgpu_device *adev)
 {
-	WREG32_P(CG_GIT, CG_GICST(R600_GICST_DFLT), ~CG_GICST_MASK);
+	WREG32_P(mmCG_GIT, R600_GICST_DFLT << CG_GIT__CG_GICST__SHIFT, ~CG_GIT__CG_GICST_MASK);
 }
 
 static void si_program_tp(struct amdgpu_device *adev)
@@ -4231,54 +4250,54 @@ static void si_program_tp(struct amdgpu_device *adev)
 	enum r600_td td = R600_TD_DFLT;
 
 	for (i = 0; i < R600_PM_NUMBER_OF_TC; i++)
-		WREG32(CG_FFCT_0 + i, (UTC_0(r600_utc[i]) | DTC_0(r600_dtc[i])));
+		WREG32(mmCG_FFCT_0 + i, (r600_utc[i] << CG_FFCT_0__UTC_0__SHIFT | r600_dtc[i] << CG_FFCT_0__DTC_0__SHIFT));
 
 	if (td == R600_TD_AUTO)
-		WREG32_P(SCLK_PWRMGT_CNTL, 0, ~FIR_FORCE_TREND_SEL);
+		WREG32_P(mmSCLK_PWRMGT_CNTL, 0, ~SCLK_PWRMGT_CNTL__FIR_FORCE_TREND_SEL_MASK);
 	else
-		WREG32_P(SCLK_PWRMGT_CNTL, FIR_FORCE_TREND_SEL, ~FIR_FORCE_TREND_SEL);
+		WREG32_P(mmSCLK_PWRMGT_CNTL, SCLK_PWRMGT_CNTL__FIR_FORCE_TREND_SEL_MASK, ~SCLK_PWRMGT_CNTL__FIR_FORCE_TREND_SEL_MASK);
 
 	if (td == R600_TD_UP)
-		WREG32_P(SCLK_PWRMGT_CNTL, 0, ~FIR_TREND_MODE);
+		WREG32_P(mmSCLK_PWRMGT_CNTL, 0, ~SCLK_PWRMGT_CNTL__FIR_TREND_MODE_MASK);
 
 	if (td == R600_TD_DOWN)
-		WREG32_P(SCLK_PWRMGT_CNTL, FIR_TREND_MODE, ~FIR_TREND_MODE);
+		WREG32_P(mmSCLK_PWRMGT_CNTL, SCLK_PWRMGT_CNTL__FIR_TREND_MODE_MASK, ~SCLK_PWRMGT_CNTL__FIR_TREND_MODE_MASK);
 }
 
 static void si_program_tpp(struct amdgpu_device *adev)
 {
-	WREG32(CG_TPC, R600_TPC_DFLT);
+	WREG32(mmCG_TPC, R600_TPC_DFLT);
 }
 
 static void si_program_sstp(struct amdgpu_device *adev)
 {
-	WREG32(CG_SSP, (SSTU(R600_SSTU_DFLT) | SST(R600_SST_DFLT)));
+	WREG32(mmCG_SSP, (R600_SSTU_DFLT << CG_SSP__SSTU__SHIFT| R600_SST_DFLT << CG_SSP__SST__SHIFT));
 }
 
 static void si_enable_display_gap(struct amdgpu_device *adev)
 {
-	u32 tmp = RREG32(CG_DISPLAY_GAP_CNTL);
+	u32 tmp = RREG32(mmCG_DISPLAY_GAP_CNTL);
 
-	tmp &= ~(DISP1_GAP_MASK | DISP2_GAP_MASK);
-	tmp |= (DISP1_GAP(R600_PM_DISPLAY_GAP_IGNORE) |
-		DISP2_GAP(R600_PM_DISPLAY_GAP_IGNORE));
+	tmp &= ~(CG_DISPLAY_GAP_CNTL__DISP1_GAP_MASK | CG_DISPLAY_GAP_CNTL__DISP2_GAP_MASK);
+	tmp |= (R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP1_GAP__SHIFT |
+		R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP2_GAP__SHIFT);
 
-	tmp &= ~(DISP1_GAP_MCHG_MASK | DISP2_GAP_MCHG_MASK);
-	tmp |= (DISP1_GAP_MCHG(R600_PM_DISPLAY_GAP_VBLANK) |
-		DISP2_GAP_MCHG(R600_PM_DISPLAY_GAP_IGNORE));
-	WREG32(CG_DISPLAY_GAP_CNTL, tmp);
+	tmp &= ~(CG_DISPLAY_GAP_CNTL__DISP1_GAP_MCHG_MASK | CG_DISPLAY_GAP_CNTL__DISP2_GAP_MCHG_MASK);
+	tmp |= (R600_PM_DISPLAY_GAP_VBLANK << CG_DISPLAY_GAP_CNTL__DISP1_GAP_MCHG__SHIFT |
+		R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP2_GAP_MCHG__SHIFT);
+	WREG32(mmCG_DISPLAY_GAP_CNTL, tmp);
 }
 
 static void si_program_vc(struct amdgpu_device *adev)
 {
 	struct rv7xx_power_info *pi = rv770_get_pi(adev);
 
-	WREG32(CG_FTV, pi->vrc);
+	WREG32(mmCG_FTV, pi->vrc);
 }
 
 static void si_clear_vc(struct amdgpu_device *adev)
 {
-	WREG32(CG_FTV, 0);
+	WREG32(mmCG_FTV, 0);
 }
 
 static u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock)
@@ -4735,7 +4754,7 @@ static u32 si_calculate_memory_refresh_rate(struct amdgpu_device *adev,
 	u32 dram_rows;
 	u32 dram_refresh_rate;
 	u32 mc_arb_rfsh_rate;
-	u32 tmp = (RREG32(MC_ARB_RAMCFG) & NOOFROWS_MASK) >> NOOFROWS_SHIFT;
+	u32 tmp = (RREG32(mmMC_ARB_RAMCFG) & MC_ARB_RAMCFG__NOOFROWS_MASK) >> MC_ARB_RAMCFG__NOOFROWS__SHIFT;
 
 	if (tmp >= 4)
 		dram_rows = 16384;
@@ -4909,7 +4928,7 @@ static int si_populate_smc_initial_state(struct amdgpu_device *adev,
 
 	si_populate_initial_mvdd_value(adev, &table->initialState.level.mvdd);
 
-	reg = CG_R(0xffff) | CG_L(0);
+	reg = 0xffff << CG_AT__CG_R__SHIFT | 0 << CG_AT__CG_L__SHIFT;
 	table->initialState.level.aT = cpu_to_be32(reg);
 	table->initialState.level.bSP = cpu_to_be32(pi->dsp);
 	table->initialState.level.gen2PCIE = (u8)si_pi->boot_pcie_gen;
@@ -4935,10 +4954,13 @@ static int si_populate_smc_initial_state(struct amdgpu_device *adev,
 	table->initialState.level.dpm2.BelowSafeInc = 0;
 	table->initialState.level.dpm2.PwrEfficiencyRatio = 0;
 
-	reg = MIN_POWER_MASK | MAX_POWER_MASK;
+	reg = SQ_POWER_THROTTLE__MIN_POWER_MASK |
+		SQ_POWER_THROTTLE__MAX_POWER_MASK;
 	table->initialState.level.SQPowerThrottle = cpu_to_be32(reg);
 
-	reg = MAX_POWER_DELTA_MASK | STI_SIZE_MASK | LTI_RATIO_MASK;
+	reg = SQ_POWER_THROTTLE2__MAX_POWER_DELTA_MASK |
+		SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE_MASK |
+		SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO_MASK;
 	table->initialState.level.SQPowerThrottle_2 = cpu_to_be32(reg);
 
 	return 0;
@@ -5057,8 +5079,8 @@ static int si_populate_smc_acpi_state(struct amdgpu_device *adev,
 
 	dll_cntl &= ~(MRDCK0_BYPASS | MRDCK1_BYPASS);
 
-	spll_func_cntl_2 &= ~SCLK_MUX_SEL_MASK;
-	spll_func_cntl_2 |= SCLK_MUX_SEL(4);
+	spll_func_cntl_2 &= ~CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL_MASK;
+	spll_func_cntl_2 |= 4 << CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL__SHIFT;
 
 	table->ACPIState.level.mclk.vDLL_CNTL =
 		cpu_to_be32(dll_cntl);
@@ -5102,10 +5124,10 @@ static int si_populate_smc_acpi_state(struct amdgpu_device *adev,
 	table->ACPIState.level.dpm2.BelowSafeInc = 0;
 	table->ACPIState.level.dpm2.PwrEfficiencyRatio = 0;
 
-	reg = MIN_POWER_MASK | MAX_POWER_MASK;
+	reg = SQ_POWER_THROTTLE__MIN_POWER_MASK | SQ_POWER_THROTTLE__MAX_POWER_MASK;
 	table->ACPIState.level.SQPowerThrottle = cpu_to_be32(reg);
 
-	reg = MAX_POWER_DELTA_MASK | STI_SIZE_MASK | LTI_RATIO_MASK;
+	reg = SQ_POWER_THROTTLE2__MAX_POWER_DELTA_MASK | SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE_MASK | SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO_MASK;
 	table->ACPIState.level.SQPowerThrottle_2 = cpu_to_be32(reg);
 
 	return 0;
@@ -5250,8 +5272,8 @@ static int si_init_smc_table(struct amdgpu_device *adev)
 		if (ret)
 			return ret;
 
-		WREG32(CG_ULV_CONTROL, ulv->cg_ulv_control);
-		WREG32(CG_ULV_PARAMETER, ulv->cg_ulv_parameter);
+		WREG32(mmCG_ULV_CONTROL, ulv->cg_ulv_control);
+		WREG32(mmCG_ULV_PARAMETER, ulv->cg_ulv_parameter);
 
 		lane_width = amdgpu_get_pcie_lanes(adev);
 		si_write_smc_soft_register(adev, SI_SMC_SOFT_REGISTER_non_ulv_pcie_link_width, lane_width);
@@ -5294,16 +5316,16 @@ static int si_calculate_sclk_params(struct amdgpu_device *adev,
 	do_div(tmp, reference_clock);
 	fbdiv = (u32) tmp;
 
-	spll_func_cntl &= ~(SPLL_PDIV_A_MASK | SPLL_REF_DIV_MASK);
-	spll_func_cntl |= SPLL_REF_DIV(dividers.ref_div);
-	spll_func_cntl |= SPLL_PDIV_A(dividers.post_div);
+	spll_func_cntl &= ~(CG_SPLL_FUNC_CNTL__SPLL_PDIV_A_MASK | CG_SPLL_FUNC_CNTL__SPLL_REF_DIV_MASK);
+	spll_func_cntl |= dividers.ref_div << CG_SPLL_FUNC_CNTL__SPLL_REF_DIV__SHIFT;
+	spll_func_cntl |= dividers.post_div << CG_SPLL_FUNC_CNTL__SPLL_PDIV_A__SHIFT;
 
-	spll_func_cntl_2 &= ~SCLK_MUX_SEL_MASK;
-	spll_func_cntl_2 |= SCLK_MUX_SEL(2);
+	spll_func_cntl_2 &= ~CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL_MASK;
+	spll_func_cntl_2 |= 2 << CG_SPLL_FUNC_CNTL_2__SCLK_MUX_SEL__SHIFT;
 
-	spll_func_cntl_3 &= ~SPLL_FB_DIV_MASK;
-	spll_func_cntl_3 |= SPLL_FB_DIV(fbdiv);
-	spll_func_cntl_3 |= SPLL_DITHEN;
+	spll_func_cntl_3 &= ~CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV_MASK;
+	spll_func_cntl_3 |= fbdiv << CG_SPLL_FUNC_CNTL_3__SPLL_FB_DIV__SHIFT;
+	spll_func_cntl_3 |= CG_SPLL_FUNC_CNTL_3__SPLL_DITHEN_MASK;
 
 	if (pi->sclk_ss) {
 		struct amdgpu_atom_ss ss;
@@ -5314,12 +5336,12 @@ static int si_calculate_sclk_params(struct amdgpu_device *adev,
 			u32 clk_s = reference_clock * 5 / (reference_divider * ss.rate);
 			u32 clk_v = 4 * ss.percentage * fbdiv / (clk_s * 10000);
 
-			cg_spll_spread_spectrum &= ~CLK_S_MASK;
-			cg_spll_spread_spectrum |= CLK_S(clk_s);
-			cg_spll_spread_spectrum |= SSEN;
+			cg_spll_spread_spectrum &= ~CG_SPLL_SPREAD_SPECTRUM__CLK_S_MASK;
+			cg_spll_spread_spectrum |= clk_s << CG_SPLL_SPREAD_SPECTRUM__CLK_S__SHIFT;
+			cg_spll_spread_spectrum |= CG_SPLL_SPREAD_SPECTRUM__SSEN_MASK;
 
-			cg_spll_spread_spectrum_2 &= ~CLK_V_MASK;
-			cg_spll_spread_spectrum_2 |= CLK_V(clk_v);
+			cg_spll_spread_spectrum_2 &= ~CG_SPLL_SPREAD_SPECTRUM_2__CLK_V_MASK;
+			cg_spll_spread_spectrum_2 |= clk_v << CG_SPLL_SPREAD_SPECTRUM_2__CLK_V__SHIFT;
 		}
 	}
 
@@ -5485,7 +5507,7 @@ static int si_convert_power_level_to_smc(struct amdgpu_device *adev,
 	if (pi->mclk_stutter_mode_threshold &&
 	    (pl->mclk <= pi->mclk_stutter_mode_threshold) &&
 	    !eg_pi->uvd_enabled &&
-	    (RREG32(DPG_PIPE_STUTTER_CONTROL) & STUTTER_ENABLE) &&
+	    (RREG32(mmDPG_PIPE_STUTTER_CONTROL) & DPG_PIPE_STUTTER_CONTROL__STUTTER_ENABLE_MASK) &&
 	    (adev->pm.dpm.new_active_crtc_count <= 2)) {
 		level->mcFlags |= SISLANDS_SMC_MC_STUTTER_EN;
 	}
@@ -5579,7 +5601,7 @@ static int si_populate_smc_t(struct amdgpu_device *adev,
 		return -EINVAL;
 
 	if (state->performance_level_count < 2) {
-		a_t = CG_R(0xffff) | CG_L(0);
+		a_t = 0xffff << CG_AT__CG_R__SHIFT | 0 << CG_AT__CG_L__SHIFT;
 		smc_state->levels[0].aT = cpu_to_be32(a_t);
 		return 0;
 	}
@@ -5600,13 +5622,13 @@ static int si_populate_smc_t(struct amdgpu_device *adev,
 			t_l = (i + 1) * 1000 + 50 * R600_AH_DFLT;
 		}
 
-		a_t = be32_to_cpu(smc_state->levels[i].aT) & ~CG_R_MASK;
-		a_t |= CG_R(t_l * pi->bsp / 20000);
+		a_t = be32_to_cpu(smc_state->levels[i].aT) & ~CG_AT__CG_R_MASK;
+		a_t |= (t_l * pi->bsp / 20000) << CG_AT__CG_R__SHIFT;
 		smc_state->levels[i].aT = cpu_to_be32(a_t);
 
 		high_bsp = (i == state->performance_level_count - 2) ?
 			pi->pbsp : pi->bsp;
-		a_t = CG_R(0xffff) | CG_L(t_h * high_bsp / 20000);
+		a_t = (0xffff) << CG_AT__CG_R__SHIFT | (t_h * high_bsp / 20000) << CG_AT__CG_L__SHIFT;
 		smc_state->levels[i + 1].aT = cpu_to_be32(a_t);
 	}
 
@@ -6180,9 +6202,9 @@ static int si_upload_mc_reg_table(struct amdgpu_device *adev,
 static void si_enable_voltage_control(struct amdgpu_device *adev, bool enable)
 {
 	if (enable)
-		WREG32_P(GENERAL_PWRMGT, VOLT_PWRMGT_EN, ~VOLT_PWRMGT_EN);
+		WREG32_P(mmGENERAL_PWRMGT, GENERAL_PWRMGT__VOLT_PWRMGT_EN_MASK, ~GENERAL_PWRMGT__VOLT_PWRMGT_EN_MASK);
 	else
-		WREG32_P(GENERAL_PWRMGT, 0, ~VOLT_PWRMGT_EN);
+		WREG32_P(mmGENERAL_PWRMGT, 0, ~GENERAL_PWRMGT__VOLT_PWRMGT_EN_MASK);
 }
 
 static enum si_pcie_gen si_get_maximum_link_speed(struct amdgpu_device *adev,
@@ -6204,8 +6226,8 @@ static u16 si_get_current_pcie_speed(struct amdgpu_device *adev)
 {
 	u32 speed_cntl;
 
-	speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL) & LC_CURRENT_DATA_RATE_MASK;
-	speed_cntl >>= LC_CURRENT_DATA_RATE_SHIFT;
+	speed_cntl = RREG32_PCIE_PORT(ixPCIE_LC_SPEED_CNTL) & PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK;
+	speed_cntl >>= PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT;
 
 	return (u16)speed_cntl;
 }
@@ -6412,21 +6434,21 @@ static void si_dpm_setup_asic(struct amdgpu_device *adev)
 static int si_thermal_enable_alert(struct amdgpu_device *adev,
 				   bool enable)
 {
-	u32 thermal_int = RREG32(CG_THERMAL_INT);
+	u32 thermal_int = RREG32(mmCG_THERMAL_INT);
 
 	if (enable) {
 		PPSMC_Result result;
 
-		thermal_int &= ~(THERM_INT_MASK_HIGH | THERM_INT_MASK_LOW);
-		WREG32(CG_THERMAL_INT, thermal_int);
+		thermal_int &= ~(CG_THERMAL_INT__THERM_INT_MASK_HIGH_MASK | CG_THERMAL_INT__THERM_INT_MASK_LOW_MASK);
+		WREG32(mmCG_THERMAL_INT, thermal_int);
 		result = amdgpu_si_send_msg_to_smc(adev, PPSMC_MSG_EnableThermalInterrupt);
 		if (result != PPSMC_Result_OK) {
 			DRM_DEBUG_KMS("Could not enable thermal interrupts.\n");
 			return -EINVAL;
 		}
 	} else {
-		thermal_int |= THERM_INT_MASK_HIGH | THERM_INT_MASK_LOW;
-		WREG32(CG_THERMAL_INT, thermal_int);
+		thermal_int |= CG_THERMAL_INT__THERM_INT_MASK_HIGH_MASK | CG_THERMAL_INT__THERM_INT_MASK_LOW_MASK;
+		WREG32(mmCG_THERMAL_INT, thermal_int);
 	}
 
 	return 0;
@@ -6447,9 +6469,9 @@ static int si_thermal_set_temperature_range(struct amdgpu_device *adev,
 		return -EINVAL;
 	}
 
-	WREG32_P(CG_THERMAL_INT, DIG_THERM_INTH(high_temp / 1000), ~DIG_THERM_INTH_MASK);
-	WREG32_P(CG_THERMAL_INT, DIG_THERM_INTL(low_temp / 1000), ~DIG_THERM_INTL_MASK);
-	WREG32_P(CG_THERMAL_CTRL, DIG_THERM_DPM(high_temp / 1000), ~DIG_THERM_DPM_MASK);
+	WREG32_P(mmCG_THERMAL_INT, (high_temp / 1000) << CG_THERMAL_INT__DIG_THERM_INTH__SHIFT, ~CG_THERMAL_INT__DIG_THERM_INTH_MASK);
+	WREG32_P(mmCG_THERMAL_INT, (low_temp / 1000) << CG_THERMAL_INT__DIG_THERM_INTL__SHIFT, ~CG_THERMAL_INT__DIG_THERM_INTL_MASK);
+	WREG32_P(mmCG_THERMAL_CTRL, (high_temp / 1000) << CG_THERMAL_CTRL__DIG_THERM_DPM__SHIFT, ~CG_THERMAL_CTRL__DIG_THERM_DPM_MASK);
 
 	adev->pm.dpm.thermal.min_temp = low_temp;
 	adev->pm.dpm.thermal.max_temp = high_temp;
@@ -6463,20 +6485,20 @@ static void si_fan_ctrl_set_static_mode(struct amdgpu_device *adev, u32 mode)
 	u32 tmp;
 
 	if (si_pi->fan_ctrl_is_in_default_mode) {
-		tmp = (RREG32(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK) >> FDO_PWM_MODE_SHIFT;
+		tmp = (RREG32(mmCG_FDO_CTRL2) & CG_FDO_CTRL2__FDO_PWM_MODE_MASK) >> CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT;
 		si_pi->fan_ctrl_default_mode = tmp;
-		tmp = (RREG32(CG_FDO_CTRL2) & TMIN_MASK) >> TMIN_SHIFT;
+		tmp = (RREG32(mmCG_FDO_CTRL2) & CG_FDO_CTRL2__TMIN_MASK) >> CG_FDO_CTRL2__TMIN__SHIFT;
 		si_pi->t_min = tmp;
 		si_pi->fan_ctrl_is_in_default_mode = false;
 	}
 
-	tmp = RREG32(CG_FDO_CTRL2) & ~TMIN_MASK;
-	tmp |= TMIN(0);
-	WREG32(CG_FDO_CTRL2, tmp);
+	tmp = RREG32(mmCG_FDO_CTRL2) & ~CG_FDO_CTRL2__TMIN_MASK;
+	tmp |= 0 << CG_FDO_CTRL2__TMIN__SHIFT;
+	WREG32(mmCG_FDO_CTRL2, tmp);
 
-	tmp = RREG32(CG_FDO_CTRL2) & ~FDO_PWM_MODE_MASK;
-	tmp |= FDO_PWM_MODE(mode);
-	WREG32(CG_FDO_CTRL2, tmp);
+	tmp = RREG32(mmCG_FDO_CTRL2) & ~CG_FDO_CTRL2__FDO_PWM_MODE_MASK;
+	tmp |= mode << CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT;
+	WREG32(mmCG_FDO_CTRL2, tmp);
 }
 
 static int si_thermal_setup_fan_table(struct amdgpu_device *adev)
@@ -6495,7 +6517,7 @@ static int si_thermal_setup_fan_table(struct amdgpu_device *adev)
 		return 0;
 	}
 
-	duty100 = (RREG32(CG_FDO_CTRL1) & FMAX_DUTY100_MASK) >> FMAX_DUTY100_SHIFT;
+	duty100 = (RREG32(mmCG_FDO_CTRL1) & CG_FDO_CTRL1__FMAX_DUTY100_MASK) >> CG_FDO_CTRL1__FMAX_DUTY100__SHIFT;
 
 	if (duty100 == 0) {
 		adev->pm.dpm.fan.ucode_fan_control = false;
@@ -6531,7 +6553,7 @@ static int si_thermal_setup_fan_table(struct amdgpu_device *adev)
 						reference_clock) / 1600);
 	fan_table.fdo_max = cpu_to_be16((u16)duty100);
 
-	tmp = (RREG32(CG_MULT_THERMAL_CTRL) & TEMP_SEL_MASK) >> TEMP_SEL_SHIFT;
+	tmp = (RREG32(mmCG_MULT_THERMAL_CTRL) & CG_MULT_THERMAL_CTRL__TEMP_SEL_MASK) >> CG_MULT_THERMAL_CTRL__TEMP_SEL__SHIFT;
 	fan_table.temp_src = (uint8_t)tmp;
 
 	ret = amdgpu_si_copy_bytes_to_smc(adev,
@@ -6590,8 +6612,8 @@ static int si_dpm_get_fan_speed_pwm(void *handle,
 	if (adev->pm.no_fan)
 		return -ENOENT;
 
-	duty100 = (RREG32(CG_FDO_CTRL1) & FMAX_DUTY100_MASK) >> FMAX_DUTY100_SHIFT;
-	duty = (RREG32(CG_THERMAL_STATUS) & FDO_PWM_DUTY_MASK) >> FDO_PWM_DUTY_SHIFT;
+	duty100 = (RREG32(mmCG_FDO_CTRL1) & CG_FDO_CTRL1__FMAX_DUTY100_MASK) >> CG_FDO_CTRL1__FMAX_DUTY100__SHIFT;
+	duty = (RREG32(mmCG_THERMAL_STATUS) & CG_THERMAL_STATUS__FDO_PWM_DUTY_MASK) >> CG_THERMAL_STATUS__FDO_PWM_DUTY__SHIFT;
 
 	if (duty100 == 0)
 		return -EINVAL;
@@ -6621,7 +6643,7 @@ static int si_dpm_set_fan_speed_pwm(void *handle,
 	if (speed > 255)
 		return -EINVAL;
 
-	duty100 = (RREG32(CG_FDO_CTRL1) & FMAX_DUTY100_MASK) >> FMAX_DUTY100_SHIFT;
+	duty100 = (RREG32(mmCG_FDO_CTRL1) & CG_FDO_CTRL1__FMAX_DUTY100_MASK) >> CG_FDO_CTRL1__FMAX_DUTY100__SHIFT;
 
 	if (duty100 == 0)
 		return -EINVAL;
@@ -6630,9 +6652,9 @@ static int si_dpm_set_fan_speed_pwm(void *handle,
 	do_div(tmp64, 255);
 	duty = (u32)tmp64;
 
-	tmp = RREG32(CG_FDO_CTRL0) & ~FDO_STATIC_DUTY_MASK;
-	tmp |= FDO_STATIC_DUTY(duty);
-	WREG32(CG_FDO_CTRL0, tmp);
+	tmp = RREG32(mmCG_FDO_CTRL0) & ~CG_FDO_CTRL0__FDO_STATIC_DUTY_MASK;
+	tmp |= duty << CG_FDO_CTRL0__FDO_STATIC_DUTY__SHIFT;
+	WREG32(mmCG_FDO_CTRL0, tmp);
 
 	return 0;
 }
@@ -6672,8 +6694,8 @@ static int si_dpm_get_fan_control_mode(void *handle, u32 *fan_mode)
 	if (si_pi->fan_is_controlled_by_smc)
 		return 0;
 
-	tmp = RREG32(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK;
-	*fan_mode = (tmp >> FDO_PWM_MODE_SHIFT);
+	tmp = RREG32(mmCG_FDO_CTRL2) & CG_FDO_CTRL2__FDO_PWM_MODE_MASK;
+	*fan_mode = (tmp >> CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT);
 
 	return 0;
 }
@@ -6691,7 +6713,7 @@ static int si_fan_ctrl_get_fan_speed_rpm(struct amdgpu_device *adev,
 	if (adev->pm.fan_pulses_per_revolution == 0)
 		return -ENOENT;
 
-	tach_period = (RREG32(CG_TACH_STATUS) & TACH_PERIOD_MASK) >> TACH_PERIOD_SHIFT;
+	tach_period = (RREG32(mmCG_TACH_STATUS) & CG_TACH_STATUS__TACH_PERIOD_MASK) >> CG_TACH_STATUS__TACH_PERIOD__SHIFT;
 	if (tach_period == 0)
 		return -ENOENT;
 
@@ -6720,9 +6742,9 @@ static int si_fan_ctrl_set_fan_speed_rpm(struct amdgpu_device *adev,
 		si_fan_ctrl_stop_smc_fan_control(adev);
 
 	tach_period = 60 * xclk * 10000 / (8 * speed);
-	tmp = RREG32(CG_TACH_CTRL) & ~TARGET_PERIOD_MASK;
-	tmp |= TARGET_PERIOD(tach_period);
-	WREG32(CG_TACH_CTRL, tmp);
+	tmp = RREG32(mmCG_TACH_CTRL) & ~CG_TACH_CTRL__TARGET_PERIOD_MASK;
+	tmp |= tach_period << CG_TACH_CTRL__TARGET_PERIOD__SHIFT;
+	WREG32(mmCG_TACH_CTRL, tmp);
 
 	si_fan_ctrl_set_static_mode(adev, FDO_PWM_MODE_STATIC_RPM);
 
@@ -6736,13 +6758,13 @@ static void si_fan_ctrl_set_default_mode(struct amdgpu_device *adev)
 	u32 tmp;
 
 	if (!si_pi->fan_ctrl_is_in_default_mode) {
-		tmp = RREG32(CG_FDO_CTRL2) & ~FDO_PWM_MODE_MASK;
-		tmp |= FDO_PWM_MODE(si_pi->fan_ctrl_default_mode);
-		WREG32(CG_FDO_CTRL2, tmp);
+		tmp = RREG32(mmCG_FDO_CTRL2) & ~CG_FDO_CTRL2__FDO_PWM_MODE_MASK;
+		tmp |= si_pi->fan_ctrl_default_mode << CG_FDO_CTRL2__FDO_PWM_MODE__SHIFT;
+		WREG32(mmCG_FDO_CTRL2, tmp);
 
-		tmp = RREG32(CG_FDO_CTRL2) & ~TMIN_MASK;
-		tmp |= TMIN(si_pi->t_min);
-		WREG32(CG_FDO_CTRL2, tmp);
+		tmp = RREG32(mmCG_FDO_CTRL2) & ~CG_FDO_CTRL2__TMIN_MASK;
+		tmp |= si_pi->t_min << CG_FDO_CTRL2__TMIN__SHIFT;
+		WREG32(mmCG_FDO_CTRL2, tmp);
 		si_pi->fan_ctrl_is_in_default_mode = true;
 	}
 }
@@ -6760,14 +6782,14 @@ static void si_thermal_initialize(struct amdgpu_device *adev)
 	u32 tmp;
 
 	if (adev->pm.fan_pulses_per_revolution) {
-		tmp = RREG32(CG_TACH_CTRL) & ~EDGE_PER_REV_MASK;
-		tmp |= EDGE_PER_REV(adev->pm.fan_pulses_per_revolution -1);
-		WREG32(CG_TACH_CTRL, tmp);
+		tmp = RREG32(mmCG_TACH_CTRL) & ~CG_TACH_CTRL__EDGE_PER_REV_MASK;
+		tmp |= (adev->pm.fan_pulses_per_revolution -1) << CG_TACH_CTRL__EDGE_PER_REV__SHIFT;
+		WREG32(mmCG_TACH_CTRL, tmp);
 	}
 
-	tmp = RREG32(CG_FDO_CTRL2) & ~TACH_PWM_RESP_RATE_MASK;
-	tmp |= TACH_PWM_RESP_RATE(0x28);
-	WREG32(CG_FDO_CTRL2, tmp);
+	tmp = RREG32(mmCG_FDO_CTRL2) & ~CG_FDO_CTRL2__TACH_PWM_RESP_RATE_MASK;
+	tmp |= 0x28 << CG_FDO_CTRL2__TACH_PWM_RESP_RATE__SHIFT;
+	WREG32(mmCG_FDO_CTRL2, tmp);
 }
 
 static int si_thermal_start_thermal_controller(struct amdgpu_device *adev)
@@ -7530,8 +7552,8 @@ static void si_dpm_debugfs_print_current_performance_level(void *handle,
 	struct  si_ps *ps = si_get_ps(rps);
 	struct rv7xx_pl *pl;
 	u32 current_index =
-		(RREG32(TARGET_AND_CURRENT_PROFILE_INDEX) & CURRENT_STATE_INDEX_MASK) >>
-		CURRENT_STATE_INDEX_SHIFT;
+		(RREG32(mmTARGET_AND_CURRENT_PROFILE_INDEX) & TARGET_AND_CURRENT_PROFILE_INDEX__CURRENT_STATE_INDEX_MASK) >>
+			TARGET_AND_CURRENT_PROFILE_INDEX__CURRENT_STATE_INDEX__SHIFT;
 
 	if (current_index >= ps->performance_level_count) {
 		seq_printf(m, "invalid dpm profile %d\n", current_index);
@@ -7554,14 +7576,14 @@ static int si_dpm_set_interrupt_state(struct amdgpu_device *adev,
 	case AMDGPU_THERMAL_IRQ_LOW_TO_HIGH:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
-			cg_thermal_int = RREG32_SMC(CG_THERMAL_INT);
-			cg_thermal_int |= THERM_INT_MASK_HIGH;
-			WREG32_SMC(CG_THERMAL_INT, cg_thermal_int);
+			cg_thermal_int = RREG32_SMC(mmCG_THERMAL_INT);
+			cg_thermal_int |= CG_THERMAL_INT__THERM_INT_MASK_HIGH_MASK;
+			WREG32_SMC(mmCG_THERMAL_INT, cg_thermal_int);
 			break;
 		case AMDGPU_IRQ_STATE_ENABLE:
-			cg_thermal_int = RREG32_SMC(CG_THERMAL_INT);
-			cg_thermal_int &= ~THERM_INT_MASK_HIGH;
-			WREG32_SMC(CG_THERMAL_INT, cg_thermal_int);
+			cg_thermal_int = RREG32_SMC(mmCG_THERMAL_INT);
+			cg_thermal_int &= ~CG_THERMAL_INT__THERM_INT_MASK_HIGH_MASK;
+			WREG32_SMC(mmCG_THERMAL_INT, cg_thermal_int);
 			break;
 		default:
 			break;
@@ -7571,14 +7593,14 @@ static int si_dpm_set_interrupt_state(struct amdgpu_device *adev,
 	case AMDGPU_THERMAL_IRQ_HIGH_TO_LOW:
 		switch (state) {
 		case AMDGPU_IRQ_STATE_DISABLE:
-			cg_thermal_int = RREG32_SMC(CG_THERMAL_INT);
-			cg_thermal_int |= THERM_INT_MASK_LOW;
-			WREG32_SMC(CG_THERMAL_INT, cg_thermal_int);
+			cg_thermal_int = RREG32_SMC(mmCG_THERMAL_INT);
+			cg_thermal_int |= CG_THERMAL_INT__THERM_INT_MASK_LOW_MASK;
+			WREG32_SMC(mmCG_THERMAL_INT, cg_thermal_int);
 			break;
 		case AMDGPU_IRQ_STATE_ENABLE:
-			cg_thermal_int = RREG32_SMC(CG_THERMAL_INT);
-			cg_thermal_int &= ~THERM_INT_MASK_LOW;
-			WREG32_SMC(CG_THERMAL_INT, cg_thermal_int);
+			cg_thermal_int = RREG32_SMC(mmCG_THERMAL_INT);
+			cg_thermal_int &= ~CG_THERMAL_INT__THERM_INT_MASK_LOW_MASK;
+			WREG32_SMC(mmCG_THERMAL_INT, cg_thermal_int);
 			break;
 		default:
 			break;
@@ -7883,8 +7905,8 @@ static int si_dpm_get_temp(void *handle)
 	int actual_temp = 0;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	temp = (RREG32(CG_MULT_THERMAL_STATUS) & CTF_TEMP_MASK) >>
-		CTF_TEMP_SHIFT;
+	temp = (RREG32(mmCG_MULT_THERMAL_STATUS) & CG_MULT_THERMAL_STATUS__CTF_TEMP_MASK) >>
+		CG_MULT_THERMAL_STATUS__CTF_TEMP__SHIFT;
 
 	if (temp & 0x200)
 		actual_temp = 255;
@@ -8014,8 +8036,8 @@ static int si_dpm_read_sensor(void *handle, int idx,
 	struct  si_ps *ps = si_get_ps(rps);
 	uint32_t sclk, mclk;
 	u32 pl_index =
-		(RREG32(TARGET_AND_CURRENT_PROFILE_INDEX) & CURRENT_STATE_INDEX_MASK) >>
-		CURRENT_STATE_INDEX_SHIFT;
+		(RREG32(mmTARGET_AND_CURRENT_PROFILE_INDEX) & TARGET_AND_CURRENT_PROFILE_INDEX__CURRENT_STATE_INDEX_MASK) >>
+		TARGET_AND_CURRENT_PROFILE_INDEX__CURRENT_STATE_INDEX__SHIFT;
 
 	/* size must be at least 4 bytes for all sensors */
 	if (*size < 4)
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c
index 8f994ffa9cd1..4e65ab9e931c 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c
@@ -30,6 +30,12 @@
 #include "amdgpu_ucode.h"
 #include "sislands_smc.h"
 
+#include "smu/smu_6_0_d.h"
+#include "smu/smu_6_0_sh_mask.h"
+
+#include "gca/gfx_6_0_d.h"
+#include "gca/gfx_6_0_sh_mask.h"
+
 static int si_set_smc_sram_address(struct amdgpu_device *adev,
 				   u32 smc_address, u32 limit)
 {
@@ -38,8 +44,8 @@ static int si_set_smc_sram_address(struct amdgpu_device *adev,
 	if ((smc_address + 3) > limit)
 		return -EINVAL;
 
-	WREG32(SMC_IND_INDEX_0, smc_address);
-	WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0);
+	WREG32(mmSMC_IND_INDEX_0, smc_address);
+	WREG32_P(mmSMC_IND_ACCESS_CNTL, 0, ~SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK);
 
 	return 0;
 }
@@ -68,7 +74,7 @@ int amdgpu_si_copy_bytes_to_smc(struct amdgpu_device *adev,
 		if (ret)
 			goto done;
 
-		WREG32(SMC_IND_DATA_0, data);
+		WREG32(mmSMC_IND_DATA_0, data);
 
 		src += 4;
 		byte_count -= 4;
@@ -83,7 +89,7 @@ int amdgpu_si_copy_bytes_to_smc(struct amdgpu_device *adev,
 		if (ret)
 			goto done;
 
-		original_data = RREG32(SMC_IND_DATA_0);
+		original_data = RREG32(mmSMC_IND_DATA_0);
 		extra_shift = 8 * (4 - byte_count);
 
 		while (byte_count > 0) {
@@ -99,7 +105,7 @@ int amdgpu_si_copy_bytes_to_smc(struct amdgpu_device *adev,
 		if (ret)
 			goto done;
 
-		WREG32(SMC_IND_DATA_0, data);
+		WREG32(mmSMC_IND_DATA_0, data);
 	}
 
 done:
@@ -121,10 +127,10 @@ void amdgpu_si_reset_smc(struct amdgpu_device *adev)
 {
 	u32 tmp;
 
-	RREG32(CB_CGTT_SCLK_CTRL);
-	RREG32(CB_CGTT_SCLK_CTRL);
-	RREG32(CB_CGTT_SCLK_CTRL);
-	RREG32(CB_CGTT_SCLK_CTRL);
+	RREG32(mmCB_CGTT_SCLK_CTRL);
+	RREG32(mmCB_CGTT_SCLK_CTRL);
+	RREG32(mmCB_CGTT_SCLK_CTRL);
+	RREG32(mmCB_CGTT_SCLK_CTRL);
 
 	tmp = RREG32_SMC(SMC_SYSCON_RESET_CNTL) |
 	      RST_REG;
@@ -170,16 +176,16 @@ PPSMC_Result amdgpu_si_send_msg_to_smc(struct amdgpu_device *adev,
 	if (!amdgpu_si_is_smc_running(adev))
 		return PPSMC_Result_Failed;
 
-	WREG32(SMC_MESSAGE_0, msg);
+	WREG32(mmSMC_MESSAGE_0, msg);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(SMC_RESP_0);
+		tmp = RREG32(mmSMC_RESP_0);
 		if (tmp != 0)
 			break;
 		udelay(1);
 	}
 
-	return (PPSMC_Result)RREG32(SMC_RESP_0);
+	return (PPSMC_Result)RREG32(mmSMC_RESP_0);
 }
 
 PPSMC_Result amdgpu_si_wait_for_smc_inactive(struct amdgpu_device *adev)
@@ -225,18 +231,18 @@ int amdgpu_si_load_smc_ucode(struct amdgpu_device *adev, u32 limit)
 		return -EINVAL;
 
 	spin_lock_irqsave(&adev->smc_idx_lock, flags);
-	WREG32(SMC_IND_INDEX_0, ucode_start_address);
-	WREG32_P(SMC_IND_ACCESS_CNTL, AUTO_INCREMENT_IND_0, ~AUTO_INCREMENT_IND_0);
+	WREG32(mmSMC_IND_INDEX_0, ucode_start_address);
+	WREG32_P(mmSMC_IND_ACCESS_CNTL, SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK, ~SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK);
 	while (ucode_size >= 4) {
 		/* SMC address space is BE */
 		data = (src[0] << 24) | (src[1] << 16) | (src[2] << 8) | src[3];
 
-		WREG32(SMC_IND_DATA_0, data);
+		WREG32(mmSMC_IND_DATA_0, data);
 
 		src += 4;
 		ucode_size -= 4;
 	}
-	WREG32_P(SMC_IND_ACCESS_CNTL, 0, ~AUTO_INCREMENT_IND_0);
+	WREG32_P(mmSMC_IND_ACCESS_CNTL, 0, ~SMC_IND_ACCESS_CNTL__AUTO_INCREMENT_IND_0_MASK);
 	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
 
 	return 0;
@@ -251,7 +257,7 @@ int amdgpu_si_read_smc_sram_dword(struct amdgpu_device *adev, u32 smc_address,
 	spin_lock_irqsave(&adev->smc_idx_lock, flags);
 	ret = si_set_smc_sram_address(adev, smc_address, limit);
 	if (ret == 0)
-		*value = RREG32(SMC_IND_DATA_0);
+		*value = RREG32(mmSMC_IND_DATA_0);
 	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
 
 	return ret;
@@ -266,7 +272,7 @@ int amdgpu_si_write_smc_sram_dword(struct amdgpu_device *adev, u32 smc_address,
 	spin_lock_irqsave(&adev->smc_idx_lock, flags);
 	ret = si_set_smc_sram_address(adev, smc_address, limit);
 	if (ret == 0)
-		WREG32(SMC_IND_DATA_0, value);
+		WREG32(mmSMC_IND_DATA_0, value);
 	spin_unlock_irqrestore(&adev->smc_idx_lock, flags);
 
 	return ret;
diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
index 4bd92fd782be..8d40ed0f0e83 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
@@ -143,6 +143,10 @@ int atomctrl_initialize_mc_reg_table(
 	vram_info = (ATOM_VRAM_INFO_HEADER_V2_1 *)
 		smu_atom_get_data_table(hwmgr->adev,
 				GetIndexIntoMasterTable(DATA, VRAM_Info), &size, &frev, &crev);
+	if (!vram_info) {
+		pr_err("Could not retrieve the VramInfo table!");
+		return -EINVAL;
+	}
 
 	if (module_index >= vram_info->ucNumOfVRAMModule) {
 		pr_err("Invalid VramInfo table.");
@@ -180,6 +184,10 @@ int atomctrl_initialize_mc_reg_table_v2_2(
 	vram_info = (ATOM_VRAM_INFO_HEADER_V2_2 *)
 		smu_atom_get_data_table(hwmgr->adev,
 				GetIndexIntoMasterTable(DATA, VRAM_Info), &size, &frev, &crev);
+	if (!vram_info) {
+		pr_err("Could not retrieve the VramInfo table!");
+		return -EINVAL;
+	}
 
 	if (module_index >= vram_info->ucNumOfVRAMModule) {
 		pr_err("Invalid VramInfo table.");
diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
index 5a010cd38303..baf51cd82a35 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
@@ -46,42 +46,6 @@ static int smu7_set_smc_sram_address(struct pp_hwmgr *hwmgr, uint32_t smc_addr,
 }
 
 
-int smu7_copy_bytes_from_smc(struct pp_hwmgr *hwmgr, uint32_t smc_start_address, uint32_t *dest, uint32_t byte_count, uint32_t limit)
-{
-	uint32_t data;
-	uint32_t addr;
-	uint8_t *dest_byte;
-	uint8_t i, data_byte[4] = {0};
-	uint32_t *pdata = (uint32_t *)&data_byte;
-
-	PP_ASSERT_WITH_CODE((0 == (3 & smc_start_address)), "SMC address must be 4 byte aligned.", return -EINVAL);
-	PP_ASSERT_WITH_CODE((limit > (smc_start_address + byte_count)), "SMC address is beyond the SMC RAM area.", return -EINVAL);
-
-	addr = smc_start_address;
-
-	while (byte_count >= 4) {
-		smu7_read_smc_sram_dword(hwmgr, addr, &data, limit);
-
-		*dest = PP_SMC_TO_HOST_UL(data);
-
-		dest += 1;
-		byte_count -= 4;
-		addr += 4;
-	}
-
-	if (byte_count) {
-		smu7_read_smc_sram_dword(hwmgr, addr, &data, limit);
-		*pdata = PP_SMC_TO_HOST_UL(data);
-	/* Cast dest into byte type in dest_byte.  This way, we don't overflow if the allocated memory is not 4-byte aligned. */
-		dest_byte = (uint8_t *)dest;
-		for (i = 0; i < byte_count; i++)
-			dest_byte[i] = data_byte[i];
-	}
-
-	return 0;
-}
-
-
 int smu7_copy_bytes_to_smc(struct pp_hwmgr *hwmgr, uint32_t smc_start_address,
 				const uint8_t *src, uint32_t byte_count, uint32_t limit)
 {
diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.h
index e7303dc8c260..63e428ceaee4 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.h
@@ -53,8 +53,6 @@ struct smu7_smumgr {
 };
 
 
-int smu7_copy_bytes_from_smc(struct pp_hwmgr *hwmgr, uint32_t smc_start_address,
-				uint32_t *dest, uint32_t byte_count, uint32_t limit);
 int smu7_copy_bytes_to_smc(struct pp_hwmgr *hwmgr, uint32_t smc_start_address,
 			const uint8_t *src, uint32_t byte_count, uint32_t limit);
 int smu7_program_jump_on_start(struct pp_hwmgr *hwmgr);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 033c3229b555..f24a1d8c77db 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -2398,7 +2398,11 @@ static int smu_switch_power_profile(void *handle,
 			smu_power_profile_mode_get(smu, type);
 		else
 			smu_power_profile_mode_put(smu, type);
-		ret = smu_bump_power_profile_mode(smu, NULL, 0);
+		/* don't switch the active workload when paused */
+		if (smu->pause_workload)
+			ret = 0;
+		else
+			ret = smu_bump_power_profile_mode(smu, NULL, 0);
 		if (ret) {
 			if (enable)
 				smu_power_profile_mode_put(smu, type);
@@ -2411,6 +2415,35 @@ static int smu_switch_power_profile(void *handle,
 	return 0;
 }
 
+static int smu_pause_power_profile(void *handle,
+				   bool pause)
+{
+	struct smu_context *smu = handle;
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
+	u32 workload_mask = 1 << PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+	int ret;
+
+	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
+		return -EOPNOTSUPP;
+
+	if (smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL &&
+	    smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_PERF_DETERMINISM) {
+		smu->pause_workload = pause;
+
+		/* force to bootup default profile */
+		if (smu->pause_workload && smu->ppt_funcs->set_power_profile_mode)
+			ret = smu->ppt_funcs->set_power_profile_mode(smu,
+								     workload_mask,
+								     NULL,
+								     0);
+		else
+			ret = smu_bump_power_profile_mode(smu, NULL, 0);
+		return ret;
+	}
+
+	return 0;
+}
+
 static enum amd_dpm_forced_level smu_get_performance_level(void *handle)
 {
 	struct smu_context *smu = handle;
@@ -3399,15 +3432,15 @@ bool smu_mode1_reset_is_support(struct smu_context *smu)
 	return ret;
 }
 
-bool smu_mode2_reset_is_support(struct smu_context *smu)
+bool smu_link_reset_is_support(struct smu_context *smu)
 {
 	bool ret = false;
 
 	if (!smu->pm_enabled)
 		return false;
 
-	if (smu->ppt_funcs && smu->ppt_funcs->mode2_reset_is_support)
-		ret = smu->ppt_funcs->mode2_reset_is_support(smu);
+	if (smu->ppt_funcs && smu->ppt_funcs->link_reset_is_support)
+		ret = smu->ppt_funcs->link_reset_is_support(smu);
 
 	return ret;
 }
@@ -3442,6 +3475,19 @@ static int smu_mode2_reset(void *handle)
 	return ret;
 }
 
+int smu_link_reset(struct smu_context *smu)
+{
+	int ret = 0;
+
+	if (!smu->pm_enabled)
+		return -EOPNOTSUPP;
+
+	if (smu->ppt_funcs->link_reset)
+		ret = smu->ppt_funcs->link_reset(smu);
+
+	return ret;
+}
+
 static int smu_enable_gfx_features(void *handle)
 {
 	struct smu_context *smu = handle;
@@ -3733,6 +3779,7 @@ static const struct amd_pm_funcs swsmu_pm_funcs = {
 	.get_pp_table            = smu_sys_get_pp_table,
 	.set_pp_table            = smu_sys_set_pp_table,
 	.switch_power_profile    = smu_switch_power_profile,
+	.pause_power_profile     = smu_pause_power_profile,
 	/* export to amdgpu */
 	.dispatch_tasks          = smu_handle_dpm_task,
 	.load_firmware           = smu_load_microcode,
@@ -3941,3 +3988,11 @@ int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask)
 
 	return ret;
 }
+
+int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask)
+{
+	if (smu->ppt_funcs && smu->ppt_funcs->dpm_reset_vcn)
+		smu->ppt_funcs->dpm_reset_vcn(smu, inst_mask);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 3ba169639f54..d47e32ae4671 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -438,9 +438,11 @@ struct mclock_latency_table {
 };
 
 enum smu_reset_mode {
-    SMU_RESET_MODE_0,
-    SMU_RESET_MODE_1,
-    SMU_RESET_MODE_2,
+	SMU_RESET_MODE_0,
+	SMU_RESET_MODE_1,
+	SMU_RESET_MODE_2,
+	SMU_RESET_MODE_3,
+	SMU_RESET_MODE_4,
 };
 
 enum smu_baco_state {
@@ -558,6 +560,7 @@ struct smu_context {
 
 	/* asic agnostic workload mask */
 	uint32_t workload_mask;
+	bool pause_workload;
 	/* default/user workload preference */
 	uint32_t power_profile_mode;
 	uint32_t workload_refcount[PP_SMC_POWER_PROFILE_COUNT];
@@ -1228,10 +1231,11 @@ struct pptable_funcs {
 	 * @mode1_reset_is_support: Check if GPU supports mode1 reset.
 	 */
 	bool (*mode1_reset_is_support)(struct smu_context *smu);
+
 	/**
-	 * @mode2_reset_is_support: Check if GPU supports mode2 reset.
+	 * @link_reset_is_support: Check if GPU supports link reset.
 	 */
-	bool (*mode2_reset_is_support)(struct smu_context *smu);
+	bool (*link_reset_is_support)(struct smu_context *smu);
 
 	/**
 	 * @mode1_reset: Perform mode1 reset.
@@ -1251,6 +1255,13 @@ struct pptable_funcs {
 	int (*enable_gfx_features)(struct smu_context *smu);
 
 	/**
+	 * @link_reset: Perform link reset.
+	 *
+	 * The gfx device driver reset
+	 */
+	int (*link_reset)(struct smu_context *smu);
+
+	/**
 	 * @get_dpm_ultimate_freq: Get the hard frequency range of a clock
 	 *                         domain in MHz.
 	 */
@@ -1382,6 +1393,11 @@ struct pptable_funcs {
 	bool (*reset_sdma_is_supported)(struct smu_context *smu);
 
 	/**
+	 * @reset_vcn: message SMU to soft reset vcn instance.
+	 */
+	int (*dpm_reset_vcn)(struct smu_context *smu, uint32_t inst_mask);
+
+	/**
 	 * @get_ecc_table:  message SMU to get ECC INFO table.
 	 */
 	ssize_t (*get_ecc_info)(struct smu_context *smu, void *table);
@@ -1600,8 +1616,9 @@ int smu_get_power_limit(void *handle,
 			enum pp_power_type pp_power_type);
 
 bool smu_mode1_reset_is_support(struct smu_context *smu);
-bool smu_mode2_reset_is_support(struct smu_context *smu);
+bool smu_link_reset_is_support(struct smu_context *smu);
 int smu_mode1_reset(struct smu_context *smu);
+int smu_link_reset(struct smu_context *smu);
 
 extern const struct amd_ip_funcs smu_ip_funcs;
 
@@ -1642,6 +1659,7 @@ int smu_send_hbm_bad_channel_flag(struct smu_context *smu, uint32_t size);
 int smu_send_rma_reason(struct smu_context *smu);
 int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask);
 bool smu_reset_sdma_is_supported(struct smu_context *smu);
+int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask);
 int smu_set_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type,
 		      int level);
 ssize_t smu_get_pm_policy_info(struct smu_context *smu,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
index d26f35119a12..3d9e5e967c94 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
@@ -459,4 +459,11 @@ typedef struct __attribute__((packed, aligned(4))) {
   uint64_t AccGfxclkBelowHostLimit;
 } VfMetricsTable_t;
 
+#pragma pack(push, 4)
+typedef struct {
+  // Telemetry
+  uint32_t InputTelemetryVoltageInmV;
+} StaticMetricsTable_t;
+#pragma pack(pop)
+
 #endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
index 288b2576432b..41f268313613 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
@@ -94,7 +94,9 @@
 #define PPSMC_MSG_RmaDueToBadPageThreshold          0x43
 #define PPSMC_MSG_SetThrottlingPolicy               0x44
 #define PPSMC_MSG_ResetSDMA                         0x4D
-#define PPSMC_Message_Count                         0x4E
+#define PPSMC_MSG_ResetVCN                          0x4E
+#define PPSMC_MSG_GetStaticMetricsTable             0x59
+#define PPSMC_Message_Count                         0x5A
 
 //PPSMC Reset Types for driver msg argument
 #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET        0x1
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
index c9dee09395e3..eefdaa0b5df6 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
@@ -277,6 +277,7 @@
 	__SMU_DUMMY_MAP(MALLPowerController), \
 	__SMU_DUMMY_MAP(MALLPowerState), \
 	__SMU_DUMMY_MAP(ResetSDMA), \
+	__SMU_DUMMY_MAP(ResetVCN), \
 	__SMU_DUMMY_MAP(GetStaticMetricsTable),
 
 #undef __SMU_DUMMY_MAP
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h
index ed8304d82831..56ae555bb52a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h
@@ -281,11 +281,6 @@ int smu_v11_0_set_single_dpm_table(struct smu_context *smu,
 				   enum smu_clk_type clk_type,
 				   struct smu_11_0_dpm_table *single_dpm_table);
 
-int smu_v11_0_get_dpm_level_range(struct smu_context *smu,
-				  enum smu_clk_type clk_type,
-				  uint32_t *min_value,
-				  uint32_t *max_value);
-
 int smu_v11_0_get_current_pcie_link_width_level(struct smu_context *smu);
 
 uint16_t smu_v11_0_get_current_pcie_link_width(struct smu_context *smu);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
index cd03caffe317..4263798d716b 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
@@ -112,6 +112,7 @@ struct smu_13_0_dpm_context {
 	uint32_t                    workload_policy_mask;
 	uint32_t                    dcef_min_ds_clk;
 	uint64_t                    caps;
+	uint32_t                    board_volt;
 };
 
 enum smu_13_0_power_state {
@@ -162,8 +163,6 @@ int smu_v13_0_notify_memory_pool_location(struct smu_context *smu);
 int smu_v13_0_system_features_control(struct smu_context *smu,
 				      bool en);
 
-int smu_v13_0_init_display_count(struct smu_context *smu, uint32_t count);
-
 int smu_v13_0_set_allowed_mask(struct smu_context *smu);
 
 int smu_v13_0_notify_display_change(struct smu_context *smu);
@@ -183,13 +182,6 @@ int smu_v13_0_disable_thermal_alert(struct smu_context *smu);
 
 int smu_v13_0_get_gfx_vdd(struct smu_context *smu, uint32_t *value);
 
-int smu_v13_0_set_min_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk);
-
-int
-smu_v13_0_display_clock_voltage_request(struct smu_context *smu,
-					struct pp_display_clock_request
-					*clock_req);
-
 uint32_t
 smu_v13_0_get_fan_control_mode(struct smu_context *smu);
 
@@ -226,11 +218,6 @@ int smu_v13_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c
 int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type,
 					  uint32_t min, uint32_t max, bool automatic);
 
-int smu_v13_0_set_hard_freq_limited_range(struct smu_context *smu,
-					  enum smu_clk_type clk_type,
-					  uint32_t min,
-					  uint32_t max);
-
 int smu_v13_0_set_performance_level(struct smu_context *smu,
 				    enum amd_dpm_forced_level level);
 
@@ -310,14 +297,6 @@ int smu_v13_0_get_boot_freq_by_index(struct smu_context *smu,
 				     uint32_t *value);
 
 void smu_v13_0_interrupt_work(struct smu_context *smu);
-bool smu_v13_0_12_is_dpm_running(struct smu_context *smu);
-int smu_v13_0_12_get_max_metrics_size(void);
-int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu);
-int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
-				      MetricsMember_t member,
-				      uint32_t *value);
-ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table);
-extern const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[];
-extern const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[];
+void smu_v13_0_reset_custom_level(struct smu_context *smu);
 #endif
 #endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
index 453952cdc353..9ad46f545d15 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
@@ -1347,7 +1347,7 @@ static int arcturus_get_power_limit(struct smu_context *smu,
 		*default_power_limit = power_limit;
 	if (max_power_limit)
 		*max_power_limit = power_limit;
-	/**
+	/*
 	 * No lower bound is imposed on the limit. Any unreasonable limit set
 	 * will result in frequent throttling.
 	 */
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index 19a25fdc2f5b..115e3fa456bc 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -3089,11 +3089,6 @@ static int sienna_cichlid_stb_get_data_direct(struct smu_context *smu,
 	return 0;
 }
 
-static bool sienna_cichlid_is_mode2_reset_supported(struct smu_context *smu)
-{
-	return true;
-}
-
 static int sienna_cichlid_mode2_reset(struct smu_context *smu)
 {
 	int ret = 0, index;
@@ -3229,7 +3224,6 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
 	.get_default_config_table_settings = sienna_cichlid_get_default_config_table_settings,
 	.set_config_table = sienna_cichlid_set_config_table,
 	.get_unique_id = sienna_cichlid_get_unique_id,
-	.mode2_reset_is_support = sienna_cichlid_is_mode2_reset_supported,
 	.mode2_reset = sienna_cichlid_mode2_reset,
 };
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
index 78391d8f35a9..78e4186d06cc 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c
@@ -1204,7 +1204,7 @@ int smu_v11_0_set_fan_speed_rpm(struct smu_context *smu,
 	uint32_t crystal_clock_freq = 2500;
 	uint32_t tach_period;
 
-	if (speed == 0)
+	if (!speed || speed > UINT_MAX/8)
 		return -EINVAL;
 	/*
 	 * To prevent from possible overheat, some ASICs may have requirement
@@ -2059,45 +2059,6 @@ int smu_v11_0_set_single_dpm_table(struct smu_context *smu,
 	return 0;
 }
 
-int smu_v11_0_get_dpm_level_range(struct smu_context *smu,
-				  enum smu_clk_type clk_type,
-				  uint32_t *min_value,
-				  uint32_t *max_value)
-{
-	uint32_t level_count = 0;
-	int ret = 0;
-
-	if (!min_value && !max_value)
-		return -EINVAL;
-
-	if (min_value) {
-		/* by default, level 0 clock value as min value */
-		ret = smu_v11_0_get_dpm_freq_by_index(smu,
-						      clk_type,
-						      0,
-						      min_value);
-		if (ret)
-			return ret;
-	}
-
-	if (max_value) {
-		ret = smu_v11_0_get_dpm_level_count(smu,
-						    clk_type,
-						    &level_count);
-		if (ret)
-			return ret;
-
-		ret = smu_v11_0_get_dpm_freq_by_index(smu,
-						      clk_type,
-						      level_count - 1,
-						      max_value);
-		if (ret)
-			return ret;
-	}
-
-	return ret;
-}
-
 int smu_v11_0_get_current_pcie_link_width_level(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
index 83163d7c7f00..6de653d2ed62 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
@@ -1270,6 +1270,7 @@ static int aldebaran_set_performance_level(struct smu_context *smu,
 	struct smu_13_0_dpm_table *gfx_table =
 		&dpm_context->dpm_tables.gfx_table;
 	struct smu_umd_pstate_table *pstate_table = &smu->pstate_table;
+	int r;
 
 	/* Disable determinism if switching to another mode */
 	if ((smu_dpm->dpm_level == AMD_DPM_FORCED_LEVEL_PERF_DETERMINISM) &&
@@ -1282,7 +1283,11 @@ static int aldebaran_set_performance_level(struct smu_context *smu,
 
 	case AMD_DPM_FORCED_LEVEL_PERF_DETERMINISM:
 		return 0;
-
+	case AMD_DPM_FORCED_LEVEL_AUTO:
+		r = smu_v13_0_set_performance_level(smu, level);
+		if (!r)
+			smu_v13_0_reset_custom_level(smu);
+		return r;
 	case AMD_DPM_FORCED_LEVEL_HIGH:
 	case AMD_DPM_FORCED_LEVEL_LOW:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
@@ -1423,7 +1428,11 @@ static int aldebaran_usr_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_
 			min_clk = dpm_context->dpm_tables.gfx_table.min;
 			max_clk = dpm_context->dpm_tables.gfx_table.max;
 
-			return aldebaran_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk, false);
+			ret = aldebaran_set_soft_freq_limited_range(
+				smu, SMU_GFXCLK, min_clk, max_clk, false);
+			if (ret)
+				return ret;
+			smu_v13_0_reset_custom_level(smu);
 		}
 		break;
 	case PP_OD_COMMIT_DPM_TABLE:
@@ -1976,11 +1985,6 @@ static bool aldebaran_is_mode1_reset_supported(struct smu_context *smu)
 	return true;
 }
 
-static bool aldebaran_is_mode2_reset_supported(struct smu_context *smu)
-{
-	return true;
-}
-
 static int aldebaran_set_mp1_state(struct smu_context *smu,
 				   enum pp_mp1_state mp1_state)
 {
@@ -2086,7 +2090,6 @@ static const struct pptable_funcs aldebaran_ppt_funcs = {
 	.set_pp_feature_mask = smu_cmn_set_pp_feature_mask,
 	.get_gpu_metrics = aldebaran_get_gpu_metrics,
 	.mode1_reset_is_support = aldebaran_is_mode1_reset_supported,
-	.mode2_reset_is_support = aldebaran_is_mode2_reset_supported,
 	.smu_handle_passthrough_sbr = aldebaran_smu_handle_passthrough_sbr,
 	.mode1_reset = aldebaran_mode1_reset,
 	.set_mp1_state = aldebaran_set_mp1_state,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index ba5a9012dbd5..a7167668d189 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -709,18 +709,6 @@ int smu_v13_0_notify_memory_pool_location(struct smu_context *smu)
 	return ret;
 }
 
-int smu_v13_0_set_min_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk)
-{
-	int ret;
-
-	ret = smu_cmn_send_smc_msg_with_param(smu,
-					      SMU_MSG_SetMinDeepSleepDcefclk, clk, NULL);
-	if (ret)
-		dev_err(smu->adev->dev, "SMU13 attempt to set divider for DCEFCLK Failed!");
-
-	return ret;
-}
-
 int smu_v13_0_set_driver_table_location(struct smu_context *smu)
 {
 	struct smu_table *driver_table = &smu->smu_table.driver_table;
@@ -761,18 +749,6 @@ int smu_v13_0_set_tool_table_location(struct smu_context *smu)
 	return ret;
 }
 
-int smu_v13_0_init_display_count(struct smu_context *smu, uint32_t count)
-{
-	int ret = 0;
-
-	if (!smu->pm_enabled)
-		return ret;
-
-	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, count, NULL);
-
-	return ret;
-}
-
 int smu_v13_0_set_allowed_mask(struct smu_context *smu)
 {
 	struct smu_feature *feature = &smu->smu_feature;
@@ -1073,56 +1049,6 @@ int smu_v13_0_get_gfx_vdd(struct smu_context *smu, uint32_t *value)
 
 }
 
-int
-smu_v13_0_display_clock_voltage_request(struct smu_context *smu,
-					struct pp_display_clock_request
-					*clock_req)
-{
-	enum amd_pp_clock_type clk_type = clock_req->clock_type;
-	int ret = 0;
-	enum smu_clk_type clk_select = 0;
-	uint32_t clk_freq = clock_req->clock_freq_in_khz / 1000;
-
-	if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) ||
-	    smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT)) {
-		switch (clk_type) {
-		case amd_pp_dcef_clock:
-			clk_select = SMU_DCEFCLK;
-			break;
-		case amd_pp_disp_clock:
-			clk_select = SMU_DISPCLK;
-			break;
-		case amd_pp_pixel_clock:
-			clk_select = SMU_PIXCLK;
-			break;
-		case amd_pp_phy_clock:
-			clk_select = SMU_PHYCLK;
-			break;
-		case amd_pp_mem_clock:
-			clk_select = SMU_UCLK;
-			break;
-		default:
-			dev_info(smu->adev->dev, "[%s] Invalid Clock Type!", __func__);
-			ret = -EINVAL;
-			break;
-		}
-
-		if (ret)
-			goto failed;
-
-		if (clk_select == SMU_UCLK && smu->disable_uclk_switch)
-			return 0;
-
-		ret = smu_v13_0_set_hard_freq_limited_range(smu, clk_select, clk_freq, 0);
-
-		if (clk_select == SMU_UCLK)
-			smu->hard_min_uclk_req_from_dal = clk_freq;
-	}
-
-failed:
-	return ret;
-}
-
 uint32_t smu_v13_0_get_fan_control_mode(struct smu_context *smu)
 {
 	if (!smu_cmn_feature_is_enabled(smu, SMU_FEATURE_FAN_CONTROL_BIT))
@@ -1647,45 +1573,6 @@ out:
 	return ret;
 }
 
-int smu_v13_0_set_hard_freq_limited_range(struct smu_context *smu,
-					  enum smu_clk_type clk_type,
-					  uint32_t min,
-					  uint32_t max)
-{
-	int ret = 0, clk_id = 0;
-	uint32_t param;
-
-	if (min <= 0 && max <= 0)
-		return -EINVAL;
-
-	if (!smu_cmn_clk_dpm_is_enabled(smu, clk_type))
-		return 0;
-
-	clk_id = smu_cmn_to_asic_specific_index(smu,
-						CMN2ASIC_MAPPING_CLK,
-						clk_type);
-	if (clk_id < 0)
-		return clk_id;
-
-	if (max > 0) {
-		param = (uint32_t)((clk_id << 16) | (max & 0xffff));
-		ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetHardMaxByFreq,
-						      param, NULL);
-		if (ret)
-			return ret;
-	}
-
-	if (min > 0) {
-		param = (uint32_t)((clk_id << 16) | (min & 0xffff));
-		ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetHardMinByFreq,
-						      param, NULL);
-		if (ret)
-			return ret;
-	}
-
-	return ret;
-}
-
 int smu_v13_0_set_performance_level(struct smu_context *smu,
 				    enum amd_dpm_forced_level level)
 {
@@ -2595,3 +2482,13 @@ int smu_v13_0_set_wbrf_exclusion_ranges(struct smu_context *smu,
 
 	return ret;
 }
+
+void smu_v13_0_reset_custom_level(struct smu_context *smu)
+{
+	struct smu_umd_pstate_table *pstate_table = &smu->pstate_table;
+
+	pstate_table->uclk_pstate.custom.min = 0;
+	pstate_table->uclk_pstate.custom.max = 0;
+	pstate_table->gfxclk_pstate.custom.min = 0;
+	pstate_table->gfxclk_pstate.custom.max = 0;
+}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 238bd71baa6d..533d58e57d05 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -187,26 +187,6 @@ int smu_v13_0_12_get_max_metrics_size(void)
 	return max(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t));
 }
 
-static int smu_v13_0_12_get_static_metrics_table(struct smu_context *smu)
-{
-	struct smu_table_context *smu_table = &smu->smu_table;
-	uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size;
-	struct smu_table *table = &smu_table->driver_table;
-	int ret;
-
-	ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetStaticMetricsTable, NULL);
-	if (ret) {
-		dev_info(smu->adev->dev,
-			 "Failed to export static metrics table!\n");
-		return ret;
-	}
-
-	amdgpu_asic_invalidate_hdp(smu->adev, NULL);
-	memcpy(smu_table->metrics_table, table->cpu_addr, table_size);
-
-	return 0;
-}
-
 int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
@@ -217,7 +197,7 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu)
 	int ret, i;
 
 	if (!pptable->Init) {
-		ret = smu_v13_0_12_get_static_metrics_table(smu);
+		ret = smu_v13_0_6_get_static_metrics_table(smu);
 		if (ret)
 			return ret;
 
@@ -345,8 +325,8 @@ int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
 ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
-	struct gpu_metrics_v1_7 *gpu_metrics =
-		(struct gpu_metrics_v1_7 *)smu_table->gpu_metrics_table;
+	struct gpu_metrics_v1_8 *gpu_metrics =
+		(struct gpu_metrics_v1_8 *)smu_table->gpu_metrics_table;
 	int ret = 0, xcc_id, inst, i, j, k, idx;
 	struct amdgpu_device *adev = smu->adev;
 	u8 num_jpeg_rings_gpu_metrics;
@@ -357,7 +337,7 @@ ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table)
 	metrics = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL);
 	memcpy(metrics, smu_table->metrics_table, sizeof(MetricsTable_t));
 
-	smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 7);
+	smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 8);
 
 	gpu_metrics->temperature_hotspot =
 		SMUQ10_ROUND(metrics->MaxSocketTemperature);
@@ -474,6 +454,16 @@ ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table)
 				SMUQ10_ROUND(metrics->GfxBusy[inst]);
 			gpu_metrics->xcp_stats[i].gfx_busy_acc[idx] =
 				SMUQ10_ROUND(metrics->GfxBusyAcc[inst]);
+			if (smu_v13_0_6_cap_supported(smu, SMU_CAP(HST_LIMIT_METRICS))) {
+				gpu_metrics->xcp_stats[i].gfx_below_host_limit_ppt_acc[idx] =
+					SMUQ10_ROUND(metrics->GfxclkBelowHostLimitPptAcc[inst]);
+				gpu_metrics->xcp_stats[i].gfx_below_host_limit_thm_acc[idx] =
+					SMUQ10_ROUND(metrics->GfxclkBelowHostLimitThmAcc[inst]);
+				gpu_metrics->xcp_stats[i].gfx_low_utilization_acc[idx] =
+					SMUQ10_ROUND(metrics->GfxclkLowUtilizationAcc[inst]);
+				gpu_metrics->xcp_stats[i].gfx_below_host_limit_total_acc[idx] =
+					SMUQ10_ROUND(metrics->GfxclkBelowHostLimitTotalAcc[inst]);
+			}
 			idx++;
 		}
 	}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index c478b3be37af..7d4ff09be7e8 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -101,24 +101,6 @@ MODULE_FIRMWARE("amdgpu/smu_13_0_14.bin");
 #define MCA_BANK_IPID(_ip, _hwid, _type) \
 	[AMDGPU_MCA_IP_##_ip] = { .hwid = _hwid, .mcatype = _type, }
 
-#define SMU_CAP(x) SMU_13_0_6_CAPS_##x
-
-enum smu_v13_0_6_caps {
-	SMU_CAP(DPM),
-	SMU_CAP(DPM_POLICY),
-	SMU_CAP(OTHER_END_METRICS),
-	SMU_CAP(SET_UCLK_MAX),
-	SMU_CAP(PCIE_METRICS),
-	SMU_CAP(MCA_DEBUG_MODE),
-	SMU_CAP(PER_INST_METRICS),
-	SMU_CAP(CTF_LIMIT),
-	SMU_CAP(RMA_MSG),
-	SMU_CAP(ACA_SYND),
-	SMU_CAP(SDMA_RESET),
-	SMU_CAP(STATIC_METRICS),
-	SMU_CAP(ALL),
-};
-
 struct mca_bank_ipid {
 	enum amdgpu_mca_ip ip;
 	uint16_t hwid;
@@ -194,6 +176,8 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU
 	MSG_MAP(RmaDueToBadPageThreshold,            PPSMC_MSG_RmaDueToBadPageThreshold,        0),
 	MSG_MAP(SetThrottlingPolicy,                 PPSMC_MSG_SetThrottlingPolicy,             0),
 	MSG_MAP(ResetSDMA,                           PPSMC_MSG_ResetSDMA,                       0),
+	MSG_MAP(ResetVCN,                            PPSMC_MSG_ResetVCN,                       0),
+	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           0),
 };
 
 // clang-format on
@@ -299,8 +283,8 @@ static inline void smu_v13_0_6_cap_clear(struct smu_context *smu,
 	dpm_context->caps &= ~BIT_ULL(cap);
 }
 
-static inline bool smu_v13_0_6_cap_supported(struct smu_context *smu,
-					     enum smu_v13_0_6_caps cap)
+bool smu_v13_0_6_cap_supported(struct smu_context *smu,
+			       enum smu_v13_0_6_caps cap)
 {
 	struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context;
 
@@ -353,6 +337,9 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu)
 
 	if (fw_ver >= 0x00561E00)
 		smu_v13_0_6_cap_set(smu, SMU_CAP(STATIC_METRICS));
+
+	if (fw_ver >= 0x00562500)
+		smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS));
 }
 
 static void smu_v13_0_6_init_caps(struct smu_context *smu)
@@ -402,6 +389,13 @@ static void smu_v13_0_6_init_caps(struct smu_context *smu)
 			smu_v13_0_6_cap_clear(smu, SMU_CAP(RMA_MSG));
 		if (fw_ver < 0x00555600)
 			smu_v13_0_6_cap_clear(smu, SMU_CAP(ACA_SYND));
+		if ((pgm == 7 && fw_ver >= 0x7550E00) ||
+		    (pgm == 0 && fw_ver >= 0x00557E00))
+			smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS));
+		if (fw_ver >= 0x00557F01) {
+			smu_v13_0_6_cap_set(smu, SMU_CAP(STATIC_METRICS));
+			smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE));
+		}
 	}
 	if (((pgm == 7) && (fw_ver >= 0x7550700)) ||
 	    ((pgm == 0) && (fw_ver >= 0x00557900)) ||
@@ -525,7 +519,7 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 		return -ENOMEM;
 	smu_table->metrics_time = 0;
 
-	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_7);
+	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_8);
 	smu_table->gpu_metrics_table =
 		kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL);
 	if (!smu_table->gpu_metrics_table) {
@@ -747,9 +741,43 @@ static ssize_t smu_v13_0_6_get_pm_metrics(struct smu_context *smu,
 	return pm_metrics->common_header.structure_size;
 }
 
+static void smu_v13_0_6_fill_static_metrics_table(struct smu_context *smu,
+						  StaticMetricsTable_t *static_metrics)
+{
+	struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context;
+
+	if (!static_metrics->InputTelemetryVoltageInmV) {
+		dev_warn(smu->adev->dev, "Invalid board voltage %d\n",
+				static_metrics->InputTelemetryVoltageInmV);
+	}
+
+	dpm_context->board_volt = static_metrics->InputTelemetryVoltageInmV;
+}
+
+int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size;
+	struct smu_table *table = &smu_table->driver_table;
+	int ret;
+
+	ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetStaticMetricsTable, NULL);
+	if (ret) {
+		dev_info(smu->adev->dev,
+				"Failed to export static metrics table!\n");
+		return ret;
+	}
+
+	amdgpu_asic_invalidate_hdp(smu->adev, NULL);
+	memcpy(smu_table->metrics_table, table->cpu_addr, table_size);
+
+	return 0;
+}
+
 static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
+	StaticMetricsTable_t *static_metrics = (StaticMetricsTable_t *)smu_table->metrics_table;
 	MetricsTableV0_t *metrics_v0 = (MetricsTableV0_t *)smu_table->metrics_table;
 	MetricsTableV1_t *metrics_v1 = (MetricsTableV1_t *)smu_table->metrics_table;
 	MetricsTableV2_t *metrics_v2 = (MetricsTableV2_t *)smu_table->metrics_table;
@@ -759,7 +787,8 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu)
 	int ret, i, retry = 100;
 	uint32_t table_version;
 
-	if (smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) &&
+	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
 		return smu_v13_0_12_setup_driver_pptable(smu);
 
 	/* Store one-time values in driver PPTable */
@@ -813,6 +842,12 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu)
 			GET_METRIC_FIELD(PublicSerialNumber_AID, version)[0];
 
 		pptable->Init = true;
+		if (smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) {
+			ret = smu_v13_0_6_get_static_metrics_table(smu);
+			if (ret)
+				return ret;
+			smu_v13_0_6_fill_static_metrics_table(smu, static_metrics);
+		}
 	}
 
 	return 0;
@@ -1142,7 +1177,8 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu,
 	if (ret)
 		return ret;
 
-	if (smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) &&
+	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
 		return smu_v13_0_12_get_smu_metrics_data(smu, member, value);
 
 	/* For clocks with multiple instances, only report the first one */
@@ -1616,6 +1652,7 @@ static int smu_v13_0_6_read_sensor(struct smu_context *smu,
 				   enum amd_pp_sensors sensor, void *data,
 				   uint32_t *size)
 {
+	struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context;
 	int ret = 0;
 
 	if (amdgpu_ras_intr_triggered())
@@ -1660,6 +1697,15 @@ static int smu_v13_0_6_read_sensor(struct smu_context *smu,
 		ret = smu_v13_0_get_gfx_vdd(smu, (uint32_t *)data);
 		*size = 4;
 		break;
+	case AMDGPU_PP_SENSOR_VDDBOARD:
+		if (smu_v13_0_6_cap_supported(smu, SMU_CAP(BOARD_VOLTAGE))) {
+			*(uint32_t *)data = dpm_context->board_volt;
+			*size = 4;
+			break;
+		} else {
+			ret = -EOPNOTSUPP;
+			break;
+		}
 	case AMDGPU_PP_SENSOR_GPU_AVG_POWER:
 	default:
 		ret = -EOPNOTSUPP;
@@ -1927,7 +1973,7 @@ static int smu_v13_0_6_set_performance_level(struct smu_context *smu,
 				return ret;
 			pstate_table->uclk_pstate.curr.max = uclk_table->max;
 		}
-		pstate_table->uclk_pstate.custom.max = 0;
+		smu_v13_0_reset_custom_level(smu);
 
 		return 0;
 	case AMD_DPM_FORCED_LEVEL_MANUAL:
@@ -2140,7 +2186,7 @@ static int smu_v13_0_6_usr_edit_dpm_table(struct smu_context *smu,
 				smu, SMU_UCLK, min_clk, max_clk, false);
 			if (ret)
 				return ret;
-			pstate_table->uclk_pstate.custom.max = 0;
+			smu_v13_0_reset_custom_level(smu);
 		}
 		break;
 	case PP_OD_COMMIT_DPM_TABLE:
@@ -2486,8 +2532,8 @@ static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu)
 static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
-	struct gpu_metrics_v1_7 *gpu_metrics =
-		(struct gpu_metrics_v1_7 *)smu_table->gpu_metrics_table;
+	struct gpu_metrics_v1_8 *gpu_metrics =
+		(struct gpu_metrics_v1_8 *)smu_table->gpu_metrics_table;
 	int version = smu_v13_0_6_get_metrics_version(smu);
 	int ret = 0, xcc_id, inst, i, j, k, idx;
 	struct amdgpu_device *adev = smu->adev;
@@ -2507,13 +2553,14 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 		return ret;
 	}
 
-	if (smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) &&
+	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
 		return smu_v13_0_12_get_gpu_metrics(smu, table);
 
 	metrics_v1 = (MetricsTableV1_t *)metrics_v0;
 	metrics_v2 = (MetricsTableV2_t *)metrics_v0;
 
-	smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 7);
+	smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 8);
 
 	gpu_metrics->temperature_hotspot =
 		SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature, version));
@@ -2666,6 +2713,20 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 				gpu_metrics->xcp_stats[i].gfx_busy_acc[idx] =
 					SMUQ10_ROUND(GET_GPU_METRIC_FIELD(GfxBusyAcc,
 									  version)[inst]);
+				if (smu_v13_0_6_cap_supported(smu, SMU_CAP(HST_LIMIT_METRICS))) {
+					gpu_metrics->xcp_stats[i].gfx_below_host_limit_ppt_acc[idx] =
+						SMUQ10_ROUND
+						(metrics_v0->GfxclkBelowHostLimitPptAcc[inst]);
+					gpu_metrics->xcp_stats[i].gfx_below_host_limit_thm_acc[idx] =
+						SMUQ10_ROUND
+						(metrics_v0->GfxclkBelowHostLimitThmAcc[inst]);
+					gpu_metrics->xcp_stats[i].gfx_low_utilization_acc[idx] =
+						SMUQ10_ROUND
+						(metrics_v0->GfxclkLowUtilizationAcc[inst]);
+					gpu_metrics->xcp_stats[i].gfx_below_host_limit_total_acc[idx] =
+						SMUQ10_ROUND
+						(metrics_v0->GfxclkBelowHostLimitTotalAcc[inst]);
+				}
 				idx++;
 			}
 		}
@@ -2844,14 +2905,29 @@ static int smu_v13_0_6_mode1_reset(struct smu_context *smu)
 	return ret;
 }
 
+static int smu_v13_0_6_link_reset(struct smu_context *smu)
+{
+	int ret = 0;
+
+	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset,
+					      SMU_RESET_MODE_4, NULL);
+	return ret;
+}
+
 static bool smu_v13_0_6_is_mode1_reset_supported(struct smu_context *smu)
 {
 	return true;
 }
 
-static bool smu_v13_0_6_is_mode2_reset_supported(struct smu_context *smu)
+static inline bool smu_v13_0_6_is_link_reset_supported(struct smu_context *smu)
 {
-	return true;
+	struct amdgpu_device *adev = smu->adev;
+	int var = (adev->pdev->device & 0xF);
+
+	if (var == 0x1)
+		return true;
+
+	return false;
 }
 
 static int smu_v13_0_6_smu_send_hbm_bad_page_num(struct smu_context *smu,
@@ -2924,6 +3000,19 @@ static int smu_v13_0_6_reset_sdma(struct smu_context *smu, uint32_t inst_mask)
 	return ret;
 }
 
+static int smu_v13_0_6_reset_vcn(struct smu_context *smu, uint32_t inst_mask)
+{
+	int ret = 0;
+
+	ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_ResetVCN, inst_mask, NULL);
+	if (ret)
+		dev_err(smu->adev->dev,
+			"failed to send ResetVCN event with mask 0x%x\n",
+			inst_mask);
+	return ret;
+}
+
+
 static int mca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable)
 {
 	struct smu_context *smu = adev->powerplay.pp_handle;
@@ -3586,9 +3675,10 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.get_pm_metrics = smu_v13_0_6_get_pm_metrics,
 	.get_thermal_temperature_range = smu_v13_0_6_get_thermal_temperature_range,
 	.mode1_reset_is_support = smu_v13_0_6_is_mode1_reset_supported,
-	.mode2_reset_is_support = smu_v13_0_6_is_mode2_reset_supported,
+	.link_reset_is_support = smu_v13_0_6_is_link_reset_supported,
 	.mode1_reset = smu_v13_0_6_mode1_reset,
 	.mode2_reset = smu_v13_0_6_mode2_reset,
+	.link_reset = smu_v13_0_6_link_reset,
 	.wait_for_event = smu_v13_0_wait_for_event,
 	.i2c_init = smu_v13_0_6_i2c_control_init,
 	.i2c_fini = smu_v13_0_6_i2c_control_fini,
@@ -3596,6 +3686,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.send_rma_reason = smu_v13_0_6_send_rma_reason,
 	.reset_sdma = smu_v13_0_6_reset_sdma,
 	.reset_sdma_is_supported = smu_v13_0_6_reset_sdma_is_supported,
+	.dpm_reset_vcn = smu_v13_0_6_reset_vcn,
 };
 
 void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index 83745909e564..d151bcd0cca7 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -26,6 +26,7 @@
 #define SMU_13_0_6_UMD_PSTATE_GFXCLK_LEVEL 0x2
 #define SMU_13_0_6_UMD_PSTATE_SOCCLK_LEVEL 0x4
 #define SMU_13_0_6_UMD_PSTATE_MCLK_LEVEL 0x2
+#define SMU_CAP(x) SMU_13_0_6_CAPS_##x
 
 typedef enum {
 /*0*/   METRICS_VERSION_V0                  = 0,
@@ -51,6 +52,34 @@ struct PPTable_t {
 	bool Init;
 };
 
+enum smu_v13_0_6_caps {
+	SMU_CAP(DPM),
+	SMU_CAP(DPM_POLICY),
+	SMU_CAP(OTHER_END_METRICS),
+	SMU_CAP(SET_UCLK_MAX),
+	SMU_CAP(PCIE_METRICS),
+	SMU_CAP(MCA_DEBUG_MODE),
+	SMU_CAP(PER_INST_METRICS),
+	SMU_CAP(CTF_LIMIT),
+	SMU_CAP(RMA_MSG),
+	SMU_CAP(ACA_SYND),
+	SMU_CAP(SDMA_RESET),
+	SMU_CAP(STATIC_METRICS),
+	SMU_CAP(HST_LIMIT_METRICS),
+	SMU_CAP(BOARD_VOLTAGE),
+	SMU_CAP(ALL),
+};
+
 extern void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu);
+bool smu_v13_0_6_cap_supported(struct smu_context *smu, enum smu_v13_0_6_caps cap);
+int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu);
 
+bool smu_v13_0_12_is_dpm_running(struct smu_context *smu);
+int smu_v13_0_12_get_max_metrics_size(void);
+int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu);
+int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
+				      MetricsMember_t member, uint32_t *value);
+ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table);
+extern const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[];
+extern const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[];
 #endif
diff --git a/drivers/gpu/drm/ast/ast_cursor.c b/drivers/gpu/drm/ast/ast_cursor.c
index 139ab00dee8f..2d3ad7610c2e 100644
--- a/drivers/gpu/drm/ast/ast_cursor.c
+++ b/drivers/gpu/drm/ast/ast_cursor.c
@@ -37,6 +37,7 @@
  */
 
 /* define for signature structure */
+#define AST_HWC_SIGNATURE_SIZE		SZ_32
 #define AST_HWC_SIGNATURE_CHECKSUM	0x00
 #define AST_HWC_SIGNATURE_SizeX		0x04
 #define AST_HWC_SIGNATURE_SizeY		0x08
@@ -45,6 +46,21 @@
 #define AST_HWC_SIGNATURE_HOTSPOTX	0x14
 #define AST_HWC_SIGNATURE_HOTSPOTY	0x18
 
+static unsigned long ast_cursor_vram_size(void)
+{
+	return AST_HWC_SIZE + AST_HWC_SIGNATURE_SIZE;
+}
+
+long ast_cursor_vram_offset(struct ast_device *ast)
+{
+	unsigned long size = ast_cursor_vram_size();
+
+	if (size > ast->vram_size)
+		return -EINVAL;
+
+	return ALIGN_DOWN(ast->vram_size - size, SZ_8);
+}
+
 static u32 ast_cursor_calculate_checksum(const void *src, unsigned int width, unsigned int height)
 {
 	u32 csum = 0;
@@ -75,7 +91,7 @@ static u32 ast_cursor_calculate_checksum(const void *src, unsigned int width, un
 static void ast_set_cursor_image(struct ast_device *ast, const u8 *src,
 				 unsigned int width, unsigned int height)
 {
-	u8 __iomem *dst = ast->cursor_plane.base.vaddr;
+	u8 __iomem *dst = ast_plane_vaddr(&ast->cursor_plane.base);
 	u32 csum;
 
 	csum = ast_cursor_calculate_checksum(src, width, height);
@@ -177,7 +193,7 @@ static void ast_cursor_plane_helper_atomic_update(struct drm_plane *plane,
 	struct ast_device *ast = to_ast_device(plane->dev);
 	struct drm_rect damage;
 	u64 dst_off = ast_plane->offset;
-	u8 __iomem *dst = ast_plane->vaddr; /* TODO: Use mapping abstraction properly */
+	u8 __iomem *dst = ast_plane_vaddr(ast_plane); /* TODO: Use mapping abstraction properly */
 	u8 __iomem *sig = dst + AST_HWC_SIZE; /* TODO: Use mapping abstraction properly */
 	unsigned int offset_x, offset_y;
 	u16 x, y;
@@ -274,25 +290,16 @@ int ast_cursor_plane_init(struct ast_device *ast)
 	struct ast_cursor_plane *ast_cursor_plane = &ast->cursor_plane;
 	struct ast_plane *ast_plane = &ast_cursor_plane->base;
 	struct drm_plane *cursor_plane = &ast_plane->base;
-	size_t size;
-	void __iomem *vaddr;
-	u64 offset;
+	unsigned long size;
+	long offset;
 	int ret;
 
-	/*
-	 * Allocate backing storage for cursors. The BOs are permanently
-	 * pinned to the top end of the VRAM.
-	 */
-
-	size = roundup(AST_HWC_SIZE + AST_HWC_SIGNATURE_SIZE, PAGE_SIZE);
+	size = ast_cursor_vram_size();
+	offset = ast_cursor_vram_offset(ast);
+	if (offset < 0)
+		return offset;
 
-	if (ast->vram_fb_available < size)
-		return -ENOMEM;
-
-	vaddr = ast->vram + ast->vram_fb_available - size;
-	offset = ast->vram_fb_available - size;
-
-	ret = ast_plane_init(dev, ast_plane, vaddr, offset, size,
+	ret = ast_plane_init(dev, ast_plane, offset, size,
 			     0x01, &ast_cursor_plane_funcs,
 			     ast_cursor_plane_formats, ARRAY_SIZE(ast_cursor_plane_formats),
 			     NULL, DRM_PLANE_TYPE_CURSOR);
@@ -303,7 +310,5 @@ int ast_cursor_plane_init(struct ast_device *ast)
 	drm_plane_helper_add(cursor_plane, &ast_cursor_plane_helper_funcs);
 	drm_plane_enable_fb_damage_clips(cursor_plane);
 
-	ast->vram_fb_available -= size;
-
 	return 0;
 }
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index d2c2605d2728..2ee402096cd9 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -112,12 +112,9 @@ enum ast_config_mode {
 
 #define AST_MAX_HWC_WIDTH	64
 #define AST_MAX_HWC_HEIGHT	64
-
 #define AST_HWC_PITCH		(AST_MAX_HWC_WIDTH * SZ_2)
 #define AST_HWC_SIZE		(AST_MAX_HWC_HEIGHT * AST_HWC_PITCH)
 
-#define AST_HWC_SIGNATURE_SIZE	32
-
 /*
  * Planes
  */
@@ -125,7 +122,6 @@ enum ast_config_mode {
 struct ast_plane {
 	struct drm_plane base;
 
-	void __iomem *vaddr;
 	u64 offset;
 	unsigned long size;
 };
@@ -183,7 +179,6 @@ struct ast_device {
 	void __iomem	*vram;
 	unsigned long	vram_base;
 	unsigned long	vram_size;
-	unsigned long	vram_fb_available;
 
 	struct mutex modeset_lock; /* Protects access to modeset I/O registers in ioregs */
 
@@ -340,14 +335,6 @@ static inline void ast_set_index_reg_mask(struct ast_device *ast, u32 base, u8 i
 	__ast_write8_i_masked(ast->ioregs, base, index, preserve_mask, val);
 }
 
-#define AST_VIDMEM_SIZE_8M    0x00800000
-#define AST_VIDMEM_SIZE_16M   0x01000000
-#define AST_VIDMEM_SIZE_32M   0x02000000
-#define AST_VIDMEM_SIZE_64M   0x04000000
-#define AST_VIDMEM_SIZE_128M  0x08000000
-
-#define AST_VIDMEM_DEFAULT_SIZE AST_VIDMEM_SIZE_8M
-
 struct ast_vbios_stdtable {
 	u8 misc;
 	u8 seq[4];
@@ -440,6 +427,7 @@ int ast_vga_output_init(struct ast_device *ast);
 int ast_sil164_output_init(struct ast_device *ast);
 
 /* ast_cursor.c */
+long ast_cursor_vram_offset(struct ast_device *ast);
 int ast_cursor_plane_init(struct ast_device *ast);
 
 /* ast dp501 */
@@ -454,11 +442,12 @@ int ast_astdp_output_init(struct ast_device *ast);
 /* ast_mode.c */
 int ast_mode_config_init(struct ast_device *ast);
 int ast_plane_init(struct drm_device *dev, struct ast_plane *ast_plane,
-		   void __iomem *vaddr, u64 offset, unsigned long size,
+		   u64 offset, unsigned long size,
 		   uint32_t possible_crtcs,
 		   const struct drm_plane_funcs *funcs,
 		   const uint32_t *formats, unsigned int format_count,
 		   const uint64_t *format_modifiers,
 		   enum drm_plane_type type);
+void __iomem *ast_plane_vaddr(struct ast_plane *ast);
 
 #endif
diff --git a/drivers/gpu/drm/ast/ast_mm.c b/drivers/gpu/drm/ast/ast_mm.c
index 6dfe6d9777d4..0bc140319464 100644
--- a/drivers/gpu/drm/ast/ast_mm.c
+++ b/drivers/gpu/drm/ast/ast_mm.c
@@ -35,36 +35,35 @@
 
 static u32 ast_get_vram_size(struct ast_device *ast)
 {
-	u8 jreg;
 	u32 vram_size;
+	u8 vgacr99, vgacraa;
 
-	vram_size = AST_VIDMEM_DEFAULT_SIZE;
-	jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xaa, 0xff);
-	switch (jreg & 3) {
+	vgacraa = ast_get_index_reg(ast, AST_IO_VGACRI, 0xaa);
+	switch (vgacraa & AST_IO_VGACRAA_VGAMEM_SIZE_MASK) {
 	case 0:
-		vram_size = AST_VIDMEM_SIZE_8M;
+		vram_size = SZ_8M;
 		break;
 	case 1:
-		vram_size = AST_VIDMEM_SIZE_16M;
+		vram_size = SZ_16M;
 		break;
 	case 2:
-		vram_size = AST_VIDMEM_SIZE_32M;
+		vram_size = SZ_32M;
 		break;
 	case 3:
-		vram_size = AST_VIDMEM_SIZE_64M;
+		vram_size = SZ_64M;
 		break;
 	}
 
-	jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0x99, 0xff);
-	switch (jreg & 0x03) {
+	vgacr99 = ast_get_index_reg(ast, AST_IO_VGACRI, 0x99);
+	switch (vgacr99 & AST_IO_VGACR99_VGAMEM_RSRV_MASK) {
 	case 1:
-		vram_size -= 0x100000;
+		vram_size -= SZ_1M;
 		break;
 	case 2:
-		vram_size -= 0x200000;
+		vram_size -= SZ_2M;
 		break;
 	case 3:
-		vram_size -= 0x400000;
+		vram_size -= SZ_4M;
 		break;
 	}
 
@@ -93,7 +92,6 @@ int ast_mm_init(struct ast_device *ast)
 
 	ast->vram_base = base;
 	ast->vram_size = vram_size;
-	ast->vram_fb_available = vram_size;
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index c3b950675485..1de832964e92 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -51,6 +51,26 @@
 
 #define AST_LUT_SIZE 256
 
+#define AST_PRIMARY_PLANE_MAX_OFFSET	(BIT(16) - 1)
+
+static unsigned long ast_fb_vram_offset(void)
+{
+	return 0; // with shmem, the primary plane is always at offset 0
+}
+
+static unsigned long ast_fb_vram_size(struct ast_device *ast)
+{
+	struct drm_device *dev = &ast->base;
+	unsigned long offset = ast_fb_vram_offset(); // starts at offset
+	long cursor_offset = ast_cursor_vram_offset(ast); // ends at cursor offset
+
+	if (cursor_offset < 0)
+		cursor_offset = ast->vram_size; // no cursor; it's all ours
+	if (drm_WARN_ON_ONCE(dev, offset > cursor_offset))
+		return 0; // cannot legally happen; signal error
+	return cursor_offset - offset;
+}
+
 static inline void ast_load_palette_index(struct ast_device *ast,
 				     u8 index, u8 red, u8 green,
 				     u8 blue)
@@ -439,7 +459,7 @@ static void ast_wait_for_vretrace(struct ast_device *ast)
  */
 
 int ast_plane_init(struct drm_device *dev, struct ast_plane *ast_plane,
-		   void __iomem *vaddr, u64 offset, unsigned long size,
+		   u64 offset, unsigned long size,
 		   uint32_t possible_crtcs,
 		   const struct drm_plane_funcs *funcs,
 		   const uint32_t *formats, unsigned int format_count,
@@ -448,7 +468,6 @@ int ast_plane_init(struct drm_device *dev, struct ast_plane *ast_plane,
 {
 	struct drm_plane *plane = &ast_plane->base;
 
-	ast_plane->vaddr = vaddr;
 	ast_plane->offset = offset;
 	ast_plane->size = size;
 
@@ -457,6 +476,13 @@ int ast_plane_init(struct drm_device *dev, struct ast_plane *ast_plane,
 					type, NULL);
 }
 
+void __iomem *ast_plane_vaddr(struct ast_plane *ast_plane)
+{
+	struct ast_device *ast = to_ast_device(ast_plane->base.dev);
+
+	return ast->vram + ast_plane->offset;
+}
+
 /*
  * Primary plane
  */
@@ -503,7 +529,7 @@ static void ast_handle_damage(struct ast_plane *ast_plane, struct iosys_map *src
 			      struct drm_framebuffer *fb,
 			      const struct drm_rect *clip)
 {
-	struct iosys_map dst = IOSYS_MAP_INIT_VADDR_IOMEM(ast_plane->vaddr);
+	struct iosys_map dst = IOSYS_MAP_INIT_VADDR_IOMEM(ast_plane_vaddr(ast_plane));
 
 	iosys_map_incr(&dst, drm_fb_clip_offset(fb->pitches[0], fb->format, clip));
 	drm_fb_memcpy(&dst, fb->pitches, src, fb, clip);
@@ -576,12 +602,12 @@ static int ast_primary_plane_helper_get_scanout_buffer(struct drm_plane *plane,
 {
 	struct ast_plane *ast_plane = to_ast_plane(plane);
 
-	if (plane->state && plane->state->fb && ast_plane->vaddr) {
+	if (plane->state && plane->state->fb) {
 		sb->format = plane->state->fb->format;
 		sb->width = plane->state->fb->width;
 		sb->height = plane->state->fb->height;
 		sb->pitch[0] = plane->state->fb->pitches[0];
-		iosys_map_set_vaddr_iomem(&sb->map[0], ast_plane->vaddr);
+		iosys_map_set_vaddr_iomem(&sb->map[0], ast_plane_vaddr(ast_plane));
 		return 0;
 	}
 	return -ENODEV;
@@ -608,13 +634,11 @@ static int ast_primary_plane_init(struct ast_device *ast)
 	struct drm_device *dev = &ast->base;
 	struct ast_plane *ast_primary_plane = &ast->primary_plane;
 	struct drm_plane *primary_plane = &ast_primary_plane->base;
-	void __iomem *vaddr = ast->vram;
-	u64 offset = 0; /* with shmem, the primary plane is always at offset 0 */
-	unsigned long cursor_size = roundup(AST_HWC_SIZE + AST_HWC_SIGNATURE_SIZE, PAGE_SIZE);
-	unsigned long size = ast->vram_fb_available - cursor_size;
+	u64 offset = ast_fb_vram_offset();
+	unsigned long size = ast_fb_vram_size(ast);
 	int ret;
 
-	ret = ast_plane_init(dev, ast_primary_plane, vaddr, offset, size,
+	ret = ast_plane_init(dev, ast_primary_plane, offset, size,
 			     0x01, &ast_primary_plane_funcs,
 			     ast_primary_plane_formats, ARRAY_SIZE(ast_primary_plane_formats),
 			     NULL, DRM_PLANE_TYPE_PRIMARY);
@@ -922,9 +946,9 @@ static void ast_mode_config_helper_atomic_commit_tail(struct drm_atomic_state *s
 
 	/*
 	 * Concurrent operations could possibly trigger a call to
-	 * drm_connector_helper_funcs.get_modes by trying to read the
-	 * display modes. Protect access to I/O registers by acquiring
-	 * the I/O-register lock. Released in atomic_flush().
+	 * drm_connector_helper_funcs.get_modes by reading the display
+	 * modes. Protect access to registers by acquiring the modeset
+	 * lock.
 	 */
 	mutex_lock(&ast->modeset_lock);
 	drm_atomic_helper_commit_tail(state);
@@ -938,16 +962,20 @@ static const struct drm_mode_config_helper_funcs ast_mode_config_helper_funcs =
 static enum drm_mode_status ast_mode_config_mode_valid(struct drm_device *dev,
 						       const struct drm_display_mode *mode)
 {
-	static const unsigned long max_bpp = 4; /* DRM_FORMAT_XRGB8888 */
+	const struct drm_format_info *info = drm_format_info(DRM_FORMAT_XRGB8888);
 	struct ast_device *ast = to_ast_device(dev);
-	unsigned long fbsize, fbpages, max_fbpages;
-
-	max_fbpages = (ast->vram_fb_available) >> PAGE_SHIFT;
-
-	fbsize = mode->hdisplay * mode->vdisplay * max_bpp;
-	fbpages = DIV_ROUND_UP(fbsize, PAGE_SIZE);
-
-	if (fbpages > max_fbpages)
+	unsigned long max_fb_size = ast_fb_vram_size(ast);
+	u64 pitch;
+
+	if (drm_WARN_ON_ONCE(dev, !info))
+		return MODE_ERROR; /* driver bug */
+
+	pitch = drm_format_info_min_pitch(info, 0, mode->hdisplay);
+	if (!pitch)
+		return MODE_BAD_WIDTH;
+	if (pitch > AST_PRIMARY_PLANE_MAX_OFFSET)
+		return MODE_BAD_WIDTH; /* maximum programmable pitch */
+	if (pitch > max_fb_size / mode->vdisplay)
 		return MODE_MEM;
 
 	return MODE_OK;
@@ -1018,10 +1046,7 @@ int ast_mode_config_init(struct ast_device *ast)
 		return ret;
 
 	drm_mode_config_reset(dev);
-
-	ret = drmm_kms_helper_poll_init(dev);
-	if (ret)
-		return ret;
+	drmm_kms_helper_poll_init(dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c
index 91e85e457bdf..37568cf3822c 100644
--- a/drivers/gpu/drm/ast/ast_post.c
+++ b/drivers/gpu/drm/ast/ast_post.c
@@ -1075,16 +1075,16 @@ static void get_ddr3_info(struct ast_device *ast, struct ast2300_dram_param *par
 
 	switch (param->vram_size) {
 	default:
-	case AST_VIDMEM_SIZE_8M:
+	case SZ_8M:
 		param->dram_config |= 0x00;
 		break;
-	case AST_VIDMEM_SIZE_16M:
+	case SZ_16M:
 		param->dram_config |= 0x04;
 		break;
-	case AST_VIDMEM_SIZE_32M:
+	case SZ_32M:
 		param->dram_config |= 0x08;
 		break;
-	case AST_VIDMEM_SIZE_64M:
+	case SZ_64M:
 		param->dram_config |= 0x0c;
 		break;
 	}
@@ -1446,16 +1446,16 @@ static void get_ddr2_info(struct ast_device *ast, struct ast2300_dram_param *par
 
 	switch (param->vram_size) {
 	default:
-	case AST_VIDMEM_SIZE_8M:
+	case SZ_8M:
 		param->dram_config |= 0x00;
 		break;
-	case AST_VIDMEM_SIZE_16M:
+	case SZ_16M:
 		param->dram_config |= 0x04;
 		break;
-	case AST_VIDMEM_SIZE_32M:
+	case SZ_32M:
 		param->dram_config |= 0x08;
 		break;
-	case AST_VIDMEM_SIZE_64M:
+	case SZ_64M:
 		param->dram_config |= 0x0c;
 		break;
 	}
@@ -1635,19 +1635,19 @@ static void ast_post_chip_2300(struct ast_device *ast)
                 switch (temp & 0x0c) {
                 default:
 		case 0x00:
-			param.vram_size = AST_VIDMEM_SIZE_8M;
+			param.vram_size = SZ_8M;
 			break;
 
 		case 0x04:
-			param.vram_size = AST_VIDMEM_SIZE_16M;
+			param.vram_size = SZ_16M;
 			break;
 
 		case 0x08:
-			param.vram_size = AST_VIDMEM_SIZE_32M;
+			param.vram_size = SZ_32M;
 			break;
 
 		case 0x0c:
-			param.vram_size = AST_VIDMEM_SIZE_64M;
+			param.vram_size = SZ_64M;
 			break;
 		}
 
diff --git a/drivers/gpu/drm/ast/ast_reg.h b/drivers/gpu/drm/ast/ast_reg.h
index bb2cc1d8b84e..e15adaf3a80e 100644
--- a/drivers/gpu/drm/ast/ast_reg.h
+++ b/drivers/gpu/drm/ast/ast_reg.h
@@ -30,9 +30,11 @@
 
 #define AST_IO_VGACRI			(0x54)
 #define AST_IO_VGACR80_PASSWORD		(0xa8)
+#define AST_IO_VGACR99_VGAMEM_RSRV_MASK	GENMASK(1, 0)
 #define AST_IO_VGACRA1_VGAIO_DISABLED	BIT(1)
 #define AST_IO_VGACRA1_MMIO_ENABLED	BIT(2)
 #define AST_IO_VGACRA3_DVO_ENABLED	BIT(7)
+#define AST_IO_VGACRAA_VGAMEM_SIZE_MASK	GENMASK(1, 0)
 #define AST_IO_VGACRB6_HSYNC_OFF	BIT(0)
 #define AST_IO_VGACRB6_VSYNC_OFF	BIT(1)
 #define AST_IO_VGACRCB_HWC_16BPP	BIT(0) /* set: ARGB4444, cleared: 2bpp palette */
diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 09a1be234f71..b9e0ca85226a 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -16,6 +16,7 @@ config DRM_AUX_BRIDGE
 	tristate
 	depends on DRM_BRIDGE && OF
 	select AUXILIARY_BUS
+	select DRM_KMS_HELPER
 	select DRM_PANEL_BRIDGE
 	help
 	  Simple transparent bridge that is used by several non-DRM drivers to
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index 050dae338ffe..1257009e850c 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -948,13 +948,14 @@ static enum drm_mode_status adv7511_bridge_mode_valid(struct drm_bridge *bridge,
 }
 
 static int adv7511_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct adv7511 *adv = bridge_to_adv7511(bridge);
 	int ret = 0;
 
 	if (adv->next_bridge) {
-		ret = drm_bridge_attach(bridge->encoder, adv->next_bridge, bridge,
+		ret = drm_bridge_attach(encoder, adv->next_bridge, bridge,
 					flags | DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
index 83d711ee3a2e..f3fe47b12edc 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
@@ -143,35 +143,7 @@ static int anx6345_dp_link_training(struct anx6345 *anx6345)
 	if (err)
 		return err;
 
-	/*
-	 * Power up the sink (DP_SET_POWER register is only available on DPCD
-	 * v1.1 and later).
-	 */
-	if (anx6345->dpcd[DP_DPCD_REV] >= 0x11) {
-		err = drm_dp_dpcd_readb(&anx6345->aux, DP_SET_POWER, &dpcd[0]);
-		if (err < 0) {
-			DRM_ERROR("Failed to read DP_SET_POWER register: %d\n",
-				  err);
-			return err;
-		}
-
-		dpcd[0] &= ~DP_SET_POWER_MASK;
-		dpcd[0] |= DP_SET_POWER_D0;
-
-		err = drm_dp_dpcd_writeb(&anx6345->aux, DP_SET_POWER, dpcd[0]);
-		if (err < 0) {
-			DRM_ERROR("Failed to power up DisplayPort link: %d\n",
-				  err);
-			return err;
-		}
-
-		/*
-		 * According to the DP 1.1 specification, a "Sink Device must
-		 * exit the power saving state within 1 ms" (Section 2.5.3.1,
-		 * Table 5-52, "Sink Control Field" (register 0x600).
-		 */
-		usleep_range(1000, 2000);
-	}
+	drm_dp_link_power_up(&anx6345->aux, anx6345->dpcd[DP_DPCD_REV]);
 
 	/* Possibly enable downspread on the sink */
 	err = regmap_write(anx6345->map[I2C_IDX_DPTX],
@@ -517,6 +489,7 @@ static const struct drm_connector_funcs anx6345_connector_funcs = {
 };
 
 static int anx6345_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct anx6345 *anx6345 = bridge_to_anx6345(bridge);
@@ -553,7 +526,7 @@ static int anx6345_bridge_attach(struct drm_bridge *bridge,
 	anx6345->connector.polled = DRM_CONNECTOR_POLL_HPD;
 
 	err = drm_connector_attach_encoder(&anx6345->connector,
-					   bridge->encoder);
+					   encoder);
 	if (err) {
 		DRM_ERROR("Failed to link up connector to encoder: %d\n", err);
 		goto connector_cleanup;
@@ -691,9 +664,10 @@ static int anx6345_i2c_probe(struct i2c_client *client)
 	struct device *dev;
 	int i, err;
 
-	anx6345 = devm_kzalloc(&client->dev, sizeof(*anx6345), GFP_KERNEL);
-	if (!anx6345)
-		return -ENOMEM;
+	anx6345 = devm_drm_bridge_alloc(&client->dev, struct anx6345, bridge,
+					&anx6345_bridge_funcs);
+	if (IS_ERR(anx6345))
+		return PTR_ERR(anx6345);
 
 	mutex_init(&anx6345->lock);
 
@@ -765,7 +739,6 @@ static int anx6345_i2c_probe(struct i2c_client *client)
 	/* Look for supported chip ID */
 	anx6345_poweron(anx6345);
 	if (anx6345_get_chip_id(anx6345)) {
-		anx6345->bridge.funcs = &anx6345_bridge_funcs;
 		drm_bridge_add(&anx6345->bridge);
 
 		return 0;
diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c
index f74694bb9c50..a83020d6576f 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c
@@ -656,35 +656,7 @@ static int anx78xx_dp_link_training(struct anx78xx *anx78xx)
 	if (err)
 		return err;
 
-	/*
-	 * Power up the sink (DP_SET_POWER register is only available on DPCD
-	 * v1.1 and later).
-	 */
-	if (anx78xx->dpcd[DP_DPCD_REV] >= 0x11) {
-		err = drm_dp_dpcd_readb(&anx78xx->aux, DP_SET_POWER, &dpcd[0]);
-		if (err < 0) {
-			DRM_ERROR("Failed to read DP_SET_POWER register: %d\n",
-				  err);
-			return err;
-		}
-
-		dpcd[0] &= ~DP_SET_POWER_MASK;
-		dpcd[0] |= DP_SET_POWER_D0;
-
-		err = drm_dp_dpcd_writeb(&anx78xx->aux, DP_SET_POWER, dpcd[0]);
-		if (err < 0) {
-			DRM_ERROR("Failed to power up DisplayPort link: %d\n",
-				  err);
-			return err;
-		}
-
-		/*
-		 * According to the DP 1.1 specification, a "Sink Device must
-		 * exit the power saving state within 1 ms" (Section 2.5.3.1,
-		 * Table 5-52, "Sink Control Field" (register 0x600).
-		 */
-		usleep_range(1000, 2000);
-	}
+	drm_dp_link_power_up(&anx78xx->aux, anx78xx->dpcd[DP_DPCD_REV]);
 
 	/* Possibly enable downspread on the sink */
 	err = regmap_write(anx78xx->map[I2C_IDX_TX_P0],
@@ -888,6 +860,7 @@ static const struct drm_connector_funcs anx78xx_connector_funcs = {
 };
 
 static int anx78xx_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct anx78xx *anx78xx = bridge_to_anx78xx(bridge);
@@ -924,7 +897,7 @@ static int anx78xx_bridge_attach(struct drm_bridge *bridge,
 	anx78xx->connector.polled = DRM_CONNECTOR_POLL_HPD;
 
 	err = drm_connector_attach_encoder(&anx78xx->connector,
-					   bridge->encoder);
+					   encoder);
 	if (err) {
 		DRM_ERROR("Failed to link up connector to encoder: %d\n", err);
 		goto connector_cleanup;
diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
index 071168aa0c3b..a761941bc3c2 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c
@@ -838,10 +838,7 @@ static int analogix_dp_commit(struct analogix_dp_device *dp)
 	int ret;
 
 	/* Keep the panel disabled while we configure video */
-	if (dp->plat_data->panel) {
-		if (drm_panel_disable(dp->plat_data->panel))
-			DRM_ERROR("failed to disable the panel\n");
-	}
+	drm_panel_disable(dp->plat_data->panel);
 
 	ret = analogix_dp_train_link(dp);
 	if (ret) {
@@ -863,13 +860,7 @@ static int analogix_dp_commit(struct analogix_dp_device *dp)
 	}
 
 	/* Safe to enable the panel now */
-	if (dp->plat_data->panel) {
-		ret = drm_panel_enable(dp->plat_data->panel);
-		if (ret) {
-			DRM_ERROR("failed to enable the panel\n");
-			return ret;
-		}
-	}
+	drm_panel_enable(dp->plat_data->panel);
 
 	/* Check whether panel supports fast training */
 	ret = analogix_dp_fast_link_train_detection(dp);
@@ -955,67 +946,15 @@ static int analogix_dp_disable_psr(struct analogix_dp_device *dp)
 	return analogix_dp_send_psr_spd(dp, &psr_vsc, true);
 }
 
-/*
- * This function is a bit of a catch-all for panel preparation, hopefully
- * simplifying the logic of functions that need to prepare/unprepare the panel
- * below.
- *
- * If @prepare is true, this function will prepare the panel. Conversely, if it
- * is false, the panel will be unprepared.
- *
- * If @is_modeset_prepare is true, the function will disregard the current state
- * of the panel and either prepare/unprepare the panel based on @prepare. Once
- * it finishes, it will update dp->panel_is_modeset to reflect the current state
- * of the panel.
- */
-static int analogix_dp_prepare_panel(struct analogix_dp_device *dp,
-				     bool prepare, bool is_modeset_prepare)
-{
-	int ret = 0;
-
-	if (!dp->plat_data->panel)
-		return 0;
-
-	mutex_lock(&dp->panel_lock);
-
-	/*
-	 * Exit early if this is a temporary prepare/unprepare and we're already
-	 * modeset (since we neither want to prepare twice or unprepare early).
-	 */
-	if (dp->panel_is_modeset && !is_modeset_prepare)
-		goto out;
-
-	if (prepare)
-		ret = drm_panel_prepare(dp->plat_data->panel);
-	else
-		ret = drm_panel_unprepare(dp->plat_data->panel);
-
-	if (ret)
-		goto out;
-
-	if (is_modeset_prepare)
-		dp->panel_is_modeset = prepare;
-
-out:
-	mutex_unlock(&dp->panel_lock);
-	return ret;
-}
-
 static int analogix_dp_get_modes(struct drm_connector *connector)
 {
 	struct analogix_dp_device *dp = to_dp(connector);
 	const struct drm_edid *drm_edid;
-	int ret, num_modes = 0;
+	int num_modes = 0;
 
 	if (dp->plat_data->panel) {
 		num_modes += drm_panel_get_modes(dp->plat_data->panel, connector);
 	} else {
-		ret = analogix_dp_prepare_panel(dp, true, false);
-		if (ret) {
-			DRM_ERROR("Failed to prepare panel (%d)\n", ret);
-			return 0;
-		}
-
 		drm_edid = drm_edid_read_ddc(connector, &dp->aux.ddc);
 
 		drm_edid_connector_update(&dp->connector, drm_edid);
@@ -1024,10 +963,6 @@ static int analogix_dp_get_modes(struct drm_connector *connector)
 			num_modes += drm_edid_connector_add_modes(&dp->connector);
 			drm_edid_free(drm_edid);
 		}
-
-		ret = analogix_dp_prepare_panel(dp, false, false);
-		if (ret)
-			DRM_ERROR("Failed to unprepare panel (%d)\n", ret);
 	}
 
 	if (dp->plat_data->get_modes)
@@ -1082,24 +1017,13 @@ analogix_dp_detect(struct drm_connector *connector, bool force)
 {
 	struct analogix_dp_device *dp = to_dp(connector);
 	enum drm_connector_status status = connector_status_disconnected;
-	int ret;
 
 	if (dp->plat_data->panel)
 		return connector_status_connected;
 
-	ret = analogix_dp_prepare_panel(dp, true, false);
-	if (ret) {
-		DRM_ERROR("Failed to prepare panel (%d)\n", ret);
-		return connector_status_disconnected;
-	}
-
 	if (!analogix_dp_detect_hpd(dp))
 		status = connector_status_connected;
 
-	ret = analogix_dp_prepare_panel(dp, false, false);
-	if (ret)
-		DRM_ERROR("Failed to unprepare panel (%d)\n", ret);
-
 	return status;
 }
 
@@ -1113,10 +1037,10 @@ static const struct drm_connector_funcs analogix_dp_connector_funcs = {
 };
 
 static int analogix_dp_bridge_attach(struct drm_bridge *bridge,
+				     struct drm_encoder *encoder,
 				     enum drm_bridge_attach_flags flags)
 {
 	struct analogix_dp_device *dp = bridge->driver_private;
-	struct drm_encoder *encoder = dp->encoder;
 	struct drm_connector *connector = NULL;
 	int ret = 0;
 
@@ -1203,7 +1127,6 @@ static void analogix_dp_bridge_atomic_pre_enable(struct drm_bridge *bridge,
 	struct analogix_dp_device *dp = bridge->driver_private;
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *old_crtc_state;
-	int ret;
 
 	crtc = analogix_dp_get_new_crtc(dp, old_state);
 	if (!crtc)
@@ -1214,9 +1137,7 @@ static void analogix_dp_bridge_atomic_pre_enable(struct drm_bridge *bridge,
 	if (old_crtc_state && old_crtc_state->self_refresh_active)
 		return;
 
-	ret = analogix_dp_prepare_panel(dp, true, true);
-	if (ret)
-		DRM_ERROR("failed to setup the panel ret = %d\n", ret);
+	drm_panel_prepare(dp->plat_data->panel);
 }
 
 static int analogix_dp_set_bridge(struct analogix_dp_device *dp)
@@ -1296,17 +1217,11 @@ static void analogix_dp_bridge_atomic_enable(struct drm_bridge *bridge,
 static void analogix_dp_bridge_disable(struct drm_bridge *bridge)
 {
 	struct analogix_dp_device *dp = bridge->driver_private;
-	int ret;
 
 	if (dp->dpms_mode != DRM_MODE_DPMS_ON)
 		return;
 
-	if (dp->plat_data->panel) {
-		if (drm_panel_disable(dp->plat_data->panel)) {
-			DRM_ERROR("failed to disable the panel\n");
-			return;
-		}
-	}
+	drm_panel_disable(dp->plat_data->panel);
 
 	disable_irq(dp->irq);
 
@@ -1314,9 +1229,7 @@ static void analogix_dp_bridge_disable(struct drm_bridge *bridge)
 
 	pm_runtime_put_sync(dp->dev);
 
-	ret = analogix_dp_prepare_panel(dp, false, true);
-	if (ret)
-		DRM_ERROR("failed to setup the panel ret = %d\n", ret);
+	drm_panel_unprepare(dp->plat_data->panel);
 
 	dp->fast_train_enable = false;
 	dp->psr_supported = false;
@@ -1505,6 +1418,10 @@ static int analogix_dp_dt_parse_pdata(struct analogix_dp_device *dp)
 		video_info->max_link_rate = 0x0A;
 		video_info->max_lane_count = 0x04;
 		break;
+	case RK3588_EDP:
+		video_info->max_link_rate = 0x14;
+		video_info->max_lane_count = 0x04;
+		break;
 	case EXYNOS_DP:
 		/*
 		 * NOTE: those property parseing code is used for
@@ -1540,6 +1457,26 @@ out:
 	return ret;
 }
 
+static int analogix_dpaux_wait_hpd_asserted(struct drm_dp_aux *aux, unsigned long wait_us)
+{
+	struct analogix_dp_device *dp = to_dp(aux);
+	int val;
+	int ret;
+
+	if (dp->force_hpd)
+		return 0;
+
+	pm_runtime_get_sync(dp->dev);
+
+	ret = readx_poll_timeout(analogix_dp_get_plug_in_status, dp, val, !val,
+				 wait_us / 100, wait_us);
+
+	pm_runtime_mark_last_busy(dp->dev);
+	pm_runtime_put_autosuspend(dp->dev);
+
+	return ret;
+}
+
 struct analogix_dp_device *
 analogix_dp_probe(struct device *dev, struct analogix_dp_plat_data *plat_data)
 {
@@ -1560,9 +1497,6 @@ analogix_dp_probe(struct device *dev, struct analogix_dp_plat_data *plat_data)
 	dp->dev = &pdev->dev;
 	dp->dpms_mode = DRM_MODE_DPMS_OFF;
 
-	mutex_init(&dp->panel_lock);
-	dp->panel_is_modeset = false;
-
 	/*
 	 * platform dp driver need containor_of the plat_data to get
 	 * the driver private data, so we need to store the point of
@@ -1625,10 +1559,10 @@ analogix_dp_probe(struct device *dev, struct analogix_dp_plat_data *plat_data)
 		 * that we can get the current state of the GPIO.
 		 */
 		dp->irq = gpiod_to_irq(dp->hpd_gpiod);
-		irq_flags = IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING;
+		irq_flags = IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_NO_AUTOEN;
 	} else {
 		dp->irq = platform_get_irq(pdev, 0);
-		irq_flags = 0;
+		irq_flags = IRQF_NO_AUTOEN;
 	}
 
 	if (dp->irq == -ENXIO) {
@@ -1645,7 +1579,18 @@ analogix_dp_probe(struct device *dev, struct analogix_dp_plat_data *plat_data)
 		dev_err(&pdev->dev, "failed to request irq\n");
 		goto err_disable_clk;
 	}
-	disable_irq(dp->irq);
+
+	dp->aux.name = "DP-AUX";
+	dp->aux.transfer = analogix_dpaux_transfer;
+	dp->aux.wait_hpd_asserted = analogix_dpaux_wait_hpd_asserted;
+	dp->aux.dev = dp->dev;
+	drm_dp_aux_init(&dp->aux);
+
+	pm_runtime_use_autosuspend(dp->dev);
+	pm_runtime_set_autosuspend_delay(dp->dev, 100);
+	ret = devm_pm_runtime_enable(dp->dev);
+	if (ret)
+		goto err_disable_clk;
 
 	return dp;
 
@@ -1681,6 +1626,7 @@ int analogix_dp_resume(struct analogix_dp_device *dp)
 	if (dp->plat_data->power_on)
 		dp->plat_data->power_on(dp->plat_data);
 
+	phy_set_mode(dp->phy, PHY_MODE_DP);
 	phy_power_on(dp->phy);
 
 	analogix_dp_init_dp(dp);
@@ -1696,25 +1642,12 @@ int analogix_dp_bind(struct analogix_dp_device *dp, struct drm_device *drm_dev)
 	dp->drm_dev = drm_dev;
 	dp->encoder = dp->plat_data->encoder;
 
-	if (IS_ENABLED(CONFIG_PM)) {
-		pm_runtime_use_autosuspend(dp->dev);
-		pm_runtime_set_autosuspend_delay(dp->dev, 100);
-		pm_runtime_enable(dp->dev);
-	} else {
-		ret = analogix_dp_resume(dp);
-		if (ret)
-			return ret;
-	}
-
-	dp->aux.name = "DP-AUX";
-	dp->aux.transfer = analogix_dpaux_transfer;
-	dp->aux.dev = dp->dev;
 	dp->aux.drm_dev = drm_dev;
 
 	ret = drm_dp_aux_register(&dp->aux);
 	if (ret) {
 		DRM_ERROR("failed to register AUX (%d)\n", ret);
-		goto err_disable_pm_runtime;
+		return ret;
 	}
 
 	ret = analogix_dp_create_bridge(drm_dev, dp);
@@ -1727,13 +1660,6 @@ int analogix_dp_bind(struct analogix_dp_device *dp, struct drm_device *drm_dev)
 
 err_unregister_aux:
 	drm_dp_aux_unregister(&dp->aux);
-err_disable_pm_runtime:
-	if (IS_ENABLED(CONFIG_PM)) {
-		pm_runtime_dont_use_autosuspend(dp->dev);
-		pm_runtime_disable(dp->dev);
-	} else {
-		analogix_dp_suspend(dp);
-	}
 
 	return ret;
 }
@@ -1744,19 +1670,9 @@ void analogix_dp_unbind(struct analogix_dp_device *dp)
 	analogix_dp_bridge_disable(dp->bridge);
 	dp->connector.funcs->destroy(&dp->connector);
 
-	if (dp->plat_data->panel) {
-		if (drm_panel_unprepare(dp->plat_data->panel))
-			DRM_ERROR("failed to turnoff the panel\n");
-	}
+	drm_panel_unprepare(dp->plat_data->panel);
 
 	drm_dp_aux_unregister(&dp->aux);
-
-	if (IS_ENABLED(CONFIG_PM)) {
-		pm_runtime_dont_use_autosuspend(dp->dev);
-		pm_runtime_disable(dp->dev);
-	} else {
-		analogix_dp_suspend(dp);
-	}
 }
 EXPORT_SYMBOL_GPL(analogix_dp_unbind);
 
@@ -1782,6 +1698,20 @@ int analogix_dp_stop_crc(struct drm_connector *connector)
 }
 EXPORT_SYMBOL_GPL(analogix_dp_stop_crc);
 
+struct analogix_dp_plat_data *analogix_dp_aux_to_plat_data(struct drm_dp_aux *aux)
+{
+	struct analogix_dp_device *dp = to_dp(aux);
+
+	return dp->plat_data;
+}
+EXPORT_SYMBOL_GPL(analogix_dp_aux_to_plat_data);
+
+struct drm_dp_aux *analogix_dp_get_aux(struct analogix_dp_device *dp)
+{
+	return &dp->aux;
+}
+EXPORT_SYMBOL_GPL(analogix_dp_get_aux);
+
 MODULE_AUTHOR("Jingoo Han <jg1.han@samsung.com>");
 MODULE_DESCRIPTION("Analogix DP Core Driver");
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h
index 774d11574b09..2b54120ba4a3 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h
@@ -169,9 +169,6 @@ struct analogix_dp_device {
 	bool			fast_train_enable;
 	bool			psr_supported;
 
-	struct mutex		panel_lock;
-	bool			panel_is_modeset;
-
 	struct analogix_dp_plat_data *plat_data;
 };
 
diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
index 3afc73c858c4..38fd8d5014d2 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
@@ -11,6 +11,7 @@
 #include <linux/gpio/consumer.h>
 #include <linux/io.h>
 #include <linux/iopoll.h>
+#include <linux/phy/phy.h>
 
 #include <drm/bridge/analogix_dp.h>
 
@@ -513,10 +514,24 @@ void analogix_dp_enable_sw_function(struct analogix_dp_device *dp)
 void analogix_dp_set_link_bandwidth(struct analogix_dp_device *dp, u32 bwtype)
 {
 	u32 reg;
+	int ret;
 
 	reg = bwtype;
 	if ((bwtype == DP_LINK_BW_2_7) || (bwtype == DP_LINK_BW_1_62))
 		writel(reg, dp->reg_base + ANALOGIX_DP_LINK_BW_SET);
+
+	if (dp->phy) {
+		union phy_configure_opts phy_cfg = {0};
+
+		phy_cfg.dp.link_rate =
+			drm_dp_bw_code_to_link_rate(dp->link_train.link_rate) / 100;
+		phy_cfg.dp.set_rate = true;
+		ret = phy_configure(dp->phy, &phy_cfg);
+		if (ret && ret != -EOPNOTSUPP) {
+			dev_err(dp->dev, "%s: phy_configure() failed: %d\n", __func__, ret);
+			return;
+		}
+	}
 }
 
 void analogix_dp_get_link_bandwidth(struct analogix_dp_device *dp, u32 *bwtype)
@@ -530,9 +545,22 @@ void analogix_dp_get_link_bandwidth(struct analogix_dp_device *dp, u32 *bwtype)
 void analogix_dp_set_lane_count(struct analogix_dp_device *dp, u32 count)
 {
 	u32 reg;
+	int ret;
 
 	reg = count;
 	writel(reg, dp->reg_base + ANALOGIX_DP_LANE_COUNT_SET);
+
+	if (dp->phy) {
+		union phy_configure_opts phy_cfg = {0};
+
+		phy_cfg.dp.lanes = dp->link_train.lane_count;
+		phy_cfg.dp.set_lanes = true;
+		ret = phy_configure(dp->phy, &phy_cfg);
+		if (ret && ret != -EOPNOTSUPP) {
+			dev_err(dp->dev, "%s: phy_configure() failed: %d\n", __func__, ret);
+			return;
+		}
+	}
 }
 
 void analogix_dp_get_lane_count(struct analogix_dp_device *dp, u32 *count)
@@ -546,10 +574,34 @@ void analogix_dp_get_lane_count(struct analogix_dp_device *dp, u32 *count)
 void analogix_dp_set_lane_link_training(struct analogix_dp_device *dp)
 {
 	u8 lane;
+	int ret;
 
 	for (lane = 0; lane < dp->link_train.lane_count; lane++)
 		writel(dp->link_train.training_lane[lane],
 		       dp->reg_base + ANALOGIX_DP_LN0_LINK_TRAINING_CTL + 4 * lane);
+
+	if (dp->phy) {
+		union phy_configure_opts phy_cfg = {0};
+
+		for (lane = 0; lane < dp->link_train.lane_count; lane++) {
+			u8 training_lane = dp->link_train.training_lane[lane];
+			u8 vs, pe;
+
+			vs = (training_lane & DP_TRAIN_VOLTAGE_SWING_MASK) >>
+			     DP_TRAIN_VOLTAGE_SWING_SHIFT;
+			pe = (training_lane & DP_TRAIN_PRE_EMPHASIS_MASK) >>
+			     DP_TRAIN_PRE_EMPHASIS_SHIFT;
+			phy_cfg.dp.voltage[lane] = vs;
+			phy_cfg.dp.pre[lane] = pe;
+		}
+
+		phy_cfg.dp.set_voltages = true;
+		ret = phy_configure(dp->phy, &phy_cfg);
+		if (ret && ret != -EOPNOTSUPP) {
+			dev_err(dp->dev, "%s: phy_configure() failed: %d\n", __func__, ret);
+			return;
+		}
+	}
 }
 
 u32 analogix_dp_get_lane_link_training(struct analogix_dp_device *dp, u8 lane)
diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c
index 0b97b66de577..8a9079c2ed5c 100644
--- a/drivers/gpu/drm/bridge/analogix/anx7625.c
+++ b/drivers/gpu/drm/bridge/analogix/anx7625.c
@@ -1257,10 +1257,10 @@ static void anx7625_power_on(struct anx7625_data *ctx)
 	usleep_range(11000, 12000);
 
 	/* Power on pin enable */
-	gpiod_set_value(ctx->pdata.gpio_p_on, 1);
+	gpiod_set_value_cansleep(ctx->pdata.gpio_p_on, 1);
 	usleep_range(10000, 11000);
 	/* Power reset pin enable */
-	gpiod_set_value(ctx->pdata.gpio_reset, 1);
+	gpiod_set_value_cansleep(ctx->pdata.gpio_reset, 1);
 	usleep_range(10000, 11000);
 
 	DRM_DEV_DEBUG_DRIVER(dev, "power on !\n");
@@ -1280,9 +1280,9 @@ static void anx7625_power_standby(struct anx7625_data *ctx)
 		return;
 	}
 
-	gpiod_set_value(ctx->pdata.gpio_reset, 0);
+	gpiod_set_value_cansleep(ctx->pdata.gpio_reset, 0);
 	usleep_range(1000, 1100);
-	gpiod_set_value(ctx->pdata.gpio_p_on, 0);
+	gpiod_set_value_cansleep(ctx->pdata.gpio_p_on, 0);
 	usleep_range(1000, 1100);
 
 	ret = regulator_bulk_disable(ARRAY_SIZE(ctx->pdata.supplies),
@@ -1814,9 +1814,6 @@ static enum drm_connector_status anx7625_sink_detect(struct anx7625_data *ctx)
 
 	DRM_DEV_DEBUG_DRIVER(dev, "sink detect\n");
 
-	if (ctx->pdata.panel_bridge)
-		return connector_status_connected;
-
 	return ctx->hpd_status ? connector_status_connected :
 				     connector_status_disconnected;
 }
@@ -2141,6 +2138,7 @@ static void hdcp_check_work_func(struct work_struct *work)
 }
 
 static int anx7625_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct anx7625_data *ctx = bridge_to_anx7625(bridge);
@@ -2159,7 +2157,7 @@ static int anx7625_bridge_attach(struct drm_bridge *bridge,
 	}
 
 	if (ctx->pdata.panel_bridge) {
-		err = drm_bridge_attach(bridge->encoder,
+		err = drm_bridge_attach(encoder,
 					ctx->pdata.panel_bridge,
 					&ctx->bridge, flags);
 		if (err)
@@ -2474,6 +2472,22 @@ static const struct drm_edid *anx7625_bridge_edid_read(struct drm_bridge *bridge
 	return anx7625_edid_read(ctx);
 }
 
+static void anx7625_bridge_hpd_enable(struct drm_bridge *bridge)
+{
+	struct anx7625_data *ctx = bridge_to_anx7625(bridge);
+	struct device *dev = ctx->dev;
+
+	pm_runtime_get_sync(dev);
+}
+
+static void anx7625_bridge_hpd_disable(struct drm_bridge *bridge)
+{
+	struct anx7625_data *ctx = bridge_to_anx7625(bridge);
+	struct device *dev = ctx->dev;
+
+	pm_runtime_put_sync(dev);
+}
+
 static const struct drm_bridge_funcs anx7625_bridge_funcs = {
 	.attach = anx7625_bridge_attach,
 	.detach = anx7625_bridge_detach,
@@ -2487,6 +2501,8 @@ static const struct drm_bridge_funcs anx7625_bridge_funcs = {
 	.atomic_reset = drm_atomic_helper_bridge_reset,
 	.detect = anx7625_bridge_detect,
 	.edid_read = anx7625_bridge_edid_read,
+	.hpd_enable = anx7625_bridge_hpd_enable,
+	.hpd_disable = anx7625_bridge_hpd_disable,
 };
 
 static int anx7625_register_i2c_dummy_clients(struct anx7625_data *ctx,
@@ -2568,12 +2584,6 @@ static const struct dev_pm_ops anx7625_pm_ops = {
 			   anx7625_runtime_pm_resume, NULL)
 };
 
-static void anx7625_runtime_disable(void *data)
-{
-	pm_runtime_dont_use_autosuspend(data);
-	pm_runtime_disable(data);
-}
-
 static int anx7625_link_bridge(struct drm_dp_aux *aux)
 {
 	struct anx7625_data *platform = container_of(aux, struct anx7625_data, aux);
@@ -2590,9 +2600,8 @@ static int anx7625_link_bridge(struct drm_dp_aux *aux)
 	platform->bridge.of_node = dev->of_node;
 	if (!anx7625_of_panel_on_aux_bus(dev))
 		platform->bridge.ops |= DRM_BRIDGE_OP_EDID;
-	if (!platform->pdata.panel_bridge)
-		platform->bridge.ops |= DRM_BRIDGE_OP_HPD |
-					DRM_BRIDGE_OP_DETECT;
+	if (!platform->pdata.panel_bridge || !anx7625_of_panel_on_aux_bus(dev))
+		platform->bridge.ops |= DRM_BRIDGE_OP_HPD | DRM_BRIDGE_OP_DETECT;
 	platform->bridge.type = platform->pdata.panel_bridge ?
 				    DRM_MODE_CONNECTOR_eDP :
 				    DRM_MODE_CONNECTOR_DisplayPort;
@@ -2707,11 +2716,10 @@ static int anx7625_i2c_probe(struct i2c_client *client)
 		goto free_wq;
 	}
 
-	pm_runtime_enable(dev);
 	pm_runtime_set_autosuspend_delay(dev, 1000);
 	pm_runtime_use_autosuspend(dev);
 	pm_suspend_ignore_children(dev, true);
-	ret = devm_add_action_or_reset(dev, anx7625_runtime_disable, dev);
+	ret = devm_pm_runtime_enable(dev);
 	if (ret)
 		goto free_wq;
 
@@ -2771,7 +2779,6 @@ static void anx7625_i2c_remove(struct i2c_client *client)
 
 	if (platform->hdcp_workqueue) {
 		cancel_delayed_work(&platform->hdcp_work);
-		flush_workqueue(platform->hdcp_workqueue);
 		destroy_workqueue(platform->hdcp_workqueue);
 	}
 
diff --git a/drivers/gpu/drm/bridge/aux-bridge.c b/drivers/gpu/drm/bridge/aux-bridge.c
index 015983c015e5..c179b86d208f 100644
--- a/drivers/gpu/drm/bridge/aux-bridge.c
+++ b/drivers/gpu/drm/bridge/aux-bridge.c
@@ -86,6 +86,7 @@ struct drm_aux_bridge_data {
 };
 
 static int drm_aux_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct drm_aux_bridge_data *data;
@@ -95,7 +96,7 @@ static int drm_aux_bridge_attach(struct drm_bridge *bridge,
 
 	data = container_of(bridge, struct drm_aux_bridge_data, bridge);
 
-	return drm_bridge_attach(bridge->encoder, data->next_bridge, bridge,
+	return drm_bridge_attach(encoder, data->next_bridge, bridge,
 				 DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 }
 
diff --git a/drivers/gpu/drm/bridge/aux-hpd-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-bridge.c
index 48f297c78ee6..b3f588b71a7d 100644
--- a/drivers/gpu/drm/bridge/aux-hpd-bridge.c
+++ b/drivers/gpu/drm/bridge/aux-hpd-bridge.c
@@ -156,6 +156,7 @@ void drm_aux_hpd_bridge_notify(struct device *dev, enum drm_connector_status sta
 EXPORT_SYMBOL_GPL(drm_aux_hpd_bridge_notify);
 
 static int drm_aux_hpd_bridge_attach(struct drm_bridge *bridge,
+				     struct drm_encoder *encoder,
 				     enum drm_bridge_attach_flags flags)
 {
 	return flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR ? 0 : -EINVAL;
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index c7a0247e06ad..b022dd6e6b6e 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -425,6 +425,17 @@
 #define DSI_NULL_FRAME_OVERHEAD		6
 #define DSI_EOT_PKT_SIZE		4
 
+struct cdns_dsi_bridge_state {
+	struct drm_bridge_state base;
+	struct cdns_dsi_cfg dsi_cfg;
+};
+
+static inline struct cdns_dsi_bridge_state *
+to_cdns_dsi_bridge_state(struct drm_bridge_state *bridge_state)
+{
+	return container_of(bridge_state, struct cdns_dsi_bridge_state, base);
+}
+
 static inline struct cdns_dsi *input_to_dsi(struct cdns_dsi_input *input)
 {
 	return container_of(input, struct cdns_dsi, input);
@@ -568,15 +579,18 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 	struct phy_configure_opts_mipi_dphy *phy_cfg = &output->phy_opts.mipi_dphy;
 	unsigned long dsi_hss_hsa_hse_hbp;
 	unsigned int nlanes = output->dev->lanes;
+	int mode_clock = (mode_valid_check ? mode->clock : mode->crtc_clock);
 	int ret;
 
 	ret = cdns_dsi_mode2cfg(dsi, mode, dsi_cfg, mode_valid_check);
 	if (ret)
 		return ret;
 
-	phy_mipi_dphy_get_default_config(mode->crtc_clock * 1000,
-					 mipi_dsi_pixel_format_to_bpp(output->dev->format),
-					 nlanes, phy_cfg);
+	ret = phy_mipi_dphy_get_default_config(mode_clock * 1000,
+					       mipi_dsi_pixel_format_to_bpp(output->dev->format),
+					       nlanes, phy_cfg);
+	if (ret)
+		return ret;
 
 	ret = cdns_dsi_adjust_phy_config(dsi, dsi_cfg, phy_cfg, mode, mode_valid_check);
 	if (ret)
@@ -605,6 +619,7 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 }
 
 static int cdns_dsi_bridge_attach(struct drm_bridge *bridge,
+				  struct drm_encoder *encoder,
 				  enum drm_bridge_attach_flags flags)
 {
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
@@ -617,7 +632,7 @@ static int cdns_dsi_bridge_attach(struct drm_bridge *bridge,
 		return -ENOTSUPP;
 	}
 
-	return drm_bridge_attach(bridge->encoder, output->bridge, bridge,
+	return drm_bridge_attach(encoder, output->bridge, bridge,
 				 flags);
 }
 
@@ -655,7 +670,8 @@ cdns_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 	return MODE_OK;
 }
 
-static void cdns_dsi_bridge_disable(struct drm_bridge *bridge)
+static void cdns_dsi_bridge_atomic_disable(struct drm_bridge *bridge,
+					   struct drm_atomic_state *state)
 {
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
@@ -675,11 +691,17 @@ static void cdns_dsi_bridge_disable(struct drm_bridge *bridge)
 	pm_runtime_put(dsi->base.dev);
 }
 
-static void cdns_dsi_bridge_post_disable(struct drm_bridge *bridge)
+static void cdns_dsi_bridge_atomic_post_disable(struct drm_bridge *bridge,
+						struct drm_atomic_state *state)
 {
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
 
+	dsi->phy_initialized = false;
+	dsi->link_initialized = false;
+	phy_power_off(dsi->dphy);
+	phy_exit(dsi->dphy);
+
 	pm_runtime_put(dsi->base.dev);
 }
 
@@ -752,32 +774,59 @@ static void cdns_dsi_init_link(struct cdns_dsi *dsi)
 	dsi->link_initialized = true;
 }
 
-static void cdns_dsi_bridge_enable(struct drm_bridge *bridge)
+static void cdns_dsi_bridge_atomic_enable(struct drm_bridge *bridge,
+					  struct drm_atomic_state *state)
 {
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
 	struct cdns_dsi_output *output = &dsi->output;
+	struct drm_connector_state *conn_state;
+	struct drm_crtc_state *crtc_state;
+	struct cdns_dsi_bridge_state *dsi_state;
+	struct drm_bridge_state *new_bridge_state;
 	struct drm_display_mode *mode;
 	struct phy_configure_opts_mipi_dphy *phy_cfg = &output->phy_opts.mipi_dphy;
+	struct drm_connector *connector;
 	unsigned long tx_byte_period;
 	struct cdns_dsi_cfg dsi_cfg;
-	u32 tmp, reg_wakeup, div;
+	u32 tmp, reg_wakeup, div, status;
 	int nlanes;
 
 	if (WARN_ON(pm_runtime_get_sync(dsi->base.dev) < 0))
 		return;
 
+	new_bridge_state = drm_atomic_get_new_bridge_state(state, bridge);
+	if (WARN_ON(!new_bridge_state))
+		return;
+
+	dsi_state = to_cdns_dsi_bridge_state(new_bridge_state);
+	dsi_cfg = dsi_state->dsi_cfg;
+
 	if (dsi->platform_ops && dsi->platform_ops->enable)
 		dsi->platform_ops->enable(dsi);
 
-	mode = &bridge->encoder->crtc->state->adjusted_mode;
+	connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	conn_state = drm_atomic_get_new_connector_state(state, connector);
+	crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc);
+	mode = &crtc_state->adjusted_mode;
 	nlanes = output->dev->lanes;
 
-	WARN_ON_ONCE(cdns_dsi_check_conf(dsi, mode, &dsi_cfg, false));
-
 	cdns_dsi_hs_init(dsi);
 	cdns_dsi_init_link(dsi);
 
+	/*
+	 * Now that the DSI Link and DSI Phy are initialized,
+	 * wait for the CLK and Data Lanes to be ready.
+	 */
+	tmp = CLK_LANE_RDY;
+	for (int i = 0; i < nlanes; i++)
+		tmp |= DATA_LANE_RDY(i);
+
+	if (readl_poll_timeout(dsi->regs + MCTL_MAIN_STS, status,
+			       (tmp == (status & tmp)), 100, 500000))
+		dev_err(dsi->base.dev,
+			"Timed Out: DSI-DPhy Clock and Data Lanes not ready.\n");
+
 	writel(HBP_LEN(dsi_cfg.hbp) | HSA_LEN(dsi_cfg.hsa),
 	       dsi->regs + VID_HSIZE1);
 	writel(HFP_LEN(dsi_cfg.hfp) | HACT_LEN(dsi_cfg.hact),
@@ -892,7 +941,8 @@ static void cdns_dsi_bridge_enable(struct drm_bridge *bridge)
 	writel(tmp, dsi->regs + MCTL_MAIN_EN);
 }
 
-static void cdns_dsi_bridge_pre_enable(struct drm_bridge *bridge)
+static void cdns_dsi_bridge_atomic_pre_enable(struct drm_bridge *bridge,
+					      struct drm_atomic_state *state)
 {
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
@@ -904,13 +954,109 @@ static void cdns_dsi_bridge_pre_enable(struct drm_bridge *bridge)
 	cdns_dsi_hs_init(dsi);
 }
 
+static u32 *cdns_dsi_bridge_get_input_bus_fmts(struct drm_bridge *bridge,
+					       struct drm_bridge_state *bridge_state,
+					       struct drm_crtc_state *crtc_state,
+					       struct drm_connector_state *conn_state,
+					       u32 output_fmt,
+					       unsigned int *num_input_fmts)
+{
+	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
+	struct cdns_dsi *dsi = input_to_dsi(input);
+	struct cdns_dsi_output *output = &dsi->output;
+	u32 *input_fmts;
+
+	*num_input_fmts = 0;
+
+	input_fmts = kzalloc(sizeof(*input_fmts), GFP_KERNEL);
+	if (!input_fmts)
+		return NULL;
+
+	input_fmts[0] = drm_mipi_dsi_get_input_bus_fmt(output->dev->format);
+	if (!input_fmts[0])
+		return NULL;
+
+	*num_input_fmts = 1;
+
+	return input_fmts;
+}
+
+static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
+					struct drm_bridge_state *bridge_state,
+					struct drm_crtc_state *crtc_state,
+					struct drm_connector_state *conn_state)
+{
+	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
+	struct cdns_dsi *dsi = input_to_dsi(input);
+	struct cdns_dsi_bridge_state *dsi_state = to_cdns_dsi_bridge_state(bridge_state);
+	const struct drm_display_mode *mode = &crtc_state->mode;
+	struct cdns_dsi_cfg *dsi_cfg = &dsi_state->dsi_cfg;
+
+	return cdns_dsi_check_conf(dsi, mode, dsi_cfg, false);
+}
+
+static struct drm_bridge_state *
+cdns_dsi_bridge_atomic_duplicate_state(struct drm_bridge *bridge)
+{
+	struct cdns_dsi_bridge_state *dsi_state, *old_dsi_state;
+	struct drm_bridge_state *bridge_state;
+
+	if (WARN_ON(!bridge->base.state))
+		return NULL;
+
+	bridge_state = drm_priv_to_bridge_state(bridge->base.state);
+	old_dsi_state = to_cdns_dsi_bridge_state(bridge_state);
+
+	dsi_state = kzalloc(sizeof(*dsi_state), GFP_KERNEL);
+	if (!dsi_state)
+		return NULL;
+
+	__drm_atomic_helper_bridge_duplicate_state(bridge, &dsi_state->base);
+
+	memcpy(&dsi_state->dsi_cfg, &old_dsi_state->dsi_cfg,
+	       sizeof(dsi_state->dsi_cfg));
+
+	return &dsi_state->base;
+}
+
+static void
+cdns_dsi_bridge_atomic_destroy_state(struct drm_bridge *bridge,
+				     struct drm_bridge_state *state)
+{
+	struct cdns_dsi_bridge_state *dsi_state;
+
+	dsi_state = to_cdns_dsi_bridge_state(state);
+
+	kfree(dsi_state);
+}
+
+static struct drm_bridge_state *
+cdns_dsi_bridge_atomic_reset(struct drm_bridge *bridge)
+{
+	struct cdns_dsi_bridge_state *dsi_state;
+
+	dsi_state = kzalloc(sizeof(*dsi_state), GFP_KERNEL);
+	if (!dsi_state)
+		return NULL;
+
+	memset(dsi_state, 0, sizeof(*dsi_state));
+	dsi_state->base.bridge = bridge;
+
+	return &dsi_state->base;
+}
+
 static const struct drm_bridge_funcs cdns_dsi_bridge_funcs = {
 	.attach = cdns_dsi_bridge_attach,
 	.mode_valid = cdns_dsi_bridge_mode_valid,
-	.disable = cdns_dsi_bridge_disable,
-	.pre_enable = cdns_dsi_bridge_pre_enable,
-	.enable = cdns_dsi_bridge_enable,
-	.post_disable = cdns_dsi_bridge_post_disable,
+	.atomic_disable = cdns_dsi_bridge_atomic_disable,
+	.atomic_pre_enable = cdns_dsi_bridge_atomic_pre_enable,
+	.atomic_enable = cdns_dsi_bridge_atomic_enable,
+	.atomic_post_disable = cdns_dsi_bridge_atomic_post_disable,
+	.atomic_check = cdns_dsi_bridge_atomic_check,
+	.atomic_reset = cdns_dsi_bridge_atomic_reset,
+	.atomic_duplicate_state = cdns_dsi_bridge_atomic_duplicate_state,
+	.atomic_destroy_state = cdns_dsi_bridge_atomic_destroy_state,
+	.atomic_get_input_bus_fmts = cdns_dsi_bridge_get_input_bus_fmts,
 };
 
 static int cdns_dsi_attach(struct mipi_dsi_host *host,
@@ -920,8 +1066,6 @@ static int cdns_dsi_attach(struct mipi_dsi_host *host,
 	struct cdns_dsi_output *output = &dsi->output;
 	struct cdns_dsi_input *input = &dsi->input;
 	struct drm_bridge *bridge;
-	struct drm_panel *panel;
-	struct device_node *np;
 	int ret;
 
 	/*
@@ -939,26 +1083,10 @@ static int cdns_dsi_attach(struct mipi_dsi_host *host,
 	/*
 	 * The host <-> device link might be described using an OF-graph
 	 * representation, in this case we extract the device of_node from
-	 * this representation, otherwise we use dsidev->dev.of_node which
-	 * should have been filled by the core.
+	 * this representation.
 	 */
-	np = of_graph_get_remote_node(dsi->base.dev->of_node, DSI_OUTPUT_PORT,
-				      dev->channel);
-	if (!np)
-		np = of_node_get(dev->dev.of_node);
-
-	panel = of_drm_find_panel(np);
-	if (!IS_ERR(panel)) {
-		bridge = drm_panel_bridge_add_typed(panel,
-						    DRM_MODE_CONNECTOR_DSI);
-	} else {
-		bridge = of_drm_find_bridge(dev->dev.of_node);
-		if (!bridge)
-			bridge = ERR_PTR(-EINVAL);
-	}
-
-	of_node_put(np);
-
+	bridge = devm_drm_of_get_bridge(dsi->base.dev, dsi->base.dev->of_node,
+					DSI_OUTPUT_PORT, dev->channel);
 	if (IS_ERR(bridge)) {
 		ret = PTR_ERR(bridge);
 		dev_err(host->dev, "failed to add DSI device %s (err = %d)",
@@ -968,7 +1096,6 @@ static int cdns_dsi_attach(struct mipi_dsi_host *host,
 
 	output->dev = dev;
 	output->bridge = bridge;
-	output->panel = panel;
 
 	/*
 	 * The DSI output has been properly configured, we can now safely
@@ -984,12 +1111,9 @@ static int cdns_dsi_detach(struct mipi_dsi_host *host,
 			   struct mipi_dsi_device *dev)
 {
 	struct cdns_dsi *dsi = to_cdns_dsi(host);
-	struct cdns_dsi_output *output = &dsi->output;
 	struct cdns_dsi_input *input = &dsi->input;
 
 	drm_bridge_remove(&input->bridge);
-	if (output->panel)
-		drm_panel_bridge_remove(output->bridge);
 
 	return 0;
 }
@@ -1152,7 +1276,6 @@ static int __maybe_unused cdns_dsi_suspend(struct device *dev)
 	clk_disable_unprepare(dsi->dsi_sys_clk);
 	clk_disable_unprepare(dsi->dsi_p_clk);
 	reset_control_assert(dsi->dsi_p_rst);
-	dsi->link_initialized = false;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.h b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.h
index ca7ea2da635c..5db5dbbbcaad 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.h
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.h
@@ -10,7 +10,6 @@
 
 #include <drm/drm_bridge.h>
 #include <drm/drm_mipi_dsi.h>
-#include <drm/drm_panel.h>
 
 #include <linux/bits.h>
 #include <linux/completion.h>
@@ -21,7 +20,6 @@ struct reset_control;
 
 struct cdns_dsi_output {
 	struct mipi_dsi_device *dev;
-	struct drm_panel *panel;
 	struct drm_bridge *bridge;
 	union phy_configure_opts phy_opts;
 };
diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
index 81fad14c2cd5..b431e7efd1f0 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
@@ -546,76 +546,6 @@ out:
 }
 
 /**
- * cdns_mhdp_link_power_up() - power up a DisplayPort link
- * @aux: DisplayPort AUX channel
- * @link: pointer to a structure containing the link configuration
- *
- * Returns 0 on success or a negative error code on failure.
- */
-static
-int cdns_mhdp_link_power_up(struct drm_dp_aux *aux, struct cdns_mhdp_link *link)
-{
-	u8 value;
-	int err;
-
-	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
-	if (link->revision < 0x11)
-		return 0;
-
-	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
-	if (err < 0)
-		return err;
-
-	value &= ~DP_SET_POWER_MASK;
-	value |= DP_SET_POWER_D0;
-
-	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
-	if (err < 0)
-		return err;
-
-	/*
-	 * According to the DP 1.1 specification, a "Sink Device must exit the
-	 * power saving state within 1 ms" (Section 2.5.3.1, Table 5-52, "Sink
-	 * Control Field" (register 0x600).
-	 */
-	usleep_range(1000, 2000);
-
-	return 0;
-}
-
-/**
- * cdns_mhdp_link_power_down() - power down a DisplayPort link
- * @aux: DisplayPort AUX channel
- * @link: pointer to a structure containing the link configuration
- *
- * Returns 0 on success or a negative error code on failure.
- */
-static
-int cdns_mhdp_link_power_down(struct drm_dp_aux *aux,
-			      struct cdns_mhdp_link *link)
-{
-	u8 value;
-	int err;
-
-	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
-	if (link->revision < 0x11)
-		return 0;
-
-	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
-	if (err < 0)
-		return err;
-
-	value &= ~DP_SET_POWER_MASK;
-	value |= DP_SET_POWER_D3;
-
-	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
-	if (err < 0)
-		return err;
-
-	return 0;
-}
-
-/**
  * cdns_mhdp_link_configure() - configure a DisplayPort link
  * @aux: DisplayPort AUX channel
  * @link: pointer to a structure containing the link configuration
@@ -1453,7 +1383,7 @@ static int cdns_mhdp_link_up(struct cdns_mhdp_device *mhdp)
 		mhdp->link.capabilities |= DP_LINK_CAP_ENHANCED_FRAMING;
 
 	dev_dbg(mhdp->dev, "Set sink device power state via DPCD\n");
-	cdns_mhdp_link_power_up(&mhdp->aux, &mhdp->link);
+	drm_dp_link_power_up(&mhdp->aux, mhdp->link.revision);
 
 	cdns_mhdp_fill_sink_caps(mhdp, dpcd);
 
@@ -1500,7 +1430,7 @@ static void cdns_mhdp_link_down(struct cdns_mhdp_device *mhdp)
 	WARN_ON(!mutex_is_locked(&mhdp->link_mutex));
 
 	if (mhdp->plugged)
-		cdns_mhdp_link_power_down(&mhdp->aux, &mhdp->link);
+		drm_dp_link_power_down(&mhdp->aux, mhdp->link.revision);
 
 	mhdp->link_up = false;
 }
@@ -1726,6 +1656,7 @@ static int cdns_mhdp_connector_init(struct cdns_mhdp_device *mhdp)
 }
 
 static int cdns_mhdp_attach(struct drm_bridge *bridge,
+			    struct drm_encoder *encoder,
 			    enum drm_bridge_attach_flags flags)
 {
 	struct cdns_mhdp_device *mhdp = bridge_to_mhdp(bridge);
@@ -2305,7 +2236,7 @@ static int cdns_mhdp_update_link_status(struct cdns_mhdp_device *mhdp)
 		 * If everything looks fine, just return, as we don't handle
 		 * DP IRQs.
 		 */
-		if (ret > 0 &&
+		if (!ret &&
 		    drm_dp_channel_eq_ok(status, mhdp->link.num_lanes) &&
 		    drm_dp_clock_recovery_ok(status, mhdp->link.num_lanes))
 			goto out;
diff --git a/drivers/gpu/drm/bridge/chipone-icn6211.c b/drivers/gpu/drm/bridge/chipone-icn6211.c
index 81f7c701961f..634c5b030667 100644
--- a/drivers/gpu/drm/bridge/chipone-icn6211.c
+++ b/drivers/gpu/drm/bridge/chipone-icn6211.c
@@ -580,11 +580,13 @@ static int chipone_dsi_host_attach(struct chipone *icn)
 	return ret;
 }
 
-static int chipone_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags)
+static int chipone_attach(struct drm_bridge *bridge,
+			  struct drm_encoder *encoder,
+			  enum drm_bridge_attach_flags flags)
 {
 	struct chipone *icn = bridge_to_chipone(bridge);
 
-	return drm_bridge_attach(bridge->encoder, icn->panel_bridge, bridge, flags);
+	return drm_bridge_attach(encoder, icn->panel_bridge, bridge, flags);
 }
 
 #define MAX_INPUT_SEL_FORMATS	1
diff --git a/drivers/gpu/drm/bridge/chrontel-ch7033.c b/drivers/gpu/drm/bridge/chrontel-ch7033.c
index da17f0978a79..210c45c1efd4 100644
--- a/drivers/gpu/drm/bridge/chrontel-ch7033.c
+++ b/drivers/gpu/drm/bridge/chrontel-ch7033.c
@@ -268,13 +268,14 @@ static void ch7033_hpd_event(void *arg, enum drm_connector_status status)
 }
 
 static int ch7033_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct ch7033_priv *priv = bridge_to_ch7033_priv(bridge);
 	struct drm_connector *connector = &priv->connector;
 	int ret;
 
-	ret = drm_bridge_attach(bridge->encoder, priv->next_bridge, bridge,
+	ret = drm_bridge_attach(encoder, priv->next_bridge, bridge,
 				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret)
 		return ret;
@@ -305,7 +306,7 @@ static int ch7033_bridge_attach(struct drm_bridge *bridge,
 		return ret;
 	}
 
-	return drm_connector_attach_encoder(&priv->connector, bridge->encoder);
+	return drm_connector_attach_encoder(&priv->connector, encoder);
 }
 
 static void ch7033_bridge_detach(struct drm_bridge *bridge)
diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c
index 72bc508d4e6e..badd2c7f91a1 100644
--- a/drivers/gpu/drm/bridge/display-connector.c
+++ b/drivers/gpu/drm/bridge/display-connector.c
@@ -34,6 +34,7 @@ to_display_connector(struct drm_bridge *bridge)
 }
 
 static int display_connector_attach(struct drm_bridge *bridge,
+				    struct drm_encoder *encoder,
 				    enum drm_bridge_attach_flags flags)
 {
 	return flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR ? 0 : -EINVAL;
@@ -209,9 +210,10 @@ static int display_connector_probe(struct platform_device *pdev)
 	const char *label = NULL;
 	int ret;
 
-	conn = devm_kzalloc(&pdev->dev, sizeof(*conn), GFP_KERNEL);
-	if (!conn)
-		return -ENOMEM;
+	conn = devm_drm_bridge_alloc(&pdev->dev, struct display_connector, bridge,
+				     &display_connector_bridge_funcs);
+	if (IS_ERR(conn))
+		return PTR_ERR(conn);
 
 	platform_set_drvdata(pdev, conn);
 
@@ -361,7 +363,6 @@ static int display_connector_probe(struct platform_device *pdev)
 		}
 	}
 
-	conn->bridge.funcs = &display_connector_bridge_funcs;
 	conn->bridge.of_node = pdev->dev.of_node;
 
 	if (conn->bridge.ddc)
diff --git a/drivers/gpu/drm/bridge/fsl-ldb.c b/drivers/gpu/drm/bridge/fsl-ldb.c
index 26ae1ab5237f..2cb6dfc7a6d3 100644
--- a/drivers/gpu/drm/bridge/fsl-ldb.c
+++ b/drivers/gpu/drm/bridge/fsl-ldb.c
@@ -113,11 +113,12 @@ static unsigned long fsl_ldb_link_frequency(struct fsl_ldb *fsl_ldb, int clock)
 }
 
 static int fsl_ldb_attach(struct drm_bridge *bridge,
+			  struct drm_encoder *encoder,
 			  enum drm_bridge_attach_flags flags)
 {
 	struct fsl_ldb *fsl_ldb = to_fsl_ldb(bridge);
 
-	return drm_bridge_attach(bridge->encoder, fsl_ldb->panel_bridge,
+	return drm_bridge_attach(encoder, fsl_ldb->panel_bridge,
 				 bridge, flags);
 }
 
@@ -180,9 +181,9 @@ static void fsl_ldb_atomic_enable(struct drm_bridge *bridge,
 
 	configured_link_freq = clk_get_rate(fsl_ldb->clk);
 	if (configured_link_freq != requested_link_freq)
-		dev_warn(fsl_ldb->dev, "Configured LDB clock (%lu Hz) does not match requested LVDS clock: %lu Hz\n",
-			 configured_link_freq,
-			 requested_link_freq);
+		dev_warn(fsl_ldb->dev,
+			 "Configured %pC clock (%lu Hz) does not match requested LVDS clock: %lu Hz\n",
+			 fsl_ldb->clk, configured_link_freq, requested_link_freq);
 
 	clk_prepare_enable(fsl_ldb->clk);
 
diff --git a/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c b/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c
index 9b5bebbe357d..6149ba141a38 100644
--- a/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c
+++ b/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c
@@ -104,7 +104,7 @@ void ldb_bridge_disable_helper(struct drm_bridge *bridge)
 }
 EXPORT_SYMBOL_GPL(ldb_bridge_disable_helper);
 
-int ldb_bridge_attach_helper(struct drm_bridge *bridge,
+int ldb_bridge_attach_helper(struct drm_bridge *bridge, struct drm_encoder *encoder,
 			     enum drm_bridge_attach_flags flags)
 {
 	struct ldb_channel *ldb_ch = bridge->driver_private;
@@ -116,9 +116,8 @@ int ldb_bridge_attach_helper(struct drm_bridge *bridge,
 		return -EINVAL;
 	}
 
-	return drm_bridge_attach(bridge->encoder,
-				ldb_ch->next_bridge, bridge,
-				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+	return drm_bridge_attach(encoder, ldb_ch->next_bridge, bridge,
+				 DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 }
 EXPORT_SYMBOL_GPL(ldb_bridge_attach_helper);
 
@@ -191,8 +190,7 @@ int ldb_find_next_bridge_helper(struct ldb *ldb)
 }
 EXPORT_SYMBOL_GPL(ldb_find_next_bridge_helper);
 
-void ldb_add_bridge_helper(struct ldb *ldb,
-			   const struct drm_bridge_funcs *bridge_funcs)
+void ldb_add_bridge_helper(struct ldb *ldb)
 {
 	struct ldb_channel *ldb_ch;
 	int i;
@@ -204,7 +202,6 @@ void ldb_add_bridge_helper(struct ldb *ldb,
 			continue;
 
 		ldb_ch->bridge.driver_private = ldb_ch;
-		ldb_ch->bridge.funcs = bridge_funcs;
 		ldb_ch->bridge.of_node = ldb_ch->np;
 
 		drm_bridge_add(&ldb_ch->bridge);
diff --git a/drivers/gpu/drm/bridge/imx/imx-ldb-helper.h b/drivers/gpu/drm/bridge/imx/imx-ldb-helper.h
index a0a5cde27fbc..de187e326999 100644
--- a/drivers/gpu/drm/bridge/imx/imx-ldb-helper.h
+++ b/drivers/gpu/drm/bridge/imx/imx-ldb-helper.h
@@ -81,15 +81,14 @@ void ldb_bridge_enable_helper(struct drm_bridge *bridge);
 
 void ldb_bridge_disable_helper(struct drm_bridge *bridge);
 
-int ldb_bridge_attach_helper(struct drm_bridge *bridge,
+int ldb_bridge_attach_helper(struct drm_bridge *bridge, struct drm_encoder *encoder,
 			     enum drm_bridge_attach_flags flags);
 
 int ldb_init_helper(struct ldb *ldb);
 
 int ldb_find_next_bridge_helper(struct ldb *ldb);
 
-void ldb_add_bridge_helper(struct ldb *ldb,
-			   const struct drm_bridge_funcs *bridge_funcs);
+void ldb_add_bridge_helper(struct ldb *ldb);
 
 void ldb_remove_bridge_helper(struct ldb *ldb);
 
diff --git a/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c b/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c
index 3ebf0b9866de..f072c6ed39ef 100644
--- a/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c
+++ b/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c
@@ -23,7 +23,8 @@ struct imx_legacy_bridge {
 #define to_imx_legacy_bridge(bridge)	container_of(bridge, struct imx_legacy_bridge, base)
 
 static int imx_legacy_bridge_attach(struct drm_bridge *bridge,
-				   enum drm_bridge_attach_flags flags)
+				    struct drm_encoder *encoder,
+				    enum drm_bridge_attach_flags flags)
 {
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
@@ -76,9 +77,9 @@ struct drm_bridge *devm_imx_drm_legacy_bridge(struct device *dev,
 	imx_bridge->base.ops = DRM_BRIDGE_OP_MODES;
 	imx_bridge->base.type = type;
 
-       ret = devm_drm_bridge_add(dev, &imx_bridge->base);
-       if (ret)
-               return ERR_PTR(ret);
+	ret = devm_drm_bridge_add(dev, &imx_bridge->base);
+	if (ret)
+		return ERR_PTR(ret);
 
 	return &imx_bridge->base;
 }
diff --git a/drivers/gpu/drm/bridge/imx/imx8mp-hdmi-pvi.c b/drivers/gpu/drm/bridge/imx/imx8mp-hdmi-pvi.c
index a17433a7c755..8a4fd7d77a8d 100644
--- a/drivers/gpu/drm/bridge/imx/imx8mp-hdmi-pvi.c
+++ b/drivers/gpu/drm/bridge/imx/imx8mp-hdmi-pvi.c
@@ -40,11 +40,12 @@ to_imx8mp_hdmi_pvi(struct drm_bridge *bridge)
 }
 
 static int imx8mp_hdmi_pvi_bridge_attach(struct drm_bridge *bridge,
+					 struct drm_encoder *encoder,
 					 enum drm_bridge_attach_flags flags)
 {
 	struct imx8mp_hdmi_pvi *pvi = to_imx8mp_hdmi_pvi(bridge);
 
-	return drm_bridge_attach(bridge->encoder, pvi->next_bridge,
+	return drm_bridge_attach(encoder, pvi->next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/imx/imx8qm-ldb.c b/drivers/gpu/drm/bridge/imx/imx8qm-ldb.c
index 524aac751359..47aa65938e6a 100644
--- a/drivers/gpu/drm/bridge/imx/imx8qm-ldb.c
+++ b/drivers/gpu/drm/bridge/imx/imx8qm-ldb.c
@@ -47,7 +47,7 @@ struct imx8qm_ldb_channel {
 struct imx8qm_ldb {
 	struct ldb base;
 	struct device *dev;
-	struct imx8qm_ldb_channel channel[MAX_LDB_CHAN_NUM];
+	struct imx8qm_ldb_channel *channel[MAX_LDB_CHAN_NUM];
 	struct clk *clk_pixel;
 	struct clk *clk_bypass;
 	int active_chno;
@@ -107,7 +107,7 @@ static int imx8qm_ldb_bridge_atomic_check(struct drm_bridge *bridge,
 
 	if (is_split) {
 		imx8qm_ldb_ch =
-			&imx8qm_ldb->channel[imx8qm_ldb->active_chno ^ 1];
+			imx8qm_ldb->channel[imx8qm_ldb->active_chno ^ 1];
 		imx8qm_ldb_set_phy_cfg(imx8qm_ldb, di_clk, is_split, true,
 				       phy_cfg);
 		ret = phy_validate(imx8qm_ldb_ch->phy, PHY_MODE_LVDS, 0, &opts);
@@ -158,7 +158,7 @@ imx8qm_ldb_bridge_mode_set(struct drm_bridge *bridge,
 
 	if (is_split) {
 		imx8qm_ldb_ch =
-			&imx8qm_ldb->channel[imx8qm_ldb->active_chno ^ 1];
+			imx8qm_ldb->channel[imx8qm_ldb->active_chno ^ 1];
 		imx8qm_ldb_set_phy_cfg(imx8qm_ldb, di_clk, is_split, true,
 				       phy_cfg);
 		ret = phy_configure(imx8qm_ldb_ch->phy, &opts);
@@ -226,13 +226,13 @@ static void imx8qm_ldb_bridge_atomic_enable(struct drm_bridge *bridge,
 	}
 
 	if (is_split) {
-		ret = phy_power_on(imx8qm_ldb->channel[0].phy);
+		ret = phy_power_on(imx8qm_ldb->channel[0]->phy);
 		if (ret)
 			DRM_DEV_ERROR(dev,
 				      "failed to power on channel0 PHY: %d\n",
 				      ret);
 
-		ret = phy_power_on(imx8qm_ldb->channel[1].phy);
+		ret = phy_power_on(imx8qm_ldb->channel[1]->phy);
 		if (ret)
 			DRM_DEV_ERROR(dev,
 				      "failed to power on channel1 PHY: %d\n",
@@ -261,12 +261,12 @@ static void imx8qm_ldb_bridge_atomic_disable(struct drm_bridge *bridge,
 	ldb_bridge_disable_helper(bridge);
 
 	if (is_split) {
-		ret = phy_power_off(imx8qm_ldb->channel[0].phy);
+		ret = phy_power_off(imx8qm_ldb->channel[0]->phy);
 		if (ret)
 			DRM_DEV_ERROR(dev,
 				      "failed to power off channel0 PHY: %d\n",
 				      ret);
-		ret = phy_power_off(imx8qm_ldb->channel[1].phy);
+		ret = phy_power_off(imx8qm_ldb->channel[1]->phy);
 		if (ret)
 			DRM_DEV_ERROR(dev,
 				      "failed to power off channel1 PHY: %d\n",
@@ -412,7 +412,7 @@ static int imx8qm_ldb_get_phy(struct imx8qm_ldb *imx8qm_ldb)
 	int i, ret;
 
 	for (i = 0; i < MAX_LDB_CHAN_NUM; i++) {
-		imx8qm_ldb_ch = &imx8qm_ldb->channel[i];
+		imx8qm_ldb_ch = imx8qm_ldb->channel[i];
 		ldb_ch = &imx8qm_ldb_ch->base;
 
 		if (!ldb_ch->is_available)
@@ -448,6 +448,14 @@ static int imx8qm_ldb_probe(struct platform_device *pdev)
 	if (!imx8qm_ldb)
 		return -ENOMEM;
 
+	for (i = 0; i < MAX_LDB_CHAN_NUM; i++) {
+		imx8qm_ldb->channel[i] =
+			devm_drm_bridge_alloc(dev, struct imx8qm_ldb_channel, base.bridge,
+					      &imx8qm_ldb_bridge_funcs);
+		if (IS_ERR(imx8qm_ldb->channel[i]))
+			return PTR_ERR(imx8qm_ldb->channel[i]);
+	}
+
 	imx8qm_ldb->clk_pixel = devm_clk_get(dev, "pixel");
 	if (IS_ERR(imx8qm_ldb->clk_pixel)) {
 		ret = PTR_ERR(imx8qm_ldb->clk_pixel);
@@ -473,7 +481,7 @@ static int imx8qm_ldb_probe(struct platform_device *pdev)
 	ldb->ctrl_reg = 0xe0;
 
 	for (i = 0; i < MAX_LDB_CHAN_NUM; i++)
-		ldb->channel[i] = &imx8qm_ldb->channel[i].base;
+		ldb->channel[i] = &imx8qm_ldb->channel[i]->base;
 
 	ret = ldb_init_helper(ldb);
 	if (ret)
@@ -499,12 +507,12 @@ static int imx8qm_ldb_probe(struct platform_device *pdev)
 		}
 
 		imx8qm_ldb->active_chno = 0;
-		imx8qm_ldb_ch = &imx8qm_ldb->channel[0];
+		imx8qm_ldb_ch = imx8qm_ldb->channel[0];
 		ldb_ch = &imx8qm_ldb_ch->base;
 		ldb_ch->link_type = pixel_order;
 	} else {
 		for (i = 0; i < MAX_LDB_CHAN_NUM; i++) {
-			imx8qm_ldb_ch = &imx8qm_ldb->channel[i];
+			imx8qm_ldb_ch = imx8qm_ldb->channel[i];
 			ldb_ch = &imx8qm_ldb_ch->base;
 
 			if (ldb_ch->is_available) {
@@ -525,7 +533,7 @@ static int imx8qm_ldb_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, imx8qm_ldb);
 	pm_runtime_enable(dev);
 
-	ldb_add_bridge_helper(ldb, &imx8qm_ldb_bridge_funcs);
+	ldb_add_bridge_helper(ldb);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/bridge/imx/imx8qxp-ldb.c b/drivers/gpu/drm/bridge/imx/imx8qxp-ldb.c
index 3cb484773ddf..5d272916e200 100644
--- a/drivers/gpu/drm/bridge/imx/imx8qxp-ldb.c
+++ b/drivers/gpu/drm/bridge/imx/imx8qxp-ldb.c
@@ -44,7 +44,7 @@ struct imx8qxp_ldb_channel {
 struct imx8qxp_ldb {
 	struct ldb base;
 	struct device *dev;
-	struct imx8qxp_ldb_channel channel[MAX_LDB_CHAN_NUM];
+	struct imx8qxp_ldb_channel *channel[MAX_LDB_CHAN_NUM];
 	struct clk *clk_pixel;
 	struct clk *clk_bypass;
 	struct drm_bridge *companion;
@@ -410,7 +410,7 @@ static const struct drm_bridge_funcs imx8qxp_ldb_bridge_funcs = {
 static int imx8qxp_ldb_set_di_id(struct imx8qxp_ldb *imx8qxp_ldb)
 {
 	struct imx8qxp_ldb_channel *imx8qxp_ldb_ch =
-			 &imx8qxp_ldb->channel[imx8qxp_ldb->active_chno];
+			 imx8qxp_ldb->channel[imx8qxp_ldb->active_chno];
 	struct ldb_channel *ldb_ch = &imx8qxp_ldb_ch->base;
 	struct device_node *ep, *remote;
 	struct device *dev = imx8qxp_ldb->dev;
@@ -456,7 +456,7 @@ imx8qxp_ldb_check_chno_and_dual_link(struct ldb_channel *ldb_ch, int link)
 static int imx8qxp_ldb_parse_dt_companion(struct imx8qxp_ldb *imx8qxp_ldb)
 {
 	struct imx8qxp_ldb_channel *imx8qxp_ldb_ch =
-			 &imx8qxp_ldb->channel[imx8qxp_ldb->active_chno];
+			 imx8qxp_ldb->channel[imx8qxp_ldb->active_chno];
 	struct ldb_channel *ldb_ch = &imx8qxp_ldb_ch->base;
 	struct ldb_channel *companion_ldb_ch;
 	struct device_node *companion;
@@ -586,6 +586,14 @@ static int imx8qxp_ldb_probe(struct platform_device *pdev)
 	if (!imx8qxp_ldb)
 		return -ENOMEM;
 
+	for (i = 0; i < MAX_LDB_CHAN_NUM; i++) {
+		imx8qxp_ldb->channel[i] =
+			devm_drm_bridge_alloc(dev, struct imx8qxp_ldb_channel, base.bridge,
+					      &imx8qxp_ldb_bridge_funcs);
+		if (IS_ERR(imx8qxp_ldb->channel[i]))
+			return PTR_ERR(imx8qxp_ldb->channel[i]);
+	}
+
 	imx8qxp_ldb->clk_pixel = devm_clk_get(dev, "pixel");
 	if (IS_ERR(imx8qxp_ldb->clk_pixel)) {
 		ret = PTR_ERR(imx8qxp_ldb->clk_pixel);
@@ -611,7 +619,7 @@ static int imx8qxp_ldb_probe(struct platform_device *pdev)
 	ldb->ctrl_reg = 0xe0;
 
 	for (i = 0; i < MAX_LDB_CHAN_NUM; i++)
-		ldb->channel[i] = &imx8qxp_ldb->channel[i].base;
+		ldb->channel[i] = &imx8qxp_ldb->channel[i]->base;
 
 	ret = ldb_init_helper(ldb);
 	if (ret)
@@ -627,7 +635,7 @@ static int imx8qxp_ldb_probe(struct platform_device *pdev)
 	}
 
 	for (i = 0; i < MAX_LDB_CHAN_NUM; i++) {
-		imx8qxp_ldb_ch = &imx8qxp_ldb->channel[i];
+		imx8qxp_ldb_ch = imx8qxp_ldb->channel[i];
 		ldb_ch = &imx8qxp_ldb_ch->base;
 
 		if (ldb_ch->is_available) {
@@ -660,9 +668,9 @@ static int imx8qxp_ldb_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, imx8qxp_ldb);
 	pm_runtime_enable(dev);
 
-	ldb_add_bridge_helper(ldb, &imx8qxp_ldb_bridge_funcs);
+	ldb_add_bridge_helper(ldb);
 
-	return ret;
+	return 0;
 }
 
 static void imx8qxp_ldb_remove(struct platform_device *pdev)
diff --git a/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-combiner.c b/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-combiner.c
index 1d9529dc7f2a..1f6fd488e703 100644
--- a/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-combiner.c
+++ b/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-combiner.c
@@ -108,6 +108,7 @@ imx8qxp_pc_bridge_mode_valid(struct drm_bridge *bridge,
 }
 
 static int imx8qxp_pc_bridge_attach(struct drm_bridge *bridge,
+				    struct drm_encoder *encoder,
 				    enum drm_bridge_attach_flags flags)
 {
 	struct imx8qxp_pc_channel *ch = bridge->driver_private;
@@ -119,7 +120,7 @@ static int imx8qxp_pc_bridge_attach(struct drm_bridge *bridge,
 		return -EINVAL;
 	}
 
-	return drm_bridge_attach(bridge->encoder,
+	return drm_bridge_attach(encoder,
 				 ch->next_bridge, bridge,
 				 DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 }
diff --git a/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-link.c b/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-link.c
index cd6818db0fd3..e092c9ea99b0 100644
--- a/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-link.c
+++ b/drivers/gpu/drm/bridge/imx/imx8qxp-pixel-link.c
@@ -128,6 +128,7 @@ static void imx8qxp_pixel_link_set_mst_addr(struct imx8qxp_pixel_link *pl)
 }
 
 static int imx8qxp_pixel_link_bridge_attach(struct drm_bridge *bridge,
+					    struct drm_encoder *encoder,
 					    enum drm_bridge_attach_flags flags)
 {
 	struct imx8qxp_pixel_link *pl = bridge->driver_private;
@@ -138,7 +139,7 @@ static int imx8qxp_pixel_link_bridge_attach(struct drm_bridge *bridge,
 		return -EINVAL;
 	}
 
-	return drm_bridge_attach(bridge->encoder,
+	return drm_bridge_attach(encoder,
 				 pl->next_bridge, bridge,
 				 DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 }
diff --git a/drivers/gpu/drm/bridge/imx/imx8qxp-pxl2dpi.c b/drivers/gpu/drm/bridge/imx/imx8qxp-pxl2dpi.c
index 49dd4f96d52c..da138ab51b3b 100644
--- a/drivers/gpu/drm/bridge/imx/imx8qxp-pxl2dpi.c
+++ b/drivers/gpu/drm/bridge/imx/imx8qxp-pxl2dpi.c
@@ -48,6 +48,7 @@ struct imx8qxp_pxl2dpi {
 #define bridge_to_p2d(b)	container_of(b, struct imx8qxp_pxl2dpi, bridge)
 
 static int imx8qxp_pxl2dpi_bridge_attach(struct drm_bridge *bridge,
+					 struct drm_encoder *encoder,
 					 enum drm_bridge_attach_flags flags)
 {
 	struct imx8qxp_pxl2dpi *p2d = bridge->driver_private;
@@ -58,7 +59,7 @@ static int imx8qxp_pxl2dpi_bridge_attach(struct drm_bridge *bridge,
 		return -EINVAL;
 	}
 
-	return drm_bridge_attach(bridge->encoder,
+	return drm_bridge_attach(encoder,
 				 p2d->next_bridge, bridge,
 				 DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 }
diff --git a/drivers/gpu/drm/bridge/ite-it6263.c b/drivers/gpu/drm/bridge/ite-it6263.c
index 21152a1c28f7..a3a63a977b0a 100644
--- a/drivers/gpu/drm/bridge/ite-it6263.c
+++ b/drivers/gpu/drm/bridge/ite-it6263.c
@@ -665,13 +665,14 @@ it6263_bridge_mode_valid(struct drm_bridge *bridge,
 }
 
 static int it6263_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct it6263 *it = bridge_to_it6263(bridge);
 	struct drm_connector *connector;
 	int ret;
 
-	ret = drm_bridge_attach(bridge->encoder, it->next_bridge, bridge,
+	ret = drm_bridge_attach(encoder, it->next_bridge, bridge,
 				flags | DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret < 0)
 		return ret;
@@ -679,7 +680,7 @@ static int it6263_bridge_attach(struct drm_bridge *bridge,
 	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)
 		return 0;
 
-	connector = drm_bridge_connector_init(bridge->dev, bridge->encoder);
+	connector = drm_bridge_connector_init(bridge->dev, encoder);
 	if (IS_ERR(connector)) {
 		ret = PTR_ERR(connector);
 		dev_err(it->dev, "failed to initialize bridge connector: %d\n",
@@ -687,7 +688,7 @@ static int it6263_bridge_attach(struct drm_bridge *bridge,
 		return ret;
 	}
 
-	drm_connector_attach_encoder(connector, bridge->encoder);
+	drm_connector_attach_encoder(connector, encoder);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/bridge/ite-it6505.c b/drivers/gpu/drm/bridge/ite-it6505.c
index 8a607558ac89..1383d1e21afe 100644
--- a/drivers/gpu/drm/bridge/ite-it6505.c
+++ b/drivers/gpu/drm/bridge/ite-it6505.c
@@ -771,40 +771,6 @@ static void it6505_calc_video_info(struct it6505 *it6505)
 			     DRM_MODE_ARG(&it6505->video_info));
 }
 
-static int it6505_drm_dp_link_set_power(struct drm_dp_aux *aux,
-					struct it6505_drm_dp_link *link,
-					u8 mode)
-{
-	u8 value;
-	int err;
-
-	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
-	if (link->revision < DPCD_V_1_1)
-		return 0;
-
-	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
-	if (err < 0)
-		return err;
-
-	value &= ~DP_SET_POWER_MASK;
-	value |= mode;
-
-	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
-	if (err < 0)
-		return err;
-
-	if (mode == DP_SET_POWER_D0) {
-		/*
-		 * According to the DP 1.1 specification, a "Sink Device must
-		 * exit the power saving state within 1 ms" (Section 2.5.3.1,
-		 * Table 5-52, "Sink Control Field" (register 0x600).
-		 */
-		usleep_range(1000, 2000);
-	}
-
-	return 0;
-}
-
 static void it6505_clear_int(struct it6505 *it6505)
 {
 	it6505_write(it6505, INT_STATUS_01, 0xFF);
@@ -2578,8 +2544,7 @@ static void it6505_irq_hpd(struct it6505 *it6505)
 		}
 		it6505->auto_train_retry = AUTO_TRAIN_RETRY;
 
-		it6505_drm_dp_link_set_power(&it6505->aux, &it6505->link,
-					     DP_SET_POWER_D0);
+		drm_dp_link_power_up(&it6505->aux, it6505->link.revision);
 		dp_sink_count = it6505_dpcd_read(it6505, DP_SINK_COUNT);
 		it6505->sink_count = DP_GET_SINK_COUNT(dp_sink_count);
 
@@ -2910,8 +2875,7 @@ static enum drm_connector_status it6505_detect(struct it6505 *it6505)
 	}
 
 	if (it6505->hpd_state) {
-		it6505_drm_dp_link_set_power(&it6505->aux, &it6505->link,
-					     DP_SET_POWER_D0);
+		drm_dp_link_power_up(&it6505->aux, it6505->link.revision);
 		dp_sink_count = it6505_dpcd_read(it6505, DP_SINK_COUNT);
 		it6505->sink_count = DP_GET_SINK_COUNT(dp_sink_count);
 		DRM_DEV_DEBUG_DRIVER(dev, "it6505->sink_count:%d branch:%d",
@@ -3124,6 +3088,7 @@ static inline struct it6505 *bridge_to_it6505(struct drm_bridge *bridge)
 }
 
 static int it6505_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct it6505 *it6505 = bridge_to_it6505(bridge);
@@ -3233,8 +3198,7 @@ static void it6505_bridge_atomic_enable(struct drm_bridge *bridge,
 	it6505_int_mask_enable(it6505);
 	it6505_video_reset(it6505);
 
-	it6505_drm_dp_link_set_power(&it6505->aux, &it6505->link,
-				     DP_SET_POWER_D0);
+	drm_dp_link_power_up(&it6505->aux, it6505->link.revision);
 }
 
 static void it6505_bridge_atomic_disable(struct drm_bridge *bridge,
@@ -3246,8 +3210,7 @@ static void it6505_bridge_atomic_disable(struct drm_bridge *bridge,
 	DRM_DEV_DEBUG_DRIVER(dev, "start");
 
 	if (it6505->powered) {
-		it6505_drm_dp_link_set_power(&it6505->aux, &it6505->link,
-					     DP_SET_POWER_D3);
+		drm_dp_link_power_down(&it6505->aux, it6505->link.revision);
 		it6505_video_disable(it6505);
 	}
 }
diff --git a/drivers/gpu/drm/bridge/ite-it66121.c b/drivers/gpu/drm/bridge/ite-it66121.c
index b9f90f32145d..7b110ae53291 100644
--- a/drivers/gpu/drm/bridge/ite-it66121.c
+++ b/drivers/gpu/drm/bridge/ite-it66121.c
@@ -586,6 +586,7 @@ static bool it66121_is_hpd_detect(struct it66121_ctx *ctx)
 }
 
 static int it66121_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct it66121_ctx *ctx = container_of(bridge, struct it66121_ctx, bridge);
@@ -594,7 +595,7 @@ static int it66121_bridge_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	ret = drm_bridge_attach(bridge->encoder, ctx->next_bridge, bridge, flags);
+	ret = drm_bridge_attach(encoder, ctx->next_bridge, bridge, flags);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/bridge/lontium-lt8912b.c b/drivers/gpu/drm/bridge/lontium-lt8912b.c
index 52da204f5740..3e49d855b364 100644
--- a/drivers/gpu/drm/bridge/lontium-lt8912b.c
+++ b/drivers/gpu/drm/bridge/lontium-lt8912b.c
@@ -543,12 +543,13 @@ exit:
 }
 
 static int lt8912_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct lt8912 *lt = bridge_to_lt8912(bridge);
 	int ret;
 
-	ret = drm_bridge_attach(bridge->encoder, lt->hdmi_port, bridge,
+	ret = drm_bridge_attach(encoder, lt->hdmi_port, bridge,
 				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret < 0) {
 		dev_err(lt->dev, "Failed to attach next bridge (%d)\n", ret);
diff --git a/drivers/gpu/drm/bridge/lontium-lt9211.c b/drivers/gpu/drm/bridge/lontium-lt9211.c
index 0fc5ea18fe6a..9b2dac9bd63c 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9211.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9211.c
@@ -99,11 +99,12 @@ static struct lt9211 *bridge_to_lt9211(struct drm_bridge *bridge)
 }
 
 static int lt9211_attach(struct drm_bridge *bridge,
+			 struct drm_encoder *encoder,
 			 enum drm_bridge_attach_flags flags)
 {
 	struct lt9211 *ctx = bridge_to_lt9211(bridge);
 
-	return drm_bridge_attach(bridge->encoder, ctx->panel_bridge,
+	return drm_bridge_attach(encoder, ctx->panel_bridge,
 				 &ctx->bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c
index 026803034231..a35a8b8ca89c 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9611.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9611.c
@@ -740,11 +740,12 @@ static struct mipi_dsi_device *lt9611_attach_dsi(struct lt9611 *lt9611,
 }
 
 static int lt9611_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct lt9611 *lt9611 = bridge_to_lt9611(bridge);
 
-	return drm_bridge_attach(bridge->encoder, lt9611->next_bridge,
+	return drm_bridge_attach(encoder, lt9611->next_bridge,
 				 bridge, flags);
 }
 
@@ -1130,7 +1131,7 @@ static int lt9611_probe(struct i2c_client *client)
 	lt9611->bridge.of_node = client->dev.of_node;
 	lt9611->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID |
 			     DRM_BRIDGE_OP_HPD | DRM_BRIDGE_OP_MODES |
-			     DRM_BRIDGE_OP_HDMI;
+			     DRM_BRIDGE_OP_HDMI | DRM_BRIDGE_OP_HDMI_AUDIO;
 	lt9611->bridge.type = DRM_MODE_CONNECTOR_HDMIA;
 	lt9611->bridge.vendor = "Lontium";
 	lt9611->bridge.product = "LT9611";
diff --git a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c
index f4c3ff1fdc69..766da2cb45a7 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c
@@ -280,11 +280,12 @@ static struct mipi_dsi_device *lt9611uxc_attach_dsi(struct lt9611uxc *lt9611uxc,
 }
 
 static int lt9611uxc_bridge_attach(struct drm_bridge *bridge,
+				   struct drm_encoder *encoder,
 				   enum drm_bridge_attach_flags flags)
 {
 	struct lt9611uxc *lt9611uxc = bridge_to_lt9611uxc(bridge);
 
-	return drm_bridge_attach(bridge->encoder, lt9611uxc->next_bridge,
+	return drm_bridge_attach(encoder, lt9611uxc->next_bridge,
 				 bridge, flags);
 }
 
@@ -774,9 +775,9 @@ static int lt9611uxc_probe(struct i2c_client *client)
 		return -ENODEV;
 	}
 
-	lt9611uxc = devm_kzalloc(dev, sizeof(*lt9611uxc), GFP_KERNEL);
-	if (!lt9611uxc)
-		return -ENOMEM;
+	lt9611uxc = devm_drm_bridge_alloc(dev, struct lt9611uxc, bridge, &lt9611uxc_bridge_funcs);
+	if (IS_ERR(lt9611uxc))
+		return PTR_ERR(lt9611uxc);
 
 	lt9611uxc->dev = dev;
 	lt9611uxc->client = client;
@@ -855,7 +856,6 @@ retry:
 
 	i2c_set_clientdata(client, lt9611uxc);
 
-	lt9611uxc->bridge.funcs = &lt9611uxc_bridge_funcs;
 	lt9611uxc->bridge.of_node = client->dev.of_node;
 	lt9611uxc->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID;
 	if (lt9611uxc->hpd_supported)
@@ -880,7 +880,11 @@ retry:
 		}
 	}
 
-	return lt9611uxc_audio_init(dev, lt9611uxc);
+	ret = lt9611uxc_audio_init(dev, lt9611uxc);
+	if (ret)
+		goto err_remove_bridge;
+
+	return 0;
 
 err_remove_bridge:
 	free_irq(client->irq, lt9611uxc);
diff --git a/drivers/gpu/drm/bridge/lvds-codec.c b/drivers/gpu/drm/bridge/lvds-codec.c
index 389af0233fcd..1646e454e0b0 100644
--- a/drivers/gpu/drm/bridge/lvds-codec.c
+++ b/drivers/gpu/drm/bridge/lvds-codec.c
@@ -34,11 +34,12 @@ static inline struct lvds_codec *to_lvds_codec(struct drm_bridge *bridge)
 }
 
 static int lvds_codec_attach(struct drm_bridge *bridge,
+			     struct drm_encoder *encoder,
 			     enum drm_bridge_attach_flags flags)
 {
 	struct lvds_codec *lvds_codec = to_lvds_codec(bridge);
 
-	return drm_bridge_attach(bridge->encoder, lvds_codec->panel_bridge,
+	return drm_bridge_attach(encoder, lvds_codec->panel_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
index a47aabf134fd..15a5a1f644fc 100644
--- a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
+++ b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
@@ -190,6 +190,7 @@ static irqreturn_t ge_b850v3_lvds_irq_handler(int irq, void *dev_id)
 }
 
 static int ge_b850v3_lvds_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct i2c_client *stdp4028_i2c
diff --git a/drivers/gpu/drm/bridge/microchip-lvds.c b/drivers/gpu/drm/bridge/microchip-lvds.c
index 53dd140a1b8d..1d4ae0097df8 100644
--- a/drivers/gpu/drm/bridge/microchip-lvds.c
+++ b/drivers/gpu/drm/bridge/microchip-lvds.c
@@ -104,11 +104,12 @@ static void lvds_serialiser_on(struct mchp_lvds *lvds)
 }
 
 static int mchp_lvds_attach(struct drm_bridge *bridge,
+			    struct drm_encoder *encoder,
 			    enum drm_bridge_attach_flags flags)
 {
 	struct mchp_lvds *lvds = bridge_to_lvds(bridge);
 
-	return drm_bridge_attach(bridge->encoder, lvds->panel_bridge,
+	return drm_bridge_attach(encoder, lvds->panel_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
index d04c62a0cb9f..55912ae11f46 100644
--- a/drivers/gpu/drm/bridge/nwl-dsi.c
+++ b/drivers/gpu/drm/bridge/nwl-dsi.c
@@ -910,6 +910,7 @@ static void nwl_dsi_bridge_atomic_enable(struct drm_bridge *bridge,
 }
 
 static int nwl_dsi_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct nwl_dsi *dsi = bridge_to_dsi(bridge);
@@ -919,7 +920,7 @@ static int nwl_dsi_bridge_attach(struct drm_bridge *bridge,
 	if (IS_ERR(panel_bridge))
 		return PTR_ERR(panel_bridge);
 
-	return drm_bridge_attach(bridge->encoder, panel_bridge, bridge, flags);
+	return drm_bridge_attach(encoder, panel_bridge, bridge, flags);
 }
 
 static u32 *nwl_bridge_atomic_get_input_bus_fmts(struct drm_bridge *bridge,
diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c
index 27261b2ac9c8..25d7c415478b 100644
--- a/drivers/gpu/drm/bridge/nxp-ptn3460.c
+++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c
@@ -214,13 +214,14 @@ static const struct drm_connector_funcs ptn3460_connector_funcs = {
 };
 
 static int ptn3460_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct ptn3460_bridge *ptn_bridge = bridge_to_ptn3460(bridge);
 	int ret;
 
 	/* Let this driver create connector if requested */
-	ret = drm_bridge_attach(bridge->encoder, ptn_bridge->panel_bridge,
+	ret = drm_bridge_attach(encoder, ptn_bridge->panel_bridge,
 				bridge, flags | DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret < 0)
 		return ret;
@@ -239,7 +240,7 @@ static int ptn3460_bridge_attach(struct drm_bridge *bridge,
 					&ptn3460_connector_helper_funcs);
 	drm_connector_register(&ptn_bridge->connector);
 	drm_connector_attach_encoder(&ptn_bridge->connector,
-							bridge->encoder);
+							encoder);
 
 	drm_helper_hpd_irq_event(ptn_bridge->connector.dev);
 
diff --git a/drivers/gpu/drm/bridge/panel.c b/drivers/gpu/drm/bridge/panel.c
index 258c85c83a28..79b009ab9396 100644
--- a/drivers/gpu/drm/bridge/panel.c
+++ b/drivers/gpu/drm/bridge/panel.c
@@ -58,6 +58,7 @@ static const struct drm_connector_funcs panel_bridge_connector_funcs = {
 };
 
 static int panel_bridge_attach(struct drm_bridge *bridge,
+			       struct drm_encoder *encoder,
 			       enum drm_bridge_attach_flags flags)
 {
 	struct panel_bridge *panel_bridge = drm_bridge_to_panel_bridge(bridge);
@@ -81,7 +82,7 @@ static int panel_bridge_attach(struct drm_bridge *bridge,
 	drm_panel_bridge_set_orientation(connector, bridge);
 
 	drm_connector_attach_encoder(&panel_bridge->connector,
-					  bridge->encoder);
+					  encoder);
 
 	if (bridge->dev->registered) {
 		if (connector->funcs->reset)
diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c
index 13ada42a5514..8726fefc5c65 100644
--- a/drivers/gpu/drm/bridge/parade-ps8622.c
+++ b/drivers/gpu/drm/bridge/parade-ps8622.c
@@ -418,6 +418,7 @@ static void ps8622_post_disable(struct drm_bridge *bridge)
 }
 
 static int ps8622_attach(struct drm_bridge *bridge,
+			 struct drm_encoder *encoder,
 			 enum drm_bridge_attach_flags flags)
 {
 	struct ps8622_bridge *ps8622 = bridge_to_ps8622(bridge);
diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c
index a42138b33258..2422ff68c104 100644
--- a/drivers/gpu/drm/bridge/parade-ps8640.c
+++ b/drivers/gpu/drm/bridge/parade-ps8640.c
@@ -494,6 +494,7 @@ static void ps8640_atomic_post_disable(struct drm_bridge *bridge,
 }
 
 static int ps8640_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct ps8640 *ps_bridge = bridge_to_ps8640(bridge);
@@ -518,7 +519,7 @@ static int ps8640_bridge_attach(struct drm_bridge *bridge,
 	}
 
 	/* Attach the panel-bridge to the dsi bridge */
-	ret = drm_bridge_attach(bridge->encoder, ps_bridge->panel_bridge,
+	ret = drm_bridge_attach(encoder, ps_bridge->panel_bridge,
 				&ps_bridge->bridge, flags);
 	if (ret)
 		goto err_bridge_attach;
diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 54de6ed2fae8..0014c497e3fe 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -1640,11 +1640,12 @@ static void samsung_dsim_mode_set(struct drm_bridge *bridge,
 }
 
 static int samsung_dsim_attach(struct drm_bridge *bridge,
+			       struct drm_encoder *encoder,
 			       enum drm_bridge_attach_flags flags)
 {
 	struct samsung_dsim *dsi = bridge_to_dsi(bridge);
 
-	return drm_bridge_attach(bridge->encoder, dsi->out_bridge, bridge,
+	return drm_bridge_attach(encoder, dsi->out_bridge, bridge,
 				 flags);
 }
 
@@ -1935,9 +1936,9 @@ int samsung_dsim_probe(struct platform_device *pdev)
 	struct samsung_dsim *dsi;
 	int ret, i;
 
-	dsi = devm_kzalloc(dev, sizeof(*dsi), GFP_KERNEL);
-	if (!dsi)
-		return -ENOMEM;
+	dsi = devm_drm_bridge_alloc(dev, struct samsung_dsim, bridge, &samsung_dsim_bridge_funcs);
+	if (IS_ERR(dsi))
+		return PTR_ERR(dsi);
 
 	init_completion(&dsi->completed);
 	spin_lock_init(&dsi->transfer_lock);
@@ -2007,7 +2008,6 @@ int samsung_dsim_probe(struct platform_device *pdev)
 
 	pm_runtime_enable(dev);
 
-	dsi->bridge.funcs = &samsung_dsim_bridge_funcs;
 	dsi->bridge.of_node = dev->of_node;
 	dsi->bridge.type = DRM_MODE_CONNECTOR_DSI;
 
diff --git a/drivers/gpu/drm/bridge/sii902x.c b/drivers/gpu/drm/bridge/sii902x.c
index 914a2609a685..6de61d9fe064 100644
--- a/drivers/gpu/drm/bridge/sii902x.c
+++ b/drivers/gpu/drm/bridge/sii902x.c
@@ -416,6 +416,7 @@ out:
 }
 
 static int sii902x_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct sii902x *sii902x = bridge_to_sii902x(bridge);
@@ -424,7 +425,7 @@ static int sii902x_bridge_attach(struct drm_bridge *bridge,
 	int ret;
 
 	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)
-		return drm_bridge_attach(bridge->encoder, sii902x->next_bridge,
+		return drm_bridge_attach(encoder, sii902x->next_bridge,
 					 bridge, flags);
 
 	drm_connector_helper_add(&sii902x->connector,
@@ -452,7 +453,7 @@ static int sii902x_bridge_attach(struct drm_bridge *bridge,
 	if (ret)
 		return ret;
 
-	drm_connector_attach_encoder(&sii902x->connector, bridge->encoder);
+	drm_connector_attach_encoder(&sii902x->connector, encoder);
 
 	return 0;
 }
@@ -1138,6 +1139,7 @@ static int sii902x_init(struct sii902x *sii902x)
 	sii902x->bridge.of_node = dev->of_node;
 	sii902x->bridge.timings = &default_sii902x_timings;
 	sii902x->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID;
+	sii902x->bridge.type = DRM_MODE_CONNECTOR_HDMIA;
 
 	if (sii902x->i2c->irq > 0)
 		sii902x->bridge.ops |= DRM_BRIDGE_OP_HPD;
diff --git a/drivers/gpu/drm/bridge/sil-sii8620.c b/drivers/gpu/drm/bridge/sil-sii8620.c
index 28a2e1ee04b2..3af650dc92a1 100644
--- a/drivers/gpu/drm/bridge/sil-sii8620.c
+++ b/drivers/gpu/drm/bridge/sil-sii8620.c
@@ -2203,6 +2203,7 @@ static inline struct sii8620 *bridge_to_sii8620(struct drm_bridge *bridge)
 }
 
 static int sii8620_attach(struct drm_bridge *bridge,
+			  struct drm_encoder *encoder,
 			  enum drm_bridge_attach_flags flags)
 {
 	struct sii8620 *ctx = bridge_to_sii8620(bridge);
diff --git a/drivers/gpu/drm/bridge/simple-bridge.c b/drivers/gpu/drm/bridge/simple-bridge.c
index ab0b0e36e97a..70db5b99e5bb 100644
--- a/drivers/gpu/drm/bridge/simple-bridge.c
+++ b/drivers/gpu/drm/bridge/simple-bridge.c
@@ -103,12 +103,13 @@ static const struct drm_connector_funcs simple_bridge_con_funcs = {
 };
 
 static int simple_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct simple_bridge *sbridge = drm_bridge_to_simple_bridge(bridge);
 	int ret;
 
-	ret = drm_bridge_attach(bridge->encoder, sbridge->next_bridge, bridge,
+	ret = drm_bridge_attach(encoder, sbridge->next_bridge, bridge,
 				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret < 0)
 		return ret;
@@ -127,7 +128,7 @@ static int simple_bridge_attach(struct drm_bridge *bridge,
 		return ret;
 	}
 
-	drm_connector_attach_encoder(&sbridge->connector, bridge->encoder);
+	drm_connector_attach_encoder(&sbridge->connector, encoder);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c
index 6166f197e37b..5e5f8c2f95be 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c
@@ -1077,6 +1077,7 @@ struct dw_hdmi_qp *dw_hdmi_qp_bind(struct platform_device *pdev,
 	hdmi->bridge.ops = DRM_BRIDGE_OP_DETECT |
 			   DRM_BRIDGE_OP_EDID |
 			   DRM_BRIDGE_OP_HDMI |
+			   DRM_BRIDGE_OP_HDMI_AUDIO |
 			   DRM_BRIDGE_OP_HPD;
 	hdmi->bridge.of_node = pdev->dev.of_node;
 	hdmi->bridge.type = DRM_MODE_CONNECTOR_HDMIA;
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
index 0890add5f707..8791408dd1ff 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
@@ -22,8 +22,8 @@
 
 #include <media/cec-notifier.h>
 
-#include <uapi/linux/media-bus-format.h>
-#include <uapi/linux/videodev2.h>
+#include <linux/media-bus-format.h>
+#include <linux/videodev2.h>
 
 #include <drm/bridge/dw_hdmi.h>
 #include <drm/display/drm_hdmi_helper.h>
@@ -2889,12 +2889,13 @@ static int dw_hdmi_bridge_atomic_check(struct drm_bridge *bridge,
 }
 
 static int dw_hdmi_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct dw_hdmi *hdmi = bridge->driver_private;
 
 	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)
-		return drm_bridge_attach(bridge->encoder, hdmi->next_bridge,
+		return drm_bridge_attach(encoder, hdmi->next_bridge,
 					 bridge, flags);
 
 	return dw_hdmi_connector_create(hdmi);
@@ -3332,9 +3333,9 @@ struct dw_hdmi *dw_hdmi_probe(struct platform_device *pdev,
 	u8 config0;
 	u8 config3;
 
-	hdmi = devm_kzalloc(dev, sizeof(*hdmi), GFP_KERNEL);
-	if (!hdmi)
-		return ERR_PTR(-ENOMEM);
+	hdmi = devm_drm_bridge_alloc(dev, struct dw_hdmi, bridge, &dw_hdmi_bridge_funcs);
+	if (IS_ERR(hdmi))
+		return hdmi;
 
 	hdmi->plat_data = plat_data;
 	hdmi->dev = dev;
@@ -3494,7 +3495,6 @@ struct dw_hdmi *dw_hdmi_probe(struct platform_device *pdev,
 	}
 
 	hdmi->bridge.driver_private = hdmi;
-	hdmi->bridge.funcs = &dw_hdmi_bridge_funcs;
 	hdmi->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID
 			 | DRM_BRIDGE_OP_HPD;
 	hdmi->bridge.interlace_allowed = true;
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
index 2b6e70a49f43..b08ada920a50 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c
@@ -1072,15 +1072,16 @@ dw_mipi_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 }
 
 static int dw_mipi_dsi_bridge_attach(struct drm_bridge *bridge,
+				     struct drm_encoder *encoder,
 				     enum drm_bridge_attach_flags flags)
 {
 	struct dw_mipi_dsi *dsi = bridge_to_dsi(bridge);
 
 	/* Set the encoder type as caller does not know it */
-	bridge->encoder->encoder_type = DRM_MODE_ENCODER_DSI;
+	encoder->encoder_type = DRM_MODE_ENCODER_DSI;
 
 	/* Attach the panel-bridge to the dsi bridge */
-	return drm_bridge_attach(bridge->encoder, dsi->panel_bridge, bridge,
+	return drm_bridge_attach(encoder, dsi->panel_bridge, bridge,
 				 flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi2.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi2.c
index 5fd7a459efdd..c76f5f2e74d1 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi2.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi2.c
@@ -870,15 +870,16 @@ dw_mipi_dsi2_bridge_mode_valid(struct drm_bridge *bridge,
 }
 
 static int dw_mipi_dsi2_bridge_attach(struct drm_bridge *bridge,
+				      struct drm_encoder *encoder,
 				      enum drm_bridge_attach_flags flags)
 {
 	struct dw_mipi_dsi2 *dsi2 = bridge_to_dsi2(bridge);
 
 	/* Set the encoder type as caller does not know it */
-	bridge->encoder->encoder_type = DRM_MODE_ENCODER_DSI;
+	encoder->encoder_type = DRM_MODE_ENCODER_DSI;
 
 	/* Attach the panel-bridge to the dsi bridge */
-	return drm_bridge_attach(bridge->encoder, dsi2->panel_bridge, bridge,
+	return drm_bridge_attach(encoder, dsi2->panel_bridge, bridge,
 				 flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/tc358762.c b/drivers/gpu/drm/bridge/tc358762.c
index 49c76027f831..edf01476f2ef 100644
--- a/drivers/gpu/drm/bridge/tc358762.c
+++ b/drivers/gpu/drm/bridge/tc358762.c
@@ -202,11 +202,12 @@ static void tc358762_enable(struct drm_bridge *bridge,
 }
 
 static int tc358762_attach(struct drm_bridge *bridge,
+			   struct drm_encoder *encoder,
 			   enum drm_bridge_attach_flags flags)
 {
 	struct tc358762 *ctx = bridge_to_tc358762(bridge);
 
-	return drm_bridge_attach(bridge->encoder, ctx->panel_bridge,
+	return drm_bridge_attach(encoder, ctx->panel_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/bridge/tc358764.c b/drivers/gpu/drm/bridge/tc358764.c
index 3d3d135b4348..3f76c890fad9 100644
--- a/drivers/gpu/drm/bridge/tc358764.c
+++ b/drivers/gpu/drm/bridge/tc358764.c
@@ -295,11 +295,12 @@ static void tc358764_pre_enable(struct drm_bridge *bridge)
 }
 
 static int tc358764_attach(struct drm_bridge *bridge,
+			   struct drm_encoder *encoder,
 			   enum drm_bridge_attach_flags flags)
 {
 	struct tc358764 *ctx = bridge_to_tc358764(bridge);
 
-	return drm_bridge_attach(bridge->encoder, ctx->next_bridge, bridge, flags);
+	return drm_bridge_attach(encoder, ctx->next_bridge, bridge, flags);
 }
 
 static const struct drm_bridge_funcs tc358764_bridge_funcs = {
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 39e2d3a7a27d..7e5449fb86a3 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -1795,6 +1795,7 @@ static const struct drm_connector_funcs tc_connector_funcs = {
 };
 
 static int tc_dpi_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct tc_data *tc = bridge_to_tc(bridge);
@@ -1807,6 +1808,7 @@ static int tc_dpi_bridge_attach(struct drm_bridge *bridge,
 }
 
 static int tc_edp_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24;
diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index ec79b0dd0e2c..063f217a17b6 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -554,6 +554,7 @@ static const struct mipi_dsi_host_ops tc358768_dsi_host_ops = {
 };
 
 static int tc358768_bridge_attach(struct drm_bridge *bridge,
+				  struct drm_encoder *encoder,
 				  enum drm_bridge_attach_flags flags)
 {
 	struct tc358768_priv *priv = bridge_to_tc358768(bridge);
@@ -563,7 +564,7 @@ static int tc358768_bridge_attach(struct drm_bridge *bridge,
 		return -ENOTSUPP;
 	}
 
-	return drm_bridge_attach(bridge->encoder, priv->output.bridge, bridge,
+	return drm_bridge_attach(encoder, priv->output.bridge, bridge,
 				 flags);
 }
 
@@ -580,7 +581,8 @@ tc358768_bridge_mode_valid(struct drm_bridge *bridge,
 	return MODE_OK;
 }
 
-static void tc358768_bridge_disable(struct drm_bridge *bridge)
+static void tc358768_bridge_atomic_disable(struct drm_bridge *bridge,
+					   struct drm_atomic_state *state)
 {
 	struct tc358768_priv *priv = bridge_to_tc358768(bridge);
 	int ret;
@@ -602,7 +604,8 @@ static void tc358768_bridge_disable(struct drm_bridge *bridge)
 		dev_warn(priv->dev, "Software disable failed: %d\n", ret);
 }
 
-static void tc358768_bridge_post_disable(struct drm_bridge *bridge)
+static void tc358768_bridge_atomic_post_disable(struct drm_bridge *bridge,
+						struct drm_atomic_state *state)
 {
 	struct tc358768_priv *priv = bridge_to_tc358768(bridge);
 
@@ -682,13 +685,17 @@ static u32 tc358768_dsi_bytes_to_ns(struct tc358768_priv *priv, u32 val)
 	return (u32)div_u64(m, n);
 }
 
-static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
+static void tc358768_bridge_atomic_pre_enable(struct drm_bridge *bridge,
+					      struct drm_atomic_state *state)
 {
 	struct tc358768_priv *priv = bridge_to_tc358768(bridge);
 	struct mipi_dsi_device *dsi_dev = priv->output.dev;
 	unsigned long mode_flags = dsi_dev->mode_flags;
 	u32 val, val2, lptxcnt, hact, data_type;
 	s32 raw_val;
+	struct drm_crtc_state *crtc_state;
+	struct drm_connector_state *conn_state;
+	struct drm_connector *connector;
 	const struct drm_display_mode *mode;
 	u32 hsbyteclk_ps, dsiclk_ps, ui_ps;
 	u32 dsiclk, hsbyteclk;
@@ -719,7 +726,10 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 		return;
 	}
 
-	mode = &bridge->encoder->crtc->state->adjusted_mode;
+	connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	conn_state = drm_atomic_get_new_connector_state(state, connector);
+	crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc);
+	mode = &crtc_state->adjusted_mode;
 	ret = tc358768_setup_pll(priv, mode);
 	if (ret) {
 		dev_err(dev, "PLL setup failed: %d\n", ret);
@@ -1076,14 +1086,12 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	tc358768_write(priv, TC358768_DSI_CONFW, val);
 
 	ret = tc358768_clear_error(priv);
-	if (ret) {
+	if (ret)
 		dev_err(dev, "Bridge pre_enable failed: %d\n", ret);
-		tc358768_bridge_disable(bridge);
-		tc358768_bridge_post_disable(bridge);
-	}
 }
 
-static void tc358768_bridge_enable(struct drm_bridge *bridge)
+static void tc358768_bridge_atomic_enable(struct drm_bridge *bridge,
+					  struct drm_atomic_state *state)
 {
 	struct tc358768_priv *priv = bridge_to_tc358768(bridge);
 	int ret;
@@ -1100,11 +1108,8 @@ static void tc358768_bridge_enable(struct drm_bridge *bridge)
 	tc358768_update_bits(priv, TC358768_CONFCTL, BIT(6), BIT(6));
 
 	ret = tc358768_clear_error(priv);
-	if (ret) {
+	if (ret)
 		dev_err(priv->dev, "Bridge enable failed: %d\n", ret);
-		tc358768_bridge_disable(bridge);
-		tc358768_bridge_post_disable(bridge);
-	}
 }
 
 #define MAX_INPUT_SEL_FORMATS	1
@@ -1166,10 +1171,10 @@ static const struct drm_bridge_funcs tc358768_bridge_funcs = {
 	.attach = tc358768_bridge_attach,
 	.mode_valid = tc358768_bridge_mode_valid,
 	.mode_fixup = tc358768_mode_fixup,
-	.pre_enable = tc358768_bridge_pre_enable,
-	.enable = tc358768_bridge_enable,
-	.disable = tc358768_bridge_disable,
-	.post_disable = tc358768_bridge_post_disable,
+	.atomic_pre_enable = tc358768_bridge_atomic_pre_enable,
+	.atomic_enable = tc358768_bridge_atomic_enable,
+	.atomic_disable = tc358768_bridge_atomic_disable,
+	.atomic_post_disable = tc358768_bridge_atomic_post_disable,
 
 	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
diff --git a/drivers/gpu/drm/bridge/tc358775.c b/drivers/gpu/drm/bridge/tc358775.c
index c89757bec4e6..1b10e6ee1724 100644
--- a/drivers/gpu/drm/bridge/tc358775.c
+++ b/drivers/gpu/drm/bridge/tc358775.c
@@ -286,7 +286,8 @@ static inline struct tc_data *bridge_to_tc(struct drm_bridge *b)
 	return container_of(b, struct tc_data, bridge);
 }
 
-static void tc_bridge_pre_enable(struct drm_bridge *bridge)
+static void tc_bridge_atomic_pre_enable(struct drm_bridge *bridge,
+					struct drm_atomic_state *state)
 {
 	struct tc_data *tc = bridge_to_tc(bridge);
 	struct device *dev = &tc->dsi->dev;
@@ -309,7 +310,8 @@ static void tc_bridge_pre_enable(struct drm_bridge *bridge)
 	usleep_range(10, 20);
 }
 
-static void tc_bridge_post_disable(struct drm_bridge *bridge)
+static void tc_bridge_atomic_post_disable(struct drm_bridge *bridge,
+					  struct drm_atomic_state *state)
 {
 	struct tc_data *tc = bridge_to_tc(bridge);
 	struct device *dev = &tc->dsi->dev;
@@ -368,30 +370,21 @@ static void d2l_write(struct i2c_client *i2c, u16 addr, u32 val)
 			ret, addr);
 }
 
-/* helper function to access bus_formats */
-static struct drm_connector *get_connector(struct drm_encoder *encoder)
-{
-	struct drm_device *dev = encoder->dev;
-	struct drm_connector *connector;
-
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
-		if (connector->encoder == encoder)
-			return connector;
-
-	return NULL;
-}
-
-static void tc_bridge_enable(struct drm_bridge *bridge)
+static void tc_bridge_atomic_enable(struct drm_bridge *bridge,
+				    struct drm_atomic_state *state)
 {
 	struct tc_data *tc = bridge_to_tc(bridge);
 	u32 hback_porch, hsync_len, hfront_porch, hactive, htime1, htime2;
 	u32 vback_porch, vsync_len, vfront_porch, vactive, vtime1, vtime2;
 	u32 val = 0;
 	u16 dsiclk, clkdiv, byteclk, t1, t2, t3, vsdelay;
-	struct drm_display_mode *mode;
-	struct drm_connector *connector = get_connector(bridge->encoder);
-
-	mode = &bridge->encoder->crtc->state->adjusted_mode;
+	struct drm_connector *connector =
+		drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	struct drm_connector_state *conn_state =
+		drm_atomic_get_new_connector_state(state, connector);
+	struct drm_crtc_state *crtc_state =
+		drm_atomic_get_new_crtc_state(state, conn_state->crtc);
+	struct drm_display_mode *mode = &crtc_state->adjusted_mode;
 
 	hback_porch = mode->htotal - mode->hsync_end;
 	hsync_len  = mode->hsync_end - mode->hsync_start;
@@ -589,21 +582,25 @@ static int tc358775_parse_dt(struct device_node *np, struct tc_data *tc)
 }
 
 static int tc_bridge_attach(struct drm_bridge *bridge,
+			    struct drm_encoder *encoder,
 			    enum drm_bridge_attach_flags flags)
 {
 	struct tc_data *tc = bridge_to_tc(bridge);
 
 	/* Attach the panel-bridge to the dsi bridge */
-	return drm_bridge_attach(bridge->encoder, tc->panel_bridge,
+	return drm_bridge_attach(encoder, tc->panel_bridge,
 				 &tc->bridge, flags);
 }
 
 static const struct drm_bridge_funcs tc_bridge_funcs = {
 	.attach = tc_bridge_attach,
-	.pre_enable = tc_bridge_pre_enable,
-	.enable = tc_bridge_enable,
+	.atomic_pre_enable = tc_bridge_atomic_pre_enable,
+	.atomic_enable = tc_bridge_atomic_enable,
 	.mode_valid = tc_mode_valid,
-	.post_disable = tc_bridge_post_disable,
+	.atomic_post_disable = tc_bridge_atomic_post_disable,
+	.atomic_reset = drm_atomic_helper_bridge_reset,
+	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
 };
 
 static int tc_attach_host(struct tc_data *tc)
diff --git a/drivers/gpu/drm/bridge/tda998x_drv.c b/drivers/gpu/drm/bridge/tda998x_drv.c
index 20658258fb51..850909f78a7b 100644
--- a/drivers/gpu/drm/bridge/tda998x_drv.c
+++ b/drivers/gpu/drm/bridge/tda998x_drv.c
@@ -1365,6 +1365,7 @@ static int tda998x_connector_init(struct tda998x_priv *priv,
 /* DRM bridge functions */
 
 static int tda998x_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct tda998x_priv *priv = bridge_to_tda998x_priv(bridge);
@@ -1780,9 +1781,9 @@ static int tda998x_create(struct device *dev)
 	u32 video;
 	int rev_lo, rev_hi, ret;
 
-	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
+	priv = devm_drm_bridge_alloc(dev, struct tda998x_priv, bridge, &tda998x_bridge_funcs);
+	if (IS_ERR(priv))
+		return PTR_ERR(priv);
 
 	dev_set_drvdata(dev, priv);
 
@@ -1947,7 +1948,6 @@ static int tda998x_create(struct device *dev)
 			tda998x_audio_codec_init(priv, &client->dev);
 	}
 
-	priv->bridge.funcs = &tda998x_bridge_funcs;
 #ifdef CONFIG_OF
 	priv->bridge.of_node = dev->of_node;
 #endif
diff --git a/drivers/gpu/drm/bridge/thc63lvd1024.c b/drivers/gpu/drm/bridge/thc63lvd1024.c
index bba10cf9b4f9..e2fc78adebcf 100644
--- a/drivers/gpu/drm/bridge/thc63lvd1024.c
+++ b/drivers/gpu/drm/bridge/thc63lvd1024.c
@@ -43,11 +43,12 @@ static inline struct thc63_dev *to_thc63(struct drm_bridge *bridge)
 }
 
 static int thc63_attach(struct drm_bridge *bridge,
+			struct drm_encoder *encoder,
 			enum drm_bridge_attach_flags flags)
 {
 	struct thc63_dev *thc63 = to_thc63(bridge);
 
-	return drm_bridge_attach(bridge->encoder, thc63->next, bridge, flags);
+	return drm_bridge_attach(encoder, thc63->next, bridge, flags);
 }
 
 static enum drm_mode_status thc63_mode_valid(struct drm_bridge *bridge,
diff --git a/drivers/gpu/drm/bridge/ti-dlpc3433.c b/drivers/gpu/drm/bridge/ti-dlpc3433.c
index 85f2a0e74a1c..47638d1c96ec 100644
--- a/drivers/gpu/drm/bridge/ti-dlpc3433.c
+++ b/drivers/gpu/drm/bridge/ti-dlpc3433.c
@@ -242,12 +242,12 @@ static void dlpc_mode_set(struct drm_bridge *bridge,
 	drm_mode_copy(&dlpc->mode, adjusted_mode);
 }
 
-static int dlpc_attach(struct drm_bridge *bridge,
+static int dlpc_attach(struct drm_bridge *bridge, struct drm_encoder *encoder,
 		       enum drm_bridge_attach_flags flags)
 {
 	struct dlpc *dlpc = bridge_to_dlpc(bridge);
 
-	return drm_bridge_attach(bridge->encoder, dlpc->next_bridge, bridge, flags);
+	return drm_bridge_attach(encoder, dlpc->next_bridge, bridge, flags);
 }
 
 static const struct drm_bridge_funcs dlpc_bridge_funcs = {
diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
index 95563aa1b450..033c44326552 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
@@ -40,7 +40,7 @@
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
-#include <drm/drm_drv.h> /* DRM_MODESET_LOCK_ALL_BEGIN() needs drm_drv_uses_atomic_modeset() */
+#include <drm/drm_bridge_helper.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_of.h>
 #include <drm/drm_print.h>
@@ -290,11 +290,12 @@ static struct sn65dsi83 *bridge_to_sn65dsi83(struct drm_bridge *bridge)
 }
 
 static int sn65dsi83_attach(struct drm_bridge *bridge,
+			    struct drm_encoder *encoder,
 			    enum drm_bridge_attach_flags flags)
 {
 	struct sn65dsi83 *ctx = bridge_to_sn65dsi83(bridge);
 
-	return drm_bridge_attach(bridge->encoder, ctx->panel_bridge,
+	return drm_bridge_attach(encoder, ctx->panel_bridge,
 				 &ctx->bridge, flags);
 }
 
@@ -370,7 +371,6 @@ static u8 sn65dsi83_get_dsi_div(struct sn65dsi83 *ctx)
 
 static int sn65dsi83_reset_pipe(struct sn65dsi83 *sn65dsi83)
 {
-	struct drm_device *dev = sn65dsi83->bridge.dev;
 	struct drm_modeset_acquire_ctx ctx;
 	int err;
 
@@ -385,26 +385,21 @@ static int sn65dsi83_reset_pipe(struct sn65dsi83 *sn65dsi83)
 	 * Keep the lock during the whole operation to be atomic.
 	 */
 
-	DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx, 0, err);
-
-	if (!sn65dsi83->bridge.encoder->crtc) {
-		/*
-		 * No CRTC attached -> No CRTC active outputs to reset
-		 * This can happen when the SN65DSI83 is reset. Simply do
-		 * nothing without returning any errors.
-		 */
-		err = 0;
-		goto end;
-	}
+	drm_modeset_acquire_init(&ctx, 0);
 
 	dev_warn(sn65dsi83->dev, "reset the pipe\n");
 
-	err = drm_atomic_helper_reset_crtc(sn65dsi83->bridge.encoder->crtc, &ctx);
+retry:
+	err = drm_bridge_helper_reset_crtc(&sn65dsi83->bridge, &ctx);
+	if (err == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry;
+	}
 
-end:
-	DRM_MODESET_LOCK_ALL_END(dev, ctx, err);
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
 
-	return err;
+	return 0;
 }
 
 static void sn65dsi83_reset_work(struct work_struct *ws)
@@ -946,9 +941,9 @@ static int sn65dsi83_probe(struct i2c_client *client)
 	struct sn65dsi83 *ctx;
 	int ret;
 
-	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	ctx = devm_drm_bridge_alloc(dev, struct sn65dsi83, bridge, &sn65dsi83_funcs);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
 
 	ctx->dev = dev;
 	INIT_WORK(&ctx->reset_work, sn65dsi83_reset_work);
@@ -988,7 +983,6 @@ static int sn65dsi83_probe(struct i2c_client *client)
 	dev_set_drvdata(dev, ctx);
 	i2c_set_clientdata(client, ctx);
 
-	ctx->bridge.funcs = &sn65dsi83_funcs;
 	ctx->bridge.of_node = dev->of_node;
 	ctx->bridge.pre_enable_prev_first = true;
 	ctx->bridge.type = DRM_MODE_CONNECTOR_LVDS;
diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
index 01d456b955ab..60224f476e1d 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
@@ -35,6 +35,7 @@
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
+#define SN_DEVICE_ID_REGS			0x00	/* up to 0x07 */
 #define SN_DEVICE_REV_REG			0x08
 #define SN_DPPLL_SRC_REG			0x0A
 #define  DPPLL_CLK_SRC_DSICLK			BIT(0)
@@ -243,11 +244,26 @@ static void ti_sn65dsi86_write_u16(struct ti_sn65dsi86 *pdata,
 	regmap_bulk_write(pdata->regmap, reg, buf, ARRAY_SIZE(buf));
 }
 
-static u32 ti_sn_bridge_get_dsi_freq(struct ti_sn65dsi86 *pdata)
+static struct drm_display_mode *
+get_new_adjusted_display_mode(struct drm_bridge *bridge,
+			      struct drm_atomic_state *state)
+{
+	struct drm_connector *connector =
+		drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	struct drm_connector_state *conn_state =
+		drm_atomic_get_new_connector_state(state, connector);
+	struct drm_crtc_state *crtc_state =
+		drm_atomic_get_new_crtc_state(state, conn_state->crtc);
+
+	return &crtc_state->adjusted_mode;
+}
+
+static u32 ti_sn_bridge_get_dsi_freq(struct ti_sn65dsi86 *pdata,
+				     struct drm_atomic_state *state)
 {
 	u32 bit_rate_khz, clk_freq_khz;
 	struct drm_display_mode *mode =
-		&pdata->bridge.encoder->crtc->state->adjusted_mode;
+		get_new_adjusted_display_mode(&pdata->bridge, state);
 
 	bit_rate_khz = mode->clock *
 			mipi_dsi_pixel_format_to_bpp(pdata->dsi->format);
@@ -274,7 +290,8 @@ static const u32 ti_sn_bridge_dsiclk_lut[] = {
 	460800000,
 };
 
-static void ti_sn_bridge_set_refclk_freq(struct ti_sn65dsi86 *pdata)
+static void ti_sn_bridge_set_refclk_freq(struct ti_sn65dsi86 *pdata,
+					 struct drm_atomic_state *state)
 {
 	int i;
 	u32 refclk_rate;
@@ -287,7 +304,7 @@ static void ti_sn_bridge_set_refclk_freq(struct ti_sn65dsi86 *pdata)
 		refclk_lut_size = ARRAY_SIZE(ti_sn_bridge_refclk_lut);
 		clk_prepare_enable(pdata->refclk);
 	} else {
-		refclk_rate = ti_sn_bridge_get_dsi_freq(pdata) * 1000;
+		refclk_rate = ti_sn_bridge_get_dsi_freq(pdata, state) * 1000;
 		refclk_lut = ti_sn_bridge_dsiclk_lut;
 		refclk_lut_size = ARRAY_SIZE(ti_sn_bridge_dsiclk_lut);
 	}
@@ -311,12 +328,13 @@ static void ti_sn_bridge_set_refclk_freq(struct ti_sn65dsi86 *pdata)
 	pdata->pwm_refclk_freq = ti_sn_bridge_refclk_lut[i];
 }
 
-static void ti_sn65dsi86_enable_comms(struct ti_sn65dsi86 *pdata)
+static void ti_sn65dsi86_enable_comms(struct ti_sn65dsi86 *pdata,
+				      struct drm_atomic_state *state)
 {
 	mutex_lock(&pdata->comms_mutex);
 
 	/* configure bridge ref_clk */
-	ti_sn_bridge_set_refclk_freq(pdata);
+	ti_sn_bridge_set_refclk_freq(pdata, state);
 
 	/*
 	 * HPD on this bridge chip is a bit useless.  This is an eDP bridge
@@ -376,7 +394,7 @@ static int __maybe_unused ti_sn65dsi86_resume(struct device *dev)
 	 * clock so reading early doesn't work.
 	 */
 	if (pdata->refclk)
-		ti_sn65dsi86_enable_comms(pdata);
+		ti_sn65dsi86_enable_comms(pdata, NULL);
 
 	return ret;
 }
@@ -423,36 +441,8 @@ static int status_show(struct seq_file *s, void *data)
 
 	return 0;
 }
-
 DEFINE_SHOW_ATTRIBUTE(status);
 
-static void ti_sn65dsi86_debugfs_remove(void *data)
-{
-	debugfs_remove_recursive(data);
-}
-
-static void ti_sn65dsi86_debugfs_init(struct ti_sn65dsi86 *pdata)
-{
-	struct device *dev = pdata->dev;
-	struct dentry *debugfs;
-	int ret;
-
-	debugfs = debugfs_create_dir(dev_name(dev), NULL);
-
-	/*
-	 * We might get an error back if debugfs wasn't enabled in the kernel
-	 * so let's just silently return upon failure.
-	 */
-	if (IS_ERR_OR_NULL(debugfs))
-		return;
-
-	ret = devm_add_action_or_reset(dev, ti_sn65dsi86_debugfs_remove, debugfs);
-	if (ret)
-		return;
-
-	debugfs_create_file("status", 0600, debugfs, pdata, &status_fops);
-}
-
 /* -----------------------------------------------------------------------------
  * Auxiliary Devices (*not* AUX)
  */
@@ -732,6 +722,7 @@ static int ti_sn_attach_host(struct auxiliary_device *adev, struct ti_sn65dsi86
 }
 
 static int ti_sn_bridge_attach(struct drm_bridge *bridge,
+			       struct drm_encoder *encoder,
 			       enum drm_bridge_attach_flags flags)
 {
 	struct ti_sn65dsi86 *pdata = bridge_to_ti_sn65dsi86(bridge);
@@ -748,7 +739,7 @@ static int ti_sn_bridge_attach(struct drm_bridge *bridge,
 	 * Attach the next bridge.
 	 * We never want the next bridge to *also* create a connector.
 	 */
-	ret = drm_bridge_attach(bridge->encoder, pdata->next_bridge,
+	ret = drm_bridge_attach(encoder, pdata->next_bridge,
 				&pdata->bridge, flags | DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret < 0)
 		goto err_initted_aux;
@@ -821,12 +812,13 @@ static void ti_sn_bridge_atomic_disable(struct drm_bridge *bridge,
 	regmap_update_bits(pdata->regmap, SN_ENH_FRAME_REG, VSTREAM_ENABLE, 0);
 }
 
-static void ti_sn_bridge_set_dsi_rate(struct ti_sn65dsi86 *pdata)
+static void ti_sn_bridge_set_dsi_rate(struct ti_sn65dsi86 *pdata,
+				      struct drm_atomic_state *state)
 {
 	unsigned int bit_rate_mhz, clk_freq_mhz;
 	unsigned int val;
 	struct drm_display_mode *mode =
-		&pdata->bridge.encoder->crtc->state->adjusted_mode;
+		get_new_adjusted_display_mode(&pdata->bridge, state);
 
 	/* set DSIA clk frequency */
 	bit_rate_mhz = (mode->clock / 1000) *
@@ -856,12 +848,14 @@ static const unsigned int ti_sn_bridge_dp_rate_lut[] = {
 	0, 1620, 2160, 2430, 2700, 3240, 4320, 5400
 };
 
-static int ti_sn_bridge_calc_min_dp_rate_idx(struct ti_sn65dsi86 *pdata, unsigned int bpp)
+static int ti_sn_bridge_calc_min_dp_rate_idx(struct ti_sn65dsi86 *pdata,
+					     struct drm_atomic_state *state,
+					     unsigned int bpp)
 {
 	unsigned int bit_rate_khz, dp_rate_mhz;
 	unsigned int i;
 	struct drm_display_mode *mode =
-		&pdata->bridge.encoder->crtc->state->adjusted_mode;
+		get_new_adjusted_display_mode(&pdata->bridge, state);
 
 	/* Calculate minimum bit rate based on our pixel clock. */
 	bit_rate_khz = mode->clock * bpp;
@@ -960,10 +954,11 @@ static unsigned int ti_sn_bridge_read_valid_rates(struct ti_sn65dsi86 *pdata)
 	return valid_rates;
 }
 
-static void ti_sn_bridge_set_video_timings(struct ti_sn65dsi86 *pdata)
+static void ti_sn_bridge_set_video_timings(struct ti_sn65dsi86 *pdata,
+					   struct drm_atomic_state *state)
 {
 	struct drm_display_mode *mode =
-		&pdata->bridge.encoder->crtc->state->adjusted_mode;
+		get_new_adjusted_display_mode(&pdata->bridge, state);
 	u8 hsync_polarity = 0, vsync_polarity = 0;
 
 	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
@@ -1105,7 +1100,7 @@ static void ti_sn_bridge_atomic_enable(struct drm_bridge *bridge,
 			   pdata->ln_polrs << LN_POLRS_OFFSET);
 
 	/* set dsi clk frequency value */
-	ti_sn_bridge_set_dsi_rate(pdata);
+	ti_sn_bridge_set_dsi_rate(pdata, state);
 
 	/*
 	 * The SN65DSI86 only supports ASSR Display Authentication method and
@@ -1140,7 +1135,7 @@ static void ti_sn_bridge_atomic_enable(struct drm_bridge *bridge,
 	valid_rates = ti_sn_bridge_read_valid_rates(pdata);
 
 	/* Train until we run out of rates */
-	for (dp_rate_idx = ti_sn_bridge_calc_min_dp_rate_idx(pdata, bpp);
+	for (dp_rate_idx = ti_sn_bridge_calc_min_dp_rate_idx(pdata, state, bpp);
 	     dp_rate_idx < ARRAY_SIZE(ti_sn_bridge_dp_rate_lut);
 	     dp_rate_idx++) {
 		if (!(valid_rates & BIT(dp_rate_idx)))
@@ -1156,7 +1151,7 @@ static void ti_sn_bridge_atomic_enable(struct drm_bridge *bridge,
 	}
 
 	/* config video parameters */
-	ti_sn_bridge_set_video_timings(pdata);
+	ti_sn_bridge_set_video_timings(pdata, state);
 
 	/* enable video stream */
 	regmap_update_bits(pdata->regmap, SN_ENH_FRAME_REG, VSTREAM_ENABLE,
@@ -1171,7 +1166,7 @@ static void ti_sn_bridge_atomic_pre_enable(struct drm_bridge *bridge,
 	pm_runtime_get_sync(pdata->dev);
 
 	if (!pdata->refclk)
-		ti_sn65dsi86_enable_comms(pdata);
+		ti_sn65dsi86_enable_comms(pdata, state);
 
 	/* td7: min 100 us after enable before DSI data */
 	usleep_range(100, 110);
@@ -1216,6 +1211,15 @@ static const struct drm_edid *ti_sn_bridge_edid_read(struct drm_bridge *bridge,
 	return drm_edid_read_ddc(connector, &pdata->aux.ddc);
 }
 
+static void ti_sn65dsi86_debugfs_init(struct drm_bridge *bridge, struct dentry *root)
+{
+	struct ti_sn65dsi86 *pdata = bridge_to_ti_sn65dsi86(bridge);
+	struct dentry *debugfs;
+
+	debugfs = debugfs_create_dir(dev_name(pdata->dev), root);
+	debugfs_create_file("status", 0600, debugfs, pdata, &status_fops);
+}
+
 static const struct drm_bridge_funcs ti_sn_bridge_funcs = {
 	.attach = ti_sn_bridge_attach,
 	.detach = ti_sn_bridge_detach,
@@ -1229,6 +1233,7 @@ static const struct drm_bridge_funcs ti_sn_bridge_funcs = {
 	.atomic_reset = drm_atomic_helper_bridge_reset,
 	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
+	.debugfs_init = ti_sn65dsi86_debugfs_init,
 };
 
 static void ti_sn_bridge_parse_lanes(struct ti_sn65dsi86 *pdata,
@@ -1312,7 +1317,6 @@ static int ti_sn_bridge_probe(struct auxiliary_device *adev,
 	if (ret)
 		return ret;
 
-	pdata->bridge.funcs = &ti_sn_bridge_funcs;
 	pdata->bridge.of_node = np;
 	pdata->bridge.type = pdata->next_bridge->type == DRM_MODE_CONNECTOR_DisplayPort
 			   ? DRM_MODE_CONNECTOR_DisplayPort : DRM_MODE_CONNECTOR_eDP;
@@ -1894,6 +1898,7 @@ static int ti_sn65dsi86_probe(struct i2c_client *client)
 {
 	struct device *dev = &client->dev;
 	struct ti_sn65dsi86 *pdata;
+	u8 id_buf[8];
 	int ret;
 
 	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) {
@@ -1901,9 +1906,9 @@ static int ti_sn65dsi86_probe(struct i2c_client *client)
 		return -ENODEV;
 	}
 
-	pdata = devm_kzalloc(dev, sizeof(struct ti_sn65dsi86), GFP_KERNEL);
-	if (!pdata)
-		return -ENOMEM;
+	pdata = devm_drm_bridge_alloc(dev, struct ti_sn65dsi86, bridge, &ti_sn_bridge_funcs);
+	if (IS_ERR(pdata))
+		return PTR_ERR(pdata);
 	dev_set_drvdata(dev, pdata);
 	pdata->dev = dev;
 
@@ -1937,7 +1942,15 @@ static int ti_sn65dsi86_probe(struct i2c_client *client)
 	if (ret)
 		return ret;
 
-	ti_sn65dsi86_debugfs_init(pdata);
+	pm_runtime_get_sync(dev);
+	ret = regmap_bulk_read(pdata->regmap, SN_DEVICE_ID_REGS, id_buf, ARRAY_SIZE(id_buf));
+	pm_runtime_put_autosuspend(dev);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to read device id\n");
+
+	/* The ID string is stored backwards */
+	if (strncmp(id_buf, "68ISD   ", ARRAY_SIZE(id_buf)))
+		return dev_err_probe(dev, -EOPNOTSUPP, "unsupported device id\n");
 
 	/*
 	 * Break ourselves up into a collection of aux devices. The only real
diff --git a/drivers/gpu/drm/bridge/ti-tdp158.c b/drivers/gpu/drm/bridge/ti-tdp158.c
index 22316382451f..cca75443f012 100644
--- a/drivers/gpu/drm/bridge/ti-tdp158.c
+++ b/drivers/gpu/drm/bridge/ti-tdp158.c
@@ -45,11 +45,13 @@ static void tdp158_disable(struct drm_bridge *bridge,
 	regulator_disable(tdp158->vcc);
 }
 
-static int tdp158_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags)
+static int tdp158_attach(struct drm_bridge *bridge,
+			 struct drm_encoder *encoder,
+			 enum drm_bridge_attach_flags flags)
 {
 	struct tdp158 *tdp158 = bridge->driver_private;
 
-	return drm_bridge_attach(bridge->encoder, tdp158->next, bridge, flags);
+	return drm_bridge_attach(encoder, tdp158->next, bridge, flags);
 }
 
 static const struct drm_bridge_funcs tdp158_bridge_funcs = {
diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c
index 79ab5da827e1..e15d232ddbac 100644
--- a/drivers/gpu/drm/bridge/ti-tfp410.c
+++ b/drivers/gpu/drm/bridge/ti-tfp410.c
@@ -120,12 +120,13 @@ static void tfp410_hpd_callback(void *arg, enum drm_connector_status status)
 }
 
 static int tfp410_attach(struct drm_bridge *bridge,
+			 struct drm_encoder *encoder,
 			 enum drm_bridge_attach_flags flags)
 {
 	struct tfp410 *dvi = drm_bridge_to_tfp410(bridge);
 	int ret;
 
-	ret = drm_bridge_attach(bridge->encoder, dvi->next_bridge, bridge,
+	ret = drm_bridge_attach(encoder, dvi->next_bridge, bridge,
 				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 	if (ret < 0)
 		return ret;
@@ -159,7 +160,7 @@ static int tfp410_attach(struct drm_bridge *bridge,
 	drm_display_info_set_bus_formats(&dvi->connector.display_info,
 					 &dvi->bus_format, 1);
 
-	drm_connector_attach_encoder(&dvi->connector, bridge->encoder);
+	drm_connector_attach_encoder(&dvi->connector, encoder);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/bridge/ti-tpd12s015.c b/drivers/gpu/drm/bridge/ti-tpd12s015.c
index 47b74cb25b14..1c289051a598 100644
--- a/drivers/gpu/drm/bridge/ti-tpd12s015.c
+++ b/drivers/gpu/drm/bridge/ti-tpd12s015.c
@@ -38,6 +38,7 @@ static inline struct tpd12s015_device *to_tpd12s015(struct drm_bridge *bridge)
 }
 
 static int tpd12s015_attach(struct drm_bridge *bridge,
+			    struct drm_encoder *encoder,
 			    enum drm_bridge_attach_flags flags)
 {
 	struct tpd12s015_device *tpd = to_tpd12s015(bridge);
@@ -46,7 +47,7 @@ static int tpd12s015_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	ret = drm_bridge_attach(bridge->encoder, tpd->next_bridge,
+	ret = drm_bridge_attach(encoder, tpd->next_bridge,
 				bridge, flags);
 	if (ret < 0)
 		return ret;
diff --git a/drivers/gpu/drm/ci/arm64.config b/drivers/gpu/drm/ci/arm64.config
index a8fca079921b..fddfbd4d2493 100644
--- a/drivers/gpu/drm/ci/arm64.config
+++ b/drivers/gpu/drm/ci/arm64.config
@@ -193,6 +193,8 @@ CONFIG_PWM_MTK_DISP=y
 CONFIG_MTK_CMDQ=y
 CONFIG_REGULATOR_DA9211=y
 CONFIG_DRM_ANALOGIX_ANX7625=y
+CONFIG_PHY_MTK_HDMI=y
+CONFIG_PHY_MTK_MIPI_DSI=y
 
 # For nouveau.  Note that DRM must be a module so that it's loaded after NFS is up to provide the firmware.
 CONFIG_ARCH_TEGRA=y
diff --git a/drivers/gpu/drm/ci/build-igt.sh b/drivers/gpu/drm/ci/build-igt.sh
index eddb5f782a5e..caa2f4804ed5 100644
--- a/drivers/gpu/drm/ci/build-igt.sh
+++ b/drivers/gpu/drm/ci/build-igt.sh
@@ -71,4 +71,4 @@ tar -cf artifacts/igt.tar /igt
 # Pass needed files to the test stage
 S3_ARTIFACT_NAME="igt.tar.gz"
 gzip -c artifacts/igt.tar > ${S3_ARTIFACT_NAME}
-ci-fairy s3cp --token-file "${S3_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/${S3_ARTIFACT_NAME}
+s3_upload ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/
diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh
index 19fe01257ab9..6fb74c51abe2 100644
--- a/drivers/gpu/drm/ci/build.sh
+++ b/drivers/gpu/drm/ci/build.sh
@@ -98,14 +98,14 @@ done
 
 make ${KERNEL_IMAGE_NAME}
 
-mkdir -p /lava-files/
+mkdir -p /kernel/
 for image in ${KERNEL_IMAGE_NAME}; do
-    cp arch/${KERNEL_ARCH}/boot/${image} /lava-files/.
+    cp arch/${KERNEL_ARCH}/boot/${image} /kernel/.
 done
 
 if [[ -n ${DEVICE_TREES} ]]; then
     make dtbs
-    cp ${DEVICE_TREES} /lava-files/.
+    cp ${DEVICE_TREES} /kernel/.
 fi
 
 make modules
@@ -121,11 +121,11 @@ if [[ ${DEBIAN_ARCH} = "arm64" ]]; then
         -d arch/arm64/boot/Image.lzma \
         -C lzma\
         -b arch/arm64/boot/dts/qcom/sdm845-cheza-r3.dtb \
-        /lava-files/cheza-kernel
+        /kernel/cheza-kernel
     KERNEL_IMAGE_NAME+=" cheza-kernel"
 
     # Make a gzipped copy of the Image for db410c.
-    gzip -k /lava-files/Image
+    gzip -k /kernel/Image
     KERNEL_IMAGE_NAME+=" Image.gz"
 fi
 
@@ -139,7 +139,7 @@ cp -rfv drivers/gpu/drm/ci/* install/.
 . .gitlab-ci/container/container_post_build.sh
 
 if [[ "$UPLOAD_TO_MINIO" = "1" ]]; then
-    xz -7 -c -T${FDO_CI_CONCURRENT:-4} vmlinux > /lava-files/vmlinux.xz
+    xz -7 -c -T${FDO_CI_CONCURRENT:-4} vmlinux > /kernel/vmlinux.xz
     FILES_TO_UPLOAD="$KERNEL_IMAGE_NAME vmlinux.xz"
 
     if [[ -n $DEVICE_TREES ]]; then
@@ -148,13 +148,13 @@ if [[ "$UPLOAD_TO_MINIO" = "1" ]]; then
 
     ls -l "${S3_JWT_FILE}"
     for f in $FILES_TO_UPLOAD; do
-        ci-fairy s3cp --token-file "${S3_JWT_FILE}" /lava-files/$f \
-                https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/$f
+        s3_upload /kernel/$f \
+                https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/
     done
 
     S3_ARTIFACT_NAME="kernel-files.tar.zst"
     tar --zstd -cf $S3_ARTIFACT_NAME install
-    ci-fairy s3cp --token-file "${S3_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/${S3_ARTIFACT_NAME}
+    s3_upload ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/
 
     echo "Download vmlinux.xz from https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/vmlinux.xz"
 fi
@@ -165,7 +165,7 @@ ln -s common artifacts/install/ci-common
 cp .config artifacts/${CI_JOB_NAME}_config
 
 for image in ${KERNEL_IMAGE_NAME}; do
-    cp /lava-files/$image artifacts/install/.
+    cp /kernel/$image artifacts/install/.
 done
 
 tar -C artifacts -cf artifacts/install.tar install
diff --git a/drivers/gpu/drm/ci/build.yml b/drivers/gpu/drm/ci/build.yml
index 274f118533a7..8eb56ebcf4aa 100644
--- a/drivers/gpu/drm/ci/build.yml
+++ b/drivers/gpu/drm/ci/build.yml
@@ -67,7 +67,7 @@ testing:arm32:
     #
     # db410c and db820c don't boot with KASAN_INLINE, probably due to the kernel
     # becoming too big for their bootloaders.
-    ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT"
+    ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT DEBUG_WW_MUTEX_SLOWPATH"
     UPLOAD_TO_MINIO: 1
     MERGE_FRAGMENT: arm.config
 
@@ -79,7 +79,7 @@ testing:arm64:
     #
     # db410c and db820c don't boot with KASAN_INLINE, probably due to the kernel
     # becoming too big for their bootloaders.
-    ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT"
+    ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT DEBUG_WW_MUTEX_SLOWPATH"
     UPLOAD_TO_MINIO: 1
     MERGE_FRAGMENT: arm64.config
 
@@ -91,7 +91,7 @@ testing:x86_64:
     #
     # db410c and db820c don't boot with KASAN_INLINE, probably due to the kernel
     # becoming too big for their bootloaders.
-    ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT"
+    ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT DEBUG_WW_MUTEX_SLOWPATH"
     UPLOAD_TO_MINIO: 1
     MERGE_FRAGMENT: x86_64.config
 
@@ -143,6 +143,10 @@ debian-arm64-release:
   rules:
     - when: never
 
+debian-arm64-ubsan:
+  rules:
+    - when: never
+
 debian-build-testing:
   rules:
     - when: never
@@ -183,6 +187,10 @@ debian-testing-msan:
   rules:
     - when: never
 
+debian-testing-ubsan:
+  rules:
+    - when: never
+
 debian-vulkan:
   rules:
     - when: never
diff --git a/drivers/gpu/drm/ci/container.yml b/drivers/gpu/drm/ci/container.yml
index 07dc13ff865d..56c95c2f91ae 100644
--- a/drivers/gpu/drm/ci/container.yml
+++ b/drivers/gpu/drm/ci/container.yml
@@ -24,6 +24,18 @@ alpine/x86_64_build:
   rules:
     - when: never
 
+debian/arm32_test-base:
+  rules:
+    - when: never
+
+debian/arm32_test-gl:
+  rules:
+    - when: never
+
+debian/arm32_test-vk:
+  rules:
+    - when: never
+
 debian/arm64_test-gl:
   rules:
     - when: never
@@ -32,6 +44,10 @@ debian/arm64_test-vk:
   rules:
     - when: never
 
+debian/baremetal_arm32_test:
+  rules:
+    - when: never
+
 debian/ppc64el_build:
   rules:
     - when: never
@@ -40,6 +56,14 @@ debian/s390x_build:
   rules:
     - when: never
 
+debian/x86_32_build:
+  rules:
+    - when: never
+
+debian/x86_64_test-android:
+  rules:
+    - when: never
+
 debian/x86_64_test-vk:
   rules:
     - when: never
diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml
index f04aabe8327c..ba75b3a7eca4 100644
--- a/drivers/gpu/drm/ci/gitlab-ci.yml
+++ b/drivers/gpu/drm/ci/gitlab-ci.yml
@@ -1,11 +1,11 @@
 variables:
   DRM_CI_PROJECT_PATH: &drm-ci-project-path mesa/mesa
-  DRM_CI_COMMIT_SHA: &drm-ci-commit-sha 7d3062470f3ccc6cb40540e772e902c7e2248024
+  DRM_CI_COMMIT_SHA: &drm-ci-commit-sha f73132f1215a37ce8ffc711a0136c90649aaf128
 
   UPSTREAM_REPO: https://gitlab.freedesktop.org/drm/kernel.git
   TARGET_BRANCH: drm-next
 
-  IGT_VERSION: 33adea9ebafd059ac88a5ccfec60536394f36c7c
+  IGT_VERSION: 04bedb9238586b81d4d4ca62b02e584f6cfc77af
 
   DEQP_RUNNER_GIT_URL: https://gitlab.freedesktop.org/mesa/deqp-runner.git
   DEQP_RUNNER_GIT_TAG: v0.20.0
@@ -20,8 +20,10 @@ variables:
           rm download-git-cache.sh
           set +o xtrace
   S3_JWT_FILE: /s3_jwt
+  S3_JWT_HEADER_FILE: /s3_jwt_header
   S3_JWT_FILE_SCRIPT: |-
       echo -n '${S3_JWT}' > '${S3_JWT_FILE}' &&
+      echo -n "Authorization: Bearer ${S3_JWT}" > '${S3_JWT_HEADER_FILE}' &&
       unset CI_JOB_JWT S3_JWT  # Unsetting vulnerable env variables
   S3_HOST: s3.freedesktop.org
   # This bucket is used to fetch the kernel image
@@ -143,11 +145,11 @@ stages:
     # Pre-merge pipeline
     - if: &is-pre-merge $CI_PIPELINE_SOURCE == "merge_request_event"
     # Push to a branch on a fork
-    - if: &is-fork-push $CI_PROJECT_NAMESPACE != "mesa" && $CI_PIPELINE_SOURCE == "push"
+    - if: &is-fork-push $CI_PIPELINE_SOURCE == "push"
     # nightly pipeline
     - if: &is-scheduled-pipeline $CI_PIPELINE_SOURCE == "schedule"
     # pipeline for direct pushes that bypassed the CI
-    - if: &is-direct-push $CI_PROJECT_NAMESPACE == "mesa" && $CI_PIPELINE_SOURCE == "push" && $GITLAB_USER_LOGIN != "marge-bot"
+    - if: &is-direct-push $CI_PIPELINE_SOURCE == "push" && $GITLAB_USER_LOGIN != "marge-bot"
 
 
 # Rules applied to every job in the pipeline
@@ -170,29 +172,48 @@ stages:
     - !reference [.disable-farm-mr-rules, rules]
     # Never run immediately after merging, as we just ran everything
     - !reference [.never-post-merge-rules, rules]
-    # Build everything in merge pipelines, if any files affecting the pipeline
-    # were changed
+    # Build everything in merge pipelines
     - if: *is-merge-attempt
-      changes: &all_paths
-      - drivers/gpu/drm/ci/**/*
       when: on_success
     # Same as above, but for pre-merge pipelines
     - if: *is-pre-merge
-      changes:
-        *all_paths
       when: manual
-    # Skip everything for pre-merge and merge pipelines which don't change
-    # anything in the build
+    # Build everything after someone bypassed the CI
+    - if: *is-direct-push
+      when: manual
+    # Build everything in scheduled pipelines
+    - if: *is-scheduled-pipeline
+      when: on_success
+    # Allow building everything in fork pipelines, but build nothing unless
+    # manually triggered
+    - when: manual
+
+
+# Repeat of the above but with `when: on_success` replaced with
+# `when: delayed` + `start_in:`, for build-only jobs.
+# Note: make sure the branches in this list are the same as in
+# `.container+build-rules` above.
+.build-only-delayed-rules:
+  rules:
+    - !reference [.common-rules, rules]
+    # Run when re-enabling a disabled farm, but not when disabling it
+    - !reference [.disable-farm-mr-rules, rules]
+    # Never run immediately after merging, as we just ran everything
+    - !reference [.never-post-merge-rules, rules]
+    # Build everything in merge pipelines
     - if: *is-merge-attempt
-      when: never
+      when: delayed
+      start_in: &build-delay 5 minutes
+    # Same as above, but for pre-merge pipelines
     - if: *is-pre-merge
-      when: never
+      when: manual
     # Build everything after someone bypassed the CI
     - if: *is-direct-push
-      when: on_success
+      when: manual
     # Build everything in scheduled pipelines
     - if: *is-scheduled-pipeline
-      when: on_success
+      when: delayed
+      start_in: *build-delay
     # Allow building everything in fork pipelines, but build nothing unless
     # manually triggered
     - when: manual
@@ -232,7 +253,7 @@ make git archive:
     - tar -cvzf ../$CI_PROJECT_NAME.tar.gz .
 
     # Use id_tokens for JWT auth
-    - ci-fairy s3cp --token-file "${S3_JWT_FILE}" ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/${S3_GITCACHE_BUCKET}/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/$CI_PROJECT_NAME.tar.gz
+    - s3_upload ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/${S3_GITCACHE_BUCKET}/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/
 
 
 # Sanity checks of MR settings and commit logs
diff --git a/drivers/gpu/drm/ci/igt_runner.sh b/drivers/gpu/drm/ci/igt_runner.sh
index 68b042e43b7f..2a0599f12c58 100755
--- a/drivers/gpu/drm/ci/igt_runner.sh
+++ b/drivers/gpu/drm/ci/igt_runner.sh
@@ -85,5 +85,16 @@ deqp-runner junit \
    --limit 50 \
    --template "See $ARTIFACTS_BASE_URL/results/{{testcase}}.xml"
 
+# Check if /proc/lockdep_stats exists
+if [ -f /proc/lockdep_stats ]; then
+    # If debug_locks is 0, it indicates lockdep is detected and it turns itself off.
+    debug_locks=$(grep 'debug_locks:' /proc/lockdep_stats | awk '{print $2}')
+    if [ "$debug_locks" -eq 0 ] && [ "$ret" -eq 0 ]; then
+        echo "Warning: LOCKDEP issue detected. Please check dmesg logs for more information."
+        cat /proc/lockdep_stats
+        ret=101
+    fi
+fi
+
 cd $oldpath
 exit $ret
diff --git a/drivers/gpu/drm/ci/image-tags.yml b/drivers/gpu/drm/ci/image-tags.yml
index 20049f3626b2..53fe34b86578 100644
--- a/drivers/gpu/drm/ci/image-tags.yml
+++ b/drivers/gpu/drm/ci/image-tags.yml
@@ -1,5 +1,5 @@
 variables:
-   CONTAINER_TAG: "20250204-mesa-uprev"
+   CONTAINER_TAG: "20250328-mesa-uprev"
    DEBIAN_X86_64_BUILD_BASE_IMAGE: "debian/x86_64_build-base"
    DEBIAN_BASE_TAG: "${CONTAINER_TAG}"
 
@@ -20,3 +20,5 @@ variables:
    DEBIAN_PYUTILS_TAG: "${CONTAINER_TAG}"
 
    ALPINE_X86_64_LAVA_SSH_TAG: "${CONTAINER_TAG}"
+
+   CONDITIONAL_BUILD_ANGLE_TAG: fec96cc945650c5fe9f7188cabe80d8a
diff --git a/drivers/gpu/drm/ci/lava-submit.sh b/drivers/gpu/drm/ci/lava-submit.sh
index 6e5ac51e8c0a..a1e8b34fb2d4 100755
--- a/drivers/gpu/drm/ci/lava-submit.sh
+++ b/drivers/gpu/drm/ci/lava-submit.sh
@@ -48,12 +48,13 @@ ROOTFS_URL="$(get_path_to_artifact lava-rootfs.tar.zst)"
 rm -rf results
 mkdir -p results/job-rootfs-overlay/
 
-artifacts/ci-common/generate-env.sh > results/job-rootfs-overlay/set-job-env-vars.sh
+artifacts/ci-common/export-gitlab-job-env-for-dut.sh \
+    > results/job-rootfs-overlay/set-job-env-vars.sh
 cp artifacts/ci-common/init-*.sh results/job-rootfs-overlay/
 cp "$SCRIPTS_DIR"/setup-test-env.sh results/job-rootfs-overlay/
 
 tar zcf job-rootfs-overlay.tar.gz -C results/job-rootfs-overlay/ .
-ci-fairy s3cp --token-file "${S3_JWT_FILE}" job-rootfs-overlay.tar.gz "https://${JOB_ROOTFS_OVERLAY_PATH}"
+s3_upload job-rootfs-overlay.tar.gz "https://${JOB_ARTIFACTS_BASE}"
 
 # Prepare env vars for upload.
 section_switch variables "Environment variables passed through to device:"
diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml
index 6a1e059858e5..84a25f0e783b 100644
--- a/drivers/gpu/drm/ci/test.yml
+++ b/drivers/gpu/drm/ci/test.yml
@@ -1,6 +1,14 @@
+.allow_failure_lockdep:
+  variables:
+    FF_USE_NEW_BASH_EVAL_STRATEGY: 'true'
+  allow_failure:
+    exit_codes:
+      - 101
+
 .lava-test:
   extends:
     - .container+build-rules
+    - .allow_failure_lockdep
   timeout: "1h30m"
   rules:
     - !reference [.scheduled_pipeline-rules, rules]
@@ -69,6 +77,7 @@
   extends:
     - .baremetal-test-arm64
     - .use-debian/baremetal_arm64_test
+    - .allow_failure_lockdep
   timeout: "1h30m"
   rules:
     - !reference [.scheduled_pipeline-rules, rules]
@@ -89,6 +98,28 @@
   tags:
     - $RUNNER_TAG
 
+.software-driver:
+  stage: software-driver
+  extends:
+    - .allow_failure_lockdep
+  timeout: "1h30m"
+  rules:
+    - !reference [.scheduled_pipeline-rules, rules]
+    - when: on_success
+  extends:
+    - .test-gl
+  tags:
+    - kvm
+  script:
+    - ln -sf $CI_PROJECT_DIR/install /install
+    - mv install/bzImage /kernel/bzImage
+    - mkdir -p /lib/modules
+    - install/crosvm-runner.sh install/igt_runner.sh
+  needs:
+    - debian/x86_64_test-gl
+    - testing:x86_64
+    - igt:x86_64
+
 .msm-sc7180:
   extends:
     - .lava-igt:arm64
@@ -133,7 +164,7 @@ msm:apq8016:
     BM_KERNEL_EXTRA_ARGS: clk_ignore_unused
     RUNNER_TAG: google-freedreno-db410c
   script:
-    - ./install/bare-metal/fastboot.sh
+    - ./install/bare-metal/fastboot.sh || exit $?
 
 msm:apq8096:
   extends:
@@ -147,7 +178,7 @@ msm:apq8096:
     GPU_VERSION: apq8096
     RUNNER_TAG: google-freedreno-db820c
   script:
-    - ./install/bare-metal/fastboot.sh
+    - ./install/bare-metal/fastboot.sh || exit $?
 
 msm:sdm845:
   extends:
@@ -161,7 +192,7 @@ msm:sdm845:
     GPU_VERSION: sdm845
     RUNNER_TAG: google-freedreno-cheza
   script:
-    - ./install/bare-metal/cros-servo.sh
+    - ./install/bare-metal/cros-servo.sh || exit $?
 
 msm:sm8350-hdk:
   extends:
@@ -440,47 +471,16 @@ panfrost:g12b:
     - .panfrost-gpu
 
 virtio_gpu:none:
-  stage: software-driver
-  timeout: "1h30m"
-  rules:
-    - !reference [.scheduled_pipeline-rules, rules]
-    - when: on_success
+  extends:
+    - .software-driver
   variables:
     CROSVM_GALLIUM_DRIVER: llvmpipe
     DRIVER_NAME: virtio_gpu
     GPU_VERSION: none
-  extends:
-    - .test-gl
-  tags:
-    - kvm
-  script:
-    - ln -sf $CI_PROJECT_DIR/install /install
-    - mv install/bzImage /lava-files/bzImage
-    - install/crosvm-runner.sh install/igt_runner.sh
-  needs:
-    - debian/x86_64_test-gl
-    - testing:x86_64
-    - igt:x86_64
 
 vkms:none:
-  stage: software-driver
-  timeout: "1h30m"
-  rules:
-    - !reference [.scheduled_pipeline-rules, rules]
-    - when: on_success
+  extends:
+    - .software-driver
   variables:
     DRIVER_NAME: vkms
     GPU_VERSION: none
-  extends:
-    - .test-gl
-  tags:
-    - kvm
-  script:
-    - ln -sf $CI_PROJECT_DIR/install /install
-    - mv install/bzImage /lava-files/bzImage
-    - mkdir -p /lib/modules
-    - ./install/crosvm-runner.sh ./install/igt_runner.sh
-  needs:
-    - debian/x86_64_test-gl
-    - testing:x86_64
-    - igt:x86_64
diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt
index 75374085f40f..f44dbce3151a 100644
--- a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt
@@ -14,16 +14,10 @@ amdgpu/amd_plane@mpo-scale-nv12,Fail
 amdgpu/amd_plane@mpo-scale-p010,Fail
 amdgpu/amd_plane@mpo-scale-rgb,Crash
 amdgpu/amd_plane@mpo-swizzle-toggle,Fail
-amdgpu/amd_uvd_dec@amdgpu_uvd_decode,Crash
+amdgpu/amd_uvd_dec@amdgpu_uvd_decode,Fail
 kms_addfb_basic@bad-pitch-65536,Fail
 kms_addfb_basic@bo-too-small,Fail
 kms_addfb_basic@too-high,Fail
-kms_async_flips@alternate-sync-async-flip,Fail
-kms_async_flips@alternate-sync-async-flip-atomic,Fail
-kms_async_flips@test-cursor,Fail
-kms_async_flips@test-cursor-atomic,Fail
-kms_async_flips@test-time-stamp,Fail
-kms_async_flips@test-time-stamp-atomic,Fail
 kms_atomic_transition@plane-all-modeset-transition-internal-panels,Fail
 kms_atomic_transition@plane-all-transition,Fail
 kms_atomic_transition@plane-all-transition-nonblocking,Fail
diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt
index 3879c4812a22..902d54027506 100644
--- a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt
@@ -14,6 +14,7 @@ gem_.*
 i915_.*
 xe_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt b/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt
index a29cea4f234c..8e2b5504004e 100644
--- a/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt
@@ -1,22 +1,18 @@
-core_setmaster@master-drop-set-shared-fd,Fail
+core_setmaster_vs_auth,Fail
 i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
 i915_module_load@resize-bar,Fail
 i915_pm_rpm@gem-execbuf-stress,Timeout
 i915_pm_rpm@module-reload,Fail
-kms_async_flips@test-time-stamp,Timeout
-kms_ccs@crc-sprite-planes-basic-y-tiled-ccs,Timeout
-kms_flip@dpms-off-confusion-interruptible,Timeout
-kms_flip@wf_vblank-ts-check-interruptible,Fail
+kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc,Timeout
+kms_fb_coherency@memset-crc,Crash
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail
-kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail
-kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail
@@ -31,12 +27,18 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail
+kms_frontbuffer_tracking@fbc-rgb101010-draw-mmap-cpu,Timeout
 kms_lease@lease-uevent,Fail
-kms_lease@page-flip-implicit-plane,Timeout
 kms_plane_alpha_blend@alpha-basic,Fail
 kms_plane_alpha_blend@alpha-opaque-fb,Fail
 kms_plane_alpha_blend@alpha-transparent-fb,Fail
 kms_plane_alpha_blend@constant-alpha-max,Fail
+kms_plane_scaling@planes-upscale-factor-0-25,Timeout
+kms_pm_backlight@brightness-with-dpms,Crash
+kms_pm_backlight@fade,Crash
+kms_prop_blob@invalid-set-prop-any,Fail
+kms_properties@connector-properties-legacy,Timeout
+kms_universal_plane@disable-primary-vs-flip,Timeout
 perf@i915-ref-count,Fail
 perf_pmu@module-unload,Fail
 perf_pmu@rc6,Crash
@@ -44,8 +46,3 @@ sysfs_heartbeat_interval@long,Timeout
 sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt b/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt
index 2ef1dc35a7fa..922327632eff 100644
--- a/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt
@@ -11,6 +11,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt
index ee11999e3da1..7353ab11e940 100644
--- a/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt
@@ -1,4 +1,3 @@
-core_setmaster@master-drop-set-user,Fail
 i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
@@ -16,6 +15,7 @@ kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail
+kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail
@@ -30,7 +30,6 @@ kms_plane_alpha_blend@alpha-opaque-fb,Fail
 kms_plane_alpha_blend@alpha-transparent-fb,Fail
 kms_plane_alpha_blend@constant-alpha-max,Fail
 kms_pm_backlight@basic-brightness,Fail
-kms_pm_backlight@brightness-with-dpms,Crash
 kms_pm_backlight@fade,Fail
 kms_pm_backlight@fade-with-dpms,Fail
 kms_pm_rpm@modeset-stress-extra-wait,Timeout
@@ -43,8 +42,3 @@ sysfs_heartbeat_interval@long,Timeout
 sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt
index 4f50e0240ff4..80bf2741866c 100644
--- a/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt
@@ -13,6 +13,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt
index 47b3f1d42bb6..6fef7c1e56ea 100644
--- a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt
@@ -1,4 +1,4 @@
-core_setmaster@master-drop-set-shared-fd,Fail
+core_setmaster_vs_auth,Fail
 i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
@@ -8,8 +8,9 @@ i915_pipe_stress@stress-xrgb8888-ytiled,Fail
 i915_pm_rpm@gem-execbuf-stress,Timeout
 i915_pm_rpm@module-reload,Fail
 i915_pm_rpm@system-suspend-execbuf,Timeout
-kms_async_flips@test-time-stamp,Timeout
-kms_ccs@crc-sprite-planes-basic-y-tiled-ccs,Timeout
+kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc,Timeout
+kms_cursor_crc@cursor-suspend,Timeout
+kms_fb_coherency@memset-crc,Crash
 kms_flip@busy-flip,Timeout
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail
@@ -34,17 +35,22 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail
 kms_lease@lease-uevent,Fail
-kms_lease@page-flip-implicit-plane,Timeout
+kms_pipe_stress@stress-xrgb8888-untiled,Fail
+kms_pipe_stress@stress-xrgb8888-ytiled,Fail
 kms_plane_alpha_blend@alpha-basic,Fail
 kms_plane_alpha_blend@alpha-opaque-fb,Fail
 kms_plane_alpha_blend@alpha-transparent-fb,Fail
 kms_plane_alpha_blend@constant-alpha-max,Fail
+kms_plane_scaling@planes-upscale-factor-0-25,Timeout
+kms_pm_backlight@brightness-with-dpms,Crash
+kms_pm_backlight@fade,Crash
+kms_prop_blob@invalid-set-prop-any,Fail
+kms_properties@connector-properties-legacy,Timeout
 kms_psr2_sf@cursor-plane-update-sf,Fail
 kms_psr2_sf@overlay-plane-update-continuous-sf,Fail
 kms_psr2_sf@overlay-plane-update-sf-dmg-area,Fail
 kms_psr2_sf@overlay-primary-update-sf-dmg-area,Fail
 kms_psr2_sf@plane-move-sf-dmg-area,Fail
-kms_psr2_sf@pr-cursor-plane-update-sf,Timeout
 kms_psr2_sf@primary-plane-update-sf-dmg-area,Fail
 kms_psr2_sf@primary-plane-update-sf-dmg-area-big-fb,Fail
 kms_psr2_sf@psr2-cursor-plane-update-sf,Fail
@@ -57,6 +63,7 @@ kms_psr2_sf@psr2-primary-plane-update-sf-dmg-area-big-fb,Fail
 kms_psr2_su@page_flip-NV12,Fail
 kms_psr2_su@page_flip-P010,Fail
 kms_setmode@basic,Fail
+kms_universal_plane@disable-primary-vs-flip,Timeout
 perf@i915-ref-count,Fail
 perf_pmu@module-unload,Fail
 perf_pmu@rc6,Crash
@@ -65,6 +72,3 @@ sysfs_heartbeat_interval@long,Timeout
 sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt b/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt
index c87ff8b40e99..c393a138b8a6 100644
--- a/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt
@@ -9,6 +9,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
@@ -16,7 +17,6 @@ gem_.*
 # Hangs the machine and timeout occurs
 i915_pm_rc6_residency.*
 i915_suspend.*
-xe_module_load.*
 api_intel_allocator.*
 kms_cursor_legacy.*
 i915_pm_rpm.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt b/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt
index 843c363b42f5..8adf5f0a6e80 100644
--- a/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt
@@ -1,39 +1,47 @@
-core_setmaster@master-drop-set-shared-fd,Fail
-core_setmaster@master-drop-set-user,Fail
+core_setmaster_vs_auth,Fail
 gen9_exec_parse@unaligned-access,Fail
 i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
 i915_module_load@resize-bar,Fail
-kms_dirtyfb@default-dirtyfb-ioctl,Fail
 kms_dirtyfb@drrs-dirtyfb-ioctl,Fail
-kms_dirtyfb@fbc-dirtyfb-ioctl,Fail
+kms_flip@blocking-wf_vblank,Fail
+kms_flip@wf_vblank-ts-check,Fail
 kms_flip@wf_vblank-ts-check-interruptible,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail
-kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail
+kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail
+kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail
-kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail
+kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail
+kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail
+kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail
+kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail
-kms_frontbuffer_tracking@fbcdrrs-tiling-linear,Fail
+kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail
+kms_frontbuffer_tracking@fbc-rgb101010-draw-mmap-cpu,Timeout
+kms_frontbuffer_tracking@fbc-tiling-linear,Fail
 kms_lease@lease-uevent,Fail
 kms_plane_alpha_blend@alpha-opaque-fb,Fail
+kms_plane_scaling@planes-upscale-factor-0-25,Timeout
+kms_pm_backlight@brightness-with-dpms,Crash
+kms_pm_backlight@fade,Crash
+kms_prop_blob@invalid-set-prop-any,Fail
+kms_properties@connector-properties-legacy,Timeout
+kms_rotation_crc@multiplane-rotation,Fail
 kms_rotation_crc@multiplane-rotation-cropping-top,Fail
+kms_universal_plane@disable-primary-vs-flip,Timeout
 perf@non-zero-reason,Timeout
 sysfs_heartbeat_interval@long,Timeout
+sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt b/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt
index 219ae839323a..2e4ef9f35654 100644
--- a/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt
@@ -12,6 +12,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt
index 0e08fff741aa..57453e340040 100644
--- a/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt
@@ -3,12 +3,13 @@ i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
 i915_module_load@resize-bar,Fail
+i915_pm_rpm@gem-execbuf-stress,Timeout
 kms_flip@dpms-off-confusion,Fail
+kms_flip@nonexisting-fb,Fail
 kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset,Fail
-kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset-interruptible,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail
-kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail
+kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,UnexpectedImprovement(Skip)
 kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail
@@ -28,7 +29,6 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail
-kms_frontbuffer_tracking@fbc-rgb565-draw-blt,Timeout
 kms_lease@lease-uevent,Fail
 kms_pm_rpm@modeset-stress-extra-wait,Timeout
 kms_rotation_crc@bad-pixel-format,Fail
@@ -37,13 +37,10 @@ kms_rotation_crc@multiplane-rotation-cropping-bottom,Fail
 kms_rotation_crc@multiplane-rotation-cropping-top,Fail
 perf@i915-ref-count,Fail
 perf_pmu@module-unload,Fail
+perf_pmu@most-busy-idle-check-all,Fail
 perf_pmu@rc6,Crash
+prime_busy@before-wait,Fail
 sysfs_heartbeat_interval@long,Timeout
 sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-jsl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-jsl-skips.txt
index 1a3d87c0ca6e..8dec57da1bb3 100644
--- a/drivers/gpu/drm/ci/xfails/i915-jsl-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-jsl-skips.txt
@@ -9,6 +9,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt
index d4fba4f55ec1..117098bc95d9 100644
--- a/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt
@@ -21,8 +21,3 @@ sysfs_heartbeat_interval@long,Timeout
 sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt
index dc722d6a774e..e287462a491a 100644
--- a/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt
@@ -12,6 +12,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt
index 93d42b146df9..462c050a8b2d 100644
--- a/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt
@@ -1,13 +1,14 @@
 api_intel_allocator@reopen,Timeout
 api_intel_bb@destroy-bb,Timeout
 core_hotunplug@hotrebind-lateclose,Timeout
-drm_read@short-buffer-block,Timeout
 dumb_buffer@map-valid,Timeout
 i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
 i915_module_load@resize-bar,Fail
+i915_pm_rpm@gem-execbuf-stress,Timeout
 i915_pm_rps@engine-order,Timeout
+i915_pm_rps@waitboost,Fail
 kms_lease@lease-uevent,Fail
 kms_rotation_crc@multiplane-rotation,Fail
 perf@i915-ref-count,Fail
@@ -16,6 +17,7 @@ perf_pmu@enable-race,Timeout
 perf_pmu@module-unload,Fail
 perf_pmu@rc6,Crash
 perf_pmu@semaphore-wait-idle,Timeout
+prime_busy@before,Fail
 prime_mmap@test_refcounting,Timeout
 sriov_basic@enable-vfs-bind-unbind-each-numvfs-all,Timeout
 syncobj_basic@illegal-fd-to-handle,Timeout
@@ -26,8 +28,3 @@ syncobj_wait@multi-wait-all-submitted,Timeout
 syncobj_wait@multi-wait-for-submit-submitted-signaled,Timeout
 syncobj_wait@wait-any-complex,Timeout
 syncobj_wait@wait-delayed-signal,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt
index 938377896841..429dc3c731df 100644
--- a/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt
@@ -18,6 +18,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt
index 1cb6978c86dc..0f167cfd503c 100644
--- a/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt
@@ -1,4 +1,4 @@
-core_setmaster@master-drop-set-shared-fd,Fail
+core_setmaster_vs_auth,Fail
 i915_module_load@load,Fail
 i915_module_load@reload,Fail
 i915_module_load@reload-no-display,Fail
@@ -6,10 +6,9 @@ i915_module_load@resize-bar,Fail
 i915_pm_rpm@gem-execbuf-stress,Timeout
 i915_pm_rpm@module-reload,Fail
 i915_pm_rpm@system-suspend-execbuf,Timeout
-kms_async_flips@test-time-stamp,Timeout
-kms_ccs@crc-sprite-planes-basic-y-tiled-ccs,Timeout
-kms_dirtyfb@default-dirtyfb-ioctl,Fail
-kms_dirtyfb@fbc-dirtyfb-ioctl,Fail
+kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc,Timeout
+kms_cursor_crc@cursor-suspend,Timeout
+kms_fb_coherency@memset-crc,Crash
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail
 kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail
@@ -30,13 +29,19 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail
 kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail
+kms_frontbuffer_tracking@fbc-rgb101010-draw-mmap-cpu,Timeout
 kms_frontbuffer_tracking@fbc-tiling-linear,Fail
 kms_lease@lease-uevent,Fail
-kms_lease@page-flip-implicit-plane,Timeout
 kms_plane_alpha_blend@alpha-basic,Fail
 kms_plane_alpha_blend@alpha-opaque-fb,Fail
 kms_plane_alpha_blend@alpha-transparent-fb,Fail
 kms_plane_alpha_blend@constant-alpha-max,Fail
+kms_plane_scaling@planes-upscale-factor-0-25,Timeout
+kms_pm_backlight@brightness-with-dpms,Crash
+kms_pm_backlight@fade,Crash
+kms_prop_blob@invalid-set-prop-any,Fail
+kms_properties@connector-properties-legacy,Timeout
+kms_universal_plane@disable-primary-vs-flip,Timeout
 perf@i915-ref-count,Fail
 perf_pmu@module-unload,Fail
 perf_pmu@rc6,Crash
@@ -45,8 +50,3 @@ sysfs_heartbeat_interval@long,Timeout
 sysfs_heartbeat_interval@off,Timeout
 sysfs_preempt_timeout@off,Timeout
 sysfs_timeslice_duration@off,Timeout
-xe_module_load@force-load,Fail
-xe_module_load@load,Fail
-xe_module_load@many-reload,Fail
-xe_module_load@reload,Fail
-xe_module_load@reload-no-display,Fail
diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt
index 29bff8922ae1..7e7374ebf3d1 100644
--- a/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt
@@ -9,6 +9,7 @@ nouveau_.*
 ^v3d.*
 ^vc4.*
 ^vmwgfx*
+^xe.*
 
 # GEM tests takes ~1000 hours, so skip it
 gem_.*
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt
index 4f176c04ec4e..592d7d69e6fc 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt
@@ -17,8 +17,28 @@ kms_bw@linear-tiling-2-displays-3840x2160p,Fail
 kms_color@invalid-gamma-lut-sizes,Fail
 kms_cursor_legacy@cursor-vs-flip-atomic,Fail
 kms_cursor_legacy@cursor-vs-flip-legacy,Fail
+kms_cursor_legacy@flip-vs-cursor-atomic,Fail
+kms_cursor_legacy@flip-vs-cursor-legacy,Fail
+kms_cursor_legacy@flip-vs-cursor-toggle,Fail
+kms_cursor_legacy@flip-vs-cursor-varying-size,Fail
+kms_flip@basic-plain-flip,Fail
+kms_flip@dpms-off-confusion,Fail
+kms_flip@dpms-off-confusion-interruptible,Fail
+kms_flip@flip-vs-absolute-wf_vblank,Fail
+kms_flip@flip-vs-absolute-wf_vblank-interruptible,Fail
+kms_flip@flip-vs-blocking-wf-vblank,Fail
+kms_flip@flip-vs-expired-vblank,Fail
+kms_flip@flip-vs-expired-vblank-interruptible,Fail
 kms_flip@flip-vs-modeset-vs-hang,Fail
+kms_flip@flip-vs-panning,Fail
+kms_flip@flip-vs-panning-interruptible,Fail
 kms_flip@flip-vs-panning-vs-hang,Fail
 kms_flip@flip-vs-suspend,Fail
 kms_flip@flip-vs-suspend-interruptible,Fail
+kms_flip@plain-flip-fb-recreate,Fail
+kms_flip@plain-flip-fb-recreate-interruptible,Fail
+kms_flip@plain-flip-interruptible,Fail
+kms_flip@plain-flip-ts-check,Fail
+kms_flip@plain-flip-ts-check-interruptible,Fail
+kms_invalid_mode@overflow-vrefresh,Fail
 kms_lease@lease-uevent,Fail
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt
index 2956567c3048..443596d9e662 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt
@@ -46,3 +46,10 @@ kms_prop_blob@invalid-set-prop
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_bw@connected-linear-tiling-1-displays-2160x1440p
+
+# Board Name: mt8173-elm-hana
+# Bug Report: https://lore.kernel.org/linux-mediatek/d25442b9-0b6b-433c-8e23-997840fad305@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_flip@flip-vs-wf_vblank-interruptible
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt
index d0db51874aef..b5ee7323a160 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt
@@ -11,6 +11,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt
index 5a063361d7f2..184d0cccc318 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt
@@ -1,12 +1,38 @@
-core_setmaster@master-drop-set-user,Fail
 dumb_buffer@create-clear,Crash
 kms_bw@connected-linear-tiling-1-displays-1920x1080p,Fail
+kms_bw@connected-linear-tiling-1-displays-2160x1440p,Fail
 kms_bw@connected-linear-tiling-1-displays-2560x1440p,Fail
 kms_bw@connected-linear-tiling-1-displays-3840x2160p,Fail
 kms_bw@linear-tiling-1-displays-1920x1080p,Fail
 kms_bw@linear-tiling-1-displays-2160x1440p,Fail
 kms_bw@linear-tiling-1-displays-3840x2160p,Fail
+kms_color@invalid-gamma-lut-sizes,Fail
 kms_cursor_legacy@cursor-vs-flip-atomic,Fail
+kms_cursor_legacy@cursor-vs-flip-legacy,Fail
+kms_cursor_legacy@flip-vs-cursor-atomic,Fail
+kms_cursor_legacy@flip-vs-cursor-legacy,Fail
+kms_cursor_legacy@flip-vs-cursor-toggle,Fail
+kms_cursor_legacy@flip-vs-cursor-varying-size,Fail
+kms_flip@basic-flip-vs-wf_vblank,Fail
+kms_flip@basic-plain-flip,Fail
+kms_flip@dpms-off-confusion,Fail
+kms_flip@dpms-off-confusion-interruptible,Fail
+kms_flip@flip-vs-absolute-wf_vblank,Fail
+kms_flip@flip-vs-absolute-wf_vblank-interruptible,Fail
+kms_flip@flip-vs-blocking-wf-vblank,Fail
+kms_flip@flip-vs-expired-vblank,Fail
+kms_flip@flip-vs-expired-vblank-interruptible,Fail
+kms_flip@flip-vs-modeset-vs-hang,Fail
+kms_flip@flip-vs-panning,Fail
+kms_flip@flip-vs-panning-interruptible,Fail
 kms_flip@flip-vs-panning-vs-hang,Fail
 kms_flip@flip-vs-suspend,Fail
+kms_flip@flip-vs-suspend-interruptible,Fail
+kms_flip@flip-vs-wf_vblank-interruptible,Fail
+kms_flip@plain-flip-fb-recreate,Fail
+kms_flip@plain-flip-fb-recreate-interruptible,Fail
+kms_flip@plain-flip-interruptible,Fail
+kms_flip@plain-flip-ts-check,Fail
+kms_flip@plain-flip-ts-check-interruptible,Fail
+kms_invalid_mode@overflow-vrefresh,Fail
 kms_lease@lease-uevent,Fail
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt
index df7e5ce7a036..0c67fec92450 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt
@@ -18,3 +18,24 @@ kms_cursor_legacy@cursor-vs-flip-atomic-transitions
 # IGT Version: 1.28-gf13702b8e
 # Linux Version: 6.10.0-rc5
 fbdev@write
+
+# Board Name: mt8183-kukui-jacuzzi-juniper-sku16
+# Bug Report: https://lore.kernel.org/linux-mediatek/a520d1d6-95b3-4573-b8f2-689f05bc2230@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_flip@basic-flip-vs-modeset
+
+# Board Name: mt8183-kukui-jacuzzi-juniper-sku16
+# Bug Report: https://lore.kernel.org/linux-mediatek/ca960a82-00fc-4183-b983-998f7ac2fbb5@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_atomic_transition@plane-all-modeset-transition-internal-panels
+
+# Board Name: mt8183-kukui-jacuzzi-juniper-sku16
+# Bug Report: https://lore.kernel.org/linux-mediatek/da578eed-224f-4374-853a-1ff0aa20d03b@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_atomic_transition@plane-toggle-modeset-transition
diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt
index d0db51874aef..b5ee7323a160 100644
--- a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt
@@ -11,6 +11,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/meson-g12b-skips.txt b/drivers/gpu/drm/ci/xfails/meson-g12b-skips.txt
index 8198e06344a3..9fd44a4b962a 100644
--- a/drivers/gpu/drm/ci/xfails/meson-g12b-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/meson-g12b-skips.txt
@@ -11,6 +11,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt b/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt
index 7752adff05c1..72c469021b66 100644
--- a/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt
@@ -1,8 +1,4 @@
 kms_3d,Fail
-kms_cursor_legacy@forked-bo,Fail
-kms_cursor_legacy@forked-move,Fail
-kms_cursor_legacy@single-bo,Fail
-kms_cursor_legacy@torture-bo,Fail
 kms_force_connector_basic@force-edid,Fail
 kms_hdmi_inject@inject-4k,Fail
 kms_lease@lease-uevent,Fail
diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8016-skips.txt b/drivers/gpu/drm/ci/xfails/msm-apq8016-skips.txt
index 1674c8e214d6..87724413174c 100644
--- a/drivers/gpu/drm/ci/xfails/msm-apq8016-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-apq8016-skips.txt
@@ -10,6 +10,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt b/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt
index 5550be5486ed..a4d2f2a7963a 100644
--- a/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt
@@ -13,6 +13,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt
index 8910afb6acf2..d270af1cca52 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt
@@ -32,3 +32,10 @@ kms_lease@page-flip-implicit-plane
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_plane@plane-position-hole-dpms
+
+# Board Name: sc7180-trogdor-kingoftown
+# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/73
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_plane@plane-position-covered
diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt
index 478d7c161616..d4b8ba3a54a9 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt
@@ -13,6 +13,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
@@ -28,3 +29,6 @@ kms_cursor_crc@cursor-random-max-size
 # https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162
 kms_display_modes@extended-mode-basic
 kms_display_modes@mst-extended-mode-negative
+
+# It causes other tests to fail, so skip it.
+kms_invalid_mode@overflow-vrefresh
diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-flakes.txt
index cd3d3b0befe4..cafc802cecea 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-flakes.txt
@@ -11,3 +11,10 @@ msm/msm_mapping@shadow
 # IGT Version: 1.28-gf13702b8e
 # Linux Version: 6.10.0-rc5
 kms_lease@page-flip-implicit-plane
+
+# Board Name: sc7180-trogdor-lazor-limozeen-nots-r5
+# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/74
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_cursor_crc@cursor-random-128x128
diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt
index ef9318afcd89..022db559cc7d 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt
@@ -13,6 +13,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt
index 38ec0305c1f4..e32d73c6c98e 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt
@@ -130,3 +130,10 @@ kms_lease@page-flip-implicit-plane
 # IGT Version: 1.28-ga73311079
 # Linux Version: 6.11.0-rc5
 kms_flip@flip-vs-expired-vblank
+
+# Board Name: sdm845-cheza-r3
+# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/75
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_flip@plain-flip-ts-check-interruptible
diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt
index 2ce7f7e23a01..6c86d1953e11 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt
@@ -18,6 +18,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
@@ -35,3 +36,315 @@ kms_content_protection@uevent
 # https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162
 kms_display_modes@extended-mode-basic
 kms_display_modes@mst-extended-mode-negative
+
+# Kernel panic
+msm/msm_recovery@hangcheck
+# DEBUG - Begin test msm/msm_recovery@hangcheck
+# Console: switching to colour dummy device 80x25
+# [  489.526286] [IGT] msm_recovery: executing
+# [  489.531926] [IGT] msm_recovery: starting subtest hangcheck
+# [  492.808574] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  492.820358] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  492.831154] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  493.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  493.844177] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  493.854971] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  494.824633] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  494.836237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  494.847034] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  495.816570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  495.828170] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  495.838966] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  496.804643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  496.816246] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  496.827041] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  497.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  497.844170] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  497.854963] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  498.820636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  498.832232] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  498.843024] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  499.816568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  499.828163] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  499.838958] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  500.808570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  500.820165] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  500.830960] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  501.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  501.844175] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  501.854965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  502.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  502.836171] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  502.846965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  503.816570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  503.828176] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  503.838969] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  504.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  504.816237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  504.827033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  505.828643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  505.840247] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  505.851043] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  506.820637] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  506.832233] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  506.843026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  507.816567] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  507.828171] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  507.838965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  508.808568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  508.820173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  508.830969] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  509.832568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  509.844173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  509.854967] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  510.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  510.836162] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  510.846954] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  511.816569] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  511.828173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  511.838968] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  512.804641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  512.816246] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  512.827040] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  513.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  513.840239] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  513.851035] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  514.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  514.836164] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  514.846959] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  515.812640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  515.824235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  515.835030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  515.912427] rcu: INFO: rcu_preempt self-detected stall on CPU
+# [  515.918398] rcu: 	0-....: (6452 ticks this GP) idle=6afc/1/0x4000000000000000 softirq=12492/12697 fqs=3179
+# [  515.929296] rcu: 	(t=6505 jiffies g=36205 q=58 ncpus=8)
+# [  515.934709] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G        W          6.14.0-rc4-gdddf15cff632 #1
+# [  515.934727] Tainted: [W]=WARN
+# [  515.934732] Hardware name: Google Cheza (rev3+) (DT)
+# [  515.934739] pstate: 00400009 (nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
+# [  515.934751] pc : rcu_core+0x59c/0xe68
+# [  515.934769] lr : rcu_core+0x74/0xe68
+# [  515.934781] sp : ffff800080003e50
+# [  515.934785] x29: ffff800080003e50 x28: ffff225d038e9bc0 x27: 0000000000000002
+# [  515.934805] x26: ffffc171a8ee6108 x25: ffffc171a85bc2c0 x24: ffff60ecd691e000
+# [  515.934820] x23: ffffc171a85d15c0 x22: ffffc171a8f8d780 x21: ffff225e7eeef5c0
+# [  515.934835] x20: ffffc171a8ef0e80 x19: ffffc171a85d15d1 x18: ffffc171a9461e70
+# [  515.934850] x17: ffff60ecd691e000 x16: ffff800080000000 x15: 0000000000000000
+# [  515.934866] x14: ffffc171a85d0780 x13: 0000000000000400 x12: 0000000000000000
+# [  515.934880] x11: ffffc171a85ce900 x10: ffffc171a8ef5000 x9 : ffffc171a8ef0000
+# [  515.934894] x8 : ffff800080003d88 x7 : ffffc171a8ee6100 x6 : ffff800080003de0
+# [  515.934909] x5 : ffff800080003dc8 x4 : 0000000000000003 x3 : 0000000000000000
+# [  515.934923] x2 : 0000000000000101 x1 : 0000000000000000 x0 : ffff225d038e9bc0
+# [  515.934939] Call trace:
+# [  515.934945]  rcu_core+0x59c/0xe68 (P)
+# [  515.934962]  rcu_core_si+0x10/0x1c
+# [  515.934976]  handle_softirqs+0x118/0x4b8
+# [  515.934994]  __do_softirq+0x14/0x20
+# [  515.935007]  ____do_softirq+0x10/0x1c
+# [  515.935021]  call_on_irq_stack+0x24/0x4c
+# [  515.935034]  do_softirq_own_stack+0x1c/0x28
+# [  515.935048]  __irq_exit_rcu+0x174/0x1b4
+# [  515.935063]  irq_exit_rcu+0x10/0x38
+# [  515.935077]  el1_interrupt+0x38/0x64
+# [  515.935092]  el1h_64_irq_handler+0x18/0x24
+# [  515.935104]  el1h_64_irq+0x6c/0x70
+# [  515.935115]  lock_acquire+0x1e0/0x338 (P)
+# [  515.935129]  __mutex_lock+0xa8/0x4b8
+# [  515.935144]  mutex_lock_nested+0x24/0x30
+# [  515.935159]  _find_opp_table_unlocked+0x40/0xfc
+# [  515.935174]  _find_key+0x64/0x16c
+# [  515.935184]  dev_pm_opp_find_freq_exact+0x4c/0x74
+# [  515.935197]  qcom_cpufreq_hw_target_index+0xe8/0x128
+# [  515.935211]  __cpufreq_driver_target+0x144/0x29c
+# [  515.935227]  sugov_work+0x58/0x74
+# [  515.935239]  kthread_worker_fn+0xf4/0x324
+# [  515.935254]  kthread+0x12c/0x208
+# [  515.935266]  ret_from_fork+0x10/0x20
+# [  516.808569] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  516.820174] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  516.830968] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  517.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  517.840236] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  517.851032] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  518.820642] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  518.832237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  518.843030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  519.812636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  519.824231] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  519.835026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  520.808570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  520.820165] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  520.830959] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  521.828643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  521.840238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  521.851033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  522.820636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  522.832232] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  522.843027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  523.812639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  523.824239] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  523.835034] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  524.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  524.816235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  524.827026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  525.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  525.840236] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  525.851031] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  526.820641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  526.832244] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  526.843041] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  527.812642] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  527.824242] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  527.835038] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  528.804639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  528.816234] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  528.827027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  529.832634] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  529.844231] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  529.855017] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  530.820646] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  530.832270] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  530.843065] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  531.812640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  531.824238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  531.835030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  532.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  532.816237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  532.827031] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  533.828640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  533.840243] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  533.851037] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  534.820640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  534.832245] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  534.843038] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  535.812641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  535.824238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  535.835033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  536.804639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  536.816235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  536.827030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  537.828640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  537.840234] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  537.851020] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  538.820640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  538.832235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  538.843027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  539.812644] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0!
+# [  539.824247] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     completed fence: 45605
+# [  539.835040] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2:     submitted fence: 45611
+# [  540.124426] watchdog: BUG: soft lockup - CPU#0 stuck for 49s! [sugov:0:126]
+# [  540.124439] Modules linked in:
+# [  540.124448] irq event stamp: 9912389
+# [  540.124453] hardirqs last  enabled at (9912388): [<ffffc171a767a24c>] exit_to_kernel_mode+0x38/0x130
+# [  540.124473] hardirqs last disabled at (9912389): [<ffffc171a767a368>] el1_interrupt+0x24/0x64
+# [  540.124486] softirqs last  enabled at (9898068): [<ffffc171a62bc290>] handle_softirqs+0x4a0/0x4b8
+# [  540.124505] softirqs last disabled at (9898071): [<ffffc171a62105b0>] __do_softirq+0x14/0x20
+# [  540.124525] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G        W          6.14.0-rc4-gdddf15cff632 #1
+# [  540.124540] Tainted: [W]=WARN
+# [  540.124544] Hardware name: Google Cheza (rev3+) (DT)
+# [  540.124549] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
+# [  540.124560] pc : xhci_urb_enqueue+0xbc/0x32c
+# [  540.124573] lr : xhci_urb_enqueue+0xb4/0x32c
+# [  540.124581] sp : ffff800080003c20
+# [  540.124586] x29: ffff800080003c20 x28: 0000000000000000 x27: ffff225d00b1e6a0
+# [  540.124602] x26: ffff225d01c3d800 x25: 0000000000000001 x24: 0000000000000006
+# [  540.124617] x23: ffff225d044dc000 x22: ffff225d044dc000 x21: 0000000000000001
+# [  540.124632] x20: ffff225d002d7280 x19: ffff225d0573a780 x18: ffff225e7eff0f50
+# [  540.124647] x17: 000000000000cab0 x16: 0000000000000000 x15: ffff225d0353a000
+# [  540.124661] x14: 0000000000000000 x13: 0000000000000820 x12: 0000000000000000
+# [  540.124674] x11: ffff800080003a30 x10: 0000000000000001 x9 : 0000000000000000
+# [  540.124689] x8 : ffff225d002d7300 x7 : 0000000000000000 x6 : 000000000000003f
+# [  540.124702] x5 : 00000000ffffffff x4 : 0000000000000920 x3 : 0000000000000080
+# [  540.124716] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff225d002d7280
+# [  540.124731] Call trace:
+# [  540.124736]  xhci_urb_enqueue+0xbc/0x32c (P)
+# [  540.124751]  usb_hcd_submit_urb+0x98/0x7fc
+# [  540.124766]  usb_submit_urb+0x294/0x560
+# [  540.124780]  intr_callback+0x78/0x1fc
+# [  540.124798]  __usb_hcd_giveback_urb+0x68/0x128
+# [  540.124812]  usb_giveback_urb_bh+0xa8/0x140
+# [  540.124825]  process_one_work+0x208/0x5e8
+# [  540.124840]  bh_worker+0x1a8/0x20c
+# [  540.124853]  workqueue_softirq_action+0x78/0x88
+# [  540.124868]  tasklet_hi_action+0x14/0x3c
+# [  540.124883]  handle_softirqs+0x118/0x4b8
+# [  540.124897]  __do_softirq+0x14/0x20
+# [  540.124908]  ____do_softirq+0x10/0x1c
+# [  540.124922]  call_on_irq_stack+0x24/0x4c
+# [  540.124934]  do_softirq_own_stack+0x1c/0x28
+# [  540.124947]  __irq_exit_rcu+0x174/0x1b4
+# [  540.124961]  irq_exit_rcu+0x10/0x38
+# [  540.124976]  el1_interrupt+0x38/0x64
+# [  540.124987]  el1h_64_irq_handler+0x18/0x24
+# [  540.124998]  el1h_64_irq+0x6c/0x70
+# [  540.125009]  lock_acquire+0x1e0/0x338 (P)
+# [  540.125023]  __mutex_lock+0xa8/0x4b8
+# [  540.125038]  mutex_lock_nested+0x24/0x30
+# [  540.125052]  _find_opp_table_unlocked+0x40/0xfc
+# [  540.125067]  _find_key+0x64/0x16c
+# [  540.125078]  dev_pm_opp_find_freq_exact+0x4c/0x74
+# [  540.125090]  qcom_cpufreq_hw_target_index+0xe8/0x128
+# [  540.125105]  __cpufreq_driver_target+0x144/0x29c
+# [  540.125121]  sugov_work+0x58/0x74
+# [  540.125133]  kthread_worker_fn+0xf4/0x324
+# [  540.125148]  kthread+0x12c/0x208
+# [  540.125160]  ret_from_fork+0x10/0x20
+# [  540.125176] Kernel panic - not syncing: softlockup: hung tasks
+# [  540.423567] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G        W    L     6.14.0-rc4-gdddf15cff632 #1
+# [  540.433411] Tainted: [W]=WARN, [L]=SOFTLOCKUP
+# [  540.437901] Hardware name: Google Cheza (rev3+) (DT)
+# [  540.443022] Call trace:
+# [  540.445559]  show_stack+0x18/0x24 (C)
+# [  540.449357]  dump_stack_lvl+0x38/0xd0
+# [  540.453157]  dump_stack+0x18/0x24
+# [  540.456599]  panic+0x3bc/0x41c
+# [  540.459767]  watchdog_timer_fn+0x254/0x2e4
+# [  540.464005]  __hrtimer_run_queues+0x3c4/0x440
+# [  540.468508]  hrtimer_interrupt+0xe4/0x244
+# [  540.472662]  arch_timer_handler_phys+0x2c/0x44
+# [  540.477256]  handle_percpu_devid_irq+0x90/0x1f0
+# [  540.481943]  handle_irq_desc+0x40/0x58
+# [  540.485829]  generic_handle_domain_irq+0x1c/0x28
+# [  540.490604]  gic_handle_irq+0x4c/0x11c
+# [  540.494483]  do_interrupt_handler+0x50/0x84
+# [  540.498811]  el1_interrupt+0x34/0x64
+# [  540.502518]  el1h_64_irq_handler+0x18/0x24
+# [  540.506758]  el1h_64_irq+0x6c/0x70
+# [  540.510279]  xhci_urb_enqueue+0xbc/0x32c (P)
+# [  540.514693]  usb_hcd_submit_urb+0x98/0x7fc
+# [  540.518932]  usb_submit_urb+0x294/0x560
+# [  540.522901]  intr_callback+0x78/0x1fc
+# [  540.526700]  __usb_hcd_giveback_urb+0x68/0x128
+# [  540.531288]  usb_giveback_urb_bh+0xa8/0x140
+# [  540.535614]  process_one_work+0x208/0x5e8
+# [  540.539769]  bh_worker+0x1a8/0x20c
+# [  540.543293]  workqueue_softirq_action+0x78/0x88
+# [  540.547980]  tasklet_hi_action+0x14/0x3c
+# [  540.552038]  handle_softirqs+0x118/0x4b8
+# [  540.556096]  __do_softirq+0x14/0x20
+# [  540.559705]  ____do_softirq+0x10/0x1c
+# [  540.563500]  call_on_irq_stack+0x24/0x4c
+# [  540.567554]  do_softirq_own_stack+0x1c/0x28
+# [  540.571878]  __irq_exit_rcu+0x174/0x1b4
+# [  540.575849]  irq_exit_rcu+0x10/0x38
+# [  540.579462]  el1_interrupt+0x38/0x64
+# [  540.583158]  el1h_64_irq_handler+0x18/0x24
+# [  540.587397]  el1h_64_irq+0x6c/0x70
+# [  540.590918]  lock_acquire+0x1e0/0x338 (P)
+# [  540.595060]  __mutex_lock+0xa8/0x4b8
+# [  540.598760]  mutex_lock_nested+0x24/0x30
+# [  540.602818]  _find_opp_table_unlocked+0x40/0xfc
+# [  540.607503]  _find_key+0x64/0x16c
+# [  540.610940]  dev_pm_opp_find_freq_exact+0x4c/0x74
+# [  540.615798]  qcom_cpufreq_hw_target_index+0xe8/0x128
+# [  540.620924]  __cpufreq_driver_target+0x144/0x29c
+# [  540.625698]  sugov_work+0x58/0x74
+# [  540.629134]  kthread_worker_fn+0xf4/0x324
+# [  540.633278]  kthread+0x12c/0x208
+# [  540.636619]  ret_from_fork+0x10/0x20
+# [  540.640321] SMP: stopping secondary CPUs
+# [  540.644518] Kernel Offset: 0x417126200000 from 0xffff800080000000
+# [  540.650848] PHYS_OFFSET: 0xfff0dda400000000
+# [  540.655170] CPU features: 0x000,00000100,00901250,8200721b
+# [  540.660829] Memory Limit: none
+# [  540.663999] ---[ end Kernel panic - not syncing: softlockup: hung tasks ]---
diff --git a/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-skips.txt
index 329770c520d9..9450f2a002fd 100644
--- a/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-skips.txt
@@ -10,6 +10,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/panfrost-g12b-skips.txt b/drivers/gpu/drm/ci/xfails/panfrost-g12b-skips.txt
index 3c7e494857b5..198deea3faa9 100644
--- a/drivers/gpu/drm/ci/xfails/panfrost-g12b-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/panfrost-g12b-skips.txt
@@ -10,6 +10,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Panfrost is not a KMS driver, so skip the KMS tests
 kms_.*
diff --git a/drivers/gpu/drm/ci/xfails/panfrost-mt8183-skips.txt b/drivers/gpu/drm/ci/xfails/panfrost-mt8183-skips.txt
index 3c7e494857b5..198deea3faa9 100644
--- a/drivers/gpu/drm/ci/xfails/panfrost-mt8183-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/panfrost-mt8183-skips.txt
@@ -10,6 +10,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Panfrost is not a KMS driver, so skip the KMS tests
 kms_.*
diff --git a/drivers/gpu/drm/ci/xfails/panfrost-rk3288-skips.txt b/drivers/gpu/drm/ci/xfails/panfrost-rk3288-skips.txt
index feeed89b6c3f..af99ac54c3a5 100644
--- a/drivers/gpu/drm/ci/xfails/panfrost-rk3288-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/panfrost-rk3288-skips.txt
@@ -13,6 +13,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Panfrost is not a KMS driver, so skip the KMS tests
 kms_.*
diff --git a/drivers/gpu/drm/ci/xfails/panfrost-rk3399-skips.txt b/drivers/gpu/drm/ci/xfails/panfrost-rk3399-skips.txt
index feeed89b6c3f..af99ac54c3a5 100644
--- a/drivers/gpu/drm/ci/xfails/panfrost-rk3399-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/panfrost-rk3399-skips.txt
@@ -13,6 +13,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Panfrost is not a KMS driver, so skip the KMS tests
 kms_.*
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt
index ba9160d4d8eb..61122ea7f008 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt
@@ -5,6 +5,5 @@ core_setmaster_vs_auth,Crash
 dumb_buffer@create-clear,Crash
 fbdev@pan,Crash
 kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail
-kms_flip@flip-vs-modeset-vs-hang,Crash
 kms_prop_blob@invalid-set-prop,Crash
 kms_prop_blob@invalid-set-prop-any,Crash
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt
index eb16b29dee48..71418ea35a17 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt
@@ -14,6 +14,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt
index 2803d0d80192..45dd8d493f6e 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt
@@ -2,7 +2,6 @@ dumb_buffer@create-clear,Crash
 kms_atomic_transition@modeset-transition,Fail
 kms_atomic_transition@modeset-transition-fencing,Fail
 kms_atomic_transition@plane-toggle-modeset-transition,Fail
-kms_bw@connected-linear-tiling-1-displays-2160x1440p,Fail
 kms_color@gamma,Fail
 kms_color@legacy-gamma,Fail
 kms_cursor_crc@cursor-alpha-opaque,Fail
@@ -55,6 +54,7 @@ kms_flip@plain-flip-ts-check,Fail
 kms_flip@plain-flip-ts-check-interruptible,Fail
 kms_flip@wf_vblank-ts-check-interruptible,Fail
 kms_invalid_mode@int-max-clock,Fail
+kms_invalid_mode@overflow-vrefresh,Fail
 kms_lease@lease-uevent,Fail
 kms_lease@page-flip-implicit-plane,Fail
 kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt
index 348b4ce7eb4b..b467991d4094 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt
@@ -75,58 +75,72 @@ kms_bw@linear-tiling-2-displays-2160x1440p
 # Linux Version: 6.11.0-rc5
 kms_flip@flip-vs-expired-vblank
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/f944dd08-c88c-49ae-aff0-274374550a93@collabora.com/T/#u
 # Failure Rate: 40
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_bw@linear-tiling-1-displays-2160x1440p
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/afa2d3bf-29f2-488d-8cc9-f30d461444b0@collabora.com/T/#u
 # Failure Rate: 80
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_plane_multiple@tiling-none
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/6fdaa97f-c1a5-4216-831f-dbb7c5f90498@collabora.com/T/#u
 # Failure Rate: 100
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_bw@linear-tiling-1-displays-1920x1080p
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/616aa015-9574-4527-9d07-d8d698bbcc3c@collabora.com/T/#u
 # Failure Rate: 100
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_plane@plane-position-hole-dpms
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/7a1b888f-d7db-4ed7-96cd-3975ace837fb@collabora.com/T/#u
 # Failure Rate: 100
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_flip@flip-vs-absolute-wf_vblank
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/f17fffb6-abc4-464e-8465-395311b01f6a@collabora.com/T/#u
 # Failure Rate: 100
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_flip@flip-vs-blocking-wf-vblank
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/9b590b26-1bf9-4951-b6a3-ef6c67e6a1c6@collabora.com/T/#u
 # Failure Rate: 60
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_bw@linear-tiling-2-displays-1920x1080p
 
-# Board Name: hp-11A-G6-EE-grunt
+# Board Name: rk3399-gru-kevin
 # Bug Report: https://lore.kernel.org/dri-devel/059545fa-65b1-4f5c-a13e-4d2898679f51@collabora.com/T/#u
 # Failure Rate: 20
 # IGT Version: 1.29-g33adea9eb
 # Linux Version: 6.13.0-rc2
 kms_flip@modeset-vs-vblank-race-interruptible
+
+# Board Name: rk3399-gru-kevin
+# Bug Report: https://lore.kernel.org/dri-devel/eece9a80-42f3-41f4-86cc-69d8a51b976a@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_bw@connected-linear-tiling-1-displays-2160x1440p
+
+# Board Name: rk3399-gru-kevin
+# Bug Report: https://lore.kernel.org/dri-devel/63dfd5b7-8a54-44a3-9530-f8dcd77a21d1@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_bw@linear-tiling-1-displays-3840x2160p
diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt
index e8e994d92557..b83ec75161b2 100644
--- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt
@@ -14,6 +14,7 @@ nouveau_.*
 gem_.*
 i915_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt
index c72fee70e739..9749ddb75121 100644
--- a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt
+++ b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt
@@ -157,6 +157,7 @@ kms_flip@plain-flip-ts-check-interruptible,Fail
 kms_flip@wf_vblank-ts-check,Fail
 kms_flip@wf_vblank-ts-check-interruptible,Fail
 kms_invalid_mode@int-max-clock,Fail
+kms_invalid_mode@overflow-vrefresh,Fail
 kms_lease@cursor-implicit-plane,Fail
 kms_lease@lease-uevent,Fail
 kms_lease@page-flip-implicit-plane,Fail
diff --git a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt
index adbcdd0f28d2..28e37185bac0 100644
--- a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt
@@ -19,6 +19,7 @@ gem_.*
 i915_.*
 xe_.*
 tools_test.*
+kms_dp_link_training.*
 
 # Currently fails and causes coverage loss for other tests
 # since core_getversion also fails.
diff --git a/drivers/gpu/drm/ci/xfails/vkms-none-flakes.txt b/drivers/gpu/drm/ci/xfails/vkms-none-flakes.txt
index 62428f3c8f31..e3ca6da8cde7 100644
--- a/drivers/gpu/drm/ci/xfails/vkms-none-flakes.txt
+++ b/drivers/gpu/drm/ci/xfails/vkms-none-flakes.txt
@@ -88,3 +88,31 @@ kms_flip@flip-vs-expired-vblank
 # IGT Version: 1.28-gf13702b8e
 # Linux Version: 6.10.0-rc5
 kms_pipe_crc_basic@nonblocking-crc-frame-sequence
+
+# Board Name: vkms
+# Bug Report: https://lore.kernel.org/dri-devel/2364a6bf-e6bc-4741-8c78-cea8bdb06e03@collabora.com/T/#u
+# Failure Rate: 20
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_flip@modeset-vs-vblank-race
+
+# Board Name: vkms
+# Bug Report: https://lore.kernel.org/dri-devel/f7d72ed9-a783-46d7-b75d-54072bda32a3@collabora.com/T/#u
+# Failure Rate: 100
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_pipe_crc_basic@suspend-read-crc
+
+# Board Name: vkms
+# Bug Report: https://lore.kernel.org/dri-devel/98d3ba54-bcb9-41ab-adb1-a18ba61ee2e4@collabora.com/T/#u
+# Failure Rate: 100
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_plane@plane-panning-bottom-right-suspend
+
+# Board Name: vkms
+# Bug Report: https://lore.kernel.org/dri-devel/b58d15eb-094d-4ac2-aad3-83e518c2f55d@collabora.com/T/#u
+# Failure Rate: 100
+# IGT Version: 1.30-g04bedb923
+# Linux Version: 6.14.0-rc4
+kms_vblank@ts-continuation-dpms-suspend
diff --git a/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt b/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt
index 319789806271..716d2d4e452d 100644
--- a/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt
+++ b/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt
@@ -1,5 +1,6 @@
 # keeps printing vkms_vblank_simulate: vblank timer overrun and never ends
 kms_invalid_mode@int-max-clock
+kms_invalid_mode@overflow-vrefresh
 
 # kernel panic seen with kms_cursor_crc tests
 kms_cursor_crc.*
@@ -802,6 +803,7 @@ gem_.*
 i915_.*
 xe_.*
 tools_test.*
+kms_dp_link_training.*
 
 # IGT issue. is_joiner_mode() should return false for non-Intel hardware.
 # https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162
diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c
index 30c736fc0067..7d2e499ea5de 100644
--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -98,6 +98,21 @@ struct drm_bridge_connector {
 	 * HDMI connector infrastructure, if any (see &DRM_BRIDGE_OP_HDMI).
 	 */
 	struct drm_bridge *bridge_hdmi;
+	/**
+	 * @bridge_hdmi_audio:
+	 *
+	 * The bridge in the chain that implements necessary support for the
+	 * HDMI Audio infrastructure, if any (see &DRM_BRIDGE_OP_HDMI_AUDIO).
+	 */
+	struct drm_bridge *bridge_hdmi_audio;
+	/**
+	 * @bridge_dp_audio:
+	 *
+	 * The bridge in the chain that implements necessary support for the
+	 * DisplayPort Audio infrastructure, if any (see
+	 * &DRM_BRIDGE_OP_DP_AUDIO).
+	 */
+	struct drm_bridge *bridge_dp_audio;
 };
 
 #define to_drm_bridge_connector(x) \
@@ -433,14 +448,25 @@ static int drm_bridge_connector_audio_startup(struct drm_connector *connector)
 		to_drm_bridge_connector(connector);
 	struct drm_bridge *bridge;
 
-	bridge = bridge_connector->bridge_hdmi;
-	if (!bridge)
-		return -EINVAL;
+	if (bridge_connector->bridge_hdmi_audio) {
+		bridge = bridge_connector->bridge_hdmi_audio;
+
+		if (!bridge->funcs->hdmi_audio_startup)
+			return 0;
+
+		return bridge->funcs->hdmi_audio_startup(connector, bridge);
+	}
+
+	if (bridge_connector->bridge_dp_audio) {
+		bridge = bridge_connector->bridge_dp_audio;
 
-	if (!bridge->funcs->hdmi_audio_startup)
-		return 0;
+		if (!bridge->funcs->dp_audio_startup)
+			return 0;
 
-	return bridge->funcs->hdmi_audio_startup(connector, bridge);
+		return bridge->funcs->dp_audio_startup(connector, bridge);
+	}
+
+	return -EINVAL;
 }
 
 static int drm_bridge_connector_audio_prepare(struct drm_connector *connector,
@@ -451,11 +477,19 @@ static int drm_bridge_connector_audio_prepare(struct drm_connector *connector,
 		to_drm_bridge_connector(connector);
 	struct drm_bridge *bridge;
 
-	bridge = bridge_connector->bridge_hdmi;
-	if (!bridge)
-		return -EINVAL;
+	if (bridge_connector->bridge_hdmi_audio) {
+		bridge = bridge_connector->bridge_hdmi_audio;
+
+		return bridge->funcs->hdmi_audio_prepare(connector, bridge, fmt, hparms);
+	}
 
-	return bridge->funcs->hdmi_audio_prepare(connector, bridge, fmt, hparms);
+	if (bridge_connector->bridge_dp_audio) {
+		bridge = bridge_connector->bridge_dp_audio;
+
+		return bridge->funcs->dp_audio_prepare(connector, bridge, fmt, hparms);
+	}
+
+	return -EINVAL;
 }
 
 static void drm_bridge_connector_audio_shutdown(struct drm_connector *connector)
@@ -464,11 +498,15 @@ static void drm_bridge_connector_audio_shutdown(struct drm_connector *connector)
 		to_drm_bridge_connector(connector);
 	struct drm_bridge *bridge;
 
-	bridge = bridge_connector->bridge_hdmi;
-	if (!bridge)
-		return;
+	if (bridge_connector->bridge_hdmi_audio) {
+		bridge = bridge_connector->bridge_hdmi_audio;
+		bridge->funcs->hdmi_audio_shutdown(connector, bridge);
+	}
 
-	bridge->funcs->hdmi_audio_shutdown(connector, bridge);
+	if (bridge_connector->bridge_dp_audio) {
+		bridge = bridge_connector->bridge_dp_audio;
+		bridge->funcs->dp_audio_shutdown(connector, bridge);
+	}
 }
 
 static int drm_bridge_connector_audio_mute_stream(struct drm_connector *connector,
@@ -478,15 +516,27 @@ static int drm_bridge_connector_audio_mute_stream(struct drm_connector *connecto
 		to_drm_bridge_connector(connector);
 	struct drm_bridge *bridge;
 
-	bridge = bridge_connector->bridge_hdmi;
-	if (!bridge)
-		return -EINVAL;
+	if (bridge_connector->bridge_hdmi_audio) {
+		bridge = bridge_connector->bridge_hdmi_audio;
+
+		if (!bridge->funcs->hdmi_audio_mute_stream)
+			return -ENOTSUPP;
 
-	if (bridge->funcs->hdmi_audio_mute_stream)
 		return bridge->funcs->hdmi_audio_mute_stream(connector, bridge,
 							     enable, direction);
-	else
-		return -ENOTSUPP;
+	}
+
+	if (bridge_connector->bridge_dp_audio) {
+		bridge = bridge_connector->bridge_dp_audio;
+
+		if (!bridge->funcs->dp_audio_mute_stream)
+			return -ENOTSUPP;
+
+		return bridge->funcs->dp_audio_mute_stream(connector, bridge,
+							   enable, direction);
+	}
+
+	return -EINVAL;
 }
 
 static const struct drm_connector_hdmi_audio_funcs drm_bridge_connector_hdmi_audio_funcs = {
@@ -576,6 +626,42 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 				max_bpc = bridge->max_bpc;
 		}
 
+		if (bridge->ops & DRM_BRIDGE_OP_HDMI_AUDIO) {
+			if (bridge_connector->bridge_hdmi_audio)
+				return ERR_PTR(-EBUSY);
+
+			if (bridge_connector->bridge_dp_audio)
+				return ERR_PTR(-EBUSY);
+
+			if (!bridge->hdmi_audio_max_i2s_playback_channels &&
+			    !bridge->hdmi_audio_spdif_playback)
+				return ERR_PTR(-EINVAL);
+
+			if (!bridge->funcs->hdmi_audio_prepare ||
+			    !bridge->funcs->hdmi_audio_shutdown)
+				return ERR_PTR(-EINVAL);
+
+			bridge_connector->bridge_hdmi_audio = bridge;
+		}
+
+		if (bridge->ops & DRM_BRIDGE_OP_DP_AUDIO) {
+			if (bridge_connector->bridge_dp_audio)
+				return ERR_PTR(-EBUSY);
+
+			if (bridge_connector->bridge_hdmi_audio)
+				return ERR_PTR(-EBUSY);
+
+			if (!bridge->hdmi_audio_max_i2s_playback_channels &&
+			    !bridge->hdmi_audio_spdif_playback)
+				return ERR_PTR(-EINVAL);
+
+			if (!bridge->funcs->dp_audio_prepare ||
+			    !bridge->funcs->dp_audio_shutdown)
+				return ERR_PTR(-EINVAL);
+
+			bridge_connector->bridge_dp_audio = bridge;
+		}
+
 		if (!drm_bridge_get_next_bridge(bridge))
 			connector_type = bridge->type;
 
@@ -611,22 +697,6 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 					       max_bpc);
 		if (ret)
 			return ERR_PTR(ret);
-
-		if (bridge->hdmi_audio_max_i2s_playback_channels ||
-		    bridge->hdmi_audio_spdif_playback) {
-			if (!bridge->funcs->hdmi_audio_prepare ||
-			    !bridge->funcs->hdmi_audio_shutdown)
-				return ERR_PTR(-EINVAL);
-
-			ret = drm_connector_hdmi_audio_init(connector,
-							    bridge->hdmi_audio_dev,
-							    &drm_bridge_connector_hdmi_audio_funcs,
-							    bridge->hdmi_audio_max_i2s_playback_channels,
-							    bridge->hdmi_audio_spdif_playback,
-							    bridge->hdmi_audio_dai_port);
-			if (ret)
-				return ERR_PTR(ret);
-		}
 	} else {
 		ret = drmm_connector_init(drm, connector,
 					  &drm_bridge_connector_funcs,
@@ -635,6 +705,24 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 			return ERR_PTR(ret);
 	}
 
+	if (bridge_connector->bridge_hdmi_audio ||
+	    bridge_connector->bridge_dp_audio) {
+		struct device *dev;
+
+		if (bridge_connector->bridge_hdmi_audio)
+			dev = bridge_connector->bridge_hdmi_audio->hdmi_audio_dev;
+		else
+			dev = bridge_connector->bridge_dp_audio->hdmi_audio_dev;
+
+		ret = drm_connector_hdmi_audio_init(connector, dev,
+						    &drm_bridge_connector_hdmi_audio_funcs,
+						    bridge->hdmi_audio_max_i2s_playback_channels,
+						    bridge->hdmi_audio_spdif_playback,
+						    bridge->hdmi_audio_dai_port);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+
 	drm_connector_helper_add(connector, &drm_bridge_connector_helper_funcs);
 
 	if (bridge_connector->bridge_hpd)
diff --git a/drivers/gpu/drm/display/drm_dp_cec.c b/drivers/gpu/drm/display/drm_dp_cec.c
index 56a4965e518c..ed31471bd0e2 100644
--- a/drivers/gpu/drm/display/drm_dp_cec.c
+++ b/drivers/gpu/drm/display/drm_dp_cec.c
@@ -96,7 +96,7 @@ static int drm_dp_cec_adap_enable(struct cec_adapter *adap, bool enable)
 	u32 val = enable ? DP_CEC_TUNNELING_ENABLE : 0;
 	ssize_t err = 0;
 
-	err = drm_dp_dpcd_writeb(aux, DP_CEC_TUNNELING_CONTROL, val);
+	err = drm_dp_dpcd_write_byte(aux, DP_CEC_TUNNELING_CONTROL, val);
 	return (enable && err < 0) ? err : 0;
 }
 
@@ -112,7 +112,7 @@ static int drm_dp_cec_adap_log_addr(struct cec_adapter *adap, u8 addr)
 		la_mask |= adap->log_addrs.log_addr_mask | (1 << addr);
 	mask[0] = la_mask & 0xff;
 	mask[1] = la_mask >> 8;
-	err = drm_dp_dpcd_write(aux, DP_CEC_LOGICAL_ADDRESS_MASK, mask, 2);
+	err = drm_dp_dpcd_write_data(aux, DP_CEC_LOGICAL_ADDRESS_MASK, mask, 2);
 	return (addr != CEC_LOG_ADDR_INVALID && err < 0) ? err : 0;
 }
 
@@ -123,15 +123,14 @@ static int drm_dp_cec_adap_transmit(struct cec_adapter *adap, u8 attempts,
 	unsigned int retries = min(5, attempts - 1);
 	ssize_t err;
 
-	err = drm_dp_dpcd_write(aux, DP_CEC_TX_MESSAGE_BUFFER,
-				msg->msg, msg->len);
+	err = drm_dp_dpcd_write_data(aux, DP_CEC_TX_MESSAGE_BUFFER,
+				     msg->msg, msg->len);
 	if (err < 0)
 		return err;
 
-	err = drm_dp_dpcd_writeb(aux, DP_CEC_TX_MESSAGE_INFO,
-				 (msg->len - 1) | (retries << 4) |
-				 DP_CEC_TX_MESSAGE_SEND);
-	return err < 0 ? err : 0;
+	return drm_dp_dpcd_write_byte(aux, DP_CEC_TX_MESSAGE_INFO,
+				      (msg->len - 1) | (retries << 4) |
+				      DP_CEC_TX_MESSAGE_SEND);
 }
 
 static int drm_dp_cec_adap_monitor_all_enable(struct cec_adapter *adap,
@@ -144,13 +143,13 @@ static int drm_dp_cec_adap_monitor_all_enable(struct cec_adapter *adap,
 	if (!(adap->capabilities & CEC_CAP_MONITOR_ALL))
 		return 0;
 
-	err = drm_dp_dpcd_readb(aux, DP_CEC_TUNNELING_CONTROL, &val);
-	if (err >= 0) {
+	err = drm_dp_dpcd_read_byte(aux, DP_CEC_TUNNELING_CONTROL, &val);
+	if (!err) {
 		if (enable)
 			val |= DP_CEC_SNOOPING_ENABLE;
 		else
 			val &= ~DP_CEC_SNOOPING_ENABLE;
-		err = drm_dp_dpcd_writeb(aux, DP_CEC_TUNNELING_CONTROL, val);
+		err = drm_dp_dpcd_write_byte(aux, DP_CEC_TUNNELING_CONTROL, val);
 	}
 	return (enable && err < 0) ? err : 0;
 }
@@ -194,7 +193,7 @@ static int drm_dp_cec_received(struct drm_dp_aux *aux)
 	u8 rx_msg_info;
 	ssize_t err;
 
-	err = drm_dp_dpcd_readb(aux, DP_CEC_RX_MESSAGE_INFO, &rx_msg_info);
+	err = drm_dp_dpcd_read_byte(aux, DP_CEC_RX_MESSAGE_INFO, &rx_msg_info);
 	if (err < 0)
 		return err;
 
@@ -202,7 +201,7 @@ static int drm_dp_cec_received(struct drm_dp_aux *aux)
 		return 0;
 
 	msg.len = (rx_msg_info & DP_CEC_RX_MESSAGE_LEN_MASK) + 1;
-	err = drm_dp_dpcd_read(aux, DP_CEC_RX_MESSAGE_BUFFER, msg.msg, msg.len);
+	err = drm_dp_dpcd_read_data(aux, DP_CEC_RX_MESSAGE_BUFFER, msg.msg, msg.len);
 	if (err < 0)
 		return err;
 
@@ -215,7 +214,7 @@ static void drm_dp_cec_handle_irq(struct drm_dp_aux *aux)
 	struct cec_adapter *adap = aux->cec.adap;
 	u8 flags;
 
-	if (drm_dp_dpcd_readb(aux, DP_CEC_TUNNELING_IRQ_FLAGS, &flags) < 0)
+	if (drm_dp_dpcd_read_byte(aux, DP_CEC_TUNNELING_IRQ_FLAGS, &flags) < 0)
 		return;
 
 	if (flags & DP_CEC_RX_MESSAGE_INFO_VALID)
@@ -230,7 +229,7 @@ static void drm_dp_cec_handle_irq(struct drm_dp_aux *aux)
 		 (DP_CEC_TX_ADDRESS_NACK_ERROR | DP_CEC_TX_DATA_NACK_ERROR))
 		cec_transmit_attempt_done(adap, CEC_TX_STATUS_NACK |
 						CEC_TX_STATUS_MAX_RETRIES);
-	drm_dp_dpcd_writeb(aux, DP_CEC_TUNNELING_IRQ_FLAGS, flags);
+	drm_dp_dpcd_write_byte(aux, DP_CEC_TUNNELING_IRQ_FLAGS, flags);
 }
 
 /**
@@ -253,13 +252,13 @@ void drm_dp_cec_irq(struct drm_dp_aux *aux)
 	if (!aux->cec.adap)
 		goto unlock;
 
-	ret = drm_dp_dpcd_readb(aux, DP_DEVICE_SERVICE_IRQ_VECTOR_ESI1,
-				&cec_irq);
+	ret = drm_dp_dpcd_read_byte(aux, DP_DEVICE_SERVICE_IRQ_VECTOR_ESI1,
+				    &cec_irq);
 	if (ret < 0 || !(cec_irq & DP_CEC_IRQ))
 		goto unlock;
 
 	drm_dp_cec_handle_irq(aux);
-	drm_dp_dpcd_writeb(aux, DP_DEVICE_SERVICE_IRQ_VECTOR_ESI1, DP_CEC_IRQ);
+	drm_dp_dpcd_write_byte(aux, DP_DEVICE_SERVICE_IRQ_VECTOR_ESI1, DP_CEC_IRQ);
 unlock:
 	mutex_unlock(&aux->cec.lock);
 }
@@ -269,7 +268,7 @@ static bool drm_dp_cec_cap(struct drm_dp_aux *aux, u8 *cec_cap)
 {
 	u8 cap = 0;
 
-	if (drm_dp_dpcd_readb(aux, DP_CEC_TUNNELING_CAPABILITY, &cap) != 1 ||
+	if (drm_dp_dpcd_read_byte(aux, DP_CEC_TUNNELING_CAPABILITY, &cap) < 0 ||
 	    !(cap & DP_CEC_TUNNELING_CAPABLE))
 		return false;
 	if (cec_cap)
diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
index dbce1c3f4969..f2a6559a2710 100644
--- a/drivers/gpu/drm/display/drm_dp_helper.c
+++ b/drivers/gpu/drm/display/drm_dp_helper.c
@@ -327,7 +327,7 @@ static int __read_delay(struct drm_dp_aux *aux, const u8 dpcd[DP_RECEIVER_CAP_SI
 	if (offset < DP_RECEIVER_CAP_SIZE) {
 		rd_interval = dpcd[offset];
 	} else {
-		if (drm_dp_dpcd_readb(aux, offset, &rd_interval) != 1) {
+		if (drm_dp_dpcd_read_byte(aux, offset, &rd_interval) < 0) {
 			drm_dbg_kms(aux->drm_dev, "%s: failed rd interval read\n",
 				    aux->name);
 			/* arbitrary default delay */
@@ -358,7 +358,7 @@ int drm_dp_128b132b_read_aux_rd_interval(struct drm_dp_aux *aux)
 	int unit;
 	u8 val;
 
-	if (drm_dp_dpcd_readb(aux, DP_128B132B_TRAINING_AUX_RD_INTERVAL, &val) != 1) {
+	if (drm_dp_dpcd_read_byte(aux, DP_128B132B_TRAINING_AUX_RD_INTERVAL, &val) < 0) {
 		drm_err(aux->drm_dev, "%s: failed rd interval read\n",
 			aux->name);
 		/* default to max */
@@ -704,6 +704,8 @@ EXPORT_SYMBOL(drm_dp_dpcd_set_powered);
  * function returns -EPROTO. Errors from the underlying AUX channel transfer
  * function, with the exception of -EBUSY (which causes the transaction to
  * be retried), are propagated to the caller.
+ *
+ * In most of the cases you want to use drm_dp_dpcd_read_data() instead.
  */
 ssize_t drm_dp_dpcd_read(struct drm_dp_aux *aux, unsigned int offset,
 			 void *buffer, size_t size)
@@ -752,6 +754,8 @@ EXPORT_SYMBOL(drm_dp_dpcd_read);
  * function returns -EPROTO. Errors from the underlying AUX channel transfer
  * function, with the exception of -EBUSY (which causes the transaction to
  * be retried), are propagated to the caller.
+ *
+ * In most of the cases you want to use drm_dp_dpcd_write_data() instead.
  */
 ssize_t drm_dp_dpcd_write(struct drm_dp_aux *aux, unsigned int offset,
 			  void *buffer, size_t size)
@@ -774,14 +778,13 @@ EXPORT_SYMBOL(drm_dp_dpcd_write);
  * @aux: DisplayPort AUX channel
  * @status: buffer to store the link status in (must be at least 6 bytes)
  *
- * Returns the number of bytes transferred on success or a negative error
- * code on failure.
+ * Returns a negative error code on failure or 0 on success.
  */
 int drm_dp_dpcd_read_link_status(struct drm_dp_aux *aux,
 				 u8 status[DP_LINK_STATUS_SIZE])
 {
-	return drm_dp_dpcd_read(aux, DP_LANE0_1_STATUS, status,
-				DP_LINK_STATUS_SIZE);
+	return drm_dp_dpcd_read_data(aux, DP_LANE0_1_STATUS, status,
+				     DP_LINK_STATUS_SIZE);
 }
 EXPORT_SYMBOL(drm_dp_dpcd_read_link_status);
 
@@ -804,30 +807,20 @@ int drm_dp_dpcd_read_phy_link_status(struct drm_dp_aux *aux,
 {
 	int ret;
 
-	if (dp_phy == DP_PHY_DPRX) {
-		ret = drm_dp_dpcd_read(aux,
-				       DP_LANE0_1_STATUS,
-				       link_status,
-				       DP_LINK_STATUS_SIZE);
-
-		if (ret < 0)
-			return ret;
-
-		WARN_ON(ret != DP_LINK_STATUS_SIZE);
-
-		return 0;
-	}
+	if (dp_phy == DP_PHY_DPRX)
+		return drm_dp_dpcd_read_data(aux,
+					     DP_LANE0_1_STATUS,
+					     link_status,
+					     DP_LINK_STATUS_SIZE);
 
-	ret = drm_dp_dpcd_read(aux,
-			       DP_LANE0_1_STATUS_PHY_REPEATER(dp_phy),
-			       link_status,
-			       DP_LINK_STATUS_SIZE - 1);
+	ret = drm_dp_dpcd_read_data(aux,
+				    DP_LANE0_1_STATUS_PHY_REPEATER(dp_phy),
+				    link_status,
+				    DP_LINK_STATUS_SIZE - 1);
 
 	if (ret < 0)
 		return ret;
 
-	WARN_ON(ret != DP_LINK_STATUS_SIZE - 1);
-
 	/* Convert the LTTPR to the sink PHY link status layout */
 	memmove(&link_status[DP_SINK_STATUS - DP_LANE0_1_STATUS + 1],
 		&link_status[DP_SINK_STATUS - DP_LANE0_1_STATUS],
@@ -838,12 +831,81 @@ int drm_dp_dpcd_read_phy_link_status(struct drm_dp_aux *aux,
 }
 EXPORT_SYMBOL(drm_dp_dpcd_read_phy_link_status);
 
+/**
+ * drm_dp_link_power_up() - power up a DisplayPort link
+ * @aux: DisplayPort AUX channel
+ * @revision: DPCD revision supported on the link
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int drm_dp_link_power_up(struct drm_dp_aux *aux, unsigned char revision)
+{
+	u8 value;
+	int err;
+
+	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
+	if (revision < DP_DPCD_REV_11)
+		return 0;
+
+	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
+	if (err < 0)
+		return err;
+
+	value &= ~DP_SET_POWER_MASK;
+	value |= DP_SET_POWER_D0;
+
+	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
+	if (err < 0)
+		return err;
+
+	/*
+	 * According to the DP 1.1 specification, a "Sink Device must exit the
+	 * power saving state within 1 ms" (Section 2.5.3.1, Table 5-52, "Sink
+	 * Control Field" (register 0x600).
+	 */
+	usleep_range(1000, 2000);
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_dp_link_power_up);
+
+/**
+ * drm_dp_link_power_down() - power down a DisplayPort link
+ * @aux: DisplayPort AUX channel
+ * @revision: DPCD revision supported on the link
+ *
+ * Returns 0 on success or a negative error code on failure.
+ */
+int drm_dp_link_power_down(struct drm_dp_aux *aux, unsigned char revision)
+{
+	u8 value;
+	int err;
+
+	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
+	if (revision < DP_DPCD_REV_11)
+		return 0;
+
+	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
+	if (err < 0)
+		return err;
+
+	value &= ~DP_SET_POWER_MASK;
+	value |= DP_SET_POWER_D3;
+
+	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_dp_link_power_down);
+
 static int read_payload_update_status(struct drm_dp_aux *aux)
 {
 	int ret;
 	u8 status;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PAYLOAD_TABLE_UPDATE_STATUS, &status);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PAYLOAD_TABLE_UPDATE_STATUS, &status);
 	if (ret < 0)
 		return ret;
 
@@ -870,21 +932,21 @@ int drm_dp_dpcd_write_payload(struct drm_dp_aux *aux,
 	int ret;
 	int retries = 0;
 
-	drm_dp_dpcd_writeb(aux, DP_PAYLOAD_TABLE_UPDATE_STATUS,
-			   DP_PAYLOAD_TABLE_UPDATED);
+	drm_dp_dpcd_write_byte(aux, DP_PAYLOAD_TABLE_UPDATE_STATUS,
+			       DP_PAYLOAD_TABLE_UPDATED);
 
 	payload_alloc[0] = vcpid;
 	payload_alloc[1] = start_time_slot;
 	payload_alloc[2] = time_slot_count;
 
-	ret = drm_dp_dpcd_write(aux, DP_PAYLOAD_ALLOCATE_SET, payload_alloc, 3);
-	if (ret != 3) {
+	ret = drm_dp_dpcd_write_data(aux, DP_PAYLOAD_ALLOCATE_SET, payload_alloc, 3);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "failed to write payload allocation %d\n", ret);
 		goto fail;
 	}
 
 retry:
-	ret = drm_dp_dpcd_readb(aux, DP_PAYLOAD_TABLE_UPDATE_STATUS, &status);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PAYLOAD_TABLE_UPDATE_STATUS, &status);
 	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "failed to read payload table status %d\n", ret);
 		goto fail;
@@ -1040,15 +1102,15 @@ bool drm_dp_send_real_edid_checksum(struct drm_dp_aux *aux,
 {
 	u8 link_edid_read = 0, auto_test_req = 0, test_resp = 0;
 
-	if (drm_dp_dpcd_read(aux, DP_DEVICE_SERVICE_IRQ_VECTOR,
-			     &auto_test_req, 1) < 1) {
+	if (drm_dp_dpcd_read_byte(aux, DP_DEVICE_SERVICE_IRQ_VECTOR,
+				  &auto_test_req) < 0) {
 		drm_err(aux->drm_dev, "%s: DPCD failed read at register 0x%x\n",
 			aux->name, DP_DEVICE_SERVICE_IRQ_VECTOR);
 		return false;
 	}
 	auto_test_req &= DP_AUTOMATED_TEST_REQUEST;
 
-	if (drm_dp_dpcd_read(aux, DP_TEST_REQUEST, &link_edid_read, 1) < 1) {
+	if (drm_dp_dpcd_read_byte(aux, DP_TEST_REQUEST, &link_edid_read) < 0) {
 		drm_err(aux->drm_dev, "%s: DPCD failed read at register 0x%x\n",
 			aux->name, DP_TEST_REQUEST);
 		return false;
@@ -1061,23 +1123,23 @@ bool drm_dp_send_real_edid_checksum(struct drm_dp_aux *aux,
 		return false;
 	}
 
-	if (drm_dp_dpcd_write(aux, DP_DEVICE_SERVICE_IRQ_VECTOR,
-			      &auto_test_req, 1) < 1) {
+	if (drm_dp_dpcd_write_byte(aux, DP_DEVICE_SERVICE_IRQ_VECTOR,
+				   auto_test_req) < 0) {
 		drm_err(aux->drm_dev, "%s: DPCD failed write at register 0x%x\n",
 			aux->name, DP_DEVICE_SERVICE_IRQ_VECTOR);
 		return false;
 	}
 
 	/* send back checksum for the last edid extension block data */
-	if (drm_dp_dpcd_write(aux, DP_TEST_EDID_CHECKSUM,
-			      &real_edid_checksum, 1) < 1) {
+	if (drm_dp_dpcd_write_byte(aux, DP_TEST_EDID_CHECKSUM,
+				   real_edid_checksum) < 0) {
 		drm_err(aux->drm_dev, "%s: DPCD failed write at register 0x%x\n",
 			aux->name, DP_TEST_EDID_CHECKSUM);
 		return false;
 	}
 
 	test_resp |= DP_TEST_EDID_CHECKSUM_WRITE;
-	if (drm_dp_dpcd_write(aux, DP_TEST_RESPONSE, &test_resp, 1) < 1) {
+	if (drm_dp_dpcd_write_byte(aux, DP_TEST_RESPONSE, test_resp) < 0) {
 		drm_err(aux->drm_dev, "%s: DPCD failed write at register 0x%x\n",
 			aux->name, DP_TEST_RESPONSE);
 		return false;
@@ -1114,12 +1176,10 @@ static int drm_dp_read_extended_dpcd_caps(struct drm_dp_aux *aux,
 	      DP_EXTENDED_RECEIVER_CAP_FIELD_PRESENT))
 		return 0;
 
-	ret = drm_dp_dpcd_read(aux, DP_DP13_DPCD_REV, &dpcd_ext,
-			       sizeof(dpcd_ext));
+	ret = drm_dp_dpcd_read_data(aux, DP_DP13_DPCD_REV, &dpcd_ext,
+				    sizeof(dpcd_ext));
 	if (ret < 0)
 		return ret;
-	if (ret != sizeof(dpcd_ext))
-		return -EIO;
 
 	if (dpcd[DP_DPCD_REV] > dpcd_ext[DP_DPCD_REV]) {
 		drm_dbg_kms(aux->drm_dev,
@@ -1156,10 +1216,10 @@ int drm_dp_read_dpcd_caps(struct drm_dp_aux *aux,
 {
 	int ret;
 
-	ret = drm_dp_dpcd_read(aux, DP_DPCD_REV, dpcd, DP_RECEIVER_CAP_SIZE);
+	ret = drm_dp_dpcd_read_data(aux, DP_DPCD_REV, dpcd, DP_RECEIVER_CAP_SIZE);
 	if (ret < 0)
 		return ret;
-	if (ret != DP_RECEIVER_CAP_SIZE || dpcd[DP_DPCD_REV] == 0)
+	if (dpcd[DP_DPCD_REV] == 0)
 		return -EIO;
 
 	ret = drm_dp_read_extended_dpcd_caps(aux, dpcd);
@@ -1209,11 +1269,9 @@ int drm_dp_read_downstream_info(struct drm_dp_aux *aux,
 	if (dpcd[DP_DOWNSTREAMPORT_PRESENT] & DP_DETAILED_CAP_INFO_AVAILABLE)
 		len *= 4;
 
-	ret = drm_dp_dpcd_read(aux, DP_DOWNSTREAM_PORT_0, downstream_ports, len);
+	ret = drm_dp_dpcd_read_data(aux, DP_DOWNSTREAM_PORT_0, downstream_ports, len);
 	if (ret < 0)
 		return ret;
-	if (ret != len)
-		return -EIO;
 
 	drm_dbg_kms(aux->drm_dev, "%s: DPCD DFP: %*ph\n", aux->name, len, downstream_ports);
 
@@ -1570,7 +1628,7 @@ EXPORT_SYMBOL(drm_dp_downstream_mode);
  */
 int drm_dp_downstream_id(struct drm_dp_aux *aux, char id[6])
 {
-	return drm_dp_dpcd_read(aux, DP_BRANCH_ID, id, 6);
+	return drm_dp_dpcd_read_data(aux, DP_BRANCH_ID, id, 6);
 }
 EXPORT_SYMBOL(drm_dp_downstream_id);
 
@@ -1635,13 +1693,13 @@ void drm_dp_downstream_debug(struct seq_file *m,
 	drm_dp_downstream_id(aux, id);
 	seq_printf(m, "\t\tID: %s\n", id);
 
-	len = drm_dp_dpcd_read(aux, DP_BRANCH_HW_REV, &rev[0], 1);
-	if (len > 0)
+	len = drm_dp_dpcd_read_data(aux, DP_BRANCH_HW_REV, &rev[0], 1);
+	if (!len)
 		seq_printf(m, "\t\tHW: %d.%d\n",
 			   (rev[0] & 0xf0) >> 4, rev[0] & 0xf);
 
-	len = drm_dp_dpcd_read(aux, DP_BRANCH_SW_REV, rev, 2);
-	if (len > 0)
+	len = drm_dp_dpcd_read_data(aux, DP_BRANCH_SW_REV, rev, 2);
+	if (!len)
 		seq_printf(m, "\t\tSW: %d.%d\n", rev[0], rev[1]);
 
 	if (detailed_cap_info) {
@@ -1779,11 +1837,9 @@ int drm_dp_read_sink_count(struct drm_dp_aux *aux)
 	u8 count;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_SINK_COUNT, &count);
+	ret = drm_dp_dpcd_read_byte(aux, DP_SINK_COUNT, &count);
 	if (ret < 0)
 		return ret;
-	if (ret != 1)
-		return -EIO;
 
 	return DP_GET_SINK_COUNT(count);
 }
@@ -2081,14 +2137,17 @@ static int drm_dp_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 
 	for (i = 0; i < num; i++) {
 		msg.address = msgs[i].addr;
-		drm_dp_i2c_msg_set_request(&msg, &msgs[i]);
-		/* Send a bare address packet to start the transaction.
-		 * Zero sized messages specify an address only (bare
-		 * address) transaction.
-		 */
-		msg.buffer = NULL;
-		msg.size = 0;
-		err = drm_dp_i2c_do_msg(aux, &msg);
+
+		if (!aux->no_zero_sized) {
+			drm_dp_i2c_msg_set_request(&msg, &msgs[i]);
+			/* Send a bare address packet to start the transaction.
+			 * Zero sized messages specify an address only (bare
+			 * address) transaction.
+			 */
+			msg.buffer = NULL;
+			msg.size = 0;
+			err = drm_dp_i2c_do_msg(aux, &msg);
+		}
 
 		/*
 		 * Reset msg.request in case in case it got
@@ -2107,6 +2166,8 @@ static int drm_dp_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 			msg.buffer = msgs[i].buf + j;
 			msg.size = min(transfer_size, msgs[i].len - j);
 
+			if (j + msg.size == msgs[i].len && aux->no_zero_sized)
+				msg.request &= ~DP_AUX_I2C_MOT;
 			err = drm_dp_i2c_drain_msg(aux, &msg);
 
 			/*
@@ -2124,15 +2185,17 @@ static int drm_dp_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 	}
 	if (err >= 0)
 		err = num;
-	/* Send a bare address packet to close out the transaction.
-	 * Zero sized messages specify an address only (bare
-	 * address) transaction.
-	 */
-	msg.request &= ~DP_AUX_I2C_MOT;
-	msg.buffer = NULL;
-	msg.size = 0;
-	(void)drm_dp_i2c_do_msg(aux, &msg);
 
+	if (!aux->no_zero_sized) {
+		/* Send a bare address packet to close out the transaction.
+		 * Zero sized messages specify an address only (bare
+		 * address) transaction.
+		 */
+		msg.request &= ~DP_AUX_I2C_MOT;
+		msg.buffer = NULL;
+		msg.size = 0;
+		(void)drm_dp_i2c_do_msg(aux, &msg);
+	}
 	return err;
 }
 
@@ -2172,13 +2235,13 @@ static int drm_dp_aux_get_crc(struct drm_dp_aux *aux, u8 *crc)
 	u8 buf, count;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_TEST_SINK, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_TEST_SINK, &buf);
 	if (ret < 0)
 		return ret;
 
 	WARN_ON(!(buf & DP_TEST_SINK_START));
 
-	ret = drm_dp_dpcd_readb(aux, DP_TEST_SINK_MISC, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_TEST_SINK_MISC, &buf);
 	if (ret < 0)
 		return ret;
 
@@ -2192,11 +2255,7 @@ static int drm_dp_aux_get_crc(struct drm_dp_aux *aux, u8 *crc)
 	 * At DP_TEST_CRC_R_CR, there's 6 bytes containing CRC data, 2 bytes
 	 * per component (RGB or CrYCb).
 	 */
-	ret = drm_dp_dpcd_read(aux, DP_TEST_CRC_R_CR, crc, 6);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_dpcd_read_data(aux, DP_TEST_CRC_R_CR, crc, 6);
 }
 
 static void drm_dp_aux_crc_work(struct work_struct *work)
@@ -2395,11 +2454,11 @@ int drm_dp_start_crc(struct drm_dp_aux *aux, struct drm_crtc *crtc)
 	u8 buf;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_TEST_SINK, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_TEST_SINK, &buf);
 	if (ret < 0)
 		return ret;
 
-	ret = drm_dp_dpcd_writeb(aux, DP_TEST_SINK, buf | DP_TEST_SINK_START);
+	ret = drm_dp_dpcd_write_byte(aux, DP_TEST_SINK, buf | DP_TEST_SINK_START);
 	if (ret < 0)
 		return ret;
 
@@ -2422,11 +2481,11 @@ int drm_dp_stop_crc(struct drm_dp_aux *aux)
 	u8 buf;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_TEST_SINK, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_TEST_SINK, &buf);
 	if (ret < 0)
 		return ret;
 
-	ret = drm_dp_dpcd_writeb(aux, DP_TEST_SINK, buf & ~DP_TEST_SINK_START);
+	ret = drm_dp_dpcd_write_byte(aux, DP_TEST_SINK, buf & ~DP_TEST_SINK_START);
 	if (ret < 0)
 		return ret;
 
@@ -2512,11 +2571,7 @@ drm_dp_get_quirks(const struct drm_dp_dpcd_ident *ident, bool is_branch)
 static int drm_dp_read_ident(struct drm_dp_aux *aux, unsigned int offset,
 			     struct drm_dp_dpcd_ident *ident)
 {
-	int ret;
-
-	ret = drm_dp_dpcd_read(aux, offset, ident, sizeof(*ident));
-
-	return ret < 0 ? ret : 0;
+	return drm_dp_dpcd_read_data(aux, offset, ident, sizeof(*ident));
 }
 
 static void drm_dp_dump_desc(struct drm_dp_aux *aux,
@@ -2774,13 +2829,11 @@ static int drm_dp_read_lttpr_regs(struct drm_dp_aux *aux,
 	int ret;
 
 	for (offset = 0; offset < buf_size; offset += block_size) {
-		ret = drm_dp_dpcd_read(aux,
-				       address + offset,
-				       &buf[offset], block_size);
+		ret = drm_dp_dpcd_read_data(aux,
+					    address + offset,
+					    &buf[offset], block_size);
 		if (ret < 0)
 			return ret;
-
-		WARN_ON(ret != block_size);
 	}
 
 	return 0;
@@ -2995,12 +3048,12 @@ int drm_dp_get_phy_test_pattern(struct drm_dp_aux *aux,
 	int err;
 	u8 rate, lanes;
 
-	err = drm_dp_dpcd_readb(aux, DP_TEST_LINK_RATE, &rate);
+	err = drm_dp_dpcd_read_byte(aux, DP_TEST_LINK_RATE, &rate);
 	if (err < 0)
 		return err;
 	data->link_rate = drm_dp_bw_code_to_link_rate(rate);
 
-	err = drm_dp_dpcd_readb(aux, DP_TEST_LANE_COUNT, &lanes);
+	err = drm_dp_dpcd_read_byte(aux, DP_TEST_LANE_COUNT, &lanes);
 	if (err < 0)
 		return err;
 	data->num_lanes = lanes & DP_MAX_LANE_COUNT_MASK;
@@ -3008,22 +3061,22 @@ int drm_dp_get_phy_test_pattern(struct drm_dp_aux *aux,
 	if (lanes & DP_ENHANCED_FRAME_CAP)
 		data->enhanced_frame_cap = true;
 
-	err = drm_dp_dpcd_readb(aux, DP_PHY_TEST_PATTERN, &data->phy_pattern);
+	err = drm_dp_dpcd_read_byte(aux, DP_PHY_TEST_PATTERN, &data->phy_pattern);
 	if (err < 0)
 		return err;
 
 	switch (data->phy_pattern) {
 	case DP_PHY_TEST_PATTERN_80BIT_CUSTOM:
-		err = drm_dp_dpcd_read(aux, DP_TEST_80BIT_CUSTOM_PATTERN_7_0,
-				       &data->custom80, sizeof(data->custom80));
+		err = drm_dp_dpcd_read_data(aux, DP_TEST_80BIT_CUSTOM_PATTERN_7_0,
+					    &data->custom80, sizeof(data->custom80));
 		if (err < 0)
 			return err;
 
 		break;
 	case DP_PHY_TEST_PATTERN_CP2520:
-		err = drm_dp_dpcd_read(aux, DP_TEST_HBR2_SCRAMBLER_RESET,
-				       &data->hbr2_reset,
-				       sizeof(data->hbr2_reset));
+		err = drm_dp_dpcd_read_data(aux, DP_TEST_HBR2_SCRAMBLER_RESET,
+					    &data->hbr2_reset,
+					    sizeof(data->hbr2_reset));
 		if (err < 0)
 			return err;
 	}
@@ -3050,15 +3103,15 @@ int drm_dp_set_phy_test_pattern(struct drm_dp_aux *aux,
 	if (dp_rev < 0x12) {
 		test_pattern = (test_pattern << 2) &
 			       DP_LINK_QUAL_PATTERN_11_MASK;
-		err = drm_dp_dpcd_writeb(aux, DP_TRAINING_PATTERN_SET,
-					 test_pattern);
+		err = drm_dp_dpcd_write_byte(aux, DP_TRAINING_PATTERN_SET,
+					     test_pattern);
 		if (err < 0)
 			return err;
 	} else {
 		for (i = 0; i < data->num_lanes; i++) {
-			err = drm_dp_dpcd_writeb(aux,
-						 DP_LINK_QUAL_LANE0_SET + i,
-						 test_pattern);
+			err = drm_dp_dpcd_write_byte(aux,
+						     DP_LINK_QUAL_LANE0_SET + i,
+						     test_pattern);
 			if (err < 0)
 				return err;
 		}
@@ -3265,8 +3318,8 @@ bool drm_dp_as_sdp_supported(struct drm_dp_aux *aux, const u8 dpcd[DP_RECEIVER_C
 	if (dpcd[DP_DPCD_REV] < DP_DPCD_REV_13)
 		return false;
 
-	if (drm_dp_dpcd_readb(aux, DP_DPRX_FEATURE_ENUMERATION_LIST_CONT_1,
-			      &rx_feature) != 1) {
+	if (drm_dp_dpcd_read_byte(aux, DP_DPRX_FEATURE_ENUMERATION_LIST_CONT_1,
+				  &rx_feature) < 0) {
 		drm_dbg_dp(aux->drm_dev,
 			   "Failed to read DP_DPRX_FEATURE_ENUMERATION_LIST_CONT_1\n");
 		return false;
@@ -3290,7 +3343,7 @@ bool drm_dp_vsc_sdp_supported(struct drm_dp_aux *aux, const u8 dpcd[DP_RECEIVER_
 	if (dpcd[DP_DPCD_REV] < DP_DPCD_REV_13)
 		return false;
 
-	if (drm_dp_dpcd_readb(aux, DP_DPRX_FEATURE_ENUMERATION_LIST, &rx_feature) != 1) {
+	if (drm_dp_dpcd_read_byte(aux, DP_DPRX_FEATURE_ENUMERATION_LIST, &rx_feature) < 0) {
 		drm_dbg_dp(aux->drm_dev, "failed to read DP_DPRX_FEATURE_ENUMERATION_LIST\n");
 		return false;
 	}
@@ -3421,16 +3474,13 @@ EXPORT_SYMBOL(drm_dp_get_pcon_max_frl_bw);
  */
 int drm_dp_pcon_frl_prepare(struct drm_dp_aux *aux, bool enable_frl_ready_hpd)
 {
-	int ret;
 	u8 buf = DP_PCON_ENABLE_SOURCE_CTL_MODE |
 		 DP_PCON_ENABLE_LINK_FRL_MODE;
 
 	if (enable_frl_ready_hpd)
 		buf |= DP_PCON_ENABLE_HPD_READY;
 
-	ret = drm_dp_dpcd_writeb(aux, DP_PCON_HDMI_LINK_CONFIG_1, buf);
-
-	return ret;
+	return drm_dp_dpcd_write_byte(aux, DP_PCON_HDMI_LINK_CONFIG_1, buf);
 }
 EXPORT_SYMBOL(drm_dp_pcon_frl_prepare);
 
@@ -3445,7 +3495,7 @@ bool drm_dp_pcon_is_frl_ready(struct drm_dp_aux *aux)
 	int ret;
 	u8 buf;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PCON_HDMI_TX_LINK_STATUS, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PCON_HDMI_TX_LINK_STATUS, &buf);
 	if (ret < 0)
 		return false;
 
@@ -3474,7 +3524,7 @@ int drm_dp_pcon_frl_configure_1(struct drm_dp_aux *aux, int max_frl_gbps,
 	int ret;
 	u8 buf;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PCON_HDMI_LINK_CONFIG_1, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PCON_HDMI_LINK_CONFIG_1, &buf);
 	if (ret < 0)
 		return ret;
 
@@ -3509,11 +3559,7 @@ int drm_dp_pcon_frl_configure_1(struct drm_dp_aux *aux, int max_frl_gbps,
 		return -EINVAL;
 	}
 
-	ret = drm_dp_dpcd_writeb(aux, DP_PCON_HDMI_LINK_CONFIG_1, buf);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_dpcd_write_byte(aux, DP_PCON_HDMI_LINK_CONFIG_1, buf);
 }
 EXPORT_SYMBOL(drm_dp_pcon_frl_configure_1);
 
@@ -3539,7 +3585,7 @@ int drm_dp_pcon_frl_configure_2(struct drm_dp_aux *aux, int max_frl_mask,
 	else
 		buf &= ~DP_PCON_FRL_LINK_TRAIN_EXTENDED;
 
-	ret = drm_dp_dpcd_writeb(aux, DP_PCON_HDMI_LINK_CONFIG_2, buf);
+	return drm_dp_dpcd_write_byte(aux, DP_PCON_HDMI_LINK_CONFIG_2, buf);
 	if (ret < 0)
 		return ret;
 
@@ -3555,13 +3601,7 @@ EXPORT_SYMBOL(drm_dp_pcon_frl_configure_2);
  */
 int drm_dp_pcon_reset_frl_config(struct drm_dp_aux *aux)
 {
-	int ret;
-
-	ret = drm_dp_dpcd_writeb(aux, DP_PCON_HDMI_LINK_CONFIG_1, 0x0);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_dpcd_write_byte(aux, DP_PCON_HDMI_LINK_CONFIG_1, 0x0);
 }
 EXPORT_SYMBOL(drm_dp_pcon_reset_frl_config);
 
@@ -3576,7 +3616,7 @@ int drm_dp_pcon_frl_enable(struct drm_dp_aux *aux)
 	int ret;
 	u8 buf = 0;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PCON_HDMI_LINK_CONFIG_1, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PCON_HDMI_LINK_CONFIG_1, &buf);
 	if (ret < 0)
 		return ret;
 	if (!(buf & DP_PCON_ENABLE_SOURCE_CTL_MODE)) {
@@ -3585,11 +3625,7 @@ int drm_dp_pcon_frl_enable(struct drm_dp_aux *aux)
 		return -EINVAL;
 	}
 	buf |= DP_PCON_ENABLE_HDMI_LINK;
-	ret = drm_dp_dpcd_writeb(aux, DP_PCON_HDMI_LINK_CONFIG_1, buf);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_dpcd_write_byte(aux, DP_PCON_HDMI_LINK_CONFIG_1, buf);
 }
 EXPORT_SYMBOL(drm_dp_pcon_frl_enable);
 
@@ -3604,7 +3640,7 @@ bool drm_dp_pcon_hdmi_link_active(struct drm_dp_aux *aux)
 	u8 buf;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PCON_HDMI_TX_LINK_STATUS, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PCON_HDMI_TX_LINK_STATUS, &buf);
 	if (ret < 0)
 		return false;
 
@@ -3629,7 +3665,7 @@ int drm_dp_pcon_hdmi_link_mode(struct drm_dp_aux *aux, u8 *frl_trained_mask)
 	int mode;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PCON_HDMI_POST_FRL_STATUS, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PCON_HDMI_POST_FRL_STATUS, &buf);
 	if (ret < 0)
 		return ret;
 
@@ -3658,7 +3694,7 @@ void drm_dp_pcon_hdmi_frl_link_error_count(struct drm_dp_aux *aux,
 	struct drm_hdmi_info *hdmi = &connector->display_info.hdmi;
 
 	for (i = 0; i < hdmi->max_lanes; i++) {
-		if (drm_dp_dpcd_readb(aux, DP_PCON_HDMI_ERROR_STATUS_LN0 + i, &buf) < 0)
+		if (drm_dp_dpcd_read_byte(aux, DP_PCON_HDMI_ERROR_STATUS_LN0 + i, &buf) < 0)
 			return;
 
 		error_count = buf & DP_PCON_HDMI_ERROR_COUNT_MASK;
@@ -3793,7 +3829,7 @@ int drm_dp_pcon_configure_dsc_enc(struct drm_dp_aux *aux, u8 pps_buf_config)
 	u8 buf;
 	int ret;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, &buf);
 	if (ret < 0)
 		return ret;
 
@@ -3804,11 +3840,7 @@ int drm_dp_pcon_configure_dsc_enc(struct drm_dp_aux *aux, u8 pps_buf_config)
 		buf |= pps_buf_config << 2;
 	}
 
-	ret = drm_dp_dpcd_writeb(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, buf);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_dpcd_write_byte(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, buf);
 }
 
 /**
@@ -3820,13 +3852,7 @@ int drm_dp_pcon_configure_dsc_enc(struct drm_dp_aux *aux, u8 pps_buf_config)
  */
 int drm_dp_pcon_pps_default(struct drm_dp_aux *aux)
 {
-	int ret;
-
-	ret = drm_dp_pcon_configure_dsc_enc(aux, DP_PCON_ENC_PPS_OVERRIDE_DISABLED);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_pcon_configure_dsc_enc(aux, DP_PCON_ENC_PPS_OVERRIDE_DISABLED);
 }
 EXPORT_SYMBOL(drm_dp_pcon_pps_default);
 
@@ -3842,15 +3868,11 @@ int drm_dp_pcon_pps_override_buf(struct drm_dp_aux *aux, u8 pps_buf[128])
 {
 	int ret;
 
-	ret = drm_dp_dpcd_write(aux, DP_PCON_HDMI_PPS_OVERRIDE_BASE, &pps_buf, 128);
+	ret = drm_dp_dpcd_write_data(aux, DP_PCON_HDMI_PPS_OVERRIDE_BASE, &pps_buf, 128);
 	if (ret < 0)
 		return ret;
 
-	ret = drm_dp_pcon_configure_dsc_enc(aux, DP_PCON_ENC_PPS_OVERRIDE_EN_BUFFER);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_pcon_configure_dsc_enc(aux, DP_PCON_ENC_PPS_OVERRIDE_EN_BUFFER);
 }
 EXPORT_SYMBOL(drm_dp_pcon_pps_override_buf);
 
@@ -3867,21 +3889,17 @@ int drm_dp_pcon_pps_override_param(struct drm_dp_aux *aux, u8 pps_param[6])
 {
 	int ret;
 
-	ret = drm_dp_dpcd_write(aux, DP_PCON_HDMI_PPS_OVRD_SLICE_HEIGHT, &pps_param[0], 2);
+	ret = drm_dp_dpcd_write_data(aux, DP_PCON_HDMI_PPS_OVRD_SLICE_HEIGHT, &pps_param[0], 2);
 	if (ret < 0)
 		return ret;
-	ret = drm_dp_dpcd_write(aux, DP_PCON_HDMI_PPS_OVRD_SLICE_WIDTH, &pps_param[2], 2);
+	ret = drm_dp_dpcd_write_data(aux, DP_PCON_HDMI_PPS_OVRD_SLICE_WIDTH, &pps_param[2], 2);
 	if (ret < 0)
 		return ret;
-	ret = drm_dp_dpcd_write(aux, DP_PCON_HDMI_PPS_OVRD_BPP, &pps_param[4], 2);
+	ret = drm_dp_dpcd_write_data(aux, DP_PCON_HDMI_PPS_OVRD_BPP, &pps_param[4], 2);
 	if (ret < 0)
 		return ret;
 
-	ret = drm_dp_pcon_configure_dsc_enc(aux, DP_PCON_ENC_PPS_OVERRIDE_EN_BUFFER);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_pcon_configure_dsc_enc(aux, DP_PCON_ENC_PPS_OVERRIDE_EN_BUFFER);
 }
 EXPORT_SYMBOL(drm_dp_pcon_pps_override_param);
 
@@ -3897,7 +3915,7 @@ int drm_dp_pcon_convert_rgb_to_ycbcr(struct drm_dp_aux *aux, u8 color_spc)
 	int ret;
 	u8 buf;
 
-	ret = drm_dp_dpcd_readb(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, &buf);
+	ret = drm_dp_dpcd_read_byte(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, &buf);
 	if (ret < 0)
 		return ret;
 
@@ -3906,11 +3924,7 @@ int drm_dp_pcon_convert_rgb_to_ycbcr(struct drm_dp_aux *aux, u8 color_spc)
 	else
 		buf &= ~DP_CONVERSION_RGB_YCBCR_MASK;
 
-	ret = drm_dp_dpcd_writeb(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, buf);
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	return drm_dp_dpcd_write_byte(aux, DP_PROTOCOL_CONVERTER_CONTROL_2, buf);
 }
 EXPORT_SYMBOL(drm_dp_pcon_convert_rgb_to_ycbcr);
 
@@ -3942,12 +3956,12 @@ int drm_edp_backlight_set_level(struct drm_dp_aux *aux, const struct drm_edp_bac
 		buf[0] = level;
 	}
 
-	ret = drm_dp_dpcd_write(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, buf, sizeof(buf));
-	if (ret != sizeof(buf)) {
+	ret = drm_dp_dpcd_write_data(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, buf, sizeof(buf));
+	if (ret < 0) {
 		drm_err(aux->drm_dev,
 			"%s: Failed to write aux backlight level: %d\n",
 			aux->name, ret);
-		return ret < 0 ? ret : -EIO;
+		return ret;
 	}
 
 	return 0;
@@ -3965,22 +3979,22 @@ drm_edp_backlight_set_enable(struct drm_dp_aux *aux, const struct drm_edp_backli
 	if (!bl->aux_enable)
 		return 0;
 
-	ret = drm_dp_dpcd_readb(aux, DP_EDP_DISPLAY_CONTROL_REGISTER, &buf);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_read_byte(aux, DP_EDP_DISPLAY_CONTROL_REGISTER, &buf);
+	if (ret < 0) {
 		drm_err(aux->drm_dev, "%s: Failed to read eDP display control register: %d\n",
 			aux->name, ret);
-		return ret < 0 ? ret : -EIO;
+		return ret;
 	}
 	if (enable)
 		buf |= DP_EDP_BACKLIGHT_ENABLE;
 	else
 		buf &= ~DP_EDP_BACKLIGHT_ENABLE;
 
-	ret = drm_dp_dpcd_writeb(aux, DP_EDP_DISPLAY_CONTROL_REGISTER, buf);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_write_byte(aux, DP_EDP_DISPLAY_CONTROL_REGISTER, buf);
+	if (ret < 0) {
 		drm_err(aux->drm_dev, "%s: Failed to write eDP display control register: %d\n",
 			aux->name, ret);
-		return ret < 0 ? ret : -EIO;
+		return ret;
 	}
 
 	return 0;
@@ -4016,15 +4030,16 @@ int drm_edp_backlight_enable(struct drm_dp_aux *aux, const struct drm_edp_backli
 		dpcd_buf = DP_EDP_BACKLIGHT_CONTROL_MODE_PWM;
 
 	if (bl->pwmgen_bit_count) {
-		ret = drm_dp_dpcd_writeb(aux, DP_EDP_PWMGEN_BIT_COUNT, bl->pwmgen_bit_count);
-		if (ret != 1)
+		ret = drm_dp_dpcd_write_byte(aux, DP_EDP_PWMGEN_BIT_COUNT, bl->pwmgen_bit_count);
+		if (ret < 0)
 			drm_dbg_kms(aux->drm_dev, "%s: Failed to write aux pwmgen bit count: %d\n",
 				    aux->name, ret);
 	}
 
 	if (bl->pwm_freq_pre_divider) {
-		ret = drm_dp_dpcd_writeb(aux, DP_EDP_BACKLIGHT_FREQ_SET, bl->pwm_freq_pre_divider);
-		if (ret != 1)
+		ret = drm_dp_dpcd_write_byte(aux, DP_EDP_BACKLIGHT_FREQ_SET,
+					     bl->pwm_freq_pre_divider);
+		if (ret < 0)
 			drm_dbg_kms(aux->drm_dev,
 				    "%s: Failed to write aux backlight frequency: %d\n",
 				    aux->name, ret);
@@ -4032,8 +4047,8 @@ int drm_edp_backlight_enable(struct drm_dp_aux *aux, const struct drm_edp_backli
 			dpcd_buf |= DP_EDP_BACKLIGHT_FREQ_AUX_SET_ENABLE;
 	}
 
-	ret = drm_dp_dpcd_writeb(aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, dpcd_buf);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_write_byte(aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, dpcd_buf);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "%s: Failed to write aux backlight mode: %d\n",
 			    aux->name, ret);
 		return ret < 0 ? ret : -EIO;
@@ -4088,8 +4103,8 @@ drm_edp_backlight_probe_max(struct drm_dp_aux *aux, struct drm_edp_backlight_inf
 	if (!bl->aux_set)
 		return 0;
 
-	ret = drm_dp_dpcd_readb(aux, DP_EDP_PWMGEN_BIT_COUNT, &pn);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_read_byte(aux, DP_EDP_PWMGEN_BIT_COUNT, &pn);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "%s: Failed to read pwmgen bit count cap: %d\n",
 			    aux->name, ret);
 		return -ENODEV;
@@ -4122,14 +4137,14 @@ drm_edp_backlight_probe_max(struct drm_dp_aux *aux, struct drm_edp_backlight_inf
 	 * - FxP is within 25% of desired value.
 	 *   Note: 25% is arbitrary value and may need some tweak.
 	 */
-	ret = drm_dp_dpcd_readb(aux, DP_EDP_PWMGEN_BIT_COUNT_CAP_MIN, &pn_min);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_read_byte(aux, DP_EDP_PWMGEN_BIT_COUNT_CAP_MIN, &pn_min);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "%s: Failed to read pwmgen bit count cap min: %d\n",
 			    aux->name, ret);
 		return 0;
 	}
-	ret = drm_dp_dpcd_readb(aux, DP_EDP_PWMGEN_BIT_COUNT_CAP_MAX, &pn_max);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_read_byte(aux, DP_EDP_PWMGEN_BIT_COUNT_CAP_MAX, &pn_max);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "%s: Failed to read pwmgen bit count cap max: %d\n",
 			    aux->name, ret);
 		return 0;
@@ -4154,8 +4169,8 @@ drm_edp_backlight_probe_max(struct drm_dp_aux *aux, struct drm_edp_backlight_inf
 			break;
 	}
 
-	ret = drm_dp_dpcd_writeb(aux, DP_EDP_PWMGEN_BIT_COUNT, pn);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_write_byte(aux, DP_EDP_PWMGEN_BIT_COUNT, pn);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "%s: Failed to write aux pwmgen bit count: %d\n",
 			    aux->name, ret);
 		return 0;
@@ -4180,8 +4195,8 @@ drm_edp_backlight_probe_state(struct drm_dp_aux *aux, struct drm_edp_backlight_i
 	u8 buf[2];
 	u8 mode_reg;
 
-	ret = drm_dp_dpcd_readb(aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, &mode_reg);
-	if (ret != 1) {
+	ret = drm_dp_dpcd_read_byte(aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, &mode_reg);
+	if (ret < 0) {
 		drm_dbg_kms(aux->drm_dev, "%s: Failed to read backlight mode: %d\n",
 			    aux->name, ret);
 		return ret < 0 ? ret : -EIO;
@@ -4194,11 +4209,11 @@ drm_edp_backlight_probe_state(struct drm_dp_aux *aux, struct drm_edp_backlight_i
 	if (*current_mode == DP_EDP_BACKLIGHT_CONTROL_MODE_DPCD) {
 		int size = 1 + bl->lsb_reg_used;
 
-		ret = drm_dp_dpcd_read(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, buf, size);
-		if (ret != size) {
+		ret = drm_dp_dpcd_read_data(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, buf, size);
+		if (ret < 0) {
 			drm_dbg_kms(aux->drm_dev, "%s: Failed to read backlight level: %d\n",
 				    aux->name, ret);
-			return ret < 0 ? ret : -EIO;
+			return ret;
 		}
 
 		if (bl->lsb_reg_used)
@@ -4343,8 +4358,8 @@ int drm_panel_dp_aux_backlight(struct drm_panel *panel, struct drm_dp_aux *aux)
 	if (!panel || !panel->dev || !aux)
 		return -EINVAL;
 
-	ret = drm_dp_dpcd_read(aux, DP_EDP_DPCD_REV, edp_dpcd,
-			       EDP_DISPLAY_CTL_CAP_SIZE);
+	ret = drm_dp_dpcd_read_data(aux, DP_EDP_DPCD_REV, edp_dpcd,
+				    EDP_DISPLAY_CTL_CAP_SIZE);
 	if (ret < 0)
 		return ret;
 
@@ -4385,8 +4400,9 @@ EXPORT_SYMBOL(drm_panel_dp_aux_backlight);
 #endif
 
 /* See DP Standard v2.1 2.6.4.4.1.1, 2.8.4.4, 2.8.7 */
-static int drm_dp_link_symbol_cycles(int lane_count, int pixels, int bpp_x16,
-				     int symbol_size, bool is_mst)
+static int drm_dp_link_data_symbol_cycles(int lane_count, int pixels,
+					  int bpp_x16, int symbol_size,
+					  bool is_mst)
 {
 	int cycles = DIV_ROUND_UP(pixels * bpp_x16, 16 * symbol_size * lane_count);
 	int align = is_mst ? 4 / lane_count : 1;
@@ -4394,22 +4410,42 @@ static int drm_dp_link_symbol_cycles(int lane_count, int pixels, int bpp_x16,
 	return ALIGN(cycles, align);
 }
 
-static int drm_dp_link_dsc_symbol_cycles(int lane_count, int pixels, int slice_count,
-					 int bpp_x16, int symbol_size, bool is_mst)
+/**
+ * drm_dp_link_symbol_cycles - calculate the link symbol count with/without dsc
+ * @lane_count: DP link lane count
+ * @pixels: number of pixels in a scanline
+ * @dsc_slice_count: number of slices for DSC or '0' for non-DSC
+ * @bpp_x16: bits per pixel in .4 binary fixed format
+ * @symbol_size: DP symbol size
+ * @is_mst: %true for MST and %false for SST
+ *
+ * Calculate the link symbol cycles for both DSC (@dsc_slice_count !=0) and
+ * non-DSC case (@dsc_slice_count == 0) and return the count.
+ */
+int drm_dp_link_symbol_cycles(int lane_count, int pixels, int dsc_slice_count,
+			      int bpp_x16, int symbol_size, bool is_mst)
 {
+	int slice_count = dsc_slice_count ? : 1;
 	int slice_pixels = DIV_ROUND_UP(pixels, slice_count);
-	int slice_data_cycles = drm_dp_link_symbol_cycles(lane_count, slice_pixels,
-							  bpp_x16, symbol_size, is_mst);
-	int slice_eoc_cycles = is_mst ? 4 / lane_count : 1;
+	int slice_data_cycles = drm_dp_link_data_symbol_cycles(lane_count,
+							       slice_pixels,
+							       bpp_x16,
+							       symbol_size,
+							       is_mst);
+	int slice_eoc_cycles = 0;
+
+	if (dsc_slice_count)
+		slice_eoc_cycles = is_mst ? 4 / lane_count : 1;
 
 	return slice_count * (slice_data_cycles + slice_eoc_cycles);
 }
+EXPORT_SYMBOL(drm_dp_link_symbol_cycles);
 
 /**
  * drm_dp_bw_overhead - Calculate the BW overhead of a DP link stream
  * @lane_count: DP link lane count
  * @hactive: pixel count of the active period in one scanline of the stream
- * @dsc_slice_count: DSC slice count if @flags/DRM_DP_LINK_BW_OVERHEAD_DSC is set
+ * @dsc_slice_count: number of slices for DSC or '0' for non-DSC
  * @bpp_x16: bits per pixel in .4 binary fixed point
  * @flags: DRM_DP_OVERHEAD_x flags
  *
@@ -4423,7 +4459,7 @@ static int drm_dp_link_dsc_symbol_cycles(int lane_count, int pixels, int slice_c
  * as well as the stream's
  * - @hactive timing
  * - @bpp_x16 color depth
- * - compression mode (@flags / %DRM_DP_OVERHEAD_DSC).
+ * - compression mode (@dsc_slice_count != 0)
  * Note that this overhead doesn't account for the 8b/10b, 128b/132b
  * channel coding efficiency, for that see
  * @drm_dp_link_bw_channel_coding_efficiency().
@@ -4478,15 +4514,10 @@ int drm_dp_bw_overhead(int lane_count, int hactive,
 	WARN_ON((flags & DRM_DP_BW_OVERHEAD_UHBR) &&
 		(flags & DRM_DP_BW_OVERHEAD_FEC));
 
-	if (flags & DRM_DP_BW_OVERHEAD_DSC)
-		symbol_cycles = drm_dp_link_dsc_symbol_cycles(lane_count, hactive,
-							      dsc_slice_count,
-							      bpp_x16, symbol_size,
-							      is_mst);
-	else
-		symbol_cycles = drm_dp_link_symbol_cycles(lane_count, hactive,
-							  bpp_x16, symbol_size,
-							  is_mst);
+	symbol_cycles = drm_dp_link_symbol_cycles(lane_count, hactive,
+						  dsc_slice_count,
+						  bpp_x16, symbol_size,
+						  is_mst);
 
 	return DIV_ROUND_UP_ULL(mul_u32_u32(symbol_cycles * symbol_size * lane_count,
 					    overhead * 16),
diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c
index 3a1f1ffc7b55..a89f38fd3218 100644
--- a/drivers/gpu/drm/display/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c
@@ -2192,24 +2192,20 @@ static int drm_dp_check_mstb_guid(struct drm_dp_mst_branch *mstb, guid_t *guid)
 	guid_copy(&mstb->guid, guid);
 
 	if (!drm_dp_validate_guid(mstb->mgr, &mstb->guid)) {
+		struct drm_dp_aux *aux;
 		u8 buf[UUID_SIZE];
 
 		export_guid(buf, &mstb->guid);
 
-		if (mstb->port_parent) {
-			ret = drm_dp_send_dpcd_write(mstb->mgr,
-						     mstb->port_parent,
-						     DP_GUID, sizeof(buf), buf);
-		} else {
-			ret = drm_dp_dpcd_write(mstb->mgr->aux,
-						DP_GUID, buf, sizeof(buf));
-		}
-	}
+		if (mstb->port_parent)
+			aux = &mstb->port_parent->aux;
+		else
+			aux = mstb->mgr->aux;
 
-	if (ret < 16 && ret > 0)
-		return -EPROTO;
+		ret = drm_dp_dpcd_write_data(aux, DP_GUID, buf, sizeof(buf));
+	}
 
-	return ret == 16 ? 0 : ret;
+	return ret;
 }
 
 static void build_mst_prop_path(const struct drm_dp_mst_branch *mstb,
@@ -2744,14 +2740,13 @@ retry:
 	do {
 		tosend = min3(mgr->max_dpcd_transaction_bytes, 16, total);
 
-		ret = drm_dp_dpcd_write(mgr->aux, regbase + offset,
-					&msg[offset],
-					tosend);
-		if (ret != tosend) {
-			if (ret == -EIO && retries < 5) {
-				retries++;
-				goto retry;
-			}
+		ret = drm_dp_dpcd_write_data(mgr->aux, regbase + offset,
+					     &msg[offset],
+					     tosend);
+		if (ret == -EIO && retries < 5) {
+			retries++;
+			goto retry;
+		} else if (ret < 0) {
 			drm_dbg_kms(mgr->dev, "failed to dpcd write %d %d\n", tosend, ret);
 
 			return -EIO;
@@ -3624,7 +3619,7 @@ enum drm_dp_mst_mode drm_dp_read_mst_cap(struct drm_dp_aux *aux,
 	if (dpcd[DP_DPCD_REV] < DP_DPCD_REV_12)
 		return DRM_DP_SST;
 
-	if (drm_dp_dpcd_readb(aux, DP_MSTM_CAP, &mstm_cap) != 1)
+	if (drm_dp_dpcd_read_byte(aux, DP_MSTM_CAP, &mstm_cap) < 0)
 		return DRM_DP_SST;
 
 	if (mstm_cap & DP_MST_CAP)
@@ -3679,10 +3674,10 @@ int drm_dp_mst_topology_mgr_set_mst(struct drm_dp_mst_topology_mgr *mgr, bool ms
 		mgr->mst_primary = mstb;
 		drm_dp_mst_topology_get_mstb(mgr->mst_primary);
 
-		ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
-					 DP_MST_EN |
-					 DP_UP_REQ_EN |
-					 DP_UPSTREAM_IS_SRC);
+		ret = drm_dp_dpcd_write_byte(mgr->aux, DP_MSTM_CTRL,
+					     DP_MST_EN |
+					     DP_UP_REQ_EN |
+					     DP_UPSTREAM_IS_SRC);
 		if (ret < 0)
 			goto out_unlock;
 
@@ -3697,7 +3692,7 @@ int drm_dp_mst_topology_mgr_set_mst(struct drm_dp_mst_topology_mgr *mgr, bool ms
 		mstb = mgr->mst_primary;
 		mgr->mst_primary = NULL;
 		/* this can fail if the device is gone */
-		drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL, 0);
+		drm_dp_dpcd_write_byte(mgr->aux, DP_MSTM_CTRL, 0);
 		ret = 0;
 		mgr->payload_id_table_cleared = false;
 
@@ -3763,8 +3758,8 @@ EXPORT_SYMBOL(drm_dp_mst_topology_queue_probe);
 void drm_dp_mst_topology_mgr_suspend(struct drm_dp_mst_topology_mgr *mgr)
 {
 	mutex_lock(&mgr->lock);
-	drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
-			   DP_MST_EN | DP_UPSTREAM_IS_SRC);
+	drm_dp_dpcd_write_byte(mgr->aux, DP_MSTM_CTRL,
+			       DP_MST_EN | DP_UPSTREAM_IS_SRC);
 	mutex_unlock(&mgr->lock);
 	flush_work(&mgr->up_req_work);
 	flush_work(&mgr->work);
@@ -3813,18 +3808,18 @@ int drm_dp_mst_topology_mgr_resume(struct drm_dp_mst_topology_mgr *mgr,
 		goto out_fail;
 	}
 
-	ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
-				 DP_MST_EN |
-				 DP_UP_REQ_EN |
-				 DP_UPSTREAM_IS_SRC);
+	ret = drm_dp_dpcd_write_byte(mgr->aux, DP_MSTM_CTRL,
+				     DP_MST_EN |
+				     DP_UP_REQ_EN |
+				     DP_UPSTREAM_IS_SRC);
 	if (ret < 0) {
 		drm_dbg_kms(mgr->dev, "mst write failed - undocked during suspend?\n");
 		goto out_fail;
 	}
 
 	/* Some hubs forget their guids after they resume */
-	ret = drm_dp_dpcd_read(mgr->aux, DP_GUID, buf, sizeof(buf));
-	if (ret != sizeof(buf)) {
+	ret = drm_dp_dpcd_read_data(mgr->aux, DP_GUID, buf, sizeof(buf));
+	if (ret < 0) {
 		drm_dbg_kms(mgr->dev, "dpcd read failed - undocked during suspend?\n");
 		goto out_fail;
 	}
@@ -3883,8 +3878,8 @@ drm_dp_get_one_sb_msg(struct drm_dp_mst_topology_mgr *mgr, bool up,
 		*mstb = NULL;
 
 	len = min(mgr->max_dpcd_transaction_bytes, 16);
-	ret = drm_dp_dpcd_read(mgr->aux, basereg, replyblock, len);
-	if (ret != len) {
+	ret = drm_dp_dpcd_read_data(mgr->aux, basereg, replyblock, len);
+	if (ret < 0) {
 		drm_dbg_kms(mgr->dev, "failed to read DPCD down rep %d %d\n", len, ret);
 		return false;
 	}
@@ -3922,9 +3917,9 @@ drm_dp_get_one_sb_msg(struct drm_dp_mst_topology_mgr *mgr, bool up,
 	curreply = len;
 	while (replylen > 0) {
 		len = min3(replylen, mgr->max_dpcd_transaction_bytes, 16);
-		ret = drm_dp_dpcd_read(mgr->aux, basereg + curreply,
-				    replyblock, len);
-		if (ret != len) {
+		ret = drm_dp_dpcd_read_data(mgr->aux, basereg + curreply,
+					    replyblock, len);
+		if (ret < 0) {
 			drm_dbg_kms(mgr->dev, "failed to read a chunk (len %d, ret %d)\n",
 				    len, ret);
 			return false;
@@ -4881,9 +4876,9 @@ static bool dump_dp_payload_table(struct drm_dp_mst_topology_mgr *mgr,
 	int i;
 
 	for (i = 0; i < DP_PAYLOAD_TABLE_SIZE; i += 16) {
-		if (drm_dp_dpcd_read(mgr->aux,
-				     DP_PAYLOAD_TABLE_UPDATE_STATUS + i,
-				     &buf[i], 16) != 16)
+		if (drm_dp_dpcd_read_data(mgr->aux,
+					  DP_PAYLOAD_TABLE_UPDATE_STATUS + i,
+					  &buf[i], 16) < 0)
 			return false;
 	}
 	return true;
@@ -4972,23 +4967,24 @@ void drm_dp_mst_dump_topology(struct seq_file *m,
 		}
 		seq_printf(m, "dpcd: %*ph\n", DP_RECEIVER_CAP_SIZE, buf);
 
-		ret = drm_dp_dpcd_read(mgr->aux, DP_FAUX_CAP, buf, 2);
-		if (ret != 2) {
+		ret = drm_dp_dpcd_read_data(mgr->aux, DP_FAUX_CAP, buf, 2);
+		if (ret < 0) {
 			seq_printf(m, "faux/mst read failed\n");
 			goto out;
 		}
 		seq_printf(m, "faux/mst: %*ph\n", 2, buf);
 
-		ret = drm_dp_dpcd_read(mgr->aux, DP_MSTM_CTRL, buf, 1);
-		if (ret != 1) {
+		ret = drm_dp_dpcd_read_data(mgr->aux, DP_MSTM_CTRL, buf, 1);
+		if (ret < 0) {
 			seq_printf(m, "mst ctrl read failed\n");
 			goto out;
 		}
 		seq_printf(m, "mst ctrl: %*ph\n", 1, buf);
 
 		/* dump the standard OUI branch header */
-		ret = drm_dp_dpcd_read(mgr->aux, DP_BRANCH_OUI, buf, DP_BRANCH_OUI_HEADER_SIZE);
-		if (ret != DP_BRANCH_OUI_HEADER_SIZE) {
+		ret = drm_dp_dpcd_read_data(mgr->aux, DP_BRANCH_OUI, buf,
+					    DP_BRANCH_OUI_HEADER_SIZE);
+		if (ret < 0) {
 			seq_printf(m, "branch oui read failed\n");
 			goto out;
 		}
@@ -6112,14 +6108,14 @@ struct drm_dp_aux *drm_dp_mst_dsc_aux_for_port(struct drm_dp_mst_port *port)
 
 	/* DP-to-DP peer device */
 	if (drm_dp_mst_is_virtual_dpcd(immediate_upstream_port)) {
-		if (drm_dp_dpcd_read(&port->aux,
-				     DP_DSC_SUPPORT, &endpoint_dsc, 1) != 1)
+		if (drm_dp_dpcd_read_data(&port->aux,
+					  DP_DSC_SUPPORT, &endpoint_dsc, 1) < 0)
 			return NULL;
-		if (drm_dp_dpcd_read(&port->aux,
-				     DP_FEC_CAPABILITY, &endpoint_fec, 1) != 1)
+		if (drm_dp_dpcd_read_data(&port->aux,
+					  DP_FEC_CAPABILITY, &endpoint_fec, 1) < 0)
 			return NULL;
-		if (drm_dp_dpcd_read(&immediate_upstream_port->aux,
-				     DP_DSC_SUPPORT, &upstream_dsc, 1) != 1)
+		if (drm_dp_dpcd_read_data(&immediate_upstream_port->aux,
+					  DP_DSC_SUPPORT, &upstream_dsc, 1) < 0)
 			return NULL;
 
 		/* Enpoint decompression with DP-to-DP peer device */
@@ -6157,8 +6153,8 @@ struct drm_dp_aux *drm_dp_mst_dsc_aux_for_port(struct drm_dp_mst_port *port)
 	if (drm_dp_has_quirk(&desc, DP_DPCD_QUIRK_DSC_WITHOUT_VIRTUAL_DPCD)) {
 		u8 dpcd_ext[DP_RECEIVER_CAP_SIZE];
 
-		if (drm_dp_dpcd_read(immediate_upstream_aux,
-				     DP_DSC_SUPPORT, &upstream_dsc, 1) != 1)
+		if (drm_dp_dpcd_read_data(immediate_upstream_aux,
+					  DP_DSC_SUPPORT, &upstream_dsc, 1) < 0)
 			return NULL;
 
 		if (!(upstream_dsc & DP_DSC_DECOMPRESSION_IS_SUPPORTED))
@@ -6180,11 +6176,11 @@ struct drm_dp_aux *drm_dp_mst_dsc_aux_for_port(struct drm_dp_mst_port *port)
 	 * therefore the endpoint needs to be
 	 * both DSC and FEC capable.
 	 */
-	if (drm_dp_dpcd_read(&port->aux,
-	   DP_DSC_SUPPORT, &endpoint_dsc, 1) != 1)
+	if (drm_dp_dpcd_read_data(&port->aux,
+				  DP_DSC_SUPPORT, &endpoint_dsc, 1) < 0)
 		return NULL;
-	if (drm_dp_dpcd_read(&port->aux,
-	   DP_FEC_CAPABILITY, &endpoint_fec, 1) != 1)
+	if (drm_dp_dpcd_read_data(&port->aux,
+				  DP_FEC_CAPABILITY, &endpoint_fec, 1) < 0)
 		return NULL;
 	if ((endpoint_dsc & DP_DSC_DECOMPRESSION_IS_SUPPORTED) &&
 	   (endpoint_fec & DP_FEC_CAPABLE))
diff --git a/drivers/gpu/drm/display/drm_dp_tunnel.c b/drivers/gpu/drm/display/drm_dp_tunnel.c
index 90fe07a89260..076edf161048 100644
--- a/drivers/gpu/drm/display/drm_dp_tunnel.c
+++ b/drivers/gpu/drm/display/drm_dp_tunnel.c
@@ -222,7 +222,7 @@ static int read_tunnel_regs(struct drm_dp_aux *aux, struct drm_dp_tunnel_regs *r
 	while ((len = next_reg_area(&offset))) {
 		int address = DP_TUNNELING_BASE + offset;
 
-		if (drm_dp_dpcd_read(aux, address, tunnel_reg_ptr(regs, address), len) < 0)
+		if (drm_dp_dpcd_read_data(aux, address, tunnel_reg_ptr(regs, address), len) < 0)
 			return -EIO;
 
 		offset += len;
@@ -913,7 +913,7 @@ static int set_bw_alloc_mode(struct drm_dp_tunnel *tunnel, bool enable)
 	u8 mask = DP_DISPLAY_DRIVER_BW_ALLOCATION_MODE_ENABLE | DP_UNMASK_BW_ALLOCATION_IRQ;
 	u8 val;
 
-	if (drm_dp_dpcd_readb(tunnel->aux, DP_DPTX_BW_ALLOCATION_MODE_CONTROL, &val) < 0)
+	if (drm_dp_dpcd_read_byte(tunnel->aux, DP_DPTX_BW_ALLOCATION_MODE_CONTROL, &val) < 0)
 		goto out_err;
 
 	if (enable)
@@ -921,7 +921,7 @@ static int set_bw_alloc_mode(struct drm_dp_tunnel *tunnel, bool enable)
 	else
 		val &= ~mask;
 
-	if (drm_dp_dpcd_writeb(tunnel->aux, DP_DPTX_BW_ALLOCATION_MODE_CONTROL, val) < 0)
+	if (drm_dp_dpcd_write_byte(tunnel->aux, DP_DPTX_BW_ALLOCATION_MODE_CONTROL, val) < 0)
 		goto out_err;
 
 	tunnel->bw_alloc_enabled = enable;
@@ -1039,7 +1039,7 @@ static int clear_bw_req_state(struct drm_dp_aux *aux)
 {
 	u8 bw_req_mask = DP_BW_REQUEST_SUCCEEDED | DP_BW_REQUEST_FAILED;
 
-	if (drm_dp_dpcd_writeb(aux, DP_TUNNELING_STATUS, bw_req_mask) < 0)
+	if (drm_dp_dpcd_write_byte(aux, DP_TUNNELING_STATUS, bw_req_mask) < 0)
 		return -EIO;
 
 	return 0;
@@ -1052,7 +1052,7 @@ static int bw_req_complete(struct drm_dp_aux *aux, bool *status_changed)
 	u8 val;
 	int err;
 
-	if (drm_dp_dpcd_readb(aux, DP_TUNNELING_STATUS, &val) < 0)
+	if (drm_dp_dpcd_read_byte(aux, DP_TUNNELING_STATUS, &val) < 0)
 		return -EIO;
 
 	*status_changed = val & status_change_mask;
@@ -1095,7 +1095,7 @@ static int allocate_tunnel_bw(struct drm_dp_tunnel *tunnel, int bw)
 	if (err)
 		goto out;
 
-	if (drm_dp_dpcd_writeb(tunnel->aux, DP_REQUEST_BW, request_bw) < 0) {
+	if (drm_dp_dpcd_write_byte(tunnel->aux, DP_REQUEST_BW, request_bw) < 0) {
 		err = -EIO;
 		goto out;
 	}
@@ -1196,13 +1196,13 @@ static int check_and_clear_status_change(struct drm_dp_tunnel *tunnel)
 	u8 mask = DP_BW_ALLOCATION_CAPABILITY_CHANGED | DP_ESTIMATED_BW_CHANGED;
 	u8 val;
 
-	if (drm_dp_dpcd_readb(tunnel->aux, DP_TUNNELING_STATUS, &val) < 0)
+	if (drm_dp_dpcd_read_byte(tunnel->aux, DP_TUNNELING_STATUS, &val) < 0)
 		goto out_err;
 
 	val &= mask;
 
 	if (val) {
-		if (drm_dp_dpcd_writeb(tunnel->aux, DP_TUNNELING_STATUS, val) < 0)
+		if (drm_dp_dpcd_write_byte(tunnel->aux, DP_TUNNELING_STATUS, val) < 0)
 			goto out_err;
 
 		return 1;
@@ -1215,7 +1215,7 @@ static int check_and_clear_status_change(struct drm_dp_tunnel *tunnel)
 	 * Check for estimated BW changes explicitly to account for lost
 	 * BW change notifications.
 	 */
-	if (drm_dp_dpcd_readb(tunnel->aux, DP_ESTIMATED_BW, &val) < 0)
+	if (drm_dp_dpcd_read_byte(tunnel->aux, DP_ESTIMATED_BW, &val) < 0)
 		goto out_err;
 
 	if (val * tunnel->bw_granularity != tunnel->estimated_bw)
@@ -1300,7 +1300,7 @@ int drm_dp_tunnel_handle_irq(struct drm_dp_tunnel_mgr *mgr, struct drm_dp_aux *a
 {
 	u8 val;
 
-	if (drm_dp_dpcd_readb(aux, DP_TUNNELING_STATUS, &val) < 0)
+	if (drm_dp_dpcd_read_byte(aux, DP_TUNNELING_STATUS, &val) < 0)
 		return -EIO;
 
 	if (val & (DP_BW_REQUEST_SUCCEEDED | DP_BW_REQUEST_FAILED))
diff --git a/drivers/gpu/drm/display/drm_hdmi_helper.c b/drivers/gpu/drm/display/drm_hdmi_helper.c
index 74dd4d01dd9b..855cb02b827d 100644
--- a/drivers/gpu/drm/display/drm_hdmi_helper.c
+++ b/drivers/gpu/drm/display/drm_hdmi_helper.c
@@ -256,3 +256,171 @@ drm_hdmi_compute_mode_clock(const struct drm_display_mode *mode,
 	return DIV_ROUND_CLOSEST_ULL(clock * bpc, 8);
 }
 EXPORT_SYMBOL(drm_hdmi_compute_mode_clock);
+
+struct drm_hdmi_acr_n_cts_entry {
+	unsigned int n;
+	unsigned int cts;
+};
+
+struct drm_hdmi_acr_data {
+	unsigned long tmds_clock_khz;
+	struct drm_hdmi_acr_n_cts_entry n_cts_32k,
+					n_cts_44k1,
+					n_cts_48k;
+};
+
+static const struct drm_hdmi_acr_data hdmi_acr_n_cts[] = {
+	{
+		/* "Other" entry */
+		.n_cts_32k =  { .n = 4096, },
+		.n_cts_44k1 = { .n = 6272, },
+		.n_cts_48k =  { .n = 6144, },
+	}, {
+		.tmds_clock_khz = 25175,
+		.n_cts_32k =  { .n = 4576,  .cts = 28125, },
+		.n_cts_44k1 = { .n = 7007,  .cts = 31250, },
+		.n_cts_48k =  { .n = 6864,  .cts = 28125, },
+	}, {
+		.tmds_clock_khz = 25200,
+		.n_cts_32k =  { .n = 4096,  .cts = 25200, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 28000, },
+		.n_cts_48k =  { .n = 6144,  .cts = 25200, },
+	}, {
+		.tmds_clock_khz = 27000,
+		.n_cts_32k =  { .n = 4096,  .cts = 27000, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 30000, },
+		.n_cts_48k =  { .n = 6144,  .cts = 27000, },
+	}, {
+		.tmds_clock_khz = 27027,
+		.n_cts_32k =  { .n = 4096,  .cts = 27027, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 30030, },
+		.n_cts_48k =  { .n = 6144,  .cts = 27027, },
+	}, {
+		.tmds_clock_khz = 54000,
+		.n_cts_32k =  { .n = 4096,  .cts = 54000, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 60000, },
+		.n_cts_48k =  { .n = 6144,  .cts = 54000, },
+	}, {
+		.tmds_clock_khz = 54054,
+		.n_cts_32k =  { .n = 4096,  .cts = 54054, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 60060, },
+		.n_cts_48k =  { .n = 6144,  .cts = 54054, },
+	}, {
+		.tmds_clock_khz = 74176,
+		.n_cts_32k =  { .n = 11648, .cts = 210937, }, /* and 210938 */
+		.n_cts_44k1 = { .n = 17836, .cts = 234375, },
+		.n_cts_48k =  { .n = 11648, .cts = 140625, },
+	}, {
+		.tmds_clock_khz = 74250,
+		.n_cts_32k =  { .n = 4096,  .cts = 74250, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 82500, },
+		.n_cts_48k =  { .n = 6144,  .cts = 74250, },
+	}, {
+		.tmds_clock_khz = 148352,
+		.n_cts_32k =  { .n = 11648, .cts = 421875, },
+		.n_cts_44k1 = { .n = 8918,  .cts = 234375, },
+		.n_cts_48k =  { .n = 5824,  .cts = 140625, },
+	}, {
+		.tmds_clock_khz = 148500,
+		.n_cts_32k =  { .n = 4096,  .cts = 148500, },
+		.n_cts_44k1 = { .n = 6272,  .cts = 165000, },
+		.n_cts_48k =  { .n = 6144,  .cts = 148500, },
+	}, {
+		.tmds_clock_khz = 296703,
+		.n_cts_32k =  { .n = 5824,  .cts = 421875, },
+		.n_cts_44k1 = { .n = 4459,  .cts = 234375, },
+		.n_cts_48k =  { .n = 5824,  .cts = 281250, },
+	}, {
+		.tmds_clock_khz = 297000,
+		.n_cts_32k =  { .n = 3072,  .cts = 222750, },
+		.n_cts_44k1 = { .n = 4704,  .cts = 247500, },
+		.n_cts_48k =  { .n = 5120,  .cts = 247500, },
+	}, {
+		.tmds_clock_khz = 593407,
+		.n_cts_32k =  { .n = 5824,  .cts = 843750, },
+		.n_cts_44k1 = { .n = 8918,  .cts = 937500, },
+		.n_cts_48k =  { .n = 5824,  .cts = 562500, },
+	}, {
+		.tmds_clock_khz = 594000,
+		.n_cts_32k =  { .n = 3072,  .cts = 445500, },
+		.n_cts_44k1 = { .n = 9408,  .cts = 990000, },
+		.n_cts_48k =  { .n = 6144,  .cts = 594000, },
+	},
+};
+
+static int drm_hdmi_acr_find_tmds_entry(unsigned long tmds_clock_khz)
+{
+	int i;
+
+	/* skip the "other" entry */
+	for (i = 1; i < ARRAY_SIZE(hdmi_acr_n_cts); i++) {
+		if (hdmi_acr_n_cts[i].tmds_clock_khz == tmds_clock_khz)
+			return i;
+	}
+
+	return 0;
+}
+
+/**
+ * drm_hdmi_acr_get_n_cts() - get N and CTS values for Audio Clock Regeneration
+ *
+ * @tmds_char_rate: TMDS clock (char rate) as used by the HDMI connector
+ * @sample_rate: audio sample rate
+ * @out_n: a pointer to write the N value
+ * @out_cts: a pointer to write the CTS value
+ *
+ * Get the N and CTS values (either by calculating them or by returning data
+ * from the tables. This follows the HDMI 1.4b Section 7.2 "Audio Sample Clock
+ * Capture and Regeneration".
+ *
+ * Note, @sample_rate corresponds to the Fs value, see sections 7.2.4 - 7.2.6
+ * on how to select Fs for non-L-PCM formats.
+ */
+void
+drm_hdmi_acr_get_n_cts(unsigned long long tmds_char_rate,
+		       unsigned int sample_rate,
+		       unsigned int *out_n,
+		       unsigned int *out_cts)
+{
+	/* be a bit more tolerant, especially for the 1.001 entries */
+	unsigned long tmds_clock_khz = DIV_ROUND_CLOSEST_ULL(tmds_char_rate, 1000);
+	const struct drm_hdmi_acr_n_cts_entry *entry;
+	unsigned int n, cts, mult;
+	int tmds_idx;
+
+	tmds_idx = drm_hdmi_acr_find_tmds_entry(tmds_clock_khz);
+
+	/*
+	 * Don't change the order, 192 kHz is divisible by 48k and 32k, but it
+	 * should use 48k entry.
+	 */
+	if (sample_rate % 48000 == 0) {
+		entry = &hdmi_acr_n_cts[tmds_idx].n_cts_48k;
+		mult = sample_rate / 48000;
+	} else if (sample_rate % 44100 == 0) {
+		entry = &hdmi_acr_n_cts[tmds_idx].n_cts_44k1;
+		mult = sample_rate / 44100;
+	} else if (sample_rate % 32000 == 0) {
+		entry = &hdmi_acr_n_cts[tmds_idx].n_cts_32k;
+		mult = sample_rate / 32000;
+	} else {
+		entry = NULL;
+	}
+
+	if (entry) {
+		n = entry->n * mult;
+		cts = entry->cts;
+	} else {
+		/* Recommended optimal value, HDMI 1.4b, Section 7.2.1 */
+		n = 128 * sample_rate / 1000;
+		cts = 0;
+	}
+
+	if (!cts)
+		cts = DIV_ROUND_CLOSEST_ULL(tmds_char_rate * n,
+					    128 * sample_rate);
+
+	*out_n = n;
+	*out_cts = cts;
+}
+EXPORT_SYMBOL(drm_hdmi_acr_get_n_cts);
diff --git a/drivers/gpu/drm/display/drm_hdmi_state_helper.c b/drivers/gpu/drm/display/drm_hdmi_state_helper.c
index c205f37da1e1..d9d9948b29e9 100644
--- a/drivers/gpu/drm/display/drm_hdmi_state_helper.c
+++ b/drivers/gpu/drm/display/drm_hdmi_state_helper.c
@@ -10,6 +10,298 @@
 #include <drm/display/drm_hdmi_state_helper.h>
 
 /**
+ * DOC: hdmi helpers
+ *
+ * These functions contain an implementation of the HDMI specification
+ * in the form of KMS helpers.
+ *
+ * It contains TMDS character rate computation, automatic selection of
+ * output formats, infoframes generation, etc.
+ *
+ * Infoframes Compliance
+ * ~~~~~~~~~~~~~~~~~~~~~
+ *
+ * Drivers using the helpers will expose the various infoframes
+ * generated according to the HDMI specification in debugfs.
+ *
+ * Compliance can then be tested using ``edid-decode`` from the ``v4l-utils`` project
+ * (https://git.linuxtv.org/v4l-utils.git/). A sample run would look like:
+ *
+ * .. code-block:: bash
+ *
+ *	# edid-decode \
+ *		-I /sys/kernel/debug/dri/1/HDMI-A-1/infoframes/audio \
+ *		-I /sys/kernel/debug/dri/1/HDMI-A-1/infoframes/avi \
+ *		-I /sys/kernel/debug/dri/1/HDMI-A-1/infoframes/hdmi \
+ *		-I /sys/kernel/debug/dri/1/HDMI-A-1/infoframes/hdr_drm \
+ *		-I /sys/kernel/debug/dri/1/HDMI-A-1/infoframes/spd \
+ *		/sys/class/drm/card1-HDMI-A-1/edid \
+ *		-c
+ *
+ *	edid-decode (hex):
+ *
+ *	00 ff ff ff ff ff ff 00 1e 6d f4 5b 1e ef 06 00
+ *	07 20 01 03 80 2f 34 78 ea 24 05 af 4f 42 ab 25
+ *	0f 50 54 21 08 00 d1 c0 61 40 45 40 01 01 01 01
+ *	01 01 01 01 01 01 98 d0 00 40 a1 40 d4 b0 30 20
+ *	3a 00 d1 0b 12 00 00 1a 00 00 00 fd 00 3b 3d 1e
+ *	b2 31 00 0a 20 20 20 20 20 20 00 00 00 fc 00 4c
+ *	47 20 53 44 51 48 44 0a 20 20 20 20 00 00 00 ff
+ *	00 32 30 37 4e 54 52 4c 44 43 34 33 30 0a 01 46
+ *
+ *	02 03 42 72 23 09 07 07 4d 01 03 04 90 12 13 1f
+ *	22 5d 5e 5f 60 61 83 01 00 00 6d 03 0c 00 10 00
+ *	b8 3c 20 00 60 01 02 03 67 d8 5d c4 01 78 80 03
+ *	e3 0f 00 18 e2 00 6a e3 05 c0 00 e6 06 05 01 52
+ *	52 51 11 5d 00 a0 a0 40 29 b0 30 20 3a 00 d1 0b
+ *	12 00 00 1a 00 00 00 00 00 00 00 00 00 00 00 00
+ *	00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
+ *	00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 c3
+ *
+ *	----------------
+ *
+ *	Block 0, Base EDID:
+ *	  EDID Structure Version & Revision: 1.3
+ *	  Vendor & Product Identification:
+ *	    Manufacturer: GSM
+ *	    Model: 23540
+ *	    Serial Number: 454430 (0x0006ef1e)
+ *	    Made in: week 7 of 2022
+ *	  Basic Display Parameters & Features:
+ *	    Digital display
+ *	    Maximum image size: 47 cm x 52 cm
+ *	    Gamma: 2.20
+ *	    DPMS levels: Standby Suspend Off
+ *	    RGB color display
+ *	    First detailed timing is the preferred timing
+ *	  Color Characteristics:
+ *	    Red  : 0.6835, 0.3105
+ *	    Green: 0.2587, 0.6679
+ *	    Blue : 0.1445, 0.0585
+ *	    White: 0.3134, 0.3291
+ *	  Established Timings I & II:
+ *	    DMT 0x04:   640x480    59.940476 Hz   4:3     31.469 kHz     25.175000 MHz
+ *	    DMT 0x09:   800x600    60.316541 Hz   4:3     37.879 kHz     40.000000 MHz
+ *	    DMT 0x10:  1024x768    60.003840 Hz   4:3     48.363 kHz     65.000000 MHz
+ *	  Standard Timings:
+ *	    DMT 0x52:  1920x1080   60.000000 Hz  16:9     67.500 kHz    148.500000 MHz
+ *	    DMT 0x10:  1024x768    60.003840 Hz   4:3     48.363 kHz     65.000000 MHz
+ *	    DMT 0x09:   800x600    60.316541 Hz   4:3     37.879 kHz     40.000000 MHz
+ *	  Detailed Timing Descriptors:
+ *	    DTD 1:  2560x2880   59.966580 Hz   8:9    185.417 kHz    534.000000 MHz (465 mm x 523 mm)
+ *	                 Hfront   48 Hsync  32 Hback  240 Hpol P
+ *	                 Vfront    3 Vsync  10 Vback  199 Vpol N
+ *	    Display Range Limits:
+ *	      Monitor ranges (GTF): 59-61 Hz V, 30-178 kHz H, max dotclock 490 MHz
+ *	    Display Product Name: 'LG SDQHD'
+ *	    Display Product Serial Number: '207NTRLDC430'
+ *	  Extension blocks: 1
+ *	Checksum: 0x46
+ *
+ *	----------------
+ *
+ *	Block 1, CTA-861 Extension Block:
+ *	  Revision: 3
+ *	  Basic audio support
+ *	  Supports YCbCr 4:4:4
+ *	  Supports YCbCr 4:2:2
+ *	  Native detailed modes: 2
+ *	  Audio Data Block:
+ *	    Linear PCM:
+ *	      Max channels: 2
+ *	      Supported sample rates (kHz): 48 44.1 32
+ *	      Supported sample sizes (bits): 24 20 16
+ *	  Video Data Block:
+ *	    VIC   1:   640x480    59.940476 Hz   4:3     31.469 kHz     25.175000 MHz
+ *	    VIC   3:   720x480    59.940060 Hz  16:9     31.469 kHz     27.000000 MHz
+ *	    VIC   4:  1280x720    60.000000 Hz  16:9     45.000 kHz     74.250000 MHz
+ *	    VIC  16:  1920x1080   60.000000 Hz  16:9     67.500 kHz    148.500000 MHz (native)
+ *	    VIC  18:   720x576    50.000000 Hz  16:9     31.250 kHz     27.000000 MHz
+ *	    VIC  19:  1280x720    50.000000 Hz  16:9     37.500 kHz     74.250000 MHz
+ *	    VIC  31:  1920x1080   50.000000 Hz  16:9     56.250 kHz    148.500000 MHz
+ *	    VIC  34:  1920x1080   30.000000 Hz  16:9     33.750 kHz     74.250000 MHz
+ *	    VIC  93:  3840x2160   24.000000 Hz  16:9     54.000 kHz    297.000000 MHz
+ *	    VIC  94:  3840x2160   25.000000 Hz  16:9     56.250 kHz    297.000000 MHz
+ *	    VIC  95:  3840x2160   30.000000 Hz  16:9     67.500 kHz    297.000000 MHz
+ *	    VIC  96:  3840x2160   50.000000 Hz  16:9    112.500 kHz    594.000000 MHz
+ *	    VIC  97:  3840x2160   60.000000 Hz  16:9    135.000 kHz    594.000000 MHz
+ *	  Speaker Allocation Data Block:
+ *	    FL/FR - Front Left/Right
+ *	  Vendor-Specific Data Block (HDMI), OUI 00-0C-03:
+ *	    Source physical address: 1.0.0.0
+ *	    Supports_AI
+ *	    DC_36bit
+ *	    DC_30bit
+ *	    DC_Y444
+ *	    Maximum TMDS clock: 300 MHz
+ *	    Extended HDMI video details:
+ *	      HDMI VICs:
+ *	        HDMI VIC 1:  3840x2160   30.000000 Hz  16:9     67.500 kHz    297.000000 MHz
+ *	        HDMI VIC 2:  3840x2160   25.000000 Hz  16:9     56.250 kHz    297.000000 MHz
+ *	        HDMI VIC 3:  3840x2160   24.000000 Hz  16:9     54.000 kHz    297.000000 MHz
+ *	  Vendor-Specific Data Block (HDMI Forum), OUI C4-5D-D8:
+ *	    Version: 1
+ *	    Maximum TMDS Character Rate: 600 MHz
+ *	    SCDC Present
+ *	    Supports 12-bits/component Deep Color 4:2:0 Pixel Encoding
+ *	    Supports 10-bits/component Deep Color 4:2:0 Pixel Encoding
+ *	  YCbCr 4:2:0 Capability Map Data Block:
+ *	    VIC  96:  3840x2160   50.000000 Hz  16:9    112.500 kHz    594.000000 MHz
+ *	    VIC  97:  3840x2160   60.000000 Hz  16:9    135.000 kHz    594.000000 MHz
+ *	  Video Capability Data Block:
+ *	    YCbCr quantization: No Data
+ *	    RGB quantization: Selectable (via AVI Q)
+ *	    PT scan behavior: Always Underscanned
+ *	    IT scan behavior: Always Underscanned
+ *	    CE scan behavior: Always Underscanned
+ *	  Colorimetry Data Block:
+ *	    BT2020YCC
+ *	    BT2020RGB
+ *	  HDR Static Metadata Data Block:
+ *	    Electro optical transfer functions:
+ *	      Traditional gamma - SDR luminance range
+ *	      SMPTE ST2084
+ *	    Supported static metadata descriptors:
+ *	      Static metadata type 1
+ *	    Desired content max luminance: 82 (295.365 cd/m^2)
+ *	    Desired content max frame-average luminance: 82 (295.365 cd/m^2)
+ *	    Desired content min luminance: 81 (0.298 cd/m^2)
+ *	  Detailed Timing Descriptors:
+ *	    DTD 2:  2560x2880   29.986961 Hz   8:9     87.592 kHz    238.250000 MHz (465 mm x 523 mm)
+ *	                 Hfront   48 Hsync  32 Hback   80 Hpol P
+ *	                 Vfront    3 Vsync  10 Vback   28 Vpol N
+ *	Checksum: 0xc3  Unused space in Extension Block: 43 bytes
+ *
+ *	----------------
+ *
+ *	edid-decode 1.29.0-5346
+ *	edid-decode SHA: c363e9aa6d70 2025-03-11 11:41:18
+ *
+ *	Warnings:
+ *
+ *	Block 1, CTA-861 Extension Block:
+ *	  IT Video Formats are overscanned by default, but normally this should be underscanned.
+ *	  Video Data Block: VIC 1 and the first DTD are not identical. Is this intended?
+ *	  Video Data Block: All VICs are in ascending order, and the first (preferred) VIC <= 4, is that intended?
+ *	  Video Capability Data Block: Set Selectable YCbCr Quantization to avoid interop issues.
+ *	  Video Capability Data Block: S_PT is equal to S_IT and S_CE, so should be set to 0 instead.
+ *	  Colorimetry Data Block: Set the sRGB colorimetry bit to avoid interop issues.
+ *	  Display Product Serial Number is set, so the Serial Number in the Base EDID should be 0.
+ *	EDID:
+ *	  Base EDID: Some timings are out of range of the Monitor Ranges:
+ *	    Vertical Freq: 24.000 - 60.317 Hz (Monitor: 59.000 - 61.000 Hz)
+ *	    Horizontal Freq: 31.250 - 185.416 kHz (Monitor: 30.000 - 178.000 kHz)
+ *	    Maximum Clock: 594.000 MHz (Monitor: 490.000 MHz)
+ *
+ *	Failures:
+ *
+ *	Block 1, CTA-861 Extension Block:
+ *	  Video Capability Data Block: IT video formats are always underscanned, but bit 7 of Byte 3 of the CTA-861 Extension header is set to overscanned.
+ *	EDID:
+ *	  CTA-861: Native progressive timings are a mix of several resolutions.
+ *
+ *	EDID conformity: FAIL
+ *
+ *	================
+ *
+ *	InfoFrame of '/sys/kernel/debug/dri/1/HDMI-A-1/infoframes/audio' was empty.
+ *
+ *	================
+ *
+ *	edid-decode InfoFrame (hex):
+ *
+ *	82 02 0d 31 12 28 04 00 00 00 00 00 00 00 00 00
+ *	00
+ *
+ *	----------------
+ *
+ *	HDMI InfoFrame Checksum: 0x31
+ *
+ *	AVI InfoFrame
+ *	  Version: 2
+ *	  Length: 13
+ *	  Y: Color Component Sample Format: RGB
+ *	  A: Active Format Information Present: Yes
+ *	  B: Bar Data Present: Bar Data not present
+ *	  S: Scan Information: Composed for an underscanned display
+ *	  C: Colorimetry: No Data
+ *	  M: Picture Aspect Ratio: 16:9
+ *	  R: Active Portion Aspect Ratio: 8
+ *	  ITC: IT Content: No Data
+ *	  EC: Extended Colorimetry: xvYCC601
+ *	  Q: RGB Quantization Range: Limited Range
+ *	  SC: Non-Uniform Picture Scaling: No Known non-uniform scaling
+ *	  YQ: YCC Quantization Range: Limited Range
+ *	  CN: IT Content Type: Graphics
+ *	  PR: Pixel Data Repetition Count: 0
+ *	  Line Number of End of Top Bar: 0
+ *	  Line Number of Start of Bottom Bar: 0
+ *	  Pixel Number of End of Left Bar: 0
+ *	  Pixel Number of Start of Right Bar: 0
+ *
+ *	----------------
+ *
+ *	AVI InfoFrame conformity: PASS
+ *
+ *	================
+ *
+ *	edid-decode InfoFrame (hex):
+ *
+ *	81 01 05 49 03 0c 00 20 01
+ *
+ *	----------------
+ *
+ *	HDMI InfoFrame Checksum: 0x49
+ *
+ *	Vendor-Specific InfoFrame (HDMI), OUI 00-0C-03
+ *	  Version: 1
+ *	  Length: 5
+ *	  HDMI Video Format: HDMI_VIC is present
+ *	  HDMI VIC 1:  3840x2160   30.000000 Hz  16:9     67.500 kHz    297.000000 MHz
+ *
+ *	----------------
+ *
+ *	Vendor-Specific InfoFrame (HDMI), OUI 00-0C-03 conformity: PASS
+ *
+ *	================
+ *
+ *	InfoFrame of '/sys/kernel/debug/dri/1/HDMI-A-1/infoframes/hdr_drm' was empty.
+ *
+ *	================
+ *
+ *	edid-decode InfoFrame (hex):
+ *
+ *	83 01 19 93 42 72 6f 61 64 63 6f 6d 56 69 64 65
+ *	6f 63 6f 72 65 00 00 00 00 00 00 00 09
+ *
+ *	----------------
+ *
+ *	HDMI InfoFrame Checksum: 0x93
+ *
+ *	Source Product Description InfoFrame
+ *	  Version: 1
+ *	  Length: 25
+ *	  Vendor Name: 'Broadcom'
+ *	  Product Description: 'Videocore'
+ *	  Source Information: PC general
+ *
+ *	----------------
+ *
+ *	Source Product Description InfoFrame conformity: PASS
+ *
+ * Testing
+ * ~~~~~~~
+ *
+ * The helpers have unit testing and can be tested using kunit with:
+ *
+ * .. code-block:: bash
+ *
+ *	$ ./tools/testing/kunit/kunit.py run \
+ *		--kunitconfig=drivers/gpu/drm/tests \
+ *		drm_atomic_helper_connector_hdmi_*
+ */
+
+/**
  * __drm_atomic_helper_connector_hdmi_reset() - Initializes all HDMI @drm_connector_state resources
  * @connector: DRM connector
  * @new_conn_state: connector state to reset
@@ -816,7 +1108,7 @@ drm_atomic_helper_connector_hdmi_update(struct drm_connector *connector,
  * @status: Connection status
  *
  * This function should be called as a part of the .detect() / .detect_ctx()
- * callbacks, updating the HDMI-specific connector's data.
+ * callbacks for all status changes.
  */
 void drm_atomic_helper_connector_hdmi_hotplug(struct drm_connector *connector,
 					      enum drm_connector_status status)
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 9ea2611770f4..0138cf0b8b63 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -933,6 +933,9 @@ EXPORT_SYMBOL(drm_atomic_get_new_private_obj_state);
  * state). This is especially true in enable hooks because the pipeline has
  * changed.
  *
+ * If you don't have access to the atomic state, see
+ * drm_atomic_get_connector_for_encoder().
+ *
  * Returns: The old connector connected to @encoder, or NULL if the encoder is
  * not connected.
  */
@@ -967,6 +970,9 @@ EXPORT_SYMBOL(drm_atomic_get_old_connector_for_encoder);
  * attached to @encoder vs ones that do (and to inspect their state). This is
  * especially true in disable hooks because the pipeline will change.
  *
+ * If you don't have access to the atomic state, see
+ * drm_atomic_get_connector_for_encoder().
+ *
  * Returns: The new connector connected to @encoder, or NULL if the encoder is
  * not connected.
  */
@@ -988,6 +994,59 @@ drm_atomic_get_new_connector_for_encoder(const struct drm_atomic_state *state,
 EXPORT_SYMBOL(drm_atomic_get_new_connector_for_encoder);
 
 /**
+ * drm_atomic_get_connector_for_encoder - Get connector currently assigned to an encoder
+ * @encoder: The encoder to find the connector of
+ * @ctx: Modeset locking context
+ *
+ * This function finds and returns the connector currently assigned to
+ * an @encoder.
+ *
+ * It is similar to the drm_atomic_get_old_connector_for_encoder() and
+ * drm_atomic_get_new_connector_for_encoder() helpers, but doesn't
+ * require access to the atomic state. If you have access to it, prefer
+ * using these. This helper is typically useful in situations where you
+ * don't have access to the atomic state, like detect, link repair,
+ * threaded interrupt handlers, or hooks from other frameworks (ALSA,
+ * CEC, etc.).
+ *
+ * Returns:
+ * The connector connected to @encoder, or an error pointer otherwise.
+ * When the error is EDEADLK, a deadlock has been detected and the
+ * sequence must be restarted.
+ */
+struct drm_connector *
+drm_atomic_get_connector_for_encoder(const struct drm_encoder *encoder,
+				     struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_connector_list_iter conn_iter;
+	struct drm_connector *out_connector = ERR_PTR(-EINVAL);
+	struct drm_connector *connector;
+	struct drm_device *dev = encoder->dev;
+	int ret;
+
+	ret = drm_modeset_lock(&dev->mode_config.connection_mutex, ctx);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_connector_list_iter_begin(dev, &conn_iter);
+	drm_for_each_connector_iter(connector, &conn_iter) {
+		if (!connector->state)
+			continue;
+
+		if (encoder == connector->state->best_encoder) {
+			out_connector = connector;
+			break;
+		}
+	}
+	drm_connector_list_iter_end(&conn_iter);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
+
+	return out_connector;
+}
+EXPORT_SYMBOL(drm_atomic_get_connector_for_encoder);
+
+
+/**
  * drm_atomic_get_old_crtc_for_encoder - Get old crtc for an encoder
  * @state: Atomic state
  * @encoder: The encoder to fetch the crtc state for
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 5302ab324898..ee64ca1b1bec 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -3409,6 +3409,9 @@ EXPORT_SYMBOL(drm_atomic_helper_disable_all);
  * This implies a reset of all active components available between the CRTC and
  * connectors.
  *
+ * A variant of this function exists with
+ * drm_bridge_helper_reset_crtc(), dedicated to bridges.
+ *
  * NOTE: This relies on resetting &drm_crtc_state.connectors_changed.
  * For drivers which optimize out unnecessary modesets this will result in
  * a no-op commit, achieving nothing.
diff --git a/drivers/gpu/drm/drm_blend.c b/drivers/gpu/drm/drm_blend.c
index 6e74de833466..6852d73c931c 100644
--- a/drivers/gpu/drm/drm_blend.c
+++ b/drivers/gpu/drm/drm_blend.c
@@ -75,6 +75,12 @@
  * 	the currently visible vertical area of the &drm_crtc.
  * FB_ID:
  * 	Mode object ID of the &drm_framebuffer this plane should scan out.
+ *
+ *	When a KMS client is performing front-buffer rendering, it should set
+ *	FB_ID to the same front-buffer FB on each atomic commit. This implies
+ *	to the driver that it needs to re-read the same FB again. Otherwise
+ *	drivers which do not employ continuously repeated scanout cycles might
+ *	not update the screen.
  * CRTC_ID:
  * 	Mode object ID of the &drm_crtc this plane should be connected to.
  *
diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c
index fa2794217a90..b4c89ec01998 100644
--- a/drivers/gpu/drm/drm_bridge.c
+++ b/drivers/gpu/drm/drm_bridge.c
@@ -21,6 +21,7 @@
  * DEALINGS IN THE SOFTWARE.
  */
 
+#include <linux/debugfs.h>
 #include <linux/err.h>
 #include <linux/media-bus-format.h>
 #include <linux/module.h>
@@ -198,10 +199,96 @@
 static DEFINE_MUTEX(bridge_lock);
 static LIST_HEAD(bridge_list);
 
+static void __drm_bridge_free(struct kref *kref)
+{
+	struct drm_bridge *bridge = container_of(kref, struct drm_bridge, refcount);
+
+	kfree(bridge->container);
+}
+
+/**
+ * drm_bridge_get - Acquire a bridge reference
+ * @bridge: DRM bridge
+ *
+ * This function increments the bridge's refcount.
+ *
+ * Returns:
+ * Pointer to @bridge.
+ */
+struct drm_bridge *drm_bridge_get(struct drm_bridge *bridge)
+{
+	if (bridge)
+		kref_get(&bridge->refcount);
+
+	return bridge;
+}
+EXPORT_SYMBOL(drm_bridge_get);
+
+/**
+ * drm_bridge_put - Release a bridge reference
+ * @bridge: DRM bridge
+ *
+ * This function decrements the bridge's reference count and frees the
+ * object if the reference count drops to zero.
+ */
+void drm_bridge_put(struct drm_bridge *bridge)
+{
+	if (bridge)
+		kref_put(&bridge->refcount, __drm_bridge_free);
+}
+EXPORT_SYMBOL(drm_bridge_put);
+
+/**
+ * drm_bridge_put_void - wrapper to drm_bridge_put() taking a void pointer
+ *
+ * @data: pointer to @struct drm_bridge, cast to a void pointer
+ *
+ * Wrapper of drm_bridge_put() to be used when a function taking a void
+ * pointer is needed, for example as a devm action.
+ */
+static void drm_bridge_put_void(void *data)
+{
+	struct drm_bridge *bridge = (struct drm_bridge *)data;
+
+	drm_bridge_put(bridge);
+}
+
+void *__devm_drm_bridge_alloc(struct device *dev, size_t size, size_t offset,
+			      const struct drm_bridge_funcs *funcs)
+{
+	void *container;
+	struct drm_bridge *bridge;
+	int err;
+
+	if (!funcs) {
+		dev_warn(dev, "Missing funcs pointer\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	container = kzalloc(size, GFP_KERNEL);
+	if (!container)
+		return ERR_PTR(-ENOMEM);
+
+	bridge = container + offset;
+	bridge->container = container;
+	bridge->funcs = funcs;
+	kref_init(&bridge->refcount);
+
+	err = devm_add_action_or_reset(dev, drm_bridge_put_void, bridge);
+	if (err)
+		return ERR_PTR(err);
+
+	return container;
+}
+EXPORT_SYMBOL(__devm_drm_bridge_alloc);
+
 /**
  * drm_bridge_add - add the given bridge to the global bridge list
  *
  * @bridge: bridge control structure
+ *
+ * The bridge to be added must have been allocated by
+ * devm_drm_bridge_alloc().
  */
 void drm_bridge_add(struct drm_bridge *bridge)
 {
@@ -280,6 +367,11 @@ static const struct drm_private_state_funcs drm_bridge_priv_state_funcs = {
 	.atomic_destroy_state = drm_bridge_atomic_destroy_priv_state,
 };
 
+static bool drm_bridge_is_atomic(struct drm_bridge *bridge)
+{
+	return bridge->funcs->atomic_reset != NULL;
+}
+
 /**
  * drm_bridge_attach - attach the bridge to an encoder's chain
  *
@@ -327,12 +419,12 @@ int drm_bridge_attach(struct drm_encoder *encoder, struct drm_bridge *bridge,
 		list_add(&bridge->chain_node, &encoder->bridge_chain);
 
 	if (bridge->funcs->attach) {
-		ret = bridge->funcs->attach(bridge, flags);
+		ret = bridge->funcs->attach(bridge, encoder, flags);
 		if (ret < 0)
 			goto err_reset_bridge;
 	}
 
-	if (bridge->funcs->atomic_reset) {
+	if (drm_bridge_is_atomic(bridge)) {
 		struct drm_bridge_state *state;
 
 		state = bridge->funcs->atomic_reset(bridge);
@@ -377,7 +469,7 @@ void drm_bridge_detach(struct drm_bridge *bridge)
 	if (WARN_ON(!bridge->dev))
 		return;
 
-	if (bridge->funcs->atomic_reset)
+	if (drm_bridge_is_atomic(bridge))
 		drm_atomic_private_obj_fini(&bridge->base);
 
 	if (bridge->funcs->detach)
@@ -1300,6 +1392,75 @@ struct drm_bridge *of_drm_find_bridge(struct device_node *np)
 EXPORT_SYMBOL(of_drm_find_bridge);
 #endif
 
+static void drm_bridge_debugfs_show_bridge(struct drm_printer *p,
+					   struct drm_bridge *bridge,
+					   unsigned int idx)
+{
+	drm_printf(p, "bridge[%u]: %ps\n", idx, bridge->funcs);
+	drm_printf(p, "\ttype: [%d] %s\n",
+		   bridge->type,
+		   drm_get_connector_type_name(bridge->type));
+
+	if (bridge->of_node)
+		drm_printf(p, "\tOF: %pOFfc\n", bridge->of_node);
+
+	drm_printf(p, "\tops: [0x%x]", bridge->ops);
+	if (bridge->ops & DRM_BRIDGE_OP_DETECT)
+		drm_puts(p, " detect");
+	if (bridge->ops & DRM_BRIDGE_OP_EDID)
+		drm_puts(p, " edid");
+	if (bridge->ops & DRM_BRIDGE_OP_HPD)
+		drm_puts(p, " hpd");
+	if (bridge->ops & DRM_BRIDGE_OP_MODES)
+		drm_puts(p, " modes");
+	if (bridge->ops & DRM_BRIDGE_OP_HDMI)
+		drm_puts(p, " hdmi");
+	drm_puts(p, "\n");
+}
+
+static int allbridges_show(struct seq_file *m, void *data)
+{
+	struct drm_printer p = drm_seq_file_printer(m);
+	struct drm_bridge *bridge;
+	unsigned int idx = 0;
+
+	mutex_lock(&bridge_lock);
+
+	list_for_each_entry(bridge, &bridge_list, list)
+		drm_bridge_debugfs_show_bridge(&p, bridge, idx++);
+
+	mutex_unlock(&bridge_lock);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(allbridges);
+
+static int encoder_bridges_show(struct seq_file *m, void *data)
+{
+	struct drm_encoder *encoder = m->private;
+	struct drm_printer p = drm_seq_file_printer(m);
+	struct drm_bridge *bridge;
+	unsigned int idx = 0;
+
+	drm_for_each_bridge_in_chain(encoder, bridge)
+		drm_bridge_debugfs_show_bridge(&p, bridge, idx++);
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(encoder_bridges);
+
+void drm_bridge_debugfs_params(struct dentry *root)
+{
+	debugfs_create_file("bridges", 0444, root, NULL, &allbridges_fops);
+}
+
+void drm_bridge_debugfs_encoder_params(struct dentry *root,
+				       struct drm_encoder *encoder)
+{
+	/* bridges list */
+	debugfs_create_file("bridges", 0444, root, encoder, &encoder_bridges_fops);
+}
+
 MODULE_AUTHOR("Ajay Kumar <ajaykumar.rs@samsung.com>");
 MODULE_DESCRIPTION("DRM bridge infrastructure");
 MODULE_LICENSE("GPL and additional rights");
diff --git a/drivers/gpu/drm/drm_bridge_helper.c b/drivers/gpu/drm/drm_bridge_helper.c
new file mode 100644
index 000000000000..af80d2496194
--- /dev/null
+++ b/drivers/gpu/drm/drm_bridge_helper.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_bridge_helper.h>
+#include <drm/drm_modeset_lock.h>
+
+/**
+ * drm_bridge_helper_reset_crtc - Reset the pipeline feeding a bridge
+ * @bridge: DRM bridge to reset
+ * @ctx: lock acquisition context
+ *
+ * Reset a @bridge pipeline. It will power-cycle all active components
+ * between the CRTC and connector that bridge is connected to.
+ *
+ * As it relies on drm_atomic_helper_reset_crtc(), the same limitations
+ * apply.
+ *
+ * Returns:
+ *
+ * 0 on success or a negative error code on failure. If the error
+ * returned is EDEADLK, the whole atomic sequence must be restarted.
+ */
+int drm_bridge_helper_reset_crtc(struct drm_bridge *bridge,
+				 struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_connector *connector;
+	struct drm_encoder *encoder = bridge->encoder;
+	struct drm_device *dev = encoder->dev;
+	struct drm_crtc *crtc;
+	int ret;
+
+	ret = drm_modeset_lock(&dev->mode_config.connection_mutex, ctx);
+	if (ret)
+		return ret;
+
+	connector = drm_atomic_get_connector_for_encoder(encoder, ctx);
+	if (IS_ERR(connector)) {
+		ret = PTR_ERR(connector);
+		goto out;
+	}
+
+	if (!connector->state) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	crtc = connector->state->crtc;
+	ret = drm_atomic_helper_reset_crtc(crtc, ctx);
+	if (ret)
+		goto out;
+
+out:
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
+	return ret;
+}
+EXPORT_SYMBOL(drm_bridge_helper_reset_crtc);
diff --git a/drivers/gpu/drm/drm_client.c b/drivers/gpu/drm/drm_client.c
index 549b28a5918c..f1de7faf9fb4 100644
--- a/drivers/gpu/drm/drm_client.c
+++ b/drivers/gpu/drm/drm_client.c
@@ -174,7 +174,7 @@ EXPORT_SYMBOL(drm_client_release);
 static void drm_client_buffer_delete(struct drm_client_buffer *buffer)
 {
 	if (buffer->gem) {
-		drm_gem_vunmap_unlocked(buffer->gem, &buffer->map);
+		drm_gem_vunmap(buffer->gem, &buffer->map);
 		drm_gem_object_put(buffer->gem);
 	}
 
@@ -252,7 +252,7 @@ int drm_client_buffer_vmap_local(struct drm_client_buffer *buffer,
 
 	drm_gem_lock(gem);
 
-	ret = drm_gem_vmap(gem, map);
+	ret = drm_gem_vmap_locked(gem, map);
 	if (ret)
 		goto err_drm_gem_vmap_unlocked;
 	*map_copy = *map;
@@ -278,7 +278,7 @@ void drm_client_buffer_vunmap_local(struct drm_client_buffer *buffer)
 	struct drm_gem_object *gem = buffer->gem;
 	struct iosys_map *map = &buffer->map;
 
-	drm_gem_vunmap(gem, map);
+	drm_gem_vunmap_locked(gem, map);
 	drm_gem_unlock(gem);
 }
 EXPORT_SYMBOL(drm_client_buffer_vunmap_local);
@@ -316,7 +316,7 @@ drm_client_buffer_vmap(struct drm_client_buffer *buffer,
 	ret = drm_gem_pin_locked(gem);
 	if (ret)
 		goto err_drm_gem_pin_locked;
-	ret = drm_gem_vmap(gem, map);
+	ret = drm_gem_vmap_locked(gem, map);
 	if (ret)
 		goto err_drm_gem_vmap;
 
@@ -348,7 +348,7 @@ void drm_client_buffer_vunmap(struct drm_client_buffer *buffer)
 	struct iosys_map *map = &buffer->map;
 
 	drm_gem_lock(gem);
-	drm_gem_vunmap(gem, map);
+	drm_gem_vunmap_locked(gem, map);
 	drm_gem_unpin_locked(gem);
 	drm_gem_unlock(gem);
 }
diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c
index aca442c25209..0f9d5ba36c81 100644
--- a/drivers/gpu/drm/drm_client_modeset.c
+++ b/drivers/gpu/drm/drm_client_modeset.c
@@ -39,7 +39,7 @@ int drm_client_modeset_create(struct drm_client_dev *client)
 	unsigned int max_connector_count = 1;
 	struct drm_mode_set *modeset;
 	struct drm_crtc *crtc;
-	unsigned int i = 0;
+	int i = 0;
 
 	/* Add terminating zero entry to enable index less iteration */
 	client->modesets = kcalloc(num_crtc + 1, sizeof(*client->modesets), GFP_KERNEL);
@@ -73,9 +73,10 @@ err_free:
 static void drm_client_modeset_release(struct drm_client_dev *client)
 {
 	struct drm_mode_set *modeset;
-	unsigned int i;
 
 	drm_client_for_each_modeset(modeset, client) {
+		int i;
+
 		drm_mode_destroy(client->dev, modeset->mode);
 		modeset->mode = NULL;
 		modeset->fb = NULL;
@@ -117,10 +118,10 @@ drm_client_find_modeset(struct drm_client_dev *client, struct drm_crtc *crtc)
 	return NULL;
 }
 
-static struct drm_display_mode *
+static const struct drm_display_mode *
 drm_connector_get_tiled_mode(struct drm_connector *connector)
 {
-	struct drm_display_mode *mode;
+	const struct drm_display_mode *mode;
 
 	list_for_each_entry(mode, &connector->modes, head) {
 		if (mode->hdisplay == connector->tile_h_size &&
@@ -130,10 +131,10 @@ drm_connector_get_tiled_mode(struct drm_connector *connector)
 	return NULL;
 }
 
-static struct drm_display_mode *
+static const struct drm_display_mode *
 drm_connector_fallback_non_tiled_mode(struct drm_connector *connector)
 {
-	struct drm_display_mode *mode;
+	const struct drm_display_mode *mode;
 
 	list_for_each_entry(mode, &connector->modes, head) {
 		if (mode->hdisplay == connector->tile_h_size &&
@@ -144,10 +145,10 @@ drm_connector_fallback_non_tiled_mode(struct drm_connector *connector)
 	return NULL;
 }
 
-static struct drm_display_mode *
+static const struct drm_display_mode *
 drm_connector_preferred_mode(struct drm_connector *connector, int width, int height)
 {
-	struct drm_display_mode *mode;
+	const struct drm_display_mode *mode;
 
 	list_for_each_entry(mode, &connector->modes, head) {
 		if (mode->hdisplay > width ||
@@ -159,16 +160,18 @@ drm_connector_preferred_mode(struct drm_connector *connector, int width, int hei
 	return NULL;
 }
 
-static struct drm_display_mode *drm_connector_first_mode(struct drm_connector *connector)
+static const struct drm_display_mode *
+drm_connector_first_mode(struct drm_connector *connector)
 {
 	return list_first_entry_or_null(&connector->modes,
 					struct drm_display_mode, head);
 }
 
-static struct drm_display_mode *drm_connector_pick_cmdline_mode(struct drm_connector *connector)
+static const struct drm_display_mode *
+drm_connector_pick_cmdline_mode(struct drm_connector *connector)
 {
-	struct drm_cmdline_mode *cmdline_mode;
-	struct drm_display_mode *mode;
+	const struct drm_cmdline_mode *cmdline_mode;
+	const struct drm_display_mode *mode;
 	bool prefer_non_interlace;
 
 	/*
@@ -237,9 +240,9 @@ static bool drm_connector_enabled(struct drm_connector *connector, bool strict)
 	return enable;
 }
 
-static void drm_client_connectors_enabled(struct drm_connector **connectors,
+static void drm_client_connectors_enabled(struct drm_connector *connectors[],
 					  unsigned int connector_count,
-					  bool *enabled)
+					  bool enabled[])
 {
 	bool any_enabled = false;
 	struct drm_connector *connector;
@@ -263,16 +266,35 @@ static void drm_client_connectors_enabled(struct drm_connector **connectors,
 		enabled[i] = drm_connector_enabled(connectors[i], false);
 }
 
+static void mode_replace(struct drm_device *dev,
+			 const struct drm_display_mode **dst,
+			 const struct drm_display_mode *src)
+{
+	drm_mode_destroy(dev, (struct drm_display_mode *)*dst);
+
+	*dst = src ? drm_mode_duplicate(dev, src) : NULL;
+}
+
+static void modes_destroy(struct drm_device *dev,
+			  const struct drm_display_mode *modes[],
+			  int count)
+{
+	int i;
+
+	for (i = 0; i < count; i++)
+		mode_replace(dev, &modes[i], NULL);
+}
+
 static bool drm_client_target_cloned(struct drm_device *dev,
-				     struct drm_connector **connectors,
+				     struct drm_connector *connectors[],
 				     unsigned int connector_count,
-				     struct drm_display_mode **modes,
-				     struct drm_client_offset *offsets,
-				     bool *enabled, int width, int height)
+				     const struct drm_display_mode *modes[],
+				     struct drm_client_offset offsets[],
+				     bool enabled[], int width, int height)
 {
-	int count, i, j;
+	int count, i;
 	bool can_clone = false;
-	struct drm_display_mode *dmt_mode, *mode;
+	struct drm_display_mode *dmt_mode;
 
 	/* only contemplate cloning in the single crtc case */
 	if (dev->mode_config.num_crtc > 1)
@@ -291,9 +313,13 @@ static bool drm_client_target_cloned(struct drm_device *dev,
 	/* check the command line or if nothing common pick 1024x768 */
 	can_clone = true;
 	for (i = 0; i < connector_count; i++) {
+		int j;
+
 		if (!enabled[i])
 			continue;
-		modes[i] = drm_connector_pick_cmdline_mode(connectors[i]);
+
+		mode_replace(dev, &modes[i],
+			     drm_connector_pick_cmdline_mode(connectors[i]));
 		if (!modes[i]) {
 			can_clone = false;
 			break;
@@ -323,6 +349,8 @@ static bool drm_client_target_cloned(struct drm_device *dev,
 		goto fail;
 
 	for (i = 0; i < connector_count; i++) {
+		const struct drm_display_mode *mode;
+
 		if (!enabled[i])
 			continue;
 
@@ -332,7 +360,7 @@ static bool drm_client_target_cloned(struct drm_device *dev,
 					   DRM_MODE_MATCH_CLOCK |
 					   DRM_MODE_MATCH_FLAGS |
 					   DRM_MODE_MATCH_3D_FLAGS))
-				modes[i] = mode;
+				mode_replace(dev, &modes[i], mode);
 		}
 		if (!modes[i])
 			can_clone = false;
@@ -349,19 +377,19 @@ fail:
 }
 
 static int drm_client_get_tile_offsets(struct drm_device *dev,
-				       struct drm_connector **connectors,
+				       struct drm_connector *connectors[],
 				       unsigned int connector_count,
-				       struct drm_display_mode **modes,
-				       struct drm_client_offset *offsets,
+				       const struct drm_display_mode *modes[],
+				       struct drm_client_offset offsets[],
 				       int idx,
 				       int h_idx, int v_idx)
 {
-	struct drm_connector *connector;
 	int i;
 	int hoffset = 0, voffset = 0;
 
 	for (i = 0; i < connector_count; i++) {
-		connector = connectors[i];
+		struct drm_connector *connector = connectors[i];
+
 		if (!connector->has_tile)
 			continue;
 
@@ -384,14 +412,13 @@ static int drm_client_get_tile_offsets(struct drm_device *dev,
 }
 
 static bool drm_client_target_preferred(struct drm_device *dev,
-					struct drm_connector **connectors,
+					struct drm_connector *connectors[],
 					unsigned int connector_count,
-					struct drm_display_mode **modes,
-					struct drm_client_offset *offsets,
-					bool *enabled, int width, int height)
+					const struct drm_display_mode *modes[],
+					struct drm_client_offset offsets[],
+					bool enabled[], int width, int height)
 {
 	const u64 mask = BIT_ULL(connector_count) - 1;
-	struct drm_connector *connector;
 	u64 conn_configured = 0;
 	int tile_pass = 0;
 	int num_tiled_conns = 0;
@@ -405,7 +432,9 @@ static bool drm_client_target_preferred(struct drm_device *dev,
 
 retry:
 	for (i = 0; i < connector_count; i++) {
-		connector = connectors[i];
+		struct drm_connector *connector = connectors[i];
+		const char *mode_type;
+
 
 		if (conn_configured & BIT_ULL(i))
 			continue;
@@ -438,20 +467,23 @@ retry:
 						    modes, offsets, i,
 						    connector->tile_h_loc, connector->tile_v_loc);
 		}
-		drm_dbg_kms(dev, "[CONNECTOR:%d:%s] looking for cmdline mode\n",
-			    connector->base.id, connector->name);
 
-		/* got for command line mode first */
-		modes[i] = drm_connector_pick_cmdline_mode(connector);
+		mode_type = "cmdline";
+		mode_replace(dev, &modes[i],
+			     drm_connector_pick_cmdline_mode(connector));
+
 		if (!modes[i]) {
-			drm_dbg_kms(dev, "[CONNECTOR:%d:%s] looking for preferred mode, tile %d\n",
-				    connector->base.id, connector->name,
-				    connector->tile_group ? connector->tile_group->id : 0);
-			modes[i] = drm_connector_preferred_mode(connector, width, height);
+			mode_type = "preferred";
+			mode_replace(dev, &modes[i],
+				     drm_connector_preferred_mode(connector, width, height));
 		}
-		/* No preferred modes, pick one off the list */
-		if (!modes[i])
-			modes[i] = drm_connector_first_mode(connector);
+
+		if (!modes[i]) {
+			mode_type = "first";
+			mode_replace(dev, &modes[i],
+				     drm_connector_first_mode(connector));
+		}
+
 		/*
 		 * In case of tiled mode if all tiles not present fallback to
 		 * first available non tiled mode.
@@ -466,18 +498,24 @@ retry:
 			    (connector->tile_h_loc == 0 &&
 			     connector->tile_v_loc == 0 &&
 			     !drm_connector_get_tiled_mode(connector))) {
-				drm_dbg_kms(dev,
-					    "[CONNECTOR:%d:%s] Falling back to non-tiled mode\n",
-					    connector->base.id, connector->name);
-				modes[i] = drm_connector_fallback_non_tiled_mode(connector);
+				mode_type = "non tiled";
+				mode_replace(dev, &modes[i],
+					     drm_connector_fallback_non_tiled_mode(connector));
 			} else {
-				modes[i] = drm_connector_get_tiled_mode(connector);
+				mode_type = "tiled";
+				mode_replace(dev, &modes[i],
+					     drm_connector_get_tiled_mode(connector));
 			}
 		}
 
-		drm_dbg_kms(dev, "[CONNECTOR:%d:%s] Found mode %s\n",
-			    connector->base.id, connector->name,
-			    modes[i] ? modes[i]->name : "none");
+		if (modes[i])
+			drm_dbg_kms(dev, "[CONNECTOR:%d:%s] found %s mode: %s\n",
+				    connector->base.id, connector->name,
+				    mode_type, modes[i]->name);
+		else
+			drm_dbg_kms(dev, "[CONNECTOR:%d:%s] no mode found\n",
+				    connector->base.id, connector->name);
+
 		conn_configured |= BIT_ULL(i);
 	}
 
@@ -502,18 +540,17 @@ static bool connector_has_possible_crtc(struct drm_connector *connector,
 }
 
 static int drm_client_pick_crtcs(struct drm_client_dev *client,
-				 struct drm_connector **connectors,
+				 struct drm_connector *connectors[],
 				 unsigned int connector_count,
-				 struct drm_crtc **best_crtcs,
-				 struct drm_display_mode **modes,
+				 struct drm_crtc *best_crtcs[],
+				 const struct drm_display_mode *modes[],
 				 int n, int width, int height)
 {
 	struct drm_device *dev = client->dev;
 	struct drm_connector *connector;
 	int my_score, best_score, score;
-	struct drm_crtc **crtcs, *crtc;
+	struct drm_crtc **crtcs;
 	struct drm_mode_set *modeset;
-	int o;
 
 	if (n == connector_count)
 		return 0;
@@ -543,7 +580,8 @@ static int drm_client_pick_crtcs(struct drm_client_dev *client,
 	 * remaining connectors
 	 */
 	drm_client_for_each_modeset(modeset, client) {
-		crtc = modeset->crtc;
+		struct drm_crtc *crtc = modeset->crtc;
+		int o;
 
 		if (!connector_has_possible_crtc(connector, crtc))
 			continue;
@@ -577,17 +615,17 @@ static int drm_client_pick_crtcs(struct drm_client_dev *client,
 
 /* Try to read the BIOS display configuration and use it for the initial config */
 static bool drm_client_firmware_config(struct drm_client_dev *client,
-				       struct drm_connector **connectors,
+				       struct drm_connector *connectors[],
 				       unsigned int connector_count,
-				       struct drm_crtc **crtcs,
-				       struct drm_display_mode **modes,
-				       struct drm_client_offset *offsets,
-				       bool *enabled, int width, int height)
+				       struct drm_crtc *crtcs[],
+				       const struct drm_display_mode *modes[],
+				       struct drm_client_offset offsets[],
+				       bool enabled[], int width, int height)
 {
 	const int count = min_t(unsigned int, connector_count, BITS_PER_LONG);
 	unsigned long conn_configured, conn_seq, mask;
 	struct drm_device *dev = client->dev;
-	int i, j;
+	int i;
 	bool *save_enabled;
 	bool fallback = true, ret = true;
 	int num_connectors_enabled = 0;
@@ -621,11 +659,11 @@ static bool drm_client_firmware_config(struct drm_client_dev *client,
 retry:
 	conn_seq = conn_configured;
 	for (i = 0; i < count; i++) {
-		struct drm_connector *connector;
+		struct drm_connector *connector = connectors[i];
 		struct drm_encoder *encoder;
-		struct drm_crtc *new_crtc;
-
-		connector = connectors[i];
+		struct drm_crtc *crtc;
+		const char *mode_type;
+		int j;
 
 		if (conn_configured & BIT(i))
 			continue;
@@ -664,7 +702,7 @@ retry:
 
 		num_connectors_enabled++;
 
-		new_crtc = connector->state->crtc;
+		crtc = connector->state->crtc;
 
 		/*
 		 * Make sure we're not trying to drive multiple connectors
@@ -672,69 +710,52 @@ retry:
 		 * match the BIOS.
 		 */
 		for (j = 0; j < count; j++) {
-			if (crtcs[j] == new_crtc) {
-				drm_dbg_kms(dev, "fallback: cloned configuration\n");
+			if (crtcs[j] == crtc) {
+				drm_dbg_kms(dev, "[CONNECTOR:%d:%s] fallback: cloned configuration\n",
+					    connector->base.id, connector->name);
 				goto bail;
 			}
 		}
 
-		drm_dbg_kms(dev, "[CONNECTOR:%d:%s] looking for cmdline mode\n",
-			    connector->base.id, connector->name);
-
-		/* go for command line mode first */
-		modes[i] = drm_connector_pick_cmdline_mode(connector);
+		mode_type = "cmdline";
+		mode_replace(dev, &modes[i],
+			     drm_connector_pick_cmdline_mode(connector));
 
-		/* try for preferred next */
 		if (!modes[i]) {
-			drm_dbg_kms(dev,
-				    "[CONNECTOR:%d:%s] looking for preferred mode, has tile: %s\n",
-				    connector->base.id, connector->name,
-				    str_yes_no(connector->has_tile));
-			modes[i] = drm_connector_preferred_mode(connector, width, height);
+			mode_type = "preferred";
+			mode_replace(dev, &modes[i],
+				     drm_connector_preferred_mode(connector, width, height));
 		}
 
-		/* No preferred mode marked by the EDID? Are there any modes? */
-		if (!modes[i] && !list_empty(&connector->modes)) {
-			drm_dbg_kms(dev, "[CONNECTOR:%d:%s] using first listed mode\n",
-				    connector->base.id, connector->name);
-			modes[i] = drm_connector_first_mode(connector);
+		if (!modes[i]) {
+			mode_type = "first";
+			mode_replace(dev, &modes[i],
+				     drm_connector_first_mode(connector));
 		}
 
 		/* last resort: use current mode */
 		if (!modes[i]) {
-			/*
-			 * IMPORTANT: We want to use the adjusted mode (i.e.
-			 * after the panel fitter upscaling) as the initial
-			 * config, not the input mode, which is what crtc->mode
-			 * usually contains. But since our current
-			 * code puts a mode derived from the post-pfit timings
-			 * into crtc->mode this works out correctly.
-			 *
-			 * This is crtc->mode and not crtc->state->mode for the
-			 * fastboot check to work correctly.
-			 */
-			drm_dbg_kms(dev, "[CONNECTOR:%d:%s] looking for current mode\n",
-				    connector->base.id, connector->name);
-			modes[i] = &connector->state->crtc->mode;
+			mode_type = "current";
+			mode_replace(dev, &modes[i],
+				     &crtc->state->mode);
 		}
+
 		/*
 		 * In case of tiled modes, if all tiles are not present
 		 * then fallback to a non tiled mode.
 		 */
 		if (connector->has_tile &&
 		    num_tiled_conns < connector->num_h_tile * connector->num_v_tile) {
-			drm_dbg_kms(dev, "[CONNECTOR:%d:%s] Falling back to non-tiled mode\n",
-				    connector->base.id, connector->name);
-			modes[i] = drm_connector_fallback_non_tiled_mode(connector);
+			mode_type = "non tiled";
+			mode_replace(dev, &modes[i],
+				     drm_connector_fallback_non_tiled_mode(connector));
 		}
-		crtcs[i] = new_crtc;
+		crtcs[i] = crtc;
 
-		drm_dbg_kms(dev, "[CONNECTOR:%d:%s] on [CRTC:%d:%s]: %dx%d%s\n",
+		drm_dbg_kms(dev, "[CONNECTOR::%d:%s] on [CRTC:%d:%s] using %s mode: %s\n",
 			    connector->base.id, connector->name,
-			    connector->state->crtc->base.id,
-			    connector->state->crtc->name,
-			    modes[i]->hdisplay, modes[i]->vdisplay,
-			    modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" : "");
+			    crtc->base.id, crtc->name,
+			    mode_type, modes[i]->name);
 
 		fallback = false;
 		conn_configured |= BIT(i);
@@ -799,8 +820,7 @@ int drm_client_modeset_probe(struct drm_client_dev *client, unsigned int width,
 	unsigned int total_modes_count = 0;
 	struct drm_client_offset *offsets;
 	unsigned int connector_count = 0;
-	/* points to modes protected by mode_config.mutex */
-	struct drm_display_mode **modes;
+	const struct drm_display_mode **modes;
 	struct drm_crtc **crtcs;
 	int i, ret = 0;
 	bool *enabled;
@@ -851,7 +871,7 @@ int drm_client_modeset_probe(struct drm_client_dev *client, unsigned int width,
 
 	if (!drm_client_firmware_config(client, connectors, connector_count, crtcs,
 					modes, offsets, enabled, width, height)) {
-		memset(modes, 0, connector_count * sizeof(*modes));
+		modes_destroy(dev, modes, connector_count);
 		memset(crtcs, 0, connector_count * sizeof(*crtcs));
 		memset(offsets, 0, connector_count * sizeof(*offsets));
 
@@ -868,10 +888,12 @@ int drm_client_modeset_probe(struct drm_client_dev *client, unsigned int width,
 				      crtcs, modes, 0, width, height);
 	}
 
+	mutex_unlock(&dev->mode_config.mutex);
+
 	drm_client_modeset_release(client);
 
 	for (i = 0; i < connector_count; i++) {
-		struct drm_display_mode *mode = modes[i];
+		const struct drm_display_mode *mode = modes[i];
 		struct drm_crtc *crtc = crtcs[i];
 		struct drm_client_offset *offset = &offsets[i];
 
@@ -902,11 +924,11 @@ int drm_client_modeset_probe(struct drm_client_dev *client, unsigned int width,
 			modeset->y = offset->y;
 		}
 	}
-	mutex_unlock(&dev->mode_config.mutex);
 
 	mutex_unlock(&client->modeset_mutex);
 out:
 	kfree(crtcs);
+	modes_destroy(dev, modes, connector_count);
 	kfree(modes);
 	kfree(offsets);
 	kfree(enabled);
@@ -938,7 +960,7 @@ bool drm_client_rotation(struct drm_mode_set *modeset, unsigned int *rotation)
 	struct drm_plane *plane = modeset->crtc->primary;
 	struct drm_cmdline_mode *cmdline;
 	u64 valid_mask = 0;
-	unsigned int i;
+	int i;
 
 	if (!modeset->num_connectors)
 		return false;
@@ -1219,11 +1241,12 @@ static void drm_client_modeset_dpms_legacy(struct drm_client_dev *client, int dp
 	struct drm_connector *connector;
 	struct drm_mode_set *modeset;
 	struct drm_modeset_acquire_ctx ctx;
-	int j;
 	int ret;
 
 	DRM_MODESET_LOCK_ALL_BEGIN(dev, ctx, 0, ret);
 	drm_client_for_each_modeset(modeset, client) {
+		int j;
+
 		if (!modeset->crtc->enabled)
 			continue;
 
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 0955f1c385dd..39497493f74c 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -334,7 +334,6 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 		if (!encoder_funcs)
 			continue;
 
-		encoder_funcs = encoder->helper_private;
 		if (encoder_funcs->mode_fixup) {
 			if (!(ret = encoder_funcs->mode_fixup(encoder, mode,
 							      adjusted_mode))) {
diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c
index 6b2178864c7e..3dfd8b34dceb 100644
--- a/drivers/gpu/drm/drm_debugfs.c
+++ b/drivers/gpu/drm/drm_debugfs.c
@@ -740,40 +740,6 @@ void drm_debugfs_crtc_remove(struct drm_crtc *crtc)
 	crtc->debugfs_entry = NULL;
 }
 
-static int bridges_show(struct seq_file *m, void *data)
-{
-	struct drm_encoder *encoder = m->private;
-	struct drm_printer p = drm_seq_file_printer(m);
-	struct drm_bridge *bridge;
-	unsigned int idx = 0;
-
-	drm_for_each_bridge_in_chain(encoder, bridge) {
-		drm_printf(&p, "bridge[%u]: %ps\n", idx++, bridge->funcs);
-		drm_printf(&p, "\ttype: [%d] %s\n",
-			   bridge->type,
-			   drm_get_connector_type_name(bridge->type));
-
-		if (bridge->of_node)
-			drm_printf(&p, "\tOF: %pOFfc\n", bridge->of_node);
-
-		drm_printf(&p, "\tops: [0x%x]", bridge->ops);
-		if (bridge->ops & DRM_BRIDGE_OP_DETECT)
-			drm_puts(&p, " detect");
-		if (bridge->ops & DRM_BRIDGE_OP_EDID)
-			drm_puts(&p, " edid");
-		if (bridge->ops & DRM_BRIDGE_OP_HPD)
-			drm_puts(&p, " hpd");
-		if (bridge->ops & DRM_BRIDGE_OP_MODES)
-			drm_puts(&p, " modes");
-		if (bridge->ops & DRM_BRIDGE_OP_HDMI)
-			drm_puts(&p, " hdmi");
-		drm_puts(&p, "\n");
-	}
-
-	return 0;
-}
-DEFINE_SHOW_ATTRIBUTE(bridges);
-
 void drm_debugfs_encoder_add(struct drm_encoder *encoder)
 {
 	struct drm_minor *minor = encoder->dev->primary;
@@ -789,9 +755,7 @@ void drm_debugfs_encoder_add(struct drm_encoder *encoder)
 
 	encoder->debugfs_entry = root;
 
-	/* bridges list */
-	debugfs_create_file("bridges", 0444, root, encoder,
-			    &bridges_fops);
+	drm_bridge_debugfs_encoder_params(root, encoder);
 
 	if (encoder->funcs && encoder->funcs->debugfs_init)
 		encoder->funcs->debugfs_init(encoder, root);
diff --git a/drivers/gpu/drm/drm_displayid_internal.h b/drivers/gpu/drm/drm_displayid_internal.h
index aee1b86a73c1..957dd0619f5c 100644
--- a/drivers/gpu/drm/drm_displayid_internal.h
+++ b/drivers/gpu/drm/drm_displayid_internal.h
@@ -66,6 +66,7 @@ struct drm_edid;
 #define DATA_BLOCK_2_STEREO_DISPLAY_INTERFACE	0x27
 #define DATA_BLOCK_2_TILED_DISPLAY_TOPOLOGY	0x28
 #define DATA_BLOCK_2_CONTAINER_ID		0x29
+#define DATA_BLOCK_2_TYPE_10_FORMULA_TIMING	0x2a
 #define DATA_BLOCK_2_VENDOR_SPECIFIC		0x7e
 #define DATA_BLOCK_2_CTA_DISPLAY_ID		0x81
 
@@ -114,20 +115,32 @@ struct displayid_tiled_block {
 struct displayid_detailed_timings_1 {
 	u8 pixel_clock[3];
 	u8 flags;
-	u8 hactive[2];
-	u8 hblank[2];
-	u8 hsync[2];
-	u8 hsw[2];
-	u8 vactive[2];
-	u8 vblank[2];
-	u8 vsync[2];
-	u8 vsw[2];
+	__le16 hactive;
+	__le16 hblank;
+	__le16 hsync;
+	__le16 hsw;
+	__le16 vactive;
+	__le16 vblank;
+	__le16 vsync;
+	__le16 vsw;
 } __packed;
 
 struct displayid_detailed_timing_block {
 	struct displayid_block base;
 	struct displayid_detailed_timings_1 timings[];
-};
+} __packed;
+
+struct displayid_formula_timings_9 {
+	u8 flags;
+	__le16 hactive;
+	__le16 vactive;
+	u8 vrefresh;
+} __packed;
+
+struct displayid_formula_timing_block {
+	struct displayid_block base;
+	struct displayid_formula_timings_9 timings[];
+} __packed;
 
 #define DISPLAYID_VESA_MSO_OVERLAP	GENMASK(3, 0)
 #define DISPLAYID_VESA_MSO_MODE		GENMASK(6, 5)
diff --git a/drivers/gpu/drm/drm_draw.c b/drivers/gpu/drm/drm_draw.c
index 385eb5e10047..9dc0408fbbea 100644
--- a/drivers/gpu/drm/drm_draw.c
+++ b/drivers/gpu/drm/drm_draw.c
@@ -13,85 +13,7 @@
 #include <drm/drm_fourcc.h>
 
 #include "drm_draw_internal.h"
-
-/*
- * Conversions from xrgb8888
- */
-
-static u16 convert_xrgb8888_to_rgb565(u32 pix)
-{
-	return ((pix & 0x00F80000) >> 8) |
-	       ((pix & 0x0000FC00) >> 5) |
-	       ((pix & 0x000000F8) >> 3);
-}
-
-static u16 convert_xrgb8888_to_rgba5551(u32 pix)
-{
-	return ((pix & 0x00f80000) >> 8) |
-	       ((pix & 0x0000f800) >> 5) |
-	       ((pix & 0x000000f8) >> 2) |
-	       BIT(0); /* set alpha bit */
-}
-
-static u16 convert_xrgb8888_to_xrgb1555(u32 pix)
-{
-	return ((pix & 0x00f80000) >> 9) |
-	       ((pix & 0x0000f800) >> 6) |
-	       ((pix & 0x000000f8) >> 3);
-}
-
-static u16 convert_xrgb8888_to_argb1555(u32 pix)
-{
-	return BIT(15) | /* set alpha bit */
-	       ((pix & 0x00f80000) >> 9) |
-	       ((pix & 0x0000f800) >> 6) |
-	       ((pix & 0x000000f8) >> 3);
-}
-
-static u32 convert_xrgb8888_to_argb8888(u32 pix)
-{
-	return pix | GENMASK(31, 24); /* fill alpha bits */
-}
-
-static u32 convert_xrgb8888_to_xbgr8888(u32 pix)
-{
-	return ((pix & 0x00ff0000) >> 16) <<  0 |
-	       ((pix & 0x0000ff00) >>  8) <<  8 |
-	       ((pix & 0x000000ff) >>  0) << 16 |
-	       ((pix & 0xff000000) >> 24) << 24;
-}
-
-static u32 convert_xrgb8888_to_abgr8888(u32 pix)
-{
-	return ((pix & 0x00ff0000) >> 16) <<  0 |
-	       ((pix & 0x0000ff00) >>  8) <<  8 |
-	       ((pix & 0x000000ff) >>  0) << 16 |
-	       GENMASK(31, 24); /* fill alpha bits */
-}
-
-static u32 convert_xrgb8888_to_xrgb2101010(u32 pix)
-{
-	pix = ((pix & 0x000000FF) << 2) |
-	      ((pix & 0x0000FF00) << 4) |
-	      ((pix & 0x00FF0000) << 6);
-	return pix | ((pix >> 8) & 0x00300C03);
-}
-
-static u32 convert_xrgb8888_to_argb2101010(u32 pix)
-{
-	pix = ((pix & 0x000000FF) << 2) |
-	      ((pix & 0x0000FF00) << 4) |
-	      ((pix & 0x00FF0000) << 6);
-	return GENMASK(31, 30) /* set alpha bits */ | pix | ((pix >> 8) & 0x00300C03);
-}
-
-static u32 convert_xrgb8888_to_abgr2101010(u32 pix)
-{
-	pix = ((pix & 0x00FF0000) >> 14) |
-	      ((pix & 0x0000FF00) << 4) |
-	      ((pix & 0x000000FF) << 22);
-	return GENMASK(31, 30) /* set alpha bits */ | pix | ((pix >> 8) & 0x00300C03);
-}
+#include "drm_format_internal.h"
 
 /**
  * drm_draw_color_from_xrgb8888 - convert one pixel from xrgb8888 to the desired format
@@ -106,28 +28,28 @@ u32 drm_draw_color_from_xrgb8888(u32 color, u32 format)
 {
 	switch (format) {
 	case DRM_FORMAT_RGB565:
-		return convert_xrgb8888_to_rgb565(color);
+		return drm_pixel_xrgb8888_to_rgb565(color);
 	case DRM_FORMAT_RGBA5551:
-		return convert_xrgb8888_to_rgba5551(color);
+		return drm_pixel_xrgb8888_to_rgba5551(color);
 	case DRM_FORMAT_XRGB1555:
-		return convert_xrgb8888_to_xrgb1555(color);
+		return drm_pixel_xrgb8888_to_xrgb1555(color);
 	case DRM_FORMAT_ARGB1555:
-		return convert_xrgb8888_to_argb1555(color);
+		return drm_pixel_xrgb8888_to_argb1555(color);
 	case DRM_FORMAT_RGB888:
 	case DRM_FORMAT_XRGB8888:
 		return color;
 	case DRM_FORMAT_ARGB8888:
-		return convert_xrgb8888_to_argb8888(color);
+		return drm_pixel_xrgb8888_to_argb8888(color);
 	case DRM_FORMAT_XBGR8888:
-		return convert_xrgb8888_to_xbgr8888(color);
+		return drm_pixel_xrgb8888_to_xbgr8888(color);
 	case DRM_FORMAT_ABGR8888:
-		return convert_xrgb8888_to_abgr8888(color);
+		return drm_pixel_xrgb8888_to_abgr8888(color);
 	case DRM_FORMAT_XRGB2101010:
-		return convert_xrgb8888_to_xrgb2101010(color);
+		return drm_pixel_xrgb8888_to_xrgb2101010(color);
 	case DRM_FORMAT_ARGB2101010:
-		return convert_xrgb8888_to_argb2101010(color);
+		return drm_pixel_xrgb8888_to_argb2101010(color);
 	case DRM_FORMAT_ABGR2101010:
-		return convert_xrgb8888_to_abgr2101010(color);
+		return drm_pixel_xrgb8888_to_abgr2101010(color);
 	default:
 		WARN_ONCE(1, "Can't convert to %p4cc\n", &format);
 		return 0;
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index ebb648f1c7a9..4589ab4bb1ec 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -40,6 +40,7 @@
 #include <linux/xarray.h>
 
 #include <drm/drm_accel.h>
+#include <drm/drm_bridge.h>
 #include <drm/drm_cache.h>
 #include <drm/drm_client_event.h>
 #include <drm/drm_color_mgmt.h>
@@ -500,6 +501,25 @@ void drm_dev_unplug(struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_dev_unplug);
 
+/**
+ * drm_dev_set_dma_dev - set the DMA device for a DRM device
+ * @dev: DRM device
+ * @dma_dev: DMA device or NULL
+ *
+ * Sets the DMA device of the given DRM device. Only required if
+ * the DMA device is different from the DRM device's parent. After
+ * calling this function, the DRM device holds a reference on
+ * @dma_dev. Pass NULL to clear the DMA device.
+ */
+void drm_dev_set_dma_dev(struct drm_device *dev, struct device *dma_dev)
+{
+	dma_dev = get_device(dma_dev);
+
+	put_device(dev->dma_dev);
+	dev->dma_dev = dma_dev;
+}
+EXPORT_SYMBOL(drm_dev_set_dma_dev);
+
 /*
  * Available recovery methods for wedged device. To be sent along with device
  * wedged uevent.
@@ -654,6 +674,8 @@ static void drm_dev_init_release(struct drm_device *dev, void *res)
 {
 	drm_fs_inode_free(dev->anon_inode);
 
+	put_device(dev->dma_dev);
+	dev->dma_dev = NULL;
 	put_device(dev->dev);
 	/* Prevent use-after-free in drm_managed_release when debugging is
 	 * enabled. Slightly awkward, but can't really be helped. */
@@ -1214,6 +1236,7 @@ static int __init drm_core_init(void)
 	}
 
 	drm_debugfs_root = debugfs_create_dir("dri", NULL);
+	drm_bridge_debugfs_params(drm_debugfs_root);
 
 	ret = register_chrdev(DRM_MAJOR, "drm", &drm_stub_fops);
 	if (ret < 0)
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 13bc4c290b17..eab5196140d7 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -6760,23 +6760,23 @@ out:
 }
 
 static struct drm_display_mode *drm_mode_displayid_detailed(struct drm_device *dev,
-							    struct displayid_detailed_timings_1 *timings,
+							    const struct displayid_detailed_timings_1 *timings,
 							    bool type_7)
 {
 	struct drm_display_mode *mode;
-	unsigned pixel_clock = (timings->pixel_clock[0] |
-				(timings->pixel_clock[1] << 8) |
-				(timings->pixel_clock[2] << 16)) + 1;
-	unsigned hactive = (timings->hactive[0] | timings->hactive[1] << 8) + 1;
-	unsigned hblank = (timings->hblank[0] | timings->hblank[1] << 8) + 1;
-	unsigned hsync = (timings->hsync[0] | (timings->hsync[1] & 0x7f) << 8) + 1;
-	unsigned hsync_width = (timings->hsw[0] | timings->hsw[1] << 8) + 1;
-	unsigned vactive = (timings->vactive[0] | timings->vactive[1] << 8) + 1;
-	unsigned vblank = (timings->vblank[0] | timings->vblank[1] << 8) + 1;
-	unsigned vsync = (timings->vsync[0] | (timings->vsync[1] & 0x7f) << 8) + 1;
-	unsigned vsync_width = (timings->vsw[0] | timings->vsw[1] << 8) + 1;
-	bool hsync_positive = (timings->hsync[1] >> 7) & 0x1;
-	bool vsync_positive = (timings->vsync[1] >> 7) & 0x1;
+	unsigned int pixel_clock = (timings->pixel_clock[0] |
+				    (timings->pixel_clock[1] << 8) |
+				    (timings->pixel_clock[2] << 16)) + 1;
+	unsigned int hactive = le16_to_cpu(timings->hactive) + 1;
+	unsigned int hblank = le16_to_cpu(timings->hblank) + 1;
+	unsigned int hsync = (le16_to_cpu(timings->hsync) & 0x7fff) + 1;
+	unsigned int hsync_width = le16_to_cpu(timings->hsw) + 1;
+	unsigned int vactive = le16_to_cpu(timings->vactive) + 1;
+	unsigned int vblank = le16_to_cpu(timings->vblank) + 1;
+	unsigned int vsync = (le16_to_cpu(timings->vsync) & 0x7fff) + 1;
+	unsigned int vsync_width = le16_to_cpu(timings->vsw) + 1;
+	bool hsync_positive = le16_to_cpu(timings->hsync) & (1 << 15);
+	bool vsync_positive = le16_to_cpu(timings->vsync) & (1 << 15);
 
 	mode = drm_mode_create(dev);
 	if (!mode)
@@ -6833,6 +6833,66 @@ static int add_displayid_detailed_1_modes(struct drm_connector *connector,
 	return num_modes;
 }
 
+static struct drm_display_mode *drm_mode_displayid_formula(struct drm_device *dev,
+							   const struct displayid_formula_timings_9 *timings,
+							   bool type_10)
+{
+	struct drm_display_mode *mode;
+	u16 hactive = le16_to_cpu(timings->hactive) + 1;
+	u16 vactive = le16_to_cpu(timings->vactive) + 1;
+	u8 timing_formula = timings->flags & 0x7;
+
+	/* TODO: support RB-v2 & RB-v3 */
+	if (timing_formula > 1)
+		return NULL;
+
+	/* TODO: support video-optimized refresh rate */
+	if (timings->flags & (1 << 4))
+		drm_dbg_kms(dev, "Fractional vrefresh is not implemented, proceeding with non-video-optimized refresh rate");
+
+	mode = drm_cvt_mode(dev, hactive, vactive, timings->vrefresh + 1, timing_formula == 1, false, false);
+	if (!mode)
+		return NULL;
+
+	/* TODO: interpret S3D flags */
+
+	mode->type = DRM_MODE_TYPE_DRIVER;
+	drm_mode_set_name(mode);
+
+	return mode;
+}
+
+static int add_displayid_formula_modes(struct drm_connector *connector,
+				       const struct displayid_block *block)
+{
+	const struct displayid_formula_timing_block *formula_block = (struct displayid_formula_timing_block *)block;
+	int num_timings;
+	struct drm_display_mode *newmode;
+	int num_modes = 0;
+	bool type_10 = block->tag == DATA_BLOCK_2_TYPE_10_FORMULA_TIMING;
+	int timing_size = 6 + ((formula_block->base.rev & 0x70) >> 4);
+
+	/* extended blocks are not supported yet */
+	if (timing_size != 6)
+		return 0;
+
+	if (block->num_bytes % timing_size)
+		return 0;
+
+	num_timings = block->num_bytes / timing_size;
+	for (int i = 0; i < num_timings; i++) {
+		const struct displayid_formula_timings_9 *timings = &formula_block->timings[i];
+
+		newmode = drm_mode_displayid_formula(connector->dev, timings, type_10);
+		if (!newmode)
+			continue;
+
+		drm_mode_probed_add(connector, newmode);
+		num_modes++;
+	}
+	return num_modes;
+}
+
 static int add_displayid_detailed_modes(struct drm_connector *connector,
 					const struct drm_edid *drm_edid)
 {
@@ -6845,6 +6905,9 @@ static int add_displayid_detailed_modes(struct drm_connector *connector,
 		if (block->tag == DATA_BLOCK_TYPE_1_DETAILED_TIMING ||
 		    block->tag == DATA_BLOCK_2_TYPE_7_DETAILED_TIMING)
 			num_modes += add_displayid_detailed_1_modes(connector, block);
+		else if (block->tag == DATA_BLOCK_2_TYPE_9_FORMULA_TIMING ||
+			 block->tag == DATA_BLOCK_2_TYPE_10_FORMULA_TIMING)
+			num_modes += add_displayid_formula_modes(connector, block);
 	}
 	displayid_iter_end(&iter);
 
@@ -7099,18 +7162,12 @@ EXPORT_SYMBOL(drm_add_edid_modes);
  * Return: The number of modes added or 0 if we couldn't find any.
  */
 int drm_add_modes_noedid(struct drm_connector *connector,
-			int hdisplay, int vdisplay)
+			 unsigned int hdisplay, unsigned int vdisplay)
 {
-	int i, count, num_modes = 0;
+	int i, count = ARRAY_SIZE(drm_dmt_modes), num_modes = 0;
 	struct drm_display_mode *mode;
 	struct drm_device *dev = connector->dev;
 
-	count = ARRAY_SIZE(drm_dmt_modes);
-	if (hdisplay < 0)
-		hdisplay = 0;
-	if (vdisplay < 0)
-		vdisplay = 0;
-
 	for (i = 0; i < count; i++) {
 		const struct drm_display_mode *ptr = &drm_dmt_modes[i];
 
diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
index c299cd94d3f7..246cf845e2c9 100644
--- a/drivers/gpu/drm/drm_file.c
+++ b/drivers/gpu/drm/drm_file.c
@@ -964,6 +964,10 @@ void drm_show_fdinfo(struct seq_file *m, struct file *f)
 	struct drm_file *file = f->private_data;
 	struct drm_device *dev = file->minor->dev;
 	struct drm_printer p = drm_seq_file_printer(m);
+	int idx;
+
+	if (!drm_dev_enter(dev, &idx))
+		return;
 
 	drm_printf(&p, "drm-driver:\t%s\n", dev->driver->name);
 	drm_printf(&p, "drm-client-id:\t%llu\n", file->client_id);
@@ -983,10 +987,46 @@ void drm_show_fdinfo(struct seq_file *m, struct file *f)
 
 	if (dev->driver->show_fdinfo)
 		dev->driver->show_fdinfo(&p, file);
+
+	drm_dev_exit(idx);
 }
 EXPORT_SYMBOL(drm_show_fdinfo);
 
 /**
+ * drm_file_err - log process name, pid and client_name associated with a drm_file
+ * @file_priv: context of interest for process name and pid
+ * @fmt: printf() like format string
+ *
+ * Helper function for clients which needs to log process details such
+ * as name and pid etc along with user logs.
+ */
+void drm_file_err(struct drm_file *file_priv, const char *fmt, ...)
+{
+	va_list args;
+	struct va_format vaf;
+	struct pid *pid;
+	struct task_struct *task;
+	struct drm_device *dev = file_priv->minor->dev;
+
+	va_start(args, fmt);
+	vaf.fmt = fmt;
+	vaf.va = &args;
+
+	mutex_lock(&file_priv->client_name_lock);
+	rcu_read_lock();
+	pid = rcu_dereference(file_priv->pid);
+	task = pid_task(pid, PIDTYPE_TGID);
+
+	drm_err(dev, "comm: %s pid: %d client: %s ... %pV", task ? task->comm : "Unset",
+		task ? task->pid : 0, file_priv->client_name ?: "Unset", &vaf);
+
+	va_end(args);
+	rcu_read_unlock();
+	mutex_unlock(&file_priv->client_name_lock);
+}
+EXPORT_SYMBOL(drm_file_err);
+
+/**
  * mock_drm_getfile - Create a new struct file for the drm device
  * @minor: drm minor to wrap (e.g. #drm_device.primary)
  * @flags: file creation mode (O_RDWR etc)
diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c
index 01d3ab307ac3..d36e6cacc575 100644
--- a/drivers/gpu/drm/drm_format_helper.c
+++ b/drivers/gpu/drm/drm_format_helper.c
@@ -20,6 +20,8 @@
 #include <drm/drm_print.h>
 #include <drm/drm_rect.h>
 
+#include "drm_format_internal.h"
+
 /**
  * drm_format_conv_state_init - Initialize format-conversion state
  * @state: The state to initialize
@@ -244,6 +246,152 @@ static int drm_fb_xfrm(struct iosys_map *dst,
 				     xfrm_line);
 }
 
+#define ALIGN_DOWN_PIXELS(end, n, a) \
+	((end) - ((n) & ((a) - 1)))
+
+static __always_inline void drm_fb_xfrm_line_32to8(void *dbuf, const void *sbuf,
+						   unsigned int pixels,
+						   u32 (*xfrm_pixel)(u32))
+{
+	__le32 *dbuf32 = dbuf;
+	u8 *dbuf8;
+	const __le32 *sbuf32 = sbuf;
+	const __le32 *send32 = sbuf32 + pixels;
+
+	/* write 4 pixels at once */
+	while (sbuf32 < ALIGN_DOWN_PIXELS(send32, pixels, 4)) {
+		u32 pix[4] = {
+			le32_to_cpup(sbuf32),
+			le32_to_cpup(sbuf32 + 1),
+			le32_to_cpup(sbuf32 + 2),
+			le32_to_cpup(sbuf32 + 3),
+		};
+		/* write output bytes in reverse order for little endianness */
+		u32 val32 = xfrm_pixel(pix[0]) |
+			   (xfrm_pixel(pix[1]) << 8) |
+			   (xfrm_pixel(pix[2]) << 16) |
+			   (xfrm_pixel(pix[3]) << 24);
+		*dbuf32++ = cpu_to_le32(val32);
+		sbuf32 += ARRAY_SIZE(pix);
+	}
+
+	/* write trailing pixels */
+	dbuf8 = (u8 __force *)dbuf32;
+	while (sbuf32 < send32)
+		*dbuf8++ = xfrm_pixel(le32_to_cpup(sbuf32++));
+}
+
+static __always_inline void drm_fb_xfrm_line_32to16(void *dbuf, const void *sbuf,
+						    unsigned int pixels,
+						    u32 (*xfrm_pixel)(u32))
+{
+	__le64 *dbuf64 = dbuf;
+	__le32 *dbuf32;
+	__le16 *dbuf16;
+	const __le32 *sbuf32 = sbuf;
+	const __le32 *send32 = sbuf32 + pixels;
+
+#if defined(CONFIG_64BIT)
+	/* write 4 pixels at once */
+	while (sbuf32 < ALIGN_DOWN_PIXELS(send32, pixels, 4)) {
+		u32 pix[4] = {
+			le32_to_cpup(sbuf32),
+			le32_to_cpup(sbuf32 + 1),
+			le32_to_cpup(sbuf32 + 2),
+			le32_to_cpup(sbuf32 + 3),
+		};
+		/* write output bytes in reverse order for little endianness */
+		u64 val64 = ((u64)xfrm_pixel(pix[0])) |
+			    ((u64)xfrm_pixel(pix[1]) << 16) |
+			    ((u64)xfrm_pixel(pix[2]) << 32) |
+			    ((u64)xfrm_pixel(pix[3]) << 48);
+		*dbuf64++ = cpu_to_le64(val64);
+		sbuf32 += ARRAY_SIZE(pix);
+	}
+#endif
+
+	/* write 2 pixels at once */
+	dbuf32 = (__le32 __force *)dbuf64;
+	while (sbuf32 < ALIGN_DOWN_PIXELS(send32, pixels, 2)) {
+		u32 pix[2] = {
+			le32_to_cpup(sbuf32),
+			le32_to_cpup(sbuf32 + 1),
+		};
+		/* write output bytes in reverse order for little endianness */
+		u32 val32 = xfrm_pixel(pix[0]) |
+			   (xfrm_pixel(pix[1]) << 16);
+		*dbuf32++ = cpu_to_le32(val32);
+		sbuf32 += ARRAY_SIZE(pix);
+	}
+
+	/* write trailing pixel */
+	dbuf16 = (__le16 __force *)dbuf32;
+	while (sbuf32 < send32)
+		*dbuf16++ = cpu_to_le16(xfrm_pixel(le32_to_cpup(sbuf32++)));
+}
+
+static __always_inline void drm_fb_xfrm_line_32to24(void *dbuf, const void *sbuf,
+						    unsigned int pixels,
+						    u32 (*xfrm_pixel)(u32))
+{
+	__le32 *dbuf32 = dbuf;
+	u8 *dbuf8;
+	const __le32 *sbuf32 = sbuf;
+	const __le32 *send32 = sbuf32 + pixels;
+
+	/* write pixels in chunks of 4 */
+	while (sbuf32 < ALIGN_DOWN_PIXELS(send32, pixels, 4)) {
+		u32 val24[4] = {
+			xfrm_pixel(le32_to_cpup(sbuf32)),
+			xfrm_pixel(le32_to_cpup(sbuf32 + 1)),
+			xfrm_pixel(le32_to_cpup(sbuf32 + 2)),
+			xfrm_pixel(le32_to_cpup(sbuf32 + 3)),
+		};
+		u32 out32[3] = {
+			/* write output bytes in reverse order for little endianness */
+			((val24[0] & 0x000000ff)) |
+			((val24[0] & 0x0000ff00)) |
+			((val24[0] & 0x00ff0000)) |
+			((val24[1] & 0x000000ff) << 24),
+			((val24[1] & 0x0000ff00) >> 8) |
+			((val24[1] & 0x00ff0000) >> 8) |
+			((val24[2] & 0x000000ff) << 16) |
+			((val24[2] & 0x0000ff00) << 16),
+			((val24[2] & 0x00ff0000) >> 16) |
+			((val24[3] & 0x000000ff) << 8) |
+			((val24[3] & 0x0000ff00) << 8) |
+			((val24[3] & 0x00ff0000) << 8),
+		};
+
+		*dbuf32++ = cpu_to_le32(out32[0]);
+		*dbuf32++ = cpu_to_le32(out32[1]);
+		*dbuf32++ = cpu_to_le32(out32[2]);
+		sbuf32 += ARRAY_SIZE(val24);
+	}
+
+	/* write trailing pixel */
+	dbuf8 = (u8 __force *)dbuf32;
+	while (sbuf32 < send32) {
+		u32 val24 = xfrm_pixel(le32_to_cpup(sbuf32++));
+		/* write output in reverse order for little endianness */
+		*dbuf8++ = (val24 & 0x000000ff);
+		*dbuf8++ = (val24 & 0x0000ff00) >>  8;
+		*dbuf8++ = (val24 & 0x00ff0000) >> 16;
+	}
+}
+
+static __always_inline void drm_fb_xfrm_line_32to32(void *dbuf, const void *sbuf,
+						    unsigned int pixels,
+						    u32 (*xfrm_pixel)(u32))
+{
+	__le32 *dbuf32 = dbuf;
+	const __le32 *sbuf32 = sbuf;
+	const __le32 *send32 = sbuf32 + pixels;
+
+	while (sbuf32 < send32)
+		*dbuf32++ = cpu_to_le32(xfrm_pixel(le32_to_cpup(sbuf32++)));
+}
+
 /**
  * drm_fb_memcpy - Copy clip buffer
  * @dst: Array of destination buffers
@@ -368,17 +516,7 @@ EXPORT_SYMBOL(drm_fb_swab);
 
 static void drm_fb_xrgb8888_to_rgb332_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	u8 *dbuf8 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		dbuf8[x] = ((pix & 0x00e00000) >> 16) |
-			   ((pix & 0x0000e000) >> 11) |
-			   ((pix & 0x000000c0) >> 6);
-	}
+	drm_fb_xfrm_line_32to8(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_rgb332);
 }
 
 /**
@@ -417,38 +555,19 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb332);
 
 static void drm_fb_xrgb8888_to_rgb565_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le16 *dbuf16 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u16 val16;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val16 = ((pix & 0x00F80000) >> 8) |
-			((pix & 0x0000FC00) >> 5) |
-			((pix & 0x000000F8) >> 3);
-		dbuf16[x] = cpu_to_le16(val16);
-	}
+	drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_rgb565);
+}
+
+static __always_inline u32 drm_xrgb8888_to_rgb565_swab(u32 pix)
+{
+	return swab16(drm_pixel_xrgb8888_to_rgb565(pix));
 }
 
 /* TODO: implement this helper as conversion to RGB565|BIG_ENDIAN */
 static void drm_fb_xrgb8888_to_rgb565_swab_line(void *dbuf, const void *sbuf,
 						unsigned int pixels)
 {
-	__le16 *dbuf16 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u16 val16;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val16 = ((pix & 0x00F80000) >> 8) |
-			((pix & 0x0000FC00) >> 5) |
-			((pix & 0x000000F8) >> 3);
-		dbuf16[x] = cpu_to_le16(swab16(val16));
-	}
+	drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_xrgb8888_to_rgb565_swab);
 }
 
 /**
@@ -495,19 +614,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565);
 
 static void drm_fb_xrgb8888_to_xrgb1555_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le16 *dbuf16 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u16 val16;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val16 = ((pix & 0x00f80000) >> 9) |
-			((pix & 0x0000f800) >> 6) |
-			((pix & 0x000000f8) >> 3);
-		dbuf16[x] = cpu_to_le16(val16);
-	}
+	drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_xrgb1555);
 }
 
 /**
@@ -547,20 +654,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_xrgb1555);
 
 static void drm_fb_xrgb8888_to_argb1555_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le16 *dbuf16 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u16 val16;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val16 = BIT(15) | /* set alpha bit */
-			((pix & 0x00f80000) >> 9) |
-			((pix & 0x0000f800) >> 6) |
-			((pix & 0x000000f8) >> 3);
-		dbuf16[x] = cpu_to_le16(val16);
-	}
+	drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_argb1555);
 }
 
 /**
@@ -600,20 +694,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_argb1555);
 
 static void drm_fb_xrgb8888_to_rgba5551_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le16 *dbuf16 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u16 val16;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val16 = ((pix & 0x00f80000) >> 8) |
-			((pix & 0x0000f800) >> 5) |
-			((pix & 0x000000f8) >> 2) |
-			BIT(0); /* set alpha bit */
-		dbuf16[x] = cpu_to_le16(val16);
-	}
+	drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_rgba5551);
 }
 
 /**
@@ -653,18 +734,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgba5551);
 
 static void drm_fb_xrgb8888_to_rgb888_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	u8 *dbuf8 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		/* write blue-green-red to output in little endianness */
-		*dbuf8++ = (pix & 0x000000FF) >>  0;
-		*dbuf8++ = (pix & 0x0000FF00) >>  8;
-		*dbuf8++ = (pix & 0x00FF0000) >> 16;
-	}
+	drm_fb_xfrm_line_32to24(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_rgb888);
 }
 
 /**
@@ -704,18 +774,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb888);
 
 static void drm_fb_xrgb8888_to_bgr888_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	u8 *dbuf8 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		/* write red-green-blue to output in little endianness */
-		*dbuf8++ = (pix & 0x00ff0000) >> 16;
-		*dbuf8++ = (pix & 0x0000ff00) >> 8;
-		*dbuf8++ = (pix & 0x000000ff) >> 0;
-	}
+	drm_fb_xfrm_line_32to24(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_bgr888);
 }
 
 /**
@@ -755,16 +814,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_bgr888);
 
 static void drm_fb_xrgb8888_to_argb8888_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le32 *dbuf32 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		pix |= GENMASK(31, 24); /* fill alpha bits */
-		dbuf32[x] = cpu_to_le32(pix);
-	}
+	drm_fb_xfrm_line_32to32(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_argb8888);
 }
 
 /**
@@ -804,19 +854,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_argb8888);
 
 static void drm_fb_xrgb8888_to_abgr8888_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le32 *dbuf32 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		pix = ((pix & 0x00ff0000) >> 16) <<  0 |
-		      ((pix & 0x0000ff00) >>  8) <<  8 |
-		      ((pix & 0x000000ff) >>  0) << 16 |
-		      GENMASK(31, 24); /* fill alpha bits */
-		*dbuf32++ = cpu_to_le32(pix);
-	}
+	drm_fb_xfrm_line_32to32(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_abgr8888);
 }
 
 static void drm_fb_xrgb8888_to_abgr8888(struct iosys_map *dst, const unsigned int *dst_pitch,
@@ -835,19 +873,7 @@ static void drm_fb_xrgb8888_to_abgr8888(struct iosys_map *dst, const unsigned in
 
 static void drm_fb_xrgb8888_to_xbgr8888_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le32 *dbuf32 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		pix = ((pix & 0x00ff0000) >> 16) <<  0 |
-		      ((pix & 0x0000ff00) >>  8) <<  8 |
-		      ((pix & 0x000000ff) >>  0) << 16 |
-		      ((pix & 0xff000000) >> 24) << 24;
-		*dbuf32++ = cpu_to_le32(pix);
-	}
+	drm_fb_xfrm_line_32to32(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_xbgr8888);
 }
 
 static void drm_fb_xrgb8888_to_xbgr8888(struct iosys_map *dst, const unsigned int *dst_pitch,
@@ -866,20 +892,7 @@ static void drm_fb_xrgb8888_to_xbgr8888(struct iosys_map *dst, const unsigned in
 
 static void drm_fb_xrgb8888_to_xrgb2101010_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le32 *dbuf32 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 val32;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val32 = ((pix & 0x000000FF) << 2) |
-			((pix & 0x0000FF00) << 4) |
-			((pix & 0x00FF0000) << 6);
-		pix = val32 | ((val32 >> 8) & 0x00300C03);
-		*dbuf32++ = cpu_to_le32(pix);
-	}
+	drm_fb_xfrm_line_32to32(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_xrgb2101010);
 }
 
 /**
@@ -920,21 +933,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_xrgb2101010);
 
 static void drm_fb_xrgb8888_to_argb2101010_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	__le32 *dbuf32 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 val32;
-	u32 pix;
-
-	for (x = 0; x < pixels; x++) {
-		pix = le32_to_cpu(sbuf32[x]);
-		val32 = ((pix & 0x000000ff) << 2) |
-			((pix & 0x0000ff00) << 4) |
-			((pix & 0x00ff0000) << 6);
-		pix = GENMASK(31, 30) | /* set alpha bits */
-		      val32 | ((val32 >> 8) & 0x00300c03);
-		*dbuf32++ = cpu_to_le32(pix);
-	}
+	drm_fb_xfrm_line_32to32(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_argb2101010);
 }
 
 /**
@@ -975,19 +974,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_argb2101010);
 
 static void drm_fb_xrgb8888_to_gray8_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	u8 *dbuf8 = dbuf;
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-
-	for (x = 0; x < pixels; x++) {
-		u32 pix = le32_to_cpu(sbuf32[x]);
-		u8 r = (pix & 0x00ff0000) >> 16;
-		u8 g = (pix & 0x0000ff00) >> 8;
-		u8 b =  pix & 0x000000ff;
-
-		/* ITU BT.601: Y = 0.299 R + 0.587 G + 0.114 B */
-		*dbuf8++ = (3 * r + 6 * g + b) / 10;
-	}
+	drm_fb_xfrm_line_32to8(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_r8_bt601);
 }
 
 /**
@@ -1031,36 +1018,7 @@ EXPORT_SYMBOL(drm_fb_xrgb8888_to_gray8);
 
 static void drm_fb_argb8888_to_argb4444_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
-	unsigned int pixels2 = pixels & ~GENMASK_ULL(0, 0);
-	__le32 *dbuf32 = dbuf;
-	__le16 *dbuf16 = dbuf + pixels2 * sizeof(*dbuf16);
-	const __le32 *sbuf32 = sbuf;
-	unsigned int x;
-	u32 val32;
-	u16 val16;
-	u32 pix[2];
-
-	for (x = 0; x < pixels2; x += 2, ++dbuf32) {
-		pix[0] = le32_to_cpu(sbuf32[x]);
-		pix[1] = le32_to_cpu(sbuf32[x + 1]);
-		val32 = ((pix[0] & 0xf0000000) >> 16) |
-			((pix[0] & 0x00f00000) >> 12) |
-			((pix[0] & 0x0000f000) >> 8) |
-			((pix[0] & 0x000000f0) >> 4) |
-			((pix[1] & 0xf0000000) >> 0) |
-			((pix[1] & 0x00f00000) << 4) |
-			((pix[1] & 0x0000f000) << 8) |
-			((pix[1] & 0x000000f0) << 12);
-		*dbuf32 = cpu_to_le32(val32);
-	}
-	for (; x < pixels; x++) {
-		pix[0] = le32_to_cpu(sbuf32[x]);
-		val16 = ((pix[0] & 0xf0000000) >> 16) |
-			((pix[0] & 0x00f00000) >> 12) |
-			((pix[0] & 0x0000f000) >> 8) |
-			((pix[0] & 0x000000f0) >> 4);
-		dbuf16[x] = cpu_to_le16(val16);
-	}
+	drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_argb8888_to_argb4444);
 }
 
 /**
diff --git a/drivers/gpu/drm/drm_format_internal.h b/drivers/gpu/drm/drm_format_internal.h
new file mode 100644
index 000000000000..9f857bfa368d
--- /dev/null
+++ b/drivers/gpu/drm/drm_format_internal.h
@@ -0,0 +1,160 @@
+/* SPDX-License-Identifier: GPL-2.0 or MIT */
+
+#ifndef DRM_FORMAT_INTERNAL_H
+#define DRM_FORMAT_INTERNAL_H
+
+#include <linux/bits.h>
+#include <linux/types.h>
+
+/*
+ * Each pixel-format conversion helper takes a raw pixel in a
+ * specific input format and returns a raw pixel in a specific
+ * output format. All pixels are in little-endian byte order.
+ *
+ * Function names are
+ *
+ *   drm_pixel_<input>_to_<output>_<algorithm>()
+ *
+ * where <input> and <output> refer to pixel formats. The
+ * <algorithm> is optional and hints to the method used for the
+ * conversion. Helpers with no algorithm given apply pixel-bit
+ * shifting.
+ *
+ * The argument type is u32. We expect this to be wide enough to
+ * hold all conversion input from 32-bit RGB to any output format.
+ * The Linux kernel should avoid format conversion for anything
+ * but XRGB8888 input data. Converting from other format can still
+ * be acceptable in some cases.
+ *
+ * The return type is u32. It is wide enough to hold all conversion
+ * output from XRGB8888. For output formats wider than 32 bit, a
+ * return type of u64 would be acceptable.
+ */
+
+/*
+ * Conversions from XRGB8888
+ */
+
+static inline u32 drm_pixel_xrgb8888_to_r8_bt601(u32 pix)
+{
+	u32 r = (pix & 0x00ff0000) >> 16;
+	u32 g = (pix & 0x0000ff00) >> 8;
+	u32 b =  pix & 0x000000ff;
+
+	/* ITU-R BT.601: Y = 0.299 R + 0.587 G + 0.114 B */
+	return (3 * r + 6 * g + b) / 10;
+}
+
+static inline u32 drm_pixel_xrgb8888_to_rgb332(u32 pix)
+{
+	return ((pix & 0x00e00000) >> 16) |
+	       ((pix & 0x0000e000) >> 11) |
+	       ((pix & 0x000000c0) >> 6);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_rgb565(u32 pix)
+{
+	return ((pix & 0x00f80000) >> 8) |
+	       ((pix & 0x0000fc00) >> 5) |
+	       ((pix & 0x000000f8) >> 3);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_rgbx5551(u32 pix)
+{
+	return ((pix & 0x00f80000) >> 8) |
+	       ((pix & 0x0000f800) >> 5) |
+	       ((pix & 0x000000f8) >> 2);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_rgba5551(u32 pix)
+{
+	return drm_pixel_xrgb8888_to_rgbx5551(pix) |
+	       BIT(0); /* set alpha bit */
+}
+
+static inline u32 drm_pixel_xrgb8888_to_xrgb1555(u32 pix)
+{
+	return ((pix & 0x00f80000) >> 9) |
+	       ((pix & 0x0000f800) >> 6) |
+	       ((pix & 0x000000f8) >> 3);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_argb1555(u32 pix)
+{
+	return BIT(15) | /* set alpha bit */
+	       drm_pixel_xrgb8888_to_xrgb1555(pix);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_rgb888(u32 pix)
+{
+	return pix & GENMASK(23, 0);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_bgr888(u32 pix)
+{
+	return ((pix & 0x00ff0000) >> 16) |
+	       ((pix & 0x0000ff00)) |
+	       ((pix & 0x000000ff) << 16);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_argb8888(u32 pix)
+{
+	return GENMASK(31, 24) | /* fill alpha bits */
+	       pix;
+}
+
+static inline u32 drm_pixel_xrgb8888_to_xbgr8888(u32 pix)
+{
+	return ((pix & 0xff000000)) | /* also copy filler bits */
+	       ((pix & 0x00ff0000) >> 16) |
+	       ((pix & 0x0000ff00)) |
+	       ((pix & 0x000000ff) << 16);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_abgr8888(u32 pix)
+{
+	return GENMASK(31, 24) | /* fill alpha bits */
+	       drm_pixel_xrgb8888_to_xbgr8888(pix);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_xrgb2101010(u32 pix)
+{
+	pix = ((pix & 0x000000ff) << 2) |
+	      ((pix & 0x0000ff00) << 4) |
+	      ((pix & 0x00ff0000) << 6);
+	return pix | ((pix >> 8) & 0x00300c03);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_argb2101010(u32 pix)
+{
+	return GENMASK(31, 30) | /* set alpha bits */
+	       drm_pixel_xrgb8888_to_xrgb2101010(pix);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_xbgr2101010(u32 pix)
+{
+	pix = ((pix & 0x00ff0000) >> 14) |
+	      ((pix & 0x0000ff00) << 4) |
+	      ((pix & 0x000000ff) << 22);
+	return pix | ((pix >> 8) & 0x00300c03);
+}
+
+static inline u32 drm_pixel_xrgb8888_to_abgr2101010(u32 pix)
+{
+	return GENMASK(31, 30) | /* set alpha bits */
+	       drm_pixel_xrgb8888_to_xbgr2101010(pix);
+}
+
+/*
+ * Conversion from ARGB8888
+ */
+
+static inline u32 drm_pixel_argb8888_to_argb4444(u32 pix)
+{
+	return ((pix & 0xf0000000) >> 16) |
+	       ((pix & 0x00f00000) >> 12) |
+	       ((pix & 0x0000f000) >> 8) |
+	       ((pix & 0x000000f0) >> 4);
+}
+
+#endif
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index c6240bab3fa5..1e659d2660f7 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -1216,7 +1216,7 @@ void drm_gem_unpin(struct drm_gem_object *obj)
 	dma_resv_unlock(obj->resv);
 }
 
-int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map)
+int drm_gem_vmap_locked(struct drm_gem_object *obj, struct iosys_map *map)
 {
 	int ret;
 
@@ -1233,9 +1233,9 @@ int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map)
 
 	return 0;
 }
-EXPORT_SYMBOL(drm_gem_vmap);
+EXPORT_SYMBOL(drm_gem_vmap_locked);
 
-void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map)
+void drm_gem_vunmap_locked(struct drm_gem_object *obj, struct iosys_map *map)
 {
 	dma_resv_assert_held(obj->resv);
 
@@ -1248,7 +1248,7 @@ void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map)
 	/* Always set the mapping to NULL. Callers may rely on this. */
 	iosys_map_clear(map);
 }
-EXPORT_SYMBOL(drm_gem_vunmap);
+EXPORT_SYMBOL(drm_gem_vunmap_locked);
 
 void drm_gem_lock(struct drm_gem_object *obj)
 {
@@ -1262,25 +1262,25 @@ void drm_gem_unlock(struct drm_gem_object *obj)
 }
 EXPORT_SYMBOL(drm_gem_unlock);
 
-int drm_gem_vmap_unlocked(struct drm_gem_object *obj, struct iosys_map *map)
+int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map)
 {
 	int ret;
 
 	dma_resv_lock(obj->resv, NULL);
-	ret = drm_gem_vmap(obj, map);
+	ret = drm_gem_vmap_locked(obj, map);
 	dma_resv_unlock(obj->resv);
 
 	return ret;
 }
-EXPORT_SYMBOL(drm_gem_vmap_unlocked);
+EXPORT_SYMBOL(drm_gem_vmap);
 
-void drm_gem_vunmap_unlocked(struct drm_gem_object *obj, struct iosys_map *map)
+void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map)
 {
 	dma_resv_lock(obj->resv, NULL);
-	drm_gem_vunmap(obj, map);
+	drm_gem_vunmap_locked(obj, map);
 	dma_resv_unlock(obj->resv);
 }
-EXPORT_SYMBOL(drm_gem_vunmap_unlocked);
+EXPORT_SYMBOL(drm_gem_vunmap);
 
 /**
  * drm_gem_lock_reservations - Sets up the ww context and acquires
@@ -1543,10 +1543,10 @@ tail:
 EXPORT_SYMBOL(drm_gem_lru_scan);
 
 /**
- * drm_gem_evict - helper to evict backing pages for a GEM object
+ * drm_gem_evict_locked - helper to evict backing pages for a GEM object
  * @obj: obj in question
  */
-int drm_gem_evict(struct drm_gem_object *obj)
+int drm_gem_evict_locked(struct drm_gem_object *obj)
 {
 	dma_resv_assert_held(obj->resv);
 
@@ -1558,4 +1558,4 @@ int drm_gem_evict(struct drm_gem_object *obj)
 
 	return 0;
 }
-EXPORT_SYMBOL(drm_gem_evict);
+EXPORT_SYMBOL(drm_gem_evict_locked);
diff --git a/drivers/gpu/drm/drm_gem_framebuffer_helper.c b/drivers/gpu/drm/drm_gem_framebuffer_helper.c
index 0fbeb686e561..6f72e7a0f427 100644
--- a/drivers/gpu/drm/drm_gem_framebuffer_helper.c
+++ b/drivers/gpu/drm/drm_gem_framebuffer_helper.c
@@ -362,7 +362,7 @@ int drm_gem_fb_vmap(struct drm_framebuffer *fb, struct iosys_map *map,
 			ret = -EINVAL;
 			goto err_drm_gem_vunmap;
 		}
-		ret = drm_gem_vmap_unlocked(obj, &map[i]);
+		ret = drm_gem_vmap(obj, &map[i]);
 		if (ret)
 			goto err_drm_gem_vunmap;
 	}
@@ -384,7 +384,7 @@ err_drm_gem_vunmap:
 		obj = drm_gem_fb_get_obj(fb, i);
 		if (!obj)
 			continue;
-		drm_gem_vunmap_unlocked(obj, &map[i]);
+		drm_gem_vunmap(obj, &map[i]);
 	}
 	return ret;
 }
@@ -411,7 +411,7 @@ void drm_gem_fb_vunmap(struct drm_framebuffer *fb, struct iosys_map *map)
 			continue;
 		if (iosys_map_is_null(&map[i]))
 			continue;
-		drm_gem_vunmap_unlocked(obj, &map[i]);
+		drm_gem_vunmap(obj, &map[i]);
 	}
 }
 EXPORT_SYMBOL(drm_gem_fb_vunmap);
diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c
index d99dee67353a..aa43265f4f4f 100644
--- a/drivers/gpu/drm/drm_gem_shmem_helper.c
+++ b/drivers/gpu/drm/drm_gem_shmem_helper.c
@@ -165,7 +165,7 @@ void drm_gem_shmem_free(struct drm_gem_shmem_object *shmem)
 	} else {
 		dma_resv_lock(shmem->base.resv, NULL);
 
-		drm_WARN_ON(obj->dev, shmem->vmap_use_count);
+		drm_WARN_ON(obj->dev, refcount_read(&shmem->vmap_use_count));
 
 		if (shmem->sgt) {
 			dma_unmap_sgtable(obj->dev->dev, shmem->sgt,
@@ -174,9 +174,10 @@ void drm_gem_shmem_free(struct drm_gem_shmem_object *shmem)
 			kfree(shmem->sgt);
 		}
 		if (shmem->pages)
-			drm_gem_shmem_put_pages(shmem);
+			drm_gem_shmem_put_pages_locked(shmem);
 
-		drm_WARN_ON(obj->dev, shmem->pages_use_count);
+		drm_WARN_ON(obj->dev, refcount_read(&shmem->pages_use_count));
+		drm_WARN_ON(obj->dev, refcount_read(&shmem->pages_pin_count));
 
 		dma_resv_unlock(shmem->base.resv);
 	}
@@ -186,21 +187,20 @@ void drm_gem_shmem_free(struct drm_gem_shmem_object *shmem)
 }
 EXPORT_SYMBOL_GPL(drm_gem_shmem_free);
 
-static int drm_gem_shmem_get_pages(struct drm_gem_shmem_object *shmem)
+static int drm_gem_shmem_get_pages_locked(struct drm_gem_shmem_object *shmem)
 {
 	struct drm_gem_object *obj = &shmem->base;
 	struct page **pages;
 
 	dma_resv_assert_held(shmem->base.resv);
 
-	if (shmem->pages_use_count++ > 0)
+	if (refcount_inc_not_zero(&shmem->pages_use_count))
 		return 0;
 
 	pages = drm_gem_get_pages(obj);
 	if (IS_ERR(pages)) {
 		drm_dbg_kms(obj->dev, "Failed to get pages (%ld)\n",
 			    PTR_ERR(pages));
-		shmem->pages_use_count = 0;
 		return PTR_ERR(pages);
 	}
 
@@ -216,38 +216,36 @@ static int drm_gem_shmem_get_pages(struct drm_gem_shmem_object *shmem)
 
 	shmem->pages = pages;
 
+	refcount_set(&shmem->pages_use_count, 1);
+
 	return 0;
 }
 
 /*
- * drm_gem_shmem_put_pages - Decrease use count on the backing pages for a shmem GEM object
+ * drm_gem_shmem_put_pages_locked - Decrease use count on the backing pages for a shmem GEM object
  * @shmem: shmem GEM object
  *
  * This function decreases the use count and puts the backing pages when use drops to zero.
  */
-void drm_gem_shmem_put_pages(struct drm_gem_shmem_object *shmem)
+void drm_gem_shmem_put_pages_locked(struct drm_gem_shmem_object *shmem)
 {
 	struct drm_gem_object *obj = &shmem->base;
 
 	dma_resv_assert_held(shmem->base.resv);
 
-	if (drm_WARN_ON_ONCE(obj->dev, !shmem->pages_use_count))
-		return;
-
-	if (--shmem->pages_use_count > 0)
-		return;
-
+	if (refcount_dec_and_test(&shmem->pages_use_count)) {
 #ifdef CONFIG_X86
-	if (shmem->map_wc)
-		set_pages_array_wb(shmem->pages, obj->size >> PAGE_SHIFT);
+		if (shmem->map_wc)
+			set_pages_array_wb(shmem->pages, obj->size >> PAGE_SHIFT);
 #endif
 
-	drm_gem_put_pages(obj, shmem->pages,
-			  shmem->pages_mark_dirty_on_put,
-			  shmem->pages_mark_accessed_on_put);
-	shmem->pages = NULL;
+		drm_gem_put_pages(obj, shmem->pages,
+				  shmem->pages_mark_dirty_on_put,
+				  shmem->pages_mark_accessed_on_put);
+		shmem->pages = NULL;
+	}
 }
-EXPORT_SYMBOL(drm_gem_shmem_put_pages);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_put_pages_locked);
 
 int drm_gem_shmem_pin_locked(struct drm_gem_shmem_object *shmem)
 {
@@ -257,7 +255,12 @@ int drm_gem_shmem_pin_locked(struct drm_gem_shmem_object *shmem)
 
 	drm_WARN_ON(shmem->base.dev, drm_gem_is_imported(&shmem->base));
 
-	ret = drm_gem_shmem_get_pages(shmem);
+	if (refcount_inc_not_zero(&shmem->pages_pin_count))
+		return 0;
+
+	ret = drm_gem_shmem_get_pages_locked(shmem);
+	if (!ret)
+		refcount_set(&shmem->pages_pin_count, 1);
 
 	return ret;
 }
@@ -267,7 +270,8 @@ void drm_gem_shmem_unpin_locked(struct drm_gem_shmem_object *shmem)
 {
 	dma_resv_assert_held(shmem->base.resv);
 
-	drm_gem_shmem_put_pages(shmem);
+	if (refcount_dec_and_test(&shmem->pages_pin_count))
+		drm_gem_shmem_put_pages_locked(shmem);
 }
 EXPORT_SYMBOL(drm_gem_shmem_unpin_locked);
 
@@ -288,6 +292,9 @@ int drm_gem_shmem_pin(struct drm_gem_shmem_object *shmem)
 
 	drm_WARN_ON(obj->dev, drm_gem_is_imported(obj));
 
+	if (refcount_inc_not_zero(&shmem->pages_pin_count))
+		return 0;
+
 	ret = dma_resv_lock_interruptible(shmem->base.resv, NULL);
 	if (ret)
 		return ret;
@@ -296,7 +303,7 @@ int drm_gem_shmem_pin(struct drm_gem_shmem_object *shmem)
 
 	return ret;
 }
-EXPORT_SYMBOL(drm_gem_shmem_pin);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_pin);
 
 /**
  * drm_gem_shmem_unpin - Unpin backing pages for a shmem GEM object
@@ -311,14 +318,17 @@ void drm_gem_shmem_unpin(struct drm_gem_shmem_object *shmem)
 
 	drm_WARN_ON(obj->dev, drm_gem_is_imported(obj));
 
+	if (refcount_dec_not_one(&shmem->pages_pin_count))
+		return;
+
 	dma_resv_lock(shmem->base.resv, NULL);
 	drm_gem_shmem_unpin_locked(shmem);
 	dma_resv_unlock(shmem->base.resv);
 }
-EXPORT_SYMBOL(drm_gem_shmem_unpin);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_unpin);
 
 /*
- * drm_gem_shmem_vmap - Create a virtual mapping for a shmem GEM object
+ * drm_gem_shmem_vmap_locked - Create a virtual mapping for a shmem GEM object
  * @shmem: shmem GEM object
  * @map: Returns the kernel virtual address of the SHMEM GEM object's backing
  *       store.
@@ -327,47 +337,43 @@ EXPORT_SYMBOL(drm_gem_shmem_unpin);
  * exists for the buffer backing the shmem GEM object. It hides the differences
  * between dma-buf imported and natively allocated objects.
  *
- * Acquired mappings should be cleaned up by calling drm_gem_shmem_vunmap().
+ * Acquired mappings should be cleaned up by calling drm_gem_shmem_vunmap_locked().
  *
  * Returns:
  * 0 on success or a negative error code on failure.
  */
-int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem,
-		       struct iosys_map *map)
+int drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem,
+			      struct iosys_map *map)
 {
 	struct drm_gem_object *obj = &shmem->base;
 	int ret = 0;
 
 	if (drm_gem_is_imported(obj)) {
 		ret = dma_buf_vmap(obj->dma_buf, map);
-		if (!ret) {
-			if (drm_WARN_ON(obj->dev, map->is_iomem)) {
-				dma_buf_vunmap(obj->dma_buf, map);
-				return -EIO;
-			}
-		}
 	} else {
 		pgprot_t prot = PAGE_KERNEL;
 
 		dma_resv_assert_held(shmem->base.resv);
 
-		if (shmem->vmap_use_count++ > 0) {
+		if (refcount_inc_not_zero(&shmem->vmap_use_count)) {
 			iosys_map_set_vaddr(map, shmem->vaddr);
 			return 0;
 		}
 
-		ret = drm_gem_shmem_get_pages(shmem);
+		ret = drm_gem_shmem_pin_locked(shmem);
 		if (ret)
-			goto err_zero_use;
+			return ret;
 
 		if (shmem->map_wc)
 			prot = pgprot_writecombine(prot);
 		shmem->vaddr = vmap(shmem->pages, obj->size >> PAGE_SHIFT,
 				    VM_MAP, prot);
-		if (!shmem->vaddr)
+		if (!shmem->vaddr) {
 			ret = -ENOMEM;
-		else
+		} else {
 			iosys_map_set_vaddr(map, shmem->vaddr);
+			refcount_set(&shmem->vmap_use_count, 1);
+		}
 	}
 
 	if (ret) {
@@ -379,28 +385,26 @@ int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem,
 
 err_put_pages:
 	if (!drm_gem_is_imported(obj))
-		drm_gem_shmem_put_pages(shmem);
-err_zero_use:
-	shmem->vmap_use_count = 0;
+		drm_gem_shmem_unpin_locked(shmem);
 
 	return ret;
 }
-EXPORT_SYMBOL(drm_gem_shmem_vmap);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_vmap_locked);
 
 /*
- * drm_gem_shmem_vunmap - Unmap a virtual mapping for a shmem GEM object
+ * drm_gem_shmem_vunmap_locked - Unmap a virtual mapping for a shmem GEM object
  * @shmem: shmem GEM object
  * @map: Kernel virtual address where the SHMEM GEM object was mapped
  *
  * This function cleans up a kernel virtual address mapping acquired by
- * drm_gem_shmem_vmap(). The mapping is only removed when the use count drops to
- * zero.
+ * drm_gem_shmem_vmap_locked(). The mapping is only removed when the use count
+ * drops to zero.
  *
  * This function hides the differences between dma-buf imported and natively
  * allocated objects.
  */
-void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem,
-			  struct iosys_map *map)
+void drm_gem_shmem_vunmap_locked(struct drm_gem_shmem_object *shmem,
+				 struct iosys_map *map)
 {
 	struct drm_gem_object *obj = &shmem->base;
 
@@ -409,19 +413,15 @@ void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem,
 	} else {
 		dma_resv_assert_held(shmem->base.resv);
 
-		if (drm_WARN_ON_ONCE(obj->dev, !shmem->vmap_use_count))
-			return;
-
-		if (--shmem->vmap_use_count > 0)
-			return;
+		if (refcount_dec_and_test(&shmem->vmap_use_count)) {
+			vunmap(shmem->vaddr);
+			shmem->vaddr = NULL;
 
-		vunmap(shmem->vaddr);
-		drm_gem_shmem_put_pages(shmem);
+			drm_gem_shmem_unpin_locked(shmem);
+		}
 	}
-
-	shmem->vaddr = NULL;
 }
-EXPORT_SYMBOL(drm_gem_shmem_vunmap);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_vunmap_locked);
 
 static int
 drm_gem_shmem_create_with_handle(struct drm_file *file_priv,
@@ -449,7 +449,7 @@ drm_gem_shmem_create_with_handle(struct drm_file *file_priv,
 /* Update madvise status, returns true if not purged, else
  * false or -errno.
  */
-int drm_gem_shmem_madvise(struct drm_gem_shmem_object *shmem, int madv)
+int drm_gem_shmem_madvise_locked(struct drm_gem_shmem_object *shmem, int madv)
 {
 	dma_resv_assert_held(shmem->base.resv);
 
@@ -460,9 +460,9 @@ int drm_gem_shmem_madvise(struct drm_gem_shmem_object *shmem, int madv)
 
 	return (madv >= 0);
 }
-EXPORT_SYMBOL(drm_gem_shmem_madvise);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_madvise_locked);
 
-void drm_gem_shmem_purge(struct drm_gem_shmem_object *shmem)
+void drm_gem_shmem_purge_locked(struct drm_gem_shmem_object *shmem)
 {
 	struct drm_gem_object *obj = &shmem->base;
 	struct drm_device *dev = obj->dev;
@@ -476,7 +476,7 @@ void drm_gem_shmem_purge(struct drm_gem_shmem_object *shmem)
 	kfree(shmem->sgt);
 	shmem->sgt = NULL;
 
-	drm_gem_shmem_put_pages(shmem);
+	drm_gem_shmem_put_pages_locked(shmem);
 
 	shmem->madv = -1;
 
@@ -492,7 +492,7 @@ void drm_gem_shmem_purge(struct drm_gem_shmem_object *shmem)
 
 	invalidate_mapping_pages(file_inode(obj->filp)->i_mapping, 0, (loff_t)-1);
 }
-EXPORT_SYMBOL(drm_gem_shmem_purge);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_purge_locked);
 
 /**
  * drm_gem_shmem_dumb_create - Create a dumb shmem buffer object
@@ -575,8 +575,8 @@ static void drm_gem_shmem_vm_open(struct vm_area_struct *vma)
 	 * mmap'd, vm_open() just grabs an additional reference for the new
 	 * mm the vma is getting copied into (ie. on fork()).
 	 */
-	if (!drm_WARN_ON_ONCE(obj->dev, !shmem->pages_use_count))
-		shmem->pages_use_count++;
+	drm_WARN_ON_ONCE(obj->dev,
+			 !refcount_inc_not_zero(&shmem->pages_use_count));
 
 	dma_resv_unlock(shmem->base.resv);
 
@@ -589,7 +589,7 @@ static void drm_gem_shmem_vm_close(struct vm_area_struct *vma)
 	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
 
 	dma_resv_lock(shmem->base.resv, NULL);
-	drm_gem_shmem_put_pages(shmem);
+	drm_gem_shmem_put_pages_locked(shmem);
 	dma_resv_unlock(shmem->base.resv);
 
 	drm_gem_vm_close(vma);
@@ -639,7 +639,7 @@ int drm_gem_shmem_mmap(struct drm_gem_shmem_object *shmem, struct vm_area_struct
 		return -EINVAL;
 
 	dma_resv_lock(shmem->base.resv, NULL);
-	ret = drm_gem_shmem_get_pages(shmem);
+	ret = drm_gem_shmem_get_pages_locked(shmem);
 	dma_resv_unlock(shmem->base.resv);
 
 	if (ret)
@@ -666,11 +666,12 @@ void drm_gem_shmem_print_info(const struct drm_gem_shmem_object *shmem,
 	if (drm_gem_is_imported(&shmem->base))
 		return;
 
-	drm_printf_indent(p, indent, "pages_use_count=%u\n", shmem->pages_use_count);
-	drm_printf_indent(p, indent, "vmap_use_count=%u\n", shmem->vmap_use_count);
+	drm_printf_indent(p, indent, "pages_pin_count=%u\n", refcount_read(&shmem->pages_pin_count));
+	drm_printf_indent(p, indent, "pages_use_count=%u\n", refcount_read(&shmem->pages_use_count));
+	drm_printf_indent(p, indent, "vmap_use_count=%u\n", refcount_read(&shmem->vmap_use_count));
 	drm_printf_indent(p, indent, "vaddr=%p\n", shmem->vaddr);
 }
-EXPORT_SYMBOL(drm_gem_shmem_print_info);
+EXPORT_SYMBOL_GPL(drm_gem_shmem_print_info);
 
 /**
  * drm_gem_shmem_get_sg_table - Provide a scatter/gather table of pinned
@@ -707,7 +708,7 @@ static struct sg_table *drm_gem_shmem_get_pages_sgt_locked(struct drm_gem_shmem_
 
 	drm_WARN_ON(obj->dev, drm_gem_is_imported(obj));
 
-	ret = drm_gem_shmem_get_pages(shmem);
+	ret = drm_gem_shmem_get_pages_locked(shmem);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -729,7 +730,7 @@ err_free_sgt:
 	sg_free_table(sgt);
 	kfree(sgt);
 err_put_pages:
-	drm_gem_shmem_put_pages(shmem);
+	drm_gem_shmem_put_pages_locked(shmem);
 	return ERR_PTR(ret);
 }
 
diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index 38431e8360e7..f3ac2c78e3b2 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -1330,7 +1330,6 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
 	unsigned long num_dma_mapped;
 	unsigned int order = 0;
 	unsigned long *pfns;
-	struct page **pages;
 	int err = 0;
 	struct dev_pagemap *pagemap;
 	struct drm_pagemap *dpagemap;
@@ -1369,7 +1368,6 @@ retry:
 	if (err)
 		goto err_free;
 
-	pages = (struct page **)pfns;
 map_pages:
 	/*
 	 * Perform all dma mappings under the notifier lock to not
@@ -1444,8 +1442,6 @@ map_pages:
 				err = -EFAULT;
 				goto err_unmap;
 			}
-
-			pages[i] = page;
 		} else {
 			dma_addr_t addr;
 
@@ -1469,9 +1465,9 @@ map_pages:
 		}
 		i += 1 << order;
 		num_dma_mapped = i;
+		range->flags.has_dma_mapping = true;
 	}
 
-	range->flags.has_dma_mapping = true;
 	if (zdd) {
 		range->flags.has_devmem_pages = true;
 		range->dpagemap = dpagemap;
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index b2b6a8e49dda..e44f28fd81d3 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -179,8 +179,8 @@ int drm_gem_pin_locked(struct drm_gem_object *obj);
 void drm_gem_unpin_locked(struct drm_gem_object *obj);
 int drm_gem_pin(struct drm_gem_object *obj);
 void drm_gem_unpin(struct drm_gem_object *obj);
-int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map);
-void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map);
+int drm_gem_vmap_locked(struct drm_gem_object *obj, struct iosys_map *map);
+void drm_gem_vunmap_locked(struct drm_gem_object *obj, struct iosys_map *map);
 
 /* drm_debugfs.c drm_debugfs_crc.c */
 #if defined(CONFIG_DEBUG_FS)
diff --git a/drivers/gpu/drm/drm_mipi_dbi.c b/drivers/gpu/drm/drm_mipi_dbi.c
index 89e05a5bed1d..a4cd476f9b30 100644
--- a/drivers/gpu/drm/drm_mipi_dbi.c
+++ b/drivers/gpu/drm/drm_mipi_dbi.c
@@ -404,12 +404,16 @@ static void mipi_dbi_blank(struct mipi_dbi_dev *dbidev)
 	u16 height = drm->mode_config.min_height;
 	u16 width = drm->mode_config.min_width;
 	struct mipi_dbi *dbi = &dbidev->dbi;
-	size_t len = width * height * 2;
+	const struct drm_format_info *dst_format;
+	size_t len;
 	int idx;
 
 	if (!drm_dev_enter(drm, &idx))
 		return;
 
+	dst_format = drm_format_info(dbidev->pixel_format);
+	len = drm_format_info_min_pitch(dst_format, 0, width) * height;
+
 	memset(dbidev->tx_buf, 0, len);
 
 	mipi_dbi_set_window_address(dbidev, 0, width - 1, 0, height - 1);
diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index dfa595556320..e5184a0c2465 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -36,6 +36,8 @@
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_print.h>
 
+#include <linux/media-bus-format.h>
+
 #include <video/mipi_display.h>
 
 /**
@@ -871,6 +873,41 @@ ssize_t mipi_dsi_generic_read(struct mipi_dsi_device *dsi, const void *params,
 EXPORT_SYMBOL(mipi_dsi_generic_read);
 
 /**
+ * drm_mipi_dsi_get_input_bus_fmt() - Get the required MEDIA_BUS_FMT_* based
+ *				      input pixel format for a given DSI output
+ *				      pixel format
+ * @dsi_format: pixel format that a DSI host needs to output
+ *
+ * Various DSI hosts can use this function during their
+ * &drm_bridge_funcs.atomic_get_input_bus_fmts operation to ascertain
+ * the MEDIA_BUS_FMT_* pixel format required as input.
+ *
+ * RETURNS:
+ * a 32-bit MEDIA_BUS_FMT_* value on success or 0 in case of failure.
+ */
+u32 drm_mipi_dsi_get_input_bus_fmt(enum mipi_dsi_pixel_format dsi_format)
+{
+	switch (dsi_format) {
+	case MIPI_DSI_FMT_RGB888:
+		return MEDIA_BUS_FMT_RGB888_1X24;
+
+	case MIPI_DSI_FMT_RGB666:
+		return MEDIA_BUS_FMT_RGB666_1X24_CPADHI;
+
+	case MIPI_DSI_FMT_RGB666_PACKED:
+		return MEDIA_BUS_FMT_RGB666_1X18;
+
+	case MIPI_DSI_FMT_RGB565:
+		return MEDIA_BUS_FMT_RGB565_1X16;
+
+	default:
+		/* Unsupported DSI Format */
+		return 0;
+	}
+}
+EXPORT_SYMBOL(drm_mipi_dsi_get_input_bus_fmt);
+
+/**
  * mipi_dsi_dcs_write_buffer() - transmit a DCS command with payload
  * @dsi: DSI peripheral device
  * @data: buffer containing data to be transmitted
diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c
index 8642a2fb25a9..b4239fd04e9d 100644
--- a/drivers/gpu/drm/drm_mode_config.c
+++ b/drivers/gpu/drm/drm_mode_config.c
@@ -383,6 +383,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev)
 
 	prop = drm_property_create(dev,
 				   DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB,
+				   "IN_FORMATS_ASYNC", 0);
+	if (!prop)
+		return -ENOMEM;
+	dev->mode_config.async_modifiers_property = prop;
+
+	prop = drm_property_create(dev,
+				   DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB,
 				   "SIZE_HINTS", 0);
 	if (!prop)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/drm_panel.c b/drivers/gpu/drm/drm_panel.c
index c627e42a7ce7..650de4da0853 100644
--- a/drivers/gpu/drm/drm_panel.c
+++ b/drivers/gpu/drm/drm_panel.c
@@ -74,8 +74,9 @@ EXPORT_SYMBOL(drm_panel_init);
  * drm_panel_add - add a panel to the global registry
  * @panel: panel to add
  *
- * Add a panel to the global registry so that it can be looked up by display
- * drivers.
+ * Add a panel to the global registry so that it can be looked
+ * up by display drivers. The panel to be added must have been
+ * allocated by devm_drm_panel_alloc().
  */
 void drm_panel_add(struct drm_panel *panel)
 {
@@ -105,21 +106,21 @@ EXPORT_SYMBOL(drm_panel_remove);
  *
  * Calling this function will enable power and deassert any reset signals to
  * the panel. After this has completed it is possible to communicate with any
- * integrated circuitry via a command bus.
- *
- * Return: 0 on success or a negative error code on failure.
+ * integrated circuitry via a command bus. This function cannot fail (as it is
+ * called from the pre_enable call chain). There will always be a call to
+ * drm_panel_disable() afterwards.
  */
-int drm_panel_prepare(struct drm_panel *panel)
+void drm_panel_prepare(struct drm_panel *panel)
 {
 	struct drm_panel_follower *follower;
 	int ret;
 
 	if (!panel)
-		return -EINVAL;
+		return;
 
 	if (panel->prepared) {
 		dev_warn(panel->dev, "Skipping prepare of already prepared panel\n");
-		return 0;
+		return;
 	}
 
 	mutex_lock(&panel->follower_lock);
@@ -138,11 +139,8 @@ int drm_panel_prepare(struct drm_panel *panel)
 				 follower->funcs->panel_prepared, ret);
 	}
 
-	ret = 0;
 exit:
 	mutex_unlock(&panel->follower_lock);
-
-	return ret;
 }
 EXPORT_SYMBOL(drm_panel_prepare);
 
@@ -154,16 +152,14 @@ EXPORT_SYMBOL(drm_panel_prepare);
  * reset, turn off power supplies, ...). After this function has completed, it
  * is usually no longer possible to communicate with the panel until another
  * call to drm_panel_prepare().
- *
- * Return: 0 on success or a negative error code on failure.
  */
-int drm_panel_unprepare(struct drm_panel *panel)
+void drm_panel_unprepare(struct drm_panel *panel)
 {
 	struct drm_panel_follower *follower;
 	int ret;
 
 	if (!panel)
-		return -EINVAL;
+		return;
 
 	/*
 	 * If you are seeing the warning below it likely means one of two things:
@@ -176,7 +172,7 @@ int drm_panel_unprepare(struct drm_panel *panel)
 	 */
 	if (!panel->prepared) {
 		dev_warn(panel->dev, "Skipping unprepare of already unprepared panel\n");
-		return 0;
+		return;
 	}
 
 	mutex_lock(&panel->follower_lock);
@@ -195,11 +191,8 @@ int drm_panel_unprepare(struct drm_panel *panel)
 	}
 	panel->prepared = false;
 
-	ret = 0;
 exit:
 	mutex_unlock(&panel->follower_lock);
-
-	return ret;
 }
 EXPORT_SYMBOL(drm_panel_unprepare);
 
@@ -209,26 +202,26 @@ EXPORT_SYMBOL(drm_panel_unprepare);
  *
  * Calling this function will cause the panel display drivers to be turned on
  * and the backlight to be enabled. Content will be visible on screen after
- * this call completes.
- *
- * Return: 0 on success or a negative error code on failure.
+ * this call completes. This function cannot fail (as it is called from the
+ * enable call chain). There will always be a call to drm_panel_disable()
+ * afterwards.
  */
-int drm_panel_enable(struct drm_panel *panel)
+void drm_panel_enable(struct drm_panel *panel)
 {
 	int ret;
 
 	if (!panel)
-		return -EINVAL;
+		return;
 
 	if (panel->enabled) {
 		dev_warn(panel->dev, "Skipping enable of already enabled panel\n");
-		return 0;
+		return;
 	}
 
 	if (panel->funcs && panel->funcs->enable) {
 		ret = panel->funcs->enable(panel);
 		if (ret < 0)
-			return ret;
+			return;
 	}
 	panel->enabled = true;
 
@@ -236,8 +229,6 @@ int drm_panel_enable(struct drm_panel *panel)
 	if (ret < 0)
 		DRM_DEV_INFO(panel->dev, "failed to enable backlight: %d\n",
 			     ret);
-
-	return 0;
 }
 EXPORT_SYMBOL(drm_panel_enable);
 
@@ -248,15 +239,13 @@ EXPORT_SYMBOL(drm_panel_enable);
  * This will typically turn off the panel's backlight or disable the display
  * drivers. For smart panels it should still be possible to communicate with
  * the integrated circuitry via any command bus after this call.
- *
- * Return: 0 on success or a negative error code on failure.
  */
-int drm_panel_disable(struct drm_panel *panel)
+void drm_panel_disable(struct drm_panel *panel)
 {
 	int ret;
 
 	if (!panel)
-		return -EINVAL;
+		return;
 
 	/*
 	 * If you are seeing the warning below it likely means one of two things:
@@ -269,7 +258,7 @@ int drm_panel_disable(struct drm_panel *panel)
 	 */
 	if (!panel->enabled) {
 		dev_warn(panel->dev, "Skipping disable of already disabled panel\n");
-		return 0;
+		return;
 	}
 
 	ret = backlight_disable(panel->backlight);
@@ -280,11 +269,9 @@ int drm_panel_disable(struct drm_panel *panel)
 	if (panel->funcs && panel->funcs->disable) {
 		ret = panel->funcs->disable(panel);
 		if (ret < 0)
-			return ret;
+			return;
 	}
 	panel->enabled = false;
-
-	return 0;
 }
 EXPORT_SYMBOL(drm_panel_disable);
 
@@ -317,6 +304,93 @@ int drm_panel_get_modes(struct drm_panel *panel,
 }
 EXPORT_SYMBOL(drm_panel_get_modes);
 
+static void __drm_panel_free(struct kref *kref)
+{
+	struct drm_panel *panel = container_of(kref, struct drm_panel, refcount);
+
+	kfree(panel->container);
+}
+
+/**
+ * drm_panel_get - Acquire a panel reference
+ * @panel: DRM panel
+ *
+ * This function increments the panel's refcount.
+ * Returns:
+ * Pointer to @panel
+ */
+struct drm_panel *drm_panel_get(struct drm_panel *panel)
+{
+	if (!panel)
+		return panel;
+
+	kref_get(&panel->refcount);
+
+	return panel;
+}
+EXPORT_SYMBOL(drm_panel_get);
+
+/**
+ * drm_panel_put - Release a panel reference
+ * @panel: DRM panel
+ *
+ * This function decrements the panel's reference count and frees the
+ * object if the reference count drops to zero.
+ */
+void drm_panel_put(struct drm_panel *panel)
+{
+	if (panel)
+		kref_put(&panel->refcount, __drm_panel_free);
+}
+EXPORT_SYMBOL(drm_panel_put);
+
+/**
+ * drm_panel_put_void - wrapper to drm_panel_put() taking a void pointer
+ *
+ * @data: pointer to @struct drm_panel, cast to a void pointer
+ *
+ * Wrapper of drm_panel_put() to be used when a function taking a void
+ * pointer is needed, for example as a devm action.
+ */
+static void drm_panel_put_void(void *data)
+{
+	struct drm_panel *panel = (struct drm_panel *)data;
+
+	drm_panel_put(panel);
+}
+
+void *__devm_drm_panel_alloc(struct device *dev, size_t size, size_t offset,
+			     const struct drm_panel_funcs *funcs,
+			     int connector_type)
+{
+	void *container;
+	struct drm_panel *panel;
+	int err;
+
+	if (!funcs) {
+		dev_warn(dev, "Missing funcs pointer\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	container = kzalloc(size, GFP_KERNEL);
+	if (!container)
+		return ERR_PTR(-ENOMEM);
+
+	panel = container + offset;
+	panel->container = container;
+	panel->funcs = funcs;
+	kref_init(&panel->refcount);
+
+	err = devm_add_action_or_reset(dev, drm_panel_put_void, panel);
+	if (err)
+		return ERR_PTR(err);
+
+	drm_panel_init(panel, dev, funcs, connector_type);
+
+	return container;
+}
+EXPORT_SYMBOL(__devm_drm_panel_alloc);
+
 #ifdef CONFIG_OF
 /**
  * of_drm_find_panel - look up a panel using a device tree node
diff --git a/drivers/gpu/drm/drm_panel_orientation_quirks.c b/drivers/gpu/drm/drm_panel_orientation_quirks.c
index c554ad8f246b..7ac0fd5391fe 100644
--- a/drivers/gpu/drm/drm_panel_orientation_quirks.c
+++ b/drivers/gpu/drm/drm_panel_orientation_quirks.c
@@ -517,6 +517,12 @@ static const struct dmi_system_id orientation_data[] = {
 		  DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "LTH17"),
 		},
 		.driver_data = (void *)&lcd800x1280_rightside_up,
+	}, {	/* ZOTAC Gaming Zone */
+		.matches = {
+		  DMI_EXACT_MATCH(DMI_SYS_VENDOR, "ZOTAC"),
+		  DMI_EXACT_MATCH(DMI_BOARD_NAME, "G0A1W"),
+		},
+		.driver_data = (void *)&lcd1080x1920_leftside_up,
 	}, {	/* One Mix 2S (generic strings, also match on bios date) */
 		.matches = {
 		  DMI_EXACT_MATCH(DMI_SYS_VENDOR, "Default string"),
diff --git a/drivers/gpu/drm/drm_panic.c b/drivers/gpu/drm/drm_panic.c
index b47ea25fdfaa..b4de79583805 100644
--- a/drivers/gpu/drm/drm_panic.c
+++ b/drivers/gpu/drm/drm_panic.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/font.h>
+#include <linux/highmem.h>
 #include <linux/init.h>
 #include <linux/iosys-map.h>
 #include <linux/kdebug.h>
@@ -154,6 +155,90 @@ static void drm_panic_blit_pixel(struct drm_scanout_buffer *sb, struct drm_rect
 				sb->set_pixel(sb, clip->x1 + x, clip->y1 + y, fg_color);
 }
 
+static void drm_panic_write_pixel16(void *vaddr, unsigned int offset, u16 color)
+{
+	u16 *p = vaddr + offset;
+
+	*p = color;
+}
+
+static void drm_panic_write_pixel24(void *vaddr, unsigned int offset, u32 color)
+{
+	u8 *p = vaddr + offset;
+
+	*p++ = color & 0xff;
+	color >>= 8;
+	*p++ = color & 0xff;
+	color >>= 8;
+	*p = color & 0xff;
+}
+
+static void drm_panic_write_pixel32(void *vaddr, unsigned int offset, u32 color)
+{
+	u32 *p = vaddr + offset;
+
+	*p = color;
+}
+
+static void drm_panic_write_pixel(void *vaddr, unsigned int offset, u32 color, unsigned int cpp)
+{
+	switch (cpp) {
+	case 2:
+		drm_panic_write_pixel16(vaddr, offset, color);
+		break;
+	case 3:
+		drm_panic_write_pixel24(vaddr, offset, color);
+		break;
+	case 4:
+		drm_panic_write_pixel32(vaddr, offset, color);
+		break;
+	default:
+		pr_debug_once("Can't blit with pixel width %d\n", cpp);
+	}
+}
+
+/*
+ * The scanout buffer pages are not mapped, so for each pixel,
+ * use kmap_local_page_try_from_panic() to map the page, and write the pixel.
+ * Try to keep the map from the previous pixel, to avoid too much map/unmap.
+ */
+static void drm_panic_blit_page(struct page **pages, unsigned int dpitch,
+				unsigned int cpp, const u8 *sbuf8,
+				unsigned int spitch, struct drm_rect *clip,
+				unsigned int scale, u32 fg32)
+{
+	unsigned int y, x;
+	unsigned int page = ~0;
+	unsigned int height = drm_rect_height(clip);
+	unsigned int width = drm_rect_width(clip);
+	void *vaddr = NULL;
+
+	for (y = 0; y < height; y++) {
+		for (x = 0; x < width; x++) {
+			if (drm_draw_is_pixel_fg(sbuf8, spitch, x / scale, y / scale)) {
+				unsigned int new_page;
+				unsigned int offset;
+
+				offset = (y + clip->y1) * dpitch + (x + clip->x1) * cpp;
+				new_page = offset >> PAGE_SHIFT;
+				offset = offset % PAGE_SIZE;
+				if (new_page != page) {
+					if (!pages[new_page])
+						continue;
+					if (vaddr)
+						kunmap_local(vaddr);
+					page = new_page;
+					vaddr = kmap_local_page_try_from_panic(pages[page]);
+				}
+				if (vaddr)
+					drm_panic_write_pixel(vaddr, offset, fg32, cpp);
+			}
+		}
+	}
+	if (vaddr)
+		kunmap_local(vaddr);
+}
+
 /*
  * drm_panic_blit - convert a monochrome image to a linear framebuffer
  * @sb: destination scanout buffer
@@ -177,6 +262,10 @@ static void drm_panic_blit(struct drm_scanout_buffer *sb, struct drm_rect *clip,
 	if (sb->set_pixel)
 		return drm_panic_blit_pixel(sb, clip, sbuf8, spitch, scale, fg_color);
 
+	if (sb->pages)
+		return drm_panic_blit_page(sb->pages, sb->pitch[0], sb->format->cpp[0],
+					   sbuf8, spitch, clip, scale, fg_color);
+
 	map = sb->map[0];
 	iosys_map_incr(&map, clip->y1 * sb->pitch[0] + clip->x1 * sb->format->cpp[0]);
 
@@ -209,6 +298,35 @@ static void drm_panic_fill_pixel(struct drm_scanout_buffer *sb,
 			sb->set_pixel(sb, clip->x1 + x, clip->y1 + y, color);
 }
 
+static void drm_panic_fill_page(struct page **pages, unsigned int dpitch,
+				unsigned int cpp, struct drm_rect *clip,
+				u32 color)
+{
+	unsigned int y, x;
+	unsigned int page = ~0;
+	void *vaddr = NULL;
+
+	for (y = clip->y1; y < clip->y2; y++) {
+		for (x = clip->x1; x < clip->x2; x++) {
+			unsigned int new_page;
+			unsigned int offset;
+
+			offset = y * dpitch + x * cpp;
+			new_page = offset >> PAGE_SHIFT;
+			offset = offset % PAGE_SIZE;
+			if (new_page != page) {
+				if (vaddr)
+					kunmap_local(vaddr);
+				page = new_page;
+				vaddr = kmap_local_page_try_from_panic(pages[page]);
+			}
+			drm_panic_write_pixel(vaddr, offset, color, cpp);
+		}
+	}
+	if (vaddr)
+		kunmap_local(vaddr);
+}
+
 /*
  * drm_panic_fill - Fill a rectangle with a color
  * @sb: destination scanout buffer
@@ -225,6 +343,10 @@ static void drm_panic_fill(struct drm_scanout_buffer *sb, struct drm_rect *clip,
 	if (sb->set_pixel)
 		return drm_panic_fill_pixel(sb, clip, color);
 
+	if (sb->pages)
+		return drm_panic_fill_page(sb->pages, sb->pitch[0], sb->format->cpp[0],
+					   clip, color);
+
 	map = sb->map[0];
 	iosys_map_incr(&map, clip->y1 * sb->pitch[0] + clip->x1 * sb->format->cpp[0]);
 
@@ -709,16 +831,24 @@ static void draw_panic_plane(struct drm_plane *plane, const char *description)
 	if (!drm_panic_trylock(plane->dev, flags))
 		return;
 
+	ret = plane->helper_private->get_scanout_buffer(plane, &sb);
+
+	if (ret || !drm_panic_is_format_supported(sb.format))
+		goto unlock;
+
+	/* One of these should be set, or it can't draw pixels */
+	if (!sb.set_pixel && !sb.pages && iosys_map_is_null(&sb.map[0]))
+		goto unlock;
+
 	drm_panic_set_description(description);
 
-	ret = plane->helper_private->get_scanout_buffer(plane, &sb);
+	draw_panic_dispatch(&sb);
+	if (plane->helper_private->panic_flush)
+		plane->helper_private->panic_flush(plane);
 
-	if (!ret && drm_panic_is_format_supported(sb.format)) {
-		draw_panic_dispatch(&sb);
-		if (plane->helper_private->panic_flush)
-			plane->helper_private->panic_flush(plane);
-	}
 	drm_panic_clear_description();
+
+unlock:
 	drm_panic_unlock(plane->dev, flags);
 }
 
diff --git a/drivers/gpu/drm/drm_panic_qr.rs b/drivers/gpu/drm/drm_panic_qr.rs
index f2a99681b998..dd55b1cb764d 100644
--- a/drivers/gpu/drm/drm_panic_qr.rs
+++ b/drivers/gpu/drm/drm_panic_qr.rs
@@ -5,7 +5,7 @@
 //! It is called from a panic handler, so it should't allocate memory and
 //! does all the work on the stack or on the provided buffers. For
 //! simplification, it only supports low error correction, and applies the
-//! first mask (checkerboard). It will draw the smallest QRcode that can
+//! first mask (checkerboard). It will draw the smallest QR code that can
 //! contain the string passed as parameter. To get the most compact
 //! QR code, the start of the URL is encoded as binary, and the
 //! compressed kmsg is encoded as numeric.
@@ -315,7 +315,7 @@ impl Segment<'_> {
         }
     }
 
-    // Returns the size of the length field in bits, depending on QR Version.
+    /// Returns the size of the length field in bits, depending on QR Version.
     fn length_bits_count(&self, version: Version) -> usize {
         let Version(v) = version;
         match self {
@@ -331,7 +331,7 @@ impl Segment<'_> {
         }
     }
 
-    // Number of characters in the segment.
+    /// Number of characters in the segment.
     fn character_count(&self) -> usize {
         match self {
             Segment::Binary(data) => data.len(),
@@ -366,8 +366,48 @@ impl Segment<'_> {
         SegmentIterator {
             segment: self,
             offset: 0,
-            carry: 0,
-            carry_len: 0,
+            decfifo: Default::default(),
+        }
+    }
+}
+
+/// Max fifo size is 17 (max push) + 2 (max remaining)
+const MAX_FIFO_SIZE: usize = 19;
+
+/// A simple Decimal digit FIFO
+#[derive(Default)]
+struct DecFifo {
+    decimals: [u8; MAX_FIFO_SIZE],
+    len: usize,
+}
+
+impl DecFifo {
+    fn push(&mut self, data: u64, len: usize) {
+        let mut chunk = data;
+        for i in (0..self.len).rev() {
+            self.decimals[i + len] = self.decimals[i];
+        }
+        for i in 0..len {
+            self.decimals[i] = (chunk % 10) as u8;
+            chunk /= 10;
+        }
+        self.len += len;
+    }
+
+    /// Pop 3 decimal digits from the FIFO
+    fn pop3(&mut self) -> Option<(u16, usize)> {
+        if self.len == 0 {
+            None
+        } else {
+            let poplen = 3.min(self.len);
+            self.len -= poplen;
+            let mut out = 0;
+            let mut exp = 1;
+            for i in 0..poplen {
+                out += self.decimals[self.len + i] as u16 * exp;
+                exp *= 10;
+            }
+            Some((out, NUM_CHARS_BITS[poplen]))
         }
     }
 }
@@ -375,8 +415,7 @@ impl Segment<'_> {
 struct SegmentIterator<'a> {
     segment: &'a Segment<'a>,
     offset: usize,
-    carry: u64,
-    carry_len: usize,
+    decfifo: DecFifo,
 }
 
 impl Iterator for SegmentIterator<'_> {
@@ -394,31 +433,17 @@ impl Iterator for SegmentIterator<'_> {
                 }
             }
             Segment::Numeric(data) => {
-                if self.carry_len < 3 && self.offset < data.len() {
-                    // If there are less than 3 decimal digits in the carry,
-                    // take the next 7 bytes of input, and add them to the carry.
+                if self.decfifo.len < 3 && self.offset < data.len() {
+                    // If there are less than 3 decimal digits in the fifo,
+                    // take the next 7 bytes of input, and push them to the fifo.
                     let mut buf = [0u8; 8];
                     let len = 7.min(data.len() - self.offset);
                     buf[..len].copy_from_slice(&data[self.offset..self.offset + len]);
                     let chunk = u64::from_le_bytes(buf);
-                    let pow = u64::pow(10, BYTES_TO_DIGITS[len] as u32);
-                    self.carry = chunk + self.carry * pow;
+                    self.decfifo.push(chunk, BYTES_TO_DIGITS[len]);
                     self.offset += len;
-                    self.carry_len += BYTES_TO_DIGITS[len];
-                }
-                match self.carry_len {
-                    0 => None,
-                    len => {
-                        // take the next 3 decimal digits of the carry
-                        // and return 10bits of numeric data.
-                        let out_len = 3.min(len);
-                        self.carry_len -= out_len;
-                        let pow = u64::pow(10, self.carry_len as u32);
-                        let out = (self.carry / pow) as u16;
-                        self.carry = self.carry % pow;
-                        Some((out, NUM_CHARS_BITS[out_len]))
-                    }
                 }
+                self.decfifo.pop3()
             }
         }
     }
@@ -569,8 +594,8 @@ struct EncodedMsgIterator<'a> {
 impl Iterator for EncodedMsgIterator<'_> {
     type Item = u8;
 
-    // Send the bytes in interleaved mode, first byte of first block of group1,
-    // then first byte of second block of group1, ...
+    /// Send the bytes in interleaved mode, first byte of first block of group1,
+    /// then first byte of second block of group1, ...
     fn next(&mut self) -> Option<Self::Item> {
         let em = self.em;
         let blocks = em.g1_blocks + em.g2_blocks;
@@ -638,7 +663,7 @@ impl QrImage<'_> {
         self.data.fill(0);
     }
 
-    // Set pixel to light color.
+    /// Set pixel to light color.
     fn set(&mut self, x: u8, y: u8) {
         let off = y as usize * self.stride as usize + x as usize / 8;
         let mut v = self.data[off];
@@ -646,13 +671,13 @@ impl QrImage<'_> {
         self.data[off] = v;
     }
 
-    // Invert a module color.
+    /// Invert a module color.
     fn xor(&mut self, x: u8, y: u8) {
         let off = y as usize * self.stride as usize + x as usize / 8;
         self.data[off] ^= 0x80 >> (x % 8);
     }
 
-    // Draw a light square at (x, y) top left corner.
+    /// Draw a light square at (x, y) top left corner.
     fn draw_square(&mut self, x: u8, y: u8, size: u8) {
         for k in 0..size {
             self.set(x + k, y);
@@ -784,7 +809,7 @@ impl QrImage<'_> {
         vinfo != 0 && ((x >= pos && x < pos + 3 && y < 6) || (y >= pos && y < pos + 3 && x < 6))
     }
 
-    // Returns true if the module is reserved (Not usable for data and EC).
+    /// Returns true if the module is reserved (Not usable for data and EC).
     fn is_reserved(&self, x: u8, y: u8) -> bool {
         self.is_alignment(x, y)
             || self.is_finder(x, y)
@@ -793,13 +818,14 @@ impl QrImage<'_> {
             || self.is_version_info(x, y)
     }
 
-    // Last module to draw, at bottom left corner.
+    /// Last module to draw, at bottom left corner.
     fn is_last(&self, x: u8, y: u8) -> bool {
         x == 0 && y == self.width - 1
     }
 
-    // Move to the next module according to QR code order.
-    // From bottom right corner, to bottom left corner.
+    /// Move to the next module according to QR code order.
+    ///
+    /// From bottom right corner, to bottom left corner.
     fn next(&self, x: u8, y: u8) -> (u8, u8) {
         let x_adj = if x <= 6 { x + 1 } else { x };
         let column_type = (self.width - x_adj) % 4;
@@ -812,7 +838,7 @@ impl QrImage<'_> {
         }
     }
 
-    // Find next module that can hold data.
+    /// Find next module that can hold data.
     fn next_available(&self, x: u8, y: u8) -> (u8, u8) {
         let (mut x, mut y) = self.next(x, y);
         while self.is_reserved(x, y) && !self.is_last(x, y) {
@@ -841,7 +867,7 @@ impl QrImage<'_> {
         }
     }
 
-    // Apply checkerboard mask to all non-reserved modules.
+    /// Apply checkerboard mask to all non-reserved modules.
     fn apply_mask(&mut self) {
         for x in 0..self.width {
             for y in 0..self.width {
@@ -852,7 +878,7 @@ impl QrImage<'_> {
         }
     }
 
-    // Draw the QR code with the provided data iterator.
+    /// Draw the QR code with the provided data iterator.
     fn draw_all(&mut self, data: impl Iterator<Item = u8>) {
         // First clear the table, as it may have already some data.
         self.clear();
@@ -876,7 +902,7 @@ impl QrImage<'_> {
 ///   will be encoded as binary segment, otherwise it will be encoded
 ///   efficiently as a numeric segment, and appended to the URL.
 /// * `data_len`: Length of the data, that needs to be encoded, must be less
-///   than data_size.
+///   than `data_size`.
 /// * `data_size`: Size of data buffer, it should be at least 4071 bytes to hold
 ///   a V40 QR code. It will then be overwritten with the QR code image.
 /// * `tmp`: A temporary buffer that the QR code encoder will use, to write the
diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c
index a28b22fdd7a4..04992dfd4c79 100644
--- a/drivers/gpu/drm/drm_plane.c
+++ b/drivers/gpu/drm/drm_plane.c
@@ -141,6 +141,14 @@
  *     various bugs in this area with inconsistencies between the capability
  *     flag and per-plane properties.
  *
+ * IN_FORMATS_ASYNC:
+ *     Blob property which contains the set of buffer format and modifier
+ *     pairs supported by this plane for asynchronous flips. The blob is a struct
+ *     drm_format_modifier_blob. Userspace cannot change this property. This is an
+ *     optional property and if not present then user should expect a failure in
+ *     atomic ioctl when the modifier/format is not supported by that plane under
+ *     asynchronous flip.
+ *
  * SIZE_HINTS:
  *     Blob property which contains the set of recommended plane size
  *     which can used for simple "cursor like" use cases (eg. no scaling).
@@ -185,9 +193,13 @@ modifiers_ptr(struct drm_format_modifier_blob *blob)
 	return (struct drm_format_modifier *)(((char *)blob) + blob->modifiers_offset);
 }
 
-static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane)
+static struct drm_property_blob *create_in_format_blob(struct drm_device *dev,
+						       struct drm_plane *plane,
+						       bool (*format_mod_supported)
+						       (struct drm_plane *plane,
+							u32 format,
+							u64 modifier))
 {
-	const struct drm_mode_config *config = &dev->mode_config;
 	struct drm_property_blob *blob;
 	struct drm_format_modifier *mod;
 	size_t blob_size, formats_size, modifiers_size;
@@ -213,7 +225,7 @@ static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane
 
 	blob = drm_property_create_blob(dev, blob_size, NULL);
 	if (IS_ERR(blob))
-		return -1;
+		return NULL;
 
 	blob_data = blob->data;
 	blob_data->version = FORMAT_BLOB_CURRENT;
@@ -229,10 +241,10 @@ static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane
 	mod = modifiers_ptr(blob_data);
 	for (i = 0; i < plane->modifier_count; i++) {
 		for (j = 0; j < plane->format_count; j++) {
-			if (!plane->funcs->format_mod_supported ||
-			    plane->funcs->format_mod_supported(plane,
-							       plane->format_types[j],
-							       plane->modifiers[i])) {
+			if (!format_mod_supported ||
+			    format_mod_supported(plane,
+						 plane->format_types[j],
+						 plane->modifiers[i])) {
 				mod->formats |= 1ULL << j;
 			}
 		}
@@ -243,10 +255,7 @@ static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane
 		mod++;
 	}
 
-	drm_object_attach_property(&plane->base, config->modifiers_property,
-				   blob->base.id);
-
-	return 0;
+	return blob;
 }
 
 /**
@@ -358,6 +367,7 @@ static int __drm_universal_plane_init(struct drm_device *dev,
 				      const char *name, va_list ap)
 {
 	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_property_blob *blob;
 	static const uint64_t default_modifiers[] = {
 		DRM_FORMAT_MOD_LINEAR,
 	};
@@ -469,8 +479,24 @@ static int __drm_universal_plane_init(struct drm_device *dev,
 		drm_plane_create_hotspot_properties(plane);
 	}
 
-	if (format_modifier_count)
-		create_in_format_blob(dev, plane);
+	if (format_modifier_count) {
+		blob = create_in_format_blob(dev, plane,
+					     plane->funcs->format_mod_supported);
+		if (!IS_ERR(blob))
+			drm_object_attach_property(&plane->base,
+						   config->modifiers_property,
+						   blob->base.id);
+	}
+
+	if (plane->funcs->format_mod_supported_async) {
+		blob = create_in_format_blob(dev, plane,
+					     plane->funcs->format_mod_supported_async);
+		if (!IS_ERR(blob))
+			drm_object_attach_property(&plane->base,
+						   config->async_modifiers_property,
+						   blob->base.id);
+	}
+
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index bdb51c8f262e..d828502268b8 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -707,7 +707,7 @@ int drm_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct iosys_map *map)
 {
 	struct drm_gem_object *obj = dma_buf->priv;
 
-	return drm_gem_vmap(obj, map);
+	return drm_gem_vmap_locked(obj, map);
 }
 EXPORT_SYMBOL(drm_gem_dmabuf_vmap);
 
@@ -723,7 +723,7 @@ void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, struct iosys_map *map)
 {
 	struct drm_gem_object *obj = dma_buf->priv;
 
-	drm_gem_vunmap(obj, map);
+	drm_gem_vunmap_locked(obj, map);
 }
 EXPORT_SYMBOL(drm_gem_dmabuf_vunmap);
 
@@ -804,7 +804,6 @@ int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma)
 EXPORT_SYMBOL(drm_gem_dmabuf_mmap);
 
 static const struct dma_buf_ops drm_gem_prime_dmabuf_ops =  {
-	.cache_sgt_mapping = true,
 	.attach = drm_gem_map_attach,
 	.detach = drm_gem_map_detach,
 	.map_dma_buf = drm_gem_map_dma_buf,
@@ -998,7 +997,7 @@ EXPORT_SYMBOL(drm_gem_prime_import_dev);
 struct drm_gem_object *drm_gem_prime_import(struct drm_device *dev,
 					    struct dma_buf *dma_buf)
 {
-	return drm_gem_prime_import_dev(dev, dma_buf, dev->dev);
+	return drm_gem_prime_import_dev(dev, dma_buf, drm_dev_dma_dev(dev));
 }
 EXPORT_SYMBOL(drm_gem_prime_import);
 
diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c
index 7ba16323e7c2..6b3541159c0f 100644
--- a/drivers/gpu/drm/drm_probe_helper.c
+++ b/drivers/gpu/drm/drm_probe_helper.c
@@ -958,15 +958,16 @@ static void drm_kms_helper_poll_init_release(struct drm_device *dev, void *res)
  * cleaned up when the DRM device goes away.
  *
  * See drm_kms_helper_poll_init() for more information.
- *
- * Returns:
- * 0 on success, or a negative errno code otherwise.
  */
-int drmm_kms_helper_poll_init(struct drm_device *dev)
+void drmm_kms_helper_poll_init(struct drm_device *dev)
 {
+	int ret;
+
 	drm_kms_helper_poll_init(dev);
 
-	return drmm_add_action_or_reset(dev, drm_kms_helper_poll_init_release, dev);
+	ret = drmm_add_action_or_reset(dev, drm_kms_helper_poll_init_release, dev);
+	if (ret)
+		drm_warn(dev, "Connector status will not be updated, error %d\n", ret);
 }
 EXPORT_SYMBOL(drmm_kms_helper_poll_init);
 
diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c
index 4f2ab8a7b50f..636cd83ca29e 100644
--- a/drivers/gpu/drm/drm_syncobj.c
+++ b/drivers/gpu/drm/drm_syncobj.c
@@ -741,7 +741,7 @@ static int drm_syncobj_fd_to_handle(struct drm_file *file_private,
 }
 
 static int drm_syncobj_import_sync_file_fence(struct drm_file *file_private,
-					      int fd, int handle)
+					      int fd, int handle, u64 point)
 {
 	struct dma_fence *fence = sync_file_get_fence(fd);
 	struct drm_syncobj *syncobj;
@@ -755,14 +755,24 @@ static int drm_syncobj_import_sync_file_fence(struct drm_file *file_private,
 		return -ENOENT;
 	}
 
-	drm_syncobj_replace_fence(syncobj, fence);
+	if (point) {
+		struct dma_fence_chain *chain = dma_fence_chain_alloc();
+
+		if (!chain)
+			return -ENOMEM;
+
+		drm_syncobj_add_point(syncobj, chain, fence, point);
+	} else {
+		drm_syncobj_replace_fence(syncobj, fence);
+	}
+
 	dma_fence_put(fence);
 	drm_syncobj_put(syncobj);
 	return 0;
 }
 
 static int drm_syncobj_export_sync_file(struct drm_file *file_private,
-					int handle, int *p_fd)
+					int handle, u64 point, int *p_fd)
 {
 	int ret;
 	struct dma_fence *fence;
@@ -772,7 +782,7 @@ static int drm_syncobj_export_sync_file(struct drm_file *file_private,
 	if (fd < 0)
 		return fd;
 
-	ret = drm_syncobj_find_fence(file_private, handle, 0, 0, &fence);
+	ret = drm_syncobj_find_fence(file_private, handle, point, 0, &fence);
 	if (ret)
 		goto err_put_fd;
 
@@ -869,6 +879,9 @@ drm_syncobj_handle_to_fd_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file_private)
 {
 	struct drm_syncobj_handle *args = data;
+	unsigned int valid_flags = DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_TIMELINE |
+				   DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_EXPORT_SYNC_FILE;
+	u64 point = 0;
 
 	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
 		return -EOPNOTSUPP;
@@ -876,13 +889,18 @@ drm_syncobj_handle_to_fd_ioctl(struct drm_device *dev, void *data,
 	if (args->pad)
 		return -EINVAL;
 
-	if (args->flags != 0 &&
-	    args->flags != DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_EXPORT_SYNC_FILE)
+	if (args->flags & ~valid_flags)
 		return -EINVAL;
 
+	if (args->flags & DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_TIMELINE)
+		point = args->point;
+
 	if (args->flags & DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_EXPORT_SYNC_FILE)
 		return drm_syncobj_export_sync_file(file_private, args->handle,
-						    &args->fd);
+						    point, &args->fd);
+
+	if (args->point)
+		return -EINVAL;
 
 	return drm_syncobj_handle_to_fd(file_private, args->handle,
 					&args->fd);
@@ -893,6 +911,9 @@ drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file_private)
 {
 	struct drm_syncobj_handle *args = data;
+	unsigned int valid_flags = DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_TIMELINE |
+				   DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_IMPORT_SYNC_FILE;
+	u64 point = 0;
 
 	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
 		return -EOPNOTSUPP;
@@ -900,14 +921,20 @@ drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 	if (args->pad)
 		return -EINVAL;
 
-	if (args->flags != 0 &&
-	    args->flags != DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_IMPORT_SYNC_FILE)
+	if (args->flags & ~valid_flags)
 		return -EINVAL;
 
+	if (args->flags & DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_TIMELINE)
+		point = args->point;
+
 	if (args->flags & DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_IMPORT_SYNC_FILE)
 		return drm_syncobj_import_sync_file_fence(file_private,
 							  args->fd,
-							  args->handle);
+							  args->handle,
+							  point);
+
+	if (args->point)
+		return -EINVAL;
 
 	return drm_syncobj_fd_to_handle(file_private, args->fd,
 					&args->handle);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
index 42e57d142554..917ad527c961 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
@@ -39,7 +39,7 @@ int etnaviv_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map)
 
 int etnaviv_gem_prime_pin(struct drm_gem_object *obj)
 {
-	if (!obj->import_attach) {
+	if (!drm_gem_is_imported(obj)) {
 		struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj);
 
 		mutex_lock(&etnaviv_obj->lock);
@@ -51,7 +51,7 @@ int etnaviv_gem_prime_pin(struct drm_gem_object *obj)
 
 void etnaviv_gem_prime_unpin(struct drm_gem_object *obj)
 {
-	if (!obj->import_attach) {
+	if (!drm_gem_is_imported(obj)) {
 		struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj);
 
 		mutex_lock(&etnaviv_obj->lock);
@@ -65,7 +65,7 @@ static void etnaviv_gem_prime_release(struct etnaviv_gem_object *etnaviv_obj)
 	struct iosys_map map = IOSYS_MAP_INIT_VADDR(etnaviv_obj->vaddr);
 
 	if (etnaviv_obj->vaddr)
-		dma_buf_vunmap_unlocked(etnaviv_obj->base.import_attach->dmabuf, &map);
+		dma_buf_vunmap_unlocked(etnaviv_obj->base.dma_buf, &map);
 
 	/* Don't drop the pages for imported dmabuf, as they are not
 	 * ours, just free the array we allocated:
@@ -82,7 +82,7 @@ static void *etnaviv_gem_prime_vmap_impl(struct etnaviv_gem_object *etnaviv_obj)
 
 	lockdep_assert_held(&etnaviv_obj->lock);
 
-	ret = dma_buf_vmap(etnaviv_obj->base.import_attach->dmabuf, &map);
+	ret = dma_buf_vmap(etnaviv_obj->base.dma_buf, &map);
 	if (ret)
 		return NULL;
 	return map.vaddr;
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index 5170f72b0830..f91daefa9d2b 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -43,13 +43,13 @@ struct decon_data {
 	unsigned int wincon_burstlen_shift;
 };
 
-static struct decon_data exynos7_decon_data = {
+static const struct decon_data exynos7_decon_data = {
 	.vidw_buf_start_base = 0x80,
 	.shadowcon_win_protect_shift = 10,
 	.wincon_burstlen_shift = 11,
 };
 
-static struct decon_data exynos7870_decon_data = {
+static const struct decon_data exynos7870_decon_data = {
 	.vidw_buf_start_base = 0x880,
 	.shadowcon_win_protect_shift = 8,
 	.wincon_burstlen_shift = 10,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index f313ae7bc3a3..6cc7bf77bcac 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -355,8 +355,7 @@ static void exynos_drm_platform_shutdown(struct platform_device *pdev)
 {
 	struct drm_device *drm = platform_get_drvdata(pdev);
 
-	if (drm)
-		drm_atomic_helper_shutdown(drm);
+	drm_atomic_helper_shutdown(drm);
 }
 
 static struct platform_driver exynos_drm_platform_driver = {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index b150cfd92f6e..09e33a26caaf 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -908,7 +908,7 @@ static void fimc_dst_set_buf_seq(struct fimc_context *ctx, u32 buf_id,
 	u32 buf_num;
 	u32 cfg;
 
-	DRM_DEV_DEBUG_KMS(ctx->dev, "buf_id[%d]enqueu[%d]\n", buf_id, enqueue);
+	DRM_DEV_DEBUG_KMS(ctx->dev, "buf_id[%d]enqueue[%d]\n", buf_id, enqueue);
 
 	spin_lock_irqsave(&ctx->lock, flags);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 1ad87584b1c2..c394cc702d7d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -731,7 +731,7 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win,
 	/*
 	 * Setting dma-burst to 16Word causes permanent tearing for very small
 	 * buffers, e.g. cursor buffer. Burst Mode switching which based on
-	 * plane size is not recommended as plane size varies alot towards the
+	 * plane size is not recommended as plane size varies a lot towards the
 	 * end of the screen and rapid movement causes unstable DMA, but it is
 	 * still better to change dma-burst than displaying garbage.
 	 */
diff --git a/drivers/gpu/drm/exynos/exynos_drm_mic.c b/drivers/gpu/drm/exynos/exynos_drm_mic.c
index b34ec6728337..29a8366513fa 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_mic.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_mic.c
@@ -379,11 +379,11 @@ static int exynos_mic_probe(struct platform_device *pdev)
 	struct resource res;
 	int ret, i;
 
-	mic = devm_kzalloc(dev, sizeof(*mic), GFP_KERNEL);
-	if (!mic) {
+	mic = devm_drm_bridge_alloc(dev, struct exynos_mic, bridge, &mic_bridge_funcs);
+	if (IS_ERR(mic)) {
 		DRM_DEV_ERROR(dev,
 			      "mic: Failed to allocate memory for MIC object\n");
-		ret = -ENOMEM;
+		ret = PTR_ERR(mic);
 		goto err;
 	}
 
@@ -421,7 +421,6 @@ static int exynos_mic_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, mic);
 
-	mic->bridge.funcs = &mic_bridge_funcs;
 	mic->bridge.of_node = dev->of_node;
 
 	drm_bridge_add(&mic->bridge);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 08cf79a62025..e644e2382d77 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -312,9 +312,6 @@ static int vidi_get_modes(struct drm_connector *connector)
 	else
 		drm_edid = drm_edid_alloc(fake_edid_info, sizeof(fake_edid_info));
 
-	if (!drm_edid)
-		return 0;
-
 	drm_edid_connector_update(connector, drm_edid);
 
 	count = drm_edid_connector_add_modes(connector);
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
index 03b076db9381..3bbfc1b56a65 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
@@ -260,7 +260,6 @@ static int fsl_dcu_drm_probe(struct platform_device *pdev)
 	struct fsl_dcu_drm_device *fsl_dev;
 	struct drm_device *drm;
 	struct device *dev = &pdev->dev;
-	struct resource *res;
 	void __iomem *base;
 	struct clk *pix_clk_in;
 	char pix_clk_name[32];
@@ -278,8 +277,7 @@ static int fsl_dcu_drm_probe(struct platform_device *pdev)
 		return -ENODEV;
 	fsl_dev->soc = id->data;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	base = devm_ioremap_resource(dev, res);
+	base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(base)) {
 		ret = PTR_ERR(base);
 		return ret;
diff --git a/drivers/gpu/drm/gma500/mmu.c b/drivers/gpu/drm/gma500/mmu.c
index 4d78b33eaa82..e6753282e70e 100644
--- a/drivers/gpu/drm/gma500/mmu.c
+++ b/drivers/gpu/drm/gma500/mmu.c
@@ -730,44 +730,3 @@ out:
 
 	return ret;
 }
-
-int psb_mmu_virtual_to_pfn(struct psb_mmu_pd *pd, uint32_t virtual,
-			   unsigned long *pfn)
-{
-	int ret;
-	struct psb_mmu_pt *pt;
-	uint32_t tmp;
-	spinlock_t *lock = &pd->driver->lock;
-
-	down_read(&pd->driver->sem);
-	pt = psb_mmu_pt_map_lock(pd, virtual);
-	if (!pt) {
-		uint32_t *v;
-
-		spin_lock(lock);
-		v = kmap_atomic(pd->p);
-		tmp = v[psb_mmu_pd_index(virtual)];
-		kunmap_atomic(v);
-		spin_unlock(lock);
-
-		if (tmp != pd->invalid_pde || !(tmp & PSB_PTE_VALID) ||
-		    !(pd->invalid_pte & PSB_PTE_VALID)) {
-			ret = -EINVAL;
-			goto out;
-		}
-		ret = 0;
-		*pfn = pd->invalid_pte >> PAGE_SHIFT;
-		goto out;
-	}
-	tmp = pt->v[psb_mmu_pt_index(virtual)];
-	if (!(tmp & PSB_PTE_VALID)) {
-		ret = -EINVAL;
-	} else {
-		ret = 0;
-		*pfn = tmp >> PAGE_SHIFT;
-	}
-	psb_mmu_pt_unmap_unlock(pt);
-out:
-	up_read(&pd->driver->sem);
-	return ret;
-}
diff --git a/drivers/gpu/drm/gma500/mmu.h b/drivers/gpu/drm/gma500/mmu.h
index d4b5720ef08e..e6d39703718c 100644
--- a/drivers/gpu/drm/gma500/mmu.h
+++ b/drivers/gpu/drm/gma500/mmu.h
@@ -71,8 +71,6 @@ extern int psb_mmu_insert_pfn_sequence(struct psb_mmu_pd *pd,
 				       uint32_t start_pfn,
 				       unsigned long address,
 				       uint32_t num_pages, int type);
-extern int psb_mmu_virtual_to_pfn(struct psb_mmu_pd *pd, uint32_t virtual,
-				  unsigned long *pfn);
 extern void psb_mmu_set_pd_context(struct psb_mmu_pd *pd, int hw_context);
 extern int psb_mmu_insert_pages(struct psb_mmu_pd *pd, struct page **pages,
 				unsigned long address, uint32_t num_pages,
diff --git a/drivers/gpu/drm/gma500/oaktrail_crtc.c b/drivers/gpu/drm/gma500/oaktrail_crtc.c
index de8ccfe9890f..ea9b41af0867 100644
--- a/drivers/gpu/drm/gma500/oaktrail_crtc.c
+++ b/drivers/gpu/drm/gma500/oaktrail_crtc.c
@@ -658,10 +658,3 @@ const struct drm_crtc_helper_funcs oaktrail_helper_funcs = {
 	.prepare = gma_crtc_prepare,
 	.commit = gma_crtc_commit,
 };
-
-/* Not used yet */
-const struct gma_clock_funcs mrst_clock_funcs = {
-	.clock = mrst_lvds_clock,
-	.limit = mrst_limit,
-	.pll_is_valid = gma_pll_is_valid,
-};
diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h
index 9dc9dcd1b09f..979ea8ecf0d5 100644
--- a/drivers/gpu/drm/gma500/psb_intel_drv.h
+++ b/drivers/gpu/drm/gma500/psb_intel_drv.h
@@ -182,7 +182,6 @@ struct gma_i2c_chan *gma_i2c_create(struct drm_device *dev, const u32 reg,
 void gma_i2c_destroy(struct gma_i2c_chan *chan);
 int psb_intel_ddc_get_modes(struct drm_connector *connector,
 			    struct i2c_adapter *adapter);
-extern bool psb_intel_ddc_probe(struct i2c_adapter *adapter);
 
 extern void psb_intel_crtc_init(struct drm_device *dev, int pipe,
 			    struct psb_intel_mode_device *mode_dev);
diff --git a/drivers/gpu/drm/gma500/psb_intel_modes.c b/drivers/gpu/drm/gma500/psb_intel_modes.c
index 8be0ec340de5..45b10f30a2a9 100644
--- a/drivers/gpu/drm/gma500/psb_intel_modes.c
+++ b/drivers/gpu/drm/gma500/psb_intel_modes.c
@@ -12,37 +12,6 @@
 #include "psb_intel_drv.h"
 
 /**
- * psb_intel_ddc_probe
- * @adapter:   Associated I2C adaptor
- */
-bool psb_intel_ddc_probe(struct i2c_adapter *adapter)
-{
-	u8 out_buf[] = { 0x0, 0x0 };
-	u8 buf[2];
-	int ret;
-	struct i2c_msg msgs[] = {
-		{
-		 .addr = 0x50,
-		 .flags = 0,
-		 .len = 1,
-		 .buf = out_buf,
-		 },
-		{
-		 .addr = 0x50,
-		 .flags = I2C_M_RD,
-		 .len = 1,
-		 .buf = buf,
-		 }
-	};
-
-	ret = i2c_transfer(adapter, msgs, 2);
-	if (ret == 2)
-		return true;
-
-	return false;
-}
-
-/**
  * psb_intel_ddc_get_modes - get modelist from monitor
  * @connector: DRM connector device to use
  * @adapter:   Associated I2C adaptor
diff --git a/drivers/gpu/drm/gud/gud_drv.c b/drivers/gpu/drm/gud/gud_drv.c
index cb405771d6e2..5385a2126e45 100644
--- a/drivers/gpu/drm/gud/gud_drv.c
+++ b/drivers/gpu/drm/gud/gud_drv.c
@@ -309,21 +309,6 @@ out:
 	return ret;
 }
 
-/*
- * FIXME: Dma-buf sharing requires DMA support by the importing device.
- *        This function is a workaround to make USB devices work as well.
- *        See todo.rst for how to fix the issue in the dma-buf framework.
- */
-static struct drm_gem_object *gud_gem_prime_import(struct drm_device *drm, struct dma_buf *dma_buf)
-{
-	struct gud_device *gdrm = to_gud_device(drm);
-
-	if (!gdrm->dmadev)
-		return ERR_PTR(-ENODEV);
-
-	return drm_gem_prime_import_dev(drm, dma_buf, gdrm->dmadev);
-}
-
 static int gud_stats_debugfs(struct seq_file *m, void *data)
 {
 	struct drm_debugfs_entry *entry = m->private;
@@ -376,7 +361,6 @@ static const struct drm_driver gud_drm_driver = {
 	.driver_features	= DRIVER_MODESET | DRIVER_GEM | DRIVER_ATOMIC,
 	.fops			= &gud_fops,
 	DRM_GEM_SHMEM_DRIVER_OPS,
-	.gem_prime_import	= gud_gem_prime_import,
 	DRM_FBDEV_SHMEM_DRIVER_OPS,
 
 	.name			= "gud",
@@ -434,6 +418,7 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id)
 	size_t max_buffer_size = 0;
 	struct gud_device *gdrm;
 	struct drm_device *drm;
+	struct device *dma_dev;
 	u8 *formats_dev;
 	u32 *formats;
 	int ret, i;
@@ -609,17 +594,19 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id)
 
 	usb_set_intfdata(intf, gdrm);
 
-	gdrm->dmadev = usb_intf_get_dma_device(intf);
-	if (!gdrm->dmadev)
-		dev_warn(dev, "buffer sharing not supported");
+	dma_dev = usb_intf_get_dma_device(intf);
+	if (dma_dev) {
+		drm_dev_set_dma_dev(drm, dma_dev);
+		put_device(dma_dev);
+	} else {
+		dev_warn(dev, "buffer sharing not supported"); /* not an error */
+	}
 
 	drm_debugfs_add_file(drm, "stats", gud_stats_debugfs, NULL);
 
 	ret = drm_dev_register(drm, 0);
-	if (ret) {
-		put_device(gdrm->dmadev);
+	if (ret)
 		return ret;
-	}
 
 	drm_kms_helper_poll_init(drm);
 
@@ -638,8 +625,6 @@ static void gud_disconnect(struct usb_interface *interface)
 	drm_kms_helper_poll_fini(drm);
 	drm_dev_unplug(drm);
 	drm_atomic_helper_shutdown(drm);
-	put_device(gdrm->dmadev);
-	gdrm->dmadev = NULL;
 }
 
 static int gud_suspend(struct usb_interface *intf, pm_message_t message)
diff --git a/drivers/gpu/drm/gud/gud_internal.h b/drivers/gpu/drm/gud/gud_internal.h
index 0d148a6f27aa..d6fb25388722 100644
--- a/drivers/gpu/drm/gud/gud_internal.h
+++ b/drivers/gpu/drm/gud/gud_internal.h
@@ -16,7 +16,6 @@
 struct gud_device {
 	struct drm_device drm;
 	struct drm_simple_display_pipe pipe;
-	struct device *dmadev;
 	struct work_struct work;
 	u32 flags;
 	const struct drm_format_info *xrgb8888_emulation_format;
diff --git a/drivers/gpu/drm/hisilicon/hibmc/Makefile b/drivers/gpu/drm/hisilicon/hibmc/Makefile
index 95a4ed599d98..1f65c683282f 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/Makefile
+++ b/drivers/gpu/drm/hisilicon/hibmc/Makefile
@@ -1,5 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0-only
 hibmc-drm-y := hibmc_drm_drv.o hibmc_drm_de.o hibmc_drm_vdac.o hibmc_drm_i2c.o \
-	       dp/dp_aux.o dp/dp_link.o dp/dp_hw.o hibmc_drm_dp.o
+	       dp/dp_aux.o dp/dp_link.o dp/dp_hw.o dp/dp_serdes.o hibmc_drm_dp.o \
+	       hibmc_drm_debugfs.o
 
 obj-$(CONFIG_DRM_HISI_HIBMC) += hibmc-drm.o
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_aux.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_aux.c
index 0a903cce1fa9..8732cd1d8cb6 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_aux.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_aux.c
@@ -8,6 +8,7 @@
 #include <drm/drm_print.h>
 #include "dp_comm.h"
 #include "dp_reg.h"
+#include "dp_hw.h"
 
 #define HIBMC_AUX_CMD_REQ_LEN		GENMASK(7, 4)
 #define HIBMC_AUX_CMD_ADDR		GENMASK(27, 8)
@@ -124,7 +125,8 @@ static int hibmc_dp_aux_parse_xfer(struct hibmc_dp_dev *dp, struct drm_dp_aux_ms
 /* ret >= 0 ,ret is size; ret < 0, ret is err code */
 static ssize_t hibmc_dp_aux_xfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 {
-	struct hibmc_dp_dev *dp = container_of(aux, struct hibmc_dp_dev, aux);
+	struct hibmc_dp *dp_priv = container_of(aux, struct hibmc_dp, aux);
+	struct hibmc_dp_dev *dp = dp_priv->dp_dev;
 	u32 aux_cmd;
 	int ret;
 	u32 val; /* val will be assigned at the beginning of readl_poll_timeout function */
@@ -151,14 +153,16 @@ static ssize_t hibmc_dp_aux_xfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *
 	return hibmc_dp_aux_parse_xfer(dp, msg);
 }
 
-void hibmc_dp_aux_init(struct hibmc_dp_dev *dp)
+void hibmc_dp_aux_init(struct hibmc_dp *dp)
 {
-	hibmc_dp_reg_write_field(dp, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_SYNC_LEN_SEL, 0x0);
-	hibmc_dp_reg_write_field(dp, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_TIMER_TIMEOUT, 0x1);
-	hibmc_dp_reg_write_field(dp, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_MIN_PULSE_NUM,
+	hibmc_dp_reg_write_field(dp->dp_dev, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_SYNC_LEN_SEL, 0x0);
+	hibmc_dp_reg_write_field(dp->dp_dev, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_TIMER_TIMEOUT, 0x1);
+	hibmc_dp_reg_write_field(dp->dp_dev, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_MIN_PULSE_NUM,
 				 HIBMC_DP_MIN_PULSE_NUM);
 
 	dp->aux.transfer = hibmc_dp_aux_xfer;
-	dp->aux.is_remote = 0;
+	dp->aux.name = "HIBMC DRM dp aux";
+	dp->aux.drm_dev = dp->drm_dev;
 	drm_dp_aux_init(&dp->aux);
+	dp->dp_dev->aux = &dp->aux;
 }
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h
index 2c52a4476c4d..4add05c7f161 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_comm.h
@@ -13,6 +13,8 @@
 #include <linux/io.h>
 #include <drm/display/drm_dp_helper.h>
 
+#include "dp_hw.h"
+
 #define HIBMC_DP_LANE_NUM_MAX 2
 
 struct hibmc_link_status {
@@ -32,12 +34,13 @@ struct hibmc_dp_link {
 };
 
 struct hibmc_dp_dev {
-	struct drm_dp_aux aux;
+	struct drm_dp_aux *aux;
 	struct drm_device *dev;
 	void __iomem *base;
 	struct mutex lock; /* protects concurrent RW in hibmc_dp_reg_write_field() */
 	struct hibmc_dp_link link;
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
+	void __iomem *serdes_base;
 };
 
 #define dp_field_modify(reg_value, mask, val)				\
@@ -57,7 +60,10 @@ struct hibmc_dp_dev {
 		mutex_unlock(&_dp->lock);				\
 	} while (0)
 
-void hibmc_dp_aux_init(struct hibmc_dp_dev *dp);
+void hibmc_dp_aux_init(struct hibmc_dp *dp);
 int hibmc_dp_link_training(struct hibmc_dp_dev *dp);
+int hibmc_dp_serdes_init(struct hibmc_dp_dev *dp);
+int hibmc_dp_serdes_rate_switch(u8 rate, struct hibmc_dp_dev *dp);
+int hibmc_dp_serdes_set_tx_cfg(struct hibmc_dp_dev *dp, u8 train_set[HIBMC_DP_LANE_NUM_MAX]);
 
 #endif
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_config.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_config.h
index 74dd9956144e..08f9e1caf7fc 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_config.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_config.h
@@ -15,5 +15,7 @@
 #define HIBMC_DP_CLK_EN			0x7
 #define HIBMC_DP_SYNC_EN_MASK		0x3
 #define HIBMC_DP_LINK_RATE_CAL		27
+#define HIBMC_DP_SYNC_DELAY(lanes)	((lanes) == 0x2 ? 86 : 46)
+#define HIBMC_DP_INT_ENABLE		0xc
 
 #endif
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
index a8d543881c09..8f0daec7d174 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.c
@@ -72,6 +72,9 @@ static void hibmc_dp_set_sst(struct hibmc_dp_dev *dp, struct drm_display_mode *m
 				 HIBMC_DP_CFG_STREAM_HTOTAL_SIZE, htotal_size);
 	hibmc_dp_reg_write_field(dp, HIBMC_DP_VIDEO_HORIZONTAL_SIZE,
 				 HIBMC_DP_CFG_STREAM_HBLANK_SIZE, hblank_size);
+	hibmc_dp_reg_write_field(dp, HIBMC_DP_VIDEO_PACKET,
+				 HIBMC_DP_CFG_STREAM_SYNC_CALIBRATION,
+				 HIBMC_DP_SYNC_DELAY(dp->link.cap.lanes));
 }
 
 static void hibmc_dp_link_cfg(struct hibmc_dp_dev *dp, struct drm_display_mode *mode)
@@ -151,6 +154,7 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 {
 	struct drm_device *drm_dev = dp->drm_dev;
 	struct hibmc_dp_dev *dp_dev;
+	int ret;
 
 	dp_dev = devm_kzalloc(drm_dev->dev, sizeof(struct hibmc_dp_dev), GFP_KERNEL);
 	if (!dp_dev)
@@ -163,10 +167,14 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 	dp_dev->dev = drm_dev;
 	dp_dev->base = dp->mmio + HIBMC_DP_OFFSET;
 
-	hibmc_dp_aux_init(dp_dev);
+	hibmc_dp_aux_init(dp);
+
+	ret = hibmc_dp_serdes_init(dp_dev);
+	if (ret)
+		return ret;
 
 	dp_dev->link.cap.lanes = 0x2;
-	dp_dev->link.cap.link_rate = DP_LINK_BW_2_7;
+	dp_dev->link.cap.link_rate = DP_LINK_BW_8_1;
 
 	/* hdcp data */
 	writel(HIBMC_DP_HDCP, dp_dev->base + HIBMC_DP_HDCP_CFG);
@@ -181,6 +189,36 @@ int hibmc_dp_hw_init(struct hibmc_dp *dp)
 	return 0;
 }
 
+void hibmc_dp_enable_int(struct hibmc_dp *dp)
+{
+	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
+
+	writel(HIBMC_DP_INT_ENABLE, dp_dev->base + HIBMC_DP_INTR_ENABLE);
+}
+
+void hibmc_dp_disable_int(struct hibmc_dp *dp)
+{
+	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
+
+	writel(0, dp_dev->base + HIBMC_DP_INTR_ENABLE);
+	writel(HIBMC_DP_INT_RST, dp_dev->base + HIBMC_DP_INTR_ORIGINAL_STATUS);
+}
+
+void hibmc_dp_hpd_cfg(struct hibmc_dp *dp)
+{
+	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
+
+	hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_SYNC_LEN_SEL, 0x0);
+	hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_TIMER_TIMEOUT, 0x1);
+	hibmc_dp_reg_write_field(dp->dp_dev, HIBMC_DP_AUX_REQ, HIBMC_DP_CFG_AUX_MIN_PULSE_NUM, 0x9);
+	writel(HIBMC_DP_HDCP, dp_dev->base + HIBMC_DP_HDCP_CFG);
+	writel(0, dp_dev->base + HIBMC_DP_INTR_ENABLE);
+	writel(HIBMC_DP_INT_RST, dp_dev->base + HIBMC_DP_INTR_ORIGINAL_STATUS);
+	writel(HIBMC_DP_INT_ENABLE, dp_dev->base + HIBMC_DP_INTR_ENABLE);
+	writel(HIBMC_DP_DPTX_RST, dp_dev->base + HIBMC_DP_DPTX_RST_CTRL);
+	writel(HIBMC_DP_CLK_EN, dp_dev->base + HIBMC_DP_DPTX_CLK_CTRL);
+}
+
 void hibmc_dp_display_en(struct hibmc_dp *dp, bool enable)
 {
 	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
@@ -218,3 +256,52 @@ int hibmc_dp_mode_set(struct hibmc_dp *dp, struct drm_display_mode *mode)
 
 	return 0;
 }
+
+void hibmc_dp_reset_link(struct hibmc_dp *dp)
+{
+	dp->dp_dev->link.status.clock_recovered = false;
+	dp->dp_dev->link.status.channel_equalized = false;
+}
+
+static const struct hibmc_dp_color_raw g_rgb_raw[] = {
+	{CBAR_COLOR_BAR, 0x000, 0x000, 0x000},
+	{CBAR_WHITE,     0xfff, 0xfff, 0xfff},
+	{CBAR_RED,       0xfff, 0x000, 0x000},
+	{CBAR_ORANGE,    0xfff, 0x800, 0x000},
+	{CBAR_YELLOW,    0xfff, 0xfff, 0x000},
+	{CBAR_GREEN,     0x000, 0xfff, 0x000},
+	{CBAR_CYAN,      0x000, 0x800, 0x800},
+	{CBAR_BLUE,      0x000, 0x000, 0xfff},
+	{CBAR_PURPLE,    0x800, 0x000, 0x800},
+	{CBAR_BLACK,     0x000, 0x000, 0x000},
+};
+
+void hibmc_dp_set_cbar(struct hibmc_dp *dp, const struct hibmc_dp_cbar_cfg *cfg)
+{
+	struct hibmc_dp_dev *dp_dev = dp->dp_dev;
+	struct hibmc_dp_color_raw raw_data;
+
+	if (cfg->enable) {
+		hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL, BIT(9),
+					 cfg->self_timing);
+		hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL, GENMASK(8, 1),
+					 cfg->dynamic_rate);
+		if (cfg->pattern == CBAR_COLOR_BAR) {
+			hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL, BIT(10), 0);
+		} else {
+			raw_data = g_rgb_raw[cfg->pattern];
+			drm_dbg_dp(dp->drm_dev, "r:%x g:%x b:%x\n", raw_data.r_value,
+				   raw_data.g_value, raw_data.b_value);
+			hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL, BIT(10), 1);
+			hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL, GENMASK(23, 12),
+						 raw_data.r_value);
+			hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL1, GENMASK(23, 12),
+						 raw_data.g_value);
+			hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL1, GENMASK(11, 0),
+						 raw_data.b_value);
+		}
+	}
+
+	hibmc_dp_reg_write_field(dp_dev, HIBMC_DP_COLOR_BAR_CTRL, BIT(0), cfg->enable);
+	writel(HIBMC_DP_SYNC_EN_MASK, dp_dev->base + HIBMC_DP_TIMING_SYNC_CTRL);
+}
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
index 4dc13b3d9875..665f5b166dfb 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_hw.h
@@ -10,19 +10,55 @@
 #include <drm/drm_encoder.h>
 #include <drm/drm_connector.h>
 #include <drm/drm_print.h>
+#include <drm/display/drm_dp_helper.h>
 
 struct hibmc_dp_dev;
 
+enum hibmc_dp_cbar_pattern {
+	CBAR_COLOR_BAR,
+	CBAR_WHITE,
+	CBAR_RED,
+	CBAR_ORANGE,
+	CBAR_YELLOW,
+	CBAR_GREEN,
+	CBAR_CYAN,
+	CBAR_BLUE,
+	CBAR_PURPLE,
+	CBAR_BLACK,
+};
+
+struct hibmc_dp_color_raw {
+	enum hibmc_dp_cbar_pattern pattern;
+	u32 r_value;
+	u32 g_value;
+	u32 b_value;
+};
+
+struct hibmc_dp_cbar_cfg {
+	u8 enable;
+	u8 self_timing;
+	u8 dynamic_rate; /* 0:static, 1-255(frame):dynamic */
+	enum hibmc_dp_cbar_pattern pattern;
+};
+
 struct hibmc_dp {
 	struct hibmc_dp_dev *dp_dev;
 	struct drm_device *drm_dev;
 	struct drm_encoder encoder;
 	struct drm_connector connector;
 	void __iomem *mmio;
+	struct drm_dp_aux aux;
+	struct hibmc_dp_cbar_cfg cfg;
+	u32 irq_status;
 };
 
 int hibmc_dp_hw_init(struct hibmc_dp *dp);
 int hibmc_dp_mode_set(struct hibmc_dp *dp, struct drm_display_mode *mode);
 void hibmc_dp_display_en(struct hibmc_dp *dp, bool enable);
+void hibmc_dp_set_cbar(struct hibmc_dp *dp, const struct hibmc_dp_cbar_cfg *cfg);
+void hibmc_dp_reset_link(struct hibmc_dp *dp);
+void hibmc_dp_hpd_cfg(struct hibmc_dp *dp);
+void hibmc_dp_enable_int(struct hibmc_dp *dp);
+void hibmc_dp_disable_int(struct hibmc_dp *dp);
 
 #endif
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c
index f6355c16cc0a..74f7832ea53e 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.c
@@ -9,6 +9,22 @@
 
 #define HIBMC_EQ_MAX_RETRY 5
 
+static inline int hibmc_dp_get_serdes_rate_cfg(struct hibmc_dp_dev *dp)
+{
+	switch (dp->link.cap.link_rate) {
+	case DP_LINK_BW_1_62:
+		return DP_SERDES_BW_1_62;
+	case DP_LINK_BW_2_7:
+		return DP_SERDES_BW_2_7;
+	case DP_LINK_BW_5_4:
+		return DP_SERDES_BW_5_4;
+	case DP_LINK_BW_8_1:
+		return DP_SERDES_BW_8_1;
+	default:
+		return -EINVAL;
+	}
+}
+
 static int hibmc_dp_link_training_configure(struct hibmc_dp_dev *dp)
 {
 	u8 buf[2];
@@ -26,7 +42,7 @@ static int hibmc_dp_link_training_configure(struct hibmc_dp_dev *dp)
 	/* set rate and lane count */
 	buf[0] = dp->link.cap.link_rate;
 	buf[1] = DP_LANE_COUNT_ENHANCED_FRAME_EN | dp->link.cap.lanes;
-	ret = drm_dp_dpcd_write(&dp->aux, DP_LINK_BW_SET, buf, sizeof(buf));
+	ret = drm_dp_dpcd_write(dp->aux, DP_LINK_BW_SET, buf, sizeof(buf));
 	if (ret != sizeof(buf)) {
 		drm_dbg_dp(dp->dev, "dp aux write link rate and lanes failed, ret: %d\n", ret);
 		return ret >= 0 ? -EIO : ret;
@@ -35,17 +51,13 @@ static int hibmc_dp_link_training_configure(struct hibmc_dp_dev *dp)
 	/* set 8b/10b and downspread */
 	buf[0] = DP_SPREAD_AMP_0_5;
 	buf[1] = DP_SET_ANSI_8B10B;
-	ret = drm_dp_dpcd_write(&dp->aux, DP_DOWNSPREAD_CTRL, buf, sizeof(buf));
+	ret = drm_dp_dpcd_write(dp->aux, DP_DOWNSPREAD_CTRL, buf, sizeof(buf));
 	if (ret != sizeof(buf)) {
 		drm_dbg_dp(dp->dev, "dp aux write 8b/10b and downspread failed, ret: %d\n", ret);
 		return ret >= 0 ? -EIO : ret;
 	}
 
-	ret = drm_dp_read_dpcd_caps(&dp->aux, dp->dpcd);
-	if (ret)
-		drm_err(dp->dev, "dp aux read dpcd failed, ret: %d\n", ret);
-
-	return ret;
+	return 0;
 }
 
 static int hibmc_dp_link_set_pattern(struct hibmc_dp_dev *dp, int pattern)
@@ -84,7 +96,7 @@ static int hibmc_dp_link_set_pattern(struct hibmc_dp_dev *dp, int pattern)
 
 	hibmc_dp_reg_write_field(dp, HIBMC_DP_PHYIF_CTRL0, HIBMC_DP_CFG_PAT_SEL, val);
 
-	ret = drm_dp_dpcd_write(&dp->aux, DP_TRAINING_PATTERN_SET, &buf, sizeof(buf));
+	ret = drm_dp_dpcd_write(dp->aux, DP_TRAINING_PATTERN_SET, &buf, sizeof(buf));
 	if (ret != sizeof(buf)) {
 		drm_dbg_dp(dp->dev, "dp aux write training pattern set failed\n");
 		return ret >= 0 ? -EIO : ret;
@@ -108,9 +120,13 @@ static int hibmc_dp_link_training_cr_pre(struct hibmc_dp_dev *dp)
 		return ret;
 
 	for (i = 0; i < dp->link.cap.lanes; i++)
-		train_set[i] = DP_TRAIN_VOLTAGE_SWING_LEVEL_2;
+		train_set[i] = DP_TRAIN_VOLTAGE_SWING_LEVEL_0;
+
+	ret = hibmc_dp_serdes_set_tx_cfg(dp, dp->link.train_set);
+	if (ret)
+		return ret;
 
-	ret = drm_dp_dpcd_write(&dp->aux, DP_TRAINING_LANE0_SET, train_set, dp->link.cap.lanes);
+	ret = drm_dp_dpcd_write(dp->aux, DP_TRAINING_LANE0_SET, train_set, dp->link.cap.lanes);
 	if (ret != dp->link.cap.lanes) {
 		drm_dbg_dp(dp->dev, "dp aux write training lane set failed\n");
 		return ret >= 0 ? -EIO : ret;
@@ -137,21 +153,29 @@ static bool hibmc_dp_link_get_adjust_train(struct hibmc_dp_dev *dp,
 	return false;
 }
 
-static inline int hibmc_dp_link_reduce_rate(struct hibmc_dp_dev *dp)
+static int hibmc_dp_link_reduce_rate(struct hibmc_dp_dev *dp)
 {
+	int ret;
+
 	switch (dp->link.cap.link_rate) {
 	case DP_LINK_BW_2_7:
 		dp->link.cap.link_rate = DP_LINK_BW_1_62;
-		return 0;
+		break;
 	case DP_LINK_BW_5_4:
 		dp->link.cap.link_rate = DP_LINK_BW_2_7;
-		return 0;
+		break;
 	case DP_LINK_BW_8_1:
 		dp->link.cap.link_rate = DP_LINK_BW_5_4;
-		return 0;
+		break;
 	default:
 		return -EINVAL;
 	}
+
+	ret = hibmc_dp_get_serdes_rate_cfg(dp);
+	if (ret < 0)
+		return ret;
+
+	return hibmc_dp_serdes_rate_switch(ret, dp);
 }
 
 static inline int hibmc_dp_link_reduce_lane(struct hibmc_dp_dev *dp)
@@ -159,6 +183,7 @@ static inline int hibmc_dp_link_reduce_lane(struct hibmc_dp_dev *dp)
 	switch (dp->link.cap.lanes) {
 	case 0x2:
 		dp->link.cap.lanes--;
+		drm_dbg_dp(dp->dev, "dp link training reduce to 1 lane\n");
 		break;
 	case 0x1:
 		drm_err(dp->dev, "dp link training reduce lane failed, already reach minimum\n");
@@ -185,10 +210,10 @@ static int hibmc_dp_link_training_cr(struct hibmc_dp_dev *dp)
 
 	voltage_tries = 1;
 	for (cr_tries = 0; cr_tries < 80; cr_tries++) {
-		drm_dp_link_train_clock_recovery_delay(&dp->aux, dp->dpcd);
+		drm_dp_link_train_clock_recovery_delay(dp->aux, dp->dpcd);
 
-		ret = drm_dp_dpcd_read_link_status(&dp->aux, lane_status);
-		if (ret != DP_LINK_STATUS_SIZE) {
+		ret = drm_dp_dpcd_read_link_status(dp->aux, lane_status);
+		if (ret) {
 			drm_err(dp->dev, "Get lane status failed\n");
 			return ret;
 		}
@@ -206,7 +231,12 @@ static int hibmc_dp_link_training_cr(struct hibmc_dp_dev *dp)
 		}
 
 		level_changed = hibmc_dp_link_get_adjust_train(dp, lane_status);
-		ret = drm_dp_dpcd_write(&dp->aux, DP_TRAINING_LANE0_SET, dp->link.train_set,
+
+		ret = hibmc_dp_serdes_set_tx_cfg(dp, dp->link.train_set);
+		if (ret)
+			return ret;
+
+		ret = drm_dp_dpcd_write(dp->aux, DP_TRAINING_LANE0_SET, dp->link.train_set,
 					dp->link.cap.lanes);
 		if (ret != dp->link.cap.lanes) {
 			drm_dbg_dp(dp->dev, "Update link training failed\n");
@@ -233,10 +263,10 @@ static int hibmc_dp_link_training_channel_eq(struct hibmc_dp_dev *dp)
 		return ret;
 
 	for (eq_tries = 0; eq_tries < HIBMC_EQ_MAX_RETRY; eq_tries++) {
-		drm_dp_link_train_channel_eq_delay(&dp->aux, dp->dpcd);
+		drm_dp_link_train_channel_eq_delay(dp->aux, dp->dpcd);
 
-		ret = drm_dp_dpcd_read_link_status(&dp->aux, lane_status);
-		if (ret != DP_LINK_STATUS_SIZE) {
+		ret = drm_dp_dpcd_read_link_status(dp->aux, lane_status);
+		if (ret) {
 			drm_err(dp->dev, "get lane status failed\n");
 			break;
 		}
@@ -255,7 +285,12 @@ static int hibmc_dp_link_training_channel_eq(struct hibmc_dp_dev *dp)
 		}
 
 		hibmc_dp_link_get_adjust_train(dp, lane_status);
-		ret = drm_dp_dpcd_write(&dp->aux, DP_TRAINING_LANE0_SET,
+
+		ret = hibmc_dp_serdes_set_tx_cfg(dp, dp->link.train_set);
+		if (ret)
+			return ret;
+
+		ret = drm_dp_dpcd_write(dp->aux, DP_TRAINING_LANE0_SET,
 					dp->link.train_set, dp->link.cap.lanes);
 		if (ret != dp->link.cap.lanes) {
 			drm_dbg_dp(dp->dev, "Update link training failed\n");
@@ -295,6 +330,21 @@ int hibmc_dp_link_training(struct hibmc_dp_dev *dp)
 	struct hibmc_dp_link *link = &dp->link;
 	int ret;
 
+	ret = drm_dp_read_dpcd_caps(dp->aux, dp->dpcd);
+	if (ret)
+		drm_err(dp->dev, "dp aux read dpcd failed, ret: %d\n", ret);
+
+	dp->link.cap.link_rate = dp->dpcd[DP_MAX_LINK_RATE];
+	dp->link.cap.lanes = 0x2;
+
+	ret = hibmc_dp_get_serdes_rate_cfg(dp);
+	if (ret < 0)
+		return ret;
+
+	ret = hibmc_dp_serdes_rate_switch(ret, dp);
+	if (ret)
+		return ret;
+
 	while (true) {
 		ret = hibmc_dp_link_training_cr_pre(dp);
 		if (ret)
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_reg.h b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_reg.h
index 4a515c726d52..394b1e933c3a 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_reg.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_reg.h
@@ -5,72 +5,128 @@
 #define DP_REG_H
 
 #define HIBMC_DP_AUX_CMD_ADDR			0x50
+
 #define HIBMC_DP_AUX_WR_DATA0			0x54
 #define HIBMC_DP_AUX_WR_DATA1			0x58
 #define HIBMC_DP_AUX_WR_DATA2			0x5c
 #define HIBMC_DP_AUX_WR_DATA3			0x60
 #define HIBMC_DP_AUX_RD_DATA0			0x64
+
 #define HIBMC_DP_AUX_REQ			0x74
+#define HIBMC_DP_CFG_AUX_REQ			BIT(0)
+#define HIBMC_DP_CFG_AUX_SYNC_LEN_SEL		BIT(1)
+#define HIBMC_DP_CFG_AUX_TIMER_TIMEOUT		BIT(2)
+#define HIBMC_DP_CFG_AUX_MIN_PULSE_NUM		GENMASK(13, 9)
+
 #define HIBMC_DP_AUX_STATUS			0x78
+#define HIBMC_DP_CFG_AUX_TIMEOUT		BIT(0)
+#define HIBMC_DP_CFG_AUX_STATUS			GENMASK(11, 4)
+#define HIBMC_DP_CFG_AUX_READY_DATA_BYTE	GENMASK(16, 12)
+#define HIBMC_DP_CFG_AUX			GENMASK(24, 17)
+
 #define HIBMC_DP_PHYIF_CTRL0			0xa0
+#define HIBMC_DP_CFG_SCRAMBLE_EN		BIT(0)
+#define HIBMC_DP_CFG_PAT_SEL			GENMASK(7, 4)
+#define HIBMC_DP_CFG_LANE_DATA_EN		GENMASK(11, 8)
+
 #define HIBMC_DP_VIDEO_CTRL			0x100
+#define HIBMC_DP_CFG_STREAM_RGB_ENABLE		BIT(1)
+#define HIBMC_DP_CFG_STREAM_VIDEO_MAPPING	GENMASK(5, 2)
+#define HIBMC_DP_CFG_STREAM_FRAME_MODE		BIT(6)
+#define HIBMC_DP_CFG_STREAM_HSYNC_POLARITY	BIT(7)
+#define HIBMC_DP_CFG_STREAM_VSYNC_POLARITY	BIT(8)
+
 #define HIBMC_DP_VIDEO_CONFIG0			0x104
+#define HIBMC_DP_CFG_STREAM_HACTIVE		GENMASK(31, 16)
+#define HIBMC_DP_CFG_STREAM_HBLANK		GENMASK(15, 0)
+
 #define HIBMC_DP_VIDEO_CONFIG1			0x108
+#define HIBMC_DP_CFG_STREAM_VACTIVE		GENMASK(31, 16)
+#define HIBMC_DP_CFG_STREAM_VBLANK		GENMASK(15, 0)
+
 #define HIBMC_DP_VIDEO_CONFIG2			0x10c
+#define HIBMC_DP_CFG_STREAM_HSYNC_WIDTH		GENMASK(15, 0)
+
 #define HIBMC_DP_VIDEO_CONFIG3			0x110
+#define HIBMC_DP_CFG_STREAM_VSYNC_WIDTH		GENMASK(15, 0)
+#define HIBMC_DP_CFG_STREAM_VFRONT_PORCH	GENMASK(31, 16)
+
 #define HIBMC_DP_VIDEO_PACKET			0x114
+#define HIBMC_DP_CFG_STREAM_TU_SYMBOL_SIZE	GENMASK(5, 0)
+#define HIBMC_DP_CFG_STREAM_TU_SYMBOL_FRAC_SIZE	GENMASK(9, 6)
+#define HIBMC_DP_CFG_STREAM_SYNC_CALIBRATION	GENMASK(31, 20)
+
 #define HIBMC_DP_VIDEO_MSA0			0x118
+#define HIBMC_DP_CFG_STREAM_VSTART		GENMASK(31, 16)
+#define HIBMC_DP_CFG_STREAM_HSTART		GENMASK(15, 0)
+
 #define HIBMC_DP_VIDEO_MSA1			0x11c
 #define HIBMC_DP_VIDEO_MSA2			0x120
+
 #define HIBMC_DP_VIDEO_HORIZONTAL_SIZE		0X124
+#define HIBMC_DP_CFG_STREAM_HTOTAL_SIZE		GENMASK(31, 16)
+#define HIBMC_DP_CFG_STREAM_HBLANK_SIZE		GENMASK(15, 0)
+
+#define HIBMC_DP_COLOR_BAR_CTRL			0x260
+#define HIBMC_DP_COLOR_BAR_CTRL1		0x264
+
 #define HIBMC_DP_TIMING_GEN_CONFIG0		0x26c
+#define HIBMC_DP_CFG_TIMING_GEN0_HACTIVE	GENMASK(31, 16)
+#define HIBMC_DP_CFG_TIMING_GEN0_HBLANK		GENMASK(15, 0)
+
 #define HIBMC_DP_TIMING_GEN_CONFIG2		0x274
+#define HIBMC_DP_CFG_TIMING_GEN0_VACTIVE	GENMASK(31, 16)
+#define HIBMC_DP_CFG_TIMING_GEN0_VBLANK		GENMASK(15, 0)
+
 #define HIBMC_DP_TIMING_GEN_CONFIG3		0x278
+#define HIBMC_DP_CFG_TIMING_GEN0_VFRONT_PORCH	GENMASK(31, 16)
+
 #define HIBMC_DP_HDCP_CFG			0x600
+
 #define HIBMC_DP_DPTX_RST_CTRL			0x700
+#define HIBMC_DP_CFG_AUX_RST_N			BIT(4)
+
 #define HIBMC_DP_DPTX_CLK_CTRL			0x704
+
 #define HIBMC_DP_DPTX_GCTL0			0x708
+#define HIBMC_DP_CFG_PHY_LANE_NUM		GENMASK(2, 1)
+
 #define HIBMC_DP_INTR_ENABLE			0x720
 #define HIBMC_DP_INTR_ORIGINAL_STATUS		0x728
+
 #define HIBMC_DP_TIMING_MODEL_CTRL		0x884
+#define HIBMC_DP_CFG_PIXEL_NUM_TIMING_MODE_SEL1	GENMASK(31, 16)
+
 #define HIBMC_DP_TIMING_SYNC_CTRL		0xFF0
 
-#define HIBMC_DP_CFG_AUX_SYNC_LEN_SEL		BIT(1)
-#define HIBMC_DP_CFG_AUX_TIMER_TIMEOUT		BIT(2)
-#define HIBMC_DP_CFG_STREAM_FRAME_MODE		BIT(6)
-#define HIBMC_DP_CFG_AUX_MIN_PULSE_NUM		GENMASK(13, 9)
-#define HIBMC_DP_CFG_LANE_DATA_EN		GENMASK(11, 8)
-#define HIBMC_DP_CFG_PHY_LANE_NUM		GENMASK(2, 1)
-#define HIBMC_DP_CFG_AUX_REQ			BIT(0)
-#define HIBMC_DP_CFG_AUX_RST_N			BIT(4)
-#define HIBMC_DP_CFG_AUX_TIMEOUT		BIT(0)
-#define HIBMC_DP_CFG_AUX_READY_DATA_BYTE	GENMASK(16, 12)
-#define HIBMC_DP_CFG_AUX			GENMASK(24, 17)
-#define HIBMC_DP_CFG_AUX_STATUS			GENMASK(11, 4)
-#define HIBMC_DP_CFG_SCRAMBLE_EN		BIT(0)
-#define HIBMC_DP_CFG_PAT_SEL			GENMASK(7, 4)
-#define HIBMC_DP_CFG_TIMING_GEN0_HACTIVE	GENMASK(31, 16)
-#define HIBMC_DP_CFG_TIMING_GEN0_HBLANK		GENMASK(15, 0)
-#define HIBMC_DP_CFG_TIMING_GEN0_VACTIVE	GENMASK(31, 16)
-#define HIBMC_DP_CFG_TIMING_GEN0_VBLANK		GENMASK(15, 0)
-#define HIBMC_DP_CFG_TIMING_GEN0_VFRONT_PORCH	GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_HACTIVE		GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_HBLANK		GENMASK(15, 0)
-#define HIBMC_DP_CFG_STREAM_HSYNC_WIDTH		GENMASK(15, 0)
-#define HIBMC_DP_CFG_STREAM_VACTIVE		GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_VBLANK		GENMASK(15, 0)
-#define HIBMC_DP_CFG_STREAM_VFRONT_PORCH	GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_VSYNC_WIDTH		GENMASK(15, 0)
-#define HIBMC_DP_CFG_STREAM_VSTART		GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_HSTART		GENMASK(15, 0)
-#define HIBMC_DP_CFG_STREAM_VSYNC_POLARITY	BIT(8)
-#define HIBMC_DP_CFG_STREAM_HSYNC_POLARITY	BIT(7)
-#define HIBMC_DP_CFG_STREAM_RGB_ENABLE		BIT(1)
-#define HIBMC_DP_CFG_STREAM_VIDEO_MAPPING	GENMASK(5, 2)
-#define HIBMC_DP_CFG_PIXEL_NUM_TIMING_MODE_SEL1	GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_TU_SYMBOL_SIZE	GENMASK(5, 0)
-#define HIBMC_DP_CFG_STREAM_TU_SYMBOL_FRAC_SIZE	GENMASK(9, 6)
-#define HIBMC_DP_CFG_STREAM_HTOTAL_SIZE		GENMASK(31, 16)
-#define HIBMC_DP_CFG_STREAM_HBLANK_SIZE		GENMASK(15, 0)
+#define HIBMC_DP_INTSTAT			0x1e0724
+#define HIBMC_DP_INTCLR				0x1e0728
+
+/* dp serdes reg */
+#define HIBMC_DP_HOST_OFFSET		0x10000
+#define HIBMC_DP_LANE0_RATE_OFFSET	0x4
+#define HIBMC_DP_LANE1_RATE_OFFSET	0xc
+#define HIBMC_DP_LANE_STATUS_OFFSET	0x10
+#define HIBMC_DP_PMA_LANE0_OFFSET	0x18
+#define HIBMC_DP_PMA_LANE1_OFFSET	0x1c
+#define HIBMC_DP_HOST_SERDES_CTRL	0x1f001c
+#define HIBMC_DP_PMA_TXDEEMPH		GENMASK(18, 1)
+#define DP_SERDES_DONE			0x3
+
+/* dp serdes TX-Deempth Configuration */
+#define DP_SERDES_VOL0_PRE0		0x280
+#define DP_SERDES_VOL0_PRE1		0x2300
+#define DP_SERDES_VOL0_PRE2		0x53c0
+#define DP_SERDES_VOL0_PRE3		0x8400
+#define DP_SERDES_VOL1_PRE0		0x380
+#define DP_SERDES_VOL1_PRE1		0x3440
+#define DP_SERDES_VOL1_PRE2		0x6480
+#define DP_SERDES_VOL2_PRE0		0x4c1
+#define DP_SERDES_VOL2_PRE1		0x4500
+#define DP_SERDES_VOL3_PRE0		0x600
+#define DP_SERDES_BW_8_1		0x3
+#define DP_SERDES_BW_5_4		0x2
+#define DP_SERDES_BW_2_7		0x1
+#define DP_SERDES_BW_1_62		0x0
 
 #endif
diff --git a/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c
new file mode 100644
index 000000000000..676059d4c1e6
--- /dev/null
+++ b/drivers/gpu/drm/hisilicon/hibmc/dp/dp_serdes.c
@@ -0,0 +1,71 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+// Copyright (c) 2025 Hisilicon Limited.
+
+#include <linux/delay.h>
+#include <drm/drm_device.h>
+#include <drm/drm_print.h>
+#include "dp_comm.h"
+#include "dp_config.h"
+#include "dp_reg.h"
+
+int hibmc_dp_serdes_set_tx_cfg(struct hibmc_dp_dev *dp, u8 train_set[HIBMC_DP_LANE_NUM_MAX])
+{
+	static const u32 serdes_tx_cfg[4][4] = { {DP_SERDES_VOL0_PRE0, DP_SERDES_VOL0_PRE1,
+						  DP_SERDES_VOL0_PRE2, DP_SERDES_VOL0_PRE3},
+						 {DP_SERDES_VOL1_PRE0, DP_SERDES_VOL1_PRE1,
+						  DP_SERDES_VOL1_PRE2}, {DP_SERDES_VOL2_PRE0,
+						  DP_SERDES_VOL2_PRE1}, {DP_SERDES_VOL3_PRE0}};
+	int cfg[2];
+	int i;
+
+	for (i = 0; i < HIBMC_DP_LANE_NUM_MAX; i++) {
+		cfg[i] = serdes_tx_cfg[FIELD_GET(DP_TRAIN_VOLTAGE_SWING_MASK, train_set[i])]
+				      [FIELD_GET(DP_TRAIN_PRE_EMPHASIS_MASK, train_set[i])];
+		if (!cfg[i])
+			return -EINVAL;
+
+		/* lane1 offset is 4 */
+		writel(FIELD_PREP(HIBMC_DP_PMA_TXDEEMPH, cfg[i]),
+		       dp->serdes_base + HIBMC_DP_PMA_LANE0_OFFSET + i * 4);
+	}
+
+	usleep_range(300, 500);
+
+	if (readl(dp->serdes_base + HIBMC_DP_LANE_STATUS_OFFSET) != DP_SERDES_DONE) {
+		drm_dbg_dp(dp->dev, "dp serdes cfg failed\n");
+		return -EAGAIN;
+	}
+
+	return 0;
+}
+
+int hibmc_dp_serdes_rate_switch(u8 rate, struct hibmc_dp_dev *dp)
+{
+	writel(rate, dp->serdes_base + HIBMC_DP_LANE0_RATE_OFFSET);
+	writel(rate, dp->serdes_base + HIBMC_DP_LANE1_RATE_OFFSET);
+
+	usleep_range(300, 500);
+
+	if (readl(dp->serdes_base + HIBMC_DP_LANE_STATUS_OFFSET) != DP_SERDES_DONE) {
+		drm_dbg_dp(dp->dev, "dp serdes rate switching failed\n");
+		return -EAGAIN;
+	}
+
+	if (rate < DP_SERDES_BW_8_1)
+		drm_dbg_dp(dp->dev, "reducing serdes rate to :%d\n",
+			   rate ? rate * HIBMC_DP_LINK_RATE_CAL * 10 : 162);
+
+	return 0;
+}
+
+int hibmc_dp_serdes_init(struct hibmc_dp_dev *dp)
+{
+	dp->serdes_base = dp->base + HIBMC_DP_HOST_OFFSET;
+
+	writel(FIELD_PREP(HIBMC_DP_PMA_TXDEEMPH, DP_SERDES_VOL0_PRE0),
+	       dp->serdes_base + HIBMC_DP_PMA_LANE0_OFFSET);
+	writel(FIELD_PREP(HIBMC_DP_PMA_TXDEEMPH, DP_SERDES_VOL0_PRE0),
+	       dp->serdes_base + HIBMC_DP_PMA_LANE1_OFFSET);
+
+	return hibmc_dp_serdes_rate_switch(DP_SERDES_BW_8_1, dp);
+}
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_debugfs.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_debugfs.c
new file mode 100644
index 000000000000..f585387c3a49
--- /dev/null
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_debugfs.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+// Copyright (c) 2024 Hisilicon Limited.
+
+#include <linux/debugfs.h>
+#include <linux/device.h>
+#include <linux/seq_file.h>
+#include <linux/pci.h>
+
+#include <drm/drm_drv.h>
+#include <drm/drm_file.h>
+#include <drm/drm_debugfs.h>
+#include <drm/drm_edid.h>
+
+#include "hibmc_drm_drv.h"
+
+#define MAX_BUF_SIZE 12
+
+static ssize_t hibmc_control_write(struct file *file, const char __user *user_buf,
+				   size_t count, loff_t *ppos)
+{
+	struct hibmc_drm_private *priv = file_inode(file)->i_private;
+	struct hibmc_dp_cbar_cfg *cfg = &priv->dp.cfg;
+	int ret, idx;
+	u8 buf[MAX_BUF_SIZE];
+
+	if (count >= MAX_BUF_SIZE)
+		return -EINVAL;
+
+	if (copy_from_user(buf, user_buf, count))
+		return -EFAULT;
+
+	buf[count] = '\0';
+
+	/* Only 4 parameters is allowed, the ranger are as follow:
+	 * [0] enable/disable colorbar feature
+	       0: enable colorbar, 1: disable colorbar
+	 * [1] the timing source of colorbar displaying
+	       0: timing follows XDP, 1: internal self timing
+	 * [2] the movment of colorbar displaying
+	       0: static colorbar image,
+	 *     1~255: right shifting a type of color per (1~255)frames
+	 * [3] the color type of colorbar displaying
+	       0~9: color bar, white, red, orange,
+	 *          yellow, green, cyan, bule, pupper, black
+	 */
+	if (sscanf(buf, "%hhu %hhu %hhu %u", &cfg->enable, &cfg->self_timing,
+		   &cfg->dynamic_rate, &cfg->pattern) != 4) {
+		return -EINVAL;
+	}
+
+	if (cfg->pattern > 9 || cfg->enable > 1 || cfg->self_timing > 1)
+		return -EINVAL;
+
+	ret = drm_dev_enter(&priv->dev, &idx);
+	if (!ret)
+		return -ENODEV;
+
+	hibmc_dp_set_cbar(&priv->dp, cfg);
+
+	drm_dev_exit(idx);
+
+	return count;
+}
+
+static int hibmc_dp_dbgfs_show(struct seq_file *m, void *arg)
+{
+	struct hibmc_drm_private *priv = m->private;
+	struct hibmc_dp_cbar_cfg *cfg = &priv->dp.cfg;
+	int idx;
+
+	if (!drm_dev_enter(&priv->dev, &idx))
+		return -ENODEV;
+
+	seq_printf(m, "hibmc dp colorbar cfg: %u %u %u %u\n", cfg->enable, cfg->self_timing,
+		   cfg->dynamic_rate, cfg->pattern);
+
+	drm_dev_exit(idx);
+
+	return 0;
+}
+
+static int hibmc_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, hibmc_dp_dbgfs_show, inode->i_private);
+}
+
+static const struct file_operations hibmc_dbg_fops = {
+	.owner   = THIS_MODULE,
+	.write   = hibmc_control_write,
+	.read    = seq_read,
+	.open    = hibmc_open,
+	.llseek  = seq_lseek,
+	.release = single_release,
+};
+
+void hibmc_debugfs_init(struct drm_connector *connector, struct dentry *root)
+{
+	struct drm_device *dev = connector->dev;
+	struct hibmc_drm_private *priv = to_hibmc_drm_private(dev);
+
+	/* create the file in drm directory, so we don't need to remove manually */
+	debugfs_create_file("colorbar-cfg", 0200,
+			    root, priv, &hibmc_dbg_fops);
+}
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
index 603d6b198a54..d06832e62e96 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_dp.c
@@ -13,27 +13,64 @@
 #include "hibmc_drm_drv.h"
 #include "dp/dp_hw.h"
 
+#define DP_MASKED_SINK_HPD_PLUG_INT	BIT(2)
+
 static int hibmc_dp_connector_get_modes(struct drm_connector *connector)
 {
+	const struct drm_edid *drm_edid;
 	int count;
 
-	count = drm_add_modes_noedid(connector, connector->dev->mode_config.max_width,
-				     connector->dev->mode_config.max_height);
-	drm_set_preferred_mode(connector, 1024, 768); // temporary implementation
+	drm_edid = drm_edid_read(connector);
+
+	drm_edid_connector_update(connector, drm_edid);
+
+	count = drm_edid_connector_add_modes(connector);
+
+	drm_edid_free(drm_edid);
 
 	return count;
 }
 
+static int hibmc_dp_detect(struct drm_connector *connector,
+			   struct drm_modeset_acquire_ctx *ctx, bool force)
+{
+	mdelay(200);
+
+	return drm_connector_helper_detect_from_ddc(connector, ctx, force);
+}
+
 static const struct drm_connector_helper_funcs hibmc_dp_conn_helper_funcs = {
 	.get_modes = hibmc_dp_connector_get_modes,
+	.detect_ctx = hibmc_dp_detect,
 };
 
+static int hibmc_dp_late_register(struct drm_connector *connector)
+{
+	struct hibmc_dp *dp = to_hibmc_dp(connector);
+
+	hibmc_dp_enable_int(dp);
+
+	return drm_dp_aux_register(&dp->aux);
+}
+
+static void hibmc_dp_early_unregister(struct drm_connector *connector)
+{
+	struct hibmc_dp *dp = to_hibmc_dp(connector);
+
+	drm_dp_aux_unregister(&dp->aux);
+
+	hibmc_dp_disable_int(dp);
+}
+
 static const struct drm_connector_funcs hibmc_dp_conn_funcs = {
 	.reset = drm_atomic_helper_connector_reset,
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.destroy = drm_connector_cleanup,
 	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+	.late_register = hibmc_dp_late_register,
+	.early_unregister = hibmc_dp_early_unregister,
+	.debugfs_init = hibmc_debugfs_init,
 };
 
 static inline int hibmc_dp_prepare(struct hibmc_dp *dp, struct drm_display_mode *mode)
@@ -74,6 +111,31 @@ static const struct drm_encoder_helper_funcs hibmc_dp_encoder_helper_funcs = {
 	.atomic_disable = hibmc_dp_encoder_disable,
 };
 
+irqreturn_t hibmc_dp_hpd_isr(int irq, void *arg)
+{
+	struct drm_device *dev = (struct drm_device *)arg;
+	struct hibmc_drm_private *priv = to_hibmc_drm_private(dev);
+	int idx;
+
+	if (!drm_dev_enter(dev, &idx))
+		return -ENODEV;
+
+	if (priv->dp.irq_status & DP_MASKED_SINK_HPD_PLUG_INT) {
+		drm_dbg_dp(&priv->dev, "HPD IN isr occur!\n");
+		hibmc_dp_hpd_cfg(&priv->dp);
+	} else {
+		drm_dbg_dp(&priv->dev, "HPD OUT isr occur!\n");
+		hibmc_dp_reset_link(&priv->dp);
+	}
+
+	if (dev->registered)
+		drm_connector_helper_hpd_irq_event(&priv->dp.connector);
+
+	drm_dev_exit(idx);
+
+	return IRQ_HANDLED;
+}
+
 int hibmc_dp_init(struct hibmc_drm_private *priv)
 {
 	struct drm_device *dev = &priv->dev;
@@ -103,8 +165,8 @@ int hibmc_dp_init(struct hibmc_drm_private *priv)
 
 	drm_encoder_helper_add(encoder, &hibmc_dp_encoder_helper_funcs);
 
-	ret = drm_connector_init(dev, connector, &hibmc_dp_conn_funcs,
-				 DRM_MODE_CONNECTOR_DisplayPort);
+	ret = drm_connector_init_with_ddc(dev, connector, &hibmc_dp_conn_funcs,
+					  DRM_MODE_CONNECTOR_DisplayPort, &dp->aux.ddc);
 	if (ret) {
 		drm_err(dev, "init dp connector failed: %d\n", ret);
 		return ret;
@@ -114,5 +176,7 @@ int hibmc_dp_init(struct hibmc_drm_private *priv)
 
 	drm_connector_attach_encoder(connector, encoder);
 
+	connector->polled = DRM_CONNECTOR_POLL_HPD;
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
index e6de6d5edf6b..768b97f9e74a 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
@@ -28,12 +28,12 @@
 #include "hibmc_drm_drv.h"
 #include "hibmc_drm_regs.h"
 
-#define HIBMC_DP_HOST_SERDES_CTRL		0x1f001c
-#define HIBMC_DP_HOST_SERDES_CTRL_VAL		0x8a00
-#define HIBMC_DP_HOST_SERDES_CTRL_MASK		0x7ffff
+#include "dp/dp_reg.h"
 
 DEFINE_DRM_GEM_FOPS(hibmc_fops);
 
+static const char *g_irqs_names_map[HIBMC_MAX_VECTORS] = { "vblank", "hpd" };
+
 static irqreturn_t hibmc_interrupt(int irq, void *arg)
 {
 	struct drm_device *dev = (struct drm_device *)arg;
@@ -51,6 +51,22 @@ static irqreturn_t hibmc_interrupt(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
+static irqreturn_t hibmc_dp_interrupt(int irq, void *arg)
+{
+	struct drm_device *dev = (struct drm_device *)arg;
+	struct hibmc_drm_private *priv = to_hibmc_drm_private(dev);
+	u32 status;
+
+	status = readl(priv->mmio + HIBMC_DP_INTSTAT);
+	if (status) {
+		priv->dp.irq_status = status;
+		writel(status, priv->mmio + HIBMC_DP_INTCLR);
+		return IRQ_WAKE_THREAD;
+	}
+
+	return IRQ_HANDLED;
+}
+
 static int hibmc_dumb_create(struct drm_file *file, struct drm_device *dev,
 			     struct drm_mode_create_dumb *args)
 {
@@ -121,9 +137,12 @@ static int hibmc_kms_init(struct hibmc_drm_private *priv)
 		return ret;
 	}
 
-	/* if DP existed, init DP */
-	if ((readl(priv->mmio + HIBMC_DP_HOST_SERDES_CTRL) &
-	     HIBMC_DP_HOST_SERDES_CTRL_MASK) == HIBMC_DP_HOST_SERDES_CTRL_VAL) {
+	/*
+	 * If the serdes reg is readable and is not equal to 0,
+	 * DP block exists and initializes it.
+	 */
+	ret = readl(priv->mmio + HIBMC_DP_HOST_SERDES_CTRL);
+	if (ret) {
 		ret = hibmc_dp_init(priv);
 		if (ret)
 			drm_err(dev, "failed to init dp: %d\n", ret);
@@ -250,15 +269,48 @@ static int hibmc_hw_init(struct hibmc_drm_private *priv)
 	return 0;
 }
 
-static int hibmc_unload(struct drm_device *dev)
+static void hibmc_unload(struct drm_device *dev)
 {
-	struct pci_dev *pdev = to_pci_dev(dev->dev);
-
 	drm_atomic_helper_shutdown(dev);
+}
 
-	free_irq(pdev->irq, dev);
+static int hibmc_msi_init(struct drm_device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev->dev);
+	char name[32] = {0};
+	int valid_irq_num;
+	int irq;
+	int ret;
 
-	pci_disable_msi(to_pci_dev(dev->dev));
+	ret = pci_alloc_irq_vectors(pdev, HIBMC_MIN_VECTORS,
+				    HIBMC_MAX_VECTORS, PCI_IRQ_MSI);
+	if (ret < 0) {
+		drm_err(dev, "enabling MSI failed: %d\n", ret);
+		return ret;
+	}
+
+	valid_irq_num = ret;
+
+	for (int i = 0; i < valid_irq_num; i++) {
+		snprintf(name, ARRAY_SIZE(name) - 1, "%s-%s-%s",
+			 dev->driver->name, pci_name(pdev), g_irqs_names_map[i]);
+
+		irq = pci_irq_vector(pdev, i);
+
+		if (i)
+			/* PCI devices require shared interrupts. */
+			ret = devm_request_threaded_irq(&pdev->dev, irq,
+							hibmc_dp_interrupt,
+							hibmc_dp_hpd_isr,
+							IRQF_SHARED, name, dev);
+		else
+			ret = devm_request_irq(&pdev->dev, irq, hibmc_interrupt,
+					       IRQF_SHARED, name, dev);
+		if (ret) {
+			drm_err(dev, "install irq failed: %d\n", ret);
+			return ret;
+		}
+	}
 
 	return 0;
 }
@@ -290,15 +342,10 @@ static int hibmc_load(struct drm_device *dev)
 		goto err;
 	}
 
-	ret = pci_enable_msi(pdev);
+	ret = hibmc_msi_init(dev);
 	if (ret) {
-		drm_warn(dev, "enabling MSI failed: %d\n", ret);
-	} else {
-		/* PCI devices require shared interrupts. */
-		ret = request_irq(pdev->irq, hibmc_interrupt, IRQF_SHARED,
-				  dev->driver->name, dev);
-		if (ret)
-			drm_warn(dev, "install irq failed: %d\n", ret);
+		drm_err(dev, "hibmc msi init failed, ret:%d\n", ret);
+		goto err;
 	}
 
 	/* reset all the states of crtc/plane/encoder/connector */
@@ -374,7 +421,7 @@ static void hibmc_pci_remove(struct pci_dev *pdev)
 
 static void hibmc_pci_shutdown(struct pci_dev *pdev)
 {
-	drm_atomic_helper_shutdown(pci_get_drvdata(pdev));
+	hibmc_pci_remove(pdev);
 }
 
 static const struct pci_device_id hibmc_pci_table[] = {
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
index d982f1e4b958..274feabe7df0 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
@@ -22,6 +22,9 @@
 
 #include "dp/dp_hw.h"
 
+#define HIBMC_MIN_VECTORS	1
+#define HIBMC_MAX_VECTORS	2
+
 struct hibmc_vdac {
 	struct drm_device *dev;
 	struct drm_encoder encoder;
@@ -47,6 +50,11 @@ static inline struct hibmc_vdac *to_hibmc_vdac(struct drm_connector *connector)
 	return container_of(connector, struct hibmc_vdac, connector);
 }
 
+static inline struct hibmc_dp *to_hibmc_dp(struct drm_connector *connector)
+{
+	return container_of(connector, struct hibmc_dp, connector);
+}
+
 static inline struct hibmc_drm_private *to_hibmc_drm_private(struct drm_device *dev)
 {
 	return container_of(dev, struct hibmc_drm_private, dev);
@@ -64,4 +72,8 @@ int hibmc_ddc_create(struct drm_device *drm_dev, struct hibmc_vdac *connector);
 
 int hibmc_dp_init(struct hibmc_drm_private *priv);
 
+void hibmc_debugfs_init(struct drm_connector *connector, struct dentry *root);
+
+irqreturn_t hibmc_dp_hpd_isr(int irq, void *arg);
+
 #endif
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
index 05e19ea4c9f9..e8a527ede854 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c
@@ -60,6 +60,7 @@ static void hibmc_connector_destroy(struct drm_connector *connector)
 static const struct drm_connector_helper_funcs
 	hibmc_connector_helper_funcs = {
 	.get_modes = hibmc_connector_get_modes,
+	.detect_ctx = drm_connector_helper_detect_from_ddc,
 };
 
 static const struct drm_connector_funcs hibmc_connector_funcs = {
@@ -127,5 +128,7 @@ int hibmc_vdac_init(struct hibmc_drm_private *priv)
 
 	drm_connector_attach_encoder(connector, encoder);
 
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
index 2eea9fb0e76b..e80debdc4176 100644
--- a/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
+++ b/drivers/gpu/drm/hisilicon/kirin/dw_drm_dsi.c
@@ -825,7 +825,6 @@ static const struct component_ops dsi_ops = {
 static int dsi_parse_dt(struct platform_device *pdev, struct dw_dsi *dsi)
 {
 	struct dsi_hw_ctx *ctx = dsi->ctx;
-	struct resource *res;
 
 	ctx->pclk = devm_clk_get(&pdev->dev, "pclk");
 	if (IS_ERR(ctx->pclk)) {
@@ -833,8 +832,7 @@ static int dsi_parse_dt(struct platform_device *pdev, struct dw_dsi *dsi)
 		return PTR_ERR(ctx->pclk);
 	}
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	ctx->base = devm_ioremap_resource(&pdev->dev, res);
+	ctx->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(ctx->base)) {
 		DRM_ERROR("failed to remap dsi io region\n");
 		return PTR_ERR(ctx->base);
diff --git a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
index 2eb49177ac42..45c4eb008ad5 100644
--- a/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
+++ b/drivers/gpu/drm/hisilicon/kirin/kirin_drm_ade.c
@@ -844,7 +844,6 @@ static struct drm_plane_funcs ade_plane_funcs = {
 static void *ade_hw_ctx_alloc(struct platform_device *pdev,
 			      struct drm_crtc *crtc)
 {
-	struct resource *res;
 	struct device *dev = &pdev->dev;
 	struct device_node *np = pdev->dev.of_node;
 	struct ade_hw_ctx *ctx = NULL;
@@ -856,8 +855,7 @@ static void *ade_hw_ctx_alloc(struct platform_device *pdev,
 		return ERR_PTR(-ENOMEM);
 	}
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	ctx->base = devm_ioremap_resource(dev, res);
+	ctx->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(ctx->base)) {
 		DRM_ERROR("failed to remap ade io base\n");
 		return ERR_PTR(-EIO);
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index ed05b131ed3a..13d4a16f7d33 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -52,7 +52,6 @@ i915-y += \
 i915-y += \
 	soc/intel_dram.o \
 	soc/intel_gmch.o \
-	soc/intel_pch.o \
 	soc/intel_rom.o
 
 # core library code
@@ -247,6 +246,7 @@ i915-y += \
 	display/intel_display_power_map.o \
 	display/intel_display_power_well.o \
 	display/intel_display_reset.o \
+	display/intel_display_rpm.o \
 	display/intel_display_rps.o \
 	display/intel_display_snapshot.o \
 	display/intel_display_wa.o \
@@ -281,6 +281,7 @@ i915-y += \
 	display/intel_modeset_setup.o \
 	display/intel_modeset_verify.o \
 	display/intel_overlay.o \
+	display/intel_pch.o \
 	display/intel_pch_display.o \
 	display/intel_pch_refclk.o \
 	display/intel_plane_initial.o \
diff --git a/drivers/gpu/drm/i915/display/dvo_ch7017.c b/drivers/gpu/drm/i915/display/dvo_ch7017.c
index 206818f9ad49..f10c0fb8d2c8 100644
--- a/drivers/gpu/drm/i915/display/dvo_ch7017.c
+++ b/drivers/gpu/drm/i915/display/dvo_ch7017.c
@@ -25,6 +25,8 @@
  *
  */
 
+#include <drm/drm_print.h>
+
 #include "intel_display_types.h"
 #include "intel_dvo_dev.h"
 
diff --git a/drivers/gpu/drm/i915/display/dvo_ch7xxx.c b/drivers/gpu/drm/i915/display/dvo_ch7xxx.c
index 10ab3cc73e58..49f02aca818b 100644
--- a/drivers/gpu/drm/i915/display/dvo_ch7xxx.c
+++ b/drivers/gpu/drm/i915/display/dvo_ch7xxx.c
@@ -26,6 +26,8 @@ SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 **************************************************************************/
 
+#include <drm/drm_print.h>
+
 #include "intel_display_types.h"
 #include "intel_dvo_dev.h"
 
diff --git a/drivers/gpu/drm/i915/display/dvo_ivch.c b/drivers/gpu/drm/i915/display/dvo_ivch.c
index d9c3152d4338..0713b2709412 100644
--- a/drivers/gpu/drm/i915/display/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/display/dvo_ivch.c
@@ -29,6 +29,8 @@
  *
  */
 
+#include <drm/drm_print.h>
+
 #include "intel_display_types.h"
 #include "intel_dvo_dev.h"
 
diff --git a/drivers/gpu/drm/i915/display/dvo_ns2501.c b/drivers/gpu/drm/i915/display/dvo_ns2501.c
index 92d32d6b5bce..80b71bd6a837 100644
--- a/drivers/gpu/drm/i915/display/dvo_ns2501.c
+++ b/drivers/gpu/drm/i915/display/dvo_ns2501.c
@@ -26,6 +26,8 @@
  *
  */
 
+#include <drm/drm_print.h>
+
 #include "intel_display_types.h"
 #include "intel_dvo_dev.h"
 
diff --git a/drivers/gpu/drm/i915/display/dvo_sil164.c b/drivers/gpu/drm/i915/display/dvo_sil164.c
index b42c717085f3..017b617a8069 100644
--- a/drivers/gpu/drm/i915/display/dvo_sil164.c
+++ b/drivers/gpu/drm/i915/display/dvo_sil164.c
@@ -26,6 +26,8 @@ SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 **************************************************************************/
 
+#include <drm/drm_print.h>
+
 #include "intel_display_types.h"
 #include "intel_dvo_dev.h"
 
diff --git a/drivers/gpu/drm/i915/display/dvo_tfp410.c b/drivers/gpu/drm/i915/display/dvo_tfp410.c
index 280699438526..ed560e3438db 100644
--- a/drivers/gpu/drm/i915/display/dvo_tfp410.c
+++ b/drivers/gpu/drm/i915/display/dvo_tfp410.c
@@ -25,6 +25,8 @@
  *
  */
 
+#include <drm/drm_print.h>
+
 #include "intel_display_types.h"
 #include "intel_dvo_dev.h"
 
diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c
index 55b9e9bfcc4d..e0a98e6fd6d1 100644
--- a/drivers/gpu/drm/i915/display/g4x_dp.c
+++ b/drivers/gpu/drm/i915/display/g4x_dp.c
@@ -7,9 +7,11 @@
 
 #include <linux/string_helpers.h>
 
+#include <drm/drm_print.h>
+
 #include "g4x_dp.h"
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_audio.h"
 #include "intel_backlight.h"
 #include "intel_connector.h"
@@ -28,7 +30,6 @@
 #include "intel_hotplug.h"
 #include "intel_pch_display.h"
 #include "intel_pps.h"
-#include "vlv_sideband.h"
 
 static const struct dpll g4x_dpll[] = {
 	{ .dot = 162000, .p1 = 2, .p2 = 10, .n = 2, .m1 = 23, .m2 = 8, },
@@ -60,14 +61,13 @@ static void g4x_dp_set_clock(struct intel_encoder *encoder,
 			     struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	const struct dpll *divisor = NULL;
 	int i, count = 0;
 
 	if (display->platform.g4x) {
 		divisor = g4x_dpll;
 		count = ARRAY_SIZE(g4x_dpll);
-	} else if (HAS_PCH_SPLIT(dev_priv)) {
+	} else if (HAS_PCH_SPLIT(display)) {
 		divisor = pch_dpll;
 		count = ARRAY_SIZE(pch_dpll);
 	} else if (display->platform.cherryview) {
@@ -93,7 +93,6 @@ static void intel_dp_prepare(struct intel_encoder *encoder,
 			     const struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	enum port port = encoder->port;
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
@@ -141,7 +140,7 @@ static void intel_dp_prepare(struct intel_encoder *encoder,
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
 		intel_dp->DP |= DP_PIPE_SEL_IVB(crtc->pipe);
-	} else if (HAS_PCH_CPT(dev_priv) && port != PORT_A) {
+	} else if (HAS_PCH_CPT(display) && port != PORT_A) {
 		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 
 		intel_de_rmw(display, TRANS_DP_CTL(crtc->pipe),
@@ -183,7 +182,7 @@ static void assert_dp_port(struct intel_dp *intel_dp, bool state)
 
 static void assert_edp_pll(struct intel_display *display, bool state)
 {
-	bool cur_state = intel_de_read(display, DP_A) & DP_PLL_ENABLE;
+	bool cur_state = intel_de_read(display, DP_A) & EDP_PLL_ENABLE;
 
 	INTEL_DISPLAY_STATE_WARN(display, cur_state != state,
 				 "eDP PLL state assertion failure (expected %s, current %s)\n",
@@ -205,12 +204,12 @@ static void ilk_edp_pll_on(struct intel_dp *intel_dp,
 	drm_dbg_kms(display->drm, "enabling eDP PLL for clock %d\n",
 		    pipe_config->port_clock);
 
-	intel_dp->DP &= ~DP_PLL_FREQ_MASK;
+	intel_dp->DP &= ~EDP_PLL_FREQ_MASK;
 
 	if (pipe_config->port_clock == 162000)
-		intel_dp->DP |= DP_PLL_FREQ_162MHZ;
+		intel_dp->DP |= EDP_PLL_FREQ_162MHZ;
 	else
-		intel_dp->DP |= DP_PLL_FREQ_270MHZ;
+		intel_dp->DP |= EDP_PLL_FREQ_270MHZ;
 
 	intel_de_write(display, DP_A, intel_dp->DP);
 	intel_de_posting_read(display, DP_A);
@@ -225,7 +224,7 @@ static void ilk_edp_pll_on(struct intel_dp *intel_dp,
 	if (display->platform.ironlake)
 		intel_wait_for_vblank_if_active(display, !crtc->pipe);
 
-	intel_dp->DP |= DP_PLL_ENABLE;
+	intel_dp->DP |= EDP_PLL_ENABLE;
 
 	intel_de_write(display, DP_A, intel_dp->DP);
 	intel_de_posting_read(display, DP_A);
@@ -243,7 +242,7 @@ static void ilk_edp_pll_off(struct intel_dp *intel_dp,
 
 	drm_dbg_kms(display->drm, "disabling eDP PLL\n");
 
-	intel_dp->DP &= ~DP_PLL_ENABLE;
+	intel_dp->DP &= ~EDP_PLL_ENABLE;
 
 	intel_de_write(display, DP_A, intel_dp->DP);
 	intel_de_posting_read(display, DP_A);
@@ -277,7 +276,6 @@ bool g4x_dp_port_enabled(struct intel_display *display,
 			 i915_reg_t dp_reg, enum port port,
 			 enum pipe *pipe)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	bool ret;
 	u32 val;
 
@@ -287,13 +285,13 @@ bool g4x_dp_port_enabled(struct intel_display *display,
 
 	/* asserts want to know the pipe even if the port is disabled */
 	if (display->platform.ivybridge && port == PORT_A)
-		*pipe = (val & DP_PIPE_SEL_MASK_IVB) >> DP_PIPE_SEL_SHIFT_IVB;
-	else if (HAS_PCH_CPT(dev_priv) && port != PORT_A)
+		*pipe = REG_FIELD_GET(DP_PIPE_SEL_MASK_IVB, val);
+	else if (HAS_PCH_CPT(display) && port != PORT_A)
 		ret &= cpt_dp_port_selected(display, port, pipe);
 	else if (display->platform.cherryview)
-		*pipe = (val & DP_PIPE_SEL_MASK_CHV) >> DP_PIPE_SEL_SHIFT_CHV;
+		*pipe = REG_FIELD_GET(DP_PIPE_SEL_MASK_CHV, val);
 	else
-		*pipe = (val & DP_PIPE_SEL_MASK) >> DP_PIPE_SEL_SHIFT;
+		*pipe = REG_FIELD_GET(DP_PIPE_SEL_MASK, val);
 
 	return ret;
 }
@@ -338,7 +336,6 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 				struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	u32 tmp, flags = 0;
 	enum port port = encoder->port;
@@ -353,7 +350,7 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 
 	pipe_config->has_audio = tmp & DP_AUDIO_OUTPUT_ENABLE && port != PORT_A;
 
-	if (HAS_PCH_CPT(dev_priv) && port != PORT_A) {
+	if (HAS_PCH_CPT(display) && port != PORT_A) {
 		u32 trans_dp = intel_de_read(display,
 					     TRANS_DP_CTL(crtc->pipe));
 
@@ -389,13 +386,12 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	if (display->platform.g4x && tmp & DP_COLOR_RANGE_16_235)
 		pipe_config->limited_color_range = true;
 
-	pipe_config->lane_count =
-		((tmp & DP_PORT_WIDTH_MASK) >> DP_PORT_WIDTH_SHIFT) + 1;
+	pipe_config->lane_count = REG_FIELD_GET(DP_PORT_WIDTH_MASK, tmp) + 1;
 
 	g4x_dp_get_m_n(pipe_config);
 
 	if (port == PORT_A) {
-		if ((intel_de_read(display, DP_A) & DP_PLL_FREQ_MASK) == DP_PLL_FREQ_162MHZ)
+		if ((intel_de_read(display, DP_A) & EDP_PLL_FREQ_MASK) == EDP_PLL_FREQ_162MHZ)
 			pipe_config->port_clock = 162000;
 		else
 			pipe_config->port_clock = 270000;
@@ -416,7 +412,6 @@ intel_dp_link_down(struct intel_encoder *encoder,
 		   const struct intel_crtc_state *old_crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->uapi.crtc);
 	enum port port = encoder->port;
@@ -429,7 +424,7 @@ intel_dp_link_down(struct intel_encoder *encoder,
 	drm_dbg_kms(display->drm, "\n");
 
 	if ((display->platform.ivybridge && port == PORT_A) ||
-	    (HAS_PCH_CPT(dev_priv) && port != PORT_A)) {
+	    (HAS_PCH_CPT(display) && port != PORT_A)) {
 		intel_dp->DP &= ~DP_LINK_TRAIN_MASK_CPT;
 		intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE_CPT;
 	} else {
@@ -448,7 +443,7 @@ intel_dp_link_down(struct intel_encoder *encoder,
 	 * to transcoder A after disabling it to allow the
 	 * matching HDMI port to be enabled on transcoder A.
 	 */
-	if (HAS_PCH_IBX(dev_priv) && crtc->pipe == PIPE_B && port != PORT_A) {
+	if (HAS_PCH_IBX(display) && crtc->pipe == PIPE_B && port != PORT_A) {
 		/*
 		 * We get CPU/PCH FIFO underruns on the other pipe when
 		 * doing the workaround. Sweep them under the rug.
@@ -519,7 +514,7 @@ static void intel_disable_dp(struct intel_atomic_state *state,
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
-	intel_dp->link_trained = false;
+	intel_dp->link.active = false;
 
 	/*
 	 * Make sure the panel is off before trying to change the mode.
@@ -581,16 +576,10 @@ static void chv_post_disable_dp(struct intel_atomic_state *state,
 				const struct intel_crtc_state *old_crtc_state,
 				const struct drm_connector_state *old_conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-
 	intel_dp_link_down(encoder, old_crtc_state);
 
-	vlv_dpio_get(dev_priv);
-
 	/* Assert data lane reset */
 	chv_data_lane_soft_reset(encoder, old_crtc_state, true);
-
-	vlv_dpio_put(dev_priv);
 }
 
 static void
@@ -1223,10 +1212,10 @@ static int g4x_dp_compute_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *crtc_state,
 				 struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	int ret;
 
-	if (HAS_PCH_SPLIT(i915) && encoder->port != PORT_A)
+	if (HAS_PCH_SPLIT(display) && encoder->port != PORT_A)
 		crtc_state->has_pch_encoder = true;
 
 	ret = intel_dp_compute_config(encoder, crtc_state, conn_state);
@@ -1279,7 +1268,6 @@ static const struct drm_encoder_funcs intel_dp_enc_funcs = {
 bool g4x_dp_init(struct intel_display *display,
 		 i915_reg_t output_reg, enum port port)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	const struct intel_bios_encoder_data *devdata;
 	struct intel_digital_port *dig_port;
 	struct intel_encoder *intel_encoder;
@@ -1353,7 +1341,7 @@ bool g4x_dp_init(struct intel_display *display,
 	intel_encoder->audio_disable = g4x_dp_audio_disable;
 
 	if ((display->platform.ivybridge && port == PORT_A) ||
-	    (HAS_PCH_CPT(dev_priv) && port != PORT_A))
+	    (HAS_PCH_CPT(display) && port != PORT_A))
 		dig_port->dp.set_link_train = cpt_set_link_train;
 	else
 		dig_port->dp.set_link_train = g4x_set_link_train;
@@ -1370,7 +1358,7 @@ bool g4x_dp_init(struct intel_display *display,
 		intel_encoder->set_signal_levels = g4x_set_signal_levels;
 
 	if (display->platform.valleyview || display->platform.cherryview ||
-	    (HAS_PCH_SPLIT(dev_priv) && port != PORT_A)) {
+	    (HAS_PCH_SPLIT(display) && port != PORT_A)) {
 		dig_port->dp.preemph_max = intel_dp_preemph_max_3;
 		dig_port->dp.voltage_max = intel_dp_voltage_max_3;
 	} else {
diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c
index 3dc2c59a3df0..1d252432d729 100644
--- a/drivers/gpu/drm/i915/display/g4x_hdmi.c
+++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c
@@ -5,8 +5,9 @@
  * HDMI support for G4x,ILK,SNB,IVB,VLV,CHV (HSW+ handled by the DDI code).
  */
 
+#include <drm/drm_print.h>
+
 #include "g4x_hdmi.h"
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
@@ -22,13 +23,11 @@
 #include "intel_hdmi.h"
 #include "intel_hotplug.h"
 #include "intel_sdvo.h"
-#include "vlv_sideband.h"
 
 static void intel_hdmi_prepare(struct intel_encoder *encoder,
 			       const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
@@ -37,7 +36,7 @@ static void intel_hdmi_prepare(struct intel_encoder *encoder,
 	intel_dp_dual_mode_set_tmds_output(intel_hdmi, true);
 
 	hdmi_val = SDVO_ENCODING_HDMI;
-	if (!HAS_PCH_SPLIT(dev_priv) && crtc_state->limited_color_range)
+	if (!HAS_PCH_SPLIT(display) && crtc_state->limited_color_range)
 		hdmi_val |= HDMI_COLOR_RANGE_16_235;
 	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
 		hdmi_val |= SDVO_VSYNC_ACTIVE_HIGH;
@@ -52,7 +51,7 @@ static void intel_hdmi_prepare(struct intel_encoder *encoder,
 	if (crtc_state->has_hdmi_sink)
 		hdmi_val |= HDMI_MODE_SELECT_HDMI;
 
-	if (HAS_PCH_CPT(dev_priv))
+	if (HAS_PCH_CPT(display))
 		hdmi_val |= SDVO_PIPE_SEL_CPT(crtc->pipe);
 	else if (display->platform.cherryview)
 		hdmi_val |= SDVO_PIPE_SEL_CHV(crtc->pipe);
@@ -134,9 +133,8 @@ static int g4x_hdmi_compute_config(struct intel_encoder *encoder,
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 
-	if (HAS_PCH_SPLIT(i915)) {
+	if (HAS_PCH_SPLIT(display)) {
 		crtc_state->has_pch_encoder = true;
 		if (!intel_fdi_compute_pipe_bpp(crtc_state))
 			return -EINVAL;
@@ -155,7 +153,6 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 				  struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	u32 tmp, flags = 0;
 	int dotclock;
@@ -186,7 +183,7 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	if (tmp & HDMI_AUDIO_ENABLE)
 		pipe_config->has_audio = true;
 
-	if (!HAS_PCH_SPLIT(dev_priv) &&
+	if (!HAS_PCH_SPLIT(display) &&
 	    tmp & HDMI_COLOR_RANGE_16_235)
 		pipe_config->limited_color_range = true;
 
@@ -383,7 +380,6 @@ static void intel_disable_hdmi(struct intel_atomic_state *state,
 			       const struct drm_connector_state *old_conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	struct intel_digital_port *dig_port =
 		hdmi_to_dig_port(intel_hdmi);
@@ -401,7 +397,7 @@ static void intel_disable_hdmi(struct intel_atomic_state *state,
 	 * to transcoder A after disabling it to allow the
 	 * matching DP port to be enabled on transcoder A.
 	 */
-	if (HAS_PCH_IBX(dev_priv) && crtc->pipe == PIPE_B) {
+	if (HAS_PCH_IBX(display) && crtc->pipe == PIPE_B) {
 		/*
 		 * We get CPU/PCH FIFO underruns on the other pipe when
 		 * doing the workaround. Sweep them under the rug.
@@ -539,15 +535,8 @@ static void chv_hdmi_post_disable(struct intel_atomic_state *state,
 				  const struct intel_crtc_state *old_crtc_state,
 				  const struct drm_connector_state *old_conn_state)
 {
-	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	vlv_dpio_get(dev_priv);
-
 	/* Assert data lane reset */
 	chv_data_lane_soft_reset(encoder, old_crtc_state, true);
-
-	vlv_dpio_put(dev_priv);
 }
 
 static void chv_hdmi_pre_enable(struct intel_atomic_state *state,
@@ -682,7 +671,6 @@ static bool assert_hdmi_port_valid(struct intel_display *display, enum port port
 bool g4x_hdmi_init(struct intel_display *display,
 		   i915_reg_t hdmi_reg, enum port port)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	const struct intel_bios_encoder_data *devdata;
 	struct intel_digital_port *dig_port;
 	struct intel_encoder *intel_encoder;
@@ -724,7 +712,7 @@ bool g4x_hdmi_init(struct intel_display *display,
 
 	intel_encoder->hotplug = intel_hdmi_hotplug;
 	intel_encoder->compute_config = g4x_hdmi_compute_config;
-	if (HAS_PCH_SPLIT(dev_priv)) {
+	if (HAS_PCH_SPLIT(display)) {
 		intel_encoder->disable = pch_disable_hdmi;
 		intel_encoder->post_disable = pch_post_disable_hdmi;
 	} else {
@@ -745,9 +733,9 @@ bool g4x_hdmi_init(struct intel_display *display,
 		intel_encoder->post_disable = vlv_hdmi_post_disable;
 	} else {
 		intel_encoder->pre_enable = intel_hdmi_pre_enable;
-		if (HAS_PCH_CPT(dev_priv))
+		if (HAS_PCH_CPT(display))
 			intel_encoder->enable = cpt_enable_hdmi;
-		else if (HAS_PCH_IBX(dev_priv))
+		else if (HAS_PCH_IBX(display))
 			intel_encoder->enable = ibx_enable_hdmi;
 		else
 			intel_encoder->enable = g4x_enable_hdmi;
diff --git a/drivers/gpu/drm/i915/display/hsw_ips.c b/drivers/gpu/drm/i915/display/hsw_ips.c
index 674a0e5f0858..4307e2ed03d9 100644
--- a/drivers/gpu/drm/i915/display/hsw_ips.c
+++ b/drivers/gpu/drm/i915/display/hsw_ips.c
@@ -10,6 +10,7 @@
 #include "i915_reg.h"
 #include "intel_color_regs.h"
 #include "intel_de.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_pcode.h"
 
@@ -344,10 +345,9 @@ static int hsw_ips_debugfs_status_show(struct seq_file *m, void *unused)
 {
 	struct intel_crtc *crtc = m->private;
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
 	seq_printf(m, "Enabled by kernel parameter: %s\n",
 		   str_yes_no(display->params.enable_ips));
@@ -361,7 +361,7 @@ static int hsw_ips_debugfs_status_show(struct seq_file *m, void *unused)
 			seq_puts(m, "Currently: disabled\n");
 	}
 
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index 013295f66d56..a2a6d52be0a5 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -7,9 +7,10 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_blend.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "i9xx_plane.h"
 #include "i9xx_plane_regs.h"
 #include "intel_atomic.h"
@@ -630,85 +631,85 @@ vlv_primary_async_flip(struct intel_dsb *dsb,
 static void
 bdw_primary_enable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 
-	spin_lock_irq(&i915->irq_lock);
-	bdw_enable_pipe_irq(i915, pipe, GEN8_PIPE_PRIMARY_FLIP_DONE);
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	bdw_enable_pipe_irq(display, pipe, GEN8_PIPE_PRIMARY_FLIP_DONE);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 bdw_primary_disable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 
-	spin_lock_irq(&i915->irq_lock);
-	bdw_disable_pipe_irq(i915, pipe, GEN8_PIPE_PRIMARY_FLIP_DONE);
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	bdw_disable_pipe_irq(display, pipe, GEN8_PIPE_PRIMARY_FLIP_DONE);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 ivb_primary_enable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 
-	spin_lock_irq(&i915->irq_lock);
-	ilk_enable_display_irq(i915, DE_PLANE_FLIP_DONE_IVB(plane->i9xx_plane));
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	ilk_enable_display_irq(display, DE_PLANE_FLIP_DONE_IVB(plane->i9xx_plane));
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 ivb_primary_disable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 
-	spin_lock_irq(&i915->irq_lock);
-	ilk_disable_display_irq(i915, DE_PLANE_FLIP_DONE_IVB(plane->i9xx_plane));
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	ilk_disable_display_irq(display, DE_PLANE_FLIP_DONE_IVB(plane->i9xx_plane));
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 ilk_primary_enable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 
-	spin_lock_irq(&i915->irq_lock);
-	ilk_enable_display_irq(i915, DE_PLANE_FLIP_DONE(plane->i9xx_plane));
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	ilk_enable_display_irq(display, DE_PLANE_FLIP_DONE(plane->i9xx_plane));
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 ilk_primary_disable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 
-	spin_lock_irq(&i915->irq_lock);
-	ilk_disable_display_irq(i915, DE_PLANE_FLIP_DONE(plane->i9xx_plane));
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	ilk_disable_display_irq(display, DE_PLANE_FLIP_DONE(plane->i9xx_plane));
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 vlv_primary_enable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 
-	spin_lock_irq(&i915->irq_lock);
-	i915_enable_pipestat(i915, pipe, PLANE_FLIP_DONE_INT_STATUS_VLV);
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	i915_enable_pipestat(display, pipe, PLANE_FLIP_DONE_INT_STATUS_VLV);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 vlv_primary_disable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 
-	spin_lock_irq(&i915->irq_lock);
-	i915_disable_pipestat(i915, pipe, PLANE_FLIP_DONE_INT_STATUS_VLV);
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	i915_disable_pipestat(display, pipe, PLANE_FLIP_DONE_INT_STATUS_VLV);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static bool i9xx_plane_can_async_flip(u64 modifier)
@@ -820,7 +821,7 @@ unsigned int vlv_plane_min_alignment(struct intel_plane *plane,
 {
 	struct intel_display *display = to_intel_display(plane);
 
-	if (intel_plane_can_async_flip(plane, fb->modifier))
+	if (intel_plane_can_async_flip(plane, fb->format->format, fb->modifier))
 		return 256 * 1024;
 
 	/* FIXME undocumented so not sure what's actually needed */
@@ -844,7 +845,7 @@ static unsigned int g4x_primary_min_alignment(struct intel_plane *plane,
 {
 	struct intel_display *display = to_intel_display(plane);
 
-	if (intel_plane_can_async_flip(plane, fb->modifier))
+	if (intel_plane_can_async_flip(plane, fb->format->format, fb->modifier))
 		return 256 * 1024;
 
 	if (intel_scanout_needs_vtd_wa(display))
@@ -889,6 +890,7 @@ static const struct drm_plane_funcs i965_plane_funcs = {
 	.atomic_duplicate_state = intel_plane_duplicate_state,
 	.atomic_destroy_state = intel_plane_destroy_state,
 	.format_mod_supported = i965_plane_format_mod_supported,
+	.format_mod_supported_async = intel_plane_format_mod_supported_async,
 };
 
 static const struct drm_plane_funcs i8xx_plane_funcs = {
@@ -898,6 +900,7 @@ static const struct drm_plane_funcs i8xx_plane_funcs = {
 	.atomic_duplicate_state = intel_plane_duplicate_state,
 	.atomic_destroy_state = intel_plane_destroy_state,
 	.format_mod_supported = i8xx_plane_format_mod_supported,
+	.format_mod_supported_async = intel_plane_format_mod_supported_async,
 };
 
 struct intel_plane *
diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c
index 7c80e37c1c5f..77876ef735b7 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.c
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.c
@@ -9,6 +9,7 @@
 #include "i9xx_wm_regs.h"
 #include "intel_atomic.h"
 #include "intel_bo.h"
+#include "intel_de.h"
 #include "intel_display.h"
 #include "intel_display_trace.h"
 #include "intel_fb.h"
@@ -81,13 +82,14 @@ static const struct cxsr_latency cxsr_latency_table[] = {
 	{0, 1, 400, 800, 6042, 36042, 6584, 36584},    /* DDR3-800 SC */
 };
 
-static const struct cxsr_latency *pnv_get_cxsr_latency(struct drm_i915_private *i915)
+static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *display)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(cxsr_latency_table); i++) {
 		const struct cxsr_latency *latency = &cxsr_latency_table[i];
-		bool is_desktop = !IS_MOBILE(i915);
+		bool is_desktop = !display->platform.mobile;
 
 		if (is_desktop == latency->is_desktop &&
 		    i915->is_ddr3 == latency->is_ddr3 &&
@@ -96,15 +98,16 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct drm_i915_private *
 			return latency;
 	}
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "Could not find CxSR latency for DDR%s, FSB %u kHz, MEM %u kHz\n",
 		    i915->is_ddr3 ? "3" : "2", i915->fsb_freq, i915->mem_freq);
 
 	return NULL;
 }
 
-static void chv_set_memory_dvfs(struct drm_i915_private *dev_priv, bool enable)
+static void chv_set_memory_dvfs(struct intel_display *display, bool enable)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 val;
 
 	vlv_punit_get(dev_priv);
@@ -120,14 +123,15 @@ static void chv_set_memory_dvfs(struct drm_i915_private *dev_priv, bool enable)
 
 	if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2) &
 		      FORCE_DDR_FREQ_REQ_ACK) == 0, 3))
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"timed out waiting for Punit DDR DVFS request\n");
 
 	vlv_punit_put(dev_priv);
 }
 
-static void chv_set_memory_pm5(struct drm_i915_private *dev_priv, bool enable)
+static void chv_set_memory_pm5(struct intel_display *display, bool enable)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 val;
 
 	vlv_punit_get(dev_priv);
@@ -145,53 +149,52 @@ static void chv_set_memory_pm5(struct drm_i915_private *dev_priv, bool enable)
 #define FW_WM(value, plane) \
 	(((value) << DSPFW_ ## plane ## _SHIFT) & DSPFW_ ## plane ## _MASK)
 
-static bool _intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enable)
+static bool _intel_set_memory_cxsr(struct intel_display *display, bool enable)
 {
-	struct intel_display *display = &dev_priv->display;
 	bool was_enabled;
 	u32 val;
 
-	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
-		was_enabled = intel_uncore_read(&dev_priv->uncore, FW_BLC_SELF_VLV) & FW_CSPWRDWNEN;
-		intel_uncore_write(&dev_priv->uncore, FW_BLC_SELF_VLV, enable ? FW_CSPWRDWNEN : 0);
-		intel_uncore_posting_read(&dev_priv->uncore, FW_BLC_SELF_VLV);
-	} else if (IS_G4X(dev_priv) || IS_I965GM(dev_priv)) {
-		was_enabled = intel_uncore_read(&dev_priv->uncore, FW_BLC_SELF) & FW_BLC_SELF_EN;
-		intel_uncore_write(&dev_priv->uncore, FW_BLC_SELF, enable ? FW_BLC_SELF_EN : 0);
-		intel_uncore_posting_read(&dev_priv->uncore, FW_BLC_SELF);
-	} else if (IS_PINEVIEW(dev_priv)) {
-		val = intel_uncore_read(&dev_priv->uncore, DSPFW3(dev_priv));
+	if (display->platform.valleyview || display->platform.cherryview) {
+		was_enabled = intel_de_read(display, FW_BLC_SELF_VLV) & FW_CSPWRDWNEN;
+		intel_de_write(display, FW_BLC_SELF_VLV, enable ? FW_CSPWRDWNEN : 0);
+		intel_de_posting_read(display, FW_BLC_SELF_VLV);
+	} else if (display->platform.g4x || display->platform.i965gm) {
+		was_enabled = intel_de_read(display, FW_BLC_SELF) & FW_BLC_SELF_EN;
+		intel_de_write(display, FW_BLC_SELF, enable ? FW_BLC_SELF_EN : 0);
+		intel_de_posting_read(display, FW_BLC_SELF);
+	} else if (display->platform.pineview) {
+		val = intel_de_read(display, DSPFW3(display));
 		was_enabled = val & PINEVIEW_SELF_REFRESH_EN;
 		if (enable)
 			val |= PINEVIEW_SELF_REFRESH_EN;
 		else
 			val &= ~PINEVIEW_SELF_REFRESH_EN;
-		intel_uncore_write(&dev_priv->uncore, DSPFW3(dev_priv), val);
-		intel_uncore_posting_read(&dev_priv->uncore, DSPFW3(dev_priv));
-	} else if (IS_I945G(dev_priv) || IS_I945GM(dev_priv)) {
-		was_enabled = intel_uncore_read(&dev_priv->uncore, FW_BLC_SELF) & FW_BLC_SELF_EN;
+		intel_de_write(display, DSPFW3(display), val);
+		intel_de_posting_read(display, DSPFW3(display));
+	} else if (display->platform.i945g || display->platform.i945gm) {
+		was_enabled = intel_de_read(display, FW_BLC_SELF) & FW_BLC_SELF_EN;
 		val = enable ? _MASKED_BIT_ENABLE(FW_BLC_SELF_EN) :
 			       _MASKED_BIT_DISABLE(FW_BLC_SELF_EN);
-		intel_uncore_write(&dev_priv->uncore, FW_BLC_SELF, val);
-		intel_uncore_posting_read(&dev_priv->uncore, FW_BLC_SELF);
-	} else if (IS_I915GM(dev_priv)) {
+		intel_de_write(display, FW_BLC_SELF, val);
+		intel_de_posting_read(display, FW_BLC_SELF);
+	} else if (display->platform.i915gm) {
 		/*
 		 * FIXME can't find a bit like this for 915G, and
 		 * yet it does have the related watermark in
 		 * FW_BLC_SELF. What's going on?
 		 */
-		was_enabled = intel_uncore_read(&dev_priv->uncore, INSTPM) & INSTPM_SELF_EN;
+		was_enabled = intel_de_read(display, INSTPM) & INSTPM_SELF_EN;
 		val = enable ? _MASKED_BIT_ENABLE(INSTPM_SELF_EN) :
 			       _MASKED_BIT_DISABLE(INSTPM_SELF_EN);
-		intel_uncore_write(&dev_priv->uncore, INSTPM, val);
-		intel_uncore_posting_read(&dev_priv->uncore, INSTPM);
+		intel_de_write(display, INSTPM, val);
+		intel_de_posting_read(display, INSTPM);
 	} else {
 		return false;
 	}
 
 	trace_intel_memory_cxsr(display, was_enabled, enable);
 
-	drm_dbg_kms(&dev_priv->drm, "memory self-refresh is %s (was %s)\n",
+	drm_dbg_kms(display->drm, "memory self-refresh is %s (was %s)\n",
 		    str_enabled_disabled(enable),
 		    str_enabled_disabled(was_enabled));
 
@@ -200,7 +203,7 @@ static bool _intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enabl
 
 /**
  * intel_set_memory_cxsr - Configure CxSR state
- * @dev_priv: i915 device
+ * @display: display device
  * @enable: Allow vs. disallow CxSR
  *
  * Allow or disallow the system to enter a special CxSR
@@ -235,17 +238,17 @@ static bool _intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enabl
  * the hardware w.r.t. HPLL SR when writing to plane registers.
  * Disallowing just CxSR is sufficient.
  */
-bool intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enable)
+bool intel_set_memory_cxsr(struct intel_display *display, bool enable)
 {
 	bool ret;
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
-	ret = _intel_set_memory_cxsr(dev_priv, enable);
-	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		dev_priv->display.wm.vlv.cxsr = enable;
-	else if (IS_G4X(dev_priv))
-		dev_priv->display.wm.g4x.cxsr = enable;
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
+	ret = _intel_set_memory_cxsr(display, enable);
+	if (display->platform.valleyview || display->platform.cherryview)
+		display->wm.vlv.cxsr = enable;
+	else if (display->platform.g4x)
+		display->wm.g4x.cxsr = enable;
+	mutex_unlock(&display->wm.wm_mutex);
 
 	return ret;
 }
@@ -271,8 +274,8 @@ static const int pessimal_latency_ns = 5000;
 
 static void vlv_get_fifo_size(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct vlv_fifo_state *fifo_state = &crtc_state->wm.vlv.fifo_state;
 	enum pipe pipe = crtc->pipe;
 	int sprite0_start, sprite1_start;
@@ -280,22 +283,20 @@ static void vlv_get_fifo_size(struct intel_crtc_state *crtc_state)
 
 	switch (pipe) {
 	case PIPE_A:
-		dsparb = intel_uncore_read(&dev_priv->uncore,
-					   DSPARB(dev_priv));
-		dsparb2 = intel_uncore_read(&dev_priv->uncore, DSPARB2);
+		dsparb = intel_de_read(display, DSPARB(display));
+		dsparb2 = intel_de_read(display, DSPARB2);
 		sprite0_start = VLV_FIFO_START(dsparb, dsparb2, 0, 0);
 		sprite1_start = VLV_FIFO_START(dsparb, dsparb2, 8, 4);
 		break;
 	case PIPE_B:
-		dsparb = intel_uncore_read(&dev_priv->uncore,
-					   DSPARB(dev_priv));
-		dsparb2 = intel_uncore_read(&dev_priv->uncore, DSPARB2);
+		dsparb = intel_de_read(display, DSPARB(display));
+		dsparb2 = intel_de_read(display, DSPARB2);
 		sprite0_start = VLV_FIFO_START(dsparb, dsparb2, 16, 8);
 		sprite1_start = VLV_FIFO_START(dsparb, dsparb2, 24, 12);
 		break;
 	case PIPE_C:
-		dsparb2 = intel_uncore_read(&dev_priv->uncore, DSPARB2);
-		dsparb3 = intel_uncore_read(&dev_priv->uncore, DSPARB3);
+		dsparb2 = intel_de_read(display, DSPARB2);
+		dsparb3 = intel_de_read(display, DSPARB3);
 		sprite0_start = VLV_FIFO_START(dsparb3, dsparb2, 0, 16);
 		sprite1_start = VLV_FIFO_START(dsparb3, dsparb2, 8, 20);
 		break;
@@ -310,26 +311,26 @@ static void vlv_get_fifo_size(struct intel_crtc_state *crtc_state)
 	fifo_state->plane[PLANE_CURSOR] = 63;
 }
 
-static int i9xx_get_fifo_size(struct drm_i915_private *dev_priv,
+static int i9xx_get_fifo_size(struct intel_display *display,
 			      enum i9xx_plane_id i9xx_plane)
 {
-	u32 dsparb = intel_uncore_read(&dev_priv->uncore, DSPARB(dev_priv));
+	u32 dsparb = intel_de_read(display, DSPARB(display));
 	int size;
 
 	size = dsparb & 0x7f;
 	if (i9xx_plane == PLANE_B)
 		size = ((dsparb >> DSPARB_CSTART_SHIFT) & 0x7f) - size;
 
-	drm_dbg_kms(&dev_priv->drm, "FIFO size - (0x%08x) %c: %d\n",
+	drm_dbg_kms(display->drm, "FIFO size - (0x%08x) %c: %d\n",
 		    dsparb, plane_name(i9xx_plane), size);
 
 	return size;
 }
 
-static int i830_get_fifo_size(struct drm_i915_private *dev_priv,
+static int i830_get_fifo_size(struct intel_display *display,
 			      enum i9xx_plane_id i9xx_plane)
 {
-	u32 dsparb = intel_uncore_read(&dev_priv->uncore, DSPARB(dev_priv));
+	u32 dsparb = intel_de_read(display, DSPARB(display));
 	int size;
 
 	size = dsparb & 0x1ff;
@@ -337,22 +338,22 @@ static int i830_get_fifo_size(struct drm_i915_private *dev_priv,
 		size = ((dsparb >> DSPARB_BEND_SHIFT) & 0x1ff) - size;
 	size >>= 1; /* Convert to cachelines */
 
-	drm_dbg_kms(&dev_priv->drm, "FIFO size - (0x%08x) %c: %d\n",
+	drm_dbg_kms(display->drm, "FIFO size - (0x%08x) %c: %d\n",
 		    dsparb, plane_name(i9xx_plane), size);
 
 	return size;
 }
 
-static int i845_get_fifo_size(struct drm_i915_private *dev_priv,
+static int i845_get_fifo_size(struct intel_display *display,
 			      enum i9xx_plane_id i9xx_plane)
 {
-	u32 dsparb = intel_uncore_read(&dev_priv->uncore, DSPARB(dev_priv));
+	u32 dsparb = intel_de_read(display, DSPARB(display));
 	int size;
 
 	size = dsparb & 0x7f;
 	size >>= 2; /* Convert to cachelines */
 
-	drm_dbg_kms(&dev_priv->drm, "FIFO size - (0x%08x) %c: %d\n",
+	drm_dbg_kms(display->drm, "FIFO size - (0x%08x) %c: %d\n",
 		    dsparb, plane_name(i9xx_plane), size);
 
 	return size;
@@ -537,7 +538,7 @@ static unsigned int intel_wm_method2(unsigned int pixel_rate,
 
 /**
  * intel_calculate_wm - calculate watermark level
- * @i915: the device
+ * @display: display device
  * @pixel_rate: pixel clock
  * @wm: chip FIFO params
  * @fifo_size: size of the FIFO buffer
@@ -555,7 +556,7 @@ static unsigned int intel_wm_method2(unsigned int pixel_rate,
  * past the watermark point.  If the FIFO drains completely, a FIFO underrun
  * will occur, and a display engine hang could result.
  */
-static unsigned int intel_calculate_wm(struct drm_i915_private *i915,
+static unsigned int intel_calculate_wm(struct intel_display *display,
 				       int pixel_rate,
 				       const struct intel_watermark_params *wm,
 				       int fifo_size, int cpp,
@@ -573,10 +574,10 @@ static unsigned int intel_calculate_wm(struct drm_i915_private *i915,
 				   latency_ns / 100);
 	entries = DIV_ROUND_UP(entries, wm->cacheline_size) +
 		wm->guard_size;
-	drm_dbg_kms(&i915->drm, "FIFO entries required for mode: %d\n", entries);
+	drm_dbg_kms(display->drm, "FIFO entries required for mode: %d\n", entries);
 
 	wm_size = fifo_size - entries;
-	drm_dbg_kms(&i915->drm, "FIFO watermark level: %d\n", wm_size);
+	drm_dbg_kms(display->drm, "FIFO watermark level: %d\n", wm_size);
 
 	/* Don't promote wm_size to unsigned... */
 	if (wm_size > wm->max_wm)
@@ -626,11 +627,11 @@ static bool intel_crtc_active(struct intel_crtc *crtc)
 		crtc->config->hw.adjusted_mode.crtc_clock;
 }
 
-static struct intel_crtc *single_enabled_crtc(struct drm_i915_private *dev_priv)
+static struct intel_crtc *single_enabled_crtc(struct intel_display *display)
 {
 	struct intel_crtc *crtc, *enabled = NULL;
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		if (intel_crtc_active(crtc)) {
 			if (enabled)
 				return NULL;
@@ -641,21 +642,21 @@ static struct intel_crtc *single_enabled_crtc(struct drm_i915_private *dev_priv)
 	return enabled;
 }
 
-static void pnv_update_wm(struct drm_i915_private *dev_priv)
+static void pnv_update_wm(struct intel_display *display)
 {
 	struct intel_crtc *crtc;
 	const struct cxsr_latency *latency;
 	u32 reg;
 	unsigned int wm;
 
-	latency = pnv_get_cxsr_latency(dev_priv);
+	latency = pnv_get_cxsr_latency(display);
 	if (!latency) {
-		drm_dbg_kms(&dev_priv->drm, "Unknown FSB/MEM, disabling CxSR\n");
-		intel_set_memory_cxsr(dev_priv, false);
+		drm_dbg_kms(display->drm, "Unknown FSB/MEM, disabling CxSR\n");
+		intel_set_memory_cxsr(display, false);
 		return;
 	}
 
-	crtc = single_enabled_crtc(dev_priv);
+	crtc = single_enabled_crtc(display);
 	if (crtc) {
 		const struct drm_framebuffer *fb =
 			crtc->base.primary->state->fb;
@@ -663,47 +664,46 @@ static void pnv_update_wm(struct drm_i915_private *dev_priv)
 		int cpp = fb->format->cpp[0];
 
 		/* Display SR */
-		wm = intel_calculate_wm(dev_priv, pixel_rate,
+		wm = intel_calculate_wm(display, pixel_rate,
 					&pnv_display_wm,
 					pnv_display_wm.fifo_size,
 					cpp, latency->display_sr);
-		reg = intel_uncore_read(&dev_priv->uncore, DSPFW1(dev_priv));
+		reg = intel_de_read(display, DSPFW1(display));
 		reg &= ~DSPFW_SR_MASK;
 		reg |= FW_WM(wm, SR);
-		intel_uncore_write(&dev_priv->uncore, DSPFW1(dev_priv), reg);
-		drm_dbg_kms(&dev_priv->drm, "DSPFW1 register is %x\n", reg);
+		intel_de_write(display, DSPFW1(display), reg);
+		drm_dbg_kms(display->drm, "DSPFW1 register is %x\n", reg);
 
 		/* cursor SR */
-		wm = intel_calculate_wm(dev_priv, pixel_rate,
+		wm = intel_calculate_wm(display, pixel_rate,
 					&pnv_cursor_wm,
 					pnv_display_wm.fifo_size,
 					4, latency->cursor_sr);
-		intel_uncore_rmw(&dev_priv->uncore, DSPFW3(dev_priv),
-				 DSPFW_CURSOR_SR_MASK,
-				 FW_WM(wm, CURSOR_SR));
+		intel_de_rmw(display, DSPFW3(display),
+			     DSPFW_CURSOR_SR_MASK, FW_WM(wm, CURSOR_SR));
 
 		/* Display HPLL off SR */
-		wm = intel_calculate_wm(dev_priv, pixel_rate,
+		wm = intel_calculate_wm(display, pixel_rate,
 					&pnv_display_hplloff_wm,
 					pnv_display_hplloff_wm.fifo_size,
 					cpp, latency->display_hpll_disable);
-		intel_uncore_rmw(&dev_priv->uncore, DSPFW3(dev_priv),
-				 DSPFW_HPLL_SR_MASK, FW_WM(wm, HPLL_SR));
+		intel_de_rmw(display, DSPFW3(display),
+			     DSPFW_HPLL_SR_MASK, FW_WM(wm, HPLL_SR));
 
 		/* cursor HPLL off SR */
-		wm = intel_calculate_wm(dev_priv, pixel_rate,
+		wm = intel_calculate_wm(display, pixel_rate,
 					&pnv_cursor_hplloff_wm,
 					pnv_display_hplloff_wm.fifo_size,
 					4, latency->cursor_hpll_disable);
-		reg = intel_uncore_read(&dev_priv->uncore, DSPFW3(dev_priv));
+		reg = intel_de_read(display, DSPFW3(display));
 		reg &= ~DSPFW_HPLL_CURSOR_MASK;
 		reg |= FW_WM(wm, HPLL_CURSOR);
-		intel_uncore_write(&dev_priv->uncore, DSPFW3(dev_priv), reg);
-		drm_dbg_kms(&dev_priv->drm, "DSPFW3 register is %x\n", reg);
+		intel_de_write(display, DSPFW3(display), reg);
+		drm_dbg_kms(display->drm, "DSPFW3 register is %x\n", reg);
 
-		intel_set_memory_cxsr(dev_priv, true);
+		intel_set_memory_cxsr(display, true);
 	} else {
-		intel_set_memory_cxsr(dev_priv, false);
+		intel_set_memory_cxsr(display, false);
 	}
 }
 
@@ -794,53 +794,51 @@ static unsigned int g4x_tlb_miss_wa(int fifo_size, int width, int cpp)
 	return max(0, tlb_miss);
 }
 
-static void g4x_write_wm_values(struct drm_i915_private *dev_priv,
+static void g4x_write_wm_values(struct intel_display *display,
 				const struct g4x_wm_values *wm)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe)
+	for_each_pipe(display, pipe)
 		trace_g4x_wm(intel_crtc_for_pipe(display, pipe), wm);
 
-	intel_uncore_write(&dev_priv->uncore, DSPFW1(dev_priv),
-			   FW_WM(wm->sr.plane, SR) |
-			   FW_WM(wm->pipe[PIPE_B].plane[PLANE_CURSOR], CURSORB) |
-			   FW_WM(wm->pipe[PIPE_B].plane[PLANE_PRIMARY], PLANEB) |
-			   FW_WM(wm->pipe[PIPE_A].plane[PLANE_PRIMARY], PLANEA));
-	intel_uncore_write(&dev_priv->uncore, DSPFW2(dev_priv),
-			   (wm->fbc_en ? DSPFW_FBC_SR_EN : 0) |
-			   FW_WM(wm->sr.fbc, FBC_SR) |
-			   FW_WM(wm->hpll.fbc, FBC_HPLL_SR) |
-			   FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE0], SPRITEB) |
-			   FW_WM(wm->pipe[PIPE_A].plane[PLANE_CURSOR], CURSORA) |
-			   FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE0], SPRITEA));
-	intel_uncore_write(&dev_priv->uncore, DSPFW3(dev_priv),
-			   (wm->hpll_en ? DSPFW_HPLL_SR_EN : 0) |
-			   FW_WM(wm->sr.cursor, CURSOR_SR) |
-			   FW_WM(wm->hpll.cursor, HPLL_CURSOR) |
-			   FW_WM(wm->hpll.plane, HPLL_SR));
-
-	intel_uncore_posting_read(&dev_priv->uncore, DSPFW1(dev_priv));
+	intel_de_write(display, DSPFW1(display),
+		       FW_WM(wm->sr.plane, SR) |
+		       FW_WM(wm->pipe[PIPE_B].plane[PLANE_CURSOR], CURSORB) |
+		       FW_WM(wm->pipe[PIPE_B].plane[PLANE_PRIMARY], PLANEB) |
+		       FW_WM(wm->pipe[PIPE_A].plane[PLANE_PRIMARY], PLANEA));
+	intel_de_write(display, DSPFW2(display),
+		       (wm->fbc_en ? DSPFW_FBC_SR_EN : 0) |
+		       FW_WM(wm->sr.fbc, FBC_SR) |
+		       FW_WM(wm->hpll.fbc, FBC_HPLL_SR) |
+		       FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE0], SPRITEB) |
+		       FW_WM(wm->pipe[PIPE_A].plane[PLANE_CURSOR], CURSORA) |
+		       FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE0], SPRITEA));
+	intel_de_write(display, DSPFW3(display),
+		       (wm->hpll_en ? DSPFW_HPLL_SR_EN : 0) |
+		       FW_WM(wm->sr.cursor, CURSOR_SR) |
+		       FW_WM(wm->hpll.cursor, HPLL_CURSOR) |
+		       FW_WM(wm->hpll.plane, HPLL_SR));
+
+	intel_de_posting_read(display, DSPFW1(display));
 }
 
 #define FW_WM_VLV(value, plane) \
 	(((value) << DSPFW_ ## plane ## _SHIFT) & DSPFW_ ## plane ## _MASK_VLV)
 
-static void vlv_write_wm_values(struct drm_i915_private *dev_priv,
+static void vlv_write_wm_values(struct intel_display *display,
 				const struct vlv_wm_values *wm)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		trace_vlv_wm(intel_crtc_for_pipe(display, pipe), wm);
 
-		intel_uncore_write(&dev_priv->uncore, VLV_DDL(pipe),
-				   (wm->ddl[pipe].plane[PLANE_CURSOR] << DDL_CURSOR_SHIFT) |
-				   (wm->ddl[pipe].plane[PLANE_SPRITE1] << DDL_SPRITE_SHIFT(1)) |
-				   (wm->ddl[pipe].plane[PLANE_SPRITE0] << DDL_SPRITE_SHIFT(0)) |
-				   (wm->ddl[pipe].plane[PLANE_PRIMARY] << DDL_PLANE_SHIFT));
+		intel_de_write(display, VLV_DDL(pipe),
+			       (wm->ddl[pipe].plane[PLANE_CURSOR] << DDL_CURSOR_SHIFT) |
+			       (wm->ddl[pipe].plane[PLANE_SPRITE1] << DDL_SPRITE_SHIFT(1)) |
+			       (wm->ddl[pipe].plane[PLANE_SPRITE0] << DDL_SPRITE_SHIFT(0)) |
+			       (wm->ddl[pipe].plane[PLANE_PRIMARY] << DDL_PLANE_SHIFT));
 	}
 
 	/*
@@ -848,72 +846,72 @@ static void vlv_write_wm_values(struct drm_i915_private *dev_priv,
 	 * high order bits so that there are no out of bounds values
 	 * present in the registers during the reprogramming.
 	 */
-	intel_uncore_write(&dev_priv->uncore, DSPHOWM, 0);
-	intel_uncore_write(&dev_priv->uncore, DSPHOWM1, 0);
-	intel_uncore_write(&dev_priv->uncore, DSPFW4, 0);
-	intel_uncore_write(&dev_priv->uncore, DSPFW5, 0);
-	intel_uncore_write(&dev_priv->uncore, DSPFW6, 0);
-
-	intel_uncore_write(&dev_priv->uncore, DSPFW1(dev_priv),
-			   FW_WM(wm->sr.plane, SR) |
-			   FW_WM(wm->pipe[PIPE_B].plane[PLANE_CURSOR], CURSORB) |
-			   FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_PRIMARY], PLANEB) |
-			   FW_WM_VLV(wm->pipe[PIPE_A].plane[PLANE_PRIMARY], PLANEA));
-	intel_uncore_write(&dev_priv->uncore, DSPFW2(dev_priv),
-			   FW_WM_VLV(wm->pipe[PIPE_A].plane[PLANE_SPRITE1], SPRITEB) |
-			   FW_WM(wm->pipe[PIPE_A].plane[PLANE_CURSOR], CURSORA) |
-			   FW_WM_VLV(wm->pipe[PIPE_A].plane[PLANE_SPRITE0], SPRITEA));
-	intel_uncore_write(&dev_priv->uncore, DSPFW3(dev_priv),
-			   FW_WM(wm->sr.cursor, CURSOR_SR));
-
-	if (IS_CHERRYVIEW(dev_priv)) {
-		intel_uncore_write(&dev_priv->uncore, DSPFW7_CHV,
-				   FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE1], SPRITED) |
-				   FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE0], SPRITEC));
-		intel_uncore_write(&dev_priv->uncore, DSPFW8_CHV,
-				   FW_WM_VLV(wm->pipe[PIPE_C].plane[PLANE_SPRITE1], SPRITEF) |
-				   FW_WM_VLV(wm->pipe[PIPE_C].plane[PLANE_SPRITE0], SPRITEE));
-		intel_uncore_write(&dev_priv->uncore, DSPFW9_CHV,
-				   FW_WM_VLV(wm->pipe[PIPE_C].plane[PLANE_PRIMARY], PLANEC) |
-				   FW_WM(wm->pipe[PIPE_C].plane[PLANE_CURSOR], CURSORC));
-		intel_uncore_write(&dev_priv->uncore, DSPHOWM,
-				   FW_WM(wm->sr.plane >> 9, SR_HI) |
-				   FW_WM(wm->pipe[PIPE_C].plane[PLANE_SPRITE1] >> 8, SPRITEF_HI) |
-				   FW_WM(wm->pipe[PIPE_C].plane[PLANE_SPRITE0] >> 8, SPRITEE_HI) |
-				   FW_WM(wm->pipe[PIPE_C].plane[PLANE_PRIMARY] >> 8, PLANEC_HI) |
-				   FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE1] >> 8, SPRITED_HI) |
-				   FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE0] >> 8, SPRITEC_HI) |
-				   FW_WM(wm->pipe[PIPE_B].plane[PLANE_PRIMARY] >> 8, PLANEB_HI) |
-				   FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE1] >> 8, SPRITEB_HI) |
-				   FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE0] >> 8, SPRITEA_HI) |
-				   FW_WM(wm->pipe[PIPE_A].plane[PLANE_PRIMARY] >> 8, PLANEA_HI));
+	intel_de_write(display, DSPHOWM, 0);
+	intel_de_write(display, DSPHOWM1, 0);
+	intel_de_write(display, DSPFW4, 0);
+	intel_de_write(display, DSPFW5, 0);
+	intel_de_write(display, DSPFW6, 0);
+
+	intel_de_write(display, DSPFW1(display),
+		       FW_WM(wm->sr.plane, SR) |
+		       FW_WM(wm->pipe[PIPE_B].plane[PLANE_CURSOR], CURSORB) |
+		       FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_PRIMARY], PLANEB) |
+		       FW_WM_VLV(wm->pipe[PIPE_A].plane[PLANE_PRIMARY], PLANEA));
+	intel_de_write(display, DSPFW2(display),
+		       FW_WM_VLV(wm->pipe[PIPE_A].plane[PLANE_SPRITE1], SPRITEB) |
+		       FW_WM(wm->pipe[PIPE_A].plane[PLANE_CURSOR], CURSORA) |
+		       FW_WM_VLV(wm->pipe[PIPE_A].plane[PLANE_SPRITE0], SPRITEA));
+	intel_de_write(display, DSPFW3(display),
+		       FW_WM(wm->sr.cursor, CURSOR_SR));
+
+	if (display->platform.cherryview) {
+		intel_de_write(display, DSPFW7_CHV,
+			       FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE1], SPRITED) |
+			       FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE0], SPRITEC));
+		intel_de_write(display, DSPFW8_CHV,
+			       FW_WM_VLV(wm->pipe[PIPE_C].plane[PLANE_SPRITE1], SPRITEF) |
+			       FW_WM_VLV(wm->pipe[PIPE_C].plane[PLANE_SPRITE0], SPRITEE));
+		intel_de_write(display, DSPFW9_CHV,
+			       FW_WM_VLV(wm->pipe[PIPE_C].plane[PLANE_PRIMARY], PLANEC) |
+			       FW_WM(wm->pipe[PIPE_C].plane[PLANE_CURSOR], CURSORC));
+		intel_de_write(display, DSPHOWM,
+			       FW_WM(wm->sr.plane >> 9, SR_HI) |
+			       FW_WM(wm->pipe[PIPE_C].plane[PLANE_SPRITE1] >> 8, SPRITEF_HI) |
+			       FW_WM(wm->pipe[PIPE_C].plane[PLANE_SPRITE0] >> 8, SPRITEE_HI) |
+			       FW_WM(wm->pipe[PIPE_C].plane[PLANE_PRIMARY] >> 8, PLANEC_HI) |
+			       FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE1] >> 8, SPRITED_HI) |
+			       FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE0] >> 8, SPRITEC_HI) |
+			       FW_WM(wm->pipe[PIPE_B].plane[PLANE_PRIMARY] >> 8, PLANEB_HI) |
+			       FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE1] >> 8, SPRITEB_HI) |
+			       FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE0] >> 8, SPRITEA_HI) |
+			       FW_WM(wm->pipe[PIPE_A].plane[PLANE_PRIMARY] >> 8, PLANEA_HI));
 	} else {
-		intel_uncore_write(&dev_priv->uncore, DSPFW7,
-				   FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE1], SPRITED) |
-				   FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE0], SPRITEC));
-		intel_uncore_write(&dev_priv->uncore, DSPHOWM,
-				   FW_WM(wm->sr.plane >> 9, SR_HI) |
-				   FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE1] >> 8, SPRITED_HI) |
-				   FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE0] >> 8, SPRITEC_HI) |
-				   FW_WM(wm->pipe[PIPE_B].plane[PLANE_PRIMARY] >> 8, PLANEB_HI) |
-				   FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE1] >> 8, SPRITEB_HI) |
-				   FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE0] >> 8, SPRITEA_HI) |
-				   FW_WM(wm->pipe[PIPE_A].plane[PLANE_PRIMARY] >> 8, PLANEA_HI));
+		intel_de_write(display, DSPFW7,
+			       FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE1], SPRITED) |
+			       FW_WM_VLV(wm->pipe[PIPE_B].plane[PLANE_SPRITE0], SPRITEC));
+		intel_de_write(display, DSPHOWM,
+			       FW_WM(wm->sr.plane >> 9, SR_HI) |
+			       FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE1] >> 8, SPRITED_HI) |
+			       FW_WM(wm->pipe[PIPE_B].plane[PLANE_SPRITE0] >> 8, SPRITEC_HI) |
+			       FW_WM(wm->pipe[PIPE_B].plane[PLANE_PRIMARY] >> 8, PLANEB_HI) |
+			       FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE1] >> 8, SPRITEB_HI) |
+			       FW_WM(wm->pipe[PIPE_A].plane[PLANE_SPRITE0] >> 8, SPRITEA_HI) |
+			       FW_WM(wm->pipe[PIPE_A].plane[PLANE_PRIMARY] >> 8, PLANEA_HI));
 	}
 
-	intel_uncore_posting_read(&dev_priv->uncore, DSPFW1(dev_priv));
+	intel_de_posting_read(display, DSPFW1(display));
 }
 
 #undef FW_WM_VLV
 
-static void g4x_setup_wm_latency(struct drm_i915_private *dev_priv)
+static void g4x_setup_wm_latency(struct intel_display *display)
 {
 	/* all latencies in usec */
-	dev_priv->display.wm.pri_latency[G4X_WM_LEVEL_NORMAL] = 5;
-	dev_priv->display.wm.pri_latency[G4X_WM_LEVEL_SR] = 12;
-	dev_priv->display.wm.pri_latency[G4X_WM_LEVEL_HPLL] = 35;
+	display->wm.pri_latency[G4X_WM_LEVEL_NORMAL] = 5;
+	display->wm.pri_latency[G4X_WM_LEVEL_SR] = 12;
+	display->wm.pri_latency[G4X_WM_LEVEL_HPLL] = 35;
 
-	dev_priv->display.wm.num_levels = G4X_WM_LEVEL_HPLL + 1;
+	display->wm.num_levels = G4X_WM_LEVEL_HPLL + 1;
 }
 
 static int g4x_plane_fifo_size(enum plane_id plane_id, int level)
@@ -962,11 +960,11 @@ static u16 g4x_compute_wm(const struct intel_crtc_state *crtc_state,
 			  const struct intel_plane_state *plane_state,
 			  int level)
 {
+	struct intel_display *display = to_intel_display(plane_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
 	const struct drm_display_mode *pipe_mode =
 		&crtc_state->hw.pipe_mode;
-	unsigned int latency = dev_priv->display.wm.pri_latency[level] * 10;
+	unsigned int latency = display->wm.pri_latency[level] * 10;
 	unsigned int pixel_rate, htotal, cpp, width, wm;
 
 	if (latency == 0)
@@ -1017,10 +1015,10 @@ static u16 g4x_compute_wm(const struct intel_crtc_state *crtc_state,
 static bool g4x_raw_plane_wm_set(struct intel_crtc_state *crtc_state,
 				 int level, enum plane_id plane_id, u16 value)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	bool dirty = false;
 
-	for (; level < dev_priv->display.wm.num_levels; level++) {
+	for (; level < display->wm.num_levels; level++) {
 		struct g4x_pipe_wm *raw = &crtc_state->wm.g4x.raw[level];
 
 		dirty |= raw->plane[plane_id] != value;
@@ -1033,13 +1031,13 @@ static bool g4x_raw_plane_wm_set(struct intel_crtc_state *crtc_state,
 static bool g4x_raw_fbc_wm_set(struct intel_crtc_state *crtc_state,
 			       int level, u16 value)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	bool dirty = false;
 
 	/* NORMAL level doesn't have an FBC watermark */
 	level = max(level, G4X_WM_LEVEL_SR);
 
-	for (; level < dev_priv->display.wm.num_levels; level++) {
+	for (; level < display->wm.num_levels; level++) {
 		struct g4x_pipe_wm *raw = &crtc_state->wm.g4x.raw[level];
 
 		dirty |= raw->fbc != value;
@@ -1056,8 +1054,8 @@ static u32 ilk_compute_fbc_wm(const struct intel_crtc_state *crtc_state,
 static bool g4x_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 				     const struct intel_plane_state *plane_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	enum plane_id plane_id = plane->id;
 	bool dirty = false;
 	int level;
@@ -1069,7 +1067,7 @@ static bool g4x_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 		goto out;
 	}
 
-	for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+	for (level = 0; level < display->wm.num_levels; level++) {
 		struct g4x_pipe_wm *raw = &crtc_state->wm.g4x.raw[level];
 		int wm, max_wm;
 
@@ -1109,7 +1107,7 @@ static bool g4x_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 
  out:
 	if (dirty) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "%s watermarks: normal=%d, SR=%d, HPLL=%d\n",
 			    plane->base.name,
 			    crtc_state->wm.g4x.raw[G4X_WM_LEVEL_NORMAL].plane[plane_id],
@@ -1117,7 +1115,7 @@ static bool g4x_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 			    crtc_state->wm.g4x.raw[G4X_WM_LEVEL_HPLL].plane[plane_id]);
 
 		if (plane_id == PLANE_PRIMARY)
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "FBC watermarks: SR=%d, HPLL=%d\n",
 				    crtc_state->wm.g4x.raw[G4X_WM_LEVEL_SR].fbc,
 				    crtc_state->wm.g4x.raw[G4X_WM_LEVEL_HPLL].fbc);
@@ -1137,9 +1135,9 @@ static bool g4x_raw_plane_wm_is_valid(const struct intel_crtc_state *crtc_state,
 static bool g4x_raw_crtc_wm_is_valid(const struct intel_crtc_state *crtc_state,
 				     int level)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
-	if (level >= dev_priv->display.wm.num_levels)
+	if (level >= display->wm.num_levels)
 		return false;
 
 	return g4x_raw_plane_wm_is_valid(crtc_state, PLANE_PRIMARY, level) &&
@@ -1281,7 +1279,7 @@ static int g4x_compute_pipe_wm(struct intel_atomic_state *state,
 static int g4x_compute_intermediate_wm(struct intel_atomic_state *state,
 				       struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_crtc_state *old_crtc_state =
@@ -1311,7 +1309,7 @@ static int g4x_compute_intermediate_wm(struct intel_atomic_state *state,
 			max(optimal->wm.plane[plane_id],
 			    active->wm.plane[plane_id]);
 
-		drm_WARN_ON(&dev_priv->drm, intermediate->wm.plane[plane_id] >
+		drm_WARN_ON(display->drm, intermediate->wm.plane[plane_id] >
 			    g4x_plane_fifo_size(plane_id, G4X_WM_LEVEL_NORMAL));
 	}
 
@@ -1329,23 +1327,23 @@ static int g4x_compute_intermediate_wm(struct intel_atomic_state *state,
 	intermediate->hpll.fbc = max(optimal->hpll.fbc,
 				     active->hpll.fbc);
 
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(display->drm,
 		    (intermediate->sr.plane >
 		     g4x_plane_fifo_size(PLANE_PRIMARY, G4X_WM_LEVEL_SR) ||
 		     intermediate->sr.cursor >
 		     g4x_plane_fifo_size(PLANE_CURSOR, G4X_WM_LEVEL_SR)) &&
 		    intermediate->cxsr);
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(display->drm,
 		    (intermediate->sr.plane >
 		     g4x_plane_fifo_size(PLANE_PRIMARY, G4X_WM_LEVEL_HPLL) ||
 		     intermediate->sr.cursor >
 		     g4x_plane_fifo_size(PLANE_CURSOR, G4X_WM_LEVEL_HPLL)) &&
 		    intermediate->hpll_en);
 
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(display->drm,
 		    intermediate->sr.fbc > g4x_fbc_fifo_size(1) &&
 		    intermediate->fbc_en && intermediate->cxsr);
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(display->drm,
 		    intermediate->hpll.fbc > g4x_fbc_fifo_size(2) &&
 		    intermediate->fbc_en && intermediate->hpll_en);
 
@@ -1376,7 +1374,7 @@ static int g4x_compute_watermarks(struct intel_atomic_state *state,
 	return 0;
 }
 
-static void g4x_merge_wm(struct drm_i915_private *dev_priv,
+static void g4x_merge_wm(struct intel_display *display,
 			 struct g4x_wm_values *wm)
 {
 	struct intel_crtc *crtc;
@@ -1386,7 +1384,7 @@ static void g4x_merge_wm(struct drm_i915_private *dev_priv,
 	wm->hpll_en = true;
 	wm->fbc_en = true;
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct g4x_wm_state *wm_state = &crtc->wm.active.g4x;
 
 		if (!crtc->active)
@@ -1408,7 +1406,7 @@ static void g4x_merge_wm(struct drm_i915_private *dev_priv,
 		wm->fbc_en = false;
 	}
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct g4x_wm_state *wm_state = &crtc->wm.active.g4x;
 		enum pipe pipe = crtc->pipe;
 
@@ -1420,23 +1418,23 @@ static void g4x_merge_wm(struct drm_i915_private *dev_priv,
 	}
 }
 
-static void g4x_program_watermarks(struct drm_i915_private *dev_priv)
+static void g4x_program_watermarks(struct intel_display *display)
 {
-	struct g4x_wm_values *old_wm = &dev_priv->display.wm.g4x;
+	struct g4x_wm_values *old_wm = &display->wm.g4x;
 	struct g4x_wm_values new_wm = {};
 
-	g4x_merge_wm(dev_priv, &new_wm);
+	g4x_merge_wm(display, &new_wm);
 
 	if (memcmp(old_wm, &new_wm, sizeof(new_wm)) == 0)
 		return;
 
 	if (is_disabling(old_wm->cxsr, new_wm.cxsr, true))
-		_intel_set_memory_cxsr(dev_priv, false);
+		_intel_set_memory_cxsr(display, false);
 
-	g4x_write_wm_values(dev_priv, &new_wm);
+	g4x_write_wm_values(display, &new_wm);
 
 	if (is_enabling(old_wm->cxsr, new_wm.cxsr, true))
-		_intel_set_memory_cxsr(dev_priv, true);
+		_intel_set_memory_cxsr(display, true);
 
 	*old_wm = new_wm;
 }
@@ -1444,30 +1442,30 @@ static void g4x_program_watermarks(struct drm_i915_private *dev_priv)
 static void g4x_initial_watermarks(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 	crtc->wm.active.g4x = crtc_state->wm.g4x.intermediate;
-	g4x_program_watermarks(dev_priv);
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	g4x_program_watermarks(display);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
 static void g4x_optimize_watermarks(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
 	if (!crtc_state->wm.need_postvbl_update)
 		return;
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 	crtc->wm.active.g4x = crtc_state->wm.g4x.optimal;
-	g4x_program_watermarks(dev_priv);
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	g4x_program_watermarks(display);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
 /* latency must be in 0.1us units. */
@@ -1486,18 +1484,18 @@ static unsigned int vlv_wm_method2(unsigned int pixel_rate,
 	return ret;
 }
 
-static void vlv_setup_wm_latency(struct drm_i915_private *dev_priv)
+static void vlv_setup_wm_latency(struct intel_display *display)
 {
 	/* all latencies in usec */
-	dev_priv->display.wm.pri_latency[VLV_WM_LEVEL_PM2] = 3;
+	display->wm.pri_latency[VLV_WM_LEVEL_PM2] = 3;
 
-	dev_priv->display.wm.num_levels = VLV_WM_LEVEL_PM2 + 1;
+	display->wm.num_levels = VLV_WM_LEVEL_PM2 + 1;
 
-	if (IS_CHERRYVIEW(dev_priv)) {
-		dev_priv->display.wm.pri_latency[VLV_WM_LEVEL_PM5] = 12;
-		dev_priv->display.wm.pri_latency[VLV_WM_LEVEL_DDR_DVFS] = 33;
+	if (display->platform.cherryview) {
+		display->wm.pri_latency[VLV_WM_LEVEL_PM5] = 12;
+		display->wm.pri_latency[VLV_WM_LEVEL_DDR_DVFS] = 33;
 
-		dev_priv->display.wm.num_levels = VLV_WM_LEVEL_DDR_DVFS + 1;
+		display->wm.num_levels = VLV_WM_LEVEL_DDR_DVFS + 1;
 	}
 }
 
@@ -1505,13 +1503,13 @@ static u16 vlv_compute_wm_level(const struct intel_crtc_state *crtc_state,
 				const struct intel_plane_state *plane_state,
 				int level)
 {
+	struct intel_display *display = to_intel_display(plane_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
 	const struct drm_display_mode *pipe_mode =
 		&crtc_state->hw.pipe_mode;
 	unsigned int pixel_rate, htotal, cpp, width, wm;
 
-	if (dev_priv->display.wm.pri_latency[level] == 0)
+	if (display->wm.pri_latency[level] == 0)
 		return USHRT_MAX;
 
 	if (!intel_wm_plane_visible(crtc_state, plane_state))
@@ -1532,7 +1530,7 @@ static u16 vlv_compute_wm_level(const struct intel_crtc_state *crtc_state,
 		wm = 63;
 	} else {
 		wm = vlv_wm_method2(pixel_rate, htotal, width, cpp,
-				    dev_priv->display.wm.pri_latency[level] * 10);
+				    display->wm.pri_latency[level] * 10);
 	}
 
 	return min_t(unsigned int, wm, USHRT_MAX);
@@ -1546,8 +1544,8 @@ static bool vlv_need_sprite0_fifo_workaround(unsigned int active_planes)
 
 static int vlv_compute_fifo(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct g4x_pipe_wm *raw =
 		&crtc_state->wm.vlv.raw[VLV_WM_LEVEL_PM2];
 	struct vlv_fifo_state *fifo_state = &crtc_state->wm.vlv.fifo_state;
@@ -1616,11 +1614,11 @@ static int vlv_compute_fifo(struct intel_crtc_state *crtc_state)
 		fifo_left -= plane_extra;
 	}
 
-	drm_WARN_ON(&dev_priv->drm, active_planes != 0 && fifo_left != 0);
+	drm_WARN_ON(display->drm, active_planes != 0 && fifo_left != 0);
 
 	/* give it all to the first plane if none are active */
 	if (active_planes == 0) {
-		drm_WARN_ON(&dev_priv->drm, fifo_left != fifo_size);
+		drm_WARN_ON(display->drm, fifo_left != fifo_size);
 		fifo_state->plane[PLANE_PRIMARY] = fifo_left;
 	}
 
@@ -1631,9 +1629,9 @@ static int vlv_compute_fifo(struct intel_crtc_state *crtc_state)
 static void vlv_invalidate_wms(struct intel_crtc *crtc,
 			       struct vlv_wm_state *wm_state, int level)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 
-	for (; level < dev_priv->display.wm.num_levels; level++) {
+	for (; level < display->wm.num_levels; level++) {
 		enum plane_id plane_id;
 
 		for_each_plane_id_on_crtc(crtc, plane_id)
@@ -1659,10 +1657,10 @@ static u16 vlv_invert_wm_value(u16 wm, u16 fifo_size)
 static bool vlv_raw_plane_wm_set(struct intel_crtc_state *crtc_state,
 				 int level, enum plane_id plane_id, u16 value)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	bool dirty = false;
 
-	for (; level < dev_priv->display.wm.num_levels; level++) {
+	for (; level < display->wm.num_levels; level++) {
 		struct g4x_pipe_wm *raw = &crtc_state->wm.vlv.raw[level];
 
 		dirty |= raw->plane[plane_id] != value;
@@ -1675,8 +1673,8 @@ static bool vlv_raw_plane_wm_set(struct intel_crtc_state *crtc_state,
 static bool vlv_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 				     const struct intel_plane_state *plane_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	enum plane_id plane_id = plane->id;
 	int level;
 	bool dirty = false;
@@ -1686,7 +1684,7 @@ static bool vlv_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 		goto out;
 	}
 
-	for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+	for (level = 0; level < display->wm.num_levels; level++) {
 		struct g4x_pipe_wm *raw = &crtc_state->wm.vlv.raw[level];
 		int wm = vlv_compute_wm_level(crtc_state, plane_state, level);
 		int max_wm = plane_id == PLANE_CURSOR ? 63 : 511;
@@ -1703,7 +1701,7 @@ static bool vlv_raw_plane_wm_compute(struct intel_crtc_state *crtc_state,
 
 out:
 	if (dirty)
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "%s watermarks: PM2=%d, PM5=%d, DDR DVFS=%d\n",
 			    plane->base.name,
 			    crtc_state->wm.vlv.raw[VLV_WM_LEVEL_PM2].plane[plane_id],
@@ -1734,8 +1732,8 @@ static bool vlv_raw_crtc_wm_is_valid(const struct intel_crtc_state *crtc_state,
 
 static int _vlv_compute_pipe_wm(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct vlv_wm_state *wm_state = &crtc_state->wm.vlv.optimal;
 	const struct vlv_fifo_state *fifo_state =
 		&crtc_state->wm.vlv.fifo_state;
@@ -1745,7 +1743,7 @@ static int _vlv_compute_pipe_wm(struct intel_crtc_state *crtc_state)
 	int level;
 
 	/* initially allow all levels */
-	wm_state->num_levels = dev_priv->display.wm.num_levels;
+	wm_state->num_levels = display->wm.num_levels;
 	/*
 	 * Note that enabling cxsr with no primary/sprite planes
 	 * enabled can wedge the pipe. Hence we only allow cxsr
@@ -1755,7 +1753,7 @@ static int _vlv_compute_pipe_wm(struct intel_crtc_state *crtc_state)
 
 	for (level = 0; level < wm_state->num_levels; level++) {
 		const struct g4x_pipe_wm *raw = &crtc_state->wm.vlv.raw[level];
-		const int sr_fifo_size = INTEL_NUM_PIPES(dev_priv) * 512 - 1;
+		const int sr_fifo_size = INTEL_NUM_PIPES(display) * 512 - 1;
 
 		if (!vlv_raw_crtc_wm_is_valid(crtc_state, level))
 			break;
@@ -1855,6 +1853,7 @@ static int vlv_compute_pipe_wm(struct intel_atomic_state *state,
 static void vlv_atomic_update_fifo(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
+	struct intel_display *display = to_intel_display(crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_uncore *uncore = &dev_priv->uncore;
 	const struct intel_crtc_state *crtc_state =
@@ -1871,8 +1870,8 @@ static void vlv_atomic_update_fifo(struct intel_atomic_state *state,
 	sprite1_start = fifo_state->plane[PLANE_SPRITE0] + sprite0_start;
 	fifo_size = fifo_state->plane[PLANE_SPRITE1] + sprite1_start;
 
-	drm_WARN_ON(&dev_priv->drm, fifo_state->plane[PLANE_CURSOR] != 63);
-	drm_WARN_ON(&dev_priv->drm, fifo_size != 511);
+	drm_WARN_ON(display->drm, fifo_state->plane[PLANE_CURSOR] != 63);
+	drm_WARN_ON(display->drm, fifo_size != 511);
 
 	trace_vlv_fifo_size(crtc, sprite0_start, sprite1_start, fifo_size);
 
@@ -1889,8 +1888,8 @@ static void vlv_atomic_update_fifo(struct intel_atomic_state *state,
 
 	switch (crtc->pipe) {
 	case PIPE_A:
-		dsparb = intel_uncore_read_fw(uncore, DSPARB(dev_priv));
-		dsparb2 = intel_uncore_read_fw(uncore, DSPARB2);
+		dsparb = intel_de_read_fw(display, DSPARB(display));
+		dsparb2 = intel_de_read_fw(display, DSPARB2);
 
 		dsparb &= ~(VLV_FIFO(SPRITEA, 0xff) |
 			    VLV_FIFO(SPRITEB, 0xff));
@@ -1902,12 +1901,12 @@ static void vlv_atomic_update_fifo(struct intel_atomic_state *state,
 		dsparb2 |= (VLV_FIFO(SPRITEA_HI, sprite0_start >> 8) |
 			   VLV_FIFO(SPRITEB_HI, sprite1_start >> 8));
 
-		intel_uncore_write_fw(uncore, DSPARB(dev_priv), dsparb);
-		intel_uncore_write_fw(uncore, DSPARB2, dsparb2);
+		intel_de_write_fw(display, DSPARB(display), dsparb);
+		intel_de_write_fw(display, DSPARB2, dsparb2);
 		break;
 	case PIPE_B:
-		dsparb = intel_uncore_read_fw(uncore, DSPARB(dev_priv));
-		dsparb2 = intel_uncore_read_fw(uncore, DSPARB2);
+		dsparb = intel_de_read_fw(display, DSPARB(display));
+		dsparb2 = intel_de_read_fw(display, DSPARB2);
 
 		dsparb &= ~(VLV_FIFO(SPRITEC, 0xff) |
 			    VLV_FIFO(SPRITED, 0xff));
@@ -1919,12 +1918,12 @@ static void vlv_atomic_update_fifo(struct intel_atomic_state *state,
 		dsparb2 |= (VLV_FIFO(SPRITEC_HI, sprite0_start >> 8) |
 			   VLV_FIFO(SPRITED_HI, sprite1_start >> 8));
 
-		intel_uncore_write_fw(uncore, DSPARB(dev_priv), dsparb);
-		intel_uncore_write_fw(uncore, DSPARB2, dsparb2);
+		intel_de_write_fw(display, DSPARB(display), dsparb);
+		intel_de_write_fw(display, DSPARB2, dsparb2);
 		break;
 	case PIPE_C:
-		dsparb3 = intel_uncore_read_fw(uncore, DSPARB3);
-		dsparb2 = intel_uncore_read_fw(uncore, DSPARB2);
+		dsparb3 = intel_de_read_fw(display, DSPARB3);
+		dsparb2 = intel_de_read_fw(display, DSPARB2);
 
 		dsparb3 &= ~(VLV_FIFO(SPRITEE, 0xff) |
 			     VLV_FIFO(SPRITEF, 0xff));
@@ -1936,14 +1935,14 @@ static void vlv_atomic_update_fifo(struct intel_atomic_state *state,
 		dsparb2 |= (VLV_FIFO(SPRITEE_HI, sprite0_start >> 8) |
 			   VLV_FIFO(SPRITEF_HI, sprite1_start >> 8));
 
-		intel_uncore_write_fw(uncore, DSPARB3, dsparb3);
-		intel_uncore_write_fw(uncore, DSPARB2, dsparb2);
+		intel_de_write_fw(display, DSPARB3, dsparb3);
+		intel_de_write_fw(display, DSPARB2, dsparb2);
 		break;
 	default:
 		break;
 	}
 
-	intel_uncore_posting_read_fw(uncore, DSPARB(dev_priv));
+	intel_de_read_fw(display, DSPARB(display));
 
 	spin_unlock(&uncore->lock);
 }
@@ -2018,16 +2017,16 @@ static int vlv_compute_watermarks(struct intel_atomic_state *state,
 	return 0;
 }
 
-static void vlv_merge_wm(struct drm_i915_private *dev_priv,
+static void vlv_merge_wm(struct intel_display *display,
 			 struct vlv_wm_values *wm)
 {
 	struct intel_crtc *crtc;
 	int num_active_pipes = 0;
 
-	wm->level = dev_priv->display.wm.num_levels - 1;
+	wm->level = display->wm.num_levels - 1;
 	wm->cxsr = true;
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct vlv_wm_state *wm_state = &crtc->wm.active.vlv;
 
 		if (!crtc->active)
@@ -2046,7 +2045,7 @@ static void vlv_merge_wm(struct drm_i915_private *dev_priv,
 	if (num_active_pipes > 1)
 		wm->level = VLV_WM_LEVEL_PM2;
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct vlv_wm_state *wm_state = &crtc->wm.active.vlv;
 		enum pipe pipe = crtc->pipe;
 
@@ -2061,35 +2060,35 @@ static void vlv_merge_wm(struct drm_i915_private *dev_priv,
 	}
 }
 
-static void vlv_program_watermarks(struct drm_i915_private *dev_priv)
+static void vlv_program_watermarks(struct intel_display *display)
 {
-	struct vlv_wm_values *old_wm = &dev_priv->display.wm.vlv;
+	struct vlv_wm_values *old_wm = &display->wm.vlv;
 	struct vlv_wm_values new_wm = {};
 
-	vlv_merge_wm(dev_priv, &new_wm);
+	vlv_merge_wm(display, &new_wm);
 
 	if (memcmp(old_wm, &new_wm, sizeof(new_wm)) == 0)
 		return;
 
 	if (is_disabling(old_wm->level, new_wm.level, VLV_WM_LEVEL_DDR_DVFS))
-		chv_set_memory_dvfs(dev_priv, false);
+		chv_set_memory_dvfs(display, false);
 
 	if (is_disabling(old_wm->level, new_wm.level, VLV_WM_LEVEL_PM5))
-		chv_set_memory_pm5(dev_priv, false);
+		chv_set_memory_pm5(display, false);
 
 	if (is_disabling(old_wm->cxsr, new_wm.cxsr, true))
-		_intel_set_memory_cxsr(dev_priv, false);
+		_intel_set_memory_cxsr(display, false);
 
-	vlv_write_wm_values(dev_priv, &new_wm);
+	vlv_write_wm_values(display, &new_wm);
 
 	if (is_enabling(old_wm->cxsr, new_wm.cxsr, true))
-		_intel_set_memory_cxsr(dev_priv, true);
+		_intel_set_memory_cxsr(display, true);
 
 	if (is_enabling(old_wm->level, new_wm.level, VLV_WM_LEVEL_PM5))
-		chv_set_memory_pm5(dev_priv, true);
+		chv_set_memory_pm5(display, true);
 
 	if (is_enabling(old_wm->level, new_wm.level, VLV_WM_LEVEL_DDR_DVFS))
-		chv_set_memory_dvfs(dev_priv, true);
+		chv_set_memory_dvfs(display, true);
 
 	*old_wm = new_wm;
 }
@@ -2097,33 +2096,33 @@ static void vlv_program_watermarks(struct drm_i915_private *dev_priv)
 static void vlv_initial_watermarks(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 	crtc->wm.active.vlv = crtc_state->wm.vlv.intermediate;
-	vlv_program_watermarks(dev_priv);
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	vlv_program_watermarks(display);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
 static void vlv_optimize_watermarks(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
 	if (!crtc_state->wm.need_postvbl_update)
 		return;
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 	crtc->wm.active.vlv = crtc_state->wm.vlv.optimal;
-	vlv_program_watermarks(dev_priv);
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	vlv_program_watermarks(display);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
-static void i965_update_wm(struct drm_i915_private *dev_priv)
+static void i965_update_wm(struct intel_display *display)
 {
 	struct intel_crtc *crtc;
 	int srwm = 1;
@@ -2131,7 +2130,7 @@ static void i965_update_wm(struct drm_i915_private *dev_priv)
 	bool cxsr_enabled;
 
 	/* Calc sr entries for one plane configs */
-	crtc = single_enabled_crtc(dev_priv);
+	crtc = single_enabled_crtc(display);
 	if (crtc) {
 		/* self-refresh has much higher latency */
 		static const int sr_latency_ns = 12000;
@@ -2152,7 +2151,7 @@ static void i965_update_wm(struct drm_i915_private *dev_priv)
 		if (srwm < 0)
 			srwm = 1;
 		srwm &= 0x1ff;
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "self-refresh entries: %d, wm: %d\n",
 			    entries, srwm);
 
@@ -2167,7 +2166,7 @@ static void i965_update_wm(struct drm_i915_private *dev_priv)
 		if (cursor_sr > i965_cursor_wm_info.max_wm)
 			cursor_sr = i965_cursor_wm_info.max_wm;
 
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "self-refresh watermark: display plane %d "
 			    "cursor %d\n", srwm, cursor_sr);
 
@@ -2175,39 +2174,38 @@ static void i965_update_wm(struct drm_i915_private *dev_priv)
 	} else {
 		cxsr_enabled = false;
 		/* Turn off self refresh if both pipes are enabled */
-		intel_set_memory_cxsr(dev_priv, false);
+		intel_set_memory_cxsr(display, false);
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Setting FIFO watermarks - A: 8, B: 8, C: 8, SR %d\n",
 		    srwm);
 
 	/* 965 has limitations... */
-	intel_uncore_write(&dev_priv->uncore, DSPFW1(dev_priv),
-			   FW_WM(srwm, SR) |
-			   FW_WM(8, CURSORB) |
-			   FW_WM(8, PLANEB) |
-			   FW_WM(8, PLANEA));
-	intel_uncore_write(&dev_priv->uncore, DSPFW2(dev_priv),
-			   FW_WM(8, CURSORA) |
-			   FW_WM(8, PLANEC_OLD));
+	intel_de_write(display, DSPFW1(display),
+		       FW_WM(srwm, SR) |
+		       FW_WM(8, CURSORB) |
+		       FW_WM(8, PLANEB) |
+		       FW_WM(8, PLANEA));
+	intel_de_write(display, DSPFW2(display),
+		       FW_WM(8, CURSORA) |
+		       FW_WM(8, PLANEC_OLD));
 	/* update cursor SR watermark */
-	intel_uncore_write(&dev_priv->uncore, DSPFW3(dev_priv),
-			   FW_WM(cursor_sr, CURSOR_SR));
+	intel_de_write(display, DSPFW3(display),
+		       FW_WM(cursor_sr, CURSOR_SR));
 
 	if (cxsr_enabled)
-		intel_set_memory_cxsr(dev_priv, true);
+		intel_set_memory_cxsr(display, true);
 }
 
 #undef FW_WM
 
-static struct intel_crtc *intel_crtc_for_plane(struct drm_i915_private *i915,
+static struct intel_crtc *intel_crtc_for_plane(struct intel_display *display,
 					       enum i9xx_plane_id i9xx_plane)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_plane *plane;
 
-	for_each_intel_plane(&i915->drm, plane) {
+	for_each_intel_plane(display->drm, plane) {
 		if (plane->id == PLANE_PRIMARY &&
 		    plane->i9xx_plane == i9xx_plane)
 			return intel_crtc_for_pipe(display, plane->pipe);
@@ -2216,7 +2214,7 @@ static struct intel_crtc *intel_crtc_for_plane(struct drm_i915_private *i915,
 	return NULL;
 }
 
-static void i9xx_update_wm(struct drm_i915_private *dev_priv)
+static void i9xx_update_wm(struct intel_display *display)
 {
 	const struct intel_watermark_params *wm_info;
 	u32 fwater_lo;
@@ -2226,29 +2224,29 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 	int planea_wm, planeb_wm;
 	struct intel_crtc *crtc;
 
-	if (IS_I945GM(dev_priv))
+	if (display->platform.i945gm)
 		wm_info = &i945_wm_info;
-	else if (DISPLAY_VER(dev_priv) != 2)
+	else if (DISPLAY_VER(display) != 2)
 		wm_info = &i915_wm_info;
 	else
 		wm_info = &i830_a_wm_info;
 
-	if (DISPLAY_VER(dev_priv) == 2)
-		fifo_size = i830_get_fifo_size(dev_priv, PLANE_A);
+	if (DISPLAY_VER(display) == 2)
+		fifo_size = i830_get_fifo_size(display, PLANE_A);
 	else
-		fifo_size = i9xx_get_fifo_size(dev_priv, PLANE_A);
-	crtc = intel_crtc_for_plane(dev_priv, PLANE_A);
+		fifo_size = i9xx_get_fifo_size(display, PLANE_A);
+	crtc = intel_crtc_for_plane(display, PLANE_A);
 	if (intel_crtc_active(crtc)) {
 		const struct drm_framebuffer *fb =
 			crtc->base.primary->state->fb;
 		int cpp;
 
-		if (DISPLAY_VER(dev_priv) == 2)
+		if (DISPLAY_VER(display) == 2)
 			cpp = 4;
 		else
 			cpp = fb->format->cpp[0];
 
-		planea_wm = intel_calculate_wm(dev_priv, crtc->config->pixel_rate,
+		planea_wm = intel_calculate_wm(display, crtc->config->pixel_rate,
 					       wm_info, fifo_size, cpp,
 					       pessimal_latency_ns);
 	} else {
@@ -2257,25 +2255,25 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 			planea_wm = wm_info->max_wm;
 	}
 
-	if (DISPLAY_VER(dev_priv) == 2)
+	if (DISPLAY_VER(display) == 2)
 		wm_info = &i830_bc_wm_info;
 
-	if (DISPLAY_VER(dev_priv) == 2)
-		fifo_size = i830_get_fifo_size(dev_priv, PLANE_B);
+	if (DISPLAY_VER(display) == 2)
+		fifo_size = i830_get_fifo_size(display, PLANE_B);
 	else
-		fifo_size = i9xx_get_fifo_size(dev_priv, PLANE_B);
-	crtc = intel_crtc_for_plane(dev_priv, PLANE_B);
+		fifo_size = i9xx_get_fifo_size(display, PLANE_B);
+	crtc = intel_crtc_for_plane(display, PLANE_B);
 	if (intel_crtc_active(crtc)) {
 		const struct drm_framebuffer *fb =
 			crtc->base.primary->state->fb;
 		int cpp;
 
-		if (DISPLAY_VER(dev_priv) == 2)
+		if (DISPLAY_VER(display) == 2)
 			cpp = 4;
 		else
 			cpp = fb->format->cpp[0];
 
-		planeb_wm = intel_calculate_wm(dev_priv, crtc->config->pixel_rate,
+		planeb_wm = intel_calculate_wm(display, crtc->config->pixel_rate,
 					       wm_info, fifo_size, cpp,
 					       pessimal_latency_ns);
 	} else {
@@ -2284,11 +2282,11 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 			planeb_wm = wm_info->max_wm;
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "FIFO watermarks - A: %d, B: %d\n", planea_wm, planeb_wm);
 
-	crtc = single_enabled_crtc(dev_priv);
-	if (IS_I915GM(dev_priv) && crtc) {
+	crtc = single_enabled_crtc(display);
+	if (display->platform.i915gm && crtc) {
 		struct drm_gem_object *obj;
 
 		obj = intel_fb_bo(crtc->base.primary->state->fb);
@@ -2304,10 +2302,10 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 	cwm = 2;
 
 	/* Play safe and disable self-refresh before adjusting watermarks. */
-	intel_set_memory_cxsr(dev_priv, false);
+	intel_set_memory_cxsr(display, false);
 
 	/* Calc sr entries for one plane configs */
-	if (HAS_FW_BLC(dev_priv) && crtc) {
+	if (HAS_FW_BLC(display) && crtc) {
 		/* self-refresh has much higher latency */
 		static const int sr_latency_ns = 6000;
 		const struct drm_display_mode *pipe_mode =
@@ -2320,7 +2318,7 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 		int cpp;
 		int entries;
 
-		if (IS_I915GM(dev_priv) || IS_I945GM(dev_priv))
+		if (display->platform.i915gm || display->platform.i945gm)
 			cpp = 4;
 		else
 			cpp = fb->format->cpp[0];
@@ -2328,20 +2326,20 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 		entries = intel_wm_method2(pixel_rate, htotal, width, cpp,
 					   sr_latency_ns / 100);
 		entries = DIV_ROUND_UP(entries, wm_info->cacheline_size);
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "self-refresh entries: %d\n", entries);
 		srwm = wm_info->fifo_size - entries;
 		if (srwm < 0)
 			srwm = 1;
 
-		if (IS_I945G(dev_priv) || IS_I945GM(dev_priv))
-			intel_uncore_write(&dev_priv->uncore, FW_BLC_SELF,
-				   FW_BLC_SELF_FIFO_MASK | (srwm & 0xff));
+		if (display->platform.i945g || display->platform.i945gm)
+			intel_de_write(display, FW_BLC_SELF,
+				       FW_BLC_SELF_FIFO_MASK | (srwm & 0xff));
 		else
-			intel_uncore_write(&dev_priv->uncore, FW_BLC_SELF, srwm & 0x3f);
+			intel_de_write(display, FW_BLC_SELF, srwm & 0x3f);
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Setting FIFO watermarks - A: %d, B: %d, C: %d, SR %d\n",
 		     planea_wm, planeb_wm, cwm, srwm);
 
@@ -2352,34 +2350,34 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv)
 	fwater_lo = fwater_lo | (1 << 24) | (1 << 8);
 	fwater_hi = fwater_hi | (1 << 8);
 
-	intel_uncore_write(&dev_priv->uncore, FW_BLC, fwater_lo);
-	intel_uncore_write(&dev_priv->uncore, FW_BLC2, fwater_hi);
+	intel_de_write(display, FW_BLC, fwater_lo);
+	intel_de_write(display, FW_BLC2, fwater_hi);
 
 	if (crtc)
-		intel_set_memory_cxsr(dev_priv, true);
+		intel_set_memory_cxsr(display, true);
 }
 
-static void i845_update_wm(struct drm_i915_private *dev_priv)
+static void i845_update_wm(struct intel_display *display)
 {
 	struct intel_crtc *crtc;
 	u32 fwater_lo;
 	int planea_wm;
 
-	crtc = single_enabled_crtc(dev_priv);
+	crtc = single_enabled_crtc(display);
 	if (crtc == NULL)
 		return;
 
-	planea_wm = intel_calculate_wm(dev_priv, crtc->config->pixel_rate,
+	planea_wm = intel_calculate_wm(display, crtc->config->pixel_rate,
 				       &i845_wm_info,
-				       i845_get_fifo_size(dev_priv, PLANE_A),
+				       i845_get_fifo_size(display, PLANE_A),
 				       4, pessimal_latency_ns);
-	fwater_lo = intel_uncore_read(&dev_priv->uncore, FW_BLC) & ~0xfff;
+	fwater_lo = intel_de_read(display, FW_BLC) & ~0xfff;
 	fwater_lo |= (3<<8) | planea_wm;
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Setting FIFO watermarks - A: %d\n", planea_wm);
 
-	intel_uncore_write(&dev_priv->uncore, FW_BLC, fwater_lo);
+	intel_de_write(display, FW_BLC, fwater_lo);
 }
 
 /* latency must be in 0.1us units. */
@@ -2534,24 +2532,24 @@ static u32 ilk_compute_fbc_wm(const struct intel_crtc_state *crtc_state,
 }
 
 static unsigned int
-ilk_display_fifo_size(const struct drm_i915_private *dev_priv)
+ilk_display_fifo_size(struct intel_display *display)
 {
-	if (DISPLAY_VER(dev_priv) >= 8)
+	if (DISPLAY_VER(display) >= 8)
 		return 3072;
-	else if (DISPLAY_VER(dev_priv) >= 7)
+	else if (DISPLAY_VER(display) >= 7)
 		return 768;
 	else
 		return 512;
 }
 
 static unsigned int
-ilk_plane_wm_reg_max(const struct drm_i915_private *dev_priv,
+ilk_plane_wm_reg_max(struct intel_display *display,
 		     int level, bool is_sprite)
 {
-	if (DISPLAY_VER(dev_priv) >= 8)
+	if (DISPLAY_VER(display) >= 8)
 		/* BDW primary/sprite plane watermarks */
 		return level == 0 ? 255 : 2047;
-	else if (DISPLAY_VER(dev_priv) >= 7)
+	else if (DISPLAY_VER(display) >= 7)
 		/* IVB/HSW primary/sprite plane watermarks */
 		return level == 0 ? 127 : 1023;
 	else if (!is_sprite)
@@ -2563,30 +2561,30 @@ ilk_plane_wm_reg_max(const struct drm_i915_private *dev_priv,
 }
 
 static unsigned int
-ilk_cursor_wm_reg_max(const struct drm_i915_private *dev_priv, int level)
+ilk_cursor_wm_reg_max(struct intel_display *display, int level)
 {
-	if (DISPLAY_VER(dev_priv) >= 7)
+	if (DISPLAY_VER(display) >= 7)
 		return level == 0 ? 63 : 255;
 	else
 		return level == 0 ? 31 : 63;
 }
 
-static unsigned int ilk_fbc_wm_reg_max(const struct drm_i915_private *dev_priv)
+static unsigned int ilk_fbc_wm_reg_max(struct intel_display *display)
 {
-	if (DISPLAY_VER(dev_priv) >= 8)
+	if (DISPLAY_VER(display) >= 8)
 		return 31;
 	else
 		return 15;
 }
 
 /* Calculate the maximum primary/sprite plane watermark */
-static unsigned int ilk_plane_wm_max(const struct drm_i915_private *dev_priv,
+static unsigned int ilk_plane_wm_max(struct intel_display *display,
 				     int level,
 				     const struct intel_wm_config *config,
 				     enum intel_ddb_partitioning ddb_partitioning,
 				     bool is_sprite)
 {
-	unsigned int fifo_size = ilk_display_fifo_size(dev_priv);
+	unsigned int fifo_size = ilk_display_fifo_size(display);
 
 	/* if sprites aren't enabled, sprites get nothing */
 	if (is_sprite && !config->sprites_enabled)
@@ -2594,14 +2592,14 @@ static unsigned int ilk_plane_wm_max(const struct drm_i915_private *dev_priv,
 
 	/* HSW allows LP1+ watermarks even with multiple pipes */
 	if (level == 0 || config->num_pipes_active > 1) {
-		fifo_size /= INTEL_NUM_PIPES(dev_priv);
+		fifo_size /= INTEL_NUM_PIPES(display);
 
 		/*
 		 * For some reason the non self refresh
 		 * FIFO size is only half of the self
 		 * refresh FIFO size on ILK/SNB.
 		 */
-		if (DISPLAY_VER(dev_priv) < 7)
+		if (DISPLAY_VER(display) < 7)
 			fifo_size /= 2;
 	}
 
@@ -2617,11 +2615,11 @@ static unsigned int ilk_plane_wm_max(const struct drm_i915_private *dev_priv,
 	}
 
 	/* clamp to max that the registers can hold */
-	return min(fifo_size, ilk_plane_wm_reg_max(dev_priv, level, is_sprite));
+	return min(fifo_size, ilk_plane_wm_reg_max(display, level, is_sprite));
 }
 
 /* Calculate the maximum cursor plane watermark */
-static unsigned int ilk_cursor_wm_max(const struct drm_i915_private *dev_priv,
+static unsigned int ilk_cursor_wm_max(struct intel_display *display,
 				      int level,
 				      const struct intel_wm_config *config)
 {
@@ -2630,32 +2628,32 @@ static unsigned int ilk_cursor_wm_max(const struct drm_i915_private *dev_priv,
 		return 64;
 
 	/* otherwise just report max that registers can hold */
-	return ilk_cursor_wm_reg_max(dev_priv, level);
+	return ilk_cursor_wm_reg_max(display, level);
 }
 
-static void ilk_compute_wm_maximums(const struct drm_i915_private *dev_priv,
+static void ilk_compute_wm_maximums(struct intel_display *display,
 				    int level,
 				    const struct intel_wm_config *config,
 				    enum intel_ddb_partitioning ddb_partitioning,
 				    struct ilk_wm_maximums *max)
 {
-	max->pri = ilk_plane_wm_max(dev_priv, level, config, ddb_partitioning, false);
-	max->spr = ilk_plane_wm_max(dev_priv, level, config, ddb_partitioning, true);
-	max->cur = ilk_cursor_wm_max(dev_priv, level, config);
-	max->fbc = ilk_fbc_wm_reg_max(dev_priv);
+	max->pri = ilk_plane_wm_max(display, level, config, ddb_partitioning, false);
+	max->spr = ilk_plane_wm_max(display, level, config, ddb_partitioning, true);
+	max->cur = ilk_cursor_wm_max(display, level, config);
+	max->fbc = ilk_fbc_wm_reg_max(display);
 }
 
-static void ilk_compute_wm_reg_maximums(const struct drm_i915_private *dev_priv,
+static void ilk_compute_wm_reg_maximums(struct intel_display *display,
 					int level,
 					struct ilk_wm_maximums *max)
 {
-	max->pri = ilk_plane_wm_reg_max(dev_priv, level, false);
-	max->spr = ilk_plane_wm_reg_max(dev_priv, level, true);
-	max->cur = ilk_cursor_wm_reg_max(dev_priv, level);
-	max->fbc = ilk_fbc_wm_reg_max(dev_priv);
+	max->pri = ilk_plane_wm_reg_max(display, level, false);
+	max->spr = ilk_plane_wm_reg_max(display, level, true);
+	max->cur = ilk_cursor_wm_reg_max(display, level);
+	max->fbc = ilk_fbc_wm_reg_max(display);
 }
 
-static bool ilk_validate_wm_level(struct drm_i915_private *i915,
+static bool ilk_validate_wm_level(struct intel_display *display,
 				  int level,
 				  const struct ilk_wm_maximums *max,
 				  struct intel_wm_level *result)
@@ -2679,15 +2677,15 @@ static bool ilk_validate_wm_level(struct drm_i915_private *i915,
 	 */
 	if (level == 0 && !result->enable) {
 		if (result->pri_val > max->pri)
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "Primary WM%d too large %u (max %u)\n",
 				    level, result->pri_val, max->pri);
 		if (result->spr_val > max->spr)
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "Sprite WM%d too large %u (max %u)\n",
 				    level, result->spr_val, max->spr);
 		if (result->cur_val > max->cur)
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "Cursor WM%d too large %u (max %u)\n",
 				    level, result->cur_val, max->cur);
 
@@ -2700,7 +2698,7 @@ static bool ilk_validate_wm_level(struct drm_i915_private *i915,
 	return ret;
 }
 
-static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
+static void ilk_compute_wm_level(struct intel_display *display,
 				 const struct intel_crtc *crtc,
 				 int level,
 				 struct intel_crtc_state *crtc_state,
@@ -2709,9 +2707,9 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 				 const struct intel_plane_state *curstate,
 				 struct intel_wm_level *result)
 {
-	u16 pri_latency = dev_priv->display.wm.pri_latency[level];
-	u16 spr_latency = dev_priv->display.wm.spr_latency[level];
-	u16 cur_latency = dev_priv->display.wm.cur_latency[level];
+	u16 pri_latency = display->wm.pri_latency[level];
+	u16 spr_latency = display->wm.spr_latency[level];
+	u16 cur_latency = display->wm.cur_latency[level];
 
 	/* WM1+ latency values stored in 0.5us units */
 	if (level > 0) {
@@ -2735,11 +2733,12 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 	result->enable = true;
 }
 
-static void hsw_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
+static void hsw_read_wm_latency(struct intel_display *display, u16 wm[])
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u64 sskpd;
 
-	i915->display.wm.num_levels = 5;
+	display->wm.num_levels = 5;
 
 	sskpd = intel_uncore_read64(&i915->uncore, MCH_SSKPD);
 
@@ -2752,11 +2751,12 @@ static void hsw_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
 	wm[4] = REG_FIELD_GET64(SSKPD_WM4_MASK_HSW, sskpd);
 }
 
-static void snb_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
+static void snb_read_wm_latency(struct intel_display *display, u16 wm[])
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 sskpd;
 
-	i915->display.wm.num_levels = 4;
+	display->wm.num_levels = 4;
 
 	sskpd = intel_uncore_read(&i915->uncore, MCH_SSKPD);
 
@@ -2766,11 +2766,12 @@ static void snb_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
 	wm[3] = REG_FIELD_GET(SSKPD_WM3_MASK_SNB, sskpd);
 }
 
-static void ilk_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
+static void ilk_read_wm_latency(struct intel_display *display, u16 wm[])
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 mltr;
 
-	i915->display.wm.num_levels = 3;
+	display->wm.num_levels = 3;
 
 	mltr = intel_uncore_read(&i915->uncore, MLTR_ILK);
 
@@ -2780,24 +2781,21 @@ static void ilk_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
 	wm[2] = REG_FIELD_GET(MLTR_WM2_MASK, mltr);
 }
 
-static void intel_fixup_spr_wm_latency(struct drm_i915_private *dev_priv,
-				       u16 wm[5])
+static void intel_fixup_spr_wm_latency(struct intel_display *display, u16 wm[5])
 {
 	/* ILK sprite LP0 latency is 1300 ns */
-	if (DISPLAY_VER(dev_priv) == 5)
+	if (DISPLAY_VER(display) == 5)
 		wm[0] = 13;
 }
 
-static void intel_fixup_cur_wm_latency(struct drm_i915_private *dev_priv,
-				       u16 wm[5])
+static void intel_fixup_cur_wm_latency(struct intel_display *display, u16 wm[5])
 {
 	/* ILK cursor LP0 latency is 1300 ns */
-	if (DISPLAY_VER(dev_priv) == 5)
+	if (DISPLAY_VER(display) == 5)
 		wm[0] = 13;
 }
 
-static bool ilk_increase_wm_latency(struct drm_i915_private *dev_priv,
-				    u16 wm[5], u16 min)
+static bool ilk_increase_wm_latency(struct intel_display *display, u16 wm[5], u16 min)
 {
 	int level;
 
@@ -2805,13 +2803,13 @@ static bool ilk_increase_wm_latency(struct drm_i915_private *dev_priv,
 		return false;
 
 	wm[0] = max(wm[0], min);
-	for (level = 1; level < dev_priv->display.wm.num_levels; level++)
+	for (level = 1; level < display->wm.num_levels; level++)
 		wm[level] = max_t(u16, wm[level], DIV_ROUND_UP(min, 5));
 
 	return true;
 }
 
-static void snb_wm_latency_quirk(struct drm_i915_private *dev_priv)
+static void snb_wm_latency_quirk(struct intel_display *display)
 {
 	bool changed;
 
@@ -2819,21 +2817,21 @@ static void snb_wm_latency_quirk(struct drm_i915_private *dev_priv)
 	 * The BIOS provided WM memory latency values are often
 	 * inadequate for high resolution displays. Adjust them.
 	 */
-	changed = ilk_increase_wm_latency(dev_priv, dev_priv->display.wm.pri_latency, 12);
-	changed |= ilk_increase_wm_latency(dev_priv, dev_priv->display.wm.spr_latency, 12);
-	changed |= ilk_increase_wm_latency(dev_priv, dev_priv->display.wm.cur_latency, 12);
+	changed = ilk_increase_wm_latency(display, display->wm.pri_latency, 12);
+	changed |= ilk_increase_wm_latency(display, display->wm.spr_latency, 12);
+	changed |= ilk_increase_wm_latency(display, display->wm.cur_latency, 12);
 
 	if (!changed)
 		return;
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "WM latency values increased to avoid potential underruns\n");
-	intel_print_wm_latency(dev_priv, "Primary", dev_priv->display.wm.pri_latency);
-	intel_print_wm_latency(dev_priv, "Sprite", dev_priv->display.wm.spr_latency);
-	intel_print_wm_latency(dev_priv, "Cursor", dev_priv->display.wm.cur_latency);
+	intel_print_wm_latency(display, "Primary", display->wm.pri_latency);
+	intel_print_wm_latency(display, "Sprite", display->wm.spr_latency);
+	intel_print_wm_latency(display, "Cursor", display->wm.cur_latency);
 }
 
-static void snb_wm_lp3_irq_quirk(struct drm_i915_private *dev_priv)
+static void snb_wm_lp3_irq_quirk(struct intel_display *display)
 {
 	/*
 	 * On some SNB machines (Thinkpad X220 Tablet at least)
@@ -2846,50 +2844,50 @@ static void snb_wm_lp3_irq_quirk(struct drm_i915_private *dev_priv)
 	 * interrupts only. To play it safe we disable LP3
 	 * watermarks entirely.
 	 */
-	if (dev_priv->display.wm.pri_latency[3] == 0 &&
-	    dev_priv->display.wm.spr_latency[3] == 0 &&
-	    dev_priv->display.wm.cur_latency[3] == 0)
+	if (display->wm.pri_latency[3] == 0 &&
+	    display->wm.spr_latency[3] == 0 &&
+	    display->wm.cur_latency[3] == 0)
 		return;
 
-	dev_priv->display.wm.pri_latency[3] = 0;
-	dev_priv->display.wm.spr_latency[3] = 0;
-	dev_priv->display.wm.cur_latency[3] = 0;
+	display->wm.pri_latency[3] = 0;
+	display->wm.spr_latency[3] = 0;
+	display->wm.cur_latency[3] = 0;
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "LP3 watermarks disabled due to potential for lost interrupts\n");
-	intel_print_wm_latency(dev_priv, "Primary", dev_priv->display.wm.pri_latency);
-	intel_print_wm_latency(dev_priv, "Sprite", dev_priv->display.wm.spr_latency);
-	intel_print_wm_latency(dev_priv, "Cursor", dev_priv->display.wm.cur_latency);
+	intel_print_wm_latency(display, "Primary", display->wm.pri_latency);
+	intel_print_wm_latency(display, "Sprite", display->wm.spr_latency);
+	intel_print_wm_latency(display, "Cursor", display->wm.cur_latency);
 }
 
-static void ilk_setup_wm_latency(struct drm_i915_private *dev_priv)
+static void ilk_setup_wm_latency(struct intel_display *display)
 {
-	if (IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv))
-		hsw_read_wm_latency(dev_priv, dev_priv->display.wm.pri_latency);
-	else if (DISPLAY_VER(dev_priv) >= 6)
-		snb_read_wm_latency(dev_priv, dev_priv->display.wm.pri_latency);
+	if (display->platform.broadwell || display->platform.haswell)
+		hsw_read_wm_latency(display, display->wm.pri_latency);
+	else if (DISPLAY_VER(display) >= 6)
+		snb_read_wm_latency(display, display->wm.pri_latency);
 	else
-		ilk_read_wm_latency(dev_priv, dev_priv->display.wm.pri_latency);
+		ilk_read_wm_latency(display, display->wm.pri_latency);
 
-	memcpy(dev_priv->display.wm.spr_latency, dev_priv->display.wm.pri_latency,
-	       sizeof(dev_priv->display.wm.pri_latency));
-	memcpy(dev_priv->display.wm.cur_latency, dev_priv->display.wm.pri_latency,
-	       sizeof(dev_priv->display.wm.pri_latency));
+	memcpy(display->wm.spr_latency, display->wm.pri_latency,
+	       sizeof(display->wm.pri_latency));
+	memcpy(display->wm.cur_latency, display->wm.pri_latency,
+	       sizeof(display->wm.pri_latency));
 
-	intel_fixup_spr_wm_latency(dev_priv, dev_priv->display.wm.spr_latency);
-	intel_fixup_cur_wm_latency(dev_priv, dev_priv->display.wm.cur_latency);
+	intel_fixup_spr_wm_latency(display, display->wm.spr_latency);
+	intel_fixup_cur_wm_latency(display, display->wm.cur_latency);
 
-	intel_print_wm_latency(dev_priv, "Primary", dev_priv->display.wm.pri_latency);
-	intel_print_wm_latency(dev_priv, "Sprite", dev_priv->display.wm.spr_latency);
-	intel_print_wm_latency(dev_priv, "Cursor", dev_priv->display.wm.cur_latency);
+	intel_print_wm_latency(display, "Primary", display->wm.pri_latency);
+	intel_print_wm_latency(display, "Sprite", display->wm.spr_latency);
+	intel_print_wm_latency(display, "Cursor", display->wm.cur_latency);
 
-	if (DISPLAY_VER(dev_priv) == 6) {
-		snb_wm_latency_quirk(dev_priv);
-		snb_wm_lp3_irq_quirk(dev_priv);
+	if (DISPLAY_VER(display) == 6) {
+		snb_wm_latency_quirk(display);
+		snb_wm_lp3_irq_quirk(display);
 	}
 }
 
-static bool ilk_validate_pipe_wm(struct drm_i915_private *dev_priv,
+static bool ilk_validate_pipe_wm(struct intel_display *display,
 				 struct intel_pipe_wm *pipe_wm)
 {
 	/* LP0 watermark maximums depend on this pipe alone */
@@ -2901,11 +2899,11 @@ static bool ilk_validate_pipe_wm(struct drm_i915_private *dev_priv,
 	struct ilk_wm_maximums max;
 
 	/* LP0 watermarks always use 1/2 DDB partitioning */
-	ilk_compute_wm_maximums(dev_priv, 0, &config, INTEL_DDB_PART_1_2, &max);
+	ilk_compute_wm_maximums(display, 0, &config, INTEL_DDB_PART_1_2, &max);
 
 	/* At least LP0 must be valid */
-	if (!ilk_validate_wm_level(dev_priv, 0, &max, &pipe_wm->wm[0])) {
-		drm_dbg_kms(&dev_priv->drm, "LP0 watermark invalid\n");
+	if (!ilk_validate_wm_level(display, 0, &max, &pipe_wm->wm[0])) {
+		drm_dbg_kms(display->drm, "LP0 watermark invalid\n");
 		return false;
 	}
 
@@ -2916,7 +2914,7 @@ static bool ilk_validate_pipe_wm(struct drm_i915_private *dev_priv,
 static int ilk_compute_pipe_wm(struct intel_atomic_state *state,
 			       struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_pipe_wm *pipe_wm;
@@ -2943,10 +2941,10 @@ static int ilk_compute_pipe_wm(struct intel_atomic_state *state,
 	pipe_wm->sprites_enabled = crtc_state->active_planes & BIT(PLANE_SPRITE0);
 	pipe_wm->sprites_scaled = crtc_state->scaled_planes & BIT(PLANE_SPRITE0);
 
-	usable_level = dev_priv->display.wm.num_levels - 1;
+	usable_level = display->wm.num_levels - 1;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
-	if (DISPLAY_VER(dev_priv) < 7 && pipe_wm->sprites_enabled)
+	if (DISPLAY_VER(display) < 7 && pipe_wm->sprites_enabled)
 		usable_level = 1;
 
 	/* ILK/SNB/IVB: LP1+ watermarks only w/o scaling */
@@ -2954,18 +2952,18 @@ static int ilk_compute_pipe_wm(struct intel_atomic_state *state,
 		usable_level = 0;
 
 	memset(&pipe_wm->wm, 0, sizeof(pipe_wm->wm));
-	ilk_compute_wm_level(dev_priv, crtc, 0, crtc_state,
+	ilk_compute_wm_level(display, crtc, 0, crtc_state,
 			     pristate, sprstate, curstate, &pipe_wm->wm[0]);
 
-	if (!ilk_validate_pipe_wm(dev_priv, pipe_wm))
+	if (!ilk_validate_pipe_wm(display, pipe_wm))
 		return -EINVAL;
 
-	ilk_compute_wm_reg_maximums(dev_priv, 1, &max);
+	ilk_compute_wm_reg_maximums(display, 1, &max);
 
 	for (level = 1; level <= usable_level; level++) {
 		struct intel_wm_level *wm = &pipe_wm->wm[level];
 
-		ilk_compute_wm_level(dev_priv, crtc, level, crtc_state,
+		ilk_compute_wm_level(display, crtc, level, crtc_state,
 				     pristate, sprstate, curstate, wm);
 
 		/*
@@ -2973,7 +2971,7 @@ static int ilk_compute_pipe_wm(struct intel_atomic_state *state,
 		 * register maximums since such watermarks are
 		 * always invalid.
 		 */
-		if (!ilk_validate_wm_level(dev_priv, level, &max, wm)) {
+		if (!ilk_validate_wm_level(display, level, &max, wm)) {
 			memset(wm, 0, sizeof(*wm));
 			break;
 		}
@@ -2990,7 +2988,7 @@ static int ilk_compute_pipe_wm(struct intel_atomic_state *state,
 static int ilk_compute_intermediate_wm(struct intel_atomic_state *state,
 				       struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_crtc_state *old_crtc_state =
@@ -3015,7 +3013,7 @@ static int ilk_compute_intermediate_wm(struct intel_atomic_state *state,
 	intermediate->sprites_enabled |= active->sprites_enabled;
 	intermediate->sprites_scaled |= active->sprites_scaled;
 
-	for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+	for (level = 0; level < display->wm.num_levels; level++) {
 		struct intel_wm_level *intermediate_wm = &intermediate->wm[level];
 		const struct intel_wm_level *active_wm = &active->wm[level];
 
@@ -3036,7 +3034,7 @@ static int ilk_compute_intermediate_wm(struct intel_atomic_state *state,
 	 * there's no safe way to transition from the old state to
 	 * the new state, so we need to fail the atomic transaction.
 	 */
-	if (!ilk_validate_pipe_wm(dev_priv, intermediate))
+	if (!ilk_validate_pipe_wm(display, intermediate))
 		return -EINVAL;
 
 	/*
@@ -3068,7 +3066,7 @@ static int ilk_compute_watermarks(struct intel_atomic_state *state,
 /*
  * Merge the watermarks from all active pipes for a specific level.
  */
-static void ilk_merge_wm_level(struct drm_i915_private *dev_priv,
+static void ilk_merge_wm_level(struct intel_display *display,
 			       int level,
 			       struct intel_wm_level *ret_wm)
 {
@@ -3076,7 +3074,7 @@ static void ilk_merge_wm_level(struct drm_i915_private *dev_priv,
 
 	ret_wm->enable = true;
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct intel_pipe_wm *active = &crtc->wm.active.ilk;
 		const struct intel_wm_level *wm = &active->wm[level];
 
@@ -3101,31 +3099,31 @@ static void ilk_merge_wm_level(struct drm_i915_private *dev_priv,
 /*
  * Merge all low power watermarks for all active pipes.
  */
-static void ilk_wm_merge(struct drm_i915_private *dev_priv,
+static void ilk_wm_merge(struct intel_display *display,
 			 const struct intel_wm_config *config,
 			 const struct ilk_wm_maximums *max,
 			 struct intel_pipe_wm *merged)
 {
-	int level, num_levels = dev_priv->display.wm.num_levels;
+	int level, num_levels = display->wm.num_levels;
 	int last_enabled_level = num_levels - 1;
 
 	/* ILK/SNB/IVB: LP1+ watermarks only w/ single pipe */
-	if ((DISPLAY_VER(dev_priv) < 7 || IS_IVYBRIDGE(dev_priv)) &&
+	if ((DISPLAY_VER(display) < 7 || display->platform.ivybridge) &&
 	    config->num_pipes_active > 1)
 		last_enabled_level = 0;
 
 	/* ILK: FBC WM must be disabled always */
-	merged->fbc_wm_enabled = DISPLAY_VER(dev_priv) >= 6;
+	merged->fbc_wm_enabled = DISPLAY_VER(display) >= 6;
 
 	/* merge each WM1+ level */
 	for (level = 1; level < num_levels; level++) {
 		struct intel_wm_level *wm = &merged->wm[level];
 
-		ilk_merge_wm_level(dev_priv, level, wm);
+		ilk_merge_wm_level(display, level, wm);
 
 		if (level > last_enabled_level)
 			wm->enable = false;
-		else if (!ilk_validate_wm_level(dev_priv, level, max, wm))
+		else if (!ilk_validate_wm_level(display, level, max, wm))
 			/* make sure all following levels get disabled */
 			last_enabled_level = level - 1;
 
@@ -3141,8 +3139,8 @@ static void ilk_wm_merge(struct drm_i915_private *dev_priv,
 	}
 
 	/* ILK: LP2+ must be disabled when FBC WM is disabled but FBC enabled */
-	if (DISPLAY_VER(dev_priv) == 5 && HAS_FBC(dev_priv) &&
-	    dev_priv->display.params.enable_fbc && !merged->fbc_wm_enabled) {
+	if (DISPLAY_VER(display) == 5 && HAS_FBC(display) &&
+	    display->params.enable_fbc && !merged->fbc_wm_enabled) {
 		for (level = 2; level < num_levels; level++) {
 			struct intel_wm_level *wm = &merged->wm[level];
 
@@ -3158,16 +3156,16 @@ static int ilk_wm_lp_to_level(int wm_lp, const struct intel_pipe_wm *pipe_wm)
 }
 
 /* The value we need to program into the WM_LPx latency field */
-static unsigned int ilk_wm_lp_latency(struct drm_i915_private *dev_priv,
+static unsigned int ilk_wm_lp_latency(struct intel_display *display,
 				      int level)
 {
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (display->platform.haswell || display->platform.broadwell)
 		return 2 * level;
 	else
-		return dev_priv->display.wm.pri_latency[level];
+		return display->wm.pri_latency[level];
 }
 
-static void ilk_compute_wm_results(struct drm_i915_private *dev_priv,
+static void ilk_compute_wm_results(struct intel_display *display,
 				   const struct intel_pipe_wm *merged,
 				   enum intel_ddb_partitioning partitioning,
 				   struct ilk_wm_values *results)
@@ -3191,14 +3189,14 @@ static void ilk_compute_wm_results(struct drm_i915_private *dev_priv,
 		 * disabled. Doing otherwise could cause underruns.
 		 */
 		results->wm_lp[wm_lp - 1] =
-			WM_LP_LATENCY(ilk_wm_lp_latency(dev_priv, level)) |
+			WM_LP_LATENCY(ilk_wm_lp_latency(display, level)) |
 			WM_LP_PRIMARY(r->pri_val) |
 			WM_LP_CURSOR(r->cur_val);
 
 		if (r->enable)
 			results->wm_lp[wm_lp - 1] |= WM_LP_ENABLE;
 
-		if (DISPLAY_VER(dev_priv) >= 8)
+		if (DISPLAY_VER(display) >= 8)
 			results->wm_lp[wm_lp - 1] |= WM_LP_FBC_BDW(r->fbc_val);
 		else
 			results->wm_lp[wm_lp - 1] |= WM_LP_FBC_ILK(r->fbc_val);
@@ -3209,19 +3207,19 @@ static void ilk_compute_wm_results(struct drm_i915_private *dev_priv,
 		 * Always set WM_LP_SPRITE_EN when spr_val != 0, even if the
 		 * level is disabled. Doing otherwise could cause underruns.
 		 */
-		if (DISPLAY_VER(dev_priv) < 7 && r->spr_val) {
-			drm_WARN_ON(&dev_priv->drm, wm_lp != 1);
+		if (DISPLAY_VER(display) < 7 && r->spr_val) {
+			drm_WARN_ON(display->drm, wm_lp != 1);
 			results->wm_lp_spr[wm_lp - 1] |= WM_LP_SPRITE_ENABLE;
 		}
 	}
 
 	/* LP0 register values */
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		enum pipe pipe = crtc->pipe;
 		const struct intel_pipe_wm *pipe_wm = &crtc->wm.active.ilk;
 		const struct intel_wm_level *r = &pipe_wm->wm[0];
 
-		if (drm_WARN_ON(&dev_priv->drm, !r->enable))
+		if (drm_WARN_ON(display->drm, !r->enable))
 			continue;
 
 		results->wm_pipe[pipe] =
@@ -3236,13 +3234,13 @@ static void ilk_compute_wm_results(struct drm_i915_private *dev_priv,
  * case both are at the same level. Prefer r1 in case they're the same.
  */
 static struct intel_pipe_wm *
-ilk_find_best_result(struct drm_i915_private *dev_priv,
+ilk_find_best_result(struct intel_display *display,
 		     struct intel_pipe_wm *r1,
 		     struct intel_pipe_wm *r2)
 {
 	int level, level1 = 0, level2 = 0;
 
-	for (level = 1; level < dev_priv->display.wm.num_levels; level++) {
+	for (level = 1; level < display->wm.num_levels; level++) {
 		if (r1->wm[level].enable)
 			level1 = level;
 		if (r2->wm[level].enable)
@@ -3268,7 +3266,7 @@ ilk_find_best_result(struct drm_i915_private *dev_priv,
 #define WM_DIRTY_FBC (1 << 24)
 #define WM_DIRTY_DDB (1 << 25)
 
-static unsigned int ilk_compute_wm_dirty(struct drm_i915_private *dev_priv,
+static unsigned int ilk_compute_wm_dirty(struct intel_display *display,
 					 const struct ilk_wm_values *old,
 					 const struct ilk_wm_values *new)
 {
@@ -3276,7 +3274,7 @@ static unsigned int ilk_compute_wm_dirty(struct drm_i915_private *dev_priv,
 	enum pipe pipe;
 	int wm_lp;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		if (old->wm_pipe[pipe] != new->wm_pipe[pipe]) {
 			dirty |= WM_DIRTY_PIPE(pipe);
 			/* Must disable LP1+ watermarks too */
@@ -3314,25 +3312,25 @@ static unsigned int ilk_compute_wm_dirty(struct drm_i915_private *dev_priv,
 	return dirty;
 }
 
-static bool _ilk_disable_lp_wm(struct drm_i915_private *dev_priv,
+static bool _ilk_disable_lp_wm(struct intel_display *display,
 			       unsigned int dirty)
 {
-	struct ilk_wm_values *previous = &dev_priv->display.wm.hw;
+	struct ilk_wm_values *previous = &display->wm.hw;
 	bool changed = false;
 
 	if (dirty & WM_DIRTY_LP(3) && previous->wm_lp[2] & WM_LP_ENABLE) {
 		previous->wm_lp[2] &= ~WM_LP_ENABLE;
-		intel_uncore_write(&dev_priv->uncore, WM3_LP_ILK, previous->wm_lp[2]);
+		intel_de_write(display, WM3_LP_ILK, previous->wm_lp[2]);
 		changed = true;
 	}
 	if (dirty & WM_DIRTY_LP(2) && previous->wm_lp[1] & WM_LP_ENABLE) {
 		previous->wm_lp[1] &= ~WM_LP_ENABLE;
-		intel_uncore_write(&dev_priv->uncore, WM2_LP_ILK, previous->wm_lp[1]);
+		intel_de_write(display, WM2_LP_ILK, previous->wm_lp[1]);
 		changed = true;
 	}
 	if (dirty & WM_DIRTY_LP(1) && previous->wm_lp[0] & WM_LP_ENABLE) {
 		previous->wm_lp[0] &= ~WM_LP_ENABLE;
-		intel_uncore_write(&dev_priv->uncore, WM1_LP_ILK, previous->wm_lp[0]);
+		intel_de_write(display, WM1_LP_ILK, previous->wm_lp[0]);
 		changed = true;
 	}
 
@@ -3348,73 +3346,73 @@ static bool _ilk_disable_lp_wm(struct drm_i915_private *dev_priv,
  * The spec says we shouldn't write when we don't need, because every write
  * causes WMs to be re-evaluated, expending some power.
  */
-static void ilk_write_wm_values(struct drm_i915_private *dev_priv,
+static void ilk_write_wm_values(struct intel_display *display,
 				struct ilk_wm_values *results)
 {
-	struct ilk_wm_values *previous = &dev_priv->display.wm.hw;
+	struct ilk_wm_values *previous = &display->wm.hw;
 	unsigned int dirty;
 
-	dirty = ilk_compute_wm_dirty(dev_priv, previous, results);
+	dirty = ilk_compute_wm_dirty(display, previous, results);
 	if (!dirty)
 		return;
 
-	_ilk_disable_lp_wm(dev_priv, dirty);
+	_ilk_disable_lp_wm(display, dirty);
 
 	if (dirty & WM_DIRTY_PIPE(PIPE_A))
-		intel_uncore_write(&dev_priv->uncore, WM0_PIPE_ILK(PIPE_A), results->wm_pipe[0]);
+		intel_de_write(display, WM0_PIPE_ILK(PIPE_A), results->wm_pipe[0]);
 	if (dirty & WM_DIRTY_PIPE(PIPE_B))
-		intel_uncore_write(&dev_priv->uncore, WM0_PIPE_ILK(PIPE_B), results->wm_pipe[1]);
+		intel_de_write(display, WM0_PIPE_ILK(PIPE_B), results->wm_pipe[1]);
 	if (dirty & WM_DIRTY_PIPE(PIPE_C))
-		intel_uncore_write(&dev_priv->uncore, WM0_PIPE_ILK(PIPE_C), results->wm_pipe[2]);
+		intel_de_write(display, WM0_PIPE_ILK(PIPE_C), results->wm_pipe[2]);
 
 	if (dirty & WM_DIRTY_DDB) {
-		if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
-			intel_uncore_rmw(&dev_priv->uncore, WM_MISC, WM_MISC_DATA_PARTITION_5_6,
-					 results->partitioning == INTEL_DDB_PART_1_2 ? 0 :
-					 WM_MISC_DATA_PARTITION_5_6);
+		if (display->platform.haswell || display->platform.broadwell)
+			intel_de_rmw(display, WM_MISC, WM_MISC_DATA_PARTITION_5_6,
+				     results->partitioning == INTEL_DDB_PART_1_2 ? 0 :
+				     WM_MISC_DATA_PARTITION_5_6);
 		else
-			intel_uncore_rmw(&dev_priv->uncore, DISP_ARB_CTL2, DISP_DATA_PARTITION_5_6,
-					 results->partitioning == INTEL_DDB_PART_1_2 ? 0 :
-					 DISP_DATA_PARTITION_5_6);
+			intel_de_rmw(display, DISP_ARB_CTL2, DISP_DATA_PARTITION_5_6,
+				     results->partitioning == INTEL_DDB_PART_1_2 ? 0 :
+				     DISP_DATA_PARTITION_5_6);
 	}
 
 	if (dirty & WM_DIRTY_FBC)
-		intel_uncore_rmw(&dev_priv->uncore, DISP_ARB_CTL, DISP_FBC_WM_DIS,
-				 results->enable_fbc_wm ? 0 : DISP_FBC_WM_DIS);
+		intel_de_rmw(display, DISP_ARB_CTL, DISP_FBC_WM_DIS,
+			     results->enable_fbc_wm ? 0 : DISP_FBC_WM_DIS);
 
 	if (dirty & WM_DIRTY_LP(1) &&
 	    previous->wm_lp_spr[0] != results->wm_lp_spr[0])
-		intel_uncore_write(&dev_priv->uncore, WM1S_LP_ILK, results->wm_lp_spr[0]);
+		intel_de_write(display, WM1S_LP_ILK, results->wm_lp_spr[0]);
 
-	if (DISPLAY_VER(dev_priv) >= 7) {
+	if (DISPLAY_VER(display) >= 7) {
 		if (dirty & WM_DIRTY_LP(2) && previous->wm_lp_spr[1] != results->wm_lp_spr[1])
-			intel_uncore_write(&dev_priv->uncore, WM2S_LP_IVB, results->wm_lp_spr[1]);
+			intel_de_write(display, WM2S_LP_IVB, results->wm_lp_spr[1]);
 		if (dirty & WM_DIRTY_LP(3) && previous->wm_lp_spr[2] != results->wm_lp_spr[2])
-			intel_uncore_write(&dev_priv->uncore, WM3S_LP_IVB, results->wm_lp_spr[2]);
+			intel_de_write(display, WM3S_LP_IVB, results->wm_lp_spr[2]);
 	}
 
 	if (dirty & WM_DIRTY_LP(1) && previous->wm_lp[0] != results->wm_lp[0])
-		intel_uncore_write(&dev_priv->uncore, WM1_LP_ILK, results->wm_lp[0]);
+		intel_de_write(display, WM1_LP_ILK, results->wm_lp[0]);
 	if (dirty & WM_DIRTY_LP(2) && previous->wm_lp[1] != results->wm_lp[1])
-		intel_uncore_write(&dev_priv->uncore, WM2_LP_ILK, results->wm_lp[1]);
+		intel_de_write(display, WM2_LP_ILK, results->wm_lp[1]);
 	if (dirty & WM_DIRTY_LP(3) && previous->wm_lp[2] != results->wm_lp[2])
-		intel_uncore_write(&dev_priv->uncore, WM3_LP_ILK, results->wm_lp[2]);
+		intel_de_write(display, WM3_LP_ILK, results->wm_lp[2]);
 
-	dev_priv->display.wm.hw = *results;
+	display->wm.hw = *results;
 }
 
-bool ilk_disable_cxsr(struct drm_i915_private *dev_priv)
+bool ilk_disable_cxsr(struct intel_display *display)
 {
-	return _ilk_disable_lp_wm(dev_priv, WM_DIRTY_LP_ALL);
+	return _ilk_disable_lp_wm(display, WM_DIRTY_LP_ALL);
 }
 
-static void ilk_compute_wm_config(struct drm_i915_private *dev_priv,
+static void ilk_compute_wm_config(struct intel_display *display,
 				  struct intel_wm_config *config)
 {
 	struct intel_crtc *crtc;
 
 	/* Compute the currently _active_ config */
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct intel_pipe_wm *wm = &crtc->wm.active.ilk;
 
 		if (!wm->pipe_enabled)
@@ -3426,7 +3424,7 @@ static void ilk_compute_wm_config(struct drm_i915_private *dev_priv,
 	}
 }
 
-static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
+static void ilk_program_watermarks(struct intel_display *display)
 {
 	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct ilk_wm_maximums max;
@@ -3434,18 +3432,18 @@ static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
 	struct ilk_wm_values results = {};
 	enum intel_ddb_partitioning partitioning;
 
-	ilk_compute_wm_config(dev_priv, &config);
+	ilk_compute_wm_config(display, &config);
 
-	ilk_compute_wm_maximums(dev_priv, 1, &config, INTEL_DDB_PART_1_2, &max);
-	ilk_wm_merge(dev_priv, &config, &max, &lp_wm_1_2);
+	ilk_compute_wm_maximums(display, 1, &config, INTEL_DDB_PART_1_2, &max);
+	ilk_wm_merge(display, &config, &max, &lp_wm_1_2);
 
 	/* 5/6 split only in single pipe config on IVB+ */
-	if (DISPLAY_VER(dev_priv) >= 7 &&
+	if (DISPLAY_VER(display) >= 7 &&
 	    config.num_pipes_active == 1 && config.sprites_enabled) {
-		ilk_compute_wm_maximums(dev_priv, 1, &config, INTEL_DDB_PART_5_6, &max);
-		ilk_wm_merge(dev_priv, &config, &max, &lp_wm_5_6);
+		ilk_compute_wm_maximums(display, 1, &config, INTEL_DDB_PART_5_6, &max);
+		ilk_wm_merge(display, &config, &max, &lp_wm_5_6);
 
-		best_lp_wm = ilk_find_best_result(dev_priv, &lp_wm_1_2, &lp_wm_5_6);
+		best_lp_wm = ilk_find_best_result(display, &lp_wm_1_2, &lp_wm_5_6);
 	} else {
 		best_lp_wm = &lp_wm_1_2;
 	}
@@ -3453,50 +3451,49 @@ static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
 	partitioning = (best_lp_wm == &lp_wm_1_2) ?
 		       INTEL_DDB_PART_1_2 : INTEL_DDB_PART_5_6;
 
-	ilk_compute_wm_results(dev_priv, best_lp_wm, partitioning, &results);
+	ilk_compute_wm_results(display, best_lp_wm, partitioning, &results);
 
-	ilk_write_wm_values(dev_priv, &results);
+	ilk_write_wm_values(display, &results);
 }
 
 static void ilk_initial_watermarks(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 	crtc->wm.active.ilk = crtc_state->wm.ilk.intermediate;
-	ilk_program_watermarks(dev_priv);
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	ilk_program_watermarks(display);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
 static void ilk_optimize_watermarks(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
 	if (!crtc_state->wm.need_postvbl_update)
 		return;
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 	crtc->wm.active.ilk = crtc_state->wm.ilk.optimal;
-	ilk_program_watermarks(dev_priv);
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	ilk_program_watermarks(display);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
 static void ilk_pipe_wm_get_hw_state(struct intel_crtc *crtc)
 {
-	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct ilk_wm_values *hw = &dev_priv->display.wm.hw;
+	struct intel_display *display = to_intel_display(crtc);
+	struct ilk_wm_values *hw = &display->wm.hw;
 	struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state);
 	struct intel_pipe_wm *active = &crtc_state->wm.ilk.optimal;
 	enum pipe pipe = crtc->pipe;
 
-	hw->wm_pipe[pipe] = intel_uncore_read(&dev_priv->uncore, WM0_PIPE_ILK(pipe));
+	hw->wm_pipe[pipe] = intel_de_read(display, WM0_PIPE_ILK(pipe));
 
 	memset(active, 0, sizeof(*active));
 
@@ -3523,7 +3520,7 @@ static void ilk_pipe_wm_get_hw_state(struct intel_crtc *crtc)
 		 * should be marked as enabled but zeroed,
 		 * which is what we'd compute them to.
 		 */
-		for (level = 0; level < dev_priv->display.wm.num_levels; level++)
+		for (level = 0; level < display->wm.num_levels; level++)
 			active->wm[level].enable = true;
 	}
 
@@ -3572,7 +3569,7 @@ static int ilk_sanitize_watermarks_add_affected(struct drm_atomic_state *state)
  * through the atomic check code to calculate new watermark values in the
  * state object.
  */
-void ilk_wm_sanitize(struct drm_i915_private *dev_priv)
+void ilk_wm_sanitize(struct intel_display *display)
 {
 	struct drm_atomic_state *state;
 	struct intel_atomic_state *intel_state;
@@ -3583,14 +3580,14 @@ void ilk_wm_sanitize(struct drm_i915_private *dev_priv)
 	int i;
 
 	/* Only supported on platforms that use atomic watermark design */
-	if (!dev_priv->display.funcs.wm->optimize_watermarks)
+	if (!display->funcs.wm->optimize_watermarks)
 		return;
 
-	if (drm_WARN_ON(&dev_priv->drm, DISPLAY_VER(dev_priv) >= 9))
+	if (drm_WARN_ON(display->drm, DISPLAY_VER(display) >= 9))
 		return;
 
-	state = drm_atomic_state_alloc(&dev_priv->drm);
-	if (drm_WARN_ON(&dev_priv->drm, !state))
+	state = drm_atomic_state_alloc(display->drm);
+	if (drm_WARN_ON(display->drm, !state))
 		return;
 
 	intel_state = to_intel_atomic_state(state);
@@ -3606,14 +3603,14 @@ retry:
 	 * intermediate watermarks (since we don't trust the current
 	 * watermarks).
 	 */
-	if (!HAS_GMCH(dev_priv))
+	if (!HAS_GMCH(display))
 		intel_state->skip_intermediate_wm = true;
 
 	ret = ilk_sanitize_watermarks_add_affected(state);
 	if (ret)
 		goto fail;
 
-	ret = intel_atomic_check(&dev_priv->drm, state);
+	ret = intel_atomic_check(display->drm, state);
 	if (ret)
 		goto fail;
 
@@ -3643,7 +3640,7 @@ fail:
 	 * If this actually happens, we'll have to just leave the
 	 * BIOS-programmed watermarks untouched and hope for the best.
 	 */
-	drm_WARN(&dev_priv->drm, ret,
+	drm_WARN(display->drm, ret,
 		 "Could not determine valid watermarks for inherited state\n");
 
 	drm_atomic_state_put(state);
@@ -3657,18 +3654,18 @@ fail:
 #define _FW_WM_VLV(value, plane) \
 	(((value) & DSPFW_ ## plane ## _MASK_VLV) >> DSPFW_ ## plane ## _SHIFT)
 
-static void g4x_read_wm_values(struct drm_i915_private *dev_priv,
+static void g4x_read_wm_values(struct intel_display *display,
 			       struct g4x_wm_values *wm)
 {
 	u32 tmp;
 
-	tmp = intel_uncore_read(&dev_priv->uncore, DSPFW1(dev_priv));
+	tmp = intel_de_read(display, DSPFW1(display));
 	wm->sr.plane = _FW_WM(tmp, SR);
 	wm->pipe[PIPE_B].plane[PLANE_CURSOR] = _FW_WM(tmp, CURSORB);
 	wm->pipe[PIPE_B].plane[PLANE_PRIMARY] = _FW_WM(tmp, PLANEB);
 	wm->pipe[PIPE_A].plane[PLANE_PRIMARY] = _FW_WM(tmp, PLANEA);
 
-	tmp = intel_uncore_read(&dev_priv->uncore, DSPFW2(dev_priv));
+	tmp = intel_de_read(display, DSPFW2(display));
 	wm->fbc_en = tmp & DSPFW_FBC_SR_EN;
 	wm->sr.fbc = _FW_WM(tmp, FBC_SR);
 	wm->hpll.fbc = _FW_WM(tmp, FBC_HPLL_SR);
@@ -3676,21 +3673,21 @@ static void g4x_read_wm_values(struct drm_i915_private *dev_priv,
 	wm->pipe[PIPE_A].plane[PLANE_CURSOR] = _FW_WM(tmp, CURSORA);
 	wm->pipe[PIPE_A].plane[PLANE_SPRITE0] = _FW_WM(tmp, SPRITEA);
 
-	tmp = intel_uncore_read(&dev_priv->uncore, DSPFW3(dev_priv));
+	tmp = intel_de_read(display, DSPFW3(display));
 	wm->hpll_en = tmp & DSPFW_HPLL_SR_EN;
 	wm->sr.cursor = _FW_WM(tmp, CURSOR_SR);
 	wm->hpll.cursor = _FW_WM(tmp, HPLL_CURSOR);
 	wm->hpll.plane = _FW_WM(tmp, HPLL_SR);
 }
 
-static void vlv_read_wm_values(struct drm_i915_private *dev_priv,
+static void vlv_read_wm_values(struct intel_display *display,
 			       struct vlv_wm_values *wm)
 {
 	enum pipe pipe;
 	u32 tmp;
 
-	for_each_pipe(dev_priv, pipe) {
-		tmp = intel_uncore_read(&dev_priv->uncore, VLV_DDL(pipe));
+	for_each_pipe(display, pipe) {
+		tmp = intel_de_read(display, VLV_DDL(pipe));
 
 		wm->ddl[pipe].plane[PLANE_PRIMARY] =
 			(tmp >> DDL_PLANE_SHIFT) & (DDL_PRECISION_HIGH | DRAIN_LATENCY_MASK);
@@ -3702,34 +3699,34 @@ static void vlv_read_wm_values(struct drm_i915_private *dev_priv,
 			(tmp >> DDL_SPRITE_SHIFT(1)) & (DDL_PRECISION_HIGH | DRAIN_LATENCY_MASK);
 	}
 
-	tmp = intel_uncore_read(&dev_priv->uncore, DSPFW1(dev_priv));
+	tmp = intel_de_read(display, DSPFW1(display));
 	wm->sr.plane = _FW_WM(tmp, SR);
 	wm->pipe[PIPE_B].plane[PLANE_CURSOR] = _FW_WM(tmp, CURSORB);
 	wm->pipe[PIPE_B].plane[PLANE_PRIMARY] = _FW_WM_VLV(tmp, PLANEB);
 	wm->pipe[PIPE_A].plane[PLANE_PRIMARY] = _FW_WM_VLV(tmp, PLANEA);
 
-	tmp = intel_uncore_read(&dev_priv->uncore, DSPFW2(dev_priv));
+	tmp = intel_de_read(display, DSPFW2(display));
 	wm->pipe[PIPE_A].plane[PLANE_SPRITE1] = _FW_WM_VLV(tmp, SPRITEB);
 	wm->pipe[PIPE_A].plane[PLANE_CURSOR] = _FW_WM(tmp, CURSORA);
 	wm->pipe[PIPE_A].plane[PLANE_SPRITE0] = _FW_WM_VLV(tmp, SPRITEA);
 
-	tmp = intel_uncore_read(&dev_priv->uncore, DSPFW3(dev_priv));
+	tmp = intel_de_read(display, DSPFW3(display));
 	wm->sr.cursor = _FW_WM(tmp, CURSOR_SR);
 
-	if (IS_CHERRYVIEW(dev_priv)) {
-		tmp = intel_uncore_read(&dev_priv->uncore, DSPFW7_CHV);
+	if (display->platform.cherryview) {
+		tmp = intel_de_read(display, DSPFW7_CHV);
 		wm->pipe[PIPE_B].plane[PLANE_SPRITE1] = _FW_WM_VLV(tmp, SPRITED);
 		wm->pipe[PIPE_B].plane[PLANE_SPRITE0] = _FW_WM_VLV(tmp, SPRITEC);
 
-		tmp = intel_uncore_read(&dev_priv->uncore, DSPFW8_CHV);
+		tmp = intel_de_read(display, DSPFW8_CHV);
 		wm->pipe[PIPE_C].plane[PLANE_SPRITE1] = _FW_WM_VLV(tmp, SPRITEF);
 		wm->pipe[PIPE_C].plane[PLANE_SPRITE0] = _FW_WM_VLV(tmp, SPRITEE);
 
-		tmp = intel_uncore_read(&dev_priv->uncore, DSPFW9_CHV);
+		tmp = intel_de_read(display, DSPFW9_CHV);
 		wm->pipe[PIPE_C].plane[PLANE_PRIMARY] = _FW_WM_VLV(tmp, PLANEC);
 		wm->pipe[PIPE_C].plane[PLANE_CURSOR] = _FW_WM(tmp, CURSORC);
 
-		tmp = intel_uncore_read(&dev_priv->uncore, DSPHOWM);
+		tmp = intel_de_read(display, DSPHOWM);
 		wm->sr.plane |= _FW_WM(tmp, SR_HI) << 9;
 		wm->pipe[PIPE_C].plane[PLANE_SPRITE1] |= _FW_WM(tmp, SPRITEF_HI) << 8;
 		wm->pipe[PIPE_C].plane[PLANE_SPRITE0] |= _FW_WM(tmp, SPRITEE_HI) << 8;
@@ -3741,11 +3738,11 @@ static void vlv_read_wm_values(struct drm_i915_private *dev_priv,
 		wm->pipe[PIPE_A].plane[PLANE_SPRITE0] |= _FW_WM(tmp, SPRITEA_HI) << 8;
 		wm->pipe[PIPE_A].plane[PLANE_PRIMARY] |= _FW_WM(tmp, PLANEA_HI) << 8;
 	} else {
-		tmp = intel_uncore_read(&dev_priv->uncore, DSPFW7);
+		tmp = intel_de_read(display, DSPFW7);
 		wm->pipe[PIPE_B].plane[PLANE_SPRITE1] = _FW_WM_VLV(tmp, SPRITED);
 		wm->pipe[PIPE_B].plane[PLANE_SPRITE0] = _FW_WM_VLV(tmp, SPRITEC);
 
-		tmp = intel_uncore_read(&dev_priv->uncore, DSPHOWM);
+		tmp = intel_de_read(display, DSPHOWM);
 		wm->sr.plane |= _FW_WM(tmp, SR_HI) << 9;
 		wm->pipe[PIPE_B].plane[PLANE_SPRITE1] |= _FW_WM(tmp, SPRITED_HI) << 8;
 		wm->pipe[PIPE_B].plane[PLANE_SPRITE0] |= _FW_WM(tmp, SPRITEC_HI) << 8;
@@ -3759,16 +3756,16 @@ static void vlv_read_wm_values(struct drm_i915_private *dev_priv,
 #undef _FW_WM
 #undef _FW_WM_VLV
 
-static void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv)
+static void g4x_wm_get_hw_state(struct intel_display *display)
 {
-	struct g4x_wm_values *wm = &dev_priv->display.wm.g4x;
+	struct g4x_wm_values *wm = &display->wm.g4x;
 	struct intel_crtc *crtc;
 
-	g4x_read_wm_values(dev_priv, wm);
+	g4x_read_wm_values(display, wm);
 
-	wm->cxsr = intel_uncore_read(&dev_priv->uncore, FW_BLC_SELF) & FW_BLC_SELF_EN;
+	wm->cxsr = intel_de_read(display, FW_BLC_SELF) & FW_BLC_SELF_EN;
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct g4x_wm_state *active = &crtc->wm.active.g4x;
@@ -3833,7 +3830,7 @@ static void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv)
 		crtc_state->wm.g4x.optimal = *active;
 		crtc_state->wm.g4x.intermediate = *active;
 
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Initial watermarks: pipe %c, plane=%d, cursor=%d, sprite=%d\n",
 			    pipe_name(pipe),
 			    wm->pipe[pipe].plane[PLANE_PRIMARY],
@@ -3841,26 +3838,25 @@ static void g4x_wm_get_hw_state(struct drm_i915_private *dev_priv)
 			    wm->pipe[pipe].plane[PLANE_SPRITE0]);
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Initial SR watermarks: plane=%d, cursor=%d fbc=%d\n",
 		    wm->sr.plane, wm->sr.cursor, wm->sr.fbc);
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Initial HPLL watermarks: plane=%d, SR cursor=%d fbc=%d\n",
 		    wm->hpll.plane, wm->hpll.cursor, wm->hpll.fbc);
-	drm_dbg_kms(&dev_priv->drm, "Initial SR=%s HPLL=%s FBC=%s\n",
+	drm_dbg_kms(display->drm, "Initial SR=%s HPLL=%s FBC=%s\n",
 		    str_yes_no(wm->cxsr), str_yes_no(wm->hpll_en),
 		    str_yes_no(wm->fbc_en));
 }
 
-static void g4x_wm_sanitize(struct drm_i915_private *dev_priv)
+static void g4x_wm_sanitize(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_plane *plane;
 	struct intel_crtc *crtc;
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 
-	for_each_intel_plane(&dev_priv->drm, plane) {
+	for_each_intel_plane(display->drm, plane) {
 		struct intel_crtc *crtc =
 			intel_crtc_for_pipe(display, plane->pipe);
 		struct intel_crtc_state *crtc_state =
@@ -3873,7 +3869,7 @@ static void g4x_wm_sanitize(struct drm_i915_private *dev_priv)
 		if (plane_state->uapi.visible)
 			continue;
 
-		for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+		for (level = 0; level < display->wm.num_levels; level++) {
 			struct g4x_pipe_wm *raw =
 				&crtc_state->wm.g4x.raw[level];
 
@@ -3884,36 +3880,37 @@ static void g4x_wm_sanitize(struct drm_i915_private *dev_priv)
 		}
 	}
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		int ret;
 
 		ret = _g4x_compute_pipe_wm(crtc_state);
-		drm_WARN_ON(&dev_priv->drm, ret);
+		drm_WARN_ON(display->drm, ret);
 
 		crtc_state->wm.g4x.intermediate =
 			crtc_state->wm.g4x.optimal;
 		crtc->wm.active.g4x = crtc_state->wm.g4x.optimal;
 	}
 
-	g4x_program_watermarks(dev_priv);
+	g4x_program_watermarks(display);
 
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
-static void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv)
+static void vlv_wm_get_hw_state(struct intel_display *display)
 {
-	struct vlv_wm_values *wm = &dev_priv->display.wm.vlv;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	struct vlv_wm_values *wm = &display->wm.vlv;
 	struct intel_crtc *crtc;
 	u32 val;
 
-	vlv_read_wm_values(dev_priv, wm);
+	vlv_read_wm_values(display, wm);
 
-	wm->cxsr = intel_uncore_read(&dev_priv->uncore, FW_BLC_SELF_VLV) & FW_CSPWRDWNEN;
+	wm->cxsr = intel_de_read(display, FW_BLC_SELF_VLV) & FW_CSPWRDWNEN;
 	wm->level = VLV_WM_LEVEL_PM2;
 
-	if (IS_CHERRYVIEW(dev_priv)) {
+	if (display->platform.cherryview) {
 		vlv_punit_get(dev_priv);
 
 		val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM);
@@ -3935,10 +3932,10 @@ static void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv)
 
 		if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2) &
 			      FORCE_DDR_FREQ_REQ_ACK) == 0, 3)) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "Punit not acking DDR DVFS request, "
 				    "assuming DDR DVFS is disabled\n");
-			dev_priv->display.wm.num_levels = VLV_WM_LEVEL_PM5 + 1;
+			display->wm.num_levels = VLV_WM_LEVEL_PM5 + 1;
 		} else {
 			val = vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2);
 			if ((val & FORCE_DDR_HIGH_FREQ) == 0)
@@ -3948,7 +3945,7 @@ static void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv)
 		vlv_punit_put(dev_priv);
 	}
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct vlv_wm_state *active = &crtc->wm.active.vlv;
@@ -3988,7 +3985,7 @@ static void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv)
 		crtc_state->wm.vlv.optimal = *active;
 		crtc_state->wm.vlv.intermediate = *active;
 
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Initial watermarks: pipe %c, plane=%d, cursor=%d, sprite0=%d, sprite1=%d\n",
 			    pipe_name(pipe),
 			    wm->pipe[pipe].plane[PLANE_PRIMARY],
@@ -3997,20 +3994,19 @@ static void vlv_wm_get_hw_state(struct drm_i915_private *dev_priv)
 			    wm->pipe[pipe].plane[PLANE_SPRITE1]);
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Initial watermarks: SR plane=%d, SR cursor=%d level=%d cxsr=%d\n",
 		    wm->sr.plane, wm->sr.cursor, wm->level, wm->cxsr);
 }
 
-static void vlv_wm_sanitize(struct drm_i915_private *dev_priv)
+static void vlv_wm_sanitize(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_plane *plane;
 	struct intel_crtc *crtc;
 
-	mutex_lock(&dev_priv->display.wm.wm_mutex);
+	mutex_lock(&display->wm.wm_mutex);
 
-	for_each_intel_plane(&dev_priv->drm, plane) {
+	for_each_intel_plane(display->drm, plane) {
 		struct intel_crtc *crtc =
 			intel_crtc_for_pipe(display, plane->pipe);
 		struct intel_crtc_state *crtc_state =
@@ -4023,7 +4019,7 @@ static void vlv_wm_sanitize(struct drm_i915_private *dev_priv)
 		if (plane_state->uapi.visible)
 			continue;
 
-		for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+		for (level = 0; level < display->wm.num_levels; level++) {
 			struct g4x_pipe_wm *raw =
 				&crtc_state->wm.vlv.raw[level];
 
@@ -4031,33 +4027,33 @@ static void vlv_wm_sanitize(struct drm_i915_private *dev_priv)
 		}
 	}
 
-	for_each_intel_crtc(&dev_priv->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		int ret;
 
 		ret = _vlv_compute_pipe_wm(crtc_state);
-		drm_WARN_ON(&dev_priv->drm, ret);
+		drm_WARN_ON(display->drm, ret);
 
 		crtc_state->wm.vlv.intermediate =
 			crtc_state->wm.vlv.optimal;
 		crtc->wm.active.vlv = crtc_state->wm.vlv.optimal;
 	}
 
-	vlv_program_watermarks(dev_priv);
+	vlv_program_watermarks(display);
 
-	mutex_unlock(&dev_priv->display.wm.wm_mutex);
+	mutex_unlock(&display->wm.wm_mutex);
 }
 
 /*
  * FIXME should probably kill this and improve
  * the real watermark readout/sanitation instead
  */
-static void ilk_init_lp_watermarks(struct drm_i915_private *dev_priv)
+static void ilk_init_lp_watermarks(struct intel_display *display)
 {
-	intel_uncore_rmw(&dev_priv->uncore, WM3_LP_ILK, WM_LP_ENABLE, 0);
-	intel_uncore_rmw(&dev_priv->uncore, WM2_LP_ILK, WM_LP_ENABLE, 0);
-	intel_uncore_rmw(&dev_priv->uncore, WM1_LP_ILK, WM_LP_ENABLE, 0);
+	intel_de_rmw(display, WM3_LP_ILK, WM_LP_ENABLE, 0);
+	intel_de_rmw(display, WM2_LP_ILK, WM_LP_ENABLE, 0);
+	intel_de_rmw(display, WM1_LP_ILK, WM_LP_ENABLE, 0);
 
 	/*
 	 * Don't touch WM_LP_SPRITE_ENABLE here.
@@ -4065,37 +4061,37 @@ static void ilk_init_lp_watermarks(struct drm_i915_private *dev_priv)
 	 */
 }
 
-static void ilk_wm_get_hw_state(struct drm_i915_private *dev_priv)
+static void ilk_wm_get_hw_state(struct intel_display *display)
 {
-	struct ilk_wm_values *hw = &dev_priv->display.wm.hw;
+	struct ilk_wm_values *hw = &display->wm.hw;
 	struct intel_crtc *crtc;
 
-	ilk_init_lp_watermarks(dev_priv);
+	ilk_init_lp_watermarks(display);
 
-	for_each_intel_crtc(&dev_priv->drm, crtc)
+	for_each_intel_crtc(display->drm, crtc)
 		ilk_pipe_wm_get_hw_state(crtc);
 
-	hw->wm_lp[0] = intel_uncore_read(&dev_priv->uncore, WM1_LP_ILK);
-	hw->wm_lp[1] = intel_uncore_read(&dev_priv->uncore, WM2_LP_ILK);
-	hw->wm_lp[2] = intel_uncore_read(&dev_priv->uncore, WM3_LP_ILK);
+	hw->wm_lp[0] = intel_de_read(display, WM1_LP_ILK);
+	hw->wm_lp[1] = intel_de_read(display, WM2_LP_ILK);
+	hw->wm_lp[2] = intel_de_read(display, WM3_LP_ILK);
 
-	hw->wm_lp_spr[0] = intel_uncore_read(&dev_priv->uncore, WM1S_LP_ILK);
-	if (DISPLAY_VER(dev_priv) >= 7) {
-		hw->wm_lp_spr[1] = intel_uncore_read(&dev_priv->uncore, WM2S_LP_IVB);
-		hw->wm_lp_spr[2] = intel_uncore_read(&dev_priv->uncore, WM3S_LP_IVB);
+	hw->wm_lp_spr[0] = intel_de_read(display, WM1S_LP_ILK);
+	if (DISPLAY_VER(display) >= 7) {
+		hw->wm_lp_spr[1] = intel_de_read(display, WM2S_LP_IVB);
+		hw->wm_lp_spr[2] = intel_de_read(display, WM3S_LP_IVB);
 	}
 
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
-		hw->partitioning = (intel_uncore_read(&dev_priv->uncore, WM_MISC) &
+	if (display->platform.haswell || display->platform.broadwell)
+		hw->partitioning = (intel_de_read(display, WM_MISC) &
 				    WM_MISC_DATA_PARTITION_5_6) ?
 			INTEL_DDB_PART_5_6 : INTEL_DDB_PART_1_2;
-	else if (IS_IVYBRIDGE(dev_priv))
-		hw->partitioning = (intel_uncore_read(&dev_priv->uncore, DISP_ARB_CTL2) &
+	else if (display->platform.ivybridge)
+		hw->partitioning = (intel_de_read(display, DISP_ARB_CTL2) &
 				    DISP_DATA_PARTITION_5_6) ?
 			INTEL_DDB_PART_5_6 : INTEL_DDB_PART_1_2;
 
 	hw->enable_fbc_wm =
-		!(intel_uncore_read(&dev_priv->uncore, DISP_ARB_CTL) & DISP_FBC_WM_DIS);
+		!(intel_de_read(display, DISP_ARB_CTL) & DISP_FBC_WM_DIS);
 }
 
 static const struct intel_wm_funcs ilk_wm_funcs = {
@@ -4145,39 +4141,39 @@ static const struct intel_wm_funcs i845_wm_funcs = {
 static const struct intel_wm_funcs nop_funcs = {
 };
 
-void i9xx_wm_init(struct drm_i915_private *dev_priv)
+void i9xx_wm_init(struct intel_display *display)
 {
 	/* For FIFO watermark updates */
-	if (HAS_PCH_SPLIT(dev_priv)) {
-		ilk_setup_wm_latency(dev_priv);
-		dev_priv->display.funcs.wm = &ilk_wm_funcs;
-	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
-		vlv_setup_wm_latency(dev_priv);
-		dev_priv->display.funcs.wm = &vlv_wm_funcs;
-	} else if (IS_G4X(dev_priv)) {
-		g4x_setup_wm_latency(dev_priv);
-		dev_priv->display.funcs.wm = &g4x_wm_funcs;
-	} else if (IS_PINEVIEW(dev_priv)) {
-		if (!pnv_get_cxsr_latency(dev_priv)) {
-			drm_info(&dev_priv->drm, "Unknown FSB/MEM, disabling CxSR\n");
+	if (HAS_PCH_SPLIT(display)) {
+		ilk_setup_wm_latency(display);
+		display->funcs.wm = &ilk_wm_funcs;
+	} else if (display->platform.valleyview || display->platform.cherryview) {
+		vlv_setup_wm_latency(display);
+		display->funcs.wm = &vlv_wm_funcs;
+	} else if (display->platform.g4x) {
+		g4x_setup_wm_latency(display);
+		display->funcs.wm = &g4x_wm_funcs;
+	} else if (display->platform.pineview) {
+		if (!pnv_get_cxsr_latency(display)) {
+			drm_info(display->drm, "Unknown FSB/MEM, disabling CxSR\n");
 			/* Disable CxSR and never update its watermark again */
-			intel_set_memory_cxsr(dev_priv, false);
-			dev_priv->display.funcs.wm = &nop_funcs;
+			intel_set_memory_cxsr(display, false);
+			display->funcs.wm = &nop_funcs;
 		} else {
-			dev_priv->display.funcs.wm = &pnv_wm_funcs;
+			display->funcs.wm = &pnv_wm_funcs;
 		}
-	} else if (DISPLAY_VER(dev_priv) == 4) {
-		dev_priv->display.funcs.wm = &i965_wm_funcs;
-	} else if (DISPLAY_VER(dev_priv) == 3) {
-		dev_priv->display.funcs.wm = &i9xx_wm_funcs;
-	} else if (DISPLAY_VER(dev_priv) == 2) {
-		if (INTEL_NUM_PIPES(dev_priv) == 1)
-			dev_priv->display.funcs.wm = &i845_wm_funcs;
+	} else if (DISPLAY_VER(display) == 4) {
+		display->funcs.wm = &i965_wm_funcs;
+	} else if (DISPLAY_VER(display) == 3) {
+		display->funcs.wm = &i9xx_wm_funcs;
+	} else if (DISPLAY_VER(display) == 2) {
+		if (INTEL_NUM_PIPES(display) == 1)
+			display->funcs.wm = &i845_wm_funcs;
 		else
-			dev_priv->display.funcs.wm = &i9xx_wm_funcs;
+			display->funcs.wm = &i9xx_wm_funcs;
 	} else {
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"unexpected fall-through in %s\n", __func__);
-		dev_priv->display.funcs.wm = &nop_funcs;
+		display->funcs.wm = &nop_funcs;
 	}
 }
diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.h b/drivers/gpu/drm/i915/display/i9xx_wm.h
index 06ac37c6c94b..7bb363b2a756 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.h
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.h
@@ -8,28 +8,28 @@
 
 #include <linux/types.h>
 
-struct drm_i915_private;
 struct intel_crtc_state;
+struct intel_display;
 struct intel_plane_state;
 
 #ifdef I915
-bool ilk_disable_cxsr(struct drm_i915_private *i915);
-void ilk_wm_sanitize(struct drm_i915_private *i915);
-bool intel_set_memory_cxsr(struct drm_i915_private *i915, bool enable);
-void i9xx_wm_init(struct drm_i915_private *i915);
+bool ilk_disable_cxsr(struct intel_display *display);
+void ilk_wm_sanitize(struct intel_display *display);
+bool intel_set_memory_cxsr(struct intel_display *display, bool enable);
+void i9xx_wm_init(struct intel_display *display);
 #else
-static inline bool ilk_disable_cxsr(struct drm_i915_private *i915)
+static inline bool ilk_disable_cxsr(struct intel_display *display)
 {
 	return false;
 }
-static inline void ilk_wm_sanitize(struct drm_i915_private *i915)
+static inline void ilk_wm_sanitize(struct intel_display *display)
 {
 }
-static inline bool intel_set_memory_cxsr(struct drm_i915_private *i915, bool enable)
+static inline bool intel_set_memory_cxsr(struct intel_display *display, bool enable)
 {
 	return false;
 }
-static inline void i9xx_wm_init(struct drm_i915_private *i915)
+static inline void i9xx_wm_init(struct intel_display *display)
 {
 }
 #endif
diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c
index 402b7b2e1829..ca7033251e91 100644
--- a/drivers/gpu/drm/i915/display/icl_dsi.c
+++ b/drivers/gpu/drm/i915/display/icl_dsi.c
@@ -29,6 +29,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fixed.h>
 #include <drm/drm_mipi_dsi.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
 #include "i915_reg.h"
@@ -1826,107 +1827,56 @@ static const struct mipi_dsi_host_ops gen11_dsi_host_ops = {
 	.transfer = gen11_dsi_host_transfer,
 };
 
-#define ICL_PREPARE_CNT_MAX	0x7
-#define ICL_CLK_ZERO_CNT_MAX	0xf
-#define ICL_TRAIL_CNT_MAX	0x7
-#define ICL_TCLK_PRE_CNT_MAX	0x3
-#define ICL_TCLK_POST_CNT_MAX	0x7
-#define ICL_HS_ZERO_CNT_MAX	0xf
-#define ICL_EXIT_ZERO_CNT_MAX	0x7
-
 static void icl_dphy_param_init(struct intel_dsi *intel_dsi)
 {
-	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct intel_connector *connector = intel_dsi->attached_connector;
 	struct mipi_config *mipi_config = connector->panel.vbt.dsi.config;
 	u32 tlpx_ns;
-	u32 prepare_cnt, exit_zero_cnt, clk_zero_cnt, trail_cnt;
-	u32 ths_prepare_ns, tclk_trail_ns;
-	u32 hs_zero_cnt;
-	u32 tclk_pre_cnt;
+	u32 tclk_prepare_esc_clk, tclk_zero_esc_clk, tclk_pre_esc_clk;
+	u32 ths_prepare_esc_clk, ths_zero_esc_clk, ths_exit_esc_clk;
 
 	tlpx_ns = intel_dsi_tlpx_ns(intel_dsi);
 
-	tclk_trail_ns = max(mipi_config->tclk_trail, mipi_config->ths_trail);
-	ths_prepare_ns = max(mipi_config->ths_prepare,
-			     mipi_config->tclk_prepare);
-
 	/*
-	 * prepare cnt in escape clocks
-	 * this field represents a hexadecimal value with a precision
-	 * of 1.2 – i.e. the most significant bit is the integer
-	 * and the least significant 2 bits are fraction bits.
-	 * so, the field can represent a range of 0.25 to 1.75
+	 * The clock and data lane prepare timing parameters are in expressed in
+	 * units of 1/4 escape clocks, and all the other timings parameters in
+	 * escape clocks.
 	 */
-	prepare_cnt = DIV_ROUND_UP(ths_prepare_ns * 4, tlpx_ns);
-	if (prepare_cnt > ICL_PREPARE_CNT_MAX) {
-		drm_dbg_kms(display->drm, "prepare_cnt out of range (%d)\n",
-			    prepare_cnt);
-		prepare_cnt = ICL_PREPARE_CNT_MAX;
-	}
+	tclk_prepare_esc_clk = DIV_ROUND_UP(mipi_config->tclk_prepare * 4, tlpx_ns);
+	tclk_prepare_esc_clk = min(tclk_prepare_esc_clk, 7);
 
-	/* clk zero count in escape clocks */
-	clk_zero_cnt = DIV_ROUND_UP(mipi_config->tclk_prepare_clkzero -
-				    ths_prepare_ns, tlpx_ns);
-	if (clk_zero_cnt > ICL_CLK_ZERO_CNT_MAX) {
-		drm_dbg_kms(display->drm,
-			    "clk_zero_cnt out of range (%d)\n", clk_zero_cnt);
-		clk_zero_cnt = ICL_CLK_ZERO_CNT_MAX;
-	}
+	tclk_zero_esc_clk = DIV_ROUND_UP(mipi_config->tclk_prepare_clkzero -
+					 mipi_config->tclk_prepare, tlpx_ns);
+	tclk_zero_esc_clk = min(tclk_zero_esc_clk, 15);
 
-	/* trail cnt in escape clocks*/
-	trail_cnt = DIV_ROUND_UP(tclk_trail_ns, tlpx_ns);
-	if (trail_cnt > ICL_TRAIL_CNT_MAX) {
-		drm_dbg_kms(display->drm, "trail_cnt out of range (%d)\n",
-			    trail_cnt);
-		trail_cnt = ICL_TRAIL_CNT_MAX;
-	}
+	tclk_pre_esc_clk = DIV_ROUND_UP(mipi_config->tclk_pre, tlpx_ns);
+	tclk_pre_esc_clk = min(tclk_pre_esc_clk, 3);
 
-	/* tclk pre count in escape clocks */
-	tclk_pre_cnt = DIV_ROUND_UP(mipi_config->tclk_pre, tlpx_ns);
-	if (tclk_pre_cnt > ICL_TCLK_PRE_CNT_MAX) {
-		drm_dbg_kms(display->drm,
-			    "tclk_pre_cnt out of range (%d)\n", tclk_pre_cnt);
-		tclk_pre_cnt = ICL_TCLK_PRE_CNT_MAX;
-	}
+	ths_prepare_esc_clk = DIV_ROUND_UP(mipi_config->ths_prepare * 4, tlpx_ns);
+	ths_prepare_esc_clk = min(ths_prepare_esc_clk, 7);
 
-	/* hs zero cnt in escape clocks */
-	hs_zero_cnt = DIV_ROUND_UP(mipi_config->ths_prepare_hszero -
-				   ths_prepare_ns, tlpx_ns);
-	if (hs_zero_cnt > ICL_HS_ZERO_CNT_MAX) {
-		drm_dbg_kms(display->drm, "hs_zero_cnt out of range (%d)\n",
-			    hs_zero_cnt);
-		hs_zero_cnt = ICL_HS_ZERO_CNT_MAX;
-	}
+	ths_zero_esc_clk = DIV_ROUND_UP(mipi_config->ths_prepare_hszero -
+					mipi_config->ths_prepare, tlpx_ns);
+	ths_zero_esc_clk = min(ths_zero_esc_clk, 15);
 
-	/* hs exit zero cnt in escape clocks */
-	exit_zero_cnt = DIV_ROUND_UP(mipi_config->ths_exit, tlpx_ns);
-	if (exit_zero_cnt > ICL_EXIT_ZERO_CNT_MAX) {
-		drm_dbg_kms(display->drm,
-			    "exit_zero_cnt out of range (%d)\n",
-			    exit_zero_cnt);
-		exit_zero_cnt = ICL_EXIT_ZERO_CNT_MAX;
-	}
+	ths_exit_esc_clk = DIV_ROUND_UP(mipi_config->ths_exit, tlpx_ns);
+	ths_exit_esc_clk = min(ths_exit_esc_clk, 7);
 
 	/* clock lane dphy timings */
 	intel_dsi->dphy_reg = (CLK_PREPARE_OVERRIDE |
-			       CLK_PREPARE(prepare_cnt) |
+			       CLK_PREPARE(tclk_prepare_esc_clk) |
 			       CLK_ZERO_OVERRIDE |
-			       CLK_ZERO(clk_zero_cnt) |
+			       CLK_ZERO(tclk_zero_esc_clk) |
 			       CLK_PRE_OVERRIDE |
-			       CLK_PRE(tclk_pre_cnt) |
-			       CLK_TRAIL_OVERRIDE |
-			       CLK_TRAIL(trail_cnt));
+			       CLK_PRE(tclk_pre_esc_clk));
 
 	/* data lanes dphy timings */
 	intel_dsi->dphy_data_lane_reg = (HS_PREPARE_OVERRIDE |
-					 HS_PREPARE(prepare_cnt) |
+					 HS_PREPARE(ths_prepare_esc_clk) |
 					 HS_ZERO_OVERRIDE |
-					 HS_ZERO(hs_zero_cnt) |
-					 HS_TRAIL_OVERRIDE |
-					 HS_TRAIL(trail_cnt) |
+					 HS_ZERO(ths_zero_esc_clk) |
 					 HS_EXIT_OVERRIDE |
-					 HS_EXIT(exit_zero_cnt));
+					 HS_EXIT(ths_exit_esc_clk));
 
 	intel_dsi_log_params(intel_dsi);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c
index 55f3ae1e68c9..c176bdbc19a3 100644
--- a/drivers/gpu/drm/i915/display/intel_alpm.c
+++ b/drivers/gpu/drm/i915/display/intel_alpm.c
@@ -5,12 +5,15 @@
 
 #include <linux/debugfs.h>
 
+#include <drm/drm_print.h>
+
 #include "intel_alpm.h"
 #include "intel_crtc.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
 #include "intel_dp_aux.h"
+#include "intel_psr.h"
 #include "intel_psr_regs.h"
 
 bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp)
@@ -23,7 +26,7 @@ bool intel_alpm_aux_less_wake_supported(struct intel_dp *intel_dp)
 	return intel_dp->alpm_dpcd & DP_ALPM_AUX_LESS_CAP;
 }
 
-void intel_alpm_init_dpcd(struct intel_dp *intel_dp)
+void intel_alpm_init(struct intel_dp *intel_dp)
 {
 	u8 dpcd;
 
@@ -31,6 +34,7 @@ void intel_alpm_init_dpcd(struct intel_dp *intel_dp)
 		return;
 
 	intel_dp->alpm_dpcd = dpcd;
+	mutex_init(&intel_dp->alpm_parameters.lock);
 }
 
 /*
@@ -276,6 +280,14 @@ void intel_alpm_lobf_compute_config(struct intel_dp *intel_dp,
 	int waketime_in_lines, first_sdp_position;
 	int context_latency, guardband;
 
+	if (intel_dp->alpm_parameters.lobf_disable_debug) {
+		drm_dbg_kms(display->drm, "LOBF is disabled by debug flag\n");
+		return;
+	}
+
+	if (intel_dp->alpm_parameters.sink_alpm_error)
+		return;
+
 	if (!intel_dp_is_edp(intel_dp))
 		return;
 
@@ -288,6 +300,10 @@ void intel_alpm_lobf_compute_config(struct intel_dp *intel_dp,
 	if (crtc_state->has_psr)
 		return;
 
+	if (crtc_state->vrr.vmin != crtc_state->vrr.vmax ||
+	    crtc_state->vrr.vmin != crtc_state->vrr.flipline)
+		return;
+
 	if (!(intel_alpm_aux_wake_supported(intel_dp) ||
 	      intel_alpm_aux_less_wake_supported(intel_dp)))
 		return;
@@ -316,15 +332,16 @@ static void lnl_alpm_configure(struct intel_dp *intel_dp,
 	enum port port = dp_to_dig_port(intel_dp)->base.port;
 	u32 alpm_ctl;
 
-	if (DISPLAY_VER(display) < 20 ||
-	    (!intel_dp->psr.sel_update_enabled && !intel_dp_is_edp(intel_dp)))
+	if (DISPLAY_VER(display) < 20 || (!intel_psr_needs_alpm(intel_dp, crtc_state) &&
+					  !crtc_state->has_lobf))
 		return;
 
+	mutex_lock(&intel_dp->alpm_parameters.lock);
 	/*
 	 * Panel Replay on eDP is always using ALPM aux less. I.e. no need to
 	 * check panel support at this point.
 	 */
-	if ((intel_dp->psr.panel_replay_enabled && intel_dp_is_edp(intel_dp)) ||
+	if ((crtc_state->has_panel_replay && intel_dp_is_edp(intel_dp)) ||
 	    (crtc_state->has_lobf && intel_alpm_aux_less_wake_supported(intel_dp))) {
 		alpm_ctl = ALPM_CTL_ALPM_ENABLE |
 			ALPM_CTL_ALPM_AUX_LESS_ENABLE |
@@ -353,18 +370,107 @@ static void lnl_alpm_configure(struct intel_dp *intel_dp,
 			ALPM_CTL_EXTENDED_FAST_WAKE_TIME(intel_dp->alpm_parameters.fast_wake_lines);
 	}
 
-	if (crtc_state->has_lobf)
+	if (crtc_state->has_lobf) {
 		alpm_ctl |= ALPM_CTL_LOBF_ENABLE;
+		drm_dbg_kms(display->drm, "Link off between frames (LOBF) enabled\n");
+	}
 
 	alpm_ctl |= ALPM_CTL_ALPM_ENTRY_CHECK(intel_dp->alpm_parameters.check_entry_lines);
 
 	intel_de_write(display, ALPM_CTL(display, cpu_transcoder), alpm_ctl);
+	mutex_unlock(&intel_dp->alpm_parameters.lock);
 }
 
 void intel_alpm_configure(struct intel_dp *intel_dp,
 			  const struct intel_crtc_state *crtc_state)
 {
 	lnl_alpm_configure(intel_dp, crtc_state);
+	intel_dp->alpm_parameters.transcoder = crtc_state->cpu_transcoder;
+}
+
+void intel_alpm_pre_plane_update(struct intel_atomic_state *state,
+				 struct intel_crtc *crtc)
+{
+	struct intel_display *display = to_intel_display(state);
+	const struct intel_crtc_state *crtc_state =
+		intel_atomic_get_new_crtc_state(state, crtc);
+	const struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+	struct intel_encoder *encoder;
+
+	if (DISPLAY_VER(display) < 20)
+		return;
+
+	if (crtc_state->has_lobf || crtc_state->has_lobf == old_crtc_state->has_lobf)
+		return;
+
+	for_each_intel_encoder_mask(display->drm, encoder,
+				    crtc_state->uapi.encoder_mask) {
+		struct intel_dp *intel_dp;
+
+		if (!intel_encoder_is_dp(encoder))
+			continue;
+
+		intel_dp = enc_to_intel_dp(encoder);
+
+		if (!intel_dp_is_edp(intel_dp))
+			continue;
+
+		if (old_crtc_state->has_lobf) {
+			mutex_lock(&intel_dp->alpm_parameters.lock);
+			intel_de_write(display, ALPM_CTL(display, cpu_transcoder), 0);
+			drm_dbg_kms(display->drm, "Link off between frames (LOBF) disabled\n");
+			mutex_unlock(&intel_dp->alpm_parameters.lock);
+		}
+	}
+}
+
+void intel_alpm_enable_sink(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *crtc_state)
+{
+	u8 val;
+
+	if (!intel_psr_needs_alpm(intel_dp, crtc_state) && !crtc_state->has_lobf)
+		return;
+
+	val = DP_ALPM_ENABLE | DP_ALPM_LOCK_ERROR_IRQ_HPD_ENABLE;
+
+	if (crtc_state->has_panel_replay || (crtc_state->has_lobf &&
+					     intel_alpm_aux_less_wake_supported(intel_dp)))
+		val |= DP_ALPM_MODE_AUX_LESS;
+
+	drm_dp_dpcd_writeb(&intel_dp->aux, DP_RECEIVER_ALPM_CONFIG, val);
+}
+
+void intel_alpm_post_plane_update(struct intel_atomic_state *state,
+				  struct intel_crtc *crtc)
+{
+	struct intel_display *display = to_intel_display(state);
+	const struct intel_crtc_state *crtc_state =
+		intel_atomic_get_new_crtc_state(state, crtc);
+	const struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
+	struct intel_encoder *encoder;
+
+	if (crtc_state->has_psr || !crtc_state->has_lobf ||
+	    crtc_state->has_lobf == old_crtc_state->has_lobf)
+		return;
+
+	for_each_intel_encoder_mask(display->drm, encoder,
+				    crtc_state->uapi.encoder_mask) {
+		struct intel_dp *intel_dp;
+
+		if (!intel_encoder_is_dp(encoder))
+			continue;
+
+		intel_dp = enc_to_intel_dp(encoder);
+
+		if (intel_dp_is_edp(intel_dp)) {
+			intel_alpm_enable_sink(intel_dp, crtc_state);
+			intel_alpm_configure(intel_dp, crtc_state);
+		}
+	}
 }
 
 static int i915_edp_lobf_info_show(struct seq_file *m, void *data)
@@ -403,6 +509,32 @@ out:
 
 DEFINE_SHOW_ATTRIBUTE(i915_edp_lobf_info);
 
+static int
+i915_edp_lobf_debug_get(void *data, u64 *val)
+{
+	struct intel_connector *connector = data;
+	struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder);
+
+	*val = intel_dp->alpm_parameters.lobf_disable_debug;
+
+	return 0;
+}
+
+static int
+i915_edp_lobf_debug_set(void *data, u64 val)
+{
+	struct intel_connector *connector = data;
+	struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder);
+
+	intel_dp->alpm_parameters.lobf_disable_debug = val;
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(i915_edp_lobf_debug_fops,
+			i915_edp_lobf_debug_get, i915_edp_lobf_debug_set,
+			"%llu\n");
+
 void intel_alpm_lobf_debugfs_add(struct intel_connector *connector)
 {
 	struct intel_display *display = to_intel_display(connector);
@@ -412,6 +544,55 @@ void intel_alpm_lobf_debugfs_add(struct intel_connector *connector)
 	    connector->base.connector_type != DRM_MODE_CONNECTOR_eDP)
 		return;
 
+	debugfs_create_file("i915_edp_lobf_debug", 0644, root,
+			    connector, &i915_edp_lobf_debug_fops);
+
 	debugfs_create_file("i915_edp_lobf_info", 0444, root,
 			    connector, &i915_edp_lobf_info_fops);
 }
+
+void intel_alpm_disable(struct intel_dp *intel_dp)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->alpm_parameters.transcoder;
+
+	if (DISPLAY_VER(display) < 20 || !intel_dp->alpm_dpcd)
+		return;
+
+	mutex_lock(&intel_dp->alpm_parameters.lock);
+
+	intel_de_rmw(display, ALPM_CTL(display, cpu_transcoder),
+		     ALPM_CTL_ALPM_ENABLE | ALPM_CTL_LOBF_ENABLE |
+		     ALPM_CTL_ALPM_AUX_LESS_ENABLE, 0);
+
+	intel_de_rmw(display,
+		     PORT_ALPM_CTL(cpu_transcoder),
+		     PORT_ALPM_CTL_ALPM_AUX_LESS_ENABLE, 0);
+
+	drm_dbg_kms(display->drm, "Disabling ALPM\n");
+	mutex_unlock(&intel_dp->alpm_parameters.lock);
+}
+
+bool intel_alpm_get_error(struct intel_dp *intel_dp)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+	struct drm_dp_aux *aux = &intel_dp->aux;
+	u8 val;
+	int r;
+
+	r = drm_dp_dpcd_readb(aux, DP_RECEIVER_ALPM_STATUS, &val);
+	if (r != 1) {
+		drm_err(display->drm, "Error reading ALPM status\n");
+		return true;
+	}
+
+	if (val & DP_ALPM_LOCK_TIMEOUT_ERROR) {
+		drm_dbg_kms(display->drm, "ALPM lock timeout error\n");
+
+		/* Clearing error */
+		drm_dp_dpcd_writeb(aux, DP_RECEIVER_ALPM_STATUS, val);
+		return true;
+	}
+
+	return false;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_alpm.h b/drivers/gpu/drm/i915/display/intel_alpm.h
index 8c409b10dce6..c9fe21e3e72c 100644
--- a/drivers/gpu/drm/i915/display/intel_alpm.h
+++ b/drivers/gpu/drm/i915/display/intel_alpm.h
@@ -12,8 +12,10 @@ struct intel_dp;
 struct intel_crtc_state;
 struct drm_connector_state;
 struct intel_connector;
+struct intel_atomic_state;
+struct intel_crtc;
 
-void intel_alpm_init_dpcd(struct intel_dp *intel_dp);
+void intel_alpm_init(struct intel_dp *intel_dp);
 bool intel_alpm_compute_params(struct intel_dp *intel_dp,
 			       const struct intel_crtc_state *crtc_state);
 void intel_alpm_lobf_compute_config(struct intel_dp *intel_dp,
@@ -21,7 +23,15 @@ void intel_alpm_lobf_compute_config(struct intel_dp *intel_dp,
 				    struct drm_connector_state *conn_state);
 void intel_alpm_configure(struct intel_dp *intel_dp,
 			  const struct intel_crtc_state *crtc_state);
+void intel_alpm_enable_sink(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *crtc_state);
+void intel_alpm_pre_plane_update(struct intel_atomic_state *state,
+				 struct intel_crtc *crtc);
+void intel_alpm_post_plane_update(struct intel_atomic_state *state,
+				  struct intel_crtc *crtc);
 void intel_alpm_lobf_debugfs_add(struct intel_connector *connector);
 bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp);
 bool intel_alpm_aux_less_wake_supported(struct intel_dp *intel_dp);
+void intel_alpm_disable(struct intel_dp *intel_dp);
+bool intel_alpm_get_error(struct intel_dp *intel_dp);
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c
index 03dc54c802d3..e83feca5c9c9 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic.c
@@ -33,16 +33,17 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "intel_atomic.h"
 #include "intel_cdclk.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_dp_tunnel.h"
+#include "intel_fb.h"
 #include "intel_global_state.h"
 #include "intel_hdcp.h"
 #include "intel_psr.h"
-#include "intel_fb.h"
 #include "skl_universal_plane.h"
 
 /**
@@ -59,17 +60,16 @@ int intel_digital_connector_atomic_get_property(struct drm_connector *connector,
 						struct drm_property *property,
 						u64 *val)
 {
-	struct drm_device *dev = connector->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_display *display = to_intel_display(connector->dev);
 	const struct intel_digital_connector_state *intel_conn_state =
 		to_intel_digital_connector_state(state);
 
-	if (property == dev_priv->display.properties.force_audio)
+	if (property == display->properties.force_audio)
 		*val = intel_conn_state->force_audio;
-	else if (property == dev_priv->display.properties.broadcast_rgb)
+	else if (property == display->properties.broadcast_rgb)
 		*val = intel_conn_state->broadcast_rgb;
 	else {
-		drm_dbg_atomic(&dev_priv->drm,
+		drm_dbg_atomic(display->drm,
 			       "Unknown property [PROP:%d:%s]\n",
 			       property->base.id, property->name);
 		return -EINVAL;
@@ -92,22 +92,21 @@ int intel_digital_connector_atomic_set_property(struct drm_connector *connector,
 						struct drm_property *property,
 						u64 val)
 {
-	struct drm_device *dev = connector->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_display *display = to_intel_display(connector->dev);
 	struct intel_digital_connector_state *intel_conn_state =
 		to_intel_digital_connector_state(state);
 
-	if (property == dev_priv->display.properties.force_audio) {
+	if (property == display->properties.force_audio) {
 		intel_conn_state->force_audio = val;
 		return 0;
 	}
 
-	if (property == dev_priv->display.properties.broadcast_rgb) {
+	if (property == display->properties.broadcast_rgb) {
 		intel_conn_state->broadcast_rgb = val;
 		return 0;
 	}
 
-	drm_dbg_atomic(&dev_priv->drm, "Unknown property [PROP:%d:%s]\n",
+	drm_dbg_atomic(display->drm, "Unknown property [PROP:%d:%s]\n",
 		       property->base.id, property->name);
 	return -EINVAL;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
index 7276179df878..1bcfa5f4fd63 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
@@ -174,11 +174,27 @@ bool intel_plane_needs_physical(struct intel_plane *plane)
 		DISPLAY_INFO(display)->cursor_needs_physical;
 }
 
-bool intel_plane_can_async_flip(struct intel_plane *plane, u64 modifier)
+bool intel_plane_can_async_flip(struct intel_plane *plane, u32 format,
+				u64 modifier)
 {
+	if (intel_format_info_is_yuv_semiplanar(drm_format_info(format), modifier) ||
+	    format == DRM_FORMAT_C8)
+		return false;
+
 	return plane->can_async_flip && plane->can_async_flip(modifier);
 }
 
+bool intel_plane_format_mod_supported_async(struct drm_plane *plane,
+					    u32 format,
+					    u64 modifier)
+{
+	if (!plane->funcs->format_mod_supported(plane, format, modifier))
+		return false;
+
+	return intel_plane_can_async_flip(to_intel_plane(plane),
+					format, modifier);
+}
+
 unsigned int intel_adjusted_rate(const struct drm_rect *src,
 				 const struct drm_rect *dst,
 				 unsigned int rate)
diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.h b/drivers/gpu/drm/i915/display/intel_atomic_plane.h
index 6efac923dcbc..317320c32285 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic_plane.h
+++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.h
@@ -21,7 +21,8 @@ enum plane_id;
 
 struct intel_plane *
 intel_crtc_get_plane(struct intel_crtc *crtc, enum plane_id plane_id);
-bool intel_plane_can_async_flip(struct intel_plane *plane, u64 modifier);
+bool intel_plane_can_async_flip(struct intel_plane *plane, u32 format,
+				u64 modifier);
 unsigned int intel_adjusted_rate(const struct drm_rect *src,
 				 const struct drm_rect *dst,
 				 unsigned int rate);
@@ -89,5 +90,8 @@ int intel_atomic_add_affected_planes(struct intel_atomic_state *state,
 int intel_atomic_check_planes(struct intel_atomic_state *state);
 
 u32 intel_plane_ggtt_offset(const struct intel_plane_state *plane_state);
+bool intel_plane_format_mod_supported_async(struct drm_plane *plane,
+					    u32 format,
+					    u64 modifier);
 
 #endif /* __INTEL_ATOMIC_PLANE_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c
index ea935a5d94c8..40d8bbd8107d 100644
--- a/drivers/gpu/drm/i915/display/intel_audio.c
+++ b/drivers/gpu/drm/i915/display/intel_audio.c
@@ -27,9 +27,9 @@
 #include <drm/drm_edid.h>
 #include <drm/drm_eld.h>
 #include <drm/drm_fixed.h>
+#include <drm/drm_print.h>
 #include <drm/intel/i915_component.h>
 
-#include "i915_drv.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
 #include "intel_audio_regs.h"
@@ -587,19 +587,17 @@ static void ibx_audio_regs_init(struct intel_display *display,
 				enum pipe pipe,
 				struct ibx_audio_regs *regs)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (display->platform.valleyview || display->platform.cherryview) {
 		regs->hdmiw_hdmiedid = VLV_HDMIW_HDMIEDID(pipe);
 		regs->aud_config = VLV_AUD_CFG(pipe);
 		regs->aud_cntl_st = VLV_AUD_CNTL_ST(pipe);
 		regs->aud_cntrl_st2 = VLV_AUD_CNTL_ST2;
-	} else if (HAS_PCH_CPT(i915)) {
+	} else if (HAS_PCH_CPT(display)) {
 		regs->hdmiw_hdmiedid = CPT_HDMIW_HDMIEDID(pipe);
 		regs->aud_config = CPT_AUD_CFG(pipe);
 		regs->aud_cntl_st = CPT_AUD_CNTL_ST(pipe);
 		regs->aud_cntrl_st2 = CPT_AUD_CNTRL_ST2;
-	} else if (HAS_PCH_IBX(i915)) {
+	} else if (HAS_PCH_IBX(display)) {
 		regs->hdmiw_hdmiedid = IBX_HDMIW_HDMIEDID(pipe);
 		regs->aud_config = IBX_AUD_CFG(pipe);
 		regs->aud_cntl_st = IBX_AUD_CNTL_ST(pipe);
@@ -889,12 +887,10 @@ static const struct intel_audio_funcs hsw_audio_funcs = {
  */
 void intel_audio_hooks_init(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (display->platform.g4x)
 		display->funcs.audio = &g4x_audio_funcs;
 	else if (display->platform.valleyview || display->platform.cherryview ||
-		 HAS_PCH_CPT(i915) || HAS_PCH_IBX(i915))
+		 HAS_PCH_CPT(display) || HAS_PCH_IBX(display))
 		display->funcs.audio = &ibx_audio_funcs;
 	else if (display->platform.haswell || DISPLAY_VER(display) >= 8)
 		display->funcs.audio = &hsw_audio_funcs;
diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c
index 178dc6c8de80..5827da586003 100644
--- a/drivers/gpu/drm/i915/display/intel_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_backlight.c
@@ -10,12 +10,16 @@
 
 #include <acpi/video.h>
 
-#include "i915_drv.h"
+#include <drm/drm_file.h>
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_backlight.h"
 #include "intel_backlight_regs.h"
 #include "intel_connector.h"
 #include "intel_de.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dp_aux_backlight.h"
 #include "intel_dsi_dcs_backlight.h"
@@ -472,7 +476,6 @@ static void lpt_enable_backlight(const struct intel_crtc_state *crtc_state,
 {
 	struct intel_connector *connector = to_intel_connector(conn_state->connector);
 	struct intel_display *display = to_intel_display(connector);
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_panel *panel = &connector->panel;
 	u32 pch_ctl1, pch_ctl2;
 
@@ -485,7 +488,7 @@ static void lpt_enable_backlight(const struct intel_crtc_state *crtc_state,
 		intel_de_write(display, BLC_PWM_PCH_CTL1, pch_ctl1);
 	}
 
-	if (HAS_PCH_LPT(i915))
+	if (HAS_PCH_LPT(display))
 		intel_de_rmw(display, SOUTH_CHICKEN2, LPT_PWM_GRANULARITY,
 			     panel->backlight.alternate_pwm_increment ?
 			     LPT_PWM_GRANULARITY : 0);
@@ -502,7 +505,7 @@ static void lpt_enable_backlight(const struct intel_crtc_state *crtc_state,
 		pch_ctl1 |= BLM_PCH_POLARITY;
 
 	/* After LPT, override is the default. */
-	if (HAS_PCH_LPT(i915))
+	if (HAS_PCH_LPT(display))
 		pch_ctl1 |= BLM_PCH_OVERRIDE_ENABLE;
 
 	intel_de_write(display, BLC_PWM_PCH_CTL1, pch_ctl1);
@@ -901,11 +904,9 @@ static int intel_backlight_device_get_brightness(struct backlight_device *bd)
 {
 	struct intel_connector *connector = bl_get_data(bd);
 	struct intel_display *display = to_intel_display(connector);
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	intel_wakeref_t wakeref;
 	int ret = 0;
 
-	with_intel_runtime_pm(&i915->runtime_pm, wakeref) {
+	with_intel_display_rpm(display) {
 		u32 hw_level;
 
 		drm_modeset_lock(&display->drm->mode_config.connection_mutex, NULL);
@@ -1065,7 +1066,7 @@ static u32 spt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
  */
 static u32 lpt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 {
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 	struct intel_panel *panel = &connector->panel;
 	u32 mul, clock;
 
@@ -1074,7 +1075,7 @@ static u32 lpt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
 	else
 		mul = 128;
 
-	if (HAS_PCH_LPT_H(i915))
+	if (HAS_PCH_LPT_H(display))
 		clock = MHz(135); /* LPT:H */
 	else
 		clock = MHz(24); /* LPT:LP */
@@ -1231,12 +1232,11 @@ static u32 get_backlight_min_vbt(struct intel_connector *connector)
 static int lpt_setup_backlight(struct intel_connector *connector, enum pipe unused)
 {
 	struct intel_display *display = to_intel_display(connector);
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_panel *panel = &connector->panel;
 	u32 cpu_ctl2, pch_ctl1, pch_ctl2, val;
 	bool alt, cpu_mode;
 
-	if (HAS_PCH_LPT(i915))
+	if (HAS_PCH_LPT(display))
 		alt = intel_de_read(display, SOUTH_CHICKEN2) & LPT_PWM_GRANULARITY;
 	else
 		alt = intel_de_read(display, SOUTH_CHICKEN1) & SPT_PWM_GRANULARITY;
@@ -1260,7 +1260,7 @@ static int lpt_setup_backlight(struct intel_connector *connector, enum pipe unus
 
 	panel->backlight.pwm_enabled = pch_ctl1 & BLM_PCH_PWM_ENABLE;
 
-	cpu_mode = panel->backlight.pwm_enabled && HAS_PCH_LPT(i915) &&
+	cpu_mode = panel->backlight.pwm_enabled && HAS_PCH_LPT(display) &&
 		   !(pch_ctl1 & BLM_PCH_OVERRIDE_ENABLE) &&
 		   (cpu_ctl2 & BLM_PWM_ENABLE);
 
@@ -1467,15 +1467,13 @@ bxt_setup_backlight(struct intel_connector *connector, enum pipe unused)
 
 static int cnp_num_backlight_controllers(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	if (INTEL_PCH_TYPE(i915) >= PCH_MTL)
+	if (INTEL_PCH_TYPE(display) >= PCH_MTL)
 		return 2;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_DG1)
+	if (INTEL_PCH_TYPE(display) >= PCH_DG1)
 		return 1;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
+	if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 		return 2;
 
 	return 1;
@@ -1483,14 +1481,12 @@ static int cnp_num_backlight_controllers(struct intel_display *display)
 
 static bool cnp_backlight_controller_is_valid(struct intel_display *display, int controller)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (controller < 0 || controller >= cnp_num_backlight_controllers(display))
 		return false;
 
 	if (controller == 1 &&
-	    INTEL_PCH_TYPE(i915) >= PCH_ICP &&
-	    INTEL_PCH_TYPE(i915) <= PCH_ADP)
+	    INTEL_PCH_TYPE(display) >= PCH_ICP &&
+	    INTEL_PCH_TYPE(display) <= PCH_ADP)
 		return intel_de_read(display, SOUTH_CHICKEN1) & ICP_SECOND_PPS_IO_SELECT;
 
 	return true;
@@ -1819,7 +1815,6 @@ void intel_backlight_init_funcs(struct intel_panel *panel)
 	struct intel_connector *connector =
 		container_of(panel, struct intel_connector, panel);
 	struct intel_display *display = to_intel_display(connector);
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
 	if (connector->base.connector_type == DRM_MODE_CONNECTOR_DSI &&
 	    intel_dsi_dcs_init_backlight_funcs(connector) == 0)
@@ -1827,14 +1822,14 @@ void intel_backlight_init_funcs(struct intel_panel *panel)
 
 	if (display->platform.geminilake || display->platform.broxton) {
 		panel->backlight.pwm_funcs = &bxt_pwm_funcs;
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_CNP) {
+	} else if (INTEL_PCH_TYPE(display) >= PCH_CNP) {
 		panel->backlight.pwm_funcs = &cnp_pwm_funcs;
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_LPT_H) {
-		if (HAS_PCH_LPT(i915))
+	} else if (INTEL_PCH_TYPE(display) >= PCH_LPT_H) {
+		if (HAS_PCH_LPT(display))
 			panel->backlight.pwm_funcs = &lpt_pwm_funcs;
 		else
 			panel->backlight.pwm_funcs = &spt_pwm_funcs;
-	} else if (HAS_PCH_SPLIT(i915)) {
+	} else if (HAS_PCH_SPLIT(display)) {
 		panel->backlight.pwm_funcs = &pch_pwm_funcs;
 	} else if (display->platform.valleyview || display->platform.cherryview) {
 		if (connector->base.connector_type == DRM_MODE_CONNECTOR_DSI) {
diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index a8d08d7d82b3..ba7b8938b17c 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -37,6 +37,7 @@
 
 #include "i915_drv.h"
 #include "intel_display.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_gmbus.h"
 
@@ -2244,28 +2245,27 @@ static const u8 adlp_ddc_pin_map[] = {
 
 static u8 map_ddc_pin(struct intel_display *display, u8 vbt_pin)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	const u8 *ddc_pin_map;
 	int i, n_entries;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_MTL || display->platform.alderlake_p) {
+	if (INTEL_PCH_TYPE(display) >= PCH_MTL || display->platform.alderlake_p) {
 		ddc_pin_map = adlp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(adlp_ddc_pin_map);
 	} else if (display->platform.alderlake_s) {
 		ddc_pin_map = adls_ddc_pin_map;
 		n_entries = ARRAY_SIZE(adls_ddc_pin_map);
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_DG1) {
+	} else if (INTEL_PCH_TYPE(display) >= PCH_DG1) {
 		return vbt_pin;
-	} else if (display->platform.rocketlake && INTEL_PCH_TYPE(i915) == PCH_TGP) {
+	} else if (display->platform.rocketlake && INTEL_PCH_TYPE(display) == PCH_TGP) {
 		ddc_pin_map = rkl_pch_tgp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(rkl_pch_tgp_ddc_pin_map);
-	} else if (HAS_PCH_TGP(i915) && DISPLAY_VER(display) == 9) {
+	} else if (HAS_PCH_TGP(display) && DISPLAY_VER(display) == 9) {
 		ddc_pin_map = gen9bc_tgp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(gen9bc_tgp_ddc_pin_map);
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_ICP) {
+	} else if (INTEL_PCH_TYPE(display) >= PCH_ICP) {
 		ddc_pin_map = icp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(icp_ddc_pin_map);
-	} else if (HAS_PCH_CNP(i915)) {
+	} else if (HAS_PCH_CNP(display)) {
 		ddc_pin_map = cnp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(cnp_ddc_pin_map);
 	} else {
@@ -2864,8 +2864,6 @@ parse_general_definitions(struct intel_display *display)
 static void
 init_vbt_defaults(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	display->vbt.crt_ddc_pin = GMBUS_PIN_VGADDC;
 
 	/* general features */
@@ -2882,7 +2880,7 @@ init_vbt_defaults(struct intel_display *display)
 	 * clock for LVDS.
 	 */
 	display->vbt.lvds_ssc_freq = intel_bios_ssc_frequency(display,
-							      !HAS_PCH_SPLIT(i915));
+							      !HAS_PCH_SPLIT(display));
 	drm_dbg_kms(display->drm, "Set default to SSC at %d kHz\n",
 		    display->vbt.lvds_ssc_freq);
 }
@@ -3115,7 +3113,6 @@ static const struct vbt_header *intel_bios_get_vbt(struct intel_display *display
 {
 	struct drm_i915_private *i915 = to_i915(display->drm);
 	const struct vbt_header *vbt = NULL;
-	intel_wakeref_t wakeref;
 
 	vbt = firmware_get_vbt(display, sizep);
 
@@ -3126,12 +3123,12 @@ static const struct vbt_header *intel_bios_get_vbt(struct intel_display *display
 	 * If the OpRegion does not have VBT, look in SPI flash
 	 * through MMIO or PCI mapping
 	 */
-	if (!vbt && IS_DGFX(i915))
-		with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+	if (!vbt && display->platform.dgfx)
+		with_intel_display_rpm(display)
 			vbt = oprom_get_vbt(display, intel_rom_spi(i915), sizep, "SPI flash");
 
 	if (!vbt)
-		with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+		with_intel_display_rpm(display)
 			vbt = oprom_get_vbt(display, intel_rom_pci(i915), sizep, "PCI ROM");
 
 	return vbt;
diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h
index f9841f0498c6..6cd7a011b8c4 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.h
+++ b/drivers/gpu/drm/i915/display/intel_bios.h
@@ -24,7 +24,7 @@
 /*
  * Please use intel_vbt_defs.h for VBT private data, to hide and abstract away
  * the VBT from the rest of the driver. Add the parsed, clean data to struct
- * intel_vbt_data within struct drm_i915_private.
+ * intel_vbt_data within struct intel_display.
  */
 
 #ifndef _INTEL_BIOS_H_
diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 048be2872247..a5dd2932b852 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -39,14 +39,15 @@ struct intel_qgv_info {
 	u8 deinterleave;
 };
 
-static int dg1_mchbar_read_qgv_point_info(struct drm_i915_private *dev_priv,
+static int dg1_mchbar_read_qgv_point_info(struct intel_display *display,
 					  struct intel_qgv_point *sp,
 					  int point)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 dclk_ratio, dclk_reference;
 	u32 val;
 
-	val = intel_uncore_read(&dev_priv->uncore, SA_PERF_STATUS_0_0_0_MCHBAR_PC);
+	val = intel_uncore_read(&i915->uncore, SA_PERF_STATUS_0_0_0_MCHBAR_PC);
 	dclk_ratio = REG_FIELD_GET(DG1_QCLK_RATIO_MASK, val);
 	if (val & DG1_QCLK_REFERENCE)
 		dclk_reference = 6; /* 6 * 16.666 MHz = 100 MHz */
@@ -54,18 +55,18 @@ static int dg1_mchbar_read_qgv_point_info(struct drm_i915_private *dev_priv,
 		dclk_reference = 8; /* 8 * 16.666 MHz = 133 MHz */
 	sp->dclk = DIV_ROUND_UP((16667 * dclk_ratio * dclk_reference) + 500, 1000);
 
-	val = intel_uncore_read(&dev_priv->uncore, SKL_MC_BIOS_DATA_0_0_0_MCHBAR_PCU);
+	val = intel_uncore_read(&i915->uncore, SKL_MC_BIOS_DATA_0_0_0_MCHBAR_PCU);
 	if (val & DG1_GEAR_TYPE)
 		sp->dclk *= 2;
 
 	if (sp->dclk == 0)
 		return -EINVAL;
 
-	val = intel_uncore_read(&dev_priv->uncore, MCHBAR_CH0_CR_TC_PRE_0_0_0_MCHBAR);
+	val = intel_uncore_read(&i915->uncore, MCHBAR_CH0_CR_TC_PRE_0_0_0_MCHBAR);
 	sp->t_rp = REG_FIELD_GET(DG1_DRAM_T_RP_MASK, val);
 	sp->t_rdpre = REG_FIELD_GET(DG1_DRAM_T_RDPRE_MASK, val);
 
-	val = intel_uncore_read(&dev_priv->uncore, MCHBAR_CH0_CR_TC_PRE_0_0_0_MCHBAR_HIGH);
+	val = intel_uncore_read(&i915->uncore, MCHBAR_CH0_CR_TC_PRE_0_0_0_MCHBAR_HIGH);
 	sp->t_rcd = REG_FIELD_GET(DG1_DRAM_T_RCD_MASK, val);
 	sp->t_ras = REG_FIELD_GET(DG1_DRAM_T_RAS_MASK, val);
 
@@ -74,22 +75,23 @@ static int dg1_mchbar_read_qgv_point_info(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static int icl_pcode_read_qgv_point_info(struct drm_i915_private *dev_priv,
+static int icl_pcode_read_qgv_point_info(struct intel_display *display,
 					 struct intel_qgv_point *sp,
 					 int point)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 val = 0, val2 = 0;
 	u16 dclk;
 	int ret;
 
-	ret = snb_pcode_read(&dev_priv->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO |
+	ret = snb_pcode_read(&i915->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO |
 			     ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point),
 			     &val, &val2);
 	if (ret)
 		return ret;
 
 	dclk = val & 0xffff;
-	sp->dclk = DIV_ROUND_UP((16667 * dclk) + (DISPLAY_VER(dev_priv) >= 12 ? 500 : 0),
+	sp->dclk = DIV_ROUND_UP((16667 * dclk) + (DISPLAY_VER(display) >= 12 ? 500 : 0),
 				1000);
 	sp->t_rp = (val & 0xff0000) >> 16;
 	sp->t_rcd = (val & 0xff000000) >> 24;
@@ -102,14 +104,15 @@ static int icl_pcode_read_qgv_point_info(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static int adls_pcode_read_psf_gv_point_info(struct drm_i915_private *dev_priv,
-					    struct intel_psf_gv_point *points)
+static int adls_pcode_read_psf_gv_point_info(struct intel_display *display,
+					     struct intel_psf_gv_point *points)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 val = 0;
 	int ret;
 	int i;
 
-	ret = snb_pcode_read(&dev_priv->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO |
+	ret = snb_pcode_read(&i915->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO |
 			     ADL_PCODE_MEM_SS_READ_PSF_GV_INFO, &val, NULL);
 	if (ret)
 		return ret;
@@ -122,10 +125,10 @@ static int adls_pcode_read_psf_gv_point_info(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static u16 icl_qgv_points_mask(struct drm_i915_private *i915)
+static u16 icl_qgv_points_mask(struct intel_display *display)
 {
-	unsigned int num_psf_gv_points = i915->display.bw.max[0].num_psf_gv_points;
-	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	unsigned int num_psf_gv_points = display->bw.max[0].num_psf_gv_points;
+	unsigned int num_qgv_points = display->bw.max[0].num_qgv_points;
 	u16 qgv_points = 0, psf_points = 0;
 
 	/*
@@ -142,49 +145,51 @@ static u16 icl_qgv_points_mask(struct drm_i915_private *i915)
 	return ICL_PCODE_REQ_QGV_PT(qgv_points) | ADLS_PCODE_REQ_PSF_PT(psf_points);
 }
 
-static bool is_sagv_enabled(struct drm_i915_private *i915, u16 points_mask)
+static bool is_sagv_enabled(struct intel_display *display, u16 points_mask)
 {
-	return !is_power_of_2(~points_mask & icl_qgv_points_mask(i915) &
+	return !is_power_of_2(~points_mask & icl_qgv_points_mask(display) &
 			      ICL_PCODE_REQ_QGV_PT_MASK);
 }
 
-int icl_pcode_restrict_qgv_points(struct drm_i915_private *dev_priv,
+int icl_pcode_restrict_qgv_points(struct intel_display *display,
 				  u32 points_mask)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	int ret;
 
-	if (DISPLAY_VER(dev_priv) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		return 0;
 
 	/* bspec says to keep retrying for at least 1 ms */
-	ret = skl_pcode_request(&dev_priv->uncore, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG,
+	ret = skl_pcode_request(&i915->uncore, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG,
 				points_mask,
 				ICL_PCODE_REP_QGV_MASK | ADLS_PCODE_REP_PSF_MASK,
 				ICL_PCODE_REP_QGV_SAFE | ADLS_PCODE_REP_PSF_SAFE,
 				1);
 
 	if (ret < 0) {
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"Failed to disable qgv points (0x%x) points: 0x%x\n",
 			ret, points_mask);
 		return ret;
 	}
 
-	dev_priv->display.sagv.status = is_sagv_enabled(dev_priv, points_mask) ?
+	display->sagv.status = is_sagv_enabled(display, points_mask) ?
 		I915_SAGV_ENABLED : I915_SAGV_DISABLED;
 
 	return 0;
 }
 
-static int mtl_read_qgv_point_info(struct drm_i915_private *dev_priv,
+static int mtl_read_qgv_point_info(struct intel_display *display,
 				   struct intel_qgv_point *sp, int point)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 val, val2;
 	u16 dclk;
 
-	val = intel_uncore_read(&dev_priv->uncore,
+	val = intel_uncore_read(&i915->uncore,
 				MTL_MEM_SS_INFO_QGV_POINT_LOW(point));
-	val2 = intel_uncore_read(&dev_priv->uncore,
+	val2 = intel_uncore_read(&i915->uncore,
 				 MTL_MEM_SS_INFO_QGV_POINT_HIGH(point));
 	dclk = REG_FIELD_GET(MTL_DCLK_MASK, val);
 	sp->dclk = DIV_ROUND_CLOSEST(16667 * dclk, 1000);
@@ -200,29 +205,30 @@ static int mtl_read_qgv_point_info(struct drm_i915_private *dev_priv,
 }
 
 static int
-intel_read_qgv_point_info(struct drm_i915_private *dev_priv,
+intel_read_qgv_point_info(struct intel_display *display,
 			  struct intel_qgv_point *sp,
 			  int point)
 {
-	if (DISPLAY_VER(dev_priv) >= 14)
-		return mtl_read_qgv_point_info(dev_priv, sp, point);
-	else if (IS_DG1(dev_priv))
-		return dg1_mchbar_read_qgv_point_info(dev_priv, sp, point);
+	if (DISPLAY_VER(display) >= 14)
+		return mtl_read_qgv_point_info(display, sp, point);
+	else if (display->platform.dg1)
+		return dg1_mchbar_read_qgv_point_info(display, sp, point);
 	else
-		return icl_pcode_read_qgv_point_info(dev_priv, sp, point);
+		return icl_pcode_read_qgv_point_info(display, sp, point);
 }
 
-static int icl_get_qgv_points(struct drm_i915_private *dev_priv,
+static int icl_get_qgv_points(struct intel_display *display,
 			      struct intel_qgv_info *qi,
 			      bool is_y_tile)
 {
-	const struct dram_info *dram_info = &dev_priv->dram_info;
+	struct drm_i915_private *i915 = to_i915(display->drm);
+	const struct dram_info *dram_info = &i915->dram_info;
 	int i, ret;
 
 	qi->num_points = dram_info->num_qgv_points;
 	qi->num_psf_points = dram_info->num_psf_gv_points;
 
-	if (DISPLAY_VER(dev_priv) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		switch (dram_info->type) {
 		case INTEL_DRAM_DDR4:
 			qi->t_bl = 4;
@@ -244,13 +250,14 @@ static int icl_get_qgv_points(struct drm_i915_private *dev_priv,
 			qi->deinterleave = 4;
 			break;
 		case INTEL_DRAM_GDDR:
+		case INTEL_DRAM_GDDR_ECC:
 			qi->channel_width = 32;
 			break;
 		default:
 			MISSING_CASE(dram_info->type);
 			return -EINVAL;
 		}
-	} else if (DISPLAY_VER(dev_priv) >= 12) {
+	} else if (DISPLAY_VER(display) >= 12) {
 		switch (dram_info->type) {
 		case INTEL_DRAM_DDR4:
 			qi->t_bl = is_y_tile ? 8 : 4;
@@ -265,7 +272,7 @@ static int icl_get_qgv_points(struct drm_i915_private *dev_priv,
 			qi->deinterleave = is_y_tile ? 1 : 2;
 			break;
 		case INTEL_DRAM_LPDDR4:
-			if (IS_ROCKETLAKE(dev_priv)) {
+			if (display->platform.rocketlake) {
 				qi->t_bl = 8;
 				qi->max_numchannels = 4;
 				qi->channel_width = 32;
@@ -284,39 +291,39 @@ static int icl_get_qgv_points(struct drm_i915_private *dev_priv,
 			qi->max_numchannels = 1;
 			break;
 		}
-	} else if (DISPLAY_VER(dev_priv) == 11) {
-		qi->t_bl = dev_priv->dram_info.type == INTEL_DRAM_DDR4 ? 4 : 8;
+	} else if (DISPLAY_VER(display) == 11) {
+		qi->t_bl = dram_info->type == INTEL_DRAM_DDR4 ? 4 : 8;
 		qi->max_numchannels = 1;
 	}
 
-	if (drm_WARN_ON(&dev_priv->drm,
+	if (drm_WARN_ON(display->drm,
 			qi->num_points > ARRAY_SIZE(qi->points)))
 		qi->num_points = ARRAY_SIZE(qi->points);
 
 	for (i = 0; i < qi->num_points; i++) {
 		struct intel_qgv_point *sp = &qi->points[i];
 
-		ret = intel_read_qgv_point_info(dev_priv, sp, i);
+		ret = intel_read_qgv_point_info(display, sp, i);
 		if (ret) {
-			drm_dbg_kms(&dev_priv->drm, "Could not read QGV %d info\n", i);
+			drm_dbg_kms(display->drm, "Could not read QGV %d info\n", i);
 			return ret;
 		}
 
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "QGV %d: DCLK=%d tRP=%d tRDPRE=%d tRAS=%d tRCD=%d tRC=%d\n",
 			    i, sp->dclk, sp->t_rp, sp->t_rdpre, sp->t_ras,
 			    sp->t_rcd, sp->t_rc);
 	}
 
 	if (qi->num_psf_points > 0) {
-		ret = adls_pcode_read_psf_gv_point_info(dev_priv, qi->psf_points);
+		ret = adls_pcode_read_psf_gv_point_info(display, qi->psf_points);
 		if (ret) {
-			drm_err(&dev_priv->drm, "Failed to read PSF point data; PSF points will not be considered in bandwidth calculations.\n");
+			drm_err(display->drm, "Failed to read PSF point data; PSF points will not be considered in bandwidth calculations.\n");
 			qi->num_psf_points = 0;
 		}
 
 		for (i = 0; i < qi->num_psf_points; i++)
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "PSF GV %d: CLK=%d \n",
 				    i, qi->psf_points[i].clk);
 	}
@@ -398,20 +405,34 @@ static const struct intel_sa_info xe2_hpd_sa_info = {
 	/* Other values not used by simplified algorithm */
 };
 
-static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa)
+static const struct intel_sa_info xe2_hpd_ecc_sa_info = {
+	.derating = 45,
+	.deprogbwlimit = 53,
+	/* Other values not used by simplified algorithm */
+};
+
+static const struct intel_sa_info xe3lpd_sa_info = {
+	.deburst = 32,
+	.deprogbwlimit = 65, /* GB/s */
+	.displayrtids = 256,
+	.derating = 10,
+};
+
+static int icl_get_bw_info(struct intel_display *display, const struct intel_sa_info *sa)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_qgv_info qi = {};
 	bool is_y_tile = true; /* assume y tile may be used */
-	int num_channels = max_t(u8, 1, dev_priv->dram_info.num_channels);
+	int num_channels = max_t(u8, 1, i915->dram_info.num_channels);
 	int ipqdepth, ipqdepthpch = 16;
 	int dclk_max;
 	int maxdebw;
-	int num_groups = ARRAY_SIZE(dev_priv->display.bw.max);
+	int num_groups = ARRAY_SIZE(display->bw.max);
 	int i, ret;
 
-	ret = icl_get_qgv_points(dev_priv, &qi, is_y_tile);
+	ret = icl_get_qgv_points(display, &qi, is_y_tile);
 	if (ret) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Failed to get memory subsystem information, ignoring bandwidth limits");
 		return ret;
 	}
@@ -422,7 +443,7 @@ static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 	qi.deinterleave = DIV_ROUND_UP(num_channels, is_y_tile ? 4 : 2);
 
 	for (i = 0; i < num_groups; i++) {
-		struct intel_bw_info *bi = &dev_priv->display.bw.max[i];
+		struct intel_bw_info *bi = &display->bw.max[i];
 		int clpchgroup;
 		int j;
 
@@ -449,7 +470,7 @@ static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 			bi->deratedbw[j] = min(maxdebw,
 					       bw * (100 - sa->derating) / 100);
 
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "BW%d / QGV %d: num_planes=%d deratedbw=%u\n",
 				    i, j, bi->num_planes, bi->deratedbw[j]);
 		}
@@ -460,44 +481,45 @@ static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 	 * as it will fail and pointless anyway.
 	 */
 	if (qi.num_points == 1)
-		dev_priv->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+		display->sagv.status = I915_SAGV_NOT_CONTROLLED;
 	else
-		dev_priv->display.sagv.status = I915_SAGV_ENABLED;
+		display->sagv.status = I915_SAGV_ENABLED;
 
 	return 0;
 }
 
-static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa)
+static int tgl_get_bw_info(struct intel_display *display, const struct intel_sa_info *sa)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_qgv_info qi = {};
-	const struct dram_info *dram_info = &dev_priv->dram_info;
+	const struct dram_info *dram_info = &i915->dram_info;
 	bool is_y_tile = true; /* assume y tile may be used */
-	int num_channels = max_t(u8, 1, dev_priv->dram_info.num_channels);
+	int num_channels = max_t(u8, 1, dram_info->num_channels);
 	int ipqdepth, ipqdepthpch = 16;
 	int dclk_max;
 	int maxdebw, peakbw;
 	int clperchgroup;
-	int num_groups = ARRAY_SIZE(dev_priv->display.bw.max);
+	int num_groups = ARRAY_SIZE(display->bw.max);
 	int i, ret;
 
-	ret = icl_get_qgv_points(dev_priv, &qi, is_y_tile);
+	ret = icl_get_qgv_points(display, &qi, is_y_tile);
 	if (ret) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Failed to get memory subsystem information, ignoring bandwidth limits");
 		return ret;
 	}
 
-	if (DISPLAY_VER(dev_priv) < 14 &&
+	if (DISPLAY_VER(display) < 14 &&
 	    (dram_info->type == INTEL_DRAM_LPDDR4 || dram_info->type == INTEL_DRAM_LPDDR5))
 		num_channels *= 2;
 
 	qi.deinterleave = qi.deinterleave ? : DIV_ROUND_UP(num_channels, is_y_tile ? 4 : 2);
 
-	if (num_channels < qi.max_numchannels && DISPLAY_VER(dev_priv) >= 12)
+	if (num_channels < qi.max_numchannels && DISPLAY_VER(display) >= 12)
 		qi.deinterleave = max(DIV_ROUND_UP(qi.deinterleave, 2), 1);
 
-	if (DISPLAY_VER(dev_priv) >= 12 && num_channels > qi.max_numchannels)
-		drm_warn(&dev_priv->drm, "Number of channels exceeds max number of channels.");
+	if (DISPLAY_VER(display) >= 12 && num_channels > qi.max_numchannels)
+		drm_warn(display->drm, "Number of channels exceeds max number of channels.");
 	if (qi.max_numchannels != 0)
 		num_channels = min_t(u8, num_channels, qi.max_numchannels);
 
@@ -514,7 +536,7 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 	clperchgroup = 4 * DIV_ROUND_UP(8, num_channels) * qi.deinterleave;
 
 	for (i = 0; i < num_groups; i++) {
-		struct intel_bw_info *bi = &dev_priv->display.bw.max[i];
+		struct intel_bw_info *bi = &display->bw.max[i];
 		struct intel_bw_info *bi_next;
 		int clpchgroup;
 		int j;
@@ -522,7 +544,7 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 		clpchgroup = (sa->deburst * qi.deinterleave / num_channels) << i;
 
 		if (i < num_groups - 1) {
-			bi_next = &dev_priv->display.bw.max[i + 1];
+			bi_next = &display->bw.max[i + 1];
 
 			if (clpchgroup < clperchgroup)
 				bi_next->num_planes = (ipqdepth - clpchgroup) /
@@ -554,7 +576,7 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 							  num_channels *
 							  qi.channel_width, 8);
 
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "BW%d / QGV %d: num_planes=%d deratedbw=%u peakbw: %u\n",
 				    i, j, bi->num_planes, bi->deratedbw[j],
 				    bi->peakbw[j]);
@@ -565,7 +587,7 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 
 			bi->psf_bw[j] = adl_calc_psf_bw(sp->clk);
 
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "BW%d / PSF GV %d: num_planes=%d bw=%u\n",
 				    i, j, bi->num_planes, bi->psf_bw[j]);
 		}
@@ -577,17 +599,17 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 	 * as it will fail and pointless anyway.
 	 */
 	if (qi.num_points == 1)
-		dev_priv->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+		display->sagv.status = I915_SAGV_NOT_CONTROLLED;
 	else
-		dev_priv->display.sagv.status = I915_SAGV_ENABLED;
+		display->sagv.status = I915_SAGV_ENABLED;
 
 	return 0;
 }
 
-static void dg2_get_bw_info(struct drm_i915_private *i915)
+static void dg2_get_bw_info(struct intel_display *display)
 {
-	unsigned int deratedbw = IS_DG2_G11(i915) ? 38000 : 50000;
-	int num_groups = ARRAY_SIZE(i915->display.bw.max);
+	unsigned int deratedbw = display->platform.dg2_g11 ? 38000 : 50000;
+	int num_groups = ARRAY_SIZE(display->bw.max);
 	int i;
 
 	/*
@@ -598,7 +620,7 @@ static void dg2_get_bw_info(struct drm_i915_private *i915)
 	 * whereas DG2-G11 platforms have 38 GB/s.
 	 */
 	for (i = 0; i < num_groups; i++) {
-		struct intel_bw_info *bi = &i915->display.bw.max[i];
+		struct intel_bw_info *bi = &display->bw.max[i];
 
 		bi->num_planes = 1;
 		/* Need only one dummy QGV point per group */
@@ -606,20 +628,21 @@ static void dg2_get_bw_info(struct drm_i915_private *i915)
 		bi->deratedbw[0] = deratedbw;
 	}
 
-	i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+	display->sagv.status = I915_SAGV_NOT_CONTROLLED;
 }
 
-static int xe2_hpd_get_bw_info(struct drm_i915_private *i915,
+static int xe2_hpd_get_bw_info(struct intel_display *display,
 			       const struct intel_sa_info *sa)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_qgv_info qi = {};
 	int num_channels = i915->dram_info.num_channels;
 	int peakbw, maxdebw;
 	int ret, i;
 
-	ret = icl_get_qgv_points(i915, &qi, true);
+	ret = icl_get_qgv_points(display, &qi, true);
 	if (ret) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Failed to get memory subsystem information, ignoring bandwidth limits");
 		return ret;
 	}
@@ -631,33 +654,33 @@ static int xe2_hpd_get_bw_info(struct drm_i915_private *i915,
 		const struct intel_qgv_point *point = &qi.points[i];
 		int bw = num_channels * (qi.channel_width / 8) * point->dclk;
 
-		i915->display.bw.max[0].deratedbw[i] =
+		display->bw.max[0].deratedbw[i] =
 			min(maxdebw, (100 - sa->derating) * bw / 100);
-		i915->display.bw.max[0].peakbw[i] = bw;
+		display->bw.max[0].peakbw[i] = bw;
 
-		drm_dbg_kms(&i915->drm, "QGV %d: deratedbw=%u peakbw: %u\n",
-			    i, i915->display.bw.max[0].deratedbw[i],
-			    i915->display.bw.max[0].peakbw[i]);
+		drm_dbg_kms(display->drm, "QGV %d: deratedbw=%u peakbw: %u\n",
+			    i, display->bw.max[0].deratedbw[i],
+			    display->bw.max[0].peakbw[i]);
 	}
 
 	/* Bandwidth does not depend on # of planes; set all groups the same */
-	i915->display.bw.max[0].num_planes = 1;
-	i915->display.bw.max[0].num_qgv_points = qi.num_points;
-	for (i = 1; i < ARRAY_SIZE(i915->display.bw.max); i++)
-		memcpy(&i915->display.bw.max[i], &i915->display.bw.max[0],
-		       sizeof(i915->display.bw.max[0]));
+	display->bw.max[0].num_planes = 1;
+	display->bw.max[0].num_qgv_points = qi.num_points;
+	for (i = 1; i < ARRAY_SIZE(display->bw.max); i++)
+		memcpy(&display->bw.max[i], &display->bw.max[0],
+		       sizeof(display->bw.max[0]));
 
 	/*
 	 * Xe2_HPD should always have exactly two QGV points representing
 	 * battery and plugged-in operation.
 	 */
-	drm_WARN_ON(&i915->drm, qi.num_points != 2);
-	i915->display.sagv.status = I915_SAGV_ENABLED;
+	drm_WARN_ON(display->drm, qi.num_points != 2);
+	display->sagv.status = I915_SAGV_ENABLED;
 
 	return 0;
 }
 
-static unsigned int icl_max_bw_index(struct drm_i915_private *dev_priv,
+static unsigned int icl_max_bw_index(struct intel_display *display,
 				     int num_planes, int qgv_point)
 {
 	int i;
@@ -667,9 +690,9 @@ static unsigned int icl_max_bw_index(struct drm_i915_private *dev_priv,
 	 */
 	num_planes = max(1, num_planes);
 
-	for (i = 0; i < ARRAY_SIZE(dev_priv->display.bw.max); i++) {
+	for (i = 0; i < ARRAY_SIZE(display->bw.max); i++) {
 		const struct intel_bw_info *bi =
-			&dev_priv->display.bw.max[i];
+			&display->bw.max[i];
 
 		/*
 		 * Pcode will not expose all QGV points when
@@ -685,7 +708,7 @@ static unsigned int icl_max_bw_index(struct drm_i915_private *dev_priv,
 	return UINT_MAX;
 }
 
-static unsigned int tgl_max_bw_index(struct drm_i915_private *dev_priv,
+static unsigned int tgl_max_bw_index(struct intel_display *display,
 				     int num_planes, int qgv_point)
 {
 	int i;
@@ -695,9 +718,9 @@ static unsigned int tgl_max_bw_index(struct drm_i915_private *dev_priv,
 	 */
 	num_planes = max(1, num_planes);
 
-	for (i = ARRAY_SIZE(dev_priv->display.bw.max) - 1; i >= 0; i--) {
+	for (i = ARRAY_SIZE(display->bw.max) - 1; i >= 0; i--) {
 		const struct intel_bw_info *bi =
-			&dev_priv->display.bw.max[i];
+			&display->bw.max[i];
 
 		/*
 		 * Pcode will not expose all QGV points when
@@ -713,52 +736,59 @@ static unsigned int tgl_max_bw_index(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static unsigned int adl_psf_bw(struct drm_i915_private *dev_priv,
+static unsigned int adl_psf_bw(struct intel_display *display,
 			       int psf_gv_point)
 {
 	const struct intel_bw_info *bi =
-			&dev_priv->display.bw.max[0];
+			&display->bw.max[0];
 
 	return bi->psf_bw[psf_gv_point];
 }
 
-static unsigned int icl_qgv_bw(struct drm_i915_private *i915,
+static unsigned int icl_qgv_bw(struct intel_display *display,
 			       int num_active_planes, int qgv_point)
 {
 	unsigned int idx;
 
-	if (DISPLAY_VER(i915) >= 12)
-		idx = tgl_max_bw_index(i915, num_active_planes, qgv_point);
+	if (DISPLAY_VER(display) >= 12)
+		idx = tgl_max_bw_index(display, num_active_planes, qgv_point);
 	else
-		idx = icl_max_bw_index(i915, num_active_planes, qgv_point);
+		idx = icl_max_bw_index(display, num_active_planes, qgv_point);
 
-	if (idx >= ARRAY_SIZE(i915->display.bw.max))
+	if (idx >= ARRAY_SIZE(display->bw.max))
 		return 0;
 
-	return i915->display.bw.max[idx].deratedbw[qgv_point];
+	return display->bw.max[idx].deratedbw[qgv_point];
 }
 
-void intel_bw_init_hw(struct drm_i915_private *dev_priv)
+void intel_bw_init_hw(struct intel_display *display)
 {
-	if (!HAS_DISPLAY(dev_priv))
+	const struct dram_info *dram_info = &to_i915(display->drm)->dram_info;
+
+	if (!HAS_DISPLAY(display))
 		return;
 
-	if (DISPLAY_VERx100(dev_priv) >= 1401 && IS_DGFX(dev_priv))
-		xe2_hpd_get_bw_info(dev_priv, &xe2_hpd_sa_info);
-	else if (DISPLAY_VER(dev_priv) >= 14)
-		tgl_get_bw_info(dev_priv, &mtl_sa_info);
-	else if (IS_DG2(dev_priv))
-		dg2_get_bw_info(dev_priv);
-	else if (IS_ALDERLAKE_P(dev_priv))
-		tgl_get_bw_info(dev_priv, &adlp_sa_info);
-	else if (IS_ALDERLAKE_S(dev_priv))
-		tgl_get_bw_info(dev_priv, &adls_sa_info);
-	else if (IS_ROCKETLAKE(dev_priv))
-		tgl_get_bw_info(dev_priv, &rkl_sa_info);
-	else if (DISPLAY_VER(dev_priv) == 12)
-		tgl_get_bw_info(dev_priv, &tgl_sa_info);
-	else if (DISPLAY_VER(dev_priv) == 11)
-		icl_get_bw_info(dev_priv, &icl_sa_info);
+	if (DISPLAY_VER(display) >= 30)
+		tgl_get_bw_info(display, &xe3lpd_sa_info);
+	else if (DISPLAY_VERx100(display) >= 1401 && display->platform.dgfx &&
+		 dram_info->type == INTEL_DRAM_GDDR_ECC)
+		xe2_hpd_get_bw_info(display, &xe2_hpd_ecc_sa_info);
+	else if (DISPLAY_VERx100(display) >= 1401 && display->platform.dgfx)
+		xe2_hpd_get_bw_info(display, &xe2_hpd_sa_info);
+	else if (DISPLAY_VER(display) >= 14)
+		tgl_get_bw_info(display, &mtl_sa_info);
+	else if (display->platform.dg2)
+		dg2_get_bw_info(display);
+	else if (display->platform.alderlake_p)
+		tgl_get_bw_info(display, &adlp_sa_info);
+	else if (display->platform.alderlake_s)
+		tgl_get_bw_info(display, &adls_sa_info);
+	else if (display->platform.rocketlake)
+		tgl_get_bw_info(display, &rkl_sa_info);
+	else if (DISPLAY_VER(display) == 12)
+		tgl_get_bw_info(display, &tgl_sa_info);
+	else if (DISPLAY_VER(display) == 11)
+		icl_get_bw_info(display, &icl_sa_info);
 }
 
 static unsigned int intel_bw_crtc_num_active_planes(const struct intel_crtc_state *crtc_state)
@@ -772,8 +802,8 @@ static unsigned int intel_bw_crtc_num_active_planes(const struct intel_crtc_stat
 
 static unsigned int intel_bw_crtc_data_rate(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	unsigned int data_rate = 0;
 	enum plane_id plane_id;
 
@@ -787,7 +817,7 @@ static unsigned int intel_bw_crtc_data_rate(const struct intel_crtc_state *crtc_
 
 		data_rate += crtc_state->data_rate[plane_id];
 
-		if (DISPLAY_VER(i915) < 11)
+		if (DISPLAY_VER(display) < 11)
 			data_rate += crtc_state->data_rate_y[plane_id];
 	}
 
@@ -795,39 +825,38 @@ static unsigned int intel_bw_crtc_data_rate(const struct intel_crtc_state *crtc_
 }
 
 /* "Maximum Pipe Read Bandwidth" */
-static int intel_bw_crtc_min_cdclk(const struct intel_crtc_state *crtc_state)
+static int intel_bw_crtc_min_cdclk(struct intel_display *display,
+				   unsigned int data_rate)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-
-	if (DISPLAY_VER(i915) < 12)
+	if (DISPLAY_VER(display) < 12)
 		return 0;
 
-	return DIV_ROUND_UP_ULL(mul_u32_u32(intel_bw_crtc_data_rate(crtc_state), 10), 512);
+	return DIV_ROUND_UP_ULL(mul_u32_u32(data_rate, 10), 512);
 }
 
-static unsigned int intel_bw_num_active_planes(struct drm_i915_private *dev_priv,
+static unsigned int intel_bw_num_active_planes(struct intel_display *display,
 					       const struct intel_bw_state *bw_state)
 {
 	unsigned int num_active_planes = 0;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe)
+	for_each_pipe(display, pipe)
 		num_active_planes += bw_state->num_active_planes[pipe];
 
 	return num_active_planes;
 }
 
-static unsigned int intel_bw_data_rate(struct drm_i915_private *dev_priv,
+static unsigned int intel_bw_data_rate(struct intel_display *display,
 				       const struct intel_bw_state *bw_state)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	unsigned int data_rate = 0;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe)
+	for_each_pipe(display, pipe)
 		data_rate += bw_state->data_rate[pipe];
 
-	if (DISPLAY_VER(dev_priv) >= 13 && i915_vtd_active(dev_priv))
+	if (DISPLAY_VER(display) >= 13 && i915_vtd_active(i915))
 		data_rate = DIV_ROUND_UP(data_rate * 105, 100);
 
 	return data_rate;
@@ -836,10 +865,10 @@ static unsigned int intel_bw_data_rate(struct drm_i915_private *dev_priv,
 struct intel_bw_state *
 intel_atomic_get_old_bw_state(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_global_state *bw_state;
 
-	bw_state = intel_atomic_get_old_global_obj_state(state, &dev_priv->display.bw.obj);
+	bw_state = intel_atomic_get_old_global_obj_state(state, &display->bw.obj);
 
 	return to_intel_bw_state(bw_state);
 }
@@ -847,10 +876,10 @@ intel_atomic_get_old_bw_state(struct intel_atomic_state *state)
 struct intel_bw_state *
 intel_atomic_get_new_bw_state(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_global_state *bw_state;
 
-	bw_state = intel_atomic_get_new_global_obj_state(state, &dev_priv->display.bw.obj);
+	bw_state = intel_atomic_get_new_global_obj_state(state, &display->bw.obj);
 
 	return to_intel_bw_state(bw_state);
 }
@@ -858,27 +887,27 @@ intel_atomic_get_new_bw_state(struct intel_atomic_state *state)
 struct intel_bw_state *
 intel_atomic_get_bw_state(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_global_state *bw_state;
 
-	bw_state = intel_atomic_get_global_obj_state(state, &dev_priv->display.bw.obj);
+	bw_state = intel_atomic_get_global_obj_state(state, &display->bw.obj);
 	if (IS_ERR(bw_state))
 		return ERR_CAST(bw_state);
 
 	return to_intel_bw_state(bw_state);
 }
 
-static unsigned int icl_max_bw_qgv_point_mask(struct drm_i915_private *i915,
+static unsigned int icl_max_bw_qgv_point_mask(struct intel_display *display,
 					      int num_active_planes)
 {
-	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	unsigned int num_qgv_points = display->bw.max[0].num_qgv_points;
 	unsigned int max_bw_point = 0;
 	unsigned int max_bw = 0;
 	int i;
 
 	for (i = 0; i < num_qgv_points; i++) {
 		unsigned int max_data_rate =
-			icl_qgv_bw(i915, num_active_planes, i);
+			icl_qgv_bw(display, num_active_planes, i);
 
 		/*
 		 * We need to know which qgv point gives us
@@ -897,23 +926,23 @@ static unsigned int icl_max_bw_qgv_point_mask(struct drm_i915_private *i915,
 	return max_bw_point;
 }
 
-static u16 icl_prepare_qgv_points_mask(struct drm_i915_private *i915,
+static u16 icl_prepare_qgv_points_mask(struct intel_display *display,
 				       unsigned int qgv_points,
 				       unsigned int psf_points)
 {
 	return ~(ICL_PCODE_REQ_QGV_PT(qgv_points) |
-		 ADLS_PCODE_REQ_PSF_PT(psf_points)) & icl_qgv_points_mask(i915);
+		 ADLS_PCODE_REQ_PSF_PT(psf_points)) & icl_qgv_points_mask(display);
 }
 
-static unsigned int icl_max_bw_psf_gv_point_mask(struct drm_i915_private *i915)
+static unsigned int icl_max_bw_psf_gv_point_mask(struct intel_display *display)
 {
-	unsigned int num_psf_gv_points = i915->display.bw.max[0].num_psf_gv_points;
+	unsigned int num_psf_gv_points = display->bw.max[0].num_psf_gv_points;
 	unsigned int max_bw_point_mask = 0;
 	unsigned int max_bw = 0;
 	int i;
 
 	for (i = 0; i < num_psf_gv_points; i++) {
-		unsigned int max_data_rate = adl_psf_bw(i915, i);
+		unsigned int max_data_rate = adl_psf_bw(display, i);
 
 		if (max_data_rate > max_bw) {
 			max_bw_point_mask = BIT(i);
@@ -926,29 +955,29 @@ static unsigned int icl_max_bw_psf_gv_point_mask(struct drm_i915_private *i915)
 	return max_bw_point_mask;
 }
 
-static void icl_force_disable_sagv(struct drm_i915_private *i915,
+static void icl_force_disable_sagv(struct intel_display *display,
 				   struct intel_bw_state *bw_state)
 {
-	unsigned int qgv_points = icl_max_bw_qgv_point_mask(i915, 0);
-	unsigned int psf_points = icl_max_bw_psf_gv_point_mask(i915);
+	unsigned int qgv_points = icl_max_bw_qgv_point_mask(display, 0);
+	unsigned int psf_points = icl_max_bw_psf_gv_point_mask(display);
 
-	bw_state->qgv_points_mask = icl_prepare_qgv_points_mask(i915,
+	bw_state->qgv_points_mask = icl_prepare_qgv_points_mask(display,
 								qgv_points,
 								psf_points);
 
-	drm_dbg_kms(&i915->drm, "Forcing SAGV disable: mask 0x%x\n",
+	drm_dbg_kms(display->drm, "Forcing SAGV disable: mask 0x%x\n",
 		    bw_state->qgv_points_mask);
 
-	icl_pcode_restrict_qgv_points(i915, bw_state->qgv_points_mask);
+	icl_pcode_restrict_qgv_points(display, bw_state->qgv_points_mask);
 }
 
-static int mtl_find_qgv_points(struct drm_i915_private *i915,
+static int mtl_find_qgv_points(struct intel_display *display,
 			       unsigned int data_rate,
 			       unsigned int num_active_planes,
 			       struct intel_bw_state *new_bw_state)
 {
 	unsigned int best_rate = UINT_MAX;
-	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	unsigned int num_qgv_points = display->bw.max[0].num_qgv_points;
 	unsigned int qgv_peak_bw  = 0;
 	int i;
 	int ret;
@@ -962,9 +991,9 @@ static int mtl_find_qgv_points(struct drm_i915_private *i915,
 	 * for qgv peak bw in PM Demand request. So assign UINT_MAX if SAGV is
 	 * not enabled. PM Demand code will clamp the value for the register
 	 */
-	if (!intel_can_enable_sagv(i915, new_bw_state)) {
+	if (!intel_can_enable_sagv(display, new_bw_state)) {
 		new_bw_state->qgv_point_peakbw = U16_MAX;
-		drm_dbg_kms(&i915->drm, "No SAGV, use UINT_MAX as peak bw.");
+		drm_dbg_kms(display->drm, "No SAGV, use UINT_MAX as peak bw.");
 		return 0;
 	}
 
@@ -974,27 +1003,27 @@ static int mtl_find_qgv_points(struct drm_i915_private *i915,
 	 */
 	for (i = 0; i < num_qgv_points; i++) {
 		unsigned int bw_index =
-			tgl_max_bw_index(i915, num_active_planes, i);
+			tgl_max_bw_index(display, num_active_planes, i);
 		unsigned int max_data_rate;
 
-		if (bw_index >= ARRAY_SIZE(i915->display.bw.max))
+		if (bw_index >= ARRAY_SIZE(display->bw.max))
 			continue;
 
-		max_data_rate = i915->display.bw.max[bw_index].deratedbw[i];
+		max_data_rate = display->bw.max[bw_index].deratedbw[i];
 
 		if (max_data_rate < data_rate)
 			continue;
 
 		if (max_data_rate - data_rate < best_rate) {
 			best_rate = max_data_rate - data_rate;
-			qgv_peak_bw = i915->display.bw.max[bw_index].peakbw[i];
+			qgv_peak_bw = display->bw.max[bw_index].peakbw[i];
 		}
 
-		drm_dbg_kms(&i915->drm, "QGV point %d: max bw %d required %d qgv_peak_bw: %d\n",
+		drm_dbg_kms(display->drm, "QGV point %d: max bw %d required %d qgv_peak_bw: %d\n",
 			    i, max_data_rate, data_rate, qgv_peak_bw);
 	}
 
-	drm_dbg_kms(&i915->drm, "Matching peaks QGV bw: %d for required data rate: %d\n",
+	drm_dbg_kms(display->drm, "Matching peaks QGV bw: %d for required data rate: %d\n",
 		    qgv_peak_bw, data_rate);
 
 	/*
@@ -1002,7 +1031,7 @@ static int mtl_find_qgv_points(struct drm_i915_private *i915,
 	 * satisfying the required data rate is found
 	 */
 	if (qgv_peak_bw == 0) {
-		drm_dbg_kms(&i915->drm, "No QGV points for bw %d for display configuration(%d active planes).\n",
+		drm_dbg_kms(display->drm, "No QGV points for bw %d for display configuration(%d active planes).\n",
 			    data_rate, num_active_planes);
 		return -EINVAL;
 	}
@@ -1013,14 +1042,14 @@ static int mtl_find_qgv_points(struct drm_i915_private *i915,
 	return 0;
 }
 
-static int icl_find_qgv_points(struct drm_i915_private *i915,
+static int icl_find_qgv_points(struct intel_display *display,
 			       unsigned int data_rate,
 			       unsigned int num_active_planes,
 			       const struct intel_bw_state *old_bw_state,
 			       struct intel_bw_state *new_bw_state)
 {
-	unsigned int num_psf_gv_points = i915->display.bw.max[0].num_psf_gv_points;
-	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	unsigned int num_psf_gv_points = display->bw.max[0].num_psf_gv_points;
+	unsigned int num_qgv_points = display->bw.max[0].num_qgv_points;
 	u16 psf_points = 0;
 	u16 qgv_points = 0;
 	int i;
@@ -1031,22 +1060,22 @@ static int icl_find_qgv_points(struct drm_i915_private *i915,
 		return ret;
 
 	for (i = 0; i < num_qgv_points; i++) {
-		unsigned int max_data_rate = icl_qgv_bw(i915,
+		unsigned int max_data_rate = icl_qgv_bw(display,
 							num_active_planes, i);
 		if (max_data_rate >= data_rate)
 			qgv_points |= BIT(i);
 
-		drm_dbg_kms(&i915->drm, "QGV point %d: max bw %d required %d\n",
+		drm_dbg_kms(display->drm, "QGV point %d: max bw %d required %d\n",
 			    i, max_data_rate, data_rate);
 	}
 
 	for (i = 0; i < num_psf_gv_points; i++) {
-		unsigned int max_data_rate = adl_psf_bw(i915, i);
+		unsigned int max_data_rate = adl_psf_bw(display, i);
 
 		if (max_data_rate >= data_rate)
 			psf_points |= BIT(i);
 
-		drm_dbg_kms(&i915->drm, "PSF GV point %d: max bw %d"
+		drm_dbg_kms(display->drm, "PSF GV point %d: max bw %d"
 			    " required %d\n",
 			    i, max_data_rate, data_rate);
 	}
@@ -1057,14 +1086,14 @@ static int icl_find_qgv_points(struct drm_i915_private *i915,
 	 * reasons.
 	 */
 	if (qgv_points == 0) {
-		drm_dbg_kms(&i915->drm, "No QGV points provide sufficient memory"
+		drm_dbg_kms(display->drm, "No QGV points provide sufficient memory"
 			    " bandwidth %d for display configuration(%d active planes).\n",
 			    data_rate, num_active_planes);
 		return -EINVAL;
 	}
 
 	if (num_psf_gv_points > 0 && psf_points == 0) {
-		drm_dbg_kms(&i915->drm, "No PSF GV points provide sufficient memory"
+		drm_dbg_kms(display->drm, "No PSF GV points provide sufficient memory"
 			    " bandwidth %d for display configuration(%d active planes).\n",
 			    data_rate, num_active_planes);
 		return -EINVAL;
@@ -1075,9 +1104,9 @@ static int icl_find_qgv_points(struct drm_i915_private *i915,
 	 * we can't enable SAGV due to the increased memory latency it may
 	 * cause.
 	 */
-	if (!intel_can_enable_sagv(i915, new_bw_state)) {
-		qgv_points = icl_max_bw_qgv_point_mask(i915, num_active_planes);
-		drm_dbg_kms(&i915->drm, "No SAGV, using single QGV point mask 0x%x\n",
+	if (!intel_can_enable_sagv(display, new_bw_state)) {
+		qgv_points = icl_max_bw_qgv_point_mask(display, num_active_planes);
+		drm_dbg_kms(display->drm, "No SAGV, using single QGV point mask 0x%x\n",
 			    qgv_points);
 	}
 
@@ -1085,7 +1114,7 @@ static int icl_find_qgv_points(struct drm_i915_private *i915,
 	 * We store the ones which need to be masked as that is what PCode
 	 * actually accepts as a parameter.
 	 */
-	new_bw_state->qgv_points_mask = icl_prepare_qgv_points_mask(i915,
+	new_bw_state->qgv_points_mask = icl_prepare_qgv_points_mask(display,
 								    qgv_points,
 								    psf_points);
 	/*
@@ -1101,80 +1130,90 @@ static int icl_find_qgv_points(struct drm_i915_private *i915,
 	return 0;
 }
 
-static int intel_bw_check_qgv_points(struct drm_i915_private *i915,
+static int intel_bw_check_qgv_points(struct intel_display *display,
 				     const struct intel_bw_state *old_bw_state,
 				     struct intel_bw_state *new_bw_state)
 {
-	unsigned int data_rate = intel_bw_data_rate(i915, new_bw_state);
+	unsigned int data_rate = intel_bw_data_rate(display, new_bw_state);
 	unsigned int num_active_planes =
-			intel_bw_num_active_planes(i915, new_bw_state);
+			intel_bw_num_active_planes(display, new_bw_state);
 
 	data_rate = DIV_ROUND_UP(data_rate, 1000);
 
-	if (DISPLAY_VER(i915) >= 14)
-		return mtl_find_qgv_points(i915, data_rate, num_active_planes,
+	if (DISPLAY_VER(display) >= 14)
+		return mtl_find_qgv_points(display, data_rate, num_active_planes,
 					   new_bw_state);
 	else
-		return icl_find_qgv_points(i915, data_rate, num_active_planes,
+		return icl_find_qgv_points(display, data_rate, num_active_planes,
 					   old_bw_state, new_bw_state);
 }
 
-static bool intel_bw_state_changed(struct drm_i915_private *i915,
+static bool intel_dbuf_bw_changed(struct intel_display *display,
+				  const struct intel_dbuf_bw *old_dbuf_bw,
+				  const struct intel_dbuf_bw *new_dbuf_bw)
+{
+	enum dbuf_slice slice;
+
+	for_each_dbuf_slice(display, slice) {
+		if (old_dbuf_bw->max_bw[slice] != new_dbuf_bw->max_bw[slice] ||
+		    old_dbuf_bw->active_planes[slice] != new_dbuf_bw->active_planes[slice])
+			return true;
+	}
+
+	return false;
+}
+
+static bool intel_bw_state_changed(struct intel_display *display,
 				   const struct intel_bw_state *old_bw_state,
 				   const struct intel_bw_state *new_bw_state)
 {
 	enum pipe pipe;
 
-	for_each_pipe(i915, pipe) {
-		const struct intel_dbuf_bw *old_crtc_bw =
+	for_each_pipe(display, pipe) {
+		const struct intel_dbuf_bw *old_dbuf_bw =
 			&old_bw_state->dbuf_bw[pipe];
-		const struct intel_dbuf_bw *new_crtc_bw =
+		const struct intel_dbuf_bw *new_dbuf_bw =
 			&new_bw_state->dbuf_bw[pipe];
-		enum dbuf_slice slice;
 
-		for_each_dbuf_slice(i915, slice) {
-			if (old_crtc_bw->max_bw[slice] != new_crtc_bw->max_bw[slice] ||
-			    old_crtc_bw->active_planes[slice] != new_crtc_bw->active_planes[slice])
-				return true;
-		}
+		if (intel_dbuf_bw_changed(display, old_dbuf_bw, new_dbuf_bw))
+			return true;
 
-		if (old_bw_state->min_cdclk[pipe] != new_bw_state->min_cdclk[pipe])
+		if (intel_bw_crtc_min_cdclk(display, old_bw_state->data_rate[pipe]) !=
+		    intel_bw_crtc_min_cdclk(display, new_bw_state->data_rate[pipe]))
 			return true;
 	}
 
 	return false;
 }
 
-static void skl_plane_calc_dbuf_bw(struct intel_bw_state *bw_state,
+static void skl_plane_calc_dbuf_bw(struct intel_dbuf_bw *dbuf_bw,
 				   struct intel_crtc *crtc,
 				   enum plane_id plane_id,
 				   const struct skl_ddb_entry *ddb,
 				   unsigned int data_rate)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	struct intel_dbuf_bw *crtc_bw = &bw_state->dbuf_bw[crtc->pipe];
-	unsigned int dbuf_mask = skl_ddb_dbuf_slice_mask(i915, ddb);
+	struct intel_display *display = to_intel_display(crtc);
+	unsigned int dbuf_mask = skl_ddb_dbuf_slice_mask(display, ddb);
 	enum dbuf_slice slice;
 
 	/*
 	 * The arbiter can only really guarantee an
 	 * equal share of the total bw to each plane.
 	 */
-	for_each_dbuf_slice_in_mask(i915, slice, dbuf_mask) {
-		crtc_bw->max_bw[slice] = max(crtc_bw->max_bw[slice], data_rate);
-		crtc_bw->active_planes[slice] |= BIT(plane_id);
+	for_each_dbuf_slice_in_mask(display, slice, dbuf_mask) {
+		dbuf_bw->max_bw[slice] = max(dbuf_bw->max_bw[slice], data_rate);
+		dbuf_bw->active_planes[slice] |= BIT(plane_id);
 	}
 }
 
-static void skl_crtc_calc_dbuf_bw(struct intel_bw_state *bw_state,
+static void skl_crtc_calc_dbuf_bw(struct intel_dbuf_bw *dbuf_bw,
 				  const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	struct intel_dbuf_bw *crtc_bw = &bw_state->dbuf_bw[crtc->pipe];
 	enum plane_id plane_id;
 
-	memset(crtc_bw, 0, sizeof(*crtc_bw));
+	memset(dbuf_bw, 0, sizeof(*dbuf_bw));
 
 	if (!crtc_state->hw.active)
 		return;
@@ -1187,12 +1226,12 @@ static void skl_crtc_calc_dbuf_bw(struct intel_bw_state *bw_state,
 		if (plane_id == PLANE_CURSOR)
 			continue;
 
-		skl_plane_calc_dbuf_bw(bw_state, crtc, plane_id,
+		skl_plane_calc_dbuf_bw(dbuf_bw, crtc, plane_id,
 				       &crtc_state->wm.skl.plane_ddb[plane_id],
 				       crtc_state->data_rate[plane_id]);
 
-		if (DISPLAY_VER(i915) < 11)
-			skl_plane_calc_dbuf_bw(bw_state, crtc, plane_id,
+		if (DISPLAY_VER(display) < 11)
+			skl_plane_calc_dbuf_bw(dbuf_bw, crtc, plane_id,
 					       &crtc_state->wm.skl.plane_ddb_y[plane_id],
 					       crtc_state->data_rate[plane_id]);
 	}
@@ -1200,13 +1239,13 @@ static void skl_crtc_calc_dbuf_bw(struct intel_bw_state *bw_state,
 
 /* "Maximum Data Buffer Bandwidth" */
 static int
-intel_bw_dbuf_min_cdclk(struct drm_i915_private *i915,
+intel_bw_dbuf_min_cdclk(struct intel_display *display,
 			const struct intel_bw_state *bw_state)
 {
 	unsigned int total_max_bw = 0;
 	enum dbuf_slice slice;
 
-	for_each_dbuf_slice(i915, slice) {
+	for_each_dbuf_slice(display, slice) {
 		int num_active_planes = 0;
 		unsigned int max_bw = 0;
 		enum pipe pipe;
@@ -1215,11 +1254,11 @@ intel_bw_dbuf_min_cdclk(struct drm_i915_private *i915,
 		 * The arbiter can only really guarantee an
 		 * equal share of the total bw to each plane.
 		 */
-		for_each_pipe(i915, pipe) {
-			const struct intel_dbuf_bw *crtc_bw = &bw_state->dbuf_bw[pipe];
+		for_each_pipe(display, pipe) {
+			const struct intel_dbuf_bw *dbuf_bw = &bw_state->dbuf_bw[pipe];
 
-			max_bw = max(crtc_bw->max_bw[slice], max_bw);
-			num_active_planes += hweight8(crtc_bw->active_planes[slice]);
+			max_bw = max(dbuf_bw->max_bw[slice], max_bw);
+			num_active_planes += hweight8(dbuf_bw->active_planes[slice]);
 		}
 		max_bw *= num_active_planes;
 
@@ -1229,16 +1268,18 @@ intel_bw_dbuf_min_cdclk(struct drm_i915_private *i915,
 	return DIV_ROUND_UP(total_max_bw, 64);
 }
 
-int intel_bw_min_cdclk(struct drm_i915_private *i915,
+int intel_bw_min_cdclk(struct intel_display *display,
 		       const struct intel_bw_state *bw_state)
 {
 	enum pipe pipe;
 	int min_cdclk;
 
-	min_cdclk = intel_bw_dbuf_min_cdclk(i915, bw_state);
+	min_cdclk = intel_bw_dbuf_min_cdclk(display, bw_state);
 
-	for_each_pipe(i915, pipe)
-		min_cdclk = max(min_cdclk, bw_state->min_cdclk[pipe]);
+	for_each_pipe(display, pipe)
+		min_cdclk = max(min_cdclk,
+				intel_bw_crtc_min_cdclk(display,
+							bw_state->data_rate[pipe]));
 
 	return min_cdclk;
 }
@@ -1246,42 +1287,49 @@ int intel_bw_min_cdclk(struct drm_i915_private *i915,
 int intel_bw_calc_min_cdclk(struct intel_atomic_state *state,
 			    bool *need_cdclk_calc)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_bw_state *new_bw_state = NULL;
 	const struct intel_bw_state *old_bw_state = NULL;
 	const struct intel_cdclk_state *cdclk_state;
-	const struct intel_crtc_state *crtc_state;
+	const struct intel_crtc_state *old_crtc_state;
+	const struct intel_crtc_state *new_crtc_state;
 	int old_min_cdclk, new_min_cdclk;
 	struct intel_crtc *crtc;
 	int i;
 
-	if (DISPLAY_VER(dev_priv) < 9)
+	if (DISPLAY_VER(display) < 9)
 		return 0;
 
-	for_each_new_intel_crtc_in_state(state, crtc, crtc_state, i) {
+	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
+					    new_crtc_state, i) {
+		struct intel_dbuf_bw old_dbuf_bw, new_dbuf_bw;
+
+		skl_crtc_calc_dbuf_bw(&old_dbuf_bw, old_crtc_state);
+		skl_crtc_calc_dbuf_bw(&new_dbuf_bw, new_crtc_state);
+
+		if (!intel_dbuf_bw_changed(display, &old_dbuf_bw, &new_dbuf_bw))
+			continue;
+
 		new_bw_state = intel_atomic_get_bw_state(state);
 		if (IS_ERR(new_bw_state))
 			return PTR_ERR(new_bw_state);
 
 		old_bw_state = intel_atomic_get_old_bw_state(state);
 
-		skl_crtc_calc_dbuf_bw(new_bw_state, crtc_state);
-
-		new_bw_state->min_cdclk[crtc->pipe] =
-			intel_bw_crtc_min_cdclk(crtc_state);
+		new_bw_state->dbuf_bw[crtc->pipe] = new_dbuf_bw;
 	}
 
 	if (!old_bw_state)
 		return 0;
 
-	if (intel_bw_state_changed(dev_priv, old_bw_state, new_bw_state)) {
+	if (intel_bw_state_changed(display, old_bw_state, new_bw_state)) {
 		int ret = intel_atomic_lock_global_state(&new_bw_state->base);
 		if (ret)
 			return ret;
 	}
 
-	old_min_cdclk = intel_bw_min_cdclk(dev_priv, old_bw_state);
-	new_min_cdclk = intel_bw_min_cdclk(dev_priv, new_bw_state);
+	old_min_cdclk = intel_bw_min_cdclk(display, old_bw_state);
+	new_min_cdclk = intel_bw_min_cdclk(display, new_bw_state);
 
 	/*
 	 * No need to check against the cdclk state if
@@ -1309,7 +1357,7 @@ int intel_bw_calc_min_cdclk(struct intel_atomic_state *state,
 	if (new_min_cdclk <= cdclk_state->bw_min_cdclk)
 		return 0;
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "new bandwidth min cdclk (%d kHz) > old min cdclk (%d kHz)\n",
 		    new_min_cdclk, cdclk_state->bw_min_cdclk);
 	*need_cdclk_calc = true;
@@ -1319,7 +1367,7 @@ int intel_bw_calc_min_cdclk(struct intel_atomic_state *state,
 
 static int intel_bw_check_data_rate(struct intel_atomic_state *state, bool *changed)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_crtc_state *new_crtc_state, *old_crtc_state;
 	struct intel_crtc *crtc;
 	int i;
@@ -1353,7 +1401,7 @@ static int intel_bw_check_data_rate(struct intel_atomic_state *state, bool *chan
 
 		*changed = true;
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[CRTC:%d:%s] data rate %u num active planes %u\n",
 			    crtc->base.base.id, crtc->base.name,
 			    new_bw_state->data_rate[crtc->pipe],
@@ -1363,16 +1411,103 @@ static int intel_bw_check_data_rate(struct intel_atomic_state *state, bool *chan
 	return 0;
 }
 
-int intel_bw_atomic_check(struct intel_atomic_state *state)
+static int intel_bw_modeset_checks(struct intel_atomic_state *state)
 {
+	struct intel_display *display = to_intel_display(state);
+	const struct intel_bw_state *old_bw_state;
+	struct intel_bw_state *new_bw_state;
+
+	if (DISPLAY_VER(display) < 9)
+		return 0;
+
+	new_bw_state = intel_atomic_get_bw_state(state);
+	if (IS_ERR(new_bw_state))
+		return PTR_ERR(new_bw_state);
+
+	old_bw_state = intel_atomic_get_old_bw_state(state);
+
+	new_bw_state->active_pipes =
+		intel_calc_active_pipes(state, old_bw_state->active_pipes);
+
+	if (new_bw_state->active_pipes != old_bw_state->active_pipes) {
+		int ret;
+
+		ret = intel_atomic_lock_global_state(&new_bw_state->base);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int intel_bw_check_sagv_mask(struct intel_atomic_state *state)
+{
+	struct intel_display *display = to_intel_display(state);
+	const struct intel_crtc_state *old_crtc_state;
+	const struct intel_crtc_state *new_crtc_state;
+	const struct intel_bw_state *old_bw_state = NULL;
+	struct intel_bw_state *new_bw_state = NULL;
+	struct intel_crtc *crtc;
+	int ret, i;
+
+	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
+					    new_crtc_state, i) {
+		if (intel_crtc_can_enable_sagv(old_crtc_state) ==
+		    intel_crtc_can_enable_sagv(new_crtc_state))
+			continue;
+
+		new_bw_state = intel_atomic_get_bw_state(state);
+		if (IS_ERR(new_bw_state))
+			return PTR_ERR(new_bw_state);
+
+		old_bw_state = intel_atomic_get_old_bw_state(state);
+
+		if (intel_crtc_can_enable_sagv(new_crtc_state))
+			new_bw_state->pipe_sagv_reject &= ~BIT(crtc->pipe);
+		else
+			new_bw_state->pipe_sagv_reject |= BIT(crtc->pipe);
+	}
+
+	if (!new_bw_state)
+		return 0;
+
+	if (intel_can_enable_sagv(display, new_bw_state) !=
+	    intel_can_enable_sagv(display, old_bw_state)) {
+		ret = intel_atomic_serialize_global_state(&new_bw_state->base);
+		if (ret)
+			return ret;
+	} else if (new_bw_state->pipe_sagv_reject != old_bw_state->pipe_sagv_reject) {
+		ret = intel_atomic_lock_global_state(&new_bw_state->base);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int intel_bw_atomic_check(struct intel_atomic_state *state, bool any_ms)
+{
+	struct intel_display *display = to_intel_display(state);
 	bool changed = false;
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	struct intel_bw_state *new_bw_state;
 	const struct intel_bw_state *old_bw_state;
 	int ret;
 
+	if (DISPLAY_VER(display) < 9)
+		return 0;
+
+	if (any_ms) {
+		ret = intel_bw_modeset_checks(state);
+		if (ret)
+			return ret;
+	}
+
+	ret = intel_bw_check_sagv_mask(state);
+	if (ret)
+		return ret;
+
 	/* FIXME earlier gens need some checks too */
-	if (DISPLAY_VER(i915) < 11)
+	if (DISPLAY_VER(display) < 11)
 		return 0;
 
 	ret = intel_bw_check_data_rate(state, &changed);
@@ -1383,9 +1518,8 @@ int intel_bw_atomic_check(struct intel_atomic_state *state)
 	new_bw_state = intel_atomic_get_new_bw_state(state);
 
 	if (new_bw_state &&
-	    (intel_can_enable_sagv(i915, old_bw_state) !=
-	     intel_can_enable_sagv(i915, new_bw_state) ||
-	     new_bw_state->force_check_qgv))
+	    intel_can_enable_sagv(display, old_bw_state) !=
+	    intel_can_enable_sagv(display, new_bw_state))
 		changed = true;
 
 	/*
@@ -1395,28 +1529,25 @@ int intel_bw_atomic_check(struct intel_atomic_state *state)
 	if (!changed)
 		return 0;
 
-	ret = intel_bw_check_qgv_points(i915, old_bw_state, new_bw_state);
+	ret = intel_bw_check_qgv_points(display, old_bw_state, new_bw_state);
 	if (ret)
 		return ret;
 
-	new_bw_state->force_check_qgv = false;
-
 	return 0;
 }
 
 static void intel_bw_crtc_update(struct intel_bw_state *bw_state,
 				 const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 
 	bw_state->data_rate[crtc->pipe] =
 		intel_bw_crtc_data_rate(crtc_state);
 	bw_state->num_active_planes[crtc->pipe] =
 		intel_bw_crtc_num_active_planes(crtc_state);
-	bw_state->force_check_qgv = true;
 
-	drm_dbg_kms(&i915->drm, "pipe %c data rate %u num active planes %u\n",
+	drm_dbg_kms(display->drm, "pipe %c data rate %u num active planes %u\n",
 		    pipe_name(crtc->pipe),
 		    bw_state->data_rate[crtc->pipe],
 		    bw_state->num_active_planes[crtc->pipe]);
@@ -1432,6 +1563,7 @@ void intel_bw_update_hw_state(struct intel_display *display)
 		return;
 
 	bw_state->active_pipes = 0;
+	bw_state->pipe_sagv_reject = 0;
 
 	for_each_intel_crtc(display->drm, crtc) {
 		const struct intel_crtc_state *crtc_state =
@@ -1443,6 +1575,11 @@ void intel_bw_update_hw_state(struct intel_display *display)
 
 		if (DISPLAY_VER(display) >= 11)
 			intel_bw_crtc_update(bw_state, crtc_state);
+
+		skl_crtc_calc_dbuf_bw(&bw_state->dbuf_bw[pipe], crtc_state);
+
+		/* initially SAGV has been forced off */
+		bw_state->pipe_sagv_reject |= BIT(pipe);
 	}
 }
 
@@ -1458,6 +1595,7 @@ void intel_bw_crtc_disable_noatomic(struct intel_crtc *crtc)
 
 	bw_state->data_rate[pipe] = 0;
 	bw_state->num_active_planes[pipe] = 0;
+	memset(&bw_state->dbuf_bw[pipe], 0, sizeof(bw_state->dbuf_bw[pipe]));
 }
 
 static struct intel_global_state *
@@ -1483,9 +1621,8 @@ static const struct intel_global_state_funcs intel_bw_funcs = {
 	.atomic_destroy_state = intel_bw_destroy_state,
 };
 
-int intel_bw_init(struct drm_i915_private *i915)
+int intel_bw_init(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_bw_state *state;
 
 	state = kzalloc(sizeof(*state), GFP_KERNEL);
@@ -1499,8 +1636,8 @@ int intel_bw_init(struct drm_i915_private *i915)
 	 * Limit this only if we have SAGV. And for Display version 14 onwards
 	 * sagv is handled though pmdemand requests
 	 */
-	if (intel_has_sagv(i915) && IS_DISPLAY_VER(i915, 11, 13))
-		icl_force_disable_sagv(i915, state);
+	if (intel_has_sagv(display) && IS_DISPLAY_VER(display, 11, 13))
+		icl_force_disable_sagv(display, state);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_bw.h b/drivers/gpu/drm/i915/display/intel_bw.h
index 3313e4eac4f0..eb2cc883e9c1 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.h
+++ b/drivers/gpu/drm/i915/display/intel_bw.h
@@ -12,7 +12,6 @@
 #include "intel_display_power.h"
 #include "intel_global_state.h"
 
-struct drm_i915_private;
 struct intel_atomic_state;
 struct intel_crtc;
 struct intel_crtc_state;
@@ -49,13 +48,6 @@ struct intel_bw_state {
 	 */
 	u16 qgv_points_mask;
 
-	/*
-	 * Flag to force the QGV comparison in atomic check right after the
-	 * hw state readout
-	 */
-	bool force_check_qgv;
-
-	int min_cdclk[I915_MAX_PIPES];
 	unsigned int data_rate[I915_MAX_PIPES];
 	u8 num_active_planes[I915_MAX_PIPES];
 };
@@ -72,14 +64,14 @@ intel_atomic_get_new_bw_state(struct intel_atomic_state *state);
 struct intel_bw_state *
 intel_atomic_get_bw_state(struct intel_atomic_state *state);
 
-void intel_bw_init_hw(struct drm_i915_private *dev_priv);
-int intel_bw_init(struct drm_i915_private *dev_priv);
-int intel_bw_atomic_check(struct intel_atomic_state *state);
-int icl_pcode_restrict_qgv_points(struct drm_i915_private *dev_priv,
+void intel_bw_init_hw(struct intel_display *display);
+int intel_bw_init(struct intel_display *display);
+int intel_bw_atomic_check(struct intel_atomic_state *state, bool any_ms);
+int icl_pcode_restrict_qgv_points(struct intel_display *display,
 				  u32 points_mask);
 int intel_bw_calc_min_cdclk(struct intel_atomic_state *state,
 			    bool *need_cdclk_calc);
-int intel_bw_min_cdclk(struct drm_i915_private *i915,
+int intel_bw_min_cdclk(struct intel_display *display,
 		       const struct intel_bw_state *bw_state);
 void intel_bw_update_hw_state(struct intel_display *display);
 void intel_bw_crtc_disable_noatomic(struct intel_crtc *crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 2a8749a0213e..b1718b491ffd 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -1972,9 +1972,7 @@ int intel_mdclk_cdclk_ratio(struct intel_display *display,
 static void xe2lpd_mdclk_cdclk_ratio_program(struct intel_display *display,
 					     const struct intel_cdclk_config *cdclk_config)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	intel_dbuf_mdclk_cdclk_ratio_update(i915,
+	intel_dbuf_mdclk_cdclk_ratio_update(display,
 					    intel_mdclk_cdclk_ratio(display, cdclk_config),
 					    cdclk_config->joined_mbus);
 }
@@ -2808,7 +2806,6 @@ static int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_stat
 static int intel_compute_min_cdclk(struct intel_atomic_state *state)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_cdclk_state *cdclk_state =
 		intel_atomic_get_new_cdclk_state(state);
 	const struct intel_bw_state *bw_state;
@@ -2836,7 +2833,7 @@ static int intel_compute_min_cdclk(struct intel_atomic_state *state)
 
 	bw_state = intel_atomic_get_new_bw_state(state);
 	if (bw_state) {
-		min_cdclk = intel_bw_min_cdclk(dev_priv, bw_state);
+		min_cdclk = intel_bw_min_cdclk(display, bw_state);
 
 		if (cdclk_state->bw_min_cdclk != min_cdclk) {
 			int ret;
@@ -3342,6 +3339,8 @@ int intel_modeset_calc_cdclk(struct intel_atomic_state *state)
 
 void intel_cdclk_update_hw_state(struct intel_display *display)
 {
+	const struct intel_bw_state *bw_state =
+		to_intel_bw_state(display->bw.obj.state);
 	struct intel_cdclk_state *cdclk_state =
 		to_intel_cdclk_state(display->cdclk.obj.state);
 	struct intel_crtc *crtc;
@@ -3359,6 +3358,8 @@ void intel_cdclk_update_hw_state(struct intel_display *display)
 		cdclk_state->min_cdclk[pipe] = intel_crtc_compute_min_cdclk(crtc_state);
 		cdclk_state->min_voltage_level[pipe] = crtc_state->min_voltage_level;
 	}
+
+	cdclk_state->bw_min_cdclk = intel_bw_min_cdclk(display, bw_state);
 }
 
 void intel_cdclk_crtc_disable_noatomic(struct intel_crtc *crtc)
@@ -3493,7 +3494,6 @@ static int dg1_rawclk(struct intel_display *display)
 
 static int cnp_rawclk(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	int divider, fraction;
 	u32 rawclk;
 
@@ -3513,7 +3513,7 @@ static int cnp_rawclk(struct intel_display *display)
 
 		rawclk |= CNP_RAWCLK_DEN(DIV_ROUND_CLOSEST(numerator * 1000,
 							   fraction) - 1);
-		if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+		if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 			rawclk |= ICP_RAWCLK_NUM(numerator);
 	}
 
@@ -3552,21 +3552,20 @@ static int i9xx_hrawclk(struct intel_display *display)
  */
 u32 intel_read_rawclk(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 freq;
 
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTL)
+	if (INTEL_PCH_TYPE(display) >= PCH_MTL)
 		/*
 		 * MTL always uses a 38.4 MHz rawclk.  The bspec tells us
 		 * "RAWCLK_FREQ defaults to the values for 38.4 and does
 		 * not need to be programmed."
 		 */
 		freq = 38400;
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1)
+	else if (INTEL_PCH_TYPE(display) >= PCH_DG1)
 		freq = dg1_rawclk(display);
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
+	else if (INTEL_PCH_TYPE(display) >= PCH_CNP)
 		freq = cnp_rawclk(display);
-	else if (HAS_PCH_SPLIT(dev_priv))
+	else if (HAS_PCH_SPLIT(display))
 		freq = pch_rawclk(display);
 	else if (display->platform.valleyview || display->platform.cherryview)
 		freq = vlv_hrawclk(display);
diff --git a/drivers/gpu/drm/i915/display/intel_cmtg.c b/drivers/gpu/drm/i915/display/intel_cmtg.c
index 07d7f4e8f60f..82606ebae1de 100644
--- a/drivers/gpu/drm/i915/display/intel_cmtg.c
+++ b/drivers/gpu/drm/i915/display/intel_cmtg.c
@@ -9,7 +9,6 @@
 #include <drm/drm_device.h>
 #include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_crtc.h"
 #include "intel_cmtg.h"
diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index cfe14162231d..98dddf72c0eb 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -22,7 +22,9 @@
  *
  */
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
+#include "i915_utils.h"
 #include "i9xx_plane_regs.h"
 #include "intel_color.h"
 #include "intel_color_regs.h"
@@ -405,14 +407,13 @@ static void icl_read_csc(struct intel_crtc_state *crtc_state)
 static bool ilk_limited_range(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	/* icl+ have dedicated output CSC */
 	if (DISPLAY_VER(display) >= 11)
 		return false;
 
 	/* pre-hsw have TRANSCONF_COLOR_RANGE_SELECT */
-	if (DISPLAY_VER(display) < 7 || IS_IVYBRIDGE(i915))
+	if (DISPLAY_VER(display) < 7 || display->platform.ivybridge)
 		return false;
 
 	return crtc_state->limited_color_range;
@@ -516,7 +517,6 @@ static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
 static void ilk_assign_csc(struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	bool limited_color_range = ilk_csc_limited_range(crtc_state);
 
 	if (crtc_state->hw.ctm) {
@@ -538,7 +538,7 @@ static void ilk_assign_csc(struct intel_crtc_state *crtc_state)
 		 * LUT is needed but CSC is not we need to load an
 		 * identity matrix.
 		 */
-		drm_WARN_ON(display->drm, !IS_GEMINILAKE(i915));
+		drm_WARN_ON(display->drm, !display->platform.geminilake);
 
 		ilk_csc_copy(display, &crtc_state->csc, &ilk_csc_matrix_identity);
 	} else {
@@ -3983,12 +3983,10 @@ int intel_color_init(struct intel_display *display)
 
 void intel_color_init_hooks(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (HAS_GMCH(display)) {
-		if (IS_CHERRYVIEW(i915))
+		if (display->platform.cherryview)
 			display->funcs.color = &chv_color_funcs;
-		else if (IS_VALLEYVIEW(i915))
+		else if (display->platform.valleyview)
 			display->funcs.color = &vlv_color_funcs;
 		else if (DISPLAY_VER(display) >= 4)
 			display->funcs.color = &i965_color_funcs;
@@ -4005,7 +4003,7 @@ void intel_color_init_hooks(struct intel_display *display)
 			display->funcs.color = &skl_color_funcs;
 		else if (DISPLAY_VER(display) == 8)
 			display->funcs.color = &bdw_color_funcs;
-		else if (IS_HASWELL(i915))
+		else if (display->platform.haswell)
 			display->funcs.color = &hsw_color_funcs;
 		else if (DISPLAY_VER(display) == 7)
 			display->funcs.color = &ivb_color_funcs;
diff --git a/drivers/gpu/drm/i915/display/intel_combo_phy.c b/drivers/gpu/drm/i915/display/intel_combo_phy.c
index 17eea244cc83..f5cc38dbe559 100644
--- a/drivers/gpu/drm/i915/display/intel_combo_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_combo_phy.c
@@ -3,6 +3,8 @@
  * Copyright © 2018 Intel Corporation
  */
 
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
 #include "i915_utils.h"
 #include "intel_combo_phy.h"
diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c
index e42357bd9e80..6c81c9f2fd09 100644
--- a/drivers/gpu/drm/i915/display/intel_connector.c
+++ b/drivers/gpu/drm/i915/display/intel_connector.c
@@ -31,8 +31,10 @@
 #include <drm/drm_probe_helper.h>
 
 #include "i915_drv.h"
+#include "i915_utils.h"
 #include "intel_backlight.h"
 #include "intel_connector.h"
+#include "intel_display_core.h"
 #include "intel_display_debugfs.h"
 #include "intel_display_types.h"
 #include "intel_hdcp.h"
@@ -154,13 +156,14 @@ void intel_connector_destroy(struct drm_connector *connector)
 int intel_connector_register(struct drm_connector *connector)
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
+	struct drm_i915_private *i915 = to_i915(connector->dev);
 	int ret;
 
 	ret = intel_backlight_device_register(intel_connector);
 	if (ret)
 		goto err;
 
-	if (i915_inject_probe_failure(to_i915(connector->dev))) {
+	if (i915_inject_probe_failure(i915)) {
 		ret = -EFAULT;
 		goto err_backlight;
 	}
@@ -204,10 +207,10 @@ bool intel_connector_get_hw_state(struct intel_connector *connector)
 
 enum pipe intel_connector_get_pipe(struct intel_connector *connector)
 {
-	struct drm_device *dev = connector->base.dev;
+	struct intel_display *display = to_intel_display(connector);
 
-	drm_WARN_ON(dev,
-		    !drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+	drm_WARN_ON(display->drm,
+		    !drm_modeset_is_locked(&display->drm->mode_config.connection_mutex));
 
 	if (!connector->base.state->crtc)
 		return INVALID_PIPE;
@@ -264,20 +267,19 @@ static const struct drm_prop_enum_list force_audio_names[] = {
 void
 intel_attach_force_audio_property(struct drm_connector *connector)
 {
-	struct drm_device *dev = connector->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_display *display = to_intel_display(connector->dev);
 	struct drm_property *prop;
 
-	prop = dev_priv->display.properties.force_audio;
+	prop = display->properties.force_audio;
 	if (prop == NULL) {
-		prop = drm_property_create_enum(dev, 0,
-					   "audio",
-					   force_audio_names,
-					   ARRAY_SIZE(force_audio_names));
+		prop = drm_property_create_enum(display->drm, 0,
+						"audio",
+						force_audio_names,
+						ARRAY_SIZE(force_audio_names));
 		if (prop == NULL)
 			return;
 
-		dev_priv->display.properties.force_audio = prop;
+		display->properties.force_audio = prop;
 	}
 	drm_object_attach_property(&connector->base, prop, 0);
 }
@@ -291,20 +293,19 @@ static const struct drm_prop_enum_list broadcast_rgb_names[] = {
 void
 intel_attach_broadcast_rgb_property(struct drm_connector *connector)
 {
-	struct drm_device *dev = connector->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_display *display = to_intel_display(connector->dev);
 	struct drm_property *prop;
 
-	prop = dev_priv->display.properties.broadcast_rgb;
+	prop = display->properties.broadcast_rgb;
 	if (prop == NULL) {
-		prop = drm_property_create_enum(dev, DRM_MODE_PROP_ENUM,
-					   "Broadcast RGB",
-					   broadcast_rgb_names,
-					   ARRAY_SIZE(broadcast_rgb_names));
+		prop = drm_property_create_enum(display->drm, DRM_MODE_PROP_ENUM,
+						"Broadcast RGB",
+						broadcast_rgb_names,
+						ARRAY_SIZE(broadcast_rgb_names));
 		if (prop == NULL)
 			return;
 
-		dev_priv->display.properties.broadcast_rgb = prop;
+		display->properties.broadcast_rgb = prop;
 	}
 
 	drm_object_attach_property(&connector->base, prop, 0);
@@ -336,14 +337,14 @@ intel_attach_dp_colorspace_property(struct drm_connector *connector)
 void
 intel_attach_scaling_mode_property(struct drm_connector *connector)
 {
-	struct drm_i915_private *i915 = to_i915(connector->dev);
+	struct intel_display *display = to_intel_display(connector->dev);
 	u32 scaling_modes;
 
 	scaling_modes = BIT(DRM_MODE_SCALE_ASPECT) |
 		BIT(DRM_MODE_SCALE_FULLSCREEN);
 
 	/* On GMCH platforms borders are only possible on the LVDS port */
-	if (!HAS_GMCH(i915) || connector->connector_type == DRM_MODE_CONNECTOR_LVDS)
+	if (!HAS_GMCH(display) || connector->connector_type == DRM_MODE_CONNECTOR_LVDS)
 		scaling_modes |= BIT(DRM_MODE_SCALE_CENTER);
 
 	drm_connector_attach_scaling_mode_property(connector, scaling_modes);
diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 76ffb3f8467c..38b50a779b6b 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -31,9 +31,9 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
-#include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
 #include "intel_connector.h"
@@ -91,13 +91,12 @@ static struct intel_crt *intel_attached_crt(struct intel_connector *connector)
 bool intel_crt_port_enabled(struct intel_display *display,
 			    i915_reg_t adpa_reg, enum pipe *pipe)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 val;
 
 	val = intel_de_read(display, adpa_reg);
 
 	/* asserts want to know the pipe even if the port is disabled */
-	if (HAS_PCH_CPT(dev_priv))
+	if (HAS_PCH_CPT(display))
 		*pipe = REG_FIELD_GET(ADPA_PIPE_SEL_MASK_CPT, val);
 	else
 		*pipe = REG_FIELD_GET(ADPA_PIPE_SEL_MASK, val);
@@ -177,7 +176,6 @@ static void intel_crt_set_dpms(struct intel_encoder *encoder,
 			       int mode)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
@@ -194,14 +192,14 @@ static void intel_crt_set_dpms(struct intel_encoder *encoder,
 		adpa |= ADPA_VSYNC_ACTIVE_HIGH;
 
 	/* For CPT allow 3 pipe config, for others just use A or B */
-	if (HAS_PCH_LPT(dev_priv))
+	if (HAS_PCH_LPT(display))
 		; /* Those bits don't exist here */
-	else if (HAS_PCH_CPT(dev_priv))
+	else if (HAS_PCH_CPT(display))
 		adpa |= ADPA_PIPE_SEL_CPT(crtc->pipe);
 	else
 		adpa |= ADPA_PIPE_SEL(crtc->pipe);
 
-	if (!HAS_PCH_SPLIT(dev_priv))
+	if (!HAS_PCH_SPLIT(display))
 		intel_de_write(display, BCLRPAT(display, crtc->pipe), 0);
 
 	switch (mode) {
@@ -356,7 +354,6 @@ intel_crt_mode_valid(struct drm_connector *connector,
 		     const struct drm_display_mode *mode)
 {
 	struct intel_display *display = to_intel_display(connector->dev);
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	int max_dotclk = display->cdclk.max_dotclk_freq;
 	enum drm_mode_status status;
 	int max_clock;
@@ -368,9 +365,9 @@ intel_crt_mode_valid(struct drm_connector *connector,
 	if (mode->clock < 25000)
 		return MODE_CLOCK_LOW;
 
-	if (HAS_PCH_LPT(dev_priv))
+	if (HAS_PCH_LPT(display))
 		max_clock = 180000;
-	else if (IS_VALLEYVIEW(dev_priv))
+	else if (display->platform.valleyview)
 		/*
 		 * 270 MHz due to current DPLL limits,
 		 * DAC limit supposedly 355 MHz.
@@ -387,7 +384,7 @@ intel_crt_mode_valid(struct drm_connector *connector,
 		return MODE_CLOCK_HIGH;
 
 	/* The FDI receiver on LPT only supports 8bpc and only has 2 lanes. */
-	if (HAS_PCH_LPT(dev_priv) &&
+	if (HAS_PCH_LPT(display) &&
 	    ilk_get_lanes_required(mode->clock, 270000, 24) > 2)
 		return MODE_CLOCK_HIGH;
 
@@ -438,7 +435,6 @@ static int hsw_crt_compute_config(struct intel_encoder *encoder,
 				  struct drm_connector_state *conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct drm_display_mode *adjusted_mode =
 		&crtc_state->hw.adjusted_mode;
 
@@ -457,7 +453,7 @@ static int hsw_crt_compute_config(struct intel_encoder *encoder,
 	crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	/* LPT FDI RX only supports 8bpc. */
-	if (HAS_PCH_LPT(dev_priv)) {
+	if (HAS_PCH_LPT(display)) {
 		/* TODO: Check crtc_state->max_link_bpp_x16 instead of bw_constrained */
 		if (crtc_state->bw_constrained && crtc_state->pipe_bpp < 24) {
 			drm_dbg_kms(display->drm,
@@ -482,13 +478,12 @@ static bool ilk_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct intel_display *display = to_intel_display(connector->dev);
 	struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector));
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	u32 adpa;
 	bool ret;
 
 	/* The first time through, trigger an explicit detection cycle */
 	if (crt->force_hotplug_required) {
-		bool turn_off_dac = HAS_PCH_SPLIT(dev_priv);
+		bool turn_off_dac = HAS_PCH_SPLIT(display);
 		u32 save_adpa;
 
 		crt->force_hotplug_required = false;
@@ -532,8 +527,6 @@ static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct intel_display *display = to_intel_display(connector->dev);
 	struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector));
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
-	bool reenable_hpd;
 	u32 adpa;
 	bool ret;
 	u32 save_adpa;
@@ -550,7 +543,7 @@ static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 	 *
 	 * Just disable HPD interrupts here to prevent this
 	 */
-	reenable_hpd = intel_hpd_disable(dev_priv, crt->base.hpd_pin);
+	intel_hpd_block(&crt->base);
 
 	save_adpa = adpa = intel_de_read(display, crt->adpa_reg);
 	drm_dbg_kms(display->drm,
@@ -577,8 +570,7 @@ static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 	drm_dbg_kms(display->drm,
 		    "valleyview hotplug adpa=0x%x, result %d\n", adpa, ret);
 
-	if (reenable_hpd)
-		intel_hpd_enable(dev_priv, crt->base.hpd_pin);
+	intel_hpd_clear_and_unblock(&crt->base);
 
 	return ret;
 }
@@ -586,15 +578,14 @@ static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
 static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct intel_display *display = to_intel_display(connector->dev);
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	u32 stat;
 	bool ret = false;
 	int i, tries = 0;
 
-	if (HAS_PCH_SPLIT(dev_priv))
+	if (HAS_PCH_SPLIT(display))
 		return ilk_crt_detect_hotplug(connector);
 
-	if (IS_VALLEYVIEW(dev_priv))
+	if (display->platform.valleyview)
 		return valleyview_crt_detect_hotplug(connector);
 
 	/*
@@ -602,14 +593,14 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 	 * to get a reliable result.
 	 */
 
-	if (IS_G45(dev_priv))
+	if (display->platform.g45)
 		tries = 2;
 	else
 		tries = 1;
 
 	for (i = 0; i < tries ; i++) {
 		/* turn on the FORCE_DETECT */
-		i915_hotplug_interrupt_update(dev_priv,
+		i915_hotplug_interrupt_update(display,
 					      CRT_HOTPLUG_FORCE_DETECT,
 					      CRT_HOTPLUG_FORCE_DETECT);
 		/* wait for FORCE_DETECT to go off */
@@ -627,7 +618,7 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 	intel_de_write(display, PORT_HOTPLUG_STAT(display),
 		       CRT_HOTPLUG_INT_STATUS);
 
-	i915_hotplug_interrupt_update(dev_priv, CRT_HOTPLUG_FORCE_DETECT, 0);
+	i915_hotplug_interrupt_update(display, CRT_HOTPLUG_FORCE_DETECT, 0);
 
 	return ret;
 }
@@ -880,7 +871,7 @@ intel_crt_detect(struct drm_connector *connector,
 
 	wakeref = intel_display_power_get(display, encoder->power_domain);
 
-	if (I915_HAS_HOTPLUG(display)) {
+	if (HAS_HOTPLUG(display)) {
 		/* We can not rely on the HPD pin always being correctly wired
 		 * up, for example many KVM do not pass it through, and so
 		 * only trust an assertion that the monitor is connected.
@@ -904,7 +895,7 @@ intel_crt_detect(struct drm_connector *connector,
 	 * broken monitor (without edid) to work behind a broken kvm (that fails
 	 * to have the right resistors for HP detection) needs to fix this up.
 	 * For now just bail out. */
-	if (I915_HAS_HOTPLUG(display)) {
+	if (HAS_HOTPLUG(display)) {
 		status = connector_status_disconnected;
 		goto out;
 	}
@@ -943,7 +934,6 @@ out:
 static int intel_crt_get_modes(struct drm_connector *connector)
 {
 	struct intel_display *display = to_intel_display(connector->dev);
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector));
 	struct intel_encoder *encoder = &crt->base;
 	intel_wakeref_t wakeref;
@@ -956,7 +946,7 @@ static int intel_crt_get_modes(struct drm_connector *connector)
 	wakeref = intel_display_power_get(display, encoder->power_domain);
 
 	ret = intel_crt_ddc_get_modes(connector, connector->ddc);
-	if (ret || !IS_G4X(dev_priv))
+	if (ret || !display->platform.g4x)
 		goto out;
 
 	/* Try to probe digital port for output in DVI-I -> VGA mode. */
@@ -1015,16 +1005,15 @@ static const struct drm_encoder_funcs intel_crt_enc_funcs = {
 
 void intel_crt_init(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_connector *connector;
 	struct intel_crt *crt;
 	i915_reg_t adpa_reg;
 	u8 ddc_pin;
 	u32 adpa;
 
-	if (HAS_PCH_SPLIT(dev_priv))
+	if (HAS_PCH_SPLIT(display))
 		adpa_reg = PCH_ADPA;
-	else if (IS_VALLEYVIEW(dev_priv))
+	else if (display->platform.valleyview)
 		adpa_reg = VLV_ADPA;
 	else
 		adpa_reg = ADPA;
@@ -1072,7 +1061,7 @@ void intel_crt_init(struct intel_display *display)
 
 	crt->base.type = INTEL_OUTPUT_ANALOG;
 	crt->base.cloneable = BIT(INTEL_OUTPUT_DVO) | BIT(INTEL_OUTPUT_HDMI);
-	if (IS_I830(dev_priv))
+	if (display->platform.i830)
 		crt->base.pipe_mask = BIT(PIPE_A);
 	else
 		crt->base.pipe_mask = ~0;
@@ -1084,7 +1073,7 @@ void intel_crt_init(struct intel_display *display)
 
 	crt->base.power_domain = POWER_DOMAIN_PORT_CRT;
 
-	if (I915_HAS_HOTPLUG(display) &&
+	if (HAS_HOTPLUG(display) &&
 	    !dmi_check_system(intel_spurious_crt_detect)) {
 		crt->base.hpd_pin = HPD_CRT;
 		crt->base.hotplug = intel_encoder_hotplug;
@@ -1112,7 +1101,7 @@ void intel_crt_init(struct intel_display *display)
 
 		intel_ddi_buf_trans_init(&crt->base);
 	} else {
-		if (HAS_PCH_SPLIT(dev_priv)) {
+		if (HAS_PCH_SPLIT(display)) {
 			crt->base.compute_config = pch_crt_compute_config;
 			crt->base.disable = pch_disable_crt;
 			crt->base.post_disable = pch_post_disable_crt;
@@ -1134,7 +1123,7 @@ void intel_crt_init(struct intel_display *display)
 	 * polarity and link reversal bits or not, instead of relying on the
 	 * BIOS.
 	 */
-	if (HAS_PCH_LPT(dev_priv)) {
+	if (HAS_PCH_LPT(display)) {
 		u32 fdi_config = FDI_RX_POLARITY_REVERSED_LPT |
 				 FDI_RX_LINK_REVERSAL_OVERRIDE;
 
diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index 5b2603ef2ff7..29cfc38f12e0 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -124,7 +124,7 @@ void intel_crtc_vblank_on(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 
-	crtc->block_dc_for_vblank = intel_psr_needs_block_dc_vblank(crtc_state);
+	crtc->vblank_psr_notify = intel_psr_needs_vblank_notification(crtc_state);
 
 	assert_vblank_disabled(&crtc->base);
 	drm_crtc_set_max_vblank_count(&crtc->base,
@@ -154,9 +154,9 @@ void intel_crtc_vblank_off(const struct intel_crtc_state *crtc_state)
 	drm_crtc_vblank_off(&crtc->base);
 	assert_vblank_disabled(&crtc->base);
 
-	crtc->block_dc_for_vblank = false;
+	crtc->vblank_psr_notify = false;
 
-	flush_work(&display->irq.vblank_dc_work);
+	flush_work(&display->irq.vblank_notify_work);
 }
 
 struct intel_crtc_state *intel_crtc_state_alloc(struct intel_crtc *crtc)
@@ -305,7 +305,6 @@ static const struct drm_crtc_funcs i8xx_crtc_funcs = {
 
 int intel_crtc_init(struct intel_display *display, enum pipe pipe)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_plane *primary, *cursor;
 	const struct drm_crtc_funcs *funcs;
 	struct intel_crtc *crtc;
@@ -333,7 +332,7 @@ int intel_crtc_init(struct intel_display *display, enum pipe pipe)
 	for_each_sprite(display, pipe, sprite) {
 		struct intel_plane *plane;
 
-		if (DISPLAY_VER(dev_priv) >= 9)
+		if (DISPLAY_VER(display) >= 9)
 			plane = skl_universal_plane_create(display, pipe, PLANE_2 + sprite);
 		else
 			plane = intel_sprite_plane_create(display, pipe, sprite);
diff --git a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
index 599ddce96371..0c7f91046996 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
@@ -5,9 +5,10 @@
 
 #include <drm/drm_edid.h>
 #include <drm/drm_eld.h>
+#include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "intel_crtc_state_dump.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_hdmi.h"
 #include "intel_vblank.h"
@@ -42,13 +43,13 @@ intel_dump_m_n_config(struct drm_printer *p,
 }
 
 static void
-intel_dump_infoframe(struct drm_i915_private *i915,
+intel_dump_infoframe(struct intel_display *display,
 		     const union hdmi_infoframe *frame)
 {
 	if (!drm_debug_enabled(DRM_UT_KMS))
 		return;
 
-	hdmi_infoframe_log(KERN_DEBUG, i915->drm.dev, frame);
+	hdmi_infoframe_log(KERN_DEBUG, display->drm->dev, frame);
 }
 
 #define OUTPUT_TYPE(x) [INTEL_OUTPUT_ ## x] = #x
@@ -136,7 +137,7 @@ static void intel_dump_plane_state(struct drm_printer *p,
 }
 
 static void
-ilk_dump_csc(struct drm_i915_private *i915,
+ilk_dump_csc(struct intel_display *display,
 	     struct drm_printer *p,
 	     const char *name,
 	     const struct intel_csc_matrix *csc)
@@ -152,7 +153,7 @@ ilk_dump_csc(struct drm_i915_private *i915,
 			   csc->coeff[3 * i + 1],
 			   csc->coeff[3 * i + 2]);
 
-	if (DISPLAY_VER(i915) < 7)
+	if (DISPLAY_VER(display) < 7)
 		return;
 
 	drm_printf(p, "%s: post offsets: 0x%04x 0x%04x 0x%04x\n", name,
@@ -178,7 +179,6 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 {
 	struct intel_display *display = to_intel_display(pipe_config);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	const struct intel_plane_state *plane_state;
 	struct intel_plane *plane;
 	struct drm_printer p;
@@ -188,7 +188,7 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 	if (!drm_debug_enabled(DRM_UT_KMS))
 		return;
 
-	p = drm_dbg_printer(&i915->drm, DRM_UT_KMS, NULL);
+	p = drm_dbg_printer(display->drm, DRM_UT_KMS, NULL);
 
 	drm_printf(&p, "[CRTC:%d:%s] enable: %s [%s]\n",
 		   crtc->base.base.id, crtc->base.name,
@@ -262,19 +262,19 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 		drm_printf(&p, "GCP: 0x%x\n", pipe_config->infoframes.gcp);
 	if (pipe_config->infoframes.enable &
 	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI))
-		intel_dump_infoframe(i915, &pipe_config->infoframes.avi);
+		intel_dump_infoframe(display, &pipe_config->infoframes.avi);
 	if (pipe_config->infoframes.enable &
 	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_SPD))
-		intel_dump_infoframe(i915, &pipe_config->infoframes.spd);
+		intel_dump_infoframe(display, &pipe_config->infoframes.spd);
 	if (pipe_config->infoframes.enable &
 	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_VENDOR))
-		intel_dump_infoframe(i915, &pipe_config->infoframes.hdmi);
+		intel_dump_infoframe(display, &pipe_config->infoframes.hdmi);
 	if (pipe_config->infoframes.enable &
 	    intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_DRM))
-		intel_dump_infoframe(i915, &pipe_config->infoframes.drm);
+		intel_dump_infoframe(display, &pipe_config->infoframes.drm);
 	if (pipe_config->infoframes.enable &
 	    intel_hdmi_infoframe_enable(HDMI_PACKET_TYPE_GAMUT_METADATA))
-		intel_dump_infoframe(i915, &pipe_config->infoframes.drm);
+		intel_dump_infoframe(display, &pipe_config->infoframes.drm);
 	if (pipe_config->infoframes.enable &
 	    intel_hdmi_infoframe_enable(DP_SDP_VSC))
 		drm_dp_vsc_sdp_log(&p, &pipe_config->infoframes.vsc);
@@ -294,8 +294,9 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 		   pipe_config->hw.adjusted_mode.crtc_vdisplay,
 		   pipe_config->framestart_delay, pipe_config->msa_timing_delay);
 
-	drm_printf(&p, "vrr: %s, vmin: %d, vmax: %d, flipline: %d, pipeline full: %d, guardband: %d vsync start: %d, vsync end: %d\n",
+	drm_printf(&p, "vrr: %s, fixed rr: %s, vmin: %d, vmax: %d, flipline: %d, pipeline full: %d, guardband: %d vsync start: %d, vsync end: %d\n",
 		   str_yes_no(pipe_config->vrr.enable),
+		   str_yes_no(intel_vrr_is_fixed_rr(pipe_config)),
 		   pipe_config->vrr.vmin, pipe_config->vrr.vmax, pipe_config->vrr.flipline,
 		   pipe_config->vrr.pipeline_full, pipe_config->vrr.guardband,
 		   pipe_config->vrr.vsync_start, pipe_config->vrr.vsync_end);
@@ -319,14 +320,14 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 	drm_printf(&p, "linetime: %d, ips linetime: %d\n",
 		   pipe_config->linetime, pipe_config->ips_linetime);
 
-	if (DISPLAY_VER(i915) >= 9)
+	if (DISPLAY_VER(display) >= 9)
 		drm_printf(&p, "num_scalers: %d, scaler_users: 0x%x, scaler_id: %d, scaling_filter: %d\n",
 			   crtc->num_scalers,
 			   pipe_config->scaler_state.scaler_users,
 			   pipe_config->scaler_state.scaler_id,
 			   pipe_config->hw.scaling_filter);
 
-	if (HAS_GMCH(i915))
+	if (HAS_GMCH(display))
 		drm_printf(&p, "gmch pfit: control: 0x%08x, ratios: 0x%08x, lvds border: 0x%08x\n",
 			   pipe_config->gmch_pfit.control,
 			   pipe_config->gmch_pfit.pgm_ratios,
@@ -343,7 +344,7 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 
 	intel_dpll_dump_hw_state(display, &p, &pipe_config->dpll_hw_state);
 
-	if (IS_CHERRYVIEW(i915))
+	if (display->platform.cherryview)
 		drm_printf(&p, "cgm_mode: 0x%x gamma_mode: 0x%x gamma_enable: %d csc_enable: %d\n",
 			   pipe_config->cgm_mode, pipe_config->gamma_mode,
 			   pipe_config->gamma_enable, pipe_config->csc_enable);
@@ -354,20 +355,20 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 
 	drm_printf(&p, "pre csc lut: %s%d entries, post csc lut: %d entries\n",
 		   pipe_config->pre_csc_lut && pipe_config->pre_csc_lut ==
-		   i915->display.color.glk_linear_degamma_lut ? "(linear) " : "",
+		   display->color.glk_linear_degamma_lut ? "(linear) " : "",
 		   pipe_config->pre_csc_lut ?
 		   drm_color_lut_size(pipe_config->pre_csc_lut) : 0,
 		   pipe_config->post_csc_lut ?
 		   drm_color_lut_size(pipe_config->post_csc_lut) : 0);
 
-	if (DISPLAY_VER(i915) >= 11)
-		ilk_dump_csc(i915, &p, "output csc", &pipe_config->output_csc);
+	if (DISPLAY_VER(display) >= 11)
+		ilk_dump_csc(display, &p, "output csc", &pipe_config->output_csc);
 
-	if (!HAS_GMCH(i915))
-		ilk_dump_csc(i915, &p, "pipe csc", &pipe_config->csc);
-	else if (IS_CHERRYVIEW(i915))
+	if (!HAS_GMCH(display))
+		ilk_dump_csc(display, &p, "pipe csc", &pipe_config->csc);
+	else if (display->platform.cherryview)
 		vlv_dump_csc(&p, "cgm csc", &pipe_config->csc);
-	else if (IS_VALLEYVIEW(i915))
+	else if (display->platform.valleyview)
 		vlv_dump_csc(&p, "wgc csc", &pipe_config->csc);
 
 	intel_vdsc_state_dump(&p, 0, pipe_config);
diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c
index 3276a5b4a9b0..2fec5ba58373 100644
--- a/drivers/gpu/drm/i915/display/intel_cursor.c
+++ b/drivers/gpu/drm/i915/display/intel_cursor.c
@@ -9,10 +9,11 @@
 #include <drm/drm_blend.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_print.h>
 #include <drm/drm_vblank.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_atomic.h"
 #include "intel_atomic_plane.h"
 #include "intel_cursor.h"
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 22595766eac5..b09f724c3046 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -6,8 +6,10 @@
 #include <linux/log2.h>
 #include <linux/math64.h>
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_cx0_phy.h"
 #include "intel_cx0_phy_regs.h"
 #include "intel_ddi.h"
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index f38c998935b9..74132c1d6385 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -30,11 +30,13 @@
 
 #include <drm/display/drm_dp_helper.h>
 #include <drm/display/drm_scdc_helper.h>
+#include <drm/drm_print.h>
 #include <drm/drm_privacy_screen_consumer.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "icl_dsi.h"
+#include "intel_alpm.h"
 #include "intel_audio.h"
 #include "intel_audio_regs.h"
 #include "intel_backlight.h"
@@ -78,6 +80,7 @@
 #include "intel_tc.h"
 #include "intel_vdsc.h"
 #include "intel_vdsc_regs.h"
+#include "intel_vrr.h"
 #include "skl_scaler.h"
 #include "skl_universal_plane.h"
 
@@ -106,14 +109,14 @@ static int intel_ddi_hdmi_level(struct intel_encoder *encoder,
 	return level;
 }
 
-static bool has_buf_trans_select(struct drm_i915_private *i915)
+static bool has_buf_trans_select(struct intel_display *display)
 {
-	return DISPLAY_VER(i915) < 10 && !IS_BROXTON(i915);
+	return DISPLAY_VER(display) < 10 && !display->platform.broxton;
 }
 
-static bool has_iboost(struct drm_i915_private *i915)
+static bool has_iboost(struct intel_display *display)
 {
-	return DISPLAY_VER(i915) == 9 && !IS_BROXTON(i915);
+	return DISPLAY_VER(display) == 9 && !display->platform.broxton;
 }
 
 /*
@@ -124,25 +127,25 @@ static bool has_iboost(struct drm_i915_private *i915)
 void hsw_prepare_dp_ddi_buffers(struct intel_encoder *encoder,
 				const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	u32 iboost_bit = 0;
 	int i, n_entries;
 	enum port port = encoder->port;
 	const struct intel_ddi_buf_trans *trans;
 
 	trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
-	if (drm_WARN_ON_ONCE(&dev_priv->drm, !trans))
+	if (drm_WARN_ON_ONCE(display->drm, !trans))
 		return;
 
 	/* If we're boosting the current, set bit 31 of trans1 */
-	if (has_iboost(dev_priv) &&
+	if (has_iboost(display) &&
 	    intel_bios_dp_boost_level(encoder->devdata))
 		iboost_bit = DDI_BUF_BALANCE_LEG_ENABLE;
 
 	for (i = 0; i < n_entries; i++) {
-		intel_de_write(dev_priv, DDI_BUF_TRANS_LO(port, i),
+		intel_de_write(display, DDI_BUF_TRANS_LO(port, i),
 			       trans->entries[i].hsw.trans1 | iboost_bit);
-		intel_de_write(dev_priv, DDI_BUF_TRANS_HI(port, i),
+		intel_de_write(display, DDI_BUF_TRANS_HI(port, i),
 			       trans->entries[i].hsw.trans2);
 	}
 }
@@ -155,7 +158,7 @@ void hsw_prepare_dp_ddi_buffers(struct intel_encoder *encoder,
 static void hsw_prepare_hdmi_ddi_buffers(struct intel_encoder *encoder,
 					 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	int level = intel_ddi_level(encoder, crtc_state, 0);
 	u32 iboost_bit = 0;
 	int n_entries;
@@ -163,27 +166,25 @@ static void hsw_prepare_hdmi_ddi_buffers(struct intel_encoder *encoder,
 	const struct intel_ddi_buf_trans *trans;
 
 	trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
-	if (drm_WARN_ON_ONCE(&dev_priv->drm, !trans))
+	if (drm_WARN_ON_ONCE(display->drm, !trans))
 		return;
 
 	/* If we're boosting the current, set bit 31 of trans1 */
-	if (has_iboost(dev_priv) &&
+	if (has_iboost(display) &&
 	    intel_bios_hdmi_boost_level(encoder->devdata))
 		iboost_bit = DDI_BUF_BALANCE_LEG_ENABLE;
 
 	/* Entry 9 is for HDMI: */
-	intel_de_write(dev_priv, DDI_BUF_TRANS_LO(port, 9),
+	intel_de_write(display, DDI_BUF_TRANS_LO(port, 9),
 		       trans->entries[level].hsw.trans1 | iboost_bit);
-	intel_de_write(dev_priv, DDI_BUF_TRANS_HI(port, 9),
+	intel_de_write(display, DDI_BUF_TRANS_HI(port, 9),
 		       trans->entries[level].hsw.trans2);
 }
 
 static i915_reg_t intel_ddi_buf_status_reg(struct intel_display *display, enum port port)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (DISPLAY_VER(display) >= 14)
-		return XELPDP_PORT_BUF_CTL1(i915, port);
+		return XELPDP_PORT_BUF_CTL1(display, port);
 	else
 		return DDI_BUF_CTL(port);
 }
@@ -346,7 +347,6 @@ static void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder,
 				      const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 
@@ -359,14 +359,14 @@ static void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder,
 	if (dig_port->ddi_a_4_lanes)
 		intel_dp->DP |= DDI_A_4_LANES;
 
-	if (DISPLAY_VER(i915) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		if (intel_dp_is_uhbr(crtc_state))
 			intel_dp->DP |= DDI_BUF_PORT_DATA_40BIT;
 		else
 			intel_dp->DP |= DDI_BUF_PORT_DATA_10BIT;
 	}
 
-	if (IS_ALDERLAKE_P(i915) && intel_encoder_is_tc(encoder)) {
+	if (display->platform.alderlake_p && intel_encoder_is_tc(encoder)) {
 		intel_dp->DP |= ddi_buf_phy_link_rate(crtc_state->port_clock);
 		if (!intel_tc_port_in_tbt_alt_mode(dig_port))
 			intel_dp->DP |= DDI_BUF_CTL_TC_PHY_OWNERSHIP;
@@ -379,8 +379,7 @@ static void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder,
 	}
 }
 
-static int icl_calc_tbt_pll_link(struct intel_display *display,
-				 enum port port)
+static int icl_calc_tbt_pll_link(struct intel_display *display, enum port port)
 {
 	u32 val = intel_de_read(display, DDI_CLK_SEL(port)) & DDI_CLK_SEL_MASK;
 
@@ -414,15 +413,14 @@ static void ddi_dotclock_get(struct intel_crtc_state *pipe_config)
 void intel_ddi_set_dp_msa(const struct intel_crtc_state *crtc_state,
 			  const struct drm_connector_state *conn_state)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 temp;
 
 	if (!intel_crtc_has_dp_encoder(crtc_state))
 		return;
 
-	drm_WARN_ON(&dev_priv->drm, transcoder_is_dsi(cpu_transcoder));
+	drm_WARN_ON(display->drm, transcoder_is_dsi(cpu_transcoder));
 
 	temp = DP_MSA_MISC_SYNC_CLOCK;
 
@@ -445,7 +443,7 @@ void intel_ddi_set_dp_msa(const struct intel_crtc_state *crtc_state,
 	}
 
 	/* nonsense combination */
-	drm_WARN_ON(&dev_priv->drm, crtc_state->limited_color_range &&
+	drm_WARN_ON(display->drm, crtc_state->limited_color_range &&
 		    crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB);
 
 	if (crtc_state->limited_color_range)
@@ -468,7 +466,7 @@ void intel_ddi_set_dp_msa(const struct intel_crtc_state *crtc_state,
 	if (intel_dp_needs_vsc_sdp(crtc_state, conn_state))
 		temp |= DP_MSA_MISC_COLOR_VSC_SDP;
 
-	intel_de_write(dev_priv, TRANS_MSA_MISC(dev_priv, cpu_transcoder),
+	intel_de_write(display, TRANS_MSA_MISC(display, cpu_transcoder),
 		       temp);
 }
 
@@ -507,8 +505,8 @@ static u32
 intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 				      const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	enum port port = encoder->port;
@@ -516,7 +514,7 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 
 	/* Enable TRANS_DDI_FUNC_CTL for the pipe to work in HDMI mode */
 	temp = TRANS_DDI_FUNC_ENABLE;
-	if (DISPLAY_VER(dev_priv) >= 12)
+	if (DISPLAY_VER(display) >= 12)
 		temp |= TGL_TRANS_DDI_SELECT_PORT(port);
 	else
 		temp |= TRANS_DDI_SELECT_PORT(port);
@@ -578,7 +576,7 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 			temp |= TRANS_DDI_HDMI_SCRAMBLING;
 		if (crtc_state->hdmi_high_tmds_clock_ratio)
 			temp |= TRANS_DDI_HIGH_TMDS_CHAR_RATE;
-		if (DISPLAY_VER(dev_priv) >= 14)
+		if (DISPLAY_VER(display) >= 14)
 			temp |= TRANS_DDI_PORT_WIDTH(crtc_state->lane_count);
 	} else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) {
 		temp |= TRANS_DDI_MODE_SELECT_FDI_OR_128B132B;
@@ -591,11 +589,11 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 			temp |= TRANS_DDI_MODE_SELECT_DP_MST;
 		temp |= DDI_PORT_WIDTH(crtc_state->lane_count);
 
-		if (DISPLAY_VER(dev_priv) >= 12) {
+		if (DISPLAY_VER(display) >= 12) {
 			enum transcoder master;
 
 			master = crtc_state->mst_master_transcoder;
-			drm_WARN_ON(&dev_priv->drm,
+			drm_WARN_ON(display->drm,
 				    master == INVALID_TRANSCODER);
 			temp |= TRANS_DDI_MST_TRANSPORT_SELECT(master);
 		}
@@ -604,7 +602,7 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 		temp |= DDI_PORT_WIDTH(crtc_state->lane_count);
 	}
 
-	if (IS_DISPLAY_VER(dev_priv, 8, 10) &&
+	if (IS_DISPLAY_VER(display, 8, 10) &&
 	    crtc_state->master_transcoder != INVALID_TRANSCODER) {
 		u8 master_select =
 			bdw_trans_port_sync_master_select(crtc_state->master_transcoder);
@@ -619,11 +617,10 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 void intel_ddi_enable_transcoder_func(struct intel_encoder *encoder,
 				      const struct intel_crtc_state *crtc_state)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 
-	if (DISPLAY_VER(dev_priv) >= 11) {
+	if (DISPLAY_VER(display) >= 11) {
 		enum transcoder master_transcoder = crtc_state->master_transcoder;
 		u32 ctl2 = 0;
 
@@ -635,12 +632,12 @@ void intel_ddi_enable_transcoder_func(struct intel_encoder *encoder,
 				PORT_SYNC_MODE_MASTER_SELECT(master_select);
 		}
 
-		intel_de_write(dev_priv,
-			       TRANS_DDI_FUNC_CTL2(dev_priv, cpu_transcoder),
+		intel_de_write(display,
+			       TRANS_DDI_FUNC_CTL2(display, cpu_transcoder),
 			       ctl2);
 	}
 
-	intel_de_write(dev_priv, TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder),
+	intel_de_write(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder),
 		       intel_ddi_transcoder_func_reg_val_get(encoder,
 							     crtc_state));
 }
@@ -654,8 +651,7 @@ void
 intel_ddi_config_transcoder_func(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 ctl;
 
@@ -663,7 +659,7 @@ intel_ddi_config_transcoder_func(struct intel_encoder *encoder,
 
 	ctl = intel_ddi_transcoder_func_reg_val_get(encoder, crtc_state);
 	ctl &= ~TRANS_DDI_FUNC_ENABLE;
-	intel_de_write(dev_priv, TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder),
+	intel_de_write(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder),
 		       ctl);
 }
 
@@ -677,27 +673,26 @@ void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 ctl;
 
-	if (DISPLAY_VER(dev_priv) >= 11)
-		intel_de_write(dev_priv,
-			       TRANS_DDI_FUNC_CTL2(dev_priv, cpu_transcoder),
+	if (DISPLAY_VER(display) >= 11)
+		intel_de_write(display,
+			       TRANS_DDI_FUNC_CTL2(display, cpu_transcoder),
 			       0);
 
-	ctl = intel_de_read(dev_priv,
-			    TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder));
+	ctl = intel_de_read(display,
+			    TRANS_DDI_FUNC_CTL(display, cpu_transcoder));
 
 	drm_WARN_ON(crtc->base.dev, ctl & TRANS_DDI_HDCP_SIGNALLING);
 
 	ctl &= ~TRANS_DDI_FUNC_ENABLE;
 
-	if (IS_DISPLAY_VER(dev_priv, 8, 10))
+	if (IS_DISPLAY_VER(display, 8, 10))
 		ctl &= ~(TRANS_DDI_PORT_SYNC_ENABLE |
 			 TRANS_DDI_PORT_SYNC_MASTER_SELECT_MASK);
 
-	if (DISPLAY_VER(dev_priv) >= 12) {
+	if (DISPLAY_VER(display) >= 12) {
 		if (!intel_dp_mst_is_master_trans(crtc_state)) {
 			ctl &= ~(TGL_TRANS_DDI_PORT_MASK |
 				 TRANS_DDI_MODE_SELECT_MASK);
@@ -706,7 +701,7 @@ void intel_ddi_disable_transcoder_func(const struct intel_crtc_state *crtc_state
 		ctl &= ~(TRANS_DDI_PORT_MASK | TRANS_DDI_MODE_SELECT_MASK);
 	}
 
-	intel_de_write(dev_priv, TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder),
+	intel_de_write(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder),
 		       ctl);
 
 	if (intel_dp_mst_is_slave_trans(crtc_state))
@@ -725,17 +720,15 @@ int intel_ddi_toggle_hdcp_bits(struct intel_encoder *intel_encoder,
 			       bool enable, u32 hdcp_mask)
 {
 	struct intel_display *display = to_intel_display(intel_encoder);
-	struct drm_device *dev = intel_encoder->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	intel_wakeref_t wakeref;
 	int ret = 0;
 
 	wakeref = intel_display_power_get_if_enabled(display,
 						     intel_encoder->power_domain);
-	if (drm_WARN_ON(dev, !wakeref))
+	if (drm_WARN_ON(display->drm, !wakeref))
 		return -ENXIO;
 
-	intel_de_rmw(dev_priv, TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder),
+	intel_de_rmw(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder),
 		     hdcp_mask, enable ? hdcp_mask : 0);
 	intel_display_power_put(display, intel_encoder->power_domain, wakeref);
 	return ret;
@@ -744,7 +737,6 @@ int intel_ddi_toggle_hdcp_bits(struct intel_encoder *intel_encoder,
 bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector)
 {
 	struct intel_display *display = to_intel_display(intel_connector);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_encoder *encoder = intel_attached_encoder(intel_connector);
 	int type = intel_connector->base.connector_type;
 	enum port port = encoder->port;
@@ -765,12 +757,12 @@ bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector)
 		goto out;
 	}
 
-	if (HAS_TRANSCODER(dev_priv, TRANSCODER_EDP) && port == PORT_A)
+	if (HAS_TRANSCODER(display, TRANSCODER_EDP) && port == PORT_A)
 		cpu_transcoder = TRANSCODER_EDP;
 	else
 		cpu_transcoder = (enum transcoder) pipe;
 
-	ddi_mode = intel_de_read(dev_priv, TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder)) &
+	ddi_mode = intel_de_read(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder)) &
 		TRANS_DDI_MODE_SELECT_MASK;
 
 	if (ddi_mode == TRANS_DDI_MODE_SELECT_HDMI ||
@@ -804,7 +796,6 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 					u8 *pipe_mask, bool *is_dp_mst)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum port port = encoder->port;
 	intel_wakeref_t wakeref;
 	enum pipe p;
@@ -819,13 +810,13 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 	if (!wakeref)
 		return;
 
-	tmp = intel_de_read(dev_priv, DDI_BUF_CTL(port));
+	tmp = intel_de_read(display, DDI_BUF_CTL(port));
 	if (!(tmp & DDI_BUF_CTL_ENABLE))
 		goto out;
 
-	if (HAS_TRANSCODER(dev_priv, TRANSCODER_EDP) && port == PORT_A) {
-		tmp = intel_de_read(dev_priv,
-				    TRANS_DDI_FUNC_CTL(dev_priv, TRANSCODER_EDP));
+	if (HAS_TRANSCODER(display, TRANSCODER_EDP) && port == PORT_A) {
+		tmp = intel_de_read(display,
+				    TRANS_DDI_FUNC_CTL(display, TRANSCODER_EDP));
 
 		switch (tmp & TRANS_DDI_EDP_INPUT_MASK) {
 		default:
@@ -846,7 +837,7 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 		goto out;
 	}
 
-	for_each_pipe(dev_priv, p) {
+	for_each_pipe(display, p) {
 		enum transcoder cpu_transcoder = (enum transcoder)p;
 		u32 port_mask, ddi_select, ddi_mode;
 		intel_wakeref_t trans_wakeref;
@@ -856,7 +847,7 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 		if (!trans_wakeref)
 			continue;
 
-		if (DISPLAY_VER(dev_priv) >= 12) {
+		if (DISPLAY_VER(display) >= 12) {
 			port_mask = TGL_TRANS_DDI_PORT_MASK;
 			ddi_select = TGL_TRANS_DDI_SELECT_PORT(port);
 		} else {
@@ -864,8 +855,8 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 			ddi_select = TRANS_DDI_SELECT_PORT(port);
 		}
 
-		tmp = intel_de_read(dev_priv,
-				    TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder));
+		tmp = intel_de_read(display,
+				    TRANS_DDI_FUNC_CTL(display, cpu_transcoder));
 		intel_display_power_put(display, POWER_DOMAIN_TRANSCODER(cpu_transcoder),
 					trans_wakeref);
 
@@ -883,12 +874,12 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 	}
 
 	if (!*pipe_mask)
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "No pipe for [ENCODER:%d:%s] found\n",
 			    encoder->base.base.id, encoder->base.name);
 
 	if (!mst_pipe_mask && dp128b132b_pipe_mask) {
-		struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
 		/*
 		 * If we don't have 8b/10b MST, but have more than one
@@ -901,12 +892,12 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 		 * can assume it's SST.
 		 */
 		if (hweight8(dp128b132b_pipe_mask) > 1 ||
-		    intel_dp_mst_encoder_active_links(dig_port))
+		    intel_dp_mst_active_streams(intel_dp))
 			mst_pipe_mask = dp128b132b_pipe_mask;
 	}
 
 	if (!mst_pipe_mask && hweight8(*pipe_mask) > 1) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Multiple pipes for [ENCODER:%d:%s] (pipe_mask %02x)\n",
 			    encoder->base.base.id, encoder->base.name,
 			    *pipe_mask);
@@ -914,7 +905,7 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 	}
 
 	if (mst_pipe_mask && mst_pipe_mask != *pipe_mask)
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Conflicting MST and non-MST state for [ENCODER:%d:%s] (pipe masks: all %02x, MST %02x, 128b/132b %02x)\n",
 			    encoder->base.base.id, encoder->base.name,
 			    *pipe_mask, mst_pipe_mask, dp128b132b_pipe_mask);
@@ -922,12 +913,12 @@ static void intel_ddi_get_encoder_pipes(struct intel_encoder *encoder,
 		*is_dp_mst = mst_pipe_mask;
 
 out:
-	if (*pipe_mask && (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))) {
-		tmp = intel_de_read(dev_priv, BXT_PHY_CTL(port));
+	if (*pipe_mask && (display->platform.geminilake || display->platform.broxton)) {
+		tmp = intel_de_read(display, BXT_PHY_CTL(port));
 		if ((tmp & (BXT_PHY_CMNLANE_POWERDOWN_ACK |
 			    BXT_PHY_LANE_POWERDOWN_ACK |
 			    BXT_PHY_LANE_ENABLED)) != BXT_PHY_LANE_ENABLED)
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"[ENCODER:%d:%s] enabled but PHY powered down? (PHY_CTL %08x)\n",
 				encoder->base.base.id, encoder->base.name, tmp);
 	}
@@ -1041,8 +1032,7 @@ static void intel_ddi_get_power_domains(struct intel_encoder *encoder,
 void intel_ddi_enable_transcoder_clock(struct intel_encoder *encoder,
 				       const struct intel_crtc_state *crtc_state)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	enum phy phy = intel_encoder_to_phy(encoder);
 	u32 val;
@@ -1050,53 +1040,53 @@ void intel_ddi_enable_transcoder_clock(struct intel_encoder *encoder,
 	if (cpu_transcoder == TRANSCODER_EDP)
 		return;
 
-	if (DISPLAY_VER(dev_priv) >= 13)
+	if (DISPLAY_VER(display) >= 13)
 		val = TGL_TRANS_CLK_SEL_PORT(phy);
-	else if (DISPLAY_VER(dev_priv) >= 12)
+	else if (DISPLAY_VER(display) >= 12)
 		val = TGL_TRANS_CLK_SEL_PORT(encoder->port);
 	else
 		val = TRANS_CLK_SEL_PORT(encoder->port);
 
-	intel_de_write(dev_priv, TRANS_CLK_SEL(cpu_transcoder), val);
+	intel_de_write(display, TRANS_CLK_SEL(cpu_transcoder), val);
 }
 
 void intel_ddi_disable_transcoder_clock(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 val;
 
 	if (cpu_transcoder == TRANSCODER_EDP)
 		return;
 
-	if (DISPLAY_VER(dev_priv) >= 12)
+	if (DISPLAY_VER(display) >= 12)
 		val = TGL_TRANS_CLK_SEL_DISABLED;
 	else
 		val = TRANS_CLK_SEL_DISABLED;
 
-	intel_de_write(dev_priv, TRANS_CLK_SEL(cpu_transcoder), val);
+	intel_de_write(display, TRANS_CLK_SEL(cpu_transcoder), val);
 }
 
-static void _skl_ddi_set_iboost(struct drm_i915_private *dev_priv,
+static void _skl_ddi_set_iboost(struct intel_display *display,
 				enum port port, u8 iboost)
 {
 	u32 tmp;
 
-	tmp = intel_de_read(dev_priv, DISPIO_CR_TX_BMU_CR0);
+	tmp = intel_de_read(display, DISPIO_CR_TX_BMU_CR0);
 	tmp &= ~(BALANCE_LEG_MASK(port) | BALANCE_LEG_DISABLE(port));
 	if (iboost)
 		tmp |= iboost << BALANCE_LEG_SHIFT(port);
 	else
 		tmp |= BALANCE_LEG_DISABLE(port);
-	intel_de_write(dev_priv, DISPIO_CR_TX_BMU_CR0, tmp);
+	intel_de_write(display, DISPIO_CR_TX_BMU_CR0, tmp);
 }
 
 static void skl_ddi_set_iboost(struct intel_encoder *encoder,
 			       const struct intel_crtc_state *crtc_state,
 			       int level)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u8 iboost;
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
@@ -1109,7 +1099,7 @@ static void skl_ddi_set_iboost(struct intel_encoder *encoder,
 		int n_entries;
 
 		trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
-		if (drm_WARN_ON_ONCE(&dev_priv->drm, !trans))
+		if (drm_WARN_ON_ONCE(display->drm, !trans))
 			return;
 
 		iboost = trans->entries[level].hsw.i_boost;
@@ -1117,28 +1107,28 @@ static void skl_ddi_set_iboost(struct intel_encoder *encoder,
 
 	/* Make sure that the requested I_boost is valid */
 	if (iboost && iboost != 0x1 && iboost != 0x3 && iboost != 0x7) {
-		drm_err(&dev_priv->drm, "Invalid I_boost value %u\n", iboost);
+		drm_err(display->drm, "Invalid I_boost value %u\n", iboost);
 		return;
 	}
 
-	_skl_ddi_set_iboost(dev_priv, encoder->port, iboost);
+	_skl_ddi_set_iboost(display, encoder->port, iboost);
 
 	if (encoder->port == PORT_A && dig_port->max_lanes == 4)
-		_skl_ddi_set_iboost(dev_priv, PORT_E, iboost);
+		_skl_ddi_set_iboost(display, PORT_E, iboost);
 }
 
 static u8 intel_ddi_dp_voltage_max(struct intel_dp *intel_dp,
 				   const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	int n_entries;
 
 	encoder->get_buf_trans(encoder, crtc_state, &n_entries);
 
-	if (drm_WARN_ON(&dev_priv->drm, n_entries < 1))
+	if (drm_WARN_ON(display->drm, n_entries < 1))
 		n_entries = 1;
-	if (drm_WARN_ON(&dev_priv->drm,
+	if (drm_WARN_ON(display->drm,
 			n_entries > ARRAY_SIZE(index_to_dp_signal_levels)))
 		n_entries = ARRAY_SIZE(index_to_dp_signal_levels);
 
@@ -1171,14 +1161,14 @@ static u32 icl_combo_phy_loadgen_select(const struct intel_crtc_state *crtc_stat
 static void icl_ddi_combo_vswing_program(struct intel_encoder *encoder,
 					 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_ddi_buf_trans *trans;
 	enum phy phy = intel_encoder_to_phy(encoder);
 	int n_entries, ln;
 	u32 val;
 
 	trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
-	if (drm_WARN_ON_ONCE(&dev_priv->drm, !trans))
+	if (drm_WARN_ON_ONCE(display->drm, !trans))
 		return;
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP)) {
@@ -1186,25 +1176,25 @@ static void icl_ddi_combo_vswing_program(struct intel_encoder *encoder,
 
 		val = EDP4K2K_MODE_OVRD_EN | EDP4K2K_MODE_OVRD_OPTIMIZED;
 		intel_dp->hobl_active = is_hobl_buf_trans(trans);
-		intel_de_rmw(dev_priv, ICL_PORT_CL_DW10(phy), val,
+		intel_de_rmw(display, ICL_PORT_CL_DW10(phy), val,
 			     intel_dp->hobl_active ? val : 0);
 	}
 
 	/* Set PORT_TX_DW5 */
-	val = intel_de_read(dev_priv, ICL_PORT_TX_DW5_LN(0, phy));
+	val = intel_de_read(display, ICL_PORT_TX_DW5_LN(0, phy));
 	val &= ~(SCALING_MODE_SEL_MASK | RTERM_SELECT_MASK |
 		 COEFF_POLARITY | CURSOR_PROGRAM |
 		 TAP2_DISABLE | TAP3_DISABLE);
 	val |= SCALING_MODE_SEL(0x2);
 	val |= RTERM_SELECT(0x6);
 	val |= TAP3_DISABLE;
-	intel_de_write(dev_priv, ICL_PORT_TX_DW5_GRP(phy), val);
+	intel_de_write(display, ICL_PORT_TX_DW5_GRP(phy), val);
 
 	/* Program PORT_TX_DW2 */
 	for (ln = 0; ln < 4; ln++) {
 		int level = intel_ddi_level(encoder, crtc_state, ln);
 
-		intel_de_rmw(dev_priv, ICL_PORT_TX_DW2_LN(ln, phy),
+		intel_de_rmw(display, ICL_PORT_TX_DW2_LN(ln, phy),
 			     SWING_SEL_UPPER_MASK | SWING_SEL_LOWER_MASK | RCOMP_SCALAR_MASK,
 			     SWING_SEL_UPPER(trans->entries[level].icl.dw2_swing_sel) |
 			     SWING_SEL_LOWER(trans->entries[level].icl.dw2_swing_sel) |
@@ -1216,7 +1206,7 @@ static void icl_ddi_combo_vswing_program(struct intel_encoder *encoder,
 	for (ln = 0; ln < 4; ln++) {
 		int level = intel_ddi_level(encoder, crtc_state, ln);
 
-		intel_de_rmw(dev_priv, ICL_PORT_TX_DW4_LN(ln, phy),
+		intel_de_rmw(display, ICL_PORT_TX_DW4_LN(ln, phy),
 			     POST_CURSOR_1_MASK | POST_CURSOR_2_MASK | CURSOR_COEFF_MASK,
 			     POST_CURSOR_1(trans->entries[level].icl.dw4_post_cursor_1) |
 			     POST_CURSOR_2(trans->entries[level].icl.dw4_post_cursor_2) |
@@ -1227,7 +1217,7 @@ static void icl_ddi_combo_vswing_program(struct intel_encoder *encoder,
 	for (ln = 0; ln < 4; ln++) {
 		int level = intel_ddi_level(encoder, crtc_state, ln);
 
-		intel_de_rmw(dev_priv, ICL_PORT_TX_DW7_LN(ln, phy),
+		intel_de_rmw(display, ICL_PORT_TX_DW7_LN(ln, phy),
 			     N_SCALAR_MASK,
 			     N_SCALAR(trans->entries[level].icl.dw7_n_scalar));
 	}
@@ -1236,7 +1226,7 @@ static void icl_ddi_combo_vswing_program(struct intel_encoder *encoder,
 static void icl_combo_phy_set_signal_levels(struct intel_encoder *encoder,
 					    const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 	u32 val;
 	int ln;
@@ -1246,12 +1236,12 @@ static void icl_combo_phy_set_signal_levels(struct intel_encoder *encoder,
 	 * set PORT_PCS_DW1 cmnkeeper_enable to 1b,
 	 * else clear to 0b.
 	 */
-	val = intel_de_read(dev_priv, ICL_PORT_PCS_DW1_LN(0, phy));
+	val = intel_de_read(display, ICL_PORT_PCS_DW1_LN(0, phy));
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
 		val &= ~COMMON_KEEPER_EN;
 	else
 		val |= COMMON_KEEPER_EN;
-	intel_de_write(dev_priv, ICL_PORT_PCS_DW1_GRP(phy), val);
+	intel_de_write(display, ICL_PORT_PCS_DW1_GRP(phy), val);
 
 	/* 2. Program loadgen select */
 	/*
@@ -1261,33 +1251,33 @@ static void icl_combo_phy_set_signal_levels(struct intel_encoder *encoder,
 	 * > 6 GHz (LN0=0, LN1=0, LN2=0, LN3=0)
 	 */
 	for (ln = 0; ln < 4; ln++) {
-		intel_de_rmw(dev_priv, ICL_PORT_TX_DW4_LN(ln, phy),
+		intel_de_rmw(display, ICL_PORT_TX_DW4_LN(ln, phy),
 			     LOADGEN_SELECT,
 			     icl_combo_phy_loadgen_select(crtc_state, ln));
 	}
 
 	/* 3. Set PORT_CL_DW5 SUS Clock Config to 11b */
-	intel_de_rmw(dev_priv, ICL_PORT_CL_DW5(phy),
+	intel_de_rmw(display, ICL_PORT_CL_DW5(phy),
 		     0, SUS_CLOCK_CONFIG);
 
 	/* 4. Clear training enable to change swing values */
-	val = intel_de_read(dev_priv, ICL_PORT_TX_DW5_LN(0, phy));
+	val = intel_de_read(display, ICL_PORT_TX_DW5_LN(0, phy));
 	val &= ~TX_TRAINING_EN;
-	intel_de_write(dev_priv, ICL_PORT_TX_DW5_GRP(phy), val);
+	intel_de_write(display, ICL_PORT_TX_DW5_GRP(phy), val);
 
 	/* 5. Program swing and de-emphasis */
 	icl_ddi_combo_vswing_program(encoder, crtc_state);
 
 	/* 6. Set training enable to trigger update */
-	val = intel_de_read(dev_priv, ICL_PORT_TX_DW5_LN(0, phy));
+	val = intel_de_read(display, ICL_PORT_TX_DW5_LN(0, phy));
 	val |= TX_TRAINING_EN;
-	intel_de_write(dev_priv, ICL_PORT_TX_DW5_GRP(phy), val);
+	intel_de_write(display, ICL_PORT_TX_DW5_GRP(phy), val);
 }
 
 static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 					 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum tc_port tc_port = intel_encoder_to_tc(encoder);
 	const struct intel_ddi_buf_trans *trans;
 	int n_entries, ln;
@@ -1296,13 +1286,13 @@ static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 		return;
 
 	trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
-	if (drm_WARN_ON_ONCE(&dev_priv->drm, !trans))
+	if (drm_WARN_ON_ONCE(display->drm, !trans))
 		return;
 
 	for (ln = 0; ln < 2; ln++) {
-		intel_de_rmw(dev_priv, MG_TX1_LINK_PARAMS(ln, tc_port),
+		intel_de_rmw(display, MG_TX1_LINK_PARAMS(ln, tc_port),
 			     CRI_USE_FS32, 0);
-		intel_de_rmw(dev_priv, MG_TX2_LINK_PARAMS(ln, tc_port),
+		intel_de_rmw(display, MG_TX2_LINK_PARAMS(ln, tc_port),
 			     CRI_USE_FS32, 0);
 	}
 
@@ -1312,13 +1302,13 @@ static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 
 		level = intel_ddi_level(encoder, crtc_state, 2*ln+0);
 
-		intel_de_rmw(dev_priv, MG_TX1_SWINGCTRL(ln, tc_port),
+		intel_de_rmw(display, MG_TX1_SWINGCTRL(ln, tc_port),
 			     CRI_TXDEEMPH_OVERRIDE_17_12_MASK,
 			     CRI_TXDEEMPH_OVERRIDE_17_12(trans->entries[level].mg.cri_txdeemph_override_17_12));
 
 		level = intel_ddi_level(encoder, crtc_state, 2*ln+1);
 
-		intel_de_rmw(dev_priv, MG_TX2_SWINGCTRL(ln, tc_port),
+		intel_de_rmw(display, MG_TX2_SWINGCTRL(ln, tc_port),
 			     CRI_TXDEEMPH_OVERRIDE_17_12_MASK,
 			     CRI_TXDEEMPH_OVERRIDE_17_12(trans->entries[level].mg.cri_txdeemph_override_17_12));
 	}
@@ -1329,7 +1319,7 @@ static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 
 		level = intel_ddi_level(encoder, crtc_state, 2*ln+0);
 
-		intel_de_rmw(dev_priv, MG_TX1_DRVCTRL(ln, tc_port),
+		intel_de_rmw(display, MG_TX1_DRVCTRL(ln, tc_port),
 			     CRI_TXDEEMPH_OVERRIDE_11_6_MASK |
 			     CRI_TXDEEMPH_OVERRIDE_5_0_MASK,
 			     CRI_TXDEEMPH_OVERRIDE_11_6(trans->entries[level].mg.cri_txdeemph_override_11_6) |
@@ -1338,7 +1328,7 @@ static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 
 		level = intel_ddi_level(encoder, crtc_state, 2*ln+1);
 
-		intel_de_rmw(dev_priv, MG_TX2_DRVCTRL(ln, tc_port),
+		intel_de_rmw(display, MG_TX2_DRVCTRL(ln, tc_port),
 			     CRI_TXDEEMPH_OVERRIDE_11_6_MASK |
 			     CRI_TXDEEMPH_OVERRIDE_5_0_MASK,
 			     CRI_TXDEEMPH_OVERRIDE_11_6(trans->entries[level].mg.cri_txdeemph_override_11_6) |
@@ -1354,21 +1344,21 @@ static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 	 * values from table for which TX1 and TX2 enabled.
 	 */
 	for (ln = 0; ln < 2; ln++) {
-		intel_de_rmw(dev_priv, MG_CLKHUB(ln, tc_port),
+		intel_de_rmw(display, MG_CLKHUB(ln, tc_port),
 			     CFG_LOW_RATE_LKREN_EN,
 			     crtc_state->port_clock < 300000 ? CFG_LOW_RATE_LKREN_EN : 0);
 	}
 
 	/* Program the MG_TX_DCC<LN, port being used> based on the link frequency */
 	for (ln = 0; ln < 2; ln++) {
-		intel_de_rmw(dev_priv, MG_TX1_DCC(ln, tc_port),
+		intel_de_rmw(display, MG_TX1_DCC(ln, tc_port),
 			     CFG_AMI_CK_DIV_OVERRIDE_VAL_MASK |
 			     CFG_AMI_CK_DIV_OVERRIDE_EN,
 			     crtc_state->port_clock > 500000 ?
 			     CFG_AMI_CK_DIV_OVERRIDE_VAL(1) |
 			     CFG_AMI_CK_DIV_OVERRIDE_EN : 0);
 
-		intel_de_rmw(dev_priv, MG_TX2_DCC(ln, tc_port),
+		intel_de_rmw(display, MG_TX2_DCC(ln, tc_port),
 			     CFG_AMI_CK_DIV_OVERRIDE_VAL_MASK |
 			     CFG_AMI_CK_DIV_OVERRIDE_EN,
 			     crtc_state->port_clock > 500000 ?
@@ -1378,9 +1368,9 @@ static void icl_mg_phy_set_signal_levels(struct intel_encoder *encoder,
 
 	/* Program MG_TX_PISO_READLOAD with values from vswing table */
 	for (ln = 0; ln < 2; ln++) {
-		intel_de_rmw(dev_priv, MG_TX1_PISO_READLOAD(ln, tc_port),
+		intel_de_rmw(display, MG_TX1_PISO_READLOAD(ln, tc_port),
 			     0, CRI_CALCINIT);
-		intel_de_rmw(dev_priv, MG_TX2_PISO_READLOAD(ln, tc_port),
+		intel_de_rmw(display, MG_TX2_PISO_READLOAD(ln, tc_port),
 			     0, CRI_CALCINIT);
 	}
 }
@@ -1490,12 +1480,12 @@ int intel_ddi_level(struct intel_encoder *encoder,
 		    const struct intel_crtc_state *crtc_state,
 		    int lane)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_ddi_buf_trans *trans;
 	int level, n_entries;
 
 	trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
-	if (drm_WARN_ON_ONCE(&i915->drm, !trans))
+	if (drm_WARN_ON_ONCE(display->drm, !trans))
 		return 0;
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
@@ -1504,7 +1494,7 @@ int intel_ddi_level(struct intel_encoder *encoder,
 		level = intel_ddi_dp_level(enc_to_intel_dp(encoder), crtc_state,
 					   lane);
 
-	if (drm_WARN_ON_ONCE(&i915->drm, level >= n_entries))
+	if (drm_WARN_ON_ONCE(display->drm, level >= n_entries))
 		level = n_entries - 1;
 
 	return level;
@@ -1514,13 +1504,13 @@ static void
 hsw_set_signal_levels(struct intel_encoder *encoder,
 		      const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	int level = intel_ddi_level(encoder, crtc_state, 0);
 	enum port port = encoder->port;
 	u32 signal_levels;
 
-	if (has_iboost(dev_priv))
+	if (has_iboost(display))
 		skl_ddi_set_iboost(encoder, crtc_state, level);
 
 	/* HDMI ignores the rest */
@@ -1529,46 +1519,46 @@ hsw_set_signal_levels(struct intel_encoder *encoder,
 
 	signal_levels = DDI_BUF_TRANS_SELECT(level);
 
-	drm_dbg_kms(&dev_priv->drm, "Using signal levels %08x\n",
+	drm_dbg_kms(display->drm, "Using signal levels %08x\n",
 		    signal_levels);
 
 	intel_dp->DP &= ~DDI_BUF_EMP_MASK;
 	intel_dp->DP |= signal_levels;
 
-	intel_de_write(dev_priv, DDI_BUF_CTL(port), intel_dp->DP);
-	intel_de_posting_read(dev_priv, DDI_BUF_CTL(port));
+	intel_de_write(display, DDI_BUF_CTL(port), intel_dp->DP);
+	intel_de_posting_read(display, DDI_BUF_CTL(port));
 }
 
-static void _icl_ddi_enable_clock(struct drm_i915_private *i915, i915_reg_t reg,
+static void _icl_ddi_enable_clock(struct intel_display *display, i915_reg_t reg,
 				  u32 clk_sel_mask, u32 clk_sel, u32 clk_off)
 {
-	mutex_lock(&i915->display.dpll.lock);
+	mutex_lock(&display->dpll.lock);
 
-	intel_de_rmw(i915, reg, clk_sel_mask, clk_sel);
+	intel_de_rmw(display, reg, clk_sel_mask, clk_sel);
 
 	/*
 	 * "This step and the step before must be
 	 *  done with separate register writes."
 	 */
-	intel_de_rmw(i915, reg, clk_off, 0);
+	intel_de_rmw(display, reg, clk_off, 0);
 
-	mutex_unlock(&i915->display.dpll.lock);
+	mutex_unlock(&display->dpll.lock);
 }
 
-static void _icl_ddi_disable_clock(struct drm_i915_private *i915, i915_reg_t reg,
+static void _icl_ddi_disable_clock(struct intel_display *display, i915_reg_t reg,
 				   u32 clk_off)
 {
-	mutex_lock(&i915->display.dpll.lock);
+	mutex_lock(&display->dpll.lock);
 
-	intel_de_rmw(i915, reg, 0, clk_off);
+	intel_de_rmw(display, reg, 0, clk_off);
 
-	mutex_unlock(&i915->display.dpll.lock);
+	mutex_unlock(&display->dpll.lock);
 }
 
-static bool _icl_ddi_is_clock_enabled(struct drm_i915_private *i915, i915_reg_t reg,
+static bool _icl_ddi_is_clock_enabled(struct intel_display *display, i915_reg_t reg,
 				      u32 clk_off)
 {
-	return !(intel_de_read(i915, reg) & clk_off);
+	return !(intel_de_read(display, reg) & clk_off);
 }
 
 static struct intel_shared_dpll *
@@ -1585,14 +1575,14 @@ _icl_ddi_get_pll(struct intel_display *display, i915_reg_t reg,
 static void adls_ddi_enable_clock(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
-	_icl_ddi_enable_clock(i915, ADLS_DPCLKA_CFGCR(phy),
+	_icl_ddi_enable_clock(display, ADLS_DPCLKA_CFGCR(phy),
 			      ADLS_DPCLKA_CFGCR_DDI_CLK_SEL_MASK(phy),
 			      pll->info->id << ADLS_DPCLKA_CFGCR_DDI_SHIFT(phy),
 			      ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
@@ -1600,19 +1590,19 @@ static void adls_ddi_enable_clock(struct intel_encoder *encoder,
 
 static void adls_ddi_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	_icl_ddi_disable_clock(i915, ADLS_DPCLKA_CFGCR(phy),
+	_icl_ddi_disable_clock(display, ADLS_DPCLKA_CFGCR(phy),
 			       ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
 static bool adls_ddi_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	return _icl_ddi_is_clock_enabled(i915, ADLS_DPCLKA_CFGCR(phy),
+	return _icl_ddi_is_clock_enabled(display, ADLS_DPCLKA_CFGCR(phy),
 					 ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
@@ -1629,14 +1619,14 @@ static struct intel_shared_dpll *adls_ddi_get_pll(struct intel_encoder *encoder)
 static void rkl_ddi_enable_clock(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
-	_icl_ddi_enable_clock(i915, ICL_DPCLKA_CFGCR0,
+	_icl_ddi_enable_clock(display, ICL_DPCLKA_CFGCR0,
 			      RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy),
 			      RKL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, phy),
 			      RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
@@ -1644,19 +1634,19 @@ static void rkl_ddi_enable_clock(struct intel_encoder *encoder,
 
 static void rkl_ddi_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	_icl_ddi_disable_clock(i915, ICL_DPCLKA_CFGCR0,
+	_icl_ddi_disable_clock(display, ICL_DPCLKA_CFGCR0,
 			       RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
 static bool rkl_ddi_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	return _icl_ddi_is_clock_enabled(i915, ICL_DPCLKA_CFGCR0,
+	return _icl_ddi_is_clock_enabled(display, ICL_DPCLKA_CFGCR0,
 					 RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
@@ -1673,23 +1663,23 @@ static struct intel_shared_dpll *rkl_ddi_get_pll(struct intel_encoder *encoder)
 static void dg1_ddi_enable_clock(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
 	/*
 	 * If we fail this, something went very wrong: first 2 PLLs should be
 	 * used by first 2 phys and last 2 PLLs by last phys
 	 */
-	if (drm_WARN_ON(&i915->drm,
+	if (drm_WARN_ON(display->drm,
 			(pll->info->id < DPLL_ID_DG1_DPLL2 && phy >= PHY_C) ||
 			(pll->info->id >= DPLL_ID_DG1_DPLL2 && phy < PHY_C)))
 		return;
 
-	_icl_ddi_enable_clock(i915, DG1_DPCLKA_CFGCR0(phy),
+	_icl_ddi_enable_clock(display, DG1_DPCLKA_CFGCR0(phy),
 			      DG1_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy),
 			      DG1_DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, phy),
 			      DG1_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
@@ -1697,19 +1687,19 @@ static void dg1_ddi_enable_clock(struct intel_encoder *encoder,
 
 static void dg1_ddi_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	_icl_ddi_disable_clock(i915, DG1_DPCLKA_CFGCR0(phy),
+	_icl_ddi_disable_clock(display, DG1_DPCLKA_CFGCR0(phy),
 			       DG1_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
 static bool dg1_ddi_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	return _icl_ddi_is_clock_enabled(i915, DG1_DPCLKA_CFGCR0(phy),
+	return _icl_ddi_is_clock_enabled(display, DG1_DPCLKA_CFGCR0(phy),
 					 DG1_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
@@ -1739,14 +1729,14 @@ static struct intel_shared_dpll *dg1_ddi_get_pll(struct intel_encoder *encoder)
 static void icl_ddi_combo_enable_clock(struct intel_encoder *encoder,
 				       const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
-	_icl_ddi_enable_clock(i915, ICL_DPCLKA_CFGCR0,
+	_icl_ddi_enable_clock(display, ICL_DPCLKA_CFGCR0,
 			      ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy),
 			      ICL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, phy),
 			      ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
@@ -1754,19 +1744,19 @@ static void icl_ddi_combo_enable_clock(struct intel_encoder *encoder,
 
 static void icl_ddi_combo_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	_icl_ddi_disable_clock(i915, ICL_DPCLKA_CFGCR0,
+	_icl_ddi_disable_clock(display, ICL_DPCLKA_CFGCR0,
 			       ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
 static bool icl_ddi_combo_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
-	return _icl_ddi_is_clock_enabled(i915, ICL_DPCLKA_CFGCR0,
+	return _icl_ddi_is_clock_enabled(display, ICL_DPCLKA_CFGCR0,
 					 ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy));
 }
 
@@ -1783,39 +1773,39 @@ struct intel_shared_dpll *icl_ddi_combo_get_pll(struct intel_encoder *encoder)
 static void jsl_ddi_tc_enable_clock(struct intel_encoder *encoder,
 				    const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum port port = encoder->port;
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
 	/*
 	 * "For DDIC and DDID, program DDI_CLK_SEL to map the MG clock to the port.
 	 *  MG does not exist, but the programming is required to ungate DDIC and DDID."
 	 */
-	intel_de_write(i915, DDI_CLK_SEL(port), DDI_CLK_SEL_MG);
+	intel_de_write(display, DDI_CLK_SEL(port), DDI_CLK_SEL_MG);
 
 	icl_ddi_combo_enable_clock(encoder, crtc_state);
 }
 
 static void jsl_ddi_tc_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 
 	icl_ddi_combo_disable_clock(encoder);
 
-	intel_de_write(i915, DDI_CLK_SEL(port), DDI_CLK_SEL_NONE);
+	intel_de_write(display, DDI_CLK_SEL(port), DDI_CLK_SEL_NONE);
 }
 
 static bool jsl_ddi_tc_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 	u32 tmp;
 
-	tmp = intel_de_read(i915, DDI_CLK_SEL(port));
+	tmp = intel_de_read(display, DDI_CLK_SEL(port));
 
 	if ((tmp & DDI_CLK_SEL_MASK) == DDI_CLK_SEL_NONE)
 		return false;
@@ -1826,54 +1816,54 @@ static bool jsl_ddi_tc_is_clock_enabled(struct intel_encoder *encoder)
 static void icl_ddi_tc_enable_clock(struct intel_encoder *encoder,
 				    const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum tc_port tc_port = intel_encoder_to_tc(encoder);
 	enum port port = encoder->port;
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
-	intel_de_write(i915, DDI_CLK_SEL(port),
+	intel_de_write(display, DDI_CLK_SEL(port),
 		       icl_pll_to_ddi_clk_sel(encoder, crtc_state));
 
-	mutex_lock(&i915->display.dpll.lock);
+	mutex_lock(&display->dpll.lock);
 
-	intel_de_rmw(i915, ICL_DPCLKA_CFGCR0,
+	intel_de_rmw(display, ICL_DPCLKA_CFGCR0,
 		     ICL_DPCLKA_CFGCR0_TC_CLK_OFF(tc_port), 0);
 
-	mutex_unlock(&i915->display.dpll.lock);
+	mutex_unlock(&display->dpll.lock);
 }
 
 static void icl_ddi_tc_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum tc_port tc_port = intel_encoder_to_tc(encoder);
 	enum port port = encoder->port;
 
-	mutex_lock(&i915->display.dpll.lock);
+	mutex_lock(&display->dpll.lock);
 
-	intel_de_rmw(i915, ICL_DPCLKA_CFGCR0,
+	intel_de_rmw(display, ICL_DPCLKA_CFGCR0,
 		     0, ICL_DPCLKA_CFGCR0_TC_CLK_OFF(tc_port));
 
-	mutex_unlock(&i915->display.dpll.lock);
+	mutex_unlock(&display->dpll.lock);
 
-	intel_de_write(i915, DDI_CLK_SEL(port), DDI_CLK_SEL_NONE);
+	intel_de_write(display, DDI_CLK_SEL(port), DDI_CLK_SEL_NONE);
 }
 
 static bool icl_ddi_tc_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum tc_port tc_port = intel_encoder_to_tc(encoder);
 	enum port port = encoder->port;
 	u32 tmp;
 
-	tmp = intel_de_read(i915, DDI_CLK_SEL(port));
+	tmp = intel_de_read(display, DDI_CLK_SEL(port));
 
 	if ((tmp & DDI_CLK_SEL_MASK) == DDI_CLK_SEL_NONE)
 		return false;
 
-	tmp = intel_de_read(i915, ICL_DPCLKA_CFGCR0);
+	tmp = intel_de_read(display, ICL_DPCLKA_CFGCR0);
 
 	return !(tmp & ICL_DPCLKA_CFGCR0_TC_CLK_OFF(tc_port));
 }
@@ -1934,47 +1924,47 @@ static struct intel_shared_dpll *bxt_ddi_get_pll(struct intel_encoder *encoder)
 static void skl_ddi_enable_clock(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum port port = encoder->port;
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
-	mutex_lock(&i915->display.dpll.lock);
+	mutex_lock(&display->dpll.lock);
 
-	intel_de_rmw(i915, DPLL_CTRL2,
+	intel_de_rmw(display, DPLL_CTRL2,
 		     DPLL_CTRL2_DDI_CLK_OFF(port) |
 		     DPLL_CTRL2_DDI_CLK_SEL_MASK(port),
 		     DPLL_CTRL2_DDI_CLK_SEL(pll->info->id, port) |
 		     DPLL_CTRL2_DDI_SEL_OVERRIDE(port));
 
-	mutex_unlock(&i915->display.dpll.lock);
+	mutex_unlock(&display->dpll.lock);
 }
 
 static void skl_ddi_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 
-	mutex_lock(&i915->display.dpll.lock);
+	mutex_lock(&display->dpll.lock);
 
-	intel_de_rmw(i915, DPLL_CTRL2,
+	intel_de_rmw(display, DPLL_CTRL2,
 		     0, DPLL_CTRL2_DDI_CLK_OFF(port));
 
-	mutex_unlock(&i915->display.dpll.lock);
+	mutex_unlock(&display->dpll.lock);
 }
 
 static bool skl_ddi_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 
 	/*
 	 * FIXME Not sure if the override affects both
 	 * the PLL selection and the CLK_OFF bit.
 	 */
-	return !(intel_de_read(i915, DPLL_CTRL2) & DPLL_CTRL2_DDI_CLK_OFF(port));
+	return !(intel_de_read(display, DPLL_CTRL2) & DPLL_CTRL2_DDI_CLK_OFF(port));
 }
 
 static struct intel_shared_dpll *skl_ddi_get_pll(struct intel_encoder *encoder)
@@ -2002,30 +1992,30 @@ static struct intel_shared_dpll *skl_ddi_get_pll(struct intel_encoder *encoder)
 void hsw_ddi_enable_clock(struct intel_encoder *encoder,
 			  const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 	enum port port = encoder->port;
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return;
 
-	intel_de_write(i915, PORT_CLK_SEL(port), hsw_pll_to_ddi_pll_sel(pll));
+	intel_de_write(display, PORT_CLK_SEL(port), hsw_pll_to_ddi_pll_sel(pll));
 }
 
 void hsw_ddi_disable_clock(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 
-	intel_de_write(i915, PORT_CLK_SEL(port), PORT_CLK_SEL_NONE);
+	intel_de_write(display, PORT_CLK_SEL(port), PORT_CLK_SEL_NONE);
 }
 
 bool hsw_ddi_is_clock_enabled(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 
-	return intel_de_read(i915, PORT_CLK_SEL(port)) != PORT_CLK_SEL_NONE;
+	return intel_de_read(display, PORT_CLK_SEL(port)) != PORT_CLK_SEL_NONE;
 }
 
 static struct intel_shared_dpll *hsw_ddi_get_pll(struct intel_encoder *encoder)
@@ -2081,7 +2071,7 @@ void intel_ddi_disable_clock(struct intel_encoder *encoder)
 
 void intel_ddi_sanitize_encoder_pll_mapping(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	u32 port_mask;
 	bool ddi_clk_needed;
 
@@ -2101,7 +2091,7 @@ void intel_ddi_sanitize_encoder_pll_mapping(struct intel_encoder *encoder)
 		 * In the unlikely case that BIOS enables DP in MST mode, just
 		 * warn since our MST HW readout is incomplete.
 		 */
-		if (drm_WARN_ON(&i915->drm, is_mst))
+		if (drm_WARN_ON(display->drm, is_mst))
 			return;
 	}
 
@@ -2116,11 +2106,11 @@ void intel_ddi_sanitize_encoder_pll_mapping(struct intel_encoder *encoder)
 		 * Sanity check that we haven't incorrectly registered another
 		 * encoder using any of the ports of this DSI encoder.
 		 */
-		for_each_intel_encoder(&i915->drm, other_encoder) {
+		for_each_intel_encoder(display->drm, other_encoder) {
 			if (other_encoder == encoder)
 				continue;
 
-			if (drm_WARN_ON(&i915->drm,
+			if (drm_WARN_ON(display->drm,
 					port_mask & BIT(other_encoder->port)))
 				return;
 		}
@@ -2135,7 +2125,7 @@ void intel_ddi_sanitize_encoder_pll_mapping(struct intel_encoder *encoder)
 	    !encoder->is_clock_enabled(encoder))
 		return;
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "[ENCODER:%d:%s] is disabled/in DSI mode with an ungated DDI clock, gate it\n",
 		    encoder->base.base.id, encoder->base.name);
 
@@ -2255,10 +2245,10 @@ tgl_dp_tp_transcoder(const struct intel_crtc_state *crtc_state)
 i915_reg_t dp_tp_ctl_reg(struct intel_encoder *encoder,
 			 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (DISPLAY_VER(dev_priv) >= 12)
-		return TGL_DP_TP_CTL(dev_priv,
+	if (DISPLAY_VER(display) >= 12)
+		return TGL_DP_TP_CTL(display,
 				     tgl_dp_tp_transcoder(crtc_state));
 	else
 		return DP_TP_CTL(encoder->port);
@@ -2267,10 +2257,10 @@ i915_reg_t dp_tp_ctl_reg(struct intel_encoder *encoder,
 static i915_reg_t dp_tp_status_reg(struct intel_encoder *encoder,
 				   const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (DISPLAY_VER(dev_priv) >= 12)
-		return TGL_DP_TP_STATUS(dev_priv,
+	if (DISPLAY_VER(display) >= 12)
+		return TGL_DP_TP_STATUS(display,
 					tgl_dp_tp_transcoder(crtc_state));
 	else
 		return DP_TP_STATUS(encoder->port);
@@ -2445,14 +2435,14 @@ static void intel_ddi_enable_fec(struct intel_encoder *encoder,
 static void intel_ddi_disable_fec(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
 	if (!crtc_state->fec_enable)
 		return;
 
-	intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state),
+	intel_de_rmw(display, dp_tp_ctl_reg(encoder, crtc_state),
 		     DP_TP_CTL_FEC_ENABLE, 0);
-	intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
+	intel_de_posting_read(display, dp_tp_ctl_reg(encoder, crtc_state));
 }
 
 static void intel_ddi_power_up_lanes(struct intel_encoder *encoder,
@@ -2474,11 +2464,11 @@ static void intel_ddi_power_up_lanes(struct intel_encoder *encoder,
  * Splitter enable for eDP MSO is limited to certain pipes, on certain
  * platforms.
  */
-static u8 intel_ddi_splitter_pipe_mask(struct drm_i915_private *i915)
+static u8 intel_ddi_splitter_pipe_mask(struct intel_display *display)
 {
-	if (DISPLAY_VER(i915) > 20)
+	if (DISPLAY_VER(display) > 20)
 		return ~0;
-	else if (IS_ALDERLAKE_P(i915))
+	else if (display->platform.alderlake_p)
 		return BIT(PIPE_A) | BIT(PIPE_B);
 	else
 		return BIT(PIPE_A);
@@ -2487,28 +2477,28 @@ static u8 intel_ddi_splitter_pipe_mask(struct drm_i915_private *i915)
 static void intel_ddi_mso_get_config(struct intel_encoder *encoder,
 				     struct intel_crtc_state *pipe_config)
 {
+	struct intel_display *display = to_intel_display(pipe_config);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	u32 dss1;
 
-	if (!HAS_MSO(i915))
+	if (!HAS_MSO(display))
 		return;
 
-	dss1 = intel_de_read(i915, ICL_PIPE_DSS_CTL1(pipe));
+	dss1 = intel_de_read(display, ICL_PIPE_DSS_CTL1(pipe));
 
 	pipe_config->splitter.enable = dss1 & SPLITTER_ENABLE;
 	if (!pipe_config->splitter.enable)
 		return;
 
-	if (drm_WARN_ON(&i915->drm, !(intel_ddi_splitter_pipe_mask(i915) & BIT(pipe)))) {
+	if (drm_WARN_ON(display->drm, !(intel_ddi_splitter_pipe_mask(display) & BIT(pipe)))) {
 		pipe_config->splitter.enable = false;
 		return;
 	}
 
 	switch (dss1 & SPLITTER_CONFIGURATION_MASK) {
 	default:
-		drm_WARN(&i915->drm, true,
+		drm_WARN(display->drm, true,
 			 "Invalid splitter configuration, dss1=0x%08x\n", dss1);
 		fallthrough;
 	case SPLITTER_CONFIGURATION_2_SEGMENT:
@@ -2524,12 +2514,12 @@ static void intel_ddi_mso_get_config(struct intel_encoder *encoder,
 
 static void intel_ddi_mso_configure(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	u32 dss1 = 0;
 
-	if (!HAS_MSO(i915))
+	if (!HAS_MSO(display))
 		return;
 
 	if (crtc_state->splitter.enable) {
@@ -2541,7 +2531,7 @@ static void intel_ddi_mso_configure(const struct intel_crtc_state *crtc_state)
 			dss1 |= SPLITTER_CONFIGURATION_4_SEGMENT;
 	}
 
-	intel_de_rmw(i915, ICL_PIPE_DSS_CTL1(pipe),
+	intel_de_rmw(display, ICL_PIPE_DSS_CTL1(pipe),
 		     SPLITTER_ENABLE | SPLITTER_CONFIGURATION_MASK |
 		     OVERLAP_PIXELS_MASK, dss1);
 }
@@ -2549,27 +2539,27 @@ static void intel_ddi_mso_configure(const struct intel_crtc_state *crtc_state)
 static void
 mtl_ddi_enable_d2d(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 	i915_reg_t reg;
 	u32 set_bits, wait_bits;
 
-	if (DISPLAY_VER(dev_priv) < 14)
+	if (DISPLAY_VER(display) < 14)
 		return;
 
-	if (DISPLAY_VER(dev_priv) >= 20) {
+	if (DISPLAY_VER(display) >= 20) {
 		reg = DDI_BUF_CTL(port);
 		set_bits = XE2LPD_DDI_BUF_D2D_LINK_ENABLE;
 		wait_bits = XE2LPD_DDI_BUF_D2D_LINK_STATE;
 	} else {
-		reg = XELPDP_PORT_BUF_CTL1(dev_priv, port);
+		reg = XELPDP_PORT_BUF_CTL1(display, port);
 		set_bits = XELPDP_PORT_BUF_D2D_LINK_ENABLE;
 		wait_bits = XELPDP_PORT_BUF_D2D_LINK_STATE;
 	}
 
-	intel_de_rmw(dev_priv, reg, 0, set_bits);
-	if (wait_for_us(intel_de_read(dev_priv, reg) & wait_bits, 100)) {
-		drm_err(&dev_priv->drm, "Timeout waiting for D2D Link enable for DDI/PORT_BUF_CTL %c\n",
+	intel_de_rmw(display, reg, 0, set_bits);
+	if (wait_for_us(intel_de_read(display, reg) & wait_bits, 100)) {
+		drm_err(display->drm, "Timeout waiting for D2D Link enable for DDI/PORT_BUF_CTL %c\n",
 			port_name(port));
 	}
 }
@@ -2599,13 +2589,13 @@ static void mtl_port_buf_ctl_program(struct intel_encoder *encoder,
 
 static void mtl_port_buf_ctl_io_selection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	u32 val;
 
 	val = intel_tc_port_in_tbt_alt_mode(dig_port) ?
 	      XELPDP_PORT_BUF_IO_SELECT_TBT : 0;
-	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(i915, encoder->port),
+	intel_de_rmw(display, XELPDP_PORT_BUF_CTL1(display, encoder->port),
 		     XELPDP_PORT_BUF_IO_SELECT_TBT, val);
 }
 
@@ -2734,7 +2724,6 @@ static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state,
 {
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST);
 	int ret;
@@ -2778,7 +2767,7 @@ static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state,
 
 	/* 5. If IO power is controlled through PWR_WELL_CTL, Enable IO Power */
 	if (!intel_tc_port_in_tbt_alt_mode(dig_port)) {
-		drm_WARN_ON(&dev_priv->drm, dig_port->ddi_io_wakeref);
+		drm_WARN_ON(display->drm, dig_port->ddi_io_wakeref);
 		dig_port->ddi_io_wakeref = intel_display_power_get(display,
 								   dig_port->ddi_io_power_domain);
 	}
@@ -2882,16 +2871,15 @@ static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state,
 {
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum port port = encoder->port;
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST);
 
-	if (DISPLAY_VER(dev_priv) < 11)
-		drm_WARN_ON(&dev_priv->drm,
+	if (DISPLAY_VER(display) < 11)
+		drm_WARN_ON(display->drm,
 			    is_mst && (port == PORT_A || port == PORT_E));
 	else
-		drm_WARN_ON(&dev_priv->drm, is_mst && port == PORT_A);
+		drm_WARN_ON(display->drm, is_mst && port == PORT_A);
 
 	intel_dp_set_link_params(intel_dp,
 				 crtc_state->port_clock,
@@ -2908,14 +2896,14 @@ static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state,
 	intel_ddi_enable_clock(encoder, crtc_state);
 
 	if (!intel_tc_port_in_tbt_alt_mode(dig_port)) {
-		drm_WARN_ON(&dev_priv->drm, dig_port->ddi_io_wakeref);
+		drm_WARN_ON(display->drm, dig_port->ddi_io_wakeref);
 		dig_port->ddi_io_wakeref = intel_display_power_get(display,
 								   dig_port->ddi_io_power_domain);
 	}
 
 	icl_program_mg_dp_mode(dig_port, crtc_state);
 
-	if (has_buf_trans_select(dev_priv))
+	if (has_buf_trans_select(display))
 		hsw_prepare_dp_ddi_buffers(encoder, crtc_state);
 
 	encoder->set_signal_levels(encoder, crtc_state);
@@ -2931,7 +2919,7 @@ static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state,
 						   crtc_state);
 	intel_dp_sink_set_fec_ready(intel_dp, crtc_state, true);
 	intel_dp_start_link_train(state, intel_dp, crtc_state);
-	if ((port != PORT_A || DISPLAY_VER(dev_priv) >= 9) &&
+	if ((port != PORT_A || DISPLAY_VER(display) >= 9) &&
 	    !is_trans_port_sync_mode(crtc_state))
 		intel_dp_stop_link_train(intel_dp, crtc_state);
 
@@ -2979,12 +2967,11 @@ static void intel_ddi_pre_enable_hdmi(struct intel_atomic_state *state,
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	struct intel_hdmi *intel_hdmi = &dig_port->hdmi;
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
 	intel_dp_dual_mode_set_tmds_output(intel_hdmi, true);
 	intel_ddi_enable_clock(encoder, crtc_state);
 
-	drm_WARN_ON(&dev_priv->drm, dig_port->ddi_io_wakeref);
+	drm_WARN_ON(display->drm, dig_port->ddi_io_wakeref);
 	dig_port->ddi_io_wakeref = intel_display_power_get(display,
 							   dig_port->ddi_io_power_domain);
 
@@ -3022,10 +3009,9 @@ static void intel_ddi_pre_enable(struct intel_atomic_state *state,
 {
 	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
-	drm_WARN_ON(&dev_priv->drm, crtc_state->has_pch_encoder);
+	drm_WARN_ON(display->drm, crtc_state->has_pch_encoder);
 
 	intel_set_cpu_fifo_underrun_reporting(display, pipe, true);
 
@@ -3050,27 +3036,27 @@ static void intel_ddi_pre_enable(struct intel_atomic_state *state,
 static void
 mtl_ddi_disable_d2d(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 	i915_reg_t reg;
 	u32 clr_bits, wait_bits;
 
-	if (DISPLAY_VER(dev_priv) < 14)
+	if (DISPLAY_VER(display) < 14)
 		return;
 
-	if (DISPLAY_VER(dev_priv) >= 20) {
+	if (DISPLAY_VER(display) >= 20) {
 		reg = DDI_BUF_CTL(port);
 		clr_bits = XE2LPD_DDI_BUF_D2D_LINK_ENABLE;
 		wait_bits = XE2LPD_DDI_BUF_D2D_LINK_STATE;
 	} else {
-		reg = XELPDP_PORT_BUF_CTL1(dev_priv, port);
+		reg = XELPDP_PORT_BUF_CTL1(display, port);
 		clr_bits = XELPDP_PORT_BUF_D2D_LINK_ENABLE;
 		wait_bits = XELPDP_PORT_BUF_D2D_LINK_STATE;
 	}
 
-	intel_de_rmw(dev_priv, reg, clr_bits, 0);
-	if (wait_for_us(!(intel_de_read(dev_priv, reg) & wait_bits), 100))
-		drm_err(&dev_priv->drm, "Timeout waiting for D2D Link disable for DDI/PORT_BUF_CTL %c\n",
+	intel_de_rmw(display, reg, clr_bits, 0);
+	if (wait_for_us(!(intel_de_read(display, reg) & wait_bits), 100))
+		drm_err(display->drm, "Timeout waiting for D2D Link disable for DDI/PORT_BUF_CTL %c\n",
 			port_name(port));
 }
 
@@ -3089,10 +3075,9 @@ static void intel_ddi_buf_disable(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum port port = encoder->port;
 
-	intel_de_rmw(dev_priv, DDI_BUF_CTL(port), DDI_BUF_CTL_ENABLE, 0);
+	intel_de_rmw(display, DDI_BUF_CTL(port), DDI_BUF_CTL_ENABLE, 0);
 
 	if (DISPLAY_VER(display) >= 14)
 		intel_wait_ddi_buf_idle(display, port);
@@ -3100,7 +3085,7 @@ static void intel_ddi_buf_disable(struct intel_encoder *encoder,
 	mtl_ddi_disable_d2d(encoder);
 
 	if (intel_crtc_has_dp_encoder(crtc_state)) {
-		intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state),
+		intel_de_rmw(display, dp_tp_ctl_reg(encoder, crtc_state),
 			     DP_TP_CTL_ENABLE, 0);
 	}
 
@@ -3118,7 +3103,6 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 				      const struct drm_connector_state *old_conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	struct intel_dp *intel_dp = &dig_port->dp;
 	intel_wakeref_t wakeref;
@@ -3135,12 +3119,12 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 	 */
 	intel_dp_set_power(intel_dp, DP_SET_POWER_D3);
 
-	if (DISPLAY_VER(dev_priv) >= 12) {
+	if (DISPLAY_VER(display) >= 12) {
 		if (is_mst || intel_dp_is_uhbr(old_crtc_state)) {
 			enum transcoder cpu_transcoder = old_crtc_state->cpu_transcoder;
 
-			intel_de_rmw(dev_priv,
-				     TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder),
+			intel_de_rmw(display,
+				     TRANS_DDI_FUNC_CTL(display, cpu_transcoder),
 				     TGL_TRANS_DDI_PORT_MASK | TRANS_DDI_MODE_SELECT_MASK,
 				     0);
 		}
@@ -3160,7 +3144,7 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 	 * Configure Transcoder Clock select to direct no clock to the
 	 * transcoder"
 	 */
-	if (DISPLAY_VER(dev_priv) >= 12)
+	if (DISPLAY_VER(display) >= 12)
 		intel_ddi_disable_transcoder_clock(old_crtc_state);
 
 	intel_pps_vdd_on(intel_dp);
@@ -3176,8 +3160,8 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 	intel_ddi_disable_clock(encoder);
 
 	/* De-select Thunderbolt */
-	if (DISPLAY_VER(dev_priv) >= 14)
-		intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(dev_priv, encoder->port),
+	if (DISPLAY_VER(display) >= 14)
+		intel_de_rmw(display, XELPDP_PORT_BUF_CTL1(display, encoder->port),
 			     XELPDP_PORT_BUF_IO_SELECT_TBT, 0);
 }
 
@@ -3187,7 +3171,6 @@ static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state,
 					const struct drm_connector_state *old_conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	struct intel_hdmi *intel_hdmi = &dig_port->hdmi;
 	intel_wakeref_t wakeref;
@@ -3195,12 +3178,12 @@ static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state,
 	dig_port->set_infoframes(encoder, false,
 				 old_crtc_state, old_conn_state);
 
-	if (DISPLAY_VER(dev_priv) < 12)
+	if (DISPLAY_VER(display) < 12)
 		intel_ddi_disable_transcoder_clock(old_crtc_state);
 
 	intel_ddi_buf_disable(encoder, old_crtc_state);
 
-	if (DISPLAY_VER(dev_priv) >= 12)
+	if (DISPLAY_VER(display) >= 12)
 		intel_ddi_disable_transcoder_clock(old_crtc_state);
 
 	wakeref = fetch_and_zero(&dig_port->ddi_io_wakeref);
@@ -3220,7 +3203,6 @@ static void intel_ddi_post_disable_hdmi_or_sst(struct intel_atomic_state *state,
 					       const struct drm_connector_state *old_conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct intel_crtc *pipe_crtc;
 	bool is_hdmi = intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_HDMI);
@@ -3249,6 +3231,8 @@ static void intel_ddi_post_disable_hdmi_or_sst(struct intel_atomic_state *state,
 		drm_dp_dpcd_poll_act_handled(&intel_dp->aux, 0);
 	}
 
+	intel_vrr_transcoder_disable(old_crtc_state);
+
 	intel_ddi_disable_transcoder_func(old_crtc_state);
 
 	for_each_pipe_crtc_modeset_disable(display, pipe_crtc, old_crtc_state, i) {
@@ -3257,7 +3241,7 @@ static void intel_ddi_post_disable_hdmi_or_sst(struct intel_atomic_state *state,
 
 		intel_dsc_disable(old_pipe_crtc_state);
 
-		if (DISPLAY_VER(dev_priv) >= 9)
+		if (DISPLAY_VER(display) >= 9)
 			skl_scaler_disable(old_pipe_crtc_state);
 		else
 			ilk_pfit_disable(old_pipe_crtc_state);
@@ -3359,12 +3343,12 @@ static void intel_ddi_enable_dp(struct intel_atomic_state *state,
 				const struct intel_crtc_state *crtc_state,
 				const struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	enum port port = encoder->port;
 
-	if (port == PORT_A && DISPLAY_VER(dev_priv) < 9)
+	if (port == PORT_A && DISPLAY_VER(display) < 9)
 		intel_dp_stop_link_train(intel_dp, crtc_state);
 
 	drm_connector_update_privacy_screen(conn_state);
@@ -3401,7 +3385,6 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 				  const struct drm_connector_state *conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	struct drm_connector *connector = conn_state->connector;
 	enum port port = encoder->port;
@@ -3410,11 +3393,11 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 	if (!intel_hdmi_handle_sink_scrambling(encoder, connector,
 					       crtc_state->hdmi_high_tmds_clock_ratio,
 					       crtc_state->hdmi_scrambling))
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "[CONNECTOR:%d:%s] Failed to configure sink scrambling/TMDS bit clock ratio\n",
 			    connector->base.id, connector->name);
 
-	if (has_buf_trans_select(dev_priv))
+	if (has_buf_trans_select(display))
 		hsw_prepare_hdmi_ddi_buffers(encoder, crtc_state);
 
 	/* e. Enable D2D Link for C10/C20 Phy */
@@ -3423,7 +3406,7 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 	encoder->set_signal_levels(encoder, crtc_state);
 
 	/* Display WA #1143: skl,kbl,cfl */
-	if (DISPLAY_VER(dev_priv) == 9 && !IS_BROXTON(dev_priv)) {
+	if (DISPLAY_VER(display) == 9 && !display->platform.broxton) {
 		/*
 		 * For some reason these chicken bits have been
 		 * stuffed into a transcoder register, event though
@@ -3433,7 +3416,7 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 		i915_reg_t reg = gen9_chicken_trans_reg_by_port(display, port);
 		u32 val;
 
-		val = intel_de_read(dev_priv, reg);
+		val = intel_de_read(display, reg);
 
 		if (port == PORT_E)
 			val |= DDIE_TRAINING_OVERRIDE_ENABLE |
@@ -3442,8 +3425,8 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 			val |= DDI_TRAINING_OVERRIDE_ENABLE |
 				DDI_TRAINING_OVERRIDE_VALUE;
 
-		intel_de_write(dev_priv, reg, val);
-		intel_de_posting_read(dev_priv, reg);
+		intel_de_write(display, reg, val);
+		intel_de_posting_read(display, reg);
 
 		udelay(1);
 
@@ -3454,7 +3437,7 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 			val &= ~(DDI_TRAINING_OVERRIDE_ENABLE |
 				 DDI_TRAINING_OVERRIDE_VALUE);
 
-		intel_de_write(dev_priv, reg, val);
+		intel_de_write(display, reg, val);
 	}
 
 	intel_ddi_power_up_lanes(encoder, crtc_state);
@@ -3475,7 +3458,7 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 	if (dig_port->ddi_a_4_lanes)
 		buf_ctl |= DDI_A_4_LANES;
 
-	if (DISPLAY_VER(dev_priv) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		u32 port_buf = 0;
 
 		port_buf |= XELPDP_PORT_WIDTH(crtc_state->lane_count);
@@ -3483,15 +3466,15 @@ static void intel_ddi_enable_hdmi(struct intel_atomic_state *state,
 		if (dig_port->lane_reversal)
 			port_buf |= XELPDP_PORT_REVERSAL;
 
-		intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(dev_priv, port),
+		intel_de_rmw(display, XELPDP_PORT_BUF_CTL1(display, port),
 			     XELPDP_PORT_WIDTH_MASK | XELPDP_PORT_REVERSAL, port_buf);
 
 		buf_ctl |= DDI_PORT_WIDTH(crtc_state->lane_count);
 
-		if (DISPLAY_VER(dev_priv) >= 20)
+		if (DISPLAY_VER(display) >= 20)
 			buf_ctl |= XE2LPD_DDI_BUF_D2D_LINK_ENABLE;
-	} else if (IS_ALDERLAKE_P(dev_priv) && intel_encoder_is_tc(encoder)) {
-		drm_WARN_ON(&dev_priv->drm, !intel_tc_port_in_legacy_mode(dig_port));
+	} else if (display->platform.alderlake_p && intel_encoder_is_tc(encoder)) {
+		drm_WARN_ON(display->drm, !intel_tc_port_in_legacy_mode(dig_port));
 		buf_ctl |= DDI_BUF_CTL_TC_PHY_OWNERSHIP;
 	}
 
@@ -3522,6 +3505,8 @@ static void intel_ddi_enable(struct intel_atomic_state *state,
 
 	intel_ddi_enable_transcoder_func(encoder, crtc_state);
 
+	intel_vrr_transcoder_enable(crtc_state);
+
 	/* Enable/Disable DP2.0 SDP split config before transcoder */
 	intel_audio_sdp_split_update(crtc_state);
 
@@ -3567,9 +3552,10 @@ static void intel_ddi_disable_dp(struct intel_atomic_state *state,
 	struct intel_connector *connector =
 		to_intel_connector(old_conn_state->connector);
 
-	intel_dp->link_trained = false;
+	intel_dp->link.active = false;
 
 	intel_psr_disable(intel_dp, old_crtc_state);
+	intel_alpm_disable(intel_dp);
 	intel_edp_backlight_off(old_conn_state);
 	/* Disable the decompression in DP Sink */
 	intel_dp_sink_disable_decompression(state,
@@ -3584,12 +3570,12 @@ static void intel_ddi_disable_hdmi(struct intel_atomic_state *state,
 				   const struct intel_crtc_state *old_crtc_state,
 				   const struct drm_connector_state *old_conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct drm_connector *connector = old_conn_state->connector;
 
 	if (!intel_hdmi_handle_sink_scrambling(encoder, connector,
 					       false, false))
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[CONNECTOR:%d:%s] Failed to reset sink scrambling/TMDS bit clock ratio\n",
 			    connector->base.id, connector->name);
 }
@@ -3653,16 +3639,16 @@ void intel_ddi_update_active_dpll(struct intel_atomic_state *state,
 				  struct intel_encoder *encoder,
 				  struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_crtc *pipe_crtc;
 
 	/* FIXME: Add MTL pll_mgr */
-	if (DISPLAY_VER(i915) >= 14 || !intel_encoder_is_tc(encoder))
+	if (DISPLAY_VER(display) >= 14 || !intel_encoder_is_tc(encoder))
 		return;
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, pipe_crtc,
+	for_each_intel_crtc_in_pipe_mask(display->drm, pipe_crtc,
 					 intel_crtc_joined_pipe_mask(crtc_state))
 		intel_update_active_dpll(state, pipe_crtc, encoder);
 }
@@ -3678,7 +3664,7 @@ intel_ddi_pre_pll_enable(struct intel_atomic_state *state,
 			 const struct intel_crtc_state *crtc_state,
 			 const struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	bool is_tc_port = intel_encoder_is_tc(encoder);
 
@@ -3697,7 +3683,7 @@ intel_ddi_pre_pll_enable(struct intel_atomic_state *state,
 		 * Type-C ports.  Skip this step for TBT.
 		 */
 		intel_tc_port_set_fia_lane_count(dig_port, crtc_state->lane_count);
-	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	else if (display->platform.geminilake || display->platform.broxton)
 		bxt_dpio_phy_set_lane_optim_mask(encoder,
 						 crtc_state->lane_lat_optim_mask);
 }
@@ -3765,10 +3751,9 @@ static void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp,
 	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct intel_encoder *encoder = &dig_port->base;
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 dp_tp_ctl;
 
-	dp_tp_ctl = intel_de_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
+	dp_tp_ctl = intel_de_read(display, dp_tp_ctl_reg(encoder, crtc_state));
 
 	drm_WARN_ON(display->drm, dp_tp_ctl & DP_TP_CTL_ENABLE);
 
@@ -3781,10 +3766,10 @@ static void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp,
 		if (crtc_state->enhanced_framing)
 			dp_tp_ctl |= DP_TP_CTL_ENHANCED_FRAME_ENABLE;
 	}
-	intel_de_write(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), dp_tp_ctl);
-	intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
+	intel_de_write(display, dp_tp_ctl_reg(encoder, crtc_state), dp_tp_ctl);
+	intel_de_posting_read(display, dp_tp_ctl_reg(encoder, crtc_state));
 
-	if (IS_ALDERLAKE_P(dev_priv) &&
+	if (display->platform.alderlake_p &&
 	    (intel_tc_port_in_dp_alt_mode(dig_port) || intel_tc_port_in_legacy_mode(dig_port)))
 		adlp_tbt_to_dp_alt_switch_wa(encoder);
 
@@ -3796,11 +3781,11 @@ static void intel_ddi_set_link_train(struct intel_dp *intel_dp,
 				     const struct intel_crtc_state *crtc_state,
 				     u8 dp_train_pat)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 temp;
 
-	temp = intel_de_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
+	temp = intel_de_read(display, dp_tp_ctl_reg(encoder, crtc_state));
 
 	temp &= ~DP_TP_CTL_LINK_TRAIN_MASK;
 	switch (intel_dp_training_pattern_symbol(dp_train_pat)) {
@@ -3821,17 +3806,17 @@ static void intel_ddi_set_link_train(struct intel_dp *intel_dp,
 		break;
 	}
 
-	intel_de_write(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), temp);
+	intel_de_write(display, dp_tp_ctl_reg(encoder, crtc_state), temp);
 }
 
 static void intel_ddi_set_idle_link_train(struct intel_dp *intel_dp,
 					  const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum port port = encoder->port;
 
-	intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state),
+	intel_de_rmw(display, dp_tp_ctl_reg(encoder, crtc_state),
 		     DP_TP_CTL_LINK_TRAIN_MASK, DP_TP_CTL_LINK_TRAIN_IDLE);
 
 	/*
@@ -3841,28 +3826,26 @@ static void intel_ddi_set_idle_link_train(struct intel_dp *intel_dp,
 	 * In this case there is requirement to wait for a minimum number of
 	 * idle patterns to be sent.
 	 */
-	if (port == PORT_A && DISPLAY_VER(dev_priv) < 12)
+	if (port == PORT_A && DISPLAY_VER(display) < 12)
 		return;
 
-	if (intel_de_wait_for_set(dev_priv,
+	if (intel_de_wait_for_set(display,
 				  dp_tp_status_reg(encoder, crtc_state),
 				  DP_TP_STATUS_IDLE_DONE, 2))
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"Timed out waiting for DP idle patterns\n");
 }
 
-static bool intel_ddi_is_audio_enabled(struct drm_i915_private *dev_priv,
+static bool intel_ddi_is_audio_enabled(struct intel_display *display,
 				       enum transcoder cpu_transcoder)
 {
-	struct intel_display *display = &dev_priv->display;
-
 	if (cpu_transcoder == TRANSCODER_EDP)
 		return false;
 
 	if (!intel_display_power_is_enabled(display, POWER_DOMAIN_AUDIO_MMIO))
 		return false;
 
-	return intel_de_read(dev_priv, HSW_AUD_PIN_ELD_CP_VLD) &
+	return intel_de_read(display, HSW_AUD_PIN_ELD_CP_VLD) &
 		AUDIO_OUTPUT_ENABLE(cpu_transcoder);
 }
 
@@ -3892,34 +3875,34 @@ static int icl_ddi_min_voltage_level(const struct intel_crtc_state *crtc_state)
 
 void intel_ddi_compute_min_voltage_level(struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
-	if (DISPLAY_VER(dev_priv) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		crtc_state->min_voltage_level = icl_ddi_min_voltage_level(crtc_state);
-	else if (DISPLAY_VER(dev_priv) >= 12)
+	else if (DISPLAY_VER(display) >= 12)
 		crtc_state->min_voltage_level = tgl_ddi_min_voltage_level(crtc_state);
-	else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv))
+	else if (display->platform.jasperlake || display->platform.elkhartlake)
 		crtc_state->min_voltage_level = jsl_ddi_min_voltage_level(crtc_state);
-	else if (DISPLAY_VER(dev_priv) >= 11)
+	else if (DISPLAY_VER(display) >= 11)
 		crtc_state->min_voltage_level = icl_ddi_min_voltage_level(crtc_state);
 }
 
-static enum transcoder bdw_transcoder_master_readout(struct drm_i915_private *dev_priv,
+static enum transcoder bdw_transcoder_master_readout(struct intel_display *display,
 						     enum transcoder cpu_transcoder)
 {
 	u32 master_select;
 
-	if (DISPLAY_VER(dev_priv) >= 11) {
-		u32 ctl2 = intel_de_read(dev_priv,
-					 TRANS_DDI_FUNC_CTL2(dev_priv, cpu_transcoder));
+	if (DISPLAY_VER(display) >= 11) {
+		u32 ctl2 = intel_de_read(display,
+					 TRANS_DDI_FUNC_CTL2(display, cpu_transcoder));
 
 		if ((ctl2 & PORT_SYNC_MODE_ENABLE) == 0)
 			return INVALID_TRANSCODER;
 
 		master_select = REG_FIELD_GET(PORT_SYNC_MODE_MASTER_SELECT_MASK, ctl2);
 	} else {
-		u32 ctl = intel_de_read(dev_priv,
-					TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder));
+		u32 ctl = intel_de_read(display,
+					TRANS_DDI_FUNC_CTL(display, cpu_transcoder));
 
 		if ((ctl & TRANS_DDI_PORT_SYNC_ENABLE) == 0)
 			return INVALID_TRANSCODER;
@@ -3936,15 +3919,14 @@ static enum transcoder bdw_transcoder_master_readout(struct drm_i915_private *de
 static void bdw_get_trans_port_sync_config(struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	u32 transcoders = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
 	enum transcoder cpu_transcoder;
 
 	crtc_state->master_transcoder =
-		bdw_transcoder_master_readout(dev_priv, crtc_state->cpu_transcoder);
+		bdw_transcoder_master_readout(display, crtc_state->cpu_transcoder);
 
-	for_each_cpu_transcoder_masked(dev_priv, cpu_transcoder, transcoders) {
+	for_each_cpu_transcoder_masked(display, cpu_transcoder, transcoders) {
 		enum intel_display_power_domain power_domain;
 		intel_wakeref_t trans_wakeref;
 
@@ -3955,14 +3937,14 @@ static void bdw_get_trans_port_sync_config(struct intel_crtc_state *crtc_state)
 		if (!trans_wakeref)
 			continue;
 
-		if (bdw_transcoder_master_readout(dev_priv, cpu_transcoder) ==
+		if (bdw_transcoder_master_readout(display, cpu_transcoder) ==
 		    crtc_state->cpu_transcoder)
 			crtc_state->sync_mode_slaves_mask |= BIT(cpu_transcoder);
 
 		intel_display_power_put(display, power_domain, trans_wakeref);
 	}
 
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(display->drm,
 		    crtc_state->master_transcoder != INVALID_TRANSCODER &&
 		    crtc_state->sync_mode_slaves_mask);
 }
@@ -4085,11 +4067,10 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder,
 				    struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum transcoder cpu_transcoder = pipe_config->cpu_transcoder;
 	u32 ddi_func_ctl, ddi_mode, flags = 0;
 
-	ddi_func_ctl = intel_de_read(dev_priv, TRANS_DDI_FUNC_CTL(dev_priv, cpu_transcoder));
+	ddi_func_ctl = intel_de_read(display, TRANS_DDI_FUNC_CTL(display, cpu_transcoder));
 	if (ddi_func_ctl & TRANS_DDI_PHSYNC)
 		flags |= DRM_MODE_FLAG_PHSYNC;
 	else
@@ -4131,13 +4112,13 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder,
 	} else if (ddi_mode == TRANS_DDI_MODE_SELECT_DP_MST) {
 		intel_ddi_read_func_ctl_dp_mst(encoder, pipe_config, ddi_func_ctl);
 	} else if (ddi_mode == TRANS_DDI_MODE_SELECT_FDI_OR_128B132B && HAS_DP20(display)) {
-		struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
 		/*
 		 * If this is true, we know we're being called from mst stream
 		 * encoder's ->get_config().
 		 */
-		if (intel_dp_mst_encoder_active_links(dig_port))
+		if (intel_dp_mst_active_streams(intel_dp))
 			intel_ddi_read_func_ctl_dp_mst(encoder, pipe_config, ddi_func_ctl);
 		else
 			intel_ddi_read_func_ctl_dp_sst(encoder, pipe_config, ddi_func_ctl);
@@ -4152,11 +4133,11 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder,
 static void intel_ddi_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *pipe_config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum transcoder cpu_transcoder = pipe_config->cpu_transcoder;
 
 	/* XXX: DSI transcoder paranoia */
-	if (drm_WARN_ON(&dev_priv->drm, transcoder_is_dsi(cpu_transcoder)))
+	if (drm_WARN_ON(display->drm, transcoder_is_dsi(cpu_transcoder)))
 		return;
 
 	intel_ddi_read_func_ctl(encoder, pipe_config);
@@ -4164,14 +4145,14 @@ static void intel_ddi_get_config(struct intel_encoder *encoder,
 	intel_ddi_mso_get_config(encoder, pipe_config);
 
 	pipe_config->has_audio =
-		intel_ddi_is_audio_enabled(dev_priv, cpu_transcoder);
+		intel_ddi_is_audio_enabled(display, cpu_transcoder);
 
 	if (encoder->type == INTEL_OUTPUT_EDP)
 		intel_edp_fixup_vbt_bpp(encoder, pipe_config->pipe_bpp);
 
 	ddi_dotclock_get(pipe_config);
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		pipe_config->lane_lat_optim_mask =
 			bxt_dpio_phy_get_lane_lat_optim_mask(encoder);
 
@@ -4192,7 +4173,7 @@ static void intel_ddi_get_config(struct intel_encoder *encoder,
 			     HDMI_INFOFRAME_TYPE_DRM,
 			     &pipe_config->infoframes.drm);
 
-	if (DISPLAY_VER(dev_priv) >= 8)
+	if (DISPLAY_VER(display) >= 8)
 		bdw_get_trans_port_sync_config(pipe_config);
 
 	intel_psr_get_config(encoder, pipe_config);
@@ -4285,10 +4266,10 @@ static enum icl_port_dpll_id
 icl_ddi_tc_port_pll_type(struct intel_encoder *encoder,
 			 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	const struct intel_shared_dpll *pll = crtc_state->shared_dpll;
 
-	if (drm_WARN_ON(&i915->drm, !pll))
+	if (drm_WARN_ON(display->drm, !pll))
 		return ICL_PORT_DPLL_DEFAULT;
 
 	if (icl_ddi_tc_pll_is_tbt(pll))
@@ -4382,11 +4363,11 @@ static void intel_ddi_sync_state(struct intel_encoder *encoder,
 static bool intel_ddi_initial_fastset_check(struct intel_encoder *encoder,
 					    struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	bool fastset = true;
 
 	if (intel_encoder_is_tc(encoder)) {
-		drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] Forcing full modeset to compute TC port DPLLs\n",
+		drm_dbg_kms(display->drm, "[ENCODER:%d:%s] Forcing full modeset to compute TC port DPLLs\n",
 			    encoder->base.base.id, encoder->base.name);
 		crtc_state->uapi.mode_changed = true;
 		fastset = false;
@@ -4421,12 +4402,12 @@ static int intel_ddi_compute_config(struct intel_encoder *encoder,
 				    struct intel_crtc_state *pipe_config,
 				    struct drm_connector_state *conn_state)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum port port = encoder->port;
 	int ret;
 
-	if (HAS_TRANSCODER(dev_priv, TRANSCODER_EDP) && port == PORT_A)
+	if (HAS_TRANSCODER(display, TRANSCODER_EDP) && port == PORT_A)
 		pipe_config->cpu_transcoder = TRANSCODER_EDP;
 
 	if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_HDMI)) {
@@ -4441,13 +4422,13 @@ static int intel_ddi_compute_config(struct intel_encoder *encoder,
 	if (ret)
 		return ret;
 
-	if (IS_HASWELL(dev_priv) && crtc->pipe == PIPE_A &&
+	if (display->platform.haswell && crtc->pipe == PIPE_A &&
 	    pipe_config->cpu_transcoder == TRANSCODER_EDP)
 		pipe_config->pch_pfit.force_thru =
 			pipe_config->pch_pfit.enabled ||
 			pipe_config->crc_enabled;
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		pipe_config->lane_lat_optim_mask =
 			bxt_dpio_phy_calc_lane_lat_optim_mask(pipe_config->lane_count);
 
@@ -4498,9 +4479,9 @@ static u8
 intel_ddi_port_sync_transcoders(const struct intel_crtc_state *ref_crtc_state,
 				int tile_group_id)
 {
+	struct intel_display *display = to_intel_display(ref_crtc_state);
 	struct drm_connector *connector;
 	const struct drm_connector_state *conn_state;
-	struct drm_i915_private *dev_priv = to_i915(ref_crtc_state->uapi.crtc->dev);
 	struct intel_atomic_state *state =
 		to_intel_atomic_state(ref_crtc_state->uapi.state);
 	u8 transcoders = 0;
@@ -4510,7 +4491,7 @@ intel_ddi_port_sync_transcoders(const struct intel_crtc_state *ref_crtc_state,
 	 * We don't enable port sync on BDW due to missing w/as and
 	 * due to not having adjusted the modeset sequence appropriately.
 	 */
-	if (DISPLAY_VER(dev_priv) < 9)
+	if (DISPLAY_VER(display) < 9)
 		return 0;
 
 	if (!intel_crtc_has_type(ref_crtc_state, INTEL_OUTPUT_DP))
@@ -4542,11 +4523,11 @@ static int intel_ddi_compute_config_late(struct intel_encoder *encoder,
 					 struct intel_crtc_state *crtc_state,
 					 struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct drm_connector *connector = conn_state->connector;
 	u8 port_sync_transcoders = 0;
 
-	drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] [CRTC:%d:%s]\n",
+	drm_dbg_kms(display->drm, "[ENCODER:%d:%s] [CRTC:%d:%s]\n",
 		    encoder->base.base.id, encoder->base.name,
 		    crtc_state->uapi.crtc->base.id, crtc_state->uapi.crtc->name);
 
@@ -4618,7 +4599,7 @@ static const struct drm_encoder_funcs intel_ddi_funcs = {
 
 static int intel_ddi_init_dp_connector(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_connector *connector;
 	enum port port = dig_port->base.port;
 
@@ -4627,7 +4608,7 @@ static int intel_ddi_init_dp_connector(struct intel_digital_port *dig_port)
 		return -ENOMEM;
 
 	dig_port->dp.output_reg = DDI_BUF_CTL(port);
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		dig_port->dp.prepare_link_retrain = mtl_ddi_prepare_link_retrain;
 	else
 		dig_port->dp.prepare_link_retrain = intel_ddi_prepare_link_retrain;
@@ -4643,15 +4624,14 @@ static int intel_ddi_init_dp_connector(struct intel_digital_port *dig_port)
 	}
 
 	if (dig_port->base.type == INTEL_OUTPUT_EDP) {
-		struct drm_device *dev = dig_port->base.base.dev;
 		struct drm_privacy_screen *privacy_screen;
 
-		privacy_screen = drm_privacy_screen_get(dev->dev, NULL);
+		privacy_screen = drm_privacy_screen_get(display->drm->dev, NULL);
 		if (!IS_ERR(privacy_screen)) {
 			drm_connector_attach_privacy_screen_provider(&connector->base,
 								     privacy_screen);
 		} else if (PTR_ERR(privacy_screen) != -ENODEV) {
-			drm_warn(dev, "Error getting privacy-screen\n");
+			drm_warn(display->drm, "Error getting privacy-screen\n");
 		}
 	}
 
@@ -4662,7 +4642,6 @@ static int intel_hdmi_reset_link(struct intel_encoder *encoder,
 				 struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_hdmi *hdmi = enc_to_intel_hdmi(encoder);
 	struct intel_connector *connector = hdmi->attached_connector;
 	struct i2c_adapter *ddc = connector->base.ddc;
@@ -4675,7 +4654,7 @@ static int intel_hdmi_reset_link(struct intel_encoder *encoder,
 	if (connector->base.status != connector_status_connected)
 		return 0;
 
-	ret = drm_modeset_lock(&dev_priv->drm.mode_config.connection_mutex,
+	ret = drm_modeset_lock(&display->drm->mode_config.connection_mutex,
 			       ctx);
 	if (ret)
 		return ret;
@@ -4692,7 +4671,7 @@ static int intel_hdmi_reset_link(struct intel_encoder *encoder,
 
 	crtc_state = to_intel_crtc_state(crtc->base.state);
 
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(display->drm,
 		    !intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI));
 
 	if (!crtc_state->hw.active)
@@ -4708,7 +4687,7 @@ static int intel_hdmi_reset_link(struct intel_encoder *encoder,
 
 	ret = drm_scdc_readb(ddc, SCDC_TMDS_CONFIG, &config);
 	if (ret < 0) {
-		drm_err(&dev_priv->drm, "[CONNECTOR:%d:%s] Failed to read TMDS config: %d\n",
+		drm_err(display->drm, "[CONNECTOR:%d:%s] Failed to read TMDS config: %d\n",
 			connector->base.base.id, connector->base.name, ret);
 		return 0;
 	}
@@ -4733,11 +4712,11 @@ static int intel_hdmi_reset_link(struct intel_encoder *encoder,
 
 static void intel_ddi_link_check(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 
 	/* TODO: Move checking the HDMI link state here as well. */
-	drm_WARN_ON(&i915->drm, !dig_port->dp.attached_connector);
+	drm_WARN_ON(display->drm, !dig_port->dp.attached_connector);
 
 	intel_dp_link_check(encoder);
 }
@@ -4800,26 +4779,26 @@ intel_ddi_hotplug(struct intel_encoder *encoder,
 
 static bool lpt_digital_port_connected(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	u32 bit = dev_priv->display.hotplug.pch_hpd[encoder->hpd_pin];
+	struct intel_display *display = to_intel_display(encoder);
+	u32 bit = display->hotplug.pch_hpd[encoder->hpd_pin];
 
-	return intel_de_read(dev_priv, SDEISR) & bit;
+	return intel_de_read(display, SDEISR) & bit;
 }
 
 static bool hsw_digital_port_connected(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	u32 bit = dev_priv->display.hotplug.hpd[encoder->hpd_pin];
+	struct intel_display *display = to_intel_display(encoder);
+	u32 bit = display->hotplug.hpd[encoder->hpd_pin];
 
-	return intel_de_read(dev_priv, DEISR) & bit;
+	return intel_de_read(display, DEISR) & bit;
 }
 
 static bool bdw_digital_port_connected(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	u32 bit = dev_priv->display.hotplug.hpd[encoder->hpd_pin];
+	struct intel_display *display = to_intel_display(encoder);
+	u32 bit = display->hotplug.hpd[encoder->hpd_pin];
 
-	return intel_de_read(dev_priv, GEN8_DE_PORT_ISR) & bit;
+	return intel_de_read(display, GEN8_DE_PORT_ISR) & bit;
 }
 
 static int intel_ddi_init_hdmi_connector(struct intel_digital_port *dig_port)
@@ -4848,7 +4827,7 @@ static int intel_ddi_init_hdmi_connector(struct intel_digital_port *dig_port)
 
 static bool intel_ddi_a_force_4_lanes(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 
 	if (dig_port->base.port != PORT_A)
 		return false;
@@ -4859,7 +4838,7 @@ static bool intel_ddi_a_force_4_lanes(struct intel_digital_port *dig_port)
 	/* Broxton/Geminilake: Bspec says that DDI_A_4_LANES is the only
 	 *                     supported configuration
 	 */
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		return true;
 
 	return false;
@@ -4868,15 +4847,15 @@ static bool intel_ddi_a_force_4_lanes(struct intel_digital_port *dig_port)
 static int
 intel_ddi_max_lanes(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	enum port port = dig_port->base.port;
 	int max_lanes = 4;
 
-	if (DISPLAY_VER(dev_priv) >= 11)
+	if (DISPLAY_VER(display) >= 11)
 		return max_lanes;
 
 	if (port == PORT_A || port == PORT_E) {
-		if (intel_de_read(dev_priv, DDI_BUF_CTL(PORT_A)) & DDI_A_4_LANES)
+		if (intel_de_read(display, DDI_BUF_CTL(PORT_A)) & DDI_A_4_LANES)
 			max_lanes = port == PORT_A ? 4 : 0;
 		else
 			/* Both A and E share 2 lanes */
@@ -4889,7 +4868,7 @@ intel_ddi_max_lanes(struct intel_digital_port *dig_port)
 	 * so we use the proper lane count for our calculations.
 	 */
 	if (intel_ddi_a_force_4_lanes(dig_port)) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Forcing DDI_A_4_LANES for port A\n");
 		dig_port->ddi_a_4_lanes = true;
 		max_lanes = 4;
@@ -4898,8 +4877,7 @@ intel_ddi_max_lanes(struct intel_digital_port *dig_port)
 	return max_lanes;
 }
 
-static enum hpd_pin xelpd_hpd_pin(struct drm_i915_private *dev_priv,
-				  enum port port)
+static enum hpd_pin xelpd_hpd_pin(struct intel_display *display, enum port port)
 {
 	if (port >= PORT_D_XELPD)
 		return HPD_PORT_D + port - PORT_D_XELPD;
@@ -4909,8 +4887,7 @@ static enum hpd_pin xelpd_hpd_pin(struct drm_i915_private *dev_priv,
 		return HPD_PORT_A + port - PORT_A;
 }
 
-static enum hpd_pin dg1_hpd_pin(struct drm_i915_private *dev_priv,
-				enum port port)
+static enum hpd_pin dg1_hpd_pin(struct intel_display *display, enum port port)
 {
 	if (port >= PORT_TC1)
 		return HPD_PORT_C + port - PORT_TC1;
@@ -4918,8 +4895,7 @@ static enum hpd_pin dg1_hpd_pin(struct drm_i915_private *dev_priv,
 		return HPD_PORT_A + port - PORT_A;
 }
 
-static enum hpd_pin tgl_hpd_pin(struct drm_i915_private *dev_priv,
-				enum port port)
+static enum hpd_pin tgl_hpd_pin(struct intel_display *display, enum port port)
 {
 	if (port >= PORT_TC1)
 		return HPD_PORT_TC1 + port - PORT_TC1;
@@ -4927,11 +4903,10 @@ static enum hpd_pin tgl_hpd_pin(struct drm_i915_private *dev_priv,
 		return HPD_PORT_A + port - PORT_A;
 }
 
-static enum hpd_pin rkl_hpd_pin(struct drm_i915_private *dev_priv,
-				enum port port)
+static enum hpd_pin rkl_hpd_pin(struct intel_display *display, enum port port)
 {
-	if (HAS_PCH_TGP(dev_priv))
-		return tgl_hpd_pin(dev_priv, port);
+	if (HAS_PCH_TGP(display))
+		return tgl_hpd_pin(display, port);
 
 	if (port >= PORT_TC1)
 		return HPD_PORT_C + port - PORT_TC1;
@@ -4939,8 +4914,7 @@ static enum hpd_pin rkl_hpd_pin(struct drm_i915_private *dev_priv,
 		return HPD_PORT_A + port - PORT_A;
 }
 
-static enum hpd_pin icl_hpd_pin(struct drm_i915_private *dev_priv,
-				enum port port)
+static enum hpd_pin icl_hpd_pin(struct intel_display *display, enum port port)
 {
 	if (port >= PORT_C)
 		return HPD_PORT_TC1 + port - PORT_C;
@@ -4948,31 +4922,30 @@ static enum hpd_pin icl_hpd_pin(struct drm_i915_private *dev_priv,
 		return HPD_PORT_A + port - PORT_A;
 }
 
-static enum hpd_pin ehl_hpd_pin(struct drm_i915_private *dev_priv,
-				enum port port)
+static enum hpd_pin ehl_hpd_pin(struct intel_display *display, enum port port)
 {
 	if (port == PORT_D)
 		return HPD_PORT_A;
 
-	if (HAS_PCH_TGP(dev_priv))
-		return icl_hpd_pin(dev_priv, port);
+	if (HAS_PCH_TGP(display))
+		return icl_hpd_pin(display, port);
 
 	return HPD_PORT_A + port - PORT_A;
 }
 
-static enum hpd_pin skl_hpd_pin(struct drm_i915_private *dev_priv, enum port port)
+static enum hpd_pin skl_hpd_pin(struct intel_display *display, enum port port)
 {
-	if (HAS_PCH_TGP(dev_priv))
-		return icl_hpd_pin(dev_priv, port);
+	if (HAS_PCH_TGP(display))
+		return icl_hpd_pin(display, port);
 
 	return HPD_PORT_A + port - PORT_A;
 }
 
-static bool intel_ddi_is_tc(struct drm_i915_private *i915, enum port port)
+static bool intel_ddi_is_tc(struct intel_display *display, enum port port)
 {
-	if (DISPLAY_VER(i915) >= 12)
+	if (DISPLAY_VER(display) >= 12)
 		return port >= PORT_TC1;
-	else if (DISPLAY_VER(i915) >= 11)
+	else if (DISPLAY_VER(display) >= 11)
 		return port >= PORT_C;
 	else
 		return false;
@@ -5015,21 +4988,21 @@ static void intel_ddi_tc_encoder_shutdown_complete(struct intel_encoder *encoder
 #define port_tc_name(port) ((port) - PORT_TC1 + '1')
 #define tc_port_name(tc_port) ((tc_port) - TC_PORT_1 + '1')
 
-static bool port_strap_detected(struct drm_i915_private *i915, enum port port)
+static bool port_strap_detected(struct intel_display *display, enum port port)
 {
 	/* straps not used on skl+ */
-	if (DISPLAY_VER(i915) >= 9)
+	if (DISPLAY_VER(display) >= 9)
 		return true;
 
 	switch (port) {
 	case PORT_A:
-		return intel_de_read(i915, DDI_BUF_CTL(PORT_A)) & DDI_INIT_DISPLAY_DETECTED;
+		return intel_de_read(display, DDI_BUF_CTL(PORT_A)) & DDI_INIT_DISPLAY_DETECTED;
 	case PORT_B:
-		return intel_de_read(i915, SFUSE_STRAP) & SFUSE_STRAP_DDIB_DETECTED;
+		return intel_de_read(display, SFUSE_STRAP) & SFUSE_STRAP_DDIB_DETECTED;
 	case PORT_C:
-		return intel_de_read(i915, SFUSE_STRAP) & SFUSE_STRAP_DDIC_DETECTED;
+		return intel_de_read(display, SFUSE_STRAP) & SFUSE_STRAP_DDIC_DETECTED;
 	case PORT_D:
-		return intel_de_read(i915, SFUSE_STRAP) & SFUSE_STRAP_DDID_DETECTED;
+		return intel_de_read(display, SFUSE_STRAP) & SFUSE_STRAP_DDID_DETECTED;
 	case PORT_E:
 		return true; /* no strap for DDI-E */
 	default:
@@ -5043,18 +5016,18 @@ static bool need_aux_ch(struct intel_encoder *encoder, bool init_dp)
 	return init_dp || intel_encoder_is_tc(encoder);
 }
 
-static bool assert_has_icl_dsi(struct drm_i915_private *i915)
+static bool assert_has_icl_dsi(struct intel_display *display)
 {
-	return !drm_WARN(&i915->drm, !IS_ALDERLAKE_P(i915) &&
-			 !IS_TIGERLAKE(i915) && DISPLAY_VER(i915) != 11,
+	return !drm_WARN(display->drm, !display->platform.alderlake_p &&
+			 !display->platform.tigerlake && DISPLAY_VER(display) != 11,
 			 "Platform does not support DSI\n");
 }
 
-static bool port_in_use(struct drm_i915_private *i915, enum port port)
+static bool port_in_use(struct intel_display *display, enum port port)
 {
 	struct intel_encoder *encoder;
 
-	for_each_intel_encoder(&i915->drm, encoder) {
+	for_each_intel_encoder(display->drm, encoder) {
 		/* FIXME what about second port for dual link DSI? */
 		if (encoder->port == port)
 			return true;
@@ -5066,7 +5039,6 @@ static bool port_in_use(struct drm_i915_private *i915, enum port port)
 void intel_ddi_init(struct intel_display *display,
 		    const struct intel_bios_encoder_data *devdata)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_digital_port *dig_port;
 	struct intel_encoder *encoder;
 	bool init_hdmi, init_dp;
@@ -5078,8 +5050,8 @@ void intel_ddi_init(struct intel_display *display,
 	if (port == PORT_NONE)
 		return;
 
-	if (!port_strap_detected(dev_priv, port)) {
-		drm_dbg_kms(&dev_priv->drm,
+	if (!port_strap_detected(display, port)) {
+		drm_dbg_kms(display->drm,
 			    "Port %c strap not detected\n", port_name(port));
 		return;
 	}
@@ -5087,15 +5059,15 @@ void intel_ddi_init(struct intel_display *display,
 	if (!assert_port_valid(display, port))
 		return;
 
-	if (port_in_use(dev_priv, port)) {
-		drm_dbg_kms(&dev_priv->drm,
+	if (port_in_use(display, port)) {
+		drm_dbg_kms(display->drm,
 			    "Port %c already claimed\n", port_name(port));
 		return;
 	}
 
 	if (intel_bios_encoder_supports_dsi(devdata)) {
 		/* BXT/GLK handled elsewhere, for now at least */
-		if (!assert_has_icl_dsi(dev_priv))
+		if (!assert_has_icl_dsi(display))
 			return;
 
 		icl_dsi_init(display, devdata);
@@ -5111,7 +5083,7 @@ void intel_ddi_init(struct intel_display *display,
 	 * outputs.
 	 */
 	if (intel_hti_uses_phy(display, phy)) {
-		drm_dbg_kms(&dev_priv->drm, "PORT %c / PHY %c reserved by HTI\n",
+		drm_dbg_kms(display->drm, "PORT %c / PHY %c reserved by HTI\n",
 			    port_name(port), phy_name(phy));
 		return;
 	}
@@ -5128,20 +5100,20 @@ void intel_ddi_init(struct intel_display *display,
 		 */
 		init_dp = true;
 		init_hdmi = false;
-		drm_dbg_kms(&dev_priv->drm, "VBT says port %c has lspcon\n",
+		drm_dbg_kms(display->drm, "VBT says port %c has lspcon\n",
 			    port_name(port));
 	}
 
 	if (!init_dp && !init_hdmi) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "VBT says port %c is not DVI/HDMI/DP compatible, respect it\n",
 			    port_name(port));
 		return;
 	}
 
 	if (intel_phy_is_snps(display, phy) &&
-	    dev_priv->display.snps.phy_failed_calibration & BIT(phy)) {
-		drm_dbg_kms(&dev_priv->drm,
+	    display->snps.phy_failed_calibration & BIT(phy)) {
+		drm_dbg_kms(display->drm,
 			    "SNPS PHY %c failed to calibrate, proceeding anyway\n",
 			    phy_name(phy));
 	}
@@ -5155,26 +5127,26 @@ void intel_ddi_init(struct intel_display *display,
 	encoder = &dig_port->base;
 	encoder->devdata = devdata;
 
-	if (DISPLAY_VER(dev_priv) >= 13 && port >= PORT_D_XELPD) {
-		drm_encoder_init(&dev_priv->drm, &encoder->base, &intel_ddi_funcs,
+	if (DISPLAY_VER(display) >= 13 && port >= PORT_D_XELPD) {
+		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
 				 DRM_MODE_ENCODER_TMDS,
 				 "DDI %c/PHY %c",
 				 port_name(port - PORT_D_XELPD + PORT_D),
 				 phy_name(phy));
-	} else if (DISPLAY_VER(dev_priv) >= 12) {
+	} else if (DISPLAY_VER(display) >= 12) {
 		enum tc_port tc_port = intel_port_to_tc(display, port);
 
-		drm_encoder_init(&dev_priv->drm, &encoder->base, &intel_ddi_funcs,
+		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
 				 DRM_MODE_ENCODER_TMDS,
 				 "DDI %s%c/PHY %s%c",
 				 port >= PORT_TC1 ? "TC" : "",
 				 port >= PORT_TC1 ? port_tc_name(port) : port_name(port),
 				 tc_port != TC_PORT_NONE ? "TC" : "",
 				 tc_port != TC_PORT_NONE ? tc_port_name(tc_port) : phy_name(phy));
-	} else if (DISPLAY_VER(dev_priv) >= 11) {
+	} else if (DISPLAY_VER(display) >= 11) {
 		enum tc_port tc_port = intel_port_to_tc(display, port);
 
-		drm_encoder_init(&dev_priv->drm, &encoder->base, &intel_ddi_funcs,
+		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
 				 DRM_MODE_ENCODER_TMDS,
 				 "DDI %c%s/PHY %s%c",
 				 port_name(port),
@@ -5182,7 +5154,7 @@ void intel_ddi_init(struct intel_display *display,
 				 tc_port != TC_PORT_NONE ? "TC" : "",
 				 tc_port != TC_PORT_NONE ? tc_port_name(tc_port) : phy_name(phy));
 	} else {
-		drm_encoder_init(&dev_priv->drm, &encoder->base, &intel_ddi_funcs,
+		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
 				 DRM_MODE_ENCODER_TMDS,
 				 "DDI %c/PHY %c", port_name(port), phy_name(phy));
 	}
@@ -5218,32 +5190,32 @@ void intel_ddi_init(struct intel_display *display,
 	encoder->cloneable = 0;
 	encoder->pipe_mask = ~0;
 
-	if (DISPLAY_VER(dev_priv) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		encoder->enable_clock = intel_mtl_pll_enable;
 		encoder->disable_clock = intel_mtl_pll_disable;
 		encoder->port_pll_type = intel_mtl_port_pll_type;
 		encoder->get_config = mtl_ddi_get_config;
-	} else if (IS_DG2(dev_priv)) {
+	} else if (display->platform.dg2) {
 		encoder->enable_clock = intel_mpllb_enable;
 		encoder->disable_clock = intel_mpllb_disable;
 		encoder->get_config = dg2_ddi_get_config;
-	} else if (IS_ALDERLAKE_S(dev_priv)) {
+	} else if (display->platform.alderlake_s) {
 		encoder->enable_clock = adls_ddi_enable_clock;
 		encoder->disable_clock = adls_ddi_disable_clock;
 		encoder->is_clock_enabled = adls_ddi_is_clock_enabled;
 		encoder->get_config = adls_ddi_get_config;
-	} else if (IS_ROCKETLAKE(dev_priv)) {
+	} else if (display->platform.rocketlake) {
 		encoder->enable_clock = rkl_ddi_enable_clock;
 		encoder->disable_clock = rkl_ddi_disable_clock;
 		encoder->is_clock_enabled = rkl_ddi_is_clock_enabled;
 		encoder->get_config = rkl_ddi_get_config;
-	} else if (IS_DG1(dev_priv)) {
+	} else if (display->platform.dg1) {
 		encoder->enable_clock = dg1_ddi_enable_clock;
 		encoder->disable_clock = dg1_ddi_disable_clock;
 		encoder->is_clock_enabled = dg1_ddi_is_clock_enabled;
 		encoder->get_config = dg1_ddi_get_config;
-	} else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) {
-		if (intel_ddi_is_tc(dev_priv, port)) {
+	} else if (display->platform.jasperlake || display->platform.elkhartlake) {
+		if (intel_ddi_is_tc(display, port)) {
 			encoder->enable_clock = jsl_ddi_tc_enable_clock;
 			encoder->disable_clock = jsl_ddi_tc_disable_clock;
 			encoder->is_clock_enabled = jsl_ddi_tc_is_clock_enabled;
@@ -5255,8 +5227,8 @@ void intel_ddi_init(struct intel_display *display,
 			encoder->is_clock_enabled = icl_ddi_combo_is_clock_enabled;
 			encoder->get_config = icl_ddi_combo_get_config;
 		}
-	} else if (DISPLAY_VER(dev_priv) >= 11) {
-		if (intel_ddi_is_tc(dev_priv, port)) {
+	} else if (DISPLAY_VER(display) >= 11) {
+		if (intel_ddi_is_tc(display, port)) {
 			encoder->enable_clock = icl_ddi_tc_enable_clock;
 			encoder->disable_clock = icl_ddi_tc_disable_clock;
 			encoder->is_clock_enabled = icl_ddi_tc_is_clock_enabled;
@@ -5268,36 +5240,36 @@ void intel_ddi_init(struct intel_display *display,
 			encoder->is_clock_enabled = icl_ddi_combo_is_clock_enabled;
 			encoder->get_config = icl_ddi_combo_get_config;
 		}
-	} else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	} else if (display->platform.geminilake || display->platform.broxton) {
 		/* BXT/GLK have fixed PLL->port mapping */
 		encoder->get_config = bxt_ddi_get_config;
-	} else if (DISPLAY_VER(dev_priv) == 9) {
+	} else if (DISPLAY_VER(display) == 9) {
 		encoder->enable_clock = skl_ddi_enable_clock;
 		encoder->disable_clock = skl_ddi_disable_clock;
 		encoder->is_clock_enabled = skl_ddi_is_clock_enabled;
 		encoder->get_config = skl_ddi_get_config;
-	} else if (IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv)) {
+	} else if (display->platform.broadwell || display->platform.haswell) {
 		encoder->enable_clock = hsw_ddi_enable_clock;
 		encoder->disable_clock = hsw_ddi_disable_clock;
 		encoder->is_clock_enabled = hsw_ddi_is_clock_enabled;
 		encoder->get_config = hsw_ddi_get_config;
 	}
 
-	if (DISPLAY_VER(dev_priv) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		encoder->set_signal_levels = intel_cx0_phy_set_signal_levels;
-	} else if (IS_DG2(dev_priv)) {
+	} else if (display->platform.dg2) {
 		encoder->set_signal_levels = intel_snps_phy_set_signal_levels;
-	} else if (DISPLAY_VER(dev_priv) >= 12) {
+	} else if (DISPLAY_VER(display) >= 12) {
 		if (intel_encoder_is_combo(encoder))
 			encoder->set_signal_levels = icl_combo_phy_set_signal_levels;
 		else
 			encoder->set_signal_levels = tgl_dkl_phy_set_signal_levels;
-	} else if (DISPLAY_VER(dev_priv) >= 11) {
+	} else if (DISPLAY_VER(display) >= 11) {
 		if (intel_encoder_is_combo(encoder))
 			encoder->set_signal_levels = icl_combo_phy_set_signal_levels;
 		else
 			encoder->set_signal_levels = icl_mg_phy_set_signal_levels;
-	} else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	} else if (display->platform.geminilake || display->platform.broxton) {
 		encoder->set_signal_levels = bxt_dpio_phy_set_signal_levels;
 	} else {
 		encoder->set_signal_levels = hsw_set_signal_levels;
@@ -5305,29 +5277,29 @@ void intel_ddi_init(struct intel_display *display,
 
 	intel_ddi_buf_trans_init(encoder);
 
-	if (DISPLAY_VER(dev_priv) >= 13)
-		encoder->hpd_pin = xelpd_hpd_pin(dev_priv, port);
-	else if (IS_DG1(dev_priv))
-		encoder->hpd_pin = dg1_hpd_pin(dev_priv, port);
-	else if (IS_ROCKETLAKE(dev_priv))
-		encoder->hpd_pin = rkl_hpd_pin(dev_priv, port);
-	else if (DISPLAY_VER(dev_priv) >= 12)
-		encoder->hpd_pin = tgl_hpd_pin(dev_priv, port);
-	else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv))
-		encoder->hpd_pin = ehl_hpd_pin(dev_priv, port);
-	else if (DISPLAY_VER(dev_priv) == 11)
-		encoder->hpd_pin = icl_hpd_pin(dev_priv, port);
-	else if (DISPLAY_VER(dev_priv) == 9 && !IS_BROXTON(dev_priv))
-		encoder->hpd_pin = skl_hpd_pin(dev_priv, port);
+	if (DISPLAY_VER(display) >= 13)
+		encoder->hpd_pin = xelpd_hpd_pin(display, port);
+	else if (display->platform.dg1)
+		encoder->hpd_pin = dg1_hpd_pin(display, port);
+	else if (display->platform.rocketlake)
+		encoder->hpd_pin = rkl_hpd_pin(display, port);
+	else if (DISPLAY_VER(display) >= 12)
+		encoder->hpd_pin = tgl_hpd_pin(display, port);
+	else if (display->platform.jasperlake || display->platform.elkhartlake)
+		encoder->hpd_pin = ehl_hpd_pin(display, port);
+	else if (DISPLAY_VER(display) == 11)
+		encoder->hpd_pin = icl_hpd_pin(display, port);
+	else if (DISPLAY_VER(display) == 9 && !display->platform.broxton)
+		encoder->hpd_pin = skl_hpd_pin(display, port);
 	else
 		encoder->hpd_pin = intel_hpd_pin_default(port);
 
-	ddi_buf_ctl = intel_de_read(dev_priv, DDI_BUF_CTL(port));
+	ddi_buf_ctl = intel_de_read(display, DDI_BUF_CTL(port));
 
 	dig_port->lane_reversal = intel_bios_encoder_lane_reversal(devdata) ||
 		ddi_buf_ctl & DDI_BUF_PORT_REVERSAL;
 
-	dig_port->ddi_a_4_lanes = DISPLAY_VER(dev_priv) < 11 && ddi_buf_ctl & DDI_A_4_LANES;
+	dig_port->ddi_a_4_lanes = DISPLAY_VER(display) < 11 && ddi_buf_ctl & DDI_A_4_LANES;
 
 	dig_port->dp.output_reg = INVALID_MMIO_REG;
 	dig_port->max_lanes = intel_ddi_max_lanes(dig_port);
@@ -5346,7 +5318,7 @@ void intel_ddi_init(struct intel_display *display,
 		if (!is_legacy && init_hdmi) {
 			is_legacy = !init_dp;
 
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "VBT says port %c is non-legacy TC and has HDMI (with DP: %s), assume it's %s\n",
 				    port_name(port),
 				    str_yes_no(init_dp),
@@ -5363,24 +5335,24 @@ void intel_ddi_init(struct intel_display *display,
 			goto err;
 	}
 
-	drm_WARN_ON(&dev_priv->drm, port > PORT_I);
+	drm_WARN_ON(display->drm, port > PORT_I);
 	dig_port->ddi_io_power_domain = intel_display_power_ddi_io_domain(display, port);
 
-	if (DISPLAY_VER(dev_priv) >= 11) {
+	if (DISPLAY_VER(display) >= 11) {
 		if (intel_encoder_is_tc(encoder))
 			dig_port->connected = intel_tc_port_connected;
 		else
 			dig_port->connected = lpt_digital_port_connected;
-	} else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	} else if (display->platform.geminilake || display->platform.broxton) {
 		dig_port->connected = bdw_digital_port_connected;
-	} else if (DISPLAY_VER(dev_priv) == 9) {
+	} else if (DISPLAY_VER(display) == 9) {
 		dig_port->connected = lpt_digital_port_connected;
-	} else if (IS_BROADWELL(dev_priv)) {
+	} else if (display->platform.broadwell) {
 		if (port == PORT_A)
 			dig_port->connected = bdw_digital_port_connected;
 		else
 			dig_port->connected = lpt_digital_port_connected;
-	} else if (IS_HASWELL(dev_priv)) {
+	} else if (display->platform.haswell) {
 		if (port == PORT_A)
 			dig_port->connected = hsw_digital_port_connected;
 		else
@@ -5396,7 +5368,7 @@ void intel_ddi_init(struct intel_display *display,
 		dig_port->hpd_pulse = intel_dp_hpd_pulse;
 
 		if (dig_port->dp.mso_link_count)
-			encoder->pipe_mask = intel_ddi_splitter_pipe_mask(dev_priv);
+			encoder->pipe_mask = intel_ddi_splitter_pipe_mask(display);
 	}
 
 	/*
diff --git a/drivers/gpu/drm/i915/display/intel_de.h b/drivers/gpu/drm/i915/display/intel_de.h
index b7399e9d11cc..54ce3e4f8fd9 100644
--- a/drivers/gpu/drm/i915/display/intel_de.h
+++ b/drivers/gpu/drm/i915/display/intel_de.h
@@ -6,7 +6,6 @@
 #ifndef __INTEL_DE_H__
 #define __INTEL_DE_H__
 
-#include "intel_display_conversion.h"
 #include "intel_display_core.h"
 #include "intel_dmc_wl.h"
 #include "intel_dsb.h"
@@ -19,7 +18,7 @@ static inline struct intel_uncore *__to_uncore(struct intel_display *display)
 }
 
 static inline u32
-__intel_de_read(struct intel_display *display, i915_reg_t reg)
+intel_de_read(struct intel_display *display, i915_reg_t reg)
 {
 	u32 val;
 
@@ -31,7 +30,6 @@ __intel_de_read(struct intel_display *display, i915_reg_t reg)
 
 	return val;
 }
-#define intel_de_read(p,...) __intel_de_read(__to_intel_display(p), __VA_ARGS__)
 
 static inline u8
 intel_de_read8(struct intel_display *display, i915_reg_t reg)
@@ -66,7 +64,7 @@ intel_de_read64_2x32(struct intel_display *display,
 }
 
 static inline void
-__intel_de_posting_read(struct intel_display *display, i915_reg_t reg)
+intel_de_posting_read(struct intel_display *display, i915_reg_t reg)
 {
 	intel_dmc_wl_get(display, reg);
 
@@ -74,10 +72,9 @@ __intel_de_posting_read(struct intel_display *display, i915_reg_t reg)
 
 	intel_dmc_wl_put(display, reg);
 }
-#define intel_de_posting_read(p,...) __intel_de_posting_read(__to_intel_display(p), __VA_ARGS__)
 
 static inline void
-__intel_de_write(struct intel_display *display, i915_reg_t reg, u32 val)
+intel_de_write(struct intel_display *display, i915_reg_t reg, u32 val)
 {
 	intel_dmc_wl_get(display, reg);
 
@@ -85,7 +82,6 @@ __intel_de_write(struct intel_display *display, i915_reg_t reg, u32 val)
 
 	intel_dmc_wl_put(display, reg);
 }
-#define intel_de_write(p,...) __intel_de_write(__to_intel_display(p), __VA_ARGS__)
 
 static inline u32
 __intel_de_rmw_nowl(struct intel_display *display, i915_reg_t reg,
@@ -95,8 +91,7 @@ __intel_de_rmw_nowl(struct intel_display *display, i915_reg_t reg,
 }
 
 static inline u32
-__intel_de_rmw(struct intel_display *display, i915_reg_t reg, u32 clear,
-	       u32 set)
+intel_de_rmw(struct intel_display *display, i915_reg_t reg, u32 clear, u32 set)
 {
 	u32 val;
 
@@ -108,7 +103,6 @@ __intel_de_rmw(struct intel_display *display, i915_reg_t reg, u32 clear,
 
 	return val;
 }
-#define intel_de_rmw(p,...) __intel_de_rmw(__to_intel_display(p), __VA_ARGS__)
 
 static inline int
 __intel_de_wait_for_register_nowl(struct intel_display *display,
@@ -181,20 +175,18 @@ intel_de_wait_custom(struct intel_display *display, i915_reg_t reg,
 }
 
 static inline int
-__intel_de_wait_for_set(struct intel_display *display, i915_reg_t reg,
-			u32 mask, unsigned int timeout)
+intel_de_wait_for_set(struct intel_display *display, i915_reg_t reg,
+		      u32 mask, unsigned int timeout)
 {
 	return intel_de_wait(display, reg, mask, mask, timeout);
 }
-#define intel_de_wait_for_set(p,...) __intel_de_wait_for_set(__to_intel_display(p), __VA_ARGS__)
 
 static inline int
-__intel_de_wait_for_clear(struct intel_display *display, i915_reg_t reg,
-			  u32 mask, unsigned int timeout)
+intel_de_wait_for_clear(struct intel_display *display, i915_reg_t reg,
+			u32 mask, unsigned int timeout)
 {
 	return intel_de_wait(display, reg, mask, 0, timeout);
 }
-#define intel_de_wait_for_clear(p,...) __intel_de_wait_for_clear(__to_intel_display(p), __VA_ARGS__)
 
 /*
  * Unlocked mmio-accessors, think carefully before using these.
@@ -205,7 +197,7 @@ __intel_de_wait_for_clear(struct intel_display *display, i915_reg_t reg,
  * a more localised lock guarding all access to that bank of registers.
  */
 static inline u32
-__intel_de_read_fw(struct intel_display *display, i915_reg_t reg)
+intel_de_read_fw(struct intel_display *display, i915_reg_t reg)
 {
 	u32 val;
 
@@ -214,15 +206,13 @@ __intel_de_read_fw(struct intel_display *display, i915_reg_t reg)
 
 	return val;
 }
-#define intel_de_read_fw(p,...) __intel_de_read_fw(__to_intel_display(p), __VA_ARGS__)
 
 static inline void
-__intel_de_write_fw(struct intel_display *display, i915_reg_t reg, u32 val)
+intel_de_write_fw(struct intel_display *display, i915_reg_t reg, u32 val)
 {
 	trace_i915_reg_rw(true, reg, val, sizeof(val), true);
 	intel_uncore_write_fw(__to_uncore(display), reg, val);
 }
-#define intel_de_write_fw(p,...) __intel_de_write_fw(__to_intel_display(p), __VA_ARGS__)
 
 static inline u32
 intel_de_read_notrace(struct intel_display *display, i915_reg_t reg)
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 3afb85fe8536..6f0a0bc71b06 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -55,6 +55,7 @@
 #include "i9xx_plane.h"
 #include "i9xx_plane_regs.h"
 #include "i9xx_wm.h"
+#include "intel_alpm.h"
 #include "intel_atomic.h"
 #include "intel_atomic_plane.h"
 #include "intel_audio.h"
@@ -73,6 +74,7 @@
 #include "intel_de.h"
 #include "intel_display_driver.h"
 #include "intel_display_power.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dmc.h"
 #include "intel_dp.h"
@@ -663,7 +665,6 @@ void intel_plane_disable_noatomic(struct intel_crtc *crtc,
 				  struct intel_plane *plane)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_crtc_state *crtc_state =
 		to_intel_crtc_state(crtc->base.state);
 	struct intel_plane_state *plane_state =
@@ -696,7 +697,7 @@ void intel_plane_disable_noatomic(struct intel_crtc *crtc,
 	 * wait-for-vblank between disabling the plane and the pipe.
 	 */
 	if (HAS_GMCH(display) &&
-	    intel_set_memory_cxsr(dev_priv, false))
+	    intel_set_memory_cxsr(display, false))
 		intel_plane_initial_vblank_wait(crtc);
 
 	/*
@@ -968,7 +969,9 @@ static bool vrr_params_changed(const struct intel_crtc_state *old_crtc_state,
 		old_crtc_state->vrr.vmin != new_crtc_state->vrr.vmin ||
 		old_crtc_state->vrr.vmax != new_crtc_state->vrr.vmax ||
 		old_crtc_state->vrr.guardband != new_crtc_state->vrr.guardband ||
-		old_crtc_state->vrr.pipeline_full != new_crtc_state->vrr.pipeline_full;
+		old_crtc_state->vrr.pipeline_full != new_crtc_state->vrr.pipeline_full ||
+		old_crtc_state->vrr.vsync_start != new_crtc_state->vrr.vsync_start ||
+		old_crtc_state->vrr.vsync_end != new_crtc_state->vrr.vsync_end;
 }
 
 static bool cmrr_params_changed(const struct intel_crtc_state *old_crtc_state,
@@ -1041,19 +1044,16 @@ static void intel_post_plane_update(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	const struct intel_crtc_state *old_crtc_state =
 		intel_atomic_get_old_crtc_state(state, crtc);
 	const struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	enum pipe pipe = crtc->pipe;
 
-	intel_psr_post_plane_update(state, crtc);
-
-	intel_frontbuffer_flip(dev_priv, new_crtc_state->fb_bits);
+	intel_frontbuffer_flip(display, new_crtc_state->fb_bits);
 
 	if (new_crtc_state->update_wm_post && new_crtc_state->hw.active)
-		intel_update_watermarks(dev_priv);
+		intel_update_watermarks(display);
 
 	intel_fbc_post_update(state, crtc);
 
@@ -1078,6 +1078,10 @@ static void intel_post_plane_update(struct intel_atomic_state *state,
 
 	if (audio_enabling(old_crtc_state, new_crtc_state))
 		intel_encoders_audio_enable(state, crtc);
+
+	intel_alpm_post_plane_update(state, crtc);
+
+	intel_psr_post_plane_update(state, crtc);
 }
 
 static void intel_post_plane_update_after_readout(struct intel_atomic_state *state,
@@ -1166,13 +1170,15 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	const struct intel_crtc_state *old_crtc_state =
 		intel_atomic_get_old_crtc_state(state, crtc);
 	const struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	enum pipe pipe = crtc->pipe;
 
+	intel_alpm_pre_plane_update(state, crtc);
+	intel_psr_pre_plane_update(state, crtc);
+
 	if (intel_crtc_vrr_disabling(state, crtc)) {
 		intel_vrr_disable(old_crtc_state);
 		intel_crtc_update_active_timings(old_crtc_state, false);
@@ -1183,8 +1189,6 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 
 	intel_drrs_deactivate(old_crtc_state);
 
-	intel_psr_pre_plane_update(state, crtc);
-
 	if (hsw_ips_pre_update(state, crtc))
 		intel_crtc_wait_for_next_vblank(crtc);
 
@@ -1220,7 +1224,7 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 	 * wait-for-vblank between disabling the plane and the pipe.
 	 */
 	if (HAS_GMCH(display) && old_crtc_state->hw.active &&
-	    new_crtc_state->disable_cxsr && intel_set_memory_cxsr(dev_priv, false))
+	    new_crtc_state->disable_cxsr && intel_set_memory_cxsr(display, false))
 		intel_crtc_wait_for_next_vblank(crtc);
 
 	/*
@@ -1231,7 +1235,7 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 	 * WaCxSRDisabledForSpriteScaling:ivb
 	 */
 	if (!HAS_GMCH(display) && old_crtc_state->hw.active &&
-	    new_crtc_state->disable_cxsr && ilk_disable_cxsr(dev_priv))
+	    new_crtc_state->disable_cxsr && ilk_disable_cxsr(display))
 		intel_crtc_wait_for_next_vblank(crtc);
 
 	/*
@@ -1255,7 +1259,7 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 		 */
 		if (!intel_initial_watermarks(state, crtc))
 			if (new_crtc_state->update_wm_pre)
-				intel_update_watermarks(dev_priv);
+				intel_update_watermarks(display);
 	}
 
 	/*
@@ -1280,7 +1284,7 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 static void intel_crtc_disable_planes(struct intel_atomic_state *state,
 				      struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	unsigned int update_mask = new_crtc_state->update_planes;
@@ -1302,7 +1306,7 @@ static void intel_crtc_disable_planes(struct intel_atomic_state *state,
 			fb_bits |= plane->frontbuffer_bit;
 	}
 
-	intel_frontbuffer_flip(dev_priv, fb_bits);
+	intel_frontbuffer_flip(display, fb_bits);
 }
 
 static void intel_encoders_update_prepare(struct intel_atomic_state *state)
@@ -1509,7 +1513,6 @@ static void ilk_crtc_enable(struct intel_atomic_state *state,
 	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
 	if (drm_WARN_ON(display->drm, crtc->active))
@@ -1561,7 +1564,7 @@ static void ilk_crtc_enable(struct intel_atomic_state *state,
 
 	intel_encoders_enable(state, crtc);
 
-	if (HAS_PCH_CPT(dev_priv))
+	if (HAS_PCH_CPT(display))
 		intel_wait_for_pipe_scanline_moving(crtc);
 
 	/*
@@ -1660,13 +1663,8 @@ static void hsw_crtc_enable(struct intel_atomic_state *state,
 
 	intel_encoders_pre_pll_enable(state, crtc);
 
-	for_each_pipe_crtc_modeset_enable(display, pipe_crtc, new_crtc_state, i) {
-		const struct intel_crtc_state *pipe_crtc_state =
-			intel_atomic_get_new_crtc_state(state, pipe_crtc);
-
-		if (pipe_crtc_state->shared_dpll)
-			intel_enable_shared_dpll(pipe_crtc_state);
-	}
+	if (new_crtc_state->shared_dpll)
+		intel_enable_shared_dpll(new_crtc_state);
 
 	intel_encoders_pre_enable(state, crtc);
 
@@ -1777,8 +1775,6 @@ static void ilk_crtc_disable(struct intel_atomic_state *state,
 
 	intel_set_cpu_fifo_underrun_reporting(display, pipe, true);
 	intel_set_pch_fifo_underrun_reporting(display, pipe, true);
-
-	intel_disable_shared_dpll(old_crtc_state);
 }
 
 static void hsw_crtc_disable(struct intel_atomic_state *state,
@@ -1797,12 +1793,7 @@ static void hsw_crtc_disable(struct intel_atomic_state *state,
 	intel_encoders_disable(state, crtc);
 	intel_encoders_post_disable(state, crtc);
 
-	for_each_pipe_crtc_modeset_disable(display, pipe_crtc, old_crtc_state, i) {
-		const struct intel_crtc_state *old_pipe_crtc_state =
-			intel_atomic_get_old_crtc_state(state, pipe_crtc);
-
-		intel_disable_shared_dpll(old_pipe_crtc_state);
-	}
+	intel_disable_shared_dpll(old_crtc_state);
 
 	intel_encoders_post_pll_disable(state, crtc);
 
@@ -2081,7 +2072,6 @@ static void i9xx_crtc_enable(struct intel_atomic_state *state,
 	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
 	if (drm_WARN_ON(display->drm, crtc->active))
@@ -2105,7 +2095,7 @@ static void i9xx_crtc_enable(struct intel_atomic_state *state,
 	intel_color_modeset(new_crtc_state);
 
 	if (!intel_initial_watermarks(state, crtc))
-		intel_update_watermarks(dev_priv);
+		intel_update_watermarks(display);
 	intel_enable_transcoder(new_crtc_state);
 
 	intel_crtc_vblank_on(new_crtc_state);
@@ -2121,7 +2111,6 @@ static void i9xx_crtc_disable(struct intel_atomic_state *state,
 			      struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc_state *old_crtc_state =
 		intel_atomic_get_old_crtc_state(state, crtc);
 	enum pipe pipe = crtc->pipe;
@@ -2145,9 +2134,9 @@ static void i9xx_crtc_disable(struct intel_atomic_state *state,
 
 	if (!intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_DSI)) {
 		if (display->platform.cherryview)
-			chv_disable_pll(dev_priv, pipe);
+			chv_disable_pll(display, pipe);
 		else if (display->platform.valleyview)
-			vlv_disable_pll(dev_priv, pipe);
+			vlv_disable_pll(display, pipe);
 		else
 			i9xx_disable_pll(old_crtc_state);
 	}
@@ -2158,7 +2147,7 @@ static void i9xx_crtc_disable(struct intel_atomic_state *state,
 		intel_set_cpu_fifo_underrun_reporting(display, pipe, false);
 
 	if (!display->funcs.wm->initial_watermarks)
-		intel_update_watermarks(dev_priv);
+		intel_update_watermarks(display);
 
 	/* clock the pipe down to 640x480@60 to potentially save power */
 	if (display->platform.i830)
@@ -2341,7 +2330,6 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 
 	intel_joiner_compute_pipe_src(crtc_state);
 
@@ -2360,7 +2348,7 @@ static int intel_crtc_compute_pipe_src(struct intel_crtc_state *crtc_state)
 		}
 
 		if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS) &&
-		    intel_is_dual_link_lvds(i915)) {
+		    intel_is_dual_link_lvds(display)) {
 			drm_dbg_kms(display->drm,
 				    "[CRTC:%d:%s] Odd pipe source width not supported with dual link LVDS\n",
 				    crtc->base.base.id, crtc->base.name);
@@ -2418,14 +2406,6 @@ static int intel_crtc_compute_pipe_mode(struct intel_crtc_state *crtc_state)
 	return 0;
 }
 
-static bool intel_crtc_needs_wa_14015401596(const struct intel_crtc_state *crtc_state)
-{
-	struct intel_display *display = to_intel_display(crtc_state);
-
-	return intel_vrr_possible(crtc_state) && crtc_state->has_psr &&
-		IS_DISPLAY_VER(display, 13, 14);
-}
-
 static int intel_crtc_vblank_delay(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
@@ -2434,9 +2414,7 @@ static int intel_crtc_vblank_delay(const struct intel_crtc_state *crtc_state)
 	if (!HAS_DSB(display))
 		return 0;
 
-	/* Wa_14015401596 */
-	if (intel_crtc_needs_wa_14015401596(crtc_state))
-		vblank_delay = max(vblank_delay, 1);
+	vblank_delay = max(vblank_delay, intel_psr_min_vblank_delay(crtc_state));
 
 	return vblank_delay;
 }
@@ -2548,15 +2526,13 @@ intel_link_compute_m_n(u16 bits_per_pixel_x16, int nlanes,
 
 void intel_panel_sanitize_ssc(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	/*
 	 * There may be no VBT; and if the BIOS enabled SSC we can
 	 * just keep using it to avoid unnecessary flicker.  Whereas if the
 	 * BIOS isn't using it, don't assume it will work even if the VBT
 	 * indicates as much.
 	 */
-	if (HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_IBX(display) || HAS_PCH_CPT(display)) {
 		bool bios_lvds_use_ssc = intel_de_read(display,
 						       PCH_DREF_CONTROL) &
 			DREF_SSC1_ENABLE;
@@ -2637,6 +2613,15 @@ void intel_cpu_transcoder_set_m2_n2(struct intel_crtc *crtc,
 		      PIPE_LINK_N2(display, transcoder));
 }
 
+static bool
+transcoder_has_vrr(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+
+	return HAS_VRR(display) && !transcoder_is_dsi(cpu_transcoder);
+}
+
 static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
@@ -2701,6 +2686,15 @@ static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_sta
 		       HSYNC_START(adjusted_mode->crtc_hsync_start - 1) |
 		       HSYNC_END(adjusted_mode->crtc_hsync_end - 1));
 
+	/*
+	 * For platforms that always use VRR Timing Generator, the VTOTAL.Vtotal
+	 * bits are not required. Since the support for these bits is going to
+	 * be deprecated in upcoming platforms, avoid writing these bits for the
+	 * platforms that do not use legacy Timing Generator.
+	 */
+	if (intel_vrr_always_use_vrr_tg(display))
+		crtc_vtotal = 1;
+
 	intel_de_write(display, TRANS_VTOTAL(display, cpu_transcoder),
 		       VACTIVE(crtc_vdisplay - 1) |
 		       VTOTAL(crtc_vtotal - 1));
@@ -2720,6 +2714,19 @@ static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_sta
 		intel_de_write(display, TRANS_VTOTAL(display, pipe),
 			       VACTIVE(crtc_vdisplay - 1) |
 			       VTOTAL(crtc_vtotal - 1));
+
+	if (DISPLAY_VER(display) >= 30) {
+		/*
+		 * Address issues for resolutions with high refresh rate that
+		 * have small Hblank, specifically where Hblank is smaller than
+		 * one MTP. Simulations indicate this will address the
+		 * jitter issues that currently causes BS to be immediately
+		 * followed by BE which DPRX devices are unable to handle.
+		 * https://groups.vesa.org/wg/DP/document/20494
+		 */
+		intel_de_write(display, DP_MIN_HBLANK_CTL(cpu_transcoder),
+			       crtc_state->min_hblank);
+	}
 }
 
 static void intel_set_transcoder_timings_lrr(const struct intel_crtc_state *crtc_state)
@@ -2762,12 +2769,24 @@ static void intel_set_transcoder_timings_lrr(const struct intel_crtc_state *crtc
 		       VBLANK_START(crtc_vblank_start - 1) |
 		       VBLANK_END(crtc_vblank_end - 1));
 	/*
+	 * For platforms that always use VRR Timing Generator, the VTOTAL.Vtotal
+	 * bits are not required. Since the support for these bits is going to
+	 * be deprecated in upcoming platforms, avoid writing these bits for the
+	 * platforms that do not use legacy Timing Generator.
+	 */
+	if (intel_vrr_always_use_vrr_tg(display))
+		crtc_vtotal = 1;
+
+	/*
 	 * The double buffer latch point for TRANS_VTOTAL
 	 * is the transcoder's undelayed vblank.
 	 */
 	intel_de_write(display, TRANS_VTOTAL(display, cpu_transcoder),
 		       VACTIVE(crtc_vdisplay - 1) |
 		       VTOTAL(crtc_vtotal - 1));
+
+	intel_vrr_set_fixed_rr_timings(crtc_state);
+	intel_vrr_transcoder_enable(crtc_state);
 }
 
 static void intel_set_pipe_src_size(const struct intel_crtc_state *crtc_state)
@@ -2851,6 +2870,10 @@ static void intel_get_transcoder_timings(struct intel_crtc *crtc,
 			adjusted_mode->crtc_vdisplay +
 			intel_de_read(display,
 				      TRANS_SET_CONTEXT_LATENCY(display, cpu_transcoder));
+
+	if (DISPLAY_VER(display) >= 30)
+		pipe_config->min_hblank = intel_de_read(display,
+							DP_MIN_HBLANK_CTL(cpu_transcoder));
 }
 
 static void intel_joiner_adjust_pipe_src(struct intel_crtc_state *crtc_state)
@@ -3833,7 +3856,6 @@ static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc,
 					 struct intel_display_power_domain_set *power_domain_set)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum transcoder cpu_transcoder;
 	enum port port;
 	u32 tmp;
@@ -3855,7 +3877,7 @@ static bool bxt_get_dsi_transcoder_state(struct intel_crtc *crtc,
 		 * registers/MIPI[BXT]. We can break out here early, since we
 		 * need the same DSI PLL to be enabled for both DSI ports.
 		 */
-		if (!bxt_dsi_pll_is_enabled(dev_priv))
+		if (!bxt_dsi_pll_is_enabled(display))
 			break;
 
 		/* XXX: this works for video mode only */
@@ -3918,7 +3940,7 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc,
 	    DISPLAY_VER(display) >= 11)
 		intel_get_transcoder_timings(crtc, pipe_config);
 
-	if (HAS_VRR(display) && !transcoder_is_dsi(pipe_config->cpu_transcoder))
+	if (transcoder_has_vrr(pipe_config))
 		intel_vrr_get_config(pipe_config);
 
 	intel_get_pipe_src_size(crtc, pipe_config);
@@ -4145,8 +4167,6 @@ static u16 hsw_ips_linetime_wm(const struct intel_crtc_state *crtc_state,
 static u16 skl_linetime_wm(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct drm_display_mode *pipe_mode =
 		&crtc_state->hw.pipe_mode;
 	int linetime_wm;
@@ -4159,7 +4179,7 @@ static u16 skl_linetime_wm(const struct intel_crtc_state *crtc_state)
 
 	/* Display WA #1135: BXT:ALL GLK:ALL */
 	if ((display->platform.geminilake || display->platform.broxton) &&
-	    skl_watermark_ipc_enabled(dev_priv))
+	    skl_watermark_ipc_enabled(display))
 		linetime_wm /= 2;
 
 	return min(linetime_wm, 0x1ff);
@@ -5204,6 +5224,8 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 	PIPE_CONF_CHECK_I(lane_count);
 	PIPE_CONF_CHECK_X(lane_lat_optim_mask);
 
+	PIPE_CONF_CHECK_I(min_hblank);
+
 	if (HAS_DOUBLE_BUFFERED_M_N(display)) {
 		if (!fastset || !pipe_config->update_m_n)
 			PIPE_CONF_CHECK_M_N(dp_m_n);
@@ -5385,8 +5407,6 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 		PIPE_CONF_CHECK_I(vrr.vmin);
 		PIPE_CONF_CHECK_I(vrr.vmax);
 		PIPE_CONF_CHECK_I(vrr.flipline);
-		PIPE_CONF_CHECK_I(vrr.pipeline_full);
-		PIPE_CONF_CHECK_I(vrr.guardband);
 		PIPE_CONF_CHECK_I(vrr.vsync_start);
 		PIPE_CONF_CHECK_I(vrr.vsync_end);
 		PIPE_CONF_CHECK_LLI(cmrr.cmrr_m);
@@ -5394,6 +5414,11 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 		PIPE_CONF_CHECK_BOOL(cmrr.enable);
 	}
 
+	if (!fastset || intel_vrr_always_use_vrr_tg(display)) {
+		PIPE_CONF_CHECK_I(vrr.pipeline_full);
+		PIPE_CONF_CHECK_I(vrr.guardband);
+	}
+
 #undef PIPE_CONF_CHECK_X
 #undef PIPE_CONF_CHECK_I
 #undef PIPE_CONF_CHECK_LLI
@@ -6005,22 +6030,16 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in
 		if (!plane->async_flip)
 			continue;
 
-		if (!intel_plane_can_async_flip(plane, new_plane_state->hw.fb->modifier)) {
+		if (!intel_plane_can_async_flip(plane, new_plane_state->hw.fb->format->format,
+						new_plane_state->hw.fb->modifier)) {
 			drm_dbg_kms(display->drm,
-				    "[PLANE:%d:%s] Modifier 0x%llx does not support async flip\n",
+				    "[PLANE:%d:%s] pixel format %p4cc / modifier 0x%llx does not support async flip\n",
 				    plane->base.base.id, plane->base.name,
+				    &new_plane_state->hw.fb->format->format,
 				    new_plane_state->hw.fb->modifier);
 			return -EINVAL;
 		}
 
-		if (intel_format_info_is_yuv_semiplanar(new_plane_state->hw.fb->format,
-							new_plane_state->hw.fb->modifier)) {
-			drm_dbg_kms(display->drm,
-				    "[PLANE:%d:%s] Planar formats do not support async flips\n",
-				    plane->base.base.id, plane->base.name);
-			return -EINVAL;
-		}
-
 		/*
 		 * We turn the first async flip request into a sync flip
 		 * so that we can reconfigure the plane (eg. change modifier).
@@ -6427,7 +6446,7 @@ int intel_atomic_check(struct drm_device *dev,
 	if (ret)
 		goto fail;
 
-	ret = intel_bw_atomic_check(state);
+	ret = intel_bw_atomic_check(state, any_ms);
 	if (ret)
 		goto fail;
 
@@ -6531,7 +6550,6 @@ static void intel_pipe_fastset(const struct intel_crtc_state *old_crtc_state,
 {
 	struct intel_display *display = to_intel_display(new_crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 
 	/*
 	 * Update pipe size and adjust fitter if needed: the reason for this is
@@ -6547,7 +6565,7 @@ static void intel_pipe_fastset(const struct intel_crtc_state *old_crtc_state,
 	if (DISPLAY_VER(display) >= 9) {
 		if (new_crtc_state->pch_pfit.enabled)
 			skl_pfit_enable(new_crtc_state);
-	} else if (HAS_PCH_SPLIT(dev_priv)) {
+	} else if (HAS_PCH_SPLIT(display)) {
 		if (new_crtc_state->pch_pfit.enabled)
 			ilk_pfit_enable(new_crtc_state);
 		else if (old_crtc_state->pch_pfit.enabled)
@@ -6648,6 +6666,8 @@ static void intel_enable_crtc(struct intel_atomic_state *state,
 		intel_crtc_update_active_timings(pipe_crtc_state, false);
 	}
 
+	intel_psr_notify_pipe_change(state, crtc, true);
+
 	display->funcs.display->crtc_enable(state, crtc);
 
 	/* vblanks work again, re-enable pipe CRC. */
@@ -6767,6 +6787,8 @@ static void intel_old_crtc_state_disables(struct intel_atomic_state *state,
 					 intel_crtc_joined_pipe_mask(old_crtc_state))
 		intel_crtc_disable_pipe_crc(pipe_crtc);
 
+	intel_psr_notify_pipe_change(state, crtc, false);
+
 	display->funcs.display->crtc_disable(state, crtc);
 
 	for_each_intel_crtc_in_pipe_mask(display->drm, pipe_crtc,
@@ -7229,7 +7251,7 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state,
 static void intel_atomic_commit_tail(struct intel_atomic_state *state)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	struct drm_i915_private __maybe_unused *dev_priv = to_i915(display->drm);
 	struct intel_crtc_state *new_crtc_state, *old_crtc_state;
 	struct intel_crtc *crtc;
 	struct intel_power_domain_mask put_domains[I915_MAX_PIPES] = {};
@@ -7443,7 +7465,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state)
 	 * toggling overhead at and above 60 FPS.
 	 */
 	intel_display_power_put_async_delay(display, POWER_DOMAIN_DC_OFF, wakeref, 17);
-	intel_runtime_pm_put(&dev_priv->runtime_pm, state->wakeref);
+	intel_display_rpm_put(display, state->wakeref);
 
 	/*
 	 * Defer the cleanup of the old state to a separate worker to not
@@ -7515,10 +7537,9 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
 {
 	struct intel_display *display = to_intel_display(dev);
 	struct intel_atomic_state *state = to_intel_atomic_state(_state);
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	int ret = 0;
 
-	state->wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	state->wakeref = intel_display_rpm_get(display);
 
 	/*
 	 * The intel_legacy_cursor_update() fast path takes care
@@ -7552,7 +7573,7 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
 	if (ret) {
 		drm_dbg_atomic(display->drm,
 			       "Preparing state failed with %i\n", ret);
-		intel_runtime_pm_put(&dev_priv->runtime_pm, state->wakeref);
+		intel_display_rpm_put(display, state->wakeref);
 		return ret;
 	}
 
@@ -7562,7 +7583,7 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
 
 	if (ret) {
 		drm_atomic_helper_unprepare_planes(dev, &state->base);
-		intel_runtime_pm_put(&dev_priv->runtime_pm, state->wakeref);
+		intel_display_rpm_put(display, state->wakeref);
 		return ret;
 	}
 
@@ -7624,15 +7645,13 @@ static bool ilk_has_edp_a(struct intel_display *display)
 
 static bool intel_ddi_crt_present(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	if (DISPLAY_VER(display) >= 9)
 		return false;
 
 	if (display->platform.haswell_ult || display->platform.broadwell_ult)
 		return false;
 
-	if (HAS_PCH_LPT_H(dev_priv) &&
+	if (HAS_PCH_LPT_H(display) &&
 	    intel_de_read(display, SFUSE_STRAP) & SFUSE_STRAP_CRT_DISABLED)
 		return false;
 
@@ -7654,7 +7673,6 @@ bool assert_port_valid(struct intel_display *display, enum port port)
 
 void intel_setup_outputs(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_encoder *encoder;
 	bool dpd_is_edp = false;
 
@@ -7670,8 +7688,8 @@ void intel_setup_outputs(struct intel_display *display)
 		intel_bios_for_each_encoder(display, intel_ddi_init);
 
 		if (display->platform.geminilake || display->platform.broxton)
-			vlv_dsi_init(dev_priv);
-	} else if (HAS_PCH_SPLIT(dev_priv)) {
+			vlv_dsi_init(display);
+	} else if (HAS_PCH_SPLIT(display)) {
 		int found;
 
 		/*
@@ -7679,7 +7697,7 @@ void intel_setup_outputs(struct intel_display *display)
 		 * to prevent the registration of both eDP and LVDS and the
 		 * incorrect sharing of the PPS.
 		 */
-		intel_lvds_init(dev_priv);
+		intel_lvds_init(display);
 		intel_crt_init(display);
 
 		dpd_is_edp = intel_dp_is_port_edp(display, PORT_D);
@@ -7754,15 +7772,15 @@ void intel_setup_outputs(struct intel_display *display)
 				g4x_hdmi_init(display, CHV_HDMID, PORT_D);
 		}
 
-		vlv_dsi_init(dev_priv);
+		vlv_dsi_init(display);
 	} else if (display->platform.pineview) {
-		intel_lvds_init(dev_priv);
+		intel_lvds_init(display);
 		intel_crt_init(display);
 	} else if (IS_DISPLAY_VER(display, 3, 4)) {
 		bool found = false;
 
 		if (display->platform.mobile)
-			intel_lvds_init(dev_priv);
+			intel_lvds_init(display);
 
 		intel_crt_init(display);
 
@@ -7804,10 +7822,10 @@ void intel_setup_outputs(struct intel_display *display)
 			intel_tv_init(display);
 	} else if (DISPLAY_VER(display) == 2) {
 		if (display->platform.i85x)
-			intel_lvds_init(dev_priv);
+			intel_lvds_init(display);
 
 		intel_crt_init(display);
-		intel_dvo_init(dev_priv);
+		intel_dvo_init(display);
 	}
 
 	for_each_intel_encoder(display->drm, encoder) {
@@ -7817,7 +7835,7 @@ void intel_setup_outputs(struct intel_display *display)
 			intel_encoder_possible_clones(encoder);
 	}
 
-	intel_init_pch_refclk(dev_priv);
+	intel_init_pch_refclk(display);
 
 	drm_helper_move_panel_connectors_to_head(display->drm);
 }
@@ -8039,13 +8057,11 @@ static const struct intel_display_funcs i9xx_display_funcs = {
  */
 void intel_init_display_hooks(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	if (DISPLAY_VER(display) >= 9) {
 		display->funcs.display = &skl_display_funcs;
 	} else if (HAS_DDI(display)) {
 		display->funcs.display = &ddi_display_funcs;
-	} else if (HAS_PCH_SPLIT(dev_priv)) {
+	} else if (HAS_PCH_SPLIT(display)) {
 		display->funcs.display = &pch_split_display_funcs;
 	} else if (display->platform.cherryview ||
 		   display->platform.valleyview) {
@@ -8081,6 +8097,9 @@ retry:
 			goto out;
 		}
 
+		if (!crtc_state->hw.active)
+			crtc_state->inherited = false;
+
 		if (crtc_state->hw.active) {
 			struct intel_encoder *encoder;
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h
index eeb7ae3eaea8..b4937e102360 100644
--- a/drivers/gpu/drm/i915/display/intel_display_core.h
+++ b/drivers/gpu/drm/i915/display/intel_display_core.h
@@ -21,17 +21,15 @@
 #include "intel_display_limits.h"
 #include "intel_display_params.h"
 #include "intel_display_power.h"
+#include "intel_dmc_wl.h"
 #include "intel_dpll_mgr.h"
 #include "intel_fbc.h"
 #include "intel_global_state.h"
 #include "intel_gmbus.h"
 #include "intel_opregion.h"
-#include "intel_dmc_wl.h"
+#include "intel_pch.h"
 #include "intel_wm_types.h"
 
-struct task_struct;
-
-struct drm_i915_private;
 struct drm_property;
 struct drm_property_blob;
 struct i915_audio_component;
@@ -52,6 +50,7 @@ struct intel_hotplug_funcs;
 struct intel_initial_plane_config;
 struct intel_opregion;
 struct intel_overlay;
+struct task_struct;
 
 /* Amount of SAGV/QGV points, BSpec precisely defines this */
 #define I915_NUM_QGV_POINTS 8
@@ -80,7 +79,7 @@ struct intel_display_funcs {
 /* functions used for watermark calcs for display. */
 struct intel_wm_funcs {
 	/* update_wm is for legacy wm management */
-	void (*update_wm)(struct drm_i915_private *dev_priv);
+	void (*update_wm)(struct intel_display *display);
 	int (*compute_watermarks)(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc);
 	void (*initial_watermarks)(struct intel_atomic_state *state,
@@ -90,8 +89,8 @@ struct intel_wm_funcs {
 	void (*optimize_watermarks)(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc);
 	int (*compute_global_watermarks)(struct intel_atomic_state *state);
-	void (*get_hw_state)(struct drm_i915_private *i915);
-	void (*sanitize)(struct drm_i915_private *i915);
+	void (*get_hw_state)(struct intel_display *display);
+	void (*sanitize)(struct intel_display *display);
 };
 
 struct intel_audio_state {
@@ -160,6 +159,7 @@ struct intel_hotplug {
 	struct {
 		unsigned long last_jiffies;
 		int count;
+		int blocked_count;
 		enum {
 			HPD_ENABLED = 0,
 			HPD_DISABLED = 1,
@@ -170,8 +170,8 @@ struct intel_hotplug {
 	u32 retry_bits;
 	struct delayed_work reenable_work;
 
-	u32 long_port_mask;
-	u32 short_port_mask;
+	u32 long_hpd_pin_mask;
+	u32 short_hpd_pin_mask;
 	struct work_struct dig_port_work;
 
 	struct work_struct poll_init_work;
@@ -179,7 +179,7 @@ struct intel_hotplug {
 
 	/*
 	 * Queuing of hotplug_work, reenable_work and poll_init_work is
-	 * enabled. Protected by drm_i915_private::irq_lock.
+	 * enabled. Protected by intel_display::irq::lock.
 	 */
 	bool detection_work_enabled;
 
@@ -288,6 +288,9 @@ struct intel_display {
 	/* Platform (and subplatform, if any) identification */
 	struct intel_display_platforms platform;
 
+	/* Intel PCH: where the south display engine lives */
+	enum intel_pch pch_type;
+
 	/* Display functions */
 	struct {
 		/* Top level crtc-ish functions */
@@ -425,7 +428,7 @@ struct intel_display {
 		 * reused when sending message to gsc cs.
 		 * this is only populated post Meteorlake
 		 */
-		struct intel_hdcp_gsc_message *hdcp_message;
+		struct intel_hdcp_gsc_context *gsc_context;
 		/* Mutex to protect the above hdcp related values. */
 		struct mutex hdcp_mutex;
 	} hdcp;
@@ -453,6 +456,9 @@ struct intel_display {
 	} ips;
 
 	struct {
+		/* protects the irq masks */
+		spinlock_t lock;
+
 		/*
 		 * Most platforms treat the display irq block as an always-on
 		 * power domain. vlv/chv can disable it at runtime and need
@@ -465,9 +471,9 @@ struct intel_display {
 		/* For i915gm/i945gm vblank irq workaround */
 		u8 vblank_enabled;
 
-		int vblank_wa_num_pipes;
+		int vblank_enable_count;
 
-		struct work_struct vblank_dc_work;
+		struct work_struct vblank_notify_work;
 
 		u32 de_irq_mask[I915_MAX_PIPES];
 		u32 pipestat_irq_mask[I915_MAX_PIPES];
@@ -574,6 +580,8 @@ struct intel_display {
 	struct intel_vbt_data vbt;
 	struct intel_dmc_wl wl;
 	struct intel_wm wm;
+
+	struct work_struct psr_dc5_dc6_wa_work;
 };
 
 #endif /* __INTEL_DISPLAY_CORE_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index fdedf65bee53..8d0a1779dd19 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -7,11 +7,12 @@
 #include <linux/string_helpers.h>
 
 #include <drm/drm_debugfs.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_file.h>
 #include <drm/drm_fourcc.h>
 
 #include "hsw_ips.h"
-#include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
 #include "i9xx_wm_regs.h"
@@ -24,6 +25,7 @@
 #include "intel_display_debugfs_params.h"
 #include "intel_display_power.h"
 #include "intel_display_power_well.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dmc.h"
 #include "intel_dp.h"
@@ -54,6 +56,8 @@ static int intel_display_caps(struct seq_file *m, void *data)
 	struct intel_display *display = node_to_intel_display(m->private);
 	struct drm_printer p = drm_seq_file_printer(m);
 
+	drm_printf(&p, "PCH type: %d\n", INTEL_PCH_TYPE(display));
+
 	intel_display_device_info_print(DISPLAY_INFO(display),
 					DISPLAY_RUNTIME_INFO(display), &p);
 	intel_display_params_dump(&display->params, display->drm->driver->name, &p);
@@ -81,7 +85,6 @@ static int i915_frontbuffer_tracking(struct seq_file *m, void *unused)
 static int i915_sr_status(struct seq_file *m, void *unused)
 {
 	struct intel_display *display = node_to_intel_display(m->private);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	intel_wakeref_t wakeref;
 	bool sr_enabled = false;
 
@@ -89,7 +92,7 @@ static int i915_sr_status(struct seq_file *m, void *unused)
 
 	if (DISPLAY_VER(display) >= 9)
 		/* no global SR status; inspect per-plane WM */;
-	else if (HAS_PCH_SPLIT(dev_priv))
+	else if (HAS_PCH_SPLIT(display))
 		sr_enabled = intel_de_read(display, WM1_LP_ILK) & WM_LP_ENABLE;
 	else if (display->platform.i965gm || display->platform.g4x ||
 		 display->platform.i945g || display->platform.i945gm)
@@ -554,6 +557,8 @@ static void intel_crtc_info(struct seq_file *m, struct intel_crtc *crtc)
 	seq_printf(m, "\tpipe src=" DRM_RECT_FMT ", dither=%s, bpp=%d\n",
 		   DRM_RECT_ARG(&crtc_state->pipe_src),
 		   str_yes_no(crtc_state->dither), crtc_state->pipe_bpp);
+	seq_printf(m, "\tport_clock=%d, lane_count=%d\n",
+		   crtc_state->port_clock, crtc_state->lane_count);
 
 	intel_scaler_info(m, crtc);
 
@@ -580,13 +585,12 @@ static void intel_crtc_info(struct seq_file *m, struct intel_crtc *crtc)
 static int i915_display_info(struct seq_file *m, void *unused)
 {
 	struct intel_display *display = node_to_intel_display(m->private);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc;
 	struct drm_connector *connector;
 	struct drm_connector_list_iter conn_iter;
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 
-	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
 	drm_modeset_lock_all(display->drm);
 
@@ -605,18 +609,7 @@ static int i915_display_info(struct seq_file *m, void *unused)
 
 	drm_modeset_unlock_all(display->drm);
 
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
-
-	return 0;
-}
-
-static int i915_display_capabilities(struct seq_file *m, void *unused)
-{
-	struct intel_display *display = node_to_intel_display(m->private);
-	struct drm_printer p = drm_seq_file_printer(m);
-
-	intel_display_device_info_print(DISPLAY_INFO(display),
-					DISPLAY_RUNTIME_INFO(display), &p);
+	intel_display_rpm_put(display, wakeref);
 
 	return 0;
 }
@@ -690,14 +683,11 @@ static bool
 intel_lpsp_power_well_enabled(struct intel_display *display,
 			      enum i915_power_well_id power_well_id)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-	intel_wakeref_t wakeref;
 	bool is_enabled;
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
-	is_enabled = intel_display_power_well_is_enabled(display,
-							 power_well_id);
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	with_intel_display_rpm(display)
+		is_enabled = intel_display_power_well_is_enabled(display,
+								 power_well_id);
 
 	return is_enabled;
 }
@@ -820,7 +810,6 @@ static const struct drm_info_list intel_display_debugfs_list[] = {
 	{"i915_gem_framebuffer", i915_gem_framebuffer_info, 0},
 	{"i915_power_domain_info", i915_power_domain_info, 0},
 	{"i915_display_info", i915_display_info, 0},
-	{"i915_display_capabilities", i915_display_capabilities, 0},
 	{"i915_shared_dplls_info", i915_shared_dplls_info, 0},
 	{"i915_dp_mst_info", i915_dp_mst_info, 0},
 	{"i915_ddb_info", i915_ddb_info, 0},
@@ -829,7 +818,6 @@ static const struct drm_info_list intel_display_debugfs_list[] = {
 
 void intel_display_debugfs_register(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct drm_minor *minor = display->drm->primary;
 
 	debugfs_create_file("i915_fifo_underrun_reset", 0644, minor->debugfs_root,
@@ -844,10 +832,10 @@ void intel_display_debugfs_register(struct intel_display *display)
 	intel_dmc_debugfs_register(display);
 	intel_dp_test_debugfs_register(display);
 	intel_fbc_debugfs_register(display);
-	intel_hpd_debugfs_register(i915);
+	intel_hpd_debugfs_register(display);
 	intel_opregion_debugfs_register(display);
 	intel_psr_debugfs_register(display);
-	intel_wm_debugfs_register(i915);
+	intel_wm_debugfs_register(display);
 	intel_display_debugfs_params(display);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index 738ae522c8f4..90d714598664 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -3,11 +3,13 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#include <drm/intel/pciids.h>
-#include <drm/drm_color_mgmt.h>
 #include <linux/pci.h>
 
-#include "i915_drv.h"
+#include <drm/drm_color_mgmt.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_print.h>
+#include <drm/intel/pciids.h>
+
 #include "i915_reg.h"
 #include "intel_cx0_phy_regs.h"
 #include "intel_de.h"
@@ -1711,7 +1713,6 @@ void intel_display_device_remove(struct intel_display *display)
 
 static void __intel_display_device_info_runtime_init(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_display_runtime_info *display_runtime = DISPLAY_RUNTIME_INFO(display);
 	enum pipe pipe;
 
@@ -1775,7 +1776,7 @@ static void __intel_display_device_info_runtime_init(struct intel_display *displ
 		goto display_fused_off;
 	}
 
-	if (IS_DISPLAY_VER(display, 7, 8) && HAS_PCH_SPLIT(i915)) {
+	if (IS_DISPLAY_VER(display, 7, 8) && HAS_PCH_SPLIT(display)) {
 		u32 fuse_strap = intel_de_read(display, FUSE_STRAP);
 		u32 sfuse_strap = intel_de_read(display, SFUSE_STRAP);
 
@@ -1790,7 +1791,7 @@ static void __intel_display_device_info_runtime_init(struct intel_display *displ
 		 */
 		if (fuse_strap & ILK_INTERNAL_DISPLAY_DISABLE ||
 		    sfuse_strap & SFUSE_STRAP_DISPLAY_DISABLED ||
-		    (HAS_PCH_CPT(i915) &&
+		    (HAS_PCH_CPT(display) &&
 		     !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) {
 			drm_info(display->drm,
 				 "Display fused off, disabling\n");
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 717286981687..87c666792c0d 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -143,9 +143,11 @@ struct intel_display_platforms {
 
 #define HAS_4TILE(__display)		((__display)->platform.dg2 || DISPLAY_VER(__display) >= 14)
 #define HAS_ASYNC_FLIPS(__display)	(DISPLAY_VER(__display) >= 5)
+#define HAS_AS_SDP(__display)		(DISPLAY_VER(__display) >= 13)
 #define HAS_BIGJOINER(__display)	(DISPLAY_VER(__display) >= 11 && HAS_DSC(__display))
 #define HAS_CDCLK_CRAWL(__display)	(DISPLAY_INFO(__display)->has_cdclk_crawl)
 #define HAS_CDCLK_SQUASH(__display)	(DISPLAY_INFO(__display)->has_cdclk_squash)
+#define HAS_CMRR(__display)		(DISPLAY_VER(__display) >= 20)
 #define HAS_CMTG(__display)		(!(__display)->platform.dg2 && DISPLAY_VER(__display) >= 13)
 #define HAS_CUR_FBC(__display)		(!HAS_GMCH(__display) && IS_DISPLAY_VER(__display, 7, 13))
 #define HAS_D12_PLANE_MINIMIZATION(__display)	((__display)->platform.rocketlake || (__display)->platform.alderlake_s)
@@ -156,19 +158,21 @@ struct intel_display_platforms {
 #define HAS_DMC_WAKELOCK(__display)	(DISPLAY_VER(__display) >= 20)
 #define HAS_DOUBLE_BUFFERED_M_N(__display)	(DISPLAY_VER(__display) >= 9 || (__display)->platform.broadwell)
 #define HAS_DOUBLE_WIDE(__display)	(DISPLAY_VER(__display) < 4)
-#define HAS_DP_MST(__display)		(DISPLAY_INFO(__display)->has_dp_mst)
 #define HAS_DP20(__display)		((__display)->platform.dg2 || DISPLAY_VER(__display) >= 14)
 #define HAS_DPT(__display)		(DISPLAY_VER(__display) >= 13)
+#define HAS_DP_MST(__display)		(DISPLAY_INFO(__display)->has_dp_mst)
 #define HAS_DSB(__display)		(DISPLAY_INFO(__display)->has_dsb)
 #define HAS_DSC(__display)		(DISPLAY_RUNTIME_INFO(__display)->has_dsc)
+#define HAS_DSC_3ENGINES(__display)	(DISPLAY_VERx100(__display) == 1401 && HAS_DSC(__display))
 #define HAS_DSC_MST(__display)		(DISPLAY_VER(__display) >= 12 && HAS_DSC(__display))
 #define HAS_FBC(__display)		(DISPLAY_RUNTIME_INFO(__display)->fbc_mask != 0)
 #define HAS_FBC_DIRTY_RECT(__display)	(DISPLAY_VER(__display) >= 30)
 #define HAS_FPGA_DBG_UNCLAIMED(__display)	(DISPLAY_INFO(__display)->has_fpga_dbg)
 #define HAS_FW_BLC(__display)		(DISPLAY_VER(__display) >= 3)
-#define HAS_GMBUS_IRQ(__display)	(DISPLAY_VER(__display) >= 4)
 #define HAS_GMBUS_BURST_READ(__display)	(DISPLAY_VER(__display) >= 10 || (__display)->platform.kabylake)
+#define HAS_GMBUS_IRQ(__display)	(DISPLAY_VER(__display) >= 4)
 #define HAS_GMCH(__display)		(DISPLAY_INFO(__display)->has_gmch)
+#define HAS_HOTPLUG(__display)		(DISPLAY_INFO(__display)->has_hotplug)
 #define HAS_HW_SAGV_WM(__display)	(DISPLAY_VER(__display) >= 13 && !(__display)->platform.dgfx)
 #define HAS_IPC(__display)		(DISPLAY_INFO(__display)->has_ipc)
 #define HAS_IPS(__display)		((__display)->platform.haswell_ult || (__display)->platform.broadwell)
@@ -189,10 +193,7 @@ struct intel_display_platforms {
 					  ((__display)->platform.dgfx && DISPLAY_VER(__display) == 14)) && \
 					 HAS_DSC(__display))
 #define HAS_VRR(__display)		(DISPLAY_VER(__display) >= 11)
-#define HAS_AS_SDP(__display)		(DISPLAY_VER(__display) >= 13)
-#define HAS_CMRR(__display)		(DISPLAY_VER(__display) >= 20)
 #define INTEL_NUM_PIPES(__display)	(hweight8(DISPLAY_RUNTIME_INFO(__display)->pipe_mask))
-#define I915_HAS_HOTPLUG(__display)	(DISPLAY_INFO(__display)->has_hotplug)
 #define OVERLAY_NEEDS_PHYSICAL(__display)	(DISPLAY_INFO(__display)->overlay_needs_physical)
 #define SUPPORTS_TV(__display)		(DISPLAY_INFO(__display)->supports_tv)
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index 31740a677dd8..5c74ab5fd1aa 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -54,6 +54,7 @@
 #include "intel_plane_initial.h"
 #include "intel_pmdemand.h"
 #include "intel_pps.h"
+#include "intel_psr.h"
 #include "intel_quirks.h"
 #include "intel_vga.h"
 #include "intel_wm.h"
@@ -82,7 +83,6 @@ bool intel_display_driver_probe_defer(struct pci_dev *pdev)
 
 void intel_display_driver_init_hw(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_cdclk_state *cdclk_state;
 
 	if (!HAS_DISPLAY(display))
@@ -94,7 +94,7 @@ void intel_display_driver_init_hw(struct intel_display *display)
 	intel_cdclk_dump_config(display, &display->cdclk.hw, "Current CDCLK");
 	cdclk_state->logical = cdclk_state->actual = display->cdclk.hw;
 
-	intel_display_wa_apply(i915);
+	intel_display_wa_apply(display);
 }
 
 static const struct drm_mode_config_funcs intel_mode_funcs = {
@@ -181,7 +181,8 @@ static void intel_plane_possible_crtcs_init(struct intel_display *display)
 
 void intel_display_driver_early_probe(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
+	/* This must be called before any calls to HAS_PCH_* */
+	intel_pch_detect(display);
 
 	if (!HAS_DISPLAY(display))
 		return;
@@ -193,12 +194,12 @@ void intel_display_driver_early_probe(struct intel_display *display)
 	mutex_init(&display->pps.mutex);
 	mutex_init(&display->hdcp.hdcp_mutex);
 
-	intel_display_irq_init(i915);
+	intel_display_irq_init(display);
 	intel_dkl_phy_init(display);
 	intel_color_init_hooks(display);
 	intel_init_cdclk_hooks(display);
 	intel_audio_hooks_init(display);
-	intel_dpll_init_clock_hook(i915);
+	intel_dpll_init_clock_hook(display);
 	intel_init_display_hooks(display);
 	intel_fdi_init_hook(display);
 	intel_dmc_wl_init(display);
@@ -226,6 +227,8 @@ int intel_display_driver_probe_noirq(struct intel_display *display)
 	if (ret)
 		goto cleanup_bios;
 
+	intel_psr_dc5_dc6_wa_init(display);
+
 	/* FIXME: completely on the wrong abstraction layer */
 	ret = intel_power_domains_init(display);
 	if (ret < 0)
@@ -255,11 +258,11 @@ int intel_display_driver_probe_noirq(struct intel_display *display)
 	if (ret)
 		goto cleanup_vga_client_pw_domain_dmc;
 
-	ret = intel_dbuf_init(i915);
+	ret = intel_dbuf_init(display);
 	if (ret)
 		goto cleanup_vga_client_pw_domain_dmc;
 
-	ret = intel_bw_init(i915);
+	ret = intel_bw_init(display);
 	if (ret)
 		goto cleanup_vga_client_pw_domain_dmc;
 
@@ -315,11 +318,9 @@ static void set_display_access(struct intel_display *display,
  */
 void intel_display_driver_enable_user_access(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	set_display_access(display, true, NULL);
 
-	intel_hpd_enable_detection_work(i915);
+	intel_hpd_enable_detection_work(display);
 }
 
 /**
@@ -341,9 +342,7 @@ void intel_display_driver_enable_user_access(struct intel_display *display)
  */
 void intel_display_driver_disable_user_access(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	intel_hpd_disable_detection_work(i915);
+	intel_hpd_disable_detection_work(display);
 
 	set_display_access(display, false, current);
 }
@@ -422,14 +421,13 @@ bool intel_display_driver_check_access(struct intel_display *display)
 /* part #2: call after irq install, but before gem init */
 int intel_display_driver_probe_nogem(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	enum pipe pipe;
 	int ret;
 
 	if (!HAS_DISPLAY(display))
 		return 0;
 
-	intel_wm_init(i915);
+	intel_wm_init(display);
 
 	intel_panel_sanitize_ssc(display);
 
@@ -460,8 +458,6 @@ int intel_display_driver_probe_nogem(struct intel_display *display)
 
 	intel_hti_init(display);
 
-	/* Just disable it once at startup */
-	intel_vga_disable(display);
 	intel_setup_outputs(display);
 
 	ret = intel_dp_tunnel_mgr_init(display);
@@ -471,7 +467,7 @@ int intel_display_driver_probe_nogem(struct intel_display *display)
 	intel_display_driver_disable_user_access(display);
 
 	drm_modeset_lock_all(display->drm);
-	intel_modeset_setup_hw_state(i915, display->drm->mode_config.acquire_ctx);
+	intel_modeset_setup_hw_state(display, display->drm->mode_config.acquire_ctx);
 	intel_acpi_assign_connector_fwnodes(display);
 	drm_modeset_unlock_all(display->drm);
 
@@ -483,7 +479,7 @@ int intel_display_driver_probe_nogem(struct intel_display *display)
 	 * since the watermark calculation done here will use pstate->fb.
 	 */
 	if (!HAS_GMCH(display))
-		ilk_wm_sanitize(i915);
+		ilk_wm_sanitize(display);
 
 	return 0;
 
@@ -498,7 +494,6 @@ err_mode_config:
 /* part #3: call after gem init */
 int intel_display_driver_probe(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	int ret;
 
 	if (!HAS_DISPLAY(display))
@@ -524,16 +519,15 @@ int intel_display_driver_probe(struct intel_display *display)
 	intel_overlay_setup(display);
 
 	/* Only enable hotplug handling once the fbdev is fully set up. */
-	intel_hpd_init(i915);
+	intel_hpd_init(display);
 
-	skl_watermark_ipc_init(i915);
+	skl_watermark_ipc_init(display);
 
 	return 0;
 }
 
 void intel_display_driver_register(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct drm_printer p = drm_dbg_printer(display->drm, DRM_UT_KMS,
 					       "i915 display info:");
 
@@ -558,9 +552,9 @@ void intel_display_driver_register(struct intel_display *display)
 	 * fbdev->async_cookie.
 	 */
 	drm_kms_helper_poll_init(display->drm);
-	intel_hpd_poll_disable(i915);
+	intel_hpd_poll_disable(display);
 
-	intel_fbdev_setup(i915);
+	intel_fbdev_setup(display);
 
 	intel_display_device_info_print(DISPLAY_INFO(display),
 					DISPLAY_RUNTIME_INFO(display), &p);
@@ -600,7 +594,7 @@ void intel_display_driver_remove_noirq(struct intel_display *display)
 	 * Due to the hpd irq storm handling the hotplug work can re-arm the
 	 * poll handlers. Hence disable polling after hpd handling is shut down.
 	 */
-	intel_hpd_poll_fini(i915);
+	intel_hpd_poll_fini(display);
 
 	intel_unregister_dsm_handler();
 
@@ -695,13 +689,11 @@ __intel_display_driver_resume(struct intel_display *display,
 			      struct drm_atomic_state *state,
 			      struct drm_modeset_acquire_ctx *ctx)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct drm_crtc_state *crtc_state;
 	struct drm_crtc *crtc;
 	int ret, i;
 
-	intel_modeset_setup_hw_state(i915, ctx);
-	intel_vga_redisable(display);
+	intel_modeset_setup_hw_state(display, ctx);
 
 	if (!state)
 		return 0;
@@ -733,7 +725,6 @@ __intel_display_driver_resume(struct intel_display *display,
 
 void intel_display_driver_resume(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct drm_atomic_state *state = display->restore.modeset_state;
 	struct drm_modeset_acquire_ctx ctx;
 	int ret;
@@ -761,7 +752,7 @@ void intel_display_driver_resume(struct intel_display *display)
 	if (!ret)
 		ret = __intel_display_driver_resume(display, state, &ctx);
 
-	skl_watermark_ipc_update(i915);
+	skl_watermark_ipc_update(display);
 	drm_modeset_drop_locks(&ctx);
 	drm_modeset_acquire_fini(&ctx);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index aa23bb817805..3e73832e5e81 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -5,7 +5,6 @@
 
 #include <drm/drm_vblank.h>
 
-#include "gt/intel_rps.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
@@ -14,6 +13,8 @@
 #include "intel_crtc.h"
 #include "intel_de.h"
 #include "intel_display_irq.h"
+#include "intel_display_rpm.h"
+#include "intel_display_rps.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
 #include "intel_dmc_wl.h"
@@ -115,9 +116,8 @@ static void intel_pipe_fault_irq_handler(struct intel_display *display,
 }
 
 static void
-intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe)
+intel_handle_vblank(struct intel_display *display, enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_crtc *crtc = intel_crtc_for_pipe(display, pipe);
 
 	drm_crtc_handle_vblank(&crtc->base);
@@ -125,59 +125,59 @@ intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe)
 
 /**
  * ilk_update_display_irq - update DEIMR
- * @dev_priv: driver private
+ * @display: display device
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-void ilk_update_display_irq(struct drm_i915_private *dev_priv,
+void ilk_update_display_irq(struct intel_display *display,
 			    u32 interrupt_mask, u32 enabled_irq_mask)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 new_val;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+	lockdep_assert_held(&display->irq.lock);
+	drm_WARN_ON(display->drm, enabled_irq_mask & ~interrupt_mask);
 
 	new_val = dev_priv->irq_mask;
 	new_val &= ~interrupt_mask;
 	new_val |= (~enabled_irq_mask & interrupt_mask);
 
 	if (new_val != dev_priv->irq_mask &&
-	    !drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) {
+	    !drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv))) {
 		dev_priv->irq_mask = new_val;
 		intel_de_write(display, DEIMR, dev_priv->irq_mask);
 		intel_de_posting_read(display, DEIMR);
 	}
 }
 
-void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits)
+void ilk_enable_display_irq(struct intel_display *display, u32 bits)
 {
-	ilk_update_display_irq(i915, bits, bits);
+	ilk_update_display_irq(display, bits, bits);
 }
 
-void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits)
+void ilk_disable_display_irq(struct intel_display *display, u32 bits)
 {
-	ilk_update_display_irq(i915, bits, 0);
+	ilk_update_display_irq(display, bits, 0);
 }
 
 /**
  * bdw_update_port_irq - update DE port interrupt
- * @dev_priv: driver private
+ * @display: display device
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-void bdw_update_port_irq(struct drm_i915_private *dev_priv,
+void bdw_update_port_irq(struct intel_display *display,
 			 u32 interrupt_mask, u32 enabled_irq_mask)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 new_val;
 	u32 old_val;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+	drm_WARN_ON(display->drm, enabled_irq_mask & ~interrupt_mask);
 
-	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
+	if (drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv)))
 		return;
 
 	old_val = intel_de_read(display, GEN8_DE_PORT_IMR);
@@ -194,93 +194,92 @@ void bdw_update_port_irq(struct drm_i915_private *dev_priv,
 
 /**
  * bdw_update_pipe_irq - update DE pipe interrupt
- * @dev_priv: driver private
+ * @display: display device
  * @pipe: pipe whose interrupt to update
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-static void bdw_update_pipe_irq(struct drm_i915_private *dev_priv,
+static void bdw_update_pipe_irq(struct intel_display *display,
 				enum pipe pipe, u32 interrupt_mask,
 				u32 enabled_irq_mask)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 new_val;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+	drm_WARN_ON(display->drm, enabled_irq_mask & ~interrupt_mask);
 
-	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
+	if (drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv)))
 		return;
 
-	new_val = dev_priv->display.irq.de_irq_mask[pipe];
+	new_val = display->irq.de_irq_mask[pipe];
 	new_val &= ~interrupt_mask;
 	new_val |= (~enabled_irq_mask & interrupt_mask);
 
-	if (new_val != dev_priv->display.irq.de_irq_mask[pipe]) {
-		dev_priv->display.irq.de_irq_mask[pipe] = new_val;
+	if (new_val != display->irq.de_irq_mask[pipe]) {
+		display->irq.de_irq_mask[pipe] = new_val;
 		intel_de_write(display, GEN8_DE_PIPE_IMR(pipe), display->irq.de_irq_mask[pipe]);
 		intel_de_posting_read(display, GEN8_DE_PIPE_IMR(pipe));
 	}
 }
 
-void bdw_enable_pipe_irq(struct drm_i915_private *i915,
+void bdw_enable_pipe_irq(struct intel_display *display,
 			 enum pipe pipe, u32 bits)
 {
-	bdw_update_pipe_irq(i915, pipe, bits, bits);
+	bdw_update_pipe_irq(display, pipe, bits, bits);
 }
 
-void bdw_disable_pipe_irq(struct drm_i915_private *i915,
+void bdw_disable_pipe_irq(struct intel_display *display,
 			  enum pipe pipe, u32 bits)
 {
-	bdw_update_pipe_irq(i915, pipe, bits, 0);
+	bdw_update_pipe_irq(display, pipe, bits, 0);
 }
 
 /**
  * ibx_display_interrupt_update - update SDEIMR
- * @dev_priv: driver private
+ * @display: display device
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
+void ibx_display_interrupt_update(struct intel_display *display,
 				  u32 interrupt_mask,
 				  u32 enabled_irq_mask)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 sdeimr = intel_de_read(display, SDEIMR);
 
 	sdeimr &= ~interrupt_mask;
 	sdeimr |= (~enabled_irq_mask & interrupt_mask);
 
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+	drm_WARN_ON(display->drm, enabled_irq_mask & ~interrupt_mask);
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
-	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
+	if (drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv)))
 		return;
 
 	intel_de_write(display, SDEIMR, sdeimr);
 	intel_de_posting_read(display, SDEIMR);
 }
 
-void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits)
+void ibx_enable_display_interrupt(struct intel_display *display, u32 bits)
 {
-	ibx_display_interrupt_update(i915, bits, bits);
+	ibx_display_interrupt_update(display, bits, bits);
 }
 
-void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits)
+void ibx_disable_display_interrupt(struct intel_display *display, u32 bits)
 {
-	ibx_display_interrupt_update(i915, bits, 0);
+	ibx_display_interrupt_update(display, bits, 0);
 }
 
 u32 i915_pipestat_enable_mask(struct intel_display *display,
 			      enum pipe pipe)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 status_mask = display->irq.pipestat_irq_mask[pipe];
 	u32 enable_mask = status_mask << 16;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	if (DISPLAY_VER(display) < 5)
 		goto out;
@@ -318,48 +317,48 @@ out:
 	return enable_mask;
 }
 
-void i915_enable_pipestat(struct drm_i915_private *dev_priv,
+void i915_enable_pipestat(struct intel_display *display,
 			  enum pipe pipe, u32 status_mask)
 {
-	struct intel_display *display = &dev_priv->display;
-	i915_reg_t reg = PIPESTAT(dev_priv, pipe);
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	i915_reg_t reg = PIPESTAT(display, pipe);
 	u32 enable_mask;
 
-	drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
+	drm_WARN_ONCE(display->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
 		      "pipe %c: status_mask=0x%x\n",
 		      pipe_name(pipe), status_mask);
 
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv));
+	lockdep_assert_held(&display->irq.lock);
+	drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv));
 
-	if ((dev_priv->display.irq.pipestat_irq_mask[pipe] & status_mask) == status_mask)
+	if ((display->irq.pipestat_irq_mask[pipe] & status_mask) == status_mask)
 		return;
 
-	dev_priv->display.irq.pipestat_irq_mask[pipe] |= status_mask;
+	display->irq.pipestat_irq_mask[pipe] |= status_mask;
 	enable_mask = i915_pipestat_enable_mask(display, pipe);
 
 	intel_de_write(display, reg, enable_mask | status_mask);
 	intel_de_posting_read(display, reg);
 }
 
-void i915_disable_pipestat(struct drm_i915_private *dev_priv,
+void i915_disable_pipestat(struct intel_display *display,
 			   enum pipe pipe, u32 status_mask)
 {
-	struct intel_display *display = &dev_priv->display;
-	i915_reg_t reg = PIPESTAT(dev_priv, pipe);
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	i915_reg_t reg = PIPESTAT(display, pipe);
 	u32 enable_mask;
 
-	drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
+	drm_WARN_ONCE(display->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
 		      "pipe %c: status_mask=0x%x\n",
 		      pipe_name(pipe), status_mask);
 
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv));
+	lockdep_assert_held(&display->irq.lock);
+	drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv));
 
-	if ((dev_priv->display.irq.pipestat_irq_mask[pipe] & status_mask) == 0)
+	if ((display->irq.pipestat_irq_mask[pipe] & status_mask) == 0)
 		return;
 
-	dev_priv->display.irq.pipestat_irq_mask[pipe] &= ~status_mask;
+	display->irq.pipestat_irq_mask[pipe] &= ~status_mask;
 	enable_mask = i915_pipestat_enable_mask(display, pipe);
 
 	intel_de_write(display, reg, enable_mask | status_mask);
@@ -368,49 +367,41 @@ void i915_disable_pipestat(struct drm_i915_private *dev_priv,
 
 static bool i915_has_legacy_blc_interrupt(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	if (IS_I85X(i915))
+	if (display->platform.i85x)
 		return true;
 
-	if (IS_PINEVIEW(i915))
+	if (display->platform.pineview)
 		return true;
 
-	return IS_DISPLAY_VER(display, 3, 4) && IS_MOBILE(i915);
+	return IS_DISPLAY_VER(display, 3, 4) && display->platform.mobile;
 }
 
-/**
- * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion
- * @dev_priv: i915 device private
- */
-void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv)
+/* enable ASLE pipestat for OpRegion */
+static void i915_enable_asle_pipestat(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
-
 	if (!intel_opregion_asle_present(display))
 		return;
 
 	if (!i915_has_legacy_blc_interrupt(display))
 		return;
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
-	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_LEGACY_BLC_EVENT_STATUS);
-	if (DISPLAY_VER(dev_priv) >= 4)
-		i915_enable_pipestat(dev_priv, PIPE_A,
+	i915_enable_pipestat(display, PIPE_B, PIPE_LEGACY_BLC_EVENT_STATUS);
+	if (DISPLAY_VER(display) >= 4)
+		i915_enable_pipestat(display, PIPE_A,
 				     PIPE_LEGACY_BLC_EVENT_STATUS);
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 #if IS_ENABLED(CONFIG_DEBUG_FS)
-static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+static void display_pipe_crc_irq_handler(struct intel_display *display,
 					 enum pipe pipe,
 					 u32 crc0, u32 crc1,
 					 u32 crc2, u32 crc3,
 					 u32 crc4)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_crtc *crtc = intel_crtc_for_pipe(display, pipe);
 	struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc;
 	u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 };
@@ -427,7 +418,7 @@ static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 	 * don't trust that one either.
 	 */
 	if (pipe_crc->skipped <= 0 ||
-	    (DISPLAY_VER(dev_priv) >= 8 && pipe_crc->skipped == 1)) {
+	    (DISPLAY_VER(display) >= 8 && pipe_crc->skipped == 1)) {
 		pipe_crc->skipped++;
 		spin_unlock(&pipe_crc->lock);
 		return;
@@ -440,20 +431,19 @@ static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 }
 #else
 static inline void
-display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+display_pipe_crc_irq_handler(struct intel_display *display,
 			     enum pipe pipe,
 			     u32 crc0, u32 crc1,
 			     u32 crc2, u32 crc3,
 			     u32 crc4) {}
 #endif
 
-static void flip_done_handler(struct drm_i915_private *i915,
+static void flip_done_handler(struct intel_display *display,
 			      enum pipe pipe)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_crtc *crtc = intel_crtc_for_pipe(display, pipe);
 
-	spin_lock(&i915->drm.event_lock);
+	spin_lock(&display->drm->event_lock);
 
 	if (crtc->flip_done_event) {
 		trace_intel_crtc_flip_done(crtc);
@@ -461,25 +451,21 @@ static void flip_done_handler(struct drm_i915_private *i915,
 		crtc->flip_done_event = NULL;
 	}
 
-	spin_unlock(&i915->drm.event_lock);
+	spin_unlock(&display->drm->event_lock);
 }
 
-static void hsw_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+static void hsw_pipe_crc_irq_handler(struct intel_display *display,
 				     enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
-
-	display_pipe_crc_irq_handler(dev_priv, pipe,
+	display_pipe_crc_irq_handler(display, pipe,
 				     intel_de_read(display, PIPE_CRC_RES_HSW(pipe)),
 				     0, 0, 0, 0);
 }
 
-static void ivb_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+static void ivb_pipe_crc_irq_handler(struct intel_display *display,
 				     enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
-
-	display_pipe_crc_irq_handler(dev_priv, pipe,
+	display_pipe_crc_irq_handler(display, pipe,
 				     intel_de_read(display, PIPE_CRC_RES_1_IVB(pipe)),
 				     intel_de_read(display, PIPE_CRC_RES_2_IVB(pipe)),
 				     intel_de_read(display, PIPE_CRC_RES_3_IVB(pipe)),
@@ -487,58 +473,55 @@ static void ivb_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 				     intel_de_read(display, PIPE_CRC_RES_5_IVB(pipe)));
 }
 
-static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+static void i9xx_pipe_crc_irq_handler(struct intel_display *display,
 				      enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 res1, res2;
 
-	if (DISPLAY_VER(dev_priv) >= 3)
-		res1 = intel_de_read(display, PIPE_CRC_RES_RES1_I915(dev_priv, pipe));
+	if (DISPLAY_VER(display) >= 3)
+		res1 = intel_de_read(display, PIPE_CRC_RES_RES1_I915(display, pipe));
 	else
 		res1 = 0;
 
-	if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv))
-		res2 = intel_de_read(display, PIPE_CRC_RES_RES2_G4X(dev_priv, pipe));
+	if (DISPLAY_VER(display) >= 5 || display->platform.g4x)
+		res2 = intel_de_read(display, PIPE_CRC_RES_RES2_G4X(display, pipe));
 	else
 		res2 = 0;
 
-	display_pipe_crc_irq_handler(dev_priv, pipe,
-				     intel_de_read(display, PIPE_CRC_RES_RED(dev_priv, pipe)),
-				     intel_de_read(display, PIPE_CRC_RES_GREEN(dev_priv, pipe)),
-				     intel_de_read(display, PIPE_CRC_RES_BLUE(dev_priv, pipe)),
+	display_pipe_crc_irq_handler(display, pipe,
+				     intel_de_read(display, PIPE_CRC_RES_RED(display, pipe)),
+				     intel_de_read(display, PIPE_CRC_RES_GREEN(display, pipe)),
+				     intel_de_read(display, PIPE_CRC_RES_BLUE(display, pipe)),
 				     res1, res2);
 }
 
-static void i9xx_pipestat_irq_reset(struct drm_i915_private *dev_priv)
+static void i9xx_pipestat_irq_reset(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		intel_de_write(display,
-			       PIPESTAT(dev_priv, pipe),
+			       PIPESTAT(display, pipe),
 			       PIPESTAT_INT_STATUS_MASK | PIPE_FIFO_UNDERRUN_STATUS);
 
-		dev_priv->display.irq.pipestat_irq_mask[pipe] = 0;
+		display->irq.pipestat_irq_mask[pipe] = 0;
 	}
 }
 
-void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv,
+void i9xx_pipestat_irq_ack(struct intel_display *display,
 			   u32 iir, u32 pipe_stats[I915_MAX_PIPES])
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 
-	spin_lock(&dev_priv->irq_lock);
+	spin_lock(&display->irq.lock);
 
-	if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
-	    !dev_priv->display.irq.vlv_display_irqs_enabled) {
-		spin_unlock(&dev_priv->irq_lock);
+	if ((display->platform.valleyview || display->platform.cherryview) &&
+	    !display->irq.vlv_display_irqs_enabled) {
+		spin_unlock(&display->irq.lock);
 		return;
 	}
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		i915_reg_t reg;
 		u32 status_mask, enable_mask, iir_bit = 0;
 
@@ -566,12 +549,12 @@ void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv,
 			break;
 		}
 		if (iir & iir_bit)
-			status_mask |= dev_priv->display.irq.pipestat_irq_mask[pipe];
+			status_mask |= display->irq.pipestat_irq_mask[pipe];
 
 		if (!status_mask)
 			continue;
 
-		reg = PIPESTAT(dev_priv, pipe);
+		reg = PIPESTAT(display, pipe);
 		pipe_stats[pipe] = intel_de_read(display, reg) & status_mask;
 		enable_mask = i915_pipestat_enable_mask(display, pipe);
 
@@ -589,25 +572,24 @@ void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv,
 			intel_de_write(display, reg, enable_mask);
 		}
 	}
-	spin_unlock(&dev_priv->irq_lock);
+	spin_unlock(&display->irq.lock);
 }
 
-void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+void i915_pipestat_irq_handler(struct intel_display *display,
 			       u32 iir, u32 pipe_stats[I915_MAX_PIPES])
 {
-	struct intel_display *display = &dev_priv->display;
 	bool blc_event = false;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
+			intel_handle_vblank(display, pipe);
 
 		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
 			blc_event = true;
 
 		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+			i9xx_pipe_crc_irq_handler(display, pipe);
 
 		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
 			intel_cpu_fifo_underrun_irq_handler(display, pipe);
@@ -617,22 +599,21 @@ void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 		intel_opregion_asle_intr(display);
 }
 
-void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+void i965_pipestat_irq_handler(struct intel_display *display,
 			       u32 iir, u32 pipe_stats[I915_MAX_PIPES])
 {
-	struct intel_display *display = &dev_priv->display;
 	bool blc_event = false;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
+			intel_handle_vblank(display, pipe);
 
 		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
 			blc_event = true;
 
 		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+			i9xx_pipe_crc_irq_handler(display, pipe);
 
 		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
 			intel_cpu_fifo_underrun_irq_handler(display, pipe);
@@ -645,21 +626,20 @@ void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 		intel_gmbus_irq_handler(display);
 }
 
-void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+void valleyview_pipestat_irq_handler(struct intel_display *display,
 				     u32 pipe_stats[I915_MAX_PIPES])
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
+			intel_handle_vblank(display, pipe);
 
 		if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV)
-			flip_done_handler(dev_priv, pipe);
+			flip_done_handler(display, pipe);
 
 		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+			i9xx_pipe_crc_irq_handler(display, pipe);
 
 		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
 			intel_cpu_fifo_underrun_irq_handler(display, pipe);
@@ -669,18 +649,17 @@ void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 		intel_gmbus_irq_handler(display);
 }
 
-static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+static void ibx_irq_handler(struct intel_display *display, u32 pch_iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK;
 
-	ibx_hpd_irq_handler(dev_priv, hotplug_trigger);
+	ibx_hpd_irq_handler(display, hotplug_trigger);
 
 	if (pch_iir & SDE_AUDIO_POWER_MASK) {
 		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >>
 			       SDE_AUDIO_POWER_SHIFT);
-		drm_dbg(&dev_priv->drm, "PCH audio power change on port %d\n",
+		drm_dbg(display->drm, "PCH audio power change on port %d\n",
 			port_name(port));
 	}
 
@@ -691,26 +670,26 @@ static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		intel_gmbus_irq_handler(display);
 
 	if (pch_iir & SDE_AUDIO_HDCP_MASK)
-		drm_dbg(&dev_priv->drm, "PCH HDCP audio interrupt\n");
+		drm_dbg(display->drm, "PCH HDCP audio interrupt\n");
 
 	if (pch_iir & SDE_AUDIO_TRANS_MASK)
-		drm_dbg(&dev_priv->drm, "PCH transcoder audio interrupt\n");
+		drm_dbg(display->drm, "PCH transcoder audio interrupt\n");
 
 	if (pch_iir & SDE_POISON)
-		drm_err(&dev_priv->drm, "PCH poison interrupt\n");
+		drm_err(display->drm, "PCH poison interrupt\n");
 
 	if (pch_iir & SDE_FDI_MASK) {
-		for_each_pipe(dev_priv, pipe)
-			drm_dbg(&dev_priv->drm, "  pipe %c FDI IIR: 0x%08x\n",
+		for_each_pipe(display, pipe)
+			drm_dbg(display->drm, "  pipe %c FDI IIR: 0x%08x\n",
 				pipe_name(pipe),
 				intel_de_read(display, FDI_RX_IIR(pipe)));
 	}
 
 	if (pch_iir & (SDE_TRANSB_CRC_DONE | SDE_TRANSA_CRC_DONE))
-		drm_dbg(&dev_priv->drm, "PCH transcoder CRC done interrupt\n");
+		drm_dbg(display->drm, "PCH transcoder CRC done interrupt\n");
 
 	if (pch_iir & (SDE_TRANSB_CRC_ERR | SDE_TRANSA_CRC_ERR))
-		drm_dbg(&dev_priv->drm,
+		drm_dbg(display->drm,
 			"PCH transcoder CRC error interrupt\n");
 
 	if (pch_iir & SDE_TRANSA_FIFO_UNDER)
@@ -753,14 +732,13 @@ static const struct pipe_fault_handler ivb_pipe_fault_handlers[] = {
 	{}
 };
 
-static void ivb_err_int_handler(struct drm_i915_private *dev_priv)
+static void ivb_err_int_handler(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 err_int = intel_de_read(display, GEN7_ERR_INT);
 	enum pipe pipe;
 
 	if (err_int & ERR_INT_POISON)
-		drm_err(&dev_priv->drm, "Poison interrupt\n");
+		drm_err(display->drm, "Poison interrupt\n");
 
 	if (err_int & ERR_INT_INVALID_GTT_PTE)
 		drm_err_ratelimited(display->drm, "Invalid GTT PTE\n");
@@ -768,17 +746,17 @@ static void ivb_err_int_handler(struct drm_i915_private *dev_priv)
 	if (err_int & ERR_INT_INVALID_PTE_DATA)
 		drm_err_ratelimited(display->drm, "Invalid PTE data\n");
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		u32 fault_errors;
 
 		if (err_int & ERR_INT_FIFO_UNDERRUN(pipe))
 			intel_cpu_fifo_underrun_irq_handler(display, pipe);
 
 		if (err_int & ERR_INT_PIPE_CRC_DONE(pipe)) {
-			if (IS_IVYBRIDGE(dev_priv))
-				ivb_pipe_crc_irq_handler(dev_priv, pipe);
+			if (display->platform.ivybridge)
+				ivb_pipe_crc_irq_handler(display, pipe);
 			else
-				hsw_pipe_crc_irq_handler(dev_priv, pipe);
+				hsw_pipe_crc_irq_handler(display, pipe);
 		}
 
 		fault_errors = err_int & ivb_err_int_pipe_fault_mask(pipe);
@@ -790,34 +768,32 @@ static void ivb_err_int_handler(struct drm_i915_private *dev_priv)
 	intel_de_write(display, GEN7_ERR_INT, err_int);
 }
 
-static void cpt_serr_int_handler(struct drm_i915_private *dev_priv)
+static void cpt_serr_int_handler(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 serr_int = intel_de_read(display, SERR_INT);
 	enum pipe pipe;
 
 	if (serr_int & SERR_INT_POISON)
-		drm_err(&dev_priv->drm, "PCH poison interrupt\n");
+		drm_err(display->drm, "PCH poison interrupt\n");
 
-	for_each_pipe(dev_priv, pipe)
+	for_each_pipe(display, pipe)
 		if (serr_int & SERR_INT_TRANS_FIFO_UNDERRUN(pipe))
 			intel_pch_fifo_underrun_irq_handler(display, pipe);
 
 	intel_de_write(display, SERR_INT, serr_int);
 }
 
-static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+static void cpt_irq_handler(struct intel_display *display, u32 pch_iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
 
-	ibx_hpd_irq_handler(dev_priv, hotplug_trigger);
+	ibx_hpd_irq_handler(display, hotplug_trigger);
 
 	if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) {
 		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >>
 			       SDE_AUDIO_POWER_SHIFT_CPT);
-		drm_dbg(&dev_priv->drm, "PCH audio power change on port %c\n",
+		drm_dbg(display->drm, "PCH audio power change on port %c\n",
 			port_name(port));
 	}
 
@@ -828,20 +804,20 @@ static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		intel_gmbus_irq_handler(display);
 
 	if (pch_iir & SDE_AUDIO_CP_REQ_CPT)
-		drm_dbg(&dev_priv->drm, "Audio CP request interrupt\n");
+		drm_dbg(display->drm, "Audio CP request interrupt\n");
 
 	if (pch_iir & SDE_AUDIO_CP_CHG_CPT)
-		drm_dbg(&dev_priv->drm, "Audio CP change interrupt\n");
+		drm_dbg(display->drm, "Audio CP change interrupt\n");
 
 	if (pch_iir & SDE_FDI_MASK_CPT) {
-		for_each_pipe(dev_priv, pipe)
-			drm_dbg(&dev_priv->drm, "  pipe %c FDI IIR: 0x%08x\n",
+		for_each_pipe(display, pipe)
+			drm_dbg(display->drm, "  pipe %c FDI IIR: 0x%08x\n",
 				pipe_name(pipe),
 				intel_de_read(display, FDI_RX_IIR(pipe)));
 	}
 
 	if (pch_iir & SDE_ERROR_CPT)
-		cpt_serr_int_handler(dev_priv);
+		cpt_serr_int_handler(display);
 }
 
 static u32 ilk_gtt_fault_pipe_fault_mask(enum pipe pipe)
@@ -894,14 +870,13 @@ static void ilk_gtt_fault_irq_handler(struct intel_display *display)
 	}
 }
 
-void ilk_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir)
+void ilk_display_irq_handler(struct intel_display *display, u32 de_iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG;
 
 	if (hotplug_trigger)
-		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
+		ilk_hpd_irq_handler(display, hotplug_trigger);
 
 	if (de_iir & DE_AUX_CHANNEL_A)
 		intel_dp_aux_irq_handler(display);
@@ -910,58 +885,57 @@ void ilk_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir)
 		intel_opregion_asle_intr(display);
 
 	if (de_iir & DE_POISON)
-		drm_err(&dev_priv->drm, "Poison interrupt\n");
+		drm_err(display->drm, "Poison interrupt\n");
 
 	if (de_iir & DE_GTT_FAULT)
 		ilk_gtt_fault_irq_handler(display);
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		if (de_iir & DE_PIPE_VBLANK(pipe))
-			intel_handle_vblank(dev_priv, pipe);
+			intel_handle_vblank(display, pipe);
 
 		if (de_iir & DE_PLANE_FLIP_DONE(pipe))
-			flip_done_handler(dev_priv, pipe);
+			flip_done_handler(display, pipe);
 
 		if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe))
 			intel_cpu_fifo_underrun_irq_handler(display, pipe);
 
 		if (de_iir & DE_PIPE_CRC_DONE(pipe))
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+			i9xx_pipe_crc_irq_handler(display, pipe);
 	}
 
 	/* check event from PCH */
 	if (de_iir & DE_PCH_EVENT) {
 		u32 pch_iir = intel_de_read(display, SDEIIR);
 
-		if (HAS_PCH_CPT(dev_priv))
-			cpt_irq_handler(dev_priv, pch_iir);
+		if (HAS_PCH_CPT(display))
+			cpt_irq_handler(display, pch_iir);
 		else
-			ibx_irq_handler(dev_priv, pch_iir);
+			ibx_irq_handler(display, pch_iir);
 
 		/* should clear PCH hotplug event before clear CPU irq */
 		intel_de_write(display, SDEIIR, pch_iir);
 	}
 
-	if (DISPLAY_VER(dev_priv) == 5 && de_iir & DE_PCU_EVENT)
-		gen5_rps_irq_handler(&to_gt(dev_priv)->rps);
+	if (DISPLAY_VER(display) == 5 && de_iir & DE_PCU_EVENT)
+		ilk_display_rps_irq_handler(display);
 }
 
-void ivb_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir)
+void ivb_display_irq_handler(struct intel_display *display, u32 de_iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB;
 
 	if (hotplug_trigger)
-		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
+		ilk_hpd_irq_handler(display, hotplug_trigger);
 
 	if (de_iir & DE_ERR_INT_IVB)
-		ivb_err_int_handler(dev_priv);
+		ivb_err_int_handler(display);
 
 	if (de_iir & DE_EDP_PSR_INT_HSW) {
 		struct intel_encoder *encoder;
 
-		for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) {
+		for_each_intel_encoder_with_psr(display->drm, encoder) {
 			struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 			u32 psr_iir;
 
@@ -977,35 +951,35 @@ void ivb_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir)
 	if (de_iir & DE_GSE_IVB)
 		intel_opregion_asle_intr(display);
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		if (de_iir & DE_PIPE_VBLANK_IVB(pipe))
-			intel_handle_vblank(dev_priv, pipe);
+			intel_handle_vblank(display, pipe);
 
 		if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe))
-			flip_done_handler(dev_priv, pipe);
+			flip_done_handler(display, pipe);
 	}
 
 	/* check event from PCH */
-	if (!HAS_PCH_NOP(dev_priv) && (de_iir & DE_PCH_EVENT_IVB)) {
+	if (!HAS_PCH_NOP(display) && (de_iir & DE_PCH_EVENT_IVB)) {
 		u32 pch_iir = intel_de_read(display, SDEIIR);
 
-		cpt_irq_handler(dev_priv, pch_iir);
+		cpt_irq_handler(display, pch_iir);
 
 		/* clear PCH hotplug event before clear CPU irq */
 		intel_de_write(display, SDEIIR, pch_iir);
 	}
 }
 
-static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
+static u32 gen8_de_port_aux_mask(struct intel_display *display)
 {
 	u32 mask;
 
-	if (DISPLAY_VER(dev_priv) >= 20)
+	if (DISPLAY_VER(display) >= 20)
 		return 0;
-	else if (DISPLAY_VER(dev_priv) >= 14)
+	else if (DISPLAY_VER(display) >= 14)
 		return TGL_DE_PORT_AUX_DDIA |
 			TGL_DE_PORT_AUX_DDIB;
-	else if (DISPLAY_VER(dev_priv) >= 13)
+	else if (DISPLAY_VER(display) >= 13)
 		return TGL_DE_PORT_AUX_DDIA |
 			TGL_DE_PORT_AUX_DDIB |
 			TGL_DE_PORT_AUX_DDIC |
@@ -1015,7 +989,7 @@ static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
 			TGL_DE_PORT_AUX_USBC2 |
 			TGL_DE_PORT_AUX_USBC3 |
 			TGL_DE_PORT_AUX_USBC4;
-	else if (DISPLAY_VER(dev_priv) >= 12)
+	else if (DISPLAY_VER(display) >= 12)
 		return TGL_DE_PORT_AUX_DDIA |
 			TGL_DE_PORT_AUX_DDIB |
 			TGL_DE_PORT_AUX_DDIC |
@@ -1027,12 +1001,12 @@ static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
 			TGL_DE_PORT_AUX_USBC6;
 
 	mask = GEN8_AUX_CHANNEL_A;
-	if (DISPLAY_VER(dev_priv) >= 9)
+	if (DISPLAY_VER(display) >= 9)
 		mask |= GEN9_AUX_CHANNEL_B |
 			GEN9_AUX_CHANNEL_C |
 			GEN9_AUX_CHANNEL_D;
 
-	if (DISPLAY_VER(dev_priv) == 11) {
+	if (DISPLAY_VER(display) == 11) {
 		mask |= ICL_AUX_CHANNEL_F;
 		mask |= ICL_AUX_CHANNEL_E;
 	}
@@ -1040,10 +1014,8 @@ static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
 	return mask;
 }
 
-static u32 gen8_de_pipe_fault_mask(struct drm_i915_private *dev_priv)
+static u32 gen8_de_pipe_fault_mask(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
-
 	if (DISPLAY_VER(display) >= 14)
 		return MTL_PIPEDMC_ATS_FAULT |
 			MTL_PLANE_ATS_FAULT |
@@ -1195,15 +1167,14 @@ gen8_pipe_fault_handlers(struct intel_display *display)
 		return bdw_pipe_fault_handlers;
 }
 
-static void intel_pmdemand_irq_handler(struct drm_i915_private *dev_priv)
+static void intel_pmdemand_irq_handler(struct intel_display *display)
 {
-	wake_up_all(&dev_priv->display.pmdemand.waitqueue);
+	wake_up_all(&display->pmdemand.waitqueue);
 }
 
 static void
-gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
+gen8_de_misc_irq_handler(struct intel_display *display, u32 iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	bool found = false;
 
 	if (HAS_DBUF_OVERLAP_DETECTION(display)) {
@@ -1213,20 +1184,20 @@ gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 		}
 	}
 
-	if (DISPLAY_VER(dev_priv) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		if (iir & (XELPDP_PMDEMAND_RSP |
 			   XELPDP_PMDEMAND_RSPTOUT_ERR)) {
 			if (iir & XELPDP_PMDEMAND_RSPTOUT_ERR)
-				drm_dbg(&dev_priv->drm,
+				drm_dbg(display->drm,
 					"Error waiting for Punit PM Demand Response\n");
 
-			intel_pmdemand_irq_handler(dev_priv);
+			intel_pmdemand_irq_handler(display);
 			found = true;
 		}
 
 		if (iir & XELPDP_RM_TIMEOUT) {
 			u32 val = intel_de_read(display, RM_TIMEOUT_REG_CAPTURE);
-			drm_warn(&dev_priv->drm, "Register Access Timeout = 0x%x\n", val);
+			drm_warn(display->drm, "Register Access Timeout = 0x%x\n", val);
 			found = true;
 		}
 	} else if (iir & GEN8_DE_MISC_GSE) {
@@ -1239,12 +1210,12 @@ gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 		u32 psr_iir;
 		i915_reg_t iir_reg;
 
-		for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) {
+		for_each_intel_encoder_with_psr(display->drm, encoder) {
 			struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
-			if (DISPLAY_VER(dev_priv) >= 12)
-				iir_reg = TRANS_PSR_IIR(dev_priv,
-						        intel_dp->psr.transcoder);
+			if (DISPLAY_VER(display) >= 12)
+				iir_reg = TRANS_PSR_IIR(display,
+							intel_dp->psr.transcoder);
 			else
 				iir_reg = EDP_PSR_IIR;
 
@@ -1256,19 +1227,18 @@ gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 			intel_psr_irq_handler(intel_dp, psr_iir);
 
 			/* prior GEN12 only have one EDP PSR */
-			if (DISPLAY_VER(dev_priv) < 12)
+			if (DISPLAY_VER(display) < 12)
 				break;
 		}
 	}
 
 	if (!found)
-		drm_err(&dev_priv->drm, "Unexpected DE Misc interrupt: 0x%08x\n", iir);
+		drm_err(display->drm, "Unexpected DE Misc interrupt: 0x%08x\n", iir);
 }
 
-static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
+static void gen11_dsi_te_interrupt_handler(struct intel_display *display,
 					   u32 te_trigger)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe = INVALID_PIPE;
 	enum transcoder dsi_trans;
 	enum port port;
@@ -1278,7 +1248,7 @@ static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
 	 * Incase of dual link, TE comes from DSI_1
 	 * this is to check if dual link is enabled
 	 */
-	val = intel_de_read(display, TRANS_DDI_FUNC_CTL2(dev_priv, TRANSCODER_DSI_0));
+	val = intel_de_read(display, TRANS_DDI_FUNC_CTL2(display, TRANSCODER_DSI_0));
 	val &= PORT_SYNC_MODE_ENABLE;
 
 	/*
@@ -1294,12 +1264,12 @@ static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
 	val = val & OP_MODE_MASK;
 
 	if (val != CMD_MODE_NO_GATE && val != CMD_MODE_TE_GATE) {
-		drm_err(&dev_priv->drm, "DSI trancoder not configured in command mode\n");
+		drm_err(display->drm, "DSI trancoder not configured in command mode\n");
 		return;
 	}
 
 	/* Get PIPE for handling VBLANK event */
-	val = intel_de_read(display, TRANS_DDI_FUNC_CTL(dev_priv, dsi_trans));
+	val = intel_de_read(display, TRANS_DDI_FUNC_CTL(display, dsi_trans));
 	switch (val & TRANS_DDI_EDP_INPUT_MASK) {
 	case TRANS_DDI_EDP_INPUT_A_ON:
 		pipe = PIPE_A;
@@ -1311,28 +1281,27 @@ static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
 		pipe = PIPE_C;
 		break;
 	default:
-		drm_err(&dev_priv->drm, "Invalid PIPE\n");
+		drm_err(display->drm, "Invalid PIPE\n");
 		return;
 	}
 
-	intel_handle_vblank(dev_priv, pipe);
+	intel_handle_vblank(display, pipe);
 
 	/* clear TE in dsi IIR */
 	port = (te_trigger & DSI1_TE) ? PORT_B : PORT_A;
 	intel_de_rmw(display, DSI_INTR_IDENT_REG(port), 0, 0);
 }
 
-static u32 gen8_de_pipe_flip_done_mask(struct drm_i915_private *i915)
+static u32 gen8_de_pipe_flip_done_mask(struct intel_display *display)
 {
-	if (DISPLAY_VER(i915) >= 9)
+	if (DISPLAY_VER(display) >= 9)
 		return GEN9_PIPE_PLANE1_FLIP_DONE;
 	else
 		return GEN8_PIPE_PRIMARY_FLIP_DONE;
 }
 
-static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_iir, u32 *pica_iir)
+static void gen8_read_and_ack_pch_irqs(struct intel_display *display, u32 *pch_iir, u32 *pica_iir)
 {
-	struct intel_display *display = &i915->display;
 	u32 pica_ier = 0;
 
 	*pica_iir = 0;
@@ -1346,7 +1315,7 @@ static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_i
 	 * their flags both in the PICA and SDE IIR.
 	 */
 	if (*pch_iir & SDE_PICAINTERRUPT) {
-		drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTL);
+		drm_WARN_ON(display->drm, INTEL_PCH_TYPE(display) < PCH_MTL);
 
 		pica_ier = intel_de_rmw(display, PICAINTERRUPT_IER, ~0, 0);
 		*pica_iir = intel_de_read(display, PICAINTERRUPT_IIR);
@@ -1359,32 +1328,31 @@ static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_i
 		intel_de_write(display, PICAINTERRUPT_IER, pica_ier);
 }
 
-void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
+void gen8_de_irq_handler(struct intel_display *display, u32 master_ctl)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 iir;
 	enum pipe pipe;
 
-	drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_DISPLAY(dev_priv));
+	drm_WARN_ON_ONCE(display->drm, !HAS_DISPLAY(display));
 
 	if (master_ctl & GEN8_DE_MISC_IRQ) {
 		iir = intel_de_read(display, GEN8_DE_MISC_IIR);
 		if (iir) {
 			intel_de_write(display, GEN8_DE_MISC_IIR, iir);
-			gen8_de_misc_irq_handler(dev_priv, iir);
+			gen8_de_misc_irq_handler(display, iir);
 		} else {
-			drm_err_ratelimited(&dev_priv->drm,
+			drm_err_ratelimited(display->drm,
 					    "The master control interrupt lied (DE MISC)!\n");
 		}
 	}
 
-	if (DISPLAY_VER(dev_priv) >= 11 && (master_ctl & GEN11_DE_HPD_IRQ)) {
+	if (DISPLAY_VER(display) >= 11 && (master_ctl & GEN11_DE_HPD_IRQ)) {
 		iir = intel_de_read(display, GEN11_DE_HPD_IIR);
 		if (iir) {
 			intel_de_write(display, GEN11_DE_HPD_IIR, iir);
-			gen11_hpd_irq_handler(dev_priv, iir);
+			gen11_hpd_irq_handler(display, iir);
 		} else {
-			drm_err_ratelimited(&dev_priv->drm,
+			drm_err_ratelimited(display->drm,
 					    "The master control interrupt lied, (DE HPD)!\n");
 		}
 	}
@@ -1396,52 +1364,52 @@ void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 
 			intel_de_write(display, GEN8_DE_PORT_IIR, iir);
 
-			if (iir & gen8_de_port_aux_mask(dev_priv)) {
+			if (iir & gen8_de_port_aux_mask(display)) {
 				intel_dp_aux_irq_handler(display);
 				found = true;
 			}
 
-			if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+			if (display->platform.geminilake || display->platform.broxton) {
 				u32 hotplug_trigger = iir & BXT_DE_PORT_HOTPLUG_MASK;
 
 				if (hotplug_trigger) {
-					bxt_hpd_irq_handler(dev_priv, hotplug_trigger);
+					bxt_hpd_irq_handler(display, hotplug_trigger);
 					found = true;
 				}
-			} else if (IS_BROADWELL(dev_priv)) {
+			} else if (display->platform.broadwell) {
 				u32 hotplug_trigger = iir & BDW_DE_PORT_HOTPLUG_MASK;
 
 				if (hotplug_trigger) {
-					ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
+					ilk_hpd_irq_handler(display, hotplug_trigger);
 					found = true;
 				}
 			}
 
-			if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
+			if ((display->platform.geminilake || display->platform.broxton) &&
 			    (iir & BXT_DE_PORT_GMBUS)) {
 				intel_gmbus_irq_handler(display);
 				found = true;
 			}
 
-			if (DISPLAY_VER(dev_priv) >= 11) {
+			if (DISPLAY_VER(display) >= 11) {
 				u32 te_trigger = iir & (DSI0_TE | DSI1_TE);
 
 				if (te_trigger) {
-					gen11_dsi_te_interrupt_handler(dev_priv, te_trigger);
+					gen11_dsi_te_interrupt_handler(display, te_trigger);
 					found = true;
 				}
 			}
 
 			if (!found)
-				drm_err_ratelimited(&dev_priv->drm,
+				drm_err_ratelimited(display->drm,
 						    "Unexpected DE Port interrupt\n");
 		} else {
-			drm_err_ratelimited(&dev_priv->drm,
+			drm_err_ratelimited(display->drm,
 					    "The master control interrupt lied (DE PORT)!\n");
 		}
 	}
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		u32 fault_errors;
 
 		if (!(master_ctl & GEN8_DE_PIPE_IRQ(pipe)))
@@ -1449,7 +1417,7 @@ void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 
 		iir = intel_de_read(display, GEN8_DE_PIPE_IIR(pipe));
 		if (!iir) {
-			drm_err_ratelimited(&dev_priv->drm,
+			drm_err_ratelimited(display->drm,
 					    "The master control interrupt lied (DE PIPE)!\n");
 			continue;
 		}
@@ -1457,36 +1425,36 @@ void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 		intel_de_write(display, GEN8_DE_PIPE_IIR(pipe), iir);
 
 		if (iir & GEN8_PIPE_VBLANK)
-			intel_handle_vblank(dev_priv, pipe);
+			intel_handle_vblank(display, pipe);
 
-		if (iir & gen8_de_pipe_flip_done_mask(dev_priv))
-			flip_done_handler(dev_priv, pipe);
+		if (iir & gen8_de_pipe_flip_done_mask(display))
+			flip_done_handler(display, pipe);
 
-		if (HAS_DSB(dev_priv)) {
+		if (HAS_DSB(display)) {
 			if (iir & GEN12_DSB_INT(INTEL_DSB_0))
-				intel_dsb_irq_handler(&dev_priv->display, pipe, INTEL_DSB_0);
+				intel_dsb_irq_handler(display, pipe, INTEL_DSB_0);
 
 			if (iir & GEN12_DSB_INT(INTEL_DSB_1))
-				intel_dsb_irq_handler(&dev_priv->display, pipe, INTEL_DSB_1);
+				intel_dsb_irq_handler(display, pipe, INTEL_DSB_1);
 
 			if (iir & GEN12_DSB_INT(INTEL_DSB_2))
-				intel_dsb_irq_handler(&dev_priv->display, pipe, INTEL_DSB_2);
+				intel_dsb_irq_handler(display, pipe, INTEL_DSB_2);
 		}
 
 		if (iir & GEN8_PIPE_CDCLK_CRC_DONE)
-			hsw_pipe_crc_irq_handler(dev_priv, pipe);
+			hsw_pipe_crc_irq_handler(display, pipe);
 
 		if (iir & GEN8_PIPE_FIFO_UNDERRUN)
 			intel_cpu_fifo_underrun_irq_handler(display, pipe);
 
-		fault_errors = iir & gen8_de_pipe_fault_mask(dev_priv);
+		fault_errors = iir & gen8_de_pipe_fault_mask(display);
 		if (fault_errors)
 			intel_pipe_fault_irq_handler(display,
 						     gen8_pipe_fault_handlers(display),
 						     pipe, fault_errors);
 	}
 
-	if (HAS_PCH_SPLIT(dev_priv) && !HAS_PCH_NOP(dev_priv) &&
+	if (HAS_PCH_SPLIT(display) && !HAS_PCH_NOP(display) &&
 	    master_ctl & GEN8_DE_PCH_IRQ) {
 		u32 pica_iir;
 
@@ -1495,31 +1463,30 @@ void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 		 * scheme also closed the SDE interrupt handling race we've seen
 		 * on older pch-split platforms. But this needs testing.
 		 */
-		gen8_read_and_ack_pch_irqs(dev_priv, &iir, &pica_iir);
+		gen8_read_and_ack_pch_irqs(display, &iir, &pica_iir);
 		if (iir) {
 			if (pica_iir)
-				xelpdp_pica_irq_handler(dev_priv, pica_iir);
+				xelpdp_pica_irq_handler(display, pica_iir);
 
-			if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-				icp_irq_handler(dev_priv, iir);
-			else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
-				spt_irq_handler(dev_priv, iir);
+			if (INTEL_PCH_TYPE(display) >= PCH_ICP)
+				icp_irq_handler(display, iir);
+			else if (INTEL_PCH_TYPE(display) >= PCH_SPT)
+				spt_irq_handler(display, iir);
 			else
-				cpt_irq_handler(dev_priv, iir);
+				cpt_irq_handler(display, iir);
 		} else {
 			/*
 			 * Like on previous PCH there seems to be something
 			 * fishy going on with forwarding PCH interrupts.
 			 */
-			drm_dbg(&dev_priv->drm,
+			drm_dbg(display->drm,
 				"The master control interrupt lied (SDE)!\n");
 		}
 	}
 }
 
-u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl)
+u32 gen11_gu_misc_irq_ack(struct intel_display *display, const u32 master_ctl)
 {
-	struct intel_display *display = &i915->display;
 	u32 iir;
 
 	if (!(master_ctl & GEN11_GU_MISC_IRQ))
@@ -1532,20 +1499,17 @@ u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl)
 	return iir;
 }
 
-void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir)
+void gen11_gu_misc_irq_handler(struct intel_display *display, const u32 iir)
 {
-	struct intel_display *display = &i915->display;
-
 	if (iir & GEN11_GU_MISC_GSE)
 		intel_opregion_asle_intr(display);
 }
 
-void gen11_display_irq_handler(struct drm_i915_private *i915)
+void gen11_display_irq_handler(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	u32 disp_ctl;
 
-	disable_rpm_wakeref_asserts(&i915->runtime_pm);
+	intel_display_rpm_assert_block(display);
 	/*
 	 * GEN11_DISPLAY_INT_CTL has same format as GEN8_MASTER_IRQ
 	 * for the display related bits.
@@ -1553,16 +1517,15 @@ void gen11_display_irq_handler(struct drm_i915_private *i915)
 	disp_ctl = intel_de_read(display, GEN11_DISPLAY_INT_CTL);
 
 	intel_de_write(display, GEN11_DISPLAY_INT_CTL, 0);
-	gen8_de_irq_handler(i915, disp_ctl);
+	gen8_de_irq_handler(display, disp_ctl);
 	intel_de_write(display, GEN11_DISPLAY_INT_CTL, GEN11_DISPLAY_IRQ_ENABLE);
 
-	enable_rpm_wakeref_asserts(&i915->runtime_pm);
+	intel_display_rpm_assert_unblock(display);
 }
 
-static void i915gm_irq_cstate_wa_enable(struct drm_i915_private *i915)
+static void i915gm_irq_cstate_wa_enable(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-	lockdep_assert_held(&i915->drm.vblank_time_lock);
+	lockdep_assert_held(&display->drm->vblank_time_lock);
 
 	/*
 	 * Vblank/CRC interrupts fail to wake the device up from C2+.
@@ -1570,117 +1533,116 @@ static void i915gm_irq_cstate_wa_enable(struct drm_i915_private *i915)
 	 * the problem. There is a small power cost so we do this
 	 * only when vblank/CRC interrupts are actually enabled.
 	 */
-	if (i915->display.irq.vblank_enabled++ == 0)
+	if (display->irq.vblank_enabled++ == 0)
 		intel_de_write(display, SCPD0,
 			       _MASKED_BIT_ENABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE));
 }
 
-static void i915gm_irq_cstate_wa_disable(struct drm_i915_private *i915)
+static void i915gm_irq_cstate_wa_disable(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-	lockdep_assert_held(&i915->drm.vblank_time_lock);
+	lockdep_assert_held(&display->drm->vblank_time_lock);
 
-	if (--i915->display.irq.vblank_enabled == 0)
+	if (--display->irq.vblank_enabled == 0)
 		intel_de_write(display, SCPD0,
 			       _MASKED_BIT_DISABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE));
 }
 
-void i915gm_irq_cstate_wa(struct drm_i915_private *i915, bool enable)
+void i915gm_irq_cstate_wa(struct intel_display *display, bool enable)
 {
-	spin_lock_irq(&i915->drm.vblank_time_lock);
+	spin_lock_irq(&display->drm->vblank_time_lock);
 
 	if (enable)
-		i915gm_irq_cstate_wa_enable(i915);
+		i915gm_irq_cstate_wa_enable(display);
 	else
-		i915gm_irq_cstate_wa_disable(i915);
+		i915gm_irq_cstate_wa_disable(display);
 
-	spin_unlock_irq(&i915->drm.vblank_time_lock);
+	spin_unlock_irq(&display->drm->vblank_time_lock);
 }
 
 int i8xx_enable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 	unsigned long irqflags;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	i915_enable_pipestat(display, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 
 	return 0;
 }
 
 void i8xx_disable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 	unsigned long irqflags;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_disable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	i915_disable_pipestat(display, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 }
 
 int i915gm_enable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 
-	i915gm_irq_cstate_wa_enable(i915);
+	i915gm_irq_cstate_wa_enable(display);
 
 	return i8xx_enable_vblank(crtc);
 }
 
 void i915gm_disable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 
 	i8xx_disable_vblank(crtc);
 
-	i915gm_irq_cstate_wa_disable(i915);
+	i915gm_irq_cstate_wa_disable(display);
 }
 
 int i965_enable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 	unsigned long irqflags;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, pipe,
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	i915_enable_pipestat(display, pipe,
 			     PIPE_START_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 
 	return 0;
 }
 
 void i965_disable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 	unsigned long irqflags;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_disable_pipestat(dev_priv, pipe,
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	i915_disable_pipestat(display, pipe,
 			      PIPE_START_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 }
 
 int ilk_enable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 	unsigned long irqflags;
-	u32 bit = DISPLAY_VER(dev_priv) >= 7 ?
+	u32 bit = DISPLAY_VER(display) >= 7 ?
 		DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe);
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	ilk_enable_display_irq(dev_priv, bit);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	ilk_enable_display_irq(display, bit);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 
 	/* Even though there is no DMC, frame counter can get stuck when
 	 * PSR is active as no frames are generated.
 	 */
-	if (HAS_PSR(dev_priv))
+	if (HAS_PSR(display))
 		drm_crtc_vblank_restore(crtc);
 
 	return 0;
@@ -1688,15 +1650,15 @@ int ilk_enable_vblank(struct drm_crtc *crtc)
 
 void ilk_disable_vblank(struct drm_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum pipe pipe = to_intel_crtc(crtc)->pipe;
 	unsigned long irqflags;
-	u32 bit = DISPLAY_VER(dev_priv) >= 7 ?
+	u32 bit = DISPLAY_VER(display) >= 7 ?
 		DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe);
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	ilk_disable_display_irq(dev_priv, bit);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	ilk_disable_display_irq(display, bit);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 }
 
 static bool gen11_dsi_configure_te(struct intel_crtc *intel_crtc,
@@ -1722,44 +1684,36 @@ static bool gen11_dsi_configure_te(struct intel_crtc *intel_crtc,
 	return true;
 }
 
-static void intel_display_vblank_dc_work(struct work_struct *work)
+static void intel_display_vblank_notify_work(struct work_struct *work)
 {
 	struct intel_display *display =
-		container_of(work, typeof(*display), irq.vblank_dc_work);
-	int vblank_wa_num_pipes = READ_ONCE(display->irq.vblank_wa_num_pipes);
+		container_of(work, typeof(*display), irq.vblank_notify_work);
+	int vblank_enable_count = READ_ONCE(display->irq.vblank_enable_count);
 
-	/*
-	 * NOTE: intel_display_power_set_target_dc_state is used only by PSR
-	 * code for DC3CO handling. DC3CO target state is currently disabled in
-	 * PSR code. If DC3CO is taken into use we need take that into account
-	 * here as well.
-	 */
-	intel_display_power_set_target_dc_state(display, vblank_wa_num_pipes ? DC_STATE_DISABLE :
-						DC_STATE_EN_UPTO_DC6);
+	intel_psr_notify_vblank_enable_disable(display, vblank_enable_count);
 }
 
 int bdw_enable_vblank(struct drm_crtc *_crtc)
 {
 	struct intel_crtc *crtc = to_intel_crtc(_crtc);
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	unsigned long irqflags;
 
 	if (gen11_dsi_configure_te(crtc, true))
 		return 0;
 
-	if (crtc->block_dc_for_vblank && display->irq.vblank_wa_num_pipes++ == 0)
-		schedule_work(&display->irq.vblank_dc_work);
+	if (crtc->vblank_psr_notify && display->irq.vblank_enable_count++ == 0)
+		schedule_work(&display->irq.vblank_notify_work);
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	bdw_enable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	bdw_enable_pipe_irq(display, pipe, GEN8_PIPE_VBLANK);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 
 	/* Even if there is no DMC, frame counter can get stuck when
 	 * PSR is active as no frames are generated, so check only for PSR.
 	 */
-	if (HAS_PSR(dev_priv))
+	if (HAS_PSR(display))
 		drm_crtc_vblank_restore(&crtc->base);
 
 	return 0;
@@ -1769,19 +1723,18 @@ void bdw_disable_vblank(struct drm_crtc *_crtc)
 {
 	struct intel_crtc *crtc = to_intel_crtc(_crtc);
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	unsigned long irqflags;
 
 	if (gen11_dsi_configure_te(crtc, false))
 		return;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	bdw_disable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	spin_lock_irqsave(&display->irq.lock, irqflags);
+	bdw_disable_pipe_irq(display, pipe, GEN8_PIPE_VBLANK);
+	spin_unlock_irqrestore(&display->irq.lock, irqflags);
 
-	if (crtc->block_dc_for_vblank && --display->irq.vblank_wa_num_pipes == 0)
-		schedule_work(&display->irq.vblank_dc_work);
+	if (crtc->vblank_psr_notify && --display->irq.vblank_enable_count == 0)
+		schedule_work(&display->irq.vblank_notify_work);
 }
 
 static u32 vlv_dpinvgtt_pipe_fault_mask(enum pipe pipe)
@@ -1892,11 +1845,11 @@ void vlv_display_error_irq_handler(struct intel_display *display,
 		vlv_page_table_error_irq_handler(display, dpinvgtt);
 }
 
-static void _vlv_display_irq_reset(struct drm_i915_private *dev_priv)
+static void _vlv_display_irq_reset(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 
-	if (IS_CHERRYVIEW(dev_priv))
+	if (display->platform.cherryview)
 		intel_de_write(display, DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
 	else
 		intel_de_write(display, DPINVGTT, DPINVGTT_STATUS_MASK_VLV);
@@ -1904,31 +1857,60 @@ static void _vlv_display_irq_reset(struct drm_i915_private *dev_priv)
 	gen2_error_reset(to_intel_uncore(display->drm),
 			 VLV_ERROR_REGS);
 
-	i915_hotplug_interrupt_update_locked(dev_priv, 0xffffffff, 0);
-	intel_de_rmw(display, PORT_HOTPLUG_STAT(dev_priv), 0, 0);
+	i915_hotplug_interrupt_update_locked(display, 0xffffffff, 0);
+	intel_de_rmw(display, PORT_HOTPLUG_STAT(display), 0, 0);
 
-	i9xx_pipestat_irq_reset(dev_priv);
+	i9xx_pipestat_irq_reset(display);
 
 	intel_display_irq_regs_reset(display, VLV_IRQ_REGS);
 	dev_priv->irq_mask = ~0u;
 }
 
-void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
+void vlv_display_irq_reset(struct intel_display *display)
 {
-	if (dev_priv->display.irq.vlv_display_irqs_enabled)
-		_vlv_display_irq_reset(dev_priv);
+	spin_lock_irq(&display->irq.lock);
+	if (display->irq.vlv_display_irqs_enabled)
+		_vlv_display_irq_reset(display);
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void i9xx_display_irq_reset(struct drm_i915_private *i915)
+void i9xx_display_irq_reset(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-
-	if (I915_HAS_HOTPLUG(i915)) {
-		i915_hotplug_interrupt_update(i915, 0xffffffff, 0);
-		intel_de_rmw(display, PORT_HOTPLUG_STAT(i915), 0, 0);
+	if (HAS_HOTPLUG(display)) {
+		i915_hotplug_interrupt_update(display, 0xffffffff, 0);
+		intel_de_rmw(display, PORT_HOTPLUG_STAT(display), 0, 0);
 	}
 
-	i9xx_pipestat_irq_reset(i915);
+	i9xx_pipestat_irq_reset(display);
+}
+
+void i915_display_irq_postinstall(struct intel_display *display)
+{
+	/*
+	 * Interrupt setup is already guaranteed to be single-threaded, this is
+	 * just to make the assert_spin_locked check happy.
+	 */
+	spin_lock_irq(&display->irq.lock);
+	i915_enable_pipestat(display, PIPE_A, PIPE_CRC_DONE_INTERRUPT_STATUS);
+	i915_enable_pipestat(display, PIPE_B, PIPE_CRC_DONE_INTERRUPT_STATUS);
+	spin_unlock_irq(&display->irq.lock);
+
+	i915_enable_asle_pipestat(display);
+}
+
+void i965_display_irq_postinstall(struct intel_display *display)
+{
+	/*
+	 * Interrupt setup is already guaranteed to be single-threaded, this is
+	 * just to make the assert_spin_locked check happy.
+	 */
+	spin_lock_irq(&display->irq.lock);
+	i915_enable_pipestat(display, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
+	i915_enable_pipestat(display, PIPE_A, PIPE_CRC_DONE_INTERRUPT_STATUS);
+	i915_enable_pipestat(display, PIPE_B, PIPE_CRC_DONE_INTERRUPT_STATUS);
+	spin_unlock_irq(&display->irq.lock);
+
+	i915_enable_asle_pipestat(display);
 }
 
 static u32 vlv_error_mask(void)
@@ -1937,17 +1919,14 @@ static u32 vlv_error_mask(void)
 	return VLV_ERROR_PAGE_TABLE;
 }
 
-void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
+static void _vlv_display_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 pipestat_mask;
 	u32 enable_mask;
 	enum pipe pipe;
 
-	if (!dev_priv->display.irq.vlv_display_irqs_enabled)
-		return;
-
-	if (IS_CHERRYVIEW(dev_priv))
+	if (display->platform.cherryview)
 		intel_de_write(display, DPINVGTT,
 			       DPINVGTT_STATUS_MASK_CHV |
 			       DPINVGTT_EN_MASK_CHV);
@@ -1961,9 +1940,9 @@ void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
 
 	pipestat_mask = PIPE_CRC_DONE_INTERRUPT_STATUS;
 
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
-	for_each_pipe(dev_priv, pipe)
-		i915_enable_pipestat(dev_priv, pipe, pipestat_mask);
+	i915_enable_pipestat(display, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
+	for_each_pipe(display, pipe)
+		i915_enable_pipestat(display, pipe, pipestat_mask);
 
 	enable_mask = I915_DISPLAY_PORT_INTERRUPT |
 		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
@@ -1972,53 +1951,76 @@ void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
 		I915_LPE_PIPE_B_INTERRUPT |
 		I915_MASTER_ERROR_INTERRUPT;
 
-	if (IS_CHERRYVIEW(dev_priv))
+	if (display->platform.cherryview)
 		enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT |
 			I915_LPE_PIPE_C_INTERRUPT;
 
-	drm_WARN_ON(&dev_priv->drm, dev_priv->irq_mask != ~0u);
+	drm_WARN_ON(display->drm, dev_priv->irq_mask != ~0u);
 
 	dev_priv->irq_mask = ~enable_mask;
 
 	intel_display_irq_regs_init(display, VLV_IRQ_REGS, dev_priv->irq_mask, enable_mask);
 }
 
-void gen8_display_irq_reset(struct drm_i915_private *dev_priv)
+void vlv_display_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
+	spin_lock_irq(&display->irq.lock);
+	if (display->irq.vlv_display_irqs_enabled)
+		_vlv_display_irq_postinstall(display);
+	spin_unlock_irq(&display->irq.lock);
+}
+
+void ibx_display_irq_reset(struct intel_display *display)
+{
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
+	if (HAS_PCH_NOP(i915))
+		return;
+
+	gen2_irq_reset(to_intel_uncore(display->drm), SDE_IRQ_REGS);
+
+	if (HAS_PCH_CPT(i915) || HAS_PCH_LPT(i915))
+		intel_de_write(display, SERR_INT, 0xffffffff);
+}
+
+void gen8_display_irq_reset(struct intel_display *display)
+{
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	enum pipe pipe;
 
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
 	intel_de_write(display, EDP_PSR_IMR, 0xffffffff);
 	intel_de_write(display, EDP_PSR_IIR, 0xffffffff);
 
-	for_each_pipe(dev_priv, pipe)
+	for_each_pipe(display, pipe)
 		if (intel_display_power_is_enabled(display,
 						   POWER_DOMAIN_PIPE(pipe)))
 			intel_display_irq_regs_reset(display, GEN8_DE_PIPE_IRQ_REGS(pipe));
 
 	intel_display_irq_regs_reset(display, GEN8_DE_PORT_IRQ_REGS);
 	intel_display_irq_regs_reset(display, GEN8_DE_MISC_IRQ_REGS);
+
+	if (HAS_PCH_SPLIT(i915))
+		ibx_display_irq_reset(display);
 }
 
-void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
+void gen11_display_irq_reset(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe pipe;
 	u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
 
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
 	intel_de_write(display, GEN11_DISPLAY_INT_CTL, 0);
 
-	if (DISPLAY_VER(dev_priv) >= 12) {
+	if (DISPLAY_VER(display) >= 12) {
 		enum transcoder trans;
 
-		for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) {
+		for_each_cpu_transcoder_masked(display, trans, trans_mask) {
 			enum intel_display_power_domain domain;
 
 			domain = POWER_DOMAIN_TRANSCODER(trans);
@@ -2026,10 +2028,10 @@ void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
 				continue;
 
 			intel_de_write(display,
-				       TRANS_PSR_IMR(dev_priv, trans),
+				       TRANS_PSR_IMR(display, trans),
 				       0xffffffff);
 			intel_de_write(display,
-				       TRANS_PSR_IIR(dev_priv, trans),
+				       TRANS_PSR_IIR(display, trans),
 				       0xffffffff);
 		}
 	} else {
@@ -2037,7 +2039,7 @@ void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
 		intel_de_write(display, EDP_PSR_IIR, 0xffffffff);
 	}
 
-	for_each_pipe(dev_priv, pipe)
+	for_each_pipe(display, pipe)
 		if (intel_display_power_is_enabled(display,
 						   POWER_DOMAIN_PIPE(pipe)))
 			intel_display_irq_regs_reset(display, GEN8_DE_PIPE_IRQ_REGS(pipe));
@@ -2045,55 +2047,55 @@ void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
 	intel_display_irq_regs_reset(display, GEN8_DE_PORT_IRQ_REGS);
 	intel_display_irq_regs_reset(display, GEN8_DE_MISC_IRQ_REGS);
 
-	if (DISPLAY_VER(dev_priv) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		intel_display_irq_regs_reset(display, PICAINTERRUPT_IRQ_REGS);
 	else
 		intel_display_irq_regs_reset(display, GEN11_DE_HPD_IRQ_REGS);
 
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+	if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 		intel_display_irq_regs_reset(display, SDE_IRQ_REGS);
 }
 
-void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
+void gen8_irq_power_well_post_enable(struct intel_display *display,
 				     u8 pipe_mask)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 extra_ier = GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN |
-		gen8_de_pipe_flip_done_mask(dev_priv);
+		gen8_de_pipe_flip_done_mask(display);
 	enum pipe pipe;
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
 	if (!intel_irqs_enabled(dev_priv)) {
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_unlock_irq(&display->irq.lock);
 		return;
 	}
 
-	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
+	for_each_pipe_masked(display, pipe, pipe_mask)
 		intel_display_irq_regs_init(display, GEN8_DE_PIPE_IRQ_REGS(pipe),
-					    dev_priv->display.irq.de_irq_mask[pipe],
-					    ~dev_priv->display.irq.de_irq_mask[pipe] | extra_ier);
+					    display->irq.de_irq_mask[pipe],
+					    ~display->irq.de_irq_mask[pipe] | extra_ier);
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
+void gen8_irq_power_well_pre_disable(struct intel_display *display,
 				     u8 pipe_mask)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum pipe pipe;
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
 	if (!intel_irqs_enabled(dev_priv)) {
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_unlock_irq(&display->irq.lock);
 		return;
 	}
 
-	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
+	for_each_pipe_masked(display, pipe, pipe_mask)
 		intel_display_irq_regs_reset(display, GEN8_DE_PIPE_IRQ_REGS(pipe));
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
 	/* make sure we're done processing display irqs */
 	intel_synchronize_irq(dev_priv);
@@ -2110,17 +2112,16 @@ void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
  * to avoid races with the irq handler, assuming we have MSI. Shared legacy
  * interrupts could still race.
  */
-static void ibx_irq_postinstall(struct drm_i915_private *dev_priv)
+static void ibx_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 mask;
 
-	if (HAS_PCH_NOP(dev_priv))
+	if (HAS_PCH_NOP(display))
 		return;
 
-	if (HAS_PCH_IBX(dev_priv))
+	if (HAS_PCH_IBX(display))
 		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
-	else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
+	else if (HAS_PCH_CPT(display) || HAS_PCH_LPT(display))
 		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
 	else
 		mask = SDE_GMBUS_CPT;
@@ -2128,40 +2129,50 @@ static void ibx_irq_postinstall(struct drm_i915_private *dev_priv)
 	intel_display_irq_regs_init(display, SDE_IRQ_REGS, ~mask, 0xffffffff);
 }
 
-void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv)
+void valleyview_enable_display_irqs(struct intel_display *display)
 {
-	lockdep_assert_held(&dev_priv->irq_lock);
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 
-	if (dev_priv->display.irq.vlv_display_irqs_enabled)
-		return;
+	spin_lock_irq(&display->irq.lock);
+
+	if (display->irq.vlv_display_irqs_enabled)
+		goto out;
 
-	dev_priv->display.irq.vlv_display_irqs_enabled = true;
+	display->irq.vlv_display_irqs_enabled = true;
 
 	if (intel_irqs_enabled(dev_priv)) {
-		_vlv_display_irq_reset(dev_priv);
-		vlv_display_irq_postinstall(dev_priv);
+		_vlv_display_irq_reset(display);
+		_vlv_display_irq_postinstall(display);
 	}
+
+out:
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv)
+void valleyview_disable_display_irqs(struct intel_display *display)
 {
-	lockdep_assert_held(&dev_priv->irq_lock);
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 
-	if (!dev_priv->display.irq.vlv_display_irqs_enabled)
-		return;
+	spin_lock_irq(&display->irq.lock);
+
+	if (!display->irq.vlv_display_irqs_enabled)
+		goto out;
 
-	dev_priv->display.irq.vlv_display_irqs_enabled = false;
+	display->irq.vlv_display_irqs_enabled = false;
 
 	if (intel_irqs_enabled(dev_priv))
-		_vlv_display_irq_reset(dev_priv);
+		_vlv_display_irq_reset(display);
+out:
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void ilk_de_irq_postinstall(struct drm_i915_private *i915)
+void ilk_de_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
 	u32 display_mask, extra_mask;
 
-	if (DISPLAY_VER(i915) >= 7) {
+	if (DISPLAY_VER(display) >= 7) {
 		display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE_IVB |
 				DE_PCH_EVENT_IVB | DE_AUX_CHANNEL_A_IVB);
 		extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB |
@@ -2182,59 +2193,57 @@ void ilk_de_irq_postinstall(struct drm_i915_private *i915)
 			      DE_DP_A_HOTPLUG);
 	}
 
-	if (IS_HASWELL(i915)) {
+	if (display->platform.haswell) {
 		intel_display_irq_regs_assert_irr_is_zero(display, EDP_PSR_IIR);
 		display_mask |= DE_EDP_PSR_INT_HSW;
 	}
 
-	if (IS_IRONLAKE_M(i915))
+	if (display->platform.ironlake && display->platform.mobile)
 		extra_mask |= DE_PCU_EVENT;
 
 	i915->irq_mask = ~display_mask;
 
-	ibx_irq_postinstall(i915);
+	ibx_irq_postinstall(display);
 
 	intel_display_irq_regs_init(display, DE_IRQ_REGS, i915->irq_mask,
 				    display_mask | extra_mask);
 }
 
-static void mtp_irq_postinstall(struct drm_i915_private *i915);
-static void icp_irq_postinstall(struct drm_i915_private *i915);
+static void mtp_irq_postinstall(struct intel_display *display);
+static void icp_irq_postinstall(struct intel_display *display);
 
-void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
+void gen8_de_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
-
-	u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) |
+	u32 de_pipe_masked = gen8_de_pipe_fault_mask(display) |
 		GEN8_PIPE_CDCLK_CRC_DONE;
 	u32 de_pipe_enables;
-	u32 de_port_masked = gen8_de_port_aux_mask(dev_priv);
+	u32 de_port_masked = gen8_de_port_aux_mask(display);
 	u32 de_port_enables;
 	u32 de_misc_masked = GEN8_DE_EDP_PSR;
 	u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
 	enum pipe pipe;
 
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
-	if (DISPLAY_VER(dev_priv) >= 14)
-		mtp_irq_postinstall(dev_priv);
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-		icp_irq_postinstall(dev_priv);
-	else if (HAS_PCH_SPLIT(dev_priv))
-		ibx_irq_postinstall(dev_priv);
+	if (DISPLAY_VER(display) >= 14)
+		mtp_irq_postinstall(display);
+	else if (INTEL_PCH_TYPE(display) >= PCH_ICP)
+		icp_irq_postinstall(display);
+	else if (HAS_PCH_SPLIT(display))
+		ibx_irq_postinstall(display);
 
-	if (DISPLAY_VER(dev_priv) < 11)
+	if (DISPLAY_VER(display) < 11)
 		de_misc_masked |= GEN8_DE_MISC_GSE;
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		de_port_masked |= BXT_DE_PORT_GMBUS;
 
-	if (DISPLAY_VER(dev_priv) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		de_misc_masked |= XELPDP_PMDEMAND_RSPTOUT_ERR |
 				  XELPDP_PMDEMAND_RSP | XELPDP_RM_TIMEOUT;
-	} else if (DISPLAY_VER(dev_priv) >= 11) {
+	} else if (DISPLAY_VER(display) >= 11) {
 		enum port port;
 
 		if (intel_bios_is_dsi_present(display, &port))
@@ -2244,25 +2253,25 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	if (HAS_DBUF_OVERLAP_DETECTION(display))
 		de_misc_masked |= XE2LPD_DBUF_OVERLAP_DETECTED;
 
-	if (HAS_DSB(dev_priv))
+	if (HAS_DSB(display))
 		de_pipe_masked |= GEN12_DSB_INT(INTEL_DSB_0) |
 			GEN12_DSB_INT(INTEL_DSB_1) |
 			GEN12_DSB_INT(INTEL_DSB_2);
 
 	de_pipe_enables = de_pipe_masked |
 		GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN |
-		gen8_de_pipe_flip_done_mask(dev_priv);
+		gen8_de_pipe_flip_done_mask(display);
 
 	de_port_enables = de_port_masked;
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		de_port_enables |= BXT_DE_PORT_HOTPLUG_MASK;
-	else if (IS_BROADWELL(dev_priv))
+	else if (display->platform.broadwell)
 		de_port_enables |= BDW_DE_PORT_HOTPLUG_MASK;
 
-	if (DISPLAY_VER(dev_priv) >= 12) {
+	if (DISPLAY_VER(display) >= 12) {
 		enum transcoder trans;
 
-		for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) {
+		for_each_cpu_transcoder_masked(display, trans, trans_mask) {
 			enum intel_display_power_domain domain;
 
 			domain = POWER_DOMAIN_TRANSCODER(trans);
@@ -2270,19 +2279,19 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 				continue;
 
 			intel_display_irq_regs_assert_irr_is_zero(display,
-								  TRANS_PSR_IIR(dev_priv, trans));
+								  TRANS_PSR_IIR(display, trans));
 		}
 	} else {
 		intel_display_irq_regs_assert_irr_is_zero(display, EDP_PSR_IIR);
 	}
 
-	for_each_pipe(dev_priv, pipe) {
-		dev_priv->display.irq.de_irq_mask[pipe] = ~de_pipe_masked;
+	for_each_pipe(display, pipe) {
+		display->irq.de_irq_mask[pipe] = ~de_pipe_masked;
 
 		if (intel_display_power_is_enabled(display,
 						   POWER_DOMAIN_PIPE(pipe)))
 			intel_display_irq_regs_init(display, GEN8_DE_PIPE_IRQ_REGS(pipe),
-						    dev_priv->display.irq.de_irq_mask[pipe],
+						    display->irq.de_irq_mask[pipe],
 						    de_pipe_enables);
 	}
 
@@ -2291,7 +2300,7 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	intel_display_irq_regs_init(display, GEN8_DE_MISC_IRQ_REGS, ~de_misc_masked,
 				    de_misc_masked);
 
-	if (IS_DISPLAY_VER(dev_priv, 11, 13)) {
+	if (IS_DISPLAY_VER(display, 11, 13)) {
 		u32 de_hpd_masked = 0;
 		u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK |
 				     GEN11_DE_TBT_HOTPLUG_MASK;
@@ -2301,9 +2310,8 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void mtp_irq_postinstall(struct drm_i915_private *i915)
+static void mtp_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	u32 sde_mask = SDE_GMBUS_ICP | SDE_PICAINTERRUPT;
 	u32 de_hpd_mask = XELPDP_AUX_TC_MASK;
 	u32 de_hpd_enables = de_hpd_mask | XELPDP_DP_ALT_HOTPLUG_MASK |
@@ -2315,43 +2323,68 @@ static void mtp_irq_postinstall(struct drm_i915_private *i915)
 	intel_display_irq_regs_init(display, SDE_IRQ_REGS, ~sde_mask, 0xffffffff);
 }
 
-static void icp_irq_postinstall(struct drm_i915_private *dev_priv)
+static void icp_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 mask = SDE_GMBUS_ICP;
 
 	intel_display_irq_regs_init(display, SDE_IRQ_REGS, ~mask, 0xffffffff);
 }
 
-void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv)
+void gen11_de_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
-
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
-	gen8_de_irq_postinstall(dev_priv);
+	gen8_de_irq_postinstall(display);
 
 	intel_de_write(display, GEN11_DISPLAY_INT_CTL, GEN11_DISPLAY_IRQ_ENABLE);
 }
 
-void dg1_de_irq_postinstall(struct drm_i915_private *i915)
+void dg1_de_irq_postinstall(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-
-	if (!HAS_DISPLAY(i915))
+	if (!HAS_DISPLAY(display))
 		return;
 
-	gen8_de_irq_postinstall(i915);
+	gen8_de_irq_postinstall(display);
 	intel_de_write(display, GEN11_DISPLAY_INT_CTL, GEN11_DISPLAY_IRQ_ENABLE);
 }
 
-void intel_display_irq_init(struct drm_i915_private *i915)
+void intel_display_irq_init(struct intel_display *display)
 {
-	i915->drm.vblank_disable_immediate = true;
+	spin_lock_init(&display->irq.lock);
 
-	intel_hotplug_irq_init(i915);
+	display->drm->vblank_disable_immediate = true;
+
+	intel_hotplug_irq_init(display);
+
+	INIT_WORK(&display->irq.vblank_notify_work,
+		  intel_display_vblank_notify_work);
+}
+
+struct intel_display_irq_snapshot {
+	u32 derrmr;
+};
+
+struct intel_display_irq_snapshot *
+intel_display_irq_snapshot_capture(struct intel_display *display)
+{
+	struct intel_display_irq_snapshot *snapshot;
+
+	snapshot = kzalloc(sizeof(*snapshot), GFP_ATOMIC);
+	if (!snapshot)
+		return NULL;
+
+	if (DISPLAY_VER(display) >= 6 && DISPLAY_VER(display) < 20 && !HAS_GMCH(display))
+		snapshot->derrmr = intel_de_read(display, DERRMR);
+
+	return snapshot;
+}
+
+void intel_display_irq_snapshot_print(const struct intel_display_irq_snapshot *snapshot,
+				      struct drm_printer *p)
+{
+	if (!snapshot)
+		return;
 
-	INIT_WORK(&i915->display.irq.vblank_dc_work,
-		  intel_display_vblank_dc_work);
+	drm_printf(p, "DERRMR: 0x%08x\n", snapshot->derrmr);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.h b/drivers/gpu/drm/i915/display/intel_display_irq.h
index d9867cd0a220..c66db3851da4 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.h
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.h
@@ -12,28 +12,29 @@
 
 enum pipe;
 struct drm_crtc;
-struct drm_i915_private;
+struct drm_printer;
 struct intel_display;
+struct intel_display_irq_snapshot;
 
-void valleyview_enable_display_irqs(struct drm_i915_private *i915);
-void valleyview_disable_display_irqs(struct drm_i915_private *i915);
+void valleyview_enable_display_irqs(struct intel_display *display);
+void valleyview_disable_display_irqs(struct intel_display *display);
 
-void ilk_update_display_irq(struct drm_i915_private *i915,
+void ilk_update_display_irq(struct intel_display *display,
 			    u32 interrupt_mask, u32 enabled_irq_mask);
-void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits);
-void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits);
+void ilk_enable_display_irq(struct intel_display *display, u32 bits);
+void ilk_disable_display_irq(struct intel_display *display, u32 bits);
 
-void bdw_update_port_irq(struct drm_i915_private *i915, u32 interrupt_mask, u32 enabled_irq_mask);
-void bdw_enable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
-void bdw_disable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
+void bdw_update_port_irq(struct intel_display *display, u32 interrupt_mask, u32 enabled_irq_mask);
+void bdw_enable_pipe_irq(struct intel_display *display, enum pipe pipe, u32 bits);
+void bdw_disable_pipe_irq(struct intel_display *display, enum pipe pipe, u32 bits);
 
-void ibx_display_interrupt_update(struct drm_i915_private *i915,
+void ibx_display_interrupt_update(struct intel_display *display,
 				  u32 interrupt_mask, u32 enabled_irq_mask);
-void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits);
-void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits);
+void ibx_enable_display_interrupt(struct intel_display *display, u32 bits);
+void ibx_disable_display_interrupt(struct intel_display *display, u32 bits);
 
-void gen8_irq_power_well_post_enable(struct drm_i915_private *i915, u8 pipe_mask);
-void gen8_irq_power_well_pre_disable(struct drm_i915_private *i915, u8 pipe_mask);
+void gen8_irq_power_well_post_enable(struct intel_display *display, u8 pipe_mask);
+void gen8_irq_power_well_pre_disable(struct intel_display *display, u8 pipe_mask);
 
 int i8xx_enable_vblank(struct drm_crtc *crtc);
 int i915gm_enable_vblank(struct drm_crtc *crtc);
@@ -46,41 +47,46 @@ void i965_disable_vblank(struct drm_crtc *crtc);
 void ilk_disable_vblank(struct drm_crtc *crtc);
 void bdw_disable_vblank(struct drm_crtc *crtc);
 
-void ivb_display_irq_handler(struct drm_i915_private *i915, u32 de_iir);
-void ilk_display_irq_handler(struct drm_i915_private *i915, u32 de_iir);
-void gen8_de_irq_handler(struct drm_i915_private *i915, u32 master_ctl);
-void gen11_display_irq_handler(struct drm_i915_private *i915);
+void ivb_display_irq_handler(struct intel_display *display, u32 de_iir);
+void ilk_display_irq_handler(struct intel_display *display, u32 de_iir);
+void gen8_de_irq_handler(struct intel_display *display, u32 master_ctl);
+void gen11_display_irq_handler(struct intel_display *display);
 
-u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl);
-void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir);
+u32 gen11_gu_misc_irq_ack(struct intel_display *display, const u32 master_ctl);
+void gen11_gu_misc_irq_handler(struct intel_display *display, const u32 iir);
 
-void i9xx_display_irq_reset(struct drm_i915_private *i915);
-void vlv_display_irq_reset(struct drm_i915_private *i915);
-void gen8_display_irq_reset(struct drm_i915_private *i915);
-void gen11_display_irq_reset(struct drm_i915_private *i915);
+void i9xx_display_irq_reset(struct intel_display *display);
+void ibx_display_irq_reset(struct intel_display *display);
+void vlv_display_irq_reset(struct intel_display *display);
+void gen8_display_irq_reset(struct intel_display *display);
+void gen11_display_irq_reset(struct intel_display *display);
 
-void vlv_display_irq_postinstall(struct drm_i915_private *i915);
-void ilk_de_irq_postinstall(struct drm_i915_private *i915);
-void gen8_de_irq_postinstall(struct drm_i915_private *i915);
-void gen11_de_irq_postinstall(struct drm_i915_private *i915);
-void dg1_de_irq_postinstall(struct drm_i915_private *i915);
+void i915_display_irq_postinstall(struct intel_display *display);
+void i965_display_irq_postinstall(struct intel_display *display);
+void vlv_display_irq_postinstall(struct intel_display *display);
+void ilk_de_irq_postinstall(struct intel_display *display);
+void gen8_de_irq_postinstall(struct intel_display *display);
+void gen11_de_irq_postinstall(struct intel_display *display);
+void dg1_de_irq_postinstall(struct intel_display *display);
 
 u32 i915_pipestat_enable_mask(struct intel_display *display, enum pipe pipe);
-void i915_enable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask);
-void i915_disable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask);
-void i915_enable_asle_pipestat(struct drm_i915_private *i915);
+void i915_enable_pipestat(struct intel_display *display, enum pipe pipe, u32 status_mask);
+void i915_disable_pipestat(struct intel_display *display, enum pipe pipe, u32 status_mask);
 
-void i9xx_pipestat_irq_ack(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
+void i9xx_pipestat_irq_ack(struct intel_display *display, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
 
-void i915_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
-void i965_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
-void valleyview_pipestat_irq_handler(struct drm_i915_private *i915, u32 pipe_stats[I915_MAX_PIPES]);
+void i915_pipestat_irq_handler(struct intel_display *display, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
+void i965_pipestat_irq_handler(struct intel_display *display, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
+void valleyview_pipestat_irq_handler(struct intel_display *display, u32 pipe_stats[I915_MAX_PIPES]);
 
 void vlv_display_error_irq_ack(struct intel_display *display, u32 *eir, u32 *dpinvgtt);
 void vlv_display_error_irq_handler(struct intel_display *display, u32 eir, u32 dpinvgtt);
 
-void intel_display_irq_init(struct drm_i915_private *i915);
+void intel_display_irq_init(struct intel_display *display);
 
-void i915gm_irq_cstate_wa(struct drm_i915_private *i915, bool enable);
+void i915gm_irq_cstate_wa(struct intel_display *display, bool enable);
+
+struct intel_display_irq_snapshot *intel_display_irq_snapshot_capture(struct intel_display *display);
+void intel_display_irq_snapshot_print(const struct intel_display_irq_snapshot *snapshot, struct drm_printer *p);
 
 #endif /* __INTEL_DISPLAY_IRQ_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index f7171e6932dc..16356523816f 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -16,6 +16,7 @@
 #include "intel_display_power.h"
 #include "intel_display_power_map.h"
 #include "intel_display_power_well.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dmc.h"
 #include "intel_mchbar_regs.h"
@@ -204,7 +205,7 @@ static bool __intel_display_power_is_enabled(struct intel_display *display,
 	struct i915_power_well *power_well;
 	bool is_enabled;
 
-	if (pm_runtime_suspended(display->drm->dev))
+	if (intel_display_rpm_suspended(display))
 		return false;
 
 	is_enabled = true;
@@ -322,6 +323,35 @@ unlock:
 	mutex_unlock(&power_domains->lock);
 }
 
+/**
+ * intel_display_power_get_current_dc_state - Set target dc state.
+ * @display: display device
+ *
+ * This function set the "DC off" power well target_dc_state,
+ * based upon this target_dc_stste, "DC off" power well will
+ * enable desired DC state.
+ */
+u32 intel_display_power_get_current_dc_state(struct intel_display *display)
+{
+	struct i915_power_well *power_well;
+	struct i915_power_domains *power_domains = &display->power.domains;
+	u32 current_dc_state = DC_STATE_DISABLE;
+
+	mutex_lock(&power_domains->lock);
+	power_well = lookup_power_well(display, SKL_DISP_DC_OFF);
+
+	if (drm_WARN_ON(display->drm, !power_well))
+		goto unlock;
+
+	current_dc_state = intel_power_well_is_enabled(display, power_well) ?
+		DC_STATE_DISABLE : power_domains->target_dc_state;
+
+unlock:
+	mutex_unlock(&power_domains->lock);
+
+	return current_dc_state;
+}
+
 static void __async_put_domains_mask(struct i915_power_domains *power_domains,
 				     struct intel_power_domain_mask *mask)
 {
@@ -455,7 +485,6 @@ static bool
 intel_display_power_grab_async_put_ref(struct intel_display *display,
 				       enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
 	struct intel_power_domain_mask async_put_mask;
 	bool ret = false;
@@ -473,8 +502,8 @@ intel_display_power_grab_async_put_ref(struct intel_display *display,
 		goto out_verify;
 
 	cancel_async_put_work(power_domains, false);
-	intel_runtime_pm_put_raw(&dev_priv->runtime_pm,
-				 fetch_and_zero(&power_domains->async_put_wakeref));
+	intel_display_rpm_put_raw(display,
+				  fetch_and_zero(&power_domains->async_put_wakeref));
 out_verify:
 	verify_async_put_domains_state(power_domains);
 
@@ -512,9 +541,10 @@ __intel_display_power_get_domain(struct intel_display *display,
 intel_wakeref_t intel_display_power_get(struct intel_display *display,
 					enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
-	intel_wakeref_t wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	struct ref_tracker *wakeref;
+
+	wakeref = intel_display_rpm_get(display);
 
 	mutex_lock(&power_domains->lock);
 	__intel_display_power_get_domain(display, domain);
@@ -539,12 +569,11 @@ intel_wakeref_t
 intel_display_power_get_if_enabled(struct intel_display *display,
 				   enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	bool is_enabled;
 
-	wakeref = intel_runtime_pm_get_if_in_use(&dev_priv->runtime_pm);
+	wakeref = intel_display_rpm_get_if_in_use(display);
 	if (!wakeref)
 		return NULL;
 
@@ -560,7 +589,7 @@ intel_display_power_get_if_enabled(struct intel_display *display,
 	mutex_unlock(&power_domains->lock);
 
 	if (!is_enabled) {
-		intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+		intel_display_rpm_put(display, wakeref);
 		wakeref = NULL;
 	}
 
@@ -623,12 +652,10 @@ release_async_put_domains(struct i915_power_domains *power_domains,
 	struct intel_display *display = container_of(power_domains,
 						     struct intel_display,
 						     power.domains);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-	struct intel_runtime_pm *rpm = &dev_priv->runtime_pm;
 	enum intel_display_power_domain domain;
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 
-	wakeref = intel_runtime_pm_get_noresume(rpm);
+	wakeref = intel_display_rpm_get_noresume(display);
 
 	for_each_power_domain(domain, mask) {
 		/* Clear before put, so put's sanity check is happy. */
@@ -636,7 +663,7 @@ release_async_put_domains(struct i915_power_domains *power_domains,
 		__intel_display_power_put_domain(display, domain);
 	}
 
-	intel_runtime_pm_put(rpm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 }
 
 static void
@@ -644,11 +671,10 @@ intel_display_power_put_async_work(struct work_struct *work)
 {
 	struct intel_display *display = container_of(work, struct intel_display,
 						     power.domains.async_put_work.work);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
-	struct intel_runtime_pm *rpm = &dev_priv->runtime_pm;
-	intel_wakeref_t new_work_wakeref = intel_runtime_pm_get_raw(rpm);
-	intel_wakeref_t old_work_wakeref = NULL;
+	struct ref_tracker *new_work_wakeref, *old_work_wakeref = NULL;
+
+	new_work_wakeref = intel_display_rpm_get_raw(display);
 
 	mutex_lock(&power_domains->lock);
 
@@ -688,9 +714,9 @@ out_verify:
 	mutex_unlock(&power_domains->lock);
 
 	if (old_work_wakeref)
-		intel_runtime_pm_put_raw(rpm, old_work_wakeref);
+		intel_display_rpm_put_raw(display, old_work_wakeref);
 	if (new_work_wakeref)
-		intel_runtime_pm_put_raw(rpm, new_work_wakeref);
+		intel_display_rpm_put_raw(display, new_work_wakeref);
 }
 
 /**
@@ -711,10 +737,10 @@ void __intel_display_power_put_async(struct intel_display *display,
 				     intel_wakeref_t wakeref,
 				     int delay_ms)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
-	struct intel_runtime_pm *rpm = &i915->runtime_pm;
-	intel_wakeref_t work_wakeref = intel_runtime_pm_get_raw(rpm);
+	struct ref_tracker *work_wakeref;
+
+	work_wakeref = intel_display_rpm_get_raw(display);
 
 	delay_ms = delay_ms >= 0 ? delay_ms : 100;
 
@@ -746,9 +772,9 @@ out_verify:
 	mutex_unlock(&power_domains->lock);
 
 	if (work_wakeref)
-		intel_runtime_pm_put_raw(rpm, work_wakeref);
+		intel_display_rpm_put_raw(display, work_wakeref);
 
-	intel_runtime_pm_put(rpm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 }
 
 /**
@@ -765,7 +791,6 @@ out_verify:
  */
 void intel_display_power_flush_work(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
 	struct intel_power_domain_mask async_put_mask;
 	intel_wakeref_t work_wakeref;
@@ -786,7 +811,7 @@ out_verify:
 	mutex_unlock(&power_domains->lock);
 
 	if (work_wakeref)
-		intel_runtime_pm_put_raw(&i915->runtime_pm, work_wakeref);
+		intel_display_rpm_put_raw(display, work_wakeref);
 }
 
 /**
@@ -824,10 +849,8 @@ void intel_display_power_put(struct intel_display *display,
 			     enum intel_display_power_domain domain,
 			     intel_wakeref_t wakeref)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	__intel_display_power_put(display, domain);
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 }
 #else
 /**
@@ -846,10 +869,8 @@ void intel_display_power_put(struct intel_display *display,
 void intel_display_power_put_unchecked(struct intel_display *display,
 				       enum intel_display_power_domain domain)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	__intel_display_power_put(display, domain);
-	intel_runtime_pm_put_unchecked(&dev_priv->runtime_pm);
+	intel_display_rpm_put_unchecked(display);
 }
 #endif
 
@@ -1373,26 +1394,24 @@ static void hsw_restore_lcpll(struct intel_display *display)
  */
 static void hsw_enable_pc8(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	drm_dbg_kms(display->drm, "Enabling package C8+\n");
 
-	if (HAS_PCH_LPT_LP(dev_priv))
+	if (HAS_PCH_LPT_LP(display))
 		intel_de_rmw(display, SOUTH_DSPCLK_GATE_D,
 			     PCH_LP_PARTITION_LEVEL_DISABLE, 0);
 
-	lpt_disable_clkout_dp(dev_priv);
+	lpt_disable_clkout_dp(display);
 	hsw_disable_lcpll(display, true, true);
 }
 
 static void hsw_disable_pc8(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	struct drm_i915_private __maybe_unused *dev_priv = to_i915(display->drm);
 
 	drm_dbg_kms(display->drm, "Disabling package C8+\n");
 
 	hsw_restore_lcpll(display);
-	intel_init_pch_refclk(dev_priv);
+	intel_init_pch_refclk(display);
 
 	/* Many display registers don't survive PC8+ */
 #ifdef I915 /* FIXME */
@@ -1423,14 +1442,13 @@ static void intel_pch_reset_handshake(struct intel_display *display,
 static void skl_display_core_init(struct intel_display *display,
 				  bool resume)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
 	struct i915_power_well *well;
 
 	gen9_set_dc_state(display, DC_STATE_DISABLE);
 
 	/* enable PCH reset handshake */
-	intel_pch_reset_handshake(display, !HAS_PCH_NOP(dev_priv));
+	intel_pch_reset_handshake(display, !HAS_PCH_NOP(display));
 
 	if (!HAS_DISPLAY(display))
 		return;
@@ -1632,20 +1650,19 @@ static void tgl_bw_buddy_init(struct intel_display *display)
 static void icl_display_core_init(struct intel_display *display,
 				  bool resume)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
 	struct i915_power_well *well;
 
 	gen9_set_dc_state(display, DC_STATE_DISABLE);
 
 	/* Wa_14011294188:ehl,jsl,tgl,rkl,adl-s */
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_TGP &&
-	    INTEL_PCH_TYPE(dev_priv) < PCH_DG1)
+	if (INTEL_PCH_TYPE(display) >= PCH_TGP &&
+	    INTEL_PCH_TYPE(display) < PCH_DG1)
 		intel_de_rmw(display, SOUTH_DSPCLK_GATE_D, 0,
 			     PCH_DPMGUNIT_CLOCK_GATE_DISABLE);
 
 	/* 1. Enable PCH reset handshake. */
-	intel_pch_reset_handshake(display, !HAS_PCH_NOP(dev_priv));
+	intel_pch_reset_handshake(display, !HAS_PCH_NOP(display));
 
 	if (!HAS_DISPLAY(display))
 		return;
@@ -1916,7 +1933,6 @@ static void intel_power_domains_verify_state(struct intel_display *display);
  */
 void intel_power_domains_init_hw(struct intel_display *display, bool resume)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
 
 	power_domains->initializing = true;
@@ -1940,9 +1956,9 @@ void intel_power_domains_init_hw(struct intel_display *display, bool resume)
 		assert_isp_power_gated(display);
 	} else if (display->platform.broadwell || display->platform.haswell) {
 		hsw_assert_cdclk(display);
-		intel_pch_reset_handshake(display, !HAS_PCH_NOP(i915));
+		intel_pch_reset_handshake(display, !HAS_PCH_NOP(display));
 	} else if (display->platform.ivybridge) {
-		intel_pch_reset_handshake(display, !HAS_PCH_NOP(i915));
+		intel_pch_reset_handshake(display, !HAS_PCH_NOP(display));
 	}
 
 	/*
@@ -1979,7 +1995,6 @@ void intel_power_domains_init_hw(struct intel_display *display, bool resume)
  */
 void intel_power_domains_driver_remove(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	intel_wakeref_t wakeref __maybe_unused =
 		fetch_and_zero(&display->power.domains.init_wakeref);
 
@@ -1993,7 +2008,7 @@ void intel_power_domains_driver_remove(struct intel_display *display)
 	intel_power_domains_verify_state(display);
 
 	/* Keep the power well enabled, but cancel its rpm wakeref. */
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 }
 
 /**
@@ -2238,8 +2253,6 @@ static void intel_power_domains_verify_state(struct intel_display *display)
 
 void intel_display_power_suspend_late(struct intel_display *display, bool s2idle)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	intel_power_domains_suspend(display, s2idle);
 
 	if (DISPLAY_VER(display) >= 11 || display->platform.geminilake ||
@@ -2250,14 +2263,12 @@ void intel_display_power_suspend_late(struct intel_display *display, bool s2idle
 	}
 
 	/* Tweaked Wa_14010685332:cnp,icp,jsp,mcc,tgp,adp */
-	if (INTEL_PCH_TYPE(i915) >= PCH_CNP && INTEL_PCH_TYPE(i915) < PCH_DG1)
-		intel_de_rmw(i915, SOUTH_CHICKEN1, SBCLK_RUN_REFCLK_DIS, SBCLK_RUN_REFCLK_DIS);
+	if (INTEL_PCH_TYPE(display) >= PCH_CNP && INTEL_PCH_TYPE(display) < PCH_DG1)
+		intel_de_rmw(display, SOUTH_CHICKEN1, SBCLK_RUN_REFCLK_DIS, SBCLK_RUN_REFCLK_DIS);
 }
 
 void intel_display_power_resume_early(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (DISPLAY_VER(display) >= 11 || display->platform.geminilake ||
 	    display->platform.broxton) {
 		gen9_sanitize_dc_state(display);
@@ -2267,8 +2278,8 @@ void intel_display_power_resume_early(struct intel_display *display)
 	}
 
 	/* Tweaked Wa_14010685332:cnp,icp,jsp,mcc,tgp,adp */
-	if (INTEL_PCH_TYPE(i915) >= PCH_CNP && INTEL_PCH_TYPE(i915) < PCH_DG1)
-		intel_de_rmw(i915, SOUTH_CHICKEN1, SBCLK_RUN_REFCLK_DIS, 0);
+	if (INTEL_PCH_TYPE(display) >= PCH_CNP && INTEL_PCH_TYPE(display) < PCH_DG1)
+		intel_de_rmw(display, SOUTH_CHICKEN1, SBCLK_RUN_REFCLK_DIS, 0);
 
 	intel_power_domains_resume(display);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.h b/drivers/gpu/drm/i915/display/intel_display_power.h
index 1b53d67f9b60..f8813b0e16df 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.h
+++ b/drivers/gpu/drm/i915/display/intel_display_power.h
@@ -183,6 +183,7 @@ void intel_display_power_suspend(struct intel_display *display);
 void intel_display_power_resume(struct intel_display *display);
 void intel_display_power_set_target_dc_state(struct intel_display *display,
 					     u32 state);
+u32 intel_display_power_get_current_dc_state(struct intel_display *display);
 
 bool intel_display_power_is_enabled(struct intel_display *display,
 				    enum intel_display_power_domain domain);
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c
index e80e1fd611ca..ab1163744bc5 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_map.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c
@@ -1696,6 +1696,7 @@ I915_DECL_PW_DOMAINS(xe3lpd_pwdoms_dc_off,
 	XE3LPD_PW_C_POWER_DOMAINS,
 	XE3LPD_PW_D_POWER_DOMAINS,
 	POWER_DOMAIN_AUDIO_MMIO,
+	POWER_DOMAIN_AUDIO_PLAYBACK,
 	POWER_DOMAIN_INIT);
 
 static const struct i915_power_well_desc xe3lpd_power_wells_dcoff[] = {
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 8ec87ffd87d2..b104bce0e14d 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -13,6 +13,7 @@
 #include "intel_de.h"
 #include "intel_display_irq.h"
 #include "intel_display_power_well.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dkl_phy.h"
 #include "intel_dkl_phy_regs.h"
@@ -24,6 +25,7 @@
 #include "intel_hotplug.h"
 #include "intel_pcode.h"
 #include "intel_pps.h"
+#include "intel_psr.h"
 #include "intel_tc.h"
 #include "intel_vga.h"
 #include "skl_watermark.h"
@@ -186,22 +188,18 @@ int intel_power_well_refcount(struct i915_power_well *power_well)
 static void hsw_power_well_post_enable(struct intel_display *display,
 				       u8 irq_pipe_mask, bool has_vga)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	if (has_vga)
 		intel_vga_reset_io_mem(display);
 
 	if (irq_pipe_mask)
-		gen8_irq_power_well_post_enable(dev_priv, irq_pipe_mask);
+		gen8_irq_power_well_post_enable(display, irq_pipe_mask);
 }
 
 static void hsw_power_well_pre_disable(struct intel_display *display,
 				       u8 irq_pipe_mask)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	if (irq_pipe_mask)
-		gen8_irq_power_well_pre_disable(dev_priv, irq_pipe_mask);
+		gen8_irq_power_well_pre_disable(display, irq_pipe_mask);
 }
 
 #define ICL_AUX_PW_TO_PHY(pw_idx)	\
@@ -752,8 +750,9 @@ void gen9_sanitize_dc_state(struct intel_display *display)
 void gen9_set_dc_state(struct intel_display *display, u32 state)
 {
 	struct i915_power_domains *power_domains = &display->power.domains;
-	u32 val;
+	bool dc6_was_enabled, enable_dc6;
 	u32 mask;
+	u32 val;
 
 	if (!HAS_DISPLAY(display))
 		return;
@@ -762,6 +761,9 @@ void gen9_set_dc_state(struct intel_display *display, u32 state)
 			     state & ~power_domains->allowed_dc_mask))
 		state &= power_domains->allowed_dc_mask;
 
+	if (!power_domains->initializing)
+		intel_psr_notify_dc5_dc6(display);
+
 	val = intel_de_read(display, DC_STATE_EN);
 	mask = gen9_dc_mask(display);
 	drm_dbg_kms(display->drm, "Setting DC state from %02x to %02x\n",
@@ -772,11 +774,19 @@ void gen9_set_dc_state(struct intel_display *display, u32 state)
 		drm_err(display->drm, "DC state mismatch (0x%x -> 0x%x)\n",
 			power_domains->dc_state, val & mask);
 
+	enable_dc6 = state & DC_STATE_EN_UPTO_DC6;
+	dc6_was_enabled = val & DC_STATE_EN_UPTO_DC6;
+	if (!dc6_was_enabled && enable_dc6)
+		intel_dmc_update_dc6_allowed_count(display, true);
+
 	val &= ~mask;
 	val |= state;
 
 	gen9_write_dc_state(display, val);
 
+	if (!enable_dc6 && dc6_was_enabled)
+		intel_dmc_update_dc6_allowed_count(display, false);
+
 	power_domains->dc_state = val & mask;
 }
 
@@ -816,7 +826,8 @@ static void assert_can_enable_dc5(struct intel_display *display)
 		      (intel_de_read(display, DC_STATE_EN) &
 		       DC_STATE_EN_UPTO_DC5),
 		      "DC5 already programmed to be enabled.\n");
-	assert_rpm_wakelock_held(&dev_priv->runtime_pm);
+
+	assert_display_rpm_held(display);
 
 	assert_dmc_loaded(display);
 }
@@ -1201,7 +1212,6 @@ static void vlv_init_display_clock_gating(struct intel_display *display)
 
 static void vlv_display_power_well_init(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_encoder *encoder;
 	enum pipe pipe;
 
@@ -1225,9 +1235,7 @@ static void vlv_display_power_well_init(struct intel_display *display)
 
 	vlv_init_display_clock_gating(display);
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	valleyview_enable_display_irqs(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	valleyview_enable_display_irqs(display);
 
 	/*
 	 * During driver initialization/resume we can avoid restoring the
@@ -1236,8 +1244,8 @@ static void vlv_display_power_well_init(struct intel_display *display)
 	if (display->power.domains.initializing)
 		return;
 
-	intel_hpd_init(dev_priv);
-	intel_hpd_poll_disable(dev_priv);
+	intel_hpd_init(display);
+	intel_hpd_poll_disable(display);
 
 	/* Re-enable the ADPA, if we have one */
 	for_each_intel_encoder(display->drm, encoder) {
@@ -1245,7 +1253,7 @@ static void vlv_display_power_well_init(struct intel_display *display)
 			intel_crt_reset(&encoder->base);
 	}
 
-	intel_vga_redisable_power_on(display);
+	intel_vga_disable(display);
 
 	intel_pps_unlock_regs_wa(display);
 }
@@ -1254,9 +1262,7 @@ static void vlv_display_power_well_deinit(struct intel_display *display)
 {
 	struct drm_i915_private *dev_priv = to_i915(display->drm);
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	valleyview_disable_display_irqs(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	valleyview_disable_display_irqs(display);
 
 	/* make sure we're done processing display irqs */
 	intel_synchronize_irq(dev_priv);
@@ -1265,7 +1271,7 @@ static void vlv_display_power_well_deinit(struct intel_display *display)
 
 	/* Prevent us from re-enabling polling on accident in late suspend */
 	if (!display->drm->dev->power.is_suspended)
-		intel_hpd_poll_enable(dev_priv);
+		intel_hpd_poll_enable(display);
 }
 
 static void vlv_display_power_well_enable(struct intel_display *display,
diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.c b/drivers/gpu/drm/i915/display/intel_display_reset.c
index 1f2798404f2c..1dbd3e841df3 100644
--- a/drivers/gpu/drm/i915/display/intel_display_reset.c
+++ b/drivers/gpu/drm/i915/display/intel_display_reset.c
@@ -107,14 +107,14 @@ void intel_display_reset_finish(struct intel_display *display, bool test_only)
 		intel_display_driver_init_hw(display);
 		intel_clock_gating_init(i915);
 		intel_cx0_pll_power_save_wa(display);
-		intel_hpd_init(i915);
+		intel_hpd_init(display);
 
 		ret = __intel_display_driver_resume(display, state, ctx);
 		if (ret)
 			drm_err(display->drm,
 				"Restoring old state failed with %i\n", ret);
 
-		intel_hpd_poll_disable(i915);
+		intel_hpd_poll_disable(display);
 	}
 
 	drm_atomic_state_put(state);
diff --git a/drivers/gpu/drm/i915/display/intel_display_rpm.c b/drivers/gpu/drm/i915/display/intel_display_rpm.c
new file mode 100644
index 000000000000..48da67dd0136
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_rpm.c
@@ -0,0 +1,68 @@
+// SPDX-License-Identifier: MIT
+/* Copyright © 2025 Intel Corporation */
+
+#include "i915_drv.h"
+#include "intel_display_rpm.h"
+#include "intel_runtime_pm.h"
+
+static struct intel_runtime_pm *display_to_rpm(struct intel_display *display)
+{
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
+	return &i915->runtime_pm;
+}
+
+struct ref_tracker *intel_display_rpm_get_raw(struct intel_display *display)
+{
+	return intel_runtime_pm_get_raw(display_to_rpm(display));
+}
+
+void intel_display_rpm_put_raw(struct intel_display *display, struct ref_tracker *wakeref)
+{
+	intel_runtime_pm_put_raw(display_to_rpm(display), wakeref);
+}
+
+struct ref_tracker *intel_display_rpm_get(struct intel_display *display)
+{
+	return intel_runtime_pm_get(display_to_rpm(display));
+}
+
+struct ref_tracker *intel_display_rpm_get_if_in_use(struct intel_display *display)
+{
+	return intel_runtime_pm_get_if_in_use(display_to_rpm(display));
+}
+
+struct ref_tracker *intel_display_rpm_get_noresume(struct intel_display *display)
+{
+	return intel_runtime_pm_get_noresume(display_to_rpm(display));
+}
+
+void intel_display_rpm_put(struct intel_display *display, struct ref_tracker *wakeref)
+{
+	intel_runtime_pm_put(display_to_rpm(display), wakeref);
+}
+
+void intel_display_rpm_put_unchecked(struct intel_display *display)
+{
+	intel_runtime_pm_put_unchecked(display_to_rpm(display));
+}
+
+bool intel_display_rpm_suspended(struct intel_display *display)
+{
+	return intel_runtime_pm_suspended(display_to_rpm(display));
+}
+
+void assert_display_rpm_held(struct intel_display *display)
+{
+	assert_rpm_wakelock_held(display_to_rpm(display));
+}
+
+void intel_display_rpm_assert_block(struct intel_display *display)
+{
+	disable_rpm_wakeref_asserts(display_to_rpm(display));
+}
+
+void intel_display_rpm_assert_unblock(struct intel_display *display)
+{
+	enable_rpm_wakeref_asserts(display_to_rpm(display));
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_rpm.h b/drivers/gpu/drm/i915/display/intel_display_rpm.h
new file mode 100644
index 000000000000..6ef48515f84b
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_rpm.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __INTEL_DISPLAY_RPM__
+#define __INTEL_DISPLAY_RPM__
+
+#include <linux/types.h>
+
+struct intel_display;
+struct ref_tracker;
+
+struct ref_tracker *intel_display_rpm_get(struct intel_display *display);
+void intel_display_rpm_put(struct intel_display *display, struct ref_tracker *wakeref);
+
+#define __with_intel_display_rpm(__display, __wakeref) \
+	for (struct ref_tracker *(__wakeref) = intel_display_rpm_get(__display); (__wakeref); \
+	     intel_display_rpm_put((__display), (__wakeref)), (__wakeref) = NULL)
+
+#define with_intel_display_rpm(__display) \
+	__with_intel_display_rpm((__display), __UNIQUE_ID(wakeref))
+
+/* Only for special cases. */
+bool intel_display_rpm_suspended(struct intel_display *display);
+
+void assert_display_rpm_held(struct intel_display *display);
+void intel_display_rpm_assert_block(struct intel_display *display);
+void intel_display_rpm_assert_unblock(struct intel_display *display);
+
+/* Only for display power implementation. */
+struct ref_tracker *intel_display_rpm_get_raw(struct intel_display *display);
+void intel_display_rpm_put_raw(struct intel_display *display, struct ref_tracker *wakeref);
+
+struct ref_tracker *intel_display_rpm_get_if_in_use(struct intel_display *display);
+struct ref_tracker *intel_display_rpm_get_noresume(struct intel_display *display);
+void intel_display_rpm_put_unchecked(struct intel_display *display);
+
+#endif /* __INTEL_DISPLAY_RPM__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_rps.c b/drivers/gpu/drm/i915/display/intel_display_rps.c
index 4074a1879828..678b24115951 100644
--- a/drivers/gpu/drm/i915/display/intel_display_rps.c
+++ b/drivers/gpu/drm/i915/display/intel_display_rps.c
@@ -8,6 +8,8 @@
 
 #include "gt/intel_rps.h"
 #include "i915_drv.h"
+#include "i915_reg.h"
+#include "intel_display_irq.h"
 #include "intel_display_rps.h"
 #include "intel_display_types.h"
 
@@ -81,3 +83,24 @@ void intel_display_rps_mark_interactive(struct intel_display *display,
 	intel_rps_mark_interactive(&to_gt(i915)->rps, interactive);
 	state->rps_interactive = interactive;
 }
+
+void ilk_display_rps_enable(struct intel_display *display)
+{
+	spin_lock(&display->irq.lock);
+	ilk_enable_display_irq(display, DE_PCU_EVENT);
+	spin_unlock(&display->irq.lock);
+}
+
+void ilk_display_rps_disable(struct intel_display *display)
+{
+	spin_lock(&display->irq.lock);
+	ilk_disable_display_irq(display, DE_PCU_EVENT);
+	spin_unlock(&display->irq.lock);
+}
+
+void ilk_display_rps_irq_handler(struct intel_display *display)
+{
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
+	gen5_rps_irq_handler(&to_gt(i915)->rps);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_rps.h b/drivers/gpu/drm/i915/display/intel_display_rps.h
index 556891edb2dd..183d154f2c7c 100644
--- a/drivers/gpu/drm/i915/display/intel_display_rps.h
+++ b/drivers/gpu/drm/i915/display/intel_display_rps.h
@@ -13,10 +13,34 @@ struct drm_crtc;
 struct intel_atomic_state;
 struct intel_display;
 
+#ifdef I915
 void intel_display_rps_boost_after_vblank(struct drm_crtc *crtc,
 					  struct dma_fence *fence);
 void intel_display_rps_mark_interactive(struct intel_display *display,
 					struct intel_atomic_state *state,
 					bool interactive);
+void ilk_display_rps_enable(struct intel_display *display);
+void ilk_display_rps_disable(struct intel_display *display);
+void ilk_display_rps_irq_handler(struct intel_display *display);
+#else
+static inline void intel_display_rps_boost_after_vblank(struct drm_crtc *crtc,
+							struct dma_fence *fence)
+{
+}
+static inline void intel_display_rps_mark_interactive(struct intel_display *display,
+						      struct intel_atomic_state *state,
+						      bool interactive)
+{
+}
+static inline void ilk_display_rps_enable(struct intel_display *display)
+{
+}
+static inline void ilk_display_rps_disable(struct intel_display *display)
+{
+}
+static inline void ilk_display_rps_irq_handler(struct intel_display *display)
+{
+}
+#endif
 
 #endif /* __INTEL_DISPLAY_RPS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_snapshot.c b/drivers/gpu/drm/i915/display/intel_display_snapshot.c
index 25ba043cbb65..66087302fdbc 100644
--- a/drivers/gpu/drm/i915/display/intel_display_snapshot.c
+++ b/drivers/gpu/drm/i915/display/intel_display_snapshot.c
@@ -7,6 +7,7 @@
 
 #include "intel_display_core.h"
 #include "intel_display_device.h"
+#include "intel_display_irq.h"
 #include "intel_display_params.h"
 #include "intel_display_snapshot.h"
 #include "intel_dmc.h"
@@ -20,6 +21,7 @@ struct intel_display_snapshot {
 	struct intel_display_params params;
 	struct intel_overlay_snapshot *overlay;
 	struct intel_dmc_snapshot *dmc;
+	struct intel_display_irq_snapshot *irq;
 };
 
 struct intel_display_snapshot *intel_display_snapshot_capture(struct intel_display *display)
@@ -38,6 +40,7 @@ struct intel_display_snapshot *intel_display_snapshot_capture(struct intel_displ
 
 	intel_display_params_copy(&snapshot->params);
 
+	snapshot->irq = intel_display_irq_snapshot_capture(display);
 	snapshot->overlay = intel_overlay_snapshot_capture(display);
 	snapshot->dmc = intel_dmc_snapshot_capture(display);
 
@@ -57,6 +60,7 @@ void intel_display_snapshot_print(const struct intel_display_snapshot *snapshot,
 	intel_display_device_info_print(&snapshot->info, &snapshot->runtime_info, p);
 	intel_display_params_dump(&snapshot->params, display->drm->driver->name, p);
 
+	intel_display_irq_snapshot_print(snapshot->irq, p);
 	intel_overlay_snapshot_print(snapshot->overlay, p);
 	intel_dmc_snapshot_print(snapshot->dmc, p);
 }
@@ -68,6 +72,7 @@ void intel_display_snapshot_free(struct intel_display_snapshot *snapshot)
 
 	intel_display_params_free(&snapshot->params);
 
+	kfree(snapshot->irq);
 	kfree(snapshot->overlay);
 	kfree(snapshot->dmc);
 	kfree(snapshot);
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 99a6fd2900b9..d6d0440dcee9 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -581,7 +581,7 @@ struct dpll {
 struct intel_atomic_state {
 	struct drm_atomic_state base;
 
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 
 	struct __intel_global_objs_state *global_objs;
 	int num_global_objs;
@@ -1114,6 +1114,7 @@ struct intel_crtc_state {
 	bool wm_level_disabled;
 	u32 dc3co_exitline;
 	u16 su_y_granularity;
+	u8 active_non_psr_pipes;
 
 	/*
 	 * Frequency the dpll for the port should run at. Differs from the
@@ -1387,7 +1388,7 @@ struct intel_crtc {
 	/* armed event for DSB based updates */
 	struct drm_pending_vblank_event *dsb_event;
 
-	/* Access to these should be protected by dev_priv->irq_lock. */
+	/* Access to these should be protected by display->irq.lock. */
 	bool cpu_fifo_underrun_disabled;
 	bool pch_fifo_underrun_disabled;
 
@@ -1439,7 +1440,7 @@ struct intel_crtc {
 	struct intel_pipe_crc pipe_crc;
 #endif
 
-	bool block_dc_for_vblank;
+	bool vblank_psr_notify;
 };
 
 struct intel_plane_error {
@@ -1620,7 +1621,7 @@ struct intel_psr {
 	bool sink_support;
 	bool source_support;
 	bool enabled;
-	bool paused;
+	int pause_counter;
 	enum pipe pipe;
 	enum transcoder transcoder;
 	bool active;
@@ -1650,6 +1651,8 @@ struct intel_psr {
 	u8 entry_setup_frames;
 
 	bool link_ok;
+
+	u8 active_non_psr_pipes;
 };
 
 struct intel_dp {
@@ -1658,7 +1661,6 @@ struct intel_dp {
 	int link_rate;
 	u8 lane_count;
 	u8 sink_count;
-	bool link_trained;
 	bool needs_modeset_retry;
 	bool use_max_params;
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
@@ -1683,6 +1685,7 @@ struct intel_dp {
 	int common_rates[DP_MAX_SUPPORTED_RATES];
 	struct {
 		/* TODO: move the rest of link specific fields to here */
+		bool active;
 		/* common rate,lane_count configs in bw order */
 		int num_configs;
 #define INTEL_DP_MAX_LANE_COUNT			4
@@ -1739,7 +1742,7 @@ struct intel_dp {
 	struct {
 		struct intel_dp_mst_encoder *stream_encoders[I915_MAX_PIPES];
 		struct drm_dp_mst_topology_mgr mgr;
-		int active_links;
+		int active_streams;
 	} mst;
 
 	u32 (*get_aux_clock_divider)(struct intel_dp *dp, int index);
@@ -1805,12 +1808,16 @@ struct intel_dp {
 	struct {
 		u8 io_wake_lines;
 		u8 fast_wake_lines;
+		enum transcoder transcoder;
+		struct mutex lock;
 
 		/* LNL and beyond */
 		u8 check_entry_lines;
 		u8 aux_less_wake_lines;
 		u8 silence_period_sym_clocks;
 		u8 lfps_half_cycle_num_of_syms;
+		bool lobf_disable_debug;
+		bool sink_alpm_error;
 	} alpm_parameters;
 
 	u8 alpm_dpcd;
diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.c b/drivers/gpu/drm/i915/display/intel_display_wa.c
index e5a8022db664..da429c332914 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.c
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.c
@@ -3,38 +3,38 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_de.h"
+#include "intel_display_core.h"
 #include "intel_display_wa.h"
 
-static void gen11_display_wa_apply(struct drm_i915_private *i915)
+static void gen11_display_wa_apply(struct intel_display *display)
 {
 	/* Wa_14010594013 */
-	intel_de_rmw(i915, GEN8_CHICKEN_DCPR_1, 0, ICL_DELAY_PMRSP);
+	intel_de_rmw(display, GEN8_CHICKEN_DCPR_1, 0, ICL_DELAY_PMRSP);
 }
 
-static void xe_d_display_wa_apply(struct drm_i915_private *i915)
+static void xe_d_display_wa_apply(struct intel_display *display)
 {
 	/* Wa_14013723622 */
-	intel_de_rmw(i915, CLKREQ_POLICY, CLKREQ_POLICY_MEM_UP_OVRD, 0);
+	intel_de_rmw(display, CLKREQ_POLICY, CLKREQ_POLICY_MEM_UP_OVRD, 0);
 }
 
-static void adlp_display_wa_apply(struct drm_i915_private *i915)
+static void adlp_display_wa_apply(struct intel_display *display)
 {
 	/* Wa_22011091694:adlp */
-	intel_de_rmw(i915, GEN9_CLKGATE_DIS_5, 0, DPCE_GATING_DIS);
+	intel_de_rmw(display, GEN9_CLKGATE_DIS_5, 0, DPCE_GATING_DIS);
 
 	/* Bspec/49189 Initialize Sequence */
-	intel_de_rmw(i915, GEN8_CHICKEN_DCPR_1, DDI_CLOCK_REG_ACCESS, 0);
+	intel_de_rmw(display, GEN8_CHICKEN_DCPR_1, DDI_CLOCK_REG_ACCESS, 0);
 }
 
-void intel_display_wa_apply(struct drm_i915_private *i915)
+void intel_display_wa_apply(struct intel_display *display)
 {
-	if (IS_ALDERLAKE_P(i915))
-		adlp_display_wa_apply(i915);
-	else if (DISPLAY_VER(i915) == 12)
-		xe_d_display_wa_apply(i915);
-	else if (DISPLAY_VER(i915) == 11)
-		gen11_display_wa_apply(i915);
+	if (display->platform.alderlake_p)
+		adlp_display_wa_apply(display);
+	else if (DISPLAY_VER(display) == 12)
+		xe_d_display_wa_apply(display);
+	else if (DISPLAY_VER(display) == 11)
+		gen11_display_wa_apply(display);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.h b/drivers/gpu/drm/i915/display/intel_display_wa.h
index be644ab6ae00..babd9d16603d 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.h
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.h
@@ -8,14 +8,17 @@
 
 #include <linux/types.h>
 
-struct drm_i915_private;
+struct intel_display;
 
-void intel_display_wa_apply(struct drm_i915_private *i915);
+void intel_display_wa_apply(struct intel_display *display);
 
 #ifdef I915
-static inline bool intel_display_needs_wa_16023588340(struct drm_i915_private *i915) { return false; }
+static inline bool intel_display_needs_wa_16023588340(struct intel_display *display)
+{
+	return false;
+}
 #else
-bool intel_display_needs_wa_16023588340(struct drm_i915_private *i915);
+bool intel_display_needs_wa_16023588340(struct intel_display *display);
 #endif
 
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy.c b/drivers/gpu/drm/i915/display/intel_dkl_phy.c
index 0813fb9b5823..dad7192132ad 100644
--- a/drivers/gpu/drm/i915/display/intel_dkl_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_dkl_phy.c
@@ -4,6 +4,7 @@
  */
 
 #include <drm/drm_device.h>
+#include <drm/drm_print.h>
 
 #include "intel_de.h"
 #include "intel_display.h"
diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c
index fa6944e55d95..b58189d24e7e 100644
--- a/drivers/gpu/drm/i915/display/intel_dmc.c
+++ b/drivers/gpu/drm/i915/display/intel_dmc.c
@@ -28,6 +28,8 @@
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_de.h"
+#include "intel_display_rpm.h"
+#include "intel_display_power_well.h"
 #include "intel_dmc.h"
 #include "intel_dmc_regs.h"
 #include "intel_step.h"
@@ -57,6 +59,10 @@ struct intel_dmc {
 	const char *fw_path;
 	u32 max_fw_size; /* bytes */
 	u32 version;
+	struct {
+		u32 dc5_start;
+		u32 count;
+	} dc6_allowed;
 	struct dmc_fw_info {
 		u32 mmio_count;
 		i915_reg_t mmioaddr[20];
@@ -167,7 +173,6 @@ MODULE_FIRMWARE(BXT_DMC_PATH);
 
 static const char *dmc_firmware_default(struct intel_display *display, u32 *size)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	const char *fw_path = NULL;
 	u32 max_fw_size = 0;
 
@@ -183,39 +188,39 @@ static const char *dmc_firmware_default(struct intel_display *display, u32 *size
 	} else if (DISPLAY_VERx100(display) == 1400) {
 		fw_path = MTL_DMC_PATH;
 		max_fw_size = XELPDP_DMC_MAX_FW_SIZE;
-	} else if (IS_DG2(i915)) {
+	} else if (display->platform.dg2) {
 		fw_path = DG2_DMC_PATH;
 		max_fw_size = DISPLAY_VER13_DMC_MAX_FW_SIZE;
-	} else if (IS_ALDERLAKE_P(i915)) {
+	} else if (display->platform.alderlake_p) {
 		fw_path = ADLP_DMC_PATH;
 		max_fw_size = DISPLAY_VER13_DMC_MAX_FW_SIZE;
-	} else if (IS_ALDERLAKE_S(i915)) {
+	} else if (display->platform.alderlake_s) {
 		fw_path = ADLS_DMC_PATH;
 		max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE;
-	} else if (IS_DG1(i915)) {
+	} else if (display->platform.dg1) {
 		fw_path = DG1_DMC_PATH;
 		max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE;
-	} else if (IS_ROCKETLAKE(i915)) {
+	} else if (display->platform.rocketlake) {
 		fw_path = RKL_DMC_PATH;
 		max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE;
-	} else if (IS_TIGERLAKE(i915)) {
+	} else if (display->platform.tigerlake) {
 		fw_path = TGL_DMC_PATH;
 		max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE;
 	} else if (DISPLAY_VER(display) == 11) {
 		fw_path = ICL_DMC_PATH;
 		max_fw_size = ICL_DMC_MAX_FW_SIZE;
-	} else if (IS_GEMINILAKE(i915)) {
+	} else if (display->platform.geminilake) {
 		fw_path = GLK_DMC_PATH;
 		max_fw_size = GLK_DMC_MAX_FW_SIZE;
-	} else if (IS_KABYLAKE(i915) ||
-		   IS_COFFEELAKE(i915) ||
-		   IS_COMETLAKE(i915)) {
+	} else if (display->platform.kabylake ||
+		   display->platform.coffeelake ||
+		   display->platform.cometlake) {
 		fw_path = KBL_DMC_PATH;
 		max_fw_size = KBL_DMC_MAX_FW_SIZE;
-	} else if (IS_SKYLAKE(i915)) {
+	} else if (display->platform.skylake) {
 		fw_path = SKL_DMC_PATH;
 		max_fw_size = SKL_DMC_MAX_FW_SIZE;
-	} else if (IS_BROXTON(i915)) {
+	} else if (display->platform.broxton) {
 		fw_path = BXT_DMC_PATH;
 		max_fw_size = BXT_DMC_MAX_FW_SIZE;
 	}
@@ -511,6 +516,54 @@ void intel_dmc_disable_pipe(struct intel_display *display, enum pipe pipe)
 		intel_de_rmw(display, PIPEDMC_CONTROL(pipe), PIPEDMC_ENABLE, 0);
 }
 
+/**
+ * intel_dmc_block_pkgc() - block PKG C-state
+ * @display: display instance
+ * @pipe: pipe which register use to block
+ * @block: block/unblock
+ *
+ * This interface is target for Wa_16025596647 usage. I.e. to set/clear
+ * PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS bit in PIPEDMC_BLOCK_PKGC_SW register.
+ */
+void intel_dmc_block_pkgc(struct intel_display *display, enum pipe pipe,
+			  bool block)
+{
+	intel_de_rmw(display, PIPEDMC_BLOCK_PKGC_SW(pipe),
+		     PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS, block ?
+		     PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS : 0);
+}
+
+/**
+ * intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank() - start of PKG
+ * C-state exit
+ * @display: display instance
+ * @pipe: pipe which register use to block
+ * @enable: enable/disable
+ *
+ * This interface is target for Wa_16025596647 usage. I.e. start the package C
+ * exit at the start of the undelayed vblank
+ */
+void intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(struct intel_display *display,
+							    enum pipe pipe, bool enable)
+{
+	u32 val;
+
+	if (enable)
+		val = DMC_EVT_CTL_ENABLE | DMC_EVT_CTL_RECURRING |
+			REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK,
+				       DMC_EVT_CTL_TYPE_EDGE_0_1) |
+			REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK,
+				       DMC_EVT_CTL_EVENT_ID_VBLANK_A);
+	else
+		val = REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK,
+				     DMC_EVT_CTL_EVENT_ID_FALSE) |
+			REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK,
+				       DMC_EVT_CTL_TYPE_EDGE_0_1);
+
+	intel_de_write(display, MTL_PIPEDMC_EVT_CTL_4(pipe),
+		       val);
+}
+
 static bool is_dmc_evt_ctl_reg(struct intel_display *display,
 			       enum intel_dmc_id dmc_id, i915_reg_t reg)
 {
@@ -535,8 +588,6 @@ static bool disable_dmc_evt(struct intel_display *display,
 			    enum intel_dmc_id dmc_id,
 			    i915_reg_t reg, u32 data)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (!is_dmc_evt_ctl_reg(display, dmc_id, reg))
 		return false;
 
@@ -545,12 +596,12 @@ static bool disable_dmc_evt(struct intel_display *display,
 		return true;
 
 	/* also disable the flip queue event on the main DMC on TGL */
-	if (IS_TIGERLAKE(i915) &&
+	if (display->platform.tigerlake &&
 	    REG_FIELD_GET(DMC_EVT_CTL_EVENT_ID_MASK, data) == DMC_EVT_CTL_EVENT_ID_CLK_MSEC)
 		return true;
 
 	/* also disable the HRR event on the main DMC on TGL/ADLS */
-	if ((IS_TIGERLAKE(i915) || IS_ALDERLAKE_S(i915)) &&
+	if ((display->platform.tigerlake || display->platform.alderlake_s) &&
 	    REG_FIELD_GET(DMC_EVT_CTL_EVENT_ID_MASK, data) == DMC_EVT_CTL_EVENT_ID_VBLANK_A)
 		return true;
 
@@ -582,7 +633,6 @@ static u32 dmc_mmiodata(struct intel_display *display,
  */
 void intel_dmc_load_program(struct intel_display *display)
 {
-	struct drm_i915_private *i915 __maybe_unused = to_i915(display->drm);
 	struct i915_power_domains *power_domains = &display->power.domains;
 	struct intel_dmc *dmc = display_to_dmc(display);
 	enum intel_dmc_id dmc_id;
@@ -595,7 +645,7 @@ void intel_dmc_load_program(struct intel_display *display)
 
 	disable_all_event_handlers(display);
 
-	assert_rpm_wakelock_held(&i915->runtime_pm);
+	assert_display_rpm_held(display);
 
 	preempt_disable();
 
@@ -1006,9 +1056,7 @@ static void intel_dmc_runtime_pm_put(struct intel_display *display)
 
 static const char *dmc_fallback_path(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	if (IS_ALDERLAKE_P(i915))
+	if (display->platform.alderlake_p)
 		return ADLP_DMC_FALLBACK_PATH;
 
 	return NULL;
@@ -1232,18 +1280,56 @@ void intel_dmc_snapshot_print(const struct intel_dmc_snapshot *snapshot, struct
 			   DMC_VERSION_MINOR(snapshot->version));
 }
 
+void intel_dmc_update_dc6_allowed_count(struct intel_display *display,
+					bool start_tracking)
+{
+	struct intel_dmc *dmc = display_to_dmc(display);
+	u32 dc5_cur_count;
+
+	if (DISPLAY_VER(dmc->display) < 14)
+		return;
+
+	dc5_cur_count = intel_de_read(dmc->display, DG1_DMC_DEBUG_DC5_COUNT);
+
+	if (!start_tracking)
+		dmc->dc6_allowed.count += dc5_cur_count - dmc->dc6_allowed.dc5_start;
+
+	dmc->dc6_allowed.dc5_start = dc5_cur_count;
+}
+
+static bool intel_dmc_get_dc6_allowed_count(struct intel_display *display, u32 *count)
+{
+	struct i915_power_domains *power_domains = &display->power.domains;
+	struct intel_dmc *dmc = display_to_dmc(display);
+	bool dc6_enabled;
+
+	if (DISPLAY_VER(display) < 14)
+		return false;
+
+	mutex_lock(&power_domains->lock);
+	dc6_enabled = intel_de_read(display, DC_STATE_EN) &
+		      DC_STATE_EN_UPTO_DC6;
+	if (dc6_enabled)
+		intel_dmc_update_dc6_allowed_count(display, false);
+
+	*count = dmc->dc6_allowed.count;
+	mutex_unlock(&power_domains->lock);
+
+	return true;
+}
+
 static int intel_dmc_debugfs_status_show(struct seq_file *m, void *unused)
 {
 	struct intel_display *display = m->private;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_dmc *dmc = display_to_dmc(display);
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	i915_reg_t dc5_reg, dc6_reg = INVALID_MMIO_REG;
+	u32 dc6_allowed_count;
 
 	if (!HAS_DMC(display))
 		return -ENODEV;
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
 	seq_printf(m, "DMC initialized: %s\n", str_yes_no(dmc));
 	seq_printf(m, "fw loaded: %s\n",
@@ -1254,7 +1340,7 @@ static int intel_dmc_debugfs_status_show(struct seq_file *m, void *unused)
 	seq_printf(m, "Pipe A fw loaded: %s\n",
 		   str_yes_no(has_dmc_id_fw(display, DMC_FW_PIPEA)));
 	seq_printf(m, "Pipe B fw needed: %s\n",
-		   str_yes_no(IS_ALDERLAKE_P(i915) ||
+		   str_yes_no(display->platform.alderlake_p ||
 			      DISPLAY_VER(display) >= 14));
 	seq_printf(m, "Pipe B fw loaded: %s\n",
 		   str_yes_no(has_dmc_id_fw(display, DMC_FW_PIPEB)));
@@ -1268,7 +1354,7 @@ static int intel_dmc_debugfs_status_show(struct seq_file *m, void *unused)
 	if (DISPLAY_VER(display) >= 12) {
 		i915_reg_t dc3co_reg;
 
-		if (IS_DGFX(i915) || DISPLAY_VER(display) >= 14) {
+		if (display->platform.dgfx || DISPLAY_VER(display) >= 14) {
 			dc3co_reg = DG1_DMC_DEBUG3;
 			dc5_reg = DG1_DMC_DEBUG_DC5_COUNT;
 		} else {
@@ -1280,14 +1366,18 @@ static int intel_dmc_debugfs_status_show(struct seq_file *m, void *unused)
 		seq_printf(m, "DC3CO count: %d\n",
 			   intel_de_read(display, dc3co_reg));
 	} else {
-		dc5_reg = IS_BROXTON(i915) ? BXT_DMC_DC3_DC5_COUNT :
+		dc5_reg = display->platform.broxton ? BXT_DMC_DC3_DC5_COUNT :
 			SKL_DMC_DC3_DC5_COUNT;
-		if (!IS_GEMINILAKE(i915) && !IS_BROXTON(i915))
+		if (!display->platform.geminilake && !display->platform.broxton)
 			dc6_reg = SKL_DMC_DC5_DC6_COUNT;
 	}
 
 	seq_printf(m, "DC3 -> DC5 count: %d\n", intel_de_read(display, dc5_reg));
-	if (i915_mmio_reg_valid(dc6_reg))
+
+	if (intel_dmc_get_dc6_allowed_count(display, &dc6_allowed_count))
+		seq_printf(m, "DC5 -> DC6 allowed count: %d\n",
+			   dc6_allowed_count);
+	else if (i915_mmio_reg_valid(dc6_reg))
 		seq_printf(m, "DC5 -> DC6 count: %d\n",
 			   intel_de_read(display, dc6_reg));
 
@@ -1299,7 +1389,7 @@ out:
 		   intel_de_read(display, DMC_SSP_BASE));
 	seq_printf(m, "htp: 0x%08x\n", intel_de_read(display, DMC_HTP_SKL));
 
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dmc.h b/drivers/gpu/drm/i915/display/intel_dmc.h
index 44cecef98e73..bd1c459b0075 100644
--- a/drivers/gpu/drm/i915/display/intel_dmc.h
+++ b/drivers/gpu/drm/i915/display/intel_dmc.h
@@ -18,6 +18,10 @@ void intel_dmc_load_program(struct intel_display *display);
 void intel_dmc_disable_program(struct intel_display *display);
 void intel_dmc_enable_pipe(struct intel_display *display, enum pipe pipe);
 void intel_dmc_disable_pipe(struct intel_display *display, enum pipe pipe);
+void intel_dmc_block_pkgc(struct intel_display *display, enum pipe pipe,
+			  bool block);
+void intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(struct intel_display *display,
+							    enum pipe pipe, bool enable);
 void intel_dmc_fini(struct intel_display *display);
 void intel_dmc_suspend(struct intel_display *display);
 void intel_dmc_resume(struct intel_display *display);
@@ -26,6 +30,7 @@ void intel_dmc_debugfs_register(struct intel_display *display);
 
 struct intel_dmc_snapshot *intel_dmc_snapshot_capture(struct intel_display *display);
 void intel_dmc_snapshot_print(const struct intel_dmc_snapshot *snapshot, struct drm_printer *p);
+void intel_dmc_update_dc6_allowed_count(struct intel_display *display, bool start_tracking);
 
 void assert_dmc_loaded(struct intel_display *display);
 
diff --git a/drivers/gpu/drm/i915/display/intel_dmc_regs.h b/drivers/gpu/drm/i915/display/intel_dmc_regs.h
index 1bf446f96a10..e16ea3f16ed8 100644
--- a/drivers/gpu/drm/i915/display/intel_dmc_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_dmc_regs.h
@@ -21,6 +21,20 @@
 #define MTL_PIPEDMC_CONTROL		_MMIO(0x45250)
 #define  PIPEDMC_ENABLE_MTL(pipe)	REG_BIT(((pipe) - PIPE_A) * 4)
 
+#define _MTL_PIPEDMC_EVT_CTL_4_A	0x5f044
+#define _MTL_PIPEDMC_EVT_CTL_4_B	0x5f444
+#define MTL_PIPEDMC_EVT_CTL_4(pipe)	_MMIO_PIPE(pipe,		\
+						   _MTL_PIPEDMC_EVT_CTL_4_A, \
+						   _MTL_PIPEDMC_EVT_CTL_4_B)
+
+#define PIPEDMC_BLOCK_PKGC_SW_A	0x5f1d0
+#define PIPEDMC_BLOCK_PKGC_SW_B	0x5F5d0
+#define PIPEDMC_BLOCK_PKGC_SW(pipe)				_MMIO_PIPE(pipe, \
+									   PIPEDMC_BLOCK_PKGC_SW_A, \
+									   PIPEDMC_BLOCK_PKGC_SW_B)
+#define PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS			BIT(31)
+#define PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_UNTIL_NEXT_FRAMESTART	BIT(15)
+
 #define _ADLP_PIPEDMC_REG_MMIO_BASE_A	0x5f000
 #define _TGL_PIPEDMC_REG_MMIO_BASE_A	0x92000
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index a236b5fc7a3d..ad1e4fc9c7fe 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -45,12 +45,13 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fixed.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
 #include "g4x_dp.h"
-#include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_alpm.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
@@ -58,10 +59,12 @@
 #include "intel_combo_phy_regs.h"
 #include "intel_connector.h"
 #include "intel_crtc.h"
+#include "intel_crtc_state_dump.h"
 #include "intel_cx0_phy.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
 #include "intel_display_driver.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
 #include "intel_dp_aux.h"
@@ -87,12 +90,10 @@
 #include "intel_pfit.h"
 #include "intel_pps.h"
 #include "intel_psr.h"
-#include "intel_runtime_pm.h"
 #include "intel_quirks.h"
 #include "intel_tc.h"
 #include "intel_vdsc.h"
 #include "intel_vrr.h"
-#include "intel_crtc_state_dump.h"
 
 /* DP DSC throughput values used for slice count calculations KPixels/s */
 #define DP_DSC_PEAK_PIXEL_RATE			2720000
@@ -172,10 +173,28 @@ int intel_dp_link_symbol_clock(int rate)
 
 static int max_dprx_rate(struct intel_dp *intel_dp)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
+	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
+	int max_rate;
+
 	if (intel_dp_tunnel_bw_alloc_is_enabled(intel_dp))
-		return drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel);
+		max_rate = drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel);
+	else
+		max_rate = drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]);
 
-	return drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]);
+	/*
+	 * Some broken eDP sinks illegally declare support for
+	 * HBR3 without TPS4, and are unable to produce a stable
+	 * output. Reject HBR3 when TPS4 is not available.
+	 */
+	if (max_rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) {
+		drm_dbg_kms(display->drm,
+			    "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n",
+			    encoder->base.base.id, encoder->base.name);
+		max_rate = 540000;
+	}
+
+	return max_rate;
 }
 
 static int max_dprx_lane_count(struct intel_dp *intel_dp)
@@ -1032,10 +1051,11 @@ u8 intel_dp_dsc_get_slice_count(const struct intel_connector *connector,
 		u8 test_slice_count = valid_dsc_slicecount[i] * num_joined_pipes;
 
 		/*
-		 * 3 DSC Slices per pipe need 3 DSC engines,
-		 * which is supported only with Ultrajoiner.
+		 * 3 DSC Slices per pipe need 3 DSC engines, which is supported only
+		 * with Ultrajoiner only for some platforms.
 		 */
-		if (valid_dsc_slicecount[i] == 3 && num_joined_pipes != 4)
+		if (valid_dsc_slicecount[i] == 3 &&
+		    (!HAS_DSC_3ENGINES(display) || num_joined_pipes != 4))
 			continue;
 
 		if (test_slice_count >
@@ -2504,6 +2524,7 @@ intel_dp_dsc_compute_pipe_bpp_limits(struct intel_dp *intel_dp,
 
 bool
 intel_dp_compute_config_limits(struct intel_dp *intel_dp,
+			       struct intel_connector *connector,
 			       struct intel_crtc_state *crtc_state,
 			       bool respect_downstream_limits,
 			       bool dsc,
@@ -2557,7 +2578,7 @@ intel_dp_compute_config_limits(struct intel_dp *intel_dp,
 	intel_dp_test_compute_config(intel_dp, crtc_state, limits);
 
 	return intel_dp_compute_config_link_bpp_limits(intel_dp,
-						       intel_dp->attached_connector,
+						       connector,
 						       crtc_state,
 						       dsc,
 						       limits);
@@ -2618,7 +2639,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 	joiner_needs_dsc = intel_dp_joiner_needs_dsc(display, num_joined_pipes);
 
 	dsc_needed = joiner_needs_dsc || intel_dp->force_dsc_en ||
-		     !intel_dp_compute_config_limits(intel_dp, pipe_config,
+		     !intel_dp_compute_config_limits(intel_dp, connector, pipe_config,
 						     respect_downstream_limits,
 						     false,
 						     &limits);
@@ -2652,7 +2673,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 			    str_yes_no(ret), str_yes_no(joiner_needs_dsc),
 			    str_yes_no(intel_dp->force_dsc_en));
 
-		if (!intel_dp_compute_config_limits(intel_dp, pipe_config,
+		if (!intel_dp_compute_config_limits(intel_dp, connector, pipe_config,
 						    respect_downstream_limits,
 						    true,
 						    &limits))
@@ -3085,6 +3106,76 @@ intel_dp_queue_modeset_retry_for_link(struct intel_atomic_state *state,
 	}
 }
 
+int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state,
+				const struct drm_connector_state *conn_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	const struct drm_display_mode *adjusted_mode =
+					&crtc_state->hw.adjusted_mode;
+	struct intel_connector *connector = to_intel_connector(conn_state->connector);
+	int symbol_size = intel_dp_is_uhbr(crtc_state) ? 32 : 8;
+	/*
+	 * min symbol cycles is 3(BS,VBID, BE) for 128b/132b and
+	 * 5(BS, VBID, MVID, MAUD, BE) for 8b/10b
+	 */
+	int min_sym_cycles = intel_dp_is_uhbr(crtc_state) ? 3 : 5;
+	bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST);
+	int num_joined_pipes = intel_crtc_num_joined_pipes(crtc_state);
+	int min_hblank;
+	int max_lane_count = 4;
+	int hactive_sym_cycles, htotal_sym_cycles;
+	int dsc_slices = 0;
+	int link_bpp_x16;
+
+	if (DISPLAY_VER(display) < 30)
+		return 0;
+
+	/* MIN_HBLANK should be set only for 8b/10b MST or for 128b/132b SST/MST */
+	if (!is_mst && !intel_dp_is_uhbr(crtc_state))
+		return 0;
+
+	if (crtc_state->dsc.compression_enable) {
+		dsc_slices = intel_dp_dsc_get_slice_count(connector,
+							  adjusted_mode->crtc_clock,
+							  adjusted_mode->crtc_hdisplay,
+							  num_joined_pipes);
+		if (!dsc_slices) {
+			drm_dbg(display->drm, "failed to calculate dsc slice count\n");
+			return -EINVAL;
+		}
+	}
+
+	if (crtc_state->dsc.compression_enable)
+		link_bpp_x16 = crtc_state->dsc.compressed_bpp_x16;
+	else
+		link_bpp_x16 = fxp_q4_from_int(intel_dp_output_bpp(crtc_state->output_format,
+								   crtc_state->pipe_bpp));
+
+	/* Calculate min Hblank Link Layer Symbol Cycle Count for 8b/10b MST & 128b/132b */
+	hactive_sym_cycles = drm_dp_link_symbol_cycles(max_lane_count,
+						       adjusted_mode->hdisplay,
+						       dsc_slices,
+						       link_bpp_x16,
+						       symbol_size, is_mst);
+	htotal_sym_cycles = adjusted_mode->htotal * hactive_sym_cycles /
+			     adjusted_mode->hdisplay;
+
+	min_hblank = htotal_sym_cycles - hactive_sym_cycles;
+	/* minimum Hblank calculation: https://groups.vesa.org/wg/DP/document/20494 */
+	min_hblank = max(min_hblank, min_sym_cycles);
+
+	/*
+	 * adjust the BlankingStart/BlankingEnd framing control from
+	 * the calculated value
+	 */
+	min_hblank = min_hblank - 2;
+
+	min_hblank = min(10, min_hblank);
+	crtc_state->min_hblank = min_hblank;
+
+	return 0;
+}
+
 int
 intel_dp_compute_config(struct intel_encoder *encoder,
 			struct intel_crtc_state *pipe_config,
@@ -3184,6 +3275,10 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 				       &pipe_config->dp_m_n);
 	}
 
+	ret = intel_dp_compute_min_hblank(pipe_config, conn_state);
+	if (ret)
+		return ret;
+
 	/* FIXME: abstract this better */
 	if (pipe_config->splitter.enable)
 		pipe_config->dp_m_n.data_m *= pipe_config->splitter.link_count;
@@ -3204,7 +3299,7 @@ void intel_dp_set_link_params(struct intel_dp *intel_dp,
 			      int link_rate, int lane_count)
 {
 	memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set));
-	intel_dp->link_trained = false;
+	intel_dp->link.active = false;
 	intel_dp->needs_modeset_retry = false;
 	intel_dp->link_rate = link_rate;
 	intel_dp->lane_count = lane_count;
@@ -3568,7 +3663,7 @@ void intel_dp_sync_state(struct intel_encoder *encoder,
 	if (crtc_state) {
 		intel_dp_reset_link_params(intel_dp);
 		intel_dp_set_link_params(intel_dp, crtc_state->port_clock, crtc_state->lane_count);
-		intel_dp->link_trained = true;
+		intel_dp->link.active = true;
 	}
 }
 
@@ -4170,6 +4265,9 @@ static void intel_edp_mso_init(struct intel_dp *intel_dp)
 static void
 intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
+	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
+
 	intel_dp->num_sink_rates = 0;
 
 	if (intel_dp->edp_dpcd[0] >= DP_EDP_14) {
@@ -4180,10 +4278,7 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 				 sink_rates, sizeof(sink_rates));
 
 		for (i = 0; i < ARRAY_SIZE(sink_rates); i++) {
-			int val = le16_to_cpu(sink_rates[i]);
-
-			if (val == 0)
-				break;
+			int rate;
 
 			/* Value read multiplied by 200kHz gives the per-lane
 			 * link rate in kHz. The source rates are, however,
@@ -4191,7 +4286,24 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 			 * back to symbols is
 			 * (val * 200kHz)*(8/10 ch. encoding)*(1/8 bit to Byte)
 			 */
-			intel_dp->sink_rates[i] = (val * 200) / 10;
+			rate = le16_to_cpu(sink_rates[i]) * 200 / 10;
+
+			if (rate == 0)
+				break;
+
+			/*
+			 * Some broken eDP sinks illegally declare support for
+			 * HBR3 without TPS4, and are unable to produce a stable
+			 * output. Reject HBR3 when TPS4 is not available.
+			 */
+			if (rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) {
+				drm_dbg_kms(display->drm,
+					    "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n",
+					    encoder->base.base.id, encoder->base.name);
+				break;
+			}
+
+			intel_dp->sink_rates[i] = rate;
 		}
 		intel_dp->num_sink_rates = i;
 	}
@@ -4969,8 +5081,6 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp)
 	bool link_ok = true;
 	bool reprobe_needed = false;
 
-	drm_WARN_ON_ONCE(display->drm, intel_dp->mst.active_links < 0);
-
 	for (;;) {
 		u8 esi[4] = {};
 		u8 ack[4] = {};
@@ -4985,7 +5095,7 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp)
 
 		drm_dbg_kms(display->drm, "DPRX ESI: %4ph\n", esi);
 
-		if (intel_dp->mst.active_links > 0 && link_ok &&
+		if (intel_dp_mst_active_streams(intel_dp) > 0 && link_ok &&
 		    esi[3] & LINK_STATUS_CHANGED) {
 			if (!intel_dp_mst_link_status(intel_dp))
 				link_ok = false;
@@ -5046,7 +5156,7 @@ intel_dp_needs_link_retrain(struct intel_dp *intel_dp)
 {
 	u8 link_status[DP_LINK_STATUS_SIZE];
 
-	if (!intel_dp->link_trained)
+	if (!intel_dp->link.active)
 		return false;
 
 	/*
@@ -5358,6 +5468,11 @@ intel_dp_short_pulse(struct intel_dp *intel_dp)
 
 	intel_psr_short_pulse(intel_dp);
 
+	if (intel_alpm_get_error(intel_dp)) {
+		intel_alpm_disable(intel_dp);
+		intel_dp->alpm_parameters.sink_alpm_error = true;
+	}
+
 	if (intel_dp_test_short_pulse(intel_dp))
 		reprobe_needed = true;
 
@@ -5793,20 +5908,21 @@ out_vdd_off:
 }
 
 static void
-intel_dp_force(struct drm_connector *connector)
+intel_dp_force(struct drm_connector *_connector)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct intel_dp *intel_dp = intel_attached_dp(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_dp *intel_dp = intel_attached_dp(connector);
 
 	drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s]\n",
-		    connector->base.id, connector->name);
+		    connector->base.base.id, connector->base.name);
 
 	if (!intel_display_driver_check_access(display))
 		return;
 
 	intel_dp_unset_edid(intel_dp);
 
-	if (connector->status != connector_status_connected)
+	if (connector->base.status != connector_status_connected)
 		return;
 
 	intel_dp_set_edid(intel_dp);
@@ -5845,24 +5961,25 @@ static int intel_dp_get_modes(struct drm_connector *_connector)
 }
 
 static int
-intel_dp_connector_register(struct drm_connector *connector)
+intel_dp_connector_register(struct drm_connector *_connector)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct intel_dp *intel_dp = intel_attached_dp(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	int ret;
 
-	ret = intel_connector_register(connector);
+	ret = intel_connector_register(&connector->base);
 	if (ret)
 		return ret;
 
 	drm_dbg_kms(display->drm, "registering %s bus for %s\n",
-		    intel_dp->aux.name, connector->kdev->kobj.name);
+		    intel_dp->aux.name, connector->base.kdev->kobj.name);
 
-	intel_dp->aux.dev = connector->kdev;
+	intel_dp->aux.dev = connector->base.kdev;
 	ret = drm_dp_aux_register(&intel_dp->aux);
 	if (!ret)
-		drm_dp_cec_register_connector(&intel_dp->aux, connector);
+		drm_dp_cec_register_connector(&intel_dp->aux, &connector->base);
 
 	if (!intel_bios_encoder_is_lspcon(dig_port->base.devdata))
 		return ret;
@@ -5873,20 +5990,21 @@ intel_dp_connector_register(struct drm_connector *connector)
 	 */
 	if (intel_lspcon_init(dig_port)) {
 		if (intel_lspcon_detect_hdr_capability(dig_port))
-			drm_connector_attach_hdr_output_metadata_property(connector);
+			drm_connector_attach_hdr_output_metadata_property(&connector->base);
 	}
 
 	return ret;
 }
 
 static void
-intel_dp_connector_unregister(struct drm_connector *connector)
+intel_dp_connector_unregister(struct drm_connector *_connector)
 {
-	struct intel_dp *intel_dp = intel_attached_dp(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_dp *intel_dp = intel_attached_dp(connector);
 
 	drm_dp_cec_unregister_connector(&intel_dp->aux);
 	drm_dp_aux_unregister(&intel_dp->aux);
-	intel_connector_unregister(connector);
+	intel_connector_unregister(&connector->base);
 }
 
 void intel_dp_connector_sync_state(struct intel_connector *connector,
@@ -5947,21 +6065,21 @@ static int intel_modeset_tile_group(struct intel_atomic_state *state,
 {
 	struct intel_display *display = to_intel_display(state);
 	struct drm_connector_list_iter conn_iter;
-	struct drm_connector *connector;
+	struct intel_connector *connector;
 	int ret = 0;
 
 	drm_connector_list_iter_begin(display->drm, &conn_iter);
-	drm_for_each_connector_iter(connector, &conn_iter) {
+	for_each_intel_connector_iter(connector, &conn_iter) {
 		struct drm_connector_state *conn_state;
 		struct intel_crtc_state *crtc_state;
 		struct intel_crtc *crtc;
 
-		if (!connector->has_tile ||
-		    connector->tile_group->id != tile_group_id)
+		if (!connector->base.has_tile ||
+		    connector->base.tile_group->id != tile_group_id)
 			continue;
 
 		conn_state = drm_atomic_get_connector_state(&state->base,
-							    connector);
+							    &connector->base);
 		if (IS_ERR(conn_state)) {
 			ret = PTR_ERR(conn_state);
 			break;
@@ -6025,10 +6143,11 @@ static int intel_modeset_affected_transcoders(struct intel_atomic_state *state,
 }
 
 static int intel_modeset_synced_crtcs(struct intel_atomic_state *state,
-				      struct drm_connector *connector)
+				      struct drm_connector *_connector)
 {
+	struct intel_connector *connector = to_intel_connector(_connector);
 	const struct drm_connector_state *old_conn_state =
-		drm_atomic_get_old_connector_state(&state->base, connector);
+		drm_atomic_get_old_connector_state(&state->base, &connector->base);
 	const struct intel_crtc_state *old_crtc_state;
 	struct intel_crtc *crtc;
 	u8 transcoders;
@@ -6050,17 +6169,18 @@ static int intel_modeset_synced_crtcs(struct intel_atomic_state *state,
 						  transcoders);
 }
 
-static int intel_dp_connector_atomic_check(struct drm_connector *conn,
+static int intel_dp_connector_atomic_check(struct drm_connector *_connector,
 					   struct drm_atomic_state *_state)
 {
-	struct intel_display *display = to_intel_display(conn->dev);
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
 	struct intel_atomic_state *state = to_intel_atomic_state(_state);
-	struct drm_connector_state *conn_state = drm_atomic_get_new_connector_state(_state, conn);
-	struct intel_connector *intel_conn = to_intel_connector(conn);
-	struct intel_dp *intel_dp = enc_to_intel_dp(intel_conn->encoder);
+	struct drm_connector_state *conn_state =
+		drm_atomic_get_new_connector_state(_state, &connector->base);
+	struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder);
 	int ret;
 
-	ret = intel_digital_connector_atomic_check(conn, &state->base);
+	ret = intel_digital_connector_atomic_check(&connector->base, &state->base);
 	if (ret)
 		return ret;
 
@@ -6070,12 +6190,12 @@ static int intel_dp_connector_atomic_check(struct drm_connector *conn,
 			return ret;
 	}
 
-	if (!intel_connector_needs_modeset(state, conn))
+	if (!intel_connector_needs_modeset(state, &connector->base))
 		return 0;
 
 	ret = intel_dp_tunnel_atomic_check_state(state,
 						 intel_dp,
-						 intel_conn);
+						 connector);
 	if (ret)
 		return ret;
 
@@ -6086,26 +6206,26 @@ static int intel_dp_connector_atomic_check(struct drm_connector *conn,
 	if (DISPLAY_VER(display) < 9)
 		return 0;
 
-	if (conn->has_tile) {
-		ret = intel_modeset_tile_group(state, conn->tile_group->id);
+	if (connector->base.has_tile) {
+		ret = intel_modeset_tile_group(state, connector->base.tile_group->id);
 		if (ret)
 			return ret;
 	}
 
-	return intel_modeset_synced_crtcs(state, conn);
+	return intel_modeset_synced_crtcs(state, &connector->base);
 }
 
-static void intel_dp_oob_hotplug_event(struct drm_connector *connector,
+static void intel_dp_oob_hotplug_event(struct drm_connector *_connector,
 				       enum drm_connector_status hpd_state)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct intel_encoder *encoder = intel_attached_encoder(to_intel_connector(connector));
-	struct drm_i915_private *i915 = to_i915(connector->dev);
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_encoder *encoder = intel_attached_encoder(connector);
 	bool hpd_high = hpd_state == connector_status_connected;
 	unsigned int hpd_pin = encoder->hpd_pin;
 	bool need_work = false;
 
-	spin_lock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 	if (hpd_high != test_bit(hpd_pin, &display->hotplug.oob_hotplug_last_state)) {
 		display->hotplug.event_bits |= BIT(hpd_pin);
 
@@ -6114,10 +6234,10 @@ static void intel_dp_oob_hotplug_event(struct drm_connector *connector,
 			     hpd_high);
 		need_work = true;
 	}
-	spin_unlock_irq(&i915->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
 	if (need_work)
-		intel_hpd_schedule_detection(i915);
+		intel_hpd_schedule_detection(display);
 }
 
 static const struct drm_connector_funcs intel_dp_connector_funcs = {
@@ -6144,13 +6264,12 @@ enum irqreturn
 intel_dp_hpd_pulse(struct intel_digital_port *dig_port, bool long_hpd)
 {
 	struct intel_display *display = to_intel_display(dig_port);
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	struct intel_dp *intel_dp = &dig_port->dp;
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
 
 	if (dig_port->base.type == INTEL_OUTPUT_EDP &&
 	    (long_hpd ||
-	     intel_runtime_pm_suspended(&i915->runtime_pm) ||
+	     intel_display_rpm_suspended(display) ||
 	     !intel_pps_have_panel_power_or_vdd(intel_dp))) {
 		/*
 		 * vdd off can generate a long/short pulse on eDP which
@@ -6247,36 +6366,37 @@ intel_dp_has_gamut_metadata_dip(struct intel_encoder *encoder)
 }
 
 static void
-intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connector)
+intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *_connector)
 {
+	struct intel_connector *connector = to_intel_connector(_connector);
 	struct intel_display *display = to_intel_display(intel_dp);
 	enum port port = dp_to_dig_port(intel_dp)->base.port;
 
 	if (!intel_dp_is_edp(intel_dp))
-		drm_connector_attach_dp_subconnector_property(connector);
+		drm_connector_attach_dp_subconnector_property(&connector->base);
 
 	if (!display->platform.g4x && port != PORT_A)
-		intel_attach_force_audio_property(connector);
+		intel_attach_force_audio_property(&connector->base);
 
-	intel_attach_broadcast_rgb_property(connector);
+	intel_attach_broadcast_rgb_property(&connector->base);
 	if (HAS_GMCH(display))
-		drm_connector_attach_max_bpc_property(connector, 6, 10);
+		drm_connector_attach_max_bpc_property(&connector->base, 6, 10);
 	else if (DISPLAY_VER(display) >= 5)
-		drm_connector_attach_max_bpc_property(connector, 6, 12);
+		drm_connector_attach_max_bpc_property(&connector->base, 6, 12);
 
 	/* Register HDMI colorspace for case of lspcon */
 	if (intel_bios_encoder_is_lspcon(dp_to_dig_port(intel_dp)->base.devdata)) {
-		drm_connector_attach_content_type_property(connector);
-		intel_attach_hdmi_colorspace_property(connector);
+		drm_connector_attach_content_type_property(&connector->base);
+		intel_attach_hdmi_colorspace_property(&connector->base);
 	} else {
-		intel_attach_dp_colorspace_property(connector);
+		intel_attach_dp_colorspace_property(&connector->base);
 	}
 
 	if (intel_dp_has_gamut_metadata_dip(&dp_to_dig_port(intel_dp)->base))
-		drm_connector_attach_hdr_output_metadata_property(connector);
+		drm_connector_attach_hdr_output_metadata_property(&connector->base);
 
 	if (HAS_VRR(display))
-		drm_connector_attach_vrr_capable_property(connector);
+		drm_connector_attach_vrr_capable_property(&connector->base);
 }
 
 static void
@@ -6311,7 +6431,6 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 				     struct intel_connector *connector)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct drm_display_mode *fixed_mode;
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	bool has_dpcd;
@@ -6326,9 +6445,9 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	 * eDP and LVDS bail out early in this case to prevent interfering
 	 * with an already powered-on LVDS power sequencer.
 	 */
-	if (intel_get_lvds_encoder(dev_priv)) {
+	if (intel_get_lvds_encoder(display)) {
 		drm_WARN_ON(display->drm,
-			    !(HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv)));
+			    !(HAS_PCH_IBX(display) || HAS_PCH_CPT(display)));
 		drm_info(display->drm,
 			 "LVDS was detected, not registering eDP\n");
 
@@ -6359,7 +6478,7 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	 */
 	intel_hpd_enable_detection(encoder);
 
-	intel_alpm_init_dpcd(intel_dp);
+	intel_alpm_init(intel_dp);
 
 	/* Cache DPCD and EDID for edp. */
 	has_dpcd = intel_edp_init_dpcd(intel_dp, connector);
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index 9189db4c2594..742ae26ac4a9 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -194,6 +194,7 @@ void intel_dp_wait_source_oui(struct intel_dp *intel_dp);
 int intel_dp_output_bpp(enum intel_output_format output_format, int bpp);
 
 bool intel_dp_compute_config_limits(struct intel_dp *intel_dp,
+				    struct intel_connector *connector,
 				    struct intel_crtc_state *crtc_state,
 				    bool respect_downstream_limits,
 				    bool dsc,
@@ -208,5 +209,7 @@ bool intel_dp_has_connector(struct intel_dp *intel_dp,
 			    const struct drm_connector_state *conn_state);
 int intel_dp_dsc_max_src_input_bpc(struct intel_display *display);
 int intel_dp_dsc_min_src_input_bpc(void);
+int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state,
+				const struct drm_connector_state *conn_state);
 
 #endif /* __INTEL_DP_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c
index ec27bbd70bcf..bf8e8e0cc19c 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c
@@ -3,8 +3,10 @@
  * Copyright © 2020-2021 Intel Corporation
  */
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
@@ -111,10 +113,9 @@ static u32 ilk_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 static u32 hsw_get_aux_clock_divider(struct intel_dp *intel_dp, int index)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 
-	if (dig_port->aux_ch != AUX_CH_A && HAS_PCH_LPT_H(i915)) {
+	if (dig_port->aux_ch != AUX_CH_A && HAS_PCH_LPT_H(display)) {
 		/* Workaround for non-ULT HSW */
 		switch (index) {
 		case 0: return 63;
@@ -177,12 +178,11 @@ static u32 g4x_get_aux_send_ctl(struct intel_dp *intel_dp,
 				int send_bytes,
 				u32 aux_clock_divider)
 {
-	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(intel_dp);
 	u32 timeout;
 
 	/* Max timeout value on G4x-BDW: 1.6ms */
-	if (IS_BROADWELL(i915))
+	if (display->platform.broadwell)
 		timeout = DP_AUX_CH_CTL_TIME_OUT_600us;
 	else
 		timeout = DP_AUX_CH_CTL_TIME_OUT_400us;
@@ -247,7 +247,7 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp,
 	u32 aux_clock_divider;
 	enum intel_display_power_domain aux_domain;
 	intel_wakeref_t aux_wakeref;
-	intel_wakeref_t pps_wakeref;
+	intel_wakeref_t pps_wakeref = NULL;
 	int i, ret, recv_bytes;
 	int try, clock = 0;
 	u32 status;
@@ -272,7 +272,20 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp,
 	aux_domain = intel_aux_power_domain(dig_port);
 
 	aux_wakeref = intel_display_power_get(display, aux_domain);
-	pps_wakeref = intel_pps_lock(intel_dp);
+
+	/*
+	 * The PPS state needs to be locked for:
+	 * - eDP on all platforms, since AUX transfers on eDP need VDD power
+	 *   (either forced or via panel power) which depends on the PPS
+	 *   state.
+	 * - non-eDP on platforms where the PPS is a pipe instance (VLV/CHV),
+	 *   since changing the PPS state (via a parallel modeset for
+	 *   instance) may interfere with the AUX transfers on a non-eDP
+	 *   output as well.
+	 */
+	if (intel_dp_is_edp(intel_dp) ||
+	    display->platform.valleyview || display->platform.cherryview)
+		pps_wakeref = intel_pps_lock(intel_dp);
 
 	/*
 	 * We will be called with VDD already enabled for dpcd/edid/oui reads.
@@ -430,7 +443,9 @@ out:
 	if (vdd)
 		intel_pps_vdd_off_unlocked(intel_dp, false);
 
-	intel_pps_unlock(intel_dp, pps_wakeref);
+	if (pps_wakeref)
+		intel_pps_unlock(intel_dp, pps_wakeref);
+
 	intel_display_power_put_async(display, aux_domain, aux_wakeref);
 out_unlock:
 	intel_digital_port_unlock(encoder);
@@ -771,7 +786,6 @@ void intel_dp_aux_fini(struct intel_dp *intel_dp)
 void intel_dp_aux_init(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct intel_encoder *encoder = &dig_port->base;
 	enum aux_ch aux_ch = dig_port->aux_ch;
@@ -786,10 +800,10 @@ void intel_dp_aux_init(struct intel_dp *intel_dp)
 	} else if (DISPLAY_VER(display) >= 9) {
 		intel_dp->aux_ch_ctl_reg = skl_aux_ctl_reg;
 		intel_dp->aux_ch_data_reg = skl_aux_data_reg;
-	} else if (HAS_PCH_SPLIT(i915)) {
+	} else if (HAS_PCH_SPLIT(display)) {
 		intel_dp->aux_ch_ctl_reg = ilk_aux_ctl_reg;
 		intel_dp->aux_ch_data_reg = ilk_aux_data_reg;
-	} else if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) {
+	} else if (display->platform.valleyview || display->platform.cherryview) {
 		intel_dp->aux_ch_ctl_reg = vlv_aux_ctl_reg;
 		intel_dp->aux_ch_data_reg = vlv_aux_data_reg;
 	} else {
@@ -799,9 +813,9 @@ void intel_dp_aux_init(struct intel_dp *intel_dp)
 
 	if (DISPLAY_VER(display) >= 9)
 		intel_dp->get_aux_clock_divider = skl_get_aux_clock_divider;
-	else if (IS_BROADWELL(i915) || IS_HASWELL(i915))
+	else if (display->platform.broadwell || display->platform.haswell)
 		intel_dp->get_aux_clock_divider = hsw_get_aux_clock_divider;
-	else if (HAS_PCH_SPLIT(i915))
+	else if (HAS_PCH_SPLIT(display))
 		intel_dp->get_aux_clock_divider = ilk_get_aux_clock_divider;
 	else
 		intel_dp->get_aux_clock_divider = g4x_get_aux_clock_divider;
diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
index 8173de8aec63..271b27c9de51 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
@@ -36,7 +36,6 @@
 
 #include <drm/drm_print.h>
 
-#include "i915_utils.h"
 #include "intel_backlight.h"
 #include "intel_display_core.h"
 #include "intel_display_types.h"
@@ -149,7 +148,7 @@ intel_dp_aux_supports_hdr_backlight(struct intel_connector *connector)
 	    !(connector->base.hdr_sink_metadata.hdmi_type1.metadata_type &
 	      BIT(HDMI_STATIC_METADATA_TYPE1))) {
 		drm_info(display->drm,
-			 "[CONNECTOR:%d:%s] Panel is missing HDR static metadata. Possible support for Intel HDR backlight interface is not used. If your backlight controls don't work try booting with i915.enable_dpcd_backlight=%d. needs this, please file a _new_ bug report on drm/i915, see " FDO_BUG_URL " for details.\n",
+			 "[CONNECTOR:%d:%s] Panel is missing HDR static metadata. Possible support for Intel HDR backlight interface is not used. If your backlight controls don't work try booting with i915.enable_dpcd_backlight=%d.\n",
 			 connector->base.base.id, connector->base.name,
 			 INTEL_DP_AUX_BACKLIGHT_FORCE_INTEL);
 		return false;
@@ -663,7 +662,8 @@ intel_dp_aux_supports_vesa_backlight(struct intel_connector *connector)
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	struct intel_panel *panel = &connector->panel;
 
-	if ((intel_dp->edp_dpcd[3] & DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE)) {
+	if ((intel_dp->edp_dpcd[3] & DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE) &&
+	    (intel_dp->edp_dpcd[3] & DP_EDP_SMOOTH_BRIGHTNESS_CAPABLE)) {
 		drm_dbg_kms(display->drm,
 			    "[CONNECTOR:%d:%s] AUX Luminance Based Backlight Control Supported!\n",
 			    connector->base.base.id, connector->base.name);
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 2966f5b39392..a479b63112ea 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -56,6 +56,8 @@
 		lt_dbg(_intel_dp, _dp_phy, "Sink disconnected: " _format, ## __VA_ARGS__); \
 } while (0)
 
+#define MAX_SEQ_TRAIN_FAILURES 2
+
 static void intel_dp_reset_lttpr_common_caps(struct intel_dp *intel_dp)
 {
 	memset(intel_dp->lttpr_common_caps, 0, sizeof(intel_dp->lttpr_common_caps));
@@ -164,7 +166,7 @@ static int intel_dp_init_lttpr_phys(struct intel_dp *intel_dp, const u8 dpcd[DP_
 	 * resetting its internal state when the mode is changed from
 	 * non-transparent to transparent.
 	 */
-	if (intel_dp->link_trained) {
+	if (intel_dp->link.active) {
 		if (lttpr_count < 0 || intel_dp_lttpr_transparent_mode_enabled(intel_dp))
 			goto out_reset_lttpr_count;
 
@@ -711,8 +713,21 @@ void intel_dp_link_training_set_mode(struct intel_dp *intel_dp, int link_rate, b
 static void intel_dp_update_downspread_ctrl(struct intel_dp *intel_dp,
 					    const struct intel_crtc_state *crtc_state)
 {
+	 /*
+	  * Currently, we set the MSA ignore bit based on vrr.in_range.
+	  * We can't really read that out during driver load since we don't have
+	  * the connector information read in yet. So if we do end up doing a
+	  * modeset during initial_commit() we'll clear the MSA ignore bit.
+	  * GOP likely wouldn't have set this bit so after the initial commit,
+	  * if there are no modesets and we enable VRR mode seamlessly
+	  * (without a full modeset), the MSA ignore bit might never get set.
+	  *
+	  * #TODO: Implement readout of vrr.in_range.
+	  * We need fastset support for setting the MSA ignore bit in DPCD,
+	  * especially on the first real commit when clearing the inherited flag.
+	  */
 	intel_dp_link_training_set_mode(intel_dp,
-					crtc_state->port_clock, crtc_state->vrr.flipline);
+					crtc_state->port_clock, crtc_state->vrr.in_range);
 }
 
 void intel_dp_link_training_set_bw(struct intel_dp *intel_dp,
@@ -1110,7 +1125,10 @@ intel_dp_128b132b_intra_hop(struct intel_dp *intel_dp,
 void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 			      const struct intel_crtc_state *crtc_state)
 {
-	intel_dp->link_trained = true;
+	struct intel_display *display = to_intel_display(intel_dp);
+	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
+
+	intel_dp->link.active = true;
 
 	intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
 	intel_dp_program_link_training_pattern(intel_dp, crtc_state, DP_PHY_DPRX,
@@ -1120,6 +1138,15 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 	    wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) {
 		lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n");
 	}
+
+	intel_hpd_unblock(encoder);
+
+	if (!display->hotplug.ignore_long_hpd &&
+	    intel_dp->link.seq_train_failures < MAX_SEQ_TRAIN_FAILURES) {
+		int delay_ms = intel_dp->link.seq_train_failures ? 0 : 2000;
+
+		intel_encoder_link_check_queue_work(encoder, delay_ms);
+	}
 }
 
 static bool
@@ -1602,7 +1629,11 @@ void intel_dp_start_link_train(struct intel_atomic_state *state,
 	 * non-transparent mode. During an earlier LTTPR detection this
 	 * could've been prevented by an active link.
 	 */
-	int lttpr_count = intel_dp_init_lttpr_and_dprx_caps(intel_dp);
+	int lttpr_count;
+
+	intel_hpd_block(encoder);
+
+	lttpr_count = intel_dp_init_lttpr_and_dprx_caps(intel_dp);
 
 	if (lttpr_count < 0)
 		/* Still continue with enabling the port and link training. */
@@ -1620,7 +1651,6 @@ void intel_dp_start_link_train(struct intel_atomic_state *state,
 		lt_dbg(intel_dp, DP_PHY_DPRX, "Forcing link training failure\n");
 	} else if (passed) {
 		intel_dp->link.seq_train_failures = 0;
-		intel_encoder_link_check_queue_work(encoder, 2000);
 		return;
 	}
 
@@ -1643,10 +1673,8 @@ void intel_dp_start_link_train(struct intel_atomic_state *state,
 		return;
 	}
 
-	if (intel_dp->link.seq_train_failures < 2) {
-		intel_encoder_link_check_queue_work(encoder, 0);
+	if (intel_dp->link.seq_train_failures < MAX_SEQ_TRAIN_FAILURES)
 		return;
-	}
 
 	if (intel_dp_schedule_fallback_link_training(state, intel_dp, crtc_state))
 		return;
@@ -1693,7 +1721,7 @@ static int i915_dp_force_link_rate_show(struct seq_file *m, void *data)
 	if (err)
 		return err;
 
-	if (intel_dp->link_trained)
+	if (intel_dp->link.active)
 		current_rate = intel_dp->link_rate;
 	force_rate = intel_dp->link.force_rate;
 
@@ -1791,7 +1819,7 @@ static int i915_dp_force_lane_count_show(struct seq_file *m, void *data)
 	if (err)
 		return err;
 
-	if (intel_dp->link_trained)
+	if (intel_dp->link.active)
 		current_lane_count = intel_dp->lane_count;
 	force_lane_count = intel_dp->link.force_lane_count;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 02f95108c637..c1fd6aceec2c 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -27,10 +27,11 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fixed.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
 #include "intel_connector.h"
@@ -51,7 +52,9 @@
 #include "intel_link_bw.h"
 #include "intel_pfit.h"
 #include "intel_psr.h"
+#include "intel_step.h"
 #include "intel_vdsc.h"
+#include "intel_vrr.h"
 #include "skl_scaler.h"
 
 /*
@@ -104,6 +107,34 @@ static struct intel_dp *to_primary_dp(struct intel_encoder *encoder)
 	return &dig_port->dp;
 }
 
+int intel_dp_mst_active_streams(struct intel_dp *intel_dp)
+{
+	return intel_dp->mst.active_streams;
+}
+
+static bool intel_dp_mst_dec_active_streams(struct intel_dp *intel_dp)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	drm_dbg_kms(display->drm, "active MST streams %d -> %d\n",
+		    intel_dp->mst.active_streams, intel_dp->mst.active_streams - 1);
+
+	if (drm_WARN_ON(display->drm, intel_dp->mst.active_streams == 0))
+		return true;
+
+	return --intel_dp->mst.active_streams == 0;
+}
+
+static bool intel_dp_mst_inc_active_streams(struct intel_dp *intel_dp)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	drm_dbg_kms(display->drm, "active MST streams %d -> %d\n",
+		    intel_dp->mst.active_streams, intel_dp->mst.active_streams + 1);
+
+	return intel_dp->mst.active_streams++ == 0;
+}
+
 static int intel_dp_mst_max_dpt_bpp(const struct intel_crtc_state *crtc_state,
 				    bool dsc)
 {
@@ -210,26 +241,6 @@ static int intel_dp_mst_dsc_get_slice_count(const struct intel_connector *connec
 					    num_joined_pipes);
 }
 
-static void intel_dp_mst_compute_min_hblank(struct intel_crtc_state *crtc_state,
-					    int bpp_x16)
-{
-	struct intel_display *display = to_intel_display(crtc_state);
-	const struct drm_display_mode *adjusted_mode =
-					&crtc_state->hw.adjusted_mode;
-	int symbol_size = intel_dp_is_uhbr(crtc_state) ? 32 : 8;
-	int hblank;
-
-	if (DISPLAY_VER(display) < 20)
-		return;
-
-	/* Calculate min Hblank Link Layer Symbol Cycle Count for 8b/10b MST & 128b/132b */
-	hblank = DIV_ROUND_UP((DIV_ROUND_UP
-			       (adjusted_mode->htotal - adjusted_mode->hdisplay, 4) * bpp_x16),
-			      symbol_size);
-
-	crtc_state->min_hblank = hblank;
-}
-
 int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp,
 				   struct intel_crtc_state *crtc_state,
 				   struct drm_connector_state *conn_state,
@@ -300,8 +311,6 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp,
 		local_bw_overhead = intel_dp_mst_bw_overhead(crtc_state,
 							     false, dsc_slice_count, link_bpp_x16);
 
-		intel_dp_mst_compute_min_hblank(crtc_state, link_bpp_x16);
-
 		intel_dp_mst_compute_m_n(crtc_state,
 					 local_bw_overhead,
 					 link_bpp_x16,
@@ -590,12 +599,13 @@ adjust_limits_for_dsc_hblank_expansion_quirk(struct intel_dp *intel_dp,
 
 static bool
 mst_stream_compute_config_limits(struct intel_dp *intel_dp,
-				 const struct intel_connector *connector,
+				 struct intel_connector *connector,
 				 struct intel_crtc_state *crtc_state,
 				 bool dsc,
 				 struct link_config_limits *limits)
 {
-	if (!intel_dp_compute_config_limits(intel_dp, crtc_state, false, dsc,
+	if (!intel_dp_compute_config_limits(intel_dp, connector,
+					    crtc_state, false, dsc,
 					    limits))
 		return false;
 
@@ -710,6 +720,12 @@ static int mst_stream_compute_config(struct intel_encoder *encoder,
 		pipe_config->lane_lat_optim_mask =
 			bxt_dpio_phy_calc_lane_lat_optim_mask(pipe_config->lane_count);
 
+	ret = intel_dp_compute_min_hblank(pipe_config, conn_state);
+	if (ret)
+		return ret;
+
+	intel_vrr_compute_config(pipe_config, conn_state);
+
 	intel_dp_audio_compute_config(encoder, pipe_config, conn_state);
 
 	intel_ddi_compute_min_voltage_level(pipe_config);
@@ -990,25 +1006,17 @@ static void mst_stream_disable(struct intel_atomic_state *state,
 			       const struct intel_crtc_state *old_crtc_state,
 			       const struct drm_connector_state *old_conn_state)
 {
-	struct intel_display *display = to_intel_display(state);
 	struct intel_dp_mst_encoder *intel_mst = enc_to_mst(encoder);
 	struct intel_dp *intel_dp = to_primary_dp(encoder);
 	struct intel_connector *connector =
 		to_intel_connector(old_conn_state->connector);
-	enum transcoder trans = old_crtc_state->cpu_transcoder;
 
-	drm_dbg_kms(display->drm, "active links %d\n",
-		    intel_dp->mst.active_links);
-
-	if (intel_dp->mst.active_links == 1)
-		intel_dp->link_trained = false;
+	if (intel_dp_mst_active_streams(intel_dp) == 1)
+		intel_dp->link.active = false;
 
 	intel_hdcp_disable(intel_mst->connector);
 
 	intel_dp_sink_disable_decompression(state, connector, old_crtc_state);
-
-	if (DISPLAY_VER(display) >= 20)
-		intel_de_write(display, DP_MIN_HBLANK_CTL(trans), 0);
 }
 
 static void mst_stream_post_disable(struct intel_atomic_state *state,
@@ -1034,8 +1042,8 @@ static void mst_stream_post_disable(struct intel_atomic_state *state,
 	bool last_mst_stream;
 	int i;
 
-	intel_dp->mst.active_links--;
-	last_mst_stream = intel_dp->mst.active_links == 0;
+	last_mst_stream = intel_dp_mst_dec_active_streams(intel_dp);
+
 	drm_WARN_ON(display->drm, DISPLAY_VER(display) >= 12 && last_mst_stream &&
 		    !intel_dp_mst_is_master_trans(old_crtc_state));
 
@@ -1062,6 +1070,8 @@ static void mst_stream_post_disable(struct intel_atomic_state *state,
 	drm_dp_remove_payload_part2(&intel_dp->mst.mgr, new_mst_state,
 				    old_payload, new_payload);
 
+	intel_vrr_transcoder_disable(old_crtc_state);
+
 	intel_ddi_disable_transcoder_func(old_crtc_state);
 
 	for_each_pipe_crtc_modeset_disable(display, pipe_crtc, old_crtc_state, i) {
@@ -1104,8 +1114,6 @@ static void mst_stream_post_disable(struct intel_atomic_state *state,
 		primary_encoder->post_disable(state, primary_encoder,
 					      old_crtc_state, NULL);
 
-	drm_dbg_kms(display->drm, "active links %d\n",
-		    intel_dp->mst.active_links);
 }
 
 static void mst_stream_post_pll_disable(struct intel_atomic_state *state,
@@ -1116,7 +1124,7 @@ static void mst_stream_post_pll_disable(struct intel_atomic_state *state,
 	struct intel_encoder *primary_encoder = to_primary_encoder(encoder);
 	struct intel_dp *intel_dp = to_primary_dp(encoder);
 
-	if (intel_dp->mst.active_links == 0 &&
+	if (intel_dp_mst_active_streams(intel_dp) == 0 &&
 	    primary_encoder->post_pll_disable)
 		primary_encoder->post_pll_disable(state, primary_encoder, old_crtc_state, old_conn_state);
 }
@@ -1129,7 +1137,7 @@ static void mst_stream_pre_pll_enable(struct intel_atomic_state *state,
 	struct intel_encoder *primary_encoder = to_primary_encoder(encoder);
 	struct intel_dp *intel_dp = to_primary_dp(encoder);
 
-	if (intel_dp->mst.active_links == 0)
+	if (intel_dp_mst_active_streams(intel_dp) == 0)
 		primary_encoder->pre_pll_enable(state, primary_encoder,
 						pipe_config, NULL);
 	else
@@ -1189,13 +1197,11 @@ static void mst_stream_pre_enable(struct intel_atomic_state *state,
 	 */
 	connector->encoder = encoder;
 	intel_mst->connector = connector;
-	first_mst_stream = intel_dp->mst.active_links == 0;
+
+	first_mst_stream = intel_dp_mst_inc_active_streams(intel_dp);
 	drm_WARN_ON(display->drm, DISPLAY_VER(display) >= 12 && first_mst_stream &&
 		    !intel_dp_mst_is_master_trans(pipe_config));
 
-	drm_dbg_kms(display->drm, "active links %d\n",
-		    intel_dp->mst.active_links);
-
 	if (first_mst_stream)
 		intel_dp_set_power(intel_dp, DP_SET_POWER_D0);
 
@@ -1210,8 +1216,6 @@ static void mst_stream_pre_enable(struct intel_atomic_state *state,
 		intel_mst_reprobe_topology(intel_dp, pipe_config);
 	}
 
-	intel_dp->mst.active_links++;
-
 	ret = drm_dp_add_payload_part1(&intel_dp->mst.mgr, mst_state,
 				       drm_atomic_get_mst_payload_state(mst_state, connector->mst.port));
 	if (ret < 0)
@@ -1279,9 +1283,9 @@ static void mst_stream_enable(struct intel_atomic_state *state,
 	struct drm_dp_mst_topology_state *mst_state =
 		drm_atomic_get_new_mst_topology_state(&state->base, &intel_dp->mst.mgr);
 	enum transcoder trans = pipe_config->cpu_transcoder;
-	bool first_mst_stream = intel_dp->mst.active_links == 1;
+	bool first_mst_stream = intel_dp_mst_active_streams(intel_dp) == 1;
 	struct intel_crtc *pipe_crtc;
-	int ret, i, min_hblank;
+	int ret, i;
 
 	drm_WARN_ON(display->drm, pipe_config->has_pch_encoder);
 
@@ -1296,41 +1300,17 @@ static void mst_stream_enable(struct intel_atomic_state *state,
 			       TRANS_DP2_VFREQ_PIXEL_CLOCK(crtc_clock_hz & 0xffffff));
 	}
 
-	if (DISPLAY_VER(display) >= 20) {
-		/*
-		 * adjust the BlankingStart/BlankingEnd framing control from
-		 * the calculated value
-		 */
-		min_hblank = pipe_config->min_hblank - 2;
-
-		/* Maximum value to be programmed is limited to 0x10 */
-		min_hblank = min(0x10, min_hblank);
-
-		/*
-		 * Minimum hblank accepted for 128b/132b would be 5 and for
-		 * 8b/10b would be 3 symbol count
-		 */
-		if (intel_dp_is_uhbr(pipe_config))
-			min_hblank = max(min_hblank, 5);
-		else
-			min_hblank = max(min_hblank, 3);
-
-		intel_de_write(display, DP_MIN_HBLANK_CTL(trans),
-			       min_hblank);
-	}
-
 	enable_bs_jitter_was(pipe_config);
 
 	intel_ddi_enable_transcoder_func(encoder, pipe_config);
 
+	intel_vrr_transcoder_enable(pipe_config);
+
 	intel_ddi_clear_act_sent(encoder, pipe_config);
 
 	intel_de_rmw(display, TRANS_DDI_FUNC_CTL(display, trans), 0,
 		     TRANS_DDI_DP_VC_PAYLOAD_ALLOC);
 
-	drm_dbg_kms(display->drm, "active links %d\n",
-		    intel_dp->mst.active_links);
-
 	intel_ddi_wait_for_act_sent(encoder, pipe_config);
 	drm_dp_check_act_status(&intel_dp->mst.mgr);
 
@@ -1870,12 +1850,6 @@ mst_stream_encoders_create(struct intel_digital_port *dig_port)
 }
 
 int
-intel_dp_mst_encoder_active_links(struct intel_digital_port *dig_port)
-{
-	return dig_port->dp.mst.active_links;
-}
-
-int
 intel_dp_mst_encoder_init(struct intel_digital_port *dig_port, int conn_base_id)
 {
 	struct intel_display *display = to_intel_display(dig_port);
@@ -2101,7 +2075,7 @@ void intel_dp_mst_prepare_probe(struct intel_dp *intel_dp)
 	u8 rate_select;
 	u8 link_bw;
 
-	if (intel_dp->link_trained)
+	if (intel_dp->link.active)
 		return;
 
 	if (intel_mst_probed_link_params_valid(intel_dp, link_rate, lane_count))
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.h b/drivers/gpu/drm/i915/display/intel_dp_mst.h
index c1bbfeb02ca9..ab09b487c6bb 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.h
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.h
@@ -18,7 +18,7 @@ struct intel_link_bw_limits;
 
 int intel_dp_mst_encoder_init(struct intel_digital_port *dig_port, int conn_id);
 void intel_dp_mst_encoder_cleanup(struct intel_digital_port *dig_port);
-int intel_dp_mst_encoder_active_links(struct intel_digital_port *dig_port);
+int intel_dp_mst_active_streams(struct intel_dp *intel_dp);
 bool intel_dp_mst_is_master_trans(const struct intel_crtc_state *crtc_state);
 bool intel_dp_mst_is_slave_trans(const struct intel_crtc_state *crtc_state);
 bool intel_dp_mst_source_support(struct intel_dp *intel_dp);
diff --git a/drivers/gpu/drm/i915/display/intel_dpio_phy.c b/drivers/gpu/drm/i915/display/intel_dpio_phy.c
index 429f89543789..69f242139420 100644
--- a/drivers/gpu/drm/i915/display/intel_dpio_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_dpio_phy.c
@@ -222,9 +222,7 @@ static const struct bxt_dpio_phy_info glk_dpio_phy_info[] = {
 static const struct bxt_dpio_phy_info *
 bxt_get_phy_list(struct intel_display *display, int *count)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	if (IS_GEMINILAKE(dev_priv)) {
+	if (display->platform.geminilake) {
 		*count =  ARRAY_SIZE(glk_dpio_phy_info);
 		return glk_dpio_phy_info;
 	} else {
@@ -808,9 +806,9 @@ void chv_set_phy_signal_level(struct intel_encoder *encoder,
 	vlv_dpio_put(dev_priv);
 }
 
-void chv_data_lane_soft_reset(struct intel_encoder *encoder,
-			      const struct intel_crtc_state *crtc_state,
-			      bool reset)
+static void __chv_data_lane_soft_reset(struct intel_encoder *encoder,
+				       const struct intel_crtc_state *crtc_state,
+				       bool reset)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
@@ -853,6 +851,17 @@ void chv_data_lane_soft_reset(struct intel_encoder *encoder,
 	}
 }
 
+void chv_data_lane_soft_reset(struct intel_encoder *encoder,
+			      const struct intel_crtc_state *crtc_state,
+			      bool reset)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	vlv_dpio_get(i915);
+	__chv_data_lane_soft_reset(encoder, crtc_state, reset);
+	vlv_dpio_put(i915);
+}
+
 void chv_phy_pre_pll_enable(struct intel_encoder *encoder,
 			    const struct intel_crtc_state *crtc_state)
 {
@@ -880,7 +889,7 @@ void chv_phy_pre_pll_enable(struct intel_encoder *encoder,
 	vlv_dpio_get(dev_priv);
 
 	/* Assert data lane reset */
-	chv_data_lane_soft_reset(encoder, crtc_state, true);
+	__chv_data_lane_soft_reset(encoder, crtc_state, true);
 
 	/* program left/right clock distribution */
 	if (pipe != PIPE_B) {
@@ -1008,7 +1017,7 @@ void chv_phy_pre_encoder_enable(struct intel_encoder *encoder,
 	}
 
 	/* Deassert data lane reset */
-	chv_data_lane_soft_reset(encoder, crtc_state, false);
+	__chv_data_lane_soft_reset(encoder, crtc_state, false);
 
 	vlv_dpio_put(dev_priv);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c
index 08a30e5aafce..a9e9b98d0bf9 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll.c
@@ -373,14 +373,14 @@ int chv_calc_dpll_params(int refclk, struct dpll *clock)
 
 static int i9xx_pll_refclk(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 
 	if ((hw_state->dpll & PLL_REF_INPUT_MASK) == PLLB_REF_INPUT_SPREADSPECTRUMIN)
-		return i915->display.vbt.lvds_ssc_freq;
-	else if (HAS_PCH_SPLIT(i915))
+		return display->vbt.lvds_ssc_freq;
+	else if (HAS_PCH_SPLIT(display))
 		return 120000;
-	else if (DISPLAY_VER(i915) != 2)
+	else if (DISPLAY_VER(display) != 2)
 		return 96000;
 	else
 		return 48000;
@@ -389,27 +389,27 @@ static int i9xx_pll_refclk(const struct intel_crtc_state *crtc_state)
 void i9xx_dpll_get_hw_state(struct intel_crtc *crtc,
 			    struct intel_dpll_hw_state *dpll_hw_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct i9xx_dpll_hw_state *hw_state = &dpll_hw_state->i9xx;
 
-	if (DISPLAY_VER(dev_priv) >= 4) {
+	if (DISPLAY_VER(display) >= 4) {
 		u32 tmp;
 
 		/* No way to read it out on pipes B and C */
-		if (IS_CHERRYVIEW(dev_priv) && crtc->pipe != PIPE_A)
-			tmp = dev_priv->display.state.chv_dpll_md[crtc->pipe];
+		if (display->platform.cherryview && crtc->pipe != PIPE_A)
+			tmp = display->state.chv_dpll_md[crtc->pipe];
 		else
-			tmp = intel_de_read(dev_priv,
-					    DPLL_MD(dev_priv, crtc->pipe));
+			tmp = intel_de_read(display,
+					    DPLL_MD(display, crtc->pipe));
 
 		hw_state->dpll_md = tmp;
 	}
 
-	hw_state->dpll = intel_de_read(dev_priv, DPLL(dev_priv, crtc->pipe));
+	hw_state->dpll = intel_de_read(display, DPLL(display, crtc->pipe));
 
-	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) {
-		hw_state->fp0 = intel_de_read(dev_priv, FP0(crtc->pipe));
-		hw_state->fp1 = intel_de_read(dev_priv, FP1(crtc->pipe));
+	if (!display->platform.valleyview && !display->platform.cherryview) {
+		hw_state->fp0 = intel_de_read(display, FP0(crtc->pipe));
+		hw_state->fp1 = intel_de_read(display, FP1(crtc->pipe));
 	} else {
 		/* Mask out read-only status bits. */
 		hw_state->dpll &= ~(DPLL_LOCK_VLV |
@@ -421,8 +421,8 @@ void i9xx_dpll_get_hw_state(struct intel_crtc *crtc,
 /* Returns the clock of the currently programmed mode of the given pipe. */
 void i9xx_crtc_clock_get(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 	u32 dpll = hw_state->dpll;
 	u32 fp;
@@ -436,7 +436,7 @@ void i9xx_crtc_clock_get(struct intel_crtc_state *crtc_state)
 		fp = hw_state->fp1;
 
 	clock.m1 = (fp & FP_M1_DIV_MASK) >> FP_M1_DIV_SHIFT;
-	if (IS_PINEVIEW(dev_priv)) {
+	if (display->platform.pineview) {
 		clock.n = ffs((fp & FP_N_PINEVIEW_DIV_MASK) >> FP_N_DIV_SHIFT) - 1;
 		clock.m2 = (fp & FP_M2_PINEVIEW_DIV_MASK) >> FP_M2_DIV_SHIFT;
 	} else {
@@ -444,8 +444,8 @@ void i9xx_crtc_clock_get(struct intel_crtc_state *crtc_state)
 		clock.m2 = (fp & FP_M2_DIV_MASK) >> FP_M2_DIV_SHIFT;
 	}
 
-	if (DISPLAY_VER(dev_priv) != 2) {
-		if (IS_PINEVIEW(dev_priv))
+	if (DISPLAY_VER(display) != 2) {
+		if (display->platform.pineview)
 			clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK_PINEVIEW) >>
 				DPLL_FPA01_P1_POST_DIV_SHIFT_PINEVIEW);
 		else
@@ -462,23 +462,23 @@ void i9xx_crtc_clock_get(struct intel_crtc_state *crtc_state)
 				7 : 14;
 			break;
 		default:
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "Unknown DPLL mode %08x in programmed "
 				    "mode\n", (int)(dpll & DPLL_MODE_MASK));
 			return;
 		}
 
-		if (IS_PINEVIEW(dev_priv))
+		if (display->platform.pineview)
 			port_clock = pnv_calc_dpll_params(refclk, &clock);
 		else
 			port_clock = i9xx_calc_dpll_params(refclk, &clock);
 	} else {
 		enum pipe lvds_pipe;
 
-		if (IS_I85X(dev_priv) &&
-		    intel_lvds_port_enabled(dev_priv, LVDS, &lvds_pipe) &&
+		if (display->platform.i85x &&
+		    intel_lvds_port_enabled(display, LVDS, &lvds_pipe) &&
 		    lvds_pipe == crtc->pipe) {
-			u32 lvds = intel_de_read(dev_priv, LVDS);
+			u32 lvds = intel_de_read(display, LVDS);
 
 			clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK_I830_LVDS) >>
 				       DPLL_FPA01_P1_POST_DIV_SHIFT);
@@ -577,7 +577,7 @@ void chv_crtc_clock_get(struct intel_crtc_state *crtc_state)
  * Returns whether the given set of divisors are valid for a given refclk with
  * the given connectors.
  */
-static bool intel_pll_is_valid(struct drm_i915_private *dev_priv,
+static bool intel_pll_is_valid(struct intel_display *display,
 			       const struct intel_limit *limit,
 			       const struct dpll *clock)
 {
@@ -590,14 +590,14 @@ static bool intel_pll_is_valid(struct drm_i915_private *dev_priv,
 	if (clock->m1 < limit->m1.min || limit->m1.max < clock->m1)
 		return false;
 
-	if (!IS_PINEVIEW(dev_priv) &&
-	    !IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv) &&
-	    !IS_BROXTON(dev_priv) && !IS_GEMINILAKE(dev_priv))
+	if (!display->platform.pineview &&
+	    !display->platform.valleyview && !display->platform.cherryview &&
+	    !display->platform.broxton && !display->platform.geminilake)
 		if (clock->m1 <= clock->m2)
 			return false;
 
-	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv) &&
-	    !IS_BROXTON(dev_priv) && !IS_GEMINILAKE(dev_priv)) {
+	if (!display->platform.valleyview && !display->platform.cherryview &&
+	    !display->platform.broxton && !display->platform.geminilake) {
 		if (clock->p < limit->p.min || limit->p.max < clock->p)
 			return false;
 		if (clock->m < limit->m.min || limit->m.max < clock->m)
@@ -620,7 +620,7 @@ i9xx_select_p2_div(const struct intel_limit *limit,
 		   const struct intel_crtc_state *crtc_state,
 		   int target)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		/*
@@ -628,7 +628,7 @@ i9xx_select_p2_div(const struct intel_limit *limit,
 		 * We haven't figured out how to reliably set up different
 		 * single/dual channel state, if we even can.
 		 */
-		if (intel_is_dual_link_lvds(dev_priv))
+		if (intel_is_dual_link_lvds(display))
 			return limit->p2.p2_fast;
 		else
 			return limit->p2.p2_slow;
@@ -656,7 +656,7 @@ i9xx_find_best_dpll(const struct intel_limit *limit,
 		    const struct dpll *match_clock,
 		    struct dpll *best_clock)
 {
-	struct drm_device *dev = crtc_state->uapi.crtc->dev;
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct dpll clock;
 	int err = target;
 
@@ -677,7 +677,7 @@ i9xx_find_best_dpll(const struct intel_limit *limit,
 					int this_err;
 
 					i9xx_calc_dpll_params(refclk, &clock);
-					if (!intel_pll_is_valid(to_i915(dev),
+					if (!intel_pll_is_valid(display,
 								limit,
 								&clock))
 						continue;
@@ -714,7 +714,7 @@ pnv_find_best_dpll(const struct intel_limit *limit,
 		   const struct dpll *match_clock,
 		   struct dpll *best_clock)
 {
-	struct drm_device *dev = crtc_state->uapi.crtc->dev;
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct dpll clock;
 	int err = target;
 
@@ -733,7 +733,7 @@ pnv_find_best_dpll(const struct intel_limit *limit,
 					int this_err;
 
 					pnv_calc_dpll_params(refclk, &clock);
-					if (!intel_pll_is_valid(to_i915(dev),
+					if (!intel_pll_is_valid(display,
 								limit,
 								&clock))
 						continue;
@@ -770,7 +770,7 @@ g4x_find_best_dpll(const struct intel_limit *limit,
 		   const struct dpll *match_clock,
 		   struct dpll *best_clock)
 {
-	struct drm_device *dev = crtc_state->uapi.crtc->dev;
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct dpll clock;
 	int max_n;
 	bool found = false;
@@ -794,7 +794,7 @@ g4x_find_best_dpll(const struct intel_limit *limit,
 					int this_err;
 
 					i9xx_calc_dpll_params(refclk, &clock);
-					if (!intel_pll_is_valid(to_i915(dev),
+					if (!intel_pll_is_valid(display,
 								limit,
 								&clock))
 						continue;
@@ -817,7 +817,7 @@ g4x_find_best_dpll(const struct intel_limit *limit,
  * Check if the calculated PLL configuration is more optimal compared to the
  * best configuration and error found so far. Return the calculated error.
  */
-static bool vlv_PLL_is_optimal(struct drm_device *dev, int target_freq,
+static bool vlv_PLL_is_optimal(struct intel_display *display, int target_freq,
 			       const struct dpll *calculated_clock,
 			       const struct dpll *best_clock,
 			       unsigned int best_error_ppm,
@@ -827,13 +827,13 @@ static bool vlv_PLL_is_optimal(struct drm_device *dev, int target_freq,
 	 * For CHV ignore the error and consider only the P value.
 	 * Prefer a bigger P value based on HW requirements.
 	 */
-	if (IS_CHERRYVIEW(to_i915(dev))) {
+	if (display->platform.cherryview) {
 		*error_ppm = 0;
 
 		return calculated_clock->p > best_clock->p;
 	}
 
-	if (drm_WARN_ON_ONCE(dev, !target_freq))
+	if (drm_WARN_ON_ONCE(display->drm, !target_freq))
 		return false;
 
 	*error_ppm = div_u64(1000000ULL *
@@ -864,8 +864,7 @@ vlv_find_best_dpll(const struct intel_limit *limit,
 		   const struct dpll *match_clock,
 		   struct dpll *best_clock)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_device *dev = crtc->base.dev;
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct dpll clock;
 	unsigned int bestppm = 1000000;
 	/* min update 19.2 MHz */
@@ -889,12 +888,12 @@ vlv_find_best_dpll(const struct intel_limit *limit,
 
 					vlv_calc_dpll_params(refclk, &clock);
 
-					if (!intel_pll_is_valid(to_i915(dev),
+					if (!intel_pll_is_valid(display,
 								limit,
 								&clock))
 						continue;
 
-					if (!vlv_PLL_is_optimal(dev, target,
+					if (!vlv_PLL_is_optimal(display, target,
 								&clock,
 								best_clock,
 								bestppm, &ppm))
@@ -922,8 +921,7 @@ chv_find_best_dpll(const struct intel_limit *limit,
 		   const struct dpll *match_clock,
 		   struct dpll *best_clock)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_device *dev = crtc->base.dev;
+	struct intel_display *display = to_intel_display(crtc_state);
 	unsigned int best_error_ppm;
 	struct dpll clock;
 	u64 m2;
@@ -958,10 +956,10 @@ chv_find_best_dpll(const struct intel_limit *limit,
 
 			chv_calc_dpll_params(refclk, &clock);
 
-			if (!intel_pll_is_valid(to_i915(dev), limit, &clock))
+			if (!intel_pll_is_valid(display, limit, &clock))
 				continue;
 
-			if (!vlv_PLL_is_optimal(dev, target, &clock, best_clock,
+			if (!vlv_PLL_is_optimal(display, target, &clock, best_clock,
 						best_error_ppm, &error_ppm))
 				continue;
 
@@ -1005,8 +1003,6 @@ static u32 i9xx_dpll(const struct intel_crtc_state *crtc_state,
 		     const struct dpll *reduced_clock)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 dpll;
 
 	dpll = DPLL_VCO_ENABLE | DPLL_VGA_MODE_DIS;
@@ -1016,8 +1012,8 @@ static u32 i9xx_dpll(const struct intel_crtc_state *crtc_state,
 	else
 		dpll |= DPLLB_MODE_DAC_SERIAL;
 
-	if (IS_I945G(dev_priv) || IS_I945GM(dev_priv) ||
-	    IS_G33(dev_priv) || IS_PINEVIEW(dev_priv)) {
+	if (display->platform.i945g || display->platform.i945gm ||
+	    display->platform.g33 || display->platform.pineview) {
 		dpll |= (crtc_state->pixel_multiplier - 1)
 			<< SDVO_MULTIPLIER_SHIFT_HIRES;
 	}
@@ -1030,10 +1026,10 @@ static u32 i9xx_dpll(const struct intel_crtc_state *crtc_state,
 		dpll |= DPLL_SDVO_HIGH_SPEED;
 
 	/* compute bitmask from p1 value */
-	if (IS_G4X(dev_priv)) {
+	if (display->platform.g4x) {
 		dpll |= (1 << (clock->p1 - 1)) << DPLL_FPA01_P1_POST_DIV_SHIFT;
 		dpll |= (1 << (reduced_clock->p1 - 1)) << DPLL_FPA1_P1_POST_DIV_SHIFT;
-	} else if (IS_PINEVIEW(dev_priv)) {
+	} else if (display->platform.pineview) {
 		dpll |= (1 << (clock->p1 - 1)) << DPLL_FPA01_P1_POST_DIV_SHIFT_PINEVIEW;
 		WARN_ON(reduced_clock->p1 != clock->p1);
 	} else {
@@ -1057,7 +1053,7 @@ static u32 i9xx_dpll(const struct intel_crtc_state *crtc_state,
 	}
 	WARN_ON(reduced_clock->p2 != clock->p2);
 
-	if (DISPLAY_VER(dev_priv) >= 4)
+	if (DISPLAY_VER(display) >= 4)
 		dpll |= (6 << PLL_LOAD_PULSE_PHASE_SHIFT);
 
 	if (crtc_state->sdvo_tv_clock)
@@ -1075,11 +1071,10 @@ static void i9xx_compute_dpll(struct intel_crtc_state *crtc_state,
 			      const struct dpll *clock,
 			      const struct dpll *reduced_clock)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 
-	if (IS_PINEVIEW(dev_priv)) {
+	if (display->platform.pineview) {
 		hw_state->fp0 = pnv_dpll_compute_fp(clock);
 		hw_state->fp1 = pnv_dpll_compute_fp(reduced_clock);
 	} else {
@@ -1089,7 +1084,7 @@ static void i9xx_compute_dpll(struct intel_crtc_state *crtc_state,
 
 	hw_state->dpll = i9xx_dpll(crtc_state, clock, reduced_clock);
 
-	if (DISPLAY_VER(dev_priv) >= 4)
+	if (DISPLAY_VER(display) >= 4)
 		hw_state->dpll_md = i965_dpll_md(crtc_state);
 }
 
@@ -1098,8 +1093,6 @@ static u32 i8xx_dpll(const struct intel_crtc_state *crtc_state,
 		     const struct dpll *reduced_clock)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 dpll;
 
 	dpll = DPLL_VCO_ENABLE | DPLL_VGA_MODE_DIS;
@@ -1129,7 +1122,7 @@ static u32 i8xx_dpll(const struct intel_crtc_state *crtc_state,
 	 * both DPLLS. The spec says we should disable the DVO 2X clock
 	 * when not needed, but this seems to work fine in practice.
 	 */
-	if (IS_I830(dev_priv) ||
+	if (display->platform.i830 ||
 	    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DVO))
 		dpll |= DPLL_DVO_2X_MODE;
 
@@ -1157,14 +1150,14 @@ static void i8xx_compute_dpll(struct intel_crtc_state *crtc_state,
 static int hsw_crtc_compute_clock(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_encoder *encoder =
 		intel_get_crtc_new_encoder(state, crtc_state);
 	int ret;
 
-	if (DISPLAY_VER(dev_priv) < 11 &&
+	if (DISPLAY_VER(display) < 11 &&
 	    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI))
 		return 0;
 
@@ -1186,13 +1179,13 @@ static int hsw_crtc_compute_clock(struct intel_atomic_state *state,
 static int hsw_crtc_get_shared_dpll(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_encoder *encoder =
 		intel_get_crtc_new_encoder(state, crtc_state);
 
-	if (DISPLAY_VER(dev_priv) < 11 &&
+	if (DISPLAY_VER(display) < 11 &&
 	    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI))
 		return 0;
 
@@ -1241,12 +1234,10 @@ static int mtl_crtc_compute_clock(struct intel_atomic_state *state,
 static int ilk_fb_cb_factor(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS) &&
-	    ((intel_panel_use_ssc(display) && i915->display.vbt.lvds_ssc_freq == 100000) ||
-	     (HAS_PCH_IBX(i915) && intel_is_dual_link_lvds(i915))))
+	    ((intel_panel_use_ssc(display) && display->vbt.lvds_ssc_freq == 100000) ||
+	     (HAS_PCH_IBX(display) && intel_is_dual_link_lvds(display))))
 		return 25;
 
 	if (crtc_state->sdvo_tv_clock)
@@ -1276,8 +1267,6 @@ static u32 ilk_dpll(const struct intel_crtc_state *crtc_state,
 		    const struct dpll *reduced_clock)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 dpll;
 
 	dpll = DPLL_VCO_ENABLE;
@@ -1311,7 +1300,7 @@ static u32 ilk_dpll(const struct intel_crtc_state *crtc_state,
 	 * clear if it''s a win or loss power wise. No point in doing
 	 * this on ILK at all since it has a fixed DPLL<->pipe mapping.
 	 */
-	if (INTEL_NUM_PIPES(dev_priv) == 3 &&
+	if (INTEL_NUM_PIPES(display) == 3 &&
 	    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG))
 		dpll |= DPLL_SDVO_HIGH_SPEED;
 
@@ -1362,7 +1351,6 @@ static int ilk_crtc_compute_clock(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_limit *limit;
@@ -1375,13 +1363,13 @@ static int ilk_crtc_compute_clock(struct intel_atomic_state *state,
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		if (intel_panel_use_ssc(display)) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "using SSC reference clock of %d kHz\n",
-				    dev_priv->display.vbt.lvds_ssc_freq);
-			refclk = dev_priv->display.vbt.lvds_ssc_freq;
+				    display->vbt.lvds_ssc_freq);
+			refclk = display->vbt.lvds_ssc_freq;
 		}
 
-		if (intel_is_dual_link_lvds(dev_priv)) {
+		if (intel_is_dual_link_lvds(display)) {
 			if (refclk == 100000)
 				limit = &ilk_limits_dual_lvds_100m;
 			else
@@ -1539,7 +1527,6 @@ static int g4x_crtc_compute_clock(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_limit *limit;
@@ -1547,13 +1534,13 @@ static int g4x_crtc_compute_clock(struct intel_atomic_state *state,
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		if (intel_panel_use_ssc(display)) {
-			refclk = dev_priv->display.vbt.lvds_ssc_freq;
-			drm_dbg_kms(&dev_priv->drm,
+			refclk = display->vbt.lvds_ssc_freq;
+			drm_dbg_kms(display->drm,
 				    "using SSC reference clock of %d kHz\n",
 				    refclk);
 		}
 
-		if (intel_is_dual_link_lvds(dev_priv))
+		if (intel_is_dual_link_lvds(display))
 			limit = &intel_limits_g4x_dual_channel_lvds;
 		else
 			limit = &intel_limits_g4x_single_channel_lvds;
@@ -1589,7 +1576,6 @@ static int pnv_crtc_compute_clock(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_limit *limit;
@@ -1597,8 +1583,8 @@ static int pnv_crtc_compute_clock(struct intel_atomic_state *state,
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		if (intel_panel_use_ssc(display)) {
-			refclk = dev_priv->display.vbt.lvds_ssc_freq;
-			drm_dbg_kms(&dev_priv->drm,
+			refclk = display->vbt.lvds_ssc_freq;
+			drm_dbg_kms(display->drm,
 				    "using SSC reference clock of %d kHz\n",
 				    refclk);
 		}
@@ -1628,7 +1614,6 @@ static int i9xx_crtc_compute_clock(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_limit *limit;
@@ -1636,8 +1621,8 @@ static int i9xx_crtc_compute_clock(struct intel_atomic_state *state,
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		if (intel_panel_use_ssc(display)) {
-			refclk = dev_priv->display.vbt.lvds_ssc_freq;
-			drm_dbg_kms(&dev_priv->drm,
+			refclk = display->vbt.lvds_ssc_freq;
+			drm_dbg_kms(display->drm,
 				    "using SSC reference clock of %d kHz\n",
 				    refclk);
 		}
@@ -1669,7 +1654,6 @@ static int i8xx_crtc_compute_clock(struct intel_atomic_state *state,
 				   struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_limit *limit;
@@ -1677,8 +1661,8 @@ static int i8xx_crtc_compute_clock(struct intel_atomic_state *state,
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_LVDS)) {
 		if (intel_panel_use_ssc(display)) {
-			refclk = dev_priv->display.vbt.lvds_ssc_freq;
-			drm_dbg_kms(&dev_priv->drm,
+			refclk = display->vbt.lvds_ssc_freq;
+			drm_dbg_kms(display->drm,
 				    "using SSC reference clock of %d kHz\n",
 				    refclk);
 		}
@@ -1751,12 +1735,12 @@ static const struct intel_dpll_funcs i8xx_dpll_funcs = {
 int intel_dpll_crtc_compute_clock(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	int ret;
 
-	drm_WARN_ON(&i915->drm, !intel_crtc_needs_modeset(crtc_state));
+	drm_WARN_ON(display->drm, !intel_crtc_needs_modeset(crtc_state));
 
 	memset(&crtc_state->dpll_hw_state, 0,
 	       sizeof(crtc_state->dpll_hw_state));
@@ -1764,9 +1748,9 @@ int intel_dpll_crtc_compute_clock(struct intel_atomic_state *state,
 	if (!crtc_state->hw.enable)
 		return 0;
 
-	ret = i915->display.funcs.dpll->crtc_compute_clock(state, crtc);
+	ret = display->funcs.dpll->crtc_compute_clock(state, crtc);
 	if (ret) {
-		drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] Couldn't calculate DPLL settings\n",
+		drm_dbg_kms(display->drm, "[CRTC:%d:%s] Couldn't calculate DPLL settings\n",
 			    crtc->base.base.id, crtc->base.name);
 		return ret;
 	}
@@ -1777,23 +1761,23 @@ int intel_dpll_crtc_compute_clock(struct intel_atomic_state *state,
 int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	int ret;
 
-	drm_WARN_ON(&i915->drm, !intel_crtc_needs_modeset(crtc_state));
-	drm_WARN_ON(&i915->drm, !crtc_state->hw.enable && crtc_state->shared_dpll);
+	drm_WARN_ON(display->drm, !intel_crtc_needs_modeset(crtc_state));
+	drm_WARN_ON(display->drm, !crtc_state->hw.enable && crtc_state->shared_dpll);
 
 	if (!crtc_state->hw.enable || crtc_state->shared_dpll)
 		return 0;
 
-	if (!i915->display.funcs.dpll->crtc_get_shared_dpll)
+	if (!display->funcs.dpll->crtc_get_shared_dpll)
 		return 0;
 
-	ret = i915->display.funcs.dpll->crtc_get_shared_dpll(state, crtc);
+	ret = display->funcs.dpll->crtc_get_shared_dpll(state, crtc);
 	if (ret) {
-		drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] Couldn't get a shared DPLL\n",
+		drm_dbg_kms(display->drm, "[CRTC:%d:%s] Couldn't get a shared DPLL\n",
 			    crtc->base.base.id, crtc->base.name);
 		return ret;
 	}
@@ -1802,43 +1786,42 @@ int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state,
 }
 
 void
-intel_dpll_init_clock_hook(struct drm_i915_private *dev_priv)
-{
-	if (DISPLAY_VER(dev_priv) >= 14)
-		dev_priv->display.funcs.dpll = &mtl_dpll_funcs;
-	else if (IS_DG2(dev_priv))
-		dev_priv->display.funcs.dpll = &dg2_dpll_funcs;
-	else if (DISPLAY_VER(dev_priv) >= 9 || HAS_DDI(dev_priv))
-		dev_priv->display.funcs.dpll = &hsw_dpll_funcs;
-	else if (HAS_PCH_SPLIT(dev_priv))
-		dev_priv->display.funcs.dpll = &ilk_dpll_funcs;
-	else if (IS_CHERRYVIEW(dev_priv))
-		dev_priv->display.funcs.dpll = &chv_dpll_funcs;
-	else if (IS_VALLEYVIEW(dev_priv))
-		dev_priv->display.funcs.dpll = &vlv_dpll_funcs;
-	else if (IS_G4X(dev_priv))
-		dev_priv->display.funcs.dpll = &g4x_dpll_funcs;
-	else if (IS_PINEVIEW(dev_priv))
-		dev_priv->display.funcs.dpll = &pnv_dpll_funcs;
-	else if (DISPLAY_VER(dev_priv) != 2)
-		dev_priv->display.funcs.dpll = &i9xx_dpll_funcs;
+intel_dpll_init_clock_hook(struct intel_display *display)
+{
+	if (DISPLAY_VER(display) >= 14)
+		display->funcs.dpll = &mtl_dpll_funcs;
+	else if (display->platform.dg2)
+		display->funcs.dpll = &dg2_dpll_funcs;
+	else if (DISPLAY_VER(display) >= 9 || HAS_DDI(display))
+		display->funcs.dpll = &hsw_dpll_funcs;
+	else if (HAS_PCH_SPLIT(display))
+		display->funcs.dpll = &ilk_dpll_funcs;
+	else if (display->platform.cherryview)
+		display->funcs.dpll = &chv_dpll_funcs;
+	else if (display->platform.valleyview)
+		display->funcs.dpll = &vlv_dpll_funcs;
+	else if (display->platform.g4x)
+		display->funcs.dpll = &g4x_dpll_funcs;
+	else if (display->platform.pineview)
+		display->funcs.dpll = &pnv_dpll_funcs;
+	else if (DISPLAY_VER(display) != 2)
+		display->funcs.dpll = &i9xx_dpll_funcs;
 	else
-		dev_priv->display.funcs.dpll = &i8xx_dpll_funcs;
+		display->funcs.dpll = &i8xx_dpll_funcs;
 }
 
-static bool i9xx_has_pps(struct drm_i915_private *dev_priv)
+static bool i9xx_has_pps(struct intel_display *display)
 {
-	if (IS_I830(dev_priv))
+	if (display->platform.i830)
 		return false;
 
-	return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv);
+	return display->platform.pineview || display->platform.mobile;
 }
 
 void i9xx_enable_pll(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 	enum pipe pipe = crtc->pipe;
 	int i;
@@ -1846,27 +1829,27 @@ void i9xx_enable_pll(const struct intel_crtc_state *crtc_state)
 	assert_transcoder_disabled(display, crtc_state->cpu_transcoder);
 
 	/* PLL is protected by panel, make sure we can write it */
-	if (i9xx_has_pps(dev_priv))
+	if (i9xx_has_pps(display))
 		assert_pps_unlocked(display, pipe);
 
-	intel_de_write(dev_priv, FP0(pipe), hw_state->fp0);
-	intel_de_write(dev_priv, FP1(pipe), hw_state->fp1);
+	intel_de_write(display, FP0(pipe), hw_state->fp0);
+	intel_de_write(display, FP1(pipe), hw_state->fp1);
 
 	/*
 	 * Apparently we need to have VGA mode enabled prior to changing
 	 * the P1/P2 dividers. Otherwise the DPLL will keep using the old
 	 * dividers, even though the register value does change.
 	 */
-	intel_de_write(dev_priv, DPLL(dev_priv, pipe),
+	intel_de_write(display, DPLL(display, pipe),
 		       hw_state->dpll & ~DPLL_VGA_MODE_DIS);
-	intel_de_write(dev_priv, DPLL(dev_priv, pipe), hw_state->dpll);
+	intel_de_write(display, DPLL(display, pipe), hw_state->dpll);
 
 	/* Wait for the clocks to stabilize. */
-	intel_de_posting_read(dev_priv, DPLL(dev_priv, pipe));
+	intel_de_posting_read(display, DPLL(display, pipe));
 	udelay(150);
 
-	if (DISPLAY_VER(dev_priv) >= 4) {
-		intel_de_write(dev_priv, DPLL_MD(dev_priv, pipe),
+	if (DISPLAY_VER(display) >= 4) {
+		intel_de_write(display, DPLL_MD(display, pipe),
 			       hw_state->dpll_md);
 	} else {
 		/* The pixel multiplier can only be updated once the
@@ -1874,20 +1857,21 @@ void i9xx_enable_pll(const struct intel_crtc_state *crtc_state)
 		 *
 		 * So write it again.
 		 */
-		intel_de_write(dev_priv, DPLL(dev_priv, pipe), hw_state->dpll);
+		intel_de_write(display, DPLL(display, pipe), hw_state->dpll);
 	}
 
 	/* We do this three times for luck */
 	for (i = 0; i < 3; i++) {
-		intel_de_write(dev_priv, DPLL(dev_priv, pipe), hw_state->dpll);
-		intel_de_posting_read(dev_priv, DPLL(dev_priv, pipe));
+		intel_de_write(display, DPLL(display, pipe), hw_state->dpll);
+		intel_de_posting_read(display, DPLL(display, pipe));
 		udelay(150); /* wait for warmup */
 	}
 }
 
-static void vlv_pllb_recal_opamp(struct drm_i915_private *dev_priv,
+static void vlv_pllb_recal_opamp(struct intel_display *display,
 				 enum dpio_phy phy, enum dpio_channel ch)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 tmp;
 
 	/*
@@ -1916,6 +1900,7 @@ static void vlv_pllb_recal_opamp(struct drm_i915_private *dev_priv,
 
 static void vlv_prepare_pll(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct dpll *clock = &crtc_state->dpll;
@@ -1930,7 +1915,7 @@ static void vlv_prepare_pll(const struct intel_crtc_state *crtc_state)
 
 	/* PLL B needs special handling */
 	if (pipe == PIPE_B)
-		vlv_pllb_recal_opamp(dev_priv, phy, ch);
+		vlv_pllb_recal_opamp(display, phy, ch);
 
 	/* Set up Tx target for periodic Rcomp update */
 	vlv_dpio_write(dev_priv, phy, VLV_PCS_DW17_BCAST, 0x0100000f);
@@ -2003,24 +1988,23 @@ static void vlv_prepare_pll(const struct intel_crtc_state *crtc_state)
 
 static void _vlv_enable_pll(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 	enum pipe pipe = crtc->pipe;
 
-	intel_de_write(dev_priv, DPLL(dev_priv, pipe), hw_state->dpll);
-	intel_de_posting_read(dev_priv, DPLL(dev_priv, pipe));
+	intel_de_write(display, DPLL(display, pipe), hw_state->dpll);
+	intel_de_posting_read(display, DPLL(display, pipe));
 	udelay(150);
 
-	if (intel_de_wait_for_set(dev_priv, DPLL(dev_priv, pipe), DPLL_LOCK_VLV, 1))
-		drm_err(&dev_priv->drm, "DPLL %d failed to lock\n", pipe);
+	if (intel_de_wait_for_set(display, DPLL(display, pipe), DPLL_LOCK_VLV, 1))
+		drm_err(display->drm, "DPLL %d failed to lock\n", pipe);
 }
 
 void vlv_enable_pll(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 	enum pipe pipe = crtc->pipe;
 
@@ -2030,7 +2014,7 @@ void vlv_enable_pll(const struct intel_crtc_state *crtc_state)
 	assert_pps_unlocked(display, pipe);
 
 	/* Enable Refclk */
-	intel_de_write(dev_priv, DPLL(dev_priv, pipe),
+	intel_de_write(display, DPLL(display, pipe),
 		       hw_state->dpll & ~(DPLL_VCO_ENABLE | DPLL_EXT_BUFFER_ENABLE_VLV));
 
 	if (hw_state->dpll & DPLL_VCO_ENABLE) {
@@ -2038,8 +2022,8 @@ void vlv_enable_pll(const struct intel_crtc_state *crtc_state)
 		_vlv_enable_pll(crtc_state);
 	}
 
-	intel_de_write(dev_priv, DPLL_MD(dev_priv, pipe), hw_state->dpll_md);
-	intel_de_posting_read(dev_priv, DPLL_MD(dev_priv, pipe));
+	intel_de_write(display, DPLL_MD(display, pipe), hw_state->dpll_md);
+	intel_de_posting_read(display, DPLL_MD(display, pipe));
 }
 
 static void chv_prepare_pll(const struct intel_crtc_state *crtc_state)
@@ -2133,6 +2117,7 @@ static void chv_prepare_pll(const struct intel_crtc_state *crtc_state)
 
 static void _chv_enable_pll(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
@@ -2156,18 +2141,17 @@ static void _chv_enable_pll(const struct intel_crtc_state *crtc_state)
 	udelay(1);
 
 	/* Enable PLL */
-	intel_de_write(dev_priv, DPLL(dev_priv, pipe), hw_state->dpll);
+	intel_de_write(display, DPLL(display, pipe), hw_state->dpll);
 
 	/* Check PLL is locked */
-	if (intel_de_wait_for_set(dev_priv, DPLL(dev_priv, pipe), DPLL_LOCK_VLV, 1))
-		drm_err(&dev_priv->drm, "PLL %d failed to lock\n", pipe);
+	if (intel_de_wait_for_set(display, DPLL(display, pipe), DPLL_LOCK_VLV, 1))
+		drm_err(display->drm, "PLL %d failed to lock\n", pipe);
 }
 
 void chv_enable_pll(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx;
 	enum pipe pipe = crtc->pipe;
 
@@ -2177,7 +2161,7 @@ void chv_enable_pll(const struct intel_crtc_state *crtc_state)
 	assert_pps_unlocked(display, pipe);
 
 	/* Enable Refclk and SSC */
-	intel_de_write(dev_priv, DPLL(dev_priv, pipe),
+	intel_de_write(display, DPLL(display, pipe),
 		       hw_state->dpll & ~DPLL_VCO_ENABLE);
 
 	if (hw_state->dpll & DPLL_VCO_ENABLE) {
@@ -2192,29 +2176,29 @@ void chv_enable_pll(const struct intel_crtc_state *crtc_state)
 		 * DPLLCMD is AWOL. Use chicken bits to propagate
 		 * the value from DPLLBMD to either pipe B or C.
 		 */
-		intel_de_write(dev_priv, CBR4_VLV, CBR_DPLLBMD_PIPE(pipe));
-		intel_de_write(dev_priv, DPLL_MD(dev_priv, PIPE_B),
+		intel_de_write(display, CBR4_VLV, CBR_DPLLBMD_PIPE(pipe));
+		intel_de_write(display, DPLL_MD(display, PIPE_B),
 			       hw_state->dpll_md);
-		intel_de_write(dev_priv, CBR4_VLV, 0);
-		dev_priv->display.state.chv_dpll_md[pipe] = hw_state->dpll_md;
+		intel_de_write(display, CBR4_VLV, 0);
+		display->state.chv_dpll_md[pipe] = hw_state->dpll_md;
 
 		/*
 		 * DPLLB VGA mode also seems to cause problems.
 		 * We should always have it disabled.
 		 */
-		drm_WARN_ON(&dev_priv->drm,
-			    (intel_de_read(dev_priv, DPLL(dev_priv, PIPE_B)) &
+		drm_WARN_ON(display->drm,
+			    (intel_de_read(display, DPLL(display, PIPE_B)) &
 			     DPLL_VGA_MODE_DIS) == 0);
 	} else {
-		intel_de_write(dev_priv, DPLL_MD(dev_priv, pipe),
+		intel_de_write(display, DPLL_MD(display, pipe),
 			       hw_state->dpll_md);
-		intel_de_posting_read(dev_priv, DPLL_MD(dev_priv, pipe));
+		intel_de_posting_read(display, DPLL_MD(display, pipe));
 	}
 }
 
 /**
  * vlv_force_pll_on - forcibly enable just the PLL
- * @dev_priv: i915 private structure
+ * @display: display device
  * @pipe: pipe PLL to enable
  * @dpll: PLL configuration
  *
@@ -2222,10 +2206,9 @@ void chv_enable_pll(const struct intel_crtc_state *crtc_state)
  * in cases where we need the PLL enabled even when @pipe is not going to
  * be enabled.
  */
-int vlv_force_pll_on(struct drm_i915_private *dev_priv, enum pipe pipe,
+int vlv_force_pll_on(struct intel_display *display, enum pipe pipe,
 		     const struct dpll *dpll)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_crtc *crtc = intel_crtc_for_pipe(display, pipe);
 	struct intel_crtc_state *crtc_state;
 
@@ -2238,7 +2221,7 @@ int vlv_force_pll_on(struct drm_i915_private *dev_priv, enum pipe pipe,
 	crtc_state->dpll = *dpll;
 	crtc_state->output_types = BIT(INTEL_OUTPUT_EDP);
 
-	if (IS_CHERRYVIEW(dev_priv)) {
+	if (display->platform.cherryview) {
 		chv_compute_dpll(crtc_state);
 		chv_enable_pll(crtc_state);
 	} else {
@@ -2251,9 +2234,8 @@ int vlv_force_pll_on(struct drm_i915_private *dev_priv, enum pipe pipe,
 	return 0;
 }
 
-void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
+void vlv_disable_pll(struct intel_display *display, enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 val;
 
 	/* Make sure the pipe isn't still relying on us */
@@ -2268,9 +2250,9 @@ void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 	intel_de_posting_read(display, DPLL(display, pipe));
 }
 
-void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
+void chv_disable_pll(struct intel_display *display, enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum dpio_channel ch = vlv_pipe_to_channel(pipe);
 	enum dpio_phy phy = vlv_pipe_to_phy(pipe);
 	u32 val;
@@ -2316,18 +2298,18 @@ void i9xx_disable_pll(const struct intel_crtc_state *crtc_state)
 
 /**
  * vlv_force_pll_off - forcibly disable just the PLL
- * @dev_priv: i915 private structure
+ * @display: display device
  * @pipe: pipe PLL to disable
  *
  * Disable the PLL for @pipe. To be used in cases where we need
  * the PLL enabled even when @pipe is not going to be enabled.
  */
-void vlv_force_pll_off(struct drm_i915_private *dev_priv, enum pipe pipe)
+void vlv_force_pll_off(struct intel_display *display, enum pipe pipe)
 {
-	if (IS_CHERRYVIEW(dev_priv))
-		chv_disable_pll(dev_priv, pipe);
+	if (display->platform.cherryview)
+		chv_disable_pll(display, pipe);
 	else
-		vlv_disable_pll(dev_priv, pipe);
+		vlv_disable_pll(display, pipe);
 }
 
 /* Only for pre-ILK configs */
diff --git a/drivers/gpu/drm/i915/display/intel_dpll.h b/drivers/gpu/drm/i915/display/intel_dpll.h
index 21d06cbd2ce7..280e90a57c87 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll.h
+++ b/drivers/gpu/drm/i915/display/intel_dpll.h
@@ -8,16 +8,15 @@
 
 #include <linux/types.h>
 
+enum pipe;
 struct dpll;
-struct drm_i915_private;
 struct intel_atomic_state;
 struct intel_crtc;
 struct intel_crtc_state;
 struct intel_display;
 struct intel_dpll_hw_state;
-enum pipe;
 
-void intel_dpll_init_clock_hook(struct drm_i915_private *dev_priv);
+void intel_dpll_init_clock_hook(struct intel_display *display);
 int intel_dpll_crtc_compute_clock(struct intel_atomic_state *state,
 				  struct intel_crtc *crtc);
 int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state,
@@ -29,14 +28,14 @@ void i9xx_dpll_get_hw_state(struct intel_crtc *crtc,
 void vlv_compute_dpll(struct intel_crtc_state *crtc_state);
 void chv_compute_dpll(struct intel_crtc_state *crtc_state);
 
-int vlv_force_pll_on(struct drm_i915_private *dev_priv, enum pipe pipe,
+int vlv_force_pll_on(struct intel_display *display, enum pipe pipe,
 		     const struct dpll *dpll);
-void vlv_force_pll_off(struct drm_i915_private *dev_priv, enum pipe pipe);
+void vlv_force_pll_off(struct intel_display *display, enum pipe pipe);
 
 void chv_enable_pll(const struct intel_crtc_state *crtc_state);
-void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe);
+void chv_disable_pll(struct intel_display *display, enum pipe pipe);
 void vlv_enable_pll(const struct intel_crtc_state *crtc_state);
-void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe);
+void vlv_disable_pll(struct intel_display *display, enum pipe pipe);
 void i9xx_enable_pll(const struct intel_crtc_state *crtc_state);
 void i9xx_disable_pll(const struct intel_crtc_state *crtc_state);
 bool bxt_find_best_dpll(struct intel_crtc_state *crtc_state,
diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index c825a507b905..9da051a3f455 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -24,9 +24,11 @@
 #include <linux/math.h>
 #include <linux/string_helpers.h>
 
+#include <drm/drm_print.h>
+
 #include "bxt_dpio_phy_regs.h"
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_cx0_phy.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
@@ -38,6 +40,7 @@
 #include "intel_hti.h"
 #include "intel_mg_phy_regs.h"
 #include "intel_pch_refclk.h"
+#include "intel_step.h"
 #include "intel_tc.h"
 
 /**
@@ -257,7 +260,7 @@ void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state)
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct intel_shared_dpll *pll = crtc_state->shared_dpll;
-	unsigned int pipe_mask = BIT(crtc->pipe);
+	unsigned int pipe_mask = intel_crtc_joined_pipe_mask(crtc_state);
 	unsigned int old_mask;
 
 	if (drm_WARN_ON(display->drm, !pll))
@@ -303,7 +306,7 @@ void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state)
 	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct intel_shared_dpll *pll = crtc_state->shared_dpll;
-	unsigned int pipe_mask = BIT(crtc->pipe);
+	unsigned int pipe_mask = intel_crtc_joined_pipe_mask(crtc_state);
 
 	/* PCH only available on ILK+ */
 	if (DISPLAY_VER(display) < 5)
@@ -609,13 +612,12 @@ static int ibx_get_dpll(struct intel_atomic_state *state,
 			struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_shared_dpll *pll;
 	enum intel_dpll_id id;
 
-	if (HAS_PCH_IBX(i915)) {
+	if (HAS_PCH_IBX(display)) {
 		/* Ironlake PCH has a fixed PLL->PCH pipe mapping. */
 		id = (enum intel_dpll_id) crtc->pipe;
 		pll = intel_get_shared_dpll_by_id(display, id);
@@ -715,7 +717,6 @@ static void hsw_ddi_spll_enable(struct intel_display *display,
 static void hsw_ddi_wrpll_disable(struct intel_display *display,
 				  struct intel_shared_dpll *pll)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	const enum intel_dpll_id id = pll->info->id;
 
 	intel_de_rmw(display, WRPLL_CTL(id), WRPLL_PLL_ENABLE, 0);
@@ -726,13 +727,12 @@ static void hsw_ddi_wrpll_disable(struct intel_display *display,
 	 * that depend on it have been shut down.
 	 */
 	if (display->dpll.pch_ssc_use & BIT(id))
-		intel_init_pch_refclk(i915);
+		intel_init_pch_refclk(display);
 }
 
 static void hsw_ddi_spll_disable(struct intel_display *display,
 				 struct intel_shared_dpll *pll)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	enum intel_dpll_id id = pll->info->id;
 
 	intel_de_rmw(display, SPLL_CTL, SPLL_PLL_ENABLE, 0);
@@ -743,7 +743,7 @@ static void hsw_ddi_spll_disable(struct intel_display *display,
 	 * that depend on it have been shut down.
 	 */
 	if (display->dpll.pch_ssc_use & BIT(id))
-		intel_init_pch_refclk(i915);
+		intel_init_pch_refclk(display);
 }
 
 static bool hsw_ddi_wrpll_get_hw_state(struct intel_display *display,
@@ -2606,10 +2606,8 @@ ehl_combo_pll_div_frac_wa_needed(struct intel_display *display)
 {
 	return ((display->platform.elkhartlake &&
 		 IS_DISPLAY_STEP(display, STEP_B0, STEP_FOREVER)) ||
-		 display->platform.tigerlake ||
-		 display->platform.alderlake_s ||
-		 display->platform.alderlake_p) &&
-		 display->dpll.ref_clks.nssc == 38400;
+		DISPLAY_VER(display) >= 12) &&
+		display->dpll.ref_clks.nssc == 38400;
 }
 
 struct icl_combo_pll_params {
@@ -4309,7 +4307,6 @@ static const struct intel_dpll_mgr adlp_pll_mgr = {
  */
 void intel_shared_dpll_init(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	const struct intel_dpll_mgr *dpll_mgr = NULL;
 	const struct dpll_info *dpll_info;
 	int i;
@@ -4339,7 +4336,7 @@ void intel_shared_dpll_init(struct intel_display *display)
 		dpll_mgr = &skl_pll_mgr;
 	else if (HAS_DDI(display))
 		dpll_mgr = &hsw_pll_mgr;
-	else if (HAS_PCH_IBX(i915) || HAS_PCH_CPT(i915))
+	else if (HAS_PCH_IBX(display) || HAS_PCH_CPT(display))
 		dpll_mgr = &pch_pll_mgr;
 
 	if (!dpll_mgr)
diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c
index 0d8ebe38226e..43bd97e4f589 100644
--- a/drivers/gpu/drm/i915/display/intel_dpt.c
+++ b/drivers/gpu/drm/i915/display/intel_dpt.c
@@ -9,6 +9,7 @@
 #include "gt/gen8_ppgtt.h"
 
 #include "i915_drv.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dpt.h"
 #include "intel_fb.h"
@@ -127,7 +128,7 @@ struct i915_vma *intel_dpt_pin_to_ggtt(struct i915_address_space *vm,
 	struct drm_i915_private *i915 = vm->i915;
 	struct intel_display *display = &i915->display;
 	struct i915_dpt *dpt = i915_vm_to_dpt(vm);
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	struct i915_vma *vma;
 	void __iomem *iomem;
 	struct i915_gem_ww_ctx ww;
@@ -137,7 +138,7 @@ struct i915_vma *intel_dpt_pin_to_ggtt(struct i915_address_space *vm,
 	if (i915_gem_object_is_stolen(dpt->obj))
 		pin_flags |= PIN_MAPPABLE;
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 	atomic_inc(&display->restore.pending_fb_pin);
 
 	for_i915_gem_ww(&ww, err, true) {
@@ -169,7 +170,7 @@ struct i915_vma *intel_dpt_pin_to_ggtt(struct i915_address_space *vm,
 	dpt->obj->mm.dirty = true;
 
 	atomic_dec(&display->restore.pending_fb_pin);
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	return err ? ERR_PTR(err) : vma;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dpt_common.c b/drivers/gpu/drm/i915/display/intel_dpt_common.c
index d2dede0a5229..ce5aa0ca0fa5 100644
--- a/drivers/gpu/drm/i915/display/intel_dpt_common.c
+++ b/drivers/gpu/drm/i915/display/intel_dpt_common.c
@@ -3,7 +3,6 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
@@ -12,9 +11,9 @@
 
 void intel_dpt_configure(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 
-	if (DISPLAY_VER(i915) == 14) {
+	if (DISPLAY_VER(display) == 14) {
 		enum pipe pipe = crtc->pipe;
 		enum plane_id plane_id;
 
@@ -22,15 +21,15 @@ void intel_dpt_configure(struct intel_crtc *crtc)
 			if (plane_id == PLANE_CURSOR)
 				continue;
 
-			intel_de_rmw(i915, PLANE_CHICKEN(pipe, plane_id),
+			intel_de_rmw(display, PLANE_CHICKEN(pipe, plane_id),
 				     PLANE_CHICKEN_DISABLE_DPT,
-				     i915->display.params.enable_dpt ? 0 :
+				     display->params.enable_dpt ? 0 :
 				     PLANE_CHICKEN_DISABLE_DPT);
 		}
-	} else if (DISPLAY_VER(i915) == 13) {
-		intel_de_rmw(i915, CHICKEN_MISC_2,
+	} else if (DISPLAY_VER(display) == 13) {
+		intel_de_rmw(display, CHICKEN_MISC_2,
 			     CHICKEN_MISC_DISABLE_DPT,
-			     i915->display.params.enable_dpt ? 0 :
+			     display->params.enable_dpt ? 0 :
 			     CHICKEN_MISC_DISABLE_DPT);
 	}
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dsb.c b/drivers/gpu/drm/i915/display/intel_dsb.c
index 9fc4003d1579..481488d1fe67 100644
--- a/drivers/gpu/drm/i915/display/intel_dsb.c
+++ b/drivers/gpu/drm/i915/display/intel_dsb.c
@@ -4,13 +4,15 @@
  *
  */
 
+#include <drm/drm_print.h>
 #include <drm/drm_vblank.h>
 
-#include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_crtc.h"
 #include "intel_de.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dsb.h"
 #include "intel_dsb_buffer.h"
@@ -142,10 +144,10 @@ static int dsb_vtotal(struct intel_atomic_state *state,
 static int dsb_dewake_scanline_start(struct intel_atomic_state *state,
 				     struct intel_crtc *crtc)
 {
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_crtc_state *crtc_state =
 		intel_pre_commit_crtc_state(state, crtc);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
-	unsigned int latency = skl_watermark_max_latency(i915, 0);
+	unsigned int latency = skl_watermark_max_latency(display, 0);
 
 	return intel_mode_vdisplay(&crtc_state->hw.adjusted_mode) -
 		intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode, latency);
@@ -795,22 +797,22 @@ struct intel_dsb *intel_dsb_prepare(struct intel_atomic_state *state,
 				    enum intel_dsb_id dsb_id,
 				    unsigned int max_cmds)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
-	intel_wakeref_t wakeref;
+	struct intel_display *display = to_intel_display(state);
+	struct ref_tracker *wakeref;
 	struct intel_dsb *dsb;
 	unsigned int size;
 
-	if (!HAS_DSB(i915))
+	if (!HAS_DSB(display))
 		return NULL;
 
-	if (!i915->display.params.enable_dsb)
+	if (!display->params.enable_dsb)
 		return NULL;
 
 	dsb = kzalloc(sizeof(*dsb), GFP_KERNEL);
 	if (!dsb)
 		goto out;
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
 	/* ~1 qword per instruction, full cachelines */
 	size = ALIGN(max_cmds * 8, CACHELINE_BYTES);
@@ -818,7 +820,7 @@ struct intel_dsb *intel_dsb_prepare(struct intel_atomic_state *state,
 	if (!intel_dsb_buffer_create(crtc, &dsb->dsb_buf, size))
 		goto out_put_rpm;
 
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	dsb->id = dsb_id;
 	dsb->crtc = crtc;
@@ -831,10 +833,10 @@ struct intel_dsb *intel_dsb_prepare(struct intel_atomic_state *state,
 	return dsb;
 
 out_put_rpm:
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 	kfree(dsb);
 out:
-	drm_info_once(&i915->drm,
+	drm_info_once(display->drm,
 		      "[CRTC:%d:%s] DSB %d queue setup failed, will fallback to MMIO for display HW programming\n",
 		      crtc->base.base.id, crtc->base.name, dsb_id);
 
diff --git a/drivers/gpu/drm/i915/display/intel_dsi.c b/drivers/gpu/drm/i915/display/intel_dsi.c
index 403151175a87..a8f012119165 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi.c
+++ b/drivers/gpu/drm/i915/display/intel_dsi.c
@@ -4,8 +4,9 @@
  */
 
 #include <drm/drm_mipi_dsi.h>
+#include <drm/drm_print.h>
 
-#include "i915_drv.h"
+#include "intel_display_core.h"
 #include "intel_dsi.h"
 #include "intel_panel.h"
 
@@ -116,14 +117,14 @@ struct intel_dsi_host *intel_dsi_host_init(struct intel_dsi *intel_dsi,
 enum drm_panel_orientation
 intel_dsi_get_panel_orientation(struct intel_connector *connector)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 	enum drm_panel_orientation orientation;
 
 	orientation = connector->panel.vbt.dsi.orientation;
 	if (orientation != DRM_MODE_PANEL_ORIENTATION_UNKNOWN)
 		return orientation;
 
-	orientation = dev_priv->display.vbt.orientation;
+	orientation = display->vbt.orientation;
 	if (orientation != DRM_MODE_PANEL_ORIENTATION_UNKNOWN)
 		return orientation;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dsi_dcs_backlight.c b/drivers/gpu/drm/i915/display/intel_dsi_dcs_backlight.c
index 049443245310..b3c453bf7d5c 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_dcs_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_dsi_dcs_backlight.c
@@ -24,9 +24,10 @@
  */
 
 #include <drm/drm_mipi_dsi.h>
+#include <drm/drm_print.h>
 #include <video/mipi_display.h>
 
-#include "i915_drv.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_dsi.h"
 #include "intel_dsi_dcs_backlight.h"
@@ -162,7 +163,7 @@ static void dcs_enable_backlight(const struct intel_crtc_state *crtc_state,
 static int dcs_setup_backlight(struct intel_connector *connector,
 			       enum pipe unused)
 {
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 	struct intel_panel *panel = &connector->panel;
 
 	if (panel->vbt.backlight.brightness_precision_bits > 8)
@@ -172,7 +173,7 @@ static int dcs_setup_backlight(struct intel_connector *connector,
 
 	panel->backlight.level = panel->backlight.max;
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "[CONNECTOR:%d:%s] Using DCS for backlight control\n",
 		    connector->base.base.id, connector->base.name);
 
diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
index 7b2ffd14ae6e..29c920983413 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
@@ -31,16 +31,16 @@
 #include <linux/pinctrl/machine.h>
 #include <linux/slab.h>
 #include <linux/string_helpers.h>
-
 #include <linux/unaligned.h>
 
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 
 #include <video/mipi_display.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dsi.h"
@@ -102,13 +102,13 @@ static enum port intel_dsi_seq_port_to_port(struct intel_dsi *intel_dsi,
 static const u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi,
 				       const u8 *data)
 {
-	struct drm_i915_private *dev_priv = to_i915(intel_dsi->base.base.dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct mipi_dsi_device *dsi_device;
 	u8 type, flags, seq_port;
 	u16 len;
 	enum port port;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	flags = *data++;
 	type = *data++;
@@ -120,12 +120,12 @@ static const u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi,
 
 	port = intel_dsi_seq_port_to_port(intel_dsi, seq_port);
 
-	if (drm_WARN_ON(&dev_priv->drm, !intel_dsi->dsi_hosts[port]))
+	if (drm_WARN_ON(display->drm, !intel_dsi->dsi_hosts[port]))
 		goto out;
 
 	dsi_device = intel_dsi->dsi_hosts[port]->device;
 	if (!dsi_device) {
-		drm_dbg_kms(&dev_priv->drm, "no dsi device for port %c\n",
+		drm_dbg_kms(display->drm, "no dsi device for port %c\n",
 			    port_name(port));
 		goto out;
 	}
@@ -150,8 +150,7 @@ static const u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi,
 	case MIPI_DSI_GENERIC_READ_REQUEST_0_PARAM:
 	case MIPI_DSI_GENERIC_READ_REQUEST_1_PARAM:
 	case MIPI_DSI_GENERIC_READ_REQUEST_2_PARAM:
-		drm_dbg(&dev_priv->drm,
-			"Generic Read not yet implemented or used\n");
+		drm_dbg_kms(display->drm, "Generic Read not yet implemented or used\n");
 		break;
 	case MIPI_DSI_GENERIC_LONG_WRITE:
 		mipi_dsi_generic_write(dsi_device, data, len);
@@ -163,15 +162,14 @@ static const u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi,
 		mipi_dsi_dcs_write_buffer(dsi_device, data, 2);
 		break;
 	case MIPI_DSI_DCS_READ:
-		drm_dbg(&dev_priv->drm,
-			"DCS Read not yet implemented or used\n");
+		drm_dbg_kms(display->drm, "DCS Read not yet implemented or used\n");
 		break;
 	case MIPI_DSI_DCS_LONG_WRITE:
 		mipi_dsi_dcs_write_buffer(dsi_device, data, len);
 		break;
 	}
 
-	if (DISPLAY_VER(dev_priv) < 11)
+	if (DISPLAY_VER(display) < 11)
 		vlv_dsi_wait_for_fifo_empty(intel_dsi, port);
 
 out:
@@ -182,10 +180,10 @@ out:
 
 static const u8 *mipi_exec_delay(struct intel_dsi *intel_dsi, const u8 *data)
 {
-	struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	u32 delay = *((const u32 *) data);
 
-	drm_dbg_kms(&i915->drm, "%d usecs\n", delay);
+	drm_dbg_kms(display->drm, "%d usecs\n", delay);
 
 	usleep_range(delay, delay + 10);
 	data += 4;
@@ -196,7 +194,7 @@ static const u8 *mipi_exec_delay(struct intel_dsi *intel_dsi, const u8 *data)
 static void soc_gpio_set_value(struct intel_connector *connector, u8 gpio_index,
 			       const char *con_id, u8 idx, bool value)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 	/* XXX: this table is a quick ugly hack. */
 	static struct gpio_desc *soc_gpio_table[U8_MAX + 1];
 	struct gpio_desc *gpio_desc = soc_gpio_table[gpio_index];
@@ -204,10 +202,10 @@ static void soc_gpio_set_value(struct intel_connector *connector, u8 gpio_index,
 	if (gpio_desc) {
 		gpiod_set_value(gpio_desc, value);
 	} else {
-		gpio_desc = devm_gpiod_get_index(dev_priv->drm.dev, con_id, idx,
+		gpio_desc = devm_gpiod_get_index(display->drm->dev, con_id, idx,
 						 value ? GPIOD_OUT_HIGH : GPIOD_OUT_LOW);
 		if (IS_ERR(gpio_desc)) {
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"GPIO index %u request failed (%pe)\n",
 				gpio_index, gpio_desc);
 			return;
@@ -242,16 +240,16 @@ static void soc_opaque_gpio_set_value(struct intel_connector *connector,
 static void vlv_gpio_set_value(struct intel_connector *connector,
 			       u8 gpio_source, u8 gpio_index, bool value)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 
 	/* XXX: this assumes vlv_gpio_table only has NC GPIOs. */
 	if (connector->panel.vbt.dsi.seq_version < 3) {
 		if (gpio_source == 1) {
-			drm_dbg_kms(&dev_priv->drm, "SC gpio not supported\n");
+			drm_dbg_kms(display->drm, "SC gpio not supported\n");
 			return;
 		}
 		if (gpio_source > 1) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "unknown gpio source %u\n", gpio_source);
 			return;
 		}
@@ -264,7 +262,7 @@ static void vlv_gpio_set_value(struct intel_connector *connector,
 static void chv_gpio_set_value(struct intel_connector *connector,
 			       u8 gpio_source, u8 gpio_index, bool value)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 
 	if (connector->panel.vbt.dsi.seq_version >= 3) {
 		if (gpio_index >= CHV_GPIO_IDX_START_SE) {
@@ -284,13 +282,13 @@ static void chv_gpio_set_value(struct intel_connector *connector,
 	} else {
 		/* XXX: The spec is unclear about CHV GPIO on seq v2 */
 		if (gpio_source != 0) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "unknown gpio source %u\n", gpio_source);
 			return;
 		}
 
 		if (gpio_index >= CHV_GPIO_IDX_START_E) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "invalid gpio index %u for GPIO N\n",
 				    gpio_index);
 			return;
@@ -320,13 +318,12 @@ enum {
 	MIPI_VIO_EN_2,
 };
 
-static void icl_native_gpio_set_value(struct drm_i915_private *dev_priv,
+static void icl_native_gpio_set_value(struct intel_display *display,
 				      int gpio, bool value)
 {
-	struct intel_display *display = &dev_priv->display;
 	int index;
 
-	if (drm_WARN_ON(&dev_priv->drm, DISPLAY_VER(dev_priv) == 11 && gpio >= MIPI_RESET_2))
+	if (drm_WARN_ON(display->drm, DISPLAY_VER(display) == 11 && gpio >= MIPI_RESET_2))
 		return;
 
 	switch (gpio) {
@@ -343,25 +340,25 @@ static void icl_native_gpio_set_value(struct drm_i915_private *dev_priv,
 		 * The locking protects against concurrent SHOTPLUG_CTL_DDI
 		 * modifications in irq setup and handling.
 		 */
-		spin_lock_irq(&dev_priv->irq_lock);
-		intel_de_rmw(dev_priv, SHOTPLUG_CTL_DDI,
+		spin_lock_irq(&display->irq.lock);
+		intel_de_rmw(display, SHOTPLUG_CTL_DDI,
 			     SHOTPLUG_CTL_DDI_HPD_ENABLE(index) |
 			     SHOTPLUG_CTL_DDI_HPD_OUTPUT_DATA(index),
 			     value ? SHOTPLUG_CTL_DDI_HPD_OUTPUT_DATA(index) : 0);
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_unlock_irq(&display->irq.lock);
 		break;
 	case MIPI_AVDD_EN_1:
 	case MIPI_AVDD_EN_2:
 		index = gpio == MIPI_AVDD_EN_1 ? 0 : 1;
 
-		intel_de_rmw(dev_priv, PP_CONTROL(dev_priv, index), PANEL_POWER_ON,
+		intel_de_rmw(display, PP_CONTROL(display, index), PANEL_POWER_ON,
 			     value ? PANEL_POWER_ON : 0);
 		break;
 	case MIPI_BKLT_EN_1:
 	case MIPI_BKLT_EN_2:
 		index = gpio == MIPI_BKLT_EN_1 ? 0 : 1;
 
-		intel_de_rmw(dev_priv, PP_CONTROL(dev_priv, index), EDP_BLC_ENABLE,
+		intel_de_rmw(display, PP_CONTROL(display, index), EDP_BLC_ENABLE,
 			     value ? EDP_BLC_ENABLE : 0);
 		break;
 	case MIPI_AVEE_EN_1:
@@ -389,13 +386,12 @@ static void icl_native_gpio_set_value(struct drm_i915_private *dev_priv,
 
 static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 {
-	struct drm_device *dev = intel_dsi->base.base.dev;
-	struct drm_i915_private *i915 = to_i915(dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct intel_connector *connector = intel_dsi->attached_connector;
 	u8 gpio_source = 0, gpio_index = 0, gpio_number;
 	bool value;
 	int size;
-	bool native = DISPLAY_VER(i915) >= 11;
+	bool native = DISPLAY_VER(display) >= 11;
 
 	if (connector->panel.vbt.dsi.seq_version >= 3) {
 		size = 3;
@@ -416,16 +412,16 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 			gpio_source = (data[1] >> 1) & 3;
 	}
 
-	drm_dbg_kms(&i915->drm, "GPIO index %u, number %u, source %u, native %s, set to %s\n",
+	drm_dbg_kms(display->drm, "GPIO index %u, number %u, source %u, native %s, set to %s\n",
 		    gpio_index, gpio_number, gpio_source, str_yes_no(native), str_on_off(value));
 
 	if (native)
-		icl_native_gpio_set_value(i915, gpio_number, value);
-	else if (DISPLAY_VER(i915) >= 9)
+		icl_native_gpio_set_value(display, gpio_number, value);
+	else if (DISPLAY_VER(display) >= 9)
 		bxt_gpio_set_value(connector, gpio_index, value);
-	else if (IS_VALLEYVIEW(i915))
+	else if (display->platform.valleyview)
 		vlv_gpio_set_value(connector, gpio_source, gpio_number, value);
-	else if (IS_CHERRYVIEW(i915))
+	else if (display->platform.cherryview)
 		chv_gpio_set_value(connector, gpio_source, gpio_number, value);
 
 	return data + size;
@@ -463,8 +459,8 @@ static int i2c_adapter_lookup(struct acpi_resource *ares, void *data)
 static void i2c_acpi_find_adapter(struct intel_dsi *intel_dsi,
 				  const u16 target_addr)
 {
-	struct drm_device *drm_dev = intel_dsi->base.base.dev;
-	struct acpi_device *adev = ACPI_COMPANION(drm_dev->dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
+	struct acpi_device *adev = ACPI_COMPANION(display->drm->dev);
 	struct i2c_adapter_lookup lookup = {
 		.target_addr = target_addr,
 		.intel_dsi = intel_dsi,
@@ -484,7 +480,7 @@ static inline void i2c_acpi_find_adapter(struct intel_dsi *intel_dsi,
 
 static const u8 *mipi_exec_i2c(struct intel_dsi *intel_dsi, const u8 *data)
 {
-	struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct i2c_adapter *adapter;
 	struct i2c_msg msg;
 	int ret;
@@ -494,7 +490,7 @@ static const u8 *mipi_exec_i2c(struct intel_dsi *intel_dsi, const u8 *data)
 	u8 payload_size = *(data + 6);
 	u8 *payload_data;
 
-	drm_dbg_kms(&i915->drm, "bus %d target-addr 0x%02x reg 0x%02x data %*ph\n",
+	drm_dbg_kms(display->drm, "bus %d target-addr 0x%02x reg 0x%02x data %*ph\n",
 		    vbt_i2c_bus_num, target_addr, reg_offset, payload_size, data + 7);
 
 	if (intel_dsi->i2c_bus_num < 0) {
@@ -504,7 +500,7 @@ static const u8 *mipi_exec_i2c(struct intel_dsi *intel_dsi, const u8 *data)
 
 	adapter = i2c_get_adapter(intel_dsi->i2c_bus_num);
 	if (!adapter) {
-		drm_err(&i915->drm, "Cannot find a valid i2c bus for xfer\n");
+		drm_err(display->drm, "Cannot find a valid i2c bus for xfer\n");
 		goto err_bus;
 	}
 
@@ -522,7 +518,7 @@ static const u8 *mipi_exec_i2c(struct intel_dsi *intel_dsi, const u8 *data)
 
 	ret = i2c_transfer(adapter, &msg, 1);
 	if (ret < 0)
-		drm_err(&i915->drm,
+		drm_err(display->drm,
 			"Failed to xfer payload of size (%u) to reg (%u)\n",
 			payload_size, reg_offset);
 
@@ -535,16 +531,16 @@ err_bus:
 
 static const u8 *mipi_exec_spi(struct intel_dsi *intel_dsi, const u8 *data)
 {
-	struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 
-	drm_dbg_kms(&i915->drm, "Skipping SPI element execution\n");
+	drm_dbg_kms(display->drm, "Skipping SPI element execution\n");
 
 	return data + *(data + 5) + 6;
 }
 
 static const u8 *mipi_exec_pmic(struct intel_dsi *intel_dsi, const u8 *data)
 {
-	struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 #ifdef CONFIG_PMIC_OPREGION
 	u32 value, mask, reg_address;
 	u16 i2c_address;
@@ -560,9 +556,9 @@ static const u8 *mipi_exec_pmic(struct intel_dsi *intel_dsi, const u8 *data)
 							reg_address,
 							value, mask);
 	if (ret)
-		drm_err(&i915->drm, "%s failed, error: %d\n", __func__, ret);
+		drm_err(display->drm, "%s failed, error: %d\n", __func__, ret);
 #else
-	drm_err(&i915->drm,
+	drm_err(display->drm,
 		"Your hardware requires CONFIG_PMIC_OPREGION and it is not set\n");
 #endif
 
@@ -612,12 +608,12 @@ static const char *sequence_name(enum mipi_seq seq_id)
 static void intel_dsi_vbt_exec(struct intel_dsi *intel_dsi,
 			       enum mipi_seq seq_id)
 {
-	struct drm_i915_private *dev_priv = to_i915(intel_dsi->base.base.dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct intel_connector *connector = intel_dsi->attached_connector;
 	const u8 *data;
 	fn_mipi_elem_exec mipi_elem_exec;
 
-	if (drm_WARN_ON(&dev_priv->drm,
+	if (drm_WARN_ON(display->drm,
 			seq_id >= ARRAY_SIZE(connector->panel.vbt.dsi.sequence)))
 		return;
 
@@ -625,9 +621,9 @@ static void intel_dsi_vbt_exec(struct intel_dsi *intel_dsi,
 	if (!data)
 		return;
 
-	drm_WARN_ON(&dev_priv->drm, *data != seq_id);
+	drm_WARN_ON(display->drm, *data != seq_id);
 
-	drm_dbg_kms(&dev_priv->drm, "Starting MIPI sequence %d - %s\n",
+	drm_dbg_kms(display->drm, "Starting MIPI sequence %d - %s\n",
 		    seq_id, sequence_name(seq_id));
 
 	/* Skip Sequence Byte. */
@@ -657,19 +653,19 @@ static void intel_dsi_vbt_exec(struct intel_dsi *intel_dsi,
 
 			/* Consistency check if we have size. */
 			if (operation_size && data != next) {
-				drm_err(&dev_priv->drm,
+				drm_err(display->drm,
 					"Inconsistent operation size\n");
 				return;
 			}
 		} else if (operation_size) {
 			/* We have size, skip. */
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "Unsupported MIPI operation byte %u\n",
 				    operation_byte);
 			data += operation_size;
 		} else {
 			/* No size, can't skip without parsing. */
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"Unsupported MIPI operation byte %u\n",
 				operation_byte);
 			return;
@@ -695,54 +691,44 @@ void intel_dsi_vbt_exec_sequence(struct intel_dsi *intel_dsi,
 
 void intel_dsi_log_params(struct intel_dsi *intel_dsi)
 {
-	struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev);
-
-	drm_dbg_kms(&i915->drm, "Pclk %d\n", intel_dsi->pclk);
-	drm_dbg_kms(&i915->drm, "Pixel overlap %d\n",
-		    intel_dsi->pixel_overlap);
-	drm_dbg_kms(&i915->drm, "Lane count %d\n", intel_dsi->lane_count);
-	drm_dbg_kms(&i915->drm, "DPHY param reg 0x%x\n", intel_dsi->dphy_reg);
-	drm_dbg_kms(&i915->drm, "Video mode format %s\n",
-		    intel_dsi->video_mode == NON_BURST_SYNC_PULSE ?
-		    "non-burst with sync pulse" :
-		    intel_dsi->video_mode == NON_BURST_SYNC_EVENTS ?
-		    "non-burst with sync events" :
-		    intel_dsi->video_mode == BURST_MODE ?
-		    "burst" : "<unknown>");
-	drm_dbg_kms(&i915->drm, "Burst mode ratio %d\n",
-		    intel_dsi->burst_mode_ratio);
-	drm_dbg_kms(&i915->drm, "Reset timer %d\n", intel_dsi->rst_timer_val);
-	drm_dbg_kms(&i915->drm, "Eot %s\n",
-		    str_enabled_disabled(intel_dsi->eotp_pkt));
-	drm_dbg_kms(&i915->drm, "Clockstop %s\n",
-		    str_enabled_disabled(!intel_dsi->clock_stop));
-	drm_dbg_kms(&i915->drm, "Mode %s\n",
-		    intel_dsi->operation_mode ? "command" : "video");
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
+	struct drm_printer p = drm_dbg_printer(display->drm, DRM_UT_KMS,
+					       "DSI parameters:");
+
+	drm_printf(&p, "Pclk %d\n", intel_dsi->pclk);
+	drm_printf(&p, "Pixel overlap %d\n", intel_dsi->pixel_overlap);
+	drm_printf(&p, "Lane count %d\n", intel_dsi->lane_count);
+	drm_printf(&p, "DPHY param reg 0x%x\n", intel_dsi->dphy_reg);
+	drm_printf(&p, "Video mode format %s\n",
+		   intel_dsi->video_mode == NON_BURST_SYNC_PULSE ?
+		   "non-burst with sync pulse" :
+		   intel_dsi->video_mode == NON_BURST_SYNC_EVENTS ?
+		   "non-burst with sync events" :
+		   intel_dsi->video_mode == BURST_MODE ?
+		   "burst" : "<unknown>");
+	drm_printf(&p, "Burst mode ratio %d\n", intel_dsi->burst_mode_ratio);
+	drm_printf(&p, "Reset timer %d\n", intel_dsi->rst_timer_val);
+	drm_printf(&p, "Eot %s\n", str_enabled_disabled(intel_dsi->eotp_pkt));
+	drm_printf(&p, "Clockstop %s\n", str_enabled_disabled(!intel_dsi->clock_stop));
+	drm_printf(&p, "Mode %s\n", intel_dsi->operation_mode ? "command" : "video");
 	if (intel_dsi->dual_link == DSI_DUAL_LINK_FRONT_BACK)
-		drm_dbg_kms(&i915->drm,
-			    "Dual link: DSI_DUAL_LINK_FRONT_BACK\n");
+		drm_printf(&p, "Dual link: DSI_DUAL_LINK_FRONT_BACK\n");
 	else if (intel_dsi->dual_link == DSI_DUAL_LINK_PIXEL_ALT)
-		drm_dbg_kms(&i915->drm,
-			    "Dual link: DSI_DUAL_LINK_PIXEL_ALT\n");
+		drm_printf(&p, "Dual link: DSI_DUAL_LINK_PIXEL_ALT\n");
 	else
-		drm_dbg_kms(&i915->drm, "Dual link: NONE\n");
-	drm_dbg_kms(&i915->drm, "Pixel Format %d\n", intel_dsi->pixel_format);
-	drm_dbg_kms(&i915->drm, "TLPX %d\n", intel_dsi->escape_clk_div);
-	drm_dbg_kms(&i915->drm, "LP RX Timeout 0x%x\n",
-		    intel_dsi->lp_rx_timeout);
-	drm_dbg_kms(&i915->drm, "Turnaround Timeout 0x%x\n",
-		    intel_dsi->turn_arnd_val);
-	drm_dbg_kms(&i915->drm, "Init Count 0x%x\n", intel_dsi->init_count);
-	drm_dbg_kms(&i915->drm, "HS to LP Count 0x%x\n",
-		    intel_dsi->hs_to_lp_count);
-	drm_dbg_kms(&i915->drm, "LP Byte Clock %d\n", intel_dsi->lp_byte_clk);
-	drm_dbg_kms(&i915->drm, "DBI BW Timer 0x%x\n", intel_dsi->bw_timer);
-	drm_dbg_kms(&i915->drm, "LP to HS Clock Count 0x%x\n",
-		    intel_dsi->clk_lp_to_hs_count);
-	drm_dbg_kms(&i915->drm, "HS to LP Clock Count 0x%x\n",
-		    intel_dsi->clk_hs_to_lp_count);
-	drm_dbg_kms(&i915->drm, "BTA %s\n",
-		    str_enabled_disabled(!(intel_dsi->video_frmt_cfg_bits & DISABLE_VIDEO_BTA)));
+		drm_printf(&p, "Dual link: NONE\n");
+	drm_printf(&p, "Pixel Format %d\n", intel_dsi->pixel_format);
+	drm_printf(&p, "TLPX %d\n", intel_dsi->escape_clk_div);
+	drm_printf(&p, "LP RX Timeout 0x%x\n", intel_dsi->lp_rx_timeout);
+	drm_printf(&p, "Turnaround Timeout 0x%x\n", intel_dsi->turn_arnd_val);
+	drm_printf(&p, "Init Count 0x%x\n", intel_dsi->init_count);
+	drm_printf(&p, "HS to LP Count 0x%x\n", intel_dsi->hs_to_lp_count);
+	drm_printf(&p, "LP Byte Clock %d\n", intel_dsi->lp_byte_clk);
+	drm_printf(&p, "DBI BW Timer 0x%x\n", intel_dsi->bw_timer);
+	drm_printf(&p, "LP to HS Clock Count 0x%x\n", intel_dsi->clk_lp_to_hs_count);
+	drm_printf(&p, "HS to LP Clock Count 0x%x\n", intel_dsi->clk_hs_to_lp_count);
+	drm_printf(&p, "BTA %s\n",
+		   str_enabled_disabled(!(intel_dsi->video_frmt_cfg_bits & DISABLE_VIDEO_BTA)));
 }
 
 static enum mipi_dsi_pixel_format vbt_to_dsi_pixel_format(unsigned int format)
@@ -764,8 +750,7 @@ static enum mipi_dsi_pixel_format vbt_to_dsi_pixel_format(unsigned int format)
 
 bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id)
 {
-	struct drm_device *dev = intel_dsi->base.base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct intel_connector *connector = intel_dsi->attached_connector;
 	struct mipi_config *mipi_config = connector->panel.vbt.dsi.config;
 	struct mipi_pps_data *pps = connector->panel.vbt.dsi.pps;
@@ -773,7 +758,7 @@ bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	u16 burst_mode_ratio;
 	enum port port;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	intel_dsi->eotp_pkt = mipi_config->eot_pkt_disabled ? 0 : 1;
 	intel_dsi->clock_stop = mipi_config->enable_clk_stop ? 1 : 0;
@@ -819,7 +804,7 @@ bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id)
 		u32 bitrate;
 
 		if (mipi_config->target_burst_mode_freq == 0) {
-			drm_err(&dev_priv->drm, "Burst mode target is not set\n");
+			drm_err(display->drm, "Burst mode target is not set\n");
 			return false;
 		}
 
@@ -836,7 +821,7 @@ bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id)
 			mipi_config->target_burst_mode_freq = bitrate;
 
 		if (mipi_config->target_burst_mode_freq < bitrate) {
-			drm_err(&dev_priv->drm, "Burst mode freq is less than computed\n");
+			drm_err(display->drm, "Burst mode freq is less than computed\n");
 			return false;
 		}
 
@@ -900,8 +885,7 @@ static const struct pinctrl_map soc_pwm_pinctrl_map[] = {
 
 void intel_dsi_vbt_gpio_init(struct intel_dsi *intel_dsi, bool panel_is_on)
 {
-	struct drm_device *dev = intel_dsi->base.base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_display *display = to_intel_display(&intel_dsi->base);
 	struct intel_connector *connector = intel_dsi->attached_connector;
 	struct mipi_config *mipi_config = connector->panel.vbt.dsi.config;
 	enum gpiod_flags flags = panel_is_on ? GPIOD_OUT_HIGH : GPIOD_OUT_LOW;
@@ -911,13 +895,13 @@ void intel_dsi_vbt_gpio_init(struct intel_dsi *intel_dsi, bool panel_is_on)
 	struct pinctrl *pinctrl;
 	int ret;
 
-	if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+	if ((display->platform.valleyview || display->platform.cherryview) &&
 	    mipi_config->pwm_blc == PPS_BLC_PMIC) {
 		gpiod_lookup_table = &pmic_panel_gpio_table;
 		want_panel_gpio = true;
 	}
 
-	if (IS_VALLEYVIEW(dev_priv) && mipi_config->pwm_blc == PPS_BLC_SOC) {
+	if (display->platform.valleyview && mipi_config->pwm_blc == PPS_BLC_SOC) {
 		gpiod_lookup_table = &soc_panel_gpio_table;
 		want_panel_gpio = true;
 		want_backlight_gpio = true;
@@ -926,12 +910,12 @@ void intel_dsi_vbt_gpio_init(struct intel_dsi *intel_dsi, bool panel_is_on)
 		ret = pinctrl_register_mappings(soc_pwm_pinctrl_map,
 					     ARRAY_SIZE(soc_pwm_pinctrl_map));
 		if (ret)
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"Failed to register pwm0 pinmux mapping\n");
 
-		pinctrl = devm_pinctrl_get_select(dev->dev, "soc_pwm0");
+		pinctrl = devm_pinctrl_get_select(display->drm->dev, "soc_pwm0");
 		if (IS_ERR(pinctrl))
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"Failed to set pinmux to PWM\n");
 	}
 
@@ -939,9 +923,9 @@ void intel_dsi_vbt_gpio_init(struct intel_dsi *intel_dsi, bool panel_is_on)
 		gpiod_add_lookup_table(gpiod_lookup_table);
 
 	if (want_panel_gpio) {
-		intel_dsi->gpio_panel = devm_gpiod_get(dev->dev, "panel", flags);
+		intel_dsi->gpio_panel = devm_gpiod_get(display->drm->dev, "panel", flags);
 		if (IS_ERR(intel_dsi->gpio_panel)) {
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"Failed to own gpio for panel control\n");
 			intel_dsi->gpio_panel = NULL;
 		}
@@ -949,9 +933,9 @@ void intel_dsi_vbt_gpio_init(struct intel_dsi *intel_dsi, bool panel_is_on)
 
 	if (want_backlight_gpio) {
 		intel_dsi->gpio_backlight =
-			devm_gpiod_get(dev->dev, "backlight", flags);
+			devm_gpiod_get(display->drm->dev, "backlight", flags);
 		if (IS_ERR(intel_dsi->gpio_backlight)) {
-			drm_err(&dev_priv->drm,
+			drm_err(display->drm,
 				"Failed to own gpio for backlight control\n");
 			intel_dsi->gpio_backlight = NULL;
 		}
diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c
index c16fb34b737d..b61520353c92 100644
--- a/drivers/gpu/drm/i915/display/intel_dvo.c
+++ b/drivers/gpu/drm/i915/display/intel_dvo.c
@@ -31,10 +31,11 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_connector.h"
 #include "intel_de.h"
 #include "intel_display_driver.h"
@@ -129,13 +130,13 @@ static struct intel_dvo *intel_attached_dvo(struct intel_connector *connector)
 
 static bool intel_dvo_connector_get_hw_state(struct intel_connector *connector)
 {
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
+	struct intel_display *display = to_intel_display(connector);
 	struct intel_encoder *encoder = intel_attached_encoder(connector);
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
 	enum port port = encoder->port;
 	u32 tmp;
 
-	tmp = intel_de_read(i915, DVO(port));
+	tmp = intel_de_read(display, DVO(port));
 
 	if (!(tmp & DVO_ENABLE))
 		return false;
@@ -146,11 +147,11 @@ static bool intel_dvo_connector_get_hw_state(struct intel_connector *connector)
 static bool intel_dvo_get_hw_state(struct intel_encoder *encoder,
 				   enum pipe *pipe)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 	u32 tmp;
 
-	tmp = intel_de_read(i915, DVO(port));
+	tmp = intel_de_read(display, DVO(port));
 
 	*pipe = REG_FIELD_GET(DVO_PIPE_SEL_MASK, tmp);
 
@@ -160,13 +161,13 @@ static bool intel_dvo_get_hw_state(struct intel_encoder *encoder,
 static void intel_dvo_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *pipe_config)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum port port = encoder->port;
 	u32 tmp, flags = 0;
 
 	pipe_config->output_types |= BIT(INTEL_OUTPUT_DVO);
 
-	tmp = intel_de_read(i915, DVO(port));
+	tmp = intel_de_read(display, DVO(port));
 	if (tmp & DVO_HSYNC_ACTIVE_HIGH)
 		flags |= DRM_MODE_FLAG_PHSYNC;
 	else
@@ -186,14 +187,14 @@ static void intel_disable_dvo(struct intel_atomic_state *state,
 			      const struct intel_crtc_state *old_crtc_state,
 			      const struct drm_connector_state *old_conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
 	enum port port = encoder->port;
 
 	intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, false);
 
-	intel_de_rmw(i915, DVO(port), DVO_ENABLE, 0);
-	intel_de_posting_read(i915, DVO(port));
+	intel_de_rmw(display, DVO(port), DVO_ENABLE, 0);
+	intel_de_posting_read(display, DVO(port));
 }
 
 static void intel_enable_dvo(struct intel_atomic_state *state,
@@ -201,7 +202,7 @@ static void intel_enable_dvo(struct intel_atomic_state *state,
 			     const struct intel_crtc_state *pipe_config,
 			     const struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
 	enum port port = encoder->port;
 
@@ -209,8 +210,8 @@ static void intel_enable_dvo(struct intel_atomic_state *state,
 					 &pipe_config->hw.mode,
 					 &pipe_config->hw.adjusted_mode);
 
-	intel_de_rmw(i915, DVO(port), 0, DVO_ENABLE);
-	intel_de_posting_read(i915, DVO(port));
+	intel_de_rmw(display, DVO(port), 0, DVO_ENABLE);
+	intel_de_posting_read(display, DVO(port));
 
 	intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, true);
 }
@@ -288,7 +289,7 @@ static void intel_dvo_pre_enable(struct intel_atomic_state *state,
 				 const struct intel_crtc_state *pipe_config,
 				 const struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
 	const struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
 	enum port port = encoder->port;
@@ -296,7 +297,7 @@ static void intel_dvo_pre_enable(struct intel_atomic_state *state,
 	u32 dvo_val;
 
 	/* Save the active data order, since I don't know what it should be set to. */
-	dvo_val = intel_de_read(i915, DVO(port)) &
+	dvo_val = intel_de_read(display, DVO(port)) &
 		  (DVO_DEDICATED_INT_ENABLE |
 		   DVO_PRESERVE_MASK | DVO_ACT_DATA_ORDER_MASK);
 	dvo_val |= DVO_DATA_ORDER_FP | DVO_BORDER_ENABLE |
@@ -309,10 +310,10 @@ static void intel_dvo_pre_enable(struct intel_atomic_state *state,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
 		dvo_val |= DVO_VSYNC_ACTIVE_HIGH;
 
-	intel_de_write(i915, DVO_SRCDIM(port),
+	intel_de_write(display, DVO_SRCDIM(port),
 		       DVO_SRCDIM_HORIZONTAL(adjusted_mode->crtc_hdisplay) |
 		       DVO_SRCDIM_VERTICAL(adjusted_mode->crtc_vdisplay));
-	intel_de_write(i915, DVO(port), dvo_val);
+	intel_de_write(display, DVO(port), dvo_val);
 }
 
 static enum drm_connector_status
@@ -320,10 +321,9 @@ intel_dvo_detect(struct drm_connector *_connector, bool force)
 {
 	struct intel_display *display = to_intel_display(_connector->dev);
 	struct intel_connector *connector = to_intel_connector(_connector);
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
 
-	drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s]\n",
+	drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s]\n",
 		    connector->base.base.id, connector->base.name);
 
 	if (!intel_display_device_enabled(display))
@@ -414,11 +414,10 @@ static int intel_dvo_connector_type(const struct intel_dvo_device *dvo)
 	}
 }
 
-static bool intel_dvo_init_dev(struct drm_i915_private *dev_priv,
+static bool intel_dvo_init_dev(struct intel_display *display,
 			       struct intel_dvo *intel_dvo,
 			       const struct intel_dvo_device *dvo)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct i2c_adapter *i2c;
 	u32 dpll[I915_MAX_PIPES];
 	enum pipe pipe;
@@ -458,15 +457,15 @@ static bool intel_dvo_init_dev(struct drm_i915_private *dev_priv,
 	 * the clock enabled before we attempt to initialize
 	 * the device.
 	 */
-	for_each_pipe(dev_priv, pipe)
-		dpll[pipe] = intel_de_rmw(dev_priv, DPLL(dev_priv, pipe), 0,
+	for_each_pipe(display, pipe)
+		dpll[pipe] = intel_de_rmw(display, DPLL(display, pipe), 0,
 					  DPLL_DVO_2X_MODE);
 
 	ret = dvo->dev_ops->init(&intel_dvo->dev, i2c);
 
 	/* restore the DVO 2x clock state to original */
-	for_each_pipe(dev_priv, pipe) {
-		intel_de_write(dev_priv, DPLL(dev_priv, pipe), dpll[pipe]);
+	for_each_pipe(display, pipe) {
+		intel_de_write(display, DPLL(display, pipe), dpll[pipe]);
 	}
 
 	intel_gmbus_force_bit(i2c, false);
@@ -474,14 +473,14 @@ static bool intel_dvo_init_dev(struct drm_i915_private *dev_priv,
 	return ret;
 }
 
-static bool intel_dvo_probe(struct drm_i915_private *i915,
+static bool intel_dvo_probe(struct intel_display *display,
 			    struct intel_dvo *intel_dvo)
 {
 	int i;
 
 	/* Now, try to find a controller */
 	for (i = 0; i < ARRAY_SIZE(intel_dvo_devices); i++) {
-		if (intel_dvo_init_dev(i915, intel_dvo,
+		if (intel_dvo_init_dev(display, intel_dvo,
 				       &intel_dvo_devices[i]))
 			return true;
 	}
@@ -489,9 +488,8 @@ static bool intel_dvo_probe(struct drm_i915_private *i915,
 	return false;
 }
 
-void intel_dvo_init(struct drm_i915_private *i915)
+void intel_dvo_init(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_connector *connector;
 	struct intel_encoder *encoder;
 	struct intel_dvo *intel_dvo;
@@ -518,7 +516,7 @@ void intel_dvo_init(struct drm_i915_private *i915)
 	encoder->pre_enable = intel_dvo_pre_enable;
 	connector->get_hw_state = intel_dvo_connector_get_hw_state;
 
-	if (!intel_dvo_probe(i915, intel_dvo)) {
+	if (!intel_dvo_probe(display, intel_dvo)) {
 		kfree(intel_dvo);
 		intel_connector_free(connector);
 		return;
@@ -535,12 +533,12 @@ void intel_dvo_init(struct drm_i915_private *i915)
 		encoder->cloneable = BIT(INTEL_OUTPUT_ANALOG) |
 			BIT(INTEL_OUTPUT_DVO);
 
-	drm_encoder_init(&i915->drm, &encoder->base,
+	drm_encoder_init(display->drm, &encoder->base,
 			 &intel_dvo_enc_funcs,
 			 intel_dvo_encoder_type(&intel_dvo->dev),
 			 "DVO %c", port_name(encoder->port));
 
-	drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] detected %s\n",
+	drm_dbg_kms(display->drm, "[ENCODER:%d:%s] detected %s\n",
 		    encoder->base.base.id, encoder->base.name,
 		    intel_dvo->dev.name);
 
@@ -549,7 +547,7 @@ void intel_dvo_init(struct drm_i915_private *i915)
 			DRM_CONNECTOR_POLL_DISCONNECT;
 	connector->base.polled = connector->polled;
 
-	drm_connector_init_with_ddc(&i915->drm, &connector->base,
+	drm_connector_init_with_ddc(display->drm, &connector->base,
 				    &intel_dvo_connector_funcs,
 				    intel_dvo_connector_type(&intel_dvo->dev),
 				    intel_gmbus_get_adapter(display, GMBUS_PIN_DPC));
diff --git a/drivers/gpu/drm/i915/display/intel_dvo.h b/drivers/gpu/drm/i915/display/intel_dvo.h
index bf7a356422ab..83776552fc87 100644
--- a/drivers/gpu/drm/i915/display/intel_dvo.h
+++ b/drivers/gpu/drm/i915/display/intel_dvo.h
@@ -6,12 +6,12 @@
 #ifndef __INTEL_DVO_H__
 #define __INTEL_DVO_H__
 
-struct drm_i915_private;
+struct intel_display;
 
 #ifdef I915
-void intel_dvo_init(struct drm_i915_private *dev_priv);
+void intel_dvo_init(struct intel_display *display);
 #else
-static inline void intel_dvo_init(struct drm_i915_private *dev_priv)
+static inline void intel_dvo_init(struct intel_display *display)
 {
 }
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 2b0e0f220442..05393bd60c98 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -560,11 +560,11 @@ static bool plane_has_modifier(struct intel_display *display,
 		return false;
 
 	if (md->modifier == I915_FORMAT_MOD_4_TILED_BMG_CCS &&
-	    (GRAPHICS_VER(i915) < 20 || !IS_DGFX(i915)))
+	    (GRAPHICS_VER(i915) < 20 || !display->platform.dgfx))
 		return false;
 
 	if (md->modifier == I915_FORMAT_MOD_4_TILED_LNL_CCS &&
-	    (GRAPHICS_VER(i915) < 20 || IS_DGFX(i915)))
+	    (GRAPHICS_VER(i915) < 20 || display->platform.dgfx))
 		return false;
 
 	return true;
diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index 30ac9b089ad6..c648ab8a93d7 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -12,6 +12,7 @@
 
 #include "i915_drv.h"
 #include "intel_atomic_plane.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_dpt.h"
 #include "intel_fb.h"
@@ -117,7 +118,7 @@ intel_fb_pin_to_ggtt(const struct drm_framebuffer *fb,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_gem_object *_obj = intel_fb_bo(fb);
 	struct drm_i915_gem_object *obj = to_intel_bo(_obj);
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	struct i915_gem_ww_ctx ww;
 	struct i915_vma *vma;
 	unsigned int pinctl;
@@ -136,7 +137,7 @@ intel_fb_pin_to_ggtt(const struct drm_framebuffer *fb,
 	 * intel_runtime_pm_put(), so it is correct to wrap only the
 	 * pin/unpin/fence and not more.
 	 */
-	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
 	atomic_inc(&display->restore.pending_fb_pin);
 
@@ -215,7 +216,7 @@ err:
 		vma = ERR_PTR(ret);
 
 	atomic_dec(&display->restore.pending_fb_pin);
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 	return vma;
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index b6978135e8ad..bed2bba20b55 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -55,6 +55,7 @@
 #include "intel_cdclk.h"
 #include "intel_de.h"
 #include "intel_display_device.h"
+#include "intel_display_rpm.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
 #include "intel_display_wa.h"
@@ -251,9 +252,12 @@ static u16 intel_fbc_override_cfb_stride(const struct intel_plane_state *plane_s
 	 * Gen9 hw miscalculates cfb stride for linear as
 	 * PLANE_STRIDE*512 instead of PLANE_STRIDE*64, so
 	 * we always need to use the override there.
+	 *
+	 * wa_14022269668 For bmg, always program the FBC_STRIDE before fbc enable
 	 */
 	if (stride != stride_aligned ||
-	    (DISPLAY_VER(display) == 9 && fb->modifier == DRM_FORMAT_MOD_LINEAR))
+	    (DISPLAY_VER(display) == 9 && fb->modifier == DRM_FORMAT_MOD_LINEAR) ||
+	    display->platform.battlemage)
 		return stride_aligned * 4 / 64;
 
 	return 0;
@@ -519,6 +523,20 @@ static void ilk_fbc_activate(struct intel_fbc *fbc)
 		       DPFC_CTL_EN | g4x_dpfc_ctl(fbc));
 }
 
+static void fbc_compressor_clkgate_disable_wa(struct intel_fbc *fbc,
+					      bool disable)
+{
+	struct intel_display *display = fbc->display;
+
+	if (display->platform.dg2)
+		intel_de_rmw(display, GEN9_CLKGATE_DIS_4, DG2_DPFC_GATING_DIS,
+			     disable ? DG2_DPFC_GATING_DIS : 0);
+	else if (DISPLAY_VER(display) >= 14)
+		intel_de_rmw(display, MTL_PIPE_CLKGATE_DIS2(fbc->id),
+			     MTL_DPFC_GATING_DIS,
+			     disable ? MTL_DPFC_GATING_DIS : 0);
+}
+
 static void ilk_fbc_deactivate(struct intel_fbc *fbc)
 {
 	struct intel_display *display = fbc->display;
@@ -532,6 +550,10 @@ static void ilk_fbc_deactivate(struct intel_fbc *fbc)
 	if (dpfc_ctl & DPFC_CTL_EN) {
 		dpfc_ctl &= ~DPFC_CTL_EN;
 		intel_de_write(display, ILK_DPFC_CONTROL(fbc->id), dpfc_ctl);
+
+		/* wa_18038517565 Enable DPFC clock gating after FBC disable */
+		if (display->platform.dg2 || DISPLAY_VER(display) >= 14)
+			fbc_compressor_clkgate_disable_wa(fbc, false);
 	}
 }
 
@@ -921,6 +943,10 @@ static void intel_fbc_program_workarounds(struct intel_fbc *fbc)
 	if (DISPLAY_VER(display) >= 11 && !display->platform.dg2)
 		intel_de_rmw(display, ILK_DPFC_CHICKEN(fbc->id),
 			     0, DPFC_CHICKEN_FORCE_SLB_INVALIDATION);
+
+	/* wa_18038517565 Disable DPFC clock gating before FBC enable */
+	if (display->platform.dg2 || DISPLAY_VER(display) >= 14)
+		fbc_compressor_clkgate_disable_wa(fbc, true);
 }
 
 static void __intel_fbc_cleanup_cfb(struct intel_fbc *fbc)
@@ -1436,7 +1462,7 @@ static int intel_fbc_check_plane(struct intel_atomic_state *state,
 		return 0;
 	}
 
-	if (intel_display_needs_wa_16023588340(i915)) {
+	if (intel_display_needs_wa_16023588340(display)) {
 		plane_state->no_fbc_reason = "Wa_16023588340";
 		return 0;
 	}
@@ -1464,14 +1490,15 @@ static int intel_fbc_check_plane(struct intel_atomic_state *state,
 	 * Recommendation is to keep this combination disabled
 	 * Bspec: 50422 HSD: 14010260002
 	 *
-	 * In Xe3, PSR2 selective fetch and FBC dirty rect feature cannot
-	 * coexist. So if PSR2 selective fetch is supported then mark that
-	 * FBC is not supported.
-	 * TODO: Need a logic to decide between PSR2 and FBC Dirty rect
+	 * TODO: Implement a logic to select between PSR2 selective fetch and
+	 * FBC based on Bspec: 68881 in xe2lpd onwards.
+	 *
+	 * As we still see some strange underruns in those platforms while
+	 * disabling PSR2, keep FBC disabled in case of selective update is on
+	 * until the selection logic is implemented.
 	 */
-	if ((IS_DISPLAY_VER(display, 12, 14) || HAS_FBC_DIRTY_RECT(display)) &&
-	    crtc_state->has_sel_update && !crtc_state->has_panel_replay) {
-		plane_state->no_fbc_reason = "PSR2 enabled";
+	if (DISPLAY_VER(display) >= 12 && crtc_state->has_sel_update) {
+		plane_state->no_fbc_reason = "Selective update enabled";
 		return 0;
 	}
 
@@ -2120,13 +2147,12 @@ static int intel_fbc_debugfs_status_show(struct seq_file *m, void *unused)
 {
 	struct intel_fbc *fbc = m->private;
 	struct intel_display *display = fbc->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_plane *plane;
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 
 	drm_modeset_lock_all(display->drm);
 
-	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 	mutex_lock(&fbc->lock);
 
 	if (fbc->active) {
@@ -2151,7 +2177,7 @@ static int intel_fbc_debugfs_status_show(struct seq_file *m, void *unused)
 	}
 
 	mutex_unlock(&fbc->lock);
-	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	drm_modeset_unlock_all(display->drm);
 
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index adc19d5607de..2dc4029d71ed 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -47,9 +47,10 @@
 #include <drm/drm_managed.h>
 #include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "i915_vma.h"
 #include "intel_bo.h"
+#include "intel_display_core.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_fb.h"
 #include "intel_fb_pin.h"
@@ -65,9 +66,9 @@ struct intel_fbdev {
 
 static struct intel_fbdev *to_intel_fbdev(struct drm_fb_helper *fb_helper)
 {
-	struct drm_i915_private *i915 = to_i915(fb_helper->client.dev);
+	struct intel_display *display = to_intel_display(fb_helper->client.dev);
 
-	return i915->display.fbdev.fbdev;
+	return display->fbdev.fbdev;
 }
 
 static struct intel_frontbuffer *to_frontbuffer(struct intel_fbdev *ifbdev)
@@ -209,11 +210,10 @@ static const struct drm_fb_helper_funcs intel_fb_helper_funcs = {
 int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 				   struct drm_fb_helper_surface_size *sizes)
 {
+	struct intel_display *display = to_intel_display(helper->dev);
 	struct intel_fbdev *ifbdev = to_intel_fbdev(helper);
 	struct intel_framebuffer *fb = ifbdev->fb;
-	struct drm_device *dev = helper->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	struct fb_info *info;
 	struct i915_vma *vma;
 	unsigned long flags = 0;
@@ -226,7 +226,7 @@ int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 	if (fb &&
 	    (sizes->fb_width > fb->base.width ||
 	     sizes->fb_height > fb->base.height)) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "BIOS fb too small (%dx%d), we require (%dx%d),"
 			    " releasing it\n",
 			    fb->base.width, fb->base.height,
@@ -234,20 +234,20 @@ int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 		drm_framebuffer_put(&fb->base);
 		fb = NULL;
 	}
-	if (!fb || drm_WARN_ON(dev, !intel_fb_bo(&fb->base))) {
-		drm_dbg_kms(&dev_priv->drm,
+	if (!fb || drm_WARN_ON(display->drm, !intel_fb_bo(&fb->base))) {
+		drm_dbg_kms(display->drm,
 			    "no BIOS fb, allocating a new one\n");
 		fb = intel_fbdev_fb_alloc(helper, sizes);
 		if (IS_ERR(fb))
 			return PTR_ERR(fb);
 	} else {
-		drm_dbg_kms(&dev_priv->drm, "re-using BIOS fb\n");
+		drm_dbg_kms(display->drm, "re-using BIOS fb\n");
 		prealloc = true;
 		sizes->fb_width = fb->base.width;
 		sizes->fb_height = fb->base.height;
 	}
 
-	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
 	/* Pin the GGTT vma for our access via info->screen_base.
 	 * This also validates that any existing fb inherited from the
@@ -265,7 +265,7 @@ int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 
 	info = drm_fb_helper_alloc_info(helper);
 	if (IS_ERR(info)) {
-		drm_err(&dev_priv->drm, "Failed to allocate fb_info (%pe)\n", info);
+		drm_err(display->drm, "Failed to allocate fb_info (%pe)\n", info);
 		ret = PTR_ERR(info);
 		goto out_unpin;
 	}
@@ -277,11 +277,11 @@ int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 
 	obj = intel_fb_bo(&fb->base);
 
-	ret = intel_fbdev_fb_fill_info(dev_priv, info, obj, vma);
+	ret = intel_fbdev_fb_fill_info(display, info, obj, vma);
 	if (ret)
 		goto out_unpin;
 
-	drm_fb_helper_fill_info(info, dev->fb_helper, sizes);
+	drm_fb_helper_fill_info(info, display->drm->fb_helper, sizes);
 
 	/* If the object is shmemfs backed, it will have given us zeroed pages.
 	 * If the object is stolen however, it will be full of whatever
@@ -292,21 +292,22 @@ int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 
 	/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */
 
-	drm_dbg_kms(&dev_priv->drm, "allocated %dx%d fb: 0x%08x\n",
+	drm_dbg_kms(display->drm, "allocated %dx%d fb: 0x%08x\n",
 		    fb->base.width, fb->base.height,
 		    i915_ggtt_offset(vma));
 	ifbdev->fb = fb;
 	ifbdev->vma = vma;
 	ifbdev->vma_flags = flags;
 
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	return 0;
 
 out_unpin:
 	intel_fb_unpin_vma(vma, flags);
 out_unlock:
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
+
 	return ret;
 }
 
@@ -319,16 +320,15 @@ out_unlock:
  * Note we only support a single fb shared across pipes for boot (mostly for
  * fbcon), so we just find the biggest and use that.
  */
-static bool intel_fbdev_init_bios(struct drm_device *dev,
+static bool intel_fbdev_init_bios(struct intel_display *display,
 				  struct intel_fbdev *ifbdev)
 {
-	struct drm_i915_private *i915 = to_i915(dev);
 	struct intel_framebuffer *fb = NULL;
 	struct intel_crtc *crtc;
 	unsigned int max_size = 0;
 
 	/* Find the largest fb */
-	for_each_intel_crtc(dev, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct intel_plane *plane =
@@ -338,21 +338,21 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		struct drm_gem_object *obj = intel_fb_bo(plane_state->uapi.fb);
 
 		if (!crtc_state->uapi.active) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[CRTC:%d:%s] not active, skipping\n",
 				    crtc->base.base.id, crtc->base.name);
 			continue;
 		}
 
 		if (!obj) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s] no fb, skipping\n",
 				    plane->base.base.id, plane->base.name);
 			continue;
 		}
 
 		if (obj->size > max_size) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "found possible fb from [PLANE:%d:%s]\n",
 				    plane->base.base.id, plane->base.name);
 			fb = to_intel_framebuffer(plane_state->uapi.fb);
@@ -361,13 +361,13 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	}
 
 	if (!fb) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "no active fbs found, not using BIOS config\n");
 		goto out;
 	}
 
 	/* Now make sure all the pipes will fit into it */
-	for_each_intel_crtc(dev, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct intel_plane *plane =
@@ -375,13 +375,13 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		unsigned int cur_size;
 
 		if (!crtc_state->uapi.active) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[CRTC:%d:%s] not active, skipping\n",
 				    crtc->base.base.id, crtc->base.name);
 			continue;
 		}
 
-		drm_dbg_kms(&i915->drm, "checking [PLANE:%d:%s] for BIOS fb\n",
+		drm_dbg_kms(display->drm, "checking [PLANE:%d:%s] for BIOS fb\n",
 			    plane->base.base.id, plane->base.name);
 
 		/*
@@ -392,7 +392,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		cur_size = crtc_state->uapi.adjusted_mode.crtc_hdisplay;
 		cur_size = cur_size * fb->base.format->cpp[0];
 		if (fb->base.pitches[0] < cur_size) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "fb not wide enough for [PLANE:%d:%s] (%d vs %d)\n",
 				    plane->base.base.id, plane->base.name,
 				    cur_size, fb->base.pitches[0]);
@@ -403,7 +403,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		cur_size = crtc_state->uapi.adjusted_mode.crtc_vdisplay;
 		cur_size = intel_fb_align_height(&fb->base, 0, cur_size);
 		cur_size *= fb->base.pitches[0];
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[CRTC:%d:%s] area: %dx%d, bpp: %d, size: %d\n",
 			    crtc->base.base.id, crtc->base.name,
 			    crtc_state->uapi.adjusted_mode.crtc_hdisplay,
@@ -412,7 +412,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 			    cur_size);
 
 		if (cur_size > max_size) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "fb not big enough for [PLANE:%d:%s] (%d vs %d)\n",
 				    plane->base.base.id, plane->base.name,
 				    cur_size, max_size);
@@ -420,14 +420,14 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 			break;
 		}
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "fb big enough [PLANE:%d:%s] (%d >= %d)\n",
 			    plane->base.base.id, plane->base.name,
 			    max_size, cur_size);
 	}
 
 	if (!fb) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "BIOS fb not suitable for all pipes, not using\n");
 		goto out;
 	}
@@ -437,7 +437,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	drm_framebuffer_get(&ifbdev->fb->base);
 
 	/* Final pass to check if any active pipes don't have fbs */
-	for_each_intel_crtc(dev, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct intel_plane *plane =
@@ -448,13 +448,13 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		if (!crtc_state->uapi.active)
 			continue;
 
-		drm_WARN(dev, !plane_state->uapi.fb,
+		drm_WARN(display->drm, !plane_state->uapi.fb,
 			 "re-used BIOS config but lost an fb on [PLANE:%d:%s]\n",
 			 plane->base.base.id, plane->base.name);
 	}
 
 
-	drm_dbg_kms(&i915->drm, "using BIOS fb for initial console\n");
+	drm_dbg_kms(display->drm, "using BIOS fb for initial console\n");
 	return true;
 
 out:
@@ -479,26 +479,25 @@ static unsigned int intel_fbdev_color_mode(const struct drm_format_info *info)
 	}
 }
 
-void intel_fbdev_setup(struct drm_i915_private *i915)
+void intel_fbdev_setup(struct intel_display *display)
 {
-	struct drm_device *dev = &i915->drm;
 	struct intel_fbdev *ifbdev;
 	unsigned int preferred_bpp = 0;
 
-	if (!HAS_DISPLAY(i915))
+	if (!HAS_DISPLAY(display))
 		return;
 
-	ifbdev = drmm_kzalloc(dev, sizeof(*ifbdev), GFP_KERNEL);
+	ifbdev = drmm_kzalloc(display->drm, sizeof(*ifbdev), GFP_KERNEL);
 	if (!ifbdev)
 		return;
 
-	i915->display.fbdev.fbdev = ifbdev;
-	if (intel_fbdev_init_bios(dev, ifbdev))
+	display->fbdev.fbdev = ifbdev;
+	if (intel_fbdev_init_bios(display, ifbdev))
 		preferred_bpp = intel_fbdev_color_mode(ifbdev->fb->base.format);
 	if (!preferred_bpp)
 		preferred_bpp = 32;
 
-	drm_client_setup_with_color_mode(dev, preferred_bpp);
+	drm_client_setup_with_color_mode(display->drm, preferred_bpp);
 }
 
 struct intel_framebuffer *intel_fbdev_framebuffer(struct intel_fbdev *fbdev)
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.h b/drivers/gpu/drm/i915/display/intel_fbdev.h
index 89bad3a2b01a..a15e3e222a0c 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.h
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.h
@@ -10,7 +10,7 @@
 
 struct drm_fb_helper;
 struct drm_fb_helper_surface_size;
-struct drm_i915_private;
+struct intel_display;
 struct intel_fbdev;
 struct intel_framebuffer;
 
@@ -19,14 +19,14 @@ int intel_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper,
 				   struct drm_fb_helper_surface_size *sizes);
 #define INTEL_FBDEV_DRIVER_OPS \
 	.fbdev_probe = intel_fbdev_driver_fbdev_probe
-void intel_fbdev_setup(struct drm_i915_private *dev_priv);
+void intel_fbdev_setup(struct intel_display *display);
 struct intel_framebuffer *intel_fbdev_framebuffer(struct intel_fbdev *fbdev);
 struct i915_vma *intel_fbdev_vma_pointer(struct intel_fbdev *fbdev);
 
 #else
 #define INTEL_FBDEV_DRIVER_OPS \
 	.fbdev_probe = NULL
-static inline void intel_fbdev_setup(struct drm_i915_private *dev_priv)
+static inline void intel_fbdev_setup(struct intel_display *display)
 {
 }
 static inline struct intel_framebuffer *intel_fbdev_framebuffer(struct intel_fbdev *fbdev)
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev_fb.c b/drivers/gpu/drm/i915/display/intel_fbdev_fb.c
index 4991c35a2632..5f4cb3328265 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev_fb.c
@@ -15,9 +15,9 @@
 struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 					       struct drm_fb_helper_surface_size *sizes)
 {
+	struct intel_display *display = to_intel_display(helper->dev);
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct drm_framebuffer *fb;
-	struct drm_device *dev = helper->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_mode_fb_cmd2 mode_cmd = {};
 	struct drm_i915_gem_object *obj;
 	int size;
@@ -50,14 +50,14 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 		 *
 		 * Also skip stolen on MTL as Wa_22018444074 mitigation.
 		 */
-		if (!(IS_METEORLAKE(dev_priv)) && size * 2 < dev_priv->dsm.usable_size)
+		if (!display->platform.meteorlake && size * 2 < dev_priv->dsm.usable_size)
 			obj = i915_gem_object_create_stolen(dev_priv, size);
 		if (IS_ERR(obj))
 			obj = i915_gem_object_create_shmem(dev_priv, size);
 	}
 
 	if (IS_ERR(obj)) {
-		drm_err(&dev_priv->drm, "failed to allocate framebuffer (%pe)\n", obj);
+		drm_err(display->drm, "failed to allocate framebuffer (%pe)\n", obj);
 		return ERR_PTR(-ENOMEM);
 	}
 
@@ -67,9 +67,10 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 	return to_intel_framebuffer(fb);
 }
 
-int intel_fbdev_fb_fill_info(struct drm_i915_private *i915, struct fb_info *info,
+int intel_fbdev_fb_fill_info(struct intel_display *display, struct fb_info *info,
 			     struct drm_gem_object *_obj, struct i915_vma *vma)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct drm_i915_gem_object *obj = to_intel_bo(_obj);
 	struct i915_gem_ww_ctx ww;
 	void __iomem *vaddr;
@@ -101,7 +102,7 @@ int intel_fbdev_fb_fill_info(struct drm_i915_private *i915, struct fb_info *info
 
 		vaddr = i915_vma_pin_iomap(vma);
 		if (IS_ERR(vaddr)) {
-			drm_err(&i915->drm,
+			drm_err(display->drm,
 				"Failed to remap framebuffer into virtual memory (%pe)\n", vaddr);
 			ret = PTR_ERR(vaddr);
 			continue;
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev_fb.h b/drivers/gpu/drm/i915/display/intel_fbdev_fb.h
index e502ae375fc0..cb7957272715 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev_fb.h
+++ b/drivers/gpu/drm/i915/display/intel_fbdev_fb.h
@@ -9,13 +9,13 @@
 struct drm_fb_helper;
 struct drm_fb_helper_surface_size;
 struct drm_gem_object;
-struct drm_i915_private;
 struct fb_info;
 struct i915_vma;
+struct intel_display;
 
 struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 					       struct drm_fb_helper_surface_size *sizes);
-int intel_fbdev_fb_fill_info(struct drm_i915_private *i915, struct fb_info *info,
+int intel_fbdev_fb_fill_info(struct intel_display *display, struct fb_info *info,
 			     struct drm_gem_object *obj, struct i915_vma *vma);
 
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_fdi.c b/drivers/gpu/drm/i915/display/intel_fdi.c
index 40deee0769ae..169bbe154b5c 100644
--- a/drivers/gpu/drm/i915/display/intel_fdi.c
+++ b/drivers/gpu/drm/i915/display/intel_fdi.c
@@ -6,15 +6,16 @@
 #include <linux/string_helpers.h>
 
 #include <drm/drm_fixed.h>
+#include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_atomic.h"
 #include "intel_crtc.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
-#include "intel_dp.h"
 #include "intel_display_types.h"
+#include "intel_dp.h"
 #include "intel_fdi.h"
 #include "intel_fdi_regs.h"
 #include "intel_link_bw.h"
@@ -464,7 +465,6 @@ static void ivb_update_fdi_bc_bifurcation(const struct intel_crtc_state *crtc_st
 void intel_fdi_normal_train(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum pipe pipe = crtc->pipe;
 	i915_reg_t reg;
 	u32 temp;
@@ -483,7 +483,7 @@ void intel_fdi_normal_train(struct intel_crtc *crtc)
 
 	reg = FDI_RX_CTL(pipe);
 	temp = intel_de_read(display, reg);
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
 		temp |= FDI_LINK_TRAIN_NORMAL_CPT;
 	} else {
@@ -607,7 +607,6 @@ static void gen6_fdi_link_train(struct intel_crtc *crtc,
 				const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum pipe pipe = crtc->pipe;
 	i915_reg_t reg;
 	u32 temp, i, retry;
@@ -647,7 +646,7 @@ static void gen6_fdi_link_train(struct intel_crtc *crtc,
 
 	reg = FDI_RX_CTL(pipe);
 	temp = intel_de_read(display, reg);
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
 		temp |= FDI_LINK_TRAIN_PATTERN_1_CPT;
 	} else {
@@ -698,7 +697,7 @@ static void gen6_fdi_link_train(struct intel_crtc *crtc,
 
 	reg = FDI_RX_CTL(pipe);
 	temp = intel_de_read(display, reg);
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
 		temp |= FDI_LINK_TRAIN_PATTERN_2_CPT;
 	} else {
@@ -1077,7 +1076,6 @@ void ilk_fdi_pll_disable(struct intel_crtc *crtc)
 void ilk_fdi_disable(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	i915_reg_t reg;
 	u32 temp;
@@ -1096,7 +1094,7 @@ void ilk_fdi_disable(struct intel_crtc *crtc)
 	udelay(100);
 
 	/* Ironlake workaround, disable clock pointer after downing FDI */
-	if (HAS_PCH_IBX(dev_priv))
+	if (HAS_PCH_IBX(display))
 		intel_de_write(display, FDI_RX_CHICKEN(pipe),
 			       FDI_RX_PHASE_SYNC_POINTER_OVR);
 
@@ -1106,7 +1104,7 @@ void ilk_fdi_disable(struct intel_crtc *crtc)
 
 	reg = FDI_RX_CTL(pipe);
 	temp = intel_de_read(display, reg);
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		temp &= ~FDI_LINK_TRAIN_PATTERN_MASK_CPT;
 		temp |= FDI_LINK_TRAIN_PATTERN_1_CPT;
 	} else {
diff --git a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
index 7a8fbff39be0..2a787897b2d3 100644
--- a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
+++ b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
@@ -25,7 +25,8 @@
  *
  */
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
 #include "intel_de.h"
 #include "intel_display_irq.h"
@@ -57,11 +58,10 @@
 
 static bool ivb_can_enable_err_int(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc;
 	enum pipe pipe;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	for_each_pipe(display, pipe) {
 		crtc = intel_crtc_for_pipe(display, pipe);
@@ -75,11 +75,10 @@ static bool ivb_can_enable_err_int(struct intel_display *display)
 
 static bool cpt_can_enable_serr_int(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum pipe pipe;
 	struct intel_crtc *crtc;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	for_each_pipe(display, pipe) {
 		crtc = intel_crtc_for_pipe(display, pipe);
@@ -94,11 +93,10 @@ static bool cpt_can_enable_serr_int(struct intel_display *display)
 static void i9xx_check_fifo_underruns(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	i915_reg_t reg = PIPESTAT(display, crtc->pipe);
 	u32 enable_mask;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	if ((intel_de_read(display, reg) & PIPE_FIFO_UNDERRUN_STATUS) == 0)
 		return;
@@ -115,10 +113,9 @@ static void i9xx_set_fifo_underrun_reporting(struct intel_display *display,
 					     enum pipe pipe,
 					     bool enable, bool old)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	i915_reg_t reg = PIPESTAT(display, pipe);
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	if (enable) {
 		u32 enable_mask = i915_pipestat_enable_mask(display, pipe);
@@ -136,24 +133,22 @@ static void i9xx_set_fifo_underrun_reporting(struct intel_display *display,
 static void ilk_set_fifo_underrun_reporting(struct intel_display *display,
 					    enum pipe pipe, bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 bit = (pipe == PIPE_A) ?
 		DE_PIPEA_FIFO_UNDERRUN : DE_PIPEB_FIFO_UNDERRUN;
 
 	if (enable)
-		ilk_enable_display_irq(dev_priv, bit);
+		ilk_enable_display_irq(display, bit);
 	else
-		ilk_disable_display_irq(dev_priv, bit);
+		ilk_disable_display_irq(display, bit);
 }
 
 static void ivb_check_fifo_underruns(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	u32 err_int = intel_de_read(display, GEN7_ERR_INT);
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	if ((err_int & ERR_INT_FIFO_UNDERRUN(pipe)) == 0)
 		return;
@@ -169,7 +164,6 @@ static void ivb_set_fifo_underrun_reporting(struct intel_display *display,
 					    enum pipe pipe, bool enable,
 					    bool old)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	if (enable) {
 		intel_de_write(display, GEN7_ERR_INT,
 			       ERR_INT_FIFO_UNDERRUN(pipe));
@@ -177,9 +171,9 @@ static void ivb_set_fifo_underrun_reporting(struct intel_display *display,
 		if (!ivb_can_enable_err_int(display))
 			return;
 
-		ilk_enable_display_irq(dev_priv, DE_ERR_INT_IVB);
+		ilk_enable_display_irq(display, DE_ERR_INT_IVB);
 	} else {
-		ilk_disable_display_irq(dev_priv, DE_ERR_INT_IVB);
+		ilk_disable_display_irq(display, DE_ERR_INT_IVB);
 
 		if (old &&
 		    intel_de_read(display, GEN7_ERR_INT) & ERR_INT_FIFO_UNDERRUN(pipe)) {
@@ -193,36 +187,32 @@ static void ivb_set_fifo_underrun_reporting(struct intel_display *display,
 static void bdw_set_fifo_underrun_reporting(struct intel_display *display,
 					    enum pipe pipe, bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	if (enable)
-		bdw_enable_pipe_irq(dev_priv, pipe, GEN8_PIPE_FIFO_UNDERRUN);
+		bdw_enable_pipe_irq(display, pipe, GEN8_PIPE_FIFO_UNDERRUN);
 	else
-		bdw_disable_pipe_irq(dev_priv, pipe, GEN8_PIPE_FIFO_UNDERRUN);
+		bdw_disable_pipe_irq(display, pipe, GEN8_PIPE_FIFO_UNDERRUN);
 }
 
 static void ibx_set_fifo_underrun_reporting(struct intel_display *display,
 					    enum pipe pch_transcoder,
 					    bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 bit = (pch_transcoder == PIPE_A) ?
 		SDE_TRANSA_FIFO_UNDER : SDE_TRANSB_FIFO_UNDER;
 
 	if (enable)
-		ibx_enable_display_interrupt(dev_priv, bit);
+		ibx_enable_display_interrupt(display, bit);
 	else
-		ibx_disable_display_interrupt(dev_priv, bit);
+		ibx_disable_display_interrupt(display, bit);
 }
 
 static void cpt_check_pch_fifo_underruns(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pch_transcoder = crtc->pipe;
 	u32 serr_int = intel_de_read(display, SERR_INT);
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	if ((serr_int & SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder)) == 0)
 		return;
@@ -240,8 +230,6 @@ static void cpt_set_fifo_underrun_reporting(struct intel_display *display,
 					    enum pipe pch_transcoder,
 					    bool enable, bool old)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
 	if (enable) {
 		intel_de_write(display, SERR_INT,
 			       SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder));
@@ -249,9 +237,9 @@ static void cpt_set_fifo_underrun_reporting(struct intel_display *display,
 		if (!cpt_can_enable_serr_int(display))
 			return;
 
-		ibx_enable_display_interrupt(dev_priv, SDE_ERROR_CPT);
+		ibx_enable_display_interrupt(display, SDE_ERROR_CPT);
 	} else {
-		ibx_disable_display_interrupt(dev_priv, SDE_ERROR_CPT);
+		ibx_disable_display_interrupt(display, SDE_ERROR_CPT);
 
 		if (old && intel_de_read(display, SERR_INT) &
 		    SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder)) {
@@ -265,11 +253,10 @@ static void cpt_set_fifo_underrun_reporting(struct intel_display *display,
 static bool __intel_set_cpu_fifo_underrun_reporting(struct intel_display *display,
 						    enum pipe pipe, bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc = intel_crtc_for_pipe(display, pipe);
 	bool old;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	old = !crtc->cpu_fifo_underrun_disabled;
 	crtc->cpu_fifo_underrun_disabled = !enable;
@@ -305,13 +292,12 @@ static bool __intel_set_cpu_fifo_underrun_reporting(struct intel_display *displa
 bool intel_set_cpu_fifo_underrun_reporting(struct intel_display *display,
 					   enum pipe pipe, bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	unsigned long flags;
 	bool ret;
 
-	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	spin_lock_irqsave(&display->irq.lock, flags);
 	ret = __intel_set_cpu_fifo_underrun_reporting(display, pipe, enable);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+	spin_unlock_irqrestore(&display->irq.lock, flags);
 
 	return ret;
 }
@@ -334,7 +320,6 @@ bool intel_set_pch_fifo_underrun_reporting(struct intel_display *display,
 					   enum pipe pch_transcoder,
 					   bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc = intel_crtc_for_pipe(display, pch_transcoder);
 	unsigned long flags;
 	bool old;
@@ -348,12 +333,12 @@ bool intel_set_pch_fifo_underrun_reporting(struct intel_display *display,
 	 * crtc on LPT won't cause issues.
 	 */
 
-	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+	spin_lock_irqsave(&display->irq.lock, flags);
 
 	old = !crtc->pch_fifo_underrun_disabled;
 	crtc->pch_fifo_underrun_disabled = !enable;
 
-	if (HAS_PCH_IBX(dev_priv))
+	if (HAS_PCH_IBX(display))
 		ibx_set_fifo_underrun_reporting(display,
 						pch_transcoder,
 						enable);
@@ -362,7 +347,7 @@ bool intel_set_pch_fifo_underrun_reporting(struct intel_display *display,
 						pch_transcoder,
 						enable, old);
 
-	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+	spin_unlock_irqrestore(&display->irq.lock, flags);
 	return old;
 }
 
@@ -429,10 +414,9 @@ void intel_pch_fifo_underrun_irq_handler(struct intel_display *display,
  */
 void intel_check_cpu_fifo_underruns(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc;
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
 	for_each_intel_crtc(display->drm, crtc) {
 		if (crtc->cpu_fifo_underrun_disabled)
@@ -444,7 +428,7 @@ void intel_check_cpu_fifo_underruns(struct intel_display *display)
 			ivb_check_fifo_underruns(crtc);
 	}
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 /**
@@ -457,28 +441,25 @@ void intel_check_cpu_fifo_underruns(struct intel_display *display)
  */
 void intel_check_pch_fifo_underruns(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_crtc *crtc;
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
 	for_each_intel_crtc(display->drm, crtc) {
 		if (crtc->pch_fifo_underrun_disabled)
 			continue;
 
-		if (HAS_PCH_CPT(dev_priv))
+		if (HAS_PCH_CPT(display))
 			cpt_check_pch_fifo_underruns(crtc);
 	}
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 void intel_init_fifo_underrun_reporting(struct intel_display *display,
 					struct intel_crtc *crtc,
 					bool enable)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	crtc->cpu_fifo_underrun_disabled = !enable;
 
 	/*
@@ -490,6 +471,6 @@ void intel_init_fifo_underrun_reporting(struct intel_display *display,
 	 * PCH transcoders B and C would prevent enabling the south
 	 * error interrupt (see cpt_can_enable_serr_int()).
 	 */
-	if (intel_has_pch_trancoder(i915, crtc->pipe))
+	if (intel_has_pch_trancoder(display, crtc->pipe))
 		crtc->pch_fifo_underrun_disabled = !enable;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.c b/drivers/gpu/drm/i915/display/intel_frontbuffer.c
index ba2f88ca6117..43be5377ddc1 100644
--- a/drivers/gpu/drm/i915/display/intel_frontbuffer.c
+++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.c
@@ -58,7 +58,6 @@
 #include <drm/drm_gem.h>
 
 #include "i915_active.h"
-#include "i915_drv.h"
 #include "i915_vma.h"
 #include "intel_bo.h"
 #include "intel_display_trace.h"
@@ -72,7 +71,7 @@
 
 /**
  * frontbuffer_flush - flush frontbuffer
- * @i915: i915 device
+ * @display: display device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  * @origin: which operation caused the flush
  *
@@ -82,16 +81,14 @@
  *
  * Can be called without any locks held.
  */
-static void frontbuffer_flush(struct drm_i915_private *i915,
+static void frontbuffer_flush(struct intel_display *display,
 			      unsigned int frontbuffer_bits,
 			      enum fb_op_origin origin)
 {
-	struct intel_display *display = &i915->display;
-
 	/* Delay flushing when rings are still busy.*/
-	spin_lock(&i915->display.fb_tracking.lock);
-	frontbuffer_bits &= ~i915->display.fb_tracking.busy_bits;
-	spin_unlock(&i915->display.fb_tracking.lock);
+	spin_lock(&display->fb_tracking.lock);
+	frontbuffer_bits &= ~display->fb_tracking.busy_bits;
+	spin_unlock(&display->fb_tracking.lock);
 
 	if (!frontbuffer_bits)
 		return;
@@ -107,7 +104,7 @@ static void frontbuffer_flush(struct drm_i915_private *i915,
 
 /**
  * intel_frontbuffer_flip_prepare - prepare asynchronous frontbuffer flip
- * @i915: i915 device
+ * @display: display device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called after scheduling a flip on @obj. The actual
@@ -117,19 +114,19 @@ static void frontbuffer_flush(struct drm_i915_private *i915,
  *
  * Can be called without any locks held.
  */
-void intel_frontbuffer_flip_prepare(struct drm_i915_private *i915,
+void intel_frontbuffer_flip_prepare(struct intel_display *display,
 				    unsigned frontbuffer_bits)
 {
-	spin_lock(&i915->display.fb_tracking.lock);
-	i915->display.fb_tracking.flip_bits |= frontbuffer_bits;
+	spin_lock(&display->fb_tracking.lock);
+	display->fb_tracking.flip_bits |= frontbuffer_bits;
 	/* Remove stale busy bits due to the old buffer. */
-	i915->display.fb_tracking.busy_bits &= ~frontbuffer_bits;
-	spin_unlock(&i915->display.fb_tracking.lock);
+	display->fb_tracking.busy_bits &= ~frontbuffer_bits;
+	spin_unlock(&display->fb_tracking.lock);
 }
 
 /**
  * intel_frontbuffer_flip_complete - complete asynchronous frontbuffer flip
- * @i915: i915 device
+ * @display: display device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called after the flip has been latched and will complete
@@ -137,22 +134,22 @@ void intel_frontbuffer_flip_prepare(struct drm_i915_private *i915,
  *
  * Can be called without any locks held.
  */
-void intel_frontbuffer_flip_complete(struct drm_i915_private *i915,
+void intel_frontbuffer_flip_complete(struct intel_display *display,
 				     unsigned frontbuffer_bits)
 {
-	spin_lock(&i915->display.fb_tracking.lock);
+	spin_lock(&display->fb_tracking.lock);
 	/* Mask any cancelled flips. */
-	frontbuffer_bits &= i915->display.fb_tracking.flip_bits;
-	i915->display.fb_tracking.flip_bits &= ~frontbuffer_bits;
-	spin_unlock(&i915->display.fb_tracking.lock);
+	frontbuffer_bits &= display->fb_tracking.flip_bits;
+	display->fb_tracking.flip_bits &= ~frontbuffer_bits;
+	spin_unlock(&display->fb_tracking.lock);
 
 	if (frontbuffer_bits)
-		frontbuffer_flush(i915, frontbuffer_bits, ORIGIN_FLIP);
+		frontbuffer_flush(display, frontbuffer_bits, ORIGIN_FLIP);
 }
 
 /**
  * intel_frontbuffer_flip - synchronous frontbuffer flip
- * @i915: i915 device
+ * @display: display device
  * @frontbuffer_bits: frontbuffer plane tracking bits
  *
  * This function gets called after scheduling a flip on @obj. This is for
@@ -161,15 +158,15 @@ void intel_frontbuffer_flip_complete(struct drm_i915_private *i915,
  *
  * Can be called without any locks held.
  */
-void intel_frontbuffer_flip(struct drm_i915_private *i915,
+void intel_frontbuffer_flip(struct intel_display *display,
 			    unsigned frontbuffer_bits)
 {
-	spin_lock(&i915->display.fb_tracking.lock);
+	spin_lock(&display->fb_tracking.lock);
 	/* Remove stale busy bits due to the old buffer. */
-	i915->display.fb_tracking.busy_bits &= ~frontbuffer_bits;
-	spin_unlock(&i915->display.fb_tracking.lock);
+	display->fb_tracking.busy_bits &= ~frontbuffer_bits;
+	spin_unlock(&display->fb_tracking.lock);
 
-	frontbuffer_flush(i915, frontbuffer_bits, ORIGIN_FLIP);
+	frontbuffer_flush(display, frontbuffer_bits, ORIGIN_FLIP);
 }
 
 void __intel_fb_invalidate(struct intel_frontbuffer *front,
@@ -198,7 +195,6 @@ void __intel_fb_flush(struct intel_frontbuffer *front,
 		      unsigned int frontbuffer_bits)
 {
 	struct intel_display *display = to_intel_display(front->obj->dev);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	if (origin == ORIGIN_CS) {
 		spin_lock(&display->fb_tracking.lock);
@@ -209,7 +205,7 @@ void __intel_fb_flush(struct intel_frontbuffer *front,
 	}
 
 	if (frontbuffer_bits)
-		frontbuffer_flush(i915, frontbuffer_bits, origin);
+		frontbuffer_flush(display, frontbuffer_bits, origin);
 }
 
 static void intel_frontbuffer_flush_work(struct work_struct *work)
@@ -280,7 +276,7 @@ static void frontbuffer_release(struct kref *ref)
 struct intel_frontbuffer *
 intel_frontbuffer_get(struct drm_gem_object *obj)
 {
-	struct drm_i915_private *i915 = to_i915(obj->dev);
+	struct intel_display *display = to_intel_display(obj->dev);
 	struct intel_frontbuffer *front, *cur;
 
 	front = intel_bo_get_frontbuffer(obj);
@@ -300,9 +296,9 @@ intel_frontbuffer_get(struct drm_gem_object *obj)
 			 I915_ACTIVE_RETIRE_SLEEPS);
 	INIT_WORK(&front->flush_work, intel_frontbuffer_flush_work);
 
-	spin_lock(&i915->display.fb_tracking.lock);
+	spin_lock(&display->fb_tracking.lock);
 	cur = intel_bo_set_frontbuffer(obj, front);
-	spin_unlock(&i915->display.fb_tracking.lock);
+	spin_unlock(&display->fb_tracking.lock);
 	if (cur != front)
 		kfree(front);
 	return cur;
diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.h b/drivers/gpu/drm/i915/display/intel_frontbuffer.h
index 6237780a9f68..2fee12eaf9b6 100644
--- a/drivers/gpu/drm/i915/display/intel_frontbuffer.h
+++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.h
@@ -31,7 +31,7 @@
 #include "i915_active_types.h"
 
 struct drm_gem_object;
-struct drm_i915_private;
+struct intel_display;
 
 enum fb_op_origin {
 	ORIGIN_CPU = 0,
@@ -68,11 +68,11 @@ struct intel_frontbuffer {
 	GENMASK(INTEL_FRONTBUFFER_BITS_PER_PIPE * ((pipe) + 1) - 1,	\
 		INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))
 
-void intel_frontbuffer_flip_prepare(struct drm_i915_private *i915,
+void intel_frontbuffer_flip_prepare(struct intel_display *display,
 				    unsigned frontbuffer_bits);
-void intel_frontbuffer_flip_complete(struct drm_i915_private *i915,
+void intel_frontbuffer_flip_complete(struct intel_display *display,
 				     unsigned frontbuffer_bits);
-void intel_frontbuffer_flip(struct drm_i915_private *i915,
+void intel_frontbuffer_flip(struct intel_display *display,
 			    unsigned frontbuffer_bits);
 
 void intel_frontbuffer_put(struct intel_frontbuffer *front);
diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c
index 8a49e2bb37fa..000a898c9480 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.c
+++ b/drivers/gpu/drm/i915/display/intel_global_state.c
@@ -3,10 +3,13 @@
  * Copyright © 2020 Intel Corporation
  */
 
+#include <linux/pci.h>
 #include <linux/string.h>
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "intel_atomic.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_global_state.h"
 
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index abf457e68ee9..d55cc77650b7 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -152,32 +152,31 @@ static const struct gmbus_pin gmbus_pins_mtp[] = {
 static const struct gmbus_pin *get_gmbus_pin(struct intel_display *display,
 					     unsigned int pin)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	const struct gmbus_pin *pins;
 	size_t size;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_MTL) {
+	if (INTEL_PCH_TYPE(display) >= PCH_MTL) {
 		pins = gmbus_pins_mtp;
 		size = ARRAY_SIZE(gmbus_pins_mtp);
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_DG2) {
+	} else if (INTEL_PCH_TYPE(display) >= PCH_DG2) {
 		pins = gmbus_pins_dg2;
 		size = ARRAY_SIZE(gmbus_pins_dg2);
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_DG1) {
+	} else if (INTEL_PCH_TYPE(display) >= PCH_DG1) {
 		pins = gmbus_pins_dg1;
 		size = ARRAY_SIZE(gmbus_pins_dg1);
-	} else if (INTEL_PCH_TYPE(i915) >= PCH_ICP) {
+	} else if (INTEL_PCH_TYPE(display) >= PCH_ICP) {
 		pins = gmbus_pins_icp;
 		size = ARRAY_SIZE(gmbus_pins_icp);
-	} else if (HAS_PCH_CNP(i915)) {
+	} else if (HAS_PCH_CNP(display)) {
 		pins = gmbus_pins_cnp;
 		size = ARRAY_SIZE(gmbus_pins_cnp);
-	} else if (IS_GEMINILAKE(i915) || IS_BROXTON(i915)) {
+	} else if (display->platform.geminilake || display->platform.broxton) {
 		pins = gmbus_pins_bxt;
 		size = ARRAY_SIZE(gmbus_pins_bxt);
 	} else if (DISPLAY_VER(display) == 9) {
 		pins = gmbus_pins_skl;
 		size = ARRAY_SIZE(gmbus_pins_skl);
-	} else if (IS_BROADWELL(i915)) {
+	} else if (display->platform.broadwell) {
 		pins = gmbus_pins_bdw;
 		size = ARRAY_SIZE(gmbus_pins_bdw);
 	} else {
@@ -240,11 +239,10 @@ static void bxt_gmbus_clock_gating(struct intel_display *display,
 static u32 get_reserved(struct intel_gmbus *bus)
 {
 	struct intel_display *display = bus->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	u32 reserved = 0;
 
 	/* On most chips, these bits must be preserved in software. */
-	if (!IS_I830(i915) && !IS_I845G(i915))
+	if (!display->platform.i830 && !display->platform.i845g)
 		reserved = intel_de_read_notrace(display, bus->gpio_reg) &
 			(GPIO_DATA_PULLUP_DISABLE | GPIO_CLOCK_PULLUP_DISABLE);
 
@@ -314,11 +312,10 @@ intel_gpio_pre_xfer(struct i2c_adapter *adapter)
 {
 	struct intel_gmbus *bus = to_intel_gmbus(adapter);
 	struct intel_display *display = bus->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	intel_gmbus_reset(display);
 
-	if (IS_PINEVIEW(i915))
+	if (display->platform.pineview)
 		pnv_gmbus_clock_gating(display, false);
 
 	set_data(bus, 1);
@@ -332,12 +329,11 @@ intel_gpio_post_xfer(struct i2c_adapter *adapter)
 {
 	struct intel_gmbus *bus = to_intel_gmbus(adapter);
 	struct intel_display *display = bus->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	set_data(bus, 1);
 	set_clock(bus, 1);
 
-	if (IS_PINEVIEW(i915))
+	if (display->platform.pineview)
 		pnv_gmbus_clock_gating(display, true);
 }
 
@@ -630,14 +626,13 @@ do_gmbus_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs, int num,
 {
 	struct intel_gmbus *bus = to_intel_gmbus(adapter);
 	struct intel_display *display = bus->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	int i = 0, inc, try = 0;
 	int ret = 0;
 
 	/* Display WA #0868: skl,bxt,kbl,cfl,glk */
-	if (IS_GEMINILAKE(i915) || IS_BROXTON(i915))
+	if (display->platform.geminilake || display->platform.broxton)
 		bxt_gmbus_clock_gating(display, false);
-	else if (HAS_PCH_SPT(i915) || HAS_PCH_CNP(i915))
+	else if (HAS_PCH_SPT(display) || HAS_PCH_CNP(display))
 		pch_gmbus_clock_gating(display, false);
 
 retry:
@@ -748,9 +743,9 @@ timeout:
 
 out:
 	/* Display WA #0868: skl,bxt,kbl,cfl,glk */
-	if (IS_GEMINILAKE(i915) || IS_BROXTON(i915))
+	if (display->platform.geminilake || display->platform.broxton)
 		bxt_gmbus_clock_gating(display, true);
-	else if (HAS_PCH_SPT(i915) || HAS_PCH_CNP(i915))
+	else if (HAS_PCH_SPT(display) || HAS_PCH_CNP(display))
 		pch_gmbus_clock_gating(display, true);
 
 	return ret;
@@ -873,12 +868,11 @@ static const struct i2c_lock_operations gmbus_lock_ops = {
  */
 int intel_gmbus_setup(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct pci_dev *pdev = to_pci_dev(display->drm->dev);
 	unsigned int pin;
 	int ret;
 
-	if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915))
+	if (display->platform.valleyview || display->platform.cherryview)
 		display->gmbus.mmio_base = VLV_DISPLAY_BASE;
 	else if (!HAS_GMCH(display))
 		/*
@@ -925,7 +919,7 @@ int intel_gmbus_setup(struct intel_display *display)
 		bus->reg0 = pin | GMBUS_RATE_100KHZ;
 
 		/* gmbus seems to be broken on i830 */
-		if (IS_I830(i915))
+		if (display->platform.i830)
 			bus->force_bit = 1;
 
 		intel_gpio_setup(bus, GPIO(display, gmbus_pin->gpio));
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 1bf424a822f3..3e3038f4ee1f 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -22,13 +22,18 @@
 #include "intel_de.h"
 #include "intel_display_power.h"
 #include "intel_display_power_well.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
+#include "intel_dp_mst.h"
 #include "intel_hdcp.h"
 #include "intel_hdcp_gsc.h"
+#include "intel_hdcp_gsc_message.h"
 #include "intel_hdcp_regs.h"
 #include "intel_hdcp_shim.h"
 #include "intel_pcode.h"
 
+#define USE_HDCP_GSC(__display)		(DISPLAY_VER(__display) >= 14)
+
 #define KEY_LOAD_TRIES	5
 #define HDCP2_LC_RETRY_CNT			3
 
@@ -136,7 +141,7 @@ intel_hdcp_required_content_stream(struct intel_atomic_state *state,
 		data->k++;
 
 		/* if there is only one active stream */
-		if (dig_port->dp.mst.active_links <= 1)
+		if (intel_dp_mst_active_streams(&dig_port->dp) <= 1)
 			break;
 	}
 	drm_connector_list_iter_end(&conn_iter);
@@ -248,8 +253,8 @@ static bool intel_hdcp2_prerequisite(struct intel_connector *connector)
 		return false;
 
 	/* If MTL+ make sure gsc is loaded and proxy is setup */
-	if (intel_hdcp_gsc_cs_required(display)) {
-		if (!intel_hdcp_gsc_check_status(display))
+	if (USE_HDCP_GSC(display)) {
+		if (!intel_hdcp_gsc_check_status(display->drm))
 			return false;
 	}
 
@@ -334,9 +339,7 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port,
 
 static bool hdcp_key_loadable(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	enum i915_power_well_id id;
-	intel_wakeref_t wakeref;
 	bool enabled = false;
 
 	/*
@@ -349,7 +352,7 @@ static bool hdcp_key_loadable(struct intel_display *display)
 		id = SKL_DISP_PW_1;
 
 	/* PG1 (power well #1) needs to be enabled */
-	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+	with_intel_display_rpm(display)
 		enabled = intel_display_power_well_is_enabled(display, id);
 
 	/*
@@ -2339,7 +2342,7 @@ static int initialize_hdcp_port_data(struct intel_connector *connector,
 
 static bool is_hdcp2_supported(struct intel_display *display)
 {
-	if (intel_hdcp_gsc_cs_required(display))
+	if (USE_HDCP_GSC(display))
 		return true;
 
 	if (!IS_ENABLED(CONFIG_INTEL_MEI_HDCP))
@@ -2363,7 +2366,7 @@ void intel_hdcp_component_init(struct intel_display *display)
 
 	display->hdcp.comp_added = true;
 	mutex_unlock(&display->hdcp.hdcp_mutex);
-	if (intel_hdcp_gsc_cs_required(display))
+	if (USE_HDCP_GSC(display))
 		ret = intel_hdcp_gsc_init(display);
 	else
 		ret = component_add_typed(display->drm->dev, &i915_hdcp_ops,
@@ -2638,7 +2641,7 @@ void intel_hdcp_component_fini(struct intel_display *display)
 	display->hdcp.comp_added = false;
 	mutex_unlock(&display->hdcp.hdcp_mutex);
 
-	if (intel_hdcp_gsc_cs_required(display))
+	if (USE_HDCP_GSC(display))
 		intel_hdcp_gsc_fini(display);
 	else
 		component_del(display->drm->dev, &i915_hdcp_ops);
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
index 55965844d829..6a22862d6be1 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
@@ -11,27 +11,22 @@
 #include "i915_drv.h"
 #include "i915_utils.h"
 #include "intel_hdcp_gsc.h"
-#include "intel_hdcp_gsc_message.h"
 
-struct intel_hdcp_gsc_message {
+struct intel_hdcp_gsc_context {
+	struct drm_i915_private *i915;
 	struct i915_vma *vma;
 	void *hdcp_cmd_in;
 	void *hdcp_cmd_out;
 };
 
-bool intel_hdcp_gsc_cs_required(struct intel_display *display)
+bool intel_hdcp_gsc_check_status(struct drm_device *drm)
 {
-	return DISPLAY_VER(display) >= 14;
-}
-
-bool intel_hdcp_gsc_check_status(struct intel_display *display)
-{
-	struct drm_i915_private *i915 = to_i915(display->drm);
+	struct drm_i915_private *i915 = to_i915(drm);
 	struct intel_gt *gt = i915->media_gt;
 	struct intel_gsc_uc *gsc = gt ? &gt->uc.gsc : NULL;
 
 	if (!gsc || !intel_uc_fw_is_running(&gsc->fw)) {
-		drm_dbg_kms(display->drm,
+		drm_dbg_kms(&i915->drm,
 			    "GSC components required for HDCP2.2 are not ready\n");
 		return false;
 	}
@@ -41,7 +36,7 @@ bool intel_hdcp_gsc_check_status(struct intel_display *display)
 
 /*This function helps allocate memory for the command that we will send to gsc cs */
 static int intel_hdcp_gsc_initialize_message(struct drm_i915_private *i915,
-					     struct intel_hdcp_gsc_message *hdcp_message)
+					     struct intel_hdcp_gsc_context *gsc_context)
 {
 	struct intel_gt *gt = i915->media_gt;
 	struct drm_i915_gem_object *obj = NULL;
@@ -78,9 +73,10 @@ static int intel_hdcp_gsc_initialize_message(struct drm_i915_private *i915,
 
 	memset(cmd_in, 0, obj->base.size);
 
-	hdcp_message->hdcp_cmd_in = cmd_in;
-	hdcp_message->hdcp_cmd_out = cmd_out;
-	hdcp_message->vma = vma;
+	gsc_context->hdcp_cmd_in = cmd_in;
+	gsc_context->hdcp_cmd_out = cmd_out;
+	gsc_context->vma = vma;
+	gsc_context->i915 = i915;
 
 	return 0;
 
@@ -91,80 +87,37 @@ out_unpin:
 	return err;
 }
 
-static const struct i915_hdcp_ops gsc_hdcp_ops = {
-	.initiate_hdcp2_session = intel_hdcp_gsc_initiate_session,
-	.verify_receiver_cert_prepare_km =
-				intel_hdcp_gsc_verify_receiver_cert_prepare_km,
-	.verify_hprime = intel_hdcp_gsc_verify_hprime,
-	.store_pairing_info = intel_hdcp_gsc_store_pairing_info,
-	.initiate_locality_check = intel_hdcp_gsc_initiate_locality_check,
-	.verify_lprime = intel_hdcp_gsc_verify_lprime,
-	.get_session_key = intel_hdcp_gsc_get_session_key,
-	.repeater_check_flow_prepare_ack =
-				intel_hdcp_gsc_repeater_check_flow_prepare_ack,
-	.verify_mprime = intel_hdcp_gsc_verify_mprime,
-	.enable_hdcp_authentication = intel_hdcp_gsc_enable_authentication,
-	.close_hdcp_session = intel_hdcp_gsc_close_session,
-};
-
-static int intel_hdcp_gsc_hdcp2_init(struct intel_display *display)
+struct intel_hdcp_gsc_context *intel_hdcp_gsc_context_alloc(struct drm_device *drm)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-	struct intel_hdcp_gsc_message *hdcp_message;
+	struct drm_i915_private *i915 = to_i915(drm);
+	struct intel_hdcp_gsc_context *gsc_context;
 	int ret;
 
-	hdcp_message = kzalloc(sizeof(*hdcp_message), GFP_KERNEL);
-
-	if (!hdcp_message)
-		return -ENOMEM;
+	gsc_context = kzalloc(sizeof(*gsc_context), GFP_KERNEL);
+	if (!gsc_context)
+		return ERR_PTR(-ENOMEM);
 
 	/*
 	 * NOTE: No need to lock the comp mutex here as it is already
 	 * going to be taken before this function called
 	 */
-	display->hdcp.hdcp_message = hdcp_message;
-	ret = intel_hdcp_gsc_initialize_message(i915, hdcp_message);
-
-	if (ret)
-		drm_err(display->drm, "Could not initialize hdcp_message\n");
-
-	return ret;
-}
-
-static void intel_hdcp_gsc_free_message(struct intel_display *display)
-{
-	struct intel_hdcp_gsc_message *hdcp_message =
-					display->hdcp.hdcp_message;
+	ret = intel_hdcp_gsc_initialize_message(i915, gsc_context);
+	if (ret) {
+		drm_err(&i915->drm, "Could not initialize gsc_context\n");
+		kfree(gsc_context);
+		gsc_context = ERR_PTR(ret);
+	}
 
-	hdcp_message->hdcp_cmd_in = NULL;
-	hdcp_message->hdcp_cmd_out = NULL;
-	i915_vma_unpin_and_release(&hdcp_message->vma, I915_VMA_RELEASE_MAP);
-	kfree(hdcp_message);
+	return gsc_context;
 }
 
-int intel_hdcp_gsc_init(struct intel_display *display)
+void intel_hdcp_gsc_context_free(struct intel_hdcp_gsc_context *gsc_context)
 {
-	struct i915_hdcp_arbiter *data;
-	int ret;
-
-	data = kzalloc(sizeof(struct i915_hdcp_arbiter), GFP_KERNEL);
-	if (!data)
-		return -ENOMEM;
+	if (!gsc_context)
+		return;
 
-	mutex_lock(&display->hdcp.hdcp_mutex);
-	display->hdcp.arbiter = data;
-	display->hdcp.arbiter->hdcp_dev = display->drm->dev;
-	display->hdcp.arbiter->ops = &gsc_hdcp_ops;
-	ret = intel_hdcp_gsc_hdcp2_init(display);
-	mutex_unlock(&display->hdcp.hdcp_mutex);
-
-	return ret;
-}
-
-void intel_hdcp_gsc_fini(struct intel_display *display)
-{
-	intel_hdcp_gsc_free_message(display);
-	kfree(display->hdcp.arbiter);
+	i915_vma_unpin_and_release(&gsc_context->vma, I915_VMA_RELEASE_MAP);
+	kfree(gsc_context);
 }
 
 static int intel_gsc_send_sync(struct drm_i915_private *i915,
@@ -211,18 +164,18 @@ static int intel_gsc_send_sync(struct drm_i915_private *i915,
 /*
  * This function can now be used for sending requests and will also handle
  * receipt of reply messages hence no different function of message retrieval
- * is required. We will initialize intel_hdcp_gsc_message structure then add
+ * is required. We will initialize intel_hdcp_gsc_context structure then add
  * gsc cs memory header as stated in specs after which the normal HDCP payload
  * will follow
  */
-ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in,
-				size_t msg_in_len, u8 *msg_out,
-				size_t msg_out_len)
+ssize_t intel_hdcp_gsc_msg_send(struct intel_hdcp_gsc_context *gsc_context,
+				void *msg_in, size_t msg_in_len,
+				void *msg_out, size_t msg_out_len)
 {
+	struct drm_i915_private *i915 = gsc_context->i915;
 	struct intel_gt *gt = i915->media_gt;
 	struct intel_gsc_mtl_header *header_in, *header_out;
 	const size_t max_msg_size = PAGE_SIZE - sizeof(*header_in);
-	struct intel_hdcp_gsc_message *hdcp_message;
 	u64 addr_in, addr_out, host_session_id;
 	u32 reply_size, msg_size_in, msg_size_out;
 	int ret, tries = 0;
@@ -235,10 +188,9 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in,
 
 	msg_size_in = msg_in_len + sizeof(*header_in);
 	msg_size_out = msg_out_len + sizeof(*header_out);
-	hdcp_message = i915->display.hdcp.hdcp_message;
-	header_in = hdcp_message->hdcp_cmd_in;
-	header_out = hdcp_message->hdcp_cmd_out;
-	addr_in = i915_ggtt_offset(hdcp_message->vma);
+	header_in = gsc_context->hdcp_cmd_in;
+	header_out = gsc_context->hdcp_cmd_out;
+	addr_in = i915_ggtt_offset(gsc_context->vma);
 	addr_out = addr_in + PAGE_SIZE;
 
 	memset(header_in, 0, msg_size_in);
@@ -246,7 +198,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in,
 	get_random_bytes(&host_session_id, sizeof(u64));
 	intel_gsc_uc_heci_cmd_emit_mtl_header(header_in, HECI_MEADDRESS_HDCP,
 					      msg_size_in, host_session_id);
-	memcpy(hdcp_message->hdcp_cmd_in + sizeof(*header_in), msg_in, msg_in_len);
+	memcpy(gsc_context->hdcp_cmd_in + sizeof(*header_in), msg_in, msg_in_len);
 
 	/*
 	 * Keep sending request in case the pending bit is set no need to add
@@ -280,7 +232,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in,
 			    reply_size, (u32)msg_out_len);
 	}
 
-	memcpy(msg_out, hdcp_message->hdcp_cmd_out + sizeof(*header_out), msg_out_len);
+	memcpy(msg_out, gsc_context->hdcp_cmd_out + sizeof(*header_out), msg_out_len);
 
 err:
 	return ret;
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h
index 5695a5e4f609..9305c14aaffe 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h
@@ -6,19 +6,17 @@
 #ifndef __INTEL_HDCP_GSC_H__
 #define __INTEL_HDCP_GSC_H__
 
-#include <linux/err.h>
 #include <linux/types.h>
 
-struct drm_i915_private;
-struct intel_display;
-struct intel_hdcp_gsc_message;
+struct drm_device;
+struct intel_hdcp_gsc_context;
 
-bool intel_hdcp_gsc_cs_required(struct intel_display *display);
-ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in,
-				size_t msg_in_len, u8 *msg_out,
-				size_t msg_out_len);
-int intel_hdcp_gsc_init(struct intel_display *display);
-void intel_hdcp_gsc_fini(struct intel_display *display);
-bool intel_hdcp_gsc_check_status(struct intel_display *display);
+ssize_t intel_hdcp_gsc_msg_send(struct intel_hdcp_gsc_context *gsc_context,
+				void *msg_in, size_t msg_in_len,
+				void *msg_out, size_t msg_out_len);
+bool intel_hdcp_gsc_check_status(struct drm_device *drm);
+
+struct intel_hdcp_gsc_context *intel_hdcp_gsc_context_alloc(struct drm_device *drm);
+void intel_hdcp_gsc_context_free(struct intel_hdcp_gsc_context *gsc_context);
 
 #endif /* __INTEL_HDCP_GCS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.c
index 129104fa9b16..98967bb148e3 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.c
@@ -4,20 +4,23 @@
  */
 
 #include <linux/err.h>
+
+#include <drm/drm_print.h>
 #include <drm/intel/i915_hdcp_interface.h>
 
-#include "i915_drv.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
+#include "intel_hdcp_gsc.h"
 #include "intel_hdcp_gsc_message.h"
 
-int
+static int
 intel_hdcp_gsc_initiate_session(struct device *dev, struct hdcp_port_data *data,
 				struct hdcp2_ake_init *ake_data)
 {
 	struct wired_cmd_initiate_hdcp2_session_in session_init_in = {};
 	struct wired_cmd_initiate_hdcp2_session_out session_init_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !ake_data)
@@ -28,7 +31,7 @@ intel_hdcp_gsc_initiate_session(struct device *dev, struct hdcp_port_data *data,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	session_init_in.header.api_version = HDCP_API_VERSION;
 	session_init_in.header.command_id = WIRED_INITIATE_HDCP2_SESSION;
@@ -41,9 +44,9 @@ intel_hdcp_gsc_initiate_session(struct device *dev, struct hdcp_port_data *data,
 	session_init_in.port.attached_transcoder = (u8)data->hdcp_transcoder;
 	session_init_in.protocol = data->protocol;
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&session_init_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &session_init_in,
 				       sizeof(session_init_in),
-				       (u8 *)&session_init_out,
+				       &session_init_out,
 				       sizeof(session_init_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -64,7 +67,7 @@ intel_hdcp_gsc_initiate_session(struct device *dev, struct hdcp_port_data *data,
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_verify_receiver_cert_prepare_km(struct device *dev,
 					       struct hdcp_port_data *data,
 					       struct hdcp2_ake_send_cert *rx_cert,
@@ -75,8 +78,8 @@ intel_hdcp_gsc_verify_receiver_cert_prepare_km(struct device *dev,
 {
 	struct wired_cmd_verify_receiver_cert_in verify_rxcert_in = {};
 	struct wired_cmd_verify_receiver_cert_out verify_rxcert_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !rx_cert || !km_stored || !ek_pub_km || !msg_sz)
@@ -87,7 +90,7 @@ intel_hdcp_gsc_verify_receiver_cert_prepare_km(struct device *dev,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	verify_rxcert_in.header.api_version = HDCP_API_VERSION;
 	verify_rxcert_in.header.command_id = WIRED_VERIFY_RECEIVER_CERT;
@@ -103,9 +106,9 @@ intel_hdcp_gsc_verify_receiver_cert_prepare_km(struct device *dev,
 	memcpy(verify_rxcert_in.r_rx, &rx_cert->r_rx, HDCP_2_2_RRX_LEN);
 	memcpy(verify_rxcert_in.rx_caps, rx_cert->rx_caps, HDCP_2_2_RXCAPS_LEN);
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&verify_rxcert_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &verify_rxcert_in,
 				       sizeof(verify_rxcert_in),
-				       (u8 *)&verify_rxcert_out,
+				       &verify_rxcert_out,
 				       sizeof(verify_rxcert_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed: %zd\n", byte);
@@ -134,14 +137,14 @@ intel_hdcp_gsc_verify_receiver_cert_prepare_km(struct device *dev,
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_verify_hprime(struct device *dev, struct hdcp_port_data *data,
 			     struct hdcp2_ake_send_hprime *rx_hprime)
 {
 	struct wired_cmd_ake_send_hprime_in send_hprime_in = {};
 	struct wired_cmd_ake_send_hprime_out send_hprime_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !rx_hprime)
@@ -152,7 +155,7 @@ intel_hdcp_gsc_verify_hprime(struct device *dev, struct hdcp_port_data *data,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	send_hprime_in.header.api_version = HDCP_API_VERSION;
 	send_hprime_in.header.command_id = WIRED_AKE_SEND_HPRIME;
@@ -166,9 +169,9 @@ intel_hdcp_gsc_verify_hprime(struct device *dev, struct hdcp_port_data *data,
 	memcpy(send_hprime_in.h_prime, rx_hprime->h_prime,
 	       HDCP_2_2_H_PRIME_LEN);
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&send_hprime_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &send_hprime_in,
 				       sizeof(send_hprime_in),
-				       (u8 *)&send_hprime_out,
+				       &send_hprime_out,
 				       sizeof(send_hprime_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -184,14 +187,14 @@ intel_hdcp_gsc_verify_hprime(struct device *dev, struct hdcp_port_data *data,
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_store_pairing_info(struct device *dev, struct hdcp_port_data *data,
 				  struct hdcp2_ake_send_pairing_info *pairing_info)
 {
 	struct wired_cmd_ake_send_pairing_info_in pairing_info_in = {};
 	struct wired_cmd_ake_send_pairing_info_out pairing_info_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !pairing_info)
@@ -202,7 +205,7 @@ intel_hdcp_gsc_store_pairing_info(struct device *dev, struct hdcp_port_data *dat
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	pairing_info_in.header.api_version = HDCP_API_VERSION;
 	pairing_info_in.header.command_id = WIRED_AKE_SEND_PAIRING_INFO;
@@ -217,9 +220,9 @@ intel_hdcp_gsc_store_pairing_info(struct device *dev, struct hdcp_port_data *dat
 	memcpy(pairing_info_in.e_kh_km, pairing_info->e_kh_km,
 	       HDCP_2_2_E_KH_KM_LEN);
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&pairing_info_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &pairing_info_in,
 				       sizeof(pairing_info_in),
-				       (u8 *)&pairing_info_out,
+				       &pairing_info_out,
 				       sizeof(pairing_info_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -236,15 +239,15 @@ intel_hdcp_gsc_store_pairing_info(struct device *dev, struct hdcp_port_data *dat
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_initiate_locality_check(struct device *dev,
 				       struct hdcp_port_data *data,
 				       struct hdcp2_lc_init *lc_init_data)
 {
 	struct wired_cmd_init_locality_check_in lc_init_in = {};
 	struct wired_cmd_init_locality_check_out lc_init_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !lc_init_data)
@@ -255,7 +258,7 @@ intel_hdcp_gsc_initiate_locality_check(struct device *dev,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	lc_init_in.header.api_version = HDCP_API_VERSION;
 	lc_init_in.header.command_id = WIRED_INIT_LOCALITY_CHECK;
@@ -266,8 +269,8 @@ intel_hdcp_gsc_initiate_locality_check(struct device *dev,
 	lc_init_in.port.physical_port = (u8)data->hdcp_ddi;
 	lc_init_in.port.attached_transcoder = (u8)data->hdcp_transcoder;
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&lc_init_in, sizeof(lc_init_in),
-				       (u8 *)&lc_init_out, sizeof(lc_init_out));
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &lc_init_in, sizeof(lc_init_in),
+				       &lc_init_out, sizeof(lc_init_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
 		return byte;
@@ -285,14 +288,14 @@ intel_hdcp_gsc_initiate_locality_check(struct device *dev,
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_verify_lprime(struct device *dev, struct hdcp_port_data *data,
 			     struct hdcp2_lc_send_lprime *rx_lprime)
 {
 	struct wired_cmd_validate_locality_in verify_lprime_in = {};
 	struct wired_cmd_validate_locality_out verify_lprime_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !rx_lprime)
@@ -303,7 +306,7 @@ intel_hdcp_gsc_verify_lprime(struct device *dev, struct hdcp_port_data *data,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	verify_lprime_in.header.api_version = HDCP_API_VERSION;
 	verify_lprime_in.header.command_id = WIRED_VALIDATE_LOCALITY;
@@ -318,9 +321,9 @@ intel_hdcp_gsc_verify_lprime(struct device *dev, struct hdcp_port_data *data,
 	memcpy(verify_lprime_in.l_prime, rx_lprime->l_prime,
 	       HDCP_2_2_L_PRIME_LEN);
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&verify_lprime_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &verify_lprime_in,
 				       sizeof(verify_lprime_in),
-				       (u8 *)&verify_lprime_out,
+				       &verify_lprime_out,
 				       sizeof(verify_lprime_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -337,14 +340,15 @@ intel_hdcp_gsc_verify_lprime(struct device *dev, struct hdcp_port_data *data,
 	return 0;
 }
 
-int intel_hdcp_gsc_get_session_key(struct device *dev,
-				   struct hdcp_port_data *data,
-				   struct hdcp2_ske_send_eks *ske_data)
+static int
+intel_hdcp_gsc_get_session_key(struct device *dev,
+			       struct hdcp_port_data *data,
+			       struct hdcp2_ske_send_eks *ske_data)
 {
 	struct wired_cmd_get_session_key_in get_skey_in = {};
 	struct wired_cmd_get_session_key_out get_skey_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data || !ske_data)
@@ -355,7 +359,7 @@ int intel_hdcp_gsc_get_session_key(struct device *dev,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	get_skey_in.header.api_version = HDCP_API_VERSION;
 	get_skey_in.header.command_id = WIRED_GET_SESSION_KEY;
@@ -366,8 +370,8 @@ int intel_hdcp_gsc_get_session_key(struct device *dev,
 	get_skey_in.port.physical_port = (u8)data->hdcp_ddi;
 	get_skey_in.port.attached_transcoder = (u8)data->hdcp_transcoder;
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&get_skey_in, sizeof(get_skey_in),
-				       (u8 *)&get_skey_out, sizeof(get_skey_out));
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &get_skey_in, sizeof(get_skey_in),
+				       &get_skey_out, sizeof(get_skey_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
 		return byte;
@@ -387,7 +391,7 @@ int intel_hdcp_gsc_get_session_key(struct device *dev,
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_repeater_check_flow_prepare_ack(struct device *dev,
 					       struct hdcp_port_data *data,
 					       struct hdcp2_rep_send_receiverid_list
@@ -397,8 +401,8 @@ intel_hdcp_gsc_repeater_check_flow_prepare_ack(struct device *dev,
 {
 	struct wired_cmd_verify_repeater_in verify_repeater_in = {};
 	struct wired_cmd_verify_repeater_out verify_repeater_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !rep_topology || !rep_send_ack || !data)
@@ -409,7 +413,7 @@ intel_hdcp_gsc_repeater_check_flow_prepare_ack(struct device *dev,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	verify_repeater_in.header.api_version = HDCP_API_VERSION;
 	verify_repeater_in.header.command_id = WIRED_VERIFY_REPEATER;
@@ -430,9 +434,9 @@ intel_hdcp_gsc_repeater_check_flow_prepare_ack(struct device *dev,
 	memcpy(verify_repeater_in.receiver_ids, rep_topology->receiver_ids,
 	       HDCP_2_2_RECEIVER_IDS_MAX_LEN);
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&verify_repeater_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &verify_repeater_in,
 				       sizeof(verify_repeater_in),
-				       (u8 *)&verify_repeater_out,
+				       &verify_repeater_out,
 				       sizeof(verify_repeater_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -453,14 +457,15 @@ intel_hdcp_gsc_repeater_check_flow_prepare_ack(struct device *dev,
 	return 0;
 }
 
-int intel_hdcp_gsc_verify_mprime(struct device *dev,
-				 struct hdcp_port_data *data,
-				 struct hdcp2_rep_stream_ready *stream_ready)
+static int
+intel_hdcp_gsc_verify_mprime(struct device *dev,
+			     struct hdcp_port_data *data,
+			     struct hdcp2_rep_stream_ready *stream_ready)
 {
 	struct wired_cmd_repeater_auth_stream_req_in *verify_mprime_in;
 	struct wired_cmd_repeater_auth_stream_req_out verify_mprime_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 	size_t cmd_size;
 
@@ -472,7 +477,7 @@ int intel_hdcp_gsc_verify_mprime(struct device *dev,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	cmd_size = struct_size(verify_mprime_in, streams, data->k);
 	if (cmd_size == SIZE_MAX)
@@ -499,8 +504,8 @@ int intel_hdcp_gsc_verify_mprime(struct device *dev,
 
 	verify_mprime_in->k = cpu_to_be16(data->k);
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)verify_mprime_in, cmd_size,
-				       (u8 *)&verify_mprime_out,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, verify_mprime_in, cmd_size,
+				       &verify_mprime_out,
 				       sizeof(verify_mprime_out));
 	kfree(verify_mprime_in);
 	if (byte < 0) {
@@ -518,13 +523,13 @@ int intel_hdcp_gsc_verify_mprime(struct device *dev,
 	return 0;
 }
 
-int intel_hdcp_gsc_enable_authentication(struct device *dev,
-					 struct hdcp_port_data *data)
+static int intel_hdcp_gsc_enable_authentication(struct device *dev,
+						struct hdcp_port_data *data)
 {
 	struct wired_cmd_enable_auth_in enable_auth_in = {};
 	struct wired_cmd_enable_auth_out enable_auth_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data)
@@ -535,7 +540,7 @@ int intel_hdcp_gsc_enable_authentication(struct device *dev,
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	enable_auth_in.header.api_version = HDCP_API_VERSION;
 	enable_auth_in.header.command_id = WIRED_ENABLE_AUTH;
@@ -547,9 +552,9 @@ int intel_hdcp_gsc_enable_authentication(struct device *dev,
 	enable_auth_in.port.attached_transcoder = (u8)data->hdcp_transcoder;
 	enable_auth_in.stream_type = data->streams[0].stream_type;
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&enable_auth_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &enable_auth_in,
 				       sizeof(enable_auth_in),
-				       (u8 *)&enable_auth_out,
+				       &enable_auth_out,
 				       sizeof(enable_auth_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -565,13 +570,13 @@ int intel_hdcp_gsc_enable_authentication(struct device *dev,
 	return 0;
 }
 
-int
+static int
 intel_hdcp_gsc_close_session(struct device *dev, struct hdcp_port_data *data)
 {
 	struct wired_cmd_close_session_in session_close_in = {};
 	struct wired_cmd_close_session_out session_close_out = {};
+	struct intel_hdcp_gsc_context *gsc_context;
 	struct intel_display *display;
-	struct drm_i915_private *i915;
 	ssize_t byte;
 
 	if (!dev || !data)
@@ -582,7 +587,7 @@ intel_hdcp_gsc_close_session(struct device *dev, struct hdcp_port_data *data)
 		dev_err(dev, "DRM not initialized, aborting HDCP.\n");
 		return -ENODEV;
 	}
-	i915 = to_i915(display->drm);
+	gsc_context = display->hdcp.gsc_context;
 
 	session_close_in.header.api_version = HDCP_API_VERSION;
 	session_close_in.header.command_id = WIRED_CLOSE_SESSION;
@@ -594,9 +599,9 @@ intel_hdcp_gsc_close_session(struct device *dev, struct hdcp_port_data *data)
 	session_close_in.port.physical_port = (u8)data->hdcp_ddi;
 	session_close_in.port.attached_transcoder = (u8)data->hdcp_transcoder;
 
-	byte = intel_hdcp_gsc_msg_send(i915, (u8 *)&session_close_in,
+	byte = intel_hdcp_gsc_msg_send(gsc_context, &session_close_in,
 				       sizeof(session_close_in),
-				       (u8 *)&session_close_out,
+				       &session_close_out,
 				       sizeof(session_close_out));
 	if (byte < 0) {
 		drm_dbg_kms(display->drm, "intel_hdcp_gsc_msg_send failed. %zd\n", byte);
@@ -611,3 +616,57 @@ intel_hdcp_gsc_close_session(struct device *dev, struct hdcp_port_data *data)
 
 	return 0;
 }
+
+static const struct i915_hdcp_ops gsc_hdcp_ops = {
+	.initiate_hdcp2_session = intel_hdcp_gsc_initiate_session,
+	.verify_receiver_cert_prepare_km =
+				intel_hdcp_gsc_verify_receiver_cert_prepare_km,
+	.verify_hprime = intel_hdcp_gsc_verify_hprime,
+	.store_pairing_info = intel_hdcp_gsc_store_pairing_info,
+	.initiate_locality_check = intel_hdcp_gsc_initiate_locality_check,
+	.verify_lprime = intel_hdcp_gsc_verify_lprime,
+	.get_session_key = intel_hdcp_gsc_get_session_key,
+	.repeater_check_flow_prepare_ack =
+				intel_hdcp_gsc_repeater_check_flow_prepare_ack,
+	.verify_mprime = intel_hdcp_gsc_verify_mprime,
+	.enable_hdcp_authentication = intel_hdcp_gsc_enable_authentication,
+	.close_hdcp_session = intel_hdcp_gsc_close_session,
+};
+
+int intel_hdcp_gsc_init(struct intel_display *display)
+{
+	struct intel_hdcp_gsc_context *gsc_context;
+	struct i915_hdcp_arbiter *arbiter;
+	int ret = 0;
+
+	arbiter = kzalloc(sizeof(*arbiter), GFP_KERNEL);
+	if (!arbiter)
+		return -ENOMEM;
+
+	mutex_lock(&display->hdcp.hdcp_mutex);
+
+	gsc_context = intel_hdcp_gsc_context_alloc(display->drm);
+	if (IS_ERR(gsc_context)) {
+		ret = PTR_ERR(gsc_context);
+		kfree(arbiter);
+		goto out;
+	}
+
+	display->hdcp.arbiter = arbiter;
+	display->hdcp.arbiter->hdcp_dev = display->drm->dev;
+	display->hdcp.arbiter->ops = &gsc_hdcp_ops;
+	display->hdcp.gsc_context = gsc_context;
+
+out:
+	mutex_unlock(&display->hdcp.hdcp_mutex);
+
+	return ret;
+}
+
+void intel_hdcp_gsc_fini(struct intel_display *display)
+{
+	intel_hdcp_gsc_context_free(display->hdcp.gsc_context);
+	display->hdcp.gsc_context = NULL;
+	kfree(display->hdcp.arbiter);
+	display->hdcp.arbiter = NULL;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.h b/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.h
index 2d597f27e931..9f54157a4a3e 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.h
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc_message.h
@@ -6,68 +6,9 @@
 #ifndef __INTEL_HDCP_GSC_MESSAGE_H__
 #define __INTEL_HDCP_GSC_MESSAGE_H__
 
-#include <linux/types.h>
-
-struct device;
-struct drm_i915_private;
-struct hdcp_port_data;
-struct hdcp2_ake_init;
-struct hdcp2_ake_send_cert;
-struct hdcp2_ake_no_stored_km;
-struct hdcp2_ake_send_hprime;
-struct hdcp2_ake_send_pairing_info;
-struct hdcp2_lc_init;
-struct hdcp2_lc_send_lprime;
-struct hdcp2_ske_send_eks;
-struct hdcp2_rep_send_receiverid_list;
-struct hdcp2_rep_send_ack;
-struct hdcp2_rep_stream_ready;
 struct intel_display;
 
-ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in,
-				size_t msg_in_len, u8 *msg_out,
-				size_t msg_out_len);
-bool intel_hdcp_gsc_check_status(struct intel_display *display);
-int
-intel_hdcp_gsc_initiate_session(struct device *dev, struct hdcp_port_data *data,
-				struct hdcp2_ake_init *ake_data);
-int
-intel_hdcp_gsc_verify_receiver_cert_prepare_km(struct device *dev,
-					       struct hdcp_port_data *data,
-					       struct hdcp2_ake_send_cert *rx_cert,
-					       bool *km_stored,
-					       struct hdcp2_ake_no_stored_km
-					       *ek_pub_km,
-					       size_t *msg_sz);
-int
-intel_hdcp_gsc_verify_hprime(struct device *dev, struct hdcp_port_data *data,
-			     struct hdcp2_ake_send_hprime *rx_hprime);
-int
-intel_hdcp_gsc_store_pairing_info(struct device *dev, struct hdcp_port_data *data,
-				  struct hdcp2_ake_send_pairing_info *pairing_info);
-int
-intel_hdcp_gsc_initiate_locality_check(struct device *dev,
-				       struct hdcp_port_data *data,
-				       struct hdcp2_lc_init *lc_init_data);
-int
-intel_hdcp_gsc_verify_lprime(struct device *dev, struct hdcp_port_data *data,
-			     struct hdcp2_lc_send_lprime *rx_lprime);
-int intel_hdcp_gsc_get_session_key(struct device *dev,
-				   struct hdcp_port_data *data,
-				   struct hdcp2_ske_send_eks *ske_data);
-int
-intel_hdcp_gsc_repeater_check_flow_prepare_ack(struct device *dev,
-					       struct hdcp_port_data *data,
-					       struct hdcp2_rep_send_receiverid_list
-					       *rep_topology,
-					       struct hdcp2_rep_send_ack
-					       *rep_send_ack);
-int intel_hdcp_gsc_verify_mprime(struct device *dev,
-				 struct hdcp_port_data *data,
-				 struct hdcp2_rep_stream_ready *stream_ready);
-int intel_hdcp_gsc_enable_authentication(struct device *dev,
-					 struct hdcp_port_data *data);
-int
-intel_hdcp_gsc_close_session(struct device *dev, struct hdcp_port_data *data);
+int intel_hdcp_gsc_init(struct intel_display *display);
+void intel_hdcp_gsc_fini(struct intel_display *display);
 
 #endif /* __INTEL_HDCP_GSC_MESSAGE_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 33b8d5229db0..98033471902c 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -38,14 +38,15 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/intel/intel_lpe_audio.h>
 
 #include <media/cec-notifier.h>
 
 #include "g4x_hdmi.h"
-#include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
 #include "intel_connector.h"
@@ -64,6 +65,7 @@
 #include "intel_panel.h"
 #include "intel_pfit.h"
 #include "intel_snps_phy.h"
+#include "intel_vrr.h"
 
 static void
 assert_hdmi_port_disabled(struct intel_hdmi *intel_hdmi)
@@ -714,7 +716,7 @@ intel_hdmi_compute_avi_infoframe(struct intel_encoder *encoder,
 	struct hdmi_avi_infoframe *frame = &crtc_state->infoframes.avi.avi;
 	const struct drm_display_mode *adjusted_mode =
 		&crtc_state->hw.adjusted_mode;
-	struct drm_connector *connector = conn_state->connector;
+	struct intel_connector *connector = to_intel_connector(conn_state->connector);
 	int ret;
 
 	if (!crtc_state->has_infoframe)
@@ -723,7 +725,7 @@ intel_hdmi_compute_avi_infoframe(struct intel_encoder *encoder,
 	crtc_state->infoframes.enable |=
 		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_AVI);
 
-	ret = drm_hdmi_avi_infoframe_from_display_mode(frame, connector,
+	ret = drm_hdmi_avi_infoframe_from_display_mode(frame, &connector->base,
 						       adjusted_mode);
 	if (ret)
 		return false;
@@ -742,7 +744,7 @@ intel_hdmi_compute_avi_infoframe(struct intel_encoder *encoder,
 		    crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB);
 
 	if (crtc_state->output_format == INTEL_OUTPUT_FORMAT_RGB) {
-		drm_hdmi_avi_infoframe_quant_range(frame, connector,
+		drm_hdmi_avi_infoframe_quant_range(frame, &connector->base,
 						   adjusted_mode,
 						   crtc_state->limited_color_range ?
 						   HDMI_QUANTIZATION_RANGE_LIMITED :
@@ -768,7 +770,7 @@ intel_hdmi_compute_spd_infoframe(struct intel_encoder *encoder,
 				 struct intel_crtc_state *crtc_state,
 				 struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct hdmi_spd_infoframe *frame = &crtc_state->infoframes.spd.spd;
 	int ret;
 
@@ -778,7 +780,7 @@ intel_hdmi_compute_spd_infoframe(struct intel_encoder *encoder,
 	crtc_state->infoframes.enable |=
 		intel_hdmi_infoframe_enable(HDMI_INFOFRAME_TYPE_SPD);
 
-	if (IS_DGFX(i915))
+	if (display->platform.dgfx)
 		ret = hdmi_spd_infoframe_init(frame, "Intel", "Discrete gfx");
 	else
 		ret = hdmi_spd_infoframe_init(frame, "Intel", "Integrated gfx");
@@ -978,7 +980,6 @@ static bool intel_hdmi_set_gcp_infoframe(struct intel_encoder *encoder,
 					 const struct drm_connector_state *conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	i915_reg_t reg;
 
@@ -988,9 +989,9 @@ static bool intel_hdmi_set_gcp_infoframe(struct intel_encoder *encoder,
 
 	if (HAS_DDI(display))
 		reg = HSW_TVIDEO_DIP_GCP(display, crtc_state->cpu_transcoder);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+	else if (display->platform.valleyview || display->platform.cherryview)
 		reg = VLV_TVIDEO_DIP_GCP(crtc->pipe);
-	else if (HAS_PCH_SPLIT(dev_priv))
+	else if (HAS_PCH_SPLIT(display))
 		reg = TVIDEO_DIP_GCP(crtc->pipe);
 	else
 		return false;
@@ -1004,7 +1005,6 @@ void intel_hdmi_read_gcp_infoframe(struct intel_encoder *encoder,
 				   struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	i915_reg_t reg;
 
@@ -1014,9 +1014,9 @@ void intel_hdmi_read_gcp_infoframe(struct intel_encoder *encoder,
 
 	if (HAS_DDI(display))
 		reg = HSW_TVIDEO_DIP_GCP(display, crtc_state->cpu_transcoder);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+	else if (display->platform.valleyview || display->platform.cherryview)
 		reg = VLV_TVIDEO_DIP_GCP(crtc->pipe);
-	else if (HAS_PCH_SPLIT(dev_priv))
+	else if (HAS_PCH_SPLIT(display))
 		reg = TVIDEO_DIP_GCP(crtc->pipe);
 	else
 		return;
@@ -1028,9 +1028,9 @@ static void intel_hdmi_compute_gcp_infoframe(struct intel_encoder *encoder,
 					     struct intel_crtc_state *crtc_state,
 					     struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (IS_G4X(dev_priv) || !crtc_state->has_infoframe)
+	if (display->platform.g4x || !crtc_state->has_infoframe)
 		return;
 
 	crtc_state->infoframes.enable |=
@@ -1538,7 +1538,6 @@ int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
 	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_hdmi *hdmi = &dig_port->hdmi;
 	struct intel_connector *connector = hdmi->attached_connector;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	int ret;
 
 	if (!enable)
@@ -1557,7 +1556,7 @@ int intel_hdmi_hdcp_toggle_signalling(struct intel_digital_port *dig_port,
 	 * WA: To fix incorrect positioning of the window of
 	 * opportunity and enc_en signalling in KABYLAKE.
 	 */
-	if (IS_KABYLAKE(dev_priv) && enable)
+	if (display->platform.kabylake && enable)
 		return kbl_repositioning_enc_en_signal(connector,
 						       cpu_transcoder);
 
@@ -1569,7 +1568,6 @@ bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port,
 				     struct intel_connector *connector)
 {
 	struct intel_display *display = to_intel_display(dig_port);
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder = connector->hdcp.cpu_transcoder;
 	int ret;
@@ -1582,15 +1580,15 @@ bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port,
 	if (ret)
 		return false;
 
-	intel_de_write(i915, HDCP_RPRIME(i915, cpu_transcoder, port), ri.reg);
+	intel_de_write(display, HDCP_RPRIME(display, cpu_transcoder, port), ri.reg);
 
 	/* Wait for Ri prime match */
-	if (wait_for((intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) &
+	if (wait_for((intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) &
 		      (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC)) ==
 		     (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1)) {
 		drm_dbg_kms(display->drm, "Ri' mismatch detected (%x)\n",
-			    intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder,
-							    port)));
+			    intel_de_read(display, HDCP_STATUS(display, cpu_transcoder,
+							       port)));
 		return false;
 	}
 	return true;
@@ -1813,14 +1811,13 @@ static const struct intel_hdcp_shim intel_hdmi_hdcp_shim = {
 static int intel_hdmi_source_max_tmds_clock(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	int max_tmds_clock, vbt_max_tmds_clock;
 
-	if (DISPLAY_VER(display) >= 13 || IS_ALDERLAKE_S(dev_priv))
+	if (DISPLAY_VER(display) >= 13 || display->platform.alderlake_s)
 		max_tmds_clock = 600000;
 	else if (DISPLAY_VER(display) >= 10)
 		max_tmds_clock = 594000;
-	else if (DISPLAY_VER(display) >= 8 || IS_HASWELL(dev_priv))
+	else if (DISPLAY_VER(display) >= 8 || display->platform.haswell)
 		max_tmds_clock = 300000;
 	else if (DISPLAY_VER(display) >= 5)
 		max_tmds_clock = 225000;
@@ -1879,7 +1876,6 @@ hdmi_port_clock_valid(struct intel_hdmi *hdmi,
 		      bool has_hdmi_sink)
 {
 	struct intel_display *display = to_intel_display(hdmi);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_encoder *encoder = &hdmi_to_dig_port(hdmi)->base;
 
 	if (clock < 25000)
@@ -1889,16 +1885,16 @@ hdmi_port_clock_valid(struct intel_hdmi *hdmi,
 		return MODE_CLOCK_HIGH;
 
 	/* GLK DPLL can't generate 446-480 MHz */
-	if (IS_GEMINILAKE(dev_priv) && clock > 446666 && clock < 480000)
+	if (display->platform.geminilake && clock > 446666 && clock < 480000)
 		return MODE_CLOCK_RANGE;
 
 	/* BXT/GLK DPLL can't generate 223-240 MHz */
-	if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
+	if ((display->platform.geminilake || display->platform.broxton) &&
 	    clock > 223333 && clock < 240000)
 		return MODE_CLOCK_RANGE;
 
 	/* CHV DPLL can't generate 216-240 MHz */
-	if (IS_CHERRYVIEW(dev_priv) && clock > 216000 && clock < 240000)
+	if (display->platform.cherryview && clock > 216000 && clock < 240000)
 		return MODE_CLOCK_RANGE;
 
 	/* ICL+ combo PHY PLL can't generate 500-533.2 MHz */
@@ -1942,11 +1938,12 @@ static bool intel_hdmi_source_bpc_possible(struct intel_display *display, int bp
 	}
 }
 
-static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector,
+static bool intel_hdmi_sink_bpc_possible(struct drm_connector *_connector,
 					 int bpc, bool has_hdmi_sink,
 					 enum intel_output_format sink_format)
 {
-	const struct drm_display_info *info = &connector->display_info;
+	struct intel_connector *connector = to_intel_connector(_connector);
+	const struct drm_display_info *info = &connector->base.display_info;
 	const struct drm_hdmi_info *hdmi = &info->hdmi;
 
 	switch (bpc) {
@@ -1975,12 +1972,13 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector,
 }
 
 static enum drm_mode_status
-intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock,
+intel_hdmi_mode_clock_valid(struct drm_connector *_connector, int clock,
 			    bool has_hdmi_sink,
 			    enum intel_output_format sink_format)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_hdmi *hdmi = intel_attached_hdmi(connector);
 	enum drm_mode_status status = MODE_OK;
 	int bpc;
 
@@ -1995,7 +1993,8 @@ intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock,
 		if (!intel_hdmi_source_bpc_possible(display, bpc))
 			continue;
 
-		if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, sink_format))
+		if (!intel_hdmi_sink_bpc_possible(&connector->base, bpc, has_hdmi_sink,
+						  sink_format))
 			continue;
 
 		status = hdmi_port_clock_valid(hdmi, tmds_clock, true, has_hdmi_sink);
@@ -2010,15 +2009,16 @@ intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock,
 }
 
 static enum drm_mode_status
-intel_hdmi_mode_valid(struct drm_connector *connector,
+intel_hdmi_mode_valid(struct drm_connector *_connector,
 		      const struct drm_display_mode *mode)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_hdmi *hdmi = intel_attached_hdmi(connector);
 	enum drm_mode_status status;
 	int clock = mode->clock;
-	int max_dotclk = to_i915(connector->dev)->display.cdclk.max_dotclk_freq;
-	bool has_hdmi_sink = intel_has_hdmi_sink(hdmi, connector->state);
+	int max_dotclk = display->cdclk.max_dotclk_freq;
+	bool has_hdmi_sink = intel_has_hdmi_sink(hdmi, connector->base.state);
 	bool ycbcr_420_only;
 	enum intel_output_format sink_format;
 
@@ -2047,22 +2047,23 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 	if (clock > 600000)
 		return MODE_CLOCK_HIGH;
 
-	ycbcr_420_only = drm_mode_is_420_only(&connector->display_info, mode);
+	ycbcr_420_only = drm_mode_is_420_only(&connector->base.display_info, mode);
 
 	if (ycbcr_420_only)
 		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
 	else
 		sink_format = INTEL_OUTPUT_FORMAT_RGB;
 
-	status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, sink_format);
+	status = intel_hdmi_mode_clock_valid(&connector->base, clock, has_hdmi_sink, sink_format);
 	if (status != MODE_OK) {
 		if (ycbcr_420_only ||
-		    !connector->ycbcr_420_allowed ||
-		    !drm_mode_is_420_also(&connector->display_info, mode))
+		    !connector->base.ycbcr_420_allowed ||
+		    !drm_mode_is_420_also(&connector->base.display_info, mode))
 			return status;
 
 		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
-		status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, sink_format);
+		status = intel_hdmi_mode_clock_valid(&connector->base, clock, has_hdmi_sink,
+						     sink_format);
 		if (status != MODE_OK)
 			return status;
 	}
@@ -2073,16 +2074,16 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state,
 			     int bpc, bool has_hdmi_sink)
 {
-	struct drm_atomic_state *state = crtc_state->uapi.state;
-	struct drm_connector_state *connector_state;
-	struct drm_connector *connector;
+	struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state);
+	struct intel_digital_connector_state *connector_state;
+	struct intel_connector *connector;
 	int i;
 
-	for_each_new_connector_in_state(state, connector, connector_state, i) {
-		if (connector_state->crtc != crtc_state->uapi.crtc)
+	for_each_new_intel_connector_in_state(state, connector, connector_state, i) {
+		if (connector_state->base.crtc != crtc_state->uapi.crtc)
 			continue;
 
-		if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink,
+		if (!intel_hdmi_sink_bpc_possible(&connector->base, bpc, has_hdmi_sink,
 						  crtc_state->sink_format))
 			return false;
 	}
@@ -2210,7 +2211,7 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state,
 				 const struct drm_connector_state *conn_state)
 {
-	struct drm_connector *connector = conn_state->connector;
+	struct intel_connector *connector = to_intel_connector(conn_state->connector);
 	const struct intel_digital_connector_state *intel_conn_state =
 		to_intel_digital_connector_state(conn_state);
 
@@ -2218,7 +2219,7 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder,
 		return false;
 
 	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
-		return connector->display_info.has_audio;
+		return connector->base.display_info.has_audio;
 	else
 		return intel_conn_state->force_audio == HDMI_AUDIO_ON;
 }
@@ -2322,14 +2323,14 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 {
 	struct intel_display *display = to_intel_display(encoder);
 	struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
-	struct drm_connector *connector = conn_state->connector;
-	struct drm_scdc *scdc = &connector->display_info.hdmi.scdc;
+	struct intel_connector *connector = to_intel_connector(conn_state->connector);
+	struct drm_scdc *scdc = &connector->base.display_info.hdmi.scdc;
 	int ret;
 
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return -EINVAL;
 
-	if (!connector->interlace_allowed &&
+	if (!connector->base.interlace_allowed &&
 	    adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE)
 		return -EINVAL;
 
@@ -2384,6 +2385,8 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 		}
 	}
 
+	intel_vrr_compute_config(pipe_config, conn_state);
+
 	intel_hdmi_compute_gcp_infoframe(encoder, pipe_config,
 					 conn_state);
 
@@ -2422,25 +2425,26 @@ void intel_hdmi_encoder_shutdown(struct intel_encoder *encoder)
 }
 
 static void
-intel_hdmi_unset_edid(struct drm_connector *connector)
+intel_hdmi_unset_edid(struct drm_connector *_connector)
 {
-	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 
 	intel_hdmi->dp_dual_mode.type = DRM_DP_DUAL_MODE_NONE;
 	intel_hdmi->dp_dual_mode.max_tmds_clock = 0;
 
-	drm_edid_free(to_intel_connector(connector)->detect_edid);
-	to_intel_connector(connector)->detect_edid = NULL;
+	drm_edid_free(connector->detect_edid);
+	connector->detect_edid = NULL;
 }
 
 static void
-intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
+intel_hdmi_dp_dual_mode_detect(struct drm_connector *_connector)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
-	struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector));
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_hdmi *hdmi = intel_attached_hdmi(connector);
 	struct intel_encoder *encoder = &hdmi_to_dig_port(hdmi)->base;
-	struct i2c_adapter *ddc = connector->ddc;
+	struct i2c_adapter *ddc = connector->base.ddc;
 	enum drm_dp_dual_mode_type type;
 
 	type = drm_dp_dual_mode_detect(display->drm, ddc);
@@ -2455,7 +2459,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
 	 * if the port is a dual mode capable DP port.
 	 */
 	if (type == DRM_DP_DUAL_MODE_UNKNOWN) {
-		if (!connector->force &&
+		if (!connector->base.force &&
 		    intel_bios_encoder_supports_dp_dual_mode(encoder->devdata)) {
 			drm_dbg_kms(display->drm,
 				    "Assuming DP dual mode adaptor presence based on VBT\n");
@@ -2478,7 +2482,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
 		    hdmi->dp_dual_mode.max_tmds_clock);
 
 	/* Older VBTs are often buggy and can't be trusted :( Play it safe. */
-	if ((DISPLAY_VER(display) >= 8 || IS_HASWELL(dev_priv)) &&
+	if ((DISPLAY_VER(display) >= 8 || display->platform.haswell) &&
 	    !intel_bios_encoder_supports_dp_dual_mode(encoder->devdata)) {
 		drm_dbg_kms(display->drm,
 			    "Ignoring DP dual mode adaptor max TMDS clock for native HDMI port\n");
@@ -2487,34 +2491,35 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
 }
 
 static bool
-intel_hdmi_set_edid(struct drm_connector *connector)
+intel_hdmi_set_edid(struct drm_connector *_connector)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
-	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(to_intel_connector(connector));
-	struct i2c_adapter *ddc = connector->ddc;
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
+	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
+	struct i2c_adapter *ddc = connector->base.ddc;
 	intel_wakeref_t wakeref;
 	const struct drm_edid *drm_edid;
 	bool connected = false;
 
 	wakeref = intel_display_power_get(display, POWER_DOMAIN_GMBUS);
 
-	drm_edid = drm_edid_read_ddc(connector, ddc);
+	drm_edid = drm_edid_read_ddc(&connector->base, ddc);
 
 	if (!drm_edid && !intel_gmbus_is_forced_bit(ddc)) {
 		drm_dbg_kms(display->drm,
 			    "HDMI GMBUS EDID read failed, retry using GPIO bit-banging\n");
 		intel_gmbus_force_bit(ddc, true);
-		drm_edid = drm_edid_read_ddc(connector, ddc);
+		drm_edid = drm_edid_read_ddc(&connector->base, ddc);
 		intel_gmbus_force_bit(ddc, false);
 	}
 
 	/* Below we depend on display info having been updated */
-	drm_edid_connector_update(connector, drm_edid);
+	drm_edid_connector_update(&connector->base, drm_edid);
 
-	to_intel_connector(connector)->detect_edid = drm_edid;
+	connector->detect_edid = drm_edid;
 
 	if (drm_edid_is_digital(drm_edid)) {
-		intel_hdmi_dp_dual_mode_detect(connector);
+		intel_hdmi_dp_dual_mode_detect(&connector->base);
 
 		connected = true;
 	}
@@ -2522,28 +2527,29 @@ intel_hdmi_set_edid(struct drm_connector *connector)
 	intel_display_power_put(display, POWER_DOMAIN_GMBUS, wakeref);
 
 	cec_notifier_set_phys_addr(intel_hdmi->cec_notifier,
-				   connector->display_info.source_physical_address);
+				   connector->base.display_info.source_physical_address);
 
 	return connected;
 }
 
 static enum drm_connector_status
-intel_hdmi_detect(struct drm_connector *connector, bool force)
+intel_hdmi_detect(struct drm_connector *_connector, bool force)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
 	enum drm_connector_status status = connector_status_disconnected;
-	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(to_intel_connector(connector));
+	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct intel_encoder *encoder = &hdmi_to_dig_port(intel_hdmi)->base;
 	intel_wakeref_t wakeref;
 
 	drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s]\n",
-		    connector->base.id, connector->name);
+		    connector->base.base.id, connector->base.name);
 
 	if (!intel_display_device_enabled(display))
 		return connector_status_disconnected;
 
 	if (!intel_display_driver_check_access(display))
-		return connector->status;
+		return connector->base.status;
 
 	wakeref = intel_display_power_get(display, POWER_DOMAIN_GMBUS);
 
@@ -2551,9 +2557,9 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	    !intel_digital_port_connected(encoder))
 		goto out;
 
-	intel_hdmi_unset_edid(connector);
+	intel_hdmi_unset_edid(&connector->base);
 
-	if (intel_hdmi_set_edid(connector))
+	if (intel_hdmi_set_edid(&connector->base))
 		status = connector_status_connected;
 
 out:
@@ -2566,49 +2572,54 @@ out:
 }
 
 static void
-intel_hdmi_force(struct drm_connector *connector)
+intel_hdmi_force(struct drm_connector *_connector)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
 
 	drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s]\n",
-		    connector->base.id, connector->name);
+		    connector->base.base.id, connector->base.name);
 
 	if (!intel_display_driver_check_access(display))
 		return;
 
-	intel_hdmi_unset_edid(connector);
+	intel_hdmi_unset_edid(&connector->base);
 
-	if (connector->status != connector_status_connected)
+	if (connector->base.status != connector_status_connected)
 		return;
 
-	intel_hdmi_set_edid(connector);
+	intel_hdmi_set_edid(&connector->base);
 }
 
-static int intel_hdmi_get_modes(struct drm_connector *connector)
+static int intel_hdmi_get_modes(struct drm_connector *_connector)
 {
+	struct intel_connector *connector = to_intel_connector(_connector);
+
 	/* drm_edid_connector_update() done in ->detect() or ->force() */
-	return drm_edid_connector_add_modes(connector);
+	return drm_edid_connector_add_modes(&connector->base);
 }
 
 static int
-intel_hdmi_connector_register(struct drm_connector *connector)
+intel_hdmi_connector_register(struct drm_connector *_connector)
 {
+	struct intel_connector *connector = to_intel_connector(_connector);
 	int ret;
 
-	ret = intel_connector_register(connector);
+	ret = intel_connector_register(&connector->base);
 	if (ret)
 		return ret;
 
 	return ret;
 }
 
-static void intel_hdmi_connector_unregister(struct drm_connector *connector)
+static void intel_hdmi_connector_unregister(struct drm_connector *_connector)
 {
-	struct cec_notifier *n = intel_attached_hdmi(to_intel_connector(connector))->cec_notifier;
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct cec_notifier *n = intel_attached_hdmi(connector)->cec_notifier;
 
 	cec_notifier_conn_unregister(n);
 
-	intel_connector_unregister(connector);
+	intel_connector_unregister(&connector->base);
 }
 
 static const struct drm_connector_funcs intel_hdmi_connector_funcs = {
@@ -2624,15 +2635,16 @@ static const struct drm_connector_funcs intel_hdmi_connector_funcs = {
 	.atomic_duplicate_state = intel_digital_connector_duplicate_state,
 };
 
-static int intel_hdmi_connector_atomic_check(struct drm_connector *connector,
+static int intel_hdmi_connector_atomic_check(struct drm_connector *_connector,
 					     struct drm_atomic_state *state)
 {
-	struct intel_display *display = to_intel_display(connector->dev);
+	struct intel_connector *connector = to_intel_connector(_connector);
+	struct intel_display *display = to_intel_display(connector);
 
 	if (HAS_DDI(display))
-		return intel_digital_connector_atomic_check(connector, state);
+		return intel_digital_connector_atomic_check(&connector->base, state);
 	else
-		return g4x_hdmi_connector_atomic_check(connector, state);
+		return g4x_hdmi_connector_atomic_check(&connector->base, state);
 }
 
 static const struct drm_connector_helper_funcs intel_hdmi_connector_helper_funcs = {
@@ -2642,22 +2654,23 @@ static const struct drm_connector_helper_funcs intel_hdmi_connector_helper_funcs
 };
 
 static void
-intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *connector)
+intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *_connector)
 {
+	struct intel_connector *connector = to_intel_connector(_connector);
 	struct intel_display *display = to_intel_display(intel_hdmi);
 
-	intel_attach_force_audio_property(connector);
-	intel_attach_broadcast_rgb_property(connector);
-	intel_attach_aspect_ratio_property(connector);
+	intel_attach_force_audio_property(&connector->base);
+	intel_attach_broadcast_rgb_property(&connector->base);
+	intel_attach_aspect_ratio_property(&connector->base);
 
-	intel_attach_hdmi_colorspace_property(connector);
-	drm_connector_attach_content_type_property(connector);
+	intel_attach_hdmi_colorspace_property(&connector->base);
+	drm_connector_attach_content_type_property(&connector->base);
 
 	if (DISPLAY_VER(display) >= 10)
-		drm_connector_attach_hdr_output_metadata_property(connector);
+		drm_connector_attach_hdr_output_metadata_property(&connector->base);
 
 	if (!HAS_GMCH(display))
-		drm_connector_attach_max_bpc_property(connector, 8, 12);
+		drm_connector_attach_max_bpc_property(&connector->base, 8, 12);
 }
 
 /*
@@ -2679,25 +2692,26 @@ intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *c
  * True on success, false on failure.
  */
 bool intel_hdmi_handle_sink_scrambling(struct intel_encoder *encoder,
-				       struct drm_connector *connector,
+				       struct drm_connector *_connector,
 				       bool high_tmds_clock_ratio,
 				       bool scrambling)
 {
+	struct intel_connector *connector = to_intel_connector(_connector);
 	struct intel_display *display = to_intel_display(encoder);
 	struct drm_scrambling *sink_scrambling =
-		&connector->display_info.hdmi.scdc.scrambling;
+		&connector->base.display_info.hdmi.scdc.scrambling;
 
 	if (!sink_scrambling->supported)
 		return true;
 
 	drm_dbg_kms(display->drm,
 		    "[CONNECTOR:%d:%s] scrambling=%s, TMDS bit clock ratio=1/%d\n",
-		    connector->base.id, connector->name,
+		    connector->base.base.id, connector->base.name,
 		    str_yes_no(scrambling), high_tmds_clock_ratio ? 40 : 10);
 
 	/* Set TMDS bit clock ratio to 1/40 or 1/10, and enable/disable scrambling */
-	return drm_scdc_set_high_tmds_clock_ratio(connector, high_tmds_clock_ratio) &&
-		drm_scdc_set_scrambling(connector, scrambling);
+	return drm_scdc_set_high_tmds_clock_ratio(&connector->base, high_tmds_clock_ratio) &&
+		drm_scdc_set_scrambling(&connector->base, scrambling);
 }
 
 static u8 chv_encoder_to_ddc_pin(struct intel_encoder *encoder)
@@ -2808,7 +2822,7 @@ static u8 mcc_encoder_to_ddc_pin(struct intel_encoder *encoder)
 
 static u8 rkl_encoder_to_ddc_pin(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
 	WARN_ON(encoder->port == PORT_C);
@@ -2819,7 +2833,7 @@ static u8 rkl_encoder_to_ddc_pin(struct intel_encoder *encoder)
 	 * combo outputs.  With CMP, the traditional DDI A-D pins are used for
 	 * all outputs.
 	 */
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_TGP && phy >= PHY_C)
+	if (INTEL_PCH_TYPE(display) >= PCH_TGP && phy >= PHY_C)
 		return GMBUS_PIN_9_TC1_ICP + phy - PHY_C;
 
 	return GMBUS_PIN_1_BXT + phy;
@@ -2828,7 +2842,6 @@ static u8 rkl_encoder_to_ddc_pin(struct intel_encoder *encoder)
 static u8 gen9bc_tgp_encoder_to_ddc_pin(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_encoder_to_phy(encoder);
 
 	drm_WARN_ON(display->drm, encoder->port == PORT_A);
@@ -2839,7 +2852,7 @@ static u8 gen9bc_tgp_encoder_to_ddc_pin(struct intel_encoder *encoder)
 	 * combo outputs.  With CMP, the traditional DDI A-D pins are used for
 	 * all outputs.
 	 */
-	if (INTEL_PCH_TYPE(i915) >= PCH_TGP && phy >= PHY_C)
+	if (INTEL_PCH_TYPE(display) >= PCH_TGP && phy >= PHY_C)
 		return GMBUS_PIN_9_TC1_ICP + phy - PHY_C;
 
 	return GMBUS_PIN_1_BXT + phy;
@@ -2892,27 +2905,26 @@ static u8 g4x_encoder_to_ddc_pin(struct intel_encoder *encoder)
 static u8 intel_hdmi_default_ddc_pin(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u8 ddc_pin;
 
-	if (IS_ALDERLAKE_S(dev_priv))
+	if (display->platform.alderlake_s)
 		ddc_pin = adls_encoder_to_ddc_pin(encoder);
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1)
+	else if (INTEL_PCH_TYPE(display) >= PCH_DG1)
 		ddc_pin = dg1_encoder_to_ddc_pin(encoder);
-	else if (IS_ROCKETLAKE(dev_priv))
+	else if (display->platform.rocketlake)
 		ddc_pin = rkl_encoder_to_ddc_pin(encoder);
-	else if (DISPLAY_VER(display) == 9 && HAS_PCH_TGP(dev_priv))
+	else if (DISPLAY_VER(display) == 9 && HAS_PCH_TGP(display))
 		ddc_pin = gen9bc_tgp_encoder_to_ddc_pin(encoder);
-	else if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) &&
-		 HAS_PCH_TGP(dev_priv))
+	else if ((display->platform.jasperlake || display->platform.elkhartlake) &&
+		 HAS_PCH_TGP(display))
 		ddc_pin = mcc_encoder_to_ddc_pin(encoder);
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+	else if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 		ddc_pin = icl_encoder_to_ddc_pin(encoder);
-	else if (HAS_PCH_CNP(dev_priv))
+	else if (HAS_PCH_CNP(display))
 		ddc_pin = cnp_encoder_to_ddc_pin(encoder);
-	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	else if (display->platform.geminilake || display->platform.broxton)
 		ddc_pin = bxt_encoder_to_ddc_pin(encoder);
-	else if (IS_CHERRYVIEW(dev_priv))
+	else if (display->platform.cherryview)
 		ddc_pin = chv_encoder_to_ddc_pin(encoder);
 	else
 		ddc_pin = g4x_encoder_to_ddc_pin(encoder);
@@ -2986,15 +2998,13 @@ static u8 intel_hdmi_ddc_pin(struct intel_encoder *encoder)
 void intel_infoframe_init(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
-	struct drm_i915_private *dev_priv =
-		to_i915(dig_port->base.base.dev);
 
-	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
+	if (display->platform.valleyview || display->platform.cherryview) {
 		dig_port->write_infoframe = vlv_write_infoframe;
 		dig_port->read_infoframe = vlv_read_infoframe;
 		dig_port->set_infoframes = vlv_set_infoframes;
 		dig_port->infoframes_enabled = vlv_infoframes_enabled;
-	} else if (IS_G4X(dev_priv)) {
+	} else if (display->platform.g4x) {
 		dig_port->write_infoframe = g4x_write_infoframe;
 		dig_port->read_infoframe = g4x_read_infoframe;
 		dig_port->set_infoframes = g4x_set_infoframes;
@@ -3011,7 +3021,7 @@ void intel_infoframe_init(struct intel_digital_port *dig_port)
 			dig_port->set_infoframes = hsw_set_infoframes;
 			dig_port->infoframes_enabled = hsw_infoframes_enabled;
 		}
-	} else if (HAS_PCH_IBX(dev_priv)) {
+	} else if (HAS_PCH_IBX(display)) {
 		dig_port->write_infoframe = ibx_write_infoframe;
 		dig_port->read_infoframe = ibx_read_infoframe;
 		dig_port->set_infoframes = ibx_set_infoframes;
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index 00d7b1ccf190..fc5d8928c37e 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -30,6 +30,7 @@
 #include "i915_irq.h"
 #include "intel_connector.h"
 #include "intel_display_power.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_hdcp.h"
 #include "intel_hotplug.h"
@@ -118,7 +119,7 @@ intel_connector_hpd_pin(struct intel_connector *connector)
 
 /**
  * intel_hpd_irq_storm_detect - gather stats and detect HPD IRQ storm on a pin
- * @dev_priv: private driver data pointer
+ * @display: display device
  * @pin: the pin to gather stats on
  * @long_hpd: whether the HPD IRQ was long or short
  *
@@ -127,13 +128,13 @@ intel_connector_hpd_pin(struct intel_connector *connector)
  * responsible for further action.
  *
  * The number of IRQs that are allowed within @HPD_STORM_DETECT_PERIOD is
- * stored in @dev_priv->display.hotplug.hpd_storm_threshold which defaults to
+ * stored in @display->hotplug.hpd_storm_threshold which defaults to
  * @HPD_STORM_DEFAULT_THRESHOLD. Long IRQs count as +10 to this threshold, and
  * short IRQs count as +1. If this threshold is exceeded, it's considered an
  * IRQ storm and the IRQ state is set to @HPD_MARK_DISABLED.
  *
  * By default, most systems will only count long IRQs towards
- * &dev_priv->display.hotplug.hpd_storm_threshold. However, some older systems also
+ * &display->hotplug.hpd_storm_threshold. However, some older systems also
  * suffer from short IRQ storms and must also track these. Because short IRQ
  * storms are naturally caused by sideband interactions with DP MST devices,
  * short IRQ detection is only enabled for systems without DP MST support.
@@ -145,10 +146,10 @@ intel_connector_hpd_pin(struct intel_connector *connector)
  *
  * Return true if an IRQ storm was detected on @pin.
  */
-static bool intel_hpd_irq_storm_detect(struct drm_i915_private *dev_priv,
+static bool intel_hpd_irq_storm_detect(struct intel_display *display,
 				       enum hpd_pin pin, bool long_hpd)
 {
-	struct intel_hotplug *hpd = &dev_priv->display.hotplug;
+	struct intel_hotplug *hpd = &display->hotplug;
 	unsigned long start = hpd->stats[pin].last_jiffies;
 	unsigned long end = start + msecs_to_jiffies(HPD_STORM_DETECT_PERIOD);
 	const int increment = long_hpd ? 10 : 1;
@@ -156,7 +157,7 @@ static bool intel_hpd_irq_storm_detect(struct drm_i915_private *dev_priv,
 	bool storm = false;
 
 	if (!threshold ||
-	    (!long_hpd && !dev_priv->display.hotplug.hpd_short_storm_enabled))
+	    (!long_hpd && !display->hotplug.hpd_short_storm_enabled))
 		return false;
 
 	if (!time_in_range(jiffies, start, end)) {
@@ -167,11 +168,11 @@ static bool intel_hpd_irq_storm_detect(struct drm_i915_private *dev_priv,
 	hpd->stats[pin].count += increment;
 	if (hpd->stats[pin].count > threshold) {
 		hpd->stats[pin].state = HPD_MARK_DISABLED;
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "HPD interrupt storm detected on PIN %d\n", pin);
 		storm = true;
 	} else {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Received HPD interrupt on PIN %d - cnt: %d\n",
 			      pin,
 			      hpd->stats[pin].count);
@@ -180,56 +181,62 @@ static bool intel_hpd_irq_storm_detect(struct drm_i915_private *dev_priv,
 	return storm;
 }
 
-static bool detection_work_enabled(struct drm_i915_private *i915)
+static bool detection_work_enabled(struct intel_display *display)
 {
-	lockdep_assert_held(&i915->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
-	return i915->display.hotplug.detection_work_enabled;
+	return display->hotplug.detection_work_enabled;
 }
 
 static bool
-mod_delayed_detection_work(struct drm_i915_private *i915, struct delayed_work *work, int delay)
+mod_delayed_detection_work(struct intel_display *display, struct delayed_work *work, int delay)
 {
-	lockdep_assert_held(&i915->irq_lock);
+	struct drm_i915_private *i915 = to_i915(display->drm);
 
-	if (!detection_work_enabled(i915))
+	lockdep_assert_held(&display->irq.lock);
+
+	if (!detection_work_enabled(display))
 		return false;
 
 	return mod_delayed_work(i915->unordered_wq, work, delay);
 }
 
 static bool
-queue_delayed_detection_work(struct drm_i915_private *i915, struct delayed_work *work, int delay)
+queue_delayed_detection_work(struct intel_display *display, struct delayed_work *work, int delay)
 {
-	lockdep_assert_held(&i915->irq_lock);
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
+	lockdep_assert_held(&display->irq.lock);
 
-	if (!detection_work_enabled(i915))
+	if (!detection_work_enabled(display))
 		return false;
 
 	return queue_delayed_work(i915->unordered_wq, work, delay);
 }
 
 static bool
-queue_detection_work(struct drm_i915_private *i915, struct work_struct *work)
+queue_detection_work(struct intel_display *display, struct work_struct *work)
 {
-	lockdep_assert_held(&i915->irq_lock);
+	struct drm_i915_private *i915 = to_i915(display->drm);
 
-	if (!detection_work_enabled(i915))
+	lockdep_assert_held(&display->irq.lock);
+
+	if (!detection_work_enabled(display))
 		return false;
 
 	return queue_work(i915->unordered_wq, work);
 }
 
 static void
-intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv)
+intel_hpd_irq_storm_switch_to_polling(struct intel_display *display)
 {
 	struct drm_connector_list_iter conn_iter;
 	struct intel_connector *connector;
 	bool hpd_disabled = false;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
-	drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		enum hpd_pin pin;
 
@@ -238,15 +245,15 @@ intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv)
 
 		pin = intel_connector_hpd_pin(connector);
 		if (pin == HPD_NONE ||
-		    dev_priv->display.hotplug.stats[pin].state != HPD_MARK_DISABLED)
+		    display->hotplug.stats[pin].state != HPD_MARK_DISABLED)
 			continue;
 
-		drm_info(&dev_priv->drm,
+		drm_info(display->drm,
 			 "HPD interrupt storm detected on connector %s: "
 			 "switching from hotplug detection to polling\n",
 			 connector->base.name);
 
-		dev_priv->display.hotplug.stats[pin].state = HPD_DISABLED;
+		display->hotplug.stats[pin].state = HPD_DISABLED;
 		connector->base.polled = DRM_CONNECTOR_POLL_CONNECT |
 			DRM_CONNECTOR_POLL_DISCONNECT;
 		hpd_disabled = true;
@@ -255,36 +262,35 @@ intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv)
 
 	/* Enable polling and queue hotplug re-enabling. */
 	if (hpd_disabled) {
-		drm_kms_helper_poll_reschedule(&dev_priv->drm);
-		mod_delayed_detection_work(dev_priv,
-					   &dev_priv->display.hotplug.reenable_work,
+		drm_kms_helper_poll_reschedule(display->drm);
+		mod_delayed_detection_work(display,
+					   &display->hotplug.reenable_work,
 					   msecs_to_jiffies(HPD_STORM_REENABLE_DELAY));
 	}
 }
 
 static void intel_hpd_irq_storm_reenable_work(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, typeof(*dev_priv),
-			     display.hotplug.reenable_work.work);
+	struct intel_display *display =
+		container_of(work, typeof(*display), hotplug.reenable_work.work);
 	struct drm_connector_list_iter conn_iter;
 	struct intel_connector *connector;
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	enum hpd_pin pin;
 
-	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
-	drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		pin = intel_connector_hpd_pin(connector);
 		if (pin == HPD_NONE ||
-		    dev_priv->display.hotplug.stats[pin].state != HPD_DISABLED)
+		    display->hotplug.stats[pin].state != HPD_DISABLED)
 			continue;
 
 		if (connector->base.polled != connector->polled)
-			drm_dbg(&dev_priv->drm,
+			drm_dbg(display->drm,
 				"Reenabling HPD on connector %s\n",
 				connector->base.name);
 		connector->base.polled = connector->polled;
@@ -292,15 +298,15 @@ static void intel_hpd_irq_storm_reenable_work(struct work_struct *work)
 	drm_connector_list_iter_end(&conn_iter);
 
 	for_each_hpd_pin(pin) {
-		if (dev_priv->display.hotplug.stats[pin].state == HPD_DISABLED)
-			dev_priv->display.hotplug.stats[pin].state = HPD_ENABLED;
+		if (display->hotplug.stats[pin].state == HPD_DISABLED)
+			display->hotplug.stats[pin].state = HPD_ENABLED;
 	}
 
-	intel_hpd_irq_setup(dev_priv);
+	intel_hpd_irq_setup(display);
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 }
 
 static enum intel_hotplug_state
@@ -349,32 +355,72 @@ static bool intel_encoder_has_hpd_pulse(struct intel_encoder *encoder)
 		enc_to_dig_port(encoder)->hpd_pulse != NULL;
 }
 
+static bool hpd_pin_has_pulse(struct intel_display *display, enum hpd_pin pin)
+{
+	struct intel_encoder *encoder;
+
+	for_each_intel_encoder(display->drm, encoder) {
+		if (encoder->hpd_pin != pin)
+			continue;
+
+		if (intel_encoder_has_hpd_pulse(encoder))
+			return true;
+	}
+
+	return false;
+}
+
+static bool hpd_pin_is_blocked(struct intel_display *display, enum hpd_pin pin)
+{
+	lockdep_assert_held(&display->irq.lock);
+
+	return display->hotplug.stats[pin].blocked_count;
+}
+
+static u32 get_blocked_hpd_pin_mask(struct intel_display *display)
+{
+	enum hpd_pin pin;
+	u32 hpd_pin_mask = 0;
+
+	for_each_hpd_pin(pin) {
+		if (hpd_pin_is_blocked(display, pin))
+			hpd_pin_mask |= BIT(pin);
+	}
+
+	return hpd_pin_mask;
+}
+
 static void i915_digport_work_func(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, struct drm_i915_private, display.hotplug.dig_port_work);
-	u32 long_port_mask, short_port_mask;
+	struct intel_display *display =
+		container_of(work, struct intel_display, hotplug.dig_port_work);
+	struct intel_hotplug *hotplug = &display->hotplug;
+	u32 long_hpd_pin_mask, short_hpd_pin_mask;
 	struct intel_encoder *encoder;
+	u32 blocked_hpd_pin_mask;
 	u32 old_bits = 0;
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	long_port_mask = dev_priv->display.hotplug.long_port_mask;
-	dev_priv->display.hotplug.long_port_mask = 0;
-	short_port_mask = dev_priv->display.hotplug.short_port_mask;
-	dev_priv->display.hotplug.short_port_mask = 0;
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
-	for_each_intel_encoder(&dev_priv->drm, encoder) {
+	blocked_hpd_pin_mask = get_blocked_hpd_pin_mask(display);
+	long_hpd_pin_mask = hotplug->long_hpd_pin_mask & ~blocked_hpd_pin_mask;
+	hotplug->long_hpd_pin_mask &= ~long_hpd_pin_mask;
+	short_hpd_pin_mask = hotplug->short_hpd_pin_mask & ~blocked_hpd_pin_mask;
+	hotplug->short_hpd_pin_mask &= ~short_hpd_pin_mask;
+
+	spin_unlock_irq(&display->irq.lock);
+
+	for_each_intel_encoder(display->drm, encoder) {
 		struct intel_digital_port *dig_port;
-		enum port port = encoder->port;
+		enum hpd_pin pin = encoder->hpd_pin;
 		bool long_hpd, short_hpd;
 		enum irqreturn ret;
 
 		if (!intel_encoder_has_hpd_pulse(encoder))
 			continue;
 
-		long_hpd = long_port_mask & BIT(port);
-		short_hpd = short_port_mask & BIT(port);
+		long_hpd = long_hpd_pin_mask & BIT(pin);
+		short_hpd = short_hpd_pin_mask & BIT(pin);
 
 		if (!long_hpd && !short_hpd)
 			continue;
@@ -384,16 +430,16 @@ static void i915_digport_work_func(struct work_struct *work)
 		ret = dig_port->hpd_pulse(dig_port, long_hpd);
 		if (ret == IRQ_NONE) {
 			/* fall back to old school hpd */
-			old_bits |= BIT(encoder->hpd_pin);
+			old_bits |= BIT(pin);
 		}
 	}
 
 	if (old_bits) {
-		spin_lock_irq(&dev_priv->irq_lock);
-		dev_priv->display.hotplug.event_bits |= old_bits;
-		queue_delayed_detection_work(dev_priv,
-					     &dev_priv->display.hotplug.hotplug_work, 0);
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_lock_irq(&display->irq.lock);
+		display->hotplug.event_bits |= old_bits;
+		queue_delayed_detection_work(display,
+					     &display->hotplug.hotplug_work, 0);
+		spin_unlock_irq(&display->irq.lock);
 	}
 }
 
@@ -406,13 +452,17 @@ static void i915_digport_work_func(struct work_struct *work)
  */
 void intel_hpd_trigger_irq(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
+	struct intel_hotplug *hotplug = &display->hotplug;
+	struct intel_encoder *encoder = &dig_port->base;
+
+	spin_lock_irq(&display->irq.lock);
 
-	spin_lock_irq(&i915->irq_lock);
-	i915->display.hotplug.short_port_mask |= BIT(dig_port->base.port);
-	spin_unlock_irq(&i915->irq_lock);
+	hotplug->short_hpd_pin_mask |= BIT(encoder->hpd_pin);
+	if (!hpd_pin_is_blocked(display, encoder->hpd_pin))
+		queue_work(hotplug->dp_wq, &hotplug->dig_port_work);
 
-	queue_work(i915->display.hotplug.dp_wq, &i915->display.hotplug.dig_port_work);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 /*
@@ -420,9 +470,9 @@ void intel_hpd_trigger_irq(struct intel_digital_port *dig_port)
  */
 static void i915_hotplug_work_func(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, struct drm_i915_private,
-			     display.hotplug.hotplug_work.work);
+	struct intel_display *display =
+		container_of(work, struct intel_display, hotplug.hotplug_work.work);
+	struct intel_hotplug *hotplug = &display->hotplug;
 	struct drm_connector_list_iter conn_iter;
 	struct intel_connector *connector;
 	u32 changed = 0, retry = 0;
@@ -430,30 +480,32 @@ static void i915_hotplug_work_func(struct work_struct *work)
 	u32 hpd_retry_bits;
 	struct drm_connector *first_changed_connector = NULL;
 	int changed_connectors = 0;
+	u32 blocked_hpd_pin_mask;
 
-	mutex_lock(&dev_priv->drm.mode_config.mutex);
-	drm_dbg_kms(&dev_priv->drm, "running encoder hotplug functions\n");
+	mutex_lock(&display->drm->mode_config.mutex);
+	drm_dbg_kms(display->drm, "running encoder hotplug functions\n");
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
-	hpd_event_bits = dev_priv->display.hotplug.event_bits;
-	dev_priv->display.hotplug.event_bits = 0;
-	hpd_retry_bits = dev_priv->display.hotplug.retry_bits;
-	dev_priv->display.hotplug.retry_bits = 0;
+	blocked_hpd_pin_mask = get_blocked_hpd_pin_mask(display);
+	hpd_event_bits = hotplug->event_bits & ~blocked_hpd_pin_mask;
+	hotplug->event_bits &= ~hpd_event_bits;
+	hpd_retry_bits = hotplug->retry_bits & ~blocked_hpd_pin_mask;
+	hotplug->retry_bits &= ~hpd_retry_bits;
 
 	/* Enable polling for connectors which had HPD IRQ storms */
-	intel_hpd_irq_storm_switch_to_polling(dev_priv);
+	intel_hpd_irq_storm_switch_to_polling(display);
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
 	/* Skip calling encode hotplug handlers if ignore long HPD set*/
-	if (dev_priv->display.hotplug.ignore_long_hpd) {
-		drm_dbg_kms(&dev_priv->drm, "Ignore HPD flag on - skip encoder hotplug handlers\n");
-		mutex_unlock(&dev_priv->drm.mode_config.mutex);
+	if (display->hotplug.ignore_long_hpd) {
+		drm_dbg_kms(display->drm, "Ignore HPD flag on - skip encoder hotplug handlers\n");
+		mutex_unlock(&display->drm->mode_config.mutex);
 		return;
 	}
 
-	drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		enum hpd_pin pin;
 		u32 hpd_bit;
@@ -472,7 +524,7 @@ static void i915_hotplug_work_func(struct work_struct *work)
 			else
 				connector->hotplug_retries++;
 
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "Connector %s (pin %i) received hotplug event. (retry %d)\n",
 				    connector->base.name, pin,
 				    connector->hotplug_retries);
@@ -495,12 +547,12 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		}
 	}
 	drm_connector_list_iter_end(&conn_iter);
-	mutex_unlock(&dev_priv->drm.mode_config.mutex);
+	mutex_unlock(&display->drm->mode_config.mutex);
 
 	if (changed_connectors == 1)
 		drm_kms_helper_connector_hotplug_event(first_changed_connector);
 	else if (changed_connectors > 0)
-		drm_kms_helper_hotplug_event(&dev_priv->drm);
+		drm_kms_helper_hotplug_event(display->drm);
 
 	if (first_changed_connector)
 		drm_connector_put(first_changed_connector);
@@ -508,20 +560,20 @@ static void i915_hotplug_work_func(struct work_struct *work)
 	/* Remove shared HPD pins that have changed */
 	retry &= ~changed;
 	if (retry) {
-		spin_lock_irq(&dev_priv->irq_lock);
-		dev_priv->display.hotplug.retry_bits |= retry;
+		spin_lock_irq(&display->irq.lock);
+		display->hotplug.retry_bits |= retry;
 
-		mod_delayed_detection_work(dev_priv,
-					   &dev_priv->display.hotplug.hotplug_work,
+		mod_delayed_detection_work(display,
+					   &display->hotplug.hotplug_work,
 					   msecs_to_jiffies(HPD_RETRY_DELAY));
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_unlock_irq(&display->irq.lock);
 	}
 }
 
 
 /**
  * intel_hpd_irq_handler - main hotplug irq handler
- * @dev_priv: drm_i915_private
+ * @display: display device
  * @pin_mask: a mask of hpd pins that have triggered the irq
  * @long_mask: a mask of hpd pins that may be long hpd pulses
  *
@@ -535,7 +587,7 @@ static void i915_hotplug_work_func(struct work_struct *work)
  * Here, we do hotplug irq storm detection and mitigation, and pass further
  * processing to appropriate bottom halves.
  */
-void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
+void intel_hpd_irq_handler(struct intel_display *display,
 			   u32 pin_mask, u32 long_mask)
 {
 	struct intel_encoder *encoder;
@@ -548,7 +600,7 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	if (!pin_mask)
 		return;
 
-	spin_lock(&dev_priv->irq_lock);
+	spin_lock(&display->irq.lock);
 
 	/*
 	 * Determine whether ->hpd_pulse() exists for each pin, and
@@ -556,8 +608,7 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	 * as each pin may have up to two encoders (HDMI and DP) and
 	 * only the one of them (DP) will have ->hpd_pulse().
 	 */
-	for_each_intel_encoder(&dev_priv->drm, encoder) {
-		enum port port = encoder->port;
+	for_each_intel_encoder(display->drm, encoder) {
 		bool long_hpd;
 
 		pin = encoder->hpd_pin;
@@ -569,18 +620,20 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 
 		long_hpd = long_mask & BIT(pin);
 
-		drm_dbg(&dev_priv->drm,
+		drm_dbg(display->drm,
 			"digital hpd on [ENCODER:%d:%s] - %s\n",
 			encoder->base.base.id, encoder->base.name,
 			long_hpd ? "long" : "short");
-		queue_dig = true;
+
+		if (!hpd_pin_is_blocked(display, pin))
+			queue_dig = true;
 
 		if (long_hpd) {
 			long_hpd_pulse_mask |= BIT(pin);
-			dev_priv->display.hotplug.long_port_mask |= BIT(port);
+			display->hotplug.long_hpd_pin_mask |= BIT(pin);
 		} else {
 			short_hpd_pulse_mask |= BIT(pin);
-			dev_priv->display.hotplug.short_port_mask |= BIT(port);
+			display->hotplug.short_hpd_pin_mask |= BIT(pin);
 		}
 	}
 
@@ -591,20 +644,20 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 		if (!(BIT(pin) & pin_mask))
 			continue;
 
-		if (dev_priv->display.hotplug.stats[pin].state == HPD_DISABLED) {
+		if (display->hotplug.stats[pin].state == HPD_DISABLED) {
 			/*
 			 * On GMCH platforms the interrupt mask bits only
 			 * prevent irq generation, not the setting of the
 			 * hotplug bits itself. So only WARN about unexpected
 			 * interrupts on saner platforms.
 			 */
-			drm_WARN_ONCE(&dev_priv->drm, !HAS_GMCH(dev_priv),
+			drm_WARN_ONCE(display->drm, !HAS_GMCH(display),
 				      "Received HPD interrupt on pin %d although disabled\n",
 				      pin);
 			continue;
 		}
 
-		if (dev_priv->display.hotplug.stats[pin].state != HPD_ENABLED)
+		if (display->hotplug.stats[pin].state != HPD_ENABLED)
 			continue;
 
 		/*
@@ -615,13 +668,15 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 		if (((short_hpd_pulse_mask | long_hpd_pulse_mask) & BIT(pin))) {
 			long_hpd = long_hpd_pulse_mask & BIT(pin);
 		} else {
-			dev_priv->display.hotplug.event_bits |= BIT(pin);
+			display->hotplug.event_bits |= BIT(pin);
 			long_hpd = true;
-			queue_hp = true;
+
+			if (!hpd_pin_is_blocked(display, pin))
+				queue_hp = true;
 		}
 
-		if (intel_hpd_irq_storm_detect(dev_priv, pin, long_hpd)) {
-			dev_priv->display.hotplug.event_bits &= ~BIT(pin);
+		if (intel_hpd_irq_storm_detect(display, pin, long_hpd)) {
+			display->hotplug.event_bits &= ~BIT(pin);
 			storm_detected = true;
 			queue_hp = true;
 		}
@@ -632,7 +687,7 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	 * happens later in our hotplug work.
 	 */
 	if (storm_detected)
-		intel_hpd_irq_setup(dev_priv);
+		intel_hpd_irq_setup(display);
 
 	/*
 	 * Our hotplug handler can grab modeset locks (by calling down into the
@@ -641,17 +696,17 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	 * deadlock.
 	 */
 	if (queue_dig)
-		queue_work(dev_priv->display.hotplug.dp_wq, &dev_priv->display.hotplug.dig_port_work);
+		queue_work(display->hotplug.dp_wq, &display->hotplug.dig_port_work);
 	if (queue_hp)
-		queue_delayed_detection_work(dev_priv,
-					     &dev_priv->display.hotplug.hotplug_work, 0);
+		queue_delayed_detection_work(display,
+					     &display->hotplug.hotplug_work, 0);
 
-	spin_unlock(&dev_priv->irq_lock);
+	spin_unlock(&display->irq.lock);
 }
 
 /**
  * intel_hpd_init - initializes and enables hpd support
- * @dev_priv: i915 device instance
+ * @display: display device instance
  *
  * This function enables the hotplug support. It requires that interrupts have
  * already been enabled with intel_irq_init_hw(). From this point on hotplug and
@@ -663,40 +718,40 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
  *
  * Also see: intel_hpd_poll_enable() and intel_hpd_poll_disable().
  */
-void intel_hpd_init(struct drm_i915_private *dev_priv)
+void intel_hpd_init(struct intel_display *display)
 {
 	int i;
 
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
 	for_each_hpd_pin(i) {
-		dev_priv->display.hotplug.stats[i].count = 0;
-		dev_priv->display.hotplug.stats[i].state = HPD_ENABLED;
+		display->hotplug.stats[i].count = 0;
+		display->hotplug.stats[i].state = HPD_ENABLED;
 	}
 
 	/*
 	 * Interrupt setup is already guaranteed to be single-threaded, this is
 	 * just to make the assert_spin_locked checks happy.
 	 */
-	spin_lock_irq(&dev_priv->irq_lock);
-	intel_hpd_irq_setup(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	intel_hpd_irq_setup(display);
+	spin_unlock_irq(&display->irq.lock);
 }
 
-static void i915_hpd_poll_detect_connectors(struct drm_i915_private *i915)
+static void i915_hpd_poll_detect_connectors(struct intel_display *display)
 {
 	struct drm_connector_list_iter conn_iter;
 	struct intel_connector *connector;
 	struct intel_connector *first_changed_connector = NULL;
 	int changed = 0;
 
-	mutex_lock(&i915->drm.mode_config.mutex);
+	mutex_lock(&display->drm->mode_config.mutex);
 
-	if (!i915->drm.mode_config.poll_enabled)
+	if (!display->drm->mode_config.poll_enabled)
 		goto out;
 
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		if (!(connector->base.polled & DRM_CONNECTOR_POLL_HPD))
 			continue;
@@ -714,7 +769,7 @@ static void i915_hpd_poll_detect_connectors(struct drm_i915_private *i915)
 	drm_connector_list_iter_end(&conn_iter);
 
 out:
-	mutex_unlock(&i915->drm.mode_config.mutex);
+	mutex_unlock(&display->drm->mode_config.mutex);
 
 	if (!changed)
 		return;
@@ -722,25 +777,23 @@ out:
 	if (changed == 1)
 		drm_kms_helper_connector_hotplug_event(&first_changed_connector->base);
 	else
-		drm_kms_helper_hotplug_event(&i915->drm);
+		drm_kms_helper_hotplug_event(display->drm);
 
 	drm_connector_put(&first_changed_connector->base);
 }
 
 static void i915_hpd_poll_init_work(struct work_struct *work)
 {
-	struct drm_i915_private *dev_priv =
-		container_of(work, struct drm_i915_private,
-			     display.hotplug.poll_init_work);
-	struct intel_display *display = &dev_priv->display;
+	struct intel_display *display =
+		container_of(work, typeof(*display), hotplug.poll_init_work);
 	struct drm_connector_list_iter conn_iter;
 	struct intel_connector *connector;
 	intel_wakeref_t wakeref;
 	bool enabled;
 
-	mutex_lock(&dev_priv->drm.mode_config.mutex);
+	mutex_lock(&display->drm->mode_config.mutex);
 
-	enabled = READ_ONCE(dev_priv->display.hotplug.poll_enabled);
+	enabled = READ_ONCE(display->hotplug.poll_enabled);
 	/*
 	 * Prevent taking a power reference from this sequence of
 	 * i915_hpd_poll_init_work() -> drm_helper_hpd_irq_event() ->
@@ -750,14 +803,14 @@ static void i915_hpd_poll_init_work(struct work_struct *work)
 	if (!enabled) {
 		wakeref = intel_display_power_get(display,
 						  POWER_DOMAIN_DISPLAY_CORE);
-		drm_WARN_ON(&dev_priv->drm,
-			    READ_ONCE(dev_priv->display.hotplug.poll_enabled));
-		cancel_work(&dev_priv->display.hotplug.poll_init_work);
+		drm_WARN_ON(display->drm,
+			    READ_ONCE(display->hotplug.poll_enabled));
+		cancel_work(&display->hotplug.poll_init_work);
 	}
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 
-	drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		enum hpd_pin pin;
 
@@ -765,7 +818,7 @@ static void i915_hpd_poll_init_work(struct work_struct *work)
 		if (pin == HPD_NONE)
 			continue;
 
-		if (dev_priv->display.hotplug.stats[pin].state == HPD_DISABLED)
+		if (display->hotplug.stats[pin].state == HPD_DISABLED)
 			continue;
 
 		connector->base.polled = connector->polled;
@@ -776,19 +829,19 @@ static void i915_hpd_poll_init_work(struct work_struct *work)
 	}
 	drm_connector_list_iter_end(&conn_iter);
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
 	if (enabled)
-		drm_kms_helper_poll_reschedule(&dev_priv->drm);
+		drm_kms_helper_poll_reschedule(display->drm);
 
-	mutex_unlock(&dev_priv->drm.mode_config.mutex);
+	mutex_unlock(&display->drm->mode_config.mutex);
 
 	/*
 	 * We might have missed any hotplugs that happened while we were
 	 * in the middle of disabling polling
 	 */
 	if (!enabled) {
-		i915_hpd_poll_detect_connectors(dev_priv);
+		i915_hpd_poll_detect_connectors(display);
 
 		intel_display_power_put(display,
 					POWER_DOMAIN_DISPLAY_CORE,
@@ -798,7 +851,7 @@ static void i915_hpd_poll_init_work(struct work_struct *work)
 
 /**
  * intel_hpd_poll_enable - enable polling for connectors with hpd
- * @dev_priv: i915 device instance
+ * @display: display device instance
  *
  * This function enables polling for all connectors which support HPD.
  * Under certain conditions HPD may not be functional. On most Intel GPUs,
@@ -812,15 +865,12 @@ static void i915_hpd_poll_init_work(struct work_struct *work)
  *
  * Also see: intel_hpd_init() and intel_hpd_poll_disable().
  */
-void intel_hpd_poll_enable(struct drm_i915_private *dev_priv)
+void intel_hpd_poll_enable(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
-
-	if (!HAS_DISPLAY(dev_priv) ||
-	    !intel_display_device_enabled(display))
+	if (!HAS_DISPLAY(display) || !intel_display_device_enabled(display))
 		return;
 
-	WRITE_ONCE(dev_priv->display.hotplug.poll_enabled, true);
+	WRITE_ONCE(display->hotplug.poll_enabled, true);
 
 	/*
 	 * We might already be holding dev->mode_config.mutex, so do this in a
@@ -828,15 +878,15 @@ void intel_hpd_poll_enable(struct drm_i915_private *dev_priv)
 	 * As well, there's no issue if we race here since we always reschedule
 	 * this worker anyway
 	 */
-	spin_lock_irq(&dev_priv->irq_lock);
-	queue_detection_work(dev_priv,
-			     &dev_priv->display.hotplug.poll_init_work);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	queue_detection_work(display,
+			     &display->hotplug.poll_init_work);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 /**
  * intel_hpd_poll_disable - disable polling for connectors with hpd
- * @dev_priv: i915 device instance
+ * @display: display device instance
  *
  * This function disables polling for all connectors which support HPD.
  * Under certain conditions HPD may not be functional. On most Intel GPUs,
@@ -853,26 +903,26 @@ void intel_hpd_poll_enable(struct drm_i915_private *dev_priv)
  *
  * Also see: intel_hpd_init() and intel_hpd_poll_enable().
  */
-void intel_hpd_poll_disable(struct drm_i915_private *dev_priv)
+void intel_hpd_poll_disable(struct intel_display *display)
 {
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
-	WRITE_ONCE(dev_priv->display.hotplug.poll_enabled, false);
+	WRITE_ONCE(display->hotplug.poll_enabled, false);
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	queue_detection_work(dev_priv,
-			     &dev_priv->display.hotplug.poll_init_work);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	queue_detection_work(display,
+			     &display->hotplug.poll_init_work);
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void intel_hpd_poll_fini(struct drm_i915_private *i915)
+void intel_hpd_poll_fini(struct intel_display *display)
 {
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 
 	/* Kill all the work that may have been queued by hpd. */
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		intel_connector_cancel_modeset_retry_work(connector);
 		intel_hdcp_cancel_works(connector);
@@ -880,157 +930,261 @@ void intel_hpd_poll_fini(struct drm_i915_private *i915)
 	drm_connector_list_iter_end(&conn_iter);
 }
 
-void intel_hpd_init_early(struct drm_i915_private *i915)
+void intel_hpd_init_early(struct intel_display *display)
 {
-	INIT_DELAYED_WORK(&i915->display.hotplug.hotplug_work,
+	INIT_DELAYED_WORK(&display->hotplug.hotplug_work,
 			  i915_hotplug_work_func);
-	INIT_WORK(&i915->display.hotplug.dig_port_work, i915_digport_work_func);
-	INIT_WORK(&i915->display.hotplug.poll_init_work, i915_hpd_poll_init_work);
-	INIT_DELAYED_WORK(&i915->display.hotplug.reenable_work,
+	INIT_WORK(&display->hotplug.dig_port_work, i915_digport_work_func);
+	INIT_WORK(&display->hotplug.poll_init_work, i915_hpd_poll_init_work);
+	INIT_DELAYED_WORK(&display->hotplug.reenable_work,
 			  intel_hpd_irq_storm_reenable_work);
 
-	i915->display.hotplug.hpd_storm_threshold = HPD_STORM_DEFAULT_THRESHOLD;
+	display->hotplug.hpd_storm_threshold = HPD_STORM_DEFAULT_THRESHOLD;
 	/* If we have MST support, we want to avoid doing short HPD IRQ storm
 	 * detection, as short HPD storms will occur as a natural part of
 	 * sideband messaging with MST.
 	 * On older platforms however, IRQ storms can occur with both long and
 	 * short pulses, as seen on some G4x systems.
 	 */
-	i915->display.hotplug.hpd_short_storm_enabled = !HAS_DP_MST(i915);
+	display->hotplug.hpd_short_storm_enabled = !HAS_DP_MST(display);
 }
 
-static bool cancel_all_detection_work(struct drm_i915_private *i915)
+static bool cancel_all_detection_work(struct intel_display *display)
 {
 	bool was_pending = false;
 
-	if (cancel_delayed_work_sync(&i915->display.hotplug.hotplug_work))
+	if (cancel_delayed_work_sync(&display->hotplug.hotplug_work))
 		was_pending = true;
-	if (cancel_work_sync(&i915->display.hotplug.poll_init_work))
+	if (cancel_work_sync(&display->hotplug.poll_init_work))
 		was_pending = true;
-	if (cancel_delayed_work_sync(&i915->display.hotplug.reenable_work))
+	if (cancel_delayed_work_sync(&display->hotplug.reenable_work))
 		was_pending = true;
 
 	return was_pending;
 }
 
-void intel_hpd_cancel_work(struct drm_i915_private *dev_priv)
+void intel_hpd_cancel_work(struct intel_display *display)
 {
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(display))
 		return;
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+
+	drm_WARN_ON(display->drm, get_blocked_hpd_pin_mask(display));
 
-	dev_priv->display.hotplug.long_port_mask = 0;
-	dev_priv->display.hotplug.short_port_mask = 0;
-	dev_priv->display.hotplug.event_bits = 0;
-	dev_priv->display.hotplug.retry_bits = 0;
+	display->hotplug.long_hpd_pin_mask = 0;
+	display->hotplug.short_hpd_pin_mask = 0;
+	display->hotplug.event_bits = 0;
+	display->hotplug.retry_bits = 0;
 
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
-	cancel_work_sync(&dev_priv->display.hotplug.dig_port_work);
+	cancel_work_sync(&display->hotplug.dig_port_work);
 
 	/*
 	 * All other work triggered by hotplug events should be canceled by
 	 * now.
 	 */
-	if (cancel_all_detection_work(dev_priv))
-		drm_dbg_kms(&dev_priv->drm, "Hotplug detection work still active\n");
+	if (cancel_all_detection_work(display))
+		drm_dbg_kms(display->drm, "Hotplug detection work still active\n");
 }
 
-bool intel_hpd_disable(struct drm_i915_private *dev_priv, enum hpd_pin pin)
+static void queue_work_for_missed_irqs(struct intel_display *display)
 {
-	bool ret = false;
+	struct intel_hotplug *hotplug = &display->hotplug;
+	bool queue_hp_work = false;
+	u32 blocked_hpd_pin_mask;
+	enum hpd_pin pin;
 
-	if (pin == HPD_NONE)
-		return false;
+	lockdep_assert_held(&display->irq.lock);
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	if (dev_priv->display.hotplug.stats[pin].state == HPD_ENABLED) {
-		dev_priv->display.hotplug.stats[pin].state = HPD_DISABLED;
-		ret = true;
+	blocked_hpd_pin_mask = get_blocked_hpd_pin_mask(display);
+	if ((hotplug->event_bits | hotplug->retry_bits) & ~blocked_hpd_pin_mask)
+		queue_hp_work = true;
+
+	for_each_hpd_pin(pin) {
+		switch (display->hotplug.stats[pin].state) {
+		case HPD_MARK_DISABLED:
+			queue_hp_work = true;
+			break;
+		case HPD_DISABLED:
+		case HPD_ENABLED:
+			break;
+		default:
+			MISSING_CASE(display->hotplug.stats[pin].state);
+		}
 	}
-	spin_unlock_irq(&dev_priv->irq_lock);
 
-	return ret;
+	if ((hotplug->long_hpd_pin_mask | hotplug->short_hpd_pin_mask) & ~blocked_hpd_pin_mask)
+		queue_work(hotplug->dp_wq, &hotplug->dig_port_work);
+
+	if (queue_hp_work)
+		queue_delayed_detection_work(display, &display->hotplug.hotplug_work, 0);
 }
 
-void intel_hpd_enable(struct drm_i915_private *dev_priv, enum hpd_pin pin)
+static bool block_hpd_pin(struct intel_display *display, enum hpd_pin pin)
 {
-	if (pin == HPD_NONE)
+	struct intel_hotplug *hotplug = &display->hotplug;
+
+	lockdep_assert_held(&display->irq.lock);
+
+	hotplug->stats[pin].blocked_count++;
+
+	return hotplug->stats[pin].blocked_count == 1;
+}
+
+static bool unblock_hpd_pin(struct intel_display *display, enum hpd_pin pin)
+{
+	struct intel_hotplug *hotplug = &display->hotplug;
+
+	lockdep_assert_held(&display->irq.lock);
+
+	if (drm_WARN_ON(display->drm, hotplug->stats[pin].blocked_count == 0))
+		return true;
+
+	hotplug->stats[pin].blocked_count--;
+
+	return hotplug->stats[pin].blocked_count == 0;
+}
+
+/**
+ * intel_hpd_block - Block handling of HPD IRQs on an HPD pin
+ * @encoder: Encoder to block the HPD handling for
+ *
+ * Blocks the handling of HPD IRQs on the HPD pin of @encoder.
+ *
+ * On return:
+ *
+ * - It's guaranteed that the blocked encoders' HPD pulse handler
+ *   (via intel_digital_port::hpd_pulse()) is not running.
+ * - The hotplug event handling (via intel_encoder::hotplug()) of an
+ *   HPD IRQ pending at the time this function is called may be still
+ *   running.
+ * - Detection on the encoder's connector (via
+ *   drm_connector_helper_funcs::detect_ctx(),
+ *   drm_connector_funcs::detect()) remains allowed, for instance as part of
+ *   userspace connector probing, or DRM core's connector polling.
+ *
+ * The call must be followed by calling intel_hpd_unblock(), or
+ * intel_hpd_clear_and_unblock().
+ *
+ * Note that the handling of HPD IRQs for another encoder using the same HPD
+ * pin as that of @encoder will be also blocked.
+ */
+void intel_hpd_block(struct intel_encoder *encoder)
+{
+	struct intel_display *display = to_intel_display(encoder);
+	struct intel_hotplug *hotplug = &display->hotplug;
+	bool do_flush = false;
+
+	if (encoder->hpd_pin == HPD_NONE)
+		return;
+
+	spin_lock_irq(&display->irq.lock);
+
+	if (block_hpd_pin(display, encoder->hpd_pin))
+		do_flush = true;
+
+	spin_unlock_irq(&display->irq.lock);
+
+	if (do_flush && hpd_pin_has_pulse(display, encoder->hpd_pin))
+		flush_work(&hotplug->dig_port_work);
+}
+
+/**
+ * intel_hpd_unblock - Unblock handling of HPD IRQs on an HPD pin
+ * @encoder: Encoder to unblock the HPD handling for
+ *
+ * Unblock the handling of HPD IRQs on the HPD pin of @encoder, which was
+ * previously blocked by intel_hpd_block(). Any HPD IRQ raised on the
+ * HPD pin while it was blocked will be handled for @encoder and for any
+ * other encoder sharing the same HPD pin.
+ */
+void intel_hpd_unblock(struct intel_encoder *encoder)
+{
+	struct intel_display *display = to_intel_display(encoder);
+
+	if (encoder->hpd_pin == HPD_NONE)
 		return;
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	dev_priv->display.hotplug.stats[pin].state = HPD_ENABLED;
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+
+	if (unblock_hpd_pin(display, encoder->hpd_pin))
+		queue_work_for_missed_irqs(display);
+
+	spin_unlock_irq(&display->irq.lock);
 }
 
-static void queue_work_for_missed_irqs(struct drm_i915_private *i915)
+/**
+ * intel_hpd_clear_and_unblock - Unblock handling of new HPD IRQs on an HPD pin
+ * @encoder: Encoder to unblock the HPD handling for
+ *
+ * Unblock the handling of HPD IRQs on the HPD pin of @encoder, which was
+ * previously blocked by intel_hpd_block(). Any HPD IRQ raised on the
+ * HPD pin while it was blocked will be cleared, handling only new IRQs.
+ */
+void intel_hpd_clear_and_unblock(struct intel_encoder *encoder)
 {
-	bool queue_work = false;
-	enum hpd_pin pin;
+	struct intel_display *display = to_intel_display(encoder);
+	struct intel_hotplug *hotplug = &display->hotplug;
+	enum hpd_pin pin = encoder->hpd_pin;
 
-	lockdep_assert_held(&i915->irq_lock);
+	if (pin == HPD_NONE)
+		return;
 
-	if (i915->display.hotplug.event_bits ||
-	    i915->display.hotplug.retry_bits)
-		queue_work = true;
+	spin_lock_irq(&display->irq.lock);
 
-	for_each_hpd_pin(pin) {
-		switch (i915->display.hotplug.stats[pin].state) {
-		case HPD_MARK_DISABLED:
-			queue_work = true;
-			break;
-		case HPD_ENABLED:
-			break;
-		default:
-			MISSING_CASE(i915->display.hotplug.stats[pin].state);
-		}
+	if (unblock_hpd_pin(display, pin)) {
+		hotplug->event_bits &= ~BIT(pin);
+		hotplug->retry_bits &= ~BIT(pin);
+		hotplug->short_hpd_pin_mask &= ~BIT(pin);
+		hotplug->long_hpd_pin_mask &= ~BIT(pin);
 	}
 
-	if (queue_work)
-		queue_delayed_detection_work(i915, &i915->display.hotplug.hotplug_work, 0);
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void intel_hpd_enable_detection_work(struct drm_i915_private *i915)
+void intel_hpd_enable_detection_work(struct intel_display *display)
 {
-	spin_lock_irq(&i915->irq_lock);
-	i915->display.hotplug.detection_work_enabled = true;
-	queue_work_for_missed_irqs(i915);
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	display->hotplug.detection_work_enabled = true;
+	queue_work_for_missed_irqs(display);
+	spin_unlock_irq(&display->irq.lock);
 }
 
-void intel_hpd_disable_detection_work(struct drm_i915_private *i915)
+void intel_hpd_disable_detection_work(struct intel_display *display)
 {
-	spin_lock_irq(&i915->irq_lock);
-	i915->display.hotplug.detection_work_enabled = false;
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	display->hotplug.detection_work_enabled = false;
+	spin_unlock_irq(&display->irq.lock);
 
-	cancel_all_detection_work(i915);
+	cancel_all_detection_work(display);
 }
 
-bool intel_hpd_schedule_detection(struct drm_i915_private *i915)
+bool intel_hpd_schedule_detection(struct intel_display *display)
 {
 	unsigned long flags;
 	bool ret;
 
-	spin_lock_irqsave(&i915->irq_lock, flags);
-	ret = queue_delayed_detection_work(i915, &i915->display.hotplug.hotplug_work, 0);
-	spin_unlock_irqrestore(&i915->irq_lock, flags);
+	spin_lock_irqsave(&display->irq.lock, flags);
+	ret = queue_delayed_detection_work(display, &display->hotplug.hotplug_work, 0);
+	spin_unlock_irqrestore(&display->irq.lock, flags);
 
 	return ret;
 }
 
 static int i915_hpd_storm_ctl_show(struct seq_file *m, void *data)
 {
-	struct drm_i915_private *dev_priv = m->private;
-	struct intel_hotplug *hotplug = &dev_priv->display.hotplug;
+	struct intel_display *display = m->private;
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	struct intel_hotplug *hotplug = &display->hotplug;
 
 	/* Synchronize with everything first in case there's been an HPD
 	 * storm, but we haven't finished handling it in the kernel yet
 	 */
 	intel_synchronize_irq(dev_priv);
-	flush_work(&dev_priv->display.hotplug.dig_port_work);
-	flush_delayed_work(&dev_priv->display.hotplug.hotplug_work);
+	flush_work(&display->hotplug.dig_port_work);
+	flush_delayed_work(&display->hotplug.hotplug_work);
 
 	seq_printf(m, "Threshold: %d\n", hotplug->hpd_storm_threshold);
 	seq_printf(m, "Detected: %s\n",
@@ -1044,8 +1198,8 @@ static ssize_t i915_hpd_storm_ctl_write(struct file *file,
 					loff_t *offp)
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *dev_priv = m->private;
-	struct intel_hotplug *hotplug = &dev_priv->display.hotplug;
+	struct intel_display *display = m->private;
+	struct intel_hotplug *hotplug = &display->hotplug;
 	unsigned int new_threshold;
 	int i;
 	char *newline;
@@ -1070,21 +1224,21 @@ static ssize_t i915_hpd_storm_ctl_write(struct file *file,
 		return -EINVAL;
 
 	if (new_threshold > 0)
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Setting HPD storm detection threshold to %d\n",
 			    new_threshold);
 	else
-		drm_dbg_kms(&dev_priv->drm, "Disabling HPD storm detection\n");
+		drm_dbg_kms(display->drm, "Disabling HPD storm detection\n");
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 	hotplug->hpd_storm_threshold = new_threshold;
 	/* Reset the HPD storm stats so we don't accidentally trigger a storm */
 	for_each_hpd_pin(i)
 		hotplug->stats[i].count = 0;
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
 	/* Re-enable hpd immediately if we were in an irq storm */
-	flush_delayed_work(&dev_priv->display.hotplug.reenable_work);
+	flush_delayed_work(&display->hotplug.reenable_work);
 
 	return len;
 }
@@ -1105,10 +1259,10 @@ static const struct file_operations i915_hpd_storm_ctl_fops = {
 
 static int i915_hpd_short_storm_ctl_show(struct seq_file *m, void *data)
 {
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 
 	seq_printf(m, "Enabled: %s\n",
-		   str_yes_no(dev_priv->display.hotplug.hpd_short_storm_enabled));
+		   str_yes_no(display->hotplug.hpd_short_storm_enabled));
 
 	return 0;
 }
@@ -1125,8 +1279,8 @@ static ssize_t i915_hpd_short_storm_ctl_write(struct file *file,
 					      size_t len, loff_t *offp)
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *dev_priv = m->private;
-	struct intel_hotplug *hotplug = &dev_priv->display.hotplug;
+	struct intel_display *display = m->private;
+	struct intel_hotplug *hotplug = &display->hotplug;
 	char *newline;
 	char tmp[16];
 	int i;
@@ -1147,22 +1301,22 @@ static ssize_t i915_hpd_short_storm_ctl_write(struct file *file,
 
 	/* Reset to the "default" state for this system */
 	if (strcmp(tmp, "reset") == 0)
-		new_state = !HAS_DP_MST(dev_priv);
+		new_state = !HAS_DP_MST(display);
 	else if (kstrtobool(tmp, &new_state) != 0)
 		return -EINVAL;
 
-	drm_dbg_kms(&dev_priv->drm, "%sabling HPD short storm detection\n",
+	drm_dbg_kms(display->drm, "%sabling HPD short storm detection\n",
 		    new_state ? "En" : "Dis");
 
-	spin_lock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
 	hotplug->hpd_short_storm_enabled = new_state;
 	/* Reset the HPD storm stats so we don't accidentally trigger a storm */
 	for_each_hpd_pin(i)
 		hotplug->stats[i].count = 0;
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_unlock_irq(&display->irq.lock);
 
 	/* Re-enable hpd immediately if we were in an irq storm */
-	flush_delayed_work(&dev_priv->display.hotplug.reenable_work);
+	flush_delayed_work(&display->hotplug.reenable_work);
 
 	return len;
 }
@@ -1176,14 +1330,14 @@ static const struct file_operations i915_hpd_short_storm_ctl_fops = {
 	.write = i915_hpd_short_storm_ctl_write,
 };
 
-void intel_hpd_debugfs_register(struct drm_i915_private *i915)
+void intel_hpd_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = i915->drm.primary;
+	struct drm_minor *minor = display->drm->primary;
 
 	debugfs_create_file("i915_hpd_storm_ctl", 0644, minor->debugfs_root,
-			    i915, &i915_hpd_storm_ctl_fops);
+			    display, &i915_hpd_storm_ctl_fops);
 	debugfs_create_file("i915_hpd_short_storm_ctl", 0644, minor->debugfs_root,
-			    i915, &i915_hpd_short_storm_ctl_fops);
+			    display, &i915_hpd_short_storm_ctl_fops);
 	debugfs_create_bool("i915_ignore_long_hpd", 0644, minor->debugfs_root,
-			    &i915->display.hotplug.ignore_long_hpd);
+			    &display->hotplug.ignore_long_hpd);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.h b/drivers/gpu/drm/i915/display/intel_hotplug.h
index d6986902b054..edc41c9d3d65 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.h
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.h
@@ -8,30 +8,31 @@
 
 #include <linux/types.h>
 
-struct drm_i915_private;
+enum port;
 struct intel_connector;
 struct intel_digital_port;
+struct intel_display;
 struct intel_encoder;
-enum port;
 
-void intel_hpd_poll_enable(struct drm_i915_private *dev_priv);
-void intel_hpd_poll_disable(struct drm_i915_private *dev_priv);
-void intel_hpd_poll_fini(struct drm_i915_private *i915);
+void intel_hpd_poll_enable(struct intel_display *display);
+void intel_hpd_poll_disable(struct intel_display *display);
+void intel_hpd_poll_fini(struct intel_display *display);
 enum intel_hotplug_state intel_encoder_hotplug(struct intel_encoder *encoder,
 					       struct intel_connector *connector);
-void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
+void intel_hpd_irq_handler(struct intel_display *display,
 			   u32 pin_mask, u32 long_mask);
 void intel_hpd_trigger_irq(struct intel_digital_port *dig_port);
-void intel_hpd_init(struct drm_i915_private *dev_priv);
-void intel_hpd_init_early(struct drm_i915_private *i915);
-void intel_hpd_cancel_work(struct drm_i915_private *dev_priv);
+void intel_hpd_init(struct intel_display *display);
+void intel_hpd_init_early(struct intel_display *display);
+void intel_hpd_cancel_work(struct intel_display *display);
 enum hpd_pin intel_hpd_pin_default(enum port port);
-bool intel_hpd_disable(struct drm_i915_private *dev_priv, enum hpd_pin pin);
-void intel_hpd_enable(struct drm_i915_private *dev_priv, enum hpd_pin pin);
-void intel_hpd_debugfs_register(struct drm_i915_private *i915);
+void intel_hpd_block(struct intel_encoder *encoder);
+void intel_hpd_unblock(struct intel_encoder *encoder);
+void intel_hpd_clear_and_unblock(struct intel_encoder *encoder);
+void intel_hpd_debugfs_register(struct intel_display *display);
 
-void intel_hpd_enable_detection_work(struct drm_i915_private *i915);
-void intel_hpd_disable_detection_work(struct drm_i915_private *i915);
-bool intel_hpd_schedule_detection(struct drm_i915_private *i915);
+void intel_hpd_enable_detection_work(struct intel_display *display);
+void intel_hpd_disable_detection_work(struct intel_display *display);
+bool intel_hpd_schedule_detection(struct intel_display *display);
 
 #endif /* __INTEL_HOTPLUG_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
index 2137ac7b882a..c024b42369c8 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
@@ -3,8 +3,10 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_irq.h"
 #include "intel_display_types.h"
@@ -131,68 +133,67 @@ static const u32 hpd_mtp[HPD_NUM_PINS] = {
 	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4),
 };
 
-static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
+static void intel_hpd_init_pins(struct intel_display *display)
 {
-	struct intel_hotplug *hpd = &dev_priv->display.hotplug;
+	struct intel_hotplug *hpd = &display->hotplug;
 
-	if (HAS_GMCH(dev_priv)) {
-		if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) ||
-		    IS_CHERRYVIEW(dev_priv))
+	if (HAS_GMCH(display)) {
+		if (display->platform.g4x || display->platform.valleyview ||
+		    display->platform.cherryview)
 			hpd->hpd = hpd_status_g4x;
 		else
 			hpd->hpd = hpd_status_i915;
 		return;
 	}
 
-	if (DISPLAY_VER(dev_priv) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		hpd->hpd = hpd_xelpdp;
-	else if (DISPLAY_VER(dev_priv) >= 11)
+	else if (DISPLAY_VER(display) >= 11)
 		hpd->hpd = hpd_gen11;
-	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	else if (display->platform.geminilake || display->platform.broxton)
 		hpd->hpd = hpd_bxt;
-	else if (DISPLAY_VER(dev_priv) == 9)
+	else if (DISPLAY_VER(display) == 9)
 		hpd->hpd = NULL; /* no north HPD on SKL */
-	else if (DISPLAY_VER(dev_priv) >= 8)
+	else if (DISPLAY_VER(display) >= 8)
 		hpd->hpd = hpd_bdw;
-	else if (DISPLAY_VER(dev_priv) >= 7)
+	else if (DISPLAY_VER(display) >= 7)
 		hpd->hpd = hpd_ivb;
 	else
 		hpd->hpd = hpd_ilk;
 
-	if ((INTEL_PCH_TYPE(dev_priv) < PCH_DG1) &&
-	    (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv)))
+	if ((INTEL_PCH_TYPE(display) < PCH_DG1) &&
+	    (!HAS_PCH_SPLIT(display) || HAS_PCH_NOP(display)))
 		return;
 
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTL)
+	if (INTEL_PCH_TYPE(display) >= PCH_MTL)
 		hpd->pch_hpd = hpd_mtp;
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1)
+	else if (INTEL_PCH_TYPE(display) >= PCH_DG1)
 		hpd->pch_hpd = hpd_sde_dg1;
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+	else if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 		hpd->pch_hpd = hpd_icp;
-	else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv))
+	else if (HAS_PCH_CNP(display) || HAS_PCH_SPT(display))
 		hpd->pch_hpd = hpd_spt;
-	else if (HAS_PCH_LPT(dev_priv) || HAS_PCH_CPT(dev_priv))
+	else if (HAS_PCH_LPT(display) || HAS_PCH_CPT(display))
 		hpd->pch_hpd = hpd_cpt;
-	else if (HAS_PCH_IBX(dev_priv))
+	else if (HAS_PCH_IBX(display))
 		hpd->pch_hpd = hpd_ibx;
 	else
-		MISSING_CASE(INTEL_PCH_TYPE(dev_priv));
+		MISSING_CASE(INTEL_PCH_TYPE(display));
 }
 
 /* For display hotplug interrupt */
-void i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv,
+void i915_hotplug_interrupt_update_locked(struct intel_display *display,
 					  u32 mask, u32 bits)
 {
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, bits & ~mask);
+	lockdep_assert_held(&display->irq.lock);
+	drm_WARN_ON(display->drm, bits & ~mask);
 
-	intel_uncore_rmw(&dev_priv->uncore, PORT_HOTPLUG_EN(dev_priv), mask,
-			 bits);
+	intel_de_rmw(display, PORT_HOTPLUG_EN(display), mask, bits);
 }
 
 /**
  * i915_hotplug_interrupt_update - update hotplug interrupt enable
- * @dev_priv: driver private
+ * @display: display device
  * @mask: bits to update
  * @bits: bits to enable
  * NOTE: the HPD enable bits are modified both inside and outside
@@ -202,13 +203,13 @@ void i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv,
  * held already, this function acquires the lock itself. A non-locking
  * version is also available.
  */
-void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
+void i915_hotplug_interrupt_update(struct intel_display *display,
 				   u32 mask,
 				   u32 bits)
 {
-	spin_lock_irq(&dev_priv->irq_lock);
-	i915_hotplug_interrupt_update_locked(dev_priv, mask, bits);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	i915_hotplug_interrupt_update_locked(display, mask, bits);
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static bool gen11_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
@@ -339,7 +340,7 @@ static bool i9xx_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
  *
  * Note that the caller is expected to zero out the masks initially.
  */
-static void intel_get_hpd_pins(struct drm_i915_private *dev_priv,
+static void intel_get_hpd_pins(struct intel_display *display,
 			       u32 *pin_mask, u32 *long_mask,
 			       u32 hotplug_trigger, u32 dig_hotplug_reg,
 			       const u32 hpd[HPD_NUM_PINS],
@@ -359,37 +360,37 @@ static void intel_get_hpd_pins(struct drm_i915_private *dev_priv,
 			*long_mask |= BIT(pin);
 	}
 
-	drm_dbg(&dev_priv->drm,
-		"hotplug event received, stat 0x%08x, dig 0x%08x, pins 0x%08x, long 0x%08x\n",
-		hotplug_trigger, dig_hotplug_reg, *pin_mask, *long_mask);
+	drm_dbg_kms(display->drm,
+		    "hotplug event received, stat 0x%08x, dig 0x%08x, pins 0x%08x, long 0x%08x\n",
+		    hotplug_trigger, dig_hotplug_reg, *pin_mask, *long_mask);
 }
 
-static u32 intel_hpd_enabled_irqs(struct drm_i915_private *dev_priv,
+static u32 intel_hpd_enabled_irqs(struct intel_display *display,
 				  const u32 hpd[HPD_NUM_PINS])
 {
 	struct intel_encoder *encoder;
 	u32 enabled_irqs = 0;
 
-	for_each_intel_encoder(&dev_priv->drm, encoder)
-		if (dev_priv->display.hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED)
+	for_each_intel_encoder(display->drm, encoder)
+		if (display->hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED)
 			enabled_irqs |= hpd[encoder->hpd_pin];
 
 	return enabled_irqs;
 }
 
-static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv,
+static u32 intel_hpd_hotplug_irqs(struct intel_display *display,
 				  const u32 hpd[HPD_NUM_PINS])
 {
 	struct intel_encoder *encoder;
 	u32 hotplug_irqs = 0;
 
-	for_each_intel_encoder(&dev_priv->drm, encoder)
+	for_each_intel_encoder(display->drm, encoder)
 		hotplug_irqs |= hpd[encoder->hpd_pin];
 
 	return hotplug_irqs;
 }
 
-static u32 intel_hpd_hotplug_mask(struct drm_i915_private *i915,
+static u32 intel_hpd_hotplug_mask(struct intel_display *display,
 				  hotplug_mask_func hotplug_mask)
 {
 	enum hpd_pin pin;
@@ -401,25 +402,25 @@ static u32 intel_hpd_hotplug_mask(struct drm_i915_private *i915,
 	return hotplug;
 }
 
-static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915,
+static u32 intel_hpd_hotplug_enables(struct intel_display *display,
 				     hotplug_enables_func hotplug_enables)
 {
 	struct intel_encoder *encoder;
 	u32 hotplug = 0;
 
-	for_each_intel_encoder(&i915->drm, encoder)
+	for_each_intel_encoder(display->drm, encoder)
 		hotplug |= hotplug_enables(encoder);
 
 	return hotplug;
 }
 
-u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
+u32 i9xx_hpd_irq_ack(struct intel_display *display)
 {
 	u32 hotplug_status = 0, hotplug_status_mask;
 	int i;
 
-	if (IS_G4X(dev_priv) ||
-	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+	if (display->platform.g4x ||
+	    display->platform.valleyview || display->platform.cherryview)
 		hotplug_status_mask = HOTPLUG_INT_STATUS_G4X |
 			DP_AUX_CHANNEL_MASK_INT_STATUS_G4X;
 	else
@@ -435,53 +436,51 @@ u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
 	 * bits can itself generate a new hotplug interrupt :(
 	 */
 	for (i = 0; i < 10; i++) {
-		u32 tmp = intel_uncore_read(&dev_priv->uncore,
-					    PORT_HOTPLUG_STAT(dev_priv)) & hotplug_status_mask;
+		u32 tmp = intel_de_read(display,
+					PORT_HOTPLUG_STAT(display)) & hotplug_status_mask;
 
 		if (tmp == 0)
 			return hotplug_status;
 
 		hotplug_status |= tmp;
-		intel_uncore_write(&dev_priv->uncore,
-				   PORT_HOTPLUG_STAT(dev_priv),
-				   hotplug_status);
+		intel_de_write(display, PORT_HOTPLUG_STAT(display),
+			       hotplug_status);
 	}
 
-	drm_WARN_ONCE(&dev_priv->drm, 1,
+	drm_WARN_ONCE(display->drm, 1,
 		      "PORT_HOTPLUG_STAT did not clear (0x%08x)\n",
-		      intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT(dev_priv)));
+		      intel_de_read(display, PORT_HOTPLUG_STAT(display)));
 
 	return hotplug_status;
 }
 
-void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_status)
+void i9xx_hpd_irq_handler(struct intel_display *display, u32 hotplug_status)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 pin_mask = 0, long_mask = 0;
 	u32 hotplug_trigger;
 
-	if (IS_G4X(dev_priv) ||
-	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+	if (display->platform.g4x ||
+	    display->platform.valleyview || display->platform.cherryview)
 		hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
 	else
 		hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
 
 	if (hotplug_trigger) {
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   hotplug_trigger, hotplug_trigger,
-				   dev_priv->display.hotplug.hpd,
+				   display->hotplug.hpd,
 				   i9xx_port_hotplug_long_detect);
 
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+		intel_hpd_irq_handler(display, pin_mask, long_mask);
 	}
 
-	if ((IS_G4X(dev_priv) ||
-	     IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+	if ((display->platform.g4x ||
+	     display->platform.valleyview || display->platform.cherryview) &&
 	    hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
 		intel_dp_aux_irq_handler(display);
 }
 
-void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
+void ibx_hpd_irq_handler(struct intel_display *display, u32 hotplug_trigger)
 {
 	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
@@ -491,7 +490,7 @@ void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
 	 * zero. Not acking leads to "The master control interrupt lied (SDE)!"
 	 * errors.
 	 */
-	dig_hotplug_reg = intel_uncore_read(&dev_priv->uncore, PCH_PORT_HOTPLUG);
+	dig_hotplug_reg = intel_de_read(display, PCH_PORT_HOTPLUG);
 	if (!hotplug_trigger) {
 		u32 mask = PORTA_HOTPLUG_STATUS_MASK |
 			PORTD_HOTPLUG_STATUS_MASK |
@@ -500,63 +499,61 @@ void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
 		dig_hotplug_reg &= ~mask;
 	}
 
-	intel_uncore_write(&dev_priv->uncore, PCH_PORT_HOTPLUG, dig_hotplug_reg);
+	intel_de_write(display, PCH_PORT_HOTPLUG, dig_hotplug_reg);
 	if (!hotplug_trigger)
 		return;
 
-	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+	intel_get_hpd_pins(display, &pin_mask, &long_mask,
 			   hotplug_trigger, dig_hotplug_reg,
-			   dev_priv->display.hotplug.pch_hpd,
+			   display->hotplug.pch_hpd,
 			   pch_port_hotplug_long_detect);
 
-	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+	intel_hpd_irq_handler(display, pin_mask, long_mask);
 }
 
-void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir)
+void xelpdp_pica_irq_handler(struct intel_display *display, u32 iir)
 {
-	struct intel_display *display = &i915->display;
 	enum hpd_pin pin;
 	u32 hotplug_trigger = iir & (XELPDP_DP_ALT_HOTPLUG_MASK | XELPDP_TBT_HOTPLUG_MASK);
 	u32 trigger_aux = iir & XELPDP_AUX_TC_MASK;
 	u32 pin_mask = 0, long_mask = 0;
 
-	if (DISPLAY_VER(i915) >= 20)
+	if (DISPLAY_VER(display) >= 20)
 		trigger_aux |= iir & XE2LPD_AUX_DDI_MASK;
 
 	for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) {
 		u32 val;
 
-		if (!(i915->display.hotplug.hpd[pin] & hotplug_trigger))
+		if (!(display->hotplug.hpd[pin] & hotplug_trigger))
 			continue;
 
 		pin_mask |= BIT(pin);
 
-		val = intel_de_read(i915, XELPDP_PORT_HOTPLUG_CTL(pin));
-		intel_de_write(i915, XELPDP_PORT_HOTPLUG_CTL(pin), val);
+		val = intel_de_read(display, XELPDP_PORT_HOTPLUG_CTL(pin));
+		intel_de_write(display, XELPDP_PORT_HOTPLUG_CTL(pin), val);
 
 		if (val & (XELPDP_DP_ALT_HPD_LONG_DETECT | XELPDP_TBT_HPD_LONG_DETECT))
 			long_mask |= BIT(pin);
 	}
 
 	if (pin_mask) {
-		drm_dbg(&i915->drm,
-			"pica hotplug event received, stat 0x%08x, pins 0x%08x, long 0x%08x\n",
-			hotplug_trigger, pin_mask, long_mask);
+		drm_dbg_kms(display->drm,
+			    "pica hotplug event received, stat 0x%08x, pins 0x%08x, long 0x%08x\n",
+			    hotplug_trigger, pin_mask, long_mask);
 
-		intel_hpd_irq_handler(i915, pin_mask, long_mask);
+		intel_hpd_irq_handler(display, pin_mask, long_mask);
 	}
 
 	if (trigger_aux)
 		intel_dp_aux_irq_handler(display);
 
 	if (!pin_mask && !trigger_aux)
-		drm_err(&i915->drm,
+		drm_err(display->drm,
 			"Unexpected DE HPD/AUX interrupt 0x%08x\n", iir);
 }
 
-void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+void icp_irq_handler(struct intel_display *display, u32 pch_iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 ddi_hotplug_trigger = pch_iir & SDE_DDI_HOTPLUG_MASK_ICP;
 	u32 tc_hotplug_trigger = pch_iir & SDE_TC_HOTPLUG_MASK_ICP;
 	u32 pin_mask = 0, long_mask = 0;
@@ -565,37 +562,36 @@ void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		u32 dig_hotplug_reg;
 
 		/* Locking due to DSI native GPIO sequences */
-		spin_lock(&dev_priv->irq_lock);
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, 0, 0);
-		spin_unlock(&dev_priv->irq_lock);
+		spin_lock(&display->irq.lock);
+		dig_hotplug_reg = intel_de_rmw(display, SHOTPLUG_CTL_DDI, 0, 0);
+		spin_unlock(&display->irq.lock);
 
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   ddi_hotplug_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
+				   display->hotplug.pch_hpd,
 				   icp_ddi_port_hotplug_long_detect);
 	}
 
 	if (tc_hotplug_trigger) {
 		u32 dig_hotplug_reg;
 
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, 0, 0);
+		dig_hotplug_reg = intel_de_rmw(display, SHOTPLUG_CTL_TC, 0, 0);
 
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   tc_hotplug_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
+				   display->hotplug.pch_hpd,
 				   icp_tc_port_hotplug_long_detect);
 	}
 
 	if (pin_mask)
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+		intel_hpd_irq_handler(display, pin_mask, long_mask);
 
 	if (pch_iir & SDE_GMBUS_ICP)
 		intel_gmbus_irq_handler(display);
 }
 
-void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+void spt_irq_handler(struct intel_display *display, u32 pch_iir)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT &
 		~SDE_PORTE_HOTPLUG_SPT;
 	u32 hotplug2_trigger = pch_iir & SDE_PORTE_HOTPLUG_SPT;
@@ -604,61 +600,61 @@ void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 	if (hotplug_trigger) {
 		u32 dig_hotplug_reg;
 
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0);
+		dig_hotplug_reg = intel_de_rmw(display, PCH_PORT_HOTPLUG, 0, 0);
 
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   hotplug_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
+				   display->hotplug.pch_hpd,
 				   spt_port_hotplug_long_detect);
 	}
 
 	if (hotplug2_trigger) {
 		u32 dig_hotplug_reg;
 
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, 0, 0);
+		dig_hotplug_reg = intel_de_rmw(display, PCH_PORT_HOTPLUG2, 0, 0);
 
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   hotplug2_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
+				   display->hotplug.pch_hpd,
 				   spt_port_hotplug2_long_detect);
 	}
 
 	if (pin_mask)
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+		intel_hpd_irq_handler(display, pin_mask, long_mask);
 
 	if (pch_iir & SDE_GMBUS_CPT)
 		intel_gmbus_irq_handler(display);
 }
 
-void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
+void ilk_hpd_irq_handler(struct intel_display *display, u32 hotplug_trigger)
 {
 	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
-	dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, 0, 0);
+	dig_hotplug_reg = intel_de_rmw(display, DIGITAL_PORT_HOTPLUG_CNTRL, 0, 0);
 
-	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+	intel_get_hpd_pins(display, &pin_mask, &long_mask,
 			   hotplug_trigger, dig_hotplug_reg,
-			   dev_priv->display.hotplug.hpd,
+			   display->hotplug.hpd,
 			   ilk_port_hotplug_long_detect);
 
-	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+	intel_hpd_irq_handler(display, pin_mask, long_mask);
 }
 
-void bxt_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
+void bxt_hpd_irq_handler(struct intel_display *display, u32 hotplug_trigger)
 {
 	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
-	dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0);
+	dig_hotplug_reg = intel_de_rmw(display, PCH_PORT_HOTPLUG, 0, 0);
 
-	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+	intel_get_hpd_pins(display, &pin_mask, &long_mask,
 			   hotplug_trigger, dig_hotplug_reg,
-			   dev_priv->display.hotplug.hpd,
+			   display->hotplug.hpd,
 			   bxt_port_hotplug_long_detect);
 
-	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+	intel_hpd_irq_handler(display, pin_mask, long_mask);
 }
 
-void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
+void gen11_hpd_irq_handler(struct intel_display *display, u32 iir)
 {
 	u32 pin_mask = 0, long_mask = 0;
 	u32 trigger_tc = iir & GEN11_DE_TC_HOTPLUG_MASK;
@@ -667,29 +663,29 @@ void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 	if (trigger_tc) {
 		u32 dig_hotplug_reg;
 
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, 0, 0);
+		dig_hotplug_reg = intel_de_rmw(display, GEN11_TC_HOTPLUG_CTL, 0, 0);
 
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   trigger_tc, dig_hotplug_reg,
-				   dev_priv->display.hotplug.hpd,
+				   display->hotplug.hpd,
 				   gen11_port_hotplug_long_detect);
 	}
 
 	if (trigger_tbt) {
 		u32 dig_hotplug_reg;
 
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, 0, 0);
+		dig_hotplug_reg = intel_de_rmw(display, GEN11_TBT_HOTPLUG_CTL, 0, 0);
 
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+		intel_get_hpd_pins(display, &pin_mask, &long_mask,
 				   trigger_tbt, dig_hotplug_reg,
-				   dev_priv->display.hotplug.hpd,
+				   display->hotplug.hpd,
 				   gen11_port_hotplug_long_detect);
 	}
 
 	if (pin_mask)
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+		intel_hpd_irq_handler(display, pin_mask, long_mask);
 	else
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"Unexpected DE HPD interrupt 0x%08x\n", iir);
 }
 
@@ -711,7 +707,7 @@ static u32 ibx_hotplug_mask(enum hpd_pin hpd_pin)
 
 static u32 ibx_hotplug_enables(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
 	switch (encoder->hpd_pin) {
 	case HPD_PORT_A:
@@ -719,7 +715,7 @@ static u32 ibx_hotplug_enables(struct intel_encoder *encoder)
 		 * When CPU and PCH are on the same package, port A
 		 * HPD must be enabled in both north and south.
 		 */
-		return HAS_PCH_LPT_LP(i915) ?
+		return HAS_PCH_LPT_LP(display) ?
 			PORTA_HOTPLUG_ENABLE : 0;
 	case HPD_PORT_B:
 		return PORTB_HOTPLUG_ENABLE |
@@ -735,37 +731,37 @@ static u32 ibx_hotplug_enables(struct intel_encoder *encoder)
 	}
 }
 
-static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void ibx_hpd_detection_setup(struct intel_display *display)
 {
 	/*
 	 * Enable digital hotplug on the PCH, and configure the DP short pulse
 	 * duration to 2ms (which is the minimum in the Display Port spec).
 	 * The pulse duration bits are reserved on LPT+.
 	 */
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 intel_hpd_hotplug_mask(dev_priv, ibx_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG,
+		     intel_hpd_hotplug_mask(display, ibx_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, ibx_hotplug_enables));
 }
 
 static void ibx_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
-			 ibx_hotplug_mask(encoder->hpd_pin),
-			 ibx_hotplug_enables(encoder));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG,
+		     ibx_hotplug_mask(encoder->hpd_pin),
+		     ibx_hotplug_enables(encoder));
 }
 
-static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void ibx_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.pch_hpd);
 
-	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+	ibx_display_interrupt_update(display, hotplug_irqs, enabled_irqs);
 
-	ibx_hpd_detection_setup(dev_priv);
+	ibx_hpd_detection_setup(display);
 }
 
 static u32 icp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
@@ -806,36 +802,36 @@ static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder)
 	return icp_tc_hotplug_mask(encoder->hpd_pin);
 }
 
-static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void icp_ddi_hpd_detection_setup(struct intel_display *display)
 {
-	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI,
-			 intel_hpd_hotplug_mask(dev_priv, icp_ddi_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables));
+	intel_de_rmw(display, SHOTPLUG_CTL_DDI,
+		     intel_hpd_hotplug_mask(display, icp_ddi_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, icp_ddi_hotplug_enables));
 }
 
 static void icp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_DDI,
-			 icp_ddi_hotplug_mask(encoder->hpd_pin),
-			 icp_ddi_hotplug_enables(encoder));
+	intel_de_rmw(display, SHOTPLUG_CTL_DDI,
+		     icp_ddi_hotplug_mask(encoder->hpd_pin),
+		     icp_ddi_hotplug_enables(encoder));
 }
 
-static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void icp_tc_hpd_detection_setup(struct intel_display *display)
 {
-	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC,
-			 intel_hpd_hotplug_mask(dev_priv, icp_tc_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables));
+	intel_de_rmw(display, SHOTPLUG_CTL_TC,
+		     intel_hpd_hotplug_mask(display, icp_tc_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, icp_tc_hotplug_enables));
 }
 
 static void icp_tc_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_TC,
-			 icp_tc_hotplug_mask(encoder->hpd_pin),
-			 icp_tc_hotplug_enables(encoder));
+	intel_de_rmw(display, SHOTPLUG_CTL_TC,
+		     icp_tc_hotplug_mask(encoder->hpd_pin),
+		     icp_tc_hotplug_enables(encoder));
 }
 
 static void icp_hpd_enable_detection(struct intel_encoder *encoder)
@@ -844,23 +840,23 @@ static void icp_hpd_enable_detection(struct intel_encoder *encoder)
 	icp_tc_hpd_enable_detection(encoder);
 }
 
-static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void icp_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.pch_hpd);
 
 	/*
 	 * We reduce the value to 250us to be able to detect SHPD when an external display
 	 * is connected. This is also expected of us as stated in DP1.4a Table 3-4.
 	 */
-	intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_250);
+	intel_de_write(display, SHPD_FILTER_CNT, SHPD_FILTER_CNT_250);
 
-	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+	ibx_display_interrupt_update(display, hotplug_irqs, enabled_irqs);
 
-	icp_ddi_hpd_detection_setup(dev_priv);
-	icp_tc_hpd_detection_setup(dev_priv);
+	icp_ddi_hpd_detection_setup(display);
+	icp_tc_hpd_detection_setup(display);
 }
 
 static u32 gen11_hotplug_mask(enum hpd_pin hpd_pin)
@@ -883,88 +879,88 @@ static u32 gen11_hotplug_enables(struct intel_encoder *encoder)
 	return gen11_hotplug_mask(encoder->hpd_pin);
 }
 
-static void dg1_hpd_invert(struct drm_i915_private *i915)
+static void dg1_hpd_invert(struct intel_display *display)
 {
 	u32 val = (INVERT_DDIA_HPD |
 		   INVERT_DDIB_HPD |
 		   INVERT_DDIC_HPD |
 		   INVERT_DDID_HPD);
-	intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, 0, val);
+	intel_de_rmw(display, SOUTH_CHICKEN1, 0, val);
 }
 
 static void dg1_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	dg1_hpd_invert(i915);
+	dg1_hpd_invert(display);
 	icp_hpd_enable_detection(encoder);
 }
 
-static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void dg1_hpd_irq_setup(struct intel_display *display)
 {
-	dg1_hpd_invert(dev_priv);
-	icp_hpd_irq_setup(dev_priv);
+	dg1_hpd_invert(display);
+	icp_hpd_irq_setup(display);
 }
 
-static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void gen11_tc_hpd_detection_setup(struct intel_display *display)
 {
-	intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL,
-			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
+	intel_de_rmw(display, GEN11_TC_HOTPLUG_CTL,
+		     intel_hpd_hotplug_mask(display, gen11_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, gen11_hotplug_enables));
 }
 
 static void gen11_tc_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, GEN11_TC_HOTPLUG_CTL,
-			 gen11_hotplug_mask(encoder->hpd_pin),
-			 gen11_hotplug_enables(encoder));
+	intel_de_rmw(display, GEN11_TC_HOTPLUG_CTL,
+		     gen11_hotplug_mask(encoder->hpd_pin),
+		     gen11_hotplug_enables(encoder));
 }
 
-static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void gen11_tbt_hpd_detection_setup(struct intel_display *display)
 {
-	intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL,
-			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
+	intel_de_rmw(display, GEN11_TBT_HOTPLUG_CTL,
+		     intel_hpd_hotplug_mask(display, gen11_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, gen11_hotplug_enables));
 }
 
 static void gen11_tbt_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, GEN11_TBT_HOTPLUG_CTL,
-			 gen11_hotplug_mask(encoder->hpd_pin),
-			 gen11_hotplug_enables(encoder));
+	intel_de_rmw(display, GEN11_TBT_HOTPLUG_CTL,
+		     gen11_hotplug_mask(encoder->hpd_pin),
+		     gen11_hotplug_enables(encoder));
 }
 
 static void gen11_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
 	gen11_tc_hpd_enable_detection(encoder);
 	gen11_tbt_hpd_enable_detection(encoder);
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
+	if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 		icp_hpd_enable_detection(encoder);
 }
 
-static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void gen11_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.hpd);
 
-	intel_uncore_rmw(&dev_priv->uncore, GEN11_DE_HPD_IMR, hotplug_irqs,
-			 ~enabled_irqs & hotplug_irqs);
-	intel_uncore_posting_read(&dev_priv->uncore, GEN11_DE_HPD_IMR);
+	intel_de_rmw(display, GEN11_DE_HPD_IMR, hotplug_irqs,
+		     ~enabled_irqs & hotplug_irqs);
+	intel_de_posting_read(display, GEN11_DE_HPD_IMR);
 
-	gen11_tc_hpd_detection_setup(dev_priv);
-	gen11_tbt_hpd_detection_setup(dev_priv);
+	gen11_tc_hpd_detection_setup(display);
+	gen11_tbt_hpd_detection_setup(display);
 
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-		icp_hpd_irq_setup(dev_priv);
+	if (INTEL_PCH_TYPE(display) >= PCH_ICP)
+		icp_hpd_irq_setup(display);
 }
 
 static u32 mtp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
@@ -1001,39 +997,39 @@ static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
 	return mtp_tc_hotplug_mask(encoder->hpd_pin);
 }
 
-static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
+static void mtp_ddi_hpd_detection_setup(struct intel_display *display)
 {
-	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
-		     intel_hpd_hotplug_mask(i915, mtp_ddi_hotplug_mask),
-		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
+	intel_de_rmw(display, SHOTPLUG_CTL_DDI,
+		     intel_hpd_hotplug_mask(display, mtp_ddi_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, mtp_ddi_hotplug_enables));
 }
 
 static void mtp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+	intel_de_rmw(display, SHOTPLUG_CTL_DDI,
 		     mtp_ddi_hotplug_mask(encoder->hpd_pin),
 		     mtp_ddi_hotplug_enables(encoder));
 }
 
-static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
+static void mtp_tc_hpd_detection_setup(struct intel_display *display)
 {
-	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
-		     intel_hpd_hotplug_mask(i915, mtp_tc_hotplug_mask),
-		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
+	intel_de_rmw(display, SHOTPLUG_CTL_TC,
+		     intel_hpd_hotplug_mask(display, mtp_tc_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, mtp_tc_hotplug_enables));
 }
 
 static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+	intel_de_rmw(display, SHOTPLUG_CTL_DDI,
 		     mtp_tc_hotplug_mask(encoder->hpd_pin),
 		     mtp_tc_hotplug_enables(encoder));
 }
 
-static void mtp_hpd_invert(struct drm_i915_private *i915)
+static void mtp_hpd_invert(struct intel_display *display)
 {
 	u32 val = (INVERT_DDIA_HPD |
 		   INVERT_DDIB_HPD |
@@ -1044,49 +1040,49 @@ static void mtp_hpd_invert(struct drm_i915_private *i915)
 		   INVERT_TC4_HPD |
 		   INVERT_DDID_HPD_MTP |
 		   INVERT_DDIE_HPD);
-	intel_de_rmw(i915, SOUTH_CHICKEN1, 0, val);
+	intel_de_rmw(display, SOUTH_CHICKEN1, 0, val);
 }
 
 static void mtp_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	mtp_hpd_invert(i915);
+	mtp_hpd_invert(display);
 	mtp_ddi_hpd_enable_detection(encoder);
 	mtp_tc_hpd_enable_detection(encoder);
 }
 
-static void mtp_hpd_irq_setup(struct drm_i915_private *i915)
+static void mtp_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.pch_hpd);
 
 	/*
 	 * Use 250us here to align with the DP1.4a(Table 3-4) spec as to what the
 	 * SHPD_FILTER_CNT value should be.
 	 */
-	intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_250);
+	intel_de_write(display, SHPD_FILTER_CNT, SHPD_FILTER_CNT_250);
 
-	mtp_hpd_invert(i915);
-	ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs);
+	mtp_hpd_invert(display);
+	ibx_display_interrupt_update(display, hotplug_irqs, enabled_irqs);
 
-	mtp_ddi_hpd_detection_setup(i915);
-	mtp_tc_hpd_detection_setup(i915);
+	mtp_ddi_hpd_detection_setup(display);
+	mtp_tc_hpd_detection_setup(display);
 }
 
-static void xe2lpd_sde_hpd_irq_setup(struct drm_i915_private *i915)
+static void xe2lpd_sde_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.pch_hpd);
 
-	ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs);
+	ibx_display_interrupt_update(display, hotplug_irqs, enabled_irqs);
 
-	mtp_ddi_hpd_detection_setup(i915);
-	mtp_tc_hpd_detection_setup(i915);
+	mtp_ddi_hpd_detection_setup(display);
+	mtp_tc_hpd_detection_setup(display);
 }
 
 static bool is_xelpdp_pica_hpd_pin(enum hpd_pin hpd_pin)
@@ -1094,7 +1090,7 @@ static bool is_xelpdp_pica_hpd_pin(enum hpd_pin hpd_pin)
 	return hpd_pin >= HPD_PORT_TC1 && hpd_pin <= HPD_PORT_TC4;
 }
 
-static void _xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915,
+static void _xelpdp_pica_hpd_detection_setup(struct intel_display *display,
 					     enum hpd_pin hpd_pin, bool enable)
 {
 	u32 mask = XELPDP_TBT_HOTPLUG_ENABLE |
@@ -1103,18 +1099,18 @@ static void _xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915,
 	if (!is_xelpdp_pica_hpd_pin(hpd_pin))
 		return;
 
-	intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(hpd_pin),
+	intel_de_rmw(display, XELPDP_PORT_HOTPLUG_CTL(hpd_pin),
 		     mask, enable ? mask : 0);
 }
 
 static void xelpdp_pica_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	_xelpdp_pica_hpd_detection_setup(i915, encoder->hpd_pin, true);
+	_xelpdp_pica_hpd_detection_setup(display, encoder->hpd_pin, true);
 }
 
-static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915)
+static void xelpdp_pica_hpd_detection_setup(struct intel_display *display)
 {
 	struct intel_encoder *encoder;
 	u32 available_pins = 0;
@@ -1122,11 +1118,11 @@ static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915)
 
 	BUILD_BUG_ON(BITS_PER_TYPE(available_pins) < HPD_NUM_PINS);
 
-	for_each_intel_encoder(&i915->drm, encoder)
+	for_each_intel_encoder(display->drm, encoder)
 		available_pins |= BIT(encoder->hpd_pin);
 
 	for_each_hpd_pin(pin)
-		_xelpdp_pica_hpd_detection_setup(i915, pin, available_pins & BIT(pin));
+		_xelpdp_pica_hpd_detection_setup(display, pin, available_pins & BIT(pin));
 }
 
 static void xelpdp_hpd_enable_detection(struct intel_encoder *encoder)
@@ -1135,23 +1131,23 @@ static void xelpdp_hpd_enable_detection(struct intel_encoder *encoder)
 	mtp_hpd_enable_detection(encoder);
 }
 
-static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915)
+static void xelpdp_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.hpd);
 
-	intel_de_rmw(i915, PICAINTERRUPT_IMR, hotplug_irqs,
+	intel_de_rmw(display, PICAINTERRUPT_IMR, hotplug_irqs,
 		     ~enabled_irqs & hotplug_irqs);
-	intel_uncore_posting_read(&i915->uncore, PICAINTERRUPT_IMR);
+	intel_de_posting_read(display, PICAINTERRUPT_IMR);
 
-	xelpdp_pica_hpd_detection_setup(i915);
+	xelpdp_pica_hpd_detection_setup(display);
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_LNL)
-		xe2lpd_sde_hpd_irq_setup(i915);
-	else if (INTEL_PCH_TYPE(i915) >= PCH_MTL)
-		mtp_hpd_irq_setup(i915);
+	if (INTEL_PCH_TYPE(display) >= PCH_LNL)
+		xe2lpd_sde_hpd_irq_setup(display);
+	else if (INTEL_PCH_TYPE(display) >= PCH_MTL)
+		mtp_hpd_irq_setup(display);
 }
 
 static u32 spt_hotplug_mask(enum hpd_pin hpd_pin)
@@ -1190,57 +1186,57 @@ static u32 spt_hotplug2_enables(struct intel_encoder *encoder)
 	return spt_hotplug2_mask(encoder->hpd_pin);
 }
 
-static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void spt_hpd_detection_setup(struct intel_display *display)
 {
 	/* Display WA #1179 WaHardHangonHotPlug: cnp */
-	if (HAS_PCH_CNP(dev_priv)) {
-		intel_uncore_rmw(&dev_priv->uncore, SOUTH_CHICKEN1, CHASSIS_CLK_REQ_DURATION_MASK,
-				 CHASSIS_CLK_REQ_DURATION(0xf));
+	if (HAS_PCH_CNP(display)) {
+		intel_de_rmw(display, SOUTH_CHICKEN1, CHASSIS_CLK_REQ_DURATION_MASK,
+			     CHASSIS_CLK_REQ_DURATION(0xf));
 	}
 
 	/* Enable digital hotplug on the PCH */
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG,
+		     intel_hpd_hotplug_mask(display, spt_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, spt_hotplug_enables));
 
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2,
-			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug2_mask),
-			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG2,
+		     intel_hpd_hotplug_mask(display, spt_hotplug2_mask),
+		     intel_hpd_hotplug_enables(display, spt_hotplug2_enables));
 }
 
 static void spt_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
 	/* Display WA #1179 WaHardHangonHotPlug: cnp */
-	if (HAS_PCH_CNP(i915)) {
-		intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1,
-				 CHASSIS_CLK_REQ_DURATION_MASK,
-				 CHASSIS_CLK_REQ_DURATION(0xf));
+	if (HAS_PCH_CNP(display)) {
+		intel_de_rmw(display, SOUTH_CHICKEN1,
+			     CHASSIS_CLK_REQ_DURATION_MASK,
+			     CHASSIS_CLK_REQ_DURATION(0xf));
 	}
 
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
-			 spt_hotplug_mask(encoder->hpd_pin),
-			 spt_hotplug_enables(encoder));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG,
+		     spt_hotplug_mask(encoder->hpd_pin),
+		     spt_hotplug_enables(encoder));
 
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG2,
-			 spt_hotplug2_mask(encoder->hpd_pin),
-			 spt_hotplug2_enables(encoder));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG2,
+		     spt_hotplug2_mask(encoder->hpd_pin),
+		     spt_hotplug2_enables(encoder));
 }
 
-static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void spt_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
-		intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
+	if (INTEL_PCH_TYPE(display) >= PCH_CNP)
+		intel_de_write(display, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
 
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.pch_hpd);
 
-	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+	ibx_display_interrupt_update(display, hotplug_irqs, enabled_irqs);
 
-	spt_hpd_detection_setup(dev_priv);
+	spt_hpd_detection_setup(display);
 }
 
 static u32 ilk_hotplug_mask(enum hpd_pin hpd_pin)
@@ -1265,44 +1261,44 @@ static u32 ilk_hotplug_enables(struct intel_encoder *encoder)
 	}
 }
 
-static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void ilk_hpd_detection_setup(struct intel_display *display)
 {
 	/*
 	 * Enable digital hotplug on the CPU, and configure the DP short pulse
 	 * duration to 2ms (which is the minimum in the Display Port spec)
 	 * The pulse duration bits are reserved on HSW+.
 	 */
-	intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
-			 intel_hpd_hotplug_mask(dev_priv, ilk_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables));
+	intel_de_rmw(display, DIGITAL_PORT_HOTPLUG_CNTRL,
+		     intel_hpd_hotplug_mask(display, ilk_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, ilk_hotplug_enables));
 }
 
 static void ilk_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
-			 ilk_hotplug_mask(encoder->hpd_pin),
-			 ilk_hotplug_enables(encoder));
+	intel_de_rmw(display, DIGITAL_PORT_HOTPLUG_CNTRL,
+		     ilk_hotplug_mask(encoder->hpd_pin),
+		     ilk_hotplug_enables(encoder));
 
 	ibx_hpd_enable_detection(encoder);
 }
 
-static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void ilk_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.hpd);
 
-	if (DISPLAY_VER(dev_priv) >= 8)
-		bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
+	if (DISPLAY_VER(display) >= 8)
+		bdw_update_port_irq(display, hotplug_irqs, enabled_irqs);
 	else
-		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
+		ilk_update_display_irq(display, hotplug_irqs, enabled_irqs);
 
-	ilk_hpd_detection_setup(dev_priv);
+	ilk_hpd_detection_setup(display);
 
-	ibx_hpd_irq_setup(dev_priv);
+	ibx_hpd_irq_setup(display);
 }
 
 static u32 bxt_hotplug_mask(enum hpd_pin hpd_pin)
@@ -1344,80 +1340,80 @@ static u32 bxt_hotplug_enables(struct intel_encoder *encoder)
 	}
 }
 
-static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv)
+static void bxt_hpd_detection_setup(struct intel_display *display)
 {
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 intel_hpd_hotplug_mask(dev_priv, bxt_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG,
+		     intel_hpd_hotplug_mask(display, bxt_hotplug_mask),
+		     intel_hpd_hotplug_enables(display, bxt_hotplug_enables));
 }
 
 static void bxt_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
-			 bxt_hotplug_mask(encoder->hpd_pin),
-			 bxt_hotplug_enables(encoder));
+	intel_de_rmw(display, PCH_PORT_HOTPLUG,
+		     bxt_hotplug_mask(encoder->hpd_pin),
+		     bxt_hotplug_enables(encoder));
 }
 
-static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void bxt_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_irqs, enabled_irqs;
 
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+	enabled_irqs = intel_hpd_enabled_irqs(display, display->hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(display, display->hotplug.hpd);
 
-	bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
+	bdw_update_port_irq(display, hotplug_irqs, enabled_irqs);
 
-	bxt_hpd_detection_setup(dev_priv);
+	bxt_hpd_detection_setup(display);
 }
 
-static void g45_hpd_peg_band_gap_wa(struct drm_i915_private *i915)
+static void g45_hpd_peg_band_gap_wa(struct intel_display *display)
 {
 	/*
 	 * For G4X desktop chip, PEG_BAND_GAP_DATA 3:0 must first be written
 	 * 0xd.  Failure to do so will result in spurious interrupts being
 	 * generated on the port when a cable is not attached.
 	 */
-	intel_de_rmw(i915, PEG_BAND_GAP_DATA, 0xf, 0xd);
+	intel_de_rmw(display, PEG_BAND_GAP_DATA, 0xf, 0xd);
 }
 
 static void i915_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	u32 hotplug_en = hpd_mask_i915[encoder->hpd_pin];
 
-	if (IS_G45(i915))
-		g45_hpd_peg_band_gap_wa(i915);
+	if (display->platform.g45)
+		g45_hpd_peg_band_gap_wa(display);
 
 	/* HPD sense and interrupt enable are one and the same */
-	i915_hotplug_interrupt_update(i915, hotplug_en, hotplug_en);
+	i915_hotplug_interrupt_update(display, hotplug_en, hotplug_en);
 }
 
-static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv)
+static void i915_hpd_irq_setup(struct intel_display *display)
 {
 	u32 hotplug_en;
 
-	lockdep_assert_held(&dev_priv->irq_lock);
+	lockdep_assert_held(&display->irq.lock);
 
 	/*
 	 * Note HDMI and DP share hotplug bits. Enable bits are the same for all
 	 * generations.
 	 */
-	hotplug_en = intel_hpd_enabled_irqs(dev_priv, hpd_mask_i915);
+	hotplug_en = intel_hpd_enabled_irqs(display, hpd_mask_i915);
 	/*
 	 * Programming the CRT detection parameters tends to generate a spurious
 	 * hotplug event about three seconds later. So just do it once.
 	 */
-	if (IS_G4X(dev_priv))
+	if (display->platform.g4x)
 		hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64;
 	hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50;
 
-	if (IS_G45(dev_priv))
-		g45_hpd_peg_band_gap_wa(dev_priv);
+	if (display->platform.g45)
+		g45_hpd_peg_band_gap_wa(display);
 
 	/* Ignore TV since it's buggy */
-	i915_hotplug_interrupt_update_locked(dev_priv,
+	i915_hotplug_interrupt_update_locked(display,
 					     HOTPLUG_INT_EN_MASK |
 					     CRT_HOTPLUG_VOLTAGE_COMPARE_MASK |
 					     CRT_HOTPLUG_ACTIVATION_PERIOD_64,
@@ -1426,7 +1422,7 @@ static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv)
 
 struct intel_hotplug_funcs {
 	/* Enable HPD sense and interrupts for all present encoders */
-	void (*hpd_irq_setup)(struct drm_i915_private *i915);
+	void (*hpd_irq_setup)(struct intel_display *display);
 	/* Enable HPD sense for a single encoder */
 	void (*hpd_enable_detection)(struct intel_encoder *encoder);
 };
@@ -1449,47 +1445,47 @@ HPD_FUNCS(ilk);
 
 void intel_hpd_enable_detection(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (i915->display.funcs.hotplug)
-		i915->display.funcs.hotplug->hpd_enable_detection(encoder);
+	if (display->funcs.hotplug)
+		display->funcs.hotplug->hpd_enable_detection(encoder);
 }
 
-void intel_hpd_irq_setup(struct drm_i915_private *i915)
+void intel_hpd_irq_setup(struct intel_display *display)
 {
-	if ((IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) &&
-	    !i915->display.irq.vlv_display_irqs_enabled)
+	if ((display->platform.valleyview || display->platform.cherryview) &&
+	    !display->irq.vlv_display_irqs_enabled)
 		return;
 
-	if (i915->display.funcs.hotplug)
-		i915->display.funcs.hotplug->hpd_irq_setup(i915);
+	if (display->funcs.hotplug)
+		display->funcs.hotplug->hpd_irq_setup(display);
 }
 
-void intel_hotplug_irq_init(struct drm_i915_private *i915)
+void intel_hotplug_irq_init(struct intel_display *display)
 {
-	intel_hpd_init_pins(i915);
+	intel_hpd_init_pins(display);
 
-	intel_hpd_init_early(i915);
+	intel_hpd_init_early(display);
 
-	if (HAS_GMCH(i915)) {
-		if (I915_HAS_HOTPLUG(i915))
-			i915->display.funcs.hotplug = &i915_hpd_funcs;
+	if (HAS_GMCH(display)) {
+		if (HAS_HOTPLUG(display))
+			display->funcs.hotplug = &i915_hpd_funcs;
 	} else {
-		if (HAS_PCH_DG2(i915))
-			i915->display.funcs.hotplug = &icp_hpd_funcs;
-		else if (HAS_PCH_DG1(i915))
-			i915->display.funcs.hotplug = &dg1_hpd_funcs;
-		else if (DISPLAY_VER(i915) >= 14)
-			i915->display.funcs.hotplug = &xelpdp_hpd_funcs;
-		else if (DISPLAY_VER(i915) >= 11)
-			i915->display.funcs.hotplug = &gen11_hpd_funcs;
-		else if (IS_GEMINILAKE(i915) || IS_BROXTON(i915))
-			i915->display.funcs.hotplug = &bxt_hpd_funcs;
-		else if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
-			i915->display.funcs.hotplug = &icp_hpd_funcs;
-		else if (INTEL_PCH_TYPE(i915) >= PCH_SPT)
-			i915->display.funcs.hotplug = &spt_hpd_funcs;
+		if (HAS_PCH_DG2(display))
+			display->funcs.hotplug = &icp_hpd_funcs;
+		else if (HAS_PCH_DG1(display))
+			display->funcs.hotplug = &dg1_hpd_funcs;
+		else if (DISPLAY_VER(display) >= 14)
+			display->funcs.hotplug = &xelpdp_hpd_funcs;
+		else if (DISPLAY_VER(display) >= 11)
+			display->funcs.hotplug = &gen11_hpd_funcs;
+		else if (display->platform.geminilake || display->platform.broxton)
+			display->funcs.hotplug = &bxt_hpd_funcs;
+		else if (INTEL_PCH_TYPE(display) >= PCH_ICP)
+			display->funcs.hotplug = &icp_hpd_funcs;
+		else if (INTEL_PCH_TYPE(display) >= PCH_SPT)
+			display->funcs.hotplug = &spt_hpd_funcs;
 		else
-			i915->display.funcs.hotplug = &ilk_hpd_funcs;
+			display->funcs.hotplug = &ilk_hpd_funcs;
 	}
 }
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.h b/drivers/gpu/drm/i915/display/intel_hotplug_irq.h
index e4db752df096..9063bb02a2e9 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.h
+++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.h
@@ -8,28 +8,28 @@
 
 #include <linux/types.h>
 
-struct drm_i915_private;
+struct intel_display;
 struct intel_encoder;
 
-u32 i9xx_hpd_irq_ack(struct drm_i915_private *i915);
+u32 i9xx_hpd_irq_ack(struct intel_display *display);
 
-void i9xx_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_status);
-void ibx_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger);
-void ilk_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger);
-void gen11_hpd_irq_handler(struct drm_i915_private *i915, u32 iir);
-void bxt_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger);
-void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir);
-void icp_irq_handler(struct drm_i915_private *i915, u32 pch_iir);
-void spt_irq_handler(struct drm_i915_private *i915, u32 pch_iir);
+void i9xx_hpd_irq_handler(struct intel_display *display, u32 hotplug_status);
+void ibx_hpd_irq_handler(struct intel_display *display, u32 hotplug_trigger);
+void ilk_hpd_irq_handler(struct intel_display *display, u32 hotplug_trigger);
+void gen11_hpd_irq_handler(struct intel_display *display, u32 iir);
+void bxt_hpd_irq_handler(struct intel_display *display, u32 hotplug_trigger);
+void xelpdp_pica_irq_handler(struct intel_display *display, u32 iir);
+void icp_irq_handler(struct intel_display *display, u32 pch_iir);
+void spt_irq_handler(struct intel_display *display, u32 pch_iir);
 
-void i915_hotplug_interrupt_update_locked(struct drm_i915_private *i915,
+void i915_hotplug_interrupt_update_locked(struct intel_display *display,
 					  u32 mask, u32 bits);
-void i915_hotplug_interrupt_update(struct drm_i915_private *i915,
+void i915_hotplug_interrupt_update(struct intel_display *display,
 				   u32 mask, u32 bits);
 
 void intel_hpd_enable_detection(struct intel_encoder *encoder);
-void intel_hpd_irq_setup(struct drm_i915_private *i915);
+void intel_hpd_irq_setup(struct intel_display *display);
 
-void intel_hotplug_irq_init(struct drm_i915_private *i915);
+void intel_hotplug_irq_init(struct intel_display *display);
 
 #endif /* __INTEL_HOTPLUG_IRQ_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_hti.c b/drivers/gpu/drm/i915/display/intel_hti.c
index fb6b84f6a81d..dc454420c134 100644
--- a/drivers/gpu/drm/i915/display/intel_hti.c
+++ b/drivers/gpu/drm/i915/display/intel_hti.c
@@ -4,6 +4,7 @@
  */
 
 #include <drm/drm_device.h>
+#include <drm/drm_print.h>
 
 #include "intel_de.h"
 #include "intel_display.h"
diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.c b/drivers/gpu/drm/i915/display/intel_load_detect.c
index 86cc03a4413c..aad52d0d83e1 100644
--- a/drivers/gpu/drm/i915/display/intel_load_detect.c
+++ b/drivers/gpu/drm/i915/display/intel_load_detect.c
@@ -6,6 +6,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_atomic_uapi.h>
+#include <drm/drm_print.h>
 
 #include "intel_atomic.h"
 #include "intel_crtc.h"
diff --git a/drivers/gpu/drm/i915/display/intel_lpe_audio.c b/drivers/gpu/drm/i915/display/intel_lpe_audio.c
index 59551c8414c2..666148a14522 100644
--- a/drivers/gpu/drm/i915/display/intel_lpe_audio.c
+++ b/drivers/gpu/drm/i915/display/intel_lpe_audio.c
@@ -179,7 +179,7 @@ static int lpe_audio_irq_init(struct intel_display *display)
 				handle_simple_irq,
 				"hdmi_lpe_audio_irq_handler");
 
-	return irq_set_chip_data(irq, dev_priv);
+	return 0;
 }
 
 static bool lpe_audio_detect(struct intel_display *display)
diff --git a/drivers/gpu/drm/i915/display/intel_lspcon.c b/drivers/gpu/drm/i915/display/intel_lspcon.c
index 63c1afa30b05..f94b7eeae20f 100644
--- a/drivers/gpu/drm/i915/display/intel_lspcon.c
+++ b/drivers/gpu/drm/i915/display/intel_lspcon.c
@@ -27,6 +27,7 @@
 #include <drm/display/drm_hdmi_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 
 #include "i915_reg.h"
 #include "i915_utils.h"
diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c
index 19f52d1659fa..8ce7c630da52 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.c
+++ b/drivers/gpu/drm/i915/display/intel_lvds.c
@@ -37,9 +37,9 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
 #include "intel_backlight.h"
@@ -84,15 +84,15 @@ static struct intel_lvds_encoder *to_lvds_encoder(struct intel_encoder *encoder)
 	return container_of(encoder, struct intel_lvds_encoder, base);
 }
 
-bool intel_lvds_port_enabled(struct drm_i915_private *i915,
+bool intel_lvds_port_enabled(struct intel_display *display,
 			     i915_reg_t lvds_reg, enum pipe *pipe)
 {
 	u32 val;
 
-	val = intel_de_read(i915, lvds_reg);
+	val = intel_de_read(display, lvds_reg);
 
 	/* asserts want to know the pipe even if the port is disabled */
-	if (HAS_PCH_CPT(i915))
+	if (HAS_PCH_CPT(display))
 		*pipe = REG_FIELD_GET(LVDS_PIPE_SEL_MASK_CPT, val);
 	else
 		*pipe = REG_FIELD_GET(LVDS_PIPE_SEL_MASK, val);
@@ -104,7 +104,6 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 				    enum pipe *pipe)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(encoder);
 	intel_wakeref_t wakeref;
 	bool ret;
@@ -113,7 +112,7 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 	if (!wakeref)
 		return false;
 
-	ret = intel_lvds_port_enabled(i915, lvds_encoder->reg, pipe);
+	ret = intel_lvds_port_enabled(display, lvds_encoder->reg, pipe);
 
 	intel_display_power_put(display, encoder->power_domain, wakeref);
 
@@ -123,13 +122,13 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 static void intel_lvds_get_config(struct intel_encoder *encoder,
 				  struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(encoder);
 	u32 tmp, flags = 0;
 
 	crtc_state->output_types |= BIT(INTEL_OUTPUT_LVDS);
 
-	tmp = intel_de_read(dev_priv, lvds_encoder->reg);
+	tmp = intel_de_read(display, lvds_encoder->reg);
 	if (tmp & LVDS_HSYNC_POLARITY)
 		flags |= DRM_MODE_FLAG_NHSYNC;
 	else
@@ -141,13 +140,13 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 
 	crtc_state->hw.adjusted_mode.flags |= flags;
 
-	if (DISPLAY_VER(dev_priv) < 5)
+	if (DISPLAY_VER(display) < 5)
 		crtc_state->gmch_pfit.lvds_border_bits =
 			tmp & LVDS_BORDER_ENABLE;
 
 	/* gen2/3 store dither state in pfit control, needs to match */
-	if (DISPLAY_VER(dev_priv) < 4) {
-		tmp = intel_de_read(dev_priv, PFIT_CONTROL(dev_priv));
+	if (DISPLAY_VER(display) < 4) {
+		tmp = intel_de_read(display, PFIT_CONTROL(display));
 
 		crtc_state->gmch_pfit.control |= tmp & PFIT_PANEL_8TO6_DITHER_ENABLE;
 	}
@@ -155,24 +154,24 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 	crtc_state->hw.adjusted_mode.crtc_clock = crtc_state->port_clock;
 }
 
-static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv,
+static void intel_lvds_pps_get_hw_state(struct intel_display *display,
 					struct intel_lvds_pps *pps)
 {
 	u32 val;
 
-	pps->powerdown_on_reset = intel_de_read(dev_priv,
-						PP_CONTROL(dev_priv, 0)) & PANEL_POWER_RESET;
+	pps->powerdown_on_reset = intel_de_read(display,
+						PP_CONTROL(display, 0)) & PANEL_POWER_RESET;
 
-	val = intel_de_read(dev_priv, PP_ON_DELAYS(dev_priv, 0));
+	val = intel_de_read(display, PP_ON_DELAYS(display, 0));
 	pps->port = REG_FIELD_GET(PANEL_PORT_SELECT_MASK, val);
 	pps->delays.power_up = REG_FIELD_GET(PANEL_POWER_UP_DELAY_MASK, val);
 	pps->delays.backlight_on = REG_FIELD_GET(PANEL_LIGHT_ON_DELAY_MASK, val);
 
-	val = intel_de_read(dev_priv, PP_OFF_DELAYS(dev_priv, 0));
+	val = intel_de_read(display, PP_OFF_DELAYS(display, 0));
 	pps->delays.power_down = REG_FIELD_GET(PANEL_POWER_DOWN_DELAY_MASK, val);
 	pps->delays.backlight_off = REG_FIELD_GET(PANEL_LIGHT_OFF_DELAY_MASK, val);
 
-	val = intel_de_read(dev_priv, PP_DIVISOR(dev_priv, 0));
+	val = intel_de_read(display, PP_DIVISOR(display, 0));
 	pps->divider = REG_FIELD_GET(PP_REFERENCE_DIVIDER_MASK, val);
 	val = REG_FIELD_GET(PANEL_POWER_CYCLE_DELAY_MASK, val);
 	/*
@@ -185,12 +184,12 @@ static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv,
 	/* Convert from 100ms to 100us units */
 	pps->delays.power_cycle = val * 1000;
 
-	if (DISPLAY_VER(dev_priv) < 5 &&
+	if (DISPLAY_VER(display) < 5 &&
 	    pps->delays.power_up == 0 &&
 	    pps->delays.backlight_on == 0 &&
 	    pps->delays.power_down == 0 &&
 	    pps->delays.backlight_off == 0) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Panel power timings uninitialized, "
 			    "setting defaults\n");
 		/* Set T2 to 40ms and T5 to 200ms in 100 usec units */
@@ -201,7 +200,7 @@ static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv,
 		pps->delays.backlight_off = 200 * 10;
 	}
 
-	drm_dbg(&dev_priv->drm, "LVDS PPS:power_up %d power_down %d power_cycle %d backlight_on %d backlight_off %d "
+	drm_dbg(display->drm, "LVDS PPS:power_up %d power_down %d power_cycle %d backlight_on %d backlight_off %d "
 		"divider %d port %d powerdown_on_reset %d\n",
 		pps->delays.power_up, pps->delays.power_down,
 		pps->delays.power_cycle, pps->delays.backlight_on,
@@ -209,28 +208,28 @@ static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv,
 		pps->port, pps->powerdown_on_reset);
 }
 
-static void intel_lvds_pps_init_hw(struct drm_i915_private *dev_priv,
+static void intel_lvds_pps_init_hw(struct intel_display *display,
 				   struct intel_lvds_pps *pps)
 {
 	u32 val;
 
-	val = intel_de_read(dev_priv, PP_CONTROL(dev_priv, 0));
-	drm_WARN_ON(&dev_priv->drm,
+	val = intel_de_read(display, PP_CONTROL(display, 0));
+	drm_WARN_ON(display->drm,
 		    (val & PANEL_UNLOCK_MASK) != PANEL_UNLOCK_REGS);
 	if (pps->powerdown_on_reset)
 		val |= PANEL_POWER_RESET;
-	intel_de_write(dev_priv, PP_CONTROL(dev_priv, 0), val);
+	intel_de_write(display, PP_CONTROL(display, 0), val);
 
-	intel_de_write(dev_priv, PP_ON_DELAYS(dev_priv, 0),
+	intel_de_write(display, PP_ON_DELAYS(display, 0),
 		       REG_FIELD_PREP(PANEL_PORT_SELECT_MASK, pps->port) |
 		       REG_FIELD_PREP(PANEL_POWER_UP_DELAY_MASK, pps->delays.power_up) |
 		       REG_FIELD_PREP(PANEL_LIGHT_ON_DELAY_MASK, pps->delays.backlight_on));
 
-	intel_de_write(dev_priv, PP_OFF_DELAYS(dev_priv, 0),
+	intel_de_write(display, PP_OFF_DELAYS(display, 0),
 		       REG_FIELD_PREP(PANEL_POWER_DOWN_DELAY_MASK, pps->delays.power_down) |
 		       REG_FIELD_PREP(PANEL_LIGHT_OFF_DELAY_MASK, pps->delays.backlight_off));
 
-	intel_de_write(dev_priv, PP_DIVISOR(dev_priv, 0),
+	intel_de_write(display, PP_DIVISOR(display, 0),
 		       REG_FIELD_PREP(PP_REFERENCE_DIVIDER_MASK, pps->divider) |
 		       REG_FIELD_PREP(PANEL_POWER_CYCLE_DELAY_MASK,
 				      DIV_ROUND_UP(pps->delays.power_cycle, 1000) + 1));
@@ -243,25 +242,24 @@ static void intel_pre_enable_lvds(struct intel_atomic_state *state,
 {
 	struct intel_display *display = to_intel_display(state);
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
 	enum pipe pipe = crtc->pipe;
 	u32 temp;
 
-	if (HAS_PCH_SPLIT(i915)) {
+	if (HAS_PCH_SPLIT(display)) {
 		assert_fdi_rx_pll_disabled(display, pipe);
 		assert_shared_dpll_disabled(display, crtc_state->shared_dpll);
 	} else {
 		assert_pll_disabled(display, pipe);
 	}
 
-	intel_lvds_pps_init_hw(i915, &lvds_encoder->init_pps);
+	intel_lvds_pps_init_hw(display, &lvds_encoder->init_pps);
 
 	temp = lvds_encoder->init_lvds_val;
 	temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP;
 
-	if (HAS_PCH_CPT(i915)) {
+	if (HAS_PCH_CPT(display)) {
 		temp &= ~LVDS_PIPE_SEL_MASK_CPT;
 		temp |= LVDS_PIPE_SEL_CPT(pipe);
 	} else {
@@ -296,7 +294,7 @@ static void intel_pre_enable_lvds(struct intel_atomic_state *state,
 	 * special lvds dither control bit on pch-split platforms, dithering is
 	 * only controlled through the TRANSCONF reg.
 	 */
-	if (DISPLAY_VER(i915) == 4) {
+	if (DISPLAY_VER(display) == 4) {
 		/*
 		 * Bspec wording suggests that LVDS port dithering only exists
 		 * for 18bpp panels.
@@ -312,7 +310,7 @@ static void intel_pre_enable_lvds(struct intel_atomic_state *state,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC)
 		temp |= LVDS_VSYNC_POLARITY;
 
-	intel_de_write(i915, lvds_encoder->reg, temp);
+	intel_de_write(display, lvds_encoder->reg, temp);
 }
 
 /*
@@ -323,16 +321,16 @@ static void intel_enable_lvds(struct intel_atomic_state *state,
 			      const struct intel_crtc_state *crtc_state,
 			      const struct drm_connector_state *conn_state)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
-	intel_de_rmw(dev_priv, lvds_encoder->reg, 0, LVDS_PORT_EN);
+	intel_de_rmw(display, lvds_encoder->reg, 0, LVDS_PORT_EN);
 
-	intel_de_rmw(dev_priv, PP_CONTROL(dev_priv, 0), 0, PANEL_POWER_ON);
-	intel_de_posting_read(dev_priv, lvds_encoder->reg);
+	intel_de_rmw(display, PP_CONTROL(display, 0), 0, PANEL_POWER_ON);
+	intel_de_posting_read(display, lvds_encoder->reg);
 
-	if (intel_de_wait_for_set(dev_priv, PP_STATUS(dev_priv, 0), PP_ON, 5000))
-		drm_err(&dev_priv->drm,
+	if (intel_de_wait_for_set(display, PP_STATUS(display, 0), PP_ON, 5000))
+		drm_err(display->drm,
 			"timed out waiting for panel to power on\n");
 
 	intel_backlight_enable(crtc_state, conn_state);
@@ -343,16 +341,16 @@ static void intel_disable_lvds(struct intel_atomic_state *state,
 			       const struct intel_crtc_state *old_crtc_state,
 			       const struct drm_connector_state *old_conn_state)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
-	intel_de_rmw(dev_priv, PP_CONTROL(dev_priv, 0), PANEL_POWER_ON, 0);
-	if (intel_de_wait_for_clear(dev_priv, PP_STATUS(dev_priv, 0), PP_ON, 1000))
-		drm_err(&dev_priv->drm,
+	intel_de_rmw(display, PP_CONTROL(display, 0), PANEL_POWER_ON, 0);
+	if (intel_de_wait_for_clear(display, PP_STATUS(display, 0), PP_ON, 1000))
+		drm_err(display->drm,
 			"timed out waiting for panel to power off\n");
 
-	intel_de_rmw(dev_priv, lvds_encoder->reg, LVDS_PORT_EN, 0);
-	intel_de_posting_read(dev_priv, lvds_encoder->reg);
+	intel_de_rmw(display, lvds_encoder->reg, LVDS_PORT_EN, 0);
+	intel_de_posting_read(display, lvds_encoder->reg);
 }
 
 static void gmch_disable_lvds(struct intel_atomic_state *state,
@@ -384,10 +382,10 @@ static void pch_post_disable_lvds(struct intel_atomic_state *state,
 
 static void intel_lvds_shutdown(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (intel_de_wait_for_clear(dev_priv, PP_STATUS(dev_priv, 0), PP_CYCLE_DELAY_ACTIVE, 5000))
-		drm_err(&dev_priv->drm,
+	if (intel_de_wait_for_clear(display, PP_STATUS(display, 0), PP_CYCLE_DELAY_ACTIVE, 5000))
+		drm_err(display->drm,
 			"timed out waiting for panel power cycle delay\n");
 }
 
@@ -420,7 +418,7 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder,
 				     struct intel_crtc_state *crtc_state,
 				     struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(encoder);
 	struct intel_connector *connector = lvds_encoder->attached_connector;
 	struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
@@ -429,12 +427,12 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder,
 	int ret;
 
 	/* Should never happen!! */
-	if (DISPLAY_VER(i915) < 4 && crtc->pipe == 0) {
-		drm_err(&i915->drm, "Can't support LVDS on pipe A\n");
+	if (DISPLAY_VER(display) < 4 && crtc->pipe == 0) {
+		drm_err(display->drm, "Can't support LVDS on pipe A\n");
 		return -EINVAL;
 	}
 
-	if (HAS_PCH_SPLIT(i915)) {
+	if (HAS_PCH_SPLIT(display)) {
 		crtc_state->has_pch_encoder = true;
 		if (!intel_fdi_compute_pipe_bpp(crtc_state))
 			return -EINVAL;
@@ -447,7 +445,7 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder,
 
 	/* TODO: Check crtc_state->max_link_bpp_x16 instead of bw_constrained */
 	if (lvds_bpp != crtc_state->pipe_bpp && !crtc_state->bw_constrained) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "forcing display bpp (was %d) to LVDS (%d)\n",
 			    crtc_state->pipe_bpp, lvds_bpp);
 		crtc_state->pipe_bpp = lvds_bpp;
@@ -775,11 +773,11 @@ static const struct dmi_system_id intel_dual_link_lvds[] = {
 	{ }	/* terminating entry */
 };
 
-struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *i915)
+struct intel_encoder *intel_get_lvds_encoder(struct intel_display *display)
 {
 	struct intel_encoder *encoder;
 
-	for_each_intel_encoder(&i915->drm, encoder) {
+	for_each_intel_encoder(display->drm, encoder) {
 		if (encoder->type == INTEL_OUTPUT_LVDS)
 			return encoder;
 	}
@@ -787,24 +785,24 @@ struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *i915)
 	return NULL;
 }
 
-bool intel_is_dual_link_lvds(struct drm_i915_private *i915)
+bool intel_is_dual_link_lvds(struct intel_display *display)
 {
-	struct intel_encoder *encoder = intel_get_lvds_encoder(i915);
+	struct intel_encoder *encoder = intel_get_lvds_encoder(display);
 
 	return encoder && to_lvds_encoder(encoder)->is_dual_link;
 }
 
 static bool compute_is_dual_link_lvds(struct intel_lvds_encoder *lvds_encoder)
 {
-	struct drm_i915_private *i915 = to_i915(lvds_encoder->base.base.dev);
+	struct intel_display *display = to_intel_display(&lvds_encoder->base);
 	struct intel_connector *connector = lvds_encoder->attached_connector;
 	const struct drm_display_mode *fixed_mode =
 		intel_panel_preferred_fixed_mode(connector);
 	unsigned int val;
 
 	/* use the module option value if specified */
-	if (i915->display.params.lvds_channel_mode > 0)
-		return i915->display.params.lvds_channel_mode == 2;
+	if (display->params.lvds_channel_mode > 0)
+		return display->params.lvds_channel_mode == 2;
 
 	/* single channel LVDS is limited to 112 MHz */
 	if (fixed_mode->clock > 112999)
@@ -819,8 +817,8 @@ static bool compute_is_dual_link_lvds(struct intel_lvds_encoder *lvds_encoder)
 	 * we need to check "the value to be set" in VBT when LVDS
 	 * register is uninitialized.
 	 */
-	val = intel_de_read(i915, lvds_encoder->reg);
-	if (HAS_PCH_CPT(i915))
+	val = intel_de_read(display, lvds_encoder->reg);
+	if (HAS_PCH_CPT(display))
 		val &= ~(LVDS_DETECTED | LVDS_PIPE_SEL_MASK_CPT);
 	else
 		val &= ~(LVDS_DETECTED | LVDS_PIPE_SEL_MASK);
@@ -837,14 +835,13 @@ static void intel_lvds_add_properties(struct drm_connector *connector)
 
 /**
  * intel_lvds_init - setup LVDS connectors on this device
- * @i915: i915 device
+ * @display: display device
  *
  * Create the connector, register the LVDS DDC bus, and try to figure out what
  * modes we can display on the LVDS panel (if present).
  */
-void intel_lvds_init(struct drm_i915_private *i915)
+void intel_lvds_init(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_lvds_encoder *lvds_encoder;
 	struct intel_connector *connector;
 	const struct drm_edid *drm_edid;
@@ -855,25 +852,25 @@ void intel_lvds_init(struct drm_i915_private *i915)
 
 	/* Skip init on machines we know falsely report LVDS */
 	if (dmi_check_system(intel_no_lvds)) {
-		drm_WARN(&i915->drm, !i915->display.vbt.int_lvds_support,
+		drm_WARN(display->drm, !display->vbt.int_lvds_support,
 			 "Useless DMI match. Internal LVDS support disabled by VBT\n");
 		return;
 	}
 
-	if (!i915->display.vbt.int_lvds_support) {
-		drm_dbg_kms(&i915->drm,
+	if (!display->vbt.int_lvds_support) {
+		drm_dbg_kms(display->drm,
 			    "Internal LVDS support disabled by VBT\n");
 		return;
 	}
 
-	if (HAS_PCH_SPLIT(i915))
+	if (HAS_PCH_SPLIT(display))
 		lvds_reg = PCH_LVDS;
 	else
 		lvds_reg = LVDS;
 
-	lvds = intel_de_read(i915, lvds_reg);
+	lvds = intel_de_read(display, lvds_reg);
 
-	if (HAS_PCH_SPLIT(i915)) {
+	if (HAS_PCH_SPLIT(display)) {
 		if ((lvds & LVDS_DETECTED) == 0)
 			return;
 	}
@@ -881,11 +878,11 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	ddc_pin = GMBUS_PIN_PANEL;
 	if (!intel_bios_is_lvds_present(display, &ddc_pin)) {
 		if ((lvds & LVDS_PORT_EN) == 0) {
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "LVDS is not present in VBT\n");
 			return;
 		}
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "LVDS is not present in VBT, but enabled anyway\n");
 	}
 
@@ -902,18 +899,18 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	lvds_encoder->attached_connector = connector;
 	encoder = &lvds_encoder->base;
 
-	drm_connector_init_with_ddc(&i915->drm, &connector->base,
+	drm_connector_init_with_ddc(display->drm, &connector->base,
 				    &intel_lvds_connector_funcs,
 				    DRM_MODE_CONNECTOR_LVDS,
 				    intel_gmbus_get_adapter(display, ddc_pin));
 
-	drm_encoder_init(&i915->drm, &encoder->base, &intel_lvds_enc_funcs,
+	drm_encoder_init(display->drm, &encoder->base, &intel_lvds_enc_funcs,
 			 DRM_MODE_ENCODER_LVDS, "LVDS");
 
 	encoder->enable = intel_enable_lvds;
 	encoder->pre_enable = intel_pre_enable_lvds;
 	encoder->compute_config = intel_lvds_compute_config;
-	if (HAS_PCH_SPLIT(i915)) {
+	if (HAS_PCH_SPLIT(display)) {
 		encoder->disable = pch_disable_lvds;
 		encoder->post_disable = pch_post_disable_lvds;
 	} else {
@@ -931,7 +928,7 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	encoder->power_domain = POWER_DOMAIN_PORT_OTHER;
 	encoder->port = PORT_NONE;
 	encoder->cloneable = 0;
-	if (DISPLAY_VER(i915) < 4)
+	if (DISPLAY_VER(display) < 4)
 		encoder->pipe_mask = BIT(PIPE_B);
 	else
 		encoder->pipe_mask = ~0;
@@ -943,7 +940,7 @@ void intel_lvds_init(struct drm_i915_private *i915)
 
 	intel_lvds_add_properties(&connector->base);
 
-	intel_lvds_pps_get_hw_state(i915, &lvds_encoder->init_pps);
+	intel_lvds_pps_get_hw_state(display, &lvds_encoder->init_pps);
 	lvds_encoder->init_lvds_val = lvds;
 
 	/*
@@ -958,7 +955,7 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	 * Attempt to get the fixed panel mode from DDC.  Assume that the
 	 * preferred mode is the right one.
 	 */
-	mutex_lock(&i915->drm.mode_config.mutex);
+	mutex_lock(&display->drm->mode_config.mutex);
 	if (vga_switcheroo_handler_flags() & VGA_SWITCHEROO_CAN_SWITCH_DDC)
 		drm_edid = drm_edid_read_switcheroo(&connector->base, connector->base.ddc);
 	else
@@ -991,7 +988,7 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	if (!intel_panel_preferred_fixed_mode(connector))
 		intel_panel_add_encoder_fixed_mode(connector, encoder);
 
-	mutex_unlock(&i915->drm.mode_config.mutex);
+	mutex_unlock(&display->drm->mode_config.mutex);
 
 	/* If we still don't have a mode after all that, give up. */
 	if (!intel_panel_preferred_fixed_mode(connector))
@@ -1002,7 +999,7 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	intel_backlight_setup(connector, INVALID_PIPE);
 
 	lvds_encoder->is_dual_link = compute_is_dual_link_lvds(lvds_encoder);
-	drm_dbg_kms(&i915->drm, "detected %s-link lvds configuration\n",
+	drm_dbg_kms(display->drm, "detected %s-link lvds configuration\n",
 		    lvds_encoder->is_dual_link ? "dual" : "single");
 
 	lvds_encoder->a3_power = lvds & LVDS_A3_POWER_MASK;
@@ -1010,7 +1007,7 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	return;
 
 failed:
-	drm_dbg_kms(&i915->drm, "No LVDS modes found, disabling.\n");
+	drm_dbg_kms(display->drm, "No LVDS modes found, disabling.\n");
 	drm_connector_cleanup(&connector->base);
 	drm_encoder_cleanup(&encoder->base);
 	kfree(lvds_encoder);
diff --git a/drivers/gpu/drm/i915/display/intel_lvds.h b/drivers/gpu/drm/i915/display/intel_lvds.h
index 7ad5fa9c0434..a6db1706a97c 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.h
+++ b/drivers/gpu/drm/i915/display/intel_lvds.h
@@ -11,28 +11,28 @@
 #include "i915_reg_defs.h"
 
 enum pipe;
-struct drm_i915_private;
+struct intel_display;
 
 #ifdef I915
-bool intel_lvds_port_enabled(struct drm_i915_private *dev_priv,
+bool intel_lvds_port_enabled(struct intel_display *display,
 			     i915_reg_t lvds_reg, enum pipe *pipe);
-void intel_lvds_init(struct drm_i915_private *dev_priv);
-struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *dev_priv);
-bool intel_is_dual_link_lvds(struct drm_i915_private *dev_priv);
+void intel_lvds_init(struct intel_display *display);
+struct intel_encoder *intel_get_lvds_encoder(struct intel_display *display);
+bool intel_is_dual_link_lvds(struct intel_display *display);
 #else
-static inline bool intel_lvds_port_enabled(struct drm_i915_private *dev_priv,
+static inline bool intel_lvds_port_enabled(struct intel_display *display,
 					   i915_reg_t lvds_reg, enum pipe *pipe)
 {
 	return false;
 }
-static inline void intel_lvds_init(struct drm_i915_private *dev_priv)
+static inline void intel_lvds_init(struct intel_display *display)
 {
 }
-static inline struct intel_encoder *intel_get_lvds_encoder(struct drm_i915_private *dev_priv)
+static inline struct intel_encoder *intel_get_lvds_encoder(struct intel_display *display)
 {
 	return NULL;
 }
-static inline bool intel_is_dual_link_lvds(struct drm_i915_private *dev_priv)
+static inline bool intel_is_dual_link_lvds(struct intel_display *display)
 {
 	return false;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 312b21b1ab59..0325b0c9506d 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -6,11 +6,11 @@
  * state.
  */
 
-#include <drm/drm_atomic_uapi.h>
 #include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_atomic_uapi.h>
+#include <drm/drm_print.h>
 #include <drm/drm_vblank.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "i9xx_wm.h"
 #include "intel_atomic.h"
@@ -31,13 +31,14 @@
 #include "intel_pmdemand.h"
 #include "intel_tc.h"
 #include "intel_vblank.h"
+#include "intel_vga.h"
 #include "intel_wm.h"
 #include "skl_watermark.h"
 
 static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 					      struct drm_modeset_acquire_ctx *ctx)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_crtc_state *crtc_state =
 		to_intel_crtc_state(crtc->base.state);
 	struct intel_plane *plane;
@@ -48,7 +49,7 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 	if (!crtc_state->hw.active)
 		return;
 
-	for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+	for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 		const struct intel_plane_state *plane_state =
 			to_intel_plane_state(plane->base.state);
 
@@ -56,9 +57,9 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 			intel_plane_disable_noatomic(crtc, plane);
 	}
 
-	state = drm_atomic_state_alloc(&i915->drm);
+	state = drm_atomic_state_alloc(display->drm);
 	if (!state) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "failed to disable [CRTC:%d:%s], out of memory",
 			    crtc->base.base.id, crtc->base.name);
 		return;
@@ -68,7 +69,7 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 	to_intel_atomic_state(state)->internal = true;
 
 	/* Everything's already locked, -EDEADLK can't happen. */
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc,
+	for_each_intel_crtc_in_pipe_mask(display->drm, temp_crtc,
 					 BIT(pipe) |
 					 intel_crtc_joiner_secondary_pipes(crtc_state)) {
 		struct intel_crtc_state *temp_crtc_state =
@@ -77,14 +78,14 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 
 		ret = drm_atomic_add_affected_connectors(state, &temp_crtc->base);
 
-		drm_WARN_ON(&i915->drm, IS_ERR(temp_crtc_state) || ret);
+		drm_WARN_ON(display->drm, IS_ERR(temp_crtc_state) || ret);
 	}
 
-	i915->display.funcs.display->crtc_disable(to_intel_atomic_state(state), crtc);
+	display->funcs.display->crtc_disable(to_intel_atomic_state(state), crtc);
 
 	drm_atomic_state_put(state);
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "[CRTC:%d:%s] hw state adjusted, was enabled, now disabled\n",
 		    crtc->base.base.id, crtc->base.name);
 
@@ -118,13 +119,12 @@ static void set_encoder_for_connector(struct intel_connector *connector,
 static void reset_encoder_connector_state(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_pmdemand_state *pmdemand_state =
-		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
+		to_intel_pmdemand_state(display->pmdemand.obj.state);
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		if (connector->base.encoder != &encoder->base)
 			continue;
@@ -143,10 +143,10 @@ static void reset_encoder_connector_state(struct intel_encoder *encoder)
 
 static void reset_crtc_encoder_state(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_encoder *encoder;
 
-	for_each_encoder_on_crtc(&i915->drm, &crtc->base, encoder) {
+	for_each_encoder_on_crtc(display->drm, &crtc->base, encoder) {
 		reset_encoder_connector_state(encoder);
 		encoder->base.crtc = NULL;
 	}
@@ -155,9 +155,8 @@ static void reset_crtc_encoder_state(struct intel_crtc *crtc)
 static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	struct intel_pmdemand_state *pmdemand_state =
-		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
+		to_intel_pmdemand_state(display->pmdemand.obj.state);
 	struct intel_crtc_state *crtc_state =
 		to_intel_crtc_state(crtc->base.state);
 	enum pipe pipe = crtc->pipe;
@@ -169,7 +168,7 @@ static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 	reset_crtc_encoder_state(crtc);
 
 	intel_fbc_disable(crtc);
-	intel_update_watermarks(i915);
+	intel_update_watermarks(display);
 
 	intel_display_power_put_all_in_set(display, &crtc->enabled_power_domains);
 
@@ -184,13 +183,13 @@ static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
  * Return all the pipes using a transcoder in @transcoder_mask.
  * For joiner configs return only the joiner primary.
  */
-static u8 get_transcoder_pipes(struct drm_i915_private *i915,
+static u8 get_transcoder_pipes(struct intel_display *display,
 			       u8 transcoder_mask)
 {
 	struct intel_crtc *temp_crtc;
 	u8 pipes = 0;
 
-	for_each_intel_crtc(&i915->drm, temp_crtc) {
+	for_each_intel_crtc(display->drm, temp_crtc) {
 		struct intel_crtc_state *temp_crtc_state =
 			to_intel_crtc_state(temp_crtc->base.state);
 
@@ -215,7 +214,6 @@ static void get_portsync_pipes(struct intel_crtc *crtc,
 			       u8 *master_pipe_mask, u8 *slave_pipes_mask)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	struct intel_crtc_state *crtc_state =
 		to_intel_crtc_state(crtc->base.state);
 	struct intel_crtc *master_crtc;
@@ -234,20 +232,20 @@ static void get_portsync_pipes(struct intel_crtc *crtc,
 	else
 		master_transcoder = crtc_state->master_transcoder;
 
-	*master_pipe_mask = get_transcoder_pipes(i915, BIT(master_transcoder));
-	drm_WARN_ON(&i915->drm, !is_power_of_2(*master_pipe_mask));
+	*master_pipe_mask = get_transcoder_pipes(display, BIT(master_transcoder));
+	drm_WARN_ON(display->drm, !is_power_of_2(*master_pipe_mask));
 
 	master_crtc = intel_crtc_for_pipe(display, ffs(*master_pipe_mask) - 1);
 	master_crtc_state = to_intel_crtc_state(master_crtc->base.state);
-	*slave_pipes_mask = get_transcoder_pipes(i915, master_crtc_state->sync_mode_slaves_mask);
+	*slave_pipes_mask = get_transcoder_pipes(display, master_crtc_state->sync_mode_slaves_mask);
 }
 
-static u8 get_joiner_secondary_pipes(struct drm_i915_private *i915, u8 primary_pipes_mask)
+static u8 get_joiner_secondary_pipes(struct intel_display *display, u8 primary_pipes_mask)
 {
 	struct intel_crtc *primary_crtc;
 	u8 pipes = 0;
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, primary_crtc, primary_pipes_mask) {
+	for_each_intel_crtc_in_pipe_mask(display->drm, primary_crtc, primary_pipes_mask) {
 		struct intel_crtc_state *primary_crtc_state =
 			to_intel_crtc_state(primary_crtc->base.state);
 
@@ -260,45 +258,45 @@ static u8 get_joiner_secondary_pipes(struct drm_i915_private *i915, u8 primary_p
 static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
 					struct drm_modeset_acquire_ctx *ctx)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
+	struct intel_crtc *temp_crtc;
 	u8 portsync_master_mask;
 	u8 portsync_slaves_mask;
 	u8 joiner_secondaries_mask;
-	struct intel_crtc *temp_crtc;
 
 	/* TODO: Add support for MST */
 	get_portsync_pipes(crtc, &portsync_master_mask, &portsync_slaves_mask);
-	joiner_secondaries_mask = get_joiner_secondary_pipes(i915,
+	joiner_secondaries_mask = get_joiner_secondary_pipes(display,
 							     portsync_master_mask |
 							     portsync_slaves_mask);
 
-	drm_WARN_ON(&i915->drm,
+	drm_WARN_ON(display->drm,
 		    portsync_master_mask & portsync_slaves_mask ||
 		    portsync_master_mask & joiner_secondaries_mask ||
 		    portsync_slaves_mask & joiner_secondaries_mask);
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, joiner_secondaries_mask)
+	for_each_intel_crtc_in_pipe_mask(display->drm, temp_crtc, joiner_secondaries_mask)
 		intel_crtc_disable_noatomic_begin(temp_crtc, ctx);
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, portsync_slaves_mask)
+	for_each_intel_crtc_in_pipe_mask(display->drm, temp_crtc, portsync_slaves_mask)
 		intel_crtc_disable_noatomic_begin(temp_crtc, ctx);
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, portsync_master_mask)
+	for_each_intel_crtc_in_pipe_mask(display->drm, temp_crtc, portsync_master_mask)
 		intel_crtc_disable_noatomic_begin(temp_crtc, ctx);
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc,
+	for_each_intel_crtc_in_pipe_mask(display->drm, temp_crtc,
 					 joiner_secondaries_mask |
 					 portsync_slaves_mask |
 					 portsync_master_mask)
 		intel_crtc_disable_noatomic_complete(temp_crtc);
 }
 
-static void intel_modeset_update_connector_atomic_state(struct drm_i915_private *i915)
+static void intel_modeset_update_connector_atomic_state(struct intel_display *display)
 {
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		struct drm_connector_state *conn_state = connector->base.state;
 		struct intel_encoder *encoder =
@@ -320,7 +318,7 @@ static void intel_modeset_update_connector_atomic_state(struct drm_i915_private
 
 static void intel_crtc_copy_hw_to_uapi_state(struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
 	if (intel_crtc_is_joiner_secondary(crtc_state))
 		return;
@@ -333,7 +331,7 @@ static void intel_crtc_copy_hw_to_uapi_state(struct intel_crtc_state *crtc_state
 	crtc_state->uapi.adjusted_mode = crtc_state->hw.adjusted_mode;
 	crtc_state->uapi.scaling_filter = crtc_state->hw.scaling_filter;
 
-	if (DISPLAY_INFO(i915)->color.degamma_lut_size) {
+	if (DISPLAY_INFO(display)->color.degamma_lut_size) {
 		/* assume 1:1 mapping */
 		drm_property_replace_blob(&crtc_state->hw.degamma_lut,
 					  crtc_state->pre_csc_lut);
@@ -348,7 +346,7 @@ static void intel_crtc_copy_hw_to_uapi_state(struct intel_crtc_state *crtc_state
 		 * to gamma_lut as that is the only valid source of LUTs
 		 * in the uapi.
 		 */
-		drm_WARN_ON(&i915->drm, crtc_state->post_csc_lut &&
+		drm_WARN_ON(display->drm, crtc_state->post_csc_lut &&
 			    crtc_state->pre_csc_lut);
 
 		drm_property_replace_blob(&crtc_state->hw.degamma_lut,
@@ -367,15 +365,14 @@ static void intel_crtc_copy_hw_to_uapi_state(struct intel_crtc_state *crtc_state
 }
 
 static void
-intel_sanitize_plane_mapping(struct drm_i915_private *i915)
+intel_sanitize_plane_mapping(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_crtc *crtc;
 
-	if (DISPLAY_VER(i915) >= 4)
+	if (DISPLAY_VER(display) >= 4)
 		return;
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_plane *plane =
 			to_intel_plane(crtc->base.primary);
 		struct intel_crtc *plane_crtc;
@@ -387,7 +384,7 @@ intel_sanitize_plane_mapping(struct drm_i915_private *i915)
 		if (pipe == crtc->pipe)
 			continue;
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[PLANE:%d:%s] attached to the wrong pipe, disabling plane\n",
 			    plane->base.base.id, plane->base.name);
 
@@ -424,12 +421,12 @@ static bool intel_crtc_needs_link_reset(struct intel_crtc *crtc)
 
 static struct intel_connector *intel_encoder_find_connector(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct drm_connector_list_iter conn_iter;
 	struct intel_connector *connector;
 	struct intel_connector *found_connector = NULL;
 
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		if (&encoder->base == connector->base.encoder) {
 			found_connector = connector;
@@ -467,7 +464,7 @@ static void intel_sanitize_fifo_underrun_reporting(const struct intel_crtc_state
 static bool intel_sanitize_crtc(struct intel_crtc *crtc,
 				struct drm_modeset_acquire_ctx *ctx)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state);
 	bool needs_link_reset;
 
@@ -475,7 +472,7 @@ static bool intel_sanitize_crtc(struct intel_crtc *crtc,
 		struct intel_plane *plane;
 
 		/* Disable everything but the primary plane */
-		for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+		for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 			const struct intel_plane_state *plane_state =
 				to_intel_plane_state(plane->base.state);
 
@@ -516,7 +513,7 @@ static bool intel_sanitize_crtc(struct intel_crtc *crtc,
 	return true;
 }
 
-static void intel_sanitize_all_crtcs(struct drm_i915_private *i915,
+static void intel_sanitize_all_crtcs(struct intel_display *display,
 				     struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_crtc *crtc;
@@ -531,7 +528,7 @@ static void intel_sanitize_all_crtcs(struct drm_i915_private *i915,
 	for (;;) {
 		u32 old_mask = crtcs_forced_off;
 
-		for_each_intel_crtc(&i915->drm, crtc) {
+		for_each_intel_crtc(display->drm, crtc) {
 			u32 crtc_mask = drm_crtc_mask(&crtc->base);
 
 			if (crtcs_forced_off & crtc_mask)
@@ -544,7 +541,7 @@ static void intel_sanitize_all_crtcs(struct drm_i915_private *i915,
 			break;
 	}
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 
@@ -554,7 +551,7 @@ static void intel_sanitize_all_crtcs(struct drm_i915_private *i915,
 
 static bool has_bogus_dpll_config(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
 	/*
 	 * Some SNB BIOSen (eg. ASUS K53SV) are known to misprogram
@@ -566,7 +563,7 @@ static bool has_bogus_dpll_config(const struct intel_crtc_state *crtc_state)
 	 * without several WARNs, but for now let's take the easy
 	 * road.
 	 */
-	return IS_SANDYBRIDGE(i915) &&
+	return display->platform.sandybridge &&
 		crtc_state->hw.active &&
 		crtc_state->shared_dpll &&
 		crtc_state->port_clock == 0;
@@ -575,13 +572,12 @@ static bool has_bogus_dpll_config(const struct intel_crtc_state *crtc_state)
 static void intel_sanitize_encoder(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_connector *connector;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	struct intel_crtc_state *crtc_state = crtc ?
 		to_intel_crtc_state(crtc->base.state) : NULL;
 	struct intel_pmdemand_state *pmdemand_state =
-		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
+		to_intel_pmdemand_state(display->pmdemand.obj.state);
 
 	/*
 	 * We need to check both for a crtc link (meaning that the encoder is
@@ -592,7 +588,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 		crtc_state->hw.active;
 
 	if (crtc_state && has_bogus_dpll_config(crtc_state)) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "BIOS has misprogrammed the hardware. Disabling pipe %c\n",
 			    pipe_name(crtc->pipe));
 		has_active_crtc = false;
@@ -600,7 +596,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 
 	connector = intel_encoder_find_connector(encoder);
 	if (connector && !has_active_crtc) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[ENCODER:%d:%s] has active connectors but no active pipe!\n",
 			    encoder->base.base.id,
 			    encoder->base.name);
@@ -617,7 +613,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 		if (crtc_state) {
 			struct drm_encoder *best_encoder;
 
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[ENCODER:%d:%s] manually disabled\n",
 				    encoder->base.base.id,
 				    encoder->base.name);
@@ -651,18 +647,17 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	/* notify opregion of the sanitized encoder state */
 	intel_opregion_notify_encoder(encoder, connector && has_active_crtc);
 
-	if (HAS_DDI(i915))
+	if (HAS_DDI(display))
 		intel_ddi_sanitize_encoder_pll_mapping(encoder);
 }
 
 /* FIXME read out full plane state for all planes */
-static void readout_plane_state(struct drm_i915_private *i915)
+static void readout_plane_state(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_plane *plane;
 	struct intel_crtc *crtc;
 
-	for_each_intel_plane(&i915->drm, plane) {
+	for_each_intel_plane(display->drm, plane) {
 		struct intel_plane_state *plane_state =
 			to_intel_plane_state(plane->base.state);
 		struct intel_crtc_state *crtc_state;
@@ -676,13 +671,13 @@ static void readout_plane_state(struct drm_i915_private *i915)
 
 		intel_set_plane_visible(crtc_state, plane_state, visible);
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[PLANE:%d:%s] hw state readout: %s, pipe %c\n",
 			    plane->base.base.id, plane->base.name,
 			    str_enabled_disabled(visible), pipe_name(pipe));
 	}
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 
@@ -690,18 +685,17 @@ static void readout_plane_state(struct drm_i915_private *i915)
 	}
 }
 
-static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
+static void intel_modeset_readout_hw_state(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_pmdemand_state *pmdemand_state =
-		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
+		to_intel_pmdemand_state(display->pmdemand.obj.state);
 	enum pipe pipe;
 	struct intel_crtc *crtc;
 	struct intel_encoder *encoder;
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 
@@ -716,15 +710,15 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 		crtc->base.enabled = crtc_state->hw.enable;
 		crtc->active = crtc_state->hw.active;
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[CRTC:%d:%s] hw state readout: %s\n",
 			    crtc->base.base.id, crtc->base.name,
 			    str_enabled_disabled(crtc_state->hw.active));
 	}
 
-	readout_plane_state(i915);
+	readout_plane_state(display);
 
-	for_each_intel_encoder(&i915->drm, encoder) {
+	for_each_intel_encoder(display->drm, encoder) {
 		struct intel_crtc_state *crtc_state = NULL;
 
 		pipe = 0;
@@ -743,7 +737,7 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 				/* encoder should read be linked to joiner primary */
 				WARN_ON(intel_crtc_is_joiner_secondary(crtc_state));
 
-				for_each_intel_crtc_in_pipe_mask(&i915->drm, secondary_crtc,
+				for_each_intel_crtc_in_pipe_mask(display->drm, secondary_crtc,
 								 intel_crtc_joiner_secondary_pipes(crtc_state)) {
 					struct intel_crtc_state *secondary_crtc_state;
 
@@ -766,7 +760,7 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 		if (encoder->sync_state)
 			encoder->sync_state(encoder, crtc_state);
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[ENCODER:%d:%s] hw state readout: %s, pipe %c\n",
 			    encoder->base.base.id, encoder->base.name,
 			    str_enabled_disabled(encoder->base.crtc),
@@ -775,7 +769,7 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 
 	intel_dpll_readout_hw_state(display);
 
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_connector_list_iter_begin(display->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		struct intel_crtc_state *crtc_state = NULL;
 
@@ -809,37 +803,37 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 		if (connector->sync_state)
 			connector->sync_state(connector, crtc_state);
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "[CONNECTOR:%d:%s] hw state readout: %s\n",
 			    connector->base.base.id, connector->base.name,
 			    str_enabled_disabled(connector->base.encoder));
 	}
 	drm_connector_list_iter_end(&conn_iter);
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct intel_plane *plane;
 
-		if (crtc_state->hw.active) {
-			/*
-			 * The initial mode needs to be set in order to keep
-			 * the atomic core happy. It wants a valid mode if the
-			 * crtc's enabled, so we do the above call.
-			 *
-			 * But we don't set all the derived state fully, hence
-			 * set a flag to indicate that a full recalculation is
-			 * needed on the next commit.
-			 */
-			crtc_state->inherited = true;
+		/*
+		 * The initial mode needs to be set in order to keep
+		 * the atomic core happy. It wants a valid mode if the
+		 * crtc's enabled, so we do the above call.
+		 *
+		 * But we don't set all the derived state fully, hence
+		 * set a flag to indicate that a full recalculation is
+		 * needed on the next commit.
+		 */
+		crtc_state->inherited = true;
 
+		if (crtc_state->hw.active) {
 			intel_crtc_update_active_timings(crtc_state,
 							 crtc_state->vrr.enable);
 
 			intel_crtc_copy_hw_to_uapi_state(crtc_state);
 		}
 
-		for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+		for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 			const struct intel_plane_state *plane_state =
 				to_intel_plane_state(plane->base.state);
 
@@ -855,14 +849,14 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 			 * use plane->min_cdclk() :(
 			 */
 			if (plane_state->uapi.visible && plane->min_cdclk) {
-				if (crtc_state->double_wide || DISPLAY_VER(i915) >= 10)
+				if (crtc_state->double_wide || DISPLAY_VER(display) >= 10)
 					crtc_state->min_cdclk[plane->id] =
 						DIV_ROUND_UP(crtc_state->pixel_rate, 2);
 				else
 					crtc_state->min_cdclk[plane->id] =
 						crtc_state->pixel_rate;
 			}
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s] min_cdclk %d kHz\n",
 				    plane->base.base.id, plane->base.name,
 				    crtc_state->min_cdclk[plane->id]);
@@ -874,7 +868,7 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 
 	/* TODO move here (or even earlier?) on all platforms */
 	if (DISPLAY_VER(display) >= 9)
-		intel_wm_get_hw_state(i915);
+		intel_wm_get_hw_state(display);
 
 	intel_bw_update_hw_state(display);
 	intel_cdclk_update_hw_state(display);
@@ -883,11 +877,11 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 }
 
 static void
-get_encoder_power_domains(struct drm_i915_private *i915)
+get_encoder_power_domains(struct intel_display *display)
 {
 	struct intel_encoder *encoder;
 
-	for_each_intel_encoder(&i915->drm, encoder) {
+	for_each_intel_encoder(display->drm, encoder) {
 		struct intel_crtc_state *crtc_state;
 
 		if (!encoder->get_power_domains)
@@ -905,49 +899,51 @@ get_encoder_power_domains(struct drm_i915_private *i915)
 	}
 }
 
-static void intel_early_display_was(struct drm_i915_private *i915)
+static void intel_early_display_was(struct intel_display *display)
 {
 	/*
 	 * Display WA #1185 WaDisableDARBFClkGating:glk,icl,ehl,tgl
 	 * Also known as Wa_14010480278.
 	 */
-	if (IS_DISPLAY_VER(i915, 10, 12))
-		intel_de_rmw(i915, GEN9_CLKGATE_DIS_0, 0, DARBF_GATING_DIS);
+	if (IS_DISPLAY_VER(display, 10, 12))
+		intel_de_rmw(display, GEN9_CLKGATE_DIS_0, 0, DARBF_GATING_DIS);
 
 	/*
 	 * WaRsPkgCStateDisplayPMReq:hsw
 	 * System hang if this isn't done before disabling all planes!
 	 */
-	if (IS_HASWELL(i915))
-		intel_de_rmw(i915, CHICKEN_PAR1_1, 0, FORCE_ARB_IDLE_PLANES);
+	if (display->platform.haswell)
+		intel_de_rmw(display, CHICKEN_PAR1_1, 0, FORCE_ARB_IDLE_PLANES);
 
-	if (IS_KABYLAKE(i915) || IS_COFFEELAKE(i915) || IS_COMETLAKE(i915)) {
+	if (display->platform.kabylake || display->platform.coffeelake ||
+	    display->platform.cometlake) {
 		/* Display WA #1142:kbl,cfl,cml */
-		intel_de_rmw(i915, CHICKEN_PAR1_1,
+		intel_de_rmw(display, CHICKEN_PAR1_1,
 			     KBL_ARB_FILL_SPARE_22, KBL_ARB_FILL_SPARE_22);
-		intel_de_rmw(i915, CHICKEN_MISC_2,
+		intel_de_rmw(display, CHICKEN_MISC_2,
 			     KBL_ARB_FILL_SPARE_13 | KBL_ARB_FILL_SPARE_14,
 			     KBL_ARB_FILL_SPARE_14);
 	}
 }
 
-void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
+void intel_modeset_setup_hw_state(struct intel_display *display,
 				  struct drm_modeset_acquire_ctx *ctx)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_encoder *encoder;
 	struct intel_crtc *crtc;
 	intel_wakeref_t wakeref;
 
 	wakeref = intel_display_power_get(display, POWER_DOMAIN_INIT);
 
-	intel_early_display_was(i915);
-	intel_modeset_readout_hw_state(i915);
+	intel_early_display_was(display);
+	intel_vga_disable(display);
+
+	intel_modeset_readout_hw_state(display);
 
 	/* HW state is read out, now we need to sanitize this mess. */
-	get_encoder_power_domains(i915);
+	get_encoder_power_domains(display);
 
-	intel_pch_sanitize(i915);
+	intel_pch_sanitize(display);
 
 	intel_cmtg_sanitize(display);
 
@@ -955,7 +951,7 @@ void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
 	 * intel_sanitize_plane_mapping() may need to do vblank
 	 * waits, so we need vblank interrupts restored beforehand.
 	 */
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 
@@ -969,35 +965,35 @@ void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
 		}
 	}
 
-	intel_fbc_sanitize(&i915->display);
+	intel_fbc_sanitize(display);
 
-	intel_sanitize_plane_mapping(i915);
+	intel_sanitize_plane_mapping(display);
 
-	for_each_intel_encoder(&i915->drm, encoder)
+	for_each_intel_encoder(display->drm, encoder)
 		intel_sanitize_encoder(encoder);
 
 	/*
 	 * Sanitizing CRTCs needs their connector atomic state to be
 	 * up-to-date, so ensure that already here.
 	 */
-	intel_modeset_update_connector_atomic_state(i915);
+	intel_modeset_update_connector_atomic_state(display);
 
-	intel_sanitize_all_crtcs(i915, ctx);
+	intel_sanitize_all_crtcs(display, ctx);
 
 	intel_dpll_sanitize_state(display);
 
 	/* TODO move earlier on all platforms */
 	if (DISPLAY_VER(display) < 9)
-		intel_wm_get_hw_state(i915);
-	intel_wm_sanitize(i915);
+		intel_wm_get_hw_state(display);
+	intel_wm_sanitize(display);
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		struct intel_power_domain_mask put_domains;
 
 		intel_modeset_get_crtc_power_domains(crtc_state, &put_domains);
-		if (drm_WARN_ON(&i915->drm, !bitmap_empty(put_domains.bits, POWER_DOMAIN_NUM)))
+		if (drm_WARN_ON(display->drm, !bitmap_empty(put_domains.bits, POWER_DOMAIN_NUM)))
 			intel_modeset_put_crtc_power_domains(crtc, &put_domains);
 	}
 
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.h b/drivers/gpu/drm/i915/display/intel_modeset_setup.h
index 3beff67b33d0..f5e6f3ae9572 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.h
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.h
@@ -6,10 +6,10 @@
 #ifndef __INTEL_MODESET_SETUP_H__
 #define __INTEL_MODESET_SETUP_H__
 
-struct drm_i915_private;
 struct drm_modeset_acquire_ctx;
+struct intel_display;
 
-void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
+void intel_modeset_setup_hw_state(struct intel_display *display,
 				  struct drm_modeset_acquire_ctx *ctx);
 
 #endif /* __INTEL_MODESET_SETUP_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_verify.c b/drivers/gpu/drm/i915/display/intel_modeset_verify.c
index a008412fdd04..766a9983665a 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_verify.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_verify.c
@@ -6,13 +6,14 @@
  */
 
 #include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_print.h>
 
-#include "i915_drv.h"
 #include "intel_atomic.h"
 #include "intel_crtc.h"
 #include "intel_crtc_state_dump.h"
 #include "intel_cx0_phy.h"
 #include "intel_display.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_fdi.h"
 #include "intel_modeset_verify.h"
@@ -28,9 +29,8 @@ static void intel_connector_verify_state(const struct intel_crtc_state *crtc_sta
 {
 	struct intel_connector *connector = to_intel_connector(conn_state->connector);
 	struct intel_display *display = to_intel_display(connector);
-	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
-	drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s]\n",
+	drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s]\n",
 		    connector->base.base.id, connector->base.name);
 
 	if (connector->get_hw_state(connector)) {
@@ -91,7 +91,6 @@ verify_connector_state(struct intel_atomic_state *state,
 static void intel_pipe_config_sanity_check(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
 	if (crtc_state->has_pch_encoder) {
 		int fdi_dotclock = intel_dotclock_calculate(intel_fdi_link_freq(display, crtc_state),
@@ -103,7 +102,7 @@ static void intel_pipe_config_sanity_check(const struct intel_crtc_state *crtc_s
 		 * Yell if the encoder disagrees. Allow for slight
 		 * rounding differences.
 		 */
-		drm_WARN(&i915->drm, abs(fdi_dotclock - dotclock) > 1,
+		drm_WARN(display->drm, abs(fdi_dotclock - dotclock) > 1,
 			 "FDI dotclock and encoder dotclock mismatch, fdi: %i, encoder: %i\n",
 			 fdi_dotclock, dotclock);
 	}
@@ -113,17 +112,16 @@ static void
 verify_encoder_state(struct intel_atomic_state *state)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	struct intel_encoder *encoder;
 	struct drm_connector *connector;
 	const struct drm_connector_state *old_conn_state, *new_conn_state;
 	int i;
 
-	for_each_intel_encoder(&i915->drm, encoder) {
+	for_each_intel_encoder(display->drm, encoder) {
 		bool enabled = false, found = false;
 		enum pipe pipe;
 
-		drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s]\n",
+		drm_dbg_kms(display->drm, "[ENCODER:%d:%s]\n",
 			    encoder->base.base.id,
 			    encoder->base.name);
 
@@ -166,7 +164,6 @@ verify_crtc_state(struct intel_atomic_state *state,
 		  struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	const struct intel_crtc_state *sw_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_crtc_state *hw_crtc_state;
@@ -185,7 +182,7 @@ verify_crtc_state(struct intel_atomic_state *state,
 	intel_crtc_get_pipe_config(hw_crtc_state);
 
 	/* we keep both pipes enabled on 830 */
-	if (IS_I830(i915) && hw_crtc_state->hw.active)
+	if (display->platform.i830 && hw_crtc_state->hw.active)
 		hw_crtc_state->hw.active = sw_crtc_state->hw.active;
 
 	INTEL_DISPLAY_STATE_WARN(display,
diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
index aff9a3455c1b..12308495afa5 100644
--- a/drivers/gpu/drm/i915/display/intel_overlay.c
+++ b/drivers/gpu/drm/i915/display/intel_overlay.c
@@ -291,7 +291,6 @@ static void intel_overlay_flip_prepare(struct intel_overlay *overlay,
 				       struct i915_vma *vma)
 {
 	struct intel_display *display = overlay->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	enum pipe pipe = overlay->crtc->pipe;
 	struct intel_frontbuffer *frontbuffer = NULL;
 
@@ -307,7 +306,7 @@ static void intel_overlay_flip_prepare(struct intel_overlay *overlay,
 		intel_frontbuffer_put(overlay->frontbuffer);
 	overlay->frontbuffer = frontbuffer;
 
-	intel_frontbuffer_flip_prepare(i915, INTEL_FRONTBUFFER_OVERLAY(pipe));
+	intel_frontbuffer_flip_prepare(display, INTEL_FRONTBUFFER_OVERLAY(pipe));
 
 	overlay->old_vma = overlay->vma;
 	if (vma)
@@ -359,14 +358,13 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 static void intel_overlay_release_old_vma(struct intel_overlay *overlay)
 {
 	struct intel_display *display = overlay->display;
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct i915_vma *vma;
 
 	vma = fetch_and_zero(&overlay->old_vma);
 	if (drm_WARN_ON(display->drm, !vma))
 		return;
 
-	intel_frontbuffer_flip_complete(i915, INTEL_FRONTBUFFER_OVERLAY(overlay->crtc->pipe));
+	intel_frontbuffer_flip_complete(display, INTEL_FRONTBUFFER_OVERLAY(overlay->crtc->pipe));
 
 	i915_vma_unpin(vma);
 	i915_vma_put(vma);
diff --git a/drivers/gpu/drm/i915/display/intel_pch.c b/drivers/gpu/drm/i915/display/intel_pch.c
new file mode 100644
index 000000000000..469e8a3cfb49
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_pch.c
@@ -0,0 +1,340 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright 2025 Intel Corporation.
+ */
+
+#include <drm/drm_print.h>
+
+#include "i915_utils.h"
+#include "intel_display_core.h"
+#include "intel_pch.h"
+
+#define INTEL_PCH_DEVICE_ID_MASK		0xff80
+#define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
+#define INTEL_PCH_CPT_DEVICE_ID_TYPE		0x1c00
+#define INTEL_PCH_PPT_DEVICE_ID_TYPE		0x1e00
+#define INTEL_PCH_LPT_DEVICE_ID_TYPE		0x8c00
+#define INTEL_PCH_LPT_LP_DEVICE_ID_TYPE		0x9c00
+#define INTEL_PCH_WPT_DEVICE_ID_TYPE		0x8c80
+#define INTEL_PCH_WPT_LP_DEVICE_ID_TYPE		0x9c80
+#define INTEL_PCH_SPT_DEVICE_ID_TYPE		0xA100
+#define INTEL_PCH_SPT_LP_DEVICE_ID_TYPE		0x9D00
+#define INTEL_PCH_KBP_DEVICE_ID_TYPE		0xA280
+#define INTEL_PCH_CNP_DEVICE_ID_TYPE		0xA300
+#define INTEL_PCH_CNP_LP_DEVICE_ID_TYPE		0x9D80
+#define INTEL_PCH_CMP_DEVICE_ID_TYPE		0x0280
+#define INTEL_PCH_CMP2_DEVICE_ID_TYPE		0x0680
+#define INTEL_PCH_CMP_V_DEVICE_ID_TYPE		0xA380
+#define INTEL_PCH_ICP_DEVICE_ID_TYPE		0x3480
+#define INTEL_PCH_ICP2_DEVICE_ID_TYPE		0x3880
+#define INTEL_PCH_MCC_DEVICE_ID_TYPE		0x4B00
+#define INTEL_PCH_TGP_DEVICE_ID_TYPE		0xA080
+#define INTEL_PCH_TGP2_DEVICE_ID_TYPE		0x4380
+#define INTEL_PCH_JSP_DEVICE_ID_TYPE		0x4D80
+#define INTEL_PCH_ADP_DEVICE_ID_TYPE		0x7A80
+#define INTEL_PCH_ADP2_DEVICE_ID_TYPE		0x5180
+#define INTEL_PCH_ADP3_DEVICE_ID_TYPE		0x7A00
+#define INTEL_PCH_ADP4_DEVICE_ID_TYPE		0x5480
+#define INTEL_PCH_P2X_DEVICE_ID_TYPE		0x7100
+#define INTEL_PCH_P3X_DEVICE_ID_TYPE		0x7000
+#define INTEL_PCH_QEMU_DEVICE_ID_TYPE		0x2900 /* qemu q35 has 2918 */
+
+/*
+ * Check for platforms where the south display is on the same PCI device or SoC
+ * die as the north display. The PCH (if it even exists) is not involved in
+ * display. Return a fake PCH type for south display handling on these
+ * platforms, without actually detecting the PCH, and PCH_NONE otherwise.
+ */
+static enum intel_pch intel_pch_fake_for_south_display(struct intel_display *display)
+{
+	enum intel_pch pch_type = PCH_NONE;
+
+	if (DISPLAY_VER(display) >= 20)
+		pch_type = PCH_LNL;
+	else if (display->platform.battlemage || display->platform.meteorlake)
+		pch_type = PCH_MTL;
+	else if (display->platform.dg2)
+		pch_type = PCH_DG2;
+	else if (display->platform.dg1)
+		pch_type = PCH_DG1;
+
+	return pch_type;
+}
+
+/* Map PCH device id to PCH type, or PCH_NONE if unknown. */
+static enum intel_pch
+intel_pch_type(const struct intel_display *display, unsigned short id)
+{
+	switch (id) {
+	case INTEL_PCH_IBX_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Ibex Peak PCH\n");
+		drm_WARN_ON(display->drm, DISPLAY_VER(display) != 5);
+		return PCH_IBX;
+	case INTEL_PCH_CPT_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found CougarPoint PCH\n");
+		drm_WARN_ON(display->drm,
+			    DISPLAY_VER(display) != 6 &&
+			    !display->platform.ivybridge);
+		return PCH_CPT;
+	case INTEL_PCH_PPT_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found PantherPoint PCH\n");
+		drm_WARN_ON(display->drm,
+			    DISPLAY_VER(display) != 6 &&
+			    !display->platform.ivybridge);
+		/* PPT is CPT compatible */
+		return PCH_CPT;
+	case INTEL_PCH_LPT_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found LynxPoint PCH\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.haswell &&
+			    !display->platform.broadwell);
+		drm_WARN_ON(display->drm,
+			    display->platform.haswell_ult ||
+			    display->platform.broadwell_ult);
+		return PCH_LPT_H;
+	case INTEL_PCH_LPT_LP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found LynxPoint LP PCH\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.haswell &&
+			    !display->platform.broadwell);
+		drm_WARN_ON(display->drm,
+			    !display->platform.haswell_ult &&
+			    !display->platform.broadwell_ult);
+		return PCH_LPT_LP;
+	case INTEL_PCH_WPT_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found WildcatPoint PCH\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.haswell &&
+			    !display->platform.broadwell);
+		drm_WARN_ON(display->drm,
+			    display->platform.haswell_ult ||
+			    display->platform.broadwell_ult);
+		/* WPT is LPT compatible */
+		return PCH_LPT_H;
+	case INTEL_PCH_WPT_LP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found WildcatPoint LP PCH\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.haswell &&
+			    !display->platform.broadwell);
+		drm_WARN_ON(display->drm,
+			    !display->platform.haswell_ult &&
+			    !display->platform.broadwell_ult);
+		/* WPT is LPT compatible */
+		return PCH_LPT_LP;
+	case INTEL_PCH_SPT_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found SunrisePoint PCH\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.skylake &&
+			    !display->platform.kabylake &&
+			    !display->platform.coffeelake);
+		return PCH_SPT;
+	case INTEL_PCH_SPT_LP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found SunrisePoint LP PCH\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.skylake &&
+			    !display->platform.kabylake &&
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake);
+		return PCH_SPT;
+	case INTEL_PCH_KBP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Kaby Lake PCH (KBP)\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.skylake &&
+			    !display->platform.kabylake &&
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake);
+		/* KBP is SPT compatible */
+		return PCH_SPT;
+	case INTEL_PCH_CNP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Cannon Lake PCH (CNP)\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake);
+		return PCH_CNP;
+	case INTEL_PCH_CNP_LP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm,
+			    "Found Cannon Lake LP PCH (CNP-LP)\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake);
+		return PCH_CNP;
+	case INTEL_PCH_CMP_DEVICE_ID_TYPE:
+	case INTEL_PCH_CMP2_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Comet Lake PCH (CMP)\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake &&
+			    !display->platform.rocketlake);
+		/* CMP is CNP compatible */
+		return PCH_CNP;
+	case INTEL_PCH_CMP_V_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Comet Lake V PCH (CMP-V)\n");
+		drm_WARN_ON(display->drm,
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake);
+		/* CMP-V is based on KBP, which is SPT compatible */
+		return PCH_SPT;
+	case INTEL_PCH_ICP_DEVICE_ID_TYPE:
+	case INTEL_PCH_ICP2_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Ice Lake PCH\n");
+		drm_WARN_ON(display->drm, !display->platform.icelake);
+		return PCH_ICP;
+	case INTEL_PCH_MCC_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Mule Creek Canyon PCH\n");
+		drm_WARN_ON(display->drm, !(display->platform.jasperlake ||
+					    display->platform.elkhartlake));
+		/* MCC is TGP compatible */
+		return PCH_TGP;
+	case INTEL_PCH_TGP_DEVICE_ID_TYPE:
+	case INTEL_PCH_TGP2_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Tiger Lake LP PCH\n");
+		drm_WARN_ON(display->drm, !display->platform.tigerlake &&
+			    !display->platform.rocketlake &&
+			    !display->platform.skylake &&
+			    !display->platform.kabylake &&
+			    !display->platform.coffeelake &&
+			    !display->platform.cometlake);
+		return PCH_TGP;
+	case INTEL_PCH_JSP_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Jasper Lake PCH\n");
+		drm_WARN_ON(display->drm, !(display->platform.jasperlake ||
+					    display->platform.elkhartlake));
+		/* JSP is ICP compatible */
+		return PCH_ICP;
+	case INTEL_PCH_ADP_DEVICE_ID_TYPE:
+	case INTEL_PCH_ADP2_DEVICE_ID_TYPE:
+	case INTEL_PCH_ADP3_DEVICE_ID_TYPE:
+	case INTEL_PCH_ADP4_DEVICE_ID_TYPE:
+		drm_dbg_kms(display->drm, "Found Alder Lake PCH\n");
+		drm_WARN_ON(display->drm, !display->platform.alderlake_s &&
+			    !display->platform.alderlake_p);
+		return PCH_ADP;
+	default:
+		return PCH_NONE;
+	}
+}
+
+static bool intel_is_virt_pch(unsigned short id,
+			      unsigned short svendor, unsigned short sdevice)
+{
+	return (id == INTEL_PCH_P2X_DEVICE_ID_TYPE ||
+		id == INTEL_PCH_P3X_DEVICE_ID_TYPE ||
+		(id == INTEL_PCH_QEMU_DEVICE_ID_TYPE &&
+		 svendor == PCI_SUBVENDOR_ID_REDHAT_QUMRANET &&
+		 sdevice == PCI_SUBDEVICE_ID_QEMU));
+}
+
+static void
+intel_virt_detect_pch(const struct intel_display *display,
+		      unsigned short *pch_id, enum intel_pch *pch_type)
+{
+	unsigned short id = 0;
+
+	/*
+	 * In a virtualized passthrough environment we can be in a
+	 * setup where the ISA bridge is not able to be passed through.
+	 * In this case, a south bridge can be emulated and we have to
+	 * make an educated guess as to which PCH is really there.
+	 */
+
+	if (display->platform.alderlake_s || display->platform.alderlake_p)
+		id = INTEL_PCH_ADP_DEVICE_ID_TYPE;
+	else if (display->platform.tigerlake || display->platform.rocketlake)
+		id = INTEL_PCH_TGP_DEVICE_ID_TYPE;
+	else if (display->platform.jasperlake || display->platform.elkhartlake)
+		id = INTEL_PCH_MCC_DEVICE_ID_TYPE;
+	else if (display->platform.icelake)
+		id = INTEL_PCH_ICP_DEVICE_ID_TYPE;
+	else if (display->platform.coffeelake ||
+		 display->platform.cometlake)
+		id = INTEL_PCH_CNP_DEVICE_ID_TYPE;
+	else if (display->platform.kabylake || display->platform.skylake)
+		id = INTEL_PCH_SPT_DEVICE_ID_TYPE;
+	else if (display->platform.haswell_ult ||
+		 display->platform.broadwell_ult)
+		id = INTEL_PCH_LPT_LP_DEVICE_ID_TYPE;
+	else if (display->platform.haswell || display->platform.broadwell)
+		id = INTEL_PCH_LPT_DEVICE_ID_TYPE;
+	else if (DISPLAY_VER(display) == 6 || display->platform.ivybridge)
+		id = INTEL_PCH_CPT_DEVICE_ID_TYPE;
+	else if (DISPLAY_VER(display) == 5)
+		id = INTEL_PCH_IBX_DEVICE_ID_TYPE;
+
+	if (id)
+		drm_dbg_kms(display->drm, "Assuming PCH ID %04x\n", id);
+	else
+		drm_dbg_kms(display->drm, "Assuming no PCH\n");
+
+	*pch_type = intel_pch_type(display, id);
+
+	/* Sanity check virtual PCH id */
+	if (drm_WARN_ON(display->drm,
+			id && *pch_type == PCH_NONE))
+		id = 0;
+
+	*pch_id = id;
+}
+
+void intel_pch_detect(struct intel_display *display)
+{
+	struct pci_dev *pch = NULL;
+	unsigned short id;
+	enum intel_pch pch_type;
+
+	pch_type = intel_pch_fake_for_south_display(display);
+	if (pch_type != PCH_NONE) {
+		display->pch_type = pch_type;
+		drm_dbg_kms(display->drm,
+			    "PCH not involved in display, using fake PCH type %d for south display\n",
+			    pch_type);
+		return;
+	}
+
+	/*
+	 * The reason to probe ISA bridge instead of Dev31:Fun0 is to
+	 * make graphics device passthrough work easy for VMM, that only
+	 * need to expose ISA bridge to let driver know the real hardware
+	 * underneath. This is a requirement from virtualization team.
+	 *
+	 * In some virtualized environments (e.g. XEN), there is irrelevant
+	 * ISA bridge in the system. To work reliably, we should scan through
+	 * all the ISA bridge devices and check for the first match, instead
+	 * of only checking the first one.
+	 */
+	while ((pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, pch))) {
+		if (pch->vendor != PCI_VENDOR_ID_INTEL)
+			continue;
+
+		id = pch->device & INTEL_PCH_DEVICE_ID_MASK;
+
+		pch_type = intel_pch_type(display, id);
+		if (pch_type != PCH_NONE) {
+			display->pch_type = pch_type;
+			break;
+		} else if (intel_is_virt_pch(id, pch->subsystem_vendor,
+					     pch->subsystem_device)) {
+			intel_virt_detect_pch(display, &id, &pch_type);
+			display->pch_type = pch_type;
+			break;
+		}
+	}
+
+	/*
+	 * Use PCH_NOP (PCH but no South Display) for PCH platforms without
+	 * display.
+	 */
+	if (pch && !HAS_DISPLAY(display)) {
+		drm_dbg_kms(display->drm,
+			    "Display disabled, reverting to NOP PCH\n");
+		display->pch_type = PCH_NOP;
+	} else if (!pch) {
+		if (i915_run_as_guest() && HAS_DISPLAY(display)) {
+			intel_virt_detect_pch(display, &id, &pch_type);
+			display->pch_type = pch_type;
+		} else {
+			drm_dbg_kms(display->drm, "No PCH found.\n");
+		}
+	}
+
+	pci_dev_put(pch);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_pch.h b/drivers/gpu/drm/i915/display/intel_pch.h
new file mode 100644
index 000000000000..cf4dab1b98bf
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_pch.h
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2025 Intel Corporation.
+ */
+
+#ifndef __INTEL_PCH__
+#define __INTEL_PCH__
+
+#include "intel_display_conversion.h"
+
+struct intel_display;
+
+/*
+ * Sorted by south display engine compatibility.
+ * If the new PCH comes with a south display engine that is not
+ * inherited from the latest item, please do not add it to the
+ * end. Instead, add it right after its "parent" PCH.
+ */
+enum intel_pch {
+	PCH_NOP = -1,	/* PCH without south display */
+	PCH_NONE = 0,	/* No PCH present */
+	PCH_IBX,	/* Ibexpeak PCH */
+	PCH_CPT,	/* Cougarpoint/Pantherpoint PCH */
+	PCH_LPT_H,	/* Lynxpoint/Wildcatpoint H PCH */
+	PCH_LPT_LP,	/* Lynxpoint/Wildcatpoint LP PCH */
+	PCH_SPT,        /* Sunrisepoint/Kaby Lake PCH */
+	PCH_CNP,        /* Cannon/Comet Lake PCH */
+	PCH_ICP,	/* Ice Lake/Jasper Lake PCH */
+	PCH_TGP,	/* Tiger Lake/Mule Creek Canyon PCH */
+	PCH_ADP,	/* Alder Lake PCH */
+
+	/* Fake PCHs, functionality handled on the same PCI dev */
+	PCH_DG1 = 1024,
+	PCH_DG2,
+	PCH_MTL,
+	PCH_LNL,
+};
+
+#define INTEL_PCH_TYPE(_display)		(__to_intel_display(_display)->pch_type)
+#define HAS_PCH_DG2(display)			(INTEL_PCH_TYPE(display) == PCH_DG2)
+#define HAS_PCH_ADP(display)			(INTEL_PCH_TYPE(display) == PCH_ADP)
+#define HAS_PCH_DG1(display)			(INTEL_PCH_TYPE(display) == PCH_DG1)
+#define HAS_PCH_TGP(display)			(INTEL_PCH_TYPE(display) == PCH_TGP)
+#define HAS_PCH_ICP(display)			(INTEL_PCH_TYPE(display) == PCH_ICP)
+#define HAS_PCH_CNP(display)			(INTEL_PCH_TYPE(display) == PCH_CNP)
+#define HAS_PCH_SPT(display)			(INTEL_PCH_TYPE(display) == PCH_SPT)
+#define HAS_PCH_LPT_H(display)			(INTEL_PCH_TYPE(display) == PCH_LPT_H)
+#define HAS_PCH_LPT_LP(display)			(INTEL_PCH_TYPE(display) == PCH_LPT_LP)
+#define HAS_PCH_LPT(display)			(INTEL_PCH_TYPE(display) == PCH_LPT_H || \
+						 INTEL_PCH_TYPE(display) == PCH_LPT_LP)
+#define HAS_PCH_CPT(display)			(INTEL_PCH_TYPE(display) == PCH_CPT)
+#define HAS_PCH_IBX(display)			(INTEL_PCH_TYPE(display) == PCH_IBX)
+#define HAS_PCH_NOP(display)			(INTEL_PCH_TYPE(display) == PCH_NOP)
+#define HAS_PCH_SPLIT(display)			(INTEL_PCH_TYPE(display) != PCH_NONE)
+
+void intel_pch_detect(struct intel_display *display);
+
+#endif /* __INTEL_PCH__ */
diff --git a/drivers/gpu/drm/i915/display/intel_pch_display.c b/drivers/gpu/drm/i915/display/intel_pch_display.c
index 99f6d6f53fa7..1743ebf551cb 100644
--- a/drivers/gpu/drm/i915/display/intel_pch_display.c
+++ b/drivers/gpu/drm/i915/display/intel_pch_display.c
@@ -3,8 +3,9 @@
  * Copyright © 2021 Intel Corporation
  */
 
+#include <drm/drm_print.h>
+
 #include "g4x_dp.h"
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_crt.h"
 #include "intel_crt_regs.h"
@@ -20,28 +21,27 @@
 #include "intel_pps.h"
 #include "intel_sdvo.h"
 
-bool intel_has_pch_trancoder(struct drm_i915_private *i915,
+bool intel_has_pch_trancoder(struct intel_display *display,
 			     enum pipe pch_transcoder)
 {
-	return HAS_PCH_IBX(i915) || HAS_PCH_CPT(i915) ||
-		(HAS_PCH_LPT_H(i915) && pch_transcoder == PIPE_A);
+	return HAS_PCH_IBX(display) || HAS_PCH_CPT(display) ||
+		(HAS_PCH_LPT_H(display) && pch_transcoder == PIPE_A);
 }
 
 enum pipe intel_crtc_pch_transcoder(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 
-	if (HAS_PCH_LPT(i915))
+	if (HAS_PCH_LPT(display))
 		return PIPE_A;
 	else
 		return crtc->pipe;
 }
 
-static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv,
+static void assert_pch_dp_disabled(struct intel_display *display,
 				   enum pipe pipe, enum port port,
 				   i915_reg_t dp_reg)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe port_pipe;
 	bool state;
 
@@ -52,16 +52,15 @@ static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv,
 				 port_name(port), pipe_name(pipe));
 
 	INTEL_DISPLAY_STATE_WARN(display,
-				 HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B,
+				 HAS_PCH_IBX(display) && !state && port_pipe == PIPE_B,
 				 "IBX PCH DP %c still using transcoder B\n",
 				 port_name(port));
 }
 
-static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
+static void assert_pch_hdmi_disabled(struct intel_display *display,
 				     enum pipe pipe, enum port port,
 				     i915_reg_t hdmi_reg)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe port_pipe;
 	bool state;
 
@@ -72,20 +71,19 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
 				 port_name(port), pipe_name(pipe));
 
 	INTEL_DISPLAY_STATE_WARN(display,
-				 HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B,
+				 HAS_PCH_IBX(display) && !state && port_pipe == PIPE_B,
 				 "IBX PCH HDMI %c still using transcoder B\n",
 				 port_name(port));
 }
 
-static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
+static void assert_pch_ports_disabled(struct intel_display *display,
 				      enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
 	enum pipe port_pipe;
 
-	assert_pch_dp_disabled(dev_priv, pipe, PORT_B, PCH_DP_B);
-	assert_pch_dp_disabled(dev_priv, pipe, PORT_C, PCH_DP_C);
-	assert_pch_dp_disabled(dev_priv, pipe, PORT_D, PCH_DP_D);
+	assert_pch_dp_disabled(display, pipe, PORT_B, PCH_DP_B);
+	assert_pch_dp_disabled(display, pipe, PORT_C, PCH_DP_C);
+	assert_pch_dp_disabled(display, pipe, PORT_D, PCH_DP_D);
 
 	INTEL_DISPLAY_STATE_WARN(display,
 				 intel_crt_port_enabled(display, PCH_ADPA, &port_pipe) && port_pipe == pipe,
@@ -93,20 +91,19 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
 				 pipe_name(pipe));
 
 	INTEL_DISPLAY_STATE_WARN(display,
-				 intel_lvds_port_enabled(dev_priv, PCH_LVDS, &port_pipe) && port_pipe == pipe,
+				 intel_lvds_port_enabled(display, PCH_LVDS, &port_pipe) && port_pipe == pipe,
 				 "PCH LVDS enabled on transcoder %c, should be disabled\n",
 				 pipe_name(pipe));
 
 	/* PCH SDVOB multiplex with HDMIB */
-	assert_pch_hdmi_disabled(dev_priv, pipe, PORT_B, PCH_HDMIB);
-	assert_pch_hdmi_disabled(dev_priv, pipe, PORT_C, PCH_HDMIC);
-	assert_pch_hdmi_disabled(dev_priv, pipe, PORT_D, PCH_HDMID);
+	assert_pch_hdmi_disabled(display, pipe, PORT_B, PCH_HDMIB);
+	assert_pch_hdmi_disabled(display, pipe, PORT_C, PCH_HDMIC);
+	assert_pch_hdmi_disabled(display, pipe, PORT_D, PCH_HDMID);
 }
 
-static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
+static void assert_pch_transcoder_disabled(struct intel_display *display,
 					   enum pipe pipe)
 {
-	struct intel_display *display = &dev_priv->display;
 	u32 val;
 	bool enabled;
 
@@ -117,45 +114,45 @@ static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
 				 pipe_name(pipe));
 }
 
-static void ibx_sanitize_pch_hdmi_port(struct drm_i915_private *dev_priv,
+static void ibx_sanitize_pch_hdmi_port(struct intel_display *display,
 				       enum port port, i915_reg_t hdmi_reg)
 {
-	u32 val = intel_de_read(dev_priv, hdmi_reg);
+	u32 val = intel_de_read(display, hdmi_reg);
 
 	if (val & SDVO_ENABLE ||
 	    (val & SDVO_PIPE_SEL_MASK) == SDVO_PIPE_SEL(PIPE_A))
 		return;
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Sanitizing transcoder select for HDMI %c\n",
 		    port_name(port));
 
 	val &= ~SDVO_PIPE_SEL_MASK;
 	val |= SDVO_PIPE_SEL(PIPE_A);
 
-	intel_de_write(dev_priv, hdmi_reg, val);
+	intel_de_write(display, hdmi_reg, val);
 }
 
-static void ibx_sanitize_pch_dp_port(struct drm_i915_private *dev_priv,
+static void ibx_sanitize_pch_dp_port(struct intel_display *display,
 				     enum port port, i915_reg_t dp_reg)
 {
-	u32 val = intel_de_read(dev_priv, dp_reg);
+	u32 val = intel_de_read(display, dp_reg);
 
 	if (val & DP_PORT_EN ||
 	    (val & DP_PIPE_SEL_MASK) == DP_PIPE_SEL(PIPE_A))
 		return;
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "Sanitizing transcoder select for DP %c\n",
 		    port_name(port));
 
 	val &= ~DP_PIPE_SEL_MASK;
 	val |= DP_PIPE_SEL(PIPE_A);
 
-	intel_de_write(dev_priv, dp_reg, val);
+	intel_de_write(display, dp_reg, val);
 }
 
-static void ibx_sanitize_pch_ports(struct drm_i915_private *dev_priv)
+static void ibx_sanitize_pch_ports(struct intel_display *display)
 {
 	/*
 	 * The BIOS may select transcoder B on some of the PCH
@@ -168,14 +165,14 @@ static void ibx_sanitize_pch_ports(struct drm_i915_private *dev_priv)
 	 * (see. intel_dp_link_down(), intel_disable_hdmi(),
 	 * intel_disable_sdvo()).
 	 */
-	ibx_sanitize_pch_dp_port(dev_priv, PORT_B, PCH_DP_B);
-	ibx_sanitize_pch_dp_port(dev_priv, PORT_C, PCH_DP_C);
-	ibx_sanitize_pch_dp_port(dev_priv, PORT_D, PCH_DP_D);
+	ibx_sanitize_pch_dp_port(display, PORT_B, PCH_DP_B);
+	ibx_sanitize_pch_dp_port(display, PORT_C, PCH_DP_C);
+	ibx_sanitize_pch_dp_port(display, PORT_D, PCH_DP_D);
 
 	/* PCH SDVOB multiplex with HDMIB */
-	ibx_sanitize_pch_hdmi_port(dev_priv, PORT_B, PCH_HDMIB);
-	ibx_sanitize_pch_hdmi_port(dev_priv, PORT_C, PCH_HDMIC);
-	ibx_sanitize_pch_hdmi_port(dev_priv, PORT_D, PCH_HDMID);
+	ibx_sanitize_pch_hdmi_port(display, PORT_B, PCH_HDMIB);
+	ibx_sanitize_pch_hdmi_port(display, PORT_C, PCH_HDMIC);
+	ibx_sanitize_pch_hdmi_port(display, PORT_D, PCH_HDMID);
 }
 
 static void intel_pch_transcoder_set_m1_n1(struct intel_crtc *crtc,
@@ -225,32 +222,30 @@ void intel_pch_transcoder_get_m2_n2(struct intel_crtc *crtc,
 static void ilk_pch_transcoder_set_timings(const struct intel_crtc_state *crtc_state,
 					   enum pipe pch_transcoder)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 
-	intel_de_write(dev_priv, PCH_TRANS_HTOTAL(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_HTOTAL(dev_priv, cpu_transcoder)));
-	intel_de_write(dev_priv, PCH_TRANS_HBLANK(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_HBLANK(dev_priv, cpu_transcoder)));
-	intel_de_write(dev_priv, PCH_TRANS_HSYNC(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_HSYNC(dev_priv, cpu_transcoder)));
-
-	intel_de_write(dev_priv, PCH_TRANS_VTOTAL(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_VTOTAL(dev_priv, cpu_transcoder)));
-	intel_de_write(dev_priv, PCH_TRANS_VBLANK(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_VBLANK(dev_priv, cpu_transcoder)));
-	intel_de_write(dev_priv, PCH_TRANS_VSYNC(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_VSYNC(dev_priv, cpu_transcoder)));
-	intel_de_write(dev_priv, PCH_TRANS_VSYNCSHIFT(pch_transcoder),
-		       intel_de_read(dev_priv, TRANS_VSYNCSHIFT(dev_priv, cpu_transcoder)));
+	intel_de_write(display, PCH_TRANS_HTOTAL(pch_transcoder),
+		       intel_de_read(display, TRANS_HTOTAL(display, cpu_transcoder)));
+	intel_de_write(display, PCH_TRANS_HBLANK(pch_transcoder),
+		       intel_de_read(display, TRANS_HBLANK(display, cpu_transcoder)));
+	intel_de_write(display, PCH_TRANS_HSYNC(pch_transcoder),
+		       intel_de_read(display, TRANS_HSYNC(display, cpu_transcoder)));
+
+	intel_de_write(display, PCH_TRANS_VTOTAL(pch_transcoder),
+		       intel_de_read(display, TRANS_VTOTAL(display, cpu_transcoder)));
+	intel_de_write(display, PCH_TRANS_VBLANK(pch_transcoder),
+		       intel_de_read(display, TRANS_VBLANK(display, cpu_transcoder)));
+	intel_de_write(display, PCH_TRANS_VSYNC(pch_transcoder),
+		       intel_de_read(display, TRANS_VSYNC(display, cpu_transcoder)));
+	intel_de_write(display, PCH_TRANS_VSYNCSHIFT(pch_transcoder),
+		       intel_de_read(display, TRANS_VSYNCSHIFT(display, cpu_transcoder)));
 }
 
 static void ilk_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	i915_reg_t reg;
 	u32 val, pipeconf_val;
@@ -262,7 +257,7 @@ static void ilk_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 	assert_fdi_tx_enabled(display, pipe);
 	assert_fdi_rx_enabled(display, pipe);
 
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		reg = TRANS_CHICKEN2(pipe);
 		val = intel_de_read(display, reg);
 		/*
@@ -280,7 +275,7 @@ static void ilk_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 	val = intel_de_read(display, reg);
 	pipeconf_val = intel_de_read(display, TRANSCONF(display, pipe));
 
-	if (HAS_PCH_IBX(dev_priv)) {
+	if (HAS_PCH_IBX(display)) {
 		/* Configure frame start delay to match the CPU */
 		val &= ~TRANS_FRAME_START_DELAY_MASK;
 		val |= TRANS_FRAME_START_DELAY(crtc_state->framestart_delay - 1);
@@ -299,7 +294,7 @@ static void ilk_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 
 	val &= ~TRANS_INTERLACE_MASK;
 	if ((pipeconf_val & TRANSCONF_INTERLACE_MASK_ILK) == TRANSCONF_INTERLACE_IF_ID_ILK) {
-		if (HAS_PCH_IBX(dev_priv) &&
+		if (HAS_PCH_IBX(display) &&
 		    intel_crtc_has_type(crtc_state, INTEL_OUTPUT_SDVO))
 			val |= TRANS_INTERLACE_LEGACY_VSYNC_IBX;
 		else
@@ -317,7 +312,6 @@ static void ilk_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 static void ilk_disable_pch_transcoder(struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	i915_reg_t reg;
 
@@ -326,18 +320,18 @@ static void ilk_disable_pch_transcoder(struct intel_crtc *crtc)
 	assert_fdi_rx_disabled(display, pipe);
 
 	/* Ports must be off as well */
-	assert_pch_ports_disabled(dev_priv, pipe);
+	assert_pch_ports_disabled(display, pipe);
 
 	reg = PCH_TRANSCONF(pipe);
-	intel_de_rmw(dev_priv, reg, TRANS_ENABLE, 0);
+	intel_de_rmw(display, reg, TRANS_ENABLE, 0);
 	/* wait for PCH transcoder off, transcoder state */
-	if (intel_de_wait_for_clear(dev_priv, reg, TRANS_STATE_ENABLE, 50))
-		drm_err(&dev_priv->drm, "failed to disable transcoder %c\n",
+	if (intel_de_wait_for_clear(display, reg, TRANS_STATE_ENABLE, 50))
+		drm_err(display->drm, "failed to disable transcoder %c\n",
 			pipe_name(pipe));
 
-	if (HAS_PCH_CPT(dev_priv))
+	if (HAS_PCH_CPT(display))
 		/* Workaround: Clear the timing override chicken bit again. */
-		intel_de_rmw(dev_priv, TRANS_CHICKEN2(pipe),
+		intel_de_rmw(display, TRANS_CHICKEN2(pipe),
 			     TRANS_CHICKEN2_TIMING_OVERRIDE, 0);
 }
 
@@ -366,14 +360,13 @@ void ilk_pch_pre_enable(struct intel_atomic_state *state,
 void ilk_pch_enable(struct intel_atomic_state *state,
 		    struct intel_crtc *crtc)
 {
-	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	enum pipe pipe = crtc->pipe;
 	u32 temp;
 
-	assert_pch_transcoder_disabled(dev_priv, pipe);
+	assert_pch_transcoder_disabled(display, pipe);
 
 	/* For PCH output, training FDI link */
 	intel_fdi_link_train(crtc, crtc_state);
@@ -382,7 +375,7 @@ void ilk_pch_enable(struct intel_atomic_state *state,
 	 * We need to program the right clock selection
 	 * before writing the pixel multiplier into the DPLL.
 	 */
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		u32 sel;
 
 		temp = intel_de_read(display, PCH_DPLL_SEL);
@@ -418,7 +411,7 @@ void ilk_pch_enable(struct intel_atomic_state *state,
 	intel_fdi_normal_train(crtc);
 
 	/* For PCH DP, enable TRANS_DP_CTL */
-	if (HAS_PCH_CPT(dev_priv) &&
+	if (HAS_PCH_CPT(display) &&
 	    intel_crtc_has_dp_encoder(crtc_state)) {
 		const struct drm_display_mode *adjusted_mode =
 			&crtc_state->hw.adjusted_mode;
@@ -459,23 +452,27 @@ void ilk_pch_disable(struct intel_atomic_state *state,
 void ilk_pch_post_disable(struct intel_atomic_state *state,
 			  struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
+	const struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
 	enum pipe pipe = crtc->pipe;
 
 	ilk_disable_pch_transcoder(crtc);
 
-	if (HAS_PCH_CPT(dev_priv)) {
+	if (HAS_PCH_CPT(display)) {
 		/* disable TRANS_DP_CTL */
-		intel_de_rmw(dev_priv, TRANS_DP_CTL(pipe),
+		intel_de_rmw(display, TRANS_DP_CTL(pipe),
 			     TRANS_DP_OUTPUT_ENABLE | TRANS_DP_PORT_SEL_MASK,
 			     TRANS_DP_PORT_SEL_NONE);
 
 		/* disable DPLL_SEL */
-		intel_de_rmw(dev_priv, PCH_DPLL_SEL,
+		intel_de_rmw(display, PCH_DPLL_SEL,
 			     TRANS_DPLL_ENABLE(pipe) | TRANS_DPLLB_SEL(pipe), 0);
 	}
 
 	ilk_fdi_pll_disable(crtc);
+
+	intel_disable_shared_dpll(old_crtc_state);
 }
 
 static void ilk_pch_clock_get(struct intel_crtc_state *crtc_state)
@@ -497,9 +494,8 @@ static void ilk_pch_clock_get(struct intel_crtc_state *crtc_state)
 
 void ilk_pch_get_config(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_shared_dpll *pll;
 	enum pipe pipe = crtc->pipe;
 	enum intel_dpll_id pll_id;
@@ -518,7 +514,7 @@ void ilk_pch_get_config(struct intel_crtc_state *crtc_state)
 	intel_cpu_transcoder_get_m1_n1(crtc, crtc_state->cpu_transcoder,
 				       &crtc_state->fdi_m_n);
 
-	if (HAS_PCH_IBX(dev_priv)) {
+	if (HAS_PCH_IBX(display)) {
 		/*
 		 * The pipe->pch transcoder and pch transcoder->pll
 		 * mapping is fixed.
@@ -550,8 +546,6 @@ void ilk_pch_get_config(struct intel_crtc_state *crtc_state)
 static void lpt_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 val, pipeconf_val;
 
@@ -559,49 +553,49 @@ static void lpt_enable_pch_transcoder(const struct intel_crtc_state *crtc_state)
 	assert_fdi_tx_enabled(display, (enum pipe)cpu_transcoder);
 	assert_fdi_rx_enabled(display, PIPE_A);
 
-	val = intel_de_read(dev_priv, TRANS_CHICKEN2(PIPE_A));
+	val = intel_de_read(display, TRANS_CHICKEN2(PIPE_A));
 	/* Workaround: set timing override bit. */
 	val |= TRANS_CHICKEN2_TIMING_OVERRIDE;
 	/* Configure frame start delay to match the CPU */
 	val &= ~TRANS_CHICKEN2_FRAME_START_DELAY_MASK;
 	val |= TRANS_CHICKEN2_FRAME_START_DELAY(crtc_state->framestart_delay - 1);
-	intel_de_write(dev_priv, TRANS_CHICKEN2(PIPE_A), val);
+	intel_de_write(display, TRANS_CHICKEN2(PIPE_A), val);
 
 	val = TRANS_ENABLE;
-	pipeconf_val = intel_de_read(dev_priv,
-				     TRANSCONF(dev_priv, cpu_transcoder));
+	pipeconf_val = intel_de_read(display,
+				     TRANSCONF(display, cpu_transcoder));
 
 	if ((pipeconf_val & TRANSCONF_INTERLACE_MASK_HSW) == TRANSCONF_INTERLACE_IF_ID_ILK)
 		val |= TRANS_INTERLACE_INTERLACED;
 	else
 		val |= TRANS_INTERLACE_PROGRESSIVE;
 
-	intel_de_write(dev_priv, LPT_TRANSCONF, val);
-	if (intel_de_wait_for_set(dev_priv, LPT_TRANSCONF,
+	intel_de_write(display, LPT_TRANSCONF, val);
+	if (intel_de_wait_for_set(display, LPT_TRANSCONF,
 				  TRANS_STATE_ENABLE, 100))
-		drm_err(&dev_priv->drm, "Failed to enable PCH transcoder\n");
+		drm_err(display->drm, "Failed to enable PCH transcoder\n");
 }
 
-static void lpt_disable_pch_transcoder(struct drm_i915_private *dev_priv)
+static void lpt_disable_pch_transcoder(struct intel_display *display)
 {
-	intel_de_rmw(dev_priv, LPT_TRANSCONF, TRANS_ENABLE, 0);
+	intel_de_rmw(display, LPT_TRANSCONF, TRANS_ENABLE, 0);
 	/* wait for PCH transcoder off, transcoder state */
-	if (intel_de_wait_for_clear(dev_priv, LPT_TRANSCONF,
+	if (intel_de_wait_for_clear(display, LPT_TRANSCONF,
 				    TRANS_STATE_ENABLE, 50))
-		drm_err(&dev_priv->drm, "Failed to disable PCH transcoder\n");
+		drm_err(display->drm, "Failed to disable PCH transcoder\n");
 
 	/* Workaround: clear timing override bit. */
-	intel_de_rmw(dev_priv, TRANS_CHICKEN2(PIPE_A), TRANS_CHICKEN2_TIMING_OVERRIDE, 0);
+	intel_de_rmw(display, TRANS_CHICKEN2(PIPE_A), TRANS_CHICKEN2_TIMING_OVERRIDE, 0);
 }
 
 void lpt_pch_enable(struct intel_atomic_state *state,
 		    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 
-	assert_pch_transcoder_disabled(dev_priv, PIPE_A);
+	assert_pch_transcoder_disabled(display, PIPE_A);
 
 	lpt_program_iclkip(crtc_state);
 
@@ -614,36 +608,36 @@ void lpt_pch_enable(struct intel_atomic_state *state,
 void lpt_pch_disable(struct intel_atomic_state *state,
 		     struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 
-	lpt_disable_pch_transcoder(dev_priv);
+	lpt_disable_pch_transcoder(display);
 
-	lpt_disable_iclkip(dev_priv);
+	lpt_disable_iclkip(display);
 }
 
 void lpt_pch_get_config(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 tmp;
 
-	if ((intel_de_read(dev_priv, LPT_TRANSCONF) & TRANS_ENABLE) == 0)
+	if ((intel_de_read(display, LPT_TRANSCONF) & TRANS_ENABLE) == 0)
 		return;
 
 	crtc_state->has_pch_encoder = true;
 
-	tmp = intel_de_read(dev_priv, FDI_RX_CTL(PIPE_A));
+	tmp = intel_de_read(display, FDI_RX_CTL(PIPE_A));
 	crtc_state->fdi_lanes = ((FDI_DP_PORT_WIDTH_MASK & tmp) >>
 				 FDI_DP_PORT_WIDTH_SHIFT) + 1;
 
 	intel_cpu_transcoder_get_m1_n1(crtc, crtc_state->cpu_transcoder,
 				       &crtc_state->fdi_m_n);
 
-	crtc_state->hw.adjusted_mode.crtc_clock = lpt_get_iclkip(dev_priv);
+	crtc_state->hw.adjusted_mode.crtc_clock = lpt_get_iclkip(display);
 }
 
-void intel_pch_sanitize(struct drm_i915_private *i915)
+void intel_pch_sanitize(struct intel_display *display)
 {
-	if (HAS_PCH_IBX(i915))
-		ibx_sanitize_pch_ports(i915);
+	if (HAS_PCH_IBX(display))
+		ibx_sanitize_pch_ports(display);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_pch_display.h b/drivers/gpu/drm/i915/display/intel_pch_display.h
index 35f8288af3d1..cd6b3ed05887 100644
--- a/drivers/gpu/drm/i915/display/intel_pch_display.h
+++ b/drivers/gpu/drm/i915/display/intel_pch_display.h
@@ -9,14 +9,14 @@
 #include <linux/types.h>
 
 enum pipe;
-struct drm_i915_private;
 struct intel_atomic_state;
 struct intel_crtc;
 struct intel_crtc_state;
+struct intel_display;
 struct intel_link_m_n;
 
 #ifdef I915
-bool intel_has_pch_trancoder(struct drm_i915_private *i915,
+bool intel_has_pch_trancoder(struct intel_display *display,
 			     enum pipe pch_transcoder);
 enum pipe intel_crtc_pch_transcoder(struct intel_crtc *crtc);
 
@@ -41,9 +41,9 @@ void intel_pch_transcoder_get_m1_n1(struct intel_crtc *crtc,
 void intel_pch_transcoder_get_m2_n2(struct intel_crtc *crtc,
 				    struct intel_link_m_n *m_n);
 
-void intel_pch_sanitize(struct drm_i915_private *i915);
+void intel_pch_sanitize(struct intel_display *display);
 #else
-static inline bool intel_has_pch_trancoder(struct drm_i915_private *i915,
+static inline bool intel_has_pch_trancoder(struct intel_display *display,
 					   enum pipe pch_transcoder)
 {
 	return false;
@@ -90,7 +90,7 @@ static inline void intel_pch_transcoder_get_m2_n2(struct intel_crtc *crtc,
 						  struct intel_link_m_n *m_n)
 {
 }
-static inline void intel_pch_sanitize(struct drm_i915_private *i915)
+static inline void intel_pch_sanitize(struct intel_display *display)
 {
 }
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_pch_refclk.c b/drivers/gpu/drm/i915/display/intel_pch_refclk.c
index 33467de3d115..693b90e3dfc3 100644
--- a/drivers/gpu/drm/i915/display/intel_pch_refclk.c
+++ b/drivers/gpu/drm/i915/display/intel_pch_refclk.c
@@ -11,27 +11,28 @@
 #include "intel_pch_refclk.h"
 #include "intel_sbi.h"
 
-static void lpt_fdi_reset_mphy(struct drm_i915_private *dev_priv)
+static void lpt_fdi_reset_mphy(struct intel_display *display)
 {
-	intel_de_rmw(dev_priv, SOUTH_CHICKEN2, 0, FDI_MPHY_IOSFSB_RESET_CTL);
+	intel_de_rmw(display, SOUTH_CHICKEN2, 0, FDI_MPHY_IOSFSB_RESET_CTL);
 
-	if (wait_for_us(intel_de_read(dev_priv, SOUTH_CHICKEN2) &
+	if (wait_for_us(intel_de_read(display, SOUTH_CHICKEN2) &
 			FDI_MPHY_IOSFSB_RESET_STATUS, 100))
-		drm_err(&dev_priv->drm, "FDI mPHY reset assert timeout\n");
+		drm_err(display->drm, "FDI mPHY reset assert timeout\n");
 
-	intel_de_rmw(dev_priv, SOUTH_CHICKEN2, FDI_MPHY_IOSFSB_RESET_CTL, 0);
+	intel_de_rmw(display, SOUTH_CHICKEN2, FDI_MPHY_IOSFSB_RESET_CTL, 0);
 
-	if (wait_for_us((intel_de_read(dev_priv, SOUTH_CHICKEN2) &
+	if (wait_for_us((intel_de_read(display, SOUTH_CHICKEN2) &
 			 FDI_MPHY_IOSFSB_RESET_STATUS) == 0, 100))
-		drm_err(&dev_priv->drm, "FDI mPHY reset de-assert timeout\n");
+		drm_err(display->drm, "FDI mPHY reset de-assert timeout\n");
 }
 
 /* WaMPhyProgramming:hsw */
-static void lpt_fdi_program_mphy(struct drm_i915_private *dev_priv)
+static void lpt_fdi_program_mphy(struct intel_display *display)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 tmp;
 
-	lpt_fdi_reset_mphy(dev_priv);
+	lpt_fdi_reset_mphy(display);
 
 	tmp = intel_sbi_read(dev_priv, 0x8008, SBI_MPHY);
 	tmp &= ~(0xFF << 24);
@@ -103,11 +104,12 @@ static void lpt_fdi_program_mphy(struct drm_i915_private *dev_priv)
 	intel_sbi_write(dev_priv, 0x21EC, tmp, SBI_MPHY);
 }
 
-void lpt_disable_iclkip(struct drm_i915_private *dev_priv)
+void lpt_disable_iclkip(struct intel_display *display)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 temp;
 
-	intel_de_write(dev_priv, PIXCLK_GATE, PIXCLK_GATE_GATE);
+	intel_de_write(display, PIXCLK_GATE, PIXCLK_GATE_GATE);
 
 	intel_sbi_lock(dev_priv);
 
@@ -175,24 +177,25 @@ int lpt_iclkip(const struct intel_crtc_state *crtc_state)
 /* Program iCLKIP clock to the desired frequency */
 void lpt_program_iclkip(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	int clock = crtc_state->hw.adjusted_mode.crtc_clock;
 	struct iclkip_params p;
 	u32 temp;
 
-	lpt_disable_iclkip(dev_priv);
+	lpt_disable_iclkip(display);
 
 	lpt_compute_iclkip(&p, clock);
-	drm_WARN_ON(&dev_priv->drm, lpt_iclkip_freq(&p) != clock);
+	drm_WARN_ON(display->drm, lpt_iclkip_freq(&p) != clock);
 
 	/* This should not happen with any sane values */
-	drm_WARN_ON(&dev_priv->drm, SBI_SSCDIVINTPHASE_DIVSEL(p.divsel) &
+	drm_WARN_ON(display->drm, SBI_SSCDIVINTPHASE_DIVSEL(p.divsel) &
 		    ~SBI_SSCDIVINTPHASE_DIVSEL_MASK);
-	drm_WARN_ON(&dev_priv->drm, SBI_SSCDIVINTPHASE_DIR(p.phasedir) &
+	drm_WARN_ON(display->drm, SBI_SSCDIVINTPHASE_DIR(p.phasedir) &
 		    ~SBI_SSCDIVINTPHASE_INCVAL_MASK);
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(display->drm,
 		    "iCLKIP clock: found settings for %dKHz refresh rate: auxdiv=%x, divsel=%x, phasedir=%x, phaseinc=%x\n",
 		    clock, p.auxdiv, p.divsel, p.phasedir, p.phaseinc);
 
@@ -224,15 +227,16 @@ void lpt_program_iclkip(const struct intel_crtc_state *crtc_state)
 	/* Wait for initialization time */
 	udelay(24);
 
-	intel_de_write(dev_priv, PIXCLK_GATE, PIXCLK_GATE_UNGATE);
+	intel_de_write(display, PIXCLK_GATE, PIXCLK_GATE_UNGATE);
 }
 
-int lpt_get_iclkip(struct drm_i915_private *dev_priv)
+int lpt_get_iclkip(struct intel_display *display)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct iclkip_params p;
 	u32 temp;
 
-	if ((intel_de_read(dev_priv, PIXCLK_GATE) & PIXCLK_GATE_UNGATE) == 0)
+	if ((intel_de_read(display, PIXCLK_GATE) & PIXCLK_GATE_UNGATE) == 0)
 		return 0;
 
 	iclkip_params_init(&p);
@@ -268,15 +272,16 @@ int lpt_get_iclkip(struct drm_i915_private *dev_priv)
  * - Sequence to enable CLKOUT_DP without spread
  * - Sequence to enable CLKOUT_DP for FDI usage and configure PCH FDI I/O
  */
-static void lpt_enable_clkout_dp(struct drm_i915_private *dev_priv,
+static void lpt_enable_clkout_dp(struct intel_display *display,
 				 bool with_spread, bool with_fdi)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 reg, tmp;
 
-	if (drm_WARN(&dev_priv->drm, with_fdi && !with_spread,
+	if (drm_WARN(display->drm, with_fdi && !with_spread,
 		     "FDI requires downspread\n"))
 		with_spread = true;
-	if (drm_WARN(&dev_priv->drm, HAS_PCH_LPT_LP(dev_priv) &&
+	if (drm_WARN(display->drm, HAS_PCH_LPT_LP(display) &&
 		     with_fdi, "LP PCH doesn't have FDI\n"))
 		with_fdi = false;
 
@@ -295,10 +300,10 @@ static void lpt_enable_clkout_dp(struct drm_i915_private *dev_priv,
 		intel_sbi_write(dev_priv, SBI_SSCCTL, tmp, SBI_ICLK);
 
 		if (with_fdi)
-			lpt_fdi_program_mphy(dev_priv);
+			lpt_fdi_program_mphy(display);
 	}
 
-	reg = HAS_PCH_LPT_LP(dev_priv) ? SBI_GEN0 : SBI_DBUFF0;
+	reg = HAS_PCH_LPT_LP(display) ? SBI_GEN0 : SBI_DBUFF0;
 	tmp = intel_sbi_read(dev_priv, reg, SBI_ICLK);
 	tmp |= SBI_GEN0_CFG_BUFFENABLE_DISABLE;
 	intel_sbi_write(dev_priv, reg, tmp, SBI_ICLK);
@@ -307,13 +312,14 @@ static void lpt_enable_clkout_dp(struct drm_i915_private *dev_priv,
 }
 
 /* Sequence to disable CLKOUT_DP */
-void lpt_disable_clkout_dp(struct drm_i915_private *dev_priv)
+void lpt_disable_clkout_dp(struct intel_display *display)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 reg, tmp;
 
 	intel_sbi_lock(dev_priv);
 
-	reg = HAS_PCH_LPT_LP(dev_priv) ? SBI_GEN0 : SBI_DBUFF0;
+	reg = HAS_PCH_LPT_LP(display) ? SBI_GEN0 : SBI_DBUFF0;
 	tmp = intel_sbi_read(dev_priv, reg, SBI_ICLK);
 	tmp &= ~SBI_GEN0_CFG_BUFFENABLE_DISABLE;
 	intel_sbi_write(dev_priv, reg, tmp, SBI_ICLK);
@@ -364,15 +370,16 @@ static const u16 sscdivintphase[] = {
  * < 0 slow down the clock, > 0 speed up the clock, 0 == no bend (135MHz)
  * change in clock period = -(steps / 10) * 5.787 ps
  */
-static void lpt_bend_clkout_dp(struct drm_i915_private *dev_priv, int steps)
+static void lpt_bend_clkout_dp(struct intel_display *display, int steps)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 tmp;
 	int idx = BEND_IDX(steps);
 
-	if (drm_WARN_ON(&dev_priv->drm, steps % 5 != 0))
+	if (drm_WARN_ON(display->drm, steps % 5 != 0))
 		return;
 
-	if (drm_WARN_ON(&dev_priv->drm, idx >= ARRAY_SIZE(sscdivintphase)))
+	if (drm_WARN_ON(display->drm, idx >= ARRAY_SIZE(sscdivintphase)))
 		return;
 
 	intel_sbi_lock(dev_priv);
@@ -393,10 +400,10 @@ static void lpt_bend_clkout_dp(struct drm_i915_private *dev_priv, int steps)
 
 #undef BEND_IDX
 
-static bool spll_uses_pch_ssc(struct drm_i915_private *dev_priv)
+static bool spll_uses_pch_ssc(struct intel_display *display)
 {
-	u32 fuse_strap = intel_de_read(dev_priv, FUSE_STRAP);
-	u32 ctl = intel_de_read(dev_priv, SPLL_CTL);
+	u32 fuse_strap = intel_de_read(display, FUSE_STRAP);
+	u32 ctl = intel_de_read(display, SPLL_CTL);
 
 	if ((ctl & SPLL_PLL_ENABLE) == 0)
 		return false;
@@ -405,18 +412,17 @@ static bool spll_uses_pch_ssc(struct drm_i915_private *dev_priv)
 	    (fuse_strap & HSW_CPU_SSC_ENABLE) == 0)
 		return true;
 
-	if (IS_BROADWELL(dev_priv) &&
+	if (display->platform.broadwell &&
 	    (ctl & SPLL_REF_MASK) == SPLL_REF_PCH_SSC_BDW)
 		return true;
 
 	return false;
 }
 
-static bool wrpll_uses_pch_ssc(struct drm_i915_private *dev_priv,
-			       enum intel_dpll_id id)
+static bool wrpll_uses_pch_ssc(struct intel_display *display, enum intel_dpll_id id)
 {
-	u32 fuse_strap = intel_de_read(dev_priv, FUSE_STRAP);
-	u32 ctl = intel_de_read(dev_priv, WRPLL_CTL(id));
+	u32 fuse_strap = intel_de_read(display, FUSE_STRAP);
+	u32 ctl = intel_de_read(display, WRPLL_CTL(id));
 
 	if ((ctl & WRPLL_PLL_ENABLE) == 0)
 		return false;
@@ -424,7 +430,7 @@ static bool wrpll_uses_pch_ssc(struct drm_i915_private *dev_priv,
 	if ((ctl & WRPLL_REF_MASK) == WRPLL_REF_PCH_SSC)
 		return true;
 
-	if ((IS_BROADWELL(dev_priv) || IS_HASWELL_ULT(dev_priv)) &&
+	if ((display->platform.broadwell || display->platform.haswell_ult) &&
 	    (ctl & WRPLL_REF_MASK) == WRPLL_REF_MUXED_SSC_BDW &&
 	    (fuse_strap & HSW_CPU_SSC_ENABLE) == 0)
 		return true;
@@ -432,12 +438,12 @@ static bool wrpll_uses_pch_ssc(struct drm_i915_private *dev_priv,
 	return false;
 }
 
-static void lpt_init_pch_refclk(struct drm_i915_private *dev_priv)
+static void lpt_init_pch_refclk(struct intel_display *display)
 {
 	struct intel_encoder *encoder;
 	bool has_fdi = false;
 
-	for_each_intel_encoder(&dev_priv->drm, encoder) {
+	for_each_intel_encoder(display->drm, encoder) {
 		switch (encoder->type) {
 		case INTEL_OUTPUT_ANALOG:
 			has_fdi = true;
@@ -462,37 +468,36 @@ static void lpt_init_pch_refclk(struct drm_i915_private *dev_priv)
 	 * clock hierarchy. That would also allow us to do
 	 * clock bending finally.
 	 */
-	dev_priv->display.dpll.pch_ssc_use = 0;
+	display->dpll.pch_ssc_use = 0;
 
-	if (spll_uses_pch_ssc(dev_priv)) {
-		drm_dbg_kms(&dev_priv->drm, "SPLL using PCH SSC\n");
-		dev_priv->display.dpll.pch_ssc_use |= BIT(DPLL_ID_SPLL);
+	if (spll_uses_pch_ssc(display)) {
+		drm_dbg_kms(display->drm, "SPLL using PCH SSC\n");
+		display->dpll.pch_ssc_use |= BIT(DPLL_ID_SPLL);
 	}
 
-	if (wrpll_uses_pch_ssc(dev_priv, DPLL_ID_WRPLL1)) {
-		drm_dbg_kms(&dev_priv->drm, "WRPLL1 using PCH SSC\n");
-		dev_priv->display.dpll.pch_ssc_use |= BIT(DPLL_ID_WRPLL1);
+	if (wrpll_uses_pch_ssc(display, DPLL_ID_WRPLL1)) {
+		drm_dbg_kms(display->drm, "WRPLL1 using PCH SSC\n");
+		display->dpll.pch_ssc_use |= BIT(DPLL_ID_WRPLL1);
 	}
 
-	if (wrpll_uses_pch_ssc(dev_priv, DPLL_ID_WRPLL2)) {
-		drm_dbg_kms(&dev_priv->drm, "WRPLL2 using PCH SSC\n");
-		dev_priv->display.dpll.pch_ssc_use |= BIT(DPLL_ID_WRPLL2);
+	if (wrpll_uses_pch_ssc(display, DPLL_ID_WRPLL2)) {
+		drm_dbg_kms(display->drm, "WRPLL2 using PCH SSC\n");
+		display->dpll.pch_ssc_use |= BIT(DPLL_ID_WRPLL2);
 	}
 
-	if (dev_priv->display.dpll.pch_ssc_use)
+	if (display->dpll.pch_ssc_use)
 		return;
 
 	if (has_fdi) {
-		lpt_bend_clkout_dp(dev_priv, 0);
-		lpt_enable_clkout_dp(dev_priv, true, true);
+		lpt_bend_clkout_dp(display, 0);
+		lpt_enable_clkout_dp(display, true, true);
 	} else {
-		lpt_disable_clkout_dp(dev_priv);
+		lpt_disable_clkout_dp(display);
 	}
 }
 
-static void ilk_init_pch_refclk(struct drm_i915_private *dev_priv)
+static void ilk_init_pch_refclk(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_encoder *encoder;
 	struct intel_shared_dpll *pll;
 	int i;
@@ -521,7 +526,7 @@ static void ilk_init_pch_refclk(struct drm_i915_private *dev_priv)
 		}
 	}
 
-	if (HAS_PCH_IBX(dev_priv)) {
+	if (HAS_PCH_IBX(display)) {
 		has_ck505 = display->vbt.display_clock_mode;
 		can_ssc = has_ck505;
 	} else {
@@ -607,7 +612,7 @@ static void ilk_init_pch_refclk(struct drm_i915_private *dev_priv)
 
 		/* SSC must be turned on before enabling the CPU output  */
 		if (intel_panel_use_ssc(display) && can_ssc) {
-			drm_dbg_kms(&dev_priv->drm, "Using SSC on panel\n");
+			drm_dbg_kms(display->drm, "Using SSC on panel\n");
 			val |= DREF_SSC1_ENABLE;
 		} else {
 			val &= ~DREF_SSC1_ENABLE;
@@ -623,7 +628,7 @@ static void ilk_init_pch_refclk(struct drm_i915_private *dev_priv)
 		/* Enable CPU source on CPU attached eDP */
 		if (has_cpu_edp) {
 			if (intel_panel_use_ssc(display) && can_ssc) {
-				drm_dbg_kms(&dev_priv->drm,
+				drm_dbg_kms(display->drm,
 					    "Using SSC on eDP\n");
 				val |= DREF_CPU_SOURCE_OUTPUT_DOWNSPREAD;
 			} else {
@@ -670,10 +675,10 @@ static void ilk_init_pch_refclk(struct drm_i915_private *dev_priv)
 /*
  * Initialize reference clocks when the driver loads
  */
-void intel_init_pch_refclk(struct drm_i915_private *dev_priv)
+void intel_init_pch_refclk(struct intel_display *display)
 {
-	if (HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv))
-		ilk_init_pch_refclk(dev_priv);
-	else if (HAS_PCH_LPT(dev_priv))
-		lpt_init_pch_refclk(dev_priv);
+	if (HAS_PCH_IBX(display) || HAS_PCH_CPT(display))
+		ilk_init_pch_refclk(display);
+	else if (HAS_PCH_LPT(display))
+		lpt_init_pch_refclk(display);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_pch_refclk.h b/drivers/gpu/drm/i915/display/intel_pch_refclk.h
index ae3403c0ced8..25cc53c568bc 100644
--- a/drivers/gpu/drm/i915/display/intel_pch_refclk.h
+++ b/drivers/gpu/drm/i915/display/intel_pch_refclk.h
@@ -8,25 +8,25 @@
 
 #include <linux/types.h>
 
-struct drm_i915_private;
 struct intel_crtc_state;
+struct intel_display;
 
 #ifdef I915
 void lpt_program_iclkip(const struct intel_crtc_state *crtc_state);
-void lpt_disable_iclkip(struct drm_i915_private *dev_priv);
-int lpt_get_iclkip(struct drm_i915_private *dev_priv);
+void lpt_disable_iclkip(struct intel_display *display);
+int lpt_get_iclkip(struct intel_display *display);
 int lpt_iclkip(const struct intel_crtc_state *crtc_state);
 
-void intel_init_pch_refclk(struct drm_i915_private *dev_priv);
-void lpt_disable_clkout_dp(struct drm_i915_private *dev_priv);
+void intel_init_pch_refclk(struct intel_display *display);
+void lpt_disable_clkout_dp(struct intel_display *display);
 #else
 static inline void lpt_program_iclkip(const struct intel_crtc_state *crtc_state)
 {
 }
-static inline void lpt_disable_iclkip(struct drm_i915_private *dev_priv)
+static inline void lpt_disable_iclkip(struct intel_display *display)
 {
 }
-static inline int lpt_get_iclkip(struct drm_i915_private *dev_priv)
+static inline int lpt_get_iclkip(struct intel_display *display)
 {
 	return 0;
 }
@@ -34,10 +34,10 @@ static inline int lpt_iclkip(const struct intel_crtc_state *crtc_state)
 {
 	return 0;
 }
-static inline void intel_init_pch_refclk(struct drm_i915_private *dev_priv)
+static inline void intel_init_pch_refclk(struct intel_display *display)
 {
 }
-static inline void lpt_disable_clkout_dp(struct drm_i915_private *dev_priv)
+static inline void lpt_disable_clkout_dp(struct intel_display *display)
 {
 }
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_pipe_crc.c b/drivers/gpu/drm/i915/display/intel_pipe_crc.c
index 10e26c3db946..6182f484b5bd 100644
--- a/drivers/gpu/drm/i915/display/intel_pipe_crc.c
+++ b/drivers/gpu/drm/i915/display/intel_pipe_crc.c
@@ -75,7 +75,7 @@ static int i8xx_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
 	return 0;
 }
 
-static void i9xx_pipe_crc_auto_source(struct drm_i915_private *dev_priv,
+static void i9xx_pipe_crc_auto_source(struct intel_display *display,
 				      enum pipe pipe,
 				      enum intel_pipe_crc_source *source)
 {
@@ -85,8 +85,8 @@ static void i9xx_pipe_crc_auto_source(struct drm_i915_private *dev_priv,
 
 	*source = INTEL_PIPE_CRC_SOURCE_PIPE;
 
-	drm_modeset_lock_all(&dev_priv->drm);
-	for_each_intel_encoder(&dev_priv->drm, encoder) {
+	drm_modeset_lock_all(display->drm);
+	for_each_intel_encoder(display->drm, encoder) {
 		if (!encoder->base.crtc)
 			continue;
 
@@ -113,7 +113,7 @@ static void i9xx_pipe_crc_auto_source(struct drm_i915_private *dev_priv,
 				*source = INTEL_PIPE_CRC_SOURCE_DP_D;
 				break;
 			default:
-				drm_WARN(&dev_priv->drm, 1, "nonexisting DP port %c\n",
+				drm_WARN(display->drm, 1, "nonexisting DP port %c\n",
 					 port_name(dig_port->base.port));
 				break;
 			}
@@ -122,10 +122,10 @@ static void i9xx_pipe_crc_auto_source(struct drm_i915_private *dev_priv,
 			break;
 		}
 	}
-	drm_modeset_unlock_all(&dev_priv->drm);
+	drm_modeset_unlock_all(display->drm);
 }
 
-static int vlv_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
+static int vlv_pipe_crc_ctl_reg(struct intel_display *display,
 				enum pipe pipe,
 				enum intel_pipe_crc_source *source,
 				u32 *val)
@@ -133,7 +133,7 @@ static int vlv_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	bool need_stable_symbols = false;
 
 	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
-		i9xx_pipe_crc_auto_source(dev_priv, pipe, source);
+		i9xx_pipe_crc_auto_source(display, pipe, source);
 
 	switch (*source) {
 	case INTEL_PIPE_CRC_SOURCE_PIPE:
@@ -148,7 +148,7 @@ static int vlv_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 		need_stable_symbols = true;
 		break;
 	case INTEL_PIPE_CRC_SOURCE_DP_D:
-		if (!IS_CHERRYVIEW(dev_priv))
+		if (!display->platform.cherryview)
 			return -EINVAL;
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_DP_D_VLV;
 		need_stable_symbols = true;
@@ -170,7 +170,7 @@ static int vlv_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	 *   - DisplayPort scrambling: used for EMI reduction
 	 */
 	if (need_stable_symbols) {
-		u32 tmp = intel_de_read(dev_priv, PORT_DFT2_G4X(dev_priv));
+		u32 tmp = intel_de_read(display, PORT_DFT2_G4X(display));
 
 		tmp |= DC_BALANCE_RESET_VLV;
 		switch (pipe) {
@@ -186,26 +186,26 @@ static int vlv_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 		default:
 			return -EINVAL;
 		}
-		intel_de_write(dev_priv, PORT_DFT2_G4X(dev_priv), tmp);
+		intel_de_write(display, PORT_DFT2_G4X(display), tmp);
 	}
 
 	return 0;
 }
 
-static int i9xx_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
+static int i9xx_pipe_crc_ctl_reg(struct intel_display *display,
 				 enum pipe pipe,
 				 enum intel_pipe_crc_source *source,
 				 u32 *val)
 {
 	if (*source == INTEL_PIPE_CRC_SOURCE_AUTO)
-		i9xx_pipe_crc_auto_source(dev_priv, pipe, source);
+		i9xx_pipe_crc_auto_source(display, pipe, source);
 
 	switch (*source) {
 	case INTEL_PIPE_CRC_SOURCE_PIPE:
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_PIPE_I9XX;
 		break;
 	case INTEL_PIPE_CRC_SOURCE_TV:
-		if (!SUPPORTS_TV(dev_priv))
+		if (!SUPPORTS_TV(display))
 			return -EINVAL;
 		*val = PIPE_CRC_ENABLE | PIPE_CRC_SOURCE_TV_PRE;
 		break;
@@ -229,10 +229,10 @@ static int i9xx_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static void vlv_undo_pipe_scramble_reset(struct drm_i915_private *dev_priv,
+static void vlv_undo_pipe_scramble_reset(struct intel_display *display,
 					 enum pipe pipe)
 {
-	u32 tmp = intel_de_read(dev_priv, PORT_DFT2_G4X(dev_priv));
+	u32 tmp = intel_de_read(display, PORT_DFT2_G4X(display));
 
 	switch (pipe) {
 	case PIPE_A:
@@ -249,7 +249,7 @@ static void vlv_undo_pipe_scramble_reset(struct drm_i915_private *dev_priv,
 	}
 	if (!(tmp & PIPE_SCRAMBLE_RESET_MASK))
 		tmp &= ~DC_BALANCE_RESET_VLV;
-	intel_de_write(dev_priv, PORT_DFT2_G4X(dev_priv), tmp);
+	intel_de_write(display, PORT_DFT2_G4X(display), tmp);
 }
 
 static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
@@ -281,18 +281,18 @@ static int ilk_pipe_crc_ctl_reg(enum intel_pipe_crc_source *source,
 static void
 intel_crtc_crc_setup_workarounds(struct intel_crtc *crtc, bool enable)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_crtc_state *pipe_config;
 	struct drm_atomic_state *state;
 	struct drm_modeset_acquire_ctx ctx;
 	int ret;
 
-	if (IS_I945GM(dev_priv) || IS_I915GM(dev_priv))
-		i915gm_irq_cstate_wa(dev_priv, enable);
+	if (display->platform.i945gm || display->platform.i915gm)
+		i915gm_irq_cstate_wa(display, enable);
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	state = drm_atomic_state_alloc(&dev_priv->drm);
+	state = drm_atomic_state_alloc(display->drm);
 	if (!state) {
 		ret = -ENOMEM;
 		goto unlock;
@@ -311,7 +311,7 @@ retry:
 	pipe_config->uapi.mode_changed = pipe_config->has_psr;
 	pipe_config->crc_enabled = enable;
 
-	if (IS_HASWELL(dev_priv) &&
+	if (display->platform.haswell &&
 	    pipe_config->hw.active && crtc->pipe == PIPE_A &&
 	    pipe_config->cpu_transcoder == TRANSCODER_EDP)
 		pipe_config->uapi.mode_changed = true;
@@ -327,13 +327,13 @@ put_state:
 
 	drm_atomic_state_put(state);
 unlock:
-	drm_WARN(&dev_priv->drm, ret,
+	drm_WARN(display->drm, ret,
 		 "Toggling workaround to %i returns %i\n", enable, ret);
 	drm_modeset_drop_locks(&ctx);
 	drm_modeset_acquire_fini(&ctx);
 }
 
-static int ivb_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
+static int ivb_pipe_crc_ctl_reg(struct intel_display *display,
 				enum pipe pipe,
 				enum intel_pipe_crc_source *source,
 				u32 *val)
@@ -361,7 +361,7 @@ static int ivb_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static int skl_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
+static int skl_pipe_crc_ctl_reg(struct intel_display *display,
 				enum pipe pipe,
 				enum intel_pipe_crc_source *source,
 				u32 *val)
@@ -404,22 +404,22 @@ static int skl_pipe_crc_ctl_reg(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
-static int get_new_crc_ctl_reg(struct drm_i915_private *dev_priv,
+static int get_new_crc_ctl_reg(struct intel_display *display,
 			       enum pipe pipe,
 			       enum intel_pipe_crc_source *source, u32 *val)
 {
-	if (DISPLAY_VER(dev_priv) == 2)
+	if (DISPLAY_VER(display) == 2)
 		return i8xx_pipe_crc_ctl_reg(source, val);
-	else if (DISPLAY_VER(dev_priv) < 5)
-		return i9xx_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		return vlv_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
-	else if (IS_IRONLAKE(dev_priv) || IS_SANDYBRIDGE(dev_priv))
+	else if (DISPLAY_VER(display) < 5)
+		return i9xx_pipe_crc_ctl_reg(display, pipe, source, val);
+	else if (display->platform.valleyview || display->platform.cherryview)
+		return vlv_pipe_crc_ctl_reg(display, pipe, source, val);
+	else if (display->platform.ironlake || display->platform.sandybridge)
 		return ilk_pipe_crc_ctl_reg(source, val);
-	else if (DISPLAY_VER(dev_priv) < 9)
-		return ivb_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
+	else if (DISPLAY_VER(display) < 9)
+		return ivb_pipe_crc_ctl_reg(display, pipe, source, val);
 	else
-		return skl_pipe_crc_ctl_reg(dev_priv, pipe, source, val);
+		return skl_pipe_crc_ctl_reg(display, pipe, source, val);
 }
 
 static int
@@ -447,7 +447,7 @@ void intel_crtc_crc_init(struct intel_crtc *crtc)
 	spin_lock_init(&pipe_crc->lock);
 }
 
-static int i8xx_crc_source_valid(struct drm_i915_private *dev_priv,
+static int i8xx_crc_source_valid(struct intel_display *display,
 				 const enum intel_pipe_crc_source source)
 {
 	switch (source) {
@@ -459,7 +459,7 @@ static int i8xx_crc_source_valid(struct drm_i915_private *dev_priv,
 	}
 }
 
-static int i9xx_crc_source_valid(struct drm_i915_private *dev_priv,
+static int i9xx_crc_source_valid(struct intel_display *display,
 				 const enum intel_pipe_crc_source source)
 {
 	switch (source) {
@@ -472,7 +472,7 @@ static int i9xx_crc_source_valid(struct drm_i915_private *dev_priv,
 	}
 }
 
-static int vlv_crc_source_valid(struct drm_i915_private *dev_priv,
+static int vlv_crc_source_valid(struct intel_display *display,
 				const enum intel_pipe_crc_source source)
 {
 	switch (source) {
@@ -487,7 +487,7 @@ static int vlv_crc_source_valid(struct drm_i915_private *dev_priv,
 	}
 }
 
-static int ilk_crc_source_valid(struct drm_i915_private *dev_priv,
+static int ilk_crc_source_valid(struct intel_display *display,
 				const enum intel_pipe_crc_source source)
 {
 	switch (source) {
@@ -501,7 +501,7 @@ static int ilk_crc_source_valid(struct drm_i915_private *dev_priv,
 	}
 }
 
-static int ivb_crc_source_valid(struct drm_i915_private *dev_priv,
+static int ivb_crc_source_valid(struct intel_display *display,
 				const enum intel_pipe_crc_source source)
 {
 	switch (source) {
@@ -515,7 +515,7 @@ static int ivb_crc_source_valid(struct drm_i915_private *dev_priv,
 	}
 }
 
-static int skl_crc_source_valid(struct drm_i915_private *dev_priv,
+static int skl_crc_source_valid(struct intel_display *display,
 				const enum intel_pipe_crc_source source)
 {
 	switch (source) {
@@ -535,21 +535,21 @@ static int skl_crc_source_valid(struct drm_i915_private *dev_priv,
 }
 
 static int
-intel_is_valid_crc_source(struct drm_i915_private *dev_priv,
+intel_is_valid_crc_source(struct intel_display *display,
 			  const enum intel_pipe_crc_source source)
 {
-	if (DISPLAY_VER(dev_priv) == 2)
-		return i8xx_crc_source_valid(dev_priv, source);
-	else if (DISPLAY_VER(dev_priv) < 5)
-		return i9xx_crc_source_valid(dev_priv, source);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		return vlv_crc_source_valid(dev_priv, source);
-	else if (IS_IRONLAKE(dev_priv) || IS_SANDYBRIDGE(dev_priv))
-		return ilk_crc_source_valid(dev_priv, source);
-	else if (DISPLAY_VER(dev_priv) < 9)
-		return ivb_crc_source_valid(dev_priv, source);
+	if (DISPLAY_VER(display) == 2)
+		return i8xx_crc_source_valid(display, source);
+	else if (DISPLAY_VER(display) < 5)
+		return i9xx_crc_source_valid(display, source);
+	else if (display->platform.valleyview || display->platform.cherryview)
+		return vlv_crc_source_valid(display, source);
+	else if (display->platform.ironlake || display->platform.sandybridge)
+		return ilk_crc_source_valid(display, source);
+	else if (DISPLAY_VER(display) < 9)
+		return ivb_crc_source_valid(display, source);
 	else
-		return skl_crc_source_valid(dev_priv, source);
+		return skl_crc_source_valid(display, source);
 }
 
 const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc,
@@ -562,16 +562,16 @@ const char *const *intel_crtc_get_crc_sources(struct drm_crtc *crtc,
 int intel_crtc_verify_crc_source(struct drm_crtc *crtc, const char *source_name,
 				 size_t *values_cnt)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_display *display = to_intel_display(crtc->dev);
 	enum intel_pipe_crc_source source;
 
 	if (display_crc_ctl_parse_source(source_name, &source) < 0) {
-		drm_dbg(&dev_priv->drm, "unknown source %s\n", source_name);
+		drm_dbg_kms(display->drm, "unknown source %s\n", source_name);
 		return -EINVAL;
 	}
 
 	if (source == INTEL_PIPE_CRC_SOURCE_AUTO ||
-	    intel_is_valid_crc_source(dev_priv, source) == 0) {
+	    intel_is_valid_crc_source(display, source) == 0) {
 		*values_cnt = 5;
 		return 0;
 	}
@@ -583,7 +583,6 @@ int intel_crtc_set_crc_source(struct drm_crtc *_crtc, const char *source_name)
 {
 	struct intel_crtc *crtc = to_intel_crtc(_crtc);
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc;
 	enum intel_display_power_domain power_domain;
 	enum intel_pipe_crc_source source;
@@ -594,14 +593,14 @@ int intel_crtc_set_crc_source(struct drm_crtc *_crtc, const char *source_name)
 	bool enable;
 
 	if (display_crc_ctl_parse_source(source_name, &source) < 0) {
-		drm_dbg(&dev_priv->drm, "unknown source %s\n", source_name);
+		drm_dbg_kms(display->drm, "unknown source %s\n", source_name);
 		return -EINVAL;
 	}
 
 	power_domain = POWER_DOMAIN_PIPE(pipe);
 	wakeref = intel_display_power_get_if_enabled(display, power_domain);
 	if (!wakeref) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "Trying to capture CRC while pipe is off\n");
 		return -EIO;
 	}
@@ -610,17 +609,17 @@ int intel_crtc_set_crc_source(struct drm_crtc *_crtc, const char *source_name)
 	if (enable)
 		intel_crtc_crc_setup_workarounds(crtc, true);
 
-	ret = get_new_crc_ctl_reg(dev_priv, pipe, &source, &val);
+	ret = get_new_crc_ctl_reg(display, pipe, &source, &val);
 	if (ret != 0)
 		goto out;
 
 	pipe_crc->source = source;
-	intel_de_write(dev_priv, PIPE_CRC_CTL(dev_priv, pipe), val);
-	intel_de_posting_read(dev_priv, PIPE_CRC_CTL(dev_priv, pipe));
+	intel_de_write(display, PIPE_CRC_CTL(display, pipe), val);
+	intel_de_posting_read(display, PIPE_CRC_CTL(display, pipe));
 
 	if (!source) {
-		if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-			vlv_undo_pipe_scramble_reset(dev_priv, pipe);
+		if (display->platform.valleyview || display->platform.cherryview)
+			vlv_undo_pipe_scramble_reset(display, pipe);
 	}
 
 	pipe_crc->skipped = 0;
@@ -636,7 +635,7 @@ out:
 
 void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc;
 	enum pipe pipe = crtc->pipe;
 	u32 val = 0;
@@ -644,19 +643,20 @@ void intel_crtc_enable_pipe_crc(struct intel_crtc *crtc)
 	if (!crtc->base.crc.opened)
 		return;
 
-	if (get_new_crc_ctl_reg(dev_priv, pipe, &pipe_crc->source, &val) < 0)
+	if (get_new_crc_ctl_reg(display, pipe, &pipe_crc->source, &val) < 0)
 		return;
 
 	/* Don't need pipe_crc->lock here, IRQs are not generated. */
 	pipe_crc->skipped = 0;
 
-	intel_de_write(dev_priv, PIPE_CRC_CTL(dev_priv, pipe), val);
-	intel_de_posting_read(dev_priv, PIPE_CRC_CTL(dev_priv, pipe));
+	intel_de_write(display, PIPE_CRC_CTL(display, pipe), val);
+	intel_de_posting_read(display, PIPE_CRC_CTL(display, pipe));
 }
 
 void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc;
 	enum pipe pipe = crtc->pipe;
 
@@ -665,7 +665,7 @@ void intel_crtc_disable_pipe_crc(struct intel_crtc *crtc)
 	pipe_crc->skipped = INT_MIN;
 	spin_unlock_irq(&pipe_crc->lock);
 
-	intel_de_write(dev_priv, PIPE_CRC_CTL(dev_priv, pipe), 0);
-	intel_de_posting_read(dev_priv, PIPE_CRC_CTL(dev_priv, pipe));
+	intel_de_write(display, PIPE_CRC_CTL(display, pipe), 0);
+	intel_de_posting_read(display, PIPE_CRC_CTL(display, pipe));
 	intel_synchronize_irq(dev_priv);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c
index b1675b46e06c..c00d9184c586 100644
--- a/drivers/gpu/drm/i915/display/intel_plane_initial.c
+++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c
@@ -52,44 +52,57 @@ intel_reuse_initial_plane_obj(struct intel_crtc *this,
 	return false;
 }
 
+static enum intel_memory_type
+initial_plane_memory_type(struct intel_display *display)
+{
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
+	if (display->platform.dgfx)
+		return INTEL_MEMORY_LOCAL;
+	else if (HAS_LMEMBAR_SMEM_STOLEN(i915))
+		return INTEL_MEMORY_STOLEN_LOCAL;
+	else
+		return INTEL_MEMORY_STOLEN_SYSTEM;
+}
+
 static bool
-initial_plane_phys_lmem(struct intel_display *display,
-			struct intel_initial_plane_config *plane_config)
+initial_plane_phys(struct intel_display *display,
+		   struct intel_initial_plane_config *plane_config)
 {
 	struct drm_i915_private *i915 = to_i915(display->drm);
-	gen8_pte_t __iomem *gte = to_gt(i915)->ggtt->gsm;
+	struct i915_ggtt *ggtt = to_gt(i915)->ggtt;
 	struct intel_memory_region *mem;
+	enum intel_memory_type mem_type;
+	bool is_present, is_local;
 	dma_addr_t dma_addr;
-	gen8_pte_t pte;
 	u32 base;
 
+	mem_type = initial_plane_memory_type(display);
+	mem = intel_memory_region_by_type(i915, mem_type);
+	if (!mem) {
+		drm_dbg_kms(display->drm,
+			    "Initial plane memory region (type %s) not initialized\n",
+			    intel_memory_type_str(mem_type));
+		return false;
+	}
+
 	base = round_down(plane_config->base, I915_GTT_MIN_ALIGNMENT);
 
-	gte += base / I915_GTT_PAGE_SIZE;
+	dma_addr = intel_ggtt_read_entry(&ggtt->vm, base, &is_present, &is_local);
 
-	pte = ioread64(gte);
-	if (!(pte & GEN12_GGTT_PTE_LM)) {
+	if (!is_present) {
 		drm_err(display->drm,
-			"Initial plane programming missing PTE_LM bit\n");
+			"Initial plane FB PTE not present\n");
 		return false;
 	}
 
-	dma_addr = pte & GEN12_GGTT_PTE_ADDR_MASK;
-
-	if (IS_DGFX(i915))
-		mem = i915->mm.regions[INTEL_REGION_LMEM_0];
-	else
-		mem = i915->mm.stolen_region;
-	if (!mem) {
-		drm_dbg_kms(display->drm,
-			    "Initial plane memory region not initialized\n");
+	if (intel_memory_type_is_local(mem->type) != is_local) {
+		drm_err(display->drm,
+			"Initial plane FB PTE unsuitable for %s\n",
+			mem->region.name);
 		return false;
 	}
 
-	/*
-	 * On lmem we don't currently expect this to
-	 * ever be placed in the stolen portion.
-	 */
 	if (dma_addr < mem->region.start || dma_addr > mem->region.end) {
 		drm_err(display->drm,
 			"Initial plane programming using invalid range, dma_addr=%pa (%s [%pa-%pa])\n",
@@ -107,42 +120,6 @@ initial_plane_phys_lmem(struct intel_display *display,
 	return true;
 }
 
-static bool
-initial_plane_phys_smem(struct intel_display *display,
-			struct intel_initial_plane_config *plane_config)
-{
-	struct drm_i915_private *i915 = to_i915(display->drm);
-	struct intel_memory_region *mem;
-	u32 base;
-
-	base = round_down(plane_config->base, I915_GTT_MIN_ALIGNMENT);
-
-	mem = i915->mm.stolen_region;
-	if (!mem) {
-		drm_dbg_kms(display->drm,
-			    "Initial plane memory region not initialized\n");
-		return false;
-	}
-
-	/* FIXME get and validate the dma_addr from the PTE */
-	plane_config->phys_base = base;
-	plane_config->mem = mem;
-
-	return true;
-}
-
-static bool
-initial_plane_phys(struct intel_display *display,
-		   struct intel_initial_plane_config *plane_config)
-{
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	if (IS_DGFX(i915) || HAS_LMEMBAR_SMEM_STOLEN(i915))
-		return initial_plane_phys_lmem(display, plane_config);
-	else
-		return initial_plane_phys_smem(display, plane_config);
-}
-
 static struct i915_vma *
 initial_plane_vma(struct intel_display *display,
 		  struct intel_initial_plane_config *plane_config)
diff --git a/drivers/gpu/drm/i915/display/intel_pmdemand.c b/drivers/gpu/drm/i915/display/intel_pmdemand.c
index 63301a01906c..1253376c7654 100644
--- a/drivers/gpu/drm/i915/display/intel_pmdemand.c
+++ b/drivers/gpu/drm/i915/display/intel_pmdemand.c
@@ -5,6 +5,8 @@
 
 #include <linux/bitops.h>
 
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
 #include "i915_utils.h"
 #include "intel_atomic.h"
@@ -476,13 +478,34 @@ static bool intel_pmdemand_req_complete(struct intel_display *display)
 		 XELPDP_PMDEMAND_REQ_ENABLE);
 }
 
-static void intel_pmdemand_wait(struct intel_display *display)
+static void intel_pmdemand_poll(struct intel_display *display)
 {
-	if (!wait_event_timeout(display->pmdemand.waitqueue,
-				intel_pmdemand_req_complete(display),
-				msecs_to_jiffies_timeout(10)))
+	const unsigned int timeout_ms = 10;
+	u32 status;
+	int ret;
+
+	ret = intel_de_wait_custom(display, XELPDP_INITIATE_PMDEMAND_REQUEST(1),
+				   XELPDP_PMDEMAND_REQ_ENABLE, 0,
+				   50, timeout_ms, &status);
+
+	if (ret == -ETIMEDOUT)
 		drm_err(display->drm,
-			"timed out waiting for Punit PM Demand Response\n");
+			"timed out waiting for Punit PM Demand Response within %ums (status 0x%08x)\n",
+			timeout_ms, status);
+}
+
+static void intel_pmdemand_wait(struct intel_display *display)
+{
+	/* Wa_14024400148 For lnl use polling method */
+	if (DISPLAY_VER(display) == 20) {
+		intel_pmdemand_poll(display);
+	} else {
+		if (!wait_event_timeout(display->pmdemand.waitqueue,
+					intel_pmdemand_req_complete(display),
+					msecs_to_jiffies_timeout(10)))
+			drm_err(display->drm,
+				"timed out waiting for Punit PM Demand Response\n");
+	}
 }
 
 /* Required to be programmed during Display Init Sequences. */
diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
index 617ce4993172..05e1e5c7e8b7 100644
--- a/drivers/gpu/drm/i915/display/intel_pps.c
+++ b/drivers/gpu/drm/i915/display/intel_pps.c
@@ -91,7 +91,6 @@ static void
 vlv_power_sequencer_kick(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	enum pipe pipe = intel_dp->pps.vlv_pps_pipe;
 	bool pll_enabled, release_cl_override = false;
@@ -134,7 +133,7 @@ vlv_power_sequencer_kick(struct intel_dp *intel_dp)
 		release_cl_override = display->platform.cherryview &&
 			!chv_phy_powergate_ch(display, phy, ch, true);
 
-		if (vlv_force_pll_on(dev_priv, pipe, vlv_get_dpll(display))) {
+		if (vlv_force_pll_on(display, pipe, vlv_get_dpll(display))) {
 			drm_err(display->drm,
 				"Failed to force on PLL for pipe %c!\n",
 				pipe_name(pipe));
@@ -158,7 +157,7 @@ vlv_power_sequencer_kick(struct intel_dp *intel_dp)
 	intel_de_posting_read(display, intel_dp->output_reg);
 
 	if (!pll_enabled) {
-		vlv_force_pll_off(dev_priv, pipe);
+		vlv_force_pll_off(display, pipe);
 
 		if (release_cl_override)
 			chv_phy_powergate_ch(display, phy, ch, false);
@@ -351,21 +350,19 @@ vlv_initial_power_sequencer_setup(struct intel_dp *intel_dp)
 
 static int intel_num_pps(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
 	if (display->platform.valleyview || display->platform.cherryview)
 		return 2;
 
 	if (display->platform.geminilake || display->platform.broxton)
 		return 2;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_MTL)
+	if (INTEL_PCH_TYPE(display) >= PCH_MTL)
 		return 2;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_DG1)
+	if (INTEL_PCH_TYPE(display) >= PCH_DG1)
 		return 1;
 
-	if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
+	if (INTEL_PCH_TYPE(display) >= PCH_ICP)
 		return 2;
 
 	return 1;
@@ -374,11 +371,10 @@ static int intel_num_pps(struct intel_display *display)
 static bool intel_pps_is_valid(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	if (intel_dp->pps.pps_idx == 1 &&
-	    INTEL_PCH_TYPE(i915) >= PCH_ICP &&
-	    INTEL_PCH_TYPE(i915) <= PCH_ADP)
+	    INTEL_PCH_TYPE(display) >= PCH_ICP &&
+	    INTEL_PCH_TYPE(display) <= PCH_ADP)
 		return intel_de_read(display, SOUTH_CHICKEN1) & ICP_SECOND_PPS_IO_SELECT;
 
 	return true;
@@ -500,7 +496,6 @@ static void intel_pps_get_registers(struct intel_dp *intel_dp,
 				    struct pps_registers *regs)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	int pps_idx;
 
 	memset(regs, 0, sizeof(*regs));
@@ -519,7 +514,7 @@ static void intel_pps_get_registers(struct intel_dp *intel_dp,
 
 	/* Cycle delay moved from PP_DIVISOR to PP_CONTROL */
 	if (display->platform.geminilake || display->platform.broxton ||
-	    INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
+	    INTEL_PCH_TYPE(display) >= PCH_CNP)
 		regs->pp_div = INVALID_MMIO_REG;
 	else
 		regs->pp_div = PP_DIVISOR(display, pps_idx);
@@ -744,11 +739,11 @@ bool intel_pps_vdd_on_unlocked(struct intel_dp *intel_dp)
 	i915_reg_t pp_stat_reg, pp_ctrl_reg;
 	bool need_to_disable = !intel_dp->pps.want_panel_vdd;
 
-	lockdep_assert_held(&display->pps.mutex);
-
 	if (!intel_dp_is_edp(intel_dp))
 		return false;
 
+	lockdep_assert_held(&display->pps.mutex);
+
 	cancel_delayed_work(&intel_dp->pps.panel_vdd_work);
 	intel_dp->pps.want_panel_vdd = true;
 
@@ -925,11 +920,11 @@ void intel_pps_vdd_off_unlocked(struct intel_dp *intel_dp, bool sync)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
 
-	lockdep_assert_held(&display->pps.mutex);
-
 	if (!intel_dp_is_edp(intel_dp))
 		return;
 
+	lockdep_assert_held(&display->pps.mutex);
+
 	INTEL_DISPLAY_STATE_WARN(display, !intel_dp->pps.want_panel_vdd,
 				 "[ENCODER:%d:%s] %s VDD not forced on",
 				 dp_to_dig_port(intel_dp)->base.base.base.id,
@@ -1592,7 +1587,6 @@ static void pps_init_delays(struct intel_dp *intel_dp)
 static void pps_init_registers(struct intel_dp *intel_dp, bool force_disable_vdd)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 pp_on, pp_off, port_sel = 0;
 	int div = DISPLAY_RUNTIME_INFO(display)->rawclk_freq / 1000;
 	struct pps_registers regs;
@@ -1639,7 +1633,7 @@ static void pps_init_registers(struct intel_dp *intel_dp, bool force_disable_vdd
 	 * power sequencer any more. */
 	if (display->platform.valleyview || display->platform.cherryview) {
 		port_sel = PANEL_PORT_SELECT_VLV(port);
-	} else if (HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv)) {
+	} else if (HAS_PCH_IBX(display) || HAS_PCH_CPT(display)) {
 		switch (port) {
 		case PORT_A:
 			port_sel = PANEL_PORT_SELECT_DPA;
@@ -1792,9 +1786,7 @@ void intel_pps_unlock_regs_wa(struct intel_display *display)
 
 void intel_pps_setup(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	if (HAS_PCH_SPLIT(i915) || display->platform.geminilake || display->platform.broxton)
+	if (HAS_PCH_SPLIT(display) || display->platform.geminilake || display->platform.broxton)
 		display->pps.mmio_base = PCH_PPS_BASE;
 	else if (display->platform.valleyview || display->platform.cherryview)
 		display->pps.mmio_base = VLV_PPS_BASE;
@@ -1837,7 +1829,6 @@ void intel_pps_connector_debugfs_add(struct intel_connector *connector)
 
 void assert_pps_unlocked(struct intel_display *display, enum pipe pipe)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	i915_reg_t pp_reg;
 	u32 val;
 	enum pipe panel_pipe = INVALID_PIPE;
@@ -1846,7 +1837,7 @@ void assert_pps_unlocked(struct intel_display *display, enum pipe pipe)
 	if (drm_WARN_ON(display->drm, HAS_DDI(display)))
 		return;
 
-	if (HAS_PCH_SPLIT(dev_priv)) {
+	if (HAS_PCH_SPLIT(display)) {
 		u32 port_sel;
 
 		pp_reg = PP_CONTROL(display, 0);
@@ -1855,7 +1846,7 @@ void assert_pps_unlocked(struct intel_display *display, enum pipe pipe)
 
 		switch (port_sel) {
 		case PANEL_PORT_SELECT_LVDS:
-			intel_lvds_port_enabled(dev_priv, PCH_LVDS, &panel_pipe);
+			intel_lvds_port_enabled(display, PCH_LVDS, &panel_pipe);
 			break;
 		case PANEL_PORT_SELECT_DPA:
 			g4x_dp_port_enabled(display, DP_A, PORT_A, &panel_pipe);
@@ -1883,7 +1874,7 @@ void assert_pps_unlocked(struct intel_display *display, enum pipe pipe)
 
 		drm_WARN_ON(display->drm,
 			    port_sel != PANEL_PORT_SELECT_LVDS);
-		intel_lvds_port_enabled(dev_priv, LVDS, &panel_pipe);
+		intel_lvds_port_enabled(display, LVDS, &panel_pipe);
 	}
 
 	val = intel_de_read(display, pp_reg);
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 4e938bad808c..430ad4ef7146 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -26,6 +26,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_debugfs.h>
+#include <drm/drm_vblank.h>
 
 #include "i915_drv.h"
 #include "i915_reg.h"
@@ -36,7 +37,9 @@
 #include "intel_ddi.h"
 #include "intel_de.h"
 #include "intel_display_irq.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
+#include "intel_dmc.h"
 #include "intel_dp.h"
 #include "intel_dp_aux.h"
 #include "intel_frontbuffer.h"
@@ -45,6 +48,7 @@
 #include "intel_psr_regs.h"
 #include "intel_snps_phy.h"
 #include "intel_vblank.h"
+#include "intel_vrr.h"
 #include "skl_universal_plane.h"
 
 /**
@@ -463,8 +467,8 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir)
 		if (DISPLAY_VER(display) >= 9) {
 			u32 val;
 
-			val = intel_de_rmw(dev_priv,
-					   PSR_EVENT(dev_priv, cpu_transcoder),
+			val = intel_de_rmw(display,
+					   PSR_EVENT(display, cpu_transcoder),
 					   0, 0);
 
 			psr_event_print(display, val, intel_dp->psr.sel_update_enabled);
@@ -689,7 +693,6 @@ void intel_psr_init_dpcd(struct intel_dp *intel_dp)
 static void hsw_psr_setup_aux(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 aux_clock_divider, aux_ctl;
 	/* write DP_SET_POWER=D0 */
@@ -704,7 +707,7 @@ static void hsw_psr_setup_aux(struct intel_dp *intel_dp)
 
 	BUILD_BUG_ON(sizeof(aux_msg) > 20);
 	for (i = 0; i < sizeof(aux_msg); i += 4)
-		intel_de_write(dev_priv,
+		intel_de_write(display,
 			       psr_aux_data_reg(display, cpu_transcoder, i >> 2),
 			       intel_dp_aux_pack(&aux_msg[i], sizeof(aux_msg) - i));
 
@@ -794,31 +797,10 @@ static void _psr_enable_sink(struct intel_dp *intel_dp,
 	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, val);
 }
 
-static void intel_psr_enable_sink_alpm(struct intel_dp *intel_dp,
-				       const struct intel_crtc_state *crtc_state)
-{
-	u8 val;
-
-	/*
-	 * eDP Panel Replay uses always ALPM
-	 * PSR2 uses ALPM but PSR1 doesn't
-	 */
-	if (!intel_dp_is_edp(intel_dp) || (!crtc_state->has_panel_replay &&
-					   !crtc_state->has_sel_update))
-		return;
-
-	val = DP_ALPM_ENABLE | DP_ALPM_LOCK_ERROR_IRQ_HPD_ENABLE;
-
-	if (crtc_state->has_panel_replay)
-		val |= DP_ALPM_MODE_AUX_LESS;
-
-	drm_dp_dpcd_writeb(&intel_dp->aux, DP_RECEIVER_ALPM_CONFIG, val);
-}
-
 static void intel_psr_enable_sink(struct intel_dp *intel_dp,
 				  const struct intel_crtc_state *crtc_state)
 {
-	intel_psr_enable_sink_alpm(intel_dp, crtc_state);
+	intel_alpm_enable_sink(intel_dp, crtc_state);
 
 	crtc_state->has_panel_replay ?
 		_panel_replay_enable_sink(intel_dp, crtc_state) :
@@ -839,7 +821,6 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_connector *connector = intel_dp->attached_connector;
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 val = 0;
 
 	if (DISPLAY_VER(display) >= 11)
@@ -873,7 +854,7 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp)
 	 * WA 0479: hsw,bdw
 	 * "Do not skip both TP1 and TP2/TP3"
 	 */
-	if (DISPLAY_VER(dev_priv) < 9 &&
+	if (DISPLAY_VER(display) < 9 &&
 	    connector->panel.vbt.psr.tp1_wakeup_time_us == 0 &&
 	    connector->panel.vbt.psr.tp2_tp3_wakeup_time_us == 0)
 		val |= EDP_PSR_TP2_TP3_TIME_100us;
@@ -906,10 +887,21 @@ static u8 psr_compute_idle_frames(struct intel_dp *intel_dp)
 	return idle_frames;
 }
 
+static bool is_dc5_dc6_blocked(struct intel_dp *intel_dp)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+	u32 current_dc_state = intel_display_power_get_current_dc_state(display);
+	struct drm_vblank_crtc *vblank = &display->drm->vblank[intel_dp->psr.pipe];
+
+	return (current_dc_state != DC_STATE_EN_UPTO_DC5 &&
+		current_dc_state != DC_STATE_EN_UPTO_DC6) ||
+		intel_dp->psr.active_non_psr_pipes ||
+		READ_ONCE(vblank->enabled);
+}
+
 static void hsw_activate_psr1(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 max_sleep_time = 0x1f;
 	u32 val = EDP_PSR_ENABLE;
@@ -919,7 +911,7 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 	if (DISPLAY_VER(display) < 20)
 		val |= EDP_PSR_MAX_SLEEP_TIME(max_sleep_time);
 
-	if (IS_HASWELL(dev_priv))
+	if (display->platform.haswell)
 		val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
 
 	if (intel_dp->psr.link_standby)
@@ -935,6 +927,14 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 
 	intel_de_rmw(display, psr_ctl_reg(display, cpu_transcoder),
 		     ~EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK, val);
+
+	/* Wa_16025596647 */
+	if ((DISPLAY_VER(display) == 20 ||
+	     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
+	    is_dc5_dc6_blocked(intel_dp))
+		intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display,
+								       intel_dp->psr.pipe,
+								       true);
 }
 
 static u32 intel_psr2_get_tp_time(struct intel_dp *intel_dp)
@@ -1013,14 +1013,21 @@ static void dg2_activate_panel_replay(struct intel_dp *intel_dp)
 static void hsw_activate_psr2(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 val = EDP_PSR2_ENABLE;
 	u32 psr_val = 0;
+	u8 idle_frames;
 
-	val |= EDP_PSR2_IDLE_FRAMES(psr_compute_idle_frames(intel_dp));
+	/* Wa_16025596647 */
+	if ((DISPLAY_VER(display) == 20 ||
+	     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
+	    is_dc5_dc6_blocked(intel_dp))
+		idle_frames = 0;
+	else
+		idle_frames = psr_compute_idle_frames(intel_dp);
+	val |= EDP_PSR2_IDLE_FRAMES(idle_frames);
 
-	if (DISPLAY_VER(display) < 14 && !IS_ALDERLAKE_P(dev_priv))
+	if (DISPLAY_VER(display) < 14 && !display->platform.alderlake_p)
 		val |= EDP_SU_TRACK_ENABLE;
 
 	if (DISPLAY_VER(display) >= 10 && DISPLAY_VER(display) < 13)
@@ -1038,7 +1045,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 	}
 
 	/* Wa_22012278275:adl-p */
-	if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(display, STEP_A0, STEP_E0)) {
+	if (display->platform.alderlake_p && IS_DISPLAY_STEP(display, STEP_A0, STEP_E0)) {
 		static const u8 map[] = {
 			2, /* 5 lines */
 			1, /* 6 lines */
@@ -1103,9 +1110,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 static bool
 transcoder_has_psr2(struct intel_display *display, enum transcoder cpu_transcoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14)
+	if (display->platform.alderlake_p || DISPLAY_VER(display) >= 14)
 		return cpu_transcoder == TRANSCODER_A || cpu_transcoder == TRANSCODER_B;
 	else if (DISPLAY_VER(display) >= 12)
 		return cpu_transcoder == TRANSCODER_A;
@@ -1183,10 +1188,9 @@ dc3co_is_pipe_port_compatible(struct intel_dp *intel_dp,
 	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	enum pipe pipe = to_intel_crtc(crtc_state->uapi.crtc)->pipe;
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum port port = dig_port->base.port;
 
-	if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14)
+	if (display->platform.alderlake_p || DISPLAY_VER(display) >= 14)
 		return pipe <= PIPE_B && port <= PORT_B;
 	else
 		return pipe == PIPE_A && port == PORT_A;
@@ -1197,7 +1201,6 @@ tgl_dc3co_exitline_compute_config(struct intel_dp *intel_dp,
 				  struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	const u32 crtc_vdisplay = crtc_state->uapi.adjusted_mode.crtc_vdisplay;
 	struct i915_power_domains *power_domains = &display->power.domains;
 	u32 exit_scanlines;
@@ -1223,7 +1226,7 @@ tgl_dc3co_exitline_compute_config(struct intel_dp *intel_dp,
 		return;
 
 	/* Wa_16011303918:adl-p */
-	if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(display, STEP_A0, STEP_B0))
+	if (display->platform.alderlake_p && IS_DISPLAY_STEP(display, STEP_A0, STEP_B0))
 		return;
 
 	/*
@@ -1264,7 +1267,6 @@ static bool psr2_granularity_check(struct intel_dp *intel_dp,
 				   struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	const struct drm_dsc_config *vdsc_cfg = &crtc_state->dsc.config;
 	const int crtc_hdisplay = crtc_state->hw.adjusted_mode.crtc_hdisplay;
 	const int crtc_vdisplay = crtc_state->hw.adjusted_mode.crtc_vdisplay;
@@ -1286,7 +1288,7 @@ static bool psr2_granularity_check(struct intel_dp *intel_dp,
 	 * For other platforms with SW tracking we can adjust the y coordinates
 	 * to match sink requirement if multiple of 4.
 	 */
-	if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14)
+	if (display->platform.alderlake_p || DISPLAY_VER(display) >= 14)
 		y_granularity = intel_dp->psr.su_y_granularity;
 	else if (intel_dp->psr.su_y_granularity <= 2)
 		y_granularity = 4;
@@ -1412,7 +1414,6 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 				    struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	int crtc_hdisplay = crtc_state->hw.adjusted_mode.crtc_hdisplay;
 	int crtc_vdisplay = crtc_state->hw.adjusted_mode.crtc_vdisplay;
 	int psr_max_h = 0, psr_max_v = 0, max_bpp = 0;
@@ -1421,20 +1422,20 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 		return false;
 
 	/* JSL and EHL only supports eDP 1.3 */
-	if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) {
+	if (display->platform.jasperlake || display->platform.elkhartlake) {
 		drm_dbg_kms(display->drm, "PSR2 not supported by phy\n");
 		return false;
 	}
 
 	/* Wa_16011181250 */
-	if (IS_ROCKETLAKE(dev_priv) || IS_ALDERLAKE_S(dev_priv) ||
-	    IS_DG2(dev_priv)) {
+	if (display->platform.rocketlake || display->platform.alderlake_s ||
+	    display->platform.dg2) {
 		drm_dbg_kms(display->drm,
 			    "PSR2 is defeatured for this platform\n");
 		return false;
 	}
 
-	if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(display, STEP_A0, STEP_B0)) {
+	if (display->platform.alderlake_p && IS_DISPLAY_STEP(display, STEP_A0, STEP_B0)) {
 		drm_dbg_kms(display->drm,
 			    "PSR2 not completely functional in this stepping\n");
 		return false;
@@ -1453,7 +1454,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 	 * over PSR2.
 	 */
 	if (crtc_state->dsc.compression_enable &&
-	    (DISPLAY_VER(display) < 14 && !IS_ALDERLAKE_P(dev_priv))) {
+	    (DISPLAY_VER(display) < 14 && !display->platform.alderlake_p)) {
 		drm_dbg_kms(display->drm,
 			    "PSR2 cannot be enabled since DSC is enabled\n");
 		return false;
@@ -1486,7 +1487,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 
 	/* Wa_16011303918:adl-p */
 	if (crtc_state->vrr.enable &&
-	    IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(display, STEP_A0, STEP_B0)) {
+	    display->platform.alderlake_p && IS_DISPLAY_STEP(display, STEP_A0, STEP_B0)) {
 		drm_dbg_kms(display->drm,
 			    "PSR2 not enabled, not compatible with HW stepping + VRR\n");
 		return false;
@@ -1604,6 +1605,12 @@ _panel_replay_compute_config(struct intel_dp *intel_dp,
 		return false;
 	}
 
+	if (crtc_state->crc_enabled) {
+		drm_dbg_kms(display->drm,
+			    "Panel Replay not enabled because it would inhibit pipe CRC calculation\n");
+		return false;
+	}
+
 	if (!intel_dp_is_edp(intel_dp))
 		return true;
 
@@ -1634,12 +1641,6 @@ _panel_replay_compute_config(struct intel_dp *intel_dp,
 	if (!alpm_config_valid(intel_dp, crtc_state, true))
 		return false;
 
-	if (crtc_state->crc_enabled) {
-		drm_dbg_kms(display->drm,
-			    "Panel Replay not enabled because it would inhibit pipe CRC calculation\n");
-		return false;
-	}
-
 	return true;
 }
 
@@ -1658,6 +1659,9 @@ void intel_psr_compute_config(struct intel_dp *intel_dp,
 {
 	struct intel_display *display = to_intel_display(intel_dp);
 	const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
+	struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state);
+	struct intel_crtc *crtc;
+	u8 active_pipes = 0;
 
 	if (!psr_global_enabled(intel_dp)) {
 		drm_dbg_kms(display->drm, "PSR disabled by flag\n");
@@ -1711,6 +1715,24 @@ void intel_psr_compute_config(struct intel_dp *intel_dp,
 		drm_dbg_kms(display->drm,
 			    "PSR disabled to workaround PSR FSM hang issue\n");
 	}
+
+	/* Rest is for Wa_16025596647 */
+	if (DISPLAY_VER(display) != 20 &&
+	    !IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0))
+		return;
+
+	/* Not needed by Panel Replay  */
+	if (crtc_state->has_panel_replay)
+		return;
+
+	/* We ignore possible secondary PSR/Panel Replay capable eDP */
+	for_each_intel_crtc(display->drm, crtc)
+		active_pipes |= crtc->active ? BIT(crtc->pipe) : 0;
+
+	active_pipes = intel_calc_active_pipes(state, active_pipes);
+
+	crtc_state->active_non_psr_pipes = active_pipes &
+		~BIT(to_intel_crtc(crtc_state->uapi.crtc)->pipe);
 }
 
 void intel_psr_get_config(struct intel_encoder *encoder,
@@ -1827,7 +1849,6 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 				    const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 mask = 0;
 
@@ -1866,7 +1887,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 		 * As a workaround leave LPSP unmasked to prevent PSR entry
 		 * when external displays are active.
 		 */
-		if (DISPLAY_VER(display) >= 8 || IS_HASWELL_ULT(dev_priv))
+		if (DISPLAY_VER(display) >= 8 || display->platform.haswell_ult)
 			mask |= EDP_PSR_DEBUG_MASK_LPSP;
 
 		if (DISPLAY_VER(display) < 20)
@@ -1880,7 +1901,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 			mask |= EDP_PSR_DEBUG_MASK_DISP_REG_WRITE;
 
 		/* allow PSR with sprite enabled */
-		if (IS_HASWELL(dev_priv))
+		if (display->platform.haswell)
 			mask |= EDP_PSR_DEBUG_MASK_SPRITE_ENABLE;
 	}
 
@@ -1903,9 +1924,6 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 			     intel_dp->psr.psr2_sel_fetch_enabled ?
 			     IGNORE_PSR2_HW_TRACKING : 0);
 
-	if (intel_dp_is_edp(intel_dp))
-		intel_alpm_configure(intel_dp, crtc_state);
-
 	/*
 	 * Wa_16013835468
 	 * Wa_14015648006
@@ -1925,7 +1943,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 		 */
 		if (!intel_dp->psr.panel_replay_enabled &&
 		    (IS_DISPLAY_VERx100_STEP(display, 1400, STEP_A0, STEP_B0) ||
-		     IS_ALDERLAKE_P(dev_priv)))
+		     display->platform.alderlake_p))
 			intel_de_rmw(display, CHICKEN_TRANS(display, cpu_transcoder),
 				     0, ADLP_1_BASED_X_GRANULARITY);
 
@@ -1936,10 +1954,18 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 				     MTL_CLKGATE_DIS_TRANS(display, cpu_transcoder),
 				     0,
 				     MTL_CLKGATE_DIS_TRANS_DMASC_GATING_DIS);
-		else if (IS_ALDERLAKE_P(dev_priv))
+		else if (display->platform.alderlake_p)
 			intel_de_rmw(display, CLKGATE_DIS_MISC, 0,
 				     CLKGATE_DIS_MISC_DMASC_GATING_DIS);
 	}
+
+	/* Wa_16025596647 */
+	if ((DISPLAY_VER(display) == 20 ||
+	     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
+	    !intel_dp->psr.panel_replay_enabled)
+		intel_dmc_block_pkgc(display, intel_dp->psr.pipe, true);
+
+	intel_alpm_configure(intel_dp, crtc_state);
 }
 
 static bool psr_interrupt_error_check(struct intel_dp *intel_dp)
@@ -1995,6 +2021,7 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp,
 	intel_dp->psr.psr2_sel_fetch_cff_enabled = false;
 	intel_dp->psr.req_psr2_sdp_prior_scanline =
 		crtc_state->req_psr2_sdp_prior_scanline;
+	intel_dp->psr.active_non_psr_pipes = crtc_state->active_non_psr_pipes;
 
 	if (!psr_interrupt_error_check(intel_dp))
 		return;
@@ -2006,8 +2033,9 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp,
 			    intel_dp->psr.sel_update_enabled ? "2" : "1");
 
 	/*
-	 * Enabling here only for PSR. Panel Replay enable bit is already
-	 * written at this point. See
+	 * Enabling sink PSR/Panel Replay here only for PSR. Panel Replay enable
+	 * bit is already written at this point. Sink ALPM is enabled here for
+	 * PSR and Panel Replay. See
 	 * intel_psr_panel_replay_enable_sink. Modifiers/options:
 	 *  - Selective Update
 	 *  - Region Early Transport
@@ -2024,7 +2052,7 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp,
 
 	intel_psr_enable_source(intel_dp, crtc_state);
 	intel_dp->psr.enabled = true;
-	intel_dp->psr.paused = false;
+	intel_dp->psr.pause_counter = 0;
 
 	/*
 	 * Link_ok is sticky and set here on PSR enable. We can assume link
@@ -2070,6 +2098,12 @@ static void intel_psr_exit(struct intel_dp *intel_dp)
 
 		drm_WARN_ON(display->drm, !(val & EDP_PSR2_ENABLE));
 	} else {
+		if (DISPLAY_VER(display) == 20 ||
+		    IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0))
+			intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display,
+								       intel_dp->psr.pipe,
+								       false);
+
 		val = intel_de_rmw(display,
 				   psr_ctl_reg(display, cpu_transcoder),
 				   EDP_PSR_ENABLE, 0);
@@ -2104,7 +2138,6 @@ static void intel_psr_wait_exit_locked(struct intel_dp *intel_dp)
 static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
 	lockdep_assert_held(&intel_dp->psr.lock);
@@ -2136,7 +2169,7 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 			intel_de_rmw(display,
 				     MTL_CLKGATE_DIS_TRANS(display, cpu_transcoder),
 				     MTL_CLKGATE_DIS_TRANS_DMASC_GATING_DIS, 0);
-		else if (IS_ALDERLAKE_P(dev_priv))
+		else if (display->platform.alderlake_p)
 			intel_de_rmw(display, CLKGATE_DIS_MISC,
 				     CLKGATE_DIS_MISC_DMASC_GATING_DIS, 0);
 	}
@@ -2144,16 +2177,8 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 	if (intel_dp_is_edp(intel_dp))
 		intel_snps_phy_update_psr_power_state(&dp_to_dig_port(intel_dp)->base, false);
 
-	/* Panel Replay on eDP is always using ALPM aux less. */
-	if (intel_dp->psr.panel_replay_enabled && intel_dp_is_edp(intel_dp)) {
-		intel_de_rmw(display, ALPM_CTL(display, cpu_transcoder),
-			     ALPM_CTL_ALPM_ENABLE |
-			     ALPM_CTL_ALPM_AUX_LESS_ENABLE, 0);
-
-		intel_de_rmw(display,
-			     PORT_ALPM_CTL(cpu_transcoder),
-			     PORT_ALPM_CTL_ALPM_AUX_LESS_ENABLE, 0);
-	}
+	if (intel_dp->psr.panel_replay_enabled && intel_dp_is_edp(intel_dp))
+		intel_alpm_disable(intel_dp);
 
 	/* Disable PSR on Sink */
 	if (!intel_dp->psr.panel_replay_enabled) {
@@ -2164,12 +2189,19 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 					   DP_RECEIVER_ALPM_CONFIG, 0);
 	}
 
+	/* Wa_16025596647 */
+	if ((DISPLAY_VER(display) == 20 ||
+	     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
+	    !intel_dp->psr.panel_replay_enabled)
+		intel_dmc_block_pkgc(display, intel_dp->psr.pipe, false);
+
 	intel_dp->psr.enabled = false;
 	intel_dp->psr.panel_replay_enabled = false;
 	intel_dp->psr.sel_update_enabled = false;
 	intel_dp->psr.psr2_sel_fetch_enabled = false;
 	intel_dp->psr.su_region_et_enabled = false;
 	intel_dp->psr.psr2_sel_fetch_cff_enabled = false;
+	intel_dp->psr.active_non_psr_pipes = 0;
 }
 
 /**
@@ -2210,7 +2242,6 @@ void intel_psr_disable(struct intel_dp *intel_dp,
  */
 void intel_psr_pause(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_psr *psr = &intel_dp->psr;
 
 	if (!CAN_PSR(intel_dp) && !CAN_PANEL_REPLAY(intel_dp))
@@ -2223,12 +2254,10 @@ void intel_psr_pause(struct intel_dp *intel_dp)
 		return;
 	}
 
-	/* If we ever hit this, we will need to add refcount to pause/resume */
-	drm_WARN_ON(display->drm, psr->paused);
-
-	intel_psr_exit(intel_dp);
-	intel_psr_wait_exit_locked(intel_dp);
-	psr->paused = true;
+	if (intel_dp->psr.pause_counter++ == 0) {
+		intel_psr_exit(intel_dp);
+		intel_psr_wait_exit_locked(intel_dp);
+	}
 
 	mutex_unlock(&psr->lock);
 
@@ -2244,6 +2273,7 @@ void intel_psr_pause(struct intel_dp *intel_dp)
  */
 void intel_psr_resume(struct intel_dp *intel_dp)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_psr *psr = &intel_dp->psr;
 
 	if (!CAN_PSR(intel_dp) && !CAN_PANEL_REPLAY(intel_dp))
@@ -2251,28 +2281,36 @@ void intel_psr_resume(struct intel_dp *intel_dp)
 
 	mutex_lock(&psr->lock);
 
-	if (!psr->paused)
-		goto unlock;
+	if (!psr->enabled)
+		goto out;
 
-	psr->paused = false;
-	intel_psr_activate(intel_dp);
+	if (!psr->pause_counter) {
+		drm_warn(display->drm, "Unbalanced PSR pause/resume!\n");
+		goto out;
+	}
 
-unlock:
+	if (--intel_dp->psr.pause_counter == 0)
+		intel_psr_activate(intel_dp);
+
+out:
 	mutex_unlock(&psr->lock);
 }
 
 /**
- * intel_psr_needs_block_dc_vblank - Check if block dc entry is needed
+ * intel_psr_needs_vblank_notification - Check if PSR need vblank enable/disable
+ * notification.
  * @crtc_state: CRTC status
  *
  * We need to block DC6 entry in case of Panel Replay as enabling VBI doesn't
  * prevent it in case of Panel Replay. Panel Replay switches main link off on
  * DC entry. This means vblank interrupts are not fired and is a problem if
- * user-space is polling for vblank events.
+ * user-space is polling for vblank events. Also Wa_16025596647 needs
+ * information when vblank is enabled/disabled.
  */
-bool intel_psr_needs_block_dc_vblank(const struct intel_crtc_state *crtc_state)
+bool intel_psr_needs_vblank_notification(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_encoder *encoder;
 
 	for_each_encoder_on_crtc(crtc->base.dev, &crtc->base, encoder) {
@@ -2283,8 +2321,15 @@ bool intel_psr_needs_block_dc_vblank(const struct intel_crtc_state *crtc_state)
 
 		intel_dp = enc_to_intel_dp(encoder);
 
-		if (intel_dp_is_edp(intel_dp) &&
-		    CAN_PANEL_REPLAY(intel_dp))
+		if (!intel_dp_is_edp(intel_dp))
+			continue;
+
+		if (CAN_PANEL_REPLAY(intel_dp))
+			return true;
+
+		if ((DISPLAY_VER(display) == 20 ||
+		     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
+		    CAN_PSR(intel_dp))
 			return true;
 	}
 
@@ -2312,37 +2357,76 @@ void intel_psr_trigger_frame_change_event(struct intel_dsb *dsb,
 				   CURSURFLIVE(display, crtc->pipe), 0);
 }
 
-static u32 man_trk_ctl_enable_bit_get(struct intel_display *display)
+/**
+ * intel_psr_min_vblank_delay - Minimum vblank delay needed by PSR
+ * @crtc_state: the crtc state
+ *
+ * Return minimum vblank delay needed by PSR.
+ */
+int intel_psr_min_vblank_delay(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
+	struct intel_display *display = to_intel_display(crtc_state);
+
+	if (!crtc_state->has_psr)
+		return 0;
+
+	/* Wa_14015401596 */
+	if (intel_vrr_possible(crtc_state) && IS_DISPLAY_VER(display, 13, 14))
+		return 1;
+
+	/* Rest is for SRD_STATUS needed on LunarLake and onwards */
+	if (DISPLAY_VER(display) < 20)
+		return 0;
+
+	/*
+	 * Comment on SRD_STATUS register in Bspec for LunarLake and onwards:
+	 *
+	 * To deterministically capture the transition of the state machine
+	 * going from SRDOFFACK to IDLE, the delayed V. Blank should be at least
+	 * one line after the non-delayed V. Blank.
+	 *
+	 * Legacy TG: TRANS_SET_CONTEXT_LATENCY > 0
+	 * VRR TG: TRANS_VRR_CTL[ VRR Guardband ] < (TRANS_VRR_VMAX[ VRR Vmax ]
+	 * - TRANS_VTOTAL[ Vertical Active ])
+	 *
+	 * SRD_STATUS is used only by PSR1 on PantherLake.
+	 * SRD_STATUS is used by PSR1 and Panel Replay DP on LunarLake.
+	 */
 
-	return IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14 ? 0 :
+	if (DISPLAY_VER(display) >= 30 && (crtc_state->has_panel_replay ||
+					   crtc_state->has_sel_update))
+		return 0;
+	else if (DISPLAY_VER(display) < 30 && (crtc_state->has_sel_update ||
+					       intel_crtc_has_type(crtc_state,
+								   INTEL_OUTPUT_EDP)))
+		return 0;
+	else
+		return 1;
+}
+
+static u32 man_trk_ctl_enable_bit_get(struct intel_display *display)
+{
+	return display->platform.alderlake_p || DISPLAY_VER(display) >= 14 ? 0 :
 		PSR2_MAN_TRK_CTL_ENABLE;
 }
 
 static u32 man_trk_ctl_single_full_frame_bit_get(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	return IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14 ?
+	return display->platform.alderlake_p || DISPLAY_VER(display) >= 14 ?
 	       ADLP_PSR2_MAN_TRK_CTL_SF_SINGLE_FULL_FRAME :
 	       PSR2_MAN_TRK_CTL_SF_SINGLE_FULL_FRAME;
 }
 
 static u32 man_trk_ctl_partial_frame_bit_get(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	return IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14 ?
+	return display->platform.alderlake_p || DISPLAY_VER(display) >= 14 ?
 	       ADLP_PSR2_MAN_TRK_CTL_SF_PARTIAL_FRAME_UPDATE :
 	       PSR2_MAN_TRK_CTL_SF_PARTIAL_FRAME_UPDATE;
 }
 
 static u32 man_trk_ctl_continuos_full_frame(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	return IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14 ?
+	return display->platform.alderlake_p || DISPLAY_VER(display) >= 14 ?
 	       ADLP_PSR2_MAN_TRK_CTL_SF_CONTINUOS_FULL_FRAME :
 	       PSR2_MAN_TRK_CTL_SF_CONTINUOS_FULL_FRAME;
 }
@@ -2405,8 +2489,6 @@ static void psr2_man_trk_ctl_calc(struct intel_crtc_state *crtc_state,
 				  bool full_update)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 val = man_trk_ctl_enable_bit_get(display);
 
 	/* SF partial frame enable has to be set even on full update */
@@ -2420,7 +2502,7 @@ static void psr2_man_trk_ctl_calc(struct intel_crtc_state *crtc_state,
 	if (crtc_state->psr2_su_area.y1 == -1)
 		goto exit;
 
-	if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14) {
+	if (display->platform.alderlake_p || DISPLAY_VER(display) >= 14) {
 		val |= ADLP_PSR2_MAN_TRK_CTL_SU_REGION_START_ADDR(crtc_state->psr2_su_area.y1);
 		val |= ADLP_PSR2_MAN_TRK_CTL_SU_REGION_END_ADDR(crtc_state->psr2_su_area.y2 - 1);
 	} else {
@@ -2474,13 +2556,12 @@ static void clip_area_update(struct drm_rect *overlap_damage_area,
 static void intel_psr2_sel_fetch_pipe_alignment(struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	const struct drm_dsc_config *vdsc_cfg = &crtc_state->dsc.config;
 	u16 y_alignment;
 
 	/* ADLP aligns the SU region to vdsc slice height in case dsc is enabled */
 	if (crtc_state->dsc.compression_enable &&
-	    (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(display) >= 14))
+	    (display->platform.alderlake_p || DISPLAY_VER(display) >= 14))
 		y_alignment = vdsc_cfg->slice_height;
 	else
 		y_alignment = crtc_state->su_y_granularity;
@@ -2601,12 +2682,11 @@ static void
 intel_psr_apply_su_area_workarounds(struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
 	/* Wa_14014971492 */
 	if (!crtc_state->has_panel_replay &&
 	    ((IS_DISPLAY_VERx100_STEP(display, 1400, STEP_A0, STEP_B0) ||
-	      IS_ALDERLAKE_P(i915) || IS_TIGERLAKE(i915))) &&
+	      display->platform.alderlake_p || display->platform.tigerlake)) &&
 	    crtc_state->splitter.enable)
 		crtc_state->psr2_su_area.y1 = 0;
 
@@ -2807,7 +2887,6 @@ void intel_psr_pre_plane_update(struct intel_atomic_state *state,
 				struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	const struct intel_crtc_state *old_crtc_state =
 		intel_atomic_get_old_crtc_state(state, crtc);
 	const struct intel_crtc_state *new_crtc_state =
@@ -2839,7 +2918,7 @@ void intel_psr_pre_plane_update(struct intel_atomic_state *state,
 			    new_crtc_state->has_sel_update != psr->sel_update_enabled ||
 			    new_crtc_state->enable_psr2_su_region_et != psr->su_region_et_enabled ||
 			    new_crtc_state->has_panel_replay != psr->panel_replay_enabled ||
-			    (DISPLAY_VER(i915) < 11 && new_crtc_state->wm_level_disabled))
+			    (DISPLAY_VER(display) < 11 && new_crtc_state->wm_level_disabled))
 				intel_psr_disable_locked(intel_dp);
 			else if (new_crtc_state->wm_level_disabled)
 				/* Wa_14015648006 */
@@ -3322,7 +3401,7 @@ void intel_psr_flush(struct intel_display *display,
 		 * we have to ensure that the PSR is not activated until
 		 * intel_psr_resume() is called.
 		 */
-		if (intel_dp->psr.paused)
+		if (intel_dp->psr.pause_counter)
 			goto unlock;
 
 		if (origin == ORIGIN_FLIP ||
@@ -3419,29 +3498,14 @@ static int psr_get_status_and_error_status(struct intel_dp *intel_dp,
 
 static void psr_alpm_check(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_dp_aux *aux = &intel_dp->aux;
 	struct intel_psr *psr = &intel_dp->psr;
-	u8 val;
-	int r;
 
 	if (!psr->sel_update_enabled)
 		return;
 
-	r = drm_dp_dpcd_readb(aux, DP_RECEIVER_ALPM_STATUS, &val);
-	if (r != 1) {
-		drm_err(display->drm, "Error reading ALPM status\n");
-		return;
-	}
-
-	if (val & DP_ALPM_LOCK_TIMEOUT_ERROR) {
+	if (intel_alpm_get_error(intel_dp)) {
 		intel_psr_disable_locked(intel_dp);
 		psr->sink_not_reliable = true;
-		drm_dbg_kms(display->drm,
-			    "ALPM lock timeout error, disabling PSR\n");
-
-		/* Clearing error */
-		drm_dp_dpcd_writeb(aux, DP_RECEIVER_ALPM_STATUS, val);
 	}
 }
 
@@ -3626,6 +3690,168 @@ void intel_psr_unlock(const struct intel_crtc_state *crtc_state)
 	}
 }
 
+/* Wa_16025596647 */
+static void intel_psr_apply_underrun_on_idle_wa_locked(struct intel_dp *intel_dp)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+	bool dc5_dc6_blocked;
+
+	if (!intel_dp->psr.active)
+		return;
+
+	dc5_dc6_blocked = is_dc5_dc6_blocked(intel_dp);
+
+	if (intel_dp->psr.sel_update_enabled)
+		psr2_program_idle_frames(intel_dp, dc5_dc6_blocked ? 0 :
+					 psr_compute_idle_frames(intel_dp));
+	else
+		intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display,
+								       intel_dp->psr.pipe,
+								       dc5_dc6_blocked);
+}
+
+static void psr_dc5_dc6_wa_work(struct work_struct *work)
+{
+	struct intel_display *display = container_of(work, typeof(*display),
+						     psr_dc5_dc6_wa_work);
+	struct intel_encoder *encoder;
+
+	for_each_intel_encoder_with_psr(display->drm, encoder) {
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+		mutex_lock(&intel_dp->psr.lock);
+
+		if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled)
+			intel_psr_apply_underrun_on_idle_wa_locked(intel_dp);
+
+		mutex_unlock(&intel_dp->psr.lock);
+	}
+}
+
+/**
+ * intel_psr_notify_dc5_dc6 - Notify PSR about enable/disable dc5/dc6
+ * @display: intel atomic state
+ *
+ * This is targeted for underrun on idle PSR HW bug (Wa_16025596647) to schedule
+ * psr_dc5_dc6_wa_work used for applying/removing the workaround.
+ */
+void intel_psr_notify_dc5_dc6(struct intel_display *display)
+{
+	if (DISPLAY_VER(display) != 20 &&
+	    !IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0))
+		return;
+
+	schedule_work(&display->psr_dc5_dc6_wa_work);
+}
+
+/**
+ * intel_psr_dc5_dc6_wa_init - Init work for underrun on idle PSR HW bug wa
+ * @display: intel atomic state
+ *
+ * This is targeted for underrun on idle PSR HW bug (Wa_16025596647) to init
+ * psr_dc5_dc6_wa_work used for applying the workaround.
+ */
+void intel_psr_dc5_dc6_wa_init(struct intel_display *display)
+{
+	if (DISPLAY_VER(display) != 20 &&
+	    !IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0))
+		return;
+
+	INIT_WORK(&display->psr_dc5_dc6_wa_work, psr_dc5_dc6_wa_work);
+}
+
+/**
+ * intel_psr_notify_pipe_change - Notify PSR about enable/disable of a pipe
+ * @state: intel atomic state
+ * @crtc: intel crtc
+ * @enable: enable/disable
+ *
+ * This is targeted for underrun on idle PSR HW bug (Wa_16025596647) to apply
+ * remove the workaround when pipe is getting enabled/disabled
+ */
+void intel_psr_notify_pipe_change(struct intel_atomic_state *state,
+				  struct intel_crtc *crtc, bool enable)
+{
+	struct intel_display *display = to_intel_display(state);
+	struct intel_encoder *encoder;
+
+	if (DISPLAY_VER(display) != 20 &&
+	    !IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0))
+		return;
+
+	for_each_intel_encoder_with_psr(display->drm, encoder) {
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+		u8 active_non_psr_pipes;
+
+		mutex_lock(&intel_dp->psr.lock);
+
+		if (!intel_dp->psr.enabled || intel_dp->psr.panel_replay_enabled)
+			goto unlock;
+
+		active_non_psr_pipes = intel_dp->psr.active_non_psr_pipes;
+
+		if (enable)
+			active_non_psr_pipes |= BIT(crtc->pipe);
+		else
+			active_non_psr_pipes &= ~BIT(crtc->pipe);
+
+		if (active_non_psr_pipes == intel_dp->psr.active_non_psr_pipes)
+			goto unlock;
+
+		if ((enable && intel_dp->psr.active_non_psr_pipes) ||
+		    (!enable && !intel_dp->psr.active_non_psr_pipes)) {
+			intel_dp->psr.active_non_psr_pipes = active_non_psr_pipes;
+			goto unlock;
+		}
+
+		intel_dp->psr.active_non_psr_pipes = active_non_psr_pipes;
+
+		intel_psr_apply_underrun_on_idle_wa_locked(intel_dp);
+unlock:
+		mutex_unlock(&intel_dp->psr.lock);
+	}
+}
+
+/**
+ * intel_psr_notify_vblank_enable_disable - Notify PSR about enable/disable of vblank
+ * @display: intel display struct
+ * @enable: enable/disable
+ *
+ * This is targeted for underrun on idle PSR HW bug (Wa_16025596647) to apply
+ * remove the workaround when vblank is getting enabled/disabled
+ */
+void intel_psr_notify_vblank_enable_disable(struct intel_display *display,
+					    bool enable)
+{
+	struct intel_encoder *encoder;
+
+	for_each_intel_encoder_with_psr(display->drm, encoder) {
+		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+		mutex_lock(&intel_dp->psr.lock);
+		if (intel_dp->psr.panel_replay_enabled) {
+			mutex_unlock(&intel_dp->psr.lock);
+			break;
+		}
+
+		if (intel_dp->psr.enabled)
+			intel_psr_apply_underrun_on_idle_wa_locked(intel_dp);
+
+		mutex_unlock(&intel_dp->psr.lock);
+		return;
+	}
+
+	/*
+	 * NOTE: intel_display_power_set_target_dc_state is used
+	 * only by PSR * code for DC3CO handling. DC3CO target
+	 * state is currently disabled in * PSR code. If DC3CO
+	 * is taken into use we need take that into account here
+	 * as well.
+	 */
+	intel_display_power_set_target_dc_state(display, enable ? DC_STATE_DISABLE :
+						DC_STATE_EN_UPTO_DC6);
+}
+
 static void
 psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 {
@@ -3634,8 +3860,8 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 	const char *status = "unknown";
 	u32 val, status_val;
 
-	if (intel_dp_is_edp(intel_dp) && (intel_dp->psr.sel_update_enabled ||
-					  intel_dp->psr.panel_replay_enabled)) {
+	if ((intel_dp_is_edp(intel_dp) || DISPLAY_VER(display) >= 30) &&
+	    (intel_dp->psr.sel_update_enabled || intel_dp->psr.panel_replay_enabled)) {
 		static const char * const live_status[] = {
 			"IDLE",
 			"CAPTURE",
@@ -3728,10 +3954,9 @@ static void intel_psr_print_mode(struct intel_dp *intel_dp,
 static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	struct intel_psr *psr = &intel_dp->psr;
-	intel_wakeref_t wakeref;
+	struct ref_tracker *wakeref;
 	bool enabled;
 	u32 val, psr2_ctl;
 
@@ -3740,7 +3965,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 	if (!(psr->sink_support || psr->sink_panel_replay_support))
 		return 0;
 
-	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
+	wakeref = intel_display_rpm_get(display);
 	mutex_lock(&psr->lock);
 
 	intel_psr_print_mode(intel_dp, m);
@@ -3822,7 +4047,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 
 unlock:
 	mutex_unlock(&psr->lock);
-	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+	intel_display_rpm_put(display, wakeref);
 
 	return 0;
 }
@@ -3853,9 +4078,7 @@ static int
 i915_edp_psr_debug_set(void *data, u64 val)
 {
 	struct intel_display *display = data;
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_encoder *encoder;
-	intel_wakeref_t wakeref;
 	int ret = -ENODEV;
 
 	if (!HAS_PSR(display))
@@ -3866,12 +4089,9 @@ i915_edp_psr_debug_set(void *data, u64 val)
 
 		drm_dbg_kms(display->drm, "Setting PSR debug to %llx\n", val);
 
-		wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
-
 		// TODO: split to each transcoder's PSR debug state
-		ret = intel_psr_debug_set(intel_dp, val);
-
-		intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
+		with_intel_display_rpm(display)
+			ret = intel_psr_debug_set(intel_dp, val);
 	}
 
 	return ret;
@@ -4004,3 +4224,13 @@ void intel_psr_connector_debugfs_add(struct intel_connector *connector)
 		debugfs_create_file("i915_psr_status", 0444, root,
 				    connector, &i915_psr_status_fops);
 }
+
+bool intel_psr_needs_alpm(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state)
+{
+	/*
+	 * eDP Panel Replay uses always ALPM
+	 * PSR2 uses ALPM but PSR1 doesn't
+	 */
+	return intel_dp_is_edp(intel_dp) && (crtc_state->has_sel_update ||
+					     crtc_state->has_panel_replay);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_psr.h b/drivers/gpu/drm/i915/display/intel_psr.h
index a43a374cff55..73c3fa40844b 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.h
+++ b/drivers/gpu/drm/i915/display/intel_psr.h
@@ -59,7 +59,13 @@ void intel_psr2_program_trans_man_trk_ctl(struct intel_dsb *dsb,
 					  const struct intel_crtc_state *crtc_state);
 void intel_psr_pause(struct intel_dp *intel_dp);
 void intel_psr_resume(struct intel_dp *intel_dp);
-bool intel_psr_needs_block_dc_vblank(const struct intel_crtc_state *crtc_state);
+bool intel_psr_needs_vblank_notification(const struct intel_crtc_state *crtc_state);
+void intel_psr_notify_pipe_change(struct intel_atomic_state *state,
+				  struct intel_crtc *crtc, bool enable);
+void intel_psr_notify_dc5_dc6(struct intel_display *display);
+void intel_psr_dc5_dc6_wa_init(struct intel_display *display);
+void intel_psr_notify_vblank_enable_disable(struct intel_display *display,
+					    bool enable);
 bool intel_psr_link_ok(struct intel_dp *intel_dp);
 
 void intel_psr_lock(const struct intel_crtc_state *crtc_state);
@@ -67,7 +73,9 @@ void intel_psr_unlock(const struct intel_crtc_state *crtc_state);
 void intel_psr_trigger_frame_change_event(struct intel_dsb *dsb,
 					  struct intel_atomic_state *state,
 					  struct intel_crtc *crtc);
+int intel_psr_min_vblank_delay(const struct intel_crtc_state *crtc_state);
 void intel_psr_connector_debugfs_add(struct intel_connector *connector);
 void intel_psr_debugfs_register(struct intel_display *display);
+bool intel_psr_needs_alpm(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state);
 
 #endif /* __INTEL_PSR_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 6e2d9929b4d7..8a38df2c0283 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -36,9 +36,9 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_eld.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
@@ -214,18 +214,17 @@ intel_sdvo_create_enhance_property(struct intel_sdvo *intel_sdvo,
 static void intel_sdvo_write_sdvox(struct intel_sdvo *intel_sdvo, u32 val)
 {
 	struct intel_display *display = to_intel_display(&intel_sdvo->base);
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 bval = val, cval = val;
 	int i;
 
-	if (HAS_PCH_SPLIT(dev_priv)) {
+	if (HAS_PCH_SPLIT(display)) {
 		intel_de_write(display, intel_sdvo->sdvo_reg, val);
 		intel_de_posting_read(display, intel_sdvo->sdvo_reg);
 		/*
 		 * HW workaround, need to write this twice for issue
 		 * that may result in first write getting masked.
 		 */
-		if (HAS_PCH_IBX(dev_priv)) {
+		if (HAS_PCH_IBX(display)) {
 			intel_de_write(display, intel_sdvo->sdvo_reg, val);
 			intel_de_posting_read(display, intel_sdvo->sdvo_reg);
 		}
@@ -1360,14 +1359,13 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder,
 				     struct drm_connector_state *conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_sdvo *intel_sdvo = to_sdvo(encoder);
 	struct intel_sdvo_connector *intel_sdvo_connector =
 		to_intel_sdvo_connector(conn_state->connector);
 	struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
 	struct drm_display_mode *mode = &pipe_config->hw.mode;
 
-	if (HAS_PCH_SPLIT(i915)) {
+	if (HAS_PCH_SPLIT(display)) {
 		pipe_config->has_pch_encoder = true;
 		if (!intel_fdi_compute_pipe_bpp(pipe_config))
 			return -EINVAL;
@@ -1527,7 +1525,6 @@ static void intel_sdvo_pre_enable(struct intel_atomic_state *state,
 				  const struct drm_connector_state *conn_state)
 {
 	struct intel_display *display = to_intel_display(intel_encoder);
-	struct drm_i915_private *dev_priv = to_i915(intel_encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
 	const struct intel_sdvo_connector_state *sdvo_state =
@@ -1634,7 +1631,7 @@ static void intel_sdvo_pre_enable(struct intel_atomic_state *state,
 		sdvox |= (9 << 19) | SDVO_BORDER_ENABLE;
 	}
 
-	if (HAS_PCH_CPT(dev_priv))
+	if (HAS_PCH_CPT(display))
 		sdvox |= SDVO_PIPE_SEL_CPT(crtc->pipe);
 	else
 		sdvox |= SDVO_PIPE_SEL(crtc->pipe);
@@ -1670,13 +1667,12 @@ static bool intel_sdvo_connector_get_hw_state(struct intel_connector *connector)
 bool intel_sdvo_port_enabled(struct intel_display *display,
 			     i915_reg_t sdvo_reg, enum pipe *pipe)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	u32 val;
 
 	val = intel_de_read(display, sdvo_reg);
 
 	/* asserts want to know the pipe even if the port is disabled */
-	if (HAS_PCH_CPT(dev_priv))
+	if (HAS_PCH_CPT(display))
 		*pipe = (val & SDVO_PIPE_SEL_MASK_CPT) >> SDVO_PIPE_SEL_SHIFT_CPT;
 	else if (display->platform.cherryview)
 		*pipe = (val & SDVO_PIPE_SEL_MASK_CHV) >> SDVO_PIPE_SEL_SHIFT_CHV;
@@ -1841,7 +1837,6 @@ static void intel_disable_sdvo(struct intel_atomic_state *state,
 			       const struct drm_connector_state *conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_sdvo *intel_sdvo = to_sdvo(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->uapi.crtc);
 	u32 temp;
@@ -1861,7 +1856,7 @@ static void intel_disable_sdvo(struct intel_atomic_state *state,
 	 * to transcoder A after disabling it to allow the
 	 * matching DP port to be enabled on transcoder A.
 	 */
-	if (HAS_PCH_IBX(dev_priv) && crtc->pipe == PIPE_B) {
+	if (HAS_PCH_IBX(display) && crtc->pipe == PIPE_B) {
 		/*
 		 * We get CPU/PCH FIFO underruns on the other pipe when
 		 * doing the workaround. Sweep them under the rug.
@@ -2036,7 +2031,7 @@ static u16 intel_sdvo_get_hotplug_support(struct intel_sdvo *intel_sdvo)
 	struct intel_display *display = to_intel_display(&intel_sdvo->base);
 	u16 hotplug;
 
-	if (!I915_HAS_HOTPLUG(display))
+	if (!HAS_HOTPLUG(display))
 		return 0;
 
 	/*
@@ -3367,9 +3362,7 @@ intel_sdvo_init_ddc_proxy(struct intel_sdvo_ddc *ddc,
 
 static bool is_sdvo_port_valid(struct intel_display *display, enum port port)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
-
-	if (HAS_PCH_SPLIT(dev_priv))
+	if (HAS_PCH_SPLIT(display))
 		return port == PORT_B;
 	else
 		return port == PORT_B || port == PORT_C;
@@ -3384,7 +3377,6 @@ static bool assert_sdvo_port_valid(struct intel_display *display, enum port port
 bool intel_sdvo_init(struct intel_display *display,
 		     i915_reg_t sdvo_reg, enum port port)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	struct intel_encoder *intel_encoder;
 	struct intel_sdvo *intel_sdvo;
 	int i;
@@ -3427,7 +3419,7 @@ bool intel_sdvo_init(struct intel_display *display,
 	}
 
 	intel_encoder->compute_config = intel_sdvo_compute_config;
-	if (HAS_PCH_SPLIT(dev_priv)) {
+	if (HAS_PCH_SPLIT(display)) {
 		intel_encoder->disable = pch_disable_sdvo;
 		intel_encoder->post_disable = pch_post_disable_sdvo;
 	} else {
diff --git a/drivers/gpu/drm/i915/display/intel_snps_phy.c b/drivers/gpu/drm/i915/display/intel_snps_phy.c
index b9acd9fe160c..2b53ac9f4935 100644
--- a/drivers/gpu/drm/i915/display/intel_snps_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_snps_phy.c
@@ -5,6 +5,8 @@
 
 #include <linux/math.h>
 
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
 #include "i915_utils.h"
 #include "intel_ddi.h"
diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c
index 1ad6c8a94b3d..fd92e6b89b43 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite.c
@@ -36,9 +36,10 @@
 #include <drm/drm_blend.h>
 #include <drm/drm_color_mgmt.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_print.h>
 #include <drm/drm_rect.h>
 
-#include "i915_drv.h"
+#include "i915_utils.h"
 #include "i9xx_plane.h"
 #include "intel_atomic_plane.h"
 #include "intel_de.h"
diff --git a/drivers/gpu/drm/i915/display/intel_sprite_uapi.c b/drivers/gpu/drm/i915/display/intel_sprite_uapi.c
index 1d0b84b464c1..4981cc34da05 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite_uapi.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite_uapi.c
@@ -3,21 +3,21 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#include "i915_drv.h"
 #include "intel_crtc.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_sprite_uapi.h"
 
-static bool has_dst_key_in_primary_plane(struct drm_i915_private *dev_priv)
+static bool has_dst_key_in_primary_plane(struct intel_display *display)
 {
-	return DISPLAY_VER(dev_priv) >= 9;
+	return DISPLAY_VER(display) >= 9;
 }
 
 static void intel_plane_set_ckey(struct intel_plane_state *plane_state,
 				 const struct drm_intel_sprite_colorkey *set)
 {
+	struct intel_display *display = to_intel_display(plane_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
 	struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 
 	*key = *set;
@@ -34,7 +34,7 @@ static void intel_plane_set_ckey(struct intel_plane_state *plane_state,
 	 * On SKL+ we want dst key enabled on
 	 * the primary and not on the sprite.
 	 */
-	if (DISPLAY_VER(dev_priv) >= 9 && plane->id != PLANE_PRIMARY &&
+	if (DISPLAY_VER(display) >= 9 && plane->id != PLANE_PRIMARY &&
 	    set->flags & I915_SET_COLORKEY_DESTINATION)
 		key->flags = 0;
 }
@@ -43,7 +43,6 @@ int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
 				    struct drm_file *file_priv)
 {
 	struct intel_display *display = to_intel_display(dev);
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_intel_sprite_colorkey *set = data;
 	struct drm_plane *plane;
 	struct drm_plane_state *plane_state;
@@ -61,7 +60,7 @@ int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
 	if ((set->flags & (I915_SET_COLORKEY_DESTINATION | I915_SET_COLORKEY_SOURCE)) == (I915_SET_COLORKEY_DESTINATION | I915_SET_COLORKEY_SOURCE))
 		return -EINVAL;
 
-	if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+	if ((display->platform.valleyview || display->platform.cherryview) &&
 	    set->flags & I915_SET_COLORKEY_DESTINATION)
 		return -EINVAL;
 
@@ -74,7 +73,7 @@ int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
 	 * Also multiple planes can't do destination keying on the same
 	 * pipe simultaneously.
 	 */
-	if (DISPLAY_VER(dev_priv) >= 9 &&
+	if (DISPLAY_VER(display) >= 9 &&
 	    to_intel_plane(plane)->id >= PLANE_3 &&
 	    set->flags & I915_SET_COLORKEY_DESTINATION)
 		return -EINVAL;
@@ -99,7 +98,7 @@ int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
 		 * On some platforms we have to configure
 		 * the dst colorkey on the primary plane.
 		 */
-		if (!ret && has_dst_key_in_primary_plane(dev_priv)) {
+		if (!ret && has_dst_key_in_primary_plane(display)) {
 			struct intel_crtc *crtc =
 				intel_crtc_for_pipe(display,
 						    to_intel_plane(plane)->pipe);
diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index b8d14ed8a56e..c1014e74791f 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -3,8 +3,10 @@
  * Copyright © 2019 Intel Corporation
  */
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_atomic.h"
 #include "intel_cx0_phy_regs.h"
 #include "intel_ddi.h"
@@ -92,11 +94,6 @@ static struct intel_tc_port *to_tc_port(struct intel_digital_port *dig_port)
 	return dig_port->tc;
 }
 
-static struct drm_i915_private *tc_to_i915(struct intel_tc_port *tc)
-{
-	return to_i915(tc->dig_port->base.base.dev);
-}
-
 static bool intel_tc_port_in_mode(struct intel_digital_port *dig_port,
 				  enum tc_port_mode mode)
 {
@@ -219,10 +216,11 @@ __tc_cold_unblock(struct intel_tc_port *tc, enum intel_display_power_domain doma
 static void
 tc_cold_unblock(struct intel_tc_port *tc, intel_wakeref_t wakeref)
 {
+	struct intel_display __maybe_unused *display = to_intel_display(tc->dig_port);
 	enum intel_display_power_domain domain = tc_phy_cold_off_domain(tc);
 
 #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_RUNTIME_PM)
-	drm_WARN_ON(&tc_to_i915(tc)->drm, tc->lock_power_domain != domain);
+	drm_WARN_ON(display->drm, tc->lock_power_domain != domain);
 #endif
 	__tc_cold_unblock(tc, domain, wakeref);
 }
@@ -266,13 +264,13 @@ assert_tc_port_power_enabled(struct intel_tc_port *tc)
 
 static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 	u32 lane_mask;
 
-	lane_mask = intel_de_read(i915, PORT_TX_DFLEXDPSP(tc->phy_fia));
+	lane_mask = intel_de_read(display, PORT_TX_DFLEXDPSP(tc->phy_fia));
 
-	drm_WARN_ON(&i915->drm, lane_mask == 0xffffffff);
+	drm_WARN_ON(display->drm, lane_mask == 0xffffffff);
 	assert_tc_cold_blocked(tc);
 
 	lane_mask &= DP_LANE_ASSIGNMENT_MASK(tc->phy_fia_idx);
@@ -281,13 +279,13 @@ static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port)
 
 u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 	u32 pin_mask;
 
-	pin_mask = intel_de_read(i915, PORT_TX_DFLEXPA1(tc->phy_fia));
+	pin_mask = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia));
 
-	drm_WARN_ON(&i915->drm, pin_mask == 0xffffffff);
+	drm_WARN_ON(display->drm, pin_mask == 0xffffffff);
 	assert_tc_cold_blocked(tc);
 
 	return (pin_mask & DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx)) >>
@@ -297,13 +295,12 @@ u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port)
 static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base);
 	intel_wakeref_t wakeref;
 	u32 val, pin_assignment;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
-		val = intel_de_read(i915, TCSS_DDI_STATUS(tc_port));
+		val = intel_de_read(display, TCSS_DDI_STATUS(tc_port));
 
 	pin_assignment =
 		REG_FIELD_GET(TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK, val);
@@ -369,7 +366,7 @@ static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 
 int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 
 	if (!intel_encoder_is_tc(&dig_port->base) || tc->mode != TC_PORT_DP_ALT)
@@ -377,10 +374,10 @@ int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 
 	assert_tc_cold_blocked(tc);
 
-	if (DISPLAY_VER(i915) >= 20)
+	if (DISPLAY_VER(display) >= 20)
 		return lnl_tc_port_get_max_lane_count(dig_port);
 
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		return mtl_tc_port_get_max_lane_count(dig_port);
 
 	return intel_tc_port_get_max_lane_count(dig_port);
@@ -389,20 +386,20 @@ int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,
 				      int required_lanes)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 	bool lane_reversal = dig_port->lane_reversal;
 	u32 val;
 
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		return;
 
-	drm_WARN_ON(&i915->drm,
+	drm_WARN_ON(display->drm,
 		    lane_reversal && tc->mode != TC_PORT_LEGACY);
 
 	assert_tc_cold_blocked(tc);
 
-	val = intel_de_read(i915, PORT_TX_DFLEXDPMLE1(tc->phy_fia));
+	val = intel_de_read(display, PORT_TX_DFLEXDPMLE1(tc->phy_fia));
 	val &= ~DFLEXDPMLE1_DPMLETC_MASK(tc->phy_fia_idx);
 
 	switch (required_lanes) {
@@ -423,16 +420,16 @@ void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,
 		MISSING_CASE(required_lanes);
 	}
 
-	intel_de_write(i915, PORT_TX_DFLEXDPMLE1(tc->phy_fia), val);
+	intel_de_write(display, PORT_TX_DFLEXDPMLE1(tc->phy_fia), val);
 }
 
 static void tc_port_fixup_legacy_flag(struct intel_tc_port *tc,
 				      u32 live_status_mask)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	u32 valid_hpd_mask;
 
-	drm_WARN_ON(&i915->drm, tc->mode != TC_PORT_DISCONNECTED);
+	drm_WARN_ON(display->drm, tc->mode != TC_PORT_DISCONNECTED);
 
 	if (hweight32(live_status_mask) != 1)
 		return;
@@ -447,7 +444,7 @@ static void tc_port_fixup_legacy_flag(struct intel_tc_port *tc,
 		return;
 
 	/* If live status mismatches the VBT flag, trust the live status. */
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "Port %s: live status %08x mismatches the legacy port flag %08x, fixing flag\n",
 		    tc->port_name, live_status_mask, valid_hpd_mask);
 
@@ -490,21 +487,20 @@ icl_tc_phy_cold_off_domain(struct intel_tc_port *tc)
 static u32 icl_tc_phy_hpd_live_status(struct intel_tc_port *tc)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
-	struct drm_i915_private *i915 = tc_to_i915(tc);
 	struct intel_digital_port *dig_port = tc->dig_port;
-	u32 isr_bit = i915->display.hotplug.pch_hpd[dig_port->base.hpd_pin];
+	u32 isr_bit = display->hotplug.pch_hpd[dig_port->base.hpd_pin];
 	intel_wakeref_t wakeref;
 	u32 fia_isr;
 	u32 pch_isr;
 	u32 mask = 0;
 
 	with_intel_display_power(display, tc_phy_cold_off_domain(tc), wakeref) {
-		fia_isr = intel_de_read(i915, PORT_TX_DFLEXDPSP(tc->phy_fia));
-		pch_isr = intel_de_read(i915, SDEISR);
+		fia_isr = intel_de_read(display, PORT_TX_DFLEXDPSP(tc->phy_fia));
+		pch_isr = intel_de_read(display, SDEISR);
 	}
 
 	if (fia_isr == 0xffffffff) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: PHY in TCCOLD, nothing connected\n",
 			    tc->port_name);
 		return mask;
@@ -531,14 +527,14 @@ static u32 icl_tc_phy_hpd_live_status(struct intel_tc_port *tc)
  */
 static bool icl_tc_phy_is_ready(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	u32 val;
 
 	assert_tc_cold_blocked(tc);
 
-	val = intel_de_read(i915, PORT_TX_DFLEXDPPMS(tc->phy_fia));
+	val = intel_de_read(display, PORT_TX_DFLEXDPPMS(tc->phy_fia));
 	if (val == 0xffffffff) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: PHY in TCCOLD, assuming not ready\n",
 			    tc->port_name);
 		return false;
@@ -550,14 +546,14 @@ static bool icl_tc_phy_is_ready(struct intel_tc_port *tc)
 static bool icl_tc_phy_take_ownership(struct intel_tc_port *tc,
 				      bool take)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	u32 val;
 
 	assert_tc_cold_blocked(tc);
 
-	val = intel_de_read(i915, PORT_TX_DFLEXDPCSSS(tc->phy_fia));
+	val = intel_de_read(display, PORT_TX_DFLEXDPCSSS(tc->phy_fia));
 	if (val == 0xffffffff) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: PHY in TCCOLD, can't %s ownership\n",
 			    tc->port_name, take ? "take" : "release");
 
@@ -568,21 +564,21 @@ static bool icl_tc_phy_take_ownership(struct intel_tc_port *tc,
 	if (take)
 		val |= DP_PHY_MODE_STATUS_NOT_SAFE(tc->phy_fia_idx);
 
-	intel_de_write(i915, PORT_TX_DFLEXDPCSSS(tc->phy_fia), val);
+	intel_de_write(display, PORT_TX_DFLEXDPCSSS(tc->phy_fia), val);
 
 	return true;
 }
 
 static bool icl_tc_phy_is_owned(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	u32 val;
 
 	assert_tc_cold_blocked(tc);
 
-	val = intel_de_read(i915, PORT_TX_DFLEXDPCSSS(tc->phy_fia));
+	val = intel_de_read(display, PORT_TX_DFLEXDPCSSS(tc->phy_fia));
 	if (val == 0xffffffff) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: PHY in TCCOLD, assume not owned\n",
 			    tc->port_name);
 		return false;
@@ -619,30 +615,30 @@ static void icl_tc_phy_get_hw_state(struct intel_tc_port *tc)
 static bool tc_phy_verify_legacy_or_dp_alt_mode(struct intel_tc_port *tc,
 						int required_lanes)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	struct intel_digital_port *dig_port = tc->dig_port;
 	int max_lanes;
 
 	max_lanes = intel_tc_port_max_lane_count(dig_port);
 	if (tc->mode == TC_PORT_LEGACY) {
-		drm_WARN_ON(&i915->drm, max_lanes != 4);
+		drm_WARN_ON(display->drm, max_lanes != 4);
 		return true;
 	}
 
-	drm_WARN_ON(&i915->drm, tc->mode != TC_PORT_DP_ALT);
+	drm_WARN_ON(display->drm, tc->mode != TC_PORT_DP_ALT);
 
 	/*
 	 * Now we have to re-check the live state, in case the port recently
 	 * became disconnected. Not necessary for legacy mode.
 	 */
 	if (!(tc_phy_hpd_live_status(tc) & BIT(TC_PORT_DP_ALT))) {
-		drm_dbg_kms(&i915->drm, "Port %s: PHY sudden disconnect\n",
+		drm_dbg_kms(display->drm, "Port %s: PHY sudden disconnect\n",
 			    tc->port_name);
 		return false;
 	}
 
 	if (max_lanes < required_lanes) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: PHY max lanes %d < required lanes %d\n",
 			    tc->port_name,
 			    max_lanes, required_lanes);
@@ -655,7 +651,7 @@ static bool tc_phy_verify_legacy_or_dp_alt_mode(struct intel_tc_port *tc,
 static bool icl_tc_phy_connect(struct intel_tc_port *tc,
 			       int required_lanes)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 
 	tc->lock_wakeref = tc_cold_block(tc);
 
@@ -664,8 +660,8 @@ static bool icl_tc_phy_connect(struct intel_tc_port *tc,
 
 	if ((!tc_phy_is_ready(tc) ||
 	     !icl_tc_phy_take_ownership(tc, true)) &&
-	    !drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_LEGACY)) {
-		drm_dbg_kms(&i915->drm, "Port %s: can't take PHY ownership (ready %s)\n",
+	    !drm_WARN_ON(display->drm, tc->mode == TC_PORT_LEGACY)) {
+		drm_dbg_kms(display->drm, "Port %s: can't take PHY ownership (ready %s)\n",
 			    tc->port_name,
 			    str_yes_no(tc_phy_is_ready(tc)));
 		goto out_unblock_tc_cold;
@@ -733,14 +729,13 @@ tgl_tc_phy_cold_off_domain(struct intel_tc_port *tc)
 static void tgl_tc_phy_init(struct intel_tc_port *tc)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
-	struct drm_i915_private *i915 = tc_to_i915(tc);
 	intel_wakeref_t wakeref;
 	u32 val;
 
 	with_intel_display_power(display, tc_phy_cold_off_domain(tc), wakeref)
-		val = intel_de_read(i915, PORT_TX_DFLEXDPSP(FIA1));
+		val = intel_de_read(display, PORT_TX_DFLEXDPSP(FIA1));
 
-	drm_WARN_ON(&i915->drm, val == 0xffffffff);
+	drm_WARN_ON(display->drm, val == 0xffffffff);
 
 	tc_phy_load_fia_params(tc, val & MODULAR_FIA_MASK);
 }
@@ -775,19 +770,18 @@ adlp_tc_phy_cold_off_domain(struct intel_tc_port *tc)
 static u32 adlp_tc_phy_hpd_live_status(struct intel_tc_port *tc)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
-	struct drm_i915_private *i915 = tc_to_i915(tc);
 	struct intel_digital_port *dig_port = tc->dig_port;
 	enum hpd_pin hpd_pin = dig_port->base.hpd_pin;
-	u32 cpu_isr_bits = i915->display.hotplug.hpd[hpd_pin];
-	u32 pch_isr_bit = i915->display.hotplug.pch_hpd[hpd_pin];
+	u32 cpu_isr_bits = display->hotplug.hpd[hpd_pin];
+	u32 pch_isr_bit = display->hotplug.pch_hpd[hpd_pin];
 	intel_wakeref_t wakeref;
 	u32 cpu_isr;
 	u32 pch_isr;
 	u32 mask = 0;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) {
-		cpu_isr = intel_de_read(i915, GEN11_DE_HPD_ISR);
-		pch_isr = intel_de_read(i915, SDEISR);
+		cpu_isr = intel_de_read(display, GEN11_DE_HPD_ISR);
+		pch_isr = intel_de_read(display, SDEISR);
 	}
 
 	if (cpu_isr & (cpu_isr_bits & GEN11_DE_TC_HOTPLUG_MASK))
@@ -810,15 +804,15 @@ static u32 adlp_tc_phy_hpd_live_status(struct intel_tc_port *tc)
  */
 static bool adlp_tc_phy_is_ready(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum tc_port tc_port = intel_encoder_to_tc(&tc->dig_port->base);
 	u32 val;
 
 	assert_display_core_power_enabled(tc);
 
-	val = intel_de_read(i915, TCSS_DDI_STATUS(tc_port));
+	val = intel_de_read(display, TCSS_DDI_STATUS(tc_port));
 	if (val == 0xffffffff) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: PHY in TCCOLD, assuming not ready\n",
 			    tc->port_name);
 		return false;
@@ -830,12 +824,12 @@ static bool adlp_tc_phy_is_ready(struct intel_tc_port *tc)
 static bool adlp_tc_phy_take_ownership(struct intel_tc_port *tc,
 				       bool take)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum port port = tc->dig_port->base.port;
 
 	assert_tc_port_power_enabled(tc);
 
-	intel_de_rmw(i915, DDI_BUF_CTL(port), DDI_BUF_CTL_TC_PHY_OWNERSHIP,
+	intel_de_rmw(display, DDI_BUF_CTL(port), DDI_BUF_CTL_TC_PHY_OWNERSHIP,
 		     take ? DDI_BUF_CTL_TC_PHY_OWNERSHIP : 0);
 
 	return true;
@@ -843,13 +837,13 @@ static bool adlp_tc_phy_take_ownership(struct intel_tc_port *tc,
 
 static bool adlp_tc_phy_is_owned(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum port port = tc->dig_port->base.port;
 	u32 val;
 
 	assert_tc_port_power_enabled(tc);
 
-	val = intel_de_read(i915, DDI_BUF_CTL(port));
+	val = intel_de_read(display, DDI_BUF_CTL(port));
 	return val & DDI_BUF_CTL_TC_PHY_OWNERSHIP;
 }
 
@@ -872,7 +866,6 @@ static void adlp_tc_phy_get_hw_state(struct intel_tc_port *tc)
 static bool adlp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
-	struct drm_i915_private *i915 = tc_to_i915(tc);
 	enum intel_display_power_domain port_power_domain =
 		tc_port_power_domain(tc);
 	intel_wakeref_t port_wakeref;
@@ -885,15 +878,15 @@ static bool adlp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 	port_wakeref = intel_display_power_get(display, port_power_domain);
 
 	if (!adlp_tc_phy_take_ownership(tc, true) &&
-	    !drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_LEGACY)) {
-		drm_dbg_kms(&i915->drm, "Port %s: can't take PHY ownership\n",
+	    !drm_WARN_ON(display->drm, tc->mode == TC_PORT_LEGACY)) {
+		drm_dbg_kms(display->drm, "Port %s: can't take PHY ownership\n",
 			    tc->port_name);
 		goto out_put_port_power;
 	}
 
 	if (!tc_phy_is_ready(tc) &&
-	    !drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_LEGACY)) {
-		drm_dbg_kms(&i915->drm, "Port %s: PHY not ready\n",
+	    !drm_WARN_ON(display->drm, tc->mode == TC_PORT_LEGACY)) {
+		drm_dbg_kms(display->drm, "Port %s: PHY not ready\n",
 			    tc->port_name);
 		goto out_release_phy;
 	}
@@ -965,19 +958,18 @@ static const struct intel_tc_phy_ops adlp_tc_phy_ops = {
 static u32 xelpdp_tc_phy_hpd_live_status(struct intel_tc_port *tc)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
-	struct drm_i915_private *i915 = tc_to_i915(tc);
 	struct intel_digital_port *dig_port = tc->dig_port;
 	enum hpd_pin hpd_pin = dig_port->base.hpd_pin;
-	u32 pica_isr_bits = i915->display.hotplug.hpd[hpd_pin];
-	u32 pch_isr_bit = i915->display.hotplug.pch_hpd[hpd_pin];
+	u32 pica_isr_bits = display->hotplug.hpd[hpd_pin];
+	u32 pch_isr_bit = display->hotplug.pch_hpd[hpd_pin];
 	intel_wakeref_t wakeref;
 	u32 pica_isr;
 	u32 pch_isr;
 	u32 mask = 0;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) {
-		pica_isr = intel_de_read(i915, PICAINTERRUPT_ISR);
-		pch_isr = intel_de_read(i915, SDEISR);
+		pica_isr = intel_de_read(display, PICAINTERRUPT_ISR);
+		pch_isr = intel_de_read(display, SDEISR);
 	}
 
 	if (pica_isr & (pica_isr_bits & XELPDP_DP_ALT_HOTPLUG_MASK))
@@ -994,22 +986,22 @@ static u32 xelpdp_tc_phy_hpd_live_status(struct intel_tc_port *tc)
 static bool
 xelpdp_tc_phy_tcss_power_is_enabled(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum port port = tc->dig_port->base.port;
-	i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port);
+	i915_reg_t reg = XELPDP_PORT_BUF_CTL1(display, port);
 
 	assert_tc_cold_blocked(tc);
 
-	return intel_de_read(i915, reg) & XELPDP_TCSS_POWER_STATE;
+	return intel_de_read(display, reg) & XELPDP_TCSS_POWER_STATE;
 }
 
 static bool
 xelpdp_tc_phy_wait_for_tcss_power(struct intel_tc_port *tc, bool enabled)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 
 	if (wait_for(xelpdp_tc_phy_tcss_power_is_enabled(tc) == enabled, 5)) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Port %s: timeout waiting for TCSS power to get %s\n",
 			    str_enabled_disabled(enabled),
 			    tc->port_name);
@@ -1069,7 +1061,7 @@ static void __xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool ena
 
 static bool xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enable)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 
 	__xelpdp_tc_phy_enable_tcss_power(tc, enable);
 
@@ -1082,7 +1074,7 @@ static bool xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enabl
 	return true;
 
 out_disable:
-	if (drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_LEGACY))
+	if (drm_WARN_ON(display->drm, tc->mode == TC_PORT_LEGACY))
 		return false;
 
 	if (!enable)
@@ -1096,35 +1088,35 @@ out_disable:
 
 static void xelpdp_tc_phy_take_ownership(struct intel_tc_port *tc, bool take)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum port port = tc->dig_port->base.port;
-	i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port);
+	i915_reg_t reg = XELPDP_PORT_BUF_CTL1(display, port);
 	u32 val;
 
 	assert_tc_cold_blocked(tc);
 
-	val = intel_de_read(i915, reg);
+	val = intel_de_read(display, reg);
 	if (take)
 		val |= XELPDP_TC_PHY_OWNERSHIP;
 	else
 		val &= ~XELPDP_TC_PHY_OWNERSHIP;
-	intel_de_write(i915, reg, val);
+	intel_de_write(display, reg, val);
 }
 
 static bool xelpdp_tc_phy_is_owned(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum port port = tc->dig_port->base.port;
-	i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port);
+	i915_reg_t reg = XELPDP_PORT_BUF_CTL1(display, port);
 
 	assert_tc_cold_blocked(tc);
 
-	return intel_de_read(i915, reg) & XELPDP_TC_PHY_OWNERSHIP;
+	return intel_de_read(display, reg) & XELPDP_TC_PHY_OWNERSHIP;
 }
 
 static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	intel_wakeref_t tc_cold_wref;
 	enum intel_display_power_domain domain;
 
@@ -1134,7 +1126,7 @@ static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc)
 	if (tc->mode != TC_PORT_DISCONNECTED)
 		tc->lock_wakeref = tc_cold_block(tc);
 
-	drm_WARN_ON(&i915->drm,
+	drm_WARN_ON(display->drm,
 		    (tc->mode == TC_PORT_DP_ALT || tc->mode == TC_PORT_LEGACY) &&
 		    !xelpdp_tc_phy_tcss_power_is_enabled(tc));
 
@@ -1207,13 +1199,13 @@ tc_phy_cold_off_domain(struct intel_tc_port *tc)
 
 static u32 tc_phy_hpd_live_status(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	u32 mask;
 
 	mask = tc->phy_ops->hpd_live_status(tc);
 
 	/* The sink can be connected only in a single mode. */
-	drm_WARN_ON_ONCE(&i915->drm, hweight32(mask) > 1);
+	drm_WARN_ON_ONCE(display->drm, hweight32(mask) > 1);
 
 	return mask;
 }
@@ -1236,9 +1228,9 @@ static void tc_phy_get_hw_state(struct intel_tc_port *tc)
 static bool tc_phy_is_ready_and_owned(struct intel_tc_port *tc,
 				      bool phy_is_ready, bool phy_is_owned)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 
-	drm_WARN_ON(&i915->drm, phy_is_owned && !phy_is_ready);
+	drm_WARN_ON(display->drm, phy_is_owned && !phy_is_ready);
 
 	return phy_is_ready && phy_is_owned;
 }
@@ -1246,8 +1238,7 @@ static bool tc_phy_is_ready_and_owned(struct intel_tc_port *tc,
 static bool tc_phy_is_connected(struct intel_tc_port *tc,
 				enum icl_port_dpll_id port_pll_type)
 {
-	struct intel_encoder *encoder = &tc->dig_port->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	bool phy_is_ready = tc_phy_is_ready(tc);
 	bool phy_is_owned = tc_phy_is_owned(tc);
 	bool is_connected;
@@ -1257,7 +1248,7 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc,
 	else
 		is_connected = port_pll_type == ICL_PORT_DPLL_DEFAULT;
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "Port %s: PHY connected: %s (ready: %s, owned: %s, pll_type: %s)\n",
 		    tc->port_name,
 		    str_yes_no(is_connected),
@@ -1270,10 +1261,10 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc,
 
 static bool tc_phy_wait_for_ready(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 
 	if (wait_for(tc_phy_is_ready(tc), 500)) {
-		drm_err(&i915->drm, "Port %s: timeout waiting for PHY ready\n",
+		drm_err(display->drm, "Port %s: timeout waiting for PHY ready\n",
 			tc->port_name);
 
 		return false;
@@ -1343,7 +1334,7 @@ get_tc_mode_in_phy_not_owned_state(struct intel_tc_port *tc,
 static enum tc_port_mode
 tc_phy_get_current_mode(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	enum tc_port_mode live_mode = tc_phy_hpd_live_mode(tc);
 	bool phy_is_ready;
 	bool phy_is_owned;
@@ -1363,11 +1354,11 @@ tc_phy_get_current_mode(struct intel_tc_port *tc)
 	if (!tc_phy_is_ready_and_owned(tc, phy_is_ready, phy_is_owned)) {
 		mode = get_tc_mode_in_phy_not_owned_state(tc, live_mode);
 	} else {
-		drm_WARN_ON(&i915->drm, live_mode == TC_PORT_TBT_ALT);
+		drm_WARN_ON(display->drm, live_mode == TC_PORT_TBT_ALT);
 		mode = get_tc_mode_in_phy_owned_state(tc, live_mode);
 	}
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "Port %s: PHY mode: %s (ready: %s, owned: %s, HPD: %s)\n",
 		    tc->port_name,
 		    tc_port_mode_name(mode),
@@ -1407,7 +1398,7 @@ tc_phy_get_target_mode(struct intel_tc_port *tc)
 
 static void tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	u32 live_status_mask = tc_phy_hpd_live_status(tc);
 	bool connected;
 
@@ -1421,7 +1412,7 @@ static void tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 		connected = tc->phy_ops->connect(tc, required_lanes);
 	}
 
-	drm_WARN_ON(&i915->drm, !connected);
+	drm_WARN_ON(display->drm, !connected);
 }
 
 static void tc_phy_disconnect(struct intel_tc_port *tc)
@@ -1491,12 +1482,12 @@ static void __intel_tc_port_put_link(struct intel_tc_port *tc)
 
 static bool tc_port_is_enabled(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	struct intel_digital_port *dig_port = tc->dig_port;
 
 	assert_tc_port_power_enabled(tc);
 
-	return intel_de_read(i915, DDI_BUF_CTL(dig_port->base.port)) &
+	return intel_de_read(display, DDI_BUF_CTL(dig_port->base.port)) &
 	       DDI_BUF_CTL_ENABLE;
 }
 
@@ -1509,15 +1500,15 @@ static bool tc_port_is_enabled(struct intel_tc_port *tc)
  */
 void intel_tc_port_init_mode(struct intel_digital_port *dig_port)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 	bool update_mode = false;
 
 	mutex_lock(&tc->lock);
 
-	drm_WARN_ON(&i915->drm, tc->mode != TC_PORT_DISCONNECTED);
-	drm_WARN_ON(&i915->drm, tc->lock_wakeref);
-	drm_WARN_ON(&i915->drm, tc->link_refcount);
+	drm_WARN_ON(display->drm, tc->mode != TC_PORT_DISCONNECTED);
+	drm_WARN_ON(display->drm, tc->lock_wakeref);
+	drm_WARN_ON(display->drm, tc->link_refcount);
 
 	tc_phy_get_hw_state(tc);
 	/*
@@ -1540,8 +1531,8 @@ void intel_tc_port_init_mode(struct intel_digital_port *dig_port)
 	if (!tc_port_is_enabled(tc)) {
 		update_mode = true;
 	} else if (tc->mode == TC_PORT_DISCONNECTED) {
-		drm_WARN_ON(&i915->drm, !tc->legacy_port);
-		drm_err(&i915->drm,
+		drm_WARN_ON(display->drm, !tc->legacy_port);
+		drm_err(display->drm,
 			"Port %s: PHY disconnected on enabled port, connecting it\n",
 			tc->port_name);
 		update_mode = true;
@@ -1556,28 +1547,28 @@ void intel_tc_port_init_mode(struct intel_digital_port *dig_port)
 	mutex_unlock(&tc->lock);
 }
 
-static bool tc_port_has_active_links(struct intel_tc_port *tc,
-				     const struct intel_crtc_state *crtc_state)
+static bool tc_port_has_active_streams(struct intel_tc_port *tc,
+				       const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	struct intel_digital_port *dig_port = tc->dig_port;
 	enum icl_port_dpll_id pll_type = ICL_PORT_DPLL_DEFAULT;
-	int active_links = 0;
+	int active_streams = 0;
 
 	if (dig_port->dp.is_mst) {
 		/* TODO: get the PLL type for MST, once HW readout is done for it. */
-		active_links = intel_dp_mst_encoder_active_links(dig_port);
+		active_streams = intel_dp_mst_active_streams(&dig_port->dp);
 	} else if (crtc_state && crtc_state->hw.active) {
 		pll_type = intel_ddi_port_pll_type(&dig_port->base, crtc_state);
-		active_links = 1;
+		active_streams = 1;
 	}
 
-	if (active_links && !tc_phy_is_connected(tc, pll_type))
-		drm_err(&i915->drm,
-			"Port %s: PHY disconnected with %d active link(s)\n",
-			tc->port_name, active_links);
+	if (active_streams && !tc_phy_is_connected(tc, pll_type))
+		drm_err(display->drm,
+			"Port %s: PHY disconnected with %d active stream(s)\n",
+			tc->port_name, active_streams);
 
-	return active_links;
+	return active_streams;
 }
 
 /**
@@ -1595,13 +1586,13 @@ static bool tc_port_has_active_links(struct intel_tc_port *tc,
 void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port,
 				 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 
 	mutex_lock(&tc->lock);
 
-	drm_WARN_ON(&i915->drm, tc->link_refcount != 1);
-	if (!tc_port_has_active_links(tc, crtc_state)) {
+	drm_WARN_ON(display->drm, tc->link_refcount != 1);
+	if (!tc_port_has_active_streams(tc, crtc_state)) {
 		/*
 		 * TBT-alt is the default mode in any case the PHY ownership is not
 		 * held (regardless of the sink's connected live state), so
@@ -1610,7 +1601,7 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port,
 		 */
 		if (tc->init_mode != TC_PORT_TBT_ALT &&
 		    tc->init_mode != TC_PORT_DISCONNECTED)
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "Port %s: PHY left in %s mode on disabled port, disconnecting it\n",
 				    tc->port_name,
 				    tc_port_mode_name(tc->init_mode));
@@ -1618,7 +1609,7 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port,
 		__intel_tc_port_put_link(tc);
 	}
 
-	drm_dbg_kms(&i915->drm, "Port %s: sanitize mode (%s)\n",
+	drm_dbg_kms(display->drm, "Port %s: sanitize mode (%s)\n",
 		    tc->port_name,
 		    tc_port_mode_name(tc->mode));
 
@@ -1637,12 +1628,12 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port,
  */
 bool intel_tc_port_connected(struct intel_encoder *encoder)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 	u32 mask = ~0;
 
-	drm_WARN_ON(&i915->drm, !intel_tc_port_ref_held(dig_port));
+	drm_WARN_ON(display->drm, !intel_tc_port_ref_held(dig_port));
 
 	if (tc->mode != TC_PORT_DISCONNECTED)
 		mask = BIT(tc->mode);
@@ -1677,14 +1668,14 @@ static int reset_link_commit(struct intel_tc_port *tc,
 			     struct intel_atomic_state *state,
 			     struct drm_modeset_acquire_ctx *ctx)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	struct intel_digital_port *dig_port = tc->dig_port;
 	struct intel_dp *intel_dp = enc_to_intel_dp(&dig_port->base);
 	struct intel_crtc *crtc;
 	u8 pipe_mask;
 	int ret;
 
-	ret = drm_modeset_lock(&i915->drm.mode_config.connection_mutex, ctx);
+	ret = drm_modeset_lock(&display->drm->mode_config.connection_mutex, ctx);
 	if (ret)
 		return ret;
 
@@ -1695,7 +1686,7 @@ static int reset_link_commit(struct intel_tc_port *tc,
 	if (!pipe_mask)
 		return 0;
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, crtc, pipe_mask) {
+	for_each_intel_crtc_in_pipe_mask(display->drm, crtc, pipe_mask) {
 		struct intel_crtc_state *crtc_state;
 
 		crtc_state = intel_atomic_get_crtc_state(&state->base, crtc);
@@ -1713,13 +1704,13 @@ static int reset_link_commit(struct intel_tc_port *tc,
 
 static int reset_link(struct intel_tc_port *tc)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	struct drm_modeset_acquire_ctx ctx;
 	struct drm_atomic_state *_state;
 	struct intel_atomic_state *state;
 	int ret;
 
-	_state = drm_atomic_state_alloc(&i915->drm);
+	_state = drm_atomic_state_alloc(display->drm);
 	if (!_state)
 		return -ENOMEM;
 
@@ -1738,21 +1729,21 @@ static void intel_tc_port_link_reset_work(struct work_struct *work)
 {
 	struct intel_tc_port *tc =
 		container_of(work, struct intel_tc_port, link_reset_work.work);
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	int ret;
 
 	if (!__intel_tc_port_link_needs_reset(tc))
 		return;
 
-	mutex_lock(&i915->drm.mode_config.mutex);
+	mutex_lock(&display->drm->mode_config.mutex);
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "Port %s: TypeC DP-alt sink disconnected, resetting link\n",
 		    tc->port_name);
 	ret = reset_link(tc);
-	drm_WARN_ON(&i915->drm, ret);
+	drm_WARN_ON(display->drm, ret);
 
-	mutex_unlock(&i915->drm.mode_config.mutex);
+	mutex_unlock(&display->drm->mode_config.mutex);
 }
 
 bool intel_tc_port_link_reset(struct intel_digital_port *dig_port)
@@ -1780,7 +1771,7 @@ void intel_tc_port_link_cancel_reset_work(struct intel_digital_port *dig_port)
 static void __intel_tc_port_lock(struct intel_tc_port *tc,
 				 int required_lanes)
 {
-	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 
 	mutex_lock(&tc->lock);
 
@@ -1790,9 +1781,8 @@ static void __intel_tc_port_lock(struct intel_tc_port *tc,
 		intel_tc_port_update_mode(tc, required_lanes,
 					  false);
 
-	drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_DISCONNECTED);
-	drm_WARN_ON(&i915->drm, tc->mode != TC_PORT_TBT_ALT &&
-				!tc_phy_is_owned(tc));
+	drm_WARN_ON(display->drm, tc->mode == TC_PORT_DISCONNECTED);
+	drm_WARN_ON(display->drm, tc->mode != TC_PORT_TBT_ALT && !tc_phy_is_owned(tc));
 }
 
 void intel_tc_port_lock(struct intel_digital_port *dig_port)
@@ -1885,12 +1875,12 @@ void intel_tc_port_put_link(struct intel_digital_port *dig_port)
 
 int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy)
 {
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc;
 	enum port port = dig_port->base.port;
 	enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base);
 
-	if (drm_WARN_ON(&i915->drm, tc_port == TC_PORT_NONE))
+	if (drm_WARN_ON(display->drm, tc_port == TC_PORT_NONE))
 		return -EINVAL;
 
 	tc = kzalloc(sizeof(*tc), GFP_KERNEL);
@@ -1900,11 +1890,11 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy)
 	dig_port->tc = tc;
 	tc->dig_port = dig_port;
 
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		tc->phy_ops = &xelpdp_tc_phy_ops;
-	else if (DISPLAY_VER(i915) >= 13)
+	else if (DISPLAY_VER(display) >= 13)
 		tc->phy_ops = &adlp_tc_phy_ops;
-	else if (DISPLAY_VER(i915) >= 12)
+	else if (DISPLAY_VER(display) >= 12)
 		tc->phy_ops = &tgl_tc_phy_ops;
 	else
 		tc->phy_ops = &icl_tc_phy_ops;
diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c
index 5dbe857ea85b..acf0b3733908 100644
--- a/drivers/gpu/drm/i915/display/intel_tv.c
+++ b/drivers/gpu/drm/i915/display/intel_tv.c
@@ -33,15 +33,15 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
+#include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_connector.h"
 #include "intel_crtc.h"
 #include "intel_de.h"
-#include "intel_display_irq.h"
 #include "intel_display_driver.h"
+#include "intel_display_irq.h"
 #include "intel_display_types.h"
 #include "intel_dpll.h"
 #include "intel_hotplug.h"
@@ -1585,19 +1585,17 @@ intel_tv_detect_type(struct intel_tv *intel_tv,
 {
 	struct intel_display *display = to_intel_display(connector->dev);
 	struct intel_crtc *crtc = to_intel_crtc(connector->state->crtc);
-	struct drm_device *dev = connector->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 tv_ctl, save_tv_ctl;
 	u32 tv_dac, save_tv_dac;
 	int type;
 
 	/* Disable TV interrupts around load detect or we'll recurse */
 	if (connector->polled & DRM_CONNECTOR_POLL_HPD) {
-		spin_lock_irq(&dev_priv->irq_lock);
-		i915_disable_pipestat(dev_priv, 0,
+		spin_lock_irq(&display->irq.lock);
+		i915_disable_pipestat(display, 0,
 				      PIPE_HOTPLUG_INTERRUPT_STATUS |
 				      PIPE_HOTPLUG_TV_INTERRUPT_STATUS);
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_unlock_irq(&display->irq.lock);
 	}
 
 	save_tv_dac = tv_dac = intel_de_read(display, TV_DAC);
@@ -1668,11 +1666,11 @@ intel_tv_detect_type(struct intel_tv *intel_tv,
 
 	/* Restore interrupt config */
 	if (connector->polled & DRM_CONNECTOR_POLL_HPD) {
-		spin_lock_irq(&dev_priv->irq_lock);
-		i915_enable_pipestat(dev_priv, 0,
+		spin_lock_irq(&display->irq.lock);
+		i915_enable_pipestat(display, 0,
 				     PIPE_HOTPLUG_INTERRUPT_STATUS |
 				     PIPE_HOTPLUG_TV_INTERRUPT_STATUS);
-		spin_unlock_irq(&dev_priv->irq_lock);
+		spin_unlock_irq(&display->irq.lock);
 	}
 
 	return type;
diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c
index 4efd4f7d497a..139fa5deba80 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.c
+++ b/drivers/gpu/drm/i915/display/intel_vblank.c
@@ -222,12 +222,15 @@ int intel_crtc_scanline_offset(const struct intel_crtc_state *crtc_state)
 	 * However if queried just before the start of vblank we'll get an
 	 * answer that's slightly in the future.
 	 */
-	if (DISPLAY_VER(display) == 2)
-		return -1;
-	else if (HAS_DDI(display) && intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
-		return 2;
-	else
+	if (DISPLAY_VER(display) >= 20 || display->platform.battlemage)
+		return 1;
+	else if (DISPLAY_VER(display) >= 9 ||
+		 display->platform.broadwell || display->platform.haswell)
+		return intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) ? 2 : 1;
+	else if (DISPLAY_VER(display) >= 3)
 		return 1;
+	else
+		return -1;
 }
 
 /*
diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
index 3ed64c17bdff..8e799e225af1 100644
--- a/drivers/gpu/drm/i915/display/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
@@ -9,6 +9,7 @@
 
 #include <drm/display/drm_dsc_helper.h>
 #include <drm/drm_fixed.h>
+#include <drm/drm_print.h>
 
 #include "i915_utils.h"
 #include "intel_crtc.h"
@@ -259,6 +260,15 @@ static int intel_dsc_slice_dimensions_valid(struct intel_crtc_state *pipe_config
 	return 0;
 }
 
+static bool is_dsi_dsc_1_1(struct intel_crtc_state *crtc_state)
+{
+	struct drm_dsc_config *vdsc_cfg = &crtc_state->dsc.config;
+
+	return vdsc_cfg->dsc_version_major == 1 &&
+		vdsc_cfg->dsc_version_minor == 1 &&
+		intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI);
+}
+
 int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(pipe_config);
@@ -317,8 +327,19 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 	 * From XE_LPD onwards we supports compression bpps in steps of 1
 	 * upto uncompressed bpp-1, hence add calculations for all the rc
 	 * parameters
+	 *
+	 * We don't want to calculate all rc parameters when the panel
+	 * is MIPI DSI and it's using DSC 1.1. The reason being that some
+	 * DSI panels vendors have hardcoded PPS params in the VBT causing
+	 * the parameters sent from the source which are derived through
+	 * interpolation to differ from the params the panel expects.
+	 * This causes a noise in the display.
+	 * Furthermore for DSI panels we are currently using  bits_per_pixel
+	 * (compressed bpp) hardcoded from VBT, (unlike other encoders where we
+	 * find the optimum compressed bpp) so dont need to rely on interpolation,
+	 * as we can get the required rc parameters from the tables.
 	 */
-	if (DISPLAY_VER(display) >= 13) {
+	if (DISPLAY_VER(display) >= 13 && !is_dsi_dsc_1_1(pipe_config)) {
 		calculate_rc_params(vdsc_cfg);
 	} else {
 		if ((compressed_bpp == 8 ||
diff --git a/drivers/gpu/drm/i915/display/intel_vga.c b/drivers/gpu/drm/i915/display/intel_vga.c
index 684b5d1bc87c..05d140c8032d 100644
--- a/drivers/gpu/drm/i915/display/intel_vga.c
+++ b/drivers/gpu/drm/i915/display/intel_vga.c
@@ -4,15 +4,20 @@
  */
 
 #include <linux/delay.h>
+#include <linux/pci.h>
 #include <linux/vgaarb.h>
 
+#include <drm/drm_device.h>
+#include <drm/drm_print.h>
 #include <video/vga.h>
+
 #include "soc/intel_gmch.h"
 
-#include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_de.h"
+#include "intel_display.h"
 #include "intel_vga.h"
+#include "intel_vga_regs.h"
 
 static i915_reg_t intel_vga_cntrl_reg(struct intel_display *display)
 {
@@ -24,16 +29,42 @@ static i915_reg_t intel_vga_cntrl_reg(struct intel_display *display)
 		return VGACNTRL;
 }
 
+static bool has_vga_pipe_sel(struct intel_display *display)
+{
+	if (display->platform.i845g ||
+	    display->platform.i865g)
+		return false;
+
+	if (display->platform.valleyview ||
+	    display->platform.cherryview)
+		return true;
+
+	return DISPLAY_VER(display) < 7;
+}
+
 /* Disable the VGA plane that we never use */
 void intel_vga_disable(struct intel_display *display)
 {
 	struct pci_dev *pdev = to_pci_dev(display->drm->dev);
 	i915_reg_t vga_reg = intel_vga_cntrl_reg(display);
+	enum pipe pipe;
+	u32 tmp;
 	u8 sr1;
 
-	if (intel_de_read(display, vga_reg) & VGA_DISP_DISABLE)
+	tmp = intel_de_read(display, vga_reg);
+	if (tmp & VGA_DISP_DISABLE)
 		return;
 
+	if (display->platform.cherryview)
+		pipe = REG_FIELD_GET(VGA_PIPE_SEL_MASK_CHV, tmp);
+	else if (has_vga_pipe_sel(display))
+		pipe = REG_FIELD_GET(VGA_PIPE_SEL_MASK, tmp);
+	else
+		pipe = PIPE_A;
+
+	drm_dbg_kms(display->drm, "Disabling VGA plane on pipe %c\n",
+		    pipe_name(pipe));
+
 	/* WaEnableVGAAccessThroughIOPort:ctg,elk,ilk,snb,ivb,vlv,hsw */
 	vga_get_uninterruptible(pdev, VGA_RSRC_LEGACY_IO);
 	outb(0x01, VGA_SEQ_I);
@@ -46,39 +77,6 @@ void intel_vga_disable(struct intel_display *display)
 	intel_de_posting_read(display, vga_reg);
 }
 
-void intel_vga_redisable_power_on(struct intel_display *display)
-{
-	i915_reg_t vga_reg = intel_vga_cntrl_reg(display);
-
-	if (!(intel_de_read(display, vga_reg) & VGA_DISP_DISABLE)) {
-		drm_dbg_kms(display->drm,
-			    "Something enabled VGA plane, disabling it\n");
-		intel_vga_disable(display);
-	}
-}
-
-void intel_vga_redisable(struct intel_display *display)
-{
-	intel_wakeref_t wakeref;
-
-	/*
-	 * This function can be called both from intel_modeset_setup_hw_state or
-	 * at a very early point in our resume sequence, where the power well
-	 * structures are not yet restored. Since this function is at a very
-	 * paranoid "someone might have enabled VGA while we were not looking"
-	 * level, just check if the power well is enabled instead of trying to
-	 * follow the "don't touch the power well if we don't need it" policy
-	 * the rest of the driver uses.
-	 */
-	wakeref = intel_display_power_get_if_enabled(display, POWER_DOMAIN_VGA);
-	if (!wakeref)
-		return;
-
-	intel_vga_redisable_power_on(display);
-
-	intel_display_power_put(display, POWER_DOMAIN_VGA, wakeref);
-}
-
 void intel_vga_reset_io_mem(struct intel_display *display)
 {
 	struct pci_dev *pdev = to_pci_dev(display->drm->dev);
diff --git a/drivers/gpu/drm/i915/display/intel_vga.h b/drivers/gpu/drm/i915/display/intel_vga.h
index 824dfc32a199..16d699f3b641 100644
--- a/drivers/gpu/drm/i915/display/intel_vga.h
+++ b/drivers/gpu/drm/i915/display/intel_vga.h
@@ -10,8 +10,6 @@ struct intel_display;
 
 void intel_vga_reset_io_mem(struct intel_display *display);
 void intel_vga_disable(struct intel_display *display);
-void intel_vga_redisable(struct intel_display *display);
-void intel_vga_redisable_power_on(struct intel_display *display);
 int intel_vga_register(struct intel_display *display);
 void intel_vga_unregister(struct intel_display *display);
 
diff --git a/drivers/gpu/drm/i915/display/intel_vga_regs.h b/drivers/gpu/drm/i915/display/intel_vga_regs.h
new file mode 100644
index 000000000000..cbacced1a69f
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_vga_regs.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef __INTEL_VGA_REGS_H__
+#define __INTEL_VGA_REGS_H__
+
+#include "intel_display_reg_defs.h"
+
+#define VGACNTRL	_MMIO(0x71400)
+#define VLV_VGACNTRL	_MMIO(VLV_DISPLAY_BASE + 0x71400)
+#define CPU_VGACNTRL	_MMIO(0x41000)
+#define   VGA_DISP_DISABLE			REG_BIT(31)
+#define   VGA_2X_MODE				REG_BIT(30) /* pre-ilk */
+#define   VGA_PIPE_SEL_MASK			REG_BIT(29) /* pre-ivb */
+#define   VGA_PIPE_SEL(pipe)			REG_FIELD_PREP(VGA_PIPE_SEL_MASK, (pipe))
+#define   VGA_PIPE_SEL_MASK_CHV			REG_GENMASK(29, 28) /* chv */
+#define   VGA_PIPE_SEL_CHV(pipe)		REG_FIELD_PREP(VGA_PIPE_SEL_MASK_CHV, (pipe))
+#define   VGA_BORDER_ENABLE			REG_BIT(26)
+#define   VGA_PIPE_CSC_ENABLE			REG_BIT(24) /* ilk+ */
+#define   VGA_CENTERING_ENABLE_MASK		REG_GENMASK(25, 24) /* pre-ilk */
+#define   VGA_PALETTE_READ_SEL			REG_BIT(23) /* pre-ivb */
+#define   VGA_PALETTE_A_WRITE_DISABLE		REG_BIT(22) /* pre-ivb */
+#define   VGA_PALETTE_B_WRITE_DISABLE		REG_BIT(21) /* pre-ivb */
+#define   VGA_LEGACY_8BIT_PALETTE_ENABLE	REG_BIT(20)
+#define   VGA_PALETTE_BYPASS			REG_BIT(19)
+#define   VGA_NINE_DOT_DISABLE			REG_BIT(18)
+#define   VGA_PALETTE_READ_SEL_HI_CHV		REG_BIT(15) /* chv */
+#define   VGA_PALETTE_C_WRITE_DISABLE_CHV	REG_BIT(14) /* chv */
+#define   VGA_ACTIVE_THROTTLING_MASK		REG_GENMASK(15, 12) /* ilk+ */
+#define   VGA_BLANK_THROTTLING_MASK		REG_GENMASK(11, 8) /* ilk+ */
+#define   VGA_BLINK_DUTY_CYCLE_MASK		REG_GENMASK(7, 6)
+#define   VGA_VSYNC_BLINK_RATE_MASK		REG_GENMASK(5, 0)
+
+#endif /* __INTEL_VGA_REGS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c
index cac49319026d..c6565baf815a 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.c
+++ b/drivers/gpu/drm/i915/display/intel_vrr.c
@@ -4,6 +4,8 @@
  *
  */
 
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
@@ -32,6 +34,8 @@ bool intel_vrr_is_capable(struct intel_connector *connector)
 			return false;
 		fallthrough;
 	case DRM_MODE_CONNECTOR_DisplayPort:
+		if (connector->mst.dp)
+			return false;
 		intel_dp = intel_attached_dp(connector);
 
 		if (!drm_dp_sink_can_do_video_without_timing_msa(intel_dp->dpcd))
@@ -182,7 +186,8 @@ is_cmrr_frac_required(struct intel_crtc_state *crtc_state)
 	int calculated_refresh_k, actual_refresh_k, pixel_clock_per_line;
 	struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
 
-	if (!HAS_CMRR(display))
+	/* Avoid CMRR for now till we have VRR with fixed timings working */
+	if (!HAS_CMRR(display) || true)
 		return false;
 
 	actual_refresh_k =
@@ -222,6 +227,121 @@ cmrr_get_vtotal(struct intel_crtc_state *crtc_state, bool video_mode_required)
 	return vtotal;
 }
 
+static
+void intel_vrr_compute_cmrr_timings(struct intel_crtc_state *crtc_state)
+{
+	crtc_state->cmrr.enable = true;
+	/*
+	 * TODO: Compute precise target refresh rate to determine
+	 * if video_mode_required should be true. Currently set to
+	 * false due to uncertainty about the precise target
+	 * refresh Rate.
+	 */
+	crtc_state->vrr.vmax = cmrr_get_vtotal(crtc_state, false);
+	crtc_state->vrr.vmin = crtc_state->vrr.vmax;
+	crtc_state->vrr.flipline = crtc_state->vrr.vmin;
+	crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
+}
+
+static
+void intel_vrr_compute_vrr_timings(struct intel_crtc_state *crtc_state)
+{
+	crtc_state->vrr.enable = true;
+	crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
+}
+
+/*
+ * For fixed refresh rate mode Vmin, Vmax and Flipline all are set to
+ * Vtotal value.
+ */
+static
+int intel_vrr_fixed_rr_vtotal(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	int crtc_vtotal = crtc_state->hw.adjusted_mode.crtc_vtotal;
+
+	if (DISPLAY_VER(display) >= 13)
+		return crtc_vtotal;
+	else
+		return crtc_vtotal -
+			intel_vrr_real_vblank_delay(crtc_state);
+}
+
+static
+int intel_vrr_fixed_rr_vmax(const struct intel_crtc_state *crtc_state)
+{
+	return intel_vrr_fixed_rr_vtotal(crtc_state);
+}
+
+static
+int intel_vrr_fixed_rr_vmin(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+
+	return intel_vrr_fixed_rr_vtotal(crtc_state) -
+		intel_vrr_flipline_offset(display);
+}
+
+static
+int intel_vrr_fixed_rr_flipline(const struct intel_crtc_state *crtc_state)
+{
+	return intel_vrr_fixed_rr_vtotal(crtc_state);
+}
+
+void intel_vrr_set_fixed_rr_timings(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+
+	if (!intel_vrr_possible(crtc_state))
+		return;
+
+	intel_de_write(display, TRANS_VRR_VMIN(display, cpu_transcoder),
+		       intel_vrr_fixed_rr_vmin(crtc_state) - 1);
+	intel_de_write(display, TRANS_VRR_VMAX(display, cpu_transcoder),
+		       intel_vrr_fixed_rr_vmax(crtc_state) - 1);
+	intel_de_write(display, TRANS_VRR_FLIPLINE(display, cpu_transcoder),
+		       intel_vrr_fixed_rr_flipline(crtc_state) - 1);
+}
+
+static
+void intel_vrr_compute_fixed_rr_timings(struct intel_crtc_state *crtc_state)
+{
+	/*
+	 * For fixed rr,  vmin = vmax = flipline.
+	 * vmin is already set to crtc_vtotal set vmax and flipline the same.
+	 */
+	crtc_state->vrr.vmax = crtc_state->hw.adjusted_mode.crtc_vtotal;
+	crtc_state->vrr.flipline = crtc_state->hw.adjusted_mode.crtc_vtotal;
+}
+
+static
+int intel_vrr_compute_vmin(struct intel_crtc_state *crtc_state)
+{
+	/*
+	 * To make fixed rr and vrr work seamless the guardband/pipeline full
+	 * should be set such that it satisfies both the fixed and variable
+	 * timings.
+	 * For this set the vmin as crtc_vtotal. With this we never need to
+	 * change anything to do with the guardband.
+	 */
+	return crtc_state->hw.adjusted_mode.crtc_vtotal;
+}
+
+static
+int intel_vrr_compute_vmax(struct intel_connector *connector,
+			   const struct drm_display_mode *adjusted_mode)
+{
+	const struct drm_display_info *info = &connector->base.display_info;
+	int vmax;
+
+	vmax = adjusted_mode->crtc_clock * 1000 /
+		(adjusted_mode->crtc_htotal * info->monitor_range.min_vfreq);
+	vmax = max_t(int, vmax, adjusted_mode->crtc_vtotal);
+
+	return vmax;
+}
+
 void
 intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 			 struct drm_connector_state *conn_state)
@@ -232,14 +352,9 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	bool is_edp = intel_dp_is_edp(intel_dp);
 	struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
-	const struct drm_display_info *info = &connector->base.display_info;
 	int vmin, vmax;
 
-	/*
-	 * FIXME all joined pipes share the same transcoder.
-	 * Need to account for that during VRR toggle/push/etc.
-	 */
-	if (crtc_state->joiner_pipes)
+	if (!HAS_VRR(display))
 		return;
 
 	if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE)
@@ -247,28 +362,40 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 
 	crtc_state->vrr.in_range =
 		intel_vrr_is_in_range(connector, drm_mode_vrefresh(adjusted_mode));
-	if (!crtc_state->vrr.in_range)
-		return;
-
-	if (HAS_LRR(display))
-		crtc_state->update_lrr = true;
 
-	vmin = DIV_ROUND_UP(adjusted_mode->crtc_clock * 1000,
-			    adjusted_mode->crtc_htotal * info->monitor_range.max_vfreq);
-	vmax = adjusted_mode->crtc_clock * 1000 /
-		(adjusted_mode->crtc_htotal * info->monitor_range.min_vfreq);
+	/*
+	 * Allow fixed refresh rate with VRR Timing Generator.
+	 * For now set the vrr.in_range to 0, to allow fixed_rr but skip actual
+	 * VRR and LRR.
+	 * #TODO For actual VRR with joiner, we need to figure out how to
+	 * correctly sequence transcoder level stuff vs. pipe level stuff
+	 * in the commit.
+	 */
+	if (crtc_state->joiner_pipes)
+		crtc_state->vrr.in_range = false;
 
-	vmin = max_t(int, vmin, adjusted_mode->crtc_vtotal);
-	vmax = max_t(int, vmax, adjusted_mode->crtc_vtotal);
+	vmin = intel_vrr_compute_vmin(crtc_state);
 
-	if (vmin >= vmax)
-		return;
+	if (crtc_state->vrr.in_range) {
+		if (HAS_LRR(display))
+			crtc_state->update_lrr = true;
+		vmax = intel_vrr_compute_vmax(connector, adjusted_mode);
+	} else {
+		vmax = vmin;
+	}
 
 	crtc_state->vrr.vmin = vmin;
 	crtc_state->vrr.vmax = vmax;
 
 	crtc_state->vrr.flipline = crtc_state->vrr.vmin;
 
+	if (crtc_state->uapi.vrr_enabled && vmin < vmax)
+		intel_vrr_compute_vrr_timings(crtc_state);
+	else if (is_cmrr_frac_required(crtc_state) && is_edp)
+		intel_vrr_compute_cmrr_timings(crtc_state);
+	else
+		intel_vrr_compute_fixed_rr_timings(crtc_state);
+
 	/*
 	 * flipline determines the min vblank length the hardware will
 	 * generate, and on ICL/TGL flipline>=vmin+1, hence we reduce
@@ -276,29 +403,6 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 	 */
 	crtc_state->vrr.vmin -= intel_vrr_flipline_offset(display);
 
-	/*
-	 * When panel is VRR capable and userspace has
-	 * not enabled adaptive sync mode then Fixed Average
-	 * Vtotal mode should be enabled.
-	 */
-	if (crtc_state->uapi.vrr_enabled) {
-		crtc_state->vrr.enable = true;
-		crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
-	} else if (is_cmrr_frac_required(crtc_state) && is_edp) {
-		crtc_state->vrr.enable = true;
-		crtc_state->cmrr.enable = true;
-		/*
-		 * TODO: Compute precise target refresh rate to determine
-		 * if video_mode_required should be true. Currently set to
-		 * false due to uncertainty about the precise target
-		 * refresh Rate.
-		 */
-		crtc_state->vrr.vmax = cmrr_get_vtotal(crtc_state, false);
-		crtc_state->vrr.vmin = crtc_state->vrr.vmax;
-		crtc_state->vrr.flipline = crtc_state->vrr.vmin;
-		crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
-	}
-
 	if (HAS_AS_SDP(display)) {
 		crtc_state->vrr.vsync_start =
 			(crtc_state->hw.adjusted_mode.crtc_vtotal -
@@ -340,7 +444,10 @@ static u32 trans_vrr_ctl(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 
-	if (DISPLAY_VER(display) >= 13)
+	if (DISPLAY_VER(display) >= 14)
+		return VRR_CTL_FLIP_LINE_EN |
+			XELPD_VRR_CTL_VRR_GUARDBAND(crtc_state->vrr.guardband);
+	else if (DISPLAY_VER(display) >= 13)
 		return VRR_CTL_IGN_MAX_SHIFT | VRR_CTL_FLIP_LINE_EN |
 			XELPD_VRR_CTL_VRR_GUARDBAND(crtc_state->vrr.guardband);
 	else
@@ -380,14 +487,11 @@ void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state)
 			       lower_32_bits(crtc_state->cmrr.cmrr_n));
 	}
 
-	intel_de_write(display, TRANS_VRR_VMIN(display, cpu_transcoder),
-		       crtc_state->vrr.vmin - 1);
-	intel_de_write(display, TRANS_VRR_VMAX(display, cpu_transcoder),
-		       crtc_state->vrr.vmax - 1);
-	intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
-		       trans_vrr_ctl(crtc_state));
-	intel_de_write(display, TRANS_VRR_FLIPLINE(display, cpu_transcoder),
-		       crtc_state->vrr.flipline - 1);
+	intel_vrr_set_fixed_rr_timings(crtc_state);
+
+	if (!intel_vrr_always_use_vrr_tg(display) && !crtc_state->vrr.enable)
+		intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
+			       trans_vrr_ctl(crtc_state));
 
 	if (HAS_AS_SDP(display))
 		intel_de_write(display,
@@ -461,6 +565,17 @@ bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state)
 	return intel_de_read(display, TRANS_PUSH(display, cpu_transcoder)) & TRANS_PUSH_SEND;
 }
 
+bool intel_vrr_always_use_vrr_tg(struct intel_display *display)
+{
+	if (!HAS_VRR(display))
+		return false;
+
+	if (DISPLAY_VER(display) >= 30)
+		return true;
+
+	return false;
+}
+
 void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
@@ -469,16 +584,25 @@ void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
 	if (!crtc_state->vrr.enable)
 		return;
 
+	intel_de_write(display, TRANS_VRR_VMIN(display, cpu_transcoder),
+		       crtc_state->vrr.vmin - 1);
+	intel_de_write(display, TRANS_VRR_VMAX(display, cpu_transcoder),
+		       crtc_state->vrr.vmax - 1);
+	intel_de_write(display, TRANS_VRR_FLIPLINE(display, cpu_transcoder),
+		       crtc_state->vrr.flipline - 1);
+
 	intel_de_write(display, TRANS_PUSH(display, cpu_transcoder),
 		       TRANS_PUSH_EN);
 
-	if (crtc_state->cmrr.enable) {
-		intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
-			       VRR_CTL_VRR_ENABLE | VRR_CTL_CMRR_ENABLE |
-			       trans_vrr_ctl(crtc_state));
-	} else {
-		intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
-			       VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state));
+	if (!intel_vrr_always_use_vrr_tg(display)) {
+		if (crtc_state->cmrr.enable) {
+			intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
+				       VRR_CTL_VRR_ENABLE | VRR_CTL_CMRR_ENABLE |
+				       trans_vrr_ctl(crtc_state));
+		} else {
+			intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
+				       VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state));
+		}
 	}
 }
 
@@ -490,24 +614,77 @@ void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state)
 	if (!old_crtc_state->vrr.enable)
 		return;
 
+	if (!intel_vrr_always_use_vrr_tg(display)) {
+		intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
+			       trans_vrr_ctl(old_crtc_state));
+		intel_de_wait_for_clear(display,
+					TRANS_VRR_STATUS(display, cpu_transcoder),
+					VRR_STATUS_VRR_EN_LIVE, 1000);
+		intel_de_write(display, TRANS_PUSH(display, cpu_transcoder), 0);
+	}
+
+	intel_vrr_set_fixed_rr_timings(old_crtc_state);
+}
+
+void intel_vrr_transcoder_enable(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+
+	if (!HAS_VRR(display))
+		return;
+
+	if (!intel_vrr_possible(crtc_state))
+		return;
+
+	if (!intel_vrr_always_use_vrr_tg(display)) {
+		intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
+			       trans_vrr_ctl(crtc_state));
+		return;
+	}
+
+	intel_de_write(display, TRANS_PUSH(display, cpu_transcoder),
+		       TRANS_PUSH_EN);
+
 	intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder),
-		       trans_vrr_ctl(old_crtc_state));
-	intel_de_wait_for_clear(display,
-				TRANS_VRR_STATUS(display, cpu_transcoder),
+		       VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state));
+}
+
+void intel_vrr_transcoder_disable(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+
+	if (!HAS_VRR(display))
+		return;
+
+	if (!intel_vrr_possible(crtc_state))
+		return;
+
+	intel_de_write(display, TRANS_VRR_CTL(display, cpu_transcoder), 0);
+
+	intel_de_wait_for_clear(display, TRANS_VRR_STATUS(display, cpu_transcoder),
 				VRR_STATUS_VRR_EN_LIVE, 1000);
 	intel_de_write(display, TRANS_PUSH(display, cpu_transcoder), 0);
 }
 
+bool intel_vrr_is_fixed_rr(const struct intel_crtc_state *crtc_state)
+{
+	return crtc_state->vrr.flipline &&
+	       crtc_state->vrr.flipline == crtc_state->vrr.vmax &&
+	       crtc_state->vrr.flipline == intel_vrr_vmin_flipline(crtc_state);
+}
+
 void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 trans_vrr_ctl, trans_vrr_vsync;
+	bool vrr_enable;
 
 	trans_vrr_ctl = intel_de_read(display,
 				      TRANS_VRR_CTL(display, cpu_transcoder));
 
-	crtc_state->vrr.enable = trans_vrr_ctl & VRR_CTL_VRR_ENABLE;
 	if (HAS_CMRR(display))
 		crtc_state->cmrr.enable = (trans_vrr_ctl & VRR_CTL_CMRR_ENABLE);
 
@@ -536,6 +713,16 @@ void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
 		crtc_state->vrr.vmin = intel_de_read(display,
 						     TRANS_VRR_VMIN(display, cpu_transcoder)) + 1;
 
+		/*
+		 * For platforms that always use VRR Timing Generator, the VTOTAL.Vtotal
+		 * bits are not filled. Since for these platforms TRAN_VMIN is always
+		 * filled with crtc_vtotal, use TRAN_VRR_VMIN to get the vtotal for
+		 * adjusted_mode.
+		 */
+		if (intel_vrr_always_use_vrr_tg(display))
+			crtc_state->hw.adjusted_mode.crtc_vtotal =
+				intel_vrr_vmin_vtotal(crtc_state);
+
 		if (HAS_AS_SDP(display)) {
 			trans_vrr_vsync =
 				intel_de_read(display,
@@ -547,6 +734,18 @@ void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
 		}
 	}
 
+	vrr_enable = trans_vrr_ctl & VRR_CTL_VRR_ENABLE;
+
+	if (intel_vrr_always_use_vrr_tg(display))
+		crtc_state->vrr.enable = vrr_enable && !intel_vrr_is_fixed_rr(crtc_state);
+	else
+		crtc_state->vrr.enable = vrr_enable;
+
+	/*
+	 * #TODO: For Both VRR and CMRR the flag I915_MODE_FLAG_VRR is set for mode_flags.
+	 * Since CMRR is currently disabled, set this flag for VRR for now.
+	 * Need to keep this in mind while re-enabling CMRR.
+	 */
 	if (crtc_state->vrr.enable)
 		crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_vrr.h b/drivers/gpu/drm/i915/display/intel_vrr.h
index 514822577e8a..38bf9996b883 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.h
+++ b/drivers/gpu/drm/i915/display/intel_vrr.h
@@ -13,6 +13,7 @@ struct intel_atomic_state;
 struct intel_connector;
 struct intel_crtc_state;
 struct intel_dsb;
+struct intel_display;
 
 bool intel_vrr_is_capable(struct intel_connector *connector);
 bool intel_vrr_is_in_range(struct intel_connector *connector, int vrefresh);
@@ -35,5 +36,10 @@ int intel_vrr_vmin_vtotal(const struct intel_crtc_state *crtc_state);
 int intel_vrr_vmax_vblank_start(const struct intel_crtc_state *crtc_state);
 int intel_vrr_vmin_vblank_start(const struct intel_crtc_state *crtc_state);
 int intel_vrr_vblank_delay(const struct intel_crtc_state *crtc_state);
+bool intel_vrr_is_fixed_rr(const struct intel_crtc_state *crtc_state);
+void intel_vrr_transcoder_enable(const struct intel_crtc_state *crtc_state);
+void intel_vrr_transcoder_disable(const struct intel_crtc_state *crtc_state);
+void intel_vrr_set_fixed_rr_timings(const struct intel_crtc_state *crtc_state);
+bool intel_vrr_always_use_vrr_tg(struct intel_display *display);
 
 #endif /* __INTEL_VRR_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_wm.c b/drivers/gpu/drm/i915/display/intel_wm.c
index f00f4cfc58e5..bba82e888db2 100644
--- a/drivers/gpu/drm/i915/display/intel_wm.c
+++ b/drivers/gpu/drm/i915/display/intel_wm.c
@@ -5,15 +5,18 @@
 
 #include <linux/debugfs.h>
 
-#include "i915_drv.h"
+#include <drm/drm_file.h>
+#include <drm/drm_print.h>
+
 #include "i9xx_wm.h"
+#include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_wm.h"
 #include "skl_watermark.h"
 
 /**
  * intel_update_watermarks - update FIFO watermark values based on current modes
- * @i915: i915 device
+ * @display: display device
  *
  * Calculate watermark values for the various WM regs based on current mode
  * and plane configuration.
@@ -44,10 +47,10 @@
  * We don't use the sprite, so we can ignore that.  And on Crestline we have
  * to set the non-SR watermarks to 8.
  */
-void intel_update_watermarks(struct drm_i915_private *i915)
+void intel_update_watermarks(struct intel_display *display)
 {
-	if (i915->display.funcs.wm->update_wm)
-		i915->display.funcs.wm->update_wm(i915);
+	if (display->funcs.wm->update_wm)
+		display->funcs.wm->update_wm(display);
 }
 
 int intel_wm_compute(struct intel_atomic_state *state,
@@ -64,10 +67,10 @@ int intel_wm_compute(struct intel_atomic_state *state,
 bool intel_initial_watermarks(struct intel_atomic_state *state,
 			      struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 
-	if (i915->display.funcs.wm->initial_watermarks) {
-		i915->display.funcs.wm->initial_watermarks(state, crtc);
+	if (display->funcs.wm->initial_watermarks) {
+		display->funcs.wm->initial_watermarks(state, crtc);
 		return true;
 	}
 
@@ -77,41 +80,41 @@ bool intel_initial_watermarks(struct intel_atomic_state *state,
 void intel_atomic_update_watermarks(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 
-	if (i915->display.funcs.wm->atomic_update_watermarks)
-		i915->display.funcs.wm->atomic_update_watermarks(state, crtc);
+	if (display->funcs.wm->atomic_update_watermarks)
+		display->funcs.wm->atomic_update_watermarks(state, crtc);
 }
 
 void intel_optimize_watermarks(struct intel_atomic_state *state,
 			       struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 
-	if (i915->display.funcs.wm->optimize_watermarks)
-		i915->display.funcs.wm->optimize_watermarks(state, crtc);
+	if (display->funcs.wm->optimize_watermarks)
+		display->funcs.wm->optimize_watermarks(state, crtc);
 }
 
 int intel_compute_global_watermarks(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 
-	if (i915->display.funcs.wm->compute_global_watermarks)
-		return i915->display.funcs.wm->compute_global_watermarks(state);
+	if (display->funcs.wm->compute_global_watermarks)
+		return display->funcs.wm->compute_global_watermarks(state);
 
 	return 0;
 }
 
-void intel_wm_get_hw_state(struct drm_i915_private *i915)
+void intel_wm_get_hw_state(struct intel_display *display)
 {
-	if (i915->display.funcs.wm->get_hw_state)
-		return i915->display.funcs.wm->get_hw_state(i915);
+	if (display->funcs.wm->get_hw_state)
+		return display->funcs.wm->get_hw_state(display);
 }
 
-void intel_wm_sanitize(struct drm_i915_private *i915)
+void intel_wm_sanitize(struct intel_display *display)
 {
-	if (i915->display.funcs.wm->sanitize)
-		return i915->display.funcs.wm->sanitize(i915);
+	if (display->funcs.wm->sanitize)
+		return display->funcs.wm->sanitize(display);
 }
 
 bool intel_wm_plane_visible(const struct intel_crtc_state *crtc_state,
@@ -137,16 +140,16 @@ bool intel_wm_plane_visible(const struct intel_crtc_state *crtc_state,
 		return plane_state->uapi.visible;
 }
 
-void intel_print_wm_latency(struct drm_i915_private *dev_priv,
+void intel_print_wm_latency(struct intel_display *display,
 			    const char *name, const u16 wm[])
 {
 	int level;
 
-	for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+	for (level = 0; level < display->wm.num_levels; level++) {
 		unsigned int latency = wm[level];
 
 		if (latency == 0) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(display->drm,
 				    "%s WM%d latency not provided\n",
 				    name, level);
 			continue;
@@ -156,43 +159,43 @@ void intel_print_wm_latency(struct drm_i915_private *dev_priv,
 		 * - latencies are in us on gen9.
 		 * - before then, WM1+ latency values are in 0.5us units
 		 */
-		if (DISPLAY_VER(dev_priv) >= 9)
+		if (DISPLAY_VER(display) >= 9)
 			latency *= 10;
 		else if (level > 0)
 			latency *= 5;
 
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(display->drm,
 			    "%s WM%d latency %u (%u.%u usec)\n", name, level,
 			    wm[level], latency / 10, latency % 10);
 	}
 }
 
-void intel_wm_init(struct drm_i915_private *i915)
+void intel_wm_init(struct intel_display *display)
 {
-	if (DISPLAY_VER(i915) >= 9)
-		skl_wm_init(i915);
+	if (DISPLAY_VER(display) >= 9)
+		skl_wm_init(display);
 	else
-		i9xx_wm_init(i915);
+		i9xx_wm_init(display);
 }
 
 static void wm_latency_show(struct seq_file *m, const u16 wm[8])
 {
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	int level;
 
-	drm_modeset_lock_all(&dev_priv->drm);
+	drm_modeset_lock_all(display->drm);
 
-	for (level = 0; level < dev_priv->display.wm.num_levels; level++) {
+	for (level = 0; level < display->wm.num_levels; level++) {
 		unsigned int latency = wm[level];
 
 		/*
 		 * - WM1+ latency values in 0.5us units
 		 * - latencies are in us on gen9/vlv/chv
 		 */
-		if (DISPLAY_VER(dev_priv) >= 9 ||
-		    IS_VALLEYVIEW(dev_priv) ||
-		    IS_CHERRYVIEW(dev_priv) ||
-		    IS_G4X(dev_priv))
+		if (DISPLAY_VER(display) >= 9 ||
+		    display->platform.valleyview ||
+		    display->platform.cherryview ||
+		    display->platform.g4x)
 			latency *= 10;
 		else if (level > 0)
 			latency *= 5;
@@ -201,18 +204,18 @@ static void wm_latency_show(struct seq_file *m, const u16 wm[8])
 			   level, wm[level], latency / 10, latency % 10);
 	}
 
-	drm_modeset_unlock_all(&dev_priv->drm);
+	drm_modeset_unlock_all(display->drm);
 }
 
 static int pri_wm_latency_show(struct seq_file *m, void *data)
 {
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	const u16 *latencies;
 
-	if (DISPLAY_VER(dev_priv) >= 9)
-		latencies = dev_priv->display.wm.skl_latency;
+	if (DISPLAY_VER(display) >= 9)
+		latencies = display->wm.skl_latency;
 	else
-		latencies = dev_priv->display.wm.pri_latency;
+		latencies = display->wm.pri_latency;
 
 	wm_latency_show(m, latencies);
 
@@ -221,13 +224,13 @@ static int pri_wm_latency_show(struct seq_file *m, void *data)
 
 static int spr_wm_latency_show(struct seq_file *m, void *data)
 {
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	const u16 *latencies;
 
-	if (DISPLAY_VER(dev_priv) >= 9)
-		latencies = dev_priv->display.wm.skl_latency;
+	if (DISPLAY_VER(display) >= 9)
+		latencies = display->wm.skl_latency;
 	else
-		latencies = dev_priv->display.wm.spr_latency;
+		latencies = display->wm.spr_latency;
 
 	wm_latency_show(m, latencies);
 
@@ -236,13 +239,13 @@ static int spr_wm_latency_show(struct seq_file *m, void *data)
 
 static int cur_wm_latency_show(struct seq_file *m, void *data)
 {
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	const u16 *latencies;
 
-	if (DISPLAY_VER(dev_priv) >= 9)
-		latencies = dev_priv->display.wm.skl_latency;
+	if (DISPLAY_VER(display) >= 9)
+		latencies = display->wm.skl_latency;
 	else
-		latencies = dev_priv->display.wm.cur_latency;
+		latencies = display->wm.cur_latency;
 
 	wm_latency_show(m, latencies);
 
@@ -251,39 +254,39 @@ static int cur_wm_latency_show(struct seq_file *m, void *data)
 
 static int pri_wm_latency_open(struct inode *inode, struct file *file)
 {
-	struct drm_i915_private *dev_priv = inode->i_private;
+	struct intel_display *display = inode->i_private;
 
-	if (DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv))
+	if (DISPLAY_VER(display) < 5 && !display->platform.g4x)
 		return -ENODEV;
 
-	return single_open(file, pri_wm_latency_show, dev_priv);
+	return single_open(file, pri_wm_latency_show, display);
 }
 
 static int spr_wm_latency_open(struct inode *inode, struct file *file)
 {
-	struct drm_i915_private *dev_priv = inode->i_private;
+	struct intel_display *display = inode->i_private;
 
-	if (HAS_GMCH(dev_priv))
+	if (HAS_GMCH(display))
 		return -ENODEV;
 
-	return single_open(file, spr_wm_latency_show, dev_priv);
+	return single_open(file, spr_wm_latency_show, display);
 }
 
 static int cur_wm_latency_open(struct inode *inode, struct file *file)
 {
-	struct drm_i915_private *dev_priv = inode->i_private;
+	struct intel_display *display = inode->i_private;
 
-	if (HAS_GMCH(dev_priv))
+	if (HAS_GMCH(display))
 		return -ENODEV;
 
-	return single_open(file, cur_wm_latency_show, dev_priv);
+	return single_open(file, cur_wm_latency_show, display);
 }
 
 static ssize_t wm_latency_write(struct file *file, const char __user *ubuf,
 				size_t len, loff_t *offp, u16 wm[8])
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	u16 new[8] = {};
 	int level;
 	int ret;
@@ -300,15 +303,15 @@ static ssize_t wm_latency_write(struct file *file, const char __user *ubuf,
 	ret = sscanf(tmp, "%hu %hu %hu %hu %hu %hu %hu %hu",
 		     &new[0], &new[1], &new[2], &new[3],
 		     &new[4], &new[5], &new[6], &new[7]);
-	if (ret != dev_priv->display.wm.num_levels)
+	if (ret != display->wm.num_levels)
 		return -EINVAL;
 
-	drm_modeset_lock_all(&dev_priv->drm);
+	drm_modeset_lock_all(display->drm);
 
-	for (level = 0; level < dev_priv->display.wm.num_levels; level++)
+	for (level = 0; level < display->wm.num_levels; level++)
 		wm[level] = new[level];
 
-	drm_modeset_unlock_all(&dev_priv->drm);
+	drm_modeset_unlock_all(display->drm);
 
 	return len;
 }
@@ -317,13 +320,13 @@ static ssize_t pri_wm_latency_write(struct file *file, const char __user *ubuf,
 				    size_t len, loff_t *offp)
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	u16 *latencies;
 
-	if (DISPLAY_VER(dev_priv) >= 9)
-		latencies = dev_priv->display.wm.skl_latency;
+	if (DISPLAY_VER(display) >= 9)
+		latencies = display->wm.skl_latency;
 	else
-		latencies = dev_priv->display.wm.pri_latency;
+		latencies = display->wm.pri_latency;
 
 	return wm_latency_write(file, ubuf, len, offp, latencies);
 }
@@ -332,13 +335,13 @@ static ssize_t spr_wm_latency_write(struct file *file, const char __user *ubuf,
 				    size_t len, loff_t *offp)
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	u16 *latencies;
 
-	if (DISPLAY_VER(dev_priv) >= 9)
-		latencies = dev_priv->display.wm.skl_latency;
+	if (DISPLAY_VER(display) >= 9)
+		latencies = display->wm.skl_latency;
 	else
-		latencies = dev_priv->display.wm.spr_latency;
+		latencies = display->wm.spr_latency;
 
 	return wm_latency_write(file, ubuf, len, offp, latencies);
 }
@@ -347,13 +350,13 @@ static ssize_t cur_wm_latency_write(struct file *file, const char __user *ubuf,
 				    size_t len, loff_t *offp)
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *dev_priv = m->private;
+	struct intel_display *display = m->private;
 	u16 *latencies;
 
-	if (DISPLAY_VER(dev_priv) >= 9)
-		latencies = dev_priv->display.wm.skl_latency;
+	if (DISPLAY_VER(display) >= 9)
+		latencies = display->wm.skl_latency;
 	else
-		latencies = dev_priv->display.wm.cur_latency;
+		latencies = display->wm.cur_latency;
 
 	return wm_latency_write(file, ubuf, len, offp, latencies);
 }
@@ -385,18 +388,18 @@ static const struct file_operations i915_cur_wm_latency_fops = {
 	.write = cur_wm_latency_write
 };
 
-void intel_wm_debugfs_register(struct drm_i915_private *i915)
+void intel_wm_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = i915->drm.primary;
+	struct drm_minor *minor = display->drm->primary;
 
 	debugfs_create_file("i915_pri_wm_latency", 0644, minor->debugfs_root,
-			    i915, &i915_pri_wm_latency_fops);
+			    display, &i915_pri_wm_latency_fops);
 
 	debugfs_create_file("i915_spr_wm_latency", 0644, minor->debugfs_root,
-			    i915, &i915_spr_wm_latency_fops);
+			    display, &i915_spr_wm_latency_fops);
 
 	debugfs_create_file("i915_cur_wm_latency", 0644, minor->debugfs_root,
-			    i915, &i915_cur_wm_latency_fops);
+			    display, &i915_cur_wm_latency_fops);
 
-	skl_watermark_debugfs_register(i915);
+	skl_watermark_debugfs_register(display);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_wm.h b/drivers/gpu/drm/i915/display/intel_wm.h
index 7d3a447054b3..9ad4e9eae5ca 100644
--- a/drivers/gpu/drm/i915/display/intel_wm.h
+++ b/drivers/gpu/drm/i915/display/intel_wm.h
@@ -8,13 +8,13 @@
 
 #include <linux/types.h>
 
-struct drm_i915_private;
 struct intel_atomic_state;
 struct intel_crtc;
 struct intel_crtc_state;
+struct intel_display;
 struct intel_plane_state;
 
-void intel_update_watermarks(struct drm_i915_private *i915);
+void intel_update_watermarks(struct intel_display *display);
 int intel_wm_compute(struct intel_atomic_state *state,
 		     struct intel_crtc *crtc);
 bool intel_initial_watermarks(struct intel_atomic_state *state,
@@ -24,13 +24,13 @@ void intel_atomic_update_watermarks(struct intel_atomic_state *state,
 void intel_optimize_watermarks(struct intel_atomic_state *state,
 			       struct intel_crtc *crtc);
 int intel_compute_global_watermarks(struct intel_atomic_state *state);
-void intel_wm_get_hw_state(struct drm_i915_private *i915);
-void intel_wm_sanitize(struct drm_i915_private *i915);
+void intel_wm_get_hw_state(struct intel_display *display);
+void intel_wm_sanitize(struct intel_display *display);
 bool intel_wm_plane_visible(const struct intel_crtc_state *crtc_state,
 			    const struct intel_plane_state *plane_state);
-void intel_print_wm_latency(struct drm_i915_private *i915,
+void intel_print_wm_latency(struct intel_display *display,
 			    const char *name, const u16 wm[]);
-void intel_wm_init(struct drm_i915_private *i915);
-void intel_wm_debugfs_register(struct drm_i915_private *i915);
+void intel_wm_init(struct intel_display *display);
+void intel_wm_debugfs_register(struct intel_display *display);
 
 #endif /* __INTEL_WM_H__ */
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index ee81220a7c88..c855426544cf 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -3,8 +3,10 @@
  * Copyright © 2020 Intel Corporation
  */
 
-#include "i915_drv.h"
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 70e550539bb2..c7b336359a5e 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -601,7 +601,7 @@ static u32 tgl_plane_min_alignment(struct intel_plane *plane,
 	 * Figure out what's going on here...
 	 */
 	if (display->platform.alderlake_p &&
-	    intel_plane_can_async_flip(plane, fb->modifier))
+	    intel_plane_can_async_flip(plane, fb->format->format, fb->modifier))
 		return mult * 16 * 1024;
 
 	switch (fb->modifier) {
@@ -2666,6 +2666,7 @@ static const struct drm_plane_funcs skl_plane_funcs = {
 	.atomic_duplicate_state = intel_plane_duplicate_state,
 	.atomic_destroy_state = intel_plane_destroy_state,
 	.format_mod_supported = skl_plane_format_mod_supported,
+	.format_mod_supported_async = intel_plane_format_mod_supported_async,
 };
 
 static const struct drm_plane_funcs icl_plane_funcs = {
@@ -2675,6 +2676,7 @@ static const struct drm_plane_funcs icl_plane_funcs = {
 	.atomic_duplicate_state = intel_plane_duplicate_state,
 	.atomic_destroy_state = intel_plane_destroy_state,
 	.format_mod_supported = icl_plane_format_mod_supported,
+	.format_mod_supported_async = intel_plane_format_mod_supported_async,
 };
 
 static const struct drm_plane_funcs tgl_plane_funcs = {
@@ -2684,28 +2686,29 @@ static const struct drm_plane_funcs tgl_plane_funcs = {
 	.atomic_duplicate_state = intel_plane_duplicate_state,
 	.atomic_destroy_state = intel_plane_destroy_state,
 	.format_mod_supported = tgl_plane_format_mod_supported,
+	.format_mod_supported_async = intel_plane_format_mod_supported_async,
 };
 
 static void
 skl_plane_enable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 
-	spin_lock_irq(&i915->irq_lock);
-	bdw_enable_pipe_irq(i915, pipe, GEN9_PIPE_PLANE_FLIP_DONE(plane->id));
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	bdw_enable_pipe_irq(display, pipe, GEN9_PIPE_PLANE_FLIP_DONE(plane->id));
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static void
 skl_plane_disable_flip_done(struct intel_plane *plane)
 {
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
+	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 
-	spin_lock_irq(&i915->irq_lock);
-	bdw_disable_pipe_irq(i915, pipe, GEN9_PIPE_PLANE_FLIP_DONE(plane->id));
-	spin_unlock_irq(&i915->irq_lock);
+	spin_lock_irq(&display->irq.lock);
+	bdw_disable_pipe_irq(display, pipe, GEN9_PIPE_PLANE_FLIP_DONE(plane->id));
+	spin_unlock_irq(&display->irq.lock);
 }
 
 static bool skl_plane_has_rc_ccs(struct intel_display *display,
diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index 621e97943542..8080f777910a 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -19,6 +19,7 @@
 #include "intel_de.h"
 #include "intel_display.h"
 #include "intel_display_power.h"
+#include "intel_display_rpm.h"
 #include "intel_display_types.h"
 #include "intel_fb.h"
 #include "intel_fixed.h"
@@ -34,7 +35,7 @@
  */
 #define DSB_EXE_TIME 100
 
-static void skl_sagv_disable(struct drm_i915_private *i915);
+static void skl_sagv_disable(struct intel_display *display);
 
 /* Stores plane specific WM parameters */
 struct skl_wm_params {
@@ -69,23 +70,21 @@ u8 intel_enabled_dbuf_slices_mask(struct intel_display *display)
  * FIXME: We still don't have the proper code detect if we need to apply the WA,
  * so assume we'll always need it in order to avoid underruns.
  */
-static bool skl_needs_memory_bw_wa(struct drm_i915_private *i915)
+static bool skl_needs_memory_bw_wa(struct intel_display *display)
 {
-	return DISPLAY_VER(i915) == 9;
+	return DISPLAY_VER(display) == 9;
 }
 
 bool
-intel_has_sagv(struct drm_i915_private *i915)
+intel_has_sagv(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-
 	return HAS_SAGV(display) && display->sagv.status != I915_SAGV_NOT_CONTROLLED;
 }
 
 static u32
-intel_sagv_block_time(struct drm_i915_private *i915)
+intel_sagv_block_time(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
+	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	if (DISPLAY_VER(display) >= 14) {
 		u32 val;
@@ -115,10 +114,8 @@ intel_sagv_block_time(struct drm_i915_private *i915)
 	}
 }
 
-static void intel_sagv_init(struct drm_i915_private *i915)
+static void intel_sagv_init(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-
 	if (!HAS_SAGV(display))
 		display->sagv.status = I915_SAGV_NOT_CONTROLLED;
 
@@ -127,14 +124,14 @@ static void intel_sagv_init(struct drm_i915_private *i915)
 	 * For icl+ this was already determined by intel_bw_init_hw().
 	 */
 	if (DISPLAY_VER(display) < 11)
-		skl_sagv_disable(i915);
+		skl_sagv_disable(display);
 
 	drm_WARN_ON(display->drm, display->sagv.status == I915_SAGV_UNKNOWN);
 
-	display->sagv.block_time_us = intel_sagv_block_time(i915);
+	display->sagv.block_time_us = intel_sagv_block_time(display);
 
 	drm_dbg_kms(display->drm, "SAGV supported: %s, original SAGV block time: %u us\n",
-		    str_yes_no(intel_has_sagv(i915)), display->sagv.block_time_us);
+		    str_yes_no(intel_has_sagv(display)), display->sagv.block_time_us);
 
 	/* avoid overflow when adding with wm0 latency/etc. */
 	if (drm_WARN(display->drm, display->sagv.block_time_us > U16_MAX,
@@ -142,7 +139,7 @@ static void intel_sagv_init(struct drm_i915_private *i915)
 		     display->sagv.block_time_us))
 		display->sagv.block_time_us = 0;
 
-	if (!intel_has_sagv(i915))
+	if (!intel_has_sagv(display))
 		display->sagv.block_time_us = 0;
 }
 
@@ -157,17 +154,18 @@ static void intel_sagv_init(struct drm_i915_private *i915)
  *  - All planes can enable watermarks for latencies >= SAGV engine block time
  *  - We're not using an interlaced display configuration
  */
-static void skl_sagv_enable(struct drm_i915_private *i915)
+static void skl_sagv_enable(struct intel_display *display)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	int ret;
 
-	if (!intel_has_sagv(i915))
+	if (!intel_has_sagv(display))
 		return;
 
-	if (i915->display.sagv.status == I915_SAGV_ENABLED)
+	if (display->sagv.status == I915_SAGV_ENABLED)
 		return;
 
-	drm_dbg_kms(&i915->drm, "Enabling SAGV\n");
+	drm_dbg_kms(display->drm, "Enabling SAGV\n");
 	ret = snb_pcode_write(&i915->uncore, GEN9_PCODE_SAGV_CONTROL,
 			      GEN9_SAGV_ENABLE);
 
@@ -177,29 +175,30 @@ static void skl_sagv_enable(struct drm_i915_private *i915)
 	 * Some skl systems, pre-release machines in particular,
 	 * don't actually have SAGV.
 	 */
-	if (IS_SKYLAKE(i915) && ret == -ENXIO) {
-		drm_dbg(&i915->drm, "No SAGV found on system, ignoring\n");
-		i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+	if (display->platform.skylake && ret == -ENXIO) {
+		drm_dbg(display->drm, "No SAGV found on system, ignoring\n");
+		display->sagv.status = I915_SAGV_NOT_CONTROLLED;
 		return;
 	} else if (ret < 0) {
-		drm_err(&i915->drm, "Failed to enable SAGV\n");
+		drm_err(display->drm, "Failed to enable SAGV\n");
 		return;
 	}
 
-	i915->display.sagv.status = I915_SAGV_ENABLED;
+	display->sagv.status = I915_SAGV_ENABLED;
 }
 
-static void skl_sagv_disable(struct drm_i915_private *i915)
+static void skl_sagv_disable(struct intel_display *display)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	int ret;
 
-	if (!intel_has_sagv(i915))
+	if (!intel_has_sagv(display))
 		return;
 
-	if (i915->display.sagv.status == I915_SAGV_DISABLED)
+	if (display->sagv.status == I915_SAGV_DISABLED)
 		return;
 
-	drm_dbg_kms(&i915->drm, "Disabling SAGV\n");
+	drm_dbg_kms(display->drm, "Disabling SAGV\n");
 	/* bspec says to keep retrying for at least 1 ms */
 	ret = skl_pcode_request(&i915->uncore, GEN9_PCODE_SAGV_CONTROL,
 				GEN9_SAGV_DISABLE,
@@ -209,47 +208,47 @@ static void skl_sagv_disable(struct drm_i915_private *i915)
 	 * Some skl systems, pre-release machines in particular,
 	 * don't actually have SAGV.
 	 */
-	if (IS_SKYLAKE(i915) && ret == -ENXIO) {
-		drm_dbg(&i915->drm, "No SAGV found on system, ignoring\n");
-		i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
+	if (display->platform.skylake && ret == -ENXIO) {
+		drm_dbg(display->drm, "No SAGV found on system, ignoring\n");
+		display->sagv.status = I915_SAGV_NOT_CONTROLLED;
 		return;
 	} else if (ret < 0) {
-		drm_err(&i915->drm, "Failed to disable SAGV (%d)\n", ret);
+		drm_err(display->drm, "Failed to disable SAGV (%d)\n", ret);
 		return;
 	}
 
-	i915->display.sagv.status = I915_SAGV_DISABLED;
+	display->sagv.status = I915_SAGV_DISABLED;
 }
 
 static void skl_sagv_pre_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_bw_state *new_bw_state =
 		intel_atomic_get_new_bw_state(state);
 
 	if (!new_bw_state)
 		return;
 
-	if (!intel_can_enable_sagv(i915, new_bw_state))
-		skl_sagv_disable(i915);
+	if (!intel_can_enable_sagv(display, new_bw_state))
+		skl_sagv_disable(display);
 }
 
 static void skl_sagv_post_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_bw_state *new_bw_state =
 		intel_atomic_get_new_bw_state(state);
 
 	if (!new_bw_state)
 		return;
 
-	if (intel_can_enable_sagv(i915, new_bw_state))
-		skl_sagv_enable(i915);
+	if (intel_can_enable_sagv(display, new_bw_state))
+		skl_sagv_enable(display);
 }
 
 static void icl_sagv_pre_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_bw_state *old_bw_state =
 		intel_atomic_get_old_bw_state(state);
 	const struct intel_bw_state *new_bw_state =
@@ -267,7 +266,7 @@ static void icl_sagv_pre_plane_update(struct intel_atomic_state *state)
 
 	WARN_ON(!new_bw_state->base.changed);
 
-	drm_dbg_kms(&i915->drm, "Restricting QGV points: 0x%x -> 0x%x\n",
+	drm_dbg_kms(display->drm, "Restricting QGV points: 0x%x -> 0x%x\n",
 		    old_mask, new_mask);
 
 	/*
@@ -276,12 +275,12 @@ static void icl_sagv_pre_plane_update(struct intel_atomic_state *state)
 	 * time. Also masking should be done before updating the configuration
 	 * and unmasking afterwards.
 	 */
-	icl_pcode_restrict_qgv_points(i915, new_mask);
+	icl_pcode_restrict_qgv_points(display, new_mask);
 }
 
 static void icl_sagv_post_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_bw_state *old_bw_state =
 		intel_atomic_get_old_bw_state(state);
 	const struct intel_bw_state *new_bw_state =
@@ -299,7 +298,7 @@ static void icl_sagv_post_plane_update(struct intel_atomic_state *state)
 
 	WARN_ON(!new_bw_state->base.changed);
 
-	drm_dbg_kms(&i915->drm, "Relaxing QGV points: 0x%x -> 0x%x\n",
+	drm_dbg_kms(display->drm, "Relaxing QGV points: 0x%x -> 0x%x\n",
 		    old_mask, new_mask);
 
 	/*
@@ -308,12 +307,12 @@ static void icl_sagv_post_plane_update(struct intel_atomic_state *state)
 	 * time. Also masking should be done before updating the configuration
 	 * and unmasking afterwards.
 	 */
-	icl_pcode_restrict_qgv_points(i915, new_mask);
+	icl_pcode_restrict_qgv_points(display, new_mask);
 }
 
 void intel_sagv_pre_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 
 	/*
 	 * Just return if we can't control SAGV or don't have it.
@@ -322,10 +321,10 @@ void intel_sagv_pre_plane_update(struct intel_atomic_state *state)
 	 * disabled in a BIOS, we are not even allowed to send a PCode request,
 	 * as it will throw an error. So have to check it here.
 	 */
-	if (!intel_has_sagv(i915))
+	if (!intel_has_sagv(display))
 		return;
 
-	if (DISPLAY_VER(i915) >= 11)
+	if (DISPLAY_VER(display) >= 11)
 		icl_sagv_pre_plane_update(state);
 	else
 		skl_sagv_pre_plane_update(state);
@@ -333,7 +332,7 @@ void intel_sagv_pre_plane_update(struct intel_atomic_state *state)
 
 void intel_sagv_post_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 
 	/*
 	 * Just return if we can't control SAGV or don't have it.
@@ -342,10 +341,10 @@ void intel_sagv_post_plane_update(struct intel_atomic_state *state)
 	 * disabled in a BIOS, we are not even allowed to send a PCode request,
 	 * as it will throw an error. So have to check it here.
 	 */
-	if (!intel_has_sagv(i915))
+	if (!intel_has_sagv(display))
 		return;
 
-	if (DISPLAY_VER(i915) >= 11)
+	if (DISPLAY_VER(display) >= 11)
 		icl_sagv_post_plane_update(state);
 	else
 		skl_sagv_post_plane_update(state);
@@ -353,12 +352,12 @@ void intel_sagv_post_plane_update(struct intel_atomic_state *state)
 
 static bool skl_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum plane_id plane_id;
 	int max_level = INT_MAX;
 
-	if (!intel_has_sagv(i915))
+	if (!intel_has_sagv(display))
 		return false;
 
 	if (!crtc_state->hw.active)
@@ -377,7 +376,7 @@ static bool skl_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state)
 			continue;
 
 		/* Find the highest enabled wm level for this plane */
-		for (level = i915->display.wm.num_levels - 1;
+		for (level = display->wm.num_levels - 1;
 		     !wm->wm[level].enable; --level)
 		     { }
 
@@ -423,104 +422,37 @@ static bool tgl_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state)
 	return true;
 }
 
-static bool intel_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state)
+bool intel_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
-	if (!i915->display.params.enable_sagv)
+	if (!display->params.enable_sagv)
 		return false;
 
-	if (DISPLAY_VER(i915) >= 12)
+	/*
+	 * SAGV is initially forced off because its current
+	 * state can't be queried from pcode. Allow SAGV to
+	 * be enabled upon the first real commit.
+	 */
+	if (crtc_state->inherited)
+		return false;
+
+	if (DISPLAY_VER(display) >= 12)
 		return tgl_crtc_can_enable_sagv(crtc_state);
 	else
 		return skl_crtc_can_enable_sagv(crtc_state);
 }
 
-bool intel_can_enable_sagv(struct drm_i915_private *i915,
+bool intel_can_enable_sagv(struct intel_display *display,
 			   const struct intel_bw_state *bw_state)
 {
-	if (DISPLAY_VER(i915) < 11 &&
+	if (DISPLAY_VER(display) < 11 &&
 	    bw_state->active_pipes && !is_power_of_2(bw_state->active_pipes))
 		return false;
 
 	return bw_state->pipe_sagv_reject == 0;
 }
 
-static int intel_compute_sagv_mask(struct intel_atomic_state *state)
-{
-	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
-	int ret;
-	struct intel_crtc *crtc;
-	struct intel_crtc_state *new_crtc_state;
-	struct intel_bw_state *new_bw_state = NULL;
-	const struct intel_bw_state *old_bw_state = NULL;
-	int i;
-
-	for_each_new_intel_crtc_in_state(state, crtc,
-					 new_crtc_state, i) {
-		struct skl_pipe_wm *pipe_wm = &new_crtc_state->wm.skl.optimal;
-
-		new_bw_state = intel_atomic_get_bw_state(state);
-		if (IS_ERR(new_bw_state))
-			return PTR_ERR(new_bw_state);
-
-		old_bw_state = intel_atomic_get_old_bw_state(state);
-
-		/*
-		 * We store use_sagv_wm in the crtc state rather than relying on
-		 * that bw state since we have no convenient way to get at the
-		 * latter from the plane commit hooks (especially in the legacy
-		 * cursor case).
-		 *
-		 * drm_atomic_check_only() gets upset if we pull more crtcs
-		 * into the state, so we have to calculate this based on the
-		 * individual intel_crtc_can_enable_sagv() rather than
-		 * the overall intel_can_enable_sagv(). Otherwise the
-		 * crtcs not included in the commit would not switch to the
-		 * SAGV watermarks when we are about to enable SAGV, and that
-		 * would lead to underruns. This does mean extra power draw
-		 * when only a subset of the crtcs are blocking SAGV as the
-		 * other crtcs can't be allowed to use the more optimal
-		 * normal (ie. non-SAGV) watermarks.
-		 */
-		pipe_wm->use_sagv_wm = !HAS_HW_SAGV_WM(display) &&
-			DISPLAY_VER(i915) >= 12 &&
-			intel_crtc_can_enable_sagv(new_crtc_state);
-
-		if (intel_crtc_can_enable_sagv(new_crtc_state))
-			new_bw_state->pipe_sagv_reject &= ~BIT(crtc->pipe);
-		else
-			new_bw_state->pipe_sagv_reject |= BIT(crtc->pipe);
-	}
-
-	if (!new_bw_state)
-		return 0;
-
-	new_bw_state->active_pipes =
-		intel_calc_active_pipes(state, old_bw_state->active_pipes);
-
-	if (new_bw_state->active_pipes != old_bw_state->active_pipes) {
-		ret = intel_atomic_lock_global_state(&new_bw_state->base);
-		if (ret)
-			return ret;
-	}
-
-	if (intel_can_enable_sagv(i915, new_bw_state) !=
-	    intel_can_enable_sagv(i915, old_bw_state)) {
-		ret = intel_atomic_serialize_global_state(&new_bw_state->base);
-		if (ret)
-			return ret;
-	} else if (new_bw_state->pipe_sagv_reject != old_bw_state->pipe_sagv_reject) {
-		ret = intel_atomic_lock_global_state(&new_bw_state->base);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
 static u16 skl_ddb_entry_init(struct skl_ddb_entry *entry,
 			      u16 start, u16 end)
 {
@@ -530,17 +462,17 @@ static u16 skl_ddb_entry_init(struct skl_ddb_entry *entry,
 	return end;
 }
 
-static int intel_dbuf_slice_size(struct drm_i915_private *i915)
+static int intel_dbuf_slice_size(struct intel_display *display)
 {
-	return DISPLAY_INFO(i915)->dbuf.size /
-		hweight8(DISPLAY_INFO(i915)->dbuf.slice_mask);
+	return DISPLAY_INFO(display)->dbuf.size /
+		hweight8(DISPLAY_INFO(display)->dbuf.slice_mask);
 }
 
 static void
-skl_ddb_entry_for_slices(struct drm_i915_private *i915, u8 slice_mask,
+skl_ddb_entry_for_slices(struct intel_display *display, u8 slice_mask,
 			 struct skl_ddb_entry *ddb)
 {
-	int slice_size = intel_dbuf_slice_size(i915);
+	int slice_size = intel_dbuf_slice_size(display);
 
 	if (!slice_mask) {
 		ddb->start = 0;
@@ -552,10 +484,10 @@ skl_ddb_entry_for_slices(struct drm_i915_private *i915, u8 slice_mask,
 	ddb->end = fls(slice_mask) * slice_size;
 
 	WARN_ON(ddb->start >= ddb->end);
-	WARN_ON(ddb->end > DISPLAY_INFO(i915)->dbuf.size);
+	WARN_ON(ddb->end > DISPLAY_INFO(display)->dbuf.size);
 }
 
-static unsigned int mbus_ddb_offset(struct drm_i915_private *i915, u8 slice_mask)
+static unsigned int mbus_ddb_offset(struct intel_display *display, u8 slice_mask)
 {
 	struct skl_ddb_entry ddb;
 
@@ -564,15 +496,15 @@ static unsigned int mbus_ddb_offset(struct drm_i915_private *i915, u8 slice_mask
 	else if (slice_mask & (BIT(DBUF_S3) | BIT(DBUF_S4)))
 		slice_mask = BIT(DBUF_S3);
 
-	skl_ddb_entry_for_slices(i915, slice_mask, &ddb);
+	skl_ddb_entry_for_slices(display, slice_mask, &ddb);
 
 	return ddb.start;
 }
 
-u32 skl_ddb_dbuf_slice_mask(struct drm_i915_private *i915,
+u32 skl_ddb_dbuf_slice_mask(struct intel_display *display,
 			    const struct skl_ddb_entry *entry)
 {
-	int slice_size = intel_dbuf_slice_size(i915);
+	int slice_size = intel_dbuf_slice_size(display);
 	enum dbuf_slice start_slice, end_slice;
 	u8 slice_mask = 0;
 
@@ -618,15 +550,14 @@ static void intel_crtc_dbuf_weights(const struct intel_dbuf_state *dbuf_state,
 				    unsigned int *weight_end,
 				    unsigned int *weight_total)
 {
-	struct drm_i915_private *i915 =
-		to_i915(dbuf_state->base.state->base.dev);
+	struct intel_display *display = to_intel_display(dbuf_state->base.state->base.dev);
 	enum pipe pipe;
 
 	*weight_start = 0;
 	*weight_end = 0;
 	*weight_total = 0;
 
-	for_each_pipe(i915, pipe) {
+	for_each_pipe(display, pipe) {
 		int weight = dbuf_state->weight[pipe];
 
 		/*
@@ -652,7 +583,7 @@ static void intel_crtc_dbuf_weights(const struct intel_dbuf_state *dbuf_state,
 static int
 skl_crtc_allocate_ddb(struct intel_atomic_state *state, struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	unsigned int weight_total, weight_start, weight_end;
 	const struct intel_dbuf_state *old_dbuf_state =
 		intel_atomic_get_old_dbuf_state(state);
@@ -674,8 +605,8 @@ skl_crtc_allocate_ddb(struct intel_atomic_state *state, struct intel_crtc *crtc)
 
 	dbuf_slice_mask = new_dbuf_state->slices[pipe];
 
-	skl_ddb_entry_for_slices(i915, dbuf_slice_mask, &ddb_slices);
-	mbus_offset = mbus_ddb_offset(i915, dbuf_slice_mask);
+	skl_ddb_entry_for_slices(display, dbuf_slice_mask, &ddb_slices);
+	mbus_offset = mbus_ddb_offset(display, dbuf_slice_mask);
 	ddb_range_size = skl_ddb_entry_size(&ddb_slices);
 
 	intel_crtc_dbuf_weights(new_dbuf_state, pipe,
@@ -709,7 +640,7 @@ out:
 	crtc_state->wm.skl.ddb.start = mbus_offset + new_dbuf_state->ddb[pipe].start;
 	crtc_state->wm.skl.ddb.end = mbus_offset + new_dbuf_state->ddb[pipe].end;
 
-	drm_dbg_kms(&i915->drm,
+	drm_dbg_kms(display->drm,
 		    "[CRTC:%d:%s] dbuf slices 0x%x -> 0x%x, ddb (%d - %d) -> (%d - %d), active pipes 0x%x -> 0x%x\n",
 		    crtc->base.base.id, crtc->base.name,
 		    old_dbuf_state->slices[pipe], new_dbuf_state->slices[pipe],
@@ -734,10 +665,10 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 				 const struct skl_wm_level *result_prev,
 				 struct skl_wm_level *result /* out */);
 
-static unsigned int skl_wm_latency(struct drm_i915_private *i915, int level,
+static unsigned int skl_wm_latency(struct intel_display *display, int level,
 				   const struct skl_wm_params *wp)
 {
-	unsigned int latency = i915->display.wm.skl_latency[level];
+	unsigned int latency = display->wm.skl_latency[level];
 
 	if (latency == 0)
 		return 0;
@@ -746,11 +677,11 @@ static unsigned int skl_wm_latency(struct drm_i915_private *i915, int level,
 	 * WaIncreaseLatencyIPCEnabled: kbl,cfl
 	 * Display WA #1141: kbl,cfl
 	 */
-	if ((IS_KABYLAKE(i915) || IS_COFFEELAKE(i915) || IS_COMETLAKE(i915)) &&
-	    skl_watermark_ipc_enabled(i915))
+	if ((display->platform.kabylake || display->platform.coffeelake ||
+	     display->platform.cometlake) && skl_watermark_ipc_enabled(display))
 		latency += 4;
 
-	if (skl_needs_memory_bw_wa(i915) && wp && wp->x_tiled)
+	if (skl_needs_memory_bw_wa(display) && wp && wp->x_tiled)
 		latency += 15;
 
 	return latency;
@@ -760,8 +691,8 @@ static unsigned int
 skl_cursor_allocation(const struct intel_crtc_state *crtc_state,
 		      int num_active)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_plane *plane = to_intel_plane(crtc_state->uapi.crtc->cursor);
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 	struct skl_wm_level wm = {};
 	int ret, min_ddb_alloc = 0;
 	struct skl_wm_params wp;
@@ -772,10 +703,10 @@ skl_cursor_allocation(const struct intel_crtc_state *crtc_state,
 				    DRM_FORMAT_MOD_LINEAR,
 				    DRM_MODE_ROTATE_0,
 				    crtc_state->pixel_rate, &wp, 0, 0);
-	drm_WARN_ON(&i915->drm, ret);
+	drm_WARN_ON(display->drm, ret);
 
-	for (level = 0; level < i915->display.wm.num_levels; level++) {
-		unsigned int latency = skl_wm_latency(i915, level, &wp);
+	for (level = 0; level < display->wm.num_levels; level++) {
+		unsigned int latency = skl_wm_latency(display, level, &wp);
 
 		skl_compute_plane_wm(crtc_state, plane, level, latency, &wp, &wm, &wm);
 		if (wm.min_ddb_alloc == U16_MAX)
@@ -797,14 +728,13 @@ static void skl_ddb_entry_init_from_hw(struct skl_ddb_entry *entry, u32 reg)
 }
 
 static void
-skl_ddb_get_hw_plane_state(struct drm_i915_private *i915,
+skl_ddb_get_hw_plane_state(struct intel_display *display,
 			   const enum pipe pipe,
 			   const enum plane_id plane_id,
 			   struct skl_ddb_entry *ddb,
 			   struct skl_ddb_entry *ddb_y,
 			   u16 *min_ddb, u16 *interim_ddb)
 {
-	struct intel_display *display = &i915->display;
 	u32 val;
 
 	/* Cursor doesn't support NV12/planar, so no extra calculation needed */
@@ -837,7 +767,6 @@ static void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc,
 				      u16 *min_ddb, u16 *interim_ddb)
 {
 	struct intel_display *display = to_intel_display(crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum intel_display_power_domain power_domain;
 	enum pipe pipe = crtc->pipe;
 	intel_wakeref_t wakeref;
@@ -849,7 +778,7 @@ static void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc,
 		return;
 
 	for_each_plane_id_on_crtc(crtc, plane_id)
-		skl_ddb_get_hw_plane_state(i915, pipe,
+		skl_ddb_get_hw_plane_state(display, pipe,
 					   plane_id,
 					   &ddb[plane_id],
 					   &ddb_y[plane_id],
@@ -1367,16 +1296,16 @@ static u8 dg2_compute_dbuf_slices(enum pipe pipe, u8 active_pipes, bool join_mbu
 
 static u8 skl_compute_dbuf_slices(struct intel_crtc *crtc, u8 active_pipes, bool join_mbus)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	enum pipe pipe = crtc->pipe;
 
-	if (IS_DG2(i915))
+	if (display->platform.dg2)
 		return dg2_compute_dbuf_slices(pipe, active_pipes, join_mbus);
-	else if (DISPLAY_VER(i915) >= 13)
+	else if (DISPLAY_VER(display) >= 13)
 		return adlp_compute_dbuf_slices(pipe, active_pipes, join_mbus);
-	else if (DISPLAY_VER(i915) == 12)
+	else if (DISPLAY_VER(display) == 12)
 		return tgl_compute_dbuf_slices(pipe, active_pipes, join_mbus);
-	else if (DISPLAY_VER(i915) == 11)
+	else if (DISPLAY_VER(display) == 11)
 		return icl_compute_dbuf_slices(pipe, active_pipes, join_mbus);
 	/*
 	 * For anything else just return one slice yet.
@@ -1416,8 +1345,8 @@ skl_plane_relative_data_rate(const struct intel_crtc_state *crtc_state,
 static u64
 skl_total_relative_data_rate(const struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum plane_id plane_id;
 	u64 data_rate = 0;
 
@@ -1427,7 +1356,7 @@ skl_total_relative_data_rate(const struct intel_crtc_state *crtc_state)
 
 		data_rate += crtc_state->rel_data_rate[plane_id];
 
-		if (DISPLAY_VER(i915) < 11)
+		if (DISPLAY_VER(display) < 11)
 			data_rate += crtc_state->rel_data_rate_y[plane_id];
 	}
 
@@ -1489,7 +1418,7 @@ skl_check_nv12_wm_level(struct skl_wm_level *wm, struct skl_wm_level *uv_wm,
 	}
 }
 
-static bool skl_need_wm_copy_wa(struct drm_i915_private *i915, int level,
+static bool skl_need_wm_copy_wa(struct intel_display *display, int level,
 				const struct skl_plane_wm *wm)
 {
 	/*
@@ -1543,7 +1472,6 @@ static int
 skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 			    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_dbuf_state *dbuf_state =
@@ -1585,7 +1513,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 	 * Find the highest watermark level for which we can satisfy the block
 	 * requirement of active planes.
 	 */
-	for (level = i915->display.wm.num_levels - 1; level >= 0; level--) {
+	for (level = display->wm.num_levels - 1; level >= 0; level--) {
 		blocks = 0;
 		for_each_plane_id_on_crtc(crtc, plane_id) {
 			const struct skl_plane_wm *wm =
@@ -1596,7 +1524,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 					&crtc_state->wm.skl.plane_ddb[plane_id];
 
 				if (wm->wm[level].min_ddb_alloc > skl_ddb_entry_size(ddb)) {
-					drm_WARN_ON(&i915->drm,
+					drm_WARN_ON(display->drm,
 						    wm->wm[level].min_ddb_alloc != U16_MAX);
 					blocks = U32_MAX;
 					break;
@@ -1615,9 +1543,9 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 	}
 
 	if (level < 0) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Requested display configuration exceeds system DDB limitations");
-		drm_dbg_kms(&i915->drm, "minimum required %d/%d\n",
+		drm_dbg_kms(display->drm, "minimum required %d/%d\n",
 			    blocks, iter.size);
 		return -EINVAL;
 	}
@@ -1645,7 +1573,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 		if (plane_id == PLANE_CURSOR)
 			continue;
 
-		if (DISPLAY_VER(i915) < 11 &&
+		if (DISPLAY_VER(display) < 11 &&
 		    crtc_state->nv12_planes & BIT(plane_id)) {
 			skl_allocate_plane_ddb(&iter, ddb_y, &wm->wm[level],
 					       crtc_state->rel_data_rate_y[plane_id]);
@@ -1661,7 +1589,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 			*interim_ddb = wm->sagv.wm0.min_ddb_alloc;
 		}
 	}
-	drm_WARN_ON(&i915->drm, iter.size != 0 || iter.data_rate != 0);
+	drm_WARN_ON(display->drm, iter.size != 0 || iter.data_rate != 0);
 
 	/*
 	 * When we calculated watermark values we didn't know how high
@@ -1669,7 +1597,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 	 * all levels as "enabled."  Go back now and disable the ones
 	 * that aren't actually possible.
 	 */
-	for (level++; level < i915->display.wm.num_levels; level++) {
+	for (level++; level < display->wm.num_levels; level++) {
 		for_each_plane_id_on_crtc(crtc, plane_id) {
 			const struct skl_ddb_entry *ddb =
 				&crtc_state->wm.skl.plane_ddb[plane_id];
@@ -1678,7 +1606,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 			struct skl_plane_wm *wm =
 				&crtc_state->wm.skl.optimal.planes[plane_id];
 
-			if (DISPLAY_VER(i915) < 11 &&
+			if (DISPLAY_VER(display) < 11 &&
 			    crtc_state->nv12_planes & BIT(plane_id))
 				skl_check_nv12_wm_level(&wm->wm[level],
 							&wm->uv_wm[level],
@@ -1686,7 +1614,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 			else
 				skl_check_wm_level(&wm->wm[level], ddb);
 
-			if (skl_need_wm_copy_wa(i915, level, wm)) {
+			if (skl_need_wm_copy_wa(display, level, wm)) {
 				wm->wm[level].blocks = wm->wm[level - 1].blocks;
 				wm->wm[level].lines = wm->wm[level - 1].lines;
 				wm->wm[level].ignore_lines = wm->wm[level - 1].ignore_lines;
@@ -1708,7 +1636,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
 		struct skl_plane_wm *wm =
 			&crtc_state->wm.skl.optimal.planes[plane_id];
 
-		if (DISPLAY_VER(i915) < 11 &&
+		if (DISPLAY_VER(display) < 11 &&
 		    crtc_state->nv12_planes & BIT(plane_id)) {
 			skl_check_wm_level(&wm->trans_wm, ddb_y);
 		} else {
@@ -1734,7 +1662,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state,
  * 2xcdclk is 1350 MHz and the pixel rate should never exceed that.
  */
 static uint_fixed_16_16_t
-skl_wm_method1(const struct drm_i915_private *i915, u32 pixel_rate,
+skl_wm_method1(struct intel_display *display, u32 pixel_rate,
 	       u8 cpp, u32 latency, u32 dbuf_block_size)
 {
 	u32 wm_intermediate_val;
@@ -1746,7 +1674,7 @@ skl_wm_method1(const struct drm_i915_private *i915, u32 pixel_rate,
 	wm_intermediate_val = latency * pixel_rate * cpp;
 	ret = div_fixed16(wm_intermediate_val, 1000 * dbuf_block_size);
 
-	if (DISPLAY_VER(i915) >= 10)
+	if (DISPLAY_VER(display) >= 10)
 		ret = add_fixed16_u32(ret, 1);
 
 	return ret;
@@ -1772,7 +1700,7 @@ skl_wm_method2(u32 pixel_rate, u32 pipe_htotal, u32 latency,
 static uint_fixed_16_16_t
 intel_get_linetime_us(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	u32 pixel_rate;
 	u32 crtc_htotal;
 	uint_fixed_16_16_t linetime_us;
@@ -1782,7 +1710,7 @@ intel_get_linetime_us(const struct intel_crtc_state *crtc_state)
 
 	pixel_rate = crtc_state->pixel_rate;
 
-	if (drm_WARN_ON(&i915->drm, pixel_rate == 0))
+	if (drm_WARN_ON(display->drm, pixel_rate == 0))
 		return u32_to_fixed16(0);
 
 	crtc_htotal = crtc_state->hw.pipe_mode.crtc_htotal;
@@ -1798,15 +1726,13 @@ skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
 		      u32 plane_pixel_rate, struct skl_wm_params *wp,
 		      int color_plane, unsigned int pan_x)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct intel_display *display = to_intel_display(crtc_state);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	u32 interm_pbpl;
 
 	/* only planar format has two planes */
 	if (color_plane == 1 &&
 	    !intel_format_info_is_yuv_semiplanar(format, modifier)) {
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Non planar format have single plane\n");
 		return -EINVAL;
 	}
@@ -1824,7 +1750,7 @@ skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
 	wp->cpp = format->cpp[color_plane];
 	wp->plane_pixel_rate = plane_pixel_rate;
 
-	if (DISPLAY_VER(i915) >= 11 &&
+	if (DISPLAY_VER(display) >= 11 &&
 	    modifier == I915_FORMAT_MOD_Yf_TILED  && wp->cpp == 1)
 		wp->dbuf_block_size = 256;
 	else
@@ -1849,7 +1775,7 @@ skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
 		wp->y_min_scanlines = 4;
 	}
 
-	if (skl_needs_memory_bw_wa(i915))
+	if (skl_needs_memory_bw_wa(display))
 		wp->y_min_scanlines *= 2;
 
 	wp->plane_bytes_per_line = wp->width * wp->cpp;
@@ -1860,7 +1786,7 @@ skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
 
 		if (DISPLAY_VER(display) >= 30)
 			interm_pbpl += (pan_x != 0);
-		else if (DISPLAY_VER(i915) >= 10)
+		else if (DISPLAY_VER(display) >= 10)
 			interm_pbpl++;
 
 		wp->plane_blocks_per_line = div_fixed16(interm_pbpl,
@@ -1869,7 +1795,7 @@ skl_compute_wm_params(const struct intel_crtc_state *crtc_state,
 		interm_pbpl = DIV_ROUND_UP(wp->plane_bytes_per_line,
 					   wp->dbuf_block_size);
 
-		if (!wp->x_tiled || DISPLAY_VER(i915) >= 10)
+		if (!wp->x_tiled || DISPLAY_VER(display) >= 10)
 			interm_pbpl++;
 
 		wp->plane_blocks_per_line = u32_to_fixed16(interm_pbpl);
@@ -1906,18 +1832,18 @@ skl_compute_plane_wm_params(const struct intel_crtc_state *crtc_state,
 				     plane_state->uapi.src.x1);
 }
 
-static bool skl_wm_has_lines(struct drm_i915_private *i915, int level)
+static bool skl_wm_has_lines(struct intel_display *display, int level)
 {
-	if (DISPLAY_VER(i915) >= 10)
+	if (DISPLAY_VER(display) >= 10)
 		return true;
 
 	/* The number of lines are ignored for the level 0 watermark. */
 	return level > 0;
 }
 
-static int skl_wm_max_lines(struct drm_i915_private *i915)
+static int skl_wm_max_lines(struct intel_display *display)
 {
-	if (DISPLAY_VER(i915) >= 13)
+	if (DISPLAY_VER(display) >= 13)
 		return 255;
 	else
 		return 31;
@@ -1938,7 +1864,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 				 const struct skl_wm_level *result_prev,
 				 struct skl_wm_level *result /* out */)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	uint_fixed_16_16_t method1, method2;
 	uint_fixed_16_16_t selected_result;
 	u32 blocks, lines, min_ddb_alloc = 0;
@@ -1950,7 +1876,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 		return;
 	}
 
-	method1 = skl_wm_method1(i915, wp->plane_pixel_rate,
+	method1 = skl_wm_method1(display, wp->plane_pixel_rate,
 				 wp->cpp, latency, wp->dbuf_block_size);
 	method2 = skl_wm_method2(wp->plane_pixel_rate,
 				 crtc_state->hw.pipe_mode.crtc_htotal,
@@ -1965,7 +1891,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 		     (wp->plane_bytes_per_line / wp->dbuf_block_size < 1)) {
 			selected_result = method2;
 		} else if (latency >= wp->linetime_us) {
-			if (DISPLAY_VER(i915) == 9)
+			if (DISPLAY_VER(display) == 9)
 				selected_result = min_fixed16(method1, method2);
 			else
 				selected_result = method2;
@@ -1975,7 +1901,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 	}
 
 	blocks = fixed16_to_u32_round_up(selected_result);
-	if (DISPLAY_VER(i915) < 30)
+	if (DISPLAY_VER(display) < 30)
 		blocks++;
 
 	/*
@@ -1994,13 +1920,13 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 	 * channels' impact on the level 0 memory latency and the relevant
 	 * wm calculations.
 	 */
-	if (skl_wm_has_lines(i915, level))
+	if (skl_wm_has_lines(display, level))
 		blocks = max(blocks,
 			     fixed16_to_u32_round_up(wp->plane_blocks_per_line));
 	lines = div_round_up_fixed16(selected_result,
 				     wp->plane_blocks_per_line);
 
-	if (DISPLAY_VER(i915) == 9) {
+	if (DISPLAY_VER(display) == 9) {
 		/* Display WA #1125: skl,bxt,kbl */
 		if (level == 0 && wp->rc_surface)
 			blocks += fixed16_to_u32_round_up(wp->y_tile_minimum);
@@ -2025,7 +1951,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 		}
 	}
 
-	if (DISPLAY_VER(i915) >= 11) {
+	if (DISPLAY_VER(display) >= 11) {
 		if (wp->y_tiled) {
 			int extra_lines;
 
@@ -2042,10 +1968,10 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 		}
 	}
 
-	if (!skl_wm_has_lines(i915, level))
+	if (!skl_wm_has_lines(display, level))
 		lines = 0;
 
-	if (lines > skl_wm_max_lines(i915)) {
+	if (lines > skl_wm_max_lines(display)) {
 		/* reject it */
 		result->min_ddb_alloc = U16_MAX;
 		return;
@@ -2064,8 +1990,8 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state,
 	result->enable = true;
 	result->auto_min_alloc_wm_enable = xe3_auto_min_alloc_capable(plane, level);
 
-	if (DISPLAY_VER(i915) < 12 && i915->display.sagv.block_time_us)
-		result->can_sagv = latency >= i915->display.sagv.block_time_us;
+	if (DISPLAY_VER(display) < 12 && display->sagv.block_time_us)
+		result->can_sagv = latency >= display->sagv.block_time_us;
 }
 
 static void
@@ -2074,13 +2000,13 @@ skl_compute_wm_levels(const struct intel_crtc_state *crtc_state,
 		      const struct skl_wm_params *wm_params,
 		      struct skl_wm_level *levels)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct skl_wm_level *result_prev = &levels[0];
 	int level;
 
-	for (level = 0; level < i915->display.wm.num_levels; level++) {
+	for (level = 0; level < display->wm.num_levels; level++) {
 		struct skl_wm_level *result = &levels[level];
-		unsigned int latency = skl_wm_latency(i915, level, wm_params);
+		unsigned int latency = skl_wm_latency(display, level, wm_params);
 
 		skl_compute_plane_wm(crtc_state, plane, level, latency,
 				     wm_params, result_prev, result);
@@ -2094,21 +2020,21 @@ static void tgl_compute_sagv_wm(const struct intel_crtc_state *crtc_state,
 				const struct skl_wm_params *wm_params,
 				struct skl_plane_wm *plane_wm)
 {
-	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct skl_wm_level *sagv_wm = &plane_wm->sagv.wm0;
 	struct skl_wm_level *levels = plane_wm->wm;
 	unsigned int latency = 0;
 
-	if (i915->display.sagv.block_time_us)
-		latency = i915->display.sagv.block_time_us +
-			skl_wm_latency(i915, 0, wm_params);
+	if (display->sagv.block_time_us)
+		latency = display->sagv.block_time_us +
+			skl_wm_latency(display, 0, wm_params);
 
 	skl_compute_plane_wm(crtc_state, plane, 0, latency,
 			     wm_params, &levels[0],
 			     sagv_wm);
 }
 
-static void skl_compute_transition_wm(struct drm_i915_private *i915,
+static void skl_compute_transition_wm(struct intel_display *display,
 				      struct skl_wm_level *trans_wm,
 				      const struct skl_wm_level *wm0,
 				      const struct skl_wm_params *wp)
@@ -2117,23 +2043,23 @@ static void skl_compute_transition_wm(struct drm_i915_private *i915,
 	u16 wm0_blocks, trans_offset, blocks;
 
 	/* Transition WM don't make any sense if ipc is disabled */
-	if (!skl_watermark_ipc_enabled(i915))
+	if (!skl_watermark_ipc_enabled(display))
 		return;
 
 	/*
 	 * WaDisableTWM:skl,kbl,cfl,bxt
 	 * Transition WM are not recommended by HW team for GEN9
 	 */
-	if (DISPLAY_VER(i915) == 9)
+	if (DISPLAY_VER(display) == 9)
 		return;
 
-	if (DISPLAY_VER(i915) >= 11)
+	if (DISPLAY_VER(display) >= 11)
 		trans_min = 4;
 	else
 		trans_min = 14;
 
 	/* Display WA #1140: glk,cnl */
-	if (DISPLAY_VER(i915) == 10)
+	if (DISPLAY_VER(display) == 10)
 		trans_amount = 0;
 	else
 		trans_amount = 10; /* This is configurable amount */
@@ -2175,8 +2101,7 @@ static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state,
 				     const struct intel_plane_state *plane_state,
 				     struct intel_plane *plane, int color_plane)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct skl_plane_wm *wm = &crtc_state->wm.skl.raw.planes[plane->id];
 	struct skl_wm_params wm_params;
 	int ret;
@@ -2188,13 +2113,13 @@ static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state,
 
 	skl_compute_wm_levels(crtc_state, plane, &wm_params, wm->wm);
 
-	skl_compute_transition_wm(i915, &wm->trans_wm,
+	skl_compute_transition_wm(display, &wm->trans_wm,
 				  &wm->wm[0], &wm_params);
 
-	if (DISPLAY_VER(i915) >= 12) {
+	if (DISPLAY_VER(display) >= 12) {
 		tgl_compute_sagv_wm(crtc_state, plane, &wm_params, wm);
 
-		skl_compute_transition_wm(i915, &wm->sagv.trans_wm,
+		skl_compute_transition_wm(display, &wm->sagv.trans_wm,
 					  &wm->sagv.wm0, &wm_params);
 	}
 
@@ -2254,8 +2179,8 @@ static int skl_build_plane_wm(struct intel_crtc_state *crtc_state,
 static int icl_build_plane_wm(struct intel_crtc_state *crtc_state,
 			      const struct intel_plane_state *plane_state)
 {
+	struct intel_display *display = to_intel_display(plane_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-	struct drm_i915_private *i915 = to_i915(plane->base.dev);
 	enum plane_id plane_id = plane->id;
 	struct skl_plane_wm *wm = &crtc_state->wm.skl.raw.planes[plane_id];
 	int ret;
@@ -2269,9 +2194,9 @@ static int icl_build_plane_wm(struct intel_crtc_state *crtc_state,
 	if (plane_state->planar_linked_plane) {
 		const struct drm_framebuffer *fb = plane_state->hw.fb;
 
-		drm_WARN_ON(&i915->drm,
+		drm_WARN_ON(display->drm,
 			    !intel_wm_plane_visible(crtc_state, plane_state));
-		drm_WARN_ON(&i915->drm, !fb->format->is_yuv ||
+		drm_WARN_ON(display->drm, !fb->format->is_yuv ||
 			    fb->format->num_planes == 1);
 
 		ret = skl_build_plane_wm_single(crtc_state, plane_state,
@@ -2411,15 +2336,14 @@ static int skl_max_wm0_lines(const struct intel_crtc_state *crtc_state)
 static int skl_max_wm_level_for_vblank(struct intel_crtc_state *crtc_state,
 				       int wm0_lines)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 	int level;
 
-	for (level = i915->display.wm.num_levels - 1; level >= 0; level--) {
+	for (level = display->wm.num_levels - 1; level >= 0; level--) {
 		int latency;
 
 		/* FIXME should we care about the latency w/a's? */
-		latency = skl_wm_latency(i915, level, NULL);
+		latency = skl_wm_latency(display, level, NULL);
 		if (latency == 0)
 			continue;
 
@@ -2436,8 +2360,8 @@ static int skl_max_wm_level_for_vblank(struct intel_crtc_state *crtc_state,
 
 static int skl_wm_check_vblank(struct intel_crtc_state *crtc_state)
 {
+	struct intel_display *display = to_intel_display(crtc_state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	int wm0_lines, level;
 
 	if (!crtc_state->hw.active)
@@ -2453,9 +2377,9 @@ static int skl_wm_check_vblank(struct intel_crtc_state *crtc_state)
 	 * PSR needs to toggle LATENCY_REPORTING_REMOVED_PIPE_*
 	 * based on whether we're limited by the vblank duration.
 	 */
-	crtc_state->wm_level_disabled = level < i915->display.wm.num_levels - 1;
+	crtc_state->wm_level_disabled = level < display->wm.num_levels - 1;
 
-	for (level++; level < i915->display.wm.num_levels; level++) {
+	for (level++; level < display->wm.num_levels; level++) {
 		enum plane_id plane_id;
 
 		for_each_plane_id_on_crtc(crtc, plane_id) {
@@ -2471,10 +2395,10 @@ static int skl_wm_check_vblank(struct intel_crtc_state *crtc_state)
 		}
 	}
 
-	if (DISPLAY_VER(i915) >= 12 &&
-	    i915->display.sagv.block_time_us &&
+	if (DISPLAY_VER(display) >= 12 &&
+	    display->sagv.block_time_us &&
 	    skl_is_vblank_too_short(crtc_state, wm0_lines,
-				    i915->display.sagv.block_time_us)) {
+				    display->sagv.block_time_us)) {
 		enum plane_id plane_id;
 
 		for_each_plane_id_on_crtc(crtc, plane_id) {
@@ -2492,7 +2416,7 @@ static int skl_wm_check_vblank(struct intel_crtc_state *crtc_state)
 static int skl_build_pipe_wm(struct intel_atomic_state *state,
 			     struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	const struct intel_plane_state *plane_state;
@@ -2508,7 +2432,7 @@ static int skl_build_pipe_wm(struct intel_atomic_state *state,
 		if (plane->pipe != crtc->pipe)
 			continue;
 
-		if (DISPLAY_VER(i915) >= 11)
+		if (DISPLAY_VER(display) >= 11)
 			ret = icl_build_plane_wm(crtc_state, plane_state);
 		else
 			ret = skl_build_plane_wm(crtc_state, plane_state);
@@ -2531,11 +2455,10 @@ static bool skl_wm_level_equals(const struct skl_wm_level *l1,
 		l1->auto_min_alloc_wm_enable == l2->auto_min_alloc_wm_enable;
 }
 
-static bool skl_plane_wm_equals(struct drm_i915_private *i915,
+static bool skl_plane_wm_equals(struct intel_display *display,
 				const struct skl_plane_wm *wm1,
 				const struct skl_plane_wm *wm2)
 {
-	struct intel_display *display = &i915->display;
 	int level;
 
 	for (level = 0; level < display->wm.num_levels; level++) {
@@ -2590,14 +2513,14 @@ static int
 skl_ddb_add_affected_planes(struct intel_atomic_state *state,
 			    struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_crtc_state *old_crtc_state =
 		intel_atomic_get_old_crtc_state(state, crtc);
 	struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_plane *plane;
 
-	for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+	for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 		struct intel_plane_state *plane_state;
 		enum plane_id plane_id = plane->id;
 
@@ -2608,7 +2531,7 @@ skl_ddb_add_affected_planes(struct intel_atomic_state *state,
 			continue;
 
 		if (new_crtc_state->do_async_flip) {
-			drm_dbg_kms(&i915->drm, "[PLANE:%d:%s] Can't change DDB during async flip\n",
+			drm_dbg_kms(display->drm, "[PLANE:%d:%s] Can't change DDB during async flip\n",
 				    plane->base.base.id, plane->base.name);
 			return -EINVAL;
 		}
@@ -2627,7 +2550,7 @@ skl_ddb_add_affected_planes(struct intel_atomic_state *state,
 
 static u8 intel_dbuf_enabled_slices(const struct intel_dbuf_state *dbuf_state)
 {
-	struct drm_i915_private *i915 = to_i915(dbuf_state->base.state->base.dev);
+	struct intel_display *display = to_intel_display(dbuf_state->base.state->base.dev);
 	u8 enabled_slices;
 	enum pipe pipe;
 
@@ -2637,7 +2560,7 @@ static u8 intel_dbuf_enabled_slices(const struct intel_dbuf_state *dbuf_state)
 	 */
 	enabled_slices = BIT(DBUF_S1);
 
-	for_each_pipe(i915, pipe)
+	for_each_pipe(display, pipe)
 		enabled_slices |= dbuf_state->slices[pipe];
 
 	return enabled_slices;
@@ -2647,7 +2570,6 @@ static int
 skl_compute_ddb(struct intel_atomic_state *state)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	const struct intel_dbuf_state *old_dbuf_state;
 	struct intel_dbuf_state *new_dbuf_state = NULL;
 	struct intel_crtc_state *new_crtc_state;
@@ -2686,7 +2608,7 @@ skl_compute_ddb(struct intel_atomic_state *state)
 		}
 	}
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		enum pipe pipe = crtc->pipe;
 
 		new_dbuf_state->slices[pipe] =
@@ -2709,11 +2631,11 @@ skl_compute_ddb(struct intel_atomic_state *state)
 		if (ret)
 			return ret;
 
-		drm_dbg_kms(&i915->drm,
+		drm_dbg_kms(display->drm,
 			    "Enabled dbuf slices 0x%x -> 0x%x (total dbuf slices 0x%x), mbus joined? %s->%s\n",
 			    old_dbuf_state->enabled_slices,
 			    new_dbuf_state->enabled_slices,
-			    DISPLAY_INFO(i915)->dbuf.slice_mask,
+			    DISPLAY_INFO(display)->dbuf.slice_mask,
 			    str_yes_no(old_dbuf_state->joined_mbus),
 			    str_yes_no(new_dbuf_state->joined_mbus));
 	}
@@ -2731,7 +2653,7 @@ skl_compute_ddb(struct intel_atomic_state *state)
 			return ret;
 	}
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		ret = skl_crtc_allocate_ddb(state, crtc);
 		if (ret)
 			return ret;
@@ -2758,7 +2680,7 @@ static char enast(bool enable)
 static void
 skl_print_wm_changes(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_crtc_state *old_crtc_state;
 	const struct intel_crtc_state *new_crtc_state;
 	struct intel_plane *plane;
@@ -2775,7 +2697,7 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 		old_pipe_wm = &old_crtc_state->wm.skl.optimal;
 		new_pipe_wm = &new_crtc_state->wm.skl.optimal;
 
-		for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+		for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 			enum plane_id plane_id = plane->id;
 			const struct skl_ddb_entry *old, *new;
 
@@ -2785,24 +2707,24 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 			if (skl_ddb_entry_equal(old, new))
 				continue;
 
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s] ddb (%4d - %4d) -> (%4d - %4d), size %4d -> %4d\n",
 				    plane->base.base.id, plane->base.name,
 				    old->start, old->end, new->start, new->end,
 				    skl_ddb_entry_size(old), skl_ddb_entry_size(new));
 		}
 
-		for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+		for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 			enum plane_id plane_id = plane->id;
 			const struct skl_plane_wm *old_wm, *new_wm;
 
 			old_wm = &old_pipe_wm->planes[plane_id];
 			new_wm = &new_pipe_wm->planes[plane_id];
 
-			if (skl_plane_wm_equals(i915, old_wm, new_wm))
+			if (skl_plane_wm_equals(display, old_wm, new_wm))
 				continue;
 
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s]   level %cwm0,%cwm1,%cwm2,%cwm3,%cwm4,%cwm5,%cwm6,%cwm7,%ctwm,%cswm,%cstwm"
 				    " -> %cwm0,%cwm1,%cwm2,%cwm3,%cwm4,%cwm5,%cwm6,%cwm7,%ctwm,%cswm,%cstwm\n",
 				    plane->base.base.id, plane->base.name,
@@ -2821,7 +2743,7 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 				    enast(new_wm->sagv.wm0.enable),
 				    enast(new_wm->sagv.trans_wm.enable));
 
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s]   lines %c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%4d"
 				      " -> %c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%3d,%c%4d\n",
 				    plane->base.base.id, plane->base.name,
@@ -2848,7 +2770,7 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 				    enast(new_wm->sagv.wm0.ignore_lines), new_wm->sagv.wm0.lines,
 				    enast(new_wm->sagv.trans_wm.ignore_lines), new_wm->sagv.trans_wm.lines);
 
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s]  blocks %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%5d"
 				    " -> %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%5d\n",
 				    plane->base.base.id, plane->base.name,
@@ -2867,7 +2789,7 @@ skl_print_wm_changes(struct intel_atomic_state *state)
 				    new_wm->sagv.wm0.blocks,
 				    new_wm->sagv.trans_wm.blocks);
 
-			drm_dbg_kms(&i915->drm,
+			drm_dbg_kms(display->drm,
 				    "[PLANE:%d:%s] min_ddb %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%5d"
 				    " -> %4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%4d,%5d\n",
 				    plane->base.base.id, plane->base.name,
@@ -2945,14 +2867,14 @@ static bool skl_plane_selected_wm_equals(struct intel_plane *plane,
 static int skl_wm_add_affected_planes(struct intel_atomic_state *state,
 				      struct intel_crtc *crtc)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_crtc_state *old_crtc_state =
 		intel_atomic_get_old_crtc_state(state, crtc);
 	struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct intel_plane *plane;
 
-	for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+	for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 		struct intel_plane_state *plane_state;
 		enum plane_id plane_id = plane->id;
 
@@ -2971,7 +2893,7 @@ static int skl_wm_add_affected_planes(struct intel_atomic_state *state,
 			continue;
 
 		if (new_crtc_state->do_async_flip) {
-			drm_dbg_kms(&i915->drm, "[PLANE:%d:%s] Can't change watermarks during async flip\n",
+			drm_dbg_kms(display->drm, "[PLANE:%d:%s] Can't change watermarks during async flip\n",
 				    plane->base.base.id, plane->base.name);
 			return -EINVAL;
 		}
@@ -3002,7 +2924,6 @@ void
 intel_program_dpkgc_latency(struct intel_atomic_state *state)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	struct intel_crtc *crtc;
 	struct intel_crtc_state *new_crtc_state;
 	u32 latency = LNL_PKG_C_LATENCY_MASK;
@@ -3028,7 +2949,7 @@ intel_program_dpkgc_latency(struct intel_atomic_state *state)
 		added_wake_time = DSB_EXE_TIME +
 			display->sagv.block_time_us;
 
-		latency = skl_watermark_max_latency(i915, 1);
+		latency = skl_watermark_max_latency(display, 1);
 
 		/* Wa_22020432604 */
 		if ((DISPLAY_VER(display) == 20 || DISPLAY_VER(display) == 30) && !latency) {
@@ -3055,6 +2976,7 @@ intel_program_dpkgc_latency(struct intel_atomic_state *state)
 static int
 skl_compute_wm(struct intel_atomic_state *state)
 {
+	struct intel_display *display = to_intel_display(state);
 	struct intel_crtc *crtc;
 	struct intel_crtc_state __maybe_unused *new_crtc_state;
 	int ret, i;
@@ -3069,16 +2991,35 @@ skl_compute_wm(struct intel_atomic_state *state)
 	if (ret)
 		return ret;
 
-	ret = intel_compute_sagv_mask(state);
-	if (ret)
-		return ret;
-
 	/*
 	 * skl_compute_ddb() will have adjusted the final watermarks
 	 * based on how much ddb is available. Now we can actually
 	 * check if the final watermarks changed.
 	 */
 	for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) {
+		struct skl_pipe_wm *pipe_wm = &new_crtc_state->wm.skl.optimal;
+
+		/*
+		 * We store use_sagv_wm in the crtc state rather than relying on
+		 * that bw state since we have no convenient way to get at the
+		 * latter from the plane commit hooks (especially in the legacy
+		 * cursor case).
+		 *
+		 * drm_atomic_check_only() gets upset if we pull more crtcs
+		 * into the state, so we have to calculate this based on the
+		 * individual intel_crtc_can_enable_sagv() rather than
+		 * the overall intel_can_enable_sagv(). Otherwise the
+		 * crtcs not included in the commit would not switch to the
+		 * SAGV watermarks when we are about to enable SAGV, and that
+		 * would lead to underruns. This does mean extra power draw
+		 * when only a subset of the crtcs are blocking SAGV as the
+		 * other crtcs can't be allowed to use the more optimal
+		 * normal (ie. non-SAGV) watermarks.
+		 */
+		pipe_wm->use_sagv_wm = !HAS_HW_SAGV_WM(display) &&
+			DISPLAY_VER(display) >= 12 &&
+			intel_crtc_can_enable_sagv(new_crtc_state);
+
 		ret = skl_wm_add_affected_planes(state, crtc);
 		if (ret)
 			return ret;
@@ -3149,11 +3090,10 @@ static void skl_pipe_wm_get_hw_state(struct intel_crtc *crtc,
 	}
 }
 
-static void skl_wm_get_hw_state(struct drm_i915_private *i915)
+static void skl_wm_get_hw_state(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_dbuf_state *dbuf_state =
-		to_intel_dbuf_state(i915->display.dbuf.obj.state);
+		to_intel_dbuf_state(display->dbuf.obj.state);
 	struct intel_crtc *crtc;
 
 	if (HAS_MBUS_JOINING(display))
@@ -3193,7 +3133,7 @@ static void skl_wm_get_hw_state(struct drm_i915_private *i915)
 			if (!crtc_state->hw.active)
 				continue;
 
-			skl_ddb_get_hw_plane_state(i915, crtc->pipe,
+			skl_ddb_get_hw_plane_state(display, crtc->pipe,
 						   plane_id, ddb, ddb_y,
 						   min_ddb, interim_ddb);
 
@@ -3209,13 +3149,13 @@ static void skl_wm_get_hw_state(struct drm_i915_private *i915)
 		 */
 		slices = skl_compute_dbuf_slices(crtc, dbuf_state->active_pipes,
 						 dbuf_state->joined_mbus);
-		mbus_offset = mbus_ddb_offset(i915, slices);
+		mbus_offset = mbus_ddb_offset(display, slices);
 		crtc_state->wm.skl.ddb.start = mbus_offset + dbuf_state->ddb[pipe].start;
 		crtc_state->wm.skl.ddb.end = mbus_offset + dbuf_state->ddb[pipe].end;
 
 		/* The slices actually used by the planes on the pipe */
 		dbuf_state->slices[pipe] =
-			skl_ddb_dbuf_slice_mask(i915, &crtc_state->wm.skl.ddb);
+			skl_ddb_dbuf_slice_mask(display, &crtc_state->wm.skl.ddb);
 
 		drm_dbg_kms(display->drm,
 			    "[CRTC:%d:%s] dbuf slices 0x%x, ddb (%d - %d), active pipes 0x%x, mbus joined: %s\n",
@@ -3228,49 +3168,52 @@ static void skl_wm_get_hw_state(struct drm_i915_private *i915)
 	dbuf_state->enabled_slices = display->dbuf.enabled_slices;
 }
 
-bool skl_watermark_ipc_enabled(struct drm_i915_private *i915)
+bool skl_watermark_ipc_enabled(struct intel_display *display)
 {
-	return i915->display.wm.ipc_enabled;
+	return display->wm.ipc_enabled;
 }
 
-void skl_watermark_ipc_update(struct drm_i915_private *i915)
+void skl_watermark_ipc_update(struct intel_display *display)
 {
-	if (!HAS_IPC(i915))
+	if (!HAS_IPC(display))
 		return;
 
-	intel_de_rmw(i915, DISP_ARB_CTL2, DISP_IPC_ENABLE,
-		     skl_watermark_ipc_enabled(i915) ? DISP_IPC_ENABLE : 0);
+	intel_de_rmw(display, DISP_ARB_CTL2, DISP_IPC_ENABLE,
+		     skl_watermark_ipc_enabled(display) ? DISP_IPC_ENABLE : 0);
 }
 
-static bool skl_watermark_ipc_can_enable(struct drm_i915_private *i915)
+static bool skl_watermark_ipc_can_enable(struct intel_display *display)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
+
 	/* Display WA #0477 WaDisableIPC: skl */
-	if (IS_SKYLAKE(i915))
+	if (display->platform.skylake)
 		return false;
 
 	/* Display WA #1141: SKL:all KBL:all CFL */
-	if (IS_KABYLAKE(i915) ||
-	    IS_COFFEELAKE(i915) ||
-	    IS_COMETLAKE(i915))
+	if (display->platform.kabylake ||
+	    display->platform.coffeelake ||
+	    display->platform.cometlake)
 		return i915->dram_info.symmetric_memory;
 
 	return true;
 }
 
-void skl_watermark_ipc_init(struct drm_i915_private *i915)
+void skl_watermark_ipc_init(struct intel_display *display)
 {
-	if (!HAS_IPC(i915))
+	if (!HAS_IPC(display))
 		return;
 
-	i915->display.wm.ipc_enabled = skl_watermark_ipc_can_enable(i915);
+	display->wm.ipc_enabled = skl_watermark_ipc_can_enable(display);
 
-	skl_watermark_ipc_update(i915);
+	skl_watermark_ipc_update(display);
 }
 
 static void
-adjust_wm_latency(struct drm_i915_private *i915,
+adjust_wm_latency(struct intel_display *display,
 		  u16 wm[], int num_levels, int read_latency)
 {
+	struct drm_i915_private *i915 = to_i915(display->drm);
 	bool wm_lv_0_adjust_needed = i915->dram_info.wm_lv_0_adjust_needed;
 	int i, level;
 
@@ -3311,31 +3254,32 @@ adjust_wm_latency(struct drm_i915_private *i915,
 		wm[0] += 1;
 }
 
-static void mtl_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
+static void mtl_read_wm_latency(struct intel_display *display, u16 wm[])
 {
-	int num_levels = i915->display.wm.num_levels;
+	int num_levels = display->wm.num_levels;
 	u32 val;
 
-	val = intel_de_read(i915, MTL_LATENCY_LP0_LP1);
+	val = intel_de_read(display, MTL_LATENCY_LP0_LP1);
 	wm[0] = REG_FIELD_GET(MTL_LATENCY_LEVEL_EVEN_MASK, val);
 	wm[1] = REG_FIELD_GET(MTL_LATENCY_LEVEL_ODD_MASK, val);
 
-	val = intel_de_read(i915, MTL_LATENCY_LP2_LP3);
+	val = intel_de_read(display, MTL_LATENCY_LP2_LP3);
 	wm[2] = REG_FIELD_GET(MTL_LATENCY_LEVEL_EVEN_MASK, val);
 	wm[3] = REG_FIELD_GET(MTL_LATENCY_LEVEL_ODD_MASK, val);
 
-	val = intel_de_read(i915, MTL_LATENCY_LP4_LP5);
+	val = intel_de_read(display, MTL_LATENCY_LP4_LP5);
 	wm[4] = REG_FIELD_GET(MTL_LATENCY_LEVEL_EVEN_MASK, val);
 	wm[5] = REG_FIELD_GET(MTL_LATENCY_LEVEL_ODD_MASK, val);
 
-	adjust_wm_latency(i915, wm, num_levels, 6);
+	adjust_wm_latency(display, wm, num_levels, 6);
 }
 
-static void skl_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
+static void skl_read_wm_latency(struct intel_display *display, u16 wm[])
 {
-	int num_levels = i915->display.wm.num_levels;
-	int read_latency = DISPLAY_VER(i915) >= 12 ? 3 : 2;
-	int mult = IS_DG2(i915) ? 2 : 1;
+	struct drm_i915_private *i915 = to_i915(display->drm);
+	int num_levels = display->wm.num_levels;
+	int read_latency = DISPLAY_VER(display) >= 12 ? 3 : 2;
+	int mult = display->platform.dg2 ? 2 : 1;
 	u32 val;
 	int ret;
 
@@ -3343,7 +3287,7 @@ static void skl_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
 	val = 0; /* data0 to be programmed to 0 for first set */
 	ret = snb_pcode_read(&i915->uncore, GEN9_PCODE_READ_MEM_LATENCY, &val, NULL);
 	if (ret) {
-		drm_err(&i915->drm, "SKL Mailbox read error = %d\n", ret);
+		drm_err(display->drm, "SKL Mailbox read error = %d\n", ret);
 		return;
 	}
 
@@ -3356,7 +3300,7 @@ static void skl_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
 	val = 1; /* data0 to be programmed to 1 for second set */
 	ret = snb_pcode_read(&i915->uncore, GEN9_PCODE_READ_MEM_LATENCY, &val, NULL);
 	if (ret) {
-		drm_err(&i915->drm, "SKL Mailbox read error = %d\n", ret);
+		drm_err(display->drm, "SKL Mailbox read error = %d\n", ret);
 		return;
 	}
 
@@ -3365,24 +3309,22 @@ static void skl_read_wm_latency(struct drm_i915_private *i915, u16 wm[])
 	wm[6] = REG_FIELD_GET(GEN9_MEM_LATENCY_LEVEL_2_6_MASK, val) * mult;
 	wm[7] = REG_FIELD_GET(GEN9_MEM_LATENCY_LEVEL_3_7_MASK, val) * mult;
 
-	adjust_wm_latency(i915, wm, num_levels, read_latency);
+	adjust_wm_latency(display, wm, num_levels, read_latency);
 }
 
-static void skl_setup_wm_latency(struct drm_i915_private *i915)
+static void skl_setup_wm_latency(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
-
 	if (HAS_HW_SAGV_WM(display))
 		display->wm.num_levels = 6;
 	else
 		display->wm.num_levels = 8;
 
 	if (DISPLAY_VER(display) >= 14)
-		mtl_read_wm_latency(i915, display->wm.skl_latency);
+		mtl_read_wm_latency(display, display->wm.skl_latency);
 	else
-		skl_read_wm_latency(i915, display->wm.skl_latency);
+		skl_read_wm_latency(display, display->wm.skl_latency);
 
-	intel_print_wm_latency(i915, "Gen9 Plane", display->wm.skl_latency);
+	intel_print_wm_latency(display, "Gen9 Plane", display->wm.skl_latency);
 }
 
 static struct intel_global_state *intel_dbuf_duplicate_state(struct intel_global_obj *obj)
@@ -3410,19 +3352,18 @@ static const struct intel_global_state_funcs intel_dbuf_funcs = {
 struct intel_dbuf_state *
 intel_atomic_get_dbuf_state(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	struct intel_global_state *dbuf_state;
 
-	dbuf_state = intel_atomic_get_global_obj_state(state, &i915->display.dbuf.obj);
+	dbuf_state = intel_atomic_get_global_obj_state(state, &display->dbuf.obj);
 	if (IS_ERR(dbuf_state))
 		return ERR_CAST(dbuf_state);
 
 	return to_intel_dbuf_state(dbuf_state);
 }
 
-int intel_dbuf_init(struct drm_i915_private *i915)
+int intel_dbuf_init(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_dbuf_state *dbuf_state;
 
 	dbuf_state = kzalloc(sizeof(*dbuf_state), GFP_KERNEL);
@@ -3457,34 +3398,34 @@ static bool xelpdp_is_only_pipe_per_dbuf_bank(enum pipe pipe, u8 active_pipes)
 static u32 pipe_mbus_dbox_ctl(const struct intel_crtc *crtc,
 			      const struct intel_dbuf_state *dbuf_state)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_display *display = to_intel_display(crtc);
 	u32 val = 0;
 
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		val |= MBUS_DBOX_I_CREDIT(2);
 
-	if (DISPLAY_VER(i915) >= 12) {
+	if (DISPLAY_VER(display) >= 12) {
 		val |= MBUS_DBOX_B2B_TRANSACTIONS_MAX(16);
 		val |= MBUS_DBOX_B2B_TRANSACTIONS_DELAY(1);
 		val |= MBUS_DBOX_REGULATE_B2B_TRANSACTIONS_EN;
 	}
 
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		val |= dbuf_state->joined_mbus ?
 			MBUS_DBOX_A_CREDIT(12) : MBUS_DBOX_A_CREDIT(8);
-	else if (IS_ALDERLAKE_P(i915))
+	else if (display->platform.alderlake_p)
 		/* Wa_22010947358:adl-p */
 		val |= dbuf_state->joined_mbus ?
 			MBUS_DBOX_A_CREDIT(6) : MBUS_DBOX_A_CREDIT(4);
 	else
 		val |= MBUS_DBOX_A_CREDIT(2);
 
-	if (DISPLAY_VER(i915) >= 14) {
+	if (DISPLAY_VER(display) >= 14) {
 		val |= MBUS_DBOX_B_CREDIT(0xA);
-	} else if (IS_ALDERLAKE_P(i915)) {
+	} else if (display->platform.alderlake_p) {
 		val |= MBUS_DBOX_BW_CREDIT(2);
 		val |= MBUS_DBOX_B_CREDIT(8);
-	} else if (DISPLAY_VER(i915) >= 12) {
+	} else if (DISPLAY_VER(display) >= 12) {
 		val |= MBUS_DBOX_BW_CREDIT(2);
 		val |= MBUS_DBOX_B_CREDIT(12);
 	} else {
@@ -3492,7 +3433,7 @@ static u32 pipe_mbus_dbox_ctl(const struct intel_crtc *crtc,
 		val |= MBUS_DBOX_B_CREDIT(8);
 	}
 
-	if (DISPLAY_VERx100(i915) == 1400) {
+	if (DISPLAY_VERx100(display) == 1400) {
 		if (xelpdp_is_only_pipe_per_dbuf_bank(crtc->pipe, dbuf_state->active_pipes))
 			val |= MBUS_DBOX_BW_8CREDITS_MTL;
 		else
@@ -3502,22 +3443,22 @@ static u32 pipe_mbus_dbox_ctl(const struct intel_crtc *crtc,
 	return val;
 }
 
-static void pipe_mbus_dbox_ctl_update(struct drm_i915_private *i915,
+static void pipe_mbus_dbox_ctl_update(struct intel_display *display,
 				      const struct intel_dbuf_state *dbuf_state)
 {
 	struct intel_crtc *crtc;
 
-	for_each_intel_crtc_in_pipe_mask(&i915->drm, crtc, dbuf_state->active_pipes)
-		intel_de_write(i915, PIPE_MBUS_DBOX_CTL(crtc->pipe),
+	for_each_intel_crtc_in_pipe_mask(display->drm, crtc, dbuf_state->active_pipes)
+		intel_de_write(display, PIPE_MBUS_DBOX_CTL(crtc->pipe),
 			       pipe_mbus_dbox_ctl(crtc, dbuf_state));
 }
 
 static void intel_mbus_dbox_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_dbuf_state *new_dbuf_state, *old_dbuf_state;
 
-	if (DISPLAY_VER(i915) < 11)
+	if (DISPLAY_VER(display) < 11)
 		return;
 
 	new_dbuf_state = intel_atomic_get_new_dbuf_state(state);
@@ -3527,7 +3468,7 @@ static void intel_mbus_dbox_update(struct intel_atomic_state *state)
 	     new_dbuf_state->active_pipes == old_dbuf_state->active_pipes))
 		return;
 
-	pipe_mbus_dbox_ctl_update(i915, new_dbuf_state);
+	pipe_mbus_dbox_ctl_update(display, new_dbuf_state);
 }
 
 int intel_dbuf_state_set_mdclk_cdclk_ratio(struct intel_atomic_state *state,
@@ -3544,10 +3485,9 @@ int intel_dbuf_state_set_mdclk_cdclk_ratio(struct intel_atomic_state *state,
 	return intel_atomic_lock_global_state(&dbuf_state->base);
 }
 
-void intel_dbuf_mdclk_cdclk_ratio_update(struct drm_i915_private *i915,
+void intel_dbuf_mdclk_cdclk_ratio_update(struct intel_display *display,
 					 int ratio, bool joined_mbus)
 {
-	struct intel_display *display = &i915->display;
 	enum dbuf_slice slice;
 
 	if (!HAS_MBUS_JOINING(display))
@@ -3571,7 +3511,7 @@ void intel_dbuf_mdclk_cdclk_ratio_update(struct drm_i915_private *i915,
 
 static void intel_dbuf_mdclk_min_tracker_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_dbuf_state *old_dbuf_state =
 		intel_atomic_get_old_dbuf_state(state);
 	const struct intel_dbuf_state *new_dbuf_state =
@@ -3586,7 +3526,7 @@ static void intel_dbuf_mdclk_min_tracker_update(struct intel_atomic_state *state
 		mdclk_cdclk_ratio = new_dbuf_state->mdclk_cdclk_ratio;
 	}
 
-	intel_dbuf_mdclk_cdclk_ratio_update(i915, mdclk_cdclk_ratio,
+	intel_dbuf_mdclk_cdclk_ratio_update(display, mdclk_cdclk_ratio,
 					    new_dbuf_state->joined_mbus);
 }
 
@@ -3594,13 +3534,12 @@ static enum pipe intel_mbus_joined_pipe(struct intel_atomic_state *state,
 					const struct intel_dbuf_state *dbuf_state)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	enum pipe pipe = ffs(dbuf_state->active_pipes) - 1;
 	const struct intel_crtc_state *new_crtc_state;
 	struct intel_crtc *crtc;
 
-	drm_WARN_ON(&i915->drm, !dbuf_state->joined_mbus);
-	drm_WARN_ON(&i915->drm, !is_power_of_2(dbuf_state->active_pipes));
+	drm_WARN_ON(display->drm, !dbuf_state->joined_mbus);
+	drm_WARN_ON(display->drm, !is_power_of_2(dbuf_state->active_pipes));
 
 	crtc = intel_crtc_for_pipe(display, pipe);
 	new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc);
@@ -3611,7 +3550,7 @@ static enum pipe intel_mbus_joined_pipe(struct intel_atomic_state *state,
 		return INVALID_PIPE;
 }
 
-static void mbus_ctl_join_update(struct drm_i915_private *i915,
+static void mbus_ctl_join_update(struct intel_display *display,
 				 const struct intel_dbuf_state *dbuf_state,
 				 enum pipe pipe)
 {
@@ -3627,7 +3566,7 @@ static void mbus_ctl_join_update(struct drm_i915_private *i915,
 	else
 		mbus_ctl |= MBUS_JOIN_PIPE_SELECT_NONE;
 
-	intel_de_rmw(i915, MBUS_CTL,
+	intel_de_rmw(display, MBUS_CTL,
 		     MBUS_HASHING_MODE_MASK | MBUS_JOIN |
 		     MBUS_JOIN_PIPE_SELECT_MASK, mbus_ctl);
 }
@@ -3635,18 +3574,18 @@ static void mbus_ctl_join_update(struct drm_i915_private *i915,
 static void intel_dbuf_mbus_join_update(struct intel_atomic_state *state,
 					enum pipe pipe)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_display *display = to_intel_display(state);
 	const struct intel_dbuf_state *old_dbuf_state =
 		intel_atomic_get_old_dbuf_state(state);
 	const struct intel_dbuf_state *new_dbuf_state =
 		intel_atomic_get_new_dbuf_state(state);
 
-	drm_dbg_kms(&i915->drm, "Changing mbus joined: %s -> %s (pipe: %c)\n",
+	drm_dbg_kms(display->drm, "Changing mbus joined: %s -> %s (pipe: %c)\n",
 		    str_yes_no(old_dbuf_state->joined_mbus),
 		    str_yes_no(new_dbuf_state->joined_mbus),
 		    pipe != INVALID_PIPE ? pipe_name(pipe) : '*');
 
-	mbus_ctl_join_update(i915, new_dbuf_state, pipe);
+	mbus_ctl_join_update(display, new_dbuf_state, pipe);
 }
 
 void intel_dbuf_mbus_pre_ddb_update(struct intel_atomic_state *state)
@@ -3751,9 +3690,8 @@ void intel_dbuf_post_plane_update(struct intel_atomic_state *state)
 	gen9_dbuf_slices_update(display, new_slices);
 }
 
-static void skl_mbus_sanitize(struct drm_i915_private *i915)
+static void skl_mbus_sanitize(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct intel_dbuf_state *dbuf_state =
 		to_intel_dbuf_state(display->dbuf.obj.state);
 
@@ -3768,28 +3706,28 @@ static void skl_mbus_sanitize(struct drm_i915_private *i915)
 		    dbuf_state->active_pipes);
 
 	dbuf_state->joined_mbus = false;
-	intel_dbuf_mdclk_cdclk_ratio_update(i915,
+	intel_dbuf_mdclk_cdclk_ratio_update(display,
 					    dbuf_state->mdclk_cdclk_ratio,
 					    dbuf_state->joined_mbus);
-	pipe_mbus_dbox_ctl_update(i915, dbuf_state);
-	mbus_ctl_join_update(i915, dbuf_state, INVALID_PIPE);
+	pipe_mbus_dbox_ctl_update(display, dbuf_state);
+	mbus_ctl_join_update(display, dbuf_state, INVALID_PIPE);
 }
 
-static bool skl_dbuf_is_misconfigured(struct drm_i915_private *i915)
+static bool skl_dbuf_is_misconfigured(struct intel_display *display)
 {
 	const struct intel_dbuf_state *dbuf_state =
-		to_intel_dbuf_state(i915->display.dbuf.obj.state);
+		to_intel_dbuf_state(display->dbuf.obj.state);
 	struct skl_ddb_entry entries[I915_MAX_PIPES] = {};
 	struct intel_crtc *crtc;
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 
 		entries[crtc->pipe] = crtc_state->wm.skl.ddb;
 	}
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		const struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
 		u8 slices;
@@ -3807,7 +3745,7 @@ static bool skl_dbuf_is_misconfigured(struct drm_i915_private *i915)
 	return false;
 }
 
-static void skl_dbuf_sanitize(struct drm_i915_private *i915)
+static void skl_dbuf_sanitize(struct intel_display *display)
 {
 	struct intel_crtc *crtc;
 
@@ -3822,12 +3760,12 @@ static void skl_dbuf_sanitize(struct drm_i915_private *i915)
 	 * all the planes so that skl_commit_modeset_enables() can
 	 * simply ignore them.
 	 */
-	if (!skl_dbuf_is_misconfigured(i915))
+	if (!skl_dbuf_is_misconfigured(display))
 		return;
 
-	drm_dbg_kms(&i915->drm, "BIOS has misprogrammed the DBUF, disabling all planes\n");
+	drm_dbg_kms(display->drm, "BIOS has misprogrammed the DBUF, disabling all planes\n");
 
-	for_each_intel_crtc(&i915->drm, crtc) {
+	for_each_intel_crtc(display->drm, crtc) {
 		struct intel_plane *plane = to_intel_plane(crtc->base.primary);
 		const struct intel_plane_state *plane_state =
 			to_intel_plane_state(plane->base.state);
@@ -3837,16 +3775,16 @@ static void skl_dbuf_sanitize(struct drm_i915_private *i915)
 		if (plane_state->uapi.visible)
 			intel_plane_disable_noatomic(crtc, plane);
 
-		drm_WARN_ON(&i915->drm, crtc_state->active_planes != 0);
+		drm_WARN_ON(display->drm, crtc_state->active_planes != 0);
 
 		memset(&crtc_state->wm.skl.ddb, 0, sizeof(crtc_state->wm.skl.ddb));
 	}
 }
 
-static void skl_wm_sanitize(struct drm_i915_private *i915)
+static void skl_wm_sanitize(struct intel_display *display)
 {
-	skl_mbus_sanitize(i915);
-	skl_dbuf_sanitize(i915);
+	skl_mbus_sanitize(display);
+	skl_dbuf_sanitize(display);
 }
 
 void skl_wm_crtc_disable_noatomic(struct intel_crtc *crtc)
@@ -3897,7 +3835,6 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 			   struct intel_crtc *crtc)
 {
 	struct intel_display *display = to_intel_display(state);
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	const struct intel_crtc_state *new_crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	struct skl_hw_state {
@@ -3912,7 +3849,7 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 	u8 hw_enabled_slices;
 	int level;
 
-	if (DISPLAY_VER(i915) < 9 || !new_crtc_state->hw.active)
+	if (DISPLAY_VER(display) < 9 || !new_crtc_state->hw.active)
 		return;
 
 	hw = kzalloc(sizeof(*hw), GFP_KERNEL);
@@ -3925,26 +3862,26 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 
 	hw_enabled_slices = intel_enabled_dbuf_slices_mask(display);
 
-	if (DISPLAY_VER(i915) >= 11 &&
-	    hw_enabled_slices != i915->display.dbuf.enabled_slices)
-		drm_err(&i915->drm,
+	if (DISPLAY_VER(display) >= 11 &&
+	    hw_enabled_slices != display->dbuf.enabled_slices)
+		drm_err(display->drm,
 			"mismatch in DBUF Slices (expected 0x%x, got 0x%x)\n",
-			i915->display.dbuf.enabled_slices,
+			display->dbuf.enabled_slices,
 			hw_enabled_slices);
 
-	for_each_intel_plane_on_crtc(&i915->drm, crtc, plane) {
+	for_each_intel_plane_on_crtc(display->drm, crtc, plane) {
 		const struct skl_ddb_entry *hw_ddb_entry, *sw_ddb_entry;
 		const struct skl_wm_level *hw_wm_level, *sw_wm_level;
 
 		/* Watermarks */
-		for (level = 0; level < i915->display.wm.num_levels; level++) {
+		for (level = 0; level < display->wm.num_levels; level++) {
 			hw_wm_level = &hw->wm.planes[plane->id].wm[level];
 			sw_wm_level = skl_plane_wm_level(sw_wm, plane->id, level);
 
 			if (skl_wm_level_equals(hw_wm_level, sw_wm_level))
 				continue;
 
-			drm_err(&i915->drm,
+			drm_err(display->drm,
 				"[PLANE:%d:%s] mismatch in WM%d (expected e=%d b=%u l=%u, got e=%d b=%u l=%u)\n",
 				plane->base.base.id, plane->base.name, level,
 				sw_wm_level->enable,
@@ -3959,7 +3896,7 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 		sw_wm_level = skl_plane_trans_wm(sw_wm, plane->id);
 
 		if (!skl_wm_level_equals(hw_wm_level, sw_wm_level)) {
-			drm_err(&i915->drm,
+			drm_err(display->drm,
 				"[PLANE:%d:%s] mismatch in trans WM (expected e=%d b=%u l=%u, got e=%d b=%u l=%u)\n",
 				plane->base.base.id, plane->base.name,
 				sw_wm_level->enable,
@@ -3975,7 +3912,7 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 
 		if (HAS_HW_SAGV_WM(display) &&
 		    !skl_wm_level_equals(hw_wm_level, sw_wm_level)) {
-			drm_err(&i915->drm,
+			drm_err(display->drm,
 				"[PLANE:%d:%s] mismatch in SAGV WM (expected e=%d b=%u l=%u, got e=%d b=%u l=%u)\n",
 				plane->base.base.id, plane->base.name,
 				sw_wm_level->enable,
@@ -3991,7 +3928,7 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 
 		if (HAS_HW_SAGV_WM(display) &&
 		    !skl_wm_level_equals(hw_wm_level, sw_wm_level)) {
-			drm_err(&i915->drm,
+			drm_err(display->drm,
 				"[PLANE:%d:%s] mismatch in SAGV trans WM (expected e=%d b=%u l=%u, got e=%d b=%u l=%u)\n",
 				plane->base.base.id, plane->base.name,
 				sw_wm_level->enable,
@@ -4007,7 +3944,7 @@ void intel_wm_state_verify(struct intel_atomic_state *state,
 		sw_ddb_entry = &new_crtc_state->wm.skl.plane_ddb[PLANE_CURSOR];
 
 		if (!skl_ddb_entry_equal(hw_ddb_entry, sw_ddb_entry)) {
-			drm_err(&i915->drm,
+			drm_err(display->drm,
 				"[PLANE:%d:%s] mismatch in DDB (expected (%u,%u), found (%u,%u))\n",
 				plane->base.base.id, plane->base.name,
 				sw_ddb_entry->start, sw_ddb_entry->end,
@@ -4024,29 +3961,29 @@ static const struct intel_wm_funcs skl_wm_funcs = {
 	.sanitize = skl_wm_sanitize,
 };
 
-void skl_wm_init(struct drm_i915_private *i915)
+void skl_wm_init(struct intel_display *display)
 {
-	intel_sagv_init(i915);
+	intel_sagv_init(display);
 
-	skl_setup_wm_latency(i915);
+	skl_setup_wm_latency(display);
 
-	i915->display.funcs.wm = &skl_wm_funcs;
+	display->funcs.wm = &skl_wm_funcs;
 }
 
 static int skl_watermark_ipc_status_show(struct seq_file *m, void *data)
 {
-	struct drm_i915_private *i915 = m->private;
+	struct intel_display *display = m->private;
 
 	seq_printf(m, "Isochronous Priority Control: %s\n",
-		   str_yes_no(skl_watermark_ipc_enabled(i915)));
+		   str_yes_no(skl_watermark_ipc_enabled(display)));
 	return 0;
 }
 
 static int skl_watermark_ipc_status_open(struct inode *inode, struct file *file)
 {
-	struct drm_i915_private *i915 = inode->i_private;
+	struct intel_display *display = inode->i_private;
 
-	return single_open(file, skl_watermark_ipc_status_show, i915);
+	return single_open(file, skl_watermark_ipc_status_show, display);
 }
 
 static ssize_t skl_watermark_ipc_status_write(struct file *file,
@@ -4054,8 +3991,7 @@ static ssize_t skl_watermark_ipc_status_write(struct file *file,
 					      size_t len, loff_t *offp)
 {
 	struct seq_file *m = file->private_data;
-	struct drm_i915_private *i915 = m->private;
-	intel_wakeref_t wakeref;
+	struct intel_display *display = m->private;
 	bool enable;
 	int ret;
 
@@ -4063,12 +3999,12 @@ static ssize_t skl_watermark_ipc_status_write(struct file *file,
 	if (ret < 0)
 		return ret;
 
-	with_intel_runtime_pm(&i915->runtime_pm, wakeref) {
-		if (!skl_watermark_ipc_enabled(i915) && enable)
-			drm_info(&i915->drm,
+	with_intel_display_rpm(display) {
+		if (!skl_watermark_ipc_enabled(display) && enable)
+			drm_info(display->drm,
 				 "Enabling IPC: WM will be proper only after next commit\n");
-		i915->display.wm.ipc_enabled = enable;
-		skl_watermark_ipc_update(i915);
+		display->wm.ipc_enabled = enable;
+		skl_watermark_ipc_update(display);
 	}
 
 	return len;
@@ -4085,7 +4021,7 @@ static const struct file_operations skl_watermark_ipc_status_fops = {
 
 static int intel_sagv_status_show(struct seq_file *m, void *unused)
 {
-	struct drm_i915_private *i915 = m->private;
+	struct intel_display *display = m->private;
 	static const char * const sagv_status[] = {
 		[I915_SAGV_UNKNOWN] = "unknown",
 		[I915_SAGV_DISABLED] = "disabled",
@@ -4093,37 +4029,36 @@ static int intel_sagv_status_show(struct seq_file *m, void *unused)
 		[I915_SAGV_NOT_CONTROLLED] = "not controlled",
 	};
 
-	seq_printf(m, "SAGV available: %s\n", str_yes_no(intel_has_sagv(i915)));
+	seq_printf(m, "SAGV available: %s\n", str_yes_no(intel_has_sagv(display)));
 	seq_printf(m, "SAGV modparam: %s\n",
-		   str_enabled_disabled(i915->display.params.enable_sagv));
-	seq_printf(m, "SAGV status: %s\n", sagv_status[i915->display.sagv.status]);
-	seq_printf(m, "SAGV block time: %d usec\n", i915->display.sagv.block_time_us);
+		   str_enabled_disabled(display->params.enable_sagv));
+	seq_printf(m, "SAGV status: %s\n", sagv_status[display->sagv.status]);
+	seq_printf(m, "SAGV block time: %d usec\n", display->sagv.block_time_us);
 
 	return 0;
 }
 
 DEFINE_SHOW_ATTRIBUTE(intel_sagv_status);
 
-void skl_watermark_debugfs_register(struct drm_i915_private *i915)
+void skl_watermark_debugfs_register(struct intel_display *display)
 {
-	struct intel_display *display = &i915->display;
 	struct drm_minor *minor = display->drm->primary;
 
 	if (HAS_IPC(display))
-		debugfs_create_file("i915_ipc_status", 0644, minor->debugfs_root, i915,
-				    &skl_watermark_ipc_status_fops);
+		debugfs_create_file("i915_ipc_status", 0644, minor->debugfs_root,
+				    display, &skl_watermark_ipc_status_fops);
 
 	if (HAS_SAGV(display))
-		debugfs_create_file("i915_sagv_status", 0444, minor->debugfs_root, i915,
-				    &intel_sagv_status_fops);
+		debugfs_create_file("i915_sagv_status", 0444, minor->debugfs_root,
+				    display, &intel_sagv_status_fops);
 }
 
-unsigned int skl_watermark_max_latency(struct drm_i915_private *i915, int initial_wm_level)
+unsigned int skl_watermark_max_latency(struct intel_display *display, int initial_wm_level)
 {
 	int level;
 
-	for (level = i915->display.wm.num_levels - 1; level >= initial_wm_level; level--) {
-		unsigned int latency = skl_wm_latency(i915, level, NULL);
+	for (level = display->wm.num_levels - 1; level >= initial_wm_level; level--) {
+		unsigned int latency = skl_wm_latency(display, level, NULL);
 
 		if (latency)
 			return latency;
diff --git a/drivers/gpu/drm/i915/display/skl_watermark.h b/drivers/gpu/drm/i915/display/skl_watermark.h
index d9cff6c54310..95b0b599d5c3 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.h
+++ b/drivers/gpu/drm/i915/display/skl_watermark.h
@@ -12,7 +12,6 @@
 #include "intel_global_state.h"
 #include "intel_wm_types.h"
 
-struct drm_i915_private;
 struct intel_atomic_state;
 struct intel_bw_state;
 struct intel_crtc;
@@ -27,11 +26,12 @@ u8 intel_enabled_dbuf_slices_mask(struct intel_display *display);
 
 void intel_sagv_pre_plane_update(struct intel_atomic_state *state);
 void intel_sagv_post_plane_update(struct intel_atomic_state *state);
-bool intel_can_enable_sagv(struct drm_i915_private *i915,
+bool intel_crtc_can_enable_sagv(const struct intel_crtc_state *crtc_state);
+bool intel_can_enable_sagv(struct intel_display *display,
 			   const struct intel_bw_state *bw_state);
-bool intel_has_sagv(struct drm_i915_private *i915);
+bool intel_has_sagv(struct intel_display *display);
 
-u32 skl_ddb_dbuf_slice_mask(struct drm_i915_private *i915,
+u32 skl_ddb_dbuf_slice_mask(struct intel_display *display,
 			    const struct skl_ddb_entry *entry);
 
 bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb,
@@ -45,14 +45,14 @@ void skl_wm_crtc_disable_noatomic(struct intel_crtc *crtc);
 void skl_wm_plane_disable_noatomic(struct intel_crtc *crtc,
 				   struct intel_plane *plane);
 
-void skl_watermark_ipc_init(struct drm_i915_private *i915);
-void skl_watermark_ipc_update(struct drm_i915_private *i915);
-bool skl_watermark_ipc_enabled(struct drm_i915_private *i915);
-void skl_watermark_debugfs_register(struct drm_i915_private *i915);
+void skl_watermark_ipc_init(struct intel_display *display);
+void skl_watermark_ipc_update(struct intel_display *display);
+bool skl_watermark_ipc_enabled(struct intel_display *display);
+void skl_watermark_debugfs_register(struct intel_display *display);
 
-unsigned int skl_watermark_max_latency(struct drm_i915_private *i915,
+unsigned int skl_watermark_max_latency(struct intel_display *display,
 				       int initial_wm_level);
-void skl_wm_init(struct drm_i915_private *i915);
+void skl_wm_init(struct intel_display *display);
 
 const struct skl_wm_level *skl_plane_wm_level(const struct skl_pipe_wm *pipe_wm,
 					      enum plane_id plane_id,
@@ -86,13 +86,13 @@ intel_atomic_get_dbuf_state(struct intel_atomic_state *state);
 #define intel_atomic_get_new_dbuf_state(state) \
 	to_intel_dbuf_state(intel_atomic_get_new_global_obj_state(state, &to_intel_display(state)->dbuf.obj))
 
-int intel_dbuf_init(struct drm_i915_private *i915);
+int intel_dbuf_init(struct intel_display *display);
 int intel_dbuf_state_set_mdclk_cdclk_ratio(struct intel_atomic_state *state,
 					   int ratio);
 
 void intel_dbuf_pre_plane_update(struct intel_atomic_state *state);
 void intel_dbuf_post_plane_update(struct intel_atomic_state *state);
-void intel_dbuf_mdclk_cdclk_ratio_update(struct drm_i915_private *i915,
+void intel_dbuf_mdclk_cdclk_ratio_update(struct intel_display *display,
 					 int ratio, bool joined_mbus);
 void intel_dbuf_mbus_pre_ddb_update(struct intel_atomic_state *state);
 void intel_dbuf_mbus_post_ddb_update(struct intel_atomic_state *state);
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index af717df83197..346737f15fa9 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -251,8 +251,10 @@ static int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs,
 	return 0;
 }
 
-static void band_gap_reset(struct drm_i915_private *dev_priv)
+static void band_gap_reset(struct intel_display *display)
 {
+	struct drm_i915_private *dev_priv = to_i915(display->drm);
+
 	vlv_flisdsi_get(dev_priv);
 
 	vlv_flisdsi_write(dev_priv, 0x08, 0x0001);
@@ -269,13 +271,13 @@ static int intel_dsi_compute_config(struct intel_encoder *encoder,
 				    struct intel_crtc_state *pipe_config,
 				    struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	struct intel_connector *intel_connector = intel_dsi->attached_connector;
 	struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
 	int ret;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
@@ -298,7 +300,7 @@ static int intel_dsi_compute_config(struct intel_encoder *encoder,
 	else
 		pipe_config->pipe_bpp = 18;
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	if (display->platform.geminilake || display->platform.broxton) {
 		/* Enable Frame time stamp based scanline reporting */
 		pipe_config->mode_flags |=
 			I915_MODE_FLAG_GET_SCANLINE_FROM_TIMESTAMP;
@@ -468,7 +470,7 @@ static void vlv_dsi_device_ready(struct intel_encoder *encoder)
 	vlv_flisdsi_put(dev_priv);
 
 	/* bandgap reset is needed after everytime we do power gate */
-	band_gap_reset(dev_priv);
+	band_gap_reset(display);
 
 	for_each_dsi_port(port, intel_dsi->ports) {
 
@@ -495,11 +497,11 @@ static void vlv_dsi_device_ready(struct intel_encoder *encoder)
 
 static void intel_dsi_device_ready(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (IS_GEMINILAKE(dev_priv))
+	if (display->platform.geminilake)
 		glk_dsi_device_ready(encoder);
-	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	else if (display->platform.geminilake || display->platform.broxton)
 		bxt_dsi_device_ready(encoder);
 	else
 		vlv_dsi_device_ready(encoder);
@@ -559,23 +561,22 @@ static void glk_dsi_clear_device_ready(struct intel_encoder *encoder)
 	glk_dsi_disable_mipi_io(encoder);
 }
 
-static i915_reg_t port_ctrl_reg(struct drm_i915_private *i915, enum port port)
+static i915_reg_t port_ctrl_reg(struct intel_display *display, enum port port)
 {
-	return IS_GEMINILAKE(i915) || IS_BROXTON(i915) ?
+	return display->platform.geminilake || display->platform.broxton ?
 		BXT_MIPI_PORT_CTRL(port) : VLV_MIPI_PORT_CTRL(port);
 }
 
 static void vlv_dsi_clear_device_ready(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 
 	drm_dbg_kms(display->drm, "\n");
 	for_each_dsi_port(port, intel_dsi->ports) {
 		/* Common bit for both MIPI Port A & MIPI Port C on VLV/CHV */
-		i915_reg_t port_ctrl = IS_BROXTON(dev_priv) ?
+		i915_reg_t port_ctrl = display->platform.broxton ?
 			BXT_MIPI_PORT_CTRL(port) : VLV_MIPI_PORT_CTRL(PORT_A);
 
 		intel_de_write(display, MIPI_DEVICE_READY(display, port),
@@ -594,7 +595,7 @@ static void vlv_dsi_clear_device_ready(struct intel_encoder *encoder)
 		 * On VLV/CHV, wait till Clock lanes are in LP-00 state for MIPI
 		 * Port A only. MIPI Port C has no similar bit for checking.
 		 */
-		if ((IS_BROXTON(dev_priv) || port == PORT_A) &&
+		if ((display->platform.broxton || port == PORT_A) &&
 		    intel_de_wait_for_clear(display, port_ctrl,
 					    AFE_LATCHOUT, 30))
 			drm_err(display->drm, "DSI LP not going Low\n");
@@ -612,7 +613,6 @@ static void intel_dsi_port_enable(struct intel_encoder *encoder,
 				  const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
@@ -620,7 +620,7 @@ static void intel_dsi_port_enable(struct intel_encoder *encoder,
 	if (intel_dsi->dual_link == DSI_DUAL_LINK_FRONT_BACK) {
 		u32 temp = intel_dsi->pixel_overlap;
 
-		if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+		if (display->platform.geminilake || display->platform.broxton) {
 			for_each_dsi_port(port, intel_dsi->ports)
 				intel_de_rmw(display, MIPI_CTRL(display, port),
 					     BXT_PIXEL_OVERLAP_CNT_MASK,
@@ -633,7 +633,7 @@ static void intel_dsi_port_enable(struct intel_encoder *encoder,
 	}
 
 	for_each_dsi_port(port, intel_dsi->ports) {
-		i915_reg_t port_ctrl = port_ctrl_reg(dev_priv, port);
+		i915_reg_t port_ctrl = port_ctrl_reg(display, port);
 		u32 temp;
 
 		temp = intel_de_read(display, port_ctrl);
@@ -644,7 +644,7 @@ static void intel_dsi_port_enable(struct intel_encoder *encoder,
 		if (intel_dsi->ports == (BIT(PORT_A) | BIT(PORT_C))) {
 			temp |= (intel_dsi->dual_link - 1)
 						<< DUAL_LINK_MODE_SHIFT;
-			if (IS_BROXTON(dev_priv))
+			if (display->platform.broxton)
 				temp |= LANE_CONFIGURATION_DUAL_LINK_A;
 			else
 				temp |= crtc->pipe ?
@@ -664,12 +664,11 @@ static void intel_dsi_port_enable(struct intel_encoder *encoder,
 static void intel_dsi_port_disable(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 
 	for_each_dsi_port(port, intel_dsi->ports) {
-		i915_reg_t port_ctrl = port_ctrl_reg(dev_priv, port);
+		i915_reg_t port_ctrl = port_ctrl_reg(display, port);
 
 		/* de-assert ip_tg_enable signal */
 		intel_de_rmw(display, port_ctrl, DPI_ENABLE, 0);
@@ -730,7 +729,6 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	enum port port;
 	bool glk_cold_boot = false;
@@ -745,7 +743,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 	 * The BIOS may leave the PLL in a wonky state where it doesn't
 	 * lock. It needs to be fully powered down to fix it.
 	 */
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	if (display->platform.geminilake || display->platform.broxton) {
 		bxt_dsi_pll_disable(encoder);
 		bxt_dsi_pll_enable(encoder, pipe_config);
 	} else {
@@ -753,7 +751,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 		vlv_dsi_pll_enable(encoder, pipe_config);
 	}
 
-	if (IS_BROXTON(dev_priv)) {
+	if (display->platform.broxton) {
 		/* Add MIPI IO reset programming for modeset */
 		intel_de_rmw(display, BXT_P_CR_GT_DISP_PWRON, 0, MIPIO_RST_CTRL);
 
@@ -762,13 +760,13 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 		intel_de_write(display, BXT_P_DSI_REGULATOR_TX_CTRL, 0);
 	}
 
-	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
+	if (display->platform.valleyview || display->platform.cherryview) {
 		/* Disable DPOunit clock gating, can stall pipe */
-		intel_de_rmw(display, DSPCLK_GATE_D(dev_priv),
+		intel_de_rmw(display, DSPCLK_GATE_D(display),
 			     0, DPOUNIT_CLOCK_GATE_DISABLE);
 	}
 
-	if (!IS_GEMINILAKE(dev_priv))
+	if (!display->platform.geminilake)
 		intel_dsi_prepare(encoder, pipe_config);
 
 	/* Give the panel time to power-on and then deassert its reset */
@@ -776,7 +774,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 	msleep(intel_dsi->panel_on_delay);
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DEASSERT_RESET);
 
-	if (IS_GEMINILAKE(dev_priv)) {
+	if (display->platform.geminilake) {
 		glk_cold_boot = glk_dsi_enable_io(encoder);
 
 		/* Prepare port in cold boot(s3/s4) scenario */
@@ -788,7 +786,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 	intel_dsi_device_ready(encoder);
 
 	/* Prepare port in normal boot scenario */
-	if (IS_GEMINILAKE(dev_priv) && !glk_cold_boot)
+	if (display->platform.geminilake && !glk_cold_boot)
 		intel_dsi_prepare(encoder, pipe_config);
 
 	/* Send initialization commands in LP mode */
@@ -836,11 +834,11 @@ static void intel_dsi_disable(struct intel_atomic_state *state,
 			      const struct intel_crtc_state *old_crtc_state,
 			      const struct drm_connector_state *old_conn_state)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 
-	drm_dbg_kms(&i915->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_BACKLIGHT_OFF);
 	intel_backlight_disable(old_conn_state);
@@ -860,9 +858,9 @@ static void intel_dsi_disable(struct intel_atomic_state *state,
 
 static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	if (IS_GEMINILAKE(dev_priv))
+	if (display->platform.geminilake)
 		glk_dsi_clear_device_ready(encoder);
 	else
 		vlv_dsi_clear_device_ready(encoder);
@@ -874,13 +872,12 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state,
 				   const struct drm_connector_state *old_conn_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 
 	drm_dbg_kms(display->drm, "\n");
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	if (display->platform.geminilake || display->platform.broxton) {
 		intel_crtc_vblank_off(old_crtc_state);
 
 		skl_scaler_disable(old_crtc_state);
@@ -907,7 +904,7 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state,
 	/* Transition to LP-00 */
 	intel_dsi_clear_device_ready(encoder);
 
-	if (IS_BROXTON(dev_priv)) {
+	if (display->platform.broxton) {
 		/* Power down DSI regulator to save power */
 		intel_de_write(display, BXT_P_DSI_REGULATOR_CFG, STAP_SELECT);
 		intel_de_write(display, BXT_P_DSI_REGULATOR_TX_CTRL,
@@ -917,12 +914,12 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state,
 		intel_de_rmw(display, BXT_P_CR_GT_DISP_PWRON, MIPIO_RST_CTRL, 0);
 	}
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	if (display->platform.geminilake || display->platform.broxton) {
 		bxt_dsi_pll_disable(encoder);
 	} else {
 		vlv_dsi_pll_disable(encoder);
 
-		intel_de_rmw(display, DSPCLK_GATE_D(dev_priv),
+		intel_de_rmw(display, DSPCLK_GATE_D(display),
 			     DPOUNIT_CLOCK_GATE_DISABLE, 0);
 	}
 
@@ -939,7 +936,6 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 				   enum pipe *pipe)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	intel_wakeref_t wakeref;
 	enum port port;
@@ -957,13 +953,13 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 	 * configuration, otherwise accessing DSI registers will hang the
 	 * machine. See BSpec North Display Engine registers/MIPI[BXT].
 	 */
-	if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
-	    !bxt_dsi_pll_is_enabled(dev_priv))
+	if ((display->platform.geminilake || display->platform.broxton) &&
+	    !bxt_dsi_pll_is_enabled(display))
 		goto out_put_power;
 
 	/* XXX: this only works for one DSI output */
 	for_each_dsi_port(port, intel_dsi->ports) {
-		i915_reg_t port_ctrl = port_ctrl_reg(dev_priv, port);
+		i915_reg_t port_ctrl = port_ctrl_reg(display, port);
 		bool enabled = intel_de_read(display, port_ctrl) & DPI_ENABLE;
 
 		/*
@@ -971,10 +967,10 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 		 * bit in port C control register does not get set. As a
 		 * workaround, check pipe B conf instead.
 		 */
-		if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+		if ((display->platform.valleyview || display->platform.cherryview) &&
 		    port == PORT_C)
 			enabled = intel_de_read(display,
-						TRANSCONF(dev_priv, PIPE_B)) & TRANSCONF_ENABLE;
+						TRANSCONF(display, PIPE_B)) & TRANSCONF_ENABLE;
 
 		/* Try command mode if video mode not enabled */
 		if (!enabled) {
@@ -989,7 +985,7 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 		if (!(intel_de_read(display, MIPI_DEVICE_READY(display, port)) & DEVICE_READY))
 			continue;
 
-		if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+		if (display->platform.geminilake || display->platform.broxton) {
 			u32 tmp = intel_de_read(display, MIPI_CTRL(display, port));
 			tmp &= BXT_PIPE_SELECT_MASK;
 			tmp >>= BXT_PIPE_SELECT_SHIFT;
@@ -1177,15 +1173,15 @@ static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder,
 static void intel_dsi_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *pipe_config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	u32 pclk;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	pipe_config->output_types |= BIT(INTEL_OUTPUT_DSI);
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	if (display->platform.geminilake || display->platform.broxton) {
 		bxt_dsi_get_pipe_config(encoder, pipe_config);
 		pclk = bxt_dsi_get_pclk(encoder, pipe_config);
 	} else {
@@ -1218,7 +1214,6 @@ static void set_dsi_timings(struct intel_encoder *encoder,
 			    const struct drm_display_mode *adjusted_mode)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 	unsigned int bpp = mipi_dsi_pixel_format_to_bpp(intel_dsi->pixel_format);
@@ -1253,7 +1248,7 @@ static void set_dsi_timings(struct intel_encoder *encoder,
 	hbp = txbyteclkhs(hbp, bpp, lane_count, intel_dsi->burst_mode_ratio);
 
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+		if (display->platform.geminilake || display->platform.broxton) {
 			/*
 			 * Program hdisplay and vdisplay on MIPI transcoder.
 			 * This is different from calculated hactive and
@@ -1307,7 +1302,6 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 			      const struct intel_crtc_state *pipe_config)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	const struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
@@ -1327,7 +1321,7 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 	}
 
 	for_each_dsi_port(port, intel_dsi->ports) {
-		if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
+		if (display->platform.valleyview || display->platform.cherryview) {
 			/*
 			 * escape clock divider, 20MHz, shared for A and C.
 			 * device ready must be off when doing this! txclkesc?
@@ -1342,7 +1336,7 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 			tmp &= ~READ_REQUEST_PRIORITY_MASK;
 			intel_de_write(display, MIPI_CTRL(display, port),
 				       tmp | READ_REQUEST_PRIORITY_HIGH);
-		} else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+		} else if (display->platform.geminilake || display->platform.broxton) {
 			enum pipe pipe = crtc->pipe;
 
 			intel_de_rmw(display, MIPI_CTRL(display, port),
@@ -1377,7 +1371,7 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 	if (intel_dsi->clock_stop)
 		tmp |= CLOCKSTOP;
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+	if (display->platform.geminilake || display->platform.broxton) {
 		tmp |= BXT_DPHY_DEFEATURE_EN;
 		if (!is_cmd_mode(intel_dsi))
 			tmp |= BXT_DEFEATURE_DPI_FIFO_CTR;
@@ -1424,7 +1418,7 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 		intel_de_write(display, MIPI_INIT_COUNT(display, port),
 			       txclkesc(intel_dsi->escape_clk_div, 100));
 
-		if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
+		if ((display->platform.geminilake || display->platform.broxton) &&
 		    !intel_dsi->dual_link) {
 			/*
 			 * BXT spec says write MIPI_INIT_COUNT for
@@ -1461,7 +1455,7 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 		intel_de_write(display, MIPI_LP_BYTECLK(display, port),
 			       intel_dsi->lp_byte_clk);
 
-		if (IS_GEMINILAKE(dev_priv)) {
+		if (display->platform.geminilake) {
 			intel_de_write(display, MIPI_TLPX_TIME_COUNT(display, port),
 				       intel_dsi->lp_byte_clk);
 			/* Shadow of DPHY reg */
@@ -1513,18 +1507,17 @@ static void intel_dsi_prepare(struct intel_encoder *encoder,
 static void intel_dsi_unprepare(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 
-	if (IS_GEMINILAKE(dev_priv))
+	if (display->platform.geminilake)
 		return;
 
 	for_each_dsi_port(port, intel_dsi->ports) {
 		/* Panel commands can be sent when clock is in LP11 */
 		intel_de_write(display, MIPI_DEVICE_READY(display, port), 0x0);
 
-		if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+		if (display->platform.geminilake || display->platform.broxton)
 			bxt_dsi_reset_clocks(encoder, port);
 		else
 			vlv_dsi_reset_clocks(encoder, port);
@@ -1596,8 +1589,8 @@ static void vlv_dsi_add_properties(struct intel_connector *connector)
 
 static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 {
-	struct drm_i915_private *dev_priv = to_i915(intel_dsi->base.base.dev);
 	struct intel_connector *connector = intel_dsi->attached_connector;
+	struct intel_display *display = to_intel_display(connector);
 	struct mipi_config *mipi_config = connector->panel.vbt.dsi.config;
 	u32 tlpx_ns, extra_byte_count, tlpx_ui;
 	u32 ui_num, ui_den;
@@ -1645,7 +1638,7 @@ static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 	 * For GEMINILAKE dphy_param_reg will be programmed in terms of
 	 * HS byte clock count for other platform in HS ddr clock count
 	 */
-	mul = IS_GEMINILAKE(dev_priv) ? 8 : 2;
+	mul = display->platform.geminilake ? 8 : 2;
 	ths_prepare_ns = max(mipi_config->ths_prepare,
 			     mipi_config->tclk_prepare);
 
@@ -1653,7 +1646,7 @@ static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 	prepare_cnt = DIV_ROUND_UP(ths_prepare_ns * ui_den, ui_num * mul);
 
 	if (prepare_cnt > PREPARE_CNT_MAX) {
-		drm_dbg_kms(&dev_priv->drm, "prepare count too high %u\n",
+		drm_dbg_kms(display->drm, "prepare count too high %u\n",
 			    prepare_cnt);
 		prepare_cnt = PREPARE_CNT_MAX;
 	}
@@ -1674,7 +1667,7 @@ static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 		exit_zero_cnt += 1;
 
 	if (exit_zero_cnt > EXIT_ZERO_CNT_MAX) {
-		drm_dbg_kms(&dev_priv->drm, "exit zero count too high %u\n",
+		drm_dbg_kms(display->drm, "exit zero count too high %u\n",
 			    exit_zero_cnt);
 		exit_zero_cnt = EXIT_ZERO_CNT_MAX;
 	}
@@ -1685,7 +1678,7 @@ static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 				* ui_den, ui_num * mul);
 
 	if (clk_zero_cnt > CLK_ZERO_CNT_MAX) {
-		drm_dbg_kms(&dev_priv->drm, "clock zero count too high %u\n",
+		drm_dbg_kms(display->drm, "clock zero count too high %u\n",
 			    clk_zero_cnt);
 		clk_zero_cnt = CLK_ZERO_CNT_MAX;
 	}
@@ -1695,7 +1688,7 @@ static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 	trail_cnt = DIV_ROUND_UP(tclk_trail_ns * ui_den, ui_num * mul);
 
 	if (trail_cnt > TRAIL_CNT_MAX) {
-		drm_dbg_kms(&dev_priv->drm, "trail count too high %u\n",
+		drm_dbg_kms(display->drm, "trail count too high %u\n",
 			    trail_cnt);
 		trail_cnt = TRAIL_CNT_MAX;
 	}
@@ -1761,7 +1754,7 @@ static void vlv_dphy_param_init(struct intel_dsi *intel_dsi)
 
 int vlv_dsi_min_cdclk(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	struct intel_display *display = to_intel_display(crtc_state);
 
 	if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI))
 		return 0;
@@ -1770,7 +1763,7 @@ int vlv_dsi_min_cdclk(const struct intel_crtc_state *crtc_state)
 	 * On Valleyview some DSI panels lose (v|h)sync when the clock is lower
 	 * than 320000KHz.
 	 */
-	if (IS_VALLEYVIEW(dev_priv))
+	if (display->platform.valleyview)
 		return 320000;
 
 	/*
@@ -1778,7 +1771,7 @@ int vlv_dsi_min_cdclk(const struct intel_crtc_state *crtc_state)
 	 * picture gets unstable, despite that values are
 	 * correct for DSI PLL and DE PLL.
 	 */
-	if (IS_GEMINILAKE(dev_priv))
+	if (display->platform.geminilake)
 		return 158400;
 
 	return 0;
@@ -1903,9 +1896,8 @@ static const struct dmi_system_id vlv_dsi_dmi_quirk_table[] = {
 	{ }
 };
 
-void vlv_dsi_init(struct drm_i915_private *dev_priv)
+void vlv_dsi_init(struct intel_display *display)
 {
-	struct intel_display *display = &dev_priv->display;
 	struct intel_dsi *intel_dsi;
 	struct intel_encoder *encoder;
 	struct intel_connector *connector;
@@ -1914,16 +1906,16 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	enum port port;
 	enum pipe pipe;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	/* There is no detection method for MIPI so rely on VBT */
 	if (!intel_bios_is_dsi_present(display, &port))
 		return;
 
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
-		dev_priv->display.dsi.mmio_base = BXT_MIPI_BASE;
+	if (display->platform.geminilake || display->platform.broxton)
+		display->dsi.mmio_base = BXT_MIPI_BASE;
 	else
-		dev_priv->display.dsi.mmio_base = VLV_MIPI_BASE;
+		display->dsi.mmio_base = VLV_MIPI_BASE;
 
 	intel_dsi = kzalloc(sizeof(*intel_dsi), GFP_KERNEL);
 	if (!intel_dsi)
@@ -1938,12 +1930,12 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	encoder = &intel_dsi->base;
 	intel_dsi->attached_connector = connector;
 
-	drm_encoder_init(&dev_priv->drm, &encoder->base, &intel_dsi_funcs,
+	drm_encoder_init(display->drm, &encoder->base, &intel_dsi_funcs,
 			 DRM_MODE_ENCODER_DSI, "DSI %c", port_name(port));
 
 	encoder->compute_config = intel_dsi_compute_config;
 	encoder->pre_enable = intel_dsi_pre_enable;
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		encoder->enable = bxt_dsi_enable;
 	encoder->disable = intel_dsi_disable;
 	encoder->post_disable = intel_dsi_post_disable;
@@ -1963,7 +1955,7 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	 * On BYT/CHV, pipe A maps to MIPI DSI port A, pipe B maps to MIPI DSI
 	 * port C. BXT isn't limited like this.
 	 */
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+	if (display->platform.geminilake || display->platform.broxton)
 		encoder->pipe_mask = ~0;
 	else if (port == PORT_A)
 		encoder->pipe_mask = BIT(PIPE_A);
@@ -1979,10 +1971,10 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	else
 		intel_dsi->ports = BIT(port);
 
-	if (drm_WARN_ON(&dev_priv->drm, connector->panel.vbt.dsi.bl_ports & ~intel_dsi->ports))
+	if (drm_WARN_ON(display->drm, connector->panel.vbt.dsi.bl_ports & ~intel_dsi->ports))
 		connector->panel.vbt.dsi.bl_ports &= intel_dsi->ports;
 
-	if (drm_WARN_ON(&dev_priv->drm, connector->panel.vbt.dsi.cabc_ports & ~intel_dsi->ports))
+	if (drm_WARN_ON(display->drm, connector->panel.vbt.dsi.cabc_ports & ~intel_dsi->ports))
 		connector->panel.vbt.dsi.cabc_ports &= intel_dsi->ports;
 
 	/* Create a DSI host (and a device) for each port. */
@@ -1998,18 +1990,18 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	}
 
 	if (!intel_dsi_vbt_init(intel_dsi, MIPI_DSI_GENERIC_PANEL_ID)) {
-		drm_dbg_kms(&dev_priv->drm, "no device found\n");
+		drm_dbg_kms(display->drm, "no device found\n");
 		goto err;
 	}
 
 	/* Use clock read-back from current hw-state for fastboot */
 	current_mode = intel_encoder_current_mode(encoder);
 	if (current_mode) {
-		drm_dbg_kms(&dev_priv->drm, "Calculated pclk %d GOP %d\n",
+		drm_dbg_kms(display->drm, "Calculated pclk %d GOP %d\n",
 			    intel_dsi->pclk, current_mode->clock);
 		if (intel_fuzzy_clock_check(intel_dsi->pclk,
 					    current_mode->clock)) {
-			drm_dbg_kms(&dev_priv->drm, "Using GOP pclk\n");
+			drm_dbg_kms(display->drm, "Using GOP pclk\n");
 			intel_dsi->pclk = current_mode->clock;
 		}
 
@@ -2021,7 +2013,7 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 	intel_dsi_vbt_gpio_init(intel_dsi,
 				intel_dsi_get_hw_state(encoder, &pipe));
 
-	drm_connector_init(&dev_priv->drm, &connector->base, &intel_dsi_connector_funcs,
+	drm_connector_init(display->drm, &connector->base, &intel_dsi_connector_funcs,
 			   DRM_MODE_CONNECTOR_DSI);
 
 	drm_connector_helper_add(&connector->base, &intel_dsi_connector_helper_funcs);
@@ -2030,12 +2022,12 @@ void vlv_dsi_init(struct drm_i915_private *dev_priv)
 
 	intel_connector_attach_encoder(connector, encoder);
 
-	mutex_lock(&dev_priv->drm.mode_config.mutex);
+	mutex_lock(&display->drm->mode_config.mutex);
 	intel_panel_add_vbt_lfp_fixed_mode(connector);
-	mutex_unlock(&dev_priv->drm.mode_config.mutex);
+	mutex_unlock(&display->drm->mode_config.mutex);
 
 	if (!intel_panel_preferred_fixed_mode(connector)) {
-		drm_dbg_kms(&dev_priv->drm, "no fixed mode\n");
+		drm_dbg_kms(display->drm, "no fixed mode\n");
 		goto err_cleanup_connector;
 	}
 
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.h b/drivers/gpu/drm/i915/display/vlv_dsi.h
index 277bacfbc551..ff349b5876c2 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.h
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.h
@@ -7,14 +7,14 @@
 #define __VLV_DSI_H__
 
 enum port;
-struct drm_i915_private;
 struct intel_crtc_state;
+struct intel_display;
 struct intel_dsi;
 
 #ifdef I915
 void vlv_dsi_wait_for_fifo_empty(struct intel_dsi *intel_dsi, enum port port);
 int vlv_dsi_min_cdclk(const struct intel_crtc_state *crtc_state);
-void vlv_dsi_init(struct drm_i915_private *dev_priv);
+void vlv_dsi_init(struct intel_display *display);
 #else
 static inline void vlv_dsi_wait_for_fifo_empty(struct intel_dsi *intel_dsi, enum port port)
 {
@@ -23,7 +23,7 @@ static inline int vlv_dsi_min_cdclk(const struct intel_crtc_state *crtc_state)
 {
 	return 0;
 }
-static inline void vlv_dsi_init(struct drm_i915_private *dev_priv)
+static inline void vlv_dsi_init(struct intel_display *display)
 {
 }
 #endif
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
index 2ed47e7d1051..7ce924a5ef90 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
@@ -57,7 +57,7 @@ static u32 dsi_clk_from_pclk(u32 pclk, enum mipi_dsi_pixel_format fmt,
 	return dsi_clk_khz;
 }
 
-static int dsi_calc_mnp(struct drm_i915_private *dev_priv,
+static int dsi_calc_mnp(struct intel_display *display,
 			struct intel_crtc_state *config,
 			int target_dsi_clk)
 {
@@ -68,11 +68,11 @@ static int dsi_calc_mnp(struct drm_i915_private *dev_priv,
 
 	/* target_dsi_clk is expected in kHz */
 	if (target_dsi_clk < 300000 || target_dsi_clk > 1150000) {
-		drm_err(&dev_priv->drm, "DSI CLK Out of Range\n");
+		drm_err(display->drm, "DSI CLK Out of Range\n");
 		return -ECHRNG;
 	}
 
-	if (IS_CHERRYVIEW(dev_priv)) {
+	if (display->platform.cherryview) {
 		ref_clk = 100000;
 		n = 4;
 		m_min = 70;
@@ -116,13 +116,13 @@ static int dsi_calc_mnp(struct drm_i915_private *dev_priv,
 static int vlv_dsi_pclk(struct intel_encoder *encoder,
 			struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	int bpp = mipi_dsi_pixel_format_to_bpp(intel_dsi->pixel_format);
 	u32 dsi_clock;
 	u32 pll_ctl, pll_div;
 	u32 m = 0, p = 0, n;
-	int refclk = IS_CHERRYVIEW(dev_priv) ? 100000 : 25000;
+	int refclk = display->platform.cherryview ? 100000 : 25000;
 	int i;
 
 	pll_ctl = config->dsi_pll.ctrl;
@@ -147,7 +147,7 @@ static int vlv_dsi_pclk(struct intel_encoder *encoder,
 	p--;
 
 	if (!p) {
-		drm_err(&dev_priv->drm, "wrong P1 divisor\n");
+		drm_err(display->drm, "wrong P1 divisor\n");
 		return 0;
 	}
 
@@ -157,7 +157,7 @@ static int vlv_dsi_pclk(struct intel_encoder *encoder,
 	}
 
 	if (i == ARRAY_SIZE(lfsr_converts)) {
-		drm_err(&dev_priv->drm, "wrong m_seed programmed\n");
+		drm_err(display->drm, "wrong m_seed programmed\n");
 		return 0;
 	}
 
@@ -175,16 +175,16 @@ static int vlv_dsi_pclk(struct intel_encoder *encoder,
 int vlv_dsi_pll_compute(struct intel_encoder *encoder,
 			struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	int pclk, dsi_clk, ret;
 
 	dsi_clk = dsi_clk_from_pclk(intel_dsi->pclk, intel_dsi->pixel_format,
 				    intel_dsi->lane_count);
 
-	ret = dsi_calc_mnp(dev_priv, config, dsi_clk);
+	ret = dsi_calc_mnp(display, config, dsi_clk);
 	if (ret) {
-		drm_dbg_kms(&dev_priv->drm, "dsi_calc_mnp failed\n");
+		drm_dbg_kms(display->drm, "dsi_calc_mnp failed\n");
 		return ret;
 	}
 
@@ -196,7 +196,7 @@ int vlv_dsi_pll_compute(struct intel_encoder *encoder,
 
 	config->dsi_pll.ctrl |= DSI_PLL_VCO_EN;
 
-	drm_dbg_kms(&dev_priv->drm, "dsi pll div %08x, ctrl %08x\n",
+	drm_dbg_kms(display->drm, "dsi pll div %08x, ctrl %08x\n",
 		    config->dsi_pll.div, config->dsi_pll.ctrl);
 
 	pclk = vlv_dsi_pclk(encoder, config);
@@ -213,9 +213,10 @@ int vlv_dsi_pll_compute(struct intel_encoder *encoder,
 void vlv_dsi_pll_enable(struct intel_encoder *encoder,
 			const struct intel_crtc_state *config)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	vlv_cck_get(dev_priv);
 
@@ -235,20 +236,21 @@ void vlv_dsi_pll_enable(struct intel_encoder *encoder,
 						DSI_PLL_LOCK, 20)) {
 
 		vlv_cck_put(dev_priv);
-		drm_err(&dev_priv->drm, "DSI PLL lock failed\n");
+		drm_err(display->drm, "DSI PLL lock failed\n");
 		return;
 	}
 	vlv_cck_put(dev_priv);
 
-	drm_dbg_kms(&dev_priv->drm, "DSI PLL locked\n");
+	drm_dbg_kms(display->drm, "DSI PLL locked\n");
 }
 
 void vlv_dsi_pll_disable(struct intel_encoder *encoder)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 tmp;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	vlv_cck_get(dev_priv);
 
@@ -260,14 +262,14 @@ void vlv_dsi_pll_disable(struct intel_encoder *encoder)
 	vlv_cck_put(dev_priv);
 }
 
-bool bxt_dsi_pll_is_enabled(struct drm_i915_private *dev_priv)
+bool bxt_dsi_pll_is_enabled(struct intel_display *display)
 {
 	bool enabled;
 	u32 val;
 	u32 mask;
 
 	mask = BXT_DSI_PLL_DO_ENABLE | BXT_DSI_PLL_LOCKED;
-	val = intel_de_read(dev_priv, BXT_DSI_PLL_ENABLE);
+	val = intel_de_read(display, BXT_DSI_PLL_ENABLE);
 	enabled = (val & mask) == mask;
 
 	if (!enabled)
@@ -281,17 +283,17 @@ bool bxt_dsi_pll_is_enabled(struct drm_i915_private *dev_priv)
 	 * times, and since accessing DSI registers with invalid dividers
 	 * causes a system hang.
 	 */
-	val = intel_de_read(dev_priv, BXT_DSI_PLL_CTL);
-	if (IS_GEMINILAKE(dev_priv)) {
+	val = intel_de_read(display, BXT_DSI_PLL_CTL);
+	if (display->platform.geminilake) {
 		if (!(val & BXT_DSIA_16X_MASK)) {
-			drm_dbg(&dev_priv->drm,
-				"Invalid PLL divider (%08x)\n", val);
+			drm_dbg_kms(display->drm,
+				    "Invalid PLL divider (%08x)\n", val);
 			enabled = false;
 		}
 	} else {
 		if (!(val & BXT_DSIA_16X_MASK) || !(val & BXT_DSIC_16X_MASK)) {
-			drm_dbg(&dev_priv->drm,
-				"Invalid PLL divider (%08x)\n", val);
+			drm_dbg_kms(display->drm,
+				    "Invalid PLL divider (%08x)\n", val);
 			enabled = false;
 		}
 	}
@@ -301,29 +303,30 @@ bool bxt_dsi_pll_is_enabled(struct drm_i915_private *dev_priv)
 
 void bxt_dsi_pll_disable(struct intel_encoder *encoder)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
-	intel_de_rmw(dev_priv, BXT_DSI_PLL_ENABLE, BXT_DSI_PLL_DO_ENABLE, 0);
+	intel_de_rmw(display, BXT_DSI_PLL_ENABLE, BXT_DSI_PLL_DO_ENABLE, 0);
 
 	/*
 	 * PLL lock should deassert within 200us.
 	 * Wait up to 1ms before timing out.
 	 */
-	if (intel_de_wait_for_clear(dev_priv, BXT_DSI_PLL_ENABLE,
+	if (intel_de_wait_for_clear(display, BXT_DSI_PLL_ENABLE,
 				    BXT_DSI_PLL_LOCKED, 1))
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"Timeout waiting for PLL lock deassertion\n");
 }
 
 u32 vlv_dsi_get_pclk(struct intel_encoder *encoder,
 		     struct intel_crtc_state *config)
 {
+	struct intel_display *display = to_intel_display(encoder);
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 pll_ctl, pll_div;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	vlv_cck_get(dev_priv);
 	pll_ctl = vlv_cck_read(dev_priv, CCK_REG_DSI_PLL_CONTROL);
@@ -352,14 +355,14 @@ static int bxt_dsi_pclk(struct intel_encoder *encoder,
 u32 bxt_dsi_get_pclk(struct intel_encoder *encoder,
 		     struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	u32 pclk;
 
-	config->dsi_pll.ctrl = intel_de_read(dev_priv, BXT_DSI_PLL_CTL);
+	config->dsi_pll.ctrl = intel_de_read(display, BXT_DSI_PLL_CTL);
 
 	pclk = bxt_dsi_pclk(encoder, config);
 
-	drm_dbg(&dev_priv->drm, "Calculated pclk=%u\n", pclk);
+	drm_dbg_kms(display->drm, "Calculated pclk=%u\n", pclk);
 	return pclk;
 }
 
@@ -375,10 +378,9 @@ void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 		       temp | intel_dsi->escape_clk_div << ESCAPE_CLOCK_DIVIDER_SHIFT);
 }
 
-static void glk_dsi_program_esc_clock(struct drm_device *dev,
-				   const struct intel_crtc_state *config)
+static void glk_dsi_program_esc_clock(struct intel_display *display,
+				      const struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 dsi_rate = 0;
 	u32 pll_ratio = 0;
 	u32 ddr_clk = 0;
@@ -415,17 +417,16 @@ static void glk_dsi_program_esc_clock(struct drm_device *dev,
 
 	txesc2_div = min_t(u32, div2_value, 10);
 
-	intel_de_write(dev_priv, MIPIO_TXESC_CLK_DIV1,
+	intel_de_write(display, MIPIO_TXESC_CLK_DIV1,
 		       (1 << (txesc1_div - 1)) & GLK_TX_ESC_CLK_DIV1_MASK);
-	intel_de_write(dev_priv, MIPIO_TXESC_CLK_DIV2,
+	intel_de_write(display, MIPIO_TXESC_CLK_DIV2,
 		       (1 << (txesc2_div - 1)) & GLK_TX_ESC_CLK_DIV2_MASK);
 }
 
 /* Program BXT Mipi clocks and dividers */
-static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port,
+static void bxt_dsi_program_clocks(struct intel_display *display, enum port port,
 				   const struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 tmp;
 	u32 dsi_rate = 0;
 	u32 pll_ratio = 0;
@@ -436,7 +437,7 @@ static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port,
 	u32 mipi_8by3_divider;
 
 	/* Clear old configurations */
-	tmp = intel_de_read(dev_priv, BXT_MIPI_CLOCK_CTL);
+	tmp = intel_de_read(display, BXT_MIPI_CLOCK_CTL);
 	tmp &= ~(BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port));
 	tmp &= ~(BXT_MIPI_RX_ESCLK_UPPER_FIXDIV_MASK(port));
 	tmp &= ~(BXT_MIPI_8X_BY3_DIVIDER_MASK(port));
@@ -472,13 +473,13 @@ static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port,
 	tmp |= BXT_MIPI_RX_ESCLK_LOWER_DIVIDER(port, rx_div_lower);
 	tmp |= BXT_MIPI_RX_ESCLK_UPPER_DIVIDER(port, rx_div_upper);
 
-	intel_de_write(dev_priv, BXT_MIPI_CLOCK_CTL, tmp);
+	intel_de_write(display, BXT_MIPI_CLOCK_CTL, tmp);
 }
 
 int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 			struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	u8 dsi_ratio, dsi_ratio_min, dsi_ratio_max;
 	u32 dsi_clk;
@@ -494,7 +495,7 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 	 */
 	dsi_ratio = DIV_ROUND_UP(dsi_clk * 2, BXT_REF_CLOCK_KHZ);
 
-	if (IS_BROXTON(dev_priv)) {
+	if (display->platform.broxton) {
 		dsi_ratio_min = BXT_DSI_PLL_RATIO_MIN;
 		dsi_ratio_max = BXT_DSI_PLL_RATIO_MAX;
 	} else {
@@ -503,11 +504,11 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 	}
 
 	if (dsi_ratio < dsi_ratio_min || dsi_ratio > dsi_ratio_max) {
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"Can't get a suitable ratio from DSI PLL ratios\n");
 		return -ECHRNG;
 	} else
-		drm_dbg_kms(&dev_priv->drm, "DSI PLL calculation is Done!!\n");
+		drm_dbg_kms(display->drm, "DSI PLL calculation is Done!!\n");
 
 	/*
 	 * Program DSI ratio and Select MIPIC and MIPIA PLL output as 8x
@@ -519,7 +520,7 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 	/* As per recommendation from hardware team,
 	 * Prog PVD ratio =1 if dsi ratio <= 50
 	 */
-	if (IS_BROXTON(dev_priv) && dsi_ratio <= 50)
+	if (display->platform.broxton && dsi_ratio <= 50)
 		config->dsi_pll.ctrl |= BXT_DSI_PLL_PVD_RATIO_1;
 
 	pclk = bxt_dsi_pclk(encoder, config);
@@ -536,46 +537,45 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 void bxt_dsi_pll_enable(struct intel_encoder *encoder,
 			const struct intel_crtc_state *config)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
 
-	drm_dbg_kms(&dev_priv->drm, "\n");
+	drm_dbg_kms(display->drm, "\n");
 
 	/* Configure PLL vales */
-	intel_de_write(dev_priv, BXT_DSI_PLL_CTL, config->dsi_pll.ctrl);
-	intel_de_posting_read(dev_priv, BXT_DSI_PLL_CTL);
+	intel_de_write(display, BXT_DSI_PLL_CTL, config->dsi_pll.ctrl);
+	intel_de_posting_read(display, BXT_DSI_PLL_CTL);
 
 	/* Program TX, RX, Dphy clocks */
-	if (IS_BROXTON(dev_priv)) {
+	if (display->platform.broxton) {
 		for_each_dsi_port(port, intel_dsi->ports)
-			bxt_dsi_program_clocks(encoder->base.dev, port, config);
+			bxt_dsi_program_clocks(display, port, config);
 	} else {
-		glk_dsi_program_esc_clock(encoder->base.dev, config);
+		glk_dsi_program_esc_clock(display, config);
 	}
 
 	/* Enable DSI PLL */
-	intel_de_rmw(dev_priv, BXT_DSI_PLL_ENABLE, 0, BXT_DSI_PLL_DO_ENABLE);
+	intel_de_rmw(display, BXT_DSI_PLL_ENABLE, 0, BXT_DSI_PLL_DO_ENABLE);
 
 	/* Timeout and fail if PLL not locked */
-	if (intel_de_wait_for_set(dev_priv, BXT_DSI_PLL_ENABLE,
+	if (intel_de_wait_for_set(display, BXT_DSI_PLL_ENABLE,
 				  BXT_DSI_PLL_LOCKED, 1)) {
-		drm_err(&dev_priv->drm,
+		drm_err(display->drm,
 			"Timed out waiting for DSI PLL to lock\n");
 		return;
 	}
 
-	drm_dbg_kms(&dev_priv->drm, "DSI PLL locked\n");
+	drm_dbg_kms(display->drm, "DSI PLL locked\n");
 }
 
 void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	u32 tmp;
 
 	/* Clear old configurations */
-	if (IS_BROXTON(dev_priv)) {
+	if (display->platform.broxton) {
 		tmp = intel_de_read(display, BXT_MIPI_CLOCK_CTL);
 		tmp &= ~(BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port));
 		tmp &= ~(BXT_MIPI_RX_ESCLK_UPPER_FIXDIV_MASK(port));
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.h b/drivers/gpu/drm/i915/display/vlv_dsi_pll.h
index f975660fa609..f26e31a7dd69 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.h
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.h
@@ -9,7 +9,6 @@
 #include <linux/types.h>
 
 enum port;
-struct drm_i915_private;
 struct intel_crtc_state;
 struct intel_display;
 struct intel_encoder;
@@ -33,11 +32,11 @@ u32 bxt_dsi_get_pclk(struct intel_encoder *encoder,
 void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port);
 
 #ifdef I915
-bool bxt_dsi_pll_is_enabled(struct drm_i915_private *dev_priv);
+bool bxt_dsi_pll_is_enabled(struct intel_display *display);
 void assert_dsi_pll_enabled(struct intel_display *display);
 void assert_dsi_pll_disabled(struct intel_display *display);
 #else
-static inline bool bxt_dsi_pll_is_enabled(struct drm_i915_private *dev_priv)
+static inline bool bxt_dsi_pll_is_enabled(struct intel_display *display)
 {
 	return false;
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_busy.c b/drivers/gpu/drm/i915/gem/i915_gem_busy.c
index ddda468241ef..6e4d0ce3952f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_busy.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_busy.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c
index 7d97ea2a653e..c4854c5b4e0f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.h b/drivers/gpu/drm/i915/gem/i915_gem_clflush.h
index e6c382973129..9d7ee1579900 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
index ab1af978911b..15835952352e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2011-2012 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.h b/drivers/gpu/drm/i915/gem/i915_gem_context.h
index e5b0f66ea1fe..6e682a6a0574 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
index 67ac2586a0f3..0267c924634b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context_types.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2019 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c
index 9473050ac842..05e440643aa2 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright 2012 Red Hat Inc
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index 75a143d996e0..7a0cc51923b3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 7796c4119ef5..ca7e9216934a 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2008,2010 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.c b/drivers/gpu/drm/i915/gem/i915_gem_internal.c
index ea7561ae6e13..232b984f60b6 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_internal.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ioctls.h b/drivers/gpu/drm/i915/gem/i915_gem_ioctls.h
index 28d6526e32ab..8044d34707b6 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ioctls.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ioctls.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2019 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_lmem.c b/drivers/gpu/drm/i915/gem/i915_gem_lmem.c
index 388f90784d8a..f566191d843b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_lmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_lmem.c
@@ -48,8 +48,7 @@ bool i915_gem_object_is_lmem(struct drm_i915_gem_object *obj)
 	    i915_gem_object_evictable(obj))
 		assert_object_held(obj);
 #endif
-	return mr && (mr->type == INTEL_MEMORY_LOCAL ||
-		      mr->type == INTEL_MEMORY_STOLEN_LOCAL);
+	return mr && intel_memory_type_is_local(mr->type);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
index c3dabb857960..f6d37dff320d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.h b/drivers/gpu/drm/i915/gem/i915_gem_mman.h
index 196417fd0f5c..946fb9825eb3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2019 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 356530b599ce..1f38e367c60b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright © 2017 Intel Corporation
  *
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index a5f34542135c..c34f41605b46 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 68413c05c812..64600aa8227f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index 8780aa243105..7f83f8bdc8fb 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c
index ef85c6dc9fd5..f9e7cab140f8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
index 900c08337942..f0857c5c96df 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2019 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.h b/drivers/gpu/drm/i915/gem/i915_gem_pm.h
index bedf1e95941a..bd5bd2c5e7f9 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pm.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2019 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index ae3343c81a64..3b81cc3ad5be 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index aec41f0f098f..b81e67504bbe 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2008-2015 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
index 9d958a6f377e..3380151edfc1 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2008-2012 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
index af85d0c28168..8814cbcde5b5 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2014-2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
index 5ac23ff3feff..5a296ba3758a 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2008 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
index 09b68713ab32..307a18eede72 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
@@ -1,9 +1,8 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2012-2014 Intel Corporation
  *
-  * Based on amdgpu_mn, which bears the following notice:
+ * Based on amdgpu_mn, which bears the following notice:
  *
  * Copyright 2014 Advanced Micro Devices, Inc.
  * All Rights Reserved.
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_wait.c b/drivers/gpu/drm/i915/gem/i915_gem_wait.c
index 1f55e62044a4..7127e90c1a8f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_wait.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_wait.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2016 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
index 46b9a17d6abc..65d84a93c525 100644
--- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
+++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
@@ -1,6 +1,5 @@
+// SPDX-License-Identifier: MIT
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2017 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.h b/drivers/gpu/drm/i915/gem/i915_gemfs.h
index 5d835e44c4f6..16d4333c9a4e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gemfs.h
+++ b/drivers/gpu/drm/i915/gem/i915_gemfs.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2017 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
index 804f74084bd4..9c3f17e51885 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
@@ -1837,6 +1837,8 @@ static int igt_mmap_revoke(void *arg)
 
 int i915_gem_mman_live_selftests(struct drm_i915_private *i915)
 {
+	int ret;
+	bool unuse_mm = false;
 	static const struct i915_subtest tests[] = {
 		SUBTEST(igt_partial_tiling),
 		SUBTEST(igt_smoke_tiling),
@@ -1848,5 +1850,15 @@ int i915_gem_mman_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_mmap_gpu),
 	};
 
-	return i915_live_subtests(tests, i915);
+	if (!current->mm) {
+		kthread_use_mm(current->active_mm);
+		unuse_mm = true;
+	}
+
+	ret = i915_live_subtests(tests, i915);
+
+	if (unuse_mm)
+		kthread_unuse_mm(current->active_mm);
+
+	return ret;
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index f6c59f20832f..46a5aa4ab9c8 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -289,6 +289,14 @@ u64 gen8_ggtt_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
+static dma_addr_t gen8_ggtt_pte_decode(u64 pte, bool *is_present, bool *is_local)
+{
+	*is_present = pte & GEN8_PAGE_PRESENT;
+	*is_local = pte & GEN12_GGTT_PTE_LM;
+
+	return pte & GEN12_GGTT_PTE_ADDR_MASK;
+}
+
 static bool should_update_ggtt_with_bind(struct i915_ggtt *ggtt)
 {
 	struct intel_gt *gt = ggtt->vm.gt;
@@ -435,6 +443,11 @@ static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte)
 	writeq(pte, addr);
 }
 
+static gen8_pte_t gen8_get_pte(void __iomem *addr)
+{
+	return readq(addr);
+}
+
 static void gen8_ggtt_insert_page(struct i915_address_space *vm,
 				  dma_addr_t addr,
 				  u64 offset,
@@ -450,6 +463,16 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm,
 	ggtt->invalidate(ggtt);
 }
 
+static dma_addr_t gen8_ggtt_read_entry(struct i915_address_space *vm,
+				       u64 offset, bool *is_present, bool *is_local)
+{
+	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
+	gen8_pte_t __iomem *pte =
+		(gen8_pte_t __iomem *)ggtt->gsm + offset / I915_GTT_PAGE_SIZE;
+
+	return ggtt->vm.pte_decode(gen8_get_pte(pte), is_present, is_local);
+}
+
 static void gen8_ggtt_insert_page_bind(struct i915_address_space *vm,
 				       dma_addr_t addr, u64 offset,
 				       unsigned int pat_index, u32 flags)
@@ -605,6 +628,17 @@ static void gen6_ggtt_insert_page(struct i915_address_space *vm,
 	ggtt->invalidate(ggtt);
 }
 
+static dma_addr_t gen6_ggtt_read_entry(struct i915_address_space *vm,
+				       u64 offset,
+				       bool *is_present, bool *is_local)
+{
+	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
+	gen6_pte_t __iomem *pte =
+		(gen6_pte_t __iomem *)ggtt->gsm + offset / I915_GTT_PAGE_SIZE;
+
+	return vm->pte_decode(ioread32(pte), is_present, is_local);
+}
+
 /*
  * Binds an object into the global gtt with the specified cache level.
  * The object will be accessible to the GPU via commands whose operands
@@ -769,6 +803,14 @@ void intel_ggtt_unbind_vma(struct i915_address_space *vm,
 	vm->clear_range(vm, vma_res->start, vma_res->vma_size);
 }
 
+dma_addr_t intel_ggtt_read_entry(struct i915_address_space *vm,
+				 u64 offset, bool *is_present, bool *is_local)
+{
+	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
+
+	return ggtt->vm.read_entry(vm, offset, is_present, is_local);
+}
+
 /*
  * Reserve the top of the GuC address space for firmware images. Addresses
  * beyond GUC_GGTT_TOP in the GuC address space are inaccessible by GuC,
@@ -1245,6 +1287,7 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 	ggtt->vm.scratch_range = gen8_ggtt_clear_range;
 
 	ggtt->vm.insert_entries = gen8_ggtt_insert_entries;
+	ggtt->vm.read_entry = gen8_ggtt_read_entry;
 
 	/*
 	 * Serialize GTT updates with aperture access on BXT if VT-d is on,
@@ -1291,6 +1334,8 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 	else
 		ggtt->vm.pte_encode = gen8_ggtt_pte_encode;
 
+	ggtt->vm.pte_decode = gen8_ggtt_pte_decode;
+
 	return ggtt_probe_common(ggtt, size);
 }
 
@@ -1390,6 +1435,14 @@ static u64 iris_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
+static dma_addr_t gen6_pte_decode(u64 pte, bool *is_present, bool *is_local)
+{
+	*is_present = pte & GEN6_PTE_VALID;
+	*is_local = false;
+
+	return ((pte & 0xff0) << 28) | (pte & ~0xfff);
+}
+
 static int gen6_gmch_probe(struct i915_ggtt *ggtt)
 {
 	struct drm_i915_private *i915 = ggtt->vm.i915;
@@ -1428,6 +1481,7 @@ static int gen6_gmch_probe(struct i915_ggtt *ggtt)
 	ggtt->vm.scratch_range = gen6_ggtt_clear_range;
 	ggtt->vm.insert_page = gen6_ggtt_insert_page;
 	ggtt->vm.insert_entries = gen6_ggtt_insert_entries;
+	ggtt->vm.read_entry = gen6_ggtt_read_entry;
 	ggtt->vm.cleanup = gen6_gmch_remove;
 
 	ggtt->invalidate = gen6_ggtt_invalidate;
@@ -1443,6 +1497,8 @@ static int gen6_gmch_probe(struct i915_ggtt *ggtt)
 	else
 		ggtt->vm.pte_encode = snb_pte_encode;
 
+	ggtt->vm.pte_decode = gen6_pte_decode;
+
 	ggtt->vm.vma_ops.bind_vma    = intel_ggtt_bind_vma;
 	ggtt->vm.vma_ops.unbind_vma  = intel_ggtt_unbind_vma;
 
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c b/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c
index 59eed0a0ce90..c5f5f0bdfb2c 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c
@@ -27,6 +27,13 @@ static void gmch_ggtt_insert_page(struct i915_address_space *vm,
 	intel_gmch_gtt_insert_page(addr, offset >> PAGE_SHIFT, flags);
 }
 
+static dma_addr_t gmch_ggtt_read_entry(struct i915_address_space *vm,
+				       u64 offset, bool *is_present, bool *is_local)
+{
+	return intel_gmch_gtt_read_entry(offset >> PAGE_SHIFT,
+					 is_present, is_local);
+}
+
 static void gmch_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct i915_vma_resource *vma_res,
 				     unsigned int pat_index,
@@ -103,6 +110,7 @@ int intel_ggtt_gmch_probe(struct i915_ggtt *ggtt)
 	ggtt->vm.insert_entries = gmch_ggtt_insert_entries;
 	ggtt->vm.clear_range = gmch_ggtt_clear_range;
 	ggtt->vm.scratch_range = gmch_ggtt_clear_range;
+	ggtt->vm.read_entry = gmch_ggtt_read_entry;
 	ggtt->vm.cleanup = gmch_ggtt_remove;
 
 	ggtt->invalidate = gmch_ggtt_invalidate;
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c
index 30b128b1fde7..afbc5c769308 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.c
@@ -176,7 +176,6 @@ static void clear_vm_list(struct list_head *list)
 			i915_vma_destroy_locked(vma);
 			i915_gem_object_put(obj);
 		}
-
 	}
 }
 
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h
index 0a36ea751b63..9d3a3ad567a0 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.h
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.h
@@ -312,6 +312,7 @@ struct i915_address_space {
 	u64 (*pte_encode)(dma_addr_t addr,
 			  unsigned int pat_index,
 			  u32 flags); /* Create a valid PTE */
+	dma_addr_t (*pte_decode)(u64 pte, bool *is_present, bool *is_local);
 #define PTE_READ_ONLY	BIT(0)
 #define PTE_LM		BIT(1)
 
@@ -340,6 +341,8 @@ struct i915_address_space {
 				   struct i915_vma_resource *vma_res,
 				   unsigned int pat_index,
 				   u32 flags);
+	dma_addr_t (*read_entry)(struct i915_address_space *vm,
+				 u64 offset, bool *is_present, bool *is_local);
 	void (*cleanup)(struct i915_address_space *vm);
 
 	void (*foreach)(struct i915_address_space *vm,
@@ -590,6 +593,9 @@ void intel_ggtt_bind_vma(struct i915_address_space *vm,
 void intel_ggtt_unbind_vma(struct i915_address_space *vm,
 			   struct i915_vma_resource *vma_res);
 
+dma_addr_t intel_ggtt_read_entry(struct i915_address_space *vm,
+				 u64 offset, bool *is_present, bool *is_local);
+
 int i915_ggtt_probe_hw(struct drm_i915_private *i915);
 int i915_ggtt_init_hw(struct drm_i915_private *i915);
 int i915_ggtt_enable_hw(struct drm_i915_private *i915);
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 51847a846002..c481b56fa67d 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -751,7 +751,6 @@ static int lrc_ring_indirect_offset(const struct intel_engine_cs *engine)
 
 static int lrc_ring_cmd_buf_cctl(const struct intel_engine_cs *engine)
 {
-
 	if (GRAPHICS_VER_FULL(engine->i915) >= IP_VER(12, 55))
 		/*
 		 * Note that the CSFE context has a dummy slot for CMD_BUF_CCTL
diff --git a/drivers/gpu/drm/i915/gt/intel_mocs.c b/drivers/gpu/drm/i915/gt/intel_mocs.c
index cf41d325712e..5dd8121f4b15 100644
--- a/drivers/gpu/drm/i915/gt/intel_mocs.c
+++ b/drivers/gpu/drm/i915/gt/intel_mocs.c
@@ -314,7 +314,6 @@ static const struct drm_i915_mocs_entry icl_mocs_table[] = {
 };
 
 static const struct drm_i915_mocs_entry dg1_mocs_table[] = {
-
 	/* UC */
 	MOCS_ENTRY(1, 0, L3_1_UC),
 	/* WB - L3 */
diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c
index 9378d5901c49..9ca42589da4d 100644
--- a/drivers/gpu/drm/i915/gt/intel_rc6.c
+++ b/drivers/gpu/drm/i915/gt/intel_rc6.c
@@ -117,21 +117,10 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6)
 			GEN6_RC_CTL_RC6_ENABLE |
 			GEN6_RC_CTL_EI_MODE(1);
 
-	/*
-	 * BSpec 52698 - Render powergating must be off.
-	 * FIXME BSpec is outdated, disabling powergating for MTL is just
-	 * temporary wa and should be removed after fixing real cause
-	 * of forcewake timeouts.
-	 */
-	if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74)))
-		pg_enable =
-			GEN9_MEDIA_PG_ENABLE |
-			GEN11_MEDIA_SAMPLER_PG_ENABLE;
-	else
-		pg_enable =
-			GEN9_RENDER_PG_ENABLE |
-			GEN9_MEDIA_PG_ENABLE |
-			GEN11_MEDIA_SAMPLER_PG_ENABLE;
+	pg_enable =
+		GEN9_RENDER_PG_ENABLE |
+		GEN9_MEDIA_PG_ENABLE |
+		GEN11_MEDIA_SAMPLER_PG_ENABLE;
 
 	if (GRAPHICS_VER(gt->i915) >= 12 && !IS_DG1(gt->i915)) {
 		for (i = 0; i < I915_MAX_VCS; i++)
diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
index 6e9977b2d180..a876a34455f1 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
@@ -365,7 +365,13 @@ static void reset_prepare(struct intel_engine_cs *engine)
 			     ENGINE_READ_FW(engine, RING_HEAD),
 			     ENGINE_READ_FW(engine, RING_TAIL),
 			     ENGINE_READ_FW(engine, RING_START));
-		if (!stop_ring(engine)) {
+		/*
+		 * Sometimes engine head failed to set to zero even after writing into it.
+		 * Use wait_for_atomic() with 20ms delay to let engine resumes from
+		 * correct RING_HEAD. Experimented different values and determined
+		 * that 20ms works best based on testing.
+		 */
+		if (wait_for_atomic((!stop_ring(engine) == 0), 20)) {
 			drm_err(&engine->i915->drm,
 				"failed to set %s head to zero "
 				"ctl %08x head %08x tail %08x start %08x\n",
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index 64e9317f58fb..5abc5fcc2514 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -8,7 +8,7 @@
 #include <drm/intel/i915_drm.h>
 
 #include "display/intel_display.h"
-#include "display/intel_display_irq.h"
+#include "display/intel_display_rps.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
@@ -550,6 +550,7 @@ static unsigned int init_emon(struct intel_uncore *uncore)
 static bool gen5_rps_enable(struct intel_rps *rps)
 {
 	struct drm_i915_private *i915 = rps_to_i915(rps);
+	struct intel_display *display = &i915->display;
 	struct intel_uncore *uncore = rps_to_uncore(rps);
 	u8 fstart, vstart;
 	u32 rgvmodectl;
@@ -607,9 +608,7 @@ static bool gen5_rps_enable(struct intel_rps *rps)
 	rps->ips.last_count2 = intel_uncore_read(uncore, GFXEC);
 	rps->ips.last_time2 = ktime_get_raw_ns();
 
-	spin_lock(&i915->irq_lock);
-	ilk_enable_display_irq(i915, DE_PCU_EVENT);
-	spin_unlock(&i915->irq_lock);
+	ilk_display_rps_enable(display);
 
 	spin_unlock_irq(&mchdev_lock);
 
@@ -621,14 +620,13 @@ static bool gen5_rps_enable(struct intel_rps *rps)
 static void gen5_rps_disable(struct intel_rps *rps)
 {
 	struct drm_i915_private *i915 = rps_to_i915(rps);
+	struct intel_display *display = &i915->display;
 	struct intel_uncore *uncore = rps_to_uncore(rps);
 	u16 rgvswctl;
 
 	spin_lock_irq(&mchdev_lock);
 
-	spin_lock(&i915->irq_lock);
-	ilk_disable_display_irq(i915, DE_PCU_EVENT);
-	spin_unlock(&i915->irq_lock);
+	ilk_display_rps_disable(display);
 
 	rgvswctl = intel_uncore_read16(uncore, MEMSWCTL);
 
@@ -1001,6 +999,10 @@ void intel_rps_dec_waiters(struct intel_rps *rps)
 	if (rps_uses_slpc(rps)) {
 		slpc = rps_to_slpc(rps);
 
+		/* Don't decrement num_waiters for req where increment was skipped */
+		if (slpc->power_profile == SLPC_POWER_PROFILES_POWER_SAVING)
+			return;
+
 		intel_guc_slpc_dec_waiters(slpc);
 	} else {
 		atomic_dec(&rps->num_waiters);
@@ -1029,11 +1031,15 @@ void intel_rps_boost(struct i915_request *rq)
 			if (slpc->power_profile == SLPC_POWER_PROFILES_POWER_SAVING)
 				return;
 
-			if (slpc->min_freq_softlimit >= slpc->boost_freq)
-				return;
-
 			/* Return if old value is non zero */
 			if (!atomic_fetch_inc(&slpc->num_waiters)) {
+				/*
+				 * Skip queuing boost work if frequency is already boosted,
+				 * but still increment num_waiters.
+				 */
+				if (slpc->min_freq_softlimit >= slpc->boost_freq)
+					return;
+
 				GT_TRACE(rps_to_gt(rps), "boost fence:%llx:%llx\n",
 					 rq->fence.context, rq->fence.seqno);
 				queue_work(rps_to_gt(rps)->i915->unordered_wq,
diff --git a/drivers/gpu/drm/i915/gt/intel_rps_types.h b/drivers/gpu/drm/i915/gt/intel_rps_types.h
index 5135b90a2a40..ece445109305 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_rps_types.h
@@ -57,7 +57,7 @@ struct intel_rps {
 
 	/*
 	 * work, interrupts_enabled and pm_iir are protected by
-	 * i915->irq_lock
+	 * gt->irq_lock
 	 */
 	struct timer_list timer;
 	struct work_struct work;
diff --git a/drivers/gpu/drm/i915/gt/intel_wopcm.h b/drivers/gpu/drm/i915/gt/intel_wopcm.h
index 17d6aa86008a..d2038b6de5e7 100644
--- a/drivers/gpu/drm/i915/gt/intel_wopcm.h
+++ b/drivers/gpu/drm/i915/gt/intel_wopcm.h
@@ -1,6 +1,5 @@
+/* SPDX-License-Identifier: MIT */
 /*
- * SPDX-License-Identifier: MIT
- *
  * Copyright © 2017-2018 Intel Corporation
  */
 
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 116683ebe074..b37e400f74e5 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -156,7 +156,7 @@ static void _wa_add(struct i915_wa_list *wal, const struct i915_wa *wa)
 	if (IS_ALIGNED(wal->count, grow)) { /* Either uninitialized or full. */
 		struct i915_wa *list;
 
-		list = kmalloc_array(ALIGN(wal->count + 1, grow), sizeof(*wa),
+		list = kmalloc_array(ALIGN(wal->count + 1, grow), sizeof(*list),
 				     GFP_KERNEL);
 		if (!list) {
 			drm_err(&i915->drm, "No space for workaround init!\n");
diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c
index 22e750108c5f..23f04f6f8fba 100644
--- a/drivers/gpu/drm/i915/gt/selftest_lrc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c
@@ -7,6 +7,7 @@
 
 #include "gem/i915_gem_internal.h"
 
+#include "i915_drv.h"
 #include "i915_selftest.h"
 #include "intel_engine_heartbeat.h"
 #include "intel_engine_pm.h"
@@ -859,6 +860,14 @@ static int live_lrc_timestamp(void *arg)
 	};
 
 	/*
+	 * This test was designed to isolate a hardware bug.
+	 * The bug was found and fixed in future generations but
+	 * now the test pollutes our CI on previous generation.
+	 */
+	if (GRAPHICS_VER(gt->i915) == 12)
+		return 0;
+
+	/*
 	 * We want to verify that the timestamp is saved and restore across
 	 * context switches and is monotonic.
 	 *
diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c
index 908483ab0bc8..41716ed454b7 100644
--- a/drivers/gpu/drm/i915/gt/selftest_rc6.c
+++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c
@@ -33,15 +33,22 @@ int live_rc6_manual(void *arg)
 {
 	struct intel_gt *gt = arg;
 	struct intel_rc6 *rc6 = &gt->rc6;
-	u64 rc0_power, rc6_power;
+	struct intel_rps *rps = &gt->rps;
 	intel_wakeref_t wakeref;
+	u64 rc0_sample_energy[2];
+	u64 rc6_sample_energy[2];
+	u64 sleep_time = 1000;
+	u32 rc0_freq = 0;
+	u32 rc6_freq = 0;
+	u64 rc0_power;
+	u64 rc6_power;
 	bool has_power;
+	u64 threshold;
 	ktime_t dt;
 	u64 res[2];
 	int err = 0;
-	u32 rc0_freq = 0;
-	u32 rc6_freq = 0;
-	struct intel_rps *rps = &gt->rps;
+	u64 diff;
+
 
 	/*
 	 * Our claim is that we can "encourage" the GPU to enter rc6 at will.
@@ -60,14 +67,15 @@ int live_rc6_manual(void *arg)
 
 	/* Force RC6 off for starters */
 	__intel_rc6_disable(rc6);
-	msleep(1); /* wakeup is not immediate, takes about 100us on icl */
+	/* wakeup is not immediate, takes about 100us on icl */
+	usleep_range(1000, 2000);
 
 	res[0] = rc6_residency(rc6);
 
 	dt = ktime_get();
-	rc0_power = librapl_energy_uJ();
-	msleep(1000);
-	rc0_power = librapl_energy_uJ() - rc0_power;
+	rc0_sample_energy[0] = librapl_energy_uJ();
+	msleep(sleep_time);
+	rc0_sample_energy[1] = librapl_energy_uJ() - rc0_sample_energy[0];
 	dt = ktime_sub(ktime_get(), dt);
 	res[1] = rc6_residency(rc6);
 	rc0_freq = intel_rps_read_actual_frequency_fw(rps);
@@ -79,11 +87,12 @@ int live_rc6_manual(void *arg)
 	}
 
 	if (has_power) {
-		rc0_power = div64_u64(NSEC_PER_SEC * rc0_power,
+		rc0_power = div64_u64(NSEC_PER_SEC * rc0_sample_energy[1],
 				      ktime_to_ns(dt));
+
 		if (!rc0_power) {
 			if (rc0_freq)
-				pr_debug("No power measured while in RC0! GPU Freq: %u in RC0\n",
+				pr_debug("No power measured while in RC0! GPU Freq: %uMHz in RC0\n",
 					 rc0_freq);
 			else
 				pr_err("No power and freq measured while in RC0\n");
@@ -98,10 +107,10 @@ int live_rc6_manual(void *arg)
 	res[0] = rc6_residency(rc6);
 	intel_uncore_forcewake_flush(rc6_to_uncore(rc6), FORCEWAKE_ALL);
 	dt = ktime_get();
-	rc6_power = librapl_energy_uJ();
-	msleep(1000);
+	rc6_sample_energy[0] = librapl_energy_uJ();
+	msleep(sleep_time);
 	rc6_freq = intel_rps_read_actual_frequency_fw(rps);
-	rc6_power = librapl_energy_uJ() - rc6_power;
+	rc6_sample_energy[1] = librapl_energy_uJ() - rc6_sample_energy[0];
 	dt = ktime_sub(ktime_get(), dt);
 	res[1] = rc6_residency(rc6);
 	if (res[1] == res[0]) {
@@ -113,13 +122,24 @@ int live_rc6_manual(void *arg)
 	}
 
 	if (has_power) {
-		rc6_power = div64_u64(NSEC_PER_SEC * rc6_power,
+		rc6_power = div64_u64(NSEC_PER_SEC * rc6_sample_energy[1],
 				      ktime_to_ns(dt));
-		pr_info("GPU consumed %llduW in RC0 and %llduW in RC6\n",
+		pr_info("GPU consumed %lluuW in RC0 and %lluuW in RC6\n",
 			rc0_power, rc6_power);
+
 		if (2 * rc6_power > rc0_power) {
-			pr_err("GPU leaked energy while in RC6! GPU Freq: %u in RC6 and %u in RC0\n",
-			       rc6_freq, rc0_freq);
+			pr_err("GPU leaked energy while in RC6!\n"
+			       "GPU Freq: %uMHz in RC6 and %uMHz in RC0\n"
+			       "RC0 energy before & after sleep respectively: %lluuJ %lluuJ\n"
+			       "RC6 energy before & after sleep respectively: %lluuJ %lluuJ\n",
+			       rc6_freq, rc0_freq, rc0_sample_energy[0], rc0_sample_energy[1],
+			       rc6_sample_energy[0], rc6_sample_energy[1]);
+
+			diff = res[1] - res[0];
+			threshold = (9 * NSEC_PER_MSEC * sleep_time) / 10;
+			if (diff < threshold)
+				pr_err("Did not enter RC6 properly, RC6 start residency=%lluns, RC6 end residency=%lluns\n",
+				       res[0], res[1]);
 			err = -EINVAL;
 			goto out_unlock;
 		}
diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c
index 3941f2d6fa47..69ed946a39e5 100644
--- a/drivers/gpu/drm/i915/gt/selftest_tlb.c
+++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c
@@ -143,7 +143,7 @@ pte_tlbinv(struct intel_context *ce,
 	if (ce->engine->class == OTHER_CLASS)
 		msleep(200);
 	else
-		msleep(10);
+		usleep_range(10000, 20000);
 
 	if (va == vb) {
 		if (!i915_request_completed(rq)) {
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
index 5dc0ccd07636..d550eb6edfb8 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
@@ -230,7 +230,7 @@ int intel_gsc_fw_get_binary_info(struct intel_uc_fw *gsc_fw, const void *data, s
 		gt_info(gt, "Invalid GSC firmware for MTL/ARL, got %d.%d.%d.%d but need 102.x.x.x",
 			gsc->release.major, gsc->release.minor,
 			gsc->release.patch, gsc->release.build);
-			return -EINVAL;
+		return -EINVAL;
 	}
 
 	if (min_ver.major) {
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index d791f9baa11d..456d3372eef8 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -317,6 +317,11 @@ void intel_huc_init_early(struct intel_huc *huc)
 	}
 }
 
+void intel_huc_fini_late(struct intel_huc *huc)
+{
+	delayed_huc_load_fini(huc);
+}
+
 #define HUC_LOAD_MODE_STRING(x) (x ? "GSC" : "legacy")
 static int check_huc_loading_mode(struct intel_huc *huc)
 {
@@ -414,12 +419,6 @@ out:
 
 void intel_huc_fini(struct intel_huc *huc)
 {
-	/*
-	 * the fence is initialized in init_early, so we need to clean it up
-	 * even if HuC loading is off.
-	 */
-	delayed_huc_load_fini(huc);
-
 	if (huc->heci_pkt)
 		i915_vma_unpin_and_release(&huc->heci_pkt, 0);
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
index d5e441b9e08d..921ad4b1687f 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
@@ -55,6 +55,7 @@ struct intel_huc {
 
 int intel_huc_sanitize(struct intel_huc *huc);
 void intel_huc_init_early(struct intel_huc *huc);
+void intel_huc_fini_late(struct intel_huc *huc);
 int intel_huc_init(struct intel_huc *huc);
 void intel_huc_fini(struct intel_huc *huc);
 int intel_huc_auth(struct intel_huc *huc, enum intel_huc_authentication_type type);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
index 90ba1b0b4c9d..4a3493e8d433 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
@@ -136,6 +136,7 @@ void intel_uc_init_late(struct intel_uc *uc)
 
 void intel_uc_driver_late_release(struct intel_uc *uc)
 {
+	intel_huc_fini_late(&uc->huc);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c
index eedd1865bb98..62d14f82256f 100644
--- a/drivers/gpu/drm/i915/gvt/aperture_gm.c
+++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c
@@ -46,6 +46,7 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm)
 	unsigned int flags;
 	u64 start, end, size;
 	struct drm_mm_node *node;
+	intel_wakeref_t wakeref;
 	int ret;
 
 	if (high_gm) {
@@ -63,12 +64,12 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm)
 	}
 
 	mutex_lock(&gt->ggtt->vm.mutex);
-	mmio_hw_access_pre(gt);
+	wakeref = mmio_hw_access_pre(gt);
 	ret = i915_gem_gtt_insert(&gt->ggtt->vm, NULL, node,
 				  size, I915_GTT_PAGE_SIZE,
 				  I915_COLOR_UNEVICTABLE,
 				  start, end, flags);
-	mmio_hw_access_post(gt);
+	mmio_hw_access_post(gt, wakeref);
 	mutex_unlock(&gt->ggtt->vm.mutex);
 	if (ret)
 		gvt_err("fail to alloc %s gm space from host\n",
@@ -226,7 +227,7 @@ out_free_fence:
 		vgpu->fence.regs[i] = NULL;
 	}
 	mutex_unlock(&gvt->gt->ggtt->vm.mutex);
-	intel_runtime_pm_put_unchecked(uncore->rpm);
+	intel_runtime_pm_put(uncore->rpm, wakeref);
 	return -ENOSPC;
 }
 
diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c
index baccbf1761b7..673534f061ef 100644
--- a/drivers/gpu/drm/i915/gvt/debugfs.c
+++ b/drivers/gpu/drm/i915/gvt/debugfs.c
@@ -91,16 +91,17 @@ static int vgpu_mmio_diff_show(struct seq_file *s, void *unused)
 		.diff = 0,
 	};
 	struct diff_mmio *node, *next;
+	intel_wakeref_t wakeref;
 
 	INIT_LIST_HEAD(&param.diff_mmio_list);
 
 	mutex_lock(&gvt->lock);
 	spin_lock_bh(&gvt->scheduler.mmio_context_lock);
 
-	mmio_hw_access_pre(gvt->gt);
+	wakeref = mmio_hw_access_pre(gvt->gt);
 	/* Recognize all the diff mmios to list. */
 	intel_gvt_for_each_tracked_mmio(gvt, mmio_diff_handler, &param);
-	mmio_hw_access_post(gvt->gt);
+	mmio_hw_access_post(gvt->gt, wakeref);
 
 	spin_unlock_bh(&gvt->scheduler.mmio_context_lock);
 	mutex_unlock(&gvt->lock);
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index 2fa7ca19ba5d..ae9b0ded3651 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -220,9 +220,11 @@ static u64 read_pte64(struct i915_ggtt *ggtt, unsigned long index)
 
 static void ggtt_invalidate(struct intel_gt *gt)
 {
-	mmio_hw_access_pre(gt);
+	intel_wakeref_t wakeref;
+
+	wakeref = mmio_hw_access_pre(gt);
 	intel_uncore_write(gt->uncore, GFX_FLSH_CNTL_GEN6, GFX_FLSH_CNTL_EN);
-	mmio_hw_access_post(gt);
+	mmio_hw_access_post(gt, wakeref);
 }
 
 static void write_pte64(struct i915_ggtt *ggtt, unsigned long index, u64 pte)
diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h
index 01d890999f25..1d10c16e6465 100644
--- a/drivers/gpu/drm/i915/gvt/gvt.h
+++ b/drivers/gpu/drm/i915/gvt/gvt.h
@@ -570,14 +570,15 @@ enum {
 	GVT_FAILSAFE_GUEST_ERR,
 };
 
-static inline void mmio_hw_access_pre(struct intel_gt *gt)
+static inline intel_wakeref_t mmio_hw_access_pre(struct intel_gt *gt)
 {
-	intel_runtime_pm_get(gt->uncore->rpm);
+	return intel_runtime_pm_get(gt->uncore->rpm);
 }
 
-static inline void mmio_hw_access_post(struct intel_gt *gt)
+static inline void mmio_hw_access_post(struct intel_gt *gt,
+				       intel_wakeref_t wakeref)
 {
-	intel_runtime_pm_put_unchecked(gt->uncore->rpm);
+	intel_runtime_pm_put(gt->uncore->rpm, wakeref);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 4efee6797873..1344e6d20a34 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -56,6 +56,7 @@
 #include "display/intel_pps_regs.h"
 #include "display/intel_psr_regs.h"
 #include "display/intel_sprite_regs.h"
+#include "display/intel_vga_regs.h"
 #include "display/skl_universal_plane_regs.h"
 #include "display/skl_watermark_regs.h"
 #include "display/vlv_dsi_pll_regs.h"
@@ -264,6 +265,7 @@ static int fence_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
 {
 	struct intel_gvt *gvt = vgpu->gvt;
 	unsigned int fence_num = offset_to_fence_num(off);
+	intel_wakeref_t wakeref;
 	int ret;
 
 	ret = sanitize_fence_mmio_access(vgpu, fence_num, p_data, bytes);
@@ -271,10 +273,10 @@ static int fence_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
 		return ret;
 	write_vreg(vgpu, off, p_data, bytes);
 
-	mmio_hw_access_pre(gvt->gt);
+	wakeref = mmio_hw_access_pre(gvt->gt);
 	intel_vgpu_write_fence(vgpu, fence_num,
 			vgpu_vreg64(vgpu, fence_num_to_offset(fence_num)));
-	mmio_hw_access_post(gvt->gt);
+	mmio_hw_access_post(gvt->gt, wakeref);
 	return 0;
 }
 
@@ -513,7 +515,7 @@ static u32 bdw_vgpu_get_dp_bitrate(struct intel_vgpu *vgpu, enum port port)
 
 		switch (wrpll_ctl & WRPLL_REF_MASK) {
 		case WRPLL_REF_PCH_SSC:
-			refclk = vgpu->gvt->gt->i915->display.dpll.ref_clks.ssc;
+			refclk = 135000;
 			break;
 		case WRPLL_REF_LCPLL:
 			refclk = 2700000;
@@ -544,7 +546,7 @@ out:
 static u32 bxt_vgpu_get_dp_bitrate(struct intel_vgpu *vgpu, enum port port)
 {
 	u32 dp_br = 0;
-	int refclk = vgpu->gvt->gt->i915->display.dpll.ref_clks.nssc;
+	int refclk = 100000;
 	enum dpio_phy phy = DPIO_PHY0;
 	enum dpio_channel ch = DPIO_CH0;
 	struct dpll clock = {};
@@ -1975,10 +1977,12 @@ static int mmio_read_from_hw(struct intel_vgpu *vgpu,
 	    vgpu == gvt->scheduler.engine_owner[engine->id] ||
 	    offset == i915_mmio_reg_offset(RING_TIMESTAMP(engine->mmio_base)) ||
 	    offset == i915_mmio_reg_offset(RING_TIMESTAMP_UDW(engine->mmio_base))) {
-		mmio_hw_access_pre(gvt->gt);
+		intel_wakeref_t wakeref;
+
+		wakeref = mmio_hw_access_pre(gvt->gt);
 		vgpu_vreg(vgpu, offset) =
 			intel_uncore_read(gvt->gt->uncore, _MMIO(offset));
-		mmio_hw_access_post(gvt->gt);
+		mmio_hw_access_post(gvt->gt, wakeref);
 	}
 
 	return intel_vgpu_default_mmio_read(vgpu, offset, p_data, bytes);
@@ -3209,10 +3213,12 @@ void intel_gvt_restore_fence(struct intel_gvt *gvt)
 	int i, id;
 
 	idr_for_each_entry(&(gvt)->vgpu_idr, vgpu, id) {
-		mmio_hw_access_pre(gvt->gt);
+		intel_wakeref_t wakeref;
+
+		wakeref = mmio_hw_access_pre(gvt->gt);
 		for (i = 0; i < vgpu_fence_sz(vgpu); i++)
 			intel_vgpu_write_fence(vgpu, i, vgpu_vreg64(vgpu, fence_num_to_offset(i)));
-		mmio_hw_access_post(gvt->gt);
+		mmio_hw_access_post(gvt->gt, wakeref);
 	}
 }
 
@@ -3233,8 +3239,10 @@ void intel_gvt_restore_mmio(struct intel_gvt *gvt)
 	int id;
 
 	idr_for_each_entry(&(gvt)->vgpu_idr, vgpu, id) {
-		mmio_hw_access_pre(gvt->gt);
+		intel_wakeref_t wakeref;
+
+		wakeref = mmio_hw_access_pre(gvt->gt);
 		intel_gvt_for_each_tracked_mmio(gvt, mmio_pm_restore_handler, vgpu);
-		mmio_hw_access_post(gvt->gt);
+		mmio_hw_access_post(gvt->gt, wakeref);
 	}
 }
diff --git a/drivers/gpu/drm/i915/gvt/opregion.c b/drivers/gpu/drm/i915/gvt/opregion.c
index 509f9ccae3a9..dbad4d853d3a 100644
--- a/drivers/gpu/drm/i915/gvt/opregion.c
+++ b/drivers/gpu/drm/i915/gvt/opregion.c
@@ -222,7 +222,6 @@ int intel_vgpu_init_opregion(struct intel_vgpu *vgpu)
 	u8 *buf;
 	struct opregion_header *header;
 	struct vbt v;
-	const char opregion_signature[16] = OPREGION_SIGNATURE;
 
 	gvt_dbg_core("init vgpu%d opregion\n", vgpu->id);
 	vgpu_opregion(vgpu)->va = (void *)__get_free_pages(GFP_KERNEL |
@@ -236,8 +235,10 @@ int intel_vgpu_init_opregion(struct intel_vgpu *vgpu)
 	/* emulated opregion with VBT mailbox only */
 	buf = (u8 *)vgpu_opregion(vgpu)->va;
 	header = (struct opregion_header *)buf;
-	memcpy(header->signature, opregion_signature,
-	       sizeof(opregion_signature));
+
+	static_assert(sizeof(header->signature) == sizeof(OPREGION_SIGNATURE) - 1);
+	memcpy(header->signature, OPREGION_SIGNATURE, sizeof(header->signature));
+
 	header->size = 0x8;
 	header->opregion_ver = 0x02000000;
 	header->mboxes = MBOX_VBT;
diff --git a/drivers/gpu/drm/i915/gvt/sched_policy.c b/drivers/gpu/drm/i915/gvt/sched_policy.c
index 9f97f743aa71..6c2d68e88266 100644
--- a/drivers/gpu/drm/i915/gvt/sched_policy.c
+++ b/drivers/gpu/drm/i915/gvt/sched_policy.c
@@ -447,6 +447,7 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu)
 	struct drm_i915_private *dev_priv = vgpu->gvt->gt->i915;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
+	intel_wakeref_t wakeref;
 
 	if (!vgpu_data->active)
 		return;
@@ -465,7 +466,7 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu)
 		scheduler->current_vgpu = NULL;
 	}
 
-	intel_runtime_pm_get(&dev_priv->runtime_pm);
+	wakeref = intel_runtime_pm_get(&dev_priv->runtime_pm);
 	spin_lock_bh(&scheduler->mmio_context_lock);
 	for_each_engine(engine, vgpu->gvt->gt, id) {
 		if (scheduler->engine_owner[engine->id] == vgpu) {
@@ -474,6 +475,6 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu)
 		}
 	}
 	spin_unlock_bh(&scheduler->mmio_context_lock);
-	intel_runtime_pm_put_unchecked(&dev_priv->runtime_pm);
+	intel_runtime_pm_put(&dev_priv->runtime_pm, wakeref);
 	mutex_unlock(&vgpu->gvt->sched_lock);
 }
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 0d9e263913ff..967c0501e91e 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -66,8 +66,6 @@ static int i915_capabilities(struct seq_file *m, void *data)
 	struct drm_i915_private *i915 = node_to_i915(m->private);
 	struct drm_printer p = drm_seq_file_printer(m);
 
-	seq_printf(m, "pch: %d\n", INTEL_PCH_TYPE(i915));
-
 	intel_device_info_print(INTEL_INFO(i915), RUNTIME_INFO(i915), &p);
 	i915_print_iommu_status(i915, &p);
 	intel_gt_info_print(&to_gt(i915)->info, &p);
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index ce3cc93ea211..273bc43468a0 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -62,7 +62,6 @@
 #include "display/intel_pch_refclk.h"
 #include "display/intel_pps.h"
 #include "display/intel_sprite_uapi.h"
-#include "display/intel_vga.h"
 #include "display/skl_watermark.h"
 
 #include "gem/i915_gem_context.h"
@@ -235,7 +234,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 
 	intel_uncore_mmio_debug_init_early(dev_priv);
 
-	spin_lock_init(&dev_priv->irq_lock);
 	spin_lock_init(&dev_priv->gpu_error.lock);
 
 	intel_sbi_init(dev_priv);
@@ -263,9 +261,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 
 	i915_gem_init_early(dev_priv);
 
-	/* This must be called before any calls to HAS_PCH_* */
-	intel_detect_pch(dev_priv);
-
 	intel_irq_init(dev_priv);
 	intel_display_driver_early_probe(display);
 	intel_clock_gating_hooks_init(dev_priv);
@@ -578,7 +573,7 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv)
 	 */
 	intel_dram_detect(dev_priv);
 
-	intel_bw_init_hw(dev_priv);
+	intel_bw_init_hw(display);
 
 	return 0;
 
@@ -622,11 +617,12 @@ static void i915_driver_hw_remove(struct drm_i915_private *dev_priv)
  * Perform any steps necessary to make the driver available via kernel
  * internal or userspace interfaces.
  */
-static void i915_driver_register(struct drm_i915_private *dev_priv)
+static int i915_driver_register(struct drm_i915_private *dev_priv)
 {
 	struct intel_display *display = &dev_priv->display;
 	struct intel_gt *gt;
 	unsigned int i;
+	int ret;
 
 	i915_gem_driver_register(dev_priv);
 	i915_pmu_register(dev_priv);
@@ -634,10 +630,14 @@ static void i915_driver_register(struct drm_i915_private *dev_priv)
 	intel_vgpu_register(dev_priv);
 
 	/* Reveal our presence to userspace */
-	if (drm_dev_register(&dev_priv->drm, 0)) {
-		drm_err(&dev_priv->drm,
-			"Failed to register driver for userspace access!\n");
-		return;
+	ret = drm_dev_register(&dev_priv->drm, 0);
+	if (ret) {
+		i915_probe_error(dev_priv,
+				 "Failed to register driver for userspace access!\n");
+		drm_dev_unregister(&dev_priv->drm);
+		i915_pmu_unregister(dev_priv);
+		i915_gem_driver_unregister(dev_priv);
+		return ret;
 	}
 
 	i915_debugfs_register(dev_priv);
@@ -660,6 +660,8 @@ static void i915_driver_register(struct drm_i915_private *dev_priv)
 
 	if (i915_switcheroo_register(dev_priv))
 		drm_err(&dev_priv->drm, "Failed to register vga switcheroo!\n");
+
+	return 0;
 }
 
 /**
@@ -834,7 +836,9 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto out_cleanup_gem;
 
-	i915_driver_register(i915);
+	ret = i915_driver_register(i915);
+	if (ret)
+		goto out_cleanup_gem;
 
 	enable_rpm_wakeref_asserts(&i915->runtime_pm);
 
@@ -845,6 +849,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	return 0;
 
 out_cleanup_gem:
+	intel_pxp_fini(i915);
 	i915_gem_suspend(i915);
 	i915_gem_driver_remove(i915);
 	i915_gem_driver_release(i915);
@@ -981,7 +986,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915)
 	intel_dp_mst_suspend(display);
 
 	intel_irq_suspend(i915);
-	intel_hpd_cancel_work(i915);
+	intel_hpd_cancel_work(display);
 
 	if (HAS_DISPLAY(i915))
 		intel_display_driver_suspend_access(display);
@@ -1064,7 +1069,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 	intel_display_driver_suspend(display);
 
 	intel_irq_suspend(dev_priv);
-	intel_hpd_cancel_work(dev_priv);
+	intel_hpd_cancel_work(display);
 
 	if (HAS_DISPLAY(dev_priv))
 		intel_display_driver_suspend_access(display);
@@ -1195,13 +1200,11 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	i9xx_display_sr_restore(display);
 
-	intel_vga_redisable(display);
-
 	intel_gmbus_reset(display);
 
 	intel_pps_unlock_regs_wa(display);
 
-	intel_init_pch_refclk(dev_priv);
+	intel_init_pch_refclk(display);
 
 	/*
 	 * Interrupts have to be enabled before any batches are run. If not the
@@ -1227,7 +1230,7 @@ static int i915_drm_resume(struct drm_device *dev)
 	if (HAS_DISPLAY(dev_priv))
 		intel_display_driver_resume_access(display);
 
-	intel_hpd_init(dev_priv);
+	intel_hpd_init(display);
 
 	intel_display_driver_resume(display);
 
@@ -1235,7 +1238,7 @@ static int i915_drm_resume(struct drm_device *dev)
 		intel_display_driver_enable_user_access(display);
 		drm_kms_helper_poll_enable(dev);
 	}
-	intel_hpd_poll_disable(dev_priv);
+	intel_hpd_poll_disable(display);
 
 	intel_opregion_resume(display);
 
@@ -1575,7 +1578,7 @@ static int intel_runtime_suspend(struct device *kdev)
 	assert_forcewakes_inactive(&dev_priv->uncore);
 
 	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv))
-		intel_hpd_poll_enable(dev_priv);
+		intel_hpd_poll_enable(display);
 
 	drm_dbg(&dev_priv->drm, "Device suspended\n");
 	return 0;
@@ -1633,11 +1636,11 @@ static int intel_runtime_resume(struct device *kdev)
 	 * everyone else do it here.
 	 */
 	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) {
-		intel_hpd_init(dev_priv);
-		intel_hpd_poll_disable(dev_priv);
+		intel_hpd_init(display);
+		intel_hpd_poll_disable(display);
 	}
 
-	skl_watermark_ipc_update(dev_priv);
+	skl_watermark_ipc_update(display);
 
 	enable_rpm_wakeref_asserts(rpm);
 
diff --git a/drivers/gpu/drm/i915/i915_driver.h b/drivers/gpu/drm/i915/i915_driver.h
index 4b67ad9a61cd..1e95ecb2a163 100644
--- a/drivers/gpu/drm/i915/i915_driver.h
+++ b/drivers/gpu/drm/i915/i915_driver.h
@@ -15,7 +15,6 @@ struct drm_printer;
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_TIMESTAMP	1695980603
 
 extern const struct dev_pm_ops i915_pm_ops;
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ffc346379cc2..d0e1980dcba2 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -49,8 +49,6 @@
 #include "gt/intel_workarounds.h"
 #include "gt/uc/intel_uc.h"
 
-#include "soc/intel_pch.h"
-
 #include "i915_drm_client.h"
 #include "i915_gem.h"
 #include "i915_gpu_error.h"
@@ -224,8 +222,6 @@ struct drm_i915_private {
 	};
 	unsigned int engine_uabi_class_count[I915_LAST_UABI_ENGINE_CLASS + 1];
 
-	/* protects the irq masks */
-	spinlock_t irq_lock;
 	bool irqs_enabled;
 
 	/* LPT/WPT IOSF sideband protection */
@@ -272,9 +268,6 @@ struct drm_i915_private {
 	/* pm private clock gating functions */
 	const struct drm_i915_clock_gating_funcs *clock_gating_funcs;
 
-	/* PCH chipset type */
-	enum intel_pch pch_type;
-
 	unsigned long gem_quirks;
 
 	struct i915_gem_mm mm;
@@ -305,6 +298,8 @@ struct drm_i915_private {
 			INTEL_DRAM_DDR5,
 			INTEL_DRAM_LPDDR5,
 			INTEL_DRAM_GDDR,
+			INTEL_DRAM_GDDR_ECC,
+			__INTEL_DRAM_TYPE_MAX,
 		} type;
 		u8 num_qgv_points;
 		u8 num_psf_gv_points;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index a6613eed3398..4f785cdbd155 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -726,13 +726,6 @@ static void err_print_gt_info(struct drm_i915_error_state_buf *m,
 	intel_sseu_print_topology(gt->_gt->i915, &gt->info.sseu, &p);
 }
 
-static void err_print_gt_display(struct drm_i915_error_state_buf *m,
-				 struct intel_gt_coredump *gt)
-{
-	err_printf(m, "IER: 0x%08x\n", gt->ier);
-	err_printf(m, "DERRMR: 0x%08x\n", gt->derrmr);
-}
-
 static void err_print_gt_global_nonguc(struct drm_i915_error_state_buf *m,
 				       struct intel_gt_coredump *gt)
 {
@@ -878,7 +871,6 @@ static void __err_print_to_sgl(struct drm_i915_error_state_buf *m,
 		if (error->gt->uc && error->gt->uc->guc.is_guc_capture)
 			print_guc_capture = true;
 
-		err_print_gt_display(m, error->gt);
 		err_print_gt_global_nonguc(m, error->gt);
 		err_print_gt_fences(m, error->gt);
 
@@ -1767,27 +1759,6 @@ gt_record_uc(struct intel_gt_coredump *gt,
 	return error_uc;
 }
 
-/* Capture display registers. */
-static void gt_record_display_regs(struct intel_gt_coredump *gt)
-{
-	struct intel_uncore *uncore = gt->_gt->uncore;
-	struct drm_i915_private *i915 = uncore->i915;
-
-	if (DISPLAY_VER(i915) >= 6 && DISPLAY_VER(i915) < 20)
-		gt->derrmr = intel_uncore_read(uncore, DERRMR);
-
-	if (GRAPHICS_VER(i915) >= 8)
-		gt->ier = intel_uncore_read(uncore, GEN8_DE_MISC_IER);
-	else if (IS_VALLEYVIEW(i915))
-		gt->ier = intel_uncore_read(uncore, VLV_IER);
-	else if (HAS_PCH_SPLIT(i915))
-		gt->ier = intel_uncore_read(uncore, DEIER);
-	else if (GRAPHICS_VER(i915) == 2)
-		gt->ier = intel_uncore_read16(uncore, GEN2_IER);
-	else
-		gt->ier = intel_uncore_read(uncore, GEN2_IER);
-}
-
 /* Capture all other registers that GuC doesn't capture. */
 static void gt_record_global_nonguc_regs(struct intel_gt_coredump *gt)
 {
@@ -1821,9 +1792,12 @@ static void gt_record_global_nonguc_regs(struct intel_gt_coredump *gt)
 			gt->gtier[i] =
 				intel_uncore_read(uncore, GEN8_GT_IER(i));
 		gt->ngtier = 4;
-	} else if (HAS_PCH_SPLIT(i915)) {
+	} else if (GRAPHICS_VER(i915) >= 5) {
 		gt->gtier[0] = intel_uncore_read(uncore, GTIER);
 		gt->ngtier = 1;
+	} else {
+		gt->gtier[0] = intel_uncore_read(uncore, GEN2_IER);
+		gt->ngtier = 1;
 	}
 
 	gt->eir = intel_uncore_read(uncore, EIR);
@@ -2043,7 +2017,6 @@ intel_gt_coredump_alloc(struct intel_gt *gt, gfp_t gfp, u32 dump_flags)
 	gc->_gt = gt;
 	gc->awake = intel_gt_pm_is_awake(gt);
 
-	gt_record_display_regs(gc);
 	gt_record_global_nonguc_regs(gc);
 
 	/*
@@ -2160,7 +2133,6 @@ i915_gpu_coredump(struct intel_gt *gt, intel_engine_mask_t engine_mask, u32 dump
 void i915_error_state_store(struct i915_gpu_coredump *error)
 {
 	struct drm_i915_private *i915;
-	static bool warned;
 
 	if (IS_ERR_OR_NULL(error))
 		return;
@@ -2174,16 +2146,8 @@ void i915_error_state_store(struct i915_gpu_coredump *error)
 
 	i915_gpu_coredump_get(error);
 
-	if (!xchg(&warned, true) &&
-	    ktime_get_real_seconds() - DRIVER_TIMESTAMP < DAY_AS_SECONDS(180)) {
-		pr_info("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
-		pr_info("Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/intel/issues/new.\n");
-		pr_info("Please see https://drm.pages.freedesktop.org/intel-docs/how-to-file-i915-bugs.html for details.\n");
-		pr_info("drm/i915 developers can then reassign to the right component if it's not a kernel issue.\n");
-		pr_info("The GPU crash dump is required to analyze GPU hangs, so please always attach it.\n");
-		pr_info("GPU crash dump saved to /sys/class/drm/card%d/error\n",
-			i915->drm.primary->index);
-	}
+	drm_info(&i915->drm, "GPU error state saved to /sys/class/drm/card%d/error\n",
+		 i915->drm.primary->index);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h
index 749e1c55613e..182324979278 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.h
+++ b/drivers/gpu/drm/i915/i915_gpu_error.h
@@ -146,7 +146,6 @@ struct intel_gt_coredump {
 	/* Generic register state */
 	u32 eir;
 	u32 pgtbl_er;
-	u32 ier;
 	u32 gtier[6], ngtier;
 	u32 forcewake;
 	u32 error; /* gen6+ */
@@ -164,8 +163,6 @@ struct intel_gt_coredump {
 	u32 clock_frequency;
 	u32 clock_period_ns;
 
-	/* Display related */
-	u32 derrmr;
 	u32 sfc_done[I915_MAX_SFC]; /* gen12 */
 
 	u32 nfence;
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 37ca4a35daf2..95042879bec4 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -277,14 +277,14 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 			intel_uncore_write(&dev_priv->uncore, GEN6_PMIIR, pm_iir);
 
 		if (iir & I915_DISPLAY_PORT_INTERRUPT)
-			hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+			hotplug_status = i9xx_hpd_irq_ack(display);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			vlv_display_error_irq_ack(display, &eir, &dpinvgtt);
 
 		/* Call regardless, as some status bits might not be
 		 * signalled in IIR */
-		i9xx_pipestat_irq_ack(dev_priv, iir, pipe_stats);
+		i9xx_pipestat_irq_ack(display, iir, pipe_stats);
 
 		if (iir & (I915_LPE_PIPE_A_INTERRUPT |
 			   I915_LPE_PIPE_B_INTERRUPT))
@@ -306,12 +306,12 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 			gen6_rps_irq_handler(&to_gt(dev_priv)->rps, pm_iir);
 
 		if (hotplug_status)
-			i9xx_hpd_irq_handler(dev_priv, hotplug_status);
+			i9xx_hpd_irq_handler(display, hotplug_status);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			vlv_display_error_irq_handler(display, eir, dpinvgtt);
 
-		valleyview_pipestat_irq_handler(dev_priv, pipe_stats);
+		valleyview_pipestat_irq_handler(display, pipe_stats);
 	} while (0);
 
 	pmu_irq_stats(dev_priv, ret);
@@ -367,14 +367,14 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 		gen8_gt_irq_handler(to_gt(dev_priv), master_ctl);
 
 		if (iir & I915_DISPLAY_PORT_INTERRUPT)
-			hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+			hotplug_status = i9xx_hpd_irq_ack(display);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			vlv_display_error_irq_ack(display, &eir, &dpinvgtt);
 
 		/* Call regardless, as some status bits might not be
 		 * signalled in IIR */
-		i9xx_pipestat_irq_ack(dev_priv, iir, pipe_stats);
+		i9xx_pipestat_irq_ack(display, iir, pipe_stats);
 
 		if (iir & (I915_LPE_PIPE_A_INTERRUPT |
 			   I915_LPE_PIPE_B_INTERRUPT |
@@ -392,12 +392,12 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 		intel_uncore_write(&dev_priv->uncore, GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
 
 		if (hotplug_status)
-			i9xx_hpd_irq_handler(dev_priv, hotplug_status);
+			i9xx_hpd_irq_handler(display, hotplug_status);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			vlv_display_error_irq_handler(display, eir, dpinvgtt);
 
-		valleyview_pipestat_irq_handler(dev_priv, pipe_stats);
+		valleyview_pipestat_irq_handler(display, pipe_stats);
 	} while (0);
 
 	pmu_irq_stats(dev_priv, ret);
@@ -418,6 +418,7 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 static irqreturn_t ilk_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *i915 = arg;
+	struct intel_display *display = &i915->display;
 	void __iomem * const regs = intel_uncore_regs(&i915->uncore);
 	u32 de_iir, gt_iir, de_ier, sde_ier = 0;
 	irqreturn_t ret = IRQ_NONE;
@@ -458,9 +459,9 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg)
 	if (de_iir) {
 		raw_reg_write(regs, DEIIR, de_iir);
 		if (DISPLAY_VER(i915) >= 7)
-			ivb_display_irq_handler(i915, de_iir);
+			ivb_display_irq_handler(display, de_iir);
 		else
-			ilk_display_irq_handler(i915, de_iir);
+			ilk_display_irq_handler(display, de_iir);
 		ret = IRQ_HANDLED;
 	}
 
@@ -506,6 +507,7 @@ static inline void gen8_master_intr_enable(void __iomem * const regs)
 static irqreturn_t gen8_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = arg;
+	struct intel_display *display = &dev_priv->display;
 	void __iomem * const regs = intel_uncore_regs(&dev_priv->uncore);
 	u32 master_ctl;
 
@@ -524,7 +526,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 	/* IRQs are synced during runtime_suspend, we don't require a wakeref */
 	if (master_ctl & ~GEN8_GT_IRQS) {
 		disable_rpm_wakeref_asserts(&dev_priv->runtime_pm);
-		gen8_de_irq_handler(dev_priv, master_ctl);
+		gen8_de_irq_handler(display, master_ctl);
 		enable_rpm_wakeref_asserts(&dev_priv->runtime_pm);
 	}
 
@@ -556,6 +558,7 @@ static inline void gen11_master_intr_enable(void __iomem * const regs)
 static irqreturn_t gen11_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *i915 = arg;
+	struct intel_display *display = &i915->display;
 	void __iomem * const regs = intel_uncore_regs(&i915->uncore);
 	struct intel_gt *gt = to_gt(i915);
 	u32 master_ctl;
@@ -575,13 +578,13 @@ static irqreturn_t gen11_irq_handler(int irq, void *arg)
 
 	/* IRQs are synced during runtime_suspend, we don't require a wakeref */
 	if (master_ctl & GEN11_DISPLAY_IRQ)
-		gen11_display_irq_handler(i915);
+		gen11_display_irq_handler(display);
 
-	gu_misc_iir = gen11_gu_misc_irq_ack(i915, master_ctl);
+	gu_misc_iir = gen11_gu_misc_irq_ack(display, master_ctl);
 
 	gen11_master_intr_enable(regs);
 
-	gen11_gu_misc_irq_handler(i915, gu_misc_iir);
+	gen11_gu_misc_irq_handler(display, gu_misc_iir);
 
 	pmu_irq_stats(i915, IRQ_HANDLED);
 
@@ -613,6 +616,7 @@ static inline void dg1_master_intr_enable(void __iomem * const regs)
 static irqreturn_t dg1_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private * const i915 = arg;
+	struct intel_display *display = &i915->display;
 	struct intel_gt *gt = to_gt(i915);
 	void __iomem * const regs = intel_uncore_regs(gt->uncore);
 	u32 master_tile_ctl, master_ctl;
@@ -641,36 +645,22 @@ static irqreturn_t dg1_irq_handler(int irq, void *arg)
 	gen11_gt_irq_handler(gt, master_ctl);
 
 	if (master_ctl & GEN11_DISPLAY_IRQ)
-		gen11_display_irq_handler(i915);
+		gen11_display_irq_handler(display);
 
-	gu_misc_iir = gen11_gu_misc_irq_ack(i915, master_ctl);
+	gu_misc_iir = gen11_gu_misc_irq_ack(display, master_ctl);
 
 	dg1_master_intr_enable(regs);
 
-	gen11_gu_misc_irq_handler(i915, gu_misc_iir);
+	gen11_gu_misc_irq_handler(display, gu_misc_iir);
 
 	pmu_irq_stats(i915, IRQ_HANDLED);
 
 	return IRQ_HANDLED;
 }
 
-static void ibx_irq_reset(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-
-	if (HAS_PCH_NOP(dev_priv))
-		return;
-
-	gen2_irq_reset(uncore, SDE_IRQ_REGS);
-
-	if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
-		intel_uncore_write(&dev_priv->uncore, SERR_INT, 0xffffffff);
-}
-
-/* drm_dma.h hooks
-*/
 static void ilk_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	gen2_irq_reset(uncore, DE_IRQ_REGS);
@@ -686,45 +676,43 @@ static void ilk_irq_reset(struct drm_i915_private *dev_priv)
 
 	gen5_gt_irq_reset(to_gt(dev_priv));
 
-	ibx_irq_reset(dev_priv);
+	ibx_display_irq_reset(display);
 }
 
 static void valleyview_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
+
 	intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, 0);
 	intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER);
 
 	gen5_gt_irq_reset(to_gt(dev_priv));
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	vlv_display_irq_reset(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	vlv_display_irq_reset(display);
 }
 
 static void gen8_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	gen8_master_intr_disable(intel_uncore_regs(uncore));
 
 	gen8_gt_irq_reset(to_gt(dev_priv));
-	gen8_display_irq_reset(dev_priv);
+	gen8_display_irq_reset(display);
 	gen2_irq_reset(uncore, GEN8_PCU_IRQ_REGS);
-
-	if (HAS_PCH_SPLIT(dev_priv))
-		ibx_irq_reset(dev_priv);
-
 }
 
 static void gen11_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_gt *gt = to_gt(dev_priv);
 	struct intel_uncore *uncore = gt->uncore;
 
 	gen11_master_intr_disable(intel_uncore_regs(&dev_priv->uncore));
 
 	gen11_gt_irq_reset(gt);
-	gen11_display_irq_reset(dev_priv);
+	gen11_display_irq_reset(display);
 
 	gen2_irq_reset(uncore, GEN11_GU_MISC_IRQ_REGS);
 	gen2_irq_reset(uncore, GEN8_PCU_IRQ_REGS);
@@ -732,6 +720,7 @@ static void gen11_irq_reset(struct drm_i915_private *dev_priv)
 
 static void dg1_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 	struct intel_gt *gt;
 	unsigned int i;
@@ -741,7 +730,7 @@ static void dg1_irq_reset(struct drm_i915_private *dev_priv)
 	for_each_gt(gt, dev_priv, i)
 		gen11_gt_irq_reset(gt);
 
-	gen11_display_irq_reset(dev_priv);
+	gen11_display_irq_reset(display);
 
 	gen2_irq_reset(uncore, GEN11_GU_MISC_IRQ_REGS);
 	gen2_irq_reset(uncore, GEN8_PCU_IRQ_REGS);
@@ -751,6 +740,7 @@ static void dg1_irq_reset(struct drm_i915_private *dev_priv)
 
 static void cherryview_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 
 	intel_uncore_write(uncore, GEN8_MASTER_IRQ, 0);
@@ -760,25 +750,25 @@ static void cherryview_irq_reset(struct drm_i915_private *dev_priv)
 
 	gen2_irq_reset(uncore, GEN8_PCU_IRQ_REGS);
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	vlv_display_irq_reset(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	vlv_display_irq_reset(display);
 }
 
 static void ilk_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
+
 	gen5_gt_irq_postinstall(to_gt(dev_priv));
 
-	ilk_de_irq_postinstall(dev_priv);
+	ilk_de_irq_postinstall(display);
 }
 
 static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
+
 	gen5_gt_irq_postinstall(to_gt(dev_priv));
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	vlv_display_irq_postinstall(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	vlv_display_irq_postinstall(display);
 
 	intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE);
 	intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER);
@@ -786,20 +776,23 @@ static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv)
 
 static void gen8_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
+
 	gen8_gt_irq_postinstall(to_gt(dev_priv));
-	gen8_de_irq_postinstall(dev_priv);
+	gen8_de_irq_postinstall(display);
 
 	gen8_master_intr_enable(intel_uncore_regs(&dev_priv->uncore));
 }
 
 static void gen11_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_gt *gt = to_gt(dev_priv);
 	struct intel_uncore *uncore = gt->uncore;
 	u32 gu_misc_masked = GEN11_GU_MISC_GSE;
 
 	gen11_gt_irq_postinstall(gt);
-	gen11_de_irq_postinstall(dev_priv);
+	gen11_de_irq_postinstall(display);
 
 	gen2_irq_init(uncore, GEN11_GU_MISC_IRQ_REGS, ~gu_misc_masked, gu_misc_masked);
 
@@ -809,6 +802,7 @@ static void gen11_irq_postinstall(struct drm_i915_private *dev_priv)
 
 static void dg1_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 gu_misc_masked = GEN11_GU_MISC_GSE;
 	struct intel_gt *gt;
@@ -819,7 +813,7 @@ static void dg1_irq_postinstall(struct drm_i915_private *dev_priv)
 
 	gen2_irq_init(uncore, GEN11_GU_MISC_IRQ_REGS, ~gu_misc_masked, gu_misc_masked);
 
-	dg1_de_irq_postinstall(dev_priv);
+	dg1_de_irq_postinstall(display);
 
 	dg1_master_intr_enable(intel_uncore_regs(uncore));
 	intel_uncore_posting_read(uncore, DG1_MSTR_TILE_INTR);
@@ -827,11 +821,11 @@ static void dg1_irq_postinstall(struct drm_i915_private *dev_priv)
 
 static void cherryview_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
+
 	gen8_gt_irq_postinstall(to_gt(dev_priv));
 
-	spin_lock_irq(&dev_priv->irq_lock);
-	vlv_display_irq_postinstall(dev_priv);
-	spin_unlock_irq(&dev_priv->irq_lock);
+	vlv_display_irq_postinstall(display);
 
 	intel_uncore_write(&dev_priv->uncore, GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL);
 	intel_uncore_posting_read(&dev_priv->uncore, GEN8_MASTER_IRQ);
@@ -900,9 +894,10 @@ static void i9xx_error_irq_handler(struct drm_i915_private *dev_priv,
 
 static void i915_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 
-	i9xx_display_irq_reset(dev_priv);
+	i9xx_display_irq_reset(display);
 
 	gen2_error_reset(uncore, GEN2_ERROR_REGS);
 	gen2_irq_reset(uncore, GEN2_IRQ_REGS);
@@ -911,6 +906,7 @@ static void i915_irq_reset(struct drm_i915_private *dev_priv)
 
 static void i915_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 enable_mask;
 
@@ -932,26 +928,20 @@ static void i915_irq_postinstall(struct drm_i915_private *dev_priv)
 		enable_mask |= I915_ASLE_INTERRUPT;
 	}
 
-	if (I915_HAS_HOTPLUG(dev_priv)) {
+	if (HAS_HOTPLUG(dev_priv)) {
 		dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT;
 		enable_mask |= I915_DISPLAY_PORT_INTERRUPT;
 	}
 
 	gen2_irq_init(uncore, GEN2_IRQ_REGS, dev_priv->irq_mask, enable_mask);
 
-	/* Interrupt setup is already guaranteed to be single-threaded, this is
-	 * just to make the assert_spin_locked check happy. */
-	spin_lock_irq(&dev_priv->irq_lock);
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_CRC_DONE_INTERRUPT_STATUS);
-	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_CRC_DONE_INTERRUPT_STATUS);
-	spin_unlock_irq(&dev_priv->irq_lock);
-
-	i915_enable_asle_pipestat(dev_priv);
+	i915_display_irq_postinstall(display);
 }
 
 static irqreturn_t i915_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = arg;
+	struct intel_display *display = &dev_priv->display;
 	irqreturn_t ret = IRQ_NONE;
 
 	if (!intel_irqs_enabled(dev_priv))
@@ -972,13 +962,13 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 
 		ret = IRQ_HANDLED;
 
-		if (I915_HAS_HOTPLUG(dev_priv) &&
+		if (HAS_HOTPLUG(dev_priv) &&
 		    iir & I915_DISPLAY_PORT_INTERRUPT)
-			hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+			hotplug_status = i9xx_hpd_irq_ack(display);
 
 		/* Call regardless, as some status bits might not be
 		 * signalled in IIR */
-		i9xx_pipestat_irq_ack(dev_priv, iir, pipe_stats);
+		i9xx_pipestat_irq_ack(display, iir, pipe_stats);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i9xx_error_irq_ack(dev_priv, &eir, &eir_stuck);
@@ -992,9 +982,9 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 			i9xx_error_irq_handler(dev_priv, eir, eir_stuck);
 
 		if (hotplug_status)
-			i9xx_hpd_irq_handler(dev_priv, hotplug_status);
+			i9xx_hpd_irq_handler(display, hotplug_status);
 
-		i915_pipestat_irq_handler(dev_priv, iir, pipe_stats);
+		i915_pipestat_irq_handler(display, iir, pipe_stats);
 	} while (0);
 
 	pmu_irq_stats(dev_priv, ret);
@@ -1006,9 +996,10 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 
 static void i965_irq_reset(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 
-	i9xx_display_irq_reset(dev_priv);
+	i9xx_display_irq_reset(display);
 
 	gen2_error_reset(uncore, GEN2_ERROR_REGS);
 	gen2_irq_reset(uncore, GEN2_IRQ_REGS);
@@ -1036,6 +1027,7 @@ static u32 i965_error_mask(struct drm_i915_private *i915)
 
 static void i965_irq_postinstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 enable_mask;
 
@@ -1061,20 +1053,13 @@ static void i965_irq_postinstall(struct drm_i915_private *dev_priv)
 
 	gen2_irq_init(uncore, GEN2_IRQ_REGS, dev_priv->irq_mask, enable_mask);
 
-	/* Interrupt setup is already guaranteed to be single-threaded, this is
-	 * just to make the assert_spin_locked check happy. */
-	spin_lock_irq(&dev_priv->irq_lock);
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_CRC_DONE_INTERRUPT_STATUS);
-	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_CRC_DONE_INTERRUPT_STATUS);
-	spin_unlock_irq(&dev_priv->irq_lock);
-
-	i915_enable_asle_pipestat(dev_priv);
+	i965_display_irq_postinstall(display);
 }
 
 static irqreturn_t i965_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = arg;
+	struct intel_display *display = &dev_priv->display;
 	irqreturn_t ret = IRQ_NONE;
 
 	if (!intel_irqs_enabled(dev_priv))
@@ -1096,11 +1081,11 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		ret = IRQ_HANDLED;
 
 		if (iir & I915_DISPLAY_PORT_INTERRUPT)
-			hotplug_status = i9xx_hpd_irq_ack(dev_priv);
+			hotplug_status = i9xx_hpd_irq_ack(display);
 
 		/* Call regardless, as some status bits might not be
 		 * signalled in IIR */
-		i9xx_pipestat_irq_ack(dev_priv, iir, pipe_stats);
+		i9xx_pipestat_irq_ack(display, iir, pipe_stats);
 
 		if (iir & I915_MASTER_ERROR_INTERRUPT)
 			i9xx_error_irq_ack(dev_priv, &eir, &eir_stuck);
@@ -1119,9 +1104,9 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 			i9xx_error_irq_handler(dev_priv, eir, eir_stuck);
 
 		if (hotplug_status)
-			i9xx_hpd_irq_handler(dev_priv, hotplug_status);
+			i9xx_hpd_irq_handler(display, hotplug_status);
 
-		i965_pipestat_irq_handler(dev_priv, iir, pipe_stats);
+		i965_pipestat_irq_handler(display, iir, pipe_stats);
 	} while (0);
 
 	pmu_irq_stats(dev_priv, IRQ_HANDLED);
@@ -1280,6 +1265,7 @@ int intel_irq_install(struct drm_i915_private *dev_priv)
  */
 void intel_irq_uninstall(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = &dev_priv->display;
 	int irq = to_pci_dev(dev_priv->drm.dev)->irq;
 
 	if (drm_WARN_ON(&dev_priv->drm, !dev_priv->irqs_enabled))
@@ -1289,7 +1275,7 @@ void intel_irq_uninstall(struct drm_i915_private *dev_priv)
 
 	free_irq(irq, dev_priv);
 
-	intel_hpd_cancel_work(dev_priv);
+	intel_hpd_cancel_work(display);
 	dev_priv->irqs_enabled = false;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c5064eebe063..2e4190da3e0d 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -866,6 +866,7 @@
 #define   FP_M2_DIV_MASK	0x0000003f
 #define   FP_M2_PINEVIEW_DIV_MASK	0x000000ff
 #define   FP_M2_DIV_SHIFT		 0
+
 #define DPLL_TEST	_MMIO(0x606c)
 #define   DPLLB_TEST_SDVO_DIV_1		(0 << 22)
 #define   DPLLB_TEST_SDVO_DIV_2		(1 << 22)
@@ -877,11 +878,13 @@
 #define   DPLLA_TEST_N_BYPASS		(1 << 3)
 #define   DPLLA_TEST_M_BYPASS		(1 << 2)
 #define   DPLLA_INPUT_BUFFER_ENABLE	(1 << 0)
+
 #define D_STATE		_MMIO(0x6104)
 #define  DSTATE_GFX_RESET_I830			(1 << 6)
 #define  DSTATE_PLL_D3_OFF			(1 << 3)
 #define  DSTATE_GFX_CLOCK_GATING		(1 << 1)
 #define  DSTATE_DOT_CLOCK_GATING		(1 << 0)
+
 #define DSPCLK_GATE_D(__i915)		_MMIO(DISPLAY_MMIO_BASE(__i915) + 0x6200)
 # define DPUNIT_B_CLOCK_GATE_DISABLE		(1 << 30) /* 965 */
 # define VSUNIT_CLOCK_GATE_DISABLE		(1 << 29) /* 965 */
@@ -1050,7 +1053,6 @@
 /*
  * Overlay regs
  */
-
 #define OVADD			_MMIO(0x30000)
 #define DOVSTA			_MMIO(0x30008)
 #define OC_BUF			(0x3 << 20)
@@ -1077,6 +1079,7 @@
 
 #define GEN9_CLKGATE_DIS_4		_MMIO(0x4653C)
 #define   BXT_GMBUS_GATING_DIS		(1 << 14)
+#define   DG2_DPFC_GATING_DIS		REG_BIT(31)
 
 #define GEN9_CLKGATE_DIS_5		_MMIO(0x46540)
 #define   DPCE_GATING_DIS		REG_BIT(17)
@@ -1105,7 +1108,6 @@
 /*
  * Display engine regs
  */
-
 /* Pipe/transcoder A timing regs */
 #define _TRANS_HTOTAL_A		0x60000
 #define _TRANS_HTOTAL_B		0x61000
@@ -1396,88 +1398,50 @@
 #define VLV_DP_B		_MMIO(VLV_DISPLAY_BASE + 0x64100)
 #define VLV_DP_C		_MMIO(VLV_DISPLAY_BASE + 0x64200)
 #define CHV_DP_D		_MMIO(VLV_DISPLAY_BASE + 0x64300)
-#define   DP_PORT_EN			(1 << 31)
-#define   DP_PIPE_SEL_SHIFT		30
-#define   DP_PIPE_SEL_MASK		(1 << 30)
-#define   DP_PIPE_SEL(pipe)		((pipe) << 30)
-#define   DP_PIPE_SEL_SHIFT_IVB		29
-#define   DP_PIPE_SEL_MASK_IVB		(3 << 29)
-#define   DP_PIPE_SEL_IVB(pipe)		((pipe) << 29)
+#define   DP_PORT_EN			REG_BIT(31)
+#define   DP_PIPE_SEL_MASK		REG_GENMASK(30, 30)
+#define   DP_PIPE_SEL(pipe)		REG_FIELD_PREP(DP_PIPE_SEL_MASK, (pipe))
+#define   DP_PIPE_SEL_MASK_IVB		REG_GENMASK(30, 29)
+#define   DP_PIPE_SEL_IVB(pipe)		REG_FIELD_PREP(DP_PIPE_SEL_MASK_IVB, (pipe))
 #define   DP_PIPE_SEL_SHIFT_CHV		16
-#define   DP_PIPE_SEL_MASK_CHV		(3 << 16)
-#define   DP_PIPE_SEL_CHV(pipe)		((pipe) << 16)
-
-/* Link training mode - select a suitable mode for each stage */
-#define   DP_LINK_TRAIN_PAT_1		(0 << 28)
-#define   DP_LINK_TRAIN_PAT_2		(1 << 28)
-#define   DP_LINK_TRAIN_PAT_IDLE	(2 << 28)
-#define   DP_LINK_TRAIN_OFF		(3 << 28)
-#define   DP_LINK_TRAIN_MASK		(3 << 28)
-#define   DP_LINK_TRAIN_SHIFT		28
-
-/* CPT Link training mode */
-#define   DP_LINK_TRAIN_PAT_1_CPT	(0 << 8)
-#define   DP_LINK_TRAIN_PAT_2_CPT	(1 << 8)
-#define   DP_LINK_TRAIN_PAT_IDLE_CPT	(2 << 8)
-#define   DP_LINK_TRAIN_OFF_CPT		(3 << 8)
-#define   DP_LINK_TRAIN_MASK_CPT	(7 << 8)
-#define   DP_LINK_TRAIN_SHIFT_CPT	8
-
-/* Signal voltages. These are mostly controlled by the other end */
-#define   DP_VOLTAGE_0_4		(0 << 25)
-#define   DP_VOLTAGE_0_6		(1 << 25)
-#define   DP_VOLTAGE_0_8		(2 << 25)
-#define   DP_VOLTAGE_1_2		(3 << 25)
-#define   DP_VOLTAGE_MASK		(7 << 25)
-#define   DP_VOLTAGE_SHIFT		25
-
-/* Signal pre-emphasis levels, like voltages, the other end tells us what
- * they want
- */
-#define   DP_PRE_EMPHASIS_0		(0 << 22)
-#define   DP_PRE_EMPHASIS_3_5		(1 << 22)
-#define   DP_PRE_EMPHASIS_6		(2 << 22)
-#define   DP_PRE_EMPHASIS_9_5		(3 << 22)
-#define   DP_PRE_EMPHASIS_MASK		(7 << 22)
-#define   DP_PRE_EMPHASIS_SHIFT		22
-
-/* How many wires to use. I guess 3 was too hard */
-#define   DP_PORT_WIDTH(width)		(((width) - 1) << 19)
-#define   DP_PORT_WIDTH_MASK		(7 << 19)
-#define   DP_PORT_WIDTH_SHIFT		19
-
-/* Mystic DPCD version 1.1 special mode */
-#define   DP_ENHANCED_FRAMING		(1 << 18)
-
-/* eDP */
-#define   DP_PLL_FREQ_270MHZ		(0 << 16)
-#define   DP_PLL_FREQ_162MHZ		(1 << 16)
-#define   DP_PLL_FREQ_MASK		(3 << 16)
-
-/* locked once port is enabled */
-#define   DP_PORT_REVERSAL		(1 << 15)
-
-/* eDP */
-#define   DP_PLL_ENABLE			(1 << 14)
-
-/* sends the clock on lane 15 of the PEG for debug */
-#define   DP_CLOCK_OUTPUT_ENABLE	(1 << 13)
-
-#define   DP_SCRAMBLING_DISABLE		(1 << 12)
-#define   DP_SCRAMBLING_DISABLE_IRONLAKE	(1 << 7)
-
-/* limit RGB values to avoid confusing TVs */
-#define   DP_COLOR_RANGE_16_235		(1 << 8)
-
-/* Turn on the audio link */
-#define   DP_AUDIO_OUTPUT_ENABLE	(1 << 6)
-
-/* vs and hs sync polarity */
-#define   DP_SYNC_VS_HIGH		(1 << 4)
-#define   DP_SYNC_HS_HIGH		(1 << 3)
-
-/* A fantasy */
-#define   DP_DETECTED			(1 << 2)
+#define   DP_PIPE_SEL_MASK_CHV		REG_GENMASK(17, 16)
+#define   DP_PIPE_SEL_CHV(pipe)		REG_FIELD_PREP(DP_PIPE_SEL_MASK_CHV, (pipe))
+#define   DP_LINK_TRAIN_MASK		REG_GENMASK(29, 28)
+#define   DP_LINK_TRAIN_PAT_1		REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 0)
+#define   DP_LINK_TRAIN_PAT_2		REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 1)
+#define   DP_LINK_TRAIN_PAT_IDLE	REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 2)
+#define   DP_LINK_TRAIN_OFF		REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 3)
+#define   DP_LINK_TRAIN_MASK_CPT	REG_GENMASK(10, 8)
+#define   DP_LINK_TRAIN_PAT_1_CPT	REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 0)
+#define   DP_LINK_TRAIN_PAT_2_CPT	REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 1)
+#define   DP_LINK_TRAIN_PAT_IDLE_CPT	REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 2)
+#define   DP_LINK_TRAIN_OFF_CPT		REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 3)
+#define   DP_VOLTAGE_MASK		REG_GENMASK(27, 25)
+#define   DP_VOLTAGE_0_4		REG_FIELD_PREP(DP_VOLTAGE_MASK, 0)
+#define   DP_VOLTAGE_0_6		REG_FIELD_PREP(DP_VOLTAGE_MASK, 1)
+#define   DP_VOLTAGE_0_8		REG_FIELD_PREP(DP_VOLTAGE_MASK, 2)
+#define   DP_VOLTAGE_1_2		REG_FIELD_PREP(DP_VOLTAGE_MASK, 3)
+#define   DP_PRE_EMPHASIS_MASK		REG_GENMASK(24, 22)
+#define   DP_PRE_EMPHASIS_0		REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 0)
+#define   DP_PRE_EMPHASIS_3_5		REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 1)
+#define   DP_PRE_EMPHASIS_6		REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 2)
+#define   DP_PRE_EMPHASIS_9_5		REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 3)
+#define   DP_PORT_WIDTH_MASK		REG_GENMASK(21, 19)
+#define   DP_PORT_WIDTH(width)		REG_FIELD_PREP(DP_PORT_WIDTH_MASK, (width) - 1)
+#define   DP_ENHANCED_FRAMING		REG_BIT(18)
+#define   EDP_PLL_FREQ_MASK		REG_GENMASK(17, 16)
+#define   EDP_PLL_FREQ_270MHZ		REG_FIELD_PREP(EDP_PLL_FREQ_MASK, 0)
+#define   EDP_PLL_FREQ_162MHZ		REG_FIELD_PREP(EDP_PLL_FREQ_MASK, 1)
+#define   DP_PORT_REVERSAL		REG_BIT(15)
+#define   EDP_PLL_ENABLE		REG_BIT(14)
+#define   DP_CLOCK_OUTPUT_ENABLE	REG_BIT(13)
+#define   DP_SCRAMBLING_DISABLE		REG_BIT(12)
+#define   DP_SCRAMBLING_DISABLE_ILK	REG_BIT(7)
+#define   DP_COLOR_RANGE_16_235		REG_BIT(8)
+#define   DP_AUDIO_OUTPUT_ENABLE	REG_BIT(6)
+#define   DP_SYNC_VS_HIGH		REG_BIT(4)
+#define   DP_SYNC_HS_HIGH		REG_BIT(3)
+#define   DP_DETECTED			REG_BIT(2)
 
 /*
  * Computing GMCH M and N values for the Display Port link
@@ -1811,18 +1775,6 @@
 #define SWF3(dev_priv, i)	_MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x72414 + (i) * 4)
 #define SWF_ILK(i)	_MMIO(0x4F000 + (i) * 4)
 
-/* VBIOS regs */
-#define VGACNTRL		_MMIO(0x71400)
-# define VGA_DISP_DISABLE			(1 << 31)
-# define VGA_2X_MODE				(1 << 30)
-# define VGA_PIPE_B_SELECT			(1 << 29)
-
-#define VLV_VGACNTRL		_MMIO(VLV_DISPLAY_BASE + 0x71400)
-
-/* Ironlake */
-
-#define CPU_VGACNTRL	_MMIO(0x41000)
-
 #define DIGITAL_PORT_HOTPLUG_CNTRL	_MMIO(0x44030)
 #define  DIGITAL_PORTA_HOTPLUG_ENABLE		(1 << 4)
 #define  DIGITAL_PORTA_PULSE_DURATION_2ms	(0 << 2) /* pre-HSW */
@@ -2783,7 +2735,6 @@
  * functionality covered in PCH_PORT_HOTPLUG is split into
  * SHOTPLUG_CTL_DDI and SHOTPLUG_CTL_TC.
  */
-
 #define SHOTPLUG_CTL_DDI				_MMIO(0xc4030)
 #define   SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin)			(0x8 << (_HPD_PIN_DDI(hpd_pin) * 4))
 #define   SHOTPLUG_CTL_DDI_HPD_OUTPUT_DATA(hpd_pin)		(0x4 << (_HPD_PIN_DDI(hpd_pin) * 4))
@@ -2863,7 +2814,6 @@
 #define  TRANS_DPLL_ENABLE(pipe)	(1 << ((pipe) * 4 + 3))
 
 /* transcoder */
-
 #define _PCH_TRANS_HTOTAL_A		0xe0000
 #define _PCH_TRANS_HTOTAL_B		0xe1000
 #define PCH_TRANS_HTOTAL(pipe)		_MMIO_PIPE(pipe, _PCH_TRANS_HTOTAL_A, _PCH_TRANS_HTOTAL_B)
@@ -3794,7 +3744,6 @@ enum skl_power_gate {
 /*
  * SKL Clocks
  */
-
 /* CDCLK_CTL */
 #define CDCLK_CTL			_MMIO(0x46000)
 #define  CDCLK_FREQ_SEL_MASK		REG_GENMASK(27, 26)
@@ -4242,6 +4191,11 @@ enum skl_power_gate {
 #define MTL_CLKGATE_DIS_TRANS(dev_priv, trans)			_MMIO_TRANS2(dev_priv, trans, _MTL_CLKGATE_DIS_TRANS_A)
 #define  MTL_CLKGATE_DIS_TRANS_DMASC_GATING_DIS		REG_BIT(7)
 
+#define _MTL_PIPE_CLKGATE_DIS2_A		0x60114
+#define _MTL_PIPE_CLKGATE_DIS2_B		0x61114
+#define MTL_PIPE_CLKGATE_DIS2(pipe)		_MMIO_PIPE(pipe, _MTL_PIPE_CLKGATE_DIS2_A, _MTL_PIPE_CLKGATE_DIS2_B)
+#define   MTL_DPFC_GATING_DIS			REG_BIT(6)
+
 #define MTL_MEM_SS_INFO_GLOBAL			_MMIO(0x45700)
 #define   MTL_N_OF_ENABLED_QGV_POINTS_MASK	REG_GENMASK(11, 8)
 #define   MTL_N_OF_POPULATED_CH_MASK		REG_GENMASK(7, 4)
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 609214231ffc..f7fb40cfdb70 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -40,8 +40,6 @@
 struct drm_i915_private;
 struct timer_list;
 
-#define FDO_BUG_URL "https://drm.pages.freedesktop.org/intel-docs/how-to-file-i915-bugs.html"
-
 #define MISSING_CASE(x) WARN(1, "Missing case (%s == %ld)\n", \
 			     __stringify(x), (long)(x))
 
diff --git a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c
index 76d84cbb8361..d581a9d2c063 100644
--- a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c
+++ b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c
@@ -21,6 +21,7 @@
 #include "display/intel_pfit_regs.h"
 #include "display/intel_psr_regs.h"
 #include "display/intel_sprite_regs.h"
+#include "display/intel_vga_regs.h"
 #include "display/skl_universal_plane_regs.h"
 #include "display/skl_watermark_regs.h"
 #include "display/vlv_dsi_pll_regs.h"
diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c
index d40ee1b42110..59bd603e6deb 100644
--- a/drivers/gpu/drm/i915/intel_memory_region.c
+++ b/drivers/gpu/drm/i915/intel_memory_region.c
@@ -171,6 +171,17 @@ intel_memory_region_by_type(struct drm_i915_private *i915,
 	return NULL;
 }
 
+bool intel_memory_type_is_local(enum intel_memory_type mem_type)
+{
+	switch (mem_type) {
+	case INTEL_MEMORY_LOCAL:
+	case INTEL_MEMORY_STOLEN_LOCAL:
+		return true;
+	default:
+		return false;
+	}
+}
+
 /**
  * intel_memory_region_reserve - Reserve a memory range
  * @mem: The region for which we want to reserve a range.
@@ -216,7 +227,7 @@ static int intel_memory_region_memtest(struct intel_memory_region *mem,
 	return err;
 }
 
-static const char *region_type_str(u16 type)
+const char *intel_memory_type_str(enum intel_memory_type type)
 {
 	switch (type) {
 	case INTEL_MEMORY_SYSTEM:
@@ -260,7 +271,7 @@ intel_memory_region_create(struct drm_i915_private *i915,
 	mem->instance = instance;
 
 	snprintf(mem->uabi_name, sizeof(mem->uabi_name), "%s%u",
-		 region_type_str(type), instance);
+		 intel_memory_type_str(type), instance);
 
 	mutex_init(&mem->objects.lock);
 	INIT_LIST_HEAD(&mem->objects.list);
diff --git a/drivers/gpu/drm/i915/intel_memory_region.h b/drivers/gpu/drm/i915/intel_memory_region.h
index 5973b6fe13cf..b3b75be9ced5 100644
--- a/drivers/gpu/drm/i915/intel_memory_region.h
+++ b/drivers/gpu/drm/i915/intel_memory_region.h
@@ -85,6 +85,8 @@ struct intel_memory_region {
 	void *region_private;
 };
 
+bool intel_memory_type_is_local(enum intel_memory_type mem_type);
+
 struct intel_memory_region *
 intel_memory_region_lookup(struct drm_i915_private *i915,
 			   u16 class, u16 instance);
@@ -107,6 +109,7 @@ void intel_memory_regions_driver_release(struct drm_i915_private *i915);
 struct intel_memory_region *
 intel_memory_region_by_type(struct drm_i915_private *i915,
 			    enum intel_memory_type mem_type);
+const char *intel_memory_type_str(enum intel_memory_type type);
 
 __printf(2, 3) void
 intel_memory_region_set_name(struct intel_memory_region *mem,
diff --git a/drivers/gpu/drm/i915/intel_wakeref.h b/drivers/gpu/drm/i915/intel_wakeref.h
index 48836ef52d40..a2894a56e18f 100644
--- a/drivers/gpu/drm/i915/intel_wakeref.h
+++ b/drivers/gpu/drm/i915/intel_wakeref.h
@@ -7,8 +7,6 @@
 #ifndef INTEL_WAKEREF_H
 #define INTEL_WAKEREF_H
 
-#include <drm/drm_print.h>
-
 #include <linux/atomic.h>
 #include <linux/bitfield.h>
 #include <linux/bits.h>
@@ -16,11 +14,13 @@
 #include <linux/mutex.h>
 #include <linux/refcount.h>
 #include <linux/ref_tracker.h>
-#include <linux/slab.h>
-#include <linux/stackdepot.h>
 #include <linux/timer.h>
 #include <linux/workqueue.h>
 
+struct drm_printer;
+struct intel_runtime_pm;
+struct intel_wakeref;
+
 typedef struct ref_tracker *intel_wakeref_t;
 
 #define INTEL_REFTRACK_DEAD_COUNT 16
@@ -32,9 +32,6 @@ typedef struct ref_tracker *intel_wakeref_t;
 #define INTEL_WAKEREF_BUG_ON(expr) BUILD_BUG_ON_INVALID(expr)
 #endif
 
-struct intel_runtime_pm;
-struct intel_wakeref;
-
 struct intel_wakeref_ops {
 	int (*get)(struct intel_wakeref *wf);
 	int (*put)(struct intel_wakeref *wf);
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
index 9aae779c4da3..4969d3de2bac 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
@@ -23,6 +23,7 @@ int intel_pxp_gsccs_init(struct intel_pxp *pxp);
 
 int intel_pxp_gsccs_create_session(struct intel_pxp *pxp, int arb_session_id);
 void intel_pxp_gsccs_end_arb_fw_session(struct intel_pxp *pxp, u32 arb_session_id);
+bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp);
 
 #else
 static inline void intel_pxp_gsccs_fini(struct intel_pxp *pxp)
@@ -34,8 +35,11 @@ static inline int intel_pxp_gsccs_init(struct intel_pxp *pxp)
 	return 0;
 }
 
-#endif
+static inline bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp)
+{
+	return false;
+}
 
-bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp);
+#endif
 
 #endif /*__INTEL_PXP_GSCCS_H__ */
diff --git a/drivers/gpu/drm/i915/selftests/i915_selftest.c b/drivers/gpu/drm/i915/selftests/i915_selftest.c
index fee76c1d2f45..889281819c5b 100644
--- a/drivers/gpu/drm/i915/selftests/i915_selftest.c
+++ b/drivers/gpu/drm/i915/selftests/i915_selftest.c
@@ -23,7 +23,9 @@
 
 #include <linux/random.h>
 
+#include "gt/intel_gt.h"
 #include "gt/intel_gt_pm.h"
+#include "gt/intel_gt_regs.h"
 #include "gt/uc/intel_gsc_fw.h"
 
 #include "i915_driver.h"
@@ -253,11 +255,27 @@ int i915_mock_selftests(void)
 int i915_live_selftests(struct pci_dev *pdev)
 {
 	struct drm_i915_private *i915 = pdev_to_i915(pdev);
+	struct intel_uncore *uncore = &i915->uncore;
 	int err;
+	u32 pg_enable;
+	intel_wakeref_t wakeref;
 
 	if (!i915_selftest.live)
 		return 0;
 
+	/*
+	 * FIXME Disable render powergating, this is temporary wa and should be removed
+	 * after fixing real cause of forcewake timeouts.
+	 */
+	with_intel_runtime_pm(uncore->rpm, wakeref) {
+		if (IS_GFX_GT_IP_RANGE(to_gt(i915), IP_VER(12, 00), IP_VER(12, 74))) {
+			pg_enable = intel_uncore_read(uncore, GEN9_PG_ENABLE);
+			if (pg_enable & GEN9_RENDER_PG_ENABLE)
+				intel_uncore_write_fw(uncore, GEN9_PG_ENABLE,
+						      pg_enable & ~GEN9_RENDER_PG_ENABLE);
+		}
+	}
+
 	__wait_gsc_proxy_completed(i915);
 	__wait_gsc_huc_load_completed(i915);
 
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 9e310f4099f4..eee5c4f45a43 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -33,8 +33,14 @@ static const char *intel_dram_type_str(enum intel_dram_type type)
 		DRAM_TYPE_STR(DDR4),
 		DRAM_TYPE_STR(LPDDR3),
 		DRAM_TYPE_STR(LPDDR4),
+		DRAM_TYPE_STR(DDR5),
+		DRAM_TYPE_STR(LPDDR5),
+		DRAM_TYPE_STR(GDDR),
+		DRAM_TYPE_STR(GDDR_ECC),
 	};
 
+	BUILD_BUG_ON(ARRAY_SIZE(str) != __INTEL_DRAM_TYPE_MAX);
+
 	if (type >= ARRAY_SIZE(str))
 		type = INTEL_DRAM_UNKNOWN;
 
@@ -444,8 +450,6 @@ skl_get_dram_info(struct drm_i915_private *i915)
 	int ret;
 
 	dram_info->type = skl_get_dram_type(i915);
-	drm_dbg_kms(&i915->drm, "DRAM type: %s\n",
-		    intel_dram_type_str(dram_info->type));
 
 	ret = skl_dram_get_channels_info(i915);
 	if (ret)
@@ -560,10 +564,9 @@ static int bxt_get_dram_info(struct drm_i915_private *i915)
 			    dram_info->type != type);
 
 		drm_dbg_kms(&i915->drm,
-			    "CH%u DIMM size: %u Gb, width: X%u, ranks: %u, type: %s\n",
+			    "CH%u DIMM size: %u Gb, width: X%u, ranks: %u\n",
 			    i - BXT_D_CR_DRP0_DUNIT_START,
-			    dimm.size, dimm.width, dimm.ranks,
-			    intel_dram_type_str(type));
+			    dimm.size, dimm.width, dimm.ranks);
 
 		if (valid_ranks == 0)
 			valid_ranks = dimm.ranks;
@@ -687,6 +690,10 @@ static int xelpdp_get_dram_info(struct drm_i915_private *i915)
 		drm_WARN_ON(&i915->drm, !IS_DGFX(i915));
 		dram_info->type = INTEL_DRAM_GDDR;
 		break;
+	case 9:
+		drm_WARN_ON(&i915->drm, !IS_DGFX(i915));
+		dram_info->type = INTEL_DRAM_GDDR_ECC;
+		break;
 	default:
 		MISSING_CASE(val);
 		return -EINVAL;
@@ -726,6 +733,10 @@ void intel_dram_detect(struct drm_i915_private *i915)
 		ret = bxt_get_dram_info(i915);
 	else
 		ret = skl_get_dram_info(i915);
+
+	drm_dbg_kms(&i915->drm, "DRAM type: %s\n",
+		    intel_dram_type_str(dram_info->type));
+
 	if (ret)
 		return;
 
diff --git a/drivers/gpu/drm/i915/soc/intel_pch.c b/drivers/gpu/drm/i915/soc/intel_pch.c
deleted file mode 100644
index 82dc7fbd1a3e..000000000000
--- a/drivers/gpu/drm/i915/soc/intel_pch.c
+++ /dev/null
@@ -1,316 +0,0 @@
-// SPDX-License-Identifier: MIT
-/*
- * Copyright 2019 Intel Corporation.
- */
-
-#include "i915_drv.h"
-#include "i915_utils.h"
-#include "intel_pch.h"
-
-#define INTEL_PCH_DEVICE_ID_MASK		0xff80
-#define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
-#define INTEL_PCH_CPT_DEVICE_ID_TYPE		0x1c00
-#define INTEL_PCH_PPT_DEVICE_ID_TYPE		0x1e00
-#define INTEL_PCH_LPT_DEVICE_ID_TYPE		0x8c00
-#define INTEL_PCH_LPT_LP_DEVICE_ID_TYPE		0x9c00
-#define INTEL_PCH_WPT_DEVICE_ID_TYPE		0x8c80
-#define INTEL_PCH_WPT_LP_DEVICE_ID_TYPE		0x9c80
-#define INTEL_PCH_SPT_DEVICE_ID_TYPE		0xA100
-#define INTEL_PCH_SPT_LP_DEVICE_ID_TYPE		0x9D00
-#define INTEL_PCH_KBP_DEVICE_ID_TYPE		0xA280
-#define INTEL_PCH_CNP_DEVICE_ID_TYPE		0xA300
-#define INTEL_PCH_CNP_LP_DEVICE_ID_TYPE		0x9D80
-#define INTEL_PCH_CMP_DEVICE_ID_TYPE		0x0280
-#define INTEL_PCH_CMP2_DEVICE_ID_TYPE		0x0680
-#define INTEL_PCH_CMP_V_DEVICE_ID_TYPE		0xA380
-#define INTEL_PCH_ICP_DEVICE_ID_TYPE		0x3480
-#define INTEL_PCH_ICP2_DEVICE_ID_TYPE		0x3880
-#define INTEL_PCH_MCC_DEVICE_ID_TYPE		0x4B00
-#define INTEL_PCH_TGP_DEVICE_ID_TYPE		0xA080
-#define INTEL_PCH_TGP2_DEVICE_ID_TYPE		0x4380
-#define INTEL_PCH_JSP_DEVICE_ID_TYPE		0x4D80
-#define INTEL_PCH_ADP_DEVICE_ID_TYPE		0x7A80
-#define INTEL_PCH_ADP2_DEVICE_ID_TYPE		0x5180
-#define INTEL_PCH_ADP3_DEVICE_ID_TYPE		0x7A00
-#define INTEL_PCH_ADP4_DEVICE_ID_TYPE		0x5480
-#define INTEL_PCH_P2X_DEVICE_ID_TYPE		0x7100
-#define INTEL_PCH_P3X_DEVICE_ID_TYPE		0x7000
-#define INTEL_PCH_QEMU_DEVICE_ID_TYPE		0x2900 /* qemu q35 has 2918 */
-
-/* Map PCH device id to PCH type, or PCH_NONE if unknown. */
-static enum intel_pch
-intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id)
-{
-	switch (id) {
-	case INTEL_PCH_IBX_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Ibex Peak PCH\n");
-		drm_WARN_ON(&dev_priv->drm, GRAPHICS_VER(dev_priv) != 5);
-		return PCH_IBX;
-	case INTEL_PCH_CPT_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found CougarPoint PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    GRAPHICS_VER(dev_priv) != 6 && !IS_IVYBRIDGE(dev_priv));
-		return PCH_CPT;
-	case INTEL_PCH_PPT_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found PantherPoint PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    GRAPHICS_VER(dev_priv) != 6 && !IS_IVYBRIDGE(dev_priv));
-		/* PPT is CPT compatible */
-		return PCH_CPT;
-	case INTEL_PCH_LPT_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found LynxPoint PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv));
-		drm_WARN_ON(&dev_priv->drm,
-			    IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv));
-		return PCH_LPT_H;
-	case INTEL_PCH_LPT_LP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found LynxPoint LP PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv));
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_HASWELL_ULT(dev_priv) && !IS_BROADWELL_ULT(dev_priv));
-		return PCH_LPT_LP;
-	case INTEL_PCH_WPT_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found WildcatPoint PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv));
-		drm_WARN_ON(&dev_priv->drm,
-			    IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv));
-		/* WPT is LPT compatible */
-		return PCH_LPT_H;
-	case INTEL_PCH_WPT_LP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found WildcatPoint LP PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv));
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_HASWELL_ULT(dev_priv) && !IS_BROADWELL_ULT(dev_priv));
-		/* WPT is LPT compatible */
-		return PCH_LPT_LP;
-	case INTEL_PCH_SPT_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found SunrisePoint PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv));
-		return PCH_SPT;
-	case INTEL_PCH_SPT_LP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found SunrisePoint LP PCH\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_SKYLAKE(dev_priv) &&
-			    !IS_KABYLAKE(dev_priv) &&
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv));
-		return PCH_SPT;
-	case INTEL_PCH_KBP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Kaby Lake PCH (KBP)\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_SKYLAKE(dev_priv) &&
-			    !IS_KABYLAKE(dev_priv) &&
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv));
-		/* KBP is SPT compatible */
-		return PCH_SPT;
-	case INTEL_PCH_CNP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Cannon Lake PCH (CNP)\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv));
-		return PCH_CNP;
-	case INTEL_PCH_CNP_LP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm,
-			    "Found Cannon Lake LP PCH (CNP-LP)\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv));
-		return PCH_CNP;
-	case INTEL_PCH_CMP_DEVICE_ID_TYPE:
-	case INTEL_PCH_CMP2_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Comet Lake PCH (CMP)\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv) &&
-			    !IS_ROCKETLAKE(dev_priv));
-		/* CMP is CNP compatible */
-		return PCH_CNP;
-	case INTEL_PCH_CMP_V_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Comet Lake V PCH (CMP-V)\n");
-		drm_WARN_ON(&dev_priv->drm,
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv));
-		/* CMP-V is based on KBP, which is SPT compatible */
-		return PCH_SPT;
-	case INTEL_PCH_ICP_DEVICE_ID_TYPE:
-	case INTEL_PCH_ICP2_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Ice Lake PCH\n");
-		drm_WARN_ON(&dev_priv->drm, !IS_ICELAKE(dev_priv));
-		return PCH_ICP;
-	case INTEL_PCH_MCC_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Mule Creek Canyon PCH\n");
-		drm_WARN_ON(&dev_priv->drm, !(IS_JASPERLAKE(dev_priv) ||
-					      IS_ELKHARTLAKE(dev_priv)));
-		/* MCC is TGP compatible */
-		return PCH_TGP;
-	case INTEL_PCH_TGP_DEVICE_ID_TYPE:
-	case INTEL_PCH_TGP2_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Tiger Lake LP PCH\n");
-		drm_WARN_ON(&dev_priv->drm, !IS_TIGERLAKE(dev_priv) &&
-			    !IS_ROCKETLAKE(dev_priv) &&
-			    !IS_SKYLAKE(dev_priv) &&
-			    !IS_KABYLAKE(dev_priv) &&
-			    !IS_COFFEELAKE(dev_priv) &&
-			    !IS_COMETLAKE(dev_priv));
-		return PCH_TGP;
-	case INTEL_PCH_JSP_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Jasper Lake PCH\n");
-		drm_WARN_ON(&dev_priv->drm, !(IS_JASPERLAKE(dev_priv) ||
-					      IS_ELKHARTLAKE(dev_priv)));
-		/* JSP is ICP compatible */
-		return PCH_ICP;
-	case INTEL_PCH_ADP_DEVICE_ID_TYPE:
-	case INTEL_PCH_ADP2_DEVICE_ID_TYPE:
-	case INTEL_PCH_ADP3_DEVICE_ID_TYPE:
-	case INTEL_PCH_ADP4_DEVICE_ID_TYPE:
-		drm_dbg_kms(&dev_priv->drm, "Found Alder Lake PCH\n");
-		drm_WARN_ON(&dev_priv->drm, !IS_ALDERLAKE_S(dev_priv) &&
-			    !IS_ALDERLAKE_P(dev_priv));
-		return PCH_ADP;
-	default:
-		return PCH_NONE;
-	}
-}
-
-static bool intel_is_virt_pch(unsigned short id,
-			      unsigned short svendor, unsigned short sdevice)
-{
-	return (id == INTEL_PCH_P2X_DEVICE_ID_TYPE ||
-		id == INTEL_PCH_P3X_DEVICE_ID_TYPE ||
-		(id == INTEL_PCH_QEMU_DEVICE_ID_TYPE &&
-		 svendor == PCI_SUBVENDOR_ID_REDHAT_QUMRANET &&
-		 sdevice == PCI_SUBDEVICE_ID_QEMU));
-}
-
-static void
-intel_virt_detect_pch(const struct drm_i915_private *dev_priv,
-		      unsigned short *pch_id, enum intel_pch *pch_type)
-{
-	unsigned short id = 0;
-
-	/*
-	 * In a virtualized passthrough environment we can be in a
-	 * setup where the ISA bridge is not able to be passed through.
-	 * In this case, a south bridge can be emulated and we have to
-	 * make an educated guess as to which PCH is really there.
-	 */
-
-	if (IS_ALDERLAKE_S(dev_priv) || IS_ALDERLAKE_P(dev_priv))
-		id = INTEL_PCH_ADP_DEVICE_ID_TYPE;
-	else if (IS_TIGERLAKE(dev_priv) || IS_ROCKETLAKE(dev_priv))
-		id = INTEL_PCH_TGP_DEVICE_ID_TYPE;
-	else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv))
-		id = INTEL_PCH_MCC_DEVICE_ID_TYPE;
-	else if (IS_ICELAKE(dev_priv))
-		id = INTEL_PCH_ICP_DEVICE_ID_TYPE;
-	else if (IS_COFFEELAKE(dev_priv) ||
-		 IS_COMETLAKE(dev_priv))
-		id = INTEL_PCH_CNP_DEVICE_ID_TYPE;
-	else if (IS_KABYLAKE(dev_priv) || IS_SKYLAKE(dev_priv))
-		id = INTEL_PCH_SPT_DEVICE_ID_TYPE;
-	else if (IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv))
-		id = INTEL_PCH_LPT_LP_DEVICE_ID_TYPE;
-	else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
-		id = INTEL_PCH_LPT_DEVICE_ID_TYPE;
-	else if (GRAPHICS_VER(dev_priv) == 6 || IS_IVYBRIDGE(dev_priv))
-		id = INTEL_PCH_CPT_DEVICE_ID_TYPE;
-	else if (GRAPHICS_VER(dev_priv) == 5)
-		id = INTEL_PCH_IBX_DEVICE_ID_TYPE;
-
-	if (id)
-		drm_dbg_kms(&dev_priv->drm, "Assuming PCH ID %04x\n", id);
-	else
-		drm_dbg_kms(&dev_priv->drm, "Assuming no PCH\n");
-
-	*pch_type = intel_pch_type(dev_priv, id);
-
-	/* Sanity check virtual PCH id */
-	if (drm_WARN_ON(&dev_priv->drm,
-			id && *pch_type == PCH_NONE))
-		id = 0;
-
-	*pch_id = id;
-}
-
-void intel_detect_pch(struct drm_i915_private *dev_priv)
-{
-	struct pci_dev *pch = NULL;
-	unsigned short id;
-	enum intel_pch pch_type;
-
-	/*
-	 * South display engine on the same PCI device: just assign the fake
-	 * PCH.
-	 */
-	if (DISPLAY_VER(dev_priv) >= 20) {
-		dev_priv->pch_type = PCH_LNL;
-		return;
-	} else if (IS_BATTLEMAGE(dev_priv) || IS_METEORLAKE(dev_priv)) {
-		/*
-		 * Both north display and south display are on the SoC die.
-		 * The real PCH (if it even exists) is uninvolved in display.
-		 */
-		dev_priv->pch_type = PCH_MTL;
-		return;
-	} else if (IS_DG2(dev_priv)) {
-		dev_priv->pch_type = PCH_DG2;
-		return;
-	} else if (IS_DG1(dev_priv)) {
-		dev_priv->pch_type = PCH_DG1;
-		return;
-	}
-
-	/*
-	 * The reason to probe ISA bridge instead of Dev31:Fun0 is to
-	 * make graphics device passthrough work easy for VMM, that only
-	 * need to expose ISA bridge to let driver know the real hardware
-	 * underneath. This is a requirement from virtualization team.
-	 *
-	 * In some virtualized environments (e.g. XEN), there is irrelevant
-	 * ISA bridge in the system. To work reliably, we should scan through
-	 * all the ISA bridge devices and check for the first match, instead
-	 * of only checking the first one.
-	 */
-	while ((pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, pch))) {
-		if (pch->vendor != PCI_VENDOR_ID_INTEL)
-			continue;
-
-		id = pch->device & INTEL_PCH_DEVICE_ID_MASK;
-
-		pch_type = intel_pch_type(dev_priv, id);
-		if (pch_type != PCH_NONE) {
-			dev_priv->pch_type = pch_type;
-			break;
-		} else if (intel_is_virt_pch(id, pch->subsystem_vendor,
-					     pch->subsystem_device)) {
-			intel_virt_detect_pch(dev_priv, &id, &pch_type);
-			dev_priv->pch_type = pch_type;
-			break;
-		}
-	}
-
-	/*
-	 * Use PCH_NOP (PCH but no South Display) for PCH platforms without
-	 * display.
-	 */
-	if (pch && !HAS_DISPLAY(dev_priv)) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "Display disabled, reverting to NOP PCH\n");
-		dev_priv->pch_type = PCH_NOP;
-	} else if (!pch) {
-		if (i915_run_as_guest() && HAS_DISPLAY(dev_priv)) {
-			intel_virt_detect_pch(dev_priv, &id, &pch_type);
-			dev_priv->pch_type = pch_type;
-		} else {
-			drm_dbg_kms(&dev_priv->drm, "No PCH found.\n");
-		}
-	}
-
-	pci_dev_put(pch);
-}
diff --git a/drivers/gpu/drm/i915/soc/intel_pch.h b/drivers/gpu/drm/i915/soc/intel_pch.h
deleted file mode 100644
index 635aea7a5539..000000000000
--- a/drivers/gpu/drm/i915/soc/intel_pch.h
+++ /dev/null
@@ -1,56 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright 2019 Intel Corporation.
- */
-
-#ifndef __INTEL_PCH__
-#define __INTEL_PCH__
-
-struct drm_i915_private;
-
-/*
- * Sorted by south display engine compatibility.
- * If the new PCH comes with a south display engine that is not
- * inherited from the latest item, please do not add it to the
- * end. Instead, add it right after its "parent" PCH.
- */
-enum intel_pch {
-	PCH_NOP = -1,	/* PCH without south display */
-	PCH_NONE = 0,	/* No PCH present */
-	PCH_IBX,	/* Ibexpeak PCH */
-	PCH_CPT,	/* Cougarpoint/Pantherpoint PCH */
-	PCH_LPT_H,	/* Lynxpoint/Wildcatpoint H PCH */
-	PCH_LPT_LP,	/* Lynxpoint/Wildcatpoint LP PCH */
-	PCH_SPT,        /* Sunrisepoint/Kaby Lake PCH */
-	PCH_CNP,        /* Cannon/Comet Lake PCH */
-	PCH_ICP,	/* Ice Lake/Jasper Lake PCH */
-	PCH_TGP,	/* Tiger Lake/Mule Creek Canyon PCH */
-	PCH_ADP,	/* Alder Lake PCH */
-
-	/* Fake PCHs, functionality handled on the same PCI dev */
-	PCH_DG1 = 1024,
-	PCH_DG2,
-	PCH_MTL,
-	PCH_LNL,
-};
-
-#define INTEL_PCH_TYPE(dev_priv)		((dev_priv)->pch_type)
-#define HAS_PCH_DG2(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_DG2)
-#define HAS_PCH_ADP(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_ADP)
-#define HAS_PCH_DG1(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_DG1)
-#define HAS_PCH_TGP(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_TGP)
-#define HAS_PCH_ICP(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_ICP)
-#define HAS_PCH_CNP(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_CNP)
-#define HAS_PCH_SPT(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_SPT)
-#define HAS_PCH_LPT_H(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_LPT_H)
-#define HAS_PCH_LPT_LP(dev_priv)		(INTEL_PCH_TYPE(dev_priv) == PCH_LPT_LP)
-#define HAS_PCH_LPT(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_LPT_H || \
-						 INTEL_PCH_TYPE(dev_priv) == PCH_LPT_LP)
-#define HAS_PCH_CPT(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_CPT)
-#define HAS_PCH_IBX(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_IBX)
-#define HAS_PCH_NOP(dev_priv)			(INTEL_PCH_TYPE(dev_priv) == PCH_NOP)
-#define HAS_PCH_SPLIT(dev_priv)			(INTEL_PCH_TYPE(dev_priv) != PCH_NONE)
-
-void intel_detect_pch(struct drm_i915_private *dev_priv);
-
-#endif /* __INTEL_PCH__ */
diff --git a/drivers/gpu/drm/imagination/Makefile b/drivers/gpu/drm/imagination/Makefile
index 3d9d4d40fb80..7cca66f00a38 100644
--- a/drivers/gpu/drm/imagination/Makefile
+++ b/drivers/gpu/drm/imagination/Makefile
@@ -12,8 +12,10 @@ powervr-y := \
 	pvr_fw.o \
 	pvr_fw_meta.o \
 	pvr_fw_mips.o \
+	pvr_fw_riscv.o \
 	pvr_fw_startstop.o \
 	pvr_fw_trace.o \
+	pvr_fw_util.o \
 	pvr_gem.o \
 	pvr_hwrt.o \
 	pvr_job.o \
diff --git a/drivers/gpu/drm/imagination/pvr_debugfs.c b/drivers/gpu/drm/imagination/pvr_debugfs.c
index 6b77c9b4bde8..c7ce7daaa87a 100644
--- a/drivers/gpu/drm/imagination/pvr_debugfs.c
+++ b/drivers/gpu/drm/imagination/pvr_debugfs.c
@@ -28,9 +28,8 @@ pvr_debugfs_init(struct drm_minor *minor)
 	struct drm_device *drm_dev = minor->dev;
 	struct pvr_device *pvr_dev = to_pvr_device(drm_dev);
 	struct dentry *root = minor->debugfs_root;
-	size_t i;
 
-	for (i = 0; i < ARRAY_SIZE(pvr_debugfs_entries); ++i) {
+	for (size_t i = 0; i < ARRAY_SIZE(pvr_debugfs_entries); ++i) {
 		const struct pvr_debugfs_entry *entry = &pvr_debugfs_entries[i];
 		struct dentry *dir;
 
diff --git a/drivers/gpu/drm/imagination/pvr_device.c b/drivers/gpu/drm/imagination/pvr_device.c
index 1704c0268589..8b9ba4983c4c 100644
--- a/drivers/gpu/drm/imagination/pvr_device.c
+++ b/drivers/gpu/drm/imagination/pvr_device.c
@@ -25,6 +25,7 @@
 #include <linux/interrupt.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/reset.h>
 #include <linux/slab.h>
 #include <linux/stddef.h>
 #include <linux/types.h>
@@ -120,6 +121,21 @@ static int pvr_device_clk_init(struct pvr_device *pvr_dev)
 	return 0;
 }
 
+static int pvr_device_reset_init(struct pvr_device *pvr_dev)
+{
+	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
+	struct reset_control *reset;
+
+	reset = devm_reset_control_get_optional_exclusive(drm_dev->dev, NULL);
+	if (IS_ERR(reset))
+		return dev_err_probe(drm_dev->dev, PTR_ERR(reset),
+				     "failed to get gpu reset line\n");
+
+	pvr_dev->reset = reset;
+
+	return 0;
+}
+
 /**
  * pvr_device_process_active_queues() - Process all queue related events.
  * @pvr_dev: PowerVR device to check
@@ -146,9 +162,61 @@ static void pvr_device_process_active_queues(struct pvr_device *pvr_dev)
 	mutex_unlock(&pvr_dev->queues.lock);
 }
 
+static bool pvr_device_safety_irq_pending(struct pvr_device *pvr_dev)
+{
+	u32 events;
+
+	WARN_ON_ONCE(!pvr_dev->has_safety_events);
+
+	events = pvr_cr_read32(pvr_dev, ROGUE_CR_EVENT_STATUS);
+
+	return (events & ROGUE_CR_EVENT_STATUS_SAFETY_EN) != 0;
+}
+
+static void pvr_device_safety_irq_clear(struct pvr_device *pvr_dev)
+{
+	WARN_ON_ONCE(!pvr_dev->has_safety_events);
+
+	pvr_cr_write32(pvr_dev, ROGUE_CR_EVENT_CLEAR,
+		       ROGUE_CR_EVENT_CLEAR_SAFETY_EN);
+}
+
+static void pvr_device_handle_safety_events(struct pvr_device *pvr_dev)
+{
+	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
+	u32 events;
+
+	WARN_ON_ONCE(!pvr_dev->has_safety_events);
+
+	events = pvr_cr_read32(pvr_dev, ROGUE_CR_SAFETY_EVENT_STATUS__ROGUEXE);
+
+	/* Handle only these events on the host and leave the rest to the FW. */
+	events &= ROGUE_CR_SAFETY_EVENT_STATUS__ROGUEXE__FAULT_FW_EN |
+		ROGUE_CR_SAFETY_EVENT_STATUS__ROGUEXE__WATCHDOG_TIMEOUT_EN;
+
+	pvr_cr_write32(pvr_dev, ROGUE_CR_SAFETY_EVENT_CLEAR__ROGUEXE, events);
+
+	if (events & ROGUE_CR_SAFETY_EVENT_STATUS__ROGUEXE__FAULT_FW_EN) {
+		u32 fault_fw = pvr_cr_read32(pvr_dev, ROGUE_CR_FAULT_FW_STATUS);
+
+		pvr_cr_write32(pvr_dev, ROGUE_CR_FAULT_FW_CLEAR, fault_fw);
+
+		drm_info(drm_dev, "Safety event: FW fault (mask=0x%08x)\n", fault_fw);
+	}
+
+	if (events & ROGUE_CR_SAFETY_EVENT_STATUS__ROGUEXE__WATCHDOG_TIMEOUT_EN) {
+		/*
+		 * The watchdog timer is disabled by the driver so this event
+		 * should never be fired.
+		 */
+		drm_info(drm_dev, "Safety event: Watchdog timeout\n");
+	}
+}
+
 static irqreturn_t pvr_device_irq_thread_handler(int irq, void *data)
 {
 	struct pvr_device *pvr_dev = data;
+	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
 	irqreturn_t ret = IRQ_NONE;
 
 	/* We are in the threaded handler, we can keep dequeuing events until we
@@ -164,30 +232,76 @@ static irqreturn_t pvr_device_irq_thread_handler(int irq, void *data)
 			pvr_device_process_active_queues(pvr_dev);
 		}
 
-		pm_runtime_mark_last_busy(from_pvr_device(pvr_dev)->dev);
+		pm_runtime_mark_last_busy(drm_dev->dev);
 
 		ret = IRQ_HANDLED;
 	}
 
-	/* Unmask FW irqs before returning, so new interrupts can be received. */
-	pvr_fw_irq_enable(pvr_dev);
+	if (pvr_dev->has_safety_events) {
+		int err;
+
+		/*
+		 * Ensure the GPU is powered on since some safety events (such
+		 * as ECC faults) can happen outside of job submissions, which
+		 * are otherwise the only time a power reference is held.
+		 */
+		err = pvr_power_get(pvr_dev);
+		if (err) {
+			drm_err_ratelimited(drm_dev,
+					    "%s: could not take power reference (%d)\n",
+					    __func__, err);
+			return ret;
+		}
+
+		while (pvr_device_safety_irq_pending(pvr_dev)) {
+			pvr_device_safety_irq_clear(pvr_dev);
+			pvr_device_handle_safety_events(pvr_dev);
+
+			ret = IRQ_HANDLED;
+		}
+
+		pvr_power_put(pvr_dev);
+	}
+
 	return ret;
 }
 
 static irqreturn_t pvr_device_irq_handler(int irq, void *data)
 {
 	struct pvr_device *pvr_dev = data;
+	bool safety_irq_pending = false;
+
+	if (pvr_dev->has_safety_events)
+		safety_irq_pending = pvr_device_safety_irq_pending(pvr_dev);
 
-	if (!pvr_fw_irq_pending(pvr_dev))
+	if (!pvr_fw_irq_pending(pvr_dev) && !safety_irq_pending)
 		return IRQ_NONE; /* Spurious IRQ - ignore. */
 
-	/* Mask the FW interrupts before waking up the thread. Will be unmasked
-	 * when the thread handler is done processing events.
-	 */
-	pvr_fw_irq_disable(pvr_dev);
 	return IRQ_WAKE_THREAD;
 }
 
+static void pvr_device_safety_irq_init(struct pvr_device *pvr_dev)
+{
+	u32 num_ecc_rams = 0;
+
+	/*
+	 * Safety events are an optional feature of the RogueXE platform. They
+	 * are only enabled if at least one of ECC memory or the watchdog timer
+	 * are present in HW. While safety events can be generated by other
+	 * systems, that will never happen if the above mentioned hardware is
+	 * not present.
+	 */
+	if (!PVR_HAS_FEATURE(pvr_dev, roguexe)) {
+		pvr_dev->has_safety_events = false;
+		return;
+	}
+
+	PVR_FEATURE_VALUE(pvr_dev, ecc_rams, &num_ecc_rams);
+
+	pvr_dev->has_safety_events =
+		num_ecc_rams > 0 || PVR_HAS_FEATURE(pvr_dev, watchdog_timer);
+}
+
 /**
  * pvr_device_irq_init() - Initialise IRQ required by a PowerVR device
  * @pvr_dev: Target PowerVR device.
@@ -205,17 +319,25 @@ pvr_device_irq_init(struct pvr_device *pvr_dev)
 
 	init_waitqueue_head(&pvr_dev->kccb.rtn_q);
 
+	pvr_device_safety_irq_init(pvr_dev);
+
 	pvr_dev->irq = platform_get_irq(plat_dev, 0);
 	if (pvr_dev->irq < 0)
 		return pvr_dev->irq;
 
 	/* Clear any pending events before requesting the IRQ line. */
 	pvr_fw_irq_clear(pvr_dev);
-	pvr_fw_irq_enable(pvr_dev);
 
+	if (pvr_dev->has_safety_events)
+		pvr_device_safety_irq_clear(pvr_dev);
+
+	/*
+	 * The ONESHOT flag ensures IRQs are masked while the thread handler is
+	 * running.
+	 */
 	return request_threaded_irq(pvr_dev->irq, pvr_device_irq_handler,
 				    pvr_device_irq_thread_handler,
-				    IRQF_SHARED, "gpu", pvr_dev);
+				    IRQF_SHARED | IRQF_ONESHOT, "gpu", pvr_dev);
 }
 
 /**
@@ -509,6 +631,11 @@ pvr_device_init(struct pvr_device *pvr_dev)
 	if (err)
 		return err;
 
+	/* Get the reset line for the GPU */
+	err = pvr_device_reset_init(pvr_dev);
+	if (err)
+		return err;
+
 	/* Explicitly power the GPU so we can access control registers before the FW is booted. */
 	err = pm_runtime_resume_and_get(dev);
 	if (err)
diff --git a/drivers/gpu/drm/imagination/pvr_device.h b/drivers/gpu/drm/imagination/pvr_device.h
index 6d0dfacb677b..7cb01c38d2a9 100644
--- a/drivers/gpu/drm/imagination/pvr_device.h
+++ b/drivers/gpu/drm/imagination/pvr_device.h
@@ -18,6 +18,7 @@
 #include <linux/bits.h>
 #include <linux/compiler_attributes.h>
 #include <linux/compiler_types.h>
+#include <linux/device.h>
 #include <linux/io.h>
 #include <linux/iopoll.h>
 #include <linux/kernel.h>
@@ -131,6 +132,22 @@ struct pvr_device {
 	 */
 	struct clk *mem_clk;
 
+	struct pvr_device_power {
+		struct device **domain_devs;
+		struct device_link **domain_links;
+
+		u32 domain_count;
+	} power;
+
+	/**
+	 * @reset: Optional reset line.
+	 *
+	 * This may be used on some platforms to provide a reset line that needs to be de-asserted
+	 * after power-up procedure. It would also need to be asserted after the power-down
+	 * procedure.
+	 */
+	struct reset_control *reset;
+
 	/** @irq: IRQ number. */
 	int irq;
 
@@ -300,6 +317,9 @@ struct pvr_device {
 	 *  struct pvr_file.
 	 */
 	spinlock_t ctx_list_lock;
+
+	/** @has_safety_events: Whether this device can raise safety events. */
+	bool has_safety_events;
 };
 
 /**
@@ -728,8 +748,22 @@ pvr_ioctl_union_padding_check(void *instance, size_t union_offset,
 					      __union_size, __member_size);  \
 	})
 
-#define PVR_FW_PROCESSOR_TYPE_META  0
-#define PVR_FW_PROCESSOR_TYPE_MIPS  1
-#define PVR_FW_PROCESSOR_TYPE_RISCV 2
+/*
+ * These utility functions should more properly be placed in pvr_fw.h, but that
+ * would cause a dependency cycle between that header and this one. Since
+ * they're primarily used in pvr_device.c, let's put them in here for now.
+ */
+
+static __always_inline bool
+pvr_fw_irq_pending(struct pvr_device *pvr_dev)
+{
+	return pvr_dev->fw_dev.defs->irq_pending(pvr_dev);
+}
+
+static __always_inline void
+pvr_fw_irq_clear(struct pvr_device *pvr_dev)
+{
+	pvr_dev->fw_dev.defs->irq_clear(pvr_dev);
+}
 
 #endif /* PVR_DEVICE_H */
diff --git a/drivers/gpu/drm/imagination/pvr_drv.c b/drivers/gpu/drm/imagination/pvr_drv.c
index 0639502137b4..b058ec183bb3 100644
--- a/drivers/gpu/drm/imagination/pvr_drv.c
+++ b/drivers/gpu/drm/imagination/pvr_drv.c
@@ -44,6 +44,7 @@
  * This driver supports the following PowerVR/IMG graphics cores from Imagination Technologies:
  *
  * * AXE-1-16M (found in Texas Instruments AM62)
+ * * BXS-4-64 MC1 (found in Texas Instruments J721S2/AM68)
  */
 
 /**
@@ -1411,6 +1412,10 @@ pvr_probe(struct platform_device *plat_dev)
 
 	platform_set_drvdata(plat_dev, drm_dev);
 
+	err = pvr_power_domains_init(pvr_dev);
+	if (err)
+		return err;
+
 	init_rwsem(&pvr_dev->reset_sem);
 
 	pvr_context_device_init(pvr_dev);
@@ -1450,6 +1455,8 @@ err_watchdog_fini:
 err_context_fini:
 	pvr_context_device_fini(pvr_dev);
 
+	pvr_power_domains_fini(pvr_dev);
+
 	return err;
 }
 
@@ -1470,9 +1477,17 @@ static void pvr_remove(struct platform_device *plat_dev)
 	pvr_watchdog_fini(pvr_dev);
 	pvr_queue_device_fini(pvr_dev);
 	pvr_context_device_fini(pvr_dev);
+	pvr_power_domains_fini(pvr_dev);
 }
 
 static const struct of_device_id dt_match[] = {
+	{ .compatible = "img,img-rogue", .data = NULL },
+
+	/*
+	 * This legacy compatible string was introduced early on before the more generic
+	 * "img,img-rogue" was added. Keep it around here for compatibility, but never use
+	 * "img,img-axe" in new devicetrees.
+	 */
 	{ .compatible = "img,img-axe", .data = NULL },
 	{}
 };
@@ -1498,3 +1513,4 @@ MODULE_DESCRIPTION(PVR_DRIVER_DESC);
 MODULE_LICENSE("Dual MIT/GPL");
 MODULE_IMPORT_NS("DMA_BUF");
 MODULE_FIRMWARE("powervr/rogue_33.15.11.3_v1.fw");
+MODULE_FIRMWARE("powervr/rogue_36.53.104.796_v1.fw");
diff --git a/drivers/gpu/drm/imagination/pvr_free_list.c b/drivers/gpu/drm/imagination/pvr_free_list.c
index 5e51bc980751..5228e214491c 100644
--- a/drivers/gpu/drm/imagination/pvr_free_list.c
+++ b/drivers/gpu/drm/imagination/pvr_free_list.c
@@ -237,11 +237,10 @@ pvr_free_list_insert_pages_locked(struct pvr_free_list *free_list,
 		dma_addr_t dma_addr = sg_page_iter_dma_address(&dma_iter);
 		u64 dma_pfn = dma_addr >>
 			       ROGUE_BIF_PM_PHYSICAL_PAGE_ALIGNSHIFT;
-		u32 dma_addr_offset;
 
 		BUILD_BUG_ON(ROGUE_BIF_PM_PHYSICAL_PAGE_SIZE > PAGE_SIZE);
 
-		for (dma_addr_offset = 0; dma_addr_offset < PAGE_SIZE;
+		for (u32 dma_addr_offset = 0; dma_addr_offset < PAGE_SIZE;
 		     dma_addr_offset += ROGUE_BIF_PM_PHYSICAL_PAGE_SIZE) {
 			WARN_ON_ONCE(dma_pfn >> 32);
 
diff --git a/drivers/gpu/drm/imagination/pvr_fw.c b/drivers/gpu/drm/imagination/pvr_fw.c
index 3debc9870a82..b2f8cba77346 100644
--- a/drivers/gpu/drm/imagination/pvr_fw.c
+++ b/drivers/gpu/drm/imagination/pvr_fw.c
@@ -50,9 +50,8 @@ pvr_fw_find_layout_entry(struct pvr_device *pvr_dev, enum pvr_fw_section_id id)
 {
 	const struct pvr_fw_layout_entry *layout_entries = pvr_dev->fw_dev.layout_entries;
 	u32 num_layout_entries = pvr_dev->fw_dev.header->layout_entry_num;
-	u32 entry;
 
-	for (entry = 0; entry < num_layout_entries; entry++) {
+	for (u32 entry = 0; entry < num_layout_entries; entry++) {
 		if (layout_entries[entry].id == id)
 			return &layout_entries[entry];
 	}
@@ -65,9 +64,8 @@ pvr_fw_find_private_data(struct pvr_device *pvr_dev)
 {
 	const struct pvr_fw_layout_entry *layout_entries = pvr_dev->fw_dev.layout_entries;
 	u32 num_layout_entries = pvr_dev->fw_dev.header->layout_entry_num;
-	u32 entry;
 
-	for (entry = 0; entry < num_layout_entries; entry++) {
+	for (u32 entry = 0; entry < num_layout_entries; entry++) {
 		if (layout_entries[entry].id == META_PRIVATE_DATA ||
 		    layout_entries[entry].id == MIPS_PRIVATE_DATA ||
 		    layout_entries[entry].id == RISCV_PRIVATE_DATA)
@@ -97,7 +95,6 @@ pvr_fw_validate(struct pvr_device *pvr_dev)
 	const u8 *fw = firmware->data;
 	u32 fw_offset = firmware->size - SZ_4K;
 	u32 layout_table_size;
-	u32 entry;
 
 	if (firmware->size < SZ_4K || (firmware->size % FW_BLOCK_SIZE))
 		return -EINVAL;
@@ -144,7 +141,7 @@ pvr_fw_validate(struct pvr_device *pvr_dev)
 		return -EINVAL;
 
 	layout_entries = (const struct pvr_fw_layout_entry *)&fw[fw_offset];
-	for (entry = 0; entry < header->layout_entry_num; entry++) {
+	for (u32 entry = 0; entry < header->layout_entry_num; entry++) {
 		u32 start_addr = layout_entries[entry].base_addr;
 		u32 end_addr = start_addr + layout_entries[entry].alloc_size;
 
@@ -233,13 +230,12 @@ pvr_fw_find_mmu_segment(struct pvr_device *pvr_dev, u32 addr, u32 size, void *fw
 	const struct pvr_fw_layout_entry *layout_entries = pvr_dev->fw_dev.layout_entries;
 	u32 num_layout_entries = pvr_dev->fw_dev.header->layout_entry_num;
 	u32 end_addr = addr + size;
-	int entry = 0;
 
 	/* Ensure requested range is not zero, and size is not causing addr to overflow. */
 	if (end_addr <= addr)
 		return -EINVAL;
 
-	for (entry = 0; entry < num_layout_entries; entry++) {
+	for (int entry = 0; entry < num_layout_entries; entry++) {
 		u32 entry_start_addr = layout_entries[entry].base_addr;
 		u32 entry_end_addr = entry_start_addr + layout_entries[entry].alloc_size;
 
@@ -441,6 +437,9 @@ fw_runtime_cfg_init(void *cpu_ptr, void *priv)
 	runtime_cfg->active_pm_latency_persistant = true;
 	WARN_ON(PVR_FEATURE_VALUE(pvr_dev, num_clusters,
 				  &runtime_cfg->default_dusts_num_init) != 0);
+
+	/* Keep watchdog timer disabled. */
+	runtime_cfg->wdg_period_us = 0;
 }
 
 static void
@@ -663,7 +662,7 @@ pvr_fw_process(struct pvr_device *pvr_dev)
 		return PTR_ERR(fw_code_ptr);
 	}
 
-	if (pvr_dev->fw_dev.defs->has_fixed_data_addr()) {
+	if (pvr_dev->fw_dev.defs->has_fixed_data_addr) {
 		u32 base_addr = private_data->base_addr & pvr_dev->fw_dev.fw_heap_info.offset_mask;
 
 		fw_data_ptr =
@@ -732,7 +731,7 @@ pvr_fw_process(struct pvr_device *pvr_dev)
 					       fw_mem->core_data, fw_mem->core_code_alloc_size);
 
 	if (err)
-		goto err_free_fw_core_data_obj;
+		goto err_free_kdata;
 
 	memcpy(fw_code_ptr, fw_mem->code, fw_mem->code_alloc_size);
 	memcpy(fw_data_ptr, fw_mem->data, fw_mem->data_alloc_size);
@@ -742,10 +741,14 @@ pvr_fw_process(struct pvr_device *pvr_dev)
 		memcpy(fw_core_data_ptr, fw_mem->core_data, fw_mem->core_data_alloc_size);
 
 	/* We're finished with the firmware section memory on the CPU, unmap. */
-	if (fw_core_data_ptr)
+	if (fw_core_data_ptr) {
 		pvr_fw_object_vunmap(fw_mem->core_data_obj);
-	if (fw_core_code_ptr)
+		fw_core_data_ptr = NULL;
+	}
+	if (fw_core_code_ptr) {
 		pvr_fw_object_vunmap(fw_mem->core_code_obj);
+		fw_core_code_ptr = NULL;
+	}
 	pvr_fw_object_vunmap(fw_mem->data_obj);
 	fw_data_ptr = NULL;
 	pvr_fw_object_vunmap(fw_mem->code_obj);
@@ -753,7 +756,7 @@ pvr_fw_process(struct pvr_device *pvr_dev)
 
 	err = pvr_fw_create_fwif_connection_ctl(pvr_dev);
 	if (err)
-		goto err_free_fw_core_data_obj;
+		goto err_free_kdata;
 
 	return 0;
 
@@ -763,13 +766,16 @@ err_free_kdata:
 	kfree(fw_mem->data);
 	kfree(fw_mem->code);
 
-err_free_fw_core_data_obj:
 	if (fw_core_data_ptr)
-		pvr_fw_object_unmap_and_destroy(fw_mem->core_data_obj);
+		pvr_fw_object_vunmap(fw_mem->core_data_obj);
+	if (fw_mem->core_data_obj)
+		pvr_fw_object_destroy(fw_mem->core_data_obj);
 
 err_free_fw_core_code_obj:
 	if (fw_core_code_ptr)
-		pvr_fw_object_unmap_and_destroy(fw_mem->core_code_obj);
+		pvr_fw_object_vunmap(fw_mem->core_code_obj);
+	if (fw_mem->core_code_obj)
+		pvr_fw_object_destroy(fw_mem->core_code_obj);
 
 err_free_fw_data_obj:
 	if (fw_data_ptr)
@@ -836,6 +842,12 @@ pvr_fw_cleanup(struct pvr_device *pvr_dev)
 	struct pvr_fw_mem *fw_mem = &pvr_dev->fw_dev.mem;
 
 	pvr_fw_fini_fwif_connection_ctl(pvr_dev);
+
+	kfree(fw_mem->core_data);
+	kfree(fw_mem->core_code);
+	kfree(fw_mem->data);
+	kfree(fw_mem->code);
+
 	if (fw_mem->core_code_obj)
 		pvr_fw_object_destroy(fw_mem->core_code_obj);
 	if (fw_mem->core_data_obj)
@@ -926,18 +938,22 @@ pvr_fw_validate_init_device_info(struct pvr_device *pvr_dev)
 int
 pvr_fw_init(struct pvr_device *pvr_dev)
 {
+	static const struct pvr_fw_defs *fw_defs[PVR_FW_PROCESSOR_TYPE_COUNT] = {
+		[PVR_FW_PROCESSOR_TYPE_META] = &pvr_fw_defs_meta,
+		[PVR_FW_PROCESSOR_TYPE_MIPS] = &pvr_fw_defs_mips,
+		[PVR_FW_PROCESSOR_TYPE_RISCV] = &pvr_fw_defs_riscv,
+	};
+
 	u32 kccb_size_log2 = ROGUE_FWIF_KCCB_NUMCMDS_LOG2_DEFAULT;
 	u32 kccb_rtn_size = (1 << kccb_size_log2) * sizeof(*pvr_dev->kccb.rtn);
 	struct pvr_fw_device *fw_dev = &pvr_dev->fw_dev;
 	int err;
 
-	if (fw_dev->processor_type == PVR_FW_PROCESSOR_TYPE_META)
-		fw_dev->defs = &pvr_fw_defs_meta;
-	else if (fw_dev->processor_type == PVR_FW_PROCESSOR_TYPE_MIPS)
-		fw_dev->defs = &pvr_fw_defs_mips;
-	else
+	if (fw_dev->processor_type >= PVR_FW_PROCESSOR_TYPE_COUNT)
 		return -EINVAL;
 
+	fw_dev->defs = fw_defs[fw_dev->processor_type];
+
 	err = fw_dev->defs->init(pvr_dev);
 	if (err)
 		return err;
@@ -1443,6 +1459,15 @@ void pvr_fw_object_get_fw_addr_offset(struct pvr_fw_object *fw_obj, u32 offset,
 	*fw_addr_out = pvr_dev->fw_dev.defs->get_fw_addr_with_offset(fw_obj, offset);
 }
 
+u64
+pvr_fw_obj_get_gpu_addr(struct pvr_fw_object *fw_obj)
+{
+	struct pvr_device *pvr_dev = to_pvr_device(gem_from_pvr_gem(fw_obj->gem)->dev);
+	struct pvr_fw_device *fw_dev = &pvr_dev->fw_dev;
+
+	return fw_dev->fw_heap_info.gpu_addr + fw_obj->fw_addr_offset;
+}
+
 /*
  * pvr_fw_hard_reset() - Re-initialise the FW code and data segments, and reset all global FW
  *                       structures
diff --git a/drivers/gpu/drm/imagination/pvr_fw.h b/drivers/gpu/drm/imagination/pvr_fw.h
index b7966bd574a9..1404dd492d7c 100644
--- a/drivers/gpu/drm/imagination/pvr_fw.h
+++ b/drivers/gpu/drm/imagination/pvr_fw.h
@@ -167,47 +167,30 @@ struct pvr_fw_defs {
 	int (*wrapper_init)(struct pvr_device *pvr_dev);
 
 	/**
-	 * @has_fixed_data_addr:
+	 * @irq_pending: Check interrupt status register for pending interrupts.
 	 *
-	 * Called to check if firmware fixed data must be loaded at the address given by the
-	 * firmware layout table.
+	 * @pvr_dev: Target PowerVR device.
 	 *
 	 * This function is mandatory.
+	 */
+	bool (*irq_pending)(struct pvr_device *pvr_dev);
+
+	/**
+	 * @irq_clear: Clear pending interrupts.
 	 *
-	 * Returns:
-	 *  * %true if firmware fixed data must be loaded at the address given by the firmware
-	 *    layout table.
-	 *  * %false otherwise.
+	 * @pvr_dev: Target PowerVR device.
+	 *
+	 * This function is mandatory.
 	 */
-	bool (*has_fixed_data_addr)(void);
+	void (*irq_clear)(struct pvr_device *pvr_dev);
 
 	/**
-	 * @irq: FW Interrupt information.
+	 * @has_fixed_data_addr: Specify whether the firmware fixed data must be loaded at the
+	 *                       address given by the firmware layout table.
 	 *
-	 * Those are processor dependent, and should be initialized by the
-	 * processor backend in pvr_fw_funcs::init().
+	 * This value is mandatory.
 	 */
-	struct {
-		/** @enable_reg: FW interrupt enable register. */
-		u32 enable_reg;
-
-		/** @status_reg: FW interrupt status register. */
-		u32 status_reg;
-
-		/**
-		 * @clear_reg: FW interrupt clear register.
-		 *
-		 * If @status_reg == @clear_reg, we clear by write a bit to zero,
-		 * otherwise we clear by writing a bit to one.
-		 */
-		u32 clear_reg;
-
-		/** @event_mask: Bitmask of events to listen for. */
-		u32 event_mask;
-
-		/** @clear_mask: Value to write to the clear_reg in order to clear FW IRQs. */
-		u32 clear_mask;
-	} irq;
+	bool has_fixed_data_addr;
 };
 
 /**
@@ -400,26 +383,16 @@ struct pvr_fw_device {
 	} fw_objs;
 };
 
-#define pvr_fw_irq_read_reg(pvr_dev, name) \
-	pvr_cr_read32((pvr_dev), (pvr_dev)->fw_dev.defs->irq.name ## _reg)
-
-#define pvr_fw_irq_write_reg(pvr_dev, name, value) \
-	pvr_cr_write32((pvr_dev), (pvr_dev)->fw_dev.defs->irq.name ## _reg, value)
-
-#define pvr_fw_irq_pending(pvr_dev) \
-	(pvr_fw_irq_read_reg(pvr_dev, status) & (pvr_dev)->fw_dev.defs->irq.event_mask)
-
-#define pvr_fw_irq_clear(pvr_dev) \
-	pvr_fw_irq_write_reg(pvr_dev, clear, (pvr_dev)->fw_dev.defs->irq.clear_mask)
-
-#define pvr_fw_irq_enable(pvr_dev) \
-	pvr_fw_irq_write_reg(pvr_dev, enable, (pvr_dev)->fw_dev.defs->irq.event_mask)
-
-#define pvr_fw_irq_disable(pvr_dev) \
-	pvr_fw_irq_write_reg(pvr_dev, enable, 0)
+enum pvr_fw_processor_type {
+	PVR_FW_PROCESSOR_TYPE_META = 0,
+	PVR_FW_PROCESSOR_TYPE_MIPS,
+	PVR_FW_PROCESSOR_TYPE_RISCV,
+	PVR_FW_PROCESSOR_TYPE_COUNT,
+};
 
 extern const struct pvr_fw_defs pvr_fw_defs_meta;
 extern const struct pvr_fw_defs pvr_fw_defs_mips;
+extern const struct pvr_fw_defs pvr_fw_defs_riscv;
 
 int pvr_fw_validate_init_device_info(struct pvr_device *pvr_dev);
 int pvr_fw_init(struct pvr_device *pvr_dev);
@@ -506,4 +479,18 @@ pvr_fw_object_get_fw_addr(struct pvr_fw_object *fw_obj, u32 *fw_addr_out)
 	pvr_fw_object_get_fw_addr_offset(fw_obj, 0, fw_addr_out);
 }
 
+u64
+pvr_fw_obj_get_gpu_addr(struct pvr_fw_object *fw_obj);
+
+static __always_inline size_t
+pvr_fw_obj_get_object_size(struct pvr_fw_object *fw_obj)
+{
+	return pvr_gem_object_size(fw_obj->gem);
+}
+
+/* Util functions defined in pvr_fw_util.c. These are intended for use in pvr_fw_<arch>.c files. */
+int
+pvr_fw_process_elf_command_stream(struct pvr_device *pvr_dev, const u8 *fw, u8 *fw_code_ptr,
+				  u8 *fw_data_ptr, u8 *fw_core_code_ptr, u8 *fw_core_data_ptr);
+
 #endif /* PVR_FW_H */
diff --git a/drivers/gpu/drm/imagination/pvr_fw_meta.c b/drivers/gpu/drm/imagination/pvr_fw_meta.c
index 6d13864851fc..60db3668ad3c 100644
--- a/drivers/gpu/drm/imagination/pvr_fw_meta.c
+++ b/drivers/gpu/drm/imagination/pvr_fw_meta.c
@@ -370,13 +370,12 @@ configure_seg_mmu(struct pvr_device *pvr_dev, u32 **boot_conf_ptr)
 	const struct pvr_fw_layout_entry *layout_entries = pvr_dev->fw_dev.layout_entries;
 	u32 num_layout_entries = pvr_dev->fw_dev.header->layout_entry_num;
 	u64 seg_out_addr_top;
-	u32 i;
 
 	seg_out_addr_top =
 		ROGUE_FW_SEGMMU_OUTADDR_TOP_SLC(MMU_CONTEXT_MAPPING_FWPRIV,
 						ROGUE_FW_SEGMMU_META_BIFDM_ID);
 
-	for (i = 0; i < num_layout_entries; i++) {
+	for (u32 i = 0; i < num_layout_entries; i++) {
 		/*
 		 * FW code is using the bootloader segment which is already
 		 * configured on boot. FW coremem code and data don't use the
@@ -534,9 +533,17 @@ pvr_meta_vm_unmap(struct pvr_device *pvr_dev, struct pvr_fw_object *fw_obj)
 }
 
 static bool
-pvr_meta_has_fixed_data_addr(void)
+pvr_meta_irq_pending(struct pvr_device *pvr_dev)
 {
-	return false;
+	return pvr_cr_read32(pvr_dev, ROGUE_CR_META_SP_MSLVIRQSTATUS) &
+	       ROGUE_CR_META_SP_MSLVIRQSTATUS_TRIGVECT2_EN;
+}
+
+static void
+pvr_meta_irq_clear(struct pvr_device *pvr_dev)
+{
+	pvr_cr_write32(pvr_dev, ROGUE_CR_META_SP_MSLVIRQSTATUS,
+		       ROGUE_CR_META_SP_MSLVIRQSTATUS_TRIGVECT2_CLRMSK);
 }
 
 const struct pvr_fw_defs pvr_fw_defs_meta = {
@@ -546,12 +553,7 @@ const struct pvr_fw_defs pvr_fw_defs_meta = {
 	.vm_unmap = pvr_meta_vm_unmap,
 	.get_fw_addr_with_offset = pvr_meta_get_fw_addr_with_offset,
 	.wrapper_init = pvr_meta_wrapper_init,
-	.has_fixed_data_addr = pvr_meta_has_fixed_data_addr,
-	.irq = {
-		.enable_reg = ROGUE_CR_META_SP_MSLVIRQENABLE,
-		.status_reg = ROGUE_CR_META_SP_MSLVIRQSTATUS,
-		.clear_reg = ROGUE_CR_META_SP_MSLVIRQSTATUS,
-		.event_mask = ROGUE_CR_META_SP_MSLVIRQSTATUS_TRIGVECT2_EN,
-		.clear_mask = ROGUE_CR_META_SP_MSLVIRQSTATUS_TRIGVECT2_CLRMSK,
-	},
+	.irq_pending = pvr_meta_irq_pending,
+	.irq_clear = pvr_meta_irq_clear,
+	.has_fixed_data_addr = false,
 };
diff --git a/drivers/gpu/drm/imagination/pvr_fw_mips.c b/drivers/gpu/drm/imagination/pvr_fw_mips.c
index 0bed0257e2ab..6914fc46db50 100644
--- a/drivers/gpu/drm/imagination/pvr_fw_mips.c
+++ b/drivers/gpu/drm/imagination/pvr_fw_mips.c
@@ -8,7 +8,6 @@
 #include "pvr_rogue_mips.h"
 #include "pvr_vm_mips.h"
 
-#include <linux/elf.h>
 #include <linux/err.h>
 #include <linux/types.h>
 
@@ -16,60 +15,6 @@
 #define ROGUE_FW_HEAP_MIPS_SHIFT 24 /* 16 MB */
 #define ROGUE_FW_HEAP_MIPS_RESERVED_SIZE SZ_1M
 
-/**
- * process_elf_command_stream() - Process ELF firmware image and populate
- *                                firmware sections
- * @pvr_dev: Device pointer.
- * @fw: Pointer to firmware image.
- * @fw_code_ptr: Pointer to FW code section.
- * @fw_data_ptr: Pointer to FW data section.
- * @fw_core_code_ptr: Pointer to FW coremem code section.
- * @fw_core_data_ptr: Pointer to FW coremem data section.
- *
- * Returns :
- *  * 0 on success, or
- *  * -EINVAL on any error in ELF command stream.
- */
-static int
-process_elf_command_stream(struct pvr_device *pvr_dev, const u8 *fw, u8 *fw_code_ptr,
-			   u8 *fw_data_ptr, u8 *fw_core_code_ptr, u8 *fw_core_data_ptr)
-{
-	struct elf32_hdr *header = (struct elf32_hdr *)fw;
-	struct elf32_phdr *program_header = (struct elf32_phdr *)(fw + header->e_phoff);
-	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
-	u32 entry;
-	int err;
-
-	for (entry = 0; entry < header->e_phnum; entry++, program_header++) {
-		void *write_addr;
-
-		/* Only consider loadable entries in the ELF segment table */
-		if (program_header->p_type != PT_LOAD)
-			continue;
-
-		err = pvr_fw_find_mmu_segment(pvr_dev, program_header->p_vaddr,
-					      program_header->p_memsz, fw_code_ptr, fw_data_ptr,
-					      fw_core_code_ptr, fw_core_data_ptr, &write_addr);
-		if (err) {
-			drm_err(drm_dev,
-				"Addr 0x%x (size: %d) not found in any firmware segment",
-				program_header->p_vaddr, program_header->p_memsz);
-			return err;
-		}
-
-		/* Write to FW allocation only if available */
-		if (write_addr) {
-			memcpy(write_addr, fw + program_header->p_offset,
-			       program_header->p_filesz);
-
-			memset((u8 *)write_addr + program_header->p_filesz, 0,
-			       program_header->p_memsz - program_header->p_filesz);
-		}
-	}
-
-	return 0;
-}
-
 static int
 pvr_mips_init(struct pvr_device *pvr_dev)
 {
@@ -97,11 +42,10 @@ pvr_mips_fw_process(struct pvr_device *pvr_dev, const u8 *fw,
 	const struct pvr_fw_layout_entry *stack_entry;
 	struct rogue_mipsfw_boot_data *boot_data;
 	dma_addr_t dma_addr;
-	u32 page_nr;
 	int err;
 
-	err = process_elf_command_stream(pvr_dev, fw, fw_code_ptr, fw_data_ptr, fw_core_code_ptr,
-					 fw_core_data_ptr);
+	err = pvr_fw_process_elf_command_stream(pvr_dev, fw, fw_code_ptr, fw_data_ptr,
+						fw_core_code_ptr, fw_core_data_ptr);
 	if (err)
 		return err;
 
@@ -132,7 +76,7 @@ pvr_mips_fw_process(struct pvr_device *pvr_dev, const u8 *fw,
 
 	boot_data->reg_base = pvr_dev->regs_resource->start;
 
-	for (page_nr = 0; page_nr < ARRAY_SIZE(boot_data->pt_phys_addr); page_nr++) {
+	for (u32 page_nr = 0; page_nr < ARRAY_SIZE(boot_data->pt_phys_addr); page_nr++) {
 		/* Firmware expects 4k pages, but host page size might be different. */
 		u32 src_page_nr = (page_nr * ROGUE_MIPSFW_PAGE_SIZE_4K) >> PAGE_SHIFT;
 		u32 page_offset = (page_nr * ROGUE_MIPSFW_PAGE_SIZE_4K) & ~PAGE_MASK;
@@ -228,9 +172,17 @@ pvr_mips_get_fw_addr_with_offset(struct pvr_fw_object *fw_obj, u32 offset)
 }
 
 static bool
-pvr_mips_has_fixed_data_addr(void)
+pvr_mips_irq_pending(struct pvr_device *pvr_dev)
+{
+	return pvr_cr_read32(pvr_dev, ROGUE_CR_MIPS_WRAPPER_IRQ_STATUS) &
+	       ROGUE_CR_MIPS_WRAPPER_IRQ_STATUS_EVENT_EN;
+}
+
+static void
+pvr_mips_irq_clear(struct pvr_device *pvr_dev)
 {
-	return true;
+	pvr_cr_write32(pvr_dev, ROGUE_CR_MIPS_WRAPPER_IRQ_CLEAR,
+		       ROGUE_CR_MIPS_WRAPPER_IRQ_CLEAR_EVENT_EN);
 }
 
 const struct pvr_fw_defs pvr_fw_defs_mips = {
@@ -241,12 +193,7 @@ const struct pvr_fw_defs pvr_fw_defs_mips = {
 	.vm_unmap = pvr_vm_mips_unmap,
 	.get_fw_addr_with_offset = pvr_mips_get_fw_addr_with_offset,
 	.wrapper_init = pvr_mips_wrapper_init,
-	.has_fixed_data_addr = pvr_mips_has_fixed_data_addr,
-	.irq = {
-		.enable_reg = ROGUE_CR_MIPS_WRAPPER_IRQ_ENABLE,
-		.status_reg = ROGUE_CR_MIPS_WRAPPER_IRQ_STATUS,
-		.clear_reg = ROGUE_CR_MIPS_WRAPPER_IRQ_CLEAR,
-		.event_mask = ROGUE_CR_MIPS_WRAPPER_IRQ_STATUS_EVENT_EN,
-		.clear_mask = ROGUE_CR_MIPS_WRAPPER_IRQ_CLEAR_EVENT_EN,
-	},
+	.irq_pending = pvr_mips_irq_pending,
+	.irq_clear = pvr_mips_irq_clear,
+	.has_fixed_data_addr = true,
 };
diff --git a/drivers/gpu/drm/imagination/pvr_fw_riscv.c b/drivers/gpu/drm/imagination/pvr_fw_riscv.c
new file mode 100644
index 000000000000..fc13d483be9a
--- /dev/null
+++ b/drivers/gpu/drm/imagination/pvr_fw_riscv.c
@@ -0,0 +1,165 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright (c) 2024 Imagination Technologies Ltd. */
+
+#include "pvr_device.h"
+#include "pvr_fw.h"
+#include "pvr_fw_info.h"
+#include "pvr_fw_mips.h"
+#include "pvr_gem.h"
+#include "pvr_rogue_cr_defs.h"
+#include "pvr_rogue_riscv.h"
+#include "pvr_vm.h"
+
+#include <linux/compiler.h>
+#include <linux/delay.h>
+#include <linux/firmware.h>
+#include <linux/ktime.h>
+#include <linux/types.h>
+
+#define ROGUE_FW_HEAP_RISCV_SHIFT 25 /* 32 MB */
+#define ROGUE_FW_HEAP_RISCV_SIZE (1u << ROGUE_FW_HEAP_RISCV_SHIFT)
+
+static int
+pvr_riscv_wrapper_init(struct pvr_device *pvr_dev)
+{
+	const u64 common_opts =
+		((u64)(ROGUE_FW_HEAP_RISCV_SIZE >> FWCORE_ADDR_REMAP_CONFIG0_SIZE_ALIGNSHIFT)
+		 << ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG0_SIZE_SHIFT) |
+		((u64)MMU_CONTEXT_MAPPING_FWPRIV
+		 << FWCORE_ADDR_REMAP_CONFIG0_MMU_CONTEXT_SHIFT);
+
+	u64 code_addr = pvr_fw_obj_get_gpu_addr(pvr_dev->fw_dev.mem.code_obj);
+	u64 data_addr = pvr_fw_obj_get_gpu_addr(pvr_dev->fw_dev.mem.data_obj);
+
+	/* This condition allows us to OR the addresses into the register directly. */
+	static_assert(ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG1_DEVVADDR_SHIFT ==
+		      ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG1_DEVVADDR_ALIGNSHIFT);
+
+	WARN_ON(code_addr & ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG1_DEVVADDR_CLRMSK);
+	WARN_ON(data_addr & ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG1_DEVVADDR_CLRMSK);
+
+	pvr_cr_write64(pvr_dev, ROGUE_RISCVFW_REGION_REMAP_CR(BOOTLDR_CODE),
+		       code_addr | common_opts | ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG0_FETCH_EN_EN);
+
+	pvr_cr_write64(pvr_dev, ROGUE_RISCVFW_REGION_REMAP_CR(BOOTLDR_DATA),
+		       data_addr | common_opts |
+			       ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG0_LOAD_STORE_EN_EN);
+
+	/* Garten IDLE bit controlled by RISC-V. */
+	pvr_cr_write64(pvr_dev, ROGUE_CR_MTS_GARTEN_WRAPPER_CONFIG,
+		       ROGUE_CR_MTS_GARTEN_WRAPPER_CONFIG_IDLE_CTRL_META);
+
+	return 0;
+}
+
+struct rogue_riscv_fw_boot_data {
+	u64 coremem_code_dev_vaddr;
+	u64 coremem_data_dev_vaddr;
+	u32 coremem_code_fw_addr;
+	u32 coremem_data_fw_addr;
+	u32 coremem_code_size;
+	u32 coremem_data_size;
+	u32 flags;
+	u32 reserved;
+};
+
+static int
+pvr_riscv_fw_process(struct pvr_device *pvr_dev, const u8 *fw,
+		     u8 *fw_code_ptr, u8 *fw_data_ptr, u8 *fw_core_code_ptr, u8 *fw_core_data_ptr,
+		     u32 core_code_alloc_size)
+{
+	struct pvr_fw_device *fw_dev = &pvr_dev->fw_dev;
+	struct pvr_fw_mem *fw_mem = &fw_dev->mem;
+	struct rogue_riscv_fw_boot_data *boot_data;
+	int err;
+
+	err = pvr_fw_process_elf_command_stream(pvr_dev, fw, fw_code_ptr, fw_data_ptr,
+						fw_core_code_ptr, fw_core_data_ptr);
+	if (err)
+		goto err_out;
+
+	boot_data = (struct rogue_riscv_fw_boot_data *)fw_data_ptr;
+
+	if (fw_mem->core_code_obj) {
+		boot_data->coremem_code_dev_vaddr = pvr_fw_obj_get_gpu_addr(fw_mem->core_code_obj);
+		pvr_fw_object_get_fw_addr(fw_mem->core_code_obj, &boot_data->coremem_code_fw_addr);
+		boot_data->coremem_code_size = pvr_fw_obj_get_object_size(fw_mem->core_code_obj);
+	}
+
+	if (fw_mem->core_data_obj) {
+		boot_data->coremem_data_dev_vaddr = pvr_fw_obj_get_gpu_addr(fw_mem->core_data_obj);
+		pvr_fw_object_get_fw_addr(fw_mem->core_data_obj, &boot_data->coremem_data_fw_addr);
+		boot_data->coremem_data_size = pvr_fw_obj_get_object_size(fw_mem->core_data_obj);
+	}
+
+	return 0;
+
+err_out:
+	return err;
+}
+
+static int
+pvr_riscv_init(struct pvr_device *pvr_dev)
+{
+	pvr_fw_heap_info_init(pvr_dev, ROGUE_FW_HEAP_RISCV_SHIFT, 0);
+
+	return 0;
+}
+
+static u32
+pvr_riscv_get_fw_addr_with_offset(struct pvr_fw_object *fw_obj, u32 offset)
+{
+	u32 fw_addr = fw_obj->fw_addr_offset + offset;
+
+	/* RISC-V cacheability is determined by address. */
+	if (fw_obj->gem->flags & PVR_BO_FW_FLAGS_DEVICE_UNCACHED)
+		fw_addr |= ROGUE_RISCVFW_REGION_BASE(SHARED_UNCACHED_DATA);
+	else
+		fw_addr |= ROGUE_RISCVFW_REGION_BASE(SHARED_CACHED_DATA);
+
+	return fw_addr;
+}
+
+static int
+pvr_riscv_vm_map(struct pvr_device *pvr_dev, struct pvr_fw_object *fw_obj)
+{
+	struct pvr_gem_object *pvr_obj = fw_obj->gem;
+
+	return pvr_vm_map(pvr_dev->kernel_vm_ctx, pvr_obj, 0, fw_obj->fw_mm_node.start,
+			  pvr_gem_object_size(pvr_obj));
+}
+
+static void
+pvr_riscv_vm_unmap(struct pvr_device *pvr_dev, struct pvr_fw_object *fw_obj)
+{
+	struct pvr_gem_object *pvr_obj = fw_obj->gem;
+
+	pvr_vm_unmap_obj(pvr_dev->kernel_vm_ctx, pvr_obj,
+			 fw_obj->fw_mm_node.start, fw_obj->fw_mm_node.size);
+}
+
+static bool
+pvr_riscv_irq_pending(struct pvr_device *pvr_dev)
+{
+	return pvr_cr_read32(pvr_dev, ROGUE_CR_IRQ_OS0_EVENT_STATUS) &
+	       ROGUE_CR_IRQ_OS0_EVENT_STATUS_SOURCE_EN;
+}
+
+static void
+pvr_riscv_irq_clear(struct pvr_device *pvr_dev)
+{
+	pvr_cr_write32(pvr_dev, ROGUE_CR_IRQ_OS0_EVENT_CLEAR,
+		       ROGUE_CR_IRQ_OS0_EVENT_CLEAR_SOURCE_EN);
+}
+
+const struct pvr_fw_defs pvr_fw_defs_riscv = {
+	.init = pvr_riscv_init,
+	.fw_process = pvr_riscv_fw_process,
+	.vm_map = pvr_riscv_vm_map,
+	.vm_unmap = pvr_riscv_vm_unmap,
+	.get_fw_addr_with_offset = pvr_riscv_get_fw_addr_with_offset,
+	.wrapper_init = pvr_riscv_wrapper_init,
+	.irq_pending = pvr_riscv_irq_pending,
+	.irq_clear = pvr_riscv_irq_clear,
+	.has_fixed_data_addr = false,
+};
diff --git a/drivers/gpu/drm/imagination/pvr_fw_startstop.c b/drivers/gpu/drm/imagination/pvr_fw_startstop.c
index 36cec227cfe3..dcbb9903e791 100644
--- a/drivers/gpu/drm/imagination/pvr_fw_startstop.c
+++ b/drivers/gpu/drm/imagination/pvr_fw_startstop.c
@@ -49,6 +49,14 @@ rogue_bif_init(struct pvr_device *pvr_dev)
 
 	pvr_cr_write64(pvr_dev, BIF_CAT_BASEX(MMU_CONTEXT_MAPPING_FWPRIV),
 		       pc_addr);
+
+	if (pvr_dev->fw_dev.processor_type == PVR_FW_PROCESSOR_TYPE_RISCV) {
+		pc_addr = (((u64)pc_dma_addr >> ROGUE_CR_FWCORE_MEM_CAT_BASE0_ADDR_ALIGNSHIFT)
+			   << ROGUE_CR_FWCORE_MEM_CAT_BASE0_ADDR_SHIFT) &
+			  ~ROGUE_CR_FWCORE_MEM_CAT_BASE0_ADDR_CLRMSK;
+
+		pvr_cr_write64(pvr_dev, FWCORE_MEM_CAT_BASEX(MMU_CONTEXT_MAPPING_FWPRIV), pc_addr);
+	}
 }
 
 static int
@@ -114,6 +122,9 @@ pvr_fw_start(struct pvr_device *pvr_dev)
 		(void)pvr_cr_read32(pvr_dev, ROGUE_CR_SYS_BUS_SECURE); /* Fence write */
 	}
 
+	if (pvr_dev->fw_dev.processor_type == PVR_FW_PROCESSOR_TYPE_RISCV)
+		pvr_cr_write32(pvr_dev, ROGUE_CR_FWCORE_BOOT, 0);
+
 	/* Set Rogue in soft-reset. */
 	pvr_cr_write64(pvr_dev, ROGUE_CR_SOFT_RESET, soft_reset_mask);
 	if (has_reset2)
@@ -167,6 +178,12 @@ pvr_fw_start(struct pvr_device *pvr_dev)
 	/* ... and afterwards. */
 	udelay(3);
 
+	if (pvr_dev->fw_dev.processor_type == PVR_FW_PROCESSOR_TYPE_RISCV) {
+		/* Boot the FW. */
+		pvr_cr_write32(pvr_dev, ROGUE_CR_FWCORE_BOOT, 1);
+		udelay(3);
+	}
+
 	return 0;
 
 err_reset:
diff --git a/drivers/gpu/drm/imagination/pvr_fw_trace.c b/drivers/gpu/drm/imagination/pvr_fw_trace.c
index 5dbb636d7d4f..a1098b521485 100644
--- a/drivers/gpu/drm/imagination/pvr_fw_trace.c
+++ b/drivers/gpu/drm/imagination/pvr_fw_trace.c
@@ -21,7 +21,6 @@ tracebuf_ctrl_init(void *cpu_ptr, void *priv)
 {
 	struct rogue_fwif_tracebuf *tracebuf_ctrl = cpu_ptr;
 	struct pvr_fw_trace *fw_trace = priv;
-	u32 thread_nr;
 
 	tracebuf_ctrl->tracebuf_size_in_dwords = ROGUE_FW_TRACE_BUF_DEFAULT_SIZE_IN_DWORDS;
 	tracebuf_ctrl->tracebuf_flags = 0;
@@ -31,7 +30,7 @@ tracebuf_ctrl_init(void *cpu_ptr, void *priv)
 	else
 		tracebuf_ctrl->log_type = ROGUE_FWIF_LOG_TYPE_NONE;
 
-	for (thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
+	for (u32 thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
 		struct rogue_fwif_tracebuf_space *tracebuf_space =
 			&tracebuf_ctrl->tracebuf[thread_nr];
 		struct pvr_fw_trace_buffer *trace_buffer = &fw_trace->buffers[thread_nr];
@@ -48,10 +47,9 @@ int pvr_fw_trace_init(struct pvr_device *pvr_dev)
 {
 	struct pvr_fw_trace *fw_trace = &pvr_dev->fw_dev.fw_trace;
 	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
-	u32 thread_nr;
 	int err;
 
-	for (thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
+	for (u32 thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
 		struct pvr_fw_trace_buffer *trace_buffer = &fw_trace->buffers[thread_nr];
 
 		trace_buffer->buf =
@@ -88,7 +86,7 @@ int pvr_fw_trace_init(struct pvr_device *pvr_dev)
 	BUILD_BUG_ON(ARRAY_SIZE(fw_trace->tracebuf_ctrl->tracebuf) !=
 		     ARRAY_SIZE(fw_trace->buffers));
 
-	for (thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
+	for (u32 thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
 		struct rogue_fwif_tracebuf_space *tracebuf_space =
 			&fw_trace->tracebuf_ctrl->tracebuf[thread_nr];
 		struct pvr_fw_trace_buffer *trace_buffer = &fw_trace->buffers[thread_nr];
@@ -99,7 +97,7 @@ int pvr_fw_trace_init(struct pvr_device *pvr_dev)
 	return 0;
 
 err_free_buf:
-	for (thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
+	for (u32 thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
 		struct pvr_fw_trace_buffer *trace_buffer = &fw_trace->buffers[thread_nr];
 
 		if (trace_buffer->buf)
@@ -112,9 +110,8 @@ err_free_buf:
 void pvr_fw_trace_fini(struct pvr_device *pvr_dev)
 {
 	struct pvr_fw_trace *fw_trace = &pvr_dev->fw_dev.fw_trace;
-	u32 thread_nr;
 
-	for (thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
+	for (u32 thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); thread_nr++) {
 		struct pvr_fw_trace_buffer *trace_buffer = &fw_trace->buffers[thread_nr];
 
 		pvr_fw_object_unmap_and_destroy(trace_buffer->buf_obj);
@@ -122,8 +119,6 @@ void pvr_fw_trace_fini(struct pvr_device *pvr_dev)
 	pvr_fw_object_unmap_and_destroy(fw_trace->tracebuf_ctrl_obj);
 }
 
-#if defined(CONFIG_DEBUG_FS)
-
 /**
  * update_logtype() - Send KCCB command to trigger FW to update logtype
  * @pvr_dev: Target PowerVR device
@@ -184,9 +179,7 @@ struct pvr_fw_trace_seq_data {
 
 static u32 find_sfid(u32 id)
 {
-	u32 i;
-
-	for (i = 0; i < ARRAY_SIZE(stid_fmts); i++) {
+	for (u32 i = 0; i < ARRAY_SIZE(stid_fmts); i++) {
 		if (stid_fmts[i].id == id)
 			return i;
 	}
@@ -285,12 +278,11 @@ static void fw_trace_get_first(struct pvr_fw_trace_seq_data *trace_seq_data)
 static void *fw_trace_seq_start(struct seq_file *s, loff_t *pos)
 {
 	struct pvr_fw_trace_seq_data *trace_seq_data = s->private;
-	u32 i;
 
 	/* Reset trace index, then advance to *pos. */
 	fw_trace_get_first(trace_seq_data);
 
-	for (i = 0; i < *pos; i++) {
+	for (u32 i = 0; i < *pos; i++) {
 		if (!fw_trace_get_next(trace_seq_data))
 			return NULL;
 	}
@@ -447,7 +439,7 @@ static const struct file_operations pvr_fw_trace_fops = {
 void
 pvr_fw_trace_mask_update(struct pvr_device *pvr_dev, u32 old_mask, u32 new_mask)
 {
-	if (old_mask != new_mask)
+	if (IS_ENABLED(CONFIG_DEBUG_FS) && old_mask != new_mask)
 		update_logtype(pvr_dev, new_mask);
 }
 
@@ -455,12 +447,14 @@ void
 pvr_fw_trace_debugfs_init(struct pvr_device *pvr_dev, struct dentry *dir)
 {
 	struct pvr_fw_trace *fw_trace = &pvr_dev->fw_dev.fw_trace;
-	u32 thread_nr;
+
+	if (!IS_ENABLED(CONFIG_DEBUG_FS))
+		return;
 
 	static_assert(ARRAY_SIZE(fw_trace->buffers) <= 10,
 		      "The filename buffer is only large enough for a single-digit thread count");
 
-	for (thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); ++thread_nr) {
+	for (u32 thread_nr = 0; thread_nr < ARRAY_SIZE(fw_trace->buffers); ++thread_nr) {
 		char filename[8];
 
 		snprintf(filename, ARRAY_SIZE(filename), "trace_%u", thread_nr);
@@ -469,4 +463,3 @@ pvr_fw_trace_debugfs_init(struct pvr_device *pvr_dev, struct dentry *dir)
 				    &pvr_fw_trace_fops);
 	}
 }
-#endif
diff --git a/drivers/gpu/drm/imagination/pvr_fw_trace.h b/drivers/gpu/drm/imagination/pvr_fw_trace.h
index 0074d2b18da0..1d0ef937427a 100644
--- a/drivers/gpu/drm/imagination/pvr_fw_trace.h
+++ b/drivers/gpu/drm/imagination/pvr_fw_trace.h
@@ -65,7 +65,6 @@ struct pvr_fw_trace {
 int pvr_fw_trace_init(struct pvr_device *pvr_dev);
 void pvr_fw_trace_fini(struct pvr_device *pvr_dev);
 
-#if defined(CONFIG_DEBUG_FS)
 /* Forward declaration from <linux/dcache.h>. */
 struct dentry;
 
@@ -73,6 +72,5 @@ void pvr_fw_trace_mask_update(struct pvr_device *pvr_dev, u32 old_mask,
 			      u32 new_mask);
 
 void pvr_fw_trace_debugfs_init(struct pvr_device *pvr_dev, struct dentry *dir);
-#endif /* defined(CONFIG_DEBUG_FS) */
 
 #endif /* PVR_FW_TRACE_H */
diff --git a/drivers/gpu/drm/imagination/pvr_fw_util.c b/drivers/gpu/drm/imagination/pvr_fw_util.c
new file mode 100644
index 000000000000..377fe72d86b8
--- /dev/null
+++ b/drivers/gpu/drm/imagination/pvr_fw_util.c
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0-only OR MIT
+/* Copyright (c) 2024 Imagination Technologies Ltd. */
+
+#include "pvr_device.h"
+#include "pvr_fw.h"
+
+#include <drm/drm_device.h>
+#include <drm/drm_print.h>
+
+#include <linux/elf.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+/**
+ * pvr_fw_process_elf_command_stream() - Process ELF firmware image and populate
+ *                                       firmware sections
+ * @pvr_dev: Device pointer.
+ * @fw: Pointer to firmware image.
+ * @fw_code_ptr: Pointer to FW code section.
+ * @fw_data_ptr: Pointer to FW data section.
+ * @fw_core_code_ptr: Pointer to FW coremem code section.
+ * @fw_core_data_ptr: Pointer to FW coremem data section.
+ *
+ * Returns :
+ *  * 0 on success, or
+ *  * -EINVAL on any error in ELF command stream.
+ */
+int
+pvr_fw_process_elf_command_stream(struct pvr_device *pvr_dev, const u8 *fw,
+				  u8 *fw_code_ptr, u8 *fw_data_ptr,
+				  u8 *fw_core_code_ptr, u8 *fw_core_data_ptr)
+{
+	struct elf32_hdr *header = (struct elf32_hdr *)fw;
+	struct elf32_phdr *program_header = (struct elf32_phdr *)(fw + header->e_phoff);
+	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
+	int err;
+
+	for (u32 entry = 0; entry < header->e_phnum; entry++, program_header++) {
+		void *write_addr;
+
+		/* Only consider loadable entries in the ELF segment table */
+		if (program_header->p_type != PT_LOAD)
+			continue;
+
+		err = pvr_fw_find_mmu_segment(pvr_dev, program_header->p_vaddr,
+					      program_header->p_memsz, fw_code_ptr, fw_data_ptr,
+					      fw_core_code_ptr, fw_core_data_ptr, &write_addr);
+		if (err) {
+			drm_err(drm_dev,
+				"Addr 0x%x (size: %d) not found in any firmware segment",
+				program_header->p_vaddr, program_header->p_memsz);
+			return err;
+		}
+
+		/* Write to FW allocation only if available */
+		if (write_addr) {
+			memcpy(write_addr, fw + program_header->p_offset,
+			       program_header->p_filesz);
+
+			memset((u8 *)write_addr + program_header->p_filesz, 0,
+			       program_header->p_memsz - program_header->p_filesz);
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/imagination/pvr_gem.c b/drivers/gpu/drm/imagination/pvr_gem.c
index 6a8c81fe8c1e..a66cf082af24 100644
--- a/drivers/gpu/drm/imagination/pvr_gem.c
+++ b/drivers/gpu/drm/imagination/pvr_gem.c
@@ -19,6 +19,7 @@
 #include <linux/log2.h>
 #include <linux/mutex.h>
 #include <linux/pagemap.h>
+#include <linux/property.h>
 #include <linux/refcount.h>
 #include <linux/scatterlist.h>
 
@@ -76,8 +77,6 @@ pvr_gem_object_flags_validate(u64 flags)
 		 DRM_PVR_BO_ALLOW_CPU_USERSPACE_ACCESS),
 	};
 
-	int i;
-
 	/*
 	 * Check for bits set in undefined regions. Reserved regions refer to
 	 * options that can only be set by the kernel. These are explicitly
@@ -91,7 +90,7 @@ pvr_gem_object_flags_validate(u64 flags)
 	 * Check for all combinations of flags marked as invalid in the array
 	 * above.
 	 */
-	for (i = 0; i < ARRAY_SIZE(invalid_combinations); ++i) {
+	for (int i = 0; i < ARRAY_SIZE(invalid_combinations); ++i) {
 		u64 combo = invalid_combinations[i];
 
 		if ((flags & combo) == combo)
@@ -203,7 +202,7 @@ pvr_gem_object_vmap(struct pvr_gem_object *pvr_obj)
 
 	dma_resv_lock(obj->resv, NULL);
 
-	err = drm_gem_shmem_vmap(shmem_obj, &map);
+	err = drm_gem_shmem_vmap_locked(shmem_obj, &map);
 	if (err)
 		goto err_unlock;
 
@@ -257,7 +256,7 @@ pvr_gem_object_vunmap(struct pvr_gem_object *pvr_obj)
 			dma_sync_sgtable_for_device(dev, shmem_obj->sgt, DMA_BIDIRECTIONAL);
 	}
 
-	drm_gem_shmem_vunmap(shmem_obj, &map);
+	drm_gem_shmem_vunmap_locked(shmem_obj, &map);
 
 	dma_resv_unlock(obj->resv);
 }
@@ -336,6 +335,7 @@ struct drm_gem_object *pvr_gem_create_object(struct drm_device *drm_dev, size_t
 struct pvr_gem_object *
 pvr_gem_object_create(struct pvr_device *pvr_dev, size_t size, u64 flags)
 {
+	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
 	struct drm_gem_shmem_object *shmem_obj;
 	struct pvr_gem_object *pvr_obj;
 	struct sg_table *sgt;
@@ -345,7 +345,10 @@ pvr_gem_object_create(struct pvr_device *pvr_dev, size_t size, u64 flags)
 	if (size == 0 || !pvr_gem_object_flags_validate(flags))
 		return ERR_PTR(-EINVAL);
 
-	shmem_obj = drm_gem_shmem_create(from_pvr_device(pvr_dev), size);
+	if (device_get_dma_attr(drm_dev->dev) == DEV_DMA_COHERENT)
+		flags |= PVR_BO_CPU_CACHED;
+
+	shmem_obj = drm_gem_shmem_create(drm_dev, size);
 	if (IS_ERR(shmem_obj))
 		return ERR_CAST(shmem_obj);
 
@@ -360,8 +363,7 @@ pvr_gem_object_create(struct pvr_device *pvr_dev, size_t size, u64 flags)
 		goto err_shmem_object_free;
 	}
 
-	dma_sync_sgtable_for_device(shmem_obj->base.dev->dev, sgt,
-				    DMA_BIDIRECTIONAL);
+	dma_sync_sgtable_for_device(drm_dev->dev, sgt, DMA_BIDIRECTIONAL);
 
 	/*
 	 * Do this last because pvr_gem_object_zero() requires a fully
diff --git a/drivers/gpu/drm/imagination/pvr_gem.h b/drivers/gpu/drm/imagination/pvr_gem.h
index e0e5ea509a2e..c99f30cc6208 100644
--- a/drivers/gpu/drm/imagination/pvr_gem.h
+++ b/drivers/gpu/drm/imagination/pvr_gem.h
@@ -44,8 +44,10 @@ struct pvr_file;
  * Bits not defined anywhere are "undefined".
  *
  * CPU mapping options
- *    :PVR_BO_CPU_CACHED: By default, all GEM objects are mapped write-combined on the CPU. Set this
- *       flag to override this behaviour and map the object cached.
+ *    :PVR_BO_CPU_CACHED: By default, all GEM objects are mapped write-combined on the CPU. Set
+ *       this flag to override this behaviour and map the object cached. If the dma_coherent
+ *       property is present in devicetree, all allocations will be mapped as if this flag was set.
+ *       This does not require any additional consideration at allocation time.
  *
  * Firmware options
  *    :PVR_BO_FW_NO_CLEAR_ON_RESET: By default, all FW objects are cleared and reinitialised on hard
diff --git a/drivers/gpu/drm/imagination/pvr_hwrt.c b/drivers/gpu/drm/imagination/pvr_hwrt.c
index 54f88d6c01e5..dc0c25fa1847 100644
--- a/drivers/gpu/drm/imagination/pvr_hwrt.c
+++ b/drivers/gpu/drm/imagination/pvr_hwrt.c
@@ -44,13 +44,12 @@ hwrt_init_kernel_structure(struct pvr_file *pvr_file,
 {
 	struct pvr_device *pvr_dev = pvr_file->pvr_dev;
 	int err;
-	int i;
 
 	hwrt->pvr_dev = pvr_dev;
 	hwrt->max_rts = args->layers;
 
 	/* Get pointers to the free lists */
-	for (i = 0; i < ARRAY_SIZE(hwrt->free_lists); i++) {
+	for (int i = 0; i < ARRAY_SIZE(hwrt->free_lists); i++) {
 		hwrt->free_lists[i] = pvr_free_list_lookup(pvr_file,  args->free_list_handles[i]);
 		if (!hwrt->free_lists[i]) {
 			err = -EINVAL;
@@ -67,7 +66,7 @@ hwrt_init_kernel_structure(struct pvr_file *pvr_file,
 	return 0;
 
 err_put_free_lists:
-	for (i = 0; i < ARRAY_SIZE(hwrt->free_lists); i++) {
+	for (int i = 0; i < ARRAY_SIZE(hwrt->free_lists); i++) {
 		pvr_free_list_put(hwrt->free_lists[i]);
 		hwrt->free_lists[i] = NULL;
 	}
@@ -78,9 +77,7 @@ err_put_free_lists:
 static void
 hwrt_fini_kernel_structure(struct pvr_hwrt_dataset *hwrt)
 {
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hwrt->free_lists); i++) {
+	for (int i = 0; i < ARRAY_SIZE(hwrt->free_lists); i++) {
 		pvr_free_list_put(hwrt->free_lists[i]);
 		hwrt->free_lists[i] = NULL;
 	}
@@ -363,13 +360,12 @@ hwrt_data_init_fw_structure(struct pvr_file *pvr_file,
 	struct drm_pvr_create_hwrt_geom_data_args *geom_data_args = &args->geom_data_args;
 	struct pvr_device *pvr_dev = pvr_file->pvr_dev;
 	struct rogue_fwif_rta_ctl *rta_ctl;
-	int free_list_i;
 	int err;
 
 	pvr_fw_object_get_fw_addr(hwrt->common_fw_obj,
 				  &hwrt_data->data.hwrt_data_common_fw_addr);
 
-	for (free_list_i = 0; free_list_i < ARRAY_SIZE(hwrt->free_lists); free_list_i++) {
+	for (int free_list_i = 0; free_list_i < ARRAY_SIZE(hwrt->free_lists); free_list_i++) {
 		pvr_fw_object_get_fw_addr(hwrt->free_lists[free_list_i]->fw_obj,
 					  &hwrt_data->data.freelists_fw_addr[free_list_i]);
 	}
diff --git a/drivers/gpu/drm/imagination/pvr_job.c b/drivers/gpu/drm/imagination/pvr_job.c
index 1cdb3cfd058d..59b334d094fa 100644
--- a/drivers/gpu/drm/imagination/pvr_job.c
+++ b/drivers/gpu/drm/imagination/pvr_job.c
@@ -671,6 +671,13 @@ pvr_jobs_link_geom_frag(struct pvr_job_data *job_data, u32 *job_count)
 		geom_job->paired_job = frag_job;
 		frag_job->paired_job = geom_job;
 
+		/* The geometry job pvr_job structure is used when the fragment
+		 * job is being prepared by the GPU scheduler. Have the fragment
+		 * job hold a reference on the geometry job to prevent it being
+		 * freed until the fragment job has finished with it.
+		 */
+		pvr_job_get(geom_job);
+
 		/* Skip the fragment job we just paired to the geometry job. */
 		i++;
 	}
diff --git a/drivers/gpu/drm/imagination/pvr_mmu.c b/drivers/gpu/drm/imagination/pvr_mmu.c
index 4fe70610ed94..450d476d183f 100644
--- a/drivers/gpu/drm/imagination/pvr_mmu.c
+++ b/drivers/gpu/drm/imagination/pvr_mmu.c
@@ -17,6 +17,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/kmemleak.h>
 #include <linux/minmax.h>
+#include <linux/property.h>
 #include <linux/sizes.h>
 
 #define PVR_SHIFT_FROM_SIZE(size_) (__builtin_ctzll(size_))
@@ -259,6 +260,7 @@ pvr_mmu_backing_page_init(struct pvr_mmu_backing_page *page,
 	struct device *dev = from_pvr_device(pvr_dev)->dev;
 
 	struct page *raw_page;
+	pgprot_t prot;
 	int err;
 
 	dma_addr_t dma_addr;
@@ -268,7 +270,11 @@ pvr_mmu_backing_page_init(struct pvr_mmu_backing_page *page,
 	if (!raw_page)
 		return -ENOMEM;
 
-	host_ptr = vmap(&raw_page, 1, VM_MAP, pgprot_writecombine(PAGE_KERNEL));
+	prot = PAGE_KERNEL;
+	if (device_get_dma_attr(dev) != DEV_DMA_COHERENT)
+		prot = pgprot_writecombine(prot);
+
+	host_ptr = vmap(&raw_page, 1, VM_MAP, prot);
 	if (!host_ptr) {
 		err = -ENOMEM;
 		goto err_free_page;
diff --git a/drivers/gpu/drm/imagination/pvr_power.c b/drivers/gpu/drm/imagination/pvr_power.c
index ba7816fd28ec..41f5d89e78b8 100644
--- a/drivers/gpu/drm/imagination/pvr_power.c
+++ b/drivers/gpu/drm/imagination/pvr_power.c
@@ -10,11 +10,15 @@
 
 #include <drm/drm_drv.h>
 #include <drm/drm_managed.h>
+#include <linux/cleanup.h>
 #include <linux/clk.h>
 #include <linux/interrupt.h>
 #include <linux/mutex.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
+#include <linux/pm_domain.h>
 #include <linux/pm_runtime.h>
+#include <linux/reset.h>
 #include <linux/timer.h>
 #include <linux/types.h>
 #include <linux/workqueue.h>
@@ -252,6 +256,8 @@ pvr_power_device_suspend(struct device *dev)
 	clk_disable_unprepare(pvr_dev->sys_clk);
 	clk_disable_unprepare(pvr_dev->core_clk);
 
+	err = reset_control_assert(pvr_dev->reset);
+
 err_drm_dev_exit:
 	drm_dev_exit(idx);
 
@@ -282,16 +288,33 @@ pvr_power_device_resume(struct device *dev)
 	if (err)
 		goto err_sys_clk_disable;
 
+	/*
+	 * According to the hardware manual, a delay of at least 32 clock
+	 * cycles is required between de-asserting the clkgen reset and
+	 * de-asserting the GPU reset. Assuming a worst-case scenario with
+	 * a very high GPU clock frequency, a delay of 1 microsecond is
+	 * sufficient to ensure this requirement is met across all
+	 * feasible GPU clock speeds.
+	 */
+	udelay(1);
+
+	err = reset_control_deassert(pvr_dev->reset);
+	if (err)
+		goto err_mem_clk_disable;
+
 	if (pvr_dev->fw_dev.booted) {
 		err = pvr_power_fw_enable(pvr_dev);
 		if (err)
-			goto err_mem_clk_disable;
+			goto err_reset_assert;
 	}
 
 	drm_dev_exit(idx);
 
 	return 0;
 
+err_reset_assert:
+	reset_control_assert(pvr_dev->reset);
+
 err_mem_clk_disable:
 	clk_disable_unprepare(pvr_dev->mem_clk);
 
@@ -431,3 +454,114 @@ pvr_watchdog_fini(struct pvr_device *pvr_dev)
 {
 	cancel_delayed_work_sync(&pvr_dev->watchdog.work);
 }
+
+int pvr_power_domains_init(struct pvr_device *pvr_dev)
+{
+	struct device *dev = from_pvr_device(pvr_dev)->dev;
+
+	struct device_link **domain_links __free(kfree) = NULL;
+	struct device **domain_devs __free(kfree) = NULL;
+	int domain_count;
+	int link_count;
+
+	char dev_name[2] = "a";
+	int err;
+	int i;
+
+	domain_count = of_count_phandle_with_args(dev->of_node, "power-domains",
+						  "#power-domain-cells");
+	if (domain_count < 0)
+		return domain_count;
+
+	if (domain_count <= 1)
+		return 0;
+
+	link_count = domain_count + (domain_count - 1);
+
+	domain_devs = kcalloc(domain_count, sizeof(*domain_devs), GFP_KERNEL);
+	if (!domain_devs)
+		return -ENOMEM;
+
+	domain_links = kcalloc(link_count, sizeof(*domain_links), GFP_KERNEL);
+	if (!domain_links)
+		return -ENOMEM;
+
+	for (i = 0; i < domain_count; i++) {
+		struct device *domain_dev;
+
+		dev_name[0] = 'a' + i;
+		domain_dev = dev_pm_domain_attach_by_name(dev, dev_name);
+		if (IS_ERR_OR_NULL(domain_dev)) {
+			err = domain_dev ? PTR_ERR(domain_dev) : -ENODEV;
+			goto err_detach;
+		}
+
+		domain_devs[i] = domain_dev;
+	}
+
+	for (i = 0; i < domain_count; i++) {
+		struct device_link *link;
+
+		link = device_link_add(dev, domain_devs[i], DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME);
+		if (!link) {
+			err = -ENODEV;
+			goto err_unlink;
+		}
+
+		domain_links[i] = link;
+	}
+
+	for (i = domain_count; i < link_count; i++) {
+		struct device_link *link;
+
+		link = device_link_add(domain_devs[i - domain_count + 1],
+				       domain_devs[i - domain_count],
+				       DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME);
+		if (!link) {
+			err = -ENODEV;
+			goto err_unlink;
+		}
+
+		domain_links[i] = link;
+	}
+
+	pvr_dev->power = (struct pvr_device_power){
+		.domain_devs = no_free_ptr(domain_devs),
+		.domain_links = no_free_ptr(domain_links),
+		.domain_count = domain_count,
+	};
+
+	return 0;
+
+err_unlink:
+	while (--i >= 0)
+		device_link_del(domain_links[i]);
+
+	i = domain_count;
+
+err_detach:
+	while (--i >= 0)
+		dev_pm_domain_detach(domain_devs[i], true);
+
+	return err;
+}
+
+void pvr_power_domains_fini(struct pvr_device *pvr_dev)
+{
+	const int domain_count = pvr_dev->power.domain_count;
+
+	int i = domain_count + (domain_count - 1);
+
+	while (--i >= 0)
+		device_link_del(pvr_dev->power.domain_links[i]);
+
+	i = domain_count;
+
+	while (--i >= 0)
+		dev_pm_domain_detach(pvr_dev->power.domain_devs[i], true);
+
+	kfree(pvr_dev->power.domain_links);
+	kfree(pvr_dev->power.domain_devs);
+
+	pvr_dev->power = (struct pvr_device_power){ 0 };
+}
diff --git a/drivers/gpu/drm/imagination/pvr_power.h b/drivers/gpu/drm/imagination/pvr_power.h
index 9a9312dcb2da..ada85674a7ca 100644
--- a/drivers/gpu/drm/imagination/pvr_power.h
+++ b/drivers/gpu/drm/imagination/pvr_power.h
@@ -38,4 +38,7 @@ pvr_power_put(struct pvr_device *pvr_dev)
 	return pm_runtime_put(drm_dev->dev);
 }
 
+int pvr_power_domains_init(struct pvr_device *pvr_dev);
+void pvr_power_domains_fini(struct pvr_device *pvr_dev);
+
 #endif /* PVR_POWER_H */
diff --git a/drivers/gpu/drm/imagination/pvr_queue.c b/drivers/gpu/drm/imagination/pvr_queue.c
index eba69309bb6c..5e9bc0992824 100644
--- a/drivers/gpu/drm/imagination/pvr_queue.c
+++ b/drivers/gpu/drm/imagination/pvr_queue.c
@@ -866,6 +866,10 @@ static void pvr_queue_free_job(struct drm_sched_job *sched_job)
 	struct pvr_job *job = container_of(sched_job, struct pvr_job, base);
 
 	drm_sched_job_cleanup(sched_job);
+
+	if (job->type == DRM_PVR_JOB_TYPE_FRAGMENT && job->paired_job)
+		pvr_job_put(job->paired_job);
+
 	job->paired_job = NULL;
 	pvr_job_put(job);
 }
diff --git a/drivers/gpu/drm/imagination/pvr_rogue_cr_defs.h b/drivers/gpu/drm/imagination/pvr_rogue_cr_defs.h
index 2a90d02796d3..790c97f80a2a 100644
--- a/drivers/gpu/drm/imagination/pvr_rogue_cr_defs.h
+++ b/drivers/gpu/drm/imagination/pvr_rogue_cr_defs.h
@@ -827,6 +827,120 @@
 #define ROGUE_CR_EVENT_STATUS_TLA_COMPLETE_CLRMSK 0xFFFFFFFEU
 #define ROGUE_CR_EVENT_STATUS_TLA_COMPLETE_EN 0x00000001U
 
+/* Register ROGUE_CR_EVENT_CLEAR */
+#define ROGUE_CR_EVENT_CLEAR 0x0138U
+#define ROGUE_CR_EVENT_CLEAR__ROGUEXE__MASKFULL 0x00000000E01DFFFFULL
+#define ROGUE_CR_EVENT_CLEAR__SIGNALS__MASKFULL 0x00000000E007FFFFULL
+#define ROGUE_CR_EVENT_CLEAR_MASKFULL 0x00000000FFFFFFFFULL
+#define ROGUE_CR_EVENT_CLEAR_TDM_FENCE_FINISHED_SHIFT 31U
+#define ROGUE_CR_EVENT_CLEAR_TDM_FENCE_FINISHED_CLRMSK 0x7FFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_TDM_FENCE_FINISHED_EN 0x80000000U
+#define ROGUE_CR_EVENT_CLEAR_TDM_BUFFER_STALL_SHIFT 30U
+#define ROGUE_CR_EVENT_CLEAR_TDM_BUFFER_STALL_CLRMSK 0xBFFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_TDM_BUFFER_STALL_EN 0x40000000U
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_SIGNAL_FAILURE_SHIFT 29U
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_SIGNAL_FAILURE_CLRMSK 0xDFFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_SIGNAL_FAILURE_EN 0x20000000U
+#define ROGUE_CR_EVENT_CLEAR_DPX_OUT_OF_MEMORY_SHIFT 28U
+#define ROGUE_CR_EVENT_CLEAR_DPX_OUT_OF_MEMORY_CLRMSK 0xEFFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_DPX_OUT_OF_MEMORY_EN 0x10000000U
+#define ROGUE_CR_EVENT_CLEAR_DPX_MMU_PAGE_FAULT_SHIFT 27U
+#define ROGUE_CR_EVENT_CLEAR_DPX_MMU_PAGE_FAULT_CLRMSK 0xF7FFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_DPX_MMU_PAGE_FAULT_EN 0x08000000U
+#define ROGUE_CR_EVENT_CLEAR_RPM_OUT_OF_MEMORY_SHIFT 26U
+#define ROGUE_CR_EVENT_CLEAR_RPM_OUT_OF_MEMORY_CLRMSK 0xFBFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_RPM_OUT_OF_MEMORY_EN 0x04000000U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC3_FINISHED_SHIFT 25U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC3_FINISHED_CLRMSK 0xFDFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC3_FINISHED_EN 0x02000000U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC2_FINISHED_SHIFT 24U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC2_FINISHED_CLRMSK 0xFEFFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC2_FINISHED_EN 0x01000000U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC1_FINISHED_SHIFT 23U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC1_FINISHED_CLRMSK 0xFF7FFFFFU
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC1_FINISHED_EN 0x00800000U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC0_FINISHED_SHIFT 22U
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC0_FINISHED_CLRMSK 0xFFBFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_FBA_FC0_FINISHED_EN 0x00400000U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC3_FINISHED_SHIFT 21U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC3_FINISHED_CLRMSK 0xFFDFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC3_FINISHED_EN 0x00200000U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC2_FINISHED_SHIFT 20U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC2_FINISHED_CLRMSK 0xFFEFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC2_FINISHED_EN 0x00100000U
+#define ROGUE_CR_EVENT_CLEAR_SAFETY_SHIFT 20U
+#define ROGUE_CR_EVENT_CLEAR_SAFETY_CLRMSK 0xFFEFFFFFU
+#define ROGUE_CR_EVENT_CLEAR_SAFETY_EN 0x00100000U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC1_FINISHED_SHIFT 19U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC1_FINISHED_CLRMSK 0xFFF7FFFFU
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC1_FINISHED_EN 0x00080000U
+#define ROGUE_CR_EVENT_CLEAR_SLAVE_REQ_SHIFT 19U
+#define ROGUE_CR_EVENT_CLEAR_SLAVE_REQ_CLRMSK 0xFFF7FFFFU
+#define ROGUE_CR_EVENT_CLEAR_SLAVE_REQ_EN 0x00080000U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC0_FINISHED_SHIFT 18U
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC0_FINISHED_CLRMSK 0xFFFBFFFFU
+#define ROGUE_CR_EVENT_CLEAR_RDM_FC0_FINISHED_EN 0x00040000U
+#define ROGUE_CR_EVENT_CLEAR_TDM_CONTEXT_STORE_FINISHED_SHIFT 18U
+#define ROGUE_CR_EVENT_CLEAR_TDM_CONTEXT_STORE_FINISHED_CLRMSK 0xFFFBFFFFU
+#define ROGUE_CR_EVENT_CLEAR_TDM_CONTEXT_STORE_FINISHED_EN 0x00040000U
+#define ROGUE_CR_EVENT_CLEAR_SHG_FINISHED_SHIFT 17U
+#define ROGUE_CR_EVENT_CLEAR_SHG_FINISHED_CLRMSK 0xFFFDFFFFU
+#define ROGUE_CR_EVENT_CLEAR_SHG_FINISHED_EN 0x00020000U
+#define ROGUE_CR_EVENT_CLEAR_SPFILTER_SIGNAL_UPDATE_SHIFT 17U
+#define ROGUE_CR_EVENT_CLEAR_SPFILTER_SIGNAL_UPDATE_CLRMSK 0xFFFDFFFFU
+#define ROGUE_CR_EVENT_CLEAR_SPFILTER_SIGNAL_UPDATE_EN 0x00020000U
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_BUFFER_STALL_SHIFT 16U
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_BUFFER_STALL_CLRMSK 0xFFFEFFFFU
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_BUFFER_STALL_EN 0x00010000U
+#define ROGUE_CR_EVENT_CLEAR_USC_TRIGGER_SHIFT 15U
+#define ROGUE_CR_EVENT_CLEAR_USC_TRIGGER_CLRMSK 0xFFFF7FFFU
+#define ROGUE_CR_EVENT_CLEAR_USC_TRIGGER_EN 0x00008000U
+#define ROGUE_CR_EVENT_CLEAR_ZLS_FINISHED_SHIFT 14U
+#define ROGUE_CR_EVENT_CLEAR_ZLS_FINISHED_CLRMSK 0xFFFFBFFFU
+#define ROGUE_CR_EVENT_CLEAR_ZLS_FINISHED_EN 0x00004000U
+#define ROGUE_CR_EVENT_CLEAR_GPIO_ACK_SHIFT 13U
+#define ROGUE_CR_EVENT_CLEAR_GPIO_ACK_CLRMSK 0xFFFFDFFFU
+#define ROGUE_CR_EVENT_CLEAR_GPIO_ACK_EN 0x00002000U
+#define ROGUE_CR_EVENT_CLEAR_GPIO_REQ_SHIFT 12U
+#define ROGUE_CR_EVENT_CLEAR_GPIO_REQ_CLRMSK 0xFFFFEFFFU
+#define ROGUE_CR_EVENT_CLEAR_GPIO_REQ_EN 0x00001000U
+#define ROGUE_CR_EVENT_CLEAR_POWER_ABORT_SHIFT 11U
+#define ROGUE_CR_EVENT_CLEAR_POWER_ABORT_CLRMSK 0xFFFFF7FFU
+#define ROGUE_CR_EVENT_CLEAR_POWER_ABORT_EN 0x00000800U
+#define ROGUE_CR_EVENT_CLEAR_POWER_COMPLETE_SHIFT 10U
+#define ROGUE_CR_EVENT_CLEAR_POWER_COMPLETE_CLRMSK 0xFFFFFBFFU
+#define ROGUE_CR_EVENT_CLEAR_POWER_COMPLETE_EN 0x00000400U
+#define ROGUE_CR_EVENT_CLEAR_MMU_PAGE_FAULT_SHIFT 9U
+#define ROGUE_CR_EVENT_CLEAR_MMU_PAGE_FAULT_CLRMSK 0xFFFFFDFFU
+#define ROGUE_CR_EVENT_CLEAR_MMU_PAGE_FAULT_EN 0x00000200U
+#define ROGUE_CR_EVENT_CLEAR_PM_3D_MEM_FREE_SHIFT 8U
+#define ROGUE_CR_EVENT_CLEAR_PM_3D_MEM_FREE_CLRMSK 0xFFFFFEFFU
+#define ROGUE_CR_EVENT_CLEAR_PM_3D_MEM_FREE_EN 0x00000100U
+#define ROGUE_CR_EVENT_CLEAR_PM_OUT_OF_MEMORY_SHIFT 7U
+#define ROGUE_CR_EVENT_CLEAR_PM_OUT_OF_MEMORY_CLRMSK 0xFFFFFF7FU
+#define ROGUE_CR_EVENT_CLEAR_PM_OUT_OF_MEMORY_EN 0x00000080U
+#define ROGUE_CR_EVENT_CLEAR_TA_TERMINATE_SHIFT 6U
+#define ROGUE_CR_EVENT_CLEAR_TA_TERMINATE_CLRMSK 0xFFFFFFBFU
+#define ROGUE_CR_EVENT_CLEAR_TA_TERMINATE_EN 0x00000040U
+#define ROGUE_CR_EVENT_CLEAR_TA_FINISHED_SHIFT 5U
+#define ROGUE_CR_EVENT_CLEAR_TA_FINISHED_CLRMSK 0xFFFFFFDFU
+#define ROGUE_CR_EVENT_CLEAR_TA_FINISHED_EN 0x00000020U
+#define ROGUE_CR_EVENT_CLEAR_ISP_END_MACROTILE_SHIFT 4U
+#define ROGUE_CR_EVENT_CLEAR_ISP_END_MACROTILE_CLRMSK 0xFFFFFFEFU
+#define ROGUE_CR_EVENT_CLEAR_ISP_END_MACROTILE_EN 0x00000010U
+#define ROGUE_CR_EVENT_CLEAR_PIXELBE_END_RENDER_SHIFT 3U
+#define ROGUE_CR_EVENT_CLEAR_PIXELBE_END_RENDER_CLRMSK 0xFFFFFFF7U
+#define ROGUE_CR_EVENT_CLEAR_PIXELBE_END_RENDER_EN 0x00000008U
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_FINISHED_SHIFT 2U
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_FINISHED_CLRMSK 0xFFFFFFFBU
+#define ROGUE_CR_EVENT_CLEAR_COMPUTE_FINISHED_EN 0x00000004U
+#define ROGUE_CR_EVENT_CLEAR_KERNEL_FINISHED_SHIFT 1U
+#define ROGUE_CR_EVENT_CLEAR_KERNEL_FINISHED_CLRMSK 0xFFFFFFFDU
+#define ROGUE_CR_EVENT_CLEAR_KERNEL_FINISHED_EN 0x00000002U
+#define ROGUE_CR_EVENT_CLEAR_TLA_COMPLETE_SHIFT 0U
+#define ROGUE_CR_EVENT_CLEAR_TLA_COMPLETE_CLRMSK 0xFFFFFFFEU
+#define ROGUE_CR_EVENT_CLEAR_TLA_COMPLETE_EN 0x00000001U
+
 /* Register ROGUE_CR_TIMER */
 #define ROGUE_CR_TIMER 0x0160U
 #define ROGUE_CR_TIMER_MASKFULL 0x8000FFFFFFFFFFFFULL
@@ -6031,25 +6145,6 @@
 #define ROGUE_CR_MULTICORE_COMPUTE_CTRL_COMMON_GPU_ENABLE_SHIFT 0U
 #define ROGUE_CR_MULTICORE_COMPUTE_CTRL_COMMON_GPU_ENABLE_CLRMSK 0xFFFFFF00U
 
-/* Register ROGUE_CR_ECC_RAM_ERR_INJ */
-#define ROGUE_CR_ECC_RAM_ERR_INJ 0xF340U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_MASKFULL 0x000000000000001FULL
-#define ROGUE_CR_ECC_RAM_ERR_INJ_SLC_SIDEKICK_SHIFT 4U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_SLC_SIDEKICK_CLRMSK 0xFFFFFFEFU
-#define ROGUE_CR_ECC_RAM_ERR_INJ_SLC_SIDEKICK_EN 0x00000010U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_USC_SHIFT 3U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_USC_CLRMSK 0xFFFFFFF7U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_USC_EN 0x00000008U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_TPU_MCU_L0_SHIFT 2U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_TPU_MCU_L0_CLRMSK 0xFFFFFFFBU
-#define ROGUE_CR_ECC_RAM_ERR_INJ_TPU_MCU_L0_EN 0x00000004U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_RASCAL_SHIFT 1U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_RASCAL_CLRMSK 0xFFFFFFFDU
-#define ROGUE_CR_ECC_RAM_ERR_INJ_RASCAL_EN 0x00000002U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_MARS_SHIFT 0U
-#define ROGUE_CR_ECC_RAM_ERR_INJ_MARS_CLRMSK 0xFFFFFFFEU
-#define ROGUE_CR_ECC_RAM_ERR_INJ_MARS_EN 0x00000001U
-
 /* Register ROGUE_CR_ECC_RAM_INIT_KICK */
 #define ROGUE_CR_ECC_RAM_INIT_KICK 0xF348U
 #define ROGUE_CR_ECC_RAM_INIT_KICK_MASKFULL 0x000000000000001FULL
@@ -6163,6 +6258,26 @@
 #define ROGUE_CR_SAFETY_EVENT_CLEAR__ROGUEXE__GPU_PAGE_FAULT_CLRMSK 0xFFFFFFFEU
 #define ROGUE_CR_SAFETY_EVENT_CLEAR__ROGUEXE__GPU_PAGE_FAULT_EN 0x00000001U
 
+/* Register ROGUE_CR_FAULT_FW_STATUS */
+#define ROGUE_CR_FAULT_FW_STATUS 0xF3B0U
+#define ROGUE_CR_FAULT_FW_STATUS_MASKFULL 0x0000000000010001ULL
+#define ROGUE_CR_FAULT_FW_STATUS_CPU_CORRECT_SHIFT 16U
+#define ROGUE_CR_FAULT_FW_STATUS_CPU_CORRECT_CLRMSK 0xFFFEFFFFU
+#define ROGUE_CR_FAULT_FW_STATUS_CPU_CORRECT_EN 0x00010000U
+#define ROGUE_CR_FAULT_FW_STATUS_CPU_DETECT_SHIFT 0U
+#define ROGUE_CR_FAULT_FW_STATUS_CPU_DETECT_CLRMSK 0xFFFFFFFEU
+#define ROGUE_CR_FAULT_FW_STATUS_CPU_DETECT_EN 0x00000001U
+
+/* Register ROGUE_CR_FAULT_FW_CLEAR */
+#define ROGUE_CR_FAULT_FW_CLEAR 0xF3B8U
+#define ROGUE_CR_FAULT_FW_CLEAR_MASKFULL 0x0000000000010001ULL
+#define ROGUE_CR_FAULT_FW_CLEAR_CPU_CORRECT_SHIFT 16U
+#define ROGUE_CR_FAULT_FW_CLEAR_CPU_CORRECT_CLRMSK 0xFFFEFFFFU
+#define ROGUE_CR_FAULT_FW_CLEAR_CPU_CORRECT_EN 0x00010000U
+#define ROGUE_CR_FAULT_FW_CLEAR_CPU_DETECT_SHIFT 0U
+#define ROGUE_CR_FAULT_FW_CLEAR_CPU_DETECT_CLRMSK 0xFFFFFFFEU
+#define ROGUE_CR_FAULT_FW_CLEAR_CPU_DETECT_EN 0x00000001U
+
 /* Register ROGUE_CR_MTS_SAFETY_EVENT_ENABLE */
 #define ROGUE_CR_MTS_SAFETY_EVENT_ENABLE__ROGUEXE 0xF3D8U
 #define ROGUE_CR_MTS_SAFETY_EVENT_ENABLE__ROGUEXE__MASKFULL 0x000000000000007FULL
diff --git a/drivers/gpu/drm/imagination/pvr_rogue_riscv.h b/drivers/gpu/drm/imagination/pvr_rogue_riscv.h
new file mode 100644
index 000000000000..9a070e24fa6a
--- /dev/null
+++ b/drivers/gpu/drm/imagination/pvr_rogue_riscv.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright (c) 2024 Imagination Technologies Ltd. */
+
+#ifndef PVR_ROGUE_RISCV_H
+#define PVR_ROGUE_RISCV_H
+
+#include "pvr_rogue_cr_defs.h"
+
+#include <linux/bitops.h>
+#include <linux/sizes.h>
+#include <linux/types.h>
+
+#define ROGUE_RISCVFW_REGION_SIZE SZ_256M
+#define ROGUE_RISCVFW_REGION_SHIFT __ffs(ROGUE_RISCVFW_REGION_SIZE)
+
+enum rogue_riscvfw_region {
+	ROGUE_RISCV_REGION__RESERVED_0 = 0,
+	ROGUE_RISCV_REGION__RESERVED_1,
+	ROGUE_RISCV_REGION_SOCIF,
+	ROGUE_RISCV_REGION__RESERVED_3,
+	ROGUE_RISCV_REGION__RESERVED_4,
+	ROGUE_RISCV_REGION_BOOTLDR_DATA,
+	ROGUE_RISCV_REGION_SHARED_CACHED_DATA,
+	ROGUE_RISCV_REGION__RESERVED_7,
+	ROGUE_RISCV_REGION_COREMEM,
+	ROGUE_RISCV_REGION__RESERVED_9,
+	ROGUE_RISCV_REGION__RESERVED_A,
+	ROGUE_RISCV_REGION__RESERVED_B,
+	ROGUE_RISCV_REGION_BOOTLDR_CODE,
+	ROGUE_RISCV_REGION_SHARED_UNCACHED_DATA,
+	ROGUE_RISCV_REGION__RESERVED_E,
+	ROGUE_RISCV_REGION__RESERVED_F,
+
+	ROGUE_RISCV_REGION__COUNT,
+};
+
+#define ROGUE_RISCVFW_REGION_BASE(r) ((u32)(ROGUE_RISCV_REGION_##r) << ROGUE_RISCVFW_REGION_SHIFT)
+#define ROGUE_RISCVFW_REGION_REMAP_CR(r) \
+	(ROGUE_CR_FWCORE_ADDR_REMAP_CONFIG0 + (u32)(ROGUE_RISCV_REGION_##r) * 8U)
+
+#endif /* PVR_ROGUE_RISCV_H */
diff --git a/drivers/gpu/drm/imagination/pvr_stream.c b/drivers/gpu/drm/imagination/pvr_stream.c
index 975336a4facf..679aa618b7a9 100644
--- a/drivers/gpu/drm/imagination/pvr_stream.c
+++ b/drivers/gpu/drm/imagination/pvr_stream.c
@@ -67,9 +67,8 @@ pvr_stream_process_1(struct pvr_device *pvr_dev, const struct pvr_stream_def *st
 		     u8 *dest, u32 dest_size, u32 *stream_offset_out)
 {
 	int err = 0;
-	u32 i;
 
-	for (i = 0; i < nr_entries; i++) {
+	for (u32 i = 0; i < nr_entries; i++) {
 		if (stream_def[i].offset >= dest_size) {
 			err = -EINVAL;
 			break;
@@ -131,7 +130,6 @@ pvr_stream_process_ext_stream(struct pvr_device *pvr_dev,
 	u32 musthave_masks[PVR_STREAM_EXTHDR_TYPE_MAX];
 	u32 ext_header;
 	int err = 0;
-	u32 i;
 
 	/* Copy "must have" mask from device. We clear this as we process the stream. */
 	memcpy(musthave_masks, pvr_dev->stream_musthave_quirks[cmd_defs->type],
@@ -159,7 +157,7 @@ pvr_stream_process_ext_stream(struct pvr_device *pvr_dev,
 
 		musthave_masks[type] &= ~data;
 
-		for (i = 0; i < header->ext_streams_num; i++) {
+		for (u32 i = 0; i < header->ext_streams_num; i++) {
 			const struct pvr_stream_ext_def *ext_def = &header->ext_streams[i];
 
 			if (!(ext_header & ext_def->header_mask))
@@ -181,7 +179,7 @@ pvr_stream_process_ext_stream(struct pvr_device *pvr_dev,
 	 * Verify that "must have" mask is now zero. If it isn't then one of the "must have" quirks
 	 * for this command was not present.
 	 */
-	for (i = 0; i < cmd_defs->ext_nr_headers; i++) {
+	for (u32 i = 0; i < cmd_defs->ext_nr_headers; i++) {
 		if (musthave_masks[i])
 			return -EINVAL;
 	}
@@ -245,13 +243,11 @@ pvr_stream_process(struct pvr_device *pvr_dev, const struct pvr_stream_cmd_defs
 		if (err)
 			return err;
 	} else {
-		u32 i;
-
 		/*
 		 * If we don't have an extension stream then there must not be any "must have"
 		 * quirks for this command.
 		 */
-		for (i = 0; i < cmd_defs->ext_nr_headers; i++) {
+		for (u32 i = 0; i < cmd_defs->ext_nr_headers; i++) {
 			if (pvr_dev->stream_musthave_quirks[cmd_defs->type][i])
 				return -EINVAL;
 		}
diff --git a/drivers/gpu/drm/imagination/pvr_vm_mips.c b/drivers/gpu/drm/imagination/pvr_vm_mips.c
index 94af854547d6..5847a1c92bea 100644
--- a/drivers/gpu/drm/imagination/pvr_vm_mips.c
+++ b/drivers/gpu/drm/imagination/pvr_vm_mips.c
@@ -100,10 +100,9 @@ pvr_vm_mips_fini(struct pvr_device *pvr_dev)
 {
 	struct pvr_fw_device *fw_dev = &pvr_dev->fw_dev;
 	struct pvr_fw_mips_data *mips_data = fw_dev->processor_data.mips_data;
-	int page_nr;
 
 	vunmap(mips_data->pt);
-	for (page_nr = PVR_MIPS_PT_PAGE_COUNT - 1; page_nr >= 0; page_nr--) {
+	for (int page_nr = PVR_MIPS_PT_PAGE_COUNT - 1; page_nr >= 0; page_nr--) {
 		dma_unmap_page(from_pvr_device(pvr_dev)->dev,
 			       mips_data->pt_dma_addr[page_nr], PAGE_SIZE, DMA_TO_DEVICE);
 
diff --git a/drivers/gpu/drm/imx/ipuv3/parallel-display.c b/drivers/gpu/drm/imx/ipuv3/parallel-display.c
index 9e66eb77b1eb..6d8325c76697 100644
--- a/drivers/gpu/drm/imx/ipuv3/parallel-display.c
+++ b/drivers/gpu/drm/imx/ipuv3/parallel-display.c
@@ -162,11 +162,12 @@ static int imx_pd_bridge_atomic_check(struct drm_bridge *bridge,
 }
 
 static int imx_pd_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct imx_parallel_display *imxpd = bridge_to_imxpd(bridge);
 
-	return drm_bridge_attach(bridge->encoder, imxpd->next_bridge, bridge, flags);
+	return drm_bridge_attach(encoder, imxpd->next_bridge, bridge, flags);
 }
 
 static const struct drm_bridge_funcs imx_pd_bridge_funcs = {
diff --git a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
index 20b93fff0239..f851e9ffdb28 100644
--- a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
+++ b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
@@ -791,11 +791,12 @@ static void ingenic_drm_encoder_atomic_mode_set(struct drm_encoder *encoder,
 }
 
 static int ingenic_drm_bridge_attach(struct drm_bridge *bridge,
+				     struct drm_encoder *encoder,
 				     enum drm_bridge_attach_flags flags)
 {
-	struct ingenic_drm_bridge *ib = to_ingenic_drm_bridge(bridge->encoder);
+	struct ingenic_drm_bridge *ib = to_ingenic_drm_bridge(encoder);
 
-	return drm_bridge_attach(bridge->encoder, ib->next_bridge,
+	return drm_bridge_attach(encoder, ib->next_bridge,
 				 &ib->bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/lima/lima_gem.c b/drivers/gpu/drm/lima/lima_gem.c
index 9bb997dbb4b9..5deec673c11e 100644
--- a/drivers/gpu/drm/lima/lima_gem.c
+++ b/drivers/gpu/drm/lima/lima_gem.c
@@ -47,7 +47,7 @@ int lima_heap_alloc(struct lima_bo *bo, struct lima_vm *vm)
 		}
 
 		bo->base.pages = pages;
-		bo->base.pages_use_count = 1;
+		refcount_set(&bo->base.pages_use_count, 1);
 
 		mapping_set_unevictable(mapping);
 	}
@@ -195,7 +195,7 @@ static int lima_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map)
 	if (bo->heap_size)
 		return -EINVAL;
 
-	return drm_gem_shmem_vmap(&bo->base, map);
+	return drm_gem_shmem_vmap_locked(&bo->base, map);
 }
 
 static int lima_gem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
diff --git a/drivers/gpu/drm/lima/lima_sched.c b/drivers/gpu/drm/lima/lima_sched.c
index 825135a26aa4..7934098e651b 100644
--- a/drivers/gpu/drm/lima/lima_sched.c
+++ b/drivers/gpu/drm/lima/lima_sched.c
@@ -371,7 +371,7 @@ static void lima_sched_build_error_task_list(struct lima_sched_task *task)
 		} else {
 			buffer_chunk->size = lima_bo_size(bo);
 
-			ret = drm_gem_vmap_unlocked(&bo->base.base, &map);
+			ret = drm_gem_vmap(&bo->base.base, &map);
 			if (ret) {
 				kvfree(et);
 				goto out;
@@ -379,7 +379,7 @@ static void lima_sched_build_error_task_list(struct lima_sched_task *task)
 
 			memcpy(buffer_chunk + 1, map.vaddr, buffer_chunk->size);
 
-			drm_gem_vunmap_unlocked(&bo->base.base, &map);
+			drm_gem_vunmap(&bo->base.base, &map);
 		}
 
 		buffer_chunk = (void *)(buffer_chunk + 1) + buffer_chunk->size;
diff --git a/drivers/gpu/drm/mcde/mcde_dsi.c b/drivers/gpu/drm/mcde/mcde_dsi.c
index 395449a72f0a..a3423459dd7a 100644
--- a/drivers/gpu/drm/mcde/mcde_dsi.c
+++ b/drivers/gpu/drm/mcde/mcde_dsi.c
@@ -1048,6 +1048,7 @@ void mcde_dsi_disable(struct drm_bridge *bridge)
 }
 
 static int mcde_dsi_bridge_attach(struct drm_bridge *bridge,
+				  struct drm_encoder *encoder,
 				  enum drm_bridge_attach_flags flags)
 {
 	struct mcde_dsi *d = bridge_to_mcde_dsi(bridge);
@@ -1059,7 +1060,7 @@ static int mcde_dsi_bridge_attach(struct drm_bridge *bridge,
 	}
 
 	/* Attach the DSI bridge to the output (panel etc) bridge */
-	return drm_bridge_attach(bridge->encoder, d->bridge_out, bridge, flags);
+	return drm_bridge_attach(encoder, d->bridge_out, bridge, flags);
 }
 
 static const struct drm_bridge_funcs mcde_dsi_bridge_funcs = {
@@ -1137,7 +1138,6 @@ static int mcde_dsi_bind(struct device *dev, struct device *master,
 	d->bridge_out = bridge;
 
 	/* Create a bridge for this DSI channel */
-	d->bridge.funcs = &mcde_dsi_bridge_funcs;
 	d->bridge.of_node = dev->of_node;
 	drm_bridge_add(&d->bridge);
 
@@ -1173,9 +1173,9 @@ static int mcde_dsi_probe(struct platform_device *pdev)
 	u32 dsi_id;
 	int ret;
 
-	d = devm_kzalloc(dev, sizeof(*d), GFP_KERNEL);
-	if (!d)
-		return -ENOMEM;
+	d = devm_drm_bridge_alloc(dev, struct mcde_dsi, bridge, &mcde_dsi_bridge_funcs);
+	if (IS_ERR(d))
+		return PTR_ERR(d);
 	d->dev = dev;
 	platform_set_drvdata(pdev, d);
 
diff --git a/drivers/gpu/drm/mediatek/Makefile b/drivers/gpu/drm/mediatek/Makefile
index 32a2ed6c0cfe..43afd0a26d14 100644
--- a/drivers/gpu/drm/mediatek/Makefile
+++ b/drivers/gpu/drm/mediatek/Makefile
@@ -21,10 +21,8 @@ mediatek-drm-y := mtk_crtc.o \
 
 obj-$(CONFIG_DRM_MEDIATEK) += mediatek-drm.o
 
-mediatek-drm-hdmi-objs := mtk_cec.o \
-			  mtk_hdmi.o \
-			  mtk_hdmi_ddc.o
-
-obj-$(CONFIG_DRM_MEDIATEK_HDMI) += mediatek-drm-hdmi.o
+obj-$(CONFIG_DRM_MEDIATEK_HDMI) += mtk_cec.o
+obj-$(CONFIG_DRM_MEDIATEK_HDMI) += mtk_hdmi.o
+obj-$(CONFIG_DRM_MEDIATEK_HDMI) += mtk_hdmi_ddc.o
 
 obj-$(CONFIG_DRM_MEDIATEK_DP) += mtk_dp.o
diff --git a/drivers/gpu/drm/mediatek/mtk_cec.c b/drivers/gpu/drm/mediatek/mtk_cec.c
index b42c0d87eba3..c7be530ca041 100644
--- a/drivers/gpu/drm/mediatek/mtk_cec.c
+++ b/drivers/gpu/drm/mediatek/mtk_cec.c
@@ -12,7 +12,6 @@
 #include <linux/platform_device.h>
 
 #include "mtk_cec.h"
-#include "mtk_hdmi.h"
 #include "mtk_drm_drv.h"
 
 #define TR_CONFIG		0x00
@@ -102,6 +101,7 @@ void mtk_cec_set_hpd_event(struct device *dev,
 	cec->hpd_event = hpd_event;
 	spin_unlock_irqrestore(&cec->lock, flags);
 }
+EXPORT_SYMBOL_NS_GPL(mtk_cec_set_hpd_event, "DRM_MTK_HDMI_V1");
 
 bool mtk_cec_hpd_high(struct device *dev)
 {
@@ -112,6 +112,7 @@ bool mtk_cec_hpd_high(struct device *dev)
 
 	return (status & (HDMI_PORD | HDMI_HTPLG)) == (HDMI_PORD | HDMI_HTPLG);
 }
+EXPORT_SYMBOL_NS_GPL(mtk_cec_hpd_high, "DRM_MTK_HDMI_V1");
 
 static void mtk_cec_htplg_irq_init(struct mtk_cec *cec)
 {
@@ -247,3 +248,7 @@ struct platform_driver mtk_cec_driver = {
 		.of_match_table = mtk_cec_of_ids,
 	},
 };
+module_platform_driver(mtk_cec_driver);
+
+MODULE_DESCRIPTION("MediaTek HDMI CEC Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c b/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
index fa0e95dd29a0..fe97bb97e004 100644
--- a/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
+++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c
@@ -492,11 +492,6 @@ static const struct of_device_id mtk_ovl_adaptor_comp_dt_ids[] = {
 	{ /* sentinel */ }
 };
 
-static int compare_of(struct device *dev, void *data)
-{
-	return dev->of_node == data;
-}
-
 static int ovl_adaptor_of_get_ddp_comp_type(struct device_node *node,
 					    enum mtk_ovl_adaptor_comp_type *ctype)
 {
@@ -567,7 +562,7 @@ static int ovl_adaptor_comp_init(struct device *dev, struct component_match **ma
 
 		priv->ovl_adaptor_comp[id] = &comp_pdev->dev;
 
-		drm_of_component_match_add(dev, match, compare_of, node);
+		drm_of_component_match_add(dev, match, component_compare_of, node);
 		dev_dbg(dev, "Adding component match for %pOF\n", node);
 	}
 
diff --git a/drivers/gpu/drm/mediatek/mtk_dp.c b/drivers/gpu/drm/mediatek/mtk_dp.c
index fed3307d3374..b2408abb9d49 100644
--- a/drivers/gpu/drm/mediatek/mtk_dp.c
+++ b/drivers/gpu/drm/mediatek/mtk_dp.c
@@ -2287,6 +2287,7 @@ static void mtk_dp_poweroff(struct mtk_dp *mtk_dp)
 }
 
 static int mtk_dp_bridge_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct mtk_dp *mtk_dp = mtk_dp_from_bridge(bridge);
@@ -2310,7 +2311,7 @@ static int mtk_dp_bridge_attach(struct drm_bridge *bridge,
 		goto err_aux_register;
 
 	if (mtk_dp->next_bridge) {
-		ret = drm_bridge_attach(bridge->encoder, mtk_dp->next_bridge,
+		ret = drm_bridge_attach(encoder, mtk_dp->next_bridge,
 					&mtk_dp->bridge, flags);
 		if (ret) {
 			drm_warn(mtk_dp->drm_dev,
diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c
index 0fd13e6dd3f1..6fb85bc6487a 100644
--- a/drivers/gpu/drm/mediatek/mtk_dpi.c
+++ b/drivers/gpu/drm/mediatek/mtk_dpi.c
@@ -59,7 +59,8 @@ enum mtk_dpi_out_channel_swap {
 
 enum mtk_dpi_out_color_format {
 	MTK_DPI_COLOR_FORMAT_RGB,
-	MTK_DPI_COLOR_FORMAT_YCBCR_422
+	MTK_DPI_COLOR_FORMAT_YCBCR_422,
+	MTK_DPI_COLOR_FORMAT_YCBCR_444
 };
 
 struct mtk_dpi {
@@ -450,9 +451,16 @@ static void mtk_dpi_config_disable_edge(struct mtk_dpi *dpi)
 static void mtk_dpi_config_color_format(struct mtk_dpi *dpi,
 					enum mtk_dpi_out_color_format format)
 {
-	mtk_dpi_config_channel_swap(dpi, MTK_DPI_OUT_CHANNEL_SWAP_RGB);
+	mtk_dpi_config_channel_swap(dpi, dpi->channel_swap);
 
-	if (format == MTK_DPI_COLOR_FORMAT_YCBCR_422) {
+	switch (format) {
+	case MTK_DPI_COLOR_FORMAT_YCBCR_444:
+		mtk_dpi_config_yuv422_enable(dpi, false);
+		mtk_dpi_config_csc_enable(dpi, true);
+		if (dpi->conf->swap_input_support)
+			mtk_dpi_config_swap_input(dpi, false);
+		break;
+	case MTK_DPI_COLOR_FORMAT_YCBCR_422:
 		mtk_dpi_config_yuv422_enable(dpi, true);
 		mtk_dpi_config_csc_enable(dpi, true);
 
@@ -463,11 +471,14 @@ static void mtk_dpi_config_color_format(struct mtk_dpi *dpi,
 		mtk_dpi_mask(dpi, DPI_MATRIX_SET, dpi->mode.hdisplay <= 720 ?
 			     MATRIX_SEL_RGB_TO_BT601 : MATRIX_SEL_RGB_TO_JPEG,
 			     INT_MATRIX_SEL_MASK);
-	} else {
+		break;
+	default:
+	case MTK_DPI_COLOR_FORMAT_RGB:
 		mtk_dpi_config_yuv422_enable(dpi, false);
 		mtk_dpi_config_csc_enable(dpi, false);
 		if (dpi->conf->swap_input_support)
 			mtk_dpi_config_swap_input(dpi, false);
+		break;
 	}
 }
 
@@ -734,6 +745,65 @@ static u32 *mtk_dpi_bridge_atomic_get_input_bus_fmts(struct drm_bridge *bridge,
 	return input_fmts;
 }
 
+static unsigned int mtk_dpi_bus_fmt_bit_num(unsigned int out_bus_format)
+{
+	switch (out_bus_format) {
+	default:
+	case MEDIA_BUS_FMT_RGB888_1X24:
+	case MEDIA_BUS_FMT_BGR888_1X24:
+	case MEDIA_BUS_FMT_RGB888_2X12_LE:
+	case MEDIA_BUS_FMT_RGB888_2X12_BE:
+	case MEDIA_BUS_FMT_YUYV8_1X16:
+	case MEDIA_BUS_FMT_YUV8_1X24:
+		return MTK_DPI_OUT_BIT_NUM_8BITS;
+	case MEDIA_BUS_FMT_RGB101010_1X30:
+	case MEDIA_BUS_FMT_YUYV10_1X20:
+	case MEDIA_BUS_FMT_YUV10_1X30:
+		return MTK_DPI_OUT_BIT_NUM_10BITS;
+	case MEDIA_BUS_FMT_YUYV12_1X24:
+		return MTK_DPI_OUT_BIT_NUM_12BITS;
+	}
+}
+
+static unsigned int mtk_dpi_bus_fmt_channel_swap(unsigned int out_bus_format)
+{
+	switch (out_bus_format) {
+	default:
+	case MEDIA_BUS_FMT_RGB888_1X24:
+	case MEDIA_BUS_FMT_RGB888_2X12_LE:
+	case MEDIA_BUS_FMT_RGB888_2X12_BE:
+	case MEDIA_BUS_FMT_RGB101010_1X30:
+	case MEDIA_BUS_FMT_YUYV8_1X16:
+	case MEDIA_BUS_FMT_YUYV10_1X20:
+	case MEDIA_BUS_FMT_YUYV12_1X24:
+		return MTK_DPI_OUT_CHANNEL_SWAP_RGB;
+	case MEDIA_BUS_FMT_BGR888_1X24:
+	case MEDIA_BUS_FMT_YUV8_1X24:
+	case MEDIA_BUS_FMT_YUV10_1X30:
+		return MTK_DPI_OUT_CHANNEL_SWAP_BGR;
+	}
+}
+
+static unsigned int mtk_dpi_bus_fmt_color_format(unsigned int out_bus_format)
+{
+	switch (out_bus_format) {
+	default:
+	case MEDIA_BUS_FMT_RGB888_1X24:
+	case MEDIA_BUS_FMT_BGR888_1X24:
+	case MEDIA_BUS_FMT_RGB888_2X12_LE:
+	case MEDIA_BUS_FMT_RGB888_2X12_BE:
+	case MEDIA_BUS_FMT_RGB101010_1X30:
+		return MTK_DPI_COLOR_FORMAT_RGB;
+	case MEDIA_BUS_FMT_YUYV8_1X16:
+	case MEDIA_BUS_FMT_YUYV10_1X20:
+	case MEDIA_BUS_FMT_YUYV12_1X24:
+		return MTK_DPI_COLOR_FORMAT_YCBCR_422;
+	case MEDIA_BUS_FMT_YUV8_1X24:
+	case MEDIA_BUS_FMT_YUV10_1X30:
+		return MTK_DPI_COLOR_FORMAT_YCBCR_444;
+	}
+}
+
 static int mtk_dpi_bridge_atomic_check(struct drm_bridge *bridge,
 				       struct drm_bridge_state *bridge_state,
 				       struct drm_crtc_state *crtc_state,
@@ -753,18 +823,16 @@ static int mtk_dpi_bridge_atomic_check(struct drm_bridge *bridge,
 		bridge_state->output_bus_cfg.format);
 
 	dpi->output_fmt = out_bus_format;
-	dpi->bit_num = MTK_DPI_OUT_BIT_NUM_8BITS;
-	dpi->channel_swap = MTK_DPI_OUT_CHANNEL_SWAP_RGB;
+	dpi->bit_num = mtk_dpi_bus_fmt_bit_num(out_bus_format);
+	dpi->channel_swap = mtk_dpi_bus_fmt_channel_swap(out_bus_format);
 	dpi->yc_map = MTK_DPI_OUT_YC_MAP_RGB;
-	if (out_bus_format == MEDIA_BUS_FMT_YUYV8_1X16)
-		dpi->color_format = MTK_DPI_COLOR_FORMAT_YCBCR_422;
-	else
-		dpi->color_format = MTK_DPI_COLOR_FORMAT_RGB;
+	dpi->color_format = mtk_dpi_bus_fmt_color_format(out_bus_format);
 
 	return 0;
 }
 
 static int mtk_dpi_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct mtk_dpi *dpi = bridge_to_dpi(bridge);
@@ -783,7 +851,7 @@ static int mtk_dpi_bridge_attach(struct drm_bridge *bridge,
 					     "Failed to get bridge\n");
 	}
 
-	return drm_bridge_attach(bridge->encoder, dpi->next_bridge,
+	return drm_bridge_attach(encoder, dpi->next_bridge,
 				 &dpi->bridge, flags);
 }
 
@@ -1026,9 +1094,29 @@ static const u32 mt8183_output_fmts[] = {
 	MEDIA_BUS_FMT_RGB888_2X12_BE,
 };
 
-static const u32 mt8195_output_fmts[] = {
+static const u32 mt8195_dpi_output_fmts[] = {
+	MEDIA_BUS_FMT_BGR888_1X24,
+	MEDIA_BUS_FMT_RGB888_1X24,
+	MEDIA_BUS_FMT_RGB888_2X12_LE,
+	MEDIA_BUS_FMT_RGB888_2X12_BE,
+	MEDIA_BUS_FMT_RGB101010_1X30,
+	MEDIA_BUS_FMT_YUYV8_1X16,
+	MEDIA_BUS_FMT_YUYV10_1X20,
+	MEDIA_BUS_FMT_YUYV12_1X24,
+	MEDIA_BUS_FMT_YUV8_1X24,
+	MEDIA_BUS_FMT_YUV10_1X30,
+};
+
+static const u32 mt8195_dp_intf_output_fmts[] = {
+	MEDIA_BUS_FMT_BGR888_1X24,
 	MEDIA_BUS_FMT_RGB888_1X24,
+	MEDIA_BUS_FMT_RGB888_2X12_LE,
+	MEDIA_BUS_FMT_RGB888_2X12_BE,
+	MEDIA_BUS_FMT_RGB101010_1X30,
 	MEDIA_BUS_FMT_YUYV8_1X16,
+	MEDIA_BUS_FMT_YUYV10_1X20,
+	MEDIA_BUS_FMT_YUV8_1X24,
+	MEDIA_BUS_FMT_YUV10_1X30,
 };
 
 static const struct mtk_dpi_factor dpi_factor_mt2701[] = {
@@ -1141,8 +1229,8 @@ static const struct mtk_dpi_conf mt8192_conf = {
 
 static const struct mtk_dpi_conf mt8195_conf = {
 	.max_clock_khz = 594000,
-	.output_fmts = mt8183_output_fmts,
-	.num_output_fmts = ARRAY_SIZE(mt8183_output_fmts),
+	.output_fmts = mt8195_dpi_output_fmts,
+	.num_output_fmts = ARRAY_SIZE(mt8195_dpi_output_fmts),
 	.pixels_per_iter = 1,
 	.is_ck_de_pol = true,
 	.swap_input_support = true,
@@ -1161,8 +1249,8 @@ static const struct mtk_dpi_conf mt8195_dpintf_conf = {
 	.dpi_factor = dpi_factor_mt8195_dp_intf,
 	.num_dpi_factor = ARRAY_SIZE(dpi_factor_mt8195_dp_intf),
 	.max_clock_khz = 600000,
-	.output_fmts = mt8195_output_fmts,
-	.num_output_fmts = ARRAY_SIZE(mt8195_output_fmts),
+	.output_fmts = mt8195_dp_intf_output_fmts,
+	.num_output_fmts = ARRAY_SIZE(mt8195_dp_intf_output_fmts),
 	.pixels_per_iter = 4,
 	.dimension_mask = DPINTF_HPW_MASK,
 	.hvsize_mask = DPINTF_HSIZE_MASK,
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c
index 74158b9d6503..7c0c12dde488 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c
@@ -470,7 +470,7 @@ static int mtk_drm_kms_init(struct drm_device *drm)
 
 	ret = drmm_mode_config_init(drm);
 	if (ret)
-		goto put_mutex_dev;
+		return ret;
 
 	drm->mode_config.min_width = 64;
 	drm->mode_config.min_height = 64;
@@ -488,8 +488,11 @@ static int mtk_drm_kms_init(struct drm_device *drm)
 	for (i = 0; i < private->data->mmsys_dev_num; i++) {
 		drm->dev_private = private->all_drm_private[i];
 		ret = component_bind_all(private->all_drm_private[i]->dev, drm);
-		if (ret)
-			goto put_mutex_dev;
+		if (ret) {
+			while (--i >= 0)
+				component_unbind_all(private->all_drm_private[i]->dev, drm);
+			return ret;
+		}
 	}
 
 	/*
@@ -582,9 +585,6 @@ static int mtk_drm_kms_init(struct drm_device *drm)
 err_component_unbind:
 	for (i = 0; i < private->data->mmsys_dev_num; i++)
 		component_unbind_all(private->all_drm_private[i]->dev, drm);
-put_mutex_dev:
-	for (i = 0; i < private->data->mmsys_dev_num; i++)
-		put_device(private->all_drm_private[i]->mutex_dev);
 
 	return ret;
 }
@@ -655,8 +655,10 @@ static int mtk_drm_bind(struct device *dev)
 		return 0;
 
 	drm = drm_dev_alloc(&mtk_drm_driver, dev);
-	if (IS_ERR(drm))
-		return PTR_ERR(drm);
+	if (IS_ERR(drm)) {
+		ret = PTR_ERR(drm);
+		goto err_put_dev;
+	}
 
 	private->drm_master = true;
 	drm->dev_private = private;
@@ -682,18 +684,31 @@ err_free:
 	drm_dev_put(drm);
 	for (i = 0; i < private->data->mmsys_dev_num; i++)
 		private->all_drm_private[i]->drm = NULL;
+err_put_dev:
+	for (i = 0; i < private->data->mmsys_dev_num; i++) {
+		/* For device_find_child in mtk_drm_get_all_priv() */
+		put_device(private->all_drm_private[i]->dev);
+	}
+	put_device(private->mutex_dev);
 	return ret;
 }
 
 static void mtk_drm_unbind(struct device *dev)
 {
 	struct mtk_drm_private *private = dev_get_drvdata(dev);
+	int i;
 
 	/* for multi mmsys dev, unregister drm dev in mmsys master */
 	if (private->drm_master) {
 		drm_dev_unregister(private->drm);
 		mtk_drm_kms_deinit(private->drm);
 		drm_dev_put(private->drm);
+
+		for (i = 0; i < private->data->mmsys_dev_num; i++) {
+			/* For device_find_child in mtk_drm_get_all_priv() */
+			put_device(private->all_drm_private[i]->dev);
+		}
+		put_device(private->mutex_dev);
 	}
 	private->mtk_drm_bound = false;
 	private->drm_master = false;
diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c
index d1f407fb7eb1..4fe1f38a3c4b 100644
--- a/drivers/gpu/drm/mediatek/mtk_dsi.c
+++ b/drivers/gpu/drm/mediatek/mtk_dsi.c
@@ -807,12 +807,13 @@ static void mtk_output_dsi_disable(struct mtk_dsi *dsi)
 }
 
 static int mtk_dsi_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct mtk_dsi *dsi = bridge_to_dsi(bridge);
 
 	/* Attach the panel or bridge to the dsi bridge */
-	return drm_bridge_attach(bridge->encoder, dsi->next_bridge,
+	return drm_bridge_attach(encoder, dsi->next_bridge,
 				 &dsi->bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.c b/drivers/gpu/drm/mediatek/mtk_hdmi.c
index 06e4fac152b7..8803cd4a8bc9 100644
--- a/drivers/gpu/drm/mediatek/mtk_hdmi.c
+++ b/drivers/gpu/drm/mediatek/mtk_hdmi.c
@@ -31,7 +31,6 @@
 #include <drm/drm_probe_helper.h>
 
 #include "mtk_cec.h"
-#include "mtk_hdmi.h"
 #include "mtk_hdmi_regs.h"
 
 #define NCTS_BYTES	7
@@ -165,7 +164,7 @@ struct mtk_hdmi {
 	bool dvi_mode;
 	struct regmap *sys_regmap;
 	unsigned int sys_offset;
-	void __iomem *regs;
+	struct regmap *regs;
 	struct platform_device *audio_pdev;
 	struct hdmi_audio_param aud_param;
 	bool audio_enable;
@@ -181,50 +180,10 @@ static inline struct mtk_hdmi *hdmi_ctx_from_bridge(struct drm_bridge *b)
 	return container_of(b, struct mtk_hdmi, bridge);
 }
 
-static u32 mtk_hdmi_read(struct mtk_hdmi *hdmi, u32 offset)
-{
-	return readl(hdmi->regs + offset);
-}
-
-static void mtk_hdmi_write(struct mtk_hdmi *hdmi, u32 offset, u32 val)
-{
-	writel(val, hdmi->regs + offset);
-}
-
-static void mtk_hdmi_clear_bits(struct mtk_hdmi *hdmi, u32 offset, u32 bits)
-{
-	void __iomem *reg = hdmi->regs + offset;
-	u32 tmp;
-
-	tmp = readl(reg);
-	tmp &= ~bits;
-	writel(tmp, reg);
-}
-
-static void mtk_hdmi_set_bits(struct mtk_hdmi *hdmi, u32 offset, u32 bits)
-{
-	void __iomem *reg = hdmi->regs + offset;
-	u32 tmp;
-
-	tmp = readl(reg);
-	tmp |= bits;
-	writel(tmp, reg);
-}
-
-static void mtk_hdmi_mask(struct mtk_hdmi *hdmi, u32 offset, u32 val, u32 mask)
-{
-	void __iomem *reg = hdmi->regs + offset;
-	u32 tmp;
-
-	tmp = readl(reg);
-	tmp = (tmp & ~mask) | (val & mask);
-	writel(tmp, reg);
-}
-
 static void mtk_hdmi_hw_vid_black(struct mtk_hdmi *hdmi, bool black)
 {
-	mtk_hdmi_mask(hdmi, VIDEO_CFG_4, black ? GEN_RGB : NORMAL_PATH,
-		      VIDEO_SOURCE_SEL);
+	regmap_update_bits(hdmi->regs, VIDEO_SOURCE_SEL,
+			   VIDEO_CFG_4, black ? GEN_RGB : NORMAL_PATH);
 }
 
 static void mtk_hdmi_hw_make_reg_writable(struct mtk_hdmi *hdmi, bool enable)
@@ -259,12 +218,12 @@ static void mtk_hdmi_hw_1p4_version_enable(struct mtk_hdmi *hdmi, bool enable)
 
 static void mtk_hdmi_hw_aud_mute(struct mtk_hdmi *hdmi)
 {
-	mtk_hdmi_set_bits(hdmi, GRL_AUDIO_CFG, AUDIO_ZERO);
+	regmap_set_bits(hdmi->regs, GRL_AUDIO_CFG, AUDIO_ZERO);
 }
 
 static void mtk_hdmi_hw_aud_unmute(struct mtk_hdmi *hdmi)
 {
-	mtk_hdmi_clear_bits(hdmi, GRL_AUDIO_CFG, AUDIO_ZERO);
+	regmap_clear_bits(hdmi->regs, GRL_AUDIO_CFG, AUDIO_ZERO);
 }
 
 static void mtk_hdmi_hw_reset(struct mtk_hdmi *hdmi)
@@ -273,25 +232,25 @@ static void mtk_hdmi_hw_reset(struct mtk_hdmi *hdmi)
 			   HDMI_RST, HDMI_RST);
 	regmap_update_bits(hdmi->sys_regmap, hdmi->sys_offset + HDMI_SYS_CFG1C,
 			   HDMI_RST, 0);
-	mtk_hdmi_clear_bits(hdmi, GRL_CFG3, CFG3_CONTROL_PACKET_DELAY);
+	regmap_clear_bits(hdmi->regs, GRL_CFG3, CFG3_CONTROL_PACKET_DELAY);
 	regmap_update_bits(hdmi->sys_regmap, hdmi->sys_offset + HDMI_SYS_CFG1C,
 			   ANLG_ON, ANLG_ON);
 }
 
 static void mtk_hdmi_hw_enable_notice(struct mtk_hdmi *hdmi, bool enable_notice)
 {
-	mtk_hdmi_mask(hdmi, GRL_CFG2, enable_notice ? CFG2_NOTICE_EN : 0,
-		      CFG2_NOTICE_EN);
+	regmap_update_bits(hdmi->regs, GRL_CFG2, CFG2_NOTICE_EN,
+			   enable_notice ? CFG2_NOTICE_EN : 0);
 }
 
 static void mtk_hdmi_hw_write_int_mask(struct mtk_hdmi *hdmi, u32 int_mask)
 {
-	mtk_hdmi_write(hdmi, GRL_INT_MASK, int_mask);
+	regmap_write(hdmi->regs, GRL_INT_MASK, int_mask);
 }
 
 static void mtk_hdmi_hw_enable_dvi_mode(struct mtk_hdmi *hdmi, bool enable)
 {
-	mtk_hdmi_mask(hdmi, GRL_CFG1, enable ? CFG1_DVI : 0, CFG1_DVI);
+	regmap_update_bits(hdmi->regs, GRL_CFG1, CFG1_DVI, enable ? CFG1_DVI : 0);
 }
 
 static void mtk_hdmi_hw_send_info_frame(struct mtk_hdmi *hdmi, u8 *buffer,
@@ -337,22 +296,22 @@ static void mtk_hdmi_hw_send_info_frame(struct mtk_hdmi *hdmi, u8 *buffer,
 		dev_err(hdmi->dev, "Unknown infoframe type %d\n", frame_type);
 		return;
 	}
-	mtk_hdmi_clear_bits(hdmi, ctrl_reg, ctrl_frame_en);
-	mtk_hdmi_write(hdmi, GRL_INFOFRM_TYPE, frame_type);
-	mtk_hdmi_write(hdmi, GRL_INFOFRM_VER, frame_ver);
-	mtk_hdmi_write(hdmi, GRL_INFOFRM_LNG, frame_len);
+	regmap_clear_bits(hdmi->regs, ctrl_reg, ctrl_frame_en);
+	regmap_write(hdmi->regs, GRL_INFOFRM_TYPE, frame_type);
+	regmap_write(hdmi->regs, GRL_INFOFRM_VER, frame_ver);
+	regmap_write(hdmi->regs, GRL_INFOFRM_LNG, frame_len);
 
-	mtk_hdmi_write(hdmi, GRL_IFM_PORT, checksum);
+	regmap_write(hdmi->regs, GRL_IFM_PORT, checksum);
 	for (i = 0; i < frame_len; i++)
-		mtk_hdmi_write(hdmi, GRL_IFM_PORT, frame_data[i]);
+		regmap_write(hdmi->regs, GRL_IFM_PORT, frame_data[i]);
 
-	mtk_hdmi_set_bits(hdmi, ctrl_reg, ctrl_frame_en);
+	regmap_set_bits(hdmi->regs, ctrl_reg, ctrl_frame_en);
 }
 
 static void mtk_hdmi_hw_send_aud_packet(struct mtk_hdmi *hdmi, bool enable)
 {
-	mtk_hdmi_mask(hdmi, GRL_SHIFT_R2, enable ? 0 : AUDIO_PACKET_OFF,
-		      AUDIO_PACKET_OFF);
+	regmap_update_bits(hdmi->regs, AUDIO_PACKET_OFF,
+			   GRL_SHIFT_R2, enable ? 0 : AUDIO_PACKET_OFF);
 }
 
 static void mtk_hdmi_hw_config_sys(struct mtk_hdmi *hdmi)
@@ -373,44 +332,44 @@ static void mtk_hdmi_hw_set_deep_color_mode(struct mtk_hdmi *hdmi)
 
 static void mtk_hdmi_hw_send_av_mute(struct mtk_hdmi *hdmi)
 {
-	mtk_hdmi_clear_bits(hdmi, GRL_CFG4, CTRL_AVMUTE);
+	regmap_clear_bits(hdmi->regs, GRL_CFG4, CTRL_AVMUTE);
 	usleep_range(2000, 4000);
-	mtk_hdmi_set_bits(hdmi, GRL_CFG4, CTRL_AVMUTE);
+	regmap_set_bits(hdmi->regs, GRL_CFG4, CTRL_AVMUTE);
 }
 
 static void mtk_hdmi_hw_send_av_unmute(struct mtk_hdmi *hdmi)
 {
-	mtk_hdmi_mask(hdmi, GRL_CFG4, CFG4_AV_UNMUTE_EN,
-		      CFG4_AV_UNMUTE_EN | CFG4_AV_UNMUTE_SET);
+	regmap_update_bits(hdmi->regs, GRL_CFG4, CFG4_AV_UNMUTE_EN | CFG4_AV_UNMUTE_SET,
+			   CFG4_AV_UNMUTE_EN);
 	usleep_range(2000, 4000);
-	mtk_hdmi_mask(hdmi, GRL_CFG4, CFG4_AV_UNMUTE_SET,
-		      CFG4_AV_UNMUTE_EN | CFG4_AV_UNMUTE_SET);
+	regmap_update_bits(hdmi->regs, GRL_CFG4, CFG4_AV_UNMUTE_EN | CFG4_AV_UNMUTE_SET,
+			   CFG4_AV_UNMUTE_SET);
 }
 
 static void mtk_hdmi_hw_ncts_enable(struct mtk_hdmi *hdmi, bool on)
 {
-	mtk_hdmi_mask(hdmi, GRL_CTS_CTRL, on ? 0 : CTS_CTRL_SOFT,
-		      CTS_CTRL_SOFT);
+	regmap_update_bits(hdmi->regs, GRL_CTS_CTRL, CTS_CTRL_SOFT,
+			   on ? 0 : CTS_CTRL_SOFT);
 }
 
 static void mtk_hdmi_hw_ncts_auto_write_enable(struct mtk_hdmi *hdmi,
 					       bool enable)
 {
-	mtk_hdmi_mask(hdmi, GRL_CTS_CTRL, enable ? NCTS_WRI_ANYTIME : 0,
-		      NCTS_WRI_ANYTIME);
+	regmap_update_bits(hdmi->regs, GRL_CTS_CTRL, NCTS_WRI_ANYTIME,
+			   enable ? NCTS_WRI_ANYTIME : 0);
 }
 
 static void mtk_hdmi_hw_msic_setting(struct mtk_hdmi *hdmi,
 				     struct drm_display_mode *mode)
 {
-	mtk_hdmi_clear_bits(hdmi, GRL_CFG4, CFG4_MHL_MODE);
+	regmap_clear_bits(hdmi->regs, GRL_CFG4, CFG4_MHL_MODE);
 
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE &&
 	    mode->clock == 74250 &&
 	    mode->vdisplay == 1080)
-		mtk_hdmi_clear_bits(hdmi, GRL_CFG2, CFG2_MHL_DE_SEL);
+		regmap_clear_bits(hdmi->regs, GRL_CFG2, CFG2_MHL_DE_SEL);
 	else
-		mtk_hdmi_set_bits(hdmi, GRL_CFG2, CFG2_MHL_DE_SEL);
+		regmap_set_bits(hdmi->regs, GRL_CFG2, CFG2_MHL_DE_SEL);
 }
 
 static void mtk_hdmi_hw_aud_set_channel_swap(struct mtk_hdmi *hdmi,
@@ -438,7 +397,7 @@ static void mtk_hdmi_hw_aud_set_channel_swap(struct mtk_hdmi *hdmi,
 		swap_bit = LFE_CC_SWAP;
 		break;
 	}
-	mtk_hdmi_mask(hdmi, GRL_CH_SWAP, swap_bit, 0xff);
+	regmap_update_bits(hdmi->regs, GRL_CH_SWAP, 0xff, swap_bit);
 }
 
 static void mtk_hdmi_hw_aud_set_bit_num(struct mtk_hdmi *hdmi,
@@ -459,7 +418,7 @@ static void mtk_hdmi_hw_aud_set_bit_num(struct mtk_hdmi *hdmi,
 		break;
 	}
 
-	mtk_hdmi_mask(hdmi, GRL_AOUT_CFG, val, AOUT_BNUM_SEL_MASK);
+	regmap_update_bits(hdmi->regs, GRL_AOUT_CFG, AOUT_BNUM_SEL_MASK, val);
 }
 
 static void mtk_hdmi_hw_aud_set_i2s_fmt(struct mtk_hdmi *hdmi,
@@ -467,7 +426,7 @@ static void mtk_hdmi_hw_aud_set_i2s_fmt(struct mtk_hdmi *hdmi,
 {
 	u32 val;
 
-	val = mtk_hdmi_read(hdmi, GRL_CFG0);
+	regmap_read(hdmi->regs, GRL_CFG0, &val);
 	val &= ~(CFG0_W_LENGTH_MASK | CFG0_I2S_MODE_MASK);
 
 	switch (i2s_fmt) {
@@ -491,7 +450,7 @@ static void mtk_hdmi_hw_aud_set_i2s_fmt(struct mtk_hdmi *hdmi,
 		val |= CFG0_I2S_MODE_I2S | CFG0_W_LENGTH_16BIT;
 		break;
 	}
-	mtk_hdmi_write(hdmi, GRL_CFG0, val);
+	regmap_write(hdmi->regs, GRL_CFG0, val);
 }
 
 static void mtk_hdmi_hw_audio_config(struct mtk_hdmi *hdmi, bool dst)
@@ -500,14 +459,14 @@ static void mtk_hdmi_hw_audio_config(struct mtk_hdmi *hdmi, bool dst)
 	u8 val;
 
 	/* Disable high bitrate, set DST packet normal/double */
-	mtk_hdmi_clear_bits(hdmi, GRL_AOUT_CFG, HIGH_BIT_RATE_PACKET_ALIGN);
+	regmap_clear_bits(hdmi->regs, GRL_AOUT_CFG, HIGH_BIT_RATE_PACKET_ALIGN);
 
 	if (dst)
 		val = DST_NORMAL_DOUBLE | SACD_DST;
 	else
 		val = 0;
 
-	mtk_hdmi_mask(hdmi, GRL_AUDIO_CFG, val, mask);
+	regmap_update_bits(hdmi->regs, GRL_AUDIO_CFG, mask, val);
 }
 
 static void mtk_hdmi_hw_aud_set_i2s_chan_num(struct mtk_hdmi *hdmi,
@@ -548,10 +507,10 @@ static void mtk_hdmi_hw_aud_set_i2s_chan_num(struct mtk_hdmi *hdmi,
 		i2s_uv = I2S_UV_CH_EN(0);
 	}
 
-	mtk_hdmi_write(hdmi, GRL_CH_SW0, ch_switch & 0xff);
-	mtk_hdmi_write(hdmi, GRL_CH_SW1, (ch_switch >> 8) & 0xff);
-	mtk_hdmi_write(hdmi, GRL_CH_SW2, (ch_switch >> 16) & 0xff);
-	mtk_hdmi_write(hdmi, GRL_I2S_UV, i2s_uv);
+	regmap_write(hdmi->regs, GRL_CH_SW0, ch_switch & 0xff);
+	regmap_write(hdmi->regs, GRL_CH_SW1, (ch_switch >> 8) & 0xff);
+	regmap_write(hdmi->regs, GRL_CH_SW2, (ch_switch >> 16) & 0xff);
+	regmap_write(hdmi->regs, GRL_I2S_UV, i2s_uv);
 }
 
 static void mtk_hdmi_hw_aud_set_input_type(struct mtk_hdmi *hdmi,
@@ -559,7 +518,7 @@ static void mtk_hdmi_hw_aud_set_input_type(struct mtk_hdmi *hdmi,
 {
 	u32 val;
 
-	val = mtk_hdmi_read(hdmi, GRL_CFG1);
+	regmap_read(hdmi->regs, GRL_CFG1, &val);
 	if (input_type == HDMI_AUD_INPUT_I2S &&
 	    (val & CFG1_SPDIF) == CFG1_SPDIF) {
 		val &= ~CFG1_SPDIF;
@@ -567,7 +526,7 @@ static void mtk_hdmi_hw_aud_set_input_type(struct mtk_hdmi *hdmi,
 		(val & CFG1_SPDIF) == 0) {
 		val |= CFG1_SPDIF;
 	}
-	mtk_hdmi_write(hdmi, GRL_CFG1, val);
+	regmap_write(hdmi->regs, GRL_CFG1, val);
 }
 
 static void mtk_hdmi_hw_aud_set_channel_status(struct mtk_hdmi *hdmi,
@@ -576,13 +535,13 @@ static void mtk_hdmi_hw_aud_set_channel_status(struct mtk_hdmi *hdmi,
 	int i;
 
 	for (i = 0; i < 5; i++) {
-		mtk_hdmi_write(hdmi, GRL_I2S_C_STA0 + i * 4, channel_status[i]);
-		mtk_hdmi_write(hdmi, GRL_L_STATUS_0 + i * 4, channel_status[i]);
-		mtk_hdmi_write(hdmi, GRL_R_STATUS_0 + i * 4, channel_status[i]);
+		regmap_write(hdmi->regs, GRL_I2S_C_STA0 + i * 4, channel_status[i]);
+		regmap_write(hdmi->regs, GRL_L_STATUS_0 + i * 4, channel_status[i]);
+		regmap_write(hdmi->regs, GRL_R_STATUS_0 + i * 4, channel_status[i]);
 	}
 	for (; i < 24; i++) {
-		mtk_hdmi_write(hdmi, GRL_L_STATUS_0 + i * 4, 0);
-		mtk_hdmi_write(hdmi, GRL_R_STATUS_0 + i * 4, 0);
+		regmap_write(hdmi->regs, GRL_L_STATUS_0 + i * 4, 0);
+		regmap_write(hdmi->regs, GRL_R_STATUS_0 + i * 4, 0);
 	}
 }
 
@@ -590,13 +549,13 @@ static void mtk_hdmi_hw_aud_src_reenable(struct mtk_hdmi *hdmi)
 {
 	u32 val;
 
-	val = mtk_hdmi_read(hdmi, GRL_MIX_CTRL);
+	regmap_read(hdmi->regs, GRL_MIX_CTRL, &val);
 	if (val & MIX_CTRL_SRC_EN) {
 		val &= ~MIX_CTRL_SRC_EN;
-		mtk_hdmi_write(hdmi, GRL_MIX_CTRL, val);
+		regmap_write(hdmi->regs, GRL_MIX_CTRL, val);
 		usleep_range(255, 512);
 		val |= MIX_CTRL_SRC_EN;
-		mtk_hdmi_write(hdmi, GRL_MIX_CTRL, val);
+		regmap_write(hdmi->regs, GRL_MIX_CTRL, val);
 	}
 }
 
@@ -604,10 +563,10 @@ static void mtk_hdmi_hw_aud_src_disable(struct mtk_hdmi *hdmi)
 {
 	u32 val;
 
-	val = mtk_hdmi_read(hdmi, GRL_MIX_CTRL);
+	regmap_read(hdmi->regs, GRL_MIX_CTRL, &val);
 	val &= ~MIX_CTRL_SRC_EN;
-	mtk_hdmi_write(hdmi, GRL_MIX_CTRL, val);
-	mtk_hdmi_write(hdmi, GRL_SHIFT_L1, 0x00);
+	regmap_write(hdmi->regs, GRL_MIX_CTRL, val);
+	regmap_write(hdmi->regs, GRL_SHIFT_L1, 0x00);
 }
 
 static void mtk_hdmi_hw_aud_set_mclk(struct mtk_hdmi *hdmi,
@@ -615,7 +574,7 @@ static void mtk_hdmi_hw_aud_set_mclk(struct mtk_hdmi *hdmi,
 {
 	u32 val;
 
-	val = mtk_hdmi_read(hdmi, GRL_CFG5);
+	regmap_read(hdmi->regs, GRL_CFG5, &val);
 	val &= CFG5_CD_RATIO_MASK;
 
 	switch (mclk) {
@@ -638,7 +597,7 @@ static void mtk_hdmi_hw_aud_set_mclk(struct mtk_hdmi *hdmi,
 		val |= CFG5_FS256;
 		break;
 	}
-	mtk_hdmi_write(hdmi, GRL_CFG5, val);
+	regmap_write(hdmi->regs, GRL_CFG5, val);
 }
 
 struct hdmi_acr_n {
@@ -716,15 +675,22 @@ static unsigned int hdmi_expected_cts(unsigned int audio_sample_rate,
 				     128 * audio_sample_rate);
 }
 
+static void mtk_hdmi_get_ncts(unsigned int sample_rate, unsigned int clock,
+			      unsigned int *n, unsigned int *cts)
+{
+	*n = hdmi_recommended_n(sample_rate, clock);
+	*cts = hdmi_expected_cts(sample_rate, clock, *n);
+}
+
 static void do_hdmi_hw_aud_set_ncts(struct mtk_hdmi *hdmi, unsigned int n,
 				    unsigned int cts)
 {
 	unsigned char val[NCTS_BYTES];
 	int i;
 
-	mtk_hdmi_write(hdmi, GRL_NCTS, 0);
-	mtk_hdmi_write(hdmi, GRL_NCTS, 0);
-	mtk_hdmi_write(hdmi, GRL_NCTS, 0);
+	regmap_write(hdmi->regs, GRL_NCTS, 0);
+	regmap_write(hdmi->regs, GRL_NCTS, 0);
+	regmap_write(hdmi->regs, GRL_NCTS, 0);
 	memset(val, 0, sizeof(val));
 
 	val[0] = (cts >> 24) & 0xff;
@@ -737,7 +703,7 @@ static void do_hdmi_hw_aud_set_ncts(struct mtk_hdmi *hdmi, unsigned int n,
 	val[6] = n & 0xff;
 
 	for (i = 0; i < NCTS_BYTES; i++)
-		mtk_hdmi_write(hdmi, GRL_NCTS, val[i]);
+		regmap_write(hdmi->regs, GRL_NCTS, val[i]);
 }
 
 static void mtk_hdmi_hw_aud_set_ncts(struct mtk_hdmi *hdmi,
@@ -746,14 +712,12 @@ static void mtk_hdmi_hw_aud_set_ncts(struct mtk_hdmi *hdmi,
 {
 	unsigned int n, cts;
 
-	n = hdmi_recommended_n(sample_rate, clock);
-	cts = hdmi_expected_cts(sample_rate, clock, n);
+	mtk_hdmi_get_ncts(sample_rate, clock, &n, &cts);
 
 	dev_dbg(hdmi->dev, "%s: sample_rate=%u, clock=%d, cts=%u, n=%u\n",
 		__func__, sample_rate, clock, n, cts);
 
-	mtk_hdmi_mask(hdmi, DUMMY_304, AUDIO_I2S_NCTS_SEL_64,
-		      AUDIO_I2S_NCTS_SEL);
+	regmap_update_bits(hdmi->regs, DUMMY_304, AUDIO_I2S_NCTS_SEL, AUDIO_I2S_NCTS_SEL_64);
 	do_hdmi_hw_aud_set_ncts(hdmi, n, cts);
 }
 
@@ -873,7 +837,7 @@ static void mtk_hdmi_aud_set_input(struct mtk_hdmi *hdmi)
 	bool dst;
 
 	mtk_hdmi_hw_aud_set_channel_swap(hdmi, HDMI_AUD_SWAP_LFE_CC);
-	mtk_hdmi_set_bits(hdmi, GRL_MIX_CTRL, MIX_CTRL_FLAT);
+	regmap_set_bits(hdmi->regs, GRL_MIX_CTRL, MIX_CTRL_FLAT);
 
 	if (hdmi->aud_param.aud_input_type == HDMI_AUD_INPUT_SPDIF &&
 	    hdmi->aud_param.aud_codec == HDMI_AUDIO_CODING_TYPE_DST) {
@@ -905,7 +869,7 @@ static int mtk_hdmi_aud_set_src(struct mtk_hdmi *hdmi,
 
 	mtk_hdmi_hw_ncts_enable(hdmi, false);
 	mtk_hdmi_hw_aud_src_disable(hdmi);
-	mtk_hdmi_clear_bits(hdmi, GRL_CFG2, CFG2_ACLK_INV);
+	regmap_clear_bits(hdmi->regs, GRL_CFG2, CFG2_ACLK_INV);
 
 	if (hdmi->aud_param.aud_input_type == HDMI_AUD_INPUT_I2S) {
 		switch (sample_rate) {
@@ -1061,20 +1025,6 @@ static int mtk_hdmi_setup_vendor_specific_infoframe(struct mtk_hdmi *hdmi,
 	return 0;
 }
 
-static int mtk_hdmi_output_init(struct mtk_hdmi *hdmi)
-{
-	struct hdmi_audio_param *aud_param = &hdmi->aud_param;
-
-	aud_param->aud_codec = HDMI_AUDIO_CODING_TYPE_PCM;
-	aud_param->aud_sample_size = HDMI_AUDIO_SAMPLE_SIZE_16;
-	aud_param->aud_input_type = HDMI_AUD_INPUT_I2S;
-	aud_param->aud_i2s_fmt = HDMI_I2S_MODE_I2S_24BIT;
-	aud_param->aud_mclk = HDMI_AUD_MCLK_128FS;
-	aud_param->aud_input_chan_type = HDMI_AUD_CHAN_TYPE_2_0;
-
-	return 0;
-}
-
 static void mtk_hdmi_audio_enable(struct mtk_hdmi *hdmi)
 {
 	mtk_hdmi_hw_send_aud_packet(hdmi, true);
@@ -1087,20 +1037,6 @@ static void mtk_hdmi_audio_disable(struct mtk_hdmi *hdmi)
 	hdmi->audio_enable = false;
 }
 
-static int mtk_hdmi_audio_set_param(struct mtk_hdmi *hdmi,
-				    struct hdmi_audio_param *param)
-{
-	if (!hdmi->audio_enable) {
-		dev_err(hdmi->dev, "hdmi audio is in disable state!\n");
-		return -EINVAL;
-	}
-	dev_dbg(hdmi->dev, "codec:%d, input:%d, channel:%d, fs:%d\n",
-		param->aud_codec, param->aud_input_type,
-		param->aud_input_chan_type, param->codec_params.sample_rate);
-	memcpy(&hdmi->aud_param, param, sizeof(*param));
-	return mtk_hdmi_aud_output_config(hdmi, &hdmi->mode);
-}
-
 static int mtk_hdmi_output_set_display_mode(struct mtk_hdmi *hdmi,
 					    struct drm_display_mode *mode)
 {
@@ -1269,6 +1205,7 @@ static const struct drm_edid *mtk_hdmi_bridge_edid_read(struct drm_bridge *bridg
 }
 
 static int mtk_hdmi_bridge_attach(struct drm_bridge *bridge,
+				  struct drm_encoder *encoder,
 				  enum drm_bridge_attach_flags flags)
 {
 	struct mtk_hdmi *hdmi = hdmi_ctx_from_bridge(bridge);
@@ -1281,7 +1218,7 @@ static int mtk_hdmi_bridge_attach(struct drm_bridge *bridge,
 	}
 
 	if (hdmi->next_bridge) {
-		ret = drm_bridge_attach(bridge->encoder, hdmi->next_bridge,
+		ret = drm_bridge_attach(encoder, hdmi->next_bridge,
 					bridge, flags);
 		if (ret)
 			return ret;
@@ -1407,30 +1344,20 @@ static const struct drm_bridge_funcs mtk_hdmi_bridge_funcs = {
 	.edid_read = mtk_hdmi_bridge_edid_read,
 };
 
-static int mtk_hdmi_dt_parse_pdata(struct mtk_hdmi *hdmi,
-				   struct platform_device *pdev)
+static int mtk_hdmi_get_cec_dev(struct mtk_hdmi *hdmi, struct device *dev, struct device_node *np)
 {
-	struct device *dev = &pdev->dev;
-	struct device_node *np = dev->of_node;
-	struct device_node *cec_np, *remote, *i2c_np;
 	struct platform_device *cec_pdev;
-	struct regmap *regmap;
+	struct device_node *cec_np;
 	int ret;
 
 	ret = mtk_hdmi_get_all_clk(hdmi, np);
-	if (ret) {
-		if (ret != -EPROBE_DEFER)
-			dev_err(dev, "Failed to get clocks: %d\n", ret);
-
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get clocks\n");
 
 	/* The CEC module handles HDMI hotplug detection */
 	cec_np = of_get_compatible_child(np->parent, "mediatek,mt8173-cec");
-	if (!cec_np) {
-		dev_err(dev, "Failed to find CEC node\n");
-		return -EINVAL;
-	}
+	if (!cec_np)
+		return dev_err_probe(dev, -EINVAL, "Failed to find CEC node\n");
 
 	cec_pdev = of_find_device_by_node(cec_np);
 	if (!cec_pdev) {
@@ -1440,82 +1367,77 @@ static int mtk_hdmi_dt_parse_pdata(struct mtk_hdmi *hdmi,
 		return -EPROBE_DEFER;
 	}
 	of_node_put(cec_np);
-	hdmi->cec_dev = &cec_pdev->dev;
 
 	/*
 	 * The mediatek,syscon-hdmi property contains a phandle link to the
 	 * MMSYS_CONFIG device and the register offset of the HDMI_SYS_CFG
 	 * registers it contains.
 	 */
-	regmap = syscon_regmap_lookup_by_phandle(np, "mediatek,syscon-hdmi");
-	ret = of_property_read_u32_index(np, "mediatek,syscon-hdmi", 1,
-					 &hdmi->sys_offset);
-	if (IS_ERR(regmap))
-		ret = PTR_ERR(regmap);
-	if (ret) {
-		dev_err(dev,
-			"Failed to get system configuration registers: %d\n",
-			ret);
-		goto put_device;
-	}
-	hdmi->sys_regmap = regmap;
+	hdmi->sys_regmap = syscon_regmap_lookup_by_phandle_args(np, "mediatek,syscon-hdmi",
+								1, &hdmi->sys_offset);
+	if (IS_ERR(hdmi->sys_regmap))
+		return dev_err_probe(dev, PTR_ERR(hdmi->sys_regmap),
+				     "Failed to get system configuration registers\n");
 
-	hdmi->regs = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(hdmi->regs)) {
-		ret = PTR_ERR(hdmi->regs);
-		goto put_device;
-	}
+	hdmi->cec_dev = &cec_pdev->dev;
+	return 0;
+}
+
+static int mtk_hdmi_dt_parse_pdata(struct mtk_hdmi *hdmi,
+				   struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *np = dev->of_node;
+	struct device_node *remote, *i2c_np;
+	int ret;
+
+	ret = mtk_hdmi_get_all_clk(hdmi, np);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get clocks\n");
+
+	hdmi->regs = device_node_to_regmap(dev->of_node);
+	if (IS_ERR(hdmi->regs))
+		return PTR_ERR(hdmi->regs);
 
 	remote = of_graph_get_remote_node(np, 1, 0);
-	if (!remote) {
-		ret = -EINVAL;
-		goto put_device;
-	}
+	if (!remote)
+		return -EINVAL;
 
 	if (!of_device_is_compatible(remote, "hdmi-connector")) {
 		hdmi->next_bridge = of_drm_find_bridge(remote);
 		if (!hdmi->next_bridge) {
 			dev_err(dev, "Waiting for external bridge\n");
 			of_node_put(remote);
-			ret = -EPROBE_DEFER;
-			goto put_device;
+			return -EPROBE_DEFER;
 		}
 	}
 
 	i2c_np = of_parse_phandle(remote, "ddc-i2c-bus", 0);
-	if (!i2c_np) {
-		dev_err(dev, "Failed to find ddc-i2c-bus node in %pOF\n",
-			remote);
-		of_node_put(remote);
-		ret = -EINVAL;
-		goto put_device;
-	}
 	of_node_put(remote);
+	if (!i2c_np)
+		return dev_err_probe(dev, -EINVAL, "No ddc-i2c-bus in connector\n");
 
 	hdmi->ddc_adpt = of_find_i2c_adapter_by_node(i2c_np);
 	of_node_put(i2c_np);
-	if (!hdmi->ddc_adpt) {
-		dev_err(dev, "Failed to get ddc i2c adapter by node\n");
-		ret = -EINVAL;
-		goto put_device;
-	}
+	if (!hdmi->ddc_adpt)
+		return dev_err_probe(dev, -EINVAL, "Failed to get ddc i2c adapter by node\n");
+
+	ret = mtk_hdmi_get_cec_dev(hdmi, dev, np);
+	if (ret)
+		return ret;
 
 	return 0;
-put_device:
-	put_device(hdmi->cec_dev);
-	return ret;
 }
 
 /*
  * HDMI audio codec callbacks
  */
 
-static int mtk_hdmi_audio_hw_params(struct device *dev, void *data,
-				    struct hdmi_codec_daifmt *daifmt,
-				    struct hdmi_codec_params *params)
+static int mtk_hdmi_audio_params(struct mtk_hdmi *hdmi,
+				 struct hdmi_codec_daifmt *daifmt,
+				 struct hdmi_codec_params *params)
 {
-	struct mtk_hdmi *hdmi = dev_get_drvdata(dev);
-	struct hdmi_audio_param hdmi_params;
+	struct hdmi_audio_param aud_params = { 0 };
 	unsigned int chan = params->cea.channels;
 
 	dev_dbg(hdmi->dev, "%s: %u Hz, %d bit, %d channels\n", __func__,
@@ -1526,16 +1448,16 @@ static int mtk_hdmi_audio_hw_params(struct device *dev, void *data,
 
 	switch (chan) {
 	case 2:
-		hdmi_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_2_0;
+		aud_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_2_0;
 		break;
 	case 4:
-		hdmi_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_4_0;
+		aud_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_4_0;
 		break;
 	case 6:
-		hdmi_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_5_1;
+		aud_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_5_1;
 		break;
 	case 8:
-		hdmi_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_7_1;
+		aud_params.aud_input_chan_type = HDMI_AUD_CHAN_TYPE_7_1;
 		break;
 	default:
 		dev_err(hdmi->dev, "channel[%d] not supported!\n", chan);
@@ -1559,27 +1481,45 @@ static int mtk_hdmi_audio_hw_params(struct device *dev, void *data,
 
 	switch (daifmt->fmt) {
 	case HDMI_I2S:
-		hdmi_params.aud_codec = HDMI_AUDIO_CODING_TYPE_PCM;
-		hdmi_params.aud_sample_size = HDMI_AUDIO_SAMPLE_SIZE_16;
-		hdmi_params.aud_input_type = HDMI_AUD_INPUT_I2S;
-		hdmi_params.aud_i2s_fmt = HDMI_I2S_MODE_I2S_24BIT;
-		hdmi_params.aud_mclk = HDMI_AUD_MCLK_128FS;
+		aud_params.aud_codec = HDMI_AUDIO_CODING_TYPE_PCM;
+		aud_params.aud_sample_size = HDMI_AUDIO_SAMPLE_SIZE_16;
+		aud_params.aud_input_type = HDMI_AUD_INPUT_I2S;
+		aud_params.aud_i2s_fmt = HDMI_I2S_MODE_I2S_24BIT;
+		aud_params.aud_mclk = HDMI_AUD_MCLK_128FS;
 		break;
 	case HDMI_SPDIF:
-		hdmi_params.aud_codec = HDMI_AUDIO_CODING_TYPE_PCM;
-		hdmi_params.aud_sample_size = HDMI_AUDIO_SAMPLE_SIZE_16;
-		hdmi_params.aud_input_type = HDMI_AUD_INPUT_SPDIF;
+		aud_params.aud_codec = HDMI_AUDIO_CODING_TYPE_PCM;
+		aud_params.aud_sample_size = HDMI_AUDIO_SAMPLE_SIZE_16;
+		aud_params.aud_input_type = HDMI_AUD_INPUT_SPDIF;
 		break;
 	default:
 		dev_err(hdmi->dev, "%s: Invalid DAI format %d\n", __func__,
 			daifmt->fmt);
 		return -EINVAL;
 	}
+	memcpy(&aud_params.codec_params, params, sizeof(aud_params.codec_params));
+	memcpy(&hdmi->aud_param, &aud_params, sizeof(aud_params));
+
+	dev_dbg(hdmi->dev, "codec:%d, input:%d, channel:%d, fs:%d\n",
+		aud_params.aud_codec, aud_params.aud_input_type,
+		aud_params.aud_input_chan_type, aud_params.codec_params.sample_rate);
 
-	memcpy(&hdmi_params.codec_params, params,
-	       sizeof(hdmi_params.codec_params));
+	return 0;
+}
+
+static int mtk_hdmi_audio_hw_params(struct device *dev, void *data,
+				    struct hdmi_codec_daifmt *daifmt,
+				    struct hdmi_codec_params *params)
+{
+	struct mtk_hdmi *hdmi = dev_get_drvdata(dev);
 
-	mtk_hdmi_audio_set_param(hdmi, &hdmi_params);
+	if (!hdmi->audio_enable) {
+		dev_err(hdmi->dev, "hdmi audio is in disable state!\n");
+		return -EINVAL;
+	}
+
+	mtk_hdmi_audio_params(hdmi, daifmt, params);
+	mtk_hdmi_aud_output_config(hdmi, &hdmi->mode);
 
 	return 0;
 }
@@ -1625,17 +1565,22 @@ static int mtk_hdmi_audio_get_eld(struct device *dev, void *data, uint8_t *buf,
 	return 0;
 }
 
-static int mtk_hdmi_audio_hook_plugged_cb(struct device *dev, void *data,
-					  hdmi_codec_plugged_cb fn,
+static void mtk_hdmi_audio_set_plugged_cb(struct mtk_hdmi *hdmi, hdmi_codec_plugged_cb fn,
 					  struct device *codec_dev)
 {
-	struct mtk_hdmi *hdmi = data;
-
 	mutex_lock(&hdmi->update_plugged_status_lock);
 	hdmi->plugged_cb = fn;
 	hdmi->codec_dev = codec_dev;
 	mutex_unlock(&hdmi->update_plugged_status_lock);
+}
+
+static int mtk_hdmi_audio_hook_plugged_cb(struct device *dev, void *data,
+					  hdmi_codec_plugged_cb fn,
+					  struct device *codec_dev)
+{
+	struct mtk_hdmi *hdmi = data;
 
+	mtk_hdmi_audio_set_plugged_cb(hdmi, fn, codec_dev);
 	mtk_hdmi_update_plugged_status(hdmi);
 
 	return 0;
@@ -1658,6 +1603,7 @@ static void mtk_hdmi_unregister_audio_driver(void *data)
 static int mtk_hdmi_register_audio_driver(struct device *dev)
 {
 	struct mtk_hdmi *hdmi = dev_get_drvdata(dev);
+	struct hdmi_audio_param *aud_param = &hdmi->aud_param;
 	struct hdmi_codec_pdata codec_data = {
 		.ops = &mtk_hdmi_audio_codec_ops,
 		.max_i2s_channels = 2,
@@ -1667,6 +1613,13 @@ static int mtk_hdmi_register_audio_driver(struct device *dev)
 	};
 	int ret;
 
+	aud_param->aud_codec = HDMI_AUDIO_CODING_TYPE_PCM;
+	aud_param->aud_sample_size = HDMI_AUDIO_SAMPLE_SIZE_16;
+	aud_param->aud_input_type = HDMI_AUD_INPUT_I2S;
+	aud_param->aud_i2s_fmt = HDMI_I2S_MODE_I2S_24BIT;
+	aud_param->aud_mclk = HDMI_AUD_MCLK_128FS;
+	aud_param->aud_input_chan_type = HDMI_AUD_CHAN_TYPE_2_0;
+
 	hdmi->audio_pdev = platform_device_register_data(dev,
 							 HDMI_CODEC_DRV_NAME,
 							 PLATFORM_DEVID_AUTO,
@@ -1708,11 +1661,6 @@ static int mtk_hdmi_probe(struct platform_device *pdev)
 	mutex_init(&hdmi->update_plugged_status_lock);
 	platform_set_drvdata(pdev, hdmi);
 
-	ret = mtk_hdmi_output_init(hdmi);
-	if (ret)
-		return dev_err_probe(dev, ret,
-				     "Failed to initialize hdmi output\n");
-
 	ret = mtk_hdmi_register_audio_driver(dev);
 	if (ret)
 		return dev_err_probe(dev, ret,
@@ -1789,28 +1737,9 @@ static struct platform_driver mtk_hdmi_driver = {
 		.pm = &mtk_hdmi_pm_ops,
 	},
 };
-
-static struct platform_driver * const mtk_hdmi_drivers[] = {
-	&mtk_hdmi_ddc_driver,
-	&mtk_cec_driver,
-	&mtk_hdmi_driver,
-};
-
-static int __init mtk_hdmitx_init(void)
-{
-	return platform_register_drivers(mtk_hdmi_drivers,
-					 ARRAY_SIZE(mtk_hdmi_drivers));
-}
-
-static void __exit mtk_hdmitx_exit(void)
-{
-	platform_unregister_drivers(mtk_hdmi_drivers,
-				    ARRAY_SIZE(mtk_hdmi_drivers));
-}
-
-module_init(mtk_hdmitx_init);
-module_exit(mtk_hdmitx_exit);
+module_platform_driver(mtk_hdmi_driver);
 
 MODULE_AUTHOR("Jie Qiu <jie.qiu@mediatek.com>");
 MODULE_DESCRIPTION("MediaTek HDMI Driver");
 MODULE_LICENSE("GPL v2");
+MODULE_IMPORT_NS("DRM_MTK_HDMI_V1");
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi.h b/drivers/gpu/drm/mediatek/mtk_hdmi.h
deleted file mode 100644
index 472bf141c92b..000000000000
--- a/drivers/gpu/drm/mediatek/mtk_hdmi.h
+++ /dev/null
@@ -1,14 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * Copyright (c) 2014 MediaTek Inc.
- * Author: Jie Qiu <jie.qiu@mediatek.com>
- */
-#ifndef _MTK_HDMI_CTRL_H
-#define _MTK_HDMI_CTRL_H
-
-struct platform_driver;
-
-extern struct platform_driver mtk_cec_driver;
-extern struct platform_driver mtk_hdmi_ddc_driver;
-
-#endif /* _MTK_HDMI_CTRL_H */
diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi_ddc.c b/drivers/gpu/drm/mediatek/mtk_hdmi_ddc.c
index 07db68067844..6358e1af69b4 100644
--- a/drivers/gpu/drm/mediatek/mtk_hdmi_ddc.c
+++ b/drivers/gpu/drm/mediatek/mtk_hdmi_ddc.c
@@ -20,7 +20,6 @@
 #include <linux/of_platform.h>
 
 #include "mtk_drm_drv.h"
-#include "mtk_hdmi.h"
 
 #define SIF1_CLOK		(288)
 #define DDC_DDCMCTL0		(0x0)
@@ -337,6 +336,7 @@ struct platform_driver mtk_hdmi_ddc_driver = {
 		.of_match_table = mtk_hdmi_ddc_match,
 	},
 };
+module_platform_driver(mtk_hdmi_ddc_driver);
 
 MODULE_AUTHOR("Jie Qiu <jie.qiu@mediatek.com>");
 MODULE_DESCRIPTION("MediaTek HDMI DDC Driver");
diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index 81d2ee37e773..49ff9f1f16d3 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -169,7 +169,7 @@ static const struct meson_drm_soc_attr meson_drm_soc_attrs[] = {
 	/* S805X/S805Y HDMI PLL won't lock for HDMI PHY freq > 1,65GHz */
 	{
 		.limits = {
-			.max_hdmi_phy_freq = 1650000,
+			.max_hdmi_phy_freq = 1650000000,
 		},
 		.attrs = (const struct soc_device_attribute []) {
 			{ .soc_id = "GXL (S805*)", },
diff --git a/drivers/gpu/drm/meson/meson_drv.h b/drivers/gpu/drm/meson/meson_drv.h
index 3f9345c14f31..be4b0e4df6e1 100644
--- a/drivers/gpu/drm/meson/meson_drv.h
+++ b/drivers/gpu/drm/meson/meson_drv.h
@@ -37,7 +37,7 @@ struct meson_drm_match_data {
 };
 
 struct meson_drm_soc_limits {
-	unsigned int max_hdmi_phy_freq;
+	unsigned long long max_hdmi_phy_freq;
 };
 
 struct meson_drm {
diff --git a/drivers/gpu/drm/meson/meson_encoder_cvbs.c b/drivers/gpu/drm/meson/meson_encoder_cvbs.c
index e79f7c3ce32e..c9678dc68fa1 100644
--- a/drivers/gpu/drm/meson/meson_encoder_cvbs.c
+++ b/drivers/gpu/drm/meson/meson_encoder_cvbs.c
@@ -83,12 +83,13 @@ meson_cvbs_get_mode(const struct drm_display_mode *req_mode)
 }
 
 static int meson_encoder_cvbs_attach(struct drm_bridge *bridge,
+				     struct drm_encoder *encoder,
 				     enum drm_bridge_attach_flags flags)
 {
 	struct meson_encoder_cvbs *meson_encoder_cvbs =
 					bridge_to_meson_encoder_cvbs(bridge);
 
-	return drm_bridge_attach(bridge->encoder, meson_encoder_cvbs->next_bridge,
+	return drm_bridge_attach(encoder, meson_encoder_cvbs->next_bridge,
 				 &meson_encoder_cvbs->bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/meson/meson_encoder_dsi.c b/drivers/gpu/drm/meson/meson_encoder_dsi.c
index fe204437bd65..3db518e5f95d 100644
--- a/drivers/gpu/drm/meson/meson_encoder_dsi.c
+++ b/drivers/gpu/drm/meson/meson_encoder_dsi.c
@@ -33,11 +33,12 @@ struct meson_encoder_dsi {
 	container_of(x, struct meson_encoder_dsi, bridge)
 
 static int meson_encoder_dsi_attach(struct drm_bridge *bridge,
+				    struct drm_encoder *encoder,
 				    enum drm_bridge_attach_flags flags)
 {
 	struct meson_encoder_dsi *encoder_dsi = bridge_to_meson_encoder_dsi(bridge);
 
-	return drm_bridge_attach(bridge->encoder, encoder_dsi->next_bridge,
+	return drm_bridge_attach(encoder, encoder_dsi->next_bridge,
 				 &encoder_dsi->bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/meson/meson_encoder_hdmi.c b/drivers/gpu/drm/meson/meson_encoder_hdmi.c
index 6d1c9262a2cf..040c04b5dff9 100644
--- a/drivers/gpu/drm/meson/meson_encoder_hdmi.c
+++ b/drivers/gpu/drm/meson/meson_encoder_hdmi.c
@@ -49,11 +49,12 @@ struct meson_encoder_hdmi {
 	container_of(x, struct meson_encoder_hdmi, bridge)
 
 static int meson_encoder_hdmi_attach(struct drm_bridge *bridge,
+				     struct drm_encoder *encoder,
 				     enum drm_bridge_attach_flags flags)
 {
 	struct meson_encoder_hdmi *encoder_hdmi = bridge_to_meson_encoder_hdmi(bridge);
 
-	return drm_bridge_attach(bridge->encoder, encoder_hdmi->next_bridge,
+	return drm_bridge_attach(encoder, encoder_hdmi->next_bridge,
 				 &encoder_hdmi->bridge, flags);
 }
 
@@ -70,12 +71,12 @@ static void meson_encoder_hdmi_set_vclk(struct meson_encoder_hdmi *encoder_hdmi,
 {
 	struct meson_drm *priv = encoder_hdmi->priv;
 	int vic = drm_match_cea_mode(mode);
-	unsigned int phy_freq;
-	unsigned int vclk_freq;
-	unsigned int venc_freq;
-	unsigned int hdmi_freq;
+	unsigned long long phy_freq;
+	unsigned long long vclk_freq;
+	unsigned long long venc_freq;
+	unsigned long long hdmi_freq;
 
-	vclk_freq = mode->clock;
+	vclk_freq = mode->clock * 1000;
 
 	/* For 420, pixel clock is half unlike venc clock */
 	if (encoder_hdmi->output_bus_fmt == MEDIA_BUS_FMT_UYYVYY8_0_5X24)
@@ -107,7 +108,8 @@ static void meson_encoder_hdmi_set_vclk(struct meson_encoder_hdmi *encoder_hdmi,
 	if (mode->flags & DRM_MODE_FLAG_DBLCLK)
 		venc_freq /= 2;
 
-	dev_dbg(priv->dev, "vclk:%d phy=%d venc=%d hdmi=%d enci=%d\n",
+	dev_dbg(priv->dev,
+		"vclk:%lluHz phy=%lluHz venc=%lluHz hdmi=%lluHz enci=%d\n",
 		phy_freq, vclk_freq, venc_freq, hdmi_freq,
 		priv->venc.hdmi_use_enci);
 
@@ -122,10 +124,11 @@ static enum drm_mode_status meson_encoder_hdmi_mode_valid(struct drm_bridge *bri
 	struct meson_encoder_hdmi *encoder_hdmi = bridge_to_meson_encoder_hdmi(bridge);
 	struct meson_drm *priv = encoder_hdmi->priv;
 	bool is_hdmi2_sink = display_info->hdmi.scdc.supported;
-	unsigned int phy_freq;
-	unsigned int vclk_freq;
-	unsigned int venc_freq;
-	unsigned int hdmi_freq;
+	unsigned long long clock = mode->clock * 1000;
+	unsigned long long phy_freq;
+	unsigned long long vclk_freq;
+	unsigned long long venc_freq;
+	unsigned long long hdmi_freq;
 	int vic = drm_match_cea_mode(mode);
 	enum drm_mode_status status;
 
@@ -144,12 +147,12 @@ static enum drm_mode_status meson_encoder_hdmi_mode_valid(struct drm_bridge *bri
 		if (status != MODE_OK)
 			return status;
 
-		return meson_vclk_dmt_supported_freq(priv, mode->clock);
+		return meson_vclk_dmt_supported_freq(priv, clock);
 	/* Check against supported VIC modes */
 	} else if (!meson_venc_hdmi_supported_vic(vic))
 		return MODE_BAD;
 
-	vclk_freq = mode->clock;
+	vclk_freq = clock;
 
 	/* For 420, pixel clock is half unlike venc clock */
 	if (drm_mode_is_420_only(display_info, mode) ||
@@ -179,7 +182,8 @@ static enum drm_mode_status meson_encoder_hdmi_mode_valid(struct drm_bridge *bri
 	if (mode->flags & DRM_MODE_FLAG_DBLCLK)
 		venc_freq /= 2;
 
-	dev_dbg(priv->dev, "%s: vclk:%d phy=%d venc=%d hdmi=%d\n",
+	dev_dbg(priv->dev,
+		"%s: vclk:%lluHz phy=%lluHz venc=%lluHz hdmi=%lluHz\n",
 		__func__, phy_freq, vclk_freq, venc_freq, hdmi_freq);
 
 	return meson_vclk_vic_supported_freq(priv, phy_freq, vclk_freq);
diff --git a/drivers/gpu/drm/meson/meson_vclk.c b/drivers/gpu/drm/meson/meson_vclk.c
index 2a942dc6a6dc..3325580d885d 100644
--- a/drivers/gpu/drm/meson/meson_vclk.c
+++ b/drivers/gpu/drm/meson/meson_vclk.c
@@ -110,7 +110,10 @@
 #define HDMI_PLL_LOCK		BIT(31)
 #define HDMI_PLL_LOCK_G12A	(3 << 30)
 
-#define FREQ_1000_1001(_freq)	DIV_ROUND_CLOSEST(_freq * 1000, 1001)
+#define PIXEL_FREQ_1000_1001(_freq)	\
+	DIV_ROUND_CLOSEST_ULL((_freq) * 1000ULL, 1001ULL)
+#define PHY_FREQ_1000_1001(_freq)	\
+	(PIXEL_FREQ_1000_1001(DIV_ROUND_DOWN_ULL(_freq, 10ULL)) * 10)
 
 /* VID PLL Dividers */
 enum {
@@ -360,11 +363,11 @@ enum {
 };
 
 struct meson_vclk_params {
-	unsigned int pll_freq;
-	unsigned int phy_freq;
-	unsigned int vclk_freq;
-	unsigned int venc_freq;
-	unsigned int pixel_freq;
+	unsigned long long pll_freq;
+	unsigned long long phy_freq;
+	unsigned long long vclk_freq;
+	unsigned long long venc_freq;
+	unsigned long long pixel_freq;
 	unsigned int pll_od1;
 	unsigned int pll_od2;
 	unsigned int pll_od3;
@@ -372,11 +375,11 @@ struct meson_vclk_params {
 	unsigned int vclk_div;
 } params[] = {
 	[MESON_VCLK_HDMI_ENCI_54000] = {
-		.pll_freq = 4320000,
-		.phy_freq = 270000,
-		.vclk_freq = 54000,
-		.venc_freq = 54000,
-		.pixel_freq = 54000,
+		.pll_freq = 4320000000,
+		.phy_freq = 270000000,
+		.vclk_freq = 54000000,
+		.venc_freq = 54000000,
+		.pixel_freq = 54000000,
 		.pll_od1 = 4,
 		.pll_od2 = 4,
 		.pll_od3 = 1,
@@ -384,11 +387,11 @@ struct meson_vclk_params {
 		.vclk_div = 1,
 	},
 	[MESON_VCLK_HDMI_DDR_54000] = {
-		.pll_freq = 4320000,
-		.phy_freq = 270000,
-		.vclk_freq = 54000,
-		.venc_freq = 54000,
-		.pixel_freq = 27000,
+		.pll_freq = 4320000000,
+		.phy_freq = 270000000,
+		.vclk_freq = 54000000,
+		.venc_freq = 54000000,
+		.pixel_freq = 27000000,
 		.pll_od1 = 4,
 		.pll_od2 = 4,
 		.pll_od3 = 1,
@@ -396,11 +399,11 @@ struct meson_vclk_params {
 		.vclk_div = 1,
 	},
 	[MESON_VCLK_HDMI_DDR_148500] = {
-		.pll_freq = 2970000,
-		.phy_freq = 742500,
-		.vclk_freq = 148500,
-		.venc_freq = 148500,
-		.pixel_freq = 74250,
+		.pll_freq = 2970000000,
+		.phy_freq = 742500000,
+		.vclk_freq = 148500000,
+		.venc_freq = 148500000,
+		.pixel_freq = 74250000,
 		.pll_od1 = 4,
 		.pll_od2 = 1,
 		.pll_od3 = 1,
@@ -408,11 +411,11 @@ struct meson_vclk_params {
 		.vclk_div = 1,
 	},
 	[MESON_VCLK_HDMI_74250] = {
-		.pll_freq = 2970000,
-		.phy_freq = 742500,
-		.vclk_freq = 74250,
-		.venc_freq = 74250,
-		.pixel_freq = 74250,
+		.pll_freq = 2970000000,
+		.phy_freq = 742500000,
+		.vclk_freq = 74250000,
+		.venc_freq = 74250000,
+		.pixel_freq = 74250000,
 		.pll_od1 = 2,
 		.pll_od2 = 2,
 		.pll_od3 = 2,
@@ -420,11 +423,11 @@ struct meson_vclk_params {
 		.vclk_div = 1,
 	},
 	[MESON_VCLK_HDMI_148500] = {
-		.pll_freq = 2970000,
-		.phy_freq = 1485000,
-		.vclk_freq = 148500,
-		.venc_freq = 148500,
-		.pixel_freq = 148500,
+		.pll_freq = 2970000000,
+		.phy_freq = 1485000000,
+		.vclk_freq = 148500000,
+		.venc_freq = 148500000,
+		.pixel_freq = 148500000,
 		.pll_od1 = 1,
 		.pll_od2 = 2,
 		.pll_od3 = 2,
@@ -432,11 +435,11 @@ struct meson_vclk_params {
 		.vclk_div = 1,
 	},
 	[MESON_VCLK_HDMI_297000] = {
-		.pll_freq = 5940000,
-		.phy_freq = 2970000,
-		.venc_freq = 297000,
-		.vclk_freq = 297000,
-		.pixel_freq = 297000,
+		.pll_freq = 5940000000,
+		.phy_freq = 2970000000,
+		.venc_freq = 297000000,
+		.vclk_freq = 297000000,
+		.pixel_freq = 297000000,
 		.pll_od1 = 2,
 		.pll_od2 = 1,
 		.pll_od3 = 1,
@@ -444,11 +447,11 @@ struct meson_vclk_params {
 		.vclk_div = 2,
 	},
 	[MESON_VCLK_HDMI_594000] = {
-		.pll_freq = 5940000,
-		.phy_freq = 5940000,
-		.venc_freq = 594000,
-		.vclk_freq = 594000,
-		.pixel_freq = 594000,
+		.pll_freq = 5940000000,
+		.phy_freq = 5940000000,
+		.venc_freq = 594000000,
+		.vclk_freq = 594000000,
+		.pixel_freq = 594000000,
 		.pll_od1 = 1,
 		.pll_od2 = 1,
 		.pll_od3 = 2,
@@ -456,11 +459,11 @@ struct meson_vclk_params {
 		.vclk_div = 1,
 	},
 	[MESON_VCLK_HDMI_594000_YUV420] = {
-		.pll_freq = 5940000,
-		.phy_freq = 2970000,
-		.venc_freq = 594000,
-		.vclk_freq = 594000,
-		.pixel_freq = 297000,
+		.pll_freq = 5940000000,
+		.phy_freq = 2970000000,
+		.venc_freq = 594000000,
+		.vclk_freq = 594000000,
+		.pixel_freq = 297000000,
 		.pll_od1 = 2,
 		.pll_od2 = 1,
 		.pll_od3 = 1,
@@ -617,16 +620,16 @@ static void meson_hdmi_pll_set_params(struct meson_drm *priv, unsigned int m,
 				3 << 20, pll_od_to_reg(od3) << 20);
 }
 
-#define XTAL_FREQ 24000
+#define XTAL_FREQ (24 * 1000 * 1000)
 
 static unsigned int meson_hdmi_pll_get_m(struct meson_drm *priv,
-					 unsigned int pll_freq)
+					 unsigned long long pll_freq)
 {
 	/* The GXBB PLL has a /2 pre-multiplier */
 	if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_GXBB))
-		pll_freq /= 2;
+		pll_freq = DIV_ROUND_DOWN_ULL(pll_freq, 2);
 
-	return pll_freq / XTAL_FREQ;
+	return DIV_ROUND_DOWN_ULL(pll_freq, XTAL_FREQ);
 }
 
 #define HDMI_FRAC_MAX_GXBB	4096
@@ -635,12 +638,13 @@ static unsigned int meson_hdmi_pll_get_m(struct meson_drm *priv,
 
 static unsigned int meson_hdmi_pll_get_frac(struct meson_drm *priv,
 					    unsigned int m,
-					    unsigned int pll_freq)
+					    unsigned long long pll_freq)
 {
-	unsigned int parent_freq = XTAL_FREQ;
+	unsigned long long parent_freq = XTAL_FREQ;
 	unsigned int frac_max = HDMI_FRAC_MAX_GXL;
 	unsigned int frac_m;
 	unsigned int frac;
+	u32 remainder;
 
 	/* The GXBB PLL has a /2 pre-multiplier and a larger FRAC width */
 	if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_GXBB)) {
@@ -652,11 +656,11 @@ static unsigned int meson_hdmi_pll_get_frac(struct meson_drm *priv,
 		frac_max = HDMI_FRAC_MAX_G12A;
 
 	/* We can have a perfect match !*/
-	if (pll_freq / m == parent_freq &&
-	    pll_freq % m == 0)
+	if (div_u64_rem(pll_freq, m, &remainder) == parent_freq &&
+	    remainder == 0)
 		return 0;
 
-	frac = div_u64((u64)pll_freq * (u64)frac_max, parent_freq);
+	frac = mul_u64_u64_div_u64(pll_freq, frac_max, parent_freq);
 	frac_m = m * frac_max;
 	if (frac_m > frac)
 		return frac_max;
@@ -666,7 +670,7 @@ static unsigned int meson_hdmi_pll_get_frac(struct meson_drm *priv,
 }
 
 static bool meson_hdmi_pll_validate_params(struct meson_drm *priv,
-					   unsigned int m,
+					   unsigned long long m,
 					   unsigned int frac)
 {
 	if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_GXBB)) {
@@ -694,7 +698,7 @@ static bool meson_hdmi_pll_validate_params(struct meson_drm *priv,
 }
 
 static bool meson_hdmi_pll_find_params(struct meson_drm *priv,
-				       unsigned int freq,
+				       unsigned long long freq,
 				       unsigned int *m,
 				       unsigned int *frac,
 				       unsigned int *od)
@@ -706,7 +710,7 @@ static bool meson_hdmi_pll_find_params(struct meson_drm *priv,
 			continue;
 		*frac = meson_hdmi_pll_get_frac(priv, *m, freq * *od);
 
-		DRM_DEBUG_DRIVER("PLL params for %dkHz: m=%x frac=%x od=%d\n",
+		DRM_DEBUG_DRIVER("PLL params for %lluHz: m=%x frac=%x od=%d\n",
 				 freq, *m, *frac, *od);
 
 		if (meson_hdmi_pll_validate_params(priv, *m, *frac))
@@ -718,7 +722,7 @@ static bool meson_hdmi_pll_find_params(struct meson_drm *priv,
 
 /* pll_freq is the frequency after the OD dividers */
 enum drm_mode_status
-meson_vclk_dmt_supported_freq(struct meson_drm *priv, unsigned int freq)
+meson_vclk_dmt_supported_freq(struct meson_drm *priv, unsigned long long freq)
 {
 	unsigned int od, m, frac;
 
@@ -741,7 +745,7 @@ EXPORT_SYMBOL_GPL(meson_vclk_dmt_supported_freq);
 
 /* pll_freq is the frequency after the OD dividers */
 static void meson_hdmi_pll_generic_set(struct meson_drm *priv,
-				       unsigned int pll_freq)
+				       unsigned long long pll_freq)
 {
 	unsigned int od, m, frac, od1, od2, od3;
 
@@ -756,7 +760,7 @@ static void meson_hdmi_pll_generic_set(struct meson_drm *priv,
 			od1 = od / od2;
 		}
 
-		DRM_DEBUG_DRIVER("PLL params for %dkHz: m=%x frac=%x od=%d/%d/%d\n",
+		DRM_DEBUG_DRIVER("PLL params for %lluHz: m=%x frac=%x od=%d/%d/%d\n",
 				 pll_freq, m, frac, od1, od2, od3);
 
 		meson_hdmi_pll_set_params(priv, m, frac, od1, od2, od3);
@@ -764,17 +768,18 @@ static void meson_hdmi_pll_generic_set(struct meson_drm *priv,
 		return;
 	}
 
-	DRM_ERROR("Fatal, unable to find parameters for PLL freq %d\n",
+	DRM_ERROR("Fatal, unable to find parameters for PLL freq %lluHz\n",
 		  pll_freq);
 }
 
 enum drm_mode_status
-meson_vclk_vic_supported_freq(struct meson_drm *priv, unsigned int phy_freq,
-			      unsigned int vclk_freq)
+meson_vclk_vic_supported_freq(struct meson_drm *priv,
+			      unsigned long long phy_freq,
+			      unsigned long long vclk_freq)
 {
 	int i;
 
-	DRM_DEBUG_DRIVER("phy_freq = %d vclk_freq = %d\n",
+	DRM_DEBUG_DRIVER("phy_freq = %lluHz vclk_freq = %lluHz\n",
 			 phy_freq, vclk_freq);
 
 	/* Check against soc revision/package limits */
@@ -785,19 +790,19 @@ meson_vclk_vic_supported_freq(struct meson_drm *priv, unsigned int phy_freq,
 	}
 
 	for (i = 0 ; params[i].pixel_freq ; ++i) {
-		DRM_DEBUG_DRIVER("i = %d pixel_freq = %d alt = %d\n",
+		DRM_DEBUG_DRIVER("i = %d pixel_freq = %lluHz alt = %lluHz\n",
 				 i, params[i].pixel_freq,
-				 FREQ_1000_1001(params[i].pixel_freq));
-		DRM_DEBUG_DRIVER("i = %d phy_freq = %d alt = %d\n",
+				 PIXEL_FREQ_1000_1001(params[i].pixel_freq));
+		DRM_DEBUG_DRIVER("i = %d phy_freq = %lluHz alt = %lluHz\n",
 				 i, params[i].phy_freq,
-				 FREQ_1000_1001(params[i].phy_freq/1000)*1000);
+				 PHY_FREQ_1000_1001(params[i].phy_freq));
 		/* Match strict frequency */
 		if (phy_freq == params[i].phy_freq &&
 		    vclk_freq == params[i].vclk_freq)
 			return MODE_OK;
 		/* Match 1000/1001 variant */
-		if (phy_freq == (FREQ_1000_1001(params[i].phy_freq/1000)*1000) &&
-		    vclk_freq == FREQ_1000_1001(params[i].vclk_freq))
+		if (phy_freq == PHY_FREQ_1000_1001(params[i].phy_freq) &&
+		    vclk_freq == PIXEL_FREQ_1000_1001(params[i].vclk_freq))
 			return MODE_OK;
 	}
 
@@ -805,8 +810,9 @@ meson_vclk_vic_supported_freq(struct meson_drm *priv, unsigned int phy_freq,
 }
 EXPORT_SYMBOL_GPL(meson_vclk_vic_supported_freq);
 
-static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq,
-			   unsigned int od1, unsigned int od2, unsigned int od3,
+static void meson_vclk_set(struct meson_drm *priv,
+			   unsigned long long pll_base_freq, unsigned int od1,
+			   unsigned int od2, unsigned int od3,
 			   unsigned int vid_pll_div, unsigned int vclk_div,
 			   unsigned int hdmi_tx_div, unsigned int venc_div,
 			   bool hdmi_use_enci, bool vic_alternate_clock)
@@ -826,15 +832,15 @@ static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq,
 		meson_hdmi_pll_generic_set(priv, pll_base_freq);
 	} else if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_GXBB)) {
 		switch (pll_base_freq) {
-		case 2970000:
+		case 2970000000:
 			m = 0x3d;
 			frac = vic_alternate_clock ? 0xd02 : 0xe00;
 			break;
-		case 4320000:
+		case 4320000000:
 			m = vic_alternate_clock ? 0x59 : 0x5a;
 			frac = vic_alternate_clock ? 0xe8f : 0;
 			break;
-		case 5940000:
+		case 5940000000:
 			m = 0x7b;
 			frac = vic_alternate_clock ? 0xa05 : 0xc00;
 			break;
@@ -844,15 +850,15 @@ static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq,
 	} else if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_GXM) ||
 		   meson_vpu_is_compatible(priv, VPU_COMPATIBLE_GXL)) {
 		switch (pll_base_freq) {
-		case 2970000:
+		case 2970000000:
 			m = 0x7b;
 			frac = vic_alternate_clock ? 0x281 : 0x300;
 			break;
-		case 4320000:
+		case 4320000000:
 			m = vic_alternate_clock ? 0xb3 : 0xb4;
 			frac = vic_alternate_clock ? 0x347 : 0;
 			break;
-		case 5940000:
+		case 5940000000:
 			m = 0xf7;
 			frac = vic_alternate_clock ? 0x102 : 0x200;
 			break;
@@ -861,15 +867,15 @@ static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq,
 		meson_hdmi_pll_set_params(priv, m, frac, od1, od2, od3);
 	} else if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_G12A)) {
 		switch (pll_base_freq) {
-		case 2970000:
+		case 2970000000:
 			m = 0x7b;
 			frac = vic_alternate_clock ? 0x140b4 : 0x18000;
 			break;
-		case 4320000:
+		case 4320000000:
 			m = vic_alternate_clock ? 0xb3 : 0xb4;
 			frac = vic_alternate_clock ? 0x1a3ee : 0;
 			break;
-		case 5940000:
+		case 5940000000:
 			m = 0xf7;
 			frac = vic_alternate_clock ? 0x8148 : 0x10000;
 			break;
@@ -1025,14 +1031,14 @@ static void meson_vclk_set(struct meson_drm *priv, unsigned int pll_base_freq,
 }
 
 void meson_vclk_setup(struct meson_drm *priv, unsigned int target,
-		      unsigned int phy_freq, unsigned int vclk_freq,
-		      unsigned int venc_freq, unsigned int dac_freq,
+		      unsigned long long phy_freq, unsigned long long vclk_freq,
+		      unsigned long long venc_freq, unsigned long long dac_freq,
 		      bool hdmi_use_enci)
 {
 	bool vic_alternate_clock = false;
-	unsigned int freq;
-	unsigned int hdmi_tx_div;
-	unsigned int venc_div;
+	unsigned long long freq;
+	unsigned long long hdmi_tx_div;
+	unsigned long long venc_div;
 
 	if (target == MESON_VCLK_TARGET_CVBS) {
 		meson_venci_cvbs_clock_config(priv);
@@ -1052,27 +1058,27 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target,
 		return;
 	}
 
-	hdmi_tx_div = vclk_freq / dac_freq;
+	hdmi_tx_div = DIV_ROUND_DOWN_ULL(vclk_freq, dac_freq);
 
 	if (hdmi_tx_div == 0) {
-		pr_err("Fatal Error, invalid HDMI-TX freq %d\n",
+		pr_err("Fatal Error, invalid HDMI-TX freq %lluHz\n",
 		       dac_freq);
 		return;
 	}
 
-	venc_div = vclk_freq / venc_freq;
+	venc_div = DIV_ROUND_DOWN_ULL(vclk_freq, venc_freq);
 
 	if (venc_div == 0) {
-		pr_err("Fatal Error, invalid HDMI venc freq %d\n",
+		pr_err("Fatal Error, invalid HDMI venc freq %lluHz\n",
 		       venc_freq);
 		return;
 	}
 
 	for (freq = 0 ; params[freq].pixel_freq ; ++freq) {
 		if ((phy_freq == params[freq].phy_freq ||
-		     phy_freq == FREQ_1000_1001(params[freq].phy_freq/1000)*1000) &&
+		     phy_freq == PHY_FREQ_1000_1001(params[freq].phy_freq)) &&
 		    (vclk_freq == params[freq].vclk_freq ||
-		     vclk_freq == FREQ_1000_1001(params[freq].vclk_freq))) {
+		     vclk_freq == PIXEL_FREQ_1000_1001(params[freq].vclk_freq))) {
 			if (vclk_freq != params[freq].vclk_freq)
 				vic_alternate_clock = true;
 			else
@@ -1098,7 +1104,8 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target,
 	}
 
 	if (!params[freq].pixel_freq) {
-		pr_err("Fatal Error, invalid HDMI vclk freq %d\n", vclk_freq);
+		pr_err("Fatal Error, invalid HDMI vclk freq %lluHz\n",
+		       vclk_freq);
 		return;
 	}
 
diff --git a/drivers/gpu/drm/meson/meson_vclk.h b/drivers/gpu/drm/meson/meson_vclk.h
index 60617aaf18dd..7ac55744e574 100644
--- a/drivers/gpu/drm/meson/meson_vclk.h
+++ b/drivers/gpu/drm/meson/meson_vclk.h
@@ -20,17 +20,18 @@ enum {
 };
 
 /* 27MHz is the CVBS Pixel Clock */
-#define MESON_VCLK_CVBS			27000
+#define MESON_VCLK_CVBS			(27 * 1000 * 1000)
 
 enum drm_mode_status
-meson_vclk_dmt_supported_freq(struct meson_drm *priv, unsigned int freq);
+meson_vclk_dmt_supported_freq(struct meson_drm *priv, unsigned long long freq);
 enum drm_mode_status
-meson_vclk_vic_supported_freq(struct meson_drm *priv, unsigned int phy_freq,
-			      unsigned int vclk_freq);
+meson_vclk_vic_supported_freq(struct meson_drm *priv,
+			      unsigned long long phy_freq,
+			      unsigned long long vclk_freq);
 
 void meson_vclk_setup(struct meson_drm *priv, unsigned int target,
-		      unsigned int phy_freq, unsigned int vclk_freq,
-		      unsigned int venc_freq, unsigned int dac_freq,
+		      unsigned long long phy_freq, unsigned long long vclk_freq,
+		      unsigned long long venc_freq, unsigned long long dac_freq,
 		      bool hdmi_use_enci);
 
 #endif /* __MESON_VCLK_H */
diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index fb71658c3117..6067d08aeee3 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -223,7 +223,7 @@ void mgag200_set_mode_regs(struct mga_device *mdev, const struct drm_display_mod
 	vsyncstr = mode->crtc_vsync_start - 1;
 	vsyncend = mode->crtc_vsync_end - 1;
 	vtotal = mode->crtc_vtotal - 2;
-	vblkstr = mode->crtc_vblank_start;
+	vblkstr = mode->crtc_vblank_start - 1;
 	vblkend = vtotal + 1;
 
 	linecomp = vdispend;
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 974bc7c0ea76..7f127e2ae442 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -104,6 +104,7 @@ config DRM_MSM_DPU
 config DRM_MSM_DP
 	bool "Enable DisplayPort support in MSM DRM driver"
 	depends on DRM_MSM
+	select DRM_DISPLAY_HDMI_AUDIO_HELPER
 	select RATIONAL
 	default y
 	help
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 5df20cbeafb8..7a2ada6e2d74 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -48,7 +48,6 @@ msm-display-$(CONFIG_DRM_MSM_MDP4) += \
 	disp/mdp4/mdp4_dsi_encoder.o \
 	disp/mdp4/mdp4_dtv_encoder.o \
 	disp/mdp4/mdp4_lcdc_encoder.o \
-	disp/mdp4/mdp4_lvds_connector.o \
 	disp/mdp4/mdp4_lvds_pll.o \
 	disp/mdp4/mdp4_irq.o \
 	disp/mdp4/mdp4_kms.o \
diff --git a/drivers/gpu/drm/msm/adreno/a2xx_catalog.c b/drivers/gpu/drm/msm/adreno/a2xx_catalog.c
index 9ddb7b31fd98..5ddd015f930d 100644
--- a/drivers/gpu/drm/msm/adreno/a2xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a2xx_catalog.c
@@ -45,8 +45,3 @@ static const struct adreno_info a2xx_gpus[] = {
 	}
 };
 DECLARE_ADRENO_GPULIST(a2xx);
-
-MODULE_FIRMWARE("qcom/leia_pfp_470.fw");
-MODULE_FIRMWARE("qcom/leia_pm4_470.fw");
-MODULE_FIRMWARE("qcom/yamato_pfp.fw");
-MODULE_FIRMWARE("qcom/yamato_pm4.fw");
diff --git a/drivers/gpu/drm/msm/adreno/a3xx_catalog.c b/drivers/gpu/drm/msm/adreno/a3xx_catalog.c
index 2eb6c3e93748..1498e6532f62 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a3xx_catalog.c
@@ -85,8 +85,3 @@ static const struct adreno_info a3xx_gpus[] = {
 	}
 };
 DECLARE_ADRENO_GPULIST(a3xx);
-
-MODULE_FIRMWARE("qcom/a300_pm4.fw");
-MODULE_FIRMWARE("qcom/a300_pfp.fw");
-MODULE_FIRMWARE("qcom/a330_pm4.fw");
-MODULE_FIRMWARE("qcom/a330_pfp.fw");
diff --git a/drivers/gpu/drm/msm/adreno/a4xx_catalog.c b/drivers/gpu/drm/msm/adreno/a4xx_catalog.c
index 93519f807f87..09f9f228b75e 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a4xx_catalog.c
@@ -45,6 +45,3 @@ static const struct adreno_info a4xx_gpus[] = {
 	}
 };
 DECLARE_ADRENO_GPULIST(a4xx);
-
-MODULE_FIRMWARE("qcom/a420_pm4.fw");
-MODULE_FIRMWARE("qcom/a420_pfp.fw");
diff --git a/drivers/gpu/drm/msm/adreno/a5xx_catalog.c b/drivers/gpu/drm/msm/adreno/a5xx_catalog.c
index 633f31539162..b48a636d8237 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_catalog.c
@@ -150,12 +150,3 @@ static const struct adreno_info a5xx_gpus[] = {
 	}
 };
 DECLARE_ADRENO_GPULIST(a5xx);
-
-MODULE_FIRMWARE("qcom/a530_pm4.fw");
-MODULE_FIRMWARE("qcom/a530_pfp.fw");
-MODULE_FIRMWARE("qcom/a530v3_gpmu.fw2");
-MODULE_FIRMWARE("qcom/a530_zap.mdt");
-MODULE_FIRMWARE("qcom/a530_zap.b00");
-MODULE_FIRMWARE("qcom/a530_zap.b01");
-MODULE_FIRMWARE("qcom/a530_zap.b02");
-MODULE_FIRMWARE("qcom/a540_gpmu.fw2");
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index 53e2ff4406d8..70f7ad806c34 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -681,6 +681,7 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_SQE] = "a630_sqe.fw",
 		},
 		.gmem = (SZ_128K + SZ_4K),
+		.quirks = ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.init = a6xx_gpu_init,
 		.zapfw = "a610_zap.mdt",
@@ -713,6 +714,7 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a630_gmu.bin",
 		},
 		.gmem = SZ_512K,
+		.quirks = ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.init = a6xx_gpu_init,
 		.zapfw = "a615_zap.mdt",
@@ -743,7 +745,8 @@ static const struct adreno_info a6xx_gpus[] = {
 		},
 		.gmem = SZ_512K,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
-		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT,
+		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
+			  ADRENO_QUIRK_4GB_VA,
 		.init = a6xx_gpu_init,
 		.zapfw = "a615_zap.mbn",
 		.a6xx = &(const struct a6xx_info) {
@@ -769,7 +772,8 @@ static const struct adreno_info a6xx_gpus[] = {
 		},
 		.gmem = SZ_512K,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
-		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT,
+		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
+			  ADRENO_QUIRK_4GB_VA,
 		.init = a6xx_gpu_init,
 		.a6xx = &(const struct a6xx_info) {
 			.protect = &a630_protect,
@@ -791,6 +795,7 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a619_gmu.bin",
 		},
 		.gmem = SZ_512K,
+		.quirks = ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.init = a6xx_gpu_init,
 		.zapfw = "a615_zap.mdt",
@@ -815,6 +820,7 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a619_gmu.bin",
 		},
 		.gmem = SZ_512K,
+		.quirks = ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.init = a6xx_gpu_init,
 		.zapfw = "a615_zap.mdt",
@@ -838,8 +844,9 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a619_gmu.bin",
 		},
 		.gmem = SZ_512K,
+		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
+			  ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
-		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT,
 		.init = a6xx_gpu_init,
 		.zapfw = "a615_zap.mdt",
 		.a6xx = &(const struct a6xx_info) {
@@ -874,7 +881,6 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020200,
 			.prim_fifo_threshold = 0x00010000,
 		},
-		.address_space_size = SZ_16G,
 		.speedbins = ADRENO_SPEEDBINS(
 			{ 0, 0 },
 			{ 137, 1 },
@@ -907,7 +913,6 @@ static const struct adreno_info a6xx_gpus[] = {
 				{ /* sentinel */ },
 			},
 		},
-		.address_space_size = SZ_16G,
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(
 			0x06030001,
@@ -920,8 +925,9 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a630_gmu.bin",
 		},
 		.gmem = SZ_1M,
+		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
+			  ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
-		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT,
 		.init = a6xx_gpu_init,
 		.zapfw = "a630_zap.mdt",
 		.a6xx = &(const struct a6xx_info) {
@@ -939,8 +945,9 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a640_gmu.bin",
 		},
 		.gmem = SZ_1M,
+		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
+			  ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
-		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT,
 		.init = a6xx_gpu_init,
 		.zapfw = "a640_zap.mdt",
 		.a6xx = &(const struct a6xx_info) {
@@ -973,7 +980,6 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020202,
 			.prim_fifo_threshold = 0x00300200,
 		},
-		.address_space_size = SZ_16G,
 		.speedbins = ADRENO_SPEEDBINS(
 			{ 0, 0 },
 			{ 1, 1 },
@@ -1000,7 +1006,6 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020000,
 			.prim_fifo_threshold = 0x00300200,
 		},
-		.address_space_size = SZ_16G,
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(0x06060300),
 		.family = ADRENO_6XX_GEN4,
@@ -1019,7 +1024,6 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020200,
 			.prim_fifo_threshold = 0x00300200,
 		},
-		.address_space_size = SZ_16G,
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(0x06030500),
 		.family = ADRENO_6XX_GEN4,
@@ -1039,7 +1043,6 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020202,
 			.prim_fifo_threshold = 0x00200200,
 		},
-		.address_space_size = SZ_16G,
 		.speedbins = ADRENO_SPEEDBINS(
 			{ 0,   0 },
 			{ 117, 0 },
@@ -1056,8 +1059,9 @@ static const struct adreno_info a6xx_gpus[] = {
 			[ADRENO_FW_GMU] = "a640_gmu.bin",
 		},
 		.gmem = SZ_2M,
+		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
+			  ADRENO_QUIRK_4GB_VA,
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
-		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT,
 		.init = a6xx_gpu_init,
 		.zapfw = "a640_zap.mdt",
 		.a6xx = &(const struct a6xx_info) {
@@ -1085,22 +1089,10 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020200,
 			.prim_fifo_threshold = 0x00800200,
 		},
-		.address_space_size = SZ_16G,
 	}
 };
 DECLARE_ADRENO_GPULIST(a6xx);
 
-MODULE_FIRMWARE("qcom/a615_zap.mbn");
-MODULE_FIRMWARE("qcom/a619_gmu.bin");
-MODULE_FIRMWARE("qcom/a630_sqe.fw");
-MODULE_FIRMWARE("qcom/a630_gmu.bin");
-MODULE_FIRMWARE("qcom/a630_zap.mbn");
-MODULE_FIRMWARE("qcom/a640_gmu.bin");
-MODULE_FIRMWARE("qcom/a650_gmu.bin");
-MODULE_FIRMWARE("qcom/a650_sqe.fw");
-MODULE_FIRMWARE("qcom/a660_gmu.bin");
-MODULE_FIRMWARE("qcom/a660_sqe.fw");
-
 static const struct adreno_reglist a702_hwcg[] = {
 	{ REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x22222222 },
 	{ REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220 },
@@ -1395,7 +1387,6 @@ static const struct adreno_info a7xx_gpus[] = {
 			.pwrup_reglist = &a7xx_pwrup_reglist,
 			.gmu_cgc_mode = 0x00020000,
 		},
-		.address_space_size = SZ_16G,
 		.preempt_record_size = 2860 * SZ_1K,
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(0x43050a01), /* "C510v2" */
@@ -1429,7 +1420,6 @@ static const struct adreno_info a7xx_gpus[] = {
 				{ /* sentinel */ },
 			},
 		},
-		.address_space_size = SZ_16G,
 		.preempt_record_size = 4192 * SZ_1K,
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(0x43050c01), /* "C512v2" */
@@ -1451,7 +1441,6 @@ static const struct adreno_info a7xx_gpus[] = {
 			.gmu_chipid = 0x7050001,
 			.gmu_cgc_mode = 0x00020202,
 		},
-		.address_space_size = SZ_256G,
 		.preempt_record_size = 4192 * SZ_1K,
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(0x43051401), /* "C520v2" */
@@ -1484,7 +1473,6 @@ static const struct adreno_info a7xx_gpus[] = {
 				{ /* sentinel */ },
 			},
 		},
-		.address_space_size = SZ_16G,
 		.preempt_record_size = 3572 * SZ_1K,
 	}
 };
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index c8711938a5f4..38c0f8ef85c3 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1064,14 +1064,6 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu)
 
 	gmu->hung = false;
 
-	/* Notify AOSS about the ACD state (unimplemented for now => disable it) */
-	if (!IS_ERR(gmu->qmp)) {
-		ret = qmp_send(gmu->qmp, "{class: gpu, res: acd, val: %d}",
-			       0 /* Hardcode ACD to be disabled for now */);
-		if (ret)
-			dev_err(gmu->dev, "failed to send GPU ACD state\n");
-	}
-
 	/* Turn on the resources */
 	pm_runtime_get_sync(gmu->dev);
 
@@ -1671,6 +1663,75 @@ static int a6xx_gmu_pwrlevels_probe(struct a6xx_gmu *gmu)
 	return a6xx_gmu_rpmh_votes_init(gmu);
 }
 
+static int a6xx_gmu_acd_probe(struct a6xx_gmu *gmu)
+{
+	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
+	struct a6xx_hfi_acd_table *cmd = &gmu->acd_table;
+	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
+	struct msm_gpu *gpu = &adreno_gpu->base;
+	int ret, i, cmd_idx = 0;
+	extern bool disable_acd;
+
+	/* Skip ACD probe if requested via module param */
+	if (disable_acd) {
+		DRM_DEV_ERROR(gmu->dev, "Skipping GPU ACD probe\n");
+		return 0;
+	}
+
+	cmd->version = 1;
+	cmd->stride = 1;
+	cmd->enable_by_level = 0;
+
+	/* Skip freq = 0 and parse acd-level for rest of the OPPs */
+	for (i = 1; i < gmu->nr_gpu_freqs; i++) {
+		struct dev_pm_opp *opp;
+		struct device_node *np;
+		unsigned long freq;
+		u32 val;
+
+		freq = gmu->gpu_freqs[i];
+		opp = dev_pm_opp_find_freq_exact(&gpu->pdev->dev, freq, true);
+		np = dev_pm_opp_get_of_node(opp);
+
+		ret = of_property_read_u32(np, "qcom,opp-acd-level", &val);
+		of_node_put(np);
+		dev_pm_opp_put(opp);
+		if (ret == -EINVAL)
+			continue;
+		else if (ret) {
+			DRM_DEV_ERROR(gmu->dev, "Unable to read acd level for freq %lu\n", freq);
+			return ret;
+		}
+
+		cmd->enable_by_level |= BIT(i);
+		cmd->data[cmd_idx++] = val;
+	}
+
+	cmd->num_levels = cmd_idx;
+
+	/* It is a problem if qmp node is unavailable when ACD is required */
+	if (cmd->enable_by_level && IS_ERR_OR_NULL(gmu->qmp)) {
+		DRM_DEV_ERROR(gmu->dev, "Unable to send ACD state to AOSS\n");
+		return -EINVAL;
+	}
+
+	/* Otherwise, nothing to do if qmp is unavailable */
+	if (IS_ERR_OR_NULL(gmu->qmp))
+		return 0;
+
+	/*
+	 * Notify AOSS about the ACD state. AOSS is supposed to assume that ACD is disabled on
+	 * system reset. So it is harmless if we couldn't notify 'OFF' state
+	 */
+	ret = qmp_send(gmu->qmp, "{class: gpu, res: acd, val: %d}", !!cmd->enable_by_level);
+	if (ret && cmd->enable_by_level) {
+		DRM_DEV_ERROR(gmu->dev, "Failed to send ACD state to AOSS\n");
+		return ret;
+	}
+
+	return 0;
+}
+
 static int a6xx_gmu_clocks_probe(struct a6xx_gmu *gmu)
 {
 	int ret = devm_clk_bulk_get_all(gmu->dev, &gmu->clocks);
@@ -1989,10 +2050,11 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 		goto detach_cxpd;
 	}
 
+	/* Other errors are handled during GPU ACD probe */
 	gmu->qmp = qmp_get(gmu->dev);
-	if (IS_ERR(gmu->qmp) && adreno_is_a7xx(adreno_gpu)) {
-		ret = PTR_ERR(gmu->qmp);
-		goto remove_device_link;
+	if (PTR_ERR_OR_ZERO(gmu->qmp) == -EPROBE_DEFER) {
+		ret = -EPROBE_DEFER;
+		goto detach_gxpd;
 	}
 
 	init_completion(&gmu->pd_gate);
@@ -2008,6 +2070,10 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 	/* Get the power levels for the GMU and GPU */
 	a6xx_gmu_pwrlevels_probe(gmu);
 
+	ret = a6xx_gmu_acd_probe(gmu);
+	if (ret)
+		goto detach_gxpd;
+
 	/* Set up the HFI queues */
 	a6xx_hfi_init(gmu);
 
@@ -2018,7 +2084,13 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 
 	return 0;
 
-remove_device_link:
+detach_gxpd:
+	if (!IS_ERR_OR_NULL(gmu->gxpd))
+		dev_pm_domain_detach(gmu->gxpd, false);
+
+	if (!IS_ERR_OR_NULL(gmu->qmp))
+		qmp_put(gmu->qmp);
+
 	device_link_del(link);
 
 detach_cxpd:
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index 0c888b326cfb..b2d4489b4024 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -93,6 +93,7 @@ struct a6xx_gmu {
 	int nr_gpu_freqs;
 	unsigned long gpu_freqs[GMU_MAX_GX_FREQS];
 	u32 gx_arc_votes[GMU_MAX_GX_FREQS];
+	struct a6xx_hfi_acd_table acd_table;
 
 	int nr_gpu_bws;
 	unsigned long gpu_bw_table[GMU_MAX_GX_FREQS];
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 06465bc2d0b4..bf3758f010f4 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -242,10 +242,10 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 				break;
 			fallthrough;
 		case MSM_SUBMIT_CMD_BUF:
-			OUT_PKT7(ring, CP_INDIRECT_BUFFER_PFE, 3);
+			OUT_PKT7(ring, CP_INDIRECT_BUFFER, 3);
 			OUT_RING(ring, lower_32_bits(submit->cmd[i].iova));
 			OUT_RING(ring, upper_32_bits(submit->cmd[i].iova));
-			OUT_RING(ring, submit->cmd[i].size);
+			OUT_RING(ring, A5XX_CP_INDIRECT_BUFFER_2_IB_SIZE(submit->cmd[i].size));
 			ibs++;
 			break;
 		}
@@ -377,10 +377,10 @@ static void a7xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 				break;
 			fallthrough;
 		case MSM_SUBMIT_CMD_BUF:
-			OUT_PKT7(ring, CP_INDIRECT_BUFFER_PFE, 3);
+			OUT_PKT7(ring, CP_INDIRECT_BUFFER, 3);
 			OUT_RING(ring, lower_32_bits(submit->cmd[i].iova));
 			OUT_RING(ring, upper_32_bits(submit->cmd[i].iova));
-			OUT_RING(ring, submit->cmd[i].size);
+			OUT_RING(ring, A5XX_CP_INDIRECT_BUFFER_2_IB_SIZE(submit->cmd[i].size));
 			ibs++;
 			break;
 		}
@@ -655,7 +655,6 @@ static void a6xx_calc_ubwc_config(struct adreno_gpu *gpu)
 	if (adreno_is_7c3(gpu)) {
 		gpu->ubwc_config.highest_bank_bit = 14;
 		gpu->ubwc_config.amsbc = 1;
-		gpu->ubwc_config.rgb565_predicator = 1;
 		gpu->ubwc_config.uavflagprd_inv = 2;
 		gpu->ubwc_config.macrotile_mode = 1;
 	}
@@ -2268,7 +2267,7 @@ a6xx_create_private_address_space(struct msm_gpu *gpu)
 		return ERR_CAST(mmu);
 
 	return msm_gem_address_space_create(mmu,
-		"gpu", 0x100000000ULL,
+		"gpu", ADRENO_VM_START,
 		adreno_private_address_space_size(gpu));
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
index 0989aee3dd2c..8e69b1e84657 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
@@ -100,16 +100,14 @@ static int a6xx_hfi_queue_write(struct a6xx_gmu *gmu,
 	return 0;
 }
 
-static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum,
-		u32 *payload, u32 payload_size)
+static int a6xx_hfi_wait_for_msg_interrupt(struct a6xx_gmu *gmu, u32 id, u32 seqnum)
 {
-	struct a6xx_hfi_queue *queue = &gmu->queues[HFI_RESPONSE_QUEUE];
-	u32 val;
 	int ret;
+	u32 val;
 
 	/* Wait for a response */
 	ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_GMU2HOST_INTR_INFO, val,
-		val & A6XX_GMU_GMU2HOST_INTR_INFO_MSGQ, 100, 5000);
+		val & A6XX_GMU_GMU2HOST_INTR_INFO_MSGQ, 100, 1000000);
 
 	if (ret) {
 		DRM_DEV_ERROR(gmu->dev,
@@ -122,6 +120,19 @@ static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum,
 	gmu_write(gmu, REG_A6XX_GMU_GMU2HOST_INTR_CLR,
 		A6XX_GMU_GMU2HOST_INTR_INFO_MSGQ);
 
+	return 0;
+}
+
+static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum,
+		u32 *payload, u32 payload_size)
+{
+	struct a6xx_hfi_queue *queue = &gmu->queues[HFI_RESPONSE_QUEUE];
+	int ret;
+
+	ret = a6xx_hfi_wait_for_msg_interrupt(gmu, id, seqnum);
+	if (ret)
+		return ret;
+
 	for (;;) {
 		struct a6xx_hfi_msg_response resp;
 
@@ -129,12 +140,18 @@ static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum,
 		ret = a6xx_hfi_queue_read(gmu, queue, (u32 *) &resp,
 			sizeof(resp) >> 2);
 
-		/* If the queue is empty our response never made it */
+		/* If the queue is empty, there may have been previous missed
+		 * responses that preceded the response to our packet. Wait
+		 * further before we give up.
+		 */
 		if (!ret) {
-			DRM_DEV_ERROR(gmu->dev,
-				"The HFI response queue is unexpectedly empty\n");
-
-			return -ENOENT;
+			ret = a6xx_hfi_wait_for_msg_interrupt(gmu, id, seqnum);
+			if (ret) {
+				DRM_DEV_ERROR(gmu->dev,
+					"The HFI response queue is unexpectedly empty\n");
+				return ret;
+			}
+			continue;
 		}
 
 		if (HFI_HEADER_ID(resp.header) == HFI_F2H_MSG_ERROR) {
@@ -748,6 +765,38 @@ send:
 		NULL, 0);
 }
 
+#define HFI_FEATURE_ACD 12
+
+static int a6xx_hfi_enable_acd(struct a6xx_gmu *gmu)
+{
+	struct a6xx_hfi_acd_table *acd_table = &gmu->acd_table;
+	struct a6xx_hfi_msg_feature_ctrl msg = {
+		.feature = HFI_FEATURE_ACD,
+		.enable = 1,
+		.data = 0,
+	};
+	int ret;
+
+	if (!acd_table->enable_by_level)
+		return 0;
+
+	/* Enable ACD feature at GMU */
+	ret = a6xx_hfi_send_msg(gmu, HFI_H2F_FEATURE_CTRL, &msg, sizeof(msg), NULL, 0);
+	if (ret) {
+		DRM_DEV_ERROR(gmu->dev, "Unable to enable ACD (%d)\n", ret);
+		return ret;
+	}
+
+	/* Send ACD table to GMU */
+	ret = a6xx_hfi_send_msg(gmu, HFI_H2F_MSG_ACD, acd_table, sizeof(*acd_table), NULL, 0);
+	if (ret) {
+		DRM_DEV_ERROR(gmu->dev, "Unable to ACD table (%d)\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int a6xx_hfi_send_test(struct a6xx_gmu *gmu)
 {
 	struct a6xx_hfi_msg_test msg = { 0 };
@@ -845,6 +894,10 @@ int a6xx_hfi_start(struct a6xx_gmu *gmu, int boot_state)
 	if (ret)
 		return ret;
 
+	ret = a6xx_hfi_enable_acd(gmu);
+	if (ret)
+		return ret;
+
 	ret = a6xx_hfi_send_core_fw_start(gmu);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h
index 52ba4a07d7b9..653ef720e2da 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h
@@ -151,12 +151,33 @@ struct a6xx_hfi_msg_test {
 	u32 header;
 };
 
+#define HFI_H2F_MSG_ACD 7
+#define MAX_ACD_STRIDE 2
+
+struct a6xx_hfi_acd_table {
+	u32 header;
+	u32 version;
+	u32 enable_by_level;
+	u32 stride;
+	u32 num_levels;
+	u32 data[16 * MAX_ACD_STRIDE];
+};
+
 #define HFI_H2F_MSG_START 10
 
 struct a6xx_hfi_msg_start {
 	u32 header;
 };
 
+#define HFI_H2F_FEATURE_CTRL 11
+
+struct a6xx_hfi_msg_feature_ctrl {
+	u32 header;
+	u32 feature;
+	u32 enable;
+	u32 data;
+};
+
 #define HFI_H2F_MSG_CORE_FW_START 14
 
 struct a6xx_hfi_msg_core_fw_start {
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 236b25c094cd..f5e1490d07c1 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -24,6 +24,10 @@ int enable_preemption = -1;
 MODULE_PARM_DESC(enable_preemption, "Enable preemption (A7xx only) (1=on , 0=disable, -1=auto (default))");
 module_param(enable_preemption, int, 0600);
 
+bool disable_acd;
+MODULE_PARM_DESC(disable_acd, "Forcefully disable GPU ACD");
+module_param_unsafe(disable_acd, bool, 0400);
+
 extern const struct adreno_gpulist a2xx_gpulist;
 extern const struct adreno_gpulist a3xx_gpulist;
 extern const struct adreno_gpulist a4xx_gpulist;
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 26db1f4b5fb9..2348ffb35f7e 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -236,14 +236,27 @@ adreno_iommu_create_address_space(struct msm_gpu *gpu,
 u64 adreno_private_address_space_size(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(&gpu->pdev->dev);
+	const struct io_pgtable_cfg *ttbr1_cfg;
 
 	if (address_space_size)
 		return address_space_size;
 
-	if (adreno_gpu->info->address_space_size)
-		return adreno_gpu->info->address_space_size;
+	if (adreno_gpu->info->quirks & ADRENO_QUIRK_4GB_VA)
+		return SZ_4G;
 
-	return SZ_4G;
+	if (!adreno_smmu || !adreno_smmu->get_ttbr1_cfg)
+		return SZ_4G;
+
+	ttbr1_cfg = adreno_smmu->get_ttbr1_cfg(adreno_smmu->cookie);
+
+	/*
+	 * Userspace VM is actually using TTBR0, but both are the same size,
+	 * with b48 (sign bit) selecting which TTBRn to use.  So if IAS is
+	 * 48, the total (kernel+user) address space size is effectively
+	 * 49 bits.  But what userspace is control of is the lower 48.
+	 */
+	return BIT(ttbr1_cfg->ias) - ADRENO_VM_START;
 }
 
 #define ARM_SMMU_FSR_TF                 BIT(1)
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 92caba3584da..a8f4bf416e64 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -57,6 +57,7 @@ enum adreno_family {
 #define ADRENO_QUIRK_HAS_HW_APRIV		BIT(3)
 #define ADRENO_QUIRK_HAS_CACHED_COHERENT	BIT(4)
 #define ADRENO_QUIRK_PREEMPTION			BIT(5)
+#define ADRENO_QUIRK_4GB_VA			BIT(6)
 
 /* Helper for formating the chip_id in the way that userspace tools like
  * crashdec expect.
@@ -104,7 +105,6 @@ struct adreno_info {
 	union {
 		const struct a6xx_info *a6xx;
 	};
-	u64 address_space_size;
 	/**
 	 * @speedbins: Optional table of fuse to speedbin mappings
 	 *
@@ -578,6 +578,8 @@ static inline int adreno_is_a7xx(struct adreno_gpu *gpu)
 	       adreno_is_a740_family(gpu);
 }
 
+/* Put vm_start above 32b to catch issues with not setting xyz_BASE_HI */
+#define ADRENO_VM_START 0x100000000ULL
 u64 adreno_private_address_space_size(struct msm_gpu *gpu);
 int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx,
 		     uint32_t param, uint64_t *value, uint32_t *len);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h
index 6ac97c378056..ffc4d4257ae5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h
@@ -27,17 +27,16 @@ static const struct dpu_mdp_cfg sm8650_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sm8650_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x1000,
-		.features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = CTL_SM8550_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x1000,
-		.features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = CTL_SM8550_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h
index 1f32807bb5e5..39027a21c6fe 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h
@@ -100,14 +100,12 @@ static const struct dpu_pingpong_cfg msm8937_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_MSM8996_MASK,
 		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_1", .id = PINGPONG_1,
 		.base = 0x70800, .len = 0xd4,
-		.features = PINGPONG_MSM8996_MASK,
 		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13),
@@ -132,7 +130,6 @@ static const struct dpu_intf_cfg msm8937_intf[] = {
 		.prog_fetch_lines_worst_case = 14,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_2", .id = INTF_2,
 		.base = 0x6b000, .len = 0x268,
@@ -141,7 +138,6 @@ static const struct dpu_intf_cfg msm8937_intf[] = {
 		.prog_fetch_lines_worst_case = 14,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
-		.intr_tear_rd_ptr = -1,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h
index 42131959ff22..8d1b43ea1663 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h
@@ -93,7 +93,6 @@ static const struct dpu_pingpong_cfg msm8917_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_MSM8996_MASK,
 		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
@@ -118,7 +117,6 @@ static const struct dpu_intf_cfg msm8917_intf[] = {
 		.prog_fetch_lines_worst_case = 14,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-		.intr_tear_rd_ptr = -1,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h
index 2b4723a5c676..16c12499b24b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h
@@ -100,14 +100,12 @@ static const struct dpu_pingpong_cfg msm8953_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_MSM8996_MASK,
 		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_1", .id = PINGPONG_1,
 		.base = 0x70800, .len = 0xd4,
-		.features = PINGPONG_MSM8996_MASK,
 		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13),
@@ -131,7 +129,6 @@ static const struct dpu_intf_cfg msm8953_intf[] = {
 		.prog_fetch_lines_worst_case = 14,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_1", .id = INTF_1,
 		.base = 0x6a800, .len = 0x268,
@@ -140,7 +137,6 @@ static const struct dpu_intf_cfg msm8953_intf[] = {
 		.prog_fetch_lines_worst_case = 14,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_2", .id = INTF_2,
 		.base = 0x6b000, .len = 0x268,
@@ -149,7 +145,6 @@ static const struct dpu_intf_cfg msm8953_intf[] = {
 		.prog_fetch_lines_worst_case = 14,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
-		.intr_tear_rd_ptr = -1,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h
index 5cf19de71f06..91f514d28ac6 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h
@@ -181,15 +181,15 @@ static const struct dpu_pingpong_cfg msm8996_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_MSM8996_TE2_MASK,
-		.sblk = &msm8996_pp_sblk_te,
+		.features = PINGPONG_MSM8996_MASK,
+		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_1", .id = PINGPONG_1,
 		.base = 0x70800, .len = 0xd4,
-		.features = PINGPONG_MSM8996_TE2_MASK,
-		.sblk = &msm8996_pp_sblk_te,
+		.features = PINGPONG_MSM8996_MASK,
+		.sblk = &msm8996_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13),
 	}, {
@@ -241,7 +241,6 @@ static const struct dpu_intf_cfg msm8996_intf[] = {
 		.prog_fetch_lines_worst_case = 25,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_1", .id = INTF_1,
 		.base = 0x6a800, .len = 0x268,
@@ -250,7 +249,6 @@ static const struct dpu_intf_cfg msm8996_intf[] = {
 		.prog_fetch_lines_worst_case = 25,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_2", .id = INTF_2,
 		.base = 0x6b000, .len = 0x268,
@@ -259,7 +257,6 @@ static const struct dpu_intf_cfg msm8996_intf[] = {
 		.prog_fetch_lines_worst_case = 25,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_3", .id = INTF_3,
 		.base = 0x6b800, .len = 0x268,
@@ -267,7 +264,6 @@ static const struct dpu_intf_cfg msm8996_intf[] = {
 		.prog_fetch_lines_worst_case = 25,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31),
-		.intr_tear_rd_ptr = -1,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 746474679ef5..413cd59dc0c4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -170,15 +170,15 @@ static const struct dpu_pingpong_cfg msm8998_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = PINGPONG_SDM845_MASK,
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_1", .id = PINGPONG_1,
 		.base = 0x70800, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = PINGPONG_SDM845_MASK,
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13),
 	}, {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h
index 4f2f68b07f20..b2eb7ca699e3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h
@@ -141,15 +141,15 @@ static const struct dpu_pingpong_cfg sdm660_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = PINGPONG_SDM845_MASK,
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_1", .id = PINGPONG_1,
 		.base = 0x70800, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = PINGPONG_SDM845_MASK,
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13),
 	}, {
@@ -202,7 +202,6 @@ static const struct dpu_intf_cfg sdm660_intf[] = {
 		.prog_fetch_lines_worst_case = 21,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_1", .id = INTF_1,
 		.base = 0x6a800, .len = 0x280,
@@ -211,7 +210,6 @@ static const struct dpu_intf_cfg sdm660_intf[] = {
 		.prog_fetch_lines_worst_case = 21,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_2", .id = INTF_2,
 		.base = 0x6b000, .len = 0x280,
@@ -220,7 +218,6 @@ static const struct dpu_intf_cfg sdm660_intf[] = {
 		.prog_fetch_lines_worst_case = 21,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
-		.intr_tear_rd_ptr = -1,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h
index c70bef025ac4..85e121ad84a0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h
@@ -115,14 +115,14 @@ static const struct dpu_pingpong_cfg sdm630_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_2", .id = PINGPONG_2,
 		.base = 0x71000, .len = 0xd4,
-		.features = PINGPONG_SDM845_MASK,
+		.features = BIT(DPU_PINGPONG_DITHER),
 		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14),
@@ -147,7 +147,6 @@ static const struct dpu_intf_cfg sdm630_intf[] = {
 		.prog_fetch_lines_worst_case = 21,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25),
-		.intr_tear_rd_ptr = -1,
 	}, {
 		.name = "intf_1", .id = INTF_1,
 		.base = 0x6a800, .len = 0x280,
@@ -156,7 +155,6 @@ static const struct dpu_intf_cfg sdm630_intf[] = {
 		.prog_fetch_lines_worst_case = 21,
 		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-		.intr_tear_rd_ptr = -1,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index ab7b4822ca63..49363d7d5b93 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -194,15 +194,15 @@ static const struct dpu_pingpong_cfg sdm845_pp[] = {
 	{
 		.name = "pingpong_0", .id = PINGPONG_0,
 		.base = 0x70000, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = PINGPONG_SDM845_MASK,
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
 	}, {
 		.name = "pingpong_1", .id = PINGPONG_1,
 		.base = 0x70800, .len = 0xd4,
-		.features = PINGPONG_SDM845_TE2_MASK,
-		.sblk = &sdm845_pp_sblk_te,
+		.features = PINGPONG_SDM845_MASK,
+		.sblk = &sdm845_pp_sblk,
 		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 		.intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13),
 	}, {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
index 979527d98fbc..08d38e1d420c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
@@ -37,17 +37,16 @@ static const struct dpu_mdp_cfg sm8150_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sm8150_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x1000, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x1200, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
@@ -76,7 +75,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -84,7 +83,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_1", .id = SSPP_VIG1,
 		.base = 0x6000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 4,
 		.type = SSPP_TYPE_VIG,
@@ -92,7 +91,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_2", .id = SSPP_VIG2,
 		.base = 0x8000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 8,
 		.type = SSPP_TYPE_VIG,
@@ -100,7 +99,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_3", .id = SSPP_VIG3,
 		.base = 0xa000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 12,
 		.type = SSPP_TYPE_VIG,
@@ -108,7 +107,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -116,7 +115,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
@@ -124,7 +123,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x1f0,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
@@ -132,7 +131,7 @@ static const struct dpu_sspp_cfg sm8150_sspp[] = {
 	}, {
 		.name = "sspp_11", .id = SSPP_DMA3,
 		.base = 0x2a000, .len = 0x1f0,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 13,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index d76b8992a6c1..d6f8b1030c68 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -41,12 +41,12 @@ static const struct dpu_ctl_cfg sc8180x_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x1000, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x1200, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
@@ -75,7 +75,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -83,7 +83,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_1", .id = SSPP_VIG1,
 		.base = 0x6000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 4,
 		.type = SSPP_TYPE_VIG,
@@ -91,7 +91,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_2", .id = SSPP_VIG2,
 		.base = 0x8000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 8,
 		.type = SSPP_TYPE_VIG,
@@ -99,7 +99,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_3", .id = SSPP_VIG3,
 		.base = 0xa000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_1_4,
 		.xin_id = 12,
 		.type = SSPP_TYPE_VIG,
@@ -107,7 +107,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -115,7 +115,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
@@ -123,7 +123,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x1f0,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
@@ -131,7 +131,7 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 	}, {
 		.name = "sspp_11", .id = SSPP_DMA3,
 		.base = 0x2a000, .len = 0x1f0,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 13,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h
index 83db11339b29..71ba48b05656 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h
@@ -38,12 +38,12 @@ static const struct dpu_ctl_cfg sm7150_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x1000, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x1200, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
@@ -72,7 +72,7 @@ static const struct dpu_sspp_cfg sm7150_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_2_4,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -80,7 +80,7 @@ static const struct dpu_sspp_cfg sm7150_sspp[] = {
 	}, {
 		.name = "sspp_1", .id = SSPP_VIG1,
 		.base = 0x6000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_2_4,
 		.xin_id = 4,
 		.type = SSPP_TYPE_VIG,
@@ -88,7 +88,7 @@ static const struct dpu_sspp_cfg sm7150_sspp[] = {
 	}, {
 		.name = "sspp_2", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -96,7 +96,7 @@ static const struct dpu_sspp_cfg sm7150_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
@@ -104,7 +104,7 @@ static const struct dpu_sspp_cfg sm7150_sspp[] = {
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x1f0,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h
index d3d3a34d0b45..fcfb3774f7a1 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h
@@ -69,7 +69,7 @@ static const struct dpu_sspp_cfg sm6125_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f0,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_2_4,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -77,7 +77,7 @@ static const struct dpu_sspp_cfg sm6125_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -85,7 +85,7 @@ static const struct dpu_sspp_cfg sm6125_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x1f0,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
index 47e01c3c242f..a86fdb33ebdd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
@@ -35,17 +35,16 @@ static const struct dpu_mdp_cfg sm8250_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sm8250_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x1000, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x1200, .len = 0x1e0,
-		.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = BIT(DPU_CTL_ACTIVE_CFG),
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index 040c94c0bb66..842fcc5887fe 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -51,7 +51,7 @@ static const struct dpu_sspp_cfg sc7180_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f8,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -59,7 +59,7 @@ static const struct dpu_sspp_cfg sc7180_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f8,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -67,7 +67,7 @@ static const struct dpu_sspp_cfg sc7180_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x1f8,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
@@ -75,7 +75,7 @@ static const struct dpu_sspp_cfg sc7180_sspp[] = {
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x1f8,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
index 43f64a005f5a..c5fd89dd7c89 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
@@ -38,7 +38,7 @@ static const struct dpu_sspp_cfg sm6115_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f8,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -46,7 +46,7 @@ static const struct dpu_sspp_cfg sm6115_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f8,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h
index 397278ba999b..a234bb289d24 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h
@@ -59,7 +59,7 @@ static const struct dpu_sspp_cfg sm6350_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f8,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -67,7 +67,7 @@ static const struct dpu_sspp_cfg sm6350_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f8,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -75,7 +75,7 @@ static const struct dpu_sspp_cfg sm6350_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x1f8,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
@@ -83,7 +83,7 @@ static const struct dpu_sspp_cfg sm6350_sspp[] = {
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x1f8,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
index 3cbb2fe8aba2..53f3be28f6f6 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
@@ -46,7 +46,7 @@ static const struct dpu_sspp_cfg qcm2290_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f8,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
index a06c8634d2d7..3a3bc8e429be 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
@@ -39,7 +39,7 @@ static const struct dpu_sspp_cfg sm6375_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x1f8,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -47,7 +47,7 @@ static const struct dpu_sspp_cfg sm6375_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x1f8,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_NO_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
index 0c860e804cab..90e86063a372 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
@@ -35,17 +35,16 @@ static const struct dpu_mdp_cfg sm8350_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sm8350_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x1e8,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x1e8,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
index fcee1c3665f8..139f11321fea 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
@@ -35,17 +35,16 @@ static const struct dpu_mdp_cfg sc8280xp_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sc8280xp_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x204,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x204,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
@@ -74,7 +73,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x2ac,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
@@ -82,7 +81,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_1", .id = SSPP_VIG1,
 		.base = 0x6000, .len = 0x2ac,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 4,
 		.type = SSPP_TYPE_VIG,
@@ -90,7 +89,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_2", .id = SSPP_VIG2,
 		.base = 0x8000, .len = 0x2ac,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 8,
 		.type = SSPP_TYPE_VIG,
@@ -98,7 +97,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_3", .id = SSPP_VIG3,
 		.base = 0xa000, .len = 0x2ac,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_0,
 		.xin_id = 12,
 		.type = SSPP_TYPE_VIG,
@@ -106,7 +105,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x2ac,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
@@ -114,7 +113,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x2ac,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
@@ -122,7 +121,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x2ac,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
@@ -130,7 +129,7 @@ static const struct dpu_sspp_cfg sc8280xp_sspp[] = {
 	}, {
 		.name = "sspp_11", .id = SSPP_DMA3,
 		.base = 0x2a000, .len = 0x2ac,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 13,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 19b2ee8bbd5f..461294143a90 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -36,17 +36,16 @@ static const struct dpu_mdp_cfg sm8450_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sm8450_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x204,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x204,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h
index 4d96ce71746f..c248b3b55c41 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h
@@ -35,17 +35,16 @@ static const struct dpu_mdp_cfg sa8775p_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sa8775p_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x204,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x204,
-		.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
+		.features = CTL_SC7280_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
index 24f988465bf6..59c7fdf28e89 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
@@ -27,17 +27,16 @@ static const struct dpu_mdp_cfg sm8550_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sm8550_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x290,
-		.features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = CTL_SM8550_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x290,
-		.features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = CTL_SM8550_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
@@ -66,70 +65,70 @@ static const struct dpu_sspp_cfg sm8550_sspp[] = {
 	{
 		.name = "sspp_0", .id = SSPP_VIG0,
 		.base = 0x4000, .len = 0x344,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_2,
 		.xin_id = 0,
 		.type = SSPP_TYPE_VIG,
 	}, {
 		.name = "sspp_1", .id = SSPP_VIG1,
 		.base = 0x6000, .len = 0x344,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_2,
 		.xin_id = 4,
 		.type = SSPP_TYPE_VIG,
 	}, {
 		.name = "sspp_2", .id = SSPP_VIG2,
 		.base = 0x8000, .len = 0x344,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_2,
 		.xin_id = 8,
 		.type = SSPP_TYPE_VIG,
 	}, {
 		.name = "sspp_3", .id = SSPP_VIG3,
 		.base = 0xa000, .len = 0x344,
-		.features = VIG_SDM845_MASK,
+		.features = VIG_SDM845_MASK_SDMA,
 		.sblk = &dpu_vig_sblk_qseed3_3_2,
 		.xin_id = 12,
 		.type = SSPP_TYPE_VIG,
 	}, {
 		.name = "sspp_8", .id = SSPP_DMA0,
 		.base = 0x24000, .len = 0x344,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 1,
 		.type = SSPP_TYPE_DMA,
 	}, {
 		.name = "sspp_9", .id = SSPP_DMA1,
 		.base = 0x26000, .len = 0x344,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 5,
 		.type = SSPP_TYPE_DMA,
 	}, {
 		.name = "sspp_10", .id = SSPP_DMA2,
 		.base = 0x28000, .len = 0x344,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 9,
 		.type = SSPP_TYPE_DMA,
 	}, {
 		.name = "sspp_11", .id = SSPP_DMA3,
 		.base = 0x2a000, .len = 0x344,
-		.features = DMA_SDM845_MASK,
+		.features = DMA_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 13,
 		.type = SSPP_TYPE_DMA,
 	}, {
 		.name = "sspp_12", .id = SSPP_DMA4,
 		.base = 0x2c000, .len = 0x344,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 14,
 		.type = SSPP_TYPE_DMA,
 	}, {
 		.name = "sspp_13", .id = SSPP_DMA5,
 		.base = 0x2e000, .len = 0x344,
-		.features = DMA_CURSOR_SDM845_MASK,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
 		.sblk = &dpu_dma_sblk,
 		.xin_id = 15,
 		.type = SSPP_TYPE_DMA,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h
new file mode 100644
index 000000000000..5667d055fbd1
--- /dev/null
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h
@@ -0,0 +1,433 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _DPU_9_1_SAR2130P_H
+#define _DPU_9_1_SAR2130P_H
+
+static const struct dpu_caps sar2130p_dpu_caps = {
+	.max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
+	.max_mixer_blendstages = 0xb,
+	.has_src_split = true,
+	.has_dim_layer = true,
+	.has_idle_pc = true,
+	.has_3d_merge = true,
+	.max_linewidth = 5120,
+	.pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE,
+};
+
+static const struct dpu_mdp_cfg sar2130p_mdp = {
+	.name = "top_0",
+	.base = 0, .len = 0x494,
+	.features = BIT(DPU_MDP_PERIPH_0_REMOVED),
+	.clk_ctrls = {
+		[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 },
+	},
+};
+
+static const struct dpu_ctl_cfg sar2130p_ctl[] = {
+	{
+		.name = "ctl_0", .id = CTL_0,
+		.base = 0x15000, .len = 0x290,
+		.features = CTL_SM8550_MASK,
+		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
+	}, {
+		.name = "ctl_1", .id = CTL_1,
+		.base = 0x16000, .len = 0x290,
+		.features = CTL_SM8550_MASK,
+		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
+	}, {
+		.name = "ctl_2", .id = CTL_2,
+		.base = 0x17000, .len = 0x290,
+		.features = CTL_SM8550_MASK,
+		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11),
+	}, {
+		.name = "ctl_3", .id = CTL_3,
+		.base = 0x18000, .len = 0x290,
+		.features = CTL_SM8550_MASK,
+		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12),
+	}, {
+		.name = "ctl_4", .id = CTL_4,
+		.base = 0x19000, .len = 0x290,
+		.features = CTL_SM8550_MASK,
+		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13),
+	}, {
+		.name = "ctl_5", .id = CTL_5,
+		.base = 0x1a000, .len = 0x290,
+		.features = CTL_SM8550_MASK,
+		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23),
+	},
+};
+
+static const struct dpu_sspp_cfg sar2130p_sspp[] = {
+	{
+		.name = "sspp_0", .id = SSPP_VIG0,
+		.base = 0x4000, .len = 0x344,
+		.features = VIG_SDM845_MASK_SDMA,
+		.sblk = &dpu_vig_sblk_qseed3_3_2,
+		.xin_id = 0,
+		.type = SSPP_TYPE_VIG,
+	}, {
+		.name = "sspp_1", .id = SSPP_VIG1,
+		.base = 0x6000, .len = 0x344,
+		.features = VIG_SDM845_MASK_SDMA,
+		.sblk = &dpu_vig_sblk_qseed3_3_2,
+		.xin_id = 4,
+		.type = SSPP_TYPE_VIG,
+	}, {
+		.name = "sspp_2", .id = SSPP_VIG2,
+		.base = 0x8000, .len = 0x344,
+		.features = VIG_SDM845_MASK_SDMA,
+		.sblk = &dpu_vig_sblk_qseed3_3_2,
+		.xin_id = 8,
+		.type = SSPP_TYPE_VIG,
+	}, {
+		.name = "sspp_3", .id = SSPP_VIG3,
+		.base = 0xa000, .len = 0x344,
+		.features = VIG_SDM845_MASK_SDMA,
+		.sblk = &dpu_vig_sblk_qseed3_3_2,
+		.xin_id = 12,
+		.type = SSPP_TYPE_VIG,
+	}, {
+		.name = "sspp_8", .id = SSPP_DMA0,
+		.base = 0x24000, .len = 0x344,
+		.features = DMA_SDM845_MASK_SDMA,
+		.sblk = &dpu_dma_sblk,
+		.xin_id = 1,
+		.type = SSPP_TYPE_DMA,
+	}, {
+		.name = "sspp_9", .id = SSPP_DMA1,
+		.base = 0x26000, .len = 0x344,
+		.features = DMA_SDM845_MASK_SDMA,
+		.sblk = &dpu_dma_sblk,
+		.xin_id = 5,
+		.type = SSPP_TYPE_DMA,
+	}, {
+		.name = "sspp_10", .id = SSPP_DMA2,
+		.base = 0x28000, .len = 0x344,
+		.features = DMA_SDM845_MASK_SDMA,
+		.sblk = &dpu_dma_sblk,
+		.xin_id = 9,
+		.type = SSPP_TYPE_DMA,
+	}, {
+		.name = "sspp_11", .id = SSPP_DMA3,
+		.base = 0x2a000, .len = 0x344,
+		.features = DMA_SDM845_MASK_SDMA,
+		.sblk = &dpu_dma_sblk,
+		.xin_id = 13,
+		.type = SSPP_TYPE_DMA,
+	}, {
+		.name = "sspp_12", .id = SSPP_DMA4,
+		.base = 0x2c000, .len = 0x344,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
+		.sblk = &dpu_dma_sblk,
+		.xin_id = 14,
+		.type = SSPP_TYPE_DMA,
+	}, {
+		.name = "sspp_13", .id = SSPP_DMA5,
+		.base = 0x2e000, .len = 0x344,
+		.features = DMA_CURSOR_SDM845_MASK_SDMA,
+		.sblk = &dpu_dma_sblk,
+		.xin_id = 15,
+		.type = SSPP_TYPE_DMA,
+	},
+};
+
+static const struct dpu_lm_cfg sar2130p_lm[] = {
+	{
+		.name = "lm_0", .id = LM_0,
+		.base = 0x44000, .len = 0x320,
+		.features = MIXER_SDM845_MASK,
+		.sblk = &sdm845_lm_sblk,
+		.lm_pair = LM_1,
+		.pingpong = PINGPONG_0,
+		.dspp = DSPP_0,
+	}, {
+		.name = "lm_1", .id = LM_1,
+		.base = 0x45000, .len = 0x320,
+		.features = MIXER_SDM845_MASK,
+		.sblk = &sdm845_lm_sblk,
+		.lm_pair = LM_0,
+		.pingpong = PINGPONG_1,
+		.dspp = DSPP_1,
+	}, {
+		.name = "lm_2", .id = LM_2,
+		.base = 0x46000, .len = 0x320,
+		.features = MIXER_SDM845_MASK,
+		.sblk = &sdm845_lm_sblk,
+		.lm_pair = LM_3,
+		.pingpong = PINGPONG_2,
+		.dspp = DSPP_2,
+	}, {
+		.name = "lm_3", .id = LM_3,
+		.base = 0x47000, .len = 0x320,
+		.features = MIXER_SDM845_MASK,
+		.sblk = &sdm845_lm_sblk,
+		.lm_pair = LM_2,
+		.pingpong = PINGPONG_3,
+		.dspp = DSPP_3,
+	}, {
+		.name = "lm_4", .id = LM_4,
+		.base = 0x48000, .len = 0x320,
+		.features = MIXER_SDM845_MASK,
+		.sblk = &sdm845_lm_sblk,
+		.lm_pair = LM_5,
+		.pingpong = PINGPONG_4,
+	}, {
+		.name = "lm_5", .id = LM_5,
+		.base = 0x49000, .len = 0x320,
+		.features = MIXER_SDM845_MASK,
+		.sblk = &sdm845_lm_sblk,
+		.lm_pair = LM_4,
+		.pingpong = PINGPONG_5,
+	},
+};
+
+static const struct dpu_dspp_cfg sar2130p_dspp[] = {
+	{
+		.name = "dspp_0", .id = DSPP_0,
+		.base = 0x54000, .len = 0x1800,
+		.features = DSPP_SC7180_MASK,
+		.sblk = &sdm845_dspp_sblk,
+	}, {
+		.name = "dspp_1", .id = DSPP_1,
+		.base = 0x56000, .len = 0x1800,
+		.features = DSPP_SC7180_MASK,
+		.sblk = &sdm845_dspp_sblk,
+	}, {
+		.name = "dspp_2", .id = DSPP_2,
+		.base = 0x58000, .len = 0x1800,
+		.features = DSPP_SC7180_MASK,
+		.sblk = &sdm845_dspp_sblk,
+	}, {
+		.name = "dspp_3", .id = DSPP_3,
+		.base = 0x5a000, .len = 0x1800,
+		.features = DSPP_SC7180_MASK,
+		.sblk = &sdm845_dspp_sblk,
+	},
+};
+static const struct dpu_pingpong_cfg sar2130p_pp[] = {
+	{
+		.name = "pingpong_0", .id = PINGPONG_0,
+		.base = 0x69000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_0,
+		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
+	}, {
+		.name = "pingpong_1", .id = PINGPONG_1,
+		.base = 0x6a000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_0,
+		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
+	}, {
+		.name = "pingpong_2", .id = PINGPONG_2,
+		.base = 0x6b000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_1,
+		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
+	}, {
+		.name = "pingpong_3", .id = PINGPONG_3,
+		.base = 0x6c000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_1,
+		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
+	}, {
+		.name = "pingpong_4", .id = PINGPONG_4,
+		.base = 0x6d000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_2,
+		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
+	}, {
+		.name = "pingpong_5", .id = PINGPONG_5,
+		.base = 0x6e000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_2,
+		.intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31),
+	}, {
+		.name = "pingpong_cwb_0", .id = PINGPONG_CWB_0,
+		.base = 0x66000, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_3,
+	}, {
+		.name = "pingpong_cwb_1", .id = PINGPONG_CWB_1,
+		.base = 0x66400, .len = 0,
+		.features = BIT(DPU_PINGPONG_DITHER),
+		.sblk = &sc7280_pp_sblk,
+		.merge_3d = MERGE_3D_3,
+	},
+};
+
+static const struct dpu_merge_3d_cfg sar2130p_merge_3d[] = {
+	{
+		.name = "merge_3d_0", .id = MERGE_3D_0,
+		.base = 0x4e000, .len = 0x8,
+	}, {
+		.name = "merge_3d_1", .id = MERGE_3D_1,
+		.base = 0x4f000, .len = 0x8,
+	}, {
+		.name = "merge_3d_2", .id = MERGE_3D_2,
+		.base = 0x50000, .len = 0x8,
+	}, {
+		.name = "merge_3d_3", .id = MERGE_3D_3,
+		.base = 0x66700, .len = 0x8,
+	},
+};
+
+/*
+ * NOTE: Each display compression engine (DCE) contains dual hard
+ * slice DSC encoders so both share same base address but with
+ * its own different sub block address.
+ */
+static const struct dpu_dsc_cfg sar2130p_dsc[] = {
+	{
+		.name = "dce_0_0", .id = DSC_0,
+		.base = 0x80000, .len = 0x4,
+		.features = BIT(DPU_DSC_HW_REV_1_2),
+		.sblk = &dsc_sblk_0,
+	}, {
+		.name = "dce_0_1", .id = DSC_1,
+		.base = 0x80000, .len = 0x4,
+		.features = BIT(DPU_DSC_HW_REV_1_2),
+		.sblk = &dsc_sblk_1,
+	}, {
+		.name = "dce_1_0", .id = DSC_2,
+		.base = 0x81000, .len = 0x4,
+		.features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN),
+		.sblk = &dsc_sblk_0,
+	}, {
+		.name = "dce_1_1", .id = DSC_3,
+		.base = 0x81000, .len = 0x4,
+		.features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN),
+		.sblk = &dsc_sblk_1,
+	},
+};
+
+static const struct dpu_wb_cfg sar2130p_wb[] = {
+	{
+		.name = "wb_2", .id = WB_2,
+		.base = 0x65000, .len = 0x2c8,
+		.features = WB_SM8250_MASK,
+		.format_list = wb2_formats_rgb_yuv,
+		.num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv),
+		.xin_id = 6,
+		.vbif_idx = VBIF_RT,
+		.maxlinewidth = 4096,
+		.intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4),
+	},
+};
+
+static const struct dpu_intf_cfg sar2130p_intf[] = {
+	{
+		.name = "intf_0", .id = INTF_0,
+		.base = 0x34000, .len = 0x280,
+		.features = INTF_SC7280_MASK,
+		.type = INTF_DP,
+		.controller_id = MSM_DP_CONTROLLER_0,
+		.prog_fetch_lines_worst_case = 24,
+		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25),
+	}, {
+		.name = "intf_1", .id = INTF_1,
+		.base = 0x35000, .len = 0x300,
+		.features = INTF_SC7280_MASK,
+		.type = INTF_DSI,
+		.controller_id = MSM_DSI_CONTROLLER_0,
+		.prog_fetch_lines_worst_case = 24,
+		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+		.intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2),
+	}, {
+		.name = "intf_2", .id = INTF_2,
+		.base = 0x36000, .len = 0x300,
+		.features = INTF_SC7280_MASK,
+		.type = INTF_DSI,
+		.controller_id = MSM_DSI_CONTROLLER_1,
+		.prog_fetch_lines_worst_case = 24,
+		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+		.intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2),
+	}, {
+		.name = "intf_3", .id = INTF_3,
+		.base = 0x37000, .len = 0x280,
+		.features = INTF_SC7280_MASK,
+		.type = INTF_DP,
+		.controller_id = MSM_DP_CONTROLLER_1,
+		.prog_fetch_lines_worst_case = 24,
+		.intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+		.intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31),
+	},
+};
+
+static const struct dpu_perf_cfg sar2130p_perf_data = {
+	.max_bw_low = 13600000,
+	.max_bw_high = 18200000,
+	.min_core_ib = 2500000,
+	.min_llcc_ib = 0,
+	.min_dram_ib = 800000,
+	.min_prefill_lines = 35,
+	/* FIXME: lut tables */
+	.danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0},
+	.safe_lut_tbl = {0xfe00, 0xfe00, 0xffff},
+	.qos_lut_tbl = {
+		{.nentry = ARRAY_SIZE(sc7180_qos_linear),
+		.entries = sc7180_qos_linear
+		},
+		{.nentry = ARRAY_SIZE(sc7180_qos_macrotile),
+		.entries = sc7180_qos_macrotile
+		},
+		{.nentry = ARRAY_SIZE(sc7180_qos_nrt),
+		.entries = sc7180_qos_nrt
+		},
+		/* TODO: macrotile-qseed is different from macrotile */
+	},
+	.cdp_cfg = {
+		{.rd_enable = 0, .wr_enable = 0},
+		{.rd_enable = 0, .wr_enable = 0}
+	},
+	.clk_inefficiency_factor = 105,
+	.bw_inefficiency_factor = 120,
+};
+
+static const struct dpu_mdss_version sar2130p_mdss_ver = {
+	.core_major_ver = 9,
+	.core_minor_ver = 1,
+};
+
+const struct dpu_mdss_cfg dpu_sar2130p_cfg = {
+	.mdss_ver = &sar2130p_mdss_ver,
+	.caps = &sar2130p_dpu_caps,
+	.mdp = &sar2130p_mdp,
+	.cdm = &dpu_cdm_5_x,
+	.ctl_count = ARRAY_SIZE(sar2130p_ctl),
+	.ctl = sar2130p_ctl,
+	.sspp_count = ARRAY_SIZE(sar2130p_sspp),
+	.sspp = sar2130p_sspp,
+	.mixer_count = ARRAY_SIZE(sar2130p_lm),
+	.mixer = sar2130p_lm,
+	.dspp_count = ARRAY_SIZE(sar2130p_dspp),
+	.dspp = sar2130p_dspp,
+	.pingpong_count = ARRAY_SIZE(sar2130p_pp),
+	.pingpong = sar2130p_pp,
+	.dsc_count = ARRAY_SIZE(sar2130p_dsc),
+	.dsc = sar2130p_dsc,
+	.merge_3d_count = ARRAY_SIZE(sar2130p_merge_3d),
+	.merge_3d = sar2130p_merge_3d,
+	.wb_count = ARRAY_SIZE(sar2130p_wb),
+	.wb = sar2130p_wb,
+	.intf_count = ARRAY_SIZE(sar2130p_intf),
+	.intf = sar2130p_intf,
+	.vbif_count = ARRAY_SIZE(sm8550_vbif),
+	.vbif = sm8550_vbif,
+	.perf = &sar2130p_perf_data,
+};
+
+#endif
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h
index 6417baa84f82..52cc10aec1f9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h
@@ -26,17 +26,16 @@ static const struct dpu_mdp_cfg x1e80100_mdp = {
 	},
 };
 
-/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg x1e80100_ctl[] = {
 	{
 		.name = "ctl_0", .id = CTL_0,
 		.base = 0x15000, .len = 0x290,
-		.features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = CTL_SM8550_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	}, {
 		.name = "ctl_1", .id = CTL_1,
 		.base = 0x16000, .len = 0x290,
-		.features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY),
+		.features = CTL_SM8550_MASK,
 		.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	}, {
 		.name = "ctl_2", .id = CTL_2,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index 0714936d8835..a4b0fe0d9899 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -445,9 +445,9 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc,
 
 	uint32_t lm_idx;
 	bool bg_alpha_enable = false;
-	DECLARE_BITMAP(fetch_active, SSPP_MAX);
+	DECLARE_BITMAP(active_fetch, SSPP_MAX);
 
-	memset(fetch_active, 0, sizeof(fetch_active));
+	memset(active_fetch, 0, sizeof(active_fetch));
 	drm_atomic_crtc_for_each_plane(plane, crtc) {
 		state = plane->state;
 		if (!state)
@@ -464,7 +464,7 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc,
 		if (pstate->stage == DPU_STAGE_BASE && format->alpha_enable)
 			bg_alpha_enable = true;
 
-		set_bit(pstate->pipe.sspp->idx, fetch_active);
+		set_bit(pstate->pipe.sspp->idx, active_fetch);
 		_dpu_crtc_blend_setup_pipe(crtc, plane,
 					   mixer, cstate->num_mixers,
 					   pstate->stage,
@@ -472,7 +472,7 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc,
 					   &pstate->pipe, 0, stage_cfg);
 
 		if (pstate->r_pipe.sspp) {
-			set_bit(pstate->r_pipe.sspp->idx, fetch_active);
+			set_bit(pstate->r_pipe.sspp->idx, active_fetch);
 			_dpu_crtc_blend_setup_pipe(crtc, plane,
 						   mixer, cstate->num_mixers,
 						   pstate->stage,
@@ -492,8 +492,8 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc,
 		}
 	}
 
-	if (ctl->ops.set_active_pipes)
-		ctl->ops.set_active_pipes(ctl, fetch_active);
+	if (ctl->ops.set_active_fetch_pipes)
+		ctl->ops.set_active_fetch_pipes(ctl, active_fetch);
 
 	_dpu_crtc_program_lm_output_roi(crtc);
 }
@@ -519,6 +519,8 @@ static void _dpu_crtc_blend_setup(struct drm_crtc *crtc)
 		if (mixer[i].lm_ctl->ops.clear_all_blendstages)
 			mixer[i].lm_ctl->ops.clear_all_blendstages(
 					mixer[i].lm_ctl);
+		if (mixer[i].lm_ctl->ops.set_active_fetch_pipes)
+			mixer[i].lm_ctl->ops.set_active_fetch_pipes(mixer[i].lm_ctl, NULL);
 	}
 
 	/* initialize stage cfg */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 8610bbf2b87c..7020098360e4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -1246,7 +1246,11 @@ static void dpu_encoder_virt_atomic_mode_set(struct drm_encoder *drm_enc,
 			return;
 		}
 
-		phys->hw_ctl = i < num_ctl ? to_dpu_hw_ctl(hw_ctl[i]) : NULL;
+		/* Use first (and only) CTL if active CTLs are supported */
+		if (num_ctl == 1)
+			phys->hw_ctl = to_dpu_hw_ctl(hw_ctl[0]);
+		else
+			phys->hw_ctl = i < num_ctl ? to_dpu_hw_ctl(hw_ctl[i]) : NULL;
 		if (!phys->hw_ctl) {
 			DPU_ERROR_ENC(dpu_enc,
 				"no ctl block assigned at idx: %d\n", i);
@@ -1666,7 +1670,7 @@ static void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc,
  */
 static void _dpu_encoder_trigger_start(struct dpu_encoder_phys *phys)
 {
-	struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(phys->parent);
+	struct dpu_encoder_virt *dpu_enc;
 
 	if (!phys) {
 		DPU_ERROR("invalid argument(s)\n");
@@ -1678,6 +1682,8 @@ static void _dpu_encoder_trigger_start(struct dpu_encoder_phys *phys)
 		return;
 	}
 
+	dpu_enc = to_dpu_encoder_virt(phys->parent);
+
 	if (phys->parent->encoder_type == DRM_MODE_ENCODER_VIRTUAL &&
 	    dpu_enc->cwb_mask) {
 		DPU_DEBUG("encoder %d CWB enabled, skipping\n", DRMID(phys->parent));
@@ -2188,6 +2194,9 @@ static void dpu_encoder_helper_reset_mixers(struct dpu_encoder_phys *phys_enc)
 		/* clear all blendstages */
 		if (ctl->ops.setup_blendstage)
 			ctl->ops.setup_blendstage(ctl, hw_mixer[i]->idx, NULL);
+
+		if (ctl->ops.set_active_fetch_pipes)
+			ctl->ops.set_active_fetch_pipes(ctl, NULL);
 	}
 }
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index da9994a79ca2..a0ba55ab3c89 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -60,6 +60,8 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg(
 		return;
 
 	intf_cfg.intf = phys_enc->hw_intf->idx;
+	if (phys_enc->split_role == ENC_ROLE_MASTER)
+		intf_cfg.intf_master = phys_enc->hw_intf->idx;
 	intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_CMD;
 	intf_cfg.stream_sel = cmd_enc->stream_sel;
 	intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
index abd6600046cb..8a618841e3ea 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
@@ -298,6 +298,8 @@ static void dpu_encoder_phys_vid_setup_timing_engine(
 	if (phys_enc->hw_cdm)
 		intf_cfg.cdm = phys_enc->hw_cdm->idx;
 	intf_cfg.intf = phys_enc->hw_intf->idx;
+	if (phys_enc->split_role == ENC_ROLE_MASTER)
+		intf_cfg.intf_master = phys_enc->hw_intf->idx;
 	intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_VID;
 	intf_cfg.stream_sel = 0; /* Don't care value for video mode */
 	intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc);
@@ -372,7 +374,8 @@ static void dpu_encoder_phys_vid_underrun_irq(void *arg)
 static bool dpu_encoder_phys_vid_needs_single_flush(
 		struct dpu_encoder_phys *phys_enc)
 {
-	return phys_enc->split_role != ENC_ROLE_SOLO;
+	return !(phys_enc->hw_ctl->caps->features & BIT(DPU_CTL_ACTIVE_CFG)) &&
+		phys_enc->split_role != ENC_ROLE_SOLO;
 }
 
 static void dpu_encoder_phys_vid_atomic_mode_set(
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 64265ca4656a..c878fe196aeb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -34,11 +34,11 @@
 #define VIG_MSM8998_MASK \
 	(VIG_MASK | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE))
 
-#define VIG_SDM845_MASK \
+#define VIG_SDM845_MASK_NO_SDMA \
 	(VIG_MASK | BIT(DPU_SSPP_QOS_8LVL) | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE))
 
 #define VIG_SDM845_MASK_SDMA \
-	(VIG_SDM845_MASK | BIT(DPU_SSPP_SMART_DMA_V2))
+	(VIG_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2))
 
 #define VIG_QCM2290_MASK (VIG_BASE_MASK | BIT(DPU_SSPP_QOS_8LVL))
 
@@ -54,24 +54,24 @@
 	BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT))
 
 #define VIG_SC7280_MASK \
-	(VIG_SDM845_MASK | BIT(DPU_SSPP_INLINE_ROTATION))
+	(VIG_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_INLINE_ROTATION))
 
 #define VIG_SC7280_MASK_SDMA \
 	(VIG_SC7280_MASK | BIT(DPU_SSPP_SMART_DMA_V2))
 
-#define DMA_SDM845_MASK \
+#define DMA_SDM845_MASK_NO_SDMA \
 	(BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_QOS_8LVL) |\
 	BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\
 	BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT))
 
-#define DMA_CURSOR_SDM845_MASK \
-	(DMA_SDM845_MASK | BIT(DPU_SSPP_CURSOR))
+#define DMA_CURSOR_SDM845_MASK_NO_SDMA \
+	(DMA_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_CURSOR))
 
 #define DMA_SDM845_MASK_SDMA \
-	(DMA_SDM845_MASK | BIT(DPU_SSPP_SMART_DMA_V2))
+	(DMA_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2))
 
 #define DMA_CURSOR_SDM845_MASK_SDMA \
-	(DMA_CURSOR_SDM845_MASK | BIT(DPU_SSPP_SMART_DMA_V2))
+	(DMA_CURSOR_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2))
 
 #define DMA_CURSOR_MSM8996_MASK \
 	(DMA_MSM8996_MASK | BIT(DPU_SSPP_CURSOR))
@@ -98,15 +98,9 @@
 #define PINGPONG_MSM8996_MASK \
 	(BIT(DPU_PINGPONG_DSC))
 
-#define PINGPONG_MSM8996_TE2_MASK \
-	(PINGPONG_MSM8996_MASK | BIT(DPU_PINGPONG_TE2))
-
 #define PINGPONG_SDM845_MASK \
 	(BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_DSC))
 
-#define PINGPONG_SDM845_TE2_MASK \
-	(PINGPONG_SDM845_MASK | BIT(DPU_PINGPONG_TE2))
-
 #define PINGPONG_SM8150_MASK \
 	(BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_DSC))
 
@@ -376,8 +370,6 @@ static const struct dpu_sspp_sub_blks dpu_dma_sblk = _DMA_SBLK();
  * MIXER sub blocks config
  *************************************************************/
 
-/* MSM8998 */
-
 static const struct dpu_lm_sub_blks msm8998_lm_sblk = {
 	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 7, /* excluding base layer */
@@ -387,8 +379,6 @@ static const struct dpu_lm_sub_blks msm8998_lm_sblk = {
 	},
 };
 
-/* SDM845 */
-
 static const struct dpu_lm_sub_blks sdm845_lm_sblk = {
 	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 11, /* excluding base layer */
@@ -398,8 +388,6 @@ static const struct dpu_lm_sub_blks sdm845_lm_sblk = {
 	},
 };
 
-/* SC7180 */
-
 static const struct dpu_lm_sub_blks sc7180_lm_sblk = {
 	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 7, /* excluding base layer */
@@ -408,8 +396,6 @@ static const struct dpu_lm_sub_blks sc7180_lm_sblk = {
 	},
 };
 
-/* QCM2290 */
-
 static const struct dpu_lm_sub_blks qcm2290_lm_sblk = {
 	.maxwidth = DEFAULT_DPU_LINE_WIDTH,
 	.maxblendstages = 4, /* excluding base layer */
@@ -434,22 +420,11 @@ static const struct dpu_dspp_sub_blks sdm845_dspp_sblk = {
 /*************************************************************
  * PINGPONG sub blocks config
  *************************************************************/
-static const struct dpu_pingpong_sub_blks msm8996_pp_sblk_te = {
-	.te2 = {.name = "te2", .base = 0x2000, .len = 0x0,
-		.version = 0x1},
-};
 
 static const struct dpu_pingpong_sub_blks msm8996_pp_sblk = {
 	/* No dither block */
 };
 
-static const struct dpu_pingpong_sub_blks sdm845_pp_sblk_te = {
-	.te2 = {.name = "te2", .base = 0x2000, .len = 0x0,
-		.version = 0x1},
-	.dither = {.name = "dither", .base = 0x30e0,
-		.len = 0x20, .version = 0x10000},
-};
-
 static const struct dpu_pingpong_sub_blks sdm845_pp_sblk = {
 	.dither = {.name = "dither", .base = 0x30e0,
 		.len = 0x20, .version = 0x10000},
@@ -759,7 +734,7 @@ static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = {
 #include "catalog/dpu_8_4_sa8775p.h"
 
 #include "catalog/dpu_9_0_sm8550.h"
-
+#include "catalog/dpu_9_1_sar2130p.h"
 #include "catalog/dpu_9_2_x1e80100.h"
 
 #include "catalog/dpu_10_0_sm8650.h"
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 4cea19e1a203..01dd6e65f777 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -115,7 +115,6 @@ enum {
 
 /**
  * PINGPONG sub-blocks
- * @DPU_PINGPONG_TE2        Additional tear check block for split pipes
  * @DPU_PINGPONG_SPLIT      PP block supports split fifo
  * @DPU_PINGPONG_SLAVE      PP block is a suitable slave for split fifo
  * @DPU_PINGPONG_DITHER     Dither blocks
@@ -123,8 +122,7 @@ enum {
  * @DPU_PINGPONG_MAX
  */
 enum {
-	DPU_PINGPONG_TE2 = 0x1,
-	DPU_PINGPONG_SPLIT,
+	DPU_PINGPONG_SPLIT = 0x1,
 	DPU_PINGPONG_SLAVE,
 	DPU_PINGPONG_DITHER,
 	DPU_PINGPONG_DSC,
@@ -404,8 +402,6 @@ struct dpu_dspp_sub_blks {
 };
 
 struct dpu_pingpong_sub_blks {
-	struct dpu_pp_blk te;
-	struct dpu_pp_blk te2;
 	struct dpu_pp_blk dither;
 };
 
@@ -841,6 +837,7 @@ extern const struct dpu_mdss_cfg dpu_msm8937_cfg;
 extern const struct dpu_mdss_cfg dpu_msm8953_cfg;
 extern const struct dpu_mdss_cfg dpu_msm8996_cfg;
 extern const struct dpu_mdss_cfg dpu_msm8998_cfg;
+extern const struct dpu_mdss_cfg dpu_sar2130p_cfg;
 extern const struct dpu_mdss_cfg dpu_sdm630_cfg;
 extern const struct dpu_mdss_cfg dpu_sdm660_cfg;
 extern const struct dpu_mdss_cfg dpu_sdm845_cfg;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index 411a7cf088eb..573e42b06ad0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -261,6 +261,12 @@ static void dpu_hw_ctl_update_pending_flush_mixer(struct dpu_hw_ctl *ctx,
 	case LM_5:
 		ctx->pending_flush_mask |= BIT(20);
 		break;
+	case LM_6:
+		ctx->pending_flush_mask |= BIT(21);
+		break;
+	case LM_7:
+		ctx->pending_flush_mask |= BIT(27);
+		break;
 	default:
 		break;
 	}
@@ -563,6 +569,7 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	u32 wb_active = 0;
 	u32 cwb_active = 0;
 	u32 mode_sel = 0;
+	u32 merge_3d_active = 0;
 
 	/* CTL_TOP[31:28] carries group_id to collate CTL paths
 	 * per VM. Explicitly disable it until VM support is
@@ -578,6 +585,7 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	wb_active = DPU_REG_READ(c, CTL_WB_ACTIVE);
 	cwb_active = DPU_REG_READ(c, CTL_CWB_ACTIVE);
 	dsc_active = DPU_REG_READ(c, CTL_DSC_ACTIVE);
+	merge_3d_active = DPU_REG_READ(c, CTL_MERGE_3D_ACTIVE);
 
 	if (cfg->intf)
 		intf_active |= BIT(cfg->intf - INTF_0);
@@ -591,15 +599,18 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	if (cfg->dsc)
 		dsc_active |= cfg->dsc;
 
+	if (cfg->merge_3d)
+		merge_3d_active |= BIT(cfg->merge_3d - MERGE_3D_0);
+
 	DPU_REG_WRITE(c, CTL_TOP, mode_sel);
 	DPU_REG_WRITE(c, CTL_INTF_ACTIVE, intf_active);
 	DPU_REG_WRITE(c, CTL_WB_ACTIVE, wb_active);
 	DPU_REG_WRITE(c, CTL_CWB_ACTIVE, cwb_active);
 	DPU_REG_WRITE(c, CTL_DSC_ACTIVE, dsc_active);
+	DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE, merge_3d_active);
 
-	if (cfg->merge_3d)
-		DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE,
-			      BIT(cfg->merge_3d - MERGE_3D_0));
+	if (cfg->intf_master)
+		DPU_REG_WRITE(c, CTL_INTF_MASTER, BIT(cfg->intf_master - INTF_0));
 
 	if (cfg->cdm)
 		DPU_REG_WRITE(c, CTL_CDM_ACTIVE, cfg->cdm);
@@ -643,6 +654,7 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 {
 	struct dpu_hw_blk_reg_map *c = &ctx->hw;
 	u32 intf_active = 0;
+	u32 intf_master = 0;
 	u32 wb_active = 0;
 	u32 cwb_active = 0;
 	u32 merge3d_active = 0;
@@ -666,10 +678,21 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 
 	dpu_hw_ctl_clear_all_blendstages(ctx);
 
+	if (ctx->ops.set_active_fetch_pipes)
+		ctx->ops.set_active_fetch_pipes(ctx, NULL);
+
 	if (cfg->intf) {
 		intf_active = DPU_REG_READ(c, CTL_INTF_ACTIVE);
 		intf_active &= ~BIT(cfg->intf - INTF_0);
 		DPU_REG_WRITE(c, CTL_INTF_ACTIVE, intf_active);
+
+		intf_master = DPU_REG_READ(c, CTL_INTF_MASTER);
+
+		/* Unset this intf as master, if it is the current master */
+		if (intf_master == BIT(cfg->intf - INTF_0)) {
+			DPU_DEBUG_DRIVER("Unsetting INTF_%d master\n", cfg->intf - INTF_0);
+			DPU_REG_WRITE(c, CTL_INTF_MASTER, 0);
+		}
 	}
 
 	if (cfg->cwb) {
@@ -697,8 +720,8 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	}
 }
 
-static void dpu_hw_ctl_set_fetch_pipe_active(struct dpu_hw_ctl *ctx,
-	unsigned long *fetch_active)
+static void dpu_hw_ctl_set_active_fetch_pipes(struct dpu_hw_ctl *ctx,
+					      unsigned long *fetch_active)
 {
 	int i;
 	u32 val = 0;
@@ -761,7 +784,7 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops,
 		ops->update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp;
 
 	if (cap & BIT(DPU_CTL_FETCH_ACTIVE))
-		ops->set_active_pipes = dpu_hw_ctl_set_fetch_pipe_active;
+		ops->set_active_fetch_pipes = dpu_hw_ctl_set_active_fetch_pipes;
 };
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
index 080a9550a0cc..feb09590bc8f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
@@ -36,6 +36,7 @@ struct dpu_hw_stage_cfg {
 /**
  * struct dpu_hw_intf_cfg :Describes how the DPU writes data to output interface
  * @intf :                 Interface id
+ * @intf_master:           Master interface id in the dual pipe topology
  * @mode_3d:               3d mux configuration
  * @merge_3d:              3d merge block used
  * @intf_mode_sel:         Interface mode, cmd / vid
@@ -46,6 +47,7 @@ struct dpu_hw_stage_cfg {
  */
 struct dpu_hw_intf_cfg {
 	enum dpu_intf intf;
+	enum dpu_intf intf_master;
 	enum dpu_wb wb;
 	enum dpu_3d_blend_mode mode_3d;
 	enum dpu_merge_3d merge_3d;
@@ -254,7 +256,7 @@ struct dpu_hw_ctl_ops {
 	void (*setup_blendstage)(struct dpu_hw_ctl *ctx,
 		enum dpu_lm lm, struct dpu_hw_stage_cfg *cfg);
 
-	void (*set_active_pipes)(struct dpu_hw_ctl *ctx,
+	void (*set_active_fetch_pipes)(struct dpu_hw_ctl *ctx,
 		unsigned long *fetch_active);
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
index 8d820cd1b554..175639c8bfbb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
@@ -125,6 +125,7 @@ enum dpu_lm {
 	LM_4,
 	LM_5,
 	LM_6,
+	LM_7,
 	LM_MAX
 };
 
@@ -169,6 +170,8 @@ enum dpu_dsc {
 	DSC_3,
 	DSC_4,
 	DSC_5,
+	DSC_6,
+	DSC_7,
 	DSC_MAX
 };
 
@@ -185,6 +188,8 @@ enum dpu_pingpong {
 	PINGPONG_3,
 	PINGPONG_4,
 	PINGPONG_5,
+	PINGPONG_6,
+	PINGPONG_7,
 	PINGPONG_CWB_0,
 	PINGPONG_CWB_1,
 	PINGPONG_CWB_2,
@@ -199,6 +204,7 @@ enum dpu_merge_3d {
 	MERGE_3D_2,
 	MERGE_3D_3,
 	MERGE_3D_4,
+	MERGE_3D_5,
 	MERGE_3D_MAX
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 3305ad0623ca..1fd82b6747e9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1512,6 +1512,7 @@ static const struct of_device_id dpu_dt_match[] = {
 	{ .compatible = "qcom,msm8998-dpu", .data = &dpu_msm8998_cfg, },
 	{ .compatible = "qcom,qcm2290-dpu", .data = &dpu_qcm2290_cfg, },
 	{ .compatible = "qcom,sa8775p-dpu", .data = &dpu_sa8775p_cfg, },
+	{ .compatible = "qcom,sar2130p-dpu", .data = &dpu_sar2130p_cfg, },
 	{ .compatible = "qcom,sdm630-mdp5", .data = &dpu_sdm630_cfg, },
 	{ .compatible = "qcom,sdm660-mdp5", .data = &dpu_sdm660_cfg, },
 	{ .compatible = "qcom,sdm670-dpu", .data = &dpu_sdm670_cfg, },
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index af3e541f60c3..421138bc3cb7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -729,12 +729,40 @@ static int dpu_plane_check_inline_rotation(struct dpu_plane *pdpu,
 static int dpu_plane_atomic_check_pipe(struct dpu_plane *pdpu,
 		struct dpu_sw_pipe *pipe,
 		struct dpu_sw_pipe_cfg *pipe_cfg,
-		const struct msm_format *fmt,
-		const struct drm_display_mode *mode)
+		const struct drm_display_mode *mode,
+		struct drm_plane_state *new_plane_state)
 {
 	uint32_t min_src_size;
 	struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base);
 	int ret;
+	const struct msm_format *fmt;
+	uint32_t supported_rotations;
+	const struct dpu_sspp_cfg *pipe_hw_caps;
+	const struct dpu_sspp_sub_blks *sblk;
+
+	pipe_hw_caps = pipe->sspp->cap;
+	sblk = pipe->sspp->cap->sblk;
+
+	/*
+	 * We already have verified scaling against platform limitations.
+	 * Now check if the SSPP supports scaling at all.
+	 */
+	if (!sblk->scaler_blk.len &&
+	    ((drm_rect_width(&new_plane_state->src) >> 16 !=
+	      drm_rect_width(&new_plane_state->dst)) ||
+	     (drm_rect_height(&new_plane_state->src) >> 16 !=
+	      drm_rect_height(&new_plane_state->dst))))
+		return -ERANGE;
+
+	fmt = msm_framebuffer_format(new_plane_state->fb);
+
+	supported_rotations = DRM_MODE_REFLECT_MASK | DRM_MODE_ROTATE_0;
+
+	if (pipe_hw_caps->features & BIT(DPU_SSPP_INLINE_ROTATION))
+		supported_rotations |= DRM_MODE_ROTATE_90;
+
+	pipe_cfg->rotation = drm_rotation_simplify(new_plane_state->rotation,
+						   supported_rotations);
 
 	min_src_size = MSM_FORMAT_IS_YUV(fmt) ? 2 : 1;
 
@@ -887,10 +915,9 @@ static int dpu_plane_atomic_check_nosspp(struct drm_plane *plane,
 	return 0;
 }
 
-static int dpu_plane_is_multirect_parallel_capable(struct dpu_hw_sspp *sspp,
-						   struct dpu_sw_pipe_cfg *pipe_cfg,
-						   const struct msm_format *fmt,
-						   uint32_t max_linewidth)
+static int dpu_plane_is_multirect_capable(struct dpu_hw_sspp *sspp,
+					  struct dpu_sw_pipe_cfg *pipe_cfg,
+					  const struct msm_format *fmt)
 {
 	if (drm_rect_width(&pipe_cfg->src_rect) != drm_rect_width(&pipe_cfg->dst_rect) ||
 	    drm_rect_height(&pipe_cfg->src_rect) != drm_rect_height(&pipe_cfg->dst_rect))
@@ -902,10 +929,6 @@ static int dpu_plane_is_multirect_parallel_capable(struct dpu_hw_sspp *sspp,
 	if (MSM_FORMAT_IS_YUV(fmt))
 		return false;
 
-	if (MSM_FORMAT_IS_UBWC(fmt) &&
-	    drm_rect_width(&pipe_cfg->src_rect) > max_linewidth / 2)
-		return false;
-
 	if (!test_bit(DPU_SSPP_SMART_DMA_V1, &sspp->cap->features) &&
 	    !test_bit(DPU_SSPP_SMART_DMA_V2, &sspp->cap->features))
 		return false;
@@ -913,6 +936,27 @@ static int dpu_plane_is_multirect_parallel_capable(struct dpu_hw_sspp *sspp,
 	return true;
 }
 
+static int dpu_plane_is_parallel_capable(struct dpu_sw_pipe_cfg *pipe_cfg,
+					 const struct msm_format *fmt,
+					 uint32_t max_linewidth)
+{
+	if (MSM_FORMAT_IS_UBWC(fmt) &&
+	    drm_rect_width(&pipe_cfg->src_rect) > max_linewidth / 2)
+		return false;
+
+	return true;
+}
+
+static int dpu_plane_is_multirect_parallel_capable(struct dpu_hw_sspp *sspp,
+						   struct dpu_sw_pipe_cfg *pipe_cfg,
+						   const struct msm_format *fmt,
+						   uint32_t max_linewidth)
+{
+	return dpu_plane_is_multirect_capable(sspp, pipe_cfg, fmt) &&
+		dpu_plane_is_parallel_capable(pipe_cfg, fmt, max_linewidth);
+}
+
+
 static int dpu_plane_atomic_check_sspp(struct drm_plane *plane,
 				       struct drm_atomic_state *state,
 				       const struct drm_crtc_state *crtc_state)
@@ -923,47 +967,20 @@ static int dpu_plane_atomic_check_sspp(struct drm_plane *plane,
 	struct dpu_plane_state *pstate = to_dpu_plane_state(new_plane_state);
 	struct dpu_sw_pipe *pipe = &pstate->pipe;
 	struct dpu_sw_pipe *r_pipe = &pstate->r_pipe;
-	const struct msm_format *fmt;
 	struct dpu_sw_pipe_cfg *pipe_cfg = &pstate->pipe_cfg;
 	struct dpu_sw_pipe_cfg *r_pipe_cfg = &pstate->r_pipe_cfg;
-	uint32_t supported_rotations;
-	const struct dpu_sspp_cfg *pipe_hw_caps;
-	const struct dpu_sspp_sub_blks *sblk;
 	int ret = 0;
 
-	pipe_hw_caps = pipe->sspp->cap;
-	sblk = pipe->sspp->cap->sblk;
-
-	/*
-	 * We already have verified scaling against platform limitations.
-	 * Now check if the SSPP supports scaling at all.
-	 */
-	if (!sblk->scaler_blk.len &&
-	    ((drm_rect_width(&new_plane_state->src) >> 16 !=
-	      drm_rect_width(&new_plane_state->dst)) ||
-	     (drm_rect_height(&new_plane_state->src) >> 16 !=
-	      drm_rect_height(&new_plane_state->dst))))
-		return -ERANGE;
-
-	fmt = msm_framebuffer_format(new_plane_state->fb);
-
-	supported_rotations = DRM_MODE_REFLECT_MASK | DRM_MODE_ROTATE_0;
-
-	if (pipe_hw_caps->features & BIT(DPU_SSPP_INLINE_ROTATION))
-		supported_rotations |= DRM_MODE_ROTATE_90;
-
-	pipe_cfg->rotation = drm_rotation_simplify(new_plane_state->rotation,
-						   supported_rotations);
-	r_pipe_cfg->rotation = pipe_cfg->rotation;
-
-	ret = dpu_plane_atomic_check_pipe(pdpu, pipe, pipe_cfg, fmt,
-					  &crtc_state->adjusted_mode);
+	ret = dpu_plane_atomic_check_pipe(pdpu, pipe, pipe_cfg,
+					  &crtc_state->adjusted_mode,
+					  new_plane_state);
 	if (ret)
 		return ret;
 
 	if (drm_rect_width(&r_pipe_cfg->src_rect) != 0) {
-		ret = dpu_plane_atomic_check_pipe(pdpu, r_pipe, r_pipe_cfg, fmt,
-						  &crtc_state->adjusted_mode);
+		ret = dpu_plane_atomic_check_pipe(pdpu, r_pipe, r_pipe_cfg,
+						  &crtc_state->adjusted_mode,
+						  new_plane_state);
 		if (ret)
 			return ret;
 	}
@@ -1001,6 +1018,69 @@ static bool dpu_plane_try_multirect_parallel(struct dpu_sw_pipe *pipe, struct dp
 	return true;
 }
 
+static int dpu_plane_try_multirect_shared(struct dpu_plane_state *pstate,
+					  struct dpu_plane_state *prev_adjacent_pstate,
+					  const struct msm_format *fmt,
+					  uint32_t max_linewidth)
+{
+	struct dpu_sw_pipe *pipe = &pstate->pipe;
+	struct dpu_sw_pipe *r_pipe = &pstate->r_pipe;
+	struct dpu_sw_pipe_cfg *pipe_cfg = &pstate->pipe_cfg;
+	struct dpu_sw_pipe *prev_pipe = &prev_adjacent_pstate->pipe;
+	struct dpu_sw_pipe_cfg *prev_pipe_cfg = &prev_adjacent_pstate->pipe_cfg;
+	const struct msm_format *prev_fmt = msm_framebuffer_format(prev_adjacent_pstate->base.fb);
+	u16 max_tile_height = 1;
+
+	if (prev_adjacent_pstate->r_pipe.sspp != NULL ||
+	    prev_pipe->multirect_mode != DPU_SSPP_MULTIRECT_NONE)
+		return false;
+
+	if (!dpu_plane_is_multirect_capable(pipe->sspp, pipe_cfg, fmt) ||
+	    !dpu_plane_is_multirect_capable(prev_pipe->sspp, prev_pipe_cfg, prev_fmt))
+		return false;
+
+	if (MSM_FORMAT_IS_UBWC(fmt))
+		max_tile_height = max(max_tile_height, fmt->tile_height);
+
+	if (MSM_FORMAT_IS_UBWC(prev_fmt))
+		max_tile_height = max(max_tile_height, prev_fmt->tile_height);
+
+	r_pipe->multirect_index = DPU_SSPP_RECT_SOLO;
+	r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+
+	r_pipe->sspp = NULL;
+
+	if (dpu_plane_is_parallel_capable(pipe_cfg, fmt, max_linewidth) &&
+	    dpu_plane_is_parallel_capable(prev_pipe_cfg, prev_fmt, max_linewidth) &&
+	    (pipe_cfg->dst_rect.x1 >= prev_pipe_cfg->dst_rect.x2 ||
+	     prev_pipe_cfg->dst_rect.x1 >= pipe_cfg->dst_rect.x2)) {
+		pipe->sspp = prev_pipe->sspp;
+
+		pipe->multirect_index = DPU_SSPP_RECT_1;
+		pipe->multirect_mode = DPU_SSPP_MULTIRECT_PARALLEL;
+
+		prev_pipe->multirect_index = DPU_SSPP_RECT_0;
+		prev_pipe->multirect_mode = DPU_SSPP_MULTIRECT_PARALLEL;
+
+		return true;
+	}
+
+	if (pipe_cfg->dst_rect.y1 >= prev_pipe_cfg->dst_rect.y2 + 2 * max_tile_height ||
+	    prev_pipe_cfg->dst_rect.y1 >= pipe_cfg->dst_rect.y2 + 2 * max_tile_height) {
+		pipe->sspp = prev_pipe->sspp;
+
+		pipe->multirect_index = DPU_SSPP_RECT_1;
+		pipe->multirect_mode = DPU_SSPP_MULTIRECT_TIME_MX;
+
+		prev_pipe->multirect_index = DPU_SSPP_RECT_0;
+		prev_pipe->multirect_mode = DPU_SSPP_MULTIRECT_TIME_MX;
+
+		return true;
+	}
+
+	return false;
+}
+
 static int dpu_plane_atomic_check(struct drm_plane *plane,
 				  struct drm_atomic_state *state)
 {
@@ -1059,6 +1139,9 @@ static int dpu_plane_virtual_atomic_check(struct drm_plane *plane,
 	struct drm_crtc_state *crtc_state;
 	int ret;
 
+	if (IS_ERR(plane_state))
+		return PTR_ERR(plane_state);
+
 	if (plane_state->crtc)
 		crtc_state = drm_atomic_get_new_crtc_state(state,
 							   plane_state->crtc);
@@ -1098,13 +1181,14 @@ static int dpu_plane_virtual_atomic_check(struct drm_plane *plane,
 static int dpu_plane_virtual_assign_resources(struct drm_crtc *crtc,
 					      struct dpu_global_state *global_state,
 					      struct drm_atomic_state *state,
-					      struct drm_plane_state *plane_state)
+					      struct drm_plane_state *plane_state,
+					      struct drm_plane_state *prev_adjacent_plane_state)
 {
 	const struct drm_crtc_state *crtc_state = NULL;
 	struct drm_plane *plane = plane_state->plane;
 	struct dpu_kms *dpu_kms = _dpu_plane_get_kms(plane);
 	struct dpu_rm_sspp_requirements reqs;
-	struct dpu_plane_state *pstate;
+	struct dpu_plane_state *pstate, *prev_adjacent_pstate;
 	struct dpu_sw_pipe *pipe;
 	struct dpu_sw_pipe *r_pipe;
 	struct dpu_sw_pipe_cfg *pipe_cfg;
@@ -1116,6 +1200,8 @@ static int dpu_plane_virtual_assign_resources(struct drm_crtc *crtc,
 							   plane_state->crtc);
 
 	pstate = to_dpu_plane_state(plane_state);
+	prev_adjacent_pstate = prev_adjacent_plane_state ?
+		to_dpu_plane_state(prev_adjacent_plane_state) : NULL;
 	pipe = &pstate->pipe;
 	r_pipe = &pstate->r_pipe;
 	pipe_cfg = &pstate->pipe_cfg;
@@ -1134,24 +1220,42 @@ static int dpu_plane_virtual_assign_resources(struct drm_crtc *crtc,
 
 	reqs.rot90 = drm_rotation_90_or_270(plane_state->rotation);
 
-	pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, &reqs);
-	if (!pipe->sspp)
-		return -ENODEV;
+	if (drm_rect_width(&r_pipe_cfg->src_rect) == 0) {
+		if (!prev_adjacent_pstate ||
+		    !dpu_plane_try_multirect_shared(pstate, prev_adjacent_pstate, fmt,
+						    dpu_kms->catalog->caps->max_linewidth)) {
+			pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, &reqs);
+			if (!pipe->sspp)
+				return -ENODEV;
 
-	if (!dpu_plane_try_multirect_parallel(pipe, pipe_cfg, r_pipe, r_pipe_cfg,
-					      pipe->sspp,
-					      msm_framebuffer_format(plane_state->fb),
-					      dpu_kms->catalog->caps->max_linewidth)) {
-		/* multirect is not possible, use two SSPP blocks */
-		r_pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, &reqs);
-		if (!r_pipe->sspp)
+			r_pipe->sspp = NULL;
+
+			pipe->multirect_index = DPU_SSPP_RECT_SOLO;
+			pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+
+			r_pipe->multirect_index = DPU_SSPP_RECT_SOLO;
+			r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+		}
+	} else {
+		pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, &reqs);
+		if (!pipe->sspp)
 			return -ENODEV;
 
-		pipe->multirect_index = DPU_SSPP_RECT_SOLO;
-		pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+		if (!dpu_plane_try_multirect_parallel(pipe, pipe_cfg, r_pipe, r_pipe_cfg,
+						      pipe->sspp,
+						      msm_framebuffer_format(plane_state->fb),
+						      dpu_kms->catalog->caps->max_linewidth)) {
+			/* multirect is not possible, use two SSPP blocks */
+			r_pipe->sspp = dpu_rm_reserve_sspp(&dpu_kms->rm, global_state, crtc, &reqs);
+			if (!r_pipe->sspp)
+				return -ENODEV;
 
-		r_pipe->multirect_index = DPU_SSPP_RECT_SOLO;
-		r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+			pipe->multirect_index = DPU_SSPP_RECT_SOLO;
+			pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+
+			r_pipe->multirect_index = DPU_SSPP_RECT_SOLO;
+			r_pipe->multirect_mode = DPU_SSPP_MULTIRECT_NONE;
+		}
 	}
 
 	return dpu_plane_atomic_check_sspp(plane, state, crtc_state);
@@ -1164,6 +1268,7 @@ int dpu_assign_plane_resources(struct dpu_global_state *global_state,
 			       unsigned int num_planes)
 {
 	unsigned int i;
+	struct drm_plane_state *prev_adjacent_plane_state = NULL;
 
 	for (i = 0; i < num_planes; i++) {
 		struct drm_plane_state *plane_state = states[i];
@@ -1173,9 +1278,12 @@ int dpu_assign_plane_resources(struct dpu_global_state *global_state,
 			continue;
 
 		int ret = dpu_plane_virtual_assign_resources(crtc, global_state,
-							 state, plane_state);
+							     state, plane_state,
+							     prev_adjacent_plane_state);
 		if (ret)
-			return ret;
+			break;
+
+		prev_adjacent_plane_state = plane_state;
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index 3efbba425ca6..2e296f79cba1 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -53,6 +53,8 @@ int dpu_rm_init(struct drm_device *dev,
 	/* Clear, setup lists */
 	memset(rm, 0, sizeof(*rm));
 
+	rm->has_legacy_ctls = (cat->mdss_ver->core_major_ver < 5);
+
 	/* Interrogate HW catalog and create tracking items for hw blocks */
 	for (i = 0; i < cat->mixer_count; i++) {
 		struct dpu_hw_mixer *hw;
@@ -434,20 +436,19 @@ static int _dpu_rm_reserve_ctls(
 	int i = 0, j, num_ctls;
 	bool needs_split_display;
 
-	/*
-	 * For non-CWB mode, each hw_intf needs its own hw_ctl to program its
-	 * control path.
-	 *
-	 * Hardcode num_ctls to 1 if CWB is enabled because in CWB, both the
-	 * writeback and real-time encoders must be driven by the same control
-	 * path
-	 */
-	if (top->cwb_enabled)
-		num_ctls = 1;
-	else
+	if (rm->has_legacy_ctls) {
+		/*
+		 * TODO: check if there is a need for special handling if
+		 * DPU < 5.0 get CWB support.
+		 */
 		num_ctls = top->num_intf;
 
-	needs_split_display = _dpu_rm_needs_split_display(top);
+		needs_split_display = _dpu_rm_needs_split_display(top);
+	} else {
+		/* use single CTL */
+		num_ctls = 1;
+		needs_split_display = false;
+	}
 
 	for (j = 0; j < ARRAY_SIZE(rm->ctl_blks); j++) {
 		const struct dpu_hw_ctl *ctl;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h
index a19dbdb1b6f4..aa62966056d4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h
@@ -24,6 +24,7 @@ struct dpu_global_state;
  * @dspp_blks: array of dspp hardware resources
  * @hw_sspp: array of sspp hardware resources
  * @cdm_blk: cdm hardware resource
+ * @has_legacy_ctls: DPU uses pre-ACTIVE CTL blocks.
  */
 struct dpu_rm {
 	struct dpu_hw_blk *pingpong_blks[PINGPONG_MAX - PINGPONG_0];
@@ -37,6 +38,7 @@ struct dpu_rm {
 	struct dpu_hw_blk *dsc_blks[DSC_MAX - DSC_0];
 	struct dpu_hw_sspp *hw_sspp[SSPP_MAX - SSPP_NONE];
 	struct dpu_hw_blk *cdm_blk;
+	bool has_legacy_ctls;
 };
 
 struct dpu_rm_sspp_requirements {
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
index c469e66cfc11..7e942c1337b3 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
@@ -6,6 +6,8 @@
 
 #include <linux/delay.h>
 
+#include <drm/drm_bridge.h>
+#include <drm/drm_bridge_connector.h>
 #include <drm/drm_vblank.h>
 
 #include "msm_drv.h"
@@ -189,7 +191,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms,
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_encoder *encoder;
 	struct drm_connector *connector;
-	struct device_node *panel_node;
+	struct drm_bridge *next_bridge;
 	int dsi_id;
 	int ret;
 
@@ -199,27 +201,43 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms,
 		 * bail out early if there is no panel node (no need to
 		 * initialize LCDC encoder and LVDS connector)
 		 */
-		panel_node = of_graph_get_remote_node(dev->dev->of_node, 0, 0);
-		if (!panel_node)
-			return 0;
+		next_bridge = devm_drm_of_get_bridge(dev->dev, dev->dev->of_node, 0, 0);
+		if (IS_ERR(next_bridge)) {
+			ret = PTR_ERR(next_bridge);
+			if (ret == -ENODEV)
+				return 0;
+			return ret;
+		}
 
-		encoder = mdp4_lcdc_encoder_init(dev, panel_node);
+		encoder = mdp4_lcdc_encoder_init(dev);
 		if (IS_ERR(encoder)) {
 			DRM_DEV_ERROR(dev->dev, "failed to construct LCDC encoder\n");
-			of_node_put(panel_node);
 			return PTR_ERR(encoder);
 		}
 
 		/* LCDC can be hooked to DMA_P (TODO: Add DMA_S later?) */
 		encoder->possible_crtcs = 1 << DMA_P;
 
-		connector = mdp4_lvds_connector_init(dev, panel_node, encoder);
+		ret = drm_bridge_attach(encoder, next_bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+		if (ret) {
+			DRM_DEV_ERROR(dev->dev, "failed to attach LVDS panel/bridge: %d\n", ret);
+
+			return ret;
+		}
+
+		connector = drm_bridge_connector_init(dev, encoder);
 		if (IS_ERR(connector)) {
 			DRM_DEV_ERROR(dev->dev, "failed to initialize LVDS connector\n");
-			of_node_put(panel_node);
 			return PTR_ERR(connector);
 		}
 
+		ret = drm_connector_attach_encoder(connector, encoder);
+		if (ret) {
+			DRM_DEV_ERROR(dev->dev, "failed to attach LVDS connector: %d\n", ret);
+
+			return ret;
+		}
+
 		break;
 	case DRM_MODE_ENCODER_TMDS:
 		encoder = mdp4_dtv_encoder_init(dev);
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
index 94b1ba92785f..f9d988076337 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
@@ -191,12 +191,7 @@ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev,
 long mdp4_dtv_round_pixclk(struct drm_encoder *encoder, unsigned long rate);
 struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev);
 
-long mdp4_lcdc_round_pixclk(struct drm_encoder *encoder, unsigned long rate);
-struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
-		struct device_node *panel_node);
-
-struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev,
-		struct device_node *panel_node, struct drm_encoder *encoder);
+struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev);
 
 #ifdef CONFIG_DRM_MSM_DSI
 struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev);
@@ -207,13 +202,6 @@ static inline struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev)
 }
 #endif
 
-#ifdef CONFIG_COMMON_CLK
-struct clk *mpd4_lvds_pll_init(struct drm_device *dev);
-#else
-static inline struct clk *mpd4_lvds_pll_init(struct drm_device *dev)
-{
-	return ERR_PTR(-ENODEV);
-}
-#endif
+struct clk *mpd4_get_lcdc_clock(struct drm_device *dev);
 
 #endif /* __MDP4_KMS_H__ */
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
index 8bbc7fb881d5..06a307c1272d 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
@@ -14,7 +14,6 @@
 
 struct mdp4_lcdc_encoder {
 	struct drm_encoder base;
-	struct device_node *panel_node;
 	struct drm_panel *panel;
 	struct clk *lcdc_clk;
 	unsigned long int pixclock;
@@ -262,19 +261,12 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder)
 	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder =
 			to_mdp4_lcdc_encoder(encoder);
 	struct mdp4_kms *mdp4_kms = get_kms(encoder);
-	struct drm_panel *panel;
 
 	if (WARN_ON(!mdp4_lcdc_encoder->enabled))
 		return;
 
 	mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 0);
 
-	panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node);
-	if (!IS_ERR(panel)) {
-		drm_panel_disable(panel);
-		drm_panel_unprepare(panel);
-	}
-
 	/*
 	 * Wait for a vsync so we know the ENABLE=0 latched before
 	 * the (connector) source of the vsync's gets disabled,
@@ -300,7 +292,6 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder)
 			to_mdp4_lcdc_encoder(encoder);
 	unsigned long pc = mdp4_lcdc_encoder->pixclock;
 	struct mdp4_kms *mdp4_kms = get_kms(encoder);
-	struct drm_panel *panel;
 	uint32_t config;
 	int ret;
 
@@ -335,12 +326,6 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder)
 	if (ret)
 		DRM_DEV_ERROR(dev->dev, "failed to enable lcdc_clk: %d\n", ret);
 
-	panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node);
-	if (!IS_ERR(panel)) {
-		drm_panel_prepare(panel);
-		drm_panel_enable(panel);
-	}
-
 	setup_phy(encoder);
 
 	mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 1);
@@ -348,22 +333,34 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder)
 	mdp4_lcdc_encoder->enabled = true;
 }
 
+static enum drm_mode_status
+mdp4_lcdc_encoder_mode_valid(struct drm_encoder *encoder,
+		const struct drm_display_mode *mode)
+{
+	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder =
+			to_mdp4_lcdc_encoder(encoder);
+	long actual, requested;
+
+	requested = 1000 * mode->clock;
+	actual = clk_round_rate(mdp4_lcdc_encoder->lcdc_clk, requested);
+
+	DBG("requested=%ld, actual=%ld", requested, actual);
+
+	if (actual != requested)
+		return MODE_CLOCK_RANGE;
+
+	return MODE_OK;
+}
+
 static const struct drm_encoder_helper_funcs mdp4_lcdc_encoder_helper_funcs = {
 	.mode_set = mdp4_lcdc_encoder_mode_set,
 	.disable = mdp4_lcdc_encoder_disable,
 	.enable = mdp4_lcdc_encoder_enable,
+	.mode_valid = mdp4_lcdc_encoder_mode_valid,
 };
 
-long mdp4_lcdc_round_pixclk(struct drm_encoder *encoder, unsigned long rate)
-{
-	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder =
-			to_mdp4_lcdc_encoder(encoder);
-	return clk_round_rate(mdp4_lcdc_encoder->lcdc_clk, rate);
-}
-
 /* initialize encoder */
-struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
-		struct device_node *panel_node)
+struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev)
 {
 	struct drm_encoder *encoder;
 	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder;
@@ -374,14 +371,11 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
 	if (IS_ERR(mdp4_lcdc_encoder))
 		return ERR_CAST(mdp4_lcdc_encoder);
 
-	mdp4_lcdc_encoder->panel_node = panel_node;
-
 	encoder = &mdp4_lcdc_encoder->base;
 
 	drm_encoder_helper_add(encoder, &mdp4_lcdc_encoder_helper_funcs);
 
-	/* TODO: do we need different pll in other cases? */
-	mdp4_lcdc_encoder->lcdc_clk = mpd4_lvds_pll_init(dev);
+	mdp4_lcdc_encoder->lcdc_clk = mpd4_get_lcdc_clock(dev);
 	if (IS_ERR(mdp4_lcdc_encoder->lcdc_clk)) {
 		DRM_DEV_ERROR(dev->dev, "failed to get lvds_clk\n");
 		return ERR_CAST(mdp4_lcdc_encoder->lcdc_clk);
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c
deleted file mode 100644
index 52e728181b52..000000000000
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c
+++ /dev/null
@@ -1,121 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * Copyright (C) 2014 Red Hat
- * Author: Rob Clark <robdclark@gmail.com>
- * Author: Vinay Simha <vinaysimha@inforcecomputing.com>
- */
-
-#include "mdp4_kms.h"
-
-struct mdp4_lvds_connector {
-	struct drm_connector base;
-	struct drm_encoder *encoder;
-	struct device_node *panel_node;
-	struct drm_panel *panel;
-};
-#define to_mdp4_lvds_connector(x) container_of(x, struct mdp4_lvds_connector, base)
-
-static enum drm_connector_status mdp4_lvds_connector_detect(
-		struct drm_connector *connector, bool force)
-{
-	struct mdp4_lvds_connector *mdp4_lvds_connector =
-			to_mdp4_lvds_connector(connector);
-
-	if (!mdp4_lvds_connector->panel) {
-		mdp4_lvds_connector->panel =
-			of_drm_find_panel(mdp4_lvds_connector->panel_node);
-		if (IS_ERR(mdp4_lvds_connector->panel))
-			mdp4_lvds_connector->panel = NULL;
-	}
-
-	return mdp4_lvds_connector->panel ?
-			connector_status_connected :
-			connector_status_disconnected;
-}
-
-static void mdp4_lvds_connector_destroy(struct drm_connector *connector)
-{
-	struct mdp4_lvds_connector *mdp4_lvds_connector =
-			to_mdp4_lvds_connector(connector);
-
-	drm_connector_cleanup(connector);
-
-	kfree(mdp4_lvds_connector);
-}
-
-static int mdp4_lvds_connector_get_modes(struct drm_connector *connector)
-{
-	struct mdp4_lvds_connector *mdp4_lvds_connector =
-			to_mdp4_lvds_connector(connector);
-	struct drm_panel *panel = mdp4_lvds_connector->panel;
-	int ret = 0;
-
-	if (panel)
-		ret = drm_panel_get_modes(panel, connector);
-
-	return ret;
-}
-
-static enum drm_mode_status
-mdp4_lvds_connector_mode_valid(struct drm_connector *connector,
-			       const struct drm_display_mode *mode)
-{
-	struct mdp4_lvds_connector *mdp4_lvds_connector =
-			to_mdp4_lvds_connector(connector);
-	struct drm_encoder *encoder = mdp4_lvds_connector->encoder;
-	long actual, requested;
-
-	requested = 1000 * mode->clock;
-	actual = mdp4_lcdc_round_pixclk(encoder, requested);
-
-	DBG("requested=%ld, actual=%ld", requested, actual);
-
-	if (actual != requested)
-		return MODE_CLOCK_RANGE;
-
-	return MODE_OK;
-}
-
-static const struct drm_connector_funcs mdp4_lvds_connector_funcs = {
-	.detect = mdp4_lvds_connector_detect,
-	.fill_modes = drm_helper_probe_single_connector_modes,
-	.destroy = mdp4_lvds_connector_destroy,
-	.reset = drm_atomic_helper_connector_reset,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-};
-
-static const struct drm_connector_helper_funcs mdp4_lvds_connector_helper_funcs = {
-	.get_modes = mdp4_lvds_connector_get_modes,
-	.mode_valid = mdp4_lvds_connector_mode_valid,
-};
-
-/* initialize connector */
-struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev,
-		struct device_node *panel_node, struct drm_encoder *encoder)
-{
-	struct drm_connector *connector = NULL;
-	struct mdp4_lvds_connector *mdp4_lvds_connector;
-
-	mdp4_lvds_connector = kzalloc(sizeof(*mdp4_lvds_connector), GFP_KERNEL);
-	if (!mdp4_lvds_connector)
-		return ERR_PTR(-ENOMEM);
-
-	mdp4_lvds_connector->encoder = encoder;
-	mdp4_lvds_connector->panel_node = panel_node;
-
-	connector = &mdp4_lvds_connector->base;
-
-	drm_connector_init(dev, connector, &mdp4_lvds_connector_funcs,
-			DRM_MODE_CONNECTOR_LVDS);
-	drm_connector_helper_add(connector, &mdp4_lvds_connector_helper_funcs);
-
-	connector->polled = 0;
-
-	connector->interlace_allowed = 0;
-	connector->doublescan_allowed = 0;
-
-	drm_connector_attach_encoder(connector, encoder);
-
-	return connector;
-}
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
index ab8c0c187fb2..fa2c29470510 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
@@ -122,40 +122,59 @@ static const struct clk_ops mpd4_lvds_pll_ops = {
 	.set_rate = mpd4_lvds_pll_set_rate,
 };
 
-static const char *mpd4_lvds_pll_parents[] = {
-	"pxo",
+static const struct clk_parent_data mpd4_lvds_pll_parents[] = {
+	{ .fw_name = "pxo", .name = "pxo", },
 };
 
 static struct clk_init_data pll_init = {
 	.name = "mpd4_lvds_pll",
 	.ops = &mpd4_lvds_pll_ops,
-	.parent_names = mpd4_lvds_pll_parents,
+	.parent_data = mpd4_lvds_pll_parents,
 	.num_parents = ARRAY_SIZE(mpd4_lvds_pll_parents),
 };
 
-struct clk *mpd4_lvds_pll_init(struct drm_device *dev)
+static struct clk_hw *mpd4_lvds_pll_init(struct drm_device *dev)
 {
 	struct mdp4_lvds_pll *lvds_pll;
-	struct clk *clk;
 	int ret;
 
 	lvds_pll = devm_kzalloc(dev->dev, sizeof(*lvds_pll), GFP_KERNEL);
-	if (!lvds_pll) {
-		ret = -ENOMEM;
-		goto fail;
-	}
+	if (!lvds_pll)
+		return ERR_PTR(-ENOMEM);
 
 	lvds_pll->dev = dev;
 
 	lvds_pll->pll_hw.init = &pll_init;
-	clk = devm_clk_register(dev->dev, &lvds_pll->pll_hw);
-	if (IS_ERR(clk)) {
-		ret = PTR_ERR(clk);
-		goto fail;
+	ret = devm_clk_hw_register(dev->dev, &lvds_pll->pll_hw);
+	if (ret)
+		return ERR_PTR(ret);
+
+	ret = devm_of_clk_add_hw_provider(dev->dev, of_clk_hw_simple_get, &lvds_pll->pll_hw);
+	if (ret)
+		return ERR_PTR(ret);
+
+	return &lvds_pll->pll_hw;
+}
+
+struct clk *mpd4_get_lcdc_clock(struct drm_device *dev)
+{
+	struct clk_hw *hw;
+	struct clk *clk;
+
+
+	/* TODO: do we need different pll in other cases? */
+	hw = mpd4_lvds_pll_init(dev);
+	if (IS_ERR(hw)) {
+		DRM_DEV_ERROR(dev->dev, "failed to register LVDS PLL\n");
+		return ERR_CAST(hw);
 	}
 
-	return clk;
+	clk = devm_clk_get(dev->dev, "lcdc_clk");
+	if (clk == ERR_PTR(-ENOENT)) {
+		drm_warn(dev, "can't get LCDC clock, using PLL directly\n");
 
-fail:
-	return ERR_PTR(ret);
+		return devm_clk_hw_get_clk(dev->dev, hw, "lcdc_clk");
+	}
+
+	return clk;
 }
diff --git a/drivers/gpu/drm/msm/dp/dp_audio.c b/drivers/gpu/drm/msm/dp/dp_audio.c
index 70fdc9fe228a..f8bfb908f9b4 100644
--- a/drivers/gpu/drm/msm/dp/dp_audio.c
+++ b/drivers/gpu/drm/msm/dp/dp_audio.c
@@ -13,13 +13,13 @@
 
 #include "dp_catalog.h"
 #include "dp_audio.h"
+#include "dp_drm.h"
 #include "dp_panel.h"
 #include "dp_reg.h"
 #include "dp_display.h"
 #include "dp_utils.h"
 
 struct msm_dp_audio_private {
-	struct platform_device *audio_pdev;
 	struct platform_device *pdev;
 	struct drm_device *drm_dev;
 	struct msm_dp_catalog *catalog;
@@ -160,24 +160,11 @@ static void msm_dp_audio_enable(struct msm_dp_audio_private *audio, bool enable)
 	msm_dp_catalog_audio_enable(catalog, enable);
 }
 
-static struct msm_dp_audio_private *msm_dp_audio_get_data(struct platform_device *pdev)
+static struct msm_dp_audio_private *msm_dp_audio_get_data(struct msm_dp *msm_dp_display)
 {
 	struct msm_dp_audio *msm_dp_audio;
-	struct msm_dp *msm_dp_display;
-
-	if (!pdev) {
-		DRM_ERROR("invalid input\n");
-		return ERR_PTR(-ENODEV);
-	}
-
-	msm_dp_display = platform_get_drvdata(pdev);
-	if (!msm_dp_display) {
-		DRM_ERROR("invalid input\n");
-		return ERR_PTR(-ENODEV);
-	}
 
 	msm_dp_audio = msm_dp_display->msm_dp_audio;
-
 	if (!msm_dp_audio) {
 		DRM_ERROR("invalid msm_dp_audio data\n");
 		return ERR_PTR(-EINVAL);
@@ -186,68 +173,16 @@ static struct msm_dp_audio_private *msm_dp_audio_get_data(struct platform_device
 	return container_of(msm_dp_audio, struct msm_dp_audio_private, msm_dp_audio);
 }
 
-static int msm_dp_audio_hook_plugged_cb(struct device *dev, void *data,
-		hdmi_codec_plugged_cb fn,
-		struct device *codec_dev)
-{
-
-	struct platform_device *pdev;
-	struct msm_dp *msm_dp_display;
-
-	pdev = to_platform_device(dev);
-	if (!pdev) {
-		pr_err("invalid input\n");
-		return -ENODEV;
-	}
-
-	msm_dp_display = platform_get_drvdata(pdev);
-	if (!msm_dp_display) {
-		pr_err("invalid input\n");
-		return -ENODEV;
-	}
-
-	return msm_dp_display_set_plugged_cb(msm_dp_display, fn, codec_dev);
-}
-
-static int msm_dp_audio_get_eld(struct device *dev,
-	void *data, uint8_t *buf, size_t len)
-{
-	struct platform_device *pdev;
-	struct msm_dp *msm_dp_display;
-
-	pdev = to_platform_device(dev);
-
-	if (!pdev) {
-		DRM_ERROR("invalid input\n");
-		return -ENODEV;
-	}
-
-	msm_dp_display = platform_get_drvdata(pdev);
-	if (!msm_dp_display) {
-		DRM_ERROR("invalid input\n");
-		return -ENODEV;
-	}
-
-	mutex_lock(&msm_dp_display->connector->eld_mutex);
-	memcpy(buf, msm_dp_display->connector->eld,
-		min(sizeof(msm_dp_display->connector->eld), len));
-	mutex_unlock(&msm_dp_display->connector->eld_mutex);
-
-	return 0;
-}
-
-int msm_dp_audio_hw_params(struct device *dev,
-	void *data,
-	struct hdmi_codec_daifmt *daifmt,
-	struct hdmi_codec_params *params)
+int msm_dp_audio_prepare(struct drm_connector *connector,
+			 struct drm_bridge *bridge,
+			 struct hdmi_codec_daifmt *daifmt,
+			 struct hdmi_codec_params *params)
 {
 	int rc = 0;
 	struct msm_dp_audio_private *audio;
-	struct platform_device *pdev;
 	struct msm_dp *msm_dp_display;
 
-	pdev = to_platform_device(dev);
-	msm_dp_display = platform_get_drvdata(pdev);
+	msm_dp_display = to_dp_bridge(bridge)->msm_dp_display;
 
 	/*
 	 * there could be cases where sound card can be opened even
@@ -262,7 +197,7 @@ int msm_dp_audio_hw_params(struct device *dev,
 		goto end;
 	}
 
-	audio = msm_dp_audio_get_data(pdev);
+	audio = msm_dp_audio_get_data(msm_dp_display);
 	if (IS_ERR(audio)) {
 		rc = PTR_ERR(audio);
 		goto end;
@@ -281,15 +216,14 @@ end:
 	return rc;
 }
 
-static void msm_dp_audio_shutdown(struct device *dev, void *data)
+void msm_dp_audio_shutdown(struct drm_connector *connector,
+			   struct drm_bridge *bridge)
 {
 	struct msm_dp_audio_private *audio;
-	struct platform_device *pdev;
 	struct msm_dp *msm_dp_display;
 
-	pdev = to_platform_device(dev);
-	msm_dp_display = platform_get_drvdata(pdev);
-	audio = msm_dp_audio_get_data(pdev);
+	msm_dp_display = to_dp_bridge(bridge)->msm_dp_display;
+	audio = msm_dp_audio_get_data(msm_dp_display);
 	if (IS_ERR(audio)) {
 		DRM_ERROR("failed to get audio data\n");
 		return;
@@ -311,47 +245,6 @@ static void msm_dp_audio_shutdown(struct device *dev, void *data)
 	msm_dp_display_signal_audio_complete(msm_dp_display);
 }
 
-static const struct hdmi_codec_ops msm_dp_audio_codec_ops = {
-	.hw_params = msm_dp_audio_hw_params,
-	.audio_shutdown = msm_dp_audio_shutdown,
-	.get_eld = msm_dp_audio_get_eld,
-	.hook_plugged_cb = msm_dp_audio_hook_plugged_cb,
-};
-
-static struct hdmi_codec_pdata codec_data = {
-	.ops = &msm_dp_audio_codec_ops,
-	.max_i2s_channels = 8,
-	.i2s = 1,
-};
-
-void msm_dp_unregister_audio_driver(struct device *dev, struct msm_dp_audio *msm_dp_audio)
-{
-	struct msm_dp_audio_private *audio_priv;
-
-	audio_priv = container_of(msm_dp_audio, struct msm_dp_audio_private, msm_dp_audio);
-
-	if (audio_priv->audio_pdev) {
-		platform_device_unregister(audio_priv->audio_pdev);
-		audio_priv->audio_pdev = NULL;
-	}
-}
-
-int msm_dp_register_audio_driver(struct device *dev,
-		struct msm_dp_audio *msm_dp_audio)
-{
-	struct msm_dp_audio_private *audio_priv;
-
-	audio_priv = container_of(msm_dp_audio,
-			struct msm_dp_audio_private, msm_dp_audio);
-
-	audio_priv->audio_pdev = platform_device_register_data(dev,
-						HDMI_CODEC_DRV_NAME,
-						PLATFORM_DEVID_AUTO,
-						&codec_data,
-						sizeof(codec_data));
-	return PTR_ERR_OR_ZERO(audio_priv->audio_pdev);
-}
-
 struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev,
 			struct msm_dp_catalog *catalog)
 {
diff --git a/drivers/gpu/drm/msm/dp/dp_audio.h b/drivers/gpu/drm/msm/dp/dp_audio.h
index beea34cbab77..58fc14693e48 100644
--- a/drivers/gpu/drm/msm/dp/dp_audio.h
+++ b/drivers/gpu/drm/msm/dp/dp_audio.h
@@ -36,23 +36,6 @@ struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev,
 			struct msm_dp_catalog *catalog);
 
 /**
- * msm_dp_register_audio_driver()
- *
- * Registers DP device with hdmi_codec interface.
- *
- * @dev: DP device instance.
- * @msm_dp_audio: an instance of msm_dp_audio module.
- *
- *
- * Returns the error code in case of failure, otherwise
- * zero on success.
- */
-int msm_dp_register_audio_driver(struct device *dev,
-		struct msm_dp_audio *msm_dp_audio);
-
-void msm_dp_unregister_audio_driver(struct device *dev, struct msm_dp_audio *msm_dp_audio);
-
-/**
  * msm_dp_audio_put()
  *
  * Cleans the msm_dp_audio instance.
@@ -61,10 +44,12 @@ void msm_dp_unregister_audio_driver(struct device *dev, struct msm_dp_audio *msm
  */
 void msm_dp_audio_put(struct msm_dp_audio *msm_dp_audio);
 
-int msm_dp_audio_hw_params(struct device *dev,
-	void *data,
-	struct hdmi_codec_daifmt *daifmt,
-	struct hdmi_codec_params *params);
+int msm_dp_audio_prepare(struct drm_connector *connector,
+			 struct drm_bridge *bridge,
+			 struct hdmi_codec_daifmt *daifmt,
+			 struct hdmi_codec_params *params);
+void msm_dp_audio_shutdown(struct drm_connector *connector,
+			   struct drm_bridge *bridge);
 
 #endif /* _DP_AUDIO_H_ */
 
diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c
index d8633a596f8d..a50bfafbb4ea 100644
--- a/drivers/gpu/drm/msm/dp/dp_ctrl.c
+++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c
@@ -1034,10 +1034,12 @@ static int msm_dp_ctrl_set_vx_px(struct msm_dp_ctrl_private *ctrl,
 	return 0;
 }
 
-static int msm_dp_ctrl_update_vx_px(struct msm_dp_ctrl_private *ctrl)
+static int msm_dp_ctrl_update_phy_vx_px(struct msm_dp_ctrl_private *ctrl,
+					enum drm_dp_phy dp_phy)
 {
 	struct msm_dp_link *link = ctrl->link;
-	int ret = 0, lane, lane_cnt;
+	int lane, lane_cnt, reg;
+	int ret = 0;
 	u8 buf[4];
 	u32 max_level_reached = 0;
 	u32 voltage_swing_level = link->phy_params.v_level;
@@ -1075,8 +1077,13 @@ static int msm_dp_ctrl_update_vx_px(struct msm_dp_ctrl_private *ctrl)
 
 	drm_dbg_dp(ctrl->drm_dev, "sink: p|v=0x%x\n",
 			voltage_swing_level | pre_emphasis_level);
-	ret = drm_dp_dpcd_write(ctrl->aux, DP_TRAINING_LANE0_SET,
-					buf, lane_cnt);
+
+	if (dp_phy == DP_PHY_DPRX)
+		reg = DP_TRAINING_LANE0_SET;
+	else
+		reg = DP_TRAINING_LANE0_SET_PHY_REPEATER(dp_phy);
+
+	ret = drm_dp_dpcd_write(ctrl->aux, reg, buf, lane_cnt);
 	if (ret == lane_cnt)
 		ret = 0;
 
@@ -1084,9 +1091,10 @@ static int msm_dp_ctrl_update_vx_px(struct msm_dp_ctrl_private *ctrl)
 }
 
 static bool msm_dp_ctrl_train_pattern_set(struct msm_dp_ctrl_private *ctrl,
-		u8 pattern)
+		u8 pattern, enum drm_dp_phy dp_phy)
 {
 	u8 buf;
+	int reg;
 	int ret = 0;
 
 	drm_dbg_dp(ctrl->drm_dev, "sink: pattern=%x\n", pattern);
@@ -1096,31 +1104,26 @@ static bool msm_dp_ctrl_train_pattern_set(struct msm_dp_ctrl_private *ctrl,
 	if (pattern && pattern != DP_TRAINING_PATTERN_4)
 		buf |= DP_LINK_SCRAMBLING_DISABLE;
 
-	ret = drm_dp_dpcd_writeb(ctrl->aux, DP_TRAINING_PATTERN_SET, buf);
-	return ret == 1;
-}
-
-static int msm_dp_ctrl_read_link_status(struct msm_dp_ctrl_private *ctrl,
-				    u8 *link_status)
-{
-	int ret = 0, len;
-
-	len = drm_dp_dpcd_read_link_status(ctrl->aux, link_status);
-	if (len != DP_LINK_STATUS_SIZE) {
-		DRM_ERROR("DP link status read failed, err: %d\n", len);
-		ret = -EINVAL;
-	}
+	if (dp_phy == DP_PHY_DPRX)
+		reg = DP_TRAINING_PATTERN_SET;
+	else
+		reg = DP_TRAINING_PATTERN_SET_PHY_REPEATER(dp_phy);
 
-	return ret;
+	ret = drm_dp_dpcd_writeb(ctrl->aux, reg, buf);
+	return ret == 1;
 }
 
 static int msm_dp_ctrl_link_train_1(struct msm_dp_ctrl_private *ctrl,
-			int *training_step)
+			int *training_step, enum drm_dp_phy dp_phy)
 {
+	int delay_us;
 	int tries, old_v_level, ret = 0;
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	int const maximum_retries = 4;
 
+	delay_us = drm_dp_read_clock_recovery_delay(ctrl->aux,
+						    ctrl->panel->dpcd, dp_phy, false);
+
 	msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0);
 
 	*training_step = DP_TRAINING_1;
@@ -1129,18 +1132,19 @@ static int msm_dp_ctrl_link_train_1(struct msm_dp_ctrl_private *ctrl,
 	if (ret)
 		return ret;
 	msm_dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_1 |
-		DP_LINK_SCRAMBLING_DISABLE);
+		DP_LINK_SCRAMBLING_DISABLE, dp_phy);
 
-	ret = msm_dp_ctrl_update_vx_px(ctrl);
+	msm_dp_link_reset_phy_params_vx_px(ctrl->link);
+	ret = msm_dp_ctrl_update_phy_vx_px(ctrl, dp_phy);
 	if (ret)
 		return ret;
 
 	tries = 0;
 	old_v_level = ctrl->link->phy_params.v_level;
 	for (tries = 0; tries < maximum_retries; tries++) {
-		drm_dp_link_train_clock_recovery_delay(ctrl->aux, ctrl->panel->dpcd);
+		fsleep(delay_us);
 
-		ret = msm_dp_ctrl_read_link_status(ctrl, link_status);
+		ret = drm_dp_dpcd_read_phy_link_status(ctrl->aux, dp_phy, link_status);
 		if (ret)
 			return ret;
 
@@ -1161,7 +1165,7 @@ static int msm_dp_ctrl_link_train_1(struct msm_dp_ctrl_private *ctrl,
 		}
 
 		msm_dp_link_adjust_levels(ctrl->link, link_status);
-		ret = msm_dp_ctrl_update_vx_px(ctrl);
+		ret = msm_dp_ctrl_update_phy_vx_px(ctrl, dp_phy);
 		if (ret)
 			return ret;
 	}
@@ -1213,21 +1217,31 @@ static int msm_dp_ctrl_link_lane_down_shift(struct msm_dp_ctrl_private *ctrl)
 	return 0;
 }
 
-static void msm_dp_ctrl_clear_training_pattern(struct msm_dp_ctrl_private *ctrl)
+static void msm_dp_ctrl_clear_training_pattern(struct msm_dp_ctrl_private *ctrl,
+					       enum drm_dp_phy dp_phy)
 {
-	msm_dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_DISABLE);
-	drm_dp_link_train_channel_eq_delay(ctrl->aux, ctrl->panel->dpcd);
+	int delay_us;
+
+	msm_dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_DISABLE, dp_phy);
+
+	delay_us = drm_dp_read_channel_eq_delay(ctrl->aux,
+						ctrl->panel->dpcd, dp_phy, false);
+	fsleep(delay_us);
 }
 
 static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl,
-			int *training_step)
+			int *training_step, enum drm_dp_phy dp_phy)
 {
+	int delay_us;
 	int tries = 0, ret = 0;
 	u8 pattern;
 	u32 state_ctrl_bit;
 	int const maximum_retries = 5;
 	u8 link_status[DP_LINK_STATUS_SIZE];
 
+	delay_us = drm_dp_read_channel_eq_delay(ctrl->aux,
+						ctrl->panel->dpcd, dp_phy, false);
+
 	msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0);
 
 	*training_step = DP_TRAINING_2;
@@ -1247,12 +1261,12 @@ static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl,
 	if (ret)
 		return ret;
 
-	msm_dp_ctrl_train_pattern_set(ctrl, pattern);
+	msm_dp_ctrl_train_pattern_set(ctrl, pattern, dp_phy);
 
 	for (tries = 0; tries <= maximum_retries; tries++) {
-		drm_dp_link_train_channel_eq_delay(ctrl->aux, ctrl->panel->dpcd);
+		fsleep(delay_us);
 
-		ret = msm_dp_ctrl_read_link_status(ctrl, link_status);
+		ret = drm_dp_dpcd_read_phy_link_status(ctrl->aux, dp_phy, link_status);
 		if (ret)
 			return ret;
 
@@ -1262,7 +1276,7 @@ static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl,
 		}
 
 		msm_dp_link_adjust_levels(ctrl->link, link_status);
-		ret = msm_dp_ctrl_update_vx_px(ctrl);
+		ret = msm_dp_ctrl_update_phy_vx_px(ctrl, dp_phy);
 		if (ret)
 			return ret;
 
@@ -1271,9 +1285,32 @@ static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl,
 	return -ETIMEDOUT;
 }
 
+static int msm_dp_ctrl_link_train_1_2(struct msm_dp_ctrl_private *ctrl,
+				      int *training_step, enum drm_dp_phy dp_phy)
+{
+	int ret;
+
+	ret = msm_dp_ctrl_link_train_1(ctrl, training_step, dp_phy);
+	if (ret) {
+		DRM_ERROR("link training #1 on phy %d failed. ret=%d\n", dp_phy, ret);
+		return ret;
+	}
+	drm_dbg_dp(ctrl->drm_dev, "link training #1 on phy %d successful\n", dp_phy);
+
+	ret = msm_dp_ctrl_link_train_2(ctrl, training_step, dp_phy);
+	if (ret) {
+		DRM_ERROR("link training #2 on phy %d failed. ret=%d\n", dp_phy, ret);
+		return ret;
+	}
+	drm_dbg_dp(ctrl->drm_dev, "link training #2 on phy %d successful\n", dp_phy);
+
+	return 0;
+}
+
 static int msm_dp_ctrl_link_train(struct msm_dp_ctrl_private *ctrl,
 			int *training_step)
 {
+	int i;
 	int ret = 0;
 	const u8 *dpcd = ctrl->panel->dpcd;
 	u8 encoding[] = { 0, DP_SET_ANSI_8B10B };
@@ -1286,8 +1323,6 @@ static int msm_dp_ctrl_link_train(struct msm_dp_ctrl_private *ctrl,
 	link_info.rate = ctrl->link->link_params.rate;
 	link_info.capabilities = DP_LINK_CAP_ENHANCED_FRAMING;
 
-	msm_dp_link_reset_phy_params_vx_px(ctrl->link);
-
 	msm_dp_aux_link_configure(ctrl->aux, &link_info);
 
 	if (drm_dp_max_downspread(dpcd))
@@ -1302,24 +1337,27 @@ static int msm_dp_ctrl_link_train(struct msm_dp_ctrl_private *ctrl,
 				&assr, 1);
 	}
 
-	ret = msm_dp_ctrl_link_train_1(ctrl, training_step);
+	for (i = ctrl->link->lttpr_count - 1; i >= 0; i--) {
+		enum drm_dp_phy dp_phy = DP_PHY_LTTPR(i);
+
+		ret = msm_dp_ctrl_link_train_1_2(ctrl, training_step, dp_phy);
+		msm_dp_ctrl_clear_training_pattern(ctrl, dp_phy);
+
+		if (ret)
+			break;
+	}
+
 	if (ret) {
-		DRM_ERROR("link training #1 failed. ret=%d\n", ret);
+		DRM_ERROR("link training of LTTPR(s) failed. ret=%d\n", ret);
 		goto end;
 	}
 
-	/* print success info as this is a result of user initiated action */
-	drm_dbg_dp(ctrl->drm_dev, "link training #1 successful\n");
-
-	ret = msm_dp_ctrl_link_train_2(ctrl, training_step);
+	ret = msm_dp_ctrl_link_train_1_2(ctrl, training_step, DP_PHY_DPRX);
 	if (ret) {
-		DRM_ERROR("link training #2 failed. ret=%d\n", ret);
+		DRM_ERROR("link training on sink failed. ret=%d\n", ret);
 		goto end;
 	}
 
-	/* print success info as this is a result of user initiated action */
-	drm_dbg_dp(ctrl->drm_dev, "link training #2 successful\n");
-
 end:
 	msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0);
 
@@ -1636,7 +1674,7 @@ static int msm_dp_ctrl_link_maintenance(struct msm_dp_ctrl_private *ctrl)
 	if (ret)
 		goto end;
 
-	msm_dp_ctrl_clear_training_pattern(ctrl);
+	msm_dp_ctrl_clear_training_pattern(ctrl, DP_PHY_DPRX);
 
 	msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_SEND_VIDEO);
 
@@ -1660,7 +1698,7 @@ static bool msm_dp_ctrl_send_phy_test_pattern(struct msm_dp_ctrl_private *ctrl)
 		return false;
 	}
 	msm_dp_catalog_ctrl_send_phy_pattern(ctrl->catalog, pattern_requested);
-	msm_dp_ctrl_update_vx_px(ctrl);
+	msm_dp_ctrl_update_phy_vx_px(ctrl, DP_PHY_DPRX);
 	msm_dp_link_send_test_response(ctrl->link);
 
 	pattern_sent = msm_dp_catalog_ctrl_read_phy_pattern(ctrl->catalog);
@@ -1805,7 +1843,7 @@ static bool msm_dp_ctrl_channel_eq_ok(struct msm_dp_ctrl_private *ctrl)
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	int num_lanes = ctrl->link->link_params.num_lanes;
 
-	msm_dp_ctrl_read_link_status(ctrl, link_status);
+	drm_dp_dpcd_read_link_status(ctrl->aux, link_status);
 
 	return drm_dp_channel_eq_ok(link_status, num_lanes);
 }
@@ -1863,7 +1901,7 @@ int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl)
 			if (!msm_dp_catalog_link_is_connected(ctrl->catalog))
 				break;
 
-			msm_dp_ctrl_read_link_status(ctrl, link_status);
+			drm_dp_dpcd_read_link_status(ctrl->aux, link_status);
 
 			rc = msm_dp_ctrl_link_rate_down_shift(ctrl);
 			if (rc < 0) { /* already in RBR = 1.6G */
@@ -1888,7 +1926,7 @@ int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl)
 			if (!msm_dp_catalog_link_is_connected(ctrl->catalog))
 				break;
 
-			msm_dp_ctrl_read_link_status(ctrl, link_status);
+			drm_dp_dpcd_read_link_status(ctrl->aux, link_status);
 
 			if (!drm_dp_clock_recovery_ok(link_status,
 					ctrl->link->link_params.num_lanes))
@@ -1902,7 +1940,7 @@ int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl)
 			}
 
 			/* stop link training before start re training  */
-			msm_dp_ctrl_clear_training_pattern(ctrl);
+			msm_dp_ctrl_clear_training_pattern(ctrl, DP_PHY_DPRX);
 		}
 
 		rc = msm_dp_ctrl_reinitialize_mainlink(ctrl);
@@ -1926,7 +1964,7 @@ int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl)
 		 * link training failed
 		 * end txing train pattern here
 		 */
-		msm_dp_ctrl_clear_training_pattern(ctrl);
+		msm_dp_ctrl_clear_training_pattern(ctrl, DP_PHY_DPRX);
 
 		msm_dp_ctrl_deinitialize_mainlink(ctrl);
 		rc = -ECONNRESET;
@@ -1997,7 +2035,7 @@ int msm_dp_ctrl_on_stream(struct msm_dp_ctrl *msm_dp_ctrl, bool force_link_train
 		msm_dp_ctrl_link_retrain(ctrl);
 
 	/* stop txing train pattern to end link training */
-	msm_dp_ctrl_clear_training_pattern(ctrl);
+	msm_dp_ctrl_clear_training_pattern(ctrl, DP_PHY_DPRX);
 
 	/*
 	 * Set up transfer unit values and set controller state to send
diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
index bbc47d86ae9e..386c4669c831 100644
--- a/drivers/gpu/drm/msm/dp/dp_display.c
+++ b/drivers/gpu/drm/msm/dp/dp_display.c
@@ -13,6 +13,7 @@
 #include <linux/delay.h>
 #include <linux/string_choices.h>
 #include <drm/display/drm_dp_aux_bus.h>
+#include <drm/display/drm_hdmi_audio_helper.h>
 #include <drm/drm_edid.h>
 
 #include "msm_drv.h"
@@ -288,13 +289,6 @@ static int msm_dp_display_bind(struct device *dev, struct device *master,
 		goto end;
 	}
 
-
-	rc = msm_dp_register_audio_driver(dev, dp->audio);
-	if (rc) {
-		DRM_ERROR("Audio registration Dp failed\n");
-		goto end;
-	}
-
 	rc = msm_dp_hpd_event_thread_start(dp);
 	if (rc) {
 		DRM_ERROR("Event thread create failed\n");
@@ -316,7 +310,6 @@ static void msm_dp_display_unbind(struct device *dev, struct device *master,
 
 	of_dp_aux_depopulate_bus(dp->aux);
 
-	msm_dp_unregister_audio_driver(dev, dp->audio);
 	msm_dp_aux_unregister(dp->aux);
 	dp->drm_dev = NULL;
 	dp->aux->drm_dev = NULL;
@@ -367,17 +360,21 @@ static int msm_dp_display_send_hpd_notification(struct msm_dp_display_private *d
 	return 0;
 }
 
-static void msm_dp_display_lttpr_init(struct msm_dp_display_private *dp)
+static int msm_dp_display_lttpr_init(struct msm_dp_display_private *dp, u8 *dpcd)
 {
-	u8 lttpr_caps[DP_LTTPR_COMMON_CAP_SIZE];
-	int rc;
+	int rc, lttpr_count;
 
-	if (drm_dp_read_lttpr_common_caps(dp->aux, dp->panel->dpcd, lttpr_caps))
-		return;
+	if (drm_dp_read_lttpr_common_caps(dp->aux, dpcd, dp->link->lttpr_common_caps))
+		return 0;
 
-	rc = drm_dp_lttpr_init(dp->aux, drm_dp_lttpr_count(lttpr_caps));
-	if (rc)
+	lttpr_count = drm_dp_lttpr_count(dp->link->lttpr_common_caps);
+	rc = drm_dp_lttpr_init(dp->aux, lttpr_count);
+	if (rc) {
 		DRM_ERROR("failed to set LTTPRs transparency mode, rc=%d\n", rc);
+		return 0;
+	}
+
+	return lttpr_count;
 }
 
 static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp)
@@ -385,12 +382,17 @@ static int msm_dp_display_process_hpd_high(struct msm_dp_display_private *dp)
 	struct drm_connector *connector = dp->msm_dp_display.connector;
 	const struct drm_display_info *info = &connector->display_info;
 	int rc = 0;
+	u8 dpcd[DP_RECEIVER_CAP_SIZE];
 
-	rc = msm_dp_panel_read_sink_caps(dp->panel, connector);
+	rc = drm_dp_read_dpcd_caps(dp->aux, dpcd);
 	if (rc)
 		goto end;
 
-	msm_dp_display_lttpr_init(dp);
+	dp->link->lttpr_count = msm_dp_display_lttpr_init(dp, dpcd);
+
+	rc = msm_dp_panel_read_sink_caps(dp->panel, connector);
+	if (rc)
+		goto end;
 
 	msm_dp_link_process_request(dp->link);
 
@@ -626,9 +628,9 @@ static void msm_dp_display_handle_plugged_change(struct msm_dp *msm_dp_display,
 			struct msm_dp_display_private, msm_dp_display);
 
 	/* notify audio subsystem only if sink supports audio */
-	if (msm_dp_display->plugged_cb && msm_dp_display->codec_dev &&
-			dp->audio_supported)
-		msm_dp_display->plugged_cb(msm_dp_display->codec_dev, plugged);
+	if (dp->audio_supported)
+		drm_connector_hdmi_audio_plugged_notify(msm_dp_display->connector,
+							plugged);
 }
 
 static int msm_dp_hpd_unplug_handle(struct msm_dp_display_private *dp, u32 data)
@@ -907,19 +909,6 @@ static int msm_dp_display_disable(struct msm_dp_display_private *dp)
 	return 0;
 }
 
-int msm_dp_display_set_plugged_cb(struct msm_dp *msm_dp_display,
-		hdmi_codec_plugged_cb fn, struct device *codec_dev)
-{
-	bool plugged;
-
-	msm_dp_display->plugged_cb = fn;
-	msm_dp_display->codec_dev = codec_dev;
-	plugged = msm_dp_display->link_ready;
-	msm_dp_display_handle_plugged_change(msm_dp_display, plugged);
-
-	return 0;
-}
-
 /**
  * msm_dp_bridge_mode_valid - callback to determine if specified mode is valid
  * @bridge: Pointer to drm bridge structure
diff --git a/drivers/gpu/drm/msm/dp/dp_display.h b/drivers/gpu/drm/msm/dp/dp_display.h
index ecbc2d92f546..cc6e2cab36e9 100644
--- a/drivers/gpu/drm/msm/dp/dp_display.h
+++ b/drivers/gpu/drm/msm/dp/dp_display.h
@@ -7,7 +7,6 @@
 #define _DP_DISPLAY_H_
 
 #include "dp_panel.h"
-#include <sound/hdmi-codec.h>
 #include "disp/msm_disp_snapshot.h"
 
 #define DP_MAX_PIXEL_CLK_KHZ	675000
@@ -15,7 +14,6 @@
 struct msm_dp {
 	struct drm_device *drm_dev;
 	struct platform_device *pdev;
-	struct device *codec_dev;
 	struct drm_connector *connector;
 	struct drm_bridge *next_bridge;
 	bool link_ready;
@@ -25,14 +23,10 @@ struct msm_dp {
 	bool is_edp;
 	bool internal_hpd;
 
-	hdmi_codec_plugged_cb plugged_cb;
-
 	struct msm_dp_audio *msm_dp_audio;
 	bool psr_supported;
 };
 
-int msm_dp_display_set_plugged_cb(struct msm_dp *msm_dp_display,
-		hdmi_codec_plugged_cb fn, struct device *codec_dev);
 int msm_dp_display_get_modes(struct msm_dp *msm_dp_display);
 bool msm_dp_display_check_video_test(struct msm_dp *msm_dp_display);
 int msm_dp_display_get_test_bpp(struct msm_dp *msm_dp_display);
diff --git a/drivers/gpu/drm/msm/dp/dp_drm.c b/drivers/gpu/drm/msm/dp/dp_drm.c
index cca57e56c906..f222d7ccaa88 100644
--- a/drivers/gpu/drm/msm/dp/dp_drm.c
+++ b/drivers/gpu/drm/msm/dp/dp_drm.c
@@ -12,6 +12,7 @@
 
 #include "msm_drv.h"
 #include "msm_kms.h"
+#include "dp_audio.h"
 #include "dp_drm.h"
 
 /**
@@ -114,6 +115,9 @@ static const struct drm_bridge_funcs msm_dp_bridge_ops = {
 	.hpd_disable  = msm_dp_bridge_hpd_disable,
 	.hpd_notify   = msm_dp_bridge_hpd_notify,
 	.debugfs_init = msm_dp_bridge_debugfs_init,
+
+	.dp_audio_prepare = msm_dp_audio_prepare,
+	.dp_audio_shutdown = msm_dp_audio_shutdown,
 };
 
 static int msm_edp_bridge_atomic_check(struct drm_bridge *drm_bridge,
@@ -296,14 +300,15 @@ int msm_dp_bridge_init(struct msm_dp *msm_dp_display, struct drm_device *dev,
 	struct msm_dp_bridge *msm_dp_bridge;
 	struct drm_bridge *bridge;
 
-	msm_dp_bridge = devm_kzalloc(dev->dev, sizeof(*msm_dp_bridge), GFP_KERNEL);
-	if (!msm_dp_bridge)
-		return -ENOMEM;
+	msm_dp_bridge = devm_drm_bridge_alloc(dev->dev, struct msm_dp_bridge, bridge,
+					      msm_dp_display->is_edp ? &msm_edp_bridge_ops :
+					      &msm_dp_bridge_ops);
+	if (IS_ERR(msm_dp_bridge))
+		return PTR_ERR(msm_dp_bridge);
 
 	msm_dp_bridge->msm_dp_display = msm_dp_display;
 
 	bridge = &msm_dp_bridge->bridge;
-	bridge->funcs = msm_dp_display->is_edp ? &msm_edp_bridge_ops : &msm_dp_bridge_ops;
 	bridge->type = msm_dp_display->connector_type;
 	bridge->ycbcr_420_allowed = yuv_supported;
 
@@ -320,9 +325,13 @@ int msm_dp_bridge_init(struct msm_dp *msm_dp_display, struct drm_device *dev,
 	 */
 	if (!msm_dp_display->is_edp) {
 		bridge->ops =
+			DRM_BRIDGE_OP_DP_AUDIO |
 			DRM_BRIDGE_OP_DETECT |
 			DRM_BRIDGE_OP_HPD |
 			DRM_BRIDGE_OP_MODES;
+		bridge->hdmi_audio_dev = &msm_dp_display->pdev->dev;
+		bridge->hdmi_audio_max_i2s_playback_channels = 8;
+		bridge->hdmi_audio_dai_port = -1;
 	}
 
 	rc = devm_drm_bridge_add(dev->dev, bridge);
diff --git a/drivers/gpu/drm/msm/dp/dp_link.c b/drivers/gpu/drm/msm/dp/dp_link.c
index 1a1fbb2d7d4f..92a9077959b3 100644
--- a/drivers/gpu/drm/msm/dp/dp_link.c
+++ b/drivers/gpu/drm/msm/dp/dp_link.c
@@ -714,21 +714,21 @@ end:
 
 static int msm_dp_link_parse_sink_status_field(struct msm_dp_link_private *link)
 {
-	int len;
+	int ret;
 
 	link->prev_sink_count = link->msm_dp_link.sink_count;
-	len = drm_dp_read_sink_count(link->aux);
-	if (len < 0) {
+	ret = drm_dp_read_sink_count(link->aux);
+	if (ret < 0) {
 		DRM_ERROR("DP parse sink count failed\n");
-		return len;
+		return ret;
 	}
-	link->msm_dp_link.sink_count = len;
+	link->msm_dp_link.sink_count = ret;
 
-	len = drm_dp_dpcd_read_link_status(link->aux,
-		link->link_status);
-	if (len < DP_LINK_STATUS_SIZE) {
+	ret = drm_dp_dpcd_read_link_status(link->aux,
+					   link->link_status);
+	if (ret < 0) {
 		DRM_ERROR("DP link status read failed\n");
-		return len;
+		return ret;
 	}
 
 	return msm_dp_link_parse_request(link);
diff --git a/drivers/gpu/drm/msm/dp/dp_link.h b/drivers/gpu/drm/msm/dp/dp_link.h
index 8db5d5698a97..ba47c6d19fbf 100644
--- a/drivers/gpu/drm/msm/dp/dp_link.h
+++ b/drivers/gpu/drm/msm/dp/dp_link.h
@@ -7,6 +7,7 @@
 #define _DP_LINK_H_
 
 #include "dp_aux.h"
+#include <drm/display/drm_dp_helper.h>
 
 #define DS_PORT_STATUS_CHANGED 0x200
 #define DP_TEST_BIT_DEPTH_UNKNOWN 0xFFFFFFFF
@@ -60,6 +61,9 @@ struct msm_dp_link_phy_params {
 };
 
 struct msm_dp_link {
+	u8 lttpr_common_caps[DP_LTTPR_COMMON_CAP_SIZE];
+	int lttpr_count;
+
 	u32 sink_request;
 	u32 test_response;
 
diff --git a/drivers/gpu/drm/msm/dp/dp_panel.c b/drivers/gpu/drm/msm/dp/dp_panel.c
index 92415bf8aa16..4e8ab75c771b 100644
--- a/drivers/gpu/drm/msm/dp/dp_panel.c
+++ b/drivers/gpu/drm/msm/dp/dp_panel.c
@@ -47,7 +47,7 @@ static void msm_dp_panel_read_psr_cap(struct msm_dp_panel_private *panel)
 
 static int msm_dp_panel_read_dpcd(struct msm_dp_panel *msm_dp_panel)
 {
-	int rc;
+	int rc, max_lttpr_lanes, max_lttpr_rate;
 	struct msm_dp_panel_private *panel;
 	struct msm_dp_link_info *link_info;
 	u8 *dpcd, major, minor;
@@ -75,6 +75,16 @@ static int msm_dp_panel_read_dpcd(struct msm_dp_panel *msm_dp_panel)
 	if (link_info->rate > msm_dp_panel->max_dp_link_rate)
 		link_info->rate = msm_dp_panel->max_dp_link_rate;
 
+	/* Limit data lanes from LTTPR capabilities, if any */
+	max_lttpr_lanes = drm_dp_lttpr_max_lane_count(panel->link->lttpr_common_caps);
+	if (max_lttpr_lanes && max_lttpr_lanes < link_info->num_lanes)
+		link_info->num_lanes = max_lttpr_lanes;
+
+	/* Limit link rate from LTTPR capabilities, if any */
+	max_lttpr_rate = drm_dp_lttpr_max_link_rate(panel->link->lttpr_common_caps);
+	if (max_lttpr_rate && max_lttpr_rate < link_info->rate)
+		link_info->rate = max_lttpr_rate;
+
 	drm_dbg_dp(panel->drm_dev, "version: %d.%d\n", major, minor);
 	drm_dbg_dp(panel->drm_dev, "link_rate=%d\n", link_info->rate);
 	drm_dbg_dp(panel->drm_dev, "lane_count=%d\n", link_info->num_lanes);
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
index 7754dcec33d0..7675558ae2e5 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
@@ -221,6 +221,22 @@ static const struct msm_dsi_config sc7280_dsi_cfg = {
 	},
 };
 
+static const struct regulator_bulk_data sa8775p_dsi_regulators[] = {
+	{ .supply = "vdda", .init_load_uA = 8300 },    /* 1.2 V */
+	{ .supply = "refgen" },
+};
+
+static const struct msm_dsi_config sa8775p_dsi_cfg = {
+	.io_offset = DSI_6G_REG_SHIFT,
+	.regulator_data = sa8775p_dsi_regulators,
+	.num_regulators = ARRAY_SIZE(sa8775p_dsi_regulators),
+	.bus_clk_names = dsi_v2_4_clk_names,
+	.num_bus_clks = ARRAY_SIZE(dsi_v2_4_clk_names),
+	.io_start = {
+		{ 0xae94000, 0xae96000 },
+	},
+};
+
 static const struct msm_dsi_host_cfg_ops msm_dsi_v2_host_ops = {
 	.link_clk_set_rate = dsi_link_clk_set_rate_v2,
 	.link_clk_enable = dsi_link_clk_enable_v2,
@@ -294,6 +310,8 @@ static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = {
 		&sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_5_0,
 		&sc7280_dsi_cfg, &msm_dsi_6g_v2_host_ops},
+	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_5_1,
+		&sa8775p_dsi_cfg, &msm_dsi_6g_v2_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_6_0,
 		&sdm845_dsi_cfg, &msm_dsi_6g_v2_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_7_0,
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
index 120cb65164c1..65b0705fac0e 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
@@ -27,6 +27,7 @@
 #define MSM_DSI_6G_VER_MINOR_V2_4_0	0x20040000
 #define MSM_DSI_6G_VER_MINOR_V2_4_1	0x20040001
 #define MSM_DSI_6G_VER_MINOR_V2_5_0	0x20050000
+#define MSM_DSI_6G_VER_MINOR_V2_5_1	0x20050001
 #define MSM_DSI_6G_VER_MINOR_V2_6_0	0x20060000
 #define MSM_DSI_6G_VER_MINOR_V2_7_0	0x20070000
 #define MSM_DSI_6G_VER_MINOR_V2_8_0	0x20080000
diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c
index 4fabb01345aa..ca400924d4ee 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_manager.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c
@@ -434,12 +434,13 @@ static enum drm_mode_status dsi_mgr_bridge_mode_valid(struct drm_bridge *bridge,
 }
 
 static int dsi_mgr_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	int id = dsi_mgr_bridge_get_id(bridge);
 	struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id);
 
-	return drm_bridge_attach(bridge->encoder, msm_dsi->next_bridge,
+	return drm_bridge_attach(encoder, msm_dsi->next_bridge,
 				 bridge, flags);
 }
 
@@ -461,15 +462,14 @@ int msm_dsi_manager_connector_init(struct msm_dsi *msm_dsi,
 	struct drm_connector *connector;
 	int ret;
 
-	dsi_bridge = devm_kzalloc(msm_dsi->dev->dev,
-				sizeof(*dsi_bridge), GFP_KERNEL);
-	if (!dsi_bridge)
-		return -ENOMEM;
+	dsi_bridge = devm_drm_bridge_alloc(msm_dsi->dev->dev, struct dsi_bridge, base,
+					   &dsi_mgr_bridge_funcs);
+	if (IS_ERR(dsi_bridge))
+		return PTR_ERR(dsi_bridge);
 
 	dsi_bridge->id = msm_dsi->id;
 
 	bridge = &dsi_bridge->base;
-	bridge->funcs = &dsi_mgr_bridge_funcs;
 
 	ret = devm_drm_bridge_add(msm_dsi->dev->dev, bridge);
 	if (ret)
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
index c0bcc6828963..5973d7325699 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -581,6 +581,10 @@ static const struct of_device_id dsi_phy_dt_match[] = {
 	  .data = &dsi_phy_7nm_cfgs },
 	{ .compatible = "qcom,dsi-phy-7nm-8150",
 	  .data = &dsi_phy_7nm_8150_cfgs },
+	{ .compatible = "qcom,sa8775p-dsi-phy-5nm",
+	  .data = &dsi_phy_5nm_8775p_cfgs },
+	{ .compatible = "qcom,sar2130p-dsi-phy-5nm",
+	  .data = &dsi_phy_5nm_sar2130p_cfgs },
 	{ .compatible = "qcom,sc7280-dsi-phy-7nm",
 	  .data = &dsi_phy_7nm_7280_cfgs },
 	{ .compatible = "qcom,sm6375-dsi-phy-7nm",
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
index 1925418d9999..7ea608f620fe 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
@@ -59,6 +59,8 @@ extern const struct msm_dsi_phy_cfg dsi_phy_7nm_8150_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_7nm_7280_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_5nm_8350_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_5nm_8450_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_5nm_8775p_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_5nm_sar2130p_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_4nm_8550_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_4nm_8650_cfgs;
 
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
index a92decbee5b5..c19890358b74 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -1147,6 +1147,10 @@ static const struct regulator_bulk_data dsi_phy_7nm_37750uA_regulators[] = {
 	{ .supply = "vdds", .init_load_uA = 37550 },
 };
 
+static const struct regulator_bulk_data dsi_phy_7nm_48000uA_regulators[] = {
+	{ .supply = "vdds", .init_load_uA = 48000 },
+};
+
 static const struct regulator_bulk_data dsi_phy_7nm_98000uA_regulators[] = {
 	{ .supply = "vdds", .init_load_uA = 98000 },
 };
@@ -1289,6 +1293,52 @@ const struct msm_dsi_phy_cfg dsi_phy_5nm_8450_cfgs = {
 	.quirks = DSI_PHY_7NM_QUIRK_V4_3,
 };
 
+const struct msm_dsi_phy_cfg dsi_phy_5nm_8775p_cfgs = {
+	.has_phy_lane = true,
+	.regulator_data = dsi_phy_7nm_48000uA_regulators,
+	.num_regulators = ARRAY_SIZE(dsi_phy_7nm_48000uA_regulators),
+	.ops = {
+		.enable = dsi_7nm_phy_enable,
+		.disable = dsi_7nm_phy_disable,
+		.pll_init = dsi_pll_7nm_init,
+		.save_pll_state = dsi_7nm_pll_save_state,
+		.restore_pll_state = dsi_7nm_pll_restore_state,
+		.set_continuous_clock = dsi_7nm_set_continuous_clock,
+		},
+	.min_pll_rate = 600000000UL,
+#ifdef CONFIG_64BIT
+	.max_pll_rate = 5000000000UL,
+#else
+	.max_pll_rate = ULONG_MAX,
+#endif
+	.io_start = { 0xae94400, 0xae96400 },
+	.num_dsi_phy = 2,
+	.quirks = DSI_PHY_7NM_QUIRK_V4_2,
+};
+
+const struct msm_dsi_phy_cfg dsi_phy_5nm_sar2130p_cfgs = {
+	.has_phy_lane = true,
+	.regulator_data = dsi_phy_7nm_97800uA_regulators,
+	.num_regulators = ARRAY_SIZE(dsi_phy_7nm_97800uA_regulators),
+	.ops = {
+		.enable = dsi_7nm_phy_enable,
+		.disable = dsi_7nm_phy_disable,
+		.pll_init = dsi_pll_7nm_init,
+		.save_pll_state = dsi_7nm_pll_save_state,
+		.restore_pll_state = dsi_7nm_pll_restore_state,
+		.set_continuous_clock = dsi_7nm_set_continuous_clock,
+	},
+	.min_pll_rate = 600000000UL,
+#ifdef CONFIG_64BIT
+	.max_pll_rate = 5000000000UL,
+#else
+	.max_pll_rate = ULONG_MAX,
+#endif
+	.io_start = { 0xae95000, 0xae97000 },
+	.num_dsi_phy = 2,
+	.quirks = DSI_PHY_7NM_QUIRK_V5_2,
+};
+
 const struct msm_dsi_phy_cfg dsi_phy_4nm_8550_cfgs = {
 	.has_phy_lane = true,
 	.regulator_data = dsi_phy_7nm_98400uA_regulators,
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index 248541ff4492..2fd388b892dc 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -8,6 +8,7 @@
 #include <linux/gpio/consumer.h>
 #include <linux/of_irq.h>
 #include <linux/of_platform.h>
+#include <linux/pinctrl/consumer.h>
 #include <linux/platform_device.h>
 
 #include <drm/drm_bridge_connector.h>
@@ -199,12 +200,6 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi,
 		goto fail;
 	}
 
-	ret = msm_hdmi_hpd_enable(hdmi->bridge);
-	if (ret < 0) {
-		DRM_DEV_ERROR(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret);
-		goto fail;
-	}
-
 	return 0;
 
 fail:
@@ -220,28 +215,24 @@ fail:
  * The hdmi device:
  */
 
-#define HDMI_CFG(item, entry) \
-	.item ## _names = item ##_names_ ## entry, \
-	.item ## _cnt   = ARRAY_SIZE(item ## _names_ ## entry)
-
-static const char *hpd_reg_names_8960[] = {"core-vdda"};
-static const char *hpd_clk_names_8960[] = {"core", "master_iface", "slave_iface"};
+static const char * const pwr_reg_names_8960[] = {"core-vdda"};
+static const char * const pwr_clk_names_8960[] = {"core", "master_iface", "slave_iface"};
 
 static const struct hdmi_platform_config hdmi_tx_8960_config = {
-		HDMI_CFG(hpd_reg, 8960),
-		HDMI_CFG(hpd_clk, 8960),
+	.pwr_reg_names = pwr_reg_names_8960,
+	.pwr_reg_cnt = ARRAY_SIZE(pwr_reg_names_8960),
+	.pwr_clk_names = pwr_clk_names_8960,
+	.pwr_clk_cnt = ARRAY_SIZE(pwr_clk_names_8960),
 };
 
-static const char *pwr_reg_names_8x74[] = {"core-vdda", "core-vcc"};
-static const char *pwr_clk_names_8x74[] = {"extp", "alt_iface"};
-static const char *hpd_clk_names_8x74[] = {"iface", "core", "mdp_core"};
-static unsigned long hpd_clk_freq_8x74[] = {0, 19200000, 0};
+static const char * const pwr_reg_names_8x74[] = {"core-vdda", "core-vcc"};
+static const char * const pwr_clk_names_8x74[] = {"iface", "core", "mdp_core", "alt_iface"};
 
 static const struct hdmi_platform_config hdmi_tx_8974_config = {
-		HDMI_CFG(pwr_reg, 8x74),
-		HDMI_CFG(pwr_clk, 8x74),
-		HDMI_CFG(hpd_clk, 8x74),
-		.hpd_freq      = hpd_clk_freq_8x74,
+	.pwr_reg_names = pwr_reg_names_8x74,
+	.pwr_reg_cnt = ARRAY_SIZE(pwr_reg_names_8x74),
+	.pwr_clk_names = pwr_clk_names_8x74,
+	.pwr_clk_cnt = ARRAY_SIZE(pwr_clk_names_8x74),
 };
 
 static int msm_hdmi_bind(struct device *dev, struct device *master, void *data)
@@ -264,9 +255,6 @@ static void msm_hdmi_unbind(struct device *dev, struct device *master,
 	struct msm_drm_private *priv = dev_get_drvdata(master);
 
 	if (priv->hdmi) {
-		if (priv->hdmi->bridge)
-			msm_hdmi_hpd_disable(priv->hdmi);
-
 		msm_hdmi_destroy(priv->hdmi);
 		priv->hdmi = NULL;
 	}
@@ -296,6 +284,7 @@ static int msm_hdmi_dev_probe(struct platform_device *pdev)
 	hdmi->pdev = pdev;
 	hdmi->config = config;
 	spin_lock_init(&hdmi->reg_lock);
+	mutex_init(&hdmi->state_mutex);
 
 	ret = drm_of_find_panel_or_bridge(pdev->dev.of_node, 1, 0, NULL, &hdmi->next_bridge);
 	if (ret && ret != -ENODEV)
@@ -322,20 +311,6 @@ static int msm_hdmi_dev_probe(struct platform_device *pdev)
 	if (hdmi->irq < 0)
 		return hdmi->irq;
 
-	hdmi->hpd_regs = devm_kcalloc(&pdev->dev,
-				      config->hpd_reg_cnt,
-				      sizeof(hdmi->hpd_regs[0]),
-				      GFP_KERNEL);
-	if (!hdmi->hpd_regs)
-		return -ENOMEM;
-
-	for (i = 0; i < config->hpd_reg_cnt; i++)
-		hdmi->hpd_regs[i].supply = config->hpd_reg_names[i];
-
-	ret = devm_regulator_bulk_get(&pdev->dev, config->hpd_reg_cnt, hdmi->hpd_regs);
-	if (ret)
-		return dev_err_probe(dev, ret, "failed to get hpd regulators\n");
-
 	hdmi->pwr_regs = devm_kcalloc(&pdev->dev,
 				      config->pwr_reg_cnt,
 				      sizeof(hdmi->pwr_regs[0]),
@@ -350,25 +325,6 @@ static int msm_hdmi_dev_probe(struct platform_device *pdev)
 	if (ret)
 		return dev_err_probe(dev, ret, "failed to get pwr regulators\n");
 
-	hdmi->hpd_clks = devm_kcalloc(&pdev->dev,
-				      config->hpd_clk_cnt,
-				      sizeof(hdmi->hpd_clks[0]),
-				      GFP_KERNEL);
-	if (!hdmi->hpd_clks)
-		return -ENOMEM;
-
-	for (i = 0; i < config->hpd_clk_cnt; i++) {
-		struct clk *clk;
-
-		clk = msm_clk_get(pdev, config->hpd_clk_names[i]);
-		if (IS_ERR(clk))
-			return dev_err_probe(dev, PTR_ERR(clk),
-					     "failed to get hpd clk: %s\n",
-					     config->hpd_clk_names[i]);
-
-		hdmi->hpd_clks[i] = clk;
-	}
-
 	hdmi->pwr_clks = devm_kcalloc(&pdev->dev,
 				      config->pwr_clk_cnt,
 				      sizeof(hdmi->pwr_clks[0]),
@@ -376,17 +332,17 @@ static int msm_hdmi_dev_probe(struct platform_device *pdev)
 	if (!hdmi->pwr_clks)
 		return -ENOMEM;
 
-	for (i = 0; i < config->pwr_clk_cnt; i++) {
-		struct clk *clk;
+	for (i = 0; i < config->pwr_clk_cnt; i++)
+		hdmi->pwr_clks[i].id = config->pwr_clk_names[i];
 
-		clk = msm_clk_get(pdev, config->pwr_clk_names[i]);
-		if (IS_ERR(clk))
-			return dev_err_probe(dev, PTR_ERR(clk),
-					     "failed to get pwr clk: %s\n",
-					     config->pwr_clk_names[i]);
+	ret = devm_clk_bulk_get(&pdev->dev, config->pwr_clk_cnt, hdmi->pwr_clks);
+	if (ret)
+		return ret;
 
-		hdmi->pwr_clks[i] = clk;
-	}
+	hdmi->extp_clk = devm_clk_get_optional(&pdev->dev, "extp");
+	if (IS_ERR(hdmi->extp_clk))
+		return dev_err_probe(dev, PTR_ERR(hdmi->extp_clk),
+				     "failed to get extp clock\n");
 
 	hdmi->hpd_gpiod = devm_gpiod_get_optional(&pdev->dev, "hpd", GPIOD_IN);
 	/* This will catch e.g. -EPROBE_DEFER */
@@ -432,6 +388,48 @@ static void msm_hdmi_dev_remove(struct platform_device *pdev)
 	msm_hdmi_put_phy(hdmi);
 }
 
+static int msm_hdmi_runtime_suspend(struct device *dev)
+{
+	struct hdmi *hdmi = dev_get_drvdata(dev);
+	const struct hdmi_platform_config *config = hdmi->config;
+
+	clk_bulk_disable_unprepare(config->pwr_clk_cnt, hdmi->pwr_clks);
+
+	pinctrl_pm_select_sleep_state(dev);
+
+	regulator_bulk_disable(config->pwr_reg_cnt, hdmi->pwr_regs);
+
+	return 0;
+}
+
+static int msm_hdmi_runtime_resume(struct device *dev)
+{
+	struct hdmi *hdmi = dev_get_drvdata(dev);
+	const struct hdmi_platform_config *config = hdmi->config;
+	int ret;
+
+	ret = regulator_bulk_enable(config->pwr_reg_cnt, hdmi->pwr_regs);
+	if (ret)
+		return ret;
+
+	ret = pinctrl_pm_select_default_state(dev);
+	if (ret)
+		goto fail;
+
+	ret = clk_bulk_prepare_enable(config->pwr_clk_cnt, hdmi->pwr_clks);
+	if (ret)
+		goto fail;
+
+	return 0;
+
+fail:
+	pinctrl_pm_select_sleep_state(dev);
+
+	return ret;
+}
+
+DEFINE_RUNTIME_DEV_PM_OPS(msm_hdmi_pm_ops, msm_hdmi_runtime_suspend, msm_hdmi_runtime_resume, NULL);
+
 static const struct of_device_id msm_hdmi_dt_match[] = {
 	{ .compatible = "qcom,hdmi-tx-8998", .data = &hdmi_tx_8974_config },
 	{ .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8974_config },
@@ -449,6 +447,7 @@ static struct platform_driver msm_hdmi_driver = {
 	.driver = {
 		.name = "hdmi_msm",
 		.of_match_table = msm_hdmi_dt_match,
+		.pm = &msm_hdmi_pm_ops,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.h b/drivers/gpu/drm/msm/hdmi/hdmi.h
index a5f481c39277..d5e572d10d6a 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.h
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.h
@@ -41,16 +41,17 @@ struct hdmi {
 
 	/* video state: */
 	bool power_on;
+	bool hpd_enabled;
+	struct mutex state_mutex; /* protects two booleans */
 	unsigned long int pixclock;
 
 	void __iomem *mmio;
 	void __iomem *qfprom_mmio;
 	phys_addr_t mmio_phy_addr;
 
-	struct regulator_bulk_data *hpd_regs;
 	struct regulator_bulk_data *pwr_regs;
-	struct clk **hpd_clks;
-	struct clk **pwr_clks;
+	struct clk_bulk_data *pwr_clks;
+	struct clk *extp_clk;
 
 	struct gpio_desc *hpd_gpiod;
 
@@ -83,21 +84,12 @@ struct hdmi {
 
 /* platform config data (ie. from DT, or pdata) */
 struct hdmi_platform_config {
-	/* regulators that need to be on for hpd: */
-	const char **hpd_reg_names;
-	int hpd_reg_cnt;
-
 	/* regulators that need to be on for screen pwr: */
-	const char **pwr_reg_names;
+	const char * const *pwr_reg_names;
 	int pwr_reg_cnt;
 
-	/* clks that need to be on for hpd: */
-	const char **hpd_clk_names;
-	const long unsigned *hpd_freq;
-	int hpd_clk_cnt;
-
-	/* clks that need to be on for screen pwr (ie pixel clk): */
-	const char **pwr_clk_names;
+	/* clks that need to be on: */
+	const char * const *pwr_clk_names;
 	int pwr_clk_cnt;
 };
 
@@ -224,8 +216,8 @@ int msm_hdmi_bridge_init(struct hdmi *hdmi);
 void msm_hdmi_hpd_irq(struct drm_bridge *bridge);
 enum drm_connector_status msm_hdmi_bridge_detect(
 		struct drm_bridge *bridge);
-int msm_hdmi_hpd_enable(struct drm_bridge *bridge);
-void msm_hdmi_hpd_disable(struct hdmi *hdmi);
+void msm_hdmi_hpd_enable(struct drm_bridge *bridge);
+void msm_hdmi_hpd_disable(struct drm_bridge *bridge);
 
 /*
  * i2c adapter for ddc:
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_audio.c b/drivers/gpu/drm/msm/hdmi/hdmi_audio.c
index 8bb975e82c17..b9ec14ef2c20 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_audio.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_audio.c
@@ -4,6 +4,7 @@
  * Author: Rob Clark <robdclark@gmail.com>
  */
 
+#include <drm/display/drm_hdmi_helper.h>
 #include <drm/display/drm_hdmi_state_helper.h>
 
 #include <linux/hdmi.h>
@@ -12,71 +13,9 @@
 
 #include "hdmi.h"
 
-/* Supported HDMI Audio sample rates */
-#define MSM_HDMI_SAMPLE_RATE_32KHZ		0
-#define MSM_HDMI_SAMPLE_RATE_44_1KHZ		1
-#define MSM_HDMI_SAMPLE_RATE_48KHZ		2
-#define MSM_HDMI_SAMPLE_RATE_88_2KHZ		3
-#define MSM_HDMI_SAMPLE_RATE_96KHZ		4
-#define MSM_HDMI_SAMPLE_RATE_176_4KHZ		5
-#define MSM_HDMI_SAMPLE_RATE_192KHZ		6
-#define MSM_HDMI_SAMPLE_RATE_MAX		7
-
-
-struct hdmi_msm_audio_acr {
-	uint32_t n;	/* N parameter for clock regeneration */
-	uint32_t cts;	/* CTS parameter for clock regeneration */
-};
-
-struct hdmi_msm_audio_arcs {
-	unsigned long int pixclock;
-	struct hdmi_msm_audio_acr lut[MSM_HDMI_SAMPLE_RATE_MAX];
-};
-
-#define HDMI_MSM_AUDIO_ARCS(pclk, ...) { (1000 * (pclk)), __VA_ARGS__ }
-
-/* Audio constants lookup table for hdmi_msm_audio_acr_setup */
-/* Valid Pixel-Clock rates: 25.2MHz, 27MHz, 27.03MHz, 74.25MHz, 148.5MHz */
-static const struct hdmi_msm_audio_arcs acr_lut[] = {
-	/*  25.200MHz  */
-	HDMI_MSM_AUDIO_ARCS(25200, {
-		{4096, 25200}, {6272, 28000}, {6144, 25200}, {12544, 28000},
-		{12288, 25200}, {25088, 28000}, {24576, 25200} }),
-	/*  27.000MHz  */
-	HDMI_MSM_AUDIO_ARCS(27000, {
-		{4096, 27000}, {6272, 30000}, {6144, 27000}, {12544, 30000},
-		{12288, 27000}, {25088, 30000}, {24576, 27000} }),
-	/*  27.027MHz */
-	HDMI_MSM_AUDIO_ARCS(27030, {
-		{4096, 27027}, {6272, 30030}, {6144, 27027}, {12544, 30030},
-		{12288, 27027}, {25088, 30030}, {24576, 27027} }),
-	/*  74.250MHz */
-	HDMI_MSM_AUDIO_ARCS(74250, {
-		{4096, 74250}, {6272, 82500}, {6144, 74250}, {12544, 82500},
-		{12288, 74250}, {25088, 82500}, {24576, 74250} }),
-	/* 148.500MHz */
-	HDMI_MSM_AUDIO_ARCS(148500, {
-		{4096, 148500}, {6272, 165000}, {6144, 148500}, {12544, 165000},
-		{12288, 148500}, {25088, 165000}, {24576, 148500} }),
-};
-
-static const struct hdmi_msm_audio_arcs *get_arcs(unsigned long int pixclock)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(acr_lut); i++) {
-		const struct hdmi_msm_audio_arcs *arcs = &acr_lut[i];
-		if (arcs->pixclock == pixclock)
-			return arcs;
-	}
-
-	return NULL;
-}
-
 int msm_hdmi_audio_update(struct hdmi *hdmi)
 {
 	struct hdmi_audio *audio = &hdmi->audio;
-	const struct hdmi_msm_audio_arcs *arcs = NULL;
 	bool enabled = audio->enabled;
 	uint32_t acr_pkt_ctrl, vbi_pkt_ctrl, aud_pkt_ctrl;
 	uint32_t audio_config;
@@ -94,15 +33,6 @@ int msm_hdmi_audio_update(struct hdmi *hdmi)
 		enabled = false;
 	}
 
-	if (enabled) {
-		arcs = get_arcs(hdmi->pixclock);
-		if (!arcs) {
-			DBG("disabling audio: unsupported pixclock: %lu",
-					hdmi->pixclock);
-			enabled = false;
-		}
-	}
-
 	/* Read first before writing */
 	acr_pkt_ctrl = hdmi_read(hdmi, REG_HDMI_ACR_PKT_CTRL);
 	vbi_pkt_ctrl = hdmi_read(hdmi, REG_HDMI_VBI_PKT_CTRL);
@@ -116,15 +46,12 @@ int msm_hdmi_audio_update(struct hdmi *hdmi)
 		uint32_t n, cts, multiplier;
 		enum hdmi_acr_cts select;
 
-		n   = arcs->lut[audio->rate].n;
-		cts = arcs->lut[audio->rate].cts;
+		drm_hdmi_acr_get_n_cts(hdmi->pixclock, audio->rate, &n, &cts);
 
-		if ((MSM_HDMI_SAMPLE_RATE_192KHZ == audio->rate) ||
-				(MSM_HDMI_SAMPLE_RATE_176_4KHZ == audio->rate)) {
+		if (audio->rate == 192000 || audio->rate == 176400) {
 			multiplier = 4;
 			n >>= 2; /* divide N by 4 and use multiplier */
-		} else if ((MSM_HDMI_SAMPLE_RATE_96KHZ == audio->rate) ||
-				(MSM_HDMI_SAMPLE_RATE_88_2KHZ == audio->rate)) {
+		} else if (audio->rate == 96000 || audio->rate == 88200) {
 			multiplier = 2;
 			n >>= 1; /* divide N by 2 and use multiplier */
 		} else {
@@ -137,13 +64,11 @@ int msm_hdmi_audio_update(struct hdmi *hdmi)
 		acr_pkt_ctrl |= HDMI_ACR_PKT_CTRL_AUDIO_PRIORITY;
 		acr_pkt_ctrl |= HDMI_ACR_PKT_CTRL_N_MULTIPLIER(multiplier);
 
-		if ((MSM_HDMI_SAMPLE_RATE_48KHZ == audio->rate) ||
-				(MSM_HDMI_SAMPLE_RATE_96KHZ == audio->rate) ||
-				(MSM_HDMI_SAMPLE_RATE_192KHZ == audio->rate))
+		if (audio->rate == 48000 || audio->rate == 96000 ||
+		    audio->rate == 192000)
 			select = ACR_48;
-		else if ((MSM_HDMI_SAMPLE_RATE_44_1KHZ == audio->rate) ||
-				(MSM_HDMI_SAMPLE_RATE_88_2KHZ == audio->rate) ||
-				(MSM_HDMI_SAMPLE_RATE_176_4KHZ == audio->rate))
+		else if (audio->rate == 44100 || audio->rate == 88200 ||
+			 audio->rate == 176400)
 			select = ACR_44;
 		else /* default to 32k */
 			select = ACR_32;
@@ -204,7 +129,6 @@ int msm_hdmi_bridge_audio_prepare(struct drm_connector *connector,
 {
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
-	unsigned int rate;
 	int ret;
 
 	drm_dbg_driver(bridge->dev, "%u Hz, %d bit, %d channels\n",
@@ -214,25 +138,12 @@ int msm_hdmi_bridge_audio_prepare(struct drm_connector *connector,
 
 	switch (params->sample_rate) {
 	case 32000:
-		rate = MSM_HDMI_SAMPLE_RATE_32KHZ;
-		break;
 	case 44100:
-		rate = MSM_HDMI_SAMPLE_RATE_44_1KHZ;
-		break;
 	case 48000:
-		rate = MSM_HDMI_SAMPLE_RATE_48KHZ;
-		break;
 	case 88200:
-		rate = MSM_HDMI_SAMPLE_RATE_88_2KHZ;
-		break;
 	case 96000:
-		rate = MSM_HDMI_SAMPLE_RATE_96KHZ;
-		break;
 	case 176400:
-		rate = MSM_HDMI_SAMPLE_RATE_176_4KHZ;
-		break;
 	case 192000:
-		rate = MSM_HDMI_SAMPLE_RATE_192KHZ;
 		break;
 	default:
 		drm_err(bridge->dev, "rate[%d] not supported!\n",
@@ -245,7 +156,7 @@ int msm_hdmi_bridge_audio_prepare(struct drm_connector *connector,
 	if (ret)
 		return ret;
 
-	hdmi->audio.rate = rate;
+	hdmi->audio.rate = params->sample_rate;
 	hdmi->audio.channels = params->cea.channels;
 	hdmi->audio.enabled = true;
 
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
index 1456354c8af4..53a7ce8cc7bc 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
@@ -18,52 +18,34 @@ static void msm_hdmi_power_on(struct drm_bridge *bridge)
 	struct drm_device *dev = bridge->dev;
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
-	const struct hdmi_platform_config *config = hdmi->config;
-	int i, ret;
-
-	pm_runtime_get_sync(&hdmi->pdev->dev);
+	int ret;
 
-	ret = regulator_bulk_enable(config->pwr_reg_cnt, hdmi->pwr_regs);
-	if (ret)
-		DRM_DEV_ERROR(dev->dev, "failed to enable pwr regulator: %d\n", ret);
+	pm_runtime_resume_and_get(&hdmi->pdev->dev);
 
-	if (config->pwr_clk_cnt > 0) {
+	if (hdmi->extp_clk) {
 		DBG("pixclock: %lu", hdmi->pixclock);
-		ret = clk_set_rate(hdmi->pwr_clks[0], hdmi->pixclock);
-		if (ret) {
-			DRM_DEV_ERROR(dev->dev, "failed to set pixel clk: %s (%d)\n",
-					config->pwr_clk_names[0], ret);
-		}
-	}
+		ret = clk_set_rate(hdmi->extp_clk, hdmi->pixclock);
+		if (ret)
+			DRM_DEV_ERROR(dev->dev, "failed to set extp clk rate: %d\n", ret);
 
-	for (i = 0; i < config->pwr_clk_cnt; i++) {
-		ret = clk_prepare_enable(hdmi->pwr_clks[i]);
-		if (ret) {
-			DRM_DEV_ERROR(dev->dev, "failed to enable pwr clk: %s (%d)\n",
-					config->pwr_clk_names[i], ret);
-		}
+		ret = clk_prepare_enable(hdmi->extp_clk);
+		if (ret)
+			DRM_DEV_ERROR(dev->dev, "failed to enable extp clk: %d\n", ret);
 	}
 }
 
 static void power_off(struct drm_bridge *bridge)
 {
-	struct drm_device *dev = bridge->dev;
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
-	const struct hdmi_platform_config *config = hdmi->config;
-	int i, ret;
 
 	/* TODO do we need to wait for final vblank somewhere before
 	 * cutting the clocks?
 	 */
 	mdelay(16 + 4);
 
-	for (i = 0; i < config->pwr_clk_cnt; i++)
-		clk_disable_unprepare(hdmi->pwr_clks[i]);
-
-	ret = regulator_bulk_disable(config->pwr_reg_cnt, hdmi->pwr_regs);
-	if (ret)
-		DRM_DEV_ERROR(dev->dev, "failed to disable pwr regulator: %d\n", ret);
+	if (hdmi->extp_clk)
+		clk_disable_unprepare(hdmi->extp_clk);
 
 	pm_runtime_put(&hdmi->pdev->dev);
 }
@@ -320,13 +302,16 @@ static void msm_hdmi_bridge_atomic_pre_enable(struct drm_bridge *bridge,
 
 	msm_hdmi_set_timings(hdmi, &crtc_state->adjusted_mode);
 
+	mutex_lock(&hdmi->state_mutex);
 	if (!hdmi->power_on) {
 		msm_hdmi_phy_resource_enable(phy);
 		msm_hdmi_power_on(bridge);
 		hdmi->power_on = true;
-		if (connector->display_info.is_hdmi)
-			msm_hdmi_audio_update(hdmi);
 	}
+	mutex_unlock(&hdmi->state_mutex);
+
+	if (connector->display_info.is_hdmi)
+		msm_hdmi_audio_update(hdmi);
 
 	drm_atomic_helper_connector_hdmi_update_infoframes(connector, state);
 
@@ -349,7 +334,10 @@ static void msm_hdmi_bridge_atomic_post_disable(struct drm_bridge *bridge,
 		msm_hdmi_hdcp_off(hdmi->hdcp_ctrl);
 
 	DBG("power down");
-	msm_hdmi_set_mode(hdmi, false);
+
+	/* Keep the HDMI enabled if the HPD is enabled */
+	mutex_lock(&hdmi->state_mutex);
+	msm_hdmi_set_mode(hdmi, hdmi->hpd_enabled);
 
 	msm_hdmi_phy_powerdown(phy);
 
@@ -360,6 +348,7 @@ static void msm_hdmi_bridge_atomic_post_disable(struct drm_bridge *bridge,
 			msm_hdmi_audio_update(hdmi);
 		msm_hdmi_phy_resource_disable(phy);
 	}
+	mutex_unlock(&hdmi->state_mutex);
 }
 
 static void msm_hdmi_set_timings(struct hdmi *hdmi,
@@ -411,9 +400,6 @@ static void msm_hdmi_set_timings(struct hdmi *hdmi,
 		frame_ctrl |= HDMI_FRAME_CTRL_INTERLACED_EN;
 	DBG("frame_ctrl=%08x", frame_ctrl);
 	hdmi_write(hdmi, REG_HDMI_FRAME_CTRL, frame_ctrl);
-
-	if (hdmi->connector->display_info.is_hdmi)
-		msm_hdmi_audio_update(hdmi);
 }
 
 static const struct drm_edid *msm_hdmi_bridge_edid_read(struct drm_bridge *bridge,
@@ -440,7 +426,6 @@ static enum drm_mode_status msm_hdmi_bridge_tmds_char_rate_valid(const struct dr
 {
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
-	const struct hdmi_platform_config *config = hdmi->config;
 	struct msm_drm_private *priv = bridge->dev->dev_private;
 	struct msm_kms *kms = priv->kms;
 	long actual;
@@ -453,8 +438,8 @@ static enum drm_mode_status msm_hdmi_bridge_tmds_char_rate_valid(const struct dr
 		actual = kms->funcs->round_pixclk(kms,
 						  tmds_rate,
 						  hdmi_bridge->hdmi->encoder);
-	else if (config->pwr_clk_cnt > 0)
-		actual = clk_round_rate(hdmi->pwr_clks[0], tmds_rate);
+	else if (hdmi->extp_clk)
+		actual = clk_round_rate(hdmi->extp_clk, tmds_rate);
 	else
 		actual = tmds_rate;
 
@@ -474,6 +459,8 @@ static const struct drm_bridge_funcs msm_hdmi_bridge_funcs = {
 	.atomic_post_disable = msm_hdmi_bridge_atomic_post_disable,
 	.edid_read = msm_hdmi_bridge_edid_read,
 	.detect = msm_hdmi_bridge_detect,
+	.hpd_enable = msm_hdmi_hpd_enable,
+	.hpd_disable = msm_hdmi_hpd_disable,
 	.hdmi_tmds_char_rate_valid = msm_hdmi_bridge_tmds_char_rate_valid,
 	.hdmi_clear_infoframe = msm_hdmi_bridge_clear_infoframe,
 	.hdmi_write_infoframe = msm_hdmi_bridge_write_infoframe,
@@ -498,16 +485,15 @@ int msm_hdmi_bridge_init(struct hdmi *hdmi)
 	struct hdmi_bridge *hdmi_bridge;
 	int ret;
 
-	hdmi_bridge = devm_kzalloc(hdmi->dev->dev,
-			sizeof(*hdmi_bridge), GFP_KERNEL);
-	if (!hdmi_bridge)
-		return -ENOMEM;
+	hdmi_bridge = devm_drm_bridge_alloc(hdmi->dev->dev, struct hdmi_bridge, base,
+					    &msm_hdmi_bridge_funcs);
+	if (IS_ERR(hdmi_bridge))
+		return PTR_ERR(hdmi_bridge);
 
 	hdmi_bridge->hdmi = hdmi;
 	INIT_WORK(&hdmi_bridge->hpd_work, msm_hdmi_hotplug_work);
 
 	bridge = &hdmi_bridge->base;
-	bridge->funcs = &msm_hdmi_bridge_funcs;
 	bridge->ddc = hdmi->i2c;
 	bridge->type = DRM_MODE_CONNECTOR_HDMIA;
 	bridge->vendor = "Qualcomm";
@@ -515,6 +501,7 @@ int msm_hdmi_bridge_init(struct hdmi *hdmi)
 	bridge->ops = DRM_BRIDGE_OP_HPD |
 		DRM_BRIDGE_OP_DETECT |
 		DRM_BRIDGE_OP_HDMI |
+		DRM_BRIDGE_OP_HDMI_AUDIO |
 		DRM_BRIDGE_OP_EDID;
 	bridge->hdmi_audio_max_i2s_playback_channels = 8;
 	bridge->hdmi_audio_dev = &hdmi->pdev->dev;
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_hpd.c b/drivers/gpu/drm/msm/hdmi/hdmi_hpd.c
index 9ce0ffa35417..407e6c449ee0 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_hpd.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_hpd.c
@@ -60,68 +60,30 @@ static void msm_hdmi_phy_reset(struct hdmi *hdmi)
 	}
 }
 
-static void enable_hpd_clocks(struct hdmi *hdmi, bool enable)
-{
-	const struct hdmi_platform_config *config = hdmi->config;
-	struct device *dev = &hdmi->pdev->dev;
-	int i, ret;
-
-	if (enable) {
-		for (i = 0; i < config->hpd_clk_cnt; i++) {
-			if (config->hpd_freq && config->hpd_freq[i]) {
-				ret = clk_set_rate(hdmi->hpd_clks[i],
-						   config->hpd_freq[i]);
-				if (ret)
-					dev_warn(dev,
-						 "failed to set clk %s (%d)\n",
-						 config->hpd_clk_names[i], ret);
-			}
-
-			ret = clk_prepare_enable(hdmi->hpd_clks[i]);
-			if (ret) {
-				DRM_DEV_ERROR(dev,
-					"failed to enable hpd clk: %s (%d)\n",
-					config->hpd_clk_names[i], ret);
-			}
-		}
-	} else {
-		for (i = config->hpd_clk_cnt - 1; i >= 0; i--)
-			clk_disable_unprepare(hdmi->hpd_clks[i]);
-	}
-}
-
-int msm_hdmi_hpd_enable(struct drm_bridge *bridge)
+void msm_hdmi_hpd_enable(struct drm_bridge *bridge)
 {
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
-	const struct hdmi_platform_config *config = hdmi->config;
 	struct device *dev = &hdmi->pdev->dev;
 	uint32_t hpd_ctrl;
 	int ret;
 	unsigned long flags;
 
-	ret = regulator_bulk_enable(config->hpd_reg_cnt, hdmi->hpd_regs);
-	if (ret) {
-		DRM_DEV_ERROR(dev, "failed to enable hpd regulators: %d\n", ret);
-		goto fail;
-	}
-
-	ret = pinctrl_pm_select_default_state(dev);
-	if (ret) {
-		DRM_DEV_ERROR(dev, "pinctrl state chg failed: %d\n", ret);
-		goto fail;
-	}
-
 	if (hdmi->hpd_gpiod)
 		gpiod_set_value_cansleep(hdmi->hpd_gpiod, 1);
 
-	pm_runtime_get_sync(dev);
-	enable_hpd_clocks(hdmi, true);
+	ret = pm_runtime_resume_and_get(dev);
+	if (WARN_ON(ret))
+		return;
 
+	mutex_lock(&hdmi->state_mutex);
 	msm_hdmi_set_mode(hdmi, false);
 	msm_hdmi_phy_reset(hdmi);
 	msm_hdmi_set_mode(hdmi, true);
 
+	hdmi->hpd_enabled = true;
+	mutex_unlock(&hdmi->state_mutex);
+
 	hdmi_write(hdmi, REG_HDMI_USEC_REFTIMER, 0x0001001b);
 
 	/* enable HPD events: */
@@ -140,34 +102,23 @@ int msm_hdmi_hpd_enable(struct drm_bridge *bridge)
 	hdmi_write(hdmi, REG_HDMI_HPD_CTRL,
 			HDMI_HPD_CTRL_ENABLE | hpd_ctrl);
 	spin_unlock_irqrestore(&hdmi->reg_lock, flags);
-
-	return 0;
-
-fail:
-	return ret;
 }
 
-void msm_hdmi_hpd_disable(struct hdmi *hdmi)
+void msm_hdmi_hpd_disable(struct drm_bridge *bridge)
 {
-	const struct hdmi_platform_config *config = hdmi->config;
+	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
+	struct hdmi *hdmi = hdmi_bridge->hdmi;
 	struct device *dev = &hdmi->pdev->dev;
-	int ret;
 
 	/* Disable HPD interrupt */
 	hdmi_write(hdmi, REG_HDMI_HPD_INT_CTRL, 0);
 
-	msm_hdmi_set_mode(hdmi, false);
+	mutex_lock(&hdmi->state_mutex);
+	hdmi->hpd_enabled = false;
+	msm_hdmi_set_mode(hdmi, hdmi->power_on);
+	mutex_unlock(&hdmi->state_mutex);
 
-	enable_hpd_clocks(hdmi, false);
 	pm_runtime_put(dev);
-
-	ret = pinctrl_pm_select_sleep_state(dev);
-	if (ret)
-		dev_warn(dev, "pinctrl state chg failed: %d\n", ret);
-
-	ret = regulator_bulk_disable(config->hpd_reg_cnt, hdmi->hpd_regs);
-	if (ret)
-		dev_warn(dev, "failed to disable hpd regulator: %d\n", ret);
 }
 
 void msm_hdmi_hpd_irq(struct drm_bridge *bridge)
@@ -202,14 +153,16 @@ void msm_hdmi_hpd_irq(struct drm_bridge *bridge)
 
 static enum drm_connector_status detect_reg(struct hdmi *hdmi)
 {
-	uint32_t hpd_int_status;
+	u32 hpd_int_status = 0;
+	int ret;
 
-	pm_runtime_get_sync(&hdmi->pdev->dev);
-	enable_hpd_clocks(hdmi, true);
+	ret = pm_runtime_resume_and_get(&hdmi->pdev->dev);
+	if (ret)
+		goto out;
 
 	hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
 
-	enable_hpd_clocks(hdmi, false);
+out:
 	pm_runtime_put(&hdmi->pdev->dev);
 
 	return (hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED) ?
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
index 7aa500d24240..ebefea4fb408 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
@@ -107,11 +107,15 @@ static int msm_hdmi_i2c_xfer(struct i2c_adapter *i2c,
 	if (num == 0)
 		return num;
 
+	ret = pm_runtime_resume_and_get(&hdmi->pdev->dev);
+	if (ret)
+		return ret;
+
 	init_ddc(hdmi_i2c);
 
 	ret = ddc_clear_irq(hdmi_i2c);
 	if (ret)
-		return ret;
+		goto fail;
 
 	for (i = 0; i < num; i++) {
 		struct i2c_msg *p = &msgs[i];
@@ -169,7 +173,7 @@ static int msm_hdmi_i2c_xfer(struct i2c_adapter *i2c,
 				hdmi_read(hdmi, REG_HDMI_DDC_SW_STATUS),
 				hdmi_read(hdmi, REG_HDMI_DDC_HW_STATUS),
 				hdmi_read(hdmi, REG_HDMI_DDC_INT_CTRL));
-		return ret;
+		goto fail;
 	}
 
 	ddc_status = hdmi_read(hdmi, REG_HDMI_DDC_SW_STATUS);
@@ -202,7 +206,13 @@ static int msm_hdmi_i2c_xfer(struct i2c_adapter *i2c,
 		}
 	}
 
+	pm_runtime_put(&hdmi->pdev->dev);
+
 	return i;
+
+fail:
+	pm_runtime_put(&hdmi->pdev->dev);
+	return ret;
 }
 
 static u32 msm_hdmi_i2c_func(struct i2c_adapter *adapter)
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy.c
index 03120c54ced6..667573f1db7c 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_phy.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy.c
@@ -58,7 +58,11 @@ int msm_hdmi_phy_resource_enable(struct hdmi_phy *phy)
 	struct device *dev = &phy->pdev->dev;
 	int i, ret = 0;
 
-	pm_runtime_get_sync(dev);
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret) {
+		DRM_DEV_ERROR(dev, "runtime resume failed: %d\n", ret);
+		return ret;
+	}
 
 	ret = regulator_bulk_enable(cfg->num_regs, phy->regs);
 	if (ret) {
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index c3588dc9e537..f316e6776f67 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -671,7 +671,7 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data,
 		ret = msm_ioctl_gem_info_set_iova(dev, file, obj, args->value);
 		break;
 	case MSM_INFO_GET_FLAGS:
-		if (obj->import_attach) {
+		if (drm_gem_is_imported(obj)) {
 			ret = -EINVAL;
 			break;
 		}
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index ebc9ba66efb8..2995e80fec3b 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -735,7 +735,7 @@ static void *get_vaddr(struct drm_gem_object *obj, unsigned madv)
 
 	msm_gem_assert_locked(obj);
 
-	if (obj->import_attach)
+	if (drm_gem_is_imported(obj))
 		return ERR_PTR(-ENODEV);
 
 	pages = msm_gem_get_pages_locked(obj, madv);
@@ -1074,7 +1074,7 @@ static void msm_gem_free_object(struct drm_gem_object *obj)
 
 	put_iova_spaces(obj, true);
 
-	if (obj->import_attach) {
+	if (drm_gem_is_imported(obj)) {
 		GEM_WARN_ON(msm_obj->vaddr);
 
 		/* Don't drop the pages for imported dmabuf, as they are not
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 85f0257e83da..ba5c4ff76292 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -224,7 +224,7 @@ msm_gem_assert_locked(struct drm_gem_object *obj)
 /* imported/exported objects are not purgeable: */
 static inline bool is_unpurgeable(struct msm_gem_object *msm_obj)
 {
-	return msm_obj->base.import_attach || msm_obj->pin_count;
+	return drm_gem_is_imported(&msm_obj->base) || msm_obj->pin_count;
 }
 
 static inline bool is_purgeable(struct msm_gem_object *msm_obj)
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
index ee267490c935..2e37913d5a6a 100644
--- a/drivers/gpu/drm/msm/msm_gem_prime.c
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -50,7 +50,7 @@ int msm_gem_prime_pin(struct drm_gem_object *obj)
 	struct page **pages;
 	int ret = 0;
 
-	if (obj->import_attach)
+	if (drm_gem_is_imported(obj))
 		return 0;
 
 	pages = msm_gem_pin_pages_locked(obj);
@@ -62,7 +62,7 @@ int msm_gem_prime_pin(struct drm_gem_object *obj)
 
 void msm_gem_prime_unpin(struct drm_gem_object *obj)
 {
-	if (obj->import_attach)
+	if (drm_gem_is_imported(obj))
 		return;
 
 	msm_gem_unpin_pages_locked(obj);
diff --git a/drivers/gpu/drm/msm/msm_gpu_devfreq.c b/drivers/gpu/drm/msm/msm_gpu_devfreq.c
index 6970b0f7f457..2e1d5c343272 100644
--- a/drivers/gpu/drm/msm/msm_gpu_devfreq.c
+++ b/drivers/gpu/drm/msm/msm_gpu_devfreq.c
@@ -156,6 +156,7 @@ void msm_devfreq_init(struct msm_gpu *gpu)
 	priv->gpu_devfreq_config.downdifferential = 10;
 
 	mutex_init(&df->lock);
+	df->suspended = true;
 
 	ret = dev_pm_qos_add_request(&gpu->pdev->dev, &df->boost_freq,
 				     DEV_PM_QOS_MIN_FREQUENCY, 0);
diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c
index dcb49fd30402..f706e44231a9 100644
--- a/drivers/gpu/drm/msm/msm_mdss.c
+++ b/drivers/gpu/drm/msm/msm_mdss.c
@@ -592,6 +592,16 @@ static const struct msm_mdss_data sa8775p_data = {
 	.reg_bus_bw = 74000,
 };
 
+static const struct msm_mdss_data sar2130p_data = {
+	.ubwc_enc_version = UBWC_3_0, /* 4.0.2 in hw */
+	.ubwc_dec_version = UBWC_4_3,
+	.ubwc_swizzle = 6,
+	.ubwc_bank_spread = true,
+	.highest_bank_bit = 0,
+	.macrotile_mode = 1,
+	.reg_bus_bw = 74000,
+};
+
 static const struct msm_mdss_data sc7180_data = {
 	.ubwc_enc_version = UBWC_2_0,
 	.ubwc_dec_version = UBWC_2_0,
@@ -738,6 +748,7 @@ static const struct of_device_id mdss_dt_match[] = {
 	{ .compatible = "qcom,msm8998-mdss", .data = &msm8998_data },
 	{ .compatible = "qcom,qcm2290-mdss", .data = &qcm2290_data },
 	{ .compatible = "qcom,sa8775p-mdss", .data = &sa8775p_data },
+	{ .compatible = "qcom,sar2130p-mdss", .data = &sar2130p_data },
 	{ .compatible = "qcom,sdm670-mdss", .data = &sdm670_data },
 	{ .compatible = "qcom,sdm845-mdss", .data = &sdm845_data },
 	{ .compatible = "qcom,sc7180-mdss", .data = &sc7180_data },
diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.c b/drivers/gpu/drm/msm/msm_ringbuffer.c
index c5651c39ac2a..89dce15eed3b 100644
--- a/drivers/gpu/drm/msm/msm_ringbuffer.c
+++ b/drivers/gpu/drm/msm/msm_ringbuffer.c
@@ -93,7 +93,7 @@ struct msm_ringbuffer *msm_ringbuffer_new(struct msm_gpu *gpu, int id,
 	}
 
 	msm_gem_object_set_name(ring->bo, "ring%d", id);
-	args.name = to_msm_bo(ring->bo)->name,
+	args.name = to_msm_bo(ring->bo)->name;
 
 	ring->end   = ring->start + (MSM_GPU_RINGBUFFER_SZ >> 2);
 	ring->next  = ring->start;
diff --git a/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml b/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml
index 55a35182858c..5a6ae9fc3194 100644
--- a/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml
@@ -2259,5 +2259,12 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	</reg32>
 </domain>
 
+<domain name="CP_INDIRECT_BUFFER" width="32" varset="chip" prefix="chip" variants="A5XX-">
+	<reg64 offset="0" name="IB_BASE" type="address"/>
+	<reg32 offset="2" name="2">
+		<bitfield name="IB_SIZE" low="0" high="19"/>
+	</reg32>
+</domain>
+
 </database>
 
diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.c b/drivers/gpu/drm/mxsfb/lcdif_drv.c
index 8ee00f59ca82..fcb2a7517377 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_drv.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_drv.c
@@ -134,7 +134,6 @@ static int lcdif_load(struct drm_device *drm)
 {
 	struct platform_device *pdev = to_platform_device(drm->dev);
 	struct lcdif_drm_private *lcdif;
-	struct resource *res;
 	int ret;
 
 	lcdif = devm_kzalloc(&pdev->dev, sizeof(*lcdif), GFP_KERNEL);
@@ -144,8 +143,7 @@ static int lcdif_load(struct drm_device *drm)
 	lcdif->drm = drm;
 	drm->dev_private = lcdif;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	lcdif->base = devm_ioremap_resource(drm->dev, res);
+	lcdif->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(lcdif->base))
 		return PTR_ERR(lcdif->base);
 
diff --git a/drivers/gpu/drm/mxsfb/mxsfb_drv.c b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
index 59020862cf65..c183b1112bc4 100644
--- a/drivers/gpu/drm/mxsfb/mxsfb_drv.c
+++ b/drivers/gpu/drm/mxsfb/mxsfb_drv.c
@@ -8,6 +8,7 @@
  * Copyright (C) 2008 Embedded Alley Solutions, Inc All Rights Reserved.
  */
 
+#include <linux/aperture.h>
 #include <linux/clk.h>
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
@@ -215,7 +216,6 @@ static int mxsfb_load(struct drm_device *drm,
 {
 	struct platform_device *pdev = to_platform_device(drm->dev);
 	struct mxsfb_drm_private *mxsfb;
-	struct resource *res;
 	int ret;
 
 	mxsfb = devm_kzalloc(&pdev->dev, sizeof(*mxsfb), GFP_KERNEL);
@@ -226,8 +226,7 @@ static int mxsfb_load(struct drm_device *drm,
 	drm->dev_private = mxsfb;
 	mxsfb->devdata = devdata;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	mxsfb->base = devm_ioremap_resource(drm->dev, res);
+	mxsfb->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(mxsfb->base))
 		return PTR_ERR(mxsfb->base);
 
@@ -361,6 +360,15 @@ static int mxsfb_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_free;
 
+	/*
+	 * Remove early framebuffers (ie. simplefb). The framebuffer can be
+	 * located anywhere in RAM
+	 */
+	ret = aperture_remove_all_conflicting_devices(mxsfb_driver.name);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret,
+				     "can't kick out existing framebuffers\n");
+
 	ret = drm_dev_register(drm, 0);
 	if (ret)
 		goto err_unload;
diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild
index 7b863355c5c6..385d24530d1e 100644
--- a/drivers/gpu/drm/nouveau/Kbuild
+++ b/drivers/gpu/drm/nouveau/Kbuild
@@ -2,6 +2,7 @@
 ccflags-y += -I $(src)/include
 ccflags-y += -I $(src)/include/nvkm
 ccflags-y += -I $(src)/nvkm
+ccflags-y += -I $(src)/nvkm/subdev/gsp
 ccflags-y += -I $(src)
 
 # NVKM - HW resource manager
@@ -68,5 +69,6 @@ nouveau-y += nv17_fence.o
 nouveau-y += nv50_fence.o
 nouveau-y += nv84_fence.o
 nouveau-y += nvc0_fence.o
+nouveau-y += gv100_fence.o
 
 obj-$(CONFIG_DRM_NOUVEAU) += nouveau.o
diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
index 67146f1e8482..c063756eaea3 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c
@@ -768,9 +768,7 @@ static void nv_crtc_destroy(struct drm_crtc *crtc)
 		disp->image[nv_crtc->index] = NULL;
 	}
 
-	nouveau_bo_unmap(nv_crtc->cursor.nvbo);
-	nouveau_bo_unpin(nv_crtc->cursor.nvbo);
-	nouveau_bo_fini(nv_crtc->cursor.nvbo);
+	nouveau_bo_unpin_del(&nv_crtc->cursor.nvbo);
 	nvif_event_dtor(&nv_crtc->vblank);
 	nvif_head_dtor(&nv_crtc->head);
 	kfree(nv_crtc);
@@ -1303,6 +1301,7 @@ nv04_crtc_vblank_handler(struct nvif_event *event, void *repv, u32 repc)
 int
 nv04_crtc_create(struct drm_device *dev, int crtc_num)
 {
+	struct nouveau_cli *cli = &nouveau_drm(dev)->client;
 	struct nouveau_display *disp = nouveau_display(dev);
 	struct nouveau_crtc *nv_crtc;
 	struct drm_plane *primary;
@@ -1336,20 +1335,9 @@ nv04_crtc_create(struct drm_device *dev, int crtc_num)
 	drm_crtc_helper_add(&nv_crtc->base, &nv04_crtc_helper_funcs);
 	drm_mode_crtc_set_gamma_size(&nv_crtc->base, 256);
 
-	ret = nouveau_bo_new(&nouveau_drm(dev)->client, 64*64*4, 0x100,
-			     NOUVEAU_GEM_DOMAIN_VRAM, 0, 0x0000, NULL, NULL,
-			     &nv_crtc->cursor.nvbo);
-	if (!ret) {
-		ret = nouveau_bo_pin(nv_crtc->cursor.nvbo,
-				     NOUVEAU_GEM_DOMAIN_VRAM, false);
-		if (!ret) {
-			ret = nouveau_bo_map(nv_crtc->cursor.nvbo);
-			if (ret)
-				nouveau_bo_unpin(nv_crtc->cursor.nvbo);
-		}
-		if (ret)
-			nouveau_bo_fini(nv_crtc->cursor.nvbo);
-	}
+	ret = nouveau_bo_new_map(cli, NOUVEAU_GEM_DOMAIN_VRAM, 64 * 64 * 4, &nv_crtc->cursor.nvbo);
+	if (ret)
+		return ret;
 
 	nv04_cursor_init(nv_crtc);
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/Kbuild b/drivers/gpu/drm/nouveau/dispnv50/Kbuild
index 28be2912ff74..d5049dee4b8c 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/Kbuild
+++ b/drivers/gpu/drm/nouveau/dispnv50/Kbuild
@@ -9,11 +9,13 @@ nouveau-y += dispnv50/core907d.o
 nouveau-y += dispnv50/core917d.o
 nouveau-y += dispnv50/corec37d.o
 nouveau-y += dispnv50/corec57d.o
+nouveau-y += dispnv50/coreca7d.o
 
 nouveau-$(CONFIG_DEBUG_FS) += dispnv50/crc.o
 nouveau-$(CONFIG_DEBUG_FS) += dispnv50/crc907d.o
 nouveau-$(CONFIG_DEBUG_FS) += dispnv50/crcc37d.o
 nouveau-$(CONFIG_DEBUG_FS) += dispnv50/crcc57d.o
+nouveau-$(CONFIG_DEBUG_FS) += dispnv50/crcca7d.o
 
 nouveau-y += dispnv50/dac507d.o
 nouveau-y += dispnv50/dac907d.o
@@ -31,6 +33,7 @@ nouveau-y += dispnv50/head907d.o
 nouveau-y += dispnv50/head917d.o
 nouveau-y += dispnv50/headc37d.o
 nouveau-y += dispnv50/headc57d.o
+nouveau-y += dispnv50/headca7d.o
 
 nouveau-y += dispnv50/wimm.o
 nouveau-y += dispnv50/wimmc37b.o
@@ -39,6 +42,7 @@ nouveau-y += dispnv50/wndw.o
 nouveau-y += dispnv50/wndwc37e.o
 nouveau-y += dispnv50/wndwc57e.o
 nouveau-y += dispnv50/wndwc67e.o
+nouveau-y += dispnv50/wndwca7e.o
 
 nouveau-y += dispnv50/base.o
 nouveau-y += dispnv50/base507c.o
diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.c b/drivers/gpu/drm/nouveau/dispnv50/core.c
index f045515696cb..c6331bf97582 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/core.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/core.c
@@ -42,6 +42,7 @@ nv50_core_new(struct nouveau_drm *drm, struct nv50_core **pcore)
 		int version;
 		int (*new)(struct nouveau_drm *, s32, struct nv50_core **);
 	} cores[] = {
+		{ GB202_DISP_CORE_CHANNEL_DMA, 0, coreca7d_new },
 		{ AD102_DISP_CORE_CHANNEL_DMA, 0, corec57d_new },
 		{ GA102_DISP_CORE_CHANNEL_DMA, 0, corec57d_new },
 		{ TU102_DISP_CORE_CHANNEL_DMA, 0, corec57d_new },
diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.h b/drivers/gpu/drm/nouveau/dispnv50/core.h
index f75088186fba..aa07a3ad5dfd 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/core.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/core.h
@@ -7,7 +7,10 @@
 
 struct nv50_core {
 	const struct nv50_core_func *func;
+	struct nv50_disp *disp;
+
 	struct nv50_dmac chan;
+
 	bool assign_windows;
 };
 
@@ -18,6 +21,7 @@ struct nv50_core_func {
 	int (*init)(struct nv50_core *);
 	void (*ntfy_init)(struct nouveau_bo *, u32 offset);
 	int (*caps_init)(struct nouveau_drm *, struct nv50_disp *);
+	u32 caps_class;
 	int (*ntfy_wait_done)(struct nouveau_bo *, u32 offset,
 			      struct nvif_device *);
 	int (*update)(struct nv50_core *, u32 *interlock, bool ntfy);
@@ -70,4 +74,6 @@ int corec37d_wndw_owner(struct nv50_core *);
 extern const struct nv50_outp_func sorc37d;
 
 int corec57d_new(struct nouveau_drm *, s32, struct nv50_core **);
+
+int coreca7d_new(struct nouveau_drm *, s32, struct nv50_core **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/dispnv50/core507d.c b/drivers/gpu/drm/nouveau/dispnv50/core507d.c
index ce2cb78bbdd3..4b947b67a844 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/core507d.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/core507d.c
@@ -165,6 +165,7 @@ core507d_new_(const struct nv50_core_func *func, struct nouveau_drm *drm,
 	if (!(core = *pcore = kzalloc(sizeof(*core), GFP_KERNEL)))
 		return -ENOMEM;
 	core->func = func;
+	core->disp = disp;
 
 	ret = nv50_dmac_create(drm,
 			       &oclass, 0, &args, sizeof(args),
diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c
index 7f637b8830be..83eec2f091f0 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c
@@ -105,7 +105,7 @@ int corec37d_caps_init(struct nouveau_drm *drm, struct nv50_disp *disp)
 	int ret;
 
 	ret = nvif_object_ctor(&disp->disp->object, "dispCaps", 0,
-			       GV100_DISP_CAPS, NULL, 0, &disp->caps);
+			       disp->core->func->caps_class, NULL, 0, &disp->caps);
 	if (ret) {
 		NV_ERROR(drm,
 			 "Failed to init notifier caps region: %d\n",
@@ -162,6 +162,7 @@ corec37d = {
 	.init = corec37d_init,
 	.ntfy_init = corec37d_ntfy_init,
 	.caps_init = corec37d_caps_init,
+	.caps_class = GV100_DISP_CAPS,
 	.ntfy_wait_done = corec37d_ntfy_wait_done,
 	.update = corec37d_update,
 	.wndw.owner = corec37d_wndw_owner,
diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec57d.c b/drivers/gpu/drm/nouveau/dispnv50/corec57d.c
index 421d0d57e1d8..39be576eadcb 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/corec57d.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/corec57d.c
@@ -22,6 +22,7 @@
 #include "core.h"
 #include "head.h"
 
+#include <nvif/class.h>
 #include <nvif/pushc37b.h>
 
 #include <nvhw/class/clc57d.h>
@@ -63,6 +64,7 @@ corec57d = {
 	.init = corec57d_init,
 	.ntfy_init = corec37d_ntfy_init,
 	.caps_init = corec37d_caps_init,
+	.caps_class = GV100_DISP_CAPS,
 	.ntfy_wait_done = corec37d_ntfy_wait_done,
 	.update = corec37d_update,
 	.wndw.owner = corec37d_wndw_owner,
diff --git a/drivers/gpu/drm/nouveau/dispnv50/coreca7d.c b/drivers/gpu/drm/nouveau/dispnv50/coreca7d.c
new file mode 100644
index 000000000000..171727be400e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/dispnv50/coreca7d.c
@@ -0,0 +1,122 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "core.h"
+#include "head.h"
+
+#include <nvif/class.h>
+#include <nvif/pushc97b.h>
+
+#include <nvhw/class/clca7d.h>
+
+#include <nouveau_bo.h>
+
+static int
+coreca7d_update(struct nv50_core *core, u32 *interlock, bool ntfy)
+{
+	const u64 ntfy_addr = core->disp->sync->offset + NV50_DISP_CORE_NTFY;
+	const u32 ntfy_hi = upper_32_bits(ntfy_addr);
+	const u32 ntfy_lo = lower_32_bits(ntfy_addr);
+	struct nvif_push *push = &core->chan.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 5 + (ntfy ? 5 + 2 : 0));
+	if (ret)
+		return ret;
+
+	if (ntfy) {
+		PUSH_MTHD(push, NVCA7D, SET_SURFACE_ADDRESS_HI_NOTIFIER, ntfy_hi,
+
+					SET_SURFACE_ADDRESS_LO_NOTIFIER,
+			  NVVAL(NVCA7D, SET_SURFACE_ADDRESS_LO_NOTIFIER, ADDRESS_LO, ntfy_lo >> 4) |
+			  NVDEF(NVCA7D, SET_SURFACE_ADDRESS_LO_NOTIFIER, TARGET, PHYSICAL_NVM) |
+			  NVDEF(NVCA7D, SET_SURFACE_ADDRESS_LO_NOTIFIER, ENABLE, ENABLE));
+
+		PUSH_MTHD(push, NVCA7D, SET_NOTIFIER_CONTROL,
+			  NVDEF(NVCA7D, SET_NOTIFIER_CONTROL, MODE, WRITE) |
+			  NVDEF(NVCA7D, SET_NOTIFIER_CONTROL, NOTIFY, ENABLE));
+	}
+
+	PUSH_MTHD(push, NVCA7D, SET_INTERLOCK_FLAGS, interlock[NV50_DISP_INTERLOCK_CURS],
+				SET_WINDOW_INTERLOCK_FLAGS, interlock[NV50_DISP_INTERLOCK_WNDW]);
+
+	PUSH_MTHD(push, NVCA7D, UPDATE,
+		  NVDEF(NVCA7D, UPDATE, RELEASE_ELV, TRUE) |
+		  NVDEF(NVCA7D, UPDATE, SPECIAL_HANDLING, NONE) |
+		  NVDEF(NVCA7D, UPDATE, INHIBIT_INTERRUPTS, FALSE));
+
+	if (ntfy) {
+		PUSH_MTHD(push, NVCA7D, SET_NOTIFIER_CONTROL,
+			  NVDEF(NVCA7D, SET_NOTIFIER_CONTROL, NOTIFY, DISABLE));
+	}
+
+	return PUSH_KICK(push);
+}
+
+static int
+coreca7d_init(struct nv50_core *core)
+{
+	struct nvif_push *push = &core->chan.push;
+	const u32 windows = 8, heads = 4;
+	int ret, i;
+
+	ret = PUSH_WAIT(push, windows * 6 + heads * 6);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < windows; i++) {
+		PUSH_MTHD(push, NVCA7D, WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS(i),
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS, RGB_PACKED1BPP, TRUE) |
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS, RGB_PACKED2BPP, TRUE) |
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS, RGB_PACKED4BPP, TRUE) |
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS, RGB_PACKED8BPP, TRUE),
+
+					WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS(i), 0x00000000);
+
+		PUSH_MTHD(push, NVCA7D, WINDOW_SET_WINDOW_USAGE_BOUNDS(i),
+			  NVVAL(NVCA7D, WINDOW_SET_WINDOW_USAGE_BOUNDS, MAX_PIXELS_FETCHED_PER_LINE, 0x7fff) |
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_USAGE_BOUNDS, ILUT_ALLOWED, TRUE) |
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_USAGE_BOUNDS, INPUT_SCALER_TAPS, TAPS_2) |
+			  NVDEF(NVCA7D, WINDOW_SET_WINDOW_USAGE_BOUNDS, UPSCALING_ALLOWED, FALSE),
+
+					WINDOW_SET_PHYSICAL(i), BIT(i));
+	}
+
+	for (i = 0; i < heads; i++) {
+		PUSH_MTHD(push, NVCA7D, HEAD_SET_HEAD_USAGE_BOUNDS(i),
+			  NVDEF(NVCA7D, HEAD_SET_HEAD_USAGE_BOUNDS, CURSOR, USAGE_W256_H256) |
+			  NVDEF(NVCA7D, HEAD_SET_HEAD_USAGE_BOUNDS, OLUT_ALLOWED, TRUE) |
+			  NVDEF(NVCA7D, HEAD_SET_HEAD_USAGE_BOUNDS, OUTPUT_SCALER_TAPS, TAPS_2) |
+			  NVDEF(NVCA7D, HEAD_SET_HEAD_USAGE_BOUNDS, UPSCALING_ALLOWED, TRUE));
+
+		PUSH_MTHD(push, NVCA7D, HEAD_SET_TILE_MASK(i), BIT(i));
+
+		PUSH_MTHD(push, NVCA7D, TILE_SET_TILE_SIZE(i), 0);
+	}
+
+	core->assign_windows = true;
+	return PUSH_KICK(push);
+}
+
+static const struct nv50_core_func
+coreca7d = {
+	.init = coreca7d_init,
+	.ntfy_init = corec37d_ntfy_init,
+	.caps_init = corec37d_caps_init,
+	.caps_class = GB202_DISP_CAPS,
+	.ntfy_wait_done = corec37d_ntfy_wait_done,
+	.update = coreca7d_update,
+	.wndw.owner = corec37d_wndw_owner,
+	.head = &headca7d,
+	.sor = &sorc37d,
+#if IS_ENABLED(CONFIG_DEBUG_FS)
+	.crc = &crcca7d,
+#endif
+};
+
+int
+coreca7d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore)
+{
+	return core507d_new_(&coreca7d, drm, oclass, pcore);
+}
diff --git a/drivers/gpu/drm/nouveau/dispnv50/crc.c b/drivers/gpu/drm/nouveau/dispnv50/crc.c
index 5936b6b3b15d..deb6af40ef32 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/crc.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/crc.c
@@ -509,6 +509,10 @@ nv50_crc_ctx_init(struct nv50_head *head, struct nvif_mmu *mmu,
 	if (ret)
 		return ret;
 
+	/* No CTXDMAs on Blackwell. */
+	if (core->chan.base.user.oclass >= GB202_DISP_CORE_CHANNEL_DMA)
+		return 0;
+
 	ret = nvif_object_ctor(&core->chan.base.user, "kmsCrcNtfyCtxDma",
 			       NV50_DISP_HANDLE_CRC_CTX(head, idx),
 			       NV_DMA_IN_MEMORY,
diff --git a/drivers/gpu/drm/nouveau/dispnv50/crc.h b/drivers/gpu/drm/nouveau/dispnv50/crc.h
index 4823f1fde2dd..75a2009e8193 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/crc.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/crc.h
@@ -94,6 +94,7 @@ void nv50_crc_atomic_clr(struct nv50_head *);
 extern const struct nv50_crc_func crc907d;
 extern const struct nv50_crc_func crcc37d;
 extern const struct nv50_crc_func crcc57d;
+extern const struct nv50_crc_func crcca7d;
 
 #else /* IS_ENABLED(CONFIG_DEBUG_FS) */
 struct nv50_crc {};
diff --git a/drivers/gpu/drm/nouveau/dispnv50/crcca7d.c b/drivers/gpu/drm/nouveau/dispnv50/crcca7d.c
new file mode 100644
index 000000000000..912f59aebe87
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/dispnv50/crcca7d.c
@@ -0,0 +1,98 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "crcc37d.h"
+#include "core.h"
+#include "head.h"
+
+#include <nvif/pushc97b.h>
+
+#include <nvhw/class/clca7d.h>
+
+static int
+crcca7d_set_ctx(struct nv50_head *head, struct nv50_crc_notifier_ctx *ctx)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, ctx ? 3 : 2);
+	if (ret)
+		return ret;
+
+	if (ctx) {
+		const u32 crc_hi = upper_32_bits(ctx->mem.addr);
+		const u32 crc_lo = lower_32_bits(ctx->mem.addr);
+
+		PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_HI_CRC(i), crc_hi,
+
+					HEAD_SET_SURFACE_ADDRESS_LO_CRC(i),
+			  NVVAL(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CRC, ADDRESS_LO, crc_lo >> 4) |
+			  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CRC, TARGET, PHYSICAL_NVM) |
+			  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CRC, ENABLE, ENABLE));
+	} else {
+		PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CRC(i),
+			  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CRC, ENABLE, DISABLE));
+	}
+
+	return 0;
+}
+
+static int
+crcca7d_set_src(struct nv50_head *head, int or, enum nv50_crc_source_type source,
+		struct nv50_crc_notifier_ctx *ctx)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int primary_crc, ret;
+
+	if (!source) {
+		ret = PUSH_WAIT(push, 1);
+		if (ret)
+			return ret;
+
+		PUSH_MTHD(push, NVCA7D, HEAD_SET_CRC_CONTROL(i), 0);
+
+		return crcca7d_set_ctx(head, NULL);
+	}
+
+	switch (source) {
+	case NV50_CRC_SOURCE_TYPE_SOR:
+		primary_crc = NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR(or);
+		break;
+	case NV50_CRC_SOURCE_TYPE_SF:
+		primary_crc = NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SF;
+		break;
+	default:
+		break;
+	}
+
+	ret = crcca7d_set_ctx(head, ctx);
+	if (ret)
+		return ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_CRC_CONTROL(i),
+		  NVDEF(NVCA7D, HEAD_SET_CRC_CONTROL, CONTROLLING_CHANNEL, CORE) |
+		  NVDEF(NVCA7D, HEAD_SET_CRC_CONTROL, EXPECT_BUFFER_COLLAPSE, FALSE) |
+		  NVVAL(NVCA7D, HEAD_SET_CRC_CONTROL, PRIMARY_CRC, primary_crc) |
+		  NVDEF(NVCA7D, HEAD_SET_CRC_CONTROL, SECONDARY_CRC, NONE) |
+		  NVDEF(NVCA7D, HEAD_SET_CRC_CONTROL, CRC_DURING_SNOOZE, DISABLE));
+
+	return 0;
+}
+
+const struct nv50_crc_func
+crcca7d = {
+	.set_src = crcca7d_set_src,
+	.set_ctx = crcca7d_set_ctx,
+	.get_entry = crcc37d_get_entry,
+	.ctx_finished = crcc37d_ctx_finished,
+	.flip_threshold = CRCC37D_FLIP_THRESHOLD,
+	.num_entries = CRCC37D_MAX_ENTRIES,
+	.notifier_len = sizeof(struct crcc37d_notifier),
+};
diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs.c b/drivers/gpu/drm/nouveau/dispnv50/curs.c
index 31d8b2e4791d..557bd05240fa 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/curs.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/curs.c
@@ -31,6 +31,7 @@ nv50_curs_new(struct nouveau_drm *drm, int head, struct nv50_wndw **pwndw)
 		int version;
 		int (*new)(struct nouveau_drm *, int, s32, struct nv50_wndw **);
 	} curses[] = {
+		{ GB202_DISP_CURSOR, 0, cursc37a_new },
 		{ GA102_DISP_CURSOR, 0, cursc37a_new },
 		{ TU102_DISP_CURSOR, 0, cursc37a_new },
 		{ GV100_DISP_CURSOR, 0, cursc37a_new },
diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c
index 504cb3f2054b..e5d37eee4301 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c
@@ -279,6 +279,16 @@ nv50_dmac_create(struct nouveau_drm *drm,
 	if (syncbuf < 0)
 		return 0;
 
+	/* No CTXDMAs on Blackwell. */
+	if (disp->oclass >= GB202_DISP) {
+		/* "handle != NULL_HANDLE" is used to determine enable status
+		 * in a number of places, so fill in some fake object handles.
+		 */
+		dmac->sync.handle = NV50_DISP_HANDLE_SYNCBUF;
+		dmac->vram.handle = NV50_DISP_HANDLE_VRAM;
+		return 0;
+	}
+
 	ret = nvif_object_ctor(&dmac->base.user, "kmsSyncCtxDma", NV50_DISP_HANDLE_SYNCBUF,
 			       NV_DMA_IN_MEMORY,
 			       &(struct nv_dma_v0) {
@@ -775,10 +785,8 @@ nv50_hdmi_enable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc,
 	union hdmi_infoframe infoframe = { 0 };
 	const u8 rekey = 56; /* binary driver, and tegra, constant */
 	u32 max_ac_packet;
-	struct {
-		struct nvif_outp_infoframe_v0 infoframe;
-		u8 data[17];
-	} args = { 0 };
+	DEFINE_RAW_FLEX(struct nvif_outp_infoframe_v0, args, data, 17);
+	const u8 data_len = __member_size(args->data);
 	int ret, size;
 
 	max_ac_packet  = mode->htotal - mode->hdisplay;
@@ -815,29 +823,29 @@ nv50_hdmi_enable(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc,
 		return;
 
 	/* AVI InfoFrame. */
-	args.infoframe.version = 0;
-	args.infoframe.head = nv_crtc->index;
+	args->version = 0;
+	args->head = nv_crtc->index;
 
 	if (!drm_hdmi_avi_infoframe_from_display_mode(&infoframe.avi, &nv_connector->base, mode)) {
 		drm_hdmi_avi_infoframe_quant_range(&infoframe.avi, &nv_connector->base, mode,
 						   HDMI_QUANTIZATION_RANGE_FULL);
 
-		size = hdmi_infoframe_pack(&infoframe, args.data, ARRAY_SIZE(args.data));
+		size = hdmi_infoframe_pack(&infoframe, args->data, data_len);
 	} else {
 		size = 0;
 	}
 
-	nvif_outp_infoframe(&nv_encoder->outp, NVIF_OUTP_INFOFRAME_V0_AVI, &args.infoframe, size);
+	nvif_outp_infoframe(&nv_encoder->outp, NVIF_OUTP_INFOFRAME_V0_AVI, args, size);
 
 	/* Vendor InfoFrame. */
-	memset(&args.data, 0, sizeof(args.data));
+	memset(args->data, 0, data_len);
 	if (!drm_hdmi_vendor_infoframe_from_display_mode(&infoframe.vendor.hdmi,
 							 &nv_connector->base, mode))
-		size = hdmi_infoframe_pack(&infoframe, args.data, ARRAY_SIZE(args.data));
+		size = hdmi_infoframe_pack(&infoframe, args->data, data_len);
 	else
 		size = 0;
 
-	nvif_outp_infoframe(&nv_encoder->outp, NVIF_OUTP_INFOFRAME_V0_VSI, &args.infoframe, size);
+	nvif_outp_infoframe(&nv_encoder->outp, NVIF_OUTP_INFOFRAME_V0_VSI, args, size);
 
 	nv_encoder->hdmi.enabled = true;
 }
@@ -2810,10 +2818,7 @@ nv50_display_destroy(struct drm_device *dev)
 	nvif_object_dtor(&disp->caps);
 	nv50_core_del(&disp->core);
 
-	nouveau_bo_unmap(disp->sync);
-	if (disp->sync)
-		nouveau_bo_unpin(disp->sync);
-	nouveau_bo_fini(disp->sync);
+	nouveau_bo_unpin_del(&disp->sync);
 
 	nouveau_display(dev)->priv = NULL;
 	kfree(disp);
@@ -2845,20 +2850,7 @@ nv50_display_create(struct drm_device *dev)
 	dev->mode_config.normalize_zpos = true;
 
 	/* small shared memory area we use for notifiers and semaphores */
-	ret = nouveau_bo_new(&drm->client, 4096, 0x1000,
-			     NOUVEAU_GEM_DOMAIN_VRAM,
-			     0, 0x0000, NULL, NULL, &disp->sync);
-	if (!ret) {
-		ret = nouveau_bo_pin(disp->sync, NOUVEAU_GEM_DOMAIN_VRAM, true);
-		if (!ret) {
-			ret = nouveau_bo_map(disp->sync);
-			if (ret)
-				nouveau_bo_unpin(disp->sync);
-		}
-		if (ret)
-			nouveau_bo_fini(disp->sync);
-	}
-
+	ret = nouveau_bo_new_map(&drm->client, NOUVEAU_GEM_DOMAIN_VRAM, PAGE_SIZE, &disp->sync);
 	if (ret)
 		goto out;
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c
index d7c74cc43ba5..3dd742b4f823 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/head.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/head.c
@@ -577,6 +577,7 @@ nv50_head_create(struct drm_device *dev, int index)
 		return ERR_PTR(-ENOMEM);
 
 	head->func = disp->core->func->head;
+	head->disp = disp;
 	head->base.index = index;
 
 	if (disp->disp->object.oclass < GF110_DISP)
diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.h b/drivers/gpu/drm/nouveau/dispnv50/head.h
index e9d17037ffcf..8bd2fcb1eff5 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/head.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/head.h
@@ -13,6 +13,8 @@
 
 struct nv50_head {
 	const struct nv50_head_func *func;
+	struct nv50_disp *disp;
+
 	struct nouveau_crtc base;
 	struct nv50_crc crc;
 	struct nv50_lut olut;
@@ -98,4 +100,7 @@ int headc37d_dither(struct nv50_head *, struct nv50_head_atom *);
 void headc37d_static_wndw_map(struct nv50_head *, struct nv50_head_atom *);
 
 extern const struct nv50_head_func headc57d;
+bool headc57d_olut(struct nv50_head *, struct nv50_head_atom *, int size);
+
+extern const struct nv50_head_func headca7d;
 #endif
diff --git a/drivers/gpu/drm/nouveau/dispnv50/headc57d.c b/drivers/gpu/drm/nouveau/dispnv50/headc57d.c
index fde4087e7691..3f8ba495de8f 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/headc57d.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/headc57d.c
@@ -182,7 +182,7 @@ headc57d_olut_load(struct drm_color_lut *in, int size, void __iomem *mem)
 	writew(readw(mem - 4), mem + 4);
 }
 
-static bool
+bool
 headc57d_olut(struct nv50_head *head, struct nv50_head_atom *asyh, int size)
 {
 	if (size != 0 && size != 256 && size != 1024)
diff --git a/drivers/gpu/drm/nouveau/dispnv50/headca7d.c b/drivers/gpu/drm/nouveau/dispnv50/headca7d.c
new file mode 100644
index 000000000000..eeaeb15aa664
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/dispnv50/headca7d.c
@@ -0,0 +1,297 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "head.h"
+#include "atom.h"
+#include "core.h"
+
+#include <nvif/pushc97b.h>
+
+#include <nvhw/class/clca7d.h>
+
+static int
+headca7d_display_id(struct nv50_head *head, u32 display_id)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_DISPLAY_ID(i, 0), display_id);
+
+	return 0;
+}
+
+static int
+headca7d_or(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	u8 depth;
+	int ret;
+
+	switch (asyh->or.depth) {
+	case 6:
+		depth = NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_30_444;
+		break;
+	case 5:
+		depth = NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_24_444;
+		break;
+	case 2:
+		depth = NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_18_444;
+		break;
+	case 0:
+		depth = NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_24_444;
+		break;
+	default:
+		WARN_ON(1);
+		return -EINVAL;
+	}
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE(i),
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE, CRC_MODE, asyh->or.crc_raster) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE, HSYNC_POLARITY, asyh->or.nhsync) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE, VSYNC_POLARITY, asyh->or.nvsync) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE, PIXEL_DEPTH, depth) |
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE, COLOR_SPACE_OVERRIDE, DISABLE) |
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_OUTPUT_RESOURCE, EXT_PACKET_WIN, NONE));
+
+	return 0;
+}
+
+static int
+headca7d_procamp(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_PROCAMP(i),
+		  NVDEF(NVCA7D, HEAD_SET_PROCAMP, COLOR_SPACE, RGB) |
+		  NVDEF(NVCA7D, HEAD_SET_PROCAMP, CHROMA_LPF, DISABLE) |
+		  NVDEF(NVCA7D, HEAD_SET_PROCAMP, DYNAMIC_RANGE, VESA));
+
+	return 0;
+}
+
+static int
+headca7d_dither(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_DITHER_CONTROL(i),
+		  NVVAL(NVCA7D, HEAD_SET_DITHER_CONTROL, ENABLE, asyh->dither.enable) |
+		  NVVAL(NVCA7D, HEAD_SET_DITHER_CONTROL, BITS, asyh->dither.bits) |
+		  NVDEF(NVCA7D, HEAD_SET_DITHER_CONTROL, OFFSET_ENABLE, DISABLE) |
+		  NVVAL(NVCA7D, HEAD_SET_DITHER_CONTROL, MODE, asyh->dither.mode) |
+		  NVVAL(NVCA7D, HEAD_SET_DITHER_CONTROL, PHASE, 0));
+
+	return 0;
+}
+
+static int
+headca7d_curs_clr(struct nv50_head *head)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 4);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_CONTROL_CURSOR(i),
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_CURSOR, ENABLE, DISABLE) |
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_CURSOR, FORMAT, A8R8G8B8));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CURSOR(i, 0),
+		  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CURSOR, ENABLE, DISABLE));
+
+	return 0;
+}
+
+static int
+headca7d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const u32 curs_hi = upper_32_bits(asyh->curs.offset);
+	const u32 curs_lo = lower_32_bits(asyh->curs.offset);
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 7);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_HI_CURSOR(i, 0), curs_hi);
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CURSOR(i, 0),
+		  NVVAL(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CURSOR, ADDRESS_LO, curs_lo >> 4) |
+		  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CURSOR, TARGET, PHYSICAL_NVM) |
+		  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_CURSOR, ENABLE, ENABLE));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_CONTROL_CURSOR(i),
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_CURSOR, ENABLE, ENABLE) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_CURSOR, FORMAT, asyh->curs.format) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_CURSOR, SIZE, asyh->curs.layout) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_CURSOR, HOT_SPOT_X, 0) |
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_CURSOR, HOT_SPOT_Y, 0),
+
+				HEAD_SET_CONTROL_CURSOR_COMPOSITION(i),
+		  NVVAL(NVCA7D, HEAD_SET_CONTROL_CURSOR_COMPOSITION, K1, 0xff) |
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_CURSOR_COMPOSITION, CURSOR_COLOR_FACTOR_SELECT,
+								     K1) |
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_CURSOR_COMPOSITION, VIEWPORT_COLOR_FACTOR_SELECT,
+								     NEG_K1_TIMES_SRC) |
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL_CURSOR_COMPOSITION, MODE, BLEND));
+
+	return 0;
+}
+
+static int
+headca7d_olut_clr(struct nv50_head *head)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_OLUT(i),
+		  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_OLUT, ENABLE, DISABLE));
+
+	return 0;
+}
+
+static int
+headca7d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const u32 olut_hi = upper_32_bits(asyh->olut.offset);
+	const u32 olut_lo = lower_32_bits(asyh->olut.offset);
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 6);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_SURFACE_ADDRESS_HI_OLUT(i), olut_hi,
+
+				HEAD_SET_SURFACE_ADDRESS_LO_OLUT(i),
+		  NVVAL(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_OLUT, ADDRESS_LO, olut_lo >> 4) |
+		  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_OLUT, TARGET, PHYSICAL_NVM) |
+		  NVDEF(NVCA7D, HEAD_SET_SURFACE_ADDRESS_LO_OLUT, ENABLE, ENABLE));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_OLUT_CONTROL(i),
+		  NVVAL(NVCA7D, HEAD_SET_OLUT_CONTROL, INTERPOLATE, asyh->olut.output_mode) |
+		  NVDEF(NVCA7D, HEAD_SET_OLUT_CONTROL, MIRROR, DISABLE) |
+		  NVVAL(NVCA7D, HEAD_SET_OLUT_CONTROL, MODE, asyh->olut.mode) |
+		  NVVAL(NVCA7D, HEAD_SET_OLUT_CONTROL, SIZE, asyh->olut.size),
+
+				HEAD_SET_OLUT_FP_NORM_SCALE(i), 0xffffffff);
+
+	return 0;
+}
+
+static int
+headca7d_mode(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	struct nv50_head_mode *m = &asyh->mode;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 11);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_RASTER_SIZE(i),
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_SIZE, WIDTH, m->h.active) |
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_SIZE, HEIGHT, m->v.active),
+
+				HEAD_SET_RASTER_SYNC_END(i),
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_SYNC_END, X, m->h.synce) |
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_SYNC_END, Y, m->v.synce),
+
+				HEAD_SET_RASTER_BLANK_END(i),
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_BLANK_END, X, m->h.blanke) |
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_BLANK_END, Y, m->v.blanke),
+
+				HEAD_SET_RASTER_BLANK_START(i),
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_BLANK_START, X, m->h.blanks) |
+		  NVVAL(NVCA7D, HEAD_SET_RASTER_BLANK_START, Y, m->v.blanks));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_CONTROL(i),
+		  NVDEF(NVCA7D, HEAD_SET_CONTROL, STRUCTURE, PROGRESSIVE));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_PIXEL_CLOCK_FREQUENCY(i),
+		  NVVAL(NVCA7D, HEAD_SET_PIXEL_CLOCK_FREQUENCY, HERTZ, m->clock * 1000));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX(i),
+		  NVVAL(NVCA7D, HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX, HERTZ, m->clock * 1000));
+
+	return 0;
+}
+
+static int
+headca7d_view(struct nv50_head *head, struct nv50_head_atom *asyh)
+{
+	struct nvif_push *push = &head->disp->core->chan.push;
+	const int i = head->base.index;
+	int ret;
+
+	ret = PUSH_WAIT(push, 4);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_VIEWPORT_SIZE_IN(i),
+		  NVVAL(NVCA7D, HEAD_SET_VIEWPORT_SIZE_IN, WIDTH, asyh->view.iW) |
+		  NVVAL(NVCA7D, HEAD_SET_VIEWPORT_SIZE_IN, HEIGHT, asyh->view.iH));
+
+	PUSH_MTHD(push, NVCA7D, HEAD_SET_VIEWPORT_SIZE_OUT(i),
+		  NVVAL(NVCA7D, HEAD_SET_VIEWPORT_SIZE_OUT, WIDTH, asyh->view.oW) |
+		  NVVAL(NVCA7D, HEAD_SET_VIEWPORT_SIZE_OUT, HEIGHT, asyh->view.oH));
+	return 0;
+}
+
+const struct nv50_head_func
+headca7d = {
+	.view = headca7d_view,
+	.mode = headca7d_mode,
+	.olut = headc57d_olut,
+	.ilut_check = head907d_ilut_check,
+	.olut_identity = true,
+	.olut_size = 1024,
+	.olut_set = headca7d_olut_set,
+	.olut_clr = headca7d_olut_clr,
+	.curs_layout = head917d_curs_layout,
+	.curs_format = headc37d_curs_format,
+	.curs_set = headca7d_curs_set,
+	.curs_clr = headca7d_curs_clr,
+	.dither = headca7d_dither,
+	.procamp = headca7d_procamp,
+	.or = headca7d_or,
+	.static_wndw_map = headc37d_static_wndw_map,
+	.display_id = headca7d_display_id,
+};
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wimm.c b/drivers/gpu/drm/nouveau/dispnv50/wimm.c
index 566fbddfc8d7..53c9ab6c138b 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wimm.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wimm.c
@@ -31,6 +31,7 @@ nv50_wimm_init(struct nouveau_drm *drm, struct nv50_wndw *wndw)
 		int version;
 		int (*init)(struct nouveau_drm *, s32, struct nv50_wndw *);
 	} wimms[] = {
+		{ GB202_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
 		{ GA102_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
 		{ TU102_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
 		{ GV100_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init },
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
index f6be426dd525..11d5b923d6e7 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c
@@ -556,14 +556,24 @@ nv50_wndw_prepare_fb(struct drm_plane *plane, struct drm_plane_state *state)
 		return ret;
 
 	if (wndw->ctxdma.parent) {
-		ctxdma = nv50_wndw_ctxdma_new(wndw, fb);
-		if (IS_ERR(ctxdma)) {
-			nouveau_bo_unpin(nvbo);
-			return PTR_ERR(ctxdma);
+		if (wndw->wndw.base.user.oclass < GB202_DISP_WINDOW_CHANNEL_DMA) {
+			ctxdma = nv50_wndw_ctxdma_new(wndw, fb);
+			if (IS_ERR(ctxdma)) {
+				nouveau_bo_unpin(nvbo);
+				return PTR_ERR(ctxdma);
+			}
+
+			if (asyw->visible)
+				asyw->image.handle[0] = ctxdma->object.handle;
+		} else {
+			/* No CTXDMAs on Blackwell. */
+			if (asyw->visible) {
+				/* "handle != NULL_HANDLE" is used to determine enable status
+				 * in a number of places, so fill in a fake object handle.
+				 */
+				asyw->image.handle[0] = NV50_DISP_HANDLE_WNDW_CTX(0);
+			}
 		}
-
-		if (asyw->visible)
-			asyw->image.handle[0] = ctxdma->object.handle;
 	}
 
 	ret = drm_gem_plane_helper_prepare_fb(plane, state);
@@ -901,6 +911,7 @@ nv50_wndw_new(struct nouveau_drm *drm, enum drm_plane_type type, int index,
 		int (*new)(struct nouveau_drm *, enum drm_plane_type,
 			   int, s32, struct nv50_wndw **);
 	} wndws[] = {
+		{ GB202_DISP_WINDOW_CHANNEL_DMA, 0, wndwca7e_new },
 		{ GA102_DISP_WINDOW_CHANNEL_DMA, 0, wndwc67e_new },
 		{ TU102_DISP_WINDOW_CHANNEL_DMA, 0, wndwc57e_new },
 		{ GV100_DISP_WINDOW_CHANNEL_DMA, 0, wndwc37e_new },
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.h b/drivers/gpu/drm/nouveau/dispnv50/wndw.h
index 76a6ae5d5652..90d100514bef 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndw.h
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.h
@@ -134,6 +134,9 @@ int wndwc57e_csc_clr(struct nv50_wndw *);
 int wndwc67e_new(struct nouveau_drm *, enum drm_plane_type, int, s32,
 		 struct nv50_wndw **);
 
+int wndwca7e_new(struct nouveau_drm *, enum drm_plane_type, int, s32,
+		 struct nv50_wndw **);
+
 int nv50_wndw_new(struct nouveau_drm *, enum drm_plane_type, int index,
 		  struct nv50_wndw **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c
index 50a7b97d37a2..554c4f91f8be 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c
@@ -25,6 +25,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <nouveau_bo.h>
 
+#include <nvif/class.h>
 #include <nvif/if0014.h>
 #include <nvif/pushc37b.h>
 
diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwca7e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwca7e.c
new file mode 100644
index 000000000000..0d8e9a9d1a57
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/dispnv50/wndwca7e.c
@@ -0,0 +1,209 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "wndw.h"
+#include "atom.h"
+
+#include <nvif/pushc97b.h>
+
+#include <nvhw/class/clca7e.h>
+
+#include <nouveau_bo.h>
+
+static int
+wndwca7e_image_clr(struct nv50_wndw *wndw)
+{
+	struct nvif_push *push = &wndw->wndw.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 4);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7E, SET_PRESENT_CONTROL,
+		  NVVAL(NVCA7E, SET_PRESENT_CONTROL, MIN_PRESENT_INTERVAL, 0) |
+		  NVDEF(NVCA7E, SET_PRESENT_CONTROL, BEGIN_MODE, NON_TEARING));
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_LO_ISO(0),
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_ISO, ENABLE, DISABLE));
+
+	return 0;
+}
+
+static int
+wndwca7e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	const u32 iso0_hi = upper_32_bits(asyw->image.offset[0]);
+	const u32 iso0_lo = lower_32_bits(asyw->image.offset[0]);
+	struct nvif_push *push = &wndw->wndw.push;
+	int ret, kind;
+
+	if (asyw->image.kind)
+		kind = NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_KIND_BLOCKLINEAR;
+	else
+		kind = NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_KIND_PITCH;
+
+	ret = PUSH_WAIT(push, 17);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_HI_ISO(0), iso0_hi);
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_LO_ISO(0),
+		  NVVAL(NVCA7E, SET_SURFACE_ADDRESS_LO_ISO, ADDRESS_LO, iso0_lo >> 4) |
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_ISO, TARGET, PHYSICAL_NVM) |
+		  NVVAL(NVCA7E, SET_SURFACE_ADDRESS_LO_ISO, KIND, kind) |
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_ISO, ENABLE, ENABLE));
+
+	PUSH_MTHD(push, NVCA7E, SET_PRESENT_CONTROL,
+		  NVVAL(NVCA7E, SET_PRESENT_CONTROL, MIN_PRESENT_INTERVAL, asyw->image.interval) |
+		  NVVAL(NVCA7E, SET_PRESENT_CONTROL, BEGIN_MODE, asyw->image.mode) |
+		  NVDEF(NVCA7E, SET_PRESENT_CONTROL, TIMESTAMP_MODE, DISABLE));
+
+	PUSH_MTHD(push, NVCA7E, SET_SIZE,
+		  NVVAL(NVCA7E, SET_SIZE, WIDTH, asyw->image.w) |
+		  NVVAL(NVCA7E, SET_SIZE, HEIGHT, asyw->image.h),
+
+				SET_STORAGE,
+		  NVVAL(NVCA7E, SET_STORAGE, BLOCK_HEIGHT, asyw->image.blockh),
+
+				SET_PARAMS,
+		  NVVAL(NVCA7E, SET_PARAMS, FORMAT, asyw->image.format) |
+		  NVDEF(NVCA7E, SET_PARAMS, CLAMP_BEFORE_BLEND, DISABLE) |
+		  NVDEF(NVCA7E, SET_PARAMS, SWAP_UV, DISABLE) |
+		  NVDEF(NVCA7E, SET_PARAMS, FMT_ROUNDING_MODE, ROUND_TO_NEAREST),
+
+				SET_PLANAR_STORAGE(0),
+		  NVVAL(NVCA7E, SET_PLANAR_STORAGE, PITCH, asyw->image.blocks[0]) |
+		  NVVAL(NVCA7E, SET_PLANAR_STORAGE, PITCH, asyw->image.pitch[0] >> 6));
+
+	PUSH_MTHD(push, NVCA7E, SET_POINT_IN(0),
+		  NVVAL(NVCA7E, SET_POINT_IN, X, asyw->state.src_x >> 16) |
+		  NVVAL(NVCA7E, SET_POINT_IN, Y, asyw->state.src_y >> 16));
+
+	PUSH_MTHD(push, NVCA7E, SET_SIZE_IN,
+		  NVVAL(NVCA7E, SET_SIZE_IN, WIDTH, asyw->state.src_w >> 16) |
+		  NVVAL(NVCA7E, SET_SIZE_IN, HEIGHT, asyw->state.src_h >> 16));
+
+	PUSH_MTHD(push, NVCA7E, SET_SIZE_OUT,
+		  NVVAL(NVCA7E, SET_SIZE_OUT, WIDTH, asyw->state.crtc_w) |
+		  NVVAL(NVCA7E, SET_SIZE_OUT, HEIGHT, asyw->state.crtc_h));
+
+	return 0;
+}
+
+static int
+wndwca7e_ilut_clr(struct nv50_wndw *wndw)
+{
+	struct nvif_push *push = &wndw->wndw.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_LO_ILUT,
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_ILUT, ENABLE, DISABLE));
+
+	return 0;
+}
+
+static int
+wndwca7e_ilut_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	const u32 ilut_hi = upper_32_bits(asyw->xlut.i.offset);
+	const u32 ilut_lo = lower_32_bits(asyw->xlut.i.offset);
+	struct nvif_push *push = &wndw->wndw.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 5);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_HI_ILUT, ilut_hi,
+
+				SET_SURFACE_ADDRESS_LO_ILUT,
+		  NVVAL(NVCA7E, SET_SURFACE_ADDRESS_LO_ILUT, ADDRESS_LO, ilut_lo >> 4) |
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_ILUT, TARGET, PHYSICAL_NVM) |
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_ILUT, ENABLE, ENABLE));
+
+	PUSH_MTHD(push, NVCA7E, SET_ILUT_CONTROL,
+		  NVVAL(NVCA7E, SET_ILUT_CONTROL, SIZE, asyw->xlut.i.size) |
+		  NVVAL(NVCA7E, SET_ILUT_CONTROL, MODE, asyw->xlut.i.mode) |
+		  NVVAL(NVCA7E, SET_ILUT_CONTROL, INTERPOLATE, asyw->xlut.i.output_mode));
+
+	return 0;
+}
+
+static int
+wndwca7e_ntfy_clr(struct nv50_wndw *wndw)
+{
+	struct nvif_push *push = &wndw->wndw.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 2);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_LO_NOTIFIER,
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_NOTIFIER, ENABLE, DISABLE));
+
+	return 0;
+}
+
+static int
+wndwca7e_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw)
+{
+	struct nv50_disp *disp = nv50_disp(wndw->plane.dev);
+	const u64 ntfy_addr = disp->sync->offset + asyw->ntfy.offset;
+	const u32 ntfy_hi = upper_32_bits(ntfy_addr);
+	const u32 ntfy_lo = lower_32_bits(ntfy_addr);
+	struct nvif_push *push = &wndw->wndw.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 5);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVCA7E, SET_SURFACE_ADDRESS_HI_NOTIFIER, ntfy_hi,
+
+				SET_SURFACE_ADDRESS_LO_NOTIFIER,
+		  NVVAL(NVCA7E, SET_SURFACE_ADDRESS_LO_NOTIFIER, ADDRESS_LO, ntfy_lo >> 4) |
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_NOTIFIER, TARGET, PHYSICAL_NVM) |
+		  NVDEF(NVCA7E, SET_SURFACE_ADDRESS_LO_NOTIFIER, ENABLE, ENABLE));
+
+	PUSH_MTHD(push, NVCA7E, SET_NOTIFIER_CONTROL,
+		  NVVAL(NVCA7E, SET_NOTIFIER_CONTROL, MODE, asyw->ntfy.awaken));
+
+	return 0;
+}
+
+static const struct nv50_wndw_func
+wndwca7e = {
+	.acquire = wndwc37e_acquire,
+	.release = wndwc37e_release,
+	.ntfy_set = wndwca7e_ntfy_set,
+	.ntfy_clr = wndwca7e_ntfy_clr,
+	.ntfy_reset = corec37d_ntfy_init,
+	.ntfy_wait_begun = base507c_ntfy_wait_begun,
+	.ilut = wndwc57e_ilut,
+	.ilut_identity = true,
+	.ilut_size = 1024,
+	.xlut_set = wndwca7e_ilut_set,
+	.xlut_clr = wndwca7e_ilut_clr,
+	.csc = base907c_csc,
+	.csc_set = wndwc57e_csc_set,
+	.csc_clr = wndwc57e_csc_clr,
+	.image_set = wndwca7e_image_set,
+	.image_clr = wndwca7e_image_clr,
+	.blend_set = wndwc37e_blend_set,
+	.update = wndwc37e_update,
+};
+
+int
+wndwca7e_new(struct nouveau_drm *drm, enum drm_plane_type type, int index,
+	     s32 oclass, struct nv50_wndw **pwndw)
+{
+	return wndwc37e_new_(&wndwca7e, drm, type, index, oclass, BIT(index >> 1), pwndw);
+}
diff --git a/drivers/gpu/drm/nouveau/gv100_fence.c b/drivers/gpu/drm/nouveau/gv100_fence.c
new file mode 100644
index 000000000000..cccdeca72002
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/gv100_fence.c
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "nouveau_drv.h"
+#include "nouveau_dma.h"
+#include "nouveau_fence.h"
+
+#include "nv50_display.h"
+
+#include <nvif/push906f.h>
+
+#include <nvhw/class/clc36f.h>
+
+static int
+gv100_fence_emit32(struct nouveau_channel *chan, u64 virtual, u32 sequence)
+{
+	struct nvif_push *push = &chan->chan.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 8);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVC36F, SEM_ADDR_LO, lower_32_bits(virtual),
+				SEM_ADDR_HI, upper_32_bits(virtual),
+				SEM_PAYLOAD_LO, sequence);
+
+	PUSH_MTHD(push, NVC36F, SEM_EXECUTE,
+		  NVDEF(NVC36F, SEM_EXECUTE, OPERATION, RELEASE) |
+		  NVDEF(NVC36F, SEM_EXECUTE, RELEASE_WFI, EN) |
+		  NVDEF(NVC36F, SEM_EXECUTE, PAYLOAD_SIZE, 32BIT) |
+		  NVDEF(NVC36F, SEM_EXECUTE, RELEASE_TIMESTAMP, DIS));
+
+	PUSH_MTHD(push, NVC36F, NON_STALL_INTERRUPT, 0);
+
+	PUSH_KICK(push);
+	return 0;
+}
+
+static int
+gv100_fence_sync32(struct nouveau_channel *chan, u64 virtual, u32 sequence)
+{
+	struct nvif_push *push = &chan->chan.push;
+	int ret;
+
+	ret = PUSH_WAIT(push, 6);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVC36F, SEM_ADDR_LO, lower_32_bits(virtual),
+				SEM_ADDR_HI, upper_32_bits(virtual),
+				SEM_PAYLOAD_LO, sequence);
+
+	PUSH_MTHD(push, NVC36F, SEM_EXECUTE,
+		  NVDEF(NVC36F, SEM_EXECUTE, OPERATION, ACQ_CIRC_GEQ) |
+		  NVDEF(NVC36F, SEM_EXECUTE, ACQUIRE_SWITCH_TSG, EN) |
+		  NVDEF(NVC36F, SEM_EXECUTE, PAYLOAD_SIZE, 32BIT));
+
+	PUSH_KICK(push);
+	return 0;
+}
+
+static int
+gv100_fence_context_new(struct nouveau_channel *chan)
+{
+	struct nv84_fence_chan *fctx;
+	int ret;
+
+	ret = nv84_fence_context_new(chan);
+	if (ret)
+		return ret;
+
+	fctx = chan->fence;
+	fctx->base.emit32 = gv100_fence_emit32;
+	fctx->base.sync32 = gv100_fence_sync32;
+	return 0;
+}
+
+int
+gv100_fence_create(struct nouveau_drm *drm)
+{
+	struct nv84_fence_priv *priv;
+	int ret;
+
+	ret = nv84_fence_create(drm);
+	if (ret)
+		return ret;
+
+	priv = drm->fence;
+	priv->base.context_new = gv100_fence_context_new;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/class/clc36f.h b/drivers/gpu/drm/nouveau/include/nvhw/class/clc36f.h
new file mode 100644
index 000000000000..8735dda4c8a7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/class/clc36f.h
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef _clc36f_h_
+#define _clc36f_h_
+
+#define NVC36F_NON_STALL_INTERRUPT                                 (0x00000020)
+#define NVC36F_NON_STALL_INTERRUPT_HANDLE                                 31:0
+#define NVC36F_SEM_ADDR_LO                                         (0x0000005c)
+#define NVC36F_SEM_ADDR_LO_OFFSET                                         31:2
+#define NVC36F_SEM_ADDR_HI                                         (0x00000060)
+#define NVC36F_SEM_ADDR_HI_OFFSET                                          7:0
+#define NVC36F_SEM_PAYLOAD_LO                                      (0x00000064)
+#define NVC36F_SEM_PAYLOAD_LO_PAYLOAD                                     31:0
+#define NVC36F_SEM_PAYLOAD_HI                                      (0x00000068)
+#define NVC36F_SEM_PAYLOAD_HI_PAYLOAD                                     31:0
+#define NVC36F_SEM_EXECUTE                                         (0x0000006c)
+#define NVC36F_SEM_EXECUTE_OPERATION                                       2:0
+#define NVC36F_SEM_EXECUTE_OPERATION_ACQUIRE                        0x00000000
+#define NVC36F_SEM_EXECUTE_OPERATION_RELEASE                        0x00000001
+#define NVC36F_SEM_EXECUTE_OPERATION_ACQ_STRICT_GEQ                 0x00000002
+#define NVC36F_SEM_EXECUTE_OPERATION_ACQ_CIRC_GEQ                   0x00000003
+#define NVC36F_SEM_EXECUTE_OPERATION_ACQ_AND                        0x00000004
+#define NVC36F_SEM_EXECUTE_OPERATION_ACQ_NOR                        0x00000005
+#define NVC36F_SEM_EXECUTE_OPERATION_REDUCTION                      0x00000006
+#define NVC36F_SEM_EXECUTE_ACQUIRE_SWITCH_TSG                            12:12
+#define NVC36F_SEM_EXECUTE_ACQUIRE_SWITCH_TSG_DIS                   0x00000000
+#define NVC36F_SEM_EXECUTE_ACQUIRE_SWITCH_TSG_EN                    0x00000001
+#define NVC36F_SEM_EXECUTE_RELEASE_WFI                                   20:20
+#define NVC36F_SEM_EXECUTE_RELEASE_WFI_DIS                          0x00000000
+#define NVC36F_SEM_EXECUTE_RELEASE_WFI_EN                           0x00000001
+#define NVC36F_SEM_EXECUTE_PAYLOAD_SIZE                                  24:24
+#define NVC36F_SEM_EXECUTE_PAYLOAD_SIZE_32BIT                       0x00000000
+#define NVC36F_SEM_EXECUTE_PAYLOAD_SIZE_64BIT                       0x00000001
+#define NVC36F_SEM_EXECUTE_RELEASE_TIMESTAMP                             25:25
+#define NVC36F_SEM_EXECUTE_RELEASE_TIMESTAMP_DIS                    0x00000000
+#define NVC36F_SEM_EXECUTE_RELEASE_TIMESTAMP_EN                     0x00000001
+#define NVC36F_SEM_EXECUTE_REDUCTION                                     30:27
+#define NVC36F_SEM_EXECUTE_REDUCTION_IMIN                           0x00000000
+#define NVC36F_SEM_EXECUTE_REDUCTION_IMAX                           0x00000001
+#define NVC36F_SEM_EXECUTE_REDUCTION_IXOR                           0x00000002
+#define NVC36F_SEM_EXECUTE_REDUCTION_IAND                           0x00000003
+#define NVC36F_SEM_EXECUTE_REDUCTION_IOR                            0x00000004
+#define NVC36F_SEM_EXECUTE_REDUCTION_IADD                           0x00000005
+#define NVC36F_SEM_EXECUTE_REDUCTION_INC                            0x00000006
+#define NVC36F_SEM_EXECUTE_REDUCTION_DEC                            0x00000007
+#define NVC36F_SEM_EXECUTE_REDUCTION_FORMAT                              31:31
+#define NVC36F_SEM_EXECUTE_REDUCTION_FORMAT_SIGNED                  0x00000000
+#define NVC36F_SEM_EXECUTE_REDUCTION_FORMAT_UNSIGNED                0x00000001
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/class/clc97b.h b/drivers/gpu/drm/nouveau/include/nvhw/class/clc97b.h
new file mode 100644
index 000000000000..092aebe9551c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/class/clc97b.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef _clc97b_h_
+#define _clc97b_h_
+
+// dma opcode instructions
+#define NVC97B_DMA
+#define NVC97B_DMA_OPCODE                                                        31:29
+#define NVC97B_DMA_OPCODE_METHOD                                            0x00000000
+#define NVC97B_DMA_OPCODE_JUMP                                              0x00000001
+#define NVC97B_DMA_OPCODE_NONINC_METHOD                                     0x00000002
+#define NVC97B_DMA_OPCODE_SET_SUBDEVICE_MASK                                0x00000003
+#define NVC97B_DMA_METHOD_COUNT                                                  27:18
+#define NVC97B_DMA_METHOD_OFFSET                                                  15:2
+#define NVC97B_DMA_DATA                                                           31:0
+#define NVC97B_DMA_DATA_NOP                                                 0x00000000
+#define NVC97B_DMA_JUMP_OFFSET                                                    15:2
+#define NVC97B_DMA_SET_SUBDEVICE_MASK_VALUE                                       11:0
+
+#endif // _clc97b_h
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/class/clca7d.h b/drivers/gpu/drm/nouveau/include/nvhw/class/clca7d.h
new file mode 100644
index 000000000000..0fec6fc21d44
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/class/clca7d.h
@@ -0,0 +1,868 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef _clca7d_h_
+#define _clca7d_h_
+
+// class methods
+#define NVCA7D_UPDATE                                                           (0x00000200)
+#define NVCA7D_UPDATE_SPECIAL_HANDLING                                          21:20
+#define NVCA7D_UPDATE_SPECIAL_HANDLING_NONE                                     (0x00000000)
+#define NVCA7D_UPDATE_SPECIAL_HANDLING_INTERRUPT_RM                             (0x00000001)
+#define NVCA7D_UPDATE_SPECIAL_HANDLING_MODE_SWITCH                              (0x00000002)
+#define NVCA7D_UPDATE_SPECIAL_HANDLING_REASON                                   19:12
+#define NVCA7D_UPDATE_INHIBIT_INTERRUPTS                                        24:24
+#define NVCA7D_UPDATE_INHIBIT_INTERRUPTS_FALSE                                  (0x00000000)
+#define NVCA7D_UPDATE_INHIBIT_INTERRUPTS_TRUE                                   (0x00000001)
+#define NVCA7D_UPDATE_RELEASE_ELV                                               0:0
+#define NVCA7D_UPDATE_RELEASE_ELV_FALSE                                         (0x00000000)
+#define NVCA7D_UPDATE_RELEASE_ELV_TRUE                                          (0x00000001)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN                                             8:4
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_NONE                               (0x00000000)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN(i)                                 (0x00000001 +(i))
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN__SIZE_1                            16
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_0                                  (0x00000001)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_1                                  (0x00000002)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_2                                  (0x00000003)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_3                                  (0x00000004)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_4                                  (0x00000005)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_5                                  (0x00000006)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_6                                  (0x00000007)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_7                                  (0x00000008)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_8                                  (0x00000009)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_9                                  (0x0000000A)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_A                                  (0x0000000B)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_B                                  (0x0000000C)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_C                                  (0x0000000D)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_D                                  (0x0000000E)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_E                                  (0x0000000F)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_LOCK_PIN_F                                  (0x00000010)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_FLIP_LOCK_0                        (0x00000014)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_FLIP_LOCK_1                        (0x00000015)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_FLIP_LOCK_2                        (0x00000016)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_FLIP_LOCK_3                        (0x00000017)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK(i)                       (0x00000018 +(i))
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK__SIZE_1                  8
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_0                        (0x00000018)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_1                        (0x00000019)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_2                        (0x0000001A)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_3                        (0x0000001B)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_4                        (0x0000001C)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_5                        (0x0000001D)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_6                        (0x0000001E)
+#define NVCA7D_UPDATE_FLIP_LOCK_PIN_INTERNAL_SCAN_LOCK_7                        (0x0000001F)
+#define NVCA7D_UPDATE_FORCE_FULLSCREEN                                          28:28
+#define NVCA7D_UPDATE_FORCE_FULLSCREEN_FALSE                                    (0x00000000)
+#define NVCA7D_UPDATE_FORCE_FULLSCREEN_TRUE                                     (0x00000001)
+#define NVCA7D_SET_NOTIFIER_CONTROL                                             (0x0000020C)
+#define NVCA7D_SET_NOTIFIER_CONTROL_MODE                                        0:0
+#define NVCA7D_SET_NOTIFIER_CONTROL_MODE_WRITE                                  (0x00000000)
+#define NVCA7D_SET_NOTIFIER_CONTROL_MODE_WRITE_AWAKEN                           (0x00000001)
+#define NVCA7D_SET_NOTIFIER_CONTROL_NOTIFY                                      12:12
+#define NVCA7D_SET_NOTIFIER_CONTROL_NOTIFY_DISABLE                              (0x00000000)
+#define NVCA7D_SET_NOTIFIER_CONTROL_NOTIFY_ENABLE                               (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS                                              (0x00000218)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR(i)                     ((i)+0):((i)+0)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR__SIZE_1                8
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR_DISABLE                (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR_ENABLE                 (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR0                       0:0
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR0_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR0_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR1                       1:1
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR1_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR1_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR2                       2:2
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR2_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR2_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR3                       3:3
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR3_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR3_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR4                       4:4
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR4_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR4_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR5                       5:5
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR5_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR5_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR6                       6:6
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR6_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR6_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR7                       7:7
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR7_DISABLE               (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CURSOR7_ENABLE                (0x00000001)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CORE                          16:16
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CORE_DISABLE                  (0x00000000)
+#define NVCA7D_SET_INTERLOCK_FLAGS_INTERLOCK_WITH_CORE_ENABLE                   (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS                                       (0x0000021C)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW(i)              ((i)+0):((i)+0)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW__SIZE_1         32
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW_DISABLE         (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW_ENABLE          (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW0                0:0
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW0_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW0_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW1                1:1
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW1_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW1_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW2                2:2
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW2_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW2_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW3                3:3
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW3_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW3_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW4                4:4
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW4_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW4_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW5                5:5
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW5_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW5_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW6                6:6
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW6_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW6_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW7                7:7
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW7_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW7_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW8                8:8
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW8_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW8_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW9                9:9
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW9_DISABLE        (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW9_ENABLE         (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW10               10:10
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW10_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW10_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW11               11:11
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW11_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW11_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW12               12:12
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW12_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW12_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW13               13:13
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW13_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW13_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW14               14:14
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW14_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW14_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW15               15:15
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW15_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW15_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW16               16:16
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW16_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW16_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW17               17:17
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW17_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW17_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW18               18:18
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW18_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW18_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW19               19:19
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW19_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW19_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW20               20:20
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW20_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW20_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW21               21:21
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW21_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW21_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW22               22:22
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW22_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW22_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW23               23:23
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW23_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW23_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW24               24:24
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW24_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW24_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW25               25:25
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW25_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW25_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW26               26:26
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW26_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW26_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW27               27:27
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW27_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW27_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW28               28:28
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW28_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW28_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW29               29:29
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW29_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW29_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW30               30:30
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW30_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW30_ENABLE        (0x00000001)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW31               31:31
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW31_DISABLE       (0x00000000)
+#define NVCA7D_SET_WINDOW_INTERLOCK_FLAGS_INTERLOCK_WITH_WINDOW31_ENABLE        (0x00000001)
+#define NVCA7D_SET_SURFACE_ADDRESS_HI_NOTIFIER                                  (0x00000260)
+#define NVCA7D_SET_SURFACE_ADDRESS_HI_NOTIFIER_ADDRESS_HI                       31:0
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER                                  (0x00000264)
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_ADDRESS_LO                       31:4
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET                           3:2
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_IOVA                      (0x00000000)
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_PHYSICAL_NVM              (0x00000001)
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_PHYSICAL_PCI              (0x00000002)
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_PHYSICAL_PCI_COHERENT     (0x00000003)
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_ENABLE                           0:0
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_ENABLE_DISABLE                   (0x00000000)
+#define NVCA7D_SET_SURFACE_ADDRESS_LO_NOTIFIER_ENABLE_ENABLE                    (0x00000001)
+
+#define NVCA7D_SOR_SET_CONTROL(a)                                               (0x00000300 + (a)*0x00000020)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK                                       7:0
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_NONE                                  (0x00000000)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD0                                 (0x00000001)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD1                                 (0x00000002)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD2                                 (0x00000004)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD3                                 (0x00000008)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD4                                 (0x00000010)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD5                                 (0x00000020)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD6                                 (0x00000040)
+#define NVCA7D_SOR_SET_CONTROL_OWNER_MASK_HEAD7                                 (0x00000080)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL                                         11:8
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_LVDS_CUSTOM                             (0x00000000)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_SINGLE_TMDS_A                           (0x00000001)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_SINGLE_TMDS_B                           (0x00000002)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_DUAL_TMDS                               (0x00000005)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_DP_A                                    (0x00000008)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_DP_B                                    (0x00000009)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_HDMI_FRL                                (0x0000000C)
+#define NVCA7D_SOR_SET_CONTROL_PROTOCOL_CUSTOM                                  (0x0000000F)
+#define NVCA7D_SOR_SET_CONTROL_DE_SYNC_POLARITY                                 16:16
+#define NVCA7D_SOR_SET_CONTROL_DE_SYNC_POLARITY_POSITIVE_TRUE                   (0x00000000)
+#define NVCA7D_SOR_SET_CONTROL_DE_SYNC_POLARITY_NEGATIVE_TRUE                   (0x00000001)
+#define NVCA7D_SOR_SET_CONTROL_PIXEL_REPLICATE_MODE                             21:20
+#define NVCA7D_SOR_SET_CONTROL_PIXEL_REPLICATE_MODE_OFF                         (0x00000000)
+#define NVCA7D_SOR_SET_CONTROL_PIXEL_REPLICATE_MODE_X2                          (0x00000001)
+#define NVCA7D_SOR_SET_CONTROL_PIXEL_REPLICATE_MODE_X4                          (0x00000002)
+
+#define NVCA7D_WINDOW_SET_CONTROL(a)                                            (0x00001000 + (a)*0x00000080)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER                                         3:0
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD(i)                                 (0x00000000 +(i))
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD__SIZE_1                            8
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD0                                   (0x00000000)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD1                                   (0x00000001)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD2                                   (0x00000002)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD3                                   (0x00000003)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD4                                   (0x00000004)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD5                                   (0x00000005)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD6                                   (0x00000006)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_HEAD7                                   (0x00000007)
+#define NVCA7D_WINDOW_SET_CONTROL_OWNER_NONE                                    (0x0000000F)
+#define NVCA7D_WINDOW_SET_CONTROL_HIDE                                          8:8
+#define NVCA7D_WINDOW_SET_CONTROL_HIDE_FALSE                                    (0x00000000)
+#define NVCA7D_WINDOW_SET_CONTROL_HIDE_TRUE                                     (0x00000001)
+#define NVCA7D_WINDOW_SET_CONTROL_DISABLE_PHYSICAL_FLIPS                        9:9
+#define NVCA7D_WINDOW_SET_CONTROL_DISABLE_PHYSICAL_FLIPS_FALSE                  (0x00000000)
+#define NVCA7D_WINDOW_SET_CONTROL_DISABLE_PHYSICAL_FLIPS_TRUE                   (0x00000001)
+#define NVCA7D_WINDOW_SET_CONTROL_ALLOW_SUPERFRAME                              10:10
+#define NVCA7D_WINDOW_SET_CONTROL_ALLOW_SUPERFRAME_FALSE                        (0x00000000)
+#define NVCA7D_WINDOW_SET_CONTROL_ALLOW_SUPERFRAME_TRUE                         (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS(a)                         (0x00001004 + (a)*0x00000080)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED1BPP             0:0
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED1BPP_FALSE       (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED1BPP_TRUE        (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED2BPP             1:1
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED2BPP_FALSE       (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED2BPP_TRUE        (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED4BPP             2:2
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED4BPP_FALSE       (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED4BPP_TRUE        (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED8BPP             3:3
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED8BPP_FALSE       (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_RGB_PACKED8BPP_TRUE        (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PACKED422              4:4
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PACKED422_FALSE        (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PACKED422_TRUE         (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PLANAR420              5:5
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PLANAR420_FALSE        (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PLANAR420_TRUE         (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PLANAR444              6:6
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PLANAR444_FALSE        (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_PLANAR444_TRUE         (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR420         7:7
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR420_FALSE   (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR420_TRUE    (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422         8:8
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422_FALSE   (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422_TRUE    (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422R        9:9
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422R_FALSE  (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422R_TRUE   (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR444         10:10
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR444_FALSE   (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR444_TRUE    (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR420          11:11
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR420_FALSE    (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR420_TRUE     (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR444          12:12
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR444_FALSE    (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR444_TRUE     (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR420     13:13
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR420_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR420_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422     14:14
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422R    15:15
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422R_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422R_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR444     16:16
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR444_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR444_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS(a)                 (0x00001008 + (a)*0x00000080)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED1BPP     0:0
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED1BPP_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED1BPP_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED2BPP     1:1
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED2BPP_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED2BPP_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED4BPP     2:2
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED4BPP_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED4BPP_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED8BPP     3:3
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED8BPP_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_RGB_PACKED8BPP_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PACKED422      4:4
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PACKED422_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PACKED422_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PLANAR420      5:5
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PLANAR420_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PLANAR420_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PLANAR444      6:6
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PLANAR444_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_PLANAR444_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR420 7:7
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR420_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR420_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422 8:8
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422R 9:9
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422R_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR422R_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR444 10:10
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR444_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_YUV_SEMI_PLANAR444_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR420  11:11
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR420_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR420_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR444  12:12
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR444_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_PLANAR444_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR420 13:13
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR420_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR420_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422 14:14
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422R 15:15
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422R_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR422R_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR444 16:16
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR444_FALSE (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_ROTATED_FORMAT_USAGE_BOUNDS_EXT_YUV_SEMI_PLANAR444_TRUE (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS(a)                                (0x00001010 + (a)*0x00000080)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_MAX_PIXELS_FETCHED_PER_LINE       14:0
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_ILUT_ALLOWED                      16:16
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_ILUT_ALLOWED_FALSE                (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_ILUT_ALLOWED_TRUE                 (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_TMO_LUT_ALLOWED                   28:28
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_TMO_LUT_ALLOWED_FALSE             (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_TMO_LUT_ALLOWED_TRUE              (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_INPUT_SCALER_TAPS                 22:20
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_INPUT_SCALER_TAPS_TAPS_2          (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_INPUT_SCALER_TAPS_TAPS_5          (0x00000004)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_UPSCALING_ALLOWED                 24:24
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_UPSCALING_ALLOWED_FALSE           (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_UPSCALING_ALLOWED_TRUE            (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_OVERFETCH_ENABLED                 30:30
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_OVERFETCH_ENABLED_FALSE           (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_OVERFETCH_ENABLED_TRUE            (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_LAYOUT                            26:25
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_LAYOUT_PITCH_BLOCKLINEAR          (0x00000000)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_LAYOUT_PITCH                      (0x00000001)
+#define NVCA7D_WINDOW_SET_WINDOW_USAGE_BOUNDS_LAYOUT_BLOCKLINEAR                (0x00000002)
+#define NVCA7D_WINDOW_SET_PHYSICAL(a)                                           (0x00001014 + (a)*0x00000080)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW                                       31:0
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_NONE                                  (0x00000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW0                               (0x00000001)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW1                               (0x00000002)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW2                               (0x00000004)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW3                               (0x00000008)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW4                               (0x00000010)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW5                               (0x00000020)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW6                               (0x00000040)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW7                               (0x00000080)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW8                               (0x00000100)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW9                               (0x00000200)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW10                              (0x00000400)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW11                              (0x00000800)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW12                              (0x00001000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW13                              (0x00002000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW14                              (0x00004000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW15                              (0x00008000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW16                              (0x00010000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW17                              (0x00020000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW18                              (0x00040000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW19                              (0x00080000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW20                              (0x00100000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW21                              (0x00200000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW22                              (0x00400000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW23                              (0x00800000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW24                              (0x01000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW25                              (0x02000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW26                              (0x04000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW27                              (0x08000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW28                              (0x10000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW29                              (0x20000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW30                              (0x40000000)
+#define NVCA7D_WINDOW_SET_PHYSICAL_WINDOW_WINDOW31                              (0x80000000)
+
+#define NVCA7D_HEAD_SET_PROCAMP(a)                                              (0x00002000 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_PROCAMP_COLOR_SPACE                                     1:0
+#define NVCA7D_HEAD_SET_PROCAMP_COLOR_SPACE_RGB                                 (0x00000000)
+#define NVCA7D_HEAD_SET_PROCAMP_COLOR_SPACE_YUV_601                             (0x00000001)
+#define NVCA7D_HEAD_SET_PROCAMP_COLOR_SPACE_YUV_709                             (0x00000002)
+#define NVCA7D_HEAD_SET_PROCAMP_COLOR_SPACE_YUV_2020                            (0x00000003)
+#define NVCA7D_HEAD_SET_PROCAMP_CHROMA_LPF                                      3:3
+#define NVCA7D_HEAD_SET_PROCAMP_CHROMA_LPF_DISABLE                              (0x00000000)
+#define NVCA7D_HEAD_SET_PROCAMP_CHROMA_LPF_ENABLE                               (0x00000001)
+#define NVCA7D_HEAD_SET_PROCAMP_CHROMA_DOWN_V                                   4:4
+#define NVCA7D_HEAD_SET_PROCAMP_CHROMA_DOWN_V_DISABLE                           (0x00000000)
+#define NVCA7D_HEAD_SET_PROCAMP_CHROMA_DOWN_V_ENABLE                            (0x00000001)
+#define NVCA7D_HEAD_SET_PROCAMP_DYNAMIC_RANGE                                   28:28
+#define NVCA7D_HEAD_SET_PROCAMP_DYNAMIC_RANGE_VESA                              (0x00000000)
+#define NVCA7D_HEAD_SET_PROCAMP_DYNAMIC_RANGE_CEA                               (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE(a)                              (0x00002004 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_CRC_MODE                        1:0
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_CRC_MODE_ACTIVE_RASTER          (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_CRC_MODE_COMPLETE_RASTER        (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_CRC_MODE_NON_ACTIVE_RASTER      (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_HSYNC_POLARITY                  2:2
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_HSYNC_POLARITY_POSITIVE_TRUE    (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_HSYNC_POLARITY_NEGATIVE_TRUE    (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_VSYNC_POLARITY                  3:3
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_VSYNC_POLARITY_POSITIVE_TRUE    (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_VSYNC_POLARITY_NEGATIVE_TRUE    (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH                     7:4
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_16_422          (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_18_444          (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_20_422          (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_24_422          (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_24_444          (0x00000004)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_30_444          (0x00000005)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_32_422          (0x00000006)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_36_444          (0x00000007)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_48_444          (0x00000008)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_16_444          (0x00000009)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_PIXEL_DEPTH_BPP_18_444NP        (0x0000000A)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_COLOR_SPACE_OVERRIDE            24:24
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_COLOR_SPACE_OVERRIDE_DISABLE    (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_COLOR_SPACE_OVERRIDE_ENABLE     (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_COLOR_SPACE_FLAG                23:12
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN                  31:26
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN0             (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN1             (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN2             (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN3             (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN4             (0x00000004)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN5             (0x00000005)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN6             (0x00000006)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN7             (0x00000007)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN8             (0x00000008)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN9             (0x00000009)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN10            (0x0000000A)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN11            (0x0000000B)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN12            (0x0000000C)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN13            (0x0000000D)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN14            (0x0000000E)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN15            (0x0000000F)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN16            (0x00000010)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN17            (0x00000011)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN18            (0x00000012)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN19            (0x00000013)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN20            (0x00000014)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN21            (0x00000015)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN22            (0x00000016)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN23            (0x00000017)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN24            (0x00000018)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN25            (0x00000019)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN26            (0x0000001A)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN27            (0x0000001B)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN28            (0x0000001C)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN29            (0x0000001D)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN30            (0x0000001E)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_WIN31            (0x0000001F)
+#define NVCA7D_HEAD_SET_CONTROL_OUTPUT_RESOURCE_EXT_PACKET_WIN_NONE             (0x0000003F)
+#define NVCA7D_HEAD_SET_CONTROL(a)                                              (0x00002008 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_CONTROL_STRUCTURE                                       1:0
+#define NVCA7D_HEAD_SET_CONTROL_STRUCTURE_PROGRESSIVE                           (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO3D_STRUCTURE                              2:2
+#define NVCA7D_HEAD_SET_CONTROL_STEREO3D_STRUCTURE_NORMAL                       (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO3D_STRUCTURE_FRAME_PACKED                 (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_YUV420PACKER                                    3:3
+#define NVCA7D_HEAD_SET_CONTROL_YUV420PACKER_DISABLE                            (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_YUV420PACKER_ENABLE                             (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_MODE                                  11:10
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_MODE_NO_LOCK                          (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_MODE_FRAME_LOCK                       (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_MODE_RASTER_LOCK                      (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN                                   8:4
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_NONE                     (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN(i)                       (0x00000001 +(i))
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN__SIZE_1                  16
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_0                        (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_1                        (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_2                        (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_3                        (0x00000004)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_4                        (0x00000005)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_5                        (0x00000006)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_6                        (0x00000007)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_7                        (0x00000008)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_8                        (0x00000009)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_9                        (0x0000000A)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_A                        (0x0000000B)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_B                        (0x0000000C)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_C                        (0x0000000D)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_D                        (0x0000000E)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_E                        (0x0000000F)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_LOCK_PIN_F                        (0x00000010)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_FLIP_LOCK_0              (0x00000014)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_FLIP_LOCK_1              (0x00000015)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_FLIP_LOCK_2              (0x00000016)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_FLIP_LOCK_3              (0x00000017)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK(i)             (0x00000018 +(i))
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK__SIZE_1        8
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_0              (0x00000018)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_1              (0x00000019)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_2              (0x0000001A)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_3              (0x0000001B)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_4              (0x0000001C)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_5              (0x0000001D)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_6              (0x0000001E)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCK_PIN_INTERNAL_SCAN_LOCK_7              (0x0000001F)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_LOCKOUT_WINDOW                             15:12
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_MODE                                23:22
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_MODE_NO_LOCK                        (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_MODE_FRAME_LOCK                     (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_MODE_RASTER_LOCK                    (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN                                 20:16
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_NONE                   (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN(i)                     (0x00000001 +(i))
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN__SIZE_1                16
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_0                      (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_1                      (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_2                      (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_3                      (0x00000004)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_4                      (0x00000005)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_5                      (0x00000006)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_6                      (0x00000007)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_7                      (0x00000008)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_8                      (0x00000009)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_9                      (0x0000000A)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_A                      (0x0000000B)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_B                      (0x0000000C)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_C                      (0x0000000D)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_D                      (0x0000000E)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_E                      (0x0000000F)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_LOCK_PIN_F                      (0x00000010)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_FLIP_LOCK_0            (0x00000014)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_FLIP_LOCK_1            (0x00000015)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_FLIP_LOCK_2            (0x00000016)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_FLIP_LOCK_3            (0x00000017)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK(i)           (0x00000018 +(i))
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK__SIZE_1      8
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_0            (0x00000018)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_1            (0x00000019)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_2            (0x0000001A)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_3            (0x0000001B)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_4            (0x0000001C)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_5            (0x0000001D)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_6            (0x0000001E)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_LOCK_PIN_INTERNAL_SCAN_LOCK_7            (0x0000001F)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN                                      28:24
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_NONE                        (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN(i)                          (0x00000001 +(i))
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN__SIZE_1                     16
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_0                           (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_1                           (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_2                           (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_3                           (0x00000004)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_4                           (0x00000005)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_5                           (0x00000006)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_6                           (0x00000007)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_7                           (0x00000008)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_8                           (0x00000009)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_9                           (0x0000000A)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_A                           (0x0000000B)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_B                           (0x0000000C)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_C                           (0x0000000D)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_D                           (0x0000000E)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_E                           (0x0000000F)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_LOCK_PIN_F                           (0x00000010)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_FLIP_LOCK_0                 (0x00000014)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_FLIP_LOCK_1                 (0x00000015)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_FLIP_LOCK_2                 (0x00000016)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_FLIP_LOCK_3                 (0x00000017)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK(i)                (0x00000018 +(i))
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK__SIZE_1           8
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_0                 (0x00000018)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_1                 (0x00000019)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_2                 (0x0000001A)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_3                 (0x0000001B)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_4                 (0x0000001C)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_5                 (0x0000001D)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_6                 (0x0000001E)
+#define NVCA7D_HEAD_SET_CONTROL_STEREO_PIN_INTERNAL_SCAN_LOCK_7                 (0x0000001F)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_STEREO_LOCK_MODE                           30:30
+#define NVCA7D_HEAD_SET_CONTROL_SINK_STEREO_LOCK_MODE_DISABLE                   (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_SINK_STEREO_LOCK_MODE_ENABLE                    (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_STEREO_LOCK_MODE                         31:31
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_STEREO_LOCK_MODE_DISABLE                 (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_SOURCE_STEREO_LOCK_MODE_ENABLE                  (0x00000001)
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY(a)                                (0x0000200C + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_HERTZ                             30:0
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_ADJ1000DIV1001                    31:31
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_ADJ1000DIV1001_FALSE              (0x00000000)
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_ADJ1000DIV1001_TRUE               (0x00000001)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL(a)                                       (0x00002018 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_ENABLE                                   0:0
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_ENABLE_DISABLE                           (0x00000000)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_ENABLE_ENABLE                            (0x00000001)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_BITS                                     5:4
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_BITS_TO_6_BITS                           (0x00000000)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_BITS_TO_8_BITS                           (0x00000001)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_BITS_TO_10_BITS                          (0x00000002)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_BITS_TO_12_BITS                          (0x00000003)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_OFFSET_ENABLE                            2:2
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_OFFSET_ENABLE_DISABLE                    (0x00000000)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_OFFSET_ENABLE_ENABLE                     (0x00000001)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE                                     10:8
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE_DYNAMIC_ERR_ACC                     (0x00000000)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE_STATIC_ERR_ACC                      (0x00000001)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE_DYNAMIC_2X2                         (0x00000002)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE_STATIC_2X2                          (0x00000003)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE_TEMPORAL                            (0x00000004)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_MODE_ROUND                               (0x00000005)
+#define NVCA7D_HEAD_SET_DITHER_CONTROL_PHASE                                    13:12
+#define NVCA7D_HEAD_SET_DISPLAY_ID(a,b)                                         (0x00002020 + (a)*0x00000800 + (b)*0x00000004)
+#define NVCA7D_HEAD_SET_DISPLAY_ID_CODE                                         31:0
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX(a)                            (0x00002028 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX_HERTZ                         30:0
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX_ADJ1000DIV1001                31:31
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX_ADJ1000DIV1001_FALSE          (0x00000000)
+#define NVCA7D_HEAD_SET_PIXEL_CLOCK_FREQUENCY_MAX_ADJ1000DIV1001_TRUE           (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS(a)                                    (0x00002030 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_CURSOR                                2:0
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_CURSOR_USAGE_NONE                     (0x00000000)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_CURSOR_USAGE_W32_H32                  (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_CURSOR_USAGE_W64_H64                  (0x00000002)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_CURSOR_USAGE_W128_H128                (0x00000003)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_CURSOR_USAGE_W256_H256                (0x00000004)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OLUT_ALLOWED                          4:4
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OLUT_ALLOWED_FALSE                    (0x00000000)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OLUT_ALLOWED_TRUE                     (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_LTM_ALLOWED                           5:5
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_LTM_ALLOWED_FALSE                     (0x00000000)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_LTM_ALLOWED_TRUE                      (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OUTPUT_SCALER_TAPS                    14:12
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OUTPUT_SCALER_TAPS_TAPS_2             (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OUTPUT_SCALER_TAPS_TAPS_5             (0x00000004)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_UPSCALING_ALLOWED                     8:8
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_UPSCALING_ALLOWED_FALSE               (0x00000000)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_UPSCALING_ALLOWED_TRUE                (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OVERFETCH_ENABLED                     16:16
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OVERFETCH_ENABLED_FALSE               (0x00000000)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_OVERFETCH_ENABLED_TRUE                (0x00000001)
+#define NVCA7D_HEAD_SET_HEAD_USAGE_BOUNDS_ELV_START                             31:17
+#define NVCA7D_HEAD_SET_VIEWPORT_SIZE_IN(a)                                     (0x0000204C + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_VIEWPORT_SIZE_IN_WIDTH                                  14:0
+#define NVCA7D_HEAD_SET_VIEWPORT_SIZE_IN_HEIGHT                                 30:16
+#define NVCA7D_HEAD_SET_VIEWPORT_SIZE_OUT(a)                                    (0x00002058 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_VIEWPORT_SIZE_OUT_WIDTH                                 14:0
+#define NVCA7D_HEAD_SET_VIEWPORT_SIZE_OUT_HEIGHT                                30:16
+#define NVCA7D_HEAD_SET_TILE_MASK(a)                                            (0x00002060 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE                                          7:0
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_NONE                                     (0x00000000)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE0                                    (0x00000001)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE1                                    (0x00000002)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE2                                    (0x00000004)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE3                                    (0x00000008)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE4                                    (0x00000010)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE5                                    (0x00000020)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE6                                    (0x00000040)
+#define NVCA7D_HEAD_SET_TILE_MASK_TILE_TILE7                                    (0x00000080)
+#define NVCA7D_HEAD_SET_RASTER_SIZE(a)                                          (0x00002064 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_RASTER_SIZE_WIDTH                                       15:0
+#define NVCA7D_HEAD_SET_RASTER_SIZE_HEIGHT                                      31:16
+#define NVCA7D_HEAD_SET_RASTER_SYNC_END(a)                                      (0x00002068 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_RASTER_SYNC_END_X                                       14:0
+#define NVCA7D_HEAD_SET_RASTER_SYNC_END_Y                                       30:16
+#define NVCA7D_HEAD_SET_RASTER_BLANK_END(a)                                     (0x0000206C + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_RASTER_BLANK_END_X                                      14:0
+#define NVCA7D_HEAD_SET_RASTER_BLANK_END_Y                                      30:16
+#define NVCA7D_HEAD_SET_RASTER_BLANK_START(a)                                   (0x00002070 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_RASTER_BLANK_START_X                                    14:0
+#define NVCA7D_HEAD_SET_RASTER_BLANK_START_Y                                    30:16
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR(a)                                       (0x0000209C + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_ENABLE                                   31:31
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_ENABLE_DISABLE                           (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_ENABLE_ENABLE                            (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_FORMAT                                   7:0
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_FORMAT_A1R5G5B5                          (0x000000E9)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_FORMAT_A8R8G8B8                          (0x000000CF)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_SIZE                                     9:8
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_SIZE_W32_H32                             (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_SIZE_W64_H64                             (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_SIZE_W128_H128                           (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_SIZE_W256_H256                           (0x00000003)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_HOT_SPOT_X                               19:12
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_HOT_SPOT_Y                               27:20
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION(a)                           (0x000020A0 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_K1                           7:0
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_CURSOR_COLOR_FACTOR_SELECT   11:8
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_CURSOR_COLOR_FACTOR_SELECT_K1 (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_CURSOR_COLOR_FACTOR_SELECT_K1_TIMES_SRC (0x00000005)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_VIEWPORT_COLOR_FACTOR_SELECT 15:12
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_VIEWPORT_COLOR_FACTOR_SELECT_ZERO (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_VIEWPORT_COLOR_FACTOR_SELECT_K1 (0x00000002)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_VIEWPORT_COLOR_FACTOR_SELECT_NEG_K1_TIMES_SRC (0x00000007)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_MODE                         16:16
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_MODE_BLEND                   (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_MODE_XOR                     (0x00000001)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_BYPASS                       20:20
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_BYPASS_DISABLE               (0x00000000)
+#define NVCA7D_HEAD_SET_CONTROL_CURSOR_COMPOSITION_BYPASS_ENABLE                (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_HI_CRC(a)                               (0x00002150 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_HI_CRC_ADDRESS_HI                       31:0
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC(a)                               (0x00002154 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_ADDRESS_LO                       31:4
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_TARGET                           3:2
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_TARGET_IOVA                      (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_TARGET_PHYSICAL_NVM              (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_TARGET_PHYSICAL_PCI              (0x00000002)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_TARGET_PHYSICAL_PCI_COHERENT     (0x00000003)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_ENABLE                           0:0
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_ENABLE_DISABLE                   (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CRC_ENABLE_ENABLE                    (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_HI_OLUT(a)                              (0x00002158 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_HI_OLUT_ADDRESS_HI                      31:0
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT(a)                              (0x0000215C + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_ADDRESS_LO                      31:4
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_TARGET                          3:2
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_TARGET_IOVA                     (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_TARGET_PHYSICAL_NVM             (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_TARGET_PHYSICAL_PCI             (0x00000002)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_TARGET_PHYSICAL_PCI_COHERENT    (0x00000003)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_ENABLE                          0:0
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_ENABLE_DISABLE                  (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_OLUT_ENABLE_ENABLE                   (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_HI_CURSOR(a,b)                          (0x00002170 + (a)*0x00000800 + (b)*0x00000004)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_HI_CURSOR_ADDRESS_HI                    31:0
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR(a,b)                          (0x00002178 + (a)*0x00000800 + (b)*0x00000004)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_ADDRESS_LO                    31:4
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_TARGET                        3:2
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_TARGET_IOVA                   (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_TARGET_PHYSICAL_NVM           (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_TARGET_PHYSICAL_PCI           (0x00000002)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_TARGET_PHYSICAL_PCI_COHERENT  (0x00000003)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_KIND                          1:1
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_KIND_PITCH                    (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_KIND_BLOCKLINEAR              (0x00000001)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_ENABLE                        0:0
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_ENABLE_DISABLE                (0x00000000)
+#define NVCA7D_HEAD_SET_SURFACE_ADDRESS_LO_CURSOR_ENABLE_ENABLE                 (0x00000001)
+#define NVCA7D_HEAD_SET_CRC_CONTROL(a)                                          (0x00002184 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL                         5:0
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_0                   (0x00000000)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_1                   (0x00000001)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_2                   (0x00000002)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_3                   (0x00000003)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_4                   (0x00000004)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_5                   (0x00000005)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_6                   (0x00000006)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_7                   (0x00000007)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_8                   (0x00000008)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_9                   (0x00000009)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_10                  (0x0000000A)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_11                  (0x0000000B)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_12                  (0x0000000C)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_13                  (0x0000000D)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_14                  (0x0000000E)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_15                  (0x0000000F)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_16                  (0x00000010)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_17                  (0x00000011)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_18                  (0x00000012)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_19                  (0x00000013)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_20                  (0x00000014)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_21                  (0x00000015)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_22                  (0x00000016)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_23                  (0x00000017)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_24                  (0x00000018)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_25                  (0x00000019)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_26                  (0x0000001A)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_27                  (0x0000001B)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_28                  (0x0000001C)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_29                  (0x0000001D)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_30                  (0x0000001E)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_WIN_31                  (0x0000001F)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CONTROLLING_CHANNEL_CORE                    (0x00000020)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_EXPECT_BUFFER_COLLAPSE                      8:8
+#define NVCA7D_HEAD_SET_CRC_CONTROL_EXPECT_BUFFER_COLLAPSE_FALSE                (0x00000000)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_EXPECT_BUFFER_COLLAPSE_TRUE                 (0x00000001)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC                                 19:12
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_NONE                            (0x00000000)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SF                              (0x00000030)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR(i)                          (0x00000050 +(i))
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR__SIZE_1                     8
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR0                            (0x00000050)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR1                            (0x00000051)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR2                            (0x00000052)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR3                            (0x00000053)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR4                            (0x00000054)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR5                            (0x00000055)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR6                            (0x00000056)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_PRIMARY_CRC_SOR7                            (0x00000057)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC                               27:20
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_NONE                          (0x00000000)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SF                            (0x00000030)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR(i)                        (0x00000050 +(i))
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR__SIZE_1                   8
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR0                          (0x00000050)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR1                          (0x00000051)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR2                          (0x00000052)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR3                          (0x00000053)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR4                          (0x00000054)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR5                          (0x00000055)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR6                          (0x00000056)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_SECONDARY_CRC_SOR7                          (0x00000057)
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CRC_DURING_SNOOZE                           9:9
+#define NVCA7D_HEAD_SET_CRC_CONTROL_CRC_DURING_SNOOZE_DISABLE                   (0x00000000)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL(a)                                         (0x00002280 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_INTERPOLATE                                0:0
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_INTERPOLATE_DISABLE                        (0x00000000)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_INTERPOLATE_ENABLE                         (0x00000001)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MIRROR                                     1:1
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MIRROR_DISABLE                             (0x00000000)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MIRROR_ENABLE                              (0x00000001)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MODE                                       3:2
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MODE_SEGMENTED                             (0x00000000)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MODE_DIRECT8                               (0x00000001)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_MODE_DIRECT10                              (0x00000002)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_SIZE                                       18:8
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_DIRECT_ROUND                               4:4
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_DIRECT_ROUND_DISABLE                       (0x00000000)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_DIRECT_ROUND_ENABLE                        (0x00000001)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_LEVEL                                      25:20
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_SEGMENT_SIZE_BITS                          5:5
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_SEGMENT_SIZE_BITS_SIZE_3BITS               (0x00000000)
+#define NVCA7D_HEAD_SET_OLUT_CONTROL_SEGMENT_SIZE_BITS_SIZE_4BITS               (0x00000001)
+#define NVCA7D_HEAD_SET_OLUT_FP_NORM_SCALE(a)                                   (0x00002284 + (a)*0x00000800)
+#define NVCA7D_HEAD_SET_OLUT_FP_NORM_SCALE_VALUE                                31:0
+
+#define NVCA7D_TILE_SET_TILE_SIZE(a)                                            (0x00006000 + (a)*0x00000200)
+#define NVCA7D_TILE_SET_TILE_SIZE_START                                         14:0
+#define NVCA7D_TILE_SET_TILE_SIZE_WIDTH                                         30:16
+
+#endif // _clca7d_h
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/class/clca7e.h b/drivers/gpu/drm/nouveau/include/nvhw/class/clca7e.h
new file mode 100644
index 000000000000..ebfb2e48a4f4
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/class/clca7e.h
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef _clca7e_h_
+#define _clca7e_h_
+
+// class methods
+#define NVCA7E_SET_NOTIFIER_CONTROL                                             (0x00000220)
+#define NVCA7E_SET_NOTIFIER_CONTROL_MODE                                        0:0
+#define NVCA7E_SET_NOTIFIER_CONTROL_MODE_WRITE                                  (0x00000000)
+#define NVCA7E_SET_NOTIFIER_CONTROL_MODE_WRITE_AWAKEN                           (0x00000001)
+#define NVCA7E_SET_SIZE                                                         (0x00000224)
+#define NVCA7E_SET_SIZE_WIDTH                                                   15:0
+#define NVCA7E_SET_SIZE_HEIGHT                                                  31:16
+#define NVCA7E_SET_STORAGE                                                      (0x00000228)
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT                                         3:0
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT_NVD_BLOCK_HEIGHT_ONE_GOB                (0x00000000)
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT_NVD_BLOCK_HEIGHT_TWO_GOBS               (0x00000001)
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT_NVD_BLOCK_HEIGHT_FOUR_GOBS              (0x00000002)
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT_NVD_BLOCK_HEIGHT_EIGHT_GOBS             (0x00000003)
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT_NVD_BLOCK_HEIGHT_SIXTEEN_GOBS           (0x00000004)
+#define NVCA7E_SET_STORAGE_BLOCK_HEIGHT_NVD_BLOCK_HEIGHT_THIRTYTWO_GOBS         (0x00000005)
+#define NVCA7E_SET_PARAMS                                                       (0x0000022C)
+#define NVCA7E_SET_PARAMS_FORMAT                                                7:0
+#define NVCA7E_SET_PARAMS_FORMAT_I8                                             (0x0000001E)
+#define NVCA7E_SET_PARAMS_FORMAT_R4G4B4A4                                       (0x0000002F)
+#define NVCA7E_SET_PARAMS_FORMAT_R5G6B5                                         (0x000000E8)
+#define NVCA7E_SET_PARAMS_FORMAT_A1R5G5B5                                       (0x000000E9)
+#define NVCA7E_SET_PARAMS_FORMAT_R5G5B5A1                                       (0x0000002E)
+#define NVCA7E_SET_PARAMS_FORMAT_A8R8G8B8                                       (0x000000CF)
+#define NVCA7E_SET_PARAMS_FORMAT_X8R8G8B8                                       (0x000000E6)
+#define NVCA7E_SET_PARAMS_FORMAT_A8B8G8R8                                       (0x000000D5)
+#define NVCA7E_SET_PARAMS_FORMAT_X8B8G8R8                                       (0x000000F9)
+#define NVCA7E_SET_PARAMS_FORMAT_A2R10G10B10                                    (0x000000DF)
+#define NVCA7E_SET_PARAMS_FORMAT_A2B10G10R10                                    (0x000000D1)
+#define NVCA7E_SET_PARAMS_FORMAT_R16_G16_B16_A16_NVBIAS                         (0x00000023)
+#define NVCA7E_SET_PARAMS_FORMAT_R16_G16_B16_A16                                (0x000000C6)
+#define NVCA7E_SET_PARAMS_FORMAT_RF16_GF16_BF16_AF16                            (0x000000CA)
+#define NVCA7E_SET_PARAMS_FORMAT_Y8_U8__Y8_V8_N422                              (0x00000028)
+#define NVCA7E_SET_PARAMS_FORMAT_U8_Y8__V8_Y8_N422                              (0x00000029)
+#define NVCA7E_SET_PARAMS_FORMAT_Y8___U8V8_N444                                 (0x00000035)
+#define NVCA7E_SET_PARAMS_FORMAT_Y8___U8V8_N422                                 (0x00000036)
+#define NVCA7E_SET_PARAMS_FORMAT_Y8___V8U8_N420                                 (0x00000038)
+#define NVCA7E_SET_PARAMS_FORMAT_Y8___U8___V8_N444                              (0x0000003A)
+#define NVCA7E_SET_PARAMS_FORMAT_Y8___U8___V8_N420                              (0x0000003B)
+#define NVCA7E_SET_PARAMS_FORMAT_Y10___U10V10_N444                              (0x00000055)
+#define NVCA7E_SET_PARAMS_FORMAT_Y10___U10V10_N422                              (0x00000056)
+#define NVCA7E_SET_PARAMS_FORMAT_Y10___V10U10_N420                              (0x00000058)
+#define NVCA7E_SET_PARAMS_FORMAT_Y12___U12V12_N444                              (0x00000075)
+#define NVCA7E_SET_PARAMS_FORMAT_Y12___U12V12_N422                              (0x00000076)
+#define NVCA7E_SET_PARAMS_FORMAT_Y12___V12U12_N420                              (0x00000078)
+#define NVCA7E_SET_PARAMS_CLAMP_BEFORE_BLEND                                    18:18
+#define NVCA7E_SET_PARAMS_CLAMP_BEFORE_BLEND_DISABLE                            (0x00000000)
+#define NVCA7E_SET_PARAMS_CLAMP_BEFORE_BLEND_ENABLE                             (0x00000001)
+#define NVCA7E_SET_PARAMS_SWAP_UV                                               19:19
+#define NVCA7E_SET_PARAMS_SWAP_UV_DISABLE                                       (0x00000000)
+#define NVCA7E_SET_PARAMS_SWAP_UV_ENABLE                                        (0x00000001)
+#define NVCA7E_SET_PARAMS_FMT_ROUNDING_MODE                                     22:22
+#define NVCA7E_SET_PARAMS_FMT_ROUNDING_MODE_ROUND_TO_NEAREST                    (0x00000000)
+#define NVCA7E_SET_PARAMS_FMT_ROUNDING_MODE_ROUND_DOWN                          (0x00000001)
+#define NVCA7E_SET_PLANAR_STORAGE(b)                                            (0x00000230 + (b)*0x00000004)
+#define NVCA7E_SET_PLANAR_STORAGE_PITCH                                         12:0
+#define NVCA7E_SET_POINT_IN(b)                                                  (0x00000290 + (b)*0x00000004)
+#define NVCA7E_SET_POINT_IN_X                                                   15:0
+#define NVCA7E_SET_POINT_IN_Y                                                   31:16
+#define NVCA7E_SET_SIZE_IN                                                      (0x00000298)
+#define NVCA7E_SET_SIZE_IN_WIDTH                                                15:0
+#define NVCA7E_SET_SIZE_IN_HEIGHT                                               31:16
+#define NVCA7E_SET_SIZE_OUT                                                     (0x000002A4)
+#define NVCA7E_SET_SIZE_OUT_WIDTH                                               15:0
+#define NVCA7E_SET_SIZE_OUT_HEIGHT                                              31:16
+#define NVCA7E_SET_PRESENT_CONTROL                                              (0x00000308)
+#define NVCA7E_SET_PRESENT_CONTROL_MIN_PRESENT_INTERVAL                         3:0
+#define NVCA7E_SET_PRESENT_CONTROL_BEGIN_MODE                                   6:4
+#define NVCA7E_SET_PRESENT_CONTROL_BEGIN_MODE_NON_TEARING                       (0x00000000)
+#define NVCA7E_SET_PRESENT_CONTROL_BEGIN_MODE_IMMEDIATE                         (0x00000001)
+#define NVCA7E_SET_PRESENT_CONTROL_TIMESTAMP_MODE                               8:8
+#define NVCA7E_SET_PRESENT_CONTROL_TIMESTAMP_MODE_DISABLE                       (0x00000000)
+#define NVCA7E_SET_PRESENT_CONTROL_TIMESTAMP_MODE_ENABLE                        (0x00000001)
+#define NVCA7E_SET_PRESENT_CONTROL_STEREO_MODE                                  13:12
+#define NVCA7E_SET_PRESENT_CONTROL_STEREO_MODE_MONO                             (0x00000000)
+#define NVCA7E_SET_PRESENT_CONTROL_STEREO_MODE_PAIR_FLIP                        (0x00000001)
+#define NVCA7E_SET_PRESENT_CONTROL_STEREO_MODE_AT_ANY_FRAME                     (0x00000002)
+#define NVCA7E_SET_ILUT_CONTROL                                                 (0x00000440)
+#define NVCA7E_SET_ILUT_CONTROL_INTERPOLATE                                     0:0
+#define NVCA7E_SET_ILUT_CONTROL_INTERPOLATE_DISABLE                             (0x00000000)
+#define NVCA7E_SET_ILUT_CONTROL_INTERPOLATE_ENABLE                              (0x00000001)
+#define NVCA7E_SET_ILUT_CONTROL_MIRROR                                          1:1
+#define NVCA7E_SET_ILUT_CONTROL_MIRROR_DISABLE                                  (0x00000000)
+#define NVCA7E_SET_ILUT_CONTROL_MIRROR_ENABLE                                   (0x00000001)
+#define NVCA7E_SET_ILUT_CONTROL_MODE                                            3:2
+#define NVCA7E_SET_ILUT_CONTROL_MODE_SEGMENTED                                  (0x00000000)
+#define NVCA7E_SET_ILUT_CONTROL_MODE_DIRECT8                                    (0x00000001)
+#define NVCA7E_SET_ILUT_CONTROL_MODE_DIRECT10                                   (0x00000002)
+#define NVCA7E_SET_ILUT_CONTROL_SIZE                                            18:8
+#define NVCA7E_SET_SURFACE_ADDRESS_HI_NOTIFIER                                  (0x00000650)
+#define NVCA7E_SET_SURFACE_ADDRESS_HI_NOTIFIER_ADDRESS_HI                       31:0
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER                                  (0x00000654)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_ADDRESS_LO                       31:4
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET                           3:2
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_IOVA                      (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_PHYSICAL_NVM              (0x00000001)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_PHYSICAL_PCI              (0x00000002)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_TARGET_PHYSICAL_PCI_COHERENT     (0x00000003)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_ENABLE                           0:0
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_ENABLE_DISABLE                   (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_NOTIFIER_ENABLE_ENABLE                    (0x00000001)
+#define NVCA7E_SET_SURFACE_ADDRESS_HI_ISO(b)                                    (0x00000658 + (b)*0x00000004)
+#define NVCA7E_SET_SURFACE_ADDRESS_HI_ISO_ADDRESS_HI                            31:0
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO(b)                                    (0x00000670 + (b)*0x00000004)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_ADDRESS_LO                            31:4
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_TARGET                                3:2
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_TARGET_IOVA                           (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_TARGET_PHYSICAL_NVM                   (0x00000001)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_TARGET_PHYSICAL_PCI                   (0x00000002)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_TARGET_PHYSICAL_PCI_COHERENT          (0x00000003)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_KIND                                  1:1
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_KIND_PITCH                            (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_KIND_BLOCKLINEAR                      (0x00000001)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_ENABLE                                0:0
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_ENABLE_DISABLE                        (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ISO_ENABLE_ENABLE                         (0x00000001)
+#define NVCA7E_SET_SURFACE_ADDRESS_HI_ILUT                                      (0x00000688)
+#define NVCA7E_SET_SURFACE_ADDRESS_HI_ILUT_ADDRESS_HI                           31:0
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT                                      (0x0000068C)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_ADDRESS_LO                           31:4
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_TARGET                               3:2
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_TARGET_IOVA                          (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_TARGET_PHYSICAL_NVM                  (0x00000001)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_TARGET_PHYSICAL_PCI                  (0x00000002)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_TARGET_PHYSICAL_PCI_COHERENT         (0x00000003)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_ENABLE                               0:0
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_ENABLE_DISABLE                       (0x00000000)
+#define NVCA7E_SET_SURFACE_ADDRESS_LO_ILUT_ENABLE_ENABLE                        (0x00000001)
+
+#endif // _clca7e_h
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gb100/dev_hshub_base.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb100/dev_hshub_base.h
new file mode 100644
index 000000000000..c9d74bd95e0b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb100/dev_hshub_base.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gb100_dev_hshub_base_h__
+#define __gb100_dev_hshub_base_h__
+
+#define NV_PFB_HSHUB0      0x00870fff:0x00870000
+
+#define NV_PFB_HSHUB                                   0x00000FFF:0x00000000 /* RW--D */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_LO              0x00000E50 /* RW-4R */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR                31:0 /* RWIVF */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR_INIT     0x00000000 /* RWI-V */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR_MASK     0xFFFFFF00 /* ----V */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_HI              0x00000E54 /* RW-4R */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR                31:0 /* RWIVF */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_INIT     0x00000000 /* RWI-V */
+#define NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_MASK     0x000FFFFF /* ----V */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_LO           0x000006C0 /* RW-4R */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR             31:0 /* RWIVF */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR_INIT  0x00000000 /* RWI-V */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR_MASK  0xFFFFFF00 /* ----V */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_HI           0x000006C4 /* RW-4R */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR             31:0 /* RWIVF */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_INIT  0x00000000 /* RWI-V */
+#define NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_MASK  0x000FFFFF /* ----V */
+
+#endif // __gb100_dev_hshub_base_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gb10b/dev_fbhub.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb10b/dev_fbhub.h
new file mode 100644
index 000000000000..4d0bb8e14298
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb10b/dev_fbhub.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gb10b_dev_fb_h__
+#define __gb10b_dev_fb_h__
+
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_LO                                0x008a1d58 /* RW-4R */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR                                  31:0 /* RWIVF */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR_INIT                       0x00000000 /* RWI-V */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR_MASK                       0xffffff00 /* RW--V */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_HI                                0x008a1d5c /* RW-4R */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR                                  31:0 /* RWIVF */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_INIT                       0x00000000 /* RWI-V */
+#define NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_MASK                       0x000fffff /* RW--V */
+
+#endif // __gb10b_dev_fb_h__
+
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gb202/dev_ce.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb202/dev_ce.h
new file mode 100644
index 000000000000..b09f04b31738
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb202/dev_ce.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gb202_dev_ce_h__
+#define __gb202_dev_ce_h__
+
+#define NV_CE_GRCE_MASK                                                       0x001040d8 /* C--4R */
+#define NV_CE_GRCE_MASK_VALUE                                                        9:0 /* C--VF */
+#define NV_CE_GRCE_MASK_VALUE_INIT                                                 0x00f /* C---V */
+
+#endif // __gb202_dev_ce_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gb202/dev_therm.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb202/dev_therm.h
new file mode 100644
index 000000000000..ed359cb528fb
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gb202/dev_therm.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gb202_dev_therm_h__
+#define __gb202_dev_therm_h__
+
+#define NV_THERM_I2CS_SCRATCH                                                  0x00ad00bc /* RW-4R */
+#define NV_THERM_I2CS_SCRATCH_DATA                                                   31:0 /* RWIVF */
+#define NV_THERM_I2CS_SCRATCH_DATA_INIT                                        0x00000000 /* RWI-V */
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE                                   NV_THERM_I2CS_SCRATCH
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS                                             31:0
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS_SUCCESS                               0x000000FF
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS_FAILED                                0x00000000
+
+#endif // __gb202_dev_therm_h__
+
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_falcon_v4.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_falcon_v4.h
new file mode 100644
index 000000000000..52171b412aa1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_falcon_v4.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_falcon_v4_h__
+#define __gh100_dev_falcon_v4_h__
+
+#define NV_PFALCON_FALCON_MAILBOX0                                                                     0x00000040     /* RW-4R */
+#define NV_PFALCON_FALCON_MAILBOX0_DATA                                                                31:0           /* RWIVF */
+#define NV_PFALCON_FALCON_MAILBOX0_DATA_INIT                                                           0x00000000     /* RWI-V */
+#define NV_PFALCON_FALCON_MAILBOX1                                                                     0x00000044     /* RW-4R */
+#define NV_PFALCON_FALCON_MAILBOX1_DATA                                                                31:0           /* RWIVF */
+#define NV_PFALCON_FALCON_MAILBOX1_DATA_INIT                                                           0x00000000     /* RWI-V */
+
+#define NV_PFALCON_FALCON_HWCFG2                                                                       0x000000f4     /* R--4R */
+#define NV_PFALCON_FALCON_HWCFG2_RISCV_BR_PRIV_LOCKDOWN                                                13:13          /* R--VF */
+#define NV_PFALCON_FALCON_HWCFG2_RISCV_BR_PRIV_LOCKDOWN_LOCK                                           0x00000001     /* R---V */
+#define NV_PFALCON_FALCON_HWCFG2_RISCV_BR_PRIV_LOCKDOWN_UNLOCK                                         0x00000000     /* R---V */
+
+#endif // __gh100_dev_falcon_v4_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_fb.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_fb.h
new file mode 100644
index 000000000000..819f09465952
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_fb.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_fb_h_
+#define __gh100_dev_fb_h_
+
+#define NV_PFB_NISO_FLUSH_SYSMEM_ADDR_SHIFT                       8 /*       */
+#define NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_LO           0x00100A34 /* RW-4R */
+#define NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_LO_ADR             31:0 /* RWIVF */
+#define NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_HI           0x00100A38 /* RW-4R */
+#define NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR             31:0 /* RWIVF */
+#define NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_HI_ADR_MASK  0x000FFFFF /* ----V */
+
+#endif // __gh100_dev_fb_h_
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_fsp_pri.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_fsp_pri.h
new file mode 100644
index 000000000000..e9507242cae5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_fsp_pri.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_fsp_pri_h__
+#define __gh100_dev_fsp_pri_h__
+
+#define NV_PFSP             0x8F3FFF:0x8F0000 /* RW--D */
+
+#define NV_PFSP_MSGQ_HEAD(i)                                                                             (0x008F2c80+(i)*8) /* RW-4A */
+#define NV_PFSP_MSGQ_HEAD__SIZE_1                                                                        8              /*       */
+#define NV_PFSP_MSGQ_HEAD_VAL                                                                            31:0           /* RWIUF */
+#define NV_PFSP_MSGQ_HEAD_VAL_INIT                                                                       0x00000000     /* RWI-V */
+#define NV_PFSP_MSGQ_TAIL(i)                                                                             (0x008F2c84+(i)*8) /* RW-4A */
+#define NV_PFSP_MSGQ_TAIL__SIZE_1                                                                        8              /*       */
+#define NV_PFSP_MSGQ_TAIL_VAL                                                                            31:0           /* RWIUF */
+#define NV_PFSP_MSGQ_TAIL_VAL_INIT                                                                       0x00000000     /* RWI-V */
+
+#define NV_PFSP_QUEUE_HEAD(i)                                                                            (0x008F2c00+(i)*8) /* RW-4A */
+#define NV_PFSP_QUEUE_HEAD__SIZE_1                                                                       8              /*       */
+#define NV_PFSP_QUEUE_HEAD_ADDRESS                                                                       31:0           /* RWIVF */
+#define NV_PFSP_QUEUE_HEAD_ADDRESS_INIT                                                                  0x00000000     /* RWI-V */
+#define NV_PFSP_QUEUE_TAIL(i)                                                                            (0x008F2c04+(i)*8) /* RW-4A */
+#define NV_PFSP_QUEUE_TAIL__SIZE_1                                                                       8              /*       */
+#define NV_PFSP_QUEUE_TAIL_ADDRESS                                                                       31:0           /* RWIVF */
+#define NV_PFSP_QUEUE_TAIL_ADDRESS_INIT                                                                  0x00000000     /* RWI-V */
+
+#endif // __gh100_dev_fsp_pri_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_mmu.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_mmu.h
new file mode 100644
index 000000000000..6707e0e3b96b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_mmu.h
@@ -0,0 +1,173 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_mmu_h__
+#define __gh100_dev_mmu_h__
+
+#define NV_MMU_PTE                                                      /* ----G */
+#define NV_MMU_PTE_APERTURE                           (1*32+2):(1*32+1) /* RWXVF */
+#define NV_MMU_PTE_APERTURE_VIDEO_MEMORY                     0x00000000 /* RW--V */
+#define NV_MMU_PTE_APERTURE_PEER_MEMORY                      0x00000001 /* RW--V */
+#define NV_MMU_PTE_APERTURE_SYSTEM_COHERENT_MEMORY           0x00000002 /* RW--V */
+#define NV_MMU_PTE_APERTURE_SYSTEM_NON_COHERENT_MEMORY       0x00000003 /* RW--V */
+#define NV_MMU_PTE_KIND                              (1*32+7):(1*32+4) /* RWXVF */
+#define NV_MMU_PTE_KIND_INVALID                       0x07 /* R---V */
+#define NV_MMU_PTE_KIND_PITCH                         0x00 /* R---V */
+#define NV_MMU_PTE_KIND_GENERIC_MEMORY                                                  0x6 /* R---V */
+#define NV_MMU_PTE_KIND_Z16                                                             0x1 /* R---V */
+#define NV_MMU_PTE_KIND_S8                                                              0x2 /* R---V */
+#define NV_MMU_PTE_KIND_S8Z24                                                           0x3 /* R---V */
+#define NV_MMU_PTE_KIND_ZF32_X24S8                                                      0x4 /* R---V */
+#define NV_MMU_PTE_KIND_Z24S8                                                           0x5 /* R---V */
+#define NV_MMU_PTE_KIND_GENERIC_MEMORY_COMPRESSIBLE                                     0x8 /* R---V */
+#define NV_MMU_PTE_KIND_GENERIC_MEMORY_COMPRESSIBLE_DISABLE_PLC                         0x9 /* R---V */
+#define NV_MMU_PTE_KIND_S8_COMPRESSIBLE_DISABLE_PLC                                     0xA /* R---V */
+#define NV_MMU_PTE_KIND_Z16_COMPRESSIBLE_DISABLE_PLC                                    0xB /* R---V */
+#define NV_MMU_PTE_KIND_S8Z24_COMPRESSIBLE_DISABLE_PLC                                  0xC /* R---V */
+#define NV_MMU_PTE_KIND_ZF32_X24S8_COMPRESSIBLE_DISABLE_PLC                             0xD /* R---V */
+#define NV_MMU_PTE_KIND_Z24S8_COMPRESSIBLE_DISABLE_PLC                                  0xE /* R---V */
+#define NV_MMU_PTE_KIND_SMSKED_MESSAGE                                                  0xF /* R---V */
+
+#define NV_MMU_VER3_PDE                                                      /* ----G */
+#define NV_MMU_VER3_PDE_IS_PTE                                           0:0 /* RWXVF */
+#define NV_MMU_VER3_PDE_IS_PTE_TRUE                                      0x1 /* RW--V */
+#define NV_MMU_VER3_PDE_IS_PTE_FALSE                                     0x0 /* RW--V */
+#define NV_MMU_VER3_PDE_VALID                                            0:0 /* RWXVF */
+#define NV_MMU_VER3_PDE_VALID_TRUE                                       0x1 /* RW--V */
+#define NV_MMU_VER3_PDE_VALID_FALSE                                      0x0 /* RW--V */
+#define NV_MMU_VER3_PDE_APERTURE                                         2:1 /* RWXVF */
+#define NV_MMU_VER3_PDE_APERTURE_INVALID                          0x00000000 /* RW--V */
+#define NV_MMU_VER3_PDE_APERTURE_VIDEO_MEMORY                     0x00000001 /* RW--V */
+#define NV_MMU_VER3_PDE_APERTURE_SYSTEM_COHERENT_MEMORY           0x00000002 /* RW--V */
+#define NV_MMU_VER3_PDE_APERTURE_SYSTEM_NON_COHERENT_MEMORY       0x00000003 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF                                                                        5:3 /* RWXVF */
+#define NV_MMU_VER3_PDE_PCF_VALID_CACHED_ATS_ALLOWED__OR__INVALID_ATS_ALLOWED               0x00000000 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_CACHED_ATS_ALLOWED                                        0x00000000 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_INVALID_ATS_ALLOWED                                             0x00000000 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_UNCACHED_ATS_ALLOWED__OR__SPARSE_ATS_ALLOWED              0x00000001 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_UNCACHED_ATS_ALLOWED                                      0x00000001 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_SPARSE_ATS_ALLOWED                                              0x00000001 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_CACHED_ATS_NOT_ALLOWED__OR__INVALID_ATS_NOT_ALLOWED       0x00000002 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_CACHED_ATS_NOT_ALLOWED                                    0x00000002 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_INVALID_ATS_NOT_ALLOWED                                         0x00000002 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_UNCACHED_ATS_NOT_ALLOWED__OR__SPARSE_ATS_NOT_ALLOWED      0x00000003 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_VALID_UNCACHED_ATS_NOT_ALLOWED                                  0x00000003 /* RW--V */
+#define NV_MMU_VER3_PDE_PCF_SPARSE_ATS_NOT_ALLOWED                                          0x00000003 /* RW--V */
+#define NV_MMU_VER3_PDE_ADDRESS                                             51:12 /* RWXVF */
+#define NV_MMU_VER3_PDE_ADDRESS_SHIFT                                  0x0000000c /*       */
+#define NV_MMU_VER3_PDE__SIZE                                              8
+
+#define NV_MMU_VER3_DUAL_PDE                                                      /* ----G */
+#define NV_MMU_VER3_DUAL_PDE_IS_PTE                                           0:0 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_IS_PTE_TRUE                                      0x1 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_IS_PTE_FALSE                                     0x0 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_VALID                                            0:0 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_VALID_TRUE                                       0x1 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_VALID_FALSE                                      0x0 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_BIG                                     2:1 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_BIG_INVALID                      0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_BIG_VIDEO_MEMORY                 0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_BIG_SYSTEM_COHERENT_MEMORY       0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_BIG_SYSTEM_NON_COHERENT_MEMORY   0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG                                                                        5:3 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_CACHED_ATS_ALLOWED__OR__INVALID_ATS_ALLOWED               0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_CACHED_ATS_ALLOWED                                        0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_INVALID_ATS_ALLOWED                                             0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_UNCACHED_ATS_ALLOWED__OR__SPARSE_ATS_ALLOWED              0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_UNCACHED_ATS_ALLOWED                                      0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_SPARSE_ATS_ALLOWED                                              0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_CACHED_ATS_NOT_ALLOWED__OR__INVALID_ATS_NOT_ALLOWED       0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_CACHED_ATS_NOT_ALLOWED                                    0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_INVALID_ATS_NOT_ALLOWED                                         0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_UNCACHED_ATS_NOT_ALLOWED__OR__SPARSE_ATS_NOT_ALLOWED      0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_VALID_UNCACHED_ATS_NOT_ALLOWED                                  0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_BIG_SPARSE_ATS_NOT_ALLOWED                                          0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_ADDRESS_BIG                                     51:8 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_SMALL                                 66:65 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_SMALL_INVALID                    0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_SMALL_VIDEO_MEMORY               0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_SMALL_SYSTEM_COHERENT_MEMORY     0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_APERTURE_SMALL_SYSTEM_NON_COHERENT_MEMORY 0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL                                                                      69:67 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_CACHED_ATS_ALLOWED__OR__INVALID_ATS_ALLOWED               0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_CACHED_ATS_ALLOWED                                        0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_INVALID_ATS_ALLOWED                                             0x00000000 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_UNCACHED_ATS_ALLOWED__OR__SPARSE_ATS_ALLOWED              0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_UNCACHED_ATS_ALLOWED                                      0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_SPARSE_ATS_ALLOWED                                              0x00000001 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_CACHED_ATS_NOT_ALLOWED__OR__INVALID_ATS_NOT_ALLOWED       0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_CACHED_ATS_NOT_ALLOWED                                    0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_INVALID_ATS_NOT_ALLOWED                                         0x00000002 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_UNCACHED_ATS_NOT_ALLOWED__OR__SPARSE_ATS_NOT_ALLOWED      0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_VALID_UNCACHED_ATS_NOT_ALLOWED                                  0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_PCF_SMALL_SPARSE_ATS_NOT_ALLOWED                                          0x00000003 /* RW--V */
+#define NV_MMU_VER3_DUAL_PDE_ADDRESS_SMALL                                 115:76 /* RWXVF */
+#define NV_MMU_VER3_DUAL_PDE_ADDRESS_SHIFT                             0x0000000c /*       */
+#define NV_MMU_VER3_DUAL_PDE_ADDRESS_BIG_SHIFT 8 /*       */
+#define NV_MMU_VER3_DUAL_PDE__SIZE                                             16
+
+#define NV_MMU_VER3_PTE                                                      /* ----G */
+#define NV_MMU_VER3_PTE_VALID                                            0:0 /* RWXVF */
+#define NV_MMU_VER3_PTE_VALID_TRUE                                       0x1 /* RW--V */
+#define NV_MMU_VER3_PTE_VALID_FALSE                                      0x0 /* RW--V */
+#define NV_MMU_VER3_PTE_APERTURE                                         2:1 /* RWXVF */
+#define NV_MMU_VER3_PTE_APERTURE_VIDEO_MEMORY                     0x00000000 /* RW--V */
+#define NV_MMU_VER3_PTE_APERTURE_PEER_MEMORY                      0x00000001 /* RW--V */
+#define NV_MMU_VER3_PTE_APERTURE_SYSTEM_COHERENT_MEMORY           0x00000002 /* RW--V */
+#define NV_MMU_VER3_PTE_APERTURE_SYSTEM_NON_COHERENT_MEMORY       0x00000003 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF                                                                        7:3 /* RWXVF */
+#define NV_MMU_VER3_PTE_PCF_INVALID                                                         0x00000000 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_SPARSE                                                          0x00000001 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_MAPPING_NOWHERE                                                 0x00000002 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_NO_VALID_4KB_PAGE                                               0x00000003 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_ATOMIC_CACHED_ACE                                    0x00000000 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_ATOMIC_UNCACHED_ACE                                  0x00000001 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_ATOMIC_CACHED_ACE                                  0x00000002 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_ATOMIC_UNCACHED_ACE                                0x00000003 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_ATOMIC_CACHED_ACE                                    0x00000004 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_ATOMIC_UNCACHED_ACE                                   0x00000005 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_ATOMIC_CACHED_ACE                                  0x00000006 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_ATOMIC_UNCACHED_ACE                                0x00000007 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_NO_ATOMIC_CACHED_ACE                                 0x00000008 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_NO_ATOMIC_UNCACHED_ACE                               0x00000009 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_NO_ATOMIC_CACHED_ACE                               0x0000000A /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_NO_ATOMIC_UNCACHED_ACE                             0x0000000B /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_NO_ATOMIC_CACHED_ACE                                 0x0000000C /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_NO_ATOMIC_UNCACHED_ACE                               0x0000000D /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_NO_ATOMIC_CACHED_ACE                               0x0000000E /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_NO_ATOMIC_UNCACHED_ACE                             0x0000000F /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_ATOMIC_CACHED_ACD                                    0x00000010 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_ATOMIC_UNCACHED_ACD                                  0x00000011 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_ATOMIC_CACHED_ACD                                  0x00000012 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_ATOMIC_UNCACHED_ACD                                0x00000013 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_ATOMIC_CACHED_ACD                                    0x00000014 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_ATOMIC_UNCACHED_ACD                                  0x00000015 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_ATOMIC_CACHED_ACD                                  0x00000016 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_ATOMIC_UNCACHED_ACD                                0x00000017 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_NO_ATOMIC_CACHED_ACD                                 0x00000018 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RW_NO_ATOMIC_UNCACHED_ACD                               0x00000019 /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_NO_ATOMIC_CACHED_ACD                               0x0000001A /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_NO_ATOMIC_UNCACHED_ACD                             0x0000001B /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_NO_ATOMIC_CACHED_ACD                                 0x0000001C /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_REGULAR_RO_NO_ATOMIC_UNCACHED_ACD                               0x0000001D /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_NO_ATOMIC_CACHED_ACD                               0x0000001E /* RW--V */
+#define NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_NO_ATOMIC_UNCACHED_ACD                             0x0000001F /* RW--V */
+#define NV_MMU_VER3_PTE_KIND                                           11:8 /* RWXVF */
+#define NV_MMU_VER3_PTE_ADDRESS                                         51:12 /* RWXVF */
+#define NV_MMU_VER3_PTE_ADDRESS_SYS                                     51:12 /* RWXVF */
+#define NV_MMU_VER3_PTE_ADDRESS_PEER                                    51:12 /* RWXVF */
+#define NV_MMU_VER3_PTE_ADDRESS_VID                                     39:12 /* RWXVF */
+#define NV_MMU_VER3_PTE_PEER_ID                63:(64-3) /* RWXVF */
+#define NV_MMU_VER3_PTE_PEER_ID_0                                 0x00000000 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_1                                 0x00000001 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_2                                 0x00000002 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_3                                 0x00000003 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_4                                 0x00000004 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_5                                 0x00000005 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_6                                 0x00000006 /* RW--V */
+#define NV_MMU_VER3_PTE_PEER_ID_7                                 0x00000007 /* RW--V */
+#define NV_MMU_VER3_PTE_ADDRESS_SHIFT                             0x0000000c /*       */
+#define NV_MMU_VER3_PTE__SIZE                                              8
+
+#endif // __gh100_dev_mmu_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_riscv_pri.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_riscv_pri.h
new file mode 100644
index 000000000000..8ff4663168d2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_riscv_pri.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_riscv_pri_h__
+#define __gh100_dev_riscv_pri_h__
+
+#define NV_PRISCV_RISCV_CPUCTL                                                                         0x00000388     /* RW-4R */
+#define NV_PRISCV_RISCV_CPUCTL_HALTED                                                                  4:4            /* R-IVF */
+#define NV_PRISCV_RISCV_CPUCTL_HALTED_INIT                                                             0x00000001     /* R-I-V */
+#define NV_PRISCV_RISCV_CPUCTL_HALTED_TRUE                                                             0x00000001     /* R---V */
+#define NV_PRISCV_RISCV_CPUCTL_HALTED_FALSE                                                            0x00000000     /* R---V */
+
+#endif // __gh100_dev_riscv_pri_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_therm.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_therm.h
new file mode 100644
index 000000000000..49b4816cb00b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_therm.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_therm_h__
+#define __gh100_dev_therm_h__
+
+#define NV_THERM_I2CS_SCRATCH                                                  0x000200bc /* RW-4R */
+#define NV_THERM_I2CS_SCRATCH_DATA                                                   31:0 /* RWIVF */
+#define NV_THERM_I2CS_SCRATCH_DATA_INIT                                        0x00000000 /* RWI-V */
+
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE                                   NV_THERM_I2CS_SCRATCH
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS                                             31:0
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS_SUCCESS                               0x000000FF
+#define NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS_FAILED                                0x00000000
+
+#endif // __gh100_dev_therm_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_xtl_ep_pri.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_xtl_ep_pri.h
new file mode 100644
index 000000000000..12b49e9894a2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/dev_xtl_ep_pri.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_dev_xtl_ep_pri_h__
+#define __gh100_dev_xtl_ep_pri_h__
+
+#define NV_EP_PCFGM                                                              0x92FFF:0x92000        /* RW--D */
+
+#endif // __gh100_dev_xtl_ep_pri_h__
diff --git a/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/pri_nv_xal_ep.h b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/pri_nv_xal_ep.h
new file mode 100644
index 000000000000..1a891bd33fa3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvhw/ref/gh100/pri_nv_xal_ep.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __gh100_pri_nv_xal_ep_h__
+#define __gh100_pri_nv_xal_ep_h__
+
+#define NV_XAL_EP_BAR0_WINDOW_BASE_SHIFT                                         0x000010
+#define NV_XAL_EP_BAR0_WINDOW_BASE                                                   21:0
+#define NV_XAL_EP_BAR0_WINDOW                                                  0x0010fd40
+
+#endif // __gh100_pri_nv_xal_ep_h__
+
diff --git a/drivers/gpu/drm/nouveau/include/nvif/chan.h b/drivers/gpu/drm/nouveau/include/nvif/chan.h
new file mode 100644
index 000000000000..c329a29068d5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvif/chan.h
@@ -0,0 +1,76 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVIF_CHAN_H__
+#define __NVIF_CHAN_H__
+#include "push.h"
+
+struct nvif_chan {
+	const struct nvif_chan_func {
+		struct {
+			u32 (*read_get)(struct nvif_chan *);
+		} push;
+
+		struct {
+			u32 (*read_get)(struct nvif_chan *);
+			void (*push)(struct nvif_chan *, bool main, u64 addr, u32 size,
+				     bool no_prefetch);
+			void (*kick)(struct nvif_chan *);
+			int (*post)(struct nvif_chan *, u32 gpptr, u32 pbptr);
+			u32 post_size;
+		} gpfifo;
+
+		struct {
+			int (*release)(struct nvif_chan *, u64 addr, u32 data);
+		} sem;
+	} *func;
+
+	struct {
+		struct nvif_map map;
+	} userd;
+
+	struct {
+		struct nvif_map map;
+		u32 cur;
+		u32 max;
+		int free;
+	} gpfifo;
+
+	struct {
+		struct nvif_map map;
+		u64 addr;
+	} sema;
+
+	struct nvif_push push;
+
+	struct nvif_user *usermode;
+	u32 doorbell_token;
+};
+
+int nvif_chan_dma_wait(struct nvif_chan *, u32 push_nr);
+
+void nvif_chan_gpfifo_ctor(const struct nvif_chan_func *, void *userd, void *gpfifo, u32 gpfifo_size,
+			   void *push, u64 push_addr, u32 push_size, struct nvif_chan *);
+int nvif_chan_gpfifo_wait(struct nvif_chan *, u32 gpfifo_nr, u32 push_nr);
+void nvif_chan_gpfifo_push(struct nvif_chan *, u64 addr, u32 size, bool no_prefetch);
+int nvif_chan_gpfifo_post(struct nvif_chan *);
+
+void nvif_chan506f_gpfifo_push(struct nvif_chan *, bool main, u64 addr, u32 size, bool no_prefetch);
+void nvif_chan506f_gpfifo_kick(struct nvif_chan *);
+
+int nvif_chan906f_ctor_(const struct nvif_chan_func *, void *userd, void *gpfifo, u32 gpfifo_size,
+			void *push, u64 push_addr, u32 push_size, void *sema, u64 sema_addr,
+			struct nvif_chan *);
+u32 nvif_chan906f_read_get(struct nvif_chan *);
+u32 nvif_chan906f_gpfifo_read_get(struct nvif_chan *);
+int nvif_chan906f_gpfifo_post(struct nvif_chan *, u32 gpptr, u32 pbptr);
+
+int nvif_chan506f_ctor(struct nvif_chan *, void *userd, void *gpfifo, u32 gpfifo_size,
+		       void *push, u64 push_addr, u32 push_size);
+int nvif_chan906f_ctor(struct nvif_chan *, void *userd, void *gpfifo, u32 gpfifo_size,
+		       void *push, u64 push_addr, u32 push_size, void *sema, u64 sema_addr);
+int nvif_chanc36f_ctor(struct nvif_chan *, void *userd, void *gpfifo, u32 gpfifo_size,
+		       void *push, u64 push_addr, u32 push_size, void *sema, u64 sema_addr,
+		       struct nvif_user *usermode, u32 doorbell_token);
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvif/cl0080.h b/drivers/gpu/drm/nouveau/include/nvif/cl0080.h
index ea937fa7bc55..ea8267e0d8da 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/cl0080.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/cl0080.h
@@ -29,6 +29,8 @@ struct nv_device_info_v0 {
 #define NV_DEVICE_INFO_V0_TURING                                           0x0c
 #define NV_DEVICE_INFO_V0_AMPERE                                           0x0d
 #define NV_DEVICE_INFO_V0_ADA                                              0x0e
+#define NV_DEVICE_INFO_V0_HOPPER                                           0x0f
+#define NV_DEVICE_INFO_V0_BLACKWELL                                        0x10
 	__u8  family;
 	__u8  pad06[2];
 	__u64 ram_size;
diff --git a/drivers/gpu/drm/nouveau/include/nvif/class.h b/drivers/gpu/drm/nouveau/include/nvif/class.h
index 824e052dcc25..ff6823cb2cd8 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/class.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/class.h
@@ -57,12 +57,15 @@
 
 #define KEPLER_INLINE_TO_MEMORY_A                                    0x0000a040
 #define KEPLER_INLINE_TO_MEMORY_B                                    0x0000a140
+#define BLACKWELL_INLINE_TO_MEMORY_A                                 0x0000cd40
 
 #define NV04_DISP                                     /* cl0046.h */ 0x00000046
 
 #define VOLTA_USERMODE_A                                             0x0000c361
 #define TURING_USERMODE_A                                            0x0000c461
 #define AMPERE_USERMODE_A                                            0x0000c561
+#define HOPPER_USERMODE_A                                            0x0000c661
+#define BLACKWELL_USERMODE_A                                         0x0000c761
 
 #define MAXWELL_FAULT_BUFFER_A                        /* clb069.h */ 0x0000b069
 #define VOLTA_FAULT_BUFFER_A                          /* clb069.h */ 0x0000c369
@@ -85,6 +88,9 @@
 #define TURING_CHANNEL_GPFIFO_A                       /* if0020.h */ 0x0000c46f
 #define AMPERE_CHANNEL_GPFIFO_A                       /* if0020.h */ 0x0000c56f
 #define AMPERE_CHANNEL_GPFIFO_B                       /* if0020.h */ 0x0000c76f
+#define HOPPER_CHANNEL_GPFIFO_A                                      0x0000c86f
+#define BLACKWELL_CHANNEL_GPFIFO_A                                   0x0000c96f
+#define BLACKWELL_CHANNEL_GPFIFO_B                                   0x0000ca6f
 
 #define NV50_DISP                                     /* if0010.h */ 0x00005070
 #define G82_DISP                                      /* if0010.h */ 0x00008270
@@ -102,8 +108,10 @@
 #define TU102_DISP                                    /* if0010.h */ 0x0000c570
 #define GA102_DISP                                    /* if0010.h */ 0x0000c670
 #define AD102_DISP                                    /* if0010.h */ 0x0000c770
+#define GB202_DISP                                                   0x0000ca70
 
 #define GV100_DISP_CAPS                                              0x0000c373
+#define GB202_DISP_CAPS                                              0x0000ca73
 
 #define NV31_MPEG                                                    0x00003174
 #define G82_MPEG                                                     0x00008274
@@ -118,6 +126,7 @@
 #define GV100_DISP_CURSOR                             /* if0014.h */ 0x0000c37a
 #define TU102_DISP_CURSOR                             /* if0014.h */ 0x0000c57a
 #define GA102_DISP_CURSOR                             /* if0014.h */ 0x0000c67a
+#define GB202_DISP_CURSOR                                            0x0000ca7a
 
 #define NV50_DISP_OVERLAY                             /* if0014.h */ 0x0000507b
 #define G82_DISP_OVERLAY                              /* if0014.h */ 0x0000827b
@@ -128,6 +137,7 @@
 #define GV100_DISP_WINDOW_IMM_CHANNEL_DMA             /* if0014.h */ 0x0000c37b
 #define TU102_DISP_WINDOW_IMM_CHANNEL_DMA             /* if0014.h */ 0x0000c57b
 #define GA102_DISP_WINDOW_IMM_CHANNEL_DMA             /* if0014.h */ 0x0000c67b
+#define GB202_DISP_WINDOW_IMM_CHANNEL_DMA                            0x0000ca7b
 
 #define NV50_DISP_BASE_CHANNEL_DMA                    /* if0014.h */ 0x0000507c
 #define G82_DISP_BASE_CHANNEL_DMA                     /* if0014.h */ 0x0000827c
@@ -153,6 +163,7 @@
 #define TU102_DISP_CORE_CHANNEL_DMA                   /* if0014.h */ 0x0000c57d
 #define GA102_DISP_CORE_CHANNEL_DMA                   /* if0014.h */ 0x0000c67d
 #define AD102_DISP_CORE_CHANNEL_DMA                   /* if0014.h */ 0x0000c77d
+#define GB202_DISP_CORE_CHANNEL_DMA                                  0x0000ca7d
 
 #define NV50_DISP_OVERLAY_CHANNEL_DMA                 /* if0014.h */ 0x0000507e
 #define G82_DISP_OVERLAY_CHANNEL_DMA                  /* if0014.h */ 0x0000827e
@@ -164,6 +175,7 @@
 #define GV100_DISP_WINDOW_CHANNEL_DMA                 /* if0014.h */ 0x0000c37e
 #define TU102_DISP_WINDOW_CHANNEL_DMA                 /* if0014.h */ 0x0000c57e
 #define GA102_DISP_WINDOW_CHANNEL_DMA                 /* if0014.h */ 0x0000c67e
+#define GB202_DISP_WINDOW_CHANNEL_DMA                                0x0000ca7e
 
 #define NV50_TESLA                                                   0x00005097
 #define G82_TESLA                                                    0x00008297
@@ -189,16 +201,25 @@
 
 #define TURING_A                                      /* cl9097.h */ 0x0000c597
 
+#define AMPERE_A                                                     0x0000c697
 #define AMPERE_B                                      /* cl9097.h */ 0x0000c797
 
 #define ADA_A                                         /* cl9097.h */ 0x0000c997
 
+#define HOPPER_A                                                     0x0000cb97
+
+#define BLACKWELL_A                                                  0x0000cd97
+#define BLACKWELL_B                                                  0x0000ce97
+
 #define NV74_BSP                                                     0x000074b0
 
+#define NVB8B0_VIDEO_DECODER                                         0x0000b8b0
 #define NVC4B0_VIDEO_DECODER                                         0x0000c4b0
 #define NVC6B0_VIDEO_DECODER                                         0x0000c6b0
 #define NVC7B0_VIDEO_DECODER                                         0x0000c7b0
 #define NVC9B0_VIDEO_DECODER                                         0x0000c9b0
+#define NVCDB0_VIDEO_DECODER                                         0x0000cdb0
+#define NVCFB0_VIDEO_DECODER                                         0x0000cfb0
 
 #define GT212_MSVLD                                                  0x000085b1
 #define IGT21A_MSVLD                                                 0x000086b1
@@ -227,10 +248,14 @@
 #define TURING_DMA_COPY_A                                            0x0000c5b5
 #define AMPERE_DMA_COPY_A                                            0x0000c6b5
 #define AMPERE_DMA_COPY_B                                            0x0000c7b5
+#define HOPPER_DMA_COPY_A                                            0x0000c8b5
+#define BLACKWELL_DMA_COPY_A                                         0x0000c9b5
+#define BLACKWELL_DMA_COPY_B                                         0x0000cab5
 
 #define NVC4B7_VIDEO_ENCODER                                         0x0000c4b7
 #define NVC7B7_VIDEO_ENCODER                                         0x0000c7b7
 #define NVC9B7_VIDEO_ENCODER                                         0x0000c9b7
+#define NVCFB7_VIDEO_ENCODER                                         0x0000cfb7
 
 #define FERMI_DECOMPRESS                                             0x000090b8
 
@@ -246,15 +271,25 @@
 #define PASCAL_COMPUTE_B                                             0x0000c1c0
 #define VOLTA_COMPUTE_A                                              0x0000c3c0
 #define TURING_COMPUTE_A                                             0x0000c5c0
+#define AMPERE_COMPUTE_A                                             0x0000c6c0
 #define AMPERE_COMPUTE_B                                             0x0000c7c0
 #define ADA_COMPUTE_A                                                0x0000c9c0
+#define HOPPER_COMPUTE_A                                             0x0000cbc0
+#define BLACKWELL_COMPUTE_A                                          0x0000cdc0
+#define BLACKWELL_COMPUTE_B                                          0x0000cec0
 
 #define NV74_CIPHER                                                  0x000074c1
 
+#define NVB8D1_VIDEO_NVJPG                                           0x0000b8d1
 #define NVC4D1_VIDEO_NVJPG                                           0x0000c4d1
 #define NVC9D1_VIDEO_NVJPG                                           0x0000c9d1
+#define NVCDD1_VIDEO_NVJPG                                           0x0000cdd1
+#define NVCFD1_VIDEO_NVJPG                                           0x0000cfd1
 
+#define NVB8FA_VIDEO_OFA                                             0x0000b8fa
 #define NVC6FA_VIDEO_OFA                                             0x0000c6fa
 #define NVC7FA_VIDEO_OFA                                             0x0000c7fa
 #define NVC9FA_VIDEO_OFA                                             0x0000c9fa
+#define NVCDFA_VIDEO_OFA                                             0x0000cdfa
+#define NVCFFA_VIDEO_OFA                                             0x0000cffa
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvif/object.h b/drivers/gpu/drm/nouveau/include/nvif/object.h
index 8d205b6af46a..1b32dc701f61 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/object.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/object.h
@@ -16,7 +16,7 @@ struct nvif_object {
 	u32 handle;
 	s32 oclass;
 	void *priv; /*XXX: hack */
-	struct {
+	struct nvif_map {
 		void __iomem *ptr;
 		u64 size;
 	} map;
diff --git a/drivers/gpu/drm/nouveau/include/nvif/push.h b/drivers/gpu/drm/nouveau/include/nvif/push.h
index 6d3a8a3d2087..a493fababe3c 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/push.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/push.h
@@ -31,6 +31,12 @@ struct nvif_push {
 	void (*kick)(struct nvif_push *push);
 
 	struct nvif_mem mem;
+	u64 addr;
+
+	struct {
+		u32 get;
+		u32 max;
+	} hw;
 
 	u32 *bgn;
 	u32 *cur;
@@ -41,7 +47,7 @@ struct nvif_push {
 static inline __must_check int
 PUSH_WAIT(struct nvif_push *push, u32 size)
 {
-	if (push->cur + size >= push->end) {
+	if (push->cur + size > push->end) {
 		int ret = push->wait(push, size);
 		if (ret)
 			return ret;
@@ -55,7 +61,11 @@ PUSH_WAIT(struct nvif_push *push, u32 size)
 static inline int
 PUSH_KICK(struct nvif_push *push)
 {
-	push->kick(push);
+	if (push->cur != push->bgn) {
+		push->kick(push);
+		push->bgn = push->cur;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/include/nvif/push906f.h b/drivers/gpu/drm/nouveau/include/nvif/push906f.h
index cc2866bc8b0a..79df71de98d2 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/push906f.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/push906f.h
@@ -7,6 +7,7 @@
 #ifndef PUSH906F_SUBC
 // Host methods
 #define PUSH906F_SUBC_NV906F	0
+#define PUSH906F_SUBC_NVC36F	0
 
 // Twod
 #define PUSH906F_SUBC_NV902D	3
diff --git a/drivers/gpu/drm/nouveau/include/nvif/pushc97b.h b/drivers/gpu/drm/nouveau/include/nvif/pushc97b.h
new file mode 100644
index 000000000000..c8d6b6319134
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvif/pushc97b.h
@@ -0,0 +1,18 @@
+#ifndef __NVIF_PUSHC97B_H__
+#define __NVIF_PUSHC97B_H__
+#include <nvif/push.h>
+
+#include <nvhw/class/clc97b.h>
+
+#define PUSH_HDR(p,m,c) do {                                                    \
+        PUSH_ASSERT(!((m) & ~DRF_SMASK(NVC97B_DMA_METHOD_OFFSET)), "mthd");     \
+        PUSH_ASSERT(!((c) & ~DRF_MASK(NVC97B_DMA_METHOD_COUNT)), "size");       \
+        PUSH_DATA__((p), NVDEF(NVC97B, DMA, OPCODE, METHOD) |                   \
+			 NVVAL(NVC97B, DMA, METHOD_COUNT, (c)) |                \
+			 NVVAL(NVC97B, DMA, METHOD_OFFSET, (m) >> 2),           \
+		    " mthd 0x%04x size %d - %s", (u32)(m), (u32)(c), __func__); \
+} while(0)
+
+#define PUSH_MTHD_HDR(p,s,m,c) PUSH_HDR(p,m,c)
+#define PUSH_MTHD_INC 4:4
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
index 46afb877a296..99579e7b9376 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
@@ -46,7 +46,10 @@ struct nvkm_device {
 		GV100    = 0x140,
 		TU100    = 0x160,
 		GA100    = 0x170,
+		GH100    = 0x180,
 		AD100    = 0x190,
+		GB10x    = 0x1a0,
+		GB20x    = 0x1b0,
 	} card_type;
 	u32 chipset;
 	u8  chiprev;
@@ -77,6 +80,13 @@ struct nvkm_device {
 struct nvkm_subdev *nvkm_device_subdev(struct nvkm_device *, int type, int inst);
 struct nvkm_engine *nvkm_device_engine(struct nvkm_device *, int type, int inst);
 
+enum nvkm_bar_id {
+	NVKM_BAR_INVALID = 0,
+	NVKM_BAR0_PRI,
+	NVKM_BAR1_FB,
+	NVKM_BAR2_INST,
+};
+
 struct nvkm_device_func {
 	struct nvkm_device_pci *(*pci)(struct nvkm_device *);
 	struct nvkm_device_tegra *(*tegra)(struct nvkm_device *);
@@ -85,8 +95,8 @@ struct nvkm_device_func {
 	int (*init)(struct nvkm_device *);
 	void (*fini)(struct nvkm_device *, bool suspend);
 	int (*irq)(struct nvkm_device *);
-	resource_size_t (*resource_addr)(struct nvkm_device *, unsigned bar);
-	resource_size_t (*resource_size)(struct nvkm_device *, unsigned bar);
+	resource_size_t (*resource_addr)(struct nvkm_device *, enum nvkm_bar_id);
+	resource_size_t (*resource_size)(struct nvkm_device *, enum nvkm_bar_id);
 	bool cpu_coherent;
 };
 
@@ -124,6 +134,9 @@ struct nvkm_device *nvkm_device_find(u64 name);
 	_temp;                                                                 \
 })
 
+#define NVKM_RD32_(p,o,dr) nvkm_rd32((p), (o) + (dr))
+#define NVKM_RD32(p,A...) DRF_RV(NVKM_RD32_, (p), 0, ##A)
+
 void nvkm_device_del(struct nvkm_device **);
 
 struct nvkm_device_oclass {
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/layout.h b/drivers/gpu/drm/nouveau/include/nvkm/core/layout.h
index 9d2a1abf64f9..d92ffd17b729 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/layout.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/layout.h
@@ -1,4 +1,5 @@
 /* SPDX-License-Identifier: MIT */
+NVKM_LAYOUT_ONCE(NVKM_SUBDEV_FSP     , struct nvkm_fsp     ,      fsp)
 NVKM_LAYOUT_ONCE(NVKM_SUBDEV_GSP     , struct nvkm_gsp     ,      gsp)
 NVKM_LAYOUT_ONCE(NVKM_SUBDEV_TOP     , struct nvkm_top     ,      top)
 NVKM_LAYOUT_ONCE(NVKM_SUBDEV_VFN     , struct nvkm_vfn     ,      vfn)
@@ -29,7 +30,7 @@ NVKM_LAYOUT_INST(NVKM_SUBDEV_IOCTRL  , struct nvkm_subdev  ,   ioctrl, 3)
 NVKM_LAYOUT_ONCE(NVKM_SUBDEV_FLA     , struct nvkm_subdev  ,      fla)
 
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_BSP     , struct nvkm_engine  ,      bsp)
-NVKM_LAYOUT_INST(NVKM_ENGINE_CE      , struct nvkm_engine  ,       ce, 10)
+NVKM_LAYOUT_INST(NVKM_ENGINE_CE      , struct nvkm_engine  ,       ce, 20)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_CIPHER  , struct nvkm_engine  ,   cipher)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_DISP    , struct nvkm_disp    ,     disp)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_DMAOBJ  , struct nvkm_dma     ,      dma)
@@ -43,9 +44,9 @@ NVKM_LAYOUT_ONCE(NVKM_ENGINE_MSPDEC  , struct nvkm_engine  ,   mspdec)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_MSPPP   , struct nvkm_engine  ,    msppp)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_MSVLD   , struct nvkm_engine  ,    msvld)
 NVKM_LAYOUT_INST(NVKM_ENGINE_NVDEC   , struct nvkm_nvdec   ,    nvdec, 8)
-NVKM_LAYOUT_INST(NVKM_ENGINE_NVENC   , struct nvkm_nvenc   ,    nvenc, 3)
+NVKM_LAYOUT_INST(NVKM_ENGINE_NVENC   , struct nvkm_nvenc   ,    nvenc, 4)
 NVKM_LAYOUT_INST(NVKM_ENGINE_NVJPG   , struct nvkm_engine  ,    nvjpg, 8)
-NVKM_LAYOUT_ONCE(NVKM_ENGINE_OFA     , struct nvkm_engine  ,      ofa)
+NVKM_LAYOUT_INST(NVKM_ENGINE_OFA     , struct nvkm_engine  ,      ofa, 2)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_SEC     , struct nvkm_engine  ,      sec)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_SEC2    , struct nvkm_sec2    ,     sec2)
 NVKM_LAYOUT_ONCE(NVKM_ENGINE_SW      , struct nvkm_sw      ,       sw)
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
index 3e8db8280e2a..7903d7470d19 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
@@ -87,5 +87,4 @@ int gp102_disp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct
 int gv100_disp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_disp **);
 int tu102_disp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_disp **);
 int ga102_disp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_disp **);
-int ad102_disp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_disp **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
index be508f65b280..96c16cfccf16 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
@@ -78,9 +78,6 @@ struct nvkm_fifo {
 	struct {
 		struct nvkm_memory *mem;
 		struct nvkm_vma *bar1;
-
-		struct mutex mutex;
-		struct list_head list;
 	} userd;
 
 	struct {
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
index 8145796ffc61..a2333cfe6955 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
@@ -55,5 +55,4 @@ int gp10b_gr_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct n
 int gv100_gr_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_gr **);
 int tu102_gr_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_gr **);
 int ga102_gr_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_gr **);
-int ad102_gr_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_gr **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
index 8d2e170883e1..ca83caa55157 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
@@ -13,7 +13,5 @@ struct nvkm_nvdec {
 
 int gm107_nvdec_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvdec **);
 int tu102_nvdec_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvdec **);
-int ga100_nvdec_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvdec **);
 int ga102_nvdec_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvdec **);
-int ad102_nvdec_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvdec **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h
index 018c58fc32ba..1f6eef13f872 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h
@@ -13,6 +13,4 @@ struct nvkm_nvenc {
 
 int gm107_nvenc_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvenc **);
 int tu102_nvenc_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvenc **);
-int ga102_nvenc_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvenc **);
-int ad102_nvenc_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_nvenc **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvjpg.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvjpg.h
deleted file mode 100644
index 80b7933a789e..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvjpg.h
+++ /dev/null
@@ -1,8 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-#ifndef __NVKM_NVJPG_H__
-#define __NVKM_NVJPG_H__
-#include <core/engine.h>
-
-int ga100_nvjpg_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_engine **);
-int ad102_nvjpg_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_engine **);
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/ofa.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/ofa.h
deleted file mode 100644
index e72e2115333b..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/ofa.h
+++ /dev/null
@@ -1,9 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-#ifndef __NVKM_OFA_H__
-#define __NVKM_OFA_H__
-#include <core/engine.h>
-
-int ga100_ofa_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_engine **);
-int ga102_ofa_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_engine **);
-int ad102_ofa_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_engine **);
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h
index 5b798a1a313d..e0d777a933e1 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h
@@ -102,6 +102,9 @@ int gv100_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct n
 int tu102_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fb **);
 int ga100_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fb **);
 int ga102_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fb **);
+int gh100_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fb **);
+int gb100_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fb **);
+int gb202_fb_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fb **);
 
 #include <subdev/bios.h>
 #include <subdev/bios/ramcfg.h>
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fsp.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fsp.h
new file mode 100644
index 000000000000..8a3dbb1cbb46
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fsp.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVKM_FSP_H__
+#define __NVKM_FSP_H__
+#include <core/subdev.h>
+#include <core/falcon.h>
+
+struct nvkm_fsp {
+	const struct nvkm_fsp_func *func;
+	struct nvkm_subdev subdev;
+
+	struct nvkm_falcon falcon;
+};
+
+bool nvkm_fsp_verify_gsp_fmc(struct nvkm_fsp *, u32 hash_size, u32 pkey_size, u32 sig_size);
+int nvkm_fsp_boot_gsp_fmc(struct nvkm_fsp *, u64 args_addr, u32 rsvd_size, bool resume,
+			  u64 img_addr, const u8 *hash, const u8 *pkey, const u8 *sig);
+
+int gh100_fsp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fsp **);
+int gb100_fsp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fsp **);
+int gb202_fsp_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fsp **);
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
index 746e126c3ecf..226c7ec56b8e 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h
@@ -17,6 +17,9 @@ struct nvkm_gsp_mem {
 	dma_addr_t addr;
 };
 
+int nvkm_gsp_mem_ctor(struct nvkm_gsp *, size_t size, struct nvkm_gsp_mem *);
+void nvkm_gsp_mem_dtor(struct nvkm_gsp_mem *);
+
 struct nvkm_gsp_radix3 {
 	struct nvkm_gsp_mem lvl0;
 	struct nvkm_gsp_mem lvl1;
@@ -31,6 +34,29 @@ typedef int (*nvkm_gsp_msg_ntfy_func)(void *priv, u32 fn, void *repv, u32 repc);
 struct nvkm_gsp_event;
 typedef void (*nvkm_gsp_event_func)(struct nvkm_gsp_event *, void *repv, u32 repc);
 
+/**
+ * DOC: GSP message handling policy
+ *
+ * When sending a GSP RPC command, there can be multiple cases of handling
+ * the GSP RPC messages, which are the reply of GSP RPC commands, according
+ * to the requirement of the callers and the nature of the GSP RPC commands.
+ *
+ * NVKM_GSP_RPC_REPLY_NOWAIT - If specified, immediately return to the
+ * caller after the GSP RPC command is issued.
+ *
+ * NVKM_GSP_RPC_REPLY_RECV - If specified, wait and receive the entire GSP
+ * RPC message after the GSP RPC command is issued.
+ *
+ * NVKM_GSP_RPC_REPLY_POLL - If specified, wait for the specific reply and
+ * discard the reply before returning to the caller.
+ *
+ */
+enum nvkm_gsp_rpc_reply_policy {
+	NVKM_GSP_RPC_REPLY_NOWAIT = 0,
+	NVKM_GSP_RPC_REPLY_RECV,
+	NVKM_GSP_RPC_REPLY_POLL,
+};
+
 struct nvkm_gsp {
 	const struct nvkm_gsp_func *func;
 	struct nvkm_subdev subdev;
@@ -42,6 +68,9 @@ struct nvkm_gsp {
 			const struct firmware *load;
 			const struct firmware *unload;
 		} booter;
+
+		const struct firmware *fmc;
+
 		const struct firmware *bl;
 		const struct firmware *rm;
 	} fws;
@@ -89,6 +118,15 @@ struct nvkm_gsp {
 
 	struct {
 		struct nvkm_gsp_mem fw;
+		u8 *hash;
+		u8 *pkey;
+		u8 *sig;
+
+		struct nvkm_gsp_mem args;
+	} fmc;
+
+	struct {
+		struct nvkm_gsp_mem fw;
 		u32 code_offset;
 		u32 data_offset;
 		u32 manifest_offset;
@@ -107,6 +145,7 @@ struct nvkm_gsp {
 		struct sg_table sgt;
 		struct nvkm_gsp_radix3 radix3;
 		struct nvkm_gsp_mem meta;
+		struct sg_table fbsr;
 	} sr;
 
 	struct {
@@ -186,31 +225,7 @@ struct nvkm_gsp {
 		u8 tpcs;
 	} gr;
 
-	const struct nvkm_gsp_rm {
-		void *(*rpc_get)(struct nvkm_gsp *, u32 fn, u32 argc);
-		void *(*rpc_push)(struct nvkm_gsp *, void *argv, bool wait, u32 repc);
-		void (*rpc_done)(struct nvkm_gsp *gsp, void *repv);
-
-		void *(*rm_ctrl_get)(struct nvkm_gsp_object *, u32 cmd, u32 argc);
-		int (*rm_ctrl_push)(struct nvkm_gsp_object *, void **argv, u32 repc);
-		void (*rm_ctrl_done)(struct nvkm_gsp_object *, void *repv);
-
-		void *(*rm_alloc_get)(struct nvkm_gsp_object *, u32 oclass, u32 argc);
-		void *(*rm_alloc_push)(struct nvkm_gsp_object *, void *argv);
-		void (*rm_alloc_done)(struct nvkm_gsp_object *, void *repv);
-
-		int (*rm_free)(struct nvkm_gsp_object *);
-
-		int (*client_ctor)(struct nvkm_gsp *, struct nvkm_gsp_client *);
-		void (*client_dtor)(struct nvkm_gsp_client *);
-
-		int (*device_ctor)(struct nvkm_gsp_client *, struct nvkm_gsp_device *);
-		void (*device_dtor)(struct nvkm_gsp_device *);
-
-		int (*event_ctor)(struct nvkm_gsp_device *, u32 handle, u32 id,
-				  nvkm_gsp_event_func, struct nvkm_gsp_event *);
-		void (*event_dtor)(struct nvkm_gsp_event *);
-	} *rm;
+	struct nvkm_rm *rm;
 
 	struct {
 		struct mutex mutex;
@@ -248,16 +263,19 @@ nvkm_gsp_rm(struct nvkm_gsp *gsp)
 	return gsp && (gsp->fws.rm || gsp->fw.img);
 }
 
+#include <rm/rm.h>
+
 static inline void *
 nvkm_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 argc)
 {
-	return gsp->rm->rpc_get(gsp, fn, argc);
+	return gsp->rm->api->rpc->get(gsp, fn, argc);
 }
 
 static inline void *
-nvkm_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, u32 repc)
+nvkm_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv,
+		  enum nvkm_gsp_rpc_reply_policy policy, u32 repc)
 {
-	return gsp->rm->rpc_push(gsp, argv, wait, repc);
+	return gsp->rm->api->rpc->push(gsp, argv, policy, repc);
 }
 
 static inline void *
@@ -268,13 +286,14 @@ nvkm_gsp_rpc_rd(struct nvkm_gsp *gsp, u32 fn, u32 argc)
 	if (IS_ERR_OR_NULL(argv))
 		return argv;
 
-	return nvkm_gsp_rpc_push(gsp, argv, true, argc);
+	return nvkm_gsp_rpc_push(gsp, argv, NVKM_GSP_RPC_REPLY_RECV, argc);
 }
 
 static inline int
-nvkm_gsp_rpc_wr(struct nvkm_gsp *gsp, void *argv, bool wait)
+nvkm_gsp_rpc_wr(struct nvkm_gsp *gsp, void *argv,
+		enum nvkm_gsp_rpc_reply_policy policy)
 {
-	void *repv = nvkm_gsp_rpc_push(gsp, argv, wait, 0);
+	void *repv = nvkm_gsp_rpc_push(gsp, argv, policy, 0);
 
 	if (IS_ERR(repv))
 		return PTR_ERR(repv);
@@ -285,19 +304,19 @@ nvkm_gsp_rpc_wr(struct nvkm_gsp *gsp, void *argv, bool wait)
 static inline void
 nvkm_gsp_rpc_done(struct nvkm_gsp *gsp, void *repv)
 {
-	gsp->rm->rpc_done(gsp, repv);
+	gsp->rm->api->rpc->done(gsp, repv);
 }
 
 static inline void *
 nvkm_gsp_rm_ctrl_get(struct nvkm_gsp_object *object, u32 cmd, u32 argc)
 {
-	return object->client->gsp->rm->rm_ctrl_get(object, cmd, argc);
+	return object->client->gsp->rm->api->ctrl->get(object, cmd, argc);
 }
 
 static inline int
 nvkm_gsp_rm_ctrl_push(struct nvkm_gsp_object *object, void *argv, u32 repc)
 {
-	return object->client->gsp->rm->rm_ctrl_push(object, argv, repc);
+	return object->client->gsp->rm->api->ctrl->push(object, argv, repc);
 }
 
 static inline void *
@@ -328,7 +347,7 @@ nvkm_gsp_rm_ctrl_wr(struct nvkm_gsp_object *object, void *argv)
 static inline void
 nvkm_gsp_rm_ctrl_done(struct nvkm_gsp_object *object, void *repv)
 {
-	object->client->gsp->rm->rm_ctrl_done(object, repv);
+	object->client->gsp->rm->api->ctrl->done(object, repv);
 }
 
 static inline void *
@@ -343,7 +362,7 @@ nvkm_gsp_rm_alloc_get(struct nvkm_gsp_object *parent, u32 handle, u32 oclass, u3
 	object->parent = parent;
 	object->handle = handle;
 
-	argv = gsp->rm->rm_alloc_get(object, oclass, argc);
+	argv = gsp->rm->api->alloc->get(object, oclass, argc);
 	if (IS_ERR_OR_NULL(argv)) {
 		object->client = NULL;
 		return argv;
@@ -355,7 +374,7 @@ nvkm_gsp_rm_alloc_get(struct nvkm_gsp_object *parent, u32 handle, u32 oclass, u3
 static inline void *
 nvkm_gsp_rm_alloc_push(struct nvkm_gsp_object *object, void *argv)
 {
-	void *repv = object->client->gsp->rm->rm_alloc_push(object, argv);
+	void *repv = object->client->gsp->rm->api->alloc->push(object, argv);
 
 	if (IS_ERR(repv))
 		object->client = NULL;
@@ -377,7 +396,7 @@ nvkm_gsp_rm_alloc_wr(struct nvkm_gsp_object *object, void *argv)
 static inline void
 nvkm_gsp_rm_alloc_done(struct nvkm_gsp_object *object, void *repv)
 {
-	object->client->gsp->rm->rm_alloc_done(object, repv);
+	object->client->gsp->rm->api->alloc->done(object, repv);
 }
 
 static inline int
@@ -395,39 +414,29 @@ nvkm_gsp_rm_alloc(struct nvkm_gsp_object *parent, u32 handle, u32 oclass, u32 ar
 static inline int
 nvkm_gsp_rm_free(struct nvkm_gsp_object *object)
 {
-	if (object->client)
-		return object->client->gsp->rm->rm_free(object);
+	if (object->client) {
+		int ret = object->client->gsp->rm->api->alloc->free(object);
+		object->client = NULL;
+		return ret;
+	}
 
 	return 0;
 }
 
-static inline int
-nvkm_gsp_client_ctor(struct nvkm_gsp *gsp, struct nvkm_gsp_client *client)
-{
-	if (WARN_ON(!gsp->rm))
-		return -ENOSYS;
-
-	return gsp->rm->client_ctor(gsp, client);
-}
-
-static inline void
-nvkm_gsp_client_dtor(struct nvkm_gsp_client *client)
-{
-	if (client->gsp)
-		client->gsp->rm->client_dtor(client);
-}
+int nvkm_gsp_client_ctor(struct nvkm_gsp *, struct nvkm_gsp_client *);
+void nvkm_gsp_client_dtor(struct nvkm_gsp_client *);
 
 static inline int
 nvkm_gsp_device_ctor(struct nvkm_gsp_client *client, struct nvkm_gsp_device *device)
 {
-	return client->gsp->rm->device_ctor(client, device);
+	return client->gsp->rm->api->device->ctor(client, device);
 }
 
 static inline void
 nvkm_gsp_device_dtor(struct nvkm_gsp_device *device)
 {
 	if (device->object.client)
-		device->object.client->gsp->rm->device_dtor(device);
+		device->object.client->gsp->rm->api->device->dtor(device);
 }
 
 static inline int
@@ -459,7 +468,9 @@ static inline int
 nvkm_gsp_device_event_ctor(struct nvkm_gsp_device *device, u32 handle, u32 id,
 			   nvkm_gsp_event_func func, struct nvkm_gsp_event *event)
 {
-	return device->object.client->gsp->rm->event_ctor(device, handle, id, func, event);
+	struct nvkm_rm *rm = device->object.client->gsp->rm;
+
+	return rm->api->device->event.ctor(device, handle, id, func, event);
 }
 
 static inline void
@@ -468,7 +479,7 @@ nvkm_gsp_event_dtor(struct nvkm_gsp_event *event)
 	struct nvkm_gsp_device *device = event->device;
 
 	if (device)
-		device->object.client->gsp->rm->event_dtor(event);
+		device->object.client->gsp->rm->api->device->event.dtor(event);
 }
 
 int nvkm_gsp_intr_stall(struct nvkm_gsp *, enum nvkm_subdev_type, int);
@@ -479,5 +490,8 @@ int tu102_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_
 int tu116_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
 int ga100_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
 int ga102_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
+int gh100_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
 int ad102_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
+int gb100_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
+int gb202_gsp_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_gsp **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h
index e10cbd9203ec..db835cf7b8ac 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/instmem.h
@@ -24,11 +24,6 @@ struct nvkm_instmem {
 	struct nvkm_ramht  *ramht;
 	struct nvkm_memory *ramro;
 	struct nvkm_memory *ramfc;
-
-	struct {
-		struct sg_table fbsr;
-		bool fbsr_valid;
-	} rm;
 };
 
 u32 nvkm_instmem_rd32(struct nvkm_instmem *, u32 addr);
@@ -41,4 +36,5 @@ int nv04_instmem_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nv
 int nv40_instmem_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_instmem **);
 int nv50_instmem_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_instmem **);
 int gk20a_instmem_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_instmem **);
+int gh100_instmem_new(struct nvkm_device *, enum nvkm_subdev_type, int, struct nvkm_instmem **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h
index 935b1cacd528..abcb0dbcde70 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h
@@ -8,7 +8,7 @@ struct nvkm_vma {
 	struct list_head head;
 	struct rb_node tree;
 	u64 addr;
-	u64 size:50;
+	u64 size;
 	bool mapref:1; /* PTs (de)referenced on (un)map (vs pre-allocated). */
 	bool sparse:1; /* Unmapped PDEs/PTEs will not trigger MMU faults. */
 #define NVKM_VMA_PAGE_NONE 7
@@ -73,6 +73,7 @@ struct nvkm_vmm {
 		struct nvkm_gsp_object object;
 
 		struct nvkm_vma *rsvd;
+		bool external;
 	} rm;
 };
 
@@ -165,4 +166,5 @@ int gp100_mmu_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct
 int gp10b_mmu_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_mmu **);
 int gv100_mmu_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_mmu **);
 int tu102_mmu_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_mmu **);
+int gh100_mmu_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_mmu **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
index 3c103101d5fc..112b674ed9c8 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
@@ -50,6 +50,7 @@ int gf100_pci_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct
 int gf106_pci_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_pci **);
 int gk104_pci_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_pci **);
 int gp100_pci_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_pci **);
+int gh100_pci_new(struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_pci **);
 
 /* pcie functions */
 int nvkm_pcie_set_link(struct nvkm_pci *, enum nvkm_pcie_speed, u8 width);
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0000.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0000.h
deleted file mode 100644
index 7a3fc023072d..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0000.h
+++ /dev/null
@@ -1,38 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl0000_h__
-#define __src_common_sdk_nvidia_inc_class_cl0000_h__
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvlimits.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2001-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV01_ROOT        (0x0U) /* finn: Evaluated from "NV0000_ALLOC_PARAMETERS_MESSAGE_ID" */
-
-typedef struct NV0000_ALLOC_PARAMETERS {
-    NvHandle hClient; /* CORERM-2934: hClient must remain the first member until all allocations use these params */
-    NvU32    processID;
-    char     processName[NV_PROC_NAME_MAX_LENGTH];
-} NV0000_ALLOC_PARAMETERS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0005.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0005.h
deleted file mode 100644
index e4de36d63666..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0005.h
+++ /dev/null
@@ -1,38 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl0005_h__
-#define __src_common_sdk_nvidia_inc_class_cl0005_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2001-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NV0005_ALLOC_PARAMETERS {
-    NvHandle hParentClient;
-    NvHandle hSrcResource;
-
-    NvV32    hClass;
-    NvV32    notifyIndex;
-    NV_DECLARE_ALIGNED(NvP64 data, 8);
-} NV0005_ALLOC_PARAMETERS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0080.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0080.h
deleted file mode 100644
index 8868118e47d6..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0080.h
+++ /dev/null
@@ -1,43 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl0080_h__
-#define __src_common_sdk_nvidia_inc_class_cl0080_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2001-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV01_DEVICE_0      (0x80U) /* finn: Evaluated from "NV0080_ALLOC_PARAMETERS_MESSAGE_ID" */
-
-typedef struct NV0080_ALLOC_PARAMETERS {
-    NvU32    deviceId;
-    NvHandle hClientShare;
-    NvHandle hTargetClient;
-    NvHandle hTargetDevice;
-    NvV32    flags;
-    NV_DECLARE_ALIGNED(NvU64 vaSpaceSize, 8);
-    NV_DECLARE_ALIGNED(NvU64 vaStartInternal, 8);
-    NV_DECLARE_ALIGNED(NvU64 vaLimitInternal, 8);
-    NvV32    vaMode;
-} NV0080_ALLOC_PARAMETERS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080.h
deleted file mode 100644
index 9040ea5608a0..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080.h
+++ /dev/null
@@ -1,35 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl2080_h__
-#define __src_common_sdk_nvidia_inc_class_cl2080_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2002-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV20_SUBDEVICE_0      (0x2080U) /* finn: Evaluated from "NV2080_ALLOC_PARAMETERS_MESSAGE_ID" */
-
-typedef struct NV2080_ALLOC_PARAMETERS {
-    NvU32 subDeviceId;
-} NV2080_ALLOC_PARAMETERS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080_notification.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080_notification.h
deleted file mode 100644
index ba659d6477d3..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080_notification.h
+++ /dev/null
@@ -1,62 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl2080_notification_h__
-#define __src_common_sdk_nvidia_inc_class_cl2080_notification_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV2080_NOTIFIERS_HOTPLUG                                   (1)
-
-#define NV2080_NOTIFIERS_DP_IRQ                                    (7)
-
-#define NV2080_ENGINE_TYPE_GRAPHICS                   (0x00000001)
-#define NV2080_ENGINE_TYPE_GR0                        NV2080_ENGINE_TYPE_GRAPHICS
-
-#define NV2080_ENGINE_TYPE_COPY0                      (0x00000009)
-
-#define NV2080_ENGINE_TYPE_BSP                        (0x00000013)
-#define NV2080_ENGINE_TYPE_NVDEC0                     NV2080_ENGINE_TYPE_BSP
-
-#define NV2080_ENGINE_TYPE_MSENC                      (0x0000001b)
-#define NV2080_ENGINE_TYPE_NVENC0                      NV2080_ENGINE_TYPE_MSENC  /* Mutually exclusive alias */
-
-#define NV2080_ENGINE_TYPE_SW                         (0x00000022)
-
-#define NV2080_ENGINE_TYPE_SEC2                       (0x00000026)
-
-#define NV2080_ENGINE_TYPE_NVJPG                      (0x0000002b)
-#define NV2080_ENGINE_TYPE_NVJPEG0                     NV2080_ENGINE_TYPE_NVJPG
-
-#define NV2080_ENGINE_TYPE_OFA                        (0x00000033)
-
-typedef struct {
-    NvU32 plugDisplayMask;
-    NvU32 unplugDisplayMask;
-} Nv2080HotplugNotification;
-
-typedef struct Nv2080DpIrqNotificationRec {
-    NvU32 displayId;
-} Nv2080DpIrqNotification;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl84a0.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl84a0.h
deleted file mode 100644
index 9eb780a1ac72..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl84a0.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl84a0_h__
-#define __src_common_sdk_nvidia_inc_class_cl84a0_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2001-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV01_MEMORY_LIST_SYSTEM (0x00000081)
-
-#define NV01_MEMORY_LIST_FBMEM  (0x00000082)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl90f1.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl90f1.h
deleted file mode 100644
index f1d21776e395..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/cl90f1.h
+++ /dev/null
@@ -1,31 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_cl90f1_h__
-#define __src_common_sdk_nvidia_inc_class_cl90f1_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2011 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define FERMI_VASPACE_A                                     (0x000090f1)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/clc0b5sw.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/clc0b5sw.h
deleted file mode 100644
index b8f32576cfaa..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/class/clc0b5sw.h
+++ /dev/null
@@ -1,34 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_class_clc0b5sw_h__
-#define __src_common_sdk_nvidia_inc_class_clc0b5sw_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2014-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NVC0B5_ALLOCATION_PARAMETERS {
-    NvU32 version;
-    NvU32 engineType;
-} NVC0B5_ALLOCATION_PARAMETERS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073common.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073common.h
deleted file mode 100644
index 58b3ba7badf1..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073common.h
+++ /dev/null
@@ -1,39 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073common_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073common_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NV0073_CTRL_CMD_DSC_CAP_PARAMS {
-    NvBool bDscSupported;
-    NvU32  encoderColorFormatMask;
-    NvU32  lineBufferSizeKB;
-    NvU32  rateBufferSizeKB;
-    NvU32  bitsPerPixelPrecision;
-    NvU32  maxNumHztSlices;
-    NvU32  lineBufferBitDepth;
-} NV0073_CTRL_CMD_DSC_CAP_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dfp.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dfp.h
deleted file mode 100644
index 596f2ea8344e..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dfp.h
+++ /dev/null
@@ -1,166 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073dfp_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073dfp_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2005-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV0073_CTRL_CMD_DFP_GET_INFO (0x731140U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_GET_INFO_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_DFP_GET_INFO_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 flags;
-    NvU32 flags2;
-} NV0073_CTRL_DFP_GET_INFO_PARAMS;
-
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL                                       2:0
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL_TMDS                       (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL_LVDS                       (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL_SDI                        (0x00000002U)
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL_DISPLAYPORT                (0x00000003U)
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL_DSI                        (0x00000004U)
-#define NV0073_CTRL_DFP_FLAGS_SIGNAL_WRBK                       (0x00000005U)
-#define NV0073_CTRL_DFP_FLAGS_LANE                                         5:3
-#define NV0073_CTRL_DFP_FLAGS_LANE_NONE                         (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_LANE_SINGLE                       (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_LANE_DUAL                         (0x00000002U)
-#define NV0073_CTRL_DFP_FLAGS_LANE_QUAD                         (0x00000003U)
-#define NV0073_CTRL_DFP_FLAGS_LANE_OCT                          (0x00000004U)
-#define NV0073_CTRL_DFP_FLAGS_LIMIT                                        6:6
-#define NV0073_CTRL_DFP_FLAGS_LIMIT_DISABLE                     (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_LIMIT_60HZ_RR                     (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER                                   7:7
-#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER_NORMAL                 (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER_DISABLE                (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE                                 8:8
-#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE_FALSE                (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE_TRUE                 (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE                        9:9
-#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE_FALSE       (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE_TRUE        (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE                         10:10
-#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE_FALSE          (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE_TRUE           (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE                    11:11
-#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE_FALSE     (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE_TRUE      (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE                    12:12
-#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE_FALSE     (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE_TRUE      (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED                               14:14
-#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED_FALSE                (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED_TRUE                 (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT                       15:15
-#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT_FALSE        (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT_TRUE         (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT                         16:16
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT_NONE           (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT_PREFER_RBR     (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW                                 19:17
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_1_62GBPS               (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_2_70GBPS               (0x00000002U)
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_5_40GBPS               (0x00000003U)
-#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_8_10GBPS               (0x00000004U)
-#define NV0073_CTRL_DFP_FLAGS_LINK                                       21:20
-#define NV0073_CTRL_DFP_FLAGS_LINK_NONE                         (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_LINK_SINGLE                       (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_LINK_DUAL                         (0x00000002U)
-#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID                           22:22
-#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID_FALSE            (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID_TRUE             (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID                              24:23
-#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_NONE            (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_A               (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_B               (0x00000002U)
-#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_GANGED          (0x00000003U)
-#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED                   25:25
-#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED_FALSE    (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED_TRUE     (0x00000001U)
-#define NV0073_CTRL_DFP_FLAGS_DP_PHY_REPEATER_COUNT                      29:26
-#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE                        30:30
-#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE_FALSE         (0x00000000U)
-#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE_TRUE          (0x00000001U)
-
-#define NV0073_CTRL_CMD_DFP_SET_ELD_AUDIO_CAPS                         (0x731144U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_SET_ELD_AUDIO_CAP_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_ELD_BUFFER                      96U
-
-typedef struct NV0073_CTRL_DFP_SET_ELD_AUDIO_CAP_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 numELDSize;
-    NvU8  bufferELD[NV0073_CTRL_DFP_ELD_AUDIO_CAPS_ELD_BUFFER];
-    NvU32 maxFreqSupported;
-    NvU32 ctrl;
-    NvU32 deviceEntry;
-} NV0073_CTRL_DFP_SET_ELD_AUDIO_CAP_PARAMS;
-
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_PD                                     0:0
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_PD_FALSE              (0x00000000U)
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_PD_TRUE               (0x00000001U)
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_ELDV                                   1:1
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_ELDV_FALSE            (0x00000000U)
-#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_ELDV_TRUE             (0x00000001U)
-
-#define NV0073_CTRL_CMD_DFP_SET_AUDIO_ENABLE                (0x731150U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_SET_AUDIO_ENABLE_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_DFP_SET_AUDIO_ENABLE_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  displayId;
-    NvBool enable;
-} NV0073_CTRL_DFP_SET_AUDIO_ENABLE_PARAMS;
-
-typedef NvU32 NV0073_CTRL_DFP_ASSIGN_SOR_LINKCONFIG;
-
-typedef struct NV0073_CTRL_DFP_ASSIGN_SOR_INFO {
-    NvU32 displayMask;
-    NvU32 sorType;
-} NV0073_CTRL_DFP_ASSIGN_SOR_INFO;
-
-#define NV0073_CTRL_CMD_DFP_ASSIGN_SOR           (0x731152U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_ASSIGN_SOR_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_CMD_DFP_ASSIGN_SOR_MAX_SORS  4U
-
-typedef struct NV0073_CTRL_DFP_ASSIGN_SOR_PARAMS {
-    NvU32                                 subDeviceInstance;
-    NvU32                                 displayId;
-    NvU8                                  sorExcludeMask;
-    NvU32                                 slaveDisplayId;
-    NV0073_CTRL_DFP_ASSIGN_SOR_LINKCONFIG forceSublinkConfig;
-    NvBool                                bIs2Head1Or;
-    NvU32                                 sorAssignList[NV0073_CTRL_CMD_DFP_ASSIGN_SOR_MAX_SORS];
-    NV0073_CTRL_DFP_ASSIGN_SOR_INFO       sorAssignListWithTag[NV0073_CTRL_CMD_DFP_ASSIGN_SOR_MAX_SORS];
-    NvU8                                  reservedSorMask;
-    NvU32                                 flags;
-} NV0073_CTRL_DFP_ASSIGN_SOR_PARAMS;
-
-#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_AUDIO                                      0:0
-#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_AUDIO_OPTIMAL                    (0x00000001U)
-#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_AUDIO_DEFAULT                    (0x00000000U)
-#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_ACTIVE_SOR_NOT_AUDIO_CAPABLE               1:1
-#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_ACTIVE_SOR_NOT_AUDIO_CAPABLE_NO  (0x00000000U)
-#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_ACTIVE_SOR_NOT_AUDIO_CAPABLE_YES (0x00000001U)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dp.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dp.h
deleted file mode 100644
index bae4b1997736..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dp.h
+++ /dev/null
@@ -1,335 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073dp_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073dp_h__
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073common.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2005-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV0073_CTRL_CMD_DP_AUXCH_CTRL      (0x731341U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_AUXCH_CTRL_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_DP_AUXCH_MAX_DATA_SIZE 16U
-
-typedef struct NV0073_CTRL_DP_AUXCH_CTRL_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  displayId;
-    NvBool bAddrOnly;
-    NvU32  cmd;
-    NvU32  addr;
-    NvU8   data[NV0073_CTRL_DP_AUXCH_MAX_DATA_SIZE];
-    NvU32  size;
-    NvU32  replyType;
-    NvU32  retryTimeMs;
-} NV0073_CTRL_DP_AUXCH_CTRL_PARAMS;
-
-#define NV0073_CTRL_DP_AUXCH_CMD_TYPE                          3:3
-#define NV0073_CTRL_DP_AUXCH_CMD_TYPE_I2C               (0x00000000U)
-#define NV0073_CTRL_DP_AUXCH_CMD_TYPE_AUX               (0x00000001U)
-#define NV0073_CTRL_DP_AUXCH_CMD_I2C_MOT                       2:2
-#define NV0073_CTRL_DP_AUXCH_CMD_I2C_MOT_FALSE          (0x00000000U)
-#define NV0073_CTRL_DP_AUXCH_CMD_I2C_MOT_TRUE           (0x00000001U)
-#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE                      1:0
-#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE_WRITE         (0x00000000U)
-#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE_READ          (0x00000001U)
-#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE_WRITE_STATUS  (0x00000002U)
-
-#define NV0073_CTRL_CMD_DP_CTRL                     (0x731343U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_CTRL_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_DP_CTRL_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 cmd;
-    NvU32 data;
-    NvU32 err;
-    NvU32 retryTimeMs;
-    NvU32 eightLaneDpcdBaseAddr;
-} NV0073_CTRL_DP_CTRL_PARAMS;
-
-#define NV0073_CTRL_DP_CMD_SET_LANE_COUNT                           0:0
-#define NV0073_CTRL_DP_CMD_SET_LANE_COUNT_FALSE                         (0x00000000U)
-#define NV0073_CTRL_DP_CMD_SET_LANE_COUNT_TRUE                          (0x00000001U)
-#define NV0073_CTRL_DP_CMD_SET_LINK_BW                              1:1
-#define NV0073_CTRL_DP_CMD_SET_LINK_BW_FALSE                            (0x00000000U)
-#define NV0073_CTRL_DP_CMD_SET_LINK_BW_TRUE                             (0x00000001U)
-#define NV0073_CTRL_DP_CMD_DISABLE_DOWNSPREAD                       2:2
-#define NV0073_CTRL_DP_CMD_DISABLE_DOWNSPREAD_FALSE                     (0x00000000U)
-#define NV0073_CTRL_DP_CMD_DISABLE_DOWNSPREAD_TRUE                      (0x00000001U)
-#define NV0073_CTRL_DP_CMD_UNUSED                                   3:3
-#define NV0073_CTRL_DP_CMD_SET_FORMAT_MODE                          4:4
-#define NV0073_CTRL_DP_CMD_SET_FORMAT_MODE_SINGLE_STREAM                (0x00000000U)
-#define NV0073_CTRL_DP_CMD_SET_FORMAT_MODE_MULTI_STREAM                 (0x00000001U)
-#define NV0073_CTRL_DP_CMD_FAST_LINK_TRAINING                       5:5
-#define NV0073_CTRL_DP_CMD_FAST_LINK_TRAINING_NO                        (0x00000000U)
-#define NV0073_CTRL_DP_CMD_FAST_LINK_TRAINING_YES                       (0x00000001U)
-#define NV0073_CTRL_DP_CMD_NO_LINK_TRAINING                         6:6
-#define NV0073_CTRL_DP_CMD_NO_LINK_TRAINING_NO                          (0x00000000U)
-#define NV0073_CTRL_DP_CMD_NO_LINK_TRAINING_YES                         (0x00000001U)
-#define NV0073_CTRL_DP_CMD_SET_ENHANCED_FRAMING                     7:7
-#define NV0073_CTRL_DP_CMD_SET_ENHANCED_FRAMING_FALSE                   (0x00000000U)
-#define NV0073_CTRL_DP_CMD_SET_ENHANCED_FRAMING_TRUE                    (0x00000001U)
-#define NV0073_CTRL_DP_CMD_USE_DOWNSPREAD_SETTING                   8:8
-#define NV0073_CTRL_DP_CMD_USE_DOWNSPREAD_SETTING_DEFAULT               (0x00000000U)
-#define NV0073_CTRL_DP_CMD_USE_DOWNSPREAD_SETTING_FORCE                 (0x00000001U)
-#define NV0073_CTRL_DP_CMD_SKIP_HW_PROGRAMMING                      9:9
-#define NV0073_CTRL_DP_CMD_SKIP_HW_PROGRAMMING_NO                       (0x00000000U)
-#define NV0073_CTRL_DP_CMD_SKIP_HW_PROGRAMMING_YES                      (0x00000001U)
-#define NV0073_CTRL_DP_CMD_POST_LT_ADJ_REQ_GRANTED                10:10
-#define NV0073_CTRL_DP_CMD_POST_LT_ADJ_REQ_GRANTED_NO                   (0x00000000U)
-#define NV0073_CTRL_DP_CMD_POST_LT_ADJ_REQ_GRANTED_YES                  (0x00000001U)
-#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING                     12:11
-#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING_NO                        (0x00000000U)
-#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING_DONOT_TOGGLE_TRANSMISSION (0x00000001U)
-#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING_TOGGLE_TRANSMISSION_ON    (0x00000002U)
-#define NV0073_CTRL_DP_CMD_TRAIN_PHY_REPEATER                     13:13
-#define NV0073_CTRL_DP_CMD_TRAIN_PHY_REPEATER_NO                        (0x00000000U)
-#define NV0073_CTRL_DP_CMD_TRAIN_PHY_REPEATER_YES                       (0x00000001U)
-#define NV0073_CTRL_DP_CMD_FALLBACK_CONFIG                        14:14
-#define NV0073_CTRL_DP_CMD_FALLBACK_CONFIG_FALSE                        (0x00000000U)
-#define NV0073_CTRL_DP_CMD_FALLBACK_CONFIG_TRUE                         (0x00000001U)
-#define NV0073_CTRL_DP_CMD_ENABLE_FEC                             15:15
-#define NV0073_CTRL_DP_CMD_ENABLE_FEC_FALSE                             (0x00000000U)
-#define NV0073_CTRL_DP_CMD_ENABLE_FEC_TRUE                              (0x00000001U)
-
-#define NV0073_CTRL_DP_CMD_BANDWIDTH_TEST                         29:29
-#define NV0073_CTRL_DP_CMD_BANDWIDTH_TEST_NO                            (0x00000000U)
-#define NV0073_CTRL_DP_CMD_BANDWIDTH_TEST_YES                           (0x00000001U)
-#define NV0073_CTRL_DP_CMD_LINK_CONFIG_CHECK_DISABLE              30:30
-#define NV0073_CTRL_DP_CMD_LINK_CONFIG_CHECK_DISABLE_FALSE              (0x00000000U)
-#define NV0073_CTRL_DP_CMD_LINK_CONFIG_CHECK_DISABLE_TRUE               (0x00000001U)
-#define NV0073_CTRL_DP_CMD_DISABLE_LINK_CONFIG                    31:31
-#define NV0073_CTRL_DP_CMD_DISABLE_LINK_CONFIG_FALSE                    (0x00000000U)
-#define NV0073_CTRL_DP_CMD_DISABLE_LINK_CONFIG_TRUE                     (0x00000001U)
-
-#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT                          4:0
-#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_0                            (0x00000000U)
-#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_1                            (0x00000001U)
-#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_2                            (0x00000002U)
-#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_4                            (0x00000004U)
-#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_8                            (0x00000008U)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW                            15:8
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_1_62GBPS                        (0x00000006U)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_2_16GBPS                        (0x00000008U)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_2_43GBPS                        (0x00000009U)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_2_70GBPS                        (0x0000000AU)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_3_24GBPS                        (0x0000000CU)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_4_32GBPS                        (0x00000010U)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_5_40GBPS                        (0x00000014U)
-#define NV0073_CTRL_DP_DATA_SET_LINK_BW_8_10GBPS                        (0x0000001EU)
-#define NV0073_CTRL_DP_DATA_SET_ENHANCED_FRAMING                  18:18
-#define NV0073_CTRL_DP_DATA_SET_ENHANCED_FRAMING_NO                     (0x00000000U)
-#define NV0073_CTRL_DP_DATA_SET_ENHANCED_FRAMING_YES                    (0x00000001U)
-#define NV0073_CTRL_DP_DATA_TARGET                                22:19
-#define NV0073_CTRL_DP_DATA_TARGET_SINK                                 (0x00000000U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_0                       (0x00000001U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_1                       (0x00000002U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_2                       (0x00000003U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_3                       (0x00000004U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_4                       (0x00000005U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_5                       (0x00000006U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_6                       (0x00000007U)
-#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_7                       (0x00000008U)
-
-#define NV0073_CTRL_MAX_LANES                                           8U
-
-typedef struct NV0073_CTRL_DP_LANE_DATA_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 numLanes;
-    NvU32 data[NV0073_CTRL_MAX_LANES];
-} NV0073_CTRL_DP_LANE_DATA_PARAMS;
-
-#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS                   1:0
-#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_NONE    (0x00000000U)
-#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_LEVEL1  (0x00000001U)
-#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_LEVEL2  (0x00000002U)
-#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_LEVEL3  (0x00000003U)
-#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT                  3:2
-#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL0 (0x00000000U)
-#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL1 (0x00000001U)
-#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL2 (0x00000002U)
-#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL3 (0x00000003U)
-
-#define NV0073_CTRL_CMD_DP_SET_LANE_DATA (0x731346U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_SET_LANE_DATA_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_CMD_DP_SET_AUDIO_MUTESTREAM      (0x731359U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 mute;
-} NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS;
-
-#define NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID  (0x73135bU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  displayId;
-    NvU32  preferredDisplayId;
-
-    NvBool force;
-    NvBool useBFM;
-
-    NvU32  displayIdAssigned;
-    NvU32  allDisplayMask;
-} NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID_PARAMS;
-
-#define NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID (0x73135cU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-} NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID_PARAMS;
-
-#define NV0073_CTRL_CMD_DP_CONFIG_STREAM                   (0x731362U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  head;
-    NvU32  sorIndex;
-    NvU32  dpLink;
-
-    NvBool bEnableOverride;
-    NvBool bMST;
-    NvU32  singleHeadMultistreamMode;
-    NvU32  hBlankSym;
-    NvU32  vBlankSym;
-    NvU32  colorFormat;
-    NvBool bEnableTwoHeadOneOr;
-
-    struct {
-        NvU32  slotStart;
-        NvU32  slotEnd;
-        NvU32  PBN;
-        NvU32  Timeslice;
-        NvBool sendACT;          // deprecated -Use NV0073_CTRL_CMD_DP_SEND_ACT
-        NvU32  singleHeadMSTPipeline;
-        NvBool bEnableAudioOverRightPanel;
-    } MST;
-
-    struct {
-        NvBool bEnhancedFraming;
-        NvU32  tuSize;
-        NvU32  waterMark;
-        NvU32  actualPclkHz;     // deprecated  -Use MvidWarParams
-        NvU32  linkClkFreqHz;    // deprecated  -Use MvidWarParams
-        NvBool bEnableAudioOverRightPanel;
-        struct {
-            NvU32  activeCnt;
-            NvU32  activeFrac;
-            NvU32  activePolarity;
-            NvBool mvidWarEnabled;
-            struct {
-                NvU32 actualPclkHz;
-                NvU32 linkClkFreqHz;
-            } MvidWarParams;
-        } Legacy;
-    } SST;
-} NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS;
-
-#define NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT                    (0x731365U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT_PARAMS {
-    NvU32 subDeviceInstance;
-} NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT_PARAMS;
-
-#define NV0073_CTRL_CMD_DP_GET_CAPS   (0x731369U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID (0x69U)
-
-typedef struct NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS {
-    NvU32                          subDeviceInstance;
-    NvU32                          sorIndex;
-    NvU32                          maxLinkRate;
-    NvU32                          dpVersionsSupported;
-    NvU32                          UHBRSupported;
-    NvBool                         bIsMultistreamSupported;
-    NvBool                         bIsSCEnabled;
-    NvBool                         bHasIncreasedWatermarkLimits;
-    NvBool                         bIsPC2Disabled;
-    NvBool                         isSingleHeadMSTSupported;
-    NvBool                         bFECSupported;
-    NvBool                         bIsTrainPhyRepeater;
-    NvBool                         bOverrideLinkBw;
-    NV0073_CTRL_CMD_DSC_CAP_PARAMS DSC;
-} NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS;
-
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2                0:0
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2_NO              (0x00000000U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2_YES             (0x00000001U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4                1:1
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4_NO              (0x00000000U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4_YES             (0x00000001U)
-
-#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE                           2:0
-#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_NONE                          (0x00000000U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_1_62                          (0x00000001U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_2_70                          (0x00000002U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_5_40                          (0x00000003U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_8_10                          (0x00000004U)
-
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_RGB                (0x00000001U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_444        (0x00000002U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_NATIVE_422 (0x00000004U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_NATIVE_420 (0x00000008U)
-
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_16           (0x00000001U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_8            (0x00000002U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_4            (0x00000003U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_2            (0x00000004U)
-#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1              (0x00000005U)
-
-#define NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES (0x731377U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES        8U
-
-typedef struct NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS {
-    // In
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU16 linkRateTbl[NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES];
-
-    // Out
-    NvU8  linkBwTbl[NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES];
-    NvU8  linkBwCount;
-} NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS;
-
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE                                   3:0
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_MONITOR_ENABLE_BEGIN     (0x00000000U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_MONITOR_ENABLE_CHALLENGE (0x00000001U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_MONITOR_ENABLE_CHECK     (0x00000002U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_DRIVER_ENABLE_BEGIN      (0x00000003U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_DRIVER_ENABLE_CHALLENGE  (0x00000004U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_DRIVER_ENABLE_CHECK      (0x00000005U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_RESET_MONITOR            (0x00000006U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_INIT_PUBLIC_INFO         (0x00000007U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_GET_PUBLIC_INFO          (0x00000008U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_STATUS_CHECK             (0x00000009U)
-
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_OK                          (0x00000000U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_PENDING                     (0x80000001U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_READ_ERROR                  (0x80000002U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_WRITE_ERROR                 (0x80000003U)
-#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_DEVICE_ERROR                (0x80000004U)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073specific.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073specific.h
deleted file mode 100644
index 954958dcf834..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073specific.h
+++ /dev/null
@@ -1,216 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073specific_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073specific_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 1993-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV0073_CTRL_CMD_SPECIFIC_GET_EDID_V2         (0x730245U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_SPECIFIC_GET_EDID_MAX_EDID_BYTES 2048U
-
-typedef struct NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 bufferSize;
-    NvU32 flags;
-    NvU8  edidBuffer[NV0073_CTRL_SPECIFIC_GET_EDID_MAX_EDID_BYTES];
-} NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_GET_CONNECTOR_DATA   (0x730250U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_MAX_CONNECTORS                    4U
-
-typedef struct NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 flags;
-    NvU32 DDCPartners;
-    NvU32 count;
-    struct {
-        NvU32 index;
-        NvU32 type;
-        NvU32 location;
-    } data[NV0073_CTRL_MAX_CONNECTORS];
-    NvU32 platform;
-} NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_ENABLE (0x730273U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_HDMI_ENABLE_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SPECIFIC_SET_HDMI_ENABLE_PARAMS {
-    NvU8  subDeviceInstance;
-    NvU32 displayId;
-    NvU8  enable;
-} NV0073_CTRL_SPECIFIC_SET_HDMI_ENABLE_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM (0x730275U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM_PARAMS {
-    NvU8  subDeviceInstance;
-    NvU32 displayId;
-    NvU8  mute;
-} NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_GET_ALL_HEAD_MASK (0x730287U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_ALL_HEAD_MASK_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SPECIFIC_GET_ALL_HEAD_MASK_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 headMask;
-} NV0073_CTRL_SPECIFIC_GET_ALL_HEAD_MASK_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_SET_OD_PACKET (0x730288U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_OD_PACKET_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_SET_OD_MAX_PACKET_SIZE     36U
-
-typedef struct NV0073_CTRL_SPECIFIC_SET_OD_PACKET_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  displayId;
-    NvU32  transmitControl;
-    NvU32  packetSize;
-    NvU32  targetHead;
-    NvBool bUsePsrHeadforSdp;
-    NvU8   aPacket[NV0073_CTRL_SET_OD_MAX_PACKET_SIZE];
-} NV0073_CTRL_SPECIFIC_SET_OD_PACKET_PARAMS;
-
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ENABLE                                     0:0
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ENABLE_NO                      (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ENABLE_YES                     (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_OTHER_FRAME                                1:1
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_OTHER_FRAME_DISABLE            (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_OTHER_FRAME_ENABLE             (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SINGLE_FRAME                               2:2
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SINGLE_FRAME_DISABLE           (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SINGLE_FRAME_ENABLE            (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ON_HBLANK                                 3:3
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ON_HBLANK_DISABLE              (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ON_HBLANK_ENABLE               (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_IMMEDIATE                                  4:4
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_IMMEDIATE_DISABLE              (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_IMMEDIATE_ENABLE               (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_VIDEO_FMT                                  5:5
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_VIDEO_FMT_SW_CONTROLLED        (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_VIDEO_FMT_HW_CONTROLLED        (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_STEREO_POLARITY                        6:6
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_STEREO_POLARITY_FALSE      (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_STEREO_POLARITY_TRUE       (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_SELF_REFRESH_SETTING                   7:7
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_SELF_REFRESH_SETTING_FALSE (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_SELF_REFRESH_SETTING_TRUE  (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_GEN_INFOFRAME_MODE                         9:8
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_GEN_INFOFRAME_MODE_INFOFRAME0  (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_GEN_INFOFRAME_MODE_INFOFRAME1  (0x0000001U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_RESERVED_LEGACY_MODE                     31:31
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_RESERVED_LEGACY_MODE_NO        (0x0000000U)
-#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_RESERVED_LEGACY_MODE_YES       (0x0000001U)
-
-#define NV0073_CTRL_CMD_SPECIFIC_OR_GET_INFO (0x73028bU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  displayId;
-    NvU32  index;
-    NvU32  type;
-    NvU32  protocol;
-    NvU32  ditherType;
-    NvU32  ditherAlgo;
-    NvU32  location;
-    NvU32  rootPortId;
-    NvU32  dcbIndex;
-    NV_DECLARE_ALIGNED(NvU64 vbiosAddress, 8);
-    NvBool bIsLitByVbios;
-    NvBool bIsDispDynamic;
-} NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS;
-
-#define NV0073_CTRL_SPECIFIC_OR_TYPE_NONE                   (0x00000000U)
-#define NV0073_CTRL_SPECIFIC_OR_TYPE_DAC                    (0x00000001U)
-#define NV0073_CTRL_SPECIFIC_OR_TYPE_SOR                    (0x00000002U)
-#define NV0073_CTRL_SPECIFIC_OR_TYPE_PIOR                   (0x00000003U)
-
-#define NV0073_CTRL_SPECIFIC_OR_TYPE_DSI                    (0x00000005U)
-
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_DAC_RGB_CRT        (0x00000000U)
-
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_LVDS_CUSTOM    (0x00000000U)
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_SINGLE_TMDS_A  (0x00000001U)
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_SINGLE_TMDS_B  (0x00000002U)
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DUAL_TMDS      (0x00000005U)
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DP_A           (0x00000008U)
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DP_B           (0x00000009U)
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DSI            (0x00000010U)
-
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_DSI                (0x00000011U)
-
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_PIOR_EXT_TMDS_ENC  (0x00000000U)
-
-#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_UNKNOWN            (0xFFFFFFFFU)
-
-#define NV0073_CTRL_CMD_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS (0x730291U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS {
-    NvU32  subDeviceInstance;
-    NvU32  displayId;
-    NvU32  brightness;
-    NvBool bUncalibrated;
-} NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS (0x730292U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS_PARAMS_MESSAGE_ID" */
-
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS (0x730293U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayId;
-    NvU32 caps;
-} NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS;
-
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_GT_340MHZ_CLOCK_SUPPORTED                           0:0
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_GT_340MHZ_CLOCK_SUPPORTED_FALSE       (0x00000000U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_GT_340MHZ_CLOCK_SUPPORTED_TRUE        (0x00000001U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_LTE_340MHZ_SCRAMBLING_SUPPORTED                     1:1
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_LTE_340MHZ_SCRAMBLING_SUPPORTED_FALSE (0x00000000U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_LTE_340MHZ_SCRAMBLING_SUPPORTED_TRUE  (0x00000001U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_SCDC_SUPPORTED                                      2:2
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_SCDC_SUPPORTED_FALSE                  (0x00000000U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_SCDC_SUPPORTED_TRUE                   (0x00000001U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED                              5:3
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_NONE           (0x00000000U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_3LANES_3G      (0x00000001U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_3LANES_6G      (0x00000002U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_6G      (0x00000003U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_8G      (0x00000004U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_10G     (0x00000005U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_12G     (0x00000006U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_12_SUPPORTED                                    6:6
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_12_SUPPORTED_FALSE                (0x00000000U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_12_SUPPORTED_TRUE                 (0x00000001U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED                          9:7
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_NONE       (0x00000000U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_3LANES_3G  (0x00000001U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_3LANES_6G  (0x00000002U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_6G  (0x00000003U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_8G  (0x00000004U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_10G (0x00000005U)
-#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_12G (0x00000006U)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073system.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073system.h
deleted file mode 100644
index d69cef3c01fd..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073system.h
+++ /dev/null
@@ -1,65 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073system_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0073_ctrl0073system_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2005-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV0073_CTRL_CMD_SYSTEM_GET_NUM_HEADS (0x730102U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_NUM_HEADS_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SYSTEM_GET_NUM_HEADS_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 flags;
-    NvU32 numHeads;
-} NV0073_CTRL_SYSTEM_GET_NUM_HEADS_PARAMS;
-
-#define NV0073_CTRL_CMD_SYSTEM_GET_SUPPORTED (0x730120U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 displayMask;
-    NvU32 displayMaskDDC;
-} NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS;
-
-#define NV0073_CTRL_CMD_SYSTEM_GET_CONNECT_STATE (0x730122U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 flags;
-    NvU32 displayMask;
-    NvU32 retryTimeMs;
-} NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS;
-
-#define NV0073_CTRL_CMD_SYSTEM_GET_ACTIVE                (0x730126U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS_MESSAGE_ID" */
-
-typedef struct NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS {
-    NvU32 subDeviceInstance;
-    NvU32 head;
-    NvU32 flags;
-    NvU32 displayId;
-} NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS;
-
-#define NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS             (16U)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gpu.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gpu.h
deleted file mode 100644
index 3db099e62364..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gpu.h
+++ /dev/null
@@ -1,48 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0080_ctrl0080gpu_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0080_ctrl0080gpu_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2004-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS {
-    NvU32  totalVFs;
-    NvU32  firstVfOffset;
-    NvU32  vfFeatureMask;
-    NV_DECLARE_ALIGNED(NvU64 FirstVFBar0Address, 8);
-    NV_DECLARE_ALIGNED(NvU64 FirstVFBar1Address, 8);
-    NV_DECLARE_ALIGNED(NvU64 FirstVFBar2Address, 8);
-    NV_DECLARE_ALIGNED(NvU64 bar0Size, 8);
-    NV_DECLARE_ALIGNED(NvU64 bar1Size, 8);
-    NV_DECLARE_ALIGNED(NvU64 bar2Size, 8);
-    NvBool b64bitBar0;
-    NvBool b64bitBar1;
-    NvBool b64bitBar2;
-    NvBool bSriovEnabled;
-    NvBool bSriovHeavyEnabled;
-    NvBool bEmulateVFBar0TlbInvalidationRegister;
-    NvBool bClientRmAllocatedCtxBuffer;
-} NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gr.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gr.h
deleted file mode 100644
index ed01df925573..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gr.h
+++ /dev/null
@@ -1,31 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0080_ctrl0080gr_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0080_ctrl0080gr_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2004-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV0080_CTRL_GR_CAPS_TBL_SIZE            23
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080bios.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080bios.h
deleted file mode 100644
index b5b7631de99b..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080bios.h
+++ /dev/null
@@ -1,40 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080bios_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080bios_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2005-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS {
-    NvU32 BoardID;
-    char  chipSKU[4];
-    char  chipSKUMod[2];
-    char  project[5];
-    char  projectSKU[5];
-    char  CDP[6];
-    char  projectSKUMod[2];
-    NvU32 businessCycle;
-} NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080ce.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080ce.h
deleted file mode 100644
index fe912d2bd183..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080ce.h
+++ /dev/null
@@ -1,35 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080ce_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080ce_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2014-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NV2080_CTRL_CE_GET_FAULT_METHOD_BUFFER_SIZE_PARAMS {
-    NvU32 size;
-} NV2080_CTRL_CE_GET_FAULT_METHOD_BUFFER_SIZE_PARAMS;
-
-#define NV2080_CTRL_CMD_CE_GET_FAULT_METHOD_BUFFER_SIZE (0x20802a08) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_CE_INTERFACE_ID << 8) | NV2080_CTRL_CE_GET_FAULT_METHOD_BUFFER_SIZE_PARAMS_MESSAGE_ID" */
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080event.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080event.h
deleted file mode 100644
index 87bc4ff92ce1..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080event.h
+++ /dev/null
@@ -1,41 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080event_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080event_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2006-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV2080_CTRL_CMD_EVENT_SET_NOTIFICATION (0x20800301) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_EVENT_INTERFACE_ID << 8) | NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS {
-    NvU32  event;
-    NvU32  action;
-    NvBool bNotifyState;
-    NvU32  info32;
-    NvU16  info16;
-} NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS;
-
-#define NV2080_CTRL_EVENT_SET_NOTIFICATION_ACTION_REPEAT  (0x00000002)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fb.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fb.h
deleted file mode 100644
index 68c81f9f803c..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fb.h
+++ /dev/null
@@ -1,51 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080fb_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080fb_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2006-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MEM_TYPES   17U
-
-typedef NvBool NV2080_CTRL_CMD_FB_GET_FB_REGION_SURFACE_MEM_TYPE_FLAG[NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MEM_TYPES];
-
-typedef struct NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO {
-    NV_DECLARE_ALIGNED(NvU64 base, 8);
-    NV_DECLARE_ALIGNED(NvU64 limit, 8);
-    NV_DECLARE_ALIGNED(NvU64 reserved, 8);
-    NvU32                                                  performance;
-    NvBool                                                 supportCompressed;
-    NvBool                                                 supportISO;
-    NvBool                                                 bProtected;
-    NV2080_CTRL_CMD_FB_GET_FB_REGION_SURFACE_MEM_TYPE_FLAG blackList;
-} NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO;
-
-#define NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MAX_ENTRIES 16U
-
-typedef struct NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS {
-    NvU32 numFBRegions;
-    NV_DECLARE_ALIGNED(NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO fbRegion[NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MAX_ENTRIES], 8);
-} NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fifo.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fifo.h
deleted file mode 100644
index bc0f63699b06..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fifo.h
+++ /dev/null
@@ -1,52 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080fifo_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080fifo_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2006-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV2080_CTRL_CMD_FIFO_GET_DEVICE_INFO_TABLE                 (0x20801112) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_FIFO_INTERFACE_ID << 8) | NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS_MESSAGE_ID" */
-
-#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_MAX_ENTRIES         32
-#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_DATA_TYPES   16
-#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_PBDMA    2
-#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_NAME_LEN 16
-
-typedef struct NV2080_CTRL_FIFO_DEVICE_ENTRY {
-    NvU32 engineData[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_DATA_TYPES];
-    NvU32 pbdmaIds[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_PBDMA];
-    NvU32 pbdmaFaultIds[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_PBDMA];
-    NvU32 numPbdmas;
-    char  engineName[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_NAME_LEN];
-} NV2080_CTRL_FIFO_DEVICE_ENTRY;
-
-typedef struct NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS {
-    NvU32                         baseIndex;
-    NvU32                         numEntries;
-    NvBool                        bMore;
-    // C form: NV2080_CTRL_FIFO_DEVICE_ENTRY entries[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_MAX_ENTRIES];
-    NV2080_CTRL_FIFO_DEVICE_ENTRY entries[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_MAX_ENTRIES];
-} NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h
deleted file mode 100644
index 29d7a1052142..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h
+++ /dev/null
@@ -1,100 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080gpu_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080gpu_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2006-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV2080_GPU_MAX_NAME_STRING_LENGTH                  (0x0000040U)
-
-#define NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_0            (0x00000000U)
-
-#define NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_3            (0x00000003U)
-
-typedef struct NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY {
-    NV_DECLARE_ALIGNED(NvU64 gpuPhysAddr, 8);
-    NV_DECLARE_ALIGNED(NvU64 gpuVirtAddr, 8);
-    NV_DECLARE_ALIGNED(NvU64 size, 8);
-    NvU32 physAttr;
-    NvU16 bufferId;
-    NvU8  bInitialize;
-    NvU8  bNonmapped;
-} NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY;
-
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_MAIN                         0U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PM                           1U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PATCH                        2U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_BUFFER_BUNDLE_CB             3U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PAGEPOOL                     4U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_ATTRIBUTE_CB                 5U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_RTV_CB_GLOBAL                6U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_GFXP_POOL                    7U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_GFXP_CTRL_BLK                8U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_FECS_EVENT                   9U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PRIV_ACCESS_MAP              10U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_UNRESTRICTED_PRIV_ACCESS_MAP 11U
-#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_GLOBAL_PRIV_ACCESS_MAP       12U
-
-#define NV2080_CTRL_GPU_PROMOTE_CONTEXT_MAX_ENTRIES                        16U
-
-#define NV2080_CTRL_CMD_GPU_PROMOTE_CTX                                    (0x2080012bU) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_GPU_INTERFACE_ID << 8) | NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS {
-    NvU32    engineType;
-    NvHandle hClient;
-    NvU32    ChID;
-    NvHandle hChanClient;
-    NvHandle hObject;
-    NvHandle hVirtMemory;
-    NV_DECLARE_ALIGNED(NvU64 virtAddress, 8);
-    NV_DECLARE_ALIGNED(NvU64 size, 8);
-    NvU32    entryCount;
-    // C form: NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY promoteEntry[NV2080_CTRL_GPU_PROMOTE_CONTEXT_MAX_ENTRIES];
-    NV_DECLARE_ALIGNED(NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY promoteEntry[NV2080_CTRL_GPU_PROMOTE_CONTEXT_MAX_ENTRIES], 8);
-} NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS;
-
-typedef struct NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS {
-    NvU32 gpcMask;
-} NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS;
-
-typedef struct NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS {
-    NvU32 gpcId;
-    NvU32 tpcMask;
-} NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS;
-
-typedef struct NV2080_CTRL_GPU_GET_FERMI_ZCULL_INFO_PARAMS {
-    NvU32 gpcId;
-    NvU32 zcullMask;
-} NV2080_CTRL_GPU_GET_FERMI_ZCULL_INFO_PARAMS;
-
-#define NV2080_GPU_MAX_GID_LENGTH             (0x000000100ULL)
-
-typedef struct NV2080_CTRL_GPU_GET_GID_INFO_PARAMS {
-    NvU32 index;
-    NvU32 flags;
-    NvU32 length;
-    NvU8  data[NV2080_GPU_MAX_GID_LENGTH];
-} NV2080_CTRL_GPU_GET_GID_INFO_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gr.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gr.h
deleted file mode 100644
index 59f8895bc5d7..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gr.h
+++ /dev/null
@@ -1,41 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080gr_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080gr_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2006-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef enum NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS {
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_MAIN = 0,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_SPILL = 1,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_PAGEPOOL = 2,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_BETACB = 3,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_RTV = 4,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL = 5,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL_CONTROL = 6,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL_CONTROL_CPU = 7,
-    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_END = 8,
-} NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h
deleted file mode 100644
index e11b2dbe5288..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h
+++ /dev/null
@@ -1,162 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080internal_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl2080_ctrl2080internal_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_GET_STATIC_INFO (0x20800a01) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS {
-    NvU32  feHwSysCap;
-    NvU32  windowPresentMask;
-    NvBool bFbRemapperEnabled;
-    NvU32  numHeads;
-    NvBool bPrimaryVga;
-    NvU32  i2cPort;
-    NvU32  internalDispActiveMask;
-} NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS;
-
-#define NV2080_CTRL_INTERNAL_GR_MAX_ENGINES         8
-
-#define NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT 0x19
-
-typedef struct NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO {
-    NvU32 size;
-    NvU32 alignment;
-} NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO;
-
-typedef struct NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO {
-    NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO engine[NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT];
-} NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO;
-
-typedef struct NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS {
-    NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO engineContextBuffersInfo[NV2080_CTRL_INTERNAL_GR_MAX_ENGINES];
-} NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO (0x20800a32) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_CONSTRUCTED_FALCON_INFO {
-    NvU32 engDesc;
-    NvU32 ctxAttr;
-    NvU32 ctxBufferSize;
-    NvU32 addrSpaceList;
-    NvU32 registerBase;
-} NV2080_CTRL_INTERNAL_CONSTRUCTED_FALCON_INFO;
-#define NV2080_CTRL_CMD_INTERNAL_MAX_CONSTRUCTED_FALCONS     0x40
-
-#define NV2080_CTRL_CMD_INTERNAL_GET_CONSTRUCTED_FALCON_INFO (0x20800a42) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_GET_CONSTRUCTED_FALCON_INFO_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_GET_CONSTRUCTED_FALCON_INFO_PARAMS {
-    NvU32                                        numConstructedFalcons;
-    NV2080_CTRL_INTERNAL_CONSTRUCTED_FALCON_INFO constructedFalconsTable[NV2080_CTRL_CMD_INTERNAL_MAX_CONSTRUCTED_FALCONS];
-} NV2080_CTRL_INTERNAL_GET_CONSTRUCTED_FALCON_INFO_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_WRITE_INST_MEM (0x20800a49) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS {
-    NV_DECLARE_ALIGNED(NvU64 instMemPhysAddr, 8);
-    NV_DECLARE_ALIGNED(NvU64 instMemSize, 8);
-    NvU32 instMemAddrSpace;
-    NvU32 instMemCpuCacheAttr;
-} NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER (0x20800a58) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS {
-    NvU32  addressSpace;
-    NV_DECLARE_ALIGNED(NvU64 physicalAddr, 8);
-    NV_DECLARE_ALIGNED(NvU64 limit, 8);
-    NvU32  cacheSnoop;
-    NvU32  hclass;
-    NvU32  channelInstance;
-    NvBool valid;
-} NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_INTR_GET_KERNEL_TABLE (0x20800a5c) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS_MESSAGE_ID" */
-
-#define NV2080_CTRL_INTERNAL_INTR_MAX_TABLE_SIZE       128
-
-typedef enum NV2080_INTR_CATEGORY {
-    NV2080_INTR_CATEGORY_DEFAULT = 0,
-    NV2080_INTR_CATEGORY_ESCHED_DRIVEN_ENGINE = 1,
-    NV2080_INTR_CATEGORY_ESCHED_DRIVEN_ENGINE_NOTIFICATION = 2,
-    NV2080_INTR_CATEGORY_RUNLIST = 3,
-    NV2080_INTR_CATEGORY_RUNLIST_NOTIFICATION = 4,
-    NV2080_INTR_CATEGORY_UVM_OWNED = 5,
-    NV2080_INTR_CATEGORY_UVM_SHARED = 6,
-    NV2080_INTR_CATEGORY_ENUM_COUNT = 7,
-} NV2080_INTR_CATEGORY;
-
-typedef struct NV2080_INTR_CATEGORY_SUBTREE_MAP {
-    NvU8 subtreeStart;
-    NvU8 subtreeEnd;
-} NV2080_INTR_CATEGORY_SUBTREE_MAP;
-
-typedef struct NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_ENTRY {
-    NvU16 engineIdx;
-    NvU32 pmcIntrMask;
-    NvU32 vectorStall;
-    NvU32 vectorNonStall;
-} NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_ENTRY;
-
-typedef struct NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS {
-    NvU32                                            tableLen;
-    NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_ENTRY table[NV2080_CTRL_INTERNAL_INTR_MAX_TABLE_SIZE];
-    NV2080_INTR_CATEGORY_SUBTREE_MAP                 subtreeMap[NV2080_INTR_CATEGORY_ENUM_COUNT];
-} NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_FBSR_INIT (0x20800ac2) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS {
-    NvU32    fbsrType;
-    NvU32    numRegions;
-    NvHandle hClient;
-    NvHandle hSysMem;
-    NV_DECLARE_ALIGNED(NvU64 gspFbAllocsSysOffset, 8);
-    NvBool   bEnteringGcoffState;
-} NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_FBSR_SEND_REGION_INFO (0x20800ac3) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_FBSR_SEND_REGION_INFO_PARAMS_MESSAGE_ID" */
-
-typedef struct NV2080_CTRL_INTERNAL_FBSR_SEND_REGION_INFO_PARAMS {
-    NvU32    fbsrType;
-    NvHandle hClient;
-    NvHandle hVidMem;
-    NV_DECLARE_ALIGNED(NvU64 vidOffset, 8);
-    NV_DECLARE_ALIGNED(NvU64 sysOffset, 8);
-    NV_DECLARE_ALIGNED(NvU64 size, 8);
-} NV2080_CTRL_INTERNAL_FBSR_SEND_REGION_INFO_PARAMS;
-
-#define NV2080_CTRL_CMD_INTERNAL_INIT_BRIGHTC_STATE_LOAD (0x20800ac6) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_INIT_BRIGHTC_STATE_LOAD_PARAMS_MESSAGE_ID" */
-
-#define NV2080_CTRL_ACPI_DSM_READ_SIZE                   (0x1000) /* finn: Evaluated from "(4 * 1024)" */
-
-typedef struct NV2080_CTRL_INTERNAL_INIT_BRIGHTC_STATE_LOAD_PARAMS {
-    NvU32 status;
-    NvU16 backLightDataSize;
-    NvU8  backLightData[NV2080_CTRL_ACPI_DSM_READ_SIZE];
-} NV2080_CTRL_INTERNAL_INIT_BRIGHTC_STATE_LOAD_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl90f1.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl90f1.h
deleted file mode 100644
index 977e59818533..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl90f1.h
+++ /dev/null
@@ -1,95 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl90f1_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl90f1_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2014-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define GMMU_FMT_MAX_LEVELS  6U
-
-#define NV90F1_CTRL_CMD_VASPACE_COPY_SERVER_RESERVED_PDES (0x90f10106U) /* finn: Evaluated from "(FINN_FERMI_VASPACE_A_VASPACE_INTERFACE_ID << 8) | NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS_MESSAGE_ID" */
-
-typedef struct NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS {
-    /*!
-     * [in] GPU sub-device handle - this API only supports unicast.
-     *      Pass 0 to use subDeviceId instead.
-     */
-    NvHandle hSubDevice;
-
-    /*!
-     * [in] GPU sub-device ID. Ignored if hSubDevice is non-zero.
-     */
-    NvU32    subDeviceId;
-
-    /*!
-     * [in] Page size (VA coverage) of the level to reserve.
-     *      This need not be a leaf (page table) page size - it can be
-     *      the coverage of an arbitrary level (including root page directory).
-     */
-    NV_DECLARE_ALIGNED(NvU64 pageSize, 8);
-
-    /*!
-     * [in] First GPU virtual address of the range to reserve.
-     *      This must be aligned to pageSize.
-     */
-    NV_DECLARE_ALIGNED(NvU64 virtAddrLo, 8);
-
-    /*!
-     * [in] Last GPU virtual address of the range to reserve.
-     *      This (+1) must be aligned to pageSize.
-     */
-    NV_DECLARE_ALIGNED(NvU64 virtAddrHi, 8);
-
-    /*! 
-     * [in] Number of PDE levels to copy.
-     */
-    NvU32    numLevelsToCopy;
-
-   /*!
-     * [in] Per-level information.
-     */
-    struct {
-        /*!
-         * Physical address of this page level instance.
-         */
-        NV_DECLARE_ALIGNED(NvU64 physAddress, 8);
-
-        /*!
-         * Size in bytes allocated for this level instance.
-         */
-        NV_DECLARE_ALIGNED(NvU64 size, 8);
-
-        /*!
-         * Aperture in which this page level instance resides.
-         */
-        NvU32 aperture;
-
-        /*!
-         * Page shift corresponding to the level
-         */
-        NvU8  pageShift;
-    } levels[GMMU_FMT_MAX_LEVELS];
-} NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrla06f/ctrla06fgpfifo.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrla06f/ctrla06fgpfifo.h
deleted file mode 100644
index 684045796232..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrla06f/ctrla06fgpfifo.h
+++ /dev/null
@@ -1,42 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrla06f_ctrla06fgpfifo_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrla06f_ctrla06fgpfifo_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2007-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- * 
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- * 
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- * 
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NVA06F_CTRL_CMD_GPFIFO_SCHEDULE (0xa06f0103) /* finn: Evaluated from "(FINN_KEPLER_CHANNEL_GPFIFO_A_GPFIFO_INTERFACE_ID << 8) | NVA06F_CTRL_GPFIFO_SCHEDULE_PARAMS_MESSAGE_ID" */
-
-typedef struct NVA06F_CTRL_GPFIFO_SCHEDULE_PARAMS {
-    NvBool bEnable;
-    NvBool bSkipSubmit;
-} NVA06F_CTRL_GPFIFO_SCHEDULE_PARAMS;
-
-#define NVA06F_CTRL_CMD_BIND (0xa06f0104) /* finn: Evaluated from "(FINN_KEPLER_CHANNEL_GPFIFO_A_GPFIFO_INTERFACE_ID << 8) | NVA06F_CTRL_BIND_PARAMS_MESSAGE_ID" */
-
-typedef struct NVA06F_CTRL_BIND_PARAMS {
-    NvU32 engineType;
-} NVA06F_CTRL_BIND_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/nvlimits.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/nvlimits.h
deleted file mode 100644
index 5c5a004a8031..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/nvlimits.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_nvlimits_h__
-#define __src_common_sdk_nvidia_inc_nvlimits_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2017 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV_MAX_SUBDEVICES       8
-
-#define NV_PROC_NAME_MAX_LENGTH 100U
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h
deleted file mode 100644
index 51b5591c603e..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h
+++ /dev/null
@@ -1,148 +0,0 @@
-#ifndef __src_common_sdk_nvidia_inc_nvos_h__
-#define __src_common_sdk_nvidia_inc_nvos_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 1993-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NVOS02_FLAGS_PHYSICALITY                                   7:4
-#define NVOS02_FLAGS_PHYSICALITY_CONTIGUOUS                        (0x00000000)
-#define NVOS02_FLAGS_PHYSICALITY_NONCONTIGUOUS                     (0x00000001)
-#define NVOS02_FLAGS_LOCATION                                      11:8
-#define NVOS02_FLAGS_LOCATION_PCI                                  (0x00000000)
-#define NVOS02_FLAGS_LOCATION_AGP                                  (0x00000001)
-#define NVOS02_FLAGS_LOCATION_VIDMEM                               (0x00000002)
-#define NVOS02_FLAGS_COHERENCY                                     15:12
-#define NVOS02_FLAGS_COHERENCY_UNCACHED                            (0x00000000)
-#define NVOS02_FLAGS_COHERENCY_CACHED                              (0x00000001)
-#define NVOS02_FLAGS_COHERENCY_WRITE_COMBINE                       (0x00000002)
-#define NVOS02_FLAGS_COHERENCY_WRITE_THROUGH                       (0x00000003)
-#define NVOS02_FLAGS_COHERENCY_WRITE_PROTECT                       (0x00000004)
-#define NVOS02_FLAGS_COHERENCY_WRITE_BACK                          (0x00000005)
-#define NVOS02_FLAGS_ALLOC                                         17:16
-#define NVOS02_FLAGS_ALLOC_NONE                                    (0x00000001)
-#define NVOS02_FLAGS_GPU_CACHEABLE                                 18:18
-#define NVOS02_FLAGS_GPU_CACHEABLE_NO                              (0x00000000)
-#define NVOS02_FLAGS_GPU_CACHEABLE_YES                             (0x00000001)
-
-#define NVOS02_FLAGS_KERNEL_MAPPING                                19:19
-#define NVOS02_FLAGS_KERNEL_MAPPING_NO_MAP                         (0x00000000)
-#define NVOS02_FLAGS_KERNEL_MAPPING_MAP                            (0x00000001)
-#define NVOS02_FLAGS_ALLOC_NISO_DISPLAY                            20:20
-#define NVOS02_FLAGS_ALLOC_NISO_DISPLAY_NO                         (0x00000000)
-#define NVOS02_FLAGS_ALLOC_NISO_DISPLAY_YES                        (0x00000001)
-
-#define NVOS02_FLAGS_ALLOC_USER_READ_ONLY                          21:21
-#define NVOS02_FLAGS_ALLOC_USER_READ_ONLY_NO                       (0x00000000)
-#define NVOS02_FLAGS_ALLOC_USER_READ_ONLY_YES                      (0x00000001)
-
-#define NVOS02_FLAGS_ALLOC_DEVICE_READ_ONLY                        22:22
-#define NVOS02_FLAGS_ALLOC_DEVICE_READ_ONLY_NO                     (0x00000000)
-#define NVOS02_FLAGS_ALLOC_DEVICE_READ_ONLY_YES                    (0x00000001)
-
-#define NVOS02_FLAGS_PEER_MAP_OVERRIDE                             23:23
-#define NVOS02_FLAGS_PEER_MAP_OVERRIDE_DEFAULT                     (0x00000000)
-#define NVOS02_FLAGS_PEER_MAP_OVERRIDE_REQUIRED                    (0x00000001)
-
-#define NVOS02_FLAGS_ALLOC_TYPE_SYNCPOINT                          24:24
-#define NVOS02_FLAGS_ALLOC_TYPE_SYNCPOINT_APERTURE                 (0x00000001)
-
-#define NVOS02_FLAGS_MEMORY_PROTECTION                             26:25
-#define NVOS02_FLAGS_MEMORY_PROTECTION_DEFAULT                     (0x00000000)
-#define NVOS02_FLAGS_MEMORY_PROTECTION_PROTECTED                   (0x00000001)
-#define NVOS02_FLAGS_MEMORY_PROTECTION_UNPROTECTED                 (0x00000002)
-
-#define NVOS02_FLAGS_MAPPING                                       31:30
-#define NVOS02_FLAGS_MAPPING_DEFAULT                               (0x00000000)
-#define NVOS02_FLAGS_MAPPING_NO_MAP                                (0x00000001)
-#define NVOS02_FLAGS_MAPPING_NEVER_MAP                             (0x00000002)
-
-#define NV01_EVENT_CLIENT_RM                                       (0x04000000)
-
-typedef struct
-{
-    NvV32    channelInstance;            // One of the n channel instances of a given channel type.
-                                         // Note that core channel has only one instance
-                                         // while all others have two (one per head).
-    NvHandle hObjectBuffer;              // ctx dma handle for DMA push buffer
-    NvHandle hObjectNotify;              // ctx dma handle for an area (of type NvNotification defined in sdk/nvidia/inc/nvtypes.h) where RM can write errors/notifications
-    NvU32    offset;                     // Initial offset for put/get, usually zero.
-    NvP64    pControl NV_ALIGN_BYTES(8); // pControl gives virt addr of UDISP GET/PUT regs
-
-    NvU32    flags;
-#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB                1:1
-#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB_YES            0x00000000
-#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB_NO             0x00000001
-
-} NV50VAIO_CHANNELDMA_ALLOCATION_PARAMETERS;
-
-typedef struct
-{
-    NvV32    channelInstance;            // One of the n channel instances of a given channel type.
-                                         // All PIO channels have two instances (one per head).
-    NvHandle hObjectNotify;              // ctx dma handle for an area (of type NvNotification defined in sdk/nvidia/inc/nvtypes.h) where RM can write errors.
-    NvP64    pControl NV_ALIGN_BYTES(8); // pControl gives virt addr of control region for PIO channel
-} NV50VAIO_CHANNELPIO_ALLOCATION_PARAMETERS;
-
-typedef struct
-{
-    NvU32 size;
-    NvU32 prohibitMultipleInstances;
-    NvU32 engineInstance;               // Select NVDEC0 or NVDEC1 or NVDEC2
-} NV_BSP_ALLOCATION_PARAMETERS;
-
-typedef struct
-{
-    NvU32 size;
-    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of MSENC?
-    NvU32 engineInstance;             // Select MSENC/NVENC0 or NVENC1 or NVENC2
-} NV_MSENC_ALLOCATION_PARAMETERS;
-
-typedef struct
-{
-    NvU32 size;
-    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of NVJPG?
-    NvU32 engineInstance;
-} NV_NVJPG_ALLOCATION_PARAMETERS;
-
-typedef struct
-{
-    NvU32 size;
-    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of OFA?
-} NV_OFA_ALLOCATION_PARAMETERS;
-
-typedef struct
-{
-    NvU32   index;
-    NvV32   flags;
-    NvU64   vaSize NV_ALIGN_BYTES(8);
-    NvU64   vaStartInternal NV_ALIGN_BYTES(8);
-    NvU64   vaLimitInternal NV_ALIGN_BYTES(8);
-    NvU32   bigPageSize;
-    NvU64   vaBase NV_ALIGN_BYTES(8);
-} NV_VASPACE_ALLOCATION_PARAMETERS;
-
-#define NV_VASPACE_ALLOCATION_INDEX_GPU_NEW                                 0x00 //<! Create new VASpace, by default
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/shared/msgq/inc/msgq/msgq_priv.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/shared/msgq/inc/msgq/msgq_priv.h
deleted file mode 100644
index 0e32e71e123f..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/shared/msgq/inc/msgq/msgq_priv.h
+++ /dev/null
@@ -1,97 +0,0 @@
-#ifndef __src_common_shared_msgq_inc_msgq_msgq_priv_h__
-#define __src_common_shared_msgq_inc_msgq_msgq_priv_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2018-2019 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-/**
- * msgqTxHeader -- TX queue data structure
- * @version: the version of this structure, must be 0
- * @size: the size of the entire queue, including this header
- * @msgSize: the padded size of queue element, 16 is minimum
- * @msgCount: the number of elements in this queue
- * @writePtr: head index of this queue
- * @flags: 1 = swap the RX pointers
- * @rxHdrOff: offset of readPtr in this structure
- * @entryOff: offset of beginning of queue (msgqRxHeader), relative to
- *          beginning of this structure
- *
- * The command queue is a queue of RPCs that are sent from the driver to the
- * GSP.  The status queue is a queue of messages/responses from GSP-RM to the
- * driver.  Although the driver allocates memory for both queues, the command
- * queue is owned by the driver and the status queue is owned by GSP-RM.  In
- * addition, the headers of the two queues must not share the same 4K page.
- *
- * Each queue is prefixed with this data structure.  The idea is that a queue
- * and its header are written to only by their owner.  That is, only the
- * driver writes to the command queue and command queue header, and only the
- * GSP writes to the status (receive) queue and its header.
- *
- * This is enforced by the concept of "swapping" the RX pointers.  This is
- * why the 'flags' field must be set to 1.  'rxHdrOff' is how the GSP knows
- * where the where the tail pointer of its status queue.
- *
- * When the driver writes a new RPC to the command queue, it updates writePtr.
- * When it reads a new message from the status queue, it updates readPtr.  In
- * this way, the GSP knows when a new command is in the queue (it polls
- * writePtr) and it knows how much free space is in the status queue (it
- * checks readPtr).  The driver never cares about how much free space is in
- * the status queue.
- *
- * As usual, producers write to the head pointer, and consumers read from the
- * tail pointer.  When head == tail, the queue is empty.
- *
- * So to summarize:
- * command.writePtr = head of command queue
- * command.readPtr = tail of status queue
- * status.writePtr = head of status queue
- * status.readPtr = tail of command queue
- */
-typedef struct
-{
-    NvU32 version;   // queue version
-    NvU32 size;      // bytes, page aligned
-    NvU32 msgSize;   // entry size, bytes, must be power-of-2, 16 is minimum
-    NvU32 msgCount;  // number of entries in queue
-    NvU32 writePtr;  // message id of next slot
-    NvU32 flags;     // if set it means "i want to swap RX"
-    NvU32 rxHdrOff;  // Offset of msgqRxHeader from start of backing store.
-    NvU32 entryOff;  // Offset of entries from start of backing store.
-} msgqTxHeader;
-
-/**
- * msgqRxHeader - RX queue data structure
- * @readPtr: tail index of the other queue
- *
- * Although this is a separate struct, it could easily be merged into
- * msgqTxHeader.  msgqTxHeader.rxHdrOff is simply the offset of readPtr
- * from the beginning of msgqTxHeader.
- */
-typedef struct
-{
-    NvU32 readPtr; // message id of last message read
-} msgqRxHeader;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/uproc/os/common/include/libos_init_args.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/uproc/os/common/include/libos_init_args.h
deleted file mode 100644
index 83cf1b2c15a3..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/uproc/os/common/include/libos_init_args.h
+++ /dev/null
@@ -1,52 +0,0 @@
-#ifndef __src_common_uproc_os_common_include_libos_init_args_h__
-#define __src_common_uproc_os_common_include_libos_init_args_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2018-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef NvU64 LibosAddress;
-
-typedef enum {
-    LIBOS_MEMORY_REGION_NONE,
-    LIBOS_MEMORY_REGION_CONTIGUOUS,
-    LIBOS_MEMORY_REGION_RADIX3
-} LibosMemoryRegionKind;
-
-typedef enum {
-    LIBOS_MEMORY_REGION_LOC_NONE,
-    LIBOS_MEMORY_REGION_LOC_SYSMEM,
-    LIBOS_MEMORY_REGION_LOC_FB
-} LibosMemoryRegionLoc;
-
-typedef struct
-{
-    LibosAddress          id8;  // Id tag.
-    LibosAddress          pa;   // Physical address.
-    LibosAddress          size; // Size of memory area.
-    NvU8                  kind; // See LibosMemoryRegionKind above.
-    NvU8                  loc;  // See LibosMemoryRegionLoc above.
-} LibosMemoryRegionInitArgument;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_sr_meta.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_sr_meta.h
deleted file mode 100644
index 73213bdfcbda..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_sr_meta.h
+++ /dev/null
@@ -1,79 +0,0 @@
-#ifndef __src_nvidia_arch_nvalloc_common_inc_gsp_gsp_fw_sr_meta_h__
-#define __src_nvidia_arch_nvalloc_common_inc_gsp_gsp_fw_sr_meta_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2022-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define GSP_FW_SR_META_MAGIC     0x8a3bb9e6c6c39d93ULL
-#define GSP_FW_SR_META_REVISION  2
-
-typedef struct
-{
-    //
-    // Magic
-    // Use for verification by Booter
-    //
-    NvU64 magic;  // = GSP_FW_SR_META_MAGIC;
-
-    //
-    // Revision number
-    // Bumped up when we change this interface so it is not backward compatible.
-    // Bumped up when we revoke GSP-RM ucode
-    //
-    NvU64 revision;  // = GSP_FW_SR_META_MAGIC_REVISION;
-
-    //
-    // ---- Members regarding data in SYSMEM ----------------------------
-    // Consumed by Booter for DMA
-    //
-    NvU64 sysmemAddrOfSuspendResumeData;
-    NvU64 sizeOfSuspendResumeData;
-
-    // ---- Members for crypto ops across S/R ---------------------------
-
-    //
-    // HMAC over the entire GspFwSRMeta structure (including padding)
-    // with the hmac field itself zeroed.
-    //
-    NvU8 hmac[32];
-
-    // Hash over GspFwWprMeta structure
-    NvU8 wprMetaHash[32];
-
-    // Hash over GspFwHeapFreeList structure. All zeros signifies no free list.
-    NvU8 heapFreeListHash[32];
-
-    // Hash over data in WPR2 (skipping over free heap chunks; see Booter for details)
-    NvU8 dataHash[32];
-
-    //
-    // Pad structure to exactly 256 bytes (1 DMA chunk).
-    // Padding initialized to zero.
-    //
-    NvU32 padding[24];
-
-} GspFwSRMeta;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_wpr_meta.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_wpr_meta.h
deleted file mode 100644
index a2e141e4b459..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_wpr_meta.h
+++ /dev/null
@@ -1,170 +0,0 @@
-#ifndef __src_nvidia_arch_nvalloc_common_inc_gsp_gsp_fw_wpr_meta_h__
-#define __src_nvidia_arch_nvalloc_common_inc_gsp_gsp_fw_wpr_meta_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2021-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct
-{
-    // Magic
-    // BL to use for verification (i.e. Booter locked it in WPR2)
-    NvU64 magic; // = 0xdc3aae21371a60b3;
-
-    // Revision number of Booter-BL-Sequencer handoff interface
-    // Bumped up when we change this interface so it is not backward compatible.
-    // Bumped up when we revoke GSP-RM ucode
-    NvU64 revision; // = 1;
-
-    // ---- Members regarding data in SYSMEM ----------------------------
-    // Consumed by Booter for DMA
-
-    NvU64 sysmemAddrOfRadix3Elf;
-    NvU64 sizeOfRadix3Elf;
-
-    NvU64 sysmemAddrOfBootloader;
-    NvU64 sizeOfBootloader;
-
-    // Offsets inside bootloader image needed by Booter
-    NvU64 bootloaderCodeOffset;
-    NvU64 bootloaderDataOffset;
-    NvU64 bootloaderManifestOffset;
-
-    union
-    {
-        // Used only at initial boot
-        struct
-        {
-            NvU64 sysmemAddrOfSignature;
-            NvU64 sizeOfSignature;
-        };
-
-        //
-        // Used at suspend/resume to read GspFwHeapFreeList
-        // Offset relative to GspFwWprMeta FBMEM PA (gspFwWprStart)
-        //
-        struct
-        {
-            NvU32 gspFwHeapFreeListWprOffset;
-            NvU32 unused0;
-            NvU64 unused1;
-        };
-    };
-
-    // ---- Members describing FB layout --------------------------------
-    NvU64 gspFwRsvdStart;
-
-    NvU64 nonWprHeapOffset;
-    NvU64 nonWprHeapSize;
-
-    NvU64 gspFwWprStart;
-
-    // GSP-RM to use to setup heap.
-    NvU64 gspFwHeapOffset;
-    NvU64 gspFwHeapSize;
-
-    // BL to use to find ELF for jump
-    NvU64 gspFwOffset;
-    // Size is sizeOfRadix3Elf above.
-
-    NvU64 bootBinOffset;
-    // Size is sizeOfBootloader above.
-
-    NvU64 frtsOffset;
-    NvU64 frtsSize;
-
-    NvU64 gspFwWprEnd;
-
-    // GSP-RM to use for fbRegionInfo?
-    NvU64 fbSize;
-
-    // ---- Other members -----------------------------------------------
-
-    // GSP-RM to use for fbRegionInfo?
-    NvU64 vgaWorkspaceOffset;
-    NvU64 vgaWorkspaceSize;
-
-    // Boot count.  Used to determine whether to load the firmware image.
-    NvU64 bootCount;
-
-    // TODO: the partitionRpc* fields below do not really belong in this
-    //       structure. The values are patched in by the partition bootstrapper
-    //       when GSP-RM is booted in a partition, and this structure was a
-    //       convenient place for the bootstrapper to access them. These should
-    //       be moved to a different comm. mechanism between the bootstrapper
-    //       and the GSP-RM tasks.
-
-    union
-    {
-	struct
-	{
-	    // Shared partition RPC memory (physical address)
-	    NvU64 partitionRpcAddr;
-
-	    // Offsets relative to partitionRpcAddr
-	    NvU16 partitionRpcRequestOffset;
-	    NvU16 partitionRpcReplyOffset;
-
-	    // Code section and dataSection offset and size.
-	    NvU32 elfCodeOffset;
-	    NvU32 elfDataOffset;
-	    NvU32 elfCodeSize;
-	    NvU32 elfDataSize;
-
-	    // Used during GSP-RM resume to check for revocation
-	    NvU32 lsUcodeVersion;
-	};
-
-        struct
-	{
-	    // Pad for the partitionRpc* fields, plus 4 bytes
-	    NvU32 partitionRpcPadding[4];
-
-            // CrashCat (contiguous) buffer size/location - occupies same bytes as the
-            // elf(Code|Data)(Offset|Size) fields above.
-            // TODO: move to GSP_FMC_INIT_PARAMS
-            NvU64 sysmemAddrOfCrashReportQueue;
-            NvU32 sizeOfCrashReportQueue;
-
-            // Pad for the lsUcodeVersion field
-            NvU32 lsUcodeVersionPadding[1];
-        };
-    };
-
-    // Number of VF partitions allocating sub-heaps from the WPR heap
-    // Used during boot to ensure the heap is adequately sized
-    NvU8 gspFwHeapVfPartitionCount;
-
-    // Pad structure to exactly 256 bytes.  Can replace padding with additional
-    // fields without incrementing revision.  Padding initialized to 0.
-    NvU8 padding[7];
-
-    // BL to use for verification (i.e. Booter says OK to boot)
-    NvU64 verified;  // 0x0 -> unverified, 0xa0a0a0a0a0a0a0a0 -> verified
-} GspFwWprMeta;
-
-#define GSP_FW_WPR_META_REVISION  1
-#define GSP_FW_WPR_META_MAGIC     0xdc3aae21371a60b3ULL
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmRiscvUcode.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmRiscvUcode.h
deleted file mode 100644
index 4eff473e8990..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmRiscvUcode.h
+++ /dev/null
@@ -1,82 +0,0 @@
-#ifndef __src_nvidia_arch_nvalloc_common_inc_rmRiscvUcode_h__
-#define __src_nvidia_arch_nvalloc_common_inc_rmRiscvUcode_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2018-2019 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct {
-    //
-    // Version 1
-    // Version 2
-    // Version 3 = for Partition boot
-    // Version 4 = for eb riscv boot
-    // Version 5 = Support signing entire RISC-V image as "code" in code section for hopper and later.
-    //
-    NvU32  version;                         // structure version
-    NvU32  bootloaderOffset;
-    NvU32  bootloaderSize;
-    NvU32  bootloaderParamOffset;
-    NvU32  bootloaderParamSize;
-    NvU32  riscvElfOffset;
-    NvU32  riscvElfSize;
-    NvU32  appVersion;                      // Changelist number associated with the image
-    //
-    // Manifest contains information about Monitor and it is
-    // input to BR
-    //
-    NvU32  manifestOffset;
-    NvU32  manifestSize;
-    //
-    // Monitor Data offset within RISCV image and size
-    //
-    NvU32  monitorDataOffset;
-    NvU32  monitorDataSize;
-    //
-    // Monitor Code offset withtin RISCV image and size
-    //
-    NvU32  monitorCodeOffset;
-    NvU32  monitorCodeSize;
-    NvU32  bIsMonitorEnabled;
-    //
-    // Swbrom Code offset within RISCV image and size
-    //
-    NvU32  swbromCodeOffset;
-    NvU32  swbromCodeSize;
-    //
-    // Swbrom Data offset within RISCV image and size
-    //
-    NvU32  swbromDataOffset;
-    NvU32  swbromDataSize;
-    //
-    // Total size of FB carveout (image and reserved space).  
-    //
-    NvU32  fbReservedSize;
-    //
-    // Indicates whether the entire RISC-V image is signed as "code" in code section.
-    //
-    NvU32  bSignedAsCode;
-} RM_RISCV_UCODE_DESC;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmgspseq.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmgspseq.h
deleted file mode 100644
index 341ab0dbeaf2..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmgspseq.h
+++ /dev/null
@@ -1,100 +0,0 @@
-#ifndef __src_nvidia_arch_nvalloc_common_inc_rmgspseq_h__
-#define __src_nvidia_arch_nvalloc_common_inc_rmgspseq_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2019-2020 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef enum GSP_SEQ_BUF_OPCODE
-{
-    GSP_SEQ_BUF_OPCODE_REG_WRITE = 0,
-    GSP_SEQ_BUF_OPCODE_REG_MODIFY,
-    GSP_SEQ_BUF_OPCODE_REG_POLL,
-    GSP_SEQ_BUF_OPCODE_DELAY_US,
-    GSP_SEQ_BUF_OPCODE_REG_STORE,
-    GSP_SEQ_BUF_OPCODE_CORE_RESET,
-    GSP_SEQ_BUF_OPCODE_CORE_START,
-    GSP_SEQ_BUF_OPCODE_CORE_WAIT_FOR_HALT,
-    GSP_SEQ_BUF_OPCODE_CORE_RESUME,
-} GSP_SEQ_BUF_OPCODE;
-
-#define GSP_SEQUENCER_PAYLOAD_SIZE_DWORDS(opcode)                       \
-    ((opcode == GSP_SEQ_BUF_OPCODE_REG_WRITE)  ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_WRITE)  / sizeof(NvU32)) : \
-     (opcode == GSP_SEQ_BUF_OPCODE_REG_MODIFY) ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_MODIFY) / sizeof(NvU32)) : \
-     (opcode == GSP_SEQ_BUF_OPCODE_REG_POLL)   ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_POLL)   / sizeof(NvU32)) : \
-     (opcode == GSP_SEQ_BUF_OPCODE_DELAY_US)   ? (sizeof(GSP_SEQ_BUF_PAYLOAD_DELAY_US)   / sizeof(NvU32)) : \
-     (opcode == GSP_SEQ_BUF_OPCODE_REG_STORE)  ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_STORE)  / sizeof(NvU32)) : \
-    /* GSP_SEQ_BUF_OPCODE_CORE_RESET */                                 \
-    /* GSP_SEQ_BUF_OPCODE_CORE_START */                                 \
-    /* GSP_SEQ_BUF_OPCODE_CORE_WAIT_FOR_HALT */                         \
-    /* GSP_SEQ_BUF_OPCODE_CORE_RESUME */                                \
-    0)
-
-typedef struct
-{
-    NvU32 addr;
-    NvU32 val;
-} GSP_SEQ_BUF_PAYLOAD_REG_WRITE;
-
-typedef struct
-{
-    NvU32 addr;
-    NvU32 mask;
-    NvU32 val;
-} GSP_SEQ_BUF_PAYLOAD_REG_MODIFY;
-
-typedef struct
-{
-    NvU32 addr;
-    NvU32 mask;
-    NvU32 val;
-    NvU32 timeout;
-    NvU32 error;
-} GSP_SEQ_BUF_PAYLOAD_REG_POLL;
-
-typedef struct
-{
-    NvU32 val;
-} GSP_SEQ_BUF_PAYLOAD_DELAY_US;
-
-typedef struct
-{
-    NvU32 addr;
-    NvU32 index;
-} GSP_SEQ_BUF_PAYLOAD_REG_STORE;
-
-typedef struct GSP_SEQUENCER_BUFFER_CMD
-{
-    GSP_SEQ_BUF_OPCODE opCode;
-    union
-    {
-        GSP_SEQ_BUF_PAYLOAD_REG_WRITE regWrite;
-        GSP_SEQ_BUF_PAYLOAD_REG_MODIFY regModify;
-        GSP_SEQ_BUF_PAYLOAD_REG_POLL regPoll;
-        GSP_SEQ_BUF_PAYLOAD_DELAY_US delayUs;
-        GSP_SEQ_BUF_PAYLOAD_REG_STORE regStore;
-    } payload;
-} GSP_SEQUENCER_BUFFER_CMD;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_allclasses.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_allclasses.h
deleted file mode 100644
index 3144e9beac61..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_allclasses.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef __src_nvidia_generated_g_allclasses_h__
-#define __src_nvidia_generated_g_allclasses_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2021-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV01_EVENT_KERNEL_CALLBACK_EX            (0x0000007e)
-
-#define NV04_DISPLAY_COMMON                      (0x00000073)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_chipset_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_chipset_nvoc.h
deleted file mode 100644
index 6b8921138c7d..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_chipset_nvoc.h
+++ /dev/null
@@ -1,38 +0,0 @@
-#ifndef __src_nvidia_generated_g_chipset_nvoc_h__
-#define __src_nvidia_generated_g_chipset_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 1993-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct
-{
-    NvU16               deviceID;           // deviceID
-    NvU16               vendorID;           // vendorID
-    NvU16               subdeviceID;        // subsystem deviceID
-    NvU16               subvendorID;        // subsystem vendorID
-    NvU8                revisionID;         // revision ID
-} BUSINFO;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_fbsr_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_fbsr_nvoc.h
deleted file mode 100644
index a5128f00225b..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_fbsr_nvoc.h
+++ /dev/null
@@ -1,31 +0,0 @@
-#ifndef __src_nvidia_generated_g_fbsr_nvoc_h__
-#define __src_nvidia_generated_g_fbsr_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2009-2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define FBSR_TYPE_DMA                                 4   // Copy using DMA. Fastest.
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_gpu_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_gpu_nvoc.h
deleted file mode 100644
index 5641a21cacca..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_gpu_nvoc.h
+++ /dev/null
@@ -1,35 +0,0 @@
-#ifndef __src_nvidia_generated_g_gpu_nvoc_h__
-#define __src_nvidia_generated_g_gpu_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2004-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef enum
-{
-    COMPUTE_BRANDING_TYPE_NONE,
-    COMPUTE_BRANDING_TYPE_TESLA,
-} COMPUTE_BRANDING_TYPE;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_kernel_channel_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_kernel_channel_nvoc.h
deleted file mode 100644
index b5ad55f854dc..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_kernel_channel_nvoc.h
+++ /dev/null
@@ -1,62 +0,0 @@
-#ifndef __src_nvidia_generated_g_kernel_channel_nvoc_h__
-#define __src_nvidia_generated_g_kernel_channel_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef enum {
-    /*!
-     * Initial state as passed in NV_CHANNEL_ALLOC_PARAMS by
-     * kernel CPU-RM clients.
-     */
-    ERROR_NOTIFIER_TYPE_UNKNOWN = 0,
-    /*! @brief Error notifier is explicitly not set.
-     *
-     * The corresponding hErrorContext or hEccErrorContext must be
-     * NV01_NULL_OBJECT.
-     */
-    ERROR_NOTIFIER_TYPE_NONE,
-    /*! @brief Error notifier is a ContextDma */
-    ERROR_NOTIFIER_TYPE_CTXDMA,
-    /*! @brief Error notifier is a NvNotification array in sysmem/vidmem */
-    ERROR_NOTIFIER_TYPE_MEMORY
-} ErrorNotifierType;
-
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE                       1:0
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_USER                  0x0
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_ADMIN                 0x1
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_KERNEL                0x2
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE             3:2
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_UNKNOWN     ERROR_NOTIFIER_TYPE_UNKNOWN
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_NONE        ERROR_NOTIFIER_TYPE_NONE
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_CTXDMA      ERROR_NOTIFIER_TYPE_CTXDMA
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_MEMORY      ERROR_NOTIFIER_TYPE_MEMORY
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE         5:4
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_UNKNOWN ERROR_NOTIFIER_TYPE_UNKNOWN
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_NONE    ERROR_NOTIFIER_TYPE_NONE
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_CTXDMA  ERROR_NOTIFIER_TYPE_CTXDMA
-#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_MEMORY  ERROR_NOTIFIER_TYPE_MEMORY
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_kernel_fifo_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_kernel_fifo_nvoc.h
deleted file mode 100644
index 946954ac5b3d..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_kernel_fifo_nvoc.h
+++ /dev/null
@@ -1,119 +0,0 @@
-#ifndef __src_nvidia_generated_g_kernel_fifo_nvoc_h__
-#define __src_nvidia_generated_g_kernel_fifo_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2021-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef enum
-{
-    /* *************************************************************************
-     * Bug 3820969
-     * THINK BEFORE CHANGING ENUM ORDER HERE.
-     * VGPU-guest uses this same ordering. Because this enum is not versioned,
-     * changing the order here WILL BREAK old-guest-on-newer-host compatibility.
-     * ************************************************************************/
-
-    // *ENG_XYZ, e.g.: ENG_GR, ENG_CE etc.,
-    ENGINE_INFO_TYPE_ENG_DESC = 0,
-
-    // HW engine ID
-    ENGINE_INFO_TYPE_FIFO_TAG,
-
-    // RM_ENGINE_TYPE_*
-    ENGINE_INFO_TYPE_RM_ENGINE_TYPE,
-
-    //
-    // runlist id (meaning varies by GPU)
-    // Valid only for Esched-driven engines
-    //
-    ENGINE_INFO_TYPE_RUNLIST,
-
-    // NV_PFIFO_INTR_MMU_FAULT_ENG_ID_*
-    ENGINE_INFO_TYPE_MMU_FAULT_ID,
-
-    // ROBUST_CHANNEL_*
-    ENGINE_INFO_TYPE_RC_MASK,
-
-    // Reset Bit Position. On Ampere, only valid if not _INVALID
-    ENGINE_INFO_TYPE_RESET,
-
-    // Interrupt Bit Position
-    ENGINE_INFO_TYPE_INTR,
-
-    // log2(MC_ENGINE_*)
-    ENGINE_INFO_TYPE_MC,
-
-    // The DEV_TYPE_ENUM for this engine
-    ENGINE_INFO_TYPE_DEV_TYPE_ENUM,
-
-    // The particular instance of this engine type
-    ENGINE_INFO_TYPE_INSTANCE_ID,
-
-    //
-    // The base address for this engine's NV_RUNLIST. Valid only on Ampere+
-    // Valid only for Esched-driven engines
-    //
-    ENGINE_INFO_TYPE_RUNLIST_PRI_BASE,
-
-    //
-    // If this entry is a host-driven engine.
-    // Update _isEngineInfoTypeValidForOnlyHostDriven when adding any new entry.
-    //
-    ENGINE_INFO_TYPE_IS_HOST_DRIVEN_ENGINE,
-
-    //
-    // The index into the per-engine NV_RUNLIST registers. Valid only on Ampere+
-    // Valid only for Esched-driven engines
-    //
-    ENGINE_INFO_TYPE_RUNLIST_ENGINE_ID,
-
-    //
-    // The base address for this engine's NV_CHRAM registers. Valid only on
-    // Ampere+
-    //
-    // Valid only for Esched-driven engines
-    //
-    ENGINE_INFO_TYPE_CHRAM_PRI_BASE,
-
-    // This entry added to copy data at RMCTRL_EXPORT() call for Kernel RM
-    ENGINE_INFO_TYPE_KERNEL_RM_MAX,
-    // Used for iterating the engine info table by the index passed.
-    ENGINE_INFO_TYPE_INVALID = ENGINE_INFO_TYPE_KERNEL_RM_MAX,
-
-    // Size of FIFO_ENGINE_LIST.engineData
-    ENGINE_INFO_TYPE_ENGINE_DATA_ARRAY_SIZE = ENGINE_INFO_TYPE_INVALID,
-
-    // Input-only parameter for kfifoEngineInfoXlate.
-    ENGINE_INFO_TYPE_PBDMA_ID
-
-    /* *************************************************************************
-     * Bug 3820969
-     * THINK BEFORE CHANGING ENUM ORDER HERE.
-     * VGPU-guest uses this same ordering. Because this enum is not versioned,
-     * changing the order here WILL BREAK old-guest-on-newer-host compatibility.
-     * ************************************************************************/
-} ENGINE_INFO_TYPE;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_mem_desc_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_mem_desc_nvoc.h
deleted file mode 100644
index daabaee41c87..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_mem_desc_nvoc.h
+++ /dev/null
@@ -1,32 +0,0 @@
-#ifndef __src_nvidia_generated_g_mem_desc_nvoc_h__
-#define __src_nvidia_generated_g_mem_desc_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 1993-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define ADDR_SYSMEM     1         // System memory (PCI)
-#define ADDR_FBMEM      2         // Frame buffer memory space
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_os_nvoc.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_os_nvoc.h
deleted file mode 100644
index 10121218f4d3..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_os_nvoc.h
+++ /dev/null
@@ -1,44 +0,0 @@
-#ifndef __src_nvidia_generated_g_os_nvoc_h__
-#define __src_nvidia_generated_g_os_nvoc_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 1993-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct PACKED_REGISTRY_ENTRY
-{
-    NvU32                   nameOffset;
-    NvU8                    type;
-    NvU32                   data;
-    NvU32                   length;
-} PACKED_REGISTRY_ENTRY;
-
-typedef struct PACKED_REGISTRY_TABLE
-{
-    NvU32                   size;
-    NvU32                   numEntries;
-    PACKED_REGISTRY_ENTRY   entries[] __counted_by(numEntries);
-} PACKED_REGISTRY_TABLE;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_rpc-structures.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_rpc-structures.h
deleted file mode 100644
index 8d925e24faea..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_rpc-structures.h
+++ /dev/null
@@ -1,124 +0,0 @@
-#ifndef __src_nvidia_generated_g_rpc_structures_h__
-#define __src_nvidia_generated_g_rpc_structures_h__
-#include <nvrm/535.113.01/nvidia/generated/g_sdk-structures.h>
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/sdk-structures.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2008-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct rpc_alloc_memory_v13_01
-{
-    NvHandle   hClient;
-    NvHandle   hDevice;
-    NvHandle   hMemory;
-    NvU32      hClass;
-    NvU32      flags;
-    NvU32      pteAdjust;
-    NvU32      format;
-    NvU64      length NV_ALIGN_BYTES(8);
-    NvU32      pageCount;
-    struct pte_desc pteDesc;
-} rpc_alloc_memory_v13_01;
-
-typedef struct rpc_free_v03_00
-{
-    NVOS00_PARAMETERS_v03_00 params;
-} rpc_free_v03_00;
-
-typedef struct rpc_unloading_guest_driver_v1F_07
-{
-    NvBool     bInPMTransition;
-    NvBool     bGc6Entering;
-    NvU32      newLevel;
-} rpc_unloading_guest_driver_v1F_07;
-
-typedef struct rpc_update_bar_pde_v15_00
-{
-    UpdateBarPde_v15_00 info;
-} rpc_update_bar_pde_v15_00;
-
-typedef struct rpc_gsp_rm_alloc_v03_00
-{
-    NvHandle   hClient;
-    NvHandle   hParent;
-    NvHandle   hObject;
-    NvU32      hClass;
-    NvU32      status;
-    NvU32      paramsSize;
-    NvU32      flags;
-    NvU8       reserved[4];
-    NvU8       params[];
-} rpc_gsp_rm_alloc_v03_00;
-
-typedef struct rpc_gsp_rm_control_v03_00
-{
-    NvHandle   hClient;
-    NvHandle   hObject;
-    NvU32      cmd;
-    NvU32      status;
-    NvU32      paramsSize;
-    NvU32      flags;
-    NvU8       params[];
-} rpc_gsp_rm_control_v03_00;
-
-typedef struct rpc_run_cpu_sequencer_v17_00
-{
-    NvU32      bufferSizeDWord;
-    NvU32      cmdIndex;
-    NvU32      regSaveArea[8];
-    NvU32      commandBuffer[];
-} rpc_run_cpu_sequencer_v17_00;
-
-typedef struct rpc_post_event_v17_00
-{
-    NvHandle   hClient;
-    NvHandle   hEvent;
-    NvU32      notifyIndex;
-    NvU32      data;
-    NvU16      info16;
-    NvU32      status;
-    NvU32      eventDataSize;
-    NvBool     bNotifyList;
-    NvU8       eventData[];
-} rpc_post_event_v17_00;
-
-typedef struct rpc_rc_triggered_v17_02
-{
-    NvU32      nv2080EngineType;
-    NvU32      chid;
-    NvU32      exceptType;
-    NvU32      scope;
-    NvU16      partitionAttributionId;
-} rpc_rc_triggered_v17_02;
-
-typedef struct rpc_os_error_log_v17_00
-{
-    NvU32      exceptType;
-    NvU32      runlistId;
-    NvU32      chid;
-    char       errString[0x100];
-} rpc_os_error_log_v17_00;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_sdk-structures.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_sdk-structures.h
deleted file mode 100644
index e9fed4140468..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/generated/g_sdk-structures.h
+++ /dev/null
@@ -1,45 +0,0 @@
-#ifndef __src_nvidia_generated_g_sdk_structures_h__
-#define __src_nvidia_generated_g_sdk_structures_h__
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_headers.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2008-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct NVOS00_PARAMETERS_v03_00
-{
-    NvHandle   hRoot;
-    NvHandle   hObjectParent;
-    NvHandle   hObjectOld;
-    NvV32      status;
-} NVOS00_PARAMETERS_v03_00;
-
-typedef struct UpdateBarPde_v15_00
-{
-    NV_RPC_UPDATE_PDE_BAR_TYPE barType;
-    NvU64      entryValue NV_ALIGN_BYTES(8);
-    NvU64      entryLevelShift NV_ALIGN_BYTES(8);
-} UpdateBarPde_v15_00;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_acpi_data.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_acpi_data.h
deleted file mode 100644
index af50b11ec3b4..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_acpi_data.h
+++ /dev/null
@@ -1,74 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_gpu_acpi_data_h__
-#define __src_nvidia_inc_kernel_gpu_gpu_acpi_data_h__
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073system.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct DOD_METHOD_DATA
-{
-    NV_STATUS status;
-    NvU32     acpiIdListLen;
-    NvU32     acpiIdList[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
-} DOD_METHOD_DATA;
-
-typedef struct JT_METHOD_DATA
-{
-    NV_STATUS status;
-    NvU32     jtCaps;
-    NvU16     jtRevId;
-    NvBool    bSBIOSCaps;
-} JT_METHOD_DATA;
-
-typedef struct MUX_METHOD_DATA_ELEMENT
-{
-    NvU32       acpiId;
-    NvU32       mode;
-    NV_STATUS   status;
-} MUX_METHOD_DATA_ELEMENT;
-
-typedef struct MUX_METHOD_DATA
-{
-    NvU32                       tableLen;
-    MUX_METHOD_DATA_ELEMENT     acpiIdMuxModeTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
-    MUX_METHOD_DATA_ELEMENT     acpiIdMuxPartTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
-} MUX_METHOD_DATA;
-
-typedef struct CAPS_METHOD_DATA
-{
-    NV_STATUS status;
-    NvU32     optimusCaps;
-} CAPS_METHOD_DATA;
-
-typedef struct ACPI_METHOD_DATA
-{
-    NvBool                                               bValid;
-    DOD_METHOD_DATA                                      dodMethodData;
-    JT_METHOD_DATA                                       jtMethodData;
-    MUX_METHOD_DATA                                      muxMethodData;
-    CAPS_METHOD_DATA                                     capsMethodData;
-} ACPI_METHOD_DATA;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_engine_type.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_engine_type.h
deleted file mode 100644
index e3160c60036d..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_engine_type.h
+++ /dev/null
@@ -1,86 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_gpu_engine_type_h__
-#define __src_nvidia_inc_kernel_gpu_gpu_engine_type_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2021-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef enum
-{
-    RM_ENGINE_TYPE_NULL                 =       (0x00000000),
-    RM_ENGINE_TYPE_GR0                  =       (0x00000001),
-    RM_ENGINE_TYPE_GR1                  =       (0x00000002),
-    RM_ENGINE_TYPE_GR2                  =       (0x00000003),
-    RM_ENGINE_TYPE_GR3                  =       (0x00000004),
-    RM_ENGINE_TYPE_GR4                  =       (0x00000005),
-    RM_ENGINE_TYPE_GR5                  =       (0x00000006),
-    RM_ENGINE_TYPE_GR6                  =       (0x00000007),
-    RM_ENGINE_TYPE_GR7                  =       (0x00000008),
-    RM_ENGINE_TYPE_COPY0                =       (0x00000009),
-    RM_ENGINE_TYPE_COPY1                =       (0x0000000a),
-    RM_ENGINE_TYPE_COPY2                =       (0x0000000b),
-    RM_ENGINE_TYPE_COPY3                =       (0x0000000c),
-    RM_ENGINE_TYPE_COPY4                =       (0x0000000d),
-    RM_ENGINE_TYPE_COPY5                =       (0x0000000e),
-    RM_ENGINE_TYPE_COPY6                =       (0x0000000f),
-    RM_ENGINE_TYPE_COPY7                =       (0x00000010),
-    RM_ENGINE_TYPE_COPY8                =       (0x00000011),
-    RM_ENGINE_TYPE_COPY9                =       (0x00000012),
-    RM_ENGINE_TYPE_NVDEC0               =       (0x0000001d),
-    RM_ENGINE_TYPE_NVDEC1               =       (0x0000001e),
-    RM_ENGINE_TYPE_NVDEC2               =       (0x0000001f),
-    RM_ENGINE_TYPE_NVDEC3               =       (0x00000020),
-    RM_ENGINE_TYPE_NVDEC4               =       (0x00000021),
-    RM_ENGINE_TYPE_NVDEC5               =       (0x00000022),
-    RM_ENGINE_TYPE_NVDEC6               =       (0x00000023),
-    RM_ENGINE_TYPE_NVDEC7               =       (0x00000024),
-    RM_ENGINE_TYPE_NVENC0               =       (0x00000025),
-    RM_ENGINE_TYPE_NVENC1               =       (0x00000026),
-    RM_ENGINE_TYPE_NVENC2               =       (0x00000027),
-    RM_ENGINE_TYPE_VP                   =       (0x00000028),
-    RM_ENGINE_TYPE_ME                   =       (0x00000029),
-    RM_ENGINE_TYPE_PPP                  =       (0x0000002a),
-    RM_ENGINE_TYPE_MPEG                 =       (0x0000002b),
-    RM_ENGINE_TYPE_SW                   =       (0x0000002c),
-    RM_ENGINE_TYPE_TSEC                 =       (0x0000002d),
-    RM_ENGINE_TYPE_VIC                  =       (0x0000002e),
-    RM_ENGINE_TYPE_MP                   =       (0x0000002f),
-    RM_ENGINE_TYPE_SEC2                 =       (0x00000030),
-    RM_ENGINE_TYPE_HOST                 =       (0x00000031),
-    RM_ENGINE_TYPE_DPU                  =       (0x00000032),
-    RM_ENGINE_TYPE_PMU                  =       (0x00000033),
-    RM_ENGINE_TYPE_FBFLCN               =       (0x00000034),
-    RM_ENGINE_TYPE_NVJPEG0              =       (0x00000035),
-    RM_ENGINE_TYPE_NVJPEG1              =       (0x00000036),
-    RM_ENGINE_TYPE_NVJPEG2              =       (0x00000037),
-    RM_ENGINE_TYPE_NVJPEG3              =       (0x00000038),
-    RM_ENGINE_TYPE_NVJPEG4              =       (0x00000039),
-    RM_ENGINE_TYPE_NVJPEG5              =       (0x0000003a),
-    RM_ENGINE_TYPE_NVJPEG6              =       (0x0000003b),
-    RM_ENGINE_TYPE_NVJPEG7              =       (0x0000003c),
-    RM_ENGINE_TYPE_OFA                  =       (0x0000003d),
-    RM_ENGINE_TYPE_LAST                 =       (0x0000003e),
-} RM_ENGINE_TYPE;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_fw_heap.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_fw_heap.h
deleted file mode 100644
index 3abec59f0cc4..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_fw_heap.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_gsp_gsp_fw_heap_h__
-#define __src_nvidia_inc_kernel_gpu_gsp_gsp_fw_heap_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2022-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define GSP_FW_HEAP_PARAM_SIZE_PER_GB_FB                  (96 << 10)   // All architectures
-
-#define GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE      ((48 << 10) * 2048)   // Support 2048 channels
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_init_args.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_init_args.h
deleted file mode 100644
index 4033a6f85a76..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_init_args.h
+++ /dev/null
@@ -1,57 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_gsp_gsp_init_args_h__
-#define __src_nvidia_inc_kernel_gpu_gsp_gsp_init_args_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct {
-    RmPhysAddr sharedMemPhysAddr;
-    NvU32 pageTableEntryCount;
-    NvLength cmdQueueOffset;
-    NvLength statQueueOffset;
-    NvLength locklessCmdQueueOffset;
-    NvLength locklessStatQueueOffset;
-} MESSAGE_QUEUE_INIT_ARGUMENTS;
-
-typedef struct {
-    NvU32 oldLevel;
-    NvU32 flags;
-    NvBool bInPMTransition;
-} GSP_SR_INIT_ARGUMENTS;
-
-typedef struct
-{
-    MESSAGE_QUEUE_INIT_ARGUMENTS      messageQueueInitArguments;
-    GSP_SR_INIT_ARGUMENTS             srInitArguments;
-    NvU32                             gpuInstance;
-
-    struct
-    {
-        NvU64                         pa;
-        NvU64                         size;
-    } profilerArgs;
-} GSP_ARGUMENTS_CACHED;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_static_config.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_static_config.h
deleted file mode 100644
index eeab25a5e290..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_static_config.h
+++ /dev/null
@@ -1,174 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_gsp_gsp_static_config_h__
-#define __src_nvidia_inc_kernel_gpu_gsp_gsp_static_config_h__
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gpu.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080gr.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080bios.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fb.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gr.h>
-#include <nvrm/535.113.01/nvidia/generated/g_chipset_nvoc.h>
-#include <nvrm/535.113.01/nvidia/generated/g_gpu_nvoc.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_acpi_data.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/nvbitmask.h>
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_headers.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2019-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-typedef struct GSP_VF_INFO
-{
-    NvU32  totalVFs;
-    NvU32  firstVFOffset;
-    NvU64  FirstVFBar0Address;
-    NvU64  FirstVFBar1Address;
-    NvU64  FirstVFBar2Address;
-    NvBool b64bitBar0;
-    NvBool b64bitBar1;
-    NvBool b64bitBar2;
-} GSP_VF_INFO;
-
-typedef struct GspSMInfo_t
-{
-    NvU32 version;
-    NvU32 regBankCount;
-    NvU32 regBankRegCount;
-    NvU32 maxWarpsPerSM;
-    NvU32 maxThreadsPerWarp;
-    NvU32 geomGsObufEntries;
-    NvU32 geomXbufEntries;
-    NvU32 maxSPPerSM;
-    NvU32 rtCoreCount;
-} GspSMInfo;
-
-typedef struct GspStaticConfigInfo_t
-{
-    NvU8 grCapsBits[NV0080_CTRL_GR_CAPS_TBL_SIZE];
-    NV2080_CTRL_GPU_GET_GID_INFO_PARAMS gidInfo;
-    NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS gpcInfo;
-    NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS tpcInfo[MAX_GPC_COUNT];
-    NV2080_CTRL_GPU_GET_FERMI_ZCULL_INFO_PARAMS zcullInfo[MAX_GPC_COUNT];
-    NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS SKUInfo;
-    NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS fbRegionInfoParams;
-    COMPUTE_BRANDING_TYPE computeBranding;
-
-    NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS sriovCaps;
-    NvU32 sriovMaxGfid;
-
-    NvU32 engineCaps[NVGPU_ENGINE_CAPS_MASK_ARRAY_MAX];
-
-    GspSMInfo SM_info;
-
-    NvBool poisonFuseEnabled;
-
-    NvU64 fb_length;
-    NvU32 fbio_mask;
-    NvU32 fb_bus_width;
-    NvU32 fb_ram_type;
-    NvU32 fbp_mask;
-    NvU32 l2_cache_size;
-
-    NvU32 gfxpBufferSize[NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL];
-    NvU32 gfxpBufferAlignment[NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL];
-
-    NvU8 gpuNameString[NV2080_GPU_MAX_NAME_STRING_LENGTH];
-    NvU8 gpuShortNameString[NV2080_GPU_MAX_NAME_STRING_LENGTH];
-    NvU16 gpuNameString_Unicode[NV2080_GPU_MAX_NAME_STRING_LENGTH];
-    NvBool bGpuInternalSku;
-    NvBool bIsQuadroGeneric;
-    NvBool bIsQuadroAd;
-    NvBool bIsNvidiaNvs;
-    NvBool bIsVgx;
-    NvBool bGeforceSmb;
-    NvBool bIsTitan;
-    NvBool bIsTesla;
-    NvBool bIsMobile;
-    NvBool bIsGc6Rtd3Allowed;
-    NvBool bIsGcOffRtd3Allowed;
-    NvBool bIsGcoffLegacyAllowed;
-
-    NvU64 bar1PdeBase;
-    NvU64 bar2PdeBase;
-
-    NvBool bVbiosValid;
-    NvU32 vbiosSubVendor;
-    NvU32 vbiosSubDevice;
-
-    NvBool bPageRetirementSupported;
-
-    NvBool bSplitVasBetweenServerClientRm;
-
-    NvBool bClRootportNeedsNosnoopWAR;
-
-    VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS displaylessMaxHeads;
-    VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS displaylessMaxResolution;
-    NvU64 displaylessMaxPixels;
-
-    // Client handle for internal RMAPI control.
-    NvHandle hInternalClient;
-
-    // Device handle for internal RMAPI control.
-    NvHandle hInternalDevice;
-
-    // Subdevice handle for internal RMAPI control.
-    NvHandle hInternalSubdevice;
-
-    NvBool bSelfHostedMode;
-    NvBool bAtsSupported;
-
-    NvBool bIsGpuUefi;
-} GspStaticConfigInfo;
-
-typedef struct GspSystemInfo
-{
-    NvU64 gpuPhysAddr;
-    NvU64 gpuPhysFbAddr;
-    NvU64 gpuPhysInstAddr;
-    NvU64 nvDomainBusDeviceFunc;
-    NvU64 simAccessBufPhysAddr;
-    NvU64 pcieAtomicsOpMask;
-    NvU64 consoleMemSize;
-    NvU64 maxUserVa;
-    NvU32 pciConfigMirrorBase;
-    NvU32 pciConfigMirrorSize;
-    NvU8 oorArch;
-    NvU64 clPdbProperties;
-    NvU32 Chipset;
-    NvBool bGpuBehindBridge;
-    NvBool bMnocAvailable;
-    NvBool bUpstreamL0sUnsupported;
-    NvBool bUpstreamL1Unsupported;
-    NvBool bUpstreamL1PorSupported;
-    NvBool bUpstreamL1PorMobileOnly;
-    NvU8   upstreamAddressValid;
-    BUSINFO FHBBusInfo;
-    BUSINFO chipsetIDInfo;
-    ACPI_METHOD_DATA acpiMethodData;
-    NvU32 hypervisorType;
-    NvBool bIsPassthru;
-    NvU64 sysTimerOffsetNs;
-    GSP_VF_INFO gspVFInfo;
-} GspSystemInfo;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/intr/engine_idx.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/intr/engine_idx.h
deleted file mode 100644
index bd5e01f9814b..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/intr/engine_idx.h
+++ /dev/null
@@ -1,57 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_intr_engine_idx_h__
-#define __src_nvidia_inc_kernel_gpu_intr_engine_idx_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 1993-2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define MC_ENGINE_IDX_DISP                          2
-
-#define MC_ENGINE_IDX_CE0                           15
-
-#define MC_ENGINE_IDX_CE9                           24
-
-#define MC_ENGINE_IDX_MSENC                         38
-
-#define MC_ENGINE_IDX_MSENC2                        40
-
-#define MC_ENGINE_IDX_GSP                           49
-#define MC_ENGINE_IDX_NVJPG                         50
-#define MC_ENGINE_IDX_NVJPEG                        MC_ENGINE_IDX_NVJPG
-#define MC_ENGINE_IDX_NVJPEG0                       MC_ENGINE_IDX_NVJPEG
-
-#define MC_ENGINE_IDX_NVJPEG7                       57
-
-#define MC_ENGINE_IDX_BSP                           64
-#define MC_ENGINE_IDX_NVDEC                         MC_ENGINE_IDX_BSP
-#define MC_ENGINE_IDX_NVDEC0                        MC_ENGINE_IDX_NVDEC
-
-#define MC_ENGINE_IDX_NVDEC7                        71
-
-#define MC_ENGINE_IDX_OFA0                          80
-
-#define MC_ENGINE_IDX_GR                            82
-#define MC_ENGINE_IDX_GR0                           MC_ENGINE_IDX_GR
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/nvbitmask.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/nvbitmask.h
deleted file mode 100644
index 366447a368bf..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/gpu/nvbitmask.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_gpu_nvbitmask_h__
-#define __src_nvidia_inc_kernel_gpu_nvbitmask_h__
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_engine_type.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2021-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NVGPU_ENGINE_CAPS_MASK_BITS                32
-#define NVGPU_ENGINE_CAPS_MASK_ARRAY_MAX           ((RM_ENGINE_TYPE_LAST-1)/NVGPU_ENGINE_CAPS_MASK_BITS + 1)
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/os/nv_memory_type.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/os/nv_memory_type.h
deleted file mode 100644
index 4a850dad4776..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/inc/kernel/os/nv_memory_type.h
+++ /dev/null
@@ -1,31 +0,0 @@
-#ifndef __src_nvidia_inc_kernel_os_nv_memory_type_h__
-#define __src_nvidia_inc_kernel_os_nv_memory_type_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2020 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define NV_MEMORY_WRITECOMBINED      2
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_headers.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_headers.h
deleted file mode 100644
index f14b23852456..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_headers.h
+++ /dev/null
@@ -1,51 +0,0 @@
-#ifndef __src_nvidia_kernel_inc_vgpu_rpc_headers_h__
-#define __src_nvidia_kernel_inc_vgpu_rpc_headers_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2017-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-#define MAX_GPC_COUNT           32
-
-typedef enum
-{
-    NV_RPC_UPDATE_PDE_BAR_1,
-    NV_RPC_UPDATE_PDE_BAR_2,
-    NV_RPC_UPDATE_PDE_BAR_INVALID,
-} NV_RPC_UPDATE_PDE_BAR_TYPE;
-
-typedef struct VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS 
-{
-    NvU32 headIndex;
-    NvU32 maxHResolution;
-    NvU32 maxVResolution;
-} VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS;
-
-typedef struct VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS 
-{
-    NvU32 numHeads;
-    NvU32 maxNumHeads;
-} VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS;
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/sdk-structures.h b/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/sdk-structures.h
deleted file mode 100644
index 7801af232dff..000000000000
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/sdk-structures.h
+++ /dev/null
@@ -1,40 +0,0 @@
-#ifndef __src_nvidia_kernel_inc_vgpu_sdk_structures_h__
-#define __src_nvidia_kernel_inc_vgpu_sdk_structures_h__
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
-
-struct pte_desc
-{
-    NvU32 idr:2;
-    NvU32 reserved1:14;
-    NvU32 length:16;
-    union {
-        NvU64 pte; // PTE when IDR==0; PDE when IDR > 0
-        NvU64 pde; // PTE when IDR==0; PDE when IDR > 0
-    } pte_pde[]  NV_ALIGN_BYTES(8); // PTE when IDR==0; PDE when IDR > 0
-};
-
-#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/nvtypes.h b/drivers/gpu/drm/nouveau/include/nvrm/nvtypes.h
index e6833df1ccc7..af11648ad9c8 100644
--- a/drivers/gpu/drm/nouveau/include/nvrm/nvtypes.h
+++ b/drivers/gpu/drm/nouveau/include/nvrm/nvtypes.h
@@ -21,4 +21,6 @@ typedef NvU64 NvLength;
 typedef NvU64 RmPhysAddr;
 
 typedef NvU32 NV_STATUS;
+
+typedef union {} rpc_generic_union;
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index 2a0617e5fe2a..a3ba07fc48a0 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -315,7 +315,7 @@ nouveau_abi16_ioctl_getparam(ABI16_IOCTL_ARGS)
 		break;
 	}
 	case NOUVEAU_GETPARAM_VRAM_BAR_SIZE:
-		getparam->value = nvkm_device->func->resource_size(nvkm_device, 1);
+		getparam->value = nvkm_device->func->resource_size(nvkm_device, NVKM_BAR1_FB);
 		break;
 	case NOUVEAU_GETPARAM_VRAM_USED: {
 		struct ttm_resource_manager *vram_mgr = ttm_manager_type(&drm->ttm.bdev, TTM_PL_VRAM);
@@ -416,7 +416,7 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 	 */
 	if (nouveau_cli_uvmm(cli)) {
 		ret = nouveau_sched_create(&chan->sched, drm, drm->sched_wq,
-					   chan->chan->dma.ib_max);
+					   chan->chan->chan.gpfifo.max);
 		if (ret)
 			goto done;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index db961eade225..b96f0555ca14 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -144,6 +144,9 @@ nouveau_bo_del_ttm(struct ttm_buffer_object *bo)
 	nouveau_bo_del_io_reserve_lru(bo);
 	nv10_bo_put_tile_region(dev, nvbo->tile, NULL);
 
+	if (bo->base.import_attach)
+		drm_prime_gem_destroy(&bo->base, bo->sg);
+
 	/*
 	 * If nouveau_bo_new() allocated this buffer, the GEM object was never
 	 * initialized, so don't attempt to release it.
@@ -398,6 +401,83 @@ nouveau_bo_new(struct nouveau_cli *cli, u64 size, int align,
 	return 0;
 }
 
+void
+nouveau_bo_unpin_del(struct nouveau_bo **pnvbo)
+{
+	struct nouveau_bo *nvbo = *pnvbo;
+
+	if (!nvbo)
+		return;
+
+	nouveau_bo_unmap(nvbo);
+	nouveau_bo_unpin(nvbo);
+	nouveau_bo_fini(nvbo);
+
+	*pnvbo = NULL;
+}
+
+int
+nouveau_bo_new_pin(struct nouveau_cli *cli, u32 domain, u32 size, struct nouveau_bo **pnvbo)
+{
+	struct nouveau_bo *nvbo;
+	int ret;
+
+	ret = nouveau_bo_new(cli, size, 0, domain, 0, 0, NULL, NULL, &nvbo);
+	if (ret)
+		return ret;
+
+	ret = nouveau_bo_pin(nvbo, domain, false);
+	if (ret) {
+		nouveau_bo_fini(nvbo);
+		return ret;
+	}
+
+	*pnvbo = nvbo;
+	return 0;
+}
+
+int
+nouveau_bo_new_map(struct nouveau_cli *cli, u32 domain, u32 size, struct nouveau_bo **pnvbo)
+{
+	struct nouveau_bo *nvbo;
+	int ret;
+
+	ret = nouveau_bo_new_pin(cli, domain, size, &nvbo);
+	if (ret)
+		return ret;
+
+	ret = nouveau_bo_map(nvbo);
+	if (ret) {
+		nouveau_bo_unpin_del(&nvbo);
+		return ret;
+	}
+
+	*pnvbo = nvbo;
+	return 0;
+}
+
+int
+nouveau_bo_new_map_gpu(struct nouveau_cli *cli, u32 domain, u32 size,
+		       struct nouveau_bo **pnvbo, struct nouveau_vma **pvma)
+{
+	struct nouveau_vmm *vmm = nouveau_cli_vmm(cli);
+	struct nouveau_bo *nvbo;
+	int ret;
+
+	ret = nouveau_bo_new_map(cli, domain, size, &nvbo);
+	if (ret)
+		return ret;
+
+	ret = nouveau_vma_new(nvbo, vmm, pvma);
+	if (ret) {
+		nouveau_bo_unpin_del(&nvbo);
+		return ret;
+	}
+
+	*pnvbo = nvbo;
+	return 0;
+}
+
 static void
 set_placement_range(struct nouveau_bo *nvbo, uint32_t domain)
 {
@@ -920,6 +1000,9 @@ nouveau_bo_move_init(struct nouveau_drm *drm)
 			    struct ttm_resource *, struct ttm_resource *);
 		int (*init)(struct nouveau_channel *, u32 handle);
 	} _methods[] = {
+		{  "COPY", 4, 0xcab5, nve0_bo_move_copy, nve0_bo_move_init },
+		{  "COPY", 4, 0xc9b5, nve0_bo_move_copy, nve0_bo_move_init },
+		{  "COPY", 4, 0xc8b5, nve0_bo_move_copy, nve0_bo_move_init },
 		{  "COPY", 4, 0xc7b5, nve0_bo_move_copy, nve0_bo_move_init },
 		{  "GRCE", 0, 0xc7b5, nve0_bo_move_copy, nvc0_bo_move_init },
 		{  "COPY", 4, 0xc6b5, nve0_bo_move_copy, nve0_bo_move_init },
@@ -1201,7 +1284,7 @@ retry:
 		fallthrough;	/* tiled memory */
 	case TTM_PL_VRAM:
 		reg->bus.offset = (reg->start << PAGE_SHIFT) +
-			device->func->resource_addr(device, 1);
+			device->func->resource_addr(device, NVKM_BAR1_FB);
 		reg->bus.is_iomem = true;
 
 		/* Some BARs do not support being ioremapped WC */
@@ -1292,7 +1375,7 @@ vm_fault_t nouveau_ttm_fault_reserve_notify(struct ttm_buffer_object *bo)
 	struct nouveau_drm *drm = nouveau_bdev(bo->bdev);
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 	struct nvkm_device *device = nvxx_device(drm);
-	u32 mappable = device->func->resource_size(device, 1) >> PAGE_SHIFT;
+	u32 mappable = device->func->resource_size(device, NVKM_BAR1_FB) >> PAGE_SHIFT;
 	int i, ret;
 
 	/* as long as the bo isn't in vram, and isn't tiled, we've got
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.h b/drivers/gpu/drm/nouveau/nouveau_bo.h
index 596a63a50a20..d59fd12268b9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.h
@@ -9,6 +9,7 @@ struct nouveau_channel;
 struct nouveau_cli;
 struct nouveau_drm;
 struct nouveau_fence;
+struct nouveau_vma;
 
 struct nouveau_bo {
 	struct ttm_buffer_object bo;
@@ -89,6 +90,12 @@ void nouveau_bo_sync_for_cpu(struct nouveau_bo *nvbo);
 void nouveau_bo_add_io_reserve_lru(struct ttm_buffer_object *bo);
 void nouveau_bo_del_io_reserve_lru(struct ttm_buffer_object *bo);
 
+int nouveau_bo_new_pin(struct nouveau_cli *, u32 domain, u32 size, struct nouveau_bo **);
+int nouveau_bo_new_map(struct nouveau_cli *, u32 domain, u32 size, struct nouveau_bo **);
+int nouveau_bo_new_map_gpu(struct nouveau_cli *, u32 domain, u32 size,
+			   struct nouveau_bo **, struct nouveau_vma **);
+void nouveau_bo_unpin_del(struct nouveau_bo **);
+
 /* TODO: submit equivalent to TTM generic API upstream? */
 static inline void __iomem *
 nvbo_kmap_obj_iovirtual(struct nouveau_bo *nvbo)
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index cd659b9fd1d9..b1e92b1f7a26 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -103,12 +103,11 @@ nouveau_channel_del(struct nouveau_channel **pchan)
 		nvif_event_dtor(&chan->kill);
 		nvif_object_dtor(&chan->user);
 		nvif_mem_dtor(&chan->mem_userd);
+		nouveau_vma_del(&chan->sema.vma);
+		nouveau_bo_unpin_del(&chan->sema.bo);
 		nvif_object_dtor(&chan->push.ctxdma);
 		nouveau_vma_del(&chan->push.vma);
-		nouveau_bo_unmap(chan->push.buffer);
-		if (chan->push.buffer && chan->push.buffer->bo.pin_count)
-			nouveau_bo_unpin(chan->push.buffer);
-		nouveau_bo_fini(chan->push.buffer);
+		nouveau_bo_unpin_del(&chan->push.buffer);
 		kfree(chan);
 	}
 	*pchan = NULL;
@@ -163,14 +162,7 @@ nouveau_channel_prep(struct nouveau_cli *cli,
 	if (nouveau_vram_pushbuf)
 		target = NOUVEAU_GEM_DOMAIN_VRAM;
 
-	ret = nouveau_bo_new(cli, size, 0, target, 0, 0, NULL, NULL,
-			    &chan->push.buffer);
-	if (ret == 0) {
-		ret = nouveau_bo_pin(chan->push.buffer, target, false);
-		if (ret == 0)
-			ret = nouveau_bo_map(chan->push.buffer);
-	}
-
+	ret = nouveau_bo_new_map(cli, target, size, &chan->push.buffer);
 	if (ret) {
 		nouveau_channel_del(pchan);
 		return ret;
@@ -199,8 +191,10 @@ nouveau_channel_prep(struct nouveau_cli *cli,
 
 		chan->push.addr = chan->push.vma->addr;
 
-		if (device->info.family >= NV_DEVICE_INFO_V0_FERMI)
-			return 0;
+		if (device->info.family >= NV_DEVICE_INFO_V0_FERMI) {
+			return nouveau_bo_new_map_gpu(cli, NOUVEAU_GEM_DOMAIN_GART, PAGE_SIZE,
+						      &chan->sema.bo, &chan->sema.vma);
+		}
 
 		args.target = NV_DMA_V0_TARGET_VM;
 		args.access = NV_DMA_V0_ACCESS_VM;
@@ -209,13 +203,15 @@ nouveau_channel_prep(struct nouveau_cli *cli,
 	} else
 	if (chan->push.buffer->bo.resource->mem_type == TTM_PL_VRAM) {
 		if (device->info.family == NV_DEVICE_INFO_V0_TNT) {
+			struct nvkm_device *nvkm_device = nvxx_device(drm);
+
 			/* nv04 vram pushbuf hack, retarget to its location in
 			 * the framebuffer bar rather than direct vram access..
 			 * nfi why this exists, it came from the -nv ddx.
 			 */
 			args.target = NV_DMA_V0_TARGET_PCI;
 			args.access = NV_DMA_V0_ACCESS_RDWR;
-			args.start = nvxx_device(drm)->func->resource_addr(nvxx_device(drm), 1);
+			args.start = nvkm_device->func->resource_addr(nvkm_device, NVKM_BAR1_FB);
 			args.limit = args.start + device->info.ram_user - 1;
 		} else {
 			args.target = NV_DMA_V0_TARGET_VRAM;
@@ -253,27 +249,27 @@ nouveau_channel_ctor(struct nouveau_cli *cli, bool priv, u64 runm,
 		     struct nouveau_channel **pchan)
 {
 	const struct nvif_mclass hosts[] = {
-		{  AMPERE_CHANNEL_GPFIFO_B, 0 },
-		{  AMPERE_CHANNEL_GPFIFO_A, 0 },
-		{  TURING_CHANNEL_GPFIFO_A, 0 },
-		{   VOLTA_CHANNEL_GPFIFO_A, 0 },
-		{  PASCAL_CHANNEL_GPFIFO_A, 0 },
-		{ MAXWELL_CHANNEL_GPFIFO_A, 0 },
-		{  KEPLER_CHANNEL_GPFIFO_B, 0 },
-		{  KEPLER_CHANNEL_GPFIFO_A, 0 },
-		{   FERMI_CHANNEL_GPFIFO  , 0 },
-		{     G82_CHANNEL_GPFIFO  , 0 },
-		{    NV50_CHANNEL_GPFIFO  , 0 },
-		{    NV40_CHANNEL_DMA     , 0 },
-		{    NV17_CHANNEL_DMA     , 0 },
-		{    NV10_CHANNEL_DMA     , 0 },
-		{    NV03_CHANNEL_DMA     , 0 },
+		{ BLACKWELL_CHANNEL_GPFIFO_B, 0 },
+		{ BLACKWELL_CHANNEL_GPFIFO_A, 0 },
+		{    HOPPER_CHANNEL_GPFIFO_A, 0 },
+		{    AMPERE_CHANNEL_GPFIFO_B, 0 },
+		{    AMPERE_CHANNEL_GPFIFO_A, 0 },
+		{    TURING_CHANNEL_GPFIFO_A, 0 },
+		{     VOLTA_CHANNEL_GPFIFO_A, 0 },
+		{    PASCAL_CHANNEL_GPFIFO_A, 0 },
+		{   MAXWELL_CHANNEL_GPFIFO_A, 0 },
+		{    KEPLER_CHANNEL_GPFIFO_B, 0 },
+		{    KEPLER_CHANNEL_GPFIFO_A, 0 },
+		{     FERMI_CHANNEL_GPFIFO  , 0 },
+		{       G82_CHANNEL_GPFIFO  , 0 },
+		{      NV50_CHANNEL_GPFIFO  , 0 },
+		{      NV40_CHANNEL_DMA     , 0 },
+		{      NV17_CHANNEL_DMA     , 0 },
+		{      NV10_CHANNEL_DMA     , 0 },
+		{      NV03_CHANNEL_DMA     , 0 },
 		{}
 	};
-	struct {
-		struct nvif_chan_v0 chan;
-		char name[TASK_COMM_LEN+16];
-	} args;
+	DEFINE_RAW_FLEX(struct nvif_chan_v0, args, name, TASK_COMM_LEN + 16);
 	struct nvif_device *device = &cli->device;
 	struct nouveau_channel *chan;
 	const u64 plength = 0x10000;
@@ -298,28 +294,28 @@ nouveau_channel_ctor(struct nouveau_cli *cli, bool priv, u64 runm,
 		return ret;
 
 	/* create channel object */
-	args.chan.version = 0;
-	args.chan.namelen = sizeof(args.name);
-	args.chan.runlist = __ffs64(runm);
-	args.chan.runq = 0;
-	args.chan.priv = priv;
-	args.chan.devm = BIT(0);
+	args->version = 0;
+	args->namelen = __member_size(args->name);
+	args->runlist = __ffs64(runm);
+	args->runq = 0;
+	args->priv = priv;
+	args->devm = BIT(0);
 	if (hosts[cid].oclass < NV50_CHANNEL_GPFIFO) {
-		args.chan.vmm = 0;
-		args.chan.ctxdma = nvif_handle(&chan->push.ctxdma);
-		args.chan.offset = chan->push.addr;
-		args.chan.length = 0;
+		args->vmm = 0;
+		args->ctxdma = nvif_handle(&chan->push.ctxdma);
+		args->offset = chan->push.addr;
+		args->length = 0;
 	} else {
-		args.chan.vmm = nvif_handle(&chan->vmm->vmm.object);
+		args->vmm = nvif_handle(&chan->vmm->vmm.object);
 		if (hosts[cid].oclass < FERMI_CHANNEL_GPFIFO)
-			args.chan.ctxdma = nvif_handle(&chan->push.ctxdma);
+			args->ctxdma = nvif_handle(&chan->push.ctxdma);
 		else
-			args.chan.ctxdma = 0;
-		args.chan.offset = ioffset + chan->push.addr;
-		args.chan.length = ilength;
+			args->ctxdma = 0;
+		args->offset = ioffset + chan->push.addr;
+		args->length = ilength;
 	}
-	args.chan.huserd = 0;
-	args.chan.ouserd = 0;
+	args->huserd = 0;
+	args->ouserd = 0;
 
 	/* allocate userd */
 	if (hosts[cid].oclass >= VOLTA_CHANNEL_GPFIFO_A) {
@@ -329,27 +325,28 @@ nouveau_channel_ctor(struct nouveau_cli *cli, bool priv, u64 runm,
 		if (ret)
 			return ret;
 
-		args.chan.huserd = nvif_handle(&chan->mem_userd.object);
-		args.chan.ouserd = 0;
+		args->huserd = nvif_handle(&chan->mem_userd.object);
+		args->ouserd = 0;
 
 		chan->userd = &chan->mem_userd.object;
 	} else {
 		chan->userd = &chan->user;
 	}
 
-	snprintf(args.name, sizeof(args.name), "%s[%d]", current->comm, task_pid_nr(current));
+	snprintf(args->name, __member_size(args->name), "%s[%d]",
+		 current->comm, task_pid_nr(current));
 
 	ret = nvif_object_ctor(&device->object, "abi16ChanUser", 0, hosts[cid].oclass,
-			       &args, sizeof(args), &chan->user);
+			       args, __struct_size(args), &chan->user);
 	if (ret) {
 		nouveau_channel_del(pchan);
 		return ret;
 	}
 
-	chan->runlist = args.chan.runlist;
-	chan->chid = args.chan.chid;
-	chan->inst = args.chan.inst;
-	chan->token = args.chan.token;
+	chan->runlist = args->runlist;
+	chan->chid = args->chid;
+	chan->inst = args->inst;
+	chan->token = args->token;
 	return 0;
 }
 
@@ -367,17 +364,17 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 		return ret;
 
 	if (chan->user.oclass >= FERMI_CHANNEL_GPFIFO) {
-		struct {
-			struct nvif_event_v0 base;
-			struct nvif_chan_event_v0 host;
-		} args;
+		DEFINE_RAW_FLEX(struct nvif_event_v0, args, data,
+				sizeof(struct nvif_chan_event_v0));
+		struct nvif_chan_event_v0 *host =
+				(struct nvif_chan_event_v0 *)args->data;
 
-		args.host.version = 0;
-		args.host.type = NVIF_CHAN_EVENT_V0_KILLED;
+		host->version = 0;
+		host->type = NVIF_CHAN_EVENT_V0_KILLED;
 
 		ret = nvif_event_ctor(&chan->user, "abi16ChanKilled", chan->chid,
 				      nouveau_channel_killed, false,
-				      &args.base, sizeof(args), &chan->kill);
+				      args, __struct_size(args), &chan->kill);
 		if (ret == 0)
 			ret = nvif_event_allow(&chan->kill);
 		if (ret) {
@@ -433,25 +430,33 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 	}
 
 	/* initialise dma tracking parameters */
-	switch (chan->user.oclass) {
-	case NV03_CHANNEL_DMA:
-	case NV10_CHANNEL_DMA:
-	case NV17_CHANNEL_DMA:
-	case NV40_CHANNEL_DMA:
+	if (chan->user.oclass < NV50_CHANNEL_GPFIFO) {
 		chan->user_put = 0x40;
 		chan->user_get = 0x44;
 		chan->dma.max = (0x10000 / 4) - 2;
-		break;
-	default:
-		chan->user_put = 0x40;
-		chan->user_get = 0x44;
-		chan->user_get_hi = 0x60;
-		chan->dma.ib_base =  0x10000 / 4;
-		chan->dma.ib_max  = NV50_DMA_IB_MAX;
-		chan->dma.ib_put  = 0;
-		chan->dma.ib_free = chan->dma.ib_max - chan->dma.ib_put;
-		chan->dma.max = chan->dma.ib_base;
-		break;
+	} else
+	if (chan->user.oclass < FERMI_CHANNEL_GPFIFO) {
+		ret = nvif_chan506f_ctor(&chan->chan, chan->userd->map.ptr,
+					 (u8*)chan->push.buffer->kmap.virtual + 0x10000, 0x2000,
+					 chan->push.buffer->kmap.virtual, chan->push.addr, 0x10000);
+		if (ret)
+			return ret;
+	} else
+	if (chan->user.oclass < VOLTA_CHANNEL_GPFIFO_A) {
+		ret = nvif_chan906f_ctor(&chan->chan, chan->userd->map.ptr,
+					 (u8*)chan->push.buffer->kmap.virtual + 0x10000, 0x2000,
+					 chan->push.buffer->kmap.virtual, chan->push.addr, 0x10000,
+					 chan->sema.bo->kmap.virtual, chan->sema.vma->addr);
+		if (ret)
+			return ret;
+	} else {
+		ret = nvif_chanc36f_ctor(&chan->chan, chan->userd->map.ptr,
+					 (u8*)chan->push.buffer->kmap.virtual + 0x10000, 0x2000,
+					 chan->push.buffer->kmap.virtual, chan->push.addr, 0x10000,
+					 chan->sema.bo->kmap.virtual, chan->sema.vma->addr,
+					 &drm->client.device.user, chan->token);
+		if (ret)
+			return ret;
 	}
 
 	chan->dma.put = 0;
@@ -520,46 +525,44 @@ nouveau_channels_fini(struct nouveau_drm *drm)
 int
 nouveau_channels_init(struct nouveau_drm *drm)
 {
-	struct {
-		struct nv_device_info_v1 m;
-		struct {
-			struct nv_device_info_v1_data channels;
-			struct nv_device_info_v1_data runlists;
-		} v;
-	} args = {
-		.m.version = 1,
-		.m.count = sizeof(args.v) / sizeof(args.v.channels),
-		.v.channels.mthd = NV_DEVICE_HOST_CHANNELS,
-		.v.runlists.mthd = NV_DEVICE_HOST_RUNLISTS,
-	};
+	DEFINE_RAW_FLEX(struct nv_device_info_v1, args, data, 2);
+	struct nv_device_info_v1_data *channels = &args->data[0];
+	struct nv_device_info_v1_data *runlists = &args->data[1];
 	struct nvif_object *device = &drm->client.device.object;
 	int ret, i;
 
-	ret = nvif_object_mthd(device, NV_DEVICE_V0_INFO, &args, sizeof(args));
+	args->version = 1;
+	args->count = __member_size(args->data) / sizeof(*args->data);
+	channels->mthd = NV_DEVICE_HOST_CHANNELS;
+	runlists->mthd = NV_DEVICE_HOST_RUNLISTS;
+
+	ret = nvif_object_mthd(device, NV_DEVICE_V0_INFO, args,
+			       __struct_size(args));
 	if (ret ||
-	    args.v.runlists.mthd == NV_DEVICE_INFO_INVALID || !args.v.runlists.data ||
-	    args.v.channels.mthd == NV_DEVICE_INFO_INVALID)
+	    runlists->mthd == NV_DEVICE_INFO_INVALID || !runlists->data ||
+	    channels->mthd == NV_DEVICE_INFO_INVALID)
 		return -ENODEV;
 
-	drm->chan_nr = drm->chan_total = args.v.channels.data;
-	drm->runl_nr = fls64(args.v.runlists.data);
+	drm->chan_nr = drm->chan_total = channels->data;
+	drm->runl_nr = fls64(runlists->data);
 	drm->runl = kcalloc(drm->runl_nr, sizeof(*drm->runl), GFP_KERNEL);
 	if (!drm->runl)
 		return -ENOMEM;
 
 	if (drm->chan_nr == 0) {
 		for (i = 0; i < drm->runl_nr; i++) {
-			if (!(args.v.runlists.data & BIT(i)))
+			if (!(runlists->data & BIT(i)))
 				continue;
 
-			args.v.channels.mthd = NV_DEVICE_HOST_RUNLIST_CHANNELS;
-			args.v.channels.data = i;
+			channels->mthd = NV_DEVICE_HOST_RUNLIST_CHANNELS;
+			channels->data = i;
 
-			ret = nvif_object_mthd(device, NV_DEVICE_V0_INFO, &args, sizeof(args));
-			if (ret || args.v.channels.mthd == NV_DEVICE_INFO_INVALID)
+			ret = nvif_object_mthd(device, NV_DEVICE_V0_INFO, args,
+					       __struct_size(args));
+			if (ret || channels->mthd == NV_DEVICE_INFO_INVALID)
 				return -ENODEV;
 
-			drm->runl[i].chan_nr = args.v.channels.data;
+			drm->runl[i].chan_nr = channels->data;
 			drm->runl[i].chan_id_base = drm->chan_total;
 			drm->runl[i].context_base = dma_fence_context_alloc(drm->runl[i].chan_nr);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.h b/drivers/gpu/drm/nouveau/nouveau_chan.h
index 016f668c0bc1..561877725aac 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.h
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.h
@@ -3,13 +3,11 @@
 #define __NOUVEAU_CHAN_H__
 #include <nvif/object.h>
 #include <nvif/event.h>
-#include <nvif/push.h>
+#include <nvif/chan.h>
 struct nvif_device;
 
 struct nouveau_channel {
-	struct {
-		struct nvif_push push;
-	} chan;
+	struct nvif_chan chan;
 
 	struct nouveau_cli *cli;
 	struct nouveau_vmm *vmm;
@@ -41,15 +39,15 @@ struct nouveau_channel {
 		int free;
 		int cur;
 		int put;
-		int ib_base;
-		int ib_max;
-		int ib_free;
-		int ib_put;
 	} dma;
-	u32 user_get_hi;
 	u32 user_get;
 	u32 user_put;
 
+	struct {
+		struct nouveau_bo *bo;
+		struct nouveau_vma *vma;
+	} sema;
+
 	struct nvif_object user;
 	struct nvif_object blit;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index 1b10c6c12f46..63621b1510f6 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -1401,6 +1401,8 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		nv_connector->aux.drm_dev = dev;
 		nv_connector->aux.transfer = nouveau_connector_aux_xfer;
 		nv_connector->aux.name = connector->name;
+		if (disp->disp.object.oclass >= GB202_DISP)
+			nv_connector->aux.no_zero_sized = true;
 		drm_dp_aux_init(&nv_connector->aux);
 		break;
 	default:
diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.c b/drivers/gpu/drm/nouveau/nouveau_dma.c
index a1f329ef0641..017a803121d4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.c
@@ -43,8 +43,6 @@ READ_GET(struct nouveau_channel *chan, uint64_t *prev_get, int *timeout)
 	uint64_t val;
 
 	val = nvif_rd32(chan->userd, chan->user_get);
-        if (chan->user_get_hi)
-		val |= (uint64_t)nvif_rd32(chan->userd, chan->user_get_hi) << 32;
 
 	/* reset counter as long as GET is still advancing, this is
 	 * to avoid misdetecting a GPU lockup if the GPU happens to
@@ -68,111 +66,12 @@ READ_GET(struct nouveau_channel *chan, uint64_t *prev_get, int *timeout)
 	return (val - chan->push.addr) >> 2;
 }
 
-void
-nv50_dma_push(struct nouveau_channel *chan, u64 offset, u32 length,
-	      bool no_prefetch)
-{
-	struct nvif_user *user = &chan->cli->drm->client.device.user;
-	struct nouveau_bo *pb = chan->push.buffer;
-	int ip = (chan->dma.ib_put * 2) + chan->dma.ib_base;
-
-	BUG_ON(chan->dma.ib_free < 1);
-	WARN_ON(length > NV50_DMA_PUSH_MAX_LENGTH);
-
-	nouveau_bo_wr32(pb, ip++, lower_32_bits(offset));
-	nouveau_bo_wr32(pb, ip++, upper_32_bits(offset) | length << 8 |
-			(no_prefetch ? (1 << 31) : 0));
-
-	chan->dma.ib_put = (chan->dma.ib_put + 1) & chan->dma.ib_max;
-
-	mb();
-	/* Flush writes. */
-	nouveau_bo_rd32(pb, 0);
-
-	nvif_wr32(chan->userd, 0x8c, chan->dma.ib_put);
-	if (user->func && user->func->doorbell)
-		user->func->doorbell(user, chan->token);
-	chan->dma.ib_free--;
-}
-
-static int
-nv50_dma_push_wait(struct nouveau_channel *chan, int count)
-{
-	uint32_t cnt = 0, prev_get = 0;
-
-	while (chan->dma.ib_free < count) {
-		uint32_t get = nvif_rd32(chan->userd, 0x88);
-		if (get != prev_get) {
-			prev_get = get;
-			cnt = 0;
-		}
-
-		if ((++cnt & 0xff) == 0) {
-			udelay(1);
-			if (cnt > 100000)
-				return -EBUSY;
-		}
-
-		chan->dma.ib_free = get - chan->dma.ib_put;
-		if (chan->dma.ib_free <= 0)
-			chan->dma.ib_free += chan->dma.ib_max;
-	}
-
-	return 0;
-}
-
-static int
-nv50_dma_wait(struct nouveau_channel *chan, int slots, int count)
-{
-	uint64_t prev_get = 0;
-	int ret, cnt = 0;
-
-	ret = nv50_dma_push_wait(chan, slots + 1);
-	if (unlikely(ret))
-		return ret;
-
-	while (chan->dma.free < count) {
-		int get = READ_GET(chan, &prev_get, &cnt);
-		if (unlikely(get < 0)) {
-			if (get == -EINVAL)
-				continue;
-
-			return get;
-		}
-
-		if (get <= chan->dma.cur) {
-			chan->dma.free = chan->dma.max - chan->dma.cur;
-			if (chan->dma.free >= count)
-				break;
-
-			FIRE_RING(chan);
-			do {
-				get = READ_GET(chan, &prev_get, &cnt);
-				if (unlikely(get < 0)) {
-					if (get == -EINVAL)
-						continue;
-					return get;
-				}
-			} while (get == 0);
-			chan->dma.cur = 0;
-			chan->dma.put = 0;
-		}
-
-		chan->dma.free = get - chan->dma.cur - 1;
-	}
-
-	return 0;
-}
-
 int
-nouveau_dma_wait(struct nouveau_channel *chan, int slots, int size)
+nouveau_dma_wait(struct nouveau_channel *chan, int size)
 {
 	uint64_t prev_get = 0;
 	int cnt = 0, get;
 
-	if (chan->dma.ib_max)
-		return nv50_dma_wait(chan, slots, size);
-
 	while (chan->dma.free < size) {
 		get = READ_GET(chan, &prev_get, &cnt);
 		if (unlikely(get == -EBUSY))
diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.h b/drivers/gpu/drm/nouveau/nouveau_dma.h
index c52cda82353e..0e27b76d1e1c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.h
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.h
@@ -30,9 +30,7 @@
 #include "nouveau_bo.h"
 #include "nouveau_chan.h"
 
-int nouveau_dma_wait(struct nouveau_channel *, int slots, int size);
-void nv50_dma_push(struct nouveau_channel *, u64 addr, u32 length,
-		   bool no_prefetch);
+int nouveau_dma_wait(struct nouveau_channel *, int size);
 
 /*
  * There's a hw race condition where you can't jump to your PUT offset,
@@ -67,7 +65,7 @@ RING_SPACE(struct nouveau_channel *chan, int size)
 {
 	int ret;
 
-	ret = nouveau_dma_wait(chan, 1, size);
+	ret = nouveau_dma_wait(chan, size);
 	if (ret)
 		return ret;
 
@@ -94,12 +92,7 @@ FIRE_RING(struct nouveau_channel *chan)
 		return;
 	chan->accel_done = true;
 
-	if (chan->dma.ib_max) {
-		nv50_dma_push(chan, chan->push.addr + (chan->dma.put << 2),
-			      (chan->dma.cur - chan->dma.put) << 2, false);
-	} else {
-		WRITE_PUT(chan->dma.cur);
-	}
+	WRITE_PUT(chan->dma.cur);
 
 	chan->dma.put = chan->dma.cur;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c
index 61d0f411ef84..ca4932a150e3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
@@ -256,20 +256,15 @@ nouveau_dmem_chunk_alloc(struct nouveau_drm *drm, struct page **ppage)
 	chunk->pagemap.ops = &nouveau_dmem_pagemap_ops;
 	chunk->pagemap.owner = drm->dev;
 
-	ret = nouveau_bo_new(&drm->client, DMEM_CHUNK_SIZE, 0,
-			     NOUVEAU_GEM_DOMAIN_VRAM, 0, 0, NULL, NULL,
-			     &chunk->bo);
+	ret = nouveau_bo_new_pin(&drm->client, NOUVEAU_GEM_DOMAIN_VRAM, DMEM_CHUNK_SIZE,
+				 &chunk->bo);
 	if (ret)
 		goto out_release;
 
-	ret = nouveau_bo_pin(chunk->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
-	if (ret)
-		goto out_bo_free;
-
 	ptr = memremap_pages(&chunk->pagemap, numa_node_id());
 	if (IS_ERR(ptr)) {
 		ret = PTR_ERR(ptr);
-		goto out_bo_unpin;
+		goto out_bo_free;
 	}
 
 	mutex_lock(&drm->dmem->mutex);
@@ -292,10 +287,8 @@ nouveau_dmem_chunk_alloc(struct nouveau_drm *drm, struct page **ppage)
 
 	return 0;
 
-out_bo_unpin:
-	nouveau_bo_unpin(chunk->bo);
 out_bo_free:
-	nouveau_bo_fini(chunk->bo);
+	nouveau_bo_unpin_del(&chunk->bo);
 out_release:
 	release_mem_region(chunk->pagemap.range.start, range_len(&chunk->pagemap.range));
 out_free:
@@ -426,8 +419,7 @@ nouveau_dmem_fini(struct nouveau_drm *drm)
 
 	list_for_each_entry_safe(chunk, tmp, &drm->dmem->chunks, list) {
 		nouveau_dmem_evict_chunk(chunk);
-		nouveau_bo_unpin(chunk->bo);
-		nouveau_bo_fini(chunk->bo);
+		nouveau_bo_unpin_del(&chunk->bo);
 		WARN_ON(chunk->callocated);
 		list_del(&chunk->list);
 		memunmap_pages(&chunk->pagemap);
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index e154d08857c5..0c82a63cd49d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -503,11 +503,16 @@ nouveau_accel_init(struct nouveau_drm *drm)
 		case KEPLER_CHANNEL_GPFIFO_B:
 		case MAXWELL_CHANNEL_GPFIFO_A:
 		case PASCAL_CHANNEL_GPFIFO_A:
+			ret = nvc0_fence_create(drm);
+			break;
 		case VOLTA_CHANNEL_GPFIFO_A:
 		case TURING_CHANNEL_GPFIFO_A:
 		case AMPERE_CHANNEL_GPFIFO_A:
 		case AMPERE_CHANNEL_GPFIFO_B:
-			ret = nvc0_fence_create(drm);
+		case HOPPER_CHANNEL_GPFIFO_A:
+		case BLACKWELL_CHANNEL_GPFIFO_A:
+		case BLACKWELL_CHANNEL_GPFIFO_B:
+			ret = gv100_fence_create(drm);
 			break;
 		default:
 			break;
@@ -1079,6 +1084,10 @@ nouveau_pmops_freeze(struct device *dev)
 {
 	struct nouveau_drm *drm = dev_get_drvdata(dev);
 
+	if (drm->dev->switch_power_state == DRM_SWITCH_POWER_OFF ||
+	    drm->dev->switch_power_state == DRM_SWITCH_POWER_DYNAMIC_OFF)
+		return 0;
+
 	return nouveau_do_suspend(drm, false);
 }
 
@@ -1087,6 +1096,10 @@ nouveau_pmops_thaw(struct device *dev)
 {
 	struct nouveau_drm *drm = dev_get_drvdata(dev);
 
+	if (drm->dev->switch_power_state == DRM_SWITCH_POWER_OFF ||
+	    drm->dev->switch_power_state == DRM_SWITCH_POWER_DYNAMIC_OFF)
+		return 0;
+
 	return nouveau_do_resume(drm, false);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_exec.c b/drivers/gpu/drm/nouveau/nouveau_exec.c
index a0b5f1b16e8b..41b7c608c905 100644
--- a/drivers/gpu/drm/nouveau/nouveau_exec.c
+++ b/drivers/gpu/drm/nouveau/nouveau_exec.c
@@ -10,6 +10,8 @@
 #include "nouveau_sched.h"
 #include "nouveau_uvmm.h"
 
+#include <nvif/class.h>
+
 /**
  * DOC: Overview
  *
@@ -131,7 +133,7 @@ nouveau_exec_job_run(struct nouveau_job *job)
 	struct nouveau_fence *fence = exec_job->fence;
 	int i, ret;
 
-	ret = nouveau_dma_wait(chan, exec_job->push.count + 1, 16);
+	ret = nvif_chan_gpfifo_wait(&chan->chan, exec_job->push.count + 1, 16);
 	if (ret) {
 		NV_PRINTK(err, job->cli, "nv50cal_space: %d\n", ret);
 		return ERR_PTR(ret);
@@ -141,9 +143,11 @@ nouveau_exec_job_run(struct nouveau_job *job)
 		struct drm_nouveau_exec_push *p = &exec_job->push.s[i];
 		bool no_prefetch = p->flags & DRM_NOUVEAU_EXEC_PUSH_NO_PREFETCH;
 
-		nv50_dma_push(chan, p->va, p->va_len, no_prefetch);
+		nvif_chan_gpfifo_push(&chan->chan, p->va, p->va_len, no_prefetch);
 	}
 
+	nvif_chan_gpfifo_post(&chan->chan);
+
 	ret = nouveau_fence_emit(fence);
 	if (ret) {
 		nouveau_fence_unref(&exec_job->fence);
@@ -375,10 +379,10 @@ nouveau_exec_ioctl_exec(struct drm_device *dev,
 	if (unlikely(atomic_read(&chan->killed)))
 		return nouveau_abi16_put(abi16, -ENODEV);
 
-	if (!chan->dma.ib_max)
+	if (chan->user.oclass < NV50_CHANNEL_GPFIFO)
 		return nouveau_abi16_put(abi16, -ENOSYS);
 
-	push_max = nouveau_exec_push_max_from_ib_max(chan->dma.ib_max);
+	push_max = nouveau_exec_push_max_from_ib_max(chan->chan.gpfifo.max);
 	if (unlikely(req->push_count > push_max)) {
 		NV_PRINTK(err, cli, "pushbuf push count exceeds limit: %d max %d\n",
 			  req->push_count, push_max);
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 7cc84472cece..6ded8c2b6d3b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -90,7 +90,7 @@ nouveau_fence_context_kill(struct nouveau_fence_chan *fctx, int error)
 	while (!list_empty(&fctx->pending)) {
 		fence = list_entry(fctx->pending.next, typeof(*fence), head);
 
-		if (error)
+		if (error && !dma_fence_is_signaled_locked(&fence->base))
 			dma_fence_set_error(&fence->base, error);
 
 		if (nouveau_fence_signal(fence))
@@ -184,10 +184,10 @@ nouveau_fence_context_new(struct nouveau_channel *chan, struct nouveau_fence_cha
 	struct nouveau_cli *cli = chan->cli;
 	struct nouveau_drm *drm = cli->drm;
 	struct nouveau_fence_priv *priv = (void*)drm->fence;
-	struct {
-		struct nvif_event_v0 base;
-		struct nvif_chan_event_v0 host;
-	} args;
+	DEFINE_RAW_FLEX(struct nvif_event_v0, args, data,
+			sizeof(struct nvif_chan_event_v0));
+	struct nvif_chan_event_v0 *host =
+				(struct nvif_chan_event_v0 *)args->data;
 	int ret;
 
 	INIT_WORK(&fctx->uevent_work, nouveau_fence_uevent_work);
@@ -207,12 +207,12 @@ nouveau_fence_context_new(struct nouveau_channel *chan, struct nouveau_fence_cha
 	if (!priv->uevent)
 		return;
 
-	args.host.version = 0;
-	args.host.type = NVIF_CHAN_EVENT_V0_NON_STALL_INTR;
+	host->version = 0;
+	host->type = NVIF_CHAN_EVENT_V0_NON_STALL_INTR;
 
 	ret = nvif_event_ctor(&chan->user, "fenceNonStallIntr", (chan->runlist << 16) | chan->chid,
 			      nouveau_fence_wait_uevent_handler, false,
-			      &args.base, sizeof(args), &fctx->event);
+			      args, __struct_size(args), &fctx->event);
 
 	WARN_ON(ret);
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h
index 8bc065acfe35..6a983dd9f7b9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.h
@@ -83,6 +83,7 @@ void nv17_fence_resume(struct nouveau_drm *drm);
 int nv50_fence_create(struct nouveau_drm *);
 int nv84_fence_create(struct nouveau_drm *);
 int nvc0_fence_create(struct nouveau_drm *);
+int gv100_fence_create(struct nouveau_drm *);
 
 struct nv84_fence_chan {
 	struct nouveau_fence_chan base;
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 9ae2cee1c7c5..690e10fbf0bd 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -87,9 +87,6 @@ nouveau_gem_object_del(struct drm_gem_object *gem)
 		return;
 	}
 
-	if (gem->import_attach)
-		drm_prime_gem_destroy(gem, nvbo->bo.sg);
-
 	ttm_bo_put(&nvbo->bo);
 
 	pm_runtime_mark_last_busy(dev);
@@ -853,8 +850,8 @@ revalidate:
 		}
 	}
 
-	if (chan->dma.ib_max) {
-		ret = nouveau_dma_wait(chan, req->nr_push + 1, 16);
+	if (chan->user.oclass >= NV50_CHANNEL_GPFIFO) {
+		ret = nvif_chan_gpfifo_wait(&chan->chan, req->nr_push + 1, 16);
 		if (ret) {
 			NV_PRINTK(err, cli, "nv50cal_space: %d\n", ret);
 			goto out;
@@ -867,8 +864,10 @@ revalidate:
 			u32 length = push[i].length & ~NOUVEAU_GEM_PUSHBUF_NO_PREFETCH;
 			bool no_prefetch = push[i].length & NOUVEAU_GEM_PUSHBUF_NO_PREFETCH;
 
-			nv50_dma_push(chan, addr, length, no_prefetch);
+			nvif_chan_gpfifo_push(&chan->chan, addr, length, no_prefetch);
 		}
+
+		nvif_chan_gpfifo_post(&chan->chan);
 	} else
 	if (drm->client.device.info.chipset >= 0x25) {
 		ret = PUSH_WAIT(&chan->chan.push, req->nr_push * 2);
@@ -961,7 +960,7 @@ out_prevalid:
 	u_free(push);
 
 out_next:
-	if (chan->dma.ib_max) {
+	if (chan->user.oclass >= NV50_CHANNEL_GPFIFO) {
 		req->suffix0 = 0x00000000;
 		req->suffix1 = 0x00000000;
 	} else
diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c
index e12e2596ed84..6fa387da0637 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -720,10 +720,7 @@ nouveau_svm_fault(struct work_struct *work)
 	struct nouveau_svm *svm = container_of(buffer, typeof(*svm), buffer[buffer->id]);
 	struct nvif_object *device = &svm->drm->client.device.object;
 	struct nouveau_svmm *svmm;
-	struct {
-		struct nouveau_pfnmap_args i;
-		u64 phys[1];
-	} args;
+	DEFINE_RAW_FLEX(struct nouveau_pfnmap_args, args, p.phys, 1);
 	unsigned long hmm_flags;
 	u64 inst, start, limit;
 	int fi, fn;
@@ -772,11 +769,11 @@ nouveau_svm_fault(struct work_struct *work)
 	mutex_unlock(&svm->mutex);
 
 	/* Process list of faults. */
-	args.i.i.version = 0;
-	args.i.i.type = NVIF_IOCTL_V0_MTHD;
-	args.i.m.version = 0;
-	args.i.m.method = NVIF_VMM_V0_PFNMAP;
-	args.i.p.version = 0;
+	args->i.version = 0;
+	args->i.type = NVIF_IOCTL_V0_MTHD;
+	args->m.version = 0;
+	args->m.method = NVIF_VMM_V0_PFNMAP;
+	args->p.version = 0;
 
 	for (fi = 0; fn = fi + 1, fi < buffer->fault_nr; fi = fn) {
 		struct svm_notifier notifier;
@@ -802,9 +799,9 @@ nouveau_svm_fault(struct work_struct *work)
 		 * fault window, determining required pages and access
 		 * permissions based on pending faults.
 		 */
-		args.i.p.addr = start;
-		args.i.p.page = PAGE_SHIFT;
-		args.i.p.size = PAGE_SIZE;
+		args->p.addr = start;
+		args->p.page = PAGE_SHIFT;
+		args->p.size = PAGE_SIZE;
 		/*
 		 * Determine required permissions based on GPU fault
 		 * access flags.
@@ -832,16 +829,16 @@ nouveau_svm_fault(struct work_struct *work)
 
 		notifier.svmm = svmm;
 		if (atomic)
-			ret = nouveau_atomic_range_fault(svmm, svm->drm,
-							 &args.i, sizeof(args),
+			ret = nouveau_atomic_range_fault(svmm, svm->drm, args,
+							 __struct_size(args),
 							 &notifier);
 		else
-			ret = nouveau_range_fault(svmm, svm->drm, &args.i,
-						  sizeof(args), hmm_flags,
-						  &notifier);
+			ret = nouveau_range_fault(svmm, svm->drm, args,
+						  __struct_size(args),
+						  hmm_flags, &notifier);
 		mmput(mm);
 
-		limit = args.i.p.addr + args.i.p.size;
+		limit = args->p.addr + args->p.size;
 		for (fn = fi; ++fn < buffer->fault_nr; ) {
 			/* It's okay to skip over duplicate addresses from the
 			 * same SVMM as faults are ordered by access type such
@@ -855,14 +852,14 @@ nouveau_svm_fault(struct work_struct *work)
 			if (buffer->fault[fn]->svmm != svmm ||
 			    buffer->fault[fn]->addr >= limit ||
 			    (buffer->fault[fi]->access == FAULT_ACCESS_READ &&
-			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_V)) ||
+			     !(args->p.phys[0] & NVIF_VMM_PFNMAP_V0_V)) ||
 			    (buffer->fault[fi]->access != FAULT_ACCESS_READ &&
 			     buffer->fault[fi]->access != FAULT_ACCESS_PREFETCH &&
-			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_W)) ||
+			     !(args->p.phys[0] & NVIF_VMM_PFNMAP_V0_W)) ||
 			    (buffer->fault[fi]->access != FAULT_ACCESS_READ &&
 			     buffer->fault[fi]->access != FAULT_ACCESS_WRITE &&
 			     buffer->fault[fi]->access != FAULT_ACCESS_PREFETCH &&
-			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_A)))
+			     !(args->p.phys[0] & NVIF_VMM_PFNMAP_V0_A)))
 				break;
 		}
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index e244927eb5d4..7d2436e5d50d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -312,8 +312,8 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 	/* VRAM init */
 	drm->gem.vram_available = drm->client.device.info.ram_user;
 
-	arch_io_reserve_memtype_wc(device->func->resource_addr(device, 1),
-				   device->func->resource_size(device, 1));
+	arch_io_reserve_memtype_wc(device->func->resource_addr(device, NVKM_BAR1_FB),
+				   device->func->resource_size(device, NVKM_BAR1_FB));
 
 	ret = nouveau_ttm_init_vram(drm);
 	if (ret) {
@@ -321,8 +321,8 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 		return ret;
 	}
 
-	drm->ttm.mtrr = arch_phys_wc_add(device->func->resource_addr(device, 1),
-					 device->func->resource_size(device, 1));
+	drm->ttm.mtrr = arch_phys_wc_add(device->func->resource_addr(device, NVKM_BAR1_FB),
+					 device->func->resource_size(device, NVKM_BAR1_FB));
 
 	/* GART init */
 	if (!drm->agp.bridge) {
@@ -357,7 +357,7 @@ nouveau_ttm_fini(struct nouveau_drm *drm)
 
 	arch_phys_wc_del(drm->ttm.mtrr);
 	drm->ttm.mtrr = 0;
-	arch_io_free_memtype_wc(device->func->resource_addr(device, 1),
-				device->func->resource_size(device, 1));
+	arch_io_free_memtype_wc(device->func->resource_addr(device, NVKM_BAR1_FB),
+				device->func->resource_size(device, NVKM_BAR1_FB));
 
 }
diff --git a/drivers/gpu/drm/nouveau/nv10_fence.c b/drivers/gpu/drm/nouveau/nv10_fence.c
index 8c73f40e3bda..40ee95340814 100644
--- a/drivers/gpu/drm/nouveau/nv10_fence.c
+++ b/drivers/gpu/drm/nouveau/nv10_fence.c
@@ -85,10 +85,8 @@ void
 nv10_fence_destroy(struct nouveau_drm *drm)
 {
 	struct nv10_fence_priv *priv = drm->fence;
-	nouveau_bo_unmap(priv->bo);
-	if (priv->bo)
-		nouveau_bo_unpin(priv->bo);
-	nouveau_bo_fini(priv->bo);
+
+	nouveau_bo_unpin_del(&priv->bo);
 	drm->fence = NULL;
 	kfree(priv);
 }
diff --git a/drivers/gpu/drm/nouveau/nv17_fence.c b/drivers/gpu/drm/nouveau/nv17_fence.c
index d09bfd11369f..1b0c0aa3c305 100644
--- a/drivers/gpu/drm/nouveau/nv17_fence.c
+++ b/drivers/gpu/drm/nouveau/nv17_fence.c
@@ -130,20 +130,7 @@ nv17_fence_create(struct nouveau_drm *drm)
 	priv->base.context_del = nv10_fence_context_del;
 	spin_lock_init(&priv->lock);
 
-	ret = nouveau_bo_new(&drm->client, 4096, 0x1000,
-			     NOUVEAU_GEM_DOMAIN_VRAM,
-			     0, 0x0000, NULL, NULL, &priv->bo);
-	if (!ret) {
-		ret = nouveau_bo_pin(priv->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
-		if (!ret) {
-			ret = nouveau_bo_map(priv->bo);
-			if (ret)
-				nouveau_bo_unpin(priv->bo);
-		}
-		if (ret)
-			nouveau_bo_fini(priv->bo);
-	}
-
+	ret = nouveau_bo_new_map(&drm->client, NOUVEAU_GEM_DOMAIN_VRAM, PAGE_SIZE, &priv->bo);
 	if (ret) {
 		nv10_fence_destroy(drm);
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/nv50_fence.c b/drivers/gpu/drm/nouveau/nv50_fence.c
index 62e28dddf87c..e1f0e8adf313 100644
--- a/drivers/gpu/drm/nouveau/nv50_fence.c
+++ b/drivers/gpu/drm/nouveau/nv50_fence.c
@@ -81,20 +81,7 @@ nv50_fence_create(struct nouveau_drm *drm)
 	priv->base.context_del = nv10_fence_context_del;
 	spin_lock_init(&priv->lock);
 
-	ret = nouveau_bo_new(&drm->client, 4096, 0x1000,
-			     NOUVEAU_GEM_DOMAIN_VRAM,
-			     0, 0x0000, NULL, NULL, &priv->bo);
-	if (!ret) {
-		ret = nouveau_bo_pin(priv->bo, NOUVEAU_GEM_DOMAIN_VRAM, false);
-		if (!ret) {
-			ret = nouveau_bo_map(priv->bo);
-			if (ret)
-				nouveau_bo_unpin(priv->bo);
-		}
-		if (ret)
-			nouveau_bo_fini(priv->bo);
-	}
-
+	ret = nouveau_bo_new_map(&drm->client, NOUVEAU_GEM_DOMAIN_VRAM, PAGE_SIZE, &priv->bo);
 	if (ret) {
 		nv10_fence_destroy(drm);
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/nv84_fence.c b/drivers/gpu/drm/nouveau/nv84_fence.c
index aa7dd0c5d917..1765b2cedaf9 100644
--- a/drivers/gpu/drm/nouveau/nv84_fence.c
+++ b/drivers/gpu/drm/nouveau/nv84_fence.c
@@ -185,10 +185,8 @@ static void
 nv84_fence_destroy(struct nouveau_drm *drm)
 {
 	struct nv84_fence_priv *priv = drm->fence;
-	nouveau_bo_unmap(priv->bo);
-	if (priv->bo)
-		nouveau_bo_unpin(priv->bo);
-	nouveau_bo_fini(priv->bo);
+
+	nouveau_bo_unpin_del(&priv->bo);
 	drm->fence = NULL;
 	kfree(priv);
 }
@@ -222,19 +220,8 @@ nv84_fence_create(struct nouveau_drm *drm)
 		  * will lose CPU/GPU coherency!
 		  */
 		NOUVEAU_GEM_DOMAIN_GART | NOUVEAU_GEM_DOMAIN_COHERENT;
-	ret = nouveau_bo_new(&drm->client, 16 * drm->chan_total, 0,
-			     domain, 0, 0, NULL, NULL, &priv->bo);
-	if (ret == 0) {
-		ret = nouveau_bo_pin(priv->bo, domain, false);
-		if (ret == 0) {
-			ret = nouveau_bo_map(priv->bo);
-			if (ret)
-				nouveau_bo_unpin(priv->bo);
-		}
-		if (ret)
-			nouveau_bo_fini(priv->bo);
-	}
 
+	ret = nouveau_bo_new_map(&drm->client, domain, 16 * drm->chan_total, &priv->bo);
 	if (ret)
 		nv84_fence_destroy(drm);
 	return ret;
diff --git a/drivers/gpu/drm/nouveau/nvif/Kbuild b/drivers/gpu/drm/nouveau/nvif/Kbuild
index b7963a39dd91..198889c20ce1 100644
--- a/drivers/gpu/drm/nouveau/nvif/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvif/Kbuild
@@ -14,6 +14,12 @@ nvif-y += nvif/outp.o
 nvif-y += nvif/timer.o
 nvif-y += nvif/vmm.o
 
+# Channel classes
+nvif-y += nvif/chan.o
+nvif-y += nvif/chan506f.o
+nvif-y += nvif/chan906f.o
+nvif-y += nvif/chanc36f.o
+
 # Usermode classes
 nvif-y += nvif/user.o
 nvif-y += nvif/userc361.o
diff --git a/drivers/gpu/drm/nouveau/nvif/chan.c b/drivers/gpu/drm/nouveau/nvif/chan.c
new file mode 100644
index 000000000000..baa10227d51a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvif/chan.c
@@ -0,0 +1,156 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <nvif/chan.h>
+
+static void
+nvif_chan_gpfifo_push_kick(struct nvif_push *push)
+{
+	struct nvif_chan *chan = container_of(push, typeof(*chan), push);
+	u32 put = push->bgn - (u32 *)chan->push.mem.object.map.ptr;
+	u32 cnt;
+
+	if (chan->func->gpfifo.post) {
+		if (push->end - push->cur < chan->func->gpfifo.post_size)
+			push->end = push->cur + chan->func->gpfifo.post_size;
+
+		WARN_ON(nvif_chan_gpfifo_post(chan));
+	}
+
+	cnt = push->cur - push->bgn;
+
+	chan->func->gpfifo.push(chan, true, chan->push.addr + (put << 2), cnt << 2, false);
+	chan->func->gpfifo.kick(chan);
+}
+
+static int
+nvif_chan_gpfifo_push_wait(struct nvif_push *push, u32 push_nr)
+{
+	struct nvif_chan *chan = container_of(push, typeof(*chan), push);
+
+	return nvif_chan_gpfifo_wait(chan, 1, push_nr);
+}
+
+int
+nvif_chan_gpfifo_post(struct nvif_chan *chan)
+{
+	const u32 *map = chan->push.mem.object.map.ptr;
+	const u32 pbptr = (chan->push.cur - map) + chan->func->gpfifo.post_size;
+	const u32 gpptr = (chan->gpfifo.cur + 1) & chan->gpfifo.max;
+
+	return chan->func->gpfifo.post(chan, gpptr, pbptr);
+}
+
+void
+nvif_chan_gpfifo_push(struct nvif_chan *chan, u64 addr, u32 size, bool no_prefetch)
+{
+	chan->func->gpfifo.push(chan, false, addr, size, no_prefetch);
+}
+
+int
+nvif_chan_gpfifo_wait(struct nvif_chan *chan, u32 gpfifo_nr, u32 push_nr)
+{
+	struct nvif_push *push = &chan->push;
+	int ret = 0, time = 1000000;
+
+	if (gpfifo_nr) {
+		/* Account for pushbuf space needed by nvif_chan_gpfifo_post(),
+		 * if used after pushing userspace GPFIFO entries.
+		 */
+		if (chan->func->gpfifo.post)
+			push_nr += chan->func->gpfifo.post_size;
+	}
+
+	/* Account for the GPFIFO entry needed to submit pushbuf. */
+	if (push_nr)
+		gpfifo_nr++;
+
+	/* Wait for space in main push buffer. */
+	if (push->cur + push_nr > push->end) {
+		ret = nvif_chan_dma_wait(chan, push_nr);
+		if (ret)
+			return ret;
+	}
+
+	/* Wait for GPFIFO space. */
+	while (chan->gpfifo.free < gpfifo_nr) {
+		chan->gpfifo.free = chan->func->gpfifo.read_get(chan) - chan->gpfifo.cur - 1;
+		if (chan->gpfifo.free < 0)
+			chan->gpfifo.free += chan->gpfifo.max + 1;
+
+		if (chan->gpfifo.free < gpfifo_nr) {
+			if (!time--)
+				return -ETIMEDOUT;
+			udelay(1);
+		}
+	}
+
+	return 0;
+}
+
+void
+nvif_chan_gpfifo_ctor(const struct nvif_chan_func *func, void *userd, void *gpfifo, u32 gpfifo_size,
+		      void *push, u64 push_addr, u32 push_size, struct nvif_chan *chan)
+{
+	chan->func = func;
+
+	chan->userd.map.ptr = userd;
+
+	chan->gpfifo.map.ptr = gpfifo;
+	chan->gpfifo.max = (gpfifo_size >> 3) - 1;
+	chan->gpfifo.free = chan->gpfifo.max;
+
+	chan->push.mem.object.map.ptr = push;
+	chan->push.wait = nvif_chan_gpfifo_push_wait;
+	chan->push.kick = nvif_chan_gpfifo_push_kick;
+	chan->push.addr = push_addr;
+	chan->push.hw.max = push_size >> 2;
+	chan->push.bgn = chan->push.cur = chan->push.end = push;
+}
+
+int
+nvif_chan_dma_wait(struct nvif_chan *chan, u32 nr)
+{
+	struct nvif_push *push = &chan->push;
+	u32 cur = push->cur - (u32 *)push->mem.object.map.ptr;
+	u32 free, time = 1000000;
+
+	nr += chan->func->gpfifo.post_size;
+
+	do {
+		u32 get = chan->func->push.read_get(chan);
+
+		if (get <= cur) {
+			free = push->hw.max - cur;
+			if (free >= nr)
+				break;
+
+			PUSH_KICK(push);
+
+			while (get == 0) {
+				get = chan->func->push.read_get(chan);
+				if (get == 0) {
+					if (!time--)
+						return -ETIMEDOUT;
+					udelay(1);
+				}
+			}
+
+			cur = 0;
+		}
+
+		free = get - cur - 1;
+
+		if (free < nr) {
+			if (!time--)
+				return -ETIMEDOUT;
+			udelay(1);
+		}
+	} while (free < nr);
+
+	push->bgn = (u32 *)push->mem.object.map.ptr + cur;
+	push->cur = push->bgn;
+	push->end = push->bgn + free - chan->func->gpfifo.post_size;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvif/chan506f.c b/drivers/gpu/drm/nouveau/nvif/chan506f.c
new file mode 100644
index 000000000000..d3900887c4a7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvif/chan506f.c
@@ -0,0 +1,72 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <nvif/chan.h>
+
+void
+nvif_chan506f_gpfifo_kick(struct nvif_chan *chan)
+{
+	wmb();
+	nvif_wr32(&chan->userd, 0x8c, chan->gpfifo.cur);
+}
+
+void
+nvif_chan506f_gpfifo_push(struct nvif_chan *chan, bool main, u64 addr, u32 size, bool no_prefetch)
+{
+	u32 gpptr = chan->gpfifo.cur << 3;
+
+	if (WARN_ON(!chan->gpfifo.free))
+		return;
+
+	nvif_wr32(&chan->gpfifo, gpptr + 0, lower_32_bits(addr));
+	nvif_wr32(&chan->gpfifo, gpptr + 4, upper_32_bits(addr) |
+					    (main ? 0 : BIT(9)) |
+					    (size >> 2) << 10 |
+					    (no_prefetch ? BIT(31) : 0));
+
+	chan->gpfifo.cur = (chan->gpfifo.cur + 1) & chan->gpfifo.max;
+	chan->gpfifo.free--;
+	if (!chan->gpfifo.free)
+		chan->push.end = chan->push.cur;
+}
+
+static u32
+nvif_chan506f_gpfifo_read_get(struct nvif_chan *chan)
+{
+	return nvif_rd32(&chan->userd, 0x88);
+}
+
+static u32
+nvif_chan506f_read_get(struct nvif_chan *chan)
+{
+	u32 tlgetlo = nvif_rd32(&chan->userd, 0x58);
+	u32 tlgethi = nvif_rd32(&chan->userd, 0x5c);
+	struct nvif_push *push = &chan->push;
+
+	/* Update cached GET pointer if TOP_LEVEL_GET is valid. */
+	if (tlgethi & BIT(31)) {
+		u64 tlget = ((u64)(tlgethi & 0xff) << 32) | tlgetlo;
+
+		push->hw.get = (tlget - push->addr) >> 2;
+	}
+
+	return push->hw.get;
+}
+
+static const struct nvif_chan_func
+nvif_chan506f = {
+	.push.read_get = nvif_chan506f_read_get,
+	.gpfifo.read_get = nvif_chan506f_gpfifo_read_get,
+	.gpfifo.push = nvif_chan506f_gpfifo_push,
+	.gpfifo.kick = nvif_chan506f_gpfifo_kick,
+};
+
+int
+nvif_chan506f_ctor(struct nvif_chan *chan, void *userd, void *gpfifo, u32 gpfifo_size,
+		   void *push, u64 push_addr, u32 push_size)
+{
+	nvif_chan_gpfifo_ctor(&nvif_chan506f, userd, gpfifo, gpfifo_size,
+			      push, push_addr, push_size, chan);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvif/chan906f.c b/drivers/gpu/drm/nouveau/nvif/chan906f.c
new file mode 100644
index 000000000000..c9cfb85179b0
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvif/chan906f.c
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <nvif/chan.h>
+#include <nvif/user.h>
+#include <nvif/push906f.h>
+
+#include <nvhw/class/cl906f.h>
+
+/* Limits GPFIFO size to 1MiB, and "main" push buffer size to 64KiB. */
+#define NVIF_CHAN906F_PBPTR_BITS  15
+#define NVIF_CHAN906F_PBPTR_MASK  ((1 << NVIF_CHAN906F_PBPTR_BITS) - 1)
+
+#define NVIF_CHAN906F_GPPTR_SHIFT NVIF_CHAN906F_PBPTR_BITS
+#define NVIF_CHAN906F_GPPTR_BITS  (32 - NVIF_CHAN906F_PBPTR_BITS)
+#define NVIF_CHAN906F_GPPTR_MASK  ((1 << NVIF_CHAN906F_GPPTR_BITS) - 1)
+
+#define NVIF_CHAN906F_SEM_RELEASE_SIZE 5
+
+static int
+nvif_chan906f_sem_release(struct nvif_chan *chan, u64 addr, u32 data)
+{
+	struct nvif_push *push = &chan->push;
+	int ret;
+
+	ret = PUSH_WAIT(push, NVIF_CHAN906F_SEM_RELEASE_SIZE);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NV906F, SEMAPHOREA,
+		  NVVAL(NV906F, SEMAPHOREA, OFFSET_UPPER, upper_32_bits(addr)),
+
+				SEMAPHOREB, lower_32_bits(addr),
+
+				SEMAPHOREC, data,
+
+				SEMAPHORED,
+		  NVDEF(NV906F, SEMAPHORED, OPERATION, RELEASE) |
+		  NVDEF(NV906F, SEMAPHORED, RELEASE_WFI, DIS) |
+		  NVDEF(NV906F, SEMAPHORED, RELEASE_SIZE, 16BYTE));
+
+	return 0;
+}
+
+int
+nvif_chan906f_gpfifo_post(struct nvif_chan *chan, u32 gpptr, u32 pbptr)
+{
+	return chan->func->sem.release(chan, chan->sema.addr,
+				       (gpptr << NVIF_CHAN906F_GPPTR_SHIFT) | pbptr);
+}
+
+u32
+nvif_chan906f_gpfifo_read_get(struct nvif_chan *chan)
+{
+	return nvif_rd32(&chan->sema, 0) >> NVIF_CHAN906F_GPPTR_SHIFT;
+}
+
+u32
+nvif_chan906f_read_get(struct nvif_chan *chan)
+{
+	return nvif_rd32(&chan->sema, 0) & NVIF_CHAN906F_PBPTR_MASK;
+}
+
+static const struct nvif_chan_func
+nvif_chan906f = {
+	.push.read_get = nvif_chan906f_read_get,
+	.gpfifo.read_get = nvif_chan906f_gpfifo_read_get,
+	.gpfifo.push = nvif_chan506f_gpfifo_push,
+	.gpfifo.kick = nvif_chan506f_gpfifo_kick,
+	.gpfifo.post = nvif_chan906f_gpfifo_post,
+	.gpfifo.post_size = NVIF_CHAN906F_SEM_RELEASE_SIZE,
+	.sem.release = nvif_chan906f_sem_release,
+};
+
+int
+nvif_chan906f_ctor_(const struct nvif_chan_func *func, void *userd, void *gpfifo, u32 gpfifo_size,
+		    void *push, u64 push_addr, u32 push_size, void *sema, u64 sema_addr,
+		    struct nvif_chan *chan)
+{
+	nvif_chan_gpfifo_ctor(func, userd, gpfifo, gpfifo_size, push, push_addr, push_size, chan);
+	chan->sema.map.ptr = sema;
+	chan->sema.addr = sema_addr;
+	return 0;
+}
+
+int
+nvif_chan906f_ctor(struct nvif_chan *chan, void *userd, void *gpfifo, u32 gpfifo_size,
+		   void *push, u64 push_addr, u32 push_size, void *sema, u64 sema_addr)
+{
+	return nvif_chan906f_ctor_(&nvif_chan906f, userd, gpfifo, gpfifo_size,
+				   push, push_addr, push_size, sema, sema_addr, chan);
+}
diff --git a/drivers/gpu/drm/nouveau/nvif/chanc36f.c b/drivers/gpu/drm/nouveau/nvif/chanc36f.c
new file mode 100644
index 000000000000..ca02b939c3fd
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvif/chanc36f.c
@@ -0,0 +1,77 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <nvif/chan.h>
+#include <nvif/user.h>
+
+#include <nvif/push906f.h>
+#include <nvhw/class/clc36f.h>
+
+static void
+nvif_chanc36f_gpfifo_kick(struct nvif_chan *chan)
+{
+	struct nvif_user *usermode = chan->usermode;
+
+	nvif_wr32(&chan->userd, 0x8c, chan->gpfifo.cur);
+
+	wmb(); /* ensure CPU writes are flushed to BAR1 */
+	nvif_rd32(&chan->userd, 0); /* ensure BAR1 writes are flushed to vidmem */
+
+	usermode->func->doorbell(usermode, chan->doorbell_token);
+}
+
+#define NVIF_CHANC36F_SEM_RELEASE_SIZE 6
+
+static int
+nvif_chanc36f_sem_release(struct nvif_chan *chan, u64 addr, u32 data)
+{
+	struct nvif_push *push = &chan->push;
+	int ret;
+
+	ret = PUSH_WAIT(push, NVIF_CHANC36F_SEM_RELEASE_SIZE);
+	if (ret)
+		return ret;
+
+	PUSH_MTHD(push, NVC36F, SEM_ADDR_LO, lower_32_bits(addr),
+
+				SEM_ADDR_HI, upper_32_bits(addr),
+
+				SEM_PAYLOAD_LO, data);
+
+	PUSH_MTHD(push, NVC36F, SEM_EXECUTE,
+		  NVDEF(NVC36F, SEM_EXECUTE, OPERATION, RELEASE) |
+		  NVDEF(NVC36F, SEM_EXECUTE, RELEASE_WFI, DIS) |
+		  NVDEF(NVC36F, SEM_EXECUTE, PAYLOAD_SIZE, 32BIT) |
+		  NVDEF(NVC36F, SEM_EXECUTE, RELEASE_TIMESTAMP, DIS));
+
+	return 0;
+}
+
+static const struct nvif_chan_func
+nvif_chanc36f = {
+	.push.read_get = nvif_chan906f_read_get,
+	.gpfifo.read_get = nvif_chan906f_gpfifo_read_get,
+	.gpfifo.push = nvif_chan506f_gpfifo_push,
+	.gpfifo.kick = nvif_chanc36f_gpfifo_kick,
+	.gpfifo.post = nvif_chan906f_gpfifo_post,
+	.gpfifo.post_size = NVIF_CHANC36F_SEM_RELEASE_SIZE,
+	.sem.release = nvif_chanc36f_sem_release,
+};
+
+int
+nvif_chanc36f_ctor(struct nvif_chan *chan, void *userd, void *gpfifo, u32 gpfifo_size,
+		   void *push, u64 push_addr, u32 push_size, void *sema, u64 sema_addr,
+		   struct nvif_user *usermode, u32 doorbell_token)
+{
+	int ret;
+
+	ret = nvif_chan906f_ctor_(&nvif_chanc36f, userd, gpfifo, gpfifo_size,
+				  push, push_addr, push_size, sema, sema_addr, chan);
+	if (ret)
+		return ret;
+
+	chan->usermode = usermode;
+	chan->doorbell_token = doorbell_token;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvif/conn.c b/drivers/gpu/drm/nouveau/nvif/conn.c
index 9ee18cb99264..5a1a83c62a2a 100644
--- a/drivers/gpu/drm/nouveau/nvif/conn.c
+++ b/drivers/gpu/drm/nouveau/nvif/conn.c
@@ -30,17 +30,17 @@ int
 nvif_conn_event_ctor(struct nvif_conn *conn, const char *name, nvif_event_func func, u8 types,
 		     struct nvif_event *event)
 {
-	struct {
-		struct nvif_event_v0 base;
-		struct nvif_conn_event_v0 conn;
-	} args;
+	DEFINE_RAW_FLEX(struct nvif_event_v0, args, data,
+			sizeof(struct nvif_conn_event_v0));
+	struct nvif_conn_event_v0 *args_conn =
+				(struct nvif_conn_event_v0 *)args->data;
 	int ret;
 
-	args.conn.version = 0;
-	args.conn.types = types;
+	args_conn->version = 0;
+	args_conn->types = types;
 
 	ret = nvif_event_ctor_(&conn->object, name ?: "nvifConnHpd", nvif_conn_id(conn),
-			       func, true, &args.base, sizeof(args), false, event);
+			       func, true, args, __struct_size(args), false, event);
 	NVIF_DEBUG(&conn->object, "[NEW EVENT:HPD types:%02x]", types);
 	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nvif/disp.c b/drivers/gpu/drm/nouveau/nvif/disp.c
index 14da22fa3b5b..fa42146252da 100644
--- a/drivers/gpu/drm/nouveau/nvif/disp.c
+++ b/drivers/gpu/drm/nouveau/nvif/disp.c
@@ -36,6 +36,7 @@ int
 nvif_disp_ctor(struct nvif_device *device, const char *name, s32 oclass, struct nvif_disp *disp)
 {
 	static const struct nvif_mclass disps[] = {
+		{ GB202_DISP, 0 },
 		{ AD102_DISP, 0 },
 		{ GA102_DISP, 0 },
 		{ TU102_DISP, 0 },
diff --git a/drivers/gpu/drm/nouveau/nvif/outp.c b/drivers/gpu/drm/nouveau/nvif/outp.c
index 6daeb7f0b09b..32f6c5eb92af 100644
--- a/drivers/gpu/drm/nouveau/nvif/outp.c
+++ b/drivers/gpu/drm/nouveau/nvif/outp.c
@@ -195,20 +195,17 @@ nvif_outp_dp_aux_pwr(struct nvif_outp *outp, bool enable)
 int
 nvif_outp_hda_eld(struct nvif_outp *outp, int head, void *data, u32 size)
 {
-	struct {
-		struct nvif_outp_hda_eld_v0 mthd;
-		u8 data[128];
-	} args;
+	DEFINE_RAW_FLEX(struct nvif_outp_hda_eld_v0, mthd, data, 128);
 	int ret;
 
-	if (WARN_ON(size > ARRAY_SIZE(args.data)))
+	if (WARN_ON(size > __member_size(mthd->data)))
 		return -EINVAL;
 
-	args.mthd.version = 0;
-	args.mthd.head = head;
+	mthd->version = 0;
+	mthd->head = head;
 
-	memcpy(args.data, data, size);
-	ret = nvif_mthd(&outp->object, NVIF_OUTP_V0_HDA_ELD, &args, sizeof(args.mthd) + size);
+	memcpy(mthd->data, data, size);
+	ret = nvif_mthd(&outp->object, NVIF_OUTP_V0_HDA_ELD, mthd, sizeof(*mthd) + size);
 	NVIF_ERRON(ret, &outp->object, "[HDA_ELD head:%d size:%d]", head, size);
 	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nvif/user.c b/drivers/gpu/drm/nouveau/nvif/user.c
index b648a5e036af..53f03fa1c9c2 100644
--- a/drivers/gpu/drm/nouveau/nvif/user.c
+++ b/drivers/gpu/drm/nouveau/nvif/user.c
@@ -41,9 +41,11 @@ nvif_user_ctor(struct nvif_device *device, const char *name)
 		int version;
 		const struct nvif_user_func *func;
 	} users[] = {
-		{ AMPERE_USERMODE_A, -1, &nvif_userc361 },
-		{ TURING_USERMODE_A, -1, &nvif_userc361 },
-		{  VOLTA_USERMODE_A, -1, &nvif_userc361 },
+		{ BLACKWELL_USERMODE_A, -1, &nvif_userc361 },
+		{    HOPPER_USERMODE_A, -1, &nvif_userc361 },
+		{    AMPERE_USERMODE_A, -1, &nvif_userc361 },
+		{    TURING_USERMODE_A, -1, &nvif_userc361 },
+		{     VOLTA_USERMODE_A, -1, &nvif_userc361 },
 		{}
 	};
 	int cid, ret;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild
index 2e48b0816670..ddcf8782d6b6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild
@@ -17,8 +17,6 @@ include $(src)/nvkm/engine/msppp/Kbuild
 include $(src)/nvkm/engine/msvld/Kbuild
 include $(src)/nvkm/engine/nvenc/Kbuild
 include $(src)/nvkm/engine/nvdec/Kbuild
-include $(src)/nvkm/engine/nvjpg/Kbuild
-include $(src)/nvkm/engine/ofa/Kbuild
 include $(src)/nvkm/engine/sec/Kbuild
 include $(src)/nvkm/engine/sec2/Kbuild
 include $(src)/nvkm/engine/sw/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
index 165d61fc5d6c..9754bac65df7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
@@ -10,5 +10,4 @@ nvkm-y += nvkm/engine/ce/gv100.o
 nvkm-y += nvkm/engine/ce/tu102.o
 nvkm-y += nvkm/engine/ce/ga100.o
 nvkm-y += nvkm/engine/ce/ga102.o
-
-nvkm-y += nvkm/engine/ce/r535.o
+nvkm-y += nvkm/engine/ce/gb202.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga100.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga100.c
index 9427a592bd16..1c0c60138706 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga100.c
@@ -90,7 +90,7 @@ ga100_ce_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 	     struct nvkm_engine **pengine)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_ce_new(&ga100_ce, device, type, inst, pengine);
+		return -ENODEV;
 
 	return nvkm_engine_new_(&ga100_ce, device, type, inst, true, pengine);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga102.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga102.c
index ce56ede7c2e9..9359c5e7aa3a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/ga102.c
@@ -44,7 +44,7 @@ ga102_ce_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 	     struct nvkm_engine **pengine)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_ce_new(&ga102_ce, device, type, inst, pengine);
+		return -ENODEV;
 
 	return nvkm_engine_new_(&ga102_ce, device, type, inst, true, pengine);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/gb202.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gb202.c
new file mode 100644
index 000000000000..37c3c619c71b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gb202.c
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gb202/dev_ce.h>
+
+u32
+gb202_ce_grce_mask(struct nvkm_device *device)
+{
+	u32 data = nvkm_rd32(device, NV_CE_GRCE_MASK);
+
+	return NVVAL_GET(data, NV_CE, GRCE_MASK, VALUE);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/ce/priv.h
index 806a76a72249..34fd2657134b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/priv.h
@@ -16,4 +16,6 @@ int ga100_ce_oneinit(struct nvkm_engine *);
 int ga100_ce_init(struct nvkm_engine *);
 int ga100_ce_fini(struct nvkm_engine *, bool);
 int ga100_ce_nonstall(struct nvkm_engine *);
+
+u32 gb202_ce_grce_mask(struct nvkm_device *);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/r535.c
deleted file mode 100644
index bd0d435dbbd3..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/r535.c
+++ /dev/null
@@ -1,108 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <core/object.h>
-#include <subdev/gsp.h>
-#include <engine/fifo.h>
-
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080_notification.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/clc0b5sw.h>
-
-struct r535_ce_obj {
-	struct nvkm_object object;
-	struct nvkm_gsp_object rm;
-};
-
-static void *
-r535_ce_obj_dtor(struct nvkm_object *object)
-{
-	struct r535_ce_obj *obj = container_of(object, typeof(*obj), object);
-
-	nvkm_gsp_rm_free(&obj->rm);
-	return obj;
-}
-
-static const struct nvkm_object_func
-r535_ce_obj = {
-	.dtor = r535_ce_obj_dtor,
-};
-
-static int
-r535_ce_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
-		 struct nvkm_object **pobject)
-{
-	struct nvkm_chan *chan = nvkm_uchan_chan(oclass->parent);
-	struct r535_ce_obj *obj;
-	NVC0B5_ALLOCATION_PARAMETERS *args;
-
-	if (!(obj = kzalloc(sizeof(*obj), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_ce_obj, oclass, &obj->object);
-	*pobject = &obj->object;
-
-	args = nvkm_gsp_rm_alloc_get(&chan->rm.object, oclass->handle, oclass->base.oclass,
-				     sizeof(*args), &obj->rm);
-	if (WARN_ON(IS_ERR(args)))
-		return PTR_ERR(args);
-
-	args->version = 1;
-	args->engineType = NV2080_ENGINE_TYPE_COPY0 + oclass->engine->subdev.inst;
-
-	return nvkm_gsp_rm_alloc_wr(&obj->rm, args);
-}
-
-static void *
-r535_ce_dtor(struct nvkm_engine *engine)
-{
-	kfree(engine->func);
-	return engine;
-}
-
-int
-r535_ce_new(const struct nvkm_engine_func *hw, struct nvkm_device *device,
-	    enum nvkm_subdev_type type, int inst, struct nvkm_engine **pengine)
-{
-	struct nvkm_engine_func *rm;
-	int nclass, ret;
-
-	for (nclass = 0; hw->sclass[nclass].oclass; nclass++);
-
-	if (!(rm = kzalloc(sizeof(*rm) + (nclass + 1) * sizeof(rm->sclass[0]), GFP_KERNEL)))
-		return -ENOMEM;
-
-	rm->dtor = r535_ce_dtor;
-	for (int i = 0; i < nclass; i++) {
-		rm->sclass[i].minver = hw->sclass[i].minver;
-		rm->sclass[i].maxver = hw->sclass[i].maxver;
-		rm->sclass[i].oclass = hw->sclass[i].oclass;
-		rm->sclass[i].ctor = r535_ce_obj_ctor;
-	}
-
-	ret = nvkm_engine_new_(rm, device, type, inst, true, pengine);
-	if (ret)
-		kfree(rm);
-
-	return ret;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu102.c
index 7c8647dcb349..67d0545cf902 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu102.c
@@ -40,7 +40,7 @@ tu102_ce_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 	     struct nvkm_engine **pengine)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_ce_new(&tu102_ce, device, type, inst, pengine);
+		return -ENODEV;
 
 	return nvkm_engine_new_(&tu102_ce, device, type, inst, true, pengine);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 9093d89b16f3..3375a59ebf1a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -2529,9 +2529,6 @@ nv170_chipset = {
 	.vfn      = { 0x00000001, ga100_vfn_new },
 	.ce       = { 0x000003ff, ga100_ce_new },
 	.fifo     = { 0x00000001, ga100_fifo_new },
-	.nvdec    = { 0x0000001f, ga100_nvdec_new },
-	.nvjpg    = { 0x00000001, ga100_nvjpg_new },
-	.ofa      = { 0x00000001, ga100_ofa_new },
 };
 
 static const struct nvkm_device_chip
@@ -2561,8 +2558,6 @@ nv172_chipset = {
 	.fifo     = { 0x00000001, ga102_fifo_new },
 	.gr       = { 0x00000001, ga102_gr_new },
 	.nvdec    = { 0x00000003, ga102_nvdec_new },
-	.nvenc    = { 0x00000001, ga102_nvenc_new },
-	.ofa      = { 0x00000001, ga102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2593,8 +2588,6 @@ nv173_chipset = {
 	.fifo     = { 0x00000001, ga102_fifo_new },
 	.gr       = { 0x00000001, ga102_gr_new },
 	.nvdec    = { 0x00000003, ga102_nvdec_new },
-	.nvenc    = { 0x00000001, ga102_nvenc_new },
-	.ofa      = { 0x00000001, ga102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2625,8 +2618,6 @@ nv174_chipset = {
 	.fifo     = { 0x00000001, ga102_fifo_new },
 	.gr       = { 0x00000001, ga102_gr_new },
 	.nvdec    = { 0x00000003, ga102_nvdec_new },
-	.nvenc    = { 0x00000001, ga102_nvenc_new },
-	.ofa      = { 0x00000001, ga102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2657,8 +2648,6 @@ nv176_chipset = {
 	.fifo     = { 0x00000001, ga102_fifo_new },
 	.gr       = { 0x00000001, ga102_gr_new },
 	.nvdec    = { 0x00000003, ga102_nvdec_new },
-	.nvenc    = { 0x00000001, ga102_nvenc_new },
-	.ofa      = { 0x00000001, ga102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2689,12 +2678,26 @@ nv177_chipset = {
 	.fifo     = { 0x00000001, ga102_fifo_new },
 	.gr       = { 0x00000001, ga102_gr_new },
 	.nvdec    = { 0x00000003, ga102_nvdec_new },
-	.nvenc    = { 0x00000001, ga102_nvenc_new },
-	.ofa      = { 0x00000001, ga102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
 static const struct nvkm_device_chip
+nv180_chipset = {
+	.name = "GH100",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fault    = { 0x00000001, tu102_fault_new },
+	.fb       = { 0x00000001, gh100_fb_new },
+	.fsp      = { 0x00000001, gh100_fsp_new },
+	.gsp      = { 0x00000001, gh100_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
 nv192_chipset = {
 	.name = "AD102",
 	.bar      = { 0x00000001, tu102_bar_new },
@@ -2709,14 +2712,9 @@ nv192_chipset = {
 	.timer    = { 0x00000001, gk20a_timer_new },
 	.vfn      = { 0x00000001, ga100_vfn_new },
 	.ce       = { 0x0000001f, ga102_ce_new },
-	.disp     = { 0x00000001, ad102_disp_new },
+	.disp     = { 0x00000001, ga102_disp_new },
 	.dma      = { 0x00000001, gv100_dma_new },
 	.fifo     = { 0x00000001, ga102_fifo_new },
-	.gr       = { 0x00000001, ad102_gr_new },
-	.nvdec    = { 0x0000000f, ad102_nvdec_new },
-	.nvenc    = { 0x00000007, ad102_nvenc_new },
-	.nvjpg    = { 0x0000000f, ad102_nvjpg_new },
-	.ofa      = { 0x00000001, ad102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2735,14 +2733,9 @@ nv193_chipset = {
 	.timer    = { 0x00000001, gk20a_timer_new },
 	.vfn      = { 0x00000001, ga100_vfn_new },
 	.ce       = { 0x0000001f, ga102_ce_new },
-	.disp     = { 0x00000001, ad102_disp_new },
+	.disp     = { 0x00000001, ga102_disp_new },
 	.dma      = { 0x00000001, gv100_dma_new },
 	.fifo     = { 0x00000001, ga102_fifo_new },
-	.gr       = { 0x00000001, ad102_gr_new },
-	.nvdec    = { 0x0000000f, ad102_nvdec_new },
-	.nvenc    = { 0x00000007, ad102_nvenc_new },
-	.nvjpg    = { 0x0000000f, ad102_nvjpg_new },
-	.ofa      = { 0x00000001, ad102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2761,14 +2754,9 @@ nv194_chipset = {
 	.timer    = { 0x00000001, gk20a_timer_new },
 	.vfn      = { 0x00000001, ga100_vfn_new },
 	.ce       = { 0x0000001f, ga102_ce_new },
-	.disp     = { 0x00000001, ad102_disp_new },
+	.disp     = { 0x00000001, ga102_disp_new },
 	.dma      = { 0x00000001, gv100_dma_new },
 	.fifo     = { 0x00000001, ga102_fifo_new },
-	.gr       = { 0x00000001, ad102_gr_new },
-	.nvdec    = { 0x0000000f, ad102_nvdec_new },
-	.nvenc    = { 0x00000007, ad102_nvenc_new },
-	.nvjpg    = { 0x0000000f, ad102_nvjpg_new },
-	.ofa      = { 0x00000001, ad102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2787,14 +2775,9 @@ nv196_chipset = {
 	.timer    = { 0x00000001, gk20a_timer_new },
 	.vfn      = { 0x00000001, ga100_vfn_new },
 	.ce       = { 0x0000001f, ga102_ce_new },
-	.disp     = { 0x00000001, ad102_disp_new },
+	.disp     = { 0x00000001, ga102_disp_new },
 	.dma      = { 0x00000001, gv100_dma_new },
 	.fifo     = { 0x00000001, ga102_fifo_new },
-	.gr       = { 0x00000001, ad102_gr_new },
-	.nvdec    = { 0x0000000f, ad102_nvdec_new },
-	.nvenc    = { 0x00000007, ad102_nvenc_new },
-	.nvjpg    = { 0x0000000f, ad102_nvjpg_new },
-	.ofa      = { 0x00000001, ad102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
@@ -2813,17 +2796,122 @@ nv197_chipset = {
 	.timer    = { 0x00000001, gk20a_timer_new },
 	.vfn      = { 0x00000001, ga100_vfn_new },
 	.ce       = { 0x0000001f, ga102_ce_new },
-	.disp     = { 0x00000001, ad102_disp_new },
+	.disp     = { 0x00000001, ga102_disp_new },
 	.dma      = { 0x00000001, gv100_dma_new },
 	.fifo     = { 0x00000001, ga102_fifo_new },
-	.gr       = { 0x00000001, ad102_gr_new },
-	.nvdec    = { 0x0000000f, ad102_nvdec_new },
-	.nvenc    = { 0x00000007, ad102_nvenc_new },
-	.nvjpg    = { 0x0000000f, ad102_nvjpg_new },
-	.ofa      = { 0x00000001, ad102_ofa_new },
 	.sec2     = { 0x00000001, ga102_sec2_new },
 };
 
+static const struct nvkm_device_chip
+nv1a0_chipset = {
+	.name = "GB100",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb100_fb_new },
+	.fsp      = { 0x00000001, gb100_fsp_new },
+	.gsp      = { 0x00000001, gb100_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
+nv1a2_chipset = {
+	.name = "GB102",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb100_fb_new },
+	.fsp      = { 0x00000001, gb100_fsp_new },
+	.gsp      = { 0x00000001, gb100_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
+nv1b2_chipset = {
+	.name = "GB202",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb202_fb_new },
+	.fsp      = { 0x00000001, gb202_fsp_new },
+	.gsp      = { 0x00000001, gb202_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.disp     = { 0x00000001, ga102_disp_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
+nv1b3_chipset = {
+	.name = "GB203",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb202_fb_new },
+	.fsp      = { 0x00000001, gb202_fsp_new },
+	.gsp      = { 0x00000001, gb202_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.disp     = { 0x00000001, ga102_disp_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
+nv1b5_chipset = {
+	.name = "GB205",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb202_fb_new },
+	.fsp      = { 0x00000001, gb202_fsp_new },
+	.gsp      = { 0x00000001, gb202_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.disp     = { 0x00000001, ga102_disp_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
+nv1b6_chipset = {
+	.name = "GB206",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb202_fb_new },
+	.fsp      = { 0x00000001, gb202_fsp_new },
+	.gsp      = { 0x00000001, gb202_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.disp     = { 0x00000001, ga102_disp_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
+static const struct nvkm_device_chip
+nv1b7_chipset = {
+	.name = "GB207",
+	.bar      = { 0x00000001, tu102_bar_new },
+	.fb       = { 0x00000001, gb202_fb_new },
+	.fsp      = { 0x00000001, gb202_fsp_new },
+	.gsp      = { 0x00000001, gb202_gsp_new },
+	.imem     = { 0x00000001, gh100_instmem_new },
+	.mmu      = { 0x00000001, gh100_mmu_new },
+	.pci      = { 0x00000001, gh100_pci_new },
+	.timer    = { 0x00000001, gk20a_timer_new },
+	.vfn      = { 0x00000001, ga100_vfn_new },
+	.disp     = { 0x00000001, ga102_disp_new },
+	.fifo     = { 0x00000001, ga102_fifo_new },
+};
+
 struct nvkm_subdev *
 nvkm_device_subdev(struct nvkm_device *device, int type, int inst)
 {
@@ -3065,8 +3153,8 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 	device->debug = nvkm_dbgopt(device->dbgopt, "device");
 	INIT_LIST_HEAD(&device->subdev);
 
-	mmio_base = device->func->resource_addr(device, 0);
-	mmio_size = device->func->resource_size(device, 0);
+	mmio_base = device->func->resource_addr(device, NVKM_BAR0_PRI);
+	mmio_size = device->func->resource_size(device, NVKM_BAR0_PRI);
 
 	device->pri = ioremap(mmio_base, mmio_size);
 	if (device->pri == NULL) {
@@ -3139,7 +3227,10 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 		case 0x140: device->card_type = GV100; break;
 		case 0x160: device->card_type = TU100; break;
 		case 0x170: device->card_type = GA100; break;
+		case 0x180: device->card_type = GH100; break;
 		case 0x190: device->card_type = AD100; break;
+		case 0x1a0: device->card_type = GB10x; break;
+		case 0x1b0: device->card_type = GB20x; break;
 		default:
 			break;
 		}
@@ -3242,11 +3333,19 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 	case 0x174: device->chip = &nv174_chipset; break;
 	case 0x176: device->chip = &nv176_chipset; break;
 	case 0x177: device->chip = &nv177_chipset; break;
+	case 0x180: device->chip = &nv180_chipset; break;
 	case 0x192: device->chip = &nv192_chipset; break;
 	case 0x193: device->chip = &nv193_chipset; break;
 	case 0x194: device->chip = &nv194_chipset; break;
 	case 0x196: device->chip = &nv196_chipset; break;
 	case 0x197: device->chip = &nv197_chipset; break;
+	case 0x1a0: device->chip = &nv1a0_chipset; break;
+	case 0x1a2: device->chip = &nv1a2_chipset; break;
+	case 0x1b2: device->chip = &nv1b2_chipset; break;
+	case 0x1b3: device->chip = &nv1b3_chipset; break;
+	case 0x1b5: device->chip = &nv1b5_chipset; break;
+	case 0x1b6: device->chip = &nv1b6_chipset; break;
+	case 0x1b7: device->chip = &nv1b7_chipset; break;
 	default:
 		if (nvkm_boolopt(device->cfgopt, "NvEnableUnsupportedChipsets", false)) {
 			switch (device->chipset) {
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
index 3ff6436007fa..8f0261a0d618 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
@@ -1560,18 +1560,42 @@ nvkm_device_pci(struct nvkm_device *device)
 	return container_of(device, struct nvkm_device_pci, device);
 }
 
+static int
+nvkm_device_pci_resource_idx(struct nvkm_device_pci *pdev, enum nvkm_bar_id bar)
+{
+	int idx = 0;
+
+	if (bar == NVKM_BAR0_PRI)
+		return idx;
+
+	idx += (pci_resource_flags(pdev->pdev, idx) & IORESOURCE_MEM_64) ? 2 : 1;
+	if (bar == NVKM_BAR1_FB)
+		return idx;
+
+	idx += (pci_resource_flags(pdev->pdev, idx) & IORESOURCE_MEM_64) ? 2 : 1;
+	if (bar == NVKM_BAR2_INST)
+		return idx;
+
+	WARN_ON(1);
+	return -1;
+}
+
 static resource_size_t
-nvkm_device_pci_resource_addr(struct nvkm_device *device, unsigned bar)
+nvkm_device_pci_resource_addr(struct nvkm_device *device, enum nvkm_bar_id bar)
 {
 	struct nvkm_device_pci *pdev = nvkm_device_pci(device);
-	return pci_resource_start(pdev->pdev, bar);
+	int idx = nvkm_device_pci_resource_idx(pdev, bar);
+
+	return idx >= 0 ? pci_resource_start(pdev->pdev, idx) : 0;
 }
 
 static resource_size_t
-nvkm_device_pci_resource_size(struct nvkm_device *device, unsigned bar)
+nvkm_device_pci_resource_size(struct nvkm_device *device, enum nvkm_bar_id bar)
 {
 	struct nvkm_device_pci *pdev = nvkm_device_pci(device);
-	return pci_resource_len(pdev->pdev, bar);
+	int idx = nvkm_device_pci_resource_idx(pdev, bar);
+
+	return idx >= 0 ? pci_resource_len(pdev->pdev, idx) : 0;
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
index e42b18820a95..75ee7506d443 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
@@ -11,6 +11,7 @@
 #include <subdev/devinit.h>
 #include <subdev/fault.h>
 #include <subdev/fb.h>
+#include <subdev/fsp.h>
 #include <subdev/fuse.h>
 #include <subdev/gpio.h>
 #include <subdev/gsp.h>
@@ -43,8 +44,6 @@
 #include <engine/msvld.h>
 #include <engine/nvenc.h>
 #include <engine/nvdec.h>
-#include <engine/nvjpg.h>
-#include <engine/ofa.h>
 #include <engine/sec.h>
 #include <engine/sec2.h>
 #include <engine/sw.h>
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
index 78a83f904bbd..0ca2dfe99676 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
@@ -186,21 +186,31 @@ nvkm_device_tegra(struct nvkm_device *device)
 }
 
 static struct resource *
-nvkm_device_tegra_resource(struct nvkm_device *device, unsigned bar)
+nvkm_device_tegra_resource(struct nvkm_device *device, enum nvkm_bar_id bar)
 {
 	struct nvkm_device_tegra *tdev = nvkm_device_tegra(device);
-	return platform_get_resource(tdev->pdev, IORESOURCE_MEM, bar);
+	int idx;
+
+	switch (bar) {
+	case NVKM_BAR0_PRI: idx = 0; break;
+	case NVKM_BAR1_FB : idx = 1; break;
+	default:
+		WARN_ON(1);
+		return ERR_PTR(-EINVAL);
+	}
+
+	return platform_get_resource(tdev->pdev, IORESOURCE_MEM, idx);
 }
 
 static resource_size_t
-nvkm_device_tegra_resource_addr(struct nvkm_device *device, unsigned bar)
+nvkm_device_tegra_resource_addr(struct nvkm_device *device, enum nvkm_bar_id bar)
 {
 	struct resource *res = nvkm_device_tegra_resource(device, bar);
 	return res ? res->start : 0;
 }
 
 static resource_size_t
-nvkm_device_tegra_resource_size(struct nvkm_device *device, unsigned bar)
+nvkm_device_tegra_resource_size(struct nvkm_device *device, enum nvkm_bar_id bar)
 {
 	struct resource *res = nvkm_device_tegra_resource(device, bar);
 	return res ? resource_size(res) : 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c
index d7f75b3a43c8..58191b7a0494 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c
@@ -148,6 +148,9 @@ nvkm_udevice_info(struct nvkm_udevice *udev, void *data, u32 size)
 	case TU100: args->v0.family = NV_DEVICE_INFO_V0_TURING; break;
 	case GA100: args->v0.family = NV_DEVICE_INFO_V0_AMPERE; break;
 	case AD100: args->v0.family = NV_DEVICE_INFO_V0_ADA; break;
+	case GH100: args->v0.family = NV_DEVICE_INFO_V0_HOPPER; break;
+	case GB10x: args->v0.family = NV_DEVICE_INFO_V0_BLACKWELL; break;
+	case GB20x: args->v0.family = NV_DEVICE_INFO_V0_BLACKWELL; break;
 	default:
 		args->v0.family = 0;
 		break;
@@ -209,8 +212,8 @@ nvkm_udevice_map(struct nvkm_object *object, void *argv, u32 argc,
 	struct nvkm_udevice *udev = nvkm_udevice(object);
 	struct nvkm_device *device = udev->device;
 	*type = NVKM_OBJECT_MAP_IO;
-	*addr = device->func->resource_addr(device, 0);
-	*size = device->func->resource_size(device, 0);
+	*addr = device->func->resource_addr(device, NVKM_BAR0_PRI);
+	*size = device->func->resource_size(device, NVKM_BAR0_PRI);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
index e346e924fee8..e1aecd3fe96c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
@@ -27,9 +27,6 @@ nvkm-y += nvkm/engine/disp/gp102.o
 nvkm-y += nvkm/engine/disp/gv100.o
 nvkm-y += nvkm/engine/disp/tu102.o
 nvkm-y += nvkm/engine/disp/ga102.o
-nvkm-y += nvkm/engine/disp/ad102.o
-
-nvkm-y += nvkm/engine/disp/r535.o
 
 nvkm-y += nvkm/engine/disp/udisp.o
 nvkm-y += nvkm/engine/disp/uconn.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ad102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ad102.c
deleted file mode 100644
index 7f300a79aa29..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ad102.c
+++ /dev/null
@@ -1,52 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-#include "chan.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_disp_func
-ad102_disp = {
-	.uevent = &gv100_disp_chan_uevent,
-	.ramht_size = 0x2000,
-	.root = {  0, 0,AD102_DISP },
-	.user = {
-		{{-1,-1,GV100_DISP_CAPS                  }, gv100_disp_caps_new },
-		{{ 0, 0,GA102_DISP_CURSOR                }, nvkm_disp_chan_new, &gv100_disp_curs },
-		{{ 0, 0,GA102_DISP_WINDOW_IMM_CHANNEL_DMA}, nvkm_disp_wndw_new, &gv100_disp_wimm },
-		{{ 0, 0,AD102_DISP_CORE_CHANNEL_DMA      }, nvkm_disp_core_new, &gv100_disp_core },
-		{{ 0, 0,GA102_DISP_WINDOW_CHANNEL_DMA    }, nvkm_disp_wndw_new, &gv100_disp_wndw },
-		{}
-	},
-};
-
-int
-ad102_disp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-	       struct nvkm_disp **pdisp)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_disp_new(&ad102_disp, device, type, inst, pdisp);
-
-	return -ENODEV;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/chan.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/chan.c
index 4e43ee383c34..9b84e357d354 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/chan.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/chan.c
@@ -49,7 +49,7 @@ nvkm_disp_chan_map(struct nvkm_object *object, void *argv, u32 argc,
 {
 	struct nvkm_disp_chan *chan = nvkm_disp_chan(object);
 	struct nvkm_device *device = chan->disp->engine.subdev.device;
-	const u64 base = device->func->resource_addr(device, 0);
+	const u64 base = device->func->resource_addr(device, NVKM_BAR0_PRI);
 
 	*type = NVKM_OBJECT_MAP_IO;
 	*addr = base + chan->func->user(chan, size);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c
index cfa3698d3a2f..614921166fba 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c
@@ -805,7 +805,7 @@ gv100_disp_caps_map(struct nvkm_object *object, void *argv, u32 argc,
 	struct gv100_disp_caps *caps = gv100_disp_caps(object);
 	struct nvkm_device *device = caps->disp->engine.subdev.device;
 	*type = NVKM_OBJECT_MAP_IO;
-	*addr = 0x640000 + device->func->resource_addr(device, 0);
+	*addr = 0x640000 + device->func->resource_addr(device, NVKM_BAR0_PRI);
 	*size = 0x1000;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
index aff92848abfe..376e9c3bcb1a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
@@ -25,8 +25,7 @@ nvkm-y += nvkm/engine/fifo/gv100.o
 nvkm-y += nvkm/engine/fifo/tu102.o
 nvkm-y += nvkm/engine/fifo/ga100.o
 nvkm-y += nvkm/engine/fifo/ga102.o
-
-nvkm-y += nvkm/engine/fifo/r535.o
+nvkm-y += nvkm/engine/fifo/gb202.o
 
 nvkm-y += nvkm/engine/fifo/ucgrp.o
 nvkm-y += nvkm/engine/fifo/uchan.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c
index 22443fe4a39f..fdffa0391b31 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/base.c
@@ -303,7 +303,7 @@ nvkm_fifo_oneinit(struct nvkm_engine *engine)
 	}
 
 	/* Allocate USERD + BAR1 polling area. */
-	if (fifo->func->chan.func->userd->bar == 1) {
+	if (fifo->func->chan.func->userd->bar == NVKM_BAR1_FB) {
 		struct nvkm_vmm *bar1 = nvkm_bar_bar1_vmm(device);
 
 		ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, fifo->chid->nr *
@@ -349,8 +349,6 @@ nvkm_fifo_dtor(struct nvkm_engine *engine)
 	nvkm_chid_unref(&fifo->cgid);
 	nvkm_chid_unref(&fifo->chid);
 
-	mutex_destroy(&fifo->userd.mutex);
-
 	nvkm_event_fini(&fifo->nonstall.event);
 	mutex_destroy(&fifo->mutex);
 
@@ -391,8 +389,5 @@ nvkm_fifo_new_(const struct nvkm_fifo_func *func, struct nvkm_device *device,
 	spin_lock_init(&fifo->lock);
 	mutex_init(&fifo->mutex);
 
-	INIT_LIST_HEAD(&fifo->userd.list);
-	mutex_init(&fifo->userd.mutex);
-
 	return nvkm_engine_ctor(&nvkm_fifo, device, type, inst, true, &fifo->engine);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c
index 7d4716dcd512..4e09985424b6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c
@@ -275,11 +275,7 @@ nvkm_chan_del(struct nvkm_chan **pchan)
 	nvkm_gpuobj_del(&chan->ramfc);
 
 	if (chan->cgrp) {
-		if (!chan->func->id_put)
-			nvkm_chid_put(chan->cgrp->runl->chid, chan->id, &chan->cgrp->lock);
-		else
-			chan->func->id_put(chan);
-
+		nvkm_chid_put(chan->cgrp->runl->chid, chan->id, &chan->cgrp->lock);
 		nvkm_cgrp_unref(&chan->cgrp);
 	}
 
@@ -359,14 +355,14 @@ nvkm_chan_new_(const struct nvkm_chan_func *func, struct nvkm_runl *runl, int ru
 	/* Validate arguments against class requirements. */
 	if ((runq && runq >= runl->func->runqs) ||
 	    (!func->inst->vmm != !vmm) ||
-	    ((func->userd->bar < 0) == !userd) ||
+	    (!func->userd->bar == !userd) ||
 	    (!func->ramfc->ctxdma != !dmaobj) ||
 	    ((func->ramfc->devm < devm) && devm != BIT(0)) ||
 	    (!func->ramfc->priv && priv)) {
 		RUNL_DEBUG(runl, "args runq:%d:%d vmm:%d:%p userd:%d:%p "
 				 "push:%d:%p devm:%08x:%08x priv:%d:%d",
 			   runl->func->runqs, runq, func->inst->vmm, vmm,
-			   func->userd->bar < 0, userd, func->ramfc->ctxdma, dmaobj,
+			   func->userd->bar, userd, func->ramfc->ctxdma, dmaobj,
 			   func->ramfc->devm, devm, func->ramfc->priv, priv);
 		return -EINVAL;
 	}
@@ -441,30 +437,26 @@ nvkm_chan_new_(const struct nvkm_chan_func *func, struct nvkm_runl *runl, int ru
 	}
 
 	/* Allocate channel ID. */
-	if (!chan->func->id_get) {
-		chan->id = nvkm_chid_get(runl->chid, chan);
-		if (chan->id >= 0) {
-			if (func->userd->bar < 0) {
-				if (ouserd + chan->func->userd->size >=
-					nvkm_memory_size(userd)) {
-					RUNL_DEBUG(runl, "ouserd %llx", ouserd);
-					return -EINVAL;
-				}
-
-				ret = nvkm_memory_kmap(userd, &chan->userd.mem);
-				if (ret) {
-					RUNL_DEBUG(runl, "userd %d", ret);
-					return ret;
-				}
-
-				chan->userd.base = ouserd;
-			} else {
-				chan->userd.mem = nvkm_memory_ref(fifo->userd.mem);
-				chan->userd.base = chan->id * chan->func->userd->size;
+	chan->id = nvkm_chid_get(runl->chid, chan);
+	if (chan->id >= 0) {
+		if (!func->userd->bar) {
+			if (ouserd + chan->func->userd->size >=
+				nvkm_memory_size(userd)) {
+				RUNL_DEBUG(runl, "ouserd %llx", ouserd);
+				return -EINVAL;
+			}
+
+			ret = nvkm_memory_kmap(userd, &chan->userd.mem);
+			if (ret) {
+				RUNL_DEBUG(runl, "userd %d", ret);
+				return ret;
 			}
+
+			chan->userd.base = ouserd;
+		} else {
+			chan->userd.mem = nvkm_memory_ref(fifo->userd.mem);
+			chan->userd.base = chan->id * chan->func->userd->size;
 		}
-	} else {
-		chan->id = chan->func->id_get(chan, userd, ouserd);
 	}
 
 	if (chan->id < 0) {
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h
index 013682a709d5..445db5dfd1e4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h
@@ -17,9 +17,6 @@ struct nvkm_cctx {
 };
 
 struct nvkm_chan_func {
-	int (*id_get)(struct nvkm_chan *, struct nvkm_memory *userd, u64 ouserd);
-	void (*id_put)(struct nvkm_chan *);
-
 	const struct nvkm_chan_func_inst {
 		u32 size;
 		bool zero;
@@ -27,7 +24,7 @@ struct nvkm_chan_func {
 	} *inst;
 
 	const struct nvkm_chan_func_userd {
-		int bar;
+		enum nvkm_bar_id bar;
 		u32 base;
 		u32 size;
 		void (*clear)(struct nvkm_chan *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gb202.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gb202.c
new file mode 100644
index 000000000000..b469e8afeb0b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gb202.c
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+#include "cgrp.h"
+#include "chan.h"
+#include "runl.h"
+
+u32
+gb202_chan_doorbell_handle(struct nvkm_chan *chan)
+{
+	return BIT(30) | (chan->cgrp->runl->id << 16) | chan->id;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c
index 6c94451d0faa..e4a4fad2eafc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c
@@ -133,7 +133,7 @@ gf100_chan_userd_clear(struct nvkm_chan *chan)
 
 static const struct nvkm_chan_func_userd
 gf100_chan_userd = {
-	.bar = 1,
+	.bar = NVKM_BAR1_FB,
 	.size = 0x1000,
 	.clear = gf100_chan_userd_clear,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
index d8a4d773a58c..5655eda52a7b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
@@ -113,7 +113,7 @@ gk104_chan_ramfc = {
 
 const struct nvkm_chan_func_userd
 gk104_chan_userd = {
-	.bar = 1,
+	.bar = NVKM_BAR1_FB,
 	.size = 0x200,
 	.clear = gf100_chan_userd_clear,
 };
@@ -745,7 +745,7 @@ gk104_fifo_init(struct nvkm_fifo *fifo)
 {
 	struct nvkm_device *device = fifo->engine.subdev.device;
 
-	if (fifo->func->chan.func->userd->bar == 1)
+	if (fifo->func->chan.func->userd->bar == NVKM_BAR1_FB)
 		nvkm_wr32(device, 0x002254, 0x10000000 | fifo->userd.bar1->addr >> 12);
 
 	nvkm_wr32(device, 0x002100, 0xffffffff);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c
index 33066c8cdc64..d7f046c03cfd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c
@@ -70,7 +70,6 @@ gv100_chan_ramfc = {
 
 const struct nvkm_chan_func_userd
 gv100_chan_userd = {
-	.bar = -1,
 	.size = 0x200,
 	.clear = gf100_chan_userd_clear,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c
index 674faf002b20..c4b8e567d86f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c
@@ -154,7 +154,7 @@ nv04_chan_ramfc = {
 
 const struct nvkm_chan_func_userd
 nv04_chan_userd = {
-	.bar = 0,
+	.bar = NVKM_BAR0_PRI,
 	.base = 0x800000,
 	.size = 0x010000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv40.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv40.c
index e50a94b6d7f8..084ca5561ee1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv40.c
@@ -93,7 +93,7 @@ nv40_chan_ramfc = {
 
 static const struct nvkm_chan_func_userd
 nv40_chan_userd = {
-	.bar = 0,
+	.bar = NVKM_BAR0_PRI,
 	.base = 0xc00000,
 	.size = 0x001000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv50.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv50.c
index 954b5f3a7d57..7bf77661157d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv50.c
@@ -124,7 +124,7 @@ nv50_chan_ramfc = {
 
 const struct nvkm_chan_func_userd
 nv50_chan_userd = {
-	.bar = 0,
+	.bar = NVKM_BAR0_PRI,
 	.base = 0xc00000,
 	.size = 0x002000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h
index a0f3277605a5..5e81ae195329 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/priv.h
@@ -6,6 +6,7 @@
 #include <core/enum.h>
 struct nvkm_cctx;
 struct nvkm_cgrp;
+struct nvkm_chan;
 struct nvkm_engn;
 struct nvkm_memory;
 struct nvkm_runl;
@@ -195,6 +196,7 @@ extern const struct nvkm_chan_func_ramfc gv100_chan_ramfc;
 
 void tu102_fifo_intr_ctxsw_timeout_info(struct nvkm_engn *, u32 info);
 extern const struct nvkm_fifo_func_mmu_fault tu102_fifo_mmu_fault;
+u32 tu102_chan_doorbell_handle(struct nvkm_chan *);
 
 int ga100_fifo_runl_ctor(struct nvkm_fifo *);
 int ga100_fifo_nonstall_ctor(struct nvkm_fifo *);
@@ -206,6 +208,8 @@ extern const struct nvkm_engn_func ga100_engn_ce;
 extern const struct nvkm_cgrp_func ga100_cgrp;
 extern const struct nvkm_chan_func ga100_chan;
 
+u32 gb202_chan_doorbell_handle(struct nvkm_chan *);
+
 int nvkm_uchan_new(struct nvkm_fifo *, struct nvkm_cgrp *, const struct nvkm_oclass *,
 		   void *argv, u32 argc, struct nvkm_object **);
 int nvkm_ucgrp_new(struct nvkm_fifo *, const struct nvkm_oclass *, void *argv, u32 argc,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu102.c
index 1d39a6840a40..c5a03298e88c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu102.c
@@ -31,7 +31,7 @@
 
 #include <nvif/class.h>
 
-static u32
+u32
 tu102_chan_doorbell_handle(struct nvkm_chan *chan)
 {
 	return (chan->cgrp->runl->id << 16) | chan->id;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/uchan.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/uchan.c
index 9e56bcc166ed..52420a1edca5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/uchan.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/uchan.c
@@ -258,7 +258,7 @@ nvkm_uchan_map(struct nvkm_object *object, void *argv, u32 argc,
 	struct nvkm_chan *chan = nvkm_uchan(object)->chan;
 	struct nvkm_device *device = chan->cgrp->runl->fifo->engine.subdev.device;
 
-	if (chan->func->userd->bar < 0)
+	if (!chan->func->userd->bar)
 		return -ENOSYS;
 
 	*type = NVKM_OBJECT_MAP_IO;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild
index 1555f8c40b4f..b5418f05ccd8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild
@@ -41,9 +41,6 @@ nvkm-y += nvkm/engine/gr/gp10b.o
 nvkm-y += nvkm/engine/gr/gv100.o
 nvkm-y += nvkm/engine/gr/tu102.o
 nvkm-y += nvkm/engine/gr/ga102.o
-nvkm-y += nvkm/engine/gr/ad102.o
-
-nvkm-y += nvkm/engine/gr/r535.o
 
 nvkm-y += nvkm/engine/gr/ctxnv40.o
 nvkm-y += nvkm/engine/gr/ctxnv50.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ga102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ga102.c
index d285c597aff9..2b51f1d0c281 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ga102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ga102.c
@@ -352,7 +352,7 @@ int
 ga102_gr_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_gr **pgr)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_gr_new(&ga102_gr, device, type, inst, pgr);
+		return -ENODEV;
 
 	return gf100_gr_new_(ga102_gr_fwif, device, type, inst, pgr);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
index b0e0c9305034..54f686ba39ac 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
@@ -445,6 +445,4 @@ void gp108_gr_acr_bld_patch(struct nvkm_acr *, u32, s64);
 
 int gf100_gr_new_(const struct gf100_gr_fwif *, struct nvkm_device *, enum nvkm_subdev_type, int,
 		  struct nvkm_gr **);
-int r535_gr_new(const struct gf100_gr_func *, struct nvkm_device *, enum nvkm_subdev_type, int,
-		struct nvkm_gr **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv20.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv20.c
index 02a8c62a0a32..13407fafe947 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv20.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv20.c
@@ -297,7 +297,7 @@ nv20_gr_init(struct nvkm_gr *base)
 	nvkm_wr32(device, NV10_PGRAPH_SURFACE, tmp);
 
 	/* begin RAM config */
-	vramsz = device->func->resource_size(device, 1) - 1;
+	vramsz = device->func->resource_size(device, NVKM_BAR1_FB) - 1;
 	nvkm_wr32(device, 0x4009A4, nvkm_rd32(device, 0x100200));
 	nvkm_wr32(device, 0x4009A8, nvkm_rd32(device, 0x100204));
 	nvkm_wr32(device, NV10_PGRAPH_RDI_INDEX, 0x00EA0000);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv40.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv40.c
index a5e1f02791b4..b609b0150ba1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/nv40.c
@@ -386,7 +386,7 @@ nv40_gr_init(struct nvkm_gr *base)
 	}
 
 	/* begin RAM config */
-	vramsz = device->func->resource_size(device, 1) - 1;
+	vramsz = device->func->resource_size(device, NVKM_BAR1_FB) - 1;
 	switch (device->chipset) {
 	case 0x40:
 		nvkm_wr32(device, 0x4009A4, nvkm_rd32(device, 0x100200));
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/r535.c
deleted file mode 100644
index f4bed3eb1ec2..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/r535.c
+++ /dev/null
@@ -1,508 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "gf100.h"
-
-#include <core/memory.h>
-#include <subdev/gsp.h>
-#include <subdev/mmu/vmm.h>
-#include <engine/fifo/priv.h>
-
-#include <nvif/if900d.h>
-
-#include <nvhw/drf.h>
-
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/alloc/alloc_channel.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080fifo.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h>
-#include <nvrm/535.113.01/nvidia/generated/g_kernel_channel_nvoc.h>
-
-#define r535_gr(p) container_of((p), struct r535_gr, base)
-
-#define R515_GR_MAX_CTXBUFS 9
-
-struct r535_gr {
-	struct nvkm_gr base;
-
-	struct {
-		u16 bufferId;
-		u32 size;
-		u8  page;
-		u8  align;
-		bool global;
-		bool init;
-		bool ro;
-	} ctxbuf[R515_GR_MAX_CTXBUFS];
-	int ctxbuf_nr;
-
-	struct nvkm_memory *ctxbuf_mem[R515_GR_MAX_CTXBUFS];
-};
-
-struct r535_gr_chan {
-	struct nvkm_object object;
-	struct r535_gr *gr;
-
-	struct nvkm_vmm *vmm;
-	struct nvkm_chan *chan;
-
-	struct nvkm_memory *mem[R515_GR_MAX_CTXBUFS];
-	struct nvkm_vma    *vma[R515_GR_MAX_CTXBUFS];
-};
-
-struct r535_gr_obj {
-	struct nvkm_object object;
-	struct nvkm_gsp_object rm;
-};
-
-static void *
-r535_gr_obj_dtor(struct nvkm_object *object)
-{
-	struct r535_gr_obj *obj = container_of(object, typeof(*obj), object);
-
-	nvkm_gsp_rm_free(&obj->rm);
-	return obj;
-}
-
-static const struct nvkm_object_func
-r535_gr_obj = {
-	.dtor = r535_gr_obj_dtor,
-};
-
-static int
-r535_gr_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
-		 struct nvkm_object **pobject)
-{
-	struct r535_gr_chan *chan = container_of(oclass->parent, typeof(*chan), object);
-	struct r535_gr_obj *obj;
-
-	if (!(obj = kzalloc(sizeof(*obj), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_gr_obj, oclass, &obj->object);
-	*pobject = &obj->object;
-
-	return nvkm_gsp_rm_alloc(&chan->chan->rm.object, oclass->handle, oclass->base.oclass, 0,
-				 &obj->rm);
-}
-
-static void *
-r535_gr_chan_dtor(struct nvkm_object *object)
-{
-	struct r535_gr_chan *grc = container_of(object, typeof(*grc), object);
-	struct r535_gr *gr = grc->gr;
-
-	for (int i = 0; i < gr->ctxbuf_nr; i++) {
-		nvkm_vmm_put(grc->vmm, &grc->vma[i]);
-		nvkm_memory_unref(&grc->mem[i]);
-	}
-
-	nvkm_vmm_unref(&grc->vmm);
-	return grc;
-}
-
-static const struct nvkm_object_func
-r535_gr_chan = {
-	.dtor = r535_gr_chan_dtor,
-};
-
-static int
-r535_gr_promote_ctx(struct r535_gr *gr, bool golden, struct nvkm_vmm *vmm,
-		    struct nvkm_memory **pmem, struct nvkm_vma **pvma,
-		    struct nvkm_gsp_object *chan)
-{
-	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
-	struct nvkm_device *device = subdev->device;
-	NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS *ctrl;
-
-	ctrl = nvkm_gsp_rm_ctrl_get(&vmm->rm.device.subdevice,
-				    NV2080_CTRL_CMD_GPU_PROMOTE_CTX, sizeof(*ctrl));
-	if (WARN_ON(IS_ERR(ctrl)))
-		return PTR_ERR(ctrl);
-
-	ctrl->engineType = 1;
-	ctrl->hChanClient = vmm->rm.client.object.handle;
-	ctrl->hObject = chan->handle;
-
-	for (int i = 0; i < gr->ctxbuf_nr; i++) {
-		NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY *entry =
-			&ctrl->promoteEntry[ctrl->entryCount];
-		const bool alloc = golden || !gr->ctxbuf[i].global;
-		int ret;
-
-		entry->bufferId = gr->ctxbuf[i].bufferId;
-		entry->bInitialize = gr->ctxbuf[i].init && alloc;
-
-		if (alloc) {
-			ret = nvkm_memory_new(device, gr->ctxbuf[i].init ?
-					      NVKM_MEM_TARGET_INST : NVKM_MEM_TARGET_INST_SR_LOST,
-					      gr->ctxbuf[i].size, 1 << gr->ctxbuf[i].page,
-					      gr->ctxbuf[i].init, &pmem[i]);
-			if (WARN_ON(ret))
-				return ret;
-
-			if (gr->ctxbuf[i].bufferId ==
-					NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PRIV_ACCESS_MAP)
-				entry->bNonmapped = 1;
-		} else {
-			if (gr->ctxbuf[i].bufferId ==
-				NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_UNRESTRICTED_PRIV_ACCESS_MAP)
-				continue;
-
-			pmem[i] = nvkm_memory_ref(gr->ctxbuf_mem[i]);
-		}
-
-		if (!entry->bNonmapped) {
-			struct gf100_vmm_map_v0 args = {
-				.priv = 1,
-				.ro   = gr->ctxbuf[i].ro,
-			};
-
-			mutex_lock(&vmm->mutex.vmm);
-			ret = nvkm_vmm_get_locked(vmm, false, true, false, 0, gr->ctxbuf[i].align,
-						  nvkm_memory_size(pmem[i]), &pvma[i]);
-			mutex_unlock(&vmm->mutex.vmm);
-			if (ret)
-				return ret;
-
-			ret = nvkm_memory_map(pmem[i], 0, vmm, pvma[i], &args, sizeof(args));
-			if (ret)
-				return ret;
-
-			entry->gpuVirtAddr = pvma[i]->addr;
-		}
-
-		if (entry->bInitialize) {
-			entry->gpuPhysAddr = nvkm_memory_addr(pmem[i]);
-			entry->size = gr->ctxbuf[i].size;
-			entry->physAttr = 4;
-		}
-
-		nvkm_debug(subdev,
-			   "promote %02d: pa %016llx/%08x sz %016llx va %016llx init:%d nm:%d\n",
-			   entry->bufferId, entry->gpuPhysAddr, entry->physAttr, entry->size,
-			   entry->gpuVirtAddr, entry->bInitialize, entry->bNonmapped);
-
-		ctrl->entryCount++;
-	}
-
-	return nvkm_gsp_rm_ctrl_wr(&vmm->rm.device.subdevice, ctrl);
-}
-
-static int
-r535_gr_chan_new(struct nvkm_gr *base, struct nvkm_chan *chan, const struct nvkm_oclass *oclass,
-		 struct nvkm_object **pobject)
-{
-	struct r535_gr *gr = r535_gr(base);
-	struct r535_gr_chan *grc;
-	int ret;
-
-	if (!(grc = kzalloc(sizeof(*grc), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_gr_chan, oclass, &grc->object);
-	grc->gr = gr;
-	grc->vmm = nvkm_vmm_ref(chan->vmm);
-	grc->chan = chan;
-	*pobject = &grc->object;
-
-	ret = r535_gr_promote_ctx(gr, false, grc->vmm, grc->mem, grc->vma, &chan->rm.object);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-static u64
-r535_gr_units(struct nvkm_gr *gr)
-{
-	struct nvkm_gsp *gsp = gr->engine.subdev.device->gsp;
-
-	return (gsp->gr.tpcs << 8) | gsp->gr.gpcs;
-}
-
-static int
-r535_gr_oneinit(struct nvkm_gr *base)
-{
-	NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS *info;
-	struct r535_gr *gr = container_of(base, typeof(*gr), base);
-	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
-	struct nvkm_device *device = subdev->device;
-	struct nvkm_gsp *gsp = device->gsp;
-	struct nvkm_mmu *mmu = device->mmu;
-	struct {
-		struct nvkm_memory *inst;
-		struct nvkm_vmm *vmm;
-		struct nvkm_gsp_object chan;
-		struct nvkm_vma *vma[R515_GR_MAX_CTXBUFS];
-	} golden = {};
-	int ret;
-
-	/* Allocate a channel to use for golden context init. */
-	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, 0x12000, 0, true, &golden.inst);
-	if (ret)
-		goto done;
-
-	ret = nvkm_vmm_new(device, 0x1000, 0, NULL, 0, NULL, "grGoldenVmm", &golden.vmm);
-	if (ret)
-		goto done;
-
-	ret = mmu->func->promote_vmm(golden.vmm);
-	if (ret)
-		goto done;
-
-	{
-		NV_CHANNELGPFIFO_ALLOCATION_PARAMETERS *args;
-
-		args = nvkm_gsp_rm_alloc_get(&golden.vmm->rm.device.object, 0xf1f00000,
-					     device->fifo->func->chan.user.oclass,
-					     sizeof(*args), &golden.chan);
-		if (IS_ERR(args)) {
-			ret = PTR_ERR(args);
-			goto done;
-		}
-
-		args->gpFifoOffset = 0;
-		args->gpFifoEntries = 0x1000 / 8;
-		args->flags =
-			NVDEF(NVOS04, FLAGS, CHANNEL_TYPE, PHYSICAL) |
-			NVDEF(NVOS04, FLAGS, VPR, FALSE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_SKIP_MAP_REFCOUNTING, FALSE) |
-			NVVAL(NVOS04, FLAGS, GROUP_CHANNEL_RUNQUEUE, 0) |
-			NVDEF(NVOS04, FLAGS, PRIVILEGED_CHANNEL, TRUE) |
-			NVDEF(NVOS04, FLAGS, DELAY_CHANNEL_SCHEDULING, FALSE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_DENY_PHYSICAL_MODE_CE, FALSE) |
-			NVVAL(NVOS04, FLAGS, CHANNEL_USERD_INDEX_VALUE, 0) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_USERD_INDEX_FIXED, FALSE) |
-			NVVAL(NVOS04, FLAGS, CHANNEL_USERD_INDEX_PAGE_VALUE, 0) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_USERD_INDEX_PAGE_FIXED, TRUE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_DENY_AUTH_LEVEL_PRIV, FALSE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_SKIP_SCRUBBER, FALSE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_CLIENT_MAP_FIFO, FALSE) |
-			NVDEF(NVOS04, FLAGS, SET_EVICT_LAST_CE_PREFETCH_CHANNEL, FALSE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_VGPU_PLUGIN_CONTEXT, FALSE) |
-			NVDEF(NVOS04, FLAGS, CHANNEL_PBDMA_ACQUIRE_TIMEOUT, FALSE) |
-			NVDEF(NVOS04, FLAGS, GROUP_CHANNEL_THREAD, DEFAULT) |
-			NVDEF(NVOS04, FLAGS, MAP_CHANNEL, FALSE) |
-			NVDEF(NVOS04, FLAGS, SKIP_CTXBUFFER_ALLOC, FALSE);
-		args->hVASpace = golden.vmm->rm.object.handle;
-		args->engineType = 1;
-		args->instanceMem.base = nvkm_memory_addr(golden.inst);
-		args->instanceMem.size = 0x1000;
-		args->instanceMem.addressSpace = 2;
-		args->instanceMem.cacheAttrib = 1;
-		args->ramfcMem.base = nvkm_memory_addr(golden.inst);
-		args->ramfcMem.size = 0x200;
-		args->ramfcMem.addressSpace = 2;
-		args->ramfcMem.cacheAttrib = 1;
-		args->userdMem.base = nvkm_memory_addr(golden.inst) + 0x1000;
-		args->userdMem.size = 0x200;
-		args->userdMem.addressSpace = 2;
-		args->userdMem.cacheAttrib = 1;
-		args->mthdbufMem.base = nvkm_memory_addr(golden.inst) + 0x2000;
-		args->mthdbufMem.size = 0x5000;
-		args->mthdbufMem.addressSpace = 2;
-		args->mthdbufMem.cacheAttrib = 1;
-		args->internalFlags =
-			NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, PRIVILEGE, ADMIN) |
-			NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, ERROR_NOTIFIER_TYPE, NONE) |
-			NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, ECC_ERROR_NOTIFIER_TYPE, NONE);
-
-		ret = nvkm_gsp_rm_alloc_wr(&golden.chan, args);
-		if (ret)
-			goto done;
-	}
-
-	/* Fetch context buffer info from RM and allocate each of them here to use
-	 * during golden context init (or later as a global context buffer).
-	 *
-	 * Also build the information that'll be used to create channel contexts.
-	 */
-	info = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
-				   NV2080_CTRL_CMD_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO,
-				   sizeof(*info));
-	if (WARN_ON(IS_ERR(info))) {
-		ret = PTR_ERR(info);
-		goto done;
-	}
-
-	for (int i = 0; i < ARRAY_SIZE(info->engineContextBuffersInfo[0].engine); i++) {
-		static const struct {
-			u32     id0; /* NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID */
-			u32     id1; /* NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID */
-			bool global;
-			bool   init;
-			bool     ro;
-		} map[] = {
-#define _A(n,N,G,I,R) { .id0 = NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_##n, \
-			.id1 = NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_##N, \
-			.global = (G), .init = (I), .ro = (R) }
-#define _B(N,G,I,R) _A(GRAPHICS_##N, N, (G), (I), (R))
-			/*                                       global   init     ro */
-			_A(           GRAPHICS,             MAIN, false,  true, false),
-			_B(                                PATCH, false,  true, false),
-			_A( GRAPHICS_BUNDLE_CB, BUFFER_BUNDLE_CB,  true, false, false),
-			_B(                             PAGEPOOL,  true, false, false),
-			_B(                         ATTRIBUTE_CB,  true, false, false),
-			_B(                        RTV_CB_GLOBAL,  true, false, false),
-			_B(                           FECS_EVENT,  true,  true, false),
-			_B(                      PRIV_ACCESS_MAP,  true,  true,  true),
-#undef _B
-#undef _A
-		};
-		u32 size = info->engineContextBuffersInfo[0].engine[i].size;
-		u8 align, page;
-		int id;
-
-		for (id = 0; id < ARRAY_SIZE(map); id++) {
-			if (map[id].id0 == i)
-				break;
-		}
-
-		nvkm_debug(subdev, "%02x: size:0x%08x %s\n", i,
-			   size, (id < ARRAY_SIZE(map)) ? "*" : "");
-		if (id >= ARRAY_SIZE(map))
-			continue;
-
-		if (map[id].id1 == NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_MAIN)
-			size = ALIGN(size, 0x1000) + 64 * 0x1000; /* per-subctx headers */
-
-		if      (size >= 1 << 21) page = 21;
-		else if (size >= 1 << 16) page = 16;
-		else			  page = 12;
-
-		if (map[id].id1 == NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_ATTRIBUTE_CB)
-			align = order_base_2(size);
-		else
-			align = page;
-
-		if (WARN_ON(gr->ctxbuf_nr == ARRAY_SIZE(gr->ctxbuf)))
-			continue;
-
-		gr->ctxbuf[gr->ctxbuf_nr].bufferId = map[id].id1;
-		gr->ctxbuf[gr->ctxbuf_nr].size     = size;
-		gr->ctxbuf[gr->ctxbuf_nr].page     = page;
-		gr->ctxbuf[gr->ctxbuf_nr].align    = align;
-		gr->ctxbuf[gr->ctxbuf_nr].global   = map[id].global;
-		gr->ctxbuf[gr->ctxbuf_nr].init     = map[id].init;
-		gr->ctxbuf[gr->ctxbuf_nr].ro       = map[id].ro;
-		gr->ctxbuf_nr++;
-
-		if (map[id].id1 == NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PRIV_ACCESS_MAP) {
-			if (WARN_ON(gr->ctxbuf_nr == ARRAY_SIZE(gr->ctxbuf)))
-				continue;
-
-			gr->ctxbuf[gr->ctxbuf_nr] = gr->ctxbuf[gr->ctxbuf_nr - 1];
-			gr->ctxbuf[gr->ctxbuf_nr].bufferId =
-				NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_UNRESTRICTED_PRIV_ACCESS_MAP;
-			gr->ctxbuf_nr++;
-		}
-	}
-
-	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, info);
-
-	/* Promote golden context to RM. */
-	ret = r535_gr_promote_ctx(gr, true, golden.vmm, gr->ctxbuf_mem, golden.vma, &golden.chan);
-	if (ret)
-		goto done;
-
-	/* Allocate 3D class on channel to trigger golden context init in RM. */
-	{
-		int i;
-
-		for (i = 0; gr->base.func->sclass[i].ctor; i++) {
-			if ((gr->base.func->sclass[i].oclass & 0xff) == 0x97) {
-				struct nvkm_gsp_object threed;
-
-				ret = nvkm_gsp_rm_alloc(&golden.chan, 0x97000000,
-							gr->base.func->sclass[i].oclass, 0,
-							&threed);
-				if (ret)
-					goto done;
-
-				nvkm_gsp_rm_free(&threed);
-				break;
-			}
-		}
-
-		if (WARN_ON(!gr->base.func->sclass[i].ctor)) {
-			ret = -EINVAL;
-			goto done;
-		}
-	}
-
-done:
-	nvkm_gsp_rm_free(&golden.chan);
-	for (int i = gr->ctxbuf_nr - 1; i >= 0; i--)
-		nvkm_vmm_put(golden.vmm, &golden.vma[i]);
-	nvkm_vmm_unref(&golden.vmm);
-	nvkm_memory_unref(&golden.inst);
-	return ret;
-
-}
-
-static void *
-r535_gr_dtor(struct nvkm_gr *base)
-{
-	struct r535_gr *gr = r535_gr(base);
-
-	while (gr->ctxbuf_nr)
-		nvkm_memory_unref(&gr->ctxbuf_mem[--gr->ctxbuf_nr]);
-
-	kfree(gr->base.func);
-	return gr;
-}
-
-int
-r535_gr_new(const struct gf100_gr_func *hw,
-	    struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_gr **pgr)
-{
-	struct nvkm_gr_func *rm;
-	struct r535_gr *gr;
-	int nclass;
-
-	for (nclass = 0; hw->sclass[nclass].oclass; nclass++);
-
-	if (!(rm = kzalloc(sizeof(*rm) + (nclass + 1) * sizeof(rm->sclass[0]), GFP_KERNEL)))
-		return -ENOMEM;
-
-	rm->dtor = r535_gr_dtor;
-	rm->oneinit = r535_gr_oneinit;
-	rm->units = r535_gr_units;
-	rm->chan_new = r535_gr_chan_new;
-
-	for (int i = 0; i < nclass; i++) {
-		rm->sclass[i].minver = hw->sclass[i].minver;
-		rm->sclass[i].maxver = hw->sclass[i].maxver;
-		rm->sclass[i].oclass = hw->sclass[i].oclass;
-		rm->sclass[i].ctor = r535_gr_obj_ctor;
-	}
-
-	if (!(gr = kzalloc(sizeof(*gr), GFP_KERNEL))) {
-		kfree(rm);
-		return -ENOMEM;
-	}
-
-	*pgr = &gr->base;
-
-	return nvkm_gr_ctor(rm, device, type, inst, true, &gr->base);
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c
index b7a458e9040a..bda8054c6b59 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c
@@ -219,7 +219,7 @@ int
 tu102_gr_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_gr **pgr)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_gr_new(&tu102_gr, device, type, inst, pgr);
+		return -ENODEV;
 
 	return gf100_gr_new_(tu102_gr_fwif, device, type, inst, pgr);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild
index 2b0e923cb755..37b0cdc760c7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild
@@ -2,8 +2,4 @@
 nvkm-y += nvkm/engine/nvdec/base.o
 nvkm-y += nvkm/engine/nvdec/gm107.o
 nvkm-y += nvkm/engine/nvdec/tu102.o
-nvkm-y += nvkm/engine/nvdec/ga100.o
 nvkm-y += nvkm/engine/nvdec/ga102.o
-nvkm-y += nvkm/engine/nvdec/ad102.o
-
-nvkm-y += nvkm/engine/nvdec/r535.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ga102.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ga102.c
index 022a9c824304..eea6368adae2 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ga102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ga102.c
@@ -23,16 +23,6 @@
 
 #include <subdev/gsp.h>
 
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ga102_nvdec_gsp = {
-	.sclass = {
-		{ -1, -1, NVC7B0_VIDEO_DECODER },
-		{}
-	}
-};
-
 static const struct nvkm_falcon_func
 ga102_nvdec_flcn = {
 	.disable = gm200_flcn_disable,
@@ -67,7 +57,7 @@ ga102_nvdec_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst
 		struct nvkm_nvdec **pnvdec)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvdec_new(&ga102_nvdec_gsp, device, type, inst, pnvdec);
+		return -ENODEV;
 
 	return nvkm_nvdec_new_(ga102_nvdec_fwif, device, type, inst, 0x848000, pnvdec);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/priv.h
index f506ae83bfd7..f8d43e913093 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/priv.h
@@ -20,7 +20,4 @@ extern const struct nvkm_nvdec_fwif gm107_nvdec_fwif[];
 
 int nvkm_nvdec_new_(const struct nvkm_nvdec_fwif *fwif, struct nvkm_device *,
 		    enum nvkm_subdev_type, int, u32 addr, struct nvkm_nvdec **);
-
-int r535_nvdec_new(const struct nvkm_engine_func *, struct nvkm_device *,
-		   enum nvkm_subdev_type, int, struct nvkm_nvdec **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/r535.c
deleted file mode 100644
index 75a24f3e6617..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/r535.c
+++ /dev/null
@@ -1,110 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <core/object.h>
-#include <subdev/gsp.h>
-#include <engine/fifo.h>
-
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-
-struct r535_nvdec_obj {
-	struct nvkm_object object;
-	struct nvkm_gsp_object rm;
-};
-
-static void *
-r535_nvdec_obj_dtor(struct nvkm_object *object)
-{
-	struct r535_nvdec_obj *obj = container_of(object, typeof(*obj), object);
-
-	nvkm_gsp_rm_free(&obj->rm);
-	return obj;
-}
-
-static const struct nvkm_object_func
-r535_nvdec_obj = {
-	.dtor = r535_nvdec_obj_dtor,
-};
-
-static int
-r535_nvdec_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
-		 struct nvkm_object **pobject)
-{
-	struct nvkm_chan *chan = nvkm_uchan_chan(oclass->parent);
-	struct r535_nvdec_obj *obj;
-	NV_BSP_ALLOCATION_PARAMETERS *args;
-
-	if (!(obj = kzalloc(sizeof(*obj), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_nvdec_obj, oclass, &obj->object);
-	*pobject = &obj->object;
-
-	args = nvkm_gsp_rm_alloc_get(&chan->rm.object, oclass->handle, oclass->base.oclass,
-				     sizeof(*args), &obj->rm);
-	if (WARN_ON(IS_ERR(args)))
-		return PTR_ERR(args);
-
-	args->size = sizeof(*args);
-	args->engineInstance = oclass->engine->subdev.inst;
-
-	return nvkm_gsp_rm_alloc_wr(&obj->rm, args);
-}
-
-static void *
-r535_nvdec_dtor(struct nvkm_engine *engine)
-{
-	struct nvkm_nvdec *nvdec = nvkm_nvdec(engine);
-
-	kfree(nvdec->engine.func);
-	return nvdec;
-}
-
-int
-r535_nvdec_new(const struct nvkm_engine_func *hw, struct nvkm_device *device,
-	       enum nvkm_subdev_type type, int inst, struct nvkm_nvdec **pnvdec)
-{
-	struct nvkm_engine_func *rm;
-	int nclass;
-
-	for (nclass = 0; hw->sclass[nclass].oclass; nclass++);
-
-	if (!(rm = kzalloc(sizeof(*rm) + (nclass + 1) * sizeof(rm->sclass[0]), GFP_KERNEL)))
-		return -ENOMEM;
-
-	rm->dtor = r535_nvdec_dtor;
-	for (int i = 0; i < nclass; i++) {
-		rm->sclass[i].minver = hw->sclass[i].minver;
-		rm->sclass[i].maxver = hw->sclass[i].maxver;
-		rm->sclass[i].oclass = hw->sclass[i].oclass;
-		rm->sclass[i].ctor = r535_nvdec_obj_ctor;
-	}
-
-	if (!(*pnvdec = kzalloc(sizeof(**pnvdec), GFP_KERNEL))) {
-		kfree(rm);
-		return -ENOMEM;
-	}
-
-	return nvkm_engine_ctor(rm, device, type, inst, true, &(*pnvdec)->engine);
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/tu102.c
index 808c8e010b9e..fe95b6e22f21 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/tu102.c
@@ -23,22 +23,12 @@
 
 #include <subdev/gsp.h>
 
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-tu102_nvdec = {
-	.sclass = {
-		{ -1, -1, NVC4B0_VIDEO_DECODER },
-		{}
-	}
-};
-
 int
 tu102_nvdec_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 		struct nvkm_nvdec **pnvdec)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvdec_new(&tu102_nvdec, device, type, inst, pnvdec);
+		return -ENODEV;
 
 	return nvkm_nvdec_new_(gm107_nvdec_fwif, device, type, inst, 0, pnvdec);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild
index 2c1495b730f3..6dcb20d1d156 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild
@@ -2,7 +2,3 @@
 nvkm-y += nvkm/engine/nvenc/base.o
 nvkm-y += nvkm/engine/nvenc/gm107.o
 nvkm-y += nvkm/engine/nvenc/tu102.o
-nvkm-y += nvkm/engine/nvenc/ga102.o
-nvkm-y += nvkm/engine/nvenc/ad102.o
-
-nvkm-y += nvkm/engine/nvenc/r535.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/ga102.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/ga102.c
deleted file mode 100644
index 6463ab8e5871..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/ga102.c
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ga102_nvenc = {
-	.sclass = {
-		{ -1, -1, NVC7B7_VIDEO_ENCODER },
-		{}
-	}
-};
-
-int
-ga102_nvenc_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-		struct nvkm_nvenc **pnvenc)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvenc_new(&ga102_nvenc, device, type, inst, pnvenc);
-
-	return -ENODEV;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/priv.h
index 7917affc6505..b097e3f2867b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/priv.h
@@ -18,7 +18,4 @@ extern const struct nvkm_nvenc_fwif gm107_nvenc_fwif[];
 
 int nvkm_nvenc_new_(const struct nvkm_nvenc_fwif *, struct nvkm_device *, enum nvkm_subdev_type,
 		    int, struct nvkm_nvenc **pnvenc);
-
-int r535_nvenc_new(const struct nvkm_engine_func *, struct nvkm_device *,
-		   enum nvkm_subdev_type, int, struct nvkm_nvenc **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/r535.c
deleted file mode 100644
index c8a2a9196ce5..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/r535.c
+++ /dev/null
@@ -1,110 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <core/object.h>
-#include <subdev/gsp.h>
-#include <engine/fifo.h>
-
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-
-struct r535_nvenc_obj {
-	struct nvkm_object object;
-	struct nvkm_gsp_object rm;
-};
-
-static void *
-r535_nvenc_obj_dtor(struct nvkm_object *object)
-{
-	struct r535_nvenc_obj *obj = container_of(object, typeof(*obj), object);
-
-	nvkm_gsp_rm_free(&obj->rm);
-	return obj;
-}
-
-static const struct nvkm_object_func
-r535_nvenc_obj = {
-	.dtor = r535_nvenc_obj_dtor,
-};
-
-static int
-r535_nvenc_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
-		 struct nvkm_object **pobject)
-{
-	struct nvkm_chan *chan = nvkm_uchan_chan(oclass->parent);
-	struct r535_nvenc_obj *obj;
-	NV_MSENC_ALLOCATION_PARAMETERS *args;
-
-	if (!(obj = kzalloc(sizeof(*obj), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_nvenc_obj, oclass, &obj->object);
-	*pobject = &obj->object;
-
-	args = nvkm_gsp_rm_alloc_get(&chan->rm.object, oclass->handle, oclass->base.oclass,
-				     sizeof(*args), &obj->rm);
-	if (WARN_ON(IS_ERR(args)))
-		return PTR_ERR(args);
-
-	args->size = sizeof(*args);
-	args->engineInstance = oclass->engine->subdev.inst;
-
-	return nvkm_gsp_rm_alloc_wr(&obj->rm, args);
-}
-
-static void *
-r535_nvenc_dtor(struct nvkm_engine *engine)
-{
-	struct nvkm_nvenc *nvenc = nvkm_nvenc(engine);
-
-	kfree(nvenc->engine.func);
-	return nvenc;
-}
-
-int
-r535_nvenc_new(const struct nvkm_engine_func *hw, struct nvkm_device *device,
-	       enum nvkm_subdev_type type, int inst, struct nvkm_nvenc **pnvenc)
-{
-	struct nvkm_engine_func *rm;
-	int nclass;
-
-	for (nclass = 0; hw->sclass[nclass].oclass; nclass++);
-
-	if (!(rm = kzalloc(sizeof(*rm) + (nclass + 1) * sizeof(rm->sclass[0]), GFP_KERNEL)))
-		return -ENOMEM;
-
-	rm->dtor = r535_nvenc_dtor;
-	for (int i = 0; i < nclass; i++) {
-		rm->sclass[i].minver = hw->sclass[i].minver;
-		rm->sclass[i].maxver = hw->sclass[i].maxver;
-		rm->sclass[i].oclass = hw->sclass[i].oclass;
-		rm->sclass[i].ctor = r535_nvenc_obj_ctor;
-	}
-
-	if (!(*pnvenc = kzalloc(sizeof(**pnvenc), GFP_KERNEL))) {
-		kfree(rm);
-		return -ENOMEM;
-	}
-
-	return nvkm_engine_ctor(rm, device, type, inst, true, &(*pnvenc)->engine);
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/tu102.c
index 933864423bb3..8a436b398749 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/tu102.c
@@ -23,22 +23,12 @@
 
 #include <subdev/gsp.h>
 
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-tu102_nvenc = {
-	.sclass = {
-		{ -1, -1, NVC4B7_VIDEO_ENCODER },
-		{}
-	}
-};
-
 int
 tu102_nvenc_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 		struct nvkm_nvenc **pnvenc)
 {
 	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvenc_new(&tu102_nvenc, device, type, inst, pnvenc);
+		return -ENODEV;
 
 	return nvkm_nvenc_new_(gm107_nvenc_fwif, device, type, inst, pnvenc);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/Kbuild
deleted file mode 100644
index 1408f664add6..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/Kbuild
+++ /dev/null
@@ -1,5 +0,0 @@
-# SPDX-License-Identifier: MIT
-nvkm-y += nvkm/engine/nvjpg/ga100.o
-nvkm-y += nvkm/engine/nvjpg/ad102.o
-
-nvkm-y += nvkm/engine/nvjpg/r535.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/ad102.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/ad102.c
deleted file mode 100644
index 62705dc6494c..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/ad102.c
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ad102_nvjpg = {
-	.sclass = {
-		{ -1, -1, NVC9D1_VIDEO_NVJPG },
-		{}
-	}
-};
-
-int
-ad102_nvjpg_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-		struct nvkm_engine **pengine)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvjpg_new(&ad102_nvjpg, device, type, inst, pengine);
-
-	return -ENODEV;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/ga100.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/ga100.c
deleted file mode 100644
index f550eb07da5a..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/ga100.c
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ga100_nvjpg = {
-	.sclass = {
-		{ -1, -1, NVC4D1_VIDEO_NVJPG },
-		{}
-	}
-};
-
-int
-ga100_nvjpg_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-		struct nvkm_engine **pengine)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvjpg_new(&ga100_nvjpg, device, type, inst, pengine);
-
-	return -ENODEV;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/priv.h
deleted file mode 100644
index 1e80cf70033a..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/priv.h
+++ /dev/null
@@ -1,8 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-#ifndef __NVKM_NVJPG_PRIV_H__
-#define __NVKM_NVJPG_PRIV_H__
-#include <engine/nvjpg.h>
-
-int r535_nvjpg_new(const struct nvkm_engine_func *, struct nvkm_device *,
-		   enum nvkm_subdev_type, int, struct nvkm_engine **);
-#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/r535.c
deleted file mode 100644
index 1babddc4eb80..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvjpg/r535.c
+++ /dev/null
@@ -1,107 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <core/object.h>
-#include <subdev/gsp.h>
-#include <engine/fifo.h>
-
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-
-struct r535_nvjpg_obj {
-	struct nvkm_object object;
-	struct nvkm_gsp_object rm;
-};
-
-static void *
-r535_nvjpg_obj_dtor(struct nvkm_object *object)
-{
-	struct r535_nvjpg_obj *obj = container_of(object, typeof(*obj), object);
-
-	nvkm_gsp_rm_free(&obj->rm);
-	return obj;
-}
-
-static const struct nvkm_object_func
-r535_nvjpg_obj = {
-	.dtor = r535_nvjpg_obj_dtor,
-};
-
-static int
-r535_nvjpg_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
-		    struct nvkm_object **pobject)
-{
-	struct nvkm_chan *chan = nvkm_uchan_chan(oclass->parent);
-	struct r535_nvjpg_obj *obj;
-	NV_NVJPG_ALLOCATION_PARAMETERS *args;
-
-	if (!(obj = kzalloc(sizeof(*obj), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_nvjpg_obj, oclass, &obj->object);
-	*pobject = &obj->object;
-
-	args = nvkm_gsp_rm_alloc_get(&chan->rm.object, oclass->handle, oclass->base.oclass,
-				     sizeof(*args), &obj->rm);
-	if (WARN_ON(IS_ERR(args)))
-		return PTR_ERR(args);
-
-	args->size = sizeof(*args);
-	args->engineInstance = oclass->engine->subdev.inst;
-
-	return nvkm_gsp_rm_alloc_wr(&obj->rm, args);
-}
-
-static void *
-r535_nvjpg_dtor(struct nvkm_engine *engine)
-{
-	kfree(engine->func);
-	return engine;
-}
-
-int
-r535_nvjpg_new(const struct nvkm_engine_func *hw, struct nvkm_device *device,
-	       enum nvkm_subdev_type type, int inst, struct nvkm_engine **pengine)
-{
-	struct nvkm_engine_func *rm;
-	int nclass, ret;
-
-	for (nclass = 0; hw->sclass[nclass].oclass; nclass++);
-
-	if (!(rm = kzalloc(sizeof(*rm) + (nclass + 1) * sizeof(rm->sclass[0]), GFP_KERNEL)))
-		return -ENOMEM;
-
-	rm->dtor = r535_nvjpg_dtor;
-	for (int i = 0; i < nclass; i++) {
-		rm->sclass[i].minver = hw->sclass[i].minver;
-		rm->sclass[i].maxver = hw->sclass[i].maxver;
-		rm->sclass[i].oclass = hw->sclass[i].oclass;
-		rm->sclass[i].ctor = r535_nvjpg_obj_ctor;
-	}
-
-	ret = nvkm_engine_new_(rm, device, type, inst, true, pengine);
-	if (ret)
-		kfree(rm);
-
-	return ret;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/ofa/Kbuild
deleted file mode 100644
index 99f1713d7e51..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/Kbuild
+++ /dev/null
@@ -1,6 +0,0 @@
-# SPDX-License-Identifier: MIT
-nvkm-y += nvkm/engine/ofa/ga100.o
-nvkm-y += nvkm/engine/ofa/ga102.o
-nvkm-y += nvkm/engine/ofa/ad102.o
-
-nvkm-y += nvkm/engine/ofa/r535.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ga100.c b/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ga100.c
deleted file mode 100644
index ef474f61a1b5..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ga100.c
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ga100_ofa = {
-	.sclass = {
-		{ -1, -1, NVC6FA_VIDEO_OFA },
-		{}
-	}
-};
-
-int
-ga100_ofa_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-	      struct nvkm_engine **pengine)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_ofa_new(&ga100_ofa, device, type, inst, pengine);
-
-	return -ENODEV;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ga102.c b/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ga102.c
deleted file mode 100644
index bea255529993..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/ga102.c
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <subdev/gsp.h>
-
-#include <nvif/class.h>
-
-static const struct nvkm_engine_func
-ga102_ofa = {
-	.sclass = {
-		{ -1, -1, NVC7FA_VIDEO_OFA },
-		{}
-	}
-};
-
-int
-ga102_ofa_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-	      struct nvkm_engine **pengine)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_ofa_new(&ga102_ofa, device, type, inst, pengine);
-
-	return -ENODEV;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/ofa/priv.h
deleted file mode 100644
index caf29e6bddb4..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/priv.h
+++ /dev/null
@@ -1,8 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-#ifndef __NVKM_OFA_PRIV_H__
-#define __NVKM_OFA_PRIV_H__
-#include <engine/ofa.h>
-
-int r535_ofa_new(const struct nvkm_engine_func *, struct nvkm_device *,
-		 enum nvkm_subdev_type, int, struct nvkm_engine **);
-#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/r535.c b/drivers/gpu/drm/nouveau/nvkm/engine/ofa/r535.c
deleted file mode 100644
index 438dc692eefe..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ofa/r535.c
+++ /dev/null
@@ -1,107 +0,0 @@
-/*
- * Copyright 2023 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- */
-#include "priv.h"
-
-#include <core/object.h>
-#include <subdev/gsp.h>
-#include <subdev/mmu.h>
-#include <engine/fifo.h>
-
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-
-struct r535_ofa_obj {
-	struct nvkm_object object;
-	struct nvkm_gsp_object rm;
-};
-
-static void *
-r535_ofa_obj_dtor(struct nvkm_object *object)
-{
-	struct r535_ofa_obj *obj = container_of(object, typeof(*obj), object);
-
-	nvkm_gsp_rm_free(&obj->rm);
-	return obj;
-}
-
-static const struct nvkm_object_func
-r535_ofa_obj = {
-	.dtor = r535_ofa_obj_dtor,
-};
-
-static int
-r535_ofa_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
-		 struct nvkm_object **pobject)
-{
-	struct nvkm_chan *chan = nvkm_uchan_chan(oclass->parent);
-	struct r535_ofa_obj *obj;
-	NV_OFA_ALLOCATION_PARAMETERS *args;
-
-	if (!(obj = kzalloc(sizeof(*obj), GFP_KERNEL)))
-		return -ENOMEM;
-
-	nvkm_object_ctor(&r535_ofa_obj, oclass, &obj->object);
-	*pobject = &obj->object;
-
-	args = nvkm_gsp_rm_alloc_get(&chan->rm.object, oclass->handle, oclass->base.oclass,
-				     sizeof(*args), &obj->rm);
-	if (WARN_ON(IS_ERR(args)))
-		return PTR_ERR(args);
-
-	args->size = sizeof(*args);
-
-	return nvkm_gsp_rm_alloc_wr(&obj->rm, args);
-}
-
-static void *
-r535_ofa_dtor(struct nvkm_engine *engine)
-{
-	kfree(engine->func);
-	return engine;
-}
-
-int
-r535_ofa_new(const struct nvkm_engine_func *hw, struct nvkm_device *device,
-	     enum nvkm_subdev_type type, int inst, struct nvkm_engine **pengine)
-{
-	struct nvkm_engine_func *rm;
-	int nclass, ret;
-
-	for (nclass = 0; hw->sclass[nclass].oclass; nclass++);
-
-	if (!(rm = kzalloc(sizeof(*rm) + (nclass + 1) * sizeof(rm->sclass[0]), GFP_KERNEL)))
-		return -ENOMEM;
-
-	rm->dtor = r535_ofa_dtor;
-	for (int i = 0; i < nclass; i++) {
-		rm->sclass[i].minver = hw->sclass[i].minver;
-		rm->sclass[i].maxver = hw->sclass[i].maxver;
-		rm->sclass[i].oclass = hw->sclass[i].oclass;
-		rm->sclass[i].ctor = r535_ofa_obj_ctor;
-	}
-
-	ret = nvkm_engine_new_(rm, device, type, inst, true, pengine);
-	if (ret)
-		kfree(rm);
-
-	return ret;
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
index 4c2f6fc4ef58..c19ea4ea9bd3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
@@ -9,6 +9,7 @@ include $(src)/nvkm/subdev/fault/Kbuild
 include $(src)/nvkm/subdev/fb/Kbuild
 include $(src)/nvkm/subdev/fuse/Kbuild
 include $(src)/nvkm/subdev/gpio/Kbuild
+include $(src)/nvkm/subdev/fsp/Kbuild
 include $(src)/nvkm/subdev/gsp/Kbuild
 include $(src)/nvkm/subdev/i2c/Kbuild
 include $(src)/nvkm/subdev/iccsense/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild
index 9754c6872543..8faee3317a74 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild
@@ -7,5 +7,3 @@ nvkm-y += nvkm/subdev/bar/gk20a.o
 nvkm-y += nvkm/subdev/bar/gm107.o
 nvkm-y += nvkm/subdev/bar/gm20b.o
 nvkm-y += nvkm/subdev/bar/tu102.o
-
-nvkm-y += nvkm/subdev/bar/r535.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
index 51070b7dda85..e5e60915029c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/gf100.c
@@ -82,7 +82,7 @@ gf100_bar_bar2_init(struct nvkm_bar *base)
 
 static int
 gf100_bar_oneinit_bar(struct gf100_bar *bar, struct gf100_barN *bar_vm,
-		      struct lock_class_key *key, int bar_nr)
+		      struct lock_class_key *key, enum nvkm_bar_id bar_id)
 {
 	struct nvkm_device *device = bar->base.subdev.device;
 	resource_size_t bar_len;
@@ -93,14 +93,14 @@ gf100_bar_oneinit_bar(struct gf100_bar *bar, struct gf100_barN *bar_vm,
 	if (ret)
 		return ret;
 
-	bar_len = device->func->resource_size(device, bar_nr);
+	bar_len = device->func->resource_size(device, bar_id);
 	if (!bar_len)
 		return -ENOMEM;
-	if (bar_nr == 3 && bar->bar2_halve)
+	if (bar_id == NVKM_BAR2_INST && bar->bar2_halve)
 		bar_len >>= 1;
 
 	ret = nvkm_vmm_new(device, 0, bar_len, NULL, 0, key,
-			   (bar_nr == 3) ? "bar2" : "bar1", &bar_vm->vmm);
+			   (bar_id == NVKM_BAR2_INST) ? "bar2" : "bar1", &bar_vm->vmm);
 	if (ret)
 		return ret;
 
@@ -110,7 +110,7 @@ gf100_bar_oneinit_bar(struct gf100_bar *bar, struct gf100_barN *bar_vm,
 	/*
 	 * Bootstrap page table lookup.
 	 */
-	if (bar_nr == 3) {
+	if (bar_id == NVKM_BAR2_INST) {
 		ret = nvkm_vmm_boot(bar_vm->vmm);
 		if (ret)
 			return ret;
@@ -129,7 +129,7 @@ gf100_bar_oneinit(struct nvkm_bar *base)
 
 	/* BAR2 */
 	if (bar->base.func->bar2.init) {
-		ret = gf100_bar_oneinit_bar(bar, &bar->bar[0], &bar2_lock, 3);
+		ret = gf100_bar_oneinit_bar(bar, &bar->bar[0], &bar2_lock, NVKM_BAR2_INST);
 		if (ret)
 			return ret;
 
@@ -138,7 +138,7 @@ gf100_bar_oneinit(struct nvkm_bar *base)
 	}
 
 	/* BAR1 */
-	ret = gf100_bar_oneinit_bar(bar, &bar->bar[1], &bar1_lock, 1);
+	ret = gf100_bar_oneinit_bar(bar, &bar->bar[1], &bar1_lock, NVKM_BAR1_FB);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c
index 27d8a1be43e4..6a881becb02c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/nv50.c
@@ -127,7 +127,7 @@ nv50_bar_oneinit(struct nvkm_bar *base)
 
 	/* BAR2 */
 	start = 0x0100000000ULL;
-	size = device->func->resource_size(device, 3);
+	size = device->func->resource_size(device, NVKM_BAR2_INST);
 	if (!size)
 		return -ENOMEM;
 	limit = start + size;
@@ -167,7 +167,7 @@ nv50_bar_oneinit(struct nvkm_bar *base)
 
 	/* BAR1 */
 	start = 0x0000000000ULL;
-	size = device->func->resource_size(device, 1);
+	size = device->func->resource_size(device, NVKM_BAR1_FB);
 	if (!size)
 		return -ENOMEM;
 	limit = start + size;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/fbmem.h b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/fbmem.h
index 6c5bbff12eb4..b918e22df5a8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/fbmem.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/fbmem.h
@@ -47,8 +47,8 @@
 static inline struct io_mapping *
 fbmem_init(struct nvkm_device *dev)
 {
-	return io_mapping_create_wc(dev->func->resource_addr(dev, 1),
-				    dev->func->resource_size(dev, 1));
+	return io_mapping_create_wc(dev->func->resource_addr(dev, NVKM_BAR1_FB),
+				    dev->func->resource_size(dev, NVKM_BAR1_FB));
 }
 
 static inline void
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c
index c123e5893d76..cd2fbc0472d8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/user.c
@@ -50,7 +50,7 @@ nvkm_ufault_map(struct nvkm_object *object, void *argv, u32 argc,
 	struct nvkm_fault_buffer *buffer = nvkm_fault_buffer(object);
 	struct nvkm_device *device = buffer->fault->subdev.device;
 	*type = NVKM_OBJECT_MAP_IO;
-	*addr = device->func->resource_addr(device, 3) + buffer->addr;
+	*addr = device->func->resource_addr(device, NVKM_BAR2_INST) + buffer->addr;
 	*size = nvkm_memory_size(buffer->mem);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild
index d1611ad3bf81..8d8a5382d1b1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/Kbuild
@@ -35,6 +35,9 @@ nvkm-y += nvkm/subdev/fb/gv100.o
 nvkm-y += nvkm/subdev/fb/tu102.o
 nvkm-y += nvkm/subdev/fb/ga100.o
 nvkm-y += nvkm/subdev/fb/ga102.o
+nvkm-y += nvkm/subdev/fb/gh100.o
+nvkm-y += nvkm/subdev/fb/gb100.o
+nvkm-y += nvkm/subdev/fb/gb202.o
 
 nvkm-y += nvkm/subdev/fb/r535.o
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ga102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ga102.c
index 25f82b372bca..2819780050d8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ga102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ga102.c
@@ -25,7 +25,7 @@
 #include <subdev/gsp.h>
 #include <engine/nvdec.h>
 
-static u64
+u64
 ga102_fb_vidmem_size(struct nvkm_fb *fb)
 {
 	return (u64)nvkm_rd32(fb->subdev.device, 0x1183a4) << 20;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gb100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gb100.c
new file mode 100644
index 000000000000..1c78c8853617
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gb100.c
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gb100/dev_hshub_base.h>
+
+static void
+gb100_fb_sysmem_flush_page_init(struct nvkm_fb *fb)
+{
+	const u32 addr_hi = upper_32_bits(fb->sysmem.flush_page_addr);
+	const u32 addr_lo = lower_32_bits(fb->sysmem.flush_page_addr);
+	const u32 hshub = DRF_LO(NV_PFB_HSHUB0);
+	struct nvkm_device *device = fb->subdev.device;
+
+	nvkm_wr32(device, hshub + NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_HI, addr_hi);
+	nvkm_wr32(device, hshub + NV_PFB_HSHUB_PCIE_FLUSH_SYSMEM_ADDR_LO, addr_lo);
+	nvkm_wr32(device, hshub + NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_HI, addr_hi);
+	nvkm_wr32(device, hshub + NV_PFB_HSHUB_EG_PCIE_FLUSH_SYSMEM_ADDR_LO, addr_lo);
+}
+
+static const struct nvkm_fb_func
+gb100_fb = {
+	.sysmem.flush_page_init = gb100_fb_sysmem_flush_page_init,
+	.vidmem.size = ga102_fb_vidmem_size,
+};
+
+int
+gb100_fb_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_fb **pfb)
+{
+	return r535_fb_new(&gb100_fb, device, type, inst, pfb);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gb202.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gb202.c
new file mode 100644
index 000000000000..848505026d02
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gb202.c
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gb10b/dev_fbhub.h>
+
+static void
+gb202_fb_sysmem_flush_page_init(struct nvkm_fb *fb)
+{
+	struct nvkm_device *device = fb->subdev.device;
+	const u64 addr = fb->sysmem.flush_page_addr;
+
+	nvkm_wr32(device, NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_HI, upper_32_bits(addr));
+	nvkm_wr32(device, NV_PFB_FBHUB0_PCIE_FLUSH_SYSMEM_ADDR_LO, lower_32_bits(addr));
+}
+
+static const struct nvkm_fb_func
+gb202_fb = {
+	.sysmem.flush_page_init = gb202_fb_sysmem_flush_page_init,
+	.vidmem.size = ga102_fb_vidmem_size,
+};
+
+int
+gb202_fb_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_fb **pfb)
+{
+	return r535_fb_new(&gb202_fb, device, type, inst, pfb);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gh100.c
new file mode 100644
index 000000000000..2d8c51f882d5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gh100.c
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gh100/dev_fb.h>
+
+static void
+gh100_fb_sysmem_flush_page_init(struct nvkm_fb *fb)
+{
+	const u64 addr = fb->sysmem.flush_page_addr >> NV_PFB_NISO_FLUSH_SYSMEM_ADDR_SHIFT;
+	struct nvkm_device *device = fb->subdev.device;
+
+	nvkm_wr32(device, NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_HI, upper_32_bits(addr));
+	nvkm_wr32(device, NV_PFB_FBHUB_PCIE_FLUSH_SYSMEM_ADDR_LO, lower_32_bits(addr));
+}
+
+static const struct nvkm_fb_func
+gh100_fb = {
+	.sysmem.flush_page_init = gh100_fb_sysmem_flush_page_init,
+	.vidmem.size = ga102_fb_vidmem_size,
+};
+
+int
+gh100_fb_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_fb **pfb)
+{
+	return r535_fb_new(&gh100_fb, device, type, inst, pfb);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h
index 35c55dfba23d..ebe996503ab2 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/priv.h
@@ -98,4 +98,6 @@ int gp102_fb_vpr_scrub(struct nvkm_fb *);
 int gv100_fb_init_page(struct nvkm_fb *);
 
 bool tu102_fb_vpr_scrub_required(struct nvkm_fb *);
+
+u64 ga102_fb_vidmem_size(struct nvkm_fb *);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/Kbuild
new file mode 100644
index 000000000000..1a9ded3a86f8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/Kbuild
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: MIT
+#
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+
+nvkm-y += nvkm/subdev/fsp/base.o
+nvkm-y += nvkm/subdev/fsp/gh100.o
+nvkm-y += nvkm/subdev/fsp/gb100.o
+nvkm-y += nvkm/subdev/fsp/gb202.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/base.c
new file mode 100644
index 000000000000..e366a980baa9
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/base.c
@@ -0,0 +1,66 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+int
+nvkm_fsp_boot_gsp_fmc(struct nvkm_fsp *fsp, u64 args_addr, u32 rsvd_size, bool resume,
+		      u64 img_addr, const u8 *hash, const u8 *pkey, const u8 *sig)
+{
+	return fsp->func->cot.boot_gsp_fmc(fsp, args_addr, rsvd_size, resume,
+					   img_addr, hash, pkey, sig);
+}
+
+bool
+nvkm_fsp_verify_gsp_fmc(struct nvkm_fsp *fsp, u32 hash_size, u32 pkey_size, u32 sig_size)
+{
+	return hash_size == fsp->func->cot.size_hash &&
+	       pkey_size == fsp->func->cot.size_pkey &&
+	        sig_size == fsp->func->cot.size_sig;
+}
+
+static int
+nvkm_fsp_preinit(struct nvkm_subdev *subdev)
+{
+	struct nvkm_fsp *fsp = nvkm_fsp(subdev);
+
+	return fsp->func->wait_secure_boot(fsp);
+}
+
+static void *
+nvkm_fsp_dtor(struct nvkm_subdev *subdev)
+{
+	struct nvkm_fsp *fsp = nvkm_fsp(subdev);
+
+	nvkm_falcon_dtor(&fsp->falcon);
+	return fsp;
+}
+
+static const struct nvkm_falcon_func
+nvkm_fsp_flcn = {
+	.emem_pio = &gp102_flcn_emem_pio,
+};
+
+static const struct nvkm_subdev_func
+nvkm_fsp = {
+	.dtor = nvkm_fsp_dtor,
+	.preinit = nvkm_fsp_preinit,
+};
+
+int
+nvkm_fsp_new_(const struct nvkm_fsp_func *func,
+	      struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+	      struct nvkm_fsp **pfsp)
+{
+	struct nvkm_fsp *fsp;
+
+	fsp = *pfsp = kzalloc(sizeof(*fsp), GFP_KERNEL);
+	if (!fsp)
+		return -ENOMEM;
+
+	fsp->func = func;
+	nvkm_subdev_ctor(&nvkm_fsp, device, type, inst, &fsp->subdev);
+
+	return nvkm_falcon_ctor(&nvkm_fsp_flcn, &fsp->subdev, "fsp", 0x8f2000, &fsp->falcon);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gb100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gb100.c
new file mode 100644
index 000000000000..e06636bf54b6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gb100.c
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+static const struct nvkm_fsp_func
+gb100_fsp = {
+	.wait_secure_boot = gh100_fsp_wait_secure_boot,
+	.cot = {
+		.version = 2,
+		.size_hash = 48,
+		.size_pkey = 97,
+		.size_sig = 96,
+		.boot_gsp_fmc = gh100_fsp_boot_gsp_fmc,
+	},
+};
+
+int
+gb100_fsp_new(struct nvkm_device *device,
+	      enum nvkm_subdev_type type, int inst, struct nvkm_fsp **pfsp)
+{
+	return nvkm_fsp_new_(&gb100_fsp, device, type, inst, pfsp);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gb202.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gb202.c
new file mode 100644
index 000000000000..3438aac6383e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gb202.c
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gb202/dev_therm.h>
+
+static int
+gb202_fsp_wait_secure_boot(struct nvkm_fsp *fsp)
+{
+	struct nvkm_device *device = fsp->subdev.device;
+	unsigned timeout_ms = 4000;
+
+	do {
+		u32 status = NVKM_RD32(device, NV_THERM, I2CS_SCRATCH, FSP_BOOT_COMPLETE_STATUS);
+
+		if (status == NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS_SUCCESS)
+			return 0;
+
+		usleep_range(1000, 2000);
+	} while (timeout_ms--);
+
+	return -ETIMEDOUT;
+}
+
+static const struct nvkm_fsp_func
+gb202_fsp = {
+	.wait_secure_boot = gb202_fsp_wait_secure_boot,
+	.cot = {
+		.version = 2,
+		.size_hash = 48,
+		.size_pkey = 97,
+		.size_sig = 96,
+		.boot_gsp_fmc = gh100_fsp_boot_gsp_fmc,
+	},
+};
+
+int
+gb202_fsp_new(struct nvkm_device *device,
+	      enum nvkm_subdev_type type, int inst, struct nvkm_fsp **pfsp)
+{
+	return nvkm_fsp_new_(&gb202_fsp, device, type, inst, pfsp);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gh100.c
new file mode 100644
index 000000000000..2815be4bf5de
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/gh100.c
@@ -0,0 +1,275 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gh100/dev_fsp_pri.h>
+#include <nvhw/ref/gh100/dev_therm.h>
+
+#include <nvrm/nvtypes.h>
+
+#define MCTP_HEADER_VERSION          3:0
+#define MCTP_HEADER_RSVD             7:4
+
+#define MCTP_HEADER_DEID            15:8
+#define MCTP_HEADER_SEID            23:16
+
+#define MCTP_HEADER_TAG             26:24
+#define MCTP_HEADER_TO              27:27
+#define MCTP_HEADER_SEQ             29:28
+#define MCTP_HEADER_EOM             30:30
+#define MCTP_HEADER_SOM             31:31
+
+#define MCTP_MSG_HEADER_TYPE         6:0
+#define MCTP_MSG_HEADER_IC           7:7
+
+#define MCTP_MSG_HEADER_VENDOR_ID   23:8
+#define MCTP_MSG_HEADER_NVDM_TYPE   31:24
+
+#define MCTP_MSG_HEADER_TYPE_VENDOR_PCI 0x7e
+#define MCTP_MSG_HEADER_VENDOR_ID_NV    0x10de
+
+#define NVDM_TYPE_COT                   0x14
+#define NVDM_TYPE_FSP_RESPONSE          0x15
+
+#pragma pack(1)
+typedef struct nvdm_payload_cot
+{
+    NvU16 version;
+    NvU16 size;
+    NvU64 gspFmcSysmemOffset;
+    NvU64 frtsSysmemOffset;
+    NvU32 frtsSysmemSize;
+
+    // Note this is an offset from the end of FB
+    NvU64 frtsVidmemOffset;
+    NvU32 frtsVidmemSize;
+
+    // Authentication related fields
+    NvU32 hash384[12];
+    NvU32 publicKey[96];
+    NvU32 signature[96];
+
+    NvU64 gspBootArgsSysmemOffset;
+} NVDM_PAYLOAD_COT;
+#pragma pack()
+
+#pragma pack(1)
+typedef struct
+{
+    NvU32 taskId;
+    NvU32 commandNvdmType;
+    NvU32 errorCode;
+} NVDM_PAYLOAD_COMMAND_RESPONSE;
+#pragma pack()
+
+static u32
+gh100_fsp_poll(struct nvkm_fsp *fsp)
+{
+	struct nvkm_device *device = fsp->subdev.device;
+	u32 head, tail;
+
+	head = nvkm_rd32(device, NV_PFSP_MSGQ_HEAD(0));
+	tail = nvkm_rd32(device, NV_PFSP_MSGQ_TAIL(0));
+
+	if (head == tail)
+		return 0;
+
+	return (tail - head) + sizeof(u32); /* TAIL points at last DWORD written. */
+}
+
+static int
+gh100_fsp_recv(struct nvkm_fsp *fsp, u8 *packet, u32 max_packet_size)
+{
+	struct nvkm_device *device = fsp->subdev.device;
+	u32 packet_size;
+	int ret;
+
+	packet_size = gh100_fsp_poll(fsp);
+	if (!packet_size || WARN_ON(packet_size % 4 || packet_size > max_packet_size))
+		return -EINVAL;
+
+	ret = nvkm_falcon_pio_rd(&fsp->falcon, 0, EMEM, 0, packet, 0, packet_size);
+	if (ret)
+		return ret;
+
+	nvkm_wr32(device, NV_PFSP_MSGQ_TAIL(0), 0);
+	nvkm_wr32(device, NV_PFSP_MSGQ_HEAD(0), 0);
+
+	return packet_size;
+}
+
+static int
+gh100_fsp_wait(struct nvkm_fsp *fsp)
+{
+	int time = 1000;
+
+	do {
+		if (gh100_fsp_poll(fsp))
+			return 0;
+
+		usleep_range(1000, 2000);
+	} while(time--);
+
+	return -ETIMEDOUT;
+}
+
+static int
+gh100_fsp_send(struct nvkm_fsp *fsp, const u8 *packet, u32 packet_size)
+{
+	struct nvkm_device *device = fsp->subdev.device;
+	int time = 1000, ret;
+
+	if (WARN_ON(packet_size % sizeof(u32)))
+		return -EINVAL;
+
+	/* Ensure any previously sent message has been consumed. */
+	do {
+		u32 head = nvkm_rd32(device, NV_PFSP_QUEUE_HEAD(0));
+		u32 tail = nvkm_rd32(device, NV_PFSP_QUEUE_TAIL(0));
+
+		if (tail == head)
+			break;
+
+		usleep_range(1000, 2000);
+	} while(time--);
+
+	if (time < 0)
+		return -ETIMEDOUT;
+
+	/* Write message to EMEM. */
+	ret = nvkm_falcon_pio_wr(&fsp->falcon, packet, 0, 0, EMEM, 0, packet_size, 0, false);
+	if (ret)
+		return ret;
+
+	/* Update queue pointers - TAIL points at last DWORD written. */
+	nvkm_wr32(device, NV_PFSP_QUEUE_TAIL(0), packet_size - sizeof(u32));
+	nvkm_wr32(device, NV_PFSP_QUEUE_HEAD(0), 0);
+	return 0;
+}
+
+static int
+gh100_fsp_send_sync(struct nvkm_fsp *fsp, u8 nvdm_type, const u8 *packet, u32 packet_size)
+{
+	struct nvkm_subdev *subdev = &fsp->subdev;
+	struct {
+		u32 mctp_header;
+		u32 nvdm_header;
+		NVDM_PAYLOAD_COMMAND_RESPONSE response;
+	} reply;
+	int ret;
+
+	ret = gh100_fsp_send(fsp, packet, packet_size);
+	if (ret)
+		return ret;
+
+	ret = gh100_fsp_wait(fsp);
+	if (ret)
+		return ret;
+
+	ret = gh100_fsp_recv(fsp, (u8 *)&reply, sizeof(reply));
+	if (ret < 0)
+		return ret;
+
+	if (NVVAL_TEST(reply.mctp_header, MCTP, HEADER, SOM, !=, 1) ||
+	    NVVAL_TEST(reply.mctp_header, MCTP, HEADER, EOM, !=, 1)) {
+		nvkm_error(subdev, "unexpected MCTP header in reply: 0x%08x\n", reply.mctp_header);
+		return -EIO;
+	}
+
+	if (NVDEF_TEST(reply.nvdm_header, MCTP, MSG_HEADER, TYPE, !=, VENDOR_PCI) ||
+	    NVDEF_TEST(reply.nvdm_header, MCTP, MSG_HEADER, VENDOR_ID, !=, NV) ||
+	    NVVAL_TEST(reply.nvdm_header, MCTP, MSG_HEADER, NVDM_TYPE, !=, NVDM_TYPE_FSP_RESPONSE)) {
+		nvkm_error(subdev, "unexpected NVDM header in reply: 0x%08x\n", reply.nvdm_header);
+		return -EIO;
+	}
+
+	if (reply.response.commandNvdmType != nvdm_type) {
+		nvkm_error(subdev, "expected NVDM type 0x%02x in reply, got 0x%02x\n",
+			   nvdm_type, reply.response.commandNvdmType);
+		return -EIO;
+	}
+
+	if (reply.response.errorCode) {
+		nvkm_error(subdev, "NVDM command 0x%02x failed with error 0x%08x\n",
+			   nvdm_type, reply.response.errorCode);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+int
+gh100_fsp_boot_gsp_fmc(struct nvkm_fsp *fsp, u64 args_addr, u32 rsvd_size, bool resume,
+		       u64 img_addr, const u8 *hash, const u8 *pkey, const u8 *sig)
+{
+	struct {
+		u32 mctp_header;
+		u32 nvdm_header;
+		NVDM_PAYLOAD_COT cot;
+	} msg = {};
+
+	msg.mctp_header = NVVAL(MCTP, HEADER, SOM, 1) |
+			  NVVAL(MCTP, HEADER, EOM, 1) |
+			  NVVAL(MCTP, HEADER, SEID, 0) |
+			  NVVAL(MCTP, HEADER, SEQ, 0);
+
+	msg.nvdm_header = NVDEF(MCTP, MSG_HEADER, TYPE, VENDOR_PCI) |
+			  NVDEF(MCTP, MSG_HEADER, VENDOR_ID, NV) |
+			  NVVAL(MCTP, MSG_HEADER, NVDM_TYPE, NVDM_TYPE_COT);
+
+	msg.cot.version = fsp->func->cot.version;
+	msg.cot.size = sizeof(msg.cot);
+	msg.cot.gspFmcSysmemOffset = img_addr;
+	if (!resume) {
+		msg.cot.frtsVidmemOffset = ALIGN(rsvd_size, 0x200000);
+		msg.cot.frtsVidmemSize = 0x100000;
+	}
+
+	memcpy(msg.cot.hash384, hash, fsp->func->cot.size_hash);
+	memcpy(msg.cot.publicKey, pkey, fsp->func->cot.size_pkey);
+	memcpy(msg.cot.signature, sig, fsp->func->cot.size_sig);
+
+	msg.cot.gspBootArgsSysmemOffset = args_addr;
+
+	return gh100_fsp_send_sync(fsp, NVDM_TYPE_COT, (const u8 *)&msg, sizeof(msg));
+}
+
+int
+gh100_fsp_wait_secure_boot(struct nvkm_fsp *fsp)
+{
+	struct nvkm_device *device = fsp->subdev.device;
+	unsigned timeout_ms = 4000;
+
+	do {
+		u32 status = NVKM_RD32(device, NV_THERM, I2CS_SCRATCH, FSP_BOOT_COMPLETE_STATUS);
+
+		if (status == NV_THERM_I2CS_SCRATCH_FSP_BOOT_COMPLETE_STATUS_SUCCESS)
+			return 0;
+
+		usleep_range(1000, 2000);
+	} while (timeout_ms--);
+
+	return -ETIMEDOUT;
+}
+
+static const struct nvkm_fsp_func
+gh100_fsp = {
+	.wait_secure_boot = gh100_fsp_wait_secure_boot,
+	.cot = {
+		.version = 1,
+		.size_hash = 48,
+		.size_pkey = 384,
+		.size_sig = 384,
+		.boot_gsp_fmc = gh100_fsp_boot_gsp_fmc,
+	},
+};
+
+int
+gh100_fsp_new(struct nvkm_device *device,
+	      enum nvkm_subdev_type type, int inst, struct nvkm_fsp **pfsp)
+{
+	return nvkm_fsp_new_(&gh100_fsp, device, type, inst, pfsp);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/priv.h
new file mode 100644
index 000000000000..f0b2c605c33d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fsp/priv.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVKM_FSP_PRIV_H__
+#define __NVKM_FSP_PRIV_H__
+#define nvkm_fsp(p) container_of((p), struct nvkm_fsp, subdev)
+#include <subdev/fsp.h>
+
+struct nvkm_fsp_func {
+	int (*wait_secure_boot)(struct nvkm_fsp *);
+
+	struct {
+		u32 version;
+		u32 size_hash;
+		u32 size_pkey;
+		u32 size_sig;
+		int (*boot_gsp_fmc)(struct nvkm_fsp *, u64 args_addr, u32 rsvd_size, bool resume,
+				    u64 img_addr, const u8 *hash, const u8 *pkey, const u8 *sig);
+	} cot;
+};
+
+int nvkm_fsp_new_(const struct nvkm_fsp_func *,
+		  struct nvkm_device *, enum nvkm_subdev_type, int inst, struct nvkm_fsp **);
+
+int gh100_fsp_wait_secure_boot(struct nvkm_fsp *);
+int gh100_fsp_boot_gsp_fmc(struct nvkm_fsp *, u64 args_addr, u32 rsvd_size, bool resume,
+			   u64 img_addr, const u8 *hash, const u8 *pkey, const u8 *sig);
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild
index 16bf2f1bb780..e9c948b67bbd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild
@@ -7,6 +7,9 @@ nvkm-y += nvkm/subdev/gsp/tu102.o
 nvkm-y += nvkm/subdev/gsp/tu116.o
 nvkm-y += nvkm/subdev/gsp/ga100.o
 nvkm-y += nvkm/subdev/gsp/ga102.o
+nvkm-y += nvkm/subdev/gsp/gh100.o
 nvkm-y += nvkm/subdev/gsp/ad102.o
+nvkm-y += nvkm/subdev/gsp/gb100.o
+nvkm-y += nvkm/subdev/gsp/gb202.o
 
-nvkm-y += nvkm/subdev/gsp/r535.o
+include $(src)/nvkm/subdev/gsp/rm/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c
index c849c6299c52..eb765da0876e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c
@@ -22,30 +22,27 @@
 #include "priv.h"
 
 static const struct nvkm_gsp_func
-ad102_gsp_r535_113_01 = {
+ad102_gsp = {
 	.flcn = &ga102_gsp_flcn,
 	.fwsec = &ga102_gsp_fwsec,
 
 	.sig_section = ".fwsignature_ad10x",
 
-	.wpr_heap.os_carveout_size = 20 << 20,
-	.wpr_heap.base_size = 8 << 20,
-	.wpr_heap.min_size = 84 << 20,
-
 	.booter.ctor = ga102_gsp_booter_ctor,
 
 	.dtor = r535_gsp_dtor,
 	.oneinit = tu102_gsp_oneinit,
-	.init = r535_gsp_init,
-	.fini = r535_gsp_fini,
+	.init = tu102_gsp_init,
+	.fini = tu102_gsp_fini,
 	.reset = ga102_gsp_reset,
 
-	.rm = &r535_gsp_rm,
+	.rm.gpu = &ad10x_gpu,
 };
 
 static struct nvkm_gsp_fwif
 ad102_gsps[] = {
-	{ 0, r535_gsp_load, &ad102_gsp_r535_113_01, "535.113.01", true },
+	{ 1, tu102_gsp_load, &ad102_gsp, &r570_rm_ga102, "570.144", true },
+	{ 0, tu102_gsp_load, &ad102_gsp, &r535_rm_ga102, "535.113.01", true },
 	{}
 };
 
@@ -55,3 +52,15 @@ ad102_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 {
 	return nvkm_gsp_new_(ad102_gsps, device, type, inst, pgsp);
 }
+
+NVKM_GSP_FIRMWARE_BOOTER(ad102, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ad103, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ad104, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ad106, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ad107, 535.113.01);
+
+NVKM_GSP_FIRMWARE_BOOTER(ad102, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ad103, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ad104, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ad106, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ad107, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c
index da1bebb896f7..d23243a83a4c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c
@@ -52,7 +52,7 @@ nvkm_gsp_fini(struct nvkm_subdev *subdev, bool suspend)
 {
 	struct nvkm_gsp *gsp = nvkm_gsp(subdev);
 
-	if (!gsp->func->fini)
+	if (!gsp->func->fini || !gsp->running)
 		return 0;
 
 	return gsp->func->fini(gsp, suspend);
@@ -80,6 +80,21 @@ nvkm_gsp_oneinit(struct nvkm_subdev *subdev)
 	return gsp->func->oneinit(gsp);
 }
 
+void
+nvkm_gsp_dtor_fws(struct nvkm_gsp *gsp)
+{
+	nvkm_firmware_put(gsp->fws.fmc);
+	gsp->fws.fmc = NULL;
+	nvkm_firmware_put(gsp->fws.bl);
+	gsp->fws.bl = NULL;
+	nvkm_firmware_put(gsp->fws.booter.unload);
+	gsp->fws.booter.unload = NULL;
+	nvkm_firmware_put(gsp->fws.booter.load);
+	gsp->fws.booter.load = NULL;
+	nvkm_firmware_put(gsp->fws.rm);
+	gsp->fws.rm = NULL;
+}
+
 static void *
 nvkm_gsp_dtor(struct nvkm_subdev *subdev)
 {
@@ -89,6 +104,7 @@ nvkm_gsp_dtor(struct nvkm_subdev *subdev)
 		gsp->func->dtor(gsp);
 
 	nvkm_falcon_dtor(&gsp->falcon);
+	kfree(gsp->rm);
 	return gsp;
 }
 
@@ -101,6 +117,16 @@ nvkm_gsp = {
 };
 
 int
+nvkm_gsp_load_fw(struct nvkm_gsp *gsp, const char *name, const char *ver,
+		 const struct firmware **pfw)
+{
+	char fwname[64];
+
+	snprintf(fwname, sizeof(fwname), "gsp/%s-%s", name, ver);
+	return nvkm_firmware_get(&gsp->subdev, fwname, 0, pfw);
+}
+
+int
 nvkm_gsp_new_(const struct nvkm_gsp_fwif *fwif, struct nvkm_device *device,
 	      enum nvkm_subdev_type type, int inst, struct nvkm_gsp **pgsp)
 {
@@ -116,7 +142,19 @@ nvkm_gsp_new_(const struct nvkm_gsp_fwif *fwif, struct nvkm_device *device,
 		return PTR_ERR(fwif);
 
 	gsp->func = fwif->func;
-	gsp->rm = gsp->func->rm;
+
+	if (fwif->rm) {
+		nvkm_info(&gsp->subdev, "RM version: %s\n", fwif->ver);
+
+		gsp->rm = kzalloc(sizeof(*gsp->rm), GFP_KERNEL);
+		if (!gsp->rm)
+			return -ENOMEM;
+
+		gsp->rm->device = device;
+		gsp->rm->gpu = fwif->func->rm.gpu;
+		gsp->rm->wpr = fwif->rm->wpr;
+		gsp->rm->api = fwif->rm->api;
+	}
 
 	return nvkm_falcon_ctor(gsp->func->flcn, &gsp->subdev, gsp->subdev.name, 0x110000,
 				&gsp->falcon);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga100.c
index 223f68b532ef..d201e8697226 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga100.c
@@ -39,29 +39,27 @@ ga100_gsp_flcn = {
 };
 
 static const struct nvkm_gsp_func
-ga100_gsp_r535_113_01 = {
+ga100_gsp = {
 	.flcn = &ga100_gsp_flcn,
 	.fwsec = &tu102_gsp_fwsec,
 
 	.sig_section = ".fwsignature_ga100",
 
-	.wpr_heap.base_size = 8 << 20,
-	.wpr_heap.min_size = 64 << 20,
-
 	.booter.ctor = tu102_gsp_booter_ctor,
 
 	.dtor = r535_gsp_dtor,
 	.oneinit = tu102_gsp_oneinit,
-	.init = r535_gsp_init,
-	.fini = r535_gsp_fini,
+	.init = tu102_gsp_init,
+	.fini = tu102_gsp_fini,
 	.reset = tu102_gsp_reset,
 
-	.rm = &r535_gsp_rm,
+	.rm.gpu = &ga100_gpu,
 };
 
 static struct nvkm_gsp_fwif
 ga100_gsps[] = {
-	{  0,  r535_gsp_load, &ga100_gsp_r535_113_01, "535.113.01" },
+	{  1, tu102_gsp_load, &ga100_gsp, &r570_rm_tu102, "570.144" },
+	{  0, tu102_gsp_load, &ga100_gsp, &r535_rm_tu102, "535.113.01" },
 	{ -1, gv100_gsp_nofw, &gv100_gsp },
 	{}
 };
@@ -72,3 +70,6 @@ ga100_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 {
 	return nvkm_gsp_new_(ga100_gsps, device, type, inst, pgsp);
 }
+
+NVKM_GSP_FIRMWARE_BOOTER(ga100, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ga100, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c
index 4c4b4168a266..917f7e2f6c46 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c
@@ -150,25 +150,21 @@ ga102_gsp_flcn = {
 };
 
 static const struct nvkm_gsp_func
-ga102_gsp_r535_113_01 = {
+ga102_gsp_r535 = {
 	.flcn = &ga102_gsp_flcn,
 	.fwsec = &ga102_gsp_fwsec,
 
 	.sig_section = ".fwsignature_ga10x",
 
-	.wpr_heap.os_carveout_size = 20 << 20,
-	.wpr_heap.base_size = 8 << 20,
-	.wpr_heap.min_size = 84 << 20,
-
 	.booter.ctor = ga102_gsp_booter_ctor,
 
 	.dtor = r535_gsp_dtor,
 	.oneinit = tu102_gsp_oneinit,
-	.init = r535_gsp_init,
-	.fini = r535_gsp_fini,
+	.init = tu102_gsp_init,
+	.fini = tu102_gsp_fini,
 	.reset = ga102_gsp_reset,
 
-	.rm = &r535_gsp_rm,
+	.rm.gpu = &ga1xx_gpu,
 };
 
 static const struct nvkm_gsp_func
@@ -178,7 +174,8 @@ ga102_gsp = {
 
 static struct nvkm_gsp_fwif
 ga102_gsps[] = {
-	{  0,  r535_gsp_load, &ga102_gsp_r535_113_01, "535.113.01" },
+	{  1, tu102_gsp_load, &ga102_gsp_r535, &r570_rm_ga102, "570.144" },
+	{  0, tu102_gsp_load, &ga102_gsp_r535, &r535_rm_ga102, "535.113.01" },
 	{ -1, gv100_gsp_nofw, &ga102_gsp },
 	{}
 };
@@ -189,3 +186,15 @@ ga102_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 {
 	return nvkm_gsp_new_(ga102_gsps, device, type, inst, pgsp);
 }
+
+NVKM_GSP_FIRMWARE_BOOTER(ga102, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ga103, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ga104, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ga106, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(ga107, 535.113.01);
+
+NVKM_GSP_FIRMWARE_BOOTER(ga102, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ga103, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ga104, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ga106, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(ga107, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c
new file mode 100644
index 000000000000..12a3f2c1ed82
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+static const struct nvkm_gsp_func
+gb100_gsp = {
+	.flcn = &ga102_gsp_flcn,
+
+	.sig_section = ".fwsignature_gb10x",
+
+	.dtor = r535_gsp_dtor,
+	.oneinit = gh100_gsp_oneinit,
+	.init = gh100_gsp_init,
+	.fini = gh100_gsp_fini,
+
+	.rm.gpu = &gb10x_gpu,
+};
+
+static struct nvkm_gsp_fwif
+gb100_gsps[] = {
+	{ 0, gh100_gsp_load, &gb100_gsp, &r570_rm_gb10x, "570.144", true },
+	{}
+};
+
+int
+gb100_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+	      struct nvkm_gsp **pgsp)
+{
+	return nvkm_gsp_new_(gb100_gsps, device, type, inst, pgsp);
+}
+
+NVKM_GSP_FIRMWARE_FMC(gb100, 570.144);
+NVKM_GSP_FIRMWARE_FMC(gb102, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c
new file mode 100644
index 000000000000..c1d718172ddf
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+static const struct nvkm_gsp_func
+gb202_gsp = {
+	.flcn = &ga102_gsp_flcn,
+
+	.sig_section = ".fwsignature_gb20x",
+
+	.dtor = r535_gsp_dtor,
+	.oneinit = gh100_gsp_oneinit,
+	.init = gh100_gsp_init,
+	.fini = gh100_gsp_fini,
+
+	.rm.gpu = &gb20x_gpu,
+};
+
+static struct nvkm_gsp_fwif
+gb202_gsps[] = {
+	{ 0, gh100_gsp_load, &gb202_gsp, &r570_rm_gb20x, "570.144", true },
+	{}
+};
+
+int
+gb202_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+	      struct nvkm_gsp **pgsp)
+{
+	return nvkm_gsp_new_(gb202_gsps, device, type, inst, pgsp);
+}
+
+NVKM_GSP_FIRMWARE_FMC(gb202, 570.144);
+NVKM_GSP_FIRMWARE_FMC(gb203, 570.144);
+NVKM_GSP_FIRMWARE_FMC(gb205, 570.144);
+NVKM_GSP_FIRMWARE_FMC(gb206, 570.144);
+NVKM_GSP_FIRMWARE_FMC(gb207, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c
new file mode 100644
index 000000000000..ce31e8248807
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c
@@ -0,0 +1,358 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <linux/elf.h>
+#include <linux/crc32.h>
+
+#include <subdev/fb.h>
+#include <subdev/fsp.h>
+
+#include <rm/r570/nvrm/gsp.h>
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gh100/dev_falcon_v4.h>
+#include <nvhw/ref/gh100/dev_riscv_pri.h>
+
+int
+gh100_gsp_fini(struct nvkm_gsp *gsp, bool suspend)
+{
+	struct nvkm_falcon *falcon = &gsp->falcon;
+	int ret, time = 4000;
+
+	/* Shutdown RM. */
+	ret = r535_gsp_fini(gsp, suspend);
+	if (ret && suspend)
+		return ret;
+
+	/* Wait for RISC-V to halt. */
+	do {
+		u32 data = nvkm_falcon_rd32(falcon, falcon->addr2 + NV_PRISCV_RISCV_CPUCTL);
+
+		if (NVVAL_GET(data, NV_PRISCV, RISCV_CPUCTL, HALTED))
+			return 0;
+
+		usleep_range(1000, 2000);
+	} while(time--);
+
+	return -ETIMEDOUT;
+}
+
+static bool
+gh100_gsp_lockdown_released(struct nvkm_gsp *gsp, u32 *mbox0)
+{
+	u32 data;
+
+	/* Wait for GSP access via BAR0 to be allowed. */
+	*mbox0 = nvkm_falcon_rd32(&gsp->falcon, NV_PFALCON_FALCON_MAILBOX0);
+
+	if (*mbox0 && (*mbox0 & 0xffffff00) == 0xbadf4100)
+		return false;
+
+	/* Check if an error code has been reported. */
+	if (*mbox0) {
+		u32 mbox1 = nvkm_falcon_rd32(&gsp->falcon, NV_PFALCON_FALCON_MAILBOX1);
+
+		/* Any value that's not GSP_FMC_BOOT_PARAMS addr is an error. */
+		if ((((u64)mbox1 << 32) | *mbox0) != gsp->fmc.args.addr)
+			return true;
+	}
+
+	/* Check if lockdown has been released. */
+	data = nvkm_falcon_rd32(&gsp->falcon, NV_PFALCON_FALCON_HWCFG2);
+	return !NVVAL_GET(data, NV_PFALCON, FALCON_HWCFG2, RISCV_BR_PRIV_LOCKDOWN);
+}
+
+int
+gh100_gsp_init(struct nvkm_gsp *gsp)
+{
+	struct nvkm_subdev *subdev = &gsp->subdev;
+	struct nvkm_device *device = subdev->device;
+	const bool resume = gsp->sr.meta.data != NULL;
+	struct nvkm_gsp_mem *meta;
+	GSP_FMC_BOOT_PARAMS *args;
+	int ret, time = 4000;
+	u32 rsvd_size;
+	u32 mbox0;
+
+	if (!resume) {
+		ret = nvkm_gsp_mem_ctor(gsp, sizeof(*args), &gsp->fmc.args);
+		if (ret)
+			return ret;
+
+		meta = &gsp->wpr_meta;
+	} else {
+		gsp->rm->api->gsp->set_rmargs(gsp, true);
+		meta = &gsp->sr.meta;
+	}
+
+	args = gsp->fmc.args.data;
+
+	args->bootGspRmParams.gspRmDescOffset = meta->addr;
+	args->bootGspRmParams.gspRmDescSize = meta->size;
+	args->bootGspRmParams.target = GSP_DMA_TARGET_COHERENT_SYSTEM;
+	args->bootGspRmParams.bIsGspRmBoot = 1;
+
+	args->gspRmParams.target = GSP_DMA_TARGET_NONCOHERENT_SYSTEM;
+	args->gspRmParams.bootArgsOffset = gsp->libos.addr;
+
+	rsvd_size = gsp->fb.heap.size;
+	if (gsp->rm->wpr->rsvd_size_pmu)
+		rsvd_size = ALIGN(rsvd_size + gsp->rm->wpr->rsvd_size_pmu, 0x200000);
+
+	ret = nvkm_fsp_boot_gsp_fmc(device->fsp, gsp->fmc.args.addr, rsvd_size, resume,
+				    gsp->fmc.fw.addr, gsp->fmc.hash, gsp->fmc.pkey, gsp->fmc.sig);
+	if (ret)
+		return ret;
+
+	do {
+		if (gh100_gsp_lockdown_released(gsp, &mbox0))
+			break;
+
+		usleep_range(1000, 2000);
+	} while(time--);
+
+	if (time < 0) {
+		nvkm_error(subdev, "GSP-FMC boot timed out\n");
+		return -ETIMEDOUT;
+	}
+
+	if (mbox0) {
+		nvkm_error(subdev, "GSP-FMC boot failed (mbox: 0x%08x)\n", mbox0);
+		return -EIO;
+	}
+
+	return r535_gsp_init(gsp);
+}
+
+static int
+gh100_gsp_wpr_meta_init(struct nvkm_gsp *gsp)
+{
+	GspFwWprMeta *meta;
+	int ret;
+
+	ret = nvkm_gsp_mem_ctor(gsp, sizeof(*meta), &gsp->wpr_meta);
+	if (ret)
+		return ret;
+
+	gsp->fb.size = nvkm_fb_vidmem_size(gsp->subdev.device);
+	gsp->fb.bios.vga_workspace.size = 128 * 1024;
+	gsp->fb.heap.size = gsp->rm->wpr->heap_size_non_wpr;
+
+	meta = gsp->wpr_meta.data;
+
+	meta->magic = GSP_FW_WPR_META_MAGIC;
+	meta->revision = GSP_FW_WPR_META_REVISION;
+
+	meta->sizeOfRadix3Elf = gsp->fw.len;
+	meta->sysmemAddrOfRadix3Elf = gsp->radix3.lvl0.addr;
+
+	meta->sizeOfBootloader = gsp->boot.fw.size;
+	meta->sysmemAddrOfBootloader = gsp->boot.fw.addr;
+	meta->bootloaderCodeOffset = gsp->boot.code_offset;
+	meta->bootloaderDataOffset = gsp->boot.data_offset;
+	meta->bootloaderManifestOffset = gsp->boot.manifest_offset;
+
+	meta->sysmemAddrOfSignature = gsp->sig.addr;
+	meta->sizeOfSignature = gsp->sig.size;
+
+	meta->nonWprHeapSize = gsp->fb.heap.size;
+	meta->gspFwHeapSize = tu102_gsp_wpr_heap_size(gsp);
+	meta->frtsSize = 0x100000;
+	meta->vgaWorkspaceSize = gsp->fb.bios.vga_workspace.size;
+	meta->pmuReservedSize = gsp->rm->wpr->rsvd_size_pmu;
+	return 0;
+}
+
+/* The sh_flags value for the binary blobs in the ELF image */
+#define FMC_SHF_FLAGS (SHF_MASKPROC | SHF_MASKOS | SHF_OS_NONCONFORMING | SHF_ALLOC)
+
+#define ELF_HDR_SIZE ((u8)sizeof(struct elf32_hdr))
+#define ELF_SHDR_SIZE ((u8)sizeof(struct elf32_shdr))
+
+/* The FMC ELF header must be exactly this */
+static const u8 elf_header[] = {
+	0x7f, 'E', 'L', 'F', 1, 1, 1, 0,
+	0, 0, 0, 0, 0, 0, 0, 0,
+
+	0, 0, 0, 0, 1, 0, 0, 0, /* e_type, e_machine, e_version */
+	0, 0, 0, 0, 0, 0, 0, 0, /* e_entry, e_phoff */
+
+	ELF_HDR_SIZE, 0, 0, 0, 0, 0, 0, 0, /* e_shoff, e_flags */
+	ELF_HDR_SIZE, 0, 0, 0, /* e_ehsize, e_phentsize */
+	0, 0, ELF_SHDR_SIZE, 0, /* e_phnum, e_shentsize */
+
+	6, 0, 1, 0, /* e_shnum, e_shstrndx */
+};
+
+/**
+ * elf_validate_sections - validate each section in the FMC ELF image
+ * @elf: ELF image
+ * @length: size of the entire ELF image
+ */
+static bool
+elf_validate_sections(const void *elf, size_t length)
+{
+	const struct elf32_hdr *ehdr = elf;
+	const struct elf32_shdr *shdr = elf + ehdr->e_shoff;
+
+	/* The offset of the first section */
+	Elf32_Off section_begin = ehdr->e_shoff + ehdr->e_shnum * ehdr->e_shentsize;
+
+	if (section_begin > length)
+		return false;
+
+	/* The first section header is the null section, so skip it */
+	for (unsigned int i = 1; i < ehdr->e_shnum; i++) {
+		if (i == ehdr->e_shstrndx) {
+			if (shdr[i].sh_type != SHT_STRTAB)
+				return false;
+			if (shdr[i].sh_flags != SHF_STRINGS)
+				return false;
+		} else {
+			if (shdr[i].sh_type != SHT_PROGBITS)
+				return false;
+			if (shdr[i].sh_flags != FMC_SHF_FLAGS)
+				return false;
+		}
+
+		/* Ensure that each section is inside the image */
+		if (shdr[i].sh_offset < section_begin ||
+		    (u64)shdr[i].sh_offset + shdr[i].sh_size > length)
+			return false;
+
+		/* Non-zero sh_info is a CRC */
+		if (shdr[i].sh_info) {
+			/* The kernel's CRC32 needs a pre- and post-xor to match standard CRCs */
+			u32 crc32 = crc32_le(~0, elf + shdr[i].sh_offset, shdr[i].sh_size) ^ ~0;
+
+			if (shdr[i].sh_info != crc32)
+				return false;
+		}
+	}
+
+	return true;
+}
+
+/**
+ * elf_section - return a pointer to the data for a given section
+ * @elf: ELF image
+ * @name: section name to search for
+ * @len: pointer to returned length of found section
+ */
+static const void *
+elf_section(const void *elf, const char *name, unsigned int *len)
+{
+	const struct elf32_hdr *ehdr = elf;
+	const struct elf32_shdr *shdr = elf + ehdr->e_shoff;
+	const char *names = elf + shdr[ehdr->e_shstrndx].sh_offset;
+
+	for (unsigned int i = 1; i < ehdr->e_shnum; i++) {
+		if (!strcmp(&names[shdr[i].sh_name], name)) {
+			*len = shdr[i].sh_size;
+			return elf + shdr[i].sh_offset;
+		}
+	}
+
+	return NULL;
+}
+
+int
+gh100_gsp_oneinit(struct nvkm_gsp *gsp)
+{
+	struct nvkm_subdev *subdev = &gsp->subdev;
+	struct nvkm_device *device = subdev->device;
+	struct nvkm_fsp *fsp = device->fsp;
+	const void *fw = gsp->fws.fmc->data;
+	const void *hash, *sig, *pkey, *img;
+	unsigned int img_len = 0, hash_len = 0, pkey_len = 0, sig_len = 0;
+	int ret;
+
+	if (gsp->fws.fmc->size < ELF_HDR_SIZE ||
+	    memcmp(fw, elf_header, sizeof(elf_header)) ||
+	    !elf_validate_sections(fw, gsp->fws.fmc->size)) {
+		nvkm_error(subdev, "fmc firmware image is invalid\n");
+		return -ENODATA;
+	}
+
+	hash = elf_section(fw, "hash", &hash_len);
+	sig = elf_section(fw, "signature", &sig_len);
+	pkey = elf_section(fw, "publickey", &pkey_len);
+	img = elf_section(fw, "image", &img_len);
+
+	if (!hash || !sig || !pkey || !img) {
+		nvkm_error(subdev, "fmc firmware image is invalid\n");
+		return -ENODATA;
+	}
+
+	if (!nvkm_fsp_verify_gsp_fmc(fsp, hash_len, pkey_len, sig_len))
+		return -EINVAL;
+
+	/* Load GSP-FMC FW into memory. */
+	ret = nvkm_gsp_mem_ctor(gsp, img_len, &gsp->fmc.fw);
+	if (ret)
+		return ret;
+
+	memcpy(gsp->fmc.fw.data, img, img_len);
+
+	gsp->fmc.hash = kmemdup(hash, hash_len, GFP_KERNEL);
+	gsp->fmc.pkey = kmemdup(pkey, pkey_len, GFP_KERNEL);
+	gsp->fmc.sig = kmemdup(sig, sig_len, GFP_KERNEL);
+	if (!gsp->fmc.hash || !gsp->fmc.pkey || !gsp->fmc.sig)
+		return -ENOMEM;
+
+	ret = r535_gsp_oneinit(gsp);
+	if (ret)
+		return ret;
+
+	return gh100_gsp_wpr_meta_init(gsp);
+}
+
+static const struct nvkm_gsp_func
+gh100_gsp = {
+	.flcn = &ga102_gsp_flcn,
+
+	.sig_section = ".fwsignature_gh100",
+
+	.dtor = r535_gsp_dtor,
+	.oneinit = gh100_gsp_oneinit,
+	.init = gh100_gsp_init,
+	.fini = gh100_gsp_fini,
+
+	.rm.gpu = &gh100_gpu,
+};
+
+int
+gh100_gsp_load(struct nvkm_gsp *gsp, int ver, const struct nvkm_gsp_fwif *fwif)
+{
+	int ret;
+
+	ret = tu102_gsp_load_rm(gsp, fwif);
+	if (ret)
+		goto done;
+
+	ret = nvkm_gsp_load_fw(gsp, "fmc", fwif->ver, &gsp->fws.fmc);
+
+done:
+	if (ret)
+		nvkm_gsp_dtor_fws(gsp);
+
+	return ret;
+}
+
+static struct nvkm_gsp_fwif
+gh100_gsps[] = {
+	{ 0, gh100_gsp_load, &gh100_gsp, &r570_rm_gh100, "570.144", true },
+	{}
+};
+
+int
+gh100_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+	      struct nvkm_gsp **pgsp)
+{
+	return nvkm_gsp_new_(gh100_gsps, device, type, inst, pgsp);
+}
+
+NVKM_GSP_FIRMWARE_FMC(gh100, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h
index 9f4a62375a27..4f14e85fc69e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h
@@ -2,6 +2,7 @@
 #ifndef __NVKM_GSP_PRIV_H__
 #define __NVKM_GSP_PRIV_H__
 #include <subdev/gsp.h>
+#include <rm/gpu.h>
 enum nvkm_acr_lsf_id;
 
 int nvkm_gsp_fwsec_frts(struct nvkm_gsp *);
@@ -11,12 +12,32 @@ struct nvkm_gsp_fwif {
 	int version;
 	int (*load)(struct nvkm_gsp *, int ver, const struct nvkm_gsp_fwif *);
 	const struct nvkm_gsp_func *func;
+	const struct nvkm_rm_impl *rm;
 	const char *ver;
 	bool enable;
 };
 
+int nvkm_gsp_load_fw(struct nvkm_gsp *, const char *name, const char *ver,
+		     const struct firmware **);
+void nvkm_gsp_dtor_fws(struct nvkm_gsp *);
+
 int gv100_gsp_nofw(struct nvkm_gsp *, int, const struct nvkm_gsp_fwif *);
-int  r535_gsp_load(struct nvkm_gsp *, int, const struct nvkm_gsp_fwif *);
+
+int tu102_gsp_load(struct nvkm_gsp *, int, const struct nvkm_gsp_fwif *);
+int tu102_gsp_load_rm(struct nvkm_gsp *, const struct nvkm_gsp_fwif *);
+
+int gh100_gsp_load(struct nvkm_gsp *, int, const struct nvkm_gsp_fwif *);
+
+#define NVKM_GSP_FIRMWARE_BOOTER(chip,vers)                      \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/booter_load-"#vers".bin");   \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/booter_unload-"#vers".bin"); \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/bootloader-"#vers".bin");    \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/gsp-"#vers".bin")
+
+#define NVKM_GSP_FIRMWARE_FMC(chip,vers)                      \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/fmc-"#vers".bin");        \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/bootloader-"#vers".bin"); \
+MODULE_FIRMWARE("nvidia/"#chip"/gsp/gsp-"#vers".bin")
 
 struct nvkm_gsp_func {
 	const struct nvkm_falcon_func *flcn;
@@ -25,12 +46,6 @@ struct nvkm_gsp_func {
 	char *sig_section;
 
 	struct {
-		u32 os_carveout_size;
-		u32 base_size;
-		u64 min_size;
-	} wpr_heap;
-
-	struct {
 		int (*ctor)(struct nvkm_gsp *, const char *name, const struct firmware *,
 			    struct nvkm_falcon *, struct nvkm_falcon_fw *);
 	} booter;
@@ -41,7 +56,9 @@ struct nvkm_gsp_func {
 	int (*fini)(struct nvkm_gsp *, bool suspend);
 	int (*reset)(struct nvkm_gsp *);
 
-	const struct nvkm_gsp_rm *rm;
+	struct {
+		const struct nvkm_rm_gpu *gpu;
+	} rm;
 };
 
 extern const struct nvkm_falcon_func tu102_gsp_flcn;
@@ -49,7 +66,10 @@ extern const struct nvkm_falcon_fw_func tu102_gsp_fwsec;
 int tu102_gsp_booter_ctor(struct nvkm_gsp *, const char *, const struct firmware *,
 			  struct nvkm_falcon *, struct nvkm_falcon_fw *);
 int tu102_gsp_oneinit(struct nvkm_gsp *);
+int tu102_gsp_init(struct nvkm_gsp *);
+int tu102_gsp_fini(struct nvkm_gsp *, bool suspend);
 int tu102_gsp_reset(struct nvkm_gsp *);
+u64 tu102_gsp_wpr_heap_size(struct nvkm_gsp *);
 
 extern const struct nvkm_falcon_func ga102_gsp_flcn;
 extern const struct nvkm_falcon_fw_func ga102_gsp_fwsec;
@@ -57,11 +77,14 @@ int ga102_gsp_booter_ctor(struct nvkm_gsp *, const char *, const struct firmware
 			  struct nvkm_falcon *, struct nvkm_falcon_fw *);
 int ga102_gsp_reset(struct nvkm_gsp *);
 
+int gh100_gsp_oneinit(struct nvkm_gsp *);
+int gh100_gsp_init(struct nvkm_gsp *);
+int gh100_gsp_fini(struct nvkm_gsp *, bool suspend);
+
 void r535_gsp_dtor(struct nvkm_gsp *);
 int r535_gsp_oneinit(struct nvkm_gsp *);
 int r535_gsp_init(struct nvkm_gsp *);
 int r535_gsp_fini(struct nvkm_gsp *, bool suspend);
-extern const struct nvkm_gsp_rm r535_gsp_rm;
 
 int nvkm_gsp_new_(const struct nvkm_gsp_fwif *, struct nvkm_device *, enum nvkm_subdev_type, int,
 		  struct nvkm_gsp **);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/Kbuild
new file mode 100644
index 000000000000..04037394a2da
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/Kbuild
@@ -0,0 +1,19 @@
+# SPDX-License-Identifier: MIT
+#
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+nvkm-y += nvkm/subdev/gsp/rm/client.o
+nvkm-y += nvkm/subdev/gsp/rm/engine.o
+nvkm-y += nvkm/subdev/gsp/rm/gr.o
+nvkm-y += nvkm/subdev/gsp/rm/nvdec.o
+nvkm-y += nvkm/subdev/gsp/rm/nvenc.o
+
+nvkm-y += nvkm/subdev/gsp/rm/tu1xx.o
+nvkm-y += nvkm/subdev/gsp/rm/ga100.o
+nvkm-y += nvkm/subdev/gsp/rm/ga1xx.o
+nvkm-y += nvkm/subdev/gsp/rm/ad10x.o
+nvkm-y += nvkm/subdev/gsp/rm/gh100.o
+nvkm-y += nvkm/subdev/gsp/rm/gb10x.o
+nvkm-y += nvkm/subdev/gsp/rm/gb20x.o
+
+include $(src)/nvkm/subdev/gsp/rm/r535/Kbuild
+include $(src)/nvkm/subdev/gsp/rm/r570/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ad10x.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ad10x.c
new file mode 100644
index 000000000000..e1ce6355c35f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ad10x.c
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+ad10x_gpu = {
+	.disp.class = {
+		.root = AD102_DISP,
+		.caps = GV100_DISP_CAPS,
+		.core = AD102_DISP_CORE_CHANNEL_DMA,
+		.wndw = GA102_DISP_WINDOW_CHANNEL_DMA,
+		.wimm = GA102_DISP_WINDOW_IMM_CHANNEL_DMA,
+		.curs = GA102_DISP_CURSOR,
+	},
+
+	.usermode.class = AMPERE_USERMODE_A,
+
+	.fifo.chan = {
+		.class = AMPERE_CHANNEL_GPFIFO_A,
+		.doorbell_handle = tu102_chan_doorbell_handle,
+	},
+
+	.ce.class = AMPERE_DMA_COPY_B,
+	.gr.class = {
+		.i2m = KEPLER_INLINE_TO_MEMORY_B,
+		.twod = FERMI_TWOD_A,
+		.threed = ADA_A,
+		.compute = ADA_COMPUTE_A,
+	},
+	.nvdec.class = NVC9B0_VIDEO_DECODER,
+	.nvenc.class = NVC9B7_VIDEO_ENCODER,
+	.ofa.class = NVC9FA_VIDEO_OFA,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/client.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/client.c
new file mode 100644
index 000000000000..72d3e3ca84c2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/client.c
@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "rm.h"
+
+void
+nvkm_gsp_client_dtor(struct nvkm_gsp_client *client)
+{
+	const unsigned int id = client->object.handle - NVKM_RM_CLIENT(0);
+	struct nvkm_gsp *gsp = client->gsp;
+
+	if (!gsp)
+		return;
+
+	if (client->object.client)
+		nvkm_gsp_rm_free(&client->object);
+
+	mutex_lock(&gsp->client_id.mutex);
+	idr_remove(&gsp->client_id.idr, id);
+	mutex_unlock(&gsp->client_id.mutex);
+
+	client->gsp = NULL;
+}
+
+int
+nvkm_gsp_client_ctor(struct nvkm_gsp *gsp, struct nvkm_gsp_client *client)
+{
+	int id, ret;
+
+	if (WARN_ON(!gsp->rm))
+		return -ENOSYS;
+
+	mutex_lock(&gsp->client_id.mutex);
+	id = idr_alloc(&gsp->client_id.idr, client, 0, NVKM_RM_CLIENT_MASK + 1, GFP_KERNEL);
+	mutex_unlock(&gsp->client_id.mutex);
+	if (id < 0)
+		return id;
+
+	client->gsp = gsp;
+	client->object.client = client;
+	INIT_LIST_HEAD(&client->events);
+
+	ret = gsp->rm->api->client->ctor(client, NVKM_RM_CLIENT(id));
+	if (ret)
+		nvkm_gsp_client_dtor(client);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/engine.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/engine.c
new file mode 100644
index 000000000000..3b0e83b2f57f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/engine.c
@@ -0,0 +1,189 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "engine.h"
+#include "gpu.h"
+
+#include <core/object.h>
+#include <engine/fifo/chan.h>
+
+struct nvkm_rm_engine {
+	struct nvkm_engine engine;
+
+	struct nvkm_engine_func func;
+};
+
+struct nvkm_rm_engine_obj {
+	struct nvkm_object object;
+	struct nvkm_gsp_object rm;
+};
+
+static void*
+nvkm_rm_engine_obj_dtor(struct nvkm_object *object)
+{
+	struct nvkm_rm_engine_obj *obj = container_of(object, typeof(*obj), object);
+
+	nvkm_gsp_rm_free(&obj->rm);
+	return obj;
+}
+
+static const struct nvkm_object_func
+nvkm_rm_engine_obj = {
+	.dtor = nvkm_rm_engine_obj_dtor,
+};
+
+int
+nvkm_rm_engine_obj_new(struct nvkm_gsp_object *chan, int chid, const struct nvkm_oclass *oclass,
+		       struct nvkm_object **pobject)
+{
+	struct nvkm_rm *rm = chan->client->gsp->rm;
+	const int inst = oclass->engine->subdev.inst;
+	const u32 class = oclass->base.oclass;
+	const u32 handle = oclass->handle;
+	struct nvkm_rm_engine_obj *obj;
+	int ret;
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return -ENOMEM;
+
+	switch (oclass->engine->subdev.type) {
+	case NVKM_ENGINE_CE:
+		ret = rm->api->ce->alloc(chan, handle, class, inst, &obj->rm);
+		break;
+	case NVKM_ENGINE_GR:
+		ret = nvkm_gsp_rm_alloc(chan, handle, class, 0, &obj->rm);
+		break;
+	case NVKM_ENGINE_NVDEC:
+		ret = rm->api->nvdec->alloc(chan, handle, class, inst, &obj->rm);
+		break;
+	case NVKM_ENGINE_NVENC:
+		ret = rm->api->nvenc->alloc(chan, handle, class, inst, &obj->rm);
+		break;
+	case NVKM_ENGINE_NVJPG:
+		ret = rm->api->nvjpg->alloc(chan, handle, class, inst, &obj->rm);
+		break;
+	case NVKM_ENGINE_OFA:
+		ret = rm->api->ofa->alloc(chan, handle, class, inst, &obj->rm);
+		break;
+	default:
+		ret = -EINVAL;
+		WARN_ON(1);
+		break;
+	}
+
+	if (ret) {
+		kfree(obj);
+		return ret;
+	}
+
+	nvkm_object_ctor(&nvkm_rm_engine_obj, oclass, &obj->object);
+	*pobject = &obj->object;
+	return 0;
+}
+
+static int
+nvkm_rm_engine_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
+			struct nvkm_object **pobject)
+{
+	struct nvkm_chan *chan = nvkm_uchan_chan(oclass->parent);
+
+	return nvkm_rm_engine_obj_new(&chan->rm.object, chan->id, oclass, pobject);
+}
+
+static void *
+nvkm_rm_engine_dtor(struct nvkm_engine *engine)
+{
+	kfree(engine->func);
+	return engine;
+}
+
+int
+nvkm_rm_engine_ctor(void *(*dtor)(struct nvkm_engine *), struct nvkm_rm *rm,
+		    enum nvkm_subdev_type type, int inst,
+		    const u32 *class, int nclass, struct nvkm_engine *engine)
+{
+	struct nvkm_engine_func *func;
+
+	func = kzalloc(struct_size(func, sclass, nclass + 1), GFP_KERNEL);
+	if (!func)
+		return -ENOMEM;
+
+	func->dtor = dtor;
+
+	for (int i = 0; i < nclass; i++) {
+		func->sclass[i].oclass = class[i];
+		func->sclass[i].minver = -1;
+		func->sclass[i].maxver = 0;
+		func->sclass[i].ctor = nvkm_rm_engine_obj_ctor;
+	}
+
+	nvkm_engine_ctor(func, rm->device, type, inst, true, engine);
+	return 0;
+}
+
+static int
+nvkm_rm_engine_new_(struct nvkm_rm *rm, enum nvkm_subdev_type type, int inst, u32 class,
+		    struct nvkm_engine **pengine)
+{
+	struct nvkm_engine *engine;
+	int ret;
+
+	engine = kzalloc(sizeof(*engine), GFP_KERNEL);
+	if (!engine)
+		return -ENOMEM;
+
+	ret = nvkm_rm_engine_ctor(nvkm_rm_engine_dtor, rm, type, inst, &class, 1, engine);
+	if (ret) {
+		kfree(engine);
+		return ret;
+	}
+
+	*pengine = engine;
+	return 0;
+}
+
+int
+nvkm_rm_engine_new(struct nvkm_rm *rm, enum nvkm_subdev_type type, int inst)
+{
+	const struct nvkm_rm_gpu *gpu = rm->gpu;
+	struct nvkm_device *device = rm->device;
+
+	switch (type) {
+	case NVKM_ENGINE_CE:
+		if (WARN_ON(inst >= ARRAY_SIZE(device->ce)))
+			return -EINVAL;
+
+		return nvkm_rm_engine_new_(rm, type, inst, gpu->ce.class, &device->ce[inst]);
+	case NVKM_ENGINE_GR:
+		if (inst != 0)
+			return -ENODEV; /* MiG not supported, just ignore. */
+
+		return nvkm_rm_gr_new(rm);
+	case NVKM_ENGINE_NVDEC:
+		if (WARN_ON(inst >= ARRAY_SIZE(device->nvdec)))
+			return -EINVAL;
+
+		return nvkm_rm_nvdec_new(rm, inst);
+	case NVKM_ENGINE_NVENC:
+		if (WARN_ON(inst >= ARRAY_SIZE(device->nvenc)))
+			return -EINVAL;
+
+		return nvkm_rm_nvenc_new(rm, inst);
+	case NVKM_ENGINE_NVJPG:
+		if (WARN_ON(inst >= ARRAY_SIZE(device->nvjpg)))
+			return -EINVAL;
+
+		return nvkm_rm_engine_new_(rm, type, inst, gpu->nvjpg.class, &device->nvjpg[inst]);
+	case NVKM_ENGINE_OFA:
+		if (WARN_ON(inst >= ARRAY_SIZE(device->ofa)))
+			return -EINVAL;
+
+		return nvkm_rm_engine_new_(rm, type, inst, gpu->ofa.class, &device->ofa[inst]);
+	default:
+		break;
+	}
+
+	return -ENODEV;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/engine.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/engine.h
new file mode 100644
index 000000000000..5b8c9c3901d4
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/engine.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVKM_RM_ENGINE_H__
+#define __NVKM_RM_ENGINE_H__
+#include "gpu.h"
+
+int nvkm_rm_engine_ctor(void *(*dtor)(struct nvkm_engine *), struct nvkm_rm *,
+			enum nvkm_subdev_type type, int inst,
+			const u32 *class, int nclass, struct nvkm_engine *);
+int nvkm_rm_engine_new(struct nvkm_rm *, enum nvkm_subdev_type, int inst);
+
+int nvkm_rm_engine_obj_new(struct nvkm_gsp_object *chan, int chid, const struct nvkm_oclass *,
+			   struct nvkm_object **);
+
+int nvkm_rm_gr_new(struct nvkm_rm *);
+int nvkm_rm_nvdec_new(struct nvkm_rm *, int inst);
+int nvkm_rm_nvenc_new(struct nvkm_rm *, int inst);
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ga100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ga100.c
new file mode 100644
index 000000000000..a48c6134075d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ga100.c
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+ga100_gpu = {
+	.usermode.class = AMPERE_USERMODE_A,
+
+	.fifo.chan = {
+		.class = AMPERE_CHANNEL_GPFIFO_A,
+		.doorbell_handle = tu102_chan_doorbell_handle,
+	},
+
+	.ce.class = AMPERE_DMA_COPY_A,
+	.gr.class = {
+		.i2m = KEPLER_INLINE_TO_MEMORY_B,
+		.twod = FERMI_TWOD_A,
+		.threed = AMPERE_A,
+		.compute = AMPERE_COMPUTE_A,
+	},
+	.nvdec.class = NVC6B0_VIDEO_DECODER,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ga1xx.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ga1xx.c
new file mode 100644
index 000000000000..50536ad7f85d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/ga1xx.c
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+ga1xx_gpu = {
+	.disp.class = {
+		.root = GA102_DISP,
+		.caps = GV100_DISP_CAPS,
+		.core = GA102_DISP_CORE_CHANNEL_DMA,
+		.wndw = GA102_DISP_WINDOW_CHANNEL_DMA,
+		.wimm = GA102_DISP_WINDOW_IMM_CHANNEL_DMA,
+		.curs = GA102_DISP_CURSOR,
+	},
+
+	.usermode.class = AMPERE_USERMODE_A,
+
+	.fifo.chan = {
+		.class = AMPERE_CHANNEL_GPFIFO_A,
+		.doorbell_handle = tu102_chan_doorbell_handle,
+	},
+
+	.ce.class = AMPERE_DMA_COPY_B,
+	.gr.class = {
+		.i2m = KEPLER_INLINE_TO_MEMORY_B,
+		.twod = FERMI_TWOD_A,
+		.threed = AMPERE_B,
+		.compute = AMPERE_COMPUTE_B,
+	},
+	.nvdec.class = NVC7B0_VIDEO_DECODER,
+	.nvenc.class = NVC7B7_VIDEO_ENCODER,
+	.ofa.class = NVC7FA_VIDEO_OFA,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gb10x.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gb10x.c
new file mode 100644
index 000000000000..2f517dcd721a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gb10x.c
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+gb10x_gpu = {
+	.usermode.class = HOPPER_USERMODE_A,
+
+	.fifo.chan = {
+		.class = BLACKWELL_CHANNEL_GPFIFO_A,
+		.doorbell_handle = tu102_chan_doorbell_handle,
+	},
+
+	.ce.class = BLACKWELL_DMA_COPY_A,
+	.gr.class = {
+		.i2m = BLACKWELL_INLINE_TO_MEMORY_A,
+		.twod = FERMI_TWOD_A,
+		.threed = BLACKWELL_A,
+		.compute = BLACKWELL_COMPUTE_A,
+	},
+	.nvdec.class = NVCDB0_VIDEO_DECODER,
+	.nvjpg.class = NVCDD1_VIDEO_NVJPG,
+	.ofa.class = NVCDFA_VIDEO_OFA,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gb20x.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gb20x.c
new file mode 100644
index 000000000000..950471d9996e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gb20x.c
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/ce/priv.h>
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+gb20x_gpu = {
+	.disp.class = {
+		.root = GB202_DISP,
+		.caps = GB202_DISP_CAPS,
+		.core = GB202_DISP_CORE_CHANNEL_DMA,
+		.wndw = GB202_DISP_WINDOW_CHANNEL_DMA,
+		.wimm = GB202_DISP_WINDOW_IMM_CHANNEL_DMA,
+		.curs = GB202_DISP_CURSOR,
+	},
+
+	.usermode.class = BLACKWELL_USERMODE_A,
+
+	.fifo.chan = {
+		.class = BLACKWELL_CHANNEL_GPFIFO_B,
+		.doorbell_handle = gb202_chan_doorbell_handle,
+	},
+
+	.ce = {
+		.class = BLACKWELL_DMA_COPY_B,
+		.grce_mask = gb202_ce_grce_mask,
+	},
+	.gr.class = {
+		.i2m = BLACKWELL_INLINE_TO_MEMORY_A,
+		.twod = FERMI_TWOD_A,
+		.threed = BLACKWELL_B,
+		.compute = BLACKWELL_COMPUTE_B,
+	},
+	.nvdec.class = NVCFB0_VIDEO_DECODER,
+	.nvenc.class = NVCFB7_VIDEO_ENCODER,
+	.nvjpg.class = NVCFD1_VIDEO_NVJPG,
+	.ofa.class = NVCFFA_VIDEO_OFA,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gh100.c
new file mode 100644
index 000000000000..49e2c54e1aa8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gh100.c
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+gh100_gpu = {
+	.usermode.class = HOPPER_USERMODE_A,
+
+	.fifo.chan = {
+		.class = HOPPER_CHANNEL_GPFIFO_A,
+		.doorbell_handle = tu102_chan_doorbell_handle,
+	},
+
+	.ce.class = HOPPER_DMA_COPY_A,
+	.gr.class = {
+		.i2m = KEPLER_INLINE_TO_MEMORY_B,
+		.twod = FERMI_TWOD_A,
+		.threed = HOPPER_A,
+		.compute = HOPPER_COMPUTE_A,
+	},
+	.nvdec.class = NVB8B0_VIDEO_DECODER,
+	.nvjpg.class = NVB8D1_VIDEO_NVJPG,
+	.ofa.class = NVB8FA_VIDEO_OFA,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gpu.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gpu.h
new file mode 100644
index 000000000000..46a6325641b7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gpu.h
@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVKM_RM_GPU_H__
+#define __NVKM_RM_GPU_H__
+#include "rm.h"
+
+struct nvkm_rm_gpu {
+	struct {
+		struct {
+			u32 root;
+			u32 caps;
+			u32 core;
+			u32 wndw;
+			u32 wimm;
+			u32 curs;
+		} class;
+	} disp;
+
+	struct {
+		u32 class;
+	} usermode;
+
+	struct {
+		struct {
+			u32 class;
+			u32 (*doorbell_handle)(struct nvkm_chan *);
+		} chan;
+	} fifo;
+
+	struct {
+		u32 class;
+		u32 (*grce_mask)(struct nvkm_device *);
+	} ce;
+
+	struct {
+		struct {
+			u32 i2m;
+			u32 twod;
+			u32 threed;
+			u32 compute;
+		} class;
+	} gr;
+
+	struct {
+		u32 class;
+	} nvdec;
+
+	struct {
+		u32 class;
+	} nvenc;
+
+	struct {
+		u32 class;
+	} nvjpg;
+
+	struct {
+		u32 class;
+	} ofa;
+};
+
+extern const struct nvkm_rm_gpu tu1xx_gpu;
+extern const struct nvkm_rm_gpu ga100_gpu;
+extern const struct nvkm_rm_gpu ga1xx_gpu;
+extern const struct nvkm_rm_gpu ad10x_gpu;
+extern const struct nvkm_rm_gpu gh100_gpu;
+extern const struct nvkm_rm_gpu gb10x_gpu;
+extern const struct nvkm_rm_gpu gb20x_gpu;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gr.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gr.c
new file mode 100644
index 000000000000..f40b8fcc2bcb
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gr.c
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gr.h"
+
+#include <engine/fifo.h>
+#include <engine/gr/priv.h>
+
+static int
+nvkm_rm_gr_obj_ctor(const struct nvkm_oclass *oclass, void *argv, u32 argc,
+		    struct nvkm_object **pobject)
+{
+	struct r535_gr_chan *chan = container_of(oclass->parent, typeof(*chan), object);
+
+	return nvkm_rm_engine_obj_new(&chan->chan->rm.object, chan->chan->id, oclass, pobject);
+}
+
+static int
+nvkm_rm_gr_fini(struct nvkm_gr *base, bool suspend)
+{
+	struct nvkm_rm *rm = base->engine.subdev.device->gsp->rm;
+	struct r535_gr *gr = container_of(base, typeof(*gr), base);
+
+	if (rm->api->gr->scrubber.fini)
+		rm->api->gr->scrubber.fini(gr);
+
+	return 0;
+}
+
+static int
+nvkm_rm_gr_init(struct nvkm_gr *base)
+{
+	struct nvkm_rm *rm = base->engine.subdev.device->gsp->rm;
+	struct r535_gr *gr = container_of(base, typeof(*gr), base);
+	int ret;
+
+	if (rm->api->gr->scrubber.init) {
+		ret = rm->api->gr->scrubber.init(gr);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int
+nvkm_rm_gr_new(struct nvkm_rm *rm)
+{
+	const u32 classes[] = {
+		rm->gpu->gr.class.i2m,
+		rm->gpu->gr.class.twod,
+		rm->gpu->gr.class.threed,
+		rm->gpu->gr.class.compute,
+	};
+	struct nvkm_gr_func *func;
+	struct r535_gr *gr;
+
+	func = kzalloc(struct_size(func, sclass, ARRAY_SIZE(classes) + 1), GFP_KERNEL);
+	if (!func)
+		return -ENOMEM;
+
+	func->dtor = r535_gr_dtor;
+	func->oneinit = r535_gr_oneinit;
+	func->init = nvkm_rm_gr_init;
+	func->fini = nvkm_rm_gr_fini;
+	func->units = r535_gr_units;
+	func->chan_new = r535_gr_chan_new;
+
+	for (int i = 0; i < ARRAY_SIZE(classes); i++) {
+		func->sclass[i].oclass = classes[i];
+		func->sclass[i].minver = -1;
+		func->sclass[i].maxver = 0;
+		func->sclass[i].ctor = nvkm_rm_gr_obj_ctor;
+	}
+
+	gr = kzalloc(sizeof(*gr), GFP_KERNEL);
+	if (!gr) {
+		kfree(func);
+		return -ENOMEM;
+	}
+
+	nvkm_gr_ctor(func, rm->device, NVKM_ENGINE_GR, 0, true, &gr->base);
+	gr->scrubber.chid = -1;
+	rm->device->gr = &gr->base;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gr.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gr.h
new file mode 100644
index 000000000000..24980f23aab9
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/gr.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: MIT */
+#ifndef __NVKM_RM_GR_H__
+#define __NVKM_RM_GR_H__
+#include "engine.h"
+
+#include <core/object.h>
+#include <engine/gr.h>
+
+#define R515_GR_MAX_CTXBUFS 9
+
+struct r535_gr_chan {
+	struct nvkm_object object;
+	struct r535_gr *gr;
+
+	struct nvkm_vmm *vmm;
+	struct nvkm_chan *chan;
+
+	struct nvkm_memory *mem[R515_GR_MAX_CTXBUFS];
+	struct nvkm_vma    *vma[R515_GR_MAX_CTXBUFS];
+};
+
+struct r535_gr {
+	struct nvkm_gr base;
+
+	struct {
+		u16 bufferId;
+		u32 size;
+		u8  page;
+		u8  align;
+		bool global;
+		bool init;
+		bool ro;
+	} ctxbuf[R515_GR_MAX_CTXBUFS];
+	int ctxbuf_nr;
+
+	struct nvkm_memory *ctxbuf_mem[R515_GR_MAX_CTXBUFS];
+
+	struct {
+		int chid;
+		struct nvkm_memory *inst;
+		struct nvkm_vmm *vmm;
+		struct nvkm_gsp_object chan;
+		struct nvkm_gsp_object threed;
+		struct {
+			struct nvkm_memory *mem[R515_GR_MAX_CTXBUFS];
+			struct nvkm_vma    *vma[R515_GR_MAX_CTXBUFS];
+		} ctxbuf;
+		bool enabled;
+	} scrubber;
+};
+
+struct NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO;
+void r535_gr_get_ctxbuf_info(struct r535_gr *, int i,
+			     struct NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO *);
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/handles.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/handles.h
new file mode 100644
index 000000000000..3bdb5ad320d7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/handles.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVKM_RM_HANDLES_H__
+#define __NVKM_RM_HANDLES_H__
+
+/* RMAPI handles for various objects allocated from GSP-RM with RM_ALLOC. */
+
+#define NVKM_RM_CLIENT(id)         (0xc1d00000 | (id))
+#define NVKM_RM_CLIENT_MASK         0x0000ffff
+#define NVKM_RM_DEVICE              0xde1d0000
+#define NVKM_RM_SUBDEVICE           0x5d1d0000
+#define NVKM_RM_DISP                0x00730000
+#define NVKM_RM_VASPACE             0x90f10000
+#define NVKM_RM_CHAN(chid)         (0xf1f00000 | (chid))
+#define NVKM_RM_THREED              0x97000000
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/nvdec.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/nvdec.c
new file mode 100644
index 000000000000..d9fbfc377864
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/nvdec.c
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "engine.h"
+#include <engine/nvdec.h>
+
+static void *
+nvkm_rm_nvdec_dtor(struct nvkm_engine *engine)
+{
+	return container_of(engine, struct nvkm_nvdec, engine);
+}
+
+int
+nvkm_rm_nvdec_new(struct nvkm_rm *rm, int inst)
+{
+	struct nvkm_nvdec *nvdec;
+	int ret;
+
+	nvdec = kzalloc(sizeof(*nvdec), GFP_KERNEL);
+	if (!nvdec)
+		return -ENOMEM;
+
+	ret = nvkm_rm_engine_ctor(nvkm_rm_nvdec_dtor, rm, NVKM_ENGINE_NVDEC, inst,
+				  &rm->gpu->nvdec.class, 1, &nvdec->engine);
+	if (ret) {
+		kfree(nvdec);
+		return ret;
+	}
+
+	rm->device->nvdec[inst] = nvdec;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/nvenc.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/nvenc.c
new file mode 100644
index 000000000000..6dfa7b789e07
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/nvenc.c
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "engine.h"
+#include <engine/nvenc.h>
+
+static void *
+nvkm_rm_nvenc_dtor(struct nvkm_engine *engine)
+{
+	return container_of(engine, struct nvkm_nvenc, engine);
+}
+
+int
+nvkm_rm_nvenc_new(struct nvkm_rm *rm, int inst)
+{
+	struct nvkm_nvenc *nvenc;
+	int ret;
+
+	nvenc = kzalloc(sizeof(*nvenc), GFP_KERNEL);
+	if (!nvenc)
+		return -ENOMEM;
+
+	ret = nvkm_rm_engine_ctor(nvkm_rm_nvenc_dtor, rm, NVKM_ENGINE_NVENC, inst,
+				  &rm->gpu->nvenc.class, 1, &nvenc->engine);
+	if (ret) {
+		kfree(nvenc);
+		return ret;
+	}
+
+	rm->device->nvenc[inst] = nvenc;
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/Kbuild
new file mode 100644
index 000000000000..a5f6b2abfd33
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/Kbuild
@@ -0,0 +1,25 @@
+# SPDX-License-Identifier: MIT
+#
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+
+nvkm-y += nvkm/subdev/gsp/rm/r535/rm.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/gsp.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/rpc.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/ctrl.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/alloc.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/client.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/device.o
+
+nvkm-y += nvkm/subdev/gsp/rm/r535/bar.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/fbsr.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/vmm.o
+
+nvkm-y += nvkm/subdev/gsp/rm/r535/disp.o
+
+nvkm-y += nvkm/subdev/gsp/rm/r535/fifo.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/ce.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/gr.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/nvdec.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/nvenc.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/nvjpg.o
+nvkm-y += nvkm/subdev/gsp/rm/r535/ofa.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/alloc.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/alloc.c
new file mode 100644
index 000000000000..46e3a29f2ad7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/alloc.c
@@ -0,0 +1,112 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/rpc.h>
+
+#include "nvrm/alloc.h"
+#include "nvrm/rpcfn.h"
+
+static int
+r535_gsp_rpc_rm_free(struct nvkm_gsp_object *object)
+{
+	struct nvkm_gsp_client *client = object->client;
+	struct nvkm_gsp *gsp = client->gsp;
+	rpc_free_v03_00 *rpc;
+
+	nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x free\n",
+		   client->object.handle, object->handle);
+
+	rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_FREE, sizeof(*rpc));
+	if (WARN_ON(IS_ERR_OR_NULL(rpc)))
+		return -EIO;
+
+	rpc->params.hRoot = client->object.handle;
+	rpc->params.hObjectParent = 0;
+	rpc->params.hObjectOld = object->handle;
+	return nvkm_gsp_rpc_wr(gsp, rpc, NVKM_GSP_RPC_REPLY_RECV);
+}
+
+static void
+r535_gsp_rpc_rm_alloc_done(struct nvkm_gsp_object *object, void *params)
+{
+	rpc_gsp_rm_alloc_v03_00 *rpc = to_payload_hdr(params, rpc);
+
+	nvkm_gsp_rpc_done(object->client->gsp, rpc);
+}
+
+static void *
+r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *params)
+{
+	rpc_gsp_rm_alloc_v03_00 *rpc = to_payload_hdr(params, rpc);
+	struct nvkm_gsp *gsp = object->client->gsp;
+	void *ret = NULL;
+
+	rpc = nvkm_gsp_rpc_push(gsp, rpc, NVKM_GSP_RPC_REPLY_RECV, sizeof(*rpc));
+	if (IS_ERR_OR_NULL(rpc))
+		return rpc;
+
+	if (rpc->status) {
+		ret = ERR_PTR(r535_rpc_status_to_errno(rpc->status));
+		if (PTR_ERR(ret) != -EAGAIN && PTR_ERR(ret) != -EBUSY)
+			nvkm_error(&gsp->subdev, "RM_ALLOC: 0x%x\n", rpc->status);
+	}
+
+	nvkm_gsp_rpc_done(gsp, rpc);
+
+	return ret;
+}
+
+static void *
+r535_gsp_rpc_rm_alloc_get(struct nvkm_gsp_object *object, u32 oclass,
+			  u32 params_size)
+{
+	struct nvkm_gsp_client *client = object->client;
+	struct nvkm_gsp *gsp = client->gsp;
+	rpc_gsp_rm_alloc_v03_00 *rpc;
+
+	nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x new obj:0x%08x\n",
+		   client->object.handle, object->parent->handle,
+		   object->handle);
+
+	nvkm_debug(&gsp->subdev, "cls:0x%08x params_size:%d\n", oclass,
+		   params_size);
+
+	rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_ALLOC,
+			       sizeof(*rpc) + params_size);
+	if (IS_ERR(rpc))
+		return rpc;
+
+	rpc->hClient = client->object.handle;
+	rpc->hParent = object->parent->handle;
+	rpc->hObject = object->handle;
+	rpc->hClass = oclass;
+	rpc->status = 0;
+	rpc->paramsSize = params_size;
+	return rpc->params;
+}
+
+const struct nvkm_rm_api_alloc
+r535_alloc = {
+	.get = r535_gsp_rpc_rm_alloc_get,
+	.push = r535_gsp_rpc_rm_alloc_push,
+	.done = r535_gsp_rpc_rm_alloc_done,
+	.free = r535_gsp_rpc_rm_free,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/bar.c
index 3a30bea30e36..d06bf95b9a4a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/bar.c
@@ -19,7 +19,7 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "gf100.h"
+#include <subdev/bar/gf100.h>
 
 #include <core/mm.h>
 #include <subdev/fb.h>
@@ -27,14 +27,20 @@
 #include <subdev/instmem.h>
 #include <subdev/mmu/vmm.h>
 
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/nvidia/generated/g_rpc-structures.h>
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_global_enums.h>
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_headers.h>
+#include "nvrm/bar.h"
+#include "nvrm/rpcfn.h"
 
 static void
 r535_bar_flush(struct nvkm_bar *bar)
 {
+	/* Use NV_PFLUSH in resume path - needed on R570 to flush writes before
+	 * BAR2 page tables have been restored.
+	 */
+	if (unlikely(!bar->bar2)) {
+		g84_bar_flush(bar);
+		return;
+	}
+
 	ioread32_native(bar->flushBAR2);
 }
 
@@ -44,7 +50,7 @@ r535_bar_bar2_wait(struct nvkm_bar *base)
 }
 
 static int
-r535_bar_bar2_update_pde(struct nvkm_gsp *gsp, u64 addr)
+r535_bar_bar2_update_pde(struct nvkm_gsp *gsp, u8 page_shift, u64 pdbe)
 {
 	rpc_update_bar_pde_v15_00 *rpc;
 
@@ -53,21 +59,22 @@ r535_bar_bar2_update_pde(struct nvkm_gsp *gsp, u64 addr)
 		return -EIO;
 
 	rpc->info.barType = NV_RPC_UPDATE_PDE_BAR_2;
-	rpc->info.entryValue = addr ? ((addr >> 4) | 2) : 0; /* PD3 entry format! */
-	rpc->info.entryLevelShift = 47; //XXX: probably fetch this from mmu!
+	rpc->info.entryValue = pdbe;
+	rpc->info.entryLevelShift = page_shift;
 
-	return nvkm_gsp_rpc_wr(gsp, rpc, true);
+	return nvkm_gsp_rpc_wr(gsp, rpc, NVKM_GSP_RPC_REPLY_RECV);
 }
 
 static void
 r535_bar_bar2_fini(struct nvkm_bar *bar)
 {
+	struct nvkm_vmm *vmm = gf100_bar(bar)->bar[0].vmm;
 	struct nvkm_gsp *gsp = bar->subdev.device->gsp;
 
 	bar->flushBAR2 = bar->flushBAR2PhysMode;
 	nvkm_done(bar->flushFBZero);
 
-	WARN_ON(r535_bar_bar2_update_pde(gsp, 0));
+	WARN_ON(r535_bar_bar2_update_pde(gsp, vmm->func->page[0].shift, 0));
 }
 
 static void
@@ -76,8 +83,18 @@ r535_bar_bar2_init(struct nvkm_bar *bar)
 	struct nvkm_device *device = bar->subdev.device;
 	struct nvkm_vmm *vmm = gf100_bar(bar)->bar[0].vmm;
 	struct nvkm_gsp *gsp = device->gsp;
-
-	WARN_ON(r535_bar_bar2_update_pde(gsp, vmm->pd->pde[0]->pt[0]->addr));
+	struct nvkm_memory *pdb = vmm->pd->pt[0]->memory;
+	u32 pdb_offset = vmm->pd->pt[0]->base;
+	u32 pdbe_lo, pdbe_hi;
+	u64 pdbe;
+
+	nvkm_kmap(pdb);
+	pdbe_lo = nvkm_ro32(pdb, pdb_offset + 0);
+	pdbe_hi = nvkm_ro32(pdb, pdb_offset + 4);
+	pdbe = ((u64)pdbe_hi << 32) | pdbe_lo;
+	nvkm_done(pdb);
+
+	WARN_ON(r535_bar_bar2_update_pde(gsp, vmm->func->page[0].shift, pdbe));
 	vmm->rm.bar2_pdb = gsp->bar.rm_bar2_pdb;
 
 	if (!bar->flushFBZero) {
@@ -174,7 +191,7 @@ r535_bar_new_(const struct nvkm_bar_func *hw, struct nvkm_device *device,
 	}
 	*pbar = bar;
 
-	bar->flushBAR2PhysMode = ioremap(device->func->resource_addr(device, 3), PAGE_SIZE);
+	bar->flushBAR2PhysMode = ioremap(device->func->resource_addr(device, NVKM_BAR2_INST), PAGE_SIZE);
 	if (!bar->flushBAR2PhysMode)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ce.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ce.c
new file mode 100644
index 000000000000..2d1ce9db2dcf
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ce.c
@@ -0,0 +1,46 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/engine.h>
+
+#include "nvrm/ce.h"
+#include "nvrm/engine.h"
+
+static int
+r535_ce_alloc(struct nvkm_gsp_object *chan, u32 handle, u32 class, int inst,
+	      struct nvkm_gsp_object *ce)
+{
+	NVC0B5_ALLOCATION_PARAMETERS *args;
+
+	args = nvkm_gsp_rm_alloc_get(chan, handle, class, sizeof(*args), ce);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
+
+	args->version = 1;
+	args->engineType = NV2080_ENGINE_TYPE_COPY0 + inst;
+
+	return nvkm_gsp_rm_alloc_wr(ce, args);
+}
+
+const struct nvkm_rm_api_engine
+r535_ce = {
+	.alloc = r535_ce_alloc,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ga100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/client.c
index 932934227b9c..ec71f683e609 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ga100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/client.c
@@ -19,26 +19,27 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "priv.h"
+#include <rm/rm.h>
 
-#include <subdev/gsp.h>
+#include "nvrm/client.h"
 
-#include <nvif/class.h>
+static int
+r535_gsp_client_ctor(struct nvkm_gsp_client *client, u32 handle)
+{
+	NV0000_ALLOC_PARAMETERS *args;
 
-static const struct nvkm_engine_func
-ga100_nvdec = {
-	.sclass = {
-		{ -1, -1, NVC6B0_VIDEO_DECODER },
-		{}
-	}
-};
+	args = nvkm_gsp_rm_alloc_get(&client->object, handle, NV01_ROOT, sizeof(*args),
+				     &client->object);
+	if (IS_ERR(args))
+		return PTR_ERR(args);
 
-int
-ga100_nvdec_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-		struct nvkm_nvdec **pnvdec)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvdec_new(&ga100_nvdec, device, type, inst, pnvdec);
+	args->hClient = client->object.handle;
+	args->processID = ~0;
 
-	return -ENODEV;
+	return nvkm_gsp_rm_alloc_wr(&client->object, args);
 }
+
+const struct nvkm_rm_api_client
+r535_client = {
+	.ctor = r535_gsp_client_ctor,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ctrl.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ctrl.c
new file mode 100644
index 000000000000..70b9ee911c5e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ctrl.c
@@ -0,0 +1,93 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/rpc.h>
+
+#include "nvrm/ctrl.h"
+#include "nvrm/rpcfn.h"
+
+static void
+r535_gsp_rpc_rm_ctrl_done(struct nvkm_gsp_object *object, void *params)
+{
+	rpc_gsp_rm_control_v03_00 *rpc = to_payload_hdr(params, rpc);
+
+	if (!params)
+		return;
+	nvkm_gsp_rpc_done(object->client->gsp, rpc);
+}
+
+static int
+r535_gsp_rpc_rm_ctrl_push(struct nvkm_gsp_object *object, void **params, u32 repc)
+{
+	rpc_gsp_rm_control_v03_00 *rpc = to_payload_hdr((*params), rpc);
+	struct nvkm_gsp *gsp = object->client->gsp;
+	int ret = 0;
+
+	rpc = nvkm_gsp_rpc_push(gsp, rpc, NVKM_GSP_RPC_REPLY_RECV, repc);
+	if (IS_ERR_OR_NULL(rpc)) {
+		*params = NULL;
+		return PTR_ERR(rpc);
+	}
+
+	if (rpc->status) {
+		ret = r535_rpc_status_to_errno(rpc->status);
+		if (ret != -EAGAIN && ret != -EBUSY)
+			nvkm_error(&gsp->subdev, "cli:0x%08x obj:0x%08x ctrl cmd:0x%08x failed: 0x%08x\n",
+				   object->client->object.handle, object->handle, rpc->cmd, rpc->status);
+	}
+
+	if (repc)
+		*params = rpc->params;
+	else
+		nvkm_gsp_rpc_done(gsp, rpc);
+
+	return ret;
+}
+
+static void *
+r535_gsp_rpc_rm_ctrl_get(struct nvkm_gsp_object *object, u32 cmd, u32 params_size)
+{
+	struct nvkm_gsp_client *client = object->client;
+	struct nvkm_gsp *gsp = client->gsp;
+	rpc_gsp_rm_control_v03_00 *rpc;
+
+	nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x ctrl cmd:0x%08x params_size:%d\n",
+		   client->object.handle, object->handle, cmd, params_size);
+
+	rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_CONTROL,
+			       sizeof(*rpc) + params_size);
+	if (IS_ERR(rpc))
+		return rpc;
+
+	rpc->hClient    = client->object.handle;
+	rpc->hObject    = object->handle;
+	rpc->cmd	= cmd;
+	rpc->status     = 0;
+	rpc->paramsSize = params_size;
+	return rpc->params;
+}
+
+const struct nvkm_rm_api_ctrl
+r535_ctrl = {
+	.get = r535_gsp_rpc_rm_ctrl_get,
+	.push = r535_gsp_rpc_rm_ctrl_push,
+	.done = r535_gsp_rpc_rm_ctrl_done,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/device.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/device.c
new file mode 100644
index 000000000000..f830e12a8f6e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/device.c
@@ -0,0 +1,148 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/rm.h>
+
+#include "nvrm/device.h"
+#include "nvrm/event.h"
+
+static void
+r535_gsp_event_dtor(struct nvkm_gsp_event *event)
+{
+	struct nvkm_gsp_device *device = event->device;
+	struct nvkm_gsp_client *client = device->object.client;
+	struct nvkm_gsp *gsp = client->gsp;
+
+	mutex_lock(&gsp->client_id.mutex);
+	if (event->func) {
+		list_del(&event->head);
+		event->func = NULL;
+	}
+	mutex_unlock(&gsp->client_id.mutex);
+
+	nvkm_gsp_rm_free(&event->object);
+	event->device = NULL;
+}
+
+static int
+r535_gsp_device_event_get(struct nvkm_gsp_event *event)
+{
+	struct nvkm_gsp_device *device = event->device;
+	NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&device->subdevice,
+				    NV2080_CTRL_CMD_EVENT_SET_NOTIFICATION, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->event = event->id;
+	ctrl->action = NV2080_CTRL_EVENT_SET_NOTIFICATION_ACTION_REPEAT;
+	return nvkm_gsp_rm_ctrl_wr(&device->subdevice, ctrl);
+}
+
+static int
+r535_gsp_device_event_ctor(struct nvkm_gsp_device *device, u32 handle, u32 id,
+			   nvkm_gsp_event_func func, struct nvkm_gsp_event *event)
+{
+	struct nvkm_gsp_client *client = device->object.client;
+	struct nvkm_gsp *gsp = client->gsp;
+	NV0005_ALLOC_PARAMETERS *args;
+	int ret;
+
+	args = nvkm_gsp_rm_alloc_get(&device->subdevice, handle,
+				     NV01_EVENT_KERNEL_CALLBACK_EX, sizeof(*args),
+				     &event->object);
+	if (IS_ERR(args))
+		return PTR_ERR(args);
+
+	args->hParentClient = client->object.handle;
+	args->hSrcResource = 0;
+	args->hClass = NV01_EVENT_KERNEL_CALLBACK_EX;
+	args->notifyIndex = NV01_EVENT_CLIENT_RM | id;
+	args->data = NULL;
+
+	ret = nvkm_gsp_rm_alloc_wr(&event->object, args);
+	if (ret)
+		return ret;
+
+	event->device = device;
+	event->id = id;
+
+	ret = r535_gsp_device_event_get(event);
+	if (ret) {
+		nvkm_gsp_event_dtor(event);
+		return ret;
+	}
+
+	mutex_lock(&gsp->client_id.mutex);
+	event->func = func;
+	list_add(&event->head, &client->events);
+	mutex_unlock(&gsp->client_id.mutex);
+	return 0;
+}
+
+static void
+r535_gsp_device_dtor(struct nvkm_gsp_device *device)
+{
+	nvkm_gsp_rm_free(&device->subdevice);
+	nvkm_gsp_rm_free(&device->object);
+}
+
+static int
+r535_gsp_subdevice_ctor(struct nvkm_gsp_device *device)
+{
+	NV2080_ALLOC_PARAMETERS *args;
+
+	return nvkm_gsp_rm_alloc(&device->object, NVKM_RM_SUBDEVICE, NV20_SUBDEVICE_0,
+				 sizeof(*args), &device->subdevice);
+}
+
+static int
+r535_gsp_device_ctor(struct nvkm_gsp_client *client, struct nvkm_gsp_device *device)
+{
+	NV0080_ALLOC_PARAMETERS *args;
+	int ret;
+
+	args = nvkm_gsp_rm_alloc_get(&client->object, NVKM_RM_DEVICE, NV01_DEVICE_0, sizeof(*args),
+				     &device->object);
+	if (IS_ERR(args))
+		return PTR_ERR(args);
+
+	args->hClientShare = client->object.handle;
+
+	ret = nvkm_gsp_rm_alloc_wr(&device->object, args);
+	if (ret)
+		return ret;
+
+	ret = r535_gsp_subdevice_ctor(device);
+	if (ret)
+		nvkm_gsp_rm_free(&device->object);
+
+	return ret;
+}
+
+const struct nvkm_rm_api_device
+r535_device = {
+	.ctor = r535_gsp_device_ctor,
+	.dtor = r535_gsp_device_dtor,
+	.event.ctor = r535_gsp_device_event_ctor,
+	.event.dtor = r535_gsp_event_dtor,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/disp.c
index 99110ab2f44d..7e9e2d3564da 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/disp.c
@@ -19,13 +19,13 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "priv.h"
-#include "chan.h"
-#include "conn.h"
-#include "dp.h"
-#include "head.h"
-#include "ior.h"
-#include "outp.h"
+#include <engine/disp/priv.h>
+#include <engine/disp/chan.h>
+#include <engine/disp/conn.h>
+#include <engine/disp/dp.h>
+#include <engine/disp/head.h>
+#include <engine/disp/ior.h>
+#include <engine/disp/outp.h>
 
 #include <core/ramht.h>
 #include <subdev/bios.h>
@@ -34,19 +34,11 @@
 #include <subdev/mmu.h>
 #include <subdev/vfn.h>
 
+#include <rm/gpu.h>
+
 #include <nvhw/drf.h>
 
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080_notification.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dfp.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073dp.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073specific.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0073/ctrl0073system.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-#include <nvrm/535.113.01/nvidia/generated/g_allclasses.h>
-#include <nvrm/535.113.01/nvidia/generated/g_mem_desc_nvoc.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/os/nv_memory_type.h>
+#include "nvrm/disp.h"
 
 #include <linux/acpi.h>
 
@@ -78,9 +70,9 @@ r535_chan_fini(struct nvkm_disp_chan *chan)
 }
 
 static int
-r535_chan_push(struct nvkm_disp_chan *chan)
+r535_disp_chan_set_pushbuf(struct nvkm_disp *disp, s32 oclass, int inst, struct nvkm_memory *memory)
 {
-	struct nvkm_gsp *gsp = chan->disp->engine.subdev.device->gsp;
+	struct nvkm_gsp *gsp = disp->rm.objcom.client->gsp;
 	NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS *ctrl;
 
 	ctrl = nvkm_gsp_rm_ctrl_get(&gsp->internal.device.subdevice,
@@ -89,8 +81,8 @@ r535_chan_push(struct nvkm_disp_chan *chan)
 	if (IS_ERR(ctrl))
 		return PTR_ERR(ctrl);
 
-	if (chan->memory) {
-		switch (nvkm_memory_target(chan->memory)) {
+	if (memory) {
+		switch (nvkm_memory_target(memory)) {
 		case NVKM_MEM_TARGET_NCOH:
 			ctrl->addressSpace = ADDR_SYSMEM;
 			ctrl->cacheSnoop = 0;
@@ -107,13 +99,13 @@ r535_chan_push(struct nvkm_disp_chan *chan)
 			return -EINVAL;
 		}
 
-		ctrl->physicalAddr = nvkm_memory_addr(chan->memory);
-		ctrl->limit = nvkm_memory_size(chan->memory) - 1;
+		ctrl->physicalAddr = nvkm_memory_addr(memory);
+		ctrl->limit = nvkm_memory_size(memory) - 1;
 	}
 
-	ctrl->hclass = chan->object.oclass;
-	ctrl->channelInstance = chan->head;
-	ctrl->valid = ((chan->object.oclass & 0xff) != 0x7a) ? 1 : 0;
+	ctrl->hclass = oclass;
+	ctrl->channelInstance = inst;
+	ctrl->valid = ((oclass & 0xff) != 0x7a) ? 1 : 0;
 
 	return nvkm_gsp_rm_ctrl_wr(&gsp->internal.device.subdevice, ctrl);
 }
@@ -121,10 +113,11 @@ r535_chan_push(struct nvkm_disp_chan *chan)
 static int
 r535_curs_init(struct nvkm_disp_chan *chan)
 {
+	const struct nvkm_rm_api *rmapi = chan->disp->rm.objcom.client->gsp->rm->api;
 	NV50VAIO_CHANNELPIO_ALLOCATION_PARAMETERS *args;
 	int ret;
 
-	ret = r535_chan_push(chan);
+	ret = rmapi->disp->chan.set_pushbuf(chan->disp, chan->object.oclass, chan->head, NULL);
 	if (ret)
 		return ret;
 
@@ -172,25 +165,34 @@ r535_dmac_fini(struct nvkm_disp_chan *chan)
 }
 
 static int
-r535_dmac_init(struct nvkm_disp_chan *chan)
+r535_dmac_alloc(struct nvkm_disp *disp, u32 oclass, int inst, u32 put_offset,
+		struct nvkm_gsp_object *dmac)
 {
 	NV50VAIO_CHANNELDMA_ALLOCATION_PARAMETERS *args;
-	int ret;
-
-	ret = r535_chan_push(chan);
-	if (ret)
-		return ret;
 
-	args = nvkm_gsp_rm_alloc_get(&chan->disp->rm.object,
-				     (chan->object.oclass << 16) | chan->head,
-				     chan->object.oclass, sizeof(*args), &chan->rm.object);
+	args = nvkm_gsp_rm_alloc_get(&disp->rm.object, (oclass << 16) | inst, oclass,
+				     sizeof(*args), dmac);
 	if (IS_ERR(args))
 		return PTR_ERR(args);
 
-	args->channelInstance = chan->head;
-	args->offset = chan->suspend_put;
+	args->channelInstance = inst;
+	args->offset = put_offset;
 
-	return nvkm_gsp_rm_alloc_wr(&chan->rm.object, args);
+	return nvkm_gsp_rm_alloc_wr(dmac, args);
+}
+
+static int
+r535_dmac_init(struct nvkm_disp_chan *chan)
+{
+	const struct nvkm_rm_api *rmapi = chan->disp->rm.objcom.client->gsp->rm->api;
+	int ret;
+
+	ret = rmapi->disp->chan.set_pushbuf(chan->disp, chan->object.oclass, chan->head, chan->memory);
+	if (ret)
+		return ret;
+
+	return rmapi->disp->chan.dmac_alloc(chan->disp, chan->object.oclass, chan->head,
+					    chan->suspend_put, &chan->rm.object);
 }
 
 static int
@@ -260,47 +262,47 @@ r535_core = {
 };
 
 static int
-r535_sor_bl_set(struct nvkm_ior *sor, int lvl)
+r535_bl_ctrl(struct nvkm_disp *disp, unsigned display_id, bool set, int *pval)
 {
-	struct nvkm_disp *disp = sor->disp;
+	u32 cmd = set ? NV0073_CTRL_CMD_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS :
+			NV0073_CTRL_CMD_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS;
 	NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS *ctrl;
+	int ret;
 
-	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
-				    NV0073_CTRL_CMD_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS,
-				    sizeof(*ctrl));
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom, cmd, sizeof(*ctrl));
 	if (IS_ERR(ctrl))
 		return PTR_ERR(ctrl);
 
-	ctrl->displayId = BIT(sor->asy.outp->index);
-	ctrl->brightness = lvl;
+	ctrl->displayId = BIT(display_id);
+	ctrl->brightness = *pval;
 
-	return nvkm_gsp_rm_ctrl_wr(&disp->rm.objcom, ctrl);
+	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
+	if (ret)
+		return ret;
+
+	*pval = ctrl->brightness;
+
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
 }
 
 static int
-r535_sor_bl_get(struct nvkm_ior *sor)
+r535_sor_bl_set(struct nvkm_ior *sor, int lvl)
 {
 	struct nvkm_disp *disp = sor->disp;
-	NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS *ctrl;
-	int ret, lvl;
-
-	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
-				    NV0073_CTRL_CMD_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS,
-				    sizeof(*ctrl));
-	if (IS_ERR(ctrl))
-		return PTR_ERR(ctrl);
+	const struct nvkm_rm_api *rmapi = disp->engine.subdev.device->gsp->rm->api;
 
-	ctrl->displayId = BIT(sor->asy.outp->index);
+	return rmapi->disp->bl_ctrl(disp, sor->asy.outp->index, true, &lvl);
+}
 
-	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
-	if (ret) {
-		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
-		return ret;
-	}
+static int
+r535_sor_bl_get(struct nvkm_ior *sor)
+{
+	struct nvkm_disp *disp = sor->disp;
+	const struct nvkm_rm_api *rmapi = disp->engine.subdev.device->gsp->rm->api;
+	int lvl, ret = rmapi->disp->bl_ctrl(disp, sor->asy.outp->index, false, &lvl);
 
-	lvl = ctrl->brightness;
-	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
-	return lvl;
+	return (ret == 0) ? lvl : ret;
 }
 
 static const struct nvkm_ior_func_bl
@@ -730,7 +732,7 @@ r535_outp_acquire(struct nvkm_outp *outp, bool hda)
 }
 
 static int
-r535_disp_head_displayid(struct nvkm_disp *disp, int head, u32 *displayid)
+r535_disp_get_active(struct nvkm_disp *disp, unsigned head, u32 *displayid)
 {
 	NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS *ctrl;
 	int ret;
@@ -763,7 +765,9 @@ r535_outp_inherit(struct nvkm_outp *outp)
 	int ret;
 
 	list_for_each_entry(head, &disp->heads, head) {
-		ret = r535_disp_head_displayid(disp, head->id, &displayid);
+		const struct nvkm_rm_api *rmapi = disp->rm.objcom.client->gsp->rm->api;
+
+		ret = rmapi->disp->get_active(disp, head->id, &displayid);
 		if (WARN_ON(ret))
 			return NULL;
 
@@ -858,10 +862,9 @@ r535_outp_dfp_get_info(struct nvkm_outp *outp)
 }
 
 static int
-r535_outp_detect(struct nvkm_outp *outp)
+r535_disp_get_connect_state(struct nvkm_disp *disp, unsigned display_id)
 {
 	NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS *ctrl;
-	struct nvkm_disp *disp = outp->disp;
 	int ret;
 
 	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
@@ -870,23 +873,29 @@ r535_outp_detect(struct nvkm_outp *outp)
 		return PTR_ERR(ctrl);
 
 	ctrl->subDeviceInstance = 0;
-	ctrl->displayMask = BIT(outp->index);
+	ctrl->displayMask = BIT(display_id);
 
 	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
-	if (ret) {
-		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
-		return ret;
-	}
+	if (ret == 0 && (ctrl->displayMask & BIT(display_id)))
+		ret = 1;
+
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return ret;
+}
 
-	if (ctrl->displayMask & BIT(outp->index)) {
+static int
+r535_outp_detect(struct nvkm_outp *outp)
+{
+	const struct nvkm_rm_api *rmapi = outp->disp->rm.objcom.client->gsp->rm->api;
+	int ret;
+
+	ret = rmapi->disp->get_connect_state(outp->disp, outp->index);
+	if (ret == 1) {
 		ret = r535_outp_dfp_get_info(outp);
 		if (ret == 0)
 			ret = 1;
-	} else {
-		ret = 0;
 	}
 
-	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
 	return ret;
 }
 
@@ -1029,15 +1038,11 @@ r535_dp_train(struct nvkm_outp *outp, bool retrain)
 }
 
 static int
-r535_dp_rates(struct nvkm_outp *outp)
+r535_dp_set_indexed_link_rates(struct nvkm_outp *outp)
 {
 	NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS *ctrl;
 	struct nvkm_disp *disp = outp->disp;
 
-	if (outp->conn->info.type != DCB_CONNECTOR_eDP ||
-	    !outp->dp.rates || outp->dp.rate[0].dpcd < 0)
-		return 0;
-
 	if (WARN_ON(outp->dp.rates > ARRAY_SIZE(ctrl->linkRateTbl)))
 		return -EINVAL;
 
@@ -1054,6 +1059,18 @@ r535_dp_rates(struct nvkm_outp *outp)
 }
 
 static int
+r535_dp_rates(struct nvkm_outp *outp)
+{
+	struct nvkm_rm *rm = outp->disp->rm.objcom.client->gsp->rm;
+
+	if (outp->conn->info.type != DCB_CONNECTOR_eDP ||
+	    !outp->dp.rates || outp->dp.rate[0].dpcd < 0)
+		return 0;
+
+	return rm->api->disp->dp.set_indexed_link_rates(outp);
+}
+
+static int
 r535_dp_aux_xfer(struct nvkm_outp *outp, u8 type, u32 addr, u8 *data, u8 *psize)
 {
 	struct nvkm_disp *disp = outp->disp;
@@ -1151,6 +1168,49 @@ r535_dp = {
 };
 
 static int
+r535_dp_get_caps(struct nvkm_disp *disp, int *plink_bw, bool *pmst, bool *pwm)
+{
+	NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS *ctrl;
+	int ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
+				    NV0073_CTRL_CMD_DP_GET_CAPS, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->sorIndex = ~0;
+
+	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
+	if (ret) {
+		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+		return ret;
+	}
+
+	switch (NVVAL_GET(ctrl->maxLinkRate, NV0073_CTRL_CMD, DP_GET_CAPS, MAX_LINK_RATE)) {
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_1_62:
+		*plink_bw = 0x06;
+		break;
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_2_70:
+		*plink_bw = 0x0a;
+		break;
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_5_40:
+		*plink_bw = 0x14;
+		break;
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_8_10:
+		*plink_bw = 0x1e;
+		break;
+	default:
+		*plink_bw = 0x00;
+		break;
+	}
+
+	*pmst = ctrl->bIsMultistreamSupported;
+	*pwm = ctrl->bHasIncreasedWatermarkLimits;
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
+}
+
+static int
 r535_tmds_edid_get(struct nvkm_outp *outp, u8 *data, u16 *psize)
 {
 	NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS *ctrl;
@@ -1194,6 +1254,7 @@ r535_tmds = {
 static int
 r535_outp_new(struct nvkm_disp *disp, u32 id)
 {
+	const struct nvkm_rm_api *rmapi = disp->rm.objcom.client->gsp->rm->api;
 	NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS *ctrl;
 	enum nvkm_ior_proto proto;
 	struct dcb_output dcbE = {};
@@ -1278,43 +1339,11 @@ r535_outp_new(struct nvkm_disp *disp, u32 id)
 		if (ret)
 			return ret;
 	} else {
-		NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS *ctrl;
 		bool mst, wm;
 
-		ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
-					    NV0073_CTRL_CMD_DP_GET_CAPS, sizeof(*ctrl));
-		if (IS_ERR(ctrl))
-			return PTR_ERR(ctrl);
-
-		ctrl->sorIndex = ~0;
-
-		ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
-		if (ret) {
-			nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+		ret = rmapi->disp->dp.get_caps(disp, &dcbE.dpconf.link_bw, &mst, &wm);
+		if (ret)
 			return ret;
-		}
-
-		switch (NVVAL_GET(ctrl->maxLinkRate, NV0073_CTRL_CMD, DP_GET_CAPS, MAX_LINK_RATE)) {
-		case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_1_62:
-			dcbE.dpconf.link_bw = 0x06;
-			break;
-		case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_2_70:
-			dcbE.dpconf.link_bw = 0x0a;
-			break;
-		case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_5_40:
-			dcbE.dpconf.link_bw = 0x14;
-			break;
-		case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_8_10:
-			dcbE.dpconf.link_bw = 0x1e;
-			break;
-		default:
-			dcbE.dpconf.link_bw = 0x00;
-			break;
-		}
-
-		mst = ctrl->bIsMultistreamSupported;
-		wm = ctrl->bHasIncreasedWatermarkLimits;
-		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
 
 		if (WARN_ON(!dcbE.dpconf.link_bw))
 			return -EINVAL;
@@ -1441,11 +1470,47 @@ r535_disp_init(struct nvkm_disp *disp)
 }
 
 static int
+r535_disp_get_supported(struct nvkm_disp *disp, unsigned long *pmask)
+{
+	NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_rd(&disp->rm.objcom,
+				   NV0073_CTRL_CMD_SYSTEM_GET_SUPPORTED, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	*pmask = ctrl->displayMask;
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
+}
+
+static int
+r535_disp_get_static_info(struct nvkm_disp *disp)
+{
+	NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS *ctrl;
+	struct nvkm_gsp *gsp = disp->rm.objcom.client->gsp;
+
+	ctrl = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
+				   NV2080_CTRL_CMD_INTERNAL_DISPLAY_GET_STATIC_INFO,
+				   sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	disp->wndw.mask = ctrl->windowPresentMask;
+	disp->wndw.nr = fls(disp->wndw.mask);
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
+	return 0;
+}
+
+static int
 r535_disp_oneinit(struct nvkm_disp *disp)
 {
 	struct nvkm_device *device = disp->engine.subdev.device;
 	struct nvkm_gsp *gsp = device->gsp;
+	const struct nvkm_rm_api *rmapi = gsp->rm->api;
 	NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS *ctrl;
+	unsigned long mask;
 	int ret, i;
 
 	/* RAMIN. */
@@ -1476,24 +1541,14 @@ r535_disp_oneinit(struct nvkm_disp *disp)
 	if (ret)
 		return ret;
 
-	ret = nvkm_gsp_rm_alloc(&disp->rm.device.object, 0x00730000, NV04_DISPLAY_COMMON, 0,
+	ret = nvkm_gsp_rm_alloc(&disp->rm.device.object, NVKM_RM_DISP, NV04_DISPLAY_COMMON, 0,
 				&disp->rm.objcom);
 	if (ret)
 		return ret;
 
-	{
-		NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS *ctrl;
-
-		ctrl = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
-					   NV2080_CTRL_CMD_INTERNAL_DISPLAY_GET_STATIC_INFO,
-					   sizeof(*ctrl));
-		if (IS_ERR(ctrl))
-			return PTR_ERR(ctrl);
-
-		disp->wndw.mask = ctrl->windowPresentMask;
-		disp->wndw.nr = fls(disp->wndw.mask);
-		nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
-	}
+	ret = rmapi->disp->get_static_info(disp);
+	if (ret)
+		return ret;
 
 	/* */
 	{
@@ -1622,25 +1677,14 @@ r535_disp_oneinit(struct nvkm_disp *disp)
 			return ret;
 	}
 
-	/* */
-	{
-		NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS *ctrl;
-		unsigned long mask;
-		int i;
-
-		ctrl = nvkm_gsp_rm_ctrl_rd(&disp->rm.objcom,
-					   NV0073_CTRL_CMD_SYSTEM_GET_SUPPORTED, sizeof(*ctrl));
-		if (IS_ERR(ctrl))
-			return PTR_ERR(ctrl);
-
-		mask = ctrl->displayMask;
-		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	ret = rmapi->disp->get_supported(disp, &mask);
+	if (ret)
+		return ret;
 
-		for_each_set_bit(i, &mask, 32) {
-			ret = r535_outp_new(disp, i);
-			if (ret)
-				return ret;
-		}
+	for_each_set_bit(i, &mask, 32) {
+		ret = r535_outp_new(disp, i);
+		if (ret)
+			return ret;
 	}
 
 	ret = nvkm_event_init(&r535_disp_event, &gsp->subdev, 3, 32, &disp->rm.event);
@@ -1686,6 +1730,7 @@ int
 r535_disp_new(const struct nvkm_disp_func *hw, struct nvkm_device *device,
 	      enum nvkm_subdev_type type, int inst, struct nvkm_disp **pdisp)
 {
+	const struct nvkm_rm_gpu *gpu = device->gsp->rm->gpu;
 	struct nvkm_disp_func *rm;
 	int ret;
 
@@ -1701,20 +1746,26 @@ r535_disp_new(const struct nvkm_disp_func *hw, struct nvkm_device *device,
 	rm->sor.new = r535_sor_new;
 	rm->ramht_size = hw->ramht_size;
 
-	rm->root = hw->root;
+	rm->root.oclass = gpu->disp.class.root;
 
-	for (int i = 0; hw->user[i].ctor; i++) {
-		switch (hw->user[i].base.oclass & 0xff) {
-		case 0x73: rm->user[i] = hw->user[i]; break;
-		case 0x7d: rm->user[i] = hw->user[i]; rm->user[i].chan = &r535_core; break;
-		case 0x7e: rm->user[i] = hw->user[i]; rm->user[i].chan = &r535_wndw; break;
-		case 0x7b: rm->user[i] = hw->user[i]; rm->user[i].chan = &r535_wimm; break;
-		case 0x7a: rm->user[i] = hw->user[i]; rm->user[i].chan = &r535_curs; break;
-		default:
-			WARN_ON(1);
-			continue;
-		}
-	}
+	rm->user[0].base.oclass = gpu->disp.class.caps;
+	rm->user[0].ctor = gv100_disp_caps_new;
+
+	rm->user[1].base.oclass = gpu->disp.class.core;
+	rm->user[1].ctor = nvkm_disp_core_new;
+	rm->user[1].chan = &r535_core;
+
+	rm->user[2].base.oclass = gpu->disp.class.wndw;
+	rm->user[2].ctor = nvkm_disp_wndw_new;
+	rm->user[2].chan = &r535_wndw;
+
+	rm->user[3].base.oclass = gpu->disp.class.wimm;
+	rm->user[3].ctor = nvkm_disp_wndw_new;
+	rm->user[3].chan = &r535_wimm;
+
+	rm->user[4].base.oclass = gpu->disp.class.curs;
+	rm->user[4].ctor = nvkm_disp_chan_new;
+	rm->user[4].chan = &r535_curs;
 
 	ret = nvkm_disp_new_(rm, device, type, inst, pdisp);
 	if (ret)
@@ -1723,3 +1774,20 @@ r535_disp_new(const struct nvkm_disp_func *hw, struct nvkm_device *device,
 	mutex_init(&(*pdisp)->super.mutex); //XXX
 	return ret;
 }
+
+const struct nvkm_rm_api_disp
+r535_disp = {
+	.get_static_info = r535_disp_get_static_info,
+	.get_supported = r535_disp_get_supported,
+	.get_connect_state = r535_disp_get_connect_state,
+	.get_active = r535_disp_get_active,
+	.bl_ctrl = r535_bl_ctrl,
+	.dp = {
+		.get_caps = r535_dp_get_caps,
+		.set_indexed_link_rates = r535_dp_set_indexed_link_rates,
+	},
+	.chan = {
+		.set_pushbuf = r535_disp_chan_set_pushbuf,
+		.dmac_alloc = r535_dmac_alloc,
+	}
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/fbsr.c
index 5f3c9c02a4c0..150e22fde2ac 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/fbsr.c
@@ -19,19 +19,13 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "priv.h"
-
+#include <subdev/instmem/priv.h>
 #include <subdev/gsp.h>
 
 #include <nvhw/drf.h>
 
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl84a0.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-#include <nvrm/535.113.01/nvidia/generated/g_fbsr_nvoc.h>
-#include <nvrm/535.113.01/nvidia/generated/g_rpc-structures.h>
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_global_enums.h>
+#include "nvrm/fbsr.h"
+#include "nvrm/rpcfn.h"
 
 struct fbsr_item {
 	const char *type;
@@ -54,9 +48,9 @@ struct fbsr {
 	u64 sys_offset;
 };
 
-static int
-fbsr_memlist(struct nvkm_gsp_device *device, u32 handle, enum nvkm_memory_target aper,
-	     u64 phys, u64 size, struct sg_table *sgt, struct nvkm_gsp_object *object)
+int
+r535_fbsr_memlist(struct nvkm_gsp_device *device, u32 handle, enum nvkm_memory_target aper,
+		  u64 phys, u64 size, struct sg_table *sgt, struct nvkm_gsp_object *object)
 {
 	struct nvkm_gsp_client *client = device->object.client;
 	struct nvkm_gsp *gsp = client->gsp;
@@ -105,7 +99,7 @@ fbsr_memlist(struct nvkm_gsp_device *device, u32 handle, enum nvkm_memory_target
 			rpc->pteDesc.pte_pde[i].pte = (phys >> 12) + i;
 	}
 
-	ret = nvkm_gsp_rpc_wr(gsp, rpc, true);
+	ret = nvkm_gsp_rpc_wr(gsp, rpc, NVKM_GSP_RPC_REPLY_POLL);
 	if (ret)
 		return ret;
 
@@ -123,8 +117,8 @@ fbsr_send(struct fbsr *fbsr, struct fbsr_item *item)
 	struct nvkm_gsp_object memlist;
 	int ret;
 
-	ret = fbsr_memlist(&fbsr->device, fbsr->hmemory, NVKM_MEM_TARGET_VRAM,
-			   item->addr, item->size, NULL, &memlist);
+	ret = r535_fbsr_memlist(&fbsr->device, fbsr->hmemory, NVKM_MEM_TARGET_VRAM,
+				item->addr, item->size, NULL, &memlist);
 	if (ret)
 		return ret;
 
@@ -161,8 +155,8 @@ fbsr_init(struct fbsr *fbsr, struct sg_table *sgt, u64 items_size)
 	struct nvkm_gsp_object memlist;
 	int ret;
 
-	ret = fbsr_memlist(&fbsr->device, fbsr->hmemory, NVKM_MEM_TARGET_HOST,
-			   0, fbsr->size, sgt, &memlist);
+	ret = r535_fbsr_memlist(&fbsr->device, fbsr->hmemory, NVKM_MEM_TARGET_HOST,
+				0, fbsr->size, sgt, &memlist);
 	if (ret)
 		return ret;
 
@@ -206,22 +200,19 @@ fbsr_inst(struct fbsr *fbsr, const char *type, struct nvkm_memory *memory)
 	return fbsr_vram(fbsr, type, nvkm_memory_addr(memory), nvkm_memory_size(memory));
 }
 
-static void
-r535_instmem_resume(struct nvkm_instmem *imem)
+void
+r535_fbsr_resume(struct nvkm_gsp *gsp)
 {
 	/* RM has restored VRAM contents already, so just need to free the sysmem buffer. */
-	if (imem->rm.fbsr_valid) {
-		nvkm_gsp_sg_free(imem->subdev.device, &imem->rm.fbsr);
-		imem->rm.fbsr_valid = false;
-	}
+	nvkm_gsp_sg_free(gsp->subdev.device, &gsp->sr.fbsr);
 }
 
 static int
-r535_instmem_suspend(struct nvkm_instmem *imem)
+r535_fbsr_suspend(struct nvkm_gsp *gsp)
 {
-	struct nvkm_subdev *subdev = &imem->subdev;
+	struct nvkm_subdev *subdev = &gsp->subdev;
 	struct nvkm_device *device = subdev->device;
-	struct nvkm_gsp *gsp = device->gsp;
+	struct nvkm_instmem *imem = device->imem;
 	struct nvkm_instobj *iobj;
 	struct fbsr fbsr = {};
 	struct fbsr_item *item, *temp;
@@ -262,7 +253,7 @@ r535_instmem_suspend(struct nvkm_instmem *imem)
 	fbsr.size += gsp->fb.bios.vga_workspace.size;
 	nvkm_debug(subdev, "fbsr: size: 0x%llx bytes\n", fbsr.size);
 
-	ret = nvkm_gsp_sg(gsp->subdev.device, fbsr.size, &imem->rm.fbsr);
+	ret = nvkm_gsp_sg(gsp->subdev.device, fbsr.size, &gsp->sr.fbsr);
 	if (ret)
 		goto done;
 
@@ -271,7 +262,7 @@ r535_instmem_suspend(struct nvkm_instmem *imem)
 	if (ret)
 		goto done_sgt;
 
-	ret = fbsr_init(&fbsr, &imem->rm.fbsr, items_size);
+	ret = fbsr_init(&fbsr, &gsp->sr.fbsr, items_size);
 	if (WARN_ON(ret))
 		goto done_sgt;
 
@@ -282,12 +273,10 @@ r535_instmem_suspend(struct nvkm_instmem *imem)
 			goto done_sgt;
 	}
 
-	imem->rm.fbsr_valid = true;
-
 	/* Cleanup everything except the sysmem backup, which will be removed after resume. */
 done_sgt:
 	if (ret) /* ... unless we failed already. */
-		nvkm_gsp_sg_free(device, &imem->rm.fbsr);
+		nvkm_gsp_sg_free(device, &gsp->sr.fbsr);
 done:
 	list_for_each_entry_safe(item, temp, &fbsr.items, head) {
 		list_del(&item->head);
@@ -299,6 +288,12 @@ done:
 	return ret;
 }
 
+const struct nvkm_rm_api_fbsr
+r535_fbsr = {
+	.suspend = r535_fbsr_suspend,
+	.resume = r535_fbsr_resume,
+};
+
 static void *
 r535_instmem_dtor(struct nvkm_instmem *imem)
 {
@@ -319,11 +314,10 @@ r535_instmem_new(const struct nvkm_instmem_func *hw,
 
 	rm->dtor = r535_instmem_dtor;
 	rm->fini = hw->fini;
-	rm->suspend = r535_instmem_suspend;
-	rm->resume  = r535_instmem_resume;
 	rm->memory_new = hw->memory_new;
 	rm->memory_wrap = hw->memory_wrap;
 	rm->zero = false;
+	rm->set_bar0_window_addr = hw->set_bar0_window_addr;
 
 	ret = nv50_instmem_new_(rm, device, type, inst, pinstmem);
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/fifo.c
index 3454c7d29502..1ac5628c5140 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/fifo.c
@@ -19,11 +19,11 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "priv.h"
-#include "cgrp.h"
-#include "chan.h"
-#include "chid.h"
-#include "runl.h"
+#include <engine/fifo/priv.h>
+#include <engine/fifo/cgrp.h>
+#include <engine/fifo/chan.h>
+#include <engine/fifo/chid.h>
+#include <engine/fifo/runl.h>
 
 #include <core/gpuobj.h>
 #include <subdev/gsp.h>
@@ -31,24 +31,19 @@
 #include <subdev/vfn.h>
 #include <engine/gr.h>
 
+#include <rm/engine.h>
+
 #include <nvhw/drf.h>
 
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/alloc/alloc_channel.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080_notification.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080ce.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080fifo.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrla06f/ctrla06fgpfifo.h>
-#include <nvrm/535.113.01/nvidia/generated/g_kernel_channel_nvoc.h>
-#include <nvrm/535.113.01/nvidia/generated/g_kernel_fifo_nvoc.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/gpu_engine_type.h>
+#include "nvrm/fifo.h"
+#include "nvrm/engine.h"
 
 static u32
 r535_chan_doorbell_handle(struct nvkm_chan *chan)
 {
-	return (chan->cgrp->runl->id << 16) | chan->id;
+	struct nvkm_gsp *gsp = chan->rm.object.client->gsp;
+
+	return gsp->rm->gpu->fifo.chan.doorbell_handle(chan);
 }
 
 static void
@@ -77,50 +72,29 @@ r535_chan_ramfc_clear(struct nvkm_chan *chan)
 #define CHID_PER_USERD 8
 
 static int
-r535_chan_ramfc_write(struct nvkm_chan *chan, u64 offset, u64 length, u32 devm, bool priv)
+r535_chan_alloc(struct nvkm_gsp_device *device, u32 handle, u32 nv2080_engine_type, u8 runq,
+		bool priv, int chid, u64 inst_addr, u64 userd_addr, u64 mthdbuf_addr,
+		struct nvkm_vmm *vmm, u64 gpfifo_offset, u32 gpfifo_length,
+		struct nvkm_gsp_object *chan)
 {
-	struct nvkm_fifo *fifo = chan->cgrp->runl->fifo;
-	struct nvkm_engn *engn;
-	struct nvkm_device *device = fifo->engine.subdev.device;
+	struct nvkm_gsp *gsp = device->object.client->gsp;
+	struct nvkm_fifo *fifo = gsp->subdev.device->fifo;
+	const int userd_p = chid / CHID_PER_USERD;
+	const int userd_i = chid % CHID_PER_USERD;
 	NV_CHANNELGPFIFO_ALLOCATION_PARAMETERS *args;
-	const int userd_p = chan->id / CHID_PER_USERD;
-	const int userd_i = chan->id % CHID_PER_USERD;
-	u32 eT = ~0;
-	int ret;
-
-	if (unlikely(device->gr && !device->gr->engine.subdev.oneinit)) {
-		ret = nvkm_subdev_oneinit(&device->gr->engine.subdev);
-		if (ret)
-			return ret;
-	}
-
-	nvkm_runl_foreach_engn(engn, chan->cgrp->runl) {
-		eT = engn->id;
-		break;
-	}
-
-	if (WARN_ON(eT == ~0))
-		return -EINVAL;
 
-	chan->rm.mthdbuf.ptr = dma_alloc_coherent(fifo->engine.subdev.device->dev,
-						  fifo->rm.mthdbuf_size,
-						  &chan->rm.mthdbuf.addr, GFP_KERNEL);
-	if (!chan->rm.mthdbuf.ptr)
-		return -ENOMEM;
-
-	args = nvkm_gsp_rm_alloc_get(&chan->vmm->rm.device.object, 0xf1f00000 | chan->id,
-				     fifo->func->chan.user.oclass, sizeof(*args),
-				     &chan->rm.object);
+	args = nvkm_gsp_rm_alloc_get(&device->object, handle,
+				     fifo->func->chan.user.oclass, sizeof(*args), chan);
 	if (WARN_ON(IS_ERR(args)))
 		return PTR_ERR(args);
 
-	args->gpFifoOffset = offset;
-	args->gpFifoEntries = length / 8;
+	args->gpFifoOffset = gpfifo_offset;
+	args->gpFifoEntries = gpfifo_length / 8;
 
 	args->flags  = NVDEF(NVOS04, FLAGS, CHANNEL_TYPE, PHYSICAL);
 	args->flags |= NVDEF(NVOS04, FLAGS, VPR, FALSE);
 	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_SKIP_MAP_REFCOUNTING, FALSE);
-	args->flags |= NVVAL(NVOS04, FLAGS, GROUP_CHANNEL_RUNQUEUE, chan->runq);
+	args->flags |= NVVAL(NVOS04, FLAGS, GROUP_CHANNEL_RUNQUEUE, runq);
 	if (!priv)
 		args->flags |= NVDEF(NVOS04, FLAGS, PRIVILEGED_CHANNEL, FALSE);
 	else
@@ -143,25 +117,25 @@ r535_chan_ramfc_write(struct nvkm_chan *chan, u64 offset, u64 length, u32 devm,
 	args->flags |= NVDEF(NVOS04, FLAGS, MAP_CHANNEL, FALSE);
 	args->flags |= NVDEF(NVOS04, FLAGS, SKIP_CTXBUFFER_ALLOC, FALSE);
 
-	args->hVASpace = chan->vmm->rm.object.handle;
-	args->engineType = eT;
+	args->hVASpace = vmm->rm.object.handle;
+	args->engineType = nv2080_engine_type;
 
-	args->instanceMem.base = chan->inst->addr;
-	args->instanceMem.size = chan->inst->size;
+	args->instanceMem.base = inst_addr;
+	args->instanceMem.size = fifo->func->chan.func->inst->size;
 	args->instanceMem.addressSpace = 2;
 	args->instanceMem.cacheAttrib = 1;
 
-	args->userdMem.base = nvkm_memory_addr(chan->userd.mem) + chan->userd.base;
+	args->userdMem.base = userd_addr;
 	args->userdMem.size = fifo->func->chan.func->userd->size;
 	args->userdMem.addressSpace = 2;
 	args->userdMem.cacheAttrib = 1;
 
-	args->ramfcMem.base = chan->inst->addr + 0;
+	args->ramfcMem.base = inst_addr;
 	args->ramfcMem.size = 0x200;
 	args->ramfcMem.addressSpace = 2;
 	args->ramfcMem.cacheAttrib = 1;
 
-	args->mthdbufMem.base = chan->rm.mthdbuf.addr;
+	args->mthdbufMem.base = mthdbuf_addr;
 	args->mthdbufMem.size = fifo->rm.mthdbuf_size;
 	args->mthdbufMem.addressSpace = 1;
 	args->mthdbufMem.cacheAttrib = 0;
@@ -173,7 +147,44 @@ r535_chan_ramfc_write(struct nvkm_chan *chan, u64 offset, u64 length, u32 devm,
 	args->internalFlags |= NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, ERROR_NOTIFIER_TYPE, NONE);
 	args->internalFlags |= NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, ECC_ERROR_NOTIFIER_TYPE, NONE);
 
-	ret = nvkm_gsp_rm_alloc_wr(&chan->rm.object, args);
+	return nvkm_gsp_rm_alloc_wr(chan, args);
+}
+
+static int
+r535_chan_ramfc_write(struct nvkm_chan *chan, u64 offset, u64 length, u32 devm, bool priv)
+{
+	struct nvkm_fifo *fifo = chan->cgrp->runl->fifo;
+	struct nvkm_engn *engn;
+	struct nvkm_device *device = fifo->engine.subdev.device;
+	const struct nvkm_rm_api *rmapi = device->gsp->rm->api;
+	u32 eT = ~0;
+	int ret;
+
+	if (unlikely(device->gr && !device->gr->engine.subdev.oneinit)) {
+		ret = nvkm_subdev_oneinit(&device->gr->engine.subdev);
+		if (ret)
+			return ret;
+	}
+
+	nvkm_runl_foreach_engn(engn, chan->cgrp->runl) {
+		eT = engn->id;
+		break;
+	}
+
+	if (WARN_ON(eT == ~0))
+		return -EINVAL;
+
+	chan->rm.mthdbuf.ptr = dma_alloc_coherent(fifo->engine.subdev.device->dev,
+						  fifo->rm.mthdbuf_size,
+						  &chan->rm.mthdbuf.addr, GFP_KERNEL);
+	if (!chan->rm.mthdbuf.ptr)
+		return -ENOMEM;
+
+	ret = rmapi->fifo->chan.alloc(&chan->vmm->rm.device, NVKM_RM_CHAN(chan->id),
+				      eT, chan->runq, priv, chan->id, chan->inst->addr,
+				      nvkm_memory_addr(chan->userd.mem) + chan->userd.base,
+				      chan->rm.mthdbuf.addr, chan->vmm, offset, length,
+				      &chan->rm.object);
 	if (ret)
 		return ret;
 
@@ -215,123 +226,8 @@ r535_chan_ramfc = {
 	.priv = true,
 };
 
-struct r535_chan_userd {
-	struct nvkm_memory *mem;
-	struct nvkm_memory *map;
-	int chid;
-	u32 used;
-
-	struct list_head head;
-} *userd;
-
-static void
-r535_chan_id_put(struct nvkm_chan *chan)
-{
-	struct nvkm_runl *runl = chan->cgrp->runl;
-	struct nvkm_fifo *fifo = runl->fifo;
-	struct r535_chan_userd *userd;
-
-	mutex_lock(&fifo->userd.mutex);
-	list_for_each_entry(userd, &fifo->userd.list, head) {
-		if (userd->map == chan->userd.mem) {
-			u32 chid = chan->userd.base / chan->func->userd->size;
-
-			userd->used &= ~BIT(chid);
-			if (!userd->used) {
-				nvkm_memory_unref(&userd->map);
-				nvkm_memory_unref(&userd->mem);
-				nvkm_chid_put(runl->chid, userd->chid, &chan->cgrp->lock);
-				list_del(&userd->head);
-				kfree(userd);
-			}
-
-			break;
-		}
-	}
-	mutex_unlock(&fifo->userd.mutex);
-
-}
-
-static int
-r535_chan_id_get_locked(struct nvkm_chan *chan, struct nvkm_memory *muserd, u64 ouserd)
-{
-	const u32 userd_size = CHID_PER_USERD * chan->func->userd->size;
-	struct nvkm_runl *runl = chan->cgrp->runl;
-	struct nvkm_fifo *fifo = runl->fifo;
-	struct r535_chan_userd *userd;
-	u32 chid;
-	int ret;
-
-	if (ouserd + chan->func->userd->size >= userd_size ||
-	    (ouserd & (chan->func->userd->size - 1))) {
-		RUNL_DEBUG(runl, "ouserd %llx", ouserd);
-		return -EINVAL;
-	}
-
-	chid = div_u64(ouserd, chan->func->userd->size);
-
-	list_for_each_entry(userd, &fifo->userd.list, head) {
-		if (userd->mem == muserd) {
-			if (userd->used & BIT(chid))
-				return -EBUSY;
-			break;
-		}
-	}
-
-	if (&userd->head == &fifo->userd.list) {
-		if (nvkm_memory_size(muserd) < userd_size) {
-			RUNL_DEBUG(runl, "userd too small");
-			return -EINVAL;
-		}
-
-		userd = kzalloc(sizeof(*userd), GFP_KERNEL);
-		if (!userd)
-			return -ENOMEM;
-
-		userd->chid = nvkm_chid_get(runl->chid, chan);
-		if (userd->chid < 0) {
-			ret = userd->chid;
-			kfree(userd);
-			return ret;
-		}
-
-		userd->mem = nvkm_memory_ref(muserd);
-
-		ret = nvkm_memory_kmap(userd->mem, &userd->map);
-		if (ret) {
-			nvkm_chid_put(runl->chid, userd->chid, &chan->cgrp->lock);
-			kfree(userd);
-			return ret;
-		}
-
-
-		list_add(&userd->head, &fifo->userd.list);
-	}
-
-	userd->used |= BIT(chid);
-
-	chan->userd.mem = nvkm_memory_ref(userd->map);
-	chan->userd.base = ouserd;
-
-	return (userd->chid * CHID_PER_USERD) + chid;
-}
-
-static int
-r535_chan_id_get(struct nvkm_chan *chan, struct nvkm_memory *muserd, u64 ouserd)
-{
-	struct nvkm_fifo *fifo = chan->cgrp->runl->fifo;
-	int ret;
-
-	mutex_lock(&fifo->userd.mutex);
-	ret = r535_chan_id_get_locked(chan, muserd, ouserd);
-	mutex_unlock(&fifo->userd.mutex);
-	return ret;
-}
-
 static const struct nvkm_chan_func
 r535_chan = {
-	.id_get = r535_chan_id_get,
-	.id_put = r535_chan_id_put,
 	.inst = &gf100_chan_inst,
 	.userd = &gv100_chan_userd,
 	.ramfc = &r535_chan_ramfc,
@@ -340,10 +236,6 @@ r535_chan = {
 	.doorbell_handle = r535_chan_doorbell_handle,
 };
 
-static const struct nvkm_cgrp_func
-r535_cgrp = {
-};
-
 static int
 r535_engn_nonstall(struct nvkm_engn *engn)
 {
@@ -356,7 +248,7 @@ r535_engn_nonstall(struct nvkm_engn *engn)
 }
 
 static const struct nvkm_engn_func
-r535_ce = {
+r535_engn_ce = {
 	.nonstall = r535_engn_nonstall,
 };
 
@@ -376,7 +268,7 @@ r535_gr_ctor(struct nvkm_engn *engn, struct nvkm_vctx *vctx, struct nvkm_chan *c
 }
 
 static const struct nvkm_engn_func
-r535_gr = {
+r535_engn_gr = {
 	.nonstall = r535_engn_nonstall,
 	.ctor2 = r535_gr_ctor,
 };
@@ -449,57 +341,86 @@ r535_runl = {
 	.allow = r535_runl_allow,
 };
 
-static int
-r535_fifo_2080_type(enum nvkm_subdev_type type, int inst)
+void
+r535_fifo_rc_chid(struct nvkm_fifo *fifo, int chid)
 {
-	switch (type) {
-	case NVKM_ENGINE_GR: return NV2080_ENGINE_TYPE_GR0;
-	case NVKM_ENGINE_CE: return NV2080_ENGINE_TYPE_COPY0 + inst;
-	case NVKM_ENGINE_SEC2: return NV2080_ENGINE_TYPE_SEC2;
-	case NVKM_ENGINE_NVDEC: return NV2080_ENGINE_TYPE_NVDEC0 + inst;
-	case NVKM_ENGINE_NVENC: return NV2080_ENGINE_TYPE_NVENC0 + inst;
-	case NVKM_ENGINE_NVJPG: return NV2080_ENGINE_TYPE_NVJPEG0 + inst;
-	case NVKM_ENGINE_OFA: return NV2080_ENGINE_TYPE_OFA;
-	case NVKM_ENGINE_SW: return NV2080_ENGINE_TYPE_SW;
-	default:
-		break;
+	struct nvkm_chan *chan;
+	unsigned long flags;
+
+	chan = nvkm_chan_get_chid(&fifo->engine, chid, &flags);
+	if (!chan) {
+		nvkm_error(&fifo->engine.subdev, "rc: chid %d not found!\n", chid);
+		return;
 	}
 
-	WARN_ON(1);
-	return -EINVAL;
+	nvkm_chan_error(chan, false);
+	nvkm_chan_put(&chan, flags);
 }
 
 static int
-r535_fifo_engn_type(RM_ENGINE_TYPE rm, enum nvkm_subdev_type *ptype)
+r535_fifo_rc_triggered(void *priv, u32 fn, void *repv, u32 repc)
 {
+	rpc_rc_triggered_v17_02 *msg = repv;
+	struct nvkm_gsp *gsp = priv;
+
+	if (WARN_ON(repc < sizeof(*msg)))
+		return -EINVAL;
+
+	nvkm_error(&gsp->subdev, "rc: engn:%08x chid:%d type:%d scope:%d part:%d\n",
+		   msg->nv2080EngineType, msg->chid, msg->exceptType, msg->scope,
+		   msg->partitionAttributionId);
+
+	r535_fifo_rc_chid(gsp->subdev.device->fifo, msg->chid);
+	return 0;
+}
+
+static int
+r535_fifo_xlat_rm_engine_type(u32 rm, enum nvkm_subdev_type *ptype, int *p2080)
+{
+#define RM_ENGINE_TYPE(RM,NVKM,INST)              \
+	RM_ENGINE_TYPE_##RM:                      \
+		*ptype = NVKM_ENGINE_##NVKM;      \
+		*p2080 = NV2080_ENGINE_TYPE_##RM; \
+		return INST
+
 	switch (rm) {
-	case RM_ENGINE_TYPE_GR0:
-		*ptype = NVKM_ENGINE_GR;
-		return 0;
-	case RM_ENGINE_TYPE_COPY0...RM_ENGINE_TYPE_COPY9:
-		*ptype = NVKM_ENGINE_CE;
-		return rm - RM_ENGINE_TYPE_COPY0;
-	case RM_ENGINE_TYPE_NVDEC0...RM_ENGINE_TYPE_NVDEC7:
-		*ptype = NVKM_ENGINE_NVDEC;
-		return rm - RM_ENGINE_TYPE_NVDEC0;
-	case RM_ENGINE_TYPE_NVENC0...RM_ENGINE_TYPE_NVENC2:
-		*ptype = NVKM_ENGINE_NVENC;
-		return rm - RM_ENGINE_TYPE_NVENC0;
-	case RM_ENGINE_TYPE_SW:
-		*ptype = NVKM_ENGINE_SW;
-		return 0;
-	case RM_ENGINE_TYPE_SEC2:
-		*ptype = NVKM_ENGINE_SEC2;
-		return 0;
-	case RM_ENGINE_TYPE_NVJPEG0...RM_ENGINE_TYPE_NVJPEG7:
-		*ptype = NVKM_ENGINE_NVJPG;
-		return rm - RM_ENGINE_TYPE_NVJPEG0;
-	case RM_ENGINE_TYPE_OFA:
-		*ptype = NVKM_ENGINE_OFA;
-		return 0;
+	case RM_ENGINE_TYPE(    GR0,    GR, 0);
+	case RM_ENGINE_TYPE(  COPY0,    CE, 0);
+	case RM_ENGINE_TYPE(  COPY1,    CE, 1);
+	case RM_ENGINE_TYPE(  COPY2,    CE, 2);
+	case RM_ENGINE_TYPE(  COPY3,    CE, 3);
+	case RM_ENGINE_TYPE(  COPY4,    CE, 4);
+	case RM_ENGINE_TYPE(  COPY5,    CE, 5);
+	case RM_ENGINE_TYPE(  COPY6,    CE, 6);
+	case RM_ENGINE_TYPE(  COPY7,    CE, 7);
+	case RM_ENGINE_TYPE(  COPY8,    CE, 8);
+	case RM_ENGINE_TYPE(  COPY9,    CE, 9);
+	case RM_ENGINE_TYPE( NVDEC0, NVDEC, 0);
+	case RM_ENGINE_TYPE( NVDEC1, NVDEC, 1);
+	case RM_ENGINE_TYPE( NVDEC2, NVDEC, 2);
+	case RM_ENGINE_TYPE( NVDEC3, NVDEC, 3);
+	case RM_ENGINE_TYPE( NVDEC4, NVDEC, 4);
+	case RM_ENGINE_TYPE( NVDEC5, NVDEC, 5);
+	case RM_ENGINE_TYPE( NVDEC6, NVDEC, 6);
+	case RM_ENGINE_TYPE( NVDEC7, NVDEC, 7);
+	case RM_ENGINE_TYPE( NVENC0, NVENC, 0);
+	case RM_ENGINE_TYPE( NVENC1, NVENC, 1);
+	case RM_ENGINE_TYPE( NVENC2, NVENC, 2);
+	case RM_ENGINE_TYPE(NVJPEG0, NVJPG, 0);
+	case RM_ENGINE_TYPE(NVJPEG1, NVJPG, 1);
+	case RM_ENGINE_TYPE(NVJPEG2, NVJPG, 2);
+	case RM_ENGINE_TYPE(NVJPEG3, NVJPG, 3);
+	case RM_ENGINE_TYPE(NVJPEG4, NVJPG, 4);
+	case RM_ENGINE_TYPE(NVJPEG5, NVJPG, 5);
+	case RM_ENGINE_TYPE(NVJPEG6, NVJPG, 6);
+	case RM_ENGINE_TYPE(NVJPEG7, NVJPG, 7);
+	case RM_ENGINE_TYPE(     SW,    SW, 0);
+	case RM_ENGINE_TYPE(   SEC2,  SEC2, 0);
+	case RM_ENGINE_TYPE(    OFA,   OFA, 0);
 	default:
 		return -EINVAL;
 	}
+#undef RM_ENGINE_TYPE
 }
 
 static int
@@ -536,16 +457,19 @@ static int
 r535_fifo_runl_ctor(struct nvkm_fifo *fifo)
 {
 	struct nvkm_subdev *subdev = &fifo->engine.subdev;
-	struct nvkm_gsp *gsp = subdev->device->gsp;
+	struct nvkm_device *device = subdev->device;
+	struct nvkm_gsp *gsp = device->gsp;
+	struct nvkm_rm *rm = gsp->rm;
 	struct nvkm_runl *runl;
 	struct nvkm_engn *engn;
-	u32 cgids = 2048;
 	u32 chids = 2048;
+	u32 first = rm->api->fifo->rsvd_chids;
+	u32 count = chids - first;
 	int ret;
 	NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS *ctrl;
 
-	if ((ret = nvkm_chid_new(&nvkm_chan_event, subdev, cgids, 0, cgids, &fifo->cgid)) ||
-	    (ret = nvkm_chid_new(&nvkm_chan_event, subdev, chids, 0, chids, &fifo->chid)))
+	if ((ret = nvkm_chid_new(&nvkm_chan_event, subdev, chids, first, count, &fifo->cgid)) ||
+	    (ret = nvkm_chid_new(&nvkm_chan_event, subdev, chids, first, count, &fifo->chid)))
 		return ret;
 
 	ctrl = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
@@ -576,25 +500,43 @@ r535_fifo_runl_ctor(struct nvkm_fifo *fifo)
 		if (!runl)
 			continue;
 
-		inst = r535_fifo_engn_type(rmid, &type);
+		inst = rm->api->fifo->xlat_rm_engine_type(rmid, &type, &nv2080);
 		if (inst < 0) {
 			nvkm_warn(subdev, "RM_ENGINE_TYPE 0x%x\n", rmid);
 			nvkm_runl_del(runl);
 			continue;
 		}
 
-		nv2080 = r535_fifo_2080_type(type, inst);
-		if (nv2080 < 0) {
+		/* Skip SW engine - there's currently no support for NV SW classes. */
+		if (type == NVKM_ENGINE_SW)
+			continue;
+
+		/* Skip lone GRCEs (ones not paired with GR on a runlist), as they
+		 * don't appear to function as async copy engines.
+		 */
+		if (type == NVKM_ENGINE_CE &&
+		     rm->gpu->ce.grce_mask &&
+		    (rm->gpu->ce.grce_mask(device) & BIT(inst)) &&
+		    !nvkm_runl_find_engn(engn, runl, engn->engine->subdev.type == NVKM_ENGINE_GR)) {
+			RUNL_DEBUG(runl, "skip LCE %d - GRCE without GR", inst);
+			nvkm_runl_del(runl);
+			continue;
+		}
+
+		ret = nvkm_rm_engine_new(gsp->rm, type, inst);
+		if (ret) {
 			nvkm_runl_del(runl);
 			continue;
 		}
 
+		engn = NULL;
+
 		switch (type) {
 		case NVKM_ENGINE_CE:
-			engn = nvkm_runl_add(runl, nv2080, &r535_ce, type, inst);
+			engn = nvkm_runl_add(runl, nv2080, &r535_engn_ce, type, inst);
 			break;
 		case NVKM_ENGINE_GR:
-			engn = nvkm_runl_add(runl, nv2080, &r535_gr, type, inst);
+			engn = nvkm_runl_add(runl, nv2080, &r535_engn_gr, type, inst);
 			break;
 		case NVKM_ENGINE_NVDEC:
 		case NVKM_ENGINE_NVENC:
@@ -633,7 +575,7 @@ r535_fifo_runl_ctor(struct nvkm_fifo *fifo)
 		nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
 	}
 
-	return r535_fifo_ectx_size(fifo);
+	return rm->api->fifo->ectx_size(fifo);
 }
 
 static void
@@ -646,6 +588,7 @@ int
 r535_fifo_new(const struct nvkm_fifo_func *hw, struct nvkm_device *device,
 	      enum nvkm_subdev_type type, int inst, struct nvkm_fifo **pfifo)
 {
+	const struct nvkm_rm_gpu *gpu = device->gsp->rm->gpu;
 	struct nvkm_fifo_func *rm;
 
 	if (!(rm = kzalloc(sizeof(*rm), GFP_KERNEL)))
@@ -654,12 +597,20 @@ r535_fifo_new(const struct nvkm_fifo_func *hw, struct nvkm_device *device,
 	rm->dtor = r535_fifo_dtor;
 	rm->runl_ctor = r535_fifo_runl_ctor;
 	rm->runl = &r535_runl;
-	rm->cgrp = hw->cgrp;
-	rm->cgrp.func = &r535_cgrp;
-	rm->chan = hw->chan;
+	rm->chan.user.oclass = gpu->fifo.chan.class;
 	rm->chan.func = &r535_chan;
 	rm->nonstall = &ga100_fifo_nonstall;
 	rm->nonstall_ctor = ga100_fifo_nonstall_ctor;
 
 	return nvkm_fifo_new_(rm, device, type, inst, pfifo);
 }
+
+const struct nvkm_rm_api_fifo
+r535_fifo = {
+	.xlat_rm_engine_type = r535_fifo_xlat_rm_engine_type,
+	.ectx_size = r535_fifo_ectx_size,
+	.rc_triggered = r535_fifo_rc_triggered,
+	.chan = {
+		.alloc = r535_chan_alloc,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gr.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gr.c
new file mode 100644
index 000000000000..ddb57d5e73d6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gr.c
@@ -0,0 +1,356 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/gr.h>
+
+#include <core/memory.h>
+#include <subdev/gsp.h>
+#include <subdev/mmu/vmm.h>
+#include <engine/fifo/priv.h>
+#include <engine/gr/priv.h>
+
+#include <nvif/if900d.h>
+
+#include <nvhw/drf.h>
+
+#include "nvrm/gr.h"
+#include "nvrm/vmm.h"
+
+#define r535_gr(p) container_of((p), struct r535_gr, base)
+
+static void *
+r535_gr_chan_dtor(struct nvkm_object *object)
+{
+	struct r535_gr_chan *grc = container_of(object, typeof(*grc), object);
+	struct r535_gr *gr = grc->gr;
+
+	for (int i = 0; i < gr->ctxbuf_nr; i++) {
+		nvkm_vmm_put(grc->vmm, &grc->vma[i]);
+		nvkm_memory_unref(&grc->mem[i]);
+	}
+
+	nvkm_vmm_unref(&grc->vmm);
+	return grc;
+}
+
+static const struct nvkm_object_func
+r535_gr_chan = {
+	.dtor = r535_gr_chan_dtor,
+};
+
+int
+r535_gr_promote_ctx(struct r535_gr *gr, bool golden, struct nvkm_vmm *vmm,
+		    struct nvkm_memory **pmem, struct nvkm_vma **pvma,
+		    struct nvkm_gsp_object *chan)
+{
+	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
+	struct nvkm_device *device = subdev->device;
+	NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&vmm->rm.device.subdevice,
+				    NV2080_CTRL_CMD_GPU_PROMOTE_CTX, sizeof(*ctrl));
+	if (WARN_ON(IS_ERR(ctrl)))
+		return PTR_ERR(ctrl);
+
+	ctrl->engineType = 1;
+	ctrl->hChanClient = vmm->rm.client.object.handle;
+	ctrl->hObject = chan->handle;
+
+	for (int i = 0; i < gr->ctxbuf_nr; i++) {
+		NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY *entry =
+			&ctrl->promoteEntry[ctrl->entryCount];
+		const bool alloc = golden || !gr->ctxbuf[i].global;
+		int ret;
+
+		entry->bufferId = gr->ctxbuf[i].bufferId;
+		entry->bInitialize = gr->ctxbuf[i].init && alloc;
+
+		if (alloc) {
+			ret = nvkm_memory_new(device, gr->ctxbuf[i].init ?
+					      NVKM_MEM_TARGET_INST : NVKM_MEM_TARGET_INST_SR_LOST,
+					      gr->ctxbuf[i].size, 1 << gr->ctxbuf[i].page,
+					      gr->ctxbuf[i].init, &pmem[i]);
+			if (WARN_ON(ret))
+				return ret;
+
+			if (gr->ctxbuf[i].bufferId ==
+					NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PRIV_ACCESS_MAP)
+				entry->bNonmapped = 1;
+		} else {
+			if (gr->ctxbuf[i].bufferId ==
+				NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_UNRESTRICTED_PRIV_ACCESS_MAP)
+				continue;
+
+			pmem[i] = nvkm_memory_ref(gr->ctxbuf_mem[i]);
+		}
+
+		if (!entry->bNonmapped) {
+			struct gf100_vmm_map_v0 args = {
+				.priv = 1,
+				.ro   = gr->ctxbuf[i].ro,
+			};
+
+			mutex_lock(&vmm->mutex.vmm);
+			ret = nvkm_vmm_get_locked(vmm, false, true, false, 0, gr->ctxbuf[i].align,
+						  nvkm_memory_size(pmem[i]), &pvma[i]);
+			mutex_unlock(&vmm->mutex.vmm);
+			if (ret)
+				return ret;
+
+			ret = nvkm_memory_map(pmem[i], 0, vmm, pvma[i], &args, sizeof(args));
+			if (ret)
+				return ret;
+
+			entry->gpuVirtAddr = pvma[i]->addr;
+		}
+
+		if (entry->bInitialize) {
+			entry->gpuPhysAddr = nvkm_memory_addr(pmem[i]);
+			entry->size = gr->ctxbuf[i].size;
+			entry->physAttr = 4;
+		}
+
+		nvkm_debug(subdev,
+			   "promote %02d: pa %016llx/%08x sz %016llx va %016llx init:%d nm:%d\n",
+			   entry->bufferId, entry->gpuPhysAddr, entry->physAttr, entry->size,
+			   entry->gpuVirtAddr, entry->bInitialize, entry->bNonmapped);
+
+		ctrl->entryCount++;
+	}
+
+	return nvkm_gsp_rm_ctrl_wr(&vmm->rm.device.subdevice, ctrl);
+}
+
+int
+r535_gr_chan_new(struct nvkm_gr *base, struct nvkm_chan *chan, const struct nvkm_oclass *oclass,
+		 struct nvkm_object **pobject)
+{
+	struct r535_gr *gr = r535_gr(base);
+	struct r535_gr_chan *grc;
+	int ret;
+
+	if (!(grc = kzalloc(sizeof(*grc), GFP_KERNEL)))
+		return -ENOMEM;
+
+	nvkm_object_ctor(&r535_gr_chan, oclass, &grc->object);
+	grc->gr = gr;
+	grc->vmm = nvkm_vmm_ref(chan->vmm);
+	grc->chan = chan;
+	*pobject = &grc->object;
+
+	ret = r535_gr_promote_ctx(gr, false, grc->vmm, grc->mem, grc->vma, &chan->rm.object);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+u64
+r535_gr_units(struct nvkm_gr *gr)
+{
+	struct nvkm_gsp *gsp = gr->engine.subdev.device->gsp;
+
+	return (gsp->gr.tpcs << 8) | gsp->gr.gpcs;
+}
+
+void
+r535_gr_get_ctxbuf_info(struct r535_gr *gr, int i,
+			struct NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO *info)
+{
+	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
+	static const struct {
+		u32     id0; /* NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID */
+		u32     id1; /* NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID */
+		bool global;
+		bool   init;
+		bool     ro;
+	} map[] = {
+#define _A(n,N,G,I,R) { .id0 = NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_##n, \
+		.id1 = NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_##N, \
+		.global = (G), .init = (I), .ro = (R) }
+#define _B(N,G,I,R) _A(GRAPHICS_##N, N, (G), (I), (R))
+		/*                                       global   init     ro */
+		_A(           GRAPHICS,             MAIN, false,  true, false),
+		_B(                                PATCH, false,  true, false),
+		_A( GRAPHICS_BUNDLE_CB, BUFFER_BUNDLE_CB,  true, false, false),
+		_B(                             PAGEPOOL,  true, false, false),
+		_B(                         ATTRIBUTE_CB,  true, false, false),
+		_B(                        RTV_CB_GLOBAL,  true, false, false),
+		_B(                           FECS_EVENT,  true,  true, false),
+		_B(                      PRIV_ACCESS_MAP,  true,  true,  true),
+#undef _B
+#undef _A
+	};
+	u32 size = info->size;
+	u8 align, page;
+	int id;
+
+	for (id = 0; id < ARRAY_SIZE(map); id++) {
+		if (map[id].id0 == i)
+			break;
+	}
+
+	nvkm_debug(subdev, "%02x: size:0x%08x %s\n", i,
+		   size, (id < ARRAY_SIZE(map)) ? "*" : "");
+	if (id >= ARRAY_SIZE(map))
+		return;
+
+	if (map[id].id1 == NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_MAIN)
+		size = ALIGN(size, 0x1000) + 64 * 0x1000; /* per-subctx headers */
+
+	if      (size >= 1 << 21) page = 21;
+	else if (size >= 1 << 16) page = 16;
+	else			  page = 12;
+
+	if (map[id].id1 == NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_ATTRIBUTE_CB)
+		align = order_base_2(size);
+	else
+		align = page;
+
+	if (WARN_ON(gr->ctxbuf_nr == ARRAY_SIZE(gr->ctxbuf)))
+		return;
+
+	gr->ctxbuf[gr->ctxbuf_nr].bufferId = map[id].id1;
+	gr->ctxbuf[gr->ctxbuf_nr].size     = size;
+	gr->ctxbuf[gr->ctxbuf_nr].page     = page;
+	gr->ctxbuf[gr->ctxbuf_nr].align    = align;
+	gr->ctxbuf[gr->ctxbuf_nr].global   = map[id].global;
+	gr->ctxbuf[gr->ctxbuf_nr].init     = map[id].init;
+	gr->ctxbuf[gr->ctxbuf_nr].ro       = map[id].ro;
+	gr->ctxbuf_nr++;
+
+	if (map[id].id1 == NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PRIV_ACCESS_MAP) {
+		if (WARN_ON(gr->ctxbuf_nr == ARRAY_SIZE(gr->ctxbuf)))
+			return;
+
+		gr->ctxbuf[gr->ctxbuf_nr] = gr->ctxbuf[gr->ctxbuf_nr - 1];
+		gr->ctxbuf[gr->ctxbuf_nr].bufferId =
+			NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_UNRESTRICTED_PRIV_ACCESS_MAP;
+		gr->ctxbuf_nr++;
+	}
+}
+
+static int
+r535_gr_get_ctxbufs_info(struct r535_gr *gr)
+{
+	NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS *info;
+	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
+	struct nvkm_gsp *gsp = subdev->device->gsp;
+
+	info = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
+				   NV2080_CTRL_CMD_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO,
+				   sizeof(*info));
+	if (WARN_ON(IS_ERR(info)))
+		return PTR_ERR(info);
+
+	for (int i = 0; i < ARRAY_SIZE(info->engineContextBuffersInfo[0].engine); i++)
+		r535_gr_get_ctxbuf_info(gr, i, &info->engineContextBuffersInfo[0].engine[i]);
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, info);
+	return 0;
+}
+
+int
+r535_gr_oneinit(struct nvkm_gr *base)
+{
+	struct r535_gr *gr = container_of(base, typeof(*gr), base);
+	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
+	struct nvkm_device *device = subdev->device;
+	struct nvkm_gsp *gsp = device->gsp;
+	struct nvkm_rm *rm = gsp->rm;
+	struct {
+		struct nvkm_memory *inst;
+		struct nvkm_vmm *vmm;
+		struct nvkm_gsp_object chan;
+		struct nvkm_vma *vma[R515_GR_MAX_CTXBUFS];
+	} golden = {};
+	struct nvkm_gsp_object threed;
+	int ret;
+
+	/* Allocate a channel to use for golden context init. */
+	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, 0x12000, 0, true, &golden.inst);
+	if (ret)
+		goto done;
+
+	ret = nvkm_vmm_new(device, 0x1000, 0, NULL, 0, NULL, "grGoldenVmm", &golden.vmm);
+	if (ret)
+		goto done;
+
+	ret = r535_mmu_vaspace_new(golden.vmm, NVKM_RM_VASPACE, false);
+	if (ret)
+		goto done;
+
+	ret = rm->api->fifo->chan.alloc(&golden.vmm->rm.device, NVKM_RM_CHAN(0),
+					1, 0, true, rm->api->fifo->rsvd_chids,
+					nvkm_memory_addr(golden.inst),
+					nvkm_memory_addr(golden.inst) + 0x1000,
+					nvkm_memory_addr(golden.inst) + 0x2000,
+					golden.vmm, 0, 0x1000, &golden.chan);
+	if (ret)
+		goto done;
+
+	/* Fetch context buffer info from RM and allocate each of them here to use
+	 * during golden context init (or later as a global context buffer).
+	 *
+	 * Also build the information that'll be used to create channel contexts.
+	 */
+	ret = rm->api->gr->get_ctxbufs_info(gr);
+	if (ret)
+		goto done;
+
+	/* Promote golden context to RM. */
+	ret = r535_gr_promote_ctx(gr, true, golden.vmm, gr->ctxbuf_mem, golden.vma, &golden.chan);
+	if (ret)
+		goto done;
+
+	/* Allocate 3D class on channel to trigger golden context init in RM. */
+	ret = nvkm_gsp_rm_alloc(&golden.chan, NVKM_RM_THREED, rm->gpu->gr.class.threed, 0, &threed);
+	if (ret)
+		goto done;
+
+	/* There's no need to keep the golden channel around, as RM caches the context. */
+	nvkm_gsp_rm_free(&threed);
+done:
+	nvkm_gsp_rm_free(&golden.chan);
+	for (int i = gr->ctxbuf_nr - 1; i >= 0; i--)
+		nvkm_vmm_put(golden.vmm, &golden.vma[i]);
+	nvkm_vmm_unref(&golden.vmm);
+	nvkm_memory_unref(&golden.inst);
+	return ret;
+
+}
+
+void *
+r535_gr_dtor(struct nvkm_gr *base)
+{
+	struct r535_gr *gr = r535_gr(base);
+
+	while (gr->ctxbuf_nr)
+		nvkm_memory_unref(&gr->ctxbuf_mem[--gr->ctxbuf_nr]);
+
+	kfree(gr->base.func);
+	return gr;
+}
+
+const struct nvkm_rm_api_gr
+r535_gr = {
+	.get_ctxbufs_info = r535_gr_get_ctxbufs_info,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
index db2602e88006..baf42339f93e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
@@ -19,9 +19,12 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
+#include <rm/rpc.h>
+
 #include "priv.h"
 
 #include <core/pci.h>
+#include <subdev/pci/priv.h>
 #include <subdev/timer.h>
 #include <subdev/vfn.h>
 #include <engine/fifo/chan.h>
@@ -30,29 +33,11 @@
 
 #include <nvfw/fw.h>
 
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0000.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0005.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl0080.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl2080.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080event.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080gpu.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl2080/ctrl2080internal.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
-#include <nvrm/535.113.01/common/shared/msgq/inc/msgq/msgq_priv.h>
-#include <nvrm/535.113.01/common/uproc/os/common/include/libos_init_args.h>
-#include <nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_sr_meta.h>
-#include <nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/gsp/gsp_fw_wpr_meta.h>
-#include <nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmRiscvUcode.h>
-#include <nvrm/535.113.01/nvidia/arch/nvalloc/common/inc/rmgspseq.h>
-#include <nvrm/535.113.01/nvidia/generated/g_allclasses.h>
-#include <nvrm/535.113.01/nvidia/generated/g_os_nvoc.h>
-#include <nvrm/535.113.01/nvidia/generated/g_rpc-structures.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_fw_heap.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_init_args.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/gsp/gsp_static_config.h>
-#include <nvrm/535.113.01/nvidia/inc/kernel/gpu/intr/engine_idx.h>
-#include <nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_global_enums.h>
+#include "nvrm/gsp.h"
+#include "nvrm/rpcfn.h"
+#include "nvrm/msgfn.h"
+#include "nvrm/event.h"
+#include "nvrm/fifo.h"
 
 #include <linux/acpi.h>
 #include <linux/ctype.h>
@@ -60,990 +45,6 @@
 
 extern struct dentry *nouveau_debugfs_root;
 
-#define GSP_MSG_MIN_SIZE GSP_PAGE_SIZE
-#define GSP_MSG_MAX_SIZE (GSP_MSG_MIN_SIZE * 16)
-
-/**
- * DOC: GSP message queue element
- *
- * https://github.com/NVIDIA/open-gpu-kernel-modules/blob/535/src/nvidia/inc/kernel/gpu/gsp/message_queue_priv.h
- *
- * The GSP command queue and status queue are message queues for the
- * communication between software and GSP. The software submits the GSP
- * RPC via the GSP command queue, GSP writes the status of the submitted
- * RPC in the status queue.
- *
- * A GSP message queue element consists of three parts:
- *
- * - message element header (struct r535_gsp_msg), which mostly maintains
- *   the metadata for queuing the element.
- *
- * - RPC message header (struct nvfw_gsp_rpc), which maintains the info
- *   of the RPC. E.g., the RPC function number.
- *
- * - The payload, where the RPC message stays. E.g. the params of a
- *   specific RPC function. Some RPC functions also have their headers
- *   in the payload. E.g. rm_alloc, rm_control.
- *
- * The memory layout of a GSP message element can be illustrated below::
- *
- *    +------------------------+
- *    | Message Element Header |
- *    |    (r535_gsp_msg)      |
- *    |                        |
- *    | (r535_gsp_msg.data)    |
- *    |          |             |
- *    |----------V-------------|
- *    |    GSP RPC Header      |
- *    |    (nvfw_gsp_rpc)      |
- *    |                        |
- *    | (nvfw_gsp_rpc.data)    |
- *    |          |             |
- *    |----------V-------------|
- *    |       Payload          |
- *    |                        |
- *    |   header(optional)     |
- *    |        params          |
- *    +------------------------+
- *
- * The max size of a message queue element is 16 pages (including the
- * headers). When a GSP message to be sent is larger than 16 pages, the
- * message should be split into multiple elements and sent accordingly.
- *
- * In the bunch of the split elements, the first element has the expected
- * function number, while the rest of the elements are sent with the
- * function number NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD.
- *
- * GSP consumes the elements from the cmdq and always writes the result
- * back to the msgq. The result is also formed as split elements.
- *
- * Terminology:
- *
- * - gsp_msg(msg): GSP message element (element header + GSP RPC header +
- *   payload)
- * - gsp_rpc(rpc): GSP RPC (RPC header + payload)
- * - gsp_rpc_buf: buffer for (GSP RPC header + payload)
- * - gsp_rpc_len: size of (GSP RPC header + payload)
- * - params_size: size of params in the payload
- * - payload_size: size of (header if exists + params) in the payload
- */
-
-struct r535_gsp_msg {
-	u8 auth_tag_buffer[16];
-	u8 aad_buffer[16];
-	u32 checksum;
-	u32 sequence;
-	u32 elem_count;
-	u32 pad;
-	u8  data[];
-};
-
-struct nvfw_gsp_rpc {
-	u32 header_version;
-	u32 signature;
-	u32 length;
-	u32 function;
-	u32 rpc_result;
-	u32 rpc_result_private;
-	u32 sequence;
-	union {
-		u32 spare;
-		u32 cpuRmGfid;
-	};
-	u8  data[];
-};
-
-#define GSP_MSG_HDR_SIZE offsetof(struct r535_gsp_msg, data)
-
-#define to_gsp_hdr(p, header) \
-	container_of((void *)p, typeof(*header), data)
-
-#define to_payload_hdr(p, header) \
-	container_of((void *)p, typeof(*header), params)
-
-static int
-r535_rpc_status_to_errno(uint32_t rpc_status)
-{
-	switch (rpc_status) {
-	case 0x55: /* NV_ERR_NOT_READY */
-	case 0x66: /* NV_ERR_TIMEOUT_RETRY */
-		return -EBUSY;
-	case 0x51: /* NV_ERR_NO_MEMORY */
-		return -ENOMEM;
-	default:
-		return -EINVAL;
-	}
-}
-
-static int
-r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime)
-{
-	u32 size, rptr = *gsp->msgq.rptr;
-	int used;
-
-	size = DIV_ROUND_UP(GSP_MSG_HDR_SIZE + gsp_rpc_len,
-			    GSP_PAGE_SIZE);
-	if (WARN_ON(!size || size >= gsp->msgq.cnt))
-		return -EINVAL;
-
-	do {
-		u32 wptr = *gsp->msgq.wptr;
-
-		used = wptr + gsp->msgq.cnt - rptr;
-		if (used >= gsp->msgq.cnt)
-			used -= gsp->msgq.cnt;
-		if (used >= size)
-			break;
-
-		usleep_range(1, 2);
-	} while (--(*ptime));
-
-	if (WARN_ON(!*ptime))
-		return -ETIMEDOUT;
-
-	return used;
-}
-
-static struct r535_gsp_msg *
-r535_gsp_msgq_get_entry(struct nvkm_gsp *gsp)
-{
-	u32 rptr = *gsp->msgq.rptr;
-
-	/* Skip the first page, which is the message queue info */
-	return (void *)((u8 *)gsp->shm.msgq.ptr + GSP_PAGE_SIZE +
-	       rptr * GSP_PAGE_SIZE);
-}
-
-/**
- * DOC: Receive a GSP message queue element
- *
- * Receiving a GSP message queue element from the message queue consists of
- * the following steps:
- *
- * - Peek the element from the queue: r535_gsp_msgq_peek().
- *   Peek the first page of the element to determine the total size of the
- *   message before allocating the proper memory.
- *
- * - Allocate memory for the message.
- *   Once the total size of the message is determined from the GSP message
- *   queue element, the caller of r535_gsp_msgq_recv() allocates the
- *   required memory.
- *
- * - Receive the message: r535_gsp_msgq_recv().
- *   Copy the message into the allocated memory. Advance the read pointer.
- *   If the message is a large GSP message, r535_gsp_msgq_recv() calls
- *   r535_gsp_msgq_recv_one_elem() repeatedly to receive continuation parts
- *   until the complete message is received.
- *   r535_gsp_msgq_recv() assembles the payloads of cotinuation parts into
- *   the return of the large GSP message.
- *
- * - Free the allocated memory: r535_gsp_msg_done().
- *   The user is responsible for freeing the memory allocated for the GSP
- *   message pages after they have been processed.
- */
-static void *
-r535_gsp_msgq_peek(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries)
-{
-	struct r535_gsp_msg *mqe;
-	int ret;
-
-	ret = r535_gsp_msgq_wait(gsp, gsp_rpc_len, retries);
-	if (ret < 0)
-		return ERR_PTR(ret);
-
-	mqe = r535_gsp_msgq_get_entry(gsp);
-
-	return mqe->data;
-}
-
-struct r535_gsp_msg_info {
-	int *retries;
-	u32 gsp_rpc_len;
-	void *gsp_rpc_buf;
-	bool continuation;
-};
-
-static void
-r535_gsp_msg_dump(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg, int lvl);
-
-static void *
-r535_gsp_msgq_recv_one_elem(struct nvkm_gsp *gsp,
-			    struct r535_gsp_msg_info *info)
-{
-	u8 *buf = info->gsp_rpc_buf;
-	u32 rptr = *gsp->msgq.rptr;
-	struct r535_gsp_msg *mqe;
-	u32 size, expected, len;
-	int ret;
-
-	expected = info->gsp_rpc_len;
-
-	ret = r535_gsp_msgq_wait(gsp, expected, info->retries);
-	if (ret < 0)
-		return ERR_PTR(ret);
-
-	mqe = r535_gsp_msgq_get_entry(gsp);
-
-	if (info->continuation) {
-		struct nvfw_gsp_rpc *rpc = (struct nvfw_gsp_rpc *)mqe->data;
-
-		if (rpc->function != NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD) {
-			nvkm_error(&gsp->subdev,
-				   "Not a continuation of a large RPC\n");
-			r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR);
-			return ERR_PTR(-EIO);
-		}
-	}
-
-	size = ALIGN(expected + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE);
-
-	len = ((gsp->msgq.cnt - rptr) * GSP_PAGE_SIZE) - sizeof(*mqe);
-	len = min_t(u32, expected, len);
-
-	if (info->continuation)
-		memcpy(buf, mqe->data + sizeof(struct nvfw_gsp_rpc),
-		       len - sizeof(struct nvfw_gsp_rpc));
-	else
-		memcpy(buf, mqe->data, len);
-
-	expected -= len;
-
-	if (expected) {
-		mqe = (void *)((u8 *)gsp->shm.msgq.ptr + 0x1000 + 0 * 0x1000);
-		memcpy(buf + len, mqe, expected);
-	}
-
-	rptr = (rptr + DIV_ROUND_UP(size, GSP_PAGE_SIZE)) % gsp->msgq.cnt;
-
-	mb();
-	(*gsp->msgq.rptr) = rptr;
-	return buf;
-}
-
-static void *
-r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries)
-{
-	struct r535_gsp_msg *mqe;
-	const u32 max_rpc_size = GSP_MSG_MAX_SIZE - sizeof(*mqe);
-	struct nvfw_gsp_rpc *rpc;
-	struct r535_gsp_msg_info info = {0};
-	u32 expected = gsp_rpc_len;
-	void *buf;
-
-	mqe = r535_gsp_msgq_get_entry(gsp);
-	rpc = (struct nvfw_gsp_rpc *)mqe->data;
-
-	if (WARN_ON(rpc->length > max_rpc_size))
-		return NULL;
-
-	buf = kvmalloc(max_t(u32, rpc->length, expected), GFP_KERNEL);
-	if (!buf)
-		return ERR_PTR(-ENOMEM);
-
-	info.gsp_rpc_buf = buf;
-	info.retries = retries;
-	info.gsp_rpc_len = rpc->length;
-
-	buf = r535_gsp_msgq_recv_one_elem(gsp, &info);
-	if (IS_ERR(buf)) {
-		kvfree(info.gsp_rpc_buf);
-		info.gsp_rpc_buf = NULL;
-		return buf;
-	}
-
-	if (expected <= max_rpc_size)
-		return buf;
-
-	info.gsp_rpc_buf += info.gsp_rpc_len;
-	expected -= info.gsp_rpc_len;
-
-	while (expected) {
-		u32 size;
-
-		rpc = r535_gsp_msgq_peek(gsp, sizeof(*rpc), info.retries);
-		if (IS_ERR_OR_NULL(rpc)) {
-			kfree(buf);
-			return rpc;
-		}
-
-		info.gsp_rpc_len = rpc->length;
-		info.continuation = true;
-
-		rpc = r535_gsp_msgq_recv_one_elem(gsp, &info);
-		if (IS_ERR_OR_NULL(rpc)) {
-			kfree(buf);
-			return rpc;
-		}
-
-		size = info.gsp_rpc_len - sizeof(*rpc);
-		expected -= size;
-		info.gsp_rpc_buf += size;
-	}
-
-	rpc = buf;
-	rpc->length = gsp_rpc_len;
-	return buf;
-}
-
-static int
-r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc)
-{
-	struct r535_gsp_msg *msg = to_gsp_hdr(rpc, msg);
-	struct r535_gsp_msg *cqe;
-	u32 gsp_rpc_len = msg->checksum;
-	u64 *ptr = (void *)msg;
-	u64 *end;
-	u64 csum = 0;
-	int free, time = 1000000;
-	u32 wptr, size, step, len;
-	u32 off = 0;
-
-	len = ALIGN(GSP_MSG_HDR_SIZE + gsp_rpc_len, GSP_PAGE_SIZE);
-
-	end = (u64 *)((char *)ptr + len);
-	msg->pad = 0;
-	msg->checksum = 0;
-	msg->sequence = gsp->cmdq.seq++;
-	msg->elem_count = DIV_ROUND_UP(len, 0x1000);
-
-	while (ptr < end)
-		csum ^= *ptr++;
-
-	msg->checksum = upper_32_bits(csum) ^ lower_32_bits(csum);
-
-	wptr = *gsp->cmdq.wptr;
-	do {
-		do {
-			free = *gsp->cmdq.rptr + gsp->cmdq.cnt - wptr - 1;
-			if (free >= gsp->cmdq.cnt)
-				free -= gsp->cmdq.cnt;
-			if (free >= 1)
-				break;
-
-			usleep_range(1, 2);
-		} while(--time);
-
-		if (WARN_ON(!time)) {
-			kvfree(msg);
-			return -ETIMEDOUT;
-		}
-
-		cqe = (void *)((u8 *)gsp->shm.cmdq.ptr + 0x1000 + wptr * 0x1000);
-		step = min_t(u32, free, (gsp->cmdq.cnt - wptr));
-		size = min_t(u32, len, step * GSP_PAGE_SIZE);
-
-		memcpy(cqe, (u8 *)msg + off, size);
-
-		wptr += DIV_ROUND_UP(size, 0x1000);
-		if (wptr == gsp->cmdq.cnt)
-			wptr = 0;
-
-		off  += size;
-		len -= size;
-	} while (len);
-
-	nvkm_trace(&gsp->subdev, "cmdq: wptr %d\n", wptr);
-	wmb();
-	(*gsp->cmdq.wptr) = wptr;
-	mb();
-
-	nvkm_falcon_wr32(&gsp->falcon, 0xc00, 0x00000000);
-
-	kvfree(msg);
-	return 0;
-}
-
-static void *
-r535_gsp_cmdq_get(struct nvkm_gsp *gsp, u32 gsp_rpc_len)
-{
-	struct r535_gsp_msg *msg;
-	u32 size = GSP_MSG_HDR_SIZE + gsp_rpc_len;
-
-	size = ALIGN(size, GSP_MSG_MIN_SIZE);
-	msg = kvzalloc(size, GFP_KERNEL);
-	if (!msg)
-		return ERR_PTR(-ENOMEM);
-
-	msg->checksum = gsp_rpc_len;
-	return msg->data;
-}
-
-static void
-r535_gsp_msg_done(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg)
-{
-	kvfree(msg);
-}
-
-static void
-r535_gsp_msg_dump(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg, int lvl)
-{
-	if (gsp->subdev.debug >= lvl) {
-		nvkm_printk__(&gsp->subdev, lvl, info,
-			      "msg fn:%d len:0x%x/0x%zx res:0x%x resp:0x%x\n",
-			      msg->function, msg->length, msg->length - sizeof(*msg),
-			      msg->rpc_result, msg->rpc_result_private);
-		print_hex_dump(KERN_INFO, "msg: ", DUMP_PREFIX_OFFSET, 16, 1,
-			       msg->data, msg->length - sizeof(*msg), true);
-	}
-}
-
-static struct nvfw_gsp_rpc *
-r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len)
-{
-	struct nvkm_subdev *subdev = &gsp->subdev;
-	struct nvfw_gsp_rpc *rpc;
-	int retries = 4000000, i;
-
-retry:
-	rpc = r535_gsp_msgq_peek(gsp, sizeof(*rpc), &retries);
-	if (IS_ERR_OR_NULL(rpc))
-		return rpc;
-
-	rpc = r535_gsp_msgq_recv(gsp, gsp_rpc_len, &retries);
-	if (IS_ERR_OR_NULL(rpc))
-		return rpc;
-
-	if (rpc->rpc_result) {
-		r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR);
-		r535_gsp_msg_done(gsp, rpc);
-		return ERR_PTR(-EINVAL);
-	}
-
-	r535_gsp_msg_dump(gsp, rpc, NV_DBG_TRACE);
-
-	if (fn && rpc->function == fn) {
-		if (gsp_rpc_len) {
-			if (rpc->length < gsp_rpc_len) {
-				nvkm_error(subdev, "rpc len %d < %d\n",
-					   rpc->length, gsp_rpc_len);
-				r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR);
-				r535_gsp_msg_done(gsp, rpc);
-				return ERR_PTR(-EIO);
-			}
-
-			return rpc;
-		}
-
-		r535_gsp_msg_done(gsp, rpc);
-		return NULL;
-	}
-
-	for (i = 0; i < gsp->msgq.ntfy_nr; i++) {
-		struct nvkm_gsp_msgq_ntfy *ntfy = &gsp->msgq.ntfy[i];
-
-		if (ntfy->fn == rpc->function) {
-			if (ntfy->func)
-				ntfy->func(ntfy->priv, ntfy->fn, rpc->data,
-					   rpc->length - sizeof(*rpc));
-			break;
-		}
-	}
-
-	if (i == gsp->msgq.ntfy_nr)
-		r535_gsp_msg_dump(gsp, rpc, NV_DBG_WARN);
-
-	r535_gsp_msg_done(gsp, rpc);
-	if (fn)
-		goto retry;
-
-	if (*gsp->msgq.rptr != *gsp->msgq.wptr)
-		goto retry;
-
-	return NULL;
-}
-
-static int
-r535_gsp_msg_ntfy_add(struct nvkm_gsp *gsp, u32 fn, nvkm_gsp_msg_ntfy_func func, void *priv)
-{
-	int ret = 0;
-
-	mutex_lock(&gsp->msgq.mutex);
-	if (WARN_ON(gsp->msgq.ntfy_nr >= ARRAY_SIZE(gsp->msgq.ntfy))) {
-		ret = -ENOSPC;
-	} else {
-		gsp->msgq.ntfy[gsp->msgq.ntfy_nr].fn = fn;
-		gsp->msgq.ntfy[gsp->msgq.ntfy_nr].func = func;
-		gsp->msgq.ntfy[gsp->msgq.ntfy_nr].priv = priv;
-		gsp->msgq.ntfy_nr++;
-	}
-	mutex_unlock(&gsp->msgq.mutex);
-	return ret;
-}
-
-static int
-r535_gsp_rpc_poll(struct nvkm_gsp *gsp, u32 fn)
-{
-	void *repv;
-
-	mutex_lock(&gsp->cmdq.mutex);
-	repv = r535_gsp_msg_recv(gsp, fn, 0);
-	mutex_unlock(&gsp->cmdq.mutex);
-	if (IS_ERR(repv))
-		return PTR_ERR(repv);
-
-	return 0;
-}
-
-static void *
-r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *payload, bool wait,
-		  u32 gsp_rpc_len)
-{
-	struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc);
-	struct nvfw_gsp_rpc *msg;
-	u32 fn = rpc->function;
-	void *repv = NULL;
-	int ret;
-
-	if (gsp->subdev.debug >= NV_DBG_TRACE) {
-		nvkm_trace(&gsp->subdev, "rpc fn:%d len:0x%x/0x%zx\n", rpc->function,
-			   rpc->length, rpc->length - sizeof(*rpc));
-		print_hex_dump(KERN_INFO, "rpc: ", DUMP_PREFIX_OFFSET, 16, 1,
-			       rpc->data, rpc->length - sizeof(*rpc), true);
-	}
-
-	ret = r535_gsp_cmdq_push(gsp, rpc);
-	if (ret)
-		return ERR_PTR(ret);
-
-	if (wait) {
-		msg = r535_gsp_msg_recv(gsp, fn, gsp_rpc_len);
-		if (!IS_ERR_OR_NULL(msg))
-			repv = msg->data;
-		else
-			repv = msg;
-	}
-
-	return repv;
-}
-
-static void
-r535_gsp_event_dtor(struct nvkm_gsp_event *event)
-{
-	struct nvkm_gsp_device *device = event->device;
-	struct nvkm_gsp_client *client = device->object.client;
-	struct nvkm_gsp *gsp = client->gsp;
-
-	mutex_lock(&gsp->client_id.mutex);
-	if (event->func) {
-		list_del(&event->head);
-		event->func = NULL;
-	}
-	mutex_unlock(&gsp->client_id.mutex);
-
-	nvkm_gsp_rm_free(&event->object);
-	event->device = NULL;
-}
-
-static int
-r535_gsp_device_event_get(struct nvkm_gsp_event *event)
-{
-	struct nvkm_gsp_device *device = event->device;
-	NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS *ctrl;
-
-	ctrl = nvkm_gsp_rm_ctrl_get(&device->subdevice,
-				    NV2080_CTRL_CMD_EVENT_SET_NOTIFICATION, sizeof(*ctrl));
-	if (IS_ERR(ctrl))
-		return PTR_ERR(ctrl);
-
-	ctrl->event = event->id;
-	ctrl->action = NV2080_CTRL_EVENT_SET_NOTIFICATION_ACTION_REPEAT;
-	return nvkm_gsp_rm_ctrl_wr(&device->subdevice, ctrl);
-}
-
-static int
-r535_gsp_device_event_ctor(struct nvkm_gsp_device *device, u32 handle, u32 id,
-			   nvkm_gsp_event_func func, struct nvkm_gsp_event *event)
-{
-	struct nvkm_gsp_client *client = device->object.client;
-	struct nvkm_gsp *gsp = client->gsp;
-	NV0005_ALLOC_PARAMETERS *args;
-	int ret;
-
-	args = nvkm_gsp_rm_alloc_get(&device->subdevice, handle,
-				     NV01_EVENT_KERNEL_CALLBACK_EX, sizeof(*args),
-				     &event->object);
-	if (IS_ERR(args))
-		return PTR_ERR(args);
-
-	args->hParentClient = client->object.handle;
-	args->hSrcResource = 0;
-	args->hClass = NV01_EVENT_KERNEL_CALLBACK_EX;
-	args->notifyIndex = NV01_EVENT_CLIENT_RM | id;
-	args->data = NULL;
-
-	ret = nvkm_gsp_rm_alloc_wr(&event->object, args);
-	if (ret)
-		return ret;
-
-	event->device = device;
-	event->id = id;
-
-	ret = r535_gsp_device_event_get(event);
-	if (ret) {
-		nvkm_gsp_event_dtor(event);
-		return ret;
-	}
-
-	mutex_lock(&gsp->client_id.mutex);
-	event->func = func;
-	list_add(&event->head, &client->events);
-	mutex_unlock(&gsp->client_id.mutex);
-	return 0;
-}
-
-static void
-r535_gsp_device_dtor(struct nvkm_gsp_device *device)
-{
-	nvkm_gsp_rm_free(&device->subdevice);
-	nvkm_gsp_rm_free(&device->object);
-}
-
-static int
-r535_gsp_subdevice_ctor(struct nvkm_gsp_device *device)
-{
-	NV2080_ALLOC_PARAMETERS *args;
-
-	return nvkm_gsp_rm_alloc(&device->object, 0x5d1d0000, NV20_SUBDEVICE_0, sizeof(*args),
-				 &device->subdevice);
-}
-
-static int
-r535_gsp_device_ctor(struct nvkm_gsp_client *client, struct nvkm_gsp_device *device)
-{
-	NV0080_ALLOC_PARAMETERS *args;
-	int ret;
-
-	args = nvkm_gsp_rm_alloc_get(&client->object, 0xde1d0000, NV01_DEVICE_0, sizeof(*args),
-				     &device->object);
-	if (IS_ERR(args))
-		return PTR_ERR(args);
-
-	args->hClientShare = client->object.handle;
-
-	ret = nvkm_gsp_rm_alloc_wr(&device->object, args);
-	if (ret)
-		return ret;
-
-	ret = r535_gsp_subdevice_ctor(device);
-	if (ret)
-		nvkm_gsp_rm_free(&device->object);
-
-	return ret;
-}
-
-static void
-r535_gsp_client_dtor(struct nvkm_gsp_client *client)
-{
-	struct nvkm_gsp *gsp = client->gsp;
-
-	nvkm_gsp_rm_free(&client->object);
-
-	mutex_lock(&gsp->client_id.mutex);
-	idr_remove(&gsp->client_id.idr, client->object.handle & 0xffff);
-	mutex_unlock(&gsp->client_id.mutex);
-
-	client->gsp = NULL;
-}
-
-static int
-r535_gsp_client_ctor(struct nvkm_gsp *gsp, struct nvkm_gsp_client *client)
-{
-	NV0000_ALLOC_PARAMETERS *args;
-	int ret;
-
-	mutex_lock(&gsp->client_id.mutex);
-	ret = idr_alloc(&gsp->client_id.idr, client, 0, 0xffff + 1, GFP_KERNEL);
-	mutex_unlock(&gsp->client_id.mutex);
-	if (ret < 0)
-		return ret;
-
-	client->gsp = gsp;
-	client->object.client = client;
-	INIT_LIST_HEAD(&client->events);
-
-	args = nvkm_gsp_rm_alloc_get(&client->object, 0xc1d00000 | ret, NV01_ROOT, sizeof(*args),
-				     &client->object);
-	if (IS_ERR(args)) {
-		r535_gsp_client_dtor(client);
-		return ret;
-	}
-
-	args->hClient = client->object.handle;
-	args->processID = ~0;
-
-	ret = nvkm_gsp_rm_alloc_wr(&client->object, args);
-	if (ret) {
-		r535_gsp_client_dtor(client);
-		return ret;
-	}
-
-	return 0;
-}
-
-static int
-r535_gsp_rpc_rm_free(struct nvkm_gsp_object *object)
-{
-	struct nvkm_gsp_client *client = object->client;
-	struct nvkm_gsp *gsp = client->gsp;
-	rpc_free_v03_00 *rpc;
-
-	nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x free\n",
-		   client->object.handle, object->handle);
-
-	rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_FREE, sizeof(*rpc));
-	if (WARN_ON(IS_ERR_OR_NULL(rpc)))
-		return -EIO;
-
-	rpc->params.hRoot = client->object.handle;
-	rpc->params.hObjectParent = 0;
-	rpc->params.hObjectOld = object->handle;
-	return nvkm_gsp_rpc_wr(gsp, rpc, true);
-}
-
-static void
-r535_gsp_rpc_rm_alloc_done(struct nvkm_gsp_object *object, void *params)
-{
-	rpc_gsp_rm_alloc_v03_00 *rpc = to_payload_hdr(params, rpc);
-
-	nvkm_gsp_rpc_done(object->client->gsp, rpc);
-}
-
-static void *
-r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *params)
-{
-	rpc_gsp_rm_alloc_v03_00 *rpc = to_payload_hdr(params, rpc);
-	struct nvkm_gsp *gsp = object->client->gsp;
-	void *ret = NULL;
-
-	rpc = nvkm_gsp_rpc_push(gsp, rpc, true, sizeof(*rpc));
-	if (IS_ERR_OR_NULL(rpc))
-		return rpc;
-
-	if (rpc->status) {
-		ret = ERR_PTR(r535_rpc_status_to_errno(rpc->status));
-		if (PTR_ERR(ret) != -EAGAIN && PTR_ERR(ret) != -EBUSY)
-			nvkm_error(&gsp->subdev, "RM_ALLOC: 0x%x\n", rpc->status);
-	}
-
-	nvkm_gsp_rpc_done(gsp, rpc);
-
-	return ret;
-}
-
-static void *
-r535_gsp_rpc_rm_alloc_get(struct nvkm_gsp_object *object, u32 oclass,
-			  u32 params_size)
-{
-	struct nvkm_gsp_client *client = object->client;
-	struct nvkm_gsp *gsp = client->gsp;
-	rpc_gsp_rm_alloc_v03_00 *rpc;
-
-	nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x new obj:0x%08x\n",
-		   client->object.handle, object->parent->handle,
-		   object->handle);
-
-	nvkm_debug(&gsp->subdev, "cls:0x%08x params_size:%d\n", oclass,
-		   params_size);
-
-	rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_ALLOC,
-			       sizeof(*rpc) + params_size);
-	if (IS_ERR(rpc))
-		return rpc;
-
-	rpc->hClient = client->object.handle;
-	rpc->hParent = object->parent->handle;
-	rpc->hObject = object->handle;
-	rpc->hClass = oclass;
-	rpc->status = 0;
-	rpc->paramsSize = params_size;
-	return rpc->params;
-}
-
-static void
-r535_gsp_rpc_rm_ctrl_done(struct nvkm_gsp_object *object, void *params)
-{
-	rpc_gsp_rm_control_v03_00 *rpc = to_payload_hdr(params, rpc);
-
-	if (!params)
-		return;
-	nvkm_gsp_rpc_done(object->client->gsp, rpc);
-}
-
-static int
-r535_gsp_rpc_rm_ctrl_push(struct nvkm_gsp_object *object, void **params, u32 repc)
-{
-	rpc_gsp_rm_control_v03_00 *rpc = to_payload_hdr((*params), rpc);
-	struct nvkm_gsp *gsp = object->client->gsp;
-	int ret = 0;
-
-	rpc = nvkm_gsp_rpc_push(gsp, rpc, true, repc);
-	if (IS_ERR_OR_NULL(rpc)) {
-		*params = NULL;
-		return PTR_ERR(rpc);
-	}
-
-	if (rpc->status) {
-		ret = r535_rpc_status_to_errno(rpc->status);
-		if (ret != -EAGAIN && ret != -EBUSY)
-			nvkm_error(&gsp->subdev, "cli:0x%08x obj:0x%08x ctrl cmd:0x%08x failed: 0x%08x\n",
-				   object->client->object.handle, object->handle, rpc->cmd, rpc->status);
-	}
-
-	if (repc)
-		*params = rpc->params;
-	else
-		nvkm_gsp_rpc_done(gsp, rpc);
-
-	return ret;
-}
-
-static void *
-r535_gsp_rpc_rm_ctrl_get(struct nvkm_gsp_object *object, u32 cmd, u32 params_size)
-{
-	struct nvkm_gsp_client *client = object->client;
-	struct nvkm_gsp *gsp = client->gsp;
-	rpc_gsp_rm_control_v03_00 *rpc;
-
-	nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x ctrl cmd:0x%08x params_size:%d\n",
-		   client->object.handle, object->handle, cmd, params_size);
-
-	rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_CONTROL,
-			       sizeof(*rpc) + params_size);
-	if (IS_ERR(rpc))
-		return rpc;
-
-	rpc->hClient    = client->object.handle;
-	rpc->hObject    = object->handle;
-	rpc->cmd	= cmd;
-	rpc->status     = 0;
-	rpc->paramsSize = params_size;
-	return rpc->params;
-}
-
-static void
-r535_gsp_rpc_done(struct nvkm_gsp *gsp, void *repv)
-{
-	struct nvfw_gsp_rpc *rpc = container_of(repv, typeof(*rpc), data);
-
-	r535_gsp_msg_done(gsp, rpc);
-}
-
-static void *
-r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 payload_size)
-{
-	struct nvfw_gsp_rpc *rpc;
-
-	rpc = r535_gsp_cmdq_get(gsp, ALIGN(sizeof(*rpc) + payload_size,
-					   sizeof(u64)));
-	if (IS_ERR(rpc))
-		return ERR_CAST(rpc);
-
-	rpc->header_version = 0x03000000;
-	rpc->signature = ('C' << 24) | ('P' << 16) | ('R' << 8) | 'V';
-	rpc->function = fn;
-	rpc->rpc_result = 0xffffffff;
-	rpc->rpc_result_private = 0xffffffff;
-	rpc->length = sizeof(*rpc) + payload_size;
-	return rpc->data;
-}
-
-static void *
-r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait,
-		  u32 gsp_rpc_len)
-{
-	struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc);
-	struct r535_gsp_msg *msg = to_gsp_hdr(rpc, msg);
-	const u32 max_rpc_size = GSP_MSG_MAX_SIZE - sizeof(*msg);
-	const u32 max_payload_size = max_rpc_size - sizeof(*rpc);
-	u32 payload_size = rpc->length - sizeof(*rpc);
-	void *repv;
-
-	mutex_lock(&gsp->cmdq.mutex);
-	if (payload_size > max_payload_size) {
-		const u32 fn = rpc->function;
-		u32 remain_payload_size = payload_size;
-
-		/* Adjust length, and send initial RPC. */
-		rpc->length = sizeof(*rpc) + max_payload_size;
-		msg->checksum = rpc->length;
-
-		repv = r535_gsp_rpc_send(gsp, payload, false, 0);
-		if (IS_ERR(repv))
-			goto done;
-
-		payload += max_payload_size;
-		remain_payload_size -= max_payload_size;
-
-		/* Remaining chunks sent as CONTINUATION_RECORD RPCs. */
-		while (remain_payload_size) {
-			u32 size = min(remain_payload_size,
-				       max_payload_size);
-			void *next;
-
-			next = r535_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD, size);
-			if (IS_ERR(next)) {
-				repv = next;
-				goto done;
-			}
-
-			memcpy(next, payload, size);
-
-			repv = r535_gsp_rpc_send(gsp, next, false, 0);
-			if (IS_ERR(repv))
-				goto done;
-
-			payload += size;
-			remain_payload_size -= size;
-		}
-
-		/* Wait for reply. */
-		rpc = r535_gsp_msg_recv(gsp, fn, payload_size +
-					sizeof(*rpc));
-		if (!IS_ERR_OR_NULL(rpc)) {
-			if (wait) {
-				repv = rpc->data;
-			} else {
-				nvkm_gsp_rpc_done(gsp, rpc);
-				repv = NULL;
-			}
-		} else {
-			repv = wait ? rpc : NULL;
-		}
-	} else {
-		repv = r535_gsp_rpc_send(gsp, payload, wait, gsp_rpc_len);
-	}
-
-done:
-	mutex_unlock(&gsp->cmdq.mutex);
-	return repv;
-}
-
-const struct nvkm_gsp_rm
-r535_gsp_rm = {
-	.rpc_get = r535_gsp_rpc_get,
-	.rpc_push = r535_gsp_rpc_push,
-	.rpc_done = r535_gsp_rpc_done,
-
-	.rm_ctrl_get = r535_gsp_rpc_rm_ctrl_get,
-	.rm_ctrl_push = r535_gsp_rpc_rm_ctrl_push,
-	.rm_ctrl_done = r535_gsp_rpc_rm_ctrl_done,
-
-	.rm_alloc_get = r535_gsp_rpc_rm_alloc_get,
-	.rm_alloc_push = r535_gsp_rpc_rm_alloc_push,
-	.rm_alloc_done = r535_gsp_rpc_rm_alloc_done,
-
-	.rm_free = r535_gsp_rpc_rm_free,
-
-	.client_ctor = r535_gsp_client_ctor,
-	.client_dtor = r535_gsp_client_dtor,
-
-	.device_ctor = r535_gsp_device_ctor,
-	.device_dtor = r535_gsp_device_dtor,
-
-	.event_ctor = r535_gsp_device_event_ctor,
-	.event_dtor = r535_gsp_event_dtor,
-};
-
 static void
 r535_gsp_msgq_work(struct work_struct *work)
 {
@@ -1086,10 +87,52 @@ r535_gsp_intr(struct nvkm_inth *inth)
 	return IRQ_HANDLED;
 }
 
+static bool
+r535_gsp_xlat_mc_engine_idx(u32 mc_engine_idx, enum nvkm_subdev_type *ptype, int *pinst)
+{
+	switch (mc_engine_idx) {
+	case MC_ENGINE_IDX_GSP:
+		*ptype = NVKM_SUBDEV_GSP;
+		*pinst = 0;
+		return true;
+	case MC_ENGINE_IDX_DISP:
+		*ptype = NVKM_ENGINE_DISP;
+		*pinst = 0;
+		return true;
+	case MC_ENGINE_IDX_CE0 ... MC_ENGINE_IDX_CE9:
+		*ptype = NVKM_ENGINE_CE;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_CE0;
+		return true;
+	case MC_ENGINE_IDX_GR0:
+		*ptype = NVKM_ENGINE_GR;
+		*pinst = 0;
+		return true;
+	case MC_ENGINE_IDX_NVDEC0 ... MC_ENGINE_IDX_NVDEC7:
+		*ptype = NVKM_ENGINE_NVDEC;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_NVDEC0;
+		return true;
+	case MC_ENGINE_IDX_MSENC ... MC_ENGINE_IDX_MSENC2:
+		*ptype = NVKM_ENGINE_NVENC;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_MSENC;
+		return true;
+	case MC_ENGINE_IDX_NVJPEG0 ... MC_ENGINE_IDX_NVJPEG7:
+		*ptype = NVKM_ENGINE_NVJPG;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_NVJPEG0;
+		return true;
+	case MC_ENGINE_IDX_OFA0:
+		*ptype = NVKM_ENGINE_OFA;
+		*pinst = 0;
+		return true;
+	default:
+		return false;
+	}
+}
+
 static int
 r535_gsp_intr_get_table(struct nvkm_gsp *gsp)
 {
 	NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS *ctrl;
+	const struct nvkm_rm_api *rmapi = gsp->rm->api;
 	int ret = 0;
 
 	ctrl = nvkm_gsp_rm_ctrl_get(&gsp->internal.device.subdevice,
@@ -1112,42 +155,8 @@ r535_gsp_intr_get_table(struct nvkm_gsp *gsp)
 			   ctrl->table[i].engineIdx, ctrl->table[i].pmcIntrMask,
 			   ctrl->table[i].vectorStall, ctrl->table[i].vectorNonStall);
 
-		switch (ctrl->table[i].engineIdx) {
-		case MC_ENGINE_IDX_GSP:
-			type = NVKM_SUBDEV_GSP;
-			inst = 0;
-			break;
-		case MC_ENGINE_IDX_DISP:
-			type = NVKM_ENGINE_DISP;
-			inst = 0;
-			break;
-		case MC_ENGINE_IDX_CE0 ... MC_ENGINE_IDX_CE9:
-			type = NVKM_ENGINE_CE;
-			inst = ctrl->table[i].engineIdx - MC_ENGINE_IDX_CE0;
-			break;
-		case MC_ENGINE_IDX_GR0:
-			type = NVKM_ENGINE_GR;
-			inst = 0;
-			break;
-		case MC_ENGINE_IDX_NVDEC0 ... MC_ENGINE_IDX_NVDEC7:
-			type = NVKM_ENGINE_NVDEC;
-			inst = ctrl->table[i].engineIdx - MC_ENGINE_IDX_NVDEC0;
-			break;
-		case MC_ENGINE_IDX_MSENC ... MC_ENGINE_IDX_MSENC2:
-			type = NVKM_ENGINE_NVENC;
-			inst = ctrl->table[i].engineIdx - MC_ENGINE_IDX_MSENC;
-			break;
-		case MC_ENGINE_IDX_NVJPEG0 ... MC_ENGINE_IDX_NVJPEG7:
-			type = NVKM_ENGINE_NVJPG;
-			inst = ctrl->table[i].engineIdx - MC_ENGINE_IDX_NVJPEG0;
-			break;
-		case MC_ENGINE_IDX_OFA0:
-			type = NVKM_ENGINE_OFA;
-			inst = 0;
-			break;
-		default:
+		if (!rmapi->gsp->xlat_mc_engine_idx(ctrl->table[i].engineIdx, &type, &inst))
 			continue;
-		}
 
 		if (WARN_ON(gsp->intr_nr == ARRAY_SIZE(gsp->intr))) {
 			ret = -ENOSPC;
@@ -1165,35 +174,14 @@ r535_gsp_intr_get_table(struct nvkm_gsp *gsp)
 	return ret;
 }
 
-static int
-r535_gsp_rpc_get_gsp_static_info(struct nvkm_gsp *gsp)
+void
+r535_gsp_get_static_info_fb(struct nvkm_gsp *gsp,
+			    const struct NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS *info)
 {
-	GspStaticConfigInfo *rpc;
 	int last_usable = -1;
 
-	rpc = nvkm_gsp_rpc_rd(gsp, NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO, sizeof(*rpc));
-	if (IS_ERR(rpc))
-		return PTR_ERR(rpc);
-
-	gsp->internal.client.object.client = &gsp->internal.client;
-	gsp->internal.client.object.parent = NULL;
-	gsp->internal.client.object.handle = rpc->hInternalClient;
-	gsp->internal.client.gsp = gsp;
-
-	gsp->internal.device.object.client = &gsp->internal.client;
-	gsp->internal.device.object.parent = &gsp->internal.client.object;
-	gsp->internal.device.object.handle = rpc->hInternalDevice;
-
-	gsp->internal.device.subdevice.client = &gsp->internal.client;
-	gsp->internal.device.subdevice.parent = &gsp->internal.device.object;
-	gsp->internal.device.subdevice.handle = rpc->hInternalSubdevice;
-
-	gsp->bar.rm_bar1_pdb = rpc->bar1PdeBase;
-	gsp->bar.rm_bar2_pdb = rpc->bar2PdeBase;
-
-	for (int i = 0; i < rpc->fbRegionInfoParams.numFBRegions; i++) {
-		NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO *reg =
-			&rpc->fbRegionInfoParams.fbRegion[i];
+	for (int i = 0; i < info->numFBRegions; i++) {
+		const NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO *reg = &info->fbRegion[i];
 
 		nvkm_debug(&gsp->subdev, "fb region %d: "
 			   "%016llx-%016llx rsvd:%016llx perf:%08x comp:%d iso:%d prot:%d\n", i,
@@ -1215,10 +203,38 @@ r535_gsp_rpc_get_gsp_static_info(struct nvkm_gsp *gsp)
 	}
 
 	if (last_usable >= 0) {
-		u32 rsvd_base = rpc->fbRegionInfoParams.fbRegion[last_usable].limit + 1;
+		u32 rsvd_base = info->fbRegion[last_usable].limit + 1;
 
 		gsp->fb.rsvd_size = gsp->fb.heap.addr - rsvd_base;
 	}
+}
+
+static int
+r535_gsp_get_static_info(struct nvkm_gsp *gsp)
+{
+	GspStaticConfigInfo *rpc;
+
+	rpc = nvkm_gsp_rpc_rd(gsp, NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO, sizeof(*rpc));
+	if (IS_ERR(rpc))
+		return PTR_ERR(rpc);
+
+	gsp->internal.client.object.client = &gsp->internal.client;
+	gsp->internal.client.object.parent = NULL;
+	gsp->internal.client.object.handle = rpc->hInternalClient;
+	gsp->internal.client.gsp = gsp;
+
+	gsp->internal.device.object.client = &gsp->internal.client;
+	gsp->internal.device.object.parent = &gsp->internal.client.object;
+	gsp->internal.device.object.handle = rpc->hInternalDevice;
+
+	gsp->internal.device.subdevice.client = &gsp->internal.client;
+	gsp->internal.device.subdevice.parent = &gsp->internal.device.object;
+	gsp->internal.device.subdevice.handle = rpc->hInternalSubdevice;
+
+	gsp->bar.rm_bar1_pdb = rpc->bar1PdeBase;
+	gsp->bar.rm_bar2_pdb = rpc->bar2PdeBase;
+
+	r535_gsp_get_static_info_fb(gsp, &rpc->fbRegionInfoParams);
 
 	for (int gpc = 0; gpc < ARRAY_SIZE(rpc->tpcInfo); gpc++) {
 		if (rpc->gpcInfo.gpcMask & BIT(gpc)) {
@@ -1231,7 +247,7 @@ r535_gsp_rpc_get_gsp_static_info(struct nvkm_gsp *gsp)
 	return 0;
 }
 
-static void
+void
 nvkm_gsp_mem_dtor(struct nvkm_gsp_mem *mem)
 {
 	if (mem->data) {
@@ -1260,7 +276,7 @@ nvkm_gsp_mem_dtor(struct nvkm_gsp_mem *mem)
  * so we take a device reference to ensure its lifetime. The reference is
  * dropped in the destructor.
  */
-static int
+int
 nvkm_gsp_mem_ctor(struct nvkm_gsp *gsp, size_t size, struct nvkm_gsp_mem *mem)
 {
 	mem->data = dma_alloc_coherent(gsp->subdev.device->dev, size, &mem->addr, GFP_KERNEL);
@@ -1277,9 +293,10 @@ static int
 r535_gsp_postinit(struct nvkm_gsp *gsp)
 {
 	struct nvkm_device *device = gsp->subdev.device;
+	const struct nvkm_rm_api *rmapi = gsp->rm->api;
 	int ret;
 
-	ret = r535_gsp_rpc_get_gsp_static_info(gsp);
+	ret = rmapi->gsp->get_static_info(gsp);
 	if (WARN_ON(ret))
 		return ret;
 
@@ -1327,7 +344,7 @@ r535_gsp_rpc_unloading_guest_driver(struct nvkm_gsp *gsp, bool suspend)
 		rpc->newLevel = NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_0;
 	}
 
-	return nvkm_gsp_rpc_wr(gsp, rpc, true);
+	return nvkm_gsp_rpc_wr(gsp, rpc, NVKM_GSP_RPC_REPLY_RECV);
 }
 
 enum registry_type {
@@ -1684,7 +701,7 @@ r535_gsp_rpc_set_registry(struct nvkm_gsp *gsp)
 
 	build_registry(gsp, rpc);
 
-	return nvkm_gsp_rpc_wr(gsp, rpc, false);
+	return nvkm_gsp_rpc_wr(gsp, rpc, NVKM_GSP_RPC_REPLY_NOWAIT);
 
 fail:
 	clean_registry(gsp);
@@ -1692,7 +709,7 @@ fail:
 }
 
 #if defined(CONFIG_ACPI) && defined(CONFIG_X86)
-static void
+void
 r535_gsp_acpi_caps(acpi_handle handle, CAPS_METHOD_DATA *caps)
 {
 	const guid_t NVOP_DSM_GUID =
@@ -1726,7 +743,7 @@ r535_gsp_acpi_caps(acpi_handle handle, CAPS_METHOD_DATA *caps)
 	kfree(argv4.buffer.pointer);
 }
 
-static void
+void
 r535_gsp_acpi_jt(acpi_handle handle, JT_METHOD_DATA *jt)
 {
 	const guid_t JT_DSM_GUID =
@@ -1818,7 +835,7 @@ r535_gsp_acpi_mux(acpi_handle handle, DOD_METHOD_DATA *dod, MUX_METHOD_DATA *mux
 	}
 }
 
-static void
+void
 r535_gsp_acpi_dod(acpi_handle handle, DOD_METHOD_DATA *dod)
 {
 	acpi_status status;
@@ -1871,7 +888,7 @@ r535_gsp_acpi_info(struct nvkm_gsp *gsp, ACPI_METHOD_DATA *acpi)
 }
 
 static int
-r535_gsp_rpc_set_system_info(struct nvkm_gsp *gsp)
+r535_gsp_set_system_info(struct nvkm_gsp *gsp)
 {
 	struct nvkm_device *device = gsp->subdev.device;
 	struct nvkm_device_pci *pdev = container_of(device, typeof(*pdev), device);
@@ -1884,16 +901,16 @@ r535_gsp_rpc_set_system_info(struct nvkm_gsp *gsp)
 	if (IS_ERR(info))
 		return PTR_ERR(info);
 
-	info->gpuPhysAddr = device->func->resource_addr(device, 0);
-	info->gpuPhysFbAddr = device->func->resource_addr(device, 1);
-	info->gpuPhysInstAddr = device->func->resource_addr(device, 3);
+	info->gpuPhysAddr = device->func->resource_addr(device, NVKM_BAR0_PRI);
+	info->gpuPhysFbAddr = device->func->resource_addr(device, NVKM_BAR1_FB);
+	info->gpuPhysInstAddr = device->func->resource_addr(device, NVKM_BAR2_INST);
 	info->nvDomainBusDeviceFunc = pci_dev_id(pdev->pdev);
 	info->maxUserVa = TASK_SIZE;
-	info->pciConfigMirrorBase = 0x088000;
-	info->pciConfigMirrorSize = 0x001000;
+	info->pciConfigMirrorBase = device->pci->func->cfg.addr;
+	info->pciConfigMirrorSize = device->pci->func->cfg.size;
 	r535_gsp_acpi_info(gsp, &info->acpiMethodData);
 
-	return nvkm_gsp_rpc_wr(gsp, info, false);
+	return nvkm_gsp_rpc_wr(gsp, info, NVKM_GSP_RPC_REPLY_NOWAIT);
 }
 
 static int
@@ -1911,33 +928,6 @@ r535_gsp_msg_os_error_log(void *priv, u32 fn, void *repv, u32 repc)
 }
 
 static int
-r535_gsp_msg_rc_triggered(void *priv, u32 fn, void *repv, u32 repc)
-{
-	rpc_rc_triggered_v17_02 *msg = repv;
-	struct nvkm_gsp *gsp = priv;
-	struct nvkm_subdev *subdev = &gsp->subdev;
-	struct nvkm_chan *chan;
-	unsigned long flags;
-
-	if (WARN_ON(repc < sizeof(*msg)))
-		return -EINVAL;
-
-	nvkm_error(subdev, "rc engn:%08x chid:%d type:%d scope:%d part:%d\n",
-		   msg->nv2080EngineType, msg->chid, msg->exceptType, msg->scope,
-		   msg->partitionAttributionId);
-
-	chan = nvkm_chan_get_chid(&subdev->device->fifo->engine, msg->chid / 8, &flags);
-	if (!chan) {
-		nvkm_error(subdev, "rc chid:%d not found!\n", msg->chid);
-		return 0;
-	}
-
-	nvkm_chan_error(chan, false);
-	nvkm_chan_put(&chan, flags);
-	return 0;
-}
-
-static int
 r535_gsp_msg_mmu_fault_queued(void *priv, u32 fn, void *repv, u32 repc)
 {
 	struct nvkm_gsp *gsp = priv;
@@ -2130,97 +1120,6 @@ r535_gsp_msg_run_cpu_sequencer(void *priv, u32 fn, void *repv, u32 repc)
 }
 
 static int
-r535_gsp_booter_unload(struct nvkm_gsp *gsp, u32 mbox0, u32 mbox1)
-{
-	struct nvkm_subdev *subdev = &gsp->subdev;
-	struct nvkm_device *device = subdev->device;
-	u32 wpr2_hi;
-	int ret;
-
-	wpr2_hi = nvkm_rd32(device, 0x1fa828);
-	if (!wpr2_hi) {
-		nvkm_debug(subdev, "WPR2 not set - skipping booter unload\n");
-		return 0;
-	}
-
-	ret = nvkm_falcon_fw_boot(&gsp->booter.unload, &gsp->subdev, true, &mbox0, &mbox1, 0, 0);
-	if (WARN_ON(ret))
-		return ret;
-
-	wpr2_hi = nvkm_rd32(device, 0x1fa828);
-	if (WARN_ON(wpr2_hi))
-		return -EIO;
-
-	return 0;
-}
-
-static int
-r535_gsp_booter_load(struct nvkm_gsp *gsp, u32 mbox0, u32 mbox1)
-{
-	int ret;
-
-	ret = nvkm_falcon_fw_boot(&gsp->booter.load, &gsp->subdev, true, &mbox0, &mbox1, 0, 0);
-	if (ret)
-		return ret;
-
-	nvkm_falcon_wr32(&gsp->falcon, 0x080, gsp->boot.app_version);
-
-	if (WARN_ON(!nvkm_falcon_riscv_active(&gsp->falcon)))
-		return -EIO;
-
-	return 0;
-}
-
-static int
-r535_gsp_wpr_meta_init(struct nvkm_gsp *gsp)
-{
-	GspFwWprMeta *meta;
-	int ret;
-
-	ret = nvkm_gsp_mem_ctor(gsp, 0x1000, &gsp->wpr_meta);
-	if (ret)
-		return ret;
-
-	meta = gsp->wpr_meta.data;
-
-	meta->magic = GSP_FW_WPR_META_MAGIC;
-	meta->revision = GSP_FW_WPR_META_REVISION;
-
-	meta->sysmemAddrOfRadix3Elf = gsp->radix3.lvl0.addr;
-	meta->sizeOfRadix3Elf = gsp->fb.wpr2.elf.size;
-
-	meta->sysmemAddrOfBootloader = gsp->boot.fw.addr;
-	meta->sizeOfBootloader = gsp->boot.fw.size;
-	meta->bootloaderCodeOffset = gsp->boot.code_offset;
-	meta->bootloaderDataOffset = gsp->boot.data_offset;
-	meta->bootloaderManifestOffset = gsp->boot.manifest_offset;
-
-	meta->sysmemAddrOfSignature = gsp->sig.addr;
-	meta->sizeOfSignature = gsp->sig.size;
-
-	meta->gspFwRsvdStart = gsp->fb.heap.addr;
-	meta->nonWprHeapOffset = gsp->fb.heap.addr;
-	meta->nonWprHeapSize = gsp->fb.heap.size;
-	meta->gspFwWprStart = gsp->fb.wpr2.addr;
-	meta->gspFwHeapOffset = gsp->fb.wpr2.heap.addr;
-	meta->gspFwHeapSize = gsp->fb.wpr2.heap.size;
-	meta->gspFwOffset = gsp->fb.wpr2.elf.addr;
-	meta->bootBinOffset = gsp->fb.wpr2.boot.addr;
-	meta->frtsOffset = gsp->fb.wpr2.frts.addr;
-	meta->frtsSize = gsp->fb.wpr2.frts.size;
-	meta->gspFwWprEnd = ALIGN_DOWN(gsp->fb.bios.vga_workspace.addr, 0x20000);
-	meta->fbSize = gsp->fb.size;
-	meta->vgaWorkspaceOffset = gsp->fb.bios.vga_workspace.addr;
-	meta->vgaWorkspaceSize = gsp->fb.bios.vga_workspace.size;
-	meta->bootCount = 0;
-	meta->partitionRpcAddr = 0;
-	meta->partitionRpcRequestOffset = 0;
-	meta->partitionRpcReplyOffset = 0;
-	meta->verified = 0;
-	return 0;
-}
-
-static int
 r535_gsp_shared_init(struct nvkm_gsp *gsp)
 {
 	struct {
@@ -2271,23 +1170,11 @@ r535_gsp_shared_init(struct nvkm_gsp *gsp)
 	return 0;
 }
 
-static int
-r535_gsp_rmargs_init(struct nvkm_gsp *gsp, bool resume)
+static void
+r535_gsp_set_rmargs(struct nvkm_gsp *gsp, bool resume)
 {
-	GSP_ARGUMENTS_CACHED *args;
-	int ret;
-
-	if (!resume) {
-		ret = r535_gsp_shared_init(gsp);
-		if (ret)
-			return ret;
+	GSP_ARGUMENTS_CACHED *args = gsp->rmargs.data;
 
-		ret = nvkm_gsp_mem_ctor(gsp, 0x1000, &gsp->rmargs);
-		if (ret)
-			return ret;
-	}
-
-	args = gsp->rmargs.data;
 	args->messageQueueInitArguments.sharedMemPhysAddr = gsp->shm.mem.addr;
 	args->messageQueueInitArguments.pageTableEntryCount = gsp->shm.ptes.nr;
 	args->messageQueueInitArguments.cmdQueueOffset =
@@ -2304,7 +1191,24 @@ r535_gsp_rmargs_init(struct nvkm_gsp *gsp, bool resume)
 		args->srInitArguments.flags = 0;
 		args->srInitArguments.bInPMTransition = 1;
 	}
+}
 
+static int
+r535_gsp_rmargs_init(struct nvkm_gsp *gsp, bool resume)
+{
+	int ret;
+
+	if (!resume) {
+		ret = r535_gsp_shared_init(gsp);
+		if (ret)
+			return ret;
+
+		ret = nvkm_gsp_mem_ctor(gsp, 0x1000, &gsp->rmargs);
+		if (ret)
+			return ret;
+	}
+
+	gsp->rm->api->gsp->set_rmargs(gsp, resume);
 	return 0;
 }
 
@@ -2797,18 +1701,22 @@ lvl1_fail:
 	return ret;
 }
 
+static u32
+r535_gsp_sr_data_size(struct nvkm_gsp *gsp)
+{
+	GspFwWprMeta *meta = gsp->wpr_meta.data;
+
+	return meta->gspFwWprEnd - meta->gspFwWprStart;
+}
+
 int
 r535_gsp_fini(struct nvkm_gsp *gsp, bool suspend)
 {
-	u32 mbox0 = 0xff, mbox1 = 0xff;
+	struct nvkm_rm *rm = gsp->rm;
 	int ret;
 
-	if (!gsp->running)
-		return 0;
-
 	if (suspend) {
-		GspFwWprMeta *meta = gsp->wpr_meta.data;
-		u64 len = meta->gspFwWprEnd - meta->gspFwWprStart;
+		u32 len = rm->api->gsp->sr_data_size(gsp);
 		GspFwSRMeta *sr;
 
 		ret = nvkm_gsp_sg(gsp->subdev.device, len, &gsp->sr.sgt);
@@ -2829,8 +1737,13 @@ r535_gsp_fini(struct nvkm_gsp *gsp, bool suspend)
 		sr->sysmemAddrOfSuspendResumeData = gsp->sr.radix3.lvl0.addr;
 		sr->sizeOfSuspendResumeData = len;
 
-		mbox0 = lower_32_bits(gsp->sr.meta.addr);
-		mbox1 = upper_32_bits(gsp->sr.meta.addr);
+		ret = rm->api->fbsr->suspend(gsp);
+		if (ret) {
+			nvkm_gsp_mem_dtor(&gsp->sr.meta);
+			nvkm_gsp_radix3_dtor(gsp, &gsp->sr.radix3);
+			nvkm_gsp_sg_free(gsp->subdev.device, &gsp->sr.sgt);
+			return ret;
+		}
 	}
 
 	ret = r535_gsp_rpc_unloading_guest_driver(gsp, suspend);
@@ -2838,18 +1751,10 @@ r535_gsp_fini(struct nvkm_gsp *gsp, bool suspend)
 		return ret;
 
 	nvkm_msec(gsp->subdev.device, 2000,
-		if (nvkm_falcon_rd32(&gsp->falcon, 0x040) & 0x80000000)
+		if (nvkm_falcon_rd32(&gsp->falcon, 0x040) == 0x80000000)
 			break;
 	);
 
-	nvkm_falcon_reset(&gsp->falcon);
-
-	ret = nvkm_gsp_fwsec_sb(gsp);
-	WARN_ON(ret);
-
-	ret = r535_gsp_booter_unload(gsp, mbox0, mbox1);
-	WARN_ON(ret);
-
 	gsp->running = false;
 	return 0;
 }
@@ -2857,23 +1762,12 @@ r535_gsp_fini(struct nvkm_gsp *gsp, bool suspend)
 int
 r535_gsp_init(struct nvkm_gsp *gsp)
 {
-	u32 mbox0, mbox1;
 	int ret;
 
-	if (!gsp->sr.meta.data) {
-		mbox0 = lower_32_bits(gsp->wpr_meta.addr);
-		mbox1 = upper_32_bits(gsp->wpr_meta.addr);
-	} else {
-		r535_gsp_rmargs_init(gsp, true);
-
-		mbox0 = lower_32_bits(gsp->sr.meta.addr);
-		mbox1 = upper_32_bits(gsp->sr.meta.addr);
-	}
+	nvkm_falcon_wr32(&gsp->falcon, 0x080, gsp->boot.app_version);
 
-	/* Execute booter to handle (eventually...) booting GSP-RM. */
-	ret = r535_gsp_booter_load(gsp, mbox0, mbox1);
-	if (WARN_ON(ret))
-		goto done;
+	if (WARN_ON(!nvkm_falcon_riscv_active(&gsp->falcon)))
+		return -EIO;
 
 	ret = r535_gsp_rpc_poll(gsp, NV_VGPU_MSG_EVENT_GSP_INIT_DONE);
 	if (ret)
@@ -2883,6 +1777,8 @@ r535_gsp_init(struct nvkm_gsp *gsp)
 
 done:
 	if (gsp->sr.meta.data) {
+		gsp->rm->api->fbsr->resume(gsp);
+
 		nvkm_gsp_mem_dtor(&gsp->sr.meta);
 		nvkm_gsp_radix3_dtor(gsp, &gsp->sr.radix3);
 		nvkm_gsp_sg_free(gsp->subdev.device, &gsp->sr.sgt);
@@ -2944,19 +1840,6 @@ r535_gsp_elf_section(struct nvkm_gsp *gsp, const char *name, const u8 **pdata, u
 	return -ENOENT;
 }
 
-static void
-r535_gsp_dtor_fws(struct nvkm_gsp *gsp)
-{
-	nvkm_firmware_put(gsp->fws.bl);
-	gsp->fws.bl = NULL;
-	nvkm_firmware_put(gsp->fws.booter.unload);
-	gsp->fws.booter.unload = NULL;
-	nvkm_firmware_put(gsp->fws.booter.load);
-	gsp->fws.booter.load = NULL;
-	nvkm_firmware_put(gsp->fws.rm);
-	gsp->fws.rm = NULL;
-}
-
 #ifdef CONFIG_DEBUG_FS
 
 struct r535_gsp_log {
@@ -3190,10 +2073,16 @@ r535_gsp_dtor(struct nvkm_gsp *gsp)
 	nvkm_falcon_fw_dtor(&gsp->booter.unload);
 	nvkm_falcon_fw_dtor(&gsp->booter.load);
 
+	nvkm_gsp_mem_dtor(&gsp->fmc.args);
+	kfree(gsp->fmc.sig);
+	kfree(gsp->fmc.pkey);
+	kfree(gsp->fmc.hash);
+	nvkm_gsp_mem_dtor(&gsp->fmc.fw);
+
 	mutex_destroy(&gsp->msgq.mutex);
 	mutex_destroy(&gsp->cmdq.mutex);
 
-	r535_gsp_dtor_fws(gsp);
+	nvkm_gsp_dtor_fws(gsp);
 
 	nvkm_gsp_mem_dtor(&gsp->rmargs);
 	nvkm_gsp_mem_dtor(&gsp->wpr_meta);
@@ -3206,10 +2095,17 @@ r535_gsp_dtor(struct nvkm_gsp *gsp)
 	nvkm_gsp_mem_dtor(&gsp->logrm);
 }
 
+static void
+r535_gsp_drop_send_user_shared_data(struct nvkm_gsp *gsp)
+{
+	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_GSP_SEND_USER_SHARED_DATA, NULL, NULL);
+}
+
 int
 r535_gsp_oneinit(struct nvkm_gsp *gsp)
 {
 	struct nvkm_device *device = gsp->subdev.device;
+	const struct nvkm_rm_api *rmapi = gsp->rm->api;
 	const u8 *data;
 	u64 size;
 	int ret;
@@ -3217,16 +2113,6 @@ r535_gsp_oneinit(struct nvkm_gsp *gsp)
 	mutex_init(&gsp->cmdq.mutex);
 	mutex_init(&gsp->msgq.mutex);
 
-	ret = gsp->func->booter.ctor(gsp, "booter-load", gsp->fws.booter.load,
-				     &device->sec2->falcon, &gsp->booter.load);
-	if (ret)
-		return ret;
-
-	ret = gsp->func->booter.ctor(gsp, "booter-unload", gsp->fws.booter.unload,
-				     &device->sec2->falcon, &gsp->booter.unload);
-	if (ret)
-		return ret;
-
 	/* Load GSP firmware from ELF image into DMA-accessible memory. */
 	ret = r535_gsp_elf_section(gsp, ".fwimage", &data, &size);
 	if (ret)
@@ -3255,65 +2141,29 @@ r535_gsp_oneinit(struct nvkm_gsp *gsp)
 	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_GSP_RUN_CPU_SEQUENCER,
 			      r535_gsp_msg_run_cpu_sequencer, gsp);
 	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_POST_EVENT, r535_gsp_msg_post_event, gsp);
-	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_RC_TRIGGERED,
-			      r535_gsp_msg_rc_triggered, gsp);
+	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_RC_TRIGGERED, rmapi->fifo->rc_triggered, gsp);
 	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_MMU_FAULT_QUEUED,
 			      r535_gsp_msg_mmu_fault_queued, gsp);
 	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_OS_ERROR_LOG, r535_gsp_msg_os_error_log, gsp);
 	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_PERF_BRIDGELESS_INFO_UPDATE, NULL, NULL);
 	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_UCODE_LIBOS_PRINT, NULL, NULL);
-	r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_GSP_SEND_USER_SHARED_DATA, NULL, NULL);
+	if (rmapi->gsp->drop_send_user_shared_data)
+		rmapi->gsp->drop_send_user_shared_data(gsp);
+	if (rmapi->gsp->drop_post_nocat_record)
+		rmapi->gsp->drop_post_nocat_record(gsp);
+
 	ret = r535_gsp_rm_boot_ctor(gsp);
 	if (ret)
 		return ret;
 
 	/* Release FW images - we've copied them to DMA buffers now. */
-	r535_gsp_dtor_fws(gsp);
-
-	/* Calculate FB layout. */
-	gsp->fb.wpr2.frts.size = 0x100000;
-	gsp->fb.wpr2.frts.addr = ALIGN_DOWN(gsp->fb.bios.addr, 0x20000) - gsp->fb.wpr2.frts.size;
-
-	gsp->fb.wpr2.boot.size = gsp->boot.fw.size;
-	gsp->fb.wpr2.boot.addr = ALIGN_DOWN(gsp->fb.wpr2.frts.addr - gsp->fb.wpr2.boot.size, 0x1000);
-
-	gsp->fb.wpr2.elf.size = gsp->fw.len;
-	gsp->fb.wpr2.elf.addr = ALIGN_DOWN(gsp->fb.wpr2.boot.addr - gsp->fb.wpr2.elf.size, 0x10000);
-
-	{
-		u32 fb_size_gb = DIV_ROUND_UP_ULL(gsp->fb.size, 1 << 30);
-
-		gsp->fb.wpr2.heap.size =
-			gsp->func->wpr_heap.os_carveout_size +
-			gsp->func->wpr_heap.base_size +
-			ALIGN(GSP_FW_HEAP_PARAM_SIZE_PER_GB_FB * fb_size_gb, 1 << 20) +
-			ALIGN(GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE, 1 << 20);
-
-		gsp->fb.wpr2.heap.size = max(gsp->fb.wpr2.heap.size, gsp->func->wpr_heap.min_size);
-	}
-
-	gsp->fb.wpr2.heap.addr = ALIGN_DOWN(gsp->fb.wpr2.elf.addr - gsp->fb.wpr2.heap.size, 0x100000);
-	gsp->fb.wpr2.heap.size = ALIGN_DOWN(gsp->fb.wpr2.elf.addr - gsp->fb.wpr2.heap.addr, 0x100000);
-
-	gsp->fb.wpr2.addr = ALIGN_DOWN(gsp->fb.wpr2.heap.addr - sizeof(GspFwWprMeta), 0x100000);
-	gsp->fb.wpr2.size = gsp->fb.wpr2.frts.addr + gsp->fb.wpr2.frts.size - gsp->fb.wpr2.addr;
-
-	gsp->fb.heap.size = 0x100000;
-	gsp->fb.heap.addr = gsp->fb.wpr2.addr - gsp->fb.heap.size;
-
-	ret = nvkm_gsp_fwsec_frts(gsp);
-	if (WARN_ON(ret))
-		return ret;
+	nvkm_gsp_dtor_fws(gsp);
 
 	ret = r535_gsp_libos_init(gsp);
 	if (WARN_ON(ret))
 		return ret;
 
-	ret = r535_gsp_wpr_meta_init(gsp);
-	if (WARN_ON(ret))
-		return ret;
-
-	ret = r535_gsp_rpc_set_system_info(gsp);
+	ret = rmapi->gsp->set_system_info(gsp);
 	if (WARN_ON(ret))
 		return ret;
 
@@ -3321,76 +2171,17 @@ r535_gsp_oneinit(struct nvkm_gsp *gsp)
 	if (WARN_ON(ret))
 		return ret;
 
-	/* Reset GSP into RISC-V mode. */
-	ret = gsp->func->reset(gsp);
-	if (WARN_ON(ret))
-		return ret;
-
-	nvkm_falcon_wr32(&gsp->falcon, 0x040, lower_32_bits(gsp->libos.addr));
-	nvkm_falcon_wr32(&gsp->falcon, 0x044, upper_32_bits(gsp->libos.addr));
-
 	mutex_init(&gsp->client_id.mutex);
 	idr_init(&gsp->client_id.idr);
 	return 0;
 }
 
-static int
-r535_gsp_load_fw(struct nvkm_gsp *gsp, const char *name, const char *ver,
-		 const struct firmware **pfw)
-{
-	char fwname[64];
-
-	snprintf(fwname, sizeof(fwname), "gsp/%s-%s", name, ver);
-	return nvkm_firmware_get(&gsp->subdev, fwname, 0, pfw);
-}
-
-int
-r535_gsp_load(struct nvkm_gsp *gsp, int ver, const struct nvkm_gsp_fwif *fwif)
-{
-	struct nvkm_subdev *subdev = &gsp->subdev;
-	int ret;
-	bool enable_gsp = fwif->enable;
-
-#if IS_ENABLED(CONFIG_DRM_NOUVEAU_GSP_DEFAULT)
-	enable_gsp = true;
-#endif
-	if (!nvkm_boolopt(subdev->device->cfgopt, "NvGspRm", enable_gsp))
-		return -EINVAL;
-
-	if ((ret = r535_gsp_load_fw(gsp, "gsp", fwif->ver, &gsp->fws.rm)) ||
-	    (ret = r535_gsp_load_fw(gsp, "booter_load", fwif->ver, &gsp->fws.booter.load)) ||
-	    (ret = r535_gsp_load_fw(gsp, "booter_unload", fwif->ver, &gsp->fws.booter.unload)) ||
-	    (ret = r535_gsp_load_fw(gsp, "bootloader", fwif->ver, &gsp->fws.bl))) {
-		r535_gsp_dtor_fws(gsp);
-		return ret;
-	}
-
-	return 0;
-}
-
-#define NVKM_GSP_FIRMWARE(chip)                                  \
-MODULE_FIRMWARE("nvidia/"#chip"/gsp/booter_load-535.113.01.bin");   \
-MODULE_FIRMWARE("nvidia/"#chip"/gsp/booter_unload-535.113.01.bin"); \
-MODULE_FIRMWARE("nvidia/"#chip"/gsp/bootloader-535.113.01.bin");    \
-MODULE_FIRMWARE("nvidia/"#chip"/gsp/gsp-535.113.01.bin")
-
-NVKM_GSP_FIRMWARE(tu102);
-NVKM_GSP_FIRMWARE(tu104);
-NVKM_GSP_FIRMWARE(tu106);
-
-NVKM_GSP_FIRMWARE(tu116);
-NVKM_GSP_FIRMWARE(tu117);
-
-NVKM_GSP_FIRMWARE(ga100);
-
-NVKM_GSP_FIRMWARE(ga102);
-NVKM_GSP_FIRMWARE(ga103);
-NVKM_GSP_FIRMWARE(ga104);
-NVKM_GSP_FIRMWARE(ga106);
-NVKM_GSP_FIRMWARE(ga107);
-
-NVKM_GSP_FIRMWARE(ad102);
-NVKM_GSP_FIRMWARE(ad103);
-NVKM_GSP_FIRMWARE(ad104);
-NVKM_GSP_FIRMWARE(ad106);
-NVKM_GSP_FIRMWARE(ad107);
+const struct nvkm_rm_api_gsp
+r535_gsp = {
+	.set_rmargs = r535_gsp_set_rmargs,
+	.set_system_info = r535_gsp_set_system_info,
+	.get_static_info = r535_gsp_get_static_info,
+	.xlat_mc_engine_idx = r535_gsp_xlat_mc_engine_idx,
+	.drop_send_user_shared_data = r535_gsp_drop_send_user_shared_data,
+	.sr_data_size = r535_gsp_sr_data_size,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/ad102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvdec.c
index 1b4619ff9e8e..a8c42ec0367b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/ad102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvdec.c
@@ -19,26 +19,27 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "priv.h"
+#include <rm/engine.h>
 
-#include <subdev/gsp.h>
+#include "nvrm/nvdec.h"
 
-#include <nvif/class.h>
+static int
+r535_nvdec_alloc(struct nvkm_gsp_object *chan, u32 handle, u32 class, int inst,
+		 struct nvkm_gsp_object *nvdec)
+{
+	NV_BSP_ALLOCATION_PARAMETERS *args;
 
-static const struct nvkm_engine_func
-ad102_nvenc = {
-	.sclass = {
-		{ -1, -1, NVC9B7_VIDEO_ENCODER },
-		{}
-	}
-};
+	args = nvkm_gsp_rm_alloc_get(chan, handle, class, sizeof(*args), nvdec);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
 
-int
-ad102_nvenc_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-		struct nvkm_nvenc **pnvenc)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvenc_new(&ad102_nvenc, device, type, inst, pnvenc);
+	args->size = sizeof(*args);
+	args->engineInstance = inst;
 
-	return -ENODEV;
+	return nvkm_gsp_rm_alloc_wr(nvdec, args);
 }
+
+const struct nvkm_rm_api_engine
+r535_nvdec = {
+	.alloc = r535_nvdec_alloc,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ad102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvenc.c
index 7bfa6240d283..acb3ce8bb9de 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ad102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvenc.c
@@ -19,28 +19,27 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "gf100.h"
+#include <rm/engine.h>
 
-#include <subdev/gsp.h>
+#include "nvrm/nvenc.h"
 
-#include <nvif/class.h>
+static int
+r535_nvenc_alloc(struct nvkm_gsp_object *chan, u32 handle, u32 class, int inst,
+		 struct nvkm_gsp_object *nvenc)
+{
+	NV_MSENC_ALLOCATION_PARAMETERS *args;
 
-static const struct gf100_gr_func
-ad102_gr = {
-	.sclass = {
-		{ -1, -1, FERMI_TWOD_A },
-		{ -1, -1, KEPLER_INLINE_TO_MEMORY_B },
-		{ -1, -1, ADA_A },
-		{ -1, -1, ADA_COMPUTE_A },
-		{}
-	}
-};
+	args = nvkm_gsp_rm_alloc_get(chan, handle, class, sizeof(*args), nvenc);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
 
-int
-ad102_gr_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst, struct nvkm_gr **pgr)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_gr_new(&ad102_gr, device, type, inst, pgr);
+	args->size = sizeof(*args);
+	args->engineInstance = inst;
 
-	return -ENODEV;
+	return nvkm_gsp_rm_alloc_wr(nvenc, args);
 }
+
+const struct nvkm_rm_api_engine
+r535_nvenc = {
+	.alloc = r535_nvenc_alloc,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvjpg.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvjpg.c
new file mode 100644
index 000000000000..fbc4080ad8d8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvjpg.c
@@ -0,0 +1,45 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/engine.h>
+
+#include "nvrm/nvjpg.h"
+
+static int
+r535_nvjpg_alloc(struct nvkm_gsp_object *chan, u32 handle, u32 class, int inst,
+		 struct nvkm_gsp_object *nvjpg)
+{
+	NV_NVJPG_ALLOCATION_PARAMETERS *args;
+
+	args = nvkm_gsp_rm_alloc_get(chan, handle, class, sizeof(*args), nvjpg);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
+
+	args->size = sizeof(*args);
+	args->engineInstance = inst;
+
+	return nvkm_gsp_rm_alloc_wr(nvjpg, args);
+}
+
+const struct nvkm_rm_api_engine
+r535_nvjpg = {
+	.alloc = r535_nvjpg_alloc,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/alloc.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/alloc.h
new file mode 100644
index 000000000000..cbc7e611fbda
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/alloc.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_ALLOC_H__
+#define __NVRM_ALLOC_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct rpc_gsp_rm_alloc_v03_00
+{
+    NvHandle   hClient;
+    NvHandle   hParent;
+    NvHandle   hObject;
+    NvU32      hClass;
+    NvU32      status;
+    NvU32      paramsSize;
+    NvU32      flags;
+    NvU8       reserved[4];
+    NvU8       params[];
+} rpc_gsp_rm_alloc_v03_00;
+
+typedef struct NVOS00_PARAMETERS_v03_00
+{
+    NvHandle   hRoot;
+    NvHandle   hObjectParent;
+    NvHandle   hObjectOld;
+    NvV32      status;
+} NVOS00_PARAMETERS_v03_00;
+
+typedef struct rpc_free_v03_00
+{
+    NVOS00_PARAMETERS_v03_00 params;
+} rpc_free_v03_00;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/bar.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/bar.h
new file mode 100644
index 000000000000..60b0b08491ee
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/bar.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_BAR_H__
+#define __NVRM_BAR_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef enum
+{
+    NV_RPC_UPDATE_PDE_BAR_1,
+    NV_RPC_UPDATE_PDE_BAR_2,
+    NV_RPC_UPDATE_PDE_BAR_INVALID,
+} NV_RPC_UPDATE_PDE_BAR_TYPE;
+
+typedef struct UpdateBarPde_v15_00
+{
+    NV_RPC_UPDATE_PDE_BAR_TYPE barType;
+    NvU64      entryValue NV_ALIGN_BYTES(8);
+    NvU64      entryLevelShift NV_ALIGN_BYTES(8);
+} UpdateBarPde_v15_00;
+
+typedef struct rpc_update_bar_pde_v15_00
+{
+    UpdateBarPde_v15_00 info;
+} rpc_update_bar_pde_v15_00;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ce.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ce.h
new file mode 100644
index 000000000000..90b0325203d2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ce.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_CE_H__
+#define __NVRM_CE_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct NVC0B5_ALLOCATION_PARAMETERS {
+    NvU32 version;
+    NvU32 engineType;
+} NVC0B5_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/client.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/client.h
new file mode 100644
index 000000000000..df0e63c0cb6b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/client.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_CLIENT_H__
+#define __NVRM_CLIENT_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV01_ROOT        (0x0U) /* finn: Evaluated from "NV0000_ALLOC_PARAMETERS_MESSAGE_ID" */
+
+#define NV_PROC_NAME_MAX_LENGTH 100U
+
+typedef struct NV0000_ALLOC_PARAMETERS {
+    NvHandle hClient; /* CORERM-2934: hClient must remain the first member until all allocations use these params */
+    NvU32    processID;
+    char     processName[NV_PROC_NAME_MAX_LENGTH];
+} NV0000_ALLOC_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ctrl.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ctrl.h
new file mode 100644
index 000000000000..77f10acd82c9
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ctrl.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_CTRL_H__
+#define __NVRM_CTRL_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct rpc_gsp_rm_control_v03_00
+{
+    NvHandle   hClient;
+    NvHandle   hObject;
+    NvU32      cmd;
+    NvU32      status;
+    NvU32      paramsSize;
+    NvU32      flags;
+    NvU8       params[];
+} rpc_gsp_rm_control_v03_00;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/device.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/device.h
new file mode 100644
index 000000000000..3933b9ad61ce
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/device.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_DEVICE_H__
+#define __NVRM_DEVICE_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV01_DEVICE_0      (0x80U) /* finn: Evaluated from "NV0080_ALLOC_PARAMETERS_MESSAGE_ID" */
+
+typedef struct NV0080_ALLOC_PARAMETERS {
+    NvU32    deviceId;
+    NvHandle hClientShare;
+    NvHandle hTargetClient;
+    NvHandle hTargetDevice;
+    NvV32    flags;
+    NV_DECLARE_ALIGNED(NvU64 vaSpaceSize, 8);
+    NV_DECLARE_ALIGNED(NvU64 vaStartInternal, 8);
+    NV_DECLARE_ALIGNED(NvU64 vaLimitInternal, 8);
+    NvV32    vaMode;
+} NV0080_ALLOC_PARAMETERS;
+
+#define NV20_SUBDEVICE_0      (0x2080U) /* finn: Evaluated from "NV2080_ALLOC_PARAMETERS_MESSAGE_ID" */
+
+typedef struct NV2080_ALLOC_PARAMETERS {
+    NvU32 subDeviceId;
+} NV2080_ALLOC_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/disp.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/disp.h
new file mode 100644
index 000000000000..7b7539639540
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/disp.h
@@ -0,0 +1,741 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_DISP_H__
+#define __NVRM_DISP_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_WRITE_INST_MEM (0x20800a49) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS {
+    NV_DECLARE_ALIGNED(NvU64 instMemPhysAddr, 8);
+    NV_DECLARE_ALIGNED(NvU64 instMemSize, 8);
+    NvU32 instMemAddrSpace;
+    NvU32 instMemCpuCacheAttr;
+} NV2080_CTRL_INTERNAL_DISPLAY_WRITE_INST_MEM_PARAMS;
+
+#define NV_MEMORY_WRITECOMBINED      2
+
+#define NV04_DISPLAY_COMMON                      (0x00000073)
+
+#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_GET_STATIC_INFO (0x20800a01) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS {
+    NvU32  feHwSysCap;
+    NvU32  windowPresentMask;
+    NvBool bFbRemapperEnabled;
+    NvU32  numHeads;
+    NvBool bPrimaryVga;
+    NvU32  i2cPort;
+    NvU32  internalDispActiveMask;
+} NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS;
+
+#define NV2080_CTRL_ACPI_DSM_READ_SIZE                   (0x1000) /* finn: Evaluated from "(4 * 1024)" */
+
+#define NV2080_CTRL_CMD_INTERNAL_INIT_BRIGHTC_STATE_LOAD (0x20800ac6) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_INIT_BRIGHTC_STATE_LOAD_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_INIT_BRIGHTC_STATE_LOAD_PARAMS {
+    NvU32 status;
+    NvU16 backLightDataSize;
+    NvU8  backLightData[NV2080_CTRL_ACPI_DSM_READ_SIZE];
+} NV2080_CTRL_INTERNAL_INIT_BRIGHTC_STATE_LOAD_PARAMS;
+
+typedef struct NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT_PARAMS {
+    NvU32 subDeviceInstance;
+} NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT                    (0x731365U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_SET_MANUAL_DISPLAYPORT_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_NUM_HEADS (0x730102U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_NUM_HEADS_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_NUM_HEADS_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 flags;
+    NvU32 numHeads;
+} NV0073_CTRL_SYSTEM_GET_NUM_HEADS_PARAMS;
+
+#define NV0073_CTRL_CMD_SPECIFIC_GET_ALL_HEAD_MASK (0x730287U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_ALL_HEAD_MASK_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_GET_ALL_HEAD_MASK_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 headMask;
+} NV0073_CTRL_SPECIFIC_GET_ALL_HEAD_MASK_PARAMS;
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_SUPPORTED (0x730120U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayMask;
+    NvU32 displayMaskDDC;
+} NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS;
+
+#define NV0073_CTRL_MAX_CONNECTORS                    4U
+
+#define NV0073_CTRL_CMD_SPECIFIC_GET_CONNECTOR_DATA   (0x730250U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 flags;
+    NvU32 DDCPartners;
+    NvU32 count;
+    struct {
+        NvU32 index;
+        NvU32 type;
+        NvU32 location;
+    } data[NV0073_CTRL_MAX_CONNECTORS];
+    NvU32 platform;
+} NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS;
+
+#define NV0073_CTRL_CMD_SPECIFIC_OR_GET_INFO (0x73028bU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvU32  index;
+    NvU32  type;
+    NvU32  protocol;
+    NvU32  ditherType;
+    NvU32  ditherAlgo;
+    NvU32  location;
+    NvU32  rootPortId;
+    NvU32  dcbIndex;
+    NV_DECLARE_ALIGNED(NvU64 vbiosAddress, 8);
+    NvBool bIsLitByVbios;
+    NvBool bIsDispDynamic;
+} NV0073_CTRL_SPECIFIC_OR_GET_INFO_PARAMS;
+
+#define NV0073_CTRL_SPECIFIC_OR_TYPE_NONE                   (0x00000000U)
+#define NV0073_CTRL_SPECIFIC_OR_TYPE_DAC                    (0x00000001U)
+#define NV0073_CTRL_SPECIFIC_OR_TYPE_SOR                    (0x00000002U)
+#define NV0073_CTRL_SPECIFIC_OR_TYPE_PIOR                   (0x00000003U)
+#define NV0073_CTRL_SPECIFIC_OR_TYPE_DSI                    (0x00000005U)
+
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_DAC_RGB_CRT        (0x00000000U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_LVDS_CUSTOM    (0x00000000U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_SINGLE_TMDS_A  (0x00000001U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_SINGLE_TMDS_B  (0x00000002U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DUAL_TMDS      (0x00000005U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DP_A           (0x00000008U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DP_B           (0x00000009U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_SOR_DSI            (0x00000010U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_DSI                (0x00000011U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_PIOR_EXT_TMDS_ENC  (0x00000000U)
+#define NV0073_CTRL_SPECIFIC_OR_PROTOCOL_UNKNOWN            (0xFFFFFFFFU)
+
+#define NV0073_CTRL_CMD_DP_GET_CAPS   (0x731369U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_CMD_DSC_CAP_PARAMS {
+    NvBool bDscSupported;
+    NvU32  encoderColorFormatMask;
+    NvU32  lineBufferSizeKB;
+    NvU32  rateBufferSizeKB;
+    NvU32  bitsPerPixelPrecision;
+    NvU32  maxNumHztSlices;
+    NvU32  lineBufferBitDepth;
+} NV0073_CTRL_CMD_DSC_CAP_PARAMS;
+
+typedef struct NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS {
+    NvU32                          subDeviceInstance;
+    NvU32                          sorIndex;
+    NvU32                          maxLinkRate;
+    NvU32                          dpVersionsSupported;
+    NvU32                          UHBRSupported;
+    NvBool                         bIsMultistreamSupported;
+    NvBool                         bIsSCEnabled;
+    NvBool                         bHasIncreasedWatermarkLimits;
+    NvBool                         bIsPC2Disabled;
+    NvBool                         isSingleHeadMSTSupported;
+    NvBool                         bFECSupported;
+    NvBool                         bIsTrainPhyRepeater;
+    NvBool                         bOverrideLinkBw;
+    NV0073_CTRL_CMD_DSC_CAP_PARAMS DSC;
+} NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_GET_CAPS   (0x731369U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID" */
+#define NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID (0x69U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2                0:0
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2_NO              (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2_YES             (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4                1:1
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4_NO              (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4_YES             (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE                           2:0
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_NONE                          (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_1_62                          (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_2_70                          (0x00000002U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_5_40                          (0x00000003U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_8_10                          (0x00000004U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_RGB                (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_444        (0x00000002U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_NATIVE_422 (0x00000004U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_NATIVE_420 (0x00000008U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_16           (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_8            (0x00000002U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_4            (0x00000003U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_2            (0x00000004U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1              (0x00000005U)
+
+#define NV2080_NOTIFIERS_HOTPLUG                                   (1)
+
+typedef struct {
+    NvU32 plugDisplayMask;
+    NvU32 unplugDisplayMask;
+} Nv2080HotplugNotification;
+
+#define NV2080_NOTIFIERS_DP_IRQ                                    (7)
+
+typedef struct Nv2080DpIrqNotificationRec {
+    NvU32 displayId;
+} Nv2080DpIrqNotification;
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_CONNECT_STATE (0x730122U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 flags;
+    NvU32 displayMask;
+    NvU32 retryTimeMs;
+} NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS;
+
+#define NV0073_CTRL_CMD_DFP_GET_INFO (0x731140U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_GET_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DFP_GET_INFO_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 flags;
+    NvU32 flags2;
+} NV0073_CTRL_DFP_GET_INFO_PARAMS;
+
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL                                       2:0
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_TMDS                       (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_LVDS                       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_SDI                        (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_DISPLAYPORT                (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_DSI                        (0x00000004U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_WRBK                       (0x00000005U)
+#define NV0073_CTRL_DFP_FLAGS_LANE                                         5:3
+#define NV0073_CTRL_DFP_FLAGS_LANE_NONE                         (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_SINGLE                       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_DUAL                         (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_QUAD                         (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_OCT                          (0x00000004U)
+#define NV0073_CTRL_DFP_FLAGS_LIMIT                                        6:6
+#define NV0073_CTRL_DFP_FLAGS_LIMIT_DISABLE                     (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_LIMIT_60HZ_RR                     (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER                                   7:7
+#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER_NORMAL                 (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER_DISABLE                (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE                                 8:8
+#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE_FALSE                (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE_TRUE                 (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE                        9:9
+#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE_FALSE       (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE_TRUE        (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE                         10:10
+#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE_FALSE          (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE_TRUE           (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE                    11:11
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE_FALSE     (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE_TRUE      (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE                    12:12
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE_FALSE     (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE_TRUE      (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED                               14:14
+#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED_FALSE                (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED_TRUE                 (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT                       15:15
+#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT_FALSE        (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT_TRUE         (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT                         16:16
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT_NONE           (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT_PREFER_RBR     (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW                                 19:17
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_1_62GBPS               (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_2_70GBPS               (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_5_40GBPS               (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_8_10GBPS               (0x00000004U)
+#define NV0073_CTRL_DFP_FLAGS_LINK                                       21:20
+#define NV0073_CTRL_DFP_FLAGS_LINK_NONE                         (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_LINK_SINGLE                       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_LINK_DUAL                         (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID                           22:22
+#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID_FALSE            (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID_TRUE             (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID                              24:23
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_NONE            (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_A               (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_B               (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_GANGED          (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED                   25:25
+#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED_FALSE    (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED_TRUE     (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_PHY_REPEATER_COUNT                      29:26
+#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE                        30:30
+#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE_FALSE         (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE_TRUE          (0x00000001U)
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_ACTIVE                (0x730126U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 head;
+    NvU32 flags;
+    NvU32 displayId;
+} NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS;
+
+typedef NvU32 NV0073_CTRL_DFP_ASSIGN_SOR_LINKCONFIG;
+
+typedef struct NV0073_CTRL_DFP_ASSIGN_SOR_INFO {
+    NvU32 displayMask;
+    NvU32 sorType;
+} NV0073_CTRL_DFP_ASSIGN_SOR_INFO;
+
+#define NV0073_CTRL_CMD_DFP_ASSIGN_SOR_MAX_SORS  4U
+
+#define NV0073_CTRL_CMD_DFP_ASSIGN_SOR           (0x731152U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_ASSIGN_SOR_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DFP_ASSIGN_SOR_PARAMS {
+    NvU32                                 subDeviceInstance;
+    NvU32                                 displayId;
+    NvU8                                  sorExcludeMask;
+    NvU32                                 slaveDisplayId;
+    NV0073_CTRL_DFP_ASSIGN_SOR_LINKCONFIG forceSublinkConfig;
+    NvBool                                bIs2Head1Or;
+    NvU32                                 sorAssignList[NV0073_CTRL_CMD_DFP_ASSIGN_SOR_MAX_SORS];
+    NV0073_CTRL_DFP_ASSIGN_SOR_INFO       sorAssignListWithTag[NV0073_CTRL_CMD_DFP_ASSIGN_SOR_MAX_SORS];
+    NvU8                                  reservedSorMask;
+    NvU32                                 flags;
+} NV0073_CTRL_DFP_ASSIGN_SOR_PARAMS;
+
+#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_AUDIO                                      0:0
+#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_AUDIO_OPTIMAL                    (0x00000001U)
+#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_AUDIO_DEFAULT                    (0x00000000U)
+#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_ACTIVE_SOR_NOT_AUDIO_CAPABLE               1:1
+#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_ACTIVE_SOR_NOT_AUDIO_CAPABLE_NO  (0x00000000U)
+#define NV0073_CTRL_DFP_ASSIGN_SOR_FLAGS_ACTIVE_SOR_NOT_AUDIO_CAPABLE_YES (0x00000001U)
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS (0x730292U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_CMD_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS (0x730291U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvU32  brightness;
+    NvBool bUncalibrated;
+} NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS;
+
+#define NV0073_CTRL_CMD_DFP_SET_ELD_AUDIO_CAPS                         (0x731144U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_SET_ELD_AUDIO_CAP_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_ELD_BUFFER                      96U
+
+typedef struct NV0073_CTRL_DFP_SET_ELD_AUDIO_CAP_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 numELDSize;
+    NvU8  bufferELD[NV0073_CTRL_DFP_ELD_AUDIO_CAPS_ELD_BUFFER];
+    NvU32 maxFreqSupported;
+    NvU32 ctrl;
+    NvU32 deviceEntry;
+} NV0073_CTRL_DFP_SET_ELD_AUDIO_CAP_PARAMS;
+
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_PD                                     0:0
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_PD_FALSE              (0x00000000U)
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_PD_TRUE               (0x00000001U)
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_ELDV                                   1:1
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_ELDV_FALSE            (0x00000000U)
+#define NV0073_CTRL_DFP_ELD_AUDIO_CAPS_CTRL_ELDV_TRUE             (0x00000001U)
+
+#define NV0073_CTRL_SPECIFIC_GET_EDID_MAX_EDID_BYTES 2048U
+
+#define NV0073_CTRL_CMD_SPECIFIC_GET_EDID_V2         (0x730245U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 bufferSize;
+    NvU32 flags;
+    NvU8  edidBuffer[NV0073_CTRL_SPECIFIC_GET_EDID_MAX_EDID_BYTES];
+} NV0073_CTRL_SPECIFIC_GET_EDID_V2_PARAMS;
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_ENABLE (0x730273U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_HDMI_ENABLE_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_SET_HDMI_ENABLE_PARAMS {
+    NvU8  subDeviceInstance;
+    NvU32 displayId;
+    NvU8  enable;
+} NV0073_CTRL_SPECIFIC_SET_HDMI_ENABLE_PARAMS;
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS (0x730293U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 caps;
+} NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS;
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS (0x730293U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_HDMI_SINK_CAPS_PARAMS_MESSAGE_ID" */
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_GT_340MHZ_CLOCK_SUPPORTED                           0:0
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_GT_340MHZ_CLOCK_SUPPORTED_FALSE       (0x00000000U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_GT_340MHZ_CLOCK_SUPPORTED_TRUE        (0x00000001U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_LTE_340MHZ_SCRAMBLING_SUPPORTED                     1:1
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_LTE_340MHZ_SCRAMBLING_SUPPORTED_FALSE (0x00000000U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_LTE_340MHZ_SCRAMBLING_SUPPORTED_TRUE  (0x00000001U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_SCDC_SUPPORTED                                      2:2
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_SCDC_SUPPORTED_FALSE                  (0x00000000U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_SCDC_SUPPORTED_TRUE                   (0x00000001U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED                              5:3
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_NONE           (0x00000000U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_3LANES_3G      (0x00000001U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_3LANES_6G      (0x00000002U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_6G      (0x00000003U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_8G      (0x00000004U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_10G     (0x00000005U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_MAX_FRL_RATE_SUPPORTED_4LANES_12G     (0x00000006U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_12_SUPPORTED                                    6:6
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_12_SUPPORTED_FALSE                (0x00000000U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_12_SUPPORTED_TRUE                 (0x00000001U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED                          9:7
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_NONE       (0x00000000U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_3LANES_3G  (0x00000001U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_3LANES_6G  (0x00000002U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_6G  (0x00000003U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_8G  (0x00000004U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_10G (0x00000005U)
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_SINK_CAPS_DSC_MAX_FRL_RATE_SUPPORTED_4LANES_12G (0x00000006U)
+
+#define NV0073_CTRL_SET_OD_MAX_PACKET_SIZE     36U
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_OD_PACKET (0x730288U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_OD_PACKET_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_SET_OD_PACKET_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvU32  transmitControl;
+    NvU32  packetSize;
+    NvU32  targetHead;
+    NvBool bUsePsrHeadforSdp;
+    NvU8   aPacket[NV0073_CTRL_SET_OD_MAX_PACKET_SIZE];
+} NV0073_CTRL_SPECIFIC_SET_OD_PACKET_PARAMS;
+
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ENABLE                                     0:0
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ENABLE_NO                      (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ENABLE_YES                     (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_OTHER_FRAME                                1:1
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_OTHER_FRAME_DISABLE            (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_OTHER_FRAME_ENABLE             (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SINGLE_FRAME                               2:2
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SINGLE_FRAME_DISABLE           (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SINGLE_FRAME_ENABLE            (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ON_HBLANK                                 3:3
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ON_HBLANK_DISABLE              (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_ON_HBLANK_ENABLE               (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_IMMEDIATE                                  4:4
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_IMMEDIATE_DISABLE              (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_IMMEDIATE_ENABLE               (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_VIDEO_FMT                                  5:5
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_VIDEO_FMT_SW_CONTROLLED        (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_VIDEO_FMT_HW_CONTROLLED        (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_STEREO_POLARITY                        6:6
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_STEREO_POLARITY_FALSE      (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_STEREO_POLARITY_TRUE       (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_SELF_REFRESH_SETTING                   7:7
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_SELF_REFRESH_SETTING_FALSE (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_SET_SELF_REFRESH_SETTING_TRUE  (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_GEN_INFOFRAME_MODE                         9:8
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_GEN_INFOFRAME_MODE_INFOFRAME0  (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_GEN_INFOFRAME_MODE_INFOFRAME1  (0x0000001U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_RESERVED_LEGACY_MODE                     31:31
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_RESERVED_LEGACY_MODE_NO        (0x0000000U)
+#define NV0073_CTRL_SPECIFIC_SET_OD_PACKET_TRANSMIT_CONTROL_RESERVED_LEGACY_MODE_YES       (0x0000001U)
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM (0x730275U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM_PARAMS {
+    NvU8  subDeviceInstance;
+    NvU32 displayId;
+    NvU8  mute;
+} NV0073_CTRL_CMD_SPECIFIC_SET_HDMI_AUDIO_MUTESTREAM_PARAMS;
+
+#define NV0073_CTRL_DP_AUXCH_MAX_DATA_SIZE 16U
+
+#define NV0073_CTRL_CMD_DP_AUXCH_CTRL      (0x731341U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_AUXCH_CTRL_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DP_AUXCH_CTRL_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvBool bAddrOnly;
+    NvU32  cmd;
+    NvU32  addr;
+    NvU8   data[NV0073_CTRL_DP_AUXCH_MAX_DATA_SIZE];
+    NvU32  size;
+    NvU32  replyType;
+    NvU32  retryTimeMs;
+} NV0073_CTRL_DP_AUXCH_CTRL_PARAMS;
+
+#define NV0073_CTRL_DP_AUXCH_CMD_TYPE                          3:3
+#define NV0073_CTRL_DP_AUXCH_CMD_TYPE_I2C               (0x00000000U)
+#define NV0073_CTRL_DP_AUXCH_CMD_TYPE_AUX               (0x00000001U)
+#define NV0073_CTRL_DP_AUXCH_CMD_I2C_MOT                       2:2
+#define NV0073_CTRL_DP_AUXCH_CMD_I2C_MOT_FALSE          (0x00000000U)
+#define NV0073_CTRL_DP_AUXCH_CMD_I2C_MOT_TRUE           (0x00000001U)
+#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE                      1:0
+#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE_WRITE         (0x00000000U)
+#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE_READ          (0x00000001U)
+#define NV0073_CTRL_DP_AUXCH_CMD_REQ_TYPE_WRITE_STATUS  (0x00000002U)
+
+#define NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES (0x731377U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES        8U
+
+typedef struct NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS {
+    // In
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU16 linkRateTbl[NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES];
+
+    // Out
+    NvU8  linkBwTbl[NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES];
+    NvU8  linkBwCount;
+} NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_CTRL                     (0x731343U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_CTRL_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DP_CTRL_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 cmd;
+    NvU32 data;
+    NvU32 err;
+    NvU32 retryTimeMs;
+    NvU32 eightLaneDpcdBaseAddr;
+} NV0073_CTRL_DP_CTRL_PARAMS;
+
+#define NV0073_CTRL_DP_CMD_SET_LANE_COUNT                           0:0
+#define NV0073_CTRL_DP_CMD_SET_LANE_COUNT_FALSE                         (0x00000000U)
+#define NV0073_CTRL_DP_CMD_SET_LANE_COUNT_TRUE                          (0x00000001U)
+#define NV0073_CTRL_DP_CMD_SET_LINK_BW                              1:1
+#define NV0073_CTRL_DP_CMD_SET_LINK_BW_FALSE                            (0x00000000U)
+#define NV0073_CTRL_DP_CMD_SET_LINK_BW_TRUE                             (0x00000001U)
+#define NV0073_CTRL_DP_CMD_DISABLE_DOWNSPREAD                       2:2
+#define NV0073_CTRL_DP_CMD_DISABLE_DOWNSPREAD_FALSE                     (0x00000000U)
+#define NV0073_CTRL_DP_CMD_DISABLE_DOWNSPREAD_TRUE                      (0x00000001U)
+#define NV0073_CTRL_DP_CMD_UNUSED                                   3:3
+#define NV0073_CTRL_DP_CMD_SET_FORMAT_MODE                          4:4
+#define NV0073_CTRL_DP_CMD_SET_FORMAT_MODE_SINGLE_STREAM                (0x00000000U)
+#define NV0073_CTRL_DP_CMD_SET_FORMAT_MODE_MULTI_STREAM                 (0x00000001U)
+#define NV0073_CTRL_DP_CMD_FAST_LINK_TRAINING                       5:5
+#define NV0073_CTRL_DP_CMD_FAST_LINK_TRAINING_NO                        (0x00000000U)
+#define NV0073_CTRL_DP_CMD_FAST_LINK_TRAINING_YES                       (0x00000001U)
+#define NV0073_CTRL_DP_CMD_NO_LINK_TRAINING                         6:6
+#define NV0073_CTRL_DP_CMD_NO_LINK_TRAINING_NO                          (0x00000000U)
+#define NV0073_CTRL_DP_CMD_NO_LINK_TRAINING_YES                         (0x00000001U)
+#define NV0073_CTRL_DP_CMD_SET_ENHANCED_FRAMING                     7:7
+#define NV0073_CTRL_DP_CMD_SET_ENHANCED_FRAMING_FALSE                   (0x00000000U)
+#define NV0073_CTRL_DP_CMD_SET_ENHANCED_FRAMING_TRUE                    (0x00000001U)
+#define NV0073_CTRL_DP_CMD_USE_DOWNSPREAD_SETTING                   8:8
+#define NV0073_CTRL_DP_CMD_USE_DOWNSPREAD_SETTING_DEFAULT               (0x00000000U)
+#define NV0073_CTRL_DP_CMD_USE_DOWNSPREAD_SETTING_FORCE                 (0x00000001U)
+#define NV0073_CTRL_DP_CMD_SKIP_HW_PROGRAMMING                      9:9
+#define NV0073_CTRL_DP_CMD_SKIP_HW_PROGRAMMING_NO                       (0x00000000U)
+#define NV0073_CTRL_DP_CMD_SKIP_HW_PROGRAMMING_YES                      (0x00000001U)
+#define NV0073_CTRL_DP_CMD_POST_LT_ADJ_REQ_GRANTED                10:10
+#define NV0073_CTRL_DP_CMD_POST_LT_ADJ_REQ_GRANTED_NO                   (0x00000000U)
+#define NV0073_CTRL_DP_CMD_POST_LT_ADJ_REQ_GRANTED_YES                  (0x00000001U)
+#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING                     12:11
+#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING_NO                        (0x00000000U)
+#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING_DONOT_TOGGLE_TRANSMISSION (0x00000001U)
+#define NV0073_CTRL_DP_CMD_FAKE_LINK_TRAINING_TOGGLE_TRANSMISSION_ON    (0x00000002U)
+#define NV0073_CTRL_DP_CMD_TRAIN_PHY_REPEATER                     13:13
+#define NV0073_CTRL_DP_CMD_TRAIN_PHY_REPEATER_NO                        (0x00000000U)
+#define NV0073_CTRL_DP_CMD_TRAIN_PHY_REPEATER_YES                       (0x00000001U)
+#define NV0073_CTRL_DP_CMD_FALLBACK_CONFIG                        14:14
+#define NV0073_CTRL_DP_CMD_FALLBACK_CONFIG_FALSE                        (0x00000000U)
+#define NV0073_CTRL_DP_CMD_FALLBACK_CONFIG_TRUE                         (0x00000001U)
+#define NV0073_CTRL_DP_CMD_ENABLE_FEC                             15:15
+#define NV0073_CTRL_DP_CMD_ENABLE_FEC_FALSE                             (0x00000000U)
+#define NV0073_CTRL_DP_CMD_ENABLE_FEC_TRUE                              (0x00000001U)
+#define NV0073_CTRL_DP_CMD_BANDWIDTH_TEST                         29:29
+#define NV0073_CTRL_DP_CMD_BANDWIDTH_TEST_NO                            (0x00000000U)
+#define NV0073_CTRL_DP_CMD_BANDWIDTH_TEST_YES                           (0x00000001U)
+#define NV0073_CTRL_DP_CMD_LINK_CONFIG_CHECK_DISABLE              30:30
+#define NV0073_CTRL_DP_CMD_LINK_CONFIG_CHECK_DISABLE_FALSE              (0x00000000U)
+#define NV0073_CTRL_DP_CMD_LINK_CONFIG_CHECK_DISABLE_TRUE               (0x00000001U)
+#define NV0073_CTRL_DP_CMD_DISABLE_LINK_CONFIG                    31:31
+#define NV0073_CTRL_DP_CMD_DISABLE_LINK_CONFIG_FALSE                    (0x00000000U)
+#define NV0073_CTRL_DP_CMD_DISABLE_LINK_CONFIG_TRUE                     (0x00000001U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE                                   3:0
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_MONITOR_ENABLE_BEGIN     (0x00000000U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_MONITOR_ENABLE_CHALLENGE (0x00000001U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_MONITOR_ENABLE_CHECK     (0x00000002U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_DRIVER_ENABLE_BEGIN      (0x00000003U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_DRIVER_ENABLE_CHALLENGE  (0x00000004U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_DRIVER_ENABLE_CHECK      (0x00000005U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_RESET_MONITOR            (0x00000006U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_INIT_PUBLIC_INFO         (0x00000007U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_GET_PUBLIC_INFO          (0x00000008U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_CMD_STAGE_STATUS_CHECK             (0x00000009U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_OK                          (0x00000000U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_PENDING                     (0x80000001U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_READ_ERROR                  (0x80000002U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_WRITE_ERROR                 (0x80000003U)
+#define NV0073_CTRL_DP_CMD_ENABLE_VRR_STATUS_DEVICE_ERROR                (0x80000004U)
+
+#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT                          4:0
+#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_0                            (0x00000000U)
+#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_1                            (0x00000001U)
+#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_2                            (0x00000002U)
+#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_4                            (0x00000004U)
+#define NV0073_CTRL_DP_DATA_SET_LANE_COUNT_8                            (0x00000008U)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW                            15:8
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_1_62GBPS                        (0x00000006U)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_2_16GBPS                        (0x00000008U)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_2_43GBPS                        (0x00000009U)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_2_70GBPS                        (0x0000000AU)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_3_24GBPS                        (0x0000000CU)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_4_32GBPS                        (0x00000010U)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_5_40GBPS                        (0x00000014U)
+#define NV0073_CTRL_DP_DATA_SET_LINK_BW_8_10GBPS                        (0x0000001EU)
+#define NV0073_CTRL_DP_DATA_SET_ENHANCED_FRAMING                  18:18
+#define NV0073_CTRL_DP_DATA_SET_ENHANCED_FRAMING_NO                     (0x00000000U)
+#define NV0073_CTRL_DP_DATA_SET_ENHANCED_FRAMING_YES                    (0x00000001U)
+#define NV0073_CTRL_DP_DATA_TARGET                                22:19
+#define NV0073_CTRL_DP_DATA_TARGET_SINK                                 (0x00000000U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_0                       (0x00000001U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_1                       (0x00000002U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_2                       (0x00000003U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_3                       (0x00000004U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_4                       (0x00000005U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_5                       (0x00000006U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_6                       (0x00000007U)
+#define NV0073_CTRL_DP_DATA_TARGET_PHY_REPEATER_7                       (0x00000008U)
+
+#define NV0073_CTRL_CMD_DP_SET_LANE_DATA (0x731346U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_SET_LANE_DATA_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_MAX_LANES                                           8U
+
+typedef struct NV0073_CTRL_DP_LANE_DATA_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 numLanes;
+    NvU32 data[NV0073_CTRL_MAX_LANES];
+} NV0073_CTRL_DP_LANE_DATA_PARAMS;
+
+#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS                   1:0
+#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_NONE    (0x00000000U)
+#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_LEVEL1  (0x00000001U)
+#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_LEVEL2  (0x00000002U)
+#define NV0073_CTRL_DP_LANE_DATA_PREEMPHASIS_LEVEL3  (0x00000003U)
+#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT                  3:2
+#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL0 (0x00000000U)
+#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL1 (0x00000001U)
+#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL2 (0x00000002U)
+#define NV0073_CTRL_DP_LANE_DATA_DRIVECURRENT_LEVEL3 (0x00000003U)
+
+#define NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID  (0x73135bU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvU32  preferredDisplayId;
+
+    NvBool force;
+    NvBool useBFM;
+
+    NvU32  displayIdAssigned;
+    NvU32  allDisplayMask;
+} NV0073_CTRL_CMD_DP_TOPOLOGY_ALLOCATE_DISPLAYID_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID (0x73135cU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+} NV0073_CTRL_CMD_DP_TOPOLOGY_FREE_DISPLAYID_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_CONFIG_STREAM                   (0x731362U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  head;
+    NvU32  sorIndex;
+    NvU32  dpLink;
+
+    NvBool bEnableOverride;
+    NvBool bMST;
+    NvU32  singleHeadMultistreamMode;
+    NvU32  hBlankSym;
+    NvU32  vBlankSym;
+    NvU32  colorFormat;
+    NvBool bEnableTwoHeadOneOr;
+
+    struct {
+        NvU32  slotStart;
+        NvU32  slotEnd;
+        NvU32  PBN;
+        NvU32  Timeslice;
+        NvBool sendACT;          // deprecated -Use NV0073_CTRL_CMD_DP_SEND_ACT
+        NvU32  singleHeadMSTPipeline;
+        NvBool bEnableAudioOverRightPanel;
+    } MST;
+
+    struct {
+        NvBool bEnhancedFraming;
+        NvU32  tuSize;
+        NvU32  waterMark;
+        NvU32  actualPclkHz;     // deprecated  -Use MvidWarParams
+        NvU32  linkClkFreqHz;    // deprecated  -Use MvidWarParams
+        NvBool bEnableAudioOverRightPanel;
+        struct {
+            NvU32  activeCnt;
+            NvU32  activeFrac;
+            NvU32  activePolarity;
+            NvBool mvidWarEnabled;
+            struct {
+                NvU32 actualPclkHz;
+                NvU32 linkClkFreqHz;
+            } MvidWarParams;
+        } Legacy;
+    } SST;
+} NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS;
+
+#define NV0073_CTRL_CMD_DFP_SET_AUDIO_ENABLE                (0x731150U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_SET_AUDIO_ENABLE_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DFP_SET_AUDIO_ENABLE_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvBool enable;
+} NV0073_CTRL_DFP_SET_AUDIO_ENABLE_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_SET_AUDIO_MUTESTREAM      (0x731359U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 mute;
+} NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS;
+
+#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER (0x20800a58) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS {
+    NvU32  addressSpace;
+    NV_DECLARE_ALIGNED(NvU64 physicalAddr, 8);
+    NV_DECLARE_ALIGNED(NvU64 limit, 8);
+    NvU32  cacheSnoop;
+    NvU32  hclass;
+    NvU32  channelInstance;
+    NvBool valid;
+} NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS;
+
+#define ADDR_SYSMEM     (1)         // System memory (PCI)
+
+#define ADDR_FBMEM      2         // Frame buffer memory space
+
+typedef struct
+{
+    NvV32    channelInstance;            // One of the n channel instances of a given channel type.
+                                         // All PIO channels have two instances (one per head).
+    NvHandle hObjectNotify;              // ctx dma handle for an area (of type NvNotification defined in sdk/nvidia/inc/nvtypes.h) where RM can write errors.
+    NvP64    pControl NV_ALIGN_BYTES(8); // pControl gives virt addr of control region for PIO channel
+} NV50VAIO_CHANNELPIO_ALLOCATION_PARAMETERS;
+
+typedef struct
+{
+    NvV32    channelInstance;            // One of the n channel instances of a given channel type.
+                                         // Note that core channel has only one instance
+                                         // while all others have two (one per head).
+    NvHandle hObjectBuffer;              // ctx dma handle for DMA push buffer
+    NvHandle hObjectNotify;              // ctx dma handle for an area (of type NvNotification defined in sdk/nvidia/inc/nvtypes.h) where RM can write errors/notifications
+    NvU32    offset;                     // Initial offset for put/get, usually zero.
+    NvP64    pControl NV_ALIGN_BYTES(8); // pControl gives virt addr of UDISP GET/PUT regs
+
+    NvU32    flags;
+#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB                1:1
+#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB_YES            0x00000000
+#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB_NO             0x00000001
+
+} NV50VAIO_CHANNELDMA_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/engine.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/engine.h
new file mode 100644
index 000000000000..b26dfc8f8087
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/engine.h
@@ -0,0 +1,260 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_ENGINE_H__
+#define __NVRM_ENGINE_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define MC_ENGINE_IDX_NULL                          0 // This must be 0
+#define MC_ENGINE_IDX_TMR                           1
+#define MC_ENGINE_IDX_DISP                          2
+#define MC_ENGINE_IDX_FB                            3
+#define MC_ENGINE_IDX_FIFO                          4
+#define MC_ENGINE_IDX_VIDEO                         5
+#define MC_ENGINE_IDX_MD                            6
+#define MC_ENGINE_IDX_BUS                           7
+#define MC_ENGINE_IDX_PMGR                          8
+#define MC_ENGINE_IDX_VP2                           9
+#define MC_ENGINE_IDX_CIPHER                        10
+#define MC_ENGINE_IDX_BIF                           11
+#define MC_ENGINE_IDX_PPP                           12
+#define MC_ENGINE_IDX_PRIVRING                      13
+#define MC_ENGINE_IDX_PMU                           14
+#define MC_ENGINE_IDX_CE0                           15
+#define MC_ENGINE_IDX_CE1                           16
+#define MC_ENGINE_IDX_CE2                           17
+#define MC_ENGINE_IDX_CE3                           18
+#define MC_ENGINE_IDX_CE4                           19
+#define MC_ENGINE_IDX_CE5                           20
+#define MC_ENGINE_IDX_CE6                           21
+#define MC_ENGINE_IDX_CE7                           22
+#define MC_ENGINE_IDX_CE8                           23
+#define MC_ENGINE_IDX_CE9                           24
+#define MC_ENGINE_IDX_CE_MAX                        MC_ENGINE_IDX_CE9
+#define MC_ENGINE_IDX_VIC                           35
+#define MC_ENGINE_IDX_ISOHUB                        36
+#define MC_ENGINE_IDX_VGPU                          37
+#define MC_ENGINE_IDX_MSENC                         38
+#define MC_ENGINE_IDX_MSENC1                        39
+#define MC_ENGINE_IDX_MSENC2                        40
+#define MC_ENGINE_IDX_C2C                           41
+#define MC_ENGINE_IDX_LTC                           42
+#define MC_ENGINE_IDX_FBHUB                         43
+#define MC_ENGINE_IDX_HDACODEC                      44
+#define MC_ENGINE_IDX_GMMU                          45
+#define MC_ENGINE_IDX_SEC2                          46
+#define MC_ENGINE_IDX_FSP                           47
+#define MC_ENGINE_IDX_NVLINK                        48
+#define MC_ENGINE_IDX_GSP                           49
+#define MC_ENGINE_IDX_NVJPG                         50
+#define MC_ENGINE_IDX_NVJPEG                        MC_ENGINE_IDX_NVJPG
+#define MC_ENGINE_IDX_NVJPEG0                       MC_ENGINE_IDX_NVJPEG
+#define MC_ENGINE_IDX_NVJPEG1                       51
+#define MC_ENGINE_IDX_NVJPEG2                       52
+#define MC_ENGINE_IDX_NVJPEG3                       53
+#define MC_ENGINE_IDX_NVJPEG4                       54
+#define MC_ENGINE_IDX_NVJPEG5                       55
+#define MC_ENGINE_IDX_NVJPEG6                       56
+#define MC_ENGINE_IDX_NVJPEG7                       57
+#define MC_ENGINE_IDX_REPLAYABLE_FAULT              58
+#define MC_ENGINE_IDX_ACCESS_CNTR                   59
+#define MC_ENGINE_IDX_NON_REPLAYABLE_FAULT          60
+#define MC_ENGINE_IDX_REPLAYABLE_FAULT_ERROR        61
+#define MC_ENGINE_IDX_NON_REPLAYABLE_FAULT_ERROR    62
+#define MC_ENGINE_IDX_INFO_FAULT                    63
+#define MC_ENGINE_IDX_BSP                           64
+#define MC_ENGINE_IDX_NVDEC                         MC_ENGINE_IDX_BSP
+#define MC_ENGINE_IDX_NVDEC0                        MC_ENGINE_IDX_NVDEC
+#define MC_ENGINE_IDX_NVDEC1                        65
+#define MC_ENGINE_IDX_NVDEC2                        66
+#define MC_ENGINE_IDX_NVDEC3                        67
+#define MC_ENGINE_IDX_NVDEC4                        68
+#define MC_ENGINE_IDX_NVDEC5                        69
+#define MC_ENGINE_IDX_NVDEC6                        70
+#define MC_ENGINE_IDX_NVDEC7                        71
+#define MC_ENGINE_IDX_CPU_DOORBELL                  72
+#define MC_ENGINE_IDX_PRIV_DOORBELL                 73
+#define MC_ENGINE_IDX_MMU_ECC_ERROR                 74
+#define MC_ENGINE_IDX_BLG                           75
+#define MC_ENGINE_IDX_PERFMON                       76
+#define MC_ENGINE_IDX_BUF_RESET                     77
+#define MC_ENGINE_IDX_XBAR                          78
+#define MC_ENGINE_IDX_ZPW                           79
+#define MC_ENGINE_IDX_OFA0                          80
+#define MC_ENGINE_IDX_TEGRA                         81
+#define MC_ENGINE_IDX_GR                            82
+#define MC_ENGINE_IDX_GR0                           MC_ENGINE_IDX_GR
+#define MC_ENGINE_IDX_GR1                           83
+#define MC_ENGINE_IDX_GR2                           84
+#define MC_ENGINE_IDX_GR3                           85
+#define MC_ENGINE_IDX_GR4                           86
+#define MC_ENGINE_IDX_GR5                           87
+#define MC_ENGINE_IDX_GR6                           88
+#define MC_ENGINE_IDX_GR7                           89
+#define MC_ENGINE_IDX_ESCHED                        90
+#define MC_ENGINE_IDX_ESCHED__SIZE                  64
+#define MC_ENGINE_IDX_GR_FECS_LOG                   154
+#define MC_ENGINE_IDX_GR0_FECS_LOG                  MC_ENGINE_IDX_GR_FECS_LOG
+#define MC_ENGINE_IDX_GR1_FECS_LOG                  155
+#define MC_ENGINE_IDX_GR2_FECS_LOG                  156
+#define MC_ENGINE_IDX_GR3_FECS_LOG                  157
+#define MC_ENGINE_IDX_GR4_FECS_LOG                  158
+#define MC_ENGINE_IDX_GR5_FECS_LOG                  159
+#define MC_ENGINE_IDX_GR6_FECS_LOG                  160
+#define MC_ENGINE_IDX_GR7_FECS_LOG                  161
+#define MC_ENGINE_IDX_TMR_SWRL                      162
+#define MC_ENGINE_IDX_DISP_GSP                      163
+#define MC_ENGINE_IDX_REPLAYABLE_FAULT_CPU          164
+#define MC_ENGINE_IDX_NON_REPLAYABLE_FAULT_CPU      165
+#define MC_ENGINE_IDX_PXUC                          166
+#define MC_ENGINE_IDX_MAX                           167 // This must be kept as the max bit if
+#define MC_ENGINE_IDX_INVALID                0xFFFFFFFF
+#define MC_ENGINE_IDX_GRn(x)            (MC_ENGINE_IDX_GR0 + (x))
+#define MC_ENGINE_IDX_GRn_FECS_LOG(x)   (MC_ENGINE_IDX_GR0_FECS_LOG + (x))
+#define MC_ENGINE_IDX_CE(x)             (MC_ENGINE_IDX_CE0 + (x))
+#define MC_ENGINE_IDX_MSENCn(x)         (MC_ENGINE_IDX_MSENC + (x))
+#define MC_ENGINE_IDX_NVDECn(x)         (MC_ENGINE_IDX_NVDEC + (x))
+#define MC_ENGINE_IDX_NVJPEGn(x)        (MC_ENGINE_IDX_NVJPEG + (x))
+#define MC_ENGINE_IDX_ESCHEDn(x)        (MC_ENGINE_IDX_ESCHED + (x))
+
+typedef enum
+{
+    RM_ENGINE_TYPE_NULL                 =       (0x00000000),
+    RM_ENGINE_TYPE_GR0                  =       (0x00000001),
+    RM_ENGINE_TYPE_GR1                  =       (0x00000002),
+    RM_ENGINE_TYPE_GR2                  =       (0x00000003),
+    RM_ENGINE_TYPE_GR3                  =       (0x00000004),
+    RM_ENGINE_TYPE_GR4                  =       (0x00000005),
+    RM_ENGINE_TYPE_GR5                  =       (0x00000006),
+    RM_ENGINE_TYPE_GR6                  =       (0x00000007),
+    RM_ENGINE_TYPE_GR7                  =       (0x00000008),
+    RM_ENGINE_TYPE_COPY0                =       (0x00000009),
+    RM_ENGINE_TYPE_COPY1                =       (0x0000000a),
+    RM_ENGINE_TYPE_COPY2                =       (0x0000000b),
+    RM_ENGINE_TYPE_COPY3                =       (0x0000000c),
+    RM_ENGINE_TYPE_COPY4                =       (0x0000000d),
+    RM_ENGINE_TYPE_COPY5                =       (0x0000000e),
+    RM_ENGINE_TYPE_COPY6                =       (0x0000000f),
+    RM_ENGINE_TYPE_COPY7                =       (0x00000010),
+    RM_ENGINE_TYPE_COPY8                =       (0x00000011),
+    RM_ENGINE_TYPE_COPY9                =       (0x00000012),
+    RM_ENGINE_TYPE_NVDEC0               =       (0x0000001d),
+    RM_ENGINE_TYPE_NVDEC1               =       (0x0000001e),
+    RM_ENGINE_TYPE_NVDEC2               =       (0x0000001f),
+    RM_ENGINE_TYPE_NVDEC3               =       (0x00000020),
+    RM_ENGINE_TYPE_NVDEC4               =       (0x00000021),
+    RM_ENGINE_TYPE_NVDEC5               =       (0x00000022),
+    RM_ENGINE_TYPE_NVDEC6               =       (0x00000023),
+    RM_ENGINE_TYPE_NVDEC7               =       (0x00000024),
+    RM_ENGINE_TYPE_NVENC0               =       (0x00000025),
+    RM_ENGINE_TYPE_NVENC1               =       (0x00000026),
+    RM_ENGINE_TYPE_NVENC2               =       (0x00000027),
+    RM_ENGINE_TYPE_VP                   =       (0x00000028),
+    RM_ENGINE_TYPE_ME                   =       (0x00000029),
+    RM_ENGINE_TYPE_PPP                  =       (0x0000002a),
+    RM_ENGINE_TYPE_MPEG                 =       (0x0000002b),
+    RM_ENGINE_TYPE_SW                   =       (0x0000002c),
+    RM_ENGINE_TYPE_TSEC                 =       (0x0000002d),
+    RM_ENGINE_TYPE_VIC                  =       (0x0000002e),
+    RM_ENGINE_TYPE_MP                   =       (0x0000002f),
+    RM_ENGINE_TYPE_SEC2                 =       (0x00000030),
+    RM_ENGINE_TYPE_HOST                 =       (0x00000031),
+    RM_ENGINE_TYPE_DPU                  =       (0x00000032),
+    RM_ENGINE_TYPE_PMU                  =       (0x00000033),
+    RM_ENGINE_TYPE_FBFLCN               =       (0x00000034),
+    RM_ENGINE_TYPE_NVJPEG0              =       (0x00000035),
+    RM_ENGINE_TYPE_NVJPEG1              =       (0x00000036),
+    RM_ENGINE_TYPE_NVJPEG2              =       (0x00000037),
+    RM_ENGINE_TYPE_NVJPEG3              =       (0x00000038),
+    RM_ENGINE_TYPE_NVJPEG4              =       (0x00000039),
+    RM_ENGINE_TYPE_NVJPEG5              =       (0x0000003a),
+    RM_ENGINE_TYPE_NVJPEG6              =       (0x0000003b),
+    RM_ENGINE_TYPE_NVJPEG7              =       (0x0000003c),
+    RM_ENGINE_TYPE_OFA                  =       (0x0000003d),
+    RM_ENGINE_TYPE_LAST                 =       (0x0000003e),
+} RM_ENGINE_TYPE;
+
+#define NV2080_ENGINE_TYPE_NULL                       (0x00000000)
+#define NV2080_ENGINE_TYPE_GRAPHICS                   (0x00000001)
+#define NV2080_ENGINE_TYPE_GR0                        NV2080_ENGINE_TYPE_GRAPHICS
+#define NV2080_ENGINE_TYPE_GR1                        (0x00000002)
+#define NV2080_ENGINE_TYPE_GR2                        (0x00000003)
+#define NV2080_ENGINE_TYPE_GR3                        (0x00000004)
+#define NV2080_ENGINE_TYPE_GR4                        (0x00000005)
+#define NV2080_ENGINE_TYPE_GR5                        (0x00000006)
+#define NV2080_ENGINE_TYPE_GR6                        (0x00000007)
+#define NV2080_ENGINE_TYPE_GR7                        (0x00000008)
+#define NV2080_ENGINE_TYPE_COPY0                      (0x00000009)
+#define NV2080_ENGINE_TYPE_COPY1                      (0x0000000a)
+#define NV2080_ENGINE_TYPE_COPY2                      (0x0000000b)
+#define NV2080_ENGINE_TYPE_COPY3                      (0x0000000c)
+#define NV2080_ENGINE_TYPE_COPY4                      (0x0000000d)
+#define NV2080_ENGINE_TYPE_COPY5                      (0x0000000e)
+#define NV2080_ENGINE_TYPE_COPY6                      (0x0000000f)
+#define NV2080_ENGINE_TYPE_COPY7                      (0x00000010)
+#define NV2080_ENGINE_TYPE_COPY8                      (0x00000011)
+#define NV2080_ENGINE_TYPE_COPY9                      (0x00000012)
+#define NV2080_ENGINE_TYPE_BSP                        (0x00000013)
+#define NV2080_ENGINE_TYPE_NVDEC0                     NV2080_ENGINE_TYPE_BSP
+#define NV2080_ENGINE_TYPE_NVDEC1                     (0x00000014)
+#define NV2080_ENGINE_TYPE_NVDEC2                     (0x00000015)
+#define NV2080_ENGINE_TYPE_NVDEC3                     (0x00000016)
+#define NV2080_ENGINE_TYPE_NVDEC4                     (0x00000017)
+#define NV2080_ENGINE_TYPE_NVDEC5                     (0x00000018)
+#define NV2080_ENGINE_TYPE_NVDEC6                     (0x00000019)
+#define NV2080_ENGINE_TYPE_NVDEC7                     (0x0000001a)
+#define NV2080_ENGINE_TYPE_MSENC                      (0x0000001b)
+#define NV2080_ENGINE_TYPE_NVENC0                      NV2080_ENGINE_TYPE_MSENC  /* Mutually exclusive alias */
+#define NV2080_ENGINE_TYPE_NVENC1                     (0x0000001c)
+#define NV2080_ENGINE_TYPE_NVENC2                     (0x0000001d)
+#define NV2080_ENGINE_TYPE_VP                         (0x0000001e)
+#define NV2080_ENGINE_TYPE_ME                         (0x0000001f)
+#define NV2080_ENGINE_TYPE_PPP                        (0x00000020)
+#define NV2080_ENGINE_TYPE_MPEG                       (0x00000021)
+#define NV2080_ENGINE_TYPE_SW                         (0x00000022)
+#define NV2080_ENGINE_TYPE_CIPHER                     (0x00000023)
+#define NV2080_ENGINE_TYPE_TSEC                       NV2080_ENGINE_TYPE_CIPHER
+#define NV2080_ENGINE_TYPE_VIC                        (0x00000024)
+#define NV2080_ENGINE_TYPE_MP                         (0x00000025)
+#define NV2080_ENGINE_TYPE_SEC2                       (0x00000026)
+#define NV2080_ENGINE_TYPE_HOST                       (0x00000027)
+#define NV2080_ENGINE_TYPE_DPU                        (0x00000028)
+#define NV2080_ENGINE_TYPE_PMU                        (0x00000029)
+#define NV2080_ENGINE_TYPE_FBFLCN                     (0x0000002a)
+#define NV2080_ENGINE_TYPE_NVJPG                      (0x0000002b)
+#define NV2080_ENGINE_TYPE_NVJPEG0                     NV2080_ENGINE_TYPE_NVJPG
+#define NV2080_ENGINE_TYPE_NVJPEG1                    (0x0000002c)
+#define NV2080_ENGINE_TYPE_NVJPEG2                    (0x0000002d)
+#define NV2080_ENGINE_TYPE_NVJPEG3                    (0x0000002e)
+#define NV2080_ENGINE_TYPE_NVJPEG4                    (0x0000002f)
+#define NV2080_ENGINE_TYPE_NVJPEG5                    (0x00000030)
+#define NV2080_ENGINE_TYPE_NVJPEG6                    (0x00000031)
+#define NV2080_ENGINE_TYPE_NVJPEG7                    (0x00000032)
+#define NV2080_ENGINE_TYPE_OFA                        (0x00000033)
+#define NV2080_ENGINE_TYPE_LAST                       (0x0000003e)
+#define NV2080_ENGINE_TYPE_ALLENGINES                 (0xffffffff)
+#define NV2080_ENGINE_TYPE_COPY_SIZE 10
+#define NV2080_ENGINE_TYPE_NVENC_SIZE 3
+#define NV2080_ENGINE_TYPE_NVJPEG_SIZE 8
+#define NV2080_ENGINE_TYPE_NVDEC_SIZE 8
+#define NV2080_ENGINE_TYPE_GR_SIZE 8
+#define NV2080_ENGINE_TYPE_COPY(i)     (NV2080_ENGINE_TYPE_COPY0+(i))
+#define NV2080_ENGINE_TYPE_IS_COPY(i)  (((i) >= NV2080_ENGINE_TYPE_COPY0) && ((i) <= NV2080_ENGINE_TYPE_COPY9))
+#define NV2080_ENGINE_TYPE_COPY_IDX(i) ((i) - NV2080_ENGINE_TYPE_COPY0)
+#define NV2080_ENGINE_TYPE_NVENC(i)    (NV2080_ENGINE_TYPE_NVENC0+(i))
+#define NV2080_ENGINE_TYPE_IS_NVENC(i)  (((i) >= NV2080_ENGINE_TYPE_NVENC0) && ((i) < NV2080_ENGINE_TYPE_NVENC(NV2080_ENGINE_TYPE_NVENC_SIZE)))
+#define NV2080_ENGINE_TYPE_NVENC_IDX(i) ((i) - NV2080_ENGINE_TYPE_NVENC0)
+#define NV2080_ENGINE_TYPE_NVDEC(i)    (NV2080_ENGINE_TYPE_NVDEC0+(i))
+#define NV2080_ENGINE_TYPE_IS_NVDEC(i)  (((i) >= NV2080_ENGINE_TYPE_NVDEC0) && ((i) < NV2080_ENGINE_TYPE_NVDEC(NV2080_ENGINE_TYPE_NVDEC_SIZE)))
+#define NV2080_ENGINE_TYPE_NVDEC_IDX(i) ((i) - NV2080_ENGINE_TYPE_NVDEC0)
+#define NV2080_ENGINE_TYPE_NVJPEG(i)    (NV2080_ENGINE_TYPE_NVJPEG0+(i))
+#define NV2080_ENGINE_TYPE_IS_NVJPEG(i)  (((i) >= NV2080_ENGINE_TYPE_NVJPEG0) && ((i) < NV2080_ENGINE_TYPE_NVJPEG(NV2080_ENGINE_TYPE_NVJPEG_SIZE)))
+#define NV2080_ENGINE_TYPE_NVJPEG_IDX(i) ((i) - NV2080_ENGINE_TYPE_NVJPEG0)
+#define NV2080_ENGINE_TYPE_GR(i)       (NV2080_ENGINE_TYPE_GR0 + (i))
+#define NV2080_ENGINE_TYPE_IS_GR(i)    (((i) >= NV2080_ENGINE_TYPE_GR0) && ((i) < NV2080_ENGINE_TYPE_GR(NV2080_ENGINE_TYPE_GR_SIZE)))
+#define NV2080_ENGINE_TYPE_GR_IDX(i)   ((i) - NV2080_ENGINE_TYPE_GR0)
+#define NV2080_ENGINE_TYPE_IS_VALID(i) (((i) > (NV2080_ENGINE_TYPE_NULL)) && ((i) < (NV2080_ENGINE_TYPE_LAST)))
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/event.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/event.h
new file mode 100644
index 000000000000..057f7220c225
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/event.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_EVENT_H__
+#define __NVRM_EVENT_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV01_EVENT_KERNEL_CALLBACK_EX            (0x0000007e)
+
+typedef struct NV0005_ALLOC_PARAMETERS {
+    NvHandle hParentClient;
+    NvHandle hSrcResource;
+
+    NvV32    hClass;
+    NvV32    notifyIndex;
+    NV_DECLARE_ALIGNED(NvP64 data, 8);
+} NV0005_ALLOC_PARAMETERS;
+
+#define NV01_EVENT_CLIENT_RM                                       (0x04000000)
+
+#define NV2080_CTRL_CMD_EVENT_SET_NOTIFICATION (0x20800301) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_EVENT_INTERFACE_ID << 8) | NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS {
+    NvU32  event;
+    NvU32  action;
+    NvBool bNotifyState;
+    NvU32  info32;
+    NvU16  info16;
+} NV2080_CTRL_EVENT_SET_NOTIFICATION_PARAMS;
+
+#define NV2080_CTRL_EVENT_SET_NOTIFICATION_ACTION_REPEAT  (0x00000002)
+
+typedef struct rpc_post_event_v17_00
+{
+    NvHandle   hClient;
+    NvHandle   hEvent;
+    NvU32      notifyIndex;
+    NvU32      data;
+    NvU16      info16;
+    NvU32      status;
+    NvU32      eventDataSize;
+    NvBool     bNotifyList;
+    NvU8       eventData[];
+} rpc_post_event_v17_00;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/fbsr.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/fbsr.h
new file mode 100644
index 000000000000..28786ef013a2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/fbsr.h
@@ -0,0 +1,106 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_FBSR_H__
+#define __NVRM_FBSR_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV01_MEMORY_LIST_FBMEM  (0x00000082)
+
+#define NV01_MEMORY_LIST_SYSTEM (0x00000081)
+
+#define NVOS02_FLAGS_PHYSICALITY                                   7:4
+#define NVOS02_FLAGS_PHYSICALITY_CONTIGUOUS                        (0x00000000)
+#define NVOS02_FLAGS_PHYSICALITY_NONCONTIGUOUS                     (0x00000001)
+#define NVOS02_FLAGS_LOCATION                                      11:8
+#define NVOS02_FLAGS_LOCATION_PCI                                  (0x00000000)
+#define NVOS02_FLAGS_LOCATION_AGP                                  (0x00000001)
+#define NVOS02_FLAGS_LOCATION_VIDMEM                               (0x00000002)
+#define NVOS02_FLAGS_COHERENCY                                     15:12
+#define NVOS02_FLAGS_COHERENCY_UNCACHED                            (0x00000000)
+#define NVOS02_FLAGS_COHERENCY_CACHED                              (0x00000001)
+#define NVOS02_FLAGS_COHERENCY_WRITE_COMBINE                       (0x00000002)
+#define NVOS02_FLAGS_COHERENCY_WRITE_THROUGH                       (0x00000003)
+#define NVOS02_FLAGS_COHERENCY_WRITE_PROTECT                       (0x00000004)
+#define NVOS02_FLAGS_COHERENCY_WRITE_BACK                          (0x00000005)
+#define NVOS02_FLAGS_ALLOC                                         17:16
+#define NVOS02_FLAGS_ALLOC_NONE                                    (0x00000001)
+#define NVOS02_FLAGS_GPU_CACHEABLE                                 18:18
+#define NVOS02_FLAGS_GPU_CACHEABLE_NO                              (0x00000000)
+#define NVOS02_FLAGS_GPU_CACHEABLE_YES                             (0x00000001)
+#define NVOS02_FLAGS_KERNEL_MAPPING                                19:19
+#define NVOS02_FLAGS_KERNEL_MAPPING_NO_MAP                         (0x00000000)
+#define NVOS02_FLAGS_KERNEL_MAPPING_MAP                            (0x00000001)
+#define NVOS02_FLAGS_ALLOC_NISO_DISPLAY                            20:20
+#define NVOS02_FLAGS_ALLOC_NISO_DISPLAY_NO                         (0x00000000)
+#define NVOS02_FLAGS_ALLOC_NISO_DISPLAY_YES                        (0x00000001)
+#define NVOS02_FLAGS_ALLOC_USER_READ_ONLY                          21:21
+#define NVOS02_FLAGS_ALLOC_USER_READ_ONLY_NO                       (0x00000000)
+#define NVOS02_FLAGS_ALLOC_USER_READ_ONLY_YES                      (0x00000001)
+#define NVOS02_FLAGS_ALLOC_DEVICE_READ_ONLY                        22:22
+#define NVOS02_FLAGS_ALLOC_DEVICE_READ_ONLY_NO                     (0x00000000)
+#define NVOS02_FLAGS_ALLOC_DEVICE_READ_ONLY_YES                    (0x00000001)
+#define NVOS02_FLAGS_PEER_MAP_OVERRIDE                             23:23
+#define NVOS02_FLAGS_PEER_MAP_OVERRIDE_DEFAULT                     (0x00000000)
+#define NVOS02_FLAGS_PEER_MAP_OVERRIDE_REQUIRED                    (0x00000001)
+#define NVOS02_FLAGS_ALLOC_TYPE_SYNCPOINT                          24:24
+#define NVOS02_FLAGS_ALLOC_TYPE_SYNCPOINT_APERTURE                 (0x00000001)
+#define NVOS02_FLAGS_MEMORY_PROTECTION                             26:25
+#define NVOS02_FLAGS_MEMORY_PROTECTION_DEFAULT                     (0x00000000)
+#define NVOS02_FLAGS_MEMORY_PROTECTION_PROTECTED                   (0x00000001)
+#define NVOS02_FLAGS_MEMORY_PROTECTION_UNPROTECTED                 (0x00000002)
+#define NVOS02_FLAGS_MAPPING                                       31:30
+#define NVOS02_FLAGS_MAPPING_DEFAULT                               (0x00000000)
+#define NVOS02_FLAGS_MAPPING_NO_MAP                                (0x00000001)
+#define NVOS02_FLAGS_MAPPING_NEVER_MAP                             (0x00000002)
+
+struct pte_desc
+{
+    NvU32 idr:2;
+    NvU32 reserved1:14;
+    NvU32 length:16;
+    union {
+        NvU64 pte; // PTE when IDR==0; PDE when IDR > 0
+        NvU64 pde; // PTE when IDR==0; PDE when IDR > 0
+    } pte_pde[]  NV_ALIGN_BYTES(8); // PTE when IDR==0; PDE when IDR > 0
+};
+
+typedef struct rpc_alloc_memory_v13_01
+{
+    NvHandle   hClient;
+    NvHandle   hDevice;
+    NvHandle   hMemory;
+    NvU32      hClass;
+    NvU32      flags;
+    NvU32      pteAdjust;
+    NvU32      format;
+    NvU64      length NV_ALIGN_BYTES(8);
+    NvU32      pageCount;
+    struct pte_desc pteDesc;
+} rpc_alloc_memory_v13_01;
+
+#define FBSR_TYPE_DMA                                 4   // Copy using DMA. Fastest.
+
+#define NV2080_CTRL_CMD_INTERNAL_FBSR_INIT (0x20800ac2) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS {
+    NvU32    fbsrType;
+    NvU32    numRegions;
+    NvHandle hClient;
+    NvHandle hSysMem;
+    NV_DECLARE_ALIGNED(NvU64 gspFbAllocsSysOffset, 8);
+    NvBool   bEnteringGcoffState;
+} NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS;
+
+#define NV2080_CTRL_CMD_INTERNAL_FBSR_SEND_REGION_INFO (0x20800ac3) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_FBSR_SEND_REGION_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_FBSR_SEND_REGION_INFO_PARAMS {
+    NvU32    fbsrType;
+    NvHandle hClient;
+    NvHandle hVidMem;
+    NV_DECLARE_ALIGNED(NvU64 vidOffset, 8);
+    NV_DECLARE_ALIGNED(NvU64 sysOffset, 8);
+    NV_DECLARE_ALIGNED(NvU64 size, 8);
+} NV2080_CTRL_INTERNAL_FBSR_SEND_REGION_INFO_PARAMS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/fifo.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/fifo.h
new file mode 100644
index 000000000000..325fdd8b6090
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/fifo.h
@@ -0,0 +1,350 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_FIFO_H__
+#define __NVRM_FIFO_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_MAX_ENTRIES         32
+
+#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_DATA_TYPES   16
+
+#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_PBDMA    2
+
+#define NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_NAME_LEN 16
+
+typedef struct NV2080_CTRL_FIFO_DEVICE_ENTRY {
+    NvU32 engineData[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_DATA_TYPES];
+    NvU32 pbdmaIds[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_PBDMA];
+    NvU32 pbdmaFaultIds[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_PBDMA];
+    NvU32 numPbdmas;
+    char  engineName[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_ENGINE_MAX_NAME_LEN];
+} NV2080_CTRL_FIFO_DEVICE_ENTRY;
+
+#define NV2080_CTRL_CMD_FIFO_GET_DEVICE_INFO_TABLE                 (0x20801112) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_FIFO_INTERFACE_ID << 8) | NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS {
+    NvU32                         baseIndex;
+    NvU32                         numEntries;
+    NvBool                        bMore;
+    // C form: NV2080_CTRL_FIFO_DEVICE_ENTRY entries[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_MAX_ENTRIES];
+    NV2080_CTRL_FIFO_DEVICE_ENTRY entries[NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_MAX_ENTRIES];
+} NV2080_CTRL_FIFO_GET_DEVICE_INFO_TABLE_PARAMS;
+
+typedef enum
+{
+    /* *************************************************************************
+     * Bug 3820969
+     * THINK BEFORE CHANGING ENUM ORDER HERE.
+     * VGPU-guest uses this same ordering. Because this enum is not versioned,
+     * changing the order here WILL BREAK old-guest-on-newer-host compatibility.
+     * ************************************************************************/
+
+    // *ENG_XYZ, e.g.: ENG_GR, ENG_CE etc.,
+    ENGINE_INFO_TYPE_ENG_DESC = 0,
+
+    // HW engine ID
+    ENGINE_INFO_TYPE_FIFO_TAG,
+
+    // RM_ENGINE_TYPE_*
+    ENGINE_INFO_TYPE_RM_ENGINE_TYPE,
+
+    //
+    // runlist id (meaning varies by GPU)
+    // Valid only for Esched-driven engines
+    //
+    ENGINE_INFO_TYPE_RUNLIST,
+
+    // NV_PFIFO_INTR_MMU_FAULT_ENG_ID_*
+    ENGINE_INFO_TYPE_MMU_FAULT_ID,
+
+    // ROBUST_CHANNEL_*
+    ENGINE_INFO_TYPE_RC_MASK,
+
+    // Reset Bit Position. On Ampere, only valid if not _INVALID
+    ENGINE_INFO_TYPE_RESET,
+
+    // Interrupt Bit Position
+    ENGINE_INFO_TYPE_INTR,
+
+    // log2(MC_ENGINE_*)
+    ENGINE_INFO_TYPE_MC,
+
+    // The DEV_TYPE_ENUM for this engine
+    ENGINE_INFO_TYPE_DEV_TYPE_ENUM,
+
+    // The particular instance of this engine type
+    ENGINE_INFO_TYPE_INSTANCE_ID,
+
+    //
+    // The base address for this engine's NV_RUNLIST. Valid only on Ampere+
+    // Valid only for Esched-driven engines
+    //
+    ENGINE_INFO_TYPE_RUNLIST_PRI_BASE,
+
+    //
+    // If this entry is a host-driven engine.
+    // Update _isEngineInfoTypeValidForOnlyHostDriven when adding any new entry.
+    //
+    ENGINE_INFO_TYPE_IS_HOST_DRIVEN_ENGINE,
+
+    //
+    // The index into the per-engine NV_RUNLIST registers. Valid only on Ampere+
+    // Valid only for Esched-driven engines
+    //
+    ENGINE_INFO_TYPE_RUNLIST_ENGINE_ID,
+
+    //
+    // The base address for this engine's NV_CHRAM registers. Valid only on
+    // Ampere+
+    //
+    // Valid only for Esched-driven engines
+    //
+    ENGINE_INFO_TYPE_CHRAM_PRI_BASE,
+
+    // This entry added to copy data at RMCTRL_EXPORT() call for Kernel RM
+    ENGINE_INFO_TYPE_KERNEL_RM_MAX,
+    // Used for iterating the engine info table by the index passed.
+    ENGINE_INFO_TYPE_INVALID = ENGINE_INFO_TYPE_KERNEL_RM_MAX,
+
+    // Size of FIFO_ENGINE_LIST.engineData
+    ENGINE_INFO_TYPE_ENGINE_DATA_ARRAY_SIZE = ENGINE_INFO_TYPE_INVALID,
+
+    // Input-only parameter for kfifoEngineInfoXlate.
+    ENGINE_INFO_TYPE_PBDMA_ID
+
+    /* *************************************************************************
+     * Bug 3820969
+     * THINK BEFORE CHANGING ENUM ORDER HERE.
+     * VGPU-guest uses this same ordering. Because this enum is not versioned,
+     * changing the order here WILL BREAK old-guest-on-newer-host compatibility.
+     * ************************************************************************/
+} ENGINE_INFO_TYPE;
+
+#define NV2080_CTRL_CMD_CE_GET_FAULT_METHOD_BUFFER_SIZE (0x20802a08) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_CE_INTERFACE_ID << 8) | NV2080_CTRL_CE_GET_FAULT_METHOD_BUFFER_SIZE_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_CE_GET_FAULT_METHOD_BUFFER_SIZE_PARAMS {
+    NvU32 size;
+} NV2080_CTRL_CE_GET_FAULT_METHOD_BUFFER_SIZE_PARAMS;
+
+#define NV2080_CTRL_CMD_INTERNAL_MAX_CONSTRUCTED_FALCONS     0x40
+
+typedef struct NV2080_CTRL_INTERNAL_CONSTRUCTED_FALCON_INFO {
+    NvU32 engDesc;
+    NvU32 ctxAttr;
+    NvU32 ctxBufferSize;
+    NvU32 addrSpaceList;
+    NvU32 registerBase;
+} NV2080_CTRL_INTERNAL_CONSTRUCTED_FALCON_INFO;
+
+#define NV2080_CTRL_CMD_INTERNAL_GET_CONSTRUCTED_FALCON_INFO (0x20800a42) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_GET_CONSTRUCTED_FALCON_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_GET_CONSTRUCTED_FALCON_INFO_PARAMS {
+    NvU32                                        numConstructedFalcons;
+    NV2080_CTRL_INTERNAL_CONSTRUCTED_FALCON_INFO constructedFalconsTable[NV2080_CTRL_CMD_INTERNAL_MAX_CONSTRUCTED_FALCONS];
+} NV2080_CTRL_INTERNAL_GET_CONSTRUCTED_FALCON_INFO_PARAMS;
+
+#define NV_MAX_SUBDEVICES       8
+
+typedef struct NV_MEMORY_DESC_PARAMS {
+    NV_DECLARE_ALIGNED(NvU64 base, 8);
+    NV_DECLARE_ALIGNED(NvU64 size, 8);
+    NvU32 addressSpace;
+    NvU32 cacheAttrib;
+} NV_MEMORY_DESC_PARAMS;
+
+#define CC_CHAN_ALLOC_IV_SIZE_DWORD    3U
+
+#define CC_CHAN_ALLOC_NONCE_SIZE_DWORD 8U
+
+typedef struct NV_CHANNEL_ALLOC_PARAMS {
+
+    NvHandle hObjectError; // error context DMA
+    NvHandle hObjectBuffer; // no longer used
+    NV_DECLARE_ALIGNED(NvU64 gpFifoOffset, 8);    // offset to beginning of GP FIFO
+    NvU32    gpFifoEntries;    // number of GP FIFO entries
+
+    NvU32    flags;
+
+
+    NvHandle hContextShare; // context share handle
+    NvHandle hVASpace; // VASpace for the channel
+
+    // handle to UserD memory object for channel, ignored if hUserdMemory[0]=0
+    NvHandle hUserdMemory[NV_MAX_SUBDEVICES];
+
+    // offset to beginning of UserD within hUserdMemory[x]
+    NV_DECLARE_ALIGNED(NvU64 userdOffset[NV_MAX_SUBDEVICES], 8);
+
+    // engine type(NV2080_ENGINE_TYPE_*) with which this channel is associated
+    NvU32    engineType;
+    // Channel identifier that is unique for the duration of a RM session
+    NvU32    cid;
+    // One-hot encoded bitmask to match SET_SUBDEVICE_MASK methods
+    NvU32    subDeviceId;
+    NvHandle hObjectEccError; // ECC error context DMA
+
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS instanceMem, 8);
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS userdMem, 8);
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS ramfcMem, 8);
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS mthdbufMem, 8);
+
+    NvHandle hPhysChannelGroup;              // reserved
+    NvU32    internalFlags;                 // reserved
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS errorNotifierMem, 8); // reserved
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS eccErrorNotifierMem, 8); // reserved
+    NvU32    ProcessID;                 // reserved
+    NvU32    SubProcessID;                 // reserved
+
+    // IV used for CPU-side encryption / GPU-side decryption.
+    NvU32    encryptIv[CC_CHAN_ALLOC_IV_SIZE_DWORD];          // reserved
+    // IV used for CPU-side decryption / GPU-side encryption.
+    NvU32    decryptIv[CC_CHAN_ALLOC_IV_SIZE_DWORD];          // reserved
+    // Nonce used CPU-side signing / GPU-side signature verification.
+    NvU32    hmacNonce[CC_CHAN_ALLOC_NONCE_SIZE_DWORD];       // reserved
+} NV_CHANNEL_ALLOC_PARAMS;
+
+typedef NV_CHANNEL_ALLOC_PARAMS NV_CHANNELGPFIFO_ALLOCATION_PARAMETERS;
+
+#define NVOS04_FLAGS_CHANNEL_TYPE                                  1:0
+#define NVOS04_FLAGS_CHANNEL_TYPE_PHYSICAL                         0x00000000
+#define NVOS04_FLAGS_CHANNEL_TYPE_VIRTUAL                          0x00000001  // OBSOLETE
+#define NVOS04_FLAGS_CHANNEL_TYPE_PHYSICAL_FOR_VIRTUAL             0x00000002  // OBSOLETE
+#define NVOS04_FLAGS_VPR                                           2:2
+#define NVOS04_FLAGS_VPR_FALSE                                     0x00000000
+#define NVOS04_FLAGS_VPR_TRUE                                      0x00000001
+#define NVOS04_FLAGS_CC_SECURE                                     2:2
+#define NVOS04_FLAGS_CC_SECURE_FALSE                               0x00000000
+#define NVOS04_FLAGS_CC_SECURE_TRUE                                0x00000001
+#define NVOS04_FLAGS_CHANNEL_SKIP_MAP_REFCOUNTING                  3:3
+#define NVOS04_FLAGS_CHANNEL_SKIP_MAP_REFCOUNTING_FALSE            0x00000000
+#define NVOS04_FLAGS_CHANNEL_SKIP_MAP_REFCOUNTING_TRUE             0x00000001
+#define NVOS04_FLAGS_GROUP_CHANNEL_RUNQUEUE                       4:4
+#define NVOS04_FLAGS_GROUP_CHANNEL_RUNQUEUE_DEFAULT               0x00000000
+#define NVOS04_FLAGS_GROUP_CHANNEL_RUNQUEUE_ONE                   0x00000001
+#define NVOS04_FLAGS_PRIVILEGED_CHANNEL                           5:5
+#define NVOS04_FLAGS_PRIVILEGED_CHANNEL_FALSE                     0x00000000
+#define NVOS04_FLAGS_PRIVILEGED_CHANNEL_TRUE                      0x00000001
+#define NVOS04_FLAGS_DELAY_CHANNEL_SCHEDULING                     6:6
+#define NVOS04_FLAGS_DELAY_CHANNEL_SCHEDULING_FALSE               0x00000000
+#define NVOS04_FLAGS_DELAY_CHANNEL_SCHEDULING_TRUE                0x00000001
+#define NVOS04_FLAGS_CHANNEL_DENY_PHYSICAL_MODE_CE                7:7
+#define NVOS04_FLAGS_CHANNEL_DENY_PHYSICAL_MODE_CE_FALSE          0x00000000
+#define NVOS04_FLAGS_CHANNEL_DENY_PHYSICAL_MODE_CE_TRUE           0x00000001
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_VALUE                    10:8
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_FIXED                    11:11
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_FIXED_FALSE              0x00000000
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_FIXED_TRUE               0x00000001
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_VALUE               20:12
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_FIXED               21:21
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_FIXED_FALSE         0x00000000
+#define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_FIXED_TRUE          0x00000001
+#define NVOS04_FLAGS_CHANNEL_DENY_AUTH_LEVEL_PRIV                 22:22
+#define NVOS04_FLAGS_CHANNEL_DENY_AUTH_LEVEL_PRIV_FALSE           0x00000000
+#define NVOS04_FLAGS_CHANNEL_DENY_AUTH_LEVEL_PRIV_TRUE            0x00000001
+#define NVOS04_FLAGS_CHANNEL_SKIP_SCRUBBER                        23:23
+#define NVOS04_FLAGS_CHANNEL_SKIP_SCRUBBER_FALSE                  0x00000000
+#define NVOS04_FLAGS_CHANNEL_SKIP_SCRUBBER_TRUE                   0x00000001
+#define NVOS04_FLAGS_CHANNEL_CLIENT_MAP_FIFO                      24:24
+#define NVOS04_FLAGS_CHANNEL_CLIENT_MAP_FIFO_FALSE                0x00000000
+#define NVOS04_FLAGS_CHANNEL_CLIENT_MAP_FIFO_TRUE                 0x00000001
+#define NVOS04_FLAGS_SET_EVICT_LAST_CE_PREFETCH_CHANNEL           25:25
+#define NVOS04_FLAGS_SET_EVICT_LAST_CE_PREFETCH_CHANNEL_FALSE     0x00000000
+#define NVOS04_FLAGS_SET_EVICT_LAST_CE_PREFETCH_CHANNEL_TRUE      0x00000001
+#define NVOS04_FLAGS_CHANNEL_VGPU_PLUGIN_CONTEXT                  26:26
+#define NVOS04_FLAGS_CHANNEL_VGPU_PLUGIN_CONTEXT_FALSE            0x00000000
+#define NVOS04_FLAGS_CHANNEL_VGPU_PLUGIN_CONTEXT_TRUE             0x00000001
+#define NVOS04_FLAGS_CHANNEL_PBDMA_ACQUIRE_TIMEOUT                 27:27
+#define NVOS04_FLAGS_CHANNEL_PBDMA_ACQUIRE_TIMEOUT_FALSE           0x00000000
+#define NVOS04_FLAGS_CHANNEL_PBDMA_ACQUIRE_TIMEOUT_TRUE            0x00000001
+#define NVOS04_FLAGS_GROUP_CHANNEL_THREAD                          29:28
+#define NVOS04_FLAGS_GROUP_CHANNEL_THREAD_DEFAULT                  0x00000000
+#define NVOS04_FLAGS_GROUP_CHANNEL_THREAD_ONE                      0x00000001
+#define NVOS04_FLAGS_GROUP_CHANNEL_THREAD_TWO                      0x00000002
+#define NVOS04_FLAGS_MAP_CHANNEL                                   30:30
+#define NVOS04_FLAGS_MAP_CHANNEL_FALSE                             0x00000000
+#define NVOS04_FLAGS_MAP_CHANNEL_TRUE                              0x00000001
+#define NVOS04_FLAGS_SKIP_CTXBUFFER_ALLOC                          31:31
+#define NVOS04_FLAGS_SKIP_CTXBUFFER_ALLOC_FALSE                    0x00000000
+#define NVOS04_FLAGS_SKIP_CTXBUFFER_ALLOC_TRUE                     0x00000001
+
+typedef enum {
+    /*!
+     * Initial state as passed in NV_CHANNEL_ALLOC_PARAMS by
+     * kernel CPU-RM clients.
+     */
+    ERROR_NOTIFIER_TYPE_UNKNOWN = 0,
+    /*! @brief Error notifier is explicitly not set.
+     *
+     * The corresponding hErrorContext or hEccErrorContext must be
+     * NV01_NULL_OBJECT.
+     */
+    ERROR_NOTIFIER_TYPE_NONE,
+    /*! @brief Error notifier is a ContextDma */
+    ERROR_NOTIFIER_TYPE_CTXDMA,
+    /*! @brief Error notifier is a NvNotification array in sysmem/vidmem */
+    ERROR_NOTIFIER_TYPE_MEMORY
+} ErrorNotifierType;
+
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE                       1:0
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_USER                  0x0
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_ADMIN                 0x1
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_KERNEL                0x2
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE             3:2
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_UNKNOWN     ERROR_NOTIFIER_TYPE_UNKNOWN
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_NONE        ERROR_NOTIFIER_TYPE_NONE
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_CTXDMA      ERROR_NOTIFIER_TYPE_CTXDMA
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_MEMORY      ERROR_NOTIFIER_TYPE_MEMORY
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE         5:4
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_UNKNOWN ERROR_NOTIFIER_TYPE_UNKNOWN
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_NONE    ERROR_NOTIFIER_TYPE_NONE
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_CTXDMA  ERROR_NOTIFIER_TYPE_CTXDMA
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_MEMORY  ERROR_NOTIFIER_TYPE_MEMORY
+
+#define NVA06F_CTRL_CMD_BIND (0xa06f0104) /* finn: Evaluated from "(FINN_KEPLER_CHANNEL_GPFIFO_A_GPFIFO_INTERFACE_ID << 8) | NVA06F_CTRL_BIND_PARAMS_MESSAGE_ID" */
+typedef struct NVA06F_CTRL_BIND_PARAMS {
+    NvU32 engineType;
+} NVA06F_CTRL_BIND_PARAMS;
+
+#define NVA06F_CTRL_CMD_GPFIFO_SCHEDULE (0xa06f0103) /* finn: Evaluated from "(FINN_KEPLER_CHANNEL_GPFIFO_A_GPFIFO_INTERFACE_ID << 8) | NVA06F_CTRL_GPFIFO_SCHEDULE_PARAMS_MESSAGE_ID" */
+typedef struct NVA06F_CTRL_GPFIFO_SCHEDULE_PARAMS {
+    NvBool bEnable;
+    NvBool bSkipSubmit;
+} NVA06F_CTRL_GPFIFO_SCHEDULE_PARAMS;
+
+#define NV2080_CTRL_GPU_PROMOTE_CONTEXT_MAX_ENTRIES                        16U
+
+typedef struct NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY {
+    NV_DECLARE_ALIGNED(NvU64 gpuPhysAddr, 8);
+    NV_DECLARE_ALIGNED(NvU64 gpuVirtAddr, 8);
+    NV_DECLARE_ALIGNED(NvU64 size, 8);
+    NvU32 physAttr;
+    NvU16 bufferId;
+    NvU8  bInitialize;
+    NvU8  bNonmapped;
+} NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY;
+
+#define NV2080_CTRL_CMD_GPU_PROMOTE_CTX                                    (0x2080012bU) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_GPU_INTERFACE_ID << 8) | NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS {
+    NvU32    engineType;
+    NvHandle hClient;
+    NvU32    ChID;
+    NvHandle hChanClient;
+    NvHandle hObject;
+    NvHandle hVirtMemory;
+    NV_DECLARE_ALIGNED(NvU64 virtAddress, 8);
+    NV_DECLARE_ALIGNED(NvU64 size, 8);
+    NvU32    entryCount;
+    // C form: NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY promoteEntry[NV2080_CTRL_GPU_PROMOTE_CONTEXT_MAX_ENTRIES];
+    NV_DECLARE_ALIGNED(NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ENTRY promoteEntry[NV2080_CTRL_GPU_PROMOTE_CONTEXT_MAX_ENTRIES], 8);
+} NV2080_CTRL_GPU_PROMOTE_CTX_PARAMS;
+
+typedef struct rpc_rc_triggered_v17_02
+{
+    NvU32      nv2080EngineType;
+    NvU32      chid;
+    NvU32      exceptType;
+    NvU32      scope;
+    NvU16      partitionAttributionId;
+} rpc_rc_triggered_v17_02;
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080fifo.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/gr.h
index 6acb3f73242d..82c5ec727bb4 100644
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl0080/ctrl0080fifo.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/gr.h
@@ -1,30 +1,31 @@
-#ifndef __src_common_sdk_nvidia_inc_ctrl_ctrl0080_ctrl0080fifo_h__
-#define __src_common_sdk_nvidia_inc_ctrl_ctrl0080_ctrl0080fifo_h__
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_GR_H__
+#define __NVRM_GR_H__
+#include <nvrm/nvtypes.h>
 
 /* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
 
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2006-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
+#define NV2080_CTRL_CMD_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO (0x20800a32) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO_PARAMS_MESSAGE_ID" */
+
+#define NV2080_CTRL_INTERNAL_GR_MAX_ENGINES         8
+
+#define NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT 0x19
+
+typedef struct NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO {
+    NvU32 size;
+    NvU32 alignment;
+} NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO;
+
+typedef struct NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO {
+    NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO engine[NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT];
+} NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO;
+
+typedef struct NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS {
+    NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO engineContextBuffersInfo[NV2080_CTRL_INTERNAL_GR_MAX_ENGINES];
+} NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS;
 
 #define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID                          4:0
 #define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS                 (0x00000000)
@@ -54,4 +55,19 @@
 #define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PRIV_ACCESS_MAP (0x00000018)
 #define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT                    (0x00000019)
 
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_MAIN                         0U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PM                           1U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PATCH                        2U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_BUFFER_BUNDLE_CB             3U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PAGEPOOL                     4U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_ATTRIBUTE_CB                 5U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_RTV_CB_GLOBAL                6U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_GFXP_POOL                    7U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_GFXP_CTRL_BLK                8U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_FECS_EVENT                   9U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_PRIV_ACCESS_MAP              10U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_UNRESTRICTED_PRIV_ACCESS_MAP 11U
+#define NV2080_CTRL_GPU_PROMOTE_CTX_BUFFER_ID_GLOBAL_PRIV_ACCESS_MAP       12U
+
+#include "fifo.h"
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/gsp.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/gsp.h
new file mode 100644
index 000000000000..b6683a5bf870
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/gsp.h
@@ -0,0 +1,825 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_GSP_H__
+#define __NVRM_GSP_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MAX_ENTRIES 16U
+
+#define NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MEM_TYPES   17U
+
+typedef NvBool NV2080_CTRL_CMD_FB_GET_FB_REGION_SURFACE_MEM_TYPE_FLAG[NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MEM_TYPES];
+
+typedef struct NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO {
+    NV_DECLARE_ALIGNED(NvU64 base, 8);
+    NV_DECLARE_ALIGNED(NvU64 limit, 8);
+    NV_DECLARE_ALIGNED(NvU64 reserved, 8);
+    NvU32                                                  performance;
+    NvBool                                                 supportCompressed;
+    NvBool                                                 supportISO;
+    NvBool                                                 bProtected;
+    NV2080_CTRL_CMD_FB_GET_FB_REGION_SURFACE_MEM_TYPE_FLAG blackList;
+} NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO;
+
+typedef struct NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS {
+    NvU32 numFBRegions;
+    NV_DECLARE_ALIGNED(NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO fbRegion[NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MAX_ENTRIES], 8);
+} NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS;
+
+#define NV0080_CTRL_GR_CAPS_TBL_SIZE            23
+
+#define NV2080_GPU_MAX_GID_LENGTH             (0x000000100ULL)
+
+typedef struct NV2080_CTRL_GPU_GET_GID_INFO_PARAMS {
+    NvU32 index;
+    NvU32 flags;
+    NvU32 length;
+    NvU8  data[NV2080_GPU_MAX_GID_LENGTH];
+} NV2080_CTRL_GPU_GET_GID_INFO_PARAMS;
+
+typedef struct NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS {
+    NvU32 gpcMask;
+} NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS;
+
+typedef struct NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS {
+    NvU32 gpcId;
+    NvU32 tpcMask;
+} NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS;
+
+typedef struct NV2080_CTRL_GPU_GET_FERMI_ZCULL_INFO_PARAMS {
+    NvU32 gpcId;
+    NvU32 zcullMask;
+} NV2080_CTRL_GPU_GET_FERMI_ZCULL_INFO_PARAMS;
+
+typedef struct NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS {
+    NvU32 BoardID;
+    char  chipSKU[4];
+    char  chipSKUMod[2];
+    char  project[5];
+    char  projectSKU[5];
+    char  CDP[6];
+    char  projectSKUMod[2];
+    NvU32 businessCycle;
+} NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS;
+
+typedef enum
+{
+    COMPUTE_BRANDING_TYPE_NONE,
+    COMPUTE_BRANDING_TYPE_TESLA,
+} COMPUTE_BRANDING_TYPE;
+
+#define MAX_GPC_COUNT           32
+
+typedef struct NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS {
+    NvU32  totalVFs;
+    NvU32  firstVfOffset;
+    NvU32  vfFeatureMask;
+    NV_DECLARE_ALIGNED(NvU64 FirstVFBar0Address, 8);
+    NV_DECLARE_ALIGNED(NvU64 FirstVFBar1Address, 8);
+    NV_DECLARE_ALIGNED(NvU64 FirstVFBar2Address, 8);
+    NV_DECLARE_ALIGNED(NvU64 bar0Size, 8);
+    NV_DECLARE_ALIGNED(NvU64 bar1Size, 8);
+    NV_DECLARE_ALIGNED(NvU64 bar2Size, 8);
+    NvBool b64bitBar0;
+    NvBool b64bitBar1;
+    NvBool b64bitBar2;
+    NvBool bSriovEnabled;
+    NvBool bSriovHeavyEnabled;
+    NvBool bEmulateVFBar0TlbInvalidationRegister;
+    NvBool bClientRmAllocatedCtxBuffer;
+} NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS;
+
+#include "engine.h"
+
+#define NVGPU_ENGINE_CAPS_MASK_BITS                32
+
+#define NVGPU_ENGINE_CAPS_MASK_ARRAY_MAX           ((RM_ENGINE_TYPE_LAST-1)/NVGPU_ENGINE_CAPS_MASK_BITS + 1)
+
+typedef struct GspSMInfo_t
+{
+    NvU32 version;
+    NvU32 regBankCount;
+    NvU32 regBankRegCount;
+    NvU32 maxWarpsPerSM;
+    NvU32 maxThreadsPerWarp;
+    NvU32 geomGsObufEntries;
+    NvU32 geomXbufEntries;
+    NvU32 maxSPPerSM;
+    NvU32 rtCoreCount;
+} GspSMInfo;
+
+typedef enum NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS {
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_MAIN = 0,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_SPILL = 1,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_PAGEPOOL = 2,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_BETACB = 3,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_RTV = 4,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL = 5,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL_CONTROL = 6,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL_CONTROL_CPU = 7,
+    NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_END = 8,
+} NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS;
+
+#define NV2080_GPU_MAX_NAME_STRING_LENGTH                  (0x0000040U)
+
+typedef struct VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS 
+{
+    NvU32 numHeads;
+    NvU32 maxNumHeads;
+} VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS;
+
+typedef struct VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS 
+{
+    NvU32 headIndex;
+    NvU32 maxHResolution;
+    NvU32 maxVResolution;
+} VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS;
+
+typedef struct GspStaticConfigInfo_t
+{
+    NvU8 grCapsBits[NV0080_CTRL_GR_CAPS_TBL_SIZE];
+    NV2080_CTRL_GPU_GET_GID_INFO_PARAMS gidInfo;
+    NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS gpcInfo;
+    NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS tpcInfo[MAX_GPC_COUNT];
+    NV2080_CTRL_GPU_GET_FERMI_ZCULL_INFO_PARAMS zcullInfo[MAX_GPC_COUNT];
+    NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS SKUInfo;
+    NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS fbRegionInfoParams;
+    COMPUTE_BRANDING_TYPE computeBranding;
+
+    NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS sriovCaps;
+    NvU32 sriovMaxGfid;
+
+    NvU32 engineCaps[NVGPU_ENGINE_CAPS_MASK_ARRAY_MAX];
+
+    GspSMInfo SM_info;
+
+    NvBool poisonFuseEnabled;
+
+    NvU64 fb_length;
+    NvU32 fbio_mask;
+    NvU32 fb_bus_width;
+    NvU32 fb_ram_type;
+    NvU32 fbp_mask;
+    NvU32 l2_cache_size;
+
+    NvU32 gfxpBufferSize[NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL];
+    NvU32 gfxpBufferAlignment[NV2080_CTRL_CMD_GR_CTXSW_PREEMPTION_BIND_BUFFERS_CONTEXT_POOL];
+
+    NvU8 gpuNameString[NV2080_GPU_MAX_NAME_STRING_LENGTH];
+    NvU8 gpuShortNameString[NV2080_GPU_MAX_NAME_STRING_LENGTH];
+    NvU16 gpuNameString_Unicode[NV2080_GPU_MAX_NAME_STRING_LENGTH];
+    NvBool bGpuInternalSku;
+    NvBool bIsQuadroGeneric;
+    NvBool bIsQuadroAd;
+    NvBool bIsNvidiaNvs;
+    NvBool bIsVgx;
+    NvBool bGeforceSmb;
+    NvBool bIsTitan;
+    NvBool bIsTesla;
+    NvBool bIsMobile;
+    NvBool bIsGc6Rtd3Allowed;
+    NvBool bIsGcOffRtd3Allowed;
+    NvBool bIsGcoffLegacyAllowed;
+
+    NvU64 bar1PdeBase;
+    NvU64 bar2PdeBase;
+
+    NvBool bVbiosValid;
+    NvU32 vbiosSubVendor;
+    NvU32 vbiosSubDevice;
+
+    NvBool bPageRetirementSupported;
+
+    NvBool bSplitVasBetweenServerClientRm;
+
+    NvBool bClRootportNeedsNosnoopWAR;
+
+    VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS displaylessMaxHeads;
+    VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS displaylessMaxResolution;
+    NvU64 displaylessMaxPixels;
+
+    // Client handle for internal RMAPI control.
+    NvHandle hInternalClient;
+
+    // Device handle for internal RMAPI control.
+    NvHandle hInternalDevice;
+
+    // Subdevice handle for internal RMAPI control.
+    NvHandle hInternalSubdevice;
+
+    NvBool bSelfHostedMode;
+    NvBool bAtsSupported;
+
+    NvBool bIsGpuUefi;
+} GspStaticConfigInfo;
+
+typedef struct rpc_unloading_guest_driver_v1F_07
+{
+    NvBool     bInPMTransition;
+    NvBool     bGc6Entering;
+    NvU32      newLevel;
+} rpc_unloading_guest_driver_v1F_07;
+
+typedef struct PACKED_REGISTRY_ENTRY
+{
+    NvU32                   nameOffset;
+    NvU8                    type;
+    NvU32                   data;
+    NvU32                   length;
+} PACKED_REGISTRY_ENTRY;
+
+typedef struct PACKED_REGISTRY_TABLE
+{
+    NvU32                   size;
+    NvU32                   numEntries;
+    PACKED_REGISTRY_ENTRY   entries[] __counted_by(numEntries);
+} PACKED_REGISTRY_TABLE;
+
+typedef struct
+{
+    NvU16               deviceID;           // deviceID
+    NvU16               vendorID;           // vendorID
+    NvU16               subdeviceID;        // subsystem deviceID
+    NvU16               subvendorID;        // subsystem vendorID
+    NvU8                revisionID;         // revision ID
+} BUSINFO;
+
+#define NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS             (16U)
+
+typedef struct DOD_METHOD_DATA
+{
+    NV_STATUS status;
+    NvU32     acpiIdListLen;
+    NvU32     acpiIdList[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
+} DOD_METHOD_DATA;
+
+typedef struct JT_METHOD_DATA
+{
+    NV_STATUS status;
+    NvU32     jtCaps;
+    NvU16     jtRevId;
+    NvBool    bSBIOSCaps;
+} JT_METHOD_DATA;
+
+typedef struct MUX_METHOD_DATA_ELEMENT
+{
+    NvU32       acpiId;
+    NvU32       mode;
+    NV_STATUS   status;
+} MUX_METHOD_DATA_ELEMENT;
+
+#define NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS             (16U)
+
+typedef struct MUX_METHOD_DATA
+{
+    NvU32                       tableLen;
+    MUX_METHOD_DATA_ELEMENT     acpiIdMuxModeTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
+    MUX_METHOD_DATA_ELEMENT     acpiIdMuxPartTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
+} MUX_METHOD_DATA;
+
+typedef struct CAPS_METHOD_DATA
+{
+    NV_STATUS status;
+    NvU32     optimusCaps;
+} CAPS_METHOD_DATA;
+
+typedef struct ACPI_METHOD_DATA
+{
+    NvBool                                               bValid;
+    DOD_METHOD_DATA                                      dodMethodData;
+    JT_METHOD_DATA                                       jtMethodData;
+    MUX_METHOD_DATA                                      muxMethodData;
+    CAPS_METHOD_DATA                                     capsMethodData;
+} ACPI_METHOD_DATA;
+
+typedef struct GSP_VF_INFO
+{
+    NvU32  totalVFs;
+    NvU32  firstVFOffset;
+    NvU64  FirstVFBar0Address;
+    NvU64  FirstVFBar1Address;
+    NvU64  FirstVFBar2Address;
+    NvBool b64bitBar0;
+    NvBool b64bitBar1;
+    NvBool b64bitBar2;
+} GSP_VF_INFO;
+
+typedef struct GspSystemInfo
+{
+    NvU64 gpuPhysAddr;
+    NvU64 gpuPhysFbAddr;
+    NvU64 gpuPhysInstAddr;
+    NvU64 nvDomainBusDeviceFunc;
+    NvU64 simAccessBufPhysAddr;
+    NvU64 pcieAtomicsOpMask;
+    NvU64 consoleMemSize;
+    NvU64 maxUserVa;
+    NvU32 pciConfigMirrorBase;
+    NvU32 pciConfigMirrorSize;
+    NvU8 oorArch;
+    NvU64 clPdbProperties;
+    NvU32 Chipset;
+    NvBool bGpuBehindBridge;
+    NvBool bMnocAvailable;
+    NvBool bUpstreamL0sUnsupported;
+    NvBool bUpstreamL1Unsupported;
+    NvBool bUpstreamL1PorSupported;
+    NvBool bUpstreamL1PorMobileOnly;
+    NvU8   upstreamAddressValid;
+    BUSINFO FHBBusInfo;
+    BUSINFO chipsetIDInfo;
+    ACPI_METHOD_DATA acpiMethodData;
+    NvU32 hypervisorType;
+    NvBool bIsPassthru;
+    NvU64 sysTimerOffsetNs;
+    GSP_VF_INFO gspVFInfo;
+} GspSystemInfo;
+
+typedef struct rpc_os_error_log_v17_00
+{
+    NvU32      exceptType;
+    NvU32      runlistId;
+    NvU32      chid;
+    char       errString[0x100];
+} rpc_os_error_log_v17_00;
+
+typedef struct rpc_run_cpu_sequencer_v17_00
+{
+    NvU32      bufferSizeDWord;
+    NvU32      cmdIndex;
+    NvU32      regSaveArea[8];
+    NvU32      commandBuffer[];
+} rpc_run_cpu_sequencer_v17_00;
+
+typedef enum GSP_SEQ_BUF_OPCODE
+{
+    GSP_SEQ_BUF_OPCODE_REG_WRITE = 0,
+    GSP_SEQ_BUF_OPCODE_REG_MODIFY,
+    GSP_SEQ_BUF_OPCODE_REG_POLL,
+    GSP_SEQ_BUF_OPCODE_DELAY_US,
+    GSP_SEQ_BUF_OPCODE_REG_STORE,
+    GSP_SEQ_BUF_OPCODE_CORE_RESET,
+    GSP_SEQ_BUF_OPCODE_CORE_START,
+    GSP_SEQ_BUF_OPCODE_CORE_WAIT_FOR_HALT,
+    GSP_SEQ_BUF_OPCODE_CORE_RESUME,
+} GSP_SEQ_BUF_OPCODE;
+
+typedef struct
+{
+    NvU32 addr;
+    NvU32 val;
+} GSP_SEQ_BUF_PAYLOAD_REG_WRITE;
+
+typedef struct
+{
+    NvU32 addr;
+    NvU32 mask;
+    NvU32 val;
+} GSP_SEQ_BUF_PAYLOAD_REG_MODIFY;
+
+typedef struct
+{
+    NvU32 addr;
+    NvU32 mask;
+    NvU32 val;
+    NvU32 timeout;
+    NvU32 error;
+} GSP_SEQ_BUF_PAYLOAD_REG_POLL;
+
+typedef struct
+{
+    NvU32 val;
+} GSP_SEQ_BUF_PAYLOAD_DELAY_US;
+
+typedef struct
+{
+    NvU32 addr;
+    NvU32 index;
+} GSP_SEQ_BUF_PAYLOAD_REG_STORE;
+
+typedef struct GSP_SEQUENCER_BUFFER_CMD
+{
+    GSP_SEQ_BUF_OPCODE opCode;
+    union
+    {
+        GSP_SEQ_BUF_PAYLOAD_REG_WRITE regWrite;
+        GSP_SEQ_BUF_PAYLOAD_REG_MODIFY regModify;
+        GSP_SEQ_BUF_PAYLOAD_REG_POLL regPoll;
+        GSP_SEQ_BUF_PAYLOAD_DELAY_US delayUs;
+        GSP_SEQ_BUF_PAYLOAD_REG_STORE regStore;
+    } payload;
+} GSP_SEQUENCER_BUFFER_CMD;
+
+typedef struct
+{
+    // Magic
+    // BL to use for verification (i.e. Booter locked it in WPR2)
+    NvU64 magic; // = 0xdc3aae21371a60b3;
+
+    // Revision number of Booter-BL-Sequencer handoff interface
+    // Bumped up when we change this interface so it is not backward compatible.
+    // Bumped up when we revoke GSP-RM ucode
+    NvU64 revision; // = 1;
+
+    // ---- Members regarding data in SYSMEM ----------------------------
+    // Consumed by Booter for DMA
+
+    NvU64 sysmemAddrOfRadix3Elf;
+    NvU64 sizeOfRadix3Elf;
+
+    NvU64 sysmemAddrOfBootloader;
+    NvU64 sizeOfBootloader;
+
+    // Offsets inside bootloader image needed by Booter
+    NvU64 bootloaderCodeOffset;
+    NvU64 bootloaderDataOffset;
+    NvU64 bootloaderManifestOffset;
+
+    union
+    {
+        // Used only at initial boot
+        struct
+        {
+            NvU64 sysmemAddrOfSignature;
+            NvU64 sizeOfSignature;
+        };
+
+        //
+        // Used at suspend/resume to read GspFwHeapFreeList
+        // Offset relative to GspFwWprMeta FBMEM PA (gspFwWprStart)
+        //
+        struct
+        {
+            NvU32 gspFwHeapFreeListWprOffset;
+            NvU32 unused0;
+            NvU64 unused1;
+        };
+    };
+
+    // ---- Members describing FB layout --------------------------------
+    NvU64 gspFwRsvdStart;
+
+    NvU64 nonWprHeapOffset;
+    NvU64 nonWprHeapSize;
+
+    NvU64 gspFwWprStart;
+
+    // GSP-RM to use to setup heap.
+    NvU64 gspFwHeapOffset;
+    NvU64 gspFwHeapSize;
+
+    // BL to use to find ELF for jump
+    NvU64 gspFwOffset;
+    // Size is sizeOfRadix3Elf above.
+
+    NvU64 bootBinOffset;
+    // Size is sizeOfBootloader above.
+
+    NvU64 frtsOffset;
+    NvU64 frtsSize;
+
+    NvU64 gspFwWprEnd;
+
+    // GSP-RM to use for fbRegionInfo?
+    NvU64 fbSize;
+
+    // ---- Other members -----------------------------------------------
+
+    // GSP-RM to use for fbRegionInfo?
+    NvU64 vgaWorkspaceOffset;
+    NvU64 vgaWorkspaceSize;
+
+    // Boot count.  Used to determine whether to load the firmware image.
+    NvU64 bootCount;
+
+    // This union is organized the way it is to start at an 8-byte boundary and achieve natural
+    // packing of the internal struct fields.
+    union
+    {
+        struct
+        {
+            // TODO: the partitionRpc* fields below do not really belong in this
+            //       structure. The values are patched in by the partition bootstrapper
+            //       when GSP-RM is booted in a partition, and this structure was a
+            //       convenient place for the bootstrapper to access them. These should
+            //       be moved to a different comm. mechanism between the bootstrapper
+            //       and the GSP-RM tasks.
+
+            // Shared partition RPC memory (physical address)
+            NvU64 partitionRpcAddr;
+
+            // Offsets relative to partitionRpcAddr
+            NvU16 partitionRpcRequestOffset;
+            NvU16 partitionRpcReplyOffset;
+
+            // Code section and dataSection offset and size.
+            NvU32 elfCodeOffset;
+            NvU32 elfDataOffset;
+            NvU32 elfCodeSize;
+            NvU32 elfDataSize;
+
+            // Used during GSP-RM resume to check for revocation
+            NvU32 lsUcodeVersion;
+        };
+
+        struct
+        {
+            // Pad for the partitionRpc* fields, plus 4 bytes
+            NvU32 partitionRpcPadding[4];
+
+            // CrashCat (contiguous) buffer size/location - occupies same bytes as the
+            // elf(Code|Data)(Offset|Size) fields above.
+            // TODO: move to GSP_FMC_INIT_PARAMS
+            NvU64 sysmemAddrOfCrashReportQueue;
+            NvU32 sizeOfCrashReportQueue;
+
+            // Pad for the lsUcodeVersion field
+            NvU32 lsUcodeVersionPadding[1];
+        };
+    };
+
+    // Number of VF partitions allocating sub-heaps from the WPR heap
+    // Used during boot to ensure the heap is adequately sized
+    NvU8 gspFwHeapVfPartitionCount;
+
+    // Pad structure to exactly 256 bytes.  Can replace padding with additional
+    // fields without incrementing revision.  Padding initialized to 0.
+    NvU8 padding[7];
+
+    // BL to use for verification (i.e. Booter says OK to boot)
+    NvU64 verified;  // 0x0 -> unverified, 0xa0a0a0a0a0a0a0a0 -> verified
+} GspFwWprMeta;
+
+#define GSP_FW_WPR_META_MAGIC     0xdc3aae21371a60b3ULL
+
+#define GSP_FW_WPR_META_REVISION  1
+
+typedef struct
+{
+    NvU32 version;   // queue version
+    NvU32 size;      // bytes, page aligned
+    NvU32 msgSize;   // entry size, bytes, must be power-of-2, 16 is minimum
+    NvU32 msgCount;  // number of entries in queue
+    NvU32 writePtr;  // message id of next slot
+    NvU32 flags;     // if set it means "i want to swap RX"
+    NvU32 rxHdrOff;  // Offset of msgqRxHeader from start of backing store.
+    NvU32 entryOff;  // Offset of entries from start of backing store.
+} msgqTxHeader;
+
+typedef struct
+{
+    NvU32 readPtr; // message id of last message read
+} msgqRxHeader;
+
+typedef struct {
+    RmPhysAddr sharedMemPhysAddr;
+    NvU32 pageTableEntryCount;
+    NvLength cmdQueueOffset;
+    NvLength statQueueOffset;
+    NvLength locklessCmdQueueOffset;
+    NvLength locklessStatQueueOffset;
+} MESSAGE_QUEUE_INIT_ARGUMENTS;
+
+typedef struct {
+    NvU32 oldLevel;
+    NvU32 flags;
+    NvBool bInPMTransition;
+} GSP_SR_INIT_ARGUMENTS;
+
+typedef struct
+{
+    MESSAGE_QUEUE_INIT_ARGUMENTS      messageQueueInitArguments;
+    GSP_SR_INIT_ARGUMENTS             srInitArguments;
+    NvU32                             gpuInstance;
+
+    struct
+    {
+        NvU64                         pa;
+        NvU64                         size;
+    } profilerArgs;
+} GSP_ARGUMENTS_CACHED;
+
+#define NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_0            (0x00000000U)
+
+#define NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_3            (0x00000003U)
+
+typedef NvU64 LibosAddress;
+
+typedef struct
+{
+    LibosAddress          id8;  // Id tag.
+    LibosAddress          pa;   // Physical address.
+    LibosAddress          size; // Size of memory area.
+    NvU8                  kind; // See LibosMemoryRegionKind above.
+    NvU8                  loc;  // See LibosMemoryRegionLoc above.
+} LibosMemoryRegionInitArgument;
+
+typedef enum {
+    LIBOS_MEMORY_REGION_NONE,
+    LIBOS_MEMORY_REGION_CONTIGUOUS,
+    LIBOS_MEMORY_REGION_RADIX3
+} LibosMemoryRegionKind;
+
+typedef enum {
+    LIBOS_MEMORY_REGION_LOC_NONE,
+    LIBOS_MEMORY_REGION_LOC_SYSMEM,
+    LIBOS_MEMORY_REGION_LOC_FB
+} LibosMemoryRegionLoc;
+
+typedef struct
+{
+    //
+    // Magic
+    // Use for verification by Booter
+    //
+    NvU64 magic;  // = GSP_FW_SR_META_MAGIC;
+
+    //
+    // Revision number
+    // Bumped up when we change this interface so it is not backward compatible.
+    // Bumped up when we revoke GSP-RM ucode
+    //
+    NvU64 revision;  // = GSP_FW_SR_META_MAGIC_REVISION;
+
+    //
+    // ---- Members regarding data in SYSMEM ----------------------------
+    // Consumed by Booter for DMA
+    //
+    NvU64 sysmemAddrOfSuspendResumeData;
+    NvU64 sizeOfSuspendResumeData;
+
+    // ---- Members for crypto ops across S/R ---------------------------
+
+    //
+    // HMAC over the entire GspFwSRMeta structure (including padding)
+    // with the hmac field itself zeroed.
+    //
+    NvU8 hmac[32];
+
+    // Hash over GspFwWprMeta structure
+    NvU8 wprMetaHash[32];
+
+    // Hash over GspFwHeapFreeList structure. All zeros signifies no free list.
+    NvU8 heapFreeListHash[32];
+
+    // Hash over data in WPR2 (skipping over free heap chunks; see Booter for details)
+    NvU8 dataHash[32];
+
+    //
+    // Pad structure to exactly 256 bytes (1 DMA chunk).
+    // Padding initialized to zero.
+    //
+    NvU32 padding[24];
+
+} GspFwSRMeta;
+
+#define GSP_FW_SR_META_MAGIC     0x8a3bb9e6c6c39d93ULL
+
+#define GSP_FW_SR_META_REVISION  2
+
+#define GSP_SEQUENCER_PAYLOAD_SIZE_DWORDS(opcode)                       \
+    ((opcode == GSP_SEQ_BUF_OPCODE_REG_WRITE)  ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_WRITE)  / sizeof(NvU32)) : \
+     (opcode == GSP_SEQ_BUF_OPCODE_REG_MODIFY) ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_MODIFY) / sizeof(NvU32)) : \
+     (opcode == GSP_SEQ_BUF_OPCODE_REG_POLL)   ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_POLL)   / sizeof(NvU32)) : \
+     (opcode == GSP_SEQ_BUF_OPCODE_DELAY_US)   ? (sizeof(GSP_SEQ_BUF_PAYLOAD_DELAY_US)   / sizeof(NvU32)) : \
+     (opcode == GSP_SEQ_BUF_OPCODE_REG_STORE)  ? (sizeof(GSP_SEQ_BUF_PAYLOAD_REG_STORE)  / sizeof(NvU32)) : \
+    /* GSP_SEQ_BUF_OPCODE_CORE_RESET */                                 \
+    /* GSP_SEQ_BUF_OPCODE_CORE_START */                                 \
+    /* GSP_SEQ_BUF_OPCODE_CORE_WAIT_FOR_HALT */                         \
+    /* GSP_SEQ_BUF_OPCODE_CORE_RESUME */                                \
+    0)
+
+typedef struct {
+    //
+    // Version 1
+    // Version 2
+    // Version 3 = for Partition boot
+    // Version 4 = for eb riscv boot
+    // Version 5 = Support signing entire RISC-V image as "code" in code section for hopper and later.
+    //
+    NvU32  version;                         // structure version
+    NvU32  bootloaderOffset;
+    NvU32  bootloaderSize;
+    NvU32  bootloaderParamOffset;
+    NvU32  bootloaderParamSize;
+    NvU32  riscvElfOffset;
+    NvU32  riscvElfSize;
+    NvU32  appVersion;                      // Changelist number associated with the image
+    //
+    // Manifest contains information about Monitor and it is
+    // input to BR
+    //
+    NvU32  manifestOffset;
+    NvU32  manifestSize;
+    //
+    // Monitor Data offset within RISCV image and size
+    //
+    NvU32  monitorDataOffset;
+    NvU32  monitorDataSize;
+    //
+    // Monitor Code offset withtin RISCV image and size
+    //
+    NvU32  monitorCodeOffset;
+    NvU32  monitorCodeSize;
+    NvU32  bIsMonitorEnabled;
+    //
+    // Swbrom Code offset within RISCV image and size
+    //
+    NvU32  swbromCodeOffset;
+    NvU32  swbromCodeSize;
+    //
+    // Swbrom Data offset within RISCV image and size
+    //
+    NvU32  swbromDataOffset;
+    NvU32  swbromDataSize;
+    //
+    // Total size of FB carveout (image and reserved space).  
+    //
+    NvU32  fbReservedSize;
+    //
+    // Indicates whether the entire RISC-V image is signed as "code" in code section.
+    //
+    NvU32  bSignedAsCode;
+} RM_RISCV_UCODE_DESC;
+
+typedef struct NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_ENTRY {
+    NvU16 engineIdx;
+    NvU32 pmcIntrMask;
+    NvU32 vectorStall;
+    NvU32 vectorNonStall;
+} NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_ENTRY;
+
+typedef struct NV2080_INTR_CATEGORY_SUBTREE_MAP {
+    NvU8 subtreeStart;
+    NvU8 subtreeEnd;
+} NV2080_INTR_CATEGORY_SUBTREE_MAP;
+
+#define NV2080_CTRL_INTERNAL_INTR_MAX_TABLE_SIZE       128
+
+typedef enum NV2080_INTR_CATEGORY {
+    NV2080_INTR_CATEGORY_DEFAULT = 0,
+    NV2080_INTR_CATEGORY_ESCHED_DRIVEN_ENGINE = 1,
+    NV2080_INTR_CATEGORY_ESCHED_DRIVEN_ENGINE_NOTIFICATION = 2,
+    NV2080_INTR_CATEGORY_RUNLIST = 3,
+    NV2080_INTR_CATEGORY_RUNLIST_NOTIFICATION = 4,
+    NV2080_INTR_CATEGORY_UVM_OWNED = 5,
+    NV2080_INTR_CATEGORY_UVM_SHARED = 6,
+    NV2080_INTR_CATEGORY_ENUM_COUNT = 7,
+} NV2080_INTR_CATEGORY;
+
+#define NV2080_CTRL_CMD_INTERNAL_INTR_GET_KERNEL_TABLE (0x20800a5c) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS {
+    NvU32                                            tableLen;
+    NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_ENTRY table[NV2080_CTRL_INTERNAL_INTR_MAX_TABLE_SIZE];
+    NV2080_INTR_CATEGORY_SUBTREE_MAP                 subtreeMap[NV2080_INTR_CATEGORY_ENUM_COUNT];
+} NV2080_CTRL_INTERNAL_INTR_GET_KERNEL_TABLE_PARAMS;
+
+#define GSP_FW_HEAP_PARAM_SIZE_PER_GB_FB                  (96 << 10)   // All architectures
+
+#define GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE      ((48 << 10) * 2048)   // Support 2048 channels
+
+typedef union rpc_message_rpc_union_field_v03_00
+{
+    NvU32      spare;
+    NvU32      cpuRmGfid;
+} rpc_message_rpc_union_field_v03_00;
+
+typedef rpc_message_rpc_union_field_v03_00 rpc_message_rpc_union_field_v;
+
+typedef struct rpc_message_header_v03_00
+{
+    NvU32      header_version;
+    NvU32      signature;
+    NvU32      length;
+    NvU32      function;
+    NvU32      rpc_result;
+    NvU32      rpc_result_private;
+    NvU32      sequence;
+    rpc_message_rpc_union_field_v u;
+    rpc_generic_union rpc_message_data[];
+} rpc_message_header_v03_00;
+
+typedef rpc_message_header_v03_00 rpc_message_header_v;
+
+typedef struct GSP_MSG_QUEUE_ELEMENT
+{
+    NvU8  authTagBuffer[16];         // Authentication tag buffer.
+    NvU8  aadBuffer[16];             // AAD buffer.
+    NvU32 checkSum;                  // Set to value needed to make checksum always zero.
+    NvU32 seqNum;                    // Sequence number maintained by the message queue.
+    NvU32 elemCount;                 // Number of message queue elements this message has.
+    NV_DECLARE_ALIGNED(rpc_message_header_v rpc, 8);
+} GSP_MSG_QUEUE_ELEMENT;
+
+#define GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS2                   (0 << 20)   // No FB heap usage
+#define GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3                  (20 << 20)
+
+#define GSP_FW_HEAP_PARAM_BASE_RM_SIZE_TU10X               (8 << 20)   // Turing thru Ada
+
+#define GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS2_MIN_MB                (64u)
+#define GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB      (84u)
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/msgfn.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/msgfn.h
new file mode 100644
index 000000000000..642c13aec325
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/msgfn.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_MSGFN_H__
+#define __NVRM_MSGFN_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#ifndef E
+#    define E(RPC) NV_VGPU_MSG_EVENT_##RPC,
+#    define DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
+enum {
+#endif
+    E(FIRST_EVENT = 0x1000)                      // 0x1000
+    E(GSP_INIT_DONE)                             // 0x1001
+    E(GSP_RUN_CPU_SEQUENCER)                     // 0x1002
+    E(POST_EVENT)                                // 0x1003
+    E(RC_TRIGGERED)                              // 0x1004
+    E(MMU_FAULT_QUEUED)                          // 0x1005
+    E(OS_ERROR_LOG)                              // 0x1006
+    E(RG_LINE_INTR)                              // 0x1007
+    E(GPUACCT_PERFMON_UTIL_SAMPLES)              // 0x1008
+    E(SIM_READ)                                  // 0x1009
+    E(SIM_WRITE)                                 // 0x100a
+    E(SEMAPHORE_SCHEDULE_CALLBACK)               // 0x100b
+    E(UCODE_LIBOS_PRINT)                         // 0x100c
+    E(VGPU_GSP_PLUGIN_TRIGGERED)                 // 0x100d
+    E(PERF_GPU_BOOST_SYNC_LIMITS_CALLBACK)       // 0x100e
+    E(PERF_BRIDGELESS_INFO_UPDATE)               // 0x100f
+    E(VGPU_CONFIG)                               // 0x1010
+    E(DISPLAY_MODESET)                           // 0x1011
+    E(EXTDEV_INTR_SERVICE)                       // 0x1012
+    E(NVLINK_INBAND_RECEIVED_DATA_256)           // 0x1013
+    E(NVLINK_INBAND_RECEIVED_DATA_512)           // 0x1014
+    E(NVLINK_INBAND_RECEIVED_DATA_1024)          // 0x1015
+    E(NVLINK_INBAND_RECEIVED_DATA_2048)          // 0x1016
+    E(NVLINK_INBAND_RECEIVED_DATA_4096)          // 0x1017
+    E(TIMED_SEMAPHORE_RELEASE)                   // 0x1018
+    E(NVLINK_IS_GPU_DEGRADED)                    // 0x1019
+    E(PFM_REQ_HNDLR_STATE_SYNC_CALLBACK)         // 0x101a
+    E(GSP_SEND_USER_SHARED_DATA)                 // 0x101b
+    E(NVLINK_FAULT_UP)                           // 0x101c
+    E(GSP_LOCKDOWN_NOTICE)                       // 0x101d
+    E(MIG_CI_CONFIG_UPDATE)                      // 0x101e
+    E(NUM_EVENTS)                                // END
+#ifdef DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
+};
+#   undef E
+#   undef DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
+#endif
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvdec.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvdec.h
new file mode 100644
index 000000000000..3a04e702677f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvdec.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_NVDEC_H__
+#define __NVRM_NVDEC_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct
+{
+    NvU32 size;
+    NvU32 prohibitMultipleInstances;
+    NvU32 engineInstance;               // Select NVDEC0 or NVDEC1 or NVDEC2
+} NV_BSP_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvenc.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvenc.h
new file mode 100644
index 000000000000..203c1d5304d9
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvenc.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_NVENC_H__
+#define __NVRM_NVENC_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct
+{
+    NvU32 size;
+    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of MSENC?
+    NvU32 engineInstance;             // Select MSENC/NVENC0 or NVENC1 or NVENC2
+} NV_MSENC_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvjpg.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvjpg.h
new file mode 100644
index 000000000000..71fc53889ec7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/nvjpg.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_NVJPG_H__
+#define __NVRM_NVJPG_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct
+{
+    NvU32 size;
+    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of NVJPG?
+    NvU32 engineInstance;
+} NV_NVJPG_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ofa.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ofa.h
new file mode 100644
index 000000000000..49d81c7673d2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/ofa.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_OFA_H__
+#define __NVRM_OFA_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+typedef struct
+{
+    NvU32 size;
+    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of OFA?
+} NV_OFA_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_global_enums.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/rpcfn.h
index 73c57f235f6a..2a037acc6b1e 100644
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/nvidia/kernel/inc/vgpu/rpc_global_enums.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/rpcfn.h
@@ -1,5 +1,10 @@
-#ifndef __src_nvidia_kernel_inc_vgpu_rpc_global_enums_h__
-#define __src_nvidia_kernel_inc_vgpu_rpc_global_enums_h__
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_RPCFN_H__
+#define __NVRM_RPCFN_H__
+#include <nvrm/nvtypes.h>
 
 /* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
 
@@ -209,54 +214,12 @@ enum {
     X(RM, CTRL_SET_HS_CREDITS)                  // 198
     X(RM, CTRL_PM_AREA_PC_SAMPLER)              // 199
     X(RM, INVALIDATE_TLB)                       // 200
+    X(RM, RESERVED_201)                         // 201
+    X(RM, ECC_NOTIFIER_WRITE_ACK)               // 202
     X(RM, NUM_FUNCTIONS)                        //END
 #ifdef DEFINING_X_IN_RPC_GLOBAL_ENUMS_H
 };
 #   undef X
 #   undef DEFINING_X_IN_RPC_GLOBAL_ENUMS_H
 #endif
-
-#ifndef E
-#    define E(RPC) NV_VGPU_MSG_EVENT_##RPC,
-#    define DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
-enum {
-#endif
-    E(FIRST_EVENT = 0x1000)                      // 0x1000
-    E(GSP_INIT_DONE)                             // 0x1001
-    E(GSP_RUN_CPU_SEQUENCER)                     // 0x1002
-    E(POST_EVENT)                                // 0x1003
-    E(RC_TRIGGERED)                              // 0x1004
-    E(MMU_FAULT_QUEUED)                          // 0x1005
-    E(OS_ERROR_LOG)                              // 0x1006
-    E(RG_LINE_INTR)                              // 0x1007
-    E(GPUACCT_PERFMON_UTIL_SAMPLES)              // 0x1008
-    E(SIM_READ)                                  // 0x1009
-    E(SIM_WRITE)                                 // 0x100a
-    E(SEMAPHORE_SCHEDULE_CALLBACK)               // 0x100b
-    E(UCODE_LIBOS_PRINT)                         // 0x100c
-    E(VGPU_GSP_PLUGIN_TRIGGERED)                 // 0x100d
-    E(PERF_GPU_BOOST_SYNC_LIMITS_CALLBACK)       // 0x100e
-    E(PERF_BRIDGELESS_INFO_UPDATE)               // 0x100f
-    E(VGPU_CONFIG)                               // 0x1010
-    E(DISPLAY_MODESET)                           // 0x1011
-    E(EXTDEV_INTR_SERVICE)                       // 0x1012
-    E(NVLINK_INBAND_RECEIVED_DATA_256)           // 0x1013
-    E(NVLINK_INBAND_RECEIVED_DATA_512)           // 0x1014
-    E(NVLINK_INBAND_RECEIVED_DATA_1024)          // 0x1015
-    E(NVLINK_INBAND_RECEIVED_DATA_2048)          // 0x1016
-    E(NVLINK_INBAND_RECEIVED_DATA_4096)          // 0x1017
-    E(TIMED_SEMAPHORE_RELEASE)                   // 0x1018
-    E(NVLINK_IS_GPU_DEGRADED)                    // 0x1019
-    E(PFM_REQ_HNDLR_STATE_SYNC_CALLBACK)         // 0x101a
-    E(GSP_SEND_USER_SHARED_DATA)                 // 0x101b
-    E(NVLINK_FAULT_UP)                           // 0x101c
-    E(GSP_LOCKDOWN_NOTICE)                       // 0x101d
-    E(MIG_CI_CONFIG_UPDATE)                      // 0x101e
-    E(NUM_EVENTS)                                // END
-#ifdef DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
-};
-#   undef E
-#   undef DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
-#endif
-
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/vmm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/vmm.h
new file mode 100644
index 000000000000..f6ec04efd119
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/nvrm/vmm.h
@@ -0,0 +1,132 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_VMM_H__
+#define __NVRM_VMM_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
+
+#define FERMI_VASPACE_A                                     (0x000090f1)
+
+typedef struct
+{
+    NvU32   index;
+    NvV32   flags;
+    NvU64   vaSize NV_ALIGN_BYTES(8);
+    NvU64   vaStartInternal NV_ALIGN_BYTES(8);
+    NvU64   vaLimitInternal NV_ALIGN_BYTES(8);
+    NvU32   bigPageSize;
+    NvU64   vaBase NV_ALIGN_BYTES(8);
+} NV_VASPACE_ALLOCATION_PARAMETERS;
+
+#define NV_VASPACE_ALLOCATION_INDEX_GPU_NEW                                 0x00 //<! Create new VASpace, by default
+
+#define NV_VASPACE_ALLOCATION_FLAGS_IS_EXTERNALLY_OWNED                   BIT(3)
+
+#define SPLIT_VAS_SERVER_RM_MANAGED_VA_START   0x100000000ULL  // 4GB
+#define SPLIT_VAS_SERVER_RM_MANAGED_VA_SIZE     0x20000000ULL  // 512MB
+
+#define GMMU_FMT_MAX_LEVELS  6U
+
+#define NV90F1_CTRL_CMD_VASPACE_COPY_SERVER_RESERVED_PDES (0x90f10106U) /* finn: Evaluated from "(FINN_FERMI_VASPACE_A_VASPACE_INTERFACE_ID << 8) | NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS_MESSAGE_ID" */
+typedef struct NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS {
+    /*!
+     * [in] GPU sub-device handle - this API only supports unicast.
+     *      Pass 0 to use subDeviceId instead.
+     */
+    NvHandle hSubDevice;
+
+    /*!
+     * [in] GPU sub-device ID. Ignored if hSubDevice is non-zero.
+     */
+    NvU32    subDeviceId;
+
+    /*!
+     * [in] Page size (VA coverage) of the level to reserve.
+     *      This need not be a leaf (page table) page size - it can be
+     *      the coverage of an arbitrary level (including root page directory).
+     */
+    NV_DECLARE_ALIGNED(NvU64 pageSize, 8);
+
+    /*!
+     * [in] First GPU virtual address of the range to reserve.
+     *      This must be aligned to pageSize.
+     */
+    NV_DECLARE_ALIGNED(NvU64 virtAddrLo, 8);
+
+    /*!
+     * [in] Last GPU virtual address of the range to reserve.
+     *      This (+1) must be aligned to pageSize.
+     */
+    NV_DECLARE_ALIGNED(NvU64 virtAddrHi, 8);
+
+    /*! 
+     * [in] Number of PDE levels to copy.
+     */
+    NvU32    numLevelsToCopy;
+
+   /*!
+     * [in] Per-level information.
+     */
+    struct {
+        /*!
+         * Physical address of this page level instance.
+         */
+        NV_DECLARE_ALIGNED(NvU64 physAddress, 8);
+
+        /*!
+         * Size in bytes allocated for this level instance.
+         */
+        NV_DECLARE_ALIGNED(NvU64 size, 8);
+
+        /*!
+         * Aperture in which this page level instance resides.
+         */
+        NvU32 aperture;
+
+        /*!
+         * Page shift corresponding to the level
+         */
+        NvU8  pageShift;
+    } levels[GMMU_FMT_MAX_LEVELS];
+} NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS;
+
+#define NV0080_CTRL_CMD_DMA_SET_PAGE_DIRECTORY (0x801813U) /* finn: Evaluated from "(FINN_NV01_DEVICE_0_DMA_INTERFACE_
+ID << 8) | NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_PARAMS {
+    NV_DECLARE_ALIGNED(NvU64 physAddress, 8);
+    NvU32    numEntries;
+    NvU32    flags;
+    NvHandle hVASpace;
+    NvU32    chId;
+    NvU32    subDeviceId; // ID+1, 0 for BC
+    NvU32    pasid;
+} NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_PARAMS;
+
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_APERTURE                  1:0
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_APERTURE_VIDMEM           (0x00000000U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_APERTURE_SYSMEM_COH       (0x00000001U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_APERTURE_SYSMEM_NONCOH    (0x00000002U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_PRESERVE_PDES             2:2
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_PRESERVE_PDES_FALSE       (0x00000000U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_PRESERVE_PDES_TRUE        (0x00000001U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_ALL_CHANNELS              3:3
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_ALL_CHANNELS_FALSE        (0x00000000U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_ALL_CHANNELS_TRUE         (0x00000001U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_IGNORE_CHANNEL_BUSY       4:4
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_IGNORE_CHANNEL_BUSY_FALSE (0x00000000U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_IGNORE_CHANNEL_BUSY_TRUE  (0x00000001U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_EXTEND_VASPACE            5:5
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_EXTEND_VASPACE_FALSE      (0x00000000U)
+#define NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_FLAGS_EXTEND_VASPACE_TRUE       (0x00000001U)
+
+#define NV0080_CTRL_CMD_DMA_UNSET_PAGE_DIRECTORY                           (0x801814U) /* finn: Evaluated from "(FINN_NV01_DEVICE_0_DMA_INTERFACE_ID << 8) | NV0080_CTRL_DMA_UNSET_PAGE_DIRECTORY_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0080_CTRL_DMA_UNSET_PAGE_DIRECTORY_PARAMS {
+    NvHandle hVASpace;
+    NvU32    subDeviceId; // ID+1, 0 for BC
+} NV0080_CTRL_DMA_UNSET_PAGE_DIRECTORY_PARAMS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ad102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ofa.c
index d72b3aae9a2b..2156808cba4f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/ad102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/ofa.c
@@ -19,26 +19,26 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "priv.h"
+#include <rm/engine.h>
 
-#include <subdev/gsp.h>
+#include "nvrm/ofa.h"
 
-#include <nvif/class.h>
+static int
+r535_ofa_alloc(struct nvkm_gsp_object *chan, u32 handle, u32 class, int inst,
+	       struct nvkm_gsp_object *ofa)
+{
+	NV_OFA_ALLOCATION_PARAMETERS *args;
 
-static const struct nvkm_engine_func
-ad102_nvdec = {
-	.sclass = {
-		{ -1, -1, NVC9B0_VIDEO_DECODER },
-		{}
-	}
-};
+	args = nvkm_gsp_rm_alloc_get(chan, handle, class, sizeof(*args), ofa);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
 
-int
-ad102_nvdec_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
-		struct nvkm_nvdec **pnvdec)
-{
-	if (nvkm_gsp_rm(device->gsp))
-		return r535_nvdec_new(&ad102_nvdec, device, type, inst, pnvdec);
+	args->size = sizeof(*args);
 
-	return -ENODEV;
+	return nvkm_gsp_rm_alloc_wr(ofa, args);
 }
+
+const struct nvkm_rm_api_engine
+r535_ofa = {
+	.alloc = r535_ofa_alloc,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rm.c
new file mode 100644
index 000000000000..a4190676e1ad
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rm.c
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/rm.h>
+
+#include "nvrm/gsp.h"
+
+static const struct nvkm_rm_wpr
+r535_wpr_libos2 = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS2,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_TU10X,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS2_MIN_MB,
+};
+
+static const struct nvkm_rm_wpr
+r535_wpr_libos3 = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_TU10X,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB,
+};
+
+static const struct nvkm_rm_api
+r535_api = {
+	.gsp = &r535_gsp,
+	.rpc = &r535_rpc,
+	.ctrl = &r535_ctrl,
+	.alloc = &r535_alloc,
+	.client = &r535_client,
+	.device = &r535_device,
+	.fbsr = &r535_fbsr,
+	.disp = &r535_disp,
+	.fifo = &r535_fifo,
+	.ce = &r535_ce,
+	.gr = &r535_gr,
+	.nvdec = &r535_nvdec,
+	.nvenc = &r535_nvenc,
+	.nvjpg = &r535_nvjpg,
+	.ofa = &r535_ofa,
+};
+
+const struct nvkm_rm_impl
+r535_rm_tu102 = {
+	.wpr = &r535_wpr_libos2,
+	.api = &r535_api,
+};
+
+const struct nvkm_rm_impl
+r535_rm_ga102 = {
+	.wpr = &r535_wpr_libos3,
+	.api = &r535_api,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c
new file mode 100644
index 000000000000..5acb98d137bd
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c
@@ -0,0 +1,691 @@
+/*
+ * Copyright 2023 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <rm/rpc.h>
+
+#include "nvrm/rpcfn.h"
+
+#define GSP_MSG_MIN_SIZE GSP_PAGE_SIZE
+#define GSP_MSG_MAX_SIZE (GSP_MSG_MIN_SIZE * 16)
+
+/**
+ * DOC: GSP message queue element
+ *
+ * https://github.com/NVIDIA/open-gpu-kernel-modules/blob/535/src/nvidia/inc/kernel/gpu/gsp/message_queue_priv.h
+ *
+ * The GSP command queue and status queue are message queues for the
+ * communication between software and GSP. The software submits the GSP
+ * RPC via the GSP command queue, GSP writes the status of the submitted
+ * RPC in the status queue.
+ *
+ * A GSP message queue element consists of three parts:
+ *
+ * - message element header (struct r535_gsp_msg), which mostly maintains
+ *   the metadata for queuing the element.
+ *
+ * - RPC message header (struct nvfw_gsp_rpc), which maintains the info
+ *   of the RPC. E.g., the RPC function number.
+ *
+ * - The payload, where the RPC message stays. E.g. the params of a
+ *   specific RPC function. Some RPC functions also have their headers
+ *   in the payload. E.g. rm_alloc, rm_control.
+ *
+ * The memory layout of a GSP message element can be illustrated below::
+ *
+ *    +------------------------+
+ *    | Message Element Header |
+ *    |    (r535_gsp_msg)      |
+ *    |                        |
+ *    | (r535_gsp_msg.data)    |
+ *    |          |             |
+ *    |----------V-------------|
+ *    |    GSP RPC Header      |
+ *    |    (nvfw_gsp_rpc)      |
+ *    |                        |
+ *    | (nvfw_gsp_rpc.data)    |
+ *    |          |             |
+ *    |----------V-------------|
+ *    |       Payload          |
+ *    |                        |
+ *    |   header(optional)     |
+ *    |        params          |
+ *    +------------------------+
+ *
+ * The max size of a message queue element is 16 pages (including the
+ * headers). When a GSP message to be sent is larger than 16 pages, the
+ * message should be split into multiple elements and sent accordingly.
+ *
+ * In the bunch of the split elements, the first element has the expected
+ * function number, while the rest of the elements are sent with the
+ * function number NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD.
+ *
+ * GSP consumes the elements from the cmdq and always writes the result
+ * back to the msgq. The result is also formed as split elements.
+ *
+ * Terminology:
+ *
+ * - gsp_msg(msg): GSP message element (element header + GSP RPC header +
+ *   payload)
+ * - gsp_rpc(rpc): GSP RPC (RPC header + payload)
+ * - gsp_rpc_buf: buffer for (GSP RPC header + payload)
+ * - gsp_rpc_len: size of (GSP RPC header + payload)
+ * - params_size: size of params in the payload
+ * - payload_size: size of (header if exists + params) in the payload
+ */
+
+struct r535_gsp_msg {
+	u8 auth_tag_buffer[16];
+	u8 aad_buffer[16];
+	u32 checksum;
+	u32 sequence;
+	u32 elem_count;
+	u32 pad;
+	u8  data[];
+};
+
+struct nvfw_gsp_rpc {
+	u32 header_version;
+	u32 signature;
+	u32 length;
+	u32 function;
+	u32 rpc_result;
+	u32 rpc_result_private;
+	u32 sequence;
+	union {
+		u32 spare;
+		u32 cpuRmGfid;
+	};
+	u8  data[];
+};
+
+#define GSP_MSG_HDR_SIZE offsetof(struct r535_gsp_msg, data)
+
+#define to_gsp_hdr(p, header) \
+	container_of((void *)p, typeof(*header), data)
+
+#define to_payload_hdr(p, header) \
+	container_of((void *)p, typeof(*header), params)
+
+int
+r535_rpc_status_to_errno(uint32_t rpc_status)
+{
+	switch (rpc_status) {
+	case 0x55: /* NV_ERR_NOT_READY */
+	case 0x66: /* NV_ERR_TIMEOUT_RETRY */
+		return -EBUSY;
+	case 0x51: /* NV_ERR_NO_MEMORY */
+		return -ENOMEM;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int
+r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime)
+{
+	u32 size, rptr = *gsp->msgq.rptr;
+	int used;
+
+	size = DIV_ROUND_UP(GSP_MSG_HDR_SIZE + gsp_rpc_len,
+			    GSP_PAGE_SIZE);
+	if (WARN_ON(!size || size >= gsp->msgq.cnt))
+		return -EINVAL;
+
+	do {
+		u32 wptr = *gsp->msgq.wptr;
+
+		used = wptr + gsp->msgq.cnt - rptr;
+		if (used >= gsp->msgq.cnt)
+			used -= gsp->msgq.cnt;
+		if (used >= size)
+			break;
+
+		usleep_range(1, 2);
+	} while (--(*ptime));
+
+	if (WARN_ON(!*ptime))
+		return -ETIMEDOUT;
+
+	return used;
+}
+
+static struct r535_gsp_msg *
+r535_gsp_msgq_get_entry(struct nvkm_gsp *gsp)
+{
+	u32 rptr = *gsp->msgq.rptr;
+
+	/* Skip the first page, which is the message queue info */
+	return (void *)((u8 *)gsp->shm.msgq.ptr + GSP_PAGE_SIZE +
+	       rptr * GSP_PAGE_SIZE);
+}
+
+/**
+ * DOC: Receive a GSP message queue element
+ *
+ * Receiving a GSP message queue element from the message queue consists of
+ * the following steps:
+ *
+ * - Peek the element from the queue: r535_gsp_msgq_peek().
+ *   Peek the first page of the element to determine the total size of the
+ *   message before allocating the proper memory.
+ *
+ * - Allocate memory for the message.
+ *   Once the total size of the message is determined from the GSP message
+ *   queue element, the caller of r535_gsp_msgq_recv() allocates the
+ *   required memory.
+ *
+ * - Receive the message: r535_gsp_msgq_recv().
+ *   Copy the message into the allocated memory. Advance the read pointer.
+ *   If the message is a large GSP message, r535_gsp_msgq_recv() calls
+ *   r535_gsp_msgq_recv_one_elem() repeatedly to receive continuation parts
+ *   until the complete message is received.
+ *   r535_gsp_msgq_recv() assembles the payloads of cotinuation parts into
+ *   the return of the large GSP message.
+ *
+ * - Free the allocated memory: r535_gsp_msg_done().
+ *   The user is responsible for freeing the memory allocated for the GSP
+ *   message pages after they have been processed.
+ */
+static void *
+r535_gsp_msgq_peek(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries)
+{
+	struct r535_gsp_msg *mqe;
+	int ret;
+
+	ret = r535_gsp_msgq_wait(gsp, gsp_rpc_len, retries);
+	if (ret < 0)
+		return ERR_PTR(ret);
+
+	mqe = r535_gsp_msgq_get_entry(gsp);
+
+	return mqe->data;
+}
+
+struct r535_gsp_msg_info {
+	int *retries;
+	u32 gsp_rpc_len;
+	void *gsp_rpc_buf;
+	bool continuation;
+};
+
+static void
+r535_gsp_msg_dump(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg, int lvl);
+
+static void *
+r535_gsp_msgq_recv_one_elem(struct nvkm_gsp *gsp,
+			    struct r535_gsp_msg_info *info)
+{
+	u8 *buf = info->gsp_rpc_buf;
+	u32 rptr = *gsp->msgq.rptr;
+	struct r535_gsp_msg *mqe;
+	u32 size, expected, len;
+	int ret;
+
+	expected = info->gsp_rpc_len;
+
+	ret = r535_gsp_msgq_wait(gsp, expected, info->retries);
+	if (ret < 0)
+		return ERR_PTR(ret);
+
+	mqe = r535_gsp_msgq_get_entry(gsp);
+
+	if (info->continuation) {
+		struct nvfw_gsp_rpc *rpc = (struct nvfw_gsp_rpc *)mqe->data;
+
+		if (rpc->function != NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD) {
+			nvkm_error(&gsp->subdev,
+				   "Not a continuation of a large RPC\n");
+			r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR);
+			return ERR_PTR(-EIO);
+		}
+	}
+
+	size = ALIGN(expected + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE);
+
+	len = ((gsp->msgq.cnt - rptr) * GSP_PAGE_SIZE) - sizeof(*mqe);
+	len = min_t(u32, expected, len);
+
+	if (info->continuation)
+		memcpy(buf, mqe->data + sizeof(struct nvfw_gsp_rpc),
+		       len - sizeof(struct nvfw_gsp_rpc));
+	else
+		memcpy(buf, mqe->data, len);
+
+	expected -= len;
+
+	if (expected) {
+		mqe = (void *)((u8 *)gsp->shm.msgq.ptr + 0x1000 + 0 * 0x1000);
+		memcpy(buf + len, mqe, expected);
+	}
+
+	rptr = (rptr + DIV_ROUND_UP(size, GSP_PAGE_SIZE)) % gsp->msgq.cnt;
+
+	mb();
+	(*gsp->msgq.rptr) = rptr;
+	return buf;
+}
+
+static void *
+r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries)
+{
+	struct r535_gsp_msg *mqe;
+	const u32 max_rpc_size = GSP_MSG_MAX_SIZE - sizeof(*mqe);
+	struct nvfw_gsp_rpc *rpc;
+	struct r535_gsp_msg_info info = {0};
+	u32 expected = gsp_rpc_len;
+	void *buf;
+
+	mqe = r535_gsp_msgq_get_entry(gsp);
+	rpc = (struct nvfw_gsp_rpc *)mqe->data;
+
+	if (WARN_ON(rpc->length > max_rpc_size))
+		return NULL;
+
+	buf = kvmalloc(max_t(u32, rpc->length, expected), GFP_KERNEL);
+	if (!buf)
+		return ERR_PTR(-ENOMEM);
+
+	info.gsp_rpc_buf = buf;
+	info.retries = retries;
+	info.gsp_rpc_len = rpc->length;
+
+	buf = r535_gsp_msgq_recv_one_elem(gsp, &info);
+	if (IS_ERR(buf)) {
+		kvfree(info.gsp_rpc_buf);
+		info.gsp_rpc_buf = NULL;
+		return buf;
+	}
+
+	if (expected <= max_rpc_size)
+		return buf;
+
+	info.gsp_rpc_buf += info.gsp_rpc_len;
+	expected -= info.gsp_rpc_len;
+
+	while (expected) {
+		u32 size;
+
+		rpc = r535_gsp_msgq_peek(gsp, sizeof(*rpc), info.retries);
+		if (IS_ERR_OR_NULL(rpc)) {
+			kfree(buf);
+			return rpc;
+		}
+
+		info.gsp_rpc_len = rpc->length;
+		info.continuation = true;
+
+		rpc = r535_gsp_msgq_recv_one_elem(gsp, &info);
+		if (IS_ERR_OR_NULL(rpc)) {
+			kfree(buf);
+			return rpc;
+		}
+
+		size = info.gsp_rpc_len - sizeof(*rpc);
+		expected -= size;
+		info.gsp_rpc_buf += size;
+	}
+
+	rpc = buf;
+	rpc->length = gsp_rpc_len;
+	return buf;
+}
+
+static int
+r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc)
+{
+	struct r535_gsp_msg *msg = to_gsp_hdr(rpc, msg);
+	struct r535_gsp_msg *cqe;
+	u32 gsp_rpc_len = msg->checksum;
+	u64 *ptr = (void *)msg;
+	u64 *end;
+	u64 csum = 0;
+	int free, time = 1000000;
+	u32 wptr, size, step, len;
+	u32 off = 0;
+
+	len = ALIGN(GSP_MSG_HDR_SIZE + gsp_rpc_len, GSP_PAGE_SIZE);
+
+	end = (u64 *)((char *)ptr + len);
+	msg->pad = 0;
+	msg->checksum = 0;
+	msg->sequence = gsp->cmdq.seq++;
+	msg->elem_count = DIV_ROUND_UP(len, 0x1000);
+
+	while (ptr < end)
+		csum ^= *ptr++;
+
+	msg->checksum = upper_32_bits(csum) ^ lower_32_bits(csum);
+
+	wptr = *gsp->cmdq.wptr;
+	do {
+		do {
+			free = *gsp->cmdq.rptr + gsp->cmdq.cnt - wptr - 1;
+			if (free >= gsp->cmdq.cnt)
+				free -= gsp->cmdq.cnt;
+			if (free >= 1)
+				break;
+
+			usleep_range(1, 2);
+		} while(--time);
+
+		if (WARN_ON(!time)) {
+			kvfree(msg);
+			return -ETIMEDOUT;
+		}
+
+		cqe = (void *)((u8 *)gsp->shm.cmdq.ptr + 0x1000 + wptr * 0x1000);
+		step = min_t(u32, free, (gsp->cmdq.cnt - wptr));
+		size = min_t(u32, len, step * GSP_PAGE_SIZE);
+
+		memcpy(cqe, (u8 *)msg + off, size);
+
+		wptr += DIV_ROUND_UP(size, 0x1000);
+		if (wptr == gsp->cmdq.cnt)
+			wptr = 0;
+
+		off  += size;
+		len -= size;
+	} while (len);
+
+	nvkm_trace(&gsp->subdev, "cmdq: wptr %d\n", wptr);
+	wmb();
+	(*gsp->cmdq.wptr) = wptr;
+	mb();
+
+	nvkm_falcon_wr32(&gsp->falcon, 0xc00, 0x00000000);
+
+	kvfree(msg);
+	return 0;
+}
+
+static void *
+r535_gsp_cmdq_get(struct nvkm_gsp *gsp, u32 gsp_rpc_len)
+{
+	struct r535_gsp_msg *msg;
+	u32 size = GSP_MSG_HDR_SIZE + gsp_rpc_len;
+
+	size = ALIGN(size, GSP_MSG_MIN_SIZE);
+	msg = kvzalloc(size, GFP_KERNEL);
+	if (!msg)
+		return ERR_PTR(-ENOMEM);
+
+	msg->checksum = gsp_rpc_len;
+	return msg->data;
+}
+
+static void
+r535_gsp_msg_done(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg)
+{
+	kvfree(msg);
+}
+
+static void
+r535_gsp_msg_dump(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg, int lvl)
+{
+	if (gsp->subdev.debug >= lvl) {
+		nvkm_printk__(&gsp->subdev, lvl, info,
+			      "msg fn:%d len:0x%x/0x%zx res:0x%x resp:0x%x\n",
+			      msg->function, msg->length, msg->length - sizeof(*msg),
+			      msg->rpc_result, msg->rpc_result_private);
+		print_hex_dump(KERN_INFO, "msg: ", DUMP_PREFIX_OFFSET, 16, 1,
+			       msg->data, msg->length - sizeof(*msg), true);
+	}
+}
+
+struct nvfw_gsp_rpc *
+r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len)
+{
+	struct nvkm_subdev *subdev = &gsp->subdev;
+	struct nvfw_gsp_rpc *rpc;
+	int retries = 4000000, i;
+
+retry:
+	rpc = r535_gsp_msgq_peek(gsp, sizeof(*rpc), &retries);
+	if (IS_ERR_OR_NULL(rpc))
+		return rpc;
+
+	rpc = r535_gsp_msgq_recv(gsp, gsp_rpc_len, &retries);
+	if (IS_ERR_OR_NULL(rpc))
+		return rpc;
+
+	if (rpc->rpc_result) {
+		r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR);
+		r535_gsp_msg_done(gsp, rpc);
+		return ERR_PTR(-EINVAL);
+	}
+
+	r535_gsp_msg_dump(gsp, rpc, NV_DBG_TRACE);
+
+	if (fn && rpc->function == fn) {
+		if (gsp_rpc_len) {
+			if (rpc->length < gsp_rpc_len) {
+				nvkm_error(subdev, "rpc len %d < %d\n",
+					   rpc->length, gsp_rpc_len);
+				r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR);
+				r535_gsp_msg_done(gsp, rpc);
+				return ERR_PTR(-EIO);
+			}
+
+			return rpc;
+		}
+
+		r535_gsp_msg_done(gsp, rpc);
+		return NULL;
+	}
+
+	for (i = 0; i < gsp->msgq.ntfy_nr; i++) {
+		struct nvkm_gsp_msgq_ntfy *ntfy = &gsp->msgq.ntfy[i];
+
+		if (ntfy->fn == rpc->function) {
+			if (ntfy->func)
+				ntfy->func(ntfy->priv, ntfy->fn, rpc->data,
+					   rpc->length - sizeof(*rpc));
+			break;
+		}
+	}
+
+	if (i == gsp->msgq.ntfy_nr)
+		r535_gsp_msg_dump(gsp, rpc, NV_DBG_WARN);
+
+	r535_gsp_msg_done(gsp, rpc);
+	if (fn)
+		goto retry;
+
+	if (*gsp->msgq.rptr != *gsp->msgq.wptr)
+		goto retry;
+
+	return NULL;
+}
+
+int
+r535_gsp_msg_ntfy_add(struct nvkm_gsp *gsp, u32 fn, nvkm_gsp_msg_ntfy_func func, void *priv)
+{
+	int ret = 0;
+
+	mutex_lock(&gsp->msgq.mutex);
+	if (WARN_ON(gsp->msgq.ntfy_nr >= ARRAY_SIZE(gsp->msgq.ntfy))) {
+		ret = -ENOSPC;
+	} else {
+		gsp->msgq.ntfy[gsp->msgq.ntfy_nr].fn = fn;
+		gsp->msgq.ntfy[gsp->msgq.ntfy_nr].func = func;
+		gsp->msgq.ntfy[gsp->msgq.ntfy_nr].priv = priv;
+		gsp->msgq.ntfy_nr++;
+	}
+	mutex_unlock(&gsp->msgq.mutex);
+	return ret;
+}
+
+int
+r535_gsp_rpc_poll(struct nvkm_gsp *gsp, u32 fn)
+{
+	void *repv;
+
+	mutex_lock(&gsp->cmdq.mutex);
+	repv = r535_gsp_msg_recv(gsp, fn, 0);
+	mutex_unlock(&gsp->cmdq.mutex);
+	if (IS_ERR(repv))
+		return PTR_ERR(repv);
+
+	return 0;
+}
+
+static void *
+r535_gsp_rpc_handle_reply(struct nvkm_gsp *gsp, u32 fn,
+			  enum nvkm_gsp_rpc_reply_policy policy,
+			  u32 gsp_rpc_len)
+{
+	struct nvfw_gsp_rpc *reply;
+	void *repv = NULL;
+
+	switch (policy) {
+	case NVKM_GSP_RPC_REPLY_NOWAIT:
+		break;
+	case NVKM_GSP_RPC_REPLY_RECV:
+		reply = r535_gsp_msg_recv(gsp, fn, gsp_rpc_len);
+		if (!IS_ERR_OR_NULL(reply))
+			repv = reply->data;
+		else
+			repv = reply;
+		break;
+	case NVKM_GSP_RPC_REPLY_POLL:
+		repv = r535_gsp_msg_recv(gsp, fn, 0);
+		break;
+	}
+
+	return repv;
+}
+
+static void *
+r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *payload,
+		  enum nvkm_gsp_rpc_reply_policy policy, u32 gsp_rpc_len)
+{
+	struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc);
+	u32 fn = rpc->function;
+	int ret;
+
+	if (gsp->subdev.debug >= NV_DBG_TRACE) {
+		nvkm_trace(&gsp->subdev, "rpc fn:%d len:0x%x/0x%zx\n", rpc->function,
+			   rpc->length, rpc->length - sizeof(*rpc));
+		print_hex_dump(KERN_INFO, "rpc: ", DUMP_PREFIX_OFFSET, 16, 1,
+			       rpc->data, rpc->length - sizeof(*rpc), true);
+	}
+
+	ret = r535_gsp_cmdq_push(gsp, rpc);
+	if (ret)
+		return ERR_PTR(ret);
+
+	return r535_gsp_rpc_handle_reply(gsp, fn, policy, gsp_rpc_len);
+}
+
+static void
+r535_gsp_rpc_done(struct nvkm_gsp *gsp, void *repv)
+{
+	struct nvfw_gsp_rpc *rpc = container_of(repv, typeof(*rpc), data);
+
+	r535_gsp_msg_done(gsp, rpc);
+}
+
+static void *
+r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 payload_size)
+{
+	struct nvfw_gsp_rpc *rpc;
+
+	rpc = r535_gsp_cmdq_get(gsp, ALIGN(sizeof(*rpc) + payload_size,
+					   sizeof(u64)));
+	if (IS_ERR(rpc))
+		return ERR_CAST(rpc);
+
+	rpc->header_version = 0x03000000;
+	rpc->signature = ('C' << 24) | ('P' << 16) | ('R' << 8) | 'V';
+	rpc->function = fn;
+	rpc->rpc_result = 0xffffffff;
+	rpc->rpc_result_private = 0xffffffff;
+	rpc->length = sizeof(*rpc) + payload_size;
+	return rpc->data;
+}
+
+static void *
+r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload,
+		  enum nvkm_gsp_rpc_reply_policy policy, u32 gsp_rpc_len)
+{
+	struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc);
+	struct r535_gsp_msg *msg = to_gsp_hdr(rpc, msg);
+	const u32 max_rpc_size = GSP_MSG_MAX_SIZE - sizeof(*msg);
+	const u32 max_payload_size = max_rpc_size - sizeof(*rpc);
+	u32 payload_size = rpc->length - sizeof(*rpc);
+	void *repv;
+
+	mutex_lock(&gsp->cmdq.mutex);
+	if (payload_size > max_payload_size) {
+		const u32 fn = rpc->function;
+		u32 remain_payload_size = payload_size;
+
+		/* Adjust length, and send initial RPC. */
+		rpc->length = sizeof(*rpc) + max_payload_size;
+		msg->checksum = rpc->length;
+
+		repv = r535_gsp_rpc_send(gsp, payload, NVKM_GSP_RPC_REPLY_NOWAIT, 0);
+		if (IS_ERR(repv))
+			goto done;
+
+		payload += max_payload_size;
+		remain_payload_size -= max_payload_size;
+
+		/* Remaining chunks sent as CONTINUATION_RECORD RPCs. */
+		while (remain_payload_size) {
+			u32 size = min(remain_payload_size,
+				       max_payload_size);
+			void *next;
+
+			next = r535_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD, size);
+			if (IS_ERR(next)) {
+				repv = next;
+				goto done;
+			}
+
+			memcpy(next, payload, size);
+
+			repv = r535_gsp_rpc_send(gsp, next, NVKM_GSP_RPC_REPLY_NOWAIT, 0);
+			if (IS_ERR(repv))
+				goto done;
+
+			payload += size;
+			remain_payload_size -= size;
+		}
+
+		/* Wait for reply. */
+		repv = r535_gsp_rpc_handle_reply(gsp, fn, policy, payload_size +
+						 sizeof(*rpc));
+	} else {
+		repv = r535_gsp_rpc_send(gsp, payload, policy, gsp_rpc_len);
+	}
+
+done:
+	mutex_unlock(&gsp->cmdq.mutex);
+	return repv;
+}
+
+const struct nvkm_rm_api_rpc
+r535_rpc = {
+	.get = r535_gsp_rpc_get,
+	.push = r535_gsp_rpc_push,
+	.done = r535_gsp_rpc_done,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/vmm.c
index d3e95453f25d..52f2e5f14517 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/vmm.c
@@ -19,15 +19,38 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#include "vmm.h"
+#include <subdev/mmu/vmm.h>
 
-#include <nvrm/nvtypes.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/class/cl90f1.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/ctrl/ctrl90f1.h>
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvos.h>
+#include <nvhw/drf.h>
+#include "nvrm/vmm.h"
 
-static int
-r535_mmu_promote_vmm(struct nvkm_vmm *vmm)
+void
+r535_mmu_vaspace_del(struct nvkm_vmm *vmm)
+{
+	if (vmm->rm.external) {
+		NV0080_CTRL_DMA_UNSET_PAGE_DIRECTORY_PARAMS *ctrl;
+
+		ctrl = nvkm_gsp_rm_ctrl_get(&vmm->rm.device.object,
+					    NV0080_CTRL_CMD_DMA_UNSET_PAGE_DIRECTORY,
+					    sizeof(*ctrl));
+		if (!IS_ERR(ctrl)) {
+			ctrl->hVASpace = vmm->rm.object.handle;
+
+			WARN_ON(nvkm_gsp_rm_ctrl_wr(&vmm->rm.device.object, ctrl));
+		}
+
+		vmm->rm.external = false;
+	}
+
+	nvkm_gsp_rm_free(&vmm->rm.object);
+	nvkm_gsp_device_dtor(&vmm->rm.device);
+	nvkm_gsp_client_dtor(&vmm->rm.client);
+
+	nvkm_vmm_put(vmm, &vmm->rm.rsvd);
+}
+
+int
+r535_mmu_vaspace_new(struct nvkm_vmm *vmm, u32 handle, bool external)
 {
 	NV_VASPACE_ALLOCATION_PARAMETERS *args;
 	int ret;
@@ -37,58 +60,103 @@ r535_mmu_promote_vmm(struct nvkm_vmm *vmm)
 	if (ret)
 		return ret;
 
-	args = nvkm_gsp_rm_alloc_get(&vmm->rm.device.object, 0x90f10000, FERMI_VASPACE_A,
+	args = nvkm_gsp_rm_alloc_get(&vmm->rm.device.object, handle, FERMI_VASPACE_A,
 				     sizeof(*args), &vmm->rm.object);
 	if (IS_ERR(args))
 		return PTR_ERR(args);
 
 	args->index = NV_VASPACE_ALLOCATION_INDEX_GPU_NEW;
+	if (external)
+		args->flags = NV_VASPACE_ALLOCATION_FLAGS_IS_EXTERNALLY_OWNED;
 
 	ret = nvkm_gsp_rm_alloc_wr(&vmm->rm.object, args);
 	if (ret)
 		return ret;
 
-	{
+	if (!external) {
 		NV90F1_CTRL_VASPACE_COPY_SERVER_RESERVED_PDES_PARAMS *ctrl;
+		u8 page_shift = 29; /* 512MiB */
+		const u64 page_size = BIT_ULL(page_shift);
+		const struct nvkm_vmm_page *page;
+		const struct nvkm_vmm_desc *desc;
+		struct nvkm_vmm_pt *pd = vmm->pd;
+
+		for (page = vmm->func->page; page->shift; page++) {
+			if (page->shift == page_shift)
+				break;
+		}
+
+		if (WARN_ON(!page->shift))
+			return -EINVAL;
 
 		mutex_lock(&vmm->mutex.vmm);
-		ret = nvkm_vmm_get_locked(vmm, true, false, false, 0x1d, 32, 0x20000000,
+		ret = nvkm_vmm_get_locked(vmm, true, false, false, page_shift, 32, page_size,
 					  &vmm->rm.rsvd);
 		mutex_unlock(&vmm->mutex.vmm);
 		if (ret)
 			return ret;
 
+		/* Some parts of RM expect the server-reserved area to be in a specific location. */
+		if (WARN_ON(vmm->rm.rsvd->addr != SPLIT_VAS_SERVER_RM_MANAGED_VA_START ||
+			    vmm->rm.rsvd->size != SPLIT_VAS_SERVER_RM_MANAGED_VA_SIZE))
+			return -EINVAL;
+
 		ctrl = nvkm_gsp_rm_ctrl_get(&vmm->rm.object,
 					    NV90F1_CTRL_CMD_VASPACE_COPY_SERVER_RESERVED_PDES,
 					    sizeof(*ctrl));
 		if (IS_ERR(ctrl))
 			return PTR_ERR(ctrl);
 
-		ctrl->pageSize = 0x20000000;
+		ctrl->pageSize = page_size;
 		ctrl->virtAddrLo = vmm->rm.rsvd->addr;
 		ctrl->virtAddrHi = vmm->rm.rsvd->addr + vmm->rm.rsvd->size - 1;
-		ctrl->numLevelsToCopy = vmm->pd->pde[0]->pde[0] ? 3 : 2;
-		ctrl->levels[0].physAddress = vmm->pd->pt[0]->addr;
-		ctrl->levels[0].size = 0x20;
-		ctrl->levels[0].aperture = 1;
-		ctrl->levels[0].pageShift = 0x2f;
-		ctrl->levels[1].physAddress = vmm->pd->pde[0]->pt[0]->addr;
-		ctrl->levels[1].size = 0x1000;
-		ctrl->levels[1].aperture = 1;
-		ctrl->levels[1].pageShift = 0x26;
-		if (vmm->pd->pde[0]->pde[0]) {
-			ctrl->levels[2].physAddress = vmm->pd->pde[0]->pde[0]->pt[0]->addr;
-			ctrl->levels[2].size = 0x1000;
-			ctrl->levels[2].aperture = 1;
-			ctrl->levels[2].pageShift = 0x1d;
+
+		for (desc = page->desc; desc->bits; desc++) {
+			ctrl->numLevelsToCopy++;
+			page_shift += desc->bits;
+		}
+		desc--;
+
+		for (int i = 0; i < ctrl->numLevelsToCopy; i++, desc--) {
+			page_shift -= desc->bits;
+
+			ctrl->levels[i].physAddress = pd->pt[0]->addr;
+			ctrl->levels[i].size = (1 << desc->bits) * desc->size;
+			ctrl->levels[i].aperture = 1;
+			ctrl->levels[i].pageShift = page_shift;
+
+			pd = pd->pde[0];
 		}
 
 		ret = nvkm_gsp_rm_ctrl_wr(&vmm->rm.object, ctrl);
+	} else {
+		NV0080_CTRL_DMA_SET_PAGE_DIRECTORY_PARAMS *ctrl;
+
+		ctrl = nvkm_gsp_rm_ctrl_get(&vmm->rm.device.object,
+					    NV0080_CTRL_CMD_DMA_SET_PAGE_DIRECTORY,
+					    sizeof(*ctrl));
+		if (IS_ERR(ctrl))
+			return PTR_ERR(ctrl);
+
+		ctrl->physAddress = vmm->pd->pt[0]->addr;
+		ctrl->numEntries = 1 << vmm->func->page[0].desc->bits;
+		ctrl->flags = NVDEF(NV0080_CTRL_DMA_SET_PAGE_DIRECTORY, FLAGS, APERTURE, VIDMEM);
+		ctrl->hVASpace = vmm->rm.object.handle;
+
+		ret = nvkm_gsp_rm_ctrl_wr(&vmm->rm.device.object, ctrl);
+		if (ret == 0)
+			vmm->rm.external = true;
 	}
 
 	return ret;
 }
 
+static int
+r535_mmu_promote_vmm(struct nvkm_vmm *vmm)
+{
+	return r535_mmu_vaspace_new(vmm, NVKM_RM_VASPACE, true);
+}
+
 static void
 r535_mmu_dtor(struct nvkm_mmu *mmu)
 {
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/Kbuild
new file mode 100644
index 000000000000..5db0e7009e1f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/Kbuild
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: MIT
+nvkm-y += nvkm/subdev/gsp/rm/r570/rm.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/gsp.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/client.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/fbsr.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/disp.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/fifo.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/gr.o
+nvkm-y += nvkm/subdev/gsp/rm/r570/ofa.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/client.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/client.c
new file mode 100644
index 000000000000..87e6240662ed
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/client.c
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/rm.h>
+
+#include "nvrm/client.h"
+
+static int
+r570_gsp_client_ctor(struct nvkm_gsp_client *client, u32 handle)
+{
+	NV0000_ALLOC_PARAMETERS *args;
+
+	args = nvkm_gsp_rm_alloc_get(&client->object, handle, NV01_ROOT, sizeof(*args),
+				     &client->object);
+	if (IS_ERR(args))
+		return PTR_ERR(args);
+
+	args->hClient = client->object.handle;
+	args->processID = ~0;
+
+	return nvkm_gsp_rm_alloc_wr(&client->object, args);
+}
+
+const struct nvkm_rm_api_client
+r570_client = {
+	.ctor = r570_gsp_client_ctor,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/disp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/disp.c
new file mode 100644
index 000000000000..a96e31c2d80b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/disp.c
@@ -0,0 +1,263 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/rm.h>
+
+#include <engine/disp.h>
+#include <engine/disp/outp.h>
+
+#include "nvhw/drf.h"
+
+#include "nvrm/disp.h"
+
+static int
+r570_dmac_alloc(struct nvkm_disp *disp, u32 oclass, int inst, u32 put_offset,
+		struct nvkm_gsp_object *dmac)
+{
+	NV50VAIO_CHANNELDMA_ALLOCATION_PARAMETERS *args;
+
+	args = nvkm_gsp_rm_alloc_get(&disp->rm.object, (oclass << 16) | inst, oclass,
+				     sizeof(*args), dmac);
+	if (IS_ERR(args))
+		return PTR_ERR(args);
+
+	args->channelInstance = inst;
+	args->offset = put_offset;
+	args->subDeviceId = BIT(0);
+
+	return nvkm_gsp_rm_alloc_wr(dmac, args);
+}
+
+static int
+r570_disp_chan_set_pushbuf(struct nvkm_disp *disp, s32 oclass, int inst, struct nvkm_memory *memory)
+{
+	struct nvkm_gsp *gsp = disp->rm.objcom.client->gsp;
+	NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&gsp->internal.device.subdevice,
+				    NV2080_CTRL_CMD_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER,
+				    sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	if (memory) {
+		switch (nvkm_memory_target(memory)) {
+		case NVKM_MEM_TARGET_NCOH:
+			ctrl->addressSpace = ADDR_SYSMEM;
+			ctrl->cacheSnoop = 0;
+			ctrl->pbTargetAperture = PHYS_PCI;
+			break;
+		case NVKM_MEM_TARGET_HOST:
+			ctrl->addressSpace = ADDR_SYSMEM;
+			ctrl->cacheSnoop = 1;
+			ctrl->pbTargetAperture = PHYS_PCI_COHERENT;
+			break;
+		case NVKM_MEM_TARGET_VRAM:
+			ctrl->addressSpace = ADDR_FBMEM;
+			ctrl->pbTargetAperture = PHYS_NVM;
+			break;
+		default:
+			WARN_ON(1);
+			return -EINVAL;
+		}
+
+		ctrl->physicalAddr = nvkm_memory_addr(memory);
+		ctrl->limit = nvkm_memory_size(memory) - 1;
+	}
+
+	ctrl->hclass = oclass;
+	ctrl->channelInstance = inst;
+	ctrl->valid = ((oclass & 0xff) != 0x7a) ? 1 : 0;
+	ctrl->subDeviceId = BIT(0);
+
+	return nvkm_gsp_rm_ctrl_wr(&gsp->internal.device.subdevice, ctrl);
+}
+
+static int
+r570_dp_set_indexed_link_rates(struct nvkm_outp *outp)
+{
+	NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS *ctrl;
+	struct nvkm_disp *disp = outp->disp;
+
+	if (WARN_ON(outp->dp.rates > ARRAY_SIZE(ctrl->linkRateTbl)))
+		return -EINVAL;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
+				    NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->displayId = BIT(outp->index);
+	for (int i = 0; i < outp->dp.rates; i++)
+		ctrl->linkRateTbl[outp->dp.rate[i].dpcd] = outp->dp.rate[i].rate * 10 / 200;
+
+	return nvkm_gsp_rm_ctrl_wr(&disp->rm.objcom, ctrl);
+}
+
+static int
+r570_dp_get_caps(struct nvkm_disp *disp, int *plink_bw, bool *pmst, bool *pwm)
+{
+	NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS *ctrl;
+	int ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
+				    NV0073_CTRL_CMD_DP_GET_CAPS, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->sorIndex = ~0;
+
+	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
+	if (ret) {
+		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+		return ret;
+	}
+
+	switch (NVVAL_GET(ctrl->maxLinkRate, NV0073_CTRL_CMD, DP_GET_CAPS, MAX_LINK_RATE)) {
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_1_62:
+		*plink_bw = 0x06;
+		break;
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_2_70:
+		*plink_bw = 0x0a;
+		break;
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_5_40:
+		*plink_bw = 0x14;
+		break;
+	case NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_8_10:
+		*plink_bw = 0x1e;
+		break;
+	default:
+		*plink_bw = 0x00;
+		break;
+	}
+
+	*pmst = ctrl->bIsMultistreamSupported;
+	*pwm = ctrl->bHasIncreasedWatermarkLimits;
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
+}
+
+static int
+r570_bl_ctrl(struct nvkm_disp *disp, unsigned display_id, bool set, int *pval)
+{
+	u32 cmd = set ? NV0073_CTRL_CMD_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS :
+			NV0073_CTRL_CMD_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS;
+	NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS *ctrl;
+	int ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom, cmd, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->displayId = BIT(display_id);
+	ctrl->brightness = *pval;
+	ctrl->brightnessType = NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_TYPE_PERCENT100;
+
+	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
+	if (ret)
+		return ret;
+
+	*pval = ctrl->brightness;
+
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
+}
+
+static int
+r570_disp_get_active(struct nvkm_disp *disp, unsigned head, u32 *displayid)
+{
+	NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS *ctrl;
+	int ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
+				    NV0073_CTRL_CMD_SYSTEM_GET_ACTIVE, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->subDeviceInstance = 0;
+	ctrl->head = head;
+
+	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
+	if (ret) {
+		nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+		return ret;
+	}
+
+	*displayid = ctrl->displayId;
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
+}
+static int
+r570_disp_get_connect_state(struct nvkm_disp *disp, unsigned display_id)
+{
+	NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS *ctrl;
+	int ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&disp->rm.objcom,
+				    NV0073_CTRL_CMD_SYSTEM_GET_CONNECT_STATE, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->subDeviceInstance = 0;
+	ctrl->displayMask = BIT(display_id);
+
+	ret = nvkm_gsp_rm_ctrl_push(&disp->rm.objcom, &ctrl, sizeof(*ctrl));
+	if (ret == 0 && (ctrl->displayMask & BIT(display_id)))
+		ret = 1;
+
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return ret;
+}
+
+static int
+r570_disp_get_supported(struct nvkm_disp *disp, unsigned long *pmask)
+{
+	NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_rd(&disp->rm.objcom,
+				   NV0073_CTRL_CMD_SYSTEM_GET_SUPPORTED, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	*pmask = ctrl->displayMask;
+
+	nvkm_gsp_rm_ctrl_done(&disp->rm.objcom, ctrl);
+	return 0;
+}
+
+static int
+r570_disp_get_static_info(struct nvkm_disp *disp)
+{
+	NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS *ctrl;
+	struct nvkm_gsp *gsp = disp->engine.subdev.device->gsp;
+
+	ctrl = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
+				   NV2080_CTRL_CMD_INTERNAL_DISPLAY_GET_STATIC_INFO,
+				   sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	disp->wndw.mask = ctrl->windowPresentMask;
+	disp->wndw.nr = fls(disp->wndw.mask);
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
+	return 0;
+}
+
+const struct nvkm_rm_api_disp
+r570_disp = {
+	.get_static_info = r570_disp_get_static_info,
+	.get_supported = r570_disp_get_supported,
+	.get_connect_state = r570_disp_get_connect_state,
+	.get_active = r570_disp_get_active,
+	.bl_ctrl = r570_bl_ctrl,
+	.dp = {
+		.get_caps = r570_dp_get_caps,
+		.set_indexed_link_rates = r570_dp_set_indexed_link_rates,
+	},
+	.chan = {
+		.set_pushbuf = r570_disp_chan_set_pushbuf,
+		.dmac_alloc = r570_dmac_alloc,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/fbsr.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/fbsr.c
new file mode 100644
index 000000000000..2945d5b4e570
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/fbsr.c
@@ -0,0 +1,149 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <subdev/instmem/priv.h>
+#include <subdev/bar.h>
+#include <subdev/gsp.h>
+#include <subdev/mmu/vmm.h>
+
+#include "nvrm/fbsr.h"
+#include "nvrm/fifo.h"
+
+static int
+r570_fbsr_suspend_channels(struct nvkm_gsp *gsp, bool suspend)
+{
+	NV2080_CTRL_CMD_INTERNAL_FIFO_TOGGLE_ACTIVE_CHANNEL_SCHEDULING_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&gsp->internal.device.subdevice,
+				    NV2080_CTRL_CMD_INTERNAL_FIFO_TOGGLE_ACTIVE_CHANNEL_SCHEDULING,
+				    sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->bDisableActiveChannels = suspend;
+
+	return nvkm_gsp_rm_ctrl_wr(&gsp->internal.device.subdevice, ctrl);
+}
+
+static void
+r570_fbsr_resume(struct nvkm_gsp *gsp)
+{
+	struct nvkm_device *device = gsp->subdev.device;
+	struct nvkm_instmem *imem = device->imem;
+	struct nvkm_instobj *iobj;
+	struct nvkm_vmm *vmm;
+
+	/* Restore BAR2 page tables via BAR0 window, and re-enable BAR2. */
+	list_for_each_entry(iobj, &imem->boot, head) {
+		if (iobj->suspend)
+			nvkm_instobj_load(iobj);
+	}
+
+	device->bar->bar2 = true;
+
+	vmm = nvkm_bar_bar2_vmm(device);
+	vmm->func->flush(vmm, 0);
+
+	/* Restore remaining BAR2 allocations (including BAR1 page tables) via BAR2. */
+	list_for_each_entry(iobj, &imem->list, head) {
+		if (iobj->suspend)
+			nvkm_instobj_load(iobj);
+	}
+
+	vmm = nvkm_bar_bar1_vmm(device);
+	vmm->func->flush(vmm, 0);
+
+	/* Resume channel scheduling. */
+	r570_fbsr_suspend_channels(device->gsp, false);
+
+	/* Finish cleaning up. */
+	r535_fbsr_resume(gsp);
+}
+
+static int
+r570_fbsr_init(struct nvkm_gsp *gsp, struct sg_table *sgt, u64 size)
+{
+	NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS *ctrl;
+	struct nvkm_gsp_object memlist;
+	int ret;
+
+	ret = r535_fbsr_memlist(&gsp->internal.device, 0xcaf00003, NVKM_MEM_TARGET_HOST,
+				0, size, sgt, &memlist);
+	if (ret)
+		return ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&gsp->internal.device.subdevice,
+				    NV2080_CTRL_CMD_INTERNAL_FBSR_INIT, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->hClient = gsp->internal.client.object.handle;
+	ctrl->hSysMem = memlist.handle;
+	ctrl->sysmemAddrOfSuspendResumeData = gsp->sr.meta.addr;
+	ctrl->bEnteringGcoffState = 1;
+
+	ret = nvkm_gsp_rm_ctrl_wr(&gsp->internal.device.subdevice, ctrl);
+	if (ret)
+		return ret;
+
+	nvkm_gsp_rm_free(&memlist);
+	return 0;
+}
+
+static int
+r570_fbsr_suspend(struct nvkm_gsp *gsp)
+{
+	struct nvkm_subdev *subdev = &gsp->subdev;
+	struct nvkm_device *device = subdev->device;
+	struct nvkm_instmem *imem = device->imem;
+	struct nvkm_instobj *iobj;
+	u64 size;
+	int ret;
+
+	/* Stop channel scheduling. */
+	r570_fbsr_suspend_channels(gsp, true);
+
+	/* Save BAR2 allocations to system memory. */
+	list_for_each_entry(iobj, &imem->list, head) {
+		if (iobj->preserve) {
+			ret = nvkm_instobj_save(iobj);
+			if (ret)
+				return ret;
+		}
+	}
+
+	list_for_each_entry(iobj, &imem->boot, head) {
+		ret = nvkm_instobj_save(iobj);
+		if (ret)
+			return ret;
+	}
+
+	/* Disable BAR2 access. */
+	device->bar->bar2 = false;
+
+	/* Allocate system memory to hold RM's VRAM allocations across suspend. */
+	size  = gsp->fb.heap.size;
+	size += gsp->fb.rsvd_size;
+	size += gsp->fb.bios.vga_workspace.size;
+	nvkm_debug(subdev, "fbsr: size: 0x%llx bytes\n", size);
+
+	ret = nvkm_gsp_sg(device, size, &gsp->sr.fbsr);
+	if (ret)
+		return ret;
+
+	/* Initialise FBSR on RM. */
+	ret = r570_fbsr_init(gsp, &gsp->sr.fbsr, size);
+	if (ret) {
+		nvkm_gsp_sg_free(device, &gsp->sr.fbsr);
+		return ret;
+	}
+
+	return 0;
+}
+
+const struct nvkm_rm_api_fbsr
+r570_fbsr = {
+	.suspend = r570_fbsr_suspend,
+	.resume = r570_fbsr_resume,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/fifo.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/fifo.c
new file mode 100644
index 000000000000..79132805cfcf
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/fifo.c
@@ -0,0 +1,217 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/rm.h>
+
+#include <subdev/mmu.h>
+#include <engine/fifo/priv.h>
+#include <engine/fifo/chan.h>
+#include <engine/fifo/runl.h>
+
+#include "nvhw/drf.h"
+
+#include "nvrm/fifo.h"
+#include "nvrm/engine.h"
+
+#define CHID_PER_USERD 8
+
+static int
+r570_chan_alloc(struct nvkm_gsp_device *device, u32 handle, u32 nv2080_engine_type, u8 runq,
+	        bool priv, int chid, u64 inst_addr, u64 userd_addr, u64 mthdbuf_addr,
+		struct nvkm_vmm *vmm, u64 gpfifo_offset, u32 gpfifo_length,
+		struct nvkm_gsp_object *chan)
+{
+	struct nvkm_gsp *gsp = device->object.client->gsp;
+	struct nvkm_fifo *fifo = gsp->subdev.device->fifo;
+	const int userd_p = chid / CHID_PER_USERD;
+	const int userd_i = chid % CHID_PER_USERD;
+	NV_CHANNELGPFIFO_ALLOCATION_PARAMETERS *args;
+
+	args = nvkm_gsp_rm_alloc_get(&device->object, handle,
+				     fifo->func->chan.user.oclass, sizeof(*args), chan);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
+
+	args->gpFifoOffset = gpfifo_offset;
+	args->gpFifoEntries = gpfifo_length / 8;
+
+	args->flags  = NVDEF(NVOS04, FLAGS, CHANNEL_TYPE, PHYSICAL);
+	args->flags |= NVDEF(NVOS04, FLAGS, VPR, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_SKIP_MAP_REFCOUNTING, FALSE);
+	args->flags |= NVVAL(NVOS04, FLAGS, GROUP_CHANNEL_RUNQUEUE, runq);
+	if (!priv)
+		args->flags |= NVDEF(NVOS04, FLAGS, PRIVILEGED_CHANNEL, FALSE);
+	else
+		args->flags |= NVDEF(NVOS04, FLAGS, PRIVILEGED_CHANNEL, TRUE);
+	args->flags |= NVDEF(NVOS04, FLAGS, DELAY_CHANNEL_SCHEDULING, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_DENY_PHYSICAL_MODE_CE, FALSE);
+
+	args->flags |= NVVAL(NVOS04, FLAGS, CHANNEL_USERD_INDEX_VALUE, userd_i);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_USERD_INDEX_FIXED, FALSE);
+	args->flags |= NVVAL(NVOS04, FLAGS, CHANNEL_USERD_INDEX_PAGE_VALUE, userd_p);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_USERD_INDEX_PAGE_FIXED, TRUE);
+
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_DENY_AUTH_LEVEL_PRIV, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_SKIP_SCRUBBER, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_CLIENT_MAP_FIFO, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, SET_EVICT_LAST_CE_PREFETCH_CHANNEL, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_VGPU_PLUGIN_CONTEXT, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, CHANNEL_PBDMA_ACQUIRE_TIMEOUT, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, GROUP_CHANNEL_THREAD, DEFAULT);
+	args->flags |= NVDEF(NVOS04, FLAGS, MAP_CHANNEL, FALSE);
+	args->flags |= NVDEF(NVOS04, FLAGS, SKIP_CTXBUFFER_ALLOC, FALSE);
+
+	args->hVASpace = vmm->rm.object.handle;
+	args->engineType = nv2080_engine_type;
+
+	args->instanceMem.base = inst_addr;
+	args->instanceMem.size = fifo->func->chan.func->inst->size;
+	args->instanceMem.addressSpace = 2;
+	args->instanceMem.cacheAttrib = 1;
+
+	args->userdMem.base = userd_addr;
+	args->userdMem.size = fifo->func->chan.func->userd->size;
+	args->userdMem.addressSpace = 2;
+	args->userdMem.cacheAttrib = 1;
+
+	args->ramfcMem.base = inst_addr;
+	args->ramfcMem.size = 0x200;
+	args->ramfcMem.addressSpace = 2;
+	args->ramfcMem.cacheAttrib = 1;
+
+	args->mthdbufMem.base = mthdbuf_addr;
+	args->mthdbufMem.size = fifo->rm.mthdbuf_size;
+	args->mthdbufMem.addressSpace = 1;
+	args->mthdbufMem.cacheAttrib = 0;
+
+	if (!priv)
+		args->internalFlags = NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, PRIVILEGE, USER);
+	else
+		args->internalFlags = NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, PRIVILEGE, ADMIN);
+	args->internalFlags |= NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, ERROR_NOTIFIER_TYPE, NONE);
+	args->internalFlags |= NVDEF(NV_KERNELCHANNEL, ALLOC_INTERNALFLAGS, ECC_ERROR_NOTIFIER_TYPE, NONE);
+
+	return nvkm_gsp_rm_alloc_wr(chan, args);
+}
+
+static int
+r570_fifo_rc_triggered(void *priv, u32 fn, void *repv, u32 repc)
+{
+	rpc_rc_triggered_v17_02 *msg = repv;
+	struct nvkm_gsp *gsp = priv;
+
+	if (WARN_ON(repc < sizeof(*msg)))
+		return -EINVAL;
+
+	nvkm_error(&gsp->subdev, "rc engn:%08x chid:%d gfid:%d level:%d type:%d scope:%d part:%d "
+				 "fault_addr:%08x%08x fault_type:%08x\n",
+		   msg->nv2080EngineType, msg->chid, msg->gfid, msg->exceptLevel, msg->exceptType,
+		   msg->scope, msg->partitionAttributionId,
+		   msg->mmuFaultAddrHi, msg->mmuFaultAddrLo, msg->mmuFaultType);
+
+	r535_fifo_rc_chid(gsp->subdev.device->fifo, msg->chid);
+	return 0;
+}
+
+static int
+r570_fifo_ectx_size(struct nvkm_fifo *fifo)
+{
+	NV2080_CTRL_GPU_GET_CONSTRUCTED_FALCON_INFO_PARAMS *ctrl;
+	struct nvkm_gsp *gsp = fifo->engine.subdev.device->gsp;
+	struct nvkm_runl *runl;
+	struct nvkm_engn *engn;
+
+	ctrl = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
+				   NV2080_CTRL_CMD_GPU_GET_CONSTRUCTED_FALCON_INFO,
+				   sizeof(*ctrl));
+	if (WARN_ON(IS_ERR(ctrl)))
+		return PTR_ERR(ctrl);
+
+	for (int i = 0; i < ctrl->numConstructedFalcons; i++) {
+		nvkm_runl_foreach(runl, fifo) {
+			nvkm_runl_foreach_engn(engn, runl) {
+				if (engn->rm.desc == ctrl->constructedFalconsTable[i].engDesc) {
+					engn->rm.size =
+						ctrl->constructedFalconsTable[i].ctxBufferSize;
+					break;
+				}
+			}
+		}
+	}
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
+	return 0;
+}
+
+static int
+r570_fifo_xlat_rm_engine_type(u32 rm, enum nvkm_subdev_type *ptype, int *p2080)
+{
+#define RM_ENGINE_TYPE(RM,NVKM,INST)              \
+	RM_ENGINE_TYPE_##RM:                      \
+		*ptype = NVKM_ENGINE_##NVKM;      \
+		*p2080 = NV2080_ENGINE_TYPE_##RM; \
+		return INST
+
+	switch (rm) {
+	case RM_ENGINE_TYPE(    GR0,    GR, 0);
+	case RM_ENGINE_TYPE(  COPY0,    CE, 0);
+	case RM_ENGINE_TYPE(  COPY1,    CE, 1);
+	case RM_ENGINE_TYPE(  COPY2,    CE, 2);
+	case RM_ENGINE_TYPE(  COPY3,    CE, 3);
+	case RM_ENGINE_TYPE(  COPY4,    CE, 4);
+	case RM_ENGINE_TYPE(  COPY5,    CE, 5);
+	case RM_ENGINE_TYPE(  COPY6,    CE, 6);
+	case RM_ENGINE_TYPE(  COPY7,    CE, 7);
+	case RM_ENGINE_TYPE(  COPY8,    CE, 8);
+	case RM_ENGINE_TYPE(  COPY9,    CE, 9);
+	case RM_ENGINE_TYPE( COPY10,    CE, 10);
+	case RM_ENGINE_TYPE( COPY11,    CE, 11);
+	case RM_ENGINE_TYPE( COPY12,    CE, 12);
+	case RM_ENGINE_TYPE( COPY13,    CE, 13);
+	case RM_ENGINE_TYPE( COPY14,    CE, 14);
+	case RM_ENGINE_TYPE( COPY15,    CE, 15);
+	case RM_ENGINE_TYPE( COPY16,    CE, 16);
+	case RM_ENGINE_TYPE( COPY17,    CE, 17);
+	case RM_ENGINE_TYPE( COPY18,    CE, 18);
+	case RM_ENGINE_TYPE( COPY19,    CE, 19);
+	case RM_ENGINE_TYPE( NVDEC0, NVDEC, 0);
+	case RM_ENGINE_TYPE( NVDEC1, NVDEC, 1);
+	case RM_ENGINE_TYPE( NVDEC2, NVDEC, 2);
+	case RM_ENGINE_TYPE( NVDEC3, NVDEC, 3);
+	case RM_ENGINE_TYPE( NVDEC4, NVDEC, 4);
+	case RM_ENGINE_TYPE( NVDEC5, NVDEC, 5);
+	case RM_ENGINE_TYPE( NVDEC6, NVDEC, 6);
+	case RM_ENGINE_TYPE( NVDEC7, NVDEC, 7);
+	case RM_ENGINE_TYPE( NVENC0, NVENC, 0);
+	case RM_ENGINE_TYPE( NVENC1, NVENC, 1);
+	case RM_ENGINE_TYPE( NVENC2, NVENC, 2);
+	case RM_ENGINE_TYPE( NVENC3, NVENC, 3);
+	case RM_ENGINE_TYPE(NVJPEG0, NVJPG, 0);
+	case RM_ENGINE_TYPE(NVJPEG1, NVJPG, 1);
+	case RM_ENGINE_TYPE(NVJPEG2, NVJPG, 2);
+	case RM_ENGINE_TYPE(NVJPEG3, NVJPG, 3);
+	case RM_ENGINE_TYPE(NVJPEG4, NVJPG, 4);
+	case RM_ENGINE_TYPE(NVJPEG5, NVJPG, 5);
+	case RM_ENGINE_TYPE(NVJPEG6, NVJPG, 6);
+	case RM_ENGINE_TYPE(NVJPEG7, NVJPG, 7);
+	case RM_ENGINE_TYPE(     SW,    SW, 0);
+	case RM_ENGINE_TYPE(   SEC2,  SEC2, 0);
+	case RM_ENGINE_TYPE(   OFA0,   OFA, 0);
+	case RM_ENGINE_TYPE(   OFA1,   OFA, 1);
+	default:
+		return -EINVAL;
+	}
+#undef RM_ENGINE_TYPE
+}
+
+const struct nvkm_rm_api_fifo
+r570_fifo = {
+	.xlat_rm_engine_type = r570_fifo_xlat_rm_engine_type,
+	.ectx_size = r570_fifo_ectx_size,
+	.rsvd_chids = 1,
+	.rc_triggered = r570_fifo_rc_triggered,
+	.chan = {
+		.alloc = r570_chan_alloc,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/gr.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/gr.c
new file mode 100644
index 000000000000..b6cced9b8aa1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/gr.c
@@ -0,0 +1,191 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/gr.h>
+
+#include <subdev/mmu.h>
+#include <engine/fifo.h>
+#include <engine/fifo/chid.h>
+#include <engine/gr/priv.h>
+
+#include "nvrm/gr.h"
+#include "nvrm/engine.h"
+
+int
+r570_gr_tpc_mask(struct nvkm_gsp *gsp, int gpc, u32 *pmask)
+{
+	NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS *ctrl;
+	int ret;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&gsp->internal.device.subdevice,
+				    NV2080_CTRL_CMD_GPU_GET_FERMI_TPC_INFO, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->gpcId = gpc;
+
+	ret = nvkm_gsp_rm_ctrl_push(&gsp->internal.device.subdevice, &ctrl, sizeof(*ctrl));
+	if (ret)
+		return ret;
+
+	*pmask = ctrl->tpcMask;
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
+	return 0;
+}
+
+int
+r570_gr_gpc_mask(struct nvkm_gsp *gsp, u32 *pmask)
+{
+	NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
+				   NV2080_CTRL_CMD_GPU_GET_FERMI_GPC_INFO, sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	*pmask = ctrl->gpcMask;
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, ctrl);
+	return 0;
+}
+
+static int
+r570_gr_scrubber_ctrl(struct r535_gr *gr, bool teardown)
+{
+	NV2080_CTRL_INTERNAL_GR_INIT_BUG4208224_WAR_PARAMS *ctrl;
+
+	ctrl = nvkm_gsp_rm_ctrl_get(&gr->scrubber.vmm->rm.device.subdevice,
+				    NV2080_CTRL_CMD_INTERNAL_KGR_INIT_BUG4208224_WAR,
+				    sizeof(*ctrl));
+	if (IS_ERR(ctrl))
+		return PTR_ERR(ctrl);
+
+	ctrl->bTeardown = teardown;
+
+	return nvkm_gsp_rm_ctrl_wr(&gr->scrubber.vmm->rm.device.subdevice, ctrl);
+}
+
+static void
+r570_gr_scrubber_fini(struct r535_gr *gr)
+{
+	/* Teardown scrubber channel on RM. */
+	if (gr->scrubber.enabled) {
+		WARN_ON(r570_gr_scrubber_ctrl(gr, true));
+		gr->scrubber.enabled = false;
+	}
+
+	/* Free scrubber channel. */
+	nvkm_gsp_rm_free(&gr->scrubber.threed);
+	nvkm_gsp_rm_free(&gr->scrubber.chan);
+
+	for (int i = 0; i < gr->ctxbuf_nr; i++) {
+		nvkm_vmm_put(gr->scrubber.vmm, &gr->scrubber.ctxbuf.vma[i]);
+		nvkm_memory_unref(&gr->scrubber.ctxbuf.mem[i]);
+	}
+
+	nvkm_vmm_unref(&gr->scrubber.vmm);
+	nvkm_memory_unref(&gr->scrubber.inst);
+}
+
+static int
+r570_gr_scrubber_init(struct r535_gr *gr)
+{
+	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
+	struct nvkm_device *device = subdev->device;
+	struct nvkm_gsp *gsp = device->gsp;
+	struct nvkm_rm *rm = gsp->rm;
+	int ret;
+
+	/* Scrubber channel only required on TU10x. */
+	switch (device->chipset) {
+	case 0x162:
+	case 0x164:
+	case 0x166:
+		break;
+	default:
+		return 0;
+	}
+
+	if (gr->scrubber.chid < 0) {
+		gr->scrubber.chid = nvkm_chid_get(device->fifo->chid, NULL);
+		if (gr->scrubber.chid < 0)
+			return gr->scrubber.chid;
+	}
+
+	/* Allocate scrubber channel. */
+	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST,
+			      0x2000 + device->fifo->rm.mthdbuf_size, 0, true,
+			      &gr->scrubber.inst);
+	if (ret)
+		goto done;
+
+	ret = nvkm_vmm_new(device, 0x1000, 0, NULL, 0, NULL, "grScrubberVmm",
+			   &gr->scrubber.vmm);
+	if (ret)
+		goto done;
+
+	ret = r535_mmu_vaspace_new(gr->scrubber.vmm, KGRAPHICS_SCRUBBER_HANDLE_VAS, false);
+	if (ret)
+		goto done;
+
+	ret = rm->api->fifo->chan.alloc(&gr->scrubber.vmm->rm.device, KGRAPHICS_SCRUBBER_HANDLE_CHANNEL,
+					NV2080_ENGINE_TYPE_GR0, 0, false, gr->scrubber.chid,
+					nvkm_memory_addr(gr->scrubber.inst),
+					nvkm_memory_addr(gr->scrubber.inst) + 0x1000,
+					nvkm_memory_addr(gr->scrubber.inst) + 0x2000,
+					gr->scrubber.vmm, 0, 0x1000, &gr->scrubber.chan);
+	if (ret)
+		goto done;
+
+	ret = r535_gr_promote_ctx(gr, false, gr->scrubber.vmm, gr->scrubber.ctxbuf.mem,
+				  gr->scrubber.ctxbuf.vma, &gr->scrubber.chan);
+	if (ret)
+		goto done;
+
+	ret = nvkm_gsp_rm_alloc(&gr->scrubber.chan, KGRAPHICS_SCRUBBER_HANDLE_3DOBJ,
+				rm->gpu->gr.class.threed, 0, &gr->scrubber.threed);
+	if (ret)
+		goto done;
+
+	/* Initialise scrubber channel on RM. */
+	ret = r570_gr_scrubber_ctrl(gr, false);
+	if (ret)
+		goto done;
+
+	gr->scrubber.enabled = true;
+
+done:
+	if (ret)
+		r570_gr_scrubber_fini(gr);
+
+	return ret;
+}
+
+static int
+r570_gr_get_ctxbufs_info(struct r535_gr *gr)
+{
+	NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS *info;
+	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
+	struct nvkm_gsp *gsp = subdev->device->gsp;
+
+	info = nvkm_gsp_rm_ctrl_rd(&gsp->internal.device.subdevice,
+				   NV2080_CTRL_CMD_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO,
+				   sizeof(*info));
+	if (WARN_ON(IS_ERR(info)))
+		return PTR_ERR(info);
+
+	for (int i = 0; i < ARRAY_SIZE(info->engineContextBuffersInfo[0].engine); i++)
+		r535_gr_get_ctxbuf_info(gr, i, &info->engineContextBuffersInfo[0].engine[i]);
+
+	nvkm_gsp_rm_ctrl_done(&gsp->internal.device.subdevice, info);
+	return 0;
+}
+
+const struct nvkm_rm_api_gr
+r570_gr = {
+	.get_ctxbufs_info = r570_gr_get_ctxbufs_info,
+	.scrubber.init = r570_gr_scrubber_init,
+	.scrubber.fini = r570_gr_scrubber_fini,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/gsp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/gsp.c
new file mode 100644
index 000000000000..9d2fa4e66d59
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/gsp.c
@@ -0,0 +1,216 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/rm.h>
+#include <rm/rpc.h>
+
+#include <asm-generic/video.h>
+
+#include "nvrm/gsp.h"
+#include "nvrm/rpcfn.h"
+#include "nvrm/msgfn.h"
+
+#include <core/pci.h>
+#include <subdev/pci/priv.h>
+
+static u32
+r570_gsp_sr_data_size(struct nvkm_gsp *gsp)
+{
+	GspFwWprMeta *meta = gsp->wpr_meta.data;
+
+	return (meta->frtsOffset + meta->frtsSize) -
+	       (meta->nonWprHeapOffset + meta->nonWprHeapSize);
+}
+
+static void
+r570_gsp_drop_post_nocat_record(struct nvkm_gsp *gsp)
+{
+	if (gsp->subdev.debug < NV_DBG_DEBUG) {
+		r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_GSP_POST_NOCAT_RECORD, NULL, NULL);
+		r535_gsp_msg_ntfy_add(gsp, NV_VGPU_MSG_EVENT_GSP_LOCKDOWN_NOTICE, NULL, NULL);
+	}
+}
+
+static bool
+r570_gsp_xlat_mc_engine_idx(u32 mc_engine_idx, enum nvkm_subdev_type *ptype, int *pinst)
+{
+	switch (mc_engine_idx) {
+	case MC_ENGINE_IDX_GSP:
+		*ptype = NVKM_SUBDEV_GSP;
+		*pinst = 0;
+		return true;
+	case MC_ENGINE_IDX_DISP:
+		*ptype = NVKM_ENGINE_DISP;
+		*pinst = 0;
+		return true;
+	case MC_ENGINE_IDX_CE0 ... MC_ENGINE_IDX_CE19:
+		*ptype = NVKM_ENGINE_CE;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_CE0;
+		return true;
+	case MC_ENGINE_IDX_GR0:
+		*ptype = NVKM_ENGINE_GR;
+		*pinst = 0;
+		return true;
+	case MC_ENGINE_IDX_NVDEC0 ... MC_ENGINE_IDX_NVDEC7:
+		*ptype = NVKM_ENGINE_NVDEC;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_NVDEC0;
+		return true;
+	case MC_ENGINE_IDX_NVENC ... MC_ENGINE_IDX_NVENC3:
+		*ptype = NVKM_ENGINE_NVENC;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_NVENC;
+		return true;
+	case MC_ENGINE_IDX_NVJPEG0 ... MC_ENGINE_IDX_NVJPEG7:
+		*ptype = NVKM_ENGINE_NVJPG;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_NVJPEG0;
+		return true;
+	case MC_ENGINE_IDX_OFA0 ... MC_ENGINE_IDX_OFA1:
+		*ptype = NVKM_ENGINE_OFA;
+		*pinst = mc_engine_idx - MC_ENGINE_IDX_OFA0;
+		return true;
+	default:
+		return false;
+	}
+}
+
+static int
+r570_gsp_get_static_info(struct nvkm_gsp *gsp)
+{
+	GspStaticConfigInfo *rpc;
+	u32 gpc_mask;
+	u32 tpc_mask;
+	int ret;
+
+	rpc = nvkm_gsp_rpc_rd(gsp, NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO, sizeof(*rpc));
+	if (IS_ERR(rpc))
+		return PTR_ERR(rpc);
+
+	gsp->internal.client.object.client = &gsp->internal.client;
+	gsp->internal.client.object.parent = NULL;
+	gsp->internal.client.object.handle = rpc->hInternalClient;
+	gsp->internal.client.gsp = gsp;
+	INIT_LIST_HEAD(&gsp->internal.client.events);
+
+	gsp->internal.device.object.client = &gsp->internal.client;
+	gsp->internal.device.object.parent = &gsp->internal.client.object;
+	gsp->internal.device.object.handle = rpc->hInternalDevice;
+
+	gsp->internal.device.subdevice.client = &gsp->internal.client;
+	gsp->internal.device.subdevice.parent = &gsp->internal.device.object;
+	gsp->internal.device.subdevice.handle = rpc->hInternalSubdevice;
+
+	gsp->bar.rm_bar1_pdb = rpc->bar1PdeBase;
+	gsp->bar.rm_bar2_pdb = rpc->bar2PdeBase;
+
+	r535_gsp_get_static_info_fb(gsp, &rpc->fbRegionInfoParams);
+
+	if (gsp->rm->wpr->offset_set_by_acr) {
+		GspFwWprMeta *meta = gsp->wpr_meta.data;
+
+		meta->nonWprHeapOffset = rpc->fwWprLayoutOffset.nonWprHeapOffset;
+		meta->frtsOffset = rpc->fwWprLayoutOffset.frtsOffset;
+	}
+
+	nvkm_gsp_rpc_done(gsp, rpc);
+
+	ret = r570_gr_gpc_mask(gsp, &gpc_mask);
+	if (ret)
+		return ret;
+
+	for (int gpc = 0; gpc < 32; gpc++) {
+		if (gpc_mask & BIT(gpc)) {
+			ret = r570_gr_tpc_mask(gsp, gpc, &tpc_mask);
+			if (ret)
+				return ret;
+
+			gsp->gr.tpcs += hweight32(tpc_mask);
+			gsp->gr.gpcs++;
+		}
+	}
+
+	return 0;
+}
+
+static void
+r570_gsp_acpi_info(struct nvkm_gsp *gsp, ACPI_METHOD_DATA *acpi)
+{
+#if defined(CONFIG_ACPI) && defined(CONFIG_X86)
+	acpi_handle handle = ACPI_HANDLE(gsp->subdev.device->dev);
+
+	if (!handle)
+		return;
+
+	acpi->bValid = 1;
+
+	r535_gsp_acpi_dod(handle, &acpi->dodMethodData);
+	r535_gsp_acpi_jt(handle, &acpi->jtMethodData);
+	r535_gsp_acpi_caps(handle, &acpi->capsMethodData);
+#endif
+}
+
+static int
+r570_gsp_set_system_info(struct nvkm_gsp *gsp)
+{
+	struct nvkm_device *device = gsp->subdev.device;
+	struct pci_dev *pdev = container_of(device, struct nvkm_device_pci, device)->pdev;
+	GspSystemInfo *info;
+
+	if (WARN_ON(device->type == NVKM_DEVICE_TEGRA))
+		return -ENOSYS;
+
+	info = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_SET_SYSTEM_INFO, sizeof(*info));
+	if (IS_ERR(info))
+		return PTR_ERR(info);
+
+	info->gpuPhysAddr = device->func->resource_addr(device, NVKM_BAR0_PRI);
+	info->gpuPhysFbAddr = device->func->resource_addr(device, NVKM_BAR1_FB);
+	info->gpuPhysInstAddr = device->func->resource_addr(device, NVKM_BAR2_INST);
+	info->nvDomainBusDeviceFunc = pci_dev_id(pdev);
+	info->maxUserVa = TASK_SIZE;
+	info->pciConfigMirrorBase = device->pci->func->cfg.addr;
+	info->pciConfigMirrorSize = device->pci->func->cfg.size;
+	info->PCIDeviceID = (pdev->device << 16) | pdev->vendor;
+	info->PCISubDeviceID = (pdev->subsystem_device << 16) | pdev->subsystem_vendor;
+	info->PCIRevisionID = pdev->revision;
+	r570_gsp_acpi_info(gsp, &info->acpiMethodData);
+	info->bIsPrimary = video_is_primary_device(device->dev);
+	info->bPreserveVideoMemoryAllocations = false;
+
+	return nvkm_gsp_rpc_wr(gsp, info, NVKM_GSP_RPC_REPLY_NOWAIT);
+}
+
+static void
+r570_gsp_set_rmargs(struct nvkm_gsp *gsp, bool resume)
+{
+	GSP_ARGUMENTS_CACHED *args;
+
+	args = gsp->rmargs.data;
+	args->messageQueueInitArguments.sharedMemPhysAddr = gsp->shm.mem.addr;
+	args->messageQueueInitArguments.pageTableEntryCount = gsp->shm.ptes.nr;
+	args->messageQueueInitArguments.cmdQueueOffset =
+		(u8 *)gsp->shm.cmdq.ptr - (u8 *)gsp->shm.mem.data;
+	args->messageQueueInitArguments.statQueueOffset =
+		(u8 *)gsp->shm.msgq.ptr - (u8 *)gsp->shm.mem.data;
+
+	if (!resume) {
+		args->srInitArguments.oldLevel = 0;
+		args->srInitArguments.flags = 0;
+		args->srInitArguments.bInPMTransition = 0;
+	} else {
+		args->srInitArguments.oldLevel = NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_3;
+		args->srInitArguments.flags = 0;
+		args->srInitArguments.bInPMTransition = 1;
+	}
+
+	args->bDmemStack = 1;
+}
+
+const struct nvkm_rm_api_gsp
+r570_gsp = {
+	.set_rmargs = r570_gsp_set_rmargs,
+	.set_system_info = r570_gsp_set_system_info,
+	.get_static_info = r570_gsp_get_static_info,
+	.xlat_mc_engine_idx = r570_gsp_xlat_mc_engine_idx,
+	.drop_post_nocat_record = r570_gsp_drop_post_nocat_record,
+	.sr_data_size = r570_gsp_sr_data_size,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/client.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/client.h
new file mode 100644
index 000000000000..e8714e0abc37
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/client.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_CLIENT_H__
+#define __NVRM_CLIENT_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define NV01_ROOT        (0x0U) /* finn: Evaluated from "NV0000_ALLOC_PARAMETERS_MESSAGE_ID" */
+
+#define NV_PROC_NAME_MAX_LENGTH       100U
+
+typedef struct NV0000_ALLOC_PARAMETERS {
+    NvHandle hClient; /* CORERM-2934: hClient must remain the first member until all allocations use these params */
+    NvU32    processID;
+    char     processName[NV_PROC_NAME_MAX_LENGTH];
+    NV_DECLARE_ALIGNED(NvP64 pOsPidInfo, 8);
+} NV0000_ALLOC_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/disp.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/disp.h
new file mode 100644
index 000000000000..06e972835d77
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/disp.h
@@ -0,0 +1,355 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_DISP_H__
+#define __NVRM_DISP_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_GET_STATIC_INFO (0x20800a01) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS {
+    NvU32  feHwSysCap;
+    NvU32  windowPresentMask;
+    NvBool bFbRemapperEnabled;
+    NvU32  numHeads;
+    NvU32  i2cPort;
+    NvU32  internalDispActiveMask;
+    NvU32  embeddedDisplayPortMask;
+    NvBool bExternalMuxSupported;
+    NvBool bInternalMuxSupported;
+    NvU32  numDispChannels;
+} NV2080_CTRL_INTERNAL_DISPLAY_GET_STATIC_INFO_PARAMS;
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_SUPPORTED (0x730107U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayMask;
+    NvU32 displayMaskDDC;
+} NV0073_CTRL_SYSTEM_GET_SUPPORTED_PARAMS;
+
+#define NV0073_CTRL_MAX_CONNECTORS                  4U
+
+#define NV0073_CTRL_CMD_SPECIFIC_GET_CONNECTOR_DATA (0x730250U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 flags;
+    NvU32 DDCPartners;
+    NvU32 count;
+    struct {
+        NvU32 index;
+        NvU32 type;
+        NvU32 location;
+    } data[NV0073_CTRL_MAX_CONNECTORS];
+    NvU32 platform;
+} NV0073_CTRL_SPECIFIC_GET_CONNECTOR_DATA_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_GET_CAPS   (0x731369U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_CMD_DSC_CAP_PARAMS {
+    NvBool bDscSupported;
+    NvU32  encoderColorFormatMask;
+    NvU32  lineBufferSizeKB;
+    NvU32  rateBufferSizeKB;
+    NvU32  bitsPerPixelPrecision;
+    NvU32  maxNumHztSlices;
+    NvU32  lineBufferBitDepth;
+} NV0073_CTRL_CMD_DSC_CAP_PARAMS;
+
+typedef struct NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS {
+    NvU32                          subDeviceInstance;
+    NvU32                          sorIndex;
+    NvU32                          maxLinkRate;
+    NvU32                          dpVersionsSupported;
+    NvU32                          UHBRSupportedByGpu;
+    NvU32                          minPClkForCompressed;
+    NvBool                         bIsMultistreamSupported;
+    NvBool                         bIsSCEnabled;
+    NvBool                         bHasIncreasedWatermarkLimits;
+    NvBool                         bIsPC2Disabled;
+    NvBool                         isSingleHeadMSTSupported;
+    NvBool                         bFECSupported;
+    NvBool                         bIsTrainPhyRepeater;
+    NvBool                         bOverrideLinkBw;
+    NvBool                         bUseRgFlushSequence;
+    NvBool                         bSupportDPDownSpread;
+    NV0073_CTRL_CMD_DSC_CAP_PARAMS DSC;
+} NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_GET_CAPS   (0x731369U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID" */
+#define NV0073_CTRL_CMD_DP_GET_CAPS_PARAMS_MESSAGE_ID (0x69U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2                0:0
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2_NO              (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_2_YES             (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4                1:1
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4_NO              (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP1_4_YES             (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP2_0                2:2
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP2_0_NO              (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DP_VERSIONS_SUPPORTED_DP2_0_YES             (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE                           2:0
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_NONE                          (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_1_62                          (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_2_70                          (0x00000002U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_5_40                          (0x00000003U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_MAX_LINK_RATE_8_10                          (0x00000004U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR10_0                     0:0
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR10_0_NO                  (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR10_0_YES                 (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR13_5                     1:1
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR13_5_NO                  (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR13_5_YES                 (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR20_0                     2:2
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR20_0_NO                  (0x00000000U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_UHBR_SUPPORTED_UHBR20_0_YES                 (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_RGB                (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_444        (0x00000002U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_NATIVE_422 (0x00000004U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_ENCODER_COLOR_FORMAT_Y_CB_CR_NATIVE_420 (0x00000008U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_16           (0x00000001U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_8            (0x00000002U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_4            (0x00000003U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1_2            (0x00000004U)
+#define NV0073_CTRL_CMD_DP_GET_CAPS_DSC_BITS_PER_PIXEL_PRECISION_1              (0x00000005U)
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_CONNECT_STATE (0x730108U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 flags;
+    NvU32 displayMask;
+    NvU32 retryTimeMs;
+} NV0073_CTRL_SYSTEM_GET_CONNECT_STATE_PARAMS;
+
+#define NV0073_CTRL_CMD_DFP_GET_INFO (0x731140U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DFP_INTERFACE_ID << 8) | NV0073_CTRL_DFP_GET_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DFP_GET_INFO_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 flags;
+    NvU32 UHBRSupportedByDfp;
+} NV0073_CTRL_DFP_GET_INFO_PARAMS;
+
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL                                       2:0
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_TMDS                       (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_LVDS                       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_SDI                        (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_DISPLAYPORT                (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_DSI                        (0x00000004U)
+#define NV0073_CTRL_DFP_FLAGS_SIGNAL_WRBK                       (0x00000005U)
+#define NV0073_CTRL_DFP_FLAGS_LANE                                         5:3
+#define NV0073_CTRL_DFP_FLAGS_LANE_NONE                         (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_SINGLE                       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_DUAL                         (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_QUAD                         (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_LANE_OCT                          (0x00000004U)
+#define NV0073_CTRL_DFP_FLAGS_LIMIT                                        6:6
+#define NV0073_CTRL_DFP_FLAGS_LIMIT_DISABLE                     (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_LIMIT_60HZ_RR                     (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER                                   7:7
+#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER_NORMAL                 (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_SLI_SCALER_DISABLE                (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE                                 8:8
+#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE_FALSE                (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_CAPABLE_TRUE                 (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE                        9:9
+#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE_FALSE       (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_LIMITED_CAPABLE_TRUE        (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE                         10:10
+#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE_FALSE          (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_RANGE_AUTO_CAPABLE_TRUE           (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE                    11:11
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE_FALSE     (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR422_CAPABLE_TRUE      (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE                    12:12
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE_FALSE     (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_FORMAT_YCBCR444_CAPABLE_TRUE      (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_TYPE_C_TO_DP_CONNECTOR                     13:13
+#define NV0073_CTRL_DFP_FLAGS_TYPE_C_TO_DP_CONNECTOR_FALSE      (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_TYPE_C_TO_DP_CONNECTOR_TRUE       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED                               14:14
+#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED_FALSE                (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_HDMI_ALLOWED_TRUE                 (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT                       15:15
+#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT_FALSE        (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_EMBEDDED_DISPLAYPORT_TRUE         (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT                         16:16
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT_NONE           (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_CONSTRAINT_PREFER_RBR     (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW                                 19:17
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_1_62GBPS               (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_2_70GBPS               (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_5_40GBPS               (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_DP_LINK_BW_8_10GBPS               (0x00000004U)
+#define NV0073_CTRL_DFP_FLAGS_LINK                                       21:20
+#define NV0073_CTRL_DFP_FLAGS_LINK_NONE                         (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_LINK_SINGLE                       (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_LINK_DUAL                         (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID                           22:22
+#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID_FALSE            (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DP_FORCE_RM_EDID_TRUE             (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID                              24:23
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_NONE            (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_A               (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_B               (0x00000002U)
+#define NV0073_CTRL_DFP_FLAGS_DSI_DEVICE_ID_DSI_GANGED          (0x00000003U)
+#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED                   25:25
+#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED_FALSE    (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DP_POST_CURSOR2_DISABLED_TRUE     (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS_DP_PHY_REPEATER_COUNT                      29:26
+#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE                        30:30
+#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE_FALSE         (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS_DYNAMIC_MUX_CAPABLE_TRUE          (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_10_0GBPS                  0:0
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_10_0GBPS_FALSE (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_10_0GBPS_TRUE  (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_13_5GBPS                  1:1
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_13_5GBPS_FALSE (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_13_5GBPS_TRUE  (0x00000001U)
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_20_0GBPS                  2:2
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_20_0GBPS_FALSE (0x00000000U)
+#define NV0073_CTRL_DFP_FLAGS2_DP_UHBR_SUPPORTED_20_0GBPS_TRUE  (0x00000001U)
+
+#define NV0073_CTRL_CMD_SYSTEM_GET_ACTIVE                (0x73010cU) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SYSTEM_INTERFACE_ID << 8) | NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 head;
+    NvU32 flags;
+    NvU32 displayId;
+} NV0073_CTRL_SYSTEM_GET_ACTIVE_PARAMS;
+
+#define NV0073_CTRL_CMD_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS (0x730292U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_SET_BACKLIGHT_BRIGHTNESS_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_CMD_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS (0x730291U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_SPECIFIC_INTERFACE_ID << 8) | NV0073_CTRL_SPECIFIC_GET_BACKLIGHT_BRIGHTNESS_PARAMS_MESSAGE_ID" */
+
+typedef struct NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  displayId;
+    NvU32  brightness;
+    NvBool bUncalibrated;
+    NvU8   brightnessType;
+} NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES (0x731377U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS_MESSAGE_ID" */
+
+#define NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES        8U
+
+typedef struct NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS {
+    // In
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU16 linkRateTbl[NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES];
+
+    // Out
+    NvU16 linkBwTbl[NV0073_CTRL_DP_MAX_INDEXED_LINK_RATES];
+    NvU8  linkBwCount;
+} NV0073_CTRL_CMD_DP_CONFIG_INDEXED_LINK_RATES_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_CTRL                         (0x731343U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_CTRL_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DP_CTRL_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 cmd;
+    NvU32 data;
+    NvU32 err;
+    NvU32 retryTimeMs;
+    NvU32 eightLaneDpcdBaseAddr;
+} NV0073_CTRL_DP_CTRL_PARAMS;
+
+typedef struct NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS {
+    NvU32  subDeviceInstance;
+    NvU32  head;
+    NvU32  sorIndex;
+    NvU32  dpLink;
+
+    NvBool bEnableOverride;
+    NvBool bMST;
+    NvU32  singleHeadMultistreamMode;
+    NvU32  hBlankSym;
+    NvU32  vBlankSym;
+    NvU32  colorFormat;
+    NvBool bEnableTwoHeadOneOr;
+
+    struct {
+        NvU32  slotStart;
+        NvU32  slotEnd;
+        NvU32  PBN;
+        NvU32  Timeslice;
+        NvBool sendACT;          // deprecated -Use NV0073_CTRL_CMD_DP_SEND_ACT
+        NvU32  singleHeadMSTPipeline;
+        NvBool bEnableAudioOverRightPanel;
+    } MST;
+
+    struct {
+        NvBool bEnhancedFraming;
+        NvU32  tuSize;
+        NvU32  waterMark;
+        NvBool bEnableAudioOverRightPanel;
+    } SST;
+} NV0073_CTRL_CMD_DP_CONFIG_STREAM_PARAMS;
+
+#define NV0073_CTRL_CMD_DP_SET_AUDIO_MUTESTREAM             (0x731359U) /* finn: Evaluated from "(FINN_NV04_DISPLAY_COMMON_DP_INTERFACE_ID << 8) | NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS_MESSAGE_ID" */
+typedef struct NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS {
+    NvU32 subDeviceInstance;
+    NvU32 displayId;
+    NvU32 mute;
+} NV0073_CTRL_DP_SET_AUDIO_MUTESTREAM_PARAMS;
+
+#define NV2080_CTRL_CMD_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER (0x20800a58) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS {
+    NvU32  addressSpace;
+    NV_DECLARE_ALIGNED(NvU64 physicalAddr, 8);
+    NV_DECLARE_ALIGNED(NvU64 limit, 8);
+    NvU32  cacheSnoop;
+    NvU32  hclass;
+    NvU32  channelInstance;
+    NvBool valid;
+    NvU32  pbTargetAperture;
+    NvU32  channelPBSize;
+    NvU32  subDeviceId;
+} NV2080_CTRL_INTERNAL_DISPLAY_CHANNEL_PUSHBUFFER_PARAMS;
+
+#define ADDR_SYSMEM                    (1)
+
+#define ADDR_FBMEM      2         // Frame buffer memory space
+
+typedef enum
+{
+    PB_SIZE_4KB = 0,
+    PB_SIZE_8KB,
+    PB_SIZE_16KB,
+    PB_SIZE_32KB,
+    PB_SIZE_64KB
+} ChannelPBSize;
+
+typedef struct
+{
+    NvV32         channelInstance;            // One of the n channel instances of a given channel type.
+                                              // Note that core channel has only one instance
+                                              // while all others have two (one per head).
+    NvHandle      hObjectBuffer;              // ctx dma handle for DMA push buffer
+    NvHandle      hObjectNotify;              // ctx dma handle for an area (of type NvNotification defined in sdk/nvidia/inc/nvtypes.h) where RM can write errors/notifications
+    NvU32         offset;                     // Initial offset for put/get, usually zero.
+    NvP64         pControl NV_ALIGN_BYTES(8); // pControl gives virt addr of UDISP GET/PUT regs
+
+    NvU32         flags;
+    ChannelPBSize channelPBSize;              // Size of Push Buffer requested by client (allowed values in enum)
+#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB                1:1
+#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB_YES            0x00000000
+#define NV50VAIO_CHANNELDMA_ALLOCATION_FLAGS_CONNECT_PB_AT_GRAB_NO             0x00000001
+
+    NvU32    subDeviceId;                // One-hot encoded subDeviceId (i.e. SDM) that will be used to address the channel in the pushbuffer stream (via SSDM method)
+} NV50VAIO_CHANNELDMA_ALLOCATION_PARAMETERS;
+
+#define NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_TYPE_PERCENT100        1
+#define NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_TYPE_PERCENT1000       2
+#define NV0073_CTRL_SPECIFIC_BACKLIGHT_BRIGHTNESS_TYPE_NITS              3
+
+typedef enum
+{
+    IOVA,
+    PHYS_NVM,
+    PHYS_PCI,
+    PHYS_PCI_COHERENT
+} PBTARGETAPERTURE;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/engine.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/engine.h
new file mode 100644
index 000000000000..7997050a4f29
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/engine.h
@@ -0,0 +1,318 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_ENGINE_H__
+#define __NVRM_ENGINE_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define MC_ENGINE_IDX_NULL                          0 // This must be 0
+#define MC_ENGINE_IDX_TMR                           1
+#define MC_ENGINE_IDX_DISP                          2
+#define MC_ENGINE_IDX_FB                            3
+#define MC_ENGINE_IDX_FIFO                          4
+#define MC_ENGINE_IDX_VIDEO                         5
+#define MC_ENGINE_IDX_MD                            6
+#define MC_ENGINE_IDX_BUS                           7
+#define MC_ENGINE_IDX_PMGR                          8
+#define MC_ENGINE_IDX_VP2                           9
+#define MC_ENGINE_IDX_CIPHER                        10
+#define MC_ENGINE_IDX_BIF                           11
+#define MC_ENGINE_IDX_PPP                           12
+#define MC_ENGINE_IDX_PRIVRING                      13
+#define MC_ENGINE_IDX_PMU                           14
+#define MC_ENGINE_IDX_CE0                           15
+#define MC_ENGINE_IDX_CE1                           16
+#define MC_ENGINE_IDX_CE2                           17
+#define MC_ENGINE_IDX_CE3                           18
+#define MC_ENGINE_IDX_CE4                           19
+#define MC_ENGINE_IDX_CE5                           20
+#define MC_ENGINE_IDX_CE6                           21
+#define MC_ENGINE_IDX_CE7                           22
+#define MC_ENGINE_IDX_CE8                           23
+#define MC_ENGINE_IDX_CE9                           24
+#define MC_ENGINE_IDX_CE10                          25
+#define MC_ENGINE_IDX_CE11                          26
+#define MC_ENGINE_IDX_CE12                          27
+#define MC_ENGINE_IDX_CE13                          28
+#define MC_ENGINE_IDX_CE14                          29
+#define MC_ENGINE_IDX_CE15                          30
+#define MC_ENGINE_IDX_CE16                          31
+#define MC_ENGINE_IDX_CE17                          32
+#define MC_ENGINE_IDX_CE18                          33
+#define MC_ENGINE_IDX_CE19                          34
+#define MC_ENGINE_IDX_CE_MAX                        MC_ENGINE_IDX_CE19
+#define MC_ENGINE_IDX_VIC                           35
+#define MC_ENGINE_IDX_ISOHUB                        36
+#define MC_ENGINE_IDX_VGPU                          37
+#define MC_ENGINE_IDX_NVENC                         38
+#define MC_ENGINE_IDX_NVENC1                        39
+#define MC_ENGINE_IDX_NVENC2                        40
+#define MC_ENGINE_IDX_NVENC3                        41
+#define MC_ENGINE_IDX_C2C                           42
+#define MC_ENGINE_IDX_LTC                           43
+#define MC_ENGINE_IDX_FBHUB                         44
+#define MC_ENGINE_IDX_HDACODEC                      45
+#define MC_ENGINE_IDX_GMMU                          46
+#define MC_ENGINE_IDX_SEC2                          47
+#define MC_ENGINE_IDX_FSP                           48
+#define MC_ENGINE_IDX_NVLINK                        49
+#define MC_ENGINE_IDX_GSP                           50
+#define MC_ENGINE_IDX_NVJPG                         51
+#define MC_ENGINE_IDX_NVJPEG                        MC_ENGINE_IDX_NVJPG
+#define MC_ENGINE_IDX_NVJPEG0                       MC_ENGINE_IDX_NVJPEG
+#define MC_ENGINE_IDX_NVJPEG1                       52
+#define MC_ENGINE_IDX_NVJPEG2                       53
+#define MC_ENGINE_IDX_NVJPEG3                       54
+#define MC_ENGINE_IDX_NVJPEG4                       55
+#define MC_ENGINE_IDX_NVJPEG5                       56
+#define MC_ENGINE_IDX_NVJPEG6                       57
+#define MC_ENGINE_IDX_NVJPEG7                       58
+#define MC_ENGINE_IDX_REPLAYABLE_FAULT              59
+#define MC_ENGINE_IDX_ACCESS_CNTR                   60
+#define MC_ENGINE_IDX_NON_REPLAYABLE_FAULT          61
+#define MC_ENGINE_IDX_REPLAYABLE_FAULT_ERROR        62
+#define MC_ENGINE_IDX_NON_REPLAYABLE_FAULT_ERROR    63
+#define MC_ENGINE_IDX_INFO_FAULT                    64
+#define MC_ENGINE_IDX_BSP                           65
+#define MC_ENGINE_IDX_NVDEC                         MC_ENGINE_IDX_BSP
+#define MC_ENGINE_IDX_NVDEC0                        MC_ENGINE_IDX_NVDEC
+#define MC_ENGINE_IDX_NVDEC1                        66
+#define MC_ENGINE_IDX_NVDEC2                        67
+#define MC_ENGINE_IDX_NVDEC3                        68
+#define MC_ENGINE_IDX_NVDEC4                        69
+#define MC_ENGINE_IDX_NVDEC5                        70
+#define MC_ENGINE_IDX_NVDEC6                        71
+#define MC_ENGINE_IDX_NVDEC7                        72
+#define MC_ENGINE_IDX_CPU_DOORBELL                  73
+#define MC_ENGINE_IDX_PRIV_DOORBELL                 74
+#define MC_ENGINE_IDX_MMU_ECC_ERROR                 75
+#define MC_ENGINE_IDX_BLG                           76
+#define MC_ENGINE_IDX_PERFMON                       77
+#define MC_ENGINE_IDX_BUF_RESET                     78
+#define MC_ENGINE_IDX_XBAR                          79
+#define MC_ENGINE_IDX_ZPW                           80
+#define MC_ENGINE_IDX_OFA0                          81
+#define MC_ENGINE_IDX_OFA1                          82
+#define MC_ENGINE_IDX_TEGRA                         83
+#define MC_ENGINE_IDX_GR                            84
+#define MC_ENGINE_IDX_GR0                           MC_ENGINE_IDX_GR
+#define MC_ENGINE_IDX_GR1                           85
+#define MC_ENGINE_IDX_GR2                           86
+#define MC_ENGINE_IDX_GR3                           87
+#define MC_ENGINE_IDX_GR4                           88
+#define MC_ENGINE_IDX_GR5                           89
+#define MC_ENGINE_IDX_GR6                           90
+#define MC_ENGINE_IDX_GR7                           91
+#define MC_ENGINE_IDX_ESCHED                        92
+#define MC_ENGINE_IDX_ESCHED__SIZE                  64
+#define MC_ENGINE_IDX_GR_FECS_LOG                   156
+#define MC_ENGINE_IDX_GR0_FECS_LOG                  MC_ENGINE_IDX_GR_FECS_LOG
+#define MC_ENGINE_IDX_GR1_FECS_LOG                  157
+#define MC_ENGINE_IDX_GR2_FECS_LOG                  158
+#define MC_ENGINE_IDX_GR3_FECS_LOG                  159
+#define MC_ENGINE_IDX_GR4_FECS_LOG                  160
+#define MC_ENGINE_IDX_GR5_FECS_LOG                  161
+#define MC_ENGINE_IDX_GR6_FECS_LOG                  162
+#define MC_ENGINE_IDX_GR7_FECS_LOG                  163
+#define MC_ENGINE_IDX_TMR_SWRL                      164
+#define MC_ENGINE_IDX_DISP_GSP                      165
+#define MC_ENGINE_IDX_REPLAYABLE_FAULT_CPU          166
+#define MC_ENGINE_IDX_NON_REPLAYABLE_FAULT_CPU      167
+#define MC_ENGINE_IDX_PXUC                          168
+#define MC_ENGINE_IDX_SYSLTC                        169
+#define MC_ENGINE_IDX_LRCC                          170
+#define MC_ENGINE_IDX_GSPLITE                       171
+#define MC_ENGINE_IDX_GSPLITE0                      MC_ENGINE_IDX_GSPLITE
+#define MC_ENGINE_IDX_GSPLITE1                      172
+#define MC_ENGINE_IDX_GSPLITE2                      173
+#define MC_ENGINE_IDX_GSPLITE3                      174
+#define MC_ENGINE_IDX_GSPLITE_MAX                   MC_ENGINE_IDX_GSPLITE3
+#define MC_ENGINE_IDX_DPAUX                         175
+#define MC_ENGINE_IDX_DISP_LOW                      176
+#define MC_ENGINE_IDX_MAX                           177
+
+typedef enum
+{
+    RM_ENGINE_TYPE_NULL                            =       (0x00000000),
+    RM_ENGINE_TYPE_GR0                             =       (0x00000001),
+    RM_ENGINE_TYPE_GR1                             =       (0x00000002),
+    RM_ENGINE_TYPE_GR2                             =       (0x00000003),
+    RM_ENGINE_TYPE_GR3                             =       (0x00000004),
+    RM_ENGINE_TYPE_GR4                             =       (0x00000005),
+    RM_ENGINE_TYPE_GR5                             =       (0x00000006),
+    RM_ENGINE_TYPE_GR6                             =       (0x00000007),
+    RM_ENGINE_TYPE_GR7                             =       (0x00000008),
+    RM_ENGINE_TYPE_COPY0                           =       (0x00000009),
+    RM_ENGINE_TYPE_COPY1                           =       (0x0000000a),
+    RM_ENGINE_TYPE_COPY2                           =       (0x0000000b),
+    RM_ENGINE_TYPE_COPY3                           =       (0x0000000c),
+    RM_ENGINE_TYPE_COPY4                           =       (0x0000000d),
+    RM_ENGINE_TYPE_COPY5                           =       (0x0000000e),
+    RM_ENGINE_TYPE_COPY6                           =       (0x0000000f),
+    RM_ENGINE_TYPE_COPY7                           =       (0x00000010),
+    RM_ENGINE_TYPE_COPY8                           =       (0x00000011),
+    RM_ENGINE_TYPE_COPY9                           =       (0x00000012),
+    RM_ENGINE_TYPE_COPY10                          =       (0x00000013),
+    RM_ENGINE_TYPE_COPY11                          =       (0x00000014),
+    RM_ENGINE_TYPE_COPY12                          =       (0x00000015),
+    RM_ENGINE_TYPE_COPY13                          =       (0x00000016),
+    RM_ENGINE_TYPE_COPY14                          =       (0x00000017),
+    RM_ENGINE_TYPE_COPY15                          =       (0x00000018),
+    RM_ENGINE_TYPE_COPY16                          =       (0x00000019),
+    RM_ENGINE_TYPE_COPY17                          =       (0x0000001a),
+    RM_ENGINE_TYPE_COPY18                          =       (0x0000001b),
+    RM_ENGINE_TYPE_COPY19                          =       (0x0000001c),
+    RM_ENGINE_TYPE_NVDEC0                          =       (0x0000001d),
+    RM_ENGINE_TYPE_NVDEC1                          =       (0x0000001e),
+    RM_ENGINE_TYPE_NVDEC2                          =       (0x0000001f),
+    RM_ENGINE_TYPE_NVDEC3                          =       (0x00000020),
+    RM_ENGINE_TYPE_NVDEC4                          =       (0x00000021),
+    RM_ENGINE_TYPE_NVDEC5                          =       (0x00000022),
+    RM_ENGINE_TYPE_NVDEC6                          =       (0x00000023),
+    RM_ENGINE_TYPE_NVDEC7                          =       (0x00000024),
+    RM_ENGINE_TYPE_NVENC0                          =       (0x00000025),
+    RM_ENGINE_TYPE_NVENC1                          =       (0x00000026),
+    RM_ENGINE_TYPE_NVENC2                          =       (0x00000027),
+    // Bug 4175886 - Use this new value for all chips once GB20X is released
+    RM_ENGINE_TYPE_NVENC3                          =       (0x00000028),
+    RM_ENGINE_TYPE_VP                              =       (0x00000029),
+    RM_ENGINE_TYPE_ME                              =       (0x0000002a),
+    RM_ENGINE_TYPE_PPP                             =       (0x0000002b),
+    RM_ENGINE_TYPE_MPEG                            =       (0x0000002c),
+    RM_ENGINE_TYPE_SW                              =       (0x0000002d),
+    RM_ENGINE_TYPE_TSEC                            =       (0x0000002e),
+    RM_ENGINE_TYPE_VIC                             =       (0x0000002f),
+    RM_ENGINE_TYPE_MP                              =       (0x00000030),
+    RM_ENGINE_TYPE_SEC2                            =       (0x00000031),
+    RM_ENGINE_TYPE_HOST                            =       (0x00000032),
+    RM_ENGINE_TYPE_DPU                             =       (0x00000033),
+    RM_ENGINE_TYPE_PMU                             =       (0x00000034),
+    RM_ENGINE_TYPE_FBFLCN                          =       (0x00000035),
+    RM_ENGINE_TYPE_NVJPEG0                         =       (0x00000036),
+    RM_ENGINE_TYPE_NVJPEG1                         =       (0x00000037),
+    RM_ENGINE_TYPE_NVJPEG2                         =       (0x00000038),
+    RM_ENGINE_TYPE_NVJPEG3                         =       (0x00000039),
+    RM_ENGINE_TYPE_NVJPEG4                         =       (0x0000003a),
+    RM_ENGINE_TYPE_NVJPEG5                         =       (0x0000003b),
+    RM_ENGINE_TYPE_NVJPEG6                         =       (0x0000003c),
+    RM_ENGINE_TYPE_NVJPEG7                         =       (0x0000003d),
+    RM_ENGINE_TYPE_OFA0                            =       (0x0000003e),
+    RM_ENGINE_TYPE_OFA1                            =       (0x0000003f),
+    RM_ENGINE_TYPE_RESERVED40                      =       (0x00000040),
+    RM_ENGINE_TYPE_RESERVED41                      =       (0x00000041),
+    RM_ENGINE_TYPE_RESERVED42                      =       (0x00000042),
+    RM_ENGINE_TYPE_RESERVED43                      =       (0x00000043),
+    RM_ENGINE_TYPE_RESERVED44                      =       (0x00000044),
+    RM_ENGINE_TYPE_RESERVED45                      =       (0x00000045),
+    RM_ENGINE_TYPE_RESERVED46                      =       (0x00000046),
+    RM_ENGINE_TYPE_RESERVED47                      =       (0x00000047),
+    RM_ENGINE_TYPE_RESERVED48                      =       (0x00000048),
+    RM_ENGINE_TYPE_RESERVED49                      =       (0x00000049),
+    RM_ENGINE_TYPE_RESERVED4a                      =       (0x0000004a),
+    RM_ENGINE_TYPE_RESERVED4b                      =       (0x0000004b),
+    RM_ENGINE_TYPE_RESERVED4c                      =       (0x0000004c),
+    RM_ENGINE_TYPE_RESERVED4d                      =       (0x0000004d),
+    RM_ENGINE_TYPE_RESERVED4e                      =       (0x0000004e),
+    RM_ENGINE_TYPE_RESERVED4f                      =       (0x0000004f),
+    RM_ENGINE_TYPE_RESERVED50                      =       (0x00000050),
+    RM_ENGINE_TYPE_RESERVED51                      =       (0x00000051),
+    RM_ENGINE_TYPE_RESERVED52                      =       (0x00000052),
+    RM_ENGINE_TYPE_RESERVED53                      =       (0x00000053),
+    RM_ENGINE_TYPE_LAST                            =       (0x00000054),
+} RM_ENGINE_TYPE;
+
+#define NV2080_ENGINE_TYPE_NULL                       (0x00000000)
+#define NV2080_ENGINE_TYPE_GRAPHICS                   (0x00000001)
+#define NV2080_ENGINE_TYPE_GR0                        NV2080_ENGINE_TYPE_GRAPHICS
+#define NV2080_ENGINE_TYPE_GR1                        (0x00000002)
+#define NV2080_ENGINE_TYPE_GR2                        (0x00000003)
+#define NV2080_ENGINE_TYPE_GR3                        (0x00000004)
+#define NV2080_ENGINE_TYPE_GR4                        (0x00000005)
+#define NV2080_ENGINE_TYPE_GR5                        (0x00000006)
+#define NV2080_ENGINE_TYPE_GR6                        (0x00000007)
+#define NV2080_ENGINE_TYPE_GR7                        (0x00000008)
+#define NV2080_ENGINE_TYPE_COPY0                      (0x00000009)
+#define NV2080_ENGINE_TYPE_COPY1                      (0x0000000a)
+#define NV2080_ENGINE_TYPE_COPY2                      (0x0000000b)
+#define NV2080_ENGINE_TYPE_COPY3                      (0x0000000c)
+#define NV2080_ENGINE_TYPE_COPY4                      (0x0000000d)
+#define NV2080_ENGINE_TYPE_COPY5                      (0x0000000e)
+#define NV2080_ENGINE_TYPE_COPY6                      (0x0000000f)
+#define NV2080_ENGINE_TYPE_COPY7                      (0x00000010)
+#define NV2080_ENGINE_TYPE_COPY8                      (0x00000011)
+#define NV2080_ENGINE_TYPE_COPY9                      (0x00000012)
+#define NV2080_ENGINE_TYPE_BSP                        (0x00000013)
+#define NV2080_ENGINE_TYPE_NVDEC0                     NV2080_ENGINE_TYPE_BSP
+#define NV2080_ENGINE_TYPE_NVDEC1                     (0x00000014)
+#define NV2080_ENGINE_TYPE_NVDEC2                     (0x00000015)
+#define NV2080_ENGINE_TYPE_NVDEC3                     (0x00000016)
+#define NV2080_ENGINE_TYPE_NVDEC4                     (0x00000017)
+#define NV2080_ENGINE_TYPE_NVDEC5                     (0x00000018)
+#define NV2080_ENGINE_TYPE_NVDEC6                     (0x00000019)
+#define NV2080_ENGINE_TYPE_NVDEC7                     (0x0000001a)
+#define NV2080_ENGINE_TYPE_MSENC                      (0x0000001b)
+#define NV2080_ENGINE_TYPE_NVENC0                      NV2080_ENGINE_TYPE_MSENC  /* Mutually exclusive alias */
+#define NV2080_ENGINE_TYPE_NVENC1                     (0x0000001c)
+#define NV2080_ENGINE_TYPE_NVENC2                     (0x0000001d)
+#define NV2080_ENGINE_TYPE_VP                         (0x0000001e)
+#define NV2080_ENGINE_TYPE_ME                         (0x0000001f)
+#define NV2080_ENGINE_TYPE_PPP                        (0x00000020)
+#define NV2080_ENGINE_TYPE_MPEG                       (0x00000021)
+#define NV2080_ENGINE_TYPE_SW                         (0x00000022)
+#define NV2080_ENGINE_TYPE_CIPHER                     (0x00000023)
+#define NV2080_ENGINE_TYPE_TSEC                       NV2080_ENGINE_TYPE_CIPHER
+#define NV2080_ENGINE_TYPE_VIC                        (0x00000024)
+#define NV2080_ENGINE_TYPE_MP                         (0x00000025)
+#define NV2080_ENGINE_TYPE_SEC2                       (0x00000026)
+#define NV2080_ENGINE_TYPE_HOST                       (0x00000027)
+#define NV2080_ENGINE_TYPE_DPU                        (0x00000028)
+#define NV2080_ENGINE_TYPE_PMU                        (0x00000029)
+#define NV2080_ENGINE_TYPE_FBFLCN                     (0x0000002a)
+#define NV2080_ENGINE_TYPE_NVJPG                      (0x0000002b)
+#define NV2080_ENGINE_TYPE_NVJPEG0                     NV2080_ENGINE_TYPE_NVJPG
+#define NV2080_ENGINE_TYPE_NVJPEG1                    (0x0000002c)
+#define NV2080_ENGINE_TYPE_NVJPEG2                    (0x0000002d)
+#define NV2080_ENGINE_TYPE_NVJPEG3                    (0x0000002e)
+#define NV2080_ENGINE_TYPE_NVJPEG4                    (0x0000002f)
+#define NV2080_ENGINE_TYPE_NVJPEG5                    (0x00000030)
+#define NV2080_ENGINE_TYPE_NVJPEG6                    (0x00000031)
+#define NV2080_ENGINE_TYPE_NVJPEG7                    (0x00000032)
+#define NV2080_ENGINE_TYPE_OFA                        (0x00000033)
+#define NV2080_ENGINE_TYPE_OFA0                       NV2080_ENGINE_TYPE_OFA
+#define NV2080_ENGINE_TYPE_COPY10                     (0x00000034)
+#define NV2080_ENGINE_TYPE_COPY11                     (0x00000035)
+#define NV2080_ENGINE_TYPE_COPY12                     (0x00000036)
+#define NV2080_ENGINE_TYPE_COPY13                     (0x00000037)
+#define NV2080_ENGINE_TYPE_COPY14                     (0x00000038)
+#define NV2080_ENGINE_TYPE_COPY15                     (0x00000039)
+#define NV2080_ENGINE_TYPE_COPY16                     (0x0000003a)
+#define NV2080_ENGINE_TYPE_COPY17                     (0x0000003b)
+#define NV2080_ENGINE_TYPE_COPY18                     (0x0000003c)
+#define NV2080_ENGINE_TYPE_COPY19                     (0x0000003d)
+#define NV2080_ENGINE_TYPE_OFA1                       (0x0000003e)
+#define NV2080_ENGINE_TYPE_NVENC3                     (0x0000003f)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY0          (0x00000040)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY1          (0x00000041)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY2          (0x00000042)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY3          (0x00000043)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY4          (0x00000044)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY5          (0x00000045)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY6          (0x00000046)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY7          (0x00000047)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY8          (0x00000048)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY9          (0x00000049)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY10         (0x0000004a)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY11         (0x0000004b)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY12         (0x0000004c)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY13         (0x0000004d)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY14         (0x0000004e)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY15         (0x0000004f)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY16         (0x00000050)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY17         (0x00000051)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY18         (0x00000052)
+#define NV2080_ENGINE_TYPE_COMP_DECOMP_COPY19         (0x00000053)
+#define NV2080_ENGINE_TYPE_LAST                       (0x00000054)
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/fbsr.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/fbsr.h
new file mode 100644
index 000000000000..8af432375f7a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/fbsr.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_FBSR_H__
+#define __NVRM_FBSR_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define NV2080_CTRL_CMD_INTERNAL_FBSR_INIT (0x20800ac2) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS {
+    NvHandle hClient;
+    NvHandle hSysMem;
+    NvBool   bEnteringGcoffState;
+    NV_DECLARE_ALIGNED(NvU64 sysmemAddrOfSuspendResumeData, 8);
+} NV2080_CTRL_INTERNAL_FBSR_INIT_PARAMS;
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/alloc/alloc_channel.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/fifo.h
index 7157c7757698..2b002ca64e0f 100644
--- a/drivers/gpu/drm/nouveau/include/nvrm/535.113.01/common/sdk/nvidia/inc/alloc/alloc_channel.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/fifo.h
@@ -1,31 +1,14 @@
-#ifndef __src_common_sdk_nvidia_inc_alloc_alloc_channel_h__
-#define __src_common_sdk_nvidia_inc_alloc_alloc_channel_h__
-#include <nvrm/535.113.01/common/sdk/nvidia/inc/nvlimits.h>
-
-/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/535.113.01 */
-
-/*
- * SPDX-FileCopyrightText: Copyright (c) 2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
- * SPDX-License-Identifier: MIT
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
- */
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_FIFO_H__
+#define __NVRM_FIFO_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define NV_MAX_SUBDEVICES             8
 
 typedef struct NV_MEMORY_DESC_PARAMS {
     NV_DECLARE_ALIGNED(NvU64 base, 8);
@@ -34,137 +17,197 @@ typedef struct NV_MEMORY_DESC_PARAMS {
     NvU32 cacheAttrib;
 } NV_MEMORY_DESC_PARAMS;
 
+#define CC_CHAN_ALLOC_IV_SIZE_DWORD    3U
+
+#define CC_CHAN_ALLOC_NONCE_SIZE_DWORD 8U
+
+typedef struct NV_CHANNEL_ALLOC_PARAMS {
+
+    NvHandle hObjectError; // error context DMA
+    NvHandle hObjectBuffer; // no longer used
+    NV_DECLARE_ALIGNED(NvU64 gpFifoOffset, 8);    // offset to beginning of GP FIFO
+    NvU32    gpFifoEntries;    // number of GP FIFO entries
+
+    NvU32    flags;
+
+
+    NvHandle hContextShare; // context share handle
+    NvHandle hVASpace; // VASpace for the channel
+
+    // handle to UserD memory object for channel, ignored if hUserdMemory[0]=0
+    NvHandle hUserdMemory[NV_MAX_SUBDEVICES];
+
+    // offset to beginning of UserD within hUserdMemory[x]
+    NV_DECLARE_ALIGNED(NvU64 userdOffset[NV_MAX_SUBDEVICES], 8);
+
+    // engine type(NV2080_ENGINE_TYPE_*) with which this channel is associated
+    NvU32    engineType;
+    // Channel identifier that is unique for the duration of a RM session
+    NvU32    cid;
+    // One-hot encoded bitmask to match SET_SUBDEVICE_MASK methods
+    NvU32    subDeviceId;
+    NvHandle hObjectEccError; // ECC error context DMA
+
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS instanceMem, 8);
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS userdMem, 8);
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS ramfcMem, 8);
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS mthdbufMem, 8);
+
+    NvHandle hPhysChannelGroup;              // reserved
+    NvU32    internalFlags;                 // reserved
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS errorNotifierMem, 8); // reserved
+    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS eccErrorNotifierMem, 8); // reserved
+    NvU32    ProcessID;                 // reserved
+    NvU32    SubProcessID;                 // reserved
+
+    // IV used for CPU-side encryption / GPU-side decryption.
+    NvU32    encryptIv[CC_CHAN_ALLOC_IV_SIZE_DWORD];          // reserved
+    // IV used for CPU-side decryption / GPU-side encryption.
+    NvU32    decryptIv[CC_CHAN_ALLOC_IV_SIZE_DWORD];          // reserved
+    // Nonce used CPU-side signing / GPU-side signature verification.
+    NvU32    hmacNonce[CC_CHAN_ALLOC_NONCE_SIZE_DWORD];       // reserved
+    NvU32    tpcConfigID; // TPC Configuration Id as supported by DTD-PG Feature
+} NV_CHANNEL_ALLOC_PARAMS;
+
+typedef NV_CHANNEL_ALLOC_PARAMS NV_CHANNELGPFIFO_ALLOCATION_PARAMETERS;
+
 #define NVOS04_FLAGS_CHANNEL_TYPE                                  1:0
 #define NVOS04_FLAGS_CHANNEL_TYPE_PHYSICAL                         0x00000000
 #define NVOS04_FLAGS_CHANNEL_TYPE_VIRTUAL                          0x00000001  // OBSOLETE
 #define NVOS04_FLAGS_CHANNEL_TYPE_PHYSICAL_FOR_VIRTUAL             0x00000002  // OBSOLETE
-
 #define NVOS04_FLAGS_VPR                                           2:2
 #define NVOS04_FLAGS_VPR_FALSE                                     0x00000000
 #define NVOS04_FLAGS_VPR_TRUE                                      0x00000001
-
 #define NVOS04_FLAGS_CC_SECURE                                     2:2
 #define NVOS04_FLAGS_CC_SECURE_FALSE                               0x00000000
 #define NVOS04_FLAGS_CC_SECURE_TRUE                                0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_SKIP_MAP_REFCOUNTING                  3:3
 #define NVOS04_FLAGS_CHANNEL_SKIP_MAP_REFCOUNTING_FALSE            0x00000000
 #define NVOS04_FLAGS_CHANNEL_SKIP_MAP_REFCOUNTING_TRUE             0x00000001
-
 #define NVOS04_FLAGS_GROUP_CHANNEL_RUNQUEUE                       4:4
 #define NVOS04_FLAGS_GROUP_CHANNEL_RUNQUEUE_DEFAULT               0x00000000
 #define NVOS04_FLAGS_GROUP_CHANNEL_RUNQUEUE_ONE                   0x00000001
-
 #define NVOS04_FLAGS_PRIVILEGED_CHANNEL                           5:5
 #define NVOS04_FLAGS_PRIVILEGED_CHANNEL_FALSE                     0x00000000
 #define NVOS04_FLAGS_PRIVILEGED_CHANNEL_TRUE                      0x00000001
-
 #define NVOS04_FLAGS_DELAY_CHANNEL_SCHEDULING                     6:6
 #define NVOS04_FLAGS_DELAY_CHANNEL_SCHEDULING_FALSE               0x00000000
 #define NVOS04_FLAGS_DELAY_CHANNEL_SCHEDULING_TRUE                0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_DENY_PHYSICAL_MODE_CE                7:7
 #define NVOS04_FLAGS_CHANNEL_DENY_PHYSICAL_MODE_CE_FALSE          0x00000000
 #define NVOS04_FLAGS_CHANNEL_DENY_PHYSICAL_MODE_CE_TRUE           0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_VALUE                    10:8
-
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_FIXED                    11:11
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_FIXED_FALSE              0x00000000
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_FIXED_TRUE               0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_VALUE               20:12
-
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_FIXED               21:21
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_FIXED_FALSE         0x00000000
 #define NVOS04_FLAGS_CHANNEL_USERD_INDEX_PAGE_FIXED_TRUE          0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_DENY_AUTH_LEVEL_PRIV                 22:22
 #define NVOS04_FLAGS_CHANNEL_DENY_AUTH_LEVEL_PRIV_FALSE           0x00000000
 #define NVOS04_FLAGS_CHANNEL_DENY_AUTH_LEVEL_PRIV_TRUE            0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_SKIP_SCRUBBER                        23:23
 #define NVOS04_FLAGS_CHANNEL_SKIP_SCRUBBER_FALSE                  0x00000000
 #define NVOS04_FLAGS_CHANNEL_SKIP_SCRUBBER_TRUE                   0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_CLIENT_MAP_FIFO                      24:24
 #define NVOS04_FLAGS_CHANNEL_CLIENT_MAP_FIFO_FALSE                0x00000000
 #define NVOS04_FLAGS_CHANNEL_CLIENT_MAP_FIFO_TRUE                 0x00000001
-
 #define NVOS04_FLAGS_SET_EVICT_LAST_CE_PREFETCH_CHANNEL           25:25
 #define NVOS04_FLAGS_SET_EVICT_LAST_CE_PREFETCH_CHANNEL_FALSE     0x00000000
 #define NVOS04_FLAGS_SET_EVICT_LAST_CE_PREFETCH_CHANNEL_TRUE      0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_VGPU_PLUGIN_CONTEXT                  26:26
 #define NVOS04_FLAGS_CHANNEL_VGPU_PLUGIN_CONTEXT_FALSE            0x00000000
 #define NVOS04_FLAGS_CHANNEL_VGPU_PLUGIN_CONTEXT_TRUE             0x00000001
-
 #define NVOS04_FLAGS_CHANNEL_PBDMA_ACQUIRE_TIMEOUT                 27:27
 #define NVOS04_FLAGS_CHANNEL_PBDMA_ACQUIRE_TIMEOUT_FALSE           0x00000000
 #define NVOS04_FLAGS_CHANNEL_PBDMA_ACQUIRE_TIMEOUT_TRUE            0x00000001
-
 #define NVOS04_FLAGS_GROUP_CHANNEL_THREAD                          29:28
 #define NVOS04_FLAGS_GROUP_CHANNEL_THREAD_DEFAULT                  0x00000000
 #define NVOS04_FLAGS_GROUP_CHANNEL_THREAD_ONE                      0x00000001
 #define NVOS04_FLAGS_GROUP_CHANNEL_THREAD_TWO                      0x00000002
-
 #define NVOS04_FLAGS_MAP_CHANNEL                                   30:30
 #define NVOS04_FLAGS_MAP_CHANNEL_FALSE                             0x00000000
 #define NVOS04_FLAGS_MAP_CHANNEL_TRUE                              0x00000001
-
 #define NVOS04_FLAGS_SKIP_CTXBUFFER_ALLOC                          31:31
 #define NVOS04_FLAGS_SKIP_CTXBUFFER_ALLOC_FALSE                    0x00000000
 #define NVOS04_FLAGS_SKIP_CTXBUFFER_ALLOC_TRUE                     0x00000001
 
-#define CC_CHAN_ALLOC_IV_SIZE_DWORD    3U
-#define CC_CHAN_ALLOC_NONCE_SIZE_DWORD 8U
-
-typedef struct NV_CHANNEL_ALLOC_PARAMS {
-
-    NvHandle hObjectError; // error context DMA
-    NvHandle hObjectBuffer; // no longer used
-    NV_DECLARE_ALIGNED(NvU64 gpFifoOffset, 8);    // offset to beginning of GP FIFO
-    NvU32    gpFifoEntries;    // number of GP FIFO entries
-
-    NvU32    flags;
-
-
-    NvHandle hContextShare; // context share handle
-    NvHandle hVASpace; // VASpace for the channel
-
-    // handle to UserD memory object for channel, ignored if hUserdMemory[0]=0
-    NvHandle hUserdMemory[NV_MAX_SUBDEVICES];
-
-    // offset to beginning of UserD within hUserdMemory[x]
-    NV_DECLARE_ALIGNED(NvU64 userdOffset[NV_MAX_SUBDEVICES], 8);
-
-    // engine type(NV2080_ENGINE_TYPE_*) with which this channel is associated
-    NvU32    engineType;
-    // Channel identifier that is unique for the duration of a RM session
-    NvU32    cid;
-    // One-hot encoded bitmask to match SET_SUBDEVICE_MASK methods
-    NvU32    subDeviceId;
-    NvHandle hObjectEccError; // ECC error context DMA
-
-    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS instanceMem, 8);
-    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS userdMem, 8);
-    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS ramfcMem, 8);
-    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS mthdbufMem, 8);
-
-    NvHandle hPhysChannelGroup;              // reserved
-    NvU32    internalFlags;                 // reserved
-    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS errorNotifierMem, 8); // reserved
-    NV_DECLARE_ALIGNED(NV_MEMORY_DESC_PARAMS eccErrorNotifierMem, 8); // reserved
-    NvU32    ProcessID;                 // reserved
-    NvU32    SubProcessID;                 // reserved
-    // IV used for CPU-side encryption / GPU-side decryption.
-    NvU32    encryptIv[CC_CHAN_ALLOC_IV_SIZE_DWORD];          // reserved
-    // IV used for CPU-side decryption / GPU-side encryption.
-    NvU32    decryptIv[CC_CHAN_ALLOC_IV_SIZE_DWORD];          // reserved
-    // Nonce used CPU-side signing / GPU-side signature verification.
-    NvU32    hmacNonce[CC_CHAN_ALLOC_NONCE_SIZE_DWORD];       // reserved
-} NV_CHANNEL_ALLOC_PARAMS;
-
-typedef NV_CHANNEL_ALLOC_PARAMS NV_CHANNELGPFIFO_ALLOCATION_PARAMETERS;
-
+typedef enum {
+    /*!
+     * Initial state as passed in NV_CHANNEL_ALLOC_PARAMS by
+     * kernel CPU-RM clients.
+     */
+    ERROR_NOTIFIER_TYPE_UNKNOWN = 0,
+    /*! @brief Error notifier is explicitly not set.
+     *
+     * The corresponding hErrorContext or hEccErrorContext must be
+     * NV01_NULL_OBJECT.
+     */
+    ERROR_NOTIFIER_TYPE_NONE,
+    /*! @brief Error notifier is a ContextDma */
+    ERROR_NOTIFIER_TYPE_CTXDMA,
+    /*! @brief Error notifier is a NvNotification array in sysmem/vidmem */
+    ERROR_NOTIFIER_TYPE_MEMORY
+} ErrorNotifierType;
+
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE                       1:0
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_USER                  0x0
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_ADMIN                 0x1
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_PRIVILEGE_KERNEL                0x2
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE             3:2
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_UNKNOWN     ERROR_NOTIFIER_TYPE_UNKNOWN
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_NONE        ERROR_NOTIFIER_TYPE_NONE
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_CTXDMA      ERROR_NOTIFIER_TYPE_CTXDMA
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ERROR_NOTIFIER_TYPE_MEMORY      ERROR_NOTIFIER_TYPE_MEMORY
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE         5:4
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_UNKNOWN ERROR_NOTIFIER_TYPE_UNKNOWN
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_NONE    ERROR_NOTIFIER_TYPE_NONE
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_CTXDMA  ERROR_NOTIFIER_TYPE_CTXDMA
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_ECC_ERROR_NOTIFIER_TYPE_MEMORY  ERROR_NOTIFIER_TYPE_MEMORY
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_GSP_OWNED                       6:6
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_GSP_OWNED_NO                    0x0
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_GSP_OWNED_YES                   0x1
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_UVM_OWNED                       7:7
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_UVM_OWNED_NO                    0x0
+#define NV_KERNELCHANNEL_ALLOC_INTERNALFLAGS_UVM_OWNED_YES                   0x1
+
+typedef struct rpc_rc_triggered_v17_02
+{
+    NvU32      nv2080EngineType;
+    NvU32      chid;
+    NvU32      gfid;
+    NvU32      exceptLevel;
+    NvU32      exceptType;
+    NvU32      scope;
+    NvU16      partitionAttributionId;
+    NvU32      mmuFaultAddrLo;
+    NvU32      mmuFaultAddrHi;
+    NvU32      mmuFaultType;
+    NvBool     bCallbackNeeded;
+    NvU32      rcJournalBufferSize;
+    NvU8       rcJournalBuffer[];
+} rpc_rc_triggered_v17_02;
+
+#define NV2080_CTRL_GPU_MAX_CONSTRUCTED_FALCONS         0x40
+
+typedef struct NV2080_CTRL_GPU_CONSTRUCTED_FALCON_INFO {
+    NvU32 engDesc;
+    NvU32 ctxAttr;
+    NvU32 ctxBufferSize;
+    NvU32 addrSpaceList;
+    NvU32 registerBase;
+} NV2080_CTRL_GPU_CONSTRUCTED_FALCON_INFO;
+
+#define NV2080_CTRL_CMD_GPU_GET_CONSTRUCTED_FALCON_INFO (0x208001b0) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_GPU_INTERFACE_ID << 8) | NV2080_CTRL_GPU_GET_CONSTRUCTED_FALCON_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_GPU_GET_CONSTRUCTED_FALCON_INFO_PARAMS {
+    NvU32                                   numConstructedFalcons;
+    NV2080_CTRL_GPU_CONSTRUCTED_FALCON_INFO constructedFalconsTable[NV2080_CTRL_GPU_MAX_CONSTRUCTED_FALCONS];
+} NV2080_CTRL_GPU_GET_CONSTRUCTED_FALCON_INFO_PARAMS;
+
+typedef struct NV2080_CTRL_CMD_INTERNAL_FIFO_TOGGLE_ACTIVE_CHANNEL_SCHEDULING_PARAMS {
+    NvBool bDisableActiveChannels;
+} NV2080_CTRL_CMD_INTERNAL_FIFO_TOGGLE_ACTIVE_CHANNEL_SCHEDULING_PARAMS;
+
+#define NV2080_CTRL_CMD_INTERNAL_FIFO_TOGGLE_ACTIVE_CHANNEL_SCHEDULING (0x20800ac3) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_CMD_INTERNAL_FIFO_TOGGLE_ACTIVE_CHANNEL_SCHEDULING_PARAMS_MESSAGE_ID" */
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/gr.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/gr.h
new file mode 100644
index 000000000000..feed1dabd9d2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/gr.h
@@ -0,0 +1,79 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_GR_H__
+#define __NVRM_GR_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define NV2080_CTRL_CMD_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO (0x20800a32) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_STATIC_KGR_GET_CONTEXT_BUFFERS_INFO_PARAMS_MESSAGE_ID" */
+
+#define NV2080_CTRL_INTERNAL_GR_MAX_ENGINES         8
+
+#define NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT 0x1a
+
+typedef struct NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO {
+    NvU32 size;
+    NvU32 alignment;
+} NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO;
+
+typedef struct NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO {
+    NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_BUFFER_INFO engine[NV2080_CTRL_INTERNAL_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT];
+} NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO;
+
+typedef struct NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS {
+    NV2080_CTRL_INTERNAL_STATIC_GR_CONTEXT_BUFFERS_INFO engineContextBuffersInfo[NV2080_CTRL_INTERNAL_GR_MAX_ENGINES];
+} NV2080_CTRL_INTERNAL_STATIC_GR_GET_CONTEXT_BUFFERS_INFO_PARAMS;
+
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID                          4:0
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS                 (0x00000000)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_VLD                      (0x00000001)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_VIDEO                    (0x00000002)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_MPEG                     (0x00000003)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_CAPTURE                  (0x00000004)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_DISPLAY                  (0x00000005)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_ENCRYPTION               (0x00000006)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_POSTPROCESS              (0x00000007)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_ZCULL           (0x00000008)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PM              (0x00000009)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COMPUTE_PREEMPT          (0x0000000a)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PREEMPT         (0x0000000b)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_SPILL           (0x0000000c)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PAGEPOOL        (0x0000000d)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_BETACB          (0x0000000e)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_RTV             (0x0000000f)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PATCH           (0x00000010)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_BUNDLE_CB       (0x00000011)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PAGEPOOL_GLOBAL (0x00000012)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_ATTRIBUTE_CB    (0x00000013)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_RTV_CB_GLOBAL   (0x00000014)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_GFXP_POOL       (0x00000015)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_GFXP_CTRL_BLK   (0x00000016)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_FECS_EVENT      (0x00000017)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_PRIV_ACCESS_MAP (0x00000018)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_GRAPHICS_SETUP           (0x00000019)
+#define NV0080_CTRL_FIFO_GET_ENGINE_CONTEXT_PROPERTIES_ENGINE_ID_COUNT                    (0x0000001a)
+
+#define NV2080_CTRL_CMD_GPU_GET_FERMI_GPC_INFO (0x20800137U) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_GPU_INTERFACE_ID << 8) | NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS {
+    NvU32 gpcMask;
+} NV2080_CTRL_GPU_GET_FERMI_GPC_INFO_PARAMS;
+
+#define NV2080_CTRL_CMD_GPU_GET_FERMI_TPC_INFO (0x20800138U) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_GPU_INTERFACE_ID << 8) | NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS_MESSAGE_ID" */
+typedef struct NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS {
+    NvU32 gpcId;
+    NvU32 tpcMask;
+} NV2080_CTRL_GPU_GET_FERMI_TPC_INFO_PARAMS;
+
+#define KGRAPHICS_SCRUBBER_HANDLE_VAS        0xdada0042
+#define KGRAPHICS_SCRUBBER_HANDLE_CHANNEL    (KGRAPHICS_SCRUBBER_HANDLE_VAS + 3)
+#define KGRAPHICS_SCRUBBER_HANDLE_3DOBJ      (KGRAPHICS_SCRUBBER_HANDLE_VAS + 4)
+
+typedef struct NV2080_CTRL_INTERNAL_GR_INIT_BUG4208224_WAR_PARAMS {
+    NvBool bTeardown;
+} NV2080_CTRL_INTERNAL_GR_INIT_BUG4208224_WAR_PARAMS;
+
+#define NV2080_CTRL_CMD_INTERNAL_KGR_INIT_BUG4208224_WAR (0x20800a46) /* finn: Evaluated from "(FINN_NV20_SUBDEVICE_0_INTERNAL_INTERFACE_ID << 8) | NV2080_CTRL_INTERNAL_KGR_INIT_BUG4208224_WAR_PARAMS_MESSAGE_ID" */
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/gsp.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/gsp.h
new file mode 100644
index 000000000000..b6075021e74f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/gsp.h
@@ -0,0 +1,634 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_GSP_H__
+#define __NVRM_GSP_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#define NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MAX_ENTRIES 16U
+
+#define NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MEM_TYPES           17U
+
+typedef NvBool NV2080_CTRL_CMD_FB_GET_FB_REGION_SURFACE_MEM_TYPE_FLAG[NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MEM_TYPES];
+
+typedef struct NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO {
+    NV_DECLARE_ALIGNED(NvU64 base, 8);
+    NV_DECLARE_ALIGNED(NvU64 limit, 8);
+    NV_DECLARE_ALIGNED(NvU64 reserved, 8);
+    NvU32                                                  performance;
+    NvBool                                                 supportCompressed;
+    NvBool                                                 supportISO;
+    NvBool                                                 bProtected;
+    NV2080_CTRL_CMD_FB_GET_FB_REGION_SURFACE_MEM_TYPE_FLAG blackList;
+} NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO;
+
+typedef struct NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS {
+    NvU32 numFBRegions;
+    NV_DECLARE_ALIGNED(NV2080_CTRL_CMD_FB_GET_FB_REGION_FB_REGION_INFO fbRegion[NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_MAX_ENTRIES], 8);
+} NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS;
+
+#define NV0080_CTRL_GR_CAPS_TBL_SIZE            23
+
+#define NV2080_GPU_MAX_GID_LENGTH             (0x000000100ULL)
+
+typedef struct NV2080_CTRL_GPU_GET_GID_INFO_PARAMS {
+    NvU32 index;
+    NvU32 flags;
+    NvU32 length;
+    NvU8  data[NV2080_GPU_MAX_GID_LENGTH];
+} NV2080_CTRL_GPU_GET_GID_INFO_PARAMS;
+
+typedef struct NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS {
+    NvU32 BoardID;
+    char  chipSKU[9];
+    char  chipSKUMod[5];
+    NvU32 skuConfigVersion;
+    char  project[5];
+    char  projectSKU[5];
+    char  CDP[6];
+    char  projectSKUMod[2];
+    NvU32 businessCycle;
+} NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS;
+
+#define MAX_GPC_COUNT           32
+
+typedef struct NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS {
+    NvU32  totalVFs;
+    NvU32  firstVfOffset;
+    NvU32  vfFeatureMask;
+    NV_DECLARE_ALIGNED(NvU64 FirstVFBar0Address, 8);
+    NV_DECLARE_ALIGNED(NvU64 FirstVFBar1Address, 8);
+    NV_DECLARE_ALIGNED(NvU64 FirstVFBar2Address, 8);
+    NV_DECLARE_ALIGNED(NvU64 bar0Size, 8);
+    NV_DECLARE_ALIGNED(NvU64 bar1Size, 8);
+    NV_DECLARE_ALIGNED(NvU64 bar2Size, 8);
+    NvBool b64bitBar0;
+    NvBool b64bitBar1;
+    NvBool b64bitBar2;
+    NvBool bSriovEnabled;
+    NvBool bSriovHeavyEnabled;
+    NvBool bEmulateVFBar0TlbInvalidationRegister;
+    NvBool bClientRmAllocatedCtxBuffer;
+    NvBool bNonPowerOf2ChannelCountSupported;
+    NvBool bVfResizableBAR1Supported;
+} NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS;
+
+#include "engine.h"
+
+#define NVGPU_ENGINE_CAPS_MASK_BITS                32
+
+#define NVGPU_ENGINE_CAPS_MASK_ARRAY_MAX           ((RM_ENGINE_TYPE_LAST-1)/NVGPU_ENGINE_CAPS_MASK_BITS + 1)
+
+#define NV2080_GPU_MAX_NAME_STRING_LENGTH                  (0x0000040U)
+
+typedef struct VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS 
+{
+    NvU32 numHeads;
+    NvU32 maxNumHeads;
+} VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS;
+
+typedef struct VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS 
+{
+    NvU32 headIndex;
+    NvU32 maxHResolution;
+    NvU32 maxVResolution;
+} VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS;
+
+#define MAX_GROUP_COUNT 2
+
+typedef struct
+{
+    NvU32 ecidLow;
+    NvU32 ecidHigh;
+    NvU32 ecidExtended;
+} EcidManufacturingInfo;
+
+typedef struct
+{
+    NvU64 nonWprHeapOffset;
+    NvU64 frtsOffset;
+} FW_WPR_LAYOUT_OFFSET;
+
+typedef struct GspStaticConfigInfo_t
+{
+    NvU8 grCapsBits[NV0080_CTRL_GR_CAPS_TBL_SIZE];
+    NV2080_CTRL_GPU_GET_GID_INFO_PARAMS gidInfo;
+    NV2080_CTRL_BIOS_GET_SKU_INFO_PARAMS SKUInfo;
+    NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS fbRegionInfoParams;
+
+    NV0080_CTRL_GPU_GET_SRIOV_CAPS_PARAMS sriovCaps;
+    NvU32 sriovMaxGfid;
+
+    NvU32 engineCaps[NVGPU_ENGINE_CAPS_MASK_ARRAY_MAX];
+
+    NvBool poisonFuseEnabled;
+
+    NvU64 fb_length;
+    NvU64 fbio_mask;
+    NvU32 fb_bus_width;
+    NvU32 fb_ram_type;
+    NvU64 fbp_mask;
+    NvU32 l2_cache_size;
+
+    NvU8 gpuNameString[NV2080_GPU_MAX_NAME_STRING_LENGTH];
+    NvU8 gpuShortNameString[NV2080_GPU_MAX_NAME_STRING_LENGTH];
+    NvU16 gpuNameString_Unicode[NV2080_GPU_MAX_NAME_STRING_LENGTH];
+    NvBool bGpuInternalSku;
+    NvBool bIsQuadroGeneric;
+    NvBool bIsQuadroAd;
+    NvBool bIsNvidiaNvs;
+    NvBool bIsVgx;
+    NvBool bGeforceSmb;
+    NvBool bIsTitan;
+    NvBool bIsTesla;
+    NvBool bIsMobile;
+    NvBool bIsGc6Rtd3Allowed;
+    NvBool bIsGc8Rtd3Allowed;
+    NvBool bIsGcOffRtd3Allowed;
+    NvBool bIsGcoffLegacyAllowed;
+    NvBool bIsMigSupported;
+
+    /* "Total Board Power" refers to power requirement of GPU,
+     * while in GC6 state. Majority of this power will be used
+     * to keep V-RAM active to preserve its content.
+     * Some energy maybe consumed by Always-on components on GPU chip.
+     * This power will be provided by 3.3v voltage rail.
+     */
+    NvU16  RTD3GC6TotalBoardPower;
+
+    /* PERST# (i.e. PCI Express Reset) is a sideband signal
+     * generated by the PCIe Host to indicate the PCIe devices,
+     * that the power-rails and the reference-clock are stable.
+     * The endpoint device typically uses this signal as a global reset.
+     */
+    NvU16  RTD3GC6PerstDelay;
+
+    NvU64 bar1PdeBase;
+    NvU64 bar2PdeBase;
+
+    NvBool bVbiosValid;
+    NvU32 vbiosSubVendor;
+    NvU32 vbiosSubDevice;
+
+    NvBool bPageRetirementSupported;
+
+    NvBool bSplitVasBetweenServerClientRm;
+
+    NvBool bClRootportNeedsNosnoopWAR;
+
+    VIRTUAL_DISPLAY_GET_NUM_HEADS_PARAMS displaylessMaxHeads;
+    VIRTUAL_DISPLAY_GET_MAX_RESOLUTION_PARAMS displaylessMaxResolution;
+    NvU64 displaylessMaxPixels;
+
+    // Client handle for internal RMAPI control.
+    NvHandle hInternalClient;
+
+    // Device handle for internal RMAPI control.
+    NvHandle hInternalDevice;
+
+    // Subdevice handle for internal RMAPI control.
+    NvHandle hInternalSubdevice;
+
+    NvBool bSelfHostedMode;
+    NvBool bAtsSupported;
+
+    NvBool bIsGpuUefi;
+    NvBool bIsEfiInit;
+
+    EcidManufacturingInfo ecidInfo[MAX_GROUP_COUNT];
+
+    FW_WPR_LAYOUT_OFFSET fwWprLayoutOffset;
+} GspStaticConfigInfo;
+
+typedef struct
+{
+    NvU16               deviceID;           // deviceID
+    NvU16               vendorID;           // vendorID
+    NvU16               subdeviceID;        // subsystem deviceID
+    NvU16               subvendorID;        // subsystem vendorID
+    NvU8                revisionID;         // revision ID
+} BUSINFO;
+
+#define NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS             (16U)
+
+typedef struct DOD_METHOD_DATA
+{
+    NV_STATUS status;
+    NvU32     acpiIdListLen;
+    NvU32     acpiIdList[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
+} DOD_METHOD_DATA;
+
+typedef struct JT_METHOD_DATA
+{
+    NV_STATUS status;
+    NvU32     jtCaps;
+    NvU16     jtRevId;
+    NvBool    bSBIOSCaps;
+} JT_METHOD_DATA;
+
+typedef struct MUX_METHOD_DATA_ELEMENT
+{
+    NvU32       acpiId;
+    NvU32       mode;
+    NV_STATUS   status;
+} MUX_METHOD_DATA_ELEMENT;
+
+#define NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS             (16U)
+
+typedef struct MUX_METHOD_DATA
+{
+    NvU32                       tableLen;
+    MUX_METHOD_DATA_ELEMENT     acpiIdMuxModeTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
+    MUX_METHOD_DATA_ELEMENT     acpiIdMuxPartTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];
+    MUX_METHOD_DATA_ELEMENT     acpiIdMuxStateTable[NV0073_CTRL_SYSTEM_ACPI_ID_MAP_MAX_DISPLAYS];    
+} MUX_METHOD_DATA;
+
+typedef struct CAPS_METHOD_DATA
+{
+    NV_STATUS status;
+    NvU32     optimusCaps;
+} CAPS_METHOD_DATA;
+
+typedef struct ACPI_METHOD_DATA
+{
+    NvBool                                               bValid;
+    DOD_METHOD_DATA                                      dodMethodData;
+    JT_METHOD_DATA                                       jtMethodData;
+    MUX_METHOD_DATA                                      muxMethodData;
+    CAPS_METHOD_DATA                                     capsMethodData;
+} ACPI_METHOD_DATA;
+
+typedef struct GSP_VF_INFO
+{
+    NvU32  totalVFs;
+    NvU32  firstVFOffset;
+    NvU64  FirstVFBar0Address;
+    NvU64  FirstVFBar1Address;
+    NvU64  FirstVFBar2Address;
+    NvBool b64bitBar0;
+    NvBool b64bitBar1;
+    NvBool b64bitBar2;
+} GSP_VF_INFO;
+
+typedef struct
+{
+    // Link capabilities
+    NvU32 linkCap;
+} GSP_PCIE_CONFIG_REG;
+
+typedef struct GspSystemInfo
+{
+    NvU64 gpuPhysAddr;
+    NvU64 gpuPhysFbAddr;
+    NvU64 gpuPhysInstAddr;
+    NvU64 gpuPhysIoAddr;
+    NvU64 nvDomainBusDeviceFunc;
+    NvU64 simAccessBufPhysAddr;
+    NvU64 notifyOpSharedSurfacePhysAddr;
+    NvU64 pcieAtomicsOpMask;
+    NvU64 consoleMemSize;
+    NvU64 maxUserVa;
+    NvU32 pciConfigMirrorBase;
+    NvU32 pciConfigMirrorSize;
+    NvU32 PCIDeviceID;
+    NvU32 PCISubDeviceID;
+    NvU32 PCIRevisionID;
+    NvU32 pcieAtomicsCplDeviceCapMask;
+    NvU8 oorArch;
+    NvU64 clPdbProperties;
+    NvU32 Chipset;
+    NvBool bGpuBehindBridge;
+    NvBool bFlrSupported;
+    NvBool b64bBar0Supported;
+    NvBool bMnocAvailable;
+    NvU32  chipsetL1ssEnable;
+    NvBool bUpstreamL0sUnsupported;
+    NvBool bUpstreamL1Unsupported;
+    NvBool bUpstreamL1PorSupported;
+    NvBool bUpstreamL1PorMobileOnly;
+    NvBool bSystemHasMux;
+    NvU8   upstreamAddressValid;
+    BUSINFO FHBBusInfo;
+    BUSINFO chipsetIDInfo;
+    ACPI_METHOD_DATA acpiMethodData;
+    NvU32 hypervisorType;
+    NvBool bIsPassthru;
+    NvU64 sysTimerOffsetNs;
+    GSP_VF_INFO gspVFInfo;
+    NvBool bIsPrimary;
+    NvBool isGridBuild;
+    GSP_PCIE_CONFIG_REG pcieConfigReg;
+    NvU32 gridBuildCsp;
+    NvBool bPreserveVideoMemoryAllocations;
+    NvBool bTdrEventSupported;
+    NvBool bFeatureStretchVblankCapable;
+    NvBool bEnableDynamicGranularityPageArrays;
+    NvBool bClockBoostSupported;
+    NvBool bRouteDispIntrsToCPU;
+    NvU64  hostPageSize;
+} GspSystemInfo;
+
+typedef struct rpc_os_error_log_v17_00
+{
+    NvU32      exceptType;
+    NvU32      runlistId;
+    NvU32      chid;
+    char       errString[0x100];
+    NvU32      preemptiveRemovalPreviousXid;
+} rpc_os_error_log_v17_00;
+
+typedef struct
+{
+    // Magic
+    // BL to use for verification (i.e. Booter locked it in WPR2)
+    NvU64 magic; // = 0xdc3aae21371a60b3;
+
+    // Revision number of Booter-BL-Sequencer handoff interface
+    // Bumped up when we change this interface so it is not backward compatible.
+    // Bumped up when we revoke GSP-RM ucode
+    NvU64 revision; // = 1;
+
+    // ---- Members regarding data in SYSMEM ----------------------------
+    // Consumed by Booter for DMA
+
+    NvU64 sysmemAddrOfRadix3Elf;
+    NvU64 sizeOfRadix3Elf;
+
+    NvU64 sysmemAddrOfBootloader;
+    NvU64 sizeOfBootloader;
+
+    // Offsets inside bootloader image needed by Booter
+    NvU64 bootloaderCodeOffset;
+    NvU64 bootloaderDataOffset;
+    NvU64 bootloaderManifestOffset;
+
+    union
+    {
+        // Used only at initial boot
+        struct
+        {
+            NvU64 sysmemAddrOfSignature;
+            NvU64 sizeOfSignature;
+        };
+
+        //
+        // Used at suspend/resume to read GspFwHeapFreeList
+        // Offset relative to GspFwWprMeta FBMEM PA (gspFwWprStart)
+        //
+        struct
+        {
+            NvU32 gspFwHeapFreeListWprOffset;
+            NvU32 unused0;
+            NvU64 unused1;
+        };
+    };
+
+    // ---- Members describing FB layout --------------------------------
+    NvU64 gspFwRsvdStart;
+
+    NvU64 nonWprHeapOffset;
+    NvU64 nonWprHeapSize;
+
+    NvU64 gspFwWprStart;
+
+    // GSP-RM to use to setup heap.
+    NvU64 gspFwHeapOffset;
+    NvU64 gspFwHeapSize;
+
+    // BL to use to find ELF for jump
+    NvU64 gspFwOffset;
+    // Size is sizeOfRadix3Elf above.
+
+    NvU64 bootBinOffset;
+    // Size is sizeOfBootloader above.
+
+    NvU64 frtsOffset;
+    NvU64 frtsSize;
+
+    NvU64 gspFwWprEnd;
+
+    // GSP-RM to use for fbRegionInfo?
+    NvU64 fbSize;
+
+    // ---- Other members -----------------------------------------------
+
+    // GSP-RM to use for fbRegionInfo?
+    NvU64 vgaWorkspaceOffset;
+    NvU64 vgaWorkspaceSize;
+
+    // Boot count.  Used to determine whether to load the firmware image.
+    NvU64 bootCount;
+
+    // This union is organized the way it is to start at an 8-byte boundary and achieve natural
+    // packing of the internal struct fields.
+    union
+    {
+        struct
+        {
+            // TODO: the partitionRpc* fields below do not really belong in this
+            //       structure. The values are patched in by the partition bootstrapper
+            //       when GSP-RM is booted in a partition, and this structure was a
+            //       convenient place for the bootstrapper to access them. These should
+            //       be moved to a different comm. mechanism between the bootstrapper
+            //       and the GSP-RM tasks.
+
+            // Shared partition RPC memory (physical address)
+            NvU64 partitionRpcAddr;
+
+            // Offsets relative to partitionRpcAddr
+            NvU16 partitionRpcRequestOffset;
+            NvU16 partitionRpcReplyOffset;
+
+            // Code section and dataSection offset and size.
+            NvU32 elfCodeOffset;
+            NvU32 elfDataOffset;
+            NvU32 elfCodeSize;
+            NvU32 elfDataSize;
+
+            // Used during GSP-RM resume to check for revocation
+            NvU32 lsUcodeVersion;
+        };
+
+        struct
+        {
+            // Pad for the partitionRpc* fields, plus 4 bytes
+            NvU32 partitionRpcPadding[4];
+
+            // CrashCat (contiguous) buffer size/location - occupies same bytes as the
+            // elf(Code|Data)(Offset|Size) fields above.
+            // TODO: move to GSP_FMC_INIT_PARAMS
+            NvU64 sysmemAddrOfCrashReportQueue;
+            NvU32 sizeOfCrashReportQueue;
+
+            // Pad for the lsUcodeVersion field
+            NvU32 lsUcodeVersionPadding[1];
+        };
+    };
+
+    // Number of VF partitions allocating sub-heaps from the WPR heap
+    // Used during boot to ensure the heap is adequately sized
+    NvU8 gspFwHeapVfPartitionCount;
+
+    // Flags to help decide GSP-FW flow.
+    NvU8 flags;
+
+    // Pad structure to exactly 256 bytes.  Can replace padding with additional
+    // fields without incrementing revision.  Padding initialized to 0.
+    NvU8 padding[2];
+
+    //
+    // Starts at gspFwWprEnd+frtsSize b/c FRTS is positioned
+    // to end where this allocation starts (when RM requests FSP to create
+    // FRTS).
+    //
+    NvU32 pmuReservedSize;
+
+    // BL to use for verification (i.e. Booter says OK to boot)
+    NvU64 verified;  // 0x0 -> unverified, 0xa0a0a0a0a0a0a0a0 -> verified
+} GspFwWprMeta;
+
+#define GSP_FW_WPR_META_MAGIC     0xdc3aae21371a60b3ULL
+
+#define GSP_FW_WPR_META_REVISION  1
+
+typedef struct {
+    NvU64 sharedMemPhysAddr;
+    NvU32 pageTableEntryCount;
+    NvLength cmdQueueOffset;
+    NvLength statQueueOffset;
+} MESSAGE_QUEUE_INIT_ARGUMENTS;
+
+typedef struct {
+    NvU32 oldLevel;
+    NvU32 flags;
+    NvBool bInPMTransition;
+} GSP_SR_INIT_ARGUMENTS;
+
+typedef struct
+{
+    MESSAGE_QUEUE_INIT_ARGUMENTS      messageQueueInitArguments;
+    GSP_SR_INIT_ARGUMENTS             srInitArguments;
+    NvU32                             gpuInstance;
+    NvBool                            bDmemStack;
+
+    struct
+    {
+        NvU64                         pa;
+        NvU64                         size;
+    } profilerArgs;
+} GSP_ARGUMENTS_CACHED;
+
+#define NV2080_CTRL_GPU_SET_POWER_STATE_GPU_LEVEL_3            (0x00000003U)
+
+typedef struct
+{
+    // Magic for verification by secure ucode
+    NvU64 magic;  // = GSP_FW_SR_META_MAGIC;
+
+    //
+    // Revision number
+    // Bumped up when we change this interface so it is not backward compatible.
+    //
+    NvU64 revision;  // = GSP_FW_SR_META_MAGIC_REVISION;
+
+    // Members regarding data in SYSMEM
+    NvU64 sysmemAddrOfSuspendResumeData;
+    NvU64 sizeOfSuspendResumeData;
+
+    //
+    // Internal members for use by secure ucode
+    // Must be exactly GSP_FW_SR_META_INTERNAL_SIZE bytes.
+    //
+    NvU32 internal[32];
+
+    // Same as flags of GspFwWprMeta
+    NvU32 flags;
+
+    // Subrevision number used by secure ucode
+    NvU32 subrevision;
+
+    //
+    // Pad structure to exactly 256 bytes (1 DMA chunk).
+    // Padding initialized to zero.
+    //
+    NvU32 padding[22];
+} GspFwSRMeta;
+
+#define GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS2                   (0 << 20)   // No FB heap usage
+
+#define GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3_BAREMETAL        (22 << 20)
+
+#define GSP_FW_HEAP_PARAM_BASE_RM_SIZE_TU10X               (8 << 20)   // Turing thru Ada
+
+#define GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS2_MIN_MB                (64u)
+
+#define BULLSEYE_ROOT_HEAP_ALLOC_RM_DATA_SECTION_SIZE_DELTA         (12u)
+
+#define BULLSEYE_ROOT_HEAP_ALLOC_BAREMETAL_LIBOS_HEAP_SIZE_DELTA    (70u)
+
+#define GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB \
+                                    (88u + (BULLSEYE_ROOT_HEAP_ALLOC_RM_DATA_SECTION_SIZE_DELTA) + \
+                                    (BULLSEYE_ROOT_HEAP_ALLOC_BAREMETAL_LIBOS_HEAP_SIZE_DELTA))
+
+typedef struct GSP_FMC_INIT_PARAMS
+{
+    // CC initialization "registry keys"
+    NvU32 regkeys;
+} GSP_FMC_INIT_PARAMS;
+
+typedef enum {
+    GSP_DMA_TARGET_LOCAL_FB,
+    GSP_DMA_TARGET_COHERENT_SYSTEM,
+    GSP_DMA_TARGET_NONCOHERENT_SYSTEM,
+    GSP_DMA_TARGET_COUNT
+} GSP_DMA_TARGET;
+
+typedef struct GSP_ACR_BOOT_GSP_RM_PARAMS
+{
+    // Physical memory aperture through which gspRmDescPa is accessed
+    GSP_DMA_TARGET target;
+    // Size in bytes of the GSP-RM descriptor structure
+    NvU32          gspRmDescSize;
+    // Physical offset in the target aperture of the GSP-RM descriptor structure
+    NvU64          gspRmDescOffset;
+    // Physical offset in FB to set the start of the WPR containing GSP-RM
+    NvU64          wprCarveoutOffset;
+    // Size in bytes of the WPR containing GSP-RM
+    NvU32          wprCarveoutSize;
+    // Whether to boot GSP-RM or GSP-Proxy through ACR
+    NvBool         bIsGspRmBoot;
+} GSP_ACR_BOOT_GSP_RM_PARAMS;
+
+typedef struct GSP_RM_PARAMS
+{
+    // Physical memory aperture through which bootArgsOffset is accessed
+    GSP_DMA_TARGET target;
+    // Physical offset in the memory aperture that will be passed to GSP-RM
+    NvU64          bootArgsOffset;
+} GSP_RM_PARAMS;
+
+typedef struct GSP_SPDM_PARAMS
+{
+    // Physical Memory Aperture through which all addresses are accessed
+    GSP_DMA_TARGET target;
+
+    // Physical offset in the memory aperture where SPDM payload is stored
+    NvU64 payloadBufferOffset;
+
+    // Size of the above payload buffer
+    NvU32 payloadBufferSize;
+} GSP_SPDM_PARAMS;
+
+typedef struct GSP_FMC_BOOT_PARAMS
+{
+    GSP_FMC_INIT_PARAMS         initParams;
+    GSP_ACR_BOOT_GSP_RM_PARAMS  bootGspRmParams;
+    GSP_RM_PARAMS               gspRmParams;
+    GSP_SPDM_PARAMS             gspSpdmParams;
+} GSP_FMC_BOOT_PARAMS;
+
+#define GSP_FW_HEAP_PARAM_BASE_RM_SIZE_GH100              (14 << 20)   // Hopper+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/msgfn.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/msgfn.h
new file mode 100644
index 000000000000..e06643f57695
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/msgfn.h
@@ -0,0 +1,57 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_MSGFN_H__
+#define __NVRM_MSGFN_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#ifndef E
+#    define E(RPC, VAL) NV_VGPU_MSG_EVENT_##RPC = VAL,
+#    define DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
+enum {
+#endif
+    E(FIRST_EVENT,                                            0x1000)
+    E(GSP_INIT_DONE,                                          0x1001)
+    E(GSP_RUN_CPU_SEQUENCER,                                  0x1002)
+    E(POST_EVENT,                                             0x1003)
+    E(RC_TRIGGERED,                                           0x1004)
+    E(MMU_FAULT_QUEUED,                                       0x1005)
+    E(OS_ERROR_LOG,                                           0x1006)
+    E(RG_LINE_INTR,                                           0x1007)
+    E(GPUACCT_PERFMON_UTIL_SAMPLES,                           0x1008)
+    E(SIM_READ,                                               0x1009)
+    E(SIM_WRITE,                                              0x100a)
+    E(SEMAPHORE_SCHEDULE_CALLBACK,                            0x100b)
+    E(UCODE_LIBOS_PRINT,                                      0x100c)
+    E(VGPU_GSP_PLUGIN_TRIGGERED,                              0x100d)
+    E(PERF_GPU_BOOST_SYNC_LIMITS_CALLBACK,                    0x100e)
+    E(PERF_BRIDGELESS_INFO_UPDATE,                            0x100f)
+    E(VGPU_CONFIG,                                            0x1010)
+    E(DISPLAY_MODESET,                                        0x1011)
+    E(EXTDEV_INTR_SERVICE,                                    0x1012)
+    E(NVLINK_INBAND_RECEIVED_DATA_256,                        0x1013)
+    E(NVLINK_INBAND_RECEIVED_DATA_512,                        0x1014)
+    E(NVLINK_INBAND_RECEIVED_DATA_1024,                       0x1015)
+    E(NVLINK_INBAND_RECEIVED_DATA_2048,                       0x1016)
+    E(NVLINK_INBAND_RECEIVED_DATA_4096,                       0x1017)
+    E(TIMED_SEMAPHORE_RELEASE,                                0x1018)
+    E(NVLINK_IS_GPU_DEGRADED,                                 0x1019)
+    E(PFM_REQ_HNDLR_STATE_SYNC_CALLBACK,                      0x101a)
+    E(NVLINK_FAULT_UP,                                        0x101b)
+    E(GSP_LOCKDOWN_NOTICE,                                    0x101c)
+    E(MIG_CI_CONFIG_UPDATE,                                   0x101d)
+    E(UPDATE_GSP_TRACE,                                       0x101e)
+    E(NVLINK_FATAL_ERROR_RECOVERY,                            0x101f)
+    E(GSP_POST_NOCAT_RECORD,                                  0x1020)
+    E(FECS_ERROR,                                             0x1021)
+    E(RECOVERY_ACTION,                                        0x1022)
+    E(NUM_EVENTS,                                             0x1023)
+#ifdef DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
+};
+#   undef E
+#   undef DEFINING_E_IN_RPC_GLOBAL_ENUMS_H
+#endif
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/ofa.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/ofa.h
new file mode 100644
index 000000000000..fcaef7f553a6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/ofa.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_OFA_H__
+#define __NVRM_OFA_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+typedef struct
+{
+    NvU32 size;
+    NvU32 prohibitMultipleInstances;  // Prohibit multiple allocations of OFA?
+    NvU32 engineInstance;
+} NV_OFA_ALLOCATION_PARAMETERS;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/rpcfn.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/rpcfn.h
new file mode 100644
index 000000000000..2d67b598c58b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/nvrm/rpcfn.h
@@ -0,0 +1,249 @@
+/* SPDX-License-Identifier: MIT */
+
+/* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __NVRM_RPCFN_H__
+#define __NVRM_RPCFN_H__
+#include <nvrm/nvtypes.h>
+
+/* Excerpt of RM headers from https://github.com/NVIDIA/open-gpu-kernel-modules/tree/570.144 */
+
+#ifndef X
+#    define X(UNIT, RPC, VAL) NV_VGPU_MSG_FUNCTION_##RPC = VAL,
+#    define DEFINING_X_IN_RPC_GLOBAL_ENUMS_H
+enum {
+#endif
+    X(RM, NOP,                                                    0)
+    X(RM, SET_GUEST_SYSTEM_INFO,                                  1)
+    X(RM, ALLOC_ROOT,                                             2)
+    X(RM, ALLOC_DEVICE,                                           3) // deprecated
+    X(RM, ALLOC_MEMORY,                                           4)
+    X(RM, ALLOC_CTX_DMA,                                          5)
+    X(RM, ALLOC_CHANNEL_DMA,                                      6)
+    X(RM, MAP_MEMORY,                                             7)
+    X(RM, BIND_CTX_DMA,                                           8) // deprecated
+    X(RM, ALLOC_OBJECT,                                           9)
+    X(RM, FREE,                                                   10)
+    X(RM, LOG,                                                    11)
+    X(RM, ALLOC_VIDMEM,                                           12)
+    X(RM, UNMAP_MEMORY,                                           13)
+    X(RM, MAP_MEMORY_DMA,                                         14)
+    X(RM, UNMAP_MEMORY_DMA,                                       15)
+    X(RM, GET_EDID,                                               16) // deprecated
+    X(RM, ALLOC_DISP_CHANNEL,                                     17)
+    X(RM, ALLOC_DISP_OBJECT,                                      18)
+    X(RM, ALLOC_SUBDEVICE,                                        19)
+    X(RM, ALLOC_DYNAMIC_MEMORY,                                   20)
+    X(RM, DUP_OBJECT,                                             21)
+    X(RM, IDLE_CHANNELS,                                          22)
+    X(RM, ALLOC_EVENT,                                            23)
+    X(RM, SEND_EVENT,                                             24) // deprecated
+    X(RM, REMAPPER_CONTROL,                                       25) // deprecated
+    X(RM, DMA_CONTROL,                                            26) // deprecated
+    X(RM, DMA_FILL_PTE_MEM,                                       27)
+    X(RM, MANAGE_HW_RESOURCE,                                     28)
+    X(RM, BIND_ARBITRARY_CTX_DMA,                                 29) // deprecated
+    X(RM, CREATE_FB_SEGMENT,                                      30)
+    X(RM, DESTROY_FB_SEGMENT,                                     31)
+    X(RM, ALLOC_SHARE_DEVICE,                                     32)
+    X(RM, DEFERRED_API_CONTROL,                                   33)
+    X(RM, REMOVE_DEFERRED_API,                                    34)
+    X(RM, SIM_ESCAPE_READ,                                        35)
+    X(RM, SIM_ESCAPE_WRITE,                                       36)
+    X(RM, SIM_MANAGE_DISPLAY_CONTEXT_DMA,                         37)
+    X(RM, FREE_VIDMEM_VIRT,                                       38)
+    X(RM, PERF_GET_PSTATE_INFO,                                   39) // deprecated
+    X(RM, PERF_GET_PERFMON_SAMPLE,                                40)
+    X(RM, PERF_GET_VIRTUAL_PSTATE_INFO,                           41) // deprecated
+    X(RM, PERF_GET_LEVEL_INFO,                                    42)
+    X(RM, MAP_SEMA_MEMORY,                                        43)
+    X(RM, UNMAP_SEMA_MEMORY,                                      44)
+    X(RM, SET_SURFACE_PROPERTIES,                                 45)
+    X(RM, CLEANUP_SURFACE,                                        46)
+    X(RM, UNLOADING_GUEST_DRIVER,                                 47)
+    X(RM, TDR_SET_TIMEOUT_STATE,                                  48)
+    X(RM, SWITCH_TO_VGA,                                          49)
+    X(RM, GPU_EXEC_REG_OPS,                                       50)
+    X(RM, GET_STATIC_INFO,                                        51)
+    X(RM, ALLOC_VIRTMEM,                                          52)
+    X(RM, UPDATE_PDE_2,                                           53)
+    X(RM, SET_PAGE_DIRECTORY,                                     54)
+    X(RM, GET_STATIC_PSTATE_INFO,                                 55)
+    X(RM, TRANSLATE_GUEST_GPU_PTES,                               56)
+    X(RM, RESERVED_57,                                            57)
+    X(RM, RESET_CURRENT_GR_CONTEXT,                               58)
+    X(RM, SET_SEMA_MEM_VALIDATION_STATE,                          59)
+    X(RM, GET_ENGINE_UTILIZATION,                                 60)
+    X(RM, UPDATE_GPU_PDES,                                        61)
+    X(RM, GET_ENCODER_CAPACITY,                                   62)
+    X(RM, VGPU_PF_REG_READ32,                                     63) // deprecated
+    X(RM, SET_GUEST_SYSTEM_INFO_EXT,                              64)
+    X(GSP, GET_GSP_STATIC_INFO,                                   65)
+    X(RM, RMFS_INIT,                                              66) // deprecated
+    X(RM, RMFS_CLOSE_QUEUE,                                       67) // deprecated
+    X(RM, RMFS_CLEANUP,                                           68) // deprecated
+    X(RM, RMFS_TEST,                                              69) // deprecated
+    X(RM, UPDATE_BAR_PDE,                                         70)
+    X(RM, CONTINUATION_RECORD,                                    71)
+    X(RM, GSP_SET_SYSTEM_INFO,                                    72)
+    X(RM, SET_REGISTRY,                                           73)
+    X(GSP, GSP_INIT_POST_OBJGPU,                                  74) // deprecated
+    X(RM, SUBDEV_EVENT_SET_NOTIFICATION,                          75) // deprecated
+    X(GSP, GSP_RM_CONTROL,                                        76)
+    X(RM, GET_STATIC_INFO2,                                       77)
+    X(RM, DUMP_PROTOBUF_COMPONENT,                                78)
+    X(RM, UNSET_PAGE_DIRECTORY,                                   79)
+    X(RM, GET_CONSOLIDATED_STATIC_INFO,                           80) // deprecated
+    X(RM, GMMU_REGISTER_FAULT_BUFFER,                             81) // deprecated
+    X(RM, GMMU_UNREGISTER_FAULT_BUFFER,                           82) // deprecated
+    X(RM, GMMU_REGISTER_CLIENT_SHADOW_FAULT_BUFFER,               83) // deprecated
+    X(RM, GMMU_UNREGISTER_CLIENT_SHADOW_FAULT_BUFFER,             84) // deprecated
+    X(RM, CTRL_SET_VGPU_FB_USAGE,                                 85)
+    X(RM, CTRL_NVFBC_SW_SESSION_UPDATE_INFO,                      86)
+    X(RM, CTRL_NVENC_SW_SESSION_UPDATE_INFO,                      87)
+    X(RM, CTRL_RESET_CHANNEL,                                     88)
+    X(RM, CTRL_RESET_ISOLATED_CHANNEL,                            89)
+    X(RM, CTRL_GPU_HANDLE_VF_PRI_FAULT,                           90)
+    X(RM, CTRL_CLK_GET_EXTENDED_INFO,                             91)
+    X(RM, CTRL_PERF_BOOST,                                        92)
+    X(RM, CTRL_PERF_VPSTATES_GET_CONTROL,                         93)
+    X(RM, CTRL_GET_ZBC_CLEAR_TABLE,                               94)
+    X(RM, CTRL_SET_ZBC_COLOR_CLEAR,                               95)
+    X(RM, CTRL_SET_ZBC_DEPTH_CLEAR,                               96)
+    X(RM, CTRL_GPFIFO_SCHEDULE,                                   97)
+    X(RM, CTRL_SET_TIMESLICE,                                     98)
+    X(RM, CTRL_PREEMPT,                                           99)
+    X(RM, CTRL_FIFO_DISABLE_CHANNELS,                             100)
+    X(RM, CTRL_SET_TSG_INTERLEAVE_LEVEL,                          101)
+    X(RM, CTRL_SET_CHANNEL_INTERLEAVE_LEVEL,                      102)
+    X(GSP, GSP_RM_ALLOC,                                          103)
+    X(RM, CTRL_GET_P2P_CAPS_V2,                                   104)
+    X(RM, CTRL_CIPHER_AES_ENCRYPT,                                105)
+    X(RM, CTRL_CIPHER_SESSION_KEY,                                106)
+    X(RM, CTRL_CIPHER_SESSION_KEY_STATUS,                         107)
+    X(RM, CTRL_DBG_CLEAR_ALL_SM_ERROR_STATES,                     108)
+    X(RM, CTRL_DBG_READ_ALL_SM_ERROR_STATES,                      109)
+    X(RM, CTRL_DBG_SET_EXCEPTION_MASK,                            110)
+    X(RM, CTRL_GPU_PROMOTE_CTX,                                   111)
+    X(RM, CTRL_GR_CTXSW_PREEMPTION_BIND,                          112)
+    X(RM, CTRL_GR_SET_CTXSW_PREEMPTION_MODE,                      113)
+    X(RM, CTRL_GR_CTXSW_ZCULL_BIND,                               114)
+    X(RM, CTRL_GPU_INITIALIZE_CTX,                                115)
+    X(RM, CTRL_VASPACE_COPY_SERVER_RESERVED_PDES,                 116)
+    X(RM, CTRL_FIFO_CLEAR_FAULTED_BIT,                            117)
+    X(RM, CTRL_GET_LATEST_ECC_ADDRESSES,                          118)
+    X(RM, CTRL_MC_SERVICE_INTERRUPTS,                             119)
+    X(RM, CTRL_DMA_SET_DEFAULT_VASPACE,                           120)
+    X(RM, CTRL_GET_CE_PCE_MASK,                                   121)
+    X(RM, CTRL_GET_ZBC_CLEAR_TABLE_ENTRY,                         122)
+    X(RM, CTRL_GET_NVLINK_PEER_ID_MASK,                           123) // deprecated
+    X(RM, CTRL_GET_NVLINK_STATUS,                                 124)
+    X(RM, CTRL_GET_P2P_CAPS,                                      125)
+    X(RM, CTRL_GET_P2P_CAPS_MATRIX,                               126)
+    X(RM, RESERVED_0,                                             127)
+    X(RM, CTRL_RESERVE_PM_AREA_SMPC,                              128)
+    X(RM, CTRL_RESERVE_HWPM_LEGACY,                               129)
+    X(RM, CTRL_B0CC_EXEC_REG_OPS,                                 130)
+    X(RM, CTRL_BIND_PM_RESOURCES,                                 131)
+    X(RM, CTRL_DBG_SUSPEND_CONTEXT,                               132)
+    X(RM, CTRL_DBG_RESUME_CONTEXT,                                133)
+    X(RM, CTRL_DBG_EXEC_REG_OPS,                                  134)
+    X(RM, CTRL_DBG_SET_MODE_MMU_DEBUG,                            135)
+    X(RM, CTRL_DBG_READ_SINGLE_SM_ERROR_STATE,                    136)
+    X(RM, CTRL_DBG_CLEAR_SINGLE_SM_ERROR_STATE,                   137)
+    X(RM, CTRL_DBG_SET_MODE_ERRBAR_DEBUG,                         138)
+    X(RM, CTRL_DBG_SET_NEXT_STOP_TRIGGER_TYPE,                    139)
+    X(RM, CTRL_ALLOC_PMA_STREAM,                                  140)
+    X(RM, CTRL_PMA_STREAM_UPDATE_GET_PUT,                         141)
+    X(RM, CTRL_FB_GET_INFO_V2,                                    142)
+    X(RM, CTRL_FIFO_SET_CHANNEL_PROPERTIES,                       143)
+    X(RM, CTRL_GR_GET_CTX_BUFFER_INFO,                            144)
+    X(RM, CTRL_KGR_GET_CTX_BUFFER_PTES,                           145)
+    X(RM, CTRL_GPU_EVICT_CTX,                                     146)
+    X(RM, CTRL_FB_GET_FS_INFO,                                    147)
+    X(RM, CTRL_GRMGR_GET_GR_FS_INFO,                              148)
+    X(RM, CTRL_STOP_CHANNEL,                                      149)
+    X(RM, CTRL_GR_PC_SAMPLING_MODE,                               150)
+    X(RM, CTRL_PERF_RATED_TDP_GET_STATUS,                         151)
+    X(RM, CTRL_PERF_RATED_TDP_SET_CONTROL,                        152)
+    X(RM, CTRL_FREE_PMA_STREAM,                                   153)
+    X(RM, CTRL_TIMER_SET_GR_TICK_FREQ,                            154)
+    X(RM, CTRL_FIFO_SETUP_VF_ZOMBIE_SUBCTX_PDB,                   155)
+    X(RM, GET_CONSOLIDATED_GR_STATIC_INFO,                        156)
+    X(RM, CTRL_DBG_SET_SINGLE_SM_SINGLE_STEP,                     157)
+    X(RM, CTRL_GR_GET_TPC_PARTITION_MODE,                         158)
+    X(RM, CTRL_GR_SET_TPC_PARTITION_MODE,                         159)
+    X(UVM, UVM_PAGING_CHANNEL_ALLOCATE,                           160)
+    X(UVM, UVM_PAGING_CHANNEL_DESTROY,                            161)
+    X(UVM, UVM_PAGING_CHANNEL_MAP,                                162)
+    X(UVM, UVM_PAGING_CHANNEL_UNMAP,                              163)
+    X(UVM, UVM_PAGING_CHANNEL_PUSH_STREAM,                        164)
+    X(UVM, UVM_PAGING_CHANNEL_SET_HANDLES,                        165)
+    X(UVM, UVM_METHOD_STREAM_GUEST_PAGES_OPERATION,               166)
+    X(RM, CTRL_INTERNAL_QUIESCE_PMA_CHANNEL,                      167)
+    X(RM, DCE_RM_INIT,                                            168)
+    X(RM, REGISTER_VIRTUAL_EVENT_BUFFER,                          169)
+    X(RM, CTRL_EVENT_BUFFER_UPDATE_GET,                           170)
+    X(RM, GET_PLCABLE_ADDRESS_KIND,                               171)
+    X(RM, CTRL_PERF_LIMITS_SET_STATUS_V2,                         172)
+    X(RM, CTRL_INTERNAL_SRIOV_PROMOTE_PMA_STREAM,                 173)
+    X(RM, CTRL_GET_MMU_DEBUG_MODE,                                174)
+    X(RM, CTRL_INTERNAL_PROMOTE_FAULT_METHOD_BUFFERS,             175)
+    X(RM, CTRL_FLCN_GET_CTX_BUFFER_SIZE,                          176)
+    X(RM, CTRL_FLCN_GET_CTX_BUFFER_INFO,                          177)
+    X(RM, DISABLE_CHANNELS,                                       178)
+    X(RM, CTRL_FABRIC_MEMORY_DESCRIBE,                            179)
+    X(RM, CTRL_FABRIC_MEM_STATS,                                  180)
+    X(RM, SAVE_HIBERNATION_DATA,                                  181)
+    X(RM, RESTORE_HIBERNATION_DATA,                               182)
+    X(RM, CTRL_INTERNAL_MEMSYS_SET_ZBC_REFERENCED,                183)
+    X(RM, CTRL_EXEC_PARTITIONS_CREATE,                            184)
+    X(RM, CTRL_EXEC_PARTITIONS_DELETE,                            185)
+    X(RM, CTRL_GPFIFO_GET_WORK_SUBMIT_TOKEN,                      186)
+    X(RM, CTRL_GPFIFO_SET_WORK_SUBMIT_TOKEN_NOTIF_INDEX,          187)
+    X(RM, PMA_SCRUBBER_SHARED_BUFFER_GUEST_PAGES_OPERATION,       188)
+    X(RM, CTRL_MASTER_GET_VIRTUAL_FUNCTION_ERROR_CONT_INTR_MASK,  189)
+    X(RM, SET_SYSMEM_DIRTY_PAGE_TRACKING_BUFFER,                  190)
+    X(RM, CTRL_SUBDEVICE_GET_P2P_CAPS,                            191)
+    X(RM, CTRL_BUS_SET_P2P_MAPPING,                               192)
+    X(RM, CTRL_BUS_UNSET_P2P_MAPPING,                             193)
+    X(RM, CTRL_FLA_SETUP_INSTANCE_MEM_BLOCK,                      194)
+    X(RM, CTRL_GPU_MIGRATABLE_OPS,                                195)
+    X(RM, CTRL_GET_TOTAL_HS_CREDITS,                              196)
+    X(RM, CTRL_GET_HS_CREDITS,                                    197)
+    X(RM, CTRL_SET_HS_CREDITS,                                    198)
+    X(RM, CTRL_PM_AREA_PC_SAMPLER,                                199)
+    X(RM, INVALIDATE_TLB,                                         200)
+    X(RM, CTRL_GPU_QUERY_ECC_STATUS,                              201) // deprecated
+    X(RM, ECC_NOTIFIER_WRITE_ACK,                                 202)
+    X(RM, CTRL_DBG_GET_MODE_MMU_DEBUG,                            203)
+    X(RM, RM_API_CONTROL,                                         204)
+    X(RM, CTRL_CMD_INTERNAL_GPU_START_FABRIC_PROBE,               205)
+    X(RM, CTRL_NVLINK_GET_INBAND_RECEIVED_DATA,                   206)
+    X(RM, GET_STATIC_DATA,                                        207)
+    X(RM, RESERVED_208,                                           208)
+    X(RM, CTRL_GPU_GET_INFO_V2,                                   209)
+    X(RM, GET_BRAND_CAPS,                                         210)
+    X(RM, CTRL_CMD_NVLINK_INBAND_SEND_DATA,                       211)
+    X(RM, UPDATE_GPM_GUEST_BUFFER_INFO,                           212)
+    X(RM, CTRL_CMD_INTERNAL_CONTROL_GSP_TRACE,                    213)
+    X(RM, CTRL_SET_ZBC_STENCIL_CLEAR,                             214)
+    X(RM, CTRL_SUBDEVICE_GET_VGPU_HEAP_STATS,                     215)
+    X(RM, CTRL_SUBDEVICE_GET_LIBOS_HEAP_STATS,                    216)
+    X(RM, CTRL_DBG_SET_MODE_MMU_GCC_DEBUG,                        217)
+    X(RM, CTRL_DBG_GET_MODE_MMU_GCC_DEBUG,                        218)
+    X(RM, CTRL_RESERVE_HES,                                       219)
+    X(RM, CTRL_RELEASE_HES,                                       220)
+    X(RM, CTRL_RESERVE_CCU_PROF,                                  221)
+    X(RM, CTRL_RELEASE_CCU_PROF,                                  222)
+    X(RM, RESERVED,                                               223)
+    X(RM, CTRL_CMD_GET_CHIPLET_HS_CREDIT_POOL,                    224)
+    X(RM, CTRL_CMD_GET_HS_CREDITS_MAPPING,                        225)
+    X(RM, CTRL_EXEC_PARTITIONS_EXPORT,                            226)
+    X(RM, NUM_FUNCTIONS,                                          227)
+#ifdef DEFINING_X_IN_RPC_GLOBAL_ENUMS_H
+};
+#   undef X
+#   undef DEFINING_X_IN_RPC_GLOBAL_ENUMS_H
+#endif
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/ofa.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/ofa.c
new file mode 100644
index 000000000000..6fb3083edde3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/ofa.c
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/engine.h>
+
+#include "nvrm/ofa.h"
+
+static int
+r570_ofa_alloc(struct nvkm_gsp_object *parent, u32 handle, u32 oclass, int inst,
+	       struct nvkm_gsp_object *ofa)
+{
+	NV_OFA_ALLOCATION_PARAMETERS *args;
+
+	args = nvkm_gsp_rm_alloc_get(parent, handle, oclass, sizeof(*args), ofa);
+	if (WARN_ON(IS_ERR(args)))
+		return PTR_ERR(args);
+
+	args->size = sizeof(*args);
+	args->engineInstance = inst;
+
+	return nvkm_gsp_rm_alloc_wr(ofa, args);
+}
+
+const struct nvkm_rm_api_engine
+r570_ofa = {
+	.alloc = r570_ofa_alloc,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/rm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/rm.c
new file mode 100644
index 000000000000..498658d0c60c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r570/rm.c
@@ -0,0 +1,99 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <rm/rm.h>
+
+#include "nvrm/gsp.h"
+
+static const struct nvkm_rm_wpr
+r570_wpr_libos2 = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS2,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_TU10X,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS2_MIN_MB,
+};
+
+static const struct nvkm_rm_wpr
+r570_wpr_libos3 = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3_BAREMETAL,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_TU10X,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB,
+};
+
+static const struct nvkm_rm_wpr
+r570_wpr_libos3_gh100 = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3_BAREMETAL,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_GH100,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB,
+	.heap_size_non_wpr = 0x200000,
+	.offset_set_by_acr = true,
+};
+
+static const struct nvkm_rm_wpr
+r570_wpr_libos3_gb10x = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3_BAREMETAL,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_GH100,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB,
+	.heap_size_non_wpr = 0x200000,
+	.rsvd_size_pmu = ALIGN(0x0800000 + 0x1000000 + 0x0001000, 0x20000),
+	.offset_set_by_acr = true,
+};
+
+static const struct nvkm_rm_wpr
+r570_wpr_libos3_gb20x = {
+	.os_carveout_size = GSP_FW_HEAP_PARAM_OS_SIZE_LIBOS3_BAREMETAL,
+	.base_size = GSP_FW_HEAP_PARAM_BASE_RM_SIZE_GH100,
+	.heap_size_min = GSP_FW_HEAP_SIZE_OVERRIDE_LIBOS3_BAREMETAL_MIN_MB,
+	.heap_size_non_wpr = 0x220000,
+	.rsvd_size_pmu = ALIGN(0x0800000 + 0x1000000 + 0x0001000, 0x20000),
+	.offset_set_by_acr = true,
+};
+
+static const struct nvkm_rm_api
+r570_api = {
+	.gsp = &r570_gsp,
+	.rpc = &r535_rpc,
+	.ctrl = &r535_ctrl,
+	.alloc = &r535_alloc,
+	.client = &r570_client,
+	.device = &r535_device,
+	.fbsr = &r570_fbsr,
+	.disp = &r570_disp,
+	.fifo = &r570_fifo,
+	.ce = &r535_ce,
+	.gr = &r570_gr,
+	.nvdec = &r535_nvdec,
+	.nvenc = &r535_nvenc,
+	.nvjpg = &r535_nvjpg,
+	.ofa = &r570_ofa,
+};
+
+const struct nvkm_rm_impl
+r570_rm_tu102 = {
+	.wpr = &r570_wpr_libos2,
+	.api = &r570_api,
+};
+
+const struct nvkm_rm_impl
+r570_rm_ga102 = {
+	.wpr = &r570_wpr_libos3,
+	.api = &r570_api,
+};
+
+const struct nvkm_rm_impl
+r570_rm_gh100 = {
+	.wpr = &r570_wpr_libos3_gh100,
+	.api = &r570_api,
+};
+
+const struct nvkm_rm_impl
+r570_rm_gb10x = {
+	.wpr = &r570_wpr_libos3_gb10x,
+	.api = &r570_api,
+};
+
+const struct nvkm_rm_impl
+r570_rm_gb20x = {
+	.wpr = &r570_wpr_libos3_gb20x,
+	.api = &r570_api,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rm.h
new file mode 100644
index 000000000000..393ea775941f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rm.h
@@ -0,0 +1,191 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include <subdev/gsp.h>
+#ifndef __NVKM_RM_H__
+#define __NVKM_RM_H__
+#include "handles.h"
+struct nvkm_outp;
+struct r535_gr;
+
+struct nvkm_rm_impl {
+	const struct nvkm_rm_wpr *wpr;
+	const struct nvkm_rm_api *api;
+};
+
+struct nvkm_rm {
+	struct nvkm_device *device;
+	const struct nvkm_rm_gpu *gpu;
+	const struct nvkm_rm_wpr *wpr;
+	const struct nvkm_rm_api *api;
+};
+
+struct nvkm_rm_wpr {
+	u32 os_carveout_size;
+	u32 base_size;
+	u64 heap_size_min;
+	u32 heap_size_non_wpr;
+	u32 rsvd_size_pmu;
+	bool offset_set_by_acr;
+};
+
+struct nvkm_rm_api {
+	const struct nvkm_rm_api_gsp {
+		void (*set_rmargs)(struct nvkm_gsp *, bool resume);
+		int (*set_system_info)(struct nvkm_gsp *);
+		int (*get_static_info)(struct nvkm_gsp *);
+		bool (*xlat_mc_engine_idx)(u32 mc_engine_idx, enum nvkm_subdev_type *, int *inst);
+		void (*drop_send_user_shared_data)(struct nvkm_gsp *);
+		void (*drop_post_nocat_record)(struct nvkm_gsp *);
+		u32 (*sr_data_size)(struct nvkm_gsp *);
+	} *gsp;
+
+	const struct nvkm_rm_api_rpc {
+		void *(*get)(struct nvkm_gsp *, u32 fn, u32 argc);
+		void *(*push)(struct nvkm_gsp *gsp, void *argv,
+			      enum nvkm_gsp_rpc_reply_policy policy, u32 repc);
+		void (*done)(struct nvkm_gsp *gsp, void *repv);
+	} *rpc;
+
+	const struct nvkm_rm_api_ctrl {
+		void *(*get)(struct nvkm_gsp_object *, u32 cmd, u32 params_size);
+		int (*push)(struct nvkm_gsp_object *, void **params, u32 repc);
+		void (*done)(struct nvkm_gsp_object *, void *params);
+	} *ctrl;
+
+	const struct nvkm_rm_api_alloc {
+		void *(*get)(struct nvkm_gsp_object *, u32 oclass, u32 params_size);
+		void *(*push)(struct nvkm_gsp_object *, void *params);
+		void (*done)(struct nvkm_gsp_object *, void *params);
+
+		int (*free)(struct nvkm_gsp_object *);
+	} *alloc;
+
+	const struct nvkm_rm_api_client {
+		int (*ctor)(struct nvkm_gsp_client *, u32 handle);
+	} *client;
+
+	const struct nvkm_rm_api_device {
+		int (*ctor)(struct nvkm_gsp_client *, struct nvkm_gsp_device *);
+		void (*dtor)(struct nvkm_gsp_device *);
+
+		struct {
+			int (*ctor)(struct nvkm_gsp_device *, u32 handle, u32 id,
+				    nvkm_gsp_event_func, struct nvkm_gsp_event *);
+			void (*dtor)(struct nvkm_gsp_event *);
+		} event;
+	} *device;
+
+	const struct nvkm_rm_api_fbsr {
+		int (*suspend)(struct nvkm_gsp *);
+		void (*resume)(struct nvkm_gsp *);
+	} *fbsr;
+
+	const struct nvkm_rm_api_disp {
+		int (*get_static_info)(struct nvkm_disp *);
+		int (*get_supported)(struct nvkm_disp *, unsigned long *display_mask);
+		int (*get_connect_state)(struct nvkm_disp *, unsigned display_id);
+		int (*get_active)(struct nvkm_disp *, unsigned head, u32 *display_id);
+
+		int (*bl_ctrl)(struct nvkm_disp *, unsigned display_id, bool set, int *val);
+
+		struct {
+			int (*get_caps)(struct nvkm_disp *, int *link_bw, bool *mst, bool *wm);
+			int (*set_indexed_link_rates)(struct nvkm_outp *);
+		} dp;
+
+		struct {
+			int (*set_pushbuf)(struct nvkm_disp *, s32 oclass, int inst,
+					   struct nvkm_memory *);
+			int (*dmac_alloc)(struct nvkm_disp *, u32 oclass, int inst, u32 put_offset,
+					  struct nvkm_gsp_object *);
+		} chan;
+	} *disp;
+
+	const struct nvkm_rm_api_fifo {
+		int (*xlat_rm_engine_type)(u32 rm_engine_type,
+					   enum nvkm_subdev_type *, int *nv2080_type);
+		int (*ectx_size)(struct nvkm_fifo *);
+		unsigned rsvd_chids;
+		int (*rc_triggered)(void *priv, u32 fn, void *repv, u32 repc);
+		struct {
+			int (*alloc)(struct nvkm_gsp_device *, u32 handle,
+				     u32 nv2080_engine_type, u8 runq, bool priv, int chid,
+				     u64 inst_addr, u64 userd_addr, u64 mthdbuf_addr,
+				     struct nvkm_vmm *, u64 gpfifo_offset, u32 gpfifo_length,
+				     struct nvkm_gsp_object *);
+		} chan;
+	} *fifo;
+
+	const struct nvkm_rm_api_engine {
+		int (*alloc)(struct nvkm_gsp_object *chan, u32 handle, u32 class, int inst,
+			     struct nvkm_gsp_object *);
+	} *ce, *nvdec, *nvenc, *nvjpg, *ofa;
+
+	const struct nvkm_rm_api_gr {
+		int (*get_ctxbufs_info)(struct r535_gr *);
+		struct {
+			int (*init)(struct r535_gr *);
+			void (*fini)(struct r535_gr *);
+		} scrubber;
+	} *gr;
+};
+
+extern const struct nvkm_rm_impl r535_rm_tu102;
+extern const struct nvkm_rm_impl r535_rm_ga102;
+extern const struct nvkm_rm_api_gsp r535_gsp;
+typedef struct DOD_METHOD_DATA DOD_METHOD_DATA;
+typedef struct JT_METHOD_DATA JT_METHOD_DATA;
+typedef struct CAPS_METHOD_DATA CAPS_METHOD_DATA;
+void r535_gsp_acpi_dod(acpi_handle, DOD_METHOD_DATA *);
+void r535_gsp_acpi_jt(acpi_handle, JT_METHOD_DATA *);
+void r535_gsp_acpi_caps(acpi_handle, CAPS_METHOD_DATA *);
+struct NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS;
+void r535_gsp_get_static_info_fb(struct nvkm_gsp *,
+				 const struct NV2080_CTRL_CMD_FB_GET_FB_REGION_INFO_PARAMS *);
+extern const struct nvkm_rm_api_rpc r535_rpc;
+extern const struct nvkm_rm_api_ctrl r535_ctrl;
+extern const struct nvkm_rm_api_alloc r535_alloc;
+extern const struct nvkm_rm_api_client r535_client;
+void r535_gsp_client_dtor(struct nvkm_gsp_client *);
+extern const struct nvkm_rm_api_device r535_device;
+int r535_mmu_vaspace_new(struct nvkm_vmm *, u32 handle, bool external);
+void r535_mmu_vaspace_del(struct nvkm_vmm *);
+extern const struct nvkm_rm_api_fbsr r535_fbsr;
+void r535_fbsr_resume(struct nvkm_gsp *);
+int r535_fbsr_memlist(struct nvkm_gsp_device *, u32 handle, enum nvkm_memory_target,
+		      u64 phys, u64 size, struct sg_table *, struct nvkm_gsp_object *);
+extern const struct nvkm_rm_api_disp r535_disp;
+extern const struct nvkm_rm_api_fifo r535_fifo;
+void r535_fifo_rc_chid(struct nvkm_fifo *, int chid);
+extern const struct nvkm_rm_api_engine r535_ce;
+extern const struct nvkm_rm_api_gr r535_gr;
+void *r535_gr_dtor(struct nvkm_gr *);
+int r535_gr_oneinit(struct nvkm_gr *);
+u64 r535_gr_units(struct nvkm_gr *);
+int r535_gr_chan_new(struct nvkm_gr *, struct nvkm_chan *, const struct nvkm_oclass *,
+		     struct nvkm_object **);
+int r535_gr_promote_ctx(struct r535_gr *, bool golden, struct nvkm_vmm *,
+			struct nvkm_memory **pctxbuf_mem, struct nvkm_vma **pctxbuf_vma,
+			struct nvkm_gsp_object *chan);
+extern const struct nvkm_rm_api_engine r535_nvdec;
+extern const struct nvkm_rm_api_engine r535_nvenc;
+extern const struct nvkm_rm_api_engine r535_nvjpg;
+extern const struct nvkm_rm_api_engine r535_ofa;
+
+extern const struct nvkm_rm_impl r570_rm_tu102;
+extern const struct nvkm_rm_impl r570_rm_ga102;
+extern const struct nvkm_rm_impl r570_rm_gh100;
+extern const struct nvkm_rm_impl r570_rm_gb10x;
+extern const struct nvkm_rm_impl r570_rm_gb20x;
+extern const struct nvkm_rm_api_gsp r570_gsp;
+extern const struct nvkm_rm_api_client r570_client;
+extern const struct nvkm_rm_api_fbsr r570_fbsr;
+extern const struct nvkm_rm_api_disp r570_disp;
+extern const struct nvkm_rm_api_fifo r570_fifo;
+extern const struct nvkm_rm_api_gr r570_gr;
+int r570_gr_gpc_mask(struct nvkm_gsp *, u32 *mask);
+int r570_gr_tpc_mask(struct nvkm_gsp *, int gpc, u32 *mask);
+extern const struct nvkm_rm_api_engine r570_ofa;
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rpc.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rpc.h
new file mode 100644
index 000000000000..4431e33b3304
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rpc.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#ifndef __NVKM_RM_RPC_H__
+#define __NVKM_RM_RPC_H__
+#include "rm.h"
+
+#define to_payload_hdr(p, header) \
+	container_of((void *)p, typeof(*header), params)
+
+int r535_gsp_rpc_poll(struct nvkm_gsp *, u32 fn);
+
+struct nvfw_gsp_rpc *r535_gsp_msg_recv(struct nvkm_gsp *, int fn, u32 gsp_rpc_len);
+int r535_gsp_msg_ntfy_add(struct nvkm_gsp *, u32 fn, nvkm_gsp_msg_ntfy_func, void *priv);
+
+int r535_rpc_status_to_errno(uint32_t rpc_status);
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/tu1xx.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/tu1xx.c
new file mode 100644
index 000000000000..423502f870db
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/tu1xx.c
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "gpu.h"
+
+#include <engine/fifo/priv.h>
+
+#include <nvif/class.h>
+
+const struct nvkm_rm_gpu
+tu1xx_gpu = {
+	.disp.class = {
+		.root = TU102_DISP,
+		.caps = GV100_DISP_CAPS,
+		.core = TU102_DISP_CORE_CHANNEL_DMA,
+		.wndw = TU102_DISP_WINDOW_CHANNEL_DMA,
+		.wimm = TU102_DISP_WINDOW_IMM_CHANNEL_DMA,
+		.curs = TU102_DISP_CURSOR,
+	},
+
+	.usermode.class = TURING_USERMODE_A,
+
+	.fifo.chan = {
+		.class = TURING_CHANNEL_GPFIFO_A,
+		.doorbell_handle = tu102_chan_doorbell_handle,
+	},
+
+	.ce.class = TURING_DMA_COPY_A,
+	.gr.class = {
+		.i2m = KEPLER_INLINE_TO_MEMORY_B,
+		.twod = FERMI_TWOD_A,
+		.threed = TURING_A,
+		.compute = TURING_COMPUTE_A,
+	},
+	.nvdec.class = NVC4B0_VIDEO_DECODER,
+	.nvenc.class = NVC4B7_VIDEO_ENCODER,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c
index 59c5f2b9172a..58e233bc53b1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c
@@ -22,11 +22,45 @@
 #include "priv.h"
 
 #include <subdev/fb.h>
+#include <engine/sec2.h>
+
+#include <rm/r535/nvrm/gsp.h>
 
 #include <nvfw/flcn.h>
 #include <nvfw/fw.h>
 #include <nvfw/hs.h>
 
+static int
+tu102_gsp_booter_unload(struct nvkm_gsp *gsp, u32 mbox0, u32 mbox1)
+{
+	struct nvkm_subdev *subdev = &gsp->subdev;
+	struct nvkm_device *device = subdev->device;
+	u32 wpr2_hi;
+	int ret;
+
+	wpr2_hi = nvkm_rd32(device, 0x1fa828);
+	if (!wpr2_hi) {
+		nvkm_debug(subdev, "WPR2 not set - skipping booter unload\n");
+		return 0;
+	}
+
+	ret = nvkm_falcon_fw_boot(&gsp->booter.unload, &gsp->subdev, true, &mbox0, &mbox1, 0, 0);
+	if (WARN_ON(ret))
+		return ret;
+
+	wpr2_hi = nvkm_rd32(device, 0x1fa828);
+	if (WARN_ON(wpr2_hi))
+		return -EIO;
+
+	return 0;
+}
+
+static int
+tu102_gsp_booter_load(struct nvkm_gsp *gsp, u32 mbox0, u32 mbox1)
+{
+	return nvkm_falcon_fw_boot(&gsp->booter.load, &gsp->subdev, true, &mbox0, &mbox1, 0, 0);
+}
+
 int
 tu102_gsp_booter_ctor(struct nvkm_gsp *gsp, const char *name, const struct firmware *blob,
 		      struct nvkm_falcon *falcon, struct nvkm_falcon_fw *fw)
@@ -114,6 +148,118 @@ tu102_gsp_reset(struct nvkm_gsp *gsp)
 	return gsp->falcon.func->reset_eng(&gsp->falcon);
 }
 
+int
+tu102_gsp_fini(struct nvkm_gsp *gsp, bool suspend)
+{
+	u32 mbox0 = 0xff, mbox1 = 0xff;
+	int ret;
+
+	ret = r535_gsp_fini(gsp, suspend);
+	if (ret && suspend)
+		return ret;
+
+	nvkm_falcon_reset(&gsp->falcon);
+
+	ret = nvkm_gsp_fwsec_sb(gsp);
+	WARN_ON(ret);
+
+	if (suspend) {
+		mbox0 = lower_32_bits(gsp->sr.meta.addr);
+		mbox1 = upper_32_bits(gsp->sr.meta.addr);
+	}
+
+	ret = tu102_gsp_booter_unload(gsp, mbox0, mbox1);
+	WARN_ON(ret);
+	return 0;
+}
+
+int
+tu102_gsp_init(struct nvkm_gsp *gsp)
+{
+	u32 mbox0, mbox1;
+	int ret;
+
+	if (!gsp->sr.meta.data) {
+		mbox0 = lower_32_bits(gsp->wpr_meta.addr);
+		mbox1 = upper_32_bits(gsp->wpr_meta.addr);
+	} else {
+		gsp->rm->api->gsp->set_rmargs(gsp, true);
+
+		mbox0 = lower_32_bits(gsp->sr.meta.addr);
+		mbox1 = upper_32_bits(gsp->sr.meta.addr);
+	}
+
+	/* Execute booter to handle (eventually...) booting GSP-RM. */
+	ret = tu102_gsp_booter_load(gsp, mbox0, mbox1);
+	if (WARN_ON(ret))
+		return ret;
+
+	return r535_gsp_init(gsp);
+}
+
+static int
+tu102_gsp_wpr_meta_init(struct nvkm_gsp *gsp)
+{
+	GspFwWprMeta *meta;
+	int ret;
+
+	ret = nvkm_gsp_mem_ctor(gsp, sizeof(*meta), &gsp->wpr_meta);
+	if (ret)
+		return ret;
+
+	meta = gsp->wpr_meta.data;
+
+	meta->magic = GSP_FW_WPR_META_MAGIC;
+	meta->revision = GSP_FW_WPR_META_REVISION;
+
+	meta->sysmemAddrOfRadix3Elf = gsp->radix3.lvl0.addr;
+	meta->sizeOfRadix3Elf = gsp->fb.wpr2.elf.size;
+
+	meta->sysmemAddrOfBootloader = gsp->boot.fw.addr;
+	meta->sizeOfBootloader = gsp->boot.fw.size;
+	meta->bootloaderCodeOffset = gsp->boot.code_offset;
+	meta->bootloaderDataOffset = gsp->boot.data_offset;
+	meta->bootloaderManifestOffset = gsp->boot.manifest_offset;
+
+	meta->sysmemAddrOfSignature = gsp->sig.addr;
+	meta->sizeOfSignature = gsp->sig.size;
+
+	meta->gspFwRsvdStart = gsp->fb.heap.addr;
+	meta->nonWprHeapOffset = gsp->fb.heap.addr;
+	meta->nonWprHeapSize = gsp->fb.heap.size;
+	meta->gspFwWprStart = gsp->fb.wpr2.addr;
+	meta->gspFwHeapOffset = gsp->fb.wpr2.heap.addr;
+	meta->gspFwHeapSize = gsp->fb.wpr2.heap.size;
+	meta->gspFwOffset = gsp->fb.wpr2.elf.addr;
+	meta->bootBinOffset = gsp->fb.wpr2.boot.addr;
+	meta->frtsOffset = gsp->fb.wpr2.frts.addr;
+	meta->frtsSize = gsp->fb.wpr2.frts.size;
+	meta->gspFwWprEnd = ALIGN_DOWN(gsp->fb.bios.vga_workspace.addr, 0x20000);
+	meta->fbSize = gsp->fb.size;
+	meta->vgaWorkspaceOffset = gsp->fb.bios.vga_workspace.addr;
+	meta->vgaWorkspaceSize = gsp->fb.bios.vga_workspace.size;
+	meta->bootCount = 0;
+	meta->partitionRpcAddr = 0;
+	meta->partitionRpcRequestOffset = 0;
+	meta->partitionRpcReplyOffset = 0;
+	meta->verified = 0;
+	return 0;
+}
+
+u64
+tu102_gsp_wpr_heap_size(struct nvkm_gsp *gsp)
+{
+	u32 fb_size_gb = DIV_ROUND_UP_ULL(gsp->fb.size, 1 << 30);
+	u64 heap_size;
+
+	heap_size = gsp->rm->wpr->os_carveout_size +
+		    gsp->rm->wpr->base_size +
+		    ALIGN(GSP_FW_HEAP_PARAM_SIZE_PER_GB_FB * fb_size_gb, 1 << 20) +
+		    ALIGN(GSP_FW_HEAP_PARAM_CLIENT_ALLOC_SIZE, 1 << 20);
+
+	return max(heap_size, gsp->rm->wpr->heap_size_min);
+}
+
 static u64
 tu102_gsp_vga_workspace_addr(struct nvkm_gsp *gsp, u64 fb_size)
 {
@@ -136,14 +282,67 @@ tu102_gsp_vga_workspace_addr(struct nvkm_gsp *gsp, u64 fb_size)
 int
 tu102_gsp_oneinit(struct nvkm_gsp *gsp)
 {
-	gsp->fb.size = nvkm_fb_vidmem_size(gsp->subdev.device);
+	struct nvkm_device *device = gsp->subdev.device;
+	int ret;
+
+	gsp->fb.size = nvkm_fb_vidmem_size(device);
 
 	gsp->fb.bios.vga_workspace.addr = tu102_gsp_vga_workspace_addr(gsp, gsp->fb.size);
 	gsp->fb.bios.vga_workspace.size = gsp->fb.size - gsp->fb.bios.vga_workspace.addr;
 	gsp->fb.bios.addr = gsp->fb.bios.vga_workspace.addr;
 	gsp->fb.bios.size = gsp->fb.bios.vga_workspace.size;
 
-	return r535_gsp_oneinit(gsp);
+	ret = gsp->func->booter.ctor(gsp, "booter-load", gsp->fws.booter.load,
+				     &device->sec2->falcon, &gsp->booter.load);
+	if (ret)
+		return ret;
+
+	ret = gsp->func->booter.ctor(gsp, "booter-unload", gsp->fws.booter.unload,
+				     &device->sec2->falcon, &gsp->booter.unload);
+	if (ret)
+		return ret;
+
+	ret = r535_gsp_oneinit(gsp);
+	if (ret)
+		return ret;
+
+	/* Calculate FB layout. */
+	gsp->fb.wpr2.frts.size = 0x100000;
+	gsp->fb.wpr2.frts.addr = ALIGN_DOWN(gsp->fb.bios.addr, 0x20000) - gsp->fb.wpr2.frts.size;
+
+	gsp->fb.wpr2.boot.size = gsp->boot.fw.size;
+	gsp->fb.wpr2.boot.addr = ALIGN_DOWN(gsp->fb.wpr2.frts.addr - gsp->fb.wpr2.boot.size, 0x1000);
+
+	gsp->fb.wpr2.elf.size = gsp->fw.len;
+	gsp->fb.wpr2.elf.addr = ALIGN_DOWN(gsp->fb.wpr2.boot.addr - gsp->fb.wpr2.elf.size, 0x10000);
+
+	gsp->fb.wpr2.heap.size = tu102_gsp_wpr_heap_size(gsp);
+
+	gsp->fb.wpr2.heap.addr = ALIGN_DOWN(gsp->fb.wpr2.elf.addr - gsp->fb.wpr2.heap.size, 0x100000);
+	gsp->fb.wpr2.heap.size = ALIGN_DOWN(gsp->fb.wpr2.elf.addr - gsp->fb.wpr2.heap.addr, 0x100000);
+
+	gsp->fb.wpr2.addr = ALIGN_DOWN(gsp->fb.wpr2.heap.addr - sizeof(GspFwWprMeta), 0x100000);
+	gsp->fb.wpr2.size = gsp->fb.wpr2.frts.addr + gsp->fb.wpr2.frts.size - gsp->fb.wpr2.addr;
+
+	gsp->fb.heap.size = 0x100000;
+	gsp->fb.heap.addr = gsp->fb.wpr2.addr - gsp->fb.heap.size;
+
+	ret = tu102_gsp_wpr_meta_init(gsp);
+	if (ret)
+		return ret;
+
+	ret = nvkm_gsp_fwsec_frts(gsp);
+	if (WARN_ON(ret))
+		return ret;
+
+	/* Reset GSP into RISC-V mode. */
+	ret = gsp->func->reset(gsp);
+	if (ret)
+		return ret;
+
+	nvkm_falcon_wr32(&gsp->falcon, 0x040, lower_32_bits(gsp->libos.addr));
+	nvkm_falcon_wr32(&gsp->falcon, 0x044, upper_32_bits(gsp->libos.addr));
+	return 0;
 }
 
 const struct nvkm_falcon_func
@@ -163,29 +362,73 @@ tu102_gsp_flcn = {
 };
 
 static const struct nvkm_gsp_func
-tu102_gsp_r535_113_01 = {
+tu102_gsp = {
 	.flcn = &tu102_gsp_flcn,
 	.fwsec = &tu102_gsp_fwsec,
 
 	.sig_section = ".fwsignature_tu10x",
 
-	.wpr_heap.base_size = 8 << 20,
-	.wpr_heap.min_size = 64 << 20,
-
 	.booter.ctor = tu102_gsp_booter_ctor,
 
 	.dtor = r535_gsp_dtor,
 	.oneinit = tu102_gsp_oneinit,
-	.init = r535_gsp_init,
-	.fini = r535_gsp_fini,
+	.init = tu102_gsp_init,
+	.fini = tu102_gsp_fini,
 	.reset = tu102_gsp_reset,
 
-	.rm = &r535_gsp_rm,
+	.rm.gpu = &tu1xx_gpu,
 };
 
+int
+tu102_gsp_load_rm(struct nvkm_gsp *gsp, const struct nvkm_gsp_fwif *fwif)
+{
+	struct nvkm_subdev *subdev = &gsp->subdev;
+	bool enable_gsp = fwif->enable;
+	int ret;
+
+#if IS_ENABLED(CONFIG_DRM_NOUVEAU_GSP_DEFAULT)
+	enable_gsp = true;
+#endif
+	if (!nvkm_boolopt(subdev->device->cfgopt, "NvGspRm", enable_gsp))
+		return -EINVAL;
+
+	ret = nvkm_gsp_load_fw(gsp, "gsp", fwif->ver, &gsp->fws.rm);
+	if (ret)
+		return ret;
+
+	ret = nvkm_gsp_load_fw(gsp, "bootloader", fwif->ver, &gsp->fws.bl);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+int
+tu102_gsp_load(struct nvkm_gsp *gsp, int ver, const struct nvkm_gsp_fwif *fwif)
+{
+	int ret;
+
+	ret = tu102_gsp_load_rm(gsp, fwif);
+	if (ret)
+		goto done;
+
+	ret = nvkm_gsp_load_fw(gsp, "booter_load", fwif->ver, &gsp->fws.booter.load);
+	if (ret)
+		goto done;
+
+	ret = nvkm_gsp_load_fw(gsp, "booter_unload", fwif->ver, &gsp->fws.booter.unload);
+
+done:
+	if (ret)
+		nvkm_gsp_dtor_fws(gsp);
+
+	return ret;
+}
+
 static struct nvkm_gsp_fwif
 tu102_gsps[] = {
-	{  0,  r535_gsp_load, &tu102_gsp_r535_113_01, "535.113.01" },
+	{  1, tu102_gsp_load, &tu102_gsp, &r570_rm_tu102, "570.144" },
+	{  0, tu102_gsp_load, &tu102_gsp, &r535_rm_tu102, "535.113.01" },
 	{ -1, gv100_gsp_nofw, &gv100_gsp },
 	{}
 };
@@ -196,3 +439,11 @@ tu102_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 {
 	return nvkm_gsp_new_(tu102_gsps, device, type, inst, pgsp);
 }
+
+NVKM_GSP_FIRMWARE_BOOTER(tu102, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(tu104, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(tu106, 535.113.01);
+
+NVKM_GSP_FIRMWARE_BOOTER(tu102, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(tu104, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(tu106, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu116.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu116.c
index 04fbd9ed28b1..97eb046c25d0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu116.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu116.c
@@ -22,29 +22,27 @@
 #include "priv.h"
 
 static const struct nvkm_gsp_func
-tu116_gsp_r535_113_01 = {
+tu116_gsp = {
 	.flcn = &tu102_gsp_flcn,
 	.fwsec = &tu102_gsp_fwsec,
 
 	.sig_section = ".fwsignature_tu11x",
 
-	.wpr_heap.base_size = 8 << 20,
-	.wpr_heap.min_size = 64 << 20,
-
 	.booter.ctor = tu102_gsp_booter_ctor,
 
 	.dtor = r535_gsp_dtor,
 	.oneinit = tu102_gsp_oneinit,
-	.init = r535_gsp_init,
-	.fini = r535_gsp_fini,
+	.init = tu102_gsp_init,
+	.fini = tu102_gsp_fini,
 	.reset = tu102_gsp_reset,
 
-	.rm = &r535_gsp_rm,
+	.rm.gpu = &tu1xx_gpu,
 };
 
 static struct nvkm_gsp_fwif
 tu116_gsps[] = {
-	{  0,  r535_gsp_load, &tu116_gsp_r535_113_01, "535.113.01" },
+	{  1, tu102_gsp_load, &tu116_gsp, &r570_rm_tu102, "570.144" },
+	{  0, tu102_gsp_load, &tu116_gsp, &r535_rm_tu102, "535.113.01" },
 	{ -1, gv100_gsp_nofw, &gv100_gsp },
 	{}
 };
@@ -55,3 +53,9 @@ tu116_gsp_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
 {
 	return nvkm_gsp_new_(tu116_gsps, device, type, inst, pgsp);
 }
+
+NVKM_GSP_FIRMWARE_BOOTER(tu116, 535.113.01);
+NVKM_GSP_FIRMWARE_BOOTER(tu117, 535.113.01);
+
+NVKM_GSP_FIRMWARE_BOOTER(tu116, 570.144);
+NVKM_GSP_FIRMWARE_BOOTER(tu117, 570.144);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/Kbuild
index 553d540f2736..fa7a2862dd1f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/Kbuild
@@ -4,5 +4,4 @@ nvkm-y += nvkm/subdev/instmem/nv04.o
 nvkm-y += nvkm/subdev/instmem/nv40.o
 nvkm-y += nvkm/subdev/instmem/nv50.o
 nvkm-y += nvkm/subdev/instmem/gk20a.o
-
-nvkm-y += nvkm/subdev/instmem/r535.o
+nvkm-y += nvkm/subdev/instmem/gh100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c
index a2cd3330efc6..2f55bab8e132 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/base.c
@@ -182,9 +182,11 @@ nvkm_instmem_fini(struct nvkm_subdev *subdev, bool suspend)
 	int ret;
 
 	if (suspend) {
-		ret = imem->func->suspend(imem);
-		if (ret)
-			return ret;
+		if (imem->func->suspend) {
+			ret = imem->func->suspend(imem);
+			if (ret)
+				return ret;
+		}
 
 		imem->suspend = true;
 	}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gh100.c
new file mode 100644
index 000000000000..8d8dd5f8a6c7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gh100.c
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/ref/gh100/pri_nv_xal_ep.h>
+
+static void
+gh100_instmem_set_bar0_window_addr(struct nvkm_device *device, u64 addr)
+{
+	nvkm_wr32(device, NV_XAL_EP_BAR0_WINDOW, addr >> NV_XAL_EP_BAR0_WINDOW_BASE_SHIFT);
+}
+
+static const struct nvkm_instmem_func
+gh100_instmem = {
+	.fini = nv50_instmem_fini,
+	.memory_new = nv50_instobj_new,
+	.memory_wrap = nv50_instobj_wrap,
+	.set_bar0_window_addr = gh100_instmem_set_bar0_window_addr,
+};
+
+int
+gh100_instmem_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+		  struct nvkm_instmem **pimem)
+{
+	return r535_instmem_new(&gh100_instmem, device, type, inst, pimem);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv40.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv40.c
index 6b462f960922..2544b9f0ec85 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv40.c
@@ -239,7 +239,6 @@ nv40_instmem_new(struct nvkm_device *device, enum nvkm_subdev_type type, int ins
 		 struct nvkm_instmem **pimem)
 {
 	struct nv40_instmem *imem;
-	int bar;
 
 	if (!(imem = kzalloc(sizeof(*imem), GFP_KERNEL)))
 		return -ENOMEM;
@@ -247,13 +246,8 @@ nv40_instmem_new(struct nvkm_device *device, enum nvkm_subdev_type type, int ins
 	*pimem = &imem->base;
 
 	/* map bar */
-	if (device->func->resource_size(device, 2))
-		bar = 2;
-	else
-		bar = 3;
-
-	imem->iomem = ioremap_wc(device->func->resource_addr(device, bar),
-				 device->func->resource_size(device, bar));
+	imem->iomem = ioremap_wc(device->func->resource_addr(device, NVKM_BAR2_INST),
+				 device->func->resource_size(device, NVKM_BAR2_INST));
 	if (!imem->iomem) {
 		nvkm_error(&imem->base.subdev, "unable to map PRAMIN BAR\n");
 		return -EFAULT;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c
index dd5b5a17ece0..4ca6fb30743d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c
@@ -65,7 +65,7 @@ nv50_instobj_wr32_slow(struct nvkm_memory *memory, u64 offset, u32 data)
 
 	spin_lock_irqsave(&imem->base.lock, flags);
 	if (unlikely(imem->addr != base)) {
-		nvkm_wr32(device, 0x001700, base >> 16);
+		imem->base.func->set_bar0_window_addr(device, base);
 		imem->addr = base;
 	}
 	nvkm_wr32(device, 0x700000 + addr, data);
@@ -85,7 +85,7 @@ nv50_instobj_rd32_slow(struct nvkm_memory *memory, u64 offset)
 
 	spin_lock_irqsave(&imem->base.lock, flags);
 	if (unlikely(imem->addr != base)) {
-		nvkm_wr32(device, 0x001700, base >> 16);
+		imem->base.func->set_bar0_window_addr(device, base);
 		imem->addr = base;
 	}
 	data = nvkm_rd32(device, 0x700000 + addr);
@@ -172,7 +172,7 @@ nv50_instobj_kmap(struct nv50_instobj *iobj, struct nvkm_vmm *vmm)
 
 	/* Make the mapping visible to the host. */
 	iobj->bar = bar;
-	iobj->map = ioremap_wc(device->func->resource_addr(device, 3) +
+	iobj->map = ioremap_wc(device->func->resource_addr(device, NVKM_BAR2_INST) +
 			       (u32)iobj->bar->addr, size);
 	if (!iobj->map) {
 		nvkm_warn(subdev, "PRAMIN ioremap failed\n");
@@ -353,7 +353,7 @@ nv50_instobj_func = {
 	.map = nv50_instobj_map,
 };
 
-static int
+int
 nv50_instobj_wrap(struct nvkm_instmem *base,
 		  struct nvkm_memory *memory, struct nvkm_memory **pmemory)
 {
@@ -373,7 +373,7 @@ nv50_instobj_wrap(struct nvkm_instmem *base,
 	return 0;
 }
 
-static int
+int
 nv50_instobj_new(struct nvkm_instmem *imem, u32 size, u32 align, bool zero,
 		 struct nvkm_memory **pmemory)
 {
@@ -395,6 +395,12 @@ nv50_instobj_new(struct nvkm_instmem *imem, u32 size, u32 align, bool zero,
  *****************************************************************************/
 
 static void
+nv50_instmem_set_bar0_window_addr(struct nvkm_device *device, u64 addr)
+{
+	nvkm_wr32(device, 0x001700, addr >> 16);
+}
+
+void
 nv50_instmem_fini(struct nvkm_instmem *base)
 {
 	nv50_instmem(base)->addr = ~0ULL;
@@ -415,6 +421,7 @@ nv50_instmem = {
 	.memory_new = nv50_instobj_new,
 	.memory_wrap = nv50_instobj_wrap,
 	.zero = false,
+	.set_bar0_window_addr = nv50_instmem_set_bar0_window_addr,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/priv.h
index 4c14c96fb60a..87bbdd786eaa 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/priv.h
@@ -16,10 +16,16 @@ struct nvkm_instmem_func {
 			  bool zero, struct nvkm_memory **);
 	int (*memory_wrap)(struct nvkm_instmem *, struct nvkm_memory *, struct nvkm_memory **);
 	bool zero;
+	void (*set_bar0_window_addr)(struct nvkm_device *, u64 addr);
 };
 
 int nv50_instmem_new_(const struct nvkm_instmem_func *, struct nvkm_device *,
 		      enum nvkm_subdev_type, int, struct nvkm_instmem **);
+void nv50_instmem_fini(struct nvkm_instmem *);
+int nv50_instobj_new(struct nvkm_instmem *, u32 size, u32 align, bool zero,
+		     struct nvkm_memory **);
+int nv50_instobj_wrap(struct nvkm_instmem *, struct nvkm_memory *vram,
+		      struct nvkm_memory **bar2);
 
 void nvkm_instmem_ctor(const struct nvkm_instmem_func *, struct nvkm_device *,
 		       enum nvkm_subdev_type, int, struct nvkm_instmem *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild
index 7ba35ea59c06..ea4848931540 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild
@@ -15,8 +15,7 @@ nvkm-y += nvkm/subdev/mmu/gp100.o
 nvkm-y += nvkm/subdev/mmu/gp10b.o
 nvkm-y += nvkm/subdev/mmu/gv100.o
 nvkm-y += nvkm/subdev/mmu/tu102.o
-
-nvkm-y += nvkm/subdev/mmu/r535.o
+nvkm-y += nvkm/subdev/mmu/gh100.o
 
 nvkm-y += nvkm/subdev/mmu/mem.o
 nvkm-y += nvkm/subdev/mmu/memnv04.o
@@ -38,6 +37,7 @@ nvkm-y += nvkm/subdev/mmu/vmmgp100.o
 nvkm-y += nvkm/subdev/mmu/vmmgp10b.o
 nvkm-y += nvkm/subdev/mmu/vmmgv100.o
 nvkm-y += nvkm/subdev/mmu/vmmtu102.o
+nvkm-y += nvkm/subdev/mmu/vmmgh100.o
 
 nvkm-y += nvkm/subdev/mmu/umem.o
 nvkm-y += nvkm/subdev/mmu/ummu.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gh100.c
new file mode 100644
index 000000000000..2918fb32cc91
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/gh100.c
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "mem.h"
+#include "vmm.h"
+
+#include <nvif/class.h>
+
+static const struct nvkm_mmu_func
+gh100_mmu = {
+	.dma_bits = 52,
+	.mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}},
+	.mem = {{ -1,  0, NVIF_CLASS_MEM_GF100}, gf100_mem_new, gf100_mem_map },
+	.vmm = {{ -1,  0, NVIF_CLASS_VMM_GP100}, gh100_vmm_new },
+	.kind = tu102_mmu_kind,
+	.kind_sys = true,
+};
+
+int
+gh100_mmu_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+	      struct nvkm_mmu **pmmu)
+{
+	return r535_mmu_new(&gh100_mmu, device, type, inst, pmmu);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memgf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memgf100.c
index d9c9bee45222..160a5749a29f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memgf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memgf100.c
@@ -60,7 +60,7 @@ gf100_mem_map(struct nvkm_mmu *mmu, struct nvkm_memory *memory, void *argv,
 	if (ret)
 		return ret;
 
-	*paddr = device->func->resource_addr(device, 1) + (*pvma)->addr;
+	*paddr = device->func->resource_addr(device, NVKM_BAR1_FB) + (*pvma)->addr;
 	*psize = (*pvma)->size;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv04.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv04.c
index 79a3b0cc9f5b..1e3db52de6cb 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv04.c
@@ -41,7 +41,7 @@ nv04_mem_map(struct nvkm_mmu *mmu, struct nvkm_memory *memory, void *argv,
 	if ((ret = nvif_unvers(ret, &argv, &argc, args->vn)))
 		return ret;
 
-	*paddr = device->func->resource_addr(device, 1) + addr;
+	*paddr = device->func->resource_addr(device, NVKM_BAR1_FB) + addr;
 	*psize = nvkm_memory_size(memory);
 	*pvma = ERR_PTR(-ENODEV);
 	return 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv50.c
index 46759b89fc1f..33b2321e9d87 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/memnv50.c
@@ -57,7 +57,7 @@ nv50_mem_map(struct nvkm_mmu *mmu, struct nvkm_memory *memory, void *argv,
 	if (ret)
 		return ret;
 
-	*paddr = device->func->resource_addr(device, 1) + (*pvma)->addr;
+	*paddr = device->func->resource_addr(device, NVKM_BAR1_FB) + (*pvma)->addr;
 	*psize = (*pvma)->size;
 	return nvkm_memory_map(memory, 0, bar, *pvma, &uvmm, sizeof(uvmm));
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h
index e9ca6537778c..90efef8f0b54 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/priv.h
@@ -53,6 +53,8 @@ const u8 *gf100_mmu_kind(struct nvkm_mmu *, int *count, u8 *invalid);
 
 const u8 *gm200_mmu_kind(struct nvkm_mmu *, int *, u8 *);
 
+const u8 *tu102_mmu_kind(struct nvkm_mmu *, int *, u8 *);
+
 struct nvkm_mmu_pt {
 	union {
 		struct nvkm_mmu_ptc *ptc;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu102.c
index df662ce4a4b0..7acff3642e20 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu102.c
@@ -28,7 +28,7 @@
 
 #include <nvif/class.h>
 
-static const u8 *
+const u8 *
 tu102_mmu_kind(struct nvkm_mmu *mmu, int *count, u8 *invalid)
 {
 	static const u8
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
index 9c97800fe037..f95c58b67633 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c
@@ -19,7 +19,7 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  */
-#define NVKM_VMM_LEVELS_MAX 5
+#define NVKM_VMM_LEVELS_MAX 6
 #include "vmm.h"
 
 #include <subdev/fb.h>
@@ -1030,12 +1030,8 @@ nvkm_vmm_dtor(struct nvkm_vmm *vmm)
 	struct nvkm_vma *vma;
 	struct rb_node *node;
 
-	if (vmm->rm.client.gsp) {
-		nvkm_gsp_rm_free(&vmm->rm.object);
-		nvkm_gsp_device_dtor(&vmm->rm.device);
-		nvkm_gsp_client_dtor(&vmm->rm.client);
-		nvkm_vmm_put(vmm, &vmm->rm.rsvd);
-	}
+	if (vmm->rm.client.gsp)
+		r535_mmu_vaspace_del(vmm);
 
 	if (0)
 		nvkm_vmm_dump(vmm);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
index f9bc30cdb2b3..4586a425dbe4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
@@ -143,6 +143,8 @@ struct nvkm_vmm_func {
 	int (*aper)(enum nvkm_memory_target);
 	int (*valid)(struct nvkm_vmm *, void *argv, u32 argc,
 		     struct nvkm_vmm_map *);
+	int (*valid2)(struct nvkm_vmm *, bool ro, bool priv, u8 kind, u8 comp,
+		      struct nvkm_vmm_map *);
 	void (*flush)(struct nvkm_vmm *, int depth);
 
 	int (*mthd)(struct nvkm_vmm *, struct nvkm_client *,
@@ -254,6 +256,8 @@ void gp100_vmm_invalidate_pdb(struct nvkm_vmm *, u64 addr);
 
 int gv100_vmm_join(struct nvkm_vmm *, struct nvkm_memory *);
 
+void tu102_vmm_flush(struct nvkm_vmm *, int depth);
+
 int nv04_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		 struct lock_class_key *, const char *, struct nvkm_vmm **);
 int nv41_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
@@ -296,6 +300,9 @@ int gv100_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 int tu102_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
 		  struct lock_class_key *, const char *,
 		  struct nvkm_vmm **);
+int gh100_vmm_new(struct nvkm_mmu *, bool, u64, u64, void *, u32,
+		  struct lock_class_key *, const char *,
+		  struct nvkm_vmm **);
 
 #define VMM_PRINT(l,v,p,f,a...) do {                                           \
 	struct nvkm_vmm *_vmm = (v);                                           \
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgh100.c
new file mode 100644
index 000000000000..5614df3432da
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgh100.c
@@ -0,0 +1,306 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "vmm.h"
+
+#include <subdev/fb.h>
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gh100/dev_mmu.h>
+
+static inline void
+gh100_vmm_pgt_pte(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes,
+		  struct nvkm_vmm_map *map, u64 addr)
+{
+	u64 data = addr | map->type;
+
+	while (ptes--) {
+		VMM_WO064(pt, vmm, ptei++ * NV_MMU_VER3_PTE__SIZE, data);
+		data += map->next;
+	}
+}
+
+static void
+gh100_vmm_pgt_sgl(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes,
+		  struct nvkm_vmm_map *map)
+{
+	VMM_MAP_ITER_SGL(vmm, pt, ptei, ptes, map, gh100_vmm_pgt_pte);
+}
+
+static void
+gh100_vmm_pgt_dma(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes,
+		  struct nvkm_vmm_map *map)
+{
+	if (map->page->shift == PAGE_SHIFT) {
+		VMM_SPAM(vmm, "DMAA %08x %08x PTE(s)", ptei, ptes);
+
+		nvkm_kmap(pt->memory);
+		while (ptes--) {
+			const u64 data = *map->dma++ | map->type;
+
+			VMM_WO064(pt, vmm, ptei++ * NV_MMU_VER3_PTE__SIZE, data);
+		}
+		nvkm_done(pt->memory);
+		return;
+	}
+
+	VMM_MAP_ITER_DMA(vmm, pt, ptei, ptes, map, gh100_vmm_pgt_pte);
+}
+
+static void
+gh100_vmm_pgt_mem(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes,
+		  struct nvkm_vmm_map *map)
+{
+	VMM_MAP_ITER_MEM(vmm, pt, ptei, ptes, map, gh100_vmm_pgt_pte);
+}
+
+static void
+gh100_vmm_pgt_sparse(struct nvkm_vmm *vmm,
+		     struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes)
+{
+	const u64 data = NVDEF(NV_MMU, VER3_PTE, PCF, SPARSE);
+
+	VMM_FO064(pt, vmm, ptei * NV_MMU_VER3_PTE__SIZE, data, ptes);
+}
+
+static const struct nvkm_vmm_desc_func
+gh100_vmm_desc_spt = {
+	.unmap = gf100_vmm_pgt_unmap,
+	.sparse = gh100_vmm_pgt_sparse,
+	.mem = gh100_vmm_pgt_mem,
+	.dma = gh100_vmm_pgt_dma,
+	.sgl = gh100_vmm_pgt_sgl,
+};
+
+static void
+gh100_vmm_lpt_invalid(struct nvkm_vmm *vmm,
+		      struct nvkm_mmu_pt *pt, u32 ptei, u32 ptes)
+{
+	const u64 data = NVDEF(NV_MMU, VER3_PTE, PCF, NO_VALID_4KB_PAGE);
+
+	VMM_FO064(pt, vmm, ptei * NV_MMU_VER3_PTE__SIZE, data, ptes);
+}
+
+static const struct nvkm_vmm_desc_func
+gh100_vmm_desc_lpt = {
+	.invalid = gh100_vmm_lpt_invalid,
+	.unmap = gf100_vmm_pgt_unmap,
+	.sparse = gh100_vmm_pgt_sparse,
+	.mem = gh100_vmm_pgt_mem,
+};
+
+static inline void
+gh100_vmm_pd0_pte(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt,
+		  u32 ptei, u32 ptes, struct nvkm_vmm_map *map, u64 addr)
+{
+	u64 data = addr | map->type;
+
+	while (ptes--) {
+		VMM_WO128(pt, vmm, ptei++ * NV_MMU_VER3_DUAL_PDE__SIZE, data, 0ULL);
+		data += map->next;
+	}
+}
+
+static void
+gh100_vmm_pd0_mem(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt,
+		  u32 ptei, u32 ptes, struct nvkm_vmm_map *map)
+{
+	VMM_MAP_ITER_MEM(vmm, pt, ptei, ptes, map, gh100_vmm_pd0_pte);
+}
+
+static inline bool
+gh100_vmm_pde(struct nvkm_mmu_pt *pt, u64 *data)
+{
+	switch (nvkm_memory_target(pt->memory)) {
+	case NVKM_MEM_TARGET_VRAM:
+		*data |= NVDEF(NV_MMU, VER3_PDE, APERTURE, VIDEO_MEMORY);
+		*data |= NVDEF(NV_MMU, VER3_PDE, PCF, VALID_CACHED_ATS_NOT_ALLOWED);
+		break;
+	case NVKM_MEM_TARGET_HOST:
+		*data |= NVDEF(NV_MMU, VER3_PDE, APERTURE, SYSTEM_COHERENT_MEMORY);
+		*data |= NVDEF(NV_MMU, VER3_PDE, PCF, VALID_UNCACHED_ATS_ALLOWED);
+		break;
+	case NVKM_MEM_TARGET_NCOH:
+		*data |= NVDEF(NV_MMU, VER3_PDE, APERTURE, SYSTEM_NON_COHERENT_MEMORY);
+		*data |= NVDEF(NV_MMU, VER3_PDE, PCF, VALID_CACHED_ATS_ALLOWED);
+		break;
+	default:
+		WARN_ON(1);
+		return false;
+	}
+
+	*data |= pt->addr;
+	return true;
+}
+
+static void
+gh100_vmm_pd0_pde(struct nvkm_vmm *vmm, struct nvkm_vmm_pt *pgd, u32 pdei)
+{
+	struct nvkm_vmm_pt *pgt = pgd->pde[pdei];
+	struct nvkm_mmu_pt *pd = pgd->pt[0];
+	u64 data[2] = {};
+
+	if (pgt->pt[0] && !gh100_vmm_pde(pgt->pt[0], &data[0]))
+		return;
+	if (pgt->pt[1] && !gh100_vmm_pde(pgt->pt[1], &data[1]))
+		return;
+
+	nvkm_kmap(pd->memory);
+	VMM_WO128(pd, vmm, pdei * NV_MMU_VER3_DUAL_PDE__SIZE, data[0], data[1]);
+	nvkm_done(pd->memory);
+}
+
+static void
+gh100_vmm_pd0_sparse(struct nvkm_vmm *vmm,
+		     struct nvkm_mmu_pt *pt, u32 pdei, u32 pdes)
+{
+	const u64 data = NVDEF(NV_MMU, VER3_DUAL_PDE, PCF_BIG, SPARSE_ATS_ALLOWED);
+
+	VMM_FO128(pt, vmm, pdei * NV_MMU_VER3_DUAL_PDE__SIZE, data, 0ULL, pdes);
+}
+
+static void
+gh100_vmm_pd0_unmap(struct nvkm_vmm *vmm,
+		    struct nvkm_mmu_pt *pt, u32 pdei, u32 pdes)
+{
+	VMM_FO128(pt, vmm, pdei * NV_MMU_VER3_DUAL_PDE__SIZE, 0ULL, 0ULL, pdes);
+}
+
+static const struct nvkm_vmm_desc_func
+gh100_vmm_desc_pd0 = {
+	.unmap = gh100_vmm_pd0_unmap,
+	.sparse = gh100_vmm_pd0_sparse,
+	.pde = gh100_vmm_pd0_pde,
+	.mem = gh100_vmm_pd0_mem,
+};
+
+static void
+gh100_vmm_pd1_pde(struct nvkm_vmm *vmm, struct nvkm_vmm_pt *pgd, u32 pdei)
+{
+	struct nvkm_vmm_pt *pgt = pgd->pde[pdei];
+	struct nvkm_mmu_pt *pd = pgd->pt[0];
+	u64 data = 0;
+
+	if (!gh100_vmm_pde(pgt->pt[0], &data))
+		return;
+
+	nvkm_kmap(pd->memory);
+	VMM_WO064(pd, vmm, pdei * NV_MMU_VER3_PDE__SIZE, data);
+	nvkm_done(pd->memory);
+}
+
+static const struct nvkm_vmm_desc_func
+gh100_vmm_desc_pd1 = {
+	.unmap = gf100_vmm_pgt_unmap,
+	.sparse = gh100_vmm_pgt_sparse,
+	.pde = gh100_vmm_pd1_pde,
+};
+
+static const struct nvkm_vmm_desc
+gh100_vmm_desc_16[] = {
+	{ LPT, 5,  8, 0x0100, &gh100_vmm_desc_lpt },
+	{ PGD, 8, 16, 0x1000, &gh100_vmm_desc_pd0 },
+	{ PGD, 9,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{ PGD, 9,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{ PGD, 9,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{ PGD, 1,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{}
+};
+
+static const struct nvkm_vmm_desc
+gh100_vmm_desc_12[] = {
+	{ SPT, 9,  8, 0x1000, &gh100_vmm_desc_spt },
+	{ PGD, 8, 16, 0x1000, &gh100_vmm_desc_pd0 },
+	{ PGD, 9,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{ PGD, 9,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{ PGD, 9,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{ PGD, 1,  8, 0x1000, &gh100_vmm_desc_pd1 },
+	{}
+};
+
+static int
+gh100_vmm_valid(struct nvkm_vmm *vmm, bool ro, bool priv, u8 kind, u8 comp,
+		struct nvkm_vmm_map *map)
+{
+	const enum nvkm_memory_target target = nvkm_memory_target(map->memory);
+	const bool vol = target == NVKM_MEM_TARGET_HOST;
+	const struct nvkm_vmm_page *page = map->page;
+	u8 kind_inv, pcf;
+	int kindn, aper;
+	const u8 *kindm;
+
+	map->next = 1ULL << page->shift;
+	map->type = 0;
+
+	aper = vmm->func->aper(target);
+	if (WARN_ON(aper < 0))
+		return aper;
+
+	kindm = vmm->mmu->func->kind(vmm->mmu, &kindn, &kind_inv);
+	if (kind >= kindn || kindm[kind] == kind_inv) {
+		VMM_DEBUG(vmm, "kind %02x", kind);
+		return -EINVAL;
+	}
+
+	if (priv) {
+		if (ro) {
+			if (vol)
+				pcf = NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_ATOMIC_UNCACHED_ACD;
+			else
+				pcf = NV_MMU_VER3_PTE_PCF_PRIVILEGE_RO_ATOMIC_CACHED_ACD;
+		} else {
+			if (vol)
+				pcf = NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_ATOMIC_UNCACHED_ACD;
+			else
+				pcf = NV_MMU_VER3_PTE_PCF_PRIVILEGE_RW_ATOMIC_CACHED_ACD;
+		}
+	} else {
+		if (ro) {
+			if (vol)
+				pcf = NV_MMU_VER3_PTE_PCF_REGULAR_RO_ATOMIC_UNCACHED_ACD;
+			else
+				pcf = NV_MMU_VER3_PTE_PCF_REGULAR_RO_ATOMIC_CACHED_ACD;
+		} else {
+			if (vol)
+				pcf = NV_MMU_VER3_PTE_PCF_REGULAR_RW_ATOMIC_UNCACHED_ACD;
+			else
+				pcf = NV_MMU_VER3_PTE_PCF_REGULAR_RW_ATOMIC_CACHED_ACD;
+		}
+	}
+
+	map->type |= NVDEF(NV_MMU, VER3_PTE, VALID, TRUE);
+	map->type |= NVVAL(NV_MMU, VER3_PTE, APERTURE, aper);
+	map->type |= NVVAL(NV_MMU, VER3_PTE, PCF, pcf);
+	map->type |= NVVAL(NV_MMU, VER3_PTE, KIND, kind);
+	return 0;
+}
+
+static const struct nvkm_vmm_func
+gh100_vmm = {
+	.join = gv100_vmm_join,
+	.part = gf100_vmm_part,
+	.aper = gf100_vmm_aper,
+	.valid = gp100_vmm_valid,
+	.valid2 = gh100_vmm_valid,
+	.flush = tu102_vmm_flush,
+	.page = {
+		{ 56, &gh100_vmm_desc_16[5], NVKM_VMM_PAGE_Sxxx },
+		{ 47, &gh100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx },
+		{ 38, &gh100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx },
+		{ 29, &gh100_vmm_desc_16[2], NVKM_VMM_PAGE_SVxC },
+		{ 21, &gh100_vmm_desc_16[1], NVKM_VMM_PAGE_SVxC },
+		{ 16, &gh100_vmm_desc_16[0], NVKM_VMM_PAGE_SVxC },
+		{ 12, &gh100_vmm_desc_12[0], NVKM_VMM_PAGE_SVHx },
+		{}
+	}
+};
+
+int
+gh100_vmm_new(struct nvkm_mmu *mmu, bool managed, u64 addr, u64 size,
+	      void *argv, u32 argc, struct lock_class_key *key,
+	      const char *name, struct nvkm_vmm **pvmm)
+{
+	return gp100_vmm_new_(&gh100_vmm, mmu, managed, addr, size,
+			      argv, argc, key, name, pvmm);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
index bddac77f48f0..851fd847a2a9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
@@ -436,6 +436,9 @@ gp100_vmm_valid(struct nvkm_vmm *vmm, void *argv, u32 argc,
 		return ret;
 	}
 
+	if (vmm->func->valid2)
+		return vmm->func->valid2(vmm, ro, priv, kind, 0, map);
+
 	aper = vmm->func->aper(target);
 	if (WARN_ON(aper < 0))
 		return aper;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu102.c
index 8379e72d77ab..4b30eab40bba 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu102.c
@@ -23,7 +23,7 @@
 
 #include <subdev/timer.h>
 
-static void
+void
 tu102_vmm_flush(struct nvkm_vmm *vmm, int depth)
 {
 	struct nvkm_device *device = vmm->mmu->subdev.device;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
index 174bdf995271..a14ea0f7b1c8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
@@ -13,3 +13,4 @@ nvkm-y += nvkm/subdev/pci/gf100.o
 nvkm-y += nvkm/subdev/pci/gf106.o
 nvkm-y += nvkm/subdev/pci/gk104.o
 nvkm-y += nvkm/subdev/pci/gp100.o
+nvkm-y += nvkm/subdev/pci/gh100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
index 5a0de45d36ce..6867934256a7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
@@ -39,26 +39,26 @@ nvkm_pci_msi_rearm(struct nvkm_device *device)
 u32
 nvkm_pci_rd32(struct nvkm_pci *pci, u16 addr)
 {
-	return pci->func->rd32(pci, addr);
+	return nvkm_rd32(pci->subdev.device, pci->func->cfg.addr + addr);
 }
 
 void
 nvkm_pci_wr08(struct nvkm_pci *pci, u16 addr, u8 data)
 {
-	pci->func->wr08(pci, addr, data);
+	nvkm_wr08(pci->subdev.device, pci->func->cfg.addr + addr, data);
 }
 
 void
 nvkm_pci_wr32(struct nvkm_pci *pci, u16 addr, u32 data)
 {
-	pci->func->wr32(pci, addr, data);
+	nvkm_wr32(pci->subdev.device, pci->func->cfg.addr + addr, data);
 }
 
 u32
 nvkm_pci_mask(struct nvkm_pci *pci, u16 addr, u32 mask, u32 value)
 {
-	u32 data = pci->func->rd32(pci, addr);
-	pci->func->wr32(pci, addr, (data & ~mask) | value);
+	u32 data = nvkm_pci_rd32(pci, addr);
+	nvkm_pci_wr32(pci, addr, (data & ~mask) | value);
 	return data;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
index 5b29aacedef3..5308f6539a3f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
@@ -132,10 +132,9 @@ g84_pcie_init(struct nvkm_pci *pci)
 
 static const struct nvkm_pci_func
 g84_pci_func = {
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
+
 	.init = g84_pci_init,
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
 	.msi_rearm = nv46_pci_msi_rearm,
 
 	.pcie.init = g84_pcie_init,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g92.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g92.c
index a9e0674009c6..8ae7aa02e675 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g92.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g92.c
@@ -33,10 +33,9 @@ g92_pcie_version_supported(struct nvkm_pci *pci)
 
 static const struct nvkm_pci_func
 g92_pci_func = {
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
+
 	.init = g84_pci_init,
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
 	.msi_rearm = nv46_pci_msi_rearm,
 
 	.pcie.init = g84_pcie_init,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
index 7bacd0693283..df745d0690ca 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
@@ -25,10 +25,9 @@
 
 static const struct nvkm_pci_func
 g94_pci_func = {
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
+
 	.init = g84_pci_init,
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
 	.msi_rearm = nv40_pci_msi_rearm,
 
 	.pcie.init = g84_pcie_init,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c
index 099906092fe1..6ce941df87b7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c
@@ -78,10 +78,9 @@ gf100_pcie_set_link(struct nvkm_pci *pci, enum nvkm_pcie_speed speed, u8 width)
 
 static const struct nvkm_pci_func
 gf100_pci_func = {
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
+
 	.init = g84_pci_init,
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
 	.msi_rearm = gf100_pci_msi_rearm,
 
 	.pcie.init = gf100_pcie_init,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf106.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf106.c
index bcde609ba866..712ca7e0959a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf106.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf106.c
@@ -25,10 +25,9 @@
 
 static const struct nvkm_pci_func
 gf106_pci_func = {
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
+
 	.init = g84_pci_init,
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
 	.msi_rearm = nv40_pci_msi_rearm,
 
 	.pcie.init = gf100_pcie_init,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gh100.c
new file mode 100644
index 000000000000..42da92d7a5fe
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gh100.c
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+ */
+#include "priv.h"
+
+#include <nvhw/drf.h>
+#include <nvhw/ref/gh100/dev_xtl_ep_pri.h>
+
+static void
+gh100_pci_msi_rearm(struct nvkm_pci *pci)
+{
+	/* Handled by top-level intr ACK. */
+}
+
+static const struct nvkm_pci_func
+gh100_pci = {
+	.cfg = {
+		.addr = DRF_LO(NV_EP_PCFGM),
+		.size = DRF_HI(NV_EP_PCFGM) - DRF_LO(NV_EP_PCFGM) + 1,
+	},
+	.msi_rearm = gh100_pci_msi_rearm,
+};
+
+int
+gh100_pci_new(struct nvkm_device *device, enum nvkm_subdev_type type, int inst,
+	      struct nvkm_pci **ppci)
+{
+	return nvkm_pci_new_(&gh100_pci, device, type, inst, ppci);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gk104.c
index 6be87ecffc89..ec6d0a7de995 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gk104.c
@@ -204,10 +204,9 @@ gk104_pcie_set_link(struct nvkm_pci *pci, enum nvkm_pcie_speed speed, u8 width)
 
 static const struct nvkm_pci_func
 gk104_pci_func = {
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
+
 	.init = g84_pci_init,
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
 	.msi_rearm = nv40_pci_msi_rearm,
 
 	.pcie.init = gk104_pcie_init,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gp100.c
index a5fafda0014d..4204316a544f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gp100.c
@@ -31,9 +31,7 @@ gp100_pci_msi_rearm(struct nvkm_pci *pci)
 
 static const struct nvkm_pci_func
 gp100_pci_func = {
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
 	.msi_rearm = gp100_pci_msi_rearm,
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv04.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv04.c
index 9ab64194b185..b8a3f6850fa7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv04.c
@@ -23,32 +23,9 @@
  */
 #include "priv.h"
 
-static u32
-nv04_pci_rd32(struct nvkm_pci *pci, u16 addr)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	return nvkm_rd32(device, 0x001800 + addr);
-}
-
-static void
-nv04_pci_wr08(struct nvkm_pci *pci, u16 addr, u8 data)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	nvkm_wr08(device, 0x001800 + addr, data);
-}
-
-static void
-nv04_pci_wr32(struct nvkm_pci *pci, u16 addr, u32 data)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	nvkm_wr32(device, 0x001800 + addr, data);
-}
-
 static const struct nvkm_pci_func
 nv04_pci_func = {
-	.rd32 = nv04_pci_rd32,
-	.wr08 = nv04_pci_wr08,
-	.wr32 = nv04_pci_wr32,
+	.cfg = { .addr = 0x001800, .size = 0x1000 },
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c
index 6a3c31cf0200..1971dbbdeb2b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c
@@ -23,27 +23,6 @@
  */
 #include "priv.h"
 
-u32
-nv40_pci_rd32(struct nvkm_pci *pci, u16 addr)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	return nvkm_rd32(device, 0x088000 + addr);
-}
-
-void
-nv40_pci_wr08(struct nvkm_pci *pci, u16 addr, u8 data)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	nvkm_wr08(device, 0x088000 + addr, data);
-}
-
-void
-nv40_pci_wr32(struct nvkm_pci *pci, u16 addr, u32 data)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	nvkm_wr32(device, 0x088000 + addr, data);
-}
-
 void
 nv40_pci_msi_rearm(struct nvkm_pci *pci)
 {
@@ -52,9 +31,7 @@ nv40_pci_msi_rearm(struct nvkm_pci *pci)
 
 static const struct nvkm_pci_func
 nv40_pci_func = {
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
 	.msi_rearm = nv40_pci_msi_rearm,
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c
index 9cad17f178ec..0093eabac9ae 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c
@@ -38,9 +38,7 @@ nv46_pci_msi_rearm(struct nvkm_pci *pci)
 
 static const struct nvkm_pci_func
 nv46_pci_func = {
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
 	.msi_rearm = nv46_pci_msi_rearm,
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv4c.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv4c.c
index 741e34bf307c..b445081bb80e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv4c.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv4c.c
@@ -25,9 +25,7 @@
 
 static const struct nvkm_pci_func
 nv4c_pci_func = {
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
+	.cfg = { .addr = 0x088000, .size = 0x1000 },
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
index 9b7583532962..988eeee1471c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
@@ -8,10 +8,12 @@ int nvkm_pci_new_(const struct nvkm_pci_func *, struct nvkm_device *, enum nvkm_
 		  struct nvkm_pci **);
 
 struct nvkm_pci_func {
+	struct {
+		u32 addr;
+		u16 size;
+	} cfg;
+
 	void (*init)(struct nvkm_pci *);
-	u32 (*rd32)(struct nvkm_pci *, u16 addr);
-	void (*wr08)(struct nvkm_pci *, u16 addr, u8 data);
-	void (*wr32)(struct nvkm_pci *, u16 addr, u32 data);
 	void (*msi_rearm)(struct nvkm_pci *);
 
 	struct {
@@ -27,9 +29,6 @@ struct nvkm_pci_func {
 	} pcie;
 };
 
-u32 nv40_pci_rd32(struct nvkm_pci *, u16);
-void nv40_pci_wr08(struct nvkm_pci *, u16, u8);
-void nv40_pci_wr32(struct nvkm_pci *, u16, u32);
 void nv40_pci_msi_rearm(struct nvkm_pci *);
 
 void nv46_pci_msi_rearm(struct nvkm_pci *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/r535.c
index dce337306cab..9446049642e1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/r535.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/r535.c
@@ -21,6 +21,8 @@
  */
 #include "priv.h"
 
+#include <rm/gpu.h>
+
 static void
 r535_vfn_dtor(struct nvkm_vfn *vfn)
 {
@@ -32,6 +34,7 @@ r535_vfn_new(const struct nvkm_vfn_func *hw,
 	     struct nvkm_device *device, enum nvkm_subdev_type type, int inst, u32 addr,
 	     struct nvkm_vfn **pvfn)
 {
+	const struct nvkm_rm_gpu *gpu = device->gsp->rm->gpu;
 	struct nvkm_vfn_func *rm;
 	int ret;
 
@@ -39,8 +42,12 @@ r535_vfn_new(const struct nvkm_vfn_func *hw,
 		return -ENOMEM;
 
 	rm->dtor = r535_vfn_dtor;
-	rm->intr = hw->intr;
-	rm->user = hw->user;
+	rm->intr = &tu102_vfn_intr,
+	rm->user.addr = 0x030000;
+	rm->user.size = 0x010000;
+	rm->user.base.minver = -1;
+	rm->user.base.maxver = -1;
+	rm->user.base.oclass = gpu->usermode.class;
 
 	ret = nvkm_vfn_new_(rm, device, type, inst, addr, pvfn);
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/uvfn.c b/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/uvfn.c
index c5460a14c541..4e64d8843373 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/uvfn.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/vfn/uvfn.c
@@ -36,7 +36,7 @@ nvkm_uvfn_map(struct nvkm_object *object, void *argv, u32 argc,
 	struct nvkm_vfn *vfn = nvkm_uvfn(object)->vfn;
 	struct nvkm_device *device = vfn->subdev.device;
 
-	*addr = device->func->resource_addr(device, 0) + vfn->addr.user;
+	*addr = device->func->resource_addr(device, NVKM_BAR0_PRI) + vfn->addr.user;
 	*size = vfn->func->user.size;
 	*type = NVKM_OBJECT_MAP_IO;
 	return 0;
diff --git a/drivers/gpu/drm/omapdrm/dss/dpi.c b/drivers/gpu/drm/omapdrm/dss/dpi.c
index b17e77f700dd..6eff97a09160 100644
--- a/drivers/gpu/drm/omapdrm/dss/dpi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dpi.c
@@ -420,6 +420,7 @@ static void dpi_init_pll(struct dpi_data *dpi)
  */
 
 static int dpi_bridge_attach(struct drm_bridge *bridge,
+			     struct drm_encoder *encoder,
 			     enum drm_bridge_attach_flags flags)
 {
 	struct dpi_data *dpi = drm_bridge_to_dpi(bridge);
@@ -429,7 +430,7 @@ static int dpi_bridge_attach(struct drm_bridge *bridge,
 
 	dpi_init_pll(dpi);
 
-	return drm_bridge_attach(bridge->encoder, dpi->output.next_bridge,
+	return drm_bridge_attach(encoder, dpi->output.next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c
index 9b9cc593790c..91ee63bfe0bc 100644
--- a/drivers/gpu/drm/omapdrm/dss/dsi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dsi.c
@@ -4617,6 +4617,7 @@ static const struct component_ops dsi_component_ops = {
  */
 
 static int dsi_bridge_attach(struct drm_bridge *bridge,
+			     struct drm_encoder *encoder,
 			     enum drm_bridge_attach_flags flags)
 {
 	struct dsi_data *dsi = drm_bridge_to_dsi(bridge);
@@ -4624,7 +4625,7 @@ static int dsi_bridge_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	return drm_bridge_attach(bridge->encoder, dsi->output.next_bridge,
+	return drm_bridge_attach(encoder, dsi->output.next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
index e1ac447221ee..a3b22952fdc3 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
@@ -314,6 +314,7 @@ void hdmi4_core_disable(struct hdmi_core_data *core)
  */
 
 static int hdmi4_bridge_attach(struct drm_bridge *bridge,
+			       struct drm_encoder *encoder,
 			       enum drm_bridge_attach_flags flags)
 {
 	struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge);
@@ -321,7 +322,7 @@ static int hdmi4_bridge_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	return drm_bridge_attach(bridge->encoder, hdmi->output.next_bridge,
+	return drm_bridge_attach(encoder, hdmi->output.next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
index fa9904e4c218..0c98444d39a9 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
@@ -312,6 +312,7 @@ static void hdmi_core_disable(struct omap_hdmi *hdmi)
  */
 
 static int hdmi5_bridge_attach(struct drm_bridge *bridge,
+			       struct drm_encoder *encoder,
 			       enum drm_bridge_attach_flags flags)
 {
 	struct omap_hdmi *hdmi = drm_bridge_to_hdmi(bridge);
@@ -319,7 +320,7 @@ static int hdmi5_bridge_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	return drm_bridge_attach(bridge->encoder, hdmi->output.next_bridge,
+	return drm_bridge_attach(encoder, hdmi->output.next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/omapdrm/dss/sdi.c b/drivers/gpu/drm/omapdrm/dss/sdi.c
index f9ae358e8e52..e78826e4b560 100644
--- a/drivers/gpu/drm/omapdrm/dss/sdi.c
+++ b/drivers/gpu/drm/omapdrm/dss/sdi.c
@@ -128,6 +128,7 @@ static void sdi_config_lcd_manager(struct sdi_device *sdi)
  */
 
 static int sdi_bridge_attach(struct drm_bridge *bridge,
+			     struct drm_encoder *encoder,
 			     enum drm_bridge_attach_flags flags)
 {
 	struct sdi_device *sdi = drm_bridge_to_sdi(bridge);
@@ -135,7 +136,7 @@ static int sdi_bridge_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	return drm_bridge_attach(bridge->encoder, sdi->output.next_bridge,
+	return drm_bridge_attach(encoder, sdi->output.next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c
index aaeef603682c..50349518eda1 100644
--- a/drivers/gpu/drm/omapdrm/dss/venc.c
+++ b/drivers/gpu/drm/omapdrm/dss/venc.c
@@ -538,6 +538,7 @@ static int venc_get_clocks(struct venc_device *venc)
  */
 
 static int venc_bridge_attach(struct drm_bridge *bridge,
+			      struct drm_encoder *encoder,
 			      enum drm_bridge_attach_flags flags)
 {
 	struct venc_device *venc = drm_bridge_to_venc(bridge);
@@ -545,7 +546,7 @@ static int venc_bridge_attach(struct drm_bridge *bridge,
 	if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR))
 		return -EINVAL;
 
-	return drm_bridge_attach(bridge->encoder, venc->output.next_bridge,
+	return drm_bridge_attach(encoder, venc->output.next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index e059b06e0239..721581d425b4 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -67,6 +67,15 @@ config DRM_PANEL_BOE_HIMAX8279D
 	  24 bit RGB per pixel. It provides a MIPI DSI interface to
 	  the host and has a built-in LED backlight.
 
+config DRM_PANEL_BOE_TD4320
+	tristate "BOE TD4320 DSI panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for BOE TD4320 1080x2340
+	  video mode panel found in Xiaomi Redmi Note 7 smartphones.
+
 config DRM_PANEL_BOE_TH101MB31UIG002_28A
 	tristate "Boe TH101MB31UIG002-28A panel"
 	depends on OF
@@ -154,6 +163,17 @@ config DRM_PANEL_LVDS
 	  handling of power supplies or control signals. It implements automatic
 	  backlight handling if the panel is attached to a backlight controller.
 
+config DRM_PANEL_HIMAX_HX8279
+	tristate "Himax HX8279-based panels"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y if you want to enable support for panels based on the
+	  Himax HX8279 controller, such as the Startek KD070FHFID078
+	  7.0" 1200x1920 IPS LCD panel that uses a MIPI-DSI interface
+	  and others.
+
 config DRM_PANEL_HIMAX_HX83102
 	tristate "Himax HX83102-based panels"
 	depends on OF
@@ -497,6 +517,16 @@ config DRM_PANEL_NOVATEK_NT36672E
 	  LCD panel module. The panel has a resolution of 1080x2408 and uses 24 bit
 	  RGB per pixel.
 
+config DRM_PANEL_NOVATEK_NT37801
+	tristate "Novatek NT37801/NT37810 AMOLED DSI panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for Novatek NT37801 (or
+	  NT37810) AMOLED DSI Video Mode LCD panel module with 1440x3200
+	  resolution.
+
 config DRM_PANEL_NOVATEK_NT39016
 	tristate "Novatek NT39016 RGB/SPI panel"
 	depends on OF && SPI
@@ -996,6 +1026,15 @@ config DRM_PANEL_TRULY_NT35597_WQXGA
 	  Say Y here if you want to enable support for Truly NT35597 WQXGA Dual DSI
 	  Video Mode panel
 
+config DRM_PANEL_VISIONOX_G2647FB105
+	tristate "Visionox G2647FB105"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for the Visionox
+	  G2647FB105 (2340x1080@60Hz) AMOLED DSI cmd mode panel.
+
 config DRM_PANEL_VISIONOX_R66451
 	tristate "Visionox R66451"
 	depends on OF
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index 1bb8ae46b59b..714cbac830e3 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -5,6 +5,7 @@ obj-$(CONFIG_DRM_PANEL_ASUS_Z00T_TM5P5_NT35596) += panel-asus-z00t-tm5p5-n35596.
 obj-$(CONFIG_DRM_PANEL_AUO_A030JTN01) += panel-auo-a030jtn01.o
 obj-$(CONFIG_DRM_PANEL_BOE_BF060Y8M_AJ0) += panel-boe-bf060y8m-aj0.o
 obj-$(CONFIG_DRM_PANEL_BOE_HIMAX8279D) += panel-boe-himax8279d.o
+obj-$(CONFIG_DRM_PANEL_BOE_TD4320) += panel-boe-td4320.o
 obj-$(CONFIG_DRM_PANEL_BOE_TH101MB31UIG002_28A) += panel-boe-th101mb31ig002-28a.o
 obj-$(CONFIG_DRM_PANEL_BOE_TV101WUM_LL2) += panel-boe-tv101wum-ll2.o
 obj-$(CONFIG_DRM_PANEL_BOE_TV101WUM_NL6) += panel-boe-tv101wum-nl6.o
@@ -16,6 +17,7 @@ obj-$(CONFIG_DRM_PANEL_EBBG_FT8719) += panel-ebbg-ft8719.o
 obj-$(CONFIG_DRM_PANEL_ELIDA_KD35T133) += panel-elida-kd35t133.o
 obj-$(CONFIG_DRM_PANEL_FEIXIN_K101_IM2BA02) += panel-feixin-k101-im2ba02.o
 obj-$(CONFIG_DRM_PANEL_FEIYANG_FY07024DI26A30D) += panel-feiyang-fy07024di26a30d.o
+obj-$(CONFIG_DRM_PANEL_HIMAX_HX8279) += panel-himax-hx8279.o
 obj-$(CONFIG_DRM_PANEL_HIMAX_HX83102) += panel-himax-hx83102.o
 obj-$(CONFIG_DRM_PANEL_HIMAX_HX83112A) += panel-himax-hx83112a.o
 obj-$(CONFIG_DRM_PANEL_HIMAX_HX8394) += panel-himax-hx8394.o
@@ -49,6 +51,7 @@ obj-$(CONFIG_DRM_PANEL_NOVATEK_NT35950) += panel-novatek-nt35950.o
 obj-$(CONFIG_DRM_PANEL_NOVATEK_NT36523) += panel-novatek-nt36523.o
 obj-$(CONFIG_DRM_PANEL_NOVATEK_NT36672A) += panel-novatek-nt36672a.o
 obj-$(CONFIG_DRM_PANEL_NOVATEK_NT36672E) += panel-novatek-nt36672e.o
+obj-$(CONFIG_DRM_PANEL_NOVATEK_NT37801) += panel-novatek-nt37801.o
 obj-$(CONFIG_DRM_PANEL_NOVATEK_NT39016) += panel-novatek-nt39016.o
 obj-$(CONFIG_DRM_PANEL_MANTIX_MLAF057WE51) += panel-mantix-mlaf057we51.o
 obj-$(CONFIG_DRM_PANEL_OLIMEX_LCD_OLINUXINO) += panel-olimex-lcd-olinuxino.o
@@ -101,6 +104,7 @@ obj-$(CONFIG_DRM_PANEL_TPO_TD028TTEC1) += panel-tpo-td028ttec1.o
 obj-$(CONFIG_DRM_PANEL_TPO_TD043MTEA1) += panel-tpo-td043mtea1.o
 obj-$(CONFIG_DRM_PANEL_TPO_TPG110) += panel-tpo-tpg110.o
 obj-$(CONFIG_DRM_PANEL_TRULY_NT35597_WQXGA) += panel-truly-nt35597.o
+obj-$(CONFIG_DRM_PANEL_VISIONOX_G2647FB105) += panel-visionox-g2647fb105.o
 obj-$(CONFIG_DRM_PANEL_VISIONOX_RM69299) += panel-visionox-rm69299.o
 obj-$(CONFIG_DRM_PANEL_VISIONOX_RM692E5) += panel-visionox-rm692e5.o
 obj-$(CONFIG_DRM_PANEL_VISIONOX_VTDR6130) += panel-visionox-vtdr6130.o
diff --git a/drivers/gpu/drm/panel/panel-abt-y030xx067a.c b/drivers/gpu/drm/panel/panel-abt-y030xx067a.c
index 4692c36fe217..87fb0fd29658 100644
--- a/drivers/gpu/drm/panel/panel-abt-y030xx067a.c
+++ b/drivers/gpu/drm/panel/panel-abt-y030xx067a.c
@@ -279,9 +279,10 @@ static int y030xx067a_probe(struct spi_device *spi)
 	struct y030xx067a *priv;
 	int err;
 
-	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
+	priv = devm_drm_panel_alloc(dev, struct y030xx067a, panel,
+				    &y030xx067a_funcs, DRM_MODE_CONNECTOR_DPI);
+	if (IS_ERR(priv))
+		return PTR_ERR(priv);
 
 	priv->spi = spi;
 	spi_set_drvdata(spi, priv);
@@ -306,9 +307,6 @@ static int y030xx067a_probe(struct spi_device *spi)
 		return dev_err_probe(dev, PTR_ERR(priv->reset_gpio),
 				     "Failed to get reset GPIO\n");
 
-	drm_panel_init(&priv->panel, dev, &y030xx067a_funcs,
-		       DRM_MODE_CONNECTOR_DPI);
-
 	err = drm_panel_of_backlight(&priv->panel);
 	if (err)
 		return err;
diff --git a/drivers/gpu/drm/panel/panel-arm-versatile.c b/drivers/gpu/drm/panel/panel-arm-versatile.c
index 503ecea72c5e..ea5119018df4 100644
--- a/drivers/gpu/drm/panel/panel-arm-versatile.c
+++ b/drivers/gpu/drm/panel/panel-arm-versatile.c
@@ -306,9 +306,11 @@ static int versatile_panel_probe(struct platform_device *pdev)
 		return PTR_ERR(map);
 	}
 
-	vpanel = devm_kzalloc(dev, sizeof(*vpanel), GFP_KERNEL);
-	if (!vpanel)
-		return -ENOMEM;
+	vpanel = devm_drm_panel_alloc(dev, struct versatile_panel, panel,
+				      &versatile_panel_drm_funcs,
+				      DRM_MODE_CONNECTOR_DPI);
+	if (IS_ERR(vpanel))
+		return PTR_ERR(vpanel);
 
 	ret = regmap_read(map, SYS_CLCD, &val);
 	if (ret) {
@@ -348,9 +350,6 @@ static int versatile_panel_probe(struct platform_device *pdev)
 			dev_info(dev, "panel mounted on IB2 daughterboard\n");
 	}
 
-	drm_panel_init(&vpanel->panel, dev, &versatile_panel_drm_funcs,
-		       DRM_MODE_CONNECTOR_DPI);
-
 	drm_panel_add(&vpanel->panel);
 
 	return 0;
diff --git a/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c b/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c
index b05a663c134c..db006576d704 100644
--- a/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c
+++ b/drivers/gpu/drm/panel/panel-asus-z00t-tm5p5-n35596.c
@@ -224,9 +224,11 @@ static int tm5p5_nt35596_probe(struct mipi_dsi_device *dsi)
 	struct tm5p5_nt35596 *ctx;
 	int ret;
 
-	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	ctx = devm_drm_panel_alloc(dev, struct tm5p5_nt35596, panel,
+				   &tm5p5_nt35596_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
 
 	ctx->supplies[0].supply = "vdd";
 	ctx->supplies[1].supply = "vddio";
@@ -253,9 +255,6 @@ static int tm5p5_nt35596_probe(struct mipi_dsi_device *dsi)
 			  MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_NO_EOT_PACKET |
 			  MIPI_DSI_CLOCK_NON_CONTINUOUS | MIPI_DSI_MODE_LPM;
 
-	drm_panel_init(&ctx->panel, dev, &tm5p5_nt35596_panel_funcs,
-		       DRM_MODE_CONNECTOR_DSI);
-
 	ctx->panel.backlight = tm5p5_nt35596_create_backlight(dsi);
 	if (IS_ERR(ctx->panel.backlight)) {
 		ret = PTR_ERR(ctx->panel.backlight);
diff --git a/drivers/gpu/drm/panel/panel-auo-a030jtn01.c b/drivers/gpu/drm/panel/panel-auo-a030jtn01.c
index 77604d6a4e72..6e52bf6830e1 100644
--- a/drivers/gpu/drm/panel/panel-auo-a030jtn01.c
+++ b/drivers/gpu/drm/panel/panel-auo-a030jtn01.c
@@ -200,9 +200,10 @@ static int a030jtn01_probe(struct spi_device *spi)
 
 	spi->mode |= SPI_MODE_3 | SPI_3WIRE;
 
-	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
+	priv = devm_drm_panel_alloc(dev, struct a030jtn01, panel,
+				    &a030jtn01_funcs, DRM_MODE_CONNECTOR_DPI);
+	if (IS_ERR(priv))
+		return PTR_ERR(priv);
 
 	priv->spi = spi;
 	spi_set_drvdata(spi, priv);
@@ -223,9 +224,6 @@ static int a030jtn01_probe(struct spi_device *spi)
 	if (IS_ERR(priv->reset_gpio))
 		return dev_err_probe(dev, PTR_ERR(priv->reset_gpio), "Failed to get reset GPIO");
 
-	drm_panel_init(&priv->panel, dev, &a030jtn01_funcs,
-		       DRM_MODE_CONNECTOR_DPI);
-
 	err = drm_panel_of_backlight(&priv->panel);
 	if (err)
 		return err;
diff --git a/drivers/gpu/drm/panel/panel-boe-bf060y8m-aj0.c b/drivers/gpu/drm/panel/panel-boe-bf060y8m-aj0.c
index 7e66db4a88bb..84c21c62a43e 100644
--- a/drivers/gpu/drm/panel/panel-boe-bf060y8m-aj0.c
+++ b/drivers/gpu/drm/panel/panel-boe-bf060y8m-aj0.c
@@ -55,77 +55,56 @@ static void boe_bf060y8m_aj0_reset(struct boe_bf060y8m_aj0 *boe)
 static int boe_bf060y8m_aj0_on(struct boe_bf060y8m_aj0 *boe)
 {
 	struct mipi_dsi_device *dsi = boe->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
-
-	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0xa5, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0xb2, 0x00, 0x4c);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_3D_CONTROL, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_POWER_SAVE, DCS_ALLOW_HBM_RANGE);
-	mipi_dsi_dcs_write_seq(dsi, 0xf8,
-			       0x00, 0x08, 0x10, 0x00, 0x22, 0x00, 0x00, 0x2d);
-
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
-		return ret;
-	}
-	msleep(30);
-
-	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0xa5, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0xc0,
-			       0x08, 0x48, 0x65, 0x33, 0x33, 0x33,
-			       0x2a, 0x31, 0x39, 0x20, 0x09);
-	mipi_dsi_dcs_write_seq(dsi, 0xc1, 0x00, 0x00, 0x00, 0x1f, 0x1f,
-			       0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f,
-			       0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f);
-	mipi_dsi_dcs_write_seq(dsi, 0xe2, 0x20, 0x04, 0x10, 0x12, 0x92,
-			       0x4f, 0x8f, 0x44, 0x84, 0x83, 0x83, 0x83,
-			       0x5c, 0x5c, 0x5c);
-	mipi_dsi_dcs_write_seq(dsi, 0xde, 0x01, 0x2c, 0x00, 0x77, 0x3e);
-
-	msleep(30);
-
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display on: %d\n", ret);
-		return ret;
-	}
-	msleep(50);
-
-	return 0;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb0, 0xa5, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb2, 0x00, 0x4c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_3D_CONTROL, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_POWER_SAVE, DCS_ALLOW_HBM_RANGE);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf8,
+				     0x00, 0x08, 0x10, 0x00, 0x22, 0x00, 0x00, 0x2d);
+
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 30);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb0, 0xa5, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc0,
+				     0x08, 0x48, 0x65, 0x33, 0x33, 0x33,
+				     0x2a, 0x31, 0x39, 0x20, 0x09);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc1, 0x00, 0x00, 0x00, 0x1f, 0x1f,
+				     0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f,
+				     0x1f, 0x1f, 0x1f, 0x1f, 0x1f, 0x1f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe2, 0x20, 0x04, 0x10, 0x12, 0x92,
+				     0x4f, 0x8f, 0x44, 0x84, 0x83, 0x83, 0x83,
+				     0x5c, 0x5c, 0x5c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xde, 0x01, 0x2c, 0x00, 0x77, 0x3e);
+
+	mipi_dsi_msleep(&dsi_ctx, 30);
+
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 50);
+
+	return dsi_ctx.accum_err;
 }
 
-static int boe_bf060y8m_aj0_off(struct boe_bf060y8m_aj0 *boe)
+static void boe_bf060y8m_aj0_off(struct boe_bf060y8m_aj0 *boe)
 {
 	struct mipi_dsi_device *dsi = boe->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
 	/* OFF commands sent in HS mode */
 	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
-	ret = mipi_dsi_dcs_set_display_off(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display off: %d\n", ret);
-		return ret;
-	}
-	msleep(20);
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 20);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to enter sleep mode: %d\n", ret);
-		return ret;
-	}
-	usleep_range(1000, 2000);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_usleep_range(&dsi_ctx, 1000, 2000);
 	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
-
-	return 0;
 }
 
 static int boe_bf060y8m_aj0_prepare(struct drm_panel *panel)
 {
 	struct boe_bf060y8m_aj0 *boe = to_boe_bf060y8m_aj0(panel);
-	struct device *dev = &boe->dsi->dev;
 	int ret;
 
 	/*
@@ -157,13 +136,14 @@ static int boe_bf060y8m_aj0_prepare(struct drm_panel *panel)
 
 	ret = boe_bf060y8m_aj0_on(boe);
 	if (ret < 0) {
-		dev_err(dev, "Failed to initialize panel: %d\n", ret);
 		gpiod_set_value_cansleep(boe->reset_gpio, 1);
-		return ret;
+		goto err_on;
 	}
 
 	return 0;
 
+err_on:
+	regulator_disable(boe->vregs[BF060Y8M_VREG_VCI].consumer);
 err_vci:
 	regulator_disable(boe->vregs[BF060Y8M_VREG_VDDIO].consumer);
 err_vddio:
@@ -178,15 +158,11 @@ err_elvss:
 static int boe_bf060y8m_aj0_unprepare(struct drm_panel *panel)
 {
 	struct boe_bf060y8m_aj0 *boe = to_boe_bf060y8m_aj0(panel);
-	struct device *dev = &boe->dsi->dev;
-	int ret;
 
-	ret = boe_bf060y8m_aj0_off(boe);
-	if (ret < 0)
-		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
+	boe_bf060y8m_aj0_off(boe);
 
 	gpiod_set_value_cansleep(boe->reset_gpio, 1);
-	ret = regulator_bulk_disable(ARRAY_SIZE(boe->vregs), boe->vregs);
+	regulator_bulk_disable(ARRAY_SIZE(boe->vregs), boe->vregs);
 
 	return 0;
 }
@@ -234,13 +210,11 @@ static int boe_bf060y8m_aj0_bl_update_status(struct backlight_device *bl)
 {
 	struct mipi_dsi_device *dsi = bl_get_data(bl);
 	u16 brightness = backlight_get_brightness(bl);
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
-	ret = mipi_dsi_dcs_set_display_brightness(dsi, brightness);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, brightness);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int boe_bf060y8m_aj0_bl_get_brightness(struct backlight_device *bl)
@@ -350,9 +324,11 @@ static int boe_bf060y8m_aj0_probe(struct mipi_dsi_device *dsi)
 	struct boe_bf060y8m_aj0 *boe;
 	int ret;
 
-	boe = devm_kzalloc(dev, sizeof(*boe), GFP_KERNEL);
-	if (!boe)
-		return -ENOMEM;
+	boe = devm_drm_panel_alloc(dev, struct boe_bf060y8m_aj0, panel,
+				   &boe_bf060y8m_aj0_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(boe))
+		return PTR_ERR(boe);
 
 	ret = boe_bf060y8m_aj0_init_vregs(boe, dev);
 	if (ret)
@@ -374,9 +350,6 @@ static int boe_bf060y8m_aj0_probe(struct mipi_dsi_device *dsi)
 			  MIPI_DSI_CLOCK_NON_CONTINUOUS |
 			  MIPI_DSI_MODE_LPM;
 
-	drm_panel_init(&boe->panel, dev, &boe_bf060y8m_aj0_panel_funcs,
-		       DRM_MODE_CONNECTOR_DSI);
-
 	boe->panel.prepare_prev_first = true;
 
 	boe->panel.backlight = boe_bf060y8m_aj0_create_backlight(dsi);
diff --git a/drivers/gpu/drm/panel/panel-boe-td4320.c b/drivers/gpu/drm/panel/panel-boe-td4320.c
new file mode 100644
index 000000000000..1956daa2c71b
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-boe-td4320.c
@@ -0,0 +1,247 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright (c) 2024 Barnabas Czeman <barnabas.czeman@mainlining.org>
+// Generated with linux-mdss-dsi-panel-driver-generator from vendor device tree:
+//   Copyright (c) 2013, The Linux Foundation. All rights reserved.
+
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/regulator/consumer.h>
+
+#include <video/mipi_display.h>
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
+
+struct boe_td4320 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct regulator_bulk_data *supplies;
+	struct gpio_desc *reset_gpio;
+};
+
+static const struct regulator_bulk_data boe_td4320_supplies[] = {
+	{ .supply = "iovcc" },
+	{ .supply = "vsn" },
+	{ .supply = "vsp" },
+};
+
+static inline struct boe_td4320 *to_boe_td4320(struct drm_panel *panel)
+{
+	return container_of(panel, struct boe_td4320, panel);
+}
+
+static void boe_td4320_reset(struct boe_td4320 *ctx)
+{
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	usleep_range(1000, 2000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(5000, 6000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	msleep(30);
+}
+
+static int boe_td4320_on(struct boe_td4320 *ctx)
+{
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi };
+
+	ctx->dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xb0, 0x04);
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xd6, 0x00);
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xb8,
+					 0x19, 0x55, 0x00, 0xbe, 0x00, 0x00,
+					 0x00);
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xb9,
+					 0x4d, 0x55, 0x05, 0xe6, 0x00, 0x02,
+					 0x03);
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xba,
+					 0x9b, 0x5b, 0x07, 0xe6, 0x00, 0x13,
+					 0x00);
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xf9,
+					 0x44, 0x3f, 0x00, 0x8d, 0xbf);
+	mipi_dsi_generic_write_seq_multi(&dsi_ctx, 0xce,
+					 0x5d, 0x00, 0x0f, 0x1f, 0x2f, 0x3f,
+					 0x4f, 0x5f, 0x6f, 0x7f, 0x8f, 0x9f,
+					 0xaf, 0xbf, 0xcf, 0xdf, 0xef, 0xff,
+					 0x04, 0x00, 0x02, 0x02, 0x42, 0x01,
+					 0x69, 0x5a, 0x40, 0x40, 0x00, 0x00,
+					 0x04, 0xfa, 0x00);
+	mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, 0x00b8);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY,
+				     0x2c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_POWER_SAVE, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb0, 0x03);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x11, 0x00);
+	mipi_dsi_msleep(&dsi_ctx, 96);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x29, 0x00);
+	mipi_dsi_msleep(&dsi_ctx, 20);
+
+	return dsi_ctx.accum_err;
+}
+
+static int boe_td4320_off(struct boe_td4320 *ctx)
+{
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi };
+
+	ctx->dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 20);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 120);
+
+	return dsi_ctx.accum_err;
+}
+
+static int boe_td4320_prepare(struct drm_panel *panel)
+{
+	struct boe_td4320 *ctx = to_boe_td4320(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(boe_td4320_supplies), ctx->supplies);
+	if (ret < 0) {
+		dev_err(dev, "Failed to enable regulators: %d\n", ret);
+		return ret;
+	}
+
+	boe_td4320_reset(ctx);
+
+	ret = boe_td4320_on(ctx);
+	if (ret < 0) {
+		dev_err(dev, "Failed to initialize panel: %d\n", ret);
+		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+		regulator_bulk_disable(ARRAY_SIZE(boe_td4320_supplies), ctx->supplies);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int boe_td4320_unprepare(struct drm_panel *panel)
+{
+	struct boe_td4320 *ctx = to_boe_td4320(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = boe_td4320_off(ctx);
+	if (ret < 0)
+		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	regulator_bulk_disable(ARRAY_SIZE(boe_td4320_supplies), ctx->supplies);
+
+	return 0;
+}
+
+static const struct drm_display_mode boe_td4320_mode = {
+	.clock = (1080 + 86 + 2 + 100) * (2340 + 4 + 4 + 60) * 60 / 1000,
+	.hdisplay = 1080,
+	.hsync_start = 1080 + 86,
+	.hsync_end = 1080 + 86 + 2,
+	.htotal = 1080 + 86 + 2 + 100,
+	.vdisplay = 2340,
+	.vsync_start = 2340 + 4,
+	.vsync_end = 2340 + 4 + 4,
+	.vtotal = 2340 + 4 + 4 + 60,
+	.width_mm = 67,
+	.height_mm = 145,
+	.type = DRM_MODE_TYPE_DRIVER,
+};
+
+static int boe_td4320_get_modes(struct drm_panel *panel,
+				struct drm_connector *connector)
+{
+	return drm_connector_helper_get_modes_fixed(connector, &boe_td4320_mode);
+}
+
+static const struct drm_panel_funcs boe_td4320_panel_funcs = {
+	.prepare = boe_td4320_prepare,
+	.unprepare = boe_td4320_unprepare,
+	.get_modes = boe_td4320_get_modes,
+};
+
+static int boe_td4320_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct boe_td4320 *ctx;
+	int ret;
+
+	ctx = devm_drm_panel_alloc(dev, struct boe_td4320, panel,
+				   &boe_td4320_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
+
+	ret = devm_regulator_bulk_get_const(dev,
+					    ARRAY_SIZE(boe_td4320_supplies),
+					    boe_td4320_supplies,
+					    &ctx->supplies);
+	if (ret < 0)
+		return ret;
+
+	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(ctx->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio),
+				     "Failed to get reset-gpios\n");
+
+	ctx->dsi = dsi;
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
+			  MIPI_DSI_CLOCK_NON_CONTINUOUS;
+
+	ctx->panel.prepare_prev_first = true;
+
+	ret = drm_panel_of_backlight(&ctx->panel);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get backlight\n");
+
+	drm_panel_add(&ctx->panel);
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		drm_panel_remove(&ctx->panel);
+		return dev_err_probe(dev, ret, "Failed to attach to DSI host\n");
+	}
+
+	return 0;
+}
+
+static void boe_td4320_remove(struct mipi_dsi_device *dsi)
+{
+	struct boe_td4320 *ctx = mipi_dsi_get_drvdata(dsi);
+	int ret;
+
+	ret = mipi_dsi_detach(dsi);
+	if (ret < 0)
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
+
+	drm_panel_remove(&ctx->panel);
+}
+
+static const struct of_device_id boe_td4320_of_match[] = {
+	{ .compatible = "boe,td4320" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, boe_td4320_of_match);
+
+static struct mipi_dsi_driver boe_td4320_driver = {
+	.probe = boe_td4320_probe,
+	.remove = boe_td4320_remove,
+	.driver = {
+		.name = "panel-boe-td4320",
+		.of_match_table = boe_td4320_of_match,
+	},
+};
+module_mipi_dsi_driver(boe_td4320_driver);
+
+MODULE_AUTHOR("Barnabas Czeman <barnabas.czeman@mainlining.org>");
+MODULE_DESCRIPTION("DRM driver for boe td4320 fhdplus video mode dsi panel");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/panel/panel-boe-th101mb31ig002-28a.c b/drivers/gpu/drm/panel/panel-boe-th101mb31ig002-28a.c
index 0b87f1e6ecae..f33d4f855929 100644
--- a/drivers/gpu/drm/panel/panel-boe-th101mb31ig002-28a.c
+++ b/drivers/gpu/drm/panel/panel-boe-th101mb31ig002-28a.c
@@ -349,9 +349,11 @@ static int boe_th101mb31ig002_dsi_probe(struct mipi_dsi_device *dsi)
 	const struct panel_desc *desc;
 	int ret;
 
-	ctx = devm_kzalloc(&dsi->dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	ctx = devm_drm_panel_alloc(&dsi->dev, struct boe_th101mb31ig002, panel,
+				   &boe_th101mb31ig002_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
 
 	mipi_dsi_set_drvdata(dsi, ctx);
 	ctx->dsi = dsi;
@@ -383,9 +385,6 @@ static int boe_th101mb31ig002_dsi_probe(struct mipi_dsi_device *dsi)
 		return dev_err_probe(&dsi->dev, ret,
 				     "Failed to get orientation\n");
 
-	drm_panel_init(&ctx->panel, &dsi->dev, &boe_th101mb31ig002_funcs,
-		       DRM_MODE_CONNECTOR_DSI);
-
 	ret = drm_panel_of_backlight(&ctx->panel);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-ll2.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-ll2.c
index 50e4a5341bc6..20b6e11a7d84 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-ll2.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-ll2.c
@@ -166,9 +166,11 @@ static int boe_tv101wum_ll2_probe(struct mipi_dsi_device *dsi)
 	struct boe_tv101wum_ll2 *ctx;
 	int ret;
 
-	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	ctx = devm_drm_panel_alloc(dev, struct boe_tv101wum_ll2, panel,
+				   &boe_tv101wum_ll2_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
 
 	ret = devm_regulator_bulk_get_const(&dsi->dev,
 					ARRAY_SIZE(boe_tv101wum_ll2_supplies),
@@ -190,8 +192,6 @@ static int boe_tv101wum_ll2_probe(struct mipi_dsi_device *dsi)
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
 			  MIPI_DSI_MODE_VIDEO_HSE;
 
-	drm_panel_init(&ctx->panel, dev, &boe_tv101wum_ll2_panel_funcs,
-		       DRM_MODE_CONNECTOR_DSI);
 	ctx->panel.prepare_prev_first = true;
 
 	ret = drm_panel_of_backlight(&ctx->panel);
diff --git a/drivers/gpu/drm/panel/panel-dsi-cm.c b/drivers/gpu/drm/panel/panel-dsi-cm.c
index 6b3f4d664d2a..ae6e9ffc46cb 100644
--- a/drivers/gpu/drm/panel/panel-dsi-cm.c
+++ b/drivers/gpu/drm/panel/panel-dsi-cm.c
@@ -511,9 +511,10 @@ static int dsicm_probe(struct mipi_dsi_device *dsi)
 
 	dev_dbg(dev, "probe\n");
 
-	ddata = devm_kzalloc(dev, sizeof(*ddata), GFP_KERNEL);
-	if (!ddata)
-		return -ENOMEM;
+	ddata = devm_drm_panel_alloc(dev, struct panel_drv_data, panel,
+				     &dsicm_panel_funcs, DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ddata))
+		return PTR_ERR(ddata);
 
 	mipi_dsi_set_drvdata(dsi, ddata);
 	ddata->dsi = dsi;
@@ -530,9 +531,6 @@ static int dsicm_probe(struct mipi_dsi_device *dsi)
 
 	dsicm_hw_reset(ddata);
 
-	drm_panel_init(&ddata->panel, dev, &dsicm_panel_funcs,
-		       DRM_MODE_CONNECTOR_DSI);
-
 	if (ddata->use_dsi_backlight) {
 		struct backlight_properties props = { 0 };
 		props.max_brightness = 255;
diff --git a/drivers/gpu/drm/panel/panel-ebbg-ft8719.c b/drivers/gpu/drm/panel/panel-ebbg-ft8719.c
index 0bfed0ec0bbc..fb9f9f42be4f 100644
--- a/drivers/gpu/drm/panel/panel-ebbg-ft8719.c
+++ b/drivers/gpu/drm/panel/panel-ebbg-ft8719.c
@@ -163,9 +163,11 @@ static int ebbg_ft8719_probe(struct mipi_dsi_device *dsi)
 	struct ebbg_ft8719 *ctx;
 	int i, ret;
 
-	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	ctx = devm_drm_panel_alloc(dev, struct ebbg_ft8719, panel,
+				   &ebbg_ft8719_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
 
 	for (i = 0; i < ARRAY_SIZE(ctx->supplies); i++)
 		ctx->supplies[i].supply = regulator_names[i];
@@ -196,9 +198,6 @@ static int ebbg_ft8719_probe(struct mipi_dsi_device *dsi)
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
 			  MIPI_DSI_CLOCK_NON_CONTINUOUS;
 
-	drm_panel_init(&ctx->panel, dev, &ebbg_ft8719_panel_funcs,
-		       DRM_MODE_CONNECTOR_DSI);
-
 	ret = drm_panel_of_backlight(&ctx->panel);
 	if (ret)
 		return dev_err_probe(dev, ret, "Failed to get backlight\n");
diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index 52028c8f8988..90e8c154a978 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -839,9 +839,10 @@ static int panel_edp_probe(struct device *dev, const struct panel_desc *desc,
 	struct device_node *ddc;
 	int err;
 
-	panel = devm_kzalloc(dev, sizeof(*panel), GFP_KERNEL);
-	if (!panel)
-		return -ENOMEM;
+	panel = devm_drm_panel_alloc(dev, struct panel_edp, base,
+				     &panel_edp_funcs, DRM_MODE_CONNECTOR_eDP);
+	if (IS_ERR(panel))
+		return PTR_ERR(panel);
 
 	panel->prepared_time = 0;
 	panel->desc = desc;
@@ -886,8 +887,6 @@ static int panel_edp_probe(struct device *dev, const struct panel_desc *desc,
 
 	dev_set_drvdata(dev, panel);
 
-	drm_panel_init(&panel->base, dev, &panel_edp_funcs, DRM_MODE_CONNECTOR_eDP);
-
 	err = drm_panel_of_backlight(&panel->base);
 	if (err)
 		goto err_finished_ddc_init;
@@ -1763,6 +1762,13 @@ static const struct panel_delay delay_80_500_e50 = {
 	.enable = 50,
 };
 
+static const struct panel_delay delay_80_500_e80_p2e200 = {
+	.hpd_absent = 80,
+	.unprepare = 500,
+	.enable = 80,
+	.prepare_to_enable = 200,
+};
+
 static const struct panel_delay delay_100_500_e200 = {
 	.hpd_absent = 100,
 	.unprepare = 500,
@@ -1878,6 +1884,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xa199, &delay_200_500_e50, "B116XAN06.1"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xa7b3, &delay_200_500_e50, "B140UAN04.4"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xc4b4, &delay_200_500_e50, "B116XAT04.1"),
+	EDP_PANEL_ENTRY('A', 'U', 'O', 0xc9a8, &delay_200_500_e50, "B140QAN08.H"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xd497, &delay_200_500_e50, "B120XAN01.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xf390, &delay_200_500_e50, "B140XTN07.7"),
 
@@ -1938,6 +1945,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0c93, &delay_200_500_e200, "Unknown"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cb6, &delay_200_500_e200, "NT116WHM-N44"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cfa, &delay_200_500_e50, "NV116WHM-A4D"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0d73, &delay_200_500_e80, "NE140WUM-N6S"),
 
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1130, &delay_200_500_e50, "N116BGE-EB2"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1132, &delay_200_500_e80_d50, "N116BGE-EA2"),
@@ -1973,6 +1981,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1103, &delay_200_500_e80_d50, "MNB601LS1-3"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50, "MNB601LS1-4"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"),
+	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1457, &delay_80_500_e80_p2e200, "MNE007QS3-8"),
 
 	EDP_PANEL_ENTRY('E', 'T', 'C', 0x0000, &delay_50_500_e200_d200_po2e335, "LP079QX1-SP0V"),
 
diff --git a/drivers/gpu/drm/panel/panel-himax-hx8279.c b/drivers/gpu/drm/panel/panel-himax-hx8279.c
new file mode 100644
index 000000000000..fb302d1f91b9
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-himax-hx8279.c
@@ -0,0 +1,1296 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Himax HX8279 DriverIC panels driver
+ *
+ * Copyright (c) 2025 Collabora Ltd.
+ *                    AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
+ */
+
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_graph.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/drm_connector.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+
+/* Page selection */
+#define HX8279_REG_PAGE			0xb0
+ #define HX8279_PAGE_SEL		GENMASK(3, 0)
+
+/* Page 0 - Driver/Module Configuration */
+#define HX8279_P0_VGHS			0xbf
+#define HX8279_P0_VGLS			0xc0
+#define HX8279_P0_VGPHS			0xc2
+#define HX8279_P0_VGNHS			0xc4
+ #define HX8279_P0_VG_SEL		GENMASK(4, 0)
+ #define HX8279_VGH_MIN_MV		8700
+ #define HX8279_VGH_STEP_MV		300
+ #define HX8279_VGL_MIN_MV		6700
+ #define HX8279_VGL_STEP_MV		300
+ #define HX8279_VGPNH_MIN_MV		4000
+ #define HX8279_VGPNX_STEP_MV		50
+ #define HX8279_VGH_VOLT_SEL(x)		((x - HX8279_VGH_MIN_MV) / HX8279_VGH_STEP_MV)
+ #define HX8279_VGL_VOLT_SEL(x)		((x - HX8279_VGL_MIN_MV) / HX8279_VGL_STEP_MV)
+ #define HX8279_VGPN_VOLT_SEL(x)	((x - HX8279_VGPNH_MIN_MV) / HX8279_VGPNX_STEP_MV)
+
+/* Page 1 - Gate driver On Array (GOA) Mux */
+#define HX8279_P1_REG_GOA_L		0xc0
+#define HX8279_P1_REG_GOUTL(x)		(HX8279_P1_REG_GOA_L + (x))
+#define HX8279_P1_REG_GOA_R		0xd4
+#define HX8279_P1_REG_GOUTR(x)		(HX8279_P1_REG_GOA_R + (x))
+ #define HX8279_GOUT_STB		GENMASK(7, 6)
+ #define HX8279_GOUT_SEL		GENMASK(5, 0)
+
+/* Page 2 - Analog Gamma Configuration */
+#define HX8279_P2_REG_ANALOG_GAMMA	0xc0
+ #define HX8279_P2_REG_GAMMA_T_PVP(x)	(HX8279_P2_REG_ANALOG_GAMMA + (x))	/* 0..16 */
+ #define HX8279_P2_REG_GAMMA_T_PVN(x)	(HX8279_P2_REG_GAMMA_T_PVP(17) + (x))	/* 0..16 */
+
+/* Page 3 - Gate driver On Array (GOA) Configuration */
+#define HX8279_P3_REG_UNKNOWN_BA	0xba
+#define HX8279_P3_REG_GOA_CKV_FALL_PREC	0xbc
+#define HX8279_P3_REG_GOA_TIMING_ODD	0xc2
+ #define HX8279_P3_REG_GOA_TO(x)	(HX8279_P3_REG_GOA_TIMING_ODD + x) /* GOA_T0..5 */
+#define HX8279_P3_REG_GOA_STVL		0xc8
+ #define HX8279_P3_GOA_STV_LEAD		GENMASK(4, 0)
+#define HX8279_P3_REG_GOA_CKVL		0xc9
+ #define HX8279_P3_GOA_CKV_LEAD		GENMASK(4, 0)
+#define HX8279_P3_REG_GOA_CKVD		0xca
+ #define HX8279_P3_GOA_CKV_NONOVERLAP	BIT(7)
+ #define HX8279_P3_GOA_CKV_RESERVED	BIT(6)
+ #define HX8279_P3_GOA_CKV_DUMMY	GENMASK(5, 0)
+#define HX8279_P3_REG_GOA_CKV_RISE_PREC	0xcb
+#define HX8279_P3_REG_GOA_CLR1_W_ADJ	0xd2
+#define HX8279_P3_REG_GOA_CLR234_W_ADJ	0xd3
+#define HX8279_P3_REG_GOA_CLR1_CFG	0xd4
+#define HX8279_P3_REG_GOA_CLR_CFG(x)	(HX8279_P3_REG_GOA_CLR1_CFG + (x)) /* CLR1..4 */
+ #define HX8279_P3_GOA_CLR_CFG_POLARITY	BIT(7)
+ #define HX8279_P3_GOA_CLR_CFG_STARTPOS	GENMASK(6, 0)
+#define HX8279_P3_REG_GOA_TIMING_EVEN	0xdd
+ #define HX8279_P3_REG_GOA_TE(x)	(HX8279_P3_REG_GOA_TIMING_EVEN + x)
+#define HX8279_P3_REG_UNKNOWN_E4	0xe4
+#define HX8279_P3_REG_UNKNOWN_E5	0xe5
+
+/* Page 5 - MIPI */
+#define HX8279_P5_REG_TIMING		0xb3
+ #define HX8279_P5_TIMING_THS_SETTLE	GENMASK(7, 5)
+ #define HX8279_P5_TIMING_LHS_SETTLE	BIT(4)
+ #define HX8279_P5_TIMING_TLPX		GENMASK(3, 0)
+#define HX8279_P5_REG_UNKNOWN_B8	0xb8
+#define HX8279_P5_REG_UNKNOWN_BC	0xbc
+#define HX8279_P5_REG_UNKNOWN_D6	0xd6
+
+/* Page 6 - Engineer */
+#define HX8279_P6_REG_ENGINEER_PWD	0xb8
+#define HX8279_P6_REG_INHOUSE_FUNC	0xc0
+ #define HX8279_P6_ENG_UNLOCK_WORD	0xa5
+#define HX8279_P6_REG_GAMMA_CHOPPER	0xbc
+ #define HX8279_P6_GAMMA_POCGM_CTL	GENMASK(6, 4)
+ #define HX8279_P6_GAMMA_POGCMD_CTL	GENMASK(2, 0)
+#define HX8279_P6_REG_VOLT_ADJ		0xc7
+ /* For VCCIFS and VCCS - 0: 1450, 1: 1500, 2: 1550, 3: 1600 uV */
+ #define HX8279_P6_VOLT_ADJ_VCCIFS	GENMASK(3, 2)
+ #define HX8279_P6_VOLT_ADJ_VCCS	GENMASK(1, 0)
+#define HX8279_P6_REG_DLY_TIME_ADJ	0xd5
+
+/* Page 7...12 - Digital Gamma Adjustment */
+#define HX8279_PG_DIGITAL_GAMMA		0xb1
+#define HX8279_DGAMMA_DGMA1_HI		GENMASK(7, 6)
+#define HX8279_DGAMMA_DGMA2_HI		GENMASK(5, 4)
+#define HX8279_DGAMMA_DGMA3_HI		GENMASK(3, 2)
+#define HX8279_DGAMMA_DGMA4_HI		GENMASK(1, 0)
+#define HX8279_PG_DGAMMA_NUM_LO_BYTES	24
+#define HX8279_PG_DGAMMA_NUM_HI_BYTES	6
+
+struct hx8279 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi[2];
+	struct regulator_bulk_data vregs[2];
+	struct gpio_desc *enable_gpio;
+	struct gpio_desc *reset_gpio;
+	const struct hx8279_panel_desc *desc;
+	u8 last_page;
+	bool skip_voltage_config;
+	bool skip_goa_config;
+	bool skip_goa_timing;
+	bool skip_goa_even_timing;
+	bool skip_mipi_timing;
+};
+
+struct hx8279_panel_mode {
+	const struct drm_display_mode mode;
+	u8 bpc;
+	bool is_video_mode;
+};
+
+/**
+ * struct hx8279_goa_mux - Gate driver On Array Muxer
+ * @gout_l: Mux GOA signal to GOUT Left pin
+ * @gout_r: Mux GOA signal to GOUT Right pin
+ */
+struct hx8279_goa_mux {
+	u8 gout_l[20];
+	u8 gout_r[20];
+};
+
+/**
+ * struct hx8279_analog_gamma - Analog Gamma Adjustment
+ * @pos: Positive gamma op's input voltage, adjusted by VGP(H/L)
+ * @neg: Negative gamma op's input voltage, adjusted by VGN(H/L)
+ *
+ * Analog Gamma correction is performed with 17+17 reference voltages,
+ * changed with resistor streams, and defined with 17 register values
+ * for positive and 17 for negative.
+ *
+ * Each register holds resistance values, in 8.5ohms per unit, for the
+ * following gamma levels:
+ * 0, 8, 16, 28, 40, 56, 80, 128, 176, 200, 216, 228, 240, 248, 252, 255.
+ */
+struct hx8279_analog_gamma {
+	u8 pos[17];
+	u8 neg[17];
+};
+
+/**
+ * struct hx8279_digital_gamma - Digital Gamma Adjustment
+ * @r: Adjustment for red component
+ * @g: Adjustment for green component
+ * @b: Adjustment for blue component
+ *
+ * The layout of this structure follows the register layout to simplify
+ * both the handling and the declaration of those values in the driver.
+ * Gamma correction is internally done with a 24 segment piecewise
+ * linear interpolation; those segments are defined with 24 ten bits
+ * values of which:
+ *   - The LOW eight bits for the first 24 registers start at the first
+ *     register (at 0xb1) of the Digital Gamma Adjustment page;
+ *   - The HIGH two bits for each of the 24 registers are contained
+ *     in the last six registers;
+ *   - The last six registers contain four groups of two-bits HI values
+ *     for each of the first 24 registers, but in an inverted fashion,
+ *     this means that the first two bits relate to the last register
+ *     of a set of four.
+ *
+ * The 24 segments refer to the following gamma levels:
+ * 0, 1, 3, 7, 11, 15, 23, 31, 47, 63, 95, 127, 128, 160,
+ * 192, 208, 224, 232, 240, 244, 248, 252, 254, 255
+ */
+struct hx8279_digital_gamma {
+	u8 r[HX8279_PG_DGAMMA_NUM_LO_BYTES + HX8279_PG_DGAMMA_NUM_HI_BYTES];
+	u8 g[HX8279_PG_DGAMMA_NUM_LO_BYTES + HX8279_PG_DGAMMA_NUM_HI_BYTES];
+	u8 b[HX8279_PG_DGAMMA_NUM_LO_BYTES + HX8279_PG_DGAMMA_NUM_HI_BYTES];
+};
+
+struct hx8279_panel_desc {
+	const struct mipi_dsi_device_info dsi_info;
+	const struct hx8279_panel_mode *mode_data;
+	u8 num_lanes;
+	u8 num_modes;
+
+	/* Page 0 */
+	unsigned int vgh_mv;
+	unsigned int vgl_mv;
+	unsigned int vgph_mv;
+	unsigned int vgnh_mv;
+
+	/* Page 1 */
+	const struct hx8279_goa_mux *gmux;
+
+	/* Page 2 */
+	const struct hx8279_analog_gamma *agamma;
+
+	/* Page 3 */
+	u8 goa_unk_ba;
+	u8 goa_odd_timing[6];
+	u8 goa_even_timing[6];
+	u8 goa_stv_lead_time_ck;
+	u8 goa_ckv_lead_time_ck;
+	u8 goa_ckv_dummy_vblank_num;
+	u8 goa_ckv_rise_precharge;
+	u8 goa_ckv_fall_precharge;
+	bool goa_ckv_non_overlap_ctl;
+	u8 goa_clr1_width_adj;
+	u8 goa_clr234_width_adj;
+	s8 goa_clr_polarity[4];
+	int goa_clr_start_pos[4];
+	u8 goa_unk_e4;
+	u8 goa_unk_e5;
+
+	/* Page 5 */
+	u8 bta_tlpx;
+	bool lhs_settle_time_by_osc25;
+	u8 ths_settle_time;
+	u8 timing_unk_b8;
+	u8 timing_unk_bc;
+	u8 timing_unk_d6;
+
+	/* Page 6 */
+	u8 gamma_ctl;
+	u8 volt_adj;
+	u8 src_delay_time_adj_ck;
+
+	/* Page 7..12 */
+	const struct hx8279_digital_gamma *dgamma;
+};
+
+static inline struct hx8279 *to_hx8279(struct drm_panel *panel)
+{
+	return container_of(panel, struct hx8279, panel);
+}
+
+static void hx8279_set_page(struct hx8279 *hx,
+			    struct mipi_dsi_multi_context *dsi_ctx, u8 page)
+{
+	const u8 cmd_set_page[] = { HX8279_REG_PAGE, page };
+
+	if (hx->last_page == page)
+		return;
+
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_page, ARRAY_SIZE(cmd_set_page));
+	if (!dsi_ctx->accum_err)
+		hx->last_page = page;
+}
+
+static void hx8279_set_module_config(struct hx8279 *hx,
+				     struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	u8 cmd_set_voltage[2];
+
+	if (hx->skip_voltage_config)
+		return;
+
+	/* Page 0 - Driver/Module Configuration */
+	hx8279_set_page(hx, dsi_ctx, 0);
+
+	if (desc->vgh_mv) {
+		cmd_set_voltage[0] = HX8279_P0_VGHS;
+		cmd_set_voltage[1] = HX8279_VGH_VOLT_SEL(desc->vgh_mv);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_voltage,
+					     ARRAY_SIZE(cmd_set_voltage));
+	}
+
+	if (desc->vgl_mv) {
+		cmd_set_voltage[0] = HX8279_P0_VGLS;
+		cmd_set_voltage[1] = HX8279_VGL_VOLT_SEL(desc->vgl_mv);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_voltage,
+					     ARRAY_SIZE(cmd_set_voltage));
+	}
+
+	if (desc->vgph_mv) {
+		cmd_set_voltage[0] = HX8279_P0_VGPHS;
+		cmd_set_voltage[1] = HX8279_VGPN_VOLT_SEL(desc->vgph_mv);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_voltage,
+					     ARRAY_SIZE(cmd_set_voltage));
+	}
+
+	if (desc->vgnh_mv) {
+		cmd_set_voltage[0] = HX8279_P0_VGNHS;
+		cmd_set_voltage[1] = HX8279_VGPN_VOLT_SEL(desc->vgnh_mv);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_voltage,
+					     ARRAY_SIZE(cmd_set_voltage));
+	}
+}
+
+static void hx8279_set_gmux(struct hx8279 *hx,
+			    struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_goa_mux *gmux = hx->desc->gmux;
+	u8 cmd_set_gmux[2];
+	int i;
+
+	if (!gmux)
+		return;
+
+	hx8279_set_page(hx, dsi_ctx, 1);
+
+	for (i = 0; i < ARRAY_SIZE(gmux->gout_l); i++) {
+		cmd_set_gmux[0] = HX8279_P1_REG_GOUTL(i);
+		cmd_set_gmux[1] = gmux->gout_l[i];
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_gmux,
+					     ARRAY_SIZE(cmd_set_gmux));
+	}
+
+	for (i = 0; i < ARRAY_SIZE(gmux->gout_r); i++) {
+		cmd_set_gmux[0] = HX8279_P1_REG_GOUTR(i);
+		cmd_set_gmux[1] = gmux->gout_r[i];
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_gmux,
+					     ARRAY_SIZE(cmd_set_gmux));
+	}
+}
+
+static void hx8279_set_analog_gamma(struct hx8279 *hx,
+				    struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_analog_gamma *agamma = hx->desc->agamma;
+	u8 cmd_set_ana_gamma[2];
+	int i;
+
+	if (!agamma)
+		return;
+
+	hx8279_set_page(hx, dsi_ctx, 2);
+
+	for (i = 0; i < ARRAY_SIZE(agamma->pos); i++) {
+		cmd_set_ana_gamma[0] = HX8279_P2_REG_GAMMA_T_PVP(i);
+		cmd_set_ana_gamma[1] = agamma->pos[i];
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_ana_gamma,
+					     ARRAY_SIZE(cmd_set_ana_gamma));
+	}
+
+	for (i = 0; i < ARRAY_SIZE(agamma->neg); i++) {
+		cmd_set_ana_gamma[0] = HX8279_P2_REG_GAMMA_T_PVN(i);
+		cmd_set_ana_gamma[1] = agamma->neg[i];
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_ana_gamma,
+					     ARRAY_SIZE(cmd_set_ana_gamma));
+	}
+}
+
+static void hx8279_set_goa_timing(struct hx8279 *hx,
+				  struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	u8 cmd_set_goa_t[2];
+	int i;
+
+	if (hx->skip_goa_timing)
+		return;
+
+	hx8279_set_page(hx, dsi_ctx, 3);
+
+	for (i = 0; i < ARRAY_SIZE(desc->goa_odd_timing); i++) {
+		cmd_set_goa_t[0] = HX8279_P3_REG_GOA_TO(i);
+		cmd_set_goa_t[1] = desc->goa_odd_timing[i];
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa_t,
+					     ARRAY_SIZE(cmd_set_goa_t));
+	}
+
+	for (i = 0; i < ARRAY_SIZE(desc->goa_even_timing); i++) {
+		cmd_set_goa_t[0] = HX8279_P3_REG_GOA_TE(i);
+		cmd_set_goa_t[1] = desc->goa_odd_timing[i];
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa_t,
+					     ARRAY_SIZE(cmd_set_goa_t));
+	}
+}
+
+static void hx8279_set_goa_cfg(struct hx8279 *hx,
+			       struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	u8 cmd_set_goa[2];
+	int i;
+
+	if (hx->skip_goa_config)
+		return;
+
+	hx8279_set_page(hx, dsi_ctx, 3);
+
+	if (desc->goa_unk_ba)  {
+		cmd_set_goa[0] = HX8279_P3_REG_UNKNOWN_BA;
+		cmd_set_goa[1] = desc->goa_unk_ba;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	if (desc->goa_stv_lead_time_ck) {
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_STVL;
+		cmd_set_goa[1] = FIELD_PREP(HX8279_P3_GOA_STV_LEAD,
+					    desc->goa_stv_lead_time_ck);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	if (desc->goa_ckv_lead_time_ck) {
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CKVL;
+		cmd_set_goa[1] = FIELD_PREP(HX8279_P3_GOA_CKV_DUMMY,
+					    desc->goa_ckv_lead_time_ck);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	if (desc->goa_ckv_dummy_vblank_num) {
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CKVD;
+		cmd_set_goa[1] = FIELD_PREP(HX8279_P3_GOA_CKV_LEAD,
+					    desc->goa_ckv_dummy_vblank_num);
+		cmd_set_goa[1] |= FIELD_PREP(HX8279_P3_GOA_CKV_NONOVERLAP,
+					     desc->goa_ckv_non_overlap_ctl);
+		/* RESERVED must be always set */
+		cmd_set_goa[1] |= HX8279_P3_GOA_CKV_RESERVED;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	/*
+	 * One of the two being more than zero means that we want to write
+	 * both of them. Anyway, the register default is zero in this case.
+	 */
+	if (desc->goa_ckv_rise_precharge || desc->goa_ckv_fall_precharge) {
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CKV_RISE_PREC;
+		cmd_set_goa[1] = desc->goa_ckv_rise_precharge;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CKV_FALL_PREC;
+		cmd_set_goa[1] = desc->goa_ckv_fall_precharge;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	if (desc->goa_clr1_width_adj) {
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CLR1_W_ADJ;
+		cmd_set_goa[1] = desc->goa_clr1_width_adj;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	if (desc->goa_clr234_width_adj) {
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CLR234_W_ADJ;
+		cmd_set_goa[1] = desc->goa_clr234_width_adj;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	/* Polarity and Start Position arrays are of the same size */
+	for (i = 0; i < ARRAY_SIZE(desc->goa_clr_polarity); i++) {
+		if (desc->goa_clr_polarity[i] < 0 || desc->goa_clr_start_pos[i] < 0)
+			continue;
+
+		cmd_set_goa[0] = HX8279_P3_REG_GOA_CLR_CFG(i);
+		cmd_set_goa[1] = FIELD_PREP(HX8279_P3_GOA_CLR_CFG_STARTPOS,
+					    desc->goa_clr_start_pos[i]);
+		cmd_set_goa[1] |= FIELD_PREP(HX8279_P3_GOA_CLR_CFG_POLARITY,
+					     desc->goa_clr_polarity[i]);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	if (desc->goa_unk_e4) {
+		cmd_set_goa[0] = HX8279_P3_REG_UNKNOWN_E4;
+		cmd_set_goa[1] = desc->goa_unk_e4;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+					     ARRAY_SIZE(cmd_set_goa));
+	}
+
+	cmd_set_goa[0] = HX8279_P3_REG_UNKNOWN_E5;
+	cmd_set_goa[1] = desc->goa_unk_e5;
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_goa,
+				     ARRAY_SIZE(cmd_set_goa));
+}
+
+static void hx8279_set_mipi_cfg(struct hx8279 *hx,
+				struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	u8 cmd_set_mipi[2];
+
+	if (hx->skip_mipi_timing)
+		return;
+
+	hx8279_set_page(hx, dsi_ctx, 5);
+
+	if (desc->bta_tlpx || desc->ths_settle_time || desc->lhs_settle_time_by_osc25) {
+		cmd_set_mipi[0] = HX8279_P5_REG_TIMING;
+		cmd_set_mipi[1] = FIELD_PREP(HX8279_P5_TIMING_TLPX, desc->bta_tlpx);
+		cmd_set_mipi[1] |= FIELD_PREP(HX8279_P5_TIMING_THS_SETTLE,
+					      desc->ths_settle_time);
+		cmd_set_mipi[1] |= FIELD_PREP(HX8279_P5_TIMING_LHS_SETTLE,
+					      desc->lhs_settle_time_by_osc25);
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_mipi,
+					     ARRAY_SIZE(cmd_set_mipi));
+	}
+
+	if (desc->timing_unk_b8) {
+		cmd_set_mipi[0] = HX8279_P5_REG_UNKNOWN_B8;
+		cmd_set_mipi[1] = desc->timing_unk_b8;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_mipi,
+					     ARRAY_SIZE(cmd_set_mipi));
+	}
+
+	if (desc->timing_unk_bc) {
+		cmd_set_mipi[0] = HX8279_P5_REG_UNKNOWN_BC;
+		cmd_set_mipi[1] = desc->timing_unk_bc;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_mipi,
+					     ARRAY_SIZE(cmd_set_mipi));
+	}
+
+	if (desc->timing_unk_d6) {
+		cmd_set_mipi[0] = HX8279_P5_REG_UNKNOWN_D6;
+		cmd_set_mipi[1] = desc->timing_unk_d6;
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_mipi,
+					     ARRAY_SIZE(cmd_set_mipi));
+	}
+}
+
+static void hx8279_set_adv_cfg(struct hx8279 *hx,
+			       struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	const u8 cmd_set_dly[] = { HX8279_P6_REG_DLY_TIME_ADJ, desc->src_delay_time_adj_ck };
+	const u8 cmd_set_gamma[] = { HX8279_P6_REG_GAMMA_CHOPPER, desc->gamma_ctl };
+	const u8 cmd_set_volt_adj[] = { HX8279_P6_REG_VOLT_ADJ, desc->volt_adj };
+	u8 cmd_set_eng[] = { HX8279_P6_REG_ENGINEER_PWD, HX8279_P6_ENG_UNLOCK_WORD };
+
+	if (!desc->gamma_ctl && !desc->src_delay_time_adj_ck && !desc->volt_adj)
+		return;
+
+	hx8279_set_page(hx, dsi_ctx, 6);
+
+	/* Unlock ENG settings: write same word to both ENGINEER_PWD and INHOUSE_FUNC */
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_eng, ARRAY_SIZE(cmd_set_eng));
+
+	cmd_set_eng[0] = HX8279_P6_REG_INHOUSE_FUNC;
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_eng, ARRAY_SIZE(cmd_set_eng));
+
+	/* Set Gamma Chopper and Gamma buffer Chopper control */
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_gamma, ARRAY_SIZE(cmd_set_gamma));
+
+	/* Set Source delay time adjustment (CKV falling to Source off) */
+	if (desc->src_delay_time_adj_ck)
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_dly,
+					     ARRAY_SIZE(cmd_set_dly));
+
+	/* Set voltage adjustment */
+	if (desc->volt_adj)
+		mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_volt_adj,
+					     ARRAY_SIZE(cmd_set_volt_adj));
+
+	/* Lock ENG settings again */
+	cmd_set_eng[0] = HX8279_P6_REG_ENGINEER_PWD;
+	cmd_set_eng[1] = 0;
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_eng, ARRAY_SIZE(cmd_set_eng));
+
+	cmd_set_eng[0] = HX8279_P6_REG_INHOUSE_FUNC;
+	mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_eng, ARRAY_SIZE(cmd_set_eng));
+}
+
+static void hx8279_set_digital_gamma(struct hx8279 *hx,
+				     struct mipi_dsi_multi_context *dsi_ctx)
+{
+	const struct hx8279_digital_gamma *dgamma = hx->desc->dgamma;
+	u8 cmd_set_dig_gamma[2];
+	int i;
+
+	if (!dgamma)
+		return;
+
+	/*
+	 * Pages 7..9 are for RGB Positive, 10..12 are for RGB Negative:
+	 * The first iteration sets all positive component registers,
+	 * the second one sets all negatives.
+	 */
+	for (i = 0; i < 2; i++) {
+		u8 pg_neg = i * 3;
+
+		hx8279_set_page(hx, dsi_ctx, 7 + pg_neg);
+
+		for (i = 0; i < ARRAY_SIZE(dgamma->r); i++) {
+			cmd_set_dig_gamma[0] = HX8279_PG_DIGITAL_GAMMA + i;
+			cmd_set_dig_gamma[1] = dgamma->r[i];
+			mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_dig_gamma,
+						     ARRAY_SIZE(cmd_set_dig_gamma));
+		}
+
+		hx8279_set_page(hx, dsi_ctx, 8 + pg_neg);
+
+		for (i = 0; i < ARRAY_SIZE(dgamma->g); i++) {
+			cmd_set_dig_gamma[0] = HX8279_PG_DIGITAL_GAMMA + i;
+			cmd_set_dig_gamma[1] = dgamma->g[i];
+			mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_dig_gamma,
+						     ARRAY_SIZE(cmd_set_dig_gamma));
+		}
+
+		hx8279_set_page(hx, dsi_ctx, 9 + pg_neg);
+
+		for (i = 0; i < ARRAY_SIZE(dgamma->b); i++) {
+			cmd_set_dig_gamma[0] = HX8279_PG_DIGITAL_GAMMA + i;
+			cmd_set_dig_gamma[1] = dgamma->b[i];
+			mipi_dsi_generic_write_multi(dsi_ctx, cmd_set_dig_gamma,
+						     ARRAY_SIZE(cmd_set_dig_gamma));
+		}
+	}
+}
+
+static int hx8279_on(struct hx8279 *hx)
+{
+	struct mipi_dsi_device *dsi = hx->dsi[0];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	/* Page 5 */
+	hx8279_set_mipi_cfg(hx, &dsi_ctx);
+
+	/* Page 1 */
+	hx8279_set_gmux(hx, &dsi_ctx);
+
+	/* Page 2 */
+	hx8279_set_analog_gamma(hx, &dsi_ctx);
+
+	/* Page 3 */
+	hx8279_set_goa_cfg(hx, &dsi_ctx);
+	hx8279_set_goa_timing(hx, &dsi_ctx);
+
+	/* Page 0 - Driver/Module Configuration */
+	hx8279_set_module_config(hx, &dsi_ctx);
+
+	/* Page 6 */
+	hx8279_set_adv_cfg(hx, &dsi_ctx);
+
+	/* Pages 7..12 */
+	hx8279_set_digital_gamma(hx, &dsi_ctx);
+
+	return dsi_ctx.accum_err;
+}
+
+static void hx8279_power_off(struct hx8279 *hx)
+{
+	gpiod_set_value_cansleep(hx->reset_gpio, 0);
+	usleep_range(100, 500);
+	gpiod_set_value_cansleep(hx->enable_gpio, 0);
+	regulator_bulk_disable(ARRAY_SIZE(hx->vregs), hx->vregs);
+}
+
+static int hx8279_disable(struct drm_panel *panel)
+{
+	struct hx8279 *hx = to_hx8279(panel);
+	struct mipi_dsi_device *dsi = hx->dsi[0];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+
+	return 0;
+}
+
+static int hx8279_enable(struct drm_panel *panel)
+{
+	struct hx8279 *hx = to_hx8279(panel);
+	struct mipi_dsi_device *dsi = hx->dsi[0];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
+
+	return 0;
+}
+
+static int hx8279_prepare(struct drm_panel *panel)
+{
+	struct hx8279 *hx = to_hx8279(panel);
+	struct mipi_dsi_device *dsi = hx->dsi[0];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+	int ret;
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(hx->vregs), hx->vregs);
+	if (ret)
+		return ret;
+
+	gpiod_set_value_cansleep(hx->enable_gpio, 1);
+	usleep_range(5000, 6000);
+	gpiod_set_value_cansleep(hx->reset_gpio, 1);
+	usleep_range(6000, 7000);
+
+	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+	if (hx->dsi[1])
+		hx->dsi[1]->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	ret = hx8279_on(hx);
+	if (ret) {
+		hx8279_power_off(hx);
+		return ret;
+	}
+
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 130);
+
+	return dsi_ctx.accum_err;
+}
+
+static int hx8279_unprepare(struct drm_panel *panel)
+{
+	struct hx8279 *hx = to_hx8279(panel);
+	struct mipi_dsi_device *dsi = hx->dsi[0];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 130);
+
+	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+	if (hx->dsi[1])
+		hx->dsi[1]->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	hx8279_power_off(hx);
+
+	return dsi_ctx.accum_err;
+}
+
+static int hx8279_get_modes(struct drm_panel *panel, struct drm_connector *connector)
+{
+	struct hx8279 *hx = to_hx8279(panel);
+	int i;
+
+	for (i = 0; i < hx->desc->num_modes; i++) {
+		struct drm_display_mode *mode;
+
+		mode = drm_mode_duplicate(connector->dev, &hx->desc->mode_data[i].mode);
+		if (!mode)
+			return -ENOMEM;
+
+		drm_mode_set_name(mode);
+
+		mode->type |= DRM_MODE_TYPE_DRIVER;
+		if (hx->desc->num_modes == 1)
+			mode->type |= DRM_MODE_TYPE_PREFERRED;
+
+		drm_mode_probed_add(connector, mode);
+	}
+
+	connector->display_info.bpc = hx->desc->mode_data[0].bpc;
+	connector->display_info.height_mm = hx->desc->mode_data[0].mode.height_mm;
+	connector->display_info.width_mm = hx->desc->mode_data[0].mode.width_mm;
+
+	return hx->desc->num_modes;
+}
+
+static const struct drm_panel_funcs hx8279_panel_funcs = {
+	.disable = hx8279_disable,
+	.unprepare = hx8279_unprepare,
+	.prepare = hx8279_prepare,
+	.enable = hx8279_enable,
+	.get_modes = hx8279_get_modes,
+};
+
+static int hx8279_init_vregs(struct hx8279 *hx, struct device *dev)
+{
+	int ret;
+
+	hx->vregs[0].supply = "vdd";
+	hx->vregs[1].supply = "iovcc";
+	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(hx->vregs),
+				      hx->vregs);
+	if (ret < 0)
+		return ret;
+
+	ret = regulator_is_supported_voltage(hx->vregs[0].consumer,
+					     3000000, 5000000);
+	if (!ret)
+		return -EINVAL;
+
+	ret = regulator_is_supported_voltage(hx->vregs[1].consumer,
+					     1700000, 1900000);
+	if (!ret)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int hx8279_check_gmux_config(struct hx8279 *hx, struct device *dev)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	const struct hx8279_goa_mux *gmux = desc->gmux;
+	int i;
+
+	/* No gmux defined means we simply skip the GOA mux configuration */
+	if (!gmux)
+		return 0;
+
+	for (i = 0; i < ARRAY_SIZE(gmux->gout_l); i++) {
+		if (gmux->gout_l[i] > (HX8279_GOUT_STB | HX8279_GOUT_SEL))
+			return dev_err_probe(dev, -EINVAL,
+					     "Invalid value found in gout_l[%d]\n", i);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(gmux->gout_r); i++) {
+		if (gmux->gout_r[i] > (HX8279_GOUT_STB | HX8279_GOUT_SEL))
+			return dev_err_probe(dev, -EINVAL,
+					     "Invalid value found in gout_r[%d]\n", i);
+	}
+
+	return 0;
+}
+
+static int hx8279_check_goa_config(struct hx8279 *hx, struct device *dev)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	bool goa_odd_valid, goa_even_valid;
+	int i, num_zero, num_clr = 0;
+
+	/* Up to 4 zero values is a valid configuration. Check them all. */
+	num_zero = 1;
+	for (i = 0; i < ARRAY_SIZE(desc->goa_odd_timing); i++) {
+		if (desc->goa_odd_timing[i])
+			num_zero++;
+	}
+
+	goa_odd_valid = (num_zero != ARRAY_SIZE(desc->goa_odd_timing));
+
+	/* Up to 3 zeroes is a valid config. Check them all. */
+	num_zero = 1;
+	for (i = 0; i < ARRAY_SIZE(desc->goa_even_timing); i++) {
+		if (desc->goa_even_timing[i])
+			num_zero++;
+	}
+
+	goa_even_valid = (num_zero != ARRAY_SIZE(desc->goa_even_timing));
+
+	/* Programming one without the other would make no sense! */
+	if (goa_odd_valid != goa_even_valid)
+		return -EINVAL;
+
+	/* We know that both are either true or false now, check just one */
+	if (!goa_odd_valid)
+		hx->skip_goa_timing = true;
+
+	if (!desc->goa_unk_ba && !desc->goa_stv_lead_time_ck &&
+	    !desc->goa_ckv_lead_time_ck && !desc->goa_ckv_dummy_vblank_num &&
+	    !desc->goa_ckv_rise_precharge && !desc->goa_ckv_fall_precharge &&
+	    !desc->goa_clr1_width_adj && !desc->goa_clr234_width_adj &&
+	    !desc->goa_unk_e4 && !desc->goa_unk_e5) {
+		hx->skip_goa_config = true;
+		return 0;
+	}
+
+	if ((desc->goa_stv_lead_time_ck > HX8279_P3_GOA_STV_LEAD) ||
+	    (desc->goa_ckv_lead_time_ck > HX8279_P3_GOA_CKV_LEAD) ||
+	    (desc->goa_ckv_dummy_vblank_num > HX8279_P3_GOA_CKV_DUMMY))
+		return dev_err_probe(dev, -EINVAL,
+				     "Invalid lead timings in GOA config\n");
+
+	/*
+	 * Don't perform zero check for polarity and start position, as
+	 * both pol=0 and start=0 are valid configuration values.
+	 */
+	for (i = 0; i < ARRAY_SIZE(desc->goa_clr_start_pos); i++) {
+		if (desc->goa_clr_start_pos[i] < 0)
+			continue;
+		else if (desc->goa_clr_start_pos[i] > HX8279_P3_GOA_CLR_CFG_STARTPOS)
+			return dev_err_probe(dev, -EINVAL,
+					     "Invalid start position for CLR%d\n", i + 1);
+		else
+			num_clr++;
+	}
+	if (!num_clr)
+		return -EINVAL;
+
+	for (i = 0; i < ARRAY_SIZE(desc->goa_clr_polarity); i++) {
+		if (num_clr < 0)
+			return -EINVAL;
+
+		if (desc->goa_clr_polarity[i] < 0)
+			continue;
+		else if (desc->goa_clr_polarity[i] > 1)
+			return dev_err_probe(dev, -EINVAL,
+					     "Invalid polarity for CLR%d\n", i + 1);
+		else
+			num_clr--;
+	}
+
+	return 0;
+}
+
+static int hx8279_check_dig_gamma(struct hx8279 *hx, struct device *dev, const u8 *component)
+{
+	u8 gamma_high_bits[4];
+	u16 prev_val = 0;
+	int i, j, k, x;
+
+	/*
+	 * The gamma values are 10 bits long and shall be incremental
+	 * to form a digital gamma correction reference curve.
+	 *
+	 * As for the registers format: the first 24 bytes contain each the
+	 * lowest 8 bits for each of the gamma level references, and the last
+	 * 6 bytes contain the high two bits of 4 registers at a time, where
+	 * the first two bits are relative to the last register, and the last
+	 * two are relative to the first register.
+	 *
+	 * Another way of saying, those are the first four LOW values:
+	 * DGMA1_LO = 0xb1, DGMA2_LO = 0xb2, DGMA3_LO = 0xb3, DGMA4_LO = 0xb4
+	 *
+	 * The high values for those four are at DGMA1_4_HI = 0xc9;
+	 * ...and DGMA1_4_HI's data contains the following bits:
+	 * [1:0] = DGMA4_HI, [3:2] = DGMA3_HI, [5:4] = DGMA2_HI, [7:6] = DGMA1_HI
+	 */
+	for (i = 0; i < HX8279_PG_DGAMMA_NUM_HI_BYTES; i++) {
+		k = HX8279_PG_DGAMMA_NUM_LO_BYTES + i;
+		j = i * 4;
+		x = 0;
+
+		gamma_high_bits[0] = FIELD_GET(HX8279_DGAMMA_DGMA1_HI, component[k]);
+		gamma_high_bits[1] = FIELD_GET(HX8279_DGAMMA_DGMA2_HI, component[k]);
+		gamma_high_bits[2] = FIELD_GET(HX8279_DGAMMA_DGMA3_HI, component[k]);
+		gamma_high_bits[3] = FIELD_GET(HX8279_DGAMMA_DGMA4_HI, component[k]);
+
+		do {
+			u16 cur_val = component[j] | (gamma_high_bits[x] << 8);
+
+			if (cur_val < prev_val)
+				return dev_err_probe(dev, -EINVAL,
+						     "Invalid dgamma values: %u < %u!\n",
+						     cur_val, prev_val);
+			prev_val = cur_val;
+			j++;
+			x++;
+		} while (x < 4);
+	};
+
+	return 0;
+}
+
+static int hx8279_check_params(struct hx8279 *hx, struct device *dev)
+{
+	const struct hx8279_panel_desc *desc = hx->desc;
+	int ret;
+
+	/* Voltages config validation */
+	if (!desc->vgh_mv && !desc->vgl_mv && !desc->vgph_mv && !desc->vgnh_mv)
+		hx->skip_voltage_config = true;
+	else if ((desc->vgh_mv && desc->vgh_mv < HX8279_VGH_MIN_MV) ||
+		 (desc->vgl_mv && desc->vgl_mv < HX8279_VGL_MIN_MV) ||
+		 (desc->vgph_mv && desc->vgph_mv < HX8279_VGPNH_MIN_MV) ||
+		 (desc->vgnh_mv && desc->vgnh_mv < HX8279_VGPNH_MIN_MV))
+		return -EINVAL;
+
+	/* GOA Muxing validation */
+	ret = hx8279_check_gmux_config(hx, dev);
+	if (ret)
+		return ret;
+
+	/* GOA Configuration validation */
+	ret = hx8279_check_goa_config(hx, dev);
+	if (ret)
+		return ret;
+
+	/* MIPI Configuration validation */
+	if (!desc->bta_tlpx && !desc->lhs_settle_time_by_osc25 &&
+	    !desc->ths_settle_time && !desc->timing_unk_b8 &&
+	    !desc->timing_unk_bc && !desc->timing_unk_d6)
+		hx->skip_mipi_timing = true;
+
+	/* ENG/Gamma Configuration validation */
+	if (desc->gamma_ctl > (HX8279_P6_GAMMA_POCGM_CTL | HX8279_P6_GAMMA_POGCMD_CTL))
+		return -EINVAL;
+
+	/* Digital Gamma values validation */
+	if (desc->dgamma) {
+		ret = hx8279_check_dig_gamma(hx, dev, desc->dgamma->r);
+		if (ret)
+			return ret;
+
+		ret = hx8279_check_dig_gamma(hx, dev, desc->dgamma->g);
+		if (ret)
+			return ret;
+
+		ret = hx8279_check_dig_gamma(hx, dev, desc->dgamma->b);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int hx8279_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct device_node *dsi_r;
+	struct hx8279 *hx;
+	int i, ret;
+
+	hx = devm_drm_panel_alloc(dev, struct hx8279, panel,
+				  &hx8279_panel_funcs, DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(hx))
+		return PTR_ERR(hx);
+
+	ret = hx8279_init_vregs(hx, dev);
+	if (ret)
+		return ret;
+
+	hx->desc = device_get_match_data(dev);
+	if (!hx->desc)
+		return -ENODEV;
+
+	/*
+	 * In some DriverICs some or all fields may be OTP: perform a
+	 * basic configuration check before writing to help avoiding
+	 * irreparable mistakes.
+	 *
+	 * Please note that this is not perfect and will only check if
+	 * the values may be plausible; values that are wrong for a
+	 * specific display, but still plausible for DrIC config will
+	 * be accepted.
+	 */
+	ret = hx8279_check_params(hx, dev);
+	if (ret)
+		return dev_err_probe(dev, ret, "Invalid DriverIC configuration\n");
+
+	/* The enable line may be always tied to VCCIO, so this is optional */
+	hx->enable_gpio = devm_gpiod_get_optional(dev, "enable", GPIOD_ASIS);
+	if (IS_ERR(hx->enable_gpio))
+		return dev_err_probe(dev, PTR_ERR(hx->enable_gpio),
+				     "Failed to get enable GPIO\n");
+
+	hx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_ASIS);
+	if (IS_ERR(hx->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(hx->reset_gpio),
+				     "Failed to get reset GPIO\n");
+
+	/* If the panel is connected on two DSIs then DSI0 left, DSI1 right */
+	dsi_r = of_graph_get_remote_node(dsi->dev.of_node, 1, -1);
+	if (dsi_r) {
+		const struct mipi_dsi_device_info *info = &hx->desc->dsi_info;
+		struct mipi_dsi_host *dsi_r_host;
+
+		dsi_r_host = of_find_mipi_dsi_host_by_node(dsi_r);
+		of_node_put(dsi_r);
+		if (!dsi_r_host)
+			return dev_err_probe(dev, -EPROBE_DEFER,
+					     "Cannot get secondary DSI host\n");
+
+		hx->dsi[1] = devm_mipi_dsi_device_register_full(dev, dsi_r_host, info);
+		if (IS_ERR(hx->dsi[1]))
+			return dev_err_probe(dev, PTR_ERR(hx->dsi[1]),
+					     "Cannot get secondary DSI node\n");
+		mipi_dsi_set_drvdata(hx->dsi[1], hx);
+	}
+
+	hx->dsi[0] = dsi;
+	mipi_dsi_set_drvdata(dsi, hx);
+
+	ret = drm_panel_of_backlight(&hx->panel);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get backlight\n");
+
+	drm_panel_add(&hx->panel);
+
+	for (i = 0; i < 2; i++) {
+		if (!hx->dsi[i])
+			continue;
+
+		hx->dsi[i]->lanes = hx->desc->num_lanes;
+		hx->dsi[i]->format = MIPI_DSI_FMT_RGB888;
+
+		hx->dsi[i]->mode_flags = MIPI_DSI_CLOCK_NON_CONTINUOUS |
+					 MIPI_DSI_MODE_LPM;
+
+		if (hx->desc->mode_data[0].is_video_mode)
+			hx->dsi[i]->mode_flags |= MIPI_DSI_MODE_VIDEO |
+						  MIPI_DSI_MODE_VIDEO_SYNC_PULSE;
+
+		ret = devm_mipi_dsi_attach(dev, hx->dsi[i]);
+		if (ret < 0) {
+			drm_panel_remove(&hx->panel);
+			return dev_err_probe(dev, ret,
+					     "Cannot attach to DSI%d host.\n", i);
+		}
+	}
+
+	return 0;
+}
+
+static void hx8279_remove(struct mipi_dsi_device *dsi)
+{
+	struct hx8279 *hx = mipi_dsi_get_drvdata(dsi);
+
+	drm_panel_remove(&hx->panel);
+}
+
+static const struct hx8279_panel_mode aoly_sl101pm1794fog_v15_modes[] = {
+	{
+		.mode = {
+			.clock = 159420,
+			.hdisplay = 1200,
+			.hsync_start = 1200 + 80,
+			.hsync_end = 1200 + 80 + 60,
+			.htotal = 1200 + 80 + 60 + 24,
+			.vdisplay = 1920,
+			.vsync_start = 1920 + 10,
+			.vsync_end = 1920 + 10 + 14,
+			.vtotal = 1920 + 10 + 14 + 4,
+			.width_mm = 136,
+			.height_mm = 217,
+			.type = DRM_MODE_TYPE_DRIVER
+		},
+		.bpc = 8,
+		.is_video_mode = true,
+	},
+};
+
+static const struct hx8279_panel_mode startek_kd070fhfid078_modes[] = {
+	{
+		.mode = {
+			.clock = 156458,
+			.hdisplay = 1200,
+			.hsync_start = 1200 + 50,
+			.hsync_end = 1200 + 50 + 24,
+			.htotal = 1200 + 50 + 24 + 66,
+			.vdisplay = 1920,
+			.vsync_start = 1920 + 14,
+			.vsync_end = 1920 + 14 + 2,
+			.vtotal = 1920 + 14 + 2 + 10,
+			.width_mm = 95,
+			.height_mm = 151,
+			.type = DRM_MODE_TYPE_DRIVER
+		},
+		.bpc = 8,
+		.is_video_mode = true,
+	},
+};
+
+static const struct hx8279_goa_mux aoly_sl101pm1794fog_v15_gmux = {
+	.gout_l = { 0x5, 0x5, 0xb, 0xb, 0x9, 0x9, 0x16, 0x16, 0xe, 0xe,
+		    0x7, 0x7, 0x26, 0x26, 0x15, 0x15, 0x1, 0x1, 0x3, 0x3 },
+	.gout_r = { 0x6, 0x6, 0xc, 0xc, 0xa, 0xa, 0x16, 0x16, 0xe, 0xe,
+		    0x8, 0x8, 0x26, 0x26, 0x15, 0x15, 0x2, 0x2, 0x4, 0x4 },
+};
+
+static const struct hx8279_analog_gamma aoly_sl101pm1794fog_v15_ana_gamma = {
+	.pos = { 0x0, 0xd, 0x17, 0x26, 0x31, 0x1c, 0x2c, 0x33, 0x31,
+		 0x37, 0x37, 0x37, 0x39, 0x2e, 0x2f, 0x2f, 0x7 },
+	.neg = { 0x0, 0xd, 0x17, 0x26, 0x31, 0x3f, 0x3f, 0x3f, 0x3f,
+		 0x37, 0x37, 0x37, 0x39, 0x2e, 0x2f, 0x2f, 0x7 },
+};
+
+static const struct hx8279_digital_gamma aoly_sl101pm1794fog_v15_dig_gamma = {
+	.r = { 0x0, 0x5, 0x10, 0x22, 0x36, 0x4a, 0x6c, 0x9a, 0xd7, 0x17,
+	       0x92, 0x15, 0x18, 0x8c, 0x0, 0x3a, 0x72, 0x8c, 0xa5, 0xb1,
+	       0xbe, 0xca, 0xd1, 0xd4, 0x0, 0x0, 0x16, 0xaf, 0xff, 0xff },
+	.g = { 0x4, 0x5, 0x11, 0x24, 0x39, 0x4e, 0x72, 0xa3, 0xe1, 0x25,
+	       0xa8, 0x2e, 0x32, 0xad, 0x28, 0x63, 0x9b, 0xb5, 0xcf, 0xdb,
+	       0xe8, 0xf5, 0xfa, 0xfc, 0x0, 0x0, 0x16, 0xaf, 0xff, 0xff },
+	.b = { 0x4, 0x4, 0xf, 0x22, 0x37, 0x4d, 0x71, 0xa2, 0xe1, 0x26,
+	       0xa9, 0x2f, 0x33, 0xac, 0x24, 0x5d, 0x94, 0xac, 0xc5, 0xd1,
+	       0xdc, 0xe8, 0xed, 0xf0, 0x0, 0x0, 0x16, 0xaf, 0xff, 0xff },
+};
+
+static const struct hx8279_panel_desc aoly_sl101pm1794fog_v15 = {
+	.dsi_info = {
+		.type = "L101PM1794FOG-V15",
+		.channel = 0,
+		.node = NULL,
+	},
+	.mode_data = aoly_sl101pm1794fog_v15_modes,
+	.num_modes = ARRAY_SIZE(aoly_sl101pm1794fog_v15_modes),
+	.num_lanes = 4,
+
+	/* Driver/Module Configuration: LC Matrix voltages */
+	.vgh_mv = 16500,
+	.vgl_mv = 11200,
+	.vgph_mv = 4600,
+	.vgnh_mv = 4600,
+
+	/* Analog Gamma correction */
+	.agamma = &aoly_sl101pm1794fog_v15_ana_gamma,
+
+	/* Gate driver On Array (GOA) Muxing */
+	.gmux = &aoly_sl101pm1794fog_v15_gmux,
+
+	/* Gate driver On Array (GOA) Mux Config */
+	.goa_unk_ba = 0xf0,
+	.goa_odd_timing = { 0, 0, 0, 42, 0, 0 },
+	.goa_even_timing = { 1, 42, 0, 0 },
+	.goa_stv_lead_time_ck = 11,
+	.goa_ckv_lead_time_ck = 7,
+	.goa_ckv_dummy_vblank_num = 3,
+	.goa_ckv_rise_precharge = 1,
+	.goa_clr1_width_adj = 0,
+	.goa_clr234_width_adj = 0,
+	.goa_clr_polarity = { 1, 0, 0, 0 },
+	.goa_clr_start_pos = { 8, 9, 3, 4 },
+	.goa_unk_e4 = 0xc0,
+	.goa_unk_e5 = 0x0d,
+
+	/* MIPI Configuration */
+	.bta_tlpx = 2,
+	.lhs_settle_time_by_osc25 = true,
+	.ths_settle_time = 2,
+	.timing_unk_b8 = 0xa5,
+	.timing_unk_bc = 0x20,
+	.timing_unk_d6 = 0x7f,
+
+	/* ENG/Gamma Configuration */
+	.gamma_ctl = 0,
+	.volt_adj = FIELD_PREP_CONST(HX8279_P6_VOLT_ADJ_VCCIFS, 3) |
+		    FIELD_PREP_CONST(HX8279_P6_VOLT_ADJ_VCCS, 3),
+	.src_delay_time_adj_ck = 50,
+
+	/* Digital Gamma Adjustment */
+	.dgamma = &aoly_sl101pm1794fog_v15_dig_gamma,
+};
+
+static const struct hx8279_goa_mux startek_kd070fhfid078_gmux = {
+	.gout_l = { 0xd, 0xd, 0x6, 0x6, 0x8, 0x8, 0xa, 0xa, 0xc, 0xc,
+		    0x0, 0x0, 0xe, 0xe, 0x1, 0x1, 0x4, 0x4, 0x0, 0x0 },
+	.gout_r = { 0xd, 0xd, 0x5, 0x5, 0x7, 0x7, 0x9, 0x9, 0xb, 0xb,
+		    0x0, 0x0, 0xe, 0xe, 0x1, 0x1, 0x3, 0x3, 0x0, 0x0 },
+};
+
+static const struct hx8279_panel_desc startek_kd070fhfid078 = {
+	.dsi_info = {
+		.type = "KD070FHFID078",
+		.channel = 0,
+		.node = NULL,
+	},
+	.mode_data = startek_kd070fhfid078_modes,
+	.num_modes = ARRAY_SIZE(startek_kd070fhfid078_modes),
+	.num_lanes = 4,
+
+	/* Driver/Module Configuration: LC Matrix voltages */
+	.vgh_mv = 18000,
+	.vgl_mv = 12100,
+	.vgph_mv = 5500,
+	.vgnh_mv = 5500,
+
+	/* Gate driver On Array (GOA) Mux Config */
+	.gmux = &startek_kd070fhfid078_gmux,
+
+	/* Gate driver On Array (GOA) Configuration */
+	.goa_unk_ba = 0xf0,
+	.goa_stv_lead_time_ck = 7,
+	.goa_ckv_lead_time_ck = 3,
+	.goa_ckv_dummy_vblank_num = 1,
+	.goa_ckv_rise_precharge = 0,
+	.goa_ckv_fall_precharge = 0,
+	.goa_clr1_width_adj = 1,
+	.goa_clr234_width_adj = 5,
+	.goa_clr_polarity = { 0, 1, -1, -1 },
+	.goa_clr_start_pos = { 5, 10, -1, -1 },
+	.goa_unk_e4 = 0xc0,
+	.goa_unk_e5 = 0x00,
+
+	/* MIPI Configuration */
+	.bta_tlpx = 2,
+	.lhs_settle_time_by_osc25 = true,
+	.ths_settle_time = 2,
+	.timing_unk_b8 = 0x7f,
+	.timing_unk_bc = 0x20,
+	.timing_unk_d6 = 0x7f,
+
+	/* ENG/Gamma Configuration */
+	.gamma_ctl = FIELD_PREP_CONST(HX8279_P6_GAMMA_POCGM_CTL, 1) |
+		     FIELD_PREP_CONST(HX8279_P6_GAMMA_POGCMD_CTL, 1),
+	.src_delay_time_adj_ck = 72,
+};
+
+static const struct of_device_id hx8279_of_match[] = {
+	{ .compatible = "aoly,sl101pm1794fog-v15", .data = &aoly_sl101pm1794fog_v15 },
+	{ .compatible = "startek,kd070fhfid078", .data = &startek_kd070fhfid078 },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, hx8279_of_match);
+
+static struct mipi_dsi_driver hx8279_driver = {
+	.probe = hx8279_probe,
+	.remove = hx8279_remove,
+	.driver = {
+		.name = "panel-himax-hx8279",
+		.of_match_table = hx8279_of_match,
+	},
+};
+module_mipi_dsi_driver(hx8279_driver);
+
+MODULE_AUTHOR("AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>");
+MODULE_DESCRIPTION("Himax HX8279 DriverIC panels driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/panel/panel-himax-hx8394.c b/drivers/gpu/drm/panel/panel-himax-hx8394.c
index 92b03a2f65a3..ff994bf0e3cc 100644
--- a/drivers/gpu/drm/panel/panel-himax-hx8394.c
+++ b/drivers/gpu/drm/panel/panel-himax-hx8394.c
@@ -80,7 +80,7 @@ struct hx8394_panel_desc {
 	unsigned int lanes;
 	unsigned long mode_flags;
 	enum mipi_dsi_pixel_format format;
-	int (*init_sequence)(struct hx8394 *ctx);
+	void (*init_sequence)(struct mipi_dsi_multi_context *dsi_ctx);
 };
 
 static inline struct hx8394 *panel_to_hx8394(struct drm_panel *panel)
@@ -88,98 +88,94 @@ static inline struct hx8394 *panel_to_hx8394(struct drm_panel *panel)
 	return container_of(panel, struct hx8394, panel);
 }
 
-static int hsd060bhw4_init_sequence(struct hx8394 *ctx)
+static void hsd060bhw4_init_sequence(struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
-
 	/* 5.19.8 SETEXTC: Set extension command (B9h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETEXTC,
-			       0xff, 0x83, 0x94);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETEXTC,
+				     0xff, 0x83, 0x94);
 
 	/* 5.19.2 SETPOWER: Set power (B1h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPOWER,
-			       0x48, 0x11, 0x71, 0x09, 0x32, 0x24, 0x71, 0x31, 0x55, 0x30);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPOWER,
+				     0x48, 0x11, 0x71, 0x09, 0x32, 0x24, 0x71, 0x31, 0x55, 0x30);
 
 	/* 5.19.9 SETMIPI: Set MIPI control (BAh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETMIPI,
-			       0x63, 0x03, 0x68, 0x6b, 0xb2, 0xc0);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETMIPI,
+				     0x63, 0x03, 0x68, 0x6b, 0xb2, 0xc0);
 
 	/* 5.19.3 SETDISP: Set display related register (B2h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETDISP,
-			       0x00, 0x80, 0x78, 0x0c, 0x07);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETDISP,
+				     0x00, 0x80, 0x78, 0x0c, 0x07);
 
 	/* 5.19.4 SETCYC: Set display waveform cycles (B4h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETCYC,
-			       0x12, 0x63, 0x12, 0x63, 0x12, 0x63, 0x01, 0x0c, 0x7c, 0x55,
-			       0x00, 0x3f, 0x12, 0x6b, 0x12, 0x6b, 0x12, 0x6b, 0x01, 0x0c,
-			       0x7c);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETCYC,
+				     0x12, 0x63, 0x12, 0x63, 0x12, 0x63, 0x01, 0x0c, 0x7c, 0x55,
+				     0x00, 0x3f, 0x12, 0x6b, 0x12, 0x6b, 0x12, 0x6b, 0x01, 0x0c,
+				     0x7c);
 
 	/* 5.19.19 SETGIP0: Set GIP Option0 (D3h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP0,
-			       0x00, 0x00, 0x00, 0x00, 0x3c, 0x1c, 0x00, 0x00, 0x32, 0x10,
-			       0x09, 0x00, 0x09, 0x32, 0x15, 0xad, 0x05, 0xad, 0x32, 0x00,
-			       0x00, 0x00, 0x00, 0x37, 0x03, 0x0b, 0x0b, 0x37, 0x00, 0x00,
-			       0x00, 0x0c, 0x40);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP0,
+				     0x00, 0x00, 0x00, 0x00, 0x3c, 0x1c, 0x00, 0x00, 0x32, 0x10,
+				     0x09, 0x00, 0x09, 0x32, 0x15, 0xad, 0x05, 0xad, 0x32, 0x00,
+				     0x00, 0x00, 0x00, 0x37, 0x03, 0x0b, 0x0b, 0x37, 0x00, 0x00,
+				     0x00, 0x0c, 0x40);
 
 	/* 5.19.20 Set GIP Option1 (D5h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP1,
-			       0x19, 0x19, 0x18, 0x18, 0x1b, 0x1b, 0x1a, 0x1a, 0x00, 0x01,
-			       0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x20, 0x21, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x24, 0x25, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP1,
+				     0x19, 0x19, 0x18, 0x18, 0x1b, 0x1b, 0x1a, 0x1a, 0x00, 0x01,
+				     0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x20, 0x21, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x24, 0x25, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18);
 
 	/* 5.19.21 Set GIP Option2 (D6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP2,
-			       0x18, 0x18, 0x19, 0x19, 0x1b, 0x1b, 0x1a, 0x1a, 0x07, 0x06,
-			       0x05, 0x04, 0x03, 0x02, 0x01, 0x00, 0x25, 0x24, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x21, 0x20, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP2,
+				     0x18, 0x18, 0x19, 0x19, 0x1b, 0x1b, 0x1a, 0x1a, 0x07, 0x06,
+				     0x05, 0x04, 0x03, 0x02, 0x01, 0x00, 0x25, 0x24, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x21, 0x20, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18);
 
 	/* 5.19.25 SETGAMMA: Set gamma curve related setting (E0h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGAMMA,
-			       0x00, 0x04, 0x0c, 0x12, 0x14, 0x18, 0x1a, 0x18, 0x31, 0x3f,
-			       0x4d, 0x4c, 0x54, 0x65, 0x6b, 0x70, 0x7f, 0x82, 0x7e, 0x8a,
-			       0x99, 0x4a, 0x48, 0x49, 0x4b, 0x4a, 0x4c, 0x4b, 0x7f, 0x00,
-			       0x04, 0x0c, 0x11, 0x13, 0x17, 0x1a, 0x18, 0x31,
-			       0x3f, 0x4d, 0x4c, 0x54, 0x65, 0x6b, 0x70, 0x7f,
-			       0x82, 0x7e, 0x8a, 0x99, 0x4a, 0x48, 0x49, 0x4b,
-			       0x4a, 0x4c, 0x4b, 0x7f);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGAMMA,
+				     0x00, 0x04, 0x0c, 0x12, 0x14, 0x18, 0x1a, 0x18, 0x31, 0x3f,
+				     0x4d, 0x4c, 0x54, 0x65, 0x6b, 0x70, 0x7f, 0x82, 0x7e, 0x8a,
+				     0x99, 0x4a, 0x48, 0x49, 0x4b, 0x4a, 0x4c, 0x4b, 0x7f, 0x00,
+				     0x04, 0x0c, 0x11, 0x13, 0x17, 0x1a, 0x18, 0x31,
+				     0x3f, 0x4d, 0x4c, 0x54, 0x65, 0x6b, 0x70, 0x7f,
+				     0x82, 0x7e, 0x8a, 0x99, 0x4a, 0x48, 0x49, 0x4b,
+				     0x4a, 0x4c, 0x4b, 0x7f);
 
 	/* 5.19.17 SETPANEL (CCh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPANEL,
-			       0x0b);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPANEL,
+				     0x0b);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN1,
-			       0x1f, 0x31);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN1,
+				     0x1f, 0x31);
 
 	/* 5.19.5 SETVCOM: Set VCOM voltage (B6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETVCOM,
-			       0x7d, 0x7d);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETVCOM,
+				     0x7d, 0x7d);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN3,
-			       0x02);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN3,
+				     0x02);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x01);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x01);
 
 	/* 5.19.2 SETPOWER: Set power (B1h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPOWER,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPOWER,
+				     0x00);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x00);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN3,
-			       0xed);
-
-	return 0;
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN3,
+				     0xed);
 }
 
 static const struct drm_display_mode hsd060bhw4_mode = {
@@ -205,114 +201,110 @@ static const struct hx8394_panel_desc hsd060bhw4_desc = {
 	.init_sequence = hsd060bhw4_init_sequence,
 };
 
-static int powkiddy_x55_init_sequence(struct hx8394 *ctx)
+static void powkiddy_x55_init_sequence(struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
-
 	/* 5.19.8 SETEXTC: Set extension command (B9h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETEXTC,
-			       0xff, 0x83, 0x94);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETEXTC,
+				     0xff, 0x83, 0x94);
 
 	/* 5.19.9 SETMIPI: Set MIPI control (BAh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETMIPI,
-			       0x63, 0x03, 0x68, 0x6b, 0xb2, 0xc0);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETMIPI,
+				     0x63, 0x03, 0x68, 0x6b, 0xb2, 0xc0);
 
 	/* 5.19.2 SETPOWER: Set power (B1h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPOWER,
-			       0x48, 0x12, 0x72, 0x09, 0x32, 0x54, 0x71, 0x71, 0x57, 0x47);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPOWER,
+				     0x48, 0x12, 0x72, 0x09, 0x32, 0x54, 0x71, 0x71, 0x57, 0x47);
 
 	/* 5.19.3 SETDISP: Set display related register (B2h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETDISP,
-			       0x00, 0x80, 0x64, 0x2c, 0x16, 0x2f);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETDISP,
+				     0x00, 0x80, 0x64, 0x2c, 0x16, 0x2f);
 
 	/* 5.19.4 SETCYC: Set display waveform cycles (B4h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETCYC,
-			       0x73, 0x74, 0x73, 0x74, 0x73, 0x74, 0x01, 0x0c, 0x86, 0x75,
-			       0x00, 0x3f, 0x73, 0x74, 0x73, 0x74, 0x73, 0x74, 0x01, 0x0c,
-			       0x86);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETCYC,
+				     0x73, 0x74, 0x73, 0x74, 0x73, 0x74, 0x01, 0x0c, 0x86, 0x75,
+				     0x00, 0x3f, 0x73, 0x74, 0x73, 0x74, 0x73, 0x74, 0x01, 0x0c,
+				     0x86);
 
 	/* 5.19.5 SETVCOM: Set VCOM voltage (B6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETVCOM,
-			       0x6e, 0x6e);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETVCOM,
+				     0x6e, 0x6e);
 
 	/* 5.19.19 SETGIP0: Set GIP Option0 (D3h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP0,
-			       0x00, 0x00, 0x07, 0x07, 0x40, 0x07, 0x0c, 0x00, 0x08, 0x10,
-			       0x08, 0x00, 0x08, 0x54, 0x15, 0x0a, 0x05, 0x0a, 0x02, 0x15,
-			       0x06, 0x05, 0x06, 0x47, 0x44, 0x0a, 0x0a, 0x4b, 0x10, 0x07,
-			       0x07, 0x0c, 0x40);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP0,
+				     0x00, 0x00, 0x07, 0x07, 0x40, 0x07, 0x0c, 0x00, 0x08, 0x10,
+				     0x08, 0x00, 0x08, 0x54, 0x15, 0x0a, 0x05, 0x0a, 0x02, 0x15,
+				     0x06, 0x05, 0x06, 0x47, 0x44, 0x0a, 0x0a, 0x4b, 0x10, 0x07,
+				     0x07, 0x0c, 0x40);
 
 	/* 5.19.20 Set GIP Option1 (D5h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP1,
-			       0x1c, 0x1c, 0x1d, 0x1d, 0x00, 0x01, 0x02, 0x03, 0x04, 0x05,
-			       0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x24, 0x25, 0x18, 0x18,
-			       0x26, 0x27, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x20, 0x21,
-			       0x18, 0x18, 0x18, 0x18);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP1,
+				     0x1c, 0x1c, 0x1d, 0x1d, 0x00, 0x01, 0x02, 0x03, 0x04, 0x05,
+				     0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x24, 0x25, 0x18, 0x18,
+				     0x26, 0x27, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x20, 0x21,
+				     0x18, 0x18, 0x18, 0x18);
 
 	/* 5.19.21 Set GIP Option2 (D6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP2,
-			       0x1c, 0x1c, 0x1d, 0x1d, 0x07, 0x06, 0x05, 0x04, 0x03, 0x02,
-			       0x01, 0x00, 0x0b, 0x0a, 0x09, 0x08, 0x21, 0x20, 0x18, 0x18,
-			       0x27, 0x26, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x25, 0x24,
-			       0x18, 0x18, 0x18, 0x18);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP2,
+				     0x1c, 0x1c, 0x1d, 0x1d, 0x07, 0x06, 0x05, 0x04, 0x03, 0x02,
+				     0x01, 0x00, 0x0b, 0x0a, 0x09, 0x08, 0x21, 0x20, 0x18, 0x18,
+				     0x27, 0x26, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x25, 0x24,
+				     0x18, 0x18, 0x18, 0x18);
 
 	/* 5.19.25 SETGAMMA: Set gamma curve related setting (E0h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGAMMA,
-			       0x00, 0x0a, 0x15, 0x1b, 0x1e, 0x21, 0x24, 0x22, 0x47, 0x56,
-			       0x65, 0x66, 0x6e, 0x82, 0x88, 0x8b, 0x9a, 0x9d, 0x98, 0xa8,
-			       0xb9, 0x5d, 0x5c, 0x61, 0x66, 0x6a, 0x6f, 0x7f, 0x7f, 0x00,
-			       0x0a, 0x15, 0x1b, 0x1e, 0x21, 0x24, 0x22, 0x47, 0x56, 0x65,
-			       0x65, 0x6e, 0x81, 0x87, 0x8b, 0x98, 0x9d, 0x99, 0xa8, 0xba,
-			       0x5d, 0x5d, 0x62, 0x67, 0x6b, 0x72, 0x7f, 0x7f);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGAMMA,
+				     0x00, 0x0a, 0x15, 0x1b, 0x1e, 0x21, 0x24, 0x22, 0x47, 0x56,
+				     0x65, 0x66, 0x6e, 0x82, 0x88, 0x8b, 0x9a, 0x9d, 0x98, 0xa8,
+				     0xb9, 0x5d, 0x5c, 0x61, 0x66, 0x6a, 0x6f, 0x7f, 0x7f, 0x00,
+				     0x0a, 0x15, 0x1b, 0x1e, 0x21, 0x24, 0x22, 0x47, 0x56, 0x65,
+				     0x65, 0x6e, 0x81, 0x87, 0x8b, 0x98, 0x9d, 0x99, 0xa8, 0xba,
+				     0x5d, 0x5d, 0x62, 0x67, 0x6b, 0x72, 0x7f, 0x7f);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN1,
-			       0x1f, 0x31);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN1,
+				     0x1f, 0x31);
 
 	/* 5.19.17 SETPANEL (CCh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPANEL,
-			       0x0b);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPANEL,
+				     0x0b);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN3,
-			       0x02);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN3,
+				     0x02);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x02);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x02);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN4,
-			       0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
-			       0xff, 0xff);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN4,
+				     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+				     0xff, 0xff);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x00);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x01);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x01);
 
 	/* 5.19.2 SETPOWER: Set power (B1h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPOWER,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPOWER,
+				     0x00);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x00);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN5,
-			       0x40, 0x81, 0x50, 0x00, 0x1a, 0xfc, 0x01);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN5,
+				     0x40, 0x81, 0x50, 0x00, 0x1a, 0xfc, 0x01);
 
 	/* Unknown command, not listed in the HX8394-F datasheet */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN2,
-			       0xed);
-
-	return 0;
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN2,
+				     0xed);
 }
 
 static const struct drm_display_mode powkiddy_x55_mode = {
@@ -339,131 +331,127 @@ static const struct hx8394_panel_desc powkiddy_x55_desc = {
 	.init_sequence = powkiddy_x55_init_sequence,
 };
 
-static int mchp_ac40t08a_init_sequence(struct hx8394 *ctx)
+static void mchp_ac40t08a_init_sequence(struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
-
 	/* DCS commands do not seem to be sent correclty without this delay */
-	msleep(20);
+	mipi_dsi_msleep(dsi_ctx, 20);
 
 	/* 5.19.8 SETEXTC: Set extension command (B9h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETEXTC,
-			       0xff, 0x83, 0x94);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETEXTC,
+				     0xff, 0x83, 0x94);
 
 	/* 5.19.9 SETMIPI: Set MIPI control (BAh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETMIPI,
-			       0x63, 0x03, 0x68, 0x6b, 0xb2, 0xc0);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETMIPI,
+				     0x63, 0x03, 0x68, 0x6b, 0xb2, 0xc0);
 
 	/* 5.19.2 SETPOWER: Set power (B1h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPOWER,
-			       0x48, 0x12, 0x72, 0x09, 0x32, 0x54,
-			       0x71, 0x71, 0x57, 0x47);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPOWER,
+				     0x48, 0x12, 0x72, 0x09, 0x32, 0x54,
+				     0x71, 0x71, 0x57, 0x47);
 
 	/* 5.19.3 SETDISP: Set display related register (B2h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETDISP,
-			       0x00, 0x80, 0x64, 0x0c, 0x0d, 0x2f);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETDISP,
+				     0x00, 0x80, 0x64, 0x0c, 0x0d, 0x2f);
 
 	/* 5.19.4 SETCYC: Set display waveform cycles (B4h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETCYC,
-			       0x73, 0x74, 0x73, 0x74, 0x73, 0x74,
-			       0x01, 0x0c, 0x86, 0x75, 0x00, 0x3f,
-			       0x73, 0x74, 0x73, 0x74, 0x73, 0x74,
-			       0x01, 0x0c, 0x86);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETCYC,
+				     0x73, 0x74, 0x73, 0x74, 0x73, 0x74,
+				     0x01, 0x0c, 0x86, 0x75, 0x00, 0x3f,
+				     0x73, 0x74, 0x73, 0x74, 0x73, 0x74,
+				     0x01, 0x0c, 0x86);
 
 	/* 5.19.5 SETVCOM: Set VCOM voltage (B6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETVCOM,
-			       0x6e, 0x6e);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETVCOM,
+				     0x6e, 0x6e);
 
 	/* 5.19.19 SETGIP0: Set GIP Option0 (D3h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP0,
-			       0x00, 0x00, 0x07, 0x07, 0x40, 0x07,
-			       0x0c, 0x00, 0x08, 0x10, 0x08, 0x00,
-			       0x08, 0x54, 0x15, 0x0a, 0x05, 0x0a,
-			       0x02, 0x15, 0x06, 0x05, 0x06, 0x47,
-			       0x44, 0x0a, 0x0a, 0x4b, 0x10, 0x07,
-			       0x07, 0x0c, 0x40);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP0,
+				     0x00, 0x00, 0x07, 0x07, 0x40, 0x07,
+				     0x0c, 0x00, 0x08, 0x10, 0x08, 0x00,
+				     0x08, 0x54, 0x15, 0x0a, 0x05, 0x0a,
+				     0x02, 0x15, 0x06, 0x05, 0x06, 0x47,
+				     0x44, 0x0a, 0x0a, 0x4b, 0x10, 0x07,
+				     0x07, 0x0c, 0x40);
 
 	/* 5.19.20 Set GIP Option1 (D5h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP1,
-			       0x1c, 0x1c, 0x1d, 0x1d, 0x00, 0x01,
-			       0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
-			       0x08, 0x09, 0x0a, 0x0b, 0x24, 0x25,
-			       0x18, 0x18, 0x26, 0x27, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x20, 0x21, 0x18, 0x18,
-			       0x18, 0x18);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP1,
+				     0x1c, 0x1c, 0x1d, 0x1d, 0x00, 0x01,
+				     0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
+				     0x08, 0x09, 0x0a, 0x0b, 0x24, 0x25,
+				     0x18, 0x18, 0x26, 0x27, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x20, 0x21, 0x18, 0x18,
+				     0x18, 0x18);
 
 	/* 5.19.21 Set GIP Option2 (D6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGIP2,
-			       0x1c, 0x1c, 0x1d, 0x1d, 0x07, 0x06,
-			       0x05, 0x04, 0x03, 0x02, 0x01, 0x00,
-			       0x0b, 0x0a, 0x09, 0x08, 0x21, 0x20,
-			       0x18, 0x18, 0x27, 0x26, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
-			       0x18, 0x18, 0x25, 0x24, 0x18, 0x18,
-			       0x18, 0x18);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGIP2,
+				     0x1c, 0x1c, 0x1d, 0x1d, 0x07, 0x06,
+				     0x05, 0x04, 0x03, 0x02, 0x01, 0x00,
+				     0x0b, 0x0a, 0x09, 0x08, 0x21, 0x20,
+				     0x18, 0x18, 0x27, 0x26, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x18, 0x18, 0x18, 0x18,
+				     0x18, 0x18, 0x25, 0x24, 0x18, 0x18,
+				     0x18, 0x18);
 
 	/* 5.19.25 SETGAMMA: Set gamma curve related setting (E0h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETGAMMA,
-			       0x00, 0x0a, 0x15, 0x1b, 0x1e, 0x21,
-			       0x24, 0x22, 0x47, 0x56, 0x65, 0x66,
-			       0x6e, 0x82, 0x88, 0x8b, 0x9a, 0x9d,
-			       0x98, 0xa8, 0xb9, 0x5d, 0x5c, 0x61,
-			       0x66, 0x6a, 0x6f, 0x7f, 0x7f, 0x00,
-			       0x0a, 0x15, 0x1b, 0x1e, 0x21, 0x24,
-			       0x22, 0x47, 0x56, 0x65, 0x65, 0x6e,
-			       0x81, 0x87, 0x8b, 0x98, 0x9d, 0x99,
-			       0xa8, 0xba, 0x5d, 0x5d, 0x62, 0x67,
-			       0x6b, 0x72, 0x7f, 0x7f);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETGAMMA,
+				     0x00, 0x0a, 0x15, 0x1b, 0x1e, 0x21,
+				     0x24, 0x22, 0x47, 0x56, 0x65, 0x66,
+				     0x6e, 0x82, 0x88, 0x8b, 0x9a, 0x9d,
+				     0x98, 0xa8, 0xb9, 0x5d, 0x5c, 0x61,
+				     0x66, 0x6a, 0x6f, 0x7f, 0x7f, 0x00,
+				     0x0a, 0x15, 0x1b, 0x1e, 0x21, 0x24,
+				     0x22, 0x47, 0x56, 0x65, 0x65, 0x6e,
+				     0x81, 0x87, 0x8b, 0x98, 0x9d, 0x99,
+				     0xa8, 0xba, 0x5d, 0x5d, 0x62, 0x67,
+				     0x6b, 0x72, 0x7f, 0x7f);
 
 	/* Unknown command, not listed in the HX8394-F datasheet (C0H) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN1,
-			       0x1f, 0x73);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN1,
+				     0x1f, 0x73);
 
 	/* Set CABC control (C9h)*/
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETCABC,
-			       0x76, 0x00, 0x30);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETCABC,
+				     0x76, 0x00, 0x30);
 
 	/* 5.19.17 SETPANEL (CCh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPANEL,
-			       0x0b);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPANEL,
+				     0x0b);
 
 	/* Unknown command, not listed in the HX8394-F datasheet (D4h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN3,
-			       0x02);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN3,
+				     0x02);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x02);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x02);
 
 	/* 5.19.11 Set register bank (D8h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN4,
-			       0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
-			       0xff, 0xff, 0xff, 0xff, 0xff, 0xff);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN4,
+				     0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+				     0xff, 0xff, 0xff, 0xff, 0xff, 0xff);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x00);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x01);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x01);
 
 	/* 5.19.2 SETPOWER: Set power (B1h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETPOWER,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETPOWER,
+				     0x00);
 
 	/* 5.19.11 Set register bank (BDh) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_SETREGBANK,
-			       0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_SETREGBANK,
+				     0x00);
 
 	/* Unknown command, not listed in the HX8394-F datasheet (C6h) */
-	mipi_dsi_dcs_write_seq(dsi, HX8394_CMD_UNKNOWN2,
-			       0xed);
-
-	return 0;
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, HX8394_CMD_UNKNOWN2,
+				     0xed);
 }
 
 static const struct drm_display_mode mchp_ac40t08a_mode = {
@@ -493,35 +481,31 @@ static int hx8394_enable(struct drm_panel *panel)
 {
 	struct hx8394 *ctx = panel_to_hx8394(panel);
 	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 	int ret;
 
-	ret = ctx->desc->init_sequence(ctx);
-	if (ret) {
-		dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret);
-		return ret;
-	}
+	ctx->desc->init_sequence(&dsi_ctx);
 
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret) {
-		dev_err(ctx->dev, "Failed to exit sleep mode: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
 
+	if (dsi_ctx.accum_err)
+		return dsi_ctx.accum_err;
 	/* Panel is operational 120 msec after reset */
 	msleep(120);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret) {
-		dev_err(ctx->dev, "Failed to turn on the display: %d\n", ret);
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
+	if (dsi_ctx.accum_err)
 		goto sleep_in;
-	}
 
 	return 0;
 
 sleep_in:
+	ret = dsi_ctx.accum_err;
+	dsi_ctx.accum_err = 0;
+
 	/* This will probably fail, but let's try orderly power off anyway. */
-	if (!mipi_dsi_dcs_enter_sleep_mode(dsi))
-		msleep(50);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 50);
 
 	return ret;
 }
@@ -530,17 +514,12 @@ static int hx8394_disable(struct drm_panel *panel)
 {
 	struct hx8394 *ctx = panel_to_hx8394(panel);
 	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
-	int ret;
-
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret) {
-		dev_err(ctx->dev, "Failed to enter sleep mode: %d\n", ret);
-		return ret;
-	}
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
-	msleep(50); /* about 3 frames */
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 50); /* about 3 frames */
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int hx8394_unprepare(struct drm_panel *panel)
diff --git a/drivers/gpu/drm/panel/panel-jadard-jd9365da-h3.c b/drivers/gpu/drm/panel/panel-jadard-jd9365da-h3.c
index 7d68a8acfe2e..eb0f8373258c 100644
--- a/drivers/gpu/drm/panel/panel-jadard-jd9365da-h3.c
+++ b/drivers/gpu/drm/panel/panel-jadard-jd9365da-h3.c
@@ -129,11 +129,11 @@ static int jadard_unprepare(struct drm_panel *panel)
 {
 	struct jadard *jadard = panel_to_jadard(panel);
 
-	gpiod_set_value(jadard->reset, 1);
+	gpiod_set_value(jadard->reset, 0);
 	msleep(120);
 
 	if (jadard->desc->reset_before_power_off_vcioo) {
-		gpiod_set_value(jadard->reset, 0);
+		gpiod_set_value(jadard->reset, 1);
 
 		usleep_range(1000, 2000);
 	}
diff --git a/drivers/gpu/drm/panel/panel-newvision-nv3051d.c b/drivers/gpu/drm/panel/panel-newvision-nv3051d.c
index 5d115ecd5dd4..b6429795e8f5 100644
--- a/drivers/gpu/drm/panel/panel-newvision-nv3051d.c
+++ b/drivers/gpu/drm/panel/panel-newvision-nv3051d.c
@@ -413,15 +413,10 @@ static int panel_nv3051d_probe(struct mipi_dsi_device *dsi)
 static void panel_nv3051d_shutdown(struct mipi_dsi_device *dsi)
 {
 	struct panel_nv3051d *ctx = mipi_dsi_get_drvdata(dsi);
-	int ret;
 
-	ret = drm_panel_unprepare(&ctx->panel);
-	if (ret < 0)
-		dev_err(&dsi->dev, "Failed to unprepare panel: %d\n", ret);
+	drm_panel_unprepare(&ctx->panel);
 
-	ret = drm_panel_disable(&ctx->panel);
-	if (ret < 0)
-		dev_err(&dsi->dev, "Failed to disable panel: %d\n", ret);
+	drm_panel_disable(&ctx->panel);
 }
 
 static void panel_nv3051d_remove(struct mipi_dsi_device *dsi)
diff --git a/drivers/gpu/drm/panel/panel-novatek-nt36523.c b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
index 04f1d2676c78..116d67bfa114 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt36523.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
@@ -23,10 +23,12 @@
 
 #define DSI_NUM_MIN 1
 
-#define mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, cmd, seq...)        \
-		do {                                                 \
-			mipi_dsi_dcs_write_seq(dsi0, cmd, seq);      \
-			mipi_dsi_dcs_write_seq(dsi1, cmd, seq);      \
+#define mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, cmd, seq...)	\
+		do {								\
+			dsi_ctx.dsi = dsi0;					\
+			mipi_dsi_dcs_write_seq_multi(&dsi_ctx, cmd, seq);	\
+			dsi_ctx.dsi = dsi1;					\
+			mipi_dsi_dcs_write_seq_multi(&dsi_ctx, cmd, seq);	\
 		} while (0)
 
 struct panel_info {
@@ -67,868 +69,829 @@ static int elish_boe_init_sequence(struct panel_info *pinfo)
 {
 	struct mipi_dsi_device *dsi0 = pinfo->dsi[0];
 	struct mipi_dsi_device *dsi1 = pinfo->dsi[1];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = NULL };
 	/* No datasheet, so write magic init sequence directly */
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb9, 0x05);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x18, 0x40);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb9, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x23);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x00, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x01, 0x84);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x05, 0x2d);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x06, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x07, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x08, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x09, 0x45);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x11, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x12, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x15, 0x83);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x16, 0x0c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x29, 0x0a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x30, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x31, 0xfe);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x32, 0xfd);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x33, 0xfb);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x34, 0xf8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x35, 0xf5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x36, 0xf3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x37, 0xf2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x38, 0xf2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x39, 0xf2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3a, 0xef);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3b, 0xec);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3d, 0xe9);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3f, 0xe5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x40, 0xe5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x41, 0xe5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2a, 0x13);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x45, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x46, 0xf4);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x47, 0xe7);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x48, 0xda);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x49, 0xcd);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4a, 0xc0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4b, 0xb3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4c, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4d, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4e, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4f, 0x99);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x50, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x51, 0x68);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x52, 0x66);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x53, 0x66);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x54, 0x66);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2b, 0x0e);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x58, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x59, 0xfb);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5a, 0xf7);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5b, 0xf3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5c, 0xef);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5d, 0xe3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5e, 0xda);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5f, 0xd8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x60, 0xd8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x61, 0xd8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x62, 0xcb);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x63, 0xbf);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x64, 0xb3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x65, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x66, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x67, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x25, 0x47);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x30, 0x47);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x39, 0x47);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x19, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1a, 0xe0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1b, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1c, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2a, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2b, 0xe0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x84, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x85, 0x0c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x51, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x91, 0x1f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x92, 0x0f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x93, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x94, 0x18);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x95, 0x03);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x96, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb0, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x19, 0x1f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1b, 0x1b);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x24);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb8, 0x28);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x27);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd0, 0x31);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd1, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd2, 0x30);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd4, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xde, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xdf, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x00, 0x81);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x01, 0xb0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x22);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9f, 0x50);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x6f, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x70, 0x11);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x73, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x74, 0x49);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x76, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x77, 0x49);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xa0, 0x3f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xa9, 0x50);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xaa, 0x28);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xab, 0x28);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xad, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb8, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb9, 0x49);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xba, 0x49);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbb, 0x49);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbe, 0x04);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbf, 0x49);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc0, 0x04);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc1, 0x59);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc2, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc5, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc6, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc7, 0x48);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xca, 0x43);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xcb, 0x3c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xce, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xcf, 0x43);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd0, 0x3c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd3, 0x43);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd4, 0x3c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd7, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xdc, 0x43);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xdd, 0x3c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xe1, 0x43);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xe2, 0x3c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xf2, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xf3, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xf4, 0x48);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x13, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x14, 0x23);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbc, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbd, 0x23);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x97, 0x3c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x98, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x99, 0x95);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9a, 0x03);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9b, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9c, 0x0b);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9d, 0x0a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9e, 0x90);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x22);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9f, 0x50);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x23);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xa3, 0x50);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xe0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x14, 0x60);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x16, 0xc0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4f, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3a, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xd0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x02, 0xaf);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x09, 0xee);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1c, 0x99);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1d, 0x09);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x51, 0x0f, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x53, 0x2c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x35, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbb, 0x13);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x11);
-	msleep(70);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x29);
-
-	return 0;
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x05);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x18, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0x84);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x05, 0x2d);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x06, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x07, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x08, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0x45);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x12, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x15, 0x83);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x16, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29, 0x0a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x31, 0xfe);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x32, 0xfd);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x33, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x34, 0xf8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0xf5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x36, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x37, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x38, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0xec);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3d, 0xe9);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3f, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x40, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x41, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x45, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x46, 0xf4);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x47, 0xe7);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x48, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x49, 0xcd);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4a, 0xc0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4b, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4c, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4d, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4e, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x50, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x68);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x52, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x54, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0x0e);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x58, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x59, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5a, 0xf7);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5b, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5c, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5d, 0xe3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5e, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5f, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x60, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x61, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x62, 0xcb);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x63, 0xbf);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x64, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x65, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x66, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x67, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x25, 0x47);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0x47);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0x47);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1a, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x84, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x85, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x91, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x92, 0x0f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x93, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x94, 0x18);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x95, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x96, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb0, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x1b);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x24);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x27);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x31);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd1, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd2, 0x30);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xde, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdf, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x81);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x22);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9f, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x6f, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x70, 0x11);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x73, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x74, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x76, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x77, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa0, 0x3f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa9, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xaa, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xab, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xad, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xba, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbe, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbf, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc0, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc1, 0x59);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc2, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc5, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc6, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc7, 0x48);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xca, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcb, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xce, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcf, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd3, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd7, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdc, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdd, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xe1, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xe2, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xf2, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xf3, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xf4, 0x48);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x13, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x14, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbc, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbd, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x97, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x98, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x99, 0x95);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9a, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9b, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9c, 0x0b);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9d, 0x0a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9e, 0x90);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x22);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9f, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa3, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x14, 0x60);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x16, 0xc0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xd0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x02, 0xaf);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0xee);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1d, 0x09);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x0f, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x2c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11);
+	mipi_dsi_msleep(&dsi_ctx, 70);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29);
+
+	return dsi_ctx.accum_err;
 }
 
 static int elish_csot_init_sequence(struct panel_info *pinfo)
 {
 	struct mipi_dsi_device *dsi0 = pinfo->dsi[0];
 	struct mipi_dsi_device *dsi1 = pinfo->dsi[1];
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = NULL };
 	/* No datasheet, so write magic init sequence directly */
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb9, 0x05);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x18, 0x40);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb9, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xd0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x02, 0xaf);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x00, 0x30);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x09, 0xee);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1c, 0x99);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1d, 0x09);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3a, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xe0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4f, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x58, 0x40);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x35, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x23);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x00, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x01, 0x84);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x05, 0x2d);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x06, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x07, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x08, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x09, 0x45);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x11, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x12, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x15, 0x83);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x16, 0x0c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x29, 0x0a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x30, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x31, 0xfe);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x32, 0xfd);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x33, 0xfb);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x34, 0xf8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x35, 0xf5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x36, 0xf3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x37, 0xf2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x38, 0xf2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x39, 0xf2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3a, 0xef);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3b, 0xec);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3d, 0xe9);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3f, 0xe5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x40, 0xe5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x41, 0xe5);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2a, 0x13);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x45, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x46, 0xf4);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x47, 0xe7);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x48, 0xda);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x49, 0xcd);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4a, 0xc0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4b, 0xb3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4c, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4d, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4e, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x4f, 0x99);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x50, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x51, 0x68);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x52, 0x66);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x53, 0x66);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x54, 0x66);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2b, 0x0e);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x58, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x59, 0xfb);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5a, 0xf7);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5b, 0xf3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5c, 0xef);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5d, 0xe3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5e, 0xda);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x5f, 0xd8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x60, 0xd8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x61, 0xd8);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x62, 0xcb);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x63, 0xbf);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x64, 0xb3);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x65, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x66, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x67, 0xb2);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x51, 0x0f, 0xff);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x53, 0x2c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x55, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbb, 0x13);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x25, 0x46);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x30, 0x46);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x39, 0x46);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x01, 0xb0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x19, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1a, 0xe0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1b, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1c, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2a, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x2b, 0xe0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x84, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x85, 0x0c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x51, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x91, 0x1f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x92, 0x0f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x93, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x94, 0x18);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x95, 0x03);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x96, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb0, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x19, 0x1f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x1b, 0x1b);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x24);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb8, 0x28);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x27);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd0, 0x31);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd1, 0x20);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd4, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xde, 0x80);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xdf, 0x02);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x00, 0x81);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x01, 0xb0);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x22);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x6f, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x70, 0x11);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x73, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x74, 0x4d);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xa0, 0x3f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xa9, 0x50);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xaa, 0x28);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xab, 0x28);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xad, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb8, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xb9, 0x4b);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xba, 0x96);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbb, 0x4b);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbe, 0x07);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbf, 0x4b);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc0, 0x07);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc1, 0x5c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc2, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc5, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc6, 0x3f);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xc7, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xca, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xcb, 0x40);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xce, 0x00);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xcf, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd0, 0x40);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd3, 0x08);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xd4, 0x40);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbc, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xbd, 0x1c);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x9a, 0x03);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x11);
-	msleep(70);
-	mipi_dsi_dual_dcs_write_seq(dsi0, dsi1, 0x29);
-
-	return 0;
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x05);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x18, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xd0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x02, 0xaf);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x30);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0xee);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1d, 0x09);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x58, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0x84);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x05, 0x2d);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x06, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x07, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x08, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0x45);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x12, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x15, 0x83);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x16, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29, 0x0a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x31, 0xfe);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x32, 0xfd);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x33, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x34, 0xf8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0xf5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x36, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x37, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x38, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0xec);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3d, 0xe9);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3f, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x40, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x41, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x45, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x46, 0xf4);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x47, 0xe7);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x48, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x49, 0xcd);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4a, 0xc0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4b, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4c, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4d, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4e, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x50, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x68);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x52, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x54, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0x0e);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x58, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x59, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5a, 0xf7);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5b, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5c, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5d, 0xe3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5e, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5f, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x60, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x61, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x62, 0xcb);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x63, 0xbf);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x64, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x65, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x66, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x67, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x0f, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x2c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x55, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x25, 0x46);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0x46);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0x46);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1a, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x84, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x85, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x91, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x92, 0x0f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x93, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x94, 0x18);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x95, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x96, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb0, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x1b);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x24);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x27);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x31);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd1, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xde, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdf, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x81);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x22);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x6f, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x70, 0x11);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x73, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x74, 0x4d);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa0, 0x3f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa9, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xaa, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xab, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xad, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x4b);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xba, 0x96);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x4b);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbe, 0x07);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbf, 0x4b);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc0, 0x07);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc1, 0x5c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc2, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc5, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc6, 0x3f);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc7, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xca, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcb, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xce, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcf, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd3, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbc, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbd, 0x1c);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9a, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11);
+	mipi_dsi_msleep(&dsi_ctx, 70);
+	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29);
+
+	return dsi_ctx.accum_err;
 }
 
 static int j606f_boe_init_sequence(struct panel_info *pinfo)
 {
 	struct mipi_dsi_device *dsi = pinfo->dsi[0];
-	struct device *dev = &dsi->dev;
-	int ret;
-
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x20);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x05, 0xd9);
-	mipi_dsi_dcs_write_seq(dsi, 0x07, 0x78);
-	mipi_dsi_dcs_write_seq(dsi, 0x08, 0x5a);
-	mipi_dsi_dcs_write_seq(dsi, 0x0d, 0x63);
-	mipi_dsi_dcs_write_seq(dsi, 0x0e, 0x91);
-	mipi_dsi_dcs_write_seq(dsi, 0x0f, 0x73);
-	mipi_dsi_dcs_write_seq(dsi, 0x95, 0xeb);
-	mipi_dsi_dcs_write_seq(dsi, 0x96, 0xeb);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0x11);
-	mipi_dsi_dcs_write_seq(dsi, 0x6d, 0x66);
-	mipi_dsi_dcs_write_seq(dsi, 0x75, 0xa2);
-	mipi_dsi_dcs_write_seq(dsi, 0x77, 0xb3);
-	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d, 0x00,
-			       0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
-	mipi_dsi_dcs_write_seq(dsi, 0xb1, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e, 0x01,
-			       0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
-	mipi_dsi_dcs_write_seq(dsi, 0xb2, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08, 0x03,
-			       0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
-	mipi_dsi_dcs_write_seq(dsi, 0xb3, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7, 0x03,
-			       0xfd, 0x03, 0xff);
-	mipi_dsi_dcs_write_seq(dsi, 0xb4, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d, 0x00,
-			       0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
-	mipi_dsi_dcs_write_seq(dsi, 0xb5, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e, 0x01,
-			       0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
-	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08, 0x03,
-			       0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
-	mipi_dsi_dcs_write_seq(dsi, 0xb7, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7, 0x03,
-			       0xfd, 0x03, 0xff);
-	mipi_dsi_dcs_write_seq(dsi, 0xb8, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d, 0x00,
-			       0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
-	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e, 0x01,
-			       0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
-	mipi_dsi_dcs_write_seq(dsi, 0xba, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08, 0x03,
-			       0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
-	mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7, 0x03,
-			       0xfd, 0x03, 0xff);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x21);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65, 0x00,
-			       0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
-	mipi_dsi_dcs_write_seq(dsi, 0xb1, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76, 0x01,
-			       0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
-	mipi_dsi_dcs_write_seq(dsi, 0xb2, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00, 0x03,
-			       0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
-	mipi_dsi_dcs_write_seq(dsi, 0xb3, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf, 0x03,
-			       0xf5, 0x03, 0xf7);
-	mipi_dsi_dcs_write_seq(dsi, 0xb4, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65, 0x00,
-			       0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
-	mipi_dsi_dcs_write_seq(dsi, 0xb5, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76, 0x01,
-			       0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
-	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00, 0x03,
-			       0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
-	mipi_dsi_dcs_write_seq(dsi, 0xb7, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf, 0x03,
-			       0xf5, 0x03, 0xf7);
-	mipi_dsi_dcs_write_seq(dsi, 0xb8, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65, 0x00,
-			       0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
-	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76, 0x01,
-			       0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
-	mipi_dsi_dcs_write_seq(dsi, 0xba, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00, 0x03,
-			       0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
-	mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf, 0x03,
-			       0xf5, 0x03, 0xf7);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x23);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x00, 0x80);
-	mipi_dsi_dcs_write_seq(dsi, 0x07, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x11, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x12, 0x77);
-	mipi_dsi_dcs_write_seq(dsi, 0x15, 0x07);
-	mipi_dsi_dcs_write_seq(dsi, 0x16, 0x07);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x24);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x00, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x01, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x02, 0x1c);
-	mipi_dsi_dcs_write_seq(dsi, 0x03, 0x1c);
-	mipi_dsi_dcs_write_seq(dsi, 0x04, 0x1d);
-	mipi_dsi_dcs_write_seq(dsi, 0x05, 0x1d);
-	mipi_dsi_dcs_write_seq(dsi, 0x06, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0x07, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0x08, 0x0f);
-	mipi_dsi_dcs_write_seq(dsi, 0x09, 0x0f);
-	mipi_dsi_dcs_write_seq(dsi, 0x0a, 0x0e);
-	mipi_dsi_dcs_write_seq(dsi, 0x0b, 0x0e);
-	mipi_dsi_dcs_write_seq(dsi, 0x0c, 0x0d);
-	mipi_dsi_dcs_write_seq(dsi, 0x0d, 0x0d);
-	mipi_dsi_dcs_write_seq(dsi, 0x0e, 0x0c);
-	mipi_dsi_dcs_write_seq(dsi, 0x0f, 0x0c);
-	mipi_dsi_dcs_write_seq(dsi, 0x10, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0x11, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0x12, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x13, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x14, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x15, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x16, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x17, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x18, 0x1c);
-	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x1c);
-	mipi_dsi_dcs_write_seq(dsi, 0x1a, 0x1d);
-	mipi_dsi_dcs_write_seq(dsi, 0x1b, 0x1d);
-	mipi_dsi_dcs_write_seq(dsi, 0x1c, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0x1d, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0x1e, 0x0f);
-	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0x0f);
-	mipi_dsi_dcs_write_seq(dsi, 0x20, 0x0e);
-	mipi_dsi_dcs_write_seq(dsi, 0x21, 0x0e);
-	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x0d);
-	mipi_dsi_dcs_write_seq(dsi, 0x23, 0x0d);
-	mipi_dsi_dcs_write_seq(dsi, 0x24, 0x0c);
-	mipi_dsi_dcs_write_seq(dsi, 0x25, 0x0c);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0x27, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0x28, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x29, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_LUT, 0x20);
-	mipi_dsi_dcs_write_seq(dsi, 0x2f, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0x44);
-	mipi_dsi_dcs_write_seq(dsi, 0x33, 0x0c);
-	mipi_dsi_dcs_write_seq(dsi, 0x34, 0x32);
-	mipi_dsi_dcs_write_seq(dsi, 0x37, 0x44);
-	mipi_dsi_dcs_write_seq(dsi, 0x38, 0x40);
-	mipi_dsi_dcs_write_seq(dsi, 0x39, 0x00);
-
-	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x9a);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set pixel format: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0x3b, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_3D_CONTROL, 0x42);
-	mipi_dsi_dcs_write_seq(dsi, 0x3f, 0x06);
-	mipi_dsi_dcs_write_seq(dsi, 0x43, 0x06);
-	mipi_dsi_dcs_write_seq(dsi, 0x47, 0x66);
-	mipi_dsi_dcs_write_seq(dsi, 0x4a, 0x9a);
-	mipi_dsi_dcs_write_seq(dsi, 0x4b, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, 0x4c, 0x91);
-	mipi_dsi_dcs_write_seq(dsi, 0x4d, 0x21);
-	mipi_dsi_dcs_write_seq(dsi, 0x4e, 0x43);
-
-	ret = mipi_dsi_dcs_set_display_brightness(dsi, 18);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display brightness: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0x52, 0x34);
-	mipi_dsi_dcs_write_seq(dsi, 0x55, 0x82, 0x02);
-	mipi_dsi_dcs_write_seq(dsi, 0x56, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0x58, 0x21);
-	mipi_dsi_dcs_write_seq(dsi, 0x59, 0x30);
-	mipi_dsi_dcs_write_seq(dsi, 0x5a, 0xba);
-	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0x00, 0x06);
-	mipi_dsi_dcs_write_seq(dsi, 0x5f, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x65, 0x82);
-	mipi_dsi_dcs_write_seq(dsi, 0x7e, 0x20);
-	mipi_dsi_dcs_write_seq(dsi, 0x7f, 0x3c);
-	mipi_dsi_dcs_write_seq(dsi, 0x82, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0x97, 0xc0);
-	mipi_dsi_dcs_write_seq(dsi, 0xb6,
-			       0x05, 0x00, 0x05, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05,
-			       0x05, 0x00, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x92, 0xc4);
-	mipi_dsi_dcs_write_seq(dsi, 0x93, 0x1a);
-	mipi_dsi_dcs_write_seq(dsi, 0x94, 0x5f);
-	mipi_dsi_dcs_write_seq(dsi, 0xd7, 0x55);
-	mipi_dsi_dcs_write_seq(dsi, 0xda, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, 0xde, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0xdb, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xdc, 0xc4);
-	mipi_dsi_dcs_write_seq(dsi, 0xdd, 0x22);
-	mipi_dsi_dcs_write_seq(dsi, 0xdf, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe0, 0xc4);
-	mipi_dsi_dcs_write_seq(dsi, 0xe1, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe2, 0xc4);
-	mipi_dsi_dcs_write_seq(dsi, 0xe3, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe4, 0xc4);
-	mipi_dsi_dcs_write_seq(dsi, 0xe5, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe6, 0xc4);
-	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x88);
-	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0x8d, 0x88);
-	mipi_dsi_dcs_write_seq(dsi, 0x8e, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0xb5, 0x90);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x25);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x05, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x07);
-	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0xba);
-	mipi_dsi_dcs_write_seq(dsi, 0x20, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0xba);
-	mipi_dsi_dcs_write_seq(dsi, 0x27, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, 0x33, 0xba);
-	mipi_dsi_dcs_write_seq(dsi, 0x34, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, 0x3f, 0xe0);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_VSYNC_TIMING, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x44, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_GET_SCANLINE, 0x40);
-	mipi_dsi_dcs_write_seq(dsi, 0x48, 0xba);
-	mipi_dsi_dcs_write_seq(dsi, 0x49, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0xd0);
-	mipi_dsi_dcs_write_seq(dsi, 0x61, 0xba);
-	mipi_dsi_dcs_write_seq(dsi, 0x62, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, 0xf1, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x2a);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x64, 0x16);
-	mipi_dsi_dcs_write_seq(dsi, 0x67, 0x16);
-	mipi_dsi_dcs_write_seq(dsi, 0x6a, 0x16);
-	mipi_dsi_dcs_write_seq(dsi, 0x70, 0x30);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_READ_PPS_START, 0xf3);
-	mipi_dsi_dcs_write_seq(dsi, 0xa3, 0xff);
-	mipi_dsi_dcs_write_seq(dsi, 0xa4, 0xff);
-	mipi_dsi_dcs_write_seq(dsi, 0xa5, 0xff);
-	mipi_dsi_dcs_write_seq(dsi, 0xd6, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x26);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x00, 0xa1);
-	mipi_dsi_dcs_write_seq(dsi, 0x0a, 0xf2);
-	mipi_dsi_dcs_write_seq(dsi, 0x04, 0x28);
-	mipi_dsi_dcs_write_seq(dsi, 0x06, 0x30);
-	mipi_dsi_dcs_write_seq(dsi, 0x0c, 0x13);
-	mipi_dsi_dcs_write_seq(dsi, 0x0d, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, 0x0f, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, 0x11, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x12, 0x50);
-	mipi_dsi_dcs_write_seq(dsi, 0x13, 0x51);
-	mipi_dsi_dcs_write_seq(dsi, 0x14, 0x65);
-	mipi_dsi_dcs_write_seq(dsi, 0x15, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x16, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, 0x17, 0xa0);
-	mipi_dsi_dcs_write_seq(dsi, 0x18, 0x86);
-	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x11);
-	mipi_dsi_dcs_write_seq(dsi, 0x1a, 0x7b);
-	mipi_dsi_dcs_write_seq(dsi, 0x1b, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, 0x1c, 0xbb);
-	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x23, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x11);
-	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x7b);
-	mipi_dsi_dcs_write_seq(dsi, 0x1d, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x1e, 0xc3);
-	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0xc3);
-	mipi_dsi_dcs_write_seq(dsi, 0x24, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x25, 0xc3);
-	mipi_dsi_dcs_write_seq(dsi, 0x2f, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0xc3);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_COLUMNS, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x32, 0xc3);
-	mipi_dsi_dcs_write_seq(dsi, 0x39, 0x00);
-
-	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0xc3);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set pixel format: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0x20, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x33, 0x11);
-	mipi_dsi_dcs_write_seq(dsi, 0x34, 0x78);
-	mipi_dsi_dcs_write_seq(dsi, 0x35, 0x16);
-	mipi_dsi_dcs_write_seq(dsi, 0xc8, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0xc9, 0x82);
-	mipi_dsi_dcs_write_seq(dsi, 0xca, 0x4e);
-	mipi_dsi_dcs_write_seq(dsi, 0xcb, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_READ_PPS_CONTINUE, 0x4c);
-	mipi_dsi_dcs_write_seq(dsi, 0xaa, 0x47);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x27);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x56, 0x06);
-	mipi_dsi_dcs_write_seq(dsi, 0x58, 0x80);
-	mipi_dsi_dcs_write_seq(dsi, 0x59, 0x53);
-	mipi_dsi_dcs_write_seq(dsi, 0x5a, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x14);
-	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0x20);
-	mipi_dsi_dcs_write_seq(dsi, 0x5f, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, 0x60, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x61, 0x1d);
-	mipi_dsi_dcs_write_seq(dsi, 0x62, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x63, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x64, 0x24);
-	mipi_dsi_dcs_write_seq(dsi, 0x65, 0x1c);
-	mipi_dsi_dcs_write_seq(dsi, 0x66, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x67, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x68, 0x25);
-	mipi_dsi_dcs_write_seq(dsi, 0x00, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x78, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0xc3, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0xd1, 0x24);
-	mipi_dsi_dcs_write_seq(dsi, 0xd2, 0x30);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x2a);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x2f);
-	mipi_dsi_dcs_write_seq(dsi, 0x23, 0x08);
-	mipi_dsi_dcs_write_seq(dsi, 0x24, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x25, 0xc3);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0xf8);
-	mipi_dsi_dcs_write_seq(dsi, 0x27, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x28, 0x1a);
-	mipi_dsi_dcs_write_seq(dsi, 0x29, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x1a);
-	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_LUT, 0x1a);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0xe0);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x14, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x16, 0xc0);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0xf0);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-
-	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x08);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set pixel format: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x24);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-
-	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x5d);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set pixel format: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0x3b, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x4a, 0x5d);
-	mipi_dsi_dcs_write_seq(dsi, 0x4b, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x5a, 0x70);
-	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x91, 0x44);
-	mipi_dsi_dcs_write_seq(dsi, 0x92, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0xdb, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xdc, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0xdd, 0x22);
-	mipi_dsi_dcs_write_seq(dsi, 0xdf, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe0, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0xe1, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe2, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0xe3, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe4, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0xe5, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0xe6, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x8d, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x8e, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x25);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0x70);
-	mipi_dsi_dcs_write_seq(dsi, 0x20, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0x70);
-	mipi_dsi_dcs_write_seq(dsi, 0x27, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x33, 0x70);
-	mipi_dsi_dcs_write_seq(dsi, 0x34, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x48, 0x70);
-	mipi_dsi_dcs_write_seq(dsi, 0x49, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x61, 0x70);
-	mipi_dsi_dcs_write_seq(dsi, 0x62, 0x60);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x26);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x02, 0x31);
-	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, 0x1a, 0x7f);
-	mipi_dsi_dcs_write_seq(dsi, 0x1b, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, 0x1c, 0x0c);
-	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x0a);
-	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x7f);
-	mipi_dsi_dcs_write_seq(dsi, 0x1e, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0x25, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_COLUMNS, 0x05);
-	mipi_dsi_dcs_write_seq(dsi, 0x32, 0x8d);
-
-	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x75);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set pixel format: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x2a);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x25, 0x75);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x20);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0x18, 0x40);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x10);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x02);
-
-	ret = mipi_dsi_dcs_set_tear_on(dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set tear on: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x13);
-	mipi_dsi_dcs_write_seq(dsi, 0x3b, 0x03, 0x5f, 0x1a, 0x04, 0x04);
-	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x10);
-	usleep_range(10000, 11000);
-	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
-
-	ret = mipi_dsi_dcs_set_display_brightness(dsi, 0);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display brightness: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x2c);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_POWER_SAVE, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0x68, 0x05, 0x01);
-
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
-		return ret;
-	}
-	msleep(100);
-
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display on: %d\n", ret);
-		return ret;
-	}
-	msleep(30);
-
-	return 0;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x05, 0xd9);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x07, 0x78);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x08, 0x5a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0d, 0x63);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0e, 0x91);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0f, 0x73);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x95, 0xeb);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x96, 0xeb);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_PARTIAL_ROWS, 0x11);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6d, 0x66);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x75, 0xa2);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x77, 0xb3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb0, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d,
+				     0x00, 0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb1, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e,
+				     0x01, 0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb2, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08,
+				     0x03, 0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb3, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7,
+				     0x03, 0xfd, 0x03, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb4, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d,
+				     0x00, 0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb5, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e,
+				     0x01, 0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb6, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08,
+				     0x03, 0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb7, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7,
+				     0x03, 0xfd, 0x03, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb8, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d,
+				     0x00, 0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb9, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e,
+				     0x01, 0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xba, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08,
+				     0x03, 0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xbb, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7,
+				     0x03, 0xfd, 0x03, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x21);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb0, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65,
+				     0x00, 0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb1, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76,
+				     0x01, 0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb2, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00,
+				     0x03, 0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb3, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf,
+				     0x03, 0xf5, 0x03, 0xf7);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb4, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65,
+				     0x00, 0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb5, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76,
+				     0x01, 0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb6, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00,
+				     0x03, 0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb7, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf,
+				     0x03, 0xf5, 0x03, 0xf7);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb8, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65,
+				     0x00, 0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb9, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76,
+				     0x01, 0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xba, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00,
+				     0x03, 0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xbb, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf,
+				     0x03, 0xf5, 0x03, 0xf7);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x23);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x00, 0x80);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x07, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x11, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x12, 0x77);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x15, 0x07);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x16, 0x07);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x24);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x01, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x02, 0x1c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x03, 0x1c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x04, 0x1d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x05, 0x1d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x06, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x07, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x08, 0x0f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x09, 0x0f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0a, 0x0e);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0b, 0x0e);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0c, 0x0d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0d, 0x0d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0e, 0x0c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0f, 0x0c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x10, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x11, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x12, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x13, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x14, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x15, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x16, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x17, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x18, 0x1c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x19, 0x1c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1a, 0x1d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1b, 0x1d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1c, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1d, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1e, 0x0f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1f, 0x0f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x20, 0x0e);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x21, 0x0e);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x22, 0x0d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x23, 0x0d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x24, 0x0c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x25, 0x0c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_GAMMA_CURVE, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x27, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x28, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x29, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2a, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2b, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_LUT, 0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2f, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_PARTIAL_ROWS, 0x44);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x33, 0x0c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x34, 0x32);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x37, 0x44);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x38, 0x40);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x39, 0x00);
+
+	mipi_dsi_dcs_set_pixel_format_multi(&dsi_ctx, 0x9a);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x3b, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_3D_CONTROL, 0x42);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x3f, 0x06);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x43, 0x06);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x47, 0x66);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4a, 0x9a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4b, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4c, 0x91);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4d, 0x21);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4e, 0x43);
+
+	mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, 18);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x52, 0x34);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x55, 0x82, 0x02);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x56, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x58, 0x21);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x59, 0x30);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5a, 0xba);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5b, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0x00, 0x06);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5f, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x65, 0x82);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x7e, 0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x7f, 0x3c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x82, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x97, 0xc0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb6,
+				     0x05, 0x00, 0x05, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05,
+				     0x05, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x92, 0xc4);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x93, 0x1a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x94, 0x5f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xd7, 0x55);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xda, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xde, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdb, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdc, 0xc4);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdd, 0x22);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdf, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe0, 0xc4);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe1, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe2, 0xc4);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe3, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe4, 0xc4);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe5, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe6, 0xc4);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5c, 0x88);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5d, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x8d, 0x88);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x8e, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb5, 0x90);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x25);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x05, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x19, 0x07);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1f, 0xba);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x20, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_GAMMA_CURVE, 0xba);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x27, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x33, 0xba);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x34, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x3f, 0xe0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_VSYNC_TIMING, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x44, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_GET_SCANLINE, 0x40);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x48, 0xba);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x49, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5b, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5c, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5d, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0xd0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x61, 0xba);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x62, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf1, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x2a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x64, 0x16);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x67, 0x16);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6a, 0x16);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x70, 0x30);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_READ_PPS_START, 0xf3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xa3, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xa4, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xa5, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xd6, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x26);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x00, 0xa1);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0a, 0xf2);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x04, 0x28);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x06, 0x30);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0c, 0x13);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0d, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x0f, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x11, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x12, 0x50);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x13, 0x51);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x14, 0x65);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x15, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x16, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x17, 0xa0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x18, 0x86);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x19, 0x11);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1a, 0x7b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1b, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1c, 0xbb);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x22, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x23, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2a, 0x11);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2b, 0x7b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1d, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1e, 0xc3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1f, 0xc3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x24, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x25, 0xc3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2f, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_PARTIAL_ROWS, 0xc3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_PARTIAL_COLUMNS, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x32, 0xc3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x39, 0x00);
+
+	mipi_dsi_dcs_set_pixel_format_multi(&dsi_ctx, 0xc3);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x20, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x33, 0x11);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x34, 0x78);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x35, 0x16);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc8, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc9, 0x82);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xca, 0x4e);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xcb, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_READ_PPS_CONTINUE, 0x4c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xaa, 0x47);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x27);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x56, 0x06);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x58, 0x80);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x59, 0x53);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5a, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5b, 0x14);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5c, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5d, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5f, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x60, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x61, 0x1d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x62, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x63, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x64, 0x24);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x65, 0x1c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x66, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x67, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x68, 0x25);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x78, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc3, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xd1, 0x24);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xd2, 0x30);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x2a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x22, 0x2f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x23, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x24, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x25, 0xc3);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_GAMMA_CURVE, 0xf8);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x27, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x28, 0x1a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x29, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2a, 0x1a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2b, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_LUT, 0x1a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0xe0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x14, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x16, 0xc0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0xf0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+
+	mipi_dsi_dcs_set_pixel_format_multi(&dsi_ctx, 0x08);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x24);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+
+	mipi_dsi_dcs_set_pixel_format_multi(&dsi_ctx, 0x5d);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x3b, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4a, 0x5d);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4b, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5a, 0x70);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5b, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x91, 0x44);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x92, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdb, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdc, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdd, 0x22);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xdf, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe0, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe1, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe2, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe3, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe4, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe5, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xe6, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5c, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5d, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x8d, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x8e, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x25);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1f, 0x70);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x20, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_GAMMA_CURVE, 0x70);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x27, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x33, 0x70);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x34, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x48, 0x70);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x49, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5b, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x61, 0x70);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x62, 0x60);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x26);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x19, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1a, 0x7f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1b, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1c, 0x0c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2a, 0x0a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2b, 0x7f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1e, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x1f, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x25, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_PARTIAL_ROWS, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_PARTIAL_COLUMNS, 0x05);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x32, 0x8d);
+
+	mipi_dsi_dcs_set_pixel_format_multi(&dsi_ctx, 0x75);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x2a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x25, 0x75);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb9, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x18, 0x40);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb9, 0x02);
+
+	mipi_dsi_dcs_set_tear_on_multi(&dsi_ctx, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xbb, 0x13);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x3b, 0x03, 0x5f, 0x1a, 0x04, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0x10);
+	mipi_dsi_usleep_range(&dsi_ctx, 10000, 11000);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb, 0x01);
+
+	mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, 0);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x2c);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_POWER_SAVE, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x68, 0x05, 0x01);
+
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 100);
+
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 30);
+
+	return dsi_ctx.accum_err;
 }
 
 static const struct drm_display_mode elish_boe_modes[] = {
@@ -1063,18 +1026,18 @@ static int nt36523_prepare(struct drm_panel *panel)
 static int nt36523_disable(struct drm_panel *panel)
 {
 	struct panel_info *pinfo = to_panel_info(panel);
-	int i, ret;
+	int i;
 
 	for (i = 0; i < DSI_NUM_MIN + pinfo->desc->is_dual_dsi; i++) {
-		ret = mipi_dsi_dcs_set_display_off(pinfo->dsi[i]);
-		if (ret < 0)
-			dev_err(&pinfo->dsi[i]->dev, "failed to set display off: %d\n", ret);
+		struct mipi_dsi_multi_context dsi_ctx = { .dsi = pinfo->dsi[i]};
+
+		mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
 	}
 
 	for (i = 0; i < DSI_NUM_MIN + pinfo->desc->is_dual_dsi; i++) {
-		ret = mipi_dsi_dcs_enter_sleep_mode(pinfo->dsi[i]);
-		if (ret < 0)
-			dev_err(&pinfo->dsi[i]->dev, "failed to enter sleep mode: %d\n", ret);
+		struct mipi_dsi_multi_context dsi_ctx = { .dsi = pinfo->dsi[i]};
+
+		mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
 	}
 
 	msleep(70);
diff --git a/drivers/gpu/drm/panel/panel-novatek-nt37801.c b/drivers/gpu/drm/panel/panel-novatek-nt37801.c
new file mode 100644
index 000000000000..84d367eab058
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-novatek-nt37801.c
@@ -0,0 +1,340 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright (c) 2024 Linaro Limited
+
+#include <linux/backlight.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/regulator/consumer.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+
+#include <drm/display/drm_dsc.h>
+#include <drm/display/drm_dsc_helper.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
+
+#include <video/mipi_display.h>
+
+struct novatek_nt37801 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct drm_dsc_config dsc;
+	struct gpio_desc *reset_gpio;
+	struct regulator_bulk_data *supplies;
+};
+
+static const struct regulator_bulk_data novatek_nt37801_supplies[] = {
+	{ .supply = "vddio" },
+	{ .supply = "vci" },
+	{ .supply = "vdd" },
+};
+
+static inline struct novatek_nt37801 *to_novatek_nt37801(struct drm_panel *panel)
+{
+	return container_of(panel, struct novatek_nt37801, panel);
+}
+
+static void novatek_nt37801_reset(struct novatek_nt37801 *ctx)
+{
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	usleep_range(10000, 21000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(10000, 21000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	usleep_range(10000, 21000);
+}
+
+#define NT37801_DCS_SWITCH_PAGE			0xf0
+
+#define novatek_nt37801_switch_page(dsi_ctx, page) \
+	mipi_dsi_dcs_write_seq_multi((dsi_ctx), NT37801_DCS_SWITCH_PAGE, \
+				     0x55, 0xaa, 0x52, 0x08, (page))
+
+static int novatek_nt37801_on(struct novatek_nt37801 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	novatek_nt37801_switch_page(&dsi_ctx, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc5, 0x0b, 0x0b, 0x0b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0xaa, 0x55, 0xa5, 0x80);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x02);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf5, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x1b);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf4, 0x55);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x18);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf8, 0x19);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x0f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfc, 0x00);
+	mipi_dsi_dcs_set_column_address_multi(&dsi_ctx, 0x0000, 0x059f);
+	mipi_dsi_dcs_set_page_address_multi(&dsi_ctx, 0x0000, 0x0c7f);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x90, 0x03, 0x03);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x91,
+				     0x89, 0x28, 0x00, 0x28, 0xc2, 0x00, 0x02,
+				     0x68, 0x04, 0x6c, 0x00, 0x0a, 0x02, 0x77,
+				     0x01, 0xe9, 0x10, 0xf0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xff, 0xaa, 0x55, 0xa5, 0x81);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x23);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfb,
+				     0x00, 0x01, 0x00, 0x11, 0x33, 0x33, 0x33,
+				     0x55, 0x57, 0xd0, 0x00, 0x00, 0x44, 0x56,
+				     0x77, 0x78, 0x9a, 0xbc, 0xdd, 0xf0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x6f, 0x06);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf3, 0xdc);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_SET_GAMMA_CURVE, 0x00);
+	mipi_dsi_dcs_set_tear_on_multi(&dsi_ctx, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x3b, 0x00, 0x18, 0x00, 0x10);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY,
+				     0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x51,
+				     0x07, 0xff, 0x07, 0xff, 0x0f, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5a, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x5f, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x9c, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_MEMORY_START);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x2f, 0x00);
+
+	novatek_nt37801_switch_page(&dsi_ctx, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb2, 0x55, 0x01, 0xff, 0x03);
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 120);
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 20);
+
+	return dsi_ctx.accum_err;
+}
+
+static int novatek_nt37801_off(struct novatek_nt37801 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 20);
+
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 120);
+
+	return dsi_ctx.accum_err;
+}
+
+static int novatek_nt37801_prepare(struct drm_panel *panel)
+{
+	struct novatek_nt37801 *ctx = to_novatek_nt37801(panel);
+	struct device *dev = &ctx->dsi->dev;
+	struct drm_dsc_picture_parameter_set pps;
+	int ret;
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(novatek_nt37801_supplies),
+				    ctx->supplies);
+	if (ret < 0)
+		return ret;
+
+	novatek_nt37801_reset(ctx);
+
+	ret = novatek_nt37801_on(ctx);
+	if (ret < 0)
+		goto err;
+
+	drm_dsc_pps_payload_pack(&pps, &ctx->dsc);
+
+	ret = mipi_dsi_picture_parameter_set(ctx->dsi, &pps);
+	if (ret < 0) {
+		dev_err(panel->dev, "failed to transmit PPS: %d\n", ret);
+		goto err;
+	}
+
+	ret = mipi_dsi_compression_mode(ctx->dsi, true);
+	if (ret < 0) {
+		dev_err(dev, "failed to enable compression mode: %d\n", ret);
+		goto err;
+	}
+
+	msleep(28);
+
+	return 0;
+
+err:
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	regulator_bulk_disable(ARRAY_SIZE(novatek_nt37801_supplies),
+			       ctx->supplies);
+
+	return ret;
+}
+
+static int novatek_nt37801_unprepare(struct drm_panel *panel)
+{
+	struct novatek_nt37801 *ctx = to_novatek_nt37801(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = novatek_nt37801_off(ctx);
+	if (ret < 0)
+		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+
+	regulator_bulk_disable(ARRAY_SIZE(novatek_nt37801_supplies),
+			       ctx->supplies);
+
+	return 0;
+}
+
+static const struct drm_display_mode novatek_nt37801_mode = {
+	.clock = (1440 + 20 + 4 + 20) * (3200 + 20 + 2 + 18) * 120 / 1000,
+	.hdisplay = 1440,
+	.hsync_start = 1440 + 20,
+	.hsync_end = 1440 + 20 + 4,
+	.htotal = 1440 + 20 + 4 + 20,
+	.vdisplay = 3200,
+	.vsync_start = 3200 + 20,
+	.vsync_end = 3200 + 20 + 2,
+	.vtotal = 3200 + 20 + 2 + 18,
+	.type = DRM_MODE_TYPE_DRIVER,
+};
+
+static int novatek_nt37801_get_modes(struct drm_panel *panel,
+				     struct drm_connector *connector)
+{
+	return drm_connector_helper_get_modes_fixed(connector,
+						    &novatek_nt37801_mode);
+}
+
+static const struct drm_panel_funcs novatek_nt37801_panel_funcs = {
+	.prepare = novatek_nt37801_prepare,
+	.unprepare = novatek_nt37801_unprepare,
+	.get_modes = novatek_nt37801_get_modes,
+};
+
+static int novatek_nt37801_bl_update_status(struct backlight_device *bl)
+{
+	struct mipi_dsi_device *dsi = bl_get_data(bl);
+	u16 brightness = backlight_get_brightness(bl);
+	int ret;
+
+	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	ret = mipi_dsi_dcs_set_display_brightness_large(dsi, brightness);
+	if (ret < 0)
+		return ret;
+
+	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	return 0;
+}
+
+static const struct backlight_ops novatek_nt37801_bl_ops = {
+	.update_status = novatek_nt37801_bl_update_status,
+};
+
+static struct backlight_device *
+novatek_nt37801_create_backlight(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	const struct backlight_properties props = {
+		.type = BACKLIGHT_RAW,
+		.brightness = 4095,
+		.max_brightness = 4095,
+	};
+
+	return devm_backlight_device_register(dev, dev_name(dev), dev, dsi,
+					      &novatek_nt37801_bl_ops, &props);
+}
+
+static int novatek_nt37801_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct novatek_nt37801 *ctx;
+	int ret;
+
+	ctx = devm_drm_panel_alloc(dev, struct novatek_nt37801, panel,
+				   &novatek_nt37801_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (!ctx)
+		return -ENOMEM;
+
+	ret = devm_regulator_bulk_get_const(dev,
+					    ARRAY_SIZE(novatek_nt37801_supplies),
+					    novatek_nt37801_supplies,
+					    &ctx->supplies);
+	if (ret < 0)
+		return ret;
+
+	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio),
+				     "Failed to get reset-gpios\n");
+
+	ctx->dsi = dsi;
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_NO_EOT_PACKET | MIPI_DSI_CLOCK_NON_CONTINUOUS;
+
+	ctx->panel.prepare_prev_first = true;
+	ctx->panel.backlight = novatek_nt37801_create_backlight(dsi);
+	if (IS_ERR(ctx->panel.backlight))
+		return dev_err_probe(dev, PTR_ERR(ctx->panel.backlight),
+				     "Failed to create backlight\n");
+
+	drm_panel_add(&ctx->panel);
+
+	/* This panel only supports DSC; unconditionally enable it */
+	dsi->dsc = &ctx->dsc;
+	ctx->dsc.dsc_version_major = 1;
+	ctx->dsc.dsc_version_minor = 1;
+	ctx->dsc.slice_height = 40;
+	ctx->dsc.slice_width = 720;
+	ctx->dsc.slice_count = 1440 / ctx->dsc.slice_width;
+	ctx->dsc.bits_per_component = 8;
+	ctx->dsc.bits_per_pixel = 8 << 4; /* 4 fractional bits */
+	ctx->dsc.block_pred_enable = true;
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		drm_panel_remove(&ctx->panel);
+		return dev_err_probe(dev, ret, "Failed to attach to DSI host\n");
+	}
+
+	return 0;
+}
+
+static void novatek_nt37801_remove(struct mipi_dsi_device *dsi)
+{
+	struct novatek_nt37801 *ctx = mipi_dsi_get_drvdata(dsi);
+	int ret;
+
+	ret = mipi_dsi_detach(dsi);
+	if (ret < 0)
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
+
+	drm_panel_remove(&ctx->panel);
+}
+
+static const struct of_device_id novatek_nt37801_of_match[] = {
+	{ .compatible = "novatek,nt37801" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, novatek_nt37801_of_match);
+
+static struct mipi_dsi_driver novatek_nt37801_driver = {
+	.probe = novatek_nt37801_probe,
+	.remove = novatek_nt37801_remove,
+	.driver = {
+		.name = "panel-novatek-nt37801",
+		.of_match_table = novatek_nt37801_of_match,
+	},
+};
+module_mipi_dsi_driver(novatek_nt37801_driver);
+
+MODULE_AUTHOR("Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>");
+MODULE_DESCRIPTION("Panel driver for the Novatek NT37801/NT37810 AMOLED DSI panel");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
index f23d8832a1ad..93f11e2e9398 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
@@ -34,8 +34,8 @@ struct s6d7aa0 {
 
 struct s6d7aa0_panel_desc {
 	unsigned int panel_type;
-	int (*init_func)(struct s6d7aa0 *ctx);
-	int (*off_func)(struct s6d7aa0 *ctx);
+	void (*init_func)(struct s6d7aa0 *ctx, struct mipi_dsi_multi_context *dsi_ctx);
+	void (*off_func)(struct mipi_dsi_multi_context *dsi_ctx);
 	const struct drm_display_mode *drm_mode;
 	unsigned long mode_flags;
 	u32 bus_flags;
@@ -62,93 +62,61 @@ static void s6d7aa0_reset(struct s6d7aa0 *ctx)
 	msleep(50);
 }
 
-static int s6d7aa0_lock(struct s6d7aa0 *ctx, bool lock)
+static void s6d7aa0_lock(struct s6d7aa0 *ctx, struct mipi_dsi_multi_context *dsi_ctx, bool lock)
 {
-	struct mipi_dsi_device *dsi = ctx->dsi;
-
 	if (lock) {
-		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD1, 0xa5, 0xa5);
-		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD2, 0xa5, 0xa5);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_PASSWD1, 0xa5, 0xa5);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_PASSWD2, 0xa5, 0xa5);
 		if (ctx->desc->use_passwd3)
-			mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD3, 0x5a, 0x5a);
+			mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_PASSWD3, 0x5a, 0x5a);
 	} else {
-		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD1, 0x5a, 0x5a);
-		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD2, 0x5a, 0x5a);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_PASSWD1, 0x5a, 0x5a);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_PASSWD2, 0x5a, 0x5a);
 		if (ctx->desc->use_passwd3)
-			mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD3, 0xa5, 0xa5);
+			mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_PASSWD3, 0xa5, 0xa5);
 	}
-
-	return 0;
 }
 
 static int s6d7aa0_on(struct s6d7aa0 *ctx)
 {
 	struct mipi_dsi_device *dsi = ctx->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
-	ret = ctx->desc->init_func(ctx);
-	if (ret < 0) {
-		dev_err(dev, "Failed to initialize panel: %d\n", ret);
-		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
-		return ret;
-	}
+	ctx->desc->init_func(ctx, &dsi_ctx);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display on: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
-static int s6d7aa0_off(struct s6d7aa0 *ctx)
+static void s6d7aa0_off(struct s6d7aa0 *ctx)
 {
 	struct mipi_dsi_device *dsi = ctx->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
-	ret = ctx->desc->off_func(ctx);
-	if (ret < 0) {
-		dev_err(dev, "Panel-specific off function failed: %d\n", ret);
-		return ret;
-	}
+	ctx->desc->off_func(&dsi_ctx);
 
-	ret = mipi_dsi_dcs_set_display_off(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display off: %d\n", ret);
-		return ret;
-	}
-	msleep(64);
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 64);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to enter sleep mode: %d\n", ret);
-		return ret;
-	}
-	msleep(120);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
 
-	return 0;
+	mipi_dsi_msleep(&dsi_ctx, 120);
 }
 
 static int s6d7aa0_prepare(struct drm_panel *panel)
 {
 	struct s6d7aa0 *ctx = panel_to_s6d7aa0(panel);
-	struct device *dev = &ctx->dsi->dev;
 	int ret;
 
 	ret = regulator_bulk_enable(ARRAY_SIZE(ctx->supplies), ctx->supplies);
-	if (ret < 0) {
-		dev_err(dev, "Failed to enable regulators: %d\n", ret);
+	if (ret < 0)
 		return ret;
-	}
 
 	s6d7aa0_reset(ctx);
 
 	ret = s6d7aa0_on(ctx);
 	if (ret < 0) {
-		dev_err(dev, "Failed to initialize panel: %d\n", ret);
 		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
 		return ret;
 	}
@@ -159,12 +127,8 @@ static int s6d7aa0_prepare(struct drm_panel *panel)
 static int s6d7aa0_disable(struct drm_panel *panel)
 {
 	struct s6d7aa0 *ctx = panel_to_s6d7aa0(panel);
-	struct device *dev = &ctx->dsi->dev;
-	int ret;
 
-	ret = s6d7aa0_off(ctx);
-	if (ret < 0)
-		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
+	s6d7aa0_off(ctx);
 
 	return 0;
 }
@@ -185,13 +149,11 @@ static int s6d7aa0_bl_update_status(struct backlight_device *bl)
 {
 	struct mipi_dsi_device *dsi = bl_get_data(bl);
 	u16 brightness = backlight_get_brightness(bl);
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
-	ret = mipi_dsi_dcs_set_display_brightness(dsi, brightness);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, brightness);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int s6d7aa0_bl_get_brightness(struct backlight_device *bl)
@@ -228,65 +190,39 @@ s6d7aa0_create_backlight(struct mipi_dsi_device *dsi)
 
 /* Initialization code and structures for LSL080AL02 panel */
 
-static int s6d7aa0_lsl080al02_init(struct s6d7aa0 *ctx)
+static void s6d7aa0_lsl080al02_init(struct s6d7aa0 *ctx, struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = ctx->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	mipi_dsi_usleep_range(dsi_ctx, 20000, 25000);
 
-	usleep_range(20000, 25000);
+	s6d7aa0_lock(ctx, dsi_ctx, false);
 
-	ret = s6d7aa0_lock(ctx, false);
-	if (ret < 0) {
-		dev_err(dev, "Failed to unlock registers: %d\n", ret);
-		return ret;
-	}
-
-	mipi_dsi_dcs_write_seq(dsi, MCS_OTP_RELOAD, 0x00, 0x10);
-	usleep_range(1000, 1500);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_OTP_RELOAD, 0x00, 0x10);
+	mipi_dsi_usleep_range(dsi_ctx, 1000, 1500);
 
 	/* SEQ_B6_PARAM_8_R01 */
-	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x10);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xb6, 0x10);
 
 	/* BL_CTL_ON */
-	mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0x40, 0x00, 0x28);
-
-	usleep_range(5000, 6000);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_BL_CTL, 0x40, 0x00, 0x28);
 
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_ADDRESS_MODE, 0x04);
+	mipi_dsi_usleep_range(dsi_ctx, 5000, 6000);
 
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, MIPI_DCS_SET_ADDRESS_MODE, 0x04);
 
-	msleep(120);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_ADDRESS_MODE, 0x00);
+	mipi_dsi_dcs_exit_sleep_mode_multi(dsi_ctx);
 
-	ret = s6d7aa0_lock(ctx, true);
-	if (ret < 0) {
-		dev_err(dev, "Failed to lock registers: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_msleep(dsi_ctx, 120);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, MIPI_DCS_SET_ADDRESS_MODE, 0x00);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display on: %d\n", ret);
-		return ret;
-	}
+	s6d7aa0_lock(ctx, dsi_ctx, true);
 
-	return 0;
+	mipi_dsi_dcs_set_display_on_multi(dsi_ctx);
 }
 
-static int s6d7aa0_lsl080al02_off(struct s6d7aa0 *ctx)
+static void s6d7aa0_lsl080al02_off(struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = ctx->dsi;
-
 	/* BL_CTL_OFF */
-	mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0x40, 0x00, 0x20);
-
-	return 0;
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_BL_CTL, 0x40, 0x00, 0x20);
 }
 
 static const struct drm_display_mode s6d7aa0_lsl080al02_mode = {
@@ -317,79 +253,51 @@ static const struct s6d7aa0_panel_desc s6d7aa0_lsl080al02_desc = {
 
 /* Initialization code and structures for LSL080AL03 panel */
 
-static int s6d7aa0_lsl080al03_init(struct s6d7aa0 *ctx)
+static void s6d7aa0_lsl080al03_init(struct s6d7aa0 *ctx, struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = ctx->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
-
-	usleep_range(20000, 25000);
+	mipi_dsi_usleep_range(dsi_ctx, 20000, 25000);
 
-	ret = s6d7aa0_lock(ctx, false);
-	if (ret < 0) {
-		dev_err(dev, "Failed to unlock registers: %d\n", ret);
-		return ret;
-	}
+	s6d7aa0_lock(ctx, dsi_ctx, false);
 
 	if (ctx->desc->panel_type == S6D7AA0_PANEL_LSL080AL03) {
-		mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0xc7, 0x00, 0x29);
-		mipi_dsi_dcs_write_seq(dsi, 0xbc, 0x01, 0x4e, 0xa0);
-		mipi_dsi_dcs_write_seq(dsi, 0xfd, 0x16, 0x10, 0x11, 0x23,
-				       0x09);
-		mipi_dsi_dcs_write_seq(dsi, 0xfe, 0x00, 0x02, 0x03, 0x21,
-				       0x80, 0x78);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_BL_CTL, 0xc7, 0x00, 0x29);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xbc, 0x01, 0x4e, 0xa0);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xfd, 0x16, 0x10, 0x11, 0x23,
+					     0x09);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xfe, 0x00, 0x02, 0x03, 0x21,
+					     0x80, 0x78);
 	} else if (ctx->desc->panel_type == S6D7AA0_PANEL_LTL101AT01) {
-		mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0x40, 0x00, 0x08);
-		mipi_dsi_dcs_write_seq(dsi, 0xbc, 0x01, 0x4e, 0x0b);
-		mipi_dsi_dcs_write_seq(dsi, 0xfd, 0x16, 0x10, 0x11, 0x23,
-				       0x09);
-		mipi_dsi_dcs_write_seq(dsi, 0xfe, 0x00, 0x02, 0x03, 0x21,
-				       0x80, 0x68);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, MCS_BL_CTL, 0x40, 0x00, 0x08);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xbc, 0x01, 0x4e, 0x0b);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xfd, 0x16, 0x10, 0x11, 0x23,
+					     0x09);
+		mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xfe, 0x00, 0x02, 0x03, 0x21,
+					     0x80, 0x68);
 	}
 
-	mipi_dsi_dcs_write_seq(dsi, 0xb3, 0x51);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x24);
-	mipi_dsi_dcs_write_seq(dsi, 0xf2, 0x02, 0x08, 0x08);
-
-	usleep_range(10000, 11000);
-
-	mipi_dsi_dcs_write_seq(dsi, 0xc0, 0x80, 0x80, 0x30);
-	mipi_dsi_dcs_write_seq(dsi, 0xcd,
-			       0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
-			       0x2e, 0x2e, 0x2e, 0x2e, 0x2e);
-	mipi_dsi_dcs_write_seq(dsi, 0xce,
-			       0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
-			       0x00, 0x00, 0x00, 0x00, 0x00);
-	mipi_dsi_dcs_write_seq(dsi, 0xc1, 0x03);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xb3, 0x51);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x24);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xf2, 0x02, 0x08, 0x08);
 
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
-		return ret;
-	}
-
-	ret = s6d7aa0_lock(ctx, true);
-	if (ret < 0) {
-		dev_err(dev, "Failed to lock registers: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_usleep_range(dsi_ctx, 10000, 11000);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display on: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xc0, 0x80, 0x80, 0x30);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xcd,
+				     0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
+				     0x2e, 0x2e, 0x2e, 0x2e, 0x2e);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xce,
+				     0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+				     0x00, 0x00, 0x00, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0xc1, 0x03);
 
-	return 0;
+	mipi_dsi_dcs_exit_sleep_mode_multi(dsi_ctx);
+	s6d7aa0_lock(ctx, dsi_ctx, true);
+	mipi_dsi_dcs_set_display_on_multi(dsi_ctx);
 }
 
-static int s6d7aa0_lsl080al03_off(struct s6d7aa0 *ctx)
+static void s6d7aa0_lsl080al03_off(struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = ctx->dsi;
-
-	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x00);
-
-	return 0;
+	mipi_dsi_dcs_write_seq_multi(dsi_ctx, 0x22, 0x00);
 }
 
 static const struct drm_display_mode s6d7aa0_lsl080al03_mode = {
diff --git a/drivers/gpu/drm/panel/panel-samsung-sofef00.c b/drivers/gpu/drm/panel/panel-samsung-sofef00.c
index 04ce925b3d9d..210a25afe82b 100644
--- a/drivers/gpu/drm/panel/panel-samsung-sofef00.c
+++ b/drivers/gpu/drm/panel/panel-samsung-sofef00.c
@@ -22,7 +22,6 @@ struct sofef00_panel {
 	struct mipi_dsi_device *dsi;
 	struct regulator *supply;
 	struct gpio_desc *reset_gpio;
-	const struct drm_display_mode *mode;
 };
 
 static inline
@@ -44,66 +43,44 @@ static void sofef00_panel_reset(struct sofef00_panel *ctx)
 static int sofef00_panel_on(struct sofef00_panel *ctx)
 {
 	struct mipi_dsi_device *dsi = ctx->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
 	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
 
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
-		return ret;
-	}
-	usleep_range(10000, 11000);
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_usleep_range(&dsi_ctx, 10000, 11000);
 
-	mipi_dsi_dcs_write_seq(dsi, 0xf0, 0x5a, 0x5a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf0, 0x5a, 0x5a);
 
-	ret = mipi_dsi_dcs_set_tear_on(dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set tear on: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_set_tear_on_multi(&dsi_ctx, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
 
-	mipi_dsi_dcs_write_seq(dsi, 0xf0, 0xa5, 0xa5);
-	mipi_dsi_dcs_write_seq(dsi, 0xf0, 0x5a, 0x5a);
-	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0x07);
-	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x12);
-	mipi_dsi_dcs_write_seq(dsi, 0xf0, 0xa5, 0xa5);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x20);
-	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_POWER_SAVE, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf0, 0xa5, 0xa5);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf0, 0x5a, 0x5a);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb0, 0x07);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb6, 0x12);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf0, 0xa5, 0xa5);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x20);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_POWER_SAVE, 0x00);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display on: %d\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int sofef00_panel_off(struct sofef00_panel *ctx)
 {
 	struct mipi_dsi_device *dsi = ctx->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
 	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
 
-	ret = mipi_dsi_dcs_set_display_off(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display off: %d\n", ret);
-		return ret;
-	}
-	msleep(40);
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 40);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to enter sleep mode: %d\n", ret);
-		return ret;
-	}
-	msleep(160);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 160);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int sofef00_panel_prepare(struct drm_panel *panel)
@@ -122,7 +99,6 @@ static int sofef00_panel_prepare(struct drm_panel *panel)
 
 	ret = sofef00_panel_on(ctx);
 	if (ret < 0) {
-		dev_err(dev, "Failed to initialize panel: %d\n", ret);
 		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
 		return ret;
 	}
@@ -133,13 +109,8 @@ static int sofef00_panel_prepare(struct drm_panel *panel)
 static int sofef00_panel_unprepare(struct drm_panel *panel)
 {
 	struct sofef00_panel *ctx = to_sofef00_panel(panel);
-	struct device *dev = &ctx->dsi->dev;
-	int ret;
-
-	ret = sofef00_panel_off(ctx);
-	if (ret < 0)
-		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
 
+	sofef00_panel_off(ctx);
 	regulator_disable(ctx->supply);
 
 	return 0;
@@ -159,26 +130,11 @@ static const struct drm_display_mode enchilada_panel_mode = {
 	.height_mm = 145,
 };
 
-static const struct drm_display_mode fajita_panel_mode = {
-	.clock = (1080 + 72 + 16 + 36) * (2340 + 32 + 4 + 18) * 60 / 1000,
-	.hdisplay = 1080,
-	.hsync_start = 1080 + 72,
-	.hsync_end = 1080 + 72 + 16,
-	.htotal = 1080 + 72 + 16 + 36,
-	.vdisplay = 2340,
-	.vsync_start = 2340 + 32,
-	.vsync_end = 2340 + 32 + 4,
-	.vtotal = 2340 + 32 + 4 + 18,
-	.width_mm = 68,
-	.height_mm = 145,
-};
-
 static int sofef00_panel_get_modes(struct drm_panel *panel, struct drm_connector *connector)
 {
 	struct drm_display_mode *mode;
-	struct sofef00_panel *ctx = to_sofef00_panel(panel);
 
-	mode = drm_mode_duplicate(connector->dev, ctx->mode);
+	mode = drm_mode_duplicate(connector->dev, &enchilada_panel_mode);
 	if (!mode)
 		return -ENOMEM;
 
@@ -239,13 +195,6 @@ static int sofef00_panel_probe(struct mipi_dsi_device *dsi)
 	if (!ctx)
 		return -ENOMEM;
 
-	ctx->mode = of_device_get_match_data(dev);
-
-	if (!ctx->mode) {
-		dev_err(dev, "Missing device mode\n");
-		return -ENODEV;
-	}
-
 	ctx->supply = devm_regulator_get(dev, "vddio");
 	if (IS_ERR(ctx->supply))
 		return dev_err_probe(dev, PTR_ERR(ctx->supply),
@@ -295,14 +244,7 @@ static void sofef00_panel_remove(struct mipi_dsi_device *dsi)
 }
 
 static const struct of_device_id sofef00_panel_of_match[] = {
-	{ // OnePlus 6 / enchilada
-		.compatible = "samsung,sofef00",
-		.data = &enchilada_panel_mode,
-	},
-	{ // OnePlus 6T / fajita
-		.compatible = "samsung,s6e3fc2x01",
-		.data = &fajita_panel_mode,
-	},
+	{ .compatible = "samsung,sofef00" },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, sofef00_panel_of_match);
diff --git a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
index 729cbb0d8403..36abfa2e65e9 100644
--- a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
+++ b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
@@ -36,60 +36,49 @@ static inline struct sharp_nt_panel *to_sharp_nt_panel(struct drm_panel *panel)
 static int sharp_nt_panel_init(struct sharp_nt_panel *sharp_nt)
 {
 	struct mipi_dsi_device *dsi = sharp_nt->dsi;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
 	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
 
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
 
-	msleep(120);
+	mipi_dsi_msleep(&dsi_ctx, 120);
 
 	/* Novatek two-lane operation */
-	ret = mipi_dsi_dcs_write(dsi, 0xae, (u8[]){ 0x03 }, 1);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xae,  0x03);
 
 	/* Set both MCU and RGB I/F to 24bpp */
-	ret = mipi_dsi_dcs_set_pixel_format(dsi, MIPI_DCS_PIXEL_FMT_24BIT |
-					(MIPI_DCS_PIXEL_FMT_24BIT << 4));
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_set_pixel_format_multi(&dsi_ctx,
+					    MIPI_DCS_PIXEL_FMT_24BIT |
+					    (MIPI_DCS_PIXEL_FMT_24BIT << 4));
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int sharp_nt_panel_on(struct sharp_nt_panel *sharp_nt)
 {
 	struct mipi_dsi_device *dsi = sharp_nt->dsi;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
 	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int sharp_nt_panel_off(struct sharp_nt_panel *sharp_nt)
 {
 	struct mipi_dsi_device *dsi = sharp_nt->dsi;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
 	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
 
-	ret = mipi_dsi_dcs_set_display_off(dsi);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret < 0)
-		return ret;
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static int sharp_nt_panel_unprepare(struct drm_panel *panel)
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 232b03c1a259..82ee2f12b8d2 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -579,9 +579,10 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc)
 	u32 bus_flags;
 	int err;
 
-	panel = devm_kzalloc(dev, sizeof(*panel), GFP_KERNEL);
-	if (!panel)
-		return -ENOMEM;
+	panel = devm_drm_panel_alloc(dev, struct panel_simple, base,
+				     &panel_simple_funcs, desc->connector_type);
+	if (IS_ERR(panel))
+		return PTR_ERR(panel);
 
 	panel->desc = desc;
 
@@ -694,8 +695,6 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc)
 	pm_runtime_set_autosuspend_delay(dev, 1000);
 	pm_runtime_use_autosuspend(dev);
 
-	drm_panel_init(&panel->base, dev, &panel_simple_funcs, connector_type);
-
 	err = drm_panel_of_backlight(&panel->base);
 	if (err) {
 		dev_err_probe(dev, err, "Could not find backlight\n");
@@ -1027,27 +1026,28 @@ static const struct panel_desc auo_g070vvn01 = {
 	},
 };
 
-static const struct drm_display_mode auo_g101evn010_mode = {
-	.clock = 68930,
-	.hdisplay = 1280,
-	.hsync_start = 1280 + 82,
-	.hsync_end = 1280 + 82 + 2,
-	.htotal = 1280 + 82 + 2 + 84,
-	.vdisplay = 800,
-	.vsync_start = 800 + 8,
-	.vsync_end = 800 + 8 + 2,
-	.vtotal = 800 + 8 + 2 + 6,
+static const struct display_timing auo_g101evn010_timing = {
+	.pixelclock = { 64000000, 68930000, 85000000 },
+	.hactive = { 1280, 1280, 1280 },
+	.hfront_porch = { 8, 64, 256 },
+	.hback_porch = { 8, 64, 256 },
+	.hsync_len = { 40, 168, 767 },
+	.vactive = { 800, 800, 800 },
+	.vfront_porch = { 4, 8, 100 },
+	.vback_porch = { 4, 8, 100 },
+	.vsync_len = { 8, 16, 223 },
 };
 
 static const struct panel_desc auo_g101evn010 = {
-	.modes = &auo_g101evn010_mode,
-	.num_modes = 1,
+	.timings = &auo_g101evn010_timing,
+	.num_timings = 1,
 	.bpc = 6,
 	.size = {
 		.width = 216,
 		.height = 135,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB666_1X7X3_SPWG,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
 	.connector_type = DRM_MODE_CONNECTOR_LVDS,
 };
 
@@ -3527,6 +3527,30 @@ static const struct panel_desc newhaven_nhd_43_480272ef_atxl = {
 	.connector_type = DRM_MODE_CONNECTOR_DPI,
 };
 
+static const struct drm_display_mode nlt_nl13676bc25_03f_mode = {
+	.clock = 75400,
+	.hdisplay = 1366,
+	.hsync_start = 1366 + 14,
+	.hsync_end = 1366 + 14 + 56,
+	.htotal = 1366 + 14 + 56 + 64,
+	.vdisplay = 768,
+	.vsync_start = 768 + 1,
+	.vsync_end = 768 + 1 + 3,
+	.vtotal = 768 + 1 + 3 + 22,
+};
+
+static const struct panel_desc nlt_nl13676bc25_03f = {
+	.modes = &nlt_nl13676bc25_03f_mode,
+	.num_modes = 1,
+	.bpc = 8,
+	.size = {
+		.width = 363,
+		.height = 215,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
 static const struct display_timing nlt_nl192108ac18_02d_timing = {
 	.pixelclock = { 130000000, 148350000, 163000000 },
 	.hactive = { 1920, 1920, 1920 },
@@ -3796,6 +3820,32 @@ static const struct panel_desc pda_91_00156_a0  = {
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
 };
 
+static const struct drm_display_mode powertip_ph128800t004_zza01_mode = {
+	.clock = 71150,
+	.hdisplay = 1280,
+	.hsync_start = 1280 + 48,
+	.hsync_end = 1280 + 48 + 32,
+	.htotal = 1280 + 48 + 32 + 80,
+	.vdisplay = 800,
+	.vsync_start = 800 + 9,
+	.vsync_end = 800 + 9 + 8,
+	.vtotal = 800 + 9 + 8 + 6,
+	.flags = DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC,
+};
+
+static const struct panel_desc powertip_ph128800t004_zza01 = {
+	.modes = &powertip_ph128800t004_zza01_mode,
+	.num_modes = 1,
+	.bpc = 8,
+	.size = {
+		.width = 216,
+		.height = 135,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
 static const struct drm_display_mode powertip_ph128800t006_zhc01_mode = {
 	.clock = 66500,
 	.hdisplay = 1280,
@@ -4393,10 +4443,10 @@ static const struct panel_desc tianma_tm070jvhg33 = {
 };
 
 /*
- * The datasheet computes total blanking as back porch + front porch, not
- * including sync pulse width. This is for both H and V. To make the total
- * blanking and period correct, subtract the pulse width from the front
- * porch.
+ * The TM070JDHG34-00 datasheet computes total blanking as back porch +
+ * front porch, not including sync pulse width. This is for both H and
+ * V. To make the total blanking and period correct, subtract the pulse
+ * width from the front porch.
  *
  * This works well for the Min and Typ values, but for Max values the sync
  * pulse width is higher than back porch + front porch, so work around that
@@ -4405,6 +4455,10 @@ static const struct panel_desc tianma_tm070jvhg33 = {
  *
  * Exact datasheet values are added as a comment where they differ from the
  * ones implemented for the above reason.
+ *
+ * The P0700WXF1MBAA datasheet is even less detailed, only listing period
+ * and total blanking time, however the resulting values are the same as
+ * the TM070JDHG34-00.
  */
 static const struct display_timing tianma_tm070jdhg34_00_timing = {
 	.pixelclock = { 68400000, 71900000, 78100000 },
@@ -4427,6 +4481,30 @@ static const struct panel_desc tianma_tm070jdhg34_00 = {
 		.width = 150, /* 149.76 */
 		.height = 94, /* 93.60 */
 	},
+	.delay = {
+		.prepare = 15,		/* Tp1 */
+		.enable = 150,		/* Tp2 */
+		.disable = 150,		/* Tp4 */
+		.unprepare = 120,	/* Tp3 */
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
+static const struct panel_desc tianma_p0700wxf1mbaa = {
+	.timings = &tianma_tm070jdhg34_00_timing,
+	.num_timings = 1,
+	.bpc = 8,
+	.size = {
+		.width = 150, /* 149.76 */
+		.height = 94, /* 93.60 */
+	},
+	.delay = {
+		.prepare = 18,		/* Tr + Tp1 */
+		.enable = 152,		/* Tp2 + Tp5 */
+		.disable = 152,		/* Tp6 + Tp4 */
+		.unprepare = 120,	/* Tp3 */
+	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
 	.connector_type = DRM_MODE_CONNECTOR_LVDS,
 };
@@ -5121,6 +5199,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "newhaven,nhd-4.3-480272ef-atxl",
 		.data = &newhaven_nhd_43_480272ef_atxl,
 	}, {
+		.compatible = "nlt,nl13676bc25-03f",
+		.data = &nlt_nl13676bc25_03f,
+	}, {
 		.compatible = "nlt,nl192108ac18-02d",
 		.data = &nlt_nl192108ac18_02d,
 	}, {
@@ -5154,6 +5235,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "pda,91-00156-a0",
 		.data = &pda_91_00156_a0,
 	}, {
+		.compatible = "powertip,ph128800t004-zza01",
+		.data = &powertip_ph128800t004_zza01,
+	}, {
 		.compatible = "powertip,ph128800t006-zhc01",
 		.data = &powertip_ph128800t006_zhc01,
 	}, {
@@ -5214,6 +5298,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "tfc,s9700rtwv43tr-01b",
 		.data = &tfc_s9700rtwv43tr_01b,
 	}, {
+		.compatible = "tianma,p0700wxf1mbaa",
+		.data = &tianma_p0700wxf1mbaa,
+	}, {
 		.compatible = "tianma,tm070jdhg30",
 		.data = &tianma_tm070jdhg30,
 	}, {
diff --git a/drivers/gpu/drm/panel/panel-synaptics-r63353.c b/drivers/gpu/drm/panel/panel-synaptics-r63353.c
index 17349825543f..b148e6cba9bd 100644
--- a/drivers/gpu/drm/panel/panel-synaptics-r63353.c
+++ b/drivers/gpu/drm/panel/panel-synaptics-r63353.c
@@ -106,53 +106,34 @@ static int r63353_panel_power_off(struct r63353_panel *rpanel)
 static int r63353_panel_activate(struct r63353_panel *rpanel)
 {
 	struct mipi_dsi_device *dsi = rpanel->dsi;
-	struct device *dev = &dsi->dev;
-	int i, ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+	int i;
 
-	ret = mipi_dsi_dcs_soft_reset(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to do Software Reset (%d)\n", ret);
-		goto fail;
-	}
+	mipi_dsi_dcs_soft_reset_multi(&dsi_ctx);
 
-	usleep_range(15000, 17000);
+	mipi_dsi_usleep_range(&dsi_ctx, 15000, 17000);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to enter sleep mode (%d)\n", ret);
-		goto fail;
-	}
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
 
 	for (i = 0; i < rpanel->pdata->init_length; i++) {
 		const struct r63353_instr *instr = &rpanel->pdata->init[i];
 
-		ret = mipi_dsi_dcs_write_buffer(dsi, instr->data, instr->len);
-		if (ret < 0)
-			goto fail;
+		mipi_dsi_dcs_write_buffer_multi(&dsi_ctx, instr->data,
+						instr->len);
 	}
 
-	msleep(120);
-
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to exit sleep mode (%d)\n", ret);
-		goto fail;
-	}
+	mipi_dsi_msleep(&dsi_ctx, 120);
 
-	usleep_range(5000, 10000);
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display ON (%d)\n", ret);
-		goto fail;
-	}
+	mipi_dsi_usleep_range(&dsi_ctx, 5000, 10000);
 
-	return 0;
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
 
-fail:
-	gpiod_set_value(rpanel->reset_gpio, 0);
+	if (dsi_ctx.accum_err)
+		gpiod_set_value(rpanel->reset_gpio, 0);
 
-	return ret;
+	return dsi_ctx.accum_err;
 }
 
 static int r63353_panel_prepare(struct drm_panel *panel)
@@ -178,27 +159,16 @@ static int r63353_panel_prepare(struct drm_panel *panel)
 	return 0;
 }
 
-static int r63353_panel_deactivate(struct r63353_panel *rpanel)
+static void r63353_panel_deactivate(struct r63353_panel *rpanel)
 {
 	struct mipi_dsi_device *dsi = rpanel->dsi;
-	struct device *dev = &dsi->dev;
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
 
-	ret = mipi_dsi_dcs_set_display_off(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to set display OFF (%d)\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
 
-	usleep_range(5000, 10000);
+	mipi_dsi_usleep_range(&dsi_ctx, 5000, 10000);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret < 0) {
-		dev_err(dev, "Failed to enter sleep mode (%d)\n", ret);
-		return ret;
-	}
-
-	return 0;
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
 }
 
 static int r63353_panel_unprepare(struct drm_panel *panel)
diff --git a/drivers/gpu/drm/panel/panel-visionox-g2647fb105.c b/drivers/gpu/drm/panel/panel-visionox-g2647fb105.c
new file mode 100644
index 000000000000..413849f7b4de
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-visionox-g2647fb105.c
@@ -0,0 +1,280 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Generated with linux-mdss-dsi-panel-driver-generator from vendor device tree:
+ *   Copyright (c) 2013, The Linux Foundation. All rights reserved.
+ *   Copyright (c) 2025, Alexander Baransky <sanyapilot496@gmail.com>
+ */
+
+#include <linux/backlight.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+
+struct visionox_g2647fb105 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct gpio_desc *reset_gpio;
+	struct regulator_bulk_data *supplies;
+};
+
+static const struct regulator_bulk_data visionox_g2647fb105_supplies[] = {
+	{ .supply = "vdd3p3" },
+	{ .supply = "vddio" },
+	{ .supply = "vsn" },
+	{ .supply = "vsp" },
+};
+
+static inline
+struct visionox_g2647fb105 *to_visionox_g2647fb105(struct drm_panel *panel)
+{
+	return container_of(panel, struct visionox_g2647fb105, panel);
+}
+
+static void visionox_g2647fb105_reset(struct visionox_g2647fb105 *ctx)
+{
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(1000, 2000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	usleep_range(10000, 11000);
+}
+
+static int visionox_g2647fb105_on(struct visionox_g2647fb105 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x70, 0x04);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfe, 0x40);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x4d, 0x32);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfe, 0x40);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xbe, 0x17);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xbf, 0xbb);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc0, 0xdd);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc1, 0xff);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfe, 0xd0);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x03, 0x24);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x04, 0x03);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfe, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xc2, 0x08);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xfe, 0x00);
+
+	mipi_dsi_dcs_set_tear_on_multi(&dsi_ctx, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
+	mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, 0x0000);
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 100);
+
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
+
+	return dsi_ctx.accum_err;
+}
+
+static int visionox_g2647fb105_off(struct visionox_g2647fb105 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi };
+
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 50);
+
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 20);
+
+	return dsi_ctx.accum_err;
+}
+
+static int visionox_g2647fb105_prepare(struct drm_panel *panel)
+{
+	struct visionox_g2647fb105 *ctx = to_visionox_g2647fb105(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(visionox_g2647fb105_supplies), ctx->supplies);
+	if (ret < 0) {
+		dev_err(dev, "Failed to enable regulators: %d\n", ret);
+		return ret;
+	}
+
+	visionox_g2647fb105_reset(ctx);
+
+	ret = visionox_g2647fb105_on(ctx);
+	if (ret < 0) {
+		dev_err(dev, "Failed to initialize panel: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int visionox_g2647fb105_unprepare(struct drm_panel *panel)
+{
+	struct visionox_g2647fb105 *ctx = to_visionox_g2647fb105(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = visionox_g2647fb105_off(ctx);
+	if (ret < 0)
+		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	regulator_bulk_disable(ARRAY_SIZE(visionox_g2647fb105_supplies), ctx->supplies);
+
+	return 0;
+}
+
+static const struct drm_display_mode visionox_g2647fb105_mode = {
+	.clock = (1080 + 28 + 4 + 36) * (2340 + 8 + 4 + 4) * 60 / 1000,
+	.hdisplay = 1080,
+	.hsync_start = 1080 + 28,
+	.hsync_end = 1080 + 28 + 4,
+	.htotal = 1080 + 28 + 4 + 36,
+	.vdisplay = 2340,
+	.vsync_start = 2340 + 8,
+	.vsync_end = 2340 + 8 + 4,
+	.vtotal = 2340 + 8 + 4 + 4,
+	.width_mm = 69,
+	.height_mm = 149,
+};
+
+static int visionox_g2647fb105_get_modes(struct drm_panel *panel,
+					struct drm_connector *connector)
+{
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(connector->dev, &visionox_g2647fb105_mode);
+	if (!mode)
+		return -ENOMEM;
+
+	drm_mode_set_name(mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+	connector->display_info.width_mm = mode->width_mm;
+	connector->display_info.height_mm = mode->height_mm;
+	drm_mode_probed_add(connector, mode);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs visionox_g2647fb105_panel_funcs = {
+	.prepare = visionox_g2647fb105_prepare,
+	.unprepare = visionox_g2647fb105_unprepare,
+	.get_modes = visionox_g2647fb105_get_modes,
+};
+
+static int visionox_g2647fb105_bl_update_status(struct backlight_device *bl)
+{
+	struct mipi_dsi_device *dsi = bl_get_data(bl);
+	u16 brightness = backlight_get_brightness(bl);
+	int ret;
+
+	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	ret = mipi_dsi_dcs_set_display_brightness_large(dsi, brightness);
+	if (ret < 0)
+		return ret;
+
+	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	return 0;
+}
+
+static const struct backlight_ops visionox_g2647fb105_bl_ops = {
+	.update_status = visionox_g2647fb105_bl_update_status,
+};
+
+static struct backlight_device *
+visionox_g2647fb105_create_backlight(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	const struct backlight_properties props = {
+		.type = BACKLIGHT_RAW,
+		.brightness = 1023,
+		.max_brightness = 2047,
+	};
+
+	return devm_backlight_device_register(dev, dev_name(dev), dev, dsi,
+					      &visionox_g2647fb105_bl_ops, &props);
+}
+
+static int visionox_g2647fb105_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct visionox_g2647fb105 *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ret = devm_regulator_bulk_get_const(dev,
+					    ARRAY_SIZE(visionox_g2647fb105_supplies),
+					    visionox_g2647fb105_supplies,
+					    &ctx->supplies);
+	if (ret < 0)
+		return dev_err_probe(dev, ret, "Failed to get regulators\n");
+
+	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(ctx->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio),
+				     "Failed to get reset-gpios\n");
+
+	ctx->dsi = dsi;
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO_BURST |
+			  MIPI_DSI_CLOCK_NON_CONTINUOUS | MIPI_DSI_MODE_LPM;
+
+	ctx->panel.prepare_prev_first = true;
+
+	drm_panel_init(&ctx->panel, dev, &visionox_g2647fb105_panel_funcs,
+		       DRM_MODE_CONNECTOR_DSI);
+	ctx->panel.prepare_prev_first = true;
+
+	ctx->panel.backlight = visionox_g2647fb105_create_backlight(dsi);
+	if (IS_ERR(ctx->panel.backlight))
+		return dev_err_probe(dev, PTR_ERR(ctx->panel.backlight),
+				     "Failed to create backlight\n");
+
+	drm_panel_add(&ctx->panel);
+
+	ret = devm_mipi_dsi_attach(dev, dsi);
+	if (ret < 0) {
+		drm_panel_remove(&ctx->panel);
+		return dev_err_probe(dev, ret, "Failed to attach to DSI host\n");
+	}
+
+	return 0;
+}
+
+static void visionox_g2647fb105_remove(struct mipi_dsi_device *dsi)
+{
+	struct visionox_g2647fb105 *ctx = mipi_dsi_get_drvdata(dsi);
+	drm_panel_remove(&ctx->panel);
+}
+
+static const struct of_device_id visionox_g2647fb105_of_match[] = {
+	{ .compatible = "visionox,g2647fb105" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, visionox_g2647fb105_of_match);
+
+static struct mipi_dsi_driver visionox_g2647fb105_driver = {
+	.probe = visionox_g2647fb105_probe,
+	.remove = visionox_g2647fb105_remove,
+	.driver = {
+		.name = "panel-visionox-g2647fb105",
+		.of_match_table = visionox_g2647fb105_of_match,
+	},
+};
+module_mipi_dsi_driver(visionox_g2647fb105_driver);
+
+MODULE_AUTHOR("Alexander Baransky <sanyapilot496@gmail.com>");
+MODULE_DESCRIPTION("DRM driver for Visionox G2647FB105 AMOLED DSI panel");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c
index a45e4addcc19..5d35076b2e6d 100644
--- a/drivers/gpu/drm/panfrost/panfrost_device.c
+++ b/drivers/gpu/drm/panfrost/panfrost_device.c
@@ -209,10 +209,20 @@ int panfrost_device_init(struct panfrost_device *pfdev)
 
 	spin_lock_init(&pfdev->cycle_counter.lock);
 
+	err = panfrost_pm_domain_init(pfdev);
+	if (err)
+		return err;
+
+	err = panfrost_reset_init(pfdev);
+	if (err) {
+		dev_err(pfdev->dev, "reset init failed %d\n", err);
+		goto out_pm_domain;
+	}
+
 	err = panfrost_clk_init(pfdev);
 	if (err) {
 		dev_err(pfdev->dev, "clk init failed %d\n", err);
-		return err;
+		goto out_reset;
 	}
 
 	err = panfrost_devfreq_init(pfdev);
@@ -229,25 +239,15 @@ int panfrost_device_init(struct panfrost_device *pfdev)
 			goto out_devfreq;
 	}
 
-	err = panfrost_reset_init(pfdev);
-	if (err) {
-		dev_err(pfdev->dev, "reset init failed %d\n", err);
-		goto out_regulator;
-	}
-
-	err = panfrost_pm_domain_init(pfdev);
-	if (err)
-		goto out_reset;
-
 	pfdev->iomem = devm_platform_ioremap_resource(pfdev->pdev, 0);
 	if (IS_ERR(pfdev->iomem)) {
 		err = PTR_ERR(pfdev->iomem);
-		goto out_pm_domain;
+		goto out_regulator;
 	}
 
 	err = panfrost_gpu_init(pfdev);
 	if (err)
-		goto out_pm_domain;
+		goto out_regulator;
 
 	err = panfrost_mmu_init(pfdev);
 	if (err)
@@ -268,16 +268,16 @@ out_mmu:
 	panfrost_mmu_fini(pfdev);
 out_gpu:
 	panfrost_gpu_fini(pfdev);
-out_pm_domain:
-	panfrost_pm_domain_fini(pfdev);
-out_reset:
-	panfrost_reset_fini(pfdev);
 out_regulator:
 	panfrost_regulator_fini(pfdev);
 out_devfreq:
 	panfrost_devfreq_fini(pfdev);
 out_clk:
 	panfrost_clk_fini(pfdev);
+out_reset:
+	panfrost_reset_fini(pfdev);
+out_pm_domain:
+	panfrost_pm_domain_fini(pfdev);
 	return err;
 }
 
@@ -287,11 +287,11 @@ void panfrost_device_fini(struct panfrost_device *pfdev)
 	panfrost_job_fini(pfdev);
 	panfrost_mmu_fini(pfdev);
 	panfrost_gpu_fini(pfdev);
-	panfrost_pm_domain_fini(pfdev);
-	panfrost_reset_fini(pfdev);
 	panfrost_devfreq_fini(pfdev);
 	panfrost_regulator_fini(pfdev);
 	panfrost_clk_fini(pfdev);
+	panfrost_reset_fini(pfdev);
+	panfrost_pm_domain_fini(pfdev);
 }
 
 #define PANFROST_EXCEPTION(id) \
@@ -406,11 +406,36 @@ void panfrost_device_reset(struct panfrost_device *pfdev)
 static int panfrost_device_runtime_resume(struct device *dev)
 {
 	struct panfrost_device *pfdev = dev_get_drvdata(dev);
+	int ret;
+
+	if (pfdev->comp->pm_features & BIT(GPU_PM_RT)) {
+		ret = reset_control_deassert(pfdev->rstc);
+		if (ret)
+			return ret;
+
+		ret = clk_enable(pfdev->clock);
+		if (ret)
+			goto err_clk;
+
+		if (pfdev->bus_clock) {
+			ret = clk_enable(pfdev->bus_clock);
+			if (ret)
+				goto err_bus_clk;
+		}
+	}
 
 	panfrost_device_reset(pfdev);
 	panfrost_devfreq_resume(pfdev);
 
 	return 0;
+
+err_bus_clk:
+	if (pfdev->comp->pm_features & BIT(GPU_PM_RT))
+		clk_disable(pfdev->clock);
+err_clk:
+	if (pfdev->comp->pm_features & BIT(GPU_PM_RT))
+		reset_control_assert(pfdev->rstc);
+	return ret;
 }
 
 static int panfrost_device_runtime_suspend(struct device *dev)
@@ -426,6 +451,14 @@ static int panfrost_device_runtime_suspend(struct device *dev)
 	panfrost_gpu_suspend_irq(pfdev);
 	panfrost_gpu_power_off(pfdev);
 
+	if (pfdev->comp->pm_features & BIT(GPU_PM_RT)) {
+		if (pfdev->bus_clock)
+			clk_disable(pfdev->bus_clock);
+
+		clk_disable(pfdev->clock);
+		reset_control_assert(pfdev->rstc);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/panfrost/panfrost_device.h b/drivers/gpu/drm/panfrost/panfrost_device.h
index cffcb0ac7c11..dcff70f905cd 100644
--- a/drivers/gpu/drm/panfrost/panfrost_device.h
+++ b/drivers/gpu/drm/panfrost/panfrost_device.h
@@ -36,10 +36,21 @@ enum panfrost_drv_comp_bits {
  * enum panfrost_gpu_pm - Supported kernel power management features
  * @GPU_PM_CLK_DIS:  Allow disabling clocks during system suspend
  * @GPU_PM_VREG_OFF: Allow turning off regulators during system suspend
+ * @GPU_PM_RT: Allow disabling clocks and asserting the reset control during
+ *  system runtime suspend
  */
 enum panfrost_gpu_pm {
 	GPU_PM_CLK_DIS,
 	GPU_PM_VREG_OFF,
+	GPU_PM_RT
+};
+
+/**
+ * enum panfrost_gpu_quirks - GPU optional quirks
+ * @GPU_QUIRK_FORCE_AARCH64_PGTABLE: Use AARCH64_4K page table format
+ */
+enum panfrost_gpu_quirks {
+	GPU_QUIRK_FORCE_AARCH64_PGTABLE,
 };
 
 struct panfrost_features {
@@ -95,6 +106,9 @@ struct panfrost_compatible {
 
 	/* Allowed PM features */
 	u8 pm_features;
+
+	/* GPU configuration quirks */
+	u8 gpu_quirks;
 };
 
 struct panfrost_device {
@@ -162,6 +176,11 @@ struct panfrost_mmu {
 	int as;
 	atomic_t as_count;
 	struct list_head list;
+	struct {
+		u64 transtab;
+		u64 memattr;
+		u64 transcfg;
+	} cfg;
 };
 
 struct panfrost_engine_usage {
diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c
index 0f3935556ac7..f1ec3b02f15a 100644
--- a/drivers/gpu/drm/panfrost/panfrost_drv.c
+++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
@@ -476,7 +476,7 @@ static int panfrost_ioctl_madvise(struct drm_device *dev, void *data,
 		}
 	}
 
-	args->retained = drm_gem_shmem_madvise(&bo->base, args->madv);
+	args->retained = drm_gem_shmem_madvise_locked(&bo->base, args->madv);
 
 	if (args->retained) {
 		if (args->madv == PANFROST_MADV_DONTNEED)
@@ -776,6 +776,13 @@ static const struct panfrost_compatible default_data = {
 	.pm_domain_names = NULL,
 };
 
+static const struct panfrost_compatible allwinner_h616_data = {
+	.num_supplies = ARRAY_SIZE(default_supplies) - 1,
+	.supply_names = default_supplies,
+	.num_pm_domains = 1,
+	.pm_features = BIT(GPU_PM_RT),
+};
+
 static const struct panfrost_compatible amlogic_data = {
 	.num_supplies = ARRAY_SIZE(default_supplies) - 1,
 	.supply_names = default_supplies,
@@ -824,6 +831,7 @@ static const struct panfrost_compatible mediatek_mt8188_data = {
 	.num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains),
 	.pm_domain_names = mediatek_mt8183_pm_domains,
 	.pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF),
+	.gpu_quirks = BIT(GPU_QUIRK_FORCE_AARCH64_PGTABLE),
 };
 
 static const char * const mediatek_mt8192_supplies[] = { "mali", NULL };
@@ -835,6 +843,7 @@ static const struct panfrost_compatible mediatek_mt8192_data = {
 	.num_pm_domains = ARRAY_SIZE(mediatek_mt8192_pm_domains),
 	.pm_domain_names = mediatek_mt8192_pm_domains,
 	.pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF),
+	.gpu_quirks = BIT(GPU_QUIRK_FORCE_AARCH64_PGTABLE),
 };
 
 static const struct of_device_id dt_match[] = {
@@ -859,6 +868,7 @@ static const struct of_device_id dt_match[] = {
 	{ .compatible = "mediatek,mt8186-mali", .data = &mediatek_mt8186_data },
 	{ .compatible = "mediatek,mt8188-mali", .data = &mediatek_mt8188_data },
 	{ .compatible = "mediatek,mt8192-mali", .data = &mediatek_mt8192_data },
+	{ .compatible = "allwinner,sun50i-h616-mali", .data = &allwinner_h616_data },
 	{}
 };
 MODULE_DEVICE_TABLE(of, dt_match);
diff --git a/drivers/gpu/drm/panfrost/panfrost_dump.c b/drivers/gpu/drm/panfrost/panfrost_dump.c
index 47751302f1bc..4042afe2fbf4 100644
--- a/drivers/gpu/drm/panfrost/panfrost_dump.c
+++ b/drivers/gpu/drm/panfrost/panfrost_dump.c
@@ -209,7 +209,7 @@ void panfrost_core_dump(struct panfrost_job *job)
 			goto dump_header;
 		}
 
-		ret = drm_gem_vmap_unlocked(&bo->base.base, &map);
+		ret = drm_gem_vmap(&bo->base.base, &map);
 		if (ret) {
 			dev_err(pfdev->dev, "Panfrost Dump: couldn't map Buffer Object\n");
 			iter.hdr->bomap.valid = 0;
@@ -228,7 +228,7 @@ void panfrost_core_dump(struct panfrost_job *job)
 		vaddr = map.vaddr;
 		memcpy(iter.data, vaddr, bo->base.base.size);
 
-		drm_gem_vunmap_unlocked(&bo->base.base, &map);
+		drm_gem_vunmap(&bo->base.base, &map);
 
 		iter.hdr->bomap.valid = 1;
 
diff --git a/drivers/gpu/drm/panfrost/panfrost_features.h b/drivers/gpu/drm/panfrost/panfrost_features.h
index 7ed0cd3ea2d4..52f9d69f6db9 100644
--- a/drivers/gpu/drm/panfrost/panfrost_features.h
+++ b/drivers/gpu/drm/panfrost/panfrost_features.h
@@ -54,6 +54,7 @@ enum panfrost_hw_feature {
 	BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \
 	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
 	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
+	BIT_ULL(HW_FEATURE_AARCH64_MMU) | \
 	BIT_ULL(HW_FEATURE_COHERENCY_REG))
 
 #define hw_features_g72 (\
@@ -64,6 +65,7 @@ enum panfrost_hw_feature {
 	BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \
 	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
 	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
+	BIT_ULL(HW_FEATURE_AARCH64_MMU) | \
 	BIT_ULL(HW_FEATURE_COHERENCY_REG))
 
 #define hw_features_g51 hw_features_g72
@@ -77,6 +79,7 @@ enum panfrost_hw_feature {
 	BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \
 	BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \
 	BIT_ULL(HW_FEATURE_IDVS_GROUP_SIZE) | \
+	BIT_ULL(HW_FEATURE_AARCH64_MMU) | \
 	BIT_ULL(HW_FEATURE_COHERENCY_REG))
 
 #define hw_features_g76 (\
diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c
index 8e0ff3efede7..963f04ba2de6 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gem.c
+++ b/drivers/gpu/drm/panfrost/panfrost_gem.c
@@ -200,7 +200,7 @@ static enum drm_gem_object_status panfrost_gem_status(struct drm_gem_object *obj
 	struct panfrost_gem_object *bo = to_panfrost_bo(obj);
 	enum drm_gem_object_status res = 0;
 
-	if (bo->base.base.import_attach || bo->base.pages)
+	if (drm_gem_is_imported(&bo->base.base) || bo->base.pages)
 		res |= DRM_GEM_OBJECT_RESIDENT;
 
 	if (bo->base.madv == PANFROST_MADV_DONTNEED)
diff --git a/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c b/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c
index 3d9f51bd48b6..02b60ea1433a 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c
+++ b/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c
@@ -51,7 +51,7 @@ static bool panfrost_gem_purge(struct drm_gem_object *obj)
 		goto unlock_mappings;
 
 	panfrost_gem_teardown_mappings_locked(bo);
-	drm_gem_shmem_purge(&bo->base);
+	drm_gem_shmem_purge_locked(&bo->base);
 	ret = true;
 
 	dma_resv_unlock(shmem->base.resv);
diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c
index b91019cd5acb..f6b91c052cfb 100644
--- a/drivers/gpu/drm/panfrost/panfrost_mmu.c
+++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c
@@ -26,6 +26,48 @@
 #define mmu_write(dev, reg, data) writel(data, dev->iomem + reg)
 #define mmu_read(dev, reg) readl(dev->iomem + reg)
 
+static u64 mair_to_memattr(u64 mair, bool coherent)
+{
+	u64 memattr = 0;
+	u32 i;
+
+	for (i = 0; i < 8; i++) {
+		u8 in_attr = mair >> (8 * i), out_attr;
+		u8 outer = in_attr >> 4, inner = in_attr & 0xf;
+
+		/* For caching to be enabled, inner and outer caching policy
+		 * have to be both write-back, if one of them is write-through
+		 * or non-cacheable, we just choose non-cacheable. Device
+		 * memory is also translated to non-cacheable.
+		 */
+		if (!(outer & 3) || !(outer & 4) || !(inner & 4)) {
+			out_attr = AS_MEMATTR_AARCH64_INNER_OUTER_NC |
+				   AS_MEMATTR_AARCH64_SH_MIDGARD_INNER |
+				   AS_MEMATTR_AARCH64_INNER_ALLOC_EXPL(false, false);
+		} else {
+			out_attr = AS_MEMATTR_AARCH64_INNER_OUTER_WB |
+				   AS_MEMATTR_AARCH64_INNER_ALLOC_EXPL(inner & 1, inner & 2);
+			/* Use SH_MIDGARD_INNER mode when device isn't coherent,
+			 * so SH_IS, which is used when IOMMU_CACHE is set, maps
+			 * to Mali's internal-shareable mode. As per the Mali
+			 * Spec, inner and outer-shareable modes aren't allowed
+			 * for WB memory when coherency is disabled.
+			 * Use SH_CPU_INNER mode when coherency is enabled, so
+			 * that SH_IS actually maps to the standard definition of
+			 * inner-shareable.
+			 */
+			if (!coherent)
+				out_attr |= AS_MEMATTR_AARCH64_SH_MIDGARD_INNER;
+			else
+				out_attr |= AS_MEMATTR_AARCH64_SH_CPU_INNER;
+		}
+
+		memattr |= (u64)out_attr << (8 * i);
+	}
+
+	return memattr;
+}
+
 static int wait_ready(struct panfrost_device *pfdev, u32 as_nr)
 {
 	int ret;
@@ -124,9 +166,9 @@ static int mmu_hw_do_operation(struct panfrost_device *pfdev,
 static void panfrost_mmu_enable(struct panfrost_device *pfdev, struct panfrost_mmu *mmu)
 {
 	int as_nr = mmu->as;
-	struct io_pgtable_cfg *cfg = &mmu->pgtbl_cfg;
-	u64 transtab = cfg->arm_mali_lpae_cfg.transtab;
-	u64 memattr = cfg->arm_mali_lpae_cfg.memattr;
+	u64 transtab = mmu->cfg.transtab;
+	u64 memattr = mmu->cfg.memattr;
+	u64 transcfg = mmu->cfg.transcfg;
 
 	mmu_hw_do_operation_locked(pfdev, as_nr, 0, ~0ULL, AS_COMMAND_FLUSH_MEM);
 
@@ -139,6 +181,9 @@ static void panfrost_mmu_enable(struct panfrost_device *pfdev, struct panfrost_m
 	mmu_write(pfdev, AS_MEMATTR_LO(as_nr), lower_32_bits(memattr));
 	mmu_write(pfdev, AS_MEMATTR_HI(as_nr), upper_32_bits(memattr));
 
+	mmu_write(pfdev, AS_TRANSCFG_LO(as_nr), lower_32_bits(transcfg));
+	mmu_write(pfdev, AS_TRANSCFG_HI(as_nr), upper_32_bits(transcfg));
+
 	write_cmd(pfdev, as_nr, AS_COMMAND_UPDATE);
 }
 
@@ -152,9 +197,67 @@ static void panfrost_mmu_disable(struct panfrost_device *pfdev, u32 as_nr)
 	mmu_write(pfdev, AS_MEMATTR_LO(as_nr), 0);
 	mmu_write(pfdev, AS_MEMATTR_HI(as_nr), 0);
 
+	mmu_write(pfdev, AS_TRANSCFG_LO(as_nr), AS_TRANSCFG_ADRMODE_UNMAPPED);
+	mmu_write(pfdev, AS_TRANSCFG_HI(as_nr), 0);
+
 	write_cmd(pfdev, as_nr, AS_COMMAND_UPDATE);
 }
 
+static int mmu_cfg_init_mali_lpae(struct panfrost_mmu *mmu)
+{
+	struct io_pgtable_cfg *pgtbl_cfg = &mmu->pgtbl_cfg;
+
+	/* TODO: The following fields are duplicated between the MMU and Page
+	 * Table config structs. Ideally, should be kept in one place.
+	 */
+	mmu->cfg.transtab = pgtbl_cfg->arm_mali_lpae_cfg.transtab;
+	mmu->cfg.memattr = pgtbl_cfg->arm_mali_lpae_cfg.memattr;
+	mmu->cfg.transcfg = AS_TRANSCFG_ADRMODE_LEGACY;
+
+	return 0;
+}
+
+static int mmu_cfg_init_aarch64_4k(struct panfrost_mmu *mmu)
+{
+	struct io_pgtable_cfg *pgtbl_cfg = &mmu->pgtbl_cfg;
+	struct panfrost_device *pfdev = mmu->pfdev;
+
+	if (drm_WARN_ON(pfdev->ddev, pgtbl_cfg->arm_lpae_s1_cfg.ttbr &
+				     ~AS_TRANSTAB_AARCH64_4K_ADDR_MASK))
+		return -EINVAL;
+
+	mmu->cfg.transtab = pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
+
+	mmu->cfg.memattr = mair_to_memattr(pgtbl_cfg->arm_lpae_s1_cfg.mair,
+					   pgtbl_cfg->coherent_walk);
+
+	mmu->cfg.transcfg = AS_TRANSCFG_PTW_MEMATTR_WB |
+			    AS_TRANSCFG_PTW_RA |
+			    AS_TRANSCFG_ADRMODE_AARCH64_4K |
+			    AS_TRANSCFG_INA_BITS(55 - pgtbl_cfg->ias);
+	if (pgtbl_cfg->coherent_walk)
+		mmu->cfg.transcfg |= AS_TRANSCFG_PTW_SH_OS;
+
+	return 0;
+}
+
+static int panfrost_mmu_cfg_init(struct panfrost_mmu *mmu,
+				 enum io_pgtable_fmt fmt)
+{
+	struct panfrost_device *pfdev = mmu->pfdev;
+
+	switch (fmt) {
+	case ARM_64_LPAE_S1:
+		return mmu_cfg_init_aarch64_4k(mmu);
+	case ARM_MALI_LPAE:
+		return mmu_cfg_init_mali_lpae(mmu);
+	default:
+		/* This should never happen */
+		drm_WARN(pfdev->ddev, 1, "Invalid pgtable format");
+		return -EINVAL;
+	}
+}
+
 u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu)
 {
 	int as;
@@ -327,7 +430,7 @@ int panfrost_mmu_map(struct panfrost_gem_mapping *mapping)
 	struct drm_gem_object *obj = &shmem->base;
 	struct panfrost_device *pfdev = to_panfrost_device(obj->dev);
 	struct sg_table *sgt;
-	int prot = IOMMU_READ | IOMMU_WRITE;
+	int prot = IOMMU_READ | IOMMU_WRITE | IOMMU_CACHE;
 
 	if (WARN_ON(mapping->active))
 		return 0;
@@ -489,7 +592,7 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as,
 			goto err_unlock;
 		}
 		bo->base.pages = pages;
-		bo->base.pages_use_count = 1;
+		refcount_set(&bo->base.pages_use_count, 1);
 	} else {
 		pages = bo->base.pages;
 		if (pages[page_offset]) {
@@ -528,7 +631,7 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as,
 		goto err_map;
 
 	mmu_map_sg(pfdev, bomapping->mmu, addr,
-		   IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt);
+		   IOMMU_WRITE | IOMMU_READ | IOMMU_CACHE | IOMMU_NOEXEC, sgt);
 
 	bomapping->active = true;
 	bo->heap_rss_size += SZ_2M;
@@ -615,7 +718,22 @@ static void panfrost_drm_mm_color_adjust(const struct drm_mm_node *node,
 
 struct panfrost_mmu *panfrost_mmu_ctx_create(struct panfrost_device *pfdev)
 {
+	u32 va_bits = GPU_MMU_FEATURES_VA_BITS(pfdev->features.mmu_features);
+	u32 pa_bits = GPU_MMU_FEATURES_PA_BITS(pfdev->features.mmu_features);
 	struct panfrost_mmu *mmu;
+	enum io_pgtable_fmt fmt;
+	int ret;
+
+	if (pfdev->comp->gpu_quirks & BIT(GPU_QUIRK_FORCE_AARCH64_PGTABLE)) {
+		if (!panfrost_has_hw_feature(pfdev, HW_FEATURE_AARCH64_MMU)) {
+			dev_err_once(pfdev->dev,
+				     "AARCH64_4K page table not supported\n");
+			return ERR_PTR(-EINVAL);
+		}
+		fmt = ARM_64_LPAE_S1;
+	} else {
+		fmt = ARM_MALI_LPAE;
+	}
 
 	mmu = kzalloc(sizeof(*mmu), GFP_KERNEL);
 	if (!mmu)
@@ -633,23 +751,33 @@ struct panfrost_mmu *panfrost_mmu_ctx_create(struct panfrost_device *pfdev)
 
 	mmu->pgtbl_cfg = (struct io_pgtable_cfg) {
 		.pgsize_bitmap	= SZ_4K | SZ_2M,
-		.ias		= FIELD_GET(0xff, pfdev->features.mmu_features),
-		.oas		= FIELD_GET(0xff00, pfdev->features.mmu_features),
+		.ias		= va_bits,
+		.oas		= pa_bits,
 		.coherent_walk	= pfdev->coherent,
 		.tlb		= &mmu_tlb_ops,
 		.iommu_dev	= pfdev->dev,
 	};
 
-	mmu->pgtbl_ops = alloc_io_pgtable_ops(ARM_MALI_LPAE, &mmu->pgtbl_cfg,
-					      mmu);
+	mmu->pgtbl_ops = alloc_io_pgtable_ops(fmt, &mmu->pgtbl_cfg, mmu);
 	if (!mmu->pgtbl_ops) {
-		kfree(mmu);
-		return ERR_PTR(-EINVAL);
+		ret = -EINVAL;
+		goto err_free_mmu;
 	}
 
+	ret = panfrost_mmu_cfg_init(mmu, fmt);
+	if (ret)
+		goto err_free_io_pgtable;
+
 	kref_init(&mmu->refcount);
 
 	return mmu;
+
+err_free_io_pgtable:
+	free_io_pgtable_ops(mmu->pgtbl_ops);
+
+err_free_mmu:
+	kfree(mmu);
+	return ERR_PTR(ret);
 }
 
 static const char *access_type_name(struct panfrost_device *pfdev,
diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
index ba9b6e2b2636..52befead08c6 100644
--- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
+++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
@@ -106,7 +106,7 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
 		goto err_close_bo;
 	}
 
-	ret = drm_gem_vmap_unlocked(&bo->base, &map);
+	ret = drm_gem_vmap(&bo->base, &map);
 	if (ret)
 		goto err_put_mapping;
 	perfcnt->buf = map.vaddr;
@@ -165,7 +165,7 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
 	return 0;
 
 err_vunmap:
-	drm_gem_vunmap_unlocked(&bo->base, &map);
+	drm_gem_vunmap(&bo->base, &map);
 err_put_mapping:
 	panfrost_gem_mapping_put(perfcnt->mapping);
 err_close_bo:
@@ -195,7 +195,7 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev,
 		  GPU_PERFCNT_CFG_MODE(GPU_PERFCNT_CFG_MODE_OFF));
 
 	perfcnt->user = NULL;
-	drm_gem_vunmap_unlocked(&perfcnt->mapping->obj->base.base, &map);
+	drm_gem_vunmap(&perfcnt->mapping->obj->base.base, &map);
 	perfcnt->buf = NULL;
 	panfrost_gem_close(&perfcnt->mapping->obj->base.base, file_priv);
 	panfrost_mmu_as_put(pfdev, perfcnt->mapping->mmu);
diff --git a/drivers/gpu/drm/panfrost/panfrost_regs.h b/drivers/gpu/drm/panfrost/panfrost_regs.h
index c7bba476ab3f..2b8f1617b836 100644
--- a/drivers/gpu/drm/panfrost/panfrost_regs.h
+++ b/drivers/gpu/drm/panfrost/panfrost_regs.h
@@ -16,6 +16,8 @@
 #define   GROUPS_L2_COHERENT		BIT(0)	/* Cores groups are l2 coherent */
 
 #define GPU_MMU_FEATURES		0x014	/* (RO) MMU features */
+#define  GPU_MMU_FEATURES_VA_BITS(x)	((x) & GENMASK(7, 0))
+#define  GPU_MMU_FEATURES_PA_BITS(x)	(((x) >> 8) & GENMASK(7, 0))
 #define GPU_AS_PRESENT			0x018	/* (RO) Address space slots present */
 #define GPU_JS_PRESENT			0x01C	/* (RO) Job slots present */
 
@@ -299,6 +301,17 @@
 #define AS_TRANSTAB_HI(as)		(MMU_AS(as) + 0x04) /* (RW) Translation Table Base Address for address space n, high word */
 #define AS_MEMATTR_LO(as)		(MMU_AS(as) + 0x08) /* (RW) Memory attributes for address space n, low word. */
 #define AS_MEMATTR_HI(as)		(MMU_AS(as) + 0x0C) /* (RW) Memory attributes for address space n, high word. */
+#define   AS_MEMATTR_AARCH64_INNER_ALLOC_IMPL		(2 << 2)
+#define   AS_MEMATTR_AARCH64_INNER_ALLOC_EXPL(w, r)	((3 << 2) | \
+							 ((w) ? BIT(0) : 0) | \
+							 ((r) ? BIT(1) : 0))
+#define   AS_MEMATTR_AARCH64_SH_MIDGARD_INNER		(0 << 4)
+#define   AS_MEMATTR_AARCH64_SH_CPU_INNER		(1 << 4)
+#define   AS_MEMATTR_AARCH64_SH_CPU_INNER_SHADER_COH	(2 << 4)
+#define   AS_MEMATTR_AARCH64_SHARED			(0 << 6)
+#define   AS_MEMATTR_AARCH64_INNER_OUTER_NC		(1 << 6)
+#define   AS_MEMATTR_AARCH64_INNER_OUTER_WB		(2 << 6)
+#define   AS_MEMATTR_AARCH64_FAULT			(3 << 6)
 #define AS_LOCKADDR_LO(as)		(MMU_AS(as) + 0x10) /* (RW) Lock region address for address space n, low word */
 #define AS_LOCKADDR_HI(as)		(MMU_AS(as) + 0x14) /* (RW) Lock region address for address space n, high word */
 #define AS_COMMAND(as)			(MMU_AS(as) + 0x18) /* (WO) MMU command register for address space n */
@@ -309,6 +322,24 @@
 /* Additional Bifrost AS registers */
 #define AS_TRANSCFG_LO(as)		(MMU_AS(as) + 0x30) /* (RW) Translation table configuration for address space n, low word */
 #define AS_TRANSCFG_HI(as)		(MMU_AS(as) + 0x34) /* (RW) Translation table configuration for address space n, high word */
+#define   AS_TRANSCFG_ADRMODE_LEGACY			(0 << 0)
+#define   AS_TRANSCFG_ADRMODE_UNMAPPED			(1 << 0)
+#define   AS_TRANSCFG_ADRMODE_IDENTITY			(2 << 0)
+#define   AS_TRANSCFG_ADRMODE_AARCH64_4K		(6 << 0)
+#define   AS_TRANSCFG_ADRMODE_AARCH64_64K		(8 << 0)
+#define   AS_TRANSCFG_INA_BITS(x)			((x) << 6)
+#define   AS_TRANSCFG_OUTA_BITS(x)			((x) << 14)
+#define   AS_TRANSCFG_SL_CONCAT				BIT(22)
+#define   AS_TRANSCFG_PTW_MEMATTR_NC			(1 << 24)
+#define   AS_TRANSCFG_PTW_MEMATTR_WB			(2 << 24)
+#define   AS_TRANSCFG_PTW_SH_NS				(0 << 28)
+#define   AS_TRANSCFG_PTW_SH_OS				(2 << 28)
+#define   AS_TRANSCFG_PTW_SH_IS				(3 << 28)
+#define   AS_TRANSCFG_PTW_RA				BIT(30)
+#define   AS_TRANSCFG_DISABLE_HIER_AP			BIT(33)
+#define   AS_TRANSCFG_DISABLE_AF_FAULT			BIT(34)
+#define   AS_TRANSCFG_WXN				BIT(35)
+#define   AS_TRANSCFG_XREADABLE				BIT(36)
 #define AS_FAULTEXTRA_LO(as)		(MMU_AS(as) + 0x38) /* (RO) Secondary fault address for address space n, low word */
 #define AS_FAULTEXTRA_HI(as)		(MMU_AS(as) + 0x3C) /* (RO) Secondary fault address for address space n, high word */
 
@@ -324,6 +355,11 @@
 #define AS_TRANSTAB_LPAE_READ_INNER		BIT(2)
 #define AS_TRANSTAB_LPAE_SHARE_OUTER		BIT(4)
 
+/*
+ * Begin AARCH64_4K MMU TRANSTAB register values
+ */
+#define AS_TRANSTAB_AARCH64_4K_ADDR_MASK	0xfffffffffffffff0
+
 #define AS_STATUS_AS_ACTIVE			0x01
 
 #define AS_FAULTSTATUS_ACCESS_TYPE_MASK		(0x3 << 8)
diff --git a/drivers/gpu/drm/panthor/panthor_device.c b/drivers/gpu/drm/panthor/panthor_device.c
index a9da1d1eeb70..f0b2da5b2b96 100644
--- a/drivers/gpu/drm/panthor/panthor_device.c
+++ b/drivers/gpu/drm/panthor/panthor_device.c
@@ -171,10 +171,6 @@ int panthor_device_init(struct panthor_device *ptdev)
 	struct page *p;
 	int ret;
 
-	ret = panthor_gpu_coherency_init(ptdev);
-	if (ret)
-		return ret;
-
 	init_completion(&ptdev->unplug.done);
 	ret = drmm_mutex_init(&ptdev->base, &ptdev->unplug.lock);
 	if (ret)
@@ -184,6 +180,11 @@ int panthor_device_init(struct panthor_device *ptdev)
 	if (ret)
 		return ret;
 
+#ifdef CONFIG_DEBUG_FS
+	drmm_mutex_init(&ptdev->base, &ptdev->gems.lock);
+	INIT_LIST_HEAD(&ptdev->gems.node);
+#endif
+
 	atomic_set(&ptdev->pm.state, PANTHOR_DEVICE_PM_STATE_SUSPENDED);
 	p = alloc_page(GFP_KERNEL | __GFP_ZERO);
 	if (!p)
@@ -247,6 +248,10 @@ int panthor_device_init(struct panthor_device *ptdev)
 	if (ret)
 		goto err_rpm_put;
 
+	ret = panthor_gpu_coherency_init(ptdev);
+	if (ret)
+		goto err_unplug_gpu;
+
 	ret = panthor_mmu_init(ptdev);
 	if (ret)
 		goto err_unplug_gpu;
diff --git a/drivers/gpu/drm/panthor/panthor_device.h b/drivers/gpu/drm/panthor/panthor_device.h
index da6574021664..465d3ab1b79e 100644
--- a/drivers/gpu/drm/panthor/panthor_device.h
+++ b/drivers/gpu/drm/panthor/panthor_device.h
@@ -205,6 +205,17 @@ struct panthor_device {
 
 	/** @fast_rate: Maximum device clock frequency. Set by DVFS */
 	unsigned long fast_rate;
+
+#ifdef CONFIG_DEBUG_FS
+	/** @gems: Device-wide list of GEM objects owned by at least one file. */
+	struct {
+		/** @gems.lock: Protects the device-wide list of GEM objects. */
+		struct mutex lock;
+
+		/** @node: Used to keep track of all the device's DRM objects */
+		struct list_head node;
+	} gems;
+#endif
 };
 
 struct panthor_gpu_usage {
@@ -383,8 +394,6 @@ static irqreturn_t panthor_ ## __name ## _irq_threaded_handler(int irq, void *da
 		if (!status)									\
 			break;									\
 												\
-		gpu_write(ptdev, __reg_prefix ## _INT_CLEAR, status);				\
-												\
 		__handler(ptdev, status);							\
 		ret = IRQ_HANDLED;								\
 	}											\
diff --git a/drivers/gpu/drm/panthor/panthor_drv.c b/drivers/gpu/drm/panthor/panthor_drv.c
index 06fe46e32073..6200cad22563 100644
--- a/drivers/gpu/drm/panthor/panthor_drv.c
+++ b/drivers/gpu/drm/panthor/panthor_drv.c
@@ -940,6 +940,7 @@ static int panthor_ioctl_bo_mmap_offset(struct drm_device *ddev, void *data,
 					struct drm_file *file)
 {
 	struct drm_panthor_bo_mmap_offset *args = data;
+	struct panthor_gem_object *bo;
 	struct drm_gem_object *obj;
 	int ret;
 
@@ -950,6 +951,12 @@ static int panthor_ioctl_bo_mmap_offset(struct drm_device *ddev, void *data,
 	if (!obj)
 		return -ENOENT;
 
+	bo = to_panthor_bo(obj);
+	if (bo->flags & DRM_PANTHOR_BO_NO_MMAP) {
+		ret = -EPERM;
+		goto out;
+	}
+
 	ret = drm_gem_create_mmap_offset(obj);
 	if (ret)
 		goto out;
@@ -1331,6 +1338,46 @@ static int panthor_ioctl_vm_get_state(struct drm_device *ddev, void *data,
 	return 0;
 }
 
+static int panthor_ioctl_bo_set_label(struct drm_device *ddev, void *data,
+				      struct drm_file *file)
+{
+	struct drm_panthor_bo_set_label *args = data;
+	struct drm_gem_object *obj;
+	const char *label = NULL;
+	int ret = 0;
+
+	if (args->pad)
+		return -EINVAL;
+
+	obj = drm_gem_object_lookup(file, args->handle);
+	if (!obj)
+		return -ENOENT;
+
+	if (args->label) {
+		label = strndup_user((const char __user *)(uintptr_t)args->label,
+				     PANTHOR_BO_LABEL_MAXLEN);
+		if (IS_ERR(label)) {
+			ret = PTR_ERR(label);
+			if (ret == -EINVAL)
+				ret = -E2BIG;
+			goto err_put_obj;
+		}
+	}
+
+	/*
+	 * We treat passing a label of length 0 and passing a NULL label
+	 * differently, because even though they might seem conceptually
+	 * similar, future uses of the BO label might expect a different
+	 * behaviour in each case.
+	 */
+	panthor_gem_bo_set_label(obj, label);
+
+err_put_obj:
+	drm_gem_object_put(obj);
+
+	return ret;
+}
+
 static int
 panthor_open(struct drm_device *ddev, struct drm_file *file)
 {
@@ -1400,6 +1447,7 @@ static const struct drm_ioctl_desc panthor_drm_driver_ioctls[] = {
 	PANTHOR_IOCTL(TILER_HEAP_CREATE, tiler_heap_create, DRM_RENDER_ALLOW),
 	PANTHOR_IOCTL(TILER_HEAP_DESTROY, tiler_heap_destroy, DRM_RENDER_ALLOW),
 	PANTHOR_IOCTL(GROUP_SUBMIT, group_submit, DRM_RENDER_ALLOW),
+	PANTHOR_IOCTL(BO_SET_LABEL, bo_set_label, DRM_RENDER_ALLOW),
 };
 
 static int panthor_mmap(struct file *filp, struct vm_area_struct *vma)
@@ -1496,9 +1544,34 @@ static const struct file_operations panthor_drm_driver_fops = {
 };
 
 #ifdef CONFIG_DEBUG_FS
+static int panthor_gems_show(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct panthor_device *ptdev = container_of(dev, struct panthor_device, base);
+
+	panthor_gem_debugfs_print_bos(ptdev, m);
+
+	return 0;
+}
+
+static struct drm_info_list panthor_debugfs_list[] = {
+	{"gems", panthor_gems_show, 0, NULL},
+};
+
+static int panthor_gems_debugfs_init(struct drm_minor *minor)
+{
+	drm_debugfs_create_files(panthor_debugfs_list,
+				 ARRAY_SIZE(panthor_debugfs_list),
+				 minor->debugfs_root, minor);
+
+	return 0;
+}
+
 static void panthor_debugfs_init(struct drm_minor *minor)
 {
 	panthor_mmu_debugfs_init(minor);
+	panthor_gems_debugfs_init(minor);
 }
 #endif
 
@@ -1509,6 +1582,7 @@ static void panthor_debugfs_init(struct drm_minor *minor)
  * - 1.2 - adds DEV_QUERY_GROUP_PRIORITIES_INFO query
  *       - adds PANTHOR_GROUP_PRIORITY_REALTIME priority
  * - 1.3 - adds DRM_PANTHOR_GROUP_STATE_INNOCENT flag
+ * - 1.4 - adds DRM_IOCTL_PANTHOR_BO_SET_LABEL ioctl
  */
 static const struct drm_driver panthor_drm_driver = {
 	.driver_features = DRIVER_RENDER | DRIVER_GEM | DRIVER_SYNCOBJ |
@@ -1522,7 +1596,7 @@ static const struct drm_driver panthor_drm_driver = {
 	.name = "panthor",
 	.desc = "Panthor DRM driver",
 	.major = 1,
-	.minor = 3,
+	.minor = 4,
 
 	.gem_create_object = panthor_gem_create_object,
 	.gem_prime_import_sg_table = drm_gem_shmem_prime_import_sg_table,
diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c
index 0f52766a3120..7bc38e635329 100644
--- a/drivers/gpu/drm/panthor/panthor_fw.c
+++ b/drivers/gpu/drm/panthor/panthor_fw.c
@@ -449,7 +449,8 @@ panthor_fw_alloc_queue_iface_mem(struct panthor_device *ptdev,
 				       DRM_PANTHOR_BO_NO_MMAP,
 				       DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC |
 				       DRM_PANTHOR_VM_BIND_OP_MAP_UNCACHED,
-				       PANTHOR_VM_KERNEL_AUTO_VA);
+				       PANTHOR_VM_KERNEL_AUTO_VA,
+				       "Queue FW interface");
 	if (IS_ERR(mem))
 		return mem;
 
@@ -481,7 +482,8 @@ panthor_fw_alloc_suspend_buf_mem(struct panthor_device *ptdev, size_t size)
 	return panthor_kernel_bo_create(ptdev, panthor_fw_vm(ptdev), size,
 					DRM_PANTHOR_BO_NO_MMAP,
 					DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC,
-					PANTHOR_VM_KERNEL_AUTO_VA);
+					PANTHOR_VM_KERNEL_AUTO_VA,
+					"FW suspend buffer");
 }
 
 static int panthor_fw_load_section_entry(struct panthor_device *ptdev,
@@ -601,7 +603,7 @@ static int panthor_fw_load_section_entry(struct panthor_device *ptdev,
 		section->mem = panthor_kernel_bo_create(ptdev, panthor_fw_vm(ptdev),
 							section_size,
 							DRM_PANTHOR_BO_NO_MMAP,
-							vm_map_flags, va);
+							vm_map_flags, va, "FW section");
 		if (IS_ERR(section->mem))
 			return PTR_ERR(section->mem);
 
@@ -1008,6 +1010,8 @@ static void panthor_fw_init_global_iface(struct panthor_device *ptdev)
 
 static void panthor_job_irq_handler(struct panthor_device *ptdev, u32 status)
 {
+	gpu_write(ptdev, JOB_INT_CLEAR, status);
+
 	if (!ptdev->fw->booted && (status & JOB_INT_GLOBAL_IF))
 		ptdev->fw->booted = true;
 
diff --git a/drivers/gpu/drm/panthor/panthor_gem.c b/drivers/gpu/drm/panthor/panthor_gem.c
index 8244a4e6c2a2..7c00fd77758b 100644
--- a/drivers/gpu/drm/panthor/panthor_gem.c
+++ b/drivers/gpu/drm/panthor/panthor_gem.c
@@ -2,6 +2,7 @@
 /* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
 /* Copyright 2023 Collabora ltd. */
 
+#include <linux/cleanup.h>
 #include <linux/dma-buf.h>
 #include <linux/dma-mapping.h>
 #include <linux/err.h>
@@ -10,14 +11,64 @@
 #include <drm/panthor_drm.h>
 
 #include "panthor_device.h"
+#include "panthor_fw.h"
 #include "panthor_gem.h"
 #include "panthor_mmu.h"
 
+#ifdef CONFIG_DEBUG_FS
+static void panthor_gem_debugfs_bo_add(struct panthor_device *ptdev,
+				       struct panthor_gem_object *bo)
+{
+	INIT_LIST_HEAD(&bo->debugfs.node);
+
+	bo->debugfs.creator.tgid = current->group_leader->pid;
+	get_task_comm(bo->debugfs.creator.process_name, current->group_leader);
+
+	mutex_lock(&ptdev->gems.lock);
+	list_add_tail(&bo->debugfs.node, &ptdev->gems.node);
+	mutex_unlock(&ptdev->gems.lock);
+}
+
+static void panthor_gem_debugfs_bo_rm(struct panthor_gem_object *bo)
+{
+	struct panthor_device *ptdev = container_of(bo->base.base.dev,
+						    struct panthor_device, base);
+
+	if (list_empty(&bo->debugfs.node))
+		return;
+
+	mutex_lock(&ptdev->gems.lock);
+	list_del_init(&bo->debugfs.node);
+	mutex_unlock(&ptdev->gems.lock);
+}
+
+static void panthor_gem_debugfs_set_usage_flags(struct panthor_gem_object *bo, u32 usage_flags)
+{
+	bo->debugfs.flags = usage_flags | PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED;
+}
+#else
+static void panthor_gem_debugfs_bo_add(struct panthor_device *ptdev,
+				       struct panthor_gem_object *bo)
+{}
+static void panthor_gem_debugfs_bo_rm(struct panthor_gem_object *bo) {}
+static void panthor_gem_debugfs_set_usage_flags(struct panthor_gem_object *bo, u32 usage_flags) {}
+#endif
+
 static void panthor_gem_free_object(struct drm_gem_object *obj)
 {
 	struct panthor_gem_object *bo = to_panthor_bo(obj);
 	struct drm_gem_object *vm_root_gem = bo->exclusive_vm_root_gem;
 
+	panthor_gem_debugfs_bo_rm(bo);
+
+	/*
+	 * Label might have been allocated with kstrdup_const(),
+	 * we need to take that into account when freeing the memory
+	 */
+	kfree_const(bo->label.str);
+
+	mutex_destroy(&bo->label.lock);
+
 	drm_gem_free_mmap_offset(&bo->base.base);
 	mutex_destroy(&bo->gpuva_list_lock);
 	drm_gem_shmem_free(&bo->base);
@@ -67,17 +118,19 @@ out_free_bo:
  * @gpu_va: GPU address assigned when mapping to the VM.
  * If gpu_va == PANTHOR_VM_KERNEL_AUTO_VA, the virtual address will be
  * automatically allocated.
+ * @name: Descriptive label of the BO's contents
  *
  * Return: A valid pointer in case of success, an ERR_PTR() otherwise.
  */
 struct panthor_kernel_bo *
 panthor_kernel_bo_create(struct panthor_device *ptdev, struct panthor_vm *vm,
 			 size_t size, u32 bo_flags, u32 vm_map_flags,
-			 u64 gpu_va)
+			 u64 gpu_va, const char *name)
 {
 	struct drm_gem_shmem_object *obj;
 	struct panthor_kernel_bo *kbo;
 	struct panthor_gem_object *bo;
+	u32 debug_flags = PANTHOR_DEBUGFS_GEM_USAGE_FLAG_KERNEL;
 	int ret;
 
 	if (drm_WARN_ON(&ptdev->base, !vm))
@@ -97,6 +150,12 @@ panthor_kernel_bo_create(struct panthor_device *ptdev, struct panthor_vm *vm,
 	kbo->obj = &obj->base;
 	bo->flags = bo_flags;
 
+	if (vm == panthor_fw_vm(ptdev))
+		debug_flags |= PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED;
+
+	panthor_gem_kernel_bo_set_label(kbo, name);
+	panthor_gem_debugfs_set_usage_flags(to_panthor_bo(kbo->obj), debug_flags);
+
 	/* The system and GPU MMU page size might differ, which becomes a
 	 * problem for FW sections that need to be mapped at explicit address
 	 * since our PAGE_SIZE alignment might cover a VA range that's
@@ -129,17 +188,6 @@ err_free_bo:
 	return ERR_PTR(ret);
 }
 
-static int panthor_gem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
-{
-	struct panthor_gem_object *bo = to_panthor_bo(obj);
-
-	/* Don't allow mmap on objects that have the NO_MMAP flag set. */
-	if (bo->flags & DRM_PANTHOR_BO_NO_MMAP)
-		return -EINVAL;
-
-	return drm_gem_shmem_object_mmap(obj, vma);
-}
-
 static struct dma_buf *
 panthor_gem_prime_export(struct drm_gem_object *obj, int flags)
 {
@@ -155,7 +203,7 @@ static enum drm_gem_object_status panthor_gem_status(struct drm_gem_object *obj)
 	struct panthor_gem_object *bo = to_panthor_bo(obj);
 	enum drm_gem_object_status res = 0;
 
-	if (bo->base.base.import_attach || bo->base.pages)
+	if (drm_gem_is_imported(&bo->base.base) || bo->base.pages)
 		res |= DRM_GEM_OBJECT_RESIDENT;
 
 	return res;
@@ -169,7 +217,7 @@ static const struct drm_gem_object_funcs panthor_gem_funcs = {
 	.get_sg_table = drm_gem_shmem_object_get_sg_table,
 	.vmap = drm_gem_shmem_object_vmap,
 	.vunmap = drm_gem_shmem_object_vunmap,
-	.mmap = panthor_gem_mmap,
+	.mmap = drm_gem_shmem_object_mmap,
 	.status = panthor_gem_status,
 	.export = panthor_gem_prime_export,
 	.vm_ops = &drm_gem_shmem_vm_ops,
@@ -196,6 +244,9 @@ struct drm_gem_object *panthor_gem_create_object(struct drm_device *ddev, size_t
 	obj->base.map_wc = !ptdev->coherent;
 	mutex_init(&obj->gpuva_list_lock);
 	drm_gem_gpuva_set_lock(&obj->base.base, &obj->gpuva_list_lock);
+	mutex_init(&obj->label.lock);
+
+	panthor_gem_debugfs_bo_add(ptdev, obj);
 
 	return &obj->base.base;
 }
@@ -245,5 +296,153 @@ panthor_gem_create_with_handle(struct drm_file *file,
 	/* drop reference from allocate - handle holds it now. */
 	drm_gem_object_put(&shmem->base);
 
+	/*
+	 * No explicit flags are needed in the call below, since the
+	 * function internally sets the INITIALIZED bit for us.
+	 */
+	panthor_gem_debugfs_set_usage_flags(bo, 0);
+
 	return ret;
 }
+
+void
+panthor_gem_bo_set_label(struct drm_gem_object *obj, const char *label)
+{
+	struct panthor_gem_object *bo = to_panthor_bo(obj);
+	const char *old_label;
+
+	scoped_guard(mutex, &bo->label.lock) {
+		old_label = bo->label.str;
+		bo->label.str = label;
+	}
+
+	kfree_const(old_label);
+}
+
+void
+panthor_gem_kernel_bo_set_label(struct panthor_kernel_bo *bo, const char *label)
+{
+	const char *str;
+
+	/* We should never attempt labelling a UM-exposed GEM object */
+	if (drm_WARN_ON(bo->obj->dev, bo->obj->handle_count > 0))
+		return;
+
+	if (!label)
+		return;
+
+	str = kstrdup_const(label, GFP_KERNEL);
+	if (!str) {
+		/* Failing to allocate memory for a label isn't a fatal condition */
+		drm_warn(bo->obj->dev, "Not enough memory to allocate BO label");
+		return;
+	}
+
+	panthor_gem_bo_set_label(bo->obj, str);
+}
+
+#ifdef CONFIG_DEBUG_FS
+struct gem_size_totals {
+	size_t size;
+	size_t resident;
+	size_t reclaimable;
+};
+
+static void panthor_gem_debugfs_print_flag_names(struct seq_file *m)
+{
+	int len;
+	int i;
+
+	static const char * const gem_state_flags_names[] = {
+		[PANTHOR_DEBUGFS_GEM_STATE_IMPORTED_BIT] = "imported",
+		[PANTHOR_DEBUGFS_GEM_STATE_EXPORTED_BIT] = "exported",
+	};
+
+	static const char * const gem_usage_flags_names[] = {
+		[PANTHOR_DEBUGFS_GEM_USAGE_KERNEL_BIT] = "kernel",
+		[PANTHOR_DEBUGFS_GEM_USAGE_FW_MAPPED_BIT] = "fw-mapped",
+	};
+
+	seq_puts(m, "GEM state flags: ");
+	for (i = 0, len = ARRAY_SIZE(gem_state_flags_names); i < len; i++) {
+		if (!gem_state_flags_names[i])
+			continue;
+		seq_printf(m, "%s (0x%x)%s", gem_state_flags_names[i],
+			   (u32)BIT(i), (i < len - 1) ? ", " : "\n");
+	}
+
+	seq_puts(m, "GEM usage flags: ");
+	for (i = 0, len = ARRAY_SIZE(gem_usage_flags_names); i < len; i++) {
+		if (!gem_usage_flags_names[i])
+			continue;
+		seq_printf(m, "%s (0x%x)%s", gem_usage_flags_names[i],
+			   (u32)BIT(i), (i < len - 1) ? ", " : "\n\n");
+	}
+}
+
+static void panthor_gem_debugfs_bo_print(struct panthor_gem_object *bo,
+					 struct seq_file *m,
+					 struct gem_size_totals *totals)
+{
+	unsigned int refcount = kref_read(&bo->base.base.refcount);
+	char creator_info[32] = {};
+	size_t resident_size;
+	u32 gem_usage_flags = bo->debugfs.flags & (u32)~PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED;
+	u32 gem_state_flags = 0;
+
+	/* Skip BOs being destroyed. */
+	if (!refcount)
+		return;
+
+	resident_size = bo->base.pages ? bo->base.base.size : 0;
+
+	snprintf(creator_info, sizeof(creator_info),
+		 "%s/%d", bo->debugfs.creator.process_name, bo->debugfs.creator.tgid);
+	seq_printf(m, "%-32s%-16d%-16d%-16zd%-16zd0x%-16lx",
+		   creator_info,
+		   bo->base.base.name,
+		   refcount,
+		   bo->base.base.size,
+		   resident_size,
+		   drm_vma_node_start(&bo->base.base.vma_node));
+
+	if (bo->base.base.import_attach)
+		gem_state_flags |= PANTHOR_DEBUGFS_GEM_STATE_FLAG_IMPORTED;
+	if (bo->base.base.dma_buf)
+		gem_state_flags |= PANTHOR_DEBUGFS_GEM_STATE_FLAG_EXPORTED;
+
+	seq_printf(m, "0x%-8x 0x%-10x", gem_state_flags, gem_usage_flags);
+
+	scoped_guard(mutex, &bo->label.lock) {
+		seq_printf(m, "%s\n", bo->label.str ? : "");
+	}
+
+	totals->size += bo->base.base.size;
+	totals->resident += resident_size;
+	if (bo->base.madv > 0)
+		totals->reclaimable += resident_size;
+}
+
+void panthor_gem_debugfs_print_bos(struct panthor_device *ptdev,
+				   struct seq_file *m)
+{
+	struct gem_size_totals totals = {0};
+	struct panthor_gem_object *bo;
+
+	panthor_gem_debugfs_print_flag_names(m);
+
+	seq_puts(m, "created-by                      global-name     refcount        size            resident-size   file-offset       state      usage       label\n");
+	seq_puts(m, "----------------------------------------------------------------------------------------------------------------------------------------------\n");
+
+	scoped_guard(mutex, &ptdev->gems.lock) {
+		list_for_each_entry(bo, &ptdev->gems.node, debugfs.node) {
+			if (bo->debugfs.flags & PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED)
+				panthor_gem_debugfs_bo_print(bo, m, &totals);
+		}
+	}
+
+	seq_puts(m, "==============================================================================================================================================\n");
+	seq_printf(m, "Total size: %zd, Total resident: %zd, Total reclaimable: %zd\n",
+		   totals.size, totals.resident, totals.reclaimable);
+}
+#endif
diff --git a/drivers/gpu/drm/panthor/panthor_gem.h b/drivers/gpu/drm/panthor/panthor_gem.h
index 5749ef2ebe03..4dd732dcd59f 100644
--- a/drivers/gpu/drm/panthor/panthor_gem.h
+++ b/drivers/gpu/drm/panthor/panthor_gem.h
@@ -13,6 +13,56 @@
 
 struct panthor_vm;
 
+#define PANTHOR_BO_LABEL_MAXLEN	4096
+
+enum panthor_debugfs_gem_state_flags {
+	PANTHOR_DEBUGFS_GEM_STATE_IMPORTED_BIT = 0,
+	PANTHOR_DEBUGFS_GEM_STATE_EXPORTED_BIT = 1,
+
+	/** @PANTHOR_DEBUGFS_GEM_STATE_FLAG_IMPORTED: GEM BO is PRIME imported. */
+	PANTHOR_DEBUGFS_GEM_STATE_FLAG_IMPORTED = BIT(PANTHOR_DEBUGFS_GEM_STATE_IMPORTED_BIT),
+
+	/** @PANTHOR_DEBUGFS_GEM_STATE_FLAG_EXPORTED: GEM BO is PRIME exported. */
+	PANTHOR_DEBUGFS_GEM_STATE_FLAG_EXPORTED = BIT(PANTHOR_DEBUGFS_GEM_STATE_EXPORTED_BIT),
+};
+
+enum panthor_debugfs_gem_usage_flags {
+	PANTHOR_DEBUGFS_GEM_USAGE_KERNEL_BIT = 0,
+	PANTHOR_DEBUGFS_GEM_USAGE_FW_MAPPED_BIT = 1,
+
+	/** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_KERNEL: BO is for kernel use only. */
+	PANTHOR_DEBUGFS_GEM_USAGE_FLAG_KERNEL = BIT(PANTHOR_DEBUGFS_GEM_USAGE_KERNEL_BIT),
+
+	/** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED: BO is mapped on the FW VM. */
+	PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED = BIT(PANTHOR_DEBUGFS_GEM_USAGE_FW_MAPPED_BIT),
+
+	/** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED: BO is ready for DebugFS display. */
+	PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED = BIT(31),
+};
+
+/**
+ * struct panthor_gem_debugfs - GEM object's DebugFS list information
+ */
+struct panthor_gem_debugfs {
+	/**
+	 * @node: Node used to insert the object in the device-wide list of
+	 * GEM objects, to display information about it through a DebugFS file.
+	 */
+	struct list_head node;
+
+	/** @creator: Information about the UM process which created the GEM. */
+	struct {
+		/** @creator.process_name: Group leader name in owning thread's process */
+		char process_name[TASK_COMM_LEN];
+
+		/** @creator.tgid: PID of the thread's group leader within its process */
+		pid_t tgid;
+	} creator;
+
+	/** @flags: Combination of panthor_debugfs_gem_usage_flags flags */
+	u32 flags;
+};
+
 /**
  * struct panthor_gem_object - Driver specific GEM object.
  */
@@ -46,6 +96,24 @@ struct panthor_gem_object {
 
 	/** @flags: Combination of drm_panthor_bo_flags flags. */
 	u32 flags;
+
+	/**
+	 * @label: BO tagging fields. The label can be assigned within the
+	 * driver itself or through a specific IOCTL.
+	 */
+	struct {
+		/**
+		 * @label.str: Pointer to NULL-terminated string,
+		 */
+		const char *str;
+
+		/** @lock.str: Protects access to the @label.str field. */
+		struct mutex lock;
+	} label;
+
+#ifdef CONFIG_DEBUG_FS
+	struct panthor_gem_debugfs debugfs;
+#endif
 };
 
 /**
@@ -91,6 +159,9 @@ panthor_gem_create_with_handle(struct drm_file *file,
 			       struct panthor_vm *exclusive_vm,
 			       u64 *size, u32 flags, uint32_t *handle);
 
+void panthor_gem_bo_set_label(struct drm_gem_object *obj, const char *label);
+void panthor_gem_kernel_bo_set_label(struct panthor_kernel_bo *bo, const char *label);
+
 static inline u64
 panthor_kernel_bo_gpuva(struct panthor_kernel_bo *bo)
 {
@@ -112,7 +183,7 @@ panthor_kernel_bo_vmap(struct panthor_kernel_bo *bo)
 	if (bo->kmap)
 		return 0;
 
-	ret = drm_gem_vmap_unlocked(bo->obj, &map);
+	ret = drm_gem_vmap(bo->obj, &map);
 	if (ret)
 		return ret;
 
@@ -126,7 +197,7 @@ panthor_kernel_bo_vunmap(struct panthor_kernel_bo *bo)
 	if (bo->kmap) {
 		struct iosys_map map = IOSYS_MAP_INIT_VADDR(bo->kmap);
 
-		drm_gem_vunmap_unlocked(bo->obj, &map);
+		drm_gem_vunmap(bo->obj, &map);
 		bo->kmap = NULL;
 	}
 }
@@ -134,8 +205,13 @@ panthor_kernel_bo_vunmap(struct panthor_kernel_bo *bo)
 struct panthor_kernel_bo *
 panthor_kernel_bo_create(struct panthor_device *ptdev, struct panthor_vm *vm,
 			 size_t size, u32 bo_flags, u32 vm_map_flags,
-			 u64 gpu_va);
+			 u64 gpu_va, const char *name);
 
 void panthor_kernel_bo_destroy(struct panthor_kernel_bo *bo);
 
+#ifdef CONFIG_DEBUG_FS
+void panthor_gem_debugfs_print_bos(struct panthor_device *pfdev,
+				   struct seq_file *m);
+#endif
+
 #endif /* __PANTHOR_GEM_H__ */
diff --git a/drivers/gpu/drm/panthor/panthor_gpu.c b/drivers/gpu/drm/panthor/panthor_gpu.c
index 671049020afa..32d678a0114e 100644
--- a/drivers/gpu/drm/panthor/panthor_gpu.c
+++ b/drivers/gpu/drm/panthor/panthor_gpu.c
@@ -150,6 +150,8 @@ static void panthor_gpu_init_info(struct panthor_device *ptdev)
 
 static void panthor_gpu_irq_handler(struct panthor_device *ptdev, u32 status)
 {
+	gpu_write(ptdev, GPU_INT_CLEAR, status);
+
 	if (status & GPU_IRQ_FAULT) {
 		u32 fault_status = gpu_read(ptdev, GPU_FAULT_STATUS);
 		u64 address = ((u64)gpu_read(ptdev, GPU_FAULT_ADDR_HI) << 32) |
diff --git a/drivers/gpu/drm/panthor/panthor_heap.c b/drivers/gpu/drm/panthor/panthor_heap.c
index 3bdf61c14264..d236e9ceade4 100644
--- a/drivers/gpu/drm/panthor/panthor_heap.c
+++ b/drivers/gpu/drm/panthor/panthor_heap.c
@@ -151,7 +151,8 @@ static int panthor_alloc_heap_chunk(struct panthor_heap_pool *pool,
 	chunk->bo = panthor_kernel_bo_create(pool->ptdev, pool->vm, heap->chunk_size,
 					     DRM_PANTHOR_BO_NO_MMAP,
 					     DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC,
-					     PANTHOR_VM_KERNEL_AUTO_VA);
+					     PANTHOR_VM_KERNEL_AUTO_VA,
+					     "Tiler heap chunk");
 	if (IS_ERR(chunk->bo)) {
 		ret = PTR_ERR(chunk->bo);
 		goto err_free_chunk;
@@ -555,7 +556,8 @@ panthor_heap_pool_create(struct panthor_device *ptdev, struct panthor_vm *vm)
 	pool->gpu_contexts = panthor_kernel_bo_create(ptdev, vm, bosize,
 						      DRM_PANTHOR_BO_NO_MMAP,
 						      DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC,
-						      PANTHOR_VM_KERNEL_AUTO_VA);
+						      PANTHOR_VM_KERNEL_AUTO_VA,
+						      "Heap pool");
 	if (IS_ERR(pool->gpu_contexts)) {
 		ret = PTR_ERR(pool->gpu_contexts);
 		goto err_destroy_pool;
diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index 12a02e28f50f..6ca9a2642a4e 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -781,6 +781,7 @@ out_enable_as:
 	if (ptdev->mmu->as.faulty_mask & panthor_mmu_as_fault_mask(ptdev, as)) {
 		gpu_write(ptdev, MMU_INT_CLEAR, panthor_mmu_as_fault_mask(ptdev, as));
 		ptdev->mmu->as.faulty_mask &= ~panthor_mmu_as_fault_mask(ptdev, as);
+		ptdev->mmu->irq.mask |= panthor_mmu_as_fault_mask(ptdev, as);
 		gpu_write(ptdev, MMU_INT_MASK, ~ptdev->mmu->as.faulty_mask);
 	}
 
@@ -1103,7 +1104,7 @@ static void panthor_vm_bo_put(struct drm_gpuvm_bo *vm_bo)
 	/* If the vm_bo object was destroyed, release the pin reference that
 	 * was hold by this object.
 	 */
-	if (unpin && !bo->base.base.import_attach)
+	if (unpin && !drm_gem_is_imported(&bo->base.base))
 		drm_gem_shmem_unpin(&bo->base);
 
 	drm_gpuvm_put(vm);
@@ -1234,7 +1235,7 @@ static int panthor_vm_prepare_map_op_ctx(struct panthor_vm_op_ctx *op_ctx,
 	if (ret)
 		goto err_cleanup;
 
-	if (!bo->base.base.import_attach) {
+	if (!drm_gem_is_imported(&bo->base.base)) {
 		/* Pre-reserve the BO pages, so the map operation doesn't have to
 		 * allocate.
 		 */
@@ -1245,7 +1246,7 @@ static int panthor_vm_prepare_map_op_ctx(struct panthor_vm_op_ctx *op_ctx,
 
 	sgt = drm_gem_shmem_get_pages_sgt(&bo->base);
 	if (IS_ERR(sgt)) {
-		if (!bo->base.base.import_attach)
+		if (!drm_gem_is_imported(&bo->base.base))
 			drm_gem_shmem_unpin(&bo->base);
 
 		ret = PTR_ERR(sgt);
@@ -1256,7 +1257,7 @@ static int panthor_vm_prepare_map_op_ctx(struct panthor_vm_op_ctx *op_ctx,
 
 	preallocated_vm_bo = drm_gpuvm_bo_create(&vm->base, &bo->base.base);
 	if (!preallocated_vm_bo) {
-		if (!bo->base.base.import_attach)
+		if (!drm_gem_is_imported(&bo->base.base))
 			drm_gem_shmem_unpin(&bo->base);
 
 		ret = -ENOMEM;
@@ -1282,7 +1283,7 @@ static int panthor_vm_prepare_map_op_ctx(struct panthor_vm_op_ctx *op_ctx,
 	 * which will be released in panthor_vm_bo_put().
 	 */
 	if (preallocated_vm_bo != op_ctx->map.vm_bo &&
-	    !bo->base.base.import_attach)
+	    !drm_gem_is_imported(&bo->base.base))
 		drm_gem_shmem_unpin(&bo->base);
 
 	op_ctx->map.bo_offset = offset;
@@ -1709,11 +1710,17 @@ static void panthor_mmu_irq_handler(struct panthor_device *ptdev, u32 status)
 			access_type, access_type_name(ptdev, fault_status),
 			source_id);
 
+		/* We don't handle VM faults at the moment, so let's just clear the
+		 * interrupt and let the writer/reader crash.
+		 * Note that COMPLETED irqs are never cleared, but this is fine
+		 * because they are always masked.
+		 */
+		gpu_write(ptdev, MMU_INT_CLEAR, mask);
+
 		/* Ignore MMU interrupts on this AS until it's been
 		 * re-enabled.
 		 */
 		ptdev->mmu->irq.mask = new_int_mask;
-		gpu_write(ptdev, MMU_INT_MASK, new_int_mask);
 
 		if (ptdev->mmu->as.slots[as].vm)
 			ptdev->mmu->as.slots[as].vm->unhandled_fault = true;
diff --git a/drivers/gpu/drm/panthor/panthor_regs.h b/drivers/gpu/drm/panthor/panthor_regs.h
index b7b3b3add166..a7a323dc5cf9 100644
--- a/drivers/gpu/drm/panthor/panthor_regs.h
+++ b/drivers/gpu/drm/panthor/panthor_regs.h
@@ -133,8 +133,8 @@
 #define GPU_COHERENCY_PROT_BIT(name)			BIT(GPU_COHERENCY_  ## name)
 
 #define GPU_COHERENCY_PROTOCOL				0x304
-#define   GPU_COHERENCY_ACE				0
-#define   GPU_COHERENCY_ACE_LITE			1
+#define   GPU_COHERENCY_ACE_LITE			0
+#define   GPU_COHERENCY_ACE				1
 #define   GPU_COHERENCY_NONE				31
 
 #define MCU_CONTROL					0x700
diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
index 4d31d1967716..43ee57728de5 100644
--- a/drivers/gpu/drm/panthor/panthor_sched.c
+++ b/drivers/gpu/drm/panthor/panthor_sched.c
@@ -840,7 +840,7 @@ panthor_queue_put_syncwait_obj(struct panthor_queue *queue)
 	if (queue->syncwait.kmap) {
 		struct iosys_map map = IOSYS_MAP_INIT_VADDR(queue->syncwait.kmap);
 
-		drm_gem_vunmap_unlocked(queue->syncwait.obj, &map);
+		drm_gem_vunmap(queue->syncwait.obj, &map);
 		queue->syncwait.kmap = NULL;
 	}
 
@@ -866,7 +866,7 @@ panthor_queue_get_syncwait_obj(struct panthor_group *group, struct panthor_queue
 		goto err_put_syncwait_obj;
 
 	queue->syncwait.obj = &bo->base.base;
-	ret = drm_gem_vmap_unlocked(queue->syncwait.obj, &map);
+	ret = drm_gem_vmap(queue->syncwait.obj, &map);
 	if (drm_WARN_ON(&ptdev->base, ret))
 		goto err_put_syncwait_obj;
 
@@ -3332,7 +3332,8 @@ group_create_queue(struct panthor_group *group,
 						  DRM_PANTHOR_BO_NO_MMAP,
 						  DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC |
 						  DRM_PANTHOR_VM_BIND_OP_MAP_UNCACHED,
-						  PANTHOR_VM_KERNEL_AUTO_VA);
+						  PANTHOR_VM_KERNEL_AUTO_VA,
+						  "CS ring buffer");
 	if (IS_ERR(queue->ringbuf)) {
 		ret = PTR_ERR(queue->ringbuf);
 		goto err_free_queue;
@@ -3362,7 +3363,8 @@ group_create_queue(struct panthor_group *group,
 					 DRM_PANTHOR_BO_NO_MMAP,
 					 DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC |
 					 DRM_PANTHOR_VM_BIND_OP_MAP_UNCACHED,
-					 PANTHOR_VM_KERNEL_AUTO_VA);
+					 PANTHOR_VM_KERNEL_AUTO_VA,
+					 "Group job stats");
 
 	if (IS_ERR(queue->profiling.slots)) {
 		ret = PTR_ERR(queue->profiling.slots);
@@ -3493,7 +3495,8 @@ int panthor_group_create(struct panthor_file *pfile,
 						   DRM_PANTHOR_BO_NO_MMAP,
 						   DRM_PANTHOR_VM_BIND_OP_MAP_NOEXEC |
 						   DRM_PANTHOR_VM_BIND_OP_MAP_UNCACHED,
-						   PANTHOR_VM_KERNEL_AUTO_VA);
+						   PANTHOR_VM_KERNEL_AUTO_VA,
+						   "Group sync objects");
 	if (IS_ERR(group->syncobjs)) {
 		ret = PTR_ERR(group->syncobjs);
 		goto err_put_group;
diff --git a/drivers/gpu/drm/pl111/pl111_versatile.c b/drivers/gpu/drm/pl111/pl111_versatile.c
index 1e4b28d03f4d..5f460b296c0c 100644
--- a/drivers/gpu/drm/pl111/pl111_versatile.c
+++ b/drivers/gpu/drm/pl111/pl111_versatile.c
@@ -501,7 +501,7 @@ int pl111_versatile_init(struct device *dev, struct pl111_drm_dev_private *priv)
 	 * if we find it, it will take precedence. This is on the Integrator/AP
 	 * which only has this option for PL110 graphics.
 	 */
-	 if (versatile_clcd_type == INTEGRATOR_CLCD_CM) {
+	if (versatile_clcd_type == INTEGRATOR_CLCD_CM) {
 		np = of_find_matching_node_and_match(NULL, impd1_clcd_of_match,
 						     &clcd_id);
 		if (np)
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index 2db40789235c..a7caac5b8ac8 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -3825,8 +3825,7 @@ typedef struct _ATOM_DPCD_INFO
 // note2: From RV770, the memory is more than 32bit addressable, so we will change 
 //        ucTableFormatRevision=1,ucTableContentRevision=4, the structure remains
 //        exactly same as 1.1 and 1.2 (1.3 is never in use), but ulStartAddrUsedByFirmware 
-//        (in offset to start of memory address) is KB aligned instead of byte aligend.
-/***********************************************************************************/	
+//        (in offset to start of memory address) is KB aligned instead of byte aligned.
 // Note3:
 /* If we change usReserved to "usFBUsedbyDrvInKB", then to VBIOS this usFBUsedbyDrvInKB is a predefined, unchanged constant across VGA or non VGA adapter,
 for CAIL, The size of FB access area is known, only thing missing is the Offset of FB Access area, so we can  have:
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index fa78824931cc..3f3c360dce4b 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -501,8 +501,8 @@ bool radeon_dp_needs_link_train(struct radeon_connector *radeon_connector)
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	struct radeon_connector_atom_dig *dig = radeon_connector->con_priv;
 
-	if (drm_dp_dpcd_read_link_status(&radeon_connector->ddc_bus->aux, link_status)
-	    <= 0)
+	if (drm_dp_dpcd_read_link_status(&radeon_connector->ddc_bus->aux,
+					 link_status) < 0)
 		return false;
 	if (drm_dp_channel_eq_ok(link_status, dig->dp_lane_count))
 		return false;
@@ -678,7 +678,7 @@ static int radeon_dp_link_train_cr(struct radeon_dp_link_train_info *dp_info)
 		drm_dp_link_train_clock_recovery_delay(dp_info->aux, dp_info->dpcd);
 
 		if (drm_dp_dpcd_read_link_status(dp_info->aux,
-						 dp_info->link_status) <= 0) {
+						 dp_info->link_status) < 0) {
 			DRM_ERROR("displayport link status failed\n");
 			break;
 		}
@@ -741,7 +741,7 @@ static int radeon_dp_link_train_ce(struct radeon_dp_link_train_info *dp_info)
 		drm_dp_link_train_channel_eq_delay(dp_info->aux, dp_info->dpcd);
 
 		if (drm_dp_dpcd_read_link_status(dp_info->aux,
-						 dp_info->link_status) <= 0) {
+						 dp_info->link_status) < 0) {
 			DRM_ERROR("displayport link status failed\n");
 			break;
 		}
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 11a492f21157..51a3e0fc2f56 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -8548,7 +8548,7 @@ int cik_suspend(struct radeon_device *rdev)
  */
 int cik_init(struct radeon_device *rdev)
 {
-	struct radeon_ring *ring;
+	struct radeon_ring *ring, *ring_cp1, *ring_cp2;
 	int r;
 
 	/* Read BIOS */
@@ -8623,19 +8623,22 @@ int cik_init(struct radeon_device *rdev)
 	ring->ring_obj = NULL;
 	r600_ring_init(rdev, ring, 1024 * 1024);
 
-	ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
-	ring->ring_obj = NULL;
-	r600_ring_init(rdev, ring, 1024 * 1024);
-	r = radeon_doorbell_get(rdev, &ring->doorbell_index);
+	ring_cp1 = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
+	ring_cp2 = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
+	ring_cp1->ring_obj = NULL;
+	ring_cp2->ring_obj = NULL;
+	ring_cp1->doorbell_index = RADEON_MAX_DOORBELLS;
+	ring_cp2->doorbell_index = RADEON_MAX_DOORBELLS;
+
+	r600_ring_init(rdev, ring_cp1, 1024 * 1024);
+	r = radeon_doorbell_get(rdev, &ring_cp1->doorbell_index);
 	if (r)
 		return r;
 
-	ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
-	ring->ring_obj = NULL;
-	r600_ring_init(rdev, ring, 1024 * 1024);
-	r = radeon_doorbell_get(rdev, &ring->doorbell_index);
+	r600_ring_init(rdev, ring_cp2, 1024 * 1024);
+	r = radeon_doorbell_get(rdev, &ring_cp2->doorbell_index);
 	if (r)
-		return r;
+		goto out;
 
 	ring = &rdev->ring[R600_RING_TYPE_DMA_INDEX];
 	ring->ring_obj = NULL;
@@ -8653,12 +8656,16 @@ int cik_init(struct radeon_device *rdev)
 
 	r = r600_pcie_gart_init(rdev);
 	if (r)
-		return r;
+		goto out;
 
 	rdev->accel_working = true;
 	r = cik_startup(rdev);
 	if (r) {
 		dev_err(rdev->dev, "disabling GPU acceleration\n");
+		radeon_doorbell_free(rdev, ring_cp1->doorbell_index);
+		radeon_doorbell_free(rdev, ring_cp2->doorbell_index);
+		ring_cp1->doorbell_index = RADEON_MAX_DOORBELLS;
+		ring_cp2->doorbell_index = RADEON_MAX_DOORBELLS;
 		cik_cp_fini(rdev);
 		cik_sdma_fini(rdev);
 		cik_irq_fini(rdev);
@@ -8678,10 +8685,16 @@ int cik_init(struct radeon_device *rdev)
 	 */
 	if (!rdev->mc_fw && !(rdev->flags & RADEON_IS_IGP)) {
 		DRM_ERROR("radeon: MC ucode required for NI+.\n");
-		return -EINVAL;
+		r = -EINVAL;
+		goto out;
 	}
 
 	return 0;
+
+out:
+	radeon_doorbell_free(rdev, ring_cp1->doorbell_index);
+	radeon_doorbell_free(rdev, ring_cp2->doorbell_index);
+	return r;
 }
 
 /**
@@ -8695,6 +8708,7 @@ int cik_init(struct radeon_device *rdev)
  */
 void cik_fini(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring;
 	radeon_pm_fini(rdev);
 	cik_cp_fini(rdev);
 	cik_sdma_fini(rdev);
@@ -8708,6 +8722,10 @@ void cik_fini(struct radeon_device *rdev)
 	radeon_ib_pool_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	uvd_v1_0_fini(rdev);
+	ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
+	radeon_doorbell_free(rdev, ring->doorbell_index);
+	ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
+	radeon_doorbell_free(rdev, ring->doorbell_index);
 	radeon_uvd_fini(rdev);
 	radeon_vce_fini(rdev);
 	cik_pcie_gart_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index 661f374f5f27..9758f3a9df75 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -290,28 +290,6 @@ int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder)
 	return result;
 }
 
-/*
- * write the audio workaround status to the hardware
- */
-void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
-{
-	struct drm_device *dev = encoder->dev;
-	struct radeon_device *rdev = dev->dev_private;
-	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
-	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	uint32_t offset = dig->afmt->offset;
-	bool hdmi_audio_workaround = false; /* FIXME */
-	u32 value;
-
-	if (!hdmi_audio_workaround ||
-	    r600_hdmi_is_audio_buffer_filled(encoder))
-		value = 0; /* disable workaround */
-	else
-		value = HDMI0_AUDIO_TEST_EN; /* enable workaround */
-	WREG32_P(HDMI0_AUDIO_PACKET_CONTROL + offset,
-		 value, ~HDMI0_AUDIO_TEST_EN);
-}
-
 void r600_hdmi_audio_set_dto(struct radeon_device *rdev,
 			     struct radeon_crtc *crtc, unsigned int clock)
 {
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 8605c074d9f7..63c47585afbc 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -394,9 +394,6 @@ long radeon_fence_wait_timeout(struct radeon_fence *fence, bool interruptible, l
 int radeon_fence_wait(struct radeon_fence *fence, bool interruptible);
 int radeon_fence_wait_next(struct radeon_device *rdev, int ring);
 int radeon_fence_wait_empty(struct radeon_device *rdev, int ring);
-int radeon_fence_wait_any(struct radeon_device *rdev,
-			  struct radeon_fence **fences,
-			  bool intr);
 struct radeon_fence *radeon_fence_ref(struct radeon_fence *fence);
 void radeon_fence_unref(struct radeon_fence **fence);
 unsigned radeon_fence_count_emitted(struct radeon_device *rdev, int ring);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 8f5e07834fcc..9e697f10f9ca 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -401,7 +401,6 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock);
 void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder, void *buffer,
 				    size_t size);
 void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock);
-void r600_hdmi_audio_workaround(struct drm_encoder *encoder);
 int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder);
 void r600_hdmi_update_audio_settings(struct drm_encoder *encoder);
 u32 r600_get_xclk(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 64b26bfeafc9..b8e6202f1d5b 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -409,7 +409,6 @@ static int cmp_size_smaller_first(void *priv, const struct list_head *a,
  * radeon_cs_parser_fini() - clean parser states
  * @parser:	parser structure holding parsing context.
  * @error:	error number
- * @backoff:	indicator to backoff the reservation
  *
  * If error is set than unvalidate buffer, otherwise just free memory
  * used by parsing context.
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 8ff4f18b51a9..5b5b54e876d4 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -575,48 +575,6 @@ int radeon_fence_wait(struct radeon_fence *fence, bool intr)
 }
 
 /**
- * radeon_fence_wait_any - wait for a fence to signal on any ring
- *
- * @rdev: radeon device pointer
- * @fences: radeon fence object(s)
- * @intr: use interruptable sleep
- *
- * Wait for any requested fence to signal (all asics).  Fence
- * array is indexed by ring id.  @intr selects whether to use
- * interruptable (true) or non-interruptable (false) sleep when
- * waiting for the fences. Used by the suballocator.
- * Returns 0 if any fence has passed, error for all other cases.
- */
-int radeon_fence_wait_any(struct radeon_device *rdev,
-			  struct radeon_fence **fences,
-			  bool intr)
-{
-	uint64_t seq[RADEON_NUM_RINGS];
-	unsigned int i, num_rings = 0;
-	long r;
-
-	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-		seq[i] = 0;
-
-		if (!fences[i])
-			continue;
-
-		seq[i] = fences[i]->seq;
-		++num_rings;
-	}
-
-	/* nothing to wait for ? */
-	if (num_rings == 0)
-		return -ENOENT;
-
-	r = radeon_fence_wait_seq_timeout(rdev, seq, intr, MAX_SCHEDULE_TIMEOUT);
-	if (r < 0)
-		return r;
-
-	return 0;
-}
-
-/**
  * radeon_fence_wait_next - wait for the next fence to signal
  *
  * @rdev: radeon device pointer
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index 65a911ddd509..f9267b026f8d 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -1109,7 +1109,7 @@
 #define		MIN_POWER_SHIFT				0
 #define		MAX_POWER(x)				((x) << 16)
 #define		MAX_POWER_MASK				(0x3fff << 16)
-#define		MAX_POWER_SHIFT				0
+#define		MAX_POWER_SHIFT				16
 #define SQ_POWER_THROTTLE2                              0x8e5c
 #define		MAX_POWER_DELTA(x)			((x) << 0)
 #define		MAX_POWER_DELTA_MASK			(0x3fff << 0)
diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c b/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c
index 79b67c406bd6..93ba115d654f 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c
@@ -32,11 +32,6 @@ struct rcar_cmm {
 	} lut;
 };
 
-static inline int rcar_cmm_read(struct rcar_cmm *rcmm, u32 reg)
-{
-	return ioread32(rcmm->base + reg);
-}
-
 static inline void rcar_cmm_write(struct rcar_cmm *rcmm, u32 reg, u32 data)
 {
 	iowrite32(data, rcmm->base + reg);
diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c
index 70d8ad065bfa..4c8fe83dd610 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c
@@ -705,7 +705,7 @@ static int rcar_du_vsps_init(struct rcar_du_device *rcdu)
 		ret = of_parse_phandle_with_fixed_args(np, vsps_prop_name,
 						       cells, i, &args);
 		if (ret < 0)
-			goto error;
+			goto done;
 
 		/*
 		 * Add the VSP to the list or update the corresponding existing
@@ -743,13 +743,11 @@ static int rcar_du_vsps_init(struct rcar_du_device *rcdu)
 		vsp->dev = rcdu;
 
 		ret = rcar_du_vsp_init(vsp, vsps[i].np, vsps[i].crtcs_mask);
-		if (ret < 0)
-			goto error;
+		if (ret)
+			goto done;
 	}
 
-	return 0;
-
-error:
+done:
 	for (i = 0; i < ARRAY_SIZE(vsps); ++i)
 		of_node_put(vsps[i].np);
 
diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
index 380a855b832a..a9145253294f 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
@@ -634,6 +634,7 @@ static bool rcar_lvds_mode_fixup(struct drm_bridge *bridge,
 }
 
 static int rcar_lvds_attach(struct drm_bridge *bridge,
+			    struct drm_encoder *encoder,
 			    enum drm_bridge_attach_flags flags)
 {
 	struct rcar_lvds *lvds = bridge_to_rcar_lvds(bridge);
@@ -641,7 +642,7 @@ static int rcar_lvds_attach(struct drm_bridge *bridge,
 	if (!lvds->next_bridge)
 		return 0;
 
-	return drm_bridge_attach(bridge->encoder, lvds->next_bridge, bridge,
+	return drm_bridge_attach(encoder, lvds->next_bridge, bridge,
 				 flags);
 }
 
diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
index d1e626068065..7ab8be46c7f6 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
@@ -799,11 +799,12 @@ static void rcar_mipi_dsi_stop_video(struct rcar_mipi_dsi *dsi)
  */
 
 static int rcar_mipi_dsi_attach(struct drm_bridge *bridge,
+				struct drm_encoder *encoder,
 				enum drm_bridge_attach_flags flags)
 {
 	struct rcar_mipi_dsi *dsi = bridge_to_rcar_mipi_dsi(bridge);
 
-	return drm_bridge_attach(bridge->encoder, dsi->next_bridge, bridge,
+	return drm_bridge_attach(encoder, dsi->next_bridge, bridge,
 				 flags);
 }
 
diff --git a/drivers/gpu/drm/renesas/rz-du/Kconfig b/drivers/gpu/drm/renesas/rz-du/Kconfig
index 7c1817240846..e57536fd6f4d 100644
--- a/drivers/gpu/drm/renesas/rz-du/Kconfig
+++ b/drivers/gpu/drm/renesas/rz-du/Kconfig
@@ -14,10 +14,15 @@ config DRM_RZG2L_DU
 	  Choose this option if you have an RZ/G2L alike chipset.
 	  If M is selected the module will be called rzg2l-du-drm.
 
-config DRM_RZG2L_MIPI_DSI
-	tristate "RZ/G2L MIPI DSI Encoder Support"
-	depends on DRM && DRM_BRIDGE && OF
-	depends on ARCH_RENESAS || COMPILE_TEST
-	select DRM_MIPI_DSI
+config DRM_RZG2L_USE_MIPI_DSI
+	bool "RZ/G2L MIPI DSI Encoder Support"
+	depends on DRM_BRIDGE && OF
+	depends on DRM_RZG2L_DU || COMPILE_TEST
+	default DRM_RZG2L_DU
 	help
 	  Enable support for the RZ/G2L Display Unit embedded MIPI DSI encoders.
+
+config DRM_RZG2L_MIPI_DSI
+	def_tristate DRM_RZG2L_DU
+	depends on DRM_RZG2L_USE_MIPI_DSI
+	select DRM_MIPI_DSI
diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_drv.c b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_drv.c
index cbd9b9841267..5e40f0c1e7b0 100644
--- a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_drv.c
+++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_drv.c
@@ -79,7 +79,7 @@ DEFINE_DRM_GEM_DMA_FOPS(rzg2l_du_fops);
 
 static const struct drm_driver rzg2l_du_driver = {
 	.driver_features	= DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC,
-	.dumb_create		= rzg2l_du_dumb_create,
+	DRM_GEM_DMA_DRIVER_OPS_WITH_DUMB_CREATE(rzg2l_du_dumb_create),
 	DRM_FBDEV_DMA_DRIVER_OPS,
 	.fops			= &rzg2l_du_fops,
 	.name			= "rzg2l-du",
diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.c b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.c
index 90c6269ccd29..55a97691e9b2 100644
--- a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.c
+++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.c
@@ -36,23 +36,129 @@
 
 static const struct rzg2l_du_format_info rzg2l_du_format_infos[] = {
 	{
-		.fourcc = DRM_FORMAT_XRGB8888,
-		.v4l2 = V4L2_PIX_FMT_XBGR32,
-		.bpp = 32,
+		.fourcc = DRM_FORMAT_RGB332,
+		.v4l2 = V4L2_PIX_FMT_RGB332,
 		.planes = 1,
 		.hsub = 1,
 	}, {
-		.fourcc = DRM_FORMAT_ARGB8888,
-		.v4l2 = V4L2_PIX_FMT_ABGR32,
-		.bpp = 32,
+		.fourcc = DRM_FORMAT_ARGB4444,
+		.v4l2 = V4L2_PIX_FMT_ARGB444,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_XRGB4444,
+		.v4l2 = V4L2_PIX_FMT_XRGB444,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_ARGB1555,
+		.v4l2 = V4L2_PIX_FMT_ARGB555,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_XRGB1555,
+		.v4l2 = V4L2_PIX_FMT_XRGB555,
+		.planes = 1,
+	}, {
+		.fourcc = DRM_FORMAT_RGB565,
+		.v4l2 = V4L2_PIX_FMT_RGB565,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_BGR888,
+		.v4l2 = V4L2_PIX_FMT_RGB24,
 		.planes = 1,
 		.hsub = 1,
 	}, {
 		.fourcc = DRM_FORMAT_RGB888,
 		.v4l2 = V4L2_PIX_FMT_BGR24,
-		.bpp = 24,
 		.planes = 1,
 		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_BGRA8888,
+		.v4l2 = V4L2_PIX_FMT_ARGB32,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_BGRX8888,
+		.v4l2 = V4L2_PIX_FMT_XRGB32,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_ARGB8888,
+		.v4l2 = V4L2_PIX_FMT_ABGR32,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_XRGB8888,
+		.v4l2 = V4L2_PIX_FMT_XBGR32,
+		.planes = 1,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_UYVY,
+		.v4l2 = V4L2_PIX_FMT_UYVY,
+		.planes = 1,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YUYV,
+		.v4l2 = V4L2_PIX_FMT_YUYV,
+		.planes = 1,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YVYU,
+		.v4l2 = V4L2_PIX_FMT_YVYU,
+		.planes = 1,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_NV12,
+		.v4l2 = V4L2_PIX_FMT_NV12M,
+		.planes = 2,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_NV21,
+		.v4l2 = V4L2_PIX_FMT_NV21M,
+		.planes = 2,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_NV16,
+		.v4l2 = V4L2_PIX_FMT_NV16M,
+		.planes = 2,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_NV61,
+		.v4l2 = V4L2_PIX_FMT_NV61M,
+		.planes = 2,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YUV420,
+		.v4l2 = V4L2_PIX_FMT_YUV420M,
+		.planes = 3,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YVU420,
+		.v4l2 = V4L2_PIX_FMT_YVU420M,
+		.planes = 3,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YUV422,
+		.v4l2 = V4L2_PIX_FMT_YUV422M,
+		.planes = 3,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YVU422,
+		.v4l2 = V4L2_PIX_FMT_YVU422M,
+		.planes = 3,
+		.hsub = 2,
+	}, {
+		.fourcc = DRM_FORMAT_YUV444,
+		.v4l2 = V4L2_PIX_FMT_YUV444M,
+		.planes = 3,
+		.hsub = 1,
+	}, {
+		.fourcc = DRM_FORMAT_YVU444,
+		.v4l2 = V4L2_PIX_FMT_YVU444M,
+		.planes = 3,
+		.hsub = 1,
 	}
 };
 
diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.h b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.h
index 876e97cfbf45..e2c599f115c6 100644
--- a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.h
+++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_kms.h
@@ -23,7 +23,6 @@ struct sg_table;
 struct rzg2l_du_format_info {
 	u32 fourcc;
 	u32 v4l2;
-	unsigned int bpp;
 	unsigned int planes;
 	unsigned int hsub;
 };
diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_vsp.c b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_vsp.c
index 8643ff2eec46..040d4e4aff00 100644
--- a/drivers/gpu/drm/renesas/rz-du/rzg2l_du_vsp.c
+++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_du_vsp.c
@@ -340,6 +340,15 @@ int rzg2l_du_vsp_init(struct rzg2l_du_vsp *vsp, struct device_node *np,
 
 		drm_plane_helper_add(&plane->plane,
 				     &rzg2l_du_vsp_plane_helper_funcs);
+
+		drm_plane_create_alpha_property(&plane->plane);
+		drm_plane_create_zpos_property(&plane->plane, i, 0,
+					       num_planes - 1);
+
+		drm_plane_create_blend_mode_property(&plane->plane,
+					BIT(DRM_MODE_BLEND_PIXEL_NONE) |
+					BIT(DRM_MODE_BLEND_PREMULTI) |
+					BIT(DRM_MODE_BLEND_COVERAGE));
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c b/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c
index 4550c6d84796..dc6ab012cdb6 100644
--- a/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c
+++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c
@@ -479,7 +479,7 @@ static int rzg2l_mipi_dsi_start_video(struct rzg2l_mipi_dsi *dsi)
 	u32 status;
 	int ret;
 
-	/* Configuration for Blanking sequence and start video input*/
+	/* Configuration for Blanking sequence and start video input */
 	vich1set0r = VICH1SET0R_HFPNOLP | VICH1SET0R_HBPNOLP |
 		     VICH1SET0R_HSANOLP | VICH1SET0R_VSTART;
 	rzg2l_mipi_dsi_link_write(dsi, VICH1SET0R, vich1set0r);
@@ -523,11 +523,12 @@ err:
  */
 
 static int rzg2l_mipi_dsi_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct rzg2l_mipi_dsi *dsi = bridge_to_rzg2l_mipi_dsi(bridge);
 
-	return drm_bridge_attach(bridge->encoder, dsi->next_bridge, bridge,
+	return drm_bridge_attach(encoder, dsi->next_bridge, bridge,
 				 flags);
 }
 
diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
index 26c4410b2407..ab525668939a 100644
--- a/drivers/gpu/drm/rockchip/Kconfig
+++ b/drivers/gpu/drm/rockchip/Kconfig
@@ -2,12 +2,14 @@
 config DRM_ROCKCHIP
 	tristate "DRM Support for Rockchip"
 	depends on DRM && ROCKCHIP_IOMMU
+	depends on OF
 	select DRM_CLIENT_SELECTION
 	select DRM_GEM_DMA_HELPER
 	select DRM_KMS_HELPER
 	select DRM_PANEL
 	select VIDEOMODE_HELPERS
 	select DRM_ANALOGIX_DP if ROCKCHIP_ANALOGIX_DP
+	select DRM_DISPLAY_DP_AUX_BUS if ROCKCHIP_ANALOGIX_DP
 	select DRM_DW_HDMI if ROCKCHIP_DW_HDMI
 	select DRM_DW_HDMI_QP if ROCKCHIP_DW_HDMI_QP
 	select DRM_DW_MIPI_DSI if ROCKCHIP_DW_MIPI_DSI
diff --git a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c
index a8265a1bf9ff..d30f0983a53a 100644
--- a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c
+++ b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c
@@ -21,6 +21,7 @@
 #include <video/of_videomode.h>
 #include <video/videomode.h>
 
+#include <drm/display/drm_dp_aux_bus.h>
 #include <drm/display/drm_dp_helper.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
@@ -51,11 +52,13 @@ struct rockchip_grf_reg_field {
 /**
  * struct rockchip_dp_chip_data - splite the grf setting of kind of chips
  * @lcdc_sel: grf register field of lcdc_sel
+ * @edp_mode: grf register field of edp_mode
  * @chip_type: specific chip type
  * @reg: register base address
  */
 struct rockchip_dp_chip_data {
 	const struct rockchip_grf_reg_field lcdc_sel;
+	const struct rockchip_grf_reg_field edp_mode;
 	u32	chip_type;
 	u32	reg;
 };
@@ -70,6 +73,7 @@ struct rockchip_dp_device {
 	struct clk               *grfclk;
 	struct regmap            *grf;
 	struct reset_control     *rst;
+	struct reset_control     *apbrst;
 
 	const struct rockchip_dp_chip_data *data;
 
@@ -115,6 +119,10 @@ static int rockchip_dp_pre_init(struct rockchip_dp_device *dp)
 	usleep_range(10, 20);
 	reset_control_deassert(dp->rst);
 
+	reset_control_assert(dp->apbrst);
+	usleep_range(10, 20);
+	reset_control_deassert(dp->apbrst);
+
 	return 0;
 }
 
@@ -136,12 +144,21 @@ static int rockchip_dp_poweron(struct analogix_dp_plat_data *plat_data)
 		return ret;
 	}
 
+	ret = rockchip_grf_field_write(dp->grf, &dp->data->edp_mode, 1);
+	if (ret != 0)
+		DRM_DEV_ERROR(dp->dev, "failed to set edp mode %d\n", ret);
+
 	return ret;
 }
 
 static int rockchip_dp_powerdown(struct analogix_dp_plat_data *plat_data)
 {
 	struct rockchip_dp_device *dp = pdata_encoder_to_dp(plat_data);
+	int ret;
+
+	ret = rockchip_grf_field_write(dp->grf, &dp->data->edp_mode, 0);
+	if (ret != 0)
+		DRM_DEV_ERROR(dp->dev, "failed to set edp mode %d\n", ret);
 
 	clk_disable_unprepare(dp->pclk);
 
@@ -205,6 +222,10 @@ static void rockchip_dp_drm_encoder_enable(struct drm_encoder *encoder,
 	struct rockchip_dp_device *dp = encoder_to_dp(encoder);
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *old_crtc_state;
+	struct of_endpoint endpoint;
+	struct device_node *remote_port, *remote_port_parent;
+	char name[32];
+	u32 port_id;
 	int ret;
 
 	crtc = rockchip_dp_drm_get_new_crtc(encoder, state);
@@ -222,13 +243,27 @@ static void rockchip_dp_drm_encoder_enable(struct drm_encoder *encoder,
 		return;
 	}
 
-	ret = drm_of_encoder_active_endpoint_id(dp->dev->of_node, encoder);
+	ret = drm_of_encoder_active_endpoint(dp->dev->of_node, encoder, &endpoint);
 	if (ret < 0)
 		return;
 
-	DRM_DEV_DEBUG(dp->dev, "vop %s output to dp\n", (ret) ? "LIT" : "BIG");
+	remote_port_parent = of_graph_get_remote_port_parent(endpoint.local_node);
+	if (remote_port_parent) {
+		if (of_get_child_by_name(remote_port_parent, "ports")) {
+			remote_port = of_graph_get_remote_port(endpoint.local_node);
+			of_property_read_u32(remote_port, "reg", &port_id);
+			of_node_put(remote_port);
+			sprintf(name, "%s vp%d", remote_port_parent->full_name, port_id);
+		} else {
+			sprintf(name, "%s %s",
+				remote_port_parent->full_name, endpoint.id ? "vopl" : "vopb");
+		}
+		of_node_put(remote_port_parent);
+
+		DRM_DEV_DEBUG(dp->dev, "vop %s output to dp\n", (ret) ? "LIT" : "BIG");
+	}
 
-	ret = rockchip_grf_field_write(dp->grf, &dp->data->lcdc_sel, ret);
+	ret = rockchip_grf_field_write(dp->grf, &dp->data->lcdc_sel, endpoint.id);
 	if (ret != 0)
 		DRM_DEV_ERROR(dp->dev, "Could not write to GRF: %d\n", ret);
 
@@ -322,6 +357,12 @@ static int rockchip_dp_of_probe(struct rockchip_dp_device *dp)
 		return PTR_ERR(dp->rst);
 	}
 
+	dp->apbrst = devm_reset_control_get_optional(dev, "apb");
+	if (IS_ERR(dp->apbrst)) {
+		DRM_DEV_ERROR(dev, "failed to get apb reset control\n");
+		return PTR_ERR(dp->apbrst);
+	}
+
 	return 0;
 }
 
@@ -392,11 +433,28 @@ static const struct component_ops rockchip_dp_component_ops = {
 	.unbind = rockchip_dp_unbind,
 };
 
+static int rockchip_dp_link_panel(struct drm_dp_aux *aux)
+{
+	struct analogix_dp_plat_data *plat_data = analogix_dp_aux_to_plat_data(aux);
+	struct rockchip_dp_device *dp = pdata_encoder_to_dp(plat_data);
+	int ret;
+
+	/*
+	 * If drm_of_find_panel_or_bridge() returns -ENODEV, there may be no valid panel
+	 * or bridge nodes. The driver should go on for the driver-free bridge or the DP
+	 * mode applications.
+	 */
+	ret = drm_of_find_panel_or_bridge(dp->dev->of_node, 1, 0, &plat_data->panel, NULL);
+	if (ret && ret != -ENODEV)
+		return ret;
+
+	return component_add(dp->dev, &rockchip_dp_component_ops);
+}
+
 static int rockchip_dp_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	const struct rockchip_dp_chip_data *dp_data;
-	struct drm_panel *panel = NULL;
 	struct rockchip_dp_device *dp;
 	struct resource *res;
 	int i;
@@ -406,10 +464,6 @@ static int rockchip_dp_probe(struct platform_device *pdev)
 	if (!dp_data)
 		return -ENODEV;
 
-	ret = drm_of_find_panel_or_bridge(dev->of_node, 1, 0, &panel, NULL);
-	if (ret < 0 && ret != -ENODEV)
-		return ret;
-
 	dp = devm_kzalloc(dev, sizeof(*dp), GFP_KERNEL);
 	if (!dp)
 		return -ENOMEM;
@@ -432,7 +486,6 @@ static int rockchip_dp_probe(struct platform_device *pdev)
 
 	dp->dev = dev;
 	dp->adp = ERR_PTR(-ENODEV);
-	dp->plat_data.panel = panel;
 	dp->plat_data.dev_type = dp->data->chip_type;
 	dp->plat_data.power_on = rockchip_dp_poweron;
 	dp->plat_data.power_off = rockchip_dp_powerdown;
@@ -448,9 +501,20 @@ static int rockchip_dp_probe(struct platform_device *pdev)
 	if (IS_ERR(dp->adp))
 		return PTR_ERR(dp->adp);
 
-	ret = component_add(dev, &rockchip_dp_component_ops);
-	if (ret)
-		return ret;
+	ret = devm_of_dp_aux_populate_bus(analogix_dp_get_aux(dp->adp), rockchip_dp_link_panel);
+	if (ret) {
+		/*
+		 * If devm_of_dp_aux_populate_bus() returns -ENODEV, the done_probing() will not
+		 * be called because there are no EP devices. Then the rockchip_dp_link_panel()
+		 * will be called directly in order to support the other valid DT configurations.
+		 *
+		 * NOTE: The devm_of_dp_aux_populate_bus() is allowed to return -EPROBE_DEFER.
+		 */
+		if (ret != -ENODEV)
+			return dev_err_probe(dp->dev, ret, "failed to populate aux bus\n");
+
+		return rockchip_dp_link_panel(analogix_dp_get_aux(dp->adp));
+	}
 
 	return 0;
 }
@@ -501,9 +565,24 @@ static const struct rockchip_dp_chip_data rk3288_dp[] = {
 	{ /* sentinel */ }
 };
 
+static const struct rockchip_dp_chip_data rk3588_edp[] = {
+	{
+		.edp_mode = GRF_REG_FIELD(0x0000, 0, 0),
+		.chip_type = RK3588_EDP,
+		.reg = 0xfdec0000,
+	},
+	{
+		.edp_mode = GRF_REG_FIELD(0x0004, 0, 0),
+		.chip_type = RK3588_EDP,
+		.reg = 0xfded0000,
+	},
+	{ /* sentinel */ }
+};
+
 static const struct of_device_id rockchip_dp_dt_ids[] = {
 	{.compatible = "rockchip,rk3288-dp", .data = &rk3288_dp },
 	{.compatible = "rockchip,rk3399-edp", .data = &rk3399_edp },
+	{.compatible = "rockchip,rk3588-edp", .data = &rk3588_edp },
 	{}
 };
 MODULE_DEVICE_TABLE(of, rockchip_dp_dt_ids);
diff --git a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c
index 3d1dddb34603..7d531b6f4c09 100644
--- a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c
+++ b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c
@@ -94,6 +94,7 @@ struct rockchip_hdmi_qp {
 	struct gpio_desc *enable_gpio;
 	struct delayed_work hpd_work;
 	int port_id;
+	const struct rockchip_hdmi_qp_ctrl_ops *ctrl_ops;
 };
 
 struct rockchip_hdmi_qp_ctrl_ops {
@@ -461,6 +462,7 @@ static int dw_hdmi_qp_rockchip_bind(struct device *dev, struct device *master,
 		return -ENODEV;
 	}
 
+	hdmi->ctrl_ops = cfg->ctrl_ops;
 	hdmi->dev = &pdev->dev;
 	hdmi->port_id = -ENODEV;
 
@@ -600,27 +602,8 @@ static void dw_hdmi_qp_rockchip_remove(struct platform_device *pdev)
 static int __maybe_unused dw_hdmi_qp_rockchip_resume(struct device *dev)
 {
 	struct rockchip_hdmi_qp *hdmi = dev_get_drvdata(dev);
-	u32 val;
 
-	val = HIWORD_UPDATE(RK3588_SCLIN_MASK, RK3588_SCLIN_MASK) |
-	      HIWORD_UPDATE(RK3588_SDAIN_MASK, RK3588_SDAIN_MASK) |
-	      HIWORD_UPDATE(RK3588_MODE_MASK, RK3588_MODE_MASK) |
-	      HIWORD_UPDATE(RK3588_I2S_SEL_MASK, RK3588_I2S_SEL_MASK);
-	regmap_write(hdmi->vo_regmap,
-		     hdmi->port_id ? RK3588_GRF_VO1_CON6 : RK3588_GRF_VO1_CON3,
-		     val);
-
-	val = HIWORD_UPDATE(RK3588_SET_HPD_PATH_MASK,
-			    RK3588_SET_HPD_PATH_MASK);
-	regmap_write(hdmi->regmap, RK3588_GRF_SOC_CON7, val);
-
-	if (hdmi->port_id)
-		val = HIWORD_UPDATE(RK3588_HDMI1_GRANT_SEL,
-				    RK3588_HDMI1_GRANT_SEL);
-	else
-		val = HIWORD_UPDATE(RK3588_HDMI0_GRANT_SEL,
-				    RK3588_HDMI0_GRANT_SEL);
-	regmap_write(hdmi->vo_regmap, RK3588_GRF_VO1_CON9, val);
+	hdmi->ctrl_ops->io_init(hdmi);
 
 	dw_hdmi_qp_resume(dev, hdmi->hdmi);
 
diff --git a/drivers/gpu/drm/rockchip/inno_hdmi.c b/drivers/gpu/drm/rockchip/inno_hdmi.c
index 483ecfeaebb0..db4b4038e51d 100644
--- a/drivers/gpu/drm/rockchip/inno_hdmi.c
+++ b/drivers/gpu/drm/rockchip/inno_hdmi.c
@@ -10,10 +10,12 @@
 #include <linux/delay.h>
 #include <linux/err.h>
 #include <linux/hdmi.h>
+#include <linux/mfd/syscon.h>
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/platform_device.h>
+#include <linux/regmap.h>
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
@@ -29,8 +31,19 @@
 
 #include "inno_hdmi.h"
 
+#define HIWORD_UPDATE(val, mask)	((val) | (mask) << 16)
+
 #define INNO_HDMI_MIN_TMDS_CLOCK  25000000U
 
+#define RK3036_GRF_SOC_CON2	0x148
+#define RK3036_HDMI_PHSYNC	BIT(4)
+#define RK3036_HDMI_PVSYNC	BIT(5)
+
+enum inno_hdmi_dev_type {
+	RK3036_HDMI,
+	RK3128_HDMI,
+};
+
 struct inno_hdmi_phy_config {
 	unsigned long pixelclock;
 	u8 pre_emphasis;
@@ -38,6 +51,7 @@ struct inno_hdmi_phy_config {
 };
 
 struct inno_hdmi_variant {
+	enum inno_hdmi_dev_type dev_type;
 	struct inno_hdmi_phy_config *phy_configs;
 	struct inno_hdmi_phy_config *default_phy_config;
 };
@@ -58,6 +72,7 @@ struct inno_hdmi {
 	struct clk *pclk;
 	struct clk *refclk;
 	void __iomem *regs;
+	struct regmap *grf;
 
 	struct drm_connector	connector;
 	struct rockchip_encoder	encoder;
@@ -374,7 +389,15 @@ static int inno_hdmi_config_video_csc(struct inno_hdmi *hdmi)
 static int inno_hdmi_config_video_timing(struct inno_hdmi *hdmi,
 					 struct drm_display_mode *mode)
 {
-	int value;
+	int value, psync;
+
+	if (hdmi->variant->dev_type == RK3036_HDMI) {
+		psync = mode->flags & DRM_MODE_FLAG_PHSYNC ? RK3036_HDMI_PHSYNC : 0;
+		value = HIWORD_UPDATE(psync, RK3036_HDMI_PHSYNC);
+		psync = mode->flags & DRM_MODE_FLAG_PVSYNC ? RK3036_HDMI_PVSYNC : 0;
+		value |= HIWORD_UPDATE(psync, RK3036_HDMI_PVSYNC);
+		regmap_write(hdmi->grf, RK3036_GRF_SOC_CON2, value);
+	}
 
 	/* Set detail external video timing polarity and interlace mode */
 	value = v_EXTERANL_VIDEO(1);
@@ -885,32 +908,34 @@ static int inno_hdmi_bind(struct device *dev, struct device *master,
 		return PTR_ERR(hdmi->regs);
 
 	hdmi->pclk = devm_clk_get(hdmi->dev, "pclk");
-	if (IS_ERR(hdmi->pclk)) {
-		DRM_DEV_ERROR(hdmi->dev, "Unable to get HDMI pclk clk\n");
-		return PTR_ERR(hdmi->pclk);
-	}
+	if (IS_ERR(hdmi->pclk))
+		return dev_err_probe(dev, PTR_ERR(hdmi->pclk), "Unable to get HDMI pclk\n");
 
 	ret = clk_prepare_enable(hdmi->pclk);
-	if (ret) {
-		DRM_DEV_ERROR(hdmi->dev,
-			      "Cannot enable HDMI pclk clock: %d\n", ret);
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "Cannot enable HDMI pclk: %d\n", ret);
 
 	hdmi->refclk = devm_clk_get_optional(hdmi->dev, "ref");
 	if (IS_ERR(hdmi->refclk)) {
-		DRM_DEV_ERROR(hdmi->dev, "Unable to get HDMI reference clock\n");
-		ret = PTR_ERR(hdmi->refclk);
+		ret = dev_err_probe(dev, PTR_ERR(hdmi->refclk), "Unable to get HDMI refclk\n");
 		goto err_disable_pclk;
 	}
 
 	ret = clk_prepare_enable(hdmi->refclk);
 	if (ret) {
-		DRM_DEV_ERROR(hdmi->dev,
-			      "Cannot enable HDMI reference clock: %d\n", ret);
+		ret = dev_err_probe(dev, ret, "Cannot enable HDMI refclk: %d\n", ret);
 		goto err_disable_pclk;
 	}
 
+	if (hdmi->variant->dev_type == RK3036_HDMI) {
+		hdmi->grf = syscon_regmap_lookup_by_phandle(dev->of_node, "rockchip,grf");
+		if (IS_ERR(hdmi->grf)) {
+			ret = dev_err_probe(dev, PTR_ERR(hdmi->grf),
+					    "Unable to get rockchip,grf\n");
+			goto err_disable_clk;
+		}
+	}
+
 	irq = platform_get_irq(pdev, 0);
 	if (irq < 0) {
 		ret = irq;
@@ -995,11 +1020,13 @@ static void inno_hdmi_remove(struct platform_device *pdev)
 }
 
 static const struct inno_hdmi_variant rk3036_inno_hdmi_variant = {
+	.dev_type = RK3036_HDMI,
 	.phy_configs = rk3036_hdmi_phy_configs,
 	.default_phy_config = &rk3036_hdmi_phy_configs[1],
 };
 
 static const struct inno_hdmi_variant rk3128_inno_hdmi_variant = {
+	.dev_type = RK3128_HDMI,
 	.phy_configs = rk3128_hdmi_phy_configs,
 	.default_phy_config = &rk3128_hdmi_phy_configs[1],
 };
diff --git a/drivers/gpu/drm/rockchip/rk3066_hdmi.c b/drivers/gpu/drm/rockchip/rk3066_hdmi.c
index f7a460190313..e7875b52f298 100644
--- a/drivers/gpu/drm/rockchip/rk3066_hdmi.c
+++ b/drivers/gpu/drm/rockchip/rk3066_hdmi.c
@@ -5,6 +5,9 @@
  */
 
 #include <drm/drm_atomic.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/display/drm_hdmi_helper.h>
+#include <drm/display/drm_hdmi_state_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
 #include <drm/drm_probe_helper.h>
@@ -46,27 +49,20 @@ struct rk3066_hdmi {
 	struct clk *hclk;
 	void __iomem *regs;
 
-	struct drm_connector connector;
+	struct drm_bridge bridge;
+	struct drm_connector *connector;
 	struct rockchip_encoder encoder;
 
 	struct rk3066_hdmi_i2c *i2c;
-	struct i2c_adapter *ddc;
 
 	unsigned int tmdsclk;
 
 	struct hdmi_data_info hdmi_data;
 };
 
-static struct rk3066_hdmi *encoder_to_rk3066_hdmi(struct drm_encoder *encoder)
+static struct rk3066_hdmi *bridge_to_rk3066_hdmi(struct drm_bridge *bridge)
 {
-	struct rockchip_encoder *rkencoder = to_rockchip_encoder(encoder);
-
-	return container_of(rkencoder, struct rk3066_hdmi, encoder);
-}
-
-static struct rk3066_hdmi *connector_to_rk3066_hdmi(struct drm_connector *connector)
-{
-	return container_of(connector, struct rk3066_hdmi, connector);
+	return container_of(bridge, struct rk3066_hdmi, bridge);
 }
 
 static inline u8 hdmi_readb(struct rk3066_hdmi *hdmi, u16 offset)
@@ -161,57 +157,40 @@ static void rk3066_hdmi_set_power_mode(struct rk3066_hdmi *hdmi, int mode)
 		hdmi->tmdsclk = DEFAULT_PLLA_RATE;
 }
 
-static int
-rk3066_hdmi_upload_frame(struct rk3066_hdmi *hdmi, int setup_rc,
-			 union hdmi_infoframe *frame, u32 frame_index,
-			 u32 mask, u32 disable, u32 enable)
+static int rk3066_hdmi_bridge_clear_infoframe(struct drm_bridge *bridge,
+					      enum hdmi_infoframe_type type)
 {
-	if (mask)
-		hdmi_modb(hdmi, HDMI_CP_AUTO_SEND_CTRL, mask, disable);
-
-	hdmi_writeb(hdmi, HDMI_CP_BUF_INDEX, frame_index);
-
-	if (setup_rc >= 0) {
-		u8 packed_frame[HDMI_MAXIMUM_INFO_FRAME_SIZE];
-		ssize_t rc, i;
+	struct rk3066_hdmi *hdmi = bridge_to_rk3066_hdmi(bridge);
 
-		rc = hdmi_infoframe_pack(frame, packed_frame,
-					 sizeof(packed_frame));
-		if (rc < 0)
-			return rc;
-
-		for (i = 0; i < rc; i++)
-			hdmi_writeb(hdmi, HDMI_CP_BUF_ACC_HB0 + i * 4,
-				    packed_frame[i]);
-
-		if (mask)
-			hdmi_modb(hdmi, HDMI_CP_AUTO_SEND_CTRL, mask, enable);
+	if (type != HDMI_INFOFRAME_TYPE_AVI) {
+		drm_err(bridge->dev, "Unsupported infoframe type: %u\n", type);
+		return 0;
 	}
 
-	return setup_rc;
+	hdmi_writeb(hdmi, HDMI_CP_BUF_INDEX, HDMI_INFOFRAME_AVI);
+
+	return 0;
 }
 
-static int rk3066_hdmi_config_avi(struct rk3066_hdmi *hdmi,
-				  struct drm_display_mode *mode)
+static int
+rk3066_hdmi_bridge_write_infoframe(struct drm_bridge *bridge,
+				   enum hdmi_infoframe_type type,
+				   const u8 *buffer, size_t len)
 {
-	union hdmi_infoframe frame;
-	int rc;
+	struct rk3066_hdmi *hdmi = bridge_to_rk3066_hdmi(bridge);
+	ssize_t i;
 
-	rc = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
-						      &hdmi->connector, mode);
+	if (type != HDMI_INFOFRAME_TYPE_AVI) {
+		drm_err(bridge->dev, "Unsupported infoframe type: %u\n", type);
+		return 0;
+	}
 
-	if (hdmi->hdmi_data.enc_out_format == HDMI_COLORSPACE_YUV444)
-		frame.avi.colorspace = HDMI_COLORSPACE_YUV444;
-	else if (hdmi->hdmi_data.enc_out_format == HDMI_COLORSPACE_YUV422)
-		frame.avi.colorspace = HDMI_COLORSPACE_YUV422;
-	else
-		frame.avi.colorspace = HDMI_COLORSPACE_RGB;
+	rk3066_hdmi_bridge_clear_infoframe(bridge, type);
 
-	frame.avi.colorimetry = hdmi->hdmi_data.colorimetry;
-	frame.avi.scan_mode = HDMI_SCAN_MODE_NONE;
+	for (i = 0; i < len; i++)
+		hdmi_writeb(hdmi, HDMI_CP_BUF_ACC_HB0 + i * 4, buffer[i]);
 
-	return rk3066_hdmi_upload_frame(hdmi, rc, &frame,
-					HDMI_INFOFRAME_AVI, 0, 0, 0);
+	return 0;
 }
 
 static int rk3066_hdmi_config_video_timing(struct rk3066_hdmi *hdmi,
@@ -324,9 +303,27 @@ static void rk3066_hdmi_config_phy(struct rk3066_hdmi *hdmi)
 }
 
 static int rk3066_hdmi_setup(struct rk3066_hdmi *hdmi,
-			     struct drm_display_mode *mode)
+			     struct drm_atomic_state *state)
 {
-	struct drm_display_info *display = &hdmi->connector.display_info;
+	struct drm_bridge *bridge = &hdmi->bridge;
+	struct drm_connector *connector;
+	struct drm_display_info *display;
+	struct drm_display_mode *mode;
+	struct drm_connector_state *new_conn_state;
+	struct drm_crtc_state *new_crtc_state;
+
+	connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+
+	new_conn_state = drm_atomic_get_new_connector_state(state, connector);
+	if (WARN_ON(!new_conn_state))
+		return -EINVAL;
+
+	new_crtc_state = drm_atomic_get_new_crtc_state(state, new_conn_state->crtc);
+	if (WARN_ON(!new_crtc_state))
+		return -EINVAL;
+
+	display = &connector->display_info;
+	mode = &new_crtc_state->adjusted_mode;
 
 	hdmi->hdmi_data.vic = drm_match_cea_mode(mode);
 	hdmi->hdmi_data.enc_out_format = HDMI_COLORSPACE_RGB;
@@ -363,7 +360,7 @@ static int rk3066_hdmi_setup(struct rk3066_hdmi *hdmi,
 	if (display->is_hdmi) {
 		hdmi_modb(hdmi, HDMI_HDCP_CTRL, HDMI_VIDEO_MODE_MASK,
 			  HDMI_VIDEO_MODE_HDMI);
-		rk3066_hdmi_config_avi(hdmi, mode);
+		drm_atomic_helper_connector_hdmi_update_infoframes(connector, state);
 	} else {
 		hdmi_modb(hdmi, HDMI_HDCP_CTRL, HDMI_VIDEO_MODE_MASK, 0);
 	}
@@ -386,15 +383,15 @@ static int rk3066_hdmi_setup(struct rk3066_hdmi *hdmi,
 	return 0;
 }
 
-static void rk3066_hdmi_encoder_enable(struct drm_encoder *encoder,
-				       struct drm_atomic_state *state)
+static void rk3066_hdmi_bridge_atomic_enable(struct drm_bridge *bridge,
+					     struct drm_atomic_state *state)
 {
-	struct rk3066_hdmi *hdmi = encoder_to_rk3066_hdmi(encoder);
+	struct rk3066_hdmi *hdmi = bridge_to_rk3066_hdmi(bridge);
 	struct drm_connector_state *conn_state;
 	struct drm_crtc_state *crtc_state;
 	int mux, val;
 
-	conn_state = drm_atomic_get_new_connector_state(state, &hdmi->connector);
+	conn_state = drm_atomic_get_new_connector_state(state, hdmi->connector);
 	if (WARN_ON(!conn_state))
 		return;
 
@@ -402,7 +399,7 @@ static void rk3066_hdmi_encoder_enable(struct drm_encoder *encoder,
 	if (WARN_ON(!crtc_state))
 		return;
 
-	mux = drm_of_encoder_active_endpoint_id(hdmi->dev->of_node, encoder);
+	mux = drm_of_encoder_active_endpoint_id(hdmi->dev->of_node, &hdmi->encoder.encoder);
 	if (mux)
 		val = (HDMI_VIDEO_SEL << 16) | HDMI_VIDEO_SEL;
 	else
@@ -413,13 +410,13 @@ static void rk3066_hdmi_encoder_enable(struct drm_encoder *encoder,
 	DRM_DEV_DEBUG(hdmi->dev, "hdmi encoder enable select: vop%s\n",
 		      (mux) ? "1" : "0");
 
-	rk3066_hdmi_setup(hdmi, &crtc_state->adjusted_mode);
+	rk3066_hdmi_setup(hdmi, state);
 }
 
-static void rk3066_hdmi_encoder_disable(struct drm_encoder *encoder,
-					struct drm_atomic_state *state)
+static void rk3066_hdmi_bridge_atomic_disable(struct drm_bridge *bridge,
+					      struct drm_atomic_state *state)
 {
-	struct rk3066_hdmi *hdmi = encoder_to_rk3066_hdmi(encoder);
+	struct rk3066_hdmi *hdmi = bridge_to_rk3066_hdmi(bridge);
 
 	DRM_DEV_DEBUG(hdmi->dev, "hdmi encoder disable\n");
 
@@ -450,39 +447,34 @@ rk3066_hdmi_encoder_atomic_check(struct drm_encoder *encoder,
 static const
 struct drm_encoder_helper_funcs rk3066_hdmi_encoder_helper_funcs = {
 	.atomic_check   = rk3066_hdmi_encoder_atomic_check,
-	.atomic_enable  = rk3066_hdmi_encoder_enable,
-	.atomic_disable = rk3066_hdmi_encoder_disable,
 };
 
 static enum drm_connector_status
-rk3066_hdmi_connector_detect(struct drm_connector *connector, bool force)
+rk3066_hdmi_bridge_detect(struct drm_bridge *bridge)
 {
-	struct rk3066_hdmi *hdmi = connector_to_rk3066_hdmi(connector);
+	struct rk3066_hdmi *hdmi = bridge_to_rk3066_hdmi(bridge);
 
 	return (hdmi_readb(hdmi, HDMI_HPG_MENS_STA) & HDMI_HPG_IN_STATUS_HIGH) ?
 		connector_status_connected : connector_status_disconnected;
 }
 
-static int rk3066_hdmi_connector_get_modes(struct drm_connector *connector)
+static const struct drm_edid *
+rk3066_hdmi_bridge_edid_read(struct drm_bridge *bridge, struct drm_connector *connector)
 {
-	struct rk3066_hdmi *hdmi = connector_to_rk3066_hdmi(connector);
+	struct rk3066_hdmi *hdmi = bridge_to_rk3066_hdmi(bridge);
 	const struct drm_edid *drm_edid;
-	int ret = 0;
-
-	if (!hdmi->ddc)
-		return 0;
 
-	drm_edid = drm_edid_read_ddc(connector, hdmi->ddc);
-	drm_edid_connector_update(connector, drm_edid);
-	ret = drm_edid_connector_add_modes(connector);
-	drm_edid_free(drm_edid);
+	drm_edid = drm_edid_read_ddc(connector, bridge->ddc);
+	if (!drm_edid)
+		dev_dbg(hdmi->dev, "failed to get edid\n");
 
-	return ret;
+	return drm_edid;
 }
 
 static enum drm_mode_status
-rk3066_hdmi_connector_mode_valid(struct drm_connector *connector,
-				 const struct drm_display_mode *mode)
+rk3066_hdmi_bridge_mode_valid(struct drm_bridge *bridge,
+			      const struct drm_display_info *info,
+			      const struct drm_display_mode *mode)
 {
 	u32 vic = drm_match_cea_mode(mode);
 
@@ -492,82 +484,19 @@ rk3066_hdmi_connector_mode_valid(struct drm_connector *connector,
 		return MODE_BAD;
 }
 
-static struct drm_encoder *
-rk3066_hdmi_connector_best_encoder(struct drm_connector *connector)
-{
-	struct rk3066_hdmi *hdmi = connector_to_rk3066_hdmi(connector);
-
-	return &hdmi->encoder.encoder;
-}
-
-static int
-rk3066_hdmi_probe_single_connector_modes(struct drm_connector *connector,
-					 uint32_t maxX, uint32_t maxY)
-{
-	if (maxX > 1920)
-		maxX = 1920;
-	if (maxY > 1080)
-		maxY = 1080;
-
-	return drm_helper_probe_single_connector_modes(connector, maxX, maxY);
-}
-
-static void rk3066_hdmi_connector_destroy(struct drm_connector *connector)
-{
-	drm_connector_unregister(connector);
-	drm_connector_cleanup(connector);
-}
-
-static const struct drm_connector_funcs rk3066_hdmi_connector_funcs = {
-	.fill_modes = rk3066_hdmi_probe_single_connector_modes,
-	.detect = rk3066_hdmi_connector_detect,
-	.destroy = rk3066_hdmi_connector_destroy,
-	.reset = drm_atomic_helper_connector_reset,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-};
-
-static const
-struct drm_connector_helper_funcs rk3066_hdmi_connector_helper_funcs = {
-	.get_modes = rk3066_hdmi_connector_get_modes,
-	.mode_valid = rk3066_hdmi_connector_mode_valid,
-	.best_encoder = rk3066_hdmi_connector_best_encoder,
+static const struct drm_bridge_funcs rk3066_hdmi_bridge_funcs = {
+	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
+	.atomic_reset = drm_atomic_helper_bridge_reset,
+	.atomic_enable = rk3066_hdmi_bridge_atomic_enable,
+	.atomic_disable = rk3066_hdmi_bridge_atomic_disable,
+	.detect = rk3066_hdmi_bridge_detect,
+	.edid_read = rk3066_hdmi_bridge_edid_read,
+	.hdmi_clear_infoframe = rk3066_hdmi_bridge_clear_infoframe,
+	.hdmi_write_infoframe = rk3066_hdmi_bridge_write_infoframe,
+	.mode_valid = rk3066_hdmi_bridge_mode_valid,
 };
 
-static int
-rk3066_hdmi_register(struct drm_device *drm, struct rk3066_hdmi *hdmi)
-{
-	struct drm_encoder *encoder = &hdmi->encoder.encoder;
-	struct device *dev = hdmi->dev;
-
-	encoder->possible_crtcs =
-		drm_of_find_possible_crtcs(drm, dev->of_node);
-
-	/*
-	 * If we failed to find the CRTC(s) which this encoder is
-	 * supposed to be connected to, it's because the CRTC has
-	 * not been registered yet.  Defer probing, and hope that
-	 * the required CRTC is added later.
-	 */
-	if (encoder->possible_crtcs == 0)
-		return -EPROBE_DEFER;
-
-	drm_encoder_helper_add(encoder, &rk3066_hdmi_encoder_helper_funcs);
-	drm_simple_encoder_init(drm, encoder, DRM_MODE_ENCODER_TMDS);
-
-	hdmi->connector.polled = DRM_CONNECTOR_POLL_HPD;
-
-	drm_connector_helper_add(&hdmi->connector,
-				 &rk3066_hdmi_connector_helper_funcs);
-	drm_connector_init_with_ddc(drm, &hdmi->connector,
-				    &rk3066_hdmi_connector_funcs,
-				    DRM_MODE_CONNECTOR_HDMIA,
-				    hdmi->ddc);
-
-	drm_connector_attach_encoder(&hdmi->connector, encoder);
-
-	return 0;
-}
 
 static irqreturn_t rk3066_hdmi_hardirq(int irq, void *dev_id)
 {
@@ -597,7 +526,7 @@ static irqreturn_t rk3066_hdmi_irq(int irq, void *dev_id)
 {
 	struct rk3066_hdmi *hdmi = dev_id;
 
-	drm_helper_hpd_irq_event(hdmi->connector.dev);
+	drm_helper_hpd_irq_event(hdmi->connector->dev);
 
 	return IRQ_HANDLED;
 }
@@ -720,7 +649,7 @@ static struct i2c_adapter *rk3066_hdmi_i2c_adapter(struct rk3066_hdmi *hdmi)
 	strscpy(adap->name, "RK3066 HDMI", sizeof(adap->name));
 	i2c_set_adapdata(adap, hdmi);
 
-	ret = i2c_add_adapter(adap);
+	ret = devm_i2c_add_adapter(hdmi->dev, adap);
 	if (ret) {
 		DRM_DEV_ERROR(hdmi->dev, "cannot add %s I2C adapter\n",
 			      adap->name);
@@ -735,6 +664,66 @@ static struct i2c_adapter *rk3066_hdmi_i2c_adapter(struct rk3066_hdmi *hdmi)
 	return adap;
 }
 
+static int
+rk3066_hdmi_register(struct drm_device *drm, struct rk3066_hdmi *hdmi)
+{
+	struct drm_encoder *encoder = &hdmi->encoder.encoder;
+	struct device *dev = hdmi->dev;
+	int ret;
+
+	encoder->possible_crtcs =
+		drm_of_find_possible_crtcs(drm, dev->of_node);
+
+	/*
+	 * If we failed to find the CRTC(s) which this encoder is
+	 * supposed to be connected to, it's because the CRTC has
+	 * not been registered yet.  Defer probing, and hope that
+	 * the required CRTC is added later.
+	 */
+	if (encoder->possible_crtcs == 0)
+		return -EPROBE_DEFER;
+
+	drm_encoder_helper_add(encoder, &rk3066_hdmi_encoder_helper_funcs);
+	drm_simple_encoder_init(drm, encoder, DRM_MODE_ENCODER_TMDS);
+
+	hdmi->bridge.driver_private = hdmi;
+	hdmi->bridge.funcs = &rk3066_hdmi_bridge_funcs;
+	hdmi->bridge.ops = DRM_BRIDGE_OP_DETECT |
+			   DRM_BRIDGE_OP_EDID |
+			   DRM_BRIDGE_OP_HDMI |
+			   DRM_BRIDGE_OP_HPD;
+	hdmi->bridge.of_node = hdmi->dev->of_node;
+	hdmi->bridge.type = DRM_MODE_CONNECTOR_HDMIA;
+	hdmi->bridge.vendor = "Rockchip";
+	hdmi->bridge.product = "RK3066 HDMI";
+
+	hdmi->bridge.ddc = rk3066_hdmi_i2c_adapter(hdmi);
+	if (IS_ERR(hdmi->bridge.ddc))
+		return PTR_ERR(hdmi->bridge.ddc);
+
+	if (IS_ERR(hdmi->bridge.ddc))
+		return PTR_ERR(hdmi->bridge.ddc);
+
+	ret = devm_drm_bridge_add(dev, &hdmi->bridge);
+	if (ret)
+		return ret;
+
+	ret = drm_bridge_attach(encoder, &hdmi->bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+	if (ret)
+		return ret;
+
+	hdmi->connector = drm_bridge_connector_init(drm, encoder);
+	if (IS_ERR(hdmi->connector)) {
+		ret = PTR_ERR(hdmi->connector);
+		dev_err(hdmi->dev, "failed to init bridge connector: %d\n", ret);
+		return ret;
+	}
+
+	drm_connector_attach_encoder(hdmi->connector, encoder);
+
+	return 0;
+}
+
 static int rk3066_hdmi_bind(struct device *dev, struct device *master,
 			    void *data)
 {
@@ -781,13 +770,6 @@ static int rk3066_hdmi_bind(struct device *dev, struct device *master,
 	/* internal hclk = hdmi_hclk / 25 */
 	hdmi_writeb(hdmi, HDMI_INTERNAL_CLK_DIVIDER, 25);
 
-	hdmi->ddc = rk3066_hdmi_i2c_adapter(hdmi);
-	if (IS_ERR(hdmi->ddc)) {
-		ret = PTR_ERR(hdmi->ddc);
-		hdmi->ddc = NULL;
-		goto err_disable_hclk;
-	}
-
 	rk3066_hdmi_set_power_mode(hdmi, HDMI_SYS_POWER_MODE_B);
 	usleep_range(999, 1000);
 	hdmi_writeb(hdmi, HDMI_INTR_MASK1, HDMI_INTR_HOTPLUG);
@@ -798,7 +780,7 @@ static int rk3066_hdmi_bind(struct device *dev, struct device *master,
 
 	ret = rk3066_hdmi_register(drm, hdmi);
 	if (ret)
-		goto err_disable_i2c;
+		goto err_disable_hclk;
 
 	dev_set_drvdata(dev, hdmi);
 
@@ -813,10 +795,7 @@ static int rk3066_hdmi_bind(struct device *dev, struct device *master,
 	return 0;
 
 err_cleanup_hdmi:
-	hdmi->connector.funcs->destroy(&hdmi->connector);
 	hdmi->encoder.encoder.funcs->destroy(&hdmi->encoder.encoder);
-err_disable_i2c:
-	i2c_put_adapter(hdmi->ddc);
 err_disable_hclk:
 	clk_disable_unprepare(hdmi->hclk);
 
@@ -828,10 +807,8 @@ static void rk3066_hdmi_unbind(struct device *dev, struct device *master,
 {
 	struct rk3066_hdmi *hdmi = dev_get_drvdata(dev);
 
-	hdmi->connector.funcs->destroy(&hdmi->connector);
 	hdmi->encoder.encoder.funcs->destroy(&hdmi->encoder.encoder);
 
-	i2c_put_adapter(hdmi->ddc);
 	clk_disable_unprepare(hdmi->hclk);
 }
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
index e3596e2b557d..ba6b0528d1e5 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
@@ -733,11 +733,10 @@ static void vop_crtc_atomic_disable(struct drm_crtc *crtc,
 
 	WARN_ON(vop->event);
 
-	if (crtc->state->self_refresh_active)
+	if (crtc->state->self_refresh_active) {
 		rockchip_drm_set_win_enabled(crtc, false);
-
-	if (crtc->state->self_refresh_active)
 		goto out;
+	}
 
 	mutex_lock(&vop->vop_lock);
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h
index 680bedbb770e..fc3ecb9fcd95 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h
@@ -710,6 +710,7 @@ enum dst_factor_mode {
 
 #define VOP2_COLOR_KEY_MASK				BIT(31)
 
+#define RK3568_OVL_CTRL__LAYERSEL_REGDONE_SEL		GENMASK(31, 30)
 #define RK3568_OVL_CTRL__LAYERSEL_REGDONE_IMD		BIT(28)
 #define RK3568_OVL_CTRL__YUV_MODE(vp)			BIT(vp)
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c b/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c
index 14958d6b3d2e..32c4ed685739 100644
--- a/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c
+++ b/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c
@@ -1754,9 +1754,9 @@ static unsigned long rk3588_set_intf_mux(struct vop2_video_port *vp, int id, u32
 		dip |= FIELD_PREP(RK3588_DSP_IF_POL__DP0_PIN_POL, polflags);
 		break;
 	case ROCKCHIP_VOP2_EP_DP1:
-		die &= ~RK3588_SYS_DSP_INFACE_EN_MIPI1_MUX;
-		die |= RK3588_SYS_DSP_INFACE_EN_MIPI1 |
-			   FIELD_PREP(RK3588_SYS_DSP_INFACE_EN_MIPI1_MUX, vp->id);
+		die &= ~RK3588_SYS_DSP_INFACE_EN_DP1_MUX;
+		die |= RK3588_SYS_DSP_INFACE_EN_DP1 |
+			   FIELD_PREP(RK3588_SYS_DSP_INFACE_EN_DP1_MUX, vp->id);
 		dip &= ~RK3588_DSP_IF_POL__DP1_PIN_POL;
 		dip |= FIELD_PREP(RK3588_DSP_IF_POL__DP1_PIN_POL, polflags);
 		break;
@@ -2070,7 +2070,10 @@ static void rk3568_vop2_setup_layer_mixer(struct vop2_video_port *vp)
 	struct rockchip_crtc_state *vcstate = to_rockchip_crtc_state(vp->crtc.state);
 
 	ovl_ctrl = vop2_readl(vop2, RK3568_OVL_CTRL);
-	ovl_ctrl |= RK3568_OVL_CTRL__LAYERSEL_REGDONE_IMD;
+	ovl_ctrl &= ~RK3568_OVL_CTRL__LAYERSEL_REGDONE_IMD;
+	ovl_ctrl &= ~RK3568_OVL_CTRL__LAYERSEL_REGDONE_SEL;
+	ovl_ctrl |= FIELD_PREP(RK3568_OVL_CTRL__LAYERSEL_REGDONE_SEL, vp->id);
+
 	if (vcstate->yuv_overlay)
 		ovl_ctrl |= RK3568_OVL_CTRL__YUV_MODE(vp->id);
 	else
diff --git a/drivers/gpu/drm/rockchip/rockchip_vop_reg.c b/drivers/gpu/drm/rockchip/rockchip_vop_reg.c
index 4e2099d86517..d1f788763318 100644
--- a/drivers/gpu/drm/rockchip/rockchip_vop_reg.c
+++ b/drivers/gpu/drm/rockchip/rockchip_vop_reg.c
@@ -906,21 +906,21 @@ static const struct vop_data rk3366_vop = {
 
 static const struct vop_output rk3399_output = {
 	.dp_dclk_pol = VOP_REG(RK3399_DSP_CTRL1, 0x1, 19),
-	.rgb_dclk_pol = VOP_REG(RK3368_DSP_CTRL1, 0x1, 19),
-	.hdmi_dclk_pol = VOP_REG(RK3368_DSP_CTRL1, 0x1, 23),
-	.edp_dclk_pol = VOP_REG(RK3368_DSP_CTRL1, 0x1, 27),
-	.mipi_dclk_pol = VOP_REG(RK3368_DSP_CTRL1, 0x1, 31),
+	.rgb_dclk_pol = VOP_REG(RK3399_DSP_CTRL1, 0x1, 19),
+	.hdmi_dclk_pol = VOP_REG(RK3399_DSP_CTRL1, 0x1, 23),
+	.edp_dclk_pol = VOP_REG(RK3399_DSP_CTRL1, 0x1, 27),
+	.mipi_dclk_pol = VOP_REG(RK3399_DSP_CTRL1, 0x1, 31),
 	.dp_pin_pol = VOP_REG(RK3399_DSP_CTRL1, 0x7, 16),
-	.rgb_pin_pol = VOP_REG(RK3368_DSP_CTRL1, 0x7, 16),
-	.hdmi_pin_pol = VOP_REG(RK3368_DSP_CTRL1, 0x7, 20),
-	.edp_pin_pol = VOP_REG(RK3368_DSP_CTRL1, 0x7, 24),
-	.mipi_pin_pol = VOP_REG(RK3368_DSP_CTRL1, 0x7, 28),
+	.rgb_pin_pol = VOP_REG(RK3399_DSP_CTRL1, 0x7, 16),
+	.hdmi_pin_pol = VOP_REG(RK3399_DSP_CTRL1, 0x7, 20),
+	.edp_pin_pol = VOP_REG(RK3399_DSP_CTRL1, 0x7, 24),
+	.mipi_pin_pol = VOP_REG(RK3399_DSP_CTRL1, 0x7, 28),
 	.dp_en = VOP_REG(RK3399_SYS_CTRL, 0x1, 11),
-	.rgb_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 12),
-	.hdmi_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 13),
-	.edp_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 14),
-	.mipi_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 15),
-	.mipi_dual_channel_en = VOP_REG(RK3288_SYS_CTRL, 0x1, 3),
+	.rgb_en = VOP_REG(RK3399_SYS_CTRL, 0x1, 12),
+	.hdmi_en = VOP_REG(RK3399_SYS_CTRL, 0x1, 13),
+	.edp_en = VOP_REG(RK3399_SYS_CTRL, 0x1, 14),
+	.mipi_en = VOP_REG(RK3399_SYS_CTRL, 0x1, 15),
+	.mipi_dual_channel_en = VOP_REG(RK3399_SYS_CTRL, 0x1, 3),
 };
 
 static const struct vop_common rk3399_common = {
@@ -975,23 +975,23 @@ static const struct vop_win_phy rk3399_win0_data = {
 	.data_formats = formats_win_full_10,
 	.nformats = ARRAY_SIZE(formats_win_full_10),
 	.format_modifiers = format_modifiers_win_full_afbc,
-	.enable = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 0),
-	.format = VOP_REG(RK3288_WIN0_CTRL0, 0x7, 1),
-	.fmt_10 = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 4),
-	.rb_swap = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 12),
-	.uv_swap = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 15),
-	.x_mir_en = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 21),
-	.y_mir_en = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 22),
-	.act_info = VOP_REG(RK3288_WIN0_ACT_INFO, 0x1fff1fff, 0),
-	.dsp_info = VOP_REG(RK3288_WIN0_DSP_INFO, 0x0fff0fff, 0),
-	.dsp_st = VOP_REG(RK3288_WIN0_DSP_ST, 0x1fff1fff, 0),
-	.yrgb_mst = VOP_REG(RK3288_WIN0_YRGB_MST, 0xffffffff, 0),
-	.uv_mst = VOP_REG(RK3288_WIN0_CBR_MST, 0xffffffff, 0),
-	.yrgb_vir = VOP_REG(RK3288_WIN0_VIR, 0x3fff, 0),
-	.uv_vir = VOP_REG(RK3288_WIN0_VIR, 0x3fff, 16),
-	.src_alpha_ctl = VOP_REG(RK3288_WIN0_SRC_ALPHA_CTRL, 0xff, 0),
-	.dst_alpha_ctl = VOP_REG(RK3288_WIN0_DST_ALPHA_CTRL, 0xff, 0),
-	.channel = VOP_REG(RK3288_WIN0_CTRL2, 0xff, 0),
+	.enable = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 0),
+	.format = VOP_REG(RK3399_WIN0_CTRL0, 0x7, 1),
+	.fmt_10 = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 4),
+	.rb_swap = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 12),
+	.uv_swap = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 15),
+	.x_mir_en = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 21),
+	.y_mir_en = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 22),
+	.act_info = VOP_REG(RK3399_WIN0_ACT_INFO, 0x1fff1fff, 0),
+	.dsp_info = VOP_REG(RK3399_WIN0_DSP_INFO, 0x0fff0fff, 0),
+	.dsp_st = VOP_REG(RK3399_WIN0_DSP_ST, 0x1fff1fff, 0),
+	.yrgb_mst = VOP_REG(RK3399_WIN0_YRGB_MST, 0xffffffff, 0),
+	.uv_mst = VOP_REG(RK3399_WIN0_CBR_MST, 0xffffffff, 0),
+	.yrgb_vir = VOP_REG(RK3399_WIN0_VIR, 0x3fff, 0),
+	.uv_vir = VOP_REG(RK3399_WIN0_VIR, 0x3fff, 16),
+	.src_alpha_ctl = VOP_REG(RK3399_WIN0_SRC_ALPHA_CTRL, 0xff, 0),
+	.dst_alpha_ctl = VOP_REG(RK3399_WIN0_DST_ALPHA_CTRL, 0xff, 0),
+	.channel = VOP_REG(RK3399_WIN0_CTRL2, 0xff, 0),
 };
 
 static const struct vop_win_phy rk3399_win1_data = {
@@ -999,23 +999,23 @@ static const struct vop_win_phy rk3399_win1_data = {
 	.data_formats = formats_win_full_10,
 	.nformats = ARRAY_SIZE(formats_win_full_10),
 	.format_modifiers = format_modifiers_win_full,
-	.enable = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 0),
-	.format = VOP_REG(RK3288_WIN0_CTRL0, 0x7, 1),
-	.fmt_10 = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 4),
-	.rb_swap = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 12),
-	.uv_swap = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 15),
-	.x_mir_en = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 21),
-	.y_mir_en = VOP_REG(RK3288_WIN0_CTRL0, 0x1, 22),
-	.act_info = VOP_REG(RK3288_WIN0_ACT_INFO, 0x1fff1fff, 0),
-	.dsp_info = VOP_REG(RK3288_WIN0_DSP_INFO, 0x0fff0fff, 0),
-	.dsp_st = VOP_REG(RK3288_WIN0_DSP_ST, 0x1fff1fff, 0),
-	.yrgb_mst = VOP_REG(RK3288_WIN0_YRGB_MST, 0xffffffff, 0),
-	.uv_mst = VOP_REG(RK3288_WIN0_CBR_MST, 0xffffffff, 0),
-	.yrgb_vir = VOP_REG(RK3288_WIN0_VIR, 0x3fff, 0),
-	.uv_vir = VOP_REG(RK3288_WIN0_VIR, 0x3fff, 16),
-	.src_alpha_ctl = VOP_REG(RK3288_WIN0_SRC_ALPHA_CTRL, 0xff, 0),
-	.dst_alpha_ctl = VOP_REG(RK3288_WIN0_DST_ALPHA_CTRL, 0xff, 0),
-	.channel = VOP_REG(RK3288_WIN0_CTRL2, 0xff, 0),
+	.enable = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 0),
+	.format = VOP_REG(RK3399_WIN0_CTRL0, 0x7, 1),
+	.fmt_10 = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 4),
+	.rb_swap = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 12),
+	.uv_swap = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 15),
+	.x_mir_en = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 21),
+	.y_mir_en = VOP_REG(RK3399_WIN0_CTRL0, 0x1, 22),
+	.act_info = VOP_REG(RK3399_WIN0_ACT_INFO, 0x1fff1fff, 0),
+	.dsp_info = VOP_REG(RK3399_WIN0_DSP_INFO, 0x0fff0fff, 0),
+	.dsp_st = VOP_REG(RK3399_WIN0_DSP_ST, 0x1fff1fff, 0),
+	.yrgb_mst = VOP_REG(RK3399_WIN0_YRGB_MST, 0xffffffff, 0),
+	.uv_mst = VOP_REG(RK3399_WIN0_CBR_MST, 0xffffffff, 0),
+	.yrgb_vir = VOP_REG(RK3399_WIN0_VIR, 0x3fff, 0),
+	.uv_vir = VOP_REG(RK3399_WIN0_VIR, 0x3fff, 16),
+	.src_alpha_ctl = VOP_REG(RK3399_WIN0_SRC_ALPHA_CTRL, 0xff, 0),
+	.dst_alpha_ctl = VOP_REG(RK3399_WIN0_DST_ALPHA_CTRL, 0xff, 0),
+	.channel = VOP_REG(RK3399_WIN0_CTRL2, 0xff, 0),
 };
 
 /*
diff --git a/drivers/gpu/drm/scheduler/.kunitconfig b/drivers/gpu/drm/scheduler/.kunitconfig
new file mode 100644
index 000000000000..cece53609fcf
--- /dev/null
+++ b/drivers/gpu/drm/scheduler/.kunitconfig
@@ -0,0 +1,12 @@
+CONFIG_KUNIT=y
+CONFIG_DRM=y
+CONFIG_DRM_SCHED_KUNIT_TEST=y
+CONFIG_EXPERT=y
+CONFIG_DEBUG_SPINLOCK=y
+CONFIG_DEBUG_MUTEXES=y
+CONFIG_DEBUG_ATOMIC_SLEEP=y
+CONFIG_LOCK_DEBUGGING_SUPPORT=y
+CONFIG_PROVE_LOCKING=y
+CONFIG_LOCKDEP=y
+CONFIG_DEBUG_LOCKDEP=y
+CONFIG_DEBUG_LIST=y
diff --git a/drivers/gpu/drm/scheduler/Makefile b/drivers/gpu/drm/scheduler/Makefile
index 53863621829f..6e13e4c63e9d 100644
--- a/drivers/gpu/drm/scheduler/Makefile
+++ b/drivers/gpu/drm/scheduler/Makefile
@@ -23,3 +23,5 @@
 gpu-sched-y := sched_main.o sched_fence.o sched_entity.o
 
 obj-$(CONFIG_DRM_SCHED) += gpu-sched.o
+
+obj-$(CONFIG_DRM_SCHED_KUNIT_TEST) += tests/
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index bfea608a7106..829579c41c6b 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -828,11 +828,15 @@ EXPORT_SYMBOL(drm_sched_job_init);
  *
  * This arms a scheduler job for execution. Specifically it initializes the
  * &drm_sched_job.s_fence of @job, so that it can be attached to struct dma_resv
- * or other places that need to track the completion of this job.
+ * or other places that need to track the completion of this job. It also
+ * initializes sequence numbers, which are fundamental for fence ordering.
  *
  * Refer to drm_sched_entity_push_job() documentation for locking
  * considerations.
  *
+ * Once this function was called, you *must* submit @job with
+ * drm_sched_entity_push_job().
+ *
  * This can only be called if drm_sched_job_init() succeeded.
  */
 void drm_sched_job_arm(struct drm_sched_job *job)
@@ -1015,13 +1019,14 @@ EXPORT_SYMBOL(drm_sched_job_has_dependency);
  * Cleans up the resources allocated with drm_sched_job_init().
  *
  * Drivers should call this from their error unwind code if @job is aborted
- * before it was submitted to an entity with drm_sched_entity_push_job().
+ * before drm_sched_job_arm() is called.
  *
- * Since calling drm_sched_job_arm() causes the job's fences to be initialized,
- * it is up to the driver to ensure that fences that were exposed to external
- * parties get signaled. drm_sched_job_cleanup() does not ensure this.
+ * drm_sched_job_arm() is a point of no return since it initializes the fences
+ * and their sequence number etc. Once that function has been called, you *must*
+ * submit it with drm_sched_entity_push_job() and cannot simply abort it by
+ * calling drm_sched_job_cleanup().
  *
- * This function must also be called in &struct drm_sched_backend_ops.free_job
+ * This function should be called in the &drm_sched_backend_ops.free_job callback.
  */
 void drm_sched_job_cleanup(struct drm_sched_job *job)
 {
@@ -1029,10 +1034,15 @@ void drm_sched_job_cleanup(struct drm_sched_job *job)
 	unsigned long index;
 
 	if (kref_read(&job->s_fence->finished.refcount)) {
-		/* drm_sched_job_arm() has been called */
+		/* The job has been processed by the scheduler, i.e.,
+		 * drm_sched_job_arm() and drm_sched_entity_push_job() have
+		 * been called.
+		 */
 		dma_fence_put(&job->s_fence->finished);
 	} else {
-		/* aborted job before arming */
+		/* The job was aborted before it has been committed to be run;
+		 * notably, drm_sched_job_arm() has not been called.
+		 */
 		drm_sched_fence_free(job->s_fence);
 	}
 
@@ -1220,20 +1230,23 @@ static void drm_sched_run_job_work(struct work_struct *w)
 	drm_sched_job_begin(sched_job);
 
 	trace_drm_run_job(sched_job, entity);
+	/*
+	 * The run_job() callback must by definition return a fence whose
+	 * refcount has been incremented for the scheduler already.
+	 */
 	fence = sched->ops->run_job(sched_job);
 	complete_all(&entity->entity_idle);
 	drm_sched_fence_scheduled(s_fence, fence);
 
 	if (!IS_ERR_OR_NULL(fence)) {
-		/* Drop for original kref_init of the fence */
-		dma_fence_put(fence);
-
 		r = dma_fence_add_callback(fence, &sched_job->cb,
 					   drm_sched_job_done_cb);
 		if (r == -ENOENT)
 			drm_sched_job_done(sched_job, fence->error);
 		else if (r)
 			DRM_DEV_ERROR(sched->dev, "fence add callback failed (%d)\n", r);
+
+		dma_fence_put(fence);
 	} else {
 		drm_sched_job_done(sched_job, IS_ERR(fence) ?
 				   PTR_ERR(fence) : 0);
diff --git a/drivers/gpu/drm/scheduler/tests/Makefile b/drivers/gpu/drm/scheduler/tests/Makefile
new file mode 100644
index 000000000000..5bf707bad373
--- /dev/null
+++ b/drivers/gpu/drm/scheduler/tests/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0
+
+drm-sched-tests-y := \
+        mock_scheduler.o \
+        tests_basic.o
+
+obj-$(CONFIG_DRM_SCHED_KUNIT_TEST) += drm-sched-tests.o
diff --git a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c
new file mode 100644
index 000000000000..f999c8859cf7
--- /dev/null
+++ b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c
@@ -0,0 +1,359 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2025 Valve Corporation */
+
+#include "sched_tests.h"
+
+/*
+ * Here we implement the mock "GPU" (or the scheduler backend) which is used by
+ * the DRM scheduler unit tests in order to exercise the core functionality.
+ *
+ * Test cases are implemented in a separate file.
+ */
+
+/**
+ * drm_mock_sched_entity_new - Create a new mock scheduler entity
+ *
+ * @test: KUnit test owning the entity
+ * @priority: Scheduling priority
+ * @sched: Mock scheduler on which the entity can be scheduled
+ *
+ * Returns: New mock scheduler entity with allocation managed by the test
+ */
+struct drm_mock_sched_entity *
+drm_mock_sched_entity_new(struct kunit *test,
+			  enum drm_sched_priority priority,
+			  struct drm_mock_scheduler *sched)
+{
+	struct drm_mock_sched_entity *entity;
+	struct drm_gpu_scheduler *drm_sched;
+	int ret;
+
+	entity = kunit_kzalloc(test, sizeof(*entity), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, entity);
+
+	drm_sched = &sched->base;
+	ret = drm_sched_entity_init(&entity->base,
+				    priority,
+				    &drm_sched, 1,
+				    NULL);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	entity->test = test;
+
+	return entity;
+}
+
+/**
+ * drm_mock_sched_entity_free - Destroys a mock scheduler entity
+ *
+ * @entity: Entity to destroy
+ *
+ * To be used from the test cases once done with the entity.
+ */
+void drm_mock_sched_entity_free(struct drm_mock_sched_entity *entity)
+{
+	drm_sched_entity_destroy(&entity->base);
+}
+
+static void drm_mock_sched_job_complete(struct drm_mock_sched_job *job)
+{
+	struct drm_mock_scheduler *sched =
+		drm_sched_to_mock_sched(job->base.sched);
+
+	lockdep_assert_held(&sched->lock);
+
+	job->flags |= DRM_MOCK_SCHED_JOB_DONE;
+	list_move_tail(&job->link, &sched->done_list);
+	dma_fence_signal(&job->hw_fence);
+	complete(&job->done);
+}
+
+static enum hrtimer_restart
+drm_mock_sched_job_signal_timer(struct hrtimer *hrtimer)
+{
+	struct drm_mock_sched_job *job =
+		container_of(hrtimer, typeof(*job), timer);
+	struct drm_mock_scheduler *sched =
+		drm_sched_to_mock_sched(job->base.sched);
+	struct drm_mock_sched_job *next;
+	ktime_t now = ktime_get();
+	unsigned long flags;
+	LIST_HEAD(signal);
+
+	spin_lock_irqsave(&sched->lock, flags);
+	list_for_each_entry_safe(job, next, &sched->job_list, link) {
+		if (!job->duration_us)
+			break;
+
+		if (ktime_before(now, job->finish_at))
+			break;
+
+		sched->hw_timeline.cur_seqno = job->hw_fence.seqno;
+		drm_mock_sched_job_complete(job);
+	}
+	spin_unlock_irqrestore(&sched->lock, flags);
+
+	return HRTIMER_NORESTART;
+}
+
+/**
+ * drm_mock_sched_job_new - Create a new mock scheduler job
+ *
+ * @test: KUnit test owning the job
+ * @entity: Scheduler entity of the job
+ *
+ * Returns: New mock scheduler job with allocation managed by the test
+ */
+struct drm_mock_sched_job *
+drm_mock_sched_job_new(struct kunit *test,
+		       struct drm_mock_sched_entity *entity)
+{
+	struct drm_mock_sched_job *job;
+	int ret;
+
+	job = kunit_kzalloc(test, sizeof(*job), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, job);
+
+	ret = drm_sched_job_init(&job->base,
+				 &entity->base,
+				 1,
+				 NULL);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	job->test = test;
+
+	init_completion(&job->done);
+	spin_lock_init(&job->lock);
+	INIT_LIST_HEAD(&job->link);
+	hrtimer_setup(&job->timer, drm_mock_sched_job_signal_timer,
+		      CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+
+	return job;
+}
+
+static const char *drm_mock_sched_hw_fence_driver_name(struct dma_fence *fence)
+{
+	return "drm_mock_sched";
+}
+
+static const char *
+drm_mock_sched_hw_fence_timeline_name(struct dma_fence *fence)
+{
+	struct drm_mock_sched_job *job =
+		container_of(fence, typeof(*job), hw_fence);
+
+	return (const char *)job->base.sched->name;
+}
+
+static void drm_mock_sched_hw_fence_release(struct dma_fence *fence)
+{
+	struct drm_mock_sched_job *job =
+		container_of(fence, typeof(*job), hw_fence);
+
+	hrtimer_cancel(&job->timer);
+
+	/* Containing job is freed by the kunit framework */
+}
+
+static const struct dma_fence_ops drm_mock_sched_hw_fence_ops = {
+	.get_driver_name = drm_mock_sched_hw_fence_driver_name,
+	.get_timeline_name = drm_mock_sched_hw_fence_timeline_name,
+	.release = drm_mock_sched_hw_fence_release,
+};
+
+static struct dma_fence *mock_sched_run_job(struct drm_sched_job *sched_job)
+{
+	struct drm_mock_scheduler *sched =
+		drm_sched_to_mock_sched(sched_job->sched);
+	struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job);
+
+	dma_fence_init(&job->hw_fence,
+		       &drm_mock_sched_hw_fence_ops,
+		       &job->lock,
+		       sched->hw_timeline.context,
+		       atomic_inc_return(&sched->hw_timeline.next_seqno));
+
+	dma_fence_get(&job->hw_fence); /* Reference for the job_list */
+
+	spin_lock_irq(&sched->lock);
+	if (job->duration_us) {
+		ktime_t prev_finish_at = 0;
+
+		if (!list_empty(&sched->job_list)) {
+			struct drm_mock_sched_job *prev =
+				list_last_entry(&sched->job_list, typeof(*prev),
+						link);
+
+			prev_finish_at = prev->finish_at;
+		}
+
+		if (!prev_finish_at)
+			prev_finish_at = ktime_get();
+
+		job->finish_at = ktime_add_us(prev_finish_at, job->duration_us);
+	}
+	list_add_tail(&job->link, &sched->job_list);
+	if (job->finish_at)
+		hrtimer_start(&job->timer, job->finish_at, HRTIMER_MODE_ABS);
+	spin_unlock_irq(&sched->lock);
+
+	return &job->hw_fence;
+}
+
+static enum drm_gpu_sched_stat
+mock_sched_timedout_job(struct drm_sched_job *sched_job)
+{
+	struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job);
+
+	job->flags |= DRM_MOCK_SCHED_JOB_TIMEDOUT;
+
+	return DRM_GPU_SCHED_STAT_NOMINAL;
+}
+
+static void mock_sched_free_job(struct drm_sched_job *sched_job)
+{
+	struct drm_mock_scheduler *sched =
+			drm_sched_to_mock_sched(sched_job->sched);
+	struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job);
+	unsigned long flags;
+
+	/* Remove from the scheduler done list. */
+	spin_lock_irqsave(&sched->lock, flags);
+	list_del(&job->link);
+	spin_unlock_irqrestore(&sched->lock, flags);
+	dma_fence_put(&job->hw_fence);
+
+	drm_sched_job_cleanup(sched_job);
+
+	/* Mock job itself is freed by the kunit framework. */
+}
+
+static const struct drm_sched_backend_ops drm_mock_scheduler_ops = {
+	.run_job = mock_sched_run_job,
+	.timedout_job = mock_sched_timedout_job,
+	.free_job = mock_sched_free_job
+};
+
+/**
+ * drm_mock_sched_new - Create a new mock scheduler
+ *
+ * @test: KUnit test owning the job
+ * @timeout: Job timeout to set
+ *
+ * Returns: New mock scheduler with allocation managed by the test
+ */
+struct drm_mock_scheduler *drm_mock_sched_new(struct kunit *test, long timeout)
+{
+	struct drm_sched_init_args args = {
+		.ops		= &drm_mock_scheduler_ops,
+		.num_rqs	= DRM_SCHED_PRIORITY_COUNT,
+		.credit_limit	= U32_MAX,
+		.hang_limit	= 1,
+		.timeout	= timeout,
+		.name		= "drm-mock-scheduler",
+	};
+	struct drm_mock_scheduler *sched;
+	int ret;
+
+	sched = kunit_kzalloc(test, sizeof(*sched), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, sched);
+
+	ret = drm_sched_init(&sched->base, &args);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	sched->test = test;
+	sched->hw_timeline.context = dma_fence_context_alloc(1);
+	atomic_set(&sched->hw_timeline.next_seqno, 0);
+	INIT_LIST_HEAD(&sched->job_list);
+	INIT_LIST_HEAD(&sched->done_list);
+	spin_lock_init(&sched->lock);
+
+	return sched;
+}
+
+/**
+ * drm_mock_sched_fini - Destroys a mock scheduler
+ *
+ * @sched: Scheduler to destroy
+ *
+ * To be used from the test cases once done with the scheduler.
+ */
+void drm_mock_sched_fini(struct drm_mock_scheduler *sched)
+{
+	struct drm_mock_sched_job *job, *next;
+	unsigned long flags;
+	LIST_HEAD(list);
+
+	drm_sched_wqueue_stop(&sched->base);
+
+	/* Force complete all unfinished jobs. */
+	spin_lock_irqsave(&sched->lock, flags);
+	list_for_each_entry_safe(job, next, &sched->job_list, link)
+		list_move_tail(&job->link, &list);
+	spin_unlock_irqrestore(&sched->lock, flags);
+
+	list_for_each_entry(job, &list, link)
+		hrtimer_cancel(&job->timer);
+
+	spin_lock_irqsave(&sched->lock, flags);
+	list_for_each_entry_safe(job, next, &list, link)
+		drm_mock_sched_job_complete(job);
+	spin_unlock_irqrestore(&sched->lock, flags);
+
+	/*
+	 * Free completed jobs and jobs not yet processed by the DRM scheduler
+	 * free worker.
+	 */
+	spin_lock_irqsave(&sched->lock, flags);
+	list_for_each_entry_safe(job, next, &sched->done_list, link)
+		list_move_tail(&job->link, &list);
+	spin_unlock_irqrestore(&sched->lock, flags);
+
+	list_for_each_entry_safe(job, next, &list, link)
+		mock_sched_free_job(&job->base);
+
+	drm_sched_fini(&sched->base);
+}
+
+/**
+ * drm_mock_sched_advance - Advances the mock scheduler timeline
+ *
+ * @sched: Scheduler timeline to advance
+ * @num: By how many jobs to advance
+ *
+ * Advancing the scheduler timeline by a number of seqnos will trigger
+ * signalling of the hardware fences and unlinking the jobs from the internal
+ * scheduler tracking.
+ *
+ * This can be used from test cases which want complete control of the simulated
+ * job execution timing. For example submitting one job with no set duration
+ * would never complete it before test cases advances the timeline by one.
+ */
+unsigned int drm_mock_sched_advance(struct drm_mock_scheduler *sched,
+				    unsigned int num)
+{
+	struct drm_mock_sched_job *job, *next;
+	unsigned int found = 0;
+	unsigned long flags;
+	LIST_HEAD(signal);
+
+	spin_lock_irqsave(&sched->lock, flags);
+	if (WARN_ON_ONCE(sched->hw_timeline.cur_seqno + num <
+			 sched->hw_timeline.cur_seqno))
+		goto unlock;
+	sched->hw_timeline.cur_seqno += num;
+	list_for_each_entry_safe(job, next, &sched->job_list, link) {
+		if (sched->hw_timeline.cur_seqno < job->hw_fence.seqno)
+			break;
+
+		drm_mock_sched_job_complete(job);
+		found++;
+	}
+unlock:
+	spin_unlock_irqrestore(&sched->lock, flags);
+
+	return found;
+}
+
+MODULE_DESCRIPTION("DRM mock scheduler and tests");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/scheduler/tests/sched_tests.h b/drivers/gpu/drm/scheduler/tests/sched_tests.h
new file mode 100644
index 000000000000..27caf8285fb7
--- /dev/null
+++ b/drivers/gpu/drm/scheduler/tests/sched_tests.h
@@ -0,0 +1,226 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Valve Corporation */
+
+#ifndef _SCHED_TESTS_H_
+#define _SCHED_TESTS_H_
+
+#include <kunit/test.h>
+#include <linux/atomic.h>
+#include <linux/completion.h>
+#include <linux/dma-fence.h>
+#include <linux/hrtimer.h>
+#include <linux/ktime.h>
+#include <linux/list.h>
+#include <linux/atomic.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+
+#include <drm/gpu_scheduler.h>
+
+/*
+ * DOC: Mock DRM scheduler data structures
+ *
+ * drm_mock_* data structures are used to implement a mock "GPU".
+ *
+ * They subclass the core DRM scheduler objects and add their data on top, which
+ * enables tracking the submitted jobs and simulating their execution with the
+ * attributes as specified by the test case.
+ */
+
+/**
+ * struct drm_mock_scheduler - implements a trivial mock GPU execution engine
+ *
+ * @base: DRM scheduler base class
+ * @test: Backpointer to owning the kunit test case
+ * @lock: Lock to protect the simulated @hw_timeline, @job_list and @done_list
+ * @job_list: List of jobs submitted to the mock GPU
+ * @done_list: List of jobs completed by the mock GPU
+ * @hw_timeline: Simulated hardware timeline has a @context, @next_seqno and
+ *		 @cur_seqno for implementing a struct dma_fence signaling the
+ *		 simulated job completion.
+ *
+ * Trivial mock GPU execution engine tracks submitted jobs and enables
+ * completing them strictly in submission order.
+ */
+struct drm_mock_scheduler {
+	struct drm_gpu_scheduler base;
+
+	struct kunit		*test;
+
+	spinlock_t		lock;
+	struct list_head	job_list;
+	struct list_head	done_list;
+
+	struct {
+		u64		context;
+		atomic_t	next_seqno;
+		unsigned int	cur_seqno;
+	} hw_timeline;
+};
+
+/**
+ * struct drm_mock_sched_entity - implements a mock GPU sched entity
+ *
+ * @base: DRM scheduler entity base class
+ * @test: Backpointer to owning the kunit test case
+ *
+ * Mock GPU sched entity is used by the test cases to submit jobs to the mock
+ * scheduler.
+ */
+struct drm_mock_sched_entity {
+	struct drm_sched_entity base;
+
+	struct kunit		*test;
+};
+
+/**
+ * struct drm_mock_sched_job - implements a mock GPU job
+ *
+ * @base: DRM sched job base class
+ * @done: Completion signaling job completion.
+ * @flags: Flags designating job state.
+ * @link: List head element used by job tracking by the drm_mock_scheduler
+ * @timer: Timer used for simulating job execution duration
+ * @duration_us: Simulated job duration in micro seconds, or zero if in manual
+ *		 timeline advance mode
+ * @finish_at: Absolute time when the jobs with set duration will complete
+ * @lock: Lock used for @hw_fence
+ * @hw_fence: Fence returned to DRM scheduler as the hardware fence
+ * @test: Backpointer to owning the kunit test case
+ *
+ * Mock GPU sched job is used by the test cases to submit jobs to the mock
+ * scheduler.
+ */
+struct drm_mock_sched_job {
+	struct drm_sched_job	base;
+
+	struct completion	done;
+
+#define DRM_MOCK_SCHED_JOB_DONE		0x1
+#define DRM_MOCK_SCHED_JOB_TIMEDOUT	0x2
+	unsigned long		flags;
+
+	struct list_head	link;
+	struct hrtimer		timer;
+
+	unsigned int		duration_us;
+	ktime_t			finish_at;
+
+	spinlock_t		lock;
+	struct dma_fence	hw_fence;
+
+	struct kunit		*test;
+};
+
+static inline struct drm_mock_scheduler *
+drm_sched_to_mock_sched(struct drm_gpu_scheduler *sched)
+{
+	return container_of(sched, struct drm_mock_scheduler, base);
+};
+
+static inline struct drm_mock_sched_entity *
+drm_sched_entity_to_mock_entity(struct drm_sched_entity *sched_entity)
+{
+	return container_of(sched_entity, struct drm_mock_sched_entity, base);
+};
+
+static inline struct drm_mock_sched_job *
+drm_sched_job_to_mock_job(struct drm_sched_job *sched_job)
+{
+	return container_of(sched_job, struct drm_mock_sched_job, base);
+};
+
+struct drm_mock_scheduler *drm_mock_sched_new(struct kunit *test,
+					      long timeout);
+void drm_mock_sched_fini(struct drm_mock_scheduler *sched);
+unsigned int drm_mock_sched_advance(struct drm_mock_scheduler *sched,
+				    unsigned int num);
+
+struct drm_mock_sched_entity *
+drm_mock_sched_entity_new(struct kunit *test,
+			  enum drm_sched_priority priority,
+			  struct drm_mock_scheduler *sched);
+void drm_mock_sched_entity_free(struct drm_mock_sched_entity *entity);
+
+struct drm_mock_sched_job *
+drm_mock_sched_job_new(struct kunit *test,
+		       struct drm_mock_sched_entity *entity);
+
+/**
+ * drm_mock_sched_job_submit - Arm and submit a job in one go
+ *
+ * @job: Job to arm and submit
+ */
+static inline void drm_mock_sched_job_submit(struct drm_mock_sched_job *job)
+{
+	drm_sched_job_arm(&job->base);
+	drm_sched_entity_push_job(&job->base);
+}
+
+/**
+ * drm_mock_sched_job_set_duration_us - Set a job duration
+ *
+ * @job: Job to set the duration for
+ * @duration_us: Duration in micro seconds
+ *
+ * Jobs with duration set will be automatically completed by the mock scheduler
+ * as the timeline progresses, unless a job without a set duration is
+ * encountered in the timelime in which case calling drm_mock_sched_advance()
+ * will be required to bump the timeline.
+ */
+static inline void
+drm_mock_sched_job_set_duration_us(struct drm_mock_sched_job *job,
+				   unsigned int duration_us)
+{
+	job->duration_us = duration_us;
+}
+
+/**
+ * drm_mock_sched_job_is_finished - Check if a job is finished
+ *
+ * @job: Job to check
+ *
+ * Returns: true if finished
+ */
+static inline bool
+drm_mock_sched_job_is_finished(struct drm_mock_sched_job *job)
+{
+	return job->flags & DRM_MOCK_SCHED_JOB_DONE;
+}
+
+/**
+ * drm_mock_sched_job_wait_finished - Wait until a job is finished
+ *
+ * @job: Job to wait for
+ * @timeout: Wait time in jiffies
+ *
+ * Returns: true if finished within the timeout provided, otherwise false
+ */
+static inline bool
+drm_mock_sched_job_wait_finished(struct drm_mock_sched_job *job, long timeout)
+{
+	if (job->flags & DRM_MOCK_SCHED_JOB_DONE)
+		return true;
+
+	return wait_for_completion_timeout(&job->done, timeout) != 0;
+}
+
+/**
+ * drm_mock_sched_job_wait_scheduled - Wait until a job is scheduled
+ *
+ * @job: Job to wait for
+ * @timeout: Wait time in jiffies
+ *
+ * Returns: true if scheduled within the timeout provided, otherwise false
+ */
+static inline bool
+drm_mock_sched_job_wait_scheduled(struct drm_mock_sched_job *job, long timeout)
+{
+	KUNIT_ASSERT_EQ(job->test, job->flags & DRM_MOCK_SCHED_JOB_DONE, 0);
+
+	return dma_fence_wait_timeout(&job->base.s_fence->scheduled,
+				      false,
+				      timeout) != 0;
+}
+
+#endif
diff --git a/drivers/gpu/drm/scheduler/tests/tests_basic.c b/drivers/gpu/drm/scheduler/tests/tests_basic.c
new file mode 100644
index 000000000000..7230057e0594
--- /dev/null
+++ b/drivers/gpu/drm/scheduler/tests/tests_basic.c
@@ -0,0 +1,476 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2025 Valve Corporation */
+
+#include <linux/delay.h>
+
+#include "sched_tests.h"
+
+/*
+ * DRM scheduler basic tests should check the basic functional correctness of
+ * the scheduler, including some very light smoke testing. More targeted tests,
+ * for example focusing on testing specific bugs and other more complicated test
+ * scenarios, should be implemented in separate source units.
+ */
+
+static int drm_sched_basic_init(struct kunit *test)
+{
+	test->priv = drm_mock_sched_new(test, MAX_SCHEDULE_TIMEOUT);
+
+	return 0;
+}
+
+static void drm_sched_basic_exit(struct kunit *test)
+{
+	struct drm_mock_scheduler *sched = test->priv;
+
+	drm_mock_sched_fini(sched);
+}
+
+static int drm_sched_timeout_init(struct kunit *test)
+{
+	test->priv = drm_mock_sched_new(test, HZ);
+
+	return 0;
+}
+
+static void drm_sched_basic_submit(struct kunit *test)
+{
+	struct drm_mock_scheduler *sched = test->priv;
+	struct drm_mock_sched_entity *entity;
+	struct drm_mock_sched_job *job;
+	unsigned int i;
+	bool done;
+
+	/*
+	 * Submit one job to the scheduler and verify that it gets scheduled
+	 * and completed only when the mock hw backend processes it.
+	 */
+
+	entity = drm_mock_sched_entity_new(test,
+					   DRM_SCHED_PRIORITY_NORMAL,
+					   sched);
+	job = drm_mock_sched_job_new(test, entity);
+
+	drm_mock_sched_job_submit(job);
+
+	done = drm_mock_sched_job_wait_scheduled(job, HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	done = drm_mock_sched_job_wait_finished(job, HZ / 2);
+	KUNIT_ASSERT_FALSE(test, done);
+
+	i = drm_mock_sched_advance(sched, 1);
+	KUNIT_ASSERT_EQ(test, i, 1);
+
+	done = drm_mock_sched_job_wait_finished(job, HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	drm_mock_sched_entity_free(entity);
+}
+
+struct drm_sched_basic_params {
+	const char *description;
+	unsigned int queue_depth;
+	unsigned int num_entities;
+	unsigned int job_us;
+	bool dep_chain;
+};
+
+static const struct drm_sched_basic_params drm_sched_basic_cases[] = {
+	{
+		.description = "A queue of jobs in a single entity",
+		.queue_depth = 100,
+		.job_us = 1000,
+		.num_entities = 1,
+	},
+	{
+		.description = "A chain of dependent jobs across multiple entities",
+		.queue_depth = 100,
+		.job_us = 1000,
+		.num_entities = 1,
+		.dep_chain = true,
+	},
+	{
+		.description = "Multiple independent job queues",
+		.queue_depth = 100,
+		.job_us = 1000,
+		.num_entities = 4,
+	},
+	{
+		.description = "Multiple inter-dependent job queues",
+		.queue_depth = 100,
+		.job_us = 1000,
+		.num_entities = 4,
+		.dep_chain = true,
+	},
+};
+
+static void
+drm_sched_basic_desc(const struct drm_sched_basic_params *params, char *desc)
+{
+	strscpy(desc, params->description, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(drm_sched_basic, drm_sched_basic_cases, drm_sched_basic_desc);
+
+static void drm_sched_basic_test(struct kunit *test)
+{
+	const struct drm_sched_basic_params *params = test->param_value;
+	struct drm_mock_scheduler *sched = test->priv;
+	struct drm_mock_sched_job *job, *prev = NULL;
+	struct drm_mock_sched_entity **entity;
+	unsigned int i, cur_ent = 0;
+	bool done;
+
+	entity = kunit_kcalloc(test, params->num_entities, sizeof(*entity),
+			       GFP_KERNEL);
+	KUNIT_ASSERT_NOT_NULL(test, entity);
+
+	for (i = 0; i < params->num_entities; i++)
+		entity[i] = drm_mock_sched_entity_new(test,
+						      DRM_SCHED_PRIORITY_NORMAL,
+						      sched);
+
+	for (i = 0; i < params->queue_depth; i++) {
+		job = drm_mock_sched_job_new(test, entity[cur_ent++]);
+		cur_ent %= params->num_entities;
+		drm_mock_sched_job_set_duration_us(job, params->job_us);
+		if (params->dep_chain && prev)
+			drm_sched_job_add_dependency(&job->base,
+						     dma_fence_get(&prev->base.s_fence->finished));
+		drm_mock_sched_job_submit(job);
+		prev = job;
+	}
+
+	done = drm_mock_sched_job_wait_finished(job, HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	for (i = 0; i < params->num_entities; i++)
+		drm_mock_sched_entity_free(entity[i]);
+}
+
+static void drm_sched_basic_entity_cleanup(struct kunit *test)
+{
+	struct drm_mock_sched_job *job, *mid, *prev = NULL;
+	struct drm_mock_scheduler *sched = test->priv;
+	struct drm_mock_sched_entity *entity[4];
+	const unsigned int qd = 100;
+	unsigned int i, cur_ent = 0;
+	bool done;
+
+	/*
+	 * Submit a queue of jobs across different entities with an explicit
+	 * chain of dependencies between them and trigger entity cleanup while
+	 * the queue is still being processed.
+	 */
+
+	for (i = 0; i < ARRAY_SIZE(entity); i++)
+		entity[i] = drm_mock_sched_entity_new(test,
+						      DRM_SCHED_PRIORITY_NORMAL,
+						      sched);
+
+	for (i = 0; i < qd; i++) {
+		job = drm_mock_sched_job_new(test, entity[cur_ent++]);
+		cur_ent %= ARRAY_SIZE(entity);
+		drm_mock_sched_job_set_duration_us(job, 1000);
+		if (prev)
+			drm_sched_job_add_dependency(&job->base,
+						     dma_fence_get(&prev->base.s_fence->finished));
+		drm_mock_sched_job_submit(job);
+		if (i == qd / 2)
+			mid = job;
+		prev = job;
+	}
+
+	done = drm_mock_sched_job_wait_finished(mid, HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	/* Exit with half of the queue still pending to be executed. */
+	for (i = 0; i < ARRAY_SIZE(entity); i++)
+		drm_mock_sched_entity_free(entity[i]);
+}
+
+static struct kunit_case drm_sched_basic_tests[] = {
+	KUNIT_CASE(drm_sched_basic_submit),
+	KUNIT_CASE_PARAM(drm_sched_basic_test, drm_sched_basic_gen_params),
+	KUNIT_CASE(drm_sched_basic_entity_cleanup),
+	{}
+};
+
+static struct kunit_suite drm_sched_basic = {
+	.name = "drm_sched_basic_tests",
+	.init = drm_sched_basic_init,
+	.exit = drm_sched_basic_exit,
+	.test_cases = drm_sched_basic_tests,
+};
+
+static void drm_sched_basic_timeout(struct kunit *test)
+{
+	struct drm_mock_scheduler *sched = test->priv;
+	struct drm_mock_sched_entity *entity;
+	struct drm_mock_sched_job *job;
+	bool done;
+
+	/*
+	 * Submit a single job against a scheduler with the timeout configured
+	 * and verify that the timeout handling will run if the backend fails
+	 * to complete it in time.
+	 */
+
+	entity = drm_mock_sched_entity_new(test,
+					   DRM_SCHED_PRIORITY_NORMAL,
+					   sched);
+	job = drm_mock_sched_job_new(test, entity);
+
+	drm_mock_sched_job_submit(job);
+
+	done = drm_mock_sched_job_wait_scheduled(job, HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	done = drm_mock_sched_job_wait_finished(job, HZ / 2);
+	KUNIT_ASSERT_FALSE(test, done);
+
+	KUNIT_ASSERT_EQ(test,
+			job->flags & DRM_MOCK_SCHED_JOB_TIMEDOUT,
+			0);
+
+	done = drm_mock_sched_job_wait_finished(job, HZ);
+	KUNIT_ASSERT_FALSE(test, done);
+
+	KUNIT_ASSERT_EQ(test,
+			job->flags & DRM_MOCK_SCHED_JOB_TIMEDOUT,
+			DRM_MOCK_SCHED_JOB_TIMEDOUT);
+
+	drm_mock_sched_entity_free(entity);
+}
+
+static struct kunit_case drm_sched_timeout_tests[] = {
+	KUNIT_CASE(drm_sched_basic_timeout),
+	{}
+};
+
+static struct kunit_suite drm_sched_timeout = {
+	.name = "drm_sched_basic_timeout_tests",
+	.init = drm_sched_timeout_init,
+	.exit = drm_sched_basic_exit,
+	.test_cases = drm_sched_timeout_tests,
+};
+
+static void drm_sched_priorities(struct kunit *test)
+{
+	struct drm_mock_sched_entity *entity[DRM_SCHED_PRIORITY_COUNT];
+	struct drm_mock_scheduler *sched = test->priv;
+	struct drm_mock_sched_job *job;
+	const unsigned int qd = 100;
+	unsigned int i, cur_ent = 0;
+	enum drm_sched_priority p;
+	bool done;
+
+	/*
+	 * Submit a bunch of jobs against entities configured with different
+	 * priorities.
+	 */
+
+	BUILD_BUG_ON(DRM_SCHED_PRIORITY_KERNEL > DRM_SCHED_PRIORITY_LOW);
+	BUILD_BUG_ON(ARRAY_SIZE(entity) != DRM_SCHED_PRIORITY_COUNT);
+
+	for (p = DRM_SCHED_PRIORITY_KERNEL; p <= DRM_SCHED_PRIORITY_LOW; p++)
+		entity[p] = drm_mock_sched_entity_new(test, p, sched);
+
+	for (i = 0; i < qd; i++) {
+		job = drm_mock_sched_job_new(test, entity[cur_ent++]);
+		cur_ent %= ARRAY_SIZE(entity);
+		drm_mock_sched_job_set_duration_us(job, 1000);
+		drm_mock_sched_job_submit(job);
+	}
+
+	done = drm_mock_sched_job_wait_finished(job, HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	for (i = 0; i < ARRAY_SIZE(entity); i++)
+		drm_mock_sched_entity_free(entity[i]);
+}
+
+static void drm_sched_change_priority(struct kunit *test)
+{
+	struct drm_mock_sched_entity *entity[DRM_SCHED_PRIORITY_COUNT];
+	struct drm_mock_scheduler *sched = test->priv;
+	struct drm_mock_sched_job *job;
+	const unsigned int qd = 1000;
+	unsigned int i, cur_ent = 0;
+	enum drm_sched_priority p;
+
+	/*
+	 * Submit a bunch of jobs against entities configured with different
+	 * priorities and while waiting for them to complete, periodically keep
+	 * changing their priorities.
+	 *
+	 * We set up the queue-depth (qd) and job duration so the priority
+	 * changing loop has some time to interact with submissions to the
+	 * backend and job completions as they progress.
+	 */
+
+	for (p = DRM_SCHED_PRIORITY_KERNEL; p <= DRM_SCHED_PRIORITY_LOW; p++)
+		entity[p] = drm_mock_sched_entity_new(test, p, sched);
+
+	for (i = 0; i < qd; i++) {
+		job = drm_mock_sched_job_new(test, entity[cur_ent++]);
+		cur_ent %= ARRAY_SIZE(entity);
+		drm_mock_sched_job_set_duration_us(job, 1000);
+		drm_mock_sched_job_submit(job);
+	}
+
+	do {
+		drm_sched_entity_set_priority(&entity[cur_ent]->base,
+					      (entity[cur_ent]->base.priority + 1) %
+					      DRM_SCHED_PRIORITY_COUNT);
+		cur_ent++;
+		cur_ent %= ARRAY_SIZE(entity);
+		usleep_range(200, 500);
+	} while (!drm_mock_sched_job_is_finished(job));
+
+	for (i = 0; i < ARRAY_SIZE(entity); i++)
+		drm_mock_sched_entity_free(entity[i]);
+}
+
+static struct kunit_case drm_sched_priority_tests[] = {
+	KUNIT_CASE(drm_sched_priorities),
+	KUNIT_CASE(drm_sched_change_priority),
+	{}
+};
+
+static struct kunit_suite drm_sched_priority = {
+	.name = "drm_sched_basic_priority_tests",
+	.init = drm_sched_basic_init,
+	.exit = drm_sched_basic_exit,
+	.test_cases = drm_sched_priority_tests,
+};
+
+static void drm_sched_test_modify_sched(struct kunit *test)
+{
+	unsigned int i, cur_ent = 0, cur_sched = 0;
+	struct drm_mock_sched_entity *entity[13];
+	struct drm_mock_scheduler *sched[3];
+	struct drm_mock_sched_job *job;
+	const unsigned int qd = 1000;
+
+	/*
+	 * Submit a bunch of jobs against entities configured with different
+	 * schedulers and while waiting for them to complete, periodically keep
+	 * changing schedulers associated with each entity.
+	 *
+	 * We set up the queue-depth (qd) and job duration so the sched modify
+	 * loop has some time to interact with submissions to the backend and
+	 * job completions as they progress.
+	 *
+	 * For the number of schedulers and entities we use primes in order to
+	 * perturb the entity->sched assignments with less of a regular pattern.
+	 */
+
+	for (i = 0; i < ARRAY_SIZE(sched); i++)
+		sched[i] = drm_mock_sched_new(test, MAX_SCHEDULE_TIMEOUT);
+
+	for (i = 0; i < ARRAY_SIZE(entity); i++)
+		entity[i] = drm_mock_sched_entity_new(test,
+						      DRM_SCHED_PRIORITY_NORMAL,
+						      sched[i % ARRAY_SIZE(sched)]);
+
+	for (i = 0; i < qd; i++) {
+		job = drm_mock_sched_job_new(test, entity[cur_ent++]);
+		cur_ent %= ARRAY_SIZE(entity);
+		drm_mock_sched_job_set_duration_us(job, 1000);
+		drm_mock_sched_job_submit(job);
+	}
+
+	do {
+		struct drm_gpu_scheduler *modify;
+
+		usleep_range(200, 500);
+		cur_ent++;
+		cur_ent %= ARRAY_SIZE(entity);
+		cur_sched++;
+		cur_sched %= ARRAY_SIZE(sched);
+		modify = &sched[cur_sched]->base;
+		drm_sched_entity_modify_sched(&entity[cur_ent]->base, &modify,
+					      1);
+	} while (!drm_mock_sched_job_is_finished(job));
+
+	for (i = 0; i < ARRAY_SIZE(entity); i++)
+		drm_mock_sched_entity_free(entity[i]);
+
+	for (i = 0; i < ARRAY_SIZE(sched); i++)
+		drm_mock_sched_fini(sched[i]);
+}
+
+static struct kunit_case drm_sched_modify_sched_tests[] = {
+	KUNIT_CASE(drm_sched_test_modify_sched),
+	{}
+};
+
+static struct kunit_suite drm_sched_modify_sched = {
+	.name = "drm_sched_basic_modify_sched_tests",
+	.test_cases = drm_sched_modify_sched_tests,
+};
+
+static void drm_sched_test_credits(struct kunit *test)
+{
+	struct drm_mock_sched_entity *entity;
+	struct drm_mock_scheduler *sched;
+	struct drm_mock_sched_job *job[2];
+	bool done;
+	int i;
+
+	/*
+	 * Check that the configured credit limit is respected.
+	 */
+
+	sched = drm_mock_sched_new(test, MAX_SCHEDULE_TIMEOUT);
+	sched->base.credit_limit = 1;
+
+	entity = drm_mock_sched_entity_new(test,
+					   DRM_SCHED_PRIORITY_NORMAL,
+					   sched);
+
+	job[0] = drm_mock_sched_job_new(test, entity);
+	job[1] = drm_mock_sched_job_new(test, entity);
+
+	drm_mock_sched_job_submit(job[0]);
+	drm_mock_sched_job_submit(job[1]);
+
+	done = drm_mock_sched_job_wait_scheduled(job[0], HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	done = drm_mock_sched_job_wait_scheduled(job[1], HZ);
+	KUNIT_ASSERT_FALSE(test, done);
+
+	i = drm_mock_sched_advance(sched, 1);
+	KUNIT_ASSERT_EQ(test, i, 1);
+
+	done = drm_mock_sched_job_wait_scheduled(job[1], HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	i = drm_mock_sched_advance(sched, 1);
+	KUNIT_ASSERT_EQ(test, i, 1);
+
+	done = drm_mock_sched_job_wait_finished(job[1], HZ);
+	KUNIT_ASSERT_TRUE(test, done);
+
+	drm_mock_sched_entity_free(entity);
+	drm_mock_sched_fini(sched);
+}
+
+static struct kunit_case drm_sched_credits_tests[] = {
+	KUNIT_CASE(drm_sched_test_credits),
+	{}
+};
+
+static struct kunit_suite drm_sched_credits = {
+	.name = "drm_sched_basic_credits_tests",
+	.test_cases = drm_sched_credits_tests,
+};
+
+kunit_test_suites(&drm_sched_basic,
+		  &drm_sched_timeout,
+		  &drm_sched_priority,
+		  &drm_sched_modify_sched,
+		  &drm_sched_credits);
diff --git a/drivers/gpu/drm/sitronix/Kconfig b/drivers/gpu/drm/sitronix/Kconfig
new file mode 100644
index 000000000000..c069d0d41775
--- /dev/null
+++ b/drivers/gpu/drm/sitronix/Kconfig
@@ -0,0 +1,51 @@
+config DRM_ST7571_I2C
+	tristate "DRM support for Sitronix ST7571 display panels (I2C)"
+	depends on DRM && I2C && MMU
+	select DRM_CLIENT_SELECTION
+	select DRM_GEM_SHMEM_HELPER
+	select DRM_KMS_HELPER
+	select REGMAP_I2C
+	help
+	  DRM driver for Sitronix ST7571 panels controlled over I2C.
+
+	  if M is selected the module will be called st7571-i2c.
+
+config TINYDRM_ST7586
+	tristate
+	default n
+
+config DRM_ST7586
+	tristate "DRM support for Sitronix ST7586 display panels"
+	depends on DRM && SPI
+	select DRM_CLIENT_SELECTION
+	select DRM_KMS_HELPER
+	select DRM_GEM_DMA_HELPER
+	select DRM_MIPI_DBI
+	default TINYDRM_ST7586
+	help
+	  DRM driver for the following Sitronix ST7586 panels:
+	  * LEGO MINDSTORMS EV3
+
+	  If M is selected the module will be called st7586.
+
+config TINYDRM_ST7735R
+	tristate
+	default n
+
+config DRM_ST7735R
+	tristate "DRM support for Sitronix ST7715R/ST7735R display panels"
+	depends on DRM && SPI
+	select DRM_CLIENT_SELECTION
+	select DRM_KMS_HELPER
+	select DRM_GEM_DMA_HELPER
+	select DRM_MIPI_DBI
+	select BACKLIGHT_CLASS_DEVICE
+	default TINYDRM_ST7735R
+	help
+	  DRM driver for Sitronix ST7715R/ST7735R with one of the following
+	  LCDs:
+	  * Jianda JD-T18003-T01 1.8" 128x160 TFT
+	  * Okaya RH128128T 1.44" 128x128 TFT
+
+	  If M is selected the module will be called st7735r.
+
diff --git a/drivers/gpu/drm/sitronix/Makefile b/drivers/gpu/drm/sitronix/Makefile
new file mode 100644
index 000000000000..bd139e5a6995
--- /dev/null
+++ b/drivers/gpu/drm/sitronix/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_DRM_ST7571_I2C)		+= st7571-i2c.o
+obj-$(CONFIG_DRM_ST7586)		+= st7586.o
+obj-$(CONFIG_DRM_ST7735R)		+= st7735r.o
diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
new file mode 100644
index 000000000000..eec846892962
--- /dev/null
+++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
@@ -0,0 +1,1000 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Sitronix ST7571, a 4 level gray scale dot matrix LCD controller
+ *
+ * Copyright (C) 2025 Marcus Folkesson <marcus.folkesson@gmail.com>
+ */
+
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/regmap.h>
+
+#include <drm/clients/drm_client_setup.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_encoder.h>
+#include <drm/drm_fb_helper.h>
+#include <drm/drm_fbdev_shmem.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_gem_atomic_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_module.h>
+#include <drm/drm_plane.h>
+#include <drm/drm_probe_helper.h>
+
+#include <video/display_timing.h>
+#include <video/of_display_timing.h>
+
+#define ST7571_COMMAND_MODE			(0x00)
+#define ST7571_DATA_MODE			(0x40)
+
+/* Normal mode command set */
+#define ST7571_DISPLAY_OFF			(0xae)
+#define ST7571_DISPLAY_ON			(0xaf)
+#define ST7571_OSC_ON				(0xab)
+#define ST7571_SET_COLUMN_LSB(c)		(0x00 | FIELD_PREP(GENMASK(3, 0), (c)))
+#define ST7571_SET_COLUMN_MSB(c)		(0x10 | FIELD_PREP(GENMASK(2, 0), (c) >> 4))
+#define ST7571_SET_COM0_LSB(x)			(FIELD_PREP(GENMASK(6, 0), (x)))
+#define ST7571_SET_COM0_MSB			(0x44)
+#define ST7571_SET_COM_SCAN_DIR(d)		(0xc0 | FIELD_PREP(GENMASK(3, 3), (d)))
+#define ST7571_SET_CONTRAST_LSB(c)		(FIELD_PREP(GENMASK(5, 0), (c)))
+#define ST7571_SET_CONTRAST_MSB			(0x81)
+#define ST7571_SET_DISPLAY_DUTY_LSB(d)		(FIELD_PREP(GENMASK(7, 0), (d)))
+#define ST7571_SET_DISPLAY_DUTY_MSB		(0x48)
+#define ST7571_SET_ENTIRE_DISPLAY_ON(p)		(0xa4 | FIELD_PREP(GENMASK(0, 0), (p)))
+#define ST7571_SET_LCD_BIAS(b)			(0x50 | FIELD_PREP(GENMASK(2, 0), (b)))
+#define ST7571_SET_MODE_LSB(m)			(FIELD_PREP(GENMASK(7, 2), (m)))
+#define ST7571_SET_MODE_MSB			(0x38)
+#define ST7571_SET_PAGE(p)			(0xb0 | FIELD_PREP(GENMASK(3, 0), (p)))
+#define ST7571_SET_POWER(p)			(0x28 | FIELD_PREP(GENMASK(2, 0), (p)))
+#define ST7571_SET_REGULATOR_REG(r)		(0x20 | FIELD_PREP(GENMASK(2, 0), (r)))
+#define ST7571_SET_REVERSE(r)			(0xa6 | FIELD_PREP(GENMASK(0, 0), (r)))
+#define ST7571_SET_SEG_SCAN_DIR(d)		(0xa0 | FIELD_PREP(GENMASK(0, 0), (d)))
+#define ST7571_SET_START_LINE_LSB(l)		(FIELD_PREP(GENMASK(6, 0), (l)))
+#define ST7571_SET_START_LINE_MSB		(0x40)
+
+/* Extension command set 3 */
+#define ST7571_COMMAND_SET_3			(0x7b)
+#define ST7571_SET_COLOR_MODE(c)		(0x10 | FIELD_PREP(GENMASK(0, 0), (c)))
+#define ST7571_COMMAND_SET_NORMAL		(0x00)
+
+#define ST7571_PAGE_HEIGHT 8
+
+#define DRIVER_NAME "st7571"
+#define DRIVER_DESC "ST7571 DRM driver"
+#define DRIVER_MAJOR 1
+#define DRIVER_MINOR 0
+
+enum st7571_color_mode {
+	ST7571_COLOR_MODE_GRAY = 0,
+	ST7571_COLOR_MODE_BLACKWHITE = 1,
+};
+
+struct st7571_device;
+
+struct st7571_panel_constraints {
+	u32 min_nlines;
+	u32 max_nlines;
+	u32 min_ncols;
+	u32 max_ncols;
+	bool support_grayscale;
+};
+
+struct st7571_panel_data {
+	int (*init)(struct st7571_device *st7571);
+	struct st7571_panel_constraints constraints;
+};
+
+struct st7571_panel_format {
+	void (*prepare_buffer)(struct st7571_device *st7571,
+			       const struct iosys_map *vmap,
+			       struct drm_framebuffer *fb,
+			       struct drm_rect *rect,
+			       struct drm_format_conv_state *fmtcnv_state);
+	int (*update_rect)(struct drm_framebuffer *fb, struct drm_rect *rect);
+	enum st7571_color_mode mode;
+	const u8 nformats;
+	const u32 formats[];
+};
+
+struct st7571_device {
+	struct drm_device dev;
+
+	struct drm_plane primary_plane;
+	struct drm_crtc crtc;
+	struct drm_encoder encoder;
+	struct drm_connector connector;
+
+	struct drm_display_mode mode;
+
+	const struct st7571_panel_format *pformat;
+	const struct st7571_panel_data *pdata;
+	struct i2c_client *client;
+	struct gpio_desc *reset;
+	struct regmap *regmap;
+
+	/*
+	 * Depending on the hardware design, the acknowledge signal may be hard to
+	 * recognize as a valid logic "0" level.
+	 * Therefor, ignore NAK if possible to stay compatible with most hardware designs
+	 * and off-the-shelf panels out there.
+	 *
+	 * From section 6.4 MICROPOCESSOR INTERFACE section in the datasheet:
+	 *
+	 * "By connecting SDA_OUT to SDA_IN externally, the SDA line becomes fully
+	 * I2C interface compatible.
+	 * Separating acknowledge-output from serial data
+	 * input is advantageous for chip-on-glass (COG) applications. In COG
+	 * applications, the ITO resistance and the pull-up resistor will form a
+	 * voltage  divider, which affects acknowledge-signal level. Larger ITO
+	 * resistance will raise the acknowledged-signal level and system cannot
+	 * recognize this level as a valid logic “0” level. By separating SDA_IN from
+	 * SDA_OUT, the IC can be used in a mode that ignores the acknowledge-bit.
+	 * For applications which check acknowledge-bit, it is necessary to minimize
+	 * the ITO resistance of the SDA_OUT trace to guarantee a valid low level."
+	 *
+	 */
+	bool ignore_nak;
+
+	bool grayscale;
+	u32 height_mm;
+	u32 width_mm;
+	u32 startline;
+	u32 nlines;
+	u32 ncols;
+	u32 bpp;
+
+	/* Intermediate buffer in LCD friendly format */
+	u8 *hwbuf;
+
+	/* Row of (transformed) pixels ready to be written to the display */
+	u8 *row;
+};
+
+static inline struct st7571_device *drm_to_st7571(struct drm_device *dev)
+{
+	return container_of(dev, struct st7571_device, dev);
+}
+
+static int st7571_regmap_write(void *context, const void *data, size_t count)
+{
+	struct i2c_client *client = context;
+	struct st7571_device *st7571 = i2c_get_clientdata(client);
+	int ret;
+
+	struct i2c_msg msg = {
+		.addr = st7571->client->addr,
+		.flags = st7571->ignore_nak ? I2C_M_IGNORE_NAK : 0,
+		.len = count,
+		.buf = (u8 *)data
+	};
+
+	ret = i2c_transfer(st7571->client->adapter, &msg, 1);
+
+	/*
+	 * Unfortunately, there is no way to check if the transfer failed because of
+	 * a NAK or something else as I2C bus drivers use different return values for NAK.
+	 *
+	 * However, if the transfer fails and ignore_nak is set, we know it is an error.
+	 */
+	if (ret < 0 && st7571->ignore_nak)
+		return ret;
+
+	return 0;
+}
+
+/* The st7571 driver does not read registers but regmap expects a .read */
+static int st7571_regmap_read(void *context, const void *reg_buf,
+			      size_t reg_size, void *val_buf, size_t val_size)
+{
+	return -EOPNOTSUPP;
+}
+
+static int st7571_send_command_list(struct st7571_device *st7571,
+				    const u8 *cmd_list, size_t len)
+{
+	int ret;
+
+	for (int i = 0; i < len; i++) {
+		ret = regmap_write(st7571->regmap, ST7571_COMMAND_MODE, cmd_list[i]);
+		if (ret < 0)
+			return ret;
+	}
+
+	return ret;
+}
+
+static inline u8 st7571_transform_xy(const char *p, int x, int y)
+{
+	int xrest = x % 8;
+	u8 result = 0;
+
+	/*
+	 * Transforms an (x, y) pixel coordinate into a vertical 8-bit
+	 * column from the framebuffer. It calculates the corresponding byte in the
+	 * framebuffer, extracts the bit at the given x position across 8 consecutive
+	 * rows, and packs those bits into a single byte.
+	 *
+	 * Return an 8-bit value representing a vertical column of pixels.
+	 */
+	x = x / 8;
+	y = (y / 8) * 8;
+
+	for (int i = 0; i < 8; i++) {
+		int row_idx = y + i;
+		u8 byte = p[row_idx * 16 + x];
+		u8 bit = (byte >> xrest) & 1;
+
+		result |= (bit << i);
+	}
+
+	return result;
+}
+
+static int st7571_set_position(struct st7571_device *st7571, int x, int y)
+{
+	u8 cmd_list[] = {
+		ST7571_SET_COLUMN_LSB(x),
+		ST7571_SET_COLUMN_MSB(x),
+		ST7571_SET_PAGE(y / ST7571_PAGE_HEIGHT),
+	};
+
+	return st7571_send_command_list(st7571, cmd_list, ARRAY_SIZE(cmd_list));
+}
+
+static int st7571_fb_clear_screen(struct st7571_device *st7571)
+{
+	u32 npixels = st7571->ncols * round_up(st7571->nlines, ST7571_PAGE_HEIGHT) * st7571->bpp;
+	char pixelvalue = 0x00;
+
+	for (int i = 0; i < npixels; i++)
+		regmap_bulk_write(st7571->regmap, ST7571_DATA_MODE, &pixelvalue, 1);
+
+	return 0;
+}
+
+static void st7571_prepare_buffer_monochrome(struct st7571_device *st7571,
+					     const struct iosys_map *vmap,
+					     struct drm_framebuffer *fb,
+					     struct drm_rect *rect,
+					     struct drm_format_conv_state *fmtcnv_state)
+{
+	unsigned int dst_pitch;
+	struct iosys_map dst;
+	u32 size;
+
+	switch (fb->format->format) {
+	case DRM_FORMAT_XRGB8888:
+		dst_pitch = DIV_ROUND_UP(drm_rect_width(rect), 8);
+		iosys_map_set_vaddr(&dst, st7571->hwbuf);
+
+		drm_fb_xrgb8888_to_mono(&dst, &dst_pitch, vmap, fb, rect, fmtcnv_state);
+		break;
+
+	case DRM_FORMAT_R1:
+		size = (rect->x2 - rect->x1) * (rect->y2 - rect->y1) / 8;
+		memcpy(st7571->hwbuf, vmap->vaddr, size);
+		break;
+	}
+}
+
+static void st7571_prepare_buffer_grayscale(struct st7571_device *st7571,
+					    const struct iosys_map *vmap,
+					    struct drm_framebuffer *fb,
+					    struct drm_rect *rect,
+					    struct drm_format_conv_state *fmtcnv_state)
+{
+	u32 size = (rect->x2 - rect->x1) * (rect->y2 - rect->y1) / 8;
+	unsigned int dst_pitch;
+	struct iosys_map dst;
+
+	switch (fb->format->format) {
+	case DRM_FORMAT_XRGB8888: /* Only support XRGB8888 in monochrome mode */
+		dst_pitch = DIV_ROUND_UP(drm_rect_width(rect), 8);
+		iosys_map_set_vaddr(&dst, st7571->hwbuf);
+
+		drm_fb_xrgb8888_to_mono(&dst, &dst_pitch, vmap, fb, rect, fmtcnv_state);
+		break;
+
+	case DRM_FORMAT_R1:
+		size = (rect->x2 - rect->x1) * (rect->y2 - rect->y1) / 8;
+		memcpy(st7571->hwbuf, vmap->vaddr, size);
+		break;
+
+	case DRM_FORMAT_R2:
+		size = (rect->x2 - rect->x1) * (rect->y2 - rect->y1) / 4;
+		memcpy(st7571->hwbuf, vmap->vaddr, size);
+		break;
+	};
+}
+
+static int st7571_fb_update_rect_monochrome(struct drm_framebuffer *fb, struct drm_rect *rect)
+{
+	struct st7571_device *st7571 = drm_to_st7571(fb->dev);
+	char *row = st7571->row;
+
+	/* Align y to display page boundaries */
+	rect->y1 = round_down(rect->y1, ST7571_PAGE_HEIGHT);
+	rect->y2 = min_t(unsigned int, round_up(rect->y2, ST7571_PAGE_HEIGHT), st7571->nlines);
+
+	for (int y = rect->y1; y < rect->y2; y += ST7571_PAGE_HEIGHT) {
+		for (int x = rect->x1; x < rect->x2; x++)
+			row[x] = st7571_transform_xy(st7571->hwbuf, x, y);
+
+		st7571_set_position(st7571, rect->x1, y);
+
+		/* TODO: Investige why we can't write multiple bytes at once */
+		for (int x = rect->x1; x < rect->x2; x++)
+			regmap_bulk_write(st7571->regmap, ST7571_DATA_MODE, row + x, 1);
+	}
+
+	return 0;
+}
+
+static int st7571_fb_update_rect_grayscale(struct drm_framebuffer *fb, struct drm_rect *rect)
+{
+	struct st7571_device *st7571 = drm_to_st7571(fb->dev);
+	u32 format = fb->format->format;
+	char *row = st7571->row;
+	int x1;
+	int x2;
+
+	/* Align y to display page boundaries */
+	rect->y1 = round_down(rect->y1, ST7571_PAGE_HEIGHT);
+	rect->y2 = min_t(unsigned int, round_up(rect->y2, ST7571_PAGE_HEIGHT), st7571->nlines);
+
+	switch (format) {
+	case DRM_FORMAT_XRGB8888:
+		/* Threated as monochrome (R1) */
+		fallthrough;
+	case DRM_FORMAT_R1:
+		x1 = rect->x1;
+		x2 = rect->x2;
+		break;
+	case DRM_FORMAT_R2:
+		x1 = rect->x1 * 2;
+		x2 = rect->x2 * 2;
+		break;
+	}
+
+	for (int y = rect->y1; y < rect->y2; y += ST7571_PAGE_HEIGHT) {
+		for (int x = x1; x < x2; x++)
+			row[x] = st7571_transform_xy(st7571->hwbuf, x, y);
+
+		st7571_set_position(st7571, rect->x1, y);
+
+		/* TODO: Investige why we can't write multiple bytes at once */
+		for (int x = x1; x < x2; x++) {
+			regmap_bulk_write(st7571->regmap, ST7571_DATA_MODE, row + x, 1);
+
+			/*
+			 * As the display supports grayscale, all pixels must be written as two bits
+			 * even if the format is monochrome.
+			 *
+			 * The bit values maps to the following grayscale:
+			 * 0 0 = White
+			 * 0 1 = Light gray
+			 * 1 0 = Dark gray
+			 * 1 1 = Black
+			 *
+			 * For monochrome formats, write the same value twice to get
+			 * either a black or white pixel.
+			 */
+			if (format == DRM_FORMAT_R1 || format == DRM_FORMAT_XRGB8888)
+				regmap_bulk_write(st7571->regmap, ST7571_DATA_MODE, row + x, 1);
+		}
+	}
+
+	return 0;
+}
+
+static int st7571_connector_get_modes(struct drm_connector *conn)
+{
+	struct st7571_device *st7571 = drm_to_st7571(conn->dev);
+
+	return drm_connector_helper_get_modes_fixed(conn, &st7571->mode);
+}
+
+static const struct drm_connector_helper_funcs st7571_connector_helper_funcs = {
+	.get_modes = st7571_connector_get_modes,
+};
+
+static const struct st7571_panel_format st7571_monochrome = {
+	.prepare_buffer = st7571_prepare_buffer_monochrome,
+	.update_rect = st7571_fb_update_rect_monochrome,
+	.mode = ST7571_COLOR_MODE_BLACKWHITE,
+	.formats = {
+		DRM_FORMAT_XRGB8888,
+		DRM_FORMAT_R1,
+	},
+	.nformats = 2,
+};
+
+static const struct st7571_panel_format st7571_grayscale = {
+	.prepare_buffer = st7571_prepare_buffer_grayscale,
+	.update_rect = st7571_fb_update_rect_grayscale,
+	.mode = ST7571_COLOR_MODE_GRAY,
+	.formats = {
+		DRM_FORMAT_XRGB8888,
+		DRM_FORMAT_R1,
+		DRM_FORMAT_R2,
+	},
+	.nformats = 3,
+};
+
+static const u64 st7571_primary_plane_fmtmods[] = {
+	DRM_FORMAT_MOD_LINEAR,
+	DRM_FORMAT_MOD_INVALID
+};
+
+static int st7571_primary_plane_helper_atomic_check(struct drm_plane *plane,
+						    struct drm_atomic_state *state)
+{
+	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(state, plane);
+	struct drm_crtc *new_crtc = new_plane_state->crtc;
+	struct drm_crtc_state *new_crtc_state = NULL;
+
+	if (new_crtc)
+		new_crtc_state = drm_atomic_get_new_crtc_state(state, new_crtc);
+
+	return drm_atomic_helper_check_plane_state(new_plane_state, new_crtc_state,
+						   DRM_PLANE_NO_SCALING,
+						   DRM_PLANE_NO_SCALING,
+						   false, false);
+}
+
+static void st7571_primary_plane_helper_atomic_update(struct drm_plane *plane,
+						      struct drm_atomic_state *state)
+{
+	struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, plane);
+	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
+	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
+	struct drm_framebuffer *fb = plane_state->fb;
+	struct drm_atomic_helper_damage_iter iter;
+	struct drm_device *dev = plane->dev;
+	struct drm_rect damage;
+	struct st7571_device *st7571 = drm_to_st7571(plane->dev);
+	int ret, idx;
+
+	if (!fb)
+		return; /* no framebuffer; plane is disabled */
+
+	ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE);
+	if (ret)
+		return;
+
+	if (!drm_dev_enter(dev, &idx))
+		goto out_drm_gem_fb_end_cpu_access;
+
+	drm_atomic_helper_damage_iter_init(&iter, old_plane_state, plane_state);
+	drm_atomic_for_each_plane_damage(&iter, &damage) {
+		st7571->pformat->prepare_buffer(st7571,
+						&shadow_plane_state->data[0],
+						fb, &damage,
+						&shadow_plane_state->fmtcnv_state);
+
+		st7571->pformat->update_rect(fb, &damage);
+	}
+
+	drm_dev_exit(idx);
+
+out_drm_gem_fb_end_cpu_access:
+	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);
+}
+
+static void st7571_primary_plane_helper_atomic_disable(struct drm_plane *plane,
+						       struct drm_atomic_state *state)
+{
+	struct drm_device *dev = plane->dev;
+	struct st7571_device *st7571 = drm_to_st7571(plane->dev);
+	int idx;
+
+	if (!drm_dev_enter(dev, &idx))
+		return;
+
+	st7571_fb_clear_screen(st7571);
+	drm_dev_exit(idx);
+}
+
+static const struct drm_plane_helper_funcs st7571_primary_plane_helper_funcs = {
+	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
+	.atomic_check = st7571_primary_plane_helper_atomic_check,
+	.atomic_update = st7571_primary_plane_helper_atomic_update,
+	.atomic_disable = st7571_primary_plane_helper_atomic_disable,
+};
+
+static const struct drm_plane_funcs st7571_primary_plane_funcs = {
+	.update_plane = drm_atomic_helper_update_plane,
+	.disable_plane = drm_atomic_helper_disable_plane,
+	.destroy = drm_plane_cleanup,
+	DRM_GEM_SHADOW_PLANE_FUNCS,
+};
+
+/*
+ * CRTC
+ */
+
+static enum drm_mode_status st7571_crtc_mode_valid(struct drm_crtc *crtc,
+						   const struct drm_display_mode *mode)
+{
+	struct st7571_device *st7571 = drm_to_st7571(crtc->dev);
+
+	return drm_crtc_helper_mode_valid_fixed(crtc, mode, &st7571->mode);
+}
+
+static const struct drm_crtc_helper_funcs st7571_crtc_helper_funcs = {
+	.atomic_check = drm_crtc_helper_atomic_check,
+	.mode_valid = st7571_crtc_mode_valid,
+};
+
+static const struct drm_crtc_funcs st7571_crtc_funcs = {
+	.reset = drm_atomic_helper_crtc_reset,
+	.destroy = drm_crtc_cleanup,
+	.set_config = drm_atomic_helper_set_config,
+	.page_flip = drm_atomic_helper_page_flip,
+	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+};
+
+/*
+ * Encoder
+ */
+
+static void ssd130x_encoder_atomic_enable(struct drm_encoder *encoder,
+					  struct drm_atomic_state *state)
+{
+	struct drm_device *drm = encoder->dev;
+	struct st7571_device *st7571 = drm_to_st7571(drm);
+	u8 command = ST7571_DISPLAY_ON;
+	int ret;
+
+	ret = st7571->pdata->init(st7571);
+	if (ret)
+		return;
+
+	st7571_send_command_list(st7571, &command, 1);
+}
+
+static void ssd130x_encoder_atomic_disable(struct drm_encoder *encoder,
+					   struct drm_atomic_state *state)
+{
+	struct drm_device *drm = encoder->dev;
+	struct st7571_device *st7571 = drm_to_st7571(drm);
+	u8 command = ST7571_DISPLAY_OFF;
+
+	st7571_send_command_list(st7571, &command, 1);
+}
+
+static const struct drm_encoder_funcs st7571_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+
+};
+
+static const struct drm_encoder_helper_funcs st7571_encoder_helper_funcs = {
+	.atomic_enable = ssd130x_encoder_atomic_enable,
+	.atomic_disable = ssd130x_encoder_atomic_disable,
+};
+
+/*
+ * Connector
+ */
+
+static const struct drm_connector_funcs st7571_connector_funcs = {
+	.reset = drm_atomic_helper_connector_reset,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = drm_connector_cleanup,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static const struct drm_mode_config_funcs st7571_mode_config_funcs = {
+	.fb_create = drm_gem_fb_create_with_dirty,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
+static struct drm_display_mode st7571_mode(struct st7571_device *st7571)
+{
+	struct drm_display_mode mode = {
+		DRM_SIMPLE_MODE(st7571->ncols, st7571->nlines,
+				st7571->width_mm, st7571->height_mm),
+	};
+
+	return mode;
+}
+
+static int st7571_mode_config_init(struct st7571_device *st7571)
+{
+	struct drm_device *dev = &st7571->dev;
+	const struct st7571_panel_constraints *constraints = &st7571->pdata->constraints;
+	int ret;
+
+	ret = drmm_mode_config_init(dev);
+	if (ret)
+		return ret;
+
+	dev->mode_config.min_width = constraints->min_ncols;
+	dev->mode_config.min_height = constraints->min_nlines;
+	dev->mode_config.max_width = constraints->max_ncols;
+	dev->mode_config.max_height = constraints->max_nlines;
+	dev->mode_config.preferred_depth = 24;
+	dev->mode_config.funcs = &st7571_mode_config_funcs;
+
+	return 0;
+}
+
+static int st7571_plane_init(struct st7571_device *st7571,
+			     const struct st7571_panel_format *pformat)
+{
+	struct drm_plane *primary_plane = &st7571->primary_plane;
+	struct drm_device *dev = &st7571->dev;
+	int ret;
+
+	ret = drm_universal_plane_init(dev, primary_plane, 0,
+				       &st7571_primary_plane_funcs,
+				       pformat->formats,
+				       pformat->nformats,
+				       st7571_primary_plane_fmtmods,
+				       DRM_PLANE_TYPE_PRIMARY, NULL);
+	if (ret)
+		return ret;
+
+	drm_plane_helper_add(primary_plane, &st7571_primary_plane_helper_funcs);
+	drm_plane_enable_fb_damage_clips(primary_plane);
+
+	return 0;
+}
+
+static int st7571_crtc_init(struct st7571_device *st7571)
+{
+	struct drm_plane *primary_plane = &st7571->primary_plane;
+	struct drm_crtc *crtc = &st7571->crtc;
+	struct drm_device *dev = &st7571->dev;
+	int ret;
+
+	ret = drm_crtc_init_with_planes(dev, crtc, primary_plane, NULL,
+					&st7571_crtc_funcs, NULL);
+	if (ret)
+		return ret;
+
+	drm_crtc_helper_add(crtc, &st7571_crtc_helper_funcs);
+
+	return 0;
+}
+
+static int st7571_encoder_init(struct st7571_device *st7571)
+{
+	struct drm_encoder *encoder = &st7571->encoder;
+	struct drm_crtc *crtc = &st7571->crtc;
+	struct drm_device *dev = &st7571->dev;
+	int ret;
+
+	ret = drm_encoder_init(dev, encoder, &st7571_encoder_funcs, DRM_MODE_ENCODER_NONE, NULL);
+	if (ret)
+		return ret;
+
+	drm_encoder_helper_add(encoder, &st7571_encoder_helper_funcs);
+
+	encoder->possible_crtcs = drm_crtc_mask(crtc);
+
+	return 0;
+}
+
+static int st7571_connector_init(struct st7571_device *st7571)
+{
+	struct drm_connector *connector = &st7571->connector;
+	struct drm_encoder *encoder = &st7571->encoder;
+	struct drm_device *dev = &st7571->dev;
+	int ret;
+
+	ret = drm_connector_init(dev, connector, &st7571_connector_funcs,
+				 DRM_MODE_CONNECTOR_Unknown);
+	if (ret)
+		return ret;
+
+	drm_connector_helper_add(connector, &st7571_connector_helper_funcs);
+
+	return drm_connector_attach_encoder(connector, encoder);
+}
+
+DEFINE_DRM_GEM_FOPS(st7571_fops);
+
+static const struct drm_driver st7571_driver = {
+	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_ATOMIC,
+
+	.name		 = DRIVER_NAME,
+	.desc		 = DRIVER_DESC,
+	.major		 = DRIVER_MAJOR,
+	.minor		 = DRIVER_MINOR,
+
+	.fops		 = &st7571_fops,
+	DRM_GEM_SHMEM_DRIVER_OPS,
+	DRM_FBDEV_SHMEM_DRIVER_OPS,
+};
+
+static const struct regmap_bus st7571_regmap_bus = {
+	.read = st7571_regmap_read,
+	.write = st7571_regmap_write,
+};
+
+static const struct regmap_config st7571_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.use_single_write = true,
+};
+
+static int st7571_validate_parameters(struct st7571_device *st7571)
+{
+	struct device *dev = st7571->dev.dev;
+	const struct st7571_panel_constraints *constraints = &st7571->pdata->constraints;
+
+	if (st7571->width_mm  == 0) {
+		dev_err(dev, "Invalid panel width\n");
+		return -EINVAL;
+	}
+
+	if (st7571->height_mm == 0) {
+		dev_err(dev, "Invalid panel height\n");
+		return -EINVAL;
+	}
+
+	if (st7571->nlines < constraints->min_nlines ||
+	    st7571->nlines > constraints->max_nlines) {
+		dev_err(dev, "Invalid timing configuration.\n");
+		return -EINVAL;
+	}
+
+	if (st7571->startline + st7571->nlines > constraints->max_nlines) {
+		dev_err(dev, "Invalid timing configuration.\n");
+		return -EINVAL;
+	}
+
+	if (st7571->ncols < constraints->min_ncols ||
+	    st7571->ncols > constraints->max_ncols) {
+		dev_err(dev, "Invalid timing configuration.\n");
+		return -EINVAL;
+	}
+
+	if (st7571->grayscale && !constraints->support_grayscale) {
+		dev_err(dev, "Grayscale not supported\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int st7571_parse_dt(struct st7571_device *st7571)
+{
+	struct device *dev = &st7571->client->dev;
+	struct device_node *np = dev->of_node;
+	struct display_timing dt;
+	int ret;
+
+	ret = of_get_display_timing(np, "panel-timing", &dt);
+	if (ret) {
+		dev_err(dev, "Failed to get display timing from DT\n");
+		return ret;
+	}
+
+	of_property_read_u32(np, "width-mm", &st7571->width_mm);
+	of_property_read_u32(np, "height-mm", &st7571->height_mm);
+	st7571->grayscale = of_property_read_bool(np, "sitronix,grayscale");
+
+	if (st7571->grayscale) {
+		st7571->pformat = &st7571_grayscale;
+		st7571->bpp = 2;
+	} else {
+		st7571->pformat = &st7571_monochrome;
+		st7571->bpp = 1;
+	}
+
+	st7571->startline = dt.vfront_porch.typ;
+	st7571->nlines = dt.vactive.typ;
+	st7571->ncols = dt.hactive.typ;
+
+	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(st7571->reset))
+		return PTR_ERR(st7571->reset);
+
+	return 0;
+}
+
+static void st7571_reset(struct st7571_device *st7571)
+{
+	gpiod_set_value_cansleep(st7571->reset, 1);
+	fsleep(20);
+	gpiod_set_value_cansleep(st7571->reset, 0);
+}
+
+static int st7571_lcd_init(struct st7571_device *st7571)
+{
+	/*
+	 * Most of the initialization sequence is taken directly from the
+	 * referential initial code in the ST7571 datasheet.
+	 */
+	u8 commands[] = {
+		ST7571_DISPLAY_OFF,
+
+		ST7571_SET_MODE_MSB,
+		ST7571_SET_MODE_LSB(0x2e),
+
+		ST7571_SET_SEG_SCAN_DIR(0),
+		ST7571_SET_COM_SCAN_DIR(1),
+
+		ST7571_SET_COM0_MSB,
+		ST7571_SET_COM0_LSB(0x00),
+
+		ST7571_SET_START_LINE_MSB,
+		ST7571_SET_START_LINE_LSB(st7571->startline),
+
+		ST7571_OSC_ON,
+		ST7571_SET_REGULATOR_REG(5),
+		ST7571_SET_CONTRAST_MSB,
+		ST7571_SET_CONTRAST_LSB(0x33),
+		ST7571_SET_LCD_BIAS(0x04),
+		ST7571_SET_DISPLAY_DUTY_MSB,
+		ST7571_SET_DISPLAY_DUTY_LSB(st7571->nlines),
+
+		ST7571_SET_POWER(0x4),	/* Power Control, VC: ON, VR: OFF, VF: OFF */
+		ST7571_SET_POWER(0x6),	/* Power Control, VC: ON, VR: ON, VF: OFF */
+		ST7571_SET_POWER(0x7),	/* Power Control, VC: ON, VR: ON, VF: ON */
+
+		ST7571_COMMAND_SET_3,
+		ST7571_SET_COLOR_MODE(st7571->pformat->mode),
+		ST7571_COMMAND_SET_NORMAL,
+
+		ST7571_SET_REVERSE(0),
+		ST7571_SET_ENTIRE_DISPLAY_ON(0),
+	};
+
+	/* Perform a reset before initializing the controller */
+	st7571_reset(st7571);
+
+	return st7571_send_command_list(st7571, commands, ARRAY_SIZE(commands));
+}
+
+static int st7571_probe(struct i2c_client *client)
+{
+	struct st7571_device *st7571;
+	struct drm_device *dev;
+	int ret;
+
+	st7571 = devm_drm_dev_alloc(&client->dev, &st7571_driver,
+				    struct st7571_device, dev);
+	if (IS_ERR(st7571))
+		return PTR_ERR(st7571);
+
+	dev = &st7571->dev;
+	st7571->client = client;
+	i2c_set_clientdata(client, st7571);
+	st7571->pdata = device_get_match_data(&client->dev);
+
+	ret = st7571_parse_dt(st7571);
+	if (ret)
+		return ret;
+
+	ret = st7571_validate_parameters(st7571);
+	if (ret)
+		return ret;
+
+	st7571->mode = st7571_mode(st7571);
+
+	/*
+	 * The hardware design could make it hard to detect a NAK on the I2C bus.
+	 * If the adapter does not support protocol mangling do
+	 * not set the I2C_M_IGNORE_NAK flag at the expense * of possible
+	 * cruft in the logs.
+	 */
+	if (i2c_check_functionality(client->adapter, I2C_FUNC_PROTOCOL_MANGLING))
+		st7571->ignore_nak = true;
+
+	st7571->regmap = devm_regmap_init(&client->dev, &st7571_regmap_bus,
+					  client, &st7571_regmap_config);
+	if (IS_ERR(st7571->regmap)) {
+		return dev_err_probe(&client->dev, PTR_ERR(st7571->regmap),
+				     "Failed to initialize regmap\n");
+	}
+
+	st7571->hwbuf = devm_kzalloc(&client->dev,
+				     (st7571->nlines * st7571->ncols * st7571->bpp) / 8,
+				     GFP_KERNEL);
+	if (!st7571->hwbuf)
+		return -ENOMEM;
+
+	st7571->row = devm_kzalloc(&client->dev,
+				   (st7571->ncols * st7571->bpp),
+				   GFP_KERNEL);
+	if (!st7571->row)
+		return -ENOMEM;
+
+	ret = st7571_mode_config_init(st7571);
+	if (ret)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to initialize mode config\n");
+
+	ret = st7571_plane_init(st7571, st7571->pformat);
+	if (ret)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to initialize primary plane\n");
+
+	ret = st7571_crtc_init(st7571);
+	if (ret < 0)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to initialize CRTC\n");
+
+	ret = st7571_encoder_init(st7571);
+	if (ret < 0)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to initialize encoder\n");
+
+	ret = st7571_connector_init(st7571);
+	if (ret < 0)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to initialize connector\n");
+
+	drm_mode_config_reset(dev);
+
+	ret = drm_dev_register(dev, 0);
+	if (ret)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to register DRM device\n");
+
+	drm_client_setup(dev, NULL);
+	return 0;
+}
+
+static void st7571_remove(struct i2c_client *client)
+{
+	struct st7571_device *st7571 = i2c_get_clientdata(client);
+
+	drm_dev_unplug(&st7571->dev);
+}
+
+struct st7571_panel_data st7571_config = {
+	.init = st7571_lcd_init,
+	.constraints = {
+		.min_nlines = 1,
+		.max_nlines = 128,
+		.min_ncols = 128,
+		.max_ncols = 128,
+		.support_grayscale = true,
+	},
+};
+
+static const struct of_device_id st7571_of_match[] = {
+	{ .compatible = "sitronix,st7571", .data = &st7571_config },
+	{},
+};
+MODULE_DEVICE_TABLE(of, st7571_of_match);
+
+static const struct i2c_device_id st7571_id[] = {
+	{ "st7571", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, st7571_id);
+
+static struct i2c_driver st7571_i2c_driver = {
+	.driver = {
+		.name = "st7571",
+		.of_match_table = st7571_of_match,
+	},
+	.probe = st7571_probe,
+	.remove = st7571_remove,
+	.id_table = st7571_id,
+};
+
+module_i2c_driver(st7571_i2c_driver);
+
+MODULE_AUTHOR("Marcus Folkesson <marcus.folkesson@gmail.com>");
+MODULE_DESCRIPTION("DRM Driver for Sitronix ST7571 LCD controller");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tiny/st7586.c b/drivers/gpu/drm/sitronix/st7586.c
index a29672d84ede..a29672d84ede 100644
--- a/drivers/gpu/drm/tiny/st7586.c
+++ b/drivers/gpu/drm/sitronix/st7586.c
diff --git a/drivers/gpu/drm/tiny/st7735r.c b/drivers/gpu/drm/sitronix/st7735r.c
index 1d60f6e5b3bc..1d60f6e5b3bc 100644
--- a/drivers/gpu/drm/tiny/st7735r.c
+++ b/drivers/gpu/drm/sitronix/st7735r.c
diff --git a/drivers/gpu/drm/sprd/sprd_dpu.c b/drivers/gpu/drm/sprd/sprd_dpu.c
index cb2816985305..a3447622a33c 100644
--- a/drivers/gpu/drm/sprd/sprd_dpu.c
+++ b/drivers/gpu/drm/sprd/sprd_dpu.c
@@ -784,19 +784,12 @@ static int sprd_dpu_context_init(struct sprd_dpu *dpu,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct dpu_context *ctx = &dpu->ctx;
-	struct resource *res;
 	int ret;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(dev, "failed to get I/O resource\n");
-		return -EINVAL;
-	}
-
-	ctx->base = devm_ioremap(dev, res->start, resource_size(res));
-	if (!ctx->base) {
+	ctx->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(ctx->base)) {
 		dev_err(dev, "failed to map dpu registers\n");
-		return -EFAULT;
+		return PTR_ERR(ctx->base);
 	}
 
 	ctx->irq = platform_get_irq(pdev, 0);
diff --git a/drivers/gpu/drm/sprd/sprd_dsi.c b/drivers/gpu/drm/sprd/sprd_dsi.c
index 8fc26479bb6b..23b0e1dc547a 100644
--- a/drivers/gpu/drm/sprd/sprd_dsi.c
+++ b/drivers/gpu/drm/sprd/sprd_dsi.c
@@ -901,18 +901,11 @@ static int sprd_dsi_context_init(struct sprd_dsi *dsi,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct dsi_context *ctx = &dsi->ctx;
-	struct resource *res;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(dev, "failed to get I/O resource\n");
-		return -EINVAL;
-	}
-
-	ctx->base = devm_ioremap(dev, res->start, resource_size(res));
-	if (!ctx->base) {
+	ctx->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(ctx->base)) {
 		drm_err(dsi->drm, "failed to map dsi host registers\n");
-		return -ENXIO;
+		return PTR_ERR(ctx->base);
 	}
 
 	ctx->regmap = devm_regmap_init(dev, &regmap_tst_io, dsi, &byte_config);
diff --git a/drivers/gpu/drm/sti/Makefile b/drivers/gpu/drm/sti/Makefile
index f203ac5514ae..f778a4eee7c9 100644
--- a/drivers/gpu/drm/sti/Makefile
+++ b/drivers/gpu/drm/sti/Makefile
@@ -7,8 +7,6 @@ sti-drm-y := \
 	sti_compositor.o \
 	sti_crtc.o \
 	sti_plane.o \
-	sti_crtc.o \
-	sti_plane.o \
 	sti_hdmi.o \
 	sti_hdmi_tx3g4c28phy.o \
 	sti_dvo.o \
diff --git a/drivers/gpu/drm/sti/sti_compositor.c b/drivers/gpu/drm/sti/sti_compositor.c
index 063f82d23d80..8c529b0cca8b 100644
--- a/drivers/gpu/drm/sti/sti_compositor.c
+++ b/drivers/gpu/drm/sti/sti_compositor.c
@@ -177,7 +177,6 @@ static int sti_compositor_probe(struct platform_device *pdev)
 	struct device_node *np = dev->of_node;
 	struct device_node *vtg_np;
 	struct sti_compositor *compo;
-	struct resource *res;
 	unsigned int i;
 
 	compo = devm_kzalloc(dev, sizeof(*compo), GFP_KERNEL);
@@ -194,17 +193,10 @@ static int sti_compositor_probe(struct platform_device *pdev)
 
 	memcpy(&compo->data, of_match_node(compositor_of_match, np)->data,
 	       sizeof(struct sti_compositor_data));
-
-	/* Get Memory ressources */
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (res == NULL) {
-		DRM_ERROR("Get memory resource failed\n");
-		return -ENXIO;
-	}
-	compo->regs = devm_ioremap(dev, res->start, resource_size(res));
-	if (compo->regs == NULL) {
+	compo->regs = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(compo->regs)) {
 		DRM_ERROR("Register mapping failed\n");
-		return -ENXIO;
+		return PTR_ERR(compo->regs);
 	}
 
 	/* Get clock resources */
diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
index 4dcddd02629b..74a1eef4674e 100644
--- a/drivers/gpu/drm/sti/sti_dvo.c
+++ b/drivers/gpu/drm/sti/sti_dvo.c
@@ -511,7 +511,6 @@ static int sti_dvo_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct sti_dvo *dvo;
-	struct resource *res;
 	struct device_node *np = dev->of_node;
 
 	DRM_INFO("%s\n", __func__);
@@ -523,16 +522,9 @@ static int sti_dvo_probe(struct platform_device *pdev)
 	}
 
 	dvo->dev = pdev->dev;
-
-	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "dvo-reg");
-	if (!res) {
-		DRM_ERROR("Invalid dvo resource\n");
-		return -ENOMEM;
-	}
-	dvo->regs = devm_ioremap(dev, res->start,
-			resource_size(res));
-	if (!dvo->regs)
-		return -ENOMEM;
+	dvo->regs = devm_platform_ioremap_resource_byname(pdev, "dvo-reg");
+	if (IS_ERR(dvo->regs))
+		return PTR_ERR(dvo->regs);
 
 	dvo->clk_pix = devm_clk_get(dev, "dvo_pix");
 	if (IS_ERR(dvo->clk_pix)) {
diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
index 14fdc00d2ba0..d202b6c1eb8f 100644
--- a/drivers/gpu/drm/sti/sti_hda.c
+++ b/drivers/gpu/drm/sti/sti_hda.c
@@ -693,7 +693,7 @@ static int sti_hda_bind(struct device *dev, struct device *master, void *data)
 
 	connector->hda = hda;
 
-		bridge = devm_kzalloc(dev, sizeof(*bridge), GFP_KERNEL);
+	bridge = devm_kzalloc(dev, sizeof(*bridge), GFP_KERNEL);
 	if (!bridge)
 		return -ENOMEM;
 
@@ -750,16 +750,9 @@ static int sti_hda_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	hda->dev = pdev->dev;
-
-	/* Get resources */
-	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hda-reg");
-	if (!res) {
-		DRM_ERROR("Invalid hda resource\n");
-		return -ENOMEM;
-	}
-	hda->regs = devm_ioremap(dev, res->start, resource_size(res));
-	if (!hda->regs)
-		return -ENOMEM;
+	hda->regs = devm_platform_ioremap_resource_byname(pdev, "hda-reg");
+	if (IS_ERR(hda->regs))
+		return PTR_ERR(hda->regs);
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
 			"video-dacs-ctrl");
diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
index 164a34d793d8..37b8d619066e 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.c
+++ b/drivers/gpu/drm/sti/sti_hdmi.c
@@ -1380,7 +1380,6 @@ static int sti_hdmi_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct sti_hdmi *hdmi;
 	struct device_node *np = dev->of_node;
-	struct resource *res;
 	struct device_node *ddc;
 	int ret;
 
@@ -1399,17 +1398,9 @@ static int sti_hdmi_probe(struct platform_device *pdev)
 	}
 
 	hdmi->dev = pdev->dev;
-
-	/* Get resources */
-	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "hdmi-reg");
-	if (!res) {
-		DRM_ERROR("Invalid hdmi resource\n");
-		ret = -ENOMEM;
-		goto release_adapter;
-	}
-	hdmi->regs = devm_ioremap(dev, res->start, resource_size(res));
-	if (!hdmi->regs) {
-		ret = -ENOMEM;
+	hdmi->regs = devm_platform_ioremap_resource_byname(pdev, "hdmi-reg");
+	if (IS_ERR(hdmi->regs)) {
+		ret = PTR_ERR(hdmi->regs);
 		goto release_adapter;
 	}
 
diff --git a/drivers/gpu/drm/sti/sti_hqvdp.c b/drivers/gpu/drm/sti/sti_hqvdp.c
index 0f658709c9d0..03684062309b 100644
--- a/drivers/gpu/drm/sti/sti_hqvdp.c
+++ b/drivers/gpu/drm/sti/sti_hqvdp.c
@@ -1356,7 +1356,6 @@ static int sti_hqvdp_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct device_node *vtg_np;
 	struct sti_hqvdp *hqvdp;
-	struct resource *res;
 
 	DRM_DEBUG_DRIVER("\n");
 
@@ -1367,17 +1366,10 @@ static int sti_hqvdp_probe(struct platform_device *pdev)
 	}
 
 	hqvdp->dev = dev;
-
-	/* Get Memory resources */
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		DRM_ERROR("Get memory resource failed\n");
-		return -ENXIO;
-	}
-	hqvdp->regs = devm_ioremap(dev, res->start, resource_size(res));
-	if (!hqvdp->regs) {
+	hqvdp->regs = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(hqvdp->regs)) {
 		DRM_ERROR("Register mapping failed\n");
-		return -ENXIO;
+		return PTR_ERR(hqvdp->regs);
 	}
 
 	/* Get clock resources */
diff --git a/drivers/gpu/drm/sti/sti_tvout.c b/drivers/gpu/drm/sti/sti_tvout.c
index af6c06f448c4..6a464b035de8 100644
--- a/drivers/gpu/drm/sti/sti_tvout.c
+++ b/drivers/gpu/drm/sti/sti_tvout.c
@@ -838,7 +838,6 @@ static int sti_tvout_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct device_node *node = dev->of_node;
 	struct sti_tvout *tvout;
-	struct resource *res;
 
 	DRM_INFO("%s\n", __func__);
 
@@ -850,16 +849,9 @@ static int sti_tvout_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	tvout->dev = dev;
-
-	/* get memory resources */
-	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "tvout-reg");
-	if (!res) {
-		DRM_ERROR("Invalid glue resource\n");
-		return -ENOMEM;
-	}
-	tvout->regs = devm_ioremap(dev, res->start, resource_size(res));
-	if (!tvout->regs)
-		return -ENOMEM;
+	tvout->regs = devm_platform_ioremap_resource_byname(pdev, "tvout-reg");
+	if (IS_ERR(tvout->regs))
+		return PTR_ERR(tvout->regs);
 
 	/* get reset resources */
 	tvout->reset = devm_reset_control_get(dev, "tvout");
diff --git a/drivers/gpu/drm/sti/sti_vtg.c b/drivers/gpu/drm/sti/sti_vtg.c
index 5ba469b711b5..ee81691b3203 100644
--- a/drivers/gpu/drm/sti/sti_vtg.c
+++ b/drivers/gpu/drm/sti/sti_vtg.c
@@ -380,23 +380,15 @@ static int vtg_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct sti_vtg *vtg;
-	struct resource *res;
 	int ret;
 
 	vtg = devm_kzalloc(dev, sizeof(*vtg), GFP_KERNEL);
 	if (!vtg)
 		return -ENOMEM;
-
-	/* Get Memory ressources */
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		DRM_ERROR("Get memory resource failed\n");
-		return -ENOMEM;
-	}
-	vtg->regs = devm_ioremap(dev, res->start, resource_size(res));
-	if (!vtg->regs) {
+	vtg->regs = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(vtg->regs)) {
 		DRM_ERROR("failed to remap I/O memory\n");
-		return -ENOMEM;
+		return PTR_ERR(vtg->regs);
 	}
 
 	vtg->irq = platform_get_irq(pdev, 0);
diff --git a/drivers/gpu/drm/stm/lvds.c b/drivers/gpu/drm/stm/lvds.c
index 4613e8e3b8fd..a3ae9a93ce66 100644
--- a/drivers/gpu/drm/stm/lvds.c
+++ b/drivers/gpu/drm/stm/lvds.c
@@ -934,28 +934,27 @@ static const struct drm_connector_funcs lvds_conn_funcs = {
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
 
-static int lvds_attach(struct drm_bridge *bridge,
+static int lvds_attach(struct drm_bridge *bridge, struct drm_encoder *encoder,
 		       enum drm_bridge_attach_flags flags)
 {
 	struct stm_lvds *lvds = bridge_to_stm_lvds(bridge);
 	struct drm_connector *connector = &lvds->connector;
-	struct drm_encoder *encoder = bridge->encoder;
 	int ret;
 
-	if (!bridge->encoder) {
+	if (!encoder) {
 		drm_err(bridge->dev, "Parent encoder object not found\n");
 		return -ENODEV;
 	}
 
 	/* Set the encoder type as caller does not know it */
-	bridge->encoder->encoder_type = DRM_MODE_ENCODER_LVDS;
+	encoder->encoder_type = DRM_MODE_ENCODER_LVDS;
 
 	/* No cloning support */
-	bridge->encoder->possible_clones = 0;
+	encoder->possible_clones = 0;
 
 	/* If we have a next bridge just attach it. */
 	if (lvds->next_bridge)
-		return drm_bridge_attach(bridge->encoder, lvds->next_bridge,
+		return drm_bridge_attach(encoder, lvds->next_bridge,
 					 bridge, flags);
 
 	if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) {
diff --git a/drivers/gpu/drm/sysfb/Kconfig b/drivers/gpu/drm/sysfb/Kconfig
new file mode 100644
index 000000000000..9c9884c7efc6
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/Kconfig
@@ -0,0 +1,76 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+menu "Drivers for system framebuffers"
+	depends on DRM
+
+config DRM_SYSFB_HELPER
+	tristate
+	depends on DRM
+
+config DRM_EFIDRM
+	tristate "EFI framebuffer driver"
+	depends on DRM && MMU && EFI && (!SYSFB_SIMPLEFB || COMPILE_TEST)
+	select APERTURE_HELPERS
+	select DRM_CLIENT_SELECTION
+	select DRM_GEM_SHMEM_HELPER
+	select DRM_KMS_HELPER
+	select DRM_SYSFB_HELPER
+	select SYSFB
+	help
+	  DRM driver for EFI framebuffers.
+
+	  This driver assumes that the display hardware has been initialized
+	  by the firmware or bootloader before the kernel boots. Scanout
+	  buffer, size, and display format must be provided via EFI interfaces.
+
+config DRM_OFDRM
+	tristate "Open Firmware display driver"
+	depends on DRM && MMU && OF && (PPC || COMPILE_TEST)
+	select APERTURE_HELPERS
+	select DRM_CLIENT_SELECTION
+	select DRM_GEM_SHMEM_HELPER
+	select DRM_KMS_HELPER
+	select DRM_SYSFB_HELPER
+	help
+	  DRM driver for Open Firmware framebuffers.
+
+	  This driver assumes that the display hardware has been initialized
+	  by the Open Firmware before the kernel boots. Scanout buffer, size,
+	  and display format must be provided via device tree.
+
+config DRM_SIMPLEDRM
+	tristate "Simple framebuffer driver"
+	depends on DRM && MMU
+	select APERTURE_HELPERS
+	select DRM_CLIENT_SELECTION
+	select DRM_GEM_SHMEM_HELPER
+	select DRM_KMS_HELPER
+	select DRM_SYSFB_HELPER
+	help
+	  DRM driver for simple platform-provided framebuffers.
+
+	  This driver assumes that the display hardware has been initialized
+	  by the firmware or bootloader before the kernel boots. Scanout
+	  buffer, size, and display format must be provided via device tree,
+	  UEFI, VESA, etc.
+
+	  On x86 BIOS or UEFI systems, you should also select SYSFB_SIMPLEFB
+	  to use UEFI and VESA framebuffers.
+
+config DRM_VESADRM
+	tristate "VESA framebuffer driver"
+	depends on DRM && MMU && X86 && (!SYSFB_SIMPLEFB || COMPILE_TEST)
+	select APERTURE_HELPERS
+	select DRM_CLIENT_SELECTION
+	select DRM_GEM_SHMEM_HELPER
+	select DRM_KMS_HELPER
+	select DRM_SYSFB_HELPER
+	select SYSFB
+	help
+	  DRM driver for VESA framebuffers.
+
+	  This driver assumes that the display hardware has been initialized
+	  by the firmware or bootloader before the kernel boots. Scanout
+	  buffer, size, and display format must be provided via VBE interfaces.
+
+endmenu
diff --git a/drivers/gpu/drm/sysfb/Makefile b/drivers/gpu/drm/sysfb/Makefile
new file mode 100644
index 000000000000..a156c496413d
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/Makefile
@@ -0,0 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+drm_sysfb_helper-y := \
+	drm_sysfb.o \
+	drm_sysfb_modeset.o
+drm_sysfb_helper-$(CONFIG_SCREEN_INFO) += drm_sysfb_screen_info.o
+obj-$(CONFIG_DRM_SYSFB_HELPER)	+= drm_sysfb_helper.o
+
+obj-$(CONFIG_DRM_EFIDRM)	+= efidrm.o
+obj-$(CONFIG_DRM_OFDRM)		+= ofdrm.o
+obj-$(CONFIG_DRM_SIMPLEDRM)	+= simpledrm.o
+obj-$(CONFIG_DRM_VESADRM)	+= vesadrm.o
diff --git a/drivers/gpu/drm/sysfb/drm_sysfb.c b/drivers/gpu/drm/sysfb/drm_sysfb.c
new file mode 100644
index 000000000000..308f82153b15
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/drm_sysfb.c
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/export.h>
+#include <linux/limits.h>
+#include <linux/minmax.h>
+#include <linux/module.h>
+
+#include <drm/drm_print.h>
+
+#include "drm_sysfb_helper.h"
+
+int drm_sysfb_get_validated_int(struct drm_device *dev, const char *name,
+				u64 value, u32 max)
+{
+	if (value > min(max, INT_MAX)) {
+		drm_warn(dev, "%s of %llu exceeds maximum of %u\n", name, value, max);
+		return -EINVAL;
+	}
+	return value;
+}
+EXPORT_SYMBOL(drm_sysfb_get_validated_int);
+
+int drm_sysfb_get_validated_int0(struct drm_device *dev, const char *name,
+				 u64 value, u32 max)
+{
+	if (!value) {
+		drm_warn(dev, "%s of 0 not allowed\n", name);
+		return -EINVAL;
+	}
+	return drm_sysfb_get_validated_int(dev, name, value, max);
+}
+EXPORT_SYMBOL(drm_sysfb_get_validated_int0);
+
+MODULE_DESCRIPTION("Helpers for DRM sysfb drivers");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_helper.h b/drivers/gpu/drm/sysfb/drm_sysfb_helper.h
new file mode 100644
index 000000000000..cb08a88242cc
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_helper.h
@@ -0,0 +1,184 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef DRM_SYSFB_HELPER_H
+#define DRM_SYSFB_HELPER_H
+
+#include <linux/container_of.h>
+#include <linux/iosys-map.h>
+
+#include <video/pixel_format.h>
+
+#include <drm/drm_crtc.h>
+#include <drm/drm_device.h>
+#include <drm/drm_modes.h>
+
+struct drm_format_info;
+struct drm_scanout_buffer;
+struct screen_info;
+
+/*
+ * Input parsing
+ */
+
+struct drm_sysfb_format {
+	struct pixel_format pixel;
+	u32 fourcc;
+};
+
+int drm_sysfb_get_validated_int(struct drm_device *dev, const char *name,
+				u64 value, u32 max);
+int drm_sysfb_get_validated_int0(struct drm_device *dev, const char *name,
+				 u64 value, u32 max);
+
+#if defined(CONFIG_SCREEN_INFO)
+int drm_sysfb_get_width_si(struct drm_device *dev, const struct screen_info *si);
+int drm_sysfb_get_height_si(struct drm_device *dev, const struct screen_info *si);
+struct resource *drm_sysfb_get_memory_si(struct drm_device *dev,
+					 const struct screen_info *si,
+					 struct resource *res);
+int drm_sysfb_get_stride_si(struct drm_device *dev, const struct screen_info *si,
+			    const struct drm_format_info *format,
+			    unsigned int width, unsigned int height, u64 size);
+u64 drm_sysfb_get_visible_size_si(struct drm_device *dev, const struct screen_info *si,
+				  unsigned int height, unsigned int stride, u64 size);
+const struct drm_format_info *drm_sysfb_get_format_si(struct drm_device *dev,
+						      const struct drm_sysfb_format *formats,
+						      size_t nformats,
+						      const struct screen_info *si);
+#endif
+
+/*
+ * Input parsing
+ */
+
+int drm_sysfb_get_validated_int(struct drm_device *dev, const char *name,
+				u64 value, u32 max);
+int drm_sysfb_get_validated_int0(struct drm_device *dev, const char *name,
+				 u64 value, u32 max);
+
+/*
+ * Display modes
+ */
+
+struct drm_display_mode drm_sysfb_mode(unsigned int width,
+				       unsigned int height,
+				       unsigned int width_mm,
+				       unsigned int height_mm);
+
+/*
+ * Device
+ */
+
+struct drm_sysfb_device {
+	struct drm_device dev;
+
+	const u8 *edid; /* can be NULL */
+
+	/* hardware settings */
+	struct drm_display_mode fb_mode;
+	const struct drm_format_info *fb_format;
+	unsigned int fb_pitch;
+	unsigned int fb_gamma_lut_size;
+
+	/* hardware-framebuffer kernel address */
+	struct iosys_map fb_addr;
+};
+
+static inline struct drm_sysfb_device *to_drm_sysfb_device(struct drm_device *dev)
+{
+	return container_of(dev, struct drm_sysfb_device, dev);
+}
+
+/*
+ * Plane
+ */
+
+int drm_sysfb_plane_helper_atomic_check(struct drm_plane *plane,
+					struct drm_atomic_state *new_state);
+void drm_sysfb_plane_helper_atomic_update(struct drm_plane *plane,
+					  struct drm_atomic_state *state);
+void drm_sysfb_plane_helper_atomic_disable(struct drm_plane *plane,
+					   struct drm_atomic_state *state);
+int drm_sysfb_plane_helper_get_scanout_buffer(struct drm_plane *plane,
+					      struct drm_scanout_buffer *sb);
+
+#define DRM_SYSFB_PLANE_NFORMATS(_num_native) \
+	((_num_native) + 1)
+
+#define DRM_SYSFB_PLANE_FORMAT_MODIFIERS \
+	DRM_FORMAT_MOD_LINEAR, \
+	DRM_FORMAT_MOD_INVALID
+
+#define DRM_SYSFB_PLANE_HELPER_FUNCS \
+	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS, \
+	.atomic_check = drm_sysfb_plane_helper_atomic_check, \
+	.atomic_update = drm_sysfb_plane_helper_atomic_update, \
+	.atomic_disable = drm_sysfb_plane_helper_atomic_disable, \
+	.get_scanout_buffer = drm_sysfb_plane_helper_get_scanout_buffer
+
+#define DRM_SYSFB_PLANE_FUNCS \
+	.update_plane = drm_atomic_helper_update_plane, \
+	.disable_plane = drm_atomic_helper_disable_plane, \
+	DRM_GEM_SHADOW_PLANE_FUNCS
+
+/*
+ * CRTC
+ */
+
+struct drm_sysfb_crtc_state {
+	struct drm_crtc_state base;
+
+	/* Primary-plane format; required for color mgmt. */
+	const struct drm_format_info *format;
+};
+
+static inline struct drm_sysfb_crtc_state *
+to_drm_sysfb_crtc_state(struct drm_crtc_state *base)
+{
+	return container_of(base, struct drm_sysfb_crtc_state, base);
+}
+
+enum drm_mode_status drm_sysfb_crtc_helper_mode_valid(struct drm_crtc *crtc,
+						      const struct drm_display_mode *mode);
+int drm_sysfb_crtc_helper_atomic_check(struct drm_crtc *crtc, struct drm_atomic_state *new_state);
+
+#define DRM_SYSFB_CRTC_HELPER_FUNCS \
+	.mode_valid = drm_sysfb_crtc_helper_mode_valid, \
+	.atomic_check = drm_sysfb_crtc_helper_atomic_check
+
+void drm_sysfb_crtc_reset(struct drm_crtc *crtc);
+struct drm_crtc_state *drm_sysfb_crtc_atomic_duplicate_state(struct drm_crtc *crtc);
+void drm_sysfb_crtc_atomic_destroy_state(struct drm_crtc *crtc, struct drm_crtc_state *crtc_state);
+
+#define DRM_SYSFB_CRTC_FUNCS \
+	.reset = drm_sysfb_crtc_reset, \
+	.set_config = drm_atomic_helper_set_config, \
+	.page_flip = drm_atomic_helper_page_flip, \
+	.atomic_duplicate_state = drm_sysfb_crtc_atomic_duplicate_state, \
+	.atomic_destroy_state = drm_sysfb_crtc_atomic_destroy_state
+
+/*
+ * Connector
+ */
+
+int drm_sysfb_connector_helper_get_modes(struct drm_connector *connector);
+
+#define DRM_SYSFB_CONNECTOR_HELPER_FUNCS \
+	.get_modes = drm_sysfb_connector_helper_get_modes
+
+#define DRM_SYSFB_CONNECTOR_FUNCS \
+	.reset = drm_atomic_helper_connector_reset, \
+	.fill_modes = drm_helper_probe_single_connector_modes, \
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, \
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state
+
+/*
+ * Mode config
+ */
+
+#define DRM_SYSFB_MODE_CONFIG_FUNCS \
+	.fb_create = drm_gem_fb_create_with_dirty, \
+	.atomic_check = drm_atomic_helper_check, \
+	.atomic_commit = drm_atomic_helper_commit
+
+#endif
diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
new file mode 100644
index 000000000000..ffaa2522ab96
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
@@ -0,0 +1,320 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/export.h>
+#include <linux/slab.h>
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_fourcc.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_gem_atomic_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_panic.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
+
+#include "drm_sysfb_helper.h"
+
+struct drm_display_mode drm_sysfb_mode(unsigned int width,
+				       unsigned int height,
+				       unsigned int width_mm,
+				       unsigned int height_mm)
+{
+	/*
+	 * Assume a monitor resolution of 96 dpi to
+	 * get a somewhat reasonable screen size.
+	 */
+	if (!width_mm)
+		width_mm = DRM_MODE_RES_MM(width, 96ul);
+	if (!height_mm)
+		height_mm = DRM_MODE_RES_MM(height, 96ul);
+
+	{
+		const struct drm_display_mode mode = {
+			DRM_MODE_INIT(60, width, height, width_mm, height_mm)
+		};
+
+		return mode;
+	}
+}
+EXPORT_SYMBOL(drm_sysfb_mode);
+
+/*
+ * Plane
+ */
+
+int drm_sysfb_plane_helper_atomic_check(struct drm_plane *plane,
+					struct drm_atomic_state *new_state)
+{
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(plane->dev);
+	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(new_state, plane);
+	struct drm_shadow_plane_state *new_shadow_plane_state =
+		to_drm_shadow_plane_state(new_plane_state);
+	struct drm_framebuffer *new_fb = new_plane_state->fb;
+	struct drm_crtc *new_crtc = new_plane_state->crtc;
+	struct drm_crtc_state *new_crtc_state = NULL;
+	struct drm_sysfb_crtc_state *new_sysfb_crtc_state;
+	int ret;
+
+	if (new_crtc)
+		new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
+
+	ret = drm_atomic_helper_check_plane_state(new_plane_state, new_crtc_state,
+						  DRM_PLANE_NO_SCALING,
+						  DRM_PLANE_NO_SCALING,
+						  false, false);
+	if (ret)
+		return ret;
+	else if (!new_plane_state->visible)
+		return 0;
+
+	if (new_fb->format != sysfb->fb_format) {
+		void *buf;
+
+		/* format conversion necessary; reserve buffer */
+		buf = drm_format_conv_state_reserve(&new_shadow_plane_state->fmtcnv_state,
+						    sysfb->fb_pitch, GFP_KERNEL);
+		if (!buf)
+			return -ENOMEM;
+	}
+
+	new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
+
+	new_sysfb_crtc_state = to_drm_sysfb_crtc_state(new_crtc_state);
+	new_sysfb_crtc_state->format = new_fb->format;
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_sysfb_plane_helper_atomic_check);
+
+void drm_sysfb_plane_helper_atomic_update(struct drm_plane *plane, struct drm_atomic_state *state)
+{
+	struct drm_device *dev = plane->dev;
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(dev);
+	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
+	struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, plane);
+	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
+	struct drm_framebuffer *fb = plane_state->fb;
+	unsigned int dst_pitch = sysfb->fb_pitch;
+	const struct drm_format_info *dst_format = sysfb->fb_format;
+	struct drm_atomic_helper_damage_iter iter;
+	struct drm_rect damage;
+	int ret, idx;
+
+	ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE);
+	if (ret)
+		return;
+
+	if (!drm_dev_enter(dev, &idx))
+		goto out_drm_gem_fb_end_cpu_access;
+
+	drm_atomic_helper_damage_iter_init(&iter, old_plane_state, plane_state);
+	drm_atomic_for_each_plane_damage(&iter, &damage) {
+		struct iosys_map dst = sysfb->fb_addr;
+		struct drm_rect dst_clip = plane_state->dst;
+
+		if (!drm_rect_intersect(&dst_clip, &damage))
+			continue;
+
+		iosys_map_incr(&dst, drm_fb_clip_offset(dst_pitch, dst_format, &dst_clip));
+		drm_fb_blit(&dst, &dst_pitch, dst_format->format, shadow_plane_state->data, fb,
+			    &damage, &shadow_plane_state->fmtcnv_state);
+	}
+
+	drm_dev_exit(idx);
+out_drm_gem_fb_end_cpu_access:
+	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);
+}
+EXPORT_SYMBOL(drm_sysfb_plane_helper_atomic_update);
+
+void drm_sysfb_plane_helper_atomic_disable(struct drm_plane *plane,
+					   struct drm_atomic_state *state)
+{
+	struct drm_device *dev = plane->dev;
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(dev);
+	struct iosys_map dst = sysfb->fb_addr;
+	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
+	void __iomem *dst_vmap = dst.vaddr_iomem; /* TODO: Use mapping abstraction */
+	unsigned int dst_pitch = sysfb->fb_pitch;
+	const struct drm_format_info *dst_format = sysfb->fb_format;
+	struct drm_rect dst_clip;
+	unsigned long lines, linepixels, i;
+	int idx;
+
+	drm_rect_init(&dst_clip,
+		      plane_state->src_x >> 16, plane_state->src_y >> 16,
+		      plane_state->src_w >> 16, plane_state->src_h >> 16);
+
+	lines = drm_rect_height(&dst_clip);
+	linepixels = drm_rect_width(&dst_clip);
+
+	if (!drm_dev_enter(dev, &idx))
+		return;
+
+	/* Clear buffer to black if disabled */
+	dst_vmap += drm_fb_clip_offset(dst_pitch, dst_format, &dst_clip);
+	for (i = 0; i < lines; ++i) {
+		memset_io(dst_vmap, 0, linepixels * dst_format->cpp[0]);
+		dst_vmap += dst_pitch;
+	}
+
+	drm_dev_exit(idx);
+}
+EXPORT_SYMBOL(drm_sysfb_plane_helper_atomic_disable);
+
+int drm_sysfb_plane_helper_get_scanout_buffer(struct drm_plane *plane,
+					      struct drm_scanout_buffer *sb)
+{
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(plane->dev);
+
+	sb->width = sysfb->fb_mode.hdisplay;
+	sb->height = sysfb->fb_mode.vdisplay;
+	sb->format = sysfb->fb_format;
+	sb->pitch[0] = sysfb->fb_pitch;
+	sb->map[0] = sysfb->fb_addr;
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_sysfb_plane_helper_get_scanout_buffer);
+
+/*
+ * CRTC
+ */
+
+static void drm_sysfb_crtc_state_destroy(struct drm_sysfb_crtc_state *sysfb_crtc_state)
+{
+	__drm_atomic_helper_crtc_destroy_state(&sysfb_crtc_state->base);
+
+	kfree(sysfb_crtc_state);
+}
+
+enum drm_mode_status drm_sysfb_crtc_helper_mode_valid(struct drm_crtc *crtc,
+						      const struct drm_display_mode *mode)
+{
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(crtc->dev);
+
+	return drm_crtc_helper_mode_valid_fixed(crtc, mode, &sysfb->fb_mode);
+}
+EXPORT_SYMBOL(drm_sysfb_crtc_helper_mode_valid);
+
+int drm_sysfb_crtc_helper_atomic_check(struct drm_crtc *crtc, struct drm_atomic_state *new_state)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(dev);
+	struct drm_crtc_state *new_crtc_state = drm_atomic_get_new_crtc_state(new_state, crtc);
+	int ret;
+
+	if (!new_crtc_state->enable)
+		return 0;
+
+	ret = drm_atomic_helper_check_crtc_primary_plane(new_crtc_state);
+	if (ret)
+		return ret;
+
+	if (new_crtc_state->color_mgmt_changed) {
+		const size_t gamma_lut_length =
+			sysfb->fb_gamma_lut_size * sizeof(struct drm_color_lut);
+		const struct drm_property_blob *gamma_lut = new_crtc_state->gamma_lut;
+
+		if (gamma_lut && (gamma_lut->length != gamma_lut_length)) {
+			drm_dbg(dev, "Incorrect gamma_lut length %zu\n", gamma_lut->length);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_sysfb_crtc_helper_atomic_check);
+
+void drm_sysfb_crtc_reset(struct drm_crtc *crtc)
+{
+	struct drm_sysfb_crtc_state *sysfb_crtc_state;
+
+	if (crtc->state)
+		drm_sysfb_crtc_state_destroy(to_drm_sysfb_crtc_state(crtc->state));
+
+	sysfb_crtc_state = kzalloc(sizeof(*sysfb_crtc_state), GFP_KERNEL);
+	if (sysfb_crtc_state)
+		__drm_atomic_helper_crtc_reset(crtc, &sysfb_crtc_state->base);
+	else
+		__drm_atomic_helper_crtc_reset(crtc, NULL);
+}
+EXPORT_SYMBOL(drm_sysfb_crtc_reset);
+
+struct drm_crtc_state *drm_sysfb_crtc_atomic_duplicate_state(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_crtc_state *crtc_state = crtc->state;
+	struct drm_sysfb_crtc_state *new_sysfb_crtc_state;
+	struct drm_sysfb_crtc_state *sysfb_crtc_state;
+
+	if (drm_WARN_ON(dev, !crtc_state))
+		return NULL;
+
+	new_sysfb_crtc_state = kzalloc(sizeof(*new_sysfb_crtc_state), GFP_KERNEL);
+	if (!new_sysfb_crtc_state)
+		return NULL;
+
+	sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
+
+	__drm_atomic_helper_crtc_duplicate_state(crtc, &new_sysfb_crtc_state->base);
+	new_sysfb_crtc_state->format = sysfb_crtc_state->format;
+
+	return &new_sysfb_crtc_state->base;
+}
+EXPORT_SYMBOL(drm_sysfb_crtc_atomic_duplicate_state);
+
+void drm_sysfb_crtc_atomic_destroy_state(struct drm_crtc *crtc, struct drm_crtc_state *crtc_state)
+{
+	drm_sysfb_crtc_state_destroy(to_drm_sysfb_crtc_state(crtc_state));
+}
+EXPORT_SYMBOL(drm_sysfb_crtc_atomic_destroy_state);
+
+/*
+ * Connector
+ */
+
+static int drm_sysfb_get_edid_block(void *data, u8 *buf, unsigned int block, size_t len)
+{
+	struct drm_sysfb_device *sysfb = data;
+	const u8 *edid = sysfb->edid;
+	size_t off = block * EDID_LENGTH;
+	size_t end = off + len;
+
+	if (!edid)
+		return -EINVAL;
+	if (end > EDID_LENGTH)
+		return -EINVAL;
+	memcpy(buf, &edid[off], len);
+
+	/*
+	 * We don't have EDID extensions available and reporting them
+	 * will upset DRM helpers. Thus clear the extension field and
+	 * update the checksum. Adding the extension flag to the checksum
+	 * does this.
+	 */
+	buf[127] += buf[126];
+	buf[126] = 0;
+
+	return 0;
+}
+
+int drm_sysfb_connector_helper_get_modes(struct drm_connector *connector)
+{
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(connector->dev);
+	const struct drm_edid *drm_edid;
+
+	if (sysfb->edid) {
+		drm_edid = drm_edid_read_custom(connector, drm_sysfb_get_edid_block, sysfb);
+		drm_edid_connector_update(connector, drm_edid);
+		drm_edid_free(drm_edid);
+	}
+
+	/* Return the fixed mode even with EDID */
+	return drm_connector_helper_get_modes_fixed(connector, &sysfb->fb_mode);
+}
+EXPORT_SYMBOL(drm_sysfb_connector_helper_get_modes);
diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c b/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c
new file mode 100644
index 000000000000..0b3fb874a51f
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c
@@ -0,0 +1,107 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/export.h>
+#include <linux/limits.h>
+#include <linux/minmax.h>
+#include <linux/screen_info.h>
+
+#include <drm/drm_fourcc.h>
+#include <drm/drm_print.h>
+
+#include "drm_sysfb_helper.h"
+
+static s64 drm_sysfb_get_validated_size0(struct drm_device *dev, const char *name,
+					 u64 value, u64 max)
+{
+	if (!value) {
+		drm_warn(dev, "%s of 0 not allowed\n", name);
+		return -EINVAL;
+	} else if (value > min(max, S64_MAX)) {
+		drm_warn(dev, "%s of %llu exceeds maximum of %llu\n", name, value, max);
+		return -EINVAL;
+	}
+	return value;
+}
+
+int drm_sysfb_get_width_si(struct drm_device *dev, const struct screen_info *si)
+{
+	return drm_sysfb_get_validated_int0(dev, "width", si->lfb_width, U16_MAX);
+}
+EXPORT_SYMBOL(drm_sysfb_get_width_si);
+
+int drm_sysfb_get_height_si(struct drm_device *dev, const struct screen_info *si)
+{
+	return drm_sysfb_get_validated_int0(dev, "height", si->lfb_height, U16_MAX);
+}
+EXPORT_SYMBOL(drm_sysfb_get_height_si);
+
+struct resource *drm_sysfb_get_memory_si(struct drm_device *dev,
+					 const struct screen_info *si,
+					 struct resource *res)
+{
+	ssize_t	num;
+
+	num = screen_info_resources(si, res, 1);
+	if (!num) {
+		drm_warn(dev, "memory resource not found\n");
+		return NULL;
+	}
+
+	return res;
+}
+EXPORT_SYMBOL(drm_sysfb_get_memory_si);
+
+int drm_sysfb_get_stride_si(struct drm_device *dev, const struct screen_info *si,
+			    const struct drm_format_info *format,
+			    unsigned int width, unsigned int height, u64 size)
+{
+	u64 lfb_linelength = si->lfb_linelength;
+
+	if (!lfb_linelength)
+		lfb_linelength = drm_format_info_min_pitch(format, 0, width);
+
+	return drm_sysfb_get_validated_int0(dev, "stride", lfb_linelength, div64_u64(size, height));
+}
+EXPORT_SYMBOL(drm_sysfb_get_stride_si);
+
+u64 drm_sysfb_get_visible_size_si(struct drm_device *dev, const struct screen_info *si,
+				  unsigned int height, unsigned int stride, u64 size)
+{
+	u64 vsize = PAGE_ALIGN(height * stride);
+
+	return drm_sysfb_get_validated_size0(dev, "visible size", vsize, size);
+}
+EXPORT_SYMBOL(drm_sysfb_get_visible_size_si);
+
+const struct drm_format_info *drm_sysfb_get_format_si(struct drm_device *dev,
+						      const struct drm_sysfb_format *formats,
+						      size_t nformats,
+						      const struct screen_info *si)
+{
+	const struct drm_format_info *format = NULL;
+	u32 bits_per_pixel;
+	size_t i;
+
+	bits_per_pixel = __screen_info_lfb_bits_per_pixel(si);
+
+	for (i = 0; i < nformats; ++i) {
+		const struct pixel_format *f = &formats[i].pixel;
+
+		if (bits_per_pixel == f->bits_per_pixel &&
+		    si->red_size == f->red.length &&
+		    si->red_pos == f->red.offset &&
+		    si->green_size == f->green.length &&
+		    si->green_pos == f->green.offset &&
+		    si->blue_size == f->blue.length &&
+		    si->blue_pos == f->blue.offset) {
+			format = drm_format_info(formats[i].fourcc);
+			break;
+		}
+	}
+
+	if (!format)
+		drm_warn(dev, "No compatible color format found\n");
+
+	return format;
+}
+EXPORT_SYMBOL(drm_sysfb_get_format_si);
diff --git a/drivers/gpu/drm/sysfb/efidrm.c b/drivers/gpu/drm/sysfb/efidrm.c
new file mode 100644
index 000000000000..46912924636a
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/efidrm.c
@@ -0,0 +1,389 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/aperture.h>
+#include <linux/efi.h>
+#include <linux/limits.h>
+#include <linux/platform_device.h>
+#include <linux/screen_info.h>
+
+#include <drm/clients/drm_client_setup.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_fbdev_shmem.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_gem_atomic_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_managed.h>
+#include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_probe_helper.h>
+
+#include <video/edid.h>
+#include <video/pixel_format.h>
+
+#include "drm_sysfb_helper.h"
+
+#define DRIVER_NAME	"efidrm"
+#define DRIVER_DESC	"DRM driver for EFI platform devices"
+#define DRIVER_MAJOR	1
+#define DRIVER_MINOR	0
+
+static const struct drm_format_info *efidrm_get_format_si(struct drm_device *dev,
+							  const struct screen_info *si)
+{
+	static const struct drm_sysfb_format formats[] = {
+		{ PIXEL_FORMAT_XRGB1555, DRM_FORMAT_XRGB1555, },
+		{ PIXEL_FORMAT_RGB565, DRM_FORMAT_RGB565, },
+		{ PIXEL_FORMAT_RGB888, DRM_FORMAT_RGB888, },
+		{ PIXEL_FORMAT_XRGB8888, DRM_FORMAT_XRGB8888, },
+		{ PIXEL_FORMAT_XBGR8888, DRM_FORMAT_XBGR8888, },
+		{ PIXEL_FORMAT_XRGB2101010, DRM_FORMAT_XRGB2101010, },
+	};
+
+	return drm_sysfb_get_format_si(dev, formats, ARRAY_SIZE(formats), si);
+}
+
+static u64 efidrm_get_mem_flags(struct drm_device *dev, resource_size_t start,
+				resource_size_t len)
+{
+	u64 attribute = EFI_MEMORY_UC | EFI_MEMORY_WC |
+			EFI_MEMORY_WT | EFI_MEMORY_WB;
+	u64 mem_flags = EFI_MEMORY_WC | EFI_MEMORY_UC;
+	resource_size_t end = start + len;
+	efi_memory_desc_t md;
+	u64 md_end;
+
+	if (!efi_enabled(EFI_MEMMAP) || efi_mem_desc_lookup(start, &md))
+		goto out;
+
+	md_end = md.phys_addr + (md.num_pages << EFI_PAGE_SHIFT);
+	if (end > md_end)
+		goto out;
+
+	attribute &= md.attribute;
+	if (attribute) {
+		mem_flags |= EFI_MEMORY_WT | EFI_MEMORY_WB;
+		mem_flags &= attribute;
+	}
+
+out:
+	return mem_flags;
+}
+
+/*
+ * EFI device
+ */
+
+struct efidrm_device {
+	struct drm_sysfb_device sysfb;
+
+	/* modesetting */
+	u32 formats[DRM_SYSFB_PLANE_NFORMATS(1)];
+	struct drm_plane primary_plane;
+	struct drm_crtc crtc;
+	struct drm_encoder encoder;
+	struct drm_connector connector;
+};
+
+/*
+ * Modesetting
+ */
+
+static const u64 efidrm_primary_plane_format_modifiers[] = {
+	DRM_SYSFB_PLANE_FORMAT_MODIFIERS,
+};
+
+static const struct drm_plane_helper_funcs efidrm_primary_plane_helper_funcs = {
+	DRM_SYSFB_PLANE_HELPER_FUNCS,
+};
+
+static const struct drm_plane_funcs efidrm_primary_plane_funcs = {
+	DRM_SYSFB_PLANE_FUNCS,
+	.destroy = drm_plane_cleanup,
+};
+
+static const struct drm_crtc_helper_funcs efidrm_crtc_helper_funcs = {
+	DRM_SYSFB_CRTC_HELPER_FUNCS,
+};
+
+static const struct drm_crtc_funcs efidrm_crtc_funcs = {
+	DRM_SYSFB_CRTC_FUNCS,
+	.destroy = drm_crtc_cleanup,
+};
+
+static const struct drm_encoder_funcs efidrm_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
+static const struct drm_connector_helper_funcs efidrm_connector_helper_funcs = {
+	DRM_SYSFB_CONNECTOR_HELPER_FUNCS,
+};
+
+static const struct drm_connector_funcs efidrm_connector_funcs = {
+	DRM_SYSFB_CONNECTOR_FUNCS,
+	.destroy = drm_connector_cleanup,
+};
+
+static const struct drm_mode_config_funcs efidrm_mode_config_funcs = {
+	DRM_SYSFB_MODE_CONFIG_FUNCS,
+};
+
+/*
+ * Init / Cleanup
+ */
+
+static struct efidrm_device *efidrm_device_create(struct drm_driver *drv,
+						  struct platform_device *pdev)
+{
+	const struct screen_info *si;
+	const struct drm_format_info *format;
+	int width, height, stride;
+	u64 vsize, mem_flags;
+	struct resource resbuf;
+	struct resource *res;
+	struct efidrm_device *efi;
+	struct drm_sysfb_device *sysfb;
+	struct drm_device *dev;
+	struct resource *mem = NULL;
+	void __iomem *screen_base = NULL;
+	struct drm_plane *primary_plane;
+	struct drm_crtc *crtc;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+	unsigned long max_width, max_height;
+	size_t nformats;
+	int ret;
+
+	si = dev_get_platdata(&pdev->dev);
+	if (!si)
+		return ERR_PTR(-ENODEV);
+	if (screen_info_video_type(si) != VIDEO_TYPE_EFI)
+		return ERR_PTR(-ENODEV);
+
+	/*
+	 * EFI DRM driver
+	 */
+
+	efi = devm_drm_dev_alloc(&pdev->dev, drv, struct efidrm_device, sysfb.dev);
+	if (IS_ERR(efi))
+		return ERR_CAST(efi);
+	sysfb = &efi->sysfb;
+	dev = &sysfb->dev;
+	platform_set_drvdata(pdev, dev);
+
+	/*
+	 * Hardware settings
+	 */
+
+	format = efidrm_get_format_si(dev, si);
+	if (!format)
+		return ERR_PTR(-EINVAL);
+	width = drm_sysfb_get_width_si(dev, si);
+	if (width < 0)
+		return ERR_PTR(width);
+	height = drm_sysfb_get_height_si(dev, si);
+	if (height < 0)
+		return ERR_PTR(height);
+	res = drm_sysfb_get_memory_si(dev, si, &resbuf);
+	if (!res)
+		return ERR_PTR(-EINVAL);
+	stride = drm_sysfb_get_stride_si(dev, si, format, width, height, resource_size(res));
+	if (stride < 0)
+		return ERR_PTR(stride);
+	vsize = drm_sysfb_get_visible_size_si(dev, si, height, stride, resource_size(res));
+	if (!vsize)
+		return ERR_PTR(-EINVAL);
+
+	drm_dbg(dev, "framebuffer format=%p4cc, size=%dx%d, stride=%d bytes\n",
+		&format->format, width, height, stride);
+
+#ifdef CONFIG_X86
+	if (drm_edid_header_is_valid(edid_info.dummy) == 8)
+		sysfb->edid = edid_info.dummy;
+#endif
+	sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
+	sysfb->fb_format = format;
+	sysfb->fb_pitch = stride;
+
+	/*
+	 * Memory management
+	 */
+
+	ret = devm_aperture_acquire_for_platform_device(pdev, res->start, vsize);
+	if (ret) {
+		drm_err(dev, "could not acquire memory range %pr: %d\n", res, ret);
+		return ERR_PTR(ret);
+	}
+
+	drm_dbg(dev, "using I/O memory framebuffer at %pr\n", res);
+
+	mem = devm_request_mem_region(&pdev->dev, res->start, vsize, drv->name);
+	if (!mem) {
+		/*
+		 * We cannot make this fatal. Sometimes this comes from magic
+		 * spaces our resource handlers simply don't know about. Use
+		 * the I/O-memory resource as-is and try to map that instead.
+		 */
+		drm_warn(dev, "could not acquire memory region %pr\n", res);
+		mem = res;
+	}
+
+	mem_flags = efidrm_get_mem_flags(dev, res->start, vsize);
+
+	if (mem_flags & EFI_MEMORY_WC)
+		screen_base = devm_ioremap_wc(&pdev->dev, mem->start, resource_size(mem));
+	else if (mem_flags & EFI_MEMORY_UC)
+		screen_base = devm_ioremap(&pdev->dev, mem->start, resource_size(mem));
+	else if (mem_flags & EFI_MEMORY_WT)
+		screen_base = devm_memremap(&pdev->dev, mem->start, resource_size(mem),
+					    MEMREMAP_WT);
+	else if (mem_flags & EFI_MEMORY_WB)
+		screen_base = devm_memremap(&pdev->dev, mem->start, resource_size(mem),
+					    MEMREMAP_WB);
+	else
+		drm_err(dev, "invalid mem_flags: 0x%llx\n", mem_flags);
+	if (!screen_base)
+		return ERR_PTR(-ENOMEM);
+	iosys_map_set_vaddr_iomem(&sysfb->fb_addr, screen_base);
+
+	/*
+	 * Modesetting
+	 */
+
+	ret = drmm_mode_config_init(dev);
+	if (ret)
+		return ERR_PTR(ret);
+
+	max_width = max_t(unsigned long, width, DRM_SHADOW_PLANE_MAX_WIDTH);
+	max_height = max_t(unsigned long, height, DRM_SHADOW_PLANE_MAX_HEIGHT);
+
+	dev->mode_config.min_width = width;
+	dev->mode_config.max_width = max_width;
+	dev->mode_config.min_height = height;
+	dev->mode_config.max_height = max_height;
+	dev->mode_config.preferred_depth = format->depth;
+	dev->mode_config.funcs = &efidrm_mode_config_funcs;
+
+	/* Primary plane */
+
+	nformats = drm_fb_build_fourcc_list(dev, &format->format, 1,
+					    efi->formats, ARRAY_SIZE(efi->formats));
+
+	primary_plane = &efi->primary_plane;
+	ret = drm_universal_plane_init(dev, primary_plane, 0, &efidrm_primary_plane_funcs,
+				       efi->formats, nformats,
+				       efidrm_primary_plane_format_modifiers,
+				       DRM_PLANE_TYPE_PRIMARY, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+	drm_plane_helper_add(primary_plane, &efidrm_primary_plane_helper_funcs);
+	drm_plane_enable_fb_damage_clips(primary_plane);
+
+	/* CRTC */
+
+	crtc = &efi->crtc;
+	ret = drm_crtc_init_with_planes(dev, crtc, primary_plane, NULL,
+					&efidrm_crtc_funcs, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+	drm_crtc_helper_add(crtc, &efidrm_crtc_helper_funcs);
+
+	/* Encoder */
+
+	encoder = &efi->encoder;
+	ret = drm_encoder_init(dev, encoder, &efidrm_encoder_funcs,
+			       DRM_MODE_ENCODER_NONE, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+	encoder->possible_crtcs = drm_crtc_mask(crtc);
+
+	/* Connector */
+
+	connector = &efi->connector;
+	ret = drm_connector_init(dev, connector, &efidrm_connector_funcs,
+				 DRM_MODE_CONNECTOR_Unknown);
+	if (ret)
+		return ERR_PTR(ret);
+	drm_connector_helper_add(connector, &efidrm_connector_helper_funcs);
+	drm_connector_set_panel_orientation_with_quirk(connector,
+						       DRM_MODE_PANEL_ORIENTATION_UNKNOWN,
+						       width, height);
+	if (sysfb->edid)
+		drm_connector_attach_edid_property(connector);
+
+	ret = drm_connector_attach_encoder(connector, encoder);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_mode_config_reset(dev);
+
+	return efi;
+}
+
+/*
+ * DRM driver
+ */
+
+DEFINE_DRM_GEM_FOPS(efidrm_fops);
+
+static struct drm_driver efidrm_driver = {
+	DRM_GEM_SHMEM_DRIVER_OPS,
+	DRM_FBDEV_SHMEM_DRIVER_OPS,
+	.name			= DRIVER_NAME,
+	.desc			= DRIVER_DESC,
+	.major			= DRIVER_MAJOR,
+	.minor			= DRIVER_MINOR,
+	.driver_features	= DRIVER_ATOMIC | DRIVER_GEM | DRIVER_MODESET,
+	.fops			= &efidrm_fops,
+};
+
+/*
+ * Platform driver
+ */
+
+static int efidrm_probe(struct platform_device *pdev)
+{
+	struct efidrm_device *efi;
+	struct drm_sysfb_device *sysfb;
+	struct drm_device *dev;
+	int ret;
+
+	efi = efidrm_device_create(&efidrm_driver, pdev);
+	if (IS_ERR(efi))
+		return PTR_ERR(efi);
+	sysfb = &efi->sysfb;
+	dev = &sysfb->dev;
+
+	ret = drm_dev_register(dev, 0);
+	if (ret)
+		return ret;
+
+	drm_client_setup(dev, sysfb->fb_format);
+
+	return 0;
+}
+
+static void efidrm_remove(struct platform_device *pdev)
+{
+	struct drm_device *dev = platform_get_drvdata(pdev);
+
+	drm_dev_unplug(dev);
+}
+
+static struct platform_driver efidrm_platform_driver = {
+	.driver = {
+		.name = "efi-framebuffer",
+	},
+	.probe = efidrm_probe,
+	.remove = efidrm_remove,
+};
+
+module_platform_driver(efidrm_platform_driver);
+
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tiny/ofdrm.c b/drivers/gpu/drm/sysfb/ofdrm.c
index 13491c0e704a..fddfe8bea9f7 100644
--- a/drivers/gpu/drm/tiny/ofdrm.c
+++ b/drivers/gpu/drm/sysfb/ofdrm.c
@@ -12,6 +12,7 @@
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_device.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_edid.h>
 #include <drm/drm_fbdev_shmem.h>
 #include <drm/drm_format_helper.h>
 #include <drm/drm_framebuffer.h>
@@ -21,7 +22,8 @@
 #include <drm/drm_managed.h>
 #include <drm/drm_modeset_helper_vtables.h>
 #include <drm/drm_probe_helper.h>
-#include <drm/drm_simple_kms_helper.h>
+
+#include "drm_sysfb_helper.h"
 
 #define DRIVER_NAME	"ofdrm"
 #define DRIVER_DESC	"DRM driver for OF platform devices"
@@ -76,20 +78,12 @@ enum ofdrm_model {
 
 static int display_get_validated_int(struct drm_device *dev, const char *name, uint32_t value)
 {
-	if (value > INT_MAX) {
-		drm_err(dev, "invalid framebuffer %s of %u\n", name, value);
-		return -EINVAL;
-	}
-	return (int)value;
+	return drm_sysfb_get_validated_int(dev, name, value, INT_MAX);
 }
 
 static int display_get_validated_int0(struct drm_device *dev, const char *name, uint32_t value)
 {
-	if (!value) {
-		drm_err(dev, "invalid framebuffer %s of %u\n", name, value);
-		return -EINVAL;
-	}
-	return display_get_validated_int(dev, name, value);
+	return drm_sysfb_get_validated_int0(dev, name, value, INT_MAX);
 }
 
 static const struct drm_format_info *display_get_validated_format(struct drm_device *dev,
@@ -226,6 +220,16 @@ static u64 display_get_address_of(struct drm_device *dev, struct device_node *of
 	return address;
 }
 
+static const u8 *display_get_edid_of(struct drm_device *dev, struct device_node *of_node,
+				     u8 buf[EDID_LENGTH])
+{
+	int ret = of_property_read_u8_array(of_node, "EDID", buf, EDID_LENGTH);
+
+	if (ret)
+		return NULL;
+	return buf;
+}
+
 static bool is_avivo(u32 vendor, u32 device)
 {
 	/* This will match most R5xx */
@@ -290,22 +294,17 @@ struct ofdrm_device_funcs {
 };
 
 struct ofdrm_device {
-	struct drm_device dev;
-	struct platform_device *pdev;
+	struct drm_sysfb_device sysfb;
 
 	const struct ofdrm_device_funcs *funcs;
 
-	/* firmware-buffer settings */
-	struct iosys_map screen_base;
-	struct drm_display_mode mode;
-	const struct drm_format_info *format;
-	unsigned int pitch;
-
 	/* colormap */
 	void __iomem *cmap_base;
 
+	u8 edid[EDID_LENGTH];
+
 	/* modesetting */
-	uint32_t formats[8];
+	u32 formats[DRM_SYSFB_PLANE_NFORMATS(1)];
 	struct drm_plane primary_plane;
 	struct drm_crtc crtc;
 	struct drm_encoder encoder;
@@ -314,7 +313,7 @@ struct ofdrm_device {
 
 static struct ofdrm_device *ofdrm_device_of_dev(struct drm_device *dev)
 {
-	return container_of(dev, struct ofdrm_device, dev);
+	return container_of(to_drm_sysfb_device(dev), struct ofdrm_device, sysfb);
 }
 
 /*
@@ -354,7 +353,7 @@ static void ofdrm_pci_release(void *data)
 
 static int ofdrm_device_init_pci(struct ofdrm_device *odev)
 {
-	struct drm_device *dev = &odev->dev;
+	struct drm_device *dev = &odev->sysfb.dev;
 	struct platform_device *pdev = to_platform_device(dev->dev);
 	struct device_node *of_node = pdev->dev.of_node;
 	struct pci_dev *pcidev;
@@ -397,7 +396,7 @@ static int ofdrm_device_init_pci(struct ofdrm_device *odev)
 static struct resource *ofdrm_find_fb_resource(struct ofdrm_device *odev,
 					       struct resource *fb_res)
 {
-	struct platform_device *pdev = to_platform_device(odev->dev.dev);
+	struct platform_device *pdev = to_platform_device(odev->sysfb.dev.dev);
 	struct resource *res, *max_res = NULL;
 	u32 i;
 
@@ -423,7 +422,7 @@ static struct resource *ofdrm_find_fb_resource(struct ofdrm_device *odev,
 static void __iomem *get_cmap_address_of(struct ofdrm_device *odev, struct device_node *of_node,
 					 int bar_no, unsigned long offset, unsigned long size)
 {
-	struct drm_device *dev = &odev->dev;
+	struct drm_device *dev = &odev->sysfb.dev;
 	const __be32 *addr_p;
 	u64 max_size, address;
 	unsigned int flags;
@@ -456,7 +455,7 @@ static void __iomem *ofdrm_mach64_cmap_ioremap(struct ofdrm_device *odev,
 					       struct device_node *of_node,
 					       u64 fb_base)
 {
-	struct drm_device *dev = &odev->dev;
+	struct drm_device *dev = &odev->sysfb.dev;
 	u64 address;
 	void __iomem *cmap_base;
 
@@ -618,7 +617,7 @@ static void __iomem *ofdrm_qemu_cmap_ioremap(struct ofdrm_device *odev,
 		cpu_to_be32(0x00),
 	};
 
-	struct drm_device *dev = &odev->dev;
+	struct drm_device *dev = &odev->sysfb.dev;
 	u64 address;
 	void __iomem *cmap_base;
 
@@ -648,7 +647,7 @@ static void ofdrm_qemu_cmap_write(struct ofdrm_device *odev, unsigned char index
 static void ofdrm_device_set_gamma_linear(struct ofdrm_device *odev,
 					  const struct drm_format_info *format)
 {
-	struct drm_device *dev = &odev->dev;
+	struct drm_device *dev = &odev->sysfb.dev;
 	int i;
 
 	switch (format->format) {
@@ -687,7 +686,7 @@ static void ofdrm_device_set_gamma(struct ofdrm_device *odev,
 				   const struct drm_format_info *format,
 				   struct drm_color_lut *lut)
 {
-	struct drm_device *dev = &odev->dev;
+	struct drm_device *dev = &odev->sysfb.dev;
 	int i;
 
 	switch (format->format) {
@@ -731,205 +730,27 @@ static void ofdrm_device_set_gamma(struct ofdrm_device *odev,
  * Modesetting
  */
 
-struct ofdrm_crtc_state {
-	struct drm_crtc_state base;
-
-	/* Primary-plane format; required for color mgmt. */
-	const struct drm_format_info *format;
+static const u64 ofdrm_primary_plane_format_modifiers[] = {
+	DRM_SYSFB_PLANE_FORMAT_MODIFIERS,
 };
 
-static struct ofdrm_crtc_state *to_ofdrm_crtc_state(struct drm_crtc_state *base)
-{
-	return container_of(base, struct ofdrm_crtc_state, base);
-}
-
-static void ofdrm_crtc_state_destroy(struct ofdrm_crtc_state *ofdrm_crtc_state)
-{
-	__drm_atomic_helper_crtc_destroy_state(&ofdrm_crtc_state->base);
-	kfree(ofdrm_crtc_state);
-}
-
-static const uint64_t ofdrm_primary_plane_format_modifiers[] = {
-	DRM_FORMAT_MOD_LINEAR,
-	DRM_FORMAT_MOD_INVALID
-};
-
-static int ofdrm_primary_plane_helper_atomic_check(struct drm_plane *plane,
-						   struct drm_atomic_state *new_state)
-{
-	struct drm_device *dev = plane->dev;
-	struct ofdrm_device *odev = ofdrm_device_of_dev(dev);
-	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(new_state, plane);
-	struct drm_shadow_plane_state *new_shadow_plane_state =
-		to_drm_shadow_plane_state(new_plane_state);
-	struct drm_framebuffer *new_fb = new_plane_state->fb;
-	struct drm_crtc *new_crtc = new_plane_state->crtc;
-	struct drm_crtc_state *new_crtc_state = NULL;
-	struct ofdrm_crtc_state *new_ofdrm_crtc_state;
-	int ret;
-
-	if (new_crtc)
-		new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
-
-	ret = drm_atomic_helper_check_plane_state(new_plane_state, new_crtc_state,
-						  DRM_PLANE_NO_SCALING,
-						  DRM_PLANE_NO_SCALING,
-						  false, false);
-	if (ret)
-		return ret;
-	else if (!new_plane_state->visible)
-		return 0;
-
-	if (new_fb->format != odev->format) {
-		void *buf;
-
-		/* format conversion necessary; reserve buffer */
-		buf = drm_format_conv_state_reserve(&new_shadow_plane_state->fmtcnv_state,
-						    odev->pitch, GFP_KERNEL);
-		if (!buf)
-			return -ENOMEM;
-	}
-
-	new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
-
-	new_ofdrm_crtc_state = to_ofdrm_crtc_state(new_crtc_state);
-	new_ofdrm_crtc_state->format = new_fb->format;
-
-	return 0;
-}
-
-static void ofdrm_primary_plane_helper_atomic_update(struct drm_plane *plane,
-						     struct drm_atomic_state *state)
-{
-	struct drm_device *dev = plane->dev;
-	struct ofdrm_device *odev = ofdrm_device_of_dev(dev);
-	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
-	struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, plane);
-	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
-	struct drm_framebuffer *fb = plane_state->fb;
-	unsigned int dst_pitch = odev->pitch;
-	const struct drm_format_info *dst_format = odev->format;
-	struct drm_atomic_helper_damage_iter iter;
-	struct drm_rect damage;
-	int ret, idx;
-
-	ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE);
-	if (ret)
-		return;
-
-	if (!drm_dev_enter(dev, &idx))
-		goto out_drm_gem_fb_end_cpu_access;
-
-	drm_atomic_helper_damage_iter_init(&iter, old_plane_state, plane_state);
-	drm_atomic_for_each_plane_damage(&iter, &damage) {
-		struct iosys_map dst = odev->screen_base;
-		struct drm_rect dst_clip = plane_state->dst;
-
-		if (!drm_rect_intersect(&dst_clip, &damage))
-			continue;
-
-		iosys_map_incr(&dst, drm_fb_clip_offset(dst_pitch, dst_format, &dst_clip));
-		drm_fb_blit(&dst, &dst_pitch, dst_format->format, shadow_plane_state->data, fb,
-			    &damage, &shadow_plane_state->fmtcnv_state);
-	}
-
-	drm_dev_exit(idx);
-out_drm_gem_fb_end_cpu_access:
-	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);
-}
-
-static void ofdrm_primary_plane_helper_atomic_disable(struct drm_plane *plane,
-						      struct drm_atomic_state *state)
-{
-	struct drm_device *dev = plane->dev;
-	struct ofdrm_device *odev = ofdrm_device_of_dev(dev);
-	struct iosys_map dst = odev->screen_base;
-	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
-	void __iomem *dst_vmap = dst.vaddr_iomem; /* TODO: Use mapping abstraction */
-	unsigned int dst_pitch = odev->pitch;
-	const struct drm_format_info *dst_format = odev->format;
-	struct drm_rect dst_clip;
-	unsigned long lines, linepixels, i;
-	int idx;
-
-	drm_rect_init(&dst_clip,
-		      plane_state->src_x >> 16, plane_state->src_y >> 16,
-		      plane_state->src_w >> 16, plane_state->src_h >> 16);
-
-	lines = drm_rect_height(&dst_clip);
-	linepixels = drm_rect_width(&dst_clip);
-
-	if (!drm_dev_enter(dev, &idx))
-		return;
-
-	/* Clear buffer to black if disabled */
-	dst_vmap += drm_fb_clip_offset(dst_pitch, dst_format, &dst_clip);
-	for (i = 0; i < lines; ++i) {
-		memset_io(dst_vmap, 0, linepixels * dst_format->cpp[0]);
-		dst_vmap += dst_pitch;
-	}
-
-	drm_dev_exit(idx);
-}
-
 static const struct drm_plane_helper_funcs ofdrm_primary_plane_helper_funcs = {
-	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
-	.atomic_check = ofdrm_primary_plane_helper_atomic_check,
-	.atomic_update = ofdrm_primary_plane_helper_atomic_update,
-	.atomic_disable = ofdrm_primary_plane_helper_atomic_disable,
+	DRM_SYSFB_PLANE_HELPER_FUNCS,
 };
 
 static const struct drm_plane_funcs ofdrm_primary_plane_funcs = {
-	.update_plane = drm_atomic_helper_update_plane,
-	.disable_plane = drm_atomic_helper_disable_plane,
+	DRM_SYSFB_PLANE_FUNCS,
 	.destroy = drm_plane_cleanup,
-	DRM_GEM_SHADOW_PLANE_FUNCS,
 };
 
-static enum drm_mode_status ofdrm_crtc_helper_mode_valid(struct drm_crtc *crtc,
-							 const struct drm_display_mode *mode)
-{
-	struct ofdrm_device *odev = ofdrm_device_of_dev(crtc->dev);
-
-	return drm_crtc_helper_mode_valid_fixed(crtc, mode, &odev->mode);
-}
-
-static int ofdrm_crtc_helper_atomic_check(struct drm_crtc *crtc,
-					  struct drm_atomic_state *new_state)
-{
-	static const size_t gamma_lut_length = OFDRM_GAMMA_LUT_SIZE * sizeof(struct drm_color_lut);
-
-	struct drm_device *dev = crtc->dev;
-	struct drm_crtc_state *new_crtc_state = drm_atomic_get_new_crtc_state(new_state, crtc);
-	int ret;
-
-	if (!new_crtc_state->enable)
-		return 0;
-
-	ret = drm_atomic_helper_check_crtc_primary_plane(new_crtc_state);
-	if (ret)
-		return ret;
-
-	if (new_crtc_state->color_mgmt_changed) {
-		struct drm_property_blob *gamma_lut = new_crtc_state->gamma_lut;
-
-		if (gamma_lut && (gamma_lut->length != gamma_lut_length)) {
-			drm_dbg(dev, "Incorrect gamma_lut length %zu\n", gamma_lut->length);
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
 static void ofdrm_crtc_helper_atomic_flush(struct drm_crtc *crtc, struct drm_atomic_state *state)
 {
 	struct ofdrm_device *odev = ofdrm_device_of_dev(crtc->dev);
 	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
-	struct ofdrm_crtc_state *ofdrm_crtc_state = to_ofdrm_crtc_state(crtc_state);
+	struct drm_sysfb_crtc_state *sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
 
 	if (crtc_state->enable && crtc_state->color_mgmt_changed) {
-		const struct drm_format_info *format = ofdrm_crtc_state->format;
+		const struct drm_format_info *format = sysfb_crtc_state->format;
 
 		if (crtc_state->gamma_lut)
 			ofdrm_device_set_gamma(odev, format, crtc_state->gamma_lut->data);
@@ -938,91 +759,31 @@ static void ofdrm_crtc_helper_atomic_flush(struct drm_crtc *crtc, struct drm_ato
 	}
 }
 
-/*
- * The CRTC is always enabled. Screen updates are performed by
- * the primary plane's atomic_update function. Disabling clears
- * the screen in the primary plane's atomic_disable function.
- */
 static const struct drm_crtc_helper_funcs ofdrm_crtc_helper_funcs = {
-	.mode_valid = ofdrm_crtc_helper_mode_valid,
-	.atomic_check = ofdrm_crtc_helper_atomic_check,
+	DRM_SYSFB_CRTC_HELPER_FUNCS,
 	.atomic_flush = ofdrm_crtc_helper_atomic_flush,
 };
 
-static void ofdrm_crtc_reset(struct drm_crtc *crtc)
-{
-	struct ofdrm_crtc_state *ofdrm_crtc_state =
-		kzalloc(sizeof(*ofdrm_crtc_state), GFP_KERNEL);
-
-	if (crtc->state)
-		ofdrm_crtc_state_destroy(to_ofdrm_crtc_state(crtc->state));
-
-	if (ofdrm_crtc_state)
-		__drm_atomic_helper_crtc_reset(crtc, &ofdrm_crtc_state->base);
-	else
-		__drm_atomic_helper_crtc_reset(crtc, NULL);
-}
-
-static struct drm_crtc_state *ofdrm_crtc_atomic_duplicate_state(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_crtc_state *crtc_state = crtc->state;
-	struct ofdrm_crtc_state *new_ofdrm_crtc_state;
-	struct ofdrm_crtc_state *ofdrm_crtc_state;
-
-	if (drm_WARN_ON(dev, !crtc_state))
-		return NULL;
-
-	new_ofdrm_crtc_state = kzalloc(sizeof(*new_ofdrm_crtc_state), GFP_KERNEL);
-	if (!new_ofdrm_crtc_state)
-		return NULL;
-
-	ofdrm_crtc_state = to_ofdrm_crtc_state(crtc_state);
-
-	__drm_atomic_helper_crtc_duplicate_state(crtc, &new_ofdrm_crtc_state->base);
-	new_ofdrm_crtc_state->format = ofdrm_crtc_state->format;
-
-	return &new_ofdrm_crtc_state->base;
-}
-
-static void ofdrm_crtc_atomic_destroy_state(struct drm_crtc *crtc,
-					    struct drm_crtc_state *crtc_state)
-{
-	ofdrm_crtc_state_destroy(to_ofdrm_crtc_state(crtc_state));
-}
-
 static const struct drm_crtc_funcs ofdrm_crtc_funcs = {
-	.reset = ofdrm_crtc_reset,
+	DRM_SYSFB_CRTC_FUNCS,
 	.destroy = drm_crtc_cleanup,
-	.set_config = drm_atomic_helper_set_config,
-	.page_flip = drm_atomic_helper_page_flip,
-	.atomic_duplicate_state = ofdrm_crtc_atomic_duplicate_state,
-	.atomic_destroy_state = ofdrm_crtc_atomic_destroy_state,
 };
 
-static int ofdrm_connector_helper_get_modes(struct drm_connector *connector)
-{
-	struct ofdrm_device *odev = ofdrm_device_of_dev(connector->dev);
-
-	return drm_connector_helper_get_modes_fixed(connector, &odev->mode);
-}
+static const struct drm_encoder_funcs ofdrm_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
 
 static const struct drm_connector_helper_funcs ofdrm_connector_helper_funcs = {
-	.get_modes = ofdrm_connector_helper_get_modes,
+	DRM_SYSFB_CONNECTOR_HELPER_FUNCS,
 };
 
 static const struct drm_connector_funcs ofdrm_connector_funcs = {
-	.reset = drm_atomic_helper_connector_reset,
-	.fill_modes = drm_helper_probe_single_connector_modes,
+	DRM_SYSFB_CONNECTOR_FUNCS,
 	.destroy = drm_connector_cleanup,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
 
 static const struct drm_mode_config_funcs ofdrm_mode_config_funcs = {
-	.fb_create = drm_gem_fb_create_with_dirty,
-	.atomic_check = drm_atomic_helper_check,
-	.atomic_commit = drm_atomic_helper_commit,
+	DRM_SYSFB_MODE_CONFIG_FUNCS,
 };
 
 /*
@@ -1072,32 +833,19 @@ static const struct ofdrm_device_funcs ofdrm_qemu_device_funcs = {
 	.cmap_write = ofdrm_qemu_cmap_write,
 };
 
-static struct drm_display_mode ofdrm_mode(unsigned int width, unsigned int height)
-{
-	/*
-	 * Assume a monitor resolution of 96 dpi to
-	 * get a somewhat reasonable screen size.
-	 */
-	const struct drm_display_mode mode = {
-		DRM_MODE_INIT(60, width, height,
-			      DRM_MODE_RES_MM(width, 96ul),
-			      DRM_MODE_RES_MM(height, 96ul))
-	};
-
-	return mode;
-}
-
 static struct ofdrm_device *ofdrm_device_create(struct drm_driver *drv,
 						struct platform_device *pdev)
 {
 	struct device_node *of_node = pdev->dev.of_node;
 	struct ofdrm_device *odev;
+	struct drm_sysfb_device *sysfb;
 	struct drm_device *dev;
 	enum ofdrm_model model;
 	bool big_endian;
 	int width, height, depth, linebytes;
 	const struct drm_format_info *format;
 	u64 address;
+	const u8 *edid;
 	resource_size_t fb_size, fb_base, fb_pgbase, fb_pgsize;
 	struct resource *res, *mem;
 	void __iomem *screen_base;
@@ -1109,10 +857,11 @@ static struct ofdrm_device *ofdrm_device_create(struct drm_driver *drv,
 	size_t nformats;
 	int ret;
 
-	odev = devm_drm_dev_alloc(&pdev->dev, drv, struct ofdrm_device, dev);
+	odev = devm_drm_dev_alloc(&pdev->dev, drv, struct ofdrm_device, sysfb.dev);
 	if (IS_ERR(odev))
 		return ERR_CAST(odev);
-	dev = &odev->dev;
+	sysfb = &odev->sysfb;
+	dev = &sysfb->dev;
 	platform_set_drvdata(pdev, dev);
 
 	ret = ofdrm_device_init_pci(odev);
@@ -1246,16 +995,22 @@ static struct ofdrm_device *ofdrm_device_create(struct drm_driver *drv,
 		}
 	}
 
+	/* EDID is optional */
+	edid = display_get_edid_of(dev, of_node, odev->edid);
+
 	/*
 	 * Firmware framebuffer
 	 */
 
-	iosys_map_set_vaddr_iomem(&odev->screen_base, screen_base);
-	odev->mode = ofdrm_mode(width, height);
-	odev->format = format;
-	odev->pitch = linebytes;
+	iosys_map_set_vaddr_iomem(&sysfb->fb_addr, screen_base);
+	sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
+	sysfb->fb_format = format;
+	sysfb->fb_pitch = linebytes;
+	if (odev->cmap_base)
+		sysfb->fb_gamma_lut_size = OFDRM_GAMMA_LUT_SIZE;
+	sysfb->edid = edid;
 
-	drm_dbg(dev, "display mode={" DRM_MODE_FMT "}\n", DRM_MODE_ARG(&odev->mode));
+	drm_dbg(dev, "display mode={" DRM_MODE_FMT "}\n", DRM_MODE_ARG(&sysfb->fb_mode));
 	drm_dbg(dev, "framebuffer format=%p4cc, size=%dx%d, linebytes=%d byte\n",
 		&format->format, width, height, linebytes);
 
@@ -1302,15 +1057,16 @@ static struct ofdrm_device *ofdrm_device_create(struct drm_driver *drv,
 		return ERR_PTR(ret);
 	drm_crtc_helper_add(crtc, &ofdrm_crtc_helper_funcs);
 
-	if (odev->cmap_base) {
-		drm_mode_crtc_set_gamma_size(crtc, OFDRM_GAMMA_LUT_SIZE);
-		drm_crtc_enable_color_mgmt(crtc, 0, false, OFDRM_GAMMA_LUT_SIZE);
+	if (sysfb->fb_gamma_lut_size) {
+		ret = drm_mode_crtc_set_gamma_size(crtc, sysfb->fb_gamma_lut_size);
+		if (!ret)
+			drm_crtc_enable_color_mgmt(crtc, 0, false, sysfb->fb_gamma_lut_size);
 	}
 
 	/* Encoder */
 
 	encoder = &odev->encoder;
-	ret = drm_simple_encoder_init(dev, encoder, DRM_MODE_ENCODER_NONE);
+	ret = drm_encoder_init(dev, encoder, &ofdrm_encoder_funcs, DRM_MODE_ENCODER_NONE, NULL);
 	if (ret)
 		return ERR_PTR(ret);
 	encoder->possible_crtcs = drm_crtc_mask(crtc);
@@ -1326,6 +1082,8 @@ static struct ofdrm_device *ofdrm_device_create(struct drm_driver *drv,
 	drm_connector_set_panel_orientation_with_quirk(connector,
 						       DRM_MODE_PANEL_ORIENTATION_UNKNOWN,
 						       width, height);
+	if (edid)
+		drm_connector_attach_edid_property(connector);
 
 	ret = drm_connector_attach_encoder(connector, encoder);
 	if (ret)
@@ -1360,19 +1118,21 @@ static struct drm_driver ofdrm_driver = {
 static int ofdrm_probe(struct platform_device *pdev)
 {
 	struct ofdrm_device *odev;
+	struct drm_sysfb_device *sysfb;
 	struct drm_device *dev;
 	int ret;
 
 	odev = ofdrm_device_create(&ofdrm_driver, pdev);
 	if (IS_ERR(odev))
 		return PTR_ERR(odev);
-	dev = &odev->dev;
+	sysfb = &odev->sysfb;
+	dev = &sysfb->dev;
 
 	ret = drm_dev_register(dev, 0);
 	if (ret)
 		return ret;
 
-	drm_client_setup(dev, odev->format);
+	drm_client_setup(dev, sysfb->fb_format);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/tiny/simpledrm.c b/drivers/gpu/drm/sysfb/simpledrm.c
index 5d9ab8adf800..a1c3119330de 100644
--- a/drivers/gpu/drm/tiny/simpledrm.c
+++ b/drivers/gpu/drm/sysfb/simpledrm.c
@@ -14,7 +14,6 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_state_helper.h>
 #include <drm/drm_connector.h>
-#include <drm/drm_crtc_helper.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_device.h>
 #include <drm/drm_drv.h>
@@ -26,9 +25,10 @@
 #include <drm/drm_gem_shmem_helper.h>
 #include <drm/drm_managed.h>
 #include <drm/drm_modeset_helper_vtables.h>
-#include <drm/drm_panic.h>
 #include <drm/drm_probe_helper.h>
 
+#include "drm_sysfb_helper.h"
+
 #define DRIVER_NAME	"simpledrm"
 #define DRIVER_DESC	"DRM driver for simple-framebuffer platform devices"
 #define DRIVER_MAJOR	1
@@ -42,24 +42,14 @@ static int
 simplefb_get_validated_int(struct drm_device *dev, const char *name,
 			   uint32_t value)
 {
-	if (value > INT_MAX) {
-		drm_err(dev, "simplefb: invalid framebuffer %s of %u\n",
-			name, value);
-		return -EINVAL;
-	}
-	return (int)value;
+	return drm_sysfb_get_validated_int(dev, name, value, INT_MAX);
 }
 
 static int
 simplefb_get_validated_int0(struct drm_device *dev, const char *name,
 			    uint32_t value)
 {
-	if (!value) {
-		drm_err(dev, "simplefb: invalid framebuffer %s of %u\n",
-			name, value);
-		return -EINVAL;
-	}
-	return simplefb_get_validated_int(dev, name, value);
+	return drm_sysfb_get_validated_int0(dev, name, value, INT_MAX);
 }
 
 static const struct drm_format_info *
@@ -217,7 +207,7 @@ simplefb_get_memory_of(struct drm_device *dev, struct device_node *of_node)
  */
 
 struct simpledrm_device {
-	struct drm_device dev;
+	struct drm_sysfb_device sysfb;
 
 	/* clocks */
 #if defined CONFIG_OF && defined CONFIG_COMMON_CLK
@@ -236,28 +226,14 @@ struct simpledrm_device {
 	struct device_link **pwr_dom_links;
 #endif
 
-	/* simplefb settings */
-	struct drm_display_mode mode;
-	const struct drm_format_info *format;
-	unsigned int pitch;
-
-	/* memory management */
-	struct iosys_map screen_base;
-
 	/* modesetting */
-	uint32_t formats[8];
-	size_t nformats;
+	u32 formats[DRM_SYSFB_PLANE_NFORMATS(1)];
 	struct drm_plane primary_plane;
 	struct drm_crtc crtc;
 	struct drm_encoder encoder;
 	struct drm_connector connector;
 };
 
-static struct simpledrm_device *simpledrm_device_of_dev(struct drm_device *dev)
-{
-	return container_of(dev, struct simpledrm_device, dev);
-}
-
 /*
  * Hardware
  */
@@ -284,7 +260,7 @@ static struct simpledrm_device *simpledrm_device_of_dev(struct drm_device *dev)
 
 static void simpledrm_device_release_clocks(void *res)
 {
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(res);
+	struct simpledrm_device *sdev = res;
 	unsigned int i;
 
 	for (i = 0; i < sdev->clk_count; ++i) {
@@ -297,7 +273,7 @@ static void simpledrm_device_release_clocks(void *res)
 
 static int simpledrm_device_init_clocks(struct simpledrm_device *sdev)
 {
-	struct drm_device *dev = &sdev->dev;
+	struct drm_device *dev = &sdev->sysfb.dev;
 	struct platform_device *pdev = to_platform_device(dev->dev);
 	struct device_node *of_node = pdev->dev.of_node;
 	struct clk *clock;
@@ -382,7 +358,7 @@ static int simpledrm_device_init_clocks(struct simpledrm_device *sdev)
 
 static void simpledrm_device_release_regulators(void *res)
 {
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(res);
+	struct simpledrm_device *sdev = res;
 	unsigned int i;
 
 	for (i = 0; i < sdev->regulator_count; ++i) {
@@ -395,7 +371,7 @@ static void simpledrm_device_release_regulators(void *res)
 
 static int simpledrm_device_init_regulators(struct simpledrm_device *sdev)
 {
-	struct drm_device *dev = &sdev->dev;
+	struct drm_device *dev = &sdev->sysfb.dev;
 	struct platform_device *pdev = to_platform_device(dev->dev);
 	struct device_node *of_node = pdev->dev.of_node;
 	struct property *prop;
@@ -516,7 +492,7 @@ static void simpledrm_device_detach_genpd(void *res)
 
 static int simpledrm_device_attach_genpd(struct simpledrm_device *sdev)
 {
-	struct device *dev = sdev->dev.dev;
+	struct device *dev = sdev->sysfb.dev.dev;
 	int i;
 
 	sdev->pwr_dom_count = of_count_phandle_with_args(dev->of_node, "power-domains",
@@ -548,7 +524,7 @@ static int simpledrm_device_attach_genpd(struct simpledrm_device *sdev)
 				simpledrm_device_detach_genpd(sdev);
 				return ret;
 			}
-			drm_warn(&sdev->dev,
+			drm_warn(&sdev->sysfb.dev,
 				 "pm_domain_attach_by_id(%u) failed: %d\n", i, ret);
 			continue;
 		}
@@ -559,7 +535,7 @@ static int simpledrm_device_attach_genpd(struct simpledrm_device *sdev)
 							 DL_FLAG_PM_RUNTIME |
 							 DL_FLAG_RPM_ACTIVE);
 		if (!sdev->pwr_dom_links[i])
-			drm_warn(&sdev->dev, "failed to link power-domain %d\n", i);
+			drm_warn(&sdev->sysfb.dev, "failed to link power-domain %d\n", i);
 	}
 
 	return devm_add_action_or_reset(dev, simpledrm_device_detach_genpd, sdev);
@@ -575,210 +551,56 @@ static int simpledrm_device_attach_genpd(struct simpledrm_device *sdev)
  * Modesetting
  */
 
-static const uint64_t simpledrm_primary_plane_format_modifiers[] = {
-	DRM_FORMAT_MOD_LINEAR,
-	DRM_FORMAT_MOD_INVALID
+static const u64 simpledrm_primary_plane_format_modifiers[] = {
+	DRM_SYSFB_PLANE_FORMAT_MODIFIERS,
 };
 
-static int simpledrm_primary_plane_helper_atomic_check(struct drm_plane *plane,
-						       struct drm_atomic_state *state)
-{
-	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(state, plane);
-	struct drm_shadow_plane_state *new_shadow_plane_state =
-		to_drm_shadow_plane_state(new_plane_state);
-	struct drm_framebuffer *new_fb = new_plane_state->fb;
-	struct drm_crtc *new_crtc = new_plane_state->crtc;
-	struct drm_crtc_state *new_crtc_state = NULL;
-	struct drm_device *dev = plane->dev;
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(dev);
-	int ret;
-
-	if (new_crtc)
-		new_crtc_state = drm_atomic_get_new_crtc_state(state, new_crtc);
-
-	ret = drm_atomic_helper_check_plane_state(new_plane_state, new_crtc_state,
-						  DRM_PLANE_NO_SCALING,
-						  DRM_PLANE_NO_SCALING,
-						  false, false);
-	if (ret)
-		return ret;
-	else if (!new_plane_state->visible)
-		return 0;
-
-	if (new_fb->format != sdev->format) {
-		void *buf;
-
-		/* format conversion necessary; reserve buffer */
-		buf = drm_format_conv_state_reserve(&new_shadow_plane_state->fmtcnv_state,
-						    sdev->pitch, GFP_KERNEL);
-		if (!buf)
-			return -ENOMEM;
-	}
-
-	return 0;
-}
-
-static void simpledrm_primary_plane_helper_atomic_update(struct drm_plane *plane,
-							 struct drm_atomic_state *state)
-{
-	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
-	struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, plane);
-	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
-	struct drm_framebuffer *fb = plane_state->fb;
-	struct drm_device *dev = plane->dev;
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(dev);
-	struct drm_atomic_helper_damage_iter iter;
-	struct drm_rect damage;
-	int ret, idx;
-
-	ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE);
-	if (ret)
-		return;
-
-	if (!drm_dev_enter(dev, &idx))
-		goto out_drm_gem_fb_end_cpu_access;
-
-	drm_atomic_helper_damage_iter_init(&iter, old_plane_state, plane_state);
-	drm_atomic_for_each_plane_damage(&iter, &damage) {
-		struct drm_rect dst_clip = plane_state->dst;
-		struct iosys_map dst = sdev->screen_base;
-
-		if (!drm_rect_intersect(&dst_clip, &damage))
-			continue;
-
-		iosys_map_incr(&dst, drm_fb_clip_offset(sdev->pitch, sdev->format, &dst_clip));
-		drm_fb_blit(&dst, &sdev->pitch, sdev->format->format, shadow_plane_state->data,
-			    fb, &damage, &shadow_plane_state->fmtcnv_state);
-	}
-
-	drm_dev_exit(idx);
-out_drm_gem_fb_end_cpu_access:
-	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);
-}
-
-static void simpledrm_primary_plane_helper_atomic_disable(struct drm_plane *plane,
-							  struct drm_atomic_state *state)
-{
-	struct drm_device *dev = plane->dev;
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(dev);
-	int idx;
-
-	if (!drm_dev_enter(dev, &idx))
-		return;
-
-	/* Clear screen to black if disabled */
-	iosys_map_memset(&sdev->screen_base, 0, 0, sdev->pitch * sdev->mode.vdisplay);
-
-	drm_dev_exit(idx);
-}
-
-static int simpledrm_primary_plane_helper_get_scanout_buffer(struct drm_plane *plane,
-							     struct drm_scanout_buffer *sb)
-{
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(plane->dev);
-
-	sb->width = sdev->mode.hdisplay;
-	sb->height = sdev->mode.vdisplay;
-	sb->format = sdev->format;
-	sb->pitch[0] = sdev->pitch;
-	sb->map[0] = sdev->screen_base;
-
-	return 0;
-}
-
 static const struct drm_plane_helper_funcs simpledrm_primary_plane_helper_funcs = {
-	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
-	.atomic_check = simpledrm_primary_plane_helper_atomic_check,
-	.atomic_update = simpledrm_primary_plane_helper_atomic_update,
-	.atomic_disable = simpledrm_primary_plane_helper_atomic_disable,
-	.get_scanout_buffer = simpledrm_primary_plane_helper_get_scanout_buffer,
+	DRM_SYSFB_PLANE_HELPER_FUNCS,
 };
 
 static const struct drm_plane_funcs simpledrm_primary_plane_funcs = {
-	.update_plane = drm_atomic_helper_update_plane,
-	.disable_plane = drm_atomic_helper_disable_plane,
+	DRM_SYSFB_PLANE_FUNCS,
 	.destroy = drm_plane_cleanup,
-	DRM_GEM_SHADOW_PLANE_FUNCS,
 };
 
-static enum drm_mode_status simpledrm_crtc_helper_mode_valid(struct drm_crtc *crtc,
-							     const struct drm_display_mode *mode)
-{
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(crtc->dev);
-
-	return drm_crtc_helper_mode_valid_fixed(crtc, mode, &sdev->mode);
-}
-
-/*
- * The CRTC is always enabled. Screen updates are performed by
- * the primary plane's atomic_update function. Disabling clears
- * the screen in the primary plane's atomic_disable function.
- */
 static const struct drm_crtc_helper_funcs simpledrm_crtc_helper_funcs = {
-	.mode_valid = simpledrm_crtc_helper_mode_valid,
-	.atomic_check = drm_crtc_helper_atomic_check,
+	DRM_SYSFB_CRTC_HELPER_FUNCS,
 };
 
 static const struct drm_crtc_funcs simpledrm_crtc_funcs = {
-	.reset = drm_atomic_helper_crtc_reset,
+	DRM_SYSFB_CRTC_FUNCS,
 	.destroy = drm_crtc_cleanup,
-	.set_config = drm_atomic_helper_set_config,
-	.page_flip = drm_atomic_helper_page_flip,
-	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
 };
 
 static const struct drm_encoder_funcs simpledrm_encoder_funcs = {
 	.destroy = drm_encoder_cleanup,
 };
 
-static int simpledrm_connector_helper_get_modes(struct drm_connector *connector)
-{
-	struct simpledrm_device *sdev = simpledrm_device_of_dev(connector->dev);
-
-	return drm_connector_helper_get_modes_fixed(connector, &sdev->mode);
-}
-
 static const struct drm_connector_helper_funcs simpledrm_connector_helper_funcs = {
-	.get_modes = simpledrm_connector_helper_get_modes,
+	DRM_SYSFB_CONNECTOR_HELPER_FUNCS,
 };
 
 static const struct drm_connector_funcs simpledrm_connector_funcs = {
-	.reset = drm_atomic_helper_connector_reset,
-	.fill_modes = drm_helper_probe_single_connector_modes,
+	DRM_SYSFB_CONNECTOR_FUNCS,
 	.destroy = drm_connector_cleanup,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
 
 static const struct drm_mode_config_funcs simpledrm_mode_config_funcs = {
-	.fb_create = drm_gem_fb_create_with_dirty,
-	.atomic_check = drm_atomic_helper_check,
-	.atomic_commit = drm_atomic_helper_commit,
+	DRM_SYSFB_MODE_CONFIG_FUNCS,
 };
 
 /*
  * Init / Cleanup
  */
 
-static struct drm_display_mode simpledrm_mode(unsigned int width,
-					      unsigned int height,
-					      unsigned int width_mm,
-					      unsigned int height_mm)
-{
-	const struct drm_display_mode mode = {
-		DRM_MODE_INIT(60, width, height, width_mm, height_mm)
-	};
-
-	return mode;
-}
-
 static struct simpledrm_device *simpledrm_device_create(struct drm_driver *drv,
 							struct platform_device *pdev)
 {
 	const struct simplefb_platform_data *pd = dev_get_platdata(&pdev->dev);
 	struct device_node *of_node = pdev->dev.of_node;
 	struct simpledrm_device *sdev;
+	struct drm_sysfb_device *sysfb;
 	struct drm_device *dev;
 	int width, height, stride;
 	int width_mm = 0, height_mm = 0;
@@ -793,10 +615,11 @@ static struct simpledrm_device *simpledrm_device_create(struct drm_driver *drv,
 	size_t nformats;
 	int ret;
 
-	sdev = devm_drm_dev_alloc(&pdev->dev, drv, struct simpledrm_device, dev);
+	sdev = devm_drm_dev_alloc(&pdev->dev, drv, struct simpledrm_device, sysfb.dev);
 	if (IS_ERR(sdev))
 		return ERR_CAST(sdev);
-	dev = &sdev->dev;
+	sysfb = &sdev->sysfb;
+	dev = &sysfb->dev;
 	platform_set_drvdata(pdev, sdev);
 
 	/*
@@ -858,20 +681,11 @@ static struct simpledrm_device *simpledrm_device_create(struct drm_driver *drv,
 			return ERR_PTR(-EINVAL);
 	}
 
-	/*
-	 * Assume a monitor resolution of 96 dpi if physical dimensions
-	 * are not specified to get a somewhat reasonable screen size.
-	 */
-	if (!width_mm)
-		width_mm = DRM_MODE_RES_MM(width, 96ul);
-	if (!height_mm)
-		height_mm = DRM_MODE_RES_MM(height, 96ul);
-
-	sdev->mode = simpledrm_mode(width, height, width_mm, height_mm);
-	sdev->format = format;
-	sdev->pitch = stride;
+	sysfb->fb_mode = drm_sysfb_mode(width, height, width_mm, height_mm);
+	sysfb->fb_format = format;
+	sysfb->fb_pitch = stride;
 
-	drm_dbg(dev, "display mode={" DRM_MODE_FMT "}\n", DRM_MODE_ARG(&sdev->mode));
+	drm_dbg(dev, "display mode={" DRM_MODE_FMT "}\n", DRM_MODE_ARG(&sysfb->fb_mode));
 	drm_dbg(dev, "framebuffer format=%p4cc, size=%dx%d, stride=%d byte\n",
 		&format->format, width, height, stride);
 
@@ -895,7 +709,7 @@ static struct simpledrm_device *simpledrm_device_create(struct drm_driver *drv,
 		if (IS_ERR(screen_base))
 			return screen_base;
 
-		iosys_map_set_vaddr(&sdev->screen_base, screen_base);
+		iosys_map_set_vaddr(&sysfb->fb_addr, screen_base);
 	} else {
 		void __iomem *screen_base;
 
@@ -928,7 +742,7 @@ static struct simpledrm_device *simpledrm_device_create(struct drm_driver *drv,
 		if (!screen_base)
 			return ERR_PTR(-ENOMEM);
 
-		iosys_map_set_vaddr_iomem(&sdev->screen_base, screen_base);
+		iosys_map_set_vaddr_iomem(&sysfb->fb_addr, screen_base);
 	}
 
 	/*
@@ -1027,19 +841,21 @@ static struct drm_driver simpledrm_driver = {
 static int simpledrm_probe(struct platform_device *pdev)
 {
 	struct simpledrm_device *sdev;
+	struct drm_sysfb_device *sysfb;
 	struct drm_device *dev;
 	int ret;
 
 	sdev = simpledrm_device_create(&simpledrm_driver, pdev);
 	if (IS_ERR(sdev))
 		return PTR_ERR(sdev);
-	dev = &sdev->dev;
+	sysfb = &sdev->sysfb;
+	dev = &sysfb->dev;
 
 	ret = drm_dev_register(dev, 0);
 	if (ret)
 		return ret;
 
-	drm_client_setup(dev, sdev->format);
+	drm_client_setup(dev, sdev->sysfb.fb_format);
 
 	return 0;
 }
@@ -1047,7 +863,7 @@ static int simpledrm_probe(struct platform_device *pdev)
 static void simpledrm_remove(struct platform_device *pdev)
 {
 	struct simpledrm_device *sdev = platform_get_drvdata(pdev);
-	struct drm_device *dev = &sdev->dev;
+	struct drm_device *dev = &sdev->sysfb.dev;
 
 	drm_dev_unplug(dev);
 }
diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
new file mode 100644
index 000000000000..4d62c78e7d1e
--- /dev/null
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -0,0 +1,554 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/aperture.h>
+#include <linux/ioport.h>
+#include <linux/limits.h>
+#include <linux/platform_device.h>
+#include <linux/screen_info.h>
+
+#include <drm/clients/drm_client_setup.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_connector.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_fbdev_shmem.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_gem_atomic_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_managed.h>
+#include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_probe_helper.h>
+
+#include <video/edid.h>
+#include <video/pixel_format.h>
+#include <video/vga.h>
+
+#include "drm_sysfb_helper.h"
+
+#define DRIVER_NAME	"vesadrm"
+#define DRIVER_DESC	"DRM driver for VESA platform devices"
+#define DRIVER_MAJOR	1
+#define DRIVER_MINOR	0
+
+#define VESADRM_GAMMA_LUT_SIZE 256
+
+static const struct drm_format_info *vesadrm_get_format_si(struct drm_device *dev,
+							   const struct screen_info *si)
+{
+	static const struct drm_sysfb_format formats[] = {
+		{ PIXEL_FORMAT_XRGB1555, DRM_FORMAT_XRGB1555, },
+		{ PIXEL_FORMAT_RGB565, DRM_FORMAT_RGB565, },
+		{ PIXEL_FORMAT_RGB888, DRM_FORMAT_RGB888, },
+		{ PIXEL_FORMAT_XRGB8888, DRM_FORMAT_XRGB8888, },
+		{ PIXEL_FORMAT_XBGR8888, DRM_FORMAT_XBGR8888, },
+	};
+
+	return drm_sysfb_get_format_si(dev, formats, ARRAY_SIZE(formats), si);
+}
+
+/*
+ * VESA device
+ */
+
+struct vesadrm_device {
+	struct drm_sysfb_device sysfb;
+
+#if defined(CONFIG_X86_32)
+	/* VESA Protected Mode interface */
+	struct {
+		const u8 *PrimaryPalette;
+	} pmi;
+#endif
+
+	void (*cmap_write)(struct vesadrm_device *vesa, unsigned int index,
+			   u16 red, u16 green, u16 blue);
+
+	/* modesetting */
+	u32 formats[DRM_SYSFB_PLANE_NFORMATS(1)];
+	struct drm_plane primary_plane;
+	struct drm_crtc crtc;
+	struct drm_encoder encoder;
+	struct drm_connector connector;
+};
+
+static struct vesadrm_device *to_vesadrm_device(struct drm_device *dev)
+{
+	return container_of(to_drm_sysfb_device(dev), struct vesadrm_device, sysfb);
+}
+
+/*
+ * Palette
+ */
+
+static void vesadrm_vga_cmap_write(struct vesadrm_device *vesa, unsigned int index,
+				   u16 red, u16 green, u16 blue)
+{
+	u8 i8, r8, g8, b8;
+
+	if (index > 255)
+		return;
+
+	i8 = index;
+	r8 = red >> 8;
+	g8 = green >> 8;
+	b8 = blue >> 8;
+
+	outb_p(i8, VGA_PEL_IW);
+	outb_p(r8, VGA_PEL_D);
+	outb_p(g8, VGA_PEL_D);
+	outb_p(b8, VGA_PEL_D);
+}
+
+#if defined(CONFIG_X86_32)
+static void vesadrm_pmi_cmap_write(struct vesadrm_device *vesa, unsigned int index,
+				   u16 red, u16 green, u16 blue)
+{
+	u32 i32 = index;
+	struct {
+		u8 b8;
+		u8 g8;
+		u8 r8;
+		u8 x8;
+	} PaletteEntry = {
+		blue >> 8,
+		green >> 8,
+		red >> 8,
+		0x00,
+	};
+
+	if (index > 255)
+		return;
+
+	__asm__ __volatile__ (
+		"call *(%%esi)"
+		: /* no return value */
+		: "a" (0x4f09),
+		  "b" (0),
+		  "c" (1),
+		  "d" (i32),
+		  "D" (&PaletteEntry),
+		  "S" (&vesa->pmi.PrimaryPalette));
+}
+#endif
+
+static void vesadrm_set_gamma_linear(struct vesadrm_device *vesa,
+				     const struct drm_format_info *format)
+{
+	struct drm_device *dev = &vesa->sysfb.dev;
+	size_t i;
+	u16 r16, g16, b16;
+
+	switch (format->format) {
+	case DRM_FORMAT_XRGB1555:
+		for (i = 0; i < 32; ++i) {
+			r16 = i * 8 + i / 4;
+			r16 |= (r16 << 8) | r16;
+			vesa->cmap_write(vesa, i, r16, r16, r16);
+		}
+		break;
+	case DRM_FORMAT_RGB565:
+		for (i = 0; i < 32; ++i) {
+			r16 = i * 8 + i / 4;
+			r16 |= (r16 << 8) | r16;
+			g16 = i * 4 + i / 16;
+			g16 |= (g16 << 8) | g16;
+			b16 = r16;
+			vesa->cmap_write(vesa, i, r16, g16, b16);
+		}
+		for (i = 32; i < 64; ++i) {
+			g16 = i * 4 + i / 16;
+			g16 |= (g16 << 8) | g16;
+			vesa->cmap_write(vesa, i, 0, g16, 0);
+		}
+		break;
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_BGRX8888:
+		for (i = 0; i < 256; ++i) {
+			r16 = (i << 8) | i;
+			vesa->cmap_write(vesa, i, r16, r16, r16);
+		}
+		break;
+	default:
+		drm_warn_once(dev, "Unsupported format %p4cc for gamma correction\n",
+			      &format->format);
+		break;
+	}
+}
+
+static void vesadrm_set_gamma_lut(struct vesadrm_device *vesa,
+				  const struct drm_format_info *format,
+				  struct drm_color_lut *lut)
+{
+	struct drm_device *dev = &vesa->sysfb.dev;
+	size_t i;
+	u16 r16, g16, b16;
+
+	switch (format->format) {
+	case DRM_FORMAT_XRGB1555:
+		for (i = 0; i < 32; ++i) {
+			r16 = lut[i * 8 + i / 4].red;
+			g16 = lut[i * 8 + i / 4].green;
+			b16 = lut[i * 8 + i / 4].blue;
+			vesa->cmap_write(vesa, i, r16, g16, b16);
+		}
+		break;
+	case DRM_FORMAT_RGB565:
+		for (i = 0; i < 32; ++i) {
+			r16 = lut[i * 8 + i / 4].red;
+			g16 = lut[i * 4 + i / 16].green;
+			b16 = lut[i * 8 + i / 4].blue;
+			vesa->cmap_write(vesa, i, r16, g16, b16);
+		}
+		for (i = 32; i < 64; ++i)
+			vesa->cmap_write(vesa, i, 0, lut[i * 4 + i / 16].green, 0);
+		break;
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_BGRX8888:
+		for (i = 0; i < 256; ++i)
+			vesa->cmap_write(vesa, i, lut[i].red, lut[i].green, lut[i].blue);
+		break;
+	default:
+		drm_warn_once(dev, "Unsupported format %p4cc for gamma correction\n",
+			      &format->format);
+		break;
+	}
+}
+
+/*
+ * Modesetting
+ */
+
+static const u64 vesadrm_primary_plane_format_modifiers[] = {
+	DRM_SYSFB_PLANE_FORMAT_MODIFIERS,
+};
+
+static const struct drm_plane_helper_funcs vesadrm_primary_plane_helper_funcs = {
+	DRM_SYSFB_PLANE_HELPER_FUNCS,
+};
+
+static const struct drm_plane_funcs vesadrm_primary_plane_funcs = {
+	DRM_SYSFB_PLANE_FUNCS,
+	.destroy = drm_plane_cleanup,
+};
+
+static void vesadrm_crtc_helper_atomic_flush(struct drm_crtc *crtc,
+					     struct drm_atomic_state *state)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(dev);
+	struct vesadrm_device *vesa = to_vesadrm_device(dev);
+	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+	struct drm_sysfb_crtc_state *sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
+
+	/*
+	 * The gamma LUT has to be reloaded after changing the primary
+	 * plane's color format.
+	 */
+	if (crtc_state->enable && crtc_state->color_mgmt_changed) {
+		if (sysfb_crtc_state->format == sysfb->fb_format) {
+			if (crtc_state->gamma_lut)
+				vesadrm_set_gamma_lut(vesa,
+						      sysfb_crtc_state->format,
+						      crtc_state->gamma_lut->data);
+			else
+				vesadrm_set_gamma_linear(vesa, sysfb_crtc_state->format);
+		} else {
+			vesadrm_set_gamma_linear(vesa, sysfb_crtc_state->format);
+		}
+	}
+}
+
+static const struct drm_crtc_helper_funcs vesadrm_crtc_helper_funcs = {
+	DRM_SYSFB_CRTC_HELPER_FUNCS,
+	.atomic_flush = vesadrm_crtc_helper_atomic_flush,
+};
+
+static const struct drm_crtc_funcs vesadrm_crtc_funcs = {
+	DRM_SYSFB_CRTC_FUNCS,
+	.destroy = drm_crtc_cleanup,
+};
+
+static const struct drm_encoder_funcs vesadrm_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
+static const struct drm_connector_helper_funcs vesadrm_connector_helper_funcs = {
+	DRM_SYSFB_CONNECTOR_HELPER_FUNCS,
+};
+
+static const struct drm_connector_funcs vesadrm_connector_funcs = {
+	DRM_SYSFB_CONNECTOR_FUNCS,
+	.destroy = drm_connector_cleanup,
+};
+
+static const struct drm_mode_config_funcs vesadrm_mode_config_funcs = {
+	DRM_SYSFB_MODE_CONFIG_FUNCS,
+};
+
+/*
+ * Init / Cleanup
+ */
+
+static struct vesadrm_device *vesadrm_device_create(struct drm_driver *drv,
+						    struct platform_device *pdev)
+{
+	const struct screen_info *si;
+	const struct drm_format_info *format;
+	int width, height, stride;
+	u64 vsize;
+	struct resource resbuf;
+	struct resource *res;
+	struct vesadrm_device *vesa;
+	struct drm_sysfb_device *sysfb;
+	struct drm_device *dev;
+	struct resource *mem = NULL;
+	void __iomem *screen_base;
+	struct drm_plane *primary_plane;
+	struct drm_crtc *crtc;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+	unsigned long max_width, max_height;
+	size_t nformats;
+	int ret;
+
+	si = dev_get_platdata(&pdev->dev);
+	if (!si)
+		return ERR_PTR(-ENODEV);
+	if (screen_info_video_type(si) != VIDEO_TYPE_VLFB)
+		return ERR_PTR(-ENODEV);
+
+	/*
+	 * VESA DRM driver
+	 */
+
+	vesa = devm_drm_dev_alloc(&pdev->dev, drv, struct vesadrm_device, sysfb.dev);
+	if (IS_ERR(vesa))
+		return ERR_CAST(vesa);
+	sysfb = &vesa->sysfb;
+	dev = &sysfb->dev;
+	platform_set_drvdata(pdev, dev);
+
+	/*
+	 * Hardware settings
+	 */
+
+	format = vesadrm_get_format_si(dev, si);
+	if (!format)
+		return ERR_PTR(-EINVAL);
+	width = drm_sysfb_get_width_si(dev, si);
+	if (width < 0)
+		return ERR_PTR(width);
+	height = drm_sysfb_get_height_si(dev, si);
+	if (height < 0)
+		return ERR_PTR(height);
+	res = drm_sysfb_get_memory_si(dev, si, &resbuf);
+	if (!res)
+		return ERR_PTR(-EINVAL);
+	stride = drm_sysfb_get_stride_si(dev, si, format, width, height, resource_size(res));
+	if (stride < 0)
+		return ERR_PTR(stride);
+	vsize = drm_sysfb_get_visible_size_si(dev, si, height, stride, resource_size(res));
+	if (!vsize)
+		return ERR_PTR(-EINVAL);
+
+	drm_dbg(dev, "framebuffer format=%p4cc, size=%dx%d, stride=%d bytes\n",
+		&format->format, width, height, stride);
+
+	if (!__screen_info_vbe_mode_nonvga(si)) {
+		vesa->cmap_write = vesadrm_vga_cmap_write;
+#if defined(CONFIG_X86_32)
+	} else {
+		phys_addr_t pmi_base = __screen_info_vesapm_info_base(si);
+		const u16 *pmi_addr = phys_to_virt(pmi_base);
+
+		vesa->pmi.PrimaryPalette = (u8 *)pmi_addr + pmi_addr[2];
+		vesa->cmap_write = vesadrm_pmi_cmap_write;
+#endif
+	}
+
+#ifdef CONFIG_X86
+	if (drm_edid_header_is_valid(edid_info.dummy) == 8)
+		sysfb->edid = edid_info.dummy;
+#endif
+	sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
+	sysfb->fb_format = format;
+	sysfb->fb_pitch = stride;
+	if (vesa->cmap_write)
+		sysfb->fb_gamma_lut_size = VESADRM_GAMMA_LUT_SIZE;
+
+	/*
+	 * Memory management
+	 */
+
+	ret = devm_aperture_acquire_for_platform_device(pdev, res->start, vsize);
+	if (ret) {
+		drm_err(dev, "could not acquire memory range %pr: %d\n", res, ret);
+		return ERR_PTR(ret);
+	}
+
+	drm_dbg(dev, "using I/O memory framebuffer at %pr\n", res);
+
+	mem = devm_request_mem_region(&pdev->dev, res->start, vsize, drv->name);
+	if (!mem) {
+		/*
+		 * We cannot make this fatal. Sometimes this comes from magic
+		 * spaces our resource handlers simply don't know about. Use
+		 * the I/O-memory resource as-is and try to map that instead.
+		 */
+		drm_warn(dev, "could not acquire memory region %pr\n", res);
+		mem = res;
+	}
+
+	screen_base = devm_ioremap_wc(&pdev->dev, mem->start, resource_size(mem));
+	if (!screen_base)
+		return ERR_PTR(-ENOMEM);
+	iosys_map_set_vaddr_iomem(&sysfb->fb_addr, screen_base);
+
+	/*
+	 * Modesetting
+	 */
+
+	ret = drmm_mode_config_init(dev);
+	if (ret)
+		return ERR_PTR(ret);
+
+	max_width = max_t(unsigned long, width, DRM_SHADOW_PLANE_MAX_WIDTH);
+	max_height = max_t(unsigned long, height, DRM_SHADOW_PLANE_MAX_HEIGHT);
+
+	dev->mode_config.min_width = width;
+	dev->mode_config.max_width = max_width;
+	dev->mode_config.min_height = height;
+	dev->mode_config.max_height = max_height;
+	dev->mode_config.preferred_depth = format->depth;
+	dev->mode_config.funcs = &vesadrm_mode_config_funcs;
+
+	/* Primary plane */
+
+	nformats = drm_fb_build_fourcc_list(dev, &format->format, 1,
+					    vesa->formats, ARRAY_SIZE(vesa->formats));
+
+	primary_plane = &vesa->primary_plane;
+	ret = drm_universal_plane_init(dev, primary_plane, 0, &vesadrm_primary_plane_funcs,
+				       vesa->formats, nformats,
+				       vesadrm_primary_plane_format_modifiers,
+				       DRM_PLANE_TYPE_PRIMARY, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+	drm_plane_helper_add(primary_plane, &vesadrm_primary_plane_helper_funcs);
+	drm_plane_enable_fb_damage_clips(primary_plane);
+
+	/* CRTC */
+
+	crtc = &vesa->crtc;
+	ret = drm_crtc_init_with_planes(dev, crtc, primary_plane, NULL,
+					&vesadrm_crtc_funcs, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+	drm_crtc_helper_add(crtc, &vesadrm_crtc_helper_funcs);
+
+	if (sysfb->fb_gamma_lut_size) {
+		ret = drm_mode_crtc_set_gamma_size(crtc, sysfb->fb_gamma_lut_size);
+		if (!ret)
+			drm_crtc_enable_color_mgmt(crtc, 0, false, sysfb->fb_gamma_lut_size);
+	}
+
+	/* Encoder */
+
+	encoder = &vesa->encoder;
+	ret = drm_encoder_init(dev, encoder, &vesadrm_encoder_funcs,
+			       DRM_MODE_ENCODER_NONE, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+	encoder->possible_crtcs = drm_crtc_mask(crtc);
+
+	/* Connector */
+
+	connector = &vesa->connector;
+	ret = drm_connector_init(dev, connector, &vesadrm_connector_funcs,
+				 DRM_MODE_CONNECTOR_Unknown);
+	if (ret)
+		return ERR_PTR(ret);
+	drm_connector_helper_add(connector, &vesadrm_connector_helper_funcs);
+	drm_connector_set_panel_orientation_with_quirk(connector,
+						       DRM_MODE_PANEL_ORIENTATION_UNKNOWN,
+						       width, height);
+	if (sysfb->edid)
+		drm_connector_attach_edid_property(connector);
+
+	ret = drm_connector_attach_encoder(connector, encoder);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_mode_config_reset(dev);
+
+	return vesa;
+}
+
+/*
+ * DRM driver
+ */
+
+DEFINE_DRM_GEM_FOPS(vesadrm_fops);
+
+static struct drm_driver vesadrm_driver = {
+	DRM_GEM_SHMEM_DRIVER_OPS,
+	DRM_FBDEV_SHMEM_DRIVER_OPS,
+	.name			= DRIVER_NAME,
+	.desc			= DRIVER_DESC,
+	.major			= DRIVER_MAJOR,
+	.minor			= DRIVER_MINOR,
+	.driver_features	= DRIVER_ATOMIC | DRIVER_GEM | DRIVER_MODESET,
+	.fops			= &vesadrm_fops,
+};
+
+/*
+ * Platform driver
+ */
+
+static int vesadrm_probe(struct platform_device *pdev)
+{
+	struct vesadrm_device *vesa;
+	struct drm_sysfb_device *sysfb;
+	struct drm_device *dev;
+	int ret;
+
+	vesa = vesadrm_device_create(&vesadrm_driver, pdev);
+	if (IS_ERR(vesa))
+		return PTR_ERR(vesa);
+	sysfb = &vesa->sysfb;
+	dev = &sysfb->dev;
+
+	ret = drm_dev_register(dev, 0);
+	if (ret)
+		return ret;
+
+	drm_client_setup(dev, sysfb->fb_format);
+
+	return 0;
+}
+
+static void vesadrm_remove(struct platform_device *pdev)
+{
+	struct drm_device *dev = platform_get_drvdata(pdev);
+
+	drm_dev_unplug(dev);
+}
+
+static struct platform_driver vesadrm_platform_driver = {
+	.driver = {
+		.name = "vesa-framebuffer",
+	},
+	.probe = vesadrm_probe,
+	.remove = vesadrm_remove,
+};
+
+module_platform_driver(vesadrm_platform_driver);
+
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index 798507a8ae56..59d5c1ba145a 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -1321,10 +1321,16 @@ static struct drm_plane *tegra_dc_add_shared_planes(struct drm_device *drm,
 		if (wgrp->dc == dc->pipe) {
 			for (j = 0; j < wgrp->num_windows; j++) {
 				unsigned int index = wgrp->windows[j];
+				enum drm_plane_type type;
+
+				if (primary)
+					type = DRM_PLANE_TYPE_OVERLAY;
+				else
+					type = DRM_PLANE_TYPE_PRIMARY;
 
 				plane = tegra_shared_plane_create(drm, dc,
 								  wgrp->index,
-								  index);
+								  index, type);
 				if (IS_ERR(plane))
 					return plane;
 
@@ -1332,10 +1338,8 @@ static struct drm_plane *tegra_dc_add_shared_planes(struct drm_device *drm,
 				 * Choose the first shared plane owned by this
 				 * head as the primary plane.
 				 */
-				if (!primary) {
-					plane->type = DRM_PLANE_TYPE_PRIMARY;
+				if (!primary)
 					primary = plane;
-				}
 			}
 		}
 	}
@@ -1389,7 +1393,10 @@ static void tegra_crtc_reset(struct drm_crtc *crtc)
 	if (crtc->state)
 		tegra_crtc_atomic_destroy_state(crtc, crtc->state);
 
-	__drm_atomic_helper_crtc_reset(crtc, &state->base);
+	if (state)
+		__drm_atomic_helper_crtc_reset(crtc, &state->base);
+	else
+		__drm_atomic_helper_crtc_reset(crtc, NULL);
 }
 
 static struct drm_crtc_state *
diff --git a/drivers/gpu/drm/tegra/dp.c b/drivers/gpu/drm/tegra/dp.c
index 08fbd8f151a1..990e744b0923 100644
--- a/drivers/gpu/drm/tegra/dp.c
+++ b/drivers/gpu/drm/tegra/dp.c
@@ -256,73 +256,6 @@ int drm_dp_link_probe(struct drm_dp_aux *aux, struct drm_dp_link *link)
 }
 
 /**
- * drm_dp_link_power_up() - power up a DisplayPort link
- * @aux: DisplayPort AUX channel
- * @link: pointer to a structure containing the link configuration
- *
- * Returns 0 on success or a negative error code on failure.
- */
-int drm_dp_link_power_up(struct drm_dp_aux *aux, struct drm_dp_link *link)
-{
-	u8 value;
-	int err;
-
-	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
-	if (link->revision < 0x11)
-		return 0;
-
-	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
-	if (err < 0)
-		return err;
-
-	value &= ~DP_SET_POWER_MASK;
-	value |= DP_SET_POWER_D0;
-
-	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
-	if (err < 0)
-		return err;
-
-	/*
-	 * According to the DP 1.1 specification, a "Sink Device must exit the
-	 * power saving state within 1 ms" (Section 2.5.3.1, Table 5-52, "Sink
-	 * Control Field" (register 0x600).
-	 */
-	usleep_range(1000, 2000);
-
-	return 0;
-}
-
-/**
- * drm_dp_link_power_down() - power down a DisplayPort link
- * @aux: DisplayPort AUX channel
- * @link: pointer to a structure containing the link configuration
- *
- * Returns 0 on success or a negative error code on failure.
- */
-int drm_dp_link_power_down(struct drm_dp_aux *aux, struct drm_dp_link *link)
-{
-	u8 value;
-	int err;
-
-	/* DP_SET_POWER register is only available on DPCD v1.1 and later */
-	if (link->revision < 0x11)
-		return 0;
-
-	err = drm_dp_dpcd_readb(aux, DP_SET_POWER, &value);
-	if (err < 0)
-		return err;
-
-	value &= ~DP_SET_POWER_MASK;
-	value |= DP_SET_POWER_D3;
-
-	err = drm_dp_dpcd_writeb(aux, DP_SET_POWER, value);
-	if (err < 0)
-		return err;
-
-	return 0;
-}
-
-/**
  * drm_dp_link_configure() - configure a DisplayPort link
  * @aux: DisplayPort AUX channel
  * @link: pointer to a structure containing the link configuration
diff --git a/drivers/gpu/drm/tegra/dp.h b/drivers/gpu/drm/tegra/dp.h
index cb12ed0c54e7..695060cafac0 100644
--- a/drivers/gpu/drm/tegra/dp.h
+++ b/drivers/gpu/drm/tegra/dp.h
@@ -164,8 +164,6 @@ int drm_dp_link_remove_rate(struct drm_dp_link *link, unsigned long rate);
 void drm_dp_link_update_rates(struct drm_dp_link *link);
 
 int drm_dp_link_probe(struct drm_dp_aux *aux, struct drm_dp_link *link);
-int drm_dp_link_power_up(struct drm_dp_aux *aux, struct drm_dp_link *link);
-int drm_dp_link_power_down(struct drm_dp_aux *aux, struct drm_dp_link *link);
 int drm_dp_link_configure(struct drm_dp_aux *aux, struct drm_dp_link *link);
 int drm_dp_link_choose(struct drm_dp_link *link,
 		       const struct drm_display_mode *mode,
diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c
index 2cd8dcb959c0..e5297ac5c0fc 100644
--- a/drivers/gpu/drm/tegra/dpaux.c
+++ b/drivers/gpu/drm/tegra/dpaux.c
@@ -501,14 +501,9 @@ static int tegra_dpaux_probe(struct platform_device *pdev)
 
 	dpaux->vdd = devm_regulator_get_optional(&pdev->dev, "vdd");
 	if (IS_ERR(dpaux->vdd)) {
-		if (PTR_ERR(dpaux->vdd) != -ENODEV) {
-			if (PTR_ERR(dpaux->vdd) != -EPROBE_DEFER)
-				dev_err(&pdev->dev,
-					"failed to get VDD supply: %ld\n",
-					PTR_ERR(dpaux->vdd));
-
-			return PTR_ERR(dpaux->vdd);
-		}
+		if (PTR_ERR(dpaux->vdd) != -ENODEV)
+			return dev_err_probe(&pdev->dev, PTR_ERR(dpaux->vdd),
+					     "failed to get VDD supply\n");
 
 		dpaux->vdd = NULL;
 	}
diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c
index 9bb077558167..b5089b772267 100644
--- a/drivers/gpu/drm/tegra/dsi.c
+++ b/drivers/gpu/drm/tegra/dsi.c
@@ -1564,7 +1564,6 @@ static int tegra_dsi_ganged_probe(struct tegra_dsi *dsi)
 static int tegra_dsi_probe(struct platform_device *pdev)
 {
 	struct tegra_dsi *dsi;
-	struct resource *regs;
 	int err;
 
 	dsi = devm_kzalloc(&pdev->dev, sizeof(*dsi), GFP_KERNEL);
@@ -1636,8 +1635,7 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 		goto remove;
 	}
 
-	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	dsi->regs = devm_ioremap_resource(&pdev->dev, regs);
+	dsi->regs = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(dsi->regs)) {
 		err = PTR_ERR(dsi->regs);
 		goto remove;
diff --git a/drivers/gpu/drm/tegra/falcon.c b/drivers/gpu/drm/tegra/falcon.c
index c0d85463eb1a..17f616bbcb45 100644
--- a/drivers/gpu/drm/tegra/falcon.c
+++ b/drivers/gpu/drm/tegra/falcon.c
@@ -30,6 +30,14 @@ int falcon_wait_idle(struct falcon *falcon)
 				  (value == 0), 10, 100000);
 }
 
+static int falcon_dma_wait_not_full(struct falcon *falcon)
+{
+	u32 value;
+
+	return readl_poll_timeout(falcon->regs + FALCON_DMATRFCMD, value,
+				  !(value & FALCON_DMATRFCMD_FULL), 10, 100000);
+}
+
 static int falcon_dma_wait_idle(struct falcon *falcon)
 {
 	u32 value;
@@ -44,6 +52,7 @@ static int falcon_copy_chunk(struct falcon *falcon,
 			     enum falcon_memory target)
 {
 	u32 cmd = FALCON_DMATRFCMD_SIZE_256B;
+	int err;
 
 	if (target == FALCON_MEMORY_IMEM)
 		cmd |= FALCON_DMATRFCMD_IMEM;
@@ -56,11 +65,15 @@ static int falcon_copy_chunk(struct falcon *falcon,
 	 */
 	cmd |= FALCON_DMATRFCMD_DMACTX(1);
 
+	err = falcon_dma_wait_not_full(falcon);
+	if (err < 0)
+		return err;
+
 	falcon_writel(falcon, offset, FALCON_DMATRFMOFFS);
 	falcon_writel(falcon, base, FALCON_DMATRFFBOFFS);
 	falcon_writel(falcon, cmd, FALCON_DMATRFCMD);
 
-	return falcon_dma_wait_idle(falcon);
+	return 0;
 }
 
 static void falcon_copy_firmware_image(struct falcon *falcon,
@@ -191,6 +204,11 @@ int falcon_boot(struct falcon *falcon)
 		falcon_copy_chunk(falcon, falcon->firmware.code.offset + offset,
 				  offset, FALCON_MEMORY_IMEM);
 
+	/* wait for DMA to complete */
+	err = falcon_dma_wait_idle(falcon);
+	if (err < 0)
+		return err;
+
 	/* setup falcon interrupts */
 	falcon_writel(falcon, FALCON_IRQMSET_EXT(0xff) |
 			      FALCON_IRQMSET_SWGEN1 |
diff --git a/drivers/gpu/drm/tegra/falcon.h b/drivers/gpu/drm/tegra/falcon.h
index 1955cf11a8a6..902bb7e4fd0f 100644
--- a/drivers/gpu/drm/tegra/falcon.h
+++ b/drivers/gpu/drm/tegra/falcon.h
@@ -47,6 +47,7 @@
 #define FALCON_DMATRFMOFFS			0x00001114
 
 #define FALCON_DMATRFCMD			0x00001118
+#define FALCON_DMATRFCMD_FULL			(1 << 0)
 #define FALCON_DMATRFCMD_IDLE			(1 << 1)
 #define FALCON_DMATRFCMD_IMEM			(1 << 4)
 #define FALCON_DMATRFCMD_SIZE_256B		(6 << 8)
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index ace3e5a805cf..dbc1394f96b8 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -17,7 +17,6 @@
 
 #include <drm/drm_drv.h>
 #include <drm/drm_prime.h>
-#include <drm/tegra_drm.h>
 
 #include "drm.h"
 #include "gem.h"
diff --git a/drivers/gpu/drm/tegra/hub.c b/drivers/gpu/drm/tegra/hub.c
index fa6140fc37fb..8f779f23dc09 100644
--- a/drivers/gpu/drm/tegra/hub.c
+++ b/drivers/gpu/drm/tegra/hub.c
@@ -755,9 +755,9 @@ static const struct drm_plane_helper_funcs tegra_shared_plane_helper_funcs = {
 struct drm_plane *tegra_shared_plane_create(struct drm_device *drm,
 					    struct tegra_dc *dc,
 					    unsigned int wgrp,
-					    unsigned int index)
+					    unsigned int index,
+					    enum drm_plane_type type)
 {
-	enum drm_plane_type type = DRM_PLANE_TYPE_OVERLAY;
 	struct tegra_drm *tegra = drm->dev_private;
 	struct tegra_display_hub *hub = tegra->hub;
 	struct tegra_shared_plane *plane;
diff --git a/drivers/gpu/drm/tegra/hub.h b/drivers/gpu/drm/tegra/hub.h
index 23c4b2115ed1..a66f18c4facc 100644
--- a/drivers/gpu/drm/tegra/hub.h
+++ b/drivers/gpu/drm/tegra/hub.h
@@ -80,7 +80,8 @@ void tegra_display_hub_cleanup(struct tegra_display_hub *hub);
 struct drm_plane *tegra_shared_plane_create(struct drm_device *drm,
 					    struct tegra_dc *dc,
 					    unsigned int wgrp,
-					    unsigned int index);
+					    unsigned int index,
+					    enum drm_plane_type type);
 
 int tegra_display_hub_atomic_check(struct drm_device *drm,
 				   struct drm_atomic_state *state);
diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index 1e8ec50b759e..ff5a749710db 100644
--- a/drivers/gpu/drm/tegra/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -200,6 +200,11 @@ static const struct drm_encoder_helper_funcs tegra_rgb_encoder_helper_funcs = {
 	.atomic_check = tegra_rgb_encoder_atomic_check,
 };
 
+static void tegra_dc_of_node_put(void *data)
+{
+	of_node_put(data);
+}
+
 int tegra_dc_rgb_probe(struct tegra_dc *dc)
 {
 	struct device_node *np;
@@ -207,7 +212,14 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc)
 	int err;
 
 	np = of_get_child_by_name(dc->dev->of_node, "rgb");
-	if (!np || !of_device_is_available(np))
+	if (!np)
+		return -ENODEV;
+
+	err = devm_add_action_or_reset(dc->dev, tegra_dc_of_node_put, np);
+	if (err < 0)
+		return err;
+
+	if (!of_device_is_available(np))
 		return -ENODEV;
 
 	rgb = devm_kzalloc(dc->dev, sizeof(*rgb), GFP_KERNEL);
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index f98f70eda906..21f3dfdcc5c9 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -2666,7 +2666,7 @@ static void tegra_sor_dp_disable(struct drm_encoder *encoder)
 	 * the AUX transactions would just be timing out.
 	 */
 	if (output->connector.status != connector_status_disconnected) {
-		err = drm_dp_link_power_down(sor->aux, &sor->link);
+		err = drm_dp_link_power_down(sor->aux, sor->link.revision);
 		if (err < 0)
 			dev_err(sor->dev, "failed to power down link: %d\n",
 				err);
@@ -2882,7 +2882,7 @@ static void tegra_sor_dp_enable(struct drm_encoder *encoder)
 	else
 		dev_dbg(sor->dev, "link training succeeded\n");
 
-	err = drm_dp_link_power_up(sor->aux, &sor->link);
+	err = drm_dp_link_power_up(sor->aux, sor->link.revision);
 	if (err < 0)
 		dev_err(sor->dev, "failed to power up DP link: %d\n", err);
 
diff --git a/drivers/gpu/drm/tests/Makefile b/drivers/gpu/drm/tests/Makefile
index 0109bcf7faa5..3afd6587df08 100644
--- a/drivers/gpu/drm/tests/Makefile
+++ b/drivers/gpu/drm/tests/Makefile
@@ -4,7 +4,9 @@ obj-$(CONFIG_DRM_KUNIT_TEST_HELPERS) += \
 	drm_kunit_helpers.o
 
 obj-$(CONFIG_DRM_KUNIT_TEST) += \
+	drm_atomic_test.o \
 	drm_atomic_state_test.o \
+	drm_bridge_test.o \
 	drm_buddy_test.o \
 	drm_cmdline_parser_test.o \
 	drm_connector_test.o \
diff --git a/drivers/gpu/drm/tests/drm_atomic_test.c b/drivers/gpu/drm/tests/drm_atomic_test.c
new file mode 100644
index 000000000000..ea91bec6569e
--- /dev/null
+++ b/drivers/gpu/drm/tests/drm_atomic_test.c
@@ -0,0 +1,153 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Kunit test for drm_atomic functions
+ */
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_atomic_uapi.h>
+#include <drm/drm_encoder.h>
+#include <drm/drm_kunit_helpers.h>
+#include <drm/drm_modeset_helper_vtables.h>
+
+#include <kunit/test.h>
+
+struct drm_atomic_test_priv {
+	struct drm_device drm;
+	struct drm_plane *plane;
+	struct drm_crtc *crtc;
+	struct drm_encoder encoder;
+	struct drm_connector connector;
+};
+
+static const struct drm_connector_helper_funcs drm_atomic_init_connector_helper_funcs = {
+};
+
+static const struct drm_connector_funcs drm_atomic_init_connector_funcs = {
+	.atomic_destroy_state	= drm_atomic_helper_connector_destroy_state,
+	.atomic_duplicate_state	= drm_atomic_helper_connector_duplicate_state,
+	.reset			= drm_atomic_helper_connector_reset,
+};
+
+static struct drm_atomic_test_priv *create_device(struct kunit *test)
+{
+	struct drm_atomic_test_priv *priv;
+	struct drm_connector *connector;
+	struct drm_encoder *enc;
+	struct drm_device *drm;
+	struct drm_plane *plane;
+	struct drm_crtc *crtc;
+	struct device *dev;
+	int ret;
+
+	dev = drm_kunit_helper_alloc_device(test);
+	if (IS_ERR(dev))
+		return ERR_CAST(dev);
+
+	priv = drm_kunit_helper_alloc_drm_device(test, dev,
+						 struct drm_atomic_test_priv, drm,
+						 DRIVER_MODESET | DRIVER_ATOMIC);
+	if (IS_ERR(priv))
+		return ERR_CAST(priv);
+
+	drm = &priv->drm;
+	plane = drm_kunit_helper_create_primary_plane(test, drm,
+						      NULL,
+						      NULL,
+						      NULL, 0,
+						      NULL);
+	if (IS_ERR(plane))
+		return ERR_CAST(plane);
+	priv->plane = plane;
+
+	crtc = drm_kunit_helper_create_crtc(test, drm,
+					    plane, NULL,
+					    NULL,
+					    NULL);
+	if (IS_ERR(crtc))
+		return ERR_CAST(crtc);
+	priv->crtc = crtc;
+
+	enc = &priv->encoder;
+	ret = drmm_encoder_init(drm, enc, NULL, DRM_MODE_ENCODER_TMDS, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+
+	enc->possible_crtcs = drm_crtc_mask(crtc);
+
+	connector = &priv->connector;
+	ret = drmm_connector_init(drm, connector,
+				  &drm_atomic_init_connector_funcs,
+				  DRM_MODE_CONNECTOR_VIRTUAL,
+				  NULL);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_connector_helper_add(connector, &drm_atomic_init_connector_helper_funcs);
+
+	drm_connector_attach_encoder(connector, enc);
+
+	drm_mode_config_reset(drm);
+
+	return priv;
+}
+
+static void drm_test_drm_atomic_get_connector_for_encoder(struct kunit *test)
+{
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_atomic_test_priv *priv;
+	struct drm_display_mode *mode;
+	struct drm_connector *curr_connector;
+	int ret;
+
+	priv = create_device(test);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv);
+
+	mode = drm_kunit_display_mode_from_cea_vic(test, &priv->drm, 16);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, mode);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_enable:
+	ret = drm_kunit_helper_enable_crtc_connector(test, &priv->drm,
+						     priv->crtc, &priv->connector,
+						     mode, &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_enable;
+	}
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_conn:
+	curr_connector = drm_atomic_get_connector_for_encoder(&priv->encoder,
+							      &ctx);
+	if (PTR_ERR(curr_connector) == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_conn;
+	}
+	KUNIT_EXPECT_PTR_EQ(test, curr_connector, &priv->connector);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+}
+
+static struct kunit_case drm_atomic_get_connector_for_encoder_tests[] = {
+	KUNIT_CASE(drm_test_drm_atomic_get_connector_for_encoder),
+	{ }
+};
+
+
+static struct kunit_suite drm_atomic_get_connector_for_encoder_test_suite = {
+	.name = "drm_test_atomic_get_connector_for_encoder",
+	.test_cases = drm_atomic_get_connector_for_encoder_tests,
+};
+
+kunit_test_suite(drm_atomic_get_connector_for_encoder_test_suite);
+
+MODULE_AUTHOR("Maxime Ripard <mripard@kernel.org>");
+MODULE_DESCRIPTION("Kunit test for drm_atomic functions");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tests/drm_bridge_test.c b/drivers/gpu/drm/tests/drm_bridge_test.c
new file mode 100644
index 000000000000..ff88ec2e911c
--- /dev/null
+++ b/drivers/gpu/drm/tests/drm_bridge_test.c
@@ -0,0 +1,417 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Kunit test for drm_bridge functions
+ */
+#include <drm/drm_atomic_state_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/drm_bridge_helper.h>
+#include <drm/drm_kunit_helpers.h>
+
+#include <kunit/test.h>
+
+struct drm_bridge_init_priv {
+	struct drm_device drm;
+	struct drm_plane *plane;
+	struct drm_crtc *crtc;
+	struct drm_encoder encoder;
+	struct drm_bridge bridge;
+	struct drm_connector *connector;
+	unsigned int enable_count;
+	unsigned int disable_count;
+};
+
+static void drm_test_bridge_enable(struct drm_bridge *bridge)
+{
+	struct drm_bridge_init_priv *priv =
+		container_of(bridge, struct drm_bridge_init_priv, bridge);
+
+	priv->enable_count++;
+}
+
+static void drm_test_bridge_disable(struct drm_bridge *bridge)
+{
+	struct drm_bridge_init_priv *priv =
+		container_of(bridge, struct drm_bridge_init_priv, bridge);
+
+	priv->disable_count++;
+}
+
+static const struct drm_bridge_funcs drm_test_bridge_legacy_funcs = {
+	.enable			= drm_test_bridge_enable,
+	.disable		= drm_test_bridge_disable,
+};
+
+static void drm_test_bridge_atomic_enable(struct drm_bridge *bridge,
+					  struct drm_atomic_state *state)
+{
+	struct drm_bridge_init_priv *priv =
+		container_of(bridge, struct drm_bridge_init_priv, bridge);
+
+	priv->enable_count++;
+}
+
+static void drm_test_bridge_atomic_disable(struct drm_bridge *bridge,
+					   struct drm_atomic_state *state)
+{
+	struct drm_bridge_init_priv *priv =
+		container_of(bridge, struct drm_bridge_init_priv, bridge);
+
+	priv->disable_count++;
+}
+
+static const struct drm_bridge_funcs drm_test_bridge_atomic_funcs = {
+	.atomic_enable		= drm_test_bridge_atomic_enable,
+	.atomic_disable		= drm_test_bridge_atomic_disable,
+	.atomic_destroy_state	= drm_atomic_helper_bridge_destroy_state,
+	.atomic_duplicate_state	= drm_atomic_helper_bridge_duplicate_state,
+	.atomic_reset		= drm_atomic_helper_bridge_reset,
+};
+
+KUNIT_DEFINE_ACTION_WRAPPER(drm_bridge_remove_wrapper,
+			    drm_bridge_remove,
+			    struct drm_bridge *);
+
+static int drm_kunit_bridge_add(struct kunit *test,
+				struct drm_bridge *bridge)
+{
+	drm_bridge_add(bridge);
+
+	return kunit_add_action_or_reset(test,
+					 drm_bridge_remove_wrapper,
+					 bridge);
+}
+
+static struct drm_bridge_init_priv *
+drm_test_bridge_init(struct kunit *test, const struct drm_bridge_funcs *funcs)
+{
+	struct drm_bridge_init_priv *priv;
+	struct drm_encoder *enc;
+	struct drm_bridge *bridge;
+	struct drm_device *drm;
+	struct device *dev;
+	int ret;
+
+	dev = drm_kunit_helper_alloc_device(test);
+	if (IS_ERR(dev))
+		return ERR_CAST(dev);
+
+	priv = drm_kunit_helper_alloc_drm_device(test, dev,
+						 struct drm_bridge_init_priv, drm,
+						 DRIVER_MODESET | DRIVER_ATOMIC);
+	if (IS_ERR(priv))
+		return ERR_CAST(priv);
+
+	drm = &priv->drm;
+	priv->plane = drm_kunit_helper_create_primary_plane(test, drm,
+							    NULL,
+							    NULL,
+							    NULL, 0,
+							    NULL);
+	if (IS_ERR(priv->plane))
+		return ERR_CAST(priv->plane);
+
+	priv->crtc = drm_kunit_helper_create_crtc(test, drm,
+						  priv->plane, NULL,
+						  NULL,
+						  NULL);
+	if (IS_ERR(priv->crtc))
+		return ERR_CAST(priv->crtc);
+
+	enc = &priv->encoder;
+	ret = drmm_encoder_init(drm, enc, NULL, DRM_MODE_ENCODER_TMDS, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+
+	enc->possible_crtcs = drm_crtc_mask(priv->crtc);
+
+	bridge = &priv->bridge;
+	bridge->type = DRM_MODE_CONNECTOR_VIRTUAL;
+	bridge->funcs = funcs;
+
+	ret = drm_kunit_bridge_add(test, bridge);
+	if (ret)
+		return ERR_PTR(ret);
+
+	ret = drm_bridge_attach(enc, bridge, NULL, 0);
+	if (ret)
+		return ERR_PTR(ret);
+
+	priv->connector = drm_bridge_connector_init(drm, enc);
+	if (IS_ERR(priv->connector))
+		return ERR_CAST(priv->connector);
+
+	drm_connector_attach_encoder(priv->connector, enc);
+
+	drm_mode_config_reset(drm);
+
+	return priv;
+}
+
+/*
+ * Test that drm_bridge_get_current_state() returns the last committed
+ * state for an atomic bridge.
+ */
+static void drm_test_drm_bridge_get_current_state_atomic(struct kunit *test)
+{
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_bridge_init_priv *priv;
+	struct drm_bridge_state *curr_bridge_state;
+	struct drm_bridge_state *bridge_state;
+	struct drm_atomic_state *state;
+	struct drm_bridge *bridge;
+	struct drm_device *drm;
+	int ret;
+
+	priv = drm_test_bridge_init(test, &drm_test_bridge_atomic_funcs);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+	drm = &priv->drm;
+	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
+
+retry_commit:
+	bridge = &priv->bridge;
+	bridge_state = drm_atomic_get_bridge_state(state, bridge);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, bridge_state);
+
+	ret = drm_atomic_commit(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		drm_modeset_backoff(&ctx);
+		goto retry_commit;
+	}
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_state:
+	ret = drm_modeset_lock(&bridge->base.lock, &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_state;
+	}
+
+	curr_bridge_state = drm_bridge_get_current_state(bridge);
+	KUNIT_EXPECT_PTR_EQ(test, curr_bridge_state, bridge_state);
+
+	drm_modeset_unlock(&bridge->base.lock);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+}
+
+/*
+ * Test that drm_bridge_get_current_state() returns NULL for a
+ * non-atomic bridge.
+ */
+static void drm_test_drm_bridge_get_current_state_legacy(struct kunit *test)
+{
+	struct drm_bridge_init_priv *priv;
+	struct drm_bridge *bridge;
+
+	priv = drm_test_bridge_init(test, &drm_test_bridge_legacy_funcs);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv);
+
+	/*
+	 * NOTE: Strictly speaking, we should take the bridge->base.lock
+	 * before calling that function. However, bridge->base is only
+	 * initialized if the bridge is atomic, while we explicitly
+	 * initialize one that isn't there.
+	 *
+	 * In order to avoid unnecessary warnings, let's skip the
+	 * locking. The function would return NULL in all cases anyway,
+	 * so we don't really have any concurrency to worry about.
+	 */
+	bridge = &priv->bridge;
+	KUNIT_EXPECT_NULL(test, drm_bridge_get_current_state(bridge));
+}
+
+static struct kunit_case drm_bridge_get_current_state_tests[] = {
+	KUNIT_CASE(drm_test_drm_bridge_get_current_state_atomic),
+	KUNIT_CASE(drm_test_drm_bridge_get_current_state_legacy),
+	{ }
+};
+
+
+static struct kunit_suite drm_bridge_get_current_state_test_suite = {
+	.name = "drm_test_bridge_get_current_state",
+	.test_cases = drm_bridge_get_current_state_tests,
+};
+
+/*
+ * Test that an atomic bridge is properly power-cycled when calling
+ * drm_bridge_helper_reset_crtc().
+ */
+static void drm_test_drm_bridge_helper_reset_crtc_atomic(struct kunit *test)
+{
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_bridge_init_priv *priv;
+	struct drm_display_mode *mode;
+	struct drm_bridge *bridge;
+	int ret;
+
+	priv = drm_test_bridge_init(test, &drm_test_bridge_atomic_funcs);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv);
+
+	mode = drm_kunit_display_mode_from_cea_vic(test, &priv->drm, 16);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, mode);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_commit:
+	ret = drm_kunit_helper_enable_crtc_connector(test,
+						     &priv->drm, priv->crtc,
+						     priv->connector,
+						     mode,
+						     &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_commit;
+	}
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	bridge = &priv->bridge;
+	KUNIT_ASSERT_EQ(test, priv->enable_count, 1);
+	KUNIT_ASSERT_EQ(test, priv->disable_count, 0);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_reset:
+	ret = drm_bridge_helper_reset_crtc(bridge, &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_reset;
+	}
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	KUNIT_EXPECT_EQ(test, priv->enable_count, 2);
+	KUNIT_EXPECT_EQ(test, priv->disable_count, 1);
+}
+
+/*
+ * Test that calling drm_bridge_helper_reset_crtc() on a disabled atomic
+ * bridge will fail and not call the enable / disable callbacks
+ */
+static void drm_test_drm_bridge_helper_reset_crtc_atomic_disabled(struct kunit *test)
+{
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_bridge_init_priv *priv;
+	struct drm_display_mode *mode;
+	struct drm_bridge *bridge;
+	int ret;
+
+	priv = drm_test_bridge_init(test, &drm_test_bridge_atomic_funcs);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv);
+
+	mode = drm_kunit_display_mode_from_cea_vic(test, &priv->drm, 16);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, mode);
+
+	bridge = &priv->bridge;
+	KUNIT_ASSERT_EQ(test, priv->enable_count, 0);
+	KUNIT_ASSERT_EQ(test, priv->disable_count, 0);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_reset:
+	ret = drm_bridge_helper_reset_crtc(bridge, &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_reset;
+	}
+	KUNIT_EXPECT_LT(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	KUNIT_EXPECT_EQ(test, priv->enable_count, 0);
+	KUNIT_EXPECT_EQ(test, priv->disable_count, 0);
+}
+
+/*
+ * Test that a non-atomic bridge is properly power-cycled when calling
+ * drm_bridge_helper_reset_crtc().
+ */
+static void drm_test_drm_bridge_helper_reset_crtc_legacy(struct kunit *test)
+{
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_bridge_init_priv *priv;
+	struct drm_display_mode *mode;
+	struct drm_bridge *bridge;
+	int ret;
+
+	priv = drm_test_bridge_init(test, &drm_test_bridge_legacy_funcs);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv);
+
+	mode = drm_kunit_display_mode_from_cea_vic(test, &priv->drm, 16);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, mode);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_commit:
+	ret = drm_kunit_helper_enable_crtc_connector(test,
+						     &priv->drm, priv->crtc,
+						     priv->connector,
+						     mode,
+						     &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_commit;
+	}
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	bridge = &priv->bridge;
+	KUNIT_ASSERT_EQ(test, priv->enable_count, 1);
+	KUNIT_ASSERT_EQ(test, priv->disable_count, 0);
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+retry_reset:
+	ret = drm_bridge_helper_reset_crtc(bridge, &ctx);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry_reset;
+	}
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	KUNIT_EXPECT_EQ(test, priv->enable_count, 2);
+	KUNIT_EXPECT_EQ(test, priv->disable_count, 1);
+}
+
+static struct kunit_case drm_bridge_helper_reset_crtc_tests[] = {
+	KUNIT_CASE(drm_test_drm_bridge_helper_reset_crtc_atomic),
+	KUNIT_CASE(drm_test_drm_bridge_helper_reset_crtc_atomic_disabled),
+	KUNIT_CASE(drm_test_drm_bridge_helper_reset_crtc_legacy),
+	{ }
+};
+
+static struct kunit_suite drm_bridge_helper_reset_crtc_test_suite = {
+	.name = "drm_test_bridge_helper_reset_crtc",
+	.test_cases = drm_bridge_helper_reset_crtc_tests,
+};
+
+kunit_test_suites(
+	&drm_bridge_get_current_state_test_suite,
+	&drm_bridge_helper_reset_crtc_test_suite,
+);
+
+MODULE_AUTHOR("Maxime Ripard <mripard@kernel.org>");
+MODULE_DESCRIPTION("Kunit test for drm_bridge functions");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/tests/drm_client_modeset_test.c b/drivers/gpu/drm/tests/drm_client_modeset_test.c
index 7516f6cb36e4..3f44fe5e92e4 100644
--- a/drivers/gpu/drm/tests/drm_client_modeset_test.c
+++ b/drivers/gpu/drm/tests/drm_client_modeset_test.c
@@ -88,13 +88,17 @@ static void drm_test_pick_cmdline_res_1920_1080_60(struct kunit *test)
 	struct drm_device *drm = priv->drm;
 	struct drm_connector *connector = &priv->connector;
 	struct drm_cmdline_mode *cmdline_mode = &connector->cmdline_mode;
-	struct drm_display_mode *expected_mode, *mode;
+	struct drm_display_mode *expected_mode;
+	const struct drm_display_mode *mode;
 	const char *cmdline = "1920x1080@60";
 	int ret;
 
 	expected_mode = drm_mode_find_dmt(priv->drm, 1920, 1080, 60, false);
 	KUNIT_ASSERT_NOT_NULL(test, expected_mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, expected_mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_ASSERT_TRUE(test,
 			  drm_mode_parse_command_line_for_connector(cmdline,
 								    connector,
@@ -129,7 +133,8 @@ static void drm_test_pick_cmdline_named(struct kunit *test)
 	struct drm_device *drm = priv->drm;
 	struct drm_connector *connector = &priv->connector;
 	struct drm_cmdline_mode *cmdline_mode = &connector->cmdline_mode;
-	const struct drm_display_mode *expected_mode, *mode;
+	const struct drm_display_mode *mode;
+	struct drm_display_mode *expected_mode;
 	const char *cmdline = params->cmdline;
 	int ret;
 
@@ -149,6 +154,9 @@ static void drm_test_pick_cmdline_named(struct kunit *test)
 	expected_mode = params->func(drm);
 	KUNIT_ASSERT_NOT_NULL(test, expected_mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, expected_mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_EXPECT_TRUE(test, drm_mode_equal(expected_mode, mode));
 }
 
diff --git a/drivers/gpu/drm/tests/drm_cmdline_parser_test.c b/drivers/gpu/drm/tests/drm_cmdline_parser_test.c
index 59c8408c453c..1cfcb597b088 100644
--- a/drivers/gpu/drm/tests/drm_cmdline_parser_test.c
+++ b/drivers/gpu/drm/tests/drm_cmdline_parser_test.c
@@ -7,6 +7,7 @@
 #include <kunit/test.h>
 
 #include <drm/drm_connector.h>
+#include <drm/drm_kunit_helpers.h>
 #include <drm/drm_modes.h>
 
 static const struct drm_connector no_connector = {};
@@ -955,8 +956,15 @@ struct drm_cmdline_tv_option_test {
 static void drm_test_cmdline_tv_options(struct kunit *test)
 {
 	const struct drm_cmdline_tv_option_test *params = test->param_value;
-	const struct drm_display_mode *expected_mode = params->mode_fn(NULL);
+	struct drm_display_mode *expected_mode;
 	struct drm_cmdline_mode mode = { };
+	int ret;
+
+	expected_mode = params->mode_fn(NULL);
+	KUNIT_ASSERT_NOT_NULL(test, expected_mode);
+
+	ret = drm_kunit_add_mode_destroy_action(test, expected_mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	KUNIT_EXPECT_TRUE(test, drm_mode_parse_command_line_for_connector(params->cmdline,
 									  &no_connector, &mode));
diff --git a/drivers/gpu/drm/tests/drm_gem_shmem_test.c b/drivers/gpu/drm/tests/drm_gem_shmem_test.c
index fd4215e2f982..68f2c3162354 100644
--- a/drivers/gpu/drm/tests/drm_gem_shmem_test.c
+++ b/drivers/gpu/drm/tests/drm_gem_shmem_test.c
@@ -134,7 +134,7 @@ static void drm_gem_shmem_test_pin_pages(struct kunit *test)
 	shmem = drm_gem_shmem_create(drm_dev, TEST_SIZE);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, shmem);
 	KUNIT_EXPECT_NULL(test, shmem->pages);
-	KUNIT_EXPECT_EQ(test, shmem->pages_use_count, 0);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->pages_use_count), 0);
 
 	ret = kunit_add_action_or_reset(test, drm_gem_shmem_free_wrapper, shmem);
 	KUNIT_ASSERT_EQ(test, ret, 0);
@@ -142,14 +142,14 @@ static void drm_gem_shmem_test_pin_pages(struct kunit *test)
 	ret = drm_gem_shmem_pin(shmem);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 	KUNIT_ASSERT_NOT_NULL(test, shmem->pages);
-	KUNIT_EXPECT_EQ(test, shmem->pages_use_count, 1);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->pages_use_count), 1);
 
 	for (i = 0; i < (shmem->base.size >> PAGE_SHIFT); i++)
 		KUNIT_ASSERT_NOT_NULL(test, shmem->pages[i]);
 
 	drm_gem_shmem_unpin(shmem);
 	KUNIT_EXPECT_NULL(test, shmem->pages);
-	KUNIT_EXPECT_EQ(test, shmem->pages_use_count, 0);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->pages_use_count), 0);
 }
 
 /*
@@ -168,24 +168,24 @@ static void drm_gem_shmem_test_vmap(struct kunit *test)
 	shmem = drm_gem_shmem_create(drm_dev, TEST_SIZE);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, shmem);
 	KUNIT_EXPECT_NULL(test, shmem->vaddr);
-	KUNIT_EXPECT_EQ(test, shmem->vmap_use_count, 0);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->vmap_use_count), 0);
 
 	ret = kunit_add_action_or_reset(test, drm_gem_shmem_free_wrapper, shmem);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
-	ret = drm_gem_shmem_vmap(shmem, &map);
+	ret = drm_gem_shmem_vmap_locked(shmem, &map);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 	KUNIT_ASSERT_NOT_NULL(test, shmem->vaddr);
 	KUNIT_ASSERT_FALSE(test, iosys_map_is_null(&map));
-	KUNIT_EXPECT_EQ(test, shmem->vmap_use_count, 1);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->vmap_use_count), 1);
 
 	iosys_map_memset(&map, 0, TEST_BYTE, TEST_SIZE);
 	for (i = 0; i < TEST_SIZE; i++)
 		KUNIT_EXPECT_EQ(test, iosys_map_rd(&map, i, u8), TEST_BYTE);
 
-	drm_gem_shmem_vunmap(shmem, &map);
+	drm_gem_shmem_vunmap_locked(shmem, &map);
 	KUNIT_EXPECT_NULL(test, shmem->vaddr);
-	KUNIT_EXPECT_EQ(test, shmem->vmap_use_count, 0);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->vmap_use_count), 0);
 }
 
 /*
@@ -216,6 +216,9 @@ static void drm_gem_shmem_test_get_pages_sgt(struct kunit *test)
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sgt);
 	KUNIT_EXPECT_NULL(test, shmem->sgt);
 
+	ret = kunit_add_action_or_reset(test, kfree_wrapper, sgt);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	ret = kunit_add_action_or_reset(test, sg_free_table_wrapper, sgt);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
@@ -251,7 +254,7 @@ static void drm_gem_shmem_test_get_sg_table(struct kunit *test)
 	sgt = drm_gem_shmem_get_pages_sgt(shmem);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sgt);
 	KUNIT_ASSERT_NOT_NULL(test, shmem->pages);
-	KUNIT_EXPECT_EQ(test, shmem->pages_use_count, 1);
+	KUNIT_EXPECT_EQ(test, refcount_read(&shmem->pages_use_count), 1);
 	KUNIT_EXPECT_PTR_EQ(test, sgt, shmem->sgt);
 
 	for_each_sgtable_sg(sgt, sg, si) {
@@ -281,17 +284,17 @@ static void drm_gem_shmem_test_madvise(struct kunit *test)
 	ret = kunit_add_action_or_reset(test, drm_gem_shmem_free_wrapper, shmem);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
-	ret = drm_gem_shmem_madvise(shmem, 1);
+	ret = drm_gem_shmem_madvise_locked(shmem, 1);
 	KUNIT_EXPECT_TRUE(test, ret);
 	KUNIT_ASSERT_EQ(test, shmem->madv, 1);
 
 	/* Set madv to a negative value */
-	ret = drm_gem_shmem_madvise(shmem, -1);
+	ret = drm_gem_shmem_madvise_locked(shmem, -1);
 	KUNIT_EXPECT_FALSE(test, ret);
 	KUNIT_ASSERT_EQ(test, shmem->madv, -1);
 
 	/* Check that madv cannot be set back to a positive value */
-	ret = drm_gem_shmem_madvise(shmem, 0);
+	ret = drm_gem_shmem_madvise_locked(shmem, 0);
 	KUNIT_EXPECT_FALSE(test, ret);
 	KUNIT_ASSERT_EQ(test, shmem->madv, -1);
 }
@@ -319,7 +322,7 @@ static void drm_gem_shmem_test_purge(struct kunit *test)
 	ret = drm_gem_shmem_is_purgeable(shmem);
 	KUNIT_EXPECT_FALSE(test, ret);
 
-	ret = drm_gem_shmem_madvise(shmem, 1);
+	ret = drm_gem_shmem_madvise_locked(shmem, 1);
 	KUNIT_EXPECT_TRUE(test, ret);
 
 	/* The scatter/gather table will be freed by drm_gem_shmem_free */
@@ -329,7 +332,7 @@ static void drm_gem_shmem_test_purge(struct kunit *test)
 	ret = drm_gem_shmem_is_purgeable(shmem);
 	KUNIT_EXPECT_TRUE(test, ret);
 
-	drm_gem_shmem_purge(shmem);
+	drm_gem_shmem_purge_locked(shmem);
 	KUNIT_EXPECT_NULL(test, shmem->pages);
 	KUNIT_EXPECT_NULL(test, shmem->sgt);
 	KUNIT_EXPECT_EQ(test, shmem->madv, -1);
diff --git a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c
index e97efd3af9ed..7ffd666753b1 100644
--- a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c
+++ b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c
@@ -55,49 +55,6 @@ static struct drm_display_mode *find_preferred_mode(struct drm_connector *connec
 	return preferred;
 }
 
-static int light_up_connector(struct kunit *test,
-			      struct drm_device *drm,
-			      struct drm_crtc *crtc,
-			      struct drm_connector *connector,
-			      struct drm_display_mode *mode,
-			      struct drm_modeset_acquire_ctx *ctx)
-{
-	struct drm_atomic_state *state;
-	struct drm_connector_state *conn_state;
-	struct drm_crtc_state *crtc_state;
-	int ret;
-
-	state = drm_kunit_helper_atomic_state_alloc(test, drm, ctx);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
-
-retry:
-	conn_state = drm_atomic_get_connector_state(state, connector);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state);
-
-	ret = drm_atomic_set_crtc_for_connector(conn_state, crtc);
-	if (ret == -EDEADLK) {
-		drm_atomic_state_clear(state);
-		ret = drm_modeset_backoff(ctx);
-		if (!ret)
-			goto retry;
-	}
-	KUNIT_EXPECT_EQ(test, ret, 0);
-
-	crtc_state = drm_atomic_get_crtc_state(state, crtc);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_state);
-
-	ret = drm_atomic_set_mode_for_crtc(crtc_state, mode);
-	KUNIT_EXPECT_EQ(test, ret, 0);
-
-	crtc_state->enable = true;
-	crtc_state->active = true;
-
-	ret = drm_atomic_commit(state);
-	KUNIT_ASSERT_EQ(test, ret, 0);
-
-	return 0;
-}
-
 static int set_connector_edid(struct kunit *test, struct drm_connector *connector,
 			      const char *edid, size_t edid_len)
 {
@@ -298,7 +255,10 @@ static void drm_test_check_broadcast_rgb_crtc_mode_changed(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -364,7 +324,10 @@ static void drm_test_check_broadcast_rgb_crtc_mode_not_changed(struct kunit *tes
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -432,7 +395,10 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -489,7 +455,10 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode_vic_1(struct kunit *test)
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	crtc = priv->crtc;
-	ret = light_up_connector(test, drm, crtc, conn, mode, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     mode,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -547,7 +516,10 @@ static void drm_test_check_broadcast_rgb_full_cea_mode(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -606,7 +578,10 @@ static void drm_test_check_broadcast_rgb_full_cea_mode_vic_1(struct kunit *test)
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	crtc = priv->crtc;
-	ret = light_up_connector(test, drm, crtc, conn, mode, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     mode,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -666,7 +641,10 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -725,7 +703,10 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode_vic_1(struct kunit *te
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
 	crtc = priv->crtc;
-	ret = light_up_connector(test, drm, crtc, conn, mode, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     mode,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -789,7 +770,10 @@ static void drm_test_check_output_bpc_crtc_mode_changed(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -865,7 +849,10 @@ static void drm_test_check_output_bpc_crtc_mode_not_changed(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
@@ -941,7 +928,10 @@ static void drm_test_check_output_bpc_dvi(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -988,7 +978,10 @@ static void drm_test_check_tmds_char_rate_rgb_8bpc(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1037,7 +1030,10 @@ static void drm_test_check_tmds_char_rate_rgb_10bpc(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1086,7 +1082,10 @@ static void drm_test_check_tmds_char_rate_rgb_12bpc(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1134,7 +1133,10 @@ static void drm_test_check_hdmi_funcs_reject_rate(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	/* You shouldn't be doing that at home. */
@@ -1208,7 +1210,10 @@ static void drm_test_check_max_tmds_rate_bpc_fallback(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1282,7 +1287,10 @@ static void drm_test_check_max_tmds_rate_format_fallback(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1347,7 +1355,10 @@ static void drm_test_check_output_bpc_format_vic_1(struct kunit *test)
 	drm_modeset_acquire_init(&ctx, 0);
 
 	crtc = priv->crtc;
-	ret = light_up_connector(test, drm, crtc, conn, mode, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     mode,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1414,7 +1425,10 @@ static void drm_test_check_output_bpc_format_driver_rgb_only(struct kunit *test)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1483,7 +1497,10 @@ static void drm_test_check_output_bpc_format_display_rgb_only(struct kunit *test
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1543,7 +1560,10 @@ static void drm_test_check_output_bpc_format_driver_8bpc_only(struct kunit *test
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1605,7 +1625,10 @@ static void drm_test_check_output_bpc_format_display_8bpc_only(struct kunit *tes
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	conn_state = conn->state;
@@ -1645,7 +1668,10 @@ static void drm_test_check_disable_connector(struct kunit *test)
 
 	drm = &priv->drm;
 	crtc = priv->crtc;
-	ret = light_up_connector(test, drm, crtc, conn, preferred, &ctx);
+	ret = drm_kunit_helper_enable_crtc_connector(test, drm,
+						     crtc, conn,
+						     preferred,
+						     &ctx);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c
index a4eb68f0decc..5f7257840d8e 100644
--- a/drivers/gpu/drm/tests/drm_kunit_helpers.c
+++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c
@@ -2,6 +2,7 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic_uapi.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
@@ -271,6 +272,66 @@ drm_kunit_helper_create_crtc(struct kunit *test,
 }
 EXPORT_SYMBOL_GPL(drm_kunit_helper_create_crtc);
 
+/**
+ * drm_kunit_helper_enable_crtc_connector - Enables a CRTC -> Connector output
+ * @test: The test context object
+ * @drm: The device to alloc the plane for
+ * @crtc: The CRTC to enable
+ * @connector: The Connector to enable
+ * @mode: The display mode to configure the CRTC with
+ * @ctx: Locking context
+ *
+ * This function creates an atomic update to enable the route from @crtc
+ * to @connector, with the given @mode.
+ *
+ * Returns:
+ *
+ * A pointer to the new CRTC, or an ERR_PTR() otherwise. If the error
+ * returned is EDEADLK, the entire atomic sequence must be restarted.
+ */
+int drm_kunit_helper_enable_crtc_connector(struct kunit *test,
+					   struct drm_device *drm,
+					   struct drm_crtc *crtc,
+					   struct drm_connector *connector,
+					   const struct drm_display_mode *mode,
+					   struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_atomic_state *state;
+	struct drm_connector_state *conn_state;
+	struct drm_crtc_state *crtc_state;
+	int ret;
+
+	state = drm_kunit_helper_atomic_state_alloc(test, drm, ctx);
+	if (IS_ERR(state))
+		return PTR_ERR(state);
+
+	conn_state = drm_atomic_get_connector_state(state, connector);
+	if (IS_ERR(conn_state))
+		return PTR_ERR(conn_state);
+
+	ret = drm_atomic_set_crtc_for_connector(conn_state, crtc);
+	if (ret)
+		return ret;
+
+	crtc_state = drm_atomic_get_crtc_state(state, crtc);
+	if (IS_ERR(crtc_state))
+		return PTR_ERR(crtc_state);
+
+	ret = drm_atomic_set_mode_for_crtc(crtc_state, mode);
+	if (ret)
+		return ret;
+
+	crtc_state->enable = true;
+	crtc_state->active = true;
+
+	ret = drm_atomic_commit(state);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(drm_kunit_helper_enable_crtc_connector);
+
 static void kunit_action_drm_mode_destroy(void *ptr)
 {
 	struct drm_display_mode *mode = ptr;
@@ -279,6 +340,28 @@ static void kunit_action_drm_mode_destroy(void *ptr)
 }
 
 /**
+ * drm_kunit_add_mode_destroy_action() - Add a drm_destroy_mode kunit action
+ * @test: The test context object
+ * @mode: The drm_display_mode to destroy eventually
+ *
+ * Registers a kunit action that will destroy the drm_display_mode at
+ * the end of the test.
+ *
+ * If an error occurs, the drm_display_mode will be destroyed.
+ *
+ * Returns:
+ * 0 on success, an error code otherwise.
+ */
+int drm_kunit_add_mode_destroy_action(struct kunit *test,
+				      struct drm_display_mode *mode)
+{
+	return kunit_add_action_or_reset(test,
+					 kunit_action_drm_mode_destroy,
+					 mode);
+}
+EXPORT_SYMBOL_GPL(drm_kunit_add_mode_destroy_action);
+
+/**
  * drm_kunit_display_mode_from_cea_vic() - return a mode for CEA VIC for a KUnit test
  * @test: The test context object
  * @dev: DRM device
diff --git a/drivers/gpu/drm/tests/drm_modes_test.c b/drivers/gpu/drm/tests/drm_modes_test.c
index 6ed51f99e133..f5b20f92df8b 100644
--- a/drivers/gpu/drm/tests/drm_modes_test.c
+++ b/drivers/gpu/drm/tests/drm_modes_test.c
@@ -40,6 +40,7 @@ static void drm_test_modes_analog_tv_ntsc_480i(struct kunit *test)
 {
 	struct drm_test_modes_priv *priv = test->priv;
 	struct drm_display_mode *mode;
+	int ret;
 
 	mode = drm_analog_tv_mode(priv->drm,
 				  DRM_MODE_TV_MODE_NTSC,
@@ -47,6 +48,9 @@ static void drm_test_modes_analog_tv_ntsc_480i(struct kunit *test)
 				  true);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_EXPECT_EQ(test, drm_mode_vrefresh(mode), 60);
 	KUNIT_EXPECT_EQ(test, mode->hdisplay, 720);
 
@@ -70,6 +74,7 @@ static void drm_test_modes_analog_tv_ntsc_480i_inlined(struct kunit *test)
 {
 	struct drm_test_modes_priv *priv = test->priv;
 	struct drm_display_mode *expected, *mode;
+	int ret;
 
 	expected = drm_analog_tv_mode(priv->drm,
 				      DRM_MODE_TV_MODE_NTSC,
@@ -77,9 +82,15 @@ static void drm_test_modes_analog_tv_ntsc_480i_inlined(struct kunit *test)
 				      true);
 	KUNIT_ASSERT_NOT_NULL(test, expected);
 
+	ret = drm_kunit_add_mode_destroy_action(test, expected);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	mode = drm_mode_analog_ntsc_480i(priv->drm);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_EXPECT_TRUE(test, drm_mode_equal(expected, mode));
 }
 
@@ -87,6 +98,7 @@ static void drm_test_modes_analog_tv_pal_576i(struct kunit *test)
 {
 	struct drm_test_modes_priv *priv = test->priv;
 	struct drm_display_mode *mode;
+	int ret;
 
 	mode = drm_analog_tv_mode(priv->drm,
 				  DRM_MODE_TV_MODE_PAL,
@@ -94,6 +106,9 @@ static void drm_test_modes_analog_tv_pal_576i(struct kunit *test)
 				  true);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_EXPECT_EQ(test, drm_mode_vrefresh(mode), 50);
 	KUNIT_EXPECT_EQ(test, mode->hdisplay, 720);
 
@@ -117,6 +132,7 @@ static void drm_test_modes_analog_tv_pal_576i_inlined(struct kunit *test)
 {
 	struct drm_test_modes_priv *priv = test->priv;
 	struct drm_display_mode *expected, *mode;
+	int ret;
 
 	expected = drm_analog_tv_mode(priv->drm,
 				      DRM_MODE_TV_MODE_PAL,
@@ -124,9 +140,15 @@ static void drm_test_modes_analog_tv_pal_576i_inlined(struct kunit *test)
 				      true);
 	KUNIT_ASSERT_NOT_NULL(test, expected);
 
+	ret = drm_kunit_add_mode_destroy_action(test, expected);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	mode = drm_mode_analog_pal_576i(priv->drm);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_EXPECT_TRUE(test, drm_mode_equal(expected, mode));
 }
 
@@ -134,6 +156,7 @@ static void drm_test_modes_analog_tv_mono_576i(struct kunit *test)
 {
 	struct drm_test_modes_priv *priv = test->priv;
 	struct drm_display_mode *mode;
+	int ret;
 
 	mode = drm_analog_tv_mode(priv->drm,
 				  DRM_MODE_TV_MODE_MONOCHROME,
@@ -141,6 +164,9 @@ static void drm_test_modes_analog_tv_mono_576i(struct kunit *test)
 				  true);
 	KUNIT_ASSERT_NOT_NULL(test, mode);
 
+	ret = drm_kunit_add_mode_destroy_action(test, mode);
+	KUNIT_ASSERT_EQ(test, ret, 0);
+
 	KUNIT_EXPECT_EQ(test, drm_mode_vrefresh(mode), 50);
 	KUNIT_EXPECT_EQ(test, mode->hdisplay, 720);
 
diff --git a/drivers/gpu/drm/tests/drm_probe_helper_test.c b/drivers/gpu/drm/tests/drm_probe_helper_test.c
index bc09ff38aca1..db0e4f5df275 100644
--- a/drivers/gpu/drm/tests/drm_probe_helper_test.c
+++ b/drivers/gpu/drm/tests/drm_probe_helper_test.c
@@ -98,7 +98,7 @@ drm_test_connector_helper_tv_get_modes_check(struct kunit *test)
 	struct drm_connector *connector = &priv->connector;
 	struct drm_cmdline_mode *cmdline = &connector->cmdline_mode;
 	struct drm_display_mode *mode;
-	const struct drm_display_mode *expected;
+	struct drm_display_mode *expected;
 	size_t len;
 	int ret;
 
@@ -134,6 +134,9 @@ drm_test_connector_helper_tv_get_modes_check(struct kunit *test)
 
 		KUNIT_EXPECT_TRUE(test, drm_mode_equal(mode, expected));
 		KUNIT_EXPECT_TRUE(test, mode->type & DRM_MODE_TYPE_PREFERRED);
+
+		ret = drm_kunit_add_mode_destroy_action(test, expected);
+		KUNIT_ASSERT_EQ(test, ret, 0);
 	}
 
 	if (params->num_expected_modes >= 2) {
@@ -145,6 +148,9 @@ drm_test_connector_helper_tv_get_modes_check(struct kunit *test)
 
 		KUNIT_EXPECT_TRUE(test, drm_mode_equal(mode, expected));
 		KUNIT_EXPECT_FALSE(test, mode->type & DRM_MODE_TYPE_PREFERRED);
+
+		ret = drm_kunit_add_mode_destroy_action(test, expected);
+		KUNIT_ASSERT_EQ(test, ret, 0);
 	}
 
 	mutex_unlock(&priv->drm->mode_config.mutex);
diff --git a/drivers/gpu/drm/tidss/tidss_encoder.c b/drivers/gpu/drm/tidss/tidss_encoder.c
index 17a86bed8054..95b4aeff2775 100644
--- a/drivers/gpu/drm/tidss/tidss_encoder.c
+++ b/drivers/gpu/drm/tidss/tidss_encoder.c
@@ -34,11 +34,12 @@ static inline struct tidss_encoder
 }
 
 static int tidss_bridge_attach(struct drm_bridge *bridge,
+			       struct drm_encoder *encoder,
 			       enum drm_bridge_attach_flags flags)
 {
 	struct tidss_encoder *t_enc = bridge_to_tidss_encoder(bridge);
 
-	return drm_bridge_attach(bridge->encoder, t_enc->next_bridge,
+	return drm_bridge_attach(encoder, t_enc->next_bridge,
 				 bridge, flags);
 }
 
diff --git a/drivers/gpu/drm/tiny/Kconfig b/drivers/gpu/drm/tiny/Kconfig
index 54c84c9801c1..6d1b3e2cb3fb 100644
--- a/drivers/gpu/drm/tiny/Kconfig
+++ b/drivers/gpu/drm/tiny/Kconfig
@@ -3,6 +3,7 @@
 config DRM_APPLETBDRM
 	tristate "DRM support for Apple Touch Bars"
 	depends on DRM && USB && MMU
+	depends on X86 || COMPILE_TEST
 	select DRM_GEM_SHMEM_HELPER
 	select DRM_KMS_HELPER
 	help
@@ -65,20 +66,6 @@ config DRM_GM12U320
 	 This is a KMS driver for projectors which use the GM12U320 chipset
 	 for video transfer over USB2/3, such as the Acer C120 mini projector.
 
-config DRM_OFDRM
-	tristate "Open Firmware display driver"
-	depends on DRM && MMU && OF && (PPC || COMPILE_TEST)
-	select APERTURE_HELPERS
-	select DRM_CLIENT_SELECTION
-	select DRM_GEM_SHMEM_HELPER
-	select DRM_KMS_HELPER
-	help
-	  DRM driver for Open Firmware framebuffers.
-
-	  This driver assumes that the display hardware has been initialized
-	  by the Open Firmware before the kernel boots. Scanout buffer, size,
-	  and display format must be provided via device tree.
-
 config DRM_PANEL_MIPI_DBI
 	tristate "DRM support for MIPI DBI compatible panels"
 	depends on DRM && SPI
@@ -95,24 +82,6 @@ config DRM_PANEL_MIPI_DBI
 	  https://github.com/notro/panel-mipi-dbi/wiki.
 	  To compile this driver as a module, choose M here.
 
-config DRM_SIMPLEDRM
-	tristate "Simple framebuffer driver"
-	depends on DRM && MMU
-	select APERTURE_HELPERS
-	select DRM_CLIENT_SELECTION
-	select DRM_GEM_SHMEM_HELPER
-	select DRM_KMS_HELPER
-	help
-	  DRM driver for simple platform-provided framebuffers.
-
-	  This driver assumes that the display hardware has been initialized
-	  by the firmware or bootloader before the kernel boots. Scanout
-	  buffer, size, and display format must be provided via device tree,
-	  UEFI, VESA, etc.
-
-	  On x86 BIOS or UEFI systems, you should also select SYSFB_SIMPLEFB
-	  to use UEFI and VESA framebuffers.
-
 config TINYDRM_HX8357D
 	tristate "DRM support for HX8357D display panels"
 	depends on DRM && SPI
@@ -230,32 +199,3 @@ config TINYDRM_SHARP_MEMORY
 	  * 4.40" Sharp Memory LCD (LS044Q7DH01)
 
 	  If M is selected the module will be called sharp_memory.
-
-config TINYDRM_ST7586
-	tristate "DRM support for Sitronix ST7586 display panels"
-	depends on DRM && SPI
-	select DRM_CLIENT_SELECTION
-	select DRM_KMS_HELPER
-	select DRM_GEM_DMA_HELPER
-	select DRM_MIPI_DBI
-	help
-	  DRM driver for the following Sitronix ST7586 panels:
-	  * LEGO MINDSTORMS EV3
-
-	  If M is selected the module will be called st7586.
-
-config TINYDRM_ST7735R
-	tristate "DRM support for Sitronix ST7715R/ST7735R display panels"
-	depends on DRM && SPI
-	select DRM_CLIENT_SELECTION
-	select DRM_KMS_HELPER
-	select DRM_GEM_DMA_HELPER
-	select DRM_MIPI_DBI
-	select BACKLIGHT_CLASS_DEVICE
-	help
-	  DRM driver for Sitronix ST7715R/ST7735R with one of the following
-	  LCDs:
-	  * Jianda JD-T18003-T01 1.8" 128x160 TFT
-	  * Okaya RH128128T 1.44" 128x128 TFT
-
-	  If M is selected the module will be called st7735r.
diff --git a/drivers/gpu/drm/tiny/Makefile b/drivers/gpu/drm/tiny/Makefile
index 0a3a7837a58b..4a9ff61ec254 100644
--- a/drivers/gpu/drm/tiny/Makefile
+++ b/drivers/gpu/drm/tiny/Makefile
@@ -5,9 +5,7 @@ obj-$(CONFIG_DRM_ARCPGU)		+= arcpgu.o
 obj-$(CONFIG_DRM_BOCHS)			+= bochs.o
 obj-$(CONFIG_DRM_CIRRUS_QEMU)		+= cirrus-qemu.o
 obj-$(CONFIG_DRM_GM12U320)		+= gm12u320.o
-obj-$(CONFIG_DRM_OFDRM)			+= ofdrm.o
 obj-$(CONFIG_DRM_PANEL_MIPI_DBI)	+= panel-mipi-dbi.o
-obj-$(CONFIG_DRM_SIMPLEDRM)		+= simpledrm.o
 obj-$(CONFIG_TINYDRM_HX8357D)		+= hx8357d.o
 obj-$(CONFIG_TINYDRM_ILI9163)		+= ili9163.o
 obj-$(CONFIG_TINYDRM_ILI9225)		+= ili9225.o
@@ -16,5 +14,3 @@ obj-$(CONFIG_TINYDRM_ILI9486)		+= ili9486.o
 obj-$(CONFIG_TINYDRM_MI0283QT)		+= mi0283qt.o
 obj-$(CONFIG_TINYDRM_REPAPER)		+= repaper.o
 obj-$(CONFIG_TINYDRM_SHARP_MEMORY)	+= sharp-memory.o
-obj-$(CONFIG_TINYDRM_ST7586)		+= st7586.o
-obj-$(CONFIG_TINYDRM_ST7735R)		+= st7735r.o
diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
index 4370ba22dd88..751b05753c94 100644
--- a/drivers/gpu/drm/tiny/appletbdrm.c
+++ b/drivers/gpu/drm/tiny/appletbdrm.c
@@ -45,7 +45,7 @@
 #define APPLETBDRM_BULK_MSG_TIMEOUT	1000
 
 #define drm_to_adev(_drm)		container_of(_drm, struct appletbdrm_device, drm)
-#define adev_to_udev(adev)		interface_to_usbdev(to_usb_interface(adev->dmadev))
+#define adev_to_udev(adev)		interface_to_usbdev(to_usb_interface((adev)->drm.dev))
 
 struct appletbdrm_msg_request_header {
 	__le16 unk_00;
@@ -123,8 +123,6 @@ struct appletbdrm_fb_request_response {
 } __packed;
 
 struct appletbdrm_device {
-	struct device *dmadev;
-
 	unsigned int in_ep;
 	unsigned int out_ep;
 
@@ -214,7 +212,7 @@ retry:
 	}
 
 	if (response->msg != expected_response) {
-		drm_err(drm, "Unexpected response from device (expected %p4cc found %p4cc)\n",
+		drm_err(drm, "Unexpected response from device (expected %p4cl found %p4cl)\n",
 			&expected_response, &response->msg);
 		return -EIO;
 	}
@@ -288,7 +286,7 @@ static int appletbdrm_get_information(struct appletbdrm_device *adev)
 	}
 
 	if (pixel_format != APPLETBDRM_PIXEL_FORMAT) {
-		drm_err(drm, "Encountered unknown pixel format (%p4cc)\n", &pixel_format);
+		drm_err(drm, "Encountered unknown pixel format (%p4cl)\n", &pixel_format);
 		ret = -EINVAL;
 		goto free_info;
 	}
@@ -612,22 +610,10 @@ static const struct drm_encoder_funcs appletbdrm_encoder_funcs = {
 	.destroy = drm_encoder_cleanup,
 };
 
-static struct drm_gem_object *appletbdrm_driver_gem_prime_import(struct drm_device *dev,
-								 struct dma_buf *dma_buf)
-{
-	struct appletbdrm_device *adev = drm_to_adev(dev);
-
-	if (!adev->dmadev)
-		return ERR_PTR(-ENODEV);
-
-	return drm_gem_prime_import_dev(dev, dma_buf, adev->dmadev);
-}
-
 DEFINE_DRM_GEM_FOPS(appletbdrm_drm_fops);
 
 static const struct drm_driver appletbdrm_drm_driver = {
 	DRM_GEM_SHMEM_DRIVER_OPS,
-	.gem_prime_import	= appletbdrm_driver_gem_prime_import,
 	.name			= "appletbdrm",
 	.desc			= "Apple Touch Bar DRM Driver",
 	.major			= 1,
@@ -747,6 +733,7 @@ static int appletbdrm_probe(struct usb_interface *intf,
 	struct device *dev = &intf->dev;
 	struct appletbdrm_device *adev;
 	struct drm_device *drm = NULL;
+	struct device *dma_dev;
 	int ret;
 
 	ret = usb_find_common_endpoints(intf->cur_altsetting, &bulk_in, &bulk_out, NULL, NULL);
@@ -761,12 +748,19 @@ static int appletbdrm_probe(struct usb_interface *intf,
 
 	adev->in_ep = bulk_in->bEndpointAddress;
 	adev->out_ep = bulk_out->bEndpointAddress;
-	adev->dmadev = dev;
 
 	drm = &adev->drm;
 
 	usb_set_intfdata(intf, adev);
 
+	dma_dev = usb_intf_get_dma_device(intf);
+	if (dma_dev) {
+		drm_dev_set_dma_dev(drm, dma_dev);
+		put_device(dma_dev);
+	} else {
+		drm_warn(drm, "buffer sharing not supported"); /* not an error */
+	}
+
 	ret = appletbdrm_get_information(adev);
 	if (ret) {
 		drm_err(drm, "Failed to get display information\n");
diff --git a/drivers/gpu/drm/tiny/cirrus-qemu.c b/drivers/gpu/drm/tiny/cirrus-qemu.c
index 52ec1e4ea9e5..97a93adc5669 100644
--- a/drivers/gpu/drm/tiny/cirrus-qemu.c
+++ b/drivers/gpu/drm/tiny/cirrus-qemu.c
@@ -70,20 +70,6 @@ struct cirrus_device {
 
 #define to_cirrus(_dev) container_of(_dev, struct cirrus_device, dev)
 
-struct cirrus_primary_plane_state {
-	struct drm_shadow_plane_state base;
-
-	/* HW scanout buffer */
-	const struct drm_format_info   *format;
-	unsigned int		       pitch;
-};
-
-static inline struct cirrus_primary_plane_state *
-to_cirrus_primary_plane_state(struct drm_plane_state *plane_state)
-{
-	return container_of(plane_state, struct cirrus_primary_plane_state, base.base);
-};
-
 /* ------------------------------------------------------------------ */
 /*
  * The meat of this driver. The core passes us a mode and we have to program
@@ -144,37 +130,6 @@ static void wreg_hdr(struct cirrus_device *cirrus, u8 val)
 	iowrite8(val, cirrus->mmio + VGA_DAC_MASK);
 }
 
-static const struct drm_format_info *cirrus_convert_to(struct drm_framebuffer *fb)
-{
-	if (fb->format->format == DRM_FORMAT_XRGB8888 && fb->pitches[0] > CIRRUS_MAX_PITCH) {
-		if (fb->width * 3 <= CIRRUS_MAX_PITCH)
-			/* convert from XR24 to RG24 */
-			return drm_format_info(DRM_FORMAT_RGB888);
-		else
-			/* convert from XR24 to RG16 */
-			return drm_format_info(DRM_FORMAT_RGB565);
-	}
-	return NULL;
-}
-
-static const struct drm_format_info *cirrus_format(struct drm_framebuffer *fb)
-{
-	const struct drm_format_info *format = cirrus_convert_to(fb);
-
-	if (format)
-		return format;
-	return fb->format;
-}
-
-static int cirrus_pitch(struct drm_framebuffer *fb)
-{
-	const struct drm_format_info *format = cirrus_convert_to(fb);
-
-	if (format)
-		return drm_format_info_min_pitch(format, 0, fb->width);
-	return fb->pitches[0];
-}
-
 static void cirrus_set_start_address(struct cirrus_device *cirrus, u32 offset)
 {
 	u32 addr;
@@ -318,7 +273,6 @@ static void cirrus_pitch_set(struct cirrus_device *cirrus, unsigned int pitch)
 	/* Enable extended blanking and pitch bits, and enable full memory */
 	cr1b = 0x22;
 	cr1b |= (pitch >> 7) & 0x10;
-	cr1b |= (pitch >> 6) & 0x40;
 	wreg_crt(cirrus, 0x1b, cr1b);
 
 	cirrus_set_start_address(cirrus, 0);
@@ -342,13 +296,10 @@ static int cirrus_primary_plane_helper_atomic_check(struct drm_plane *plane,
 						    struct drm_atomic_state *state)
 {
 	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(state, plane);
-	struct cirrus_primary_plane_state *new_primary_plane_state =
-		to_cirrus_primary_plane_state(new_plane_state);
 	struct drm_framebuffer *fb = new_plane_state->fb;
 	struct drm_crtc *new_crtc = new_plane_state->crtc;
 	struct drm_crtc_state *new_crtc_state = NULL;
 	int ret;
-	unsigned int pitch;
 
 	if (new_crtc)
 		new_crtc_state = drm_atomic_get_new_crtc_state(state, new_crtc);
@@ -362,17 +313,12 @@ static int cirrus_primary_plane_helper_atomic_check(struct drm_plane *plane,
 	else if (!new_plane_state->visible)
 		return 0;
 
-	pitch = cirrus_pitch(fb);
-
 	/* validate size constraints */
-	if (pitch > CIRRUS_MAX_PITCH)
+	if (fb->pitches[0] > CIRRUS_MAX_PITCH)
 		return -EINVAL;
-	else if (pitch * fb->height > CIRRUS_VRAM_SIZE)
+	else if (fb->pitches[0] > CIRRUS_VRAM_SIZE / fb->height)
 		return -EINVAL;
 
-	new_primary_plane_state->format = cirrus_format(fb);
-	new_primary_plane_state->pitch = pitch;
-
 	return 0;
 }
 
@@ -381,15 +327,10 @@ static void cirrus_primary_plane_helper_atomic_update(struct drm_plane *plane,
 {
 	struct cirrus_device *cirrus = to_cirrus(plane->dev);
 	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
-	struct cirrus_primary_plane_state *primary_plane_state =
-		to_cirrus_primary_plane_state(plane_state);
 	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
 	struct drm_framebuffer *fb = plane_state->fb;
-	const struct drm_format_info *format = primary_plane_state->format;
-	unsigned int pitch = primary_plane_state->pitch;
 	struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, plane);
-	struct cirrus_primary_plane_state *old_primary_plane_state =
-		to_cirrus_primary_plane_state(old_plane_state);
+	struct drm_framebuffer *old_fb = old_plane_state->fb;
 	struct iosys_map vaddr = IOSYS_MAP_INIT_VADDR_IOMEM(cirrus->vram);
 	struct drm_atomic_helper_damage_iter iter;
 	struct drm_rect damage;
@@ -401,18 +342,17 @@ static void cirrus_primary_plane_helper_atomic_update(struct drm_plane *plane,
 	if (!drm_dev_enter(&cirrus->dev, &idx))
 		return;
 
-	if (old_primary_plane_state->format != format)
-		cirrus_format_set(cirrus, format);
-	if (old_primary_plane_state->pitch != pitch)
-		cirrus_pitch_set(cirrus, pitch);
+	if (!old_fb || old_fb->format != fb->format)
+		cirrus_format_set(cirrus, fb->format);
+	if (!old_fb || old_fb->pitches[0] != fb->pitches[0])
+		cirrus_pitch_set(cirrus, fb->pitches[0]);
 
 	drm_atomic_helper_damage_iter_init(&iter, old_plane_state, plane_state);
 	drm_atomic_for_each_plane_damage(&iter, &damage) {
-		unsigned int offset = drm_fb_clip_offset(pitch, format, &damage);
+		unsigned int offset = drm_fb_clip_offset(fb->pitches[0], fb->format, &damage);
 		struct iosys_map dst = IOSYS_MAP_INIT_OFFSET(&vaddr, offset);
 
-		drm_fb_blit(&dst, &pitch, format->format, shadow_plane_state->data, fb,
-			    &damage, &shadow_plane_state->fmtcnv_state);
+		drm_fb_memcpy(&dst, fb->pitches, shadow_plane_state->data, fb, &damage);
 	}
 
 	drm_dev_exit(idx);
@@ -424,62 +364,11 @@ static const struct drm_plane_helper_funcs cirrus_primary_plane_helper_funcs = {
 	.atomic_update = cirrus_primary_plane_helper_atomic_update,
 };
 
-static struct drm_plane_state *
-cirrus_primary_plane_atomic_duplicate_state(struct drm_plane *plane)
-{
-	struct drm_plane_state *plane_state = plane->state;
-	struct cirrus_primary_plane_state *primary_plane_state =
-		to_cirrus_primary_plane_state(plane_state);
-	struct cirrus_primary_plane_state *new_primary_plane_state;
-	struct drm_shadow_plane_state *new_shadow_plane_state;
-
-	if (!plane_state)
-		return NULL;
-
-	new_primary_plane_state = kzalloc(sizeof(*new_primary_plane_state), GFP_KERNEL);
-	if (!new_primary_plane_state)
-		return NULL;
-	new_shadow_plane_state = &new_primary_plane_state->base;
-
-	__drm_gem_duplicate_shadow_plane_state(plane, new_shadow_plane_state);
-	new_primary_plane_state->format = primary_plane_state->format;
-	new_primary_plane_state->pitch = primary_plane_state->pitch;
-
-	return &new_shadow_plane_state->base;
-}
-
-static void cirrus_primary_plane_atomic_destroy_state(struct drm_plane *plane,
-						      struct drm_plane_state *plane_state)
-{
-	struct cirrus_primary_plane_state *primary_plane_state =
-		to_cirrus_primary_plane_state(plane_state);
-
-	__drm_gem_destroy_shadow_plane_state(&primary_plane_state->base);
-	kfree(primary_plane_state);
-}
-
-static void cirrus_reset_primary_plane(struct drm_plane *plane)
-{
-	struct cirrus_primary_plane_state *primary_plane_state;
-
-	if (plane->state) {
-		cirrus_primary_plane_atomic_destroy_state(plane, plane->state);
-		plane->state = NULL; /* must be set to NULL here */
-	}
-
-	primary_plane_state = kzalloc(sizeof(*primary_plane_state), GFP_KERNEL);
-	if (!primary_plane_state)
-		return;
-	__drm_gem_reset_shadow_plane(plane, &primary_plane_state->base);
-}
-
 static const struct drm_plane_funcs cirrus_primary_plane_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
 	.destroy = drm_plane_cleanup,
-	.reset = cirrus_reset_primary_plane,
-	.atomic_duplicate_state = cirrus_primary_plane_atomic_duplicate_state,
-	.atomic_destroy_state = cirrus_primary_plane_atomic_destroy_state,
+	DRM_GEM_SHADOW_PLANE_FUNCS,
 };
 
 static int cirrus_crtc_helper_atomic_check(struct drm_crtc *crtc, struct drm_atomic_state *state)
@@ -614,9 +503,17 @@ static enum drm_mode_status cirrus_mode_config_mode_valid(struct drm_device *dev
 							  const struct drm_display_mode *mode)
 {
 	const struct drm_format_info *format = drm_format_info(DRM_FORMAT_XRGB8888);
-	uint64_t pitch = drm_format_info_min_pitch(format, 0, mode->hdisplay);
+	u64 pitch;
 
-	if (pitch * mode->vdisplay > CIRRUS_VRAM_SIZE)
+	if (drm_WARN_ON_ONCE(dev, !format))
+		return MODE_ERROR; /* driver bug */
+
+	pitch = drm_format_info_min_pitch(format, 0, mode->hdisplay);
+	if (!pitch)
+		return MODE_BAD_WIDTH;
+	if (pitch > CIRRUS_MAX_PITCH)
+		return MODE_BAD_WIDTH; /* maximum programmable pitch */
+	if (pitch > CIRRUS_VRAM_SIZE / mode->vdisplay)
 		return MODE_MEM;
 
 	return MODE_OK;
@@ -681,7 +578,7 @@ static int cirrus_pci_probe(struct pci_dev *pdev,
 	if (ret)
 		return ret;
 
-	ret = pci_request_regions(pdev, DRIVER_NAME);
+	ret = pcim_request_all_regions(pdev, DRIVER_NAME);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/tiny/gm12u320.c b/drivers/gpu/drm/tiny/gm12u320.c
index 41e9bfb2e2ff..fb0004166f4a 100644
--- a/drivers/gpu/drm/tiny/gm12u320.c
+++ b/drivers/gpu/drm/tiny/gm12u320.c
@@ -86,7 +86,6 @@ MODULE_PARM_DESC(eco_mode, "Turn on Eco mode (less bright, more silent)");
 
 struct gm12u320_device {
 	struct drm_device	         dev;
-	struct device                   *dmadev;
 	struct drm_simple_display_pipe   pipe;
 	struct drm_connector	         conn;
 	unsigned char                   *cmd_buf;
@@ -602,22 +601,6 @@ static const uint64_t gm12u320_pipe_modifiers[] = {
 	DRM_FORMAT_MOD_INVALID
 };
 
-/*
- * FIXME: Dma-buf sharing requires DMA support by the importing device.
- *        This function is a workaround to make USB devices work as well.
- *        See todo.rst for how to fix the issue in the dma-buf framework.
- */
-static struct drm_gem_object *gm12u320_gem_prime_import(struct drm_device *dev,
-							struct dma_buf *dma_buf)
-{
-	struct gm12u320_device *gm12u320 = to_gm12u320(dev);
-
-	if (!gm12u320->dmadev)
-		return ERR_PTR(-ENODEV);
-
-	return drm_gem_prime_import_dev(dev, dma_buf, gm12u320->dmadev);
-}
-
 DEFINE_DRM_GEM_FOPS(gm12u320_fops);
 
 static const struct drm_driver gm12u320_drm_driver = {
@@ -630,7 +613,6 @@ static const struct drm_driver gm12u320_drm_driver = {
 
 	.fops		 = &gm12u320_fops,
 	DRM_GEM_SHMEM_DRIVER_OPS,
-	.gem_prime_import = gm12u320_gem_prime_import,
 	DRM_FBDEV_SHMEM_DRIVER_OPS,
 };
 
@@ -645,6 +627,7 @@ static int gm12u320_usb_probe(struct usb_interface *interface,
 {
 	struct gm12u320_device *gm12u320;
 	struct drm_device *dev;
+	struct device *dma_dev;
 	int ret;
 
 	/*
@@ -660,16 +643,20 @@ static int gm12u320_usb_probe(struct usb_interface *interface,
 		return PTR_ERR(gm12u320);
 	dev = &gm12u320->dev;
 
-	gm12u320->dmadev = usb_intf_get_dma_device(to_usb_interface(dev->dev));
-	if (!gm12u320->dmadev)
+	dma_dev = usb_intf_get_dma_device(interface);
+	if (dma_dev) {
+		drm_dev_set_dma_dev(dev, dma_dev);
+		put_device(dma_dev);
+	} else {
 		drm_warn(dev, "buffer sharing not supported"); /* not an error */
+	}
 
 	INIT_DELAYED_WORK(&gm12u320->fb_update.work, gm12u320_fb_update_work);
 	mutex_init(&gm12u320->fb_update.lock);
 
 	ret = drmm_mode_config_init(dev);
 	if (ret)
-		goto err_put_device;
+		return ret;
 
 	dev->mode_config.min_width = GM12U320_USER_WIDTH;
 	dev->mode_config.max_width = GM12U320_USER_WIDTH;
@@ -679,15 +666,15 @@ static int gm12u320_usb_probe(struct usb_interface *interface,
 
 	ret = gm12u320_usb_alloc(gm12u320);
 	if (ret)
-		goto err_put_device;
+		return ret;
 
 	ret = gm12u320_set_ecomode(gm12u320);
 	if (ret)
-		goto err_put_device;
+		return ret;
 
 	ret = gm12u320_conn_init(gm12u320);
 	if (ret)
-		goto err_put_device;
+		return ret;
 
 	ret = drm_simple_display_pipe_init(&gm12u320->dev,
 					   &gm12u320->pipe,
@@ -697,31 +684,24 @@ static int gm12u320_usb_probe(struct usb_interface *interface,
 					   gm12u320_pipe_modifiers,
 					   &gm12u320->conn);
 	if (ret)
-		goto err_put_device;
+		return ret;
 
 	drm_mode_config_reset(dev);
 
 	usb_set_intfdata(interface, dev);
 	ret = drm_dev_register(dev, 0);
 	if (ret)
-		goto err_put_device;
+		return ret;
 
 	drm_client_setup(dev, NULL);
 
 	return 0;
-
-err_put_device:
-	put_device(gm12u320->dmadev);
-	return ret;
 }
 
 static void gm12u320_usb_disconnect(struct usb_interface *interface)
 {
 	struct drm_device *dev = usb_get_intfdata(interface);
-	struct gm12u320_device *gm12u320 = to_gm12u320(dev);
 
-	put_device(gm12u320->dmadev);
-	gm12u320->dmadev = NULL;
 	drm_dev_unplug(dev);
 	drm_atomic_helper_shutdown(dev);
 }
diff --git a/drivers/gpu/drm/ttm/tests/ttm_bo_test.c b/drivers/gpu/drm/ttm/tests/ttm_bo_test.c
index f8f20d2f6174..e08e5a138420 100644
--- a/drivers/gpu/drm/ttm/tests/ttm_bo_test.c
+++ b/drivers/gpu/drm/ttm/tests/ttm_bo_test.c
@@ -340,7 +340,7 @@ static void ttm_bo_unreserve_bulk(struct kunit *test)
 	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
 
 	resv = kunit_kzalloc(test, sizeof(*resv), GFP_KERNEL);
-	KUNIT_ASSERT_NOT_NULL(test, ttm_dev);
+	KUNIT_ASSERT_NOT_NULL(test, resv);
 
 	err = ttm_device_kunit_init(priv, ttm_dev, false, false);
 	KUNIT_ASSERT_EQ(test, err, 0);
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 95b86003c50d..08a23ab037cb 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -235,7 +235,7 @@ static void ttm_bo_delayed_delete(struct work_struct *work)
 
 	bo = container_of(work, typeof(*bo), delayed_delete);
 
-	dma_resv_wait_timeout(bo->base.resv, DMA_RESV_USAGE_BOOKKEEP, false,
+	dma_resv_wait_timeout(&bo->base._resv, DMA_RESV_USAGE_BOOKKEEP, false,
 			      MAX_SCHEDULE_TIMEOUT);
 	dma_resv_lock(bo->base.resv, NULL);
 	ttm_bo_cleanup_memtype_use(bo);
@@ -270,7 +270,7 @@ static void ttm_bo_release(struct kref *kref)
 		drm_vma_offset_remove(bdev->vma_manager, &bo->base.vma_node);
 		ttm_mem_io_free(bdev, bo->resource);
 
-		if (!dma_resv_test_signaled(bo->base.resv,
+		if (!dma_resv_test_signaled(&bo->base._resv,
 					    DMA_RESV_USAGE_BOOKKEEP) ||
 		    (want_init_on_free() && (bo->ttm != NULL)) ||
 		    bo->type == ttm_bo_type_sg ||
@@ -1093,7 +1093,8 @@ struct ttm_bo_swapout_walk {
 	struct ttm_lru_walk walk;
 	/** @gfp_flags: The gfp flags to use for ttm_tt_swapout() */
 	gfp_t gfp_flags;
-
+	/** @hit_low: Whether we should attempt to swap BO's with low watermark threshold */
+	/** @evict_low: If we cannot swap a bo when @try_low is false (first pass) */
 	bool hit_low, evict_low;
 };
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index a194db83421d..bdfa6ecfef05 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -220,7 +220,6 @@ vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
 		struct ttm_operation_ctx ctx = {
 			.interruptible = true,
 			.no_wait_gpu = false,
-			.force_alloc = true
 		};
 
 		ttm = bo->ttm;
diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c
index 7e5a60c55813..769b0ca9be47 100644
--- a/drivers/gpu/drm/ttm/ttm_resource.c
+++ b/drivers/gpu/drm/ttm/ttm_resource.c
@@ -548,7 +548,6 @@ int ttm_resource_manager_evict_all(struct ttm_device *bdev,
 	struct ttm_operation_ctx ctx = {
 		.interruptible = false,
 		.no_wait_gpu = false,
-		.force_alloc = true
 	};
 	struct dma_fence *fence;
 	int ret;
diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c
index 05b3a152cc33..1922988625eb 100644
--- a/drivers/gpu/drm/udl/udl_drv.c
+++ b/drivers/gpu/drm/udl/udl_drv.c
@@ -22,13 +22,14 @@ static int udl_usb_suspend(struct usb_interface *interface,
 			   pm_message_t message)
 {
 	struct drm_device *dev = usb_get_intfdata(interface);
+	struct udl_device *udl = to_udl(dev);
 	int ret;
 
 	ret = drm_mode_config_helper_suspend(dev);
 	if (ret)
 		return ret;
 
-	udl_sync_pending_urbs(dev);
+	udl_sync_pending_urbs(udl);
 	return 0;
 }
 
@@ -49,22 +50,6 @@ static int udl_usb_reset_resume(struct usb_interface *interface)
 	return drm_mode_config_helper_resume(dev);
 }
 
-/*
- * FIXME: Dma-buf sharing requires DMA support by the importing device.
- *        This function is a workaround to make USB devices work as well.
- *        See todo.rst for how to fix the issue in the dma-buf framework.
- */
-static struct drm_gem_object *udl_driver_gem_prime_import(struct drm_device *dev,
-							  struct dma_buf *dma_buf)
-{
-	struct udl_device *udl = to_udl(dev);
-
-	if (!udl->dmadev)
-		return ERR_PTR(-ENODEV);
-
-	return drm_gem_prime_import_dev(dev, dma_buf, udl->dmadev);
-}
-
 DEFINE_DRM_GEM_FOPS(udl_driver_fops);
 
 static const struct drm_driver driver = {
@@ -73,7 +58,6 @@ static const struct drm_driver driver = {
 	/* GEM hooks */
 	.fops = &udl_driver_fops,
 	DRM_GEM_SHMEM_DRIVER_OPS,
-	.gem_prime_import = udl_driver_gem_prime_import,
 	DRM_FBDEV_SHMEM_DRIVER_OPS,
 
 	.name = DRIVER_NAME,
@@ -126,10 +110,10 @@ static int udl_usb_probe(struct usb_interface *interface,
 static void udl_usb_disconnect(struct usb_interface *interface)
 {
 	struct drm_device *dev = usb_get_intfdata(interface);
+	struct udl_device *udl = to_udl(dev);
 
-	drm_kms_helper_poll_fini(dev);
-	udl_drop_usb(dev);
 	drm_dev_unplug(dev);
+	udl_drop_usb(udl);
 }
 
 /*
diff --git a/drivers/gpu/drm/udl/udl_drv.h b/drivers/gpu/drm/udl/udl_drv.h
index be00dc1d87a1..145bb95ccc48 100644
--- a/drivers/gpu/drm/udl/udl_drv.h
+++ b/drivers/gpu/drm/udl/udl_drv.h
@@ -50,18 +50,14 @@ struct urb_list {
 
 struct udl_device {
 	struct drm_device drm;
-	struct device *dev;
-	struct device *dmadev;
+
+	unsigned long sku_pixel_limit;
 
 	struct drm_plane primary_plane;
 	struct drm_crtc crtc;
 	struct drm_encoder encoder;
 	struct drm_connector connector;
 
-	struct mutex gem_lock;
-
-	int sku_pixel_limit;
-
 	struct urb_list urbs;
 };
 
@@ -73,22 +69,22 @@ static inline struct usb_device *udl_to_usb_device(struct udl_device *udl)
 }
 
 /* modeset */
-int udl_modeset_init(struct drm_device *dev);
+int udl_modeset_init(struct udl_device *udl);
 struct drm_connector *udl_connector_init(struct drm_device *dev);
 
-struct urb *udl_get_urb(struct drm_device *dev);
+struct urb *udl_get_urb(struct udl_device *udl);
 
-int udl_submit_urb(struct drm_device *dev, struct urb *urb, size_t len);
-void udl_sync_pending_urbs(struct drm_device *dev);
+int udl_submit_urb(struct udl_device *udl, struct urb *urb, size_t len);
+void udl_sync_pending_urbs(struct udl_device *udl);
 void udl_urb_completion(struct urb *urb);
 
 int udl_init(struct udl_device *udl);
 
-int udl_render_hline(struct drm_device *dev, int log_bpp, struct urb **urb_ptr,
+int udl_render_hline(struct udl_device *udl, int log_bpp, struct urb **urb_ptr,
 		     const char *front, char **urb_buf_ptr,
 		     u32 byte_offset, u32 device_byte_offset, u32 byte_width);
 
-int udl_drop_usb(struct drm_device *dev);
+int udl_drop_usb(struct udl_device *udl);
 int udl_select_std_channel(struct udl_device *udl);
 
 #endif
diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index 3ebe2ce55dfd..bc58991a6f14 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -8,6 +8,8 @@
  * Copyright (C) 2009 Bernie Thompson <bernie@plugable.com>
  */
 
+#include <linux/unaligned.h>
+
 #include <drm/drm.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
@@ -23,72 +25,99 @@
 #define WRITES_IN_FLIGHT (20)
 #define MAX_VENDOR_DESCRIPTOR_SIZE 256
 
+#define UDL_SKU_PIXEL_LIMIT_DEFAULT	2080000
+
 static struct urb *udl_get_urb_locked(struct udl_device *udl, long timeout);
 
+/*
+ * Try to make sense of whatever we parse. Therefore return @end on
+ * errors, but don't fail hard.
+ */
+static const u8 *udl_parse_key_value_pair(struct udl_device *udl, const u8 *pos, const u8 *end)
+{
+	u16 key;
+	u8 len;
+
+	/* read key */
+	if (pos >= end - 2)
+		return end;
+	key = get_unaligned_le16(pos);
+	pos += 2;
+
+	/* read value length */
+	if (pos >= end - 1)
+		return end;
+	len = *pos++;
+
+	/* read value */
+	if (pos >= end - len)
+		return end;
+	switch (key) {
+	case 0x0200: { /* maximum number of pixels */
+		unsigned int sku_pixel_limit;
+
+		if (len < sizeof(__le32))
+			break;
+		sku_pixel_limit = get_unaligned_le32(pos);
+		if (sku_pixel_limit >= 16 * UDL_SKU_PIXEL_LIMIT_DEFAULT)
+			break; /* almost 100 MiB, so probably bogus */
+		udl->sku_pixel_limit = sku_pixel_limit;
+		break;
+	}
+	default:
+		break;
+	}
+	pos += len;
+
+	return pos;
+}
+
 static int udl_parse_vendor_descriptor(struct udl_device *udl)
 {
+	struct drm_device *dev = &udl->drm;
 	struct usb_device *udev = udl_to_usb_device(udl);
-	char *desc;
-	char *buf;
-	char *desc_end;
-
-	u8 total_len = 0;
+	bool detected = false;
+	void *buf;
+	int ret;
+	unsigned int len;
+	const u8 *desc;
+	const u8 *desc_end;
 
 	buf = kzalloc(MAX_VENDOR_DESCRIPTOR_SIZE, GFP_KERNEL);
 	if (!buf)
-		return false;
+		return -ENOMEM;
+
+	ret = usb_get_descriptor(udev, 0x5f, /* vendor specific */
+				 0, buf, MAX_VENDOR_DESCRIPTOR_SIZE);
+	if (ret < 0)
+		goto out;
+	len = ret;
+
+	if (len < 5)
+		goto out;
+
 	desc = buf;
+	desc_end = desc + len;
 
-	total_len = usb_get_descriptor(udev, 0x5f, /* vendor specific */
-				    0, desc, MAX_VENDOR_DESCRIPTOR_SIZE);
-	if (total_len > 5) {
-		DRM_INFO("vendor descriptor length:%x data:%11ph\n",
-			total_len, desc);
-
-		if ((desc[0] != total_len) || /* descriptor length */
-		    (desc[1] != 0x5f) ||   /* vendor descriptor type */
-		    (desc[2] != 0x01) ||   /* version (2 bytes) */
-		    (desc[3] != 0x00) ||
-		    (desc[4] != total_len - 2)) /* length after type */
-			goto unrecognized;
-
-		desc_end = desc + total_len;
-		desc += 5; /* the fixed header we've already parsed */
-
-		while (desc < desc_end) {
-			u8 length;
-			u16 key;
-
-			key = le16_to_cpu(*((u16 *) desc));
-			desc += sizeof(u16);
-			length = *desc;
-			desc++;
-
-			switch (key) {
-			case 0x0200: { /* max_area */
-				u32 max_area;
-				max_area = le32_to_cpu(*((u32 *)desc));
-				DRM_DEBUG("DL chip limited to %d pixel modes\n",
-					max_area);
-				udl->sku_pixel_limit = max_area;
-				break;
-			}
-			default:
-				break;
-			}
-			desc += length;
-		}
-	}
+	if ((desc[0] != len) ||    /* descriptor length */
+	    (desc[1] != 0x5f) ||   /* vendor descriptor type */
+	    (desc[2] != 0x01) ||   /* version (2 bytes) */
+	    (desc[3] != 0x00) ||
+	    (desc[4] != len - 2))  /* length after type */
+		goto out;
+	desc += 5;
 
-	goto success;
+	detected = true;
 
-unrecognized:
-	/* allow udlfb to load for now even if firmware unrecognized */
-	DRM_ERROR("Unrecognized vendor firmware descriptor\n");
+	while (desc < desc_end)
+		desc = udl_parse_key_value_pair(udl, desc, desc_end);
 
-success:
+out:
+	if (!detected)
+		drm_warn(dev, "Unrecognized vendor firmware descriptor\n");
 	kfree(buf);
-	return true;
+
+	return 0;
 }
 
 /*
@@ -145,9 +174,8 @@ void udl_urb_completion(struct urb *urb)
 	wake_up(&udl->urbs.sleep);
 }
 
-static void udl_free_urb_list(struct drm_device *dev)
+static void udl_free_urb_list(struct udl_device *udl)
 {
-	struct udl_device *udl = to_udl(dev);
 	struct urb_node *unode;
 	struct urb *urb;
 
@@ -172,9 +200,8 @@ static void udl_free_urb_list(struct drm_device *dev)
 	wake_up_all(&udl->urbs.sleep);
 }
 
-static int udl_alloc_urb_list(struct drm_device *dev, int count, size_t size)
+static int udl_alloc_urb_list(struct udl_device *udl, int count, size_t size)
 {
-	struct udl_device *udl = to_udl(dev);
 	struct urb *urb;
 	struct urb_node *unode;
 	char *buf;
@@ -210,7 +237,7 @@ retry:
 			usb_free_urb(urb);
 			if (size > PAGE_SIZE) {
 				size /= 2;
-				udl_free_urb_list(dev);
+				udl_free_urb_list(udl);
 				goto retry;
 			}
 			break;
@@ -259,9 +286,8 @@ static struct urb *udl_get_urb_locked(struct udl_device *udl, long timeout)
 }
 
 #define GET_URB_TIMEOUT	HZ
-struct urb *udl_get_urb(struct drm_device *dev)
+struct urb *udl_get_urb(struct udl_device *udl)
 {
-	struct udl_device *udl = to_udl(dev);
 	struct urb *urb;
 
 	spin_lock_irq(&udl->urbs.lock);
@@ -270,9 +296,8 @@ struct urb *udl_get_urb(struct drm_device *dev)
 	return urb;
 }
 
-int udl_submit_urb(struct drm_device *dev, struct urb *urb, size_t len)
+int udl_submit_urb(struct udl_device *udl, struct urb *urb, size_t len)
 {
-	struct udl_device *udl = to_udl(dev);
 	int ret;
 
 	if (WARN_ON(len > udl->urbs.size)) {
@@ -290,9 +315,9 @@ int udl_submit_urb(struct drm_device *dev, struct urb *urb, size_t len)
 }
 
 /* wait until all pending URBs have been processed */
-void udl_sync_pending_urbs(struct drm_device *dev)
+void udl_sync_pending_urbs(struct udl_device *udl)
 {
-	struct udl_device *udl = to_udl(dev);
+	struct drm_device *dev = &udl->drm;
 
 	spin_lock_irq(&udl->urbs.lock);
 	/* 2 seconds as a sane timeout */
@@ -308,53 +333,55 @@ int udl_init(struct udl_device *udl)
 {
 	struct drm_device *dev = &udl->drm;
 	int ret = -ENOMEM;
+	struct device *dma_dev;
 
 	DRM_DEBUG("\n");
 
-	udl->dmadev = usb_intf_get_dma_device(to_usb_interface(dev->dev));
-	if (!udl->dmadev)
+	dma_dev = usb_intf_get_dma_device(to_usb_interface(dev->dev));
+	if (dma_dev) {
+		drm_dev_set_dma_dev(dev, dma_dev);
+		put_device(dma_dev);
+	} else {
 		drm_warn(dev, "buffer sharing not supported"); /* not an error */
+	}
 
-	mutex_init(&udl->gem_lock);
+	/*
+	 * Not all devices provide vendor descriptors with device
+	 * information. Initialize to default values of real-world
+	 * devices. It is just enough memory for FullHD.
+	 */
+	udl->sku_pixel_limit = UDL_SKU_PIXEL_LIMIT_DEFAULT;
 
-	if (!udl_parse_vendor_descriptor(udl)) {
-		ret = -ENODEV;
-		DRM_ERROR("firmware not recognized. Assume incompatible device\n");
+	ret = udl_parse_vendor_descriptor(udl);
+	if (ret)
 		goto err;
-	}
 
 	if (udl_select_std_channel(udl))
 		DRM_ERROR("Selecting channel failed\n");
 
-	if (!udl_alloc_urb_list(dev, WRITES_IN_FLIGHT, MAX_TRANSFER)) {
+	if (!udl_alloc_urb_list(udl, WRITES_IN_FLIGHT, MAX_TRANSFER)) {
 		DRM_ERROR("udl_alloc_urb_list failed\n");
+		ret = -ENOMEM;
 		goto err;
 	}
 
 	DRM_DEBUG("\n");
-	ret = udl_modeset_init(dev);
+	ret = udl_modeset_init(udl);
 	if (ret)
 		goto err;
 
-	drm_kms_helper_poll_init(dev);
-
 	return 0;
 
 err:
 	if (udl->urbs.count)
-		udl_free_urb_list(dev);
-	put_device(udl->dmadev);
+		udl_free_urb_list(udl);
 	DRM_ERROR("%d\n", ret);
 	return ret;
 }
 
-int udl_drop_usb(struct drm_device *dev)
+int udl_drop_usb(struct udl_device *udl)
 {
-	struct udl_device *udl = to_udl(dev);
-
-	udl_free_urb_list(dev);
-	put_device(udl->dmadev);
-	udl->dmadev = NULL;
+	udl_free_urb_list(udl);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/udl/udl_modeset.c b/drivers/gpu/drm/udl/udl_modeset.c
index bbb04f98886a..231e829bd709 100644
--- a/drivers/gpu/drm/udl/udl_modeset.c
+++ b/drivers/gpu/drm/udl/udl_modeset.c
@@ -205,6 +205,7 @@ static int udl_handle_damage(struct drm_framebuffer *fb,
 			     const struct drm_rect *clip)
 {
 	struct drm_device *dev = fb->dev;
+	struct udl_device *udl = to_udl(dev);
 	void *vaddr = map->vaddr; /* TODO: Use mapping abstraction properly */
 	int i, ret;
 	char *cmd;
@@ -216,7 +217,7 @@ static int udl_handle_damage(struct drm_framebuffer *fb,
 		return ret;
 	log_bpp = ret;
 
-	urb = udl_get_urb(dev);
+	urb = udl_get_urb(udl);
 	if (!urb)
 		return -ENOMEM;
 	cmd = urb->transfer_buffer;
@@ -226,7 +227,7 @@ static int udl_handle_damage(struct drm_framebuffer *fb,
 		const int byte_offset = line_offset + (clip->x1 << log_bpp);
 		const int dev_byte_offset = (fb->width * i + clip->x1) << log_bpp;
 		const int byte_width = drm_rect_width(clip) << log_bpp;
-		ret = udl_render_hline(dev, log_bpp, &urb, (char *)vaddr,
+		ret = udl_render_hline(udl, log_bpp, &urb, (char *)vaddr,
 				       &cmd, byte_offset, dev_byte_offset,
 				       byte_width);
 		if (ret)
@@ -239,7 +240,7 @@ static int udl_handle_damage(struct drm_framebuffer *fb,
 		if (cmd < (char *)urb->transfer_buffer + urb->transfer_buffer_length)
 			*cmd++ = UDL_MSG_BULK;
 		len = cmd - (char *)urb->transfer_buffer;
-		ret = udl_submit_urb(dev, urb, len);
+		ret = udl_submit_urb(udl, urb, len);
 	} else {
 		udl_urb_completion(urb);
 	}
@@ -330,6 +331,7 @@ static const struct drm_plane_funcs udl_primary_plane_funcs = {
 static void udl_crtc_helper_atomic_enable(struct drm_crtc *crtc, struct drm_atomic_state *state)
 {
 	struct drm_device *dev = crtc->dev;
+	struct udl_device *udl = to_udl(dev);
 	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
 	struct drm_display_mode *mode = &crtc_state->mode;
 	struct urb *urb;
@@ -339,7 +341,7 @@ static void udl_crtc_helper_atomic_enable(struct drm_crtc *crtc, struct drm_atom
 	if (!drm_dev_enter(dev, &idx))
 		return;
 
-	urb = udl_get_urb(dev);
+	urb = udl_get_urb(udl);
 	if (!urb)
 		goto out;
 
@@ -355,7 +357,7 @@ static void udl_crtc_helper_atomic_enable(struct drm_crtc *crtc, struct drm_atom
 	buf = udl_vidreg_unlock(buf);
 	buf = udl_dummy_render(buf);
 
-	udl_submit_urb(dev, urb, buf - (char *)urb->transfer_buffer);
+	udl_submit_urb(udl, urb, buf - (char *)urb->transfer_buffer);
 
 out:
 	drm_dev_exit(idx);
@@ -364,6 +366,7 @@ out:
 static void udl_crtc_helper_atomic_disable(struct drm_crtc *crtc, struct drm_atomic_state *state)
 {
 	struct drm_device *dev = crtc->dev;
+	struct udl_device *udl = to_udl(dev);
 	struct urb *urb;
 	char *buf;
 	int idx;
@@ -371,7 +374,7 @@ static void udl_crtc_helper_atomic_disable(struct drm_crtc *crtc, struct drm_ato
 	if (!drm_dev_enter(dev, &idx))
 		return;
 
-	urb = udl_get_urb(dev);
+	urb = udl_get_urb(udl);
 	if (!urb)
 		goto out;
 
@@ -381,7 +384,7 @@ static void udl_crtc_helper_atomic_disable(struct drm_crtc *crtc, struct drm_ato
 	buf = udl_vidreg_unlock(buf);
 	buf = udl_dummy_render(buf);
 
-	udl_submit_urb(dev, urb, buf - (char *)urb->transfer_buffer);
+	udl_submit_urb(udl, urb, buf - (char *)urb->transfer_buffer);
 
 out:
 	drm_dev_exit(idx);
@@ -476,9 +479,9 @@ static const struct drm_mode_config_funcs udl_mode_config_funcs = {
 	.atomic_commit = drm_atomic_helper_commit,
 };
 
-int udl_modeset_init(struct drm_device *dev)
+int udl_modeset_init(struct udl_device *udl)
 {
-	struct udl_device *udl = to_udl(dev);
+	struct drm_device *dev = &udl->drm;
 	struct drm_plane *primary_plane;
 	struct drm_crtc *crtc;
 	struct drm_encoder *encoder;
@@ -535,6 +538,7 @@ int udl_modeset_init(struct drm_device *dev)
 		return ret;
 
 	drm_mode_config_reset(dev);
+	drmm_kms_helper_poll_init(dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/udl/udl_transfer.c b/drivers/gpu/drm/udl/udl_transfer.c
index 62224992988f..7d670b3a5293 100644
--- a/drivers/gpu/drm/udl/udl_transfer.c
+++ b/drivers/gpu/drm/udl/udl_transfer.c
@@ -170,7 +170,7 @@ static void udl_compress_hline16(
  * (that we can only write to, slowly, and can never read), and (optionally)
  * our shadow copy that tracks what's been sent to that hardware buffer.
  */
-int udl_render_hline(struct drm_device *dev, int log_bpp, struct urb **urb_ptr,
+int udl_render_hline(struct udl_device *udl, int log_bpp, struct urb **urb_ptr,
 		     const char *front, char **urb_buf_ptr,
 		     u32 byte_offset, u32 device_byte_offset,
 		     u32 byte_width)
@@ -199,10 +199,10 @@ int udl_render_hline(struct drm_device *dev, int log_bpp, struct urb **urb_ptr,
 
 		if (cmd >= cmd_end) {
 			int len = cmd - (u8 *) urb->transfer_buffer;
-			int ret = udl_submit_urb(dev, urb, len);
+			int ret = udl_submit_urb(udl, urb, len);
 			if (ret)
 				return ret;
-			urb = udl_get_urb(dev);
+			urb = udl_get_urb(udl);
 			if (!urb)
 				return -EAGAIN;
 			*urb_ptr = urb;
diff --git a/drivers/gpu/drm/v3d/v3d_debugfs.c b/drivers/gpu/drm/v3d/v3d_debugfs.c
index 76816f2551c1..7e789e181af0 100644
--- a/drivers/gpu/drm/v3d/v3d_debugfs.c
+++ b/drivers/gpu/drm/v3d/v3d_debugfs.c
@@ -21,74 +21,74 @@ struct v3d_reg_def {
 };
 
 static const struct v3d_reg_def v3d_hub_reg_defs[] = {
-	REGDEF(33, 42, V3D_HUB_AXICFG),
-	REGDEF(33, 71, V3D_HUB_UIFCFG),
-	REGDEF(33, 71, V3D_HUB_IDENT0),
-	REGDEF(33, 71, V3D_HUB_IDENT1),
-	REGDEF(33, 71, V3D_HUB_IDENT2),
-	REGDEF(33, 71, V3D_HUB_IDENT3),
-	REGDEF(33, 71, V3D_HUB_INT_STS),
-	REGDEF(33, 71, V3D_HUB_INT_MSK_STS),
-
-	REGDEF(33, 71, V3D_MMU_CTL),
-	REGDEF(33, 71, V3D_MMU_VIO_ADDR),
-	REGDEF(33, 71, V3D_MMU_VIO_ID),
-	REGDEF(33, 71, V3D_MMU_DEBUG_INFO),
-
-	REGDEF(71, 71, V3D_GMP_STATUS(71)),
-	REGDEF(71, 71, V3D_GMP_CFG(71)),
-	REGDEF(71, 71, V3D_GMP_VIO_ADDR(71)),
+	REGDEF(V3D_GEN_33, V3D_GEN_42, V3D_HUB_AXICFG),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_UIFCFG),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_IDENT0),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_IDENT1),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_IDENT2),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_IDENT3),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_INT_STS),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_HUB_INT_MSK_STS),
+
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_MMU_CTL),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_MMU_VIO_ADDR),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_MMU_VIO_ID),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_MMU_DEBUG_INFO),
+
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_GMP_STATUS(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_GMP_CFG(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_GMP_VIO_ADDR(71)),
 };
 
 static const struct v3d_reg_def v3d_gca_reg_defs[] = {
-	REGDEF(33, 33, V3D_GCA_SAFE_SHUTDOWN),
-	REGDEF(33, 33, V3D_GCA_SAFE_SHUTDOWN_ACK),
+	REGDEF(V3D_GEN_33, V3D_GEN_33, V3D_GCA_SAFE_SHUTDOWN),
+	REGDEF(V3D_GEN_33, V3D_GEN_33, V3D_GCA_SAFE_SHUTDOWN_ACK),
 };
 
 static const struct v3d_reg_def v3d_core_reg_defs[] = {
-	REGDEF(33, 71, V3D_CTL_IDENT0),
-	REGDEF(33, 71, V3D_CTL_IDENT1),
-	REGDEF(33, 71, V3D_CTL_IDENT2),
-	REGDEF(33, 71, V3D_CTL_MISCCFG),
-	REGDEF(33, 71, V3D_CTL_INT_STS),
-	REGDEF(33, 71, V3D_CTL_INT_MSK_STS),
-	REGDEF(33, 71, V3D_CLE_CT0CS),
-	REGDEF(33, 71, V3D_CLE_CT0CA),
-	REGDEF(33, 71, V3D_CLE_CT0EA),
-	REGDEF(33, 71, V3D_CLE_CT1CS),
-	REGDEF(33, 71, V3D_CLE_CT1CA),
-	REGDEF(33, 71, V3D_CLE_CT1EA),
-
-	REGDEF(33, 71, V3D_PTB_BPCA),
-	REGDEF(33, 71, V3D_PTB_BPCS),
-
-	REGDEF(33, 42, V3D_GMP_STATUS(33)),
-	REGDEF(33, 42, V3D_GMP_CFG(33)),
-	REGDEF(33, 42, V3D_GMP_VIO_ADDR(33)),
-
-	REGDEF(33, 71, V3D_ERR_FDBGO),
-	REGDEF(33, 71, V3D_ERR_FDBGB),
-	REGDEF(33, 71, V3D_ERR_FDBGS),
-	REGDEF(33, 71, V3D_ERR_STAT),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CTL_IDENT0),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CTL_IDENT1),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CTL_IDENT2),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CTL_MISCCFG),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CTL_INT_STS),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CTL_INT_MSK_STS),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CLE_CT0CS),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CLE_CT0CA),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CLE_CT0EA),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CLE_CT1CS),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CLE_CT1CA),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_CLE_CT1EA),
+
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_PTB_BPCA),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_PTB_BPCS),
+
+	REGDEF(V3D_GEN_33, V3D_GEN_42, V3D_GMP_STATUS(33)),
+	REGDEF(V3D_GEN_33, V3D_GEN_42, V3D_GMP_CFG(33)),
+	REGDEF(V3D_GEN_33, V3D_GEN_42, V3D_GMP_VIO_ADDR(33)),
+
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_ERR_FDBGO),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_ERR_FDBGB),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_ERR_FDBGS),
+	REGDEF(V3D_GEN_33, V3D_GEN_71, V3D_ERR_STAT),
 };
 
 static const struct v3d_reg_def v3d_csd_reg_defs[] = {
-	REGDEF(41, 71, V3D_CSD_STATUS),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG0(41)),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG1(41)),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG2(41)),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG3(41)),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG4(41)),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG5(41)),
-	REGDEF(41, 42, V3D_CSD_CURRENT_CFG6(41)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG0(71)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG1(71)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG2(71)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG3(71)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG4(71)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG5(71)),
-	REGDEF(71, 71, V3D_CSD_CURRENT_CFG6(71)),
-	REGDEF(71, 71, V3D_V7_CSD_CURRENT_CFG7),
+	REGDEF(V3D_GEN_41, V3D_GEN_71, V3D_CSD_STATUS),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG0(41)),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG1(41)),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG2(41)),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG3(41)),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG4(41)),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG5(41)),
+	REGDEF(V3D_GEN_41, V3D_GEN_42, V3D_CSD_CURRENT_CFG6(41)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG0(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG1(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG2(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG3(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG4(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG5(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_CSD_CURRENT_CFG6(71)),
+	REGDEF(V3D_GEN_71, V3D_GEN_71, V3D_V7_CSD_CURRENT_CFG7),
 };
 
 static int v3d_v3d_debugfs_regs(struct seq_file *m, void *unused)
@@ -164,7 +164,7 @@ static int v3d_v3d_debugfs_ident(struct seq_file *m, void *unused)
 		   str_yes_no(ident2 & V3D_HUB_IDENT2_WITH_MMU));
 	seq_printf(m, "TFU:        %s\n",
 		   str_yes_no(ident1 & V3D_HUB_IDENT1_WITH_TFU));
-	if (v3d->ver <= 42) {
+	if (v3d->ver <= V3D_GEN_42) {
 		seq_printf(m, "TSY:        %s\n",
 			   str_yes_no(ident1 & V3D_HUB_IDENT1_WITH_TSY));
 	}
@@ -196,11 +196,11 @@ static int v3d_v3d_debugfs_ident(struct seq_file *m, void *unused)
 		seq_printf(m, "  QPUs:         %d\n", nslc * qups);
 		seq_printf(m, "  Semaphores:   %d\n",
 			   V3D_GET_FIELD(ident1, V3D_IDENT1_NSEM));
-		if (v3d->ver <= 42) {
+		if (v3d->ver <= V3D_GEN_42) {
 			seq_printf(m, "  BCG int:      %d\n",
 				   (ident2 & V3D_IDENT2_BCG_INT) != 0);
 		}
-		if (v3d->ver < 40) {
+		if (v3d->ver < V3D_GEN_41) {
 			seq_printf(m, "  Override TMU: %d\n",
 				   (misccfg & V3D_MISCCFG_OVRTMUOUT) != 0);
 		}
@@ -234,7 +234,7 @@ static int v3d_measure_clock(struct seq_file *m, void *unused)
 	int core = 0;
 	int measure_ms = 1000;
 
-	if (v3d->ver >= 40) {
+	if (v3d->ver >= V3D_GEN_41) {
 		int cycle_count_reg = V3D_PCTR_CYCLE_COUNT(v3d->ver);
 		V3D_CORE_WRITE(core, V3D_V4_PCTR_0_SRC_0_3,
 			       V3D_SET_FIELD_VER(cycle_count_reg,
diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c
index 852015214e97..5e997ae8bc9c 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.c
+++ b/drivers/gpu/drm/v3d/v3d_drv.c
@@ -17,6 +17,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
 #include <linux/sched/clock.h>
@@ -92,7 +93,7 @@ static int v3d_get_param_ioctl(struct drm_device *dev, void *data,
 		args->value = 1;
 		return 0;
 	case DRM_V3D_PARAM_SUPPORTS_PERFMON:
-		args->value = (v3d->ver >= 40);
+		args->value = (v3d->ver >= V3D_GEN_41);
 		return 0;
 	case DRM_V3D_PARAM_SUPPORTS_MULTISYNC_EXT:
 		args->value = 1;
@@ -254,14 +255,44 @@ static const struct drm_driver v3d_drm_driver = {
 };
 
 static const struct of_device_id v3d_of_match[] = {
-	{ .compatible = "brcm,2711-v3d" },
-	{ .compatible = "brcm,2712-v3d" },
-	{ .compatible = "brcm,7268-v3d" },
-	{ .compatible = "brcm,7278-v3d" },
+	{ .compatible = "brcm,2711-v3d", .data = (void *)V3D_GEN_42 },
+	{ .compatible = "brcm,2712-v3d", .data = (void *)V3D_GEN_71 },
+	{ .compatible = "brcm,7268-v3d", .data = (void *)V3D_GEN_33 },
+	{ .compatible = "brcm,7278-v3d", .data = (void *)V3D_GEN_41 },
 	{},
 };
 MODULE_DEVICE_TABLE(of, v3d_of_match);
 
+static void
+v3d_idle_sms(struct v3d_dev *v3d)
+{
+	if (v3d->ver < V3D_GEN_71)
+		return;
+
+	V3D_SMS_WRITE(V3D_SMS_TEE_CS, V3D_SMS_CLEAR_POWER_OFF);
+
+	if (wait_for((V3D_GET_FIELD(V3D_SMS_READ(V3D_SMS_TEE_CS),
+				    V3D_SMS_STATE) == V3D_SMS_IDLE), 100)) {
+		DRM_ERROR("Failed to power up SMS\n");
+	}
+
+	v3d_reset_sms(v3d);
+}
+
+static void
+v3d_power_off_sms(struct v3d_dev *v3d)
+{
+	if (v3d->ver < V3D_GEN_71)
+		return;
+
+	V3D_SMS_WRITE(V3D_SMS_TEE_CS, V3D_SMS_POWER_OFF);
+
+	if (wait_for((V3D_GET_FIELD(V3D_SMS_READ(V3D_SMS_TEE_CS),
+				    V3D_SMS_STATE) == V3D_SMS_POWER_OFF_STATE), 100)) {
+		DRM_ERROR("Failed to power off SMS\n");
+	}
+}
+
 static int
 map_regs(struct v3d_dev *v3d, void __iomem **regs, const char *name)
 {
@@ -274,6 +305,7 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct drm_device *drm;
 	struct v3d_dev *v3d;
+	enum v3d_gen gen;
 	int ret;
 	u32 mmu_debug;
 	u32 ident1, ident3;
@@ -287,6 +319,9 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, drm);
 
+	gen = (uintptr_t)of_device_get_match_data(dev);
+	v3d->ver = gen;
+
 	ret = map_regs(v3d, &v3d->hub_regs, "hub");
 	if (ret)
 		return ret;
@@ -295,6 +330,12 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 	if (ret)
 		return ret;
 
+	if (v3d->ver >= V3D_GEN_71) {
+		ret = map_regs(v3d, &v3d->sms_regs, "sms");
+		if (ret)
+			return ret;
+	}
+
 	v3d->clk = devm_clk_get_optional(dev, NULL);
 	if (IS_ERR(v3d->clk))
 		return dev_err_probe(dev, PTR_ERR(v3d->clk), "Failed to get V3D clock\n");
@@ -305,6 +346,8 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 		return ret;
 	}
 
+	v3d_idle_sms(v3d);
+
 	mmu_debug = V3D_READ(V3D_MMU_DEBUG_INFO);
 	mask = DMA_BIT_MASK(30 + V3D_GET_FIELD(mmu_debug, V3D_MMU_PA_WIDTH));
 	ret = dma_set_mask_and_coherent(dev, mask);
@@ -316,6 +359,11 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 	ident1 = V3D_READ(V3D_HUB_IDENT1);
 	v3d->ver = (V3D_GET_FIELD(ident1, V3D_HUB_IDENT1_TVER) * 10 +
 		    V3D_GET_FIELD(ident1, V3D_HUB_IDENT1_REV));
+	/* Make sure that the V3D tech version retrieved from the HW is equal
+	 * to the one advertised by the device tree.
+	 */
+	WARN_ON(v3d->ver != gen);
+
 	v3d->cores = V3D_GET_FIELD(ident1, V3D_HUB_IDENT1_NCORES);
 	WARN_ON(v3d->cores > 1); /* multicore not yet implemented */
 
@@ -340,7 +388,7 @@ static int v3d_platform_drm_probe(struct platform_device *pdev)
 		}
 	}
 
-	if (v3d->ver < 41) {
+	if (v3d->ver < V3D_GEN_41) {
 		ret = map_regs(v3d, &v3d->gca_regs, "gca");
 		if (ret)
 			goto clk_disable;
@@ -400,6 +448,8 @@ static void v3d_platform_drm_remove(struct platform_device *pdev)
 	dma_free_wc(v3d->drm.dev, 4096, v3d->mmu_scratch,
 		    v3d->mmu_scratch_paddr);
 
+	v3d_power_off_sms(v3d);
+
 	clk_disable_unprepare(v3d->clk);
 }
 
diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index 9deaefa0f95b..b51f0b648a08 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -94,11 +94,18 @@ struct v3d_perfmon {
 	u64 values[] __counted_by(ncounters);
 };
 
+enum v3d_gen {
+	V3D_GEN_33 = 33,
+	V3D_GEN_41 = 41,
+	V3D_GEN_42 = 42,
+	V3D_GEN_71 = 71,
+};
+
 struct v3d_dev {
 	struct drm_device drm;
 
 	/* Short representation (e.g. 33, 41) of the V3D tech version */
-	int ver;
+	enum v3d_gen ver;
 
 	/* Short representation (e.g. 5, 6) of the V3D tech revision */
 	int rev;
@@ -111,6 +118,7 @@ struct v3d_dev {
 	void __iomem *core_regs[3];
 	void __iomem *bridge_regs;
 	void __iomem *gca_regs;
+	void __iomem *sms_regs;
 	struct clk *clk;
 	struct reset_control *reset;
 
@@ -199,7 +207,7 @@ to_v3d_dev(struct drm_device *dev)
 static inline bool
 v3d_has_csd(struct v3d_dev *v3d)
 {
-	return v3d->ver >= 41;
+	return v3d->ver >= V3D_GEN_41;
 }
 
 #define v3d_to_pdev(v3d) to_platform_device((v3d)->drm.dev)
@@ -261,6 +269,15 @@ to_v3d_fence(struct dma_fence *fence)
 #define V3D_GCA_READ(offset) readl(v3d->gca_regs + offset)
 #define V3D_GCA_WRITE(offset, val) writel(val, v3d->gca_regs + offset)
 
+#define V3D_SMS_IDLE				0x0
+#define V3D_SMS_ISOLATING_FOR_RESET		0xa
+#define V3D_SMS_RESETTING			0xb
+#define V3D_SMS_ISOLATING_FOR_POWER_OFF	0xc
+#define V3D_SMS_POWER_OFF_STATE		0xd
+
+#define V3D_SMS_READ(offset) readl(v3d->sms_regs + (offset))
+#define V3D_SMS_WRITE(offset, val) writel(val, v3d->sms_regs + (offset))
+
 #define V3D_CORE_READ(core, offset) readl(v3d->core_regs[core] + offset)
 #define V3D_CORE_WRITE(core, offset, val) writel(val, v3d->core_regs[core] + offset)
 
@@ -539,6 +556,7 @@ struct dma_fence *v3d_fence_create(struct v3d_dev *v3d, enum v3d_queue queue);
 /* v3d_gem.c */
 int v3d_gem_init(struct drm_device *dev);
 void v3d_gem_destroy(struct drm_device *dev);
+void v3d_reset_sms(struct v3d_dev *v3d);
 void v3d_reset(struct v3d_dev *v3d);
 void v3d_invalidate_caches(struct v3d_dev *v3d);
 void v3d_clean_caches(struct v3d_dev *v3d);
diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
index b1e681630ded..d7d16da78db3 100644
--- a/drivers/gpu/drm/v3d/v3d_gem.c
+++ b/drivers/gpu/drm/v3d/v3d_gem.c
@@ -25,7 +25,7 @@ v3d_init_core(struct v3d_dev *v3d, int core)
 	 * type.  If you want the default behavior, you can still put
 	 * "2" in the indirect texture state's output_type field.
 	 */
-	if (v3d->ver < 40)
+	if (v3d->ver < V3D_GEN_41)
 		V3D_CORE_WRITE(core, V3D_CTL_MISCCFG, V3D_MISCCFG_OVRTMUOUT);
 
 	/* Whenever we flush the L2T cache, we always want to flush
@@ -58,7 +58,7 @@ v3d_idle_axi(struct v3d_dev *v3d, int core)
 static void
 v3d_idle_gca(struct v3d_dev *v3d)
 {
-	if (v3d->ver >= 41)
+	if (v3d->ver >= V3D_GEN_41)
 		return;
 
 	V3D_GCA_WRITE(V3D_GCA_SAFE_SHUTDOWN, V3D_GCA_SAFE_SHUTDOWN_EN);
@@ -105,6 +105,22 @@ v3d_reset_v3d(struct v3d_dev *v3d)
 }
 
 void
+v3d_reset_sms(struct v3d_dev *v3d)
+{
+	if (v3d->ver < V3D_GEN_71)
+		return;
+
+	V3D_SMS_WRITE(V3D_SMS_REE_CS, V3D_SET_FIELD(0x4, V3D_SMS_STATE));
+
+	if (wait_for(!(V3D_GET_FIELD(V3D_SMS_READ(V3D_SMS_REE_CS),
+				     V3D_SMS_STATE) == V3D_SMS_ISOLATING_FOR_RESET) &&
+		     !(V3D_GET_FIELD(V3D_SMS_READ(V3D_SMS_REE_CS),
+				     V3D_SMS_STATE) == V3D_SMS_RESETTING), 100)) {
+		DRM_ERROR("Failed to wait for SMS reset\n");
+	}
+}
+
+void
 v3d_reset(struct v3d_dev *v3d)
 {
 	struct drm_device *dev = &v3d->drm;
@@ -119,6 +135,7 @@ v3d_reset(struct v3d_dev *v3d)
 		v3d_idle_axi(v3d, 0);
 
 	v3d_idle_gca(v3d);
+	v3d_reset_sms(v3d);
 	v3d_reset_v3d(v3d);
 
 	v3d_mmu_set_page_table(v3d);
@@ -132,13 +149,13 @@ v3d_reset(struct v3d_dev *v3d)
 static void
 v3d_flush_l3(struct v3d_dev *v3d)
 {
-	if (v3d->ver < 41) {
+	if (v3d->ver < V3D_GEN_41) {
 		u32 gca_ctrl = V3D_GCA_READ(V3D_GCA_CACHE_CTRL);
 
 		V3D_GCA_WRITE(V3D_GCA_CACHE_CTRL,
 			      gca_ctrl | V3D_GCA_CACHE_CTRL_FLUSH);
 
-		if (v3d->ver < 33) {
+		if (v3d->ver < V3D_GEN_33) {
 			V3D_GCA_WRITE(V3D_GCA_CACHE_CTRL,
 				      gca_ctrl & ~V3D_GCA_CACHE_CTRL_FLUSH);
 		}
@@ -151,7 +168,7 @@ v3d_flush_l3(struct v3d_dev *v3d)
 static void
 v3d_invalidate_l2c(struct v3d_dev *v3d, int core)
 {
-	if (v3d->ver > 32)
+	if (v3d->ver >= V3D_GEN_33)
 		return;
 
 	V3D_CORE_WRITE(core, V3D_CTL_L2CACTL,
diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c
index 72b6a119412f..2cca5d3a26a2 100644
--- a/drivers/gpu/drm/v3d/v3d_irq.c
+++ b/drivers/gpu/drm/v3d/v3d_irq.c
@@ -143,7 +143,7 @@ v3d_irq(int irq, void *arg)
 	/* We shouldn't be triggering these if we have GMP in
 	 * always-allowed mode.
 	 */
-	if (v3d->ver < 71 && (intsts & V3D_INT_GMPV))
+	if (v3d->ver < V3D_GEN_71 && (intsts & V3D_INT_GMPV))
 		dev_err(v3d->drm.dev, "GMP violation\n");
 
 	/* V3D 4.2 wires the hub and core IRQs together, so if we &
@@ -186,27 +186,59 @@ v3d_hub_irq(int irq, void *arg)
 		u32 axi_id = V3D_READ(V3D_MMU_VIO_ID);
 		u64 vio_addr = ((u64)V3D_READ(V3D_MMU_VIO_ADDR) <<
 				(v3d->va_width - 32));
-		static const char *const v3d41_axi_ids[] = {
-			"L2T",
-			"PTB",
-			"PSE",
-			"TLB",
-			"CLE",
-			"TFU",
-			"MMU",
-			"GMP",
+		static const struct {
+			u32 begin;
+			u32 end;
+			const char *client;
+		} v3d41_axi_ids[] = {
+			{0x00, 0x20, "L2T"},
+			{0x20, 0x21, "PTB"},
+			{0x40, 0x41, "PSE"},
+			{0x60, 0x80, "TLB"},
+			{0x80, 0x88, "CLE"},
+			{0xA0, 0xA1, "TFU"},
+			{0xC0, 0xE0, "MMU"},
+			{0xE0, 0xE1, "GMP"},
+		}, v3d71_axi_ids[] = {
+			{0x00, 0x30, "L2T"},
+			{0x30, 0x38, "CLE"},
+			{0x38, 0x39, "PTB"},
+			{0x39, 0x3A, "PSE"},
+			{0x3A, 0x3B, "CSD"},
+			{0x40, 0x60, "TLB"},
+			{0x60, 0x70, "MMU"},
+			{0x7C, 0x7E, "TFU"},
+			{0x7F, 0x80, "GMP"},
 		};
 		const char *client = "?";
 
 		V3D_WRITE(V3D_MMU_CTL, V3D_READ(V3D_MMU_CTL));
 
-		if (v3d->ver >= 41) {
-			axi_id = axi_id >> 5;
-			if (axi_id < ARRAY_SIZE(v3d41_axi_ids))
-				client = v3d41_axi_ids[axi_id];
+		if (v3d->ver >= V3D_GEN_71) {
+			size_t i;
+
+			axi_id = axi_id & 0x7F;
+			for (i = 0; i < ARRAY_SIZE(v3d71_axi_ids); i++) {
+				if (axi_id >= v3d71_axi_ids[i].begin &&
+				    axi_id < v3d71_axi_ids[i].end) {
+					client = v3d71_axi_ids[i].client;
+					break;
+				}
+			}
+		} else if (v3d->ver >= V3D_GEN_41) {
+			size_t i;
+
+			axi_id = axi_id & 0xFF;
+			for (i = 0; i < ARRAY_SIZE(v3d41_axi_ids); i++) {
+				if (axi_id >= v3d41_axi_ids[i].begin &&
+				    axi_id < v3d41_axi_ids[i].end) {
+					client = v3d41_axi_ids[i].client;
+					break;
+				}
+			}
 		}
 
-		dev_err(v3d->drm.dev, "MMU error from client %s (%d) at 0x%llx%s%s%s\n",
+		dev_err(v3d->drm.dev, "MMU error from client %s (0x%x) at 0x%llx%s%s%s\n",
 			client, axi_id, (long long)vio_addr,
 			((intsts & V3D_HUB_INT_MMU_WRV) ?
 			 ", write violation" : ""),
@@ -217,7 +249,7 @@ v3d_hub_irq(int irq, void *arg)
 		status = IRQ_HANDLED;
 	}
 
-	if (v3d->ver >= 71 && (intsts & V3D_V7_HUB_INT_GMPV)) {
+	if (v3d->ver >= V3D_GEN_71 && (intsts & V3D_V7_HUB_INT_GMPV)) {
 		dev_err(v3d->drm.dev, "GMP Violation\n");
 		status = IRQ_HANDLED;
 	}
diff --git a/drivers/gpu/drm/v3d/v3d_perfmon.c b/drivers/gpu/drm/v3d/v3d_perfmon.c
index 3ebda2fa46fc..9a3fe5255874 100644
--- a/drivers/gpu/drm/v3d/v3d_perfmon.c
+++ b/drivers/gpu/drm/v3d/v3d_perfmon.c
@@ -200,10 +200,10 @@ void v3d_perfmon_init(struct v3d_dev *v3d)
 	const struct v3d_perf_counter_desc *counters = NULL;
 	unsigned int max = 0;
 
-	if (v3d->ver >= 71) {
+	if (v3d->ver >= V3D_GEN_71) {
 		counters = v3d_v71_performance_counters;
 		max = ARRAY_SIZE(v3d_v71_performance_counters);
-	} else if (v3d->ver >= 42) {
+	} else if (v3d->ver >= V3D_GEN_42) {
 		counters = v3d_v42_performance_counters;
 		max = ARRAY_SIZE(v3d_v42_performance_counters);
 	}
diff --git a/drivers/gpu/drm/v3d/v3d_regs.h b/drivers/gpu/drm/v3d/v3d_regs.h
index 6da3c69082bd..c1870265eaee 100644
--- a/drivers/gpu/drm/v3d/v3d_regs.h
+++ b/drivers/gpu/drm/v3d/v3d_regs.h
@@ -515,4 +515,30 @@
 # define V3D_ERR_VPAERGS                               BIT(1)
 # define V3D_ERR_VPAEABB                               BIT(0)
 
+#define V3D_SMS_REE_CS                                 0x00000
+#define V3D_SMS_TEE_CS                                 0x00400
+# define V3D_SMS_INTERRUPT                             BIT(31)
+# define V3D_SMS_POWER_OFF                             BIT(30)
+# define V3D_SMS_CLEAR_POWER_OFF                       BIT(29)
+# define V3D_SMS_LOCK                                  BIT(28)
+# define V3D_SMS_CLEAR_LOCK                            BIT(27)
+# define V3D_SMS_SVP_MODE_EXIT                         BIT(26)
+# define V3D_SMS_CLEAR_SVP_MODE_EXIT                   BIT(25)
+# define V3D_SMS_SVP_MODE_ENTER                        BIT(24)
+# define V3D_SMS_CLEAR_SVP_MODE_ENTER                  BIT(23)
+# define V3D_SMS_THEIR_MODE_EXIT                       BIT(22)
+# define V3D_SMS_THEIR_MODE_ENTER                      BIT(21)
+# define V3D_SMS_OUR_MODE_EXIT                         BIT(20)
+# define V3D_SMS_CLEAR_OUR_MODE_EXIT                   BIT(19)
+# define V3D_SMS_SEQ_PC_MASK                           V3D_MASK(16, 10)
+# define V3D_SMS_SEQ_PC_SHIFT                          10
+# define V3D_SMS_HUBCORE_STATUS_MASK                   V3D_MASK(9, 8)
+# define V3D_SMS_HUBCORE_STATUS_SHIFT                  8
+# define V3D_SMS_NEW_MODE_MASK                         V3D_MASK(7, 6)
+# define V3D_SMS_NEW_MODE_SHIFT                        6
+# define V3D_SMS_OLD_MODE_MASK                         V3D_MASK(5, 4)
+# define V3D_SMS_OLD_MODE_SHIFT                        4
+# define V3D_SMS_STATE_MASK                            V3D_MASK(3, 0)
+# define V3D_SMS_STATE_SHIFT                           0
+
 #endif /* V3D_REGS_H */
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 34c42d6e12cd..466d28ceee28 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -357,11 +357,11 @@ v3d_tfu_job_run(struct drm_sched_job *sched_job)
 	V3D_WRITE(V3D_TFU_ICA(v3d->ver), job->args.ica);
 	V3D_WRITE(V3D_TFU_IUA(v3d->ver), job->args.iua);
 	V3D_WRITE(V3D_TFU_IOA(v3d->ver), job->args.ioa);
-	if (v3d->ver >= 71)
+	if (v3d->ver >= V3D_GEN_71)
 		V3D_WRITE(V3D_V7_TFU_IOC, job->args.v71.ioc);
 	V3D_WRITE(V3D_TFU_IOS(v3d->ver), job->args.ios);
 	V3D_WRITE(V3D_TFU_COEF0(v3d->ver), job->args.coef[0]);
-	if (v3d->ver >= 71 || (job->args.coef[0] & V3D_TFU_COEF0_USECOEF)) {
+	if (v3d->ver >= V3D_GEN_71 || (job->args.coef[0] & V3D_TFU_COEF0_USECOEF)) {
 		V3D_WRITE(V3D_TFU_COEF1(v3d->ver), job->args.coef[1]);
 		V3D_WRITE(V3D_TFU_COEF2(v3d->ver), job->args.coef[2]);
 		V3D_WRITE(V3D_TFU_COEF3(v3d->ver), job->args.coef[3]);
@@ -412,7 +412,7 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)
 	 *
 	 * XXX: Set the CFG7 register
 	 */
-	if (v3d->ver >= 71)
+	if (v3d->ver >= V3D_GEN_71)
 		V3D_CORE_WRITE(0, V3D_V7_CSD_QUEUED_CFG7, 0);
 
 	/* CFG0 write kicks off the job. */
@@ -428,7 +428,8 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job)
 	struct v3d_bo *bo = to_v3d_bo(job->base.bo[0]);
 	struct v3d_bo *indirect = to_v3d_bo(indirect_csd->indirect);
 	struct drm_v3d_submit_csd *args = &indirect_csd->job->args;
-	u32 *wg_counts;
+	struct v3d_dev *v3d = job->base.v3d;
+	u32 num_batches, *wg_counts;
 
 	v3d_get_bo_vaddr(bo);
 	v3d_get_bo_vaddr(indirect);
@@ -441,8 +442,17 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job)
 	args->cfg[0] = wg_counts[0] << V3D_CSD_CFG012_WG_COUNT_SHIFT;
 	args->cfg[1] = wg_counts[1] << V3D_CSD_CFG012_WG_COUNT_SHIFT;
 	args->cfg[2] = wg_counts[2] << V3D_CSD_CFG012_WG_COUNT_SHIFT;
-	args->cfg[4] = DIV_ROUND_UP(indirect_csd->wg_size, 16) *
-		       (wg_counts[0] * wg_counts[1] * wg_counts[2]) - 1;
+
+	num_batches = DIV_ROUND_UP(indirect_csd->wg_size, 16) *
+		      (wg_counts[0] * wg_counts[1] * wg_counts[2]);
+
+	/* V3D 7.1.6 and later don't subtract 1 from the number of batches */
+	if (v3d->ver < 71 || (v3d->ver == 71 && v3d->rev < 6))
+		args->cfg[4] = num_batches - 1;
+	else
+		args->cfg[4] = num_batches;
+
+	WARN_ON(args->cfg[4] == ~0);
 
 	for (int i = 0; i < 3; i++) {
 		/* 0xffffffff indicates that the uniform rewrite is not needed */
diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock_output.c b/drivers/gpu/drm/vc4/tests/vc4_mock_output.c
index e70d7c3076ac..577d9a956369 100644
--- a/drivers/gpu/drm/vc4/tests/vc4_mock_output.c
+++ b/drivers/gpu/drm/vc4/tests/vc4_mock_output.c
@@ -61,6 +61,19 @@ static const struct drm_display_mode default_mode = {
 	DRM_SIMPLE_MODE(640, 480, 64, 48)
 };
 
+/**
+ * vc4_mock_atomic_add_output() - Enables an output in a state
+ * @test: The test context object
+ * @state: Atomic state to enable the output in.
+ * @type: Type of the output encoder
+ *
+ * Adds an output CRTC and connector to a state, and enables them.
+ *
+ * Returns:
+ * 0 on success, a negative error code on failure. If the error is
+ * EDEADLK, the entire atomic sequence must be restarted. All other
+ * errors are fatal.
+ */
 int vc4_mock_atomic_add_output(struct kunit *test,
 			       struct drm_atomic_state *state,
 			       enum vc4_encoder_type type)
@@ -75,30 +88,49 @@ int vc4_mock_atomic_add_output(struct kunit *test,
 	int ret;
 
 	encoder = vc4_find_encoder_by_type(drm, type);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder);
+	if (!encoder)
+		return -ENODEV;
 
 	crtc = vc4_find_crtc_for_encoder(test, drm, encoder);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc);
+	if (!crtc)
+		return -ENODEV;
 
 	output = encoder_to_vc4_dummy_output(encoder);
 	conn = &output->connector;
 	conn_state = drm_atomic_get_connector_state(state, conn);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state);
+	if (IS_ERR(conn_state))
+		return PTR_ERR(conn_state);
 
 	ret = drm_atomic_set_crtc_for_connector(conn_state, crtc);
-	KUNIT_EXPECT_EQ(test, ret, 0);
+	if (ret)
+		return ret;
 
 	crtc_state = drm_atomic_get_crtc_state(state, crtc);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_state);
+	if (IS_ERR(crtc_state))
+		return PTR_ERR(crtc_state);
 
 	ret = drm_atomic_set_mode_for_crtc(crtc_state, &default_mode);
-	KUNIT_EXPECT_EQ(test, ret, 0);
+	if (ret)
+		return ret;
 
 	crtc_state->active = true;
 
 	return 0;
 }
 
+/**
+ * vc4_mock_atomic_del_output() - Disables an output in a state
+ * @test: The test context object
+ * @state: Atomic state to disable the output in.
+ * @type: Type of the output encoder
+ *
+ * Adds an output CRTC and connector to a state, and disables them.
+ *
+ * Returns:
+ * 0 on success, a negative error code on failure. If the error is
+ * EDEADLK, the entire atomic sequence must be restarted. All other
+ * errors are fatal.
+ */
 int vc4_mock_atomic_del_output(struct kunit *test,
 			       struct drm_atomic_state *state,
 			       enum vc4_encoder_type type)
@@ -113,26 +145,32 @@ int vc4_mock_atomic_del_output(struct kunit *test,
 	int ret;
 
 	encoder = vc4_find_encoder_by_type(drm, type);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder);
+	if (!encoder)
+		return -ENODEV;
 
 	crtc = vc4_find_crtc_for_encoder(test, drm, encoder);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc);
+	if (!crtc)
+		return -ENODEV;
 
 	crtc_state = drm_atomic_get_crtc_state(state, crtc);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_state);
+	if (IS_ERR(crtc_state))
+		return PTR_ERR(crtc_state);
 
 	crtc_state->active = false;
 
 	ret = drm_atomic_set_mode_for_crtc(crtc_state, NULL);
-	KUNIT_ASSERT_EQ(test, ret, 0);
+	if (ret)
+		return ret;
 
 	output = encoder_to_vc4_dummy_output(encoder);
 	conn = &output->connector;
 	conn_state = drm_atomic_get_connector_state(state, conn);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state);
+	if (IS_ERR(conn_state))
+		return PTR_ERR(conn_state);
 
 	ret = drm_atomic_set_crtc_for_connector(conn_state, NULL);
-	KUNIT_ASSERT_EQ(test, ret, 0);
+	if (ret)
+		return ret;
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c b/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c
index 992e8f5c5c6e..d1f694029169 100644
--- a/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c
+++ b/drivers/gpu/drm/vc4/tests/vc4_test_pv_muxing.c
@@ -20,7 +20,6 @@
 
 struct pv_muxing_priv {
 	struct vc4_dev *vc4;
-	struct drm_atomic_state *state;
 };
 
 static bool check_fifo_conflict(struct kunit *test,
@@ -677,18 +676,41 @@ static void drm_vc4_test_pv_muxing(struct kunit *test)
 {
 	const struct pv_muxing_param *params = test->param_value;
 	const struct pv_muxing_priv *priv = test->priv;
-	struct drm_atomic_state *state = priv->state;
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_atomic_state *state;
+	struct drm_device *drm;
+	struct vc4_dev *vc4;
 	unsigned int i;
 	int ret;
 
+	drm_modeset_acquire_init(&ctx, 0);
+
+	vc4 = priv->vc4;
+	drm = &vc4->base;
+
+retry:
+	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 	for (i = 0; i < params->nencoders; i++) {
 		enum vc4_encoder_type enc_type = params->encoders[i];
 
 		ret = vc4_mock_atomic_add_output(test, state, enc_type);
+		if (ret == -EDEADLK) {
+			drm_atomic_state_clear(state);
+			ret = drm_modeset_backoff(&ctx);
+			if (!ret)
+				goto retry;
+		}
 		KUNIT_ASSERT_EQ(test, ret, 0);
 	}
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry;
+	}
 	KUNIT_EXPECT_EQ(test, ret, 0);
 
 	KUNIT_EXPECT_TRUE(test,
@@ -700,33 +722,61 @@ static void drm_vc4_test_pv_muxing(struct kunit *test)
 		KUNIT_EXPECT_TRUE(test, check_channel_for_encoder(test, state, enc_type,
 								  params->check_fn));
 	}
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
 }
 
 static void drm_vc4_test_pv_muxing_invalid(struct kunit *test)
 {
 	const struct pv_muxing_param *params = test->param_value;
 	const struct pv_muxing_priv *priv = test->priv;
-	struct drm_atomic_state *state = priv->state;
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_atomic_state *state;
+	struct drm_device *drm;
+	struct vc4_dev *vc4;
 	unsigned int i;
 	int ret;
 
+	drm_modeset_acquire_init(&ctx, 0);
+
+	vc4 = priv->vc4;
+	drm = &vc4->base;
+
+retry:
+	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
+
 	for (i = 0; i < params->nencoders; i++) {
 		enum vc4_encoder_type enc_type = params->encoders[i];
 
 		ret = vc4_mock_atomic_add_output(test, state, enc_type);
+		if (ret == -EDEADLK) {
+			drm_atomic_state_clear(state);
+			ret = drm_modeset_backoff(&ctx);
+			if (!ret)
+				goto retry;
+		}
 		KUNIT_ASSERT_EQ(test, ret, 0);
 	}
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry;
+	}
 	KUNIT_EXPECT_LT(test, ret, 0);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
 }
 
 static int vc4_pv_muxing_test_init(struct kunit *test)
 {
 	const struct pv_muxing_param *params = test->param_value;
-	struct drm_modeset_acquire_ctx ctx;
 	struct pv_muxing_priv *priv;
-	struct drm_device *drm;
 	struct vc4_dev *vc4;
 
 	priv = kunit_kzalloc(test, sizeof(*priv), GFP_KERNEL);
@@ -737,15 +787,6 @@ static int vc4_pv_muxing_test_init(struct kunit *test)
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vc4);
 	priv->vc4 = vc4;
 
-	drm_modeset_acquire_init(&ctx, 0);
-
-	drm = &vc4->base;
-	priv->state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
-	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->state);
-
-	drm_modeset_drop_locks(&ctx);
-	drm_modeset_acquire_fini(&ctx);
-
 	return 0;
 }
 
@@ -800,13 +841,26 @@ static void drm_test_vc5_pv_muxing_bugs_subsequent_crtc_enable(struct kunit *tes
 	drm_modeset_acquire_init(&ctx, 0);
 
 	drm = &vc4->base;
+retry_first:
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 
 	ret = vc4_mock_atomic_add_output(test, state, VC4_ENCODER_TYPE_HDMI0);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	new_hvs_state = vc4_hvs_get_new_global_state(state);
@@ -823,13 +877,26 @@ static void drm_test_vc5_pv_muxing_bugs_subsequent_crtc_enable(struct kunit *tes
 	ret = drm_atomic_helper_swap_state(state, false);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
+retry_second:
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 
 	ret = vc4_mock_atomic_add_output(test, state, VC4_ENCODER_TYPE_HDMI1);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_second;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_second;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	new_hvs_state = vc4_hvs_get_new_global_state(state);
@@ -874,16 +941,35 @@ static void drm_test_vc5_pv_muxing_bugs_stable_fifo(struct kunit *test)
 	drm_modeset_acquire_init(&ctx, 0);
 
 	drm = &vc4->base;
+retry_first:
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 
 	ret = vc4_mock_atomic_add_output(test, state, VC4_ENCODER_TYPE_HDMI0);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = vc4_mock_atomic_add_output(test, state, VC4_ENCODER_TYPE_HDMI1);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	new_hvs_state = vc4_hvs_get_new_global_state(state);
@@ -908,13 +994,26 @@ static void drm_test_vc5_pv_muxing_bugs_stable_fifo(struct kunit *test)
 	ret = drm_atomic_helper_swap_state(state, false);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
+retry_second:
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 
 	ret = vc4_mock_atomic_del_output(test, state, VC4_ENCODER_TYPE_HDMI0);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_second;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_second;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	new_hvs_state = vc4_hvs_get_new_global_state(state);
@@ -968,25 +1067,50 @@ drm_test_vc5_pv_muxing_bugs_subsequent_crtc_enable_too_many_crtc_state(struct ku
 	drm_modeset_acquire_init(&ctx, 0);
 
 	drm = &vc4->base;
+retry_first:
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 
 	ret = vc4_mock_atomic_add_output(test, state, VC4_ENCODER_TYPE_HDMI0);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_first;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
-
 	ret = drm_atomic_helper_swap_state(state, false);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
+retry_second:
 	state = drm_kunit_helper_atomic_state_alloc(test, drm, &ctx);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state);
 
 	ret = vc4_mock_atomic_add_output(test, state, VC4_ENCODER_TYPE_HDMI1);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_second;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	ret = drm_atomic_check_only(state);
+	if (ret == -EDEADLK) {
+		drm_atomic_state_clear(state);
+		ret = drm_modeset_backoff(&ctx);
+		if (!ret)
+			goto retry_second;
+	}
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
 	new_vc4_crtc_state = get_vc4_crtc_state_for_encoder(test, state,
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index 779b22efe27b..458e5d987964 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -552,8 +552,6 @@ struct vc4_dsi {
 	struct vc4_encoder encoder;
 	struct mipi_dsi_host dsi_host;
 
-	struct kref kref;
-
 	struct platform_device *pdev;
 
 	struct drm_bridge *out_bridge;
@@ -1160,12 +1158,13 @@ static void vc4_dsi_bridge_enable(struct drm_bridge *bridge,
 }
 
 static int vc4_dsi_bridge_attach(struct drm_bridge *bridge,
+				 struct drm_encoder *encoder,
 				 enum drm_bridge_attach_flags flags)
 {
 	struct vc4_dsi *dsi = bridge_to_vc4_dsi(bridge);
 
 	/* Attach the panel or bridge to the dsi bridge */
-	return drm_bridge_attach(bridge->encoder, dsi->out_bridge,
+	return drm_bridge_attach(encoder, dsi->out_bridge,
 				 &dsi->bridge, flags);
 }
 
@@ -1621,29 +1620,11 @@ static void vc4_dsi_dma_chan_release(void *ptr)
 	dsi->reg_dma_chan = NULL;
 }
 
-static void vc4_dsi_release(struct kref *kref)
-{
-	struct vc4_dsi *dsi =
-		container_of(kref, struct vc4_dsi, kref);
-
-	kfree(dsi);
-}
-
-static void vc4_dsi_get(struct vc4_dsi *dsi)
-{
-	kref_get(&dsi->kref);
-}
-
-static void vc4_dsi_put(struct vc4_dsi *dsi)
-{
-	kref_put(&dsi->kref, &vc4_dsi_release);
-}
-
 static void vc4_dsi_release_action(struct drm_device *drm, void *ptr)
 {
 	struct vc4_dsi *dsi = ptr;
 
-	vc4_dsi_put(dsi);
+	drm_bridge_put(&dsi->bridge);
 }
 
 static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
@@ -1654,7 +1635,7 @@ static int vc4_dsi_bind(struct device *dev, struct device *master, void *data)
 	struct drm_encoder *encoder = &dsi->encoder.base;
 	int ret;
 
-	vc4_dsi_get(dsi);
+	drm_bridge_get(&dsi->bridge);
 
 	ret = drmm_add_action_or_reset(drm, vc4_dsi_release_action, dsi);
 	if (ret)
@@ -1809,15 +1790,12 @@ static int vc4_dsi_dev_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct vc4_dsi *dsi;
 
-	dsi = kzalloc(sizeof(*dsi), GFP_KERNEL);
-	if (!dsi)
-		return -ENOMEM;
+	dsi = devm_drm_bridge_alloc(&pdev->dev, struct vc4_dsi, bridge, &vc4_dsi_bridge_funcs);
+	if (IS_ERR(dsi))
+		return PTR_ERR(dsi);
 	dev_set_drvdata(dev, dsi);
 
-	kref_init(&dsi->kref);
-
 	dsi->pdev = pdev;
-	dsi->bridge.funcs = &vc4_dsi_bridge_funcs;
 #ifdef CONFIG_OF
 	dsi->bridge.of_node = dev->of_node;
 #endif
@@ -1835,7 +1813,6 @@ static void vc4_dsi_dev_remove(struct platform_device *pdev)
 	struct vc4_dsi *dsi = dev_get_drvdata(dev);
 
 	mipi_dsi_host_unregister(&dsi->dsi_host);
-	vc4_dsi_put(dsi);
 }
 
 struct platform_driver vc4_dsi_driver = {
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 37238a12baa5..a29a6ef266f9 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -51,6 +51,7 @@
 #include <linux/reset.h>
 #include <sound/dmaengine_pcm.h>
 #include <sound/hdmi-codec.h>
+#include <sound/jack.h>
 #include <sound/pcm_drm_eld.h>
 #include <sound/pcm_params.h>
 #include <sound/soc.h>
@@ -372,13 +373,13 @@ static void vc4_hdmi_handle_hotplug(struct vc4_hdmi *vc4_hdmi,
 	 * the lock for now.
 	 */
 
+	drm_atomic_helper_connector_hdmi_hotplug(connector, status);
+
 	if (status == connector_status_disconnected) {
 		cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
 		return;
 	}
 
-	drm_atomic_helper_connector_hdmi_hotplug(connector, status);
-
 	cec_s_phys_addr(vc4_hdmi->cec_adap,
 			connector->display_info.source_physical_address, false);
 
@@ -2175,6 +2176,22 @@ static const struct drm_connector_hdmi_audio_funcs vc4_hdmi_audio_funcs = {
 	.shutdown = vc4_hdmi_audio_shutdown,
 };
 
+static int vc4_hdmi_codec_init(struct snd_soc_pcm_runtime *rtd)
+{
+	struct vc4_hdmi *vc4_hdmi = snd_soc_card_get_drvdata(rtd->card);
+	struct snd_soc_component *component = snd_soc_rtd_to_codec(rtd, 0)->component;
+	int ret;
+
+	ret = snd_soc_card_jack_new(rtd->card, "HDMI Jack", SND_JACK_LINEOUT,
+				    &vc4_hdmi->hdmi_jack);
+	if (ret) {
+		dev_err(rtd->dev, "HDMI Jack creation failed: %d\n", ret);
+		return ret;
+	}
+
+	return snd_soc_component_set_jack(component, &vc4_hdmi->hdmi_jack, NULL);
+}
+
 static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 {
 	const struct vc4_hdmi_register *mai_data =
@@ -2288,6 +2305,7 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 	dai_link->cpus->dai_name = dev_name(dev);
 	dai_link->codecs->name = dev_name(&vc4_hdmi->connector.hdmi_audio.codec_pdev->dev);
 	dai_link->platforms->name = dev_name(dev);
+	dai_link->init = vc4_hdmi_codec_init;
 
 	card->dai_link = dai_link;
 	card->num_links = 1;
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
index e3d989ca302b..a31157c99bee 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.h
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
@@ -4,6 +4,7 @@
 #include <drm/drm_connector.h>
 #include <media/cec.h>
 #include <sound/dmaengine_pcm.h>
+#include <sound/hdmi-codec.h>
 #include <sound/soc.h>
 
 #include "vc4_drv.h"
@@ -211,6 +212,12 @@ struct vc4_hdmi {
 	 * KMS hooks. Protected by @mutex.
 	 */
 	enum hdmi_colorspace output_format;
+
+	/**
+	 * @hdmi_jack: Represents the connection state of the HDMI plug, for
+	 * ALSA jack detection.
+	 */
+	struct snd_soc_jack hdmi_jack;
 };
 
 #define connector_to_vc4_hdmi(_connector)				\
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index c5e84d3494d2..056d344c5411 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -2080,7 +2080,7 @@ static int vc6_plane_mode_set(struct drm_plane *plane,
 			/* HPPF plane 1 */
 			vc4_dlist_write(vc4_state, kernel);
 			/* VPPF plane 1 */
-				vc4_dlist_write(vc4_state, kernel);
+			vc4_dlist_write(vc4_state, kernel);
 		}
 	}
 
diff --git a/drivers/gpu/drm/vgem/vgem_fence.c b/drivers/gpu/drm/vgem/vgem_fence.c
index 37bb1fb58cf9..b611c7c8ca2d 100644
--- a/drivers/gpu/drm/vgem/vgem_fence.c
+++ b/drivers/gpu/drm/vgem/vgem_fence.c
@@ -53,25 +53,10 @@ static void vgem_fence_release(struct dma_fence *base)
 	dma_fence_free(&fence->base);
 }
 
-static void vgem_fence_value_str(struct dma_fence *fence, char *str, int size)
-{
-	snprintf(str, size, "%llu", fence->seqno);
-}
-
-static void vgem_fence_timeline_value_str(struct dma_fence *fence, char *str,
-					  int size)
-{
-	snprintf(str, size, "%llu",
-		 dma_fence_is_signaled(fence) ? fence->seqno : 0);
-}
-
 static const struct dma_fence_ops vgem_fence_ops = {
 	.get_driver_name = vgem_fence_get_driver_name,
 	.get_timeline_name = vgem_fence_get_timeline_name,
 	.release = vgem_fence_release,
-
-	.fence_value_str = vgem_fence_value_str,
-	.timeline_value_str = vgem_fence_timeline_value_str,
 };
 
 static void vgem_fence_timeout(struct timer_list *t)
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index 2d88e390feb4..e32e680c7197 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -128,6 +128,14 @@ static void virtio_gpu_remove(struct virtio_device *vdev)
 	drm_dev_put(dev);
 }
 
+static void virtio_gpu_shutdown(struct virtio_device *vdev)
+{
+	/*
+	 * drm does its own synchronization on shutdown.
+	 * Do nothing here, opt out of device reset.
+	 */
+}
+
 static void virtio_gpu_config_changed(struct virtio_device *vdev)
 {
 	struct drm_device *dev = vdev->priv;
@@ -162,6 +170,7 @@ static struct virtio_driver virtio_gpu_driver = {
 	.id_table = id_table,
 	.probe = virtio_gpu_probe,
 	.remove = virtio_gpu_remove,
+	.shutdown = virtio_gpu_shutdown,
 	.config_changed = virtio_gpu_config_changed
 };
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_fence.c b/drivers/gpu/drm/virtio/virtgpu_fence.c
index f28357dbde35..44c1d8ef3c4d 100644
--- a/drivers/gpu/drm/virtio/virtgpu_fence.c
+++ b/drivers/gpu/drm/virtio/virtgpu_fence.c
@@ -49,26 +49,10 @@ static bool virtio_gpu_fence_signaled(struct dma_fence *f)
 	return false;
 }
 
-static void virtio_gpu_fence_value_str(struct dma_fence *f, char *str, int size)
-{
-	snprintf(str, size, "[%llu, %llu]", f->context, f->seqno);
-}
-
-static void virtio_gpu_timeline_value_str(struct dma_fence *f, char *str,
-					  int size)
-{
-	struct virtio_gpu_fence *fence = to_virtio_gpu_fence(f);
-
-	snprintf(str, size, "%llu",
-		 (u64)atomic64_read(&fence->drv->last_fence_id));
-}
-
 static const struct dma_fence_ops virtio_gpu_fence_ops = {
 	.get_driver_name     = virtio_gpu_get_driver_name,
 	.get_timeline_name   = virtio_gpu_get_timeline_name,
 	.signaled            = virtio_gpu_fence_signaled,
-	.fence_value_str     = virtio_gpu_fence_value_str,
-	.timeline_value_str  = virtio_gpu_timeline_value_str,
 };
 
 struct virtio_gpu_fence *virtio_gpu_fence_alloc(struct virtio_gpu_device *vgdev,
diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c
index dde8fc1a3689..90c99d83c4cf 100644
--- a/drivers/gpu/drm/virtio/virtgpu_gem.c
+++ b/drivers/gpu/drm/virtio/virtgpu_gem.c
@@ -115,13 +115,14 @@ int virtio_gpu_gem_object_open(struct drm_gem_object *obj,
 	if (!vgdev->has_context_init)
 		virtio_gpu_create_context(obj->dev, file);
 
-	objs = virtio_gpu_array_alloc(1);
-	if (!objs)
-		return -ENOMEM;
-	virtio_gpu_array_add_obj(objs, obj);
+	if (vfpriv->context_created) {
+		objs = virtio_gpu_array_alloc(1);
+		if (!objs)
+			return -ENOMEM;
+		virtio_gpu_array_add_obj(objs, obj);
 
-	if (vfpriv->ctx_id)
 		virtio_gpu_cmd_context_attach_resource(vgdev, vfpriv->ctx_id, objs);
+	}
 
 out_notify:
 	virtio_gpu_notify(vgdev);
diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c
index a6f5a78f436a..698ea7adb951 100644
--- a/drivers/gpu/drm/virtio/virtgpu_plane.c
+++ b/drivers/gpu/drm/virtio/virtgpu_plane.c
@@ -366,13 +366,7 @@ static int virtio_gpu_plane_prepare_fb(struct drm_plane *plane,
 		return 0;
 
 	obj = new_state->fb->obj[0];
-	if (obj->import_attach) {
-		ret = virtio_gpu_prepare_imported_obj(plane, new_state, obj);
-		if (ret)
-			return ret;
-	}
-
-	if (bo->dumb || obj->import_attach) {
+	if (bo->dumb || drm_gem_is_imported(obj)) {
 		vgplane_st->fence = virtio_gpu_fence_alloc(vgdev,
 						     vgdev->fence_drv.context,
 						     0);
@@ -380,7 +374,21 @@ static int virtio_gpu_plane_prepare_fb(struct drm_plane *plane,
 			return -ENOMEM;
 	}
 
+	if (drm_gem_is_imported(obj)) {
+		ret = virtio_gpu_prepare_imported_obj(plane, new_state, obj);
+		if (ret)
+			goto err_fence;
+	}
+
 	return 0;
+
+err_fence:
+	if (vgplane_st->fence) {
+		dma_fence_put(&vgplane_st->fence->f);
+		vgplane_st->fence = NULL;
+	}
+
+	return ret;
 }
 
 static void virtio_gpu_cleanup_imported_obj(struct drm_gem_object *obj)
@@ -409,7 +417,7 @@ static void virtio_gpu_plane_cleanup_fb(struct drm_plane *plane,
 	}
 
 	obj = state->fb->obj[0];
-	if (obj->import_attach)
+	if (drm_gem_is_imported(obj))
 		virtio_gpu_cleanup_imported_obj(obj);
 }
 
@@ -500,11 +508,19 @@ static int virtio_drm_get_scanout_buffer(struct drm_plane *plane,
 
 	bo = gem_to_virtio_gpu_obj(plane->state->fb->obj[0]);
 
-	/* Only support mapped shmem bo */
-	if (virtio_gpu_is_vram(bo) || bo->base.base.import_attach || !bo->base.vaddr)
+	if (virtio_gpu_is_vram(bo) || drm_gem_is_imported(&bo->base.base))
 		return -ENODEV;
 
-	iosys_map_set_vaddr(&sb->map[0], bo->base.vaddr);
+	if (bo->base.vaddr) {
+		iosys_map_set_vaddr(&sb->map[0], bo->base.vaddr);
+	} else {
+		struct drm_gem_shmem_object *shmem = &bo->base;
+
+		if (!shmem->pages)
+			return -ENODEV;
+		/* map scanout buffer later */
+		sb->pages = shmem->pages;
+	}
 
 	sb->format = plane->state->fb->format;
 	sb->height = plane->state->fb->height;
diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c
index fe6a0b018571..1118a0250279 100644
--- a/drivers/gpu/drm/virtio/virtgpu_prime.c
+++ b/drivers/gpu/drm/virtio/virtgpu_prime.c
@@ -75,7 +75,6 @@ static void virtgpu_gem_unmap_dma_buf(struct dma_buf_attachment *attach,
 
 static const struct virtio_dma_buf_ops virtgpu_dmabuf_ops =  {
 	.ops = {
-		.cache_sgt_mapping = true,
 		.attach = virtio_dma_buf_attach,
 		.detach = drm_gem_map_detach,
 		.map_dma_buf = virtgpu_gem_map_dma_buf,
@@ -205,16 +204,15 @@ static void virtgpu_dma_buf_free_obj(struct drm_gem_object *obj)
 {
 	struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj);
 	struct virtio_gpu_device *vgdev = obj->dev->dev_private;
-	struct dma_buf_attachment *attach = obj->import_attach;
 
-	if (attach) {
-		struct dma_buf *dmabuf = attach->dmabuf;
+	if (drm_gem_is_imported(obj)) {
+		struct dma_buf *dmabuf = obj->dma_buf;
 
 		dma_resv_lock(dmabuf->resv, NULL);
 		virtgpu_dma_buf_unmap(bo);
 		dma_resv_unlock(dmabuf->resv);
 
-		dma_buf_detach(dmabuf, attach);
+		dma_buf_detach(dmabuf, obj->import_attach);
 		dma_buf_put(dmabuf);
 	}
 
@@ -321,6 +319,7 @@ struct drm_gem_object *virtgpu_gem_prime_import(struct drm_device *dev,
 		return ERR_PTR(-ENOMEM);
 
 	obj = &bo->base.base;
+	obj->resv = buf->resv;
 	obj->funcs = &virtgpu_gem_dma_buf_funcs;
 	drm_gem_private_object_init(dev, obj, buf->size);
 
diff --git a/drivers/gpu/drm/vkms/Kconfig b/drivers/gpu/drm/vkms/Kconfig
index 9def079f685b..3c02f928ffe6 100644
--- a/drivers/gpu/drm/vkms/Kconfig
+++ b/drivers/gpu/drm/vkms/Kconfig
@@ -14,3 +14,18 @@ config DRM_VKMS
 	  a VKMS.
 
 	  If M is selected the module will be called vkms.
+
+config DRM_VKMS_KUNIT_TEST
+	tristate "KUnit tests for VKMS" if !KUNIT_ALL_TESTS
+	depends on DRM_VKMS && KUNIT
+	default KUNIT_ALL_TESTS
+	help
+	  This builds unit tests for VKMS. This option is not useful for
+	  distributions or general kernels, but only for kernel
+	  developers working on VKMS.
+
+	  For more information on KUnit and unit tests in general,
+	  please refer to the KUnit documentation in
+	  Documentation/dev-tools/kunit/.
+
+	  If in doubt, say "N".
diff --git a/drivers/gpu/drm/vkms/Makefile b/drivers/gpu/drm/vkms/Makefile
index 1b28a6a32948..d657865e573f 100644
--- a/drivers/gpu/drm/vkms/Makefile
+++ b/drivers/gpu/drm/vkms/Makefile
@@ -6,6 +6,9 @@ vkms-y := \
 	vkms_formats.o \
 	vkms_crtc.o \
 	vkms_composer.o \
-	vkms_writeback.o
+	vkms_writeback.o \
+	vkms_connector.o \
+	vkms_config.o
 
 obj-$(CONFIG_DRM_VKMS) += vkms.o
+obj-$(CONFIG_DRM_VKMS_KUNIT_TEST) += tests/
diff --git a/drivers/gpu/drm/vkms/tests/.kunitconfig b/drivers/gpu/drm/vkms/tests/.kunitconfig
new file mode 100644
index 000000000000..6a2d87068edc
--- /dev/null
+++ b/drivers/gpu/drm/vkms/tests/.kunitconfig
@@ -0,0 +1,4 @@
+CONFIG_KUNIT=y
+CONFIG_DRM=y
+CONFIG_DRM_VKMS=y
+CONFIG_DRM_VKMS_KUNIT_TEST=y
diff --git a/drivers/gpu/drm/vkms/tests/Makefile b/drivers/gpu/drm/vkms/tests/Makefile
new file mode 100644
index 000000000000..9ded37b67a46
--- /dev/null
+++ b/drivers/gpu/drm/vkms/tests/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-$(CONFIG_DRM_VKMS_KUNIT_TEST) += vkms_config_test.o
diff --git a/drivers/gpu/drm/vkms/tests/vkms_config_test.c b/drivers/gpu/drm/vkms/tests/vkms_config_test.c
new file mode 100644
index 000000000000..ff4566cf9925
--- /dev/null
+++ b/drivers/gpu/drm/vkms/tests/vkms_config_test.c
@@ -0,0 +1,951 @@
+// SPDX-License-Identifier: GPL-2.0+
+
+#include <kunit/test.h>
+
+#include "../vkms_config.h"
+
+MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");
+
+static size_t vkms_config_get_num_planes(struct vkms_config *config)
+{
+	struct vkms_config_plane *plane_cfg;
+	size_t count = 0;
+
+	vkms_config_for_each_plane(config, plane_cfg)
+		count++;
+
+	return count;
+}
+
+static size_t vkms_config_get_num_encoders(struct vkms_config *config)
+{
+	struct vkms_config_encoder *encoder_cfg;
+	size_t count = 0;
+
+	vkms_config_for_each_encoder(config, encoder_cfg)
+		count++;
+
+	return count;
+}
+
+static size_t vkms_config_get_num_connectors(struct vkms_config *config)
+{
+	struct vkms_config_connector *connector_cfg;
+	size_t count = 0;
+
+	vkms_config_for_each_connector(config, connector_cfg)
+		count++;
+
+	return count;
+}
+
+static struct vkms_config_plane *get_first_plane(struct vkms_config *config)
+{
+	struct vkms_config_plane *plane_cfg;
+
+	vkms_config_for_each_plane(config, plane_cfg)
+		return plane_cfg;
+
+	return NULL;
+}
+
+static struct vkms_config_crtc *get_first_crtc(struct vkms_config *config)
+{
+	struct vkms_config_crtc *crtc_cfg;
+
+	vkms_config_for_each_crtc(config, crtc_cfg)
+		return crtc_cfg;
+
+	return NULL;
+}
+
+static struct vkms_config_encoder *get_first_encoder(struct vkms_config *config)
+{
+	struct vkms_config_encoder *encoder_cfg;
+
+	vkms_config_for_each_encoder(config, encoder_cfg)
+		return encoder_cfg;
+
+	return NULL;
+}
+
+static struct vkms_config_connector *get_first_connector(struct vkms_config *config)
+{
+	struct vkms_config_connector *connector_cfg;
+
+	vkms_config_for_each_connector(config, connector_cfg)
+		return connector_cfg;
+
+	return NULL;
+}
+
+struct default_config_case {
+	bool enable_cursor;
+	bool enable_writeback;
+	bool enable_overlay;
+};
+
+static void vkms_config_test_empty_config(struct kunit *test)
+{
+	struct vkms_config *config;
+	const char *dev_name = "test";
+
+	config = vkms_config_create(dev_name);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	/* The dev_name string and the config have different lifetimes */
+	dev_name = NULL;
+	KUNIT_EXPECT_STREQ(test, vkms_config_get_device_name(config), "test");
+
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_planes(config), 0);
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_crtcs(config), 0);
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_encoders(config), 0);
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_connectors(config), 0);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static struct default_config_case default_config_cases[] = {
+	{ false, false, false },
+	{ true, false, false },
+	{ true, true, false },
+	{ true, false, true },
+	{ false, true, false },
+	{ false, true, true },
+	{ false, false, true },
+	{ true, true, true },
+};
+
+KUNIT_ARRAY_PARAM(default_config, default_config_cases, NULL);
+
+static void vkms_config_test_default_config(struct kunit *test)
+{
+	const struct default_config_case *params = test->param_value;
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+	int n_primaries = 0;
+	int n_cursors = 0;
+	int n_overlays = 0;
+
+	config = vkms_config_default_create(params->enable_cursor,
+					    params->enable_writeback,
+					    params->enable_overlay);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	/* Planes */
+	vkms_config_for_each_plane(config, plane_cfg) {
+		switch (vkms_config_plane_get_type(plane_cfg)) {
+		case DRM_PLANE_TYPE_PRIMARY:
+			n_primaries++;
+			break;
+		case DRM_PLANE_TYPE_CURSOR:
+			n_cursors++;
+			break;
+		case DRM_PLANE_TYPE_OVERLAY:
+			n_overlays++;
+			break;
+		default:
+			KUNIT_FAIL_AND_ABORT(test, "Unknown plane type");
+		}
+	}
+	KUNIT_EXPECT_EQ(test, n_primaries, 1);
+	KUNIT_EXPECT_EQ(test, n_cursors, params->enable_cursor ? 1 : 0);
+	KUNIT_EXPECT_EQ(test, n_overlays, params->enable_overlay ? 8 : 0);
+
+	/* CRTCs */
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_crtcs(config), 1);
+
+	crtc_cfg = get_first_crtc(config);
+	KUNIT_EXPECT_EQ(test, vkms_config_crtc_get_writeback(crtc_cfg),
+			params->enable_writeback);
+
+	vkms_config_for_each_plane(config, plane_cfg) {
+		struct vkms_config_crtc *possible_crtc;
+		int n_possible_crtcs = 0;
+		unsigned long idx = 0;
+
+		vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) {
+			KUNIT_EXPECT_PTR_EQ(test, crtc_cfg, possible_crtc);
+			n_possible_crtcs++;
+		}
+		KUNIT_EXPECT_EQ(test, n_possible_crtcs, 1);
+	}
+
+	/* Encoders */
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_encoders(config), 1);
+
+	/* Connectors */
+	KUNIT_EXPECT_EQ(test, vkms_config_get_num_connectors(config), 1);
+
+	KUNIT_EXPECT_TRUE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_get_planes(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_plane *plane_cfg1, *plane_cfg2;
+	int n_planes = 0;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	vkms_config_for_each_plane(config, plane_cfg)
+		n_planes++;
+	KUNIT_ASSERT_EQ(test, n_planes, 0);
+
+	plane_cfg1 = vkms_config_create_plane(config);
+	vkms_config_for_each_plane(config, plane_cfg) {
+		n_planes++;
+		if (plane_cfg != plane_cfg1)
+			KUNIT_FAIL(test, "Unexpected plane");
+	}
+	KUNIT_ASSERT_EQ(test, n_planes, 1);
+	n_planes = 0;
+
+	plane_cfg2 = vkms_config_create_plane(config);
+	vkms_config_for_each_plane(config, plane_cfg) {
+		n_planes++;
+		if (plane_cfg != plane_cfg1 && plane_cfg != plane_cfg2)
+			KUNIT_FAIL(test, "Unexpected plane");
+	}
+	KUNIT_ASSERT_EQ(test, n_planes, 2);
+	n_planes = 0;
+
+	vkms_config_destroy_plane(plane_cfg1);
+	vkms_config_for_each_plane(config, plane_cfg) {
+		n_planes++;
+		if (plane_cfg != plane_cfg2)
+			KUNIT_FAIL(test, "Unexpected plane");
+	}
+	KUNIT_ASSERT_EQ(test, n_planes, 1);
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_get_crtcs(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_crtc *crtc_cfg;
+	struct vkms_config_crtc *crtc_cfg1, *crtc_cfg2;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	KUNIT_ASSERT_EQ(test, vkms_config_get_num_crtcs(config), 0);
+	vkms_config_for_each_crtc(config, crtc_cfg)
+		KUNIT_FAIL(test, "Unexpected CRTC");
+
+	crtc_cfg1 = vkms_config_create_crtc(config);
+	KUNIT_ASSERT_EQ(test, vkms_config_get_num_crtcs(config), 1);
+	vkms_config_for_each_crtc(config, crtc_cfg) {
+		if (crtc_cfg != crtc_cfg1)
+			KUNIT_FAIL(test, "Unexpected CRTC");
+	}
+
+	crtc_cfg2 = vkms_config_create_crtc(config);
+	KUNIT_ASSERT_EQ(test, vkms_config_get_num_crtcs(config), 2);
+	vkms_config_for_each_crtc(config, crtc_cfg) {
+		if (crtc_cfg != crtc_cfg1 && crtc_cfg != crtc_cfg2)
+			KUNIT_FAIL(test, "Unexpected CRTC");
+	}
+
+	vkms_config_destroy_crtc(config, crtc_cfg2);
+	KUNIT_ASSERT_EQ(test, vkms_config_get_num_crtcs(config), 1);
+	vkms_config_for_each_crtc(config, crtc_cfg) {
+		if (crtc_cfg != crtc_cfg1)
+			KUNIT_FAIL(test, "Unexpected CRTC");
+	}
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_get_encoders(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_encoder *encoder_cfg;
+	struct vkms_config_encoder *encoder_cfg1, *encoder_cfg2;
+	int n_encoders = 0;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	vkms_config_for_each_encoder(config, encoder_cfg)
+		n_encoders++;
+	KUNIT_ASSERT_EQ(test, n_encoders, 0);
+
+	encoder_cfg1 = vkms_config_create_encoder(config);
+	vkms_config_for_each_encoder(config, encoder_cfg) {
+		n_encoders++;
+		if (encoder_cfg != encoder_cfg1)
+			KUNIT_FAIL(test, "Unexpected encoder");
+	}
+	KUNIT_ASSERT_EQ(test, n_encoders, 1);
+	n_encoders = 0;
+
+	encoder_cfg2 = vkms_config_create_encoder(config);
+	vkms_config_for_each_encoder(config, encoder_cfg) {
+		n_encoders++;
+		if (encoder_cfg != encoder_cfg1 && encoder_cfg != encoder_cfg2)
+			KUNIT_FAIL(test, "Unexpected encoder");
+	}
+	KUNIT_ASSERT_EQ(test, n_encoders, 2);
+	n_encoders = 0;
+
+	vkms_config_destroy_encoder(config, encoder_cfg2);
+	vkms_config_for_each_encoder(config, encoder_cfg) {
+		n_encoders++;
+		if (encoder_cfg != encoder_cfg1)
+			KUNIT_FAIL(test, "Unexpected encoder");
+	}
+	KUNIT_ASSERT_EQ(test, n_encoders, 1);
+	n_encoders = 0;
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_get_connectors(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_connector *connector_cfg;
+	struct vkms_config_connector *connector_cfg1, *connector_cfg2;
+	int n_connectors = 0;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	vkms_config_for_each_connector(config, connector_cfg)
+		n_connectors++;
+	KUNIT_ASSERT_EQ(test, n_connectors, 0);
+
+	connector_cfg1 = vkms_config_create_connector(config);
+	vkms_config_for_each_connector(config, connector_cfg) {
+		n_connectors++;
+		if (connector_cfg != connector_cfg1)
+			KUNIT_FAIL(test, "Unexpected connector");
+	}
+	KUNIT_ASSERT_EQ(test, n_connectors, 1);
+	n_connectors = 0;
+
+	connector_cfg2 = vkms_config_create_connector(config);
+	vkms_config_for_each_connector(config, connector_cfg) {
+		n_connectors++;
+		if (connector_cfg != connector_cfg1 &&
+		    connector_cfg != connector_cfg2)
+			KUNIT_FAIL(test, "Unexpected connector");
+	}
+	KUNIT_ASSERT_EQ(test, n_connectors, 2);
+	n_connectors = 0;
+
+	vkms_config_destroy_connector(connector_cfg2);
+	vkms_config_for_each_connector(config, connector_cfg) {
+		n_connectors++;
+		if (connector_cfg != connector_cfg1)
+			KUNIT_FAIL(test, "Unexpected connector");
+	}
+	KUNIT_ASSERT_EQ(test, n_connectors, 1);
+	n_connectors = 0;
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_invalid_plane_number(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	int n;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	/* Invalid: No planes */
+	plane_cfg = get_first_plane(config);
+	vkms_config_destroy_plane(plane_cfg);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Too many planes */
+	for (n = 0; n <= 32; n++)
+		vkms_config_create_plane(config);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_valid_plane_type(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+	struct vkms_config_encoder *encoder_cfg;
+	int err;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	plane_cfg = get_first_plane(config);
+	vkms_config_destroy_plane(plane_cfg);
+
+	crtc_cfg = get_first_crtc(config);
+
+	/* Invalid: No primary plane */
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_OVERLAY);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Multiple primary planes */
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Valid: One primary plane */
+	vkms_config_destroy_plane(plane_cfg);
+	KUNIT_EXPECT_TRUE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Multiple cursor planes */
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_CURSOR);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_CURSOR);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Valid: One primary and one cursor plane */
+	vkms_config_destroy_plane(plane_cfg);
+	KUNIT_EXPECT_TRUE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Second CRTC without primary plane */
+	crtc_cfg = vkms_config_create_crtc(config);
+	encoder_cfg = vkms_config_create_encoder(config);
+	err = vkms_config_encoder_attach_crtc(encoder_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Valid: Second CRTC with a primary plane */
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_TRUE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_valid_plane_possible_crtcs(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	plane_cfg = get_first_plane(config);
+	crtc_cfg = get_first_crtc(config);
+
+	/* Invalid: Primary plane without a possible CRTC */
+	vkms_config_plane_detach_crtc(plane_cfg, crtc_cfg);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_invalid_crtc_number(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_crtc *crtc_cfg;
+	int n;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	/* Invalid: No CRTCs */
+	crtc_cfg = get_first_crtc(config);
+	vkms_config_destroy_crtc(config, crtc_cfg);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Too many CRTCs */
+	for (n = 0; n <= 32; n++)
+		vkms_config_create_crtc(config);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_invalid_encoder_number(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_encoder *encoder_cfg;
+	int n;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	/* Invalid: No encoders */
+	encoder_cfg = get_first_encoder(config);
+	vkms_config_destroy_encoder(config, encoder_cfg);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Too many encoders */
+	for (n = 0; n <= 32; n++)
+		vkms_config_create_encoder(config);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_valid_encoder_possible_crtcs(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg1, *crtc_cfg2;
+	struct vkms_config_encoder *encoder_cfg;
+	int err;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	crtc_cfg1 = get_first_crtc(config);
+
+	/* Invalid: Encoder without a possible CRTC */
+	encoder_cfg = vkms_config_create_encoder(config);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Valid: Second CRTC with shared encoder */
+	crtc_cfg2 = vkms_config_create_crtc(config);
+
+	plane_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
+	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	err = vkms_config_encoder_attach_crtc(encoder_cfg, crtc_cfg1);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	err = vkms_config_encoder_attach_crtc(encoder_cfg, crtc_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	KUNIT_EXPECT_TRUE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Second CRTC without encoders */
+	vkms_config_encoder_detach_crtc(encoder_cfg, crtc_cfg2);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Valid: First CRTC with 2 possible encoder */
+	vkms_config_destroy_plane(plane_cfg);
+	vkms_config_destroy_crtc(config, crtc_cfg2);
+	KUNIT_EXPECT_TRUE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_invalid_connector_number(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_connector *connector_cfg;
+	int n;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	/* Invalid: No connectors */
+	connector_cfg = get_first_connector(config);
+	vkms_config_destroy_connector(connector_cfg);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	/* Invalid: Too many connectors */
+	for (n = 0; n <= 32; n++)
+		connector_cfg = vkms_config_create_connector(config);
+
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_valid_connector_possible_encoders(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_encoder *encoder_cfg;
+	struct vkms_config_connector *connector_cfg;
+
+	config = vkms_config_default_create(false, false, false);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	encoder_cfg = get_first_encoder(config);
+	connector_cfg = get_first_connector(config);
+
+	/* Invalid: Connector without a possible encoder */
+	vkms_config_connector_detach_encoder(connector_cfg, encoder_cfg);
+	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_attach_different_configs(struct kunit *test)
+{
+	struct vkms_config *config1, *config2;
+	struct vkms_config_plane *plane_cfg1, *plane_cfg2;
+	struct vkms_config_crtc *crtc_cfg1, *crtc_cfg2;
+	struct vkms_config_encoder *encoder_cfg1, *encoder_cfg2;
+	struct vkms_config_connector *connector_cfg1, *connector_cfg2;
+	int err;
+
+	config1 = vkms_config_create("test1");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config1);
+
+	config2 = vkms_config_create("test2");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config2);
+
+	plane_cfg1 = vkms_config_create_plane(config1);
+	crtc_cfg1 = vkms_config_create_crtc(config1);
+	encoder_cfg1 = vkms_config_create_encoder(config1);
+	connector_cfg1 = vkms_config_create_connector(config1);
+
+	plane_cfg2 = vkms_config_create_plane(config2);
+	crtc_cfg2 = vkms_config_create_crtc(config2);
+	encoder_cfg2 = vkms_config_create_encoder(config2);
+	connector_cfg2 = vkms_config_create_connector(config2);
+
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, connector_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, connector_cfg2);
+
+	err = vkms_config_plane_attach_crtc(plane_cfg1, crtc_cfg2);
+	KUNIT_EXPECT_NE(test, err, 0);
+	err = vkms_config_plane_attach_crtc(plane_cfg2, crtc_cfg1);
+	KUNIT_EXPECT_NE(test, err, 0);
+
+	err = vkms_config_encoder_attach_crtc(encoder_cfg1, crtc_cfg2);
+	KUNIT_EXPECT_NE(test, err, 0);
+	err = vkms_config_encoder_attach_crtc(encoder_cfg2, crtc_cfg1);
+	KUNIT_EXPECT_NE(test, err, 0);
+
+	err = vkms_config_connector_attach_encoder(connector_cfg1, encoder_cfg2);
+	KUNIT_EXPECT_NE(test, err, 0);
+	err = vkms_config_connector_attach_encoder(connector_cfg2, encoder_cfg1);
+	KUNIT_EXPECT_NE(test, err, 0);
+
+	vkms_config_destroy(config1);
+	vkms_config_destroy(config2);
+}
+
+static void vkms_config_test_plane_attach_crtc(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *overlay_cfg;
+	struct vkms_config_plane *primary_cfg;
+	struct vkms_config_plane *cursor_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+	int err;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	overlay_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(overlay_cfg, DRM_PLANE_TYPE_OVERLAY);
+	primary_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(primary_cfg, DRM_PLANE_TYPE_PRIMARY);
+	cursor_cfg = vkms_config_create_plane(config);
+	vkms_config_plane_set_type(cursor_cfg, DRM_PLANE_TYPE_CURSOR);
+
+	crtc_cfg = vkms_config_create_crtc(config);
+
+	/* No primary or cursor planes */
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_primary_plane(config, crtc_cfg));
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_cursor_plane(config, crtc_cfg));
+
+	/* Overlay plane, but no primary or cursor planes */
+	err = vkms_config_plane_attach_crtc(overlay_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_primary_plane(config, crtc_cfg));
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_cursor_plane(config, crtc_cfg));
+
+	/* Primary plane, attaching it twice must fail */
+	err = vkms_config_plane_attach_crtc(primary_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	err = vkms_config_plane_attach_crtc(primary_cfg, crtc_cfg);
+	KUNIT_EXPECT_NE(test, err, 0);
+	KUNIT_EXPECT_PTR_EQ(test,
+			    vkms_config_crtc_primary_plane(config, crtc_cfg),
+			    primary_cfg);
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_cursor_plane(config, crtc_cfg));
+
+	/* Primary and cursor planes */
+	err = vkms_config_plane_attach_crtc(cursor_cfg, crtc_cfg);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	KUNIT_EXPECT_PTR_EQ(test,
+			    vkms_config_crtc_primary_plane(config, crtc_cfg),
+			    primary_cfg);
+	KUNIT_EXPECT_PTR_EQ(test,
+			    vkms_config_crtc_cursor_plane(config, crtc_cfg),
+			    cursor_cfg);
+
+	/* Detach primary and destroy cursor plane */
+	vkms_config_plane_detach_crtc(overlay_cfg, crtc_cfg);
+	vkms_config_plane_detach_crtc(primary_cfg, crtc_cfg);
+	vkms_config_destroy_plane(cursor_cfg);
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_primary_plane(config, crtc_cfg));
+	KUNIT_EXPECT_NULL(test, vkms_config_crtc_cursor_plane(config, crtc_cfg));
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_plane_get_possible_crtcs(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg1, *plane_cfg2;
+	struct vkms_config_crtc *crtc_cfg1, *crtc_cfg2;
+	struct vkms_config_crtc *possible_crtc;
+	unsigned long idx = 0;
+	int n_crtcs = 0;
+	int err;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	plane_cfg1 = vkms_config_create_plane(config);
+	plane_cfg2 = vkms_config_create_plane(config);
+	crtc_cfg1 = vkms_config_create_crtc(config);
+	crtc_cfg2 = vkms_config_create_crtc(config);
+
+	/* No possible CRTCs */
+	vkms_config_plane_for_each_possible_crtc(plane_cfg1, idx, possible_crtc)
+		KUNIT_FAIL(test, "Unexpected possible CRTC");
+
+	vkms_config_plane_for_each_possible_crtc(plane_cfg2, idx, possible_crtc)
+		KUNIT_FAIL(test, "Unexpected possible CRTC");
+
+	/* Plane 1 attached to CRTC 1 and 2 */
+	err = vkms_config_plane_attach_crtc(plane_cfg1, crtc_cfg1);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	err = vkms_config_plane_attach_crtc(plane_cfg1, crtc_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	vkms_config_plane_for_each_possible_crtc(plane_cfg1, idx, possible_crtc) {
+		n_crtcs++;
+		if (possible_crtc != crtc_cfg1 && possible_crtc != crtc_cfg2)
+			KUNIT_FAIL(test, "Unexpected possible CRTC");
+	}
+	KUNIT_ASSERT_EQ(test, n_crtcs, 2);
+	n_crtcs = 0;
+
+	vkms_config_plane_for_each_possible_crtc(plane_cfg2, idx, possible_crtc)
+		KUNIT_FAIL(test, "Unexpected possible CRTC");
+
+	/* Plane 1 attached to CRTC 1 and plane 2 to CRTC 2 */
+	vkms_config_plane_detach_crtc(plane_cfg1, crtc_cfg2);
+	vkms_config_plane_for_each_possible_crtc(plane_cfg1, idx, possible_crtc) {
+		n_crtcs++;
+		if (possible_crtc != crtc_cfg1)
+			KUNIT_FAIL(test, "Unexpected possible CRTC");
+	}
+	KUNIT_ASSERT_EQ(test, n_crtcs, 1);
+	n_crtcs = 0;
+
+	err = vkms_config_plane_attach_crtc(plane_cfg2, crtc_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	vkms_config_plane_for_each_possible_crtc(plane_cfg2, idx, possible_crtc) {
+		n_crtcs++;
+		if (possible_crtc != crtc_cfg2)
+			KUNIT_FAIL(test, "Unexpected possible CRTC");
+	}
+	KUNIT_ASSERT_EQ(test, n_crtcs, 1);
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_encoder_get_possible_crtcs(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_encoder *encoder_cfg1, *encoder_cfg2;
+	struct vkms_config_crtc *crtc_cfg1, *crtc_cfg2;
+	struct vkms_config_crtc *possible_crtc;
+	unsigned long idx = 0;
+	int n_crtcs = 0;
+	int err;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	encoder_cfg1 = vkms_config_create_encoder(config);
+	encoder_cfg2 = vkms_config_create_encoder(config);
+	crtc_cfg1 = vkms_config_create_crtc(config);
+	crtc_cfg2 = vkms_config_create_crtc(config);
+
+	/* No possible CRTCs */
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg1, idx, possible_crtc)
+		KUNIT_FAIL(test, "Unexpected possible CRTC");
+
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg2, idx, possible_crtc)
+		KUNIT_FAIL(test, "Unexpected possible CRTC");
+
+	/* Encoder 1 attached to CRTC 1 and 2 */
+	err = vkms_config_encoder_attach_crtc(encoder_cfg1, crtc_cfg1);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	err = vkms_config_encoder_attach_crtc(encoder_cfg1, crtc_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg1, idx, possible_crtc) {
+		n_crtcs++;
+		if (possible_crtc != crtc_cfg1 && possible_crtc != crtc_cfg2)
+			KUNIT_FAIL(test, "Unexpected possible CRTC");
+	}
+	KUNIT_ASSERT_EQ(test, n_crtcs, 2);
+	n_crtcs = 0;
+
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg2, idx, possible_crtc)
+		KUNIT_FAIL(test, "Unexpected possible CRTC");
+
+	/* Encoder 1 attached to CRTC 1 and encoder 2 to CRTC 2 */
+	vkms_config_encoder_detach_crtc(encoder_cfg1, crtc_cfg2);
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg1, idx, possible_crtc) {
+		n_crtcs++;
+		if (possible_crtc != crtc_cfg1)
+			KUNIT_FAIL(test, "Unexpected possible CRTC");
+	}
+	KUNIT_ASSERT_EQ(test, n_crtcs, 1);
+	n_crtcs = 0;
+
+	err = vkms_config_encoder_attach_crtc(encoder_cfg2, crtc_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg2, idx, possible_crtc) {
+		n_crtcs++;
+		if (possible_crtc != crtc_cfg2)
+			KUNIT_FAIL(test, "Unexpected possible CRTC");
+	}
+	KUNIT_ASSERT_EQ(test, n_crtcs, 1);
+
+	vkms_config_destroy(config);
+}
+
+static void vkms_config_test_connector_get_possible_encoders(struct kunit *test)
+{
+	struct vkms_config *config;
+	struct vkms_config_connector *connector_cfg1, *connector_cfg2;
+	struct vkms_config_encoder *encoder_cfg1, *encoder_cfg2;
+	struct vkms_config_encoder *possible_encoder;
+	unsigned long idx = 0;
+	int n_encoders = 0;
+	int err;
+
+	config = vkms_config_create("test");
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
+
+	connector_cfg1 = vkms_config_create_connector(config);
+	connector_cfg2 = vkms_config_create_connector(config);
+	encoder_cfg1 = vkms_config_create_encoder(config);
+	encoder_cfg2 = vkms_config_create_encoder(config);
+
+	/* No possible encoders */
+	vkms_config_connector_for_each_possible_encoder(connector_cfg1, idx,
+							possible_encoder)
+		KUNIT_FAIL(test, "Unexpected possible encoder");
+
+	vkms_config_connector_for_each_possible_encoder(connector_cfg2, idx,
+							possible_encoder)
+		KUNIT_FAIL(test, "Unexpected possible encoder");
+
+	/* Connector 1 attached to encoders 1 and 2 */
+	err = vkms_config_connector_attach_encoder(connector_cfg1, encoder_cfg1);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	err = vkms_config_connector_attach_encoder(connector_cfg1, encoder_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+
+	vkms_config_connector_for_each_possible_encoder(connector_cfg1, idx,
+							possible_encoder) {
+		n_encoders++;
+		if (possible_encoder != encoder_cfg1 &&
+		    possible_encoder != encoder_cfg2)
+			KUNIT_FAIL(test, "Unexpected possible encoder");
+	}
+	KUNIT_ASSERT_EQ(test, n_encoders, 2);
+	n_encoders = 0;
+
+	vkms_config_connector_for_each_possible_encoder(connector_cfg2, idx,
+							possible_encoder)
+		KUNIT_FAIL(test, "Unexpected possible encoder");
+
+	/* Connector 1 attached to encoder 1 and connector 2 to encoder 2 */
+	vkms_config_connector_detach_encoder(connector_cfg1, encoder_cfg2);
+	vkms_config_connector_for_each_possible_encoder(connector_cfg1, idx,
+							possible_encoder) {
+		n_encoders++;
+		if (possible_encoder != encoder_cfg1)
+			KUNIT_FAIL(test, "Unexpected possible encoder");
+	}
+	KUNIT_ASSERT_EQ(test, n_encoders, 1);
+	n_encoders = 0;
+
+	err = vkms_config_connector_attach_encoder(connector_cfg2, encoder_cfg2);
+	KUNIT_EXPECT_EQ(test, err, 0);
+	vkms_config_connector_for_each_possible_encoder(connector_cfg2, idx,
+							possible_encoder) {
+		n_encoders++;
+		if (possible_encoder != encoder_cfg2)
+			KUNIT_FAIL(test, "Unexpected possible encoder");
+	}
+	KUNIT_ASSERT_EQ(test, n_encoders, 1);
+
+	vkms_config_destroy(config);
+}
+
+static struct kunit_case vkms_config_test_cases[] = {
+	KUNIT_CASE(vkms_config_test_empty_config),
+	KUNIT_CASE_PARAM(vkms_config_test_default_config,
+			 default_config_gen_params),
+	KUNIT_CASE(vkms_config_test_get_planes),
+	KUNIT_CASE(vkms_config_test_get_crtcs),
+	KUNIT_CASE(vkms_config_test_get_encoders),
+	KUNIT_CASE(vkms_config_test_get_connectors),
+	KUNIT_CASE(vkms_config_test_invalid_plane_number),
+	KUNIT_CASE(vkms_config_test_valid_plane_type),
+	KUNIT_CASE(vkms_config_test_valid_plane_possible_crtcs),
+	KUNIT_CASE(vkms_config_test_invalid_crtc_number),
+	KUNIT_CASE(vkms_config_test_invalid_encoder_number),
+	KUNIT_CASE(vkms_config_test_valid_encoder_possible_crtcs),
+	KUNIT_CASE(vkms_config_test_invalid_connector_number),
+	KUNIT_CASE(vkms_config_test_valid_connector_possible_encoders),
+	KUNIT_CASE(vkms_config_test_attach_different_configs),
+	KUNIT_CASE(vkms_config_test_plane_attach_crtc),
+	KUNIT_CASE(vkms_config_test_plane_get_possible_crtcs),
+	KUNIT_CASE(vkms_config_test_encoder_get_possible_crtcs),
+	KUNIT_CASE(vkms_config_test_connector_get_possible_encoders),
+	{}
+};
+
+static struct kunit_suite vkms_config_test_suite = {
+	.name = "vkms-config",
+	.test_cases = vkms_config_test_cases,
+};
+
+kunit_test_suite(vkms_config_test_suite);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Kunit test for vkms config utility");
diff --git a/drivers/gpu/drm/vkms/vkms_config.c b/drivers/gpu/drm/vkms/vkms_config.c
new file mode 100644
index 000000000000..a1df5659b0fb
--- /dev/null
+++ b/drivers/gpu/drm/vkms/vkms_config.c
@@ -0,0 +1,640 @@
+// SPDX-License-Identifier: GPL-2.0+
+
+#include <linux/slab.h>
+
+#include <drm/drm_print.h>
+#include <drm/drm_debugfs.h>
+#include <kunit/visibility.h>
+
+#include "vkms_config.h"
+
+struct vkms_config *vkms_config_create(const char *dev_name)
+{
+	struct vkms_config *config;
+
+	config = kzalloc(sizeof(*config), GFP_KERNEL);
+	if (!config)
+		return ERR_PTR(-ENOMEM);
+
+	config->dev_name = kstrdup_const(dev_name, GFP_KERNEL);
+	if (!config->dev_name) {
+		kfree(config);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	INIT_LIST_HEAD(&config->planes);
+	INIT_LIST_HEAD(&config->crtcs);
+	INIT_LIST_HEAD(&config->encoders);
+	INIT_LIST_HEAD(&config->connectors);
+
+	return config;
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_create);
+
+struct vkms_config *vkms_config_default_create(bool enable_cursor,
+					       bool enable_writeback,
+					       bool enable_overlay)
+{
+	struct vkms_config *config;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+	struct vkms_config_encoder *encoder_cfg;
+	struct vkms_config_connector *connector_cfg;
+	int n;
+
+	config = vkms_config_create(DEFAULT_DEVICE_NAME);
+	if (IS_ERR(config))
+		return config;
+
+	plane_cfg = vkms_config_create_plane(config);
+	if (IS_ERR(plane_cfg))
+		goto err_alloc;
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
+
+	crtc_cfg = vkms_config_create_crtc(config);
+	if (IS_ERR(crtc_cfg))
+		goto err_alloc;
+	vkms_config_crtc_set_writeback(crtc_cfg, enable_writeback);
+
+	if (vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg))
+		goto err_alloc;
+
+	if (enable_overlay) {
+		for (n = 0; n < NUM_OVERLAY_PLANES; n++) {
+			plane_cfg = vkms_config_create_plane(config);
+			if (IS_ERR(plane_cfg))
+				goto err_alloc;
+
+			vkms_config_plane_set_type(plane_cfg,
+						   DRM_PLANE_TYPE_OVERLAY);
+
+			if (vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg))
+				goto err_alloc;
+		}
+	}
+
+	if (enable_cursor) {
+		plane_cfg = vkms_config_create_plane(config);
+		if (IS_ERR(plane_cfg))
+			goto err_alloc;
+
+		vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_CURSOR);
+
+		if (vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg))
+			goto err_alloc;
+	}
+
+	encoder_cfg = vkms_config_create_encoder(config);
+	if (IS_ERR(encoder_cfg))
+		goto err_alloc;
+
+	if (vkms_config_encoder_attach_crtc(encoder_cfg, crtc_cfg))
+		goto err_alloc;
+
+	connector_cfg = vkms_config_create_connector(config);
+	if (IS_ERR(connector_cfg))
+		goto err_alloc;
+
+	if (vkms_config_connector_attach_encoder(connector_cfg, encoder_cfg))
+		goto err_alloc;
+
+	return config;
+
+err_alloc:
+	vkms_config_destroy(config);
+	return ERR_PTR(-ENOMEM);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_default_create);
+
+void vkms_config_destroy(struct vkms_config *config)
+{
+	struct vkms_config_plane *plane_cfg, *plane_tmp;
+	struct vkms_config_crtc *crtc_cfg, *crtc_tmp;
+	struct vkms_config_encoder *encoder_cfg, *encoder_tmp;
+	struct vkms_config_connector *connector_cfg, *connector_tmp;
+
+	list_for_each_entry_safe(plane_cfg, plane_tmp, &config->planes, link)
+		vkms_config_destroy_plane(plane_cfg);
+
+	list_for_each_entry_safe(crtc_cfg, crtc_tmp, &config->crtcs, link)
+		vkms_config_destroy_crtc(config, crtc_cfg);
+
+	list_for_each_entry_safe(encoder_cfg, encoder_tmp, &config->encoders, link)
+		vkms_config_destroy_encoder(config, encoder_cfg);
+
+	list_for_each_entry_safe(connector_cfg, connector_tmp, &config->connectors, link)
+		vkms_config_destroy_connector(connector_cfg);
+
+	kfree_const(config->dev_name);
+	kfree(config);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_destroy);
+
+static bool valid_plane_number(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	size_t n_planes;
+
+	n_planes = list_count_nodes((struct list_head *)&config->planes);
+	if (n_planes <= 0 || n_planes >= 32) {
+		drm_info(dev, "The number of planes must be between 1 and 31\n");
+		return false;
+	}
+
+	return true;
+}
+
+static bool valid_planes_for_crtc(const struct vkms_config *config,
+				  struct vkms_config_crtc *crtc_cfg)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	struct vkms_config_plane *plane_cfg;
+	bool has_primary_plane = false;
+	bool has_cursor_plane = false;
+
+	vkms_config_for_each_plane(config, plane_cfg) {
+		struct vkms_config_crtc *possible_crtc;
+		unsigned long idx = 0;
+		enum drm_plane_type type;
+
+		type = vkms_config_plane_get_type(plane_cfg);
+
+		vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) {
+			if (possible_crtc != crtc_cfg)
+				continue;
+
+			if (type == DRM_PLANE_TYPE_PRIMARY) {
+				if (has_primary_plane) {
+					drm_info(dev, "Multiple primary planes\n");
+					return false;
+				}
+
+				has_primary_plane = true;
+			} else if (type == DRM_PLANE_TYPE_CURSOR) {
+				if (has_cursor_plane) {
+					drm_info(dev, "Multiple cursor planes\n");
+					return false;
+				}
+
+				has_cursor_plane = true;
+			}
+		}
+	}
+
+	if (!has_primary_plane) {
+		drm_info(dev, "Primary plane not found\n");
+		return false;
+	}
+
+	return true;
+}
+
+static bool valid_plane_possible_crtcs(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	struct vkms_config_plane *plane_cfg;
+
+	vkms_config_for_each_plane(config, plane_cfg) {
+		if (xa_empty(&plane_cfg->possible_crtcs)) {
+			drm_info(dev, "All planes must have at least one possible CRTC\n");
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static bool valid_crtc_number(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	size_t n_crtcs;
+
+	n_crtcs = list_count_nodes((struct list_head *)&config->crtcs);
+	if (n_crtcs <= 0 || n_crtcs >= 32) {
+		drm_info(dev, "The number of CRTCs must be between 1 and 31\n");
+		return false;
+	}
+
+	return true;
+}
+
+static bool valid_encoder_number(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	size_t n_encoders;
+
+	n_encoders = list_count_nodes((struct list_head *)&config->encoders);
+	if (n_encoders <= 0 || n_encoders >= 32) {
+		drm_info(dev, "The number of encoders must be between 1 and 31\n");
+		return false;
+	}
+
+	return true;
+}
+
+static bool valid_encoder_possible_crtcs(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	struct vkms_config_crtc *crtc_cfg;
+	struct vkms_config_encoder *encoder_cfg;
+
+	vkms_config_for_each_encoder(config, encoder_cfg) {
+		if (xa_empty(&encoder_cfg->possible_crtcs)) {
+			drm_info(dev, "All encoders must have at least one possible CRTC\n");
+			return false;
+		}
+	}
+
+	vkms_config_for_each_crtc(config, crtc_cfg) {
+		bool crtc_has_encoder = false;
+
+		vkms_config_for_each_encoder(config, encoder_cfg) {
+			struct vkms_config_crtc *possible_crtc;
+			unsigned long idx = 0;
+
+			vkms_config_encoder_for_each_possible_crtc(encoder_cfg,
+								   idx, possible_crtc) {
+				if (possible_crtc == crtc_cfg)
+					crtc_has_encoder = true;
+			}
+		}
+
+		if (!crtc_has_encoder) {
+			drm_info(dev, "All CRTCs must have at least one possible encoder\n");
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static bool valid_connector_number(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	size_t n_connectors;
+
+	n_connectors = list_count_nodes((struct list_head *)&config->connectors);
+	if (n_connectors <= 0 || n_connectors >= 32) {
+		drm_info(dev, "The number of connectors must be between 1 and 31\n");
+		return false;
+	}
+
+	return true;
+}
+
+static bool valid_connector_possible_encoders(const struct vkms_config *config)
+{
+	struct drm_device *dev = config->dev ? &config->dev->drm : NULL;
+	struct vkms_config_connector *connector_cfg;
+
+	vkms_config_for_each_connector(config, connector_cfg) {
+		if (xa_empty(&connector_cfg->possible_encoders)) {
+			drm_info(dev,
+				 "All connectors must have at least one possible encoder\n");
+			return false;
+		}
+	}
+
+	return true;
+}
+
+bool vkms_config_is_valid(const struct vkms_config *config)
+{
+	struct vkms_config_crtc *crtc_cfg;
+
+	if (!valid_plane_number(config))
+		return false;
+
+	if (!valid_crtc_number(config))
+		return false;
+
+	if (!valid_encoder_number(config))
+		return false;
+
+	if (!valid_connector_number(config))
+		return false;
+
+	if (!valid_plane_possible_crtcs(config))
+		return false;
+
+	vkms_config_for_each_crtc(config, crtc_cfg) {
+		if (!valid_planes_for_crtc(config, crtc_cfg))
+			return false;
+	}
+
+	if (!valid_encoder_possible_crtcs(config))
+		return false;
+
+	if (!valid_connector_possible_encoders(config))
+		return false;
+
+	return true;
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_is_valid);
+
+static int vkms_config_show(struct seq_file *m, void *data)
+{
+	struct drm_debugfs_entry *entry = m->private;
+	struct drm_device *dev = entry->dev;
+	struct vkms_device *vkmsdev = drm_device_to_vkms_device(dev);
+	const char *dev_name;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+	struct vkms_config_encoder *encoder_cfg;
+	struct vkms_config_connector *connector_cfg;
+
+	dev_name = vkms_config_get_device_name((struct vkms_config *)vkmsdev->config);
+	seq_printf(m, "dev_name=%s\n", dev_name);
+
+	vkms_config_for_each_plane(vkmsdev->config, plane_cfg) {
+		seq_puts(m, "plane:\n");
+		seq_printf(m, "\ttype=%d\n",
+			   vkms_config_plane_get_type(plane_cfg));
+	}
+
+	vkms_config_for_each_crtc(vkmsdev->config, crtc_cfg) {
+		seq_puts(m, "crtc:\n");
+		seq_printf(m, "\twriteback=%d\n",
+			   vkms_config_crtc_get_writeback(crtc_cfg));
+	}
+
+	vkms_config_for_each_encoder(vkmsdev->config, encoder_cfg)
+		seq_puts(m, "encoder\n");
+
+	vkms_config_for_each_connector(vkmsdev->config, connector_cfg)
+		seq_puts(m, "connector\n");
+
+	return 0;
+}
+
+static const struct drm_debugfs_info vkms_config_debugfs_list[] = {
+	{ "vkms_config", vkms_config_show, 0 },
+};
+
+void vkms_config_register_debugfs(struct vkms_device *vkms_device)
+{
+	drm_debugfs_add_files(&vkms_device->drm, vkms_config_debugfs_list,
+			      ARRAY_SIZE(vkms_config_debugfs_list));
+}
+
+struct vkms_config_plane *vkms_config_create_plane(struct vkms_config *config)
+{
+	struct vkms_config_plane *plane_cfg;
+
+	plane_cfg = kzalloc(sizeof(*plane_cfg), GFP_KERNEL);
+	if (!plane_cfg)
+		return ERR_PTR(-ENOMEM);
+
+	plane_cfg->config = config;
+	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_OVERLAY);
+	xa_init_flags(&plane_cfg->possible_crtcs, XA_FLAGS_ALLOC);
+
+	list_add_tail(&plane_cfg->link, &config->planes);
+
+	return plane_cfg;
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_create_plane);
+
+void vkms_config_destroy_plane(struct vkms_config_plane *plane_cfg)
+{
+	xa_destroy(&plane_cfg->possible_crtcs);
+	list_del(&plane_cfg->link);
+	kfree(plane_cfg);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_destroy_plane);
+
+int __must_check vkms_config_plane_attach_crtc(struct vkms_config_plane *plane_cfg,
+					       struct vkms_config_crtc *crtc_cfg)
+{
+	struct vkms_config_crtc *possible_crtc;
+	unsigned long idx = 0;
+	u32 crtc_idx = 0;
+
+	if (plane_cfg->config != crtc_cfg->config)
+		return -EINVAL;
+
+	vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) {
+		if (possible_crtc == crtc_cfg)
+			return -EEXIST;
+	}
+
+	return xa_alloc(&plane_cfg->possible_crtcs, &crtc_idx, crtc_cfg,
+			xa_limit_32b, GFP_KERNEL);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_plane_attach_crtc);
+
+void vkms_config_plane_detach_crtc(struct vkms_config_plane *plane_cfg,
+				   struct vkms_config_crtc *crtc_cfg)
+{
+	struct vkms_config_crtc *possible_crtc;
+	unsigned long idx = 0;
+
+	vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) {
+		if (possible_crtc == crtc_cfg)
+			xa_erase(&plane_cfg->possible_crtcs, idx);
+	}
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_plane_detach_crtc);
+
+struct vkms_config_crtc *vkms_config_create_crtc(struct vkms_config *config)
+{
+	struct vkms_config_crtc *crtc_cfg;
+
+	crtc_cfg = kzalloc(sizeof(*crtc_cfg), GFP_KERNEL);
+	if (!crtc_cfg)
+		return ERR_PTR(-ENOMEM);
+
+	crtc_cfg->config = config;
+	vkms_config_crtc_set_writeback(crtc_cfg, false);
+
+	list_add_tail(&crtc_cfg->link, &config->crtcs);
+
+	return crtc_cfg;
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_create_crtc);
+
+void vkms_config_destroy_crtc(struct vkms_config *config,
+			      struct vkms_config_crtc *crtc_cfg)
+{
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_encoder *encoder_cfg;
+
+	vkms_config_for_each_plane(config, plane_cfg)
+		vkms_config_plane_detach_crtc(plane_cfg, crtc_cfg);
+
+	vkms_config_for_each_encoder(config, encoder_cfg)
+		vkms_config_encoder_detach_crtc(encoder_cfg, crtc_cfg);
+
+	list_del(&crtc_cfg->link);
+	kfree(crtc_cfg);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_destroy_crtc);
+
+/**
+ * vkms_config_crtc_get_plane() - Return the first attached plane to a CRTC with
+ * the specific type
+ * @config: Configuration containing the CRTC and the plane
+ * @crtc_cfg: Only find planes attached to this CRTC
+ * @type: Plane type to search
+ *
+ * Returns:
+ * The first plane found attached to @crtc_cfg with the type @type.
+ */
+static struct vkms_config_plane *vkms_config_crtc_get_plane(const struct vkms_config *config,
+							    struct vkms_config_crtc *crtc_cfg,
+							    enum drm_plane_type type)
+{
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *possible_crtc;
+	enum drm_plane_type current_type;
+	unsigned long idx = 0;
+
+	vkms_config_for_each_plane(config, plane_cfg) {
+		current_type = vkms_config_plane_get_type(plane_cfg);
+
+		vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) {
+			if (possible_crtc == crtc_cfg && current_type == type)
+				return plane_cfg;
+		}
+	}
+
+	return NULL;
+}
+
+struct vkms_config_plane *vkms_config_crtc_primary_plane(const struct vkms_config *config,
+							 struct vkms_config_crtc *crtc_cfg)
+{
+	return vkms_config_crtc_get_plane(config, crtc_cfg, DRM_PLANE_TYPE_PRIMARY);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_crtc_primary_plane);
+
+struct vkms_config_plane *vkms_config_crtc_cursor_plane(const struct vkms_config *config,
+							struct vkms_config_crtc *crtc_cfg)
+{
+	return vkms_config_crtc_get_plane(config, crtc_cfg, DRM_PLANE_TYPE_CURSOR);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_crtc_cursor_plane);
+
+struct vkms_config_encoder *vkms_config_create_encoder(struct vkms_config *config)
+{
+	struct vkms_config_encoder *encoder_cfg;
+
+	encoder_cfg = kzalloc(sizeof(*encoder_cfg), GFP_KERNEL);
+	if (!encoder_cfg)
+		return ERR_PTR(-ENOMEM);
+
+	encoder_cfg->config = config;
+	xa_init_flags(&encoder_cfg->possible_crtcs, XA_FLAGS_ALLOC);
+
+	list_add_tail(&encoder_cfg->link, &config->encoders);
+
+	return encoder_cfg;
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_create_encoder);
+
+void vkms_config_destroy_encoder(struct vkms_config *config,
+				 struct vkms_config_encoder *encoder_cfg)
+{
+	struct vkms_config_connector *connector_cfg;
+
+	vkms_config_for_each_connector(config, connector_cfg)
+		vkms_config_connector_detach_encoder(connector_cfg, encoder_cfg);
+
+	xa_destroy(&encoder_cfg->possible_crtcs);
+	list_del(&encoder_cfg->link);
+	kfree(encoder_cfg);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_destroy_encoder);
+
+int __must_check vkms_config_encoder_attach_crtc(struct vkms_config_encoder *encoder_cfg,
+						 struct vkms_config_crtc *crtc_cfg)
+{
+	struct vkms_config_crtc *possible_crtc;
+	unsigned long idx = 0;
+	u32 crtc_idx = 0;
+
+	if (encoder_cfg->config != crtc_cfg->config)
+		return -EINVAL;
+
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg, idx, possible_crtc) {
+		if (possible_crtc == crtc_cfg)
+			return -EEXIST;
+	}
+
+	return xa_alloc(&encoder_cfg->possible_crtcs, &crtc_idx, crtc_cfg,
+			xa_limit_32b, GFP_KERNEL);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_encoder_attach_crtc);
+
+void vkms_config_encoder_detach_crtc(struct vkms_config_encoder *encoder_cfg,
+				     struct vkms_config_crtc *crtc_cfg)
+{
+	struct vkms_config_crtc *possible_crtc;
+	unsigned long idx = 0;
+
+	vkms_config_encoder_for_each_possible_crtc(encoder_cfg, idx, possible_crtc) {
+		if (possible_crtc == crtc_cfg)
+			xa_erase(&encoder_cfg->possible_crtcs, idx);
+	}
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_encoder_detach_crtc);
+
+struct vkms_config_connector *vkms_config_create_connector(struct vkms_config *config)
+{
+	struct vkms_config_connector *connector_cfg;
+
+	connector_cfg = kzalloc(sizeof(*connector_cfg), GFP_KERNEL);
+	if (!connector_cfg)
+		return ERR_PTR(-ENOMEM);
+
+	connector_cfg->config = config;
+	xa_init_flags(&connector_cfg->possible_encoders, XA_FLAGS_ALLOC);
+
+	list_add_tail(&connector_cfg->link, &config->connectors);
+
+	return connector_cfg;
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_create_connector);
+
+void vkms_config_destroy_connector(struct vkms_config_connector *connector_cfg)
+{
+	xa_destroy(&connector_cfg->possible_encoders);
+	list_del(&connector_cfg->link);
+	kfree(connector_cfg);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_destroy_connector);
+
+int __must_check vkms_config_connector_attach_encoder(struct vkms_config_connector *connector_cfg,
+						      struct vkms_config_encoder *encoder_cfg)
+{
+	struct vkms_config_encoder *possible_encoder;
+	unsigned long idx = 0;
+	u32 encoder_idx = 0;
+
+	if (connector_cfg->config != encoder_cfg->config)
+		return -EINVAL;
+
+	vkms_config_connector_for_each_possible_encoder(connector_cfg, idx,
+							possible_encoder) {
+		if (possible_encoder == encoder_cfg)
+			return -EEXIST;
+	}
+
+	return xa_alloc(&connector_cfg->possible_encoders, &encoder_idx,
+			encoder_cfg, xa_limit_32b, GFP_KERNEL);
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_connector_attach_encoder);
+
+void vkms_config_connector_detach_encoder(struct vkms_config_connector *connector_cfg,
+					  struct vkms_config_encoder *encoder_cfg)
+{
+	struct vkms_config_encoder *possible_encoder;
+	unsigned long idx = 0;
+
+	vkms_config_connector_for_each_possible_encoder(connector_cfg, idx,
+							possible_encoder) {
+		if (possible_encoder == encoder_cfg)
+			xa_erase(&connector_cfg->possible_encoders, idx);
+	}
+}
+EXPORT_SYMBOL_IF_KUNIT(vkms_config_connector_detach_encoder);
diff --git a/drivers/gpu/drm/vkms/vkms_config.h b/drivers/gpu/drm/vkms/vkms_config.h
new file mode 100644
index 000000000000..0118e3f99706
--- /dev/null
+++ b/drivers/gpu/drm/vkms/vkms_config.h
@@ -0,0 +1,437 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+
+#ifndef _VKMS_CONFIG_H_
+#define _VKMS_CONFIG_H_
+
+#include <linux/list.h>
+#include <linux/types.h>
+#include <linux/xarray.h>
+
+#include "vkms_drv.h"
+
+/**
+ * struct vkms_config - General configuration for VKMS driver
+ *
+ * @dev_name: Name of the device
+ * @planes: List of planes configured for the device
+ * @crtcs: List of CRTCs configured for the device
+ * @encoders: List of encoders configured for the device
+ * @connectors: List of connectors configured for the device
+ * @dev: Used to store the current VKMS device. Only set when the device is instantiated.
+ */
+struct vkms_config {
+	const char *dev_name;
+	struct list_head planes;
+	struct list_head crtcs;
+	struct list_head encoders;
+	struct list_head connectors;
+	struct vkms_device *dev;
+};
+
+/**
+ * struct vkms_config_plane
+ *
+ * @link: Link to the others planes in vkms_config
+ * @config: The vkms_config this plane belongs to
+ * @type: Type of the plane. The creator of configuration needs to ensures that
+ *        at least one primary plane is present.
+ * @possible_crtcs: Array of CRTCs that can be used with this plane
+ * @plane: Internal usage. This pointer should never be considered as valid.
+ *         It can be used to store a temporary reference to a VKMS plane during
+ *         device creation. This pointer is not managed by the configuration and
+ *         must be managed by other means.
+ */
+struct vkms_config_plane {
+	struct list_head link;
+	struct vkms_config *config;
+
+	enum drm_plane_type type;
+	struct xarray possible_crtcs;
+
+	/* Internal usage */
+	struct vkms_plane *plane;
+};
+
+/**
+ * struct vkms_config_crtc
+ *
+ * @link: Link to the others CRTCs in vkms_config
+ * @config: The vkms_config this CRTC belongs to
+ * @writeback: If true, a writeback buffer can be attached to the CRTC
+ * @crtc: Internal usage. This pointer should never be considered as valid.
+ *        It can be used to store a temporary reference to a VKMS CRTC during
+ *        device creation. This pointer is not managed by the configuration and
+ *        must be managed by other means.
+ */
+struct vkms_config_crtc {
+	struct list_head link;
+	struct vkms_config *config;
+
+	bool writeback;
+
+	/* Internal usage */
+	struct vkms_output *crtc;
+};
+
+/**
+ * struct vkms_config_encoder
+ *
+ * @link: Link to the others encoders in vkms_config
+ * @config: The vkms_config this CRTC belongs to
+ * @possible_crtcs: Array of CRTCs that can be used with this encoder
+ * @encoder: Internal usage. This pointer should never be considered as valid.
+ *           It can be used to store a temporary reference to a VKMS encoder
+ *           during device creation. This pointer is not managed by the
+ *           configuration and must be managed by other means.
+ */
+struct vkms_config_encoder {
+	struct list_head link;
+	struct vkms_config *config;
+
+	struct xarray possible_crtcs;
+
+	/* Internal usage */
+	struct drm_encoder *encoder;
+};
+
+/**
+ * struct vkms_config_connector
+ *
+ * @link: Link to the others connector in vkms_config
+ * @config: The vkms_config this connector belongs to
+ * @possible_encoders: Array of encoders that can be used with this connector
+ * @connector: Internal usage. This pointer should never be considered as valid.
+ *             It can be used to store a temporary reference to a VKMS connector
+ *             during device creation. This pointer is not managed by the
+ *             configuration and must be managed by other means.
+ */
+struct vkms_config_connector {
+	struct list_head link;
+	struct vkms_config *config;
+
+	struct xarray possible_encoders;
+
+	/* Internal usage */
+	struct vkms_connector *connector;
+};
+
+/**
+ * vkms_config_for_each_plane - Iterate over the vkms_config planes
+ * @config: &struct vkms_config pointer
+ * @plane_cfg: &struct vkms_config_plane pointer used as cursor
+ */
+#define vkms_config_for_each_plane(config, plane_cfg) \
+	list_for_each_entry((plane_cfg), &(config)->planes, link)
+
+/**
+ * vkms_config_for_each_crtc - Iterate over the vkms_config CRTCs
+ * @config: &struct vkms_config pointer
+ * @crtc_cfg: &struct vkms_config_crtc pointer used as cursor
+ */
+#define vkms_config_for_each_crtc(config, crtc_cfg) \
+	list_for_each_entry((crtc_cfg), &(config)->crtcs, link)
+
+/**
+ * vkms_config_for_each_encoder - Iterate over the vkms_config encoders
+ * @config: &struct vkms_config pointer
+ * @encoder_cfg: &struct vkms_config_encoder pointer used as cursor
+ */
+#define vkms_config_for_each_encoder(config, encoder_cfg) \
+	list_for_each_entry((encoder_cfg), &(config)->encoders, link)
+
+/**
+ * vkms_config_for_each_connector - Iterate over the vkms_config connectors
+ * @config: &struct vkms_config pointer
+ * @connector_cfg: &struct vkms_config_connector pointer used as cursor
+ */
+#define vkms_config_for_each_connector(config, connector_cfg) \
+	list_for_each_entry((connector_cfg), &(config)->connectors, link)
+
+/**
+ * vkms_config_plane_for_each_possible_crtc - Iterate over the vkms_config_plane
+ * possible CRTCs
+ * @plane_cfg: &struct vkms_config_plane pointer
+ * @idx: Index of the cursor
+ * @possible_crtc: &struct vkms_config_crtc pointer used as cursor
+ */
+#define vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) \
+	xa_for_each(&(plane_cfg)->possible_crtcs, idx, (possible_crtc))
+
+/**
+ * vkms_config_encoder_for_each_possible_crtc - Iterate over the
+ * vkms_config_encoder possible CRTCs
+ * @encoder_cfg: &struct vkms_config_encoder pointer
+ * @idx: Index of the cursor
+ * @possible_crtc: &struct vkms_config_crtc pointer used as cursor
+ */
+#define vkms_config_encoder_for_each_possible_crtc(encoder_cfg, idx, possible_crtc) \
+	xa_for_each(&(encoder_cfg)->possible_crtcs, idx, (possible_crtc))
+
+/**
+ * vkms_config_connector_for_each_possible_encoder - Iterate over the
+ * vkms_config_connector possible encoders
+ * @connector_cfg: &struct vkms_config_connector pointer
+ * @idx: Index of the cursor
+ * @possible_encoder: &struct vkms_config_encoder pointer used as cursor
+ */
+#define vkms_config_connector_for_each_possible_encoder(connector_cfg, idx, possible_encoder) \
+	xa_for_each(&(connector_cfg)->possible_encoders, idx, (possible_encoder))
+
+/**
+ * vkms_config_create() - Create a new VKMS configuration
+ * @dev_name: Name of the device
+ *
+ * Returns:
+ * The new vkms_config or an error. Call vkms_config_destroy() to free the
+ * returned configuration.
+ */
+struct vkms_config *vkms_config_create(const char *dev_name);
+
+/**
+ * vkms_config_default_create() - Create the configuration for the default device
+ * @enable_cursor: Create or not a cursor plane
+ * @enable_writeback: Create or not a writeback connector
+ * @enable_overlay: Create or not overlay planes
+ *
+ * Returns:
+ * The default vkms_config or an error. Call vkms_config_destroy() to free the
+ * returned configuration.
+ */
+struct vkms_config *vkms_config_default_create(bool enable_cursor,
+					       bool enable_writeback,
+					       bool enable_overlay);
+
+/**
+ * vkms_config_destroy() - Free a VKMS configuration
+ * @config: vkms_config to free
+ */
+void vkms_config_destroy(struct vkms_config *config);
+
+/**
+ * vkms_config_get_device_name() - Return the name of the device
+ * @config: Configuration to get the device name from
+ *
+ * Returns:
+ * The device name. Only valid while @config is valid.
+ */
+static inline const char *
+vkms_config_get_device_name(struct vkms_config *config)
+{
+	return config->dev_name;
+}
+
+/**
+ * vkms_config_get_num_crtcs() - Return the number of CRTCs in the configuration
+ * @config: Configuration to get the number of CRTCs from
+ */
+static inline size_t vkms_config_get_num_crtcs(struct vkms_config *config)
+{
+	return list_count_nodes(&config->crtcs);
+}
+
+/**
+ * vkms_config_is_valid() - Validate a configuration
+ * @config: Configuration to validate
+ *
+ * Returns:
+ * Whether the configuration is valid or not.
+ * For example, a configuration without primary planes is not valid.
+ */
+bool vkms_config_is_valid(const struct vkms_config *config);
+
+/**
+ * vkms_config_register_debugfs() - Register a debugfs file to show the device's
+ * configuration
+ * @vkms_device: Device to register
+ */
+void vkms_config_register_debugfs(struct vkms_device *vkms_device);
+
+/**
+ * vkms_config_create_plane() - Add a new plane configuration
+ * @config: Configuration to add the plane to
+ *
+ * Returns:
+ * The new plane configuration or an error. Call vkms_config_destroy_plane() to
+ * free the returned plane configuration.
+ */
+struct vkms_config_plane *vkms_config_create_plane(struct vkms_config *config);
+
+/**
+ * vkms_config_destroy_plane() - Remove and free a plane configuration
+ * @plane_cfg: Plane configuration to destroy
+ */
+void vkms_config_destroy_plane(struct vkms_config_plane *plane_cfg);
+
+/**
+ * vkms_config_plane_type() - Return the plane type
+ * @plane_cfg: Plane to get the type from
+ */
+static inline enum drm_plane_type
+vkms_config_plane_get_type(struct vkms_config_plane *plane_cfg)
+{
+	return plane_cfg->type;
+}
+
+/**
+ * vkms_config_plane_set_type() - Set the plane type
+ * @plane_cfg: Plane to set the type to
+ * @type: New plane type
+ */
+static inline void
+vkms_config_plane_set_type(struct vkms_config_plane *plane_cfg,
+			   enum drm_plane_type type)
+{
+	plane_cfg->type = type;
+}
+
+/**
+ * vkms_config_plane_attach_crtc - Attach a plane to a CRTC
+ * @plane_cfg: Plane to attach
+ * @crtc_cfg: CRTC to attach @plane_cfg to
+ */
+int __must_check vkms_config_plane_attach_crtc(struct vkms_config_plane *plane_cfg,
+					       struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_plane_detach_crtc - Detach a plane from a CRTC
+ * @plane_cfg: Plane to detach
+ * @crtc_cfg: CRTC to detach @plane_cfg from
+ */
+void vkms_config_plane_detach_crtc(struct vkms_config_plane *plane_cfg,
+				   struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_create_crtc() - Add a new CRTC configuration
+ * @config: Configuration to add the CRTC to
+ *
+ * Returns:
+ * The new CRTC configuration or an error. Call vkms_config_destroy_crtc() to
+ * free the returned CRTC configuration.
+ */
+struct vkms_config_crtc *vkms_config_create_crtc(struct vkms_config *config);
+
+/**
+ * vkms_config_destroy_crtc() - Remove and free a CRTC configuration
+ * @config: Configuration to remove the CRTC from
+ * @crtc_cfg: CRTC configuration to destroy
+ */
+void vkms_config_destroy_crtc(struct vkms_config *config,
+			      struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_crtc_get_writeback() - If a writeback connector will be created
+ * @crtc_cfg: CRTC with or without a writeback connector
+ */
+static inline bool
+vkms_config_crtc_get_writeback(struct vkms_config_crtc *crtc_cfg)
+{
+	return crtc_cfg->writeback;
+}
+
+/**
+ * vkms_config_crtc_set_writeback() - If a writeback connector will be created
+ * @crtc_cfg: Target CRTC
+ * @writeback: Enable or disable the writeback connector
+ */
+static inline void
+vkms_config_crtc_set_writeback(struct vkms_config_crtc *crtc_cfg,
+			       bool writeback)
+{
+	crtc_cfg->writeback = writeback;
+}
+
+/**
+ * vkms_config_crtc_primary_plane() - Return the primary plane for a CRTC
+ * @config: Configuration containing the CRTC
+ * @crtc_config: Target CRTC
+ *
+ * Note that, if multiple primary planes are found, the first one is returned.
+ * In this case, the configuration will be invalid. See vkms_config_is_valid().
+ *
+ * Returns:
+ * The primary plane or NULL if none is assigned yet.
+ */
+struct vkms_config_plane *vkms_config_crtc_primary_plane(const struct vkms_config *config,
+							 struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_crtc_cursor_plane() - Return the cursor plane for a CRTC
+ * @config: Configuration containing the CRTC
+ * @crtc_config: Target CRTC
+ *
+ * Note that, if multiple cursor planes are found, the first one is returned.
+ * In this case, the configuration will be invalid. See vkms_config_is_valid().
+ *
+ * Returns:
+ * The cursor plane or NULL if none is assigned yet.
+ */
+struct vkms_config_plane *vkms_config_crtc_cursor_plane(const struct vkms_config *config,
+							struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_create_encoder() - Add a new encoder configuration
+ * @config: Configuration to add the encoder to
+ *
+ * Returns:
+ * The new encoder configuration or an error. Call vkms_config_destroy_encoder()
+ * to free the returned encoder configuration.
+ */
+struct vkms_config_encoder *vkms_config_create_encoder(struct vkms_config *config);
+
+/**
+ * vkms_config_destroy_encoder() - Remove and free a encoder configuration
+ * @config: Configuration to remove the encoder from
+ * @encoder_cfg: Encoder configuration to destroy
+ */
+void vkms_config_destroy_encoder(struct vkms_config *config,
+				 struct vkms_config_encoder *encoder_cfg);
+
+/**
+ * vkms_config_encoder_attach_crtc - Attach a encoder to a CRTC
+ * @encoder_cfg: Encoder to attach
+ * @crtc_cfg: CRTC to attach @encoder_cfg to
+ */
+int __must_check vkms_config_encoder_attach_crtc(struct vkms_config_encoder *encoder_cfg,
+						 struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_encoder_detach_crtc - Detach a encoder from a CRTC
+ * @encoder_cfg: Encoder to detach
+ * @crtc_cfg: CRTC to detach @encoder_cfg from
+ */
+void vkms_config_encoder_detach_crtc(struct vkms_config_encoder *encoder_cfg,
+				     struct vkms_config_crtc *crtc_cfg);
+
+/**
+ * vkms_config_create_connector() - Add a new connector configuration
+ * @config: Configuration to add the connector to
+ *
+ * Returns:
+ * The new connector configuration or an error. Call
+ * vkms_config_destroy_connector() to free the returned connector configuration.
+ */
+struct vkms_config_connector *vkms_config_create_connector(struct vkms_config *config);
+
+/**
+ * vkms_config_destroy_connector() - Remove and free a connector configuration
+ * @connector_cfg: Connector configuration to destroy
+ */
+void vkms_config_destroy_connector(struct vkms_config_connector *connector_cfg);
+
+/**
+ * vkms_config_connector_attach_encoder - Attach a connector to an encoder
+ * @connector_cfg: Connector to attach
+ * @encoder_cfg: Encoder to attach @connector_cfg to
+ */
+int __must_check vkms_config_connector_attach_encoder(struct vkms_config_connector *connector_cfg,
+						      struct vkms_config_encoder *encoder_cfg);
+
+/**
+ * vkms_config_connector_detach_encoder - Detach a connector from an encoder
+ * @connector_cfg: Connector to detach
+ * @encoder_cfg: Encoder to detach @connector_cfg from
+ */
+void vkms_config_connector_detach_encoder(struct vkms_config_connector *connector_cfg,
+					  struct vkms_config_encoder *encoder_cfg);
+
+#endif /* _VKMS_CONFIG_H_ */
diff --git a/drivers/gpu/drm/vkms/vkms_connector.c b/drivers/gpu/drm/vkms/vkms_connector.c
new file mode 100644
index 000000000000..48b10cba322a
--- /dev/null
+++ b/drivers/gpu/drm/vkms/vkms_connector.c
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0+
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_managed.h>
+#include <drm/drm_probe_helper.h>
+
+#include "vkms_connector.h"
+
+static const struct drm_connector_funcs vkms_connector_funcs = {
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static int vkms_conn_get_modes(struct drm_connector *connector)
+{
+	int count;
+
+	/* Use the default modes list from DRM */
+	count = drm_add_modes_noedid(connector, XRES_MAX, YRES_MAX);
+	drm_set_preferred_mode(connector, XRES_DEF, YRES_DEF);
+
+	return count;
+}
+
+static struct drm_encoder *vkms_conn_best_encoder(struct drm_connector *connector)
+{
+	struct drm_encoder *encoder;
+
+	drm_connector_for_each_possible_encoder(connector, encoder)
+		return encoder;
+
+	return NULL;
+}
+
+static const struct drm_connector_helper_funcs vkms_conn_helper_funcs = {
+	.get_modes    = vkms_conn_get_modes,
+	.best_encoder = vkms_conn_best_encoder,
+};
+
+struct vkms_connector *vkms_connector_init(struct vkms_device *vkmsdev)
+{
+	struct drm_device *dev = &vkmsdev->drm;
+	struct vkms_connector *connector;
+	int ret;
+
+	connector = drmm_kzalloc(dev, sizeof(*connector), GFP_KERNEL);
+	if (!connector)
+		return ERR_PTR(-ENOMEM);
+
+	ret = drmm_connector_init(dev, &connector->base, &vkms_connector_funcs,
+				  DRM_MODE_CONNECTOR_VIRTUAL, NULL);
+	if (ret)
+		return ERR_PTR(ret);
+
+	drm_connector_helper_add(&connector->base, &vkms_conn_helper_funcs);
+
+	return connector;
+}
diff --git a/drivers/gpu/drm/vkms/vkms_connector.h b/drivers/gpu/drm/vkms/vkms_connector.h
new file mode 100644
index 000000000000..c9149c1b7af0
--- /dev/null
+++ b/drivers/gpu/drm/vkms/vkms_connector.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+
+#ifndef _VKMS_CONNECTOR_H_
+#define _VKMS_CONNECTOR_H_
+
+#include "vkms_drv.h"
+
+/**
+ * struct vkms_connector - VKMS custom type wrapping around the DRM connector
+ *
+ * @drm: Base DRM connector
+ */
+struct vkms_connector {
+	struct drm_connector base;
+};
+
+/**
+ * vkms_connector_init() - Initialize a connector
+ * @vkmsdev: VKMS device containing the connector
+ *
+ * Returns:
+ * The connector or an error on failure.
+ */
+struct vkms_connector *vkms_connector_init(struct vkms_device *vkmsdev);
+
+#endif /* _VKMS_CONNECTOR_H_ */
diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c
index 12034ec12029..8c9898b9055d 100644
--- a/drivers/gpu/drm/vkms/vkms_crtc.c
+++ b/drivers/gpu/drm/vkms/vkms_crtc.c
@@ -194,7 +194,7 @@ static int vkms_crtc_atomic_check(struct drm_crtc *crtc,
 		i++;
 	}
 
-	vkms_state->active_planes = kcalloc(i, sizeof(plane), GFP_KERNEL);
+	vkms_state->active_planes = kcalloc(i, sizeof(*vkms_state->active_planes), GFP_KERNEL);
 	if (!vkms_state->active_planes)
 		return -ENOMEM;
 	vkms_state->num_active_planes = i;
diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c
index b6de91134a22..a24d1655f7b8 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.c
+++ b/drivers/gpu/drm/vkms/vkms_drv.c
@@ -27,11 +27,9 @@
 #include <drm/drm_gem_shmem_helper.h>
 #include <drm/drm_vblank.h>
 
+#include "vkms_config.h"
 #include "vkms_drv.h"
 
-#include <drm/drm_print.h>
-#include <drm/drm_debugfs.h>
-
 #define DRIVER_NAME	"vkms"
 #define DRIVER_DESC	"Virtual Kernel Mode Setting"
 #define DRIVER_MAJOR	1
@@ -81,23 +79,6 @@ static void vkms_atomic_commit_tail(struct drm_atomic_state *old_state)
 	drm_atomic_helper_cleanup_planes(dev, old_state);
 }
 
-static int vkms_config_show(struct seq_file *m, void *data)
-{
-	struct drm_debugfs_entry *entry = m->private;
-	struct drm_device *dev = entry->dev;
-	struct vkms_device *vkmsdev = drm_device_to_vkms_device(dev);
-
-	seq_printf(m, "writeback=%d\n", vkmsdev->config->writeback);
-	seq_printf(m, "cursor=%d\n", vkmsdev->config->cursor);
-	seq_printf(m, "overlay=%d\n", vkmsdev->config->overlay);
-
-	return 0;
-}
-
-static const struct drm_debugfs_info vkms_config_debugfs_list[] = {
-	{ "vkms_config", vkms_config_show, 0 },
-};
-
 static const struct drm_driver vkms_driver = {
 	.driver_features	= DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_GEM,
 	.fops			= &vkms_driver_fops,
@@ -170,8 +151,10 @@ static int vkms_create(struct vkms_config *config)
 	int ret;
 	struct platform_device *pdev;
 	struct vkms_device *vkms_device;
+	const char *dev_name;
 
-	pdev = platform_device_register_simple(DRIVER_NAME, -1, NULL, 0);
+	dev_name = vkms_config_get_device_name(config);
+	pdev = platform_device_register_simple(dev_name, -1, NULL, 0);
 	if (IS_ERR(pdev))
 		return PTR_ERR(pdev);
 
@@ -198,7 +181,8 @@ static int vkms_create(struct vkms_config *config)
 		goto out_devres;
 	}
 
-	ret = drm_vblank_init(&vkms_device->drm, 1);
+	ret = drm_vblank_init(&vkms_device->drm,
+			      vkms_config_get_num_crtcs(config));
 	if (ret) {
 		DRM_ERROR("Failed to vblank\n");
 		goto out_devres;
@@ -208,8 +192,7 @@ static int vkms_create(struct vkms_config *config)
 	if (ret)
 		goto out_devres;
 
-	drm_debugfs_add_files(&vkms_device->drm, vkms_config_debugfs_list,
-			      ARRAY_SIZE(vkms_config_debugfs_list));
+	vkms_config_register_debugfs(vkms_device);
 
 	ret = drm_dev_register(&vkms_device->drm, 0);
 	if (ret)
@@ -231,17 +214,13 @@ static int __init vkms_init(void)
 	int ret;
 	struct vkms_config *config;
 
-	config = kmalloc(sizeof(*config), GFP_KERNEL);
-	if (!config)
-		return -ENOMEM;
-
-	config->cursor = enable_cursor;
-	config->writeback = enable_writeback;
-	config->overlay = enable_overlay;
+	config = vkms_config_default_create(enable_cursor, enable_writeback, enable_overlay);
+	if (IS_ERR(config))
+		return PTR_ERR(config);
 
 	ret = vkms_create(config);
 	if (ret) {
-		kfree(config);
+		vkms_config_destroy(config);
 		return ret;
 	}
 
@@ -275,7 +254,7 @@ static void __exit vkms_exit(void)
 		return;
 
 	vkms_destroy(default_config);
-	kfree(default_config);
+	vkms_config_destroy(default_config);
 }
 
 module_init(vkms_init);
diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h
index abbb652be2b5..a74a7fc3a056 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.h
+++ b/drivers/gpu/drm/vkms/vkms_drv.h
@@ -12,6 +12,8 @@
 #include <drm/drm_encoder.h>
 #include <drm/drm_writeback.h>
 
+#define DEFAULT_DEVICE_NAME "vkms"
+
 #define XRES_MIN    10
 #define YRES_MIN    10
 
@@ -189,20 +191,7 @@ struct vkms_output {
 	spinlock_t composer_lock;
 };
 
-/**
- * struct vkms_config - General configuration for VKMS driver
- *
- * @writeback: If true, a writeback buffer can be attached to the CRTC
- * @cursor: If true, a cursor plane is created in the VKMS device
- * @overlay: If true, NUM_OVERLAY_PLANES will be created for the VKMS device
- * @dev: Used to store the current VKMS device. Only set when the device is instantiated.
- */
-struct vkms_config {
-	bool writeback;
-	bool cursor;
-	bool overlay;
-	struct vkms_device *dev;
-};
+struct vkms_config;
 
 /**
  * struct vkms_device - Description of a VKMS device
diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c
index 22f0d678af3a..8d7ca0cdd79f 100644
--- a/drivers/gpu/drm/vkms/vkms_output.c
+++ b/drivers/gpu/drm/vkms/vkms_output.c
@@ -1,121 +1,111 @@
 // SPDX-License-Identifier: GPL-2.0+
 
+#include "vkms_config.h"
+#include "vkms_connector.h"
 #include "vkms_drv.h"
-#include <drm/drm_atomic_helper.h>
-#include <drm/drm_edid.h>
 #include <drm/drm_managed.h>
-#include <drm/drm_probe_helper.h>
-
-static const struct drm_connector_funcs vkms_connector_funcs = {
-	.fill_modes = drm_helper_probe_single_connector_modes,
-	.reset = drm_atomic_helper_connector_reset,
-	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
-};
-
-static int vkms_conn_get_modes(struct drm_connector *connector)
-{
-	int count;
-
-	/* Use the default modes list from DRM */
-	count = drm_add_modes_noedid(connector, XRES_MAX, YRES_MAX);
-	drm_set_preferred_mode(connector, XRES_DEF, YRES_DEF);
-
-	return count;
-}
-
-static const struct drm_connector_helper_funcs vkms_conn_helper_funcs = {
-	.get_modes    = vkms_conn_get_modes,
-};
 
 int vkms_output_init(struct vkms_device *vkmsdev)
 {
 	struct drm_device *dev = &vkmsdev->drm;
-	struct drm_connector *connector;
-	struct drm_encoder *encoder;
-	struct vkms_output *output;
-	struct vkms_plane *primary, *overlay, *cursor = NULL;
+	struct vkms_config_plane *plane_cfg;
+	struct vkms_config_crtc *crtc_cfg;
+	struct vkms_config_encoder *encoder_cfg;
+	struct vkms_config_connector *connector_cfg;
 	int ret;
 	int writeback;
-	unsigned int n;
-
-	/*
-	 * Initialize used plane. One primary plane is required to perform the composition.
-	 *
-	 * The overlay and cursor planes are not mandatory, but can be used to perform complex
-	 * composition.
-	 */
-	primary = vkms_plane_init(vkmsdev, DRM_PLANE_TYPE_PRIMARY);
-	if (IS_ERR(primary))
-		return PTR_ERR(primary);
-
-	if (vkmsdev->config->cursor) {
-		cursor = vkms_plane_init(vkmsdev, DRM_PLANE_TYPE_CURSOR);
-		if (IS_ERR(cursor))
-			return PTR_ERR(cursor);
-	}
 
-	output = vkms_crtc_init(dev, &primary->base,
-				cursor ? &cursor->base : NULL);
-	if (IS_ERR(output)) {
-		DRM_ERROR("Failed to allocate CRTC\n");
-		return PTR_ERR(output);
-	}
+	if (!vkms_config_is_valid(vkmsdev->config))
+		return -EINVAL;
 
-	if (vkmsdev->config->overlay) {
-		for (n = 0; n < NUM_OVERLAY_PLANES; n++) {
-			overlay = vkms_plane_init(vkmsdev, DRM_PLANE_TYPE_OVERLAY);
-			if (IS_ERR(overlay)) {
-				DRM_DEV_ERROR(dev->dev, "Failed to init vkms plane\n");
-				return PTR_ERR(overlay);
-			}
-			overlay->base.possible_crtcs = drm_crtc_mask(&output->crtc);
+	vkms_config_for_each_plane(vkmsdev->config, plane_cfg) {
+		enum drm_plane_type type;
+
+		type = vkms_config_plane_get_type(plane_cfg);
+
+		plane_cfg->plane = vkms_plane_init(vkmsdev, type);
+		if (IS_ERR(plane_cfg->plane)) {
+			DRM_DEV_ERROR(dev->dev, "Failed to init vkms plane\n");
+			return PTR_ERR(plane_cfg->plane);
 		}
 	}
 
-	connector = drmm_kzalloc(dev, sizeof(*connector), GFP_KERNEL);
-	if (!connector) {
-		DRM_ERROR("Failed to allocate connector\n");
-		return -ENOMEM;
-	}
+	vkms_config_for_each_crtc(vkmsdev->config, crtc_cfg) {
+		struct vkms_config_plane *primary, *cursor;
 
-	ret = drmm_connector_init(dev, connector, &vkms_connector_funcs,
-				  DRM_MODE_CONNECTOR_VIRTUAL, NULL);
-	if (ret) {
-		DRM_ERROR("Failed to init connector\n");
-		return ret;
-	}
+		primary = vkms_config_crtc_primary_plane(vkmsdev->config, crtc_cfg);
+		cursor = vkms_config_crtc_cursor_plane(vkmsdev->config, crtc_cfg);
 
-	drm_connector_helper_add(connector, &vkms_conn_helper_funcs);
+		crtc_cfg->crtc = vkms_crtc_init(dev, &primary->plane->base,
+						cursor ? &cursor->plane->base : NULL);
+		if (IS_ERR(crtc_cfg->crtc)) {
+			DRM_ERROR("Failed to allocate CRTC\n");
+			return PTR_ERR(crtc_cfg->crtc);
+		}
 
-	encoder = drmm_kzalloc(dev, sizeof(*encoder), GFP_KERNEL);
-	if (!encoder) {
-		DRM_ERROR("Failed to allocate encoder\n");
-		return -ENOMEM;
+		/* Initialize the writeback component */
+		if (vkms_config_crtc_get_writeback(crtc_cfg)) {
+			writeback = vkms_enable_writeback_connector(vkmsdev, crtc_cfg->crtc);
+			if (writeback)
+				DRM_ERROR("Failed to init writeback connector\n");
+		}
 	}
-	ret = drmm_encoder_init(dev, encoder, NULL,
-				DRM_MODE_ENCODER_VIRTUAL, NULL);
-	if (ret) {
-		DRM_ERROR("Failed to init encoder\n");
-		return ret;
+
+	vkms_config_for_each_plane(vkmsdev->config, plane_cfg) {
+		struct vkms_config_crtc *possible_crtc;
+		unsigned long idx = 0;
+
+		vkms_config_plane_for_each_possible_crtc(plane_cfg, idx, possible_crtc) {
+			plane_cfg->plane->base.possible_crtcs |=
+				drm_crtc_mask(&possible_crtc->crtc->crtc);
+		}
 	}
-	encoder->possible_crtcs = drm_crtc_mask(&output->crtc);
 
-	/* Attach the encoder and the connector */
-	ret = drm_connector_attach_encoder(connector, encoder);
-	if (ret) {
-		DRM_ERROR("Failed to attach connector to encoder\n");
-		return ret;
+	vkms_config_for_each_encoder(vkmsdev->config, encoder_cfg) {
+		struct vkms_config_crtc *possible_crtc;
+		unsigned long idx = 0;
+
+		encoder_cfg->encoder = drmm_kzalloc(dev, sizeof(*encoder_cfg->encoder), GFP_KERNEL);
+		if (!encoder_cfg->encoder) {
+			DRM_ERROR("Failed to allocate encoder\n");
+			return -ENOMEM;
+		}
+		ret = drmm_encoder_init(dev, encoder_cfg->encoder, NULL,
+					DRM_MODE_ENCODER_VIRTUAL, NULL);
+		if (ret) {
+			DRM_ERROR("Failed to init encoder\n");
+			return ret;
+		}
+
+		vkms_config_encoder_for_each_possible_crtc(encoder_cfg, idx, possible_crtc) {
+			encoder_cfg->encoder->possible_crtcs |=
+				drm_crtc_mask(&possible_crtc->crtc->crtc);
+		}
 	}
 
-	/* Initialize the writeback component */
-	if (vkmsdev->config->writeback) {
-		writeback = vkms_enable_writeback_connector(vkmsdev, output);
-		if (writeback)
-			DRM_ERROR("Failed to init writeback connector\n");
+	vkms_config_for_each_connector(vkmsdev->config, connector_cfg) {
+		struct vkms_config_encoder *possible_encoder;
+		unsigned long idx = 0;
+
+		connector_cfg->connector = vkms_connector_init(vkmsdev);
+		if (IS_ERR(connector_cfg->connector)) {
+			DRM_ERROR("Failed to init connector\n");
+			return PTR_ERR(connector_cfg->connector);
+		}
+
+		vkms_config_connector_for_each_possible_encoder(connector_cfg,
+								idx,
+								possible_encoder) {
+			ret = drm_connector_attach_encoder(&connector_cfg->connector->base,
+							   possible_encoder->encoder);
+			if (ret) {
+				DRM_ERROR("Failed to attach connector to encoder\n");
+				return ret;
+			}
+		}
 	}
 
 	drm_mode_config_reset(dev);
 
-	return ret;
+	return 0;
 }
diff --git a/drivers/gpu/drm/vmwgfx/Makefile b/drivers/gpu/drm/vmwgfx/Makefile
index 46a4ab688a7f..b168fd7fe9b3 100644
--- a/drivers/gpu/drm/vmwgfx/Makefile
+++ b/drivers/gpu/drm/vmwgfx/Makefile
@@ -10,6 +10,6 @@ vmwgfx-y := vmwgfx_execbuf.o vmwgfx_gmr.o vmwgfx_kms.o vmwgfx_drv.o \
 	    vmwgfx_simple_resource.o vmwgfx_va.o vmwgfx_blit.o \
 	    vmwgfx_validation.o vmwgfx_page_dirty.o vmwgfx_streamoutput.o \
 	    vmwgfx_devcaps.o ttm_object.o vmwgfx_system_manager.o \
-	    vmwgfx_gem.o vmwgfx_vkms.o
+	    vmwgfx_gem.o vmwgfx_vkms.o vmwgfx_cursor_plane.o
 
 obj-$(CONFIG_DRM_VMWGFX) := vmwgfx.o
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c b/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
index 64bd7d74854e..fa5841fda659 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
@@ -429,7 +429,7 @@ static void *map_external(struct vmw_bo *bo, struct iosys_map *map)
 	void *ptr = NULL;
 	int ret;
 
-	if (bo->tbo.base.import_attach) {
+	if (drm_gem_is_imported(&bo->tbo.base)) {
 		ret = dma_buf_vmap(bo->tbo.base.dma_buf, map);
 		if (ret) {
 			drm_dbg_driver(&vmw->drm,
@@ -447,7 +447,7 @@ out:
 
 static void unmap_external(struct vmw_bo *bo, struct iosys_map *map)
 {
-	if (bo->tbo.base.import_attach)
+	if (drm_gem_is_imported(&bo->tbo.base))
 		dma_buf_vunmap(bo->tbo.base.dma_buf, map);
 	else
 		vmw_bo_unmap(bo);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
index 9b5b8c1f063b..f031a312c783 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c
@@ -36,8 +36,7 @@ static void vmw_bo_release(struct vmw_bo *vbo)
 {
 	struct vmw_resource *res;
 
-	WARN_ON(vbo->tbo.base.funcs &&
-		kref_read(&vbo->tbo.base.refcount) != 0);
+	WARN_ON(kref_read(&vbo->tbo.base.refcount) != 0);
 	vmw_bo_unmap(vbo);
 
 	xa_destroy(&vbo->detached_resources);
@@ -51,11 +50,13 @@ static void vmw_bo_release(struct vmw_bo *vbo)
 			mutex_lock(&res->dev_priv->cmdbuf_mutex);
 			(void)vmw_resource_reserve(res, false, true);
 			vmw_resource_mob_detach(res);
+			if (res->dirty)
+				res->func->dirty_free(res);
 			if (res->coherent)
 				vmw_bo_dirty_release(res->guest_memory_bo);
 			res->guest_memory_bo = NULL;
 			res->guest_memory_offset = 0;
-			vmw_resource_unreserve(res, false, false, false, NULL,
+			vmw_resource_unreserve(res, true, false, false, NULL,
 					       0);
 			mutex_unlock(&res->dev_priv->cmdbuf_mutex);
 		}
@@ -73,9 +74,9 @@ static void vmw_bo_free(struct ttm_buffer_object *bo)
 {
 	struct vmw_bo *vbo = to_vmw_bo(&bo->base);
 
-	WARN_ON(vbo->dirty);
 	WARN_ON(!RB_EMPTY_ROOT(&vbo->res_tree));
 	vmw_bo_release(vbo);
+	WARN_ON(vbo->dirty);
 	kfree(vbo);
 }
 
@@ -467,6 +468,7 @@ int vmw_bo_create(struct vmw_private *vmw,
 	if (unlikely(ret != 0))
 		goto out_error;
 
+	(*p_bo)->tbo.base.funcs = &vmw_gem_object_funcs;
 	return ret;
 out_error:
 	*p_bo = NULL;
@@ -848,9 +850,9 @@ void vmw_bo_placement_set_default_accelerated(struct vmw_bo *bo)
 	vmw_bo_placement_set(bo, domain, domain);
 }
 
-void vmw_bo_add_detached_resource(struct vmw_bo *vbo, struct vmw_resource *res)
+int vmw_bo_add_detached_resource(struct vmw_bo *vbo, struct vmw_resource *res)
 {
-	xa_store(&vbo->detached_resources, (unsigned long)res, res, GFP_KERNEL);
+	return xa_err(xa_store(&vbo->detached_resources, (unsigned long)res, res, GFP_KERNEL));
 }
 
 void vmw_bo_del_detached_resource(struct vmw_bo *vbo, struct vmw_resource *res)
@@ -887,3 +889,9 @@ out:
 		surf = vmw_res_to_srf(res);
 	return surf;
 }
+
+s32 vmw_bo_mobid(struct vmw_bo *vbo)
+{
+	WARN_ON(vbo->tbo.resource->mem_type != VMW_PL_MOB);
+	return (s32)vbo->tbo.resource->start;
+}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h
index 11e330c7c7f5..cf84a163bfcb 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h
@@ -141,7 +141,7 @@ void vmw_bo_move_notify(struct ttm_buffer_object *bo,
 			struct ttm_resource *mem);
 void vmw_bo_swap_notify(struct ttm_buffer_object *bo);
 
-void vmw_bo_add_detached_resource(struct vmw_bo *vbo, struct vmw_resource *res);
+int vmw_bo_add_detached_resource(struct vmw_bo *vbo, struct vmw_resource *res);
 void vmw_bo_del_detached_resource(struct vmw_bo *vbo, struct vmw_resource *res);
 struct vmw_surface *vmw_bo_surface(struct vmw_bo *vbo);
 
@@ -204,12 +204,12 @@ static inline void vmw_bo_unreference(struct vmw_bo **buf)
 
 	*buf = NULL;
 	if (tmp_buf)
-		ttm_bo_put(&tmp_buf->tbo);
+		drm_gem_object_put(&tmp_buf->tbo.base);
 }
 
 static inline struct vmw_bo *vmw_bo_reference(struct vmw_bo *buf)
 {
-	ttm_bo_get(&buf->tbo);
+	drm_gem_object_get(&buf->tbo.base);
 	return buf;
 }
 
@@ -233,4 +233,6 @@ static inline struct vmw_bo *to_vmw_bo(struct drm_gem_object *gobj)
 	return container_of((gobj), struct vmw_bo, tbo.base);
 }
 
+s32 vmw_bo_mobid(struct vmw_bo *vbo);
+
 #endif // VMWGFX_BO_H
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
index a7c07692262b..98331c4c0335 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c
@@ -432,7 +432,7 @@ static int vmw_cotable_resize(struct vmw_resource *res, size_t new_size)
 	 * for the new COTable. Initially pin the buffer object to make sure
 	 * we can use tryreserve without failure.
 	 */
-	ret = vmw_gem_object_create(dev_priv, &bo_params, &buf);
+	ret = vmw_bo_create(dev_priv, &bo_params, &buf);
 	if (ret) {
 		DRM_ERROR("Failed initializing new cotable MOB.\n");
 		goto out_done;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c
new file mode 100644
index 000000000000..718832b08d96
--- /dev/null
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c
@@ -0,0 +1,844 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/**************************************************************************
+ *
+ * Copyright (c) 2024-2025 Broadcom. All Rights Reserved. The term
+ * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
+ *
+ **************************************************************************/
+#include "vmwgfx_cursor_plane.h"
+
+#include "vmwgfx_bo.h"
+#include "vmwgfx_drv.h"
+#include "vmwgfx_kms.h"
+#include "vmwgfx_resource_priv.h"
+#include "vmw_surface_cache.h"
+
+#include "drm/drm_atomic.h"
+#include "drm/drm_atomic_helper.h"
+#include "drm/drm_plane.h"
+#include <asm/page.h>
+
+#define VMW_CURSOR_SNOOP_FORMAT SVGA3D_A8R8G8B8
+#define VMW_CURSOR_SNOOP_WIDTH 64
+#define VMW_CURSOR_SNOOP_HEIGHT 64
+
+struct vmw_svga_fifo_cmd_define_cursor {
+	u32 cmd;
+	SVGAFifoCmdDefineAlphaCursor cursor;
+};
+
+/**
+ * vmw_send_define_cursor_cmd - queue a define cursor command
+ * @dev_priv: the private driver struct
+ * @image: buffer which holds the cursor image
+ * @width: width of the mouse cursor image
+ * @height: height of the mouse cursor image
+ * @hotspotX: the horizontal position of mouse hotspot
+ * @hotspotY: the vertical position of mouse hotspot
+ */
+static void vmw_send_define_cursor_cmd(struct vmw_private *dev_priv,
+				       u32 *image, u32 width, u32 height,
+				       u32 hotspotX, u32 hotspotY)
+{
+	struct vmw_svga_fifo_cmd_define_cursor *cmd;
+	const u32 image_size = width * height * sizeof(*image);
+	const u32 cmd_size = sizeof(*cmd) + image_size;
+
+	/*
+	 * Try to reserve fifocmd space and swallow any failures;
+	 * such reservations cannot be left unconsumed for long
+	 * under the risk of clogging other fifocmd users, so
+	 * we treat reservations separtely from the way we treat
+	 * other fallible KMS-atomic resources at prepare_fb
+	 */
+	cmd = VMW_CMD_RESERVE(dev_priv, cmd_size);
+
+	if (unlikely(!cmd))
+		return;
+
+	memset(cmd, 0, sizeof(*cmd));
+
+	memcpy(&cmd[1], image, image_size);
+
+	cmd->cmd = SVGA_CMD_DEFINE_ALPHA_CURSOR;
+	cmd->cursor.id = 0;
+	cmd->cursor.width = width;
+	cmd->cursor.height = height;
+	cmd->cursor.hotspotX = hotspotX;
+	cmd->cursor.hotspotY = hotspotY;
+
+	vmw_cmd_commit_flush(dev_priv, cmd_size);
+}
+
+static void
+vmw_cursor_plane_update_legacy(struct vmw_private *vmw,
+			       struct vmw_plane_state *vps)
+{
+	struct vmw_surface *surface = vmw_user_object_surface(&vps->uo);
+	s32 hotspot_x = vps->cursor.legacy.hotspot_x + vps->base.hotspot_x;
+	s32 hotspot_y = vps->cursor.legacy.hotspot_y + vps->base.hotspot_y;
+
+	if (WARN_ON(!surface || !surface->snooper.image))
+		return;
+
+	if (vps->cursor.legacy.id != surface->snooper.id) {
+		vmw_send_define_cursor_cmd(vmw, surface->snooper.image,
+					   vps->base.crtc_w, vps->base.crtc_h,
+					   hotspot_x, hotspot_y);
+		vps->cursor.legacy.id = surface->snooper.id;
+	}
+}
+
+static enum vmw_cursor_update_type
+vmw_cursor_update_type(struct vmw_private *vmw, struct vmw_plane_state *vps)
+{
+	struct vmw_surface *surface = vmw_user_object_surface(&vps->uo);
+
+	if (surface && surface->snooper.image)
+		return VMW_CURSOR_UPDATE_LEGACY;
+
+	if (vmw->has_mob) {
+		if ((vmw->capabilities2 & SVGA_CAP2_CURSOR_MOB) != 0)
+			return VMW_CURSOR_UPDATE_MOB;
+	}
+
+	return VMW_CURSOR_UPDATE_NONE;
+}
+
+static void vmw_cursor_update_mob(struct vmw_private *vmw,
+				  struct vmw_plane_state *vps)
+{
+	SVGAGBCursorHeader *header;
+	SVGAGBAlphaCursorHeader *alpha_header;
+	struct vmw_bo *bo = vmw_user_object_buffer(&vps->uo);
+	u32 *image = vmw_bo_map_and_cache(bo);
+	const u32 image_size = vps->base.crtc_w * vps->base.crtc_h * sizeof(*image);
+
+	header = vmw_bo_map_and_cache(vps->cursor.mob);
+	alpha_header = &header->header.alphaHeader;
+
+	memset(header, 0, sizeof(*header));
+
+	header->type = SVGA_ALPHA_CURSOR;
+	header->sizeInBytes = image_size;
+
+	alpha_header->hotspotX = vps->cursor.legacy.hotspot_x + vps->base.hotspot_x;
+	alpha_header->hotspotY = vps->cursor.legacy.hotspot_y + vps->base.hotspot_y;
+	alpha_header->width = vps->base.crtc_w;
+	alpha_header->height = vps->base.crtc_h;
+
+	memcpy(header + 1, image, image_size);
+	vmw_write(vmw, SVGA_REG_CURSOR_MOBID, vmw_bo_mobid(vps->cursor.mob));
+
+	vmw_bo_unmap(bo);
+	vmw_bo_unmap(vps->cursor.mob);
+}
+
+static u32 vmw_cursor_mob_size(enum vmw_cursor_update_type update_type,
+			       u32 w, u32 h)
+{
+	switch (update_type) {
+	case VMW_CURSOR_UPDATE_LEGACY:
+	case VMW_CURSOR_UPDATE_NONE:
+		return 0;
+	case VMW_CURSOR_UPDATE_MOB:
+		return w * h * sizeof(u32) + sizeof(SVGAGBCursorHeader);
+	}
+	return 0;
+}
+
+static void vmw_cursor_mob_destroy(struct vmw_bo **vbo)
+{
+	if (!(*vbo))
+		return;
+
+	ttm_bo_unpin(&(*vbo)->tbo);
+	vmw_bo_unreference(vbo);
+}
+
+/**
+ * vmw_cursor_mob_unmap - Unmaps the cursor mobs.
+ *
+ * @vps: state of the cursor plane
+ *
+ * Returns 0 on success
+ */
+
+static int
+vmw_cursor_mob_unmap(struct vmw_plane_state *vps)
+{
+	int ret = 0;
+	struct vmw_bo *vbo = vps->cursor.mob;
+
+	if (!vbo || !vbo->map.virtual)
+		return 0;
+
+	ret = ttm_bo_reserve(&vbo->tbo, true, false, NULL);
+	if (likely(ret == 0)) {
+		vmw_bo_unmap(vbo);
+		ttm_bo_unreserve(&vbo->tbo);
+	}
+
+	return ret;
+}
+
+static void vmw_cursor_mob_put(struct vmw_cursor_plane *vcp,
+			       struct vmw_plane_state *vps)
+{
+	u32 i;
+
+	if (!vps->cursor.mob)
+		return;
+
+	vmw_cursor_mob_unmap(vps);
+
+	/* Look for a free slot to return this mob to the cache. */
+	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++) {
+		if (!vcp->cursor_mobs[i]) {
+			vcp->cursor_mobs[i] = vps->cursor.mob;
+			vps->cursor.mob = NULL;
+			return;
+		}
+	}
+
+	/* Cache is full: See if this mob is bigger than an existing mob. */
+	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++) {
+		if (vcp->cursor_mobs[i]->tbo.base.size <
+		    vps->cursor.mob->tbo.base.size) {
+			vmw_cursor_mob_destroy(&vcp->cursor_mobs[i]);
+			vcp->cursor_mobs[i] = vps->cursor.mob;
+			vps->cursor.mob = NULL;
+			return;
+		}
+	}
+
+	/* Destroy it if it's not worth caching. */
+	vmw_cursor_mob_destroy(&vps->cursor.mob);
+}
+
+static int vmw_cursor_mob_get(struct vmw_cursor_plane *vcp,
+			      struct vmw_plane_state *vps)
+{
+	struct vmw_private *dev_priv = vmw_priv(vcp->base.dev);
+	u32 size = vmw_cursor_mob_size(vps->cursor.update_type,
+				       vps->base.crtc_w, vps->base.crtc_h);
+	u32 i;
+	u32 cursor_max_dim, mob_max_size;
+	struct vmw_fence_obj *fence = NULL;
+	int ret;
+
+	if (!dev_priv->has_mob ||
+	    (dev_priv->capabilities2 & SVGA_CAP2_CURSOR_MOB) == 0)
+		return -EINVAL;
+
+	mob_max_size = vmw_read(dev_priv, SVGA_REG_MOB_MAX_SIZE);
+	cursor_max_dim = vmw_read(dev_priv, SVGA_REG_CURSOR_MAX_DIMENSION);
+
+	if (size > mob_max_size || vps->base.crtc_w > cursor_max_dim ||
+	    vps->base.crtc_h > cursor_max_dim)
+		return -EINVAL;
+
+	if (vps->cursor.mob) {
+		if (vps->cursor.mob->tbo.base.size >= size)
+			return 0;
+		vmw_cursor_mob_put(vcp, vps);
+	}
+
+	/* Look for an unused mob in the cache. */
+	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++) {
+		if (vcp->cursor_mobs[i] &&
+		    vcp->cursor_mobs[i]->tbo.base.size >= size) {
+			vps->cursor.mob = vcp->cursor_mobs[i];
+			vcp->cursor_mobs[i] = NULL;
+			return 0;
+		}
+	}
+	/* Create a new mob if we can't find an existing one. */
+	ret = vmw_bo_create_and_populate(dev_priv, size, VMW_BO_DOMAIN_MOB,
+					 &vps->cursor.mob);
+
+	if (ret != 0)
+		return ret;
+
+	/* Fence the mob creation so we are guarateed to have the mob */
+	ret = ttm_bo_reserve(&vps->cursor.mob->tbo, false, false, NULL);
+	if (ret != 0)
+		goto teardown;
+
+	ret = vmw_execbuf_fence_commands(NULL, dev_priv, &fence, NULL);
+	if (ret != 0) {
+		ttm_bo_unreserve(&vps->cursor.mob->tbo);
+		goto teardown;
+	}
+
+	dma_fence_wait(&fence->base, false);
+	dma_fence_put(&fence->base);
+
+	ttm_bo_unreserve(&vps->cursor.mob->tbo);
+
+	return 0;
+
+teardown:
+	vmw_cursor_mob_destroy(&vps->cursor.mob);
+	return ret;
+}
+
+static void vmw_cursor_update_position(struct vmw_private *dev_priv,
+				       bool show, int x, int y)
+{
+	const u32 svga_cursor_on = show ? SVGA_CURSOR_ON_SHOW
+				   : SVGA_CURSOR_ON_HIDE;
+	u32 count;
+
+	spin_lock(&dev_priv->cursor_lock);
+	if (dev_priv->capabilities2 & SVGA_CAP2_EXTRA_REGS) {
+		vmw_write(dev_priv, SVGA_REG_CURSOR4_X, x);
+		vmw_write(dev_priv, SVGA_REG_CURSOR4_Y, y);
+		vmw_write(dev_priv, SVGA_REG_CURSOR4_SCREEN_ID, SVGA3D_INVALID_ID);
+		vmw_write(dev_priv, SVGA_REG_CURSOR4_ON, svga_cursor_on);
+		vmw_write(dev_priv, SVGA_REG_CURSOR4_SUBMIT, 1);
+	} else if (vmw_is_cursor_bypass3_enabled(dev_priv)) {
+		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_ON, svga_cursor_on);
+		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_X, x);
+		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_Y, y);
+		count = vmw_fifo_mem_read(dev_priv, SVGA_FIFO_CURSOR_COUNT);
+		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_COUNT, ++count);
+	} else {
+		vmw_write(dev_priv, SVGA_REG_CURSOR_X, x);
+		vmw_write(dev_priv, SVGA_REG_CURSOR_Y, y);
+		vmw_write(dev_priv, SVGA_REG_CURSOR_ON, svga_cursor_on);
+	}
+	spin_unlock(&dev_priv->cursor_lock);
+}
+
+void vmw_kms_cursor_snoop(struct vmw_surface *srf,
+			  struct ttm_object_file *tfile,
+			  struct ttm_buffer_object *bo,
+			  SVGA3dCmdHeader *header)
+{
+	struct ttm_bo_kmap_obj map;
+	unsigned long kmap_offset;
+	unsigned long kmap_num;
+	SVGA3dCopyBox *box;
+	u32 box_count;
+	void *virtual;
+	bool is_iomem;
+	struct vmw_dma_cmd {
+		SVGA3dCmdHeader header;
+		SVGA3dCmdSurfaceDMA dma;
+	} *cmd;
+	int i, ret;
+	const struct SVGA3dSurfaceDesc *desc =
+		vmw_surface_get_desc(VMW_CURSOR_SNOOP_FORMAT);
+	const u32 image_pitch = VMW_CURSOR_SNOOP_WIDTH * desc->pitchBytesPerBlock;
+
+	cmd = container_of(header, struct vmw_dma_cmd, header);
+
+	/* No snooper installed, nothing to copy */
+	if (!srf->snooper.image)
+		return;
+
+	if (cmd->dma.host.face != 0 || cmd->dma.host.mipmap != 0) {
+		DRM_ERROR("face and mipmap for cursors should never != 0\n");
+		return;
+	}
+
+	if (cmd->header.size < 64) {
+		DRM_ERROR("at least one full copy box must be given\n");
+		return;
+	}
+
+	box = (SVGA3dCopyBox *)&cmd[1];
+	box_count = (cmd->header.size - sizeof(SVGA3dCmdSurfaceDMA)) /
+			sizeof(SVGA3dCopyBox);
+
+	if (cmd->dma.guest.ptr.offset % PAGE_SIZE ||
+	    box->x != 0    || box->y != 0    || box->z != 0    ||
+	    box->srcx != 0 || box->srcy != 0 || box->srcz != 0 ||
+	    box->d != 1    || box_count != 1 ||
+	    box->w > VMW_CURSOR_SNOOP_WIDTH || box->h > VMW_CURSOR_SNOOP_HEIGHT) {
+		/* TODO handle none page aligned offsets */
+		/* TODO handle more dst & src != 0 */
+		/* TODO handle more then one copy */
+		DRM_ERROR("Can't snoop dma request for cursor!\n");
+		DRM_ERROR("(%u, %u, %u) (%u, %u, %u) (%ux%ux%u) %u %u\n",
+			  box->srcx, box->srcy, box->srcz,
+			  box->x, box->y, box->z,
+			  box->w, box->h, box->d, box_count,
+			  cmd->dma.guest.ptr.offset);
+		return;
+	}
+
+	kmap_offset = cmd->dma.guest.ptr.offset >> PAGE_SHIFT;
+	kmap_num = (VMW_CURSOR_SNOOP_HEIGHT * image_pitch) >> PAGE_SHIFT;
+
+	ret = ttm_bo_reserve(bo, true, false, NULL);
+	if (unlikely(ret != 0)) {
+		DRM_ERROR("reserve failed\n");
+		return;
+	}
+
+	ret = ttm_bo_kmap(bo, kmap_offset, kmap_num, &map);
+	if (unlikely(ret != 0))
+		goto err_unreserve;
+
+	virtual = ttm_kmap_obj_virtual(&map, &is_iomem);
+
+	if (box->w == VMW_CURSOR_SNOOP_WIDTH && cmd->dma.guest.pitch == image_pitch) {
+		memcpy(srf->snooper.image, virtual,
+		       VMW_CURSOR_SNOOP_HEIGHT * image_pitch);
+	} else {
+		/* Image is unsigned pointer. */
+		for (i = 0; i < box->h; i++)
+			memcpy(srf->snooper.image + i * image_pitch,
+			       virtual + i * cmd->dma.guest.pitch,
+			       box->w * desc->pitchBytesPerBlock);
+	}
+	srf->snooper.id++;
+
+	ttm_bo_kunmap(&map);
+err_unreserve:
+	ttm_bo_unreserve(bo);
+}
+
+void vmw_cursor_plane_destroy(struct drm_plane *plane)
+{
+	struct vmw_cursor_plane *vcp = vmw_plane_to_vcp(plane);
+	u32 i;
+
+	vmw_cursor_update_position(vmw_priv(plane->dev), false, 0, 0);
+
+	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++)
+		vmw_cursor_mob_destroy(&vcp->cursor_mobs[i]);
+
+	drm_plane_cleanup(plane);
+}
+
+/**
+ * vmw_cursor_mob_map - Maps the cursor mobs.
+ *
+ * @vps: plane_state
+ *
+ * Returns 0 on success
+ */
+
+static int
+vmw_cursor_mob_map(struct vmw_plane_state *vps)
+{
+	int ret;
+	u32 size = vmw_cursor_mob_size(vps->cursor.update_type,
+				       vps->base.crtc_w, vps->base.crtc_h);
+	struct vmw_bo *vbo = vps->cursor.mob;
+
+	if (!vbo)
+		return -EINVAL;
+
+	if (vbo->tbo.base.size < size)
+		return -EINVAL;
+
+	if (vbo->map.virtual)
+		return 0;
+
+	ret = ttm_bo_reserve(&vbo->tbo, false, false, NULL);
+	if (unlikely(ret != 0))
+		return -ENOMEM;
+
+	vmw_bo_map_and_cache(vbo);
+
+	ttm_bo_unreserve(&vbo->tbo);
+
+	return 0;
+}
+
+/**
+ * vmw_cursor_plane_cleanup_fb - Unpins the plane surface
+ *
+ * @plane: cursor plane
+ * @old_state: contains the state to clean up
+ *
+ * Unmaps all cursor bo mappings and unpins the cursor surface
+ *
+ * Returns 0 on success
+ */
+void
+vmw_cursor_plane_cleanup_fb(struct drm_plane *plane,
+			    struct drm_plane_state *old_state)
+{
+	struct vmw_cursor_plane *vcp = vmw_plane_to_vcp(plane);
+	struct vmw_plane_state *vps = vmw_plane_state_to_vps(old_state);
+
+	if (!vmw_user_object_is_null(&vps->uo))
+		vmw_user_object_unmap(&vps->uo);
+
+	vmw_cursor_mob_unmap(vps);
+	vmw_cursor_mob_put(vcp, vps);
+
+	vmw_du_plane_unpin_surf(vps);
+	vmw_user_object_unref(&vps->uo);
+}
+
+static bool
+vmw_cursor_buffer_changed(struct vmw_plane_state *new_vps,
+			  struct vmw_plane_state *old_vps)
+{
+	struct vmw_bo *new_bo = vmw_user_object_buffer(&new_vps->uo);
+	struct vmw_bo *old_bo = vmw_user_object_buffer(&old_vps->uo);
+	struct vmw_surface *surf;
+	bool dirty = false;
+	int ret;
+
+	if (new_bo != old_bo)
+		return true;
+
+	if (new_bo) {
+		if (!old_bo) {
+			return true;
+		} else if (new_bo->dirty) {
+			vmw_bo_dirty_scan(new_bo);
+			dirty = vmw_bo_is_dirty(new_bo);
+			if (dirty) {
+				surf = vmw_user_object_surface(&new_vps->uo);
+				if (surf)
+					vmw_bo_dirty_transfer_to_res(&surf->res);
+				else
+					vmw_bo_dirty_clear(new_bo);
+			}
+			return dirty;
+		} else if (new_bo != old_bo) {
+			/*
+			 * Currently unused because the top exits right away.
+			 * In most cases buffer being different will mean
+			 * that the contents is different. For the few percent
+			 * of cases where that's not true the cost of doing
+			 * the memcmp on all other seems to outweight the
+			 * benefits. Leave the conditional to be able to
+			 * trivially validate it by removing the initial
+			 * if (new_bo != old_bo) at the start.
+			 */
+			void *old_image;
+			void *new_image;
+			bool changed = false;
+			struct ww_acquire_ctx ctx;
+			const u32 size = new_vps->base.crtc_w *
+					 new_vps->base.crtc_h * sizeof(u32);
+
+			ww_acquire_init(&ctx, &reservation_ww_class);
+
+			ret = ttm_bo_reserve(&old_bo->tbo, false, false, &ctx);
+			if (ret != 0) {
+				ww_acquire_fini(&ctx);
+				return true;
+			}
+
+			ret = ttm_bo_reserve(&new_bo->tbo, false, false, &ctx);
+			if (ret != 0) {
+				ttm_bo_unreserve(&old_bo->tbo);
+				ww_acquire_fini(&ctx);
+				return true;
+			}
+
+			old_image = vmw_bo_map_and_cache(old_bo);
+			new_image = vmw_bo_map_and_cache(new_bo);
+
+			if (old_image && new_image && old_image != new_image)
+				changed = memcmp(old_image, new_image, size) !=
+					  0;
+
+			ttm_bo_unreserve(&new_bo->tbo);
+			ttm_bo_unreserve(&old_bo->tbo);
+
+			ww_acquire_fini(&ctx);
+
+			return changed;
+		}
+		return false;
+	}
+
+	return false;
+}
+
+static bool
+vmw_cursor_plane_changed(struct vmw_plane_state *new_vps,
+			 struct vmw_plane_state *old_vps)
+{
+	if (old_vps->base.crtc_w != new_vps->base.crtc_w ||
+	    old_vps->base.crtc_h != new_vps->base.crtc_h)
+		return true;
+
+	if (old_vps->base.hotspot_x != new_vps->base.hotspot_x ||
+	    old_vps->base.hotspot_y != new_vps->base.hotspot_y)
+		return true;
+
+	if (old_vps->cursor.legacy.hotspot_x !=
+		    new_vps->cursor.legacy.hotspot_x ||
+	    old_vps->cursor.legacy.hotspot_y !=
+		    new_vps->cursor.legacy.hotspot_y)
+		return true;
+
+	if (old_vps->base.fb != new_vps->base.fb)
+		return true;
+
+	return false;
+}
+
+/**
+ * vmw_cursor_plane_prepare_fb - Readies the cursor by referencing it
+ *
+ * @plane:  display plane
+ * @new_state: info on the new plane state, including the FB
+ *
+ * Returns 0 on success
+ */
+int vmw_cursor_plane_prepare_fb(struct drm_plane *plane,
+				struct drm_plane_state *new_state)
+{
+	struct drm_framebuffer *fb = new_state->fb;
+	struct vmw_cursor_plane *vcp = vmw_plane_to_vcp(plane);
+	struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state);
+	struct vmw_plane_state *old_vps = vmw_plane_state_to_vps(plane->state);
+	struct vmw_private *vmw = vmw_priv(plane->dev);
+	struct vmw_bo *bo = NULL;
+	struct vmw_surface *surface;
+	int ret = 0;
+
+	if (!vmw_user_object_is_null(&vps->uo)) {
+		vmw_user_object_unmap(&vps->uo);
+		vmw_user_object_unref(&vps->uo);
+	}
+
+	if (fb) {
+		if (vmw_framebuffer_to_vfb(fb)->bo) {
+			vps->uo.buffer = vmw_framebuffer_to_vfbd(fb)->buffer;
+			vps->uo.surface = NULL;
+		} else {
+			memcpy(&vps->uo, &vmw_framebuffer_to_vfbs(fb)->uo, sizeof(vps->uo));
+		}
+		vmw_user_object_ref(&vps->uo);
+	}
+
+	vps->cursor.update_type = vmw_cursor_update_type(vmw, vps);
+	switch (vps->cursor.update_type) {
+	case VMW_CURSOR_UPDATE_LEGACY:
+		surface = vmw_user_object_surface(&vps->uo);
+		if (!surface || vps->cursor.legacy.id == surface->snooper.id)
+			vps->cursor.update_type = VMW_CURSOR_UPDATE_NONE;
+		break;
+	case VMW_CURSOR_UPDATE_MOB: {
+		bo = vmw_user_object_buffer(&vps->uo);
+		if (bo) {
+			struct ttm_operation_ctx ctx = { false, false };
+
+			ret = ttm_bo_reserve(&bo->tbo, true, false, NULL);
+			if (ret != 0)
+				return -ENOMEM;
+
+			ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
+			if (ret != 0)
+				return -ENOMEM;
+
+			/*
+			 * vmw_bo_pin_reserved also validates, so to skip
+			 * the extra validation use ttm_bo_pin directly
+			 */
+			if (!bo->tbo.pin_count)
+				ttm_bo_pin(&bo->tbo);
+
+			if (vmw_framebuffer_to_vfb(fb)->bo) {
+				const u32 size = new_state->crtc_w *
+						 new_state->crtc_h *
+						 sizeof(u32);
+
+				(void)vmw_bo_map_and_cache_size(bo, size);
+			} else {
+				vmw_bo_map_and_cache(bo);
+			}
+			ttm_bo_unreserve(&bo->tbo);
+		}
+		if (!vmw_user_object_is_null(&vps->uo)) {
+			if (!vmw_cursor_plane_changed(vps, old_vps) &&
+			    !vmw_cursor_buffer_changed(vps, old_vps)) {
+				vps->cursor.update_type =
+					VMW_CURSOR_UPDATE_NONE;
+			} else {
+				vmw_cursor_mob_get(vcp, vps);
+				vmw_cursor_mob_map(vps);
+			}
+		}
+	}
+		break;
+	case VMW_CURSOR_UPDATE_NONE:
+		/* do nothing */
+		break;
+	}
+
+	return 0;
+}
+
+/**
+ * vmw_cursor_plane_atomic_check - check if the new state is okay
+ *
+ * @plane: cursor plane
+ * @state: info on the new plane state
+ *
+ * This is a chance to fail if the new cursor state does not fit
+ * our requirements.
+ *
+ * Returns 0 on success
+ */
+int vmw_cursor_plane_atomic_check(struct drm_plane *plane,
+				  struct drm_atomic_state *state)
+{
+	struct drm_plane_state *new_state =
+		drm_atomic_get_new_plane_state(state, plane);
+	struct vmw_private *vmw = vmw_priv(plane->dev);
+	int ret = 0;
+	struct drm_crtc_state *crtc_state = NULL;
+	struct vmw_surface *surface = NULL;
+	struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state);
+	enum vmw_cursor_update_type update_type;
+	struct drm_framebuffer *fb = new_state->fb;
+
+	if (new_state->crtc)
+		crtc_state = drm_atomic_get_new_crtc_state(new_state->state,
+							   new_state->crtc);
+
+	ret = drm_atomic_helper_check_plane_state(new_state, crtc_state,
+						  DRM_PLANE_NO_SCALING,
+						  DRM_PLANE_NO_SCALING, true,
+						  true);
+	if (ret)
+		return ret;
+
+	/* Turning off */
+	if (!fb)
+		return 0;
+
+	update_type = vmw_cursor_update_type(vmw, vps);
+	if (update_type == VMW_CURSOR_UPDATE_LEGACY) {
+		if (new_state->crtc_w != VMW_CURSOR_SNOOP_WIDTH ||
+		    new_state->crtc_h != VMW_CURSOR_SNOOP_HEIGHT) {
+			drm_warn(&vmw->drm,
+				 "Invalid cursor dimensions (%d, %d)\n",
+				 new_state->crtc_w, new_state->crtc_h);
+			return -EINVAL;
+		}
+		surface = vmw_user_object_surface(&vps->uo);
+		if (!surface || !surface->snooper.image) {
+			drm_warn(&vmw->drm,
+				 "surface not suitable for cursor\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+void
+vmw_cursor_plane_atomic_update(struct drm_plane *plane,
+			       struct drm_atomic_state *state)
+{
+	struct drm_plane_state *new_state =
+		drm_atomic_get_new_plane_state(state, plane);
+	struct drm_plane_state *old_state =
+		drm_atomic_get_old_plane_state(state, plane);
+	struct drm_crtc *crtc = new_state->crtc ?: old_state->crtc;
+	struct vmw_private *dev_priv = vmw_priv(plane->dev);
+	struct vmw_display_unit *du = vmw_crtc_to_du(crtc);
+	struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state);
+	s32 hotspot_x, hotspot_y, cursor_x, cursor_y;
+
+	/*
+	 * Hide the cursor if the new bo is null
+	 */
+	if (vmw_user_object_is_null(&vps->uo)) {
+		vmw_cursor_update_position(dev_priv, false, 0, 0);
+		return;
+	}
+
+	switch (vps->cursor.update_type) {
+	case VMW_CURSOR_UPDATE_LEGACY:
+		vmw_cursor_plane_update_legacy(dev_priv, vps);
+		break;
+	case VMW_CURSOR_UPDATE_MOB:
+		vmw_cursor_update_mob(dev_priv, vps);
+		break;
+	case VMW_CURSOR_UPDATE_NONE:
+		/* do nothing */
+		break;
+	}
+
+	/*
+	 * For all update types update the cursor position
+	 */
+	cursor_x = new_state->crtc_x + du->set_gui_x;
+	cursor_y = new_state->crtc_y + du->set_gui_y;
+
+	hotspot_x = vps->cursor.legacy.hotspot_x + new_state->hotspot_x;
+	hotspot_y = vps->cursor.legacy.hotspot_y + new_state->hotspot_y;
+
+	vmw_cursor_update_position(dev_priv, true, cursor_x + hotspot_x,
+				   cursor_y + hotspot_y);
+}
+
+int vmw_kms_cursor_bypass_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_priv)
+{
+	struct drm_vmw_cursor_bypass_arg *arg = data;
+	struct vmw_display_unit *du;
+	struct vmw_plane_state *vps;
+	struct drm_crtc *crtc;
+	int ret = 0;
+
+	mutex_lock(&dev->mode_config.mutex);
+	if (arg->flags & DRM_VMW_CURSOR_BYPASS_ALL) {
+		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+			du = vmw_crtc_to_du(crtc);
+			vps = vmw_plane_state_to_vps(du->cursor.base.state);
+			vps->cursor.legacy.hotspot_x = arg->xhot;
+			vps->cursor.legacy.hotspot_y = arg->yhot;
+		}
+
+		mutex_unlock(&dev->mode_config.mutex);
+		return 0;
+	}
+
+	crtc = drm_crtc_find(dev, file_priv, arg->crtc_id);
+	if (!crtc) {
+		ret = -ENOENT;
+		goto out;
+	}
+
+	du = vmw_crtc_to_du(crtc);
+	vps = vmw_plane_state_to_vps(du->cursor.base.state);
+	vps->cursor.legacy.hotspot_x = arg->xhot;
+	vps->cursor.legacy.hotspot_y = arg->yhot;
+
+out:
+	mutex_unlock(&dev->mode_config.mutex);
+
+	return ret;
+}
+
+void *vmw_cursor_snooper_create(struct drm_file *file_priv,
+				struct vmw_surface_metadata *metadata)
+{
+	if (!file_priv->atomic && metadata->scanout &&
+	    metadata->num_sizes == 1 &&
+	    metadata->sizes[0].width == VMW_CURSOR_SNOOP_WIDTH &&
+	    metadata->sizes[0].height == VMW_CURSOR_SNOOP_HEIGHT &&
+	    metadata->format == VMW_CURSOR_SNOOP_FORMAT) {
+		const struct SVGA3dSurfaceDesc *desc =
+			vmw_surface_get_desc(VMW_CURSOR_SNOOP_FORMAT);
+		const u32 cursor_size_bytes = VMW_CURSOR_SNOOP_WIDTH *
+					      VMW_CURSOR_SNOOP_HEIGHT *
+					      desc->pitchBytesPerBlock;
+		void *image = kzalloc(cursor_size_bytes, GFP_KERNEL);
+
+		if (!image) {
+			DRM_ERROR("Failed to allocate cursor_image\n");
+			return ERR_PTR(-ENOMEM);
+		}
+		return image;
+	}
+	return NULL;
+}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.h b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.h
new file mode 100644
index 000000000000..40694925a70e
--- /dev/null
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.h
@@ -0,0 +1,81 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/**************************************************************************
+ *
+ * Copyright (c) 2024-2025 Broadcom. All Rights Reserved. The term
+ * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
+ *
+ **************************************************************************/
+
+#ifndef VMWGFX_CURSOR_PLANE_H
+#define VMWGFX_CURSOR_PLANE_H
+
+#include "device_include/svga3d_cmd.h"
+#include "drm/drm_file.h"
+#include "drm/drm_fourcc.h"
+#include "drm/drm_plane.h"
+
+#include <linux/types.h>
+
+struct SVGA3dCmdHeader;
+struct ttm_buffer_object;
+struct vmw_bo;
+struct vmw_cursor;
+struct vmw_private;
+struct vmw_surface;
+struct vmw_user_object;
+
+#define vmw_plane_to_vcp(x) container_of(x, struct vmw_cursor_plane, base)
+
+static const u32 __maybe_unused vmw_cursor_plane_formats[] = {
+	DRM_FORMAT_ARGB8888,
+};
+
+enum vmw_cursor_update_type {
+	VMW_CURSOR_UPDATE_NONE = 0,
+	VMW_CURSOR_UPDATE_LEGACY,
+	VMW_CURSOR_UPDATE_MOB,
+};
+
+struct vmw_cursor_plane_state {
+	enum vmw_cursor_update_type update_type;
+	bool changed;
+	bool surface_changed;
+	struct vmw_bo *mob;
+	struct {
+		s32 hotspot_x;
+		s32 hotspot_y;
+		u32 id;
+	} legacy;
+};
+
+/**
+ * Derived class for cursor plane object
+ *
+ * @base DRM plane object
+ * @cursor.cursor_mobs Cursor mobs available for re-use
+ */
+struct vmw_cursor_plane {
+	struct drm_plane base;
+
+	struct vmw_bo *cursor_mobs[3];
+};
+
+struct vmw_surface_metadata;
+void *vmw_cursor_snooper_create(struct drm_file *file_priv,
+				struct vmw_surface_metadata *metadata);
+void vmw_cursor_cmd_dma_snoop(SVGA3dCmdHeader *header,
+			      struct vmw_surface *srf,
+			      struct ttm_buffer_object *bo);
+
+void vmw_cursor_plane_destroy(struct drm_plane *plane);
+
+int vmw_cursor_plane_atomic_check(struct drm_plane *plane,
+				  struct drm_atomic_state *state);
+void vmw_cursor_plane_atomic_update(struct drm_plane *plane,
+				    struct drm_atomic_state *state);
+int vmw_cursor_plane_prepare_fb(struct drm_plane *plane,
+				struct drm_plane_state *new_state);
+void vmw_cursor_plane_cleanup_fb(struct drm_plane *plane,
+				 struct drm_plane_state *old_state);
+
+#endif /* VMWGFX_CURSOR_H */
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 0f32471c8533..0695a342b1ef 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -1,31 +1,11 @@
 // SPDX-License-Identifier: GPL-2.0 OR MIT
 /**************************************************************************
  *
- * Copyright 2009-2023 VMware, Inc., Palo Alto, CA., USA
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ * Copyright (c) 2009-2025 Broadcom. All Rights Reserved. The term
+ * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
  **************************************************************************/
 
-
 #include "vmwgfx_drv.h"
 
 #include "vmwgfx_bo.h"
@@ -1324,9 +1304,6 @@ static void vmw_master_set(struct drm_device *dev,
 static void vmw_master_drop(struct drm_device *dev,
 			    struct drm_file *file_priv)
 {
-	struct vmw_private *dev_priv = vmw_priv(dev);
-
-	vmw_kms_legacy_hotspot_clear(dev_priv);
 }
 
 bool vmwgfx_supported(struct vmw_private *vmw)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 5275ef632d4b..594af8eb04c6 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -1,29 +1,9 @@
 /* SPDX-License-Identifier: GPL-2.0 OR MIT */
 /**************************************************************************
  *
- * Copyright (c) 2009-2024 Broadcom. All Rights Reserved. The term
+ * Copyright (c) 2009-2025 Broadcom. All Rights Reserved. The term
  * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
  **************************************************************************/
 
 #ifndef _VMWGFX_DRV_H_
@@ -58,7 +38,7 @@
 
 #define VMWGFX_DRIVER_NAME "vmwgfx"
 #define VMWGFX_DRIVER_MAJOR 2
-#define VMWGFX_DRIVER_MINOR 20
+#define VMWGFX_DRIVER_MINOR 21
 #define VMWGFX_DRIVER_PATCHLEVEL 0
 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024)
 #define VMWGFX_NUM_DISPLAY_UNITS 8
@@ -100,10 +80,6 @@
 #define VMW_RES_SHADER ttm_driver_type4
 #define VMW_RES_HT_ORDER 12
 
-#define VMW_CURSOR_SNOOP_FORMAT SVGA3D_A8R8G8B8
-#define VMW_CURSOR_SNOOP_WIDTH 64
-#define VMW_CURSOR_SNOOP_HEIGHT 64
-
 #define MKSSTAT_CAPACITY_LOG2 5U
 #define MKSSTAT_CAPACITY (1U << MKSSTAT_CAPACITY_LOG2)
 
@@ -201,7 +177,7 @@ enum vmw_cmdbuf_res_type {
 struct vmw_cmdbuf_res_manager;
 
 struct vmw_cursor_snooper {
-	size_t age;
+	size_t id;
 	uint32_t *image;
 };
 
@@ -846,9 +822,7 @@ static inline bool vmw_resource_mob_attached(const struct vmw_resource *res)
  * GEM related functionality - vmwgfx_gem.c
  */
 struct vmw_bo_params;
-int vmw_gem_object_create(struct vmw_private *vmw,
-			  struct vmw_bo_params *params,
-			  struct vmw_bo **p_vbo);
+extern const struct drm_gem_object_funcs vmw_gem_object_funcs;
 extern int vmw_gem_object_create_with_handle(struct vmw_private *dev_priv,
 					     struct drm_file *filp,
 					     uint32_t size,
@@ -1050,7 +1024,6 @@ int vmw_kms_init(struct vmw_private *dev_priv);
 int vmw_kms_close(struct vmw_private *dev_priv);
 int vmw_kms_cursor_bypass_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
-void vmw_kms_cursor_post_execbuf(struct vmw_private *dev_priv);
 void vmw_kms_cursor_snoop(struct vmw_surface *srf,
 			  struct ttm_object_file *tfile,
 			  struct ttm_buffer_object *bo,
@@ -1067,7 +1040,6 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 		    uint32_t num_clips);
 int vmw_kms_update_layout_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
-void vmw_kms_legacy_hotspot_clear(struct vmw_private *dev_priv);
 int vmw_kms_suspend(struct drm_device *dev);
 int vmw_kms_resume(struct drm_device *dev);
 void vmw_kms_lost_device(struct drm_device *dev);
@@ -1393,8 +1365,10 @@ int vmw_mksstat_remove_all(struct vmw_private *dev_priv);
 	DRM_DEBUG_DRIVER(fmt, ##__VA_ARGS__)
 
 /* Resource dirtying - vmwgfx_page_dirty.c */
+bool vmw_bo_is_dirty(struct vmw_bo *vbo);
 void vmw_bo_dirty_scan(struct vmw_bo *vbo);
 int vmw_bo_dirty_add(struct vmw_bo *vbo);
+void vmw_bo_dirty_clear(struct vmw_bo *vbo);
 void vmw_bo_dirty_transfer_to_res(struct vmw_resource *res);
 void vmw_bo_dirty_clear_res(struct vmw_resource *res);
 void vmw_bo_dirty_release(struct vmw_bo *vbo);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index 2e52d73eba48..e831e324e737 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -1,29 +1,11 @@
 // SPDX-License-Identifier: GPL-2.0 OR MIT
 /**************************************************************************
  *
- * Copyright 2009 - 2023 VMware, Inc., Palo Alto, CA., USA
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ * Copyright (c) 2009-2025 Broadcom. All Rights Reserved. The term
+ * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
  **************************************************************************/
+
 #include "vmwgfx_binding.h"
 #include "vmwgfx_bo.h"
 #include "vmwgfx_drv.h"
@@ -4086,6 +4068,23 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv,
 	return 0;
 }
 
+/*
+ * DMA fence callback to remove a seqno_waiter
+ */
+struct seqno_waiter_rm_context {
+	struct dma_fence_cb base;
+	struct vmw_private *dev_priv;
+};
+
+static void seqno_waiter_rm_cb(struct dma_fence *f, struct dma_fence_cb *cb)
+{
+	struct seqno_waiter_rm_context *ctx =
+		container_of(cb, struct seqno_waiter_rm_context, base);
+
+	vmw_seqno_waiter_remove(ctx->dev_priv);
+	kfree(ctx);
+}
+
 int vmw_execbuf_process(struct drm_file *file_priv,
 			struct vmw_private *dev_priv,
 			void __user *user_commands, void *kernel_commands,
@@ -4266,6 +4265,15 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		} else {
 			/* Link the fence with the FD created earlier */
 			fd_install(out_fence_fd, sync_file->file);
+			struct seqno_waiter_rm_context *ctx =
+				kmalloc(sizeof(*ctx), GFP_KERNEL);
+			ctx->dev_priv = dev_priv;
+			vmw_seqno_waiter_add(dev_priv);
+			if (dma_fence_add_callback(&fence->base, &ctx->base,
+						   seqno_waiter_rm_cb) < 0) {
+				vmw_seqno_waiter_remove(dev_priv);
+				kfree(ctx);
+			}
 		}
 	}
 
@@ -4512,8 +4520,6 @@ int vmw_execbuf_ioctl(struct drm_device *dev, void *data,
 	if (unlikely(ret != 0))
 		goto out;
 
-	vmw_kms_cursor_post_execbuf(dev_priv);
-
 out:
 	if (in_fence)
 		dma_fence_put(in_fence);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c
index ed5015ced392..c55382167c1b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c
@@ -84,11 +84,11 @@ static int vmw_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map)
 	struct ttm_buffer_object *bo = drm_gem_ttm_of_gem(obj);
 	int ret;
 
-	if (obj->import_attach) {
-		ret = dma_buf_vmap(obj->import_attach->dmabuf, map);
+	if (drm_gem_is_imported(obj)) {
+		ret = dma_buf_vmap(obj->dma_buf, map);
 		if (!ret) {
 			if (drm_WARN_ON(obj->dev, map->is_iomem)) {
-				dma_buf_vunmap(obj->import_attach->dmabuf, map);
+				dma_buf_vunmap(obj->dma_buf, map);
 				return -EIO;
 			}
 		}
@@ -101,8 +101,8 @@ static int vmw_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map)
 
 static void vmw_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map)
 {
-	if (obj->import_attach)
-		dma_buf_vunmap(obj->import_attach->dmabuf, map);
+	if (drm_gem_is_imported(obj))
+		dma_buf_vunmap(obj->dma_buf, map);
 	else
 		drm_gem_ttm_vunmap(obj, map);
 }
@@ -111,7 +111,7 @@ static int vmw_gem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
 {
 	int ret;
 
-	if (obj->import_attach) {
+	if (drm_gem_is_imported(obj)) {
 		/*
 		 * Reset both vm_ops and vm_private_data, so we don't end up with
 		 * vm_ops pointing to our implementation if the dma-buf backend
@@ -140,7 +140,7 @@ static const struct vm_operations_struct vmw_vm_ops = {
 	.close = ttm_bo_vm_close,
 };
 
-static const struct drm_gem_object_funcs vmw_gem_object_funcs = {
+const struct drm_gem_object_funcs vmw_gem_object_funcs = {
 	.free = vmw_gem_object_free,
 	.open = vmw_gem_object_open,
 	.close = vmw_gem_object_close,
@@ -154,20 +154,6 @@ static const struct drm_gem_object_funcs vmw_gem_object_funcs = {
 	.vm_ops = &vmw_vm_ops,
 };
 
-int vmw_gem_object_create(struct vmw_private *vmw,
-			  struct vmw_bo_params *params,
-			  struct vmw_bo **p_vbo)
-{
-	int ret = vmw_bo_create(vmw, params, p_vbo);
-
-	if (ret != 0)
-		goto out_no_bo;
-
-	(*p_vbo)->tbo.base.funcs = &vmw_gem_object_funcs;
-out_no_bo:
-	return ret;
-}
-
 int vmw_gem_object_create_with_handle(struct vmw_private *dev_priv,
 				      struct drm_file *filp,
 				      uint32_t size,
@@ -183,7 +169,7 @@ int vmw_gem_object_create_with_handle(struct vmw_private *dev_priv,
 		.pin = false
 	};
 
-	ret = vmw_gem_object_create(dev_priv, &params, p_vbo);
+	ret = vmw_bo_create(dev_priv, &params, p_vbo);
 	if (ret != 0)
 		goto out_no_bo;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 1912ac1cde6d..05b1c54a070c 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -1,33 +1,15 @@
 // SPDX-License-Identifier: GPL-2.0 OR MIT
 /**************************************************************************
  *
- * Copyright (c) 2009-2024 Broadcom. All Rights Reserved. The term
+ * Copyright (c) 2009-2025 Broadcom. All Rights Reserved. The term
  * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
  **************************************************************************/
+
 #include "vmwgfx_kms.h"
 
 #include "vmwgfx_bo.h"
+#include "vmwgfx_resource_priv.h"
 #include "vmwgfx_vkms.h"
 #include "vmw_surface_cache.h"
 
@@ -59,474 +41,6 @@ void vmw_du_cleanup(struct vmw_display_unit *du)
 	drm_connector_cleanup(&du->connector);
 }
 
-/*
- * Display Unit Cursor functions
- */
-
-static int vmw_du_cursor_plane_unmap_cm(struct vmw_plane_state *vps);
-static void vmw_cursor_update_mob(struct vmw_private *dev_priv,
-				  struct vmw_plane_state *vps,
-				  u32 *image, u32 width, u32 height,
-				  u32 hotspotX, u32 hotspotY);
-
-struct vmw_svga_fifo_cmd_define_cursor {
-	u32 cmd;
-	SVGAFifoCmdDefineAlphaCursor cursor;
-};
-
-/**
- * vmw_send_define_cursor_cmd - queue a define cursor command
- * @dev_priv: the private driver struct
- * @image: buffer which holds the cursor image
- * @width: width of the mouse cursor image
- * @height: height of the mouse cursor image
- * @hotspotX: the horizontal position of mouse hotspot
- * @hotspotY: the vertical position of mouse hotspot
- */
-static void vmw_send_define_cursor_cmd(struct vmw_private *dev_priv,
-				       u32 *image, u32 width, u32 height,
-				       u32 hotspotX, u32 hotspotY)
-{
-	struct vmw_svga_fifo_cmd_define_cursor *cmd;
-	const u32 image_size = width * height * sizeof(*image);
-	const u32 cmd_size = sizeof(*cmd) + image_size;
-
-	/* Try to reserve fifocmd space and swallow any failures;
-	   such reservations cannot be left unconsumed for long
-	   under the risk of clogging other fifocmd users, so
-	   we treat reservations separtely from the way we treat
-	   other fallible KMS-atomic resources at prepare_fb */
-	cmd = VMW_CMD_RESERVE(dev_priv, cmd_size);
-
-	if (unlikely(!cmd))
-		return;
-
-	memset(cmd, 0, sizeof(*cmd));
-
-	memcpy(&cmd[1], image, image_size);
-
-	cmd->cmd = SVGA_CMD_DEFINE_ALPHA_CURSOR;
-	cmd->cursor.id = 0;
-	cmd->cursor.width = width;
-	cmd->cursor.height = height;
-	cmd->cursor.hotspotX = hotspotX;
-	cmd->cursor.hotspotY = hotspotY;
-
-	vmw_cmd_commit_flush(dev_priv, cmd_size);
-}
-
-/**
- * vmw_cursor_update_image - update the cursor image on the provided plane
- * @dev_priv: the private driver struct
- * @vps: the plane state of the cursor plane
- * @image: buffer which holds the cursor image
- * @width: width of the mouse cursor image
- * @height: height of the mouse cursor image
- * @hotspotX: the horizontal position of mouse hotspot
- * @hotspotY: the vertical position of mouse hotspot
- */
-static void vmw_cursor_update_image(struct vmw_private *dev_priv,
-				    struct vmw_plane_state *vps,
-				    u32 *image, u32 width, u32 height,
-				    u32 hotspotX, u32 hotspotY)
-{
-	if (vps->cursor.bo)
-		vmw_cursor_update_mob(dev_priv, vps, image,
-				      vps->base.crtc_w, vps->base.crtc_h,
-				      hotspotX, hotspotY);
-
-	else
-		vmw_send_define_cursor_cmd(dev_priv, image, width, height,
-					   hotspotX, hotspotY);
-}
-
-
-/**
- * vmw_cursor_update_mob - Update cursor vis CursorMob mechanism
- *
- * Called from inside vmw_du_cursor_plane_atomic_update to actually
- * make the cursor-image live.
- *
- * @dev_priv: device to work with
- * @vps: the plane state of the cursor plane
- * @image: cursor source data to fill the MOB with
- * @width: source data width
- * @height: source data height
- * @hotspotX: cursor hotspot x
- * @hotspotY: cursor hotspot Y
- */
-static void vmw_cursor_update_mob(struct vmw_private *dev_priv,
-				  struct vmw_plane_state *vps,
-				  u32 *image, u32 width, u32 height,
-				  u32 hotspotX, u32 hotspotY)
-{
-	SVGAGBCursorHeader *header;
-	SVGAGBAlphaCursorHeader *alpha_header;
-	const u32 image_size = width * height * sizeof(*image);
-
-	header = vmw_bo_map_and_cache(vps->cursor.bo);
-	alpha_header = &header->header.alphaHeader;
-
-	memset(header, 0, sizeof(*header));
-
-	header->type = SVGA_ALPHA_CURSOR;
-	header->sizeInBytes = image_size;
-
-	alpha_header->hotspotX = hotspotX;
-	alpha_header->hotspotY = hotspotY;
-	alpha_header->width = width;
-	alpha_header->height = height;
-
-	memcpy(header + 1, image, image_size);
-	vmw_write(dev_priv, SVGA_REG_CURSOR_MOBID,
-		  vps->cursor.bo->tbo.resource->start);
-}
-
-
-static u32 vmw_du_cursor_mob_size(u32 w, u32 h)
-{
-	return w * h * sizeof(u32) + sizeof(SVGAGBCursorHeader);
-}
-
-/**
- * vmw_du_cursor_plane_acquire_image -- Acquire the image data
- * @vps: cursor plane state
- */
-static u32 *vmw_du_cursor_plane_acquire_image(struct vmw_plane_state *vps)
-{
-	struct vmw_surface *surf;
-
-	if (vmw_user_object_is_null(&vps->uo))
-		return NULL;
-
-	surf = vmw_user_object_surface(&vps->uo);
-	if (surf && !vmw_user_object_is_mapped(&vps->uo))
-		return surf->snooper.image;
-
-	return vmw_user_object_map(&vps->uo);
-}
-
-static bool vmw_du_cursor_plane_has_changed(struct vmw_plane_state *old_vps,
-					    struct vmw_plane_state *new_vps)
-{
-	void *old_image;
-	void *new_image;
-	u32 size;
-	bool changed;
-
-	if (old_vps->base.crtc_w != new_vps->base.crtc_w ||
-	    old_vps->base.crtc_h != new_vps->base.crtc_h)
-	    return true;
-
-	if (old_vps->cursor.hotspot_x != new_vps->cursor.hotspot_x ||
-	    old_vps->cursor.hotspot_y != new_vps->cursor.hotspot_y)
-	    return true;
-
-	size = new_vps->base.crtc_w * new_vps->base.crtc_h * sizeof(u32);
-
-	old_image = vmw_du_cursor_plane_acquire_image(old_vps);
-	new_image = vmw_du_cursor_plane_acquire_image(new_vps);
-
-	changed = false;
-	if (old_image && new_image && old_image != new_image)
-		changed = memcmp(old_image, new_image, size) != 0;
-
-	return changed;
-}
-
-static void vmw_du_destroy_cursor_mob(struct vmw_bo **vbo)
-{
-	if (!(*vbo))
-		return;
-
-	ttm_bo_unpin(&(*vbo)->tbo);
-	vmw_bo_unreference(vbo);
-}
-
-static void vmw_du_put_cursor_mob(struct vmw_cursor_plane *vcp,
-				  struct vmw_plane_state *vps)
-{
-	u32 i;
-
-	if (!vps->cursor.bo)
-		return;
-
-	vmw_du_cursor_plane_unmap_cm(vps);
-
-	/* Look for a free slot to return this mob to the cache. */
-	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++) {
-		if (!vcp->cursor_mobs[i]) {
-			vcp->cursor_mobs[i] = vps->cursor.bo;
-			vps->cursor.bo = NULL;
-			return;
-		}
-	}
-
-	/* Cache is full: See if this mob is bigger than an existing mob. */
-	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++) {
-		if (vcp->cursor_mobs[i]->tbo.base.size <
-		    vps->cursor.bo->tbo.base.size) {
-			vmw_du_destroy_cursor_mob(&vcp->cursor_mobs[i]);
-			vcp->cursor_mobs[i] = vps->cursor.bo;
-			vps->cursor.bo = NULL;
-			return;
-		}
-	}
-
-	/* Destroy it if it's not worth caching. */
-	vmw_du_destroy_cursor_mob(&vps->cursor.bo);
-}
-
-static int vmw_du_get_cursor_mob(struct vmw_cursor_plane *vcp,
-				 struct vmw_plane_state *vps)
-{
-	struct vmw_private *dev_priv = vmw_priv(vcp->base.dev);
-	u32 size = vmw_du_cursor_mob_size(vps->base.crtc_w, vps->base.crtc_h);
-	u32 i;
-	u32 cursor_max_dim, mob_max_size;
-	struct vmw_fence_obj *fence = NULL;
-	int ret;
-
-	if (!dev_priv->has_mob ||
-	    (dev_priv->capabilities2 & SVGA_CAP2_CURSOR_MOB) == 0)
-		return -EINVAL;
-
-	mob_max_size = vmw_read(dev_priv, SVGA_REG_MOB_MAX_SIZE);
-	cursor_max_dim = vmw_read(dev_priv, SVGA_REG_CURSOR_MAX_DIMENSION);
-
-	if (size > mob_max_size || vps->base.crtc_w > cursor_max_dim ||
-	    vps->base.crtc_h > cursor_max_dim)
-		return -EINVAL;
-
-	if (vps->cursor.bo) {
-		if (vps->cursor.bo->tbo.base.size >= size)
-			return 0;
-		vmw_du_put_cursor_mob(vcp, vps);
-	}
-
-	/* Look for an unused mob in the cache. */
-	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++) {
-		if (vcp->cursor_mobs[i] &&
-		    vcp->cursor_mobs[i]->tbo.base.size >= size) {
-			vps->cursor.bo = vcp->cursor_mobs[i];
-			vcp->cursor_mobs[i] = NULL;
-			return 0;
-		}
-	}
-	/* Create a new mob if we can't find an existing one. */
-	ret = vmw_bo_create_and_populate(dev_priv, size,
-					 VMW_BO_DOMAIN_MOB,
-					 &vps->cursor.bo);
-
-	if (ret != 0)
-		return ret;
-
-	/* Fence the mob creation so we are guarateed to have the mob */
-	ret = ttm_bo_reserve(&vps->cursor.bo->tbo, false, false, NULL);
-	if (ret != 0)
-		goto teardown;
-
-	ret = vmw_execbuf_fence_commands(NULL, dev_priv, &fence, NULL);
-	if (ret != 0) {
-		ttm_bo_unreserve(&vps->cursor.bo->tbo);
-		goto teardown;
-	}
-
-	dma_fence_wait(&fence->base, false);
-	dma_fence_put(&fence->base);
-
-	ttm_bo_unreserve(&vps->cursor.bo->tbo);
-	return 0;
-
-teardown:
-	vmw_du_destroy_cursor_mob(&vps->cursor.bo);
-	return ret;
-}
-
-
-static void vmw_cursor_update_position(struct vmw_private *dev_priv,
-				       bool show, int x, int y)
-{
-	const uint32_t svga_cursor_on = show ? SVGA_CURSOR_ON_SHOW
-					     : SVGA_CURSOR_ON_HIDE;
-	uint32_t count;
-
-	spin_lock(&dev_priv->cursor_lock);
-	if (dev_priv->capabilities2 & SVGA_CAP2_EXTRA_REGS) {
-		vmw_write(dev_priv, SVGA_REG_CURSOR4_X, x);
-		vmw_write(dev_priv, SVGA_REG_CURSOR4_Y, y);
-		vmw_write(dev_priv, SVGA_REG_CURSOR4_SCREEN_ID, SVGA3D_INVALID_ID);
-		vmw_write(dev_priv, SVGA_REG_CURSOR4_ON, svga_cursor_on);
-		vmw_write(dev_priv, SVGA_REG_CURSOR4_SUBMIT, 1);
-	} else if (vmw_is_cursor_bypass3_enabled(dev_priv)) {
-		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_ON, svga_cursor_on);
-		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_X, x);
-		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_Y, y);
-		count = vmw_fifo_mem_read(dev_priv, SVGA_FIFO_CURSOR_COUNT);
-		vmw_fifo_mem_write(dev_priv, SVGA_FIFO_CURSOR_COUNT, ++count);
-	} else {
-		vmw_write(dev_priv, SVGA_REG_CURSOR_X, x);
-		vmw_write(dev_priv, SVGA_REG_CURSOR_Y, y);
-		vmw_write(dev_priv, SVGA_REG_CURSOR_ON, svga_cursor_on);
-	}
-	spin_unlock(&dev_priv->cursor_lock);
-}
-
-void vmw_kms_cursor_snoop(struct vmw_surface *srf,
-			  struct ttm_object_file *tfile,
-			  struct ttm_buffer_object *bo,
-			  SVGA3dCmdHeader *header)
-{
-	struct ttm_bo_kmap_obj map;
-	unsigned long kmap_offset;
-	unsigned long kmap_num;
-	SVGA3dCopyBox *box;
-	unsigned box_count;
-	void *virtual;
-	bool is_iomem;
-	struct vmw_dma_cmd {
-		SVGA3dCmdHeader header;
-		SVGA3dCmdSurfaceDMA dma;
-	} *cmd;
-	int i, ret;
-	const struct SVGA3dSurfaceDesc *desc =
-		vmw_surface_get_desc(VMW_CURSOR_SNOOP_FORMAT);
-	const u32 image_pitch = VMW_CURSOR_SNOOP_WIDTH * desc->pitchBytesPerBlock;
-
-	cmd = container_of(header, struct vmw_dma_cmd, header);
-
-	/* No snooper installed, nothing to copy */
-	if (!srf->snooper.image)
-		return;
-
-	if (cmd->dma.host.face != 0 || cmd->dma.host.mipmap != 0) {
-		DRM_ERROR("face and mipmap for cursors should never != 0\n");
-		return;
-	}
-
-	if (cmd->header.size < 64) {
-		DRM_ERROR("at least one full copy box must be given\n");
-		return;
-	}
-
-	box = (SVGA3dCopyBox *)&cmd[1];
-	box_count = (cmd->header.size - sizeof(SVGA3dCmdSurfaceDMA)) /
-			sizeof(SVGA3dCopyBox);
-
-	if (cmd->dma.guest.ptr.offset % PAGE_SIZE ||
-	    box->x != 0    || box->y != 0    || box->z != 0    ||
-	    box->srcx != 0 || box->srcy != 0 || box->srcz != 0 ||
-	    box->d != 1    || box_count != 1 ||
-	    box->w > VMW_CURSOR_SNOOP_WIDTH || box->h > VMW_CURSOR_SNOOP_HEIGHT) {
-		/* TODO handle none page aligned offsets */
-		/* TODO handle more dst & src != 0 */
-		/* TODO handle more then one copy */
-		DRM_ERROR("Can't snoop dma request for cursor!\n");
-		DRM_ERROR("(%u, %u, %u) (%u, %u, %u) (%ux%ux%u) %u %u\n",
-			  box->srcx, box->srcy, box->srcz,
-			  box->x, box->y, box->z,
-			  box->w, box->h, box->d, box_count,
-			  cmd->dma.guest.ptr.offset);
-		return;
-	}
-
-	kmap_offset = cmd->dma.guest.ptr.offset >> PAGE_SHIFT;
-	kmap_num = (VMW_CURSOR_SNOOP_HEIGHT*image_pitch) >> PAGE_SHIFT;
-
-	ret = ttm_bo_reserve(bo, true, false, NULL);
-	if (unlikely(ret != 0)) {
-		DRM_ERROR("reserve failed\n");
-		return;
-	}
-
-	ret = ttm_bo_kmap(bo, kmap_offset, kmap_num, &map);
-	if (unlikely(ret != 0))
-		goto err_unreserve;
-
-	virtual = ttm_kmap_obj_virtual(&map, &is_iomem);
-
-	if (box->w == VMW_CURSOR_SNOOP_WIDTH && cmd->dma.guest.pitch == image_pitch) {
-		memcpy(srf->snooper.image, virtual,
-		       VMW_CURSOR_SNOOP_HEIGHT*image_pitch);
-	} else {
-		/* Image is unsigned pointer. */
-		for (i = 0; i < box->h; i++)
-			memcpy(srf->snooper.image + i * image_pitch,
-			       virtual + i * cmd->dma.guest.pitch,
-			       box->w * desc->pitchBytesPerBlock);
-	}
-
-	srf->snooper.age++;
-
-	ttm_bo_kunmap(&map);
-err_unreserve:
-	ttm_bo_unreserve(bo);
-}
-
-/**
- * vmw_kms_legacy_hotspot_clear - Clear legacy hotspots
- *
- * @dev_priv: Pointer to the device private struct.
- *
- * Clears all legacy hotspots.
- */
-void vmw_kms_legacy_hotspot_clear(struct vmw_private *dev_priv)
-{
-	struct drm_device *dev = &dev_priv->drm;
-	struct vmw_display_unit *du;
-	struct drm_crtc *crtc;
-
-	drm_modeset_lock_all(dev);
-	drm_for_each_crtc(crtc, dev) {
-		du = vmw_crtc_to_du(crtc);
-
-		du->hotspot_x = 0;
-		du->hotspot_y = 0;
-	}
-	drm_modeset_unlock_all(dev);
-}
-
-void vmw_kms_cursor_post_execbuf(struct vmw_private *dev_priv)
-{
-	struct drm_device *dev = &dev_priv->drm;
-	struct vmw_display_unit *du;
-	struct drm_crtc *crtc;
-
-	mutex_lock(&dev->mode_config.mutex);
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		du = vmw_crtc_to_du(crtc);
-		if (!du->cursor_surface ||
-		    du->cursor_age == du->cursor_surface->snooper.age ||
-		    !du->cursor_surface->snooper.image)
-			continue;
-
-		du->cursor_age = du->cursor_surface->snooper.age;
-		vmw_send_define_cursor_cmd(dev_priv,
-					   du->cursor_surface->snooper.image,
-					   VMW_CURSOR_SNOOP_WIDTH,
-					   VMW_CURSOR_SNOOP_HEIGHT,
-					   du->hotspot_x + du->core_hotspot_x,
-					   du->hotspot_y + du->core_hotspot_y);
-	}
-
-	mutex_unlock(&dev->mode_config.mutex);
-}
-
-
-void vmw_du_cursor_plane_destroy(struct drm_plane *plane)
-{
-	struct vmw_cursor_plane *vcp = vmw_plane_to_vcp(plane);
-	u32 i;
-
-	vmw_cursor_update_position(vmw_priv(plane->dev), false, 0, 0);
-
-	for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++)
-		vmw_du_destroy_cursor_mob(&vcp->cursor_mobs[i]);
-
-	drm_plane_cleanup(plane);
-}
-
 
 void vmw_du_primary_plane_destroy(struct drm_plane *plane)
 {
@@ -575,262 +89,6 @@ vmw_du_plane_cleanup_fb(struct drm_plane *plane,
 
 
 /**
- * vmw_du_cursor_plane_map_cm - Maps the cursor mobs.
- *
- * @vps: plane_state
- *
- * Returns 0 on success
- */
-
-static int
-vmw_du_cursor_plane_map_cm(struct vmw_plane_state *vps)
-{
-	int ret;
-	u32 size = vmw_du_cursor_mob_size(vps->base.crtc_w, vps->base.crtc_h);
-	struct ttm_buffer_object *bo;
-
-	if (!vps->cursor.bo)
-		return -EINVAL;
-
-	bo = &vps->cursor.bo->tbo;
-
-	if (bo->base.size < size)
-		return -EINVAL;
-
-	if (vps->cursor.bo->map.virtual)
-		return 0;
-
-	ret = ttm_bo_reserve(bo, false, false, NULL);
-	if (unlikely(ret != 0))
-		return -ENOMEM;
-
-	vmw_bo_map_and_cache(vps->cursor.bo);
-
-	ttm_bo_unreserve(bo);
-
-	if (unlikely(ret != 0))
-		return -ENOMEM;
-
-	return 0;
-}
-
-
-/**
- * vmw_du_cursor_plane_unmap_cm - Unmaps the cursor mobs.
- *
- * @vps: state of the cursor plane
- *
- * Returns 0 on success
- */
-
-static int
-vmw_du_cursor_plane_unmap_cm(struct vmw_plane_state *vps)
-{
-	int ret = 0;
-	struct vmw_bo *vbo = vps->cursor.bo;
-
-	if (!vbo || !vbo->map.virtual)
-		return 0;
-
-	ret = ttm_bo_reserve(&vbo->tbo, true, false, NULL);
-	if (likely(ret == 0)) {
-		vmw_bo_unmap(vbo);
-		ttm_bo_unreserve(&vbo->tbo);
-	}
-
-	return ret;
-}
-
-
-/**
- * vmw_du_cursor_plane_cleanup_fb - Unpins the plane surface
- *
- * @plane: cursor plane
- * @old_state: contains the state to clean up
- *
- * Unmaps all cursor bo mappings and unpins the cursor surface
- *
- * Returns 0 on success
- */
-void
-vmw_du_cursor_plane_cleanup_fb(struct drm_plane *plane,
-			       struct drm_plane_state *old_state)
-{
-	struct vmw_cursor_plane *vcp = vmw_plane_to_vcp(plane);
-	struct vmw_plane_state *vps = vmw_plane_state_to_vps(old_state);
-
-	if (!vmw_user_object_is_null(&vps->uo))
-		vmw_user_object_unmap(&vps->uo);
-
-	vmw_du_cursor_plane_unmap_cm(vps);
-	vmw_du_put_cursor_mob(vcp, vps);
-
-	vmw_du_plane_unpin_surf(vps);
-	vmw_user_object_unref(&vps->uo);
-}
-
-
-/**
- * vmw_du_cursor_plane_prepare_fb - Readies the cursor by referencing it
- *
- * @plane:  display plane
- * @new_state: info on the new plane state, including the FB
- *
- * Returns 0 on success
- */
-int
-vmw_du_cursor_plane_prepare_fb(struct drm_plane *plane,
-			       struct drm_plane_state *new_state)
-{
-	struct drm_framebuffer *fb = new_state->fb;
-	struct vmw_cursor_plane *vcp = vmw_plane_to_vcp(plane);
-	struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state);
-	struct vmw_bo *bo = NULL;
-	int ret = 0;
-
-	if (!vmw_user_object_is_null(&vps->uo)) {
-		vmw_user_object_unmap(&vps->uo);
-		vmw_user_object_unref(&vps->uo);
-	}
-
-	if (fb) {
-		if (vmw_framebuffer_to_vfb(fb)->bo) {
-			vps->uo.buffer = vmw_framebuffer_to_vfbd(fb)->buffer;
-			vps->uo.surface = NULL;
-		} else {
-			memcpy(&vps->uo, &vmw_framebuffer_to_vfbs(fb)->uo, sizeof(vps->uo));
-		}
-		vmw_user_object_ref(&vps->uo);
-	}
-
-	bo = vmw_user_object_buffer(&vps->uo);
-	if (bo) {
-		struct ttm_operation_ctx ctx = {false, false};
-
-		ret = ttm_bo_reserve(&bo->tbo, true, false, NULL);
-		if (ret != 0)
-			return -ENOMEM;
-
-		ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
-		if (ret != 0)
-			return -ENOMEM;
-
-		vmw_bo_pin_reserved(bo, true);
-		if (vmw_framebuffer_to_vfb(fb)->bo) {
-			const u32 size = new_state->crtc_w * new_state->crtc_h * sizeof(u32);
-
-			(void)vmw_bo_map_and_cache_size(bo, size);
-		} else {
-			vmw_bo_map_and_cache(bo);
-		}
-		ttm_bo_unreserve(&bo->tbo);
-	}
-
-	if (!vmw_user_object_is_null(&vps->uo)) {
-		vmw_du_get_cursor_mob(vcp, vps);
-		vmw_du_cursor_plane_map_cm(vps);
-	}
-
-	return 0;
-}
-
-
-void
-vmw_du_cursor_plane_atomic_update(struct drm_plane *plane,
-				  struct drm_atomic_state *state)
-{
-	struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state,
-									   plane);
-	struct drm_plane_state *old_state = drm_atomic_get_old_plane_state(state,
-									   plane);
-	struct drm_crtc *crtc = new_state->crtc ?: old_state->crtc;
-	struct vmw_private *dev_priv = vmw_priv(crtc->dev);
-	struct vmw_display_unit *du = vmw_crtc_to_du(crtc);
-	struct vmw_plane_state *vps = vmw_plane_state_to_vps(new_state);
-	struct vmw_plane_state *old_vps = vmw_plane_state_to_vps(old_state);
-	struct vmw_bo *old_bo = NULL;
-	struct vmw_bo *new_bo = NULL;
-	struct ww_acquire_ctx ctx;
-	s32 hotspot_x, hotspot_y;
-	int ret;
-
-	hotspot_x = du->hotspot_x + new_state->hotspot_x;
-	hotspot_y = du->hotspot_y + new_state->hotspot_y;
-
-	du->cursor_surface = vmw_user_object_surface(&vps->uo);
-
-	if (vmw_user_object_is_null(&vps->uo)) {
-		vmw_cursor_update_position(dev_priv, false, 0, 0);
-		return;
-	}
-
-	vps->cursor.hotspot_x = hotspot_x;
-	vps->cursor.hotspot_y = hotspot_y;
-
-	if (du->cursor_surface)
-		du->cursor_age = du->cursor_surface->snooper.age;
-
-	ww_acquire_init(&ctx, &reservation_ww_class);
-
-	if (!vmw_user_object_is_null(&old_vps->uo)) {
-		old_bo = vmw_user_object_buffer(&old_vps->uo);
-		ret = ttm_bo_reserve(&old_bo->tbo, false, false, &ctx);
-		if (ret != 0)
-			return;
-	}
-
-	if (!vmw_user_object_is_null(&vps->uo)) {
-		new_bo = vmw_user_object_buffer(&vps->uo);
-		if (old_bo != new_bo) {
-			ret = ttm_bo_reserve(&new_bo->tbo, false, false, &ctx);
-			if (ret != 0) {
-				if (old_bo) {
-					ttm_bo_unreserve(&old_bo->tbo);
-					ww_acquire_fini(&ctx);
-				}
-				return;
-			}
-		} else {
-			new_bo = NULL;
-		}
-	}
-	if (!vmw_du_cursor_plane_has_changed(old_vps, vps)) {
-		/*
-		 * If it hasn't changed, avoid making the device do extra
-		 * work by keeping the old cursor active.
-		 */
-		struct vmw_cursor_plane_state tmp = old_vps->cursor;
-		old_vps->cursor = vps->cursor;
-		vps->cursor = tmp;
-	} else {
-		void *image = vmw_du_cursor_plane_acquire_image(vps);
-		if (image)
-			vmw_cursor_update_image(dev_priv, vps, image,
-						new_state->crtc_w,
-						new_state->crtc_h,
-						hotspot_x, hotspot_y);
-	}
-
-	if (new_bo)
-		ttm_bo_unreserve(&new_bo->tbo);
-	if (old_bo)
-		ttm_bo_unreserve(&old_bo->tbo);
-
-	ww_acquire_fini(&ctx);
-
-	du->cursor_x = new_state->crtc_x + du->set_gui_x;
-	du->cursor_y = new_state->crtc_y + du->set_gui_y;
-
-	vmw_cursor_update_position(dev_priv, true,
-				   du->cursor_x + hotspot_x,
-				   du->cursor_y + hotspot_y);
-
-	du->core_hotspot_x = hotspot_x - du->hotspot_x;
-	du->core_hotspot_y = hotspot_y - du->hotspot_y;
-}
-
-
-/**
  * vmw_du_primary_plane_atomic_check - check if the new state is okay
  *
  * @plane: display plane
@@ -873,66 +131,6 @@ int vmw_du_primary_plane_atomic_check(struct drm_plane *plane,
 	return ret;
 }
 
-
-/**
- * vmw_du_cursor_plane_atomic_check - check if the new state is okay
- *
- * @plane: cursor plane
- * @state: info on the new plane state
- *
- * This is a chance to fail if the new cursor state does not fit
- * our requirements.
- *
- * Returns 0 on success
- */
-int vmw_du_cursor_plane_atomic_check(struct drm_plane *plane,
-				     struct drm_atomic_state *state)
-{
-	struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state,
-									   plane);
-	int ret = 0;
-	struct drm_crtc_state *crtc_state = NULL;
-	struct vmw_surface *surface = NULL;
-	struct drm_framebuffer *fb = new_state->fb;
-
-	if (new_state->crtc)
-		crtc_state = drm_atomic_get_new_crtc_state(new_state->state,
-							   new_state->crtc);
-
-	ret = drm_atomic_helper_check_plane_state(new_state, crtc_state,
-						  DRM_PLANE_NO_SCALING,
-						  DRM_PLANE_NO_SCALING,
-						  true, true);
-	if (ret)
-		return ret;
-
-	/* Turning off */
-	if (!fb)
-		return 0;
-
-	/* A lot of the code assumes this */
-	if (new_state->crtc_w != 64 || new_state->crtc_h != 64) {
-		DRM_ERROR("Invalid cursor dimensions (%d, %d)\n",
-			  new_state->crtc_w, new_state->crtc_h);
-		return -EINVAL;
-	}
-
-	if (!vmw_framebuffer_to_vfb(fb)->bo) {
-		surface = vmw_user_object_surface(&vmw_framebuffer_to_vfbs(fb)->uo);
-
-		WARN_ON(!surface);
-
-		if (!surface ||
-		    (!surface->snooper.image && !surface->res.guest_memory_bo)) {
-			DRM_ERROR("surface not suitable for cursor\n");
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
-
 int vmw_du_crtc_atomic_check(struct drm_crtc *crtc,
 			     struct drm_atomic_state *state)
 {
@@ -1076,7 +274,7 @@ vmw_du_plane_duplicate_state(struct drm_plane *plane)
 	vps->pinned = 0;
 	vps->cpp = 0;
 
-	memset(&vps->cursor, 0, sizeof(vps->cursor));
+	vps->cursor.mob = NULL;
 
 	/* Each ref counted resource needs to be acquired again */
 	vmw_user_object_ref(&vps->uo);
@@ -1221,7 +419,20 @@ static void vmw_framebuffer_surface_destroy(struct drm_framebuffer *framebuffer)
 {
 	struct vmw_framebuffer_surface *vfbs =
 		vmw_framebuffer_to_vfbs(framebuffer);
+	struct vmw_bo *bo = vmw_user_object_buffer(&vfbs->uo);
+	struct vmw_surface *surf = vmw_user_object_surface(&vfbs->uo);
 
+	if (bo) {
+		vmw_bo_dirty_release(bo);
+		/*
+		 * bo->dirty is reference counted so it being NULL
+		 * means that the surface wasn't coherent to begin
+		 * with and so we have to free the dirty tracker
+		 * in the vmw_resource
+		 */
+		if (!bo->dirty && surf && surf->res.dirty)
+			surf->res.func->dirty_free(&surf->res);
+	}
 	drm_framebuffer_cleanup(framebuffer);
 	vmw_user_object_unref(&vfbs->uo);
 
@@ -1375,6 +586,7 @@ static void vmw_framebuffer_bo_destroy(struct drm_framebuffer *framebuffer)
 	struct vmw_framebuffer_bo *vfbd =
 		vmw_framebuffer_to_vfbd(framebuffer);
 
+	vmw_bo_dirty_release(vfbd->buffer);
 	drm_framebuffer_cleanup(framebuffer);
 	vmw_bo_unreference(&vfbd->buffer);
 
@@ -1505,6 +717,8 @@ static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev,
 	struct vmw_private *dev_priv = vmw_priv(dev);
 	struct vmw_framebuffer *vfb = NULL;
 	struct vmw_user_object uo = {0};
+	struct vmw_bo *bo;
+	struct vmw_surface *surface;
 	int ret;
 
 	/* returns either a bo or surface */
@@ -1534,6 +748,8 @@ static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev,
 	}
 
 err_out:
+	bo = vmw_user_object_buffer(&uo);
+	surface = vmw_user_object_surface(&uo);
 	/* vmw_user_object_lookup takes one ref so does new_fb */
 	vmw_user_object_unref(&uo);
 
@@ -1542,6 +758,14 @@ err_out:
 		return ERR_PTR(ret);
 	}
 
+	ttm_bo_reserve(&bo->tbo, false, false, NULL);
+	ret = vmw_bo_dirty_add(bo);
+	if (!ret && surface && surface->res.func->dirty_alloc) {
+		surface->res.coherent = true;
+		ret = surface->res.func->dirty_alloc(&surface->res);
+	}
+	ttm_bo_unreserve(&bo->tbo);
+
 	return &vfb->base;
 }
 
@@ -1974,44 +1198,6 @@ int vmw_kms_close(struct vmw_private *dev_priv)
 	return ret;
 }
 
-int vmw_kms_cursor_bypass_ioctl(struct drm_device *dev, void *data,
-				struct drm_file *file_priv)
-{
-	struct drm_vmw_cursor_bypass_arg *arg = data;
-	struct vmw_display_unit *du;
-	struct drm_crtc *crtc;
-	int ret = 0;
-
-	mutex_lock(&dev->mode_config.mutex);
-	if (arg->flags & DRM_VMW_CURSOR_BYPASS_ALL) {
-
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-			du = vmw_crtc_to_du(crtc);
-			du->hotspot_x = arg->xhot;
-			du->hotspot_y = arg->yhot;
-		}
-
-		mutex_unlock(&dev->mode_config.mutex);
-		return 0;
-	}
-
-	crtc = drm_crtc_find(dev, file_priv, arg->crtc_id);
-	if (!crtc) {
-		ret = -ENOENT;
-		goto out;
-	}
-
-	du = vmw_crtc_to_du(crtc);
-
-	du->hotspot_x = arg->xhot;
-	du->hotspot_y = arg->yhot;
-
-out:
-	mutex_unlock(&dev->mode_config.mutex);
-
-	return ret;
-}
-
 int vmw_kms_write_svga(struct vmw_private *vmw_priv,
 			unsigned width, unsigned height, unsigned pitch,
 			unsigned bpp, unsigned depth)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index 4eab581883e2..511e29cdb987 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -1,40 +1,21 @@
 /* SPDX-License-Identifier: GPL-2.0 OR MIT */
 /**************************************************************************
  *
- * Copyright (c) 2009-2024 Broadcom. All Rights Reserved. The term
+ * Copyright (c) 2009-2025 Broadcom. All Rights Reserved. The term
  * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
  **************************************************************************/
 
 #ifndef VMWGFX_KMS_H_
 #define VMWGFX_KMS_H_
 
+#include "vmwgfx_cursor_plane.h"
+#include "vmwgfx_drv.h"
+
 #include <drm/drm_encoder.h>
 #include <drm/drm_framebuffer.h>
 #include <drm/drm_probe_helper.h>
 
-#include "vmwgfx_drv.h"
-
 /**
  * struct vmw_du_update_plane - Closure structure for vmw_du_helper_plane_update
  * @plane: Plane which is being updated.
@@ -235,16 +216,11 @@ static const uint32_t __maybe_unused vmw_primary_plane_formats[] = {
 	DRM_FORMAT_XRGB1555,
 };
 
-static const uint32_t __maybe_unused vmw_cursor_plane_formats[] = {
-	DRM_FORMAT_ARGB8888,
-};
-
 
 #define vmw_crtc_state_to_vcs(x) container_of(x, struct vmw_crtc_state, base)
 #define vmw_plane_state_to_vps(x) container_of(x, struct vmw_plane_state, base)
 #define vmw_connector_state_to_vcs(x) \
 		container_of(x, struct vmw_connector_state, base)
-#define vmw_plane_to_vcp(x) container_of(x, struct vmw_cursor_plane, base)
 
 /**
  * Derived class for crtc state object
@@ -255,11 +231,6 @@ struct vmw_crtc_state {
 	struct drm_crtc_state base;
 };
 
-struct vmw_cursor_plane_state {
-	struct vmw_bo *bo;
-	s32 hotspot_x;
-	s32 hotspot_y;
-};
 
 /**
  * Derived class for plane state object
@@ -283,7 +254,6 @@ struct vmw_plane_state {
 	/* For CPU Blit */
 	unsigned int cpp;
 
-	bool surf_mapped;
 	struct vmw_cursor_plane_state cursor;
 };
 
@@ -317,17 +287,6 @@ struct vmw_connector_state {
 	int gui_y;
 };
 
-/**
- * Derived class for cursor plane object
- *
- * @base DRM plane object
- * @cursor.cursor_mobs Cursor mobs available for re-use
- */
-struct vmw_cursor_plane {
-	struct drm_plane base;
-
-	struct vmw_bo *cursor_mobs[3];
-};
 
 /**
  * Base class display unit.
@@ -343,17 +302,6 @@ struct vmw_display_unit {
 	struct drm_plane primary;
 	struct vmw_cursor_plane cursor;
 
-	struct vmw_surface *cursor_surface;
-	size_t cursor_age;
-
-	int cursor_x;
-	int cursor_y;
-
-	int hotspot_x;
-	int hotspot_y;
-	s32 core_hotspot_x;
-	s32 core_hotspot_y;
-
 	unsigned unit;
 
 	/*
@@ -403,8 +351,6 @@ struct vmw_display_unit {
  */
 void vmw_du_init(struct vmw_display_unit *du);
 void vmw_du_cleanup(struct vmw_display_unit *du);
-void vmw_du_crtc_save(struct drm_crtc *crtc);
-void vmw_du_crtc_restore(struct drm_crtc *crtc);
 int vmw_du_crtc_gamma_set(struct drm_crtc *crtc,
 			   u16 *r, u16 *g, u16 *b,
 			   uint32_t size,
@@ -460,19 +406,10 @@ void vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv);
 
 /* Universal Plane Helpers */
 void vmw_du_primary_plane_destroy(struct drm_plane *plane);
-void vmw_du_cursor_plane_destroy(struct drm_plane *plane);
 
 /* Atomic Helpers */
 int vmw_du_primary_plane_atomic_check(struct drm_plane *plane,
 				      struct drm_atomic_state *state);
-int vmw_du_cursor_plane_atomic_check(struct drm_plane *plane,
-				     struct drm_atomic_state *state);
-void vmw_du_cursor_plane_atomic_update(struct drm_plane *plane,
-				       struct drm_atomic_state *state);
-int vmw_du_cursor_plane_prepare_fb(struct drm_plane *plane,
-				   struct drm_plane_state *new_state);
-void vmw_du_cursor_plane_cleanup_fb(struct drm_plane *plane,
-			     struct drm_plane_state *old_state);
 void vmw_du_plane_cleanup_fb(struct drm_plane *plane,
 			     struct drm_plane_state *old_state);
 void vmw_du_plane_reset(struct drm_plane *plane);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index f0b429525467..c23c9195f0dc 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -372,7 +372,7 @@ static const struct drm_plane_funcs vmw_ldu_plane_funcs = {
 static const struct drm_plane_funcs vmw_ldu_cursor_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
-	.destroy = vmw_du_cursor_plane_destroy,
+	.destroy = vmw_cursor_plane_destroy,
 	.reset = vmw_du_plane_reset,
 	.atomic_duplicate_state = vmw_du_plane_duplicate_state,
 	.atomic_destroy_state = vmw_du_plane_destroy_state,
@@ -383,10 +383,10 @@ static const struct drm_plane_funcs vmw_ldu_cursor_funcs = {
  */
 static const struct
 drm_plane_helper_funcs vmw_ldu_cursor_plane_helper_funcs = {
-	.atomic_check = vmw_du_cursor_plane_atomic_check,
-	.atomic_update = vmw_du_cursor_plane_atomic_update,
-	.prepare_fb = vmw_du_cursor_plane_prepare_fb,
-	.cleanup_fb = vmw_du_cursor_plane_cleanup_fb,
+	.atomic_check = vmw_cursor_plane_atomic_check,
+	.atomic_update = vmw_cursor_plane_atomic_update,
+	.prepare_fb = vmw_cursor_plane_prepare_fb,
+	.cleanup_fb = vmw_cursor_plane_cleanup_fb,
 };
 
 static const struct
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
index 7055cbefc768..d8204d4265d3 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_mob.c
@@ -282,8 +282,7 @@ out_no_setup:
 	}
 
 	vmw_bo_unpin_unlocked(&batch->otable_bo->tbo);
-	ttm_bo_put(&batch->otable_bo->tbo);
-	batch->otable_bo = NULL;
+	vmw_bo_unreference(&batch->otable_bo);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c
index 74ff2812d66a..7de20e56082c 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c
@@ -1,27 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0 OR MIT
 /**************************************************************************
  *
- * Copyright 2019-2023 VMware, Inc., Palo Alto, CA., USA
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ * Copyright (c) 2019-2025 Broadcom. All Rights Reserved. The term
+ * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
  **************************************************************************/
 #include "vmwgfx_bo.h"
@@ -71,6 +52,11 @@ struct vmw_bo_dirty {
 	unsigned long bitmap[];
 };
 
+bool vmw_bo_is_dirty(struct vmw_bo *vbo)
+{
+	return vbo->dirty && (vbo->dirty->start < vbo->dirty->end);
+}
+
 /**
  * vmw_bo_dirty_scan_pagetable - Perform a pagetable scan for dirty bits
  * @vbo: The buffer object to scan
@@ -341,6 +327,41 @@ void vmw_bo_dirty_transfer_to_res(struct vmw_resource *res)
 		dirty->end = res_start;
 }
 
+void vmw_bo_dirty_clear(struct vmw_bo *vbo)
+{
+	struct vmw_bo_dirty *dirty = vbo->dirty;
+	pgoff_t start, cur, end;
+	unsigned long res_start = 0;
+	unsigned long res_end = vbo->tbo.base.size;
+
+	WARN_ON_ONCE(res_start & ~PAGE_MASK);
+	res_start >>= PAGE_SHIFT;
+	res_end = DIV_ROUND_UP(res_end, PAGE_SIZE);
+
+	if (res_start >= dirty->end || res_end <= dirty->start)
+		return;
+
+	cur = max(res_start, dirty->start);
+	res_end = max(res_end, dirty->end);
+	while (cur < res_end) {
+		unsigned long num;
+
+		start = find_next_bit(&dirty->bitmap[0], res_end, cur);
+		if (start >= res_end)
+			break;
+
+		end = find_next_zero_bit(&dirty->bitmap[0], res_end, start + 1);
+		cur = end + 1;
+		num = end - start;
+		bitmap_clear(&dirty->bitmap[0], start, num);
+	}
+
+	if (res_start <= dirty->start && res_end > dirty->start)
+		dirty->start = res_end;
+	if (res_start < dirty->end && res_end >= dirty->end)
+		dirty->end = res_start;
+}
+
 /**
  * vmw_bo_dirty_clear_res - Clear a resource's dirty region from
  * its backing mob.
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index a73af8a355fb..388011696941 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -273,7 +273,7 @@ int vmw_user_resource_lookup_handle(struct vmw_private *dev_priv,
 		goto out_bad_resource;
 
 	res = converter->base_obj_to_res(base);
-	kref_get(&res->kref);
+	vmw_resource_reference(res);
 
 	*p_res = res;
 	ret = 0;
@@ -347,7 +347,7 @@ static int vmw_resource_buf_alloc(struct vmw_resource *res,
 		return 0;
 	}
 
-	ret = vmw_gem_object_create(res->dev_priv, &bo_params, &gbo);
+	ret = vmw_bo_create(res->dev_priv, &bo_params, &gbo);
 	if (unlikely(ret != 0))
 		goto out_no_bo;
 
@@ -531,9 +531,9 @@ vmw_resource_check_buffer(struct ww_acquire_ctx *ticket,
 	}
 
 	INIT_LIST_HEAD(&val_list);
-	ttm_bo_get(&res->guest_memory_bo->tbo);
 	val_buf->bo = &res->guest_memory_bo->tbo;
 	val_buf->num_shared = 0;
+	drm_gem_object_get(&val_buf->bo->base);
 	list_add_tail(&val_buf->head, &val_list);
 	ret = ttm_eu_reserve_buffers(ticket, &val_list, interruptible, NULL);
 	if (unlikely(ret != 0))
@@ -557,7 +557,7 @@ vmw_resource_check_buffer(struct ww_acquire_ctx *ticket,
 out_no_validate:
 	ttm_eu_backoff_reservation(ticket, &val_list);
 out_no_reserve:
-	ttm_bo_put(val_buf->bo);
+	drm_gem_object_put(&val_buf->bo->base);
 	val_buf->bo = NULL;
 	if (guest_memory_dirty)
 		vmw_user_bo_unref(&res->guest_memory_bo);
@@ -619,7 +619,7 @@ vmw_resource_backoff_reservation(struct ww_acquire_ctx *ticket,
 	INIT_LIST_HEAD(&val_list);
 	list_add_tail(&val_buf->head, &val_list);
 	ttm_eu_backoff_reservation(ticket, &val_list);
-	ttm_bo_put(val_buf->bo);
+	drm_gem_object_put(&val_buf->bo->base);
 	val_buf->bo = NULL;
 }
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index 32029d80b72b..5f5f5a94301f 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -445,7 +445,7 @@ vmw_sou_primary_plane_prepare_fb(struct drm_plane *plane,
 	 * resume the overlays, this is preferred to failing to alloc.
 	 */
 	vmw_overlay_pause_all(dev_priv);
-	ret = vmw_gem_object_create(dev_priv, &bo_params, &vps->uo.buffer);
+	ret = vmw_bo_create(dev_priv, &bo_params, &vps->uo.buffer);
 	vmw_overlay_resume_all(dev_priv);
 	if (ret)
 		return ret;
@@ -764,7 +764,7 @@ static const struct drm_plane_funcs vmw_sou_plane_funcs = {
 static const struct drm_plane_funcs vmw_sou_cursor_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
-	.destroy = vmw_du_cursor_plane_destroy,
+	.destroy = vmw_cursor_plane_destroy,
 	.reset = vmw_du_plane_reset,
 	.atomic_duplicate_state = vmw_du_plane_duplicate_state,
 	.atomic_destroy_state = vmw_du_plane_destroy_state,
@@ -775,10 +775,10 @@ static const struct drm_plane_funcs vmw_sou_cursor_funcs = {
  */
 static const struct
 drm_plane_helper_funcs vmw_sou_cursor_plane_helper_funcs = {
-	.atomic_check = vmw_du_cursor_plane_atomic_check,
-	.atomic_update = vmw_du_cursor_plane_atomic_update,
-	.prepare_fb = vmw_du_cursor_plane_prepare_fb,
-	.cleanup_fb = vmw_du_cursor_plane_cleanup_fb,
+	.atomic_check = vmw_cursor_plane_atomic_check,
+	.atomic_update = vmw_cursor_plane_atomic_update,
+	.prepare_fb = vmw_cursor_plane_prepare_fb,
+	.cleanup_fb = vmw_cursor_plane_cleanup_fb,
 };
 
 static const struct
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
index f5d2ed1b0a72..20aab725e53a 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
@@ -1482,7 +1482,7 @@ static const struct drm_plane_funcs vmw_stdu_plane_funcs = {
 static const struct drm_plane_funcs vmw_stdu_cursor_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
-	.destroy = vmw_du_cursor_plane_destroy,
+	.destroy = vmw_cursor_plane_destroy,
 	.reset = vmw_du_plane_reset,
 	.atomic_duplicate_state = vmw_du_plane_duplicate_state,
 	.atomic_destroy_state = vmw_du_plane_destroy_state,
@@ -1494,10 +1494,10 @@ static const struct drm_plane_funcs vmw_stdu_cursor_funcs = {
  */
 static const struct
 drm_plane_helper_funcs vmw_stdu_cursor_plane_helper_funcs = {
-	.atomic_check = vmw_du_cursor_plane_atomic_check,
-	.atomic_update = vmw_du_cursor_plane_atomic_update,
-	.prepare_fb = vmw_du_cursor_plane_prepare_fb,
-	.cleanup_fb = vmw_du_cursor_plane_cleanup_fb,
+	.atomic_check = vmw_cursor_plane_atomic_check,
+	.atomic_update = vmw_cursor_plane_atomic_update,
+	.prepare_fb = vmw_cursor_plane_prepare_fb,
+	.cleanup_fb = vmw_cursor_plane_cleanup_fb,
 };
 
 static const struct
@@ -1584,6 +1584,7 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit)
 	}
 
 	drm_plane_helper_add(&cursor->base, &vmw_stdu_cursor_plane_helper_funcs);
+	drm_plane_enable_fb_damage_clips(&cursor->base);
 
 	ret = drm_connector_init(dev, connector, &vmw_stdu_connector_funcs,
 				 DRM_MODE_CONNECTOR_VIRTUAL);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
index 5721c74da3e0..7e281c3c6bc5 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
@@ -1,32 +1,13 @@
 // SPDX-License-Identifier: GPL-2.0 OR MIT
 /**************************************************************************
  *
- * Copyright (c) 2009-2024 Broadcom. All Rights Reserved. The term
+ * Copyright (c) 2009-2025 Broadcom. All Rights Reserved. The term
  * “Broadcom” refers to Broadcom Inc. and/or its subsidiaries.
  *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
  **************************************************************************/
 
 #include "vmwgfx_bo.h"
+#include "vmwgfx_cursor_plane.h"
 #include "vmwgfx_drv.h"
 #include "vmwgfx_resource_priv.h"
 #include "vmwgfx_so.h"
@@ -658,7 +639,7 @@ static void vmw_user_surface_free(struct vmw_resource *res)
 	struct vmw_user_surface *user_srf =
 	    container_of(srf, struct vmw_user_surface, srf);
 
-	WARN_ON_ONCE(res->dirty);
+	WARN_ON(res->dirty);
 	if (user_srf->master)
 		drm_master_put(&user_srf->master);
 	kfree(srf->offsets);
@@ -689,8 +670,7 @@ static void vmw_user_surface_base_release(struct ttm_base_object **p_base)
 	 * Dumb buffers own the resource and they'll unref the
 	 * resource themselves
 	 */
-	if (res && res->guest_memory_bo && res->guest_memory_bo->is_dumb)
-		return;
+	WARN_ON(res && res->guest_memory_bo && res->guest_memory_bo->is_dumb);
 
 	vmw_resource_unreference(&res);
 }
@@ -818,25 +798,11 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data,
 		}
 	}
 	res->guest_memory_size = cur_bo_offset;
-	if (!file_priv->atomic &&
-	    metadata->scanout &&
-	    metadata->num_sizes == 1 &&
-	    metadata->sizes[0].width == VMW_CURSOR_SNOOP_WIDTH &&
-	    metadata->sizes[0].height == VMW_CURSOR_SNOOP_HEIGHT &&
-	    metadata->format == VMW_CURSOR_SNOOP_FORMAT) {
-		const struct SVGA3dSurfaceDesc *desc =
-			vmw_surface_get_desc(VMW_CURSOR_SNOOP_FORMAT);
-		const u32 cursor_size_bytes = VMW_CURSOR_SNOOP_WIDTH *
-					      VMW_CURSOR_SNOOP_HEIGHT *
-					      desc->pitchBytesPerBlock;
-		srf->snooper.image = kzalloc(cursor_size_bytes, GFP_KERNEL);
-		if (!srf->snooper.image) {
-			DRM_ERROR("Failed to allocate cursor_image\n");
-			ret = -ENOMEM;
-			goto out_no_copy;
-		}
-	} else {
-		srf->snooper.image = NULL;
+
+	srf->snooper.image = vmw_cursor_snooper_create(file_priv, metadata);
+	if (IS_ERR(srf->snooper.image)) {
+		ret = PTR_ERR(srf->snooper.image);
+		goto out_no_copy;
 	}
 
 	if (drm_is_primary_client(file_priv))
@@ -864,14 +830,17 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data,
 			.pin = false
 		};
 
-		ret = vmw_gem_object_create(dev_priv,
-					    &params,
-					    &res->guest_memory_bo);
+		ret = vmw_bo_create(dev_priv, &params, &res->guest_memory_bo);
+		if (unlikely(ret != 0)) {
+			vmw_resource_unreference(&res);
+			goto out_unlock;
+		}
+
+		ret = vmw_bo_add_detached_resource(res->guest_memory_bo, res);
 		if (unlikely(ret != 0)) {
 			vmw_resource_unreference(&res);
 			goto out_unlock;
 		}
-		vmw_bo_add_detached_resource(res->guest_memory_bo, res);
 	}
 
 	tmp = vmw_resource_reference(&srf->res);
@@ -1670,6 +1639,14 @@ vmw_gb_surface_define_internal(struct drm_device *dev,
 
 	}
 
+	if (res->guest_memory_bo) {
+		ret = vmw_bo_add_detached_resource(res->guest_memory_bo, res);
+		if (unlikely(ret != 0)) {
+			vmw_resource_unreference(&res);
+			goto out_unlock;
+		}
+	}
+
 	tmp = vmw_resource_reference(res);
 	ret = ttm_prime_object_init(tfile, res->guest_memory_size, &user_srf->prime,
 				    VMW_RES_SURFACE,
@@ -1684,7 +1661,6 @@ vmw_gb_surface_define_internal(struct drm_device *dev,
 	rep->handle      = user_srf->prime.base.handle;
 	rep->backup_size = res->guest_memory_size;
 	if (res->guest_memory_bo) {
-		vmw_bo_add_detached_resource(res->guest_memory_bo, res);
 		rep->buffer_map_handle =
 			drm_vma_node_offset_addr(&res->guest_memory_bo->tbo.base.vma_node);
 		rep->buffer_size = res->guest_memory_bo->tbo.base.size;
@@ -2358,12 +2334,19 @@ int vmw_dumb_create(struct drm_file *file_priv,
 	vbo = res->guest_memory_bo;
 	vbo->is_dumb = true;
 	vbo->dumb_surface = vmw_res_to_srf(res);
-
+	drm_gem_object_put(&vbo->tbo.base);
+	/*
+	 * Unset the user surface dtor since this in not actually exposed
+	 * to userspace. The suface is owned via the dumb_buffer's GEM handle
+	 */
+	struct vmw_user_surface *usurf = container_of(vbo->dumb_surface,
+						struct vmw_user_surface, srf);
+	usurf->prime.base.refcount_release = NULL;
 err:
 	if (res)
 		vmw_resource_unreference(&res);
-	if (ret)
-		ttm_ref_object_base_unref(tfile, arg.rep.handle);
+
+	ttm_ref_object_base_unref(tfile, arg.rep.handle);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
index e7625b3f71e0..7ee93e7191c7 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
@@ -262,9 +262,8 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx,
 				bo_node->hash.key);
 		}
 		val_buf = &bo_node->base;
-		val_buf->bo = ttm_bo_get_unless_zero(&vbo->tbo);
-		if (!val_buf->bo)
-			return -ESRCH;
+		vmw_bo_reference(vbo);
+		val_buf->bo = &vbo->tbo;
 		val_buf->num_shared = 0;
 		list_add_tail(&val_buf->head, &ctx->bo_list);
 	}
@@ -656,7 +655,7 @@ void vmw_validation_unref_lists(struct vmw_validation_context *ctx)
 	struct vmw_validation_res_node *val;
 
 	list_for_each_entry(entry, &ctx->bo_list, base.head) {
-		ttm_bo_put(entry->base.bo);
+		drm_gem_object_put(&entry->base.bo->base);
 		entry->base.bo = NULL;
 	}
 
diff --git a/drivers/gpu/drm/xe/Kconfig b/drivers/gpu/drm/xe/Kconfig
index 5c2f459a2925..9bce047901b2 100644
--- a/drivers/gpu/drm/xe/Kconfig
+++ b/drivers/gpu/drm/xe/Kconfig
@@ -39,7 +39,6 @@ config DRM_XE
 	select DRM_TTM_HELPER
 	select DRM_EXEC
 	select DRM_GPUVM
-	select DRM_GPUSVM if !UML && DEVICE_PRIVATE
 	select DRM_SCHED
 	select MMU_NOTIFIER
 	select WANT_DEV_COREDUMP
@@ -74,9 +73,22 @@ config DRM_XE_DP_TUNNEL
 
 	  If in doubt say "Y".
 
+config DRM_XE_GPUSVM
+	bool "Enable CPU to GPU address mirroring"
+	depends on DRM_XE
+	depends on !UML
+	depends on DEVICE_PRIVATE
+	default y
+	select DRM_GPUSVM
+	help
+	  Enable this option if you want support for CPU to GPU address
+	  mirroring.
+
+	  If in doubut say "Y".
+
 config DRM_XE_DEVMEM_MIRROR
 	bool "Enable device memory mirror"
-	depends on DRM_XE
+	depends on DRM_XE_GPUSVM
 	select GET_FREE_REGION
 	default y
 	help
diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 9699b08585f7..e4bf484d4121 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -125,12 +125,13 @@ xe-y += xe_bb.o \
 	xe_wopcm.o
 
 xe-$(CONFIG_HMM_MIRROR) += xe_hmm.o
-xe-$(CONFIG_DRM_GPUSVM) += xe_svm.o
+xe-$(CONFIG_DRM_XE_GPUSVM) += xe_svm.o
 
 # graphics hardware monitoring (HWMON) support
 xe-$(CONFIG_HWMON) += xe_hwmon.o
 
 xe-$(CONFIG_PERF_EVENTS) += xe_pmu.o
+xe-$(CONFIG_CONFIGFS_FS) += xe_configfs.o
 
 # graphics virtualization (SR-IOV) support
 xe-y += \
@@ -185,7 +186,7 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	display/intel_fbdev_fb.o \
 	display/xe_display.o \
 	display/xe_display_misc.o \
-	display/xe_display_rps.o \
+	display/xe_display_rpm.o \
 	display/xe_display_wa.o \
 	display/xe_dsb_buffer.o \
 	display/xe_fb_pin.o \
@@ -196,7 +197,6 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 # SOC code shared with i915
 xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	i915-soc/intel_dram.o \
-	i915-soc/intel_pch.o \
 	i915-soc/intel_rom.o
 
 # Display code shared with i915
@@ -271,6 +271,7 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	i915-display/intel_panel.o \
 	i915-display/intel_pfit.o \
 	i915-display/intel_pmdemand.o \
+	i915-display/intel_pch.o \
 	i915-display/intel_pps.o \
 	i915-display/intel_psr.o \
 	i915-display/intel_qp_tables.o \
diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
index ec516e838ee8..448afb86e05c 100644
--- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
@@ -141,6 +141,7 @@ enum xe_guc_action {
 	XE_GUC_ACTION_CLIENT_SOFT_RESET = 0x5507,
 	XE_GUC_ACTION_SET_ENG_UTIL_BUFF = 0x550A,
 	XE_GUC_ACTION_SET_DEVICE_ENGINE_ACTIVITY_BUFFER = 0x550C,
+	XE_GUC_ACTION_SET_FUNCTION_ENGINE_ACTIVITY_BUFFER = 0x550D,
 	XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR = 0x6000,
 	XE_GUC_ACTION_REPORT_PAGE_FAULT_REQ_DESC = 0x6002,
 	XE_GUC_ACTION_PAGE_FAULT_RES_DESC = 0x6003,
diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
index d633f1c739e4..7de8f827281f 100644
--- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
@@ -367,6 +367,7 @@ enum xe_guc_klv_ids {
 	GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE			= 0x9008,
 	GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET				= 0x9009,
 	GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO					= 0x900a,
+	GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH					= 0x900b,
 };
 
 #endif
diff --git a/drivers/gpu/drm/xe/compat-i915-headers/gt/intel_rps.h b/drivers/gpu/drm/xe/compat-i915-headers/gt/intel_rps.h
deleted file mode 100644
index 21fec9cc837c..000000000000
--- a/drivers/gpu/drm/xe/compat-i915-headers/gt/intel_rps.h
+++ /dev/null
@@ -1,11 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright © 2023 Intel Corporation
- */
-
-#ifndef __INTEL_RPS_H__
-#define __INTEL_RPS_H__
-
-#define gen5_rps_irq_handler(x) ({})
-
-#endif /* __INTEL_RPS_H__ */
diff --git a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
index dfec5108d2c3..9b7572e06f34 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
@@ -13,7 +13,6 @@
 #include <drm/drm_drv.h>
 
 #include "i915_utils.h"
-#include "intel_runtime_pm.h"
 #include "xe_device.h" /* for xe_device_has_flat_ccs() */
 #include "xe_device_types.h"
 
@@ -22,28 +21,12 @@ static inline struct drm_i915_private *to_i915(const struct drm_device *dev)
 	return container_of(dev, struct drm_i915_private, drm);
 }
 
+/* compat platform checks only for soc/ usage */
 #define IS_PLATFORM(xe, x) ((xe)->info.platform == x)
-#define INTEL_INFO(dev_priv)	(&((dev_priv)->info))
-#define IS_I830(dev_priv)	(dev_priv && 0)
-#define IS_I845G(dev_priv)	(dev_priv && 0)
-#define IS_I85X(dev_priv)	(dev_priv && 0)
-#define IS_I865G(dev_priv)	(dev_priv && 0)
 #define IS_I915G(dev_priv)	(dev_priv && 0)
 #define IS_I915GM(dev_priv)	(dev_priv && 0)
-#define IS_I945G(dev_priv)	(dev_priv && 0)
-#define IS_I945GM(dev_priv)	(dev_priv && 0)
-#define IS_I965G(dev_priv)	(dev_priv && 0)
-#define IS_I965GM(dev_priv)	(dev_priv && 0)
-#define IS_G45(dev_priv)	(dev_priv && 0)
-#define IS_GM45(dev_priv)	(dev_priv && 0)
-#define IS_G4X(dev_priv)	(dev_priv && 0)
 #define IS_PINEVIEW(dev_priv)	(dev_priv && 0)
-#define IS_G33(dev_priv)	(dev_priv && 0)
-#define IS_IRONLAKE(dev_priv)	(dev_priv && 0)
-#define IS_IRONLAKE_M(dev_priv) (dev_priv && 0)
-#define IS_SANDYBRIDGE(dev_priv)	(dev_priv && 0)
 #define IS_IVYBRIDGE(dev_priv)	(dev_priv && 0)
-#define IS_IVB_GT1(dev_priv)	(dev_priv && 0)
 #define IS_VALLEYVIEW(dev_priv)	(dev_priv && 0)
 #define IS_CHERRYVIEW(dev_priv)	(dev_priv && 0)
 #define IS_HASWELL(dev_priv)	(dev_priv && 0)
@@ -71,39 +54,10 @@ static inline struct drm_i915_private *to_i915(const struct drm_device *dev)
 
 #define IS_HASWELL_ULT(dev_priv) (dev_priv && 0)
 #define IS_BROADWELL_ULT(dev_priv) (dev_priv && 0)
-#define IS_BROADWELL_ULX(dev_priv) (dev_priv && 0)
 
 #define IS_MOBILE(xe) (xe && 0)
 
-#define IS_TIGERLAKE_UY(xe) (xe && 0)
-#define IS_COMETLAKE_ULX(xe) (xe && 0)
-#define IS_COFFEELAKE_ULX(xe) (xe && 0)
-#define IS_KABYLAKE_ULX(xe) (xe && 0)
-#define IS_SKYLAKE_ULX(xe) (xe && 0)
-#define IS_HASWELL_ULX(xe) (xe && 0)
-#define IS_COMETLAKE_ULT(xe) (xe && 0)
-#define IS_COFFEELAKE_ULT(xe) (xe && 0)
-#define IS_KABYLAKE_ULT(xe) (xe && 0)
-#define IS_SKYLAKE_ULT(xe) (xe && 0)
-
-#define IS_DG2_G10(xe) ((xe)->info.subplatform == XE_SUBPLATFORM_DG2_G10)
-#define IS_DG2_G11(xe) ((xe)->info.subplatform == XE_SUBPLATFORM_DG2_G11)
-#define IS_DG2_G12(xe) ((xe)->info.subplatform == XE_SUBPLATFORM_DG2_G12)
-#define IS_RAPTORLAKE_U(xe) ((xe)->info.subplatform == XE_SUBPLATFORM_ALDERLAKE_P_RPLU)
-#define IS_ICL_WITH_PORT_F(xe) (xe && 0)
 #define HAS_FLAT_CCS(xe) (xe_device_has_flat_ccs(xe))
-
 #define HAS_128_BYTE_Y_TILING(xe) (xe || 1)
 
-#ifdef CONFIG_ARM64
-/*
- * arm64 indirectly includes linux/rtc.h,
- * which defines a irq_lock, so include it
- * here before #define-ing it
- */
-#include <linux/rtc.h>
-#endif
-
-#define irq_lock irq.lock
-
 #endif
diff --git a/drivers/gpu/drm/xe/compat-i915-headers/intel_runtime_pm.h b/drivers/gpu/drm/xe/compat-i915-headers/intel_runtime_pm.h
deleted file mode 100644
index 274042bff1be..000000000000
--- a/drivers/gpu/drm/xe/compat-i915-headers/intel_runtime_pm.h
+++ /dev/null
@@ -1,76 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright © 2023 Intel Corporation
- */
-
-#ifndef __INTEL_RUNTIME_PM_H__
-#define __INTEL_RUNTIME_PM_H__
-
-#include "intel_wakeref.h"
-#include "xe_device_types.h"
-#include "xe_pm.h"
-
-#define intel_runtime_pm xe_runtime_pm
-
-static inline void disable_rpm_wakeref_asserts(void *rpm)
-{
-}
-
-static inline void enable_rpm_wakeref_asserts(void *rpm)
-{
-}
-
-static inline bool
-intel_runtime_pm_suspended(struct xe_runtime_pm *pm)
-{
-	struct xe_device *xe = container_of(pm, struct xe_device, runtime_pm);
-
-	return pm_runtime_suspended(xe->drm.dev);
-}
-
-static inline intel_wakeref_t intel_runtime_pm_get(struct xe_runtime_pm *pm)
-{
-	struct xe_device *xe = container_of(pm, struct xe_device, runtime_pm);
-
-	return xe_pm_runtime_resume_and_get(xe) ? INTEL_WAKEREF_DEF : NULL;
-}
-
-static inline intel_wakeref_t intel_runtime_pm_get_if_in_use(struct xe_runtime_pm *pm)
-{
-	struct xe_device *xe = container_of(pm, struct xe_device, runtime_pm);
-
-	return xe_pm_runtime_get_if_in_use(xe) ? INTEL_WAKEREF_DEF : NULL;
-}
-
-static inline intel_wakeref_t intel_runtime_pm_get_noresume(struct xe_runtime_pm *pm)
-{
-	struct xe_device *xe = container_of(pm, struct xe_device, runtime_pm);
-
-	xe_pm_runtime_get_noresume(xe);
-
-	return INTEL_WAKEREF_DEF;
-}
-
-static inline void intel_runtime_pm_put_unchecked(struct xe_runtime_pm *pm)
-{
-	struct xe_device *xe = container_of(pm, struct xe_device, runtime_pm);
-
-	xe_pm_runtime_put(xe);
-}
-
-static inline void intel_runtime_pm_put(struct xe_runtime_pm *pm, intel_wakeref_t wakeref)
-{
-	if (wakeref)
-		intel_runtime_pm_put_unchecked(pm);
-}
-
-#define intel_runtime_pm_get_raw intel_runtime_pm_get
-#define intel_runtime_pm_put_raw intel_runtime_pm_put
-#define assert_rpm_wakelock_held(x) do { } while (0)
-#define assert_rpm_raw_wakeref_held(x) do { } while (0)
-
-#define with_intel_runtime_pm(rpm, wf) \
-	for ((wf) = intel_runtime_pm_get(rpm); (wf); \
-	     intel_runtime_pm_put((rpm), (wf)), (wf) = NULL)
-
-#endif
diff --git a/drivers/gpu/drm/xe/compat-i915-headers/soc/intel_pch.h b/drivers/gpu/drm/xe/compat-i915-headers/soc/intel_pch.h
deleted file mode 100644
index 9c46556d33a4..000000000000
--- a/drivers/gpu/drm/xe/compat-i915-headers/soc/intel_pch.h
+++ /dev/null
@@ -1,6 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright © 2023 Intel Corporation
- */
-
-#include "../../../i915/soc/intel_pch.h"
diff --git a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
index 3a1e505ff182..e8191562d122 100644
--- a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
+++ b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
@@ -45,7 +45,7 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 					   NULL, size,
 					   ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
 					   XE_BO_FLAG_STOLEN |
-					   XE_BO_FLAG_GGTT | XE_BO_FLAG_PINNED);
+					   XE_BO_FLAG_GGTT);
 		if (!IS_ERR(obj))
 			drm_info(&xe->drm, "Allocated fbdev into stolen\n");
 		else
@@ -56,7 +56,7 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 		obj = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe), NULL, size,
 					   ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
 					   XE_BO_FLAG_VRAM_IF_DGFX(xe_device_get_root_tile(xe)) |
-					   XE_BO_FLAG_GGTT | XE_BO_FLAG_PINNED);
+					   XE_BO_FLAG_GGTT);
 	}
 
 	if (IS_ERR(obj)) {
@@ -79,11 +79,11 @@ err:
 	return ERR_CAST(fb);
 }
 
-int intel_fbdev_fb_fill_info(struct drm_i915_private *i915, struct fb_info *info,
-			      struct drm_gem_object *_obj, struct i915_vma *vma)
+int intel_fbdev_fb_fill_info(struct intel_display *display, struct fb_info *info,
+			     struct drm_gem_object *_obj, struct i915_vma *vma)
 {
 	struct xe_bo *obj = gem_to_xe_bo(_obj);
-	struct pci_dev *pdev = to_pci_dev(i915->drm.dev);
+	struct pci_dev *pdev = to_pci_dev(display->drm->dev);
 
 	if (!(obj->flags & XE_BO_FLAG_SYSTEM)) {
 		if (obj->flags & XE_BO_FLAG_STOLEN)
diff --git a/drivers/gpu/drm/xe/display/xe_display.c b/drivers/gpu/drm/xe/display/xe_display.c
index 0b0aca7a25af..68f064f33d4b 100644
--- a/drivers/gpu/drm/xe/display/xe_display.c
+++ b/drivers/gpu/drm/xe/display/xe_display.c
@@ -133,9 +133,6 @@ int xe_display_init_early(struct xe_device *xe)
 	/* Fake uncore lock */
 	spin_lock_init(&xe->uncore.lock);
 
-	/* This must be called before any calls to HAS_PCH_* */
-	intel_detect_pch(xe);
-
 	intel_display_driver_early_probe(display);
 
 	/* Early display init.. */
@@ -147,7 +144,7 @@ int xe_display_init_early(struct xe_device *xe)
 	 */
 	intel_dram_detect(xe);
 
-	intel_bw_init_hw(xe);
+	intel_bw_init_hw(display);
 
 	intel_display_device_info_runtime_init(display);
 
@@ -173,7 +170,7 @@ static void xe_display_fini(void *arg)
 	struct xe_device *xe = arg;
 	struct intel_display *display = &xe->display;
 
-	intel_hpd_poll_fini(xe);
+	intel_hpd_poll_fini(display);
 	intel_hdcp_component_fini(display);
 	intel_audio_deinit(display);
 	intel_display_driver_remove(display);
@@ -220,11 +217,13 @@ void xe_display_unregister(struct xe_device *xe)
 
 void xe_display_irq_handler(struct xe_device *xe, u32 master_ctl)
 {
+	struct intel_display *display = &xe->display;
+
 	if (!xe->info.probe_display)
 		return;
 
 	if (master_ctl & DISPLAY_IRQ)
-		gen11_display_irq_handler(xe);
+		gen11_display_irq_handler(display);
 }
 
 void xe_display_irq_enable(struct xe_device *xe, u32 gu_misc_iir)
@@ -240,19 +239,23 @@ void xe_display_irq_enable(struct xe_device *xe, u32 gu_misc_iir)
 
 void xe_display_irq_reset(struct xe_device *xe)
 {
+	struct intel_display *display = &xe->display;
+
 	if (!xe->info.probe_display)
 		return;
 
-	gen11_display_irq_reset(xe);
+	gen11_display_irq_reset(display);
 }
 
 void xe_display_irq_postinstall(struct xe_device *xe, struct xe_gt *gt)
 {
+	struct intel_display *display = &xe->display;
+
 	if (!xe->info.probe_display)
 		return;
 
 	if (gt->info.id == XE_GT0)
-		gen11_de_irq_postinstall(xe);
+		gen11_de_irq_postinstall(display);
 }
 
 static bool suspend_to_idle(void)
@@ -305,7 +308,7 @@ static void xe_display_enable_d3cold(struct xe_device *xe)
 	intel_dmc_suspend(display);
 
 	if (has_display(xe))
-		intel_hpd_poll_enable(xe);
+		intel_hpd_poll_enable(display);
 }
 
 static void xe_display_disable_d3cold(struct xe_device *xe)
@@ -322,10 +325,10 @@ static void xe_display_disable_d3cold(struct xe_device *xe)
 
 	intel_display_driver_init_hw(display);
 
-	intel_hpd_init(xe);
+	intel_hpd_init(display);
 
 	if (has_display(xe))
-		intel_hpd_poll_disable(xe);
+		intel_hpd_poll_disable(display);
 
 	intel_opregion_resume(display);
 
@@ -355,7 +358,7 @@ void xe_display_pm_suspend(struct xe_device *xe)
 
 	xe_display_flush_cleanup_work(xe);
 
-	intel_hpd_cancel_work(xe);
+	intel_hpd_cancel_work(display);
 
 	if (has_display(xe)) {
 		intel_display_driver_suspend_access(display);
@@ -385,7 +388,7 @@ void xe_display_pm_shutdown(struct xe_device *xe)
 
 	xe_display_flush_cleanup_work(xe);
 	intel_dp_mst_suspend(display);
-	intel_hpd_cancel_work(xe);
+	intel_hpd_cancel_work(display);
 
 	if (has_display(xe))
 		intel_display_driver_suspend_access(display);
@@ -400,6 +403,8 @@ void xe_display_pm_shutdown(struct xe_device *xe)
 
 void xe_display_pm_runtime_suspend(struct xe_device *xe)
 {
+	struct intel_display *display = &xe->display;
+
 	if (!xe->info.probe_display)
 		return;
 
@@ -408,7 +413,7 @@ void xe_display_pm_runtime_suspend(struct xe_device *xe)
 		return;
 	}
 
-	intel_hpd_poll_enable(xe);
+	intel_hpd_poll_enable(display);
 }
 
 void xe_display_pm_suspend_late(struct xe_device *xe)
@@ -482,7 +487,7 @@ void xe_display_pm_resume(struct xe_device *xe)
 	if (has_display(xe))
 		intel_display_driver_resume_access(display);
 
-	intel_hpd_init(xe);
+	intel_hpd_init(display);
 
 	if (has_display(xe)) {
 		intel_display_driver_resume(display);
@@ -491,7 +496,7 @@ void xe_display_pm_resume(struct xe_device *xe)
 	}
 
 	if (has_display(xe))
-		intel_hpd_poll_disable(xe);
+		intel_hpd_poll_disable(display);
 
 	intel_opregion_resume(display);
 
@@ -502,6 +507,8 @@ void xe_display_pm_resume(struct xe_device *xe)
 
 void xe_display_pm_runtime_resume(struct xe_device *xe)
 {
+	struct intel_display *display = &xe->display;
+
 	if (!xe->info.probe_display)
 		return;
 
@@ -510,9 +517,9 @@ void xe_display_pm_runtime_resume(struct xe_device *xe)
 		return;
 	}
 
-	intel_hpd_init(xe);
-	intel_hpd_poll_disable(xe);
-	skl_watermark_ipc_update(xe);
+	intel_hpd_init(display);
+	intel_hpd_poll_disable(display);
+	skl_watermark_ipc_update(display);
 }
 
 
diff --git a/drivers/gpu/drm/xe/display/xe_display_rpm.c b/drivers/gpu/drm/xe/display/xe_display_rpm.c
new file mode 100644
index 000000000000..1955153aadba
--- /dev/null
+++ b/drivers/gpu/drm/xe/display/xe_display_rpm.c
@@ -0,0 +1,71 @@
+// SPDX-License-Identifier: MIT
+/* Copyright © 2025 Intel Corporation */
+
+#include "intel_display_rpm.h"
+#include "xe_device_types.h"
+#include "xe_pm.h"
+
+static struct xe_device *display_to_xe(struct intel_display *display)
+{
+	return container_of(display, struct xe_device, display);
+}
+
+struct ref_tracker *intel_display_rpm_get_raw(struct intel_display *display)
+{
+	return intel_display_rpm_get(display);
+}
+
+void intel_display_rpm_put_raw(struct intel_display *display, struct ref_tracker *wakeref)
+{
+	intel_display_rpm_put(display, wakeref);
+}
+
+struct ref_tracker *intel_display_rpm_get(struct intel_display *display)
+{
+	return xe_pm_runtime_resume_and_get(display_to_xe(display)) ? INTEL_WAKEREF_DEF : NULL;
+}
+
+struct ref_tracker *intel_display_rpm_get_if_in_use(struct intel_display *display)
+{
+	return xe_pm_runtime_get_if_in_use(display_to_xe(display)) ? INTEL_WAKEREF_DEF : NULL;
+}
+
+struct ref_tracker *intel_display_rpm_get_noresume(struct intel_display *display)
+{
+	xe_pm_runtime_get_noresume(display_to_xe(display));
+
+	return INTEL_WAKEREF_DEF;
+}
+
+void intel_display_rpm_put(struct intel_display *display, struct ref_tracker *wakeref)
+{
+	if (wakeref)
+		xe_pm_runtime_put(display_to_xe(display));
+}
+
+void intel_display_rpm_put_unchecked(struct intel_display *display)
+{
+	xe_pm_runtime_put(display_to_xe(display));
+}
+
+bool intel_display_rpm_suspended(struct intel_display *display)
+{
+	struct xe_device *xe = display_to_xe(display);
+
+	return pm_runtime_suspended(xe->drm.dev);
+}
+
+void assert_display_rpm_held(struct intel_display *display)
+{
+	/* FIXME */
+}
+
+void intel_display_rpm_assert_block(struct intel_display *display)
+{
+	/* FIXME */
+}
+
+void intel_display_rpm_assert_unblock(struct intel_display *display)
+{
+	/* FIXME */
+}
diff --git a/drivers/gpu/drm/xe/display/xe_display_rps.c b/drivers/gpu/drm/xe/display/xe_display_rps.c
deleted file mode 100644
index fa616f9688a5..000000000000
--- a/drivers/gpu/drm/xe/display/xe_display_rps.c
+++ /dev/null
@@ -1,17 +0,0 @@
-// SPDX-License-Identifier: MIT
-/*
- * Copyright © 2023 Intel Corporation
- */
-
-#include "intel_display_rps.h"
-
-void intel_display_rps_boost_after_vblank(struct drm_crtc *crtc,
-					  struct dma_fence *fence)
-{
-}
-
-void intel_display_rps_mark_interactive(struct intel_display *display,
-					struct intel_atomic_state *state,
-					bool interactive)
-{
-}
diff --git a/drivers/gpu/drm/xe/display/xe_display_wa.c b/drivers/gpu/drm/xe/display/xe_display_wa.c
index 68e3d1959ad6..2933ca97d673 100644
--- a/drivers/gpu/drm/xe/display/xe_display_wa.c
+++ b/drivers/gpu/drm/xe/display/xe_display_wa.c
@@ -10,7 +10,9 @@
 
 #include <generated/xe_wa_oob.h>
 
-bool intel_display_needs_wa_16023588340(struct drm_i915_private *i915)
+bool intel_display_needs_wa_16023588340(struct intel_display *display)
 {
-	return XE_WA(xe_root_mmio_gt(i915), 16023588340);
+	struct xe_device *xe = to_xe_device(display->drm);
+
+	return XE_WA(xe_root_mmio_gt(xe), 16023588340);
 }
diff --git a/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c b/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c
index 7c02323e9531..b35a6f201d4a 100644
--- a/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c
+++ b/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c
@@ -9,7 +9,6 @@
 
 #include "abi/gsc_command_header_abi.h"
 #include "intel_hdcp_gsc.h"
-#include "intel_hdcp_gsc_message.h"
 #include "xe_bo.h"
 #include "xe_device.h"
 #include "xe_device_types.h"
@@ -22,7 +21,8 @@
 
 #define HECI_MEADDRESS_HDCP 18
 
-struct intel_hdcp_gsc_message {
+struct intel_hdcp_gsc_context {
+	struct xe_device *xe;
 	struct xe_bo *hdcp_bo;
 	u64 hdcp_cmd_in;
 	u64 hdcp_cmd_out;
@@ -30,14 +30,9 @@ struct intel_hdcp_gsc_message {
 
 #define HDCP_GSC_HEADER_SIZE sizeof(struct intel_gsc_mtl_header)
 
-bool intel_hdcp_gsc_cs_required(struct intel_display *display)
+bool intel_hdcp_gsc_check_status(struct drm_device *drm)
 {
-	return DISPLAY_VER(display) >= 14;
-}
-
-bool intel_hdcp_gsc_check_status(struct intel_display *display)
-{
-	struct xe_device *xe = to_xe_device(display->drm);
+	struct xe_device *xe = to_xe_device(drm);
 	struct xe_tile *tile = xe_device_get_root_tile(xe);
 	struct xe_gt *gt = tile->media_gt;
 	struct xe_gsc *gsc = &gt->uc.gsc;
@@ -69,10 +64,9 @@ out:
 }
 
 /*This function helps allocate memory for the command that we will send to gsc cs */
-static int intel_hdcp_gsc_initialize_message(struct intel_display *display,
-					     struct intel_hdcp_gsc_message *hdcp_message)
+static int intel_hdcp_gsc_initialize_message(struct xe_device *xe,
+					     struct intel_hdcp_gsc_context *gsc_context)
 {
-	struct xe_device *xe = to_xe_device(display->drm);
 	struct xe_bo *bo = NULL;
 	u64 cmd_in, cmd_out;
 	int ret = 0;
@@ -84,7 +78,7 @@ static int intel_hdcp_gsc_initialize_message(struct intel_display *display,
 				  XE_BO_FLAG_GGTT);
 
 	if (IS_ERR(bo)) {
-		drm_err(display->drm, "Failed to allocate bo for HDCP streaming command!\n");
+		drm_err(&xe->drm, "Failed to allocate bo for HDCP streaming command!\n");
 		ret = PTR_ERR(bo);
 		goto out;
 	}
@@ -93,104 +87,60 @@ static int intel_hdcp_gsc_initialize_message(struct intel_display *display,
 	cmd_out = cmd_in + PAGE_SIZE;
 	xe_map_memset(xe, &bo->vmap, 0, 0, bo->size);
 
-	hdcp_message->hdcp_bo = bo;
-	hdcp_message->hdcp_cmd_in = cmd_in;
-	hdcp_message->hdcp_cmd_out = cmd_out;
+	gsc_context->hdcp_bo = bo;
+	gsc_context->hdcp_cmd_in = cmd_in;
+	gsc_context->hdcp_cmd_out = cmd_out;
+	gsc_context->xe = xe;
+
 out:
 	return ret;
 }
 
-static int intel_hdcp_gsc_hdcp2_init(struct intel_display *display)
+struct intel_hdcp_gsc_context *intel_hdcp_gsc_context_alloc(struct drm_device *drm)
 {
-	struct intel_hdcp_gsc_message *hdcp_message;
+	struct xe_device *xe = to_xe_device(drm);
+	struct intel_hdcp_gsc_context *gsc_context;
 	int ret;
 
-	hdcp_message = kzalloc(sizeof(*hdcp_message), GFP_KERNEL);
-
-	if (!hdcp_message)
-		return -ENOMEM;
+	gsc_context = kzalloc(sizeof(*gsc_context), GFP_KERNEL);
+	if (!gsc_context)
+		return ERR_PTR(-ENOMEM);
 
 	/*
 	 * NOTE: No need to lock the comp mutex here as it is already
 	 * going to be taken before this function called
 	 */
-	ret = intel_hdcp_gsc_initialize_message(display, hdcp_message);
+	ret = intel_hdcp_gsc_initialize_message(xe, gsc_context);
 	if (ret) {
-		drm_err(display->drm, "Could not initialize hdcp_message\n");
-		kfree(hdcp_message);
-		return ret;
+		drm_err(&xe->drm, "Could not initialize gsc_context\n");
+		kfree(gsc_context);
+		gsc_context = ERR_PTR(ret);
 	}
 
-	display->hdcp.hdcp_message = hdcp_message;
-	return ret;
-}
-
-static const struct i915_hdcp_ops gsc_hdcp_ops = {
-	.initiate_hdcp2_session = intel_hdcp_gsc_initiate_session,
-	.verify_receiver_cert_prepare_km =
-				intel_hdcp_gsc_verify_receiver_cert_prepare_km,
-	.verify_hprime = intel_hdcp_gsc_verify_hprime,
-	.store_pairing_info = intel_hdcp_gsc_store_pairing_info,
-	.initiate_locality_check = intel_hdcp_gsc_initiate_locality_check,
-	.verify_lprime = intel_hdcp_gsc_verify_lprime,
-	.get_session_key = intel_hdcp_gsc_get_session_key,
-	.repeater_check_flow_prepare_ack =
-				intel_hdcp_gsc_repeater_check_flow_prepare_ack,
-	.verify_mprime = intel_hdcp_gsc_verify_mprime,
-	.enable_hdcp_authentication = intel_hdcp_gsc_enable_authentication,
-	.close_hdcp_session = intel_hdcp_gsc_close_session,
-};
-
-int intel_hdcp_gsc_init(struct intel_display *display)
-{
-	struct i915_hdcp_arbiter *data;
-	int ret;
-
-	data = kzalloc(sizeof(*data), GFP_KERNEL);
-	if (!data)
-		return -ENOMEM;
-
-	mutex_lock(&display->hdcp.hdcp_mutex);
-	display->hdcp.arbiter = data;
-	display->hdcp.arbiter->hdcp_dev = display->drm->dev;
-	display->hdcp.arbiter->ops = &gsc_hdcp_ops;
-	ret = intel_hdcp_gsc_hdcp2_init(display);
-	if (ret)
-		kfree(data);
-
-	mutex_unlock(&display->hdcp.hdcp_mutex);
-
-	return ret;
+	return gsc_context;
 }
 
-void intel_hdcp_gsc_fini(struct intel_display *display)
+void intel_hdcp_gsc_context_free(struct intel_hdcp_gsc_context *gsc_context)
 {
-	struct intel_hdcp_gsc_message *hdcp_message =
-					display->hdcp.hdcp_message;
-	struct i915_hdcp_arbiter *arb = display->hdcp.arbiter;
-
-	if (hdcp_message) {
-		xe_bo_unpin_map_no_vm(hdcp_message->hdcp_bo);
-		kfree(hdcp_message);
-		display->hdcp.hdcp_message = NULL;
-	}
+	if (!gsc_context)
+		return;
 
-	kfree(arb);
-	display->hdcp.arbiter = NULL;
+	xe_bo_unpin_map_no_vm(gsc_context->hdcp_bo);
+	kfree(gsc_context);
 }
 
 static int xe_gsc_send_sync(struct xe_device *xe,
-			    struct intel_hdcp_gsc_message *hdcp_message,
+			    struct intel_hdcp_gsc_context *gsc_context,
 			    u32 msg_size_in, u32 msg_size_out,
 			    u32 addr_out_off)
 {
-	struct xe_gt *gt = hdcp_message->hdcp_bo->tile->media_gt;
-	struct iosys_map *map = &hdcp_message->hdcp_bo->vmap;
+	struct xe_gt *gt = gsc_context->hdcp_bo->tile->media_gt;
+	struct iosys_map *map = &gsc_context->hdcp_bo->vmap;
 	struct xe_gsc *gsc = &gt->uc.gsc;
 	int ret;
 
-	ret = xe_gsc_pkt_submit_kernel(gsc, hdcp_message->hdcp_cmd_in, msg_size_in,
-				       hdcp_message->hdcp_cmd_out, msg_size_out);
+	ret = xe_gsc_pkt_submit_kernel(gsc, gsc_context->hdcp_cmd_in, msg_size_in,
+				       gsc_context->hdcp_cmd_out, msg_size_out);
 	if (ret) {
 		drm_err(&xe->drm, "failed to send gsc HDCP msg (%d)\n", ret);
 		return ret;
@@ -205,12 +155,12 @@ static int xe_gsc_send_sync(struct xe_device *xe,
 	return ret;
 }
 
-ssize_t intel_hdcp_gsc_msg_send(struct xe_device *xe, u8 *msg_in,
-				size_t msg_in_len, u8 *msg_out,
-				size_t msg_out_len)
+ssize_t intel_hdcp_gsc_msg_send(struct intel_hdcp_gsc_context *gsc_context,
+				void *msg_in, size_t msg_in_len,
+				void *msg_out, size_t msg_out_len)
 {
+	struct xe_device *xe = gsc_context->xe;
 	const size_t max_msg_size = PAGE_SIZE - HDCP_GSC_HEADER_SIZE;
-	struct intel_hdcp_gsc_message *hdcp_message;
 	u64 host_session_id;
 	u32 msg_size_in, msg_size_out;
 	u32 addr_out_off, addr_in_wr_off = 0;
@@ -223,15 +173,14 @@ ssize_t intel_hdcp_gsc_msg_send(struct xe_device *xe, u8 *msg_in,
 
 	msg_size_in = msg_in_len + HDCP_GSC_HEADER_SIZE;
 	msg_size_out = msg_out_len + HDCP_GSC_HEADER_SIZE;
-	hdcp_message = xe->display.hdcp.hdcp_message;
 	addr_out_off = PAGE_SIZE;
 
 	host_session_id = xe_gsc_create_host_session_id();
 	xe_pm_runtime_get_noresume(xe);
-	addr_in_wr_off = xe_gsc_emit_header(xe, &hdcp_message->hdcp_bo->vmap,
+	addr_in_wr_off = xe_gsc_emit_header(xe, &gsc_context->hdcp_bo->vmap,
 					    addr_in_wr_off, HECI_MEADDRESS_HDCP,
 					    host_session_id, msg_in_len);
-	xe_map_memcpy_to(xe, &hdcp_message->hdcp_bo->vmap, addr_in_wr_off,
+	xe_map_memcpy_to(xe, &gsc_context->hdcp_bo->vmap, addr_in_wr_off,
 			 msg_in, msg_in_len);
 	/*
 	 * Keep sending request in case the pending bit is set no need to add
@@ -240,7 +189,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct xe_device *xe, u8 *msg_in,
 	 * 20 times each message 50 ms apart
 	 */
 	do {
-		ret = xe_gsc_send_sync(xe, hdcp_message, msg_size_in, msg_size_out,
+		ret = xe_gsc_send_sync(xe, gsc_context, msg_size_in, msg_size_out,
 				       addr_out_off);
 
 		/* Only try again if gsc says so */
@@ -254,7 +203,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct xe_device *xe, u8 *msg_in,
 	if (ret)
 		goto out;
 
-	xe_map_memcpy_from(xe, msg_out, &hdcp_message->hdcp_bo->vmap,
+	xe_map_memcpy_from(xe, msg_out, &gsc_context->hdcp_bo->vmap,
 			   addr_out_off + HDCP_GSC_HEADER_SIZE,
 			   msg_out_len);
 
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index 4ca0cb571194..6502b8274173 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -83,7 +83,7 @@ initial_plane_bo(struct xe_device *xe,
 	if (plane_config->size == 0)
 		return NULL;
 
-	flags = XE_BO_FLAG_PINNED | XE_BO_FLAG_SCANOUT | XE_BO_FLAG_GGTT;
+	flags = XE_BO_FLAG_SCANOUT | XE_BO_FLAG_GGTT;
 
 	base = round_down(plane_config->base, page_size);
 	if (IS_DGFX(xe)) {
diff --git a/drivers/gpu/drm/xe/instructions/xe_alu_commands.h b/drivers/gpu/drm/xe/instructions/xe_alu_commands.h
new file mode 100644
index 000000000000..2987b10d3e16
--- /dev/null
+++ b/drivers/gpu/drm/xe/instructions/xe_alu_commands.h
@@ -0,0 +1,79 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_ALU_COMMANDS_H_
+#define _XE_ALU_COMMANDS_H_
+
+#include "instructions/xe_instr_defs.h"
+
+/* Instruction Opcodes */
+#define CS_ALU_OPCODE_NOOP			0x000
+#define CS_ALU_OPCODE_FENCE_RD			0x001
+#define CS_ALU_OPCODE_FENCE_WR			0x002
+#define CS_ALU_OPCODE_LOAD			0x080
+#define CS_ALU_OPCODE_LOADINV			0x480
+#define CS_ALU_OPCODE_LOAD0			0x081
+#define CS_ALU_OPCODE_LOAD1			0x481
+#define CS_ALU_OPCODE_LOADIND			0x082
+#define CS_ALU_OPCODE_ADD			0x100
+#define CS_ALU_OPCODE_SUB			0x101
+#define CS_ALU_OPCODE_AND			0x102
+#define CS_ALU_OPCODE_OR			0x103
+#define CS_ALU_OPCODE_XOR			0x104
+#define CS_ALU_OPCODE_SHL			0x105
+#define CS_ALU_OPCODE_SHR			0x106
+#define CS_ALU_OPCODE_SAR			0x107
+#define CS_ALU_OPCODE_STORE			0x180
+#define CS_ALU_OPCODE_STOREINV			0x580
+#define CS_ALU_OPCODE_STOREIND			0x181
+
+/* Instruction Operands */
+#define CS_ALU_OPERAND_REG(n)			REG_FIELD_PREP(GENMASK(3, 0), (n))
+#define CS_ALU_OPERAND_REG0			0x0
+#define CS_ALU_OPERAND_REG1			0x1
+#define CS_ALU_OPERAND_REG2			0x2
+#define CS_ALU_OPERAND_REG3			0x3
+#define CS_ALU_OPERAND_REG4			0x4
+#define CS_ALU_OPERAND_REG5			0x5
+#define CS_ALU_OPERAND_REG6			0x6
+#define CS_ALU_OPERAND_REG7			0x7
+#define CS_ALU_OPERAND_REG8			0x8
+#define CS_ALU_OPERAND_REG9			0x9
+#define CS_ALU_OPERAND_REG10			0xa
+#define CS_ALU_OPERAND_REG11			0xb
+#define CS_ALU_OPERAND_REG12			0xc
+#define CS_ALU_OPERAND_REG13			0xd
+#define CS_ALU_OPERAND_REG14			0xe
+#define CS_ALU_OPERAND_REG15			0xf
+#define CS_ALU_OPERAND_SRCA			0x20
+#define CS_ALU_OPERAND_SRCB			0x21
+#define CS_ALU_OPERAND_ACCU			0x31
+#define CS_ALU_OPERAND_ZF			0x32
+#define CS_ALU_OPERAND_CF			0x33
+#define CS_ALU_OPERAND_NA			0 /* N/A operand */
+
+/* Command Streamer ALU Instructions */
+#define CS_ALU_INSTR(opcode, op1, op2)		(REG_FIELD_PREP(GENMASK(31, 20), (opcode)) | \
+						 REG_FIELD_PREP(GENMASK(19, 10), (op1)) | \
+						 REG_FIELD_PREP(GENMASK(9, 0), (op2)))
+
+#define __CS_ALU_INSTR(opcode, op1, op2)	CS_ALU_INSTR(CS_ALU_OPCODE_##opcode, \
+							     CS_ALU_OPERAND_##op1, \
+							     CS_ALU_OPERAND_##op2)
+
+#define CS_ALU_INSTR_NOOP			__CS_ALU_INSTR(NOOP, NA, NA)
+#define CS_ALU_INSTR_LOAD(op1, op2)		__CS_ALU_INSTR(LOAD, op1, op2)
+#define CS_ALU_INSTR_LOADINV(op1, op2)		__CS_ALU_INSTR(LOADINV, op1, op2)
+#define CS_ALU_INSTR_LOAD0(op1)			__CS_ALU_INSTR(LOAD0, op1, NA)
+#define CS_ALU_INSTR_LOAD1(op1)			__CS_ALU_INSTR(LOAD1, op1, NA)
+#define CS_ALU_INSTR_ADD			__CS_ALU_INSTR(ADD, NA, NA)
+#define CS_ALU_INSTR_SUB			__CS_ALU_INSTR(SUB, NA, NA)
+#define CS_ALU_INSTR_AND			__CS_ALU_INSTR(AND, NA, NA)
+#define CS_ALU_INSTR_OR				__CS_ALU_INSTR(OR, NA, NA)
+#define CS_ALU_INSTR_XOR			__CS_ALU_INSTR(XOR, NA, NA)
+#define CS_ALU_INSTR_STORE(op1, op2)		__CS_ALU_INSTR(STORE, op1, op2)
+#define CS_ALU_INSTR_STOREINV(op1, op2)		__CS_ALU_INSTR(STOREINV, op1, op2)
+
+#endif
diff --git a/drivers/gpu/drm/xe/instructions/xe_gfxpipe_commands.h b/drivers/gpu/drm/xe/instructions/xe_gfxpipe_commands.h
index 31d28a67ef6a..457881af8af9 100644
--- a/drivers/gpu/drm/xe/instructions/xe_gfxpipe_commands.h
+++ b/drivers/gpu/drm/xe/instructions/xe_gfxpipe_commands.h
@@ -137,6 +137,7 @@
 #define CMD_3DSTATE_CLIP_MESH			GFXPIPE_3D_CMD(0x0, 0x81)
 #define CMD_3DSTATE_SBE_MESH			GFXPIPE_3D_CMD(0x0, 0x82)
 #define CMD_3DSTATE_CPSIZE_CONTROL_BUFFER	GFXPIPE_3D_CMD(0x0, 0x83)
+#define CMD_3DSTATE_COARSE_PIXEL		GFXPIPE_3D_CMD(0x0, 0x89)
 
 #define CMD_3DSTATE_DRAWING_RECTANGLE		GFXPIPE_3D_CMD(0x1, 0x0)
 #define CMD_3DSTATE_CHROMA_KEY			GFXPIPE_3D_CMD(0x1, 0x4)
diff --git a/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h b/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h
index a255946b6f77..8cfcd3360896 100644
--- a/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h
+++ b/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h
@@ -41,6 +41,7 @@
 
 #define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|((len)-2))
 
+#define	  PIPE_CONTROL0_L3_READ_ONLY_CACHE_INVALIDATE	BIT(10)	/* gen12 */
 #define	  PIPE_CONTROL0_HDC_PIPELINE_FLUSH		BIT(9)	/* gen12 */
 
 #define   PIPE_CONTROL_COMMAND_CACHE_INVALIDATE		(1<<29)
diff --git a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
index 167fb0f742de..eba582058d55 100644
--- a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
+++ b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
@@ -32,6 +32,7 @@
 #define MI_BATCH_BUFFER_END		__MI_INSTR(0xA)
 #define MI_TOPOLOGY_FILTER		__MI_INSTR(0xD)
 #define MI_FORCE_WAKEUP			__MI_INSTR(0x1D)
+#define MI_MATH(n)			(__MI_INSTR(0x1A) | XE_INSTR_NUM_DW((n) + 1))
 
 #define MI_STORE_DATA_IMM		__MI_INSTR(0x20)
 #define   MI_SDI_GGTT			REG_BIT(22)
@@ -61,6 +62,10 @@
 #define MI_LOAD_REGISTER_MEM		(__MI_INSTR(0x29) | XE_INSTR_NUM_DW(4))
 #define   MI_LRM_USE_GGTT		REG_BIT(22)
 
+#define MI_LOAD_REGISTER_REG		(__MI_INSTR(0x2a) | XE_INSTR_NUM_DW(3))
+#define   MI_LRR_DST_CS_MMIO		REG_BIT(19)
+#define   MI_LRR_SRC_CS_MMIO		REG_BIT(18)
+
 #define MI_COPY_MEM_MEM			(__MI_INSTR(0x2e) | XE_INSTR_NUM_DW(5))
 #define   MI_COPY_MEM_MEM_SRC_GGTT	REG_BIT(22)
 #define   MI_COPY_MEM_MEM_DST_GGTT	REG_BIT(21)
diff --git a/drivers/gpu/drm/xe/regs/xe_engine_regs.h b/drivers/gpu/drm/xe/regs/xe_engine_regs.h
index fb8ec317b6ee..da713634d6a0 100644
--- a/drivers/gpu/drm/xe/regs/xe_engine_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_engine_regs.h
@@ -188,6 +188,10 @@
 #define   PREEMPT_GPGPU_LEVEL_MASK		PREEMPT_GPGPU_LEVEL(1, 1)
 #define   PREEMPT_3D_OBJECT_LEVEL		REG_BIT(0)
 
+#define CS_GPR_DATA(base, n)			XE_REG((base) + 0x600 + (n) * 4)
+#define CS_GPR_REG(base, n)			CS_GPR_DATA((base), (n) * 2)
+#define CS_GPR_REG_UDW(base, n)			CS_GPR_DATA((base), (n) * 2 + 1)
+
 #define VDBOX_CGCTL3F08(base)			XE_REG((base) + 0x3f08)
 #define   CG3DDISHRS_CLKGATE_DIS		REG_BIT(5)
 
diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
index da1f198ac107..cbb9f7cbcfc0 100644
--- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
@@ -62,7 +62,6 @@
 #define   LE_SSE_MASK				REG_GENMASK(18, 17)
 #define   LE_SSE(value)				REG_FIELD_PREP(LE_SSE_MASK, value)
 #define   LE_COS_MASK				REG_GENMASK(16, 15)
-#define   LE_COS(value)				REG_FIELD_PREP(LE_COS_MASK)
 #define   LE_SCF_MASK				REG_BIT(14)
 #define   LE_SCF(value)				REG_FIELD_PREP(LE_SCF_MASK, value)
 #define   LE_PFM_MASK				REG_GENMASK(13, 11)
@@ -392,6 +391,18 @@
 #define XEHP_L3NODEARBCFG			XE_REG_MCR(0xb0b4)
 #define   XEHP_LNESPARE				REG_BIT(19)
 
+#define LSN_VC_REG2				XE_REG_MCR(0xb0c8)
+#define   LSN_LNI_WGT_MASK			REG_GENMASK(31, 28)
+#define   LSN_LNI_WGT(value)			REG_FIELD_PREP(LSN_LNI_WGT_MASK, value)
+#define   LSN_LNE_WGT_MASK			REG_GENMASK(27, 24)
+#define   LSN_LNE_WGT(value)			REG_FIELD_PREP(LSN_LNE_WGT_MASK, value)
+#define   LSN_DIM_X_WGT_MASK			REG_GENMASK(23, 20)
+#define   LSN_DIM_X_WGT(value)			REG_FIELD_PREP(LSN_DIM_X_WGT_MASK, value)
+#define   LSN_DIM_Y_WGT_MASK			REG_GENMASK(19, 16)
+#define   LSN_DIM_Y_WGT(value)			REG_FIELD_PREP(LSN_DIM_Y_WGT_MASK, value)
+#define   LSN_DIM_Z_WGT_MASK			REG_GENMASK(15, 12)
+#define   LSN_DIM_Z_WGT(value)			REG_FIELD_PREP(LSN_DIM_Z_WGT_MASK, value)
+
 #define L3SQCREG2				XE_REG_MCR(0xb104)
 #define   COMPMEMRD256BOVRFETCHEN		REG_BIT(20)
 
diff --git a/drivers/gpu/drm/xe/regs/xe_pcode_regs.h b/drivers/gpu/drm/xe/regs/xe_pcode_regs.h
index 8846eb9ce2a4..c7d5d782e3f9 100644
--- a/drivers/gpu/drm/xe/regs/xe_pcode_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_pcode_regs.h
@@ -21,6 +21,9 @@
 #define BMG_PACKAGE_POWER_SKU			XE_REG(0x138098)
 #define BMG_PACKAGE_POWER_SKU_UNIT		XE_REG(0x1380dc)
 #define BMG_PACKAGE_ENERGY_STATUS		XE_REG(0x138120)
+#define BMG_FAN_1_SPEED				XE_REG(0x138140)
+#define BMG_FAN_2_SPEED				XE_REG(0x138170)
+#define BMG_FAN_3_SPEED				XE_REG(0x1381a0)
 #define BMG_VRAM_TEMPERATURE			XE_REG(0x1382c0)
 #define BMG_PACKAGE_TEMPERATURE			XE_REG(0x138434)
 #define BMG_PACKAGE_RAPL_LIMIT			XE_REG(0x138440)
diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c
index 9fde67ca989f..378dcd0fb414 100644
--- a/drivers/gpu/drm/xe/tests/xe_bo.c
+++ b/drivers/gpu/drm/xe/tests/xe_bo.c
@@ -60,7 +60,7 @@ static int ccs_test_migrate(struct xe_tile *tile, struct xe_bo *bo,
 	}
 
 	/* Evict to system. CCS data should be copied. */
-	ret = xe_bo_evict(bo, true);
+	ret = xe_bo_evict(bo);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to evict bo.\n");
 		return ret;
@@ -252,7 +252,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 
 		for_each_gt(__gt, xe, id)
 			xe_gt_sanitize(__gt);
-		err = xe_bo_restore_kernel(xe);
+		err = xe_bo_restore_early(xe);
 		/*
 		 * Snapshotting the CTB and copying back a potentially old
 		 * version seems risky, depending on what might have been
@@ -273,7 +273,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 			goto cleanup_all;
 		}
 
-		err = xe_bo_restore_user(xe);
+		err = xe_bo_restore_late(xe);
 		if (err) {
 			KUNIT_FAIL(test, "restore user err=%pe\n", ERR_PTR(err));
 			goto cleanup_all;
diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
index cedd3e88a6fb..c53f67ce4b0a 100644
--- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
@@ -65,7 +65,7 @@ static void check_residency(struct kunit *test, struct xe_bo *exported,
 	 * the exporter and the importer should be the same bo.
 	 */
 	swap(exported->ttm.base.dma_buf, dmabuf);
-	ret = xe_bo_evict(exported, true);
+	ret = xe_bo_evict(exported);
 	swap(exported->ttm.base.dma_buf, dmabuf);
 	if (ret) {
 		if (ret != -EINTR && ret != -ERESTARTSYS)
diff --git a/drivers/gpu/drm/xe/tests/xe_migrate.c b/drivers/gpu/drm/xe/tests/xe_migrate.c
index d5fe0ea889ad..4a65e3103f77 100644
--- a/drivers/gpu/drm/xe/tests/xe_migrate.c
+++ b/drivers/gpu/drm/xe/tests/xe_migrate.c
@@ -202,8 +202,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test)
 
 	big = xe_bo_create_pin_map(xe, tile, m->q->vm, SZ_4M,
 				   ttm_bo_type_kernel,
-				   XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				   XE_BO_FLAG_PINNED);
+				   XE_BO_FLAG_VRAM_IF_DGFX(tile));
 	if (IS_ERR(big)) {
 		KUNIT_FAIL(test, "Failed to allocate bo: %li\n", PTR_ERR(big));
 		goto vunmap;
@@ -211,8 +210,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test)
 
 	pt = xe_bo_create_pin_map(xe, tile, m->q->vm, XE_PAGE_SIZE,
 				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				  XE_BO_FLAG_PINNED);
+				  XE_BO_FLAG_VRAM_IF_DGFX(tile));
 	if (IS_ERR(pt)) {
 		KUNIT_FAIL(test, "Failed to allocate fake pt: %li\n",
 			   PTR_ERR(pt));
@@ -222,8 +220,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test)
 	tiny = xe_bo_create_pin_map(xe, tile, m->q->vm,
 				    2 * SZ_4K,
 				    ttm_bo_type_kernel,
-				    XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				    XE_BO_FLAG_PINNED);
+				    XE_BO_FLAG_VRAM_IF_DGFX(tile));
 	if (IS_ERR(tiny)) {
 		KUNIT_FAIL(test, "Failed to allocate tiny fake pt: %li\n",
 			   PTR_ERR(tiny));
@@ -512,7 +509,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile,
 	dma_fence_put(fence);
 
 	kunit_info(test, "Evict vram buffer object\n");
-	ret = xe_bo_evict(vram_bo, true);
+	ret = xe_bo_evict(vram_bo);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to evict bo.\n");
 		return;
diff --git a/drivers/gpu/drm/xe/tests/xe_mocs.c b/drivers/gpu/drm/xe/tests/xe_mocs.c
index ef1e5256c56a..0e502feaca81 100644
--- a/drivers/gpu/drm/xe/tests/xe_mocs.c
+++ b/drivers/gpu/drm/xe/tests/xe_mocs.c
@@ -46,8 +46,11 @@ static void read_l3cc_table(struct xe_gt *gt,
 	unsigned int fw_ref, i;
 	u32 reg_val;
 
-	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
-	KUNIT_ASSERT_NE_MSG(test, fw_ref, 0, "Forcewake Failed.\n");
+	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL);
+	if (!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)) {
+		xe_force_wake_put(gt_to_fw(gt), fw_ref);
+		KUNIT_ASSERT_TRUE_MSG(test, true, "Forcewake Failed.\n");
+	}
 
 	for (i = 0; i < info->num_mocs_regs; i++) {
 		if (!(i & 1)) {
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 64f9c936eea0..d99d91fe8aa9 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -55,6 +55,8 @@ static struct ttm_placement sys_placement = {
 	.placement = &sys_placement_flags,
 };
 
+static struct ttm_placement purge_placement;
+
 static const struct ttm_place tt_placement_flags[] = {
 	{
 		.fpfn = 0,
@@ -189,11 +191,18 @@ static void try_add_system(struct xe_device *xe, struct xe_bo *bo,
 
 static bool force_contiguous(u32 bo_flags)
 {
+	if (bo_flags & XE_BO_FLAG_STOLEN)
+		return true; /* users expect this */
+	else if (bo_flags & XE_BO_FLAG_PINNED &&
+		 !(bo_flags & XE_BO_FLAG_PINNED_LATE_RESTORE))
+		return true; /* needs vmap */
+
 	/*
 	 * For eviction / restore on suspend / resume objects pinned in VRAM
 	 * must be contiguous, also only contiguous BOs support xe_bo_vmap.
 	 */
-	return bo_flags & (XE_BO_FLAG_PINNED | XE_BO_FLAG_GGTT);
+	return bo_flags & XE_BO_FLAG_NEEDS_CPU_ACCESS &&
+	       bo_flags & XE_BO_FLAG_PINNED;
 }
 
 static void add_vram(struct xe_device *xe, struct xe_bo *bo,
@@ -281,6 +290,8 @@ int xe_bo_placement_for_flags(struct xe_device *xe, struct xe_bo *bo,
 static void xe_evict_flags(struct ttm_buffer_object *tbo,
 			   struct ttm_placement *placement)
 {
+	struct xe_device *xe = container_of(tbo->bdev, typeof(*xe), ttm);
+	bool device_unplugged = drm_dev_is_unplugged(&xe->drm);
 	struct xe_bo *bo;
 
 	if (!xe_bo_is_xe_bo(tbo)) {
@@ -290,7 +301,7 @@ static void xe_evict_flags(struct ttm_buffer_object *tbo,
 			return;
 		}
 
-		*placement = sys_placement;
+		*placement = device_unplugged ? purge_placement : sys_placement;
 		return;
 	}
 
@@ -300,6 +311,11 @@ static void xe_evict_flags(struct ttm_buffer_object *tbo,
 		return;
 	}
 
+	if (device_unplugged && !tbo->base.dma_buf) {
+		*placement = purge_placement;
+		return;
+	}
+
 	/*
 	 * For xe, sg bos that are evicted to system just triggers a
 	 * rebind of the sg list upon subsequent validation to XE_PL_TT.
@@ -657,11 +673,20 @@ static int xe_bo_move_dmabuf(struct ttm_buffer_object *ttm_bo,
 	struct xe_ttm_tt *xe_tt = container_of(ttm_bo->ttm, struct xe_ttm_tt,
 					       ttm);
 	struct xe_device *xe = ttm_to_xe_device(ttm_bo->bdev);
+	bool device_unplugged = drm_dev_is_unplugged(&xe->drm);
 	struct sg_table *sg;
 
 	xe_assert(xe, attach);
 	xe_assert(xe, ttm_bo->ttm);
 
+	if (device_unplugged && new_res->mem_type == XE_PL_SYSTEM &&
+	    ttm_bo->sg) {
+		dma_resv_wait_timeout(ttm_bo->base.resv, DMA_RESV_USAGE_BOOKKEEP,
+				      false, MAX_SCHEDULE_TIMEOUT);
+		dma_buf_unmap_attachment(attach, ttm_bo->sg, DMA_BIDIRECTIONAL);
+		ttm_bo->sg = NULL;
+	}
+
 	if (new_res->mem_type == XE_PL_SYSTEM)
 		goto out;
 
@@ -898,79 +923,44 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,
 		xe_pm_runtime_get_noresume(xe);
 	}
 
-	if (xe_bo_is_pinned(bo) && !xe_bo_is_user(bo)) {
-		/*
-		 * Kernel memory that is pinned should only be moved on suspend
-		 * / resume, some of the pinned memory is required for the
-		 * device to resume / use the GPU to move other evicted memory
-		 * (user memory) around. This likely could be optimized a bit
-		 * further where we find the minimum set of pinned memory
-		 * required for resume but for simplity doing a memcpy for all
-		 * pinned memory.
-		 */
-		ret = xe_bo_vmap(bo);
-		if (!ret) {
-			ret = ttm_bo_move_memcpy(ttm_bo, ctx, new_mem);
-
-			/* Create a new VMAP once kernel BO back in VRAM */
-			if (!ret && resource_is_vram(new_mem)) {
-				struct xe_vram_region *vram = res_to_mem_region(new_mem);
-				void __iomem *new_addr = vram->mapping +
-					(new_mem->start << PAGE_SHIFT);
+	if (move_lacks_source) {
+		u32 flags = 0;
 
-				if (XE_WARN_ON(new_mem->start == XE_BO_INVALID_OFFSET)) {
-					ret = -EINVAL;
-					xe_pm_runtime_put(xe);
-					goto out;
-				}
+		if (mem_type_is_vram(new_mem->mem_type))
+			flags |= XE_MIGRATE_CLEAR_FLAG_FULL;
+		else if (handle_system_ccs)
+			flags |= XE_MIGRATE_CLEAR_FLAG_CCS_DATA;
 
-				xe_assert(xe, new_mem->start ==
-					  bo->placements->fpfn);
-
-				iosys_map_set_vaddr_iomem(&bo->vmap, new_addr);
-			}
-		}
+		fence = xe_migrate_clear(migrate, bo, new_mem, flags);
 	} else {
-		if (move_lacks_source) {
-			u32 flags = 0;
-
-			if (mem_type_is_vram(new_mem->mem_type))
-				flags |= XE_MIGRATE_CLEAR_FLAG_FULL;
-			else if (handle_system_ccs)
-				flags |= XE_MIGRATE_CLEAR_FLAG_CCS_DATA;
-
-			fence = xe_migrate_clear(migrate, bo, new_mem, flags);
-		}
-		else
-			fence = xe_migrate_copy(migrate, bo, bo, old_mem,
-						new_mem, handle_system_ccs);
-		if (IS_ERR(fence)) {
-			ret = PTR_ERR(fence);
-			xe_pm_runtime_put(xe);
-			goto out;
-		}
-		if (!move_lacks_source) {
-			ret = ttm_bo_move_accel_cleanup(ttm_bo, fence, evict,
-							true, new_mem);
-			if (ret) {
-				dma_fence_wait(fence, false);
-				ttm_bo_move_null(ttm_bo, new_mem);
-				ret = 0;
-			}
-		} else {
-			/*
-			 * ttm_bo_move_accel_cleanup() may blow up if
-			 * bo->resource == NULL, so just attach the
-			 * fence and set the new resource.
-			 */
-			dma_resv_add_fence(ttm_bo->base.resv, fence,
-					   DMA_RESV_USAGE_KERNEL);
+		fence = xe_migrate_copy(migrate, bo, bo, old_mem, new_mem,
+					handle_system_ccs);
+	}
+	if (IS_ERR(fence)) {
+		ret = PTR_ERR(fence);
+		xe_pm_runtime_put(xe);
+		goto out;
+	}
+	if (!move_lacks_source) {
+		ret = ttm_bo_move_accel_cleanup(ttm_bo, fence, evict, true,
+						new_mem);
+		if (ret) {
+			dma_fence_wait(fence, false);
 			ttm_bo_move_null(ttm_bo, new_mem);
+			ret = 0;
 		}
-
-		dma_fence_put(fence);
+	} else {
+		/*
+		 * ttm_bo_move_accel_cleanup() may blow up if
+		 * bo->resource == NULL, so just attach the
+		 * fence and set the new resource.
+		 */
+		dma_resv_add_fence(ttm_bo->base.resv, fence,
+				   DMA_RESV_USAGE_KERNEL);
+		ttm_bo_move_null(ttm_bo, new_mem);
 	}
 
+	dma_fence_put(fence);
 	xe_pm_runtime_put(xe);
 
 out:
@@ -1095,6 +1085,80 @@ out_unref:
 }
 
 /**
+ * xe_bo_notifier_prepare_pinned() - Prepare a pinned VRAM object to be backed
+ * up in system memory.
+ * @bo: The buffer object to prepare.
+ *
+ * On successful completion, the object backup pages are allocated. Expectation
+ * is that this is called from the PM notifier, prior to suspend/hibernation.
+ *
+ * Return: 0 on success. Negative error code on failure.
+ */
+int xe_bo_notifier_prepare_pinned(struct xe_bo *bo)
+{
+	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
+	struct xe_bo *backup;
+	int ret = 0;
+
+	xe_bo_lock(bo, false);
+
+	xe_assert(xe, !bo->backup_obj);
+
+	/*
+	 * Since this is called from the PM notifier we might have raced with
+	 * someone unpinning this after we dropped the pinned list lock and
+	 * grabbing the above bo lock.
+	 */
+	if (!xe_bo_is_pinned(bo))
+		goto out_unlock_bo;
+
+	if (!xe_bo_is_vram(bo))
+		goto out_unlock_bo;
+
+	if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
+		goto out_unlock_bo;
+
+	backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, bo->size,
+					DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
+					XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
+					XE_BO_FLAG_PINNED);
+	if (IS_ERR(backup)) {
+		ret = PTR_ERR(backup);
+		goto out_unlock_bo;
+	}
+
+	backup->parent_obj = xe_bo_get(bo); /* Released by bo_destroy */
+	ttm_bo_pin(&backup->ttm);
+	bo->backup_obj = backup;
+
+out_unlock_bo:
+	xe_bo_unlock(bo);
+	return ret;
+}
+
+/**
+ * xe_bo_notifier_unprepare_pinned() - Undo the previous prepare operation.
+ * @bo: The buffer object to undo the prepare for.
+ *
+ * Always returns 0. The backup object is removed, if still present. Expectation
+ * it that this called from the PM notifier when undoing the prepare step.
+ *
+ * Return: Always returns 0.
+ */
+int xe_bo_notifier_unprepare_pinned(struct xe_bo *bo)
+{
+	xe_bo_lock(bo, false);
+	if (bo->backup_obj) {
+		ttm_bo_unpin(&bo->backup_obj->ttm);
+		xe_bo_put(bo->backup_obj);
+		bo->backup_obj = NULL;
+	}
+	xe_bo_unlock(bo);
+
+	return 0;
+}
+
+/**
  * xe_bo_evict_pinned() - Evict a pinned VRAM object to system memory
  * @bo: The buffer object to move.
  *
@@ -1107,59 +1171,99 @@ out_unref:
  */
 int xe_bo_evict_pinned(struct xe_bo *bo)
 {
-	struct ttm_place place = {
-		.mem_type = XE_PL_TT,
-	};
-	struct ttm_placement placement = {
-		.placement = &place,
-		.num_placement = 1,
-	};
-	struct ttm_operation_ctx ctx = {
-		.interruptible = false,
-		.gfp_retry_mayfail = true,
-	};
-	struct ttm_resource *new_mem;
-	int ret;
+	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
+	struct xe_bo *backup = bo->backup_obj;
+	bool backup_created = false;
+	bool unmap = false;
+	int ret = 0;
 
-	xe_bo_assert_held(bo);
+	xe_bo_lock(bo, false);
 
-	if (WARN_ON(!bo->ttm.resource))
-		return -EINVAL;
+	if (WARN_ON(!bo->ttm.resource)) {
+		ret = -EINVAL;
+		goto out_unlock_bo;
+	}
 
-	if (WARN_ON(!xe_bo_is_pinned(bo)))
-		return -EINVAL;
+	if (WARN_ON(!xe_bo_is_pinned(bo))) {
+		ret = -EINVAL;
+		goto out_unlock_bo;
+	}
 
 	if (!xe_bo_is_vram(bo))
-		return 0;
+		goto out_unlock_bo;
+
+	if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
+		goto out_unlock_bo;
+
+	if (!backup) {
+		backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, bo->size,
+						DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
+						XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
+						XE_BO_FLAG_PINNED);
+		if (IS_ERR(backup)) {
+			ret = PTR_ERR(backup);
+			goto out_unlock_bo;
+		}
+		backup->parent_obj = xe_bo_get(bo); /* Released by bo_destroy */
+		backup_created = true;
+	}
 
-	ret = ttm_bo_mem_space(&bo->ttm, &placement, &new_mem, &ctx);
-	if (ret)
-		return ret;
+	if (xe_bo_is_user(bo) || (bo->flags & XE_BO_FLAG_PINNED_LATE_RESTORE)) {
+		struct xe_migrate *migrate;
+		struct dma_fence *fence;
+
+		if (bo->tile)
+			migrate = bo->tile->migrate;
+		else
+			migrate = mem_type_to_migrate(xe, bo->ttm.resource->mem_type);
+
+		ret = dma_resv_reserve_fences(bo->ttm.base.resv, 1);
+		if (ret)
+			goto out_backup;
+
+		ret = dma_resv_reserve_fences(backup->ttm.base.resv, 1);
+		if (ret)
+			goto out_backup;
 
-	if (!bo->ttm.ttm) {
-		bo->ttm.ttm = xe_ttm_tt_create(&bo->ttm, 0);
-		if (!bo->ttm.ttm) {
-			ret = -ENOMEM;
-			goto err_res_free;
+		fence = xe_migrate_copy(migrate, bo, backup, bo->ttm.resource,
+					backup->ttm.resource, false);
+		if (IS_ERR(fence)) {
+			ret = PTR_ERR(fence);
+			goto out_backup;
 		}
-	}
 
-	ret = ttm_bo_populate(&bo->ttm, &ctx);
-	if (ret)
-		goto err_res_free;
+		dma_resv_add_fence(bo->ttm.base.resv, fence,
+				   DMA_RESV_USAGE_KERNEL);
+		dma_resv_add_fence(backup->ttm.base.resv, fence,
+				   DMA_RESV_USAGE_KERNEL);
+		dma_fence_put(fence);
+	} else {
+		ret = xe_bo_vmap(backup);
+		if (ret)
+			goto out_backup;
 
-	ret = dma_resv_reserve_fences(bo->ttm.base.resv, 1);
-	if (ret)
-		goto err_res_free;
+		if (iosys_map_is_null(&bo->vmap)) {
+			ret = xe_bo_vmap(bo);
+			if (ret)
+				goto out_backup;
+			unmap = true;
+		}
 
-	ret = xe_bo_move(&bo->ttm, false, &ctx, new_mem, NULL);
-	if (ret)
-		goto err_res_free;
+		xe_map_memcpy_from(xe, backup->vmap.vaddr, &bo->vmap, 0,
+				   bo->size);
+	}
 
-	return 0;
+	if (!bo->backup_obj)
+		bo->backup_obj = backup;
 
-err_res_free:
-	ttm_resource_free(&bo->ttm, &new_mem);
+out_backup:
+	xe_bo_vunmap(backup);
+	if (ret && backup_created)
+		xe_bo_put(backup);
+out_unlock_bo:
+	if (unmap)
+		xe_bo_vunmap(bo);
+	xe_bo_unlock(bo);
 	return ret;
 }
 
@@ -1180,50 +1284,109 @@ int xe_bo_restore_pinned(struct xe_bo *bo)
 		.interruptible = false,
 		.gfp_retry_mayfail = false,
 	};
-	struct ttm_resource *new_mem;
-	struct ttm_place *place = &bo->placements[0];
+	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
+	struct xe_bo *backup = bo->backup_obj;
+	bool unmap = false;
 	int ret;
 
-	xe_bo_assert_held(bo);
+	if (!backup)
+		return 0;
 
-	if (WARN_ON(!bo->ttm.resource))
-		return -EINVAL;
+	xe_bo_lock(bo, false);
 
-	if (WARN_ON(!xe_bo_is_pinned(bo)))
-		return -EINVAL;
+	if (!xe_bo_is_pinned(backup)) {
+		ret = ttm_bo_validate(&backup->ttm, &backup->placement, &ctx);
+		if (ret)
+			goto out_unlock_bo;
+	}
 
-	if (WARN_ON(xe_bo_is_vram(bo)))
-		return -EINVAL;
+	if (xe_bo_is_user(bo) || (bo->flags & XE_BO_FLAG_PINNED_LATE_RESTORE)) {
+		struct xe_migrate *migrate;
+		struct dma_fence *fence;
 
-	if (WARN_ON(!bo->ttm.ttm && !xe_bo_is_stolen(bo)))
-		return -EINVAL;
+		if (bo->tile)
+			migrate = bo->tile->migrate;
+		else
+			migrate = mem_type_to_migrate(xe, bo->ttm.resource->mem_type);
 
-	if (!mem_type_is_vram(place->mem_type))
-		return 0;
+		ret = dma_resv_reserve_fences(bo->ttm.base.resv, 1);
+		if (ret)
+			goto out_unlock_bo;
 
-	ret = ttm_bo_mem_space(&bo->ttm, &bo->placement, &new_mem, &ctx);
-	if (ret)
-		return ret;
+		ret = dma_resv_reserve_fences(backup->ttm.base.resv, 1);
+		if (ret)
+			goto out_unlock_bo;
 
-	ret = ttm_bo_populate(&bo->ttm, &ctx);
-	if (ret)
-		goto err_res_free;
+		fence = xe_migrate_copy(migrate, backup, bo,
+					backup->ttm.resource, bo->ttm.resource,
+					false);
+		if (IS_ERR(fence)) {
+			ret = PTR_ERR(fence);
+			goto out_unlock_bo;
+		}
 
-	ret = dma_resv_reserve_fences(bo->ttm.base.resv, 1);
-	if (ret)
-		goto err_res_free;
+		dma_resv_add_fence(bo->ttm.base.resv, fence,
+				   DMA_RESV_USAGE_KERNEL);
+		dma_resv_add_fence(backup->ttm.base.resv, fence,
+				   DMA_RESV_USAGE_KERNEL);
+		dma_fence_put(fence);
+	} else {
+		ret = xe_bo_vmap(backup);
+		if (ret)
+			goto out_unlock_bo;
 
-	ret = xe_bo_move(&bo->ttm, false, &ctx, new_mem, NULL);
-	if (ret)
-		goto err_res_free;
+		if (iosys_map_is_null(&bo->vmap)) {
+			ret = xe_bo_vmap(bo);
+			if (ret)
+				goto out_backup;
+			unmap = true;
+		}
 
-	return 0;
+		xe_map_memcpy_to(xe, &bo->vmap, 0, backup->vmap.vaddr,
+				 bo->size);
+	}
+
+	bo->backup_obj = NULL;
 
-err_res_free:
-	ttm_resource_free(&bo->ttm, &new_mem);
+out_backup:
+	xe_bo_vunmap(backup);
+	if (!bo->backup_obj) {
+		if (xe_bo_is_pinned(backup))
+			ttm_bo_unpin(&backup->ttm);
+		xe_bo_put(backup);
+	}
+out_unlock_bo:
+	if (unmap)
+		xe_bo_vunmap(bo);
+	xe_bo_unlock(bo);
 	return ret;
 }
 
+int xe_bo_dma_unmap_pinned(struct xe_bo *bo)
+{
+	struct ttm_buffer_object *ttm_bo = &bo->ttm;
+	struct ttm_tt *tt = ttm_bo->ttm;
+
+	if (tt) {
+		struct xe_ttm_tt *xe_tt = container_of(tt, typeof(*xe_tt), ttm);
+
+		if (ttm_bo->type == ttm_bo_type_sg && ttm_bo->sg) {
+			dma_buf_unmap_attachment(ttm_bo->base.import_attach,
+						 ttm_bo->sg,
+						 DMA_BIDIRECTIONAL);
+			ttm_bo->sg = NULL;
+			xe_tt->sg = NULL;
+		} else if (xe_tt->sg) {
+			dma_unmap_sgtable(xe_tt->xe->drm.dev, xe_tt->sg,
+					  DMA_BIDIRECTIONAL, 0);
+			sg_free_table(xe_tt->sg);
+			xe_tt->sg = NULL;
+		}
+	}
+
+	return 0;
+}
+
 static unsigned long xe_ttm_io_mem_pfn(struct ttm_buffer_object *ttm_bo,
 				       unsigned long page_offset)
 {
@@ -1371,6 +1534,7 @@ static int xe_ttm_access_memory(struct ttm_buffer_object *ttm_bo,
 	struct xe_res_cursor cursor;
 	struct xe_vram_region *vram;
 	int bytes_left = len;
+	int err = 0;
 
 	xe_bo_assert_held(bo);
 	xe_device_assert_mem_access(xe);
@@ -1378,9 +1542,14 @@ static int xe_ttm_access_memory(struct ttm_buffer_object *ttm_bo,
 	if (!mem_type_is_vram(ttm_bo->resource->mem_type))
 		return -EIO;
 
-	/* FIXME: Use GPU for non-visible VRAM */
-	if (!xe_ttm_resource_visible(ttm_bo->resource))
-		return -EIO;
+	if (!xe_ttm_resource_visible(ttm_bo->resource) || len >= SZ_16K) {
+		struct xe_migrate *migrate =
+			mem_type_to_migrate(xe, ttm_bo->resource->mem_type);
+
+		err = xe_migrate_access_memory(migrate, bo, offset, buf, len,
+					       write);
+		goto out;
+	}
 
 	vram = res_to_mem_region(ttm_bo->resource);
 	xe_res_first(ttm_bo->resource, offset & PAGE_MASK,
@@ -1404,7 +1573,8 @@ static int xe_ttm_access_memory(struct ttm_buffer_object *ttm_bo,
 			xe_res_next(&cursor, PAGE_SIZE);
 	} while (bytes_left);
 
-	return len;
+out:
+	return err ?: len;
 }
 
 const struct ttm_device_funcs xe_ttm_funcs = {
@@ -1448,6 +1618,9 @@ static void xe_ttm_bo_destroy(struct ttm_buffer_object *ttm_bo)
 	if (bo->vm && xe_bo_is_user(bo))
 		xe_vm_put(bo->vm);
 
+	if (bo->parent_obj)
+		xe_bo_put(bo->parent_obj);
+
 	mutex_lock(&xe->mem_access.vram_userfault.lock);
 	if (!list_empty(&bo->vram_userfault_link))
 		list_del(&bo->vram_userfault_link);
@@ -1947,7 +2120,7 @@ struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
 		flags |= XE_BO_FLAG_GGTT;
 
 	bo = xe_bo_create_locked_range(xe, tile, vm, size, start, end, type,
-				       flags | XE_BO_FLAG_NEEDS_CPU_ACCESS,
+				       flags | XE_BO_FLAG_NEEDS_CPU_ACCESS | XE_BO_FLAG_PINNED,
 				       alignment);
 	if (IS_ERR(bo))
 		return bo;
@@ -2049,7 +2222,8 @@ int xe_managed_bo_reinit_in_vram(struct xe_device *xe, struct xe_tile *tile, str
 	struct xe_bo *bo;
 	u32 dst_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile) | XE_BO_FLAG_GGTT;
 
-	dst_flags |= (*src)->flags & XE_BO_FLAG_GGTT_INVALIDATE;
+	dst_flags |= (*src)->flags & (XE_BO_FLAG_GGTT_INVALIDATE |
+				      XE_BO_FLAG_PINNED_NORESTORE);
 
 	xe_assert(xe, IS_DGFX(xe));
 	xe_assert(xe, !(*src)->vmap.is_iomem);
@@ -2073,10 +2247,16 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res)
 {
 	struct xe_device *xe = ttm_to_xe_device(res->bo->bdev);
 
-	if (res->mem_type == XE_PL_STOLEN)
+	switch (res->mem_type) {
+	case XE_PL_STOLEN:
 		return xe_ttm_stolen_gpu_offset(xe);
-
-	return res_to_mem_region(res)->dpa_base;
+	case XE_PL_TT:
+	case XE_PL_SYSTEM:
+		return 0;
+	default:
+		return res_to_mem_region(res)->dpa_base;
+	}
+	return 0;
 }
 
 /**
@@ -2102,12 +2282,9 @@ int xe_bo_pin_external(struct xe_bo *bo)
 		if (err)
 			return err;
 
-		if (xe_bo_is_vram(bo)) {
-			spin_lock(&xe->pinned.lock);
-			list_add_tail(&bo->pinned_link,
-				      &xe->pinned.external_vram);
-			spin_unlock(&xe->pinned.lock);
-		}
+		spin_lock(&xe->pinned.lock);
+		list_add_tail(&bo->pinned_link, &xe->pinned.late.external);
+		spin_unlock(&xe->pinned.lock);
 	}
 
 	ttm_bo_pin(&bo->ttm);
@@ -2149,25 +2326,12 @@ int xe_bo_pin(struct xe_bo *bo)
 	if (err)
 		return err;
 
-	/*
-	 * For pinned objects in on DGFX, which are also in vram, we expect
-	 * these to be in contiguous VRAM memory. Required eviction / restore
-	 * during suspend / resume (force restore to same physical address).
-	 */
-	if (IS_DGFX(xe) && !(IS_ENABLED(CONFIG_DRM_XE_DEBUG) &&
-	    bo->flags & XE_BO_FLAG_INTERNAL_TEST)) {
-		if (mem_type_is_vram(place->mem_type)) {
-			xe_assert(xe, place->flags & TTM_PL_FLAG_CONTIGUOUS);
-
-			place->fpfn = (xe_bo_addr(bo, 0, PAGE_SIZE) -
-				       vram_region_gpu_offset(bo->ttm.resource)) >> PAGE_SHIFT;
-			place->lpfn = place->fpfn + (bo->size >> PAGE_SHIFT);
-		}
-	}
-
 	if (mem_type_is_vram(place->mem_type) || bo->flags & XE_BO_FLAG_GGTT) {
 		spin_lock(&xe->pinned.lock);
-		list_add_tail(&bo->pinned_link, &xe->pinned.kernel_bo_present);
+		if (bo->flags & XE_BO_FLAG_PINNED_LATE_RESTORE)
+			list_add_tail(&bo->pinned_link, &xe->pinned.late.kernel_bo_present);
+		else
+			list_add_tail(&bo->pinned_link, &xe->pinned.early.kernel_bo_present);
 		spin_unlock(&xe->pinned.lock);
 	}
 
@@ -2231,6 +2395,13 @@ void xe_bo_unpin(struct xe_bo *bo)
 		xe_assert(xe, !list_empty(&bo->pinned_link));
 		list_del_init(&bo->pinned_link);
 		spin_unlock(&xe->pinned.lock);
+
+		if (bo->backup_obj) {
+			if (xe_bo_is_pinned(bo->backup_obj))
+				ttm_bo_unpin(&bo->backup_obj->ttm);
+			xe_bo_put(bo->backup_obj);
+			bo->backup_obj = NULL;
+		}
 	}
 	ttm_bo_unpin(&bo->ttm);
 	if (bo->ttm.ttm && ttm_tt_is_populated(bo->ttm.ttm))
@@ -2398,7 +2569,7 @@ static int gem_create_user_ext_set_property(struct xe_device *xe,
 	int err;
 	u32 idx;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
@@ -2435,7 +2606,7 @@ static int gem_create_user_extensions(struct xe_device *xe, struct xe_bo *bo,
 	if (XE_IOCTL_DBG(xe, ext_number >= MAX_USER_EXTENSIONS))
 		return -E2BIG;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
@@ -2759,19 +2930,17 @@ int xe_bo_migrate(struct xe_bo *bo, u32 mem_type)
 /**
  * xe_bo_evict - Evict an object to evict placement
  * @bo: The buffer object to migrate.
- * @force_alloc: Set force_alloc in ttm_operation_ctx
  *
  * On successful completion, the object memory will be moved to evict
  * placement. This function blocks until the object has been fully moved.
  *
  * Return: 0 on success. Negative error code on failure.
  */
-int xe_bo_evict(struct xe_bo *bo, bool force_alloc)
+int xe_bo_evict(struct xe_bo *bo)
 {
 	struct ttm_operation_ctx ctx = {
 		.interruptible = false,
 		.no_wait_gpu = false,
-		.force_alloc = force_alloc,
 		.gfp_retry_mayfail = true,
 	};
 	struct ttm_placement placement;
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index ec3e4446d027..02ada1fb8a23 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -39,20 +39,23 @@
 #define XE_BO_FLAG_NEEDS_64K		BIT(15)
 #define XE_BO_FLAG_NEEDS_2M		BIT(16)
 #define XE_BO_FLAG_GGTT_INVALIDATE	BIT(17)
-#define XE_BO_FLAG_GGTT0                BIT(18)
-#define XE_BO_FLAG_GGTT1                BIT(19)
-#define XE_BO_FLAG_GGTT2                BIT(20)
-#define XE_BO_FLAG_GGTT3                BIT(21)
-#define XE_BO_FLAG_GGTT_ALL             (XE_BO_FLAG_GGTT0 | \
-					 XE_BO_FLAG_GGTT1 | \
-					 XE_BO_FLAG_GGTT2 | \
-					 XE_BO_FLAG_GGTT3)
-#define XE_BO_FLAG_CPU_ADDR_MIRROR	BIT(22)
+#define XE_BO_FLAG_PINNED_NORESTORE	BIT(18)
+#define XE_BO_FLAG_PINNED_LATE_RESTORE	BIT(19)
+#define XE_BO_FLAG_GGTT0		BIT(20)
+#define XE_BO_FLAG_GGTT1		BIT(21)
+#define XE_BO_FLAG_GGTT2		BIT(22)
+#define XE_BO_FLAG_GGTT3		BIT(23)
+#define XE_BO_FLAG_CPU_ADDR_MIRROR	BIT(24)
 
 /* this one is trigger internally only */
 #define XE_BO_FLAG_INTERNAL_TEST	BIT(30)
 #define XE_BO_FLAG_INTERNAL_64K		BIT(31)
 
+#define XE_BO_FLAG_GGTT_ALL		(XE_BO_FLAG_GGTT0 | \
+					 XE_BO_FLAG_GGTT1 | \
+					 XE_BO_FLAG_GGTT2 | \
+					 XE_BO_FLAG_GGTT3)
+
 #define XE_BO_FLAG_GGTTx(tile) \
 	(XE_BO_FLAG_GGTT0 << (tile)->id)
 
@@ -271,11 +274,15 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res);
 bool xe_bo_can_migrate(struct xe_bo *bo, u32 mem_type);
 
 int xe_bo_migrate(struct xe_bo *bo, u32 mem_type);
-int xe_bo_evict(struct xe_bo *bo, bool force_alloc);
+int xe_bo_evict(struct xe_bo *bo);
 
 int xe_bo_evict_pinned(struct xe_bo *bo);
+int xe_bo_notifier_prepare_pinned(struct xe_bo *bo);
+int xe_bo_notifier_unprepare_pinned(struct xe_bo *bo);
 int xe_bo_restore_pinned(struct xe_bo *bo);
 
+int xe_bo_dma_unmap_pinned(struct xe_bo *bo);
+
 extern const struct ttm_device_funcs xe_ttm_funcs;
 extern const char *const xe_mem_type_to_name[];
 
diff --git a/drivers/gpu/drm/xe/xe_bo_evict.c b/drivers/gpu/drm/xe/xe_bo_evict.c
index 6a40eedd9db1..ed3746d32b27 100644
--- a/drivers/gpu/drm/xe/xe_bo_evict.c
+++ b/drivers/gpu/drm/xe/xe_bo_evict.c
@@ -10,28 +10,103 @@
 #include "xe_ggtt.h"
 #include "xe_tile.h"
 
+typedef int (*xe_pinned_fn)(struct xe_bo *bo);
+
+static int xe_bo_apply_to_pinned(struct xe_device *xe,
+				 struct list_head *pinned_list,
+				 struct list_head *new_list,
+				 const xe_pinned_fn pinned_fn)
+{
+	LIST_HEAD(still_in_list);
+	struct xe_bo *bo;
+	int ret = 0;
+
+	spin_lock(&xe->pinned.lock);
+	while (!ret) {
+		bo = list_first_entry_or_null(pinned_list, typeof(*bo),
+					      pinned_link);
+		if (!bo)
+			break;
+		xe_bo_get(bo);
+		list_move_tail(&bo->pinned_link, &still_in_list);
+		spin_unlock(&xe->pinned.lock);
+
+		ret = pinned_fn(bo);
+		if (ret && pinned_list != new_list) {
+			spin_lock(&xe->pinned.lock);
+			/*
+			 * We might no longer be pinned, since PM notifier can
+			 * call this. If the pinned link is now empty, keep it
+			 * that way.
+			 */
+			if (!list_empty(&bo->pinned_link))
+				list_move(&bo->pinned_link, pinned_list);
+			spin_unlock(&xe->pinned.lock);
+		}
+		xe_bo_put(bo);
+		spin_lock(&xe->pinned.lock);
+	}
+	list_splice_tail(&still_in_list, new_list);
+	spin_unlock(&xe->pinned.lock);
+
+	return ret;
+}
+
 /**
- * xe_bo_evict_all - evict all BOs from VRAM
+ * xe_bo_notifier_prepare_all_pinned() - Pre-allocate the backing pages for all
+ * pinned VRAM objects which need to be saved.
+ * @xe: xe device
+ *
+ * Should be called from PM notifier when preparing for s3/s4.
  *
+ * Return: 0 on success, negative error code on error.
+ */
+int xe_bo_notifier_prepare_all_pinned(struct xe_device *xe)
+{
+	int ret;
+
+	ret = xe_bo_apply_to_pinned(xe, &xe->pinned.early.kernel_bo_present,
+				    &xe->pinned.early.kernel_bo_present,
+				    xe_bo_notifier_prepare_pinned);
+	if (!ret)
+		ret = xe_bo_apply_to_pinned(xe, &xe->pinned.late.kernel_bo_present,
+					    &xe->pinned.late.kernel_bo_present,
+					    xe_bo_notifier_prepare_pinned);
+
+	return ret;
+}
+
+/**
+ * xe_bo_notifier_unprepare_all_pinned() - Remove the backing pages for all
+ * pinned VRAM objects which have been restored.
  * @xe: xe device
  *
- * Evict non-pinned user BOs first (via GPU), evict pinned external BOs next
- * (via GPU), wait for evictions, and finally evict pinned kernel BOs via CPU.
- * All eviction magic done via TTM calls.
+ * Should be called from PM notifier after exiting s3/s4 (either on success or
+ * failure).
+ */
+void xe_bo_notifier_unprepare_all_pinned(struct xe_device *xe)
+{
+	(void)xe_bo_apply_to_pinned(xe, &xe->pinned.early.kernel_bo_present,
+				    &xe->pinned.early.kernel_bo_present,
+				    xe_bo_notifier_unprepare_pinned);
+
+	(void)xe_bo_apply_to_pinned(xe, &xe->pinned.late.kernel_bo_present,
+				    &xe->pinned.late.kernel_bo_present,
+				    xe_bo_notifier_unprepare_pinned);
+}
+
+/**
+ * xe_bo_evict_all_user - evict all non-pinned user BOs from VRAM
+ * @xe: xe device
  *
- * Evict == move VRAM BOs to temporary (typically system) memory.
+ * Evict non-pinned user BOs (via GPU).
  *
- * This function should be called before the device goes into a suspend state
- * where the VRAM loses power.
+ * Evict == move VRAM BOs to temporary (typically system) memory.
  */
-int xe_bo_evict_all(struct xe_device *xe)
+int xe_bo_evict_all_user(struct xe_device *xe)
 {
 	struct ttm_device *bdev = &xe->ttm;
-	struct xe_bo *bo;
-	struct xe_tile *tile;
-	struct list_head still_in_list;
 	u32 mem_type;
-	u8 id;
 	int ret;
 
 	/* User memory */
@@ -57,34 +132,38 @@ int xe_bo_evict_all(struct xe_device *xe)
 		}
 	}
 
-	/* Pinned user memory in VRAM */
-	INIT_LIST_HEAD(&still_in_list);
-	spin_lock(&xe->pinned.lock);
-	for (;;) {
-		bo = list_first_entry_or_null(&xe->pinned.external_vram,
-					      typeof(*bo), pinned_link);
-		if (!bo)
-			break;
-		xe_bo_get(bo);
-		list_move_tail(&bo->pinned_link, &still_in_list);
-		spin_unlock(&xe->pinned.lock);
+	return 0;
+}
 
-		xe_bo_lock(bo, false);
-		ret = xe_bo_evict_pinned(bo);
-		xe_bo_unlock(bo);
-		xe_bo_put(bo);
-		if (ret) {
-			spin_lock(&xe->pinned.lock);
-			list_splice_tail(&still_in_list,
-					 &xe->pinned.external_vram);
-			spin_unlock(&xe->pinned.lock);
-			return ret;
-		}
+/**
+ * xe_bo_evict_all - evict all BOs from VRAM
+ * @xe: xe device
+ *
+ * Evict non-pinned user BOs first (via GPU), evict pinned external BOs next
+ * (via GPU), wait for evictions, and finally evict pinned kernel BOs via CPU.
+ * All eviction magic done via TTM calls.
+ *
+ * Evict == move VRAM BOs to temporary (typically system) memory.
+ *
+ * This function should be called before the device goes into a suspend state
+ * where the VRAM loses power.
+ */
+int xe_bo_evict_all(struct xe_device *xe)
+{
+	struct xe_tile *tile;
+	u8 id;
+	int ret;
 
-		spin_lock(&xe->pinned.lock);
-	}
-	list_splice_tail(&still_in_list, &xe->pinned.external_vram);
-	spin_unlock(&xe->pinned.lock);
+	ret = xe_bo_evict_all_user(xe);
+	if (ret)
+		return ret;
+
+	ret = xe_bo_apply_to_pinned(xe, &xe->pinned.late.kernel_bo_present,
+				    &xe->pinned.late.evicted, xe_bo_evict_pinned);
+
+	if (!ret)
+		ret = xe_bo_apply_to_pinned(xe, &xe->pinned.late.kernel_bo_present,
+					    &xe->pinned.late.evicted, xe_bo_evict_pinned);
 
 	/*
 	 * Wait for all user BO to be evicted as those evictions depend on the
@@ -93,32 +172,49 @@ int xe_bo_evict_all(struct xe_device *xe)
 	for_each_tile(tile, xe, id)
 		xe_tile_migrate_wait(tile);
 
-	spin_lock(&xe->pinned.lock);
-	for (;;) {
-		bo = list_first_entry_or_null(&xe->pinned.kernel_bo_present,
-					      typeof(*bo), pinned_link);
-		if (!bo)
-			break;
-		xe_bo_get(bo);
-		list_move_tail(&bo->pinned_link, &xe->pinned.evicted);
-		spin_unlock(&xe->pinned.lock);
+	if (ret)
+		return ret;
 
-		xe_bo_lock(bo, false);
-		ret = xe_bo_evict_pinned(bo);
-		xe_bo_unlock(bo);
-		xe_bo_put(bo);
-		if (ret)
-			return ret;
+	return xe_bo_apply_to_pinned(xe, &xe->pinned.early.kernel_bo_present,
+				     &xe->pinned.early.evicted,
+				     xe_bo_evict_pinned);
+}
 
-		spin_lock(&xe->pinned.lock);
+static int xe_bo_restore_and_map_ggtt(struct xe_bo *bo)
+{
+	struct xe_device *xe = xe_bo_device(bo);
+	int ret;
+
+	ret = xe_bo_restore_pinned(bo);
+	if (ret)
+		return ret;
+
+	if (bo->flags & XE_BO_FLAG_GGTT) {
+		struct xe_tile *tile;
+		u8 id;
+
+		for_each_tile(tile, xe_bo_device(bo), id) {
+			if (tile != bo->tile && !(bo->flags & XE_BO_FLAG_GGTTx(tile)))
+				continue;
+
+			mutex_lock(&tile->mem.ggtt->lock);
+			xe_ggtt_map_bo(tile->mem.ggtt, bo);
+			mutex_unlock(&tile->mem.ggtt->lock);
+		}
 	}
-	spin_unlock(&xe->pinned.lock);
+
+	/*
+	 * We expect validate to trigger a move VRAM and our move code
+	 * should setup the iosys map.
+	 */
+	xe_assert(xe, !(bo->flags & XE_BO_FLAG_PINNED_LATE_RESTORE) ||
+		  !iosys_map_is_null(&bo->vmap));
 
 	return 0;
 }
 
 /**
- * xe_bo_restore_kernel - restore kernel BOs to VRAM
+ * xe_bo_restore_early - restore early phase kernel BOs to VRAM
  *
  * @xe: xe device
  *
@@ -128,111 +224,130 @@ int xe_bo_evict_all(struct xe_device *xe)
  * This function should be called early, before trying to init the GT, on device
  * resume.
  */
-int xe_bo_restore_kernel(struct xe_device *xe)
+int xe_bo_restore_early(struct xe_device *xe)
 {
-	struct xe_bo *bo;
-	int ret;
+	return xe_bo_apply_to_pinned(xe, &xe->pinned.early.evicted,
+				     &xe->pinned.early.kernel_bo_present,
+				     xe_bo_restore_and_map_ggtt);
+}
 
-	spin_lock(&xe->pinned.lock);
-	for (;;) {
-		bo = list_first_entry_or_null(&xe->pinned.evicted,
-					      typeof(*bo), pinned_link);
-		if (!bo)
-			break;
-		xe_bo_get(bo);
-		list_move_tail(&bo->pinned_link, &xe->pinned.kernel_bo_present);
-		spin_unlock(&xe->pinned.lock);
+/**
+ * xe_bo_restore_late - restore pinned late phase BOs
+ *
+ * @xe: xe device
+ *
+ * Move pinned user and kernel BOs which can use blitter from temporary
+ * (typically system) memory to VRAM. All moves done via TTM calls.
+ *
+ * This function should be called late, after GT init, on device resume.
+ */
+int xe_bo_restore_late(struct xe_device *xe)
+{
+	struct xe_tile *tile;
+	int ret, id;
 
-		xe_bo_lock(bo, false);
-		ret = xe_bo_restore_pinned(bo);
-		xe_bo_unlock(bo);
-		if (ret) {
-			xe_bo_put(bo);
-			return ret;
-		}
+	ret = xe_bo_apply_to_pinned(xe, &xe->pinned.late.evicted,
+				    &xe->pinned.late.kernel_bo_present,
+				    xe_bo_restore_and_map_ggtt);
 
-		if (bo->flags & XE_BO_FLAG_GGTT) {
-			struct xe_tile *tile;
-			u8 id;
+	for_each_tile(tile, xe, id)
+		xe_tile_migrate_wait(tile);
 
-			for_each_tile(tile, xe, id) {
-				if (tile != bo->tile && !(bo->flags & XE_BO_FLAG_GGTTx(tile)))
-					continue;
+	if (ret)
+		return ret;
 
-				mutex_lock(&tile->mem.ggtt->lock);
-				xe_ggtt_map_bo(tile->mem.ggtt, bo);
-				mutex_unlock(&tile->mem.ggtt->lock);
-			}
-		}
+	if (!IS_DGFX(xe))
+		return 0;
 
-		/*
-		 * We expect validate to trigger a move VRAM and our move code
-		 * should setup the iosys map.
-		 */
-		xe_assert(xe, !iosys_map_is_null(&bo->vmap));
+	/* Pinned user memory in VRAM should be validated on resume */
+	ret = xe_bo_apply_to_pinned(xe, &xe->pinned.late.external,
+				    &xe->pinned.late.external,
+				    xe_bo_restore_pinned);
 
-		xe_bo_put(bo);
+	/* Wait for restore to complete */
+	for_each_tile(tile, xe, id)
+		xe_tile_migrate_wait(tile);
 
-		spin_lock(&xe->pinned.lock);
-	}
-	spin_unlock(&xe->pinned.lock);
+	return ret;
+}
 
-	return 0;
+static void xe_bo_pci_dev_remove_pinned(struct xe_device *xe)
+{
+	struct xe_tile *tile;
+	unsigned int id;
+
+	(void)xe_bo_apply_to_pinned(xe, &xe->pinned.late.external,
+				    &xe->pinned.late.external,
+				    xe_bo_dma_unmap_pinned);
+	for_each_tile(tile, xe, id)
+		xe_tile_migrate_wait(tile);
 }
 
 /**
- * xe_bo_restore_user - restore pinned user BOs to VRAM
- *
- * @xe: xe device
+ * xe_bo_pci_dev_remove_all() - Handle bos when the pci_device is about to be removed
+ * @xe: The xe device.
  *
- * Move pinned user BOs from temporary (typically system) memory to VRAM via
- * CPU. All moves done via TTM calls.
+ * On pci_device removal we need to drop all dma mappings and move
+ * the data of exported bos out to system. This includes SVM bos and
+ * exported dma-buf bos. This is done by evicting all bos, but
+ * the evict placement in xe_evict_flags() is chosen such that all
+ * bos except those mentioned are purged, and thus their memory
+ * is released.
  *
- * This function should be called late, after GT init, on device resume.
+ * For pinned bos, we're unmapping dma.
  */
-int xe_bo_restore_user(struct xe_device *xe)
+void xe_bo_pci_dev_remove_all(struct xe_device *xe)
 {
-	struct xe_bo *bo;
-	struct xe_tile *tile;
-	struct list_head still_in_list;
-	u8 id;
-	int ret;
+	unsigned int mem_type;
 
-	if (!IS_DGFX(xe))
-		return 0;
+	/*
+	 * Move pagemap bos and exported dma-buf to system, and
+	 * purge everything else.
+	 */
+	for (mem_type = XE_PL_VRAM1; mem_type >= XE_PL_TT; --mem_type) {
+		struct ttm_resource_manager *man =
+			ttm_manager_type(&xe->ttm, mem_type);
 
-	/* Pinned user memory in VRAM should be validated on resume */
-	INIT_LIST_HEAD(&still_in_list);
-	spin_lock(&xe->pinned.lock);
-	for (;;) {
-		bo = list_first_entry_or_null(&xe->pinned.external_vram,
-					      typeof(*bo), pinned_link);
-		if (!bo)
-			break;
-		list_move_tail(&bo->pinned_link, &still_in_list);
-		xe_bo_get(bo);
-		spin_unlock(&xe->pinned.lock);
+		if (man) {
+			int ret = ttm_resource_manager_evict_all(&xe->ttm, man);
 
-		xe_bo_lock(bo, false);
-		ret = xe_bo_restore_pinned(bo);
-		xe_bo_unlock(bo);
-		xe_bo_put(bo);
-		if (ret) {
-			spin_lock(&xe->pinned.lock);
-			list_splice_tail(&still_in_list,
-					 &xe->pinned.external_vram);
-			spin_unlock(&xe->pinned.lock);
-			return ret;
+			drm_WARN_ON(&xe->drm, ret);
 		}
-
-		spin_lock(&xe->pinned.lock);
 	}
-	list_splice_tail(&still_in_list, &xe->pinned.external_vram);
-	spin_unlock(&xe->pinned.lock);
 
-	/* Wait for restore to complete */
-	for_each_tile(tile, xe, id)
-		xe_tile_migrate_wait(tile);
+	xe_bo_pci_dev_remove_pinned(xe);
+}
 
-	return 0;
+static void xe_bo_pinned_fini(void *arg)
+{
+	struct xe_device *xe = arg;
+
+	(void)xe_bo_apply_to_pinned(xe, &xe->pinned.late.kernel_bo_present,
+				    &xe->pinned.late.kernel_bo_present,
+				    xe_bo_dma_unmap_pinned);
+	(void)xe_bo_apply_to_pinned(xe, &xe->pinned.early.kernel_bo_present,
+				    &xe->pinned.early.kernel_bo_present,
+				    xe_bo_dma_unmap_pinned);
+}
+
+/**
+ * xe_bo_pinned_init() - Initialize pinned bo tracking
+ * @xe: The xe device.
+ *
+ * Initializes the lists and locks required for pinned bo
+ * tracking and registers a callback to dma-unmap
+ * any remaining pinned bos on pci device removal.
+ *
+ * Return: %0 on success, negative error code on error.
+ */
+int xe_bo_pinned_init(struct xe_device *xe)
+{
+	spin_lock_init(&xe->pinned.lock);
+	INIT_LIST_HEAD(&xe->pinned.early.kernel_bo_present);
+	INIT_LIST_HEAD(&xe->pinned.early.evicted);
+	INIT_LIST_HEAD(&xe->pinned.late.kernel_bo_present);
+	INIT_LIST_HEAD(&xe->pinned.late.evicted);
+	INIT_LIST_HEAD(&xe->pinned.late.external);
+
+	return devm_add_action_or_reset(xe->drm.dev, xe_bo_pinned_fini, xe);
 }
diff --git a/drivers/gpu/drm/xe/xe_bo_evict.h b/drivers/gpu/drm/xe/xe_bo_evict.h
index 746894798852..e8385cb7f5e9 100644
--- a/drivers/gpu/drm/xe/xe_bo_evict.h
+++ b/drivers/gpu/drm/xe/xe_bo_evict.h
@@ -9,7 +9,13 @@
 struct xe_device;
 
 int xe_bo_evict_all(struct xe_device *xe);
-int xe_bo_restore_kernel(struct xe_device *xe);
-int xe_bo_restore_user(struct xe_device *xe);
+int xe_bo_evict_all_user(struct xe_device *xe);
+int xe_bo_notifier_prepare_all_pinned(struct xe_device *xe);
+void xe_bo_notifier_unprepare_all_pinned(struct xe_device *xe);
+int xe_bo_restore_early(struct xe_device *xe);
+int xe_bo_restore_late(struct xe_device *xe);
 
+void xe_bo_pci_dev_remove_all(struct xe_device *xe);
+
+int xe_bo_pinned_init(struct xe_device *xe);
 #endif
diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index 15a92e3d4898..eb5e83c5f233 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -28,6 +28,10 @@ struct xe_vm;
 struct xe_bo {
 	/** @ttm: TTM base buffer object */
 	struct ttm_buffer_object ttm;
+	/** @backup_obj: The backup object when pinned and suspended (vram only) */
+	struct xe_bo *backup_obj;
+	/** @parent_obj: Ref to parent bo if this a backup_obj */
+	struct xe_bo *parent_obj;
 	/** @size: Size of this buffer object */
 	size_t size;
 	/** @flags: flags for this buffer object */
diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
new file mode 100644
index 000000000000..cb9f175c89a1
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -0,0 +1,250 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/configfs.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include "xe_configfs.h"
+#include "xe_module.h"
+
+/**
+ * DOC: Xe Configfs
+ *
+ * Overview
+ * =========
+ *
+ * Configfs is a filesystem-based manager of kernel objects. XE KMD registers a
+ * configfs subsystem called ``'xe'`` that creates a directory in the mounted configfs directory
+ * The user can create devices under this directory and configure them as necessary
+ * See Documentation/filesystems/configfs.rst for more information about how configfs works.
+ *
+ * Create devices
+ * ===============
+ *
+ * In order to create a device, the user has to create a directory inside ``'xe'``::
+ *
+ *	mkdir /sys/kernel/config/xe/0000:03:00.0/
+ *
+ * Every device created is populated by the driver with entries that can be
+ * used to configure it::
+ *
+ *	/sys/kernel/config/xe/
+ *		.. 0000:03:00.0/
+ *			... survivability_mode
+ *
+ * Configure Attributes
+ * ====================
+ *
+ * Survivability mode:
+ * -------------------
+ *
+ * Enable survivability mode on supported cards. This setting only takes
+ * effect when probing the device. Example to enable it::
+ *
+ *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/survivability_mode
+ *	# echo 0000:03:00.0 > /sys/bus/pci/drivers/xe/bind  (Enters survivability mode if supported)
+ *
+ * Remove devices
+ * ==============
+ *
+ * The created device directories can be removed using ``rmdir``::
+ *
+ *	rmdir /sys/kernel/config/xe/0000:03:00.0/
+ */
+
+struct xe_config_device {
+	struct config_group group;
+
+	bool survivability_mode;
+
+	/* protects attributes */
+	struct mutex lock;
+};
+
+static struct xe_config_device *to_xe_config_device(struct config_item *item)
+{
+	return container_of(to_config_group(item), struct xe_config_device, group);
+}
+
+static ssize_t survivability_mode_show(struct config_item *item, char *page)
+{
+	struct xe_config_device *dev = to_xe_config_device(item);
+
+	return sprintf(page, "%d\n", dev->survivability_mode);
+}
+
+static ssize_t survivability_mode_store(struct config_item *item, const char *page, size_t len)
+{
+	struct xe_config_device *dev = to_xe_config_device(item);
+	bool survivability_mode;
+	int ret;
+
+	ret = kstrtobool(page, &survivability_mode);
+	if (ret)
+		return ret;
+
+	mutex_lock(&dev->lock);
+	dev->survivability_mode = survivability_mode;
+	mutex_unlock(&dev->lock);
+
+	return len;
+}
+
+CONFIGFS_ATTR(, survivability_mode);
+
+static struct configfs_attribute *xe_config_device_attrs[] = {
+	&attr_survivability_mode,
+	NULL,
+};
+
+static void xe_config_device_release(struct config_item *item)
+{
+	struct xe_config_device *dev = to_xe_config_device(item);
+
+	mutex_destroy(&dev->lock);
+	kfree(dev);
+}
+
+static struct configfs_item_operations xe_config_device_ops = {
+	.release	= xe_config_device_release,
+};
+
+static const struct config_item_type xe_config_device_type = {
+	.ct_item_ops	= &xe_config_device_ops,
+	.ct_attrs	= xe_config_device_attrs,
+	.ct_owner	= THIS_MODULE,
+};
+
+static struct config_group *xe_config_make_device_group(struct config_group *group,
+							const char *name)
+{
+	unsigned int domain, bus, slot, function;
+	struct xe_config_device *dev;
+	struct pci_dev *pdev;
+	int ret;
+
+	ret = sscanf(name, "%04x:%02x:%02x.%x", &domain, &bus, &slot, &function);
+	if (ret != 4)
+		return ERR_PTR(-EINVAL);
+
+	pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function));
+	if (!pdev)
+		return ERR_PTR(-EINVAL);
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return ERR_PTR(-ENOMEM);
+
+	config_group_init_type_name(&dev->group, name, &xe_config_device_type);
+
+	mutex_init(&dev->lock);
+
+	return &dev->group;
+}
+
+static struct configfs_group_operations xe_config_device_group_ops = {
+	.make_group	= xe_config_make_device_group,
+};
+
+static const struct config_item_type xe_configfs_type = {
+	.ct_group_ops	= &xe_config_device_group_ops,
+	.ct_owner	= THIS_MODULE,
+};
+
+static struct configfs_subsystem xe_configfs = {
+	.su_group = {
+		.cg_item = {
+			.ci_namebuf = "xe",
+			.ci_type = &xe_configfs_type,
+		},
+	},
+};
+
+static struct xe_config_device *configfs_find_group(struct pci_dev *pdev)
+{
+	struct config_item *item;
+	char name[64];
+
+	snprintf(name, sizeof(name), "%04x:%02x:%02x.%x", pci_domain_nr(pdev->bus),
+		 pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn));
+
+	mutex_lock(&xe_configfs.su_mutex);
+	item = config_group_find_item(&xe_configfs.su_group, name);
+	mutex_unlock(&xe_configfs.su_mutex);
+
+	if (!item)
+		return NULL;
+
+	return to_xe_config_device(item);
+}
+
+/**
+ * xe_configfs_get_survivability_mode - get configfs survivability mode attribute
+ * @pdev: pci device
+ *
+ * find the configfs group that belongs to the pci device and return
+ * the survivability mode attribute
+ *
+ * Return: survivability mode if config group is found, false otherwise
+ */
+bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
+{
+	struct xe_config_device *dev = configfs_find_group(pdev);
+	bool mode;
+
+	if (!dev)
+		return false;
+
+	mode = dev->survivability_mode;
+	config_item_put(&dev->group.cg_item);
+
+	return mode;
+}
+
+/**
+ * xe_configfs_clear_survivability_mode - clear configfs survivability mode attribute
+ * @pdev: pci device
+ *
+ * find the configfs group that belongs to the pci device and clear survivability
+ * mode attribute
+ */
+void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
+{
+	struct xe_config_device *dev = configfs_find_group(pdev);
+
+	if (!dev)
+		return;
+
+	mutex_lock(&dev->lock);
+	dev->survivability_mode = 0;
+	mutex_unlock(&dev->lock);
+
+	config_item_put(&dev->group.cg_item);
+}
+
+int __init xe_configfs_init(void)
+{
+	struct config_group *root = &xe_configfs.su_group;
+	int ret;
+
+	config_group_init(root);
+	mutex_init(&xe_configfs.su_mutex);
+	ret = configfs_register_subsystem(&xe_configfs);
+	if (ret) {
+		pr_err("Error %d while registering %s subsystem\n",
+		       ret, root->cg_item.ci_namebuf);
+		return ret;
+	}
+
+	return 0;
+}
+
+void __exit xe_configfs_exit(void)
+{
+	configfs_unregister_subsystem(&xe_configfs);
+}
+
diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
new file mode 100644
index 000000000000..d7d041ec2611
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+#ifndef _XE_CONFIGFS_H_
+#define _XE_CONFIGFS_H_
+
+#include <linux/types.h>
+
+struct pci_dev;
+
+#if IS_ENABLED(CONFIG_CONFIGFS_FS)
+int xe_configfs_init(void);
+void xe_configfs_exit(void);
+bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
+void xe_configfs_clear_survivability_mode(struct pci_dev *pdev);
+#else
+static inline int xe_configfs_init(void) { return 0; };
+static inline void xe_configfs_exit(void) {};
+static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; };
+static inline void xe_configfs_clear_survivability_mode(struct pci_dev *pdev) {};
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c
index 81b9d9bb3f57..7a8af2311318 100644
--- a/drivers/gpu/drm/xe/xe_devcoredump.c
+++ b/drivers/gpu/drm/xe/xe_devcoredump.c
@@ -80,7 +80,8 @@ static struct xe_guc *exec_queue_to_guc(struct xe_exec_queue *q)
 	return &q->gt->uc.guc;
 }
 
-static ssize_t __xe_devcoredump_read(char *buffer, size_t count,
+static ssize_t __xe_devcoredump_read(char *buffer, ssize_t count,
+				     ssize_t start,
 				     struct xe_devcoredump *coredump)
 {
 	struct xe_device *xe;
@@ -94,7 +95,7 @@ static ssize_t __xe_devcoredump_read(char *buffer, size_t count,
 	ss = &coredump->snapshot;
 
 	iter.data = buffer;
-	iter.start = 0;
+	iter.start = start;
 	iter.remain = count;
 
 	p = drm_coredump_printer(&iter);
@@ -168,12 +169,16 @@ static void xe_devcoredump_snapshot_free(struct xe_devcoredump_snapshot *ss)
 	ss->vm = NULL;
 }
 
+#define XE_DEVCOREDUMP_CHUNK_MAX	(SZ_512M + SZ_1G)
+
 static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
 				   size_t count, void *data, size_t datalen)
 {
 	struct xe_devcoredump *coredump = data;
 	struct xe_devcoredump_snapshot *ss;
 	ssize_t byte_copied;
+	u32 chunk_offset;
+	ssize_t new_chunk_position;
 
 	if (!coredump)
 		return -ENODEV;
@@ -183,6 +188,9 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
 	/* Ensure delayed work is captured before continuing */
 	flush_work(&ss->work);
 
+	if (ss->read.size > XE_DEVCOREDUMP_CHUNK_MAX)
+		xe_pm_runtime_get(gt_to_xe(ss->gt));
+
 	mutex_lock(&coredump->lock);
 
 	if (!ss->read.buffer) {
@@ -195,12 +203,29 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
 		return 0;
 	}
 
+	new_chunk_position = div_u64_rem(offset,
+					 XE_DEVCOREDUMP_CHUNK_MAX,
+					 &chunk_offset);
+
+	if (offset >= ss->read.chunk_position + XE_DEVCOREDUMP_CHUNK_MAX ||
+	    offset < ss->read.chunk_position) {
+		ss->read.chunk_position = new_chunk_position *
+			XE_DEVCOREDUMP_CHUNK_MAX;
+
+		__xe_devcoredump_read(ss->read.buffer,
+				      XE_DEVCOREDUMP_CHUNK_MAX,
+				      ss->read.chunk_position, coredump);
+	}
+
 	byte_copied = count < ss->read.size - offset ? count :
 		ss->read.size - offset;
-	memcpy(buffer, ss->read.buffer + offset, byte_copied);
+	memcpy(buffer, ss->read.buffer + chunk_offset, byte_copied);
 
 	mutex_unlock(&coredump->lock);
 
+	if (ss->read.size > XE_DEVCOREDUMP_CHUNK_MAX)
+		xe_pm_runtime_put(gt_to_xe(ss->gt));
+
 	return byte_copied;
 }
 
@@ -254,17 +279,32 @@ static void xe_devcoredump_deferred_snap_work(struct work_struct *work)
 	xe_guc_exec_queue_snapshot_capture_delayed(ss->ge);
 	xe_force_wake_put(gt_to_fw(ss->gt), fw_ref);
 
-	xe_pm_runtime_put(xe);
+	ss->read.chunk_position = 0;
 
 	/* Calculate devcoredump size */
-	ss->read.size = __xe_devcoredump_read(NULL, INT_MAX, coredump);
-
-	ss->read.buffer = kvmalloc(ss->read.size, GFP_USER);
-	if (!ss->read.buffer)
-		return;
+	ss->read.size = __xe_devcoredump_read(NULL, LONG_MAX, 0, coredump);
+
+	if (ss->read.size > XE_DEVCOREDUMP_CHUNK_MAX) {
+		ss->read.buffer = kvmalloc(XE_DEVCOREDUMP_CHUNK_MAX,
+					   GFP_USER);
+		if (!ss->read.buffer)
+			goto put_pm;
+
+		__xe_devcoredump_read(ss->read.buffer,
+				      XE_DEVCOREDUMP_CHUNK_MAX,
+				      0, coredump);
+	} else {
+		ss->read.buffer = kvmalloc(ss->read.size, GFP_USER);
+		if (!ss->read.buffer)
+			goto put_pm;
+
+		__xe_devcoredump_read(ss->read.buffer, ss->read.size, 0,
+				      coredump);
+		xe_devcoredump_snapshot_free(ss);
+	}
 
-	__xe_devcoredump_read(ss->read.buffer, ss->read.size, coredump);
-	xe_devcoredump_snapshot_free(ss);
+put_pm:
+	xe_pm_runtime_put(xe);
 }
 
 static void devcoredump_snapshot(struct xe_devcoredump *coredump,
@@ -425,7 +465,7 @@ void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, char suffi
 	if (offset & 3)
 		drm_printf(p, "Offset not word aligned: %zu", offset);
 
-	line_buff = kzalloc(DMESG_MAX_LINE_LEN, GFP_KERNEL);
+	line_buff = kzalloc(DMESG_MAX_LINE_LEN, GFP_ATOMIC);
 	if (!line_buff) {
 		drm_printf(p, "Failed to allocate line buffer\n");
 		return;
diff --git a/drivers/gpu/drm/xe/xe_devcoredump_types.h b/drivers/gpu/drm/xe/xe_devcoredump_types.h
index 1a1d16a96b2d..a174385a6d83 100644
--- a/drivers/gpu/drm/xe/xe_devcoredump_types.h
+++ b/drivers/gpu/drm/xe/xe_devcoredump_types.h
@@ -66,6 +66,8 @@ struct xe_devcoredump_snapshot {
 	struct {
 		/** @read.size: size of devcoredump in human readable format */
 		ssize_t size;
+		/** @read.chunk_position: position of devcoredump chunk */
+		ssize_t chunk_position;
 		/** @read.buffer: buffer of devcoredump in human readable format */
 		char *buffer;
 	} read;
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 00191227bc95..c02c4c4e9412 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -23,8 +23,10 @@
 #include "regs/xe_gt_regs.h"
 #include "regs/xe_regs.h"
 #include "xe_bo.h"
+#include "xe_bo_evict.h"
 #include "xe_debugfs.h"
 #include "xe_devcoredump.h"
+#include "xe_device_sysfs.h"
 #include "xe_dma_buf.h"
 #include "xe_drm_client.h"
 #include "xe_drv.h"
@@ -467,10 +469,9 @@ struct xe_device *xe_device_create(struct pci_dev *pdev,
 			xa_erase(&xe->usm.asid_to_vm, asid);
 	}
 
-	spin_lock_init(&xe->pinned.lock);
-	INIT_LIST_HEAD(&xe->pinned.kernel_bo_present);
-	INIT_LIST_HEAD(&xe->pinned.external_vram);
-	INIT_LIST_HEAD(&xe->pinned.evicted);
+	err = xe_bo_pinned_init(xe);
+	if (err)
+		goto err;
 
 	xe->preempt_fence_wq = alloc_ordered_workqueue("xe-preempt-fence-wq",
 						       WQ_MEM_RECLAIM);
@@ -505,7 +506,15 @@ ALLOW_ERROR_INJECTION(xe_device_create, ERRNO); /* See xe_pci_probe() */
 
 static bool xe_driver_flr_disabled(struct xe_device *xe)
 {
-	return xe_mmio_read32(xe_root_tile_mmio(xe), GU_CNTL_PROTECTED) & DRIVERINT_FLR_DIS;
+	if (IS_SRIOV_VF(xe))
+		return true;
+
+	if (xe_mmio_read32(xe_root_tile_mmio(xe), GU_CNTL_PROTECTED) & DRIVERINT_FLR_DIS) {
+		drm_info(&xe->drm, "Driver-FLR disabled by BIOS\n");
+		return true;
+	}
+
+	return false;
 }
 
 /*
@@ -523,7 +532,7 @@ static bool xe_driver_flr_disabled(struct xe_device *xe)
  */
 static void __xe_driver_flr(struct xe_device *xe)
 {
-	const unsigned int flr_timeout = 3 * MICRO; /* specs recommend a 3s wait */
+	const unsigned int flr_timeout = 3 * USEC_PER_SEC; /* specs recommend a 3s wait */
 	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
 	int ret;
 
@@ -569,10 +578,8 @@ static void __xe_driver_flr(struct xe_device *xe)
 
 static void xe_driver_flr(struct xe_device *xe)
 {
-	if (xe_driver_flr_disabled(xe)) {
-		drm_info_once(&xe->drm, "BIOS Disabled Driver-FLR\n");
+	if (xe_driver_flr_disabled(xe))
 		return;
-	}
 
 	__xe_driver_flr(xe);
 }
@@ -706,7 +713,7 @@ int xe_device_probe_early(struct xe_device *xe)
 	sriov_update_device_info(xe);
 
 	err = xe_pcode_probe_early(xe);
-	if (err) {
+	if (err || xe_survivability_mode_is_requested(xe)) {
 		int save_err = err;
 
 		/*
@@ -729,6 +736,7 @@ int xe_device_probe_early(struct xe_device *xe)
 
 	return 0;
 }
+ALLOW_ERROR_INJECTION(xe_device_probe_early, ERRNO); /* See xe_pci_probe() */
 
 static int probe_has_flat_ccs(struct xe_device *xe)
 {
@@ -908,6 +916,10 @@ int xe_device_probe(struct xe_device *xe)
 	if (err)
 		goto err_unregister_display;
 
+	err = xe_device_sysfs_init(xe);
+	if (err)
+		goto err_unregister_display;
+
 	xe_debugfs_register(xe);
 
 	err = xe_hwmon_register(xe);
@@ -932,6 +944,8 @@ void xe_device_remove(struct xe_device *xe)
 	xe_display_unregister(xe);
 
 	drm_dev_unplug(&xe->drm);
+
+	xe_bo_pci_dev_remove_all(xe);
 }
 
 void xe_device_shutdown(struct xe_device *xe)
diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index 7efbd4c52791..9628e9a0a0af 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -3,14 +3,16 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include <linux/device.h>
 #include <linux/kobject.h>
 #include <linux/pci.h>
 #include <linux/sysfs.h>
 
-#include <drm/drm_managed.h>
-
 #include "xe_device.h"
 #include "xe_device_sysfs.h"
+#include "xe_mmio.h"
+#include "xe_pcode_api.h"
+#include "xe_pcode.h"
 #include "xe_pm.h"
 
 /**
@@ -63,11 +65,93 @@ vram_d3cold_threshold_store(struct device *dev, struct device_attribute *attr,
 
 static DEVICE_ATTR_RW(vram_d3cold_threshold);
 
+/**
+ * DOC: PCIe Gen5 Limitations
+ *
+ * Default link speed of discrete GPUs is determined by configuration parameters
+ * stored in their flash memory, which are subject to override through user
+ * initiated firmware updates. It has been observed that devices configured with
+ * PCIe Gen5 as their default link speed can come across link quality issues due
+ * to host or motherboard limitations and may have to auto-downgrade their link
+ * to PCIe Gen4 speed when faced with unstable link at Gen5, which makes
+ * firmware updates rather risky on such setups. It is required to ensure that
+ * the device is capable of auto-downgrading its link to PCIe Gen4 speed before
+ * pushing the firmware image with PCIe Gen5 as default configuration. This can
+ * be done by reading ``auto_link_downgrade_capable`` sysfs entry, which will
+ * denote if the device is capable of auto-downgrading its link to PCIe Gen4
+ * speed with boolean output value of ``0`` or ``1``, meaning `incapable` or
+ * `capable` respectively.
+ *
+ * .. code-block:: shell
+ *
+ *    $ cat /sys/bus/pci/devices/<bdf>/auto_link_downgrade_capable
+ *
+ * Pushing the firmware image with PCIe Gen5 as default configuration on a auto
+ * link downgrade incapable device and facing link instability due to host or
+ * motherboard limitations can result in driver failing to bind to the device,
+ * making further firmware updates impossible with RMA being the only last
+ * resort.
+ *
+ * Link downgrade status of auto link downgrade capable devices is available
+ * through ``auto_link_downgrade_status`` sysfs entry with boolean output value
+ * of ``0`` or ``1``, where ``0`` means no auto-downgrading was required during
+ * link training (which is the optimal scenario) and ``1`` means the device has
+ * auto-downgraded its link to PCIe Gen4 speed due to unstable Gen5 link.
+ *
+ * .. code-block:: shell
+ *
+ *    $ cat /sys/bus/pci/devices/<bdf>/auto_link_downgrade_status
+ */
+
+static ssize_t
+auto_link_downgrade_capable_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct xe_device *xe = pdev_to_xe_device(pdev);
+	u32 cap, val;
+
+	xe_pm_runtime_get(xe);
+	val = xe_mmio_read32(xe_root_tile_mmio(xe), BMG_PCIE_CAP);
+	xe_pm_runtime_put(xe);
+
+	cap = REG_FIELD_GET(LINK_DOWNGRADE, val);
+	return sysfs_emit(buf, "%u\n", cap == DOWNGRADE_CAPABLE ? true : false);
+}
+static DEVICE_ATTR_ADMIN_RO(auto_link_downgrade_capable);
+
+static ssize_t
+auto_link_downgrade_status_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct xe_device *xe = pdev_to_xe_device(pdev);
+	u32 val;
+	int ret;
+
+	xe_pm_runtime_get(xe);
+	ret = xe_pcode_read(xe_device_get_root_tile(xe),
+			    PCODE_MBOX(DGFX_PCODE_STATUS, DGFX_GET_INIT_STATUS, 0),
+			    &val, NULL);
+	xe_pm_runtime_put(xe);
+
+	return ret ?: sysfs_emit(buf, "%u\n", REG_FIELD_GET(DGFX_LINK_DOWNGRADE_STATUS, val));
+}
+static DEVICE_ATTR_ADMIN_RO(auto_link_downgrade_status);
+
+static const struct attribute *auto_link_downgrade_attrs[] = {
+	&dev_attr_auto_link_downgrade_capable.attr,
+	&dev_attr_auto_link_downgrade_status.attr,
+	NULL
+};
+
 static void xe_device_sysfs_fini(void *arg)
 {
 	struct xe_device *xe = arg;
 
-	sysfs_remove_file(&xe->drm.dev->kobj, &dev_attr_vram_d3cold_threshold.attr);
+	if (xe->d3cold.capable)
+		sysfs_remove_file(&xe->drm.dev->kobj, &dev_attr_vram_d3cold_threshold.attr);
+
+	if (xe->info.platform == XE_BATTLEMAGE)
+		sysfs_remove_files(&xe->drm.dev->kobj, auto_link_downgrade_attrs);
 }
 
 int xe_device_sysfs_init(struct xe_device *xe)
@@ -75,9 +159,17 @@ int xe_device_sysfs_init(struct xe_device *xe)
 	struct device *dev = xe->drm.dev;
 	int ret;
 
-	ret = sysfs_create_file(&dev->kobj, &dev_attr_vram_d3cold_threshold.attr);
-	if (ret)
-		return ret;
+	if (xe->d3cold.capable) {
+		ret = sysfs_create_file(&dev->kobj, &dev_attr_vram_d3cold_threshold.attr);
+		if (ret)
+			return ret;
+	}
+
+	if (xe->info.platform == XE_BATTLEMAGE) {
+		ret = sysfs_create_files(&dev->kobj, auto_link_downgrade_attrs);
+		if (ret)
+			return ret;
+	}
 
 	return devm_add_action_or_reset(dev, xe_device_sysfs_fini, xe);
 }
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 72ef0b6fc425..06c65dace026 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -31,7 +31,6 @@
 #endif
 
 #if IS_ENABLED(CONFIG_DRM_XE_DISPLAY)
-#include "soc/intel_pch.h"
 #include "intel_display_core.h"
 #include "intel_display_device.h"
 #endif
@@ -107,6 +106,9 @@ struct xe_vram_region {
 	resource_size_t actual_physical_size;
 	/** @mapping: pointer to VRAM mappable space */
 	void __iomem *mapping;
+	/** @ttm: VRAM TTM manager */
+	struct xe_ttm_vram_mgr ttm;
+#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR)
 	/** @pagemap: Used to remap device memory as ZONE_DEVICE */
 	struct dev_pagemap pagemap;
 	/**
@@ -120,8 +122,7 @@ struct xe_vram_region {
 	 * This is generated when remap device memory as ZONE_DEVICE
 	 */
 	resource_size_t hpa_base;
-	/** @ttm: VRAM TTM manager */
-	struct xe_ttm_vram_mgr ttm;
+#endif
 };
 
 /**
@@ -314,6 +315,8 @@ struct xe_device {
 		u8 has_atomic_enable_pte_bit:1;
 		/** @info.has_device_atomics_on_smem: Supports device atomics on SMEM */
 		u8 has_device_atomics_on_smem:1;
+		/** @info.has_fan_control: Device supports fan control */
+		u8 has_fan_control:1;
 		/** @info.has_flat_ccs: Whether flat CCS metadata is used */
 		u8 has_flat_ccs:1;
 		/** @info.has_heci_cscfi: device has heci cscfi */
@@ -332,6 +335,8 @@ struct xe_device {
 		u8 has_usm:1;
 		/** @info.is_dgfx: is discrete device */
 		u8 is_dgfx:1;
+		/** @info.needs_scratch: needs scratch page for oob prefetch to work */
+		u8 needs_scratch:1;
 		/**
 		 * @info.probe_display: Probe display hardware.  If set to
 		 * false, the driver will behave as if there is no display
@@ -418,12 +423,22 @@ struct xe_device {
 	struct {
 		/** @pinned.lock: protected pinned BO list state */
 		spinlock_t lock;
-		/** @pinned.kernel_bo_present: pinned kernel BO that are present */
-		struct list_head kernel_bo_present;
-		/** @pinned.evicted: pinned BO that have been evicted */
-		struct list_head evicted;
-		/** @pinned.external_vram: pinned external BO in vram*/
-		struct list_head external_vram;
+		/** @pinned.early: early pinned lists */
+		struct {
+			/** @pinned.early.kernel_bo_present: pinned kernel BO that are present */
+			struct list_head kernel_bo_present;
+			/** @pinned.early.evicted: pinned BO that have been evicted */
+			struct list_head evicted;
+		} early;
+		/** @pinned.late: late pinned lists */
+		struct {
+			/** @pinned.late.kernel_bo_present: pinned kernel BO that are present */
+			struct list_head kernel_bo_present;
+			/** @pinned.late.evicted: pinned BO that have been evicted */
+			struct list_head evicted;
+			/** @pinned.external: pinned external and dma-buf. */
+			struct list_head external;
+		} late;
 	} pinned;
 
 	/** @ufence_wq: user fence wait queue */
@@ -506,6 +521,9 @@ struct xe_device {
 		struct mutex lock;
 	} d3cold;
 
+	/** @pm_notifier: Our PM notifier to perform actions in response to various PM events. */
+	struct notifier_block pm_notifier;
+
 	/** @pmt: Support the PMT driver callback interface */
 	struct {
 		/** @pmt.lock: protect access for telemetry data */
@@ -570,7 +588,6 @@ struct xe_device {
 	 * migrating to the right sub-structs
 	 */
 	struct intel_display display;
-	enum intel_pch pch_type;
 
 	struct dram_info {
 		bool wm_lv_0_adjust_needed;
@@ -585,6 +602,8 @@ struct xe_device {
 			INTEL_DRAM_DDR5,
 			INTEL_DRAM_LPDDR5,
 			INTEL_DRAM_GDDR,
+			INTEL_DRAM_GDDR_ECC,
+			__INTEL_DRAM_TYPE_MAX,
 		} type;
 		u8 num_qgv_points;
 		u8 num_psf_gv_points;
diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index f67803e15a0e..346f857f3837 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -145,10 +145,7 @@ static void xe_dma_buf_unmap(struct dma_buf_attachment *attach,
 			     struct sg_table *sgt,
 			     enum dma_data_direction dir)
 {
-	struct dma_buf *dma_buf = attach->dmabuf;
-	struct xe_bo *bo = gem_to_xe_bo(dma_buf->priv);
-
-	if (!xe_bo_is_vram(bo)) {
+	if (sg_page(sgt->sgl)) {
 		dma_unmap_sgtable(attach->dev, sgt, dir, 0);
 		sg_free_table(sgt);
 		kfree(sgt);
@@ -236,7 +233,7 @@ static void xe_dma_buf_move_notify(struct dma_buf_attachment *attach)
 	struct drm_gem_object *obj = attach->importer_priv;
 	struct xe_bo *bo = gem_to_xe_bo(obj);
 
-	XE_WARN_ON(xe_bo_evict(bo, false));
+	XE_WARN_ON(xe_bo_evict(bo));
 }
 
 static const struct dma_buf_attach_ops xe_dma_buf_attach_ops = {
diff --git a/drivers/gpu/drm/xe/xe_eu_stall.c b/drivers/gpu/drm/xe/xe_eu_stall.c
index f2bb9168967c..96732613b4b7 100644
--- a/drivers/gpu/drm/xe/xe_eu_stall.c
+++ b/drivers/gpu/drm/xe/xe_eu_stall.c
@@ -52,6 +52,8 @@ struct xe_eu_stall_data_stream {
 
 	struct xe_gt *gt;
 	struct xe_bo *bo;
+	/* Lock to protect data buffer pointers */
+	struct mutex xecore_buf_lock;
 	struct per_xecore_buf *xecore_buf;
 	struct {
 		bool reported_to_user;
@@ -208,6 +210,9 @@ int xe_eu_stall_init(struct xe_gt *gt)
 	struct xe_device *xe = gt_to_xe(gt);
 	int ret;
 
+	if (!xe_eu_stall_supported_on_platform(xe))
+		return 0;
+
 	gt->eu_stall = kzalloc(sizeof(*gt->eu_stall), GFP_KERNEL);
 	if (!gt->eu_stall) {
 		ret = -ENOMEM;
@@ -278,7 +283,7 @@ static int xe_eu_stall_user_ext_set_property(struct xe_device *xe, u64 extension
 	int err;
 	u32 idx;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
@@ -308,7 +313,7 @@ static int xe_eu_stall_user_extensions(struct xe_device *xe, u64 extension,
 	if (XE_IOCTL_DBG(xe, ext_number >= MAX_USER_EXTENSIONS))
 		return -E2BIG;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
@@ -378,7 +383,7 @@ static bool eu_stall_data_buf_poll(struct xe_eu_stall_data_stream *stream)
 	u16 group, instance;
 	unsigned int xecore;
 
-	mutex_lock(&gt->eu_stall->stream_lock);
+	mutex_lock(&stream->xecore_buf_lock);
 	for_each_dss_steering(xecore, gt, group, instance) {
 		xecore_buf = &stream->xecore_buf[xecore];
 		read_ptr = xecore_buf->read;
@@ -396,7 +401,7 @@ static bool eu_stall_data_buf_poll(struct xe_eu_stall_data_stream *stream)
 			set_bit(xecore, stream->data_drop.mask);
 		xecore_buf->write = write_ptr;
 	}
-	mutex_unlock(&gt->eu_stall->stream_lock);
+	mutex_unlock(&stream->xecore_buf_lock);
 
 	return min_data_present;
 }
@@ -511,11 +516,13 @@ static ssize_t xe_eu_stall_stream_read_locked(struct xe_eu_stall_data_stream *st
 	unsigned int xecore;
 	int ret = 0;
 
+	mutex_lock(&stream->xecore_buf_lock);
 	if (bitmap_weight(stream->data_drop.mask, XE_MAX_DSS_FUSE_BITS)) {
 		if (!stream->data_drop.reported_to_user) {
 			stream->data_drop.reported_to_user = true;
 			xe_gt_dbg(gt, "EU stall data dropped in XeCores: %*pb\n",
 				  XE_MAX_DSS_FUSE_BITS, stream->data_drop.mask);
+			mutex_unlock(&stream->xecore_buf_lock);
 			return -EIO;
 		}
 		stream->data_drop.reported_to_user = false;
@@ -527,6 +534,7 @@ static ssize_t xe_eu_stall_stream_read_locked(struct xe_eu_stall_data_stream *st
 		if (ret || count == total_size)
 			break;
 	}
+	mutex_unlock(&stream->xecore_buf_lock);
 	return total_size ?: (ret ?: -EAGAIN);
 }
 
@@ -583,6 +591,7 @@ static void xe_eu_stall_stream_free(struct xe_eu_stall_data_stream *stream)
 {
 	struct xe_gt *gt = stream->gt;
 
+	mutex_destroy(&stream->xecore_buf_lock);
 	gt->eu_stall->stream = NULL;
 	kfree(stream);
 }
@@ -718,6 +727,7 @@ static int xe_eu_stall_stream_init(struct xe_eu_stall_data_stream *stream,
 	}
 
 	init_waitqueue_head(&stream->poll_wq);
+	mutex_init(&stream->xecore_buf_lock);
 	INIT_DELAYED_WORK(&stream->buf_poll_work, eu_stall_data_buf_poll_work_fn);
 	stream->per_xecore_buf_size = per_xecore_buf_size;
 	stream->sampling_rate_mult = props->sampling_rate_mult;
diff --git a/drivers/gpu/drm/xe/xe_eu_stall.h b/drivers/gpu/drm/xe/xe_eu_stall.h
index ed9d0f233566..d1c76e503799 100644
--- a/drivers/gpu/drm/xe/xe_eu_stall.h
+++ b/drivers/gpu/drm/xe/xe_eu_stall.h
@@ -7,6 +7,7 @@
 #define __XE_EU_STALL_H__
 
 #include "xe_gt_types.h"
+#include "xe_sriov.h"
 
 size_t xe_eu_stall_get_per_xecore_buf_size(void);
 size_t xe_eu_stall_data_record_size(struct xe_device *xe);
@@ -19,6 +20,6 @@ int xe_eu_stall_stream_open(struct drm_device *dev,
 
 static inline bool xe_eu_stall_supported_on_platform(struct xe_device *xe)
 {
-	return xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20;
+	return !IS_SRIOV_VF(xe) && (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20);
 }
 #endif
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index b75adfc99fb7..44364c042ad7 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -176,8 +176,8 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	}
 
 	if (xe_exec_queue_is_parallel(q)) {
-		err = __copy_from_user(addresses, addresses_user, sizeof(u64) *
-				       q->width);
+		err = copy_from_user(addresses, addresses_user, sizeof(u64) *
+				     q->width);
 		if (err) {
 			err = -EFAULT;
 			goto err_syncs;
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 606922d9dd73..21d4ced31dd9 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -479,7 +479,7 @@ static int exec_queue_user_ext_set_property(struct xe_device *xe,
 	int err;
 	u32 idx;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
@@ -518,7 +518,7 @@ static int exec_queue_user_extensions(struct xe_device *xe, struct xe_exec_queue
 	if (XE_IOCTL_DBG(xe, ext_number >= MAX_USER_EXTENSIONS))
 		return -E2BIG;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
@@ -618,9 +618,8 @@ int xe_exec_queue_create_ioctl(struct drm_device *dev, void *data,
 	if (XE_IOCTL_DBG(xe, !len || len > XE_HW_ENGINE_MAX_INSTANCE))
 		return -EINVAL;
 
-	err = __copy_from_user(eci, user_eci,
-			       sizeof(struct drm_xe_engine_class_instance) *
-			       len);
+	err = copy_from_user(eci, user_eci,
+			     sizeof(struct drm_xe_engine_class_instance) * len);
 	if (XE_IOCTL_DBG(xe, err))
 		return -EFAULT;
 
diff --git a/drivers/gpu/drm/xe/xe_force_wake.c b/drivers/gpu/drm/xe/xe_force_wake.c
index 4f6784e5abf8..8a5cba22b586 100644
--- a/drivers/gpu/drm/xe/xe_force_wake.c
+++ b/drivers/gpu/drm/xe/xe_force_wake.c
@@ -49,9 +49,6 @@ void xe_force_wake_init_gt(struct xe_gt *gt, struct xe_force_wake *fw)
 	fw->gt = gt;
 	spin_lock_init(&fw->lock);
 
-	/* Assuming gen11+ so assert this assumption is correct */
-	xe_gt_assert(gt, GRAPHICS_VER(gt_to_xe(gt)) >= 11);
-
 	if (xe->info.graphics_verx100 >= 1270) {
 		init_domain(fw, XE_FW_DOMAIN_ID_GT,
 			    FORCEWAKE_GT,
@@ -67,9 +64,6 @@ void xe_force_wake_init_engines(struct xe_gt *gt, struct xe_force_wake *fw)
 {
 	int i, j;
 
-	/* Assuming gen11+ so assert this assumption is correct */
-	xe_gt_assert(gt, GRAPHICS_VER(gt_to_xe(gt)) >= 11);
-
 	if (!xe_gt_is_media_type(gt))
 		init_domain(fw, XE_FW_DOMAIN_ID_RENDER,
 			    FORCEWAKE_RENDER,
diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index 5fcb2b4c2c13..7062115909f2 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -365,7 +365,7 @@ int xe_ggtt_init(struct xe_ggtt *ggtt)
 	 * scratch entries, rather keep the scratch page in system memory on
 	 * platforms where 64K pages are needed for VRAM.
 	 */
-	flags = XE_BO_FLAG_PINNED;
+	flags = 0;
 	if (ggtt->flags & XE_GGTT_FLAGS_64K)
 		flags |= XE_BO_FLAG_SYSTEM;
 	else
diff --git a/drivers/gpu/drm/xe/xe_gsc.c b/drivers/gpu/drm/xe/xe_gsc.c
index fd41113f8572..0bcf97063ff6 100644
--- a/drivers/gpu/drm/xe/xe_gsc.c
+++ b/drivers/gpu/drm/xe/xe_gsc.c
@@ -555,6 +555,28 @@ void xe_gsc_wait_for_worker_completion(struct xe_gsc *gsc)
 		flush_work(&gsc->work);
 }
 
+void xe_gsc_stop_prepare(struct xe_gsc *gsc)
+{
+	struct xe_gt *gt = gsc_to_gt(gsc);
+	int ret;
+
+	if (!xe_uc_fw_is_loadable(&gsc->fw) || xe_uc_fw_is_in_error_state(&gsc->fw))
+		return;
+
+	xe_force_wake_assert_held(gt_to_fw(gt), XE_FW_GSC);
+
+	/*
+	 * If the GSC FW load or the proxy init are interrupted, the only way
+	 * to recover it is to do an FLR and reload the GSC from scratch.
+	 * Therefore, let's wait for the init to complete before stopping
+	 * operations. The proxy init is the last step, so we can just wait on
+	 * that
+	 */
+	ret = xe_gsc_wait_for_proxy_init_done(gsc);
+	if (ret)
+		xe_gt_err(gt, "failed to wait for GSC init completion before uc stop\n");
+}
+
 /*
  * wa_14015076503: if the GSC FW is loaded, we need to alert it before doing a
  * GSC engine reset by writing a notification bit in the GS1 register and then
diff --git a/drivers/gpu/drm/xe/xe_gsc.h b/drivers/gpu/drm/xe/xe_gsc.h
index d99f66c38075..b8b8e0810ad9 100644
--- a/drivers/gpu/drm/xe/xe_gsc.h
+++ b/drivers/gpu/drm/xe/xe_gsc.h
@@ -16,6 +16,7 @@ struct xe_hw_engine;
 int xe_gsc_init(struct xe_gsc *gsc);
 int xe_gsc_init_post_hwconfig(struct xe_gsc *gsc);
 void xe_gsc_wait_for_worker_completion(struct xe_gsc *gsc);
+void xe_gsc_stop_prepare(struct xe_gsc *gsc);
 void xe_gsc_load_start(struct xe_gsc *gsc);
 void xe_gsc_hwe_irq_handler(struct xe_hw_engine *hwe, u16 intr_vec);
 
diff --git a/drivers/gpu/drm/xe/xe_gsc_proxy.c b/drivers/gpu/drm/xe/xe_gsc_proxy.c
index 8cf70b228ff3..d0519cd6704a 100644
--- a/drivers/gpu/drm/xe/xe_gsc_proxy.c
+++ b/drivers/gpu/drm/xe/xe_gsc_proxy.c
@@ -71,6 +71,17 @@ bool xe_gsc_proxy_init_done(struct xe_gsc *gsc)
 	       HECI1_FWSTS1_PROXY_STATE_NORMAL;
 }
 
+int xe_gsc_wait_for_proxy_init_done(struct xe_gsc *gsc)
+{
+	struct xe_gt *gt = gsc_to_gt(gsc);
+
+	/* Proxy init can take up to 500ms, so wait double that for safety */
+	return xe_mmio_wait32(&gt->mmio, HECI_FWSTS1(MTL_GSC_HECI1_BASE),
+			      HECI1_FWSTS1_CURRENT_STATE,
+			      HECI1_FWSTS1_PROXY_STATE_NORMAL,
+			      USEC_PER_SEC, NULL, false);
+}
+
 static void __gsc_proxy_irq_rmw(struct xe_gsc *gsc, u32 clr, u32 set)
 {
 	struct xe_gt *gt = gsc_to_gt(gsc);
diff --git a/drivers/gpu/drm/xe/xe_gsc_proxy.h b/drivers/gpu/drm/xe/xe_gsc_proxy.h
index fdef56995cd4..765602221dbc 100644
--- a/drivers/gpu/drm/xe/xe_gsc_proxy.h
+++ b/drivers/gpu/drm/xe/xe_gsc_proxy.h
@@ -12,6 +12,7 @@ struct xe_gsc;
 
 int xe_gsc_proxy_init(struct xe_gsc *gsc);
 bool xe_gsc_proxy_init_done(struct xe_gsc *gsc);
+int xe_gsc_wait_for_proxy_init_done(struct xe_gsc *gsc);
 int xe_gsc_proxy_start(struct xe_gsc *gsc);
 
 int xe_gsc_proxy_request_handler(struct xe_gsc *gsc);
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 10a9e3c72b36..0e5d243c9451 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -12,8 +12,10 @@
 
 #include <generated/xe_wa_oob.h>
 
+#include "instructions/xe_alu_commands.h"
 #include "instructions/xe_gfxpipe_commands.h"
 #include "instructions/xe_mi_commands.h"
+#include "regs/xe_engine_regs.h"
 #include "regs/xe_gt_regs.h"
 #include "xe_assert.h"
 #include "xe_bb.h"
@@ -176,15 +178,6 @@ static int emit_nop_job(struct xe_gt *gt, struct xe_exec_queue *q)
 	return 0;
 }
 
-/*
- * Convert back from encoded value to type-safe, only to be used when reg.mcr
- * is true
- */
-static struct xe_reg_mcr to_xe_reg_mcr(const struct xe_reg reg)
-{
-	return (const struct xe_reg_mcr){.__reg.raw = reg.raw };
-}
-
 static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q)
 {
 	struct xe_reg_sr *sr = &q->hwe->reg_lrc;
@@ -194,6 +187,7 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q)
 	struct xe_bb *bb;
 	struct dma_fence *fence;
 	long timeout;
+	int count_rmw = 0;
 	int count = 0;
 
 	if (q->hwe->class == XE_ENGINE_CLASS_RENDER)
@@ -206,30 +200,32 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q)
 	if (IS_ERR(bb))
 		return PTR_ERR(bb);
 
-	xa_for_each(&sr->xa, idx, entry)
-		++count;
+	/* count RMW registers as those will be handled separately */
+	xa_for_each(&sr->xa, idx, entry) {
+		if (entry->reg.masked || entry->clr_bits == ~0)
+			++count;
+		else
+			++count_rmw;
+	}
 
-	if (count) {
+	if (count || count_rmw)
 		xe_gt_dbg(gt, "LRC WA %s save-restore batch\n", sr->name);
 
+	if (count) {
+		/* emit single LRI with all non RMW regs */
+
 		bb->cs[bb->len++] = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(count);
 
 		xa_for_each(&sr->xa, idx, entry) {
 			struct xe_reg reg = entry->reg;
-			struct xe_reg_mcr reg_mcr = to_xe_reg_mcr(reg);
 			u32 val;
 
-			/*
-			 * Skip reading the register if it's not really needed
-			 */
 			if (reg.masked)
 				val = entry->clr_bits << 16;
-			else if (entry->clr_bits + 1)
-				val = (reg.mcr ?
-				       xe_gt_mcr_unicast_read_any(gt, reg_mcr) :
-				       xe_mmio_read32(&gt->mmio, reg)) & (~entry->clr_bits);
-			else
+			else if (entry->clr_bits == ~0)
 				val = 0;
+			else
+				continue;
 
 			val |= entry->set_bits;
 
@@ -239,6 +235,52 @@ static int emit_wa_job(struct xe_gt *gt, struct xe_exec_queue *q)
 		}
 	}
 
+	if (count_rmw) {
+		/* emit MI_MATH for each RMW reg */
+
+		xa_for_each(&sr->xa, idx, entry) {
+			if (entry->reg.masked || entry->clr_bits == ~0)
+				continue;
+
+			bb->cs[bb->len++] = MI_LOAD_REGISTER_REG | MI_LRR_DST_CS_MMIO;
+			bb->cs[bb->len++] = entry->reg.addr;
+			bb->cs[bb->len++] = CS_GPR_REG(0, 0).addr;
+
+			bb->cs[bb->len++] = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(2) |
+					    MI_LRI_LRM_CS_MMIO;
+			bb->cs[bb->len++] = CS_GPR_REG(0, 1).addr;
+			bb->cs[bb->len++] = entry->clr_bits;
+			bb->cs[bb->len++] = CS_GPR_REG(0, 2).addr;
+			bb->cs[bb->len++] = entry->set_bits;
+
+			bb->cs[bb->len++] = MI_MATH(8);
+			bb->cs[bb->len++] = CS_ALU_INSTR_LOAD(SRCA, REG0);
+			bb->cs[bb->len++] = CS_ALU_INSTR_LOADINV(SRCB, REG1);
+			bb->cs[bb->len++] = CS_ALU_INSTR_AND;
+			bb->cs[bb->len++] = CS_ALU_INSTR_STORE(REG0, ACCU);
+			bb->cs[bb->len++] = CS_ALU_INSTR_LOAD(SRCA, REG0);
+			bb->cs[bb->len++] = CS_ALU_INSTR_LOAD(SRCB, REG2);
+			bb->cs[bb->len++] = CS_ALU_INSTR_OR;
+			bb->cs[bb->len++] = CS_ALU_INSTR_STORE(REG0, ACCU);
+
+			bb->cs[bb->len++] = MI_LOAD_REGISTER_REG | MI_LRR_SRC_CS_MMIO;
+			bb->cs[bb->len++] = CS_GPR_REG(0, 0).addr;
+			bb->cs[bb->len++] = entry->reg.addr;
+
+			xe_gt_dbg(gt, "REG[%#x] = ~%#x|%#x\n",
+				  entry->reg.addr, entry->clr_bits, entry->set_bits);
+		}
+
+		/* reset used GPR */
+		bb->cs[bb->len++] = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(3) | MI_LRI_LRM_CS_MMIO;
+		bb->cs[bb->len++] = CS_GPR_REG(0, 0).addr;
+		bb->cs[bb->len++] = 0;
+		bb->cs[bb->len++] = CS_GPR_REG(0, 1).addr;
+		bb->cs[bb->len++] = 0;
+		bb->cs[bb->len++] = CS_GPR_REG(0, 2).addr;
+		bb->cs[bb->len++] = 0;
+	}
+
 	xe_lrc_emit_hwe_state_instructions(q, bb);
 
 	job = xe_bb_create_job(q, bb);
@@ -857,7 +899,7 @@ void xe_gt_suspend_prepare(struct xe_gt *gt)
 
 	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL);
 
-	xe_uc_stop_prepare(&gt->uc);
+	xe_uc_suspend_prepare(&gt->uc);
 
 	xe_force_wake_put(gt_to_fw(gt), fw_ref);
 }
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index 2d63a69cbfa3..119a55bb7580 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -92,22 +92,23 @@ static int hw_engines(struct xe_gt *gt, struct drm_printer *p)
 	struct xe_hw_engine *hwe;
 	enum xe_hw_engine_id id;
 	unsigned int fw_ref;
+	int ret = 0;
 
 	xe_pm_runtime_get(xe);
 	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL);
 	if (!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)) {
-		xe_pm_runtime_put(xe);
-		xe_force_wake_put(gt_to_fw(gt), fw_ref);
-		return -ETIMEDOUT;
+		ret = -ETIMEDOUT;
+		goto fw_put;
 	}
 
 	for_each_hw_engine(hwe, gt, id)
 		xe_hw_engine_print(hwe, p);
 
+fw_put:
 	xe_force_wake_put(gt_to_fw(gt), fw_ref);
 	xe_pm_runtime_put(xe);
 
-	return 0;
+	return ret;
 }
 
 static int powergate_info(struct xe_gt *gt, struct drm_printer *p)
@@ -299,20 +300,20 @@ static int hwconfig(struct xe_gt *gt, struct drm_printer *p)
 	return 0;
 }
 
-static const struct drm_info_list debugfs_list[] = {
-	{"hw_engines", .show = xe_gt_debugfs_simple_show, .data = hw_engines},
+/*
+ * only for GT debugfs files which can be safely used on the VF as well:
+ * - without access to the GT privileged registers
+ * - without access to the PF specific data
+ */
+static const struct drm_info_list vf_safe_debugfs_list[] = {
 	{"force_reset", .show = xe_gt_debugfs_simple_show, .data = force_reset},
 	{"force_reset_sync", .show = xe_gt_debugfs_simple_show, .data = force_reset_sync},
 	{"sa_info", .show = xe_gt_debugfs_simple_show, .data = sa_info},
 	{"topology", .show = xe_gt_debugfs_simple_show, .data = topology},
-	{"steering", .show = xe_gt_debugfs_simple_show, .data = steering},
 	{"ggtt", .show = xe_gt_debugfs_simple_show, .data = ggtt},
-	{"powergate_info", .show = xe_gt_debugfs_simple_show, .data = powergate_info},
 	{"register-save-restore", .show = xe_gt_debugfs_simple_show, .data = register_save_restore},
 	{"workarounds", .show = xe_gt_debugfs_simple_show, .data = workarounds},
 	{"tunings", .show = xe_gt_debugfs_simple_show, .data = tunings},
-	{"pat", .show = xe_gt_debugfs_simple_show, .data = pat},
-	{"mocs", .show = xe_gt_debugfs_simple_show, .data = mocs},
 	{"default_lrc_rcs", .show = xe_gt_debugfs_simple_show, .data = rcs_default_lrc},
 	{"default_lrc_ccs", .show = xe_gt_debugfs_simple_show, .data = ccs_default_lrc},
 	{"default_lrc_bcs", .show = xe_gt_debugfs_simple_show, .data = bcs_default_lrc},
@@ -322,6 +323,15 @@ static const struct drm_info_list debugfs_list[] = {
 	{"hwconfig", .show = xe_gt_debugfs_simple_show, .data = hwconfig},
 };
 
+/* everything else should be added here */
+static const struct drm_info_list pf_only_debugfs_list[] = {
+	{"hw_engines", .show = xe_gt_debugfs_simple_show, .data = hw_engines},
+	{"mocs", .show = xe_gt_debugfs_simple_show, .data = mocs},
+	{"pat", .show = xe_gt_debugfs_simple_show, .data = pat},
+	{"powergate_info", .show = xe_gt_debugfs_simple_show, .data = powergate_info},
+	{"steering", .show = xe_gt_debugfs_simple_show, .data = steering},
+};
+
 void xe_gt_debugfs_register(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
@@ -345,10 +355,15 @@ void xe_gt_debugfs_register(struct xe_gt *gt)
 	 */
 	root->d_inode->i_private = gt;
 
-	drm_debugfs_create_files(debugfs_list,
-				 ARRAY_SIZE(debugfs_list),
+	drm_debugfs_create_files(vf_safe_debugfs_list,
+				 ARRAY_SIZE(vf_safe_debugfs_list),
 				 root, minor);
 
+	if (!IS_SRIOV_VF(xe))
+		drm_debugfs_create_files(pf_only_debugfs_list,
+					 ARRAY_SIZE(pf_only_debugfs_list),
+					 root, minor);
+
 	xe_uc_debugfs_register(&gt->uc, root);
 
 	if (IS_SRIOV_PF(xe))
diff --git a/drivers/gpu/drm/xe/xe_gt_freq.c b/drivers/gpu/drm/xe/xe_gt_freq.c
index 604bdc7c8173..868a5d2c1a52 100644
--- a/drivers/gpu/drm/xe/xe_gt_freq.c
+++ b/drivers/gpu/drm/xe/xe_gt_freq.c
@@ -56,9 +56,10 @@ dev_to_xe(struct device *dev)
 	return gt_to_xe(kobj_to_gt(dev->kobj.parent));
 }
 
-static ssize_t act_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t act_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 
@@ -68,11 +69,12 @@ static ssize_t act_freq_show(struct device *dev,
 
 	return sysfs_emit(buf, "%d\n", freq);
 }
-static DEVICE_ATTR_RO(act_freq);
+static struct kobj_attribute attr_act_freq = __ATTR_RO(act_freq);
 
-static ssize_t cur_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t cur_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 	ssize_t ret;
@@ -85,11 +87,12 @@ static ssize_t cur_freq_show(struct device *dev,
 
 	return sysfs_emit(buf, "%d\n", freq);
 }
-static DEVICE_ATTR_RO(cur_freq);
+static struct kobj_attribute attr_cur_freq = __ATTR_RO(cur_freq);
 
-static ssize_t rp0_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t rp0_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 
@@ -99,11 +102,12 @@ static ssize_t rp0_freq_show(struct device *dev,
 
 	return sysfs_emit(buf, "%d\n", freq);
 }
-static DEVICE_ATTR_RO(rp0_freq);
+static struct kobj_attribute attr_rp0_freq = __ATTR_RO(rp0_freq);
 
-static ssize_t rpe_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t rpe_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 
@@ -113,11 +117,12 @@ static ssize_t rpe_freq_show(struct device *dev,
 
 	return sysfs_emit(buf, "%d\n", freq);
 }
-static DEVICE_ATTR_RO(rpe_freq);
+static struct kobj_attribute attr_rpe_freq = __ATTR_RO(rpe_freq);
 
-static ssize_t rpa_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t rpa_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 
@@ -127,20 +132,22 @@ static ssize_t rpa_freq_show(struct device *dev,
 
 	return sysfs_emit(buf, "%d\n", freq);
 }
-static DEVICE_ATTR_RO(rpa_freq);
+static struct kobj_attribute attr_rpa_freq = __ATTR_RO(rpa_freq);
 
-static ssize_t rpn_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t rpn_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 
 	return sysfs_emit(buf, "%d\n", xe_guc_pc_get_rpn_freq(pc));
 }
-static DEVICE_ATTR_RO(rpn_freq);
+static struct kobj_attribute attr_rpn_freq = __ATTR_RO(rpn_freq);
 
-static ssize_t min_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t min_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 	ssize_t ret;
@@ -154,9 +161,10 @@ static ssize_t min_freq_show(struct device *dev,
 	return sysfs_emit(buf, "%d\n", freq);
 }
 
-static ssize_t min_freq_store(struct device *dev, struct device_attribute *attr,
-			      const char *buff, size_t count)
+static ssize_t min_freq_store(struct kobject *kobj,
+			      struct kobj_attribute *attr, const char *buff, size_t count)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 	ssize_t ret;
@@ -173,11 +181,12 @@ static ssize_t min_freq_store(struct device *dev, struct device_attribute *attr,
 
 	return count;
 }
-static DEVICE_ATTR_RW(min_freq);
+static struct kobj_attribute attr_min_freq = __ATTR_RW(min_freq);
 
-static ssize_t max_freq_show(struct device *dev,
-			     struct device_attribute *attr, char *buf)
+static ssize_t max_freq_show(struct kobject *kobj,
+			     struct kobj_attribute *attr, char *buf)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 	ssize_t ret;
@@ -191,9 +200,10 @@ static ssize_t max_freq_show(struct device *dev,
 	return sysfs_emit(buf, "%d\n", freq);
 }
 
-static ssize_t max_freq_store(struct device *dev, struct device_attribute *attr,
-			      const char *buff, size_t count)
+static ssize_t max_freq_store(struct kobject *kobj,
+			      struct kobj_attribute *attr, const char *buff, size_t count)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	u32 freq;
 	ssize_t ret;
@@ -210,17 +220,17 @@ static ssize_t max_freq_store(struct device *dev, struct device_attribute *attr,
 
 	return count;
 }
-static DEVICE_ATTR_RW(max_freq);
+static struct kobj_attribute attr_max_freq = __ATTR_RW(max_freq);
 
 static const struct attribute *freq_attrs[] = {
-	&dev_attr_act_freq.attr,
-	&dev_attr_cur_freq.attr,
-	&dev_attr_rp0_freq.attr,
-	&dev_attr_rpa_freq.attr,
-	&dev_attr_rpe_freq.attr,
-	&dev_attr_rpn_freq.attr,
-	&dev_attr_min_freq.attr,
-	&dev_attr_max_freq.attr,
+	&attr_act_freq.attr,
+	&attr_cur_freq.attr,
+	&attr_rp0_freq.attr,
+	&attr_rpa_freq.attr,
+	&attr_rpe_freq.attr,
+	&attr_rpn_freq.attr,
+	&attr_min_freq.attr,
+	&attr_max_freq.attr,
 	NULL
 };
 
diff --git a/drivers/gpu/drm/xe/xe_gt_idle.c b/drivers/gpu/drm/xe/xe_gt_idle.c
index fbbace7b0b12..c11206410a4d 100644
--- a/drivers/gpu/drm/xe/xe_gt_idle.c
+++ b/drivers/gpu/drm/xe/xe_gt_idle.c
@@ -249,9 +249,10 @@ int xe_gt_idle_pg_print(struct xe_gt *gt, struct drm_printer *p)
 	return 0;
 }
 
-static ssize_t name_show(struct device *dev,
-			 struct device_attribute *attr, char *buff)
+static ssize_t name_show(struct kobject *kobj,
+			 struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt_idle *gtidle = dev_to_gtidle(dev);
 	struct xe_guc_pc *pc = gtidle_to_pc(gtidle);
 	ssize_t ret;
@@ -262,11 +263,12 @@ static ssize_t name_show(struct device *dev,
 
 	return ret;
 }
-static DEVICE_ATTR_RO(name);
+static struct kobj_attribute name_attr = __ATTR_RO(name);
 
-static ssize_t idle_status_show(struct device *dev,
-				struct device_attribute *attr, char *buff)
+static ssize_t idle_status_show(struct kobject *kobj,
+				struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt_idle *gtidle = dev_to_gtidle(dev);
 	struct xe_guc_pc *pc = gtidle_to_pc(gtidle);
 	enum xe_gt_idle_state state;
@@ -277,6 +279,7 @@ static ssize_t idle_status_show(struct device *dev,
 
 	return sysfs_emit(buff, "%s\n", gt_idle_state_to_string(state));
 }
+static struct kobj_attribute idle_status_attr = __ATTR_RO(idle_status);
 
 u64 xe_gt_idle_residency_msec(struct xe_gt_idle *gtidle)
 {
@@ -291,10 +294,11 @@ u64 xe_gt_idle_residency_msec(struct xe_gt_idle *gtidle)
 	return residency;
 }
 
-static DEVICE_ATTR_RO(idle_status);
-static ssize_t idle_residency_ms_show(struct device *dev,
-				      struct device_attribute *attr, char *buff)
+
+static ssize_t idle_residency_ms_show(struct kobject *kobj,
+				      struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt_idle *gtidle = dev_to_gtidle(dev);
 	struct xe_guc_pc *pc = gtidle_to_pc(gtidle);
 	u64 residency;
@@ -305,12 +309,12 @@ static ssize_t idle_residency_ms_show(struct device *dev,
 
 	return sysfs_emit(buff, "%llu\n", residency);
 }
-static DEVICE_ATTR_RO(idle_residency_ms);
+static struct kobj_attribute idle_residency_attr = __ATTR_RO(idle_residency_ms);
 
 static const struct attribute *gt_idle_attrs[] = {
-	&dev_attr_name.attr,
-	&dev_attr_idle_status.attr,
-	&dev_attr_idle_residency_ms.attr,
+	&name_attr.attr,
+	&idle_status_attr.attr,
+	&idle_residency_attr.attr,
 	NULL,
 };
 
diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c
index 605aad3554e7..d4d9730f0d2c 100644
--- a/drivers/gpu/drm/xe/xe_gt_mcr.c
+++ b/drivers/gpu/drm/xe/xe_gt_mcr.c
@@ -345,7 +345,8 @@ fallback:
 	 * Some older platforms don't have tables or don't have complete tables.
 	 * Newer platforms should always have the required info.
 	 */
-	if (GRAPHICS_VERx100(gt_to_xe(gt)) >= 2000)
+	if (GRAPHICS_VERx100(gt_to_xe(gt)) >= 2000 &&
+	    !gt_to_xe(gt)->info.force_execlist)
 		xe_gt_err(gt, "Slice/Subslice counts missing from hwconfig table; using typical fallback values\n");
 
 	if (gt_to_xe(gt)->info.platform == XE_PVC)
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index c5ad9a0a89c2..10622ca471a2 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -240,7 +240,7 @@ static int handle_pagefault(struct xe_gt *gt, struct pagefault *pf)
 	atomic = access_is_atomic(pf->access_type);
 
 	if (xe_vma_is_cpu_addr_mirror(vma))
-		err = xe_svm_handle_pagefault(vm, vma, gt_to_tile(gt),
+		err = xe_svm_handle_pagefault(vm, vma, gt,
 					      pf->page_addr, atomic);
 	else
 		err = handle_vma_pagefault(gt, vma, atomic);
@@ -435,9 +435,16 @@ static int xe_alloc_pf_queue(struct xe_gt *gt, struct pf_queue *pf_queue)
 	num_eus = bitmap_weight(gt->fuse_topo.eu_mask_per_dss,
 				XE_MAX_EU_FUSE_BITS) * num_dss;
 
-	/* user can issue separate page faults per EU and per CS */
+	/*
+	 * user can issue separate page faults per EU and per CS
+	 *
+	 * XXX: Multiplier required as compute UMD are getting PF queue errors
+	 * without it. Follow on why this multiplier is required.
+	 */
+#define PF_MULTIPLIER	8
 	pf_queue->num_dw =
-		(num_eus + XE_NUM_HW_ENGINES) * PF_MSG_LEN_DW;
+		(num_eus + XE_NUM_HW_ENGINES) * PF_MSG_LEN_DW * PF_MULTIPLIER;
+#undef PF_MULTIPLIER
 
 	pf_queue->gt = gt;
 	pf_queue->data = devm_kcalloc(xe->drm.dev, pf_queue->num_dw,
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index 10be109bf357..2420a548cacc 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -1444,15 +1444,23 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size)
 		return 0;
 
 	xe_gt_assert(gt, pf_get_lmem_alignment(gt) == SZ_2M);
-	bo = xe_bo_create_pin_map(xe, tile, NULL,
-				  ALIGN(size, PAGE_SIZE),
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				  XE_BO_FLAG_NEEDS_2M |
-				  XE_BO_FLAG_PINNED);
+	bo = xe_bo_create_locked(xe, tile, NULL,
+				 ALIGN(size, PAGE_SIZE),
+				 ttm_bo_type_kernel,
+				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+				 XE_BO_FLAG_NEEDS_2M |
+				 XE_BO_FLAG_PINNED |
+				 XE_BO_FLAG_PINNED_LATE_RESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
+	err = xe_bo_pin(bo);
+	xe_bo_unlock(bo);
+	if (unlikely(err)) {
+		xe_bo_put(bo);
+		return err;
+	}
+
 	config->lmem_obj = bo;
 
 	if (xe_device_has_lmtt(xe)) {
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c
index b2521dd6ec42..0fe47f41b63c 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c
@@ -51,27 +51,18 @@ static unsigned int extract_vfid(struct dentry *d)
  *      /sys/kernel/debug/dri/0/
  *      ├── gt0
  *      │   ├── pf
- *      │   │   ├── ggtt_available
- *      │   │   ├── ggtt_provisioned
  *      │   │   ├── contexts_provisioned
  *      │   │   ├── doorbells_provisioned
  *      │   │   ├── runtime_registers
  *      │   │   ├── negotiated_versions
  *      │   │   ├── adverse_events
+ *      ├── gt1
+ *      │   ├── pf
+ *      │   │   ├── ...
  */
 
 static const struct drm_info_list pf_info[] = {
 	{
-		"ggtt_available",
-		.show = xe_gt_debugfs_simple_show,
-		.data = xe_gt_sriov_pf_config_print_available_ggtt,
-	},
-	{
-		"ggtt_provisioned",
-		.show = xe_gt_debugfs_simple_show,
-		.data = xe_gt_sriov_pf_config_print_ggtt,
-	},
-	{
 		"contexts_provisioned",
 		.show = xe_gt_debugfs_simple_show,
 		.data = xe_gt_sriov_pf_config_print_ctxs,
@@ -82,11 +73,6 @@ static const struct drm_info_list pf_info[] = {
 		.data = xe_gt_sriov_pf_config_print_dbs,
 	},
 	{
-		"lmem_provisioned",
-		.show = xe_gt_debugfs_simple_show,
-		.data = xe_gt_sriov_pf_config_print_lmem,
-	},
-	{
 		"runtime_registers",
 		.show = xe_gt_debugfs_simple_show,
 		.data = xe_gt_sriov_pf_service_print_runtime,
@@ -107,6 +93,42 @@ static const struct drm_info_list pf_info[] = {
  *      /sys/kernel/debug/dri/0/
  *      ├── gt0
  *      │   ├── pf
+ *      │   │   ├── ggtt_available
+ *      │   │   ├── ggtt_provisioned
+ */
+
+static const struct drm_info_list pf_ggtt_info[] = {
+	{
+		"ggtt_available",
+		.show = xe_gt_debugfs_simple_show,
+		.data = xe_gt_sriov_pf_config_print_available_ggtt,
+	},
+	{
+		"ggtt_provisioned",
+		.show = xe_gt_debugfs_simple_show,
+		.data = xe_gt_sriov_pf_config_print_ggtt,
+	},
+};
+
+/*
+ *      /sys/kernel/debug/dri/0/
+ *      ├── gt0
+ *      │   ├── pf
+ *      │   │   ├── lmem_provisioned
+ */
+
+static const struct drm_info_list pf_lmem_info[] = {
+	{
+		"lmem_provisioned",
+		.show = xe_gt_debugfs_simple_show,
+		.data = xe_gt_sriov_pf_config_print_lmem,
+	},
+};
+
+/*
+ *      /sys/kernel/debug/dri/0/
+ *      ├── gt0
+ *      │   ├── pf
  *      │   │   ├── reset_engine
  *      │   │   ├── sample_period
  *      │   │   ├── sched_if_idle
@@ -532,6 +554,16 @@ void xe_gt_sriov_pf_debugfs_register(struct xe_gt *gt, struct dentry *root)
 	pfdentry->d_inode->i_private = gt;
 
 	drm_debugfs_create_files(pf_info, ARRAY_SIZE(pf_info), pfdentry, minor);
+	if (!xe_gt_is_media_type(gt)) {
+		drm_debugfs_create_files(pf_ggtt_info,
+					 ARRAY_SIZE(pf_ggtt_info),
+					 pfdentry, minor);
+		if (IS_DGFX(gt_to_xe(gt)))
+			drm_debugfs_create_files(pf_lmem_info,
+						 ARRAY_SIZE(pf_lmem_info),
+						 pfdentry, minor);
+	}
+
 	pf_add_policy_attrs(gt, pfdentry);
 	pf_add_config_attrs(gt, pfdentry, PFID);
 
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c
index 4efde5f46b43..821cfcc34e6b 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c
@@ -112,7 +112,6 @@ static const struct xe_reg tgl_runtime_regs[] = {
 	XELP_GT_SLICE_ENABLE,		/* _MMIO(0x9138) */
 	XELP_GT_GEOMETRY_DSS_ENABLE,	/* _MMIO(0x913c) */
 	GT_VEBOX_VDBOX_DISABLE,		/* _MMIO(0x9140) */
-	CTC_MODE,			/* _MMIO(0xa26c) */
 	HUC_KERNEL_LOAD_INFO,		/* _MMIO(0xc1dc) */
 };
 
@@ -124,7 +123,6 @@ static const struct xe_reg ats_m_runtime_regs[] = {
 	XELP_GT_GEOMETRY_DSS_ENABLE,	/* _MMIO(0x913c) */
 	GT_VEBOX_VDBOX_DISABLE,		/* _MMIO(0x9140) */
 	XEHP_GT_COMPUTE_DSS_ENABLE,	/* _MMIO(0x9144) */
-	CTC_MODE,			/* _MMIO(0xa26c) */
 	HUC_KERNEL_LOAD_INFO,		/* _MMIO(0xc1dc) */
 };
 
@@ -136,7 +134,6 @@ static const struct xe_reg pvc_runtime_regs[] = {
 	GT_VEBOX_VDBOX_DISABLE,		/* _MMIO(0x9140) */
 	XEHP_GT_COMPUTE_DSS_ENABLE,	/* _MMIO(0x9144) */
 	XEHPC_GT_COMPUTE_DSS_ENABLE_EXT,/* _MMIO(0x9148) */
-	CTC_MODE,			/* _MMIO(0xA26C) */
 	HUC_KERNEL_LOAD_INFO,		/* _MMIO(0xc1dc) */
 };
 
@@ -150,7 +147,6 @@ static const struct xe_reg ver_1270_runtime_regs[] = {
 	GT_VEBOX_VDBOX_DISABLE,		/* _MMIO(0x9140) */
 	XEHP_GT_COMPUTE_DSS_ENABLE,	/* _MMIO(0x9144) */
 	XEHPC_GT_COMPUTE_DSS_ENABLE_EXT,/* _MMIO(0x9148) */
-	CTC_MODE,			/* _MMIO(0xa26c) */
 	HUC_KERNEL_LOAD_INFO,		/* _MMIO(0xc1dc) */
 };
 
@@ -167,7 +163,6 @@ static const struct xe_reg ver_2000_runtime_regs[] = {
 	XE2_GT_COMPUTE_DSS_2,		/* _MMIO(0x914c) */
 	XE2_GT_GEOMETRY_DSS_1,		/* _MMIO(0x9150) */
 	XE2_GT_GEOMETRY_DSS_2,		/* _MMIO(0x9154) */
-	CTC_MODE,			/* _MMIO(0xa26c) */
 	HUC_KERNEL_LOAD_INFO,		/* _MMIO(0xc1dc) */
 };
 
@@ -185,7 +180,6 @@ static const struct xe_reg ver_3000_runtime_regs[] = {
 	XE2_GT_COMPUTE_DSS_2,		/* _MMIO(0x914c) */
 	XE2_GT_GEOMETRY_DSS_1,		/* _MMIO(0x9150) */
 	XE2_GT_GEOMETRY_DSS_2,		/* _MMIO(0x9154) */
-	CTC_MODE,			/* _MMIO(0xa26c) */
 	HUC_KERNEL_LOAD_INFO,		/* _MMIO(0xc1dc) */
 };
 
diff --git a/drivers/gpu/drm/xe/xe_gt_stats.c b/drivers/gpu/drm/xe/xe_gt_stats.c
index 6155ea354432..30f942671c2b 100644
--- a/drivers/gpu/drm/xe/xe_gt_stats.c
+++ b/drivers/gpu/drm/xe/xe_gt_stats.c
@@ -27,6 +27,7 @@ void xe_gt_stats_incr(struct xe_gt *gt, const enum xe_gt_stats_id id, int incr)
 }
 
 static const char *const stat_description[__XE_GT_STATS_NUM_IDS] = {
+	"svm_pagefault_count",
 	"tlb_inval_count",
 	"vma_pagefault_count",
 	"vma_pagefault_kb",
diff --git a/drivers/gpu/drm/xe/xe_gt_stats_types.h b/drivers/gpu/drm/xe/xe_gt_stats_types.h
index d556771f99d6..be3244d7133c 100644
--- a/drivers/gpu/drm/xe/xe_gt_stats_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_stats_types.h
@@ -7,6 +7,7 @@
 #define _XE_GT_STATS_TYPES_H_
 
 enum xe_gt_stats_id {
+	XE_GT_STATS_ID_SVM_PAGEFAULT_COUNT,
 	XE_GT_STATS_ID_TLB_INVAL,
 	XE_GT_STATS_ID_VMA_PAGEFAULT_COUNT,
 	XE_GT_STATS_ID_VMA_PAGEFAULT_KB,
diff --git a/drivers/gpu/drm/xe/xe_gt_throttle.c b/drivers/gpu/drm/xe/xe_gt_throttle.c
index 8db78d616b6f..aa962c783cdf 100644
--- a/drivers/gpu/drm/xe/xe_gt_throttle.c
+++ b/drivers/gpu/drm/xe/xe_gt_throttle.c
@@ -114,115 +114,115 @@ static u32 read_reason_vr_tdc(struct xe_gt *gt)
 	return tdc;
 }
 
-static ssize_t status_show(struct device *dev,
-			   struct device_attribute *attr,
-			   char *buff)
+static ssize_t status_show(struct kobject *kobj,
+			   struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool status = !!read_status(gt);
 
 	return sysfs_emit(buff, "%u\n", status);
 }
-static DEVICE_ATTR_RO(status);
+static struct kobj_attribute attr_status = __ATTR_RO(status);
 
-static ssize_t reason_pl1_show(struct device *dev,
-			       struct device_attribute *attr,
-			       char *buff)
+static ssize_t reason_pl1_show(struct kobject *kobj,
+			       struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool pl1 = !!read_reason_pl1(gt);
 
 	return sysfs_emit(buff, "%u\n", pl1);
 }
-static DEVICE_ATTR_RO(reason_pl1);
+static struct kobj_attribute attr_reason_pl1 = __ATTR_RO(reason_pl1);
 
-static ssize_t reason_pl2_show(struct device *dev,
-			       struct device_attribute *attr,
-			       char *buff)
+static ssize_t reason_pl2_show(struct kobject *kobj,
+			       struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool pl2 = !!read_reason_pl2(gt);
 
 	return sysfs_emit(buff, "%u\n", pl2);
 }
-static DEVICE_ATTR_RO(reason_pl2);
+static struct kobj_attribute attr_reason_pl2 = __ATTR_RO(reason_pl2);
 
-static ssize_t reason_pl4_show(struct device *dev,
-			       struct device_attribute *attr,
-			       char *buff)
+static ssize_t reason_pl4_show(struct kobject *kobj,
+			       struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool pl4 = !!read_reason_pl4(gt);
 
 	return sysfs_emit(buff, "%u\n", pl4);
 }
-static DEVICE_ATTR_RO(reason_pl4);
+static struct kobj_attribute attr_reason_pl4 = __ATTR_RO(reason_pl4);
 
-static ssize_t reason_thermal_show(struct device *dev,
-				   struct device_attribute *attr,
-				   char *buff)
+static ssize_t reason_thermal_show(struct kobject *kobj,
+				   struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool thermal = !!read_reason_thermal(gt);
 
 	return sysfs_emit(buff, "%u\n", thermal);
 }
-static DEVICE_ATTR_RO(reason_thermal);
+static struct kobj_attribute attr_reason_thermal = __ATTR_RO(reason_thermal);
 
-static ssize_t reason_prochot_show(struct device *dev,
-				   struct device_attribute *attr,
-				   char *buff)
+static ssize_t reason_prochot_show(struct kobject *kobj,
+				   struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool prochot = !!read_reason_prochot(gt);
 
 	return sysfs_emit(buff, "%u\n", prochot);
 }
-static DEVICE_ATTR_RO(reason_prochot);
+static struct kobj_attribute attr_reason_prochot = __ATTR_RO(reason_prochot);
 
-static ssize_t reason_ratl_show(struct device *dev,
-				struct device_attribute *attr,
-				char *buff)
+static ssize_t reason_ratl_show(struct kobject *kobj,
+				struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool ratl = !!read_reason_ratl(gt);
 
 	return sysfs_emit(buff, "%u\n", ratl);
 }
-static DEVICE_ATTR_RO(reason_ratl);
+static struct kobj_attribute attr_reason_ratl = __ATTR_RO(reason_ratl);
 
-static ssize_t reason_vr_thermalert_show(struct device *dev,
-					 struct device_attribute *attr,
-					 char *buff)
+static ssize_t reason_vr_thermalert_show(struct kobject *kobj,
+					 struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool thermalert = !!read_reason_vr_thermalert(gt);
 
 	return sysfs_emit(buff, "%u\n", thermalert);
 }
-static DEVICE_ATTR_RO(reason_vr_thermalert);
+static struct kobj_attribute attr_reason_vr_thermalert = __ATTR_RO(reason_vr_thermalert);
 
-static ssize_t reason_vr_tdc_show(struct device *dev,
-				  struct device_attribute *attr,
-				  char *buff)
+static ssize_t reason_vr_tdc_show(struct kobject *kobj,
+				  struct kobj_attribute *attr, char *buff)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_gt *gt = dev_to_gt(dev);
 	bool tdc = !!read_reason_vr_tdc(gt);
 
 	return sysfs_emit(buff, "%u\n", tdc);
 }
-static DEVICE_ATTR_RO(reason_vr_tdc);
+static struct kobj_attribute attr_reason_vr_tdc = __ATTR_RO(reason_vr_tdc);
 
 static struct attribute *throttle_attrs[] = {
-	&dev_attr_status.attr,
-	&dev_attr_reason_pl1.attr,
-	&dev_attr_reason_pl2.attr,
-	&dev_attr_reason_pl4.attr,
-	&dev_attr_reason_thermal.attr,
-	&dev_attr_reason_prochot.attr,
-	&dev_attr_reason_ratl.attr,
-	&dev_attr_reason_vr_thermalert.attr,
-	&dev_attr_reason_vr_tdc.attr,
+	&attr_status.attr,
+	&attr_reason_pl1.attr,
+	&attr_reason_pl2.attr,
+	&attr_reason_pl4.attr,
+	&attr_reason_thermal.attr,
+	&attr_reason_prochot.attr,
+	&attr_reason_ratl.attr,
+	&attr_reason_vr_thermalert.attr,
+	&attr_reason_vr_tdc.attr,
 	NULL
 };
 
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
index 03072e094991..084cbdeba8ea 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
@@ -322,6 +322,13 @@ int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt)
 	return 0;
 }
 
+/*
+ * Ensure that roundup_pow_of_two(length) doesn't overflow.
+ * Note that roundup_pow_of_two() operates on unsigned long,
+ * not on u64.
+ */
+#define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX))
+
 /**
  * xe_gt_tlb_invalidation_range - Issue a TLB invalidation on this GT for an
  * address range
@@ -346,6 +353,7 @@ int xe_gt_tlb_invalidation_range(struct xe_gt *gt,
 	struct xe_device *xe = gt_to_xe(gt);
 #define MAX_TLB_INVALIDATION_LEN	7
 	u32 action[MAX_TLB_INVALIDATION_LEN];
+	u64 length = end - start;
 	int len = 0;
 
 	xe_gt_assert(gt, fence);
@@ -358,11 +366,11 @@ int xe_gt_tlb_invalidation_range(struct xe_gt *gt,
 
 	action[len++] = XE_GUC_ACTION_TLB_INVALIDATION;
 	action[len++] = 0; /* seqno, replaced in send_tlb_invalidation */
-	if (!xe->info.has_range_tlb_invalidation) {
+	if (!xe->info.has_range_tlb_invalidation ||
+	    length > MAX_RANGE_TLB_INVALIDATION_LENGTH) {
 		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL);
 	} else {
 		u64 orig_start = start;
-		u64 length = end - start;
 		u64 align;
 
 		if (length < SZ_4K)
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index bc5714a5b36b..bac5471a1a78 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -483,7 +483,8 @@ static int guc_g2g_alloc(struct xe_guc *guc)
 					  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
 					  XE_BO_FLAG_GGTT |
 					  XE_BO_FLAG_GGTT_ALL |
-					  XE_BO_FLAG_GGTT_INVALIDATE);
+					  XE_BO_FLAG_GGTT_INVALIDATE |
+					  XE_BO_FLAG_PINNED_NORESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
@@ -1393,6 +1394,7 @@ proto:
 	/* Use data from the GuC response as our return value */
 	return FIELD_GET(GUC_HXG_RESPONSE_MSG_0_DATA0, header);
 }
+ALLOW_ERROR_INJECTION(xe_guc_mmio_send_recv, ERRNO);
 
 int xe_guc_mmio_send(struct xe_guc *guc, const u32 *request, u32 len)
 {
@@ -1508,30 +1510,32 @@ void xe_guc_print_info(struct xe_guc *guc, struct drm_printer *p)
 
 	xe_uc_fw_print(&guc->fw, p);
 
-	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
-	if (!fw_ref)
-		return;
+	if (!IS_SRIOV_VF(gt_to_xe(gt))) {
+		fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+		if (!fw_ref)
+			return;
+
+		status = xe_mmio_read32(&gt->mmio, GUC_STATUS);
+
+		drm_printf(p, "\nGuC status 0x%08x:\n", status);
+		drm_printf(p, "\tBootrom status = 0x%x\n",
+			   REG_FIELD_GET(GS_BOOTROM_MASK, status));
+		drm_printf(p, "\tuKernel status = 0x%x\n",
+			   REG_FIELD_GET(GS_UKERNEL_MASK, status));
+		drm_printf(p, "\tMIA Core status = 0x%x\n",
+			   REG_FIELD_GET(GS_MIA_MASK, status));
+		drm_printf(p, "\tLog level = %d\n",
+			   xe_guc_log_get_level(&guc->log));
+
+		drm_puts(p, "\nScratch registers:\n");
+		for (i = 0; i < SOFT_SCRATCH_COUNT; i++) {
+			drm_printf(p, "\t%2d: \t0x%x\n",
+				   i, xe_mmio_read32(&gt->mmio, SOFT_SCRATCH(i)));
+		}
 
-	status = xe_mmio_read32(&gt->mmio, GUC_STATUS);
-
-	drm_printf(p, "\nGuC status 0x%08x:\n", status);
-	drm_printf(p, "\tBootrom status = 0x%x\n",
-		   REG_FIELD_GET(GS_BOOTROM_MASK, status));
-	drm_printf(p, "\tuKernel status = 0x%x\n",
-		   REG_FIELD_GET(GS_UKERNEL_MASK, status));
-	drm_printf(p, "\tMIA Core status = 0x%x\n",
-		   REG_FIELD_GET(GS_MIA_MASK, status));
-	drm_printf(p, "\tLog level = %d\n",
-		   xe_guc_log_get_level(&guc->log));
-
-	drm_puts(p, "\nScratch registers:\n");
-	for (i = 0; i < SOFT_SCRATCH_COUNT; i++) {
-		drm_printf(p, "\t%2d: \t0x%x\n",
-			   i, xe_mmio_read32(&gt->mmio, SOFT_SCRATCH(i)));
+		xe_force_wake_put(gt_to_fw(gt), fw_ref);
 	}
 
-	xe_force_wake_put(gt_to_fw(gt), fw_ref);
-
 	drm_puts(p, "\n");
 	xe_guc_ct_print(&guc->ct, p, false);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
index e7c9e095a19f..44c1fa2fe7c8 100644
--- a/drivers/gpu/drm/xe/xe_guc_ads.c
+++ b/drivers/gpu/drm/xe/xe_guc_ads.c
@@ -376,6 +376,11 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 					GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET,
 					&offset, &remain);
 
+	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_WA(gt, 16026508708))
+		guc_waklv_enable_simple(ads,
+					GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH,
+					&offset, &remain);
+
 	size = guc_ads_waklv_size(ads) - remain;
 	if (!size)
 		return;
@@ -414,7 +419,8 @@ int xe_guc_ads_init(struct xe_guc_ads *ads)
 	bo = xe_managed_bo_create_pin_map(xe, tile, guc_ads_size(ads) + MAX_GOLDEN_LRC_SIZE,
 					  XE_BO_FLAG_SYSTEM |
 					  XE_BO_FLAG_GGTT |
-					  XE_BO_FLAG_GGTT_INVALIDATE);
+					  XE_BO_FLAG_GGTT_INVALIDATE |
+					  XE_BO_FLAG_PINNED_NORESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
@@ -490,24 +496,52 @@ static void fill_engine_enable_masks(struct xe_gt *gt,
 		       engine_enable_mask(gt, XE_ENGINE_CLASS_OTHER));
 }
 
-static void guc_prep_golden_lrc_null(struct xe_guc_ads *ads)
+/*
+ * Write the offsets corresponding to the golden LRCs. The actual data is
+ * populated later by guc_golden_lrc_populate()
+ */
+static void guc_golden_lrc_init(struct xe_guc_ads *ads)
 {
 	struct xe_device *xe = ads_to_xe(ads);
+	struct xe_gt *gt = ads_to_gt(ads);
 	struct iosys_map info_map = IOSYS_MAP_INIT_OFFSET(ads_to_map(ads),
 			offsetof(struct __guc_ads_blob, system_info));
-	u8 guc_class;
+	size_t alloc_size, real_size;
+	u32 addr_ggtt, offset;
+	int class;
+
+	offset = guc_ads_golden_lrc_offset(ads);
+	addr_ggtt = xe_bo_ggtt_addr(ads->bo) + offset;
+
+	for (class = 0; class < XE_ENGINE_CLASS_MAX; ++class) {
+		u8 guc_class;
+
+		guc_class = xe_engine_class_to_guc_class(class);
 
-	for (guc_class = 0; guc_class <= GUC_MAX_ENGINE_CLASSES; ++guc_class) {
 		if (!info_map_read(xe, &info_map,
 				   engine_enabled_masks[guc_class]))
 			continue;
 
+		real_size = xe_gt_lrc_size(gt, class);
+		alloc_size = PAGE_ALIGN(real_size);
+
+		/*
+		 * This interface is slightly confusing. We need to pass the
+		 * base address of the full golden context and the size of just
+		 * the engine state, which is the section of the context image
+		 * that starts after the execlists LRC registers. This is
+		 * required to allow the GuC to restore just the engine state
+		 * when a watchdog reset occurs.
+		 * We calculate the engine state size by removing the size of
+		 * what comes before it in the context image (which is identical
+		 * on all engines).
+		 */
 		ads_blob_write(ads, ads.eng_state_size[guc_class],
-			       guc_ads_golden_lrc_size(ads) -
-			       xe_lrc_skip_size(xe));
+			       real_size - xe_lrc_skip_size(xe));
 		ads_blob_write(ads, ads.golden_context_lrca[guc_class],
-			       xe_bo_ggtt_addr(ads->bo) +
-			       guc_ads_golden_lrc_offset(ads));
+			       addr_ggtt);
+
+		addr_ggtt += alloc_size;
 	}
 }
 
@@ -682,8 +716,8 @@ static int guc_capture_prep_lists(struct xe_guc_ads *ads)
 	}
 
 	if (ads->capture_size != PAGE_ALIGN(total_size))
-		xe_gt_dbg(gt, "ADS capture alloc size changed from %d to %d\n",
-			  ads->capture_size, PAGE_ALIGN(total_size));
+		xe_gt_dbg(gt, "Updated ADS capture size %d (was %d)\n",
+			  PAGE_ALIGN(total_size), ads->capture_size);
 	return PAGE_ALIGN(total_size);
 }
 
@@ -857,7 +891,7 @@ void xe_guc_ads_populate_minimal(struct xe_guc_ads *ads)
 
 	xe_map_memset(ads_to_xe(ads), ads_to_map(ads), 0, 0, ads->bo->size);
 	guc_policies_init(ads);
-	guc_prep_golden_lrc_null(ads);
+	guc_golden_lrc_init(ads);
 	guc_mapping_table_init_invalid(gt, &info_map);
 	guc_doorbell_init(ads);
 
@@ -883,7 +917,7 @@ void xe_guc_ads_populate(struct xe_guc_ads *ads)
 	guc_policies_init(ads);
 	fill_engine_enable_masks(gt, &info_map);
 	guc_mmio_reg_state_init(ads);
-	guc_prep_golden_lrc_null(ads);
+	guc_golden_lrc_init(ads);
 	guc_mapping_table_init(gt, &info_map);
 	guc_capture_prep_lists(ads);
 	guc_doorbell_init(ads);
@@ -903,18 +937,22 @@ void xe_guc_ads_populate(struct xe_guc_ads *ads)
 		       guc_ads_private_data_offset(ads));
 }
 
-static void guc_populate_golden_lrc(struct xe_guc_ads *ads)
+/*
+ * After the golden LRC's are recorded for each engine class by the first
+ * submission, copy them to the ADS, as initialized earlier by
+ * guc_golden_lrc_init().
+ */
+static void guc_golden_lrc_populate(struct xe_guc_ads *ads)
 {
 	struct xe_device *xe = ads_to_xe(ads);
 	struct xe_gt *gt = ads_to_gt(ads);
 	struct iosys_map info_map = IOSYS_MAP_INIT_OFFSET(ads_to_map(ads),
 			offsetof(struct __guc_ads_blob, system_info));
 	size_t total_size = 0, alloc_size, real_size;
-	u32 addr_ggtt, offset;
+	u32 offset;
 	int class;
 
 	offset = guc_ads_golden_lrc_offset(ads);
-	addr_ggtt = xe_bo_ggtt_addr(ads->bo) + offset;
 
 	for (class = 0; class < XE_ENGINE_CLASS_MAX; ++class) {
 		u8 guc_class;
@@ -931,26 +969,9 @@ static void guc_populate_golden_lrc(struct xe_guc_ads *ads)
 		alloc_size = PAGE_ALIGN(real_size);
 		total_size += alloc_size;
 
-		/*
-		 * This interface is slightly confusing. We need to pass the
-		 * base address of the full golden context and the size of just
-		 * the engine state, which is the section of the context image
-		 * that starts after the execlists LRC registers. This is
-		 * required to allow the GuC to restore just the engine state
-		 * when a watchdog reset occurs.
-		 * We calculate the engine state size by removing the size of
-		 * what comes before it in the context image (which is identical
-		 * on all engines).
-		 */
-		ads_blob_write(ads, ads.eng_state_size[guc_class],
-			       real_size - xe_lrc_skip_size(xe));
-		ads_blob_write(ads, ads.golden_context_lrca[guc_class],
-			       addr_ggtt);
-
 		xe_map_memcpy_to(xe, ads_to_map(ads), offset,
 				 gt->default_lrc[class], real_size);
 
-		addr_ggtt += alloc_size;
 		offset += alloc_size;
 	}
 
@@ -959,7 +980,7 @@ static void guc_populate_golden_lrc(struct xe_guc_ads *ads)
 
 void xe_guc_ads_populate_post_load(struct xe_guc_ads *ads)
 {
-	guc_populate_golden_lrc(ads);
+	guc_golden_lrc_populate(ads);
 }
 
 static int guc_ads_action_update_policies(struct xe_guc_ads *ads, u32 policy_offset)
diff --git a/drivers/gpu/drm/xe/xe_guc_capture.c b/drivers/gpu/drm/xe/xe_guc_capture.c
index f6d523e4c5fe..859a3ba91be5 100644
--- a/drivers/gpu/drm/xe/xe_guc_capture.c
+++ b/drivers/gpu/drm/xe/xe_guc_capture.c
@@ -105,49 +105,49 @@ struct __guc_capture_parsed_output {
  *     3. Incorrect order will trigger XE_WARN.
  */
 #define COMMON_XELP_BASE_GLOBAL \
-	{ FORCEWAKE_GT,			REG_32BIT,	0,	0,	"FORCEWAKE_GT"}
+	{ FORCEWAKE_GT,			REG_32BIT,	0,	0,	0,	"FORCEWAKE_GT"}
 
 #define COMMON_BASE_ENGINE_INSTANCE \
-	{ RING_HWSTAM(0),		REG_32BIT,	0,	0,	"HWSTAM"}, \
-	{ RING_HWS_PGA(0),		REG_32BIT,	0,	0,	"RING_HWS_PGA"}, \
-	{ RING_HEAD(0),			REG_32BIT,	0,	0,	"RING_HEAD"}, \
-	{ RING_TAIL(0),			REG_32BIT,	0,	0,	"RING_TAIL"}, \
-	{ RING_CTL(0),			REG_32BIT,	0,	0,	"RING_CTL"}, \
-	{ RING_MI_MODE(0),		REG_32BIT,	0,	0,	"RING_MI_MODE"}, \
-	{ RING_MODE(0),			REG_32BIT,	0,	0,	"RING_MODE"}, \
-	{ RING_ESR(0),			REG_32BIT,	0,	0,	"RING_ESR"}, \
-	{ RING_EMR(0),			REG_32BIT,	0,	0,	"RING_EMR"}, \
-	{ RING_EIR(0),			REG_32BIT,	0,	0,	"RING_EIR"}, \
-	{ RING_IMR(0),			REG_32BIT,	0,	0,	"RING_IMR"}, \
-	{ RING_IPEHR(0),		REG_32BIT,	0,	0,	"IPEHR"}, \
-	{ RING_INSTDONE(0),		REG_32BIT,	0,	0,	"RING_INSTDONE"}, \
-	{ INDIRECT_RING_STATE(0),	REG_32BIT,	0,	0,	"INDIRECT_RING_STATE"}, \
-	{ RING_ACTHD(0),		REG_64BIT_LOW_DW, 0,	0,	NULL}, \
-	{ RING_ACTHD_UDW(0),		REG_64BIT_HI_DW, 0,	0,	"ACTHD"}, \
-	{ RING_BBADDR(0),		REG_64BIT_LOW_DW, 0,	0,	NULL}, \
-	{ RING_BBADDR_UDW(0),		REG_64BIT_HI_DW, 0,	0,	"RING_BBADDR"}, \
-	{ RING_START(0),		REG_64BIT_LOW_DW, 0,	0,	NULL}, \
-	{ RING_START_UDW(0),		REG_64BIT_HI_DW, 0,	0,	"RING_START"}, \
-	{ RING_DMA_FADD(0),		REG_64BIT_LOW_DW, 0,	0,	NULL}, \
-	{ RING_DMA_FADD_UDW(0),		REG_64BIT_HI_DW, 0,	0,	"RING_DMA_FADD"}, \
-	{ RING_EXECLIST_STATUS_LO(0),	REG_64BIT_LOW_DW, 0,	0,	NULL}, \
-	{ RING_EXECLIST_STATUS_HI(0),	REG_64BIT_HI_DW, 0,	0,	"RING_EXECLIST_STATUS"}, \
-	{ RING_EXECLIST_SQ_CONTENTS_LO(0), REG_64BIT_LOW_DW, 0,	0,	NULL}, \
-	{ RING_EXECLIST_SQ_CONTENTS_HI(0), REG_64BIT_HI_DW, 0,	0,	"RING_EXECLIST_SQ_CONTENTS"}
+	{ RING_HWSTAM(0),		REG_32BIT,	0,	0,	0,	"HWSTAM"}, \
+	{ RING_HWS_PGA(0),		REG_32BIT,	0,	0,	0,	"RING_HWS_PGA"}, \
+	{ RING_HEAD(0),			REG_32BIT,	0,	0,	0,	"RING_HEAD"}, \
+	{ RING_TAIL(0),			REG_32BIT,	0,	0,	0,	"RING_TAIL"}, \
+	{ RING_CTL(0),			REG_32BIT,	0,	0,	0,	"RING_CTL"}, \
+	{ RING_MI_MODE(0),		REG_32BIT,	0,	0,	0,	"RING_MI_MODE"}, \
+	{ RING_MODE(0),			REG_32BIT,	0,	0,	0,	"RING_MODE"}, \
+	{ RING_ESR(0),			REG_32BIT,	0,	0,	0,	"RING_ESR"}, \
+	{ RING_EMR(0),			REG_32BIT,	0,	0,	0,	"RING_EMR"}, \
+	{ RING_EIR(0),			REG_32BIT,	0,	0,	0,	"RING_EIR"}, \
+	{ RING_IMR(0),			REG_32BIT,	0,	0,	0,	"RING_IMR"}, \
+	{ RING_IPEHR(0),		REG_32BIT,	0,	0,	0,	"IPEHR"}, \
+	{ RING_INSTDONE(0),		REG_32BIT,	0,	0,	0,	"RING_INSTDONE"}, \
+	{ INDIRECT_RING_STATE(0),	REG_32BIT,	0,	0,	0,	"INDIRECT_RING_STATE"}, \
+	{ RING_ACTHD(0),		REG_64BIT_LOW_DW, 0,	0,	0,	NULL}, \
+	{ RING_ACTHD_UDW(0),		REG_64BIT_HI_DW, 0,	0,	0,	"ACTHD"}, \
+	{ RING_BBADDR(0),		REG_64BIT_LOW_DW, 0,	0,	0,	NULL}, \
+	{ RING_BBADDR_UDW(0),		REG_64BIT_HI_DW, 0,	0,	0,	"RING_BBADDR"}, \
+	{ RING_START(0),		REG_64BIT_LOW_DW, 0,	0,	0,	NULL}, \
+	{ RING_START_UDW(0),		REG_64BIT_HI_DW, 0,	0,	0,	"RING_START"}, \
+	{ RING_DMA_FADD(0),		REG_64BIT_LOW_DW, 0,	0,	0,	NULL}, \
+	{ RING_DMA_FADD_UDW(0),		REG_64BIT_HI_DW, 0,	0,	0,	"RING_DMA_FADD"}, \
+	{ RING_EXECLIST_STATUS_LO(0),	REG_64BIT_LOW_DW, 0,	0,	0,	NULL}, \
+	{ RING_EXECLIST_STATUS_HI(0),	REG_64BIT_HI_DW, 0,	0,	0,	"RING_EXECLIST_STATUS"}, \
+	{ RING_EXECLIST_SQ_CONTENTS_LO(0), REG_64BIT_LOW_DW, 0,	0,	0,	NULL}, \
+	{ RING_EXECLIST_SQ_CONTENTS_HI(0), REG_64BIT_HI_DW, 0,	0,	0,	"RING_EXECLIST_SQ_CONTENTS"}
 
 #define COMMON_XELP_RC_CLASS \
-	{ RCU_MODE,			REG_32BIT,	0,	0,	"RCU_MODE"}
+	{ RCU_MODE,			REG_32BIT,	0,	0,	0,	"RCU_MODE"}
 
 #define COMMON_XELP_RC_CLASS_INSTDONE \
-	{ SC_INSTDONE,			REG_32BIT,	0,	0,	"SC_INSTDONE"}, \
-	{ SC_INSTDONE_EXTRA,		REG_32BIT,	0,	0,	"SC_INSTDONE_EXTRA"}, \
-	{ SC_INSTDONE_EXTRA2,		REG_32BIT,	0,	0,	"SC_INSTDONE_EXTRA2"}
+	{ SC_INSTDONE,			REG_32BIT,	0,	0,	0,	"SC_INSTDONE"}, \
+	{ SC_INSTDONE_EXTRA,		REG_32BIT,	0,	0,	0,	"SC_INSTDONE_EXTRA"}, \
+	{ SC_INSTDONE_EXTRA2,		REG_32BIT,	0,	0,	0,	"SC_INSTDONE_EXTRA2"}
 
 #define XELP_VEC_CLASS_REGS \
-	{ SFC_DONE(0),			0,	0,	0,	"SFC_DONE[0]"}, \
-	{ SFC_DONE(1),			0,	0,	0,	"SFC_DONE[1]"}, \
-	{ SFC_DONE(2),			0,	0,	0,	"SFC_DONE[2]"}, \
-	{ SFC_DONE(3),			0,	0,	0,	"SFC_DONE[3]"}
+	{ SFC_DONE(0),			0,	0,	0,	0,	"SFC_DONE[0]"}, \
+	{ SFC_DONE(1),			0,	0,	0,	0,	"SFC_DONE[1]"}, \
+	{ SFC_DONE(2),			0,	0,	0,	0,	"SFC_DONE[2]"}, \
+	{ SFC_DONE(3),			0,	0,	0,	0,	"SFC_DONE[3]"}
 
 /* XE_LP Global */
 static const struct __guc_mmio_reg_descr xe_lp_global_regs[] = {
@@ -352,15 +352,16 @@ static const struct __ext_steer_reg xehpg_extregs[] = {
 
 static void __fill_ext_reg(struct __guc_mmio_reg_descr *ext,
 			   const struct __ext_steer_reg *extlist,
-			   int slice_id, int subslice_id)
+			   u32 dss_id, u16 slice_id, u16 subslice_id)
 {
 	if (!ext || !extlist)
 		return;
 
 	ext->reg = XE_REG(extlist->reg.__reg.addr);
 	ext->flags = FIELD_PREP(GUC_REGSET_STEERING_NEEDED, 1);
-	ext->flags = FIELD_PREP(GUC_REGSET_STEERING_GROUP, slice_id);
+	ext->flags |= FIELD_PREP(GUC_REGSET_STEERING_GROUP, slice_id);
 	ext->flags |= FIELD_PREP(GUC_REGSET_STEERING_INSTANCE, subslice_id);
+	ext->dss_id = dss_id;
 	ext->regname = extlist->name;
 }
 
@@ -397,7 +398,7 @@ static void guc_capture_alloc_steered_lists(struct xe_guc *guc)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
 	u16 slice, subslice;
-	int iter, i, total = 0;
+	int dss, i, total = 0;
 	const struct __guc_mmio_reg_descr_group *lists = guc->capture->reglists;
 	const struct __guc_mmio_reg_descr_group *list;
 	struct __guc_mmio_reg_descr_group *extlists;
@@ -454,15 +455,15 @@ static void guc_capture_alloc_steered_lists(struct xe_guc *guc)
 
 	/* For steering registers, the list is generated at run-time */
 	extarray = (struct __guc_mmio_reg_descr *)extlists[0].list;
-	for_each_dss_steering(iter, gt, slice, subslice) {
+	for_each_dss_steering(dss, gt, slice, subslice) {
 		for (i = 0; i < ARRAY_SIZE(xe_extregs); ++i) {
-			__fill_ext_reg(extarray, &xe_extregs[i], slice, subslice);
+			__fill_ext_reg(extarray, &xe_extregs[i], dss, slice, subslice);
 			++extarray;
 		}
 
 		if (has_xehpg_extregs)
 			for (i = 0; i < ARRAY_SIZE(xehpg_extregs); ++i) {
-				__fill_ext_reg(extarray, &xehpg_extregs[i], slice, subslice);
+				__fill_ext_reg(extarray, &xehpg_extregs[i], dss, slice, subslice);
 				++extarray;
 			}
 	}
@@ -1672,18 +1673,16 @@ snapshot_print_by_list_order(struct xe_hw_engine_snapshot *snapshot, struct drm_
 {
 	struct xe_gt *gt = snapshot->hwe->gt;
 	struct xe_device *xe = gt_to_xe(gt);
-	struct xe_guc *guc = &gt->uc.guc;
 	struct xe_devcoredump *devcoredump = &xe->devcoredump;
 	struct xe_devcoredump_snapshot *devcore_snapshot = &devcoredump->snapshot;
 	struct gcap_reg_list_info *reginfo = NULL;
 	u32 i, last_value = 0;
-	bool is_ext, low32_ready = false;
+	bool low32_ready = false;
 
 	if (!list || !list->list || list->num_regs == 0)
 		return;
 	XE_WARN_ON(!devcore_snapshot->matched_node);
 
-	is_ext = list == guc->capture->extlists;
 	reginfo = &devcore_snapshot->matched_node->reginfo[type];
 
 	/*
@@ -1749,17 +1748,12 @@ snapshot_print_by_list_order(struct xe_hw_engine_snapshot *snapshot, struct drm_
 			 */
 			XE_WARN_ON(low32_ready);
 
-			if (is_ext) {
-				int dss, group, instance;
-
-				group = FIELD_GET(GUC_REGSET_STEERING_GROUP, reg_desc->flags);
-				instance = FIELD_GET(GUC_REGSET_STEERING_INSTANCE, reg_desc->flags);
-				dss = xe_gt_mcr_steering_info_to_dss_id(gt, group, instance);
-
-				drm_printf(p, "\t%s[%u]: 0x%08x\n", reg_desc->regname, dss, value);
-			} else {
+			if (FIELD_GET(GUC_REGSET_STEERING_NEEDED, reg_desc->flags))
+				drm_printf(p, "\t%s[%u]: 0x%08x\n", reg_desc->regname,
+					   reg_desc->dss_id, value);
+			else
 				drm_printf(p, "\t%s: 0x%08x\n", reg_desc->regname, value);
-			}
+
 			break;
 		}
 	}
diff --git a/drivers/gpu/drm/xe/xe_guc_capture_types.h b/drivers/gpu/drm/xe/xe_guc_capture_types.h
index ca2d390ccbee..6cb439115597 100644
--- a/drivers/gpu/drm/xe/xe_guc_capture_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_capture_types.h
@@ -39,6 +39,8 @@ struct __guc_mmio_reg_descr {
 	u32 flags;
 	/** @mask: The mask to apply */
 	u32 mask;
+	/** @dss_id: Cached index for steered registers */
+	u32 dss_id;
 	/** @regname: Name of the register */
 	const char *regname;
 };
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 72ad576fc18e..2447de0ebedf 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -238,7 +238,8 @@ int xe_guc_ct_init(struct xe_guc_ct *ct)
 	bo = xe_managed_bo_create_pin_map(xe, tile, guc_ct_size(),
 					  XE_BO_FLAG_SYSTEM |
 					  XE_BO_FLAG_GGTT |
-					  XE_BO_FLAG_GGTT_INVALIDATE);
+					  XE_BO_FLAG_GGTT_INVALIDATE |
+					  XE_BO_FLAG_PINNED_NORESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
@@ -1088,6 +1089,7 @@ int xe_guc_ct_send_recv(struct xe_guc_ct *ct, const u32 *action, u32 len,
 	KUNIT_STATIC_STUB_REDIRECT(xe_guc_ct_send_recv, ct, action, len, response_buffer);
 	return guc_ct_send_recv(ct, action, len, response_buffer, false);
 }
+ALLOW_ERROR_INJECTION(xe_guc_ct_send_recv, ERRNO);
 
 int xe_guc_ct_send_recv_no_fail(struct xe_guc_ct *ct, const u32 *action,
 				u32 len, u32 *response_buffer)
@@ -1828,10 +1830,10 @@ static void ct_dead_print(struct xe_dead_ct *dead)
 		return;
 	}
 
-	drm_printf(&lp, "CTB is dead - reason=0x%X\n", dead->reason);
 
 	/* Can't generate a genuine core dump at this point, so just do the good bits */
 	drm_puts(&lp, "**** Xe Device Coredump ****\n");
+	drm_printf(&lp, "Reason: CTB is dead - 0x%X\n", dead->reason);
 	xe_device_snapshot_print(xe, &lp);
 
 	drm_printf(&lp, "**** GT #%d ****\n", gt->info.id);
diff --git a/drivers/gpu/drm/xe/xe_guc_debugfs.c b/drivers/gpu/drm/xe/xe_guc_debugfs.c
index c569ff456e74..f33013f8a0f3 100644
--- a/drivers/gpu/drm/xe/xe_guc_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_guc_debugfs.c
@@ -17,101 +17,119 @@
 #include "xe_macros.h"
 #include "xe_pm.h"
 
-static struct xe_guc *node_to_guc(struct drm_info_node *node)
-{
-	return node->info_ent->data;
-}
-
-static int guc_info(struct seq_file *m, void *data)
+/*
+ * guc_debugfs_show - A show callback for struct drm_info_list
+ * @m: the &seq_file
+ * @data: data used by the drm debugfs helpers
+ *
+ * This callback can be used in struct drm_info_list to describe debugfs
+ * files that are &xe_guc specific in similar way how we handle &xe_gt
+ * specific files using &xe_gt_debugfs_simple_show.
+ *
+ * It is assumed that those debugfs files will be created on directory entry
+ * which grandparent struct dentry d_inode->i_private points to &xe_gt.
+ *
+ *      /sys/kernel/debug/dri/0/
+ *      ├── gt0			# dent->d_parent->d_parent (d_inode->i_private == gt)
+ *      │   ├── uc		# dent->d_parent
+ *      │   │   ├── guc_info	# dent
+ *      │   │   ├── guc_...
+ *
+ * This function assumes that &m->private will be set to the &struct
+ * drm_info_node corresponding to the instance of the info on a given &struct
+ * drm_minor (see struct drm_info_list.show for details).
+ *
+ * This function also assumes that struct drm_info_list.data will point to the
+ * function code that will actually print a file content::
+ *
+ *    int (*print)(struct xe_guc *, struct drm_printer *)
+ *
+ * Example::
+ *
+ *    int foo(struct xe_guc *guc, struct drm_printer *p)
+ *    {
+ *        drm_printf(p, "enabled %d\n", guc->submission_state.enabled);
+ *        return 0;
+ *    }
+ *
+ *    static const struct drm_info_list bar[] = {
+ *        { name = "foo", .show = guc_debugfs_show, .data = foo },
+ *    };
+ *
+ *    parent = debugfs_create_dir("uc", gtdir);
+ *    drm_debugfs_create_files(bar, ARRAY_SIZE(bar), parent, minor);
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+static int guc_debugfs_show(struct seq_file *m, void *data)
 {
-	struct xe_guc *guc = node_to_guc(m->private);
-	struct xe_device *xe = guc_to_xe(guc);
 	struct drm_printer p = drm_seq_file_printer(m);
+	struct drm_info_node *node = m->private;
+	struct dentry *parent = node->dent->d_parent;
+	struct dentry *grandparent = parent->d_parent;
+	struct xe_gt *gt = grandparent->d_inode->i_private;
+	struct xe_device *xe = gt_to_xe(gt);
+	int (*print)(struct xe_guc *, struct drm_printer *) = node->info_ent->data;
+	int ret;
 
 	xe_pm_runtime_get(xe);
-	xe_guc_print_info(guc, &p);
+	ret = print(&gt->uc.guc, &p);
 	xe_pm_runtime_put(xe);
 
-	return 0;
+	return ret;
 }
 
-static int guc_log(struct seq_file *m, void *data)
+static int guc_log(struct xe_guc *guc, struct drm_printer *p)
 {
-	struct xe_guc *guc = node_to_guc(m->private);
-	struct xe_device *xe = guc_to_xe(guc);
-	struct drm_printer p = drm_seq_file_printer(m);
-
-	xe_pm_runtime_get(xe);
-	xe_guc_log_print(&guc->log, &p);
-	xe_pm_runtime_put(xe);
-
+	xe_guc_log_print(&guc->log, p);
 	return 0;
 }
 
-static int guc_log_dmesg(struct seq_file *m, void *data)
+static int guc_log_dmesg(struct xe_guc *guc, struct drm_printer *p)
 {
-	struct xe_guc *guc = node_to_guc(m->private);
-	struct xe_device *xe = guc_to_xe(guc);
-
-	xe_pm_runtime_get(xe);
 	xe_guc_log_print_dmesg(&guc->log);
-	xe_pm_runtime_put(xe);
-
 	return 0;
 }
 
-static int guc_ctb(struct seq_file *m, void *data)
+static int guc_ctb(struct xe_guc *guc, struct drm_printer *p)
 {
-	struct xe_guc *guc = node_to_guc(m->private);
-	struct xe_device *xe = guc_to_xe(guc);
-	struct drm_printer p = drm_seq_file_printer(m);
-
-	xe_pm_runtime_get(xe);
-	xe_guc_ct_print(&guc->ct, &p, true);
-	xe_pm_runtime_put(xe);
-
+	xe_guc_ct_print(&guc->ct, p, true);
 	return 0;
 }
 
-static int guc_pc(struct seq_file *m, void *data)
+static int guc_pc(struct xe_guc *guc, struct drm_printer *p)
 {
-	struct xe_guc *guc = node_to_guc(m->private);
-	struct xe_device *xe = guc_to_xe(guc);
-	struct drm_printer p = drm_seq_file_printer(m);
-
-	xe_pm_runtime_get(xe);
-	xe_guc_pc_print(&guc->pc, &p);
-	xe_pm_runtime_put(xe);
-
+	xe_guc_pc_print(&guc->pc, p);
 	return 0;
 }
 
-static const struct drm_info_list debugfs_list[] = {
-	{"guc_info", guc_info, 0},
-	{"guc_log", guc_log, 0},
-	{"guc_log_dmesg", guc_log_dmesg, 0},
-	{"guc_ctb", guc_ctb, 0},
-	{"guc_pc", guc_pc, 0},
+/*
+ * only for GuC debugfs files which can be safely used on the VF as well:
+ * - without access to the GuC privileged registers
+ * - without access to the PF specific GuC objects
+ */
+static const struct drm_info_list vf_safe_debugfs_list[] = {
+	{ "guc_info", .show = guc_debugfs_show, .data = xe_guc_print_info },
+	{ "guc_ctb", .show = guc_debugfs_show, .data = guc_ctb },
+};
+
+/* everything else should be added here */
+static const struct drm_info_list pf_only_debugfs_list[] = {
+	{ "guc_log", .show = guc_debugfs_show, .data = guc_log },
+	{ "guc_log_dmesg", .show = guc_debugfs_show, .data = guc_log_dmesg },
+	{ "guc_pc", .show = guc_debugfs_show, .data = guc_pc },
 };
 
 void xe_guc_debugfs_register(struct xe_guc *guc, struct dentry *parent)
 {
 	struct drm_minor *minor = guc_to_xe(guc)->drm.primary;
-	struct drm_info_list *local;
-	int i;
 
-#define DEBUGFS_SIZE	(ARRAY_SIZE(debugfs_list) * sizeof(struct drm_info_list))
-	local = drmm_kmalloc(&guc_to_xe(guc)->drm, DEBUGFS_SIZE, GFP_KERNEL);
-	if (!local)
-		return;
-
-	memcpy(local, debugfs_list, DEBUGFS_SIZE);
-#undef DEBUGFS_SIZE
-
-	for (i = 0; i < ARRAY_SIZE(debugfs_list); ++i)
-		local[i].data = guc;
-
-	drm_debugfs_create_files(local,
-				 ARRAY_SIZE(debugfs_list),
+	drm_debugfs_create_files(vf_safe_debugfs_list,
+				 ARRAY_SIZE(vf_safe_debugfs_list),
 				 parent, minor);
+
+	if (!IS_SRIOV_VF(guc_to_xe(guc)))
+		drm_debugfs_create_files(pf_only_debugfs_list,
+					 ARRAY_SIZE(pf_only_debugfs_list),
+					 parent, minor);
 }
diff --git a/drivers/gpu/drm/xe/xe_guc_engine_activity.c b/drivers/gpu/drm/xe/xe_guc_engine_activity.c
index 2a457dcf31d5..0fb48f8f05d8 100644
--- a/drivers/gpu/drm/xe/xe_guc_engine_activity.c
+++ b/drivers/gpu/drm/xe/xe_guc_engine_activity.c
@@ -17,36 +17,61 @@
 #include "xe_hw_engine.h"
 #include "xe_map.h"
 #include "xe_mmio.h"
+#include "xe_sriov_pf_helpers.h"
 #include "xe_trace_guc.h"
 
 #define TOTAL_QUANTA 0x8000
 
-static struct iosys_map engine_activity_map(struct xe_guc *guc, struct xe_hw_engine *hwe)
+static struct iosys_map engine_activity_map(struct xe_guc *guc, struct xe_hw_engine *hwe,
+					    unsigned int index)
 {
 	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
-	struct engine_activity_buffer *buffer = &engine_activity->device_buffer;
+	struct engine_activity_buffer *buffer;
 	u16 guc_class = xe_engine_class_to_guc_class(hwe->class);
 	size_t offset;
 
-	offset = offsetof(struct guc_engine_activity_data,
+	if (engine_activity->num_functions) {
+		buffer = &engine_activity->function_buffer;
+		offset = sizeof(struct guc_engine_activity_data) * index;
+	} else {
+		buffer = &engine_activity->device_buffer;
+		offset = 0;
+	}
+
+	offset += offsetof(struct guc_engine_activity_data,
 			  engine_activity[guc_class][hwe->logical_instance]);
 
 	return IOSYS_MAP_INIT_OFFSET(&buffer->activity_bo->vmap, offset);
 }
 
-static struct iosys_map engine_metadata_map(struct xe_guc *guc)
+static struct iosys_map engine_metadata_map(struct xe_guc *guc,
+					    unsigned int index)
 {
 	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
-	struct engine_activity_buffer *buffer = &engine_activity->device_buffer;
+	struct engine_activity_buffer *buffer;
+	size_t offset;
 
-	return buffer->metadata_bo->vmap;
+	if (engine_activity->num_functions) {
+		buffer = &engine_activity->function_buffer;
+		offset = sizeof(struct guc_engine_activity_metadata) * index;
+	} else {
+		buffer = &engine_activity->device_buffer;
+		offset = 0;
+	}
+
+	return IOSYS_MAP_INIT_OFFSET(&buffer->metadata_bo->vmap, offset);
 }
 
 static int allocate_engine_activity_group(struct xe_guc *guc)
 {
 	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
 	struct xe_device *xe = guc_to_xe(guc);
-	u32 num_activity_group = 1; /* Will be modified for VF */
+	u32 num_activity_group;
+
+	/*
+	 * An additional activity group is allocated for PF
+	 */
+	num_activity_group = IS_SRIOV_PF(xe) ? xe_sriov_pf_get_totalvfs(xe) + 1 : 1;
 
 	engine_activity->eag  = drmm_kcalloc(&xe->drm, num_activity_group,
 					     sizeof(struct engine_activity_group), GFP_KERNEL);
@@ -60,10 +85,11 @@ static int allocate_engine_activity_group(struct xe_guc *guc)
 }
 
 static int allocate_engine_activity_buffers(struct xe_guc *guc,
-					    struct engine_activity_buffer *buffer)
+					    struct engine_activity_buffer *buffer,
+					    int count)
 {
-	u32 metadata_size = sizeof(struct guc_engine_activity_metadata);
-	u32 size = sizeof(struct guc_engine_activity_data);
+	u32 metadata_size = sizeof(struct guc_engine_activity_metadata) * count;
+	u32 size = sizeof(struct guc_engine_activity_data) * count;
 	struct xe_gt *gt = guc_to_gt(guc);
 	struct xe_tile *tile = gt_to_tile(gt);
 	struct xe_bo *bo, *metadata_bo;
@@ -118,10 +144,11 @@ static bool is_engine_activity_supported(struct xe_guc *guc)
 	return true;
 }
 
-static struct engine_activity *hw_engine_to_engine_activity(struct xe_hw_engine *hwe)
+static struct engine_activity *hw_engine_to_engine_activity(struct xe_hw_engine *hwe,
+							    unsigned int index)
 {
 	struct xe_guc *guc = &hwe->gt->uc.guc;
-	struct engine_activity_group *eag = &guc->engine_activity.eag[0];
+	struct engine_activity_group *eag = &guc->engine_activity.eag[index];
 	u16 guc_class = xe_engine_class_to_guc_class(hwe->class);
 
 	return &eag->engine[guc_class][hwe->logical_instance];
@@ -138,9 +165,10 @@ static u64 cpu_ns_to_guc_tsc_tick(ktime_t ns, u32 freq)
 #define read_metadata_record(xe_, map_, field_) \
 	xe_map_rd_field(xe_, map_, 0, struct guc_engine_activity_metadata, field_)
 
-static u64 get_engine_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
+static u64 get_engine_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe,
+				   unsigned int index)
 {
-	struct engine_activity *ea = hw_engine_to_engine_activity(hwe);
+	struct engine_activity *ea = hw_engine_to_engine_activity(hwe, index);
 	struct guc_engine_activity *cached_activity = &ea->activity;
 	struct guc_engine_activity_metadata *cached_metadata = &ea->metadata;
 	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
@@ -151,8 +179,8 @@ static u64 get_engine_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
 	u64 active_ticks, gpm_ts;
 	u16 change_num;
 
-	activity_map = engine_activity_map(guc, hwe);
-	metadata_map = engine_metadata_map(guc);
+	activity_map = engine_activity_map(guc, hwe, index);
+	metadata_map = engine_metadata_map(guc, index);
 	global_change_num = read_metadata_record(xe, &metadata_map, global_change_num);
 
 	/* GuC has not initialized activity data yet, return 0 */
@@ -194,9 +222,9 @@ update:
 	return ea->total + ea->active;
 }
 
-static u64 get_engine_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
+static u64 get_engine_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe, unsigned int index)
 {
-	struct engine_activity *ea = hw_engine_to_engine_activity(hwe);
+	struct engine_activity *ea = hw_engine_to_engine_activity(hwe, index);
 	struct guc_engine_activity_metadata *cached_metadata = &ea->metadata;
 	struct guc_engine_activity *cached_activity = &ea->activity;
 	struct iosys_map activity_map, metadata_map;
@@ -205,8 +233,8 @@ static u64 get_engine_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
 	u64 numerator;
 	u16 quanta_ratio;
 
-	activity_map = engine_activity_map(guc, hwe);
-	metadata_map = engine_metadata_map(guc);
+	activity_map = engine_activity_map(guc, hwe, index);
+	metadata_map = engine_metadata_map(guc, index);
 
 	if (!cached_metadata->guc_tsc_frequency_hz)
 		cached_metadata->guc_tsc_frequency_hz = read_metadata_record(xe, &metadata_map,
@@ -245,12 +273,39 @@ static int enable_engine_activity_stats(struct xe_guc *guc)
 	return xe_guc_ct_send_block(&guc->ct, action, ARRAY_SIZE(action));
 }
 
-static void engine_activity_set_cpu_ts(struct xe_guc *guc)
+static int enable_function_engine_activity_stats(struct xe_guc *guc, bool enable)
 {
 	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
-	struct engine_activity_group *eag = &engine_activity->eag[0];
+	u32 metadata_ggtt_addr = 0, ggtt_addr = 0, num_functions = 0;
+	struct engine_activity_buffer *buffer = &engine_activity->function_buffer;
+	u32 action[6];
+	int len = 0;
+
+	if (enable) {
+		metadata_ggtt_addr = xe_bo_ggtt_addr(buffer->metadata_bo);
+		ggtt_addr = xe_bo_ggtt_addr(buffer->activity_bo);
+		num_functions = engine_activity->num_functions;
+	}
+
+	action[len++] = XE_GUC_ACTION_SET_FUNCTION_ENGINE_ACTIVITY_BUFFER;
+	action[len++] = num_functions;
+	action[len++] = metadata_ggtt_addr;
+	action[len++] = 0;
+	action[len++] = ggtt_addr;
+	action[len++] = 0;
+
+	/* Blocking here to ensure the buffers are ready before reading them */
+	return xe_guc_ct_send_block(&guc->ct, action, ARRAY_SIZE(action));
+}
+
+static void engine_activity_set_cpu_ts(struct xe_guc *guc, unsigned int index)
+{
+	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
+	struct engine_activity_group *eag = &engine_activity->eag[index];
 	int i, j;
 
+	xe_gt_assert(guc_to_gt(guc), index < engine_activity->num_activity_group);
+
 	for (i = 0; i < GUC_MAX_ENGINE_CLASSES; i++)
 		for (j = 0; j < GUC_MAX_INSTANCES_PER_CLASS; j++)
 			eag->engine[i][j].last_cpu_ts = ktime_get();
@@ -265,34 +320,107 @@ static u32 gpm_timestamp_shift(struct xe_gt *gt)
 	return 3 - REG_FIELD_GET(RPM_CONFIG0_CTC_SHIFT_PARAMETER_MASK, reg);
 }
 
+static bool is_function_valid(struct xe_guc *guc, unsigned int fn_id)
+{
+	struct xe_device *xe = guc_to_xe(guc);
+	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
+
+	if (!IS_SRIOV_PF(xe) && fn_id)
+		return false;
+
+	if (engine_activity->num_functions && fn_id >= engine_activity->num_functions)
+		return false;
+
+	return true;
+}
+
+static int engine_activity_disable_function_stats(struct xe_guc *guc)
+{
+	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
+	struct engine_activity_buffer *buffer = &engine_activity->function_buffer;
+	int ret;
+
+	if (!engine_activity->num_functions)
+		return 0;
+
+	ret = enable_function_engine_activity_stats(guc, false);
+	if (ret)
+		return ret;
+
+	free_engine_activity_buffers(buffer);
+	engine_activity->num_functions = 0;
+
+	return 0;
+}
+
+static int engine_activity_enable_function_stats(struct xe_guc *guc, int num_vfs)
+{
+	struct xe_guc_engine_activity *engine_activity = &guc->engine_activity;
+	struct engine_activity_buffer *buffer = &engine_activity->function_buffer;
+	int ret, i;
+
+	if (!num_vfs)
+		return 0;
+
+	/* This includes 1 PF and num_vfs */
+	engine_activity->num_functions = num_vfs + 1;
+
+	ret = allocate_engine_activity_buffers(guc, buffer, engine_activity->num_functions);
+	if (ret)
+		return ret;
+
+	ret = enable_function_engine_activity_stats(guc, true);
+	if (ret) {
+		free_engine_activity_buffers(buffer);
+		engine_activity->num_functions = 0;
+		return ret;
+	}
+
+	/* skip PF as it was already setup */
+	for (i = 1; i < engine_activity->num_functions; i++)
+		engine_activity_set_cpu_ts(guc, i);
+
+	return 0;
+}
+
 /**
  * xe_guc_engine_activity_active_ticks - Get engine active ticks
  * @guc: The GuC object
  * @hwe: The hw_engine object
+ * @fn_id: function id to report on
  *
  * Return: accumulated ticks @hwe was active since engine activity stats were enabled.
  */
-u64 xe_guc_engine_activity_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
+u64 xe_guc_engine_activity_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe,
+					unsigned int fn_id)
 {
 	if (!xe_guc_engine_activity_supported(guc))
 		return 0;
 
-	return get_engine_active_ticks(guc, hwe);
+	if (!is_function_valid(guc, fn_id))
+		return 0;
+
+	return get_engine_active_ticks(guc, hwe, fn_id);
 }
 
 /**
  * xe_guc_engine_activity_total_ticks - Get engine total ticks
  * @guc: The GuC object
  * @hwe: The hw_engine object
+ * @fn_id: function id to report on
  *
  * Return: accumulated quanta of ticks allocated for the engine
  */
-u64 xe_guc_engine_activity_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe)
+u64 xe_guc_engine_activity_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe,
+				       unsigned int fn_id)
 {
 	if (!xe_guc_engine_activity_supported(guc))
 		return 0;
 
-	return get_engine_total_ticks(guc, hwe);
+	if (!is_function_valid(guc, fn_id))
+		return 0;
+
+	return get_engine_total_ticks(guc, hwe, fn_id);
 }
 
 /**
@@ -311,6 +439,25 @@ bool xe_guc_engine_activity_supported(struct xe_guc *guc)
 }
 
 /**
+ * xe_guc_engine_activity_function_stats - Enable/Disable per-function engine activity stats
+ * @guc: The GuC object
+ * @num_vfs: number of vfs
+ * @enable: true to enable, false otherwise
+ *
+ * Return: 0 on success, negative error code otherwise
+ */
+int xe_guc_engine_activity_function_stats(struct xe_guc *guc, int num_vfs, bool enable)
+{
+	if (!xe_guc_engine_activity_supported(guc))
+		return 0;
+
+	if (enable)
+		return engine_activity_enable_function_stats(guc, num_vfs);
+
+	return engine_activity_disable_function_stats(guc);
+}
+
+/**
  * xe_guc_engine_activity_enable_stats - Enable engine activity stats
  * @guc: The GuC object
  *
@@ -327,7 +474,7 @@ void xe_guc_engine_activity_enable_stats(struct xe_guc *guc)
 	if (ret)
 		xe_gt_err(guc_to_gt(guc), "failed to enable activity stats%d\n", ret);
 	else
-		engine_activity_set_cpu_ts(guc);
+		engine_activity_set_cpu_ts(guc, 0);
 }
 
 static void engine_activity_fini(void *arg)
@@ -360,7 +507,7 @@ int xe_guc_engine_activity_init(struct xe_guc *guc)
 		return ret;
 	}
 
-	ret = allocate_engine_activity_buffers(guc, &engine_activity->device_buffer);
+	ret = allocate_engine_activity_buffers(guc, &engine_activity->device_buffer, 1);
 	if (ret) {
 		xe_gt_err(gt, "failed to allocate engine activity buffers (%pe)\n", ERR_PTR(ret));
 		return ret;
diff --git a/drivers/gpu/drm/xe/xe_guc_engine_activity.h b/drivers/gpu/drm/xe/xe_guc_engine_activity.h
index a042d4cb404c..b32926c2d208 100644
--- a/drivers/gpu/drm/xe/xe_guc_engine_activity.h
+++ b/drivers/gpu/drm/xe/xe_guc_engine_activity.h
@@ -14,6 +14,9 @@ struct xe_guc;
 int xe_guc_engine_activity_init(struct xe_guc *guc);
 bool xe_guc_engine_activity_supported(struct xe_guc *guc);
 void xe_guc_engine_activity_enable_stats(struct xe_guc *guc);
-u64 xe_guc_engine_activity_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe);
-u64 xe_guc_engine_activity_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe);
+int xe_guc_engine_activity_function_stats(struct xe_guc *guc, int num_vfs, bool enable);
+u64 xe_guc_engine_activity_active_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe,
+					unsigned int fn_id);
+u64 xe_guc_engine_activity_total_ticks(struct xe_guc *guc, struct xe_hw_engine *hwe,
+				       unsigned int fn_id);
 #endif
diff --git a/drivers/gpu/drm/xe/xe_guc_engine_activity_types.h b/drivers/gpu/drm/xe/xe_guc_engine_activity_types.h
index 5cdd034b6b70..48f69ddefa36 100644
--- a/drivers/gpu/drm/xe/xe_guc_engine_activity_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_engine_activity_types.h
@@ -79,14 +79,24 @@ struct xe_guc_engine_activity {
 	/** @num_activity_group: number of activity groups */
 	u32 num_activity_group;
 
+	/** @num_functions: number of functions */
+	u32 num_functions;
+
 	/** @supported: indicates support for engine activity stats */
 	bool supported;
 
-	/** @eag: holds the device level engine activity data */
+	/**
+	 * @eag: holds the device level engine activity data in native mode.
+	 * In SRIOV mode, points to an array with entries which holds the engine
+	 * activity data for PF and VF's
+	 */
 	struct engine_activity_group *eag;
 
 	/** @device_buffer: buffer object for global engine activity */
 	struct engine_activity_buffer device_buffer;
+
+	/** @function_buffer: buffer object for per-function engine activity */
+	struct engine_activity_buffer function_buffer;
 };
 #endif
 
diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c
index 80514a446ba2..38039c411387 100644
--- a/drivers/gpu/drm/xe/xe_guc_log.c
+++ b/drivers/gpu/drm/xe/xe_guc_log.c
@@ -260,7 +260,8 @@ int xe_guc_log_init(struct xe_guc_log *log)
 	bo = xe_managed_bo_create_pin_map(xe, tile, guc_log_size(),
 					  XE_BO_FLAG_SYSTEM |
 					  XE_BO_FLAG_GGTT |
-					  XE_BO_FLAG_GGTT_INVALIDATE);
+					  XE_BO_FLAG_GGTT_INVALIDATE |
+					  XE_BO_FLAG_PINNED_NORESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c
index 85215313976c..18c623992035 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.c
+++ b/drivers/gpu/drm/xe/xe_guc_pc.c
@@ -462,6 +462,21 @@ static u32 get_cur_freq(struct xe_gt *gt)
 }
 
 /**
+ * xe_guc_pc_get_cur_freq_fw - With fw held, get requested frequency
+ * @pc: The GuC PC
+ *
+ * Returns: the requested frequency for that GT instance
+ */
+u32 xe_guc_pc_get_cur_freq_fw(struct xe_guc_pc *pc)
+{
+	struct xe_gt *gt = pc_to_gt(pc);
+
+	xe_force_wake_assert_held(gt_to_fw(gt), XE_FW_GT);
+
+	return get_cur_freq(gt);
+}
+
+/**
  * xe_guc_pc_get_cur_freq - Get Current requested frequency
  * @pc: The GuC PC
  * @freq: A pointer to a u32 where the freq value will be returned
@@ -1070,6 +1085,7 @@ int xe_guc_pc_start(struct xe_guc_pc *pc)
 		if (wait_for_pc_state(pc, SLPC_GLOBAL_STATE_RUNNING,
 				      SLPC_RESET_EXTENDED_TIMEOUT_MS)) {
 			xe_gt_err(gt, "GuC PC Start failed: Dynamic GT frequency control and GT sleep states are now disabled.\n");
+			ret = -EIO;
 			goto out;
 		}
 
@@ -1169,7 +1185,8 @@ int xe_guc_pc_init(struct xe_guc_pc *pc)
 	bo = xe_managed_bo_create_pin_map(xe, tile, size,
 					  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
 					  XE_BO_FLAG_GGTT |
-					  XE_BO_FLAG_GGTT_INVALIDATE);
+					  XE_BO_FLAG_GGTT_INVALIDATE |
+					  XE_BO_FLAG_PINNED_NORESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.h b/drivers/gpu/drm/xe/xe_guc_pc.h
index 39102b79602f..0a2664d5c811 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.h
+++ b/drivers/gpu/drm/xe/xe_guc_pc.h
@@ -22,6 +22,7 @@ void xe_guc_pc_print(struct xe_guc_pc *pc, struct drm_printer *p);
 
 u32 xe_guc_pc_get_act_freq(struct xe_guc_pc *pc);
 int xe_guc_pc_get_cur_freq(struct xe_guc_pc *pc, u32 *freq);
+u32 xe_guc_pc_get_cur_freq_fw(struct xe_guc_pc *pc);
 u32 xe_guc_pc_get_rp0_freq(struct xe_guc_pc *pc);
 u32 xe_guc_pc_get_rpa_freq(struct xe_guc_pc *pc);
 u32 xe_guc_pc_get_rpe_freq(struct xe_guc_pc *pc);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 31bc2022bfc2..369be36f7dc5 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -300,6 +300,8 @@ int xe_guc_submit_init(struct xe_guc *guc, unsigned int num_ids)
 
 	primelockdep(guc);
 
+	guc->submission_state.initialized = true;
+
 	return drmm_add_action_or_reset(&xe->drm, guc_submit_fini, guc);
 }
 
@@ -834,6 +836,13 @@ void xe_guc_submit_wedge(struct xe_guc *guc)
 
 	xe_gt_assert(guc_to_gt(guc), guc_to_xe(guc)->wedged.mode);
 
+	/*
+	 * If device is being wedged even before submission_state is
+	 * initialized, there's nothing to do here.
+	 */
+	if (!guc->submission_state.initialized)
+		return;
+
 	err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
 				       guc_submit_wedged_fini, guc);
 	if (err) {
@@ -1170,9 +1179,12 @@ trigger_reset:
 		process_name = q->vm->xef->process_name;
 		pid = q->vm->xef->pid;
 	}
-	xe_gt_notice(guc_to_gt(guc), "Timedout job: seqno=%u, lrc_seqno=%u, guc_id=%d, flags=0x%lx in %s [%d]",
-		     xe_sched_job_seqno(job), xe_sched_job_lrc_seqno(job),
-		     q->guc->id, q->flags, process_name, pid);
+
+	if (!exec_queue_killed(q))
+		xe_gt_notice(guc_to_gt(guc),
+			     "Timedout job: seqno=%u, lrc_seqno=%u, guc_id=%d, flags=0x%lx in %s [%d]",
+			     xe_sched_job_seqno(job), xe_sched_job_lrc_seqno(job),
+			     q->guc->id, q->flags, process_name, pid);
 
 	trace_xe_sched_job_timedout(job);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_types.h b/drivers/gpu/drm/xe/xe_guc_types.h
index 63bac64429a5..1fde7614fcc5 100644
--- a/drivers/gpu/drm/xe/xe_guc_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_types.h
@@ -89,6 +89,11 @@ struct xe_guc {
 		struct mutex lock;
 		/** @submission_state.enabled: submission is enabled */
 		bool enabled;
+		/**
+		 * @submission_state.initialized: mark when submission state is
+		 * even initialized - before that not even the lock is valid
+		 */
+		bool initialized;
 		/** @submission_state.fini_wq: submit fini wait queue */
 		wait_queue_head_t fini_wq;
 	} submission_state;
diff --git a/drivers/gpu/drm/xe/xe_hmm.c b/drivers/gpu/drm/xe/xe_hmm.c
index c3cc0fa105e8..57b71956ddf4 100644
--- a/drivers/gpu/drm/xe/xe_hmm.c
+++ b/drivers/gpu/drm/xe/xe_hmm.c
@@ -19,29 +19,6 @@ static u64 xe_npages_in_range(unsigned long start, unsigned long end)
 	return (end - start) >> PAGE_SHIFT;
 }
 
-/**
- * xe_mark_range_accessed() - mark a range is accessed, so core mm
- * have such information for memory eviction or write back to
- * hard disk
- * @range: the range to mark
- * @write: if write to this range, we mark pages in this range
- * as dirty
- */
-static void xe_mark_range_accessed(struct hmm_range *range, bool write)
-{
-	struct page *page;
-	u64 i, npages;
-
-	npages = xe_npages_in_range(range->start, range->end);
-	for (i = 0; i < npages; i++) {
-		page = hmm_pfn_to_page(range->hmm_pfns[i]);
-		if (write)
-			set_page_dirty_lock(page);
-
-		mark_page_accessed(page);
-	}
-}
-
 static int xe_alloc_sg(struct xe_device *xe, struct sg_table *st,
 		       struct hmm_range *range, struct rw_semaphore *notifier_sem)
 {
@@ -331,7 +308,6 @@ int xe_hmm_userptr_populate_range(struct xe_userptr_vma *uvma,
 	if (ret)
 		goto out_unlock;
 
-	xe_mark_range_accessed(&hmm_range, write);
 	userptr->sg = &userptr->sgt;
 	xe_hmm_userptr_set_mapped(uvma);
 	userptr->notifier_seq = hmm_range.notifier_seq;
diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c
index 8c05fd30b7df..93241fd0a4ba 100644
--- a/drivers/gpu/drm/xe/xe_hw_engine.c
+++ b/drivers/gpu/drm/xe/xe_hw_engine.c
@@ -389,12 +389,6 @@ xe_hw_engine_setup_default_lrc_state(struct xe_hw_engine *hwe)
 				 blit_cctl_val,
 				 XE_RTP_ACTION_FLAG(ENGINE_BASE)))
 		},
-		/* Use Fixed slice CCS mode */
-		{ XE_RTP_NAME("RCU_MODE_FIXED_SLICE_CCS_MODE"),
-		  XE_RTP_RULES(FUNC(xe_hw_engine_match_fixed_cslice_mode)),
-		  XE_RTP_ACTIONS(FIELD_SET(RCU_MODE, RCU_MODE_FIXED_SLICE_CCS_MODE,
-					   RCU_MODE_FIXED_SLICE_CCS_MODE))
-		},
 		/* Disable WMTP if HW doesn't support it */
 		{ XE_RTP_NAME("DISABLE_WMTP_ON_UNSUPPORTED_HW"),
 		  XE_RTP_RULES(FUNC(xe_rtp_cfeg_wmtp_disabled)),
@@ -461,6 +455,12 @@ hw_engine_setup_default_state(struct xe_hw_engine *hwe)
 		  XE_RTP_ACTIONS(SET(CSFE_CHICKEN1(0), CS_PRIORITY_MEM_READ,
 				     XE_RTP_ACTION_FLAG(ENGINE_BASE)))
 		},
+		/* Use Fixed slice CCS mode */
+		{ XE_RTP_NAME("RCU_MODE_FIXED_SLICE_CCS_MODE"),
+		  XE_RTP_RULES(FUNC(xe_hw_engine_match_fixed_cslice_mode)),
+		  XE_RTP_ACTIONS(FIELD_SET(RCU_MODE, RCU_MODE_FIXED_SLICE_CCS_MODE,
+					   RCU_MODE_FIXED_SLICE_CCS_MODE))
+		},
 	};
 
 	xe_rtp_process_to_sr(&ctx, engine_entries, ARRAY_SIZE(engine_entries), &hwe->reg_sr);
diff --git a/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c b/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c
index b53e8d2accdb..640950172088 100644
--- a/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c
@@ -32,14 +32,61 @@ bool xe_hw_engine_timeout_in_range(u64 timeout, u64 min, u64 max)
 	return timeout >= min && timeout <= max;
 }
 
-static void kobj_xe_hw_engine_release(struct kobject *kobj)
+static void xe_hw_engine_sysfs_kobj_release(struct kobject *kobj)
 {
 	kfree(kobj);
 }
 
+static ssize_t xe_hw_engine_class_sysfs_attr_show(struct kobject *kobj,
+						  struct attribute *attr,
+						  char *buf)
+{
+	struct xe_device *xe = kobj_to_xe(kobj);
+	struct kobj_attribute *kattr;
+	ssize_t ret = -EIO;
+
+	kattr = container_of(attr, struct kobj_attribute, attr);
+	if (kattr->show) {
+		xe_pm_runtime_get(xe);
+		ret = kattr->show(kobj, kattr, buf);
+		xe_pm_runtime_put(xe);
+	}
+
+	return ret;
+}
+
+static ssize_t xe_hw_engine_class_sysfs_attr_store(struct kobject *kobj,
+						   struct attribute *attr,
+						   const char *buf,
+						   size_t count)
+{
+	struct xe_device *xe = kobj_to_xe(kobj);
+	struct kobj_attribute *kattr;
+	ssize_t ret = -EIO;
+
+	kattr = container_of(attr, struct kobj_attribute, attr);
+	if (kattr->store) {
+		xe_pm_runtime_get(xe);
+		ret = kattr->store(kobj, kattr, buf, count);
+		xe_pm_runtime_put(xe);
+	}
+
+	return ret;
+}
+
+static const struct sysfs_ops xe_hw_engine_class_sysfs_ops = {
+	.show = xe_hw_engine_class_sysfs_attr_show,
+	.store = xe_hw_engine_class_sysfs_attr_store,
+};
+
 static const struct kobj_type kobj_xe_hw_engine_type = {
-	.release = kobj_xe_hw_engine_release,
-	.sysfs_ops = &kobj_sysfs_ops
+	.release = xe_hw_engine_sysfs_kobj_release,
+	.sysfs_ops = &xe_hw_engine_class_sysfs_ops,
+};
+
+static const struct kobj_type kobj_xe_hw_engine_type_def = {
+	.release = xe_hw_engine_sysfs_kobj_release,
+	.sysfs_ops = &kobj_sysfs_ops,
 };
 
 static ssize_t job_timeout_max_store(struct kobject *kobj,
@@ -543,7 +590,7 @@ static int xe_add_hw_engine_class_defaults(struct xe_device *xe,
 	if (!kobj)
 		return -ENOMEM;
 
-	kobject_init(kobj, &kobj_xe_hw_engine_type);
+	kobject_init(kobj, &kobj_xe_hw_engine_type_def);
 	err = kobject_add(kobj, parent, "%s", ".defaults");
 	if (err)
 		goto err_object;
@@ -558,58 +605,8 @@ err_object:
 	kobject_put(kobj);
 	return err;
 }
+ALLOW_ERROR_INJECTION(xe_add_hw_engine_class_defaults, ERRNO); /* See xe_pci_probe() */
 
-static void xe_hw_engine_sysfs_kobj_release(struct kobject *kobj)
-{
-	kfree(kobj);
-}
-
-static ssize_t xe_hw_engine_class_sysfs_attr_show(struct kobject *kobj,
-						  struct attribute *attr,
-						  char *buf)
-{
-	struct xe_device *xe = kobj_to_xe(kobj);
-	struct kobj_attribute *kattr;
-	ssize_t ret = -EIO;
-
-	kattr = container_of(attr, struct kobj_attribute, attr);
-	if (kattr->show) {
-		xe_pm_runtime_get(xe);
-		ret = kattr->show(kobj, kattr, buf);
-		xe_pm_runtime_put(xe);
-	}
-
-	return ret;
-}
-
-static ssize_t xe_hw_engine_class_sysfs_attr_store(struct kobject *kobj,
-						   struct attribute *attr,
-						   const char *buf,
-						   size_t count)
-{
-	struct xe_device *xe = kobj_to_xe(kobj);
-	struct kobj_attribute *kattr;
-	ssize_t ret = -EIO;
-
-	kattr = container_of(attr, struct kobj_attribute, attr);
-	if (kattr->store) {
-		xe_pm_runtime_get(xe);
-		ret = kattr->store(kobj, kattr, buf, count);
-		xe_pm_runtime_put(xe);
-	}
-
-	return ret;
-}
-
-static const struct sysfs_ops xe_hw_engine_class_sysfs_ops = {
-	.show = xe_hw_engine_class_sysfs_attr_show,
-	.store = xe_hw_engine_class_sysfs_attr_store,
-};
-
-static const struct kobj_type xe_hw_engine_sysfs_kobj_type = {
-	.release = xe_hw_engine_sysfs_kobj_release,
-	.sysfs_ops = &xe_hw_engine_class_sysfs_ops,
-};
 
 static void hw_engine_class_sysfs_fini(void *arg)
 {
@@ -640,7 +637,7 @@ int xe_hw_engine_class_sysfs_init(struct xe_gt *gt)
 	if (!kobj)
 		return -ENOMEM;
 
-	kobject_init(kobj, &xe_hw_engine_sysfs_kobj_type);
+	kobject_init(kobj, &kobj_xe_hw_engine_type);
 
 	err = kobject_add(kobj, gt->sysfs, "engines");
 	if (err)
diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c
index 48d80ffdf7bb..eb293aec36a0 100644
--- a/drivers/gpu/drm/xe/xe_hwmon.c
+++ b/drivers/gpu/drm/xe/xe_hwmon.c
@@ -5,6 +5,7 @@
 
 #include <linux/hwmon-sysfs.h>
 #include <linux/hwmon.h>
+#include <linux/jiffies.h>
 #include <linux/types.h>
 #include <linux/units.h>
 
@@ -27,6 +28,7 @@ enum xe_hwmon_reg {
 	REG_PKG_POWER_SKU_UNIT,
 	REG_GT_PERF_STATUS,
 	REG_PKG_ENERGY_STATUS,
+	REG_FAN_SPEED,
 };
 
 enum xe_hwmon_reg_operation {
@@ -42,6 +44,13 @@ enum xe_hwmon_channel {
 	CHANNEL_MAX,
 };
 
+enum xe_fan_channel {
+	FAN_1,
+	FAN_2,
+	FAN_3,
+	FAN_MAX,
+};
+
 /*
  * SF_* - scale factors for particular quantities according to hwmon spec.
  */
@@ -62,6 +71,16 @@ struct xe_hwmon_energy_info {
 };
 
 /**
+ * struct xe_hwmon_fan_info - to cache previous fan reading
+ */
+struct xe_hwmon_fan_info {
+	/** @reg_val_prev: previous fan reg val */
+	u32 reg_val_prev;
+	/** @time_prev: previous timestamp */
+	u64 time_prev;
+};
+
+/**
  * struct xe_hwmon - xe hwmon data structure
  */
 struct xe_hwmon {
@@ -79,6 +98,8 @@ struct xe_hwmon {
 	int scl_shift_time;
 	/** @ei: Energy info for energyN_input */
 	struct xe_hwmon_energy_info ei[CHANNEL_MAX];
+	/** @fi: Fan info for fanN_input */
+	struct xe_hwmon_fan_info fi[FAN_MAX];
 };
 
 static struct xe_reg xe_hwmon_get_reg(struct xe_hwmon *hwmon, enum xe_hwmon_reg hwmon_reg,
@@ -144,6 +165,14 @@ static struct xe_reg xe_hwmon_get_reg(struct xe_hwmon *hwmon, enum xe_hwmon_reg
 			return PCU_CR_PACKAGE_ENERGY_STATUS;
 		}
 		break;
+	case REG_FAN_SPEED:
+		if (channel == FAN_1)
+			return BMG_FAN_1_SPEED;
+		else if (channel == FAN_2)
+			return BMG_FAN_2_SPEED;
+		else if (channel == FAN_3)
+			return BMG_FAN_3_SPEED;
+		break;
 	default:
 		drm_warn(&xe->drm, "Unknown xe hwmon reg id: %d\n", hwmon_reg);
 		break;
@@ -454,6 +483,7 @@ static const struct hwmon_channel_info * const hwmon_info[] = {
 	HWMON_CHANNEL_INFO(curr, HWMON_C_LABEL, HWMON_C_CRIT | HWMON_C_LABEL),
 	HWMON_CHANNEL_INFO(in, HWMON_I_INPUT | HWMON_I_LABEL, HWMON_I_INPUT | HWMON_I_LABEL),
 	HWMON_CHANNEL_INFO(energy, HWMON_E_INPUT | HWMON_E_LABEL, HWMON_E_INPUT | HWMON_E_LABEL),
+	HWMON_CHANNEL_INFO(fan, HWMON_F_INPUT, HWMON_F_INPUT, HWMON_F_INPUT),
 	NULL
 };
 
@@ -480,6 +510,19 @@ static int xe_hwmon_pcode_write_i1(const struct xe_hwmon *hwmon, u32 uval)
 			      (uval & POWER_SETUP_I1_DATA_MASK));
 }
 
+static int xe_hwmon_pcode_read_fan_control(const struct xe_hwmon *hwmon, u32 subcmd, u32 *uval)
+{
+	struct xe_tile *root_tile = xe_device_get_root_tile(hwmon->xe);
+
+	/* Platforms that don't return correct value */
+	if (hwmon->xe->info.platform == XE_DG2 && subcmd == FSC_READ_NUM_FANS) {
+		*uval = 2;
+		return 0;
+	}
+
+	return xe_pcode_read(root_tile, PCODE_MBOX(FAN_SPEED_CONTROL, subcmd, 0), uval, NULL);
+}
+
 static int xe_hwmon_power_curr_crit_read(struct xe_hwmon *hwmon, int channel,
 					 long *value, u32 scale_factor)
 {
@@ -706,6 +749,75 @@ xe_hwmon_energy_read(struct xe_hwmon *hwmon, u32 attr, int channel, long *val)
 }
 
 static umode_t
+xe_hwmon_fan_is_visible(struct xe_hwmon *hwmon, u32 attr, int channel)
+{
+	u32 uval;
+
+	if (!hwmon->xe->info.has_fan_control)
+		return 0;
+
+	switch (attr) {
+	case hwmon_fan_input:
+		if (xe_hwmon_pcode_read_fan_control(hwmon, FSC_READ_NUM_FANS, &uval))
+			return 0;
+
+		return channel < uval ? 0444 : 0;
+	default:
+		return 0;
+	}
+}
+
+static int
+xe_hwmon_fan_input_read(struct xe_hwmon *hwmon, int channel, long *val)
+{
+	struct xe_mmio *mmio = xe_root_tile_mmio(hwmon->xe);
+	struct xe_hwmon_fan_info *fi = &hwmon->fi[channel];
+	u64 rotations, time_now, time;
+	u32 reg_val;
+	int ret = 0;
+
+	mutex_lock(&hwmon->hwmon_lock);
+
+	reg_val = xe_mmio_read32(mmio, xe_hwmon_get_reg(hwmon, REG_FAN_SPEED, channel));
+	time_now = get_jiffies_64();
+
+	/*
+	 * HW register value is accumulated count of pulses from PWM fan with the scale
+	 * of 2 pulses per rotation.
+	 */
+	rotations = (reg_val - fi->reg_val_prev) / 2;
+
+	time = jiffies_delta_to_msecs(time_now - fi->time_prev);
+	if (unlikely(!time)) {
+		ret = -EAGAIN;
+		goto unlock;
+	}
+
+	/*
+	 * Calculate fan speed in RPM by time averaging two subsequent readings in minutes.
+	 * RPM = number of rotations * msecs per minute / time in msecs
+	 */
+	*val = DIV_ROUND_UP_ULL(rotations * (MSEC_PER_SEC * 60), time);
+
+	fi->reg_val_prev = reg_val;
+	fi->time_prev = time_now;
+unlock:
+	mutex_unlock(&hwmon->hwmon_lock);
+	return ret;
+}
+
+static int
+xe_hwmon_fan_read(struct xe_hwmon *hwmon, u32 attr, int channel, long *val)
+{
+	switch (attr) {
+	case hwmon_fan_input:
+		return xe_hwmon_fan_input_read(hwmon, channel, val);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static umode_t
 xe_hwmon_is_visible(const void *drvdata, enum hwmon_sensor_types type,
 		    u32 attr, int channel)
 {
@@ -730,6 +842,9 @@ xe_hwmon_is_visible(const void *drvdata, enum hwmon_sensor_types type,
 	case hwmon_energy:
 		ret = xe_hwmon_energy_is_visible(hwmon, attr, channel);
 		break;
+	case hwmon_fan:
+		ret = xe_hwmon_fan_is_visible(hwmon, attr, channel);
+		break;
 	default:
 		ret = 0;
 		break;
@@ -765,6 +880,9 @@ xe_hwmon_read(struct device *dev, enum hwmon_sensor_types type, u32 attr,
 	case hwmon_energy:
 		ret = xe_hwmon_energy_read(hwmon, attr, channel, val);
 		break;
+	case hwmon_fan:
+		ret = xe_hwmon_fan_read(hwmon, attr, channel, val);
+		break;
 	default:
 		ret = -EOPNOTSUPP;
 		break;
@@ -842,7 +960,7 @@ static void
 xe_hwmon_get_preregistration_info(struct xe_hwmon *hwmon)
 {
 	struct xe_mmio *mmio = xe_root_tile_mmio(hwmon->xe);
-	long energy;
+	long energy, fan_speed;
 	u64 val_sku_unit = 0;
 	int channel;
 	struct xe_reg pkg_power_sku_unit;
@@ -866,6 +984,11 @@ xe_hwmon_get_preregistration_info(struct xe_hwmon *hwmon)
 	for (channel = 0; channel < CHANNEL_MAX; channel++)
 		if (xe_hwmon_is_visible(hwmon, hwmon_energy, hwmon_energy_input, channel))
 			xe_hwmon_energy_get(hwmon, channel, &energy);
+
+	/* Initialize 'struct xe_hwmon_fan_info' with initial fan register reading. */
+	for (channel = 0; channel < FAN_MAX; channel++)
+		if (xe_hwmon_is_visible(hwmon, hwmon_fan, hwmon_fan_input, channel))
+			xe_hwmon_fan_input_read(hwmon, channel, &fan_speed);
 }
 
 static void xe_hwmon_mutex_destroy(void *arg)
diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c
index 89393dcb53d9..63db66df064b 100644
--- a/drivers/gpu/drm/xe/xe_lmtt.c
+++ b/drivers/gpu/drm/xe/xe_lmtt.c
@@ -71,7 +71,7 @@ static struct xe_lmtt_pt *lmtt_pt_alloc(struct xe_lmtt *lmtt, unsigned int level
 					     lmtt->ops->lmtt_pte_num(level)),
 				  ttm_bo_type_kernel,
 				  XE_BO_FLAG_VRAM_IF_DGFX(lmtt_to_tile(lmtt)) |
-				  XE_BO_FLAG_NEEDS_64K | XE_BO_FLAG_PINNED);
+				  XE_BO_FLAG_NEEDS_64K);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		goto out_free_pt;
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index df3ceddede07..855c8acaf3f1 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -37,6 +37,7 @@
 #define LRC_ENGINE_CLASS			GENMASK_ULL(63, 61)
 #define LRC_ENGINE_INSTANCE			GENMASK_ULL(53, 48)
 
+#define LRC_PPHWSP_SIZE				SZ_4K
 #define LRC_INDIRECT_RING_STATE_SIZE		SZ_4K
 
 static struct xe_device *
@@ -50,19 +51,22 @@ size_t xe_gt_lrc_size(struct xe_gt *gt, enum xe_engine_class class)
 	struct xe_device *xe = gt_to_xe(gt);
 	size_t size;
 
+	/* Per-process HW status page (PPHWSP) */
+	size = LRC_PPHWSP_SIZE;
+
+	/* Engine context image */
 	switch (class) {
 	case XE_ENGINE_CLASS_RENDER:
 		if (GRAPHICS_VER(xe) >= 20)
-			size = 4 * SZ_4K;
+			size += 3 * SZ_4K;
 		else
-			size = 14 * SZ_4K;
+			size += 13 * SZ_4K;
 		break;
 	case XE_ENGINE_CLASS_COMPUTE:
-		/* 14 pages since graphics_ver == 11 */
 		if (GRAPHICS_VER(xe) >= 20)
-			size = 3 * SZ_4K;
+			size += 2 * SZ_4K;
 		else
-			size = 14 * SZ_4K;
+			size += 13 * SZ_4K;
 		break;
 	default:
 		WARN(1, "Unknown engine class: %d", class);
@@ -71,7 +75,7 @@ size_t xe_gt_lrc_size(struct xe_gt *gt, enum xe_engine_class class)
 	case XE_ENGINE_CLASS_VIDEO_DECODE:
 	case XE_ENGINE_CLASS_VIDEO_ENHANCE:
 	case XE_ENGINE_CLASS_OTHER:
-		size = 2 * SZ_4K;
+		size += 1 * SZ_4K;
 	}
 
 	/* Add indirect ring state page */
@@ -650,7 +654,6 @@ u32 xe_lrc_pphwsp_offset(struct xe_lrc *lrc)
 #define LRC_START_SEQNO_PPHWSP_OFFSET (LRC_SEQNO_PPHWSP_OFFSET + 8)
 #define LRC_CTX_JOB_TIMESTAMP_OFFSET (LRC_START_SEQNO_PPHWSP_OFFSET + 8)
 #define LRC_PARALLEL_PPHWSP_OFFSET 2048
-#define LRC_PPHWSP_SIZE SZ_4K
 
 u32 xe_lrc_regs_offset(struct xe_lrc *lrc)
 {
@@ -893,6 +896,7 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 	void *init_data = NULL;
 	u32 arb_enable;
 	u32 lrc_size;
+	u32 bo_flags;
 	int err;
 
 	kref_init(&lrc->refcount);
@@ -901,15 +905,18 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 	if (xe_gt_has_indirect_ring_state(gt))
 		lrc->flags |= XE_LRC_FLAG_INDIRECT_RING_STATE;
 
+	bo_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile) | XE_BO_FLAG_GGTT |
+		   XE_BO_FLAG_GGTT_INVALIDATE;
+	if (vm && vm->xef) /* userspace */
+		bo_flags |= XE_BO_FLAG_PINNED_LATE_RESTORE;
+
 	/*
 	 * FIXME: Perma-pinning LRC as we don't yet support moving GGTT address
 	 * via VM bind calls.
 	 */
 	lrc->bo = xe_bo_create_pin_map(xe, tile, vm, lrc_size,
 				       ttm_bo_type_kernel,
-				       XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				       XE_BO_FLAG_GGTT |
-				       XE_BO_FLAG_GGTT_INVALIDATE);
+				       bo_flags);
 	if (IS_ERR(lrc->bo))
 		return PTR_ERR(lrc->bo);
 
@@ -1445,6 +1452,7 @@ static int dump_gfxpipe_command(struct drm_printer *p,
 	MATCH3D(3DSTATE_CLIP_MESH);
 	MATCH3D(3DSTATE_SBE_MESH);
 	MATCH3D(3DSTATE_CPSIZE_CONTROL_BUFFER);
+	MATCH3D(3DSTATE_COARSE_PIXEL);
 
 	MATCH3D(3DSTATE_DRAWING_RECTANGLE);
 	MATCH3D(3DSTATE_CHROMA_KEY);
diff --git a/drivers/gpu/drm/xe/xe_memirq.c b/drivers/gpu/drm/xe/xe_memirq.c
index 404fa2a456d5..49c45ec3e83c 100644
--- a/drivers/gpu/drm/xe/xe_memirq.c
+++ b/drivers/gpu/drm/xe/xe_memirq.c
@@ -86,7 +86,7 @@ static const char *guc_name(struct xe_guc *guc)
  *   This object needs to be 4KiB aligned.
  *
  * - _`Interrupt Source Report Page`: this is the equivalent of the
- *   GEN11_GT_INTR_DWx registers, with each bit in those registers being
+ *   GT_INTR_DWx registers, with each bit in those registers being
  *   mapped to a byte here. The offsets are the same, just bytes instead
  *   of bits. This object needs to be cacheline aligned.
  *
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index df4282c71bf0..8f8e9fdfb2a8 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -97,7 +97,7 @@ struct xe_exec_queue *xe_tile_migrate_exec_queue(struct xe_tile *tile)
 	return tile->migrate->q;
 }
 
-static void xe_migrate_fini(struct drm_device *dev, void *arg)
+static void xe_migrate_fini(void *arg)
 {
 	struct xe_migrate *m = arg;
 
@@ -209,7 +209,6 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 				  num_entries * XE_PAGE_SIZE,
 				  ttm_bo_type_kernel,
 				  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				  XE_BO_FLAG_PINNED |
 				  XE_BO_FLAG_PAGETABLE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
@@ -401,7 +400,7 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 	struct xe_vm *vm;
 	int err;
 
-	m = drmm_kzalloc(&xe->drm, sizeof(*m), GFP_KERNEL);
+	m = devm_kzalloc(xe->drm.dev, sizeof(*m), GFP_KERNEL);
 	if (!m)
 		return ERR_PTR(-ENOMEM);
 
@@ -455,7 +454,7 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 	might_lock(&m->job_mutex);
 	fs_reclaim_release(GFP_KERNEL);
 
-	err = drmm_add_action_or_reset(&xe->drm, xe_migrate_fini, m);
+	err = devm_add_action_or_reset(xe->drm.dev, xe_migrate_fini, m);
 	if (err)
 		return ERR_PTR(err);
 
@@ -670,6 +669,7 @@ static void emit_copy(struct xe_gt *gt, struct xe_bb *bb,
 	u32 mocs = 0;
 	u32 tile_y = 0;
 
+	xe_gt_assert(gt, !(pitch & 3));
 	xe_gt_assert(gt, size / pitch <= S16_MAX);
 	xe_gt_assert(gt, pitch / 4 <= S16_MAX);
 	xe_gt_assert(gt, pitch <= U16_MAX);
@@ -779,10 +779,12 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 	bool dst_is_pltt = dst->mem_type == XE_PL_TT;
 	bool src_is_vram = mem_type_is_vram(src->mem_type);
 	bool dst_is_vram = mem_type_is_vram(dst->mem_type);
+	bool type_device = src_bo->ttm.type == ttm_bo_type_device;
+	bool needs_ccs_emit = type_device && xe_migrate_needs_ccs_emit(xe);
 	bool copy_ccs = xe_device_has_flat_ccs(xe) &&
 		xe_bo_needs_ccs_pages(src_bo) && xe_bo_needs_ccs_pages(dst_bo);
 	bool copy_system_ccs = copy_ccs && (!src_is_vram || !dst_is_vram);
-	bool use_comp_pat = xe_device_has_flat_ccs(xe) &&
+	bool use_comp_pat = type_device && xe_device_has_flat_ccs(xe) &&
 		GRAPHICS_VER(xe) >= 20 && src_is_vram && !dst_is_vram;
 
 	/* Copying CCS between two different BOs is not supported yet. */
@@ -839,6 +841,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 					      avail_pts, avail_pts);
 
 		if (copy_system_ccs) {
+			xe_assert(xe, type_device);
 			ccs_size = xe_device_ccs_bytes(xe, src_L0);
 			batch_size += pte_update_size(m, 0, NULL, &ccs_it, &ccs_size,
 						      &ccs_ofs, &ccs_pt, 0,
@@ -849,7 +852,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 
 		/* Add copy commands size here */
 		batch_size += ((copy_only_ccs) ? 0 : EMIT_COPY_DW) +
-			((xe_migrate_needs_ccs_emit(xe) ? EMIT_COPY_CCS_DW : 0));
+			((needs_ccs_emit ? EMIT_COPY_CCS_DW : 0));
 
 		bb = xe_bb_new(gt, batch_size, usm);
 		if (IS_ERR(bb)) {
@@ -878,7 +881,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 		if (!copy_only_ccs)
 			emit_copy(gt, bb, src_L0_ofs, dst_L0_ofs, src_L0, XE_PAGE_SIZE);
 
-		if (xe_migrate_needs_ccs_emit(xe))
+		if (needs_ccs_emit)
 			flush_flags = xe_migrate_ccs_copy(m, bb, src_L0_ofs,
 							  IS_DGFX(xe) ? src_is_vram : src_is_pltt,
 							  dst_L0_ofs,
@@ -1177,7 +1180,7 @@ err:
 err_sync:
 		/* Sync partial copies if any. FIXME: job_mutex? */
 		if (fence) {
-			dma_fence_wait(m->fence, false);
+			dma_fence_wait(fence, false);
 			dma_fence_put(fence);
 		}
 
@@ -1547,7 +1550,7 @@ void xe_migrate_wait(struct xe_migrate *m)
 static u32 pte_update_cmd_size(u64 size)
 {
 	u32 num_dword;
-	u64 entries = DIV_ROUND_UP(size, XE_PAGE_SIZE);
+	u64 entries = DIV_U64_ROUND_UP(size, XE_PAGE_SIZE);
 
 	XE_WARN_ON(size > MAX_PREEMPTDISABLE_TRANSFER);
 	/*
@@ -1558,7 +1561,7 @@ static u32 pte_update_cmd_size(u64 size)
 	 * 2 dword for the page table's physical location
 	 * 2*n dword for value of pte to fill (each pte entry is 2 dwords)
 	 */
-	num_dword = (1 + 2) * DIV_ROUND_UP(entries, 0x1ff);
+	num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, 0x1ff);
 	num_dword += entries * 2;
 
 	return num_dword;
@@ -1601,55 +1604,63 @@ enum xe_migrate_copy_dir {
 	XE_MIGRATE_COPY_TO_SRAM,
 };
 
+#define XE_CACHELINE_BYTES	64ull
+#define XE_CACHELINE_MASK	(XE_CACHELINE_BYTES - 1)
+
 static struct dma_fence *xe_migrate_vram(struct xe_migrate *m,
-					 unsigned long npages,
+					 unsigned long len,
+					 unsigned long sram_offset,
 					 dma_addr_t *sram_addr, u64 vram_addr,
 					 const enum xe_migrate_copy_dir dir)
 {
 	struct xe_gt *gt = m->tile->primary_gt;
 	struct xe_device *xe = gt_to_xe(gt);
+	bool use_usm_batch = xe->info.has_usm;
 	struct dma_fence *fence = NULL;
 	u32 batch_size = 2;
 	u64 src_L0_ofs, dst_L0_ofs;
-	u64 round_update_size;
 	struct xe_sched_job *job;
 	struct xe_bb *bb;
 	u32 update_idx, pt_slot = 0;
+	unsigned long npages = DIV_ROUND_UP(len + sram_offset, PAGE_SIZE);
+	unsigned int pitch = len >= PAGE_SIZE && !(len & ~PAGE_MASK) ?
+		PAGE_SIZE : 4;
 	int err;
 
-	if (npages * PAGE_SIZE > MAX_PREEMPTDISABLE_TRANSFER)
-		return ERR_PTR(-EINVAL);
+	if (drm_WARN_ON(&xe->drm, (len & XE_CACHELINE_MASK) ||
+			(sram_offset | vram_addr) & XE_CACHELINE_MASK))
+		return ERR_PTR(-EOPNOTSUPP);
 
-	round_update_size = npages * PAGE_SIZE;
-	batch_size += pte_update_cmd_size(round_update_size);
+	xe_assert(xe, npages * PAGE_SIZE <= MAX_PREEMPTDISABLE_TRANSFER);
+
+	batch_size += pte_update_cmd_size(len);
 	batch_size += EMIT_COPY_DW;
 
-	bb = xe_bb_new(gt, batch_size, true);
+	bb = xe_bb_new(gt, batch_size, use_usm_batch);
 	if (IS_ERR(bb)) {
 		err = PTR_ERR(bb);
 		return ERR_PTR(err);
 	}
 
 	build_pt_update_batch_sram(m, bb, pt_slot * XE_PAGE_SIZE,
-				   sram_addr, round_update_size);
+				   sram_addr, len + sram_offset);
 
 	if (dir == XE_MIGRATE_COPY_TO_VRAM) {
-		src_L0_ofs = xe_migrate_vm_addr(pt_slot, 0);
+		src_L0_ofs = xe_migrate_vm_addr(pt_slot, 0) + sram_offset;
 		dst_L0_ofs = xe_migrate_vram_ofs(xe, vram_addr, false);
 
 	} else {
 		src_L0_ofs = xe_migrate_vram_ofs(xe, vram_addr, false);
-		dst_L0_ofs = xe_migrate_vm_addr(pt_slot, 0);
+		dst_L0_ofs = xe_migrate_vm_addr(pt_slot, 0) + sram_offset;
 	}
 
 	bb->cs[bb->len++] = MI_BATCH_BUFFER_END;
 	update_idx = bb->len;
 
-	emit_copy(gt, bb, src_L0_ofs, dst_L0_ofs, round_update_size,
-		  XE_PAGE_SIZE);
+	emit_copy(gt, bb, src_L0_ofs, dst_L0_ofs, len, pitch);
 
 	job = xe_bb_create_migration_job(m->q, bb,
-					 xe_migrate_batch_base(m, true),
+					 xe_migrate_batch_base(m, use_usm_batch),
 					 update_idx);
 	if (IS_ERR(job)) {
 		err = PTR_ERR(job);
@@ -1694,7 +1705,7 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
 				     dma_addr_t *src_addr,
 				     u64 dst_addr)
 {
-	return xe_migrate_vram(m, npages, src_addr, dst_addr,
+	return xe_migrate_vram(m, npages * PAGE_SIZE, 0, src_addr, dst_addr,
 			       XE_MIGRATE_COPY_TO_VRAM);
 }
 
@@ -1715,10 +1726,193 @@ struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m,
 				       u64 src_addr,
 				       dma_addr_t *dst_addr)
 {
-	return xe_migrate_vram(m, npages, dst_addr, src_addr,
+	return xe_migrate_vram(m, npages * PAGE_SIZE, 0, dst_addr, src_addr,
 			       XE_MIGRATE_COPY_TO_SRAM);
 }
 
+static void xe_migrate_dma_unmap(struct xe_device *xe, dma_addr_t *dma_addr,
+				 int len, int write)
+{
+	unsigned long i, npages = DIV_ROUND_UP(len, PAGE_SIZE);
+
+	for (i = 0; i < npages; ++i) {
+		if (!dma_addr[i])
+			break;
+
+		dma_unmap_page(xe->drm.dev, dma_addr[i], PAGE_SIZE,
+			       write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
+	}
+	kfree(dma_addr);
+}
+
+static dma_addr_t *xe_migrate_dma_map(struct xe_device *xe,
+				      void *buf, int len, int write)
+{
+	dma_addr_t *dma_addr;
+	unsigned long i, npages = DIV_ROUND_UP(len, PAGE_SIZE);
+
+	dma_addr = kcalloc(npages, sizeof(*dma_addr), GFP_KERNEL);
+	if (!dma_addr)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < npages; ++i) {
+		dma_addr_t addr;
+		struct page *page;
+
+		if (is_vmalloc_addr(buf))
+			page = vmalloc_to_page(buf);
+		else
+			page = virt_to_page(buf);
+
+		addr = dma_map_page(xe->drm.dev,
+				    page, 0, PAGE_SIZE,
+				    write ? DMA_TO_DEVICE :
+				    DMA_FROM_DEVICE);
+		if (dma_mapping_error(xe->drm.dev, addr))
+			goto err_fault;
+
+		dma_addr[i] = addr;
+		buf += PAGE_SIZE;
+	}
+
+	return dma_addr;
+
+err_fault:
+	xe_migrate_dma_unmap(xe, dma_addr, len, write);
+	return ERR_PTR(-EFAULT);
+}
+
+/**
+ * xe_migrate_access_memory - Access memory of a BO via GPU
+ *
+ * @m: The migration context.
+ * @bo: buffer object
+ * @offset: access offset into buffer object
+ * @buf: pointer to caller memory to read into or write from
+ * @len: length of access
+ * @write: write access
+ *
+ * Access memory of a BO via GPU either reading in or writing from a passed in
+ * pointer. Pointer is dma mapped for GPU access and GPU commands are issued to
+ * read to or write from pointer.
+ *
+ * Returns:
+ * 0 if successful, negative error code on failure.
+ */
+int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
+			     unsigned long offset, void *buf, int len,
+			     int write)
+{
+	struct xe_tile *tile = m->tile;
+	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_res_cursor cursor;
+	struct dma_fence *fence = NULL;
+	dma_addr_t *dma_addr;
+	unsigned long page_offset = (unsigned long)buf & ~PAGE_MASK;
+	int bytes_left = len, current_page = 0;
+	void *orig_buf = buf;
+
+	xe_bo_assert_held(bo);
+
+	/* Use bounce buffer for small access and unaligned access */
+	if (len & XE_CACHELINE_MASK ||
+	    ((uintptr_t)buf | offset) & XE_CACHELINE_MASK) {
+		int buf_offset = 0;
+
+		/*
+		 * Less than ideal for large unaligned access but this should be
+		 * fairly rare, can fixup if this becomes common.
+		 */
+		do {
+			u8 bounce[XE_CACHELINE_BYTES];
+			void *ptr = (void *)bounce;
+			int err;
+			int copy_bytes = min_t(int, bytes_left,
+					       XE_CACHELINE_BYTES -
+					       (offset & XE_CACHELINE_MASK));
+			int ptr_offset = offset & XE_CACHELINE_MASK;
+
+			err = xe_migrate_access_memory(m, bo,
+						       offset &
+						       ~XE_CACHELINE_MASK,
+						       (void *)ptr,
+						       sizeof(bounce), 0);
+			if (err)
+				return err;
+
+			if (write) {
+				memcpy(ptr + ptr_offset, buf + buf_offset, copy_bytes);
+
+				err = xe_migrate_access_memory(m, bo,
+							       offset & ~XE_CACHELINE_MASK,
+							       (void *)ptr,
+							       sizeof(bounce), 0);
+				if (err)
+					return err;
+			} else {
+				memcpy(buf + buf_offset, ptr + ptr_offset,
+				       copy_bytes);
+			}
+
+			bytes_left -= copy_bytes;
+			buf_offset += copy_bytes;
+			offset += copy_bytes;
+		} while (bytes_left);
+
+		return 0;
+	}
+
+	dma_addr = xe_migrate_dma_map(xe, buf, len + page_offset, write);
+	if (IS_ERR(dma_addr))
+		return PTR_ERR(dma_addr);
+
+	xe_res_first(bo->ttm.resource, offset, bo->size - offset, &cursor);
+
+	do {
+		struct dma_fence *__fence;
+		u64 vram_addr = vram_region_gpu_offset(bo->ttm.resource) +
+			cursor.start;
+		int current_bytes;
+
+		if (cursor.size > MAX_PREEMPTDISABLE_TRANSFER)
+			current_bytes = min_t(int, bytes_left,
+					      MAX_PREEMPTDISABLE_TRANSFER);
+		else
+			current_bytes = min_t(int, bytes_left, cursor.size);
+
+		if (fence)
+			dma_fence_put(fence);
+
+		__fence = xe_migrate_vram(m, current_bytes,
+					  (unsigned long)buf & ~PAGE_MASK,
+					  dma_addr + current_page,
+					  vram_addr, write ?
+					  XE_MIGRATE_COPY_TO_VRAM :
+					  XE_MIGRATE_COPY_TO_SRAM);
+		if (IS_ERR(__fence)) {
+			if (fence)
+				dma_fence_wait(fence, false);
+			fence = __fence;
+			goto out_err;
+		}
+		fence = __fence;
+
+		buf += current_bytes;
+		offset += current_bytes;
+		current_page = (int)(buf - orig_buf) / PAGE_SIZE;
+		bytes_left -= current_bytes;
+		if (bytes_left)
+			xe_res_next(&cursor, current_bytes);
+	} while (bytes_left);
+
+	dma_fence_wait(fence, false);
+	dma_fence_put(fence);
+
+out_err:
+	xe_migrate_dma_unmap(xe, dma_addr, len + page_offset, write);
+	return IS_ERR(fence) ? PTR_ERR(fence) : 0;
+}
+
 #if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST)
 #include "tests/xe_migrate.c"
 #endif
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 6ff9a963425c..fb9839c1bae0 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -112,6 +112,10 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 				  struct ttm_resource *dst,
 				  bool copy_only_ccs);
 
+int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
+			     unsigned long offset, void *buf, int len,
+			     int write);
+
 #define XE_MIGRATE_CLEAR_FLAG_BO_DATA		BIT(0)
 #define XE_MIGRATE_CLEAR_FLAG_CCS_DATA		BIT(1)
 #define XE_MIGRATE_CLEAR_FLAG_FULL	(XE_MIGRATE_CLEAR_FLAG_BO_DATA | \
diff --git a/drivers/gpu/drm/xe/xe_mmio.c b/drivers/gpu/drm/xe/xe_mmio.c
index 70a36e777546..096c38cc51c8 100644
--- a/drivers/gpu/drm/xe/xe_mmio.c
+++ b/drivers/gpu/drm/xe/xe_mmio.c
@@ -138,6 +138,7 @@ int xe_mmio_probe_early(struct xe_device *xe)
 
 	return devm_add_action_or_reset(xe->drm.dev, mmio_fini, xe);
 }
+ALLOW_ERROR_INJECTION(xe_mmio_probe_early, ERRNO); /* See xe_pci_probe() */
 
 /**
  * xe_mmio_init() - Initialize an MMIO instance
@@ -204,8 +205,9 @@ void xe_mmio_write32(struct xe_mmio *mmio, struct xe_reg reg, u32 val)
 
 	trace_xe_reg_rw(mmio, true, addr, val, sizeof(val));
 
-	if (!reg.vf && mmio->sriov_vf_gt)
-		xe_gt_sriov_vf_write32(mmio->sriov_vf_gt, reg, val);
+	if (!reg.vf && IS_SRIOV_VF(mmio->tile->xe))
+		xe_gt_sriov_vf_write32(mmio->sriov_vf_gt ?:
+				       mmio->tile->primary_gt, reg, val);
 	else
 		writel(val, mmio->regs + addr);
 }
@@ -218,8 +220,9 @@ u32 xe_mmio_read32(struct xe_mmio *mmio, struct xe_reg reg)
 	/* Wa_15015404425 */
 	mmio_flush_pending_writes(mmio);
 
-	if (!reg.vf && mmio->sriov_vf_gt)
-		val = xe_gt_sriov_vf_read32(mmio->sriov_vf_gt, reg);
+	if (!reg.vf && IS_SRIOV_VF(mmio->tile->xe))
+		val = xe_gt_sriov_vf_read32(mmio->sriov_vf_gt ?:
+					    mmio->tile->primary_gt, reg);
 	else
 		val = readl(mmio->regs + addr);
 
diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c
index 9f4632e39a1a..64bf46646544 100644
--- a/drivers/gpu/drm/xe/xe_module.c
+++ b/drivers/gpu/drm/xe/xe_module.c
@@ -11,6 +11,7 @@
 #include <drm/drm_module.h>
 
 #include "xe_drv.h"
+#include "xe_configfs.h"
 #include "xe_hw_fence.h"
 #include "xe_pci.h"
 #include "xe_pm.h"
@@ -38,8 +39,8 @@ MODULE_PARM_DESC(force_execlist, "Force Execlist submission");
 module_param_named(probe_display, xe_modparam.probe_display, bool, 0444);
 MODULE_PARM_DESC(probe_display, "Probe display HW, otherwise it's left untouched (default: true)");
 
-module_param_named(vram_bar_size, xe_modparam.force_vram_bar_size, uint, 0600);
-MODULE_PARM_DESC(vram_bar_size, "Set the vram bar size(in MiB)");
+module_param_named(vram_bar_size, xe_modparam.force_vram_bar_size, int, 0600);
+MODULE_PARM_DESC(vram_bar_size, "Set the vram bar size (in MiB) - <0=disable-resize, 0=max-needed-size[default], >0=force-size");
 
 module_param_named(guc_log_level, xe_modparam.guc_log_level, int, 0600);
 MODULE_PARM_DESC(guc_log_level, "GuC firmware logging level (0=disable, 1..5=enable with verbosity min..max)");
@@ -89,6 +90,10 @@ static const struct init_funcs init_funcs[] = {
 		.init = xe_check_nomodeset,
 	},
 	{
+		.init = xe_configfs_init,
+		.exit = xe_configfs_exit,
+	},
+	{
 		.init = xe_hw_fence_module_init,
 		.exit = xe_hw_fence_module_exit,
 	},
diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c
index 7ffc98f67e69..fb842fa0552e 100644
--- a/drivers/gpu/drm/xe/xe_oa.c
+++ b/drivers/gpu/drm/xe/xe_oa.c
@@ -1301,7 +1301,7 @@ static int xe_oa_user_ext_set_property(struct xe_oa *oa, enum xe_oa_user_extn_fr
 	int err;
 	u32 idx;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(oa->xe, err))
 		return -EFAULT;
 
@@ -1338,7 +1338,7 @@ static int xe_oa_user_extensions(struct xe_oa *oa, enum xe_oa_user_extn_from fro
 	if (XE_IOCTL_DBG(oa->xe, ext_number >= MAX_USER_EXTENSIONS))
 		return -E2BIG;
 
-	err = __copy_from_user(&ext, address, sizeof(ext));
+	err = copy_from_user(&ext, address, sizeof(ext));
 	if (XE_IOCTL_DBG(oa->xe, err))
 		return -EFAULT;
 
@@ -2221,6 +2221,7 @@ addr_err:
 	kfree(oa_regs);
 	return ERR_PTR(err);
 }
+ALLOW_ERROR_INJECTION(xe_oa_alloc_regs, ERRNO);
 
 static ssize_t show_dynamic_id(struct kobject *kobj,
 			       struct kobj_attribute *attr,
@@ -2280,7 +2281,7 @@ int xe_oa_add_config_ioctl(struct drm_device *dev, u64 data, struct drm_file *fi
 		return -EACCES;
 	}
 
-	err = __copy_from_user(&param, u64_to_user_ptr(data), sizeof(param));
+	err = copy_from_user(&param, u64_to_user_ptr(data), sizeof(param));
 	if (XE_IOCTL_DBG(oa->xe, err))
 		return -EFAULT;
 
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 818f023166d5..882398e09b7e 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -62,11 +62,13 @@ struct xe_device_desc {
 	u8 is_dgfx:1;
 
 	u8 has_display:1;
+	u8 has_fan_control:1;
 	u8 has_heci_gscfi:1;
 	u8 has_heci_cscfi:1;
 	u8 has_llc:1;
 	u8 has_pxp:1;
 	u8 has_sriov:1;
+	u8 needs_scratch:1;
 	u8 skip_guc_pc:1;
 	u8 skip_mtcfg:1;
 	u8 skip_pcode:1;
@@ -302,6 +304,7 @@ static const struct xe_device_desc dg2_desc = {
 
 	DG2_FEATURES,
 	.has_display = true,
+	.has_fan_control = true,
 };
 
 static const __maybe_unused struct xe_device_desc pvc_desc = {
@@ -329,6 +332,7 @@ static const struct xe_device_desc lnl_desc = {
 	.dma_mask_size = 46,
 	.has_display = true,
 	.has_pxp = true,
+	.needs_scratch = true,
 };
 
 static const struct xe_device_desc bmg_desc = {
@@ -336,7 +340,9 @@ static const struct xe_device_desc bmg_desc = {
 	PLATFORM(BATTLEMAGE),
 	.dma_mask_size = 46,
 	.has_display = true,
+	.has_fan_control = true,
 	.has_heci_cscfi = 1,
+	.needs_scratch = true,
 };
 
 static const struct xe_device_desc ptl_desc = {
@@ -345,6 +351,7 @@ static const struct xe_device_desc ptl_desc = {
 	.has_display = true,
 	.has_sriov = true,
 	.require_force_probe = true,
+	.needs_scratch = true,
 };
 
 #undef PLATFORM
@@ -575,6 +582,7 @@ static int xe_info_init_early(struct xe_device *xe,
 
 	xe->info.dma_mask_size = desc->dma_mask_size;
 	xe->info.is_dgfx = desc->is_dgfx;
+	xe->info.has_fan_control = desc->has_fan_control;
 	xe->info.has_heci_gscfi = desc->has_heci_gscfi;
 	xe->info.has_heci_cscfi = desc->has_heci_cscfi;
 	xe->info.has_llc = desc->has_llc;
@@ -583,6 +591,7 @@ static int xe_info_init_early(struct xe_device *xe,
 	xe->info.skip_guc_pc = desc->skip_guc_pc;
 	xe->info.skip_mtcfg = desc->skip_mtcfg;
 	xe->info.skip_pcode = desc->skip_pcode;
+	xe->info.needs_scratch = desc->needs_scratch;
 
 	xe->info.probe_display = IS_ENABLED(CONFIG_DRM_XE_DISPLAY) &&
 				 xe_modparam.probe_display &&
@@ -733,7 +742,7 @@ static void xe_pci_remove(struct pci_dev *pdev)
 		return;
 
 	xe_device_remove(xe);
-	xe_pm_runtime_fini(xe);
+	xe_pm_fini(xe);
 }
 
 /*
@@ -803,18 +812,17 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return err;
 
 	err = xe_device_probe_early(xe);
-	if (err) {
-		/*
-		 * In Boot Survivability mode, no drm card is exposed and driver
-		 * is loaded with bare minimum to allow for firmware to be
-		 * flashed through mei. If early probe failed, but it managed to
-		 * enable survivability mode, return success.
-		 */
-		if (xe_survivability_mode_is_enabled(xe))
-			return 0;
+	/*
+	 * In Boot Survivability mode, no drm card is exposed and driver
+	 * is loaded with bare minimum to allow for firmware to be
+	 * flashed through mei. Return success, if survivability mode
+	 * is enabled due to pcode failure or configfs being set
+	 */
+	if (xe_survivability_mode_is_enabled(xe))
+		return 0;
 
+	if (err)
 		return err;
-	}
 
 	err = xe_info_init(xe, desc);
 	if (err)
@@ -920,6 +928,7 @@ static int xe_pci_suspend(struct device *dev)
 
 	pci_save_state(pdev);
 	pci_disable_device(pdev);
+	pci_set_power_state(pdev, PCI_D3cold);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/xe/xe_pci_sriov.c b/drivers/gpu/drm/xe/xe_pci_sriov.c
index 09ee8a06fe2e..8813efdcafbb 100644
--- a/drivers/gpu/drm/xe/xe_pci_sriov.c
+++ b/drivers/gpu/drm/xe/xe_pci_sriov.c
@@ -7,6 +7,8 @@
 #include "xe_device.h"
 #include "xe_gt_sriov_pf_config.h"
 #include "xe_gt_sriov_pf_control.h"
+#include "xe_gt_sriov_printk.h"
+#include "xe_guc_engine_activity.h"
 #include "xe_pci_sriov.h"
 #include "xe_pm.h"
 #include "xe_sriov.h"
@@ -111,6 +113,20 @@ static void pf_link_vfs(struct xe_device *xe, int num_vfs)
 	}
 }
 
+static void pf_engine_activity_stats(struct xe_device *xe, unsigned int num_vfs, bool enable)
+{
+	struct xe_gt *gt;
+	unsigned int id;
+	int ret = 0;
+
+	for_each_gt(gt, xe, id) {
+		ret = xe_guc_engine_activity_function_stats(&gt->uc.guc, num_vfs, enable);
+		if (ret)
+			xe_gt_sriov_info(gt, "Failed to %s engine activity function stats (%pe)\n",
+					 str_enable_disable(enable), ERR_PTR(ret));
+	}
+}
+
 static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
 {
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
@@ -145,6 +161,9 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
 
 	xe_sriov_info(xe, "Enabled %u of %u VF%s\n",
 		      num_vfs, total_vfs, str_plural(total_vfs));
+
+	pf_engine_activity_stats(xe, num_vfs, true);
+
 	return num_vfs;
 
 failed:
@@ -168,6 +187,8 @@ static int pf_disable_vfs(struct xe_device *xe)
 	if (!num_vfs)
 		return 0;
 
+	pf_engine_activity_stats(xe, num_vfs, false);
+
 	pci_disable_sriov(pdev);
 
 	pf_reset_vfs(xe, num_vfs);
diff --git a/drivers/gpu/drm/xe/xe_pcode.c b/drivers/gpu/drm/xe/xe_pcode.c
index 9333ce776a6e..cf955b3ed52c 100644
--- a/drivers/gpu/drm/xe/xe_pcode.c
+++ b/drivers/gpu/drm/xe/xe_pcode.c
@@ -7,6 +7,7 @@
 
 #include <linux/delay.h>
 #include <linux/errno.h>
+#include <linux/error-injection.h>
 
 #include <drm/drm_managed.h>
 
@@ -323,3 +324,4 @@ int xe_pcode_probe_early(struct xe_device *xe)
 {
 	return xe_pcode_ready(xe, false);
 }
+ALLOW_ERROR_INJECTION(xe_pcode_probe_early, ERRNO); /* See xe_pci_probe */
diff --git a/drivers/gpu/drm/xe/xe_pcode_api.h b/drivers/gpu/drm/xe/xe_pcode_api.h
index 2bae9afdbd35..127d4d26c4cf 100644
--- a/drivers/gpu/drm/xe/xe_pcode_api.h
+++ b/drivers/gpu/drm/xe/xe_pcode_api.h
@@ -34,6 +34,7 @@
 #define   DGFX_PCODE_STATUS		0x7E
 #define     DGFX_GET_INIT_STATUS	0x0
 #define     DGFX_INIT_STATUS_COMPLETE	0x1
+#define     DGFX_LINK_DOWNGRADE_STATUS	REG_BIT(31)
 
 #define   PCODE_POWER_SETUP			0x7C
 #define     POWER_SETUP_SUBCOMMAND_READ_I1	0x4
@@ -49,6 +50,9 @@
 /* Domain IDs (param2) */
 #define     PCODE_MBOX_DOMAIN_HBM		0x2
 
+#define   FAN_SPEED_CONTROL			0x7D
+#define     FSC_READ_NUM_FANS			0x4
+
 #define PCODE_SCRATCH(x)		XE_REG(0x138320 + ((x) * 4))
 /* PCODE_SCRATCH0 */
 #define   AUXINFO_REG_OFFSET		REG_GENMASK(17, 15)
@@ -63,6 +67,10 @@
 /* Auxiliary info bits */
 #define   AUXINFO_HISTORY_OFFSET	REG_GENMASK(31, 29)
 
+#define BMG_PCIE_CAP			XE_REG(0x138340)
+#define   LINK_DOWNGRADE		REG_GENMASK(1, 0)
+#define     DOWNGRADE_CAPABLE		2
+
 struct pcode_err_decode {
 	int errno;
 	const char *str;
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index 7b6b754ad6eb..693866def183 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -16,7 +16,6 @@
 #include "xe_bo.h"
 #include "xe_bo_evict.h"
 #include "xe_device.h"
-#include "xe_device_sysfs.h"
 #include "xe_ggtt.h"
 #include "xe_gt.h"
 #include "xe_guc.h"
@@ -188,7 +187,7 @@ int xe_pm_resume(struct xe_device *xe)
 	 * This only restores pinned memory which is the memory required for the
 	 * GT(s) to resume.
 	 */
-	err = xe_bo_restore_kernel(xe);
+	err = xe_bo_restore_early(xe);
 	if (err)
 		goto err;
 
@@ -199,7 +198,7 @@ int xe_pm_resume(struct xe_device *xe)
 
 	xe_display_pm_resume(xe);
 
-	err = xe_bo_restore_user(xe);
+	err = xe_bo_restore_late(xe);
 	if (err)
 		goto err;
 
@@ -273,6 +272,7 @@ int xe_pm_init_early(struct xe_device *xe)
 	if (err)
 		return err;
 
+	xe->d3cold.capable = xe_pm_pci_d3cold_capable(xe);
 	return 0;
 }
 ALLOW_ERROR_INJECTION(xe_pm_init_early, ERRNO); /* See xe_pci_probe() */
@@ -286,6 +286,42 @@ static u32 vram_threshold_value(struct xe_device *xe)
 	return DEFAULT_VRAM_THRESHOLD;
 }
 
+static int xe_pm_notifier_callback(struct notifier_block *nb,
+				   unsigned long action, void *data)
+{
+	struct xe_device *xe = container_of(nb, struct xe_device, pm_notifier);
+	int err = 0;
+
+	switch (action) {
+	case PM_HIBERNATION_PREPARE:
+	case PM_SUSPEND_PREPARE:
+		xe_pm_runtime_get(xe);
+		err = xe_bo_evict_all_user(xe);
+		if (err) {
+			drm_dbg(&xe->drm, "Notifier evict user failed (%d)\n", err);
+			xe_pm_runtime_put(xe);
+			break;
+		}
+
+		err = xe_bo_notifier_prepare_all_pinned(xe);
+		if (err) {
+			drm_dbg(&xe->drm, "Notifier prepare pin failed (%d)\n", err);
+			xe_pm_runtime_put(xe);
+		}
+		break;
+	case PM_POST_HIBERNATION:
+	case PM_POST_SUSPEND:
+		xe_bo_notifier_unprepare_all_pinned(xe);
+		xe_pm_runtime_put(xe);
+		break;
+	}
+
+	if (err)
+		return NOTIFY_BAD;
+
+	return NOTIFY_DONE;
+}
+
 /**
  * xe_pm_init - Initialize Xe Power Management
  * @xe: xe device instance
@@ -299,33 +335,31 @@ int xe_pm_init(struct xe_device *xe)
 	u32 vram_threshold;
 	int err;
 
+	xe->pm_notifier.notifier_call = xe_pm_notifier_callback;
+	err = register_pm_notifier(&xe->pm_notifier);
+	if (err)
+		return err;
+
 	/* For now suspend/resume is only allowed with GuC */
 	if (!xe_device_uc_enabled(xe))
 		return 0;
 
-	xe->d3cold.capable = xe_pm_pci_d3cold_capable(xe);
-
 	if (xe->d3cold.capable) {
-		err = xe_device_sysfs_init(xe);
-		if (err)
-			return err;
-
 		vram_threshold = vram_threshold_value(xe);
 		err = xe_pm_set_vram_threshold(xe, vram_threshold);
 		if (err)
-			return err;
+			goto err_unregister;
 	}
 
 	xe_pm_runtime_init(xe);
-
 	return 0;
+
+err_unregister:
+	unregister_pm_notifier(&xe->pm_notifier);
+	return err;
 }
 
-/**
- * xe_pm_runtime_fini - Finalize Runtime PM
- * @xe: xe device instance
- */
-void xe_pm_runtime_fini(struct xe_device *xe)
+static void xe_pm_runtime_fini(struct xe_device *xe)
 {
 	struct device *dev = xe->drm.dev;
 
@@ -333,6 +367,18 @@ void xe_pm_runtime_fini(struct xe_device *xe)
 	pm_runtime_forbid(dev);
 }
 
+/**
+ * xe_pm_fini - Finalize PM
+ * @xe: xe device instance
+ */
+void xe_pm_fini(struct xe_device *xe)
+{
+	if (xe_device_uc_enabled(xe))
+		xe_pm_runtime_fini(xe);
+
+	unregister_pm_notifier(&xe->pm_notifier);
+}
+
 static void xe_pm_write_callback_task(struct xe_device *xe,
 				      struct task_struct *task)
 {
@@ -484,7 +530,7 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 		 * This only restores pinned memory which is the memory
 		 * required for the GT(s) to resume.
 		 */
-		err = xe_bo_restore_kernel(xe);
+		err = xe_bo_restore_early(xe);
 		if (err)
 			goto out;
 	}
@@ -497,7 +543,7 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 	xe_display_pm_runtime_resume(xe);
 
 	if (xe->d3cold.allowed) {
-		err = xe_bo_restore_user(xe);
+		err = xe_bo_restore_late(xe);
 		if (err)
 			goto out;
 	}
diff --git a/drivers/gpu/drm/xe/xe_pm.h b/drivers/gpu/drm/xe/xe_pm.h
index 998d1ed64556..59678b310e55 100644
--- a/drivers/gpu/drm/xe/xe_pm.h
+++ b/drivers/gpu/drm/xe/xe_pm.h
@@ -17,7 +17,7 @@ int xe_pm_resume(struct xe_device *xe);
 
 int xe_pm_init_early(struct xe_device *xe);
 int xe_pm_init(struct xe_device *xe);
-void xe_pm_runtime_fini(struct xe_device *xe);
+void xe_pm_fini(struct xe_device *xe);
 bool xe_pm_runtime_suspended(struct xe_device *xe);
 int xe_pm_runtime_suspend(struct xe_device *xe);
 int xe_pm_runtime_resume(struct xe_device *xe);
diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c
index 4f62a6e515d6..69df0e3520a5 100644
--- a/drivers/gpu/drm/xe/xe_pmu.c
+++ b/drivers/gpu/drm/xe/xe_pmu.c
@@ -10,9 +10,11 @@
 #include "xe_force_wake.h"
 #include "xe_gt_idle.h"
 #include "xe_guc_engine_activity.h"
+#include "xe_guc_pc.h"
 #include "xe_hw_engine.h"
 #include "xe_pm.h"
 #include "xe_pmu.h"
+#include "xe_sriov_pf_helpers.h"
 
 /**
  * DOC: Xe PMU (Performance Monitoring Unit)
@@ -32,9 +34,10 @@
  *	gt[60:63]		Selects gt for the event
  *	engine_class[20:27]	Selects engine-class for event
  *	engine_instance[12:19]	Selects the engine-instance for the event
+ *	function[44:59]		Selects the function of the event (SRIOV enabled)
  *
  * For engine specific events (engine-*), gt, engine_class and engine_instance parameters must be
- * set as populated by DRM_XE_DEVICE_QUERY_ENGINES.
+ * set as populated by DRM_XE_DEVICE_QUERY_ENGINES and function if SRIOV is enabled.
  *
  * For gt specific events (gt-*) gt parameter must be passed. All other parameters will be 0.
  *
@@ -49,6 +52,7 @@
  */
 
 #define XE_PMU_EVENT_GT_MASK			GENMASK_ULL(63, 60)
+#define XE_PMU_EVENT_FUNCTION_MASK		GENMASK_ULL(59, 44)
 #define XE_PMU_EVENT_ENGINE_CLASS_MASK		GENMASK_ULL(27, 20)
 #define XE_PMU_EVENT_ENGINE_INSTANCE_MASK	GENMASK_ULL(19, 12)
 #define XE_PMU_EVENT_ID_MASK			GENMASK_ULL(11, 0)
@@ -58,6 +62,11 @@ static unsigned int config_to_event_id(u64 config)
 	return FIELD_GET(XE_PMU_EVENT_ID_MASK, config);
 }
 
+static unsigned int config_to_function_id(u64 config)
+{
+	return FIELD_GET(XE_PMU_EVENT_FUNCTION_MASK, config);
+}
+
 static unsigned int config_to_engine_class(u64 config)
 {
 	return FIELD_GET(XE_PMU_EVENT_ENGINE_CLASS_MASK, config);
@@ -76,6 +85,8 @@ static unsigned int config_to_gt_id(u64 config)
 #define XE_PMU_EVENT_GT_C6_RESIDENCY		0x01
 #define XE_PMU_EVENT_ENGINE_ACTIVE_TICKS	0x02
 #define XE_PMU_EVENT_ENGINE_TOTAL_TICKS		0x03
+#define XE_PMU_EVENT_GT_ACTUAL_FREQUENCY	0x04
+#define XE_PMU_EVENT_GT_REQUESTED_FREQUENCY	0x05
 
 static struct xe_gt *event_to_gt(struct perf_event *event)
 {
@@ -111,6 +122,14 @@ static bool is_engine_event(u64 config)
 		event_id == XE_PMU_EVENT_ENGINE_ACTIVE_TICKS);
 }
 
+static bool is_gt_frequency_event(struct perf_event *event)
+{
+	u32 id = config_to_event_id(event->attr.config);
+
+	return id == XE_PMU_EVENT_GT_ACTUAL_FREQUENCY ||
+	       id == XE_PMU_EVENT_GT_REQUESTED_FREQUENCY;
+}
+
 static bool event_gt_forcewake(struct perf_event *event)
 {
 	struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base);
@@ -118,7 +137,7 @@ static bool event_gt_forcewake(struct perf_event *event)
 	struct xe_gt *gt;
 	unsigned int *fw_ref;
 
-	if (!is_engine_event(config))
+	if (!is_engine_event(config) && !is_gt_frequency_event(event))
 		return true;
 
 	gt = xe_device_get_gt(xe, config_to_gt_id(config));
@@ -151,7 +170,7 @@ static bool event_supported(struct xe_pmu *pmu, unsigned int gt,
 static bool event_param_valid(struct perf_event *event)
 {
 	struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base);
-	unsigned int engine_class, engine_instance;
+	unsigned int engine_class, engine_instance, function_id;
 	u64 config = event->attr.config;
 	struct xe_gt *gt;
 
@@ -161,16 +180,28 @@ static bool event_param_valid(struct perf_event *event)
 
 	engine_class = config_to_engine_class(config);
 	engine_instance = config_to_engine_instance(config);
+	function_id = config_to_function_id(config);
 
 	switch (config_to_event_id(config)) {
 	case XE_PMU_EVENT_GT_C6_RESIDENCY:
-		if (engine_class || engine_instance)
+	case XE_PMU_EVENT_GT_ACTUAL_FREQUENCY:
+	case XE_PMU_EVENT_GT_REQUESTED_FREQUENCY:
+		if (engine_class || engine_instance || function_id)
 			return false;
 		break;
 	case XE_PMU_EVENT_ENGINE_ACTIVE_TICKS:
 	case XE_PMU_EVENT_ENGINE_TOTAL_TICKS:
 		if (!event_to_hwe(event))
 			return false;
+
+		/* PF(0) and total vfs when SRIOV is enabled */
+		if (IS_SRIOV_PF(xe)) {
+			if (function_id > xe_sriov_pf_get_totalvfs(xe))
+				return false;
+		} else if (function_id) {
+			return false;
+		}
+
 		break;
 	}
 
@@ -242,13 +273,17 @@ static int xe_pmu_event_init(struct perf_event *event)
 static u64 read_engine_events(struct xe_gt *gt, struct perf_event *event)
 {
 	struct xe_hw_engine *hwe;
-	u64 val = 0;
+	unsigned int function_id;
+	u64 config, val = 0;
+
+	config = event->attr.config;
+	function_id = config_to_function_id(config);
 
 	hwe = event_to_hwe(event);
-	if (config_to_event_id(event->attr.config) == XE_PMU_EVENT_ENGINE_ACTIVE_TICKS)
-		val = xe_guc_engine_activity_active_ticks(&gt->uc.guc, hwe);
+	if (config_to_event_id(config) == XE_PMU_EVENT_ENGINE_ACTIVE_TICKS)
+		val = xe_guc_engine_activity_active_ticks(&gt->uc.guc, hwe, function_id);
 	else
-		val = xe_guc_engine_activity_total_ticks(&gt->uc.guc, hwe);
+		val = xe_guc_engine_activity_total_ticks(&gt->uc.guc, hwe, function_id);
 
 	return val;
 }
@@ -266,6 +301,10 @@ static u64 __xe_pmu_event_read(struct perf_event *event)
 	case XE_PMU_EVENT_ENGINE_ACTIVE_TICKS:
 	case XE_PMU_EVENT_ENGINE_TOTAL_TICKS:
 		return read_engine_events(gt, event);
+	case XE_PMU_EVENT_GT_ACTUAL_FREQUENCY:
+		return xe_guc_pc_get_act_freq(&gt->uc.guc.pc);
+	case XE_PMU_EVENT_GT_REQUESTED_FREQUENCY:
+		return xe_guc_pc_get_cur_freq_fw(&gt->uc.guc.pc);
 	}
 
 	return 0;
@@ -281,7 +320,14 @@ static void xe_pmu_event_update(struct perf_event *event)
 		new = __xe_pmu_event_read(event);
 	} while (!local64_try_cmpxchg(&hwc->prev_count, &prev, new));
 
-	local64_add(new - prev, &event->count);
+	/*
+	 * GT frequency is not a monotonically increasing counter, so add the
+	 * instantaneous value instead.
+	 */
+	if (is_gt_frequency_event(event))
+		local64_add(new, &event->count);
+	else
+		local64_add(new - prev, &event->count);
 }
 
 static void xe_pmu_event_read(struct perf_event *event)
@@ -351,6 +397,7 @@ static void xe_pmu_event_del(struct perf_event *event, int flags)
 }
 
 PMU_FORMAT_ATTR(gt,			"config:60-63");
+PMU_FORMAT_ATTR(function,		"config:44-59");
 PMU_FORMAT_ATTR(engine_class,		"config:20-27");
 PMU_FORMAT_ATTR(engine_instance,	"config:12-19");
 PMU_FORMAT_ATTR(event,			"config:0-11");
@@ -359,6 +406,7 @@ static struct attribute *pmu_format_attrs[] = {
 	&format_attr_event.attr,
 	&format_attr_engine_class.attr,
 	&format_attr_engine_instance.attr,
+	&format_attr_function.attr,
 	&format_attr_gt.attr,
 	NULL,
 };
@@ -419,6 +467,10 @@ static ssize_t event_attr_show(struct device *dev,
 XE_EVENT_ATTR_SIMPLE(gt-c6-residency, gt_c6_residency, XE_PMU_EVENT_GT_C6_RESIDENCY, "ms");
 XE_EVENT_ATTR_NOUNIT(engine-active-ticks, engine_active_ticks, XE_PMU_EVENT_ENGINE_ACTIVE_TICKS);
 XE_EVENT_ATTR_NOUNIT(engine-total-ticks, engine_total_ticks, XE_PMU_EVENT_ENGINE_TOTAL_TICKS);
+XE_EVENT_ATTR_SIMPLE(gt-actual-frequency, gt_actual_frequency,
+		     XE_PMU_EVENT_GT_ACTUAL_FREQUENCY, "MHz");
+XE_EVENT_ATTR_SIMPLE(gt-requested-frequency, gt_requested_frequency,
+		     XE_PMU_EVENT_GT_REQUESTED_FREQUENCY, "MHz");
 
 static struct attribute *pmu_empty_event_attrs[] = {
 	/* Empty - all events are added as groups with .attr_update() */
@@ -434,6 +486,8 @@ static const struct attribute_group *pmu_events_attr_update[] = {
 	&pmu_group_gt_c6_residency,
 	&pmu_group_engine_active_ticks,
 	&pmu_group_engine_total_ticks,
+	&pmu_group_gt_actual_frequency,
+	&pmu_group_gt_requested_frequency,
 	NULL,
 };
 
@@ -442,8 +496,11 @@ static void set_supported_events(struct xe_pmu *pmu)
 	struct xe_device *xe = container_of(pmu, typeof(*xe), pmu);
 	struct xe_gt *gt = xe_device_get_gt(xe, 0);
 
-	if (!xe->info.skip_guc_pc)
+	if (!xe->info.skip_guc_pc) {
 		pmu->supported_events |= BIT_ULL(XE_PMU_EVENT_GT_C6_RESIDENCY);
+		pmu->supported_events |= BIT_ULL(XE_PMU_EVENT_GT_ACTUAL_FREQUENCY);
+		pmu->supported_events |= BIT_ULL(XE_PMU_EVENT_GT_REQUESTED_FREQUENCY);
+	}
 
 	if (xe_guc_engine_activity_supported(&gt->uc.guc)) {
 		pmu->supported_events |= BIT_ULL(XE_PMU_EVENT_ENGINE_ACTIVE_TICKS);
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index ffaf0d02dc7d..b42cf5d1b20c 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -103,6 +103,7 @@ struct xe_pt *xe_pt_create(struct xe_vm *vm, struct xe_tile *tile,
 {
 	struct xe_pt *pt;
 	struct xe_bo *bo;
+	u32 bo_flags;
 	int err;
 
 	if (level) {
@@ -115,14 +116,16 @@ struct xe_pt *xe_pt_create(struct xe_vm *vm, struct xe_tile *tile,
 	if (!pt)
 		return ERR_PTR(-ENOMEM);
 
+	bo_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+		   XE_BO_FLAG_IGNORE_MIN_PAGE_SIZE |
+		   XE_BO_FLAG_NO_RESV_EVICT | XE_BO_FLAG_PAGETABLE;
+	if (vm->xef) /* userspace */
+		bo_flags |= XE_BO_FLAG_PINNED_LATE_RESTORE;
+
 	pt->level = level;
 	bo = xe_bo_create_pin_map(vm->xe, tile, vm, SZ_4K,
 				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				  XE_BO_FLAG_IGNORE_MIN_PAGE_SIZE |
-				  XE_BO_FLAG_PINNED |
-				  XE_BO_FLAG_NO_RESV_EVICT |
-				  XE_BO_FLAG_PAGETABLE);
+				  bo_flags);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		goto err_kfree;
@@ -269,8 +272,11 @@ struct xe_pt_update {
 	bool preexisting;
 };
 
+/**
+ * struct xe_pt_stage_bind_walk - Walk state for the stage_bind walk.
+ */
 struct xe_pt_stage_bind_walk {
-	/** base: The base class. */
+	/** @base: The base class. */
 	struct xe_pt_walk base;
 
 	/* Input parameters for the walk */
@@ -278,15 +284,19 @@ struct xe_pt_stage_bind_walk {
 	struct xe_vm *vm;
 	/** @tile: The tile we're building for. */
 	struct xe_tile *tile;
-	/** @default_pte: PTE flag only template. No address is associated */
-	u64 default_pte;
+	/** @default_vram_pte: PTE flag only template for VRAM. No address is associated */
+	u64 default_vram_pte;
+	/** @default_system_pte: PTE flag only template for System. No address is associated */
+	u64 default_system_pte;
 	/** @dma_offset: DMA offset to add to the PTE. */
 	u64 dma_offset;
 	/**
-	 * @needs_64k: This address range enforces 64K alignment and
-	 * granularity.
+	 * @needs_64K: This address range enforces 64K alignment and
+	 * granularity on VRAM.
 	 */
 	bool needs_64K;
+	/** @clear_pt: clear page table entries during the bind walk */
+	bool clear_pt;
 	/**
 	 * @vma: VMA being mapped
 	 */
@@ -299,6 +309,7 @@ struct xe_pt_stage_bind_walk {
 	u64 va_curs_start;
 
 	/* Output */
+	/** @wupd: Walk output data for page-table updates. */
 	struct xe_walk_update {
 		/** @wupd.entries: Caller provided storage. */
 		struct xe_vm_pgtable_update *entries;
@@ -316,7 +327,7 @@ struct xe_pt_stage_bind_walk {
 	u64 l0_end_addr;
 	/** @addr_64K: The start address of the current 64K chunk. */
 	u64 addr_64K;
-	/** @found_64: Whether @add_64K actually points to a 64K chunk. */
+	/** @found_64K: Whether @add_64K actually points to a 64K chunk. */
 	bool found_64K;
 };
 
@@ -436,6 +447,10 @@ static bool xe_pt_hugepte_possible(u64 addr, u64 next, unsigned int level,
 	if (xe_vma_is_null(xe_walk->vma))
 		return true;
 
+	/* if we are clearing page table, no dma addresses*/
+	if (xe_walk->clear_pt)
+		return true;
+
 	/* Is the DMA address huge PTE size aligned? */
 	size = next - addr;
 	dma = addr - xe_walk->va_curs_start + xe_res_dma(xe_walk->curs);
@@ -515,24 +530,35 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset,
 	if (level == 0 || xe_pt_hugepte_possible(addr, next, level, xe_walk)) {
 		struct xe_res_cursor *curs = xe_walk->curs;
 		bool is_null = xe_vma_is_null(xe_walk->vma);
+		bool is_vram = is_null ? false : xe_res_is_vram(curs);
 
 		XE_WARN_ON(xe_walk->va_curs_start != addr);
 
-		pte = vm->pt_ops->pte_encode_vma(is_null ? 0 :
-						 xe_res_dma(curs) + xe_walk->dma_offset,
-						 xe_walk->vma, pat_index, level);
-		pte |= xe_walk->default_pte;
+		if (xe_walk->clear_pt) {
+			pte = 0;
+		} else {
+			pte = vm->pt_ops->pte_encode_vma(is_null ? 0 :
+							 xe_res_dma(curs) +
+							 xe_walk->dma_offset,
+							 xe_walk->vma,
+							 pat_index, level);
+			if (!is_null)
+				pte |= is_vram ? xe_walk->default_vram_pte :
+					xe_walk->default_system_pte;
 
-		/*
-		 * Set the XE_PTE_PS64 hint if possible, otherwise if
-		 * this device *requires* 64K PTE size for VRAM, fail.
-		 */
-		if (level == 0 && !xe_parent->is_compact) {
-			if (xe_pt_is_pte_ps64K(addr, next, xe_walk)) {
-				xe_walk->vma->gpuva.flags |= XE_VMA_PTE_64K;
-				pte |= XE_PTE_PS64;
-			} else if (XE_WARN_ON(xe_walk->needs_64K)) {
-				return -EINVAL;
+			/*
+			 * Set the XE_PTE_PS64 hint if possible, otherwise if
+			 * this device *requires* 64K PTE size for VRAM, fail.
+			 */
+			if (level == 0 && !xe_parent->is_compact) {
+				if (xe_pt_is_pte_ps64K(addr, next, xe_walk)) {
+					xe_walk->vma->gpuva.flags |=
+							XE_VMA_PTE_64K;
+					pte |= XE_PTE_PS64;
+				} else if (XE_WARN_ON(xe_walk->needs_64K &&
+					   is_vram)) {
+					return -EINVAL;
+				}
 			}
 		}
 
@@ -540,7 +566,7 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset,
 		if (unlikely(ret))
 			return ret;
 
-		if (!is_null)
+		if (!is_null && !xe_walk->clear_pt)
 			xe_res_next(curs, next - addr);
 		xe_walk->va_curs_start = next;
 		xe_walk->vma->gpuva.flags |= (XE_VMA_PTE_4K << level);
@@ -603,6 +629,44 @@ static const struct xe_pt_walk_ops xe_pt_stage_bind_ops = {
 	.pt_entry = xe_pt_stage_bind_entry,
 };
 
+/*
+ * Default atomic expectations for different allocation scenarios are as follows:
+ *
+ * 1. Traditional API: When the VM is not in LR mode:
+ *    - Device atomics are expected to function with all allocations.
+ *
+ * 2. Compute/SVM API: When the VM is in LR mode:
+ *    - Device atomics are the default behavior when the bo is placed in a single region.
+ *    - In all other cases device atomics will be disabled with AE=0 until an application
+ *      request differently using a ioctl like madvise.
+ */
+static bool xe_atomic_for_vram(struct xe_vm *vm)
+{
+	return true;
+}
+
+static bool xe_atomic_for_system(struct xe_vm *vm, struct xe_bo *bo)
+{
+	struct xe_device *xe = vm->xe;
+
+	if (!xe->info.has_device_atomics_on_smem)
+		return false;
+
+	/*
+	 * If a SMEM+LMEM allocation is backed by SMEM, a device
+	 * atomics will cause a gpu page fault and which then
+	 * gets migrated to LMEM, bind such allocations with
+	 * device atomics enabled.
+	 *
+	 * TODO: Revisit this. Perhaps add something like a
+	 * fault_on_atomics_in_system UAPI flag.
+	 * Note that this also prohibits GPU atomics in LR mode for
+	 * userptr and system memory on DGFX.
+	 */
+	return (!IS_DGFX(xe) || (!xe_vm_in_lr_mode(vm) ||
+				 (bo && xe_bo_has_single_placement(bo))));
+}
+
 /**
  * xe_pt_stage_bind() - Build a disconnected page-table tree for a given address
  * range.
@@ -612,6 +676,7 @@ static const struct xe_pt_walk_ops xe_pt_stage_bind_ops = {
  * @entries: Storage for the update entries used for connecting the tree to
  * the main tree at commit time.
  * @num_entries: On output contains the number of @entries used.
+ * @clear_pt: Clear the page table entries.
  *
  * This function builds a disconnected page-table tree for a given address
  * range. The tree is connected to the main vm tree for the gpu using
@@ -625,13 +690,13 @@ static const struct xe_pt_walk_ops xe_pt_stage_bind_ops = {
 static int
 xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 		 struct xe_svm_range *range,
-		 struct xe_vm_pgtable_update *entries, u32 *num_entries)
+		 struct xe_vm_pgtable_update *entries,
+		 u32 *num_entries, bool clear_pt)
 {
 	struct xe_device *xe = tile_to_xe(tile);
 	struct xe_bo *bo = xe_vma_bo(vma);
-	bool is_devmem = !xe_vma_is_userptr(vma) && bo &&
-		(xe_bo_is_vram(bo) || xe_bo_is_stolen_devmem(bo));
 	struct xe_res_cursor curs;
+	struct xe_vm *vm = xe_vma_vm(vma);
 	struct xe_pt_stage_bind_walk xe_walk = {
 		.base = {
 			.ops = &xe_pt_stage_bind_ops,
@@ -639,34 +704,31 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 			.max_level = XE_PT_HIGHEST_LEVEL,
 			.staging = true,
 		},
-		.vm = xe_vma_vm(vma),
+		.vm = vm,
 		.tile = tile,
 		.curs = &curs,
 		.va_curs_start = range ? range->base.itree.start :
 			xe_vma_start(vma),
 		.vma = vma,
 		.wupd.entries = entries,
+		.clear_pt = clear_pt,
 	};
-	struct xe_pt *pt = xe_vma_vm(vma)->pt_root[tile->id];
+	struct xe_pt *pt = vm->pt_root[tile->id];
 	int ret;
 
 	if (range) {
 		/* Move this entire thing to xe_svm.c? */
-		xe_svm_notifier_lock(xe_vma_vm(vma));
+		xe_svm_notifier_lock(vm);
 		if (!xe_svm_range_pages_valid(range)) {
 			xe_svm_range_debug(range, "BIND PREPARE - RETRY");
-			xe_svm_notifier_unlock(xe_vma_vm(vma));
+			xe_svm_notifier_unlock(vm);
 			return -EAGAIN;
 		}
 		if (xe_svm_range_has_dma_mapping(range)) {
 			xe_res_first_dma(range->base.dma_addr, 0,
 					 range->base.itree.last + 1 - range->base.itree.start,
 					 &curs);
-			is_devmem = xe_res_is_vram(&curs);
-			if (is_devmem)
-				xe_svm_range_debug(range, "BIND PREPARE - DMA VRAM");
-			else
-				xe_svm_range_debug(range, "BIND PREPARE - DMA");
+			xe_svm_range_debug(range, "BIND PREPARE - MIXED");
 		} else {
 			xe_assert(xe, false);
 		}
@@ -674,54 +736,21 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 		 * Note, when unlocking the resource cursor dma addresses may become
 		 * stale, but the bind will be aborted anyway at commit time.
 		 */
-		xe_svm_notifier_unlock(xe_vma_vm(vma));
+		xe_svm_notifier_unlock(vm);
 	}
 
-	xe_walk.needs_64K = (xe_vma_vm(vma)->flags & XE_VM_FLAG_64K) && is_devmem;
+	xe_walk.needs_64K = (vm->flags & XE_VM_FLAG_64K);
+	if (clear_pt)
+		goto walk_pt;
 
-	/**
-	 * Default atomic expectations for different allocation scenarios are as follows:
-	 *
-	 * 1. Traditional API: When the VM is not in LR mode:
-	 *    - Device atomics are expected to function with all allocations.
-	 *
-	 * 2. Compute/SVM API: When the VM is in LR mode:
-	 *    - Device atomics are the default behavior when the bo is placed in a single region.
-	 *    - In all other cases device atomics will be disabled with AE=0 until an application
-	 *      request differently using a ioctl like madvise.
-	 */
 	if (vma->gpuva.flags & XE_VMA_ATOMIC_PTE_BIT) {
-		if (xe_vm_in_lr_mode(xe_vma_vm(vma))) {
-			if (bo && xe_bo_has_single_placement(bo))
-				xe_walk.default_pte |= XE_USM_PPGTT_PTE_AE;
-			/**
-			 * If a SMEM+LMEM allocation is backed by SMEM, a device
-			 * atomics will cause a gpu page fault and which then
-			 * gets migrated to LMEM, bind such allocations with
-			 * device atomics enabled.
-			 */
-			else if (is_devmem)
-				xe_walk.default_pte |= XE_USM_PPGTT_PTE_AE;
-		} else {
-			xe_walk.default_pte |= XE_USM_PPGTT_PTE_AE;
-		}
-
-		/**
-		 * Unset AE if the platform(PVC) doesn't support it on an
-		 * allocation
-		 */
-		if (!xe->info.has_device_atomics_on_smem && !is_devmem)
-			xe_walk.default_pte &= ~XE_USM_PPGTT_PTE_AE;
-	}
-
-	if (is_devmem) {
-		xe_walk.default_pte |= XE_PPGTT_PTE_DM;
-		xe_walk.dma_offset = bo ? vram_region_gpu_offset(bo->ttm.resource) : 0;
+		xe_walk.default_vram_pte = xe_atomic_for_vram(vm) ? XE_USM_PPGTT_PTE_AE : 0;
+		xe_walk.default_system_pte = xe_atomic_for_system(vm, bo) ?
+			XE_USM_PPGTT_PTE_AE : 0;
 	}
 
-	if (!xe_vma_has_no_bo(vma) && xe_bo_is_stolen(bo))
-		xe_walk.dma_offset = xe_ttm_stolen_gpu_offset(xe_bo_device(bo));
-
+	xe_walk.default_vram_pte |= XE_PPGTT_PTE_DM;
+	xe_walk.dma_offset = bo ? vram_region_gpu_offset(bo->ttm.resource) : 0;
 	if (!range)
 		xe_bo_assert_held(bo);
 
@@ -739,6 +768,7 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 		curs.size = xe_vma_size(vma);
 	}
 
+walk_pt:
 	ret = xe_pt_walk_range(&pt->base, pt->level,
 			       range ? range->base.itree.start : xe_vma_start(vma),
 			       range ? range->base.itree.last + 1 : xe_vma_end(vma),
@@ -1103,12 +1133,14 @@ static void xe_pt_free_bind(struct xe_vm_pgtable_update *entries,
 static int
 xe_pt_prepare_bind(struct xe_tile *tile, struct xe_vma *vma,
 		   struct xe_svm_range *range,
-		   struct xe_vm_pgtable_update *entries, u32 *num_entries)
+		   struct xe_vm_pgtable_update *entries,
+		   u32 *num_entries, bool invalidate_on_bind)
 {
 	int err;
 
 	*num_entries = 0;
-	err = xe_pt_stage_bind(tile, vma, range, entries, num_entries);
+	err = xe_pt_stage_bind(tile, vma, range, entries, num_entries,
+			       invalidate_on_bind);
 	if (!err)
 		xe_tile_assert(tile, *num_entries);
 
@@ -1420,6 +1452,7 @@ static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
 	return err;
 }
 
+#if IS_ENABLED(CONFIG_DRM_XE_GPUSVM)
 static int xe_pt_svm_pre_commit(struct xe_migrate_pt_update *pt_update)
 {
 	struct xe_vm *vm = pt_update->vops->vm;
@@ -1453,6 +1486,7 @@ static int xe_pt_svm_pre_commit(struct xe_migrate_pt_update *pt_update)
 
 	return 0;
 }
+#endif
 
 struct invalidation_fence {
 	struct xe_gt_tlb_invalidation_fence base;
@@ -1791,7 +1825,7 @@ static int vma_reserve_fences(struct xe_device *xe, struct xe_vma *vma)
 
 static int bind_op_prepare(struct xe_vm *vm, struct xe_tile *tile,
 			   struct xe_vm_pgtable_update_ops *pt_update_ops,
-			   struct xe_vma *vma)
+			   struct xe_vma *vma, bool invalidate_on_bind)
 {
 	u32 current_op = pt_update_ops->current_op;
 	struct xe_vm_pgtable_update_op *pt_op = &pt_update_ops->ops[current_op];
@@ -1813,7 +1847,7 @@ static int bind_op_prepare(struct xe_vm *vm, struct xe_tile *tile,
 		return err;
 
 	err = xe_pt_prepare_bind(tile, vma, NULL, pt_op->entries,
-				 &pt_op->num_entries);
+				 &pt_op->num_entries, invalidate_on_bind);
 	if (!err) {
 		xe_tile_assert(tile, pt_op->num_entries <=
 			       ARRAY_SIZE(pt_op->entries));
@@ -1835,11 +1869,11 @@ static int bind_op_prepare(struct xe_vm *vm, struct xe_tile *tile,
 		 * If !rebind, and scratch enabled VMs, there is a chance the scratch
 		 * PTE is already cached in the TLB so it needs to be invalidated.
 		 * On !LR VMs this is done in the ring ops preceding a batch, but on
-		 * non-faulting LR, in particular on user-space batch buffer chaining,
-		 * it needs to be done here.
+		 * LR, in particular on user-space batch buffer chaining, it needs to
+		 * be done here.
 		 */
 		if ((!pt_op->rebind && xe_vm_has_scratch(vm) &&
-		     xe_vm_in_preempt_fence_mode(vm)))
+		     xe_vm_in_lr_mode(vm)))
 			pt_update_ops->needs_invalidation = true;
 		else if (pt_op->rebind && !xe_vm_in_lr_mode(vm))
 			/* We bump also if batch_invalidate_tlb is true */
@@ -1875,7 +1909,7 @@ static int bind_range_prepare(struct xe_vm *vm, struct xe_tile *tile,
 	pt_op->rebind = BIT(tile->id) & range->tile_present;
 
 	err = xe_pt_prepare_bind(tile, vma, range, pt_op->entries,
-				 &pt_op->num_entries);
+				 &pt_op->num_entries, false);
 	if (!err) {
 		xe_tile_assert(tile, pt_op->num_entries <=
 			       ARRAY_SIZE(pt_op->entries));
@@ -1987,11 +2021,13 @@ static int op_prepare(struct xe_vm *vm,
 
 	switch (op->base.op) {
 	case DRM_GPUVA_OP_MAP:
-		if ((!op->map.immediate && xe_vm_in_fault_mode(vm)) ||
+		if ((!op->map.immediate && xe_vm_in_fault_mode(vm) &&
+		     !op->map.invalidate_on_bind) ||
 		    op->map.is_cpu_addr_mirror)
 			break;
 
-		err = bind_op_prepare(vm, tile, pt_update_ops, op->map.vma);
+		err = bind_op_prepare(vm, tile, pt_update_ops, op->map.vma,
+				      op->map.invalidate_on_bind);
 		pt_update_ops->wait_vm_kernel = true;
 		break;
 	case DRM_GPUVA_OP_REMAP:
@@ -2005,12 +2041,12 @@ static int op_prepare(struct xe_vm *vm,
 
 		if (!err && op->remap.prev) {
 			err = bind_op_prepare(vm, tile, pt_update_ops,
-					      op->remap.prev);
+					      op->remap.prev, false);
 			pt_update_ops->wait_vm_bookkeep = true;
 		}
 		if (!err && op->remap.next) {
 			err = bind_op_prepare(vm, tile, pt_update_ops,
-					      op->remap.next);
+					      op->remap.next, false);
 			pt_update_ops->wait_vm_bookkeep = true;
 		}
 		break;
@@ -2032,7 +2068,7 @@ static int op_prepare(struct xe_vm *vm,
 		if (xe_vma_is_cpu_addr_mirror(vma))
 			break;
 
-		err = bind_op_prepare(vm, tile, pt_update_ops, vma);
+		err = bind_op_prepare(vm, tile, pt_update_ops, vma, false);
 		pt_update_ops->wait_vm_kernel = true;
 		break;
 	}
@@ -2115,7 +2151,7 @@ ALLOW_ERROR_INJECTION(xe_pt_update_ops_prepare, ERRNO);
 static void bind_op_commit(struct xe_vm *vm, struct xe_tile *tile,
 			   struct xe_vm_pgtable_update_ops *pt_update_ops,
 			   struct xe_vma *vma, struct dma_fence *fence,
-			   struct dma_fence *fence2)
+			   struct dma_fence *fence2, bool invalidate_on_bind)
 {
 	xe_tile_assert(tile, !xe_vma_is_cpu_addr_mirror(vma));
 
@@ -2132,6 +2168,8 @@ static void bind_op_commit(struct xe_vm *vm, struct xe_tile *tile,
 	}
 	vma->tile_present |= BIT(tile->id);
 	vma->tile_staged &= ~BIT(tile->id);
+	if (invalidate_on_bind)
+		vma->tile_invalidated |= BIT(tile->id);
 	if (xe_vma_is_userptr(vma)) {
 		lockdep_assert_held_read(&vm->userptr.notifier_lock);
 		to_userptr_vma(vma)->userptr.initial_bind = true;
@@ -2193,7 +2231,7 @@ static void op_commit(struct xe_vm *vm,
 			break;
 
 		bind_op_commit(vm, tile, pt_update_ops, op->map.vma, fence,
-			       fence2);
+			       fence2, op->map.invalidate_on_bind);
 		break;
 	case DRM_GPUVA_OP_REMAP:
 	{
@@ -2206,10 +2244,10 @@ static void op_commit(struct xe_vm *vm,
 
 		if (op->remap.prev)
 			bind_op_commit(vm, tile, pt_update_ops, op->remap.prev,
-				       fence, fence2);
+				       fence, fence2, false);
 		if (op->remap.next)
 			bind_op_commit(vm, tile, pt_update_ops, op->remap.next,
-				       fence, fence2);
+				       fence, fence2, false);
 		break;
 	}
 	case DRM_GPUVA_OP_UNMAP:
@@ -2227,7 +2265,7 @@ static void op_commit(struct xe_vm *vm,
 
 		if (!xe_vma_is_cpu_addr_mirror(vma))
 			bind_op_commit(vm, tile, pt_update_ops, vma, fence,
-				       fence2);
+				       fence2, false);
 		break;
 	}
 	case DRM_GPUVA_OP_DRIVER:
@@ -2257,11 +2295,15 @@ static const struct xe_migrate_pt_update_ops userptr_migrate_ops = {
 	.pre_commit = xe_pt_userptr_pre_commit,
 };
 
+#if IS_ENABLED(CONFIG_DRM_XE_GPUSVM)
 static const struct xe_migrate_pt_update_ops svm_migrate_ops = {
 	.populate = xe_vm_populate_pgtable,
 	.clear = xe_migrate_clear_pgtable_callback,
 	.pre_commit = xe_pt_svm_pre_commit,
 };
+#else
+static const struct xe_migrate_pt_update_ops svm_migrate_ops;
+#endif
 
 /**
  * xe_pt_update_ops_run() - Run PT update operations
diff --git a/drivers/gpu/drm/xe/xe_pxp_debugfs.c b/drivers/gpu/drm/xe/xe_pxp_debugfs.c
index ccfbacf08efc..525a2f6bb076 100644
--- a/drivers/gpu/drm/xe/xe_pxp_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_pxp_debugfs.c
@@ -66,9 +66,18 @@ static int pxp_terminate(struct seq_file *m, void *data)
 {
 	struct xe_pxp *pxp = node_to_pxp(m->private);
 	struct drm_printer p = drm_seq_file_printer(m);
+	int ready = xe_pxp_get_readiness_status(pxp);
 
-	if (!xe_pxp_is_enabled(pxp))
-		return -ENODEV;
+	if (ready < 0)
+		return ready; /* disabled or error occurred */
+	else if (!ready)
+		return -EBUSY; /* init still in progress */
+
+	/* no need for a termination if PXP is not active */
+	if (pxp->status != XE_PXP_ACTIVE) {
+		drm_printf(&p, "PXP not active\n");
+		return 0;
+	}
 
 	/* simulate a termination interrupt */
 	spin_lock_irq(&pxp->xe->irq.lock);
diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
index 5e65830dad25..2dbf4066d86f 100644
--- a/drivers/gpu/drm/xe/xe_query.c
+++ b/drivers/gpu/drm/xe/xe_query.c
@@ -340,7 +340,7 @@ static int query_config(struct xe_device *xe, struct drm_xe_device_query *query)
 	if (xe_device_get_root_tile(xe)->mem.vram.usable_size)
 		config->info[DRM_XE_QUERY_CONFIG_FLAGS] |=
 			DRM_XE_QUERY_CONFIG_FLAG_HAS_VRAM;
-	if (xe->info.has_usm && IS_ENABLED(CONFIG_DRM_GPUSVM))
+	if (xe->info.has_usm && IS_ENABLED(CONFIG_DRM_XE_GPUSVM))
 		config->info[DRM_XE_QUERY_CONFIG_FLAGS] |=
 			DRM_XE_QUERY_CONFIG_FLAG_HAS_CPU_ADDR_MIRROR;
 	config->info[DRM_XE_QUERY_CONFIG_FLAGS] |=
diff --git a/drivers/gpu/drm/xe/xe_reg_sr.c b/drivers/gpu/drm/xe/xe_reg_sr.c
index 9475e3f74958..fc8447a838c4 100644
--- a/drivers/gpu/drm/xe/xe_reg_sr.c
+++ b/drivers/gpu/drm/xe/xe_reg_sr.c
@@ -173,6 +173,9 @@ void xe_reg_sr_apply_mmio(struct xe_reg_sr *sr, struct xe_gt *gt)
 	if (xa_empty(&sr->xa))
 		return;
 
+	if (IS_SRIOV_VF(gt_to_xe(gt)))
+		return;
+
 	xe_gt_dbg(gt, "Applying %s save-restore MMIOs\n", sr->name);
 
 	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL);
diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
index 917fc16de866..a7582b097ae6 100644
--- a/drivers/gpu/drm/xe/xe_ring_ops.c
+++ b/drivers/gpu/drm/xe/xe_ring_ops.c
@@ -137,7 +137,8 @@ emit_pipe_control(u32 *dw, int i, u32 bit_group_0, u32 bit_group_1, u32 offset,
 static int emit_pipe_invalidate(u32 mask_flags, bool invalidate_tlb, u32 *dw,
 				int i)
 {
-	u32 flags = PIPE_CONTROL_CS_STALL |
+	u32 flags0 = 0;
+	u32 flags1 = PIPE_CONTROL_CS_STALL |
 		PIPE_CONTROL_COMMAND_CACHE_INVALIDATE |
 		PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE |
 		PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE |
@@ -148,11 +149,15 @@ static int emit_pipe_invalidate(u32 mask_flags, bool invalidate_tlb, u32 *dw,
 		PIPE_CONTROL_STORE_DATA_INDEX;
 
 	if (invalidate_tlb)
-		flags |= PIPE_CONTROL_TLB_INVALIDATE;
+		flags1 |= PIPE_CONTROL_TLB_INVALIDATE;
 
-	flags &= ~mask_flags;
+	flags1 &= ~mask_flags;
 
-	return emit_pipe_control(dw, i, 0, flags, LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0);
+	if (flags1 & PIPE_CONTROL_VF_CACHE_INVALIDATE)
+		flags0 |= PIPE_CONTROL0_L3_READ_ONLY_CACHE_INVALIDATE;
+
+	return emit_pipe_control(dw, i, flags0, flags1,
+				 LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0);
 }
 
 static int emit_store_imm_ppgtt_posted(u64 addr, u64 value,
diff --git a/drivers/gpu/drm/xe/xe_ring_ops_types.h b/drivers/gpu/drm/xe/xe_ring_ops_types.h
index 1ae56e2ee7b4..d7e3e150a9a5 100644
--- a/drivers/gpu/drm/xe/xe_ring_ops_types.h
+++ b/drivers/gpu/drm/xe/xe_ring_ops_types.h
@@ -8,7 +8,7 @@
 
 struct xe_sched_job;
 
-#define MAX_JOB_SIZE_DW 48
+#define MAX_JOB_SIZE_DW 58
 #define MAX_JOB_SIZE_BYTES (MAX_JOB_SIZE_DW * 4)
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_rtp.c b/drivers/gpu/drm/xe/xe_rtp.c
index 13bb62d3e615..29e694bb1219 100644
--- a/drivers/gpu/drm/xe/xe_rtp.c
+++ b/drivers/gpu/drm/xe/xe_rtp.c
@@ -258,9 +258,6 @@ void xe_rtp_process_to_sr(struct xe_rtp_process_ctx *ctx,
 
 	rtp_get_context(ctx, &hwe, &gt, &xe);
 
-	if (IS_SRIOV_VF(xe))
-		return;
-
 	xe_assert(xe, entries);
 
 	for (entry = entries; entry - entries < n_entries; entry++) {
diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c
index f8fe61e25518..1d43e183ca21 100644
--- a/drivers/gpu/drm/xe/xe_sa.c
+++ b/drivers/gpu/drm/xe/xe_sa.c
@@ -60,7 +60,8 @@ struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u3
 	bo = xe_managed_bo_create_pin_map(xe, tile, size,
 					  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
 					  XE_BO_FLAG_GGTT |
-					  XE_BO_FLAG_GGTT_INVALIDATE);
+					  XE_BO_FLAG_GGTT_INVALIDATE |
+					  XE_BO_FLAG_PINNED_NORESTORE);
 	if (IS_ERR(bo)) {
 		drm_err(&xe->drm, "Failed to prepare %uKiB BO for SA manager (%pe)\n",
 			size / SZ_1K, bo);
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index cb813b337fd3..1f710b3fc599 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -10,6 +10,7 @@
 #include <linux/pci.h>
 #include <linux/sysfs.h>
 
+#include "xe_configfs.h"
 #include "xe_device.h"
 #include "xe_gt.h"
 #include "xe_heci_gsc.h"
@@ -28,20 +29,32 @@
  * This is implemented by loading the driver with bare minimum (no drm card) to allow the firmware
  * to be flashed through mei and collect telemetry. The driver's probe flow is modified
  * such that it enters survivability mode when pcode initialization is incomplete and boot status
- * denotes a failure. The driver then  populates the survivability_mode PCI sysfs indicating
- * survivability mode and provides additional information required for debug
+ * denotes a failure.
  *
- * KMD exposes below admin-only readable sysfs in survivability mode
+ * Survivability mode can also be entered manually using the survivability mode attribute available
+ * through configfs which is beneficial in several usecases. It can be used to address scenarios
+ * where pcode does not detect failure or for validation purposes. It can also be used in
+ * In-Field-Repair (IFR) to repair a single card without impacting the other cards in a node.
  *
- * device/survivability_mode: The presence of this file indicates that the card is in survivability
- *			      mode. Also, provides additional information on why the driver entered
- *			      survivability mode.
+ * Use below command enable survivability mode manually::
  *
- *			      Capability Information - Provides boot status
- *			      Postcode Information   - Provides information about the failure
- *			      Overflow Information   - Provides history of previous failures
- *			      Auxiliary Information  - Certain failures may have information in
- *						       addition to postcode information
+ *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/survivability_mode
+ *
+ * Refer :ref:`xe_configfs` for more details on how to use configfs
+ *
+ * Survivability mode is indicated by the below admin-only readable sysfs which provides additional
+ * debug information::
+ *
+ *	/sys/bus/pci/devices/<device>/surivability_mode
+ *
+ * Capability Information:
+ *	Provides boot status
+ * Postcode Information:
+ *	Provides information about the failure
+ * Overflow Information
+ *	Provides history of previous failures
+ * Auxiliary Information
+ *	Certain failures may have information in addition to postcode information
  */
 
 static u32 aux_history_offset(u32 reg_value)
@@ -133,6 +146,7 @@ static void xe_survivability_mode_fini(void *arg)
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 	struct device *dev = &pdev->dev;
 
+	xe_configfs_clear_survivability_mode(pdev);
 	sysfs_remove_file(&dev->kobj, &dev_attr_survivability_mode.attr);
 }
 
@@ -186,24 +200,41 @@ bool xe_survivability_mode_is_enabled(struct xe_device *xe)
 	return xe->survivability.mode;
 }
 
-/*
- * survivability_mode_requested - check if it's possible to enable
- * survivability mode and that was requested by firmware
+/**
+ * xe_survivability_mode_is_requested - check if it's possible to enable survivability
+ *					mode that was requested by firmware or userspace
+ * @xe: xe device instance
  *
- * This function reads the boot status from Pcode.
+ * This function reads configfs and  boot status from Pcode.
  *
  * Return: true if platform support is available and boot status indicates
- * failure, false otherwise.
+ * failure or if survivability mode is requested, false otherwise.
  */
-static bool survivability_mode_requested(struct xe_device *xe)
+bool xe_survivability_mode_is_requested(struct xe_device *xe)
 {
 	struct xe_survivability *survivability = &xe->survivability;
 	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
+	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 	u32 data;
+	bool survivability_mode;
 
-	if (!IS_DGFX(xe) || xe->info.platform < XE_BATTLEMAGE || IS_SRIOV_VF(xe))
+	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
 		return false;
 
+	survivability_mode = xe_configfs_get_survivability_mode(pdev);
+
+	if (xe->info.platform < XE_BATTLEMAGE) {
+		if (survivability_mode) {
+			dev_err(&pdev->dev, "Survivability Mode is not supported on this card\n");
+			xe_configfs_clear_survivability_mode(pdev);
+		}
+		return false;
+	}
+
+	/* Enable survivability mode if set via configfs */
+	if (survivability_mode)
+		return true;
+
 	data = xe_mmio_read32(mmio, PCODE_SCRATCH(0));
 	survivability->boot_status = REG_FIELD_GET(BOOT_STATUS, data);
 
@@ -226,7 +257,7 @@ int xe_survivability_mode_enable(struct xe_device *xe)
 	struct xe_survivability_info *info;
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 
-	if (!survivability_mode_requested(xe))
+	if (!xe_survivability_mode_is_requested(xe))
 		return 0;
 
 	survivability->size = MAX_SCRATCH_MMIO;
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.h b/drivers/gpu/drm/xe/xe_survivability_mode.h
index d7e64885570d..02231c2bf008 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.h
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.h
@@ -12,5 +12,6 @@ struct xe_device;
 
 int xe_survivability_mode_enable(struct xe_device *xe);
 bool xe_survivability_mode_is_enabled(struct xe_device *xe);
+bool xe_survivability_mode_is_requested(struct xe_device *xe);
 
 #endif /* _XE_SURVIVABILITY_MODE_H_ */
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 3e829c87d7b4..d25f02c8d7fc 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -4,6 +4,7 @@
  */
 
 #include "xe_bo.h"
+#include "xe_gt_stats.h"
 #include "xe_gt_tlb_invalidation.h"
 #include "xe_migrate.h"
 #include "xe_module.h"
@@ -79,7 +80,7 @@ xe_svm_range_alloc(struct drm_gpusvm *gpusvm)
 
 	range = kzalloc(sizeof(*range), GFP_KERNEL);
 	if (!range)
-		return ERR_PTR(-ENOMEM);
+		return NULL;
 
 	INIT_LIST_HEAD(&range->garbage_collector_link);
 	xe_vm_get(gpusvm_to_vm(gpusvm));
@@ -339,6 +340,8 @@ static void xe_svm_garbage_collector_work_func(struct work_struct *w)
 	up_write(&vm->lock);
 }
 
+#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR)
+
 static struct xe_vram_region *page_to_vr(struct page *page)
 {
 	return container_of(page_pgmap(page), struct xe_vram_region, pagemap);
@@ -577,6 +580,8 @@ static const struct drm_gpusvm_devmem_ops gpusvm_devmem_ops = {
 	.copy_to_ram = xe_svm_copy_to_ram,
 };
 
+#endif
+
 static const struct drm_gpusvm_ops gpusvm_ops = {
 	.range_alloc = xe_svm_range_alloc,
 	.range_free = xe_svm_range_free,
@@ -650,6 +655,7 @@ static bool xe_svm_range_is_valid(struct xe_svm_range *range,
 	return (range->tile_present & ~range->tile_invalidated) & BIT(tile->id);
 }
 
+#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR)
 static struct xe_vram_region *tile_to_vr(struct xe_tile *tile)
 {
 	return &tile->mem.vram;
@@ -696,11 +702,14 @@ retry:
 	list_for_each_entry(block, blocks, link)
 		block->private = vr;
 
+	xe_bo_get(bo);
 	err = drm_gpusvm_migrate_to_devmem(&vm->svm.gpusvm, &range->base,
 					   &bo->devmem_allocation, ctx);
-	xe_bo_unlock(bo);
 	if (err)
-		xe_bo_put(bo);	/* Creation ref */
+		xe_svm_devmem_release(&bo->devmem_allocation);
+
+	xe_bo_unlock(bo);
+	xe_bo_put(bo);
 
 unlock:
 	mmap_read_unlock(mm);
@@ -708,12 +717,20 @@ unlock:
 
 	return err;
 }
+#else
+static int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile,
+			     struct xe_svm_range *range,
+			     const struct drm_gpusvm_ctx *ctx)
+{
+	return -EOPNOTSUPP;
+}
+#endif
 
 /**
  * xe_svm_handle_pagefault() - SVM handle page fault
  * @vm: The VM.
  * @vma: The CPU address mirror VMA.
- * @tile: The tile upon the fault occurred.
+ * @gt: The gt upon the fault occurred.
  * @fault_addr: The GPU fault address.
  * @atomic: The fault atomic access bit.
  *
@@ -723,7 +740,7 @@ unlock:
  * Return: 0 on success, negative error code on error.
  */
 int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
-			    struct xe_tile *tile, u64 fault_addr,
+			    struct xe_gt *gt, u64 fault_addr,
 			    bool atomic)
 {
 	struct drm_gpusvm_ctx ctx = {
@@ -737,12 +754,15 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 	struct drm_gpusvm_range *r;
 	struct drm_exec exec;
 	struct dma_fence *fence;
+	struct xe_tile *tile = gt_to_tile(gt);
 	ktime_t end = 0;
 	int err;
 
 	lockdep_assert_held_write(&vm->lock);
 	xe_assert(vm->xe, xe_vma_is_cpu_addr_mirror(vma));
 
+	xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_PAGEFAULT_COUNT, 1);
+
 retry:
 	/* Always process UNMAPs first so view SVM ranges is current */
 	err = xe_svm_garbage_collector(vm);
@@ -863,6 +883,7 @@ int xe_svm_bo_evict(struct xe_bo *bo)
 }
 
 #if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR)
+
 static struct drm_pagemap_device_addr
 xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,
 			  struct device *dev,
@@ -944,3 +965,15 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr)
 	return 0;
 }
 #endif
+
+/**
+ * xe_svm_flush() - SVM flush
+ * @vm: The VM.
+ *
+ * Flush all SVM actions.
+ */
+void xe_svm_flush(struct xe_vm *vm)
+{
+	if (xe_vm_in_fault_mode(vm))
+		flush_work(&vm->svm.garbage_collector.work);
+}
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index e059590e5076..2881af1e60b2 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -6,16 +6,19 @@
 #ifndef _XE_SVM_H_
 #define _XE_SVM_H_
 
+#if IS_ENABLED(CONFIG_DRM_XE_GPUSVM)
+
 #include <drm/drm_pagemap.h>
 #include <drm/drm_gpusvm.h>
 
 #define XE_INTERCONNECT_VRAM DRM_INTERCONNECT_DRIVER
 
 struct xe_bo;
-struct xe_vram_region;
+struct xe_gt;
 struct xe_tile;
 struct xe_vm;
 struct xe_vma;
+struct xe_vram_region;
 
 /** struct xe_svm_range - SVM range */
 struct xe_svm_range {
@@ -43,7 +46,6 @@ struct xe_svm_range {
 	u8 skip_migrate	:1;
 };
 
-#if IS_ENABLED(CONFIG_DRM_GPUSVM)
 /**
  * xe_svm_range_pages_valid() - SVM range pages valid
  * @range: SVM range
@@ -64,7 +66,7 @@ void xe_svm_fini(struct xe_vm *vm);
 void xe_svm_close(struct xe_vm *vm);
 
 int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
-			    struct xe_tile *tile, u64 fault_addr,
+			    struct xe_gt *gt, u64 fault_addr,
 			    bool atomic);
 
 bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end);
@@ -72,7 +74,52 @@ bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end);
 int xe_svm_bo_evict(struct xe_bo *bo);
 
 void xe_svm_range_debug(struct xe_svm_range *range, const char *operation);
+
+/**
+ * xe_svm_range_has_dma_mapping() - SVM range has DMA mapping
+ * @range: SVM range
+ *
+ * Return: True if SVM range has a DMA mapping, False otherwise
+ */
+static inline bool xe_svm_range_has_dma_mapping(struct xe_svm_range *range)
+{
+	lockdep_assert_held(&range->base.gpusvm->notifier_lock);
+	return range->base.flags.has_dma_mapping;
+}
+
+#define xe_svm_assert_in_notifier(vm__) \
+	lockdep_assert_held_write(&(vm__)->svm.gpusvm.notifier_lock)
+
+#define xe_svm_notifier_lock(vm__)	\
+	drm_gpusvm_notifier_lock(&(vm__)->svm.gpusvm)
+
+#define xe_svm_notifier_unlock(vm__)	\
+	drm_gpusvm_notifier_unlock(&(vm__)->svm.gpusvm)
+
+void xe_svm_flush(struct xe_vm *vm);
+
 #else
+#include <linux/interval_tree.h>
+
+struct drm_pagemap_device_addr;
+struct xe_bo;
+struct xe_gt;
+struct xe_vm;
+struct xe_vma;
+struct xe_tile;
+struct xe_vram_region;
+
+#define XE_INTERCONNECT_VRAM 1
+
+struct xe_svm_range {
+	struct {
+		struct interval_tree_node itree;
+		const struct drm_pagemap_device_addr *dma_addr;
+	} base;
+	u32 tile_present;
+	u32 tile_invalidated;
+};
+
 static inline bool xe_svm_range_pages_valid(struct xe_svm_range *range)
 {
 	return false;
@@ -102,7 +149,7 @@ void xe_svm_close(struct xe_vm *vm)
 
 static inline
 int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
-			    struct xe_tile *tile, u64 fault_addr,
+			    struct xe_gt *gt, u64 fault_addr,
 			    bool atomic)
 {
 	return 0;
@@ -124,27 +171,20 @@ static inline
 void xe_svm_range_debug(struct xe_svm_range *range, const char *operation)
 {
 }
-#endif
 
-/**
- * xe_svm_range_has_dma_mapping() - SVM range has DMA mapping
- * @range: SVM range
- *
- * Return: True if SVM range has a DMA mapping, False otherwise
- */
-static inline bool xe_svm_range_has_dma_mapping(struct xe_svm_range *range)
+#define xe_svm_assert_in_notifier(...) do {} while (0)
+#define xe_svm_range_has_dma_mapping(...) false
+
+static inline void xe_svm_notifier_lock(struct xe_vm *vm)
 {
-	lockdep_assert_held(&range->base.gpusvm->notifier_lock);
-	return range->base.flags.has_dma_mapping;
 }
 
-#define xe_svm_assert_in_notifier(vm__) \
-	lockdep_assert_held_write(&(vm__)->svm.gpusvm.notifier_lock)
-
-#define xe_svm_notifier_lock(vm__)	\
-	drm_gpusvm_notifier_lock(&(vm__)->svm.gpusvm)
-
-#define xe_svm_notifier_unlock(vm__)	\
-	drm_gpusvm_notifier_unlock(&(vm__)->svm.gpusvm)
+static inline void xe_svm_notifier_unlock(struct xe_vm *vm)
+{
+}
 
+static inline void xe_svm_flush(struct xe_vm *vm)
+{
+}
+#endif
 #endif
diff --git a/drivers/gpu/drm/xe/xe_uc.c b/drivers/gpu/drm/xe/xe_uc.c
index c14bd2282044..3a8751a8b92d 100644
--- a/drivers/gpu/drm/xe/xe_uc.c
+++ b/drivers/gpu/drm/xe/xe_uc.c
@@ -244,7 +244,7 @@ void xe_uc_gucrc_disable(struct xe_uc *uc)
 
 void xe_uc_stop_prepare(struct xe_uc *uc)
 {
-	xe_gsc_wait_for_worker_completion(&uc->gsc);
+	xe_gsc_stop_prepare(&uc->gsc);
 	xe_guc_stop_prepare(&uc->guc);
 }
 
@@ -278,6 +278,12 @@ again:
 		goto again;
 }
 
+void xe_uc_suspend_prepare(struct xe_uc *uc)
+{
+	xe_gsc_wait_for_worker_completion(&uc->gsc);
+	xe_guc_stop_prepare(&uc->guc);
+}
+
 int xe_uc_suspend(struct xe_uc *uc)
 {
 	/* GuC submission not enabled, nothing to do */
diff --git a/drivers/gpu/drm/xe/xe_uc.h b/drivers/gpu/drm/xe/xe_uc.h
index 3813c1ede450..c23e6f5e2514 100644
--- a/drivers/gpu/drm/xe/xe_uc.h
+++ b/drivers/gpu/drm/xe/xe_uc.h
@@ -18,6 +18,7 @@ int xe_uc_reset_prepare(struct xe_uc *uc);
 void xe_uc_stop_prepare(struct xe_uc *uc);
 void xe_uc_stop(struct xe_uc *uc);
 int xe_uc_start(struct xe_uc *uc);
+void xe_uc_suspend_prepare(struct xe_uc *uc);
 int xe_uc_suspend(struct xe_uc *uc);
 int xe_uc_sanitize_reset(struct xe_uc *uc);
 void xe_uc_declare_wedged(struct xe_uc *uc);
diff --git a/drivers/gpu/drm/xe/xe_uc_fw.c b/drivers/gpu/drm/xe/xe_uc_fw.c
index fb0eda3d5682..2741849bbf4d 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw.c
+++ b/drivers/gpu/drm/xe/xe_uc_fw.c
@@ -92,6 +92,8 @@
 
 struct uc_fw_entry {
 	enum xe_platform platform;
+	enum xe_gt_type gt_type;
+
 	struct {
 		const char *path;
 		u16 major;
@@ -106,32 +108,37 @@ struct fw_blobs_by_type {
 	u32 count;
 };
 
-#define XE_GUC_FIRMWARE_DEFS(fw_def, mmp_ver, major_ver)			\
-	fw_def(BATTLEMAGE,	major_ver(xe,	guc,	bmg,	70, 29, 2))	\
-	fw_def(LUNARLAKE,	major_ver(xe,	guc,	lnl,	70, 29, 2))	\
-	fw_def(METEORLAKE,	major_ver(i915,	guc,	mtl,	70, 29, 2))	\
-	fw_def(DG2,		major_ver(i915,	guc,	dg2,	70, 29, 2))	\
-	fw_def(DG1,		major_ver(i915,	guc,	dg1,	70, 29, 2))	\
-	fw_def(ALDERLAKE_N,	major_ver(i915,	guc,	tgl,	70, 29, 2))	\
-	fw_def(ALDERLAKE_P,	major_ver(i915,	guc,	adlp,	70, 29, 2))	\
-	fw_def(ALDERLAKE_S,	major_ver(i915,	guc,	tgl,	70, 29, 2))	\
-	fw_def(ROCKETLAKE,	major_ver(i915,	guc,	tgl,	70, 29, 2))	\
-	fw_def(TIGERLAKE,	major_ver(i915,	guc,	tgl,	70, 29, 2))
+/*
+ * Add an "ANY" define just to convey the meaning it's given here.
+ */
+#define XE_GT_TYPE_ANY XE_GT_TYPE_UNINITIALIZED
+
+#define XE_GUC_FIRMWARE_DEFS(fw_def, mmp_ver, major_ver)					\
+	fw_def(BATTLEMAGE,	GT_TYPE_ANY,	major_ver(xe,	guc,	bmg,	70, 44, 1))	\
+	fw_def(LUNARLAKE,	GT_TYPE_ANY,	major_ver(xe,	guc,	lnl,	70, 44, 1))	\
+	fw_def(METEORLAKE,	GT_TYPE_ANY,	major_ver(i915,	guc,	mtl,	70, 44, 1))	\
+	fw_def(DG2,		GT_TYPE_ANY,	major_ver(i915,	guc,	dg2,	70, 44, 1))	\
+	fw_def(DG1,		GT_TYPE_ANY,	major_ver(i915,	guc,	dg1,	70, 44, 1))	\
+	fw_def(ALDERLAKE_N,	GT_TYPE_ANY,	major_ver(i915,	guc,	tgl,	70, 44, 1))	\
+	fw_def(ALDERLAKE_P,	GT_TYPE_ANY,	major_ver(i915,	guc,	adlp,	70, 44, 1))	\
+	fw_def(ALDERLAKE_S,	GT_TYPE_ANY,	major_ver(i915,	guc,	tgl,	70, 44, 1))	\
+	fw_def(ROCKETLAKE,	GT_TYPE_ANY,	major_ver(i915,	guc,	tgl,	70, 44, 1))	\
+	fw_def(TIGERLAKE,	GT_TYPE_ANY,	major_ver(i915,	guc,	tgl,	70, 44, 1))
 
 #define XE_HUC_FIRMWARE_DEFS(fw_def, mmp_ver, no_ver)		\
-	fw_def(BATTLEMAGE,	no_ver(xe,	huc,		bmg))		\
-	fw_def(LUNARLAKE,	no_ver(xe,	huc,		lnl))		\
-	fw_def(METEORLAKE,	no_ver(i915,	huc_gsc,	mtl))		\
-	fw_def(DG1,		no_ver(i915,	huc,		dg1))		\
-	fw_def(ALDERLAKE_P,	no_ver(i915,	huc,		tgl))		\
-	fw_def(ALDERLAKE_S,	no_ver(i915,	huc,		tgl))		\
-	fw_def(ROCKETLAKE,	no_ver(i915,	huc,		tgl))		\
-	fw_def(TIGERLAKE,	no_ver(i915,	huc,		tgl))
+	fw_def(BATTLEMAGE,	GT_TYPE_ANY,	no_ver(xe,	huc,		bmg))		\
+	fw_def(LUNARLAKE,	GT_TYPE_ANY,	no_ver(xe,	huc,		lnl))		\
+	fw_def(METEORLAKE,	GT_TYPE_ANY,	no_ver(i915,	huc_gsc,	mtl))		\
+	fw_def(DG1,		GT_TYPE_ANY,	no_ver(i915,	huc,		dg1))		\
+	fw_def(ALDERLAKE_P,	GT_TYPE_ANY,	no_ver(i915,	huc,		tgl))		\
+	fw_def(ALDERLAKE_S,	GT_TYPE_ANY,	no_ver(i915,	huc,		tgl))		\
+	fw_def(ROCKETLAKE,	GT_TYPE_ANY,	no_ver(i915,	huc,		tgl))		\
+	fw_def(TIGERLAKE,	GT_TYPE_ANY,	no_ver(i915,	huc,		tgl))
 
 /* for the GSC FW we match the compatibility version and not the release one */
 #define XE_GSC_FIRMWARE_DEFS(fw_def, major_ver)		\
-	fw_def(LUNARLAKE,	major_ver(xe,	gsc,	lnl,	104, 1, 0)) \
-	fw_def(METEORLAKE,	major_ver(i915,	gsc,	mtl,	102, 1, 0))
+	fw_def(LUNARLAKE,	GT_TYPE_ANY,	major_ver(xe,	gsc,	lnl,	104, 1, 0))	\
+	fw_def(METEORLAKE,	GT_TYPE_ANY,	major_ver(i915,	gsc,	mtl,	102, 1, 0))
 
 #define MAKE_FW_PATH(dir__, uc__, shortname__, version__)			\
 	__stringify(dir__) "/" __stringify(shortname__) "_" __stringify(uc__) version__ ".bin"
@@ -159,12 +166,13 @@ struct fw_blobs_by_type {
 	  a, b, c }
 
 /* All blobs need to be declared via MODULE_FIRMWARE() */
-#define XE_UC_MODULE_FIRMWARE(platform__, fw_filename)				\
+#define XE_UC_MODULE_FIRMWARE(platform__, gt_type__, fw_filename)		\
 	MODULE_FIRMWARE(fw_filename);
 
-#define XE_UC_FW_ENTRY(platform__, entry__)					\
+#define XE_UC_FW_ENTRY(platform__, gt_type__, entry__)				\
 	{									\
 		.platform = XE_ ## platform__,					\
+		.gt_type = XE_ ## gt_type__,					\
 		entry__,							\
 	},
 
@@ -222,30 +230,38 @@ uc_fw_auto_select(struct xe_device *xe, struct xe_uc_fw *uc_fw)
 		[XE_UC_FW_TYPE_HUC] = { entries_huc, ARRAY_SIZE(entries_huc) },
 		[XE_UC_FW_TYPE_GSC] = { entries_gsc, ARRAY_SIZE(entries_gsc) },
 	};
-	static const struct uc_fw_entry *entries;
+	struct xe_gt *gt = uc_fw_to_gt(uc_fw);
 	enum xe_platform p = xe->info.platform;
+	const struct uc_fw_entry *entries;
 	u32 count;
 	int i;
 
-	xe_assert(xe, uc_fw->type < ARRAY_SIZE(blobs_all));
+	xe_gt_assert(gt, uc_fw->type < ARRAY_SIZE(blobs_all));
+	xe_gt_assert(gt, gt->info.type != XE_GT_TYPE_UNINITIALIZED);
+
 	entries = blobs_all[uc_fw->type].entries;
 	count = blobs_all[uc_fw->type].count;
 
 	for (i = 0; i < count && p <= entries[i].platform; i++) {
-		if (p == entries[i].platform) {
-			uc_fw->path = entries[i].path;
-			uc_fw->versions.wanted.major = entries[i].major;
-			uc_fw->versions.wanted.minor = entries[i].minor;
-			uc_fw->versions.wanted.patch = entries[i].patch;
-			uc_fw->full_ver_required = entries[i].full_ver_required;
-
-			if (uc_fw->type == XE_UC_FW_TYPE_GSC)
-				uc_fw->versions.wanted_type = XE_UC_FW_VER_COMPATIBILITY;
-			else
-				uc_fw->versions.wanted_type = XE_UC_FW_VER_RELEASE;
-
-			break;
-		}
+		if (p != entries[i].platform)
+			continue;
+
+		if (entries[i].gt_type != XE_GT_TYPE_ANY &&
+		    entries[i].gt_type != gt->info.type)
+			continue;
+
+		uc_fw->path = entries[i].path;
+		uc_fw->versions.wanted.major = entries[i].major;
+		uc_fw->versions.wanted.minor = entries[i].minor;
+		uc_fw->versions.wanted.patch = entries[i].patch;
+		uc_fw->full_ver_required = entries[i].full_ver_required;
+
+		if (uc_fw->type == XE_UC_FW_TYPE_GSC)
+			uc_fw->versions.wanted_type = XE_UC_FW_VER_COMPATIBILITY;
+		else
+			uc_fw->versions.wanted_type = XE_UC_FW_VER_RELEASE;
+
+		break;
 	}
 }
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 60303998bd61..79323c78130f 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2049,7 +2049,8 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data,
 		return -EINVAL;
 
 	if (XE_IOCTL_DBG(xe, args->flags & DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE &&
-			 args->flags & DRM_XE_VM_CREATE_FLAG_FAULT_MODE))
+			 args->flags & DRM_XE_VM_CREATE_FLAG_FAULT_MODE &&
+			 !xe->info.needs_scratch))
 		return -EINVAL;
 
 	if (XE_IOCTL_DBG(xe, !(args->flags & DRM_XE_VM_CREATE_FLAG_LR_MODE) &&
@@ -2201,6 +2202,20 @@ static void print_op(struct xe_device *xe, struct drm_gpuva_op *op)
 }
 #endif
 
+static bool __xe_vm_needs_clear_scratch_pages(struct xe_vm *vm, u32 bind_flags)
+{
+	if (!xe_vm_in_fault_mode(vm))
+		return false;
+
+	if (!xe_vm_has_scratch(vm))
+		return false;
+
+	if (bind_flags & DRM_XE_VM_BIND_FLAG_IMMEDIATE)
+		return false;
+
+	return true;
+}
+
 /*
  * Create operations list from IOCTL arguments, setup operations fields so parse
  * and commit steps are decoupled from IOCTL arguments. This step can fail.
@@ -2273,6 +2288,8 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_bo *bo,
 				DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR;
 			op->map.dumpable = flags & DRM_XE_VM_BIND_FLAG_DUMPABLE;
 			op->map.pat_index = pat_index;
+			op->map.invalidate_on_bind =
+				__xe_vm_needs_clear_scratch_pages(vm, flags);
 		} else if (__op->op == DRM_GPUVA_OP_PREFETCH) {
 			op->prefetch.region = prefetch_region;
 		}
@@ -2472,8 +2489,9 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 				return PTR_ERR(vma);
 
 			op->map.vma = vma;
-			if ((op->map.immediate || !xe_vm_in_fault_mode(vm)) &&
-			    !op->map.is_cpu_addr_mirror)
+			if (((op->map.immediate || !xe_vm_in_fault_mode(vm)) &&
+			     !op->map.is_cpu_addr_mirror) ||
+			    op->map.invalidate_on_bind)
 				xe_vma_ops_incr_pt_update_ops(vops,
 							      op->tile_mask);
 			break;
@@ -2726,9 +2744,10 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm,
 
 	switch (op->base.op) {
 	case DRM_GPUVA_OP_MAP:
-		err = vma_lock_and_validate(exec, op->map.vma,
-					    !xe_vm_in_fault_mode(vm) ||
-					    op->map.immediate);
+		if (!op->map.invalidate_on_bind)
+			err = vma_lock_and_validate(exec, op->map.vma,
+						    !xe_vm_in_fault_mode(vm) ||
+						    op->map.immediate);
 		break;
 	case DRM_GPUVA_OP_REMAP:
 		err = check_ufence(gpuva_to_vma(op->base.remap.unmap->va));
@@ -3082,9 +3101,9 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
 		if (!*bind_ops)
 			return args->num_binds > 1 ? -ENOBUFS : -ENOMEM;
 
-		err = __copy_from_user(*bind_ops, bind_user,
-				       sizeof(struct drm_xe_vm_bind_op) *
-				       args->num_binds);
+		err = copy_from_user(*bind_ops, bind_user,
+				     sizeof(struct drm_xe_vm_bind_op) *
+				     args->num_binds);
 		if (XE_IOCTL_DBG(xe, err)) {
 			err = -EFAULT;
 			goto free_bind_ops;
@@ -3109,7 +3128,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
 
 		if (XE_IOCTL_DBG(xe, is_cpu_addr_mirror &&
 				 (!xe_vm_in_fault_mode(vm) ||
-				 !IS_ENABLED(CONFIG_DRM_GPUSVM)))) {
+				 !IS_ENABLED(CONFIG_DRM_XE_GPUSVM)))) {
 			err = -EINVAL;
 			goto free_bind_ops;
 		}
@@ -3243,7 +3262,7 @@ static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo,
 				 XE_64K_PAGE_MASK) ||
 		    XE_IOCTL_DBG(xe, addr & XE_64K_PAGE_MASK) ||
 		    XE_IOCTL_DBG(xe, range & XE_64K_PAGE_MASK)) {
-			return  -EINVAL;
+			return -EINVAL;
 		}
 	}
 
@@ -3251,7 +3270,7 @@ static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo,
 	if (bo->cpu_caching) {
 		if (XE_IOCTL_DBG(xe, coh_mode == XE_COH_NONE &&
 				 bo->cpu_caching == DRM_XE_GEM_CPU_CACHING_WB)) {
-			return  -EINVAL;
+			return -EINVAL;
 		}
 	} else if (XE_IOCTL_DBG(xe, coh_mode == XE_COH_NONE)) {
 		/*
@@ -3260,7 +3279,7 @@ static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo,
 		 * how it was mapped on the CPU. Just assume is it
 		 * potentially cached on CPU side.
 		 */
-		return  -EINVAL;
+		return -EINVAL;
 	}
 
 	/* If a BO is protected it can only be mapped if the key is still valid */
@@ -3312,8 +3331,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	}
 
 	/* Ensure all UNMAPs visible */
-	if (xe_vm_in_fault_mode(vm))
-		flush_work(&vm->svm.garbage_collector.work);
+	xe_svm_flush(vm);
 
 	err = down_write_killable(&vm->lock);
 	if (err)
@@ -3847,6 +3865,9 @@ void xe_vm_snapshot_print(struct xe_vm_snapshot *snap, struct drm_printer *p)
 		}
 
 		drm_puts(p, "\n");
+
+		if (drm_coredump_printer_is_full(p))
+			return;
 	}
 }
 
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 84fa41b9fa20..1662604c4486 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -330,6 +330,8 @@ struct xe_vma_op_map {
 	bool is_cpu_addr_mirror;
 	/** @dumpable: whether BO is dumped on GPU hang */
 	bool dumpable;
+	/** @invalidate: invalidate the VMA before bind */
+	bool invalidate_on_bind;
 	/** @pat_index: The pat index to use for this operation. */
 	u16 pat_index;
 };
diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c
index b1f81dca610d..e421a74fb87c 100644
--- a/drivers/gpu/drm/xe/xe_vram.c
+++ b/drivers/gpu/drm/xe/xe_vram.c
@@ -49,7 +49,7 @@ _resize_bar(struct xe_device *xe, int resno, resource_size_t size)
  */
 static void resize_vram_bar(struct xe_device *xe)
 {
-	u64 force_vram_bar_size = xe_modparam.force_vram_bar_size;
+	int force_vram_bar_size = xe_modparam.force_vram_bar_size;
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 	struct pci_bus *root = pdev->bus;
 	resource_size_t current_size;
@@ -66,6 +66,9 @@ static void resize_vram_bar(struct xe_device *xe)
 	if (!bar_size_mask)
 		return;
 
+	if (force_vram_bar_size < 0)
+		return;
+
 	/* set to a specific size? */
 	if (force_vram_bar_size) {
 		u32 bar_size_bit;
diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index 24f644c0a673..6f6563cc7430 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -230,6 +230,18 @@ static const struct xe_rtp_entry_sr gt_was[] = {
 	  XE_RTP_ENTRY_FLAG(FOREACH_ENGINE),
 	},
 
+	/* Xe2_HPG */
+
+	{ XE_RTP_NAME("16025250150"),
+	  XE_RTP_RULES(GRAPHICS_VERSION(2001)),
+	  XE_RTP_ACTIONS(SET(LSN_VC_REG2,
+			     LSN_LNI_WGT(1) |
+			     LSN_LNE_WGT(1) |
+			     LSN_DIM_X_WGT(1) |
+			     LSN_DIM_Y_WGT(1) |
+			     LSN_DIM_Z_WGT(1)))
+	},
+
 	/* Xe2_HPM */
 
 	{ XE_RTP_NAME("16021867713"),
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 0c738af24f7c..9efc5accd43d 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -32,8 +32,10 @@
 		GRAPHICS_VERSION(3001)
 14022293748	GRAPHICS_VERSION(2001)
 		GRAPHICS_VERSION(2004)
+		GRAPHICS_VERSION_RANGE(3000, 3001)
 22019794406	GRAPHICS_VERSION(2001)
 		GRAPHICS_VERSION(2004)
+		GRAPHICS_VERSION_RANGE(3000, 3001)
 22019338487	MEDIA_VERSION(2000)
 		GRAPHICS_VERSION(2001)
 		MEDIA_VERSION(3000), MEDIA_STEP(A0, B0), FUNC(xe_rtp_match_not_sriov_vf)
@@ -55,3 +57,5 @@ no_media_l3	MEDIA_VERSION(3000)
 		GRAPHICS_VERSION(1260), GRAPHICS_STEP(A0, B0)
 16023105232	GRAPHICS_VERSION_RANGE(2001, 3001)
 		MEDIA_VERSION_RANGE(1301, 3000)
+16026508708	GRAPHICS_VERSION_RANGE(1200, 3001)
+		MEDIA_VERSION_RANGE(1300, 3000)
diff --git a/drivers/gpu/drm/xlnx/Kconfig b/drivers/gpu/drm/xlnx/Kconfig
index dbecca9bdd54..cfabf5e2a0bb 100644
--- a/drivers/gpu/drm/xlnx/Kconfig
+++ b/drivers/gpu/drm/xlnx/Kconfig
@@ -22,6 +22,7 @@ config DRM_ZYNQMP_DPSUB_AUDIO
 	bool "ZynqMP DisplayPort Audio Support"
 	depends on DRM_ZYNQMP_DPSUB
 	depends on SND && SND_SOC
+	depends on SND_SOC=y || DRM_ZYNQMP_DPSUB=m
 	select SND_SOC_GENERIC_DMAENGINE_PCM
 	help
 	  Choose this option to enable DisplayPort audio support in the ZynqMP
diff --git a/drivers/gpu/drm/xlnx/zynqmp_dp.c b/drivers/gpu/drm/xlnx/zynqmp_dp.c
index a6a4a871f197..238cbb49963e 100644
--- a/drivers/gpu/drm/xlnx/zynqmp_dp.c
+++ b/drivers/gpu/drm/xlnx/zynqmp_dp.c
@@ -1481,6 +1481,7 @@ static void zynqmp_dp_disp_disable(struct zynqmp_dp *dp,
  */
 
 static int zynqmp_dp_bridge_attach(struct drm_bridge *bridge,
+				   struct drm_encoder *encoder,
 				   enum drm_bridge_attach_flags flags)
 {
 	struct zynqmp_dp *dp = bridge_to_dp(bridge);
@@ -1494,7 +1495,7 @@ static int zynqmp_dp_bridge_attach(struct drm_bridge *bridge,
 	}
 
 	if (dp->next_bridge) {
-		ret = drm_bridge_attach(bridge->encoder, dp->next_bridge,
+		ret = drm_bridge_attach(encoder, dp->next_bridge,
 					bridge, flags);
 		if (ret < 0)
 			goto error;
@@ -2466,10 +2467,8 @@ int zynqmp_dp_probe(struct zynqmp_dpsub *dpsub)
 
 	dp->reset = devm_reset_control_get(dp->dev, NULL);
 	if (IS_ERR(dp->reset)) {
-		if (PTR_ERR(dp->reset) != -EPROBE_DEFER)
-			dev_err(dp->dev, "failed to get reset: %ld\n",
-				PTR_ERR(dp->reset));
-		ret = PTR_ERR(dp->reset);
+		ret = dev_err_probe(dp->dev, PTR_ERR(dp->reset),
+				    "failed to get reset\n");
 		goto err_free;
 	}
 
diff --git a/drivers/gpu/drm/xlnx/zynqmp_dp_audio.c b/drivers/gpu/drm/xlnx/zynqmp_dp_audio.c
index f07ff4eb3a6d..1a46a046103f 100644
--- a/drivers/gpu/drm/xlnx/zynqmp_dp_audio.c
+++ b/drivers/gpu/drm/xlnx/zynqmp_dp_audio.c
@@ -45,7 +45,6 @@ struct zynqmp_dpsub_audio {
 
 	struct {
 		struct snd_soc_dai_link_component cpu;
-		struct snd_soc_dai_link_component codec;
 		struct snd_soc_dai_link_component platform;
 	} components[ZYNQMP_NUM_PCMS];
 
@@ -403,10 +402,8 @@ int zynqmp_audio_init(struct zynqmp_dpsub *dpsub)
 		link->num_cpus = 1;
 		link->cpus[0].dai_name = audio->dai_name;
 
-		link->codecs = &audio->components[i].codec;
+		link->codecs = &snd_soc_dummy_dlc;
 		link->num_codecs = 1;
-		link->codecs[0].name = "snd-soc-dummy";
-		link->codecs[0].dai_name = "snd-soc-dummy-dai";
 
 		link->platforms = &audio->components[i].platform;
 		link->num_platforms = 1;
diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c
index 8e09d6d328d2..344cc9e741c1 100644
--- a/drivers/gpu/host1x/bus.c
+++ b/drivers/gpu/host1x/bus.c
@@ -41,7 +41,6 @@ static int host1x_subdev_add(struct host1x_device *device,
 			     struct device_node *np)
 {
 	struct host1x_subdev *subdev;
-	struct device_node *child;
 	int err;
 
 	subdev = kzalloc(sizeof(*subdev), GFP_KERNEL);
@@ -56,13 +55,12 @@ static int host1x_subdev_add(struct host1x_device *device,
 	mutex_unlock(&device->subdevs_lock);
 
 	/* recursively add children */
-	for_each_child_of_node(np, child) {
+	for_each_child_of_node_scoped(np, child) {
 		if (of_match_node(driver->subdevs, child) &&
 		    of_device_is_available(child)) {
 			err = host1x_subdev_add(device, driver, child);
 			if (err < 0) {
 				/* XXX cleanup? */
-				of_node_put(child);
 				return err;
 			}
 		}
@@ -90,17 +88,14 @@ static void host1x_subdev_del(struct host1x_subdev *subdev)
 static int host1x_device_parse_dt(struct host1x_device *device,
 				  struct host1x_driver *driver)
 {
-	struct device_node *np;
 	int err;
 
-	for_each_child_of_node(device->dev.parent->of_node, np) {
+	for_each_child_of_node_scoped(device->dev.parent->of_node, np) {
 		if (of_match_node(driver->subdevs, np) &&
 		    of_device_is_available(np)) {
 			err = host1x_subdev_add(device, driver, np);
-			if (err < 0) {
-				of_node_put(np);
+			if (err < 0)
 				return err;
-			}
 		}
 	}
 
diff --git a/drivers/gpu/host1x/cdma.c b/drivers/gpu/host1x/cdma.c
index 407ed9b9cf64..ba2e572567c0 100644
--- a/drivers/gpu/host1x/cdma.c
+++ b/drivers/gpu/host1x/cdma.c
@@ -247,8 +247,6 @@ static int host1x_cdma_wait_pushbuffer_space(struct host1x *host1x,
 		trace_host1x_wait_cdma(dev_name(cdma_to_channel(cdma)->dev),
 				       CDMA_EVENT_PUSH_BUFFER_SPACE);
 
-		host1x_hw_cdma_flush(host1x, cdma);
-
 		/* If somebody has managed to already start waiting, yield */
 		if (cdma->event != CDMA_EVENT_NONE) {
 			mutex_unlock(&cdma->lock);
@@ -591,7 +589,6 @@ int host1x_cdma_begin(struct host1x_cdma *cdma, struct host1x_job *job)
  */
 void host1x_cdma_push(struct host1x_cdma *cdma, u32 op1, u32 op2)
 {
-	struct host1x *host1x = cdma_to_host1x(cdma);
 	struct push_buffer *pb = &cdma->push_buffer;
 	u32 slots_free = cdma->slots_free;
 
@@ -599,11 +596,9 @@ void host1x_cdma_push(struct host1x_cdma *cdma, u32 op1, u32 op2)
 		trace_host1x_cdma_push(dev_name(cdma_to_channel(cdma)->dev),
 				       op1, op2);
 
-	if (slots_free == 0) {
-		host1x_hw_cdma_flush(host1x, cdma);
+	if (slots_free == 0)
 		slots_free = host1x_cdma_wait_locked(cdma,
 						CDMA_EVENT_PUSH_BUFFER_SPACE);
-	}
 
 	cdma->slots_free = slots_free - 1;
 	cdma->slots_used++;
diff --git a/drivers/hv/hv_common.c b/drivers/hv/hv_common.c
index b3b11be11650..59792e00cecf 100644
--- a/drivers/hv/hv_common.c
+++ b/drivers/hv/hv_common.c
@@ -307,7 +307,7 @@ void __init hv_get_partition_id(void)
 
 	local_irq_save(flags);
 	output = *this_cpu_ptr(hyperv_pcpu_input_arg);
-	status = hv_do_hypercall(HVCALL_GET_PARTITION_ID, NULL, &output);
+	status = hv_do_hypercall(HVCALL_GET_PARTITION_ID, NULL, output);
 	pt_id = output->partition_id;
 	local_irq_restore(flags);
 
@@ -566,9 +566,11 @@ int hv_common_cpu_die(unsigned int cpu)
 	 * originally allocated memory is reused in hv_common_cpu_init().
 	 */
 
-	synic_eventring_tail = this_cpu_ptr(hv_synic_eventring_tail);
-	kfree(*synic_eventring_tail);
-	*synic_eventring_tail = NULL;
+	if (hv_root_partition()) {
+		synic_eventring_tail = this_cpu_ptr(hv_synic_eventring_tail);
+		kfree(*synic_eventring_tail);
+		*synic_eventring_tail = NULL;
+	}
 
 	return 0;
 }
diff --git a/drivers/hwtracing/intel_th/Kconfig b/drivers/hwtracing/intel_th/Kconfig
index 4b6359326ede..4f7d2b6d79e2 100644
--- a/drivers/hwtracing/intel_th/Kconfig
+++ b/drivers/hwtracing/intel_th/Kconfig
@@ -60,6 +60,7 @@ config INTEL_TH_STH
 
 config INTEL_TH_MSU
 	tristate "Intel(R) Trace Hub Memory Storage Unit"
+	depends on MMU
 	help
 	  Memory Storage Unit (MSU) trace output device enables
 	  storing STP traces to system memory. It supports single
diff --git a/drivers/hwtracing/intel_th/msu.c b/drivers/hwtracing/intel_th/msu.c
index bf99d79a4192..7163950eb371 100644
--- a/drivers/hwtracing/intel_th/msu.c
+++ b/drivers/hwtracing/intel_th/msu.c
@@ -19,6 +19,7 @@
 #include <linux/io.h>
 #include <linux/workqueue.h>
 #include <linux/dma-mapping.h>
+#include <linux/pfn_t.h>
 
 #ifdef CONFIG_X86
 #include <asm/set_memory.h>
@@ -976,7 +977,6 @@ static void msc_buffer_contig_free(struct msc *msc)
 	for (off = 0; off < msc->nr_pages << PAGE_SHIFT; off += PAGE_SIZE) {
 		struct page *page = virt_to_page(msc->base + off);
 
-		page->mapping = NULL;
 		__free_page(page);
 	}
 
@@ -1158,9 +1158,6 @@ static void __msc_buffer_win_free(struct msc *msc, struct msc_window *win)
 	int i;
 
 	for_each_sg(win->sgt->sgl, sg, win->nr_segs, i) {
-		struct page *page = msc_sg_page(sg);
-
-		page->mapping = NULL;
 		dma_free_coherent(msc_dev(win->msc)->parent->parent, PAGE_SIZE,
 				  sg_virt(sg), sg_dma_address(sg));
 	}
@@ -1601,22 +1598,10 @@ static void msc_mmap_close(struct vm_area_struct *vma)
 {
 	struct msc_iter *iter = vma->vm_file->private_data;
 	struct msc *msc = iter->msc;
-	unsigned long pg;
 
 	if (!atomic_dec_and_mutex_lock(&msc->mmap_count, &msc->buf_mutex))
 		return;
 
-	/* drop page _refcounts */
-	for (pg = 0; pg < msc->nr_pages; pg++) {
-		struct page *page = msc_buffer_get_page(msc, pg);
-
-		if (WARN_ON_ONCE(!page))
-			continue;
-
-		if (page->mapping)
-			page->mapping = NULL;
-	}
-
 	/* last mapping -- drop user_count */
 	atomic_dec(&msc->user_count);
 	mutex_unlock(&msc->buf_mutex);
@@ -1626,16 +1611,14 @@ static vm_fault_t msc_mmap_fault(struct vm_fault *vmf)
 {
 	struct msc_iter *iter = vmf->vma->vm_file->private_data;
 	struct msc *msc = iter->msc;
+	struct page *page;
 
-	vmf->page = msc_buffer_get_page(msc, vmf->pgoff);
-	if (!vmf->page)
+	page = msc_buffer_get_page(msc, vmf->pgoff);
+	if (!page)
 		return VM_FAULT_SIGBUS;
 
-	get_page(vmf->page);
-	vmf->page->mapping = vmf->vma->vm_file->f_mapping;
-	vmf->page->index = vmf->pgoff;
-
-	return 0;
+	get_page(page);
+	return vmf_insert_mixed(vmf->vma, vmf->address, page_to_pfn_t(page));
 }
 
 static const struct vm_operations_struct msc_mmap_ops = {
@@ -1676,7 +1659,7 @@ out:
 		atomic_dec(&msc->user_count);
 
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
-	vm_flags_set(vma, VM_DONTEXPAND | VM_DONTCOPY);
+	vm_flags_set(vma, VM_DONTEXPAND | VM_DONTCOPY | VM_MIXEDMAP);
 	vma->vm_ops = &msc_mmap_ops;
 	return ret;
 }
diff --git a/drivers/i2c/busses/i2c-cros-ec-tunnel.c b/drivers/i2c/busses/i2c-cros-ec-tunnel.c
index 43bf90d90eeb..208ce4f9e782 100644
--- a/drivers/i2c/busses/i2c-cros-ec-tunnel.c
+++ b/drivers/i2c/busses/i2c-cros-ec-tunnel.c
@@ -247,6 +247,9 @@ static int ec_i2c_probe(struct platform_device *pdev)
 	u32 remote_bus;
 	int err;
 
+	if (!ec)
+		return dev_err_probe(dev, -EPROBE_DEFER, "couldn't find parent EC device\n");
+
 	if (!ec->cmd_xfer) {
 		dev_err(dev, "Missing sendrecv\n");
 		return -EINVAL;
diff --git a/drivers/i2c/busses/i2c-imx-lpi2c.c b/drivers/i2c/busses/i2c-imx-lpi2c.c
index 0d4b3935e687..342d47e67586 100644
--- a/drivers/i2c/busses/i2c-imx-lpi2c.c
+++ b/drivers/i2c/busses/i2c-imx-lpi2c.c
@@ -1380,9 +1380,9 @@ static int lpi2c_imx_probe(struct platform_device *pdev)
 	return 0;
 
 rpm_disable:
-	pm_runtime_put(&pdev->dev);
-	pm_runtime_disable(&pdev->dev);
 	pm_runtime_dont_use_autosuspend(&pdev->dev);
+	pm_runtime_put_sync(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
 
 	return ret;
 }
diff --git a/drivers/i2c/i2c-atr.c b/drivers/i2c/i2c-atr.c
index 8fe9ddff8e96..783fb8df2ebe 100644
--- a/drivers/i2c/i2c-atr.c
+++ b/drivers/i2c/i2c-atr.c
@@ -8,12 +8,12 @@
  * Originally based on i2c-mux.c
  */
 
-#include <linux/fwnode.h>
 #include <linux/i2c-atr.h>
 #include <linux/i2c.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <linux/property.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
diff --git a/drivers/iio/magnetometer/ak8974.c b/drivers/iio/magnetometer/ak8974.c
index 08975c60e325..7bc341c69697 100644
--- a/drivers/iio/magnetometer/ak8974.c
+++ b/drivers/iio/magnetometer/ak8974.c
@@ -535,8 +535,8 @@ static int ak8974_detect(struct ak8974 *ak8974)
 				       fab_data2, sizeof(fab_data2));
 
 		for (i = 0; i < 3; ++i) {
-			static const char axis[3] = "XYZ";
-			static const char pgaxis[6] = "ZYZXYX";
+			static const char axis[] = "XYZ";
+			static const char pgaxis[] = "ZYZXYX";
 			unsigned offz = le16_to_cpu(fab_data2[i]) & 0x7F;
 			unsigned fine = le16_to_cpu(fab_data1[i]);
 			unsigned sens = le16_to_cpu(fab_data1[i + 3]);
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index fedcdb56fb6b..ab31eefa916b 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -72,6 +72,8 @@ static const char * const cma_events[] = {
 static void cma_iboe_set_mgid(struct sockaddr *addr, union ib_gid *mgid,
 			      enum ib_gid_type gid_type);
 
+static void cma_netevent_work_handler(struct work_struct *_work);
+
 const char *__attribute_const__ rdma_event_msg(enum rdma_cm_event_type event)
 {
 	size_t index = event;
@@ -1047,6 +1049,7 @@ __rdma_create_id(struct net *net, rdma_cm_event_handler event_handler,
 	get_random_bytes(&id_priv->seq_num, sizeof id_priv->seq_num);
 	id_priv->id.route.addr.dev_addr.net = get_net(net);
 	id_priv->seq_num &= 0x00ffffff;
+	INIT_WORK(&id_priv->id.net_work, cma_netevent_work_handler);
 
 	rdma_restrack_new(&id_priv->res, RDMA_RESTRACK_CM_ID);
 	if (parent)
@@ -5241,7 +5244,6 @@ static int cma_netevent_callback(struct notifier_block *self,
 		if (!memcmp(current_id->id.route.addr.dev_addr.dst_dev_addr,
 			   neigh->ha, ETH_ALEN))
 			continue;
-		INIT_WORK(&current_id->id.net_work, cma_netevent_work_handler);
 		cma_id_get(current_id);
 		queue_work(cma_wq, &current_id->id.net_work);
 	}
diff --git a/drivers/infiniband/core/ucaps.c b/drivers/infiniband/core/ucaps.c
index 6853c6d078f9..de5cb8bf0a61 100644
--- a/drivers/infiniband/core/ucaps.c
+++ b/drivers/infiniband/core/ucaps.c
@@ -170,7 +170,7 @@ int ib_create_ucap(enum rdma_user_cap type)
 	ucap->dev.class = &ucaps_class;
 	ucap->dev.devt = MKDEV(MAJOR(ucaps_base_dev), type);
 	ucap->dev.release = ucap_dev_release;
-	ret = dev_set_name(&ucap->dev, ucap_names[type]);
+	ret = dev_set_name(&ucap->dev, "%s", ucap_names[type]);
 	if (ret)
 		goto err_device;
 
diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index e9fa22d31c23..c48ef6083020 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -76,12 +76,14 @@ static inline int ib_init_umem_odp(struct ib_umem_odp *umem_odp,
 
 		npfns = (end - start) >> PAGE_SHIFT;
 		umem_odp->pfn_list = kvcalloc(
-			npfns, sizeof(*umem_odp->pfn_list), GFP_KERNEL);
+			npfns, sizeof(*umem_odp->pfn_list),
+			GFP_KERNEL | __GFP_NOWARN);
 		if (!umem_odp->pfn_list)
 			return -ENOMEM;
 
 		umem_odp->dma_list = kvcalloc(
-			ndmas, sizeof(*umem_odp->dma_list), GFP_KERNEL);
+			ndmas, sizeof(*umem_odp->dma_list),
+			GFP_KERNEL | __GFP_NOWARN);
 		if (!umem_odp->dma_list) {
 			ret = -ENOMEM;
 			goto out_pfn_list;
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index 9082b3fd2b47..063801384b2b 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -1774,10 +1774,7 @@ int bnxt_re_destroy_srq(struct ib_srq *ib_srq, struct ib_udata *udata)
 					       ib_srq);
 	struct bnxt_re_dev *rdev = srq->rdev;
 	struct bnxt_qplib_srq *qplib_srq = &srq->qplib_srq;
-	struct bnxt_qplib_nq *nq = NULL;
 
-	if (qplib_srq->cq)
-		nq = qplib_srq->cq->nq;
 	if (rdev->chip_ctx->modes.toggle_bits & BNXT_QPLIB_SRQ_TOGGLE_BIT) {
 		free_page((unsigned long)srq->uctx_srq_page);
 		hash_del(&srq->hash_entry);
@@ -1785,8 +1782,6 @@ int bnxt_re_destroy_srq(struct ib_srq *ib_srq, struct ib_udata *udata)
 	bnxt_qplib_destroy_srq(&rdev->qplib_res, qplib_srq);
 	ib_umem_release(srq->umem);
 	atomic_dec(&rdev->stats.res.srq_count);
-	if (nq)
-		nq->budget--;
 	return 0;
 }
 
@@ -1827,7 +1822,6 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq,
 		       struct ib_udata *udata)
 {
 	struct bnxt_qplib_dev_attr *dev_attr;
-	struct bnxt_qplib_nq *nq = NULL;
 	struct bnxt_re_ucontext *uctx;
 	struct bnxt_re_dev *rdev;
 	struct bnxt_re_srq *srq;
@@ -1873,7 +1867,6 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq,
 	srq->qplib_srq.eventq_hw_ring_id = rdev->nqr->nq[0].ring_id;
 	srq->qplib_srq.sg_info.pgsize = PAGE_SIZE;
 	srq->qplib_srq.sg_info.pgshft = PAGE_SHIFT;
-	nq = &rdev->nqr->nq[0];
 
 	if (udata) {
 		rc = bnxt_re_init_user_srq(rdev, pd, srq, udata);
@@ -1908,8 +1901,6 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq,
 			goto fail;
 		}
 	}
-	if (nq)
-		nq->budget++;
 	active_srqs = atomic_inc_return(&rdev->stats.res.srq_count);
 	if (active_srqs > rdev->stats.res.srq_watermark)
 		rdev->stats.res.srq_watermark = active_srqs;
@@ -3079,7 +3070,6 @@ int bnxt_re_destroy_cq(struct ib_cq *ib_cq, struct ib_udata *udata)
 	ib_umem_release(cq->umem);
 
 	atomic_dec(&rdev->stats.res.cq_count);
-	nq->budget--;
 	kfree(cq->cql);
 	return 0;
 }
diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c
index cf89a8db4f64..8d0b63d4b50a 100644
--- a/drivers/infiniband/hw/hns/hns_roce_main.c
+++ b/drivers/infiniband/hw/hns/hns_roce_main.c
@@ -763,7 +763,7 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev)
 		if (ret)
 			return ret;
 	}
-	dma_set_max_seg_size(dev, UINT_MAX);
+	dma_set_max_seg_size(dev, SZ_2G);
 	ret = ib_register_device(ib_dev, "hns_%d", dev);
 	if (ret) {
 		dev_err(dev, "ib_register_device failed!\n");
diff --git a/drivers/infiniband/hw/mlx5/fs.c b/drivers/infiniband/hw/mlx5/fs.c
index 251246c73b33..0ff9f18a71e8 100644
--- a/drivers/infiniband/hw/mlx5/fs.c
+++ b/drivers/infiniband/hw/mlx5/fs.c
@@ -3461,7 +3461,6 @@ DECLARE_UVERBS_NAMED_OBJECT(
 	&UVERBS_METHOD(MLX5_IB_METHOD_STEERING_ANCHOR_DESTROY));
 
 const struct uapi_definition mlx5_ib_flow_defs[] = {
-#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS)
 	UAPI_DEF_CHAIN_OBJ_TREE_NAMED(
 		MLX5_IB_OBJECT_FLOW_MATCHER),
 	UAPI_DEF_CHAIN_OBJ_TREE(
@@ -3472,7 +3471,6 @@ const struct uapi_definition mlx5_ib_flow_defs[] = {
 	UAPI_DEF_CHAIN_OBJ_TREE_NAMED(
 		MLX5_IB_OBJECT_STEERING_ANCHOR,
 		UAPI_DEF_IS_OBJ_SUPPORTED(mlx5_ib_shared_ft_allowed)),
-#endif
 	{},
 };
 
diff --git a/drivers/infiniband/hw/usnic/usnic_ib_main.c b/drivers/infiniband/hw/usnic/usnic_ib_main.c
index 4ddcd5860e0f..11eca39b73a9 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_main.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_main.c
@@ -397,7 +397,7 @@ static void *usnic_ib_device_add(struct pci_dev *dev)
 	if (!us_ibdev) {
 		usnic_err("Device %s context alloc failed\n",
 				netdev_name(pci_get_drvdata(dev)));
-		return ERR_PTR(-EFAULT);
+		return NULL;
 	}
 
 	us_ibdev->ufdev = usnic_fwd_dev_alloc(dev);
@@ -517,8 +517,8 @@ static struct usnic_ib_dev *usnic_ib_discover_pf(struct usnic_vnic *vnic)
 	}
 
 	us_ibdev = usnic_ib_device_add(parent_pci);
-	if (IS_ERR_OR_NULL(us_ibdev)) {
-		us_ibdev = us_ibdev ? us_ibdev : ERR_PTR(-EFAULT);
+	if (!us_ibdev) {
+		us_ibdev = ERR_PTR(-EFAULT);
 		goto out;
 	}
 
@@ -586,10 +586,10 @@ static int usnic_ib_pci_probe(struct pci_dev *pdev,
 	}
 
 	pf = usnic_ib_discover_pf(vf->vnic);
-	if (IS_ERR_OR_NULL(pf)) {
-		usnic_err("Failed to discover pf of vnic %s with err%ld\n",
-				pci_name(pdev), PTR_ERR(pf));
-		err = pf ? PTR_ERR(pf) : -EFAULT;
+	if (IS_ERR(pf)) {
+		err = PTR_ERR(pf);
+		usnic_err("Failed to discover pf of vnic %s with err%d\n",
+				pci_name(pdev), err);
 		goto out_clean_vnic;
 	}
 
diff --git a/drivers/infiniband/sw/rxe/rxe_loc.h b/drivers/infiniband/sw/rxe/rxe_loc.h
index feb386d98d1d..0bc3fbb6554f 100644
--- a/drivers/infiniband/sw/rxe/rxe_loc.h
+++ b/drivers/infiniband/sw/rxe/rxe_loc.h
@@ -140,6 +140,12 @@ static inline int qp_mtu(struct rxe_qp *qp)
 		return IB_MTU_4096;
 }
 
+static inline bool is_odp_mr(struct rxe_mr *mr)
+{
+	return IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING) && mr->umem &&
+	       mr->umem->is_odp;
+}
+
 void free_rd_atomic_resource(struct resp_res *res);
 
 static inline void rxe_advance_resp_resource(struct rxe_qp *qp)
diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c
index 868d2f0b74e9..432d864c3ce9 100644
--- a/drivers/infiniband/sw/rxe/rxe_mr.c
+++ b/drivers/infiniband/sw/rxe/rxe_mr.c
@@ -323,7 +323,7 @@ int rxe_mr_copy(struct rxe_mr *mr, u64 iova, void *addr,
 		return err;
 	}
 
-	if (mr->umem->is_odp)
+	if (is_odp_mr(mr))
 		return rxe_odp_mr_copy(mr, iova, addr, length, dir);
 	else
 		return rxe_mr_copy_xarray(mr, iova, addr, length, dir);
@@ -536,7 +536,7 @@ int rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value)
 	u64 *va;
 
 	/* ODP is not supported right now. WIP. */
-	if (mr->umem->is_odp)
+	if (is_odp_mr(mr))
 		return RESPST_ERR_UNSUPPORTED_OPCODE;
 
 	/* See IBA oA19-28 */
diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c
index 54ba9ee1acc5..5d9174e408db 100644
--- a/drivers/infiniband/sw/rxe/rxe_resp.c
+++ b/drivers/infiniband/sw/rxe/rxe_resp.c
@@ -650,7 +650,7 @@ static enum resp_states process_flush(struct rxe_qp *qp,
 	struct resp_res *res = qp->resp.res;
 
 	/* ODP is not supported right now. WIP. */
-	if (mr->umem->is_odp)
+	if (is_odp_mr(mr))
 		return RESPST_ERR_UNSUPPORTED_OPCODE;
 
 	/* oA19-14, oA19-15 */
@@ -706,7 +706,7 @@ static enum resp_states atomic_reply(struct rxe_qp *qp,
 	if (!res->replay) {
 		u64 iova = qp->resp.va + qp->resp.offset;
 
-		if (mr->umem->is_odp)
+		if (is_odp_mr(mr))
 			err = rxe_odp_atomic_op(mr, iova, pkt->opcode,
 						atmeth_comp(pkt),
 						atmeth_swap_add(pkt),
diff --git a/drivers/input/joystick/magellan.c b/drivers/input/joystick/magellan.c
index 2eaa25c9c68c..d73389af4dd5 100644
--- a/drivers/input/joystick/magellan.c
+++ b/drivers/input/joystick/magellan.c
@@ -48,7 +48,7 @@ struct magellan {
 
 static int magellan_crunch_nibbles(unsigned char *data, int count)
 {
-	static unsigned char nibbles[16] = "0AB3D56GH9:K<MN?";
+	static unsigned char nibbles[16] __nonstring = "0AB3D56GH9:K<MN?";
 
 	do {
 		if (data[count] == nibbles[data[count] & 0xf])
diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c
index dd9e26b7b718..14aa0d77df26 100644
--- a/drivers/iommu/amd/init.c
+++ b/drivers/iommu/amd/init.c
@@ -3664,6 +3664,14 @@ found:
 	while (*uid == '0' && *(uid + 1))
 		uid++;
 
+	if (strlen(hid) >= ACPIHID_HID_LEN) {
+		pr_err("Invalid command line: hid is too long\n");
+		return 1;
+	} else if (strlen(uid) >= ACPIHID_UID_LEN) {
+		pr_err("Invalid command line: uid is too long\n");
+		return 1;
+	}
+
 	i = early_acpihid_map_size++;
 	memcpy(early_acpihid_map[i].hid, hid, strlen(hid));
 	memcpy(early_acpihid_map[i].uid, uid, strlen(uid));
diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c
index be8761bbef0f..f34209b08b4c 100644
--- a/drivers/iommu/amd/iommu.c
+++ b/drivers/iommu/amd/iommu.c
@@ -3869,6 +3869,9 @@ static int amd_ir_set_vcpu_affinity(struct irq_data *data, void *vcpu_info)
 	struct irq_2_irte *irte_info = &ir_data->irq_2_irte;
 	struct iommu_dev_data *dev_data;
 
+	if (WARN_ON_ONCE(!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir)))
+		return -EINVAL;
+
 	if (ir_data->iommu == NULL)
 		return -EINVAL;
 
@@ -3879,21 +3882,11 @@ static int amd_ir_set_vcpu_affinity(struct irq_data *data, void *vcpu_info)
 	 * we should not modify the IRTE
 	 */
 	if (!dev_data || !dev_data->use_vapic)
-		return 0;
+		return -EINVAL;
 
 	ir_data->cfg = irqd_cfg(data);
 	pi_data->ir_data = ir_data;
 
-	/* Note:
-	 * SVM tries to set up for VAPIC mode, but we are in
-	 * legacy mode. So, we force legacy mode instead.
-	 */
-	if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir)) {
-		pr_debug("%s: Fall back to using intr legacy remap\n",
-			 __func__);
-		pi_data->is_guest_mode = false;
-	}
-
 	pi_data->prev_ga_tag = ir_data->cached_ga_tag;
 	if (pi_data->is_guest_mode) {
 		ir_data->ga_root_ptr = (pi_data->base >> 12);
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
index 9ba596430e7c..980cc6b33c43 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c
@@ -411,6 +411,12 @@ struct iommu_domain *arm_smmu_sva_domain_alloc(struct device *dev,
 		return ERR_CAST(smmu_domain);
 	smmu_domain->domain.type = IOMMU_DOMAIN_SVA;
 	smmu_domain->domain.ops = &arm_smmu_sva_domain_ops;
+
+	/*
+	 * Choose page_size as the leaf page size for invalidation when
+	 * ARM_SMMU_FEAT_RANGE_INV is present
+	 */
+	smmu_domain->domain.pgsize_bitmap = PAGE_SIZE;
 	smmu_domain->smmu = smmu;
 
 	ret = xa_alloc(&arm_smmu_asid_xa, &asid, smmu_domain,
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index b4c21aaed126..48d910399a1b 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -3388,6 +3388,7 @@ static int arm_smmu_insert_master(struct arm_smmu_device *smmu,
 	mutex_lock(&smmu->streams_mutex);
 	for (i = 0; i < fwspec->num_ids; i++) {
 		struct arm_smmu_stream *new_stream = &master->streams[i];
+		struct rb_node *existing;
 		u32 sid = fwspec->ids[i];
 
 		new_stream->id = sid;
@@ -3398,11 +3399,21 @@ static int arm_smmu_insert_master(struct arm_smmu_device *smmu,
 			break;
 
 		/* Insert into SID tree */
-		if (rb_find_add(&new_stream->node, &smmu->streams,
-				arm_smmu_streams_cmp_node)) {
-			dev_warn(master->dev, "stream %u already in tree\n",
-				 sid);
-			ret = -EINVAL;
+		existing = rb_find_add(&new_stream->node, &smmu->streams,
+				       arm_smmu_streams_cmp_node);
+		if (existing) {
+			struct arm_smmu_master *existing_master =
+				rb_entry(existing, struct arm_smmu_stream, node)
+					->master;
+
+			/* Bridged PCI devices may end up with duplicated IDs */
+			if (existing_master == master)
+				continue;
+
+			dev_warn(master->dev,
+				 "Aliasing StreamID 0x%x (from %s) unsupported, expect DMA to be broken\n",
+				 sid, dev_name(existing_master->dev));
+			ret = -ENODEV;
 			break;
 		}
 	}
@@ -4429,6 +4440,8 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR3);
 	if (FIELD_GET(IDR3_RIL, reg))
 		smmu->features |= ARM_SMMU_FEAT_RANGE_INV;
+	if (FIELD_GET(IDR3_FWB, reg))
+		smmu->features |= ARM_SMMU_FEAT_S2FWB;
 
 	/* IDR5 */
 	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR5);
diff --git a/drivers/iommu/arm/arm-smmu-v3/tegra241-cmdqv.c b/drivers/iommu/arm/arm-smmu-v3/tegra241-cmdqv.c
index d525ab43a4ae..dd7d030d2e89 100644
--- a/drivers/iommu/arm/arm-smmu-v3/tegra241-cmdqv.c
+++ b/drivers/iommu/arm/arm-smmu-v3/tegra241-cmdqv.c
@@ -487,17 +487,6 @@ static int tegra241_cmdqv_hw_reset(struct arm_smmu_device *smmu)
 
 /* VCMDQ Resource Helpers */
 
-static void tegra241_vcmdq_free_smmu_cmdq(struct tegra241_vcmdq *vcmdq)
-{
-	struct arm_smmu_queue *q = &vcmdq->cmdq.q;
-	size_t nents = 1 << q->llq.max_n_shift;
-	size_t qsz = nents << CMDQ_ENT_SZ_SHIFT;
-
-	if (!q->base)
-		return;
-	dmam_free_coherent(vcmdq->cmdqv->smmu.dev, qsz, q->base, q->base_dma);
-}
-
 static int tegra241_vcmdq_alloc_smmu_cmdq(struct tegra241_vcmdq *vcmdq)
 {
 	struct arm_smmu_device *smmu = &vcmdq->cmdqv->smmu;
@@ -560,7 +549,8 @@ static void tegra241_vintf_free_lvcmdq(struct tegra241_vintf *vintf, u16 lidx)
 	struct tegra241_vcmdq *vcmdq = vintf->lvcmdqs[lidx];
 	char header[64];
 
-	tegra241_vcmdq_free_smmu_cmdq(vcmdq);
+	/* Note that the lvcmdq queue memory space is managed by devres */
+
 	tegra241_vintf_deinit_lvcmdq(vintf, lidx);
 
 	dev_dbg(vintf->cmdqv->dev,
@@ -768,13 +758,13 @@ static int tegra241_cmdqv_init_structures(struct arm_smmu_device *smmu)
 
 	vintf = kzalloc(sizeof(*vintf), GFP_KERNEL);
 	if (!vintf)
-		goto out_fallback;
+		return -ENOMEM;
 
 	/* Init VINTF0 for in-kernel use */
 	ret = tegra241_cmdqv_init_vintf(cmdqv, 0, vintf);
 	if (ret) {
 		dev_err(cmdqv->dev, "failed to init vintf0: %d\n", ret);
-		goto free_vintf;
+		return ret;
 	}
 
 	/* Preallocate logical VCMDQs to VINTF0 */
@@ -783,24 +773,12 @@ static int tegra241_cmdqv_init_structures(struct arm_smmu_device *smmu)
 
 		vcmdq = tegra241_vintf_alloc_lvcmdq(vintf, lidx);
 		if (IS_ERR(vcmdq))
-			goto free_lvcmdq;
+			return PTR_ERR(vcmdq);
 	}
 
 	/* Now, we are ready to run all the impl ops */
 	smmu->impl_ops = &tegra241_cmdqv_impl_ops;
 	return 0;
-
-free_lvcmdq:
-	for (lidx--; lidx >= 0; lidx--)
-		tegra241_vintf_free_lvcmdq(vintf, lidx);
-	tegra241_cmdqv_deinit_vintf(cmdqv, vintf->idx);
-free_vintf:
-	kfree(vintf);
-out_fallback:
-	dev_info(smmu->impl_dev, "Falling back to standard SMMU CMDQ\n");
-	smmu->options &= ~ARM_SMMU_OPT_TEGRA241_CMDQV;
-	tegra241_cmdqv_remove(smmu);
-	return 0;
 }
 
 #ifdef CONFIG_IOMMU_DEBUGFS
diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
index cb7e29dcac15..a775e4dbe06f 100644
--- a/drivers/iommu/dma-iommu.c
+++ b/drivers/iommu/dma-iommu.c
@@ -1754,7 +1754,7 @@ static size_t cookie_msi_granule(const struct iommu_domain *domain)
 		return PAGE_SIZE;
 	default:
 		BUG();
-	};
+	}
 }
 
 static struct list_head *cookie_msi_pages(const struct iommu_domain *domain)
@@ -1766,7 +1766,7 @@ static struct list_head *cookie_msi_pages(const struct iommu_domain *domain)
 		return &domain->msi_cookie->msi_page_list;
 	default:
 		BUG();
-	};
+	}
 }
 
 static struct iommu_dma_msi_page *iommu_dma_get_msi_page(struct device *dev,
diff --git a/drivers/iommu/exynos-iommu.c b/drivers/iommu/exynos-iommu.c
index 69e23e017d9e..317266aca6e2 100644
--- a/drivers/iommu/exynos-iommu.c
+++ b/drivers/iommu/exynos-iommu.c
@@ -832,7 +832,7 @@ static int __maybe_unused exynos_sysmmu_suspend(struct device *dev)
 		struct exynos_iommu_owner *owner = dev_iommu_priv_get(master);
 
 		mutex_lock(&owner->rpm_lock);
-		if (&data->domain->domain != &exynos_identity_domain) {
+		if (data->domain) {
 			dev_dbg(data->sysmmu, "saving state\n");
 			__sysmmu_disable(data);
 		}
@@ -850,7 +850,7 @@ static int __maybe_unused exynos_sysmmu_resume(struct device *dev)
 		struct exynos_iommu_owner *owner = dev_iommu_priv_get(master);
 
 		mutex_lock(&owner->rpm_lock);
-		if (&data->domain->domain != &exynos_identity_domain) {
+		if (data->domain) {
 			dev_dbg(data->sysmmu, "restoring state\n");
 			__sysmmu_enable(data);
 		}
diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
index 6e67cc66a204..cb0b993bebb4 100644
--- a/drivers/iommu/intel/iommu.c
+++ b/drivers/iommu/intel/iommu.c
@@ -3785,6 +3785,22 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
 
 	intel_iommu_debugfs_create_dev(info);
 
+	return &iommu->iommu;
+free_table:
+	intel_pasid_free_table(dev);
+clear_rbtree:
+	device_rbtree_remove(info);
+free:
+	kfree(info);
+
+	return ERR_PTR(ret);
+}
+
+static void intel_iommu_probe_finalize(struct device *dev)
+{
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
+
 	/*
 	 * The PCIe spec, in its wisdom, declares that the behaviour of the
 	 * device is undefined if you enable PASID support after ATS support.
@@ -3792,22 +3808,12 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
 	 * we can't yet know if we're ever going to use it.
 	 */
 	if (info->pasid_supported &&
-	    !pci_enable_pasid(pdev, info->pasid_supported & ~1))
+	    !pci_enable_pasid(to_pci_dev(dev), info->pasid_supported & ~1))
 		info->pasid_enabled = 1;
 
-	if (sm_supported(iommu))
+	if (sm_supported(iommu) && !dev_is_real_dma_subdevice(dev))
 		iommu_enable_pci_ats(info);
 	iommu_enable_pci_pri(info);
-
-	return &iommu->iommu;
-free_table:
-	intel_pasid_free_table(dev);
-clear_rbtree:
-	device_rbtree_remove(info);
-free:
-	kfree(info);
-
-	return ERR_PTR(ret);
 }
 
 static void intel_iommu_release_device(struct device *dev)
@@ -3835,7 +3841,6 @@ static void intel_iommu_release_device(struct device *dev)
 	intel_pasid_free_table(dev);
 	intel_iommu_debugfs_remove_dev(info);
 	kfree(info);
-	set_dma_ops(dev, NULL);
 }
 
 static void intel_iommu_get_resv_regions(struct device *device,
@@ -4392,6 +4397,7 @@ const struct iommu_ops intel_iommu_ops = {
 	.domain_alloc_sva	= intel_svm_domain_alloc,
 	.domain_alloc_nested	= intel_iommu_domain_alloc_nested,
 	.probe_device		= intel_iommu_probe_device,
+	.probe_finalize		= intel_iommu_probe_finalize,
 	.release_device		= intel_iommu_release_device,
 	.get_resv_regions	= intel_iommu_get_resv_regions,
 	.device_group		= intel_iommu_device_group,
@@ -4433,6 +4439,9 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e30, quirk_iommu_igfx);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e40, quirk_iommu_igfx);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e90, quirk_iommu_igfx);
 
+/* QM57/QS57 integrated gfx malfunctions with dmar */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_iommu_igfx);
+
 /* Broadwell igfx malfunctions with dmar */
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x1606, quirk_iommu_igfx);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x160B, quirk_iommu_igfx);
@@ -4510,7 +4519,6 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev)
 	}
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0040, quirk_calpella_no_shadow_gtt);
-DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_calpella_no_shadow_gtt);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0062, quirk_calpella_no_shadow_gtt);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x006a, quirk_calpella_no_shadow_gtt);
 
diff --git a/drivers/iommu/intel/irq_remapping.c b/drivers/iommu/intel/irq_remapping.c
index ea3ca5203919..3bc2a03cceca 100644
--- a/drivers/iommu/intel/irq_remapping.c
+++ b/drivers/iommu/intel/irq_remapping.c
@@ -1287,43 +1287,44 @@ static struct irq_chip intel_ir_chip = {
 };
 
 /*
- * With posted MSIs, all vectors are multiplexed into a single notification
- * vector. Devices MSIs are then dispatched in a demux loop where
- * EOIs can be coalesced as well.
+ * With posted MSIs, the MSI vectors are multiplexed into a single notification
+ * vector, and only the notification vector is sent to the APIC IRR.  Device
+ * MSIs are then dispatched in a demux loop that harvests the MSIs from the
+ * CPU's Posted Interrupt Request bitmap.  I.e. Posted MSIs never get sent to
+ * the APIC IRR, and thus do not need an EOI.  The notification handler instead
+ * performs a single EOI after processing the PIR.
  *
- * "INTEL-IR-POST" IRQ chip does not do EOI on ACK, thus the dummy irq_ack()
- * function. Instead EOI is performed by the posted interrupt notification
- * handler.
+ * Note!  Pending SMP/CPU affinity changes, which are per MSI, must still be
+ * honored, only the APIC EOI is omitted.
  *
  * For the example below, 3 MSIs are coalesced into one CPU notification. Only
- * one apic_eoi() is needed.
+ * one apic_eoi() is needed, but each MSI needs to process pending changes to
+ * its CPU affinity.
  *
  * __sysvec_posted_msi_notification()
  *	irq_enter();
  *		handle_edge_irq()
  *			irq_chip_ack_parent()
- *				dummy(); // No EOI
+ *				irq_move_irq(); // No EOI
  *			handle_irq_event()
  *				driver_handler()
  *		handle_edge_irq()
  *			irq_chip_ack_parent()
- *				dummy(); // No EOI
+ *				irq_move_irq(); // No EOI
  *			handle_irq_event()
  *				driver_handler()
  *		handle_edge_irq()
  *			irq_chip_ack_parent()
- *				dummy(); // No EOI
+ *				irq_move_irq(); // No EOI
  *			handle_irq_event()
  *				driver_handler()
  *	apic_eoi()
  *	irq_exit()
+ *
  */
-
-static void dummy_ack(struct irq_data *d) { }
-
 static struct irq_chip intel_ir_chip_post_msi = {
 	.name			= "INTEL-IR-POST",
-	.irq_ack		= dummy_ack,
+	.irq_ack		= irq_move_irq,
 	.irq_set_affinity	= intel_ir_set_affinity,
 	.irq_compose_msi_msg	= intel_ir_compose_msi_msg,
 	.irq_set_vcpu_affinity	= intel_ir_set_vcpu_affinity,
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index c8033ca66377..4f91a740c15f 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -538,6 +538,9 @@ static void iommu_deinit_device(struct device *dev)
 	dev->iommu_group = NULL;
 	module_put(ops->owner);
 	dev_iommu_free(dev);
+#ifdef CONFIG_IOMMU_DMA
+	dev->dma_iommu = false;
+#endif
 }
 
 static struct iommu_domain *pasid_array_entry_to_domain(void *entry)
@@ -2717,7 +2720,8 @@ int report_iommu_fault(struct iommu_domain *domain, struct device *dev,
 	 * if upper layers showed interest and installed a fault handler,
 	 * invoke it.
 	 */
-	if (domain->handler)
+	if (domain->cookie_type == IOMMU_COOKIE_FAULT_HANDLER &&
+	    domain->handler)
 		ret = domain->handler(domain, dev, iova, flags,
 						domain->handler_token);
 
diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
index 074daf1aac4e..e424b279a8cd 100644
--- a/drivers/iommu/ipmmu-vmsa.c
+++ b/drivers/iommu/ipmmu-vmsa.c
@@ -1081,31 +1081,24 @@ static int ipmmu_probe(struct platform_device *pdev)
 		}
 	}
 
+	platform_set_drvdata(pdev, mmu);
 	/*
 	 * Register the IPMMU to the IOMMU subsystem in the following cases:
 	 * - R-Car Gen2 IPMMU (all devices registered)
 	 * - R-Car Gen3 IPMMU (leaf devices only - skip root IPMMU-MM device)
 	 */
-	if (!mmu->features->has_cache_leaf_nodes || !ipmmu_is_root(mmu)) {
-		ret = iommu_device_sysfs_add(&mmu->iommu, &pdev->dev, NULL,
-					     dev_name(&pdev->dev));
-		if (ret)
-			return ret;
-
-		ret = iommu_device_register(&mmu->iommu, &ipmmu_ops, &pdev->dev);
-		if (ret)
-			return ret;
-	}
+	if (mmu->features->has_cache_leaf_nodes && ipmmu_is_root(mmu))
+		return 0;
 
-	/*
-	 * We can't create the ARM mapping here as it requires the bus to have
-	 * an IOMMU, which only happens when bus_set_iommu() is called in
-	 * ipmmu_init() after the probe function returns.
-	 */
+	ret = iommu_device_sysfs_add(&mmu->iommu, &pdev->dev, NULL, dev_name(&pdev->dev));
+	if (ret)
+		return ret;
 
-	platform_set_drvdata(pdev, mmu);
+	ret = iommu_device_register(&mmu->iommu, &ipmmu_ops, &pdev->dev);
+	if (ret)
+		iommu_device_sysfs_remove(&mmu->iommu);
 
-	return 0;
+	return ret;
 }
 
 static void ipmmu_remove(struct platform_device *pdev)
diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c
index 034b0e670384..df98d0c65f54 100644
--- a/drivers/iommu/mtk_iommu.c
+++ b/drivers/iommu/mtk_iommu.c
@@ -1372,15 +1372,6 @@ static int mtk_iommu_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, data);
 	mutex_init(&data->mutex);
 
-	ret = iommu_device_sysfs_add(&data->iommu, dev, NULL,
-				     "mtk-iommu.%pa", &ioaddr);
-	if (ret)
-		goto out_link_remove;
-
-	ret = iommu_device_register(&data->iommu, &mtk_iommu_ops, dev);
-	if (ret)
-		goto out_sysfs_remove;
-
 	if (MTK_IOMMU_HAS_FLAG(data->plat_data, SHARE_PGTABLE)) {
 		list_add_tail(&data->list, data->plat_data->hw_list);
 		data->hw_list = data->plat_data->hw_list;
@@ -1390,19 +1381,28 @@ static int mtk_iommu_probe(struct platform_device *pdev)
 		data->hw_list = &data->hw_list_head;
 	}
 
+	ret = iommu_device_sysfs_add(&data->iommu, dev, NULL,
+				     "mtk-iommu.%pa", &ioaddr);
+	if (ret)
+		goto out_list_del;
+
+	ret = iommu_device_register(&data->iommu, &mtk_iommu_ops, dev);
+	if (ret)
+		goto out_sysfs_remove;
+
 	if (MTK_IOMMU_IS_TYPE(data->plat_data, MTK_IOMMU_TYPE_MM)) {
 		ret = component_master_add_with_match(dev, &mtk_iommu_com_ops, match);
 		if (ret)
-			goto out_list_del;
+			goto out_device_unregister;
 	}
 	return ret;
 
-out_list_del:
-	list_del(&data->list);
+out_device_unregister:
 	iommu_device_unregister(&data->iommu);
 out_sysfs_remove:
 	iommu_device_sysfs_remove(&data->iommu);
-out_link_remove:
+out_list_del:
+	list_del(&data->list);
 	if (MTK_IOMMU_IS_TYPE(data->plat_data, MTK_IOMMU_TYPE_MM))
 		device_link_remove(data->smicomm_dev, dev);
 out_runtime_disable:
diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
index cec05e443083..08bb3b031f23 100644
--- a/drivers/irqchip/Kconfig
+++ b/drivers/irqchip/Kconfig
@@ -114,8 +114,8 @@ config I8259
 
 config BCM2712_MIP
 	tristate "Broadcom BCM2712 MSI-X Interrupt Peripheral support"
-	depends on ARCH_BRCMSTB || COMPILE_TEST
-	default m if ARCH_BRCMSTB
+	depends on ARCH_BRCMSTB || ARCH_BCM2835 || COMPILE_TEST
+	default m if ARCH_BRCMSTB || ARCH_BCM2835
 	depends on ARM_GIC
 	select GENERIC_IRQ_CHIP
 	select IRQ_DOMAIN_HIERARCHY
diff --git a/drivers/irqchip/irq-bcm2712-mip.c b/drivers/irqchip/irq-bcm2712-mip.c
index 49a19db2d1e1..4cce24233f0f 100644
--- a/drivers/irqchip/irq-bcm2712-mip.c
+++ b/drivers/irqchip/irq-bcm2712-mip.c
@@ -163,6 +163,7 @@ static const struct irq_domain_ops mip_middle_domain_ops = {
 static const struct msi_parent_ops mip_msi_parent_ops = {
 	.supported_flags	= MIP_MSI_FLAGS_SUPPORTED,
 	.required_flags		= MIP_MSI_FLAGS_REQUIRED,
+	.chip_flags		= MSI_CHIP_FLAG_SET_EOI | MSI_CHIP_FLAG_SET_ACK,
 	.bus_select_token       = DOMAIN_BUS_GENERIC_MSI,
 	.bus_select_mask	= MATCH_PCI_MSI,
 	.prefix			= "MIP-MSI-",
diff --git a/drivers/irqchip/irq-gic-v2m.c b/drivers/irqchip/irq-gic-v2m.c
index c69894861866..dc98c39d2b20 100644
--- a/drivers/irqchip/irq-gic-v2m.c
+++ b/drivers/irqchip/irq-gic-v2m.c
@@ -421,7 +421,7 @@ static int __init gicv2m_of_init(struct fwnode_handle *parent_handle,
 #ifdef CONFIG_ACPI
 static int acpi_num_msi;
 
-static __init struct fwnode_handle *gicv2m_get_fwnode(struct device *dev)
+static struct fwnode_handle *gicv2m_get_fwnode(struct device *dev)
 {
 	struct v2m_data *data;
 
diff --git a/drivers/irqchip/irq-qcom-mpm.c b/drivers/irqchip/irq-qcom-mpm.c
index 7942d8eb3d00..f772deb9cba5 100644
--- a/drivers/irqchip/irq-qcom-mpm.c
+++ b/drivers/irqchip/irq-qcom-mpm.c
@@ -227,6 +227,9 @@ static int qcom_mpm_alloc(struct irq_domain *domain, unsigned int virq,
 	if (ret)
 		return ret;
 
+	if (pin == GPIO_NO_WAKE_IRQ)
+		return irq_domain_disconnect_hierarchy(domain, virq);
+
 	ret = irq_domain_set_hwirq_and_chip(domain, virq, pin,
 					    &qcom_mpm_chip, priv);
 	if (ret)
diff --git a/drivers/irqchip/irq-renesas-rzv2h.c b/drivers/irqchip/irq-renesas-rzv2h.c
index 3d5b5fdf9bde..0f0fd7d4dfdf 100644
--- a/drivers/irqchip/irq-renesas-rzv2h.c
+++ b/drivers/irqchip/irq-renesas-rzv2h.c
@@ -170,6 +170,14 @@ static void rzv2h_tint_irq_endisable(struct irq_data *d, bool enable)
 	else
 		tssr &= ~ICU_TSSR_TIEN(tssel_n, priv->info->field_width);
 	writel_relaxed(tssr, priv->base + priv->info->t_offs + ICU_TSSR(k));
+
+	/*
+	 * A glitch in the edge detection circuit can cause a spurious
+	 * interrupt. Clear the status flag after setting the ICU_TSSRk
+	 * registers, which is recommended by the hardware manual as a
+	 * countermeasure.
+	 */
+	writel_relaxed(BIT(tint_nr), priv->base + priv->info->t_offs + ICU_TSCLR);
 }
 
 static void rzv2h_icu_irq_disable(struct irq_data *d)
diff --git a/drivers/irqchip/irq-sg2042-msi.c b/drivers/irqchip/irq-sg2042-msi.c
index ee682e87eb8b..375b55aa0acd 100644
--- a/drivers/irqchip/irq-sg2042-msi.c
+++ b/drivers/irqchip/irq-sg2042-msi.c
@@ -151,6 +151,7 @@ static const struct irq_domain_ops sg2042_msi_middle_domain_ops = {
 static const struct msi_parent_ops sg2042_msi_parent_ops = {
 	.required_flags		= SG2042_MSI_FLAGS_REQUIRED,
 	.supported_flags	= SG2042_MSI_FLAGS_SUPPORTED,
+	.chip_flags		= MSI_CHIP_FLAG_SET_ACK,
 	.bus_select_mask	= MATCH_PCI_MSI,
 	.bus_select_token	= DOMAIN_BUS_NEXUS,
 	.prefix			= "SG2042-",
diff --git a/drivers/mcb/mcb-parse.c b/drivers/mcb/mcb-parse.c
index 02a680c73979..bf0d7d58c8b0 100644
--- a/drivers/mcb/mcb-parse.c
+++ b/drivers/mcb/mcb-parse.c
@@ -96,7 +96,7 @@ static int chameleon_parse_gdd(struct mcb_bus *bus,
 
 	ret = mcb_device_register(bus, mdev);
 	if (ret < 0)
-		goto err;
+		return ret;
 
 	return 0;
 
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 06f809e70f15..ddb37f6670de 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -139,7 +139,7 @@ config MD_RAID456
 	tristate "RAID-4/RAID-5/RAID-6 mode"
 	depends on BLK_DEV_MD
 	select RAID6_PQ
-	select LIBCRC32C
+	select CRC32
 	select ASYNC_MEMCPY
 	select ASYNC_XOR
 	select ASYNC_PQ
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index e42f1400cea9..813b38aec3e4 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -546,7 +546,7 @@ static struct uuid_entry *uuid_find(struct cache_set *c, const char *uuid)
 
 static struct uuid_entry *uuid_find_empty(struct cache_set *c)
 {
-	static const char zero_uuid[16] = "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0";
+	static const char zero_uuid[16] = { 0 };
 
 	return uuid_find(c, zero_uuid);
 }
diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c
index 9c8ed65cd87e..f0b5a6931161 100644
--- a/drivers/md/dm-bufio.c
+++ b/drivers/md/dm-bufio.c
@@ -68,6 +68,8 @@
 #define LIST_DIRTY	1
 #define LIST_SIZE	2
 
+#define SCAN_RESCHED_CYCLE	16
+
 /*--------------------------------------------------------------*/
 
 /*
@@ -2424,7 +2426,12 @@ static void __scan(struct dm_bufio_client *c)
 
 			atomic_long_dec(&c->need_shrink);
 			freed++;
-			cond_resched();
+
+			if (unlikely(freed % SCAN_RESCHED_CYCLE == 0)) {
+				dm_bufio_unlock(c);
+				cond_resched();
+				dm_bufio_lock(c);
+			}
 		}
 	}
 }
diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c
index 2a283feb3319..cc3d3897ef42 100644
--- a/drivers/md/dm-integrity.c
+++ b/drivers/md/dm-integrity.c
@@ -5164,7 +5164,7 @@ static void dm_integrity_dtr(struct dm_target *ti)
 	BUG_ON(!RB_EMPTY_ROOT(&ic->in_progress));
 	BUG_ON(!list_empty(&ic->wait_list));
 
-	if (ic->mode == 'B')
+	if (ic->mode == 'B' && ic->bitmap_flush_work.work.func)
 		cancel_delayed_work_sync(&ic->bitmap_flush_work);
 	if (ic->metadata_wq)
 		destroy_workqueue(ic->metadata_wq);
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 35100a435c88..9e175c5e0634 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -523,9 +523,10 @@ static char **realloc_argv(unsigned int *size, char **old_argv)
 		gfp = GFP_NOIO;
 	}
 	argv = kmalloc_array(new_size, sizeof(*argv), gfp);
-	if (argv && old_argv) {
-		memcpy(argv, old_argv, *size * sizeof(*argv));
+	if (argv) {
 		*size = new_size;
+		if (old_argv)
+			memcpy(argv, old_argv, *size * sizeof(*argv));
 	}
 
 	kfree(old_argv);
@@ -1049,7 +1050,6 @@ static int dm_table_alloc_md_mempools(struct dm_table *t, struct mapped_device *
 	unsigned int min_pool_size = 0, pool_size;
 	struct dm_md_mempools *pools;
 	unsigned int bioset_flags = 0;
-	bool mempool_needs_integrity = t->integrity_supported;
 
 	if (unlikely(type == DM_TYPE_NONE)) {
 		DMERR("no table type is set, can't allocate mempools");
@@ -1074,8 +1074,6 @@ static int dm_table_alloc_md_mempools(struct dm_table *t, struct mapped_device *
 
 		per_io_data_size = max(per_io_data_size, ti->per_io_data_size);
 		min_pool_size = max(min_pool_size, ti->num_flush_bios);
-
-		mempool_needs_integrity |= ti->mempool_needs_integrity;
 	}
 	pool_size = max(dm_get_reserved_bio_based_ios(), min_pool_size);
 	front_pad = roundup(per_io_data_size,
diff --git a/drivers/md/md-bitmap.c b/drivers/md/md-bitmap.c
index 44ec9b17cfd3..37b08f26c62f 100644
--- a/drivers/md/md-bitmap.c
+++ b/drivers/md/md-bitmap.c
@@ -2357,9 +2357,8 @@ static int bitmap_get_stats(void *data, struct md_bitmap_stats *stats)
 
 	if (!bitmap)
 		return -ENOENT;
-	if (bitmap->mddev->bitmap_info.external)
-		return -ENOENT;
-	if (!bitmap->storage.sb_page) /* no superblock */
+	if (!bitmap->mddev->bitmap_info.external &&
+	    !bitmap->storage.sb_page)
 		return -EINVAL;
 	sb = kmap_local_page(bitmap->storage.sb_page);
 	stats->sync_size = le64_to_cpu(sb->sync_size);
diff --git a/drivers/md/persistent-data/Kconfig b/drivers/md/persistent-data/Kconfig
index f4f948b0e173..dbb97a7233ab 100644
--- a/drivers/md/persistent-data/Kconfig
+++ b/drivers/md/persistent-data/Kconfig
@@ -2,7 +2,7 @@
 config DM_PERSISTENT_DATA
        tristate
        depends on BLK_DEV_DM
-       select LIBCRC32C
+       select CRC32
        select DM_BUFIO
 	help
 	 Library providing immutable on-disk data structure support for
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 0efc03cea24e..de9bccbe7337 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -2200,14 +2200,9 @@ static int fix_sync_read_error(struct r1bio *r1_bio)
 				if (!rdev_set_badblocks(rdev, sect, s, 0))
 					abort = 1;
 			}
-			if (abort) {
-				conf->recovery_disabled =
-					mddev->recovery_disabled;
-				set_bit(MD_RECOVERY_INTR, &mddev->recovery);
-				md_done_sync(mddev, r1_bio->sectors, 0);
-				put_buf(r1_bio);
+			if (abort)
 				return 0;
-			}
+
 			/* Try next page */
 			sectors -= s;
 			sect += s;
@@ -2346,10 +2341,21 @@ static void sync_request_write(struct mddev *mddev, struct r1bio *r1_bio)
 	int disks = conf->raid_disks * 2;
 	struct bio *wbio;
 
-	if (!test_bit(R1BIO_Uptodate, &r1_bio->state))
-		/* ouch - failed to read all of that. */
-		if (!fix_sync_read_error(r1_bio))
+	if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
+		/*
+		 * ouch - failed to read all of that.
+		 * No need to fix read error for check/repair
+		 * because all member disks are read.
+		 */
+		if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) ||
+		    !fix_sync_read_error(r1_bio)) {
+			conf->recovery_disabled = mddev->recovery_disabled;
+			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+			md_done_sync(mddev, r1_bio->sectors, 0);
+			put_buf(r1_bio);
 			return;
+		}
+	}
 
 	if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
 		process_checks(r1_bio);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 846c5f29486e..ba32bac975b8 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1735,6 +1735,7 @@ retry_discard:
 	 * The discard bio returns only first r10bio finishes
 	 */
 	if (first_copy) {
+		md_account_bio(mddev, &bio);
 		r10_bio->master_bio = bio;
 		set_bit(R10BIO_Discard, &r10_bio->state);
 		first_copy = false;
diff --git a/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c b/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c
index 04756302b878..98d3d123004c 100644
--- a/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c
+++ b/drivers/misc/mchp_pci1xxxx/mchp_pci1xxxx_gpio.c
@@ -37,6 +37,7 @@
 struct pci1xxxx_gpio {
 	struct auxiliary_device *aux_dev;
 	void __iomem *reg_base;
+	raw_spinlock_t wa_lock;
 	struct gpio_chip gpio;
 	spinlock_t lock;
 	int irq_base;
@@ -167,7 +168,7 @@ static void pci1xxxx_gpio_irq_ack(struct irq_data *data)
 	unsigned long flags;
 
 	spin_lock_irqsave(&priv->lock, flags);
-	pci1xxx_assign_bit(priv->reg_base, INTR_STAT_OFFSET(gpio), (gpio % 32), true);
+	writel(BIT(gpio % 32), priv->reg_base + INTR_STAT_OFFSET(gpio));
 	spin_unlock_irqrestore(&priv->lock, flags);
 }
 
@@ -257,6 +258,7 @@ static irqreturn_t pci1xxxx_gpio_irq_handler(int irq, void *dev_id)
 	struct pci1xxxx_gpio *priv = dev_id;
 	struct gpio_chip *gc =  &priv->gpio;
 	unsigned long int_status = 0;
+	unsigned long wa_flags;
 	unsigned long flags;
 	u8 pincount;
 	int bit;
@@ -280,7 +282,9 @@ static irqreturn_t pci1xxxx_gpio_irq_handler(int irq, void *dev_id)
 			writel(BIT(bit), priv->reg_base + INTR_STATUS_OFFSET(gpiobank));
 			spin_unlock_irqrestore(&priv->lock, flags);
 			irq = irq_find_mapping(gc->irq.domain, (bit + (gpiobank * 32)));
-			handle_nested_irq(irq);
+			raw_spin_lock_irqsave(&priv->wa_lock, wa_flags);
+			generic_handle_irq(irq);
+			raw_spin_unlock_irqrestore(&priv->wa_lock, wa_flags);
 		}
 	}
 	spin_lock_irqsave(&priv->lock, flags);
diff --git a/drivers/misc/mei/hw-me-regs.h b/drivers/misc/mei/hw-me-regs.h
index a5f88ec97df7..bc40b940ae21 100644
--- a/drivers/misc/mei/hw-me-regs.h
+++ b/drivers/misc/mei/hw-me-regs.h
@@ -117,6 +117,7 @@
 
 #define MEI_DEV_ID_LNL_M      0xA870  /* Lunar Lake Point M */
 
+#define MEI_DEV_ID_PTL_H      0xE370  /* Panther Lake H */
 #define MEI_DEV_ID_PTL_P      0xE470  /* Panther Lake P */
 
 /*
diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c
index d6ff9d82ae94..3f9c60b579ae 100644
--- a/drivers/misc/mei/pci-me.c
+++ b/drivers/misc/mei/pci-me.c
@@ -124,6 +124,7 @@ static const struct pci_device_id mei_me_pci_tbl[] = {
 
 	{MEI_PCI_DEVICE(MEI_DEV_ID_LNL_M, MEI_ME_PCH15_CFG)},
 
+	{MEI_PCI_DEVICE(MEI_DEV_ID_PTL_H, MEI_ME_PCH15_CFG)},
 	{MEI_PCI_DEVICE(MEI_DEV_ID_PTL_P, MEI_ME_PCH15_CFG)},
 
 	/* required last entry */
diff --git a/drivers/misc/mei/vsc-tp.c b/drivers/misc/mei/vsc-tp.c
index 7be1649b1972..da26a080916c 100644
--- a/drivers/misc/mei/vsc-tp.c
+++ b/drivers/misc/mei/vsc-tp.c
@@ -36,20 +36,24 @@
 #define VSC_TP_XFER_TIMEOUT_BYTES		700
 #define VSC_TP_PACKET_PADDING_SIZE		1
 #define VSC_TP_PACKET_SIZE(pkt) \
-	(sizeof(struct vsc_tp_packet) + le16_to_cpu((pkt)->len) + VSC_TP_CRC_SIZE)
+	(sizeof(struct vsc_tp_packet_hdr) + le16_to_cpu((pkt)->hdr.len) + VSC_TP_CRC_SIZE)
 #define VSC_TP_MAX_PACKET_SIZE \
-	(sizeof(struct vsc_tp_packet) + VSC_TP_MAX_MSG_SIZE + VSC_TP_CRC_SIZE)
+	(sizeof(struct vsc_tp_packet_hdr) + VSC_TP_MAX_MSG_SIZE + VSC_TP_CRC_SIZE)
 #define VSC_TP_MAX_XFER_SIZE \
 	(VSC_TP_MAX_PACKET_SIZE + VSC_TP_XFER_TIMEOUT_BYTES)
 #define VSC_TP_NEXT_XFER_LEN(len, offset) \
-	(len + sizeof(struct vsc_tp_packet) + VSC_TP_CRC_SIZE - offset + VSC_TP_PACKET_PADDING_SIZE)
+	(len + sizeof(struct vsc_tp_packet_hdr) + VSC_TP_CRC_SIZE - offset + VSC_TP_PACKET_PADDING_SIZE)
 
-struct vsc_tp_packet {
+struct vsc_tp_packet_hdr {
 	__u8 sync;
 	__u8 cmd;
 	__le16 len;
 	__le32 seq;
-	__u8 buf[] __counted_by(len);
+};
+
+struct vsc_tp_packet {
+	struct vsc_tp_packet_hdr hdr;
+	__u8 buf[VSC_TP_MAX_XFER_SIZE - sizeof(struct vsc_tp_packet_hdr)];
 };
 
 struct vsc_tp {
@@ -67,8 +71,8 @@ struct vsc_tp {
 	u32 seq;
 
 	/* command buffer */
-	void *tx_buf;
-	void *rx_buf;
+	struct vsc_tp_packet *tx_buf;
+	struct vsc_tp_packet *rx_buf;
 
 	atomic_t assert_cnt;
 	wait_queue_head_t xfer_wait;
@@ -158,12 +162,12 @@ static int vsc_tp_dev_xfer(struct vsc_tp *tp, void *obuf, void *ibuf, size_t len
 static int vsc_tp_xfer_helper(struct vsc_tp *tp, struct vsc_tp_packet *pkt,
 			      void *ibuf, u16 ilen)
 {
-	int ret, offset = 0, cpy_len, src_len, dst_len = sizeof(struct vsc_tp_packet);
+	int ret, offset = 0, cpy_len, src_len, dst_len = sizeof(struct vsc_tp_packet_hdr);
 	int next_xfer_len = VSC_TP_PACKET_SIZE(pkt) + VSC_TP_XFER_TIMEOUT_BYTES;
-	u8 *src, *crc_src, *rx_buf = tp->rx_buf;
+	u8 *src, *crc_src, *rx_buf = (u8 *)tp->rx_buf;
 	int count_down = VSC_TP_MAX_XFER_COUNT;
 	u32 recv_crc = 0, crc = ~0;
-	struct vsc_tp_packet ack;
+	struct vsc_tp_packet_hdr ack;
 	u8 *dst = (u8 *)&ack;
 	bool synced = false;
 
@@ -280,10 +284,10 @@ int vsc_tp_xfer(struct vsc_tp *tp, u8 cmd, const void *obuf, size_t olen,
 
 	guard(mutex)(&tp->mutex);
 
-	pkt->sync = VSC_TP_PACKET_SYNC;
-	pkt->cmd = cmd;
-	pkt->len = cpu_to_le16(olen);
-	pkt->seq = cpu_to_le32(++tp->seq);
+	pkt->hdr.sync = VSC_TP_PACKET_SYNC;
+	pkt->hdr.cmd = cmd;
+	pkt->hdr.len = cpu_to_le16(olen);
+	pkt->hdr.seq = cpu_to_le32(++tp->seq);
 	memcpy(pkt->buf, obuf, olen);
 
 	crc = ~crc32(~0, (u8 *)pkt, sizeof(pkt) + olen);
@@ -320,7 +324,7 @@ int vsc_tp_rom_xfer(struct vsc_tp *tp, const void *obuf, void *ibuf, size_t len)
 	guard(mutex)(&tp->mutex);
 
 	/* rom xfer is big endian */
-	cpu_to_be32_array(tp->tx_buf, obuf, words);
+	cpu_to_be32_array((u32 *)tp->tx_buf, obuf, words);
 
 	ret = read_poll_timeout(gpiod_get_value_cansleep, ret,
 				!ret, VSC_TP_ROM_XFER_POLL_DELAY_US,
@@ -336,7 +340,7 @@ int vsc_tp_rom_xfer(struct vsc_tp *tp, const void *obuf, void *ibuf, size_t len)
 		return ret;
 
 	if (ibuf)
-		be32_to_cpu_array(ibuf, tp->rx_buf, words);
+		be32_to_cpu_array(ibuf, (u32 *)tp->rx_buf, words);
 
 	return ret;
 }
@@ -490,11 +494,11 @@ static int vsc_tp_probe(struct spi_device *spi)
 	if (!tp)
 		return -ENOMEM;
 
-	tp->tx_buf = devm_kzalloc(dev, VSC_TP_MAX_XFER_SIZE, GFP_KERNEL);
+	tp->tx_buf = devm_kzalloc(dev, sizeof(*tp->tx_buf), GFP_KERNEL);
 	if (!tp->tx_buf)
 		return -ENOMEM;
 
-	tp->rx_buf = devm_kzalloc(dev, VSC_TP_MAX_XFER_SIZE, GFP_KERNEL);
+	tp->rx_buf = devm_kzalloc(dev, sizeof(*tp->rx_buf), GFP_KERNEL);
 	if (!tp->rx_buf)
 		return -ENOMEM;
 
diff --git a/drivers/misc/pci_endpoint_test.c b/drivers/misc/pci_endpoint_test.c
index d294850a35a1..c4e5e2c977be 100644
--- a/drivers/misc/pci_endpoint_test.c
+++ b/drivers/misc/pci_endpoint_test.c
@@ -122,7 +122,6 @@ struct pci_endpoint_test {
 struct pci_endpoint_test_data {
 	enum pci_barno test_reg_bar;
 	size_t alignment;
-	int irq_type;
 };
 
 static inline u32 pci_endpoint_test_readl(struct pci_endpoint_test *test,
@@ -948,7 +947,6 @@ static int pci_endpoint_test_probe(struct pci_dev *pdev,
 		test_reg_bar = data->test_reg_bar;
 		test->test_reg_bar = test_reg_bar;
 		test->alignment = data->alignment;
-		test->irq_type = data->irq_type;
 	}
 
 	init_completion(&test->irq_raised);
@@ -970,10 +968,6 @@ static int pci_endpoint_test_probe(struct pci_dev *pdev,
 
 	pci_set_master(pdev);
 
-	ret = pci_endpoint_test_alloc_irq_vectors(test, test->irq_type);
-	if (ret)
-		goto err_disable_irq;
-
 	for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) {
 		if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
 			base = pci_ioremap_bar(pdev, bar);
@@ -1009,10 +1003,6 @@ static int pci_endpoint_test_probe(struct pci_dev *pdev,
 		goto err_ida_remove;
 	}
 
-	ret = pci_endpoint_test_request_irq(test);
-	if (ret)
-		goto err_kfree_test_name;
-
 	pci_endpoint_test_get_capabilities(test);
 
 	misc_device = &test->miscdev;
@@ -1020,7 +1010,7 @@ static int pci_endpoint_test_probe(struct pci_dev *pdev,
 	misc_device->name = kstrdup(name, GFP_KERNEL);
 	if (!misc_device->name) {
 		ret = -ENOMEM;
-		goto err_release_irq;
+		goto err_kfree_test_name;
 	}
 	misc_device->parent = &pdev->dev;
 	misc_device->fops = &pci_endpoint_test_fops;
@@ -1036,9 +1026,6 @@ static int pci_endpoint_test_probe(struct pci_dev *pdev,
 err_kfree_name:
 	kfree(misc_device->name);
 
-err_release_irq:
-	pci_endpoint_test_release_irq(test);
-
 err_kfree_test_name:
 	kfree(test->name);
 
@@ -1051,8 +1038,6 @@ err_iounmap:
 			pci_iounmap(pdev, test->bar[bar]);
 	}
 
-err_disable_irq:
-	pci_endpoint_test_free_irq_vectors(test);
 	pci_release_regions(pdev);
 
 err_disable_pdev:
@@ -1092,23 +1077,19 @@ static void pci_endpoint_test_remove(struct pci_dev *pdev)
 static const struct pci_endpoint_test_data default_data = {
 	.test_reg_bar = BAR_0,
 	.alignment = SZ_4K,
-	.irq_type = PCITEST_IRQ_TYPE_MSI,
 };
 
 static const struct pci_endpoint_test_data am654_data = {
 	.test_reg_bar = BAR_2,
 	.alignment = SZ_64K,
-	.irq_type = PCITEST_IRQ_TYPE_MSI,
 };
 
 static const struct pci_endpoint_test_data j721e_data = {
 	.alignment = 256,
-	.irq_type = PCITEST_IRQ_TYPE_MSI,
 };
 
 static const struct pci_endpoint_test_data rk3588_data = {
 	.alignment = SZ_64K,
-	.irq_type = PCITEST_IRQ_TYPE_MSI,
 };
 
 /*
diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 6824131b69b1..264e11fa58ea 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -691,8 +691,8 @@ config MMC_TMIO_CORE
 config MMC_SDHI
 	tristate "Renesas SDHI SD/SDIO controller support"
 	depends on SUPERH || ARCH_RENESAS || COMPILE_TEST
+	depends on (RESET_CONTROLLER && REGULATOR) || !OF
 	select MMC_TMIO_CORE
-	select RESET_CONTROLLER if ARCH_RENESAS
 	help
 	  This provides support for the SDHI SD/SDIO controller found in
 	  Renesas SuperH, ARM and ARM64 based SoCs
diff --git a/drivers/mmc/host/renesas_sdhi_core.c b/drivers/mmc/host/renesas_sdhi_core.c
index fa6526be3638..8c83e203c516 100644
--- a/drivers/mmc/host/renesas_sdhi_core.c
+++ b/drivers/mmc/host/renesas_sdhi_core.c
@@ -1179,7 +1179,7 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 		if (IS_ERR(rdev)) {
 			dev_err(dev, "regulator register failed err=%ld", PTR_ERR(rdev));
 			ret = PTR_ERR(rdev);
-			goto efree;
+			goto edisclk;
 		}
 		priv->rdev = rdev;
 	}
@@ -1243,26 +1243,26 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 	num_irqs = platform_irq_count(pdev);
 	if (num_irqs < 0) {
 		ret = num_irqs;
-		goto eirq;
+		goto edisclk;
 	}
 
 	/* There must be at least one IRQ source */
 	if (!num_irqs) {
 		ret = -ENXIO;
-		goto eirq;
+		goto edisclk;
 	}
 
 	for (i = 0; i < num_irqs; i++) {
 		irq = platform_get_irq(pdev, i);
 		if (irq < 0) {
 			ret = irq;
-			goto eirq;
+			goto edisclk;
 		}
 
 		ret = devm_request_irq(&pdev->dev, irq, tmio_mmc_irq, 0,
 				       dev_name(&pdev->dev), host);
 		if (ret)
-			goto eirq;
+			goto edisclk;
 	}
 
 	ret = tmio_mmc_host_probe(host);
@@ -1274,8 +1274,6 @@ int renesas_sdhi_probe(struct platform_device *pdev,
 
 	return ret;
 
-eirq:
-	tmio_mmc_host_remove(host);
 edisclk:
 	renesas_sdhi_clk_disable(host);
 efree:
diff --git a/drivers/mtd/inftlcore.c b/drivers/mtd/inftlcore.c
index 9739387cff8c..58c6e1743f5c 100644
--- a/drivers/mtd/inftlcore.c
+++ b/drivers/mtd/inftlcore.c
@@ -482,10 +482,11 @@ static inline u16 INFTL_findwriteunit(struct INFTLrecord *inftl, unsigned block)
 		silly = MAX_LOOPS;
 
 		while (thisEUN <= inftl->lastEUN) {
-			inftl_read_oob(mtd, (thisEUN * inftl->EraseSize) +
-				       blockofs, 8, &retlen, (char *)&bci);
-
-			status = bci.Status | bci.Status1;
+			if (inftl_read_oob(mtd, (thisEUN * inftl->EraseSize) +
+				       blockofs, 8, &retlen, (char *)&bci) < 0)
+				status = SECTOR_IGNORE;
+			else
+				status = bci.Status | bci.Status1;
 			pr_debug("INFTL: status of block %d in EUN %d is %x\n",
 					block , writeEUN, status);
 
diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile
index db516a45f0c5..44913ff1bf12 100644
--- a/drivers/mtd/nand/Makefile
+++ b/drivers/mtd/nand/Makefile
@@ -3,11 +3,8 @@
 nandcore-objs := core.o bbt.o
 obj-$(CONFIG_MTD_NAND_CORE) += nandcore.o
 obj-$(CONFIG_MTD_NAND_ECC_MEDIATEK) += ecc-mtk.o
-ifeq ($(CONFIG_SPI_QPIC_SNAND),y)
 obj-$(CONFIG_SPI_QPIC_SNAND) += qpic_common.o
-else
 obj-$(CONFIG_MTD_NAND_QCOM) += qpic_common.o
-endif
 obj-y	+= onenand/
 obj-y	+= raw/
 obj-y	+= spi/
diff --git a/drivers/mtd/nand/raw/r852.c b/drivers/mtd/nand/raw/r852.c
index b07c2f8b4035..918974d088cf 100644
--- a/drivers/mtd/nand/raw/r852.c
+++ b/drivers/mtd/nand/raw/r852.c
@@ -387,6 +387,9 @@ static int r852_wait(struct nand_chip *chip)
 static int r852_ready(struct nand_chip *chip)
 {
 	struct r852_device *dev = r852_get_dev(nand_to_mtd(chip));
+	if (dev->card_unstable)
+		return 0;
+
 	return !(r852_read_reg(dev, R852_CARD_STA) & R852_CARD_STA_BUSY);
 }
 
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 950d8e4d86f8..8ea183da8d53 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -850,8 +850,9 @@ static int bond_check_dev_link(struct bonding *bond,
 			       struct net_device *slave_dev, int reporting)
 {
 	const struct net_device_ops *slave_ops = slave_dev->netdev_ops;
-	struct ifreq ifr;
 	struct mii_ioctl_data *mii;
+	struct ifreq ifr;
+	int ret;
 
 	if (!reporting && !netif_running(slave_dev))
 		return 0;
@@ -860,9 +861,13 @@ static int bond_check_dev_link(struct bonding *bond,
 		return netif_carrier_ok(slave_dev) ? BMSR_LSTATUS : 0;
 
 	/* Try to get link status using Ethtool first. */
-	if (slave_dev->ethtool_ops->get_link)
-		return slave_dev->ethtool_ops->get_link(slave_dev) ?
-			BMSR_LSTATUS : 0;
+	if (slave_dev->ethtool_ops->get_link) {
+		netdev_lock_ops(slave_dev);
+		ret = slave_dev->ethtool_ops->get_link(slave_dev);
+		netdev_unlock_ops(slave_dev);
+
+		return ret ? BMSR_LSTATUS : 0;
+	}
 
 	/* Ethtool can't be used, fallback to MII ioctls. */
 	if (slave_ops->ndo_eth_ioctl) {
diff --git a/drivers/net/can/rockchip/rockchip_canfd-core.c b/drivers/net/can/rockchip/rockchip_canfd-core.c
index 46201c126703..7107a37da36c 100644
--- a/drivers/net/can/rockchip/rockchip_canfd-core.c
+++ b/drivers/net/can/rockchip/rockchip_canfd-core.c
@@ -902,15 +902,16 @@ static int rkcanfd_probe(struct platform_device *pdev)
 	priv->can.data_bittiming_const = &rkcanfd_data_bittiming_const;
 	priv->can.ctrlmode_supported = CAN_CTRLMODE_LOOPBACK |
 		CAN_CTRLMODE_BERR_REPORTING;
-	if (!(priv->devtype_data.quirks & RKCANFD_QUIRK_CANFD_BROKEN))
-		priv->can.ctrlmode_supported |= CAN_CTRLMODE_FD;
 	priv->can.do_set_mode = rkcanfd_set_mode;
 	priv->can.do_get_berr_counter = rkcanfd_get_berr_counter;
 	priv->ndev = ndev;
 
 	match = device_get_match_data(&pdev->dev);
-	if (match)
+	if (match) {
 		priv->devtype_data = *(struct rkcanfd_devtype_data *)match;
+		if (!(priv->devtype_data.quirks & RKCANFD_QUIRK_CANFD_BROKEN))
+			priv->can.ctrlmode_supported |= CAN_CTRLMODE_FD;
+	}
 
 	err = can_rx_offload_add_manual(ndev, &priv->offload,
 					RKCANFD_NAPI_WEIGHT);
diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 61d164ffb3ae..e5ba71897906 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -737,6 +737,15 @@ static void b53_enable_mib(struct b53_device *dev)
 	b53_write8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc);
 }
 
+static void b53_enable_stp(struct b53_device *dev)
+{
+	u8 gc;
+
+	b53_read8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, &gc);
+	gc |= GC_RX_BPDU_EN;
+	b53_write8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc);
+}
+
 static u16 b53_default_pvid(struct b53_device *dev)
 {
 	if (is5325(dev) || is5365(dev))
@@ -876,6 +885,7 @@ static int b53_switch_reset(struct b53_device *dev)
 	}
 
 	b53_enable_mib(dev);
+	b53_enable_stp(dev);
 
 	return b53_flush_arl(dev, FAST_AGE_STATIC);
 }
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index d70399bce5b9..c5d6628d7980 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -2419,6 +2419,9 @@ mt7531_setup_common(struct dsa_switch *ds)
 	struct mt7530_priv *priv = ds->priv;
 	int ret, i;
 
+	ds->assisted_learning_on_cpu_port = true;
+	ds->mtu_enforcement_ingress = true;
+
 	mt753x_trap_frames(priv);
 
 	/* Enable and reset MIB counters */
@@ -2571,9 +2574,6 @@ mt7531_setup(struct dsa_switch *ds)
 	if (ret)
 		return ret;
 
-	ds->assisted_learning_on_cpu_port = true;
-	ds->mtu_enforcement_ingress = true;
-
 	return 0;
 }
 
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 29a89ab4b789..08db846cda8d 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -1852,6 +1852,8 @@ static int mv88e6xxx_vtu_get(struct mv88e6xxx_chip *chip, u16 vid,
 	if (!chip->info->ops->vtu_getnext)
 		return -EOPNOTSUPP;
 
+	memset(entry, 0, sizeof(*entry));
+
 	entry->vid = vid ? vid - 1 : mv88e6xxx_max_vid(chip);
 	entry->valid = false;
 
@@ -1960,7 +1962,16 @@ static int mv88e6xxx_mst_put(struct mv88e6xxx_chip *chip, u8 sid)
 	struct mv88e6xxx_mst *mst, *tmp;
 	int err;
 
-	if (!sid)
+	/* If the SID is zero, it is for a VLAN mapped to the default MSTI,
+	 * and mv88e6xxx_stu_setup() made sure it is always present, and thus,
+	 * should not be removed here.
+	 *
+	 * If the chip lacks STU support, numerically the "sid" variable will
+	 * happen to also be zero, but we don't want to rely on that fact, so
+	 * we explicitly test that first. In that case, there is also nothing
+	 * to do here.
+	 */
+	if (!mv88e6xxx_has_stu(chip) || !sid)
 		return 0;
 
 	list_for_each_entry_safe(mst, tmp, &chip->msts, node) {
diff --git a/drivers/net/dsa/mv88e6xxx/devlink.c b/drivers/net/dsa/mv88e6xxx/devlink.c
index 795c8df7b6a7..195460a0a0d4 100644
--- a/drivers/net/dsa/mv88e6xxx/devlink.c
+++ b/drivers/net/dsa/mv88e6xxx/devlink.c
@@ -736,7 +736,8 @@ void mv88e6xxx_teardown_devlink_regions_global(struct dsa_switch *ds)
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(mv88e6xxx_regions); i++)
-		dsa_devlink_region_destroy(chip->regions[i]);
+		if (chip->regions[i])
+			dsa_devlink_region_destroy(chip->regions[i]);
 }
 
 void mv88e6xxx_teardown_devlink_regions_port(struct dsa_switch *ds, int port)
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 940f1b71226d..7b35d24c38d7 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -1543,7 +1543,7 @@ static void vsc9959_tas_clock_adjust(struct ocelot *ocelot)
 	struct tc_taprio_qopt_offload *taprio;
 	struct ocelot_port *ocelot_port;
 	struct timespec64 base_ts;
-	int port;
+	int i, port;
 	u32 val;
 
 	mutex_lock(&ocelot->fwd_domain_lock);
@@ -1575,6 +1575,9 @@ static void vsc9959_tas_clock_adjust(struct ocelot *ocelot)
 			   QSYS_PARAM_CFG_REG_3_BASE_TIME_SEC_MSB_M,
 			   QSYS_PARAM_CFG_REG_3);
 
+		for (i = 0; i < taprio->num_entries; i++)
+			vsc9959_tas_gcl_set(ocelot, i, &taprio->entries[i]);
+
 		ocelot_rmw(ocelot, QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE,
 			   QSYS_TAS_PARAM_CFG_CTRL_CONFIG_CHANGE,
 			   QSYS_TAS_PARAM_CFG_CTRL);
diff --git a/drivers/net/ethernet/amd/pds_core/adminq.c b/drivers/net/ethernet/amd/pds_core/adminq.c
index c83a0a80d533..506f682d15c1 100644
--- a/drivers/net/ethernet/amd/pds_core/adminq.c
+++ b/drivers/net/ethernet/amd/pds_core/adminq.c
@@ -5,11 +5,6 @@
 
 #include "core.h"
 
-struct pdsc_wait_context {
-	struct pdsc_qcq *qcq;
-	struct completion wait_completion;
-};
-
 static int pdsc_process_notifyq(struct pdsc_qcq *qcq)
 {
 	union pds_core_notifyq_comp *comp;
@@ -109,10 +104,10 @@ void pdsc_process_adminq(struct pdsc_qcq *qcq)
 		q_info = &q->info[q->tail_idx];
 		q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1);
 
-		/* Copy out the completion data */
-		memcpy(q_info->dest, comp, sizeof(*comp));
-
-		complete_all(&q_info->wc->wait_completion);
+		if (!completion_done(&q_info->completion)) {
+			memcpy(q_info->dest, comp, sizeof(*comp));
+			complete(&q_info->completion);
+		}
 
 		if (cq->tail_idx == cq->num_descs - 1)
 			cq->done_color = !cq->done_color;
@@ -162,8 +157,7 @@ irqreturn_t pdsc_adminq_isr(int irq, void *data)
 static int __pdsc_adminq_post(struct pdsc *pdsc,
 			      struct pdsc_qcq *qcq,
 			      union pds_core_adminq_cmd *cmd,
-			      union pds_core_adminq_comp *comp,
-			      struct pdsc_wait_context *wc)
+			      union pds_core_adminq_comp *comp)
 {
 	struct pdsc_queue *q = &qcq->q;
 	struct pdsc_q_info *q_info;
@@ -205,9 +199,9 @@ static int __pdsc_adminq_post(struct pdsc *pdsc,
 	/* Post the request */
 	index = q->head_idx;
 	q_info = &q->info[index];
-	q_info->wc = wc;
 	q_info->dest = comp;
 	memcpy(q_info->desc, cmd, sizeof(*cmd));
+	reinit_completion(&q_info->completion);
 
 	dev_dbg(pdsc->dev, "head_idx %d tail_idx %d\n",
 		q->head_idx, q->tail_idx);
@@ -231,16 +225,13 @@ int pdsc_adminq_post(struct pdsc *pdsc,
 		     union pds_core_adminq_comp *comp,
 		     bool fast_poll)
 {
-	struct pdsc_wait_context wc = {
-		.wait_completion =
-			COMPLETION_INITIALIZER_ONSTACK(wc.wait_completion),
-	};
 	unsigned long poll_interval = 1;
 	unsigned long poll_jiffies;
 	unsigned long time_limit;
 	unsigned long time_start;
 	unsigned long time_done;
 	unsigned long remaining;
+	struct completion *wc;
 	int err = 0;
 	int index;
 
@@ -250,20 +241,19 @@ int pdsc_adminq_post(struct pdsc *pdsc,
 		return -ENXIO;
 	}
 
-	wc.qcq = &pdsc->adminqcq;
-	index = __pdsc_adminq_post(pdsc, &pdsc->adminqcq, cmd, comp, &wc);
+	index = __pdsc_adminq_post(pdsc, &pdsc->adminqcq, cmd, comp);
 	if (index < 0) {
 		err = index;
 		goto err_out;
 	}
 
+	wc = &pdsc->adminqcq.q.info[index].completion;
 	time_start = jiffies;
 	time_limit = time_start + HZ * pdsc->devcmd_timeout;
 	do {
 		/* Timeslice the actual wait to catch IO errors etc early */
 		poll_jiffies = msecs_to_jiffies(poll_interval);
-		remaining = wait_for_completion_timeout(&wc.wait_completion,
-							poll_jiffies);
+		remaining = wait_for_completion_timeout(wc, poll_jiffies);
 		if (remaining)
 			break;
 
@@ -292,9 +282,11 @@ int pdsc_adminq_post(struct pdsc *pdsc,
 	dev_dbg(pdsc->dev, "%s: elapsed %d msecs\n",
 		__func__, jiffies_to_msecs(time_done - time_start));
 
-	/* Check the results */
-	if (time_after_eq(time_done, time_limit))
+	/* Check the results and clear an un-completed timeout */
+	if (time_after_eq(time_done, time_limit) && !completion_done(wc)) {
 		err = -ETIMEDOUT;
+		complete(wc);
+	}
 
 	dev_dbg(pdsc->dev, "read admin queue completion idx %d:\n", index);
 	dynamic_hex_dump("comp ", DUMP_PREFIX_OFFSET, 16, 1,
diff --git a/drivers/net/ethernet/amd/pds_core/auxbus.c b/drivers/net/ethernet/amd/pds_core/auxbus.c
index eeb72b1809ea..92f359f2b449 100644
--- a/drivers/net/ethernet/amd/pds_core/auxbus.c
+++ b/drivers/net/ethernet/amd/pds_core/auxbus.c
@@ -107,9 +107,6 @@ int pds_client_adminq_cmd(struct pds_auxiliary_dev *padev,
 	dev_dbg(pf->dev, "%s: %s opcode %d\n",
 		__func__, dev_name(&padev->aux_dev.dev), req->opcode);
 
-	if (pf->state)
-		return -ENXIO;
-
 	/* Wrap the client's request */
 	cmd.client_request.opcode = PDS_AQ_CMD_CLIENT_CMD;
 	cmd.client_request.client_id = cpu_to_le16(padev->client_id);
@@ -189,7 +186,6 @@ void pdsc_auxbus_dev_del(struct pdsc *cf, struct pdsc *pf,
 	pds_client_unregister(pf, padev->client_id);
 	auxiliary_device_delete(&padev->aux_dev);
 	auxiliary_device_uninit(&padev->aux_dev);
-	padev->client_id = 0;
 	*pd_ptr = NULL;
 
 	mutex_unlock(&pf->config_lock);
diff --git a/drivers/net/ethernet/amd/pds_core/core.c b/drivers/net/ethernet/amd/pds_core/core.c
index 1eb0d92786f7..9512aa4083f0 100644
--- a/drivers/net/ethernet/amd/pds_core/core.c
+++ b/drivers/net/ethernet/amd/pds_core/core.c
@@ -167,8 +167,10 @@ static void pdsc_q_map(struct pdsc_queue *q, void *base, dma_addr_t base_pa)
 	q->base = base;
 	q->base_pa = base_pa;
 
-	for (i = 0, cur = q->info; i < q->num_descs; i++, cur++)
+	for (i = 0, cur = q->info; i < q->num_descs; i++, cur++) {
 		cur->desc = base + (i * q->desc_size);
+		init_completion(&cur->completion);
+	}
 }
 
 static void pdsc_cq_map(struct pdsc_cq *cq, void *base, dma_addr_t base_pa)
@@ -325,10 +327,7 @@ static int pdsc_core_init(struct pdsc *pdsc)
 	size_t sz;
 	int err;
 
-	/* Scale the descriptor ring length based on number of CPUs and VFs */
-	numdescs = max_t(int, PDSC_ADMINQ_MIN_LENGTH, num_online_cpus());
-	numdescs += 2 * pci_sriov_get_totalvfs(pdsc->pdev);
-	numdescs = roundup_pow_of_two(numdescs);
+	numdescs = PDSC_ADMINQ_MAX_LENGTH;
 	err = pdsc_qcq_alloc(pdsc, PDS_CORE_QTYPE_ADMINQ, 0, "adminq",
 			     PDS_CORE_QCQ_F_CORE | PDS_CORE_QCQ_F_INTR,
 			     numdescs,
diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h
index 0bf320c43083..0b53a1fab46d 100644
--- a/drivers/net/ethernet/amd/pds_core/core.h
+++ b/drivers/net/ethernet/amd/pds_core/core.h
@@ -16,7 +16,7 @@
 
 #define PDSC_WATCHDOG_SECS	5
 #define PDSC_QUEUE_NAME_MAX_SZ  16
-#define PDSC_ADMINQ_MIN_LENGTH	16	/* must be a power of two */
+#define PDSC_ADMINQ_MAX_LENGTH	16	/* must be a power of two */
 #define PDSC_NOTIFYQ_LENGTH	64	/* must be a power of two */
 #define PDSC_TEARDOWN_RECOVERY	false
 #define PDSC_TEARDOWN_REMOVING	true
@@ -96,7 +96,7 @@ struct pdsc_q_info {
 	unsigned int bytes;
 	unsigned int nbufs;
 	struct pdsc_buf_info bufs[PDS_CORE_MAX_FRAGS];
-	struct pdsc_wait_context *wc;
+	struct completion completion;
 	void *dest;
 };
 
diff --git a/drivers/net/ethernet/amd/pds_core/debugfs.c b/drivers/net/ethernet/amd/pds_core/debugfs.c
index ac37a4e738ae..04c5e3abd8d7 100644
--- a/drivers/net/ethernet/amd/pds_core/debugfs.c
+++ b/drivers/net/ethernet/amd/pds_core/debugfs.c
@@ -154,8 +154,9 @@ void pdsc_debugfs_add_qcq(struct pdsc *pdsc, struct pdsc_qcq *qcq)
 		debugfs_create_u32("index", 0400, intr_dentry, &intr->index);
 		debugfs_create_u32("vector", 0400, intr_dentry, &intr->vector);
 
-		intr_ctrl_regset = kzalloc(sizeof(*intr_ctrl_regset),
-					   GFP_KERNEL);
+		intr_ctrl_regset = devm_kzalloc(pdsc->dev,
+						sizeof(*intr_ctrl_regset),
+						GFP_KERNEL);
 		if (!intr_ctrl_regset)
 			return;
 		intr_ctrl_regset->regs = intr_ctrl_regs;
diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c
index c5c787df61a4..d8dc39da4161 100644
--- a/drivers/net/ethernet/amd/pds_core/devlink.c
+++ b/drivers/net/ethernet/amd/pds_core/devlink.c
@@ -105,7 +105,7 @@ int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
 		.fw_control.opcode = PDS_CORE_CMD_FW_CONTROL,
 		.fw_control.oper = PDS_CORE_FW_GET_LIST,
 	};
-	struct pds_core_fw_list_info fw_list;
+	struct pds_core_fw_list_info fw_list = {};
 	struct pdsc *pdsc = devlink_priv(dl);
 	union pds_core_dev_comp comp;
 	char buf[32];
@@ -118,8 +118,6 @@ int pdsc_dl_info_get(struct devlink *dl, struct devlink_info_req *req,
 	if (!err)
 		memcpy_fromio(&fw_list, pdsc->cmd_regs->data, sizeof(fw_list));
 	mutex_unlock(&pdsc->devcmd_lock);
-	if (err && err != -EIO)
-		return err;
 
 	listlen = min(fw_list.num_fw_slots, ARRAY_SIZE(fw_list.fw_names));
 	for (i = 0; i < listlen; i++) {
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
index 230726d7b74f..d41b58fad37b 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-desc.c
@@ -373,8 +373,13 @@ static int xgbe_map_rx_buffer(struct xgbe_prv_data *pdata,
 	}
 
 	/* Set up the header page info */
-	xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
-			     XGBE_SKB_ALLOC_SIZE);
+	if (pdata->netdev->features & NETIF_F_RXCSUM) {
+		xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
+				     XGBE_SKB_ALLOC_SIZE);
+	} else {
+		xgbe_set_buffer_data(&rdata->rx.hdr, &ring->rx_hdr_pa,
+				     pdata->rx_buf_size);
+	}
 
 	/* Set up the buffer page info */
 	xgbe_set_buffer_data(&rdata->rx.buf, &ring->rx_buf_pa,
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
index f393228d41c7..f1b0fb02b3cd 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-dev.c
@@ -320,6 +320,18 @@ static void xgbe_config_sph_mode(struct xgbe_prv_data *pdata)
 	XGMAC_IOWRITE_BITS(pdata, MAC_RCR, HDSMS, XGBE_SPH_HDSMS_SIZE);
 }
 
+static void xgbe_disable_sph_mode(struct xgbe_prv_data *pdata)
+{
+	unsigned int i;
+
+	for (i = 0; i < pdata->channel_count; i++) {
+		if (!pdata->channel[i]->rx_ring)
+			break;
+
+		XGMAC_DMA_IOWRITE_BITS(pdata->channel[i], DMA_CH_CR, SPH, 0);
+	}
+}
+
 static int xgbe_write_rss_reg(struct xgbe_prv_data *pdata, unsigned int type,
 			      unsigned int index, unsigned int val)
 {
@@ -3545,8 +3557,12 @@ static int xgbe_init(struct xgbe_prv_data *pdata)
 	xgbe_config_tx_coalesce(pdata);
 	xgbe_config_rx_buffer_size(pdata);
 	xgbe_config_tso_mode(pdata);
-	xgbe_config_sph_mode(pdata);
-	xgbe_config_rss(pdata);
+
+	if (pdata->netdev->features & NETIF_F_RXCSUM) {
+		xgbe_config_sph_mode(pdata);
+		xgbe_config_rss(pdata);
+	}
+
 	desc_if->wrapper_tx_desc_init(pdata);
 	desc_if->wrapper_rx_desc_init(pdata);
 	xgbe_enable_dma_interrupts(pdata);
@@ -3702,5 +3718,9 @@ void xgbe_init_function_ptrs_dev(struct xgbe_hw_if *hw_if)
 	hw_if->disable_vxlan = xgbe_disable_vxlan;
 	hw_if->set_vxlan_id = xgbe_set_vxlan_id;
 
+	/* For Split Header*/
+	hw_if->enable_sph = xgbe_config_sph_mode;
+	hw_if->disable_sph = xgbe_disable_sph_mode;
+
 	DBGPR("<--xgbe_init_function_ptrs\n");
 }
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index d84a310dfcd4..8e09ad8fa022 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -2257,10 +2257,17 @@ static int xgbe_set_features(struct net_device *netdev,
 	if (ret)
 		return ret;
 
-	if ((features & NETIF_F_RXCSUM) && !rxcsum)
+	if ((features & NETIF_F_RXCSUM) && !rxcsum) {
+		hw_if->enable_sph(pdata);
+		hw_if->enable_vxlan(pdata);
 		hw_if->enable_rx_csum(pdata);
-	else if (!(features & NETIF_F_RXCSUM) && rxcsum)
+		schedule_work(&pdata->restart_work);
+	} else if (!(features & NETIF_F_RXCSUM) && rxcsum) {
+		hw_if->disable_sph(pdata);
+		hw_if->disable_vxlan(pdata);
 		hw_if->disable_rx_csum(pdata);
+		schedule_work(&pdata->restart_work);
+	}
 
 	if ((features & NETIF_F_HW_VLAN_CTAG_RX) && !rxvlan)
 		hw_if->enable_rx_vlan_stripping(pdata);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h
index d85386cac8d1..ed5d43c16d0e 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe.h
+++ b/drivers/net/ethernet/amd/xgbe/xgbe.h
@@ -865,6 +865,10 @@ struct xgbe_hw_if {
 	void (*enable_vxlan)(struct xgbe_prv_data *);
 	void (*disable_vxlan)(struct xgbe_prv_data *);
 	void (*set_vxlan_id)(struct xgbe_prv_data *);
+
+	/* For Split Header */
+	void (*enable_sph)(struct xgbe_prv_data *pdata);
+	void (*disable_sph)(struct xgbe_prv_data *pdata);
 };
 
 /* This structure represents implementation specific routines for an
diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig
index eeec8bf17cf4..1bd4313215d7 100644
--- a/drivers/net/ethernet/broadcom/Kconfig
+++ b/drivers/net/ethernet/broadcom/Kconfig
@@ -143,7 +143,7 @@ config BNX2X
 	depends on PTP_1588_CLOCK_OPTIONAL
 	select FW_LOADER
 	select ZLIB_INFLATE
-	select LIBCRC32C
+	select CRC32
 	select MDIO
 	help
 	  This driver supports Broadcom NetXtremeII 10 gigabit Ethernet cards.
@@ -207,7 +207,7 @@ config BNXT
 	depends on PCI
 	depends on PTP_1588_CLOCK_OPTIONAL
 	select FW_LOADER
-	select LIBCRC32C
+	select CRC32
 	select NET_DEVLINK
 	select PAGE_POOL
 	select DIMLIB
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 8725e1e13908..86a5de44b6f3 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -787,7 +787,7 @@ tx_free:
 	dev_kfree_skb_any(skb);
 tx_kick_pending:
 	if (BNXT_TX_PTP_IS_SET(lflags)) {
-		txr->tx_buf_ring[txr->tx_prod].is_ts_pkt = 0;
+		txr->tx_buf_ring[RING_TX(bp, txr->tx_prod)].is_ts_pkt = 0;
 		atomic64_inc(&bp->ptp_cfg->stats.ts_err);
 		if (!(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
 			/* set SKB to err so PTP worker will clean up */
@@ -795,7 +795,7 @@ tx_kick_pending:
 	}
 	if (txr->kick_pending)
 		bnxt_txr_db_kick(bp, txr, txr->tx_prod);
-	txr->tx_buf_ring[txr->tx_prod].skb = NULL;
+	txr->tx_buf_ring[RING_TX(bp, txr->tx_prod)].skb = NULL;
 	dev_core_stats_tx_dropped_inc(dev);
 	return NETDEV_TX_OK;
 }
@@ -2015,6 +2015,7 @@ static struct sk_buff *bnxt_rx_vlan(struct sk_buff *skb, u8 cmp_type,
 	}
 	return skb;
 vlan_err:
+	skb_mark_for_recycle(skb);
 	dev_kfree_skb(skb);
 	return NULL;
 }
@@ -3414,6 +3415,9 @@ static void bnxt_free_tx_skbs(struct bnxt *bp)
 
 		bnxt_free_one_tx_ring_skbs(bp, txr, i);
 	}
+
+	if (bp->ptp_cfg && !(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
+		bnxt_ptp_free_txts_skbs(bp->ptp_cfg);
 }
 
 static void bnxt_free_one_rx_ring(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
@@ -11599,6 +11603,9 @@ static void bnxt_init_napi(struct bnxt *bp)
 		poll_fn = bnxt_poll_p5;
 	else if (BNXT_CHIP_TYPE_NITRO_A0(bp))
 		cp_nr_rings--;
+
+	set_bit(BNXT_STATE_NAPI_DISABLED, &bp->state);
+
 	for (i = 0; i < cp_nr_rings; i++) {
 		bnapi = bp->bnapi[i];
 		netif_napi_add_config_locked(bp->dev, &bnapi->napi, poll_fn,
@@ -12318,12 +12325,15 @@ static int bnxt_hwrm_if_change(struct bnxt *bp, bool up)
 {
 	struct hwrm_func_drv_if_change_output *resp;
 	struct hwrm_func_drv_if_change_input *req;
-	bool fw_reset = !bp->irq_tbl;
 	bool resc_reinit = false;
 	bool caps_change = false;
 	int rc, retry = 0;
+	bool fw_reset;
 	u32 flags = 0;
 
+	fw_reset = (bp->fw_reset_state == BNXT_FW_RESET_STATE_ABORT);
+	bp->fw_reset_state = 0;
+
 	if (!(bp->fw_cap & BNXT_FW_CAP_IF_CHANGE))
 		return 0;
 
@@ -12392,13 +12402,8 @@ static int bnxt_hwrm_if_change(struct bnxt *bp, bool up)
 				set_bit(BNXT_STATE_ABORT_ERR, &bp->state);
 				return rc;
 			}
+			/* IRQ will be initialized later in bnxt_request_irq()*/
 			bnxt_clear_int_mode(bp);
-			rc = bnxt_init_int_mode(bp);
-			if (rc) {
-				clear_bit(BNXT_STATE_FW_RESET_DET, &bp->state);
-				netdev_err(bp->dev, "init int mode failed\n");
-				return rc;
-			}
 		}
 		rc = bnxt_cancel_reservations(bp, fw_reset);
 	}
@@ -12797,8 +12802,6 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	/* VF-reps may need to be re-opened after the PF is re-opened */
 	if (BNXT_PF(bp))
 		bnxt_vf_reps_open(bp);
-	if (bp->ptp_cfg && !(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
-		WRITE_ONCE(bp->ptp_cfg->tx_avail, BNXT_MAX_TX_TS);
 	bnxt_ptp_init_rtc(bp, true);
 	bnxt_ptp_cfg_tstamp_filters(bp);
 	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
@@ -14833,7 +14836,7 @@ static void bnxt_fw_reset_abort(struct bnxt *bp, int rc)
 	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 	if (bp->fw_reset_state != BNXT_FW_RESET_STATE_POLL_VF)
 		bnxt_dl_health_fw_status_update(bp, false);
-	bp->fw_reset_state = 0;
+	bp->fw_reset_state = BNXT_FW_RESET_STATE_ABORT;
 	netif_close(bp->dev);
 }
 
@@ -16003,8 +16006,8 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 
 	bnxt_rdma_aux_device_del(bp);
 
-	bnxt_ptp_clear(bp);
 	unregister_netdev(dev);
+	bnxt_ptp_clear(bp);
 
 	bnxt_rdma_aux_device_uninit(bp);
 
@@ -16931,10 +16934,9 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
 		if (!err)
 			result = PCI_ERS_RESULT_RECOVERED;
 
+		/* IRQ will be initialized later in bnxt_io_resume */
 		bnxt_ulp_irq_stop(bp);
 		bnxt_clear_int_mode(bp);
-		err = bnxt_init_int_mode(bp);
-		bnxt_ulp_irq_restart(bp, err);
 	}
 
 reset_exit:
@@ -16963,10 +16965,13 @@ static void bnxt_io_resume(struct pci_dev *pdev)
 
 	err = bnxt_hwrm_func_qcaps(bp);
 	if (!err) {
-		if (netif_running(netdev))
+		if (netif_running(netdev)) {
 			err = bnxt_open(netdev);
-		else
+		} else {
 			err = bnxt_reserve_rings(bp, true);
+			if (!err)
+				err = bnxt_init_int_mode(bp);
+		}
 	}
 
 	if (!err)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 21726cf56586..bc8b3b7e915d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2614,6 +2614,7 @@ struct bnxt {
 #define BNXT_FW_RESET_STATE_POLL_FW	4
 #define BNXT_FW_RESET_STATE_OPENING	5
 #define BNXT_FW_RESET_STATE_POLL_FW_DOWN	6
+#define BNXT_FW_RESET_STATE_ABORT	7
 
 	u16			fw_reset_min_dsecs;
 #define BNXT_DFLT_FW_RST_MIN_DSECS	20
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c
index 5576e7cf8463..a000d3f630bd 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c
@@ -110,20 +110,30 @@ static int bnxt_hwrm_dbg_dma_data(struct bnxt *bp, void *msg,
 			}
 		}
 
+		if (cmn_req->req_type ==
+				cpu_to_le16(HWRM_DBG_COREDUMP_RETRIEVE))
+			info->dest_buf_size += len;
+
 		if (info->dest_buf) {
 			if ((info->seg_start + off + len) <=
 			    BNXT_COREDUMP_BUF_LEN(info->buf_len)) {
-				memcpy(info->dest_buf + off, dma_buf, len);
+				u16 copylen = min_t(u16, len,
+						    info->dest_buf_size - off);
+
+				memcpy(info->dest_buf + off, dma_buf, copylen);
+				if (copylen < len)
+					break;
 			} else {
 				rc = -ENOBUFS;
+				if (cmn_req->req_type ==
+				    cpu_to_le16(HWRM_DBG_COREDUMP_LIST)) {
+					kfree(info->dest_buf);
+					info->dest_buf = NULL;
+				}
 				break;
 			}
 		}
 
-		if (cmn_req->req_type ==
-				cpu_to_le16(HWRM_DBG_COREDUMP_RETRIEVE))
-			info->dest_buf_size += len;
-
 		if (!(cmn_resp->flags & HWRM_DBG_CMN_FLAGS_MORE))
 			break;
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 48dd5922e4dd..f5d490bf997e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -2062,6 +2062,17 @@ static int bnxt_get_regs_len(struct net_device *dev)
 	return reg_len;
 }
 
+#define BNXT_PCIE_32B_ENTRY(start, end)			\
+	 { offsetof(struct pcie_ctx_hw_stats, start),	\
+	   offsetof(struct pcie_ctx_hw_stats, end) }
+
+static const struct {
+	u16 start;
+	u16 end;
+} bnxt_pcie_32b_entries[] = {
+	BNXT_PCIE_32B_ENTRY(pcie_ltssm_histogram[0], pcie_ltssm_histogram[3]),
+};
+
 static void bnxt_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 			  void *_p)
 {
@@ -2094,12 +2105,27 @@ static void bnxt_get_regs(struct net_device *dev, struct ethtool_regs *regs,
 	req->pcie_stat_host_addr = cpu_to_le64(hw_pcie_stats_addr);
 	rc = hwrm_req_send(bp, req);
 	if (!rc) {
-		__le64 *src = (__le64 *)hw_pcie_stats;
-		u64 *dst = (u64 *)(_p + BNXT_PXP_REG_LEN);
-		int i;
-
-		for (i = 0; i < sizeof(*hw_pcie_stats) / sizeof(__le64); i++)
-			dst[i] = le64_to_cpu(src[i]);
+		u8 *dst = (u8 *)(_p + BNXT_PXP_REG_LEN);
+		u8 *src = (u8 *)hw_pcie_stats;
+		int i, j;
+
+		for (i = 0, j = 0; i < sizeof(*hw_pcie_stats); ) {
+			if (i >= bnxt_pcie_32b_entries[j].start &&
+			    i <= bnxt_pcie_32b_entries[j].end) {
+				u32 *dst32 = (u32 *)(dst + i);
+
+				*dst32 = le32_to_cpu(*(__le32 *)(src + i));
+				i += 4;
+				if (i > bnxt_pcie_32b_entries[j].end &&
+				    j < ARRAY_SIZE(bnxt_pcie_32b_entries) - 1)
+					j++;
+			} else {
+				u64 *dst64 = (u64 *)(dst + i);
+
+				*dst64 = le64_to_cpu(*(__le64 *)(src + i));
+				i += 8;
+			}
+		}
 	}
 	hwrm_req_drop(bp, req);
 }
@@ -4991,6 +5017,7 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 	if (!bp->num_tests || !BNXT_PF(bp))
 		return;
 
+	memset(buf, 0, sizeof(u64) * bp->num_tests);
 	if (etest->flags & ETH_TEST_FL_OFFLINE &&
 	    bnxt_ulp_registered(bp->edev)) {
 		etest->flags |= ETH_TEST_FL_FAILED;
@@ -4998,7 +5025,6 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest,
 		return;
 	}
 
-	memset(buf, 0, sizeof(u64) * bp->num_tests);
 	if (!netif_running(dev)) {
 		etest->flags |= ETH_TEST_FL_FAILED;
 		return;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index 2d4e19b96ee7..0669d43472f5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -794,6 +794,27 @@ next_slot:
 	return HZ;
 }
 
+void bnxt_ptp_free_txts_skbs(struct bnxt_ptp_cfg *ptp)
+{
+	struct bnxt_ptp_tx_req *txts_req;
+	u16 cons = ptp->txts_cons;
+
+	/* make sure ptp aux worker finished with
+	 * possible BNXT_STATE_OPEN set
+	 */
+	ptp_cancel_worker_sync(ptp->ptp_clock);
+
+	ptp->tx_avail = BNXT_MAX_TX_TS;
+	while (cons != ptp->txts_prod) {
+		txts_req = &ptp->txts_req[cons];
+		if (!IS_ERR_OR_NULL(txts_req->tx_skb))
+			dev_kfree_skb_any(txts_req->tx_skb);
+		cons = NEXT_TXTS(cons);
+	}
+	ptp->txts_cons = cons;
+	ptp_schedule_worker(ptp->ptp_clock, 0);
+}
+
 int bnxt_ptp_get_txts_prod(struct bnxt_ptp_cfg *ptp, u16 *prod)
 {
 	spin_lock_bh(&ptp->ptp_tx_lock);
@@ -1105,7 +1126,6 @@ out:
 void bnxt_ptp_clear(struct bnxt *bp)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
-	int i;
 
 	if (!ptp)
 		return;
@@ -1117,12 +1137,5 @@ void bnxt_ptp_clear(struct bnxt *bp)
 	kfree(ptp->ptp_info.pin_config);
 	ptp->ptp_info.pin_config = NULL;
 
-	for (i = 0; i < BNXT_MAX_TX_TS; i++) {
-		if (ptp->txts_req[i].tx_skb) {
-			dev_kfree_skb_any(ptp->txts_req[i].tx_skb);
-			ptp->txts_req[i].tx_skb = NULL;
-		}
-	}
-
 	bnxt_unmap_ptp_regs(bp);
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index a95f05e9c579..0481161d26ef 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -162,6 +162,7 @@ int bnxt_ptp_cfg_tstamp_filters(struct bnxt *bp);
 void bnxt_ptp_reapply_pps(struct bnxt *bp);
 int bnxt_hwtstamp_set(struct net_device *dev, struct ifreq *ifr);
 int bnxt_hwtstamp_get(struct net_device *dev, struct ifreq *ifr);
+void bnxt_ptp_free_txts_skbs(struct bnxt_ptp_cfg *ptp);
 int bnxt_ptp_get_txts_prod(struct bnxt_ptp_cfg *ptp, u16 *prod);
 void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod);
 int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts);
diff --git a/drivers/net/ethernet/cavium/Kconfig b/drivers/net/ethernet/cavium/Kconfig
index ca742cc146d7..7dae5aad3689 100644
--- a/drivers/net/ethernet/cavium/Kconfig
+++ b/drivers/net/ethernet/cavium/Kconfig
@@ -70,8 +70,8 @@ config LIQUIDIO
 	depends on 64BIT && PCI
 	depends on PCI
 	depends on PTP_1588_CLOCK_OPTIONAL
+	select CRC32
 	select FW_LOADER
-	select LIBCRC32C
 	select LIQUIDIO_CORE
 	select NET_DEVLINK
 	help
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
index 7f3f5afa864f..1546c3db08f0 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
@@ -2270,6 +2270,7 @@ int cxgb4_init_ethtool_filters(struct adapter *adap)
 		eth_filter->port[i].bmap = bitmap_zalloc(nentries, GFP_KERNEL);
 		if (!eth_filter->port[i].bmap) {
 			ret = -ENOMEM;
+			kvfree(eth_filter->port[i].loc_array);
 			goto free_eth_finfo;
 		}
 	}
diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c
index d88fbecdab4b..232e839a9d07 100644
--- a/drivers/net/ethernet/dlink/dl2k.c
+++ b/drivers/net/ethernet/dlink/dl2k.c
@@ -352,7 +352,7 @@ parse_eeprom (struct net_device *dev)
 	eth_hw_addr_set(dev, psrom->mac_addr);
 
 	if (np->chip_id == CHIP_IP1000A) {
-		np->led_mode = psrom->led_mode;
+		np->led_mode = le16_to_cpu(psrom->led_mode);
 		return 0;
 	}
 
diff --git a/drivers/net/ethernet/dlink/dl2k.h b/drivers/net/ethernet/dlink/dl2k.h
index 195dc6cfd895..0e33e2eaae96 100644
--- a/drivers/net/ethernet/dlink/dl2k.h
+++ b/drivers/net/ethernet/dlink/dl2k.h
@@ -335,7 +335,7 @@ typedef struct t_SROM {
 	u16 sub_system_id;	/* 0x06 */
 	u16 pci_base_1;		/* 0x08 (IP1000A only) */
 	u16 pci_base_2;		/* 0x0a (IP1000A only) */
-	u16 led_mode;		/* 0x0c (IP1000A only) */
+	__le16 led_mode;	/* 0x0c (IP1000A only) */
 	u16 reserved1[9];	/* 0x0e-0x1f */
 	u8 mac_addr[6];		/* 0x20-0x25 */
 	u8 reserved2[10];	/* 0x26-0x2f */
diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c
index 2106861463e4..3ee52f4b1166 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc.c
@@ -1850,6 +1850,16 @@ static void enetc_xdp_drop(struct enetc_bdr *rx_ring, int rx_ring_first,
 	}
 }
 
+static void enetc_bulk_flip_buff(struct enetc_bdr *rx_ring, int rx_ring_first,
+				 int rx_ring_last)
+{
+	while (rx_ring_first != rx_ring_last) {
+		enetc_flip_rx_buff(rx_ring,
+				   &rx_ring->rx_swbd[rx_ring_first]);
+		enetc_bdr_idx_inc(rx_ring, &rx_ring_first);
+	}
+}
+
 static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
 				   struct napi_struct *napi, int work_limit,
 				   struct bpf_prog *prog)
@@ -1868,11 +1878,10 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
 
 	while (likely(rx_frm_cnt < work_limit)) {
 		union enetc_rx_bd *rxbd, *orig_rxbd;
-		int orig_i, orig_cleaned_cnt;
 		struct xdp_buff xdp_buff;
 		struct sk_buff *skb;
+		int orig_i, err;
 		u32 bd_status;
-		int err;
 
 		rxbd = enetc_rxbd(rx_ring, i);
 		bd_status = le32_to_cpu(rxbd->r.lstatus);
@@ -1887,7 +1896,6 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
 			break;
 
 		orig_rxbd = rxbd;
-		orig_cleaned_cnt = cleaned_cnt;
 		orig_i = i;
 
 		enetc_build_xdp_buff(rx_ring, bd_status, &rxbd, &i,
@@ -1915,15 +1923,21 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
 			rx_ring->stats.xdp_drops++;
 			break;
 		case XDP_PASS:
-			rxbd = orig_rxbd;
-			cleaned_cnt = orig_cleaned_cnt;
-			i = orig_i;
-
-			skb = enetc_build_skb(rx_ring, bd_status, &rxbd,
-					      &i, &cleaned_cnt,
-					      ENETC_RXB_DMA_SIZE_XDP);
-			if (unlikely(!skb))
+			skb = xdp_build_skb_from_buff(&xdp_buff);
+			/* Probably under memory pressure, stop NAPI */
+			if (unlikely(!skb)) {
+				enetc_xdp_drop(rx_ring, orig_i, i);
+				rx_ring->stats.xdp_drops++;
 				goto out;
+			}
+
+			enetc_get_offloads(rx_ring, orig_rxbd, skb);
+
+			/* These buffers are about to be owned by the stack.
+			 * Update our buffer cache (the rx_swbd array elements)
+			 * with their other page halves.
+			 */
+			enetc_bulk_flip_buff(rx_ring, orig_i, i);
 
 			napi_gro_receive(napi, skb);
 			break;
@@ -1965,11 +1979,7 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
 				enetc_xdp_drop(rx_ring, orig_i, i);
 				rx_ring->stats.xdp_redirect_failures++;
 			} else {
-				while (orig_i != i) {
-					enetc_flip_rx_buff(rx_ring,
-							   &rx_ring->rx_swbd[orig_i]);
-					enetc_bdr_idx_inc(rx_ring, &orig_i);
-				}
+				enetc_bulk_flip_buff(rx_ring, orig_i, i);
 				xdp_redirect_frm_cnt++;
 				rx_ring->stats.xdp_redirect++;
 			}
@@ -3362,7 +3372,8 @@ static int enetc_int_vector_init(struct enetc_ndev_priv *priv, int i,
 	bdr->buffer_offset = ENETC_RXB_PAD;
 	priv->rx_ring[i] = bdr;
 
-	err = xdp_rxq_info_reg(&bdr->xdp.rxq, priv->ndev, i, 0);
+	err = __xdp_rxq_info_reg(&bdr->xdp.rxq, priv->ndev, i, 0,
+				 ENETC_RXB_DMA_SIZE_XDP);
 	if (err)
 		goto free_vector;
 
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index a86cfebedaa8..17e9bddb9ddd 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -714,7 +714,12 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq,
 	txq->bd.cur = bdp;
 
 	/* Trigger transmission start */
-	writel(0, txq->bd.reg_desc_active);
+	if (!(fep->quirks & FEC_QUIRK_ERR007885) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active) ||
+	    !readl(txq->bd.reg_desc_active))
+		writel(0, txq->bd.reg_desc_active);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h
index f8cdab62bf85..7725cb0c5c8a 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_common.h
@@ -108,14 +108,16 @@ struct hbg_irq_info {
 	bool re_enable;
 	bool need_print;
 	bool need_reset;
-	u64 count;
 
-	void (*irq_handle)(struct hbg_priv *priv, struct hbg_irq_info *info);
+	void (*irq_handle)(struct hbg_priv *priv,
+			   const struct hbg_irq_info *info);
 };
 
 struct hbg_vector {
 	char name[HBG_VECTOR_NUM][32];
-	struct hbg_irq_info *info_array;
+
+	u64 *stats_array;
+	const struct hbg_irq_info *info_array;
 	u32 info_array_len;
 };
 
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c
index 5e0ba4d5b08d..01ad82d2f5cc 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_debugfs.c
@@ -61,7 +61,7 @@ static int hbg_dbg_irq_info(struct seq_file *s, void *unused)
 {
 	struct net_device *netdev = dev_get_drvdata(s->private);
 	struct hbg_priv *priv = netdev_priv(netdev);
-	struct hbg_irq_info *info;
+	const struct hbg_irq_info *info;
 	u32 i;
 
 	for (i = 0; i < priv->vectors.info_array_len; i++) {
@@ -73,7 +73,7 @@ static int hbg_dbg_irq_info(struct seq_file *s, void *unused)
 								info->mask)),
 			   str_true_false(info->need_reset),
 			   str_true_false(info->need_print),
-			   info->count);
+			   priv->vectors.stats_array[i]);
 	}
 
 	return 0;
@@ -106,6 +106,7 @@ static int hbg_dbg_nic_state(struct seq_file *s, void *unused)
 {
 	struct net_device *netdev = dev_get_drvdata(s->private);
 	struct hbg_priv *priv = netdev_priv(netdev);
+	bool np_link_fail;
 
 	seq_printf(s, "event handling state: %s\n",
 		   state_str_true_false(priv, HBG_NIC_STATE_EVENT_HANDLING));
@@ -117,8 +118,10 @@ static int hbg_dbg_nic_state(struct seq_file *s, void *unused)
 		   reset_type_str[priv->reset_type]);
 	seq_printf(s, "need reset state: %s\n",
 		   state_str_true_false(priv, HBG_NIC_STATE_NEED_RESET));
-	seq_printf(s, "np_link fail state: %s\n",
-		   state_str_true_false(priv, HBG_NIC_STATE_NP_LINK_FAIL));
+
+	np_link_fail = !hbg_reg_read_field(priv, HBG_REG_AN_NEG_STATE_ADDR,
+					   HBG_REG_AN_NEG_STATE_NP_LINK_OK_B);
+	seq_printf(s, "np_link fail state: %s\n", str_true_false(np_link_fail));
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c
index d61c03f34ff0..f23fb5920c3c 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_diagnose.c
@@ -234,7 +234,7 @@ static u64 hbg_get_irq_stats(struct hbg_vector *vectors, u32 mask)
 
 	for (i = 0; i < vectors->info_array_len; i++)
 		if (vectors->info_array[i].mask == mask)
-			return vectors->info_array[i].count;
+			return vectors->stats_array[i];
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c
index 4e8cb66f601c..a0bcfb5a713d 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_err.c
@@ -26,12 +26,15 @@ static void hbg_restore_mac_table(struct hbg_priv *priv)
 
 static void hbg_restore_user_def_settings(struct hbg_priv *priv)
 {
+	/* The index of host mac is always 0. */
+	u64 rx_pause_addr = ether_addr_to_u64(priv->filter.mac_table[0].addr);
 	struct ethtool_pauseparam *pause_param = &priv->user_def.pause_param;
 
 	hbg_restore_mac_table(priv);
 	hbg_hw_set_mtu(priv, priv->netdev->mtu);
 	hbg_hw_set_pause_enable(priv, pause_param->tx_pause,
 				pause_param->rx_pause);
+	hbg_hw_set_rx_pause_mac_addr(priv, rx_pause_addr);
 }
 
 int hbg_rebuild(struct hbg_priv *priv)
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c
index 74a18033b444..9b65eef62b3f 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_hw.c
@@ -234,6 +234,10 @@ void hbg_hw_set_mac_filter_enable(struct hbg_priv *priv, u32 enable)
 {
 	hbg_reg_write_field(priv, HBG_REG_REC_FILT_CTRL_ADDR,
 			    HBG_REG_REC_FILT_CTRL_UC_MATCH_EN_B, enable);
+
+	/* only uc filter is supported, so set all bits of mc mask reg to 1 */
+	hbg_reg_write64(priv, HBG_REG_STATION_ADDR_LOW_MSK_0, U64_MAX);
+	hbg_reg_write64(priv, HBG_REG_STATION_ADDR_LOW_MSK_1, U64_MAX);
 }
 
 void hbg_hw_set_pause_enable(struct hbg_priv *priv, u32 tx_en, u32 rx_en)
@@ -242,6 +246,9 @@ void hbg_hw_set_pause_enable(struct hbg_priv *priv, u32 tx_en, u32 rx_en)
 			    HBG_REG_PAUSE_ENABLE_TX_B, tx_en);
 	hbg_reg_write_field(priv, HBG_REG_PAUSE_ENABLE_ADDR,
 			    HBG_REG_PAUSE_ENABLE_RX_B, rx_en);
+
+	hbg_reg_write_field(priv, HBG_REG_REC_FILT_CTRL_ADDR,
+			    HBG_REG_REC_FILT_CTRL_PAUSE_FRM_PASS_B, rx_en);
 }
 
 void hbg_hw_get_pause_enable(struct hbg_priv *priv, u32 *tx_en, u32 *rx_en)
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c
index e79e9ab3e530..8af0bc4cca21 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_irq.c
@@ -6,7 +6,7 @@
 #include "hbg_hw.h"
 
 static void hbg_irq_handle_err(struct hbg_priv *priv,
-			       struct hbg_irq_info *irq_info)
+			       const struct hbg_irq_info *irq_info)
 {
 	if (irq_info->need_print)
 		dev_err(&priv->pdev->dev,
@@ -17,30 +17,30 @@ static void hbg_irq_handle_err(struct hbg_priv *priv,
 }
 
 static void hbg_irq_handle_tx(struct hbg_priv *priv,
-			      struct hbg_irq_info *irq_info)
+			      const struct hbg_irq_info *irq_info)
 {
 	napi_schedule(&priv->tx_ring.napi);
 }
 
 static void hbg_irq_handle_rx(struct hbg_priv *priv,
-			      struct hbg_irq_info *irq_info)
+			      const struct hbg_irq_info *irq_info)
 {
 	napi_schedule(&priv->rx_ring.napi);
 }
 
 static void hbg_irq_handle_rx_buf_val(struct hbg_priv *priv,
-				      struct hbg_irq_info *irq_info)
+				      const struct hbg_irq_info *irq_info)
 {
 	priv->stats.rx_fifo_less_empty_thrsld_cnt++;
 }
 
 #define HBG_IRQ_I(name, handle) \
-	{#name, HBG_INT_MSK_##name##_B, false, false, false, 0, handle}
+	{#name, HBG_INT_MSK_##name##_B, false, false, false, handle}
 #define HBG_ERR_IRQ_I(name, need_print, ndde_reset) \
 	{#name, HBG_INT_MSK_##name##_B, true, need_print, \
-	ndde_reset, 0, hbg_irq_handle_err}
+	ndde_reset, hbg_irq_handle_err}
 
-static struct hbg_irq_info hbg_irqs[] = {
+static const struct hbg_irq_info hbg_irqs[] = {
 	HBG_IRQ_I(RX, hbg_irq_handle_rx),
 	HBG_IRQ_I(TX, hbg_irq_handle_tx),
 	HBG_ERR_IRQ_I(TX_PKT_CPL, true, true),
@@ -64,7 +64,7 @@ static struct hbg_irq_info hbg_irqs[] = {
 
 static irqreturn_t hbg_irq_handle(int irq_num, void *p)
 {
-	struct hbg_irq_info *info;
+	const struct hbg_irq_info *info;
 	struct hbg_priv *priv = p;
 	u32 status;
 	u32 i;
@@ -79,7 +79,7 @@ static irqreturn_t hbg_irq_handle(int irq_num, void *p)
 			hbg_hw_irq_enable(priv, info->mask, false);
 			hbg_hw_irq_clear(priv, info->mask);
 
-			info->count++;
+			priv->vectors.stats_array[i]++;
 			if (info->irq_handle)
 				info->irq_handle(priv, info);
 
@@ -132,6 +132,12 @@ int hbg_irq_init(struct hbg_priv *priv)
 					     irq_names_map[i]);
 	}
 
+	vectors->stats_array = devm_kcalloc(&priv->pdev->dev,
+					    ARRAY_SIZE(hbg_irqs),
+					    sizeof(u64), GFP_KERNEL);
+	if (!vectors->stats_array)
+		return -ENOMEM;
+
 	vectors->info_array = hbg_irqs;
 	vectors->info_array_len = ARRAY_SIZE(hbg_irqs);
 	return 0;
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c
index 2ac5454338e4..2e64dc1ab355 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c
@@ -21,7 +21,7 @@
 
 static void hbg_all_irq_enable(struct hbg_priv *priv, bool enabled)
 {
-	struct hbg_irq_info *info;
+	const struct hbg_irq_info *info;
 	u32 i;
 
 	for (i = 0; i < priv->vectors.info_array_len; i++) {
@@ -203,12 +203,12 @@ static int hbg_net_change_mtu(struct net_device *netdev, int new_mtu)
 	if (netif_running(netdev))
 		return -EBUSY;
 
-	hbg_hw_set_mtu(priv, new_mtu);
-	WRITE_ONCE(netdev->mtu, new_mtu);
-
 	dev_dbg(&priv->pdev->dev,
 		"change mtu from %u to %u\n", netdev->mtu, new_mtu);
 
+	hbg_hw_set_mtu(priv, new_mtu);
+	WRITE_ONCE(netdev->mtu, new_mtu);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c
index f29a937ad087..42b0083c9193 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_mdio.c
@@ -2,6 +2,7 @@
 // Copyright (c) 2024 Hisilicon Limited.
 
 #include <linux/phy.h>
+#include <linux/rtnetlink.h>
 #include "hbg_common.h"
 #include "hbg_hw.h"
 #include "hbg_mdio.h"
@@ -133,12 +134,17 @@ void hbg_fix_np_link_fail(struct hbg_priv *priv)
 {
 	struct device *dev = &priv->pdev->dev;
 
+	rtnl_lock();
+
 	if (priv->stats.np_link_fail_cnt >= HBG_NP_LINK_FAIL_RETRY_TIMES) {
 		dev_err(dev, "failed to fix the MAC link status\n");
 		priv->stats.np_link_fail_cnt = 0;
-		return;
+		goto unlock;
 	}
 
+	if (!priv->mac.phydev->link)
+		goto unlock;
+
 	priv->stats.np_link_fail_cnt++;
 	dev_err(dev, "failed to link between MAC and PHY, try to fix...\n");
 
@@ -147,6 +153,9 @@ void hbg_fix_np_link_fail(struct hbg_priv *priv)
 	 */
 	hbg_phy_stop(priv);
 	hbg_phy_start(priv);
+
+unlock:
+	rtnl_unlock();
 }
 
 static void hbg_phy_adjust_link(struct net_device *netdev)
diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h b/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h
index cc2cc612770d..a6e7f5e62b48 100644
--- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h
+++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_reg.h
@@ -68,6 +68,7 @@
 #define HBG_REG_TRANSMIT_CTRL_AN_EN_B		BIT(5)
 #define HBG_REG_REC_FILT_CTRL_ADDR		(HBG_REG_SGMII_BASE + 0x0064)
 #define HBG_REG_REC_FILT_CTRL_UC_MATCH_EN_B	BIT(0)
+#define HBG_REG_REC_FILT_CTRL_PAUSE_FRM_PASS_B	BIT(4)
 #define HBG_REG_RX_OCTETS_TOTAL_OK_ADDR		(HBG_REG_SGMII_BASE + 0x0080)
 #define HBG_REG_RX_OCTETS_BAD_ADDR		(HBG_REG_SGMII_BASE + 0x0084)
 #define HBG_REG_RX_UC_PKTS_ADDR			(HBG_REG_SGMII_BASE + 0x0088)
@@ -134,6 +135,8 @@
 #define HBG_REG_STATION_ADDR_HIGH_4_ADDR	(HBG_REG_SGMII_BASE + 0x0224)
 #define HBG_REG_STATION_ADDR_LOW_5_ADDR		(HBG_REG_SGMII_BASE + 0x0228)
 #define HBG_REG_STATION_ADDR_HIGH_5_ADDR	(HBG_REG_SGMII_BASE + 0x022C)
+#define HBG_REG_STATION_ADDR_LOW_MSK_0		(HBG_REG_SGMII_BASE + 0x0230)
+#define HBG_REG_STATION_ADDR_LOW_MSK_1		(HBG_REG_SGMII_BASE + 0x0238)
 
 /* PCU */
 #define HBG_REG_TX_FIFO_THRSLD_ADDR		(HBG_REG_SGMII_BASE + 0x0420)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
index 09749e9f7398..4e5d8bc39a1b 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c
@@ -61,7 +61,7 @@ static struct hns3_dbg_cmd_info hns3_dbg_cmd[] = {
 		.name = "tm_qset",
 		.cmd = HNAE3_DBG_CMD_TM_QSET,
 		.dentry = HNS3_DBG_DENTRY_TM,
-		.buf_len = HNS3_DBG_READ_LEN,
+		.buf_len = HNS3_DBG_READ_LEN_1MB,
 		.init = hns3_dbg_common_file_init,
 	},
 	{
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 9ff797fb36c4..b03b8758c777 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -473,20 +473,14 @@ static void hns3_mask_vector_irq(struct hns3_enet_tqp_vector *tqp_vector,
 	writel(mask_en, tqp_vector->mask_addr);
 }
 
-static void hns3_vector_enable(struct hns3_enet_tqp_vector *tqp_vector)
+static void hns3_irq_enable(struct hns3_enet_tqp_vector *tqp_vector)
 {
 	napi_enable(&tqp_vector->napi);
 	enable_irq(tqp_vector->vector_irq);
-
-	/* enable vector */
-	hns3_mask_vector_irq(tqp_vector, 1);
 }
 
-static void hns3_vector_disable(struct hns3_enet_tqp_vector *tqp_vector)
+static void hns3_irq_disable(struct hns3_enet_tqp_vector *tqp_vector)
 {
-	/* disable vector */
-	hns3_mask_vector_irq(tqp_vector, 0);
-
 	disable_irq(tqp_vector->vector_irq);
 	napi_disable(&tqp_vector->napi);
 	cancel_work_sync(&tqp_vector->rx_group.dim.work);
@@ -707,11 +701,42 @@ static int hns3_set_rx_cpu_rmap(struct net_device *netdev)
 	return 0;
 }
 
+static void hns3_enable_irqs_and_tqps(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	u16 i;
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_irq_enable(&priv->tqp_vector[i]);
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_mask_vector_irq(&priv->tqp_vector[i], 1);
+
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_tqp_enable(h->kinfo.tqp[i]);
+}
+
+static void hns3_disable_irqs_and_tqps(struct net_device *netdev)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	u16 i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_tqp_disable(h->kinfo.tqp[i]);
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_mask_vector_irq(&priv->tqp_vector[i], 0);
+
+	for (i = 0; i < priv->vector_num; i++)
+		hns3_irq_disable(&priv->tqp_vector[i]);
+}
+
 static int hns3_nic_net_up(struct net_device *netdev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	struct hnae3_handle *h = priv->ae_handle;
-	int i, j;
 	int ret;
 
 	ret = hns3_nic_reset_all_ring(h);
@@ -720,23 +745,13 @@ static int hns3_nic_net_up(struct net_device *netdev)
 
 	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
 
-	/* enable the vectors */
-	for (i = 0; i < priv->vector_num; i++)
-		hns3_vector_enable(&priv->tqp_vector[i]);
-
-	/* enable rcb */
-	for (j = 0; j < h->kinfo.num_tqps; j++)
-		hns3_tqp_enable(h->kinfo.tqp[j]);
+	hns3_enable_irqs_and_tqps(netdev);
 
 	/* start the ae_dev */
 	ret = h->ae_algo->ops->start ? h->ae_algo->ops->start(h) : 0;
 	if (ret) {
 		set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
-		while (j--)
-			hns3_tqp_disable(h->kinfo.tqp[j]);
-
-		for (j = i - 1; j >= 0; j--)
-			hns3_vector_disable(&priv->tqp_vector[j]);
+		hns3_disable_irqs_and_tqps(netdev);
 	}
 
 	return ret;
@@ -823,17 +838,9 @@ static void hns3_reset_tx_queue(struct hnae3_handle *h)
 static void hns3_nic_net_down(struct net_device *netdev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
-	struct hnae3_handle *h = hns3_get_handle(netdev);
 	const struct hnae3_ae_ops *ops;
-	int i;
 
-	/* disable vectors */
-	for (i = 0; i < priv->vector_num; i++)
-		hns3_vector_disable(&priv->tqp_vector[i]);
-
-	/* disable rcb */
-	for (i = 0; i < h->kinfo.num_tqps; i++)
-		hns3_tqp_disable(h->kinfo.tqp[i]);
+	hns3_disable_irqs_and_tqps(netdev);
 
 	/* stop ae_dev */
 	ops = priv->ae_handle->ae_algo->ops;
@@ -5864,8 +5871,6 @@ int hns3_set_channels(struct net_device *netdev,
 void hns3_external_lb_prepare(struct net_device *ndev, bool if_running)
 {
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
-	struct hnae3_handle *h = priv->ae_handle;
-	int i;
 
 	if (!if_running)
 		return;
@@ -5876,11 +5881,7 @@ void hns3_external_lb_prepare(struct net_device *ndev, bool if_running)
 	netif_carrier_off(ndev);
 	netif_tx_disable(ndev);
 
-	for (i = 0; i < priv->vector_num; i++)
-		hns3_vector_disable(&priv->tqp_vector[i]);
-
-	for (i = 0; i < h->kinfo.num_tqps; i++)
-		hns3_tqp_disable(h->kinfo.tqp[i]);
+	hns3_disable_irqs_and_tqps(ndev);
 
 	/* delay ring buffer clearing to hns3_reset_notify_uninit_enet
 	 * during reset process, because driver may not be able
@@ -5896,7 +5897,6 @@ void hns3_external_lb_restore(struct net_device *ndev, bool if_running)
 {
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
 	struct hnae3_handle *h = priv->ae_handle;
-	int i;
 
 	if (!if_running)
 		return;
@@ -5912,11 +5912,7 @@ void hns3_external_lb_restore(struct net_device *ndev, bool if_running)
 
 	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
 
-	for (i = 0; i < priv->vector_num; i++)
-		hns3_vector_enable(&priv->tqp_vector[i]);
-
-	for (i = 0; i < h->kinfo.num_tqps; i++)
-		hns3_tqp_enable(h->kinfo.tqp[i]);
+	hns3_enable_irqs_and_tqps(ndev);
 
 	netif_tx_wake_all_queues(ndev);
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
index 59cc9221185f..ec581d4b696f 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
@@ -440,6 +440,13 @@ static int hclge_ptp_create_clock(struct hclge_dev *hdev)
 	ptp->info.settime64 = hclge_ptp_settime;
 
 	ptp->info.n_alarm = 0;
+
+	spin_lock_init(&ptp->lock);
+	ptp->io_base = hdev->hw.hw.io_base + HCLGE_PTP_REG_OFFSET;
+	ptp->ts_cfg.rx_filter = HWTSTAMP_FILTER_NONE;
+	ptp->ts_cfg.tx_type = HWTSTAMP_TX_OFF;
+	hdev->ptp = ptp;
+
 	ptp->clock = ptp_clock_register(&ptp->info, &hdev->pdev->dev);
 	if (IS_ERR(ptp->clock)) {
 		dev_err(&hdev->pdev->dev,
@@ -451,12 +458,6 @@ static int hclge_ptp_create_clock(struct hclge_dev *hdev)
 		return -ENODEV;
 	}
 
-	spin_lock_init(&ptp->lock);
-	ptp->io_base = hdev->hw.hw.io_base + HCLGE_PTP_REG_OFFSET;
-	ptp->ts_cfg.rx_filter = HWTSTAMP_FILTER_NONE;
-	ptp->ts_cfg.tx_type = HWTSTAMP_TX_OFF;
-	hdev->ptp = ptp;
-
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
index 9ba767740a04..dada42e7e0ec 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c
@@ -1292,9 +1292,8 @@ static void hclgevf_sync_vlan_filter(struct hclgevf_dev *hdev)
 	rtnl_unlock();
 }
 
-static int hclgevf_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable)
+static int hclgevf_en_hw_strip_rxvtag_cmd(struct hclgevf_dev *hdev, bool enable)
 {
-	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
 	struct hclge_vf_to_pf_msg send_msg;
 
 	hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_VLAN,
@@ -1303,6 +1302,19 @@ static int hclgevf_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable)
 	return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0);
 }
 
+static int hclgevf_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable)
+{
+	struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle);
+	int ret;
+
+	ret = hclgevf_en_hw_strip_rxvtag_cmd(hdev, enable);
+	if (ret)
+		return ret;
+
+	hdev->rxvtag_strip_en = enable;
+	return 0;
+}
+
 static int hclgevf_reset_tqp(struct hnae3_handle *handle)
 {
 #define HCLGEVF_RESET_ALL_QUEUE_DONE	1U
@@ -2204,12 +2216,13 @@ static int hclgevf_rss_init_hw(struct hclgevf_dev *hdev)
 					  tc_valid, tc_size);
 }
 
-static int hclgevf_init_vlan_config(struct hclgevf_dev *hdev)
+static int hclgevf_init_vlan_config(struct hclgevf_dev *hdev,
+				    bool rxvtag_strip_en)
 {
 	struct hnae3_handle *nic = &hdev->nic;
 	int ret;
 
-	ret = hclgevf_en_hw_strip_rxvtag(nic, true);
+	ret = hclgevf_en_hw_strip_rxvtag(nic, rxvtag_strip_en);
 	if (ret) {
 		dev_err(&hdev->pdev->dev,
 			"failed to enable rx vlan offload, ret = %d\n", ret);
@@ -2879,7 +2892,7 @@ static int hclgevf_reset_hdev(struct hclgevf_dev *hdev)
 	if (ret)
 		return ret;
 
-	ret = hclgevf_init_vlan_config(hdev);
+	ret = hclgevf_init_vlan_config(hdev, hdev->rxvtag_strip_en);
 	if (ret) {
 		dev_err(&hdev->pdev->dev,
 			"failed(%d) to initialize VLAN config\n", ret);
@@ -2994,7 +3007,7 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 		goto err_config;
 	}
 
-	ret = hclgevf_init_vlan_config(hdev);
+	ret = hclgevf_init_vlan_config(hdev, true);
 	if (ret) {
 		dev_err(&hdev->pdev->dev,
 			"failed(%d) to initialize VLAN config\n", ret);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
index cccef3228461..0208425ab594 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h
@@ -253,6 +253,7 @@ struct hclgevf_dev {
 	int *vector_irq;
 
 	bool gro_en;
+	bool rxvtag_strip_en;
 
 	unsigned long vlan_del_fail_bmap[BITS_TO_LONGS(VLAN_N_VID)];
 
diff --git a/drivers/net/ethernet/intel/ice/ice_ddp.c b/drivers/net/ethernet/intel/ice/ice_ddp.c
index 69d5b1a28491..59323c019544 100644
--- a/drivers/net/ethernet/intel/ice/ice_ddp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ddp.c
@@ -2345,15 +2345,15 @@ ice_get_set_tx_topo(struct ice_hw *hw, u8 *buf, u16 buf_size,
 			cmd->set_flags |= ICE_AQC_TX_TOPO_FLAGS_SRC_RAM |
 					  ICE_AQC_TX_TOPO_FLAGS_LOAD_NEW;
 
-		if (hw->mac_type == ICE_MAC_GENERIC_3K_E825)
-			desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
+		desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
 	} else {
 		ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_tx_topo);
 		cmd->get_flags = ICE_AQC_TX_TOPO_GET_RAM;
-	}
 
-	if (hw->mac_type != ICE_MAC_GENERIC_3K_E825)
-		desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
+		if (hw->mac_type == ICE_MAC_E810 ||
+		    hw->mac_type == ICE_MAC_GENERIC)
+			desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
+	}
 
 	status = ice_aq_send_cmd(hw, &desc, buf, buf_size, cd);
 	if (status)
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_fdir.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_fdir.c
index 7752920d7a8e..1cca9b2262e8 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_fdir.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_fdir.c
@@ -2097,6 +2097,11 @@ int ice_vc_add_fdir_fltr(struct ice_vf *vf, u8 *msg)
 	pf = vf->pf;
 	dev = ice_pf_to_dev(pf);
 	vf_vsi = ice_get_vf_vsi(vf);
+	if (!vf_vsi) {
+		dev_err(dev, "Can not get FDIR vf_vsi for VF %u\n", vf->vf_id);
+		v_ret = VIRTCHNL_STATUS_ERR_PARAM;
+		goto err_exit;
+	}
 
 #define ICE_VF_MAX_FDIR_FILTERS	128
 	if (!ice_fdir_num_avail_fltr(&pf->hw, vf_vsi) ||
diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h
index 66544faab710..aef0e9775a33 100644
--- a/drivers/net/ethernet/intel/idpf/idpf.h
+++ b/drivers/net/ethernet/intel/idpf/idpf.h
@@ -629,13 +629,13 @@ bool idpf_is_capability_ena(struct idpf_adapter *adapter, bool all,
 	VIRTCHNL2_CAP_RX_HSPLIT_AT_L4V4	|\
 	VIRTCHNL2_CAP_RX_HSPLIT_AT_L4V6)
 
-#define IDPF_CAP_RX_CSUM_L4V4 (\
-	VIRTCHNL2_CAP_RX_CSUM_L4_IPV4_TCP	|\
-	VIRTCHNL2_CAP_RX_CSUM_L4_IPV4_UDP)
+#define IDPF_CAP_TX_CSUM_L4V4 (\
+	VIRTCHNL2_CAP_TX_CSUM_L4_IPV4_TCP	|\
+	VIRTCHNL2_CAP_TX_CSUM_L4_IPV4_UDP)
 
-#define IDPF_CAP_RX_CSUM_L4V6 (\
-	VIRTCHNL2_CAP_RX_CSUM_L4_IPV6_TCP	|\
-	VIRTCHNL2_CAP_RX_CSUM_L4_IPV6_UDP)
+#define IDPF_CAP_TX_CSUM_L4V6 (\
+	VIRTCHNL2_CAP_TX_CSUM_L4_IPV6_TCP	|\
+	VIRTCHNL2_CAP_TX_CSUM_L4_IPV6_UDP)
 
 #define IDPF_CAP_RX_CSUM (\
 	VIRTCHNL2_CAP_RX_CSUM_L3_IPV4		|\
@@ -644,11 +644,9 @@ bool idpf_is_capability_ena(struct idpf_adapter *adapter, bool all,
 	VIRTCHNL2_CAP_RX_CSUM_L4_IPV6_TCP	|\
 	VIRTCHNL2_CAP_RX_CSUM_L4_IPV6_UDP)
 
-#define IDPF_CAP_SCTP_CSUM (\
+#define IDPF_CAP_TX_SCTP_CSUM (\
 	VIRTCHNL2_CAP_TX_CSUM_L4_IPV4_SCTP	|\
-	VIRTCHNL2_CAP_TX_CSUM_L4_IPV6_SCTP	|\
-	VIRTCHNL2_CAP_RX_CSUM_L4_IPV4_SCTP	|\
-	VIRTCHNL2_CAP_RX_CSUM_L4_IPV6_SCTP)
+	VIRTCHNL2_CAP_TX_CSUM_L4_IPV6_SCTP)
 
 #define IDPF_CAP_TUNNEL_TX_CSUM (\
 	VIRTCHNL2_CAP_TX_CSUM_L3_SINGLE_TUNNEL	|\
diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c
index aa755dedb41d..82f09b4030bc 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c
@@ -703,8 +703,10 @@ static int idpf_cfg_netdev(struct idpf_vport *vport)
 {
 	struct idpf_adapter *adapter = vport->adapter;
 	struct idpf_vport_config *vport_config;
+	netdev_features_t other_offloads = 0;
+	netdev_features_t csum_offloads = 0;
+	netdev_features_t tso_offloads = 0;
 	netdev_features_t dflt_features;
-	netdev_features_t offloads = 0;
 	struct idpf_netdev_priv *np;
 	struct net_device *netdev;
 	u16 idx = vport->idx;
@@ -766,53 +768,32 @@ static int idpf_cfg_netdev(struct idpf_vport *vport)
 
 	if (idpf_is_cap_ena_all(adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS))
 		dflt_features |= NETIF_F_RXHASH;
-	if (idpf_is_cap_ena_all(adapter, IDPF_CSUM_CAPS, IDPF_CAP_RX_CSUM_L4V4))
-		dflt_features |= NETIF_F_IP_CSUM;
-	if (idpf_is_cap_ena_all(adapter, IDPF_CSUM_CAPS, IDPF_CAP_RX_CSUM_L4V6))
-		dflt_features |= NETIF_F_IPV6_CSUM;
+	if (idpf_is_cap_ena_all(adapter, IDPF_CSUM_CAPS, IDPF_CAP_TX_CSUM_L4V4))
+		csum_offloads |= NETIF_F_IP_CSUM;
+	if (idpf_is_cap_ena_all(adapter, IDPF_CSUM_CAPS, IDPF_CAP_TX_CSUM_L4V6))
+		csum_offloads |= NETIF_F_IPV6_CSUM;
 	if (idpf_is_cap_ena(adapter, IDPF_CSUM_CAPS, IDPF_CAP_RX_CSUM))
-		dflt_features |= NETIF_F_RXCSUM;
-	if (idpf_is_cap_ena_all(adapter, IDPF_CSUM_CAPS, IDPF_CAP_SCTP_CSUM))
-		dflt_features |= NETIF_F_SCTP_CRC;
+		csum_offloads |= NETIF_F_RXCSUM;
+	if (idpf_is_cap_ena_all(adapter, IDPF_CSUM_CAPS, IDPF_CAP_TX_SCTP_CSUM))
+		csum_offloads |= NETIF_F_SCTP_CRC;
 
 	if (idpf_is_cap_ena(adapter, IDPF_SEG_CAPS, VIRTCHNL2_CAP_SEG_IPV4_TCP))
-		dflt_features |= NETIF_F_TSO;
+		tso_offloads |= NETIF_F_TSO;
 	if (idpf_is_cap_ena(adapter, IDPF_SEG_CAPS, VIRTCHNL2_CAP_SEG_IPV6_TCP))
-		dflt_features |= NETIF_F_TSO6;
+		tso_offloads |= NETIF_F_TSO6;
 	if (idpf_is_cap_ena_all(adapter, IDPF_SEG_CAPS,
 				VIRTCHNL2_CAP_SEG_IPV4_UDP |
 				VIRTCHNL2_CAP_SEG_IPV6_UDP))
-		dflt_features |= NETIF_F_GSO_UDP_L4;
+		tso_offloads |= NETIF_F_GSO_UDP_L4;
 	if (idpf_is_cap_ena_all(adapter, IDPF_RSC_CAPS, IDPF_CAP_RSC))
-		offloads |= NETIF_F_GRO_HW;
-	/* advertise to stack only if offloads for encapsulated packets is
-	 * supported
-	 */
-	if (idpf_is_cap_ena(vport->adapter, IDPF_SEG_CAPS,
-			    VIRTCHNL2_CAP_SEG_TX_SINGLE_TUNNEL)) {
-		offloads |= NETIF_F_GSO_UDP_TUNNEL	|
-			    NETIF_F_GSO_GRE		|
-			    NETIF_F_GSO_GRE_CSUM	|
-			    NETIF_F_GSO_PARTIAL		|
-			    NETIF_F_GSO_UDP_TUNNEL_CSUM	|
-			    NETIF_F_GSO_IPXIP4		|
-			    NETIF_F_GSO_IPXIP6		|
-			    0;
-
-		if (!idpf_is_cap_ena_all(vport->adapter, IDPF_CSUM_CAPS,
-					 IDPF_CAP_TUNNEL_TX_CSUM))
-			netdev->gso_partial_features |=
-				NETIF_F_GSO_UDP_TUNNEL_CSUM;
-
-		netdev->gso_partial_features |= NETIF_F_GSO_GRE_CSUM;
-		offloads |= NETIF_F_TSO_MANGLEID;
-	}
+		other_offloads |= NETIF_F_GRO_HW;
 	if (idpf_is_cap_ena(adapter, IDPF_OTHER_CAPS, VIRTCHNL2_CAP_LOOPBACK))
-		offloads |= NETIF_F_LOOPBACK;
+		other_offloads |= NETIF_F_LOOPBACK;
 
-	netdev->features |= dflt_features;
-	netdev->hw_features |= dflt_features | offloads;
-	netdev->hw_enc_features |= dflt_features | offloads;
+	netdev->features |= dflt_features | csum_offloads | tso_offloads;
+	netdev->hw_features |=  netdev->features | other_offloads;
+	netdev->vlan_features |= netdev->features | other_offloads;
+	netdev->hw_enc_features |= dflt_features | other_offloads;
 	idpf_set_ethtool_ops(netdev);
 	netif_set_affinity_auto(netdev);
 	SET_NETDEV_DEV(netdev, &adapter->pdev->dev);
@@ -1132,11 +1113,9 @@ static struct idpf_vport *idpf_vport_alloc(struct idpf_adapter *adapter,
 
 	num_max_q = max(max_q->max_txq, max_q->max_rxq);
 	vport->q_vector_idxs = kcalloc(num_max_q, sizeof(u16), GFP_KERNEL);
-	if (!vport->q_vector_idxs) {
-		kfree(vport);
+	if (!vport->q_vector_idxs)
+		goto free_vport;
 
-		return NULL;
-	}
 	idpf_vport_init(vport, max_q);
 
 	/* This alloc is done separate from the LUT because it's not strictly
@@ -1146,11 +1125,9 @@ static struct idpf_vport *idpf_vport_alloc(struct idpf_adapter *adapter,
 	 */
 	rss_data = &adapter->vport_config[idx]->user_config.rss_data;
 	rss_data->rss_key = kzalloc(rss_data->rss_key_size, GFP_KERNEL);
-	if (!rss_data->rss_key) {
-		kfree(vport);
+	if (!rss_data->rss_key)
+		goto free_vector_idxs;
 
-		return NULL;
-	}
 	/* Initialize default rss key */
 	netdev_rss_key_fill((void *)rss_data->rss_key, rss_data->rss_key_size);
 
@@ -1163,6 +1140,13 @@ static struct idpf_vport *idpf_vport_alloc(struct idpf_adapter *adapter,
 	adapter->next_vport = idpf_get_free_slot(adapter);
 
 	return vport;
+
+free_vector_idxs:
+	kfree(vport->q_vector_idxs);
+free_vport:
+	kfree(vport);
+
+	return NULL;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/idpf/idpf_main.c b/drivers/net/ethernet/intel/idpf/idpf_main.c
index bec4a02c5373..b35713036a54 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_main.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_main.c
@@ -89,6 +89,7 @@ static void idpf_shutdown(struct pci_dev *pdev)
 {
 	struct idpf_adapter *adapter = pci_get_drvdata(pdev);
 
+	cancel_delayed_work_sync(&adapter->serv_task);
 	cancel_delayed_work_sync(&adapter->vc_event_task);
 	idpf_vc_core_deinit(adapter);
 	idpf_deinit_dflt_mbx(adapter);
diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index c35cc5cb1185..2f265c0959c7 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -319,6 +319,7 @@ struct igc_adapter {
 	struct timespec64 prev_ptp_time; /* Pre-reset PTP clock */
 	ktime_t ptp_reset_start; /* Reset time in clock mono */
 	struct system_time_snapshot snapshot;
+	struct mutex ptm_lock; /* Only allow one PTM transaction at a time */
 
 	char fw_version[32];
 
diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h
index 8e449904aa7d..d19325b0e6e0 100644
--- a/drivers/net/ethernet/intel/igc/igc_defines.h
+++ b/drivers/net/ethernet/intel/igc/igc_defines.h
@@ -574,7 +574,10 @@
 #define IGC_PTM_CTRL_SHRT_CYC(usec)	(((usec) & 0x3f) << 2)
 #define IGC_PTM_CTRL_PTM_TO(usec)	(((usec) & 0xff) << 8)
 
-#define IGC_PTM_SHORT_CYC_DEFAULT	1   /* Default short cycle interval */
+/* A short cycle time of 1us theoretically should work, but appears to be too
+ * short in practice.
+ */
+#define IGC_PTM_SHORT_CYC_DEFAULT	4   /* Default short cycle interval */
 #define IGC_PTM_CYC_TIME_DEFAULT	5   /* Default PTM cycle time */
 #define IGC_PTM_TIMEOUT_DEFAULT		255 /* Default timeout for PTM errors */
 
@@ -593,6 +596,7 @@
 #define IGC_PTM_STAT_T4M1_OVFL		BIT(3) /* T4 minus T1 overflow */
 #define IGC_PTM_STAT_ADJUST_1ST		BIT(4) /* 1588 timer adjusted during 1st PTM cycle */
 #define IGC_PTM_STAT_ADJUST_CYC		BIT(5) /* 1588 timer adjusted during non-1st PTM cycle */
+#define IGC_PTM_STAT_ALL		GENMASK(5, 0) /* Used to clear all status */
 
 /* PCIe PTM Cycle Control */
 #define IGC_PTM_CYCLE_CTRL_CYC_TIME(msec)	((msec) & 0x3ff) /* PTM Cycle Time (msec) */
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index f1330379e6bb..b1669d7cf435 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -7231,6 +7231,7 @@ static int igc_probe(struct pci_dev *pdev,
 
 err_register:
 	igc_release_hw_control(adapter);
+	igc_ptp_stop(adapter);
 err_eeprom:
 	if (!igc_check_reset_block(hw))
 		igc_reset_phy(hw);
diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c
index 946edbad4302..efc7b30e4211 100644
--- a/drivers/net/ethernet/intel/igc/igc_ptp.c
+++ b/drivers/net/ethernet/intel/igc/igc_ptp.c
@@ -974,45 +974,62 @@ static void igc_ptm_log_error(struct igc_adapter *adapter, u32 ptm_stat)
 	}
 }
 
+/* The PTM lock: adapter->ptm_lock must be held when calling igc_ptm_trigger() */
+static void igc_ptm_trigger(struct igc_hw *hw)
+{
+	u32 ctrl;
+
+	/* To "manually" start the PTM cycle we need to set the
+	 * trigger (TRIG) bit
+	 */
+	ctrl = rd32(IGC_PTM_CTRL);
+	ctrl |= IGC_PTM_CTRL_TRIG;
+	wr32(IGC_PTM_CTRL, ctrl);
+	/* Perform flush after write to CTRL register otherwise
+	 * transaction may not start
+	 */
+	wrfl();
+}
+
+/* The PTM lock: adapter->ptm_lock must be held when calling igc_ptm_reset() */
+static void igc_ptm_reset(struct igc_hw *hw)
+{
+	u32 ctrl;
+
+	ctrl = rd32(IGC_PTM_CTRL);
+	ctrl &= ~IGC_PTM_CTRL_TRIG;
+	wr32(IGC_PTM_CTRL, ctrl);
+	/* Write to clear all status */
+	wr32(IGC_PTM_STAT, IGC_PTM_STAT_ALL);
+}
+
 static int igc_phc_get_syncdevicetime(ktime_t *device,
 				      struct system_counterval_t *system,
 				      void *ctx)
 {
-	u32 stat, t2_curr_h, t2_curr_l, ctrl;
 	struct igc_adapter *adapter = ctx;
 	struct igc_hw *hw = &adapter->hw;
+	u32 stat, t2_curr_h, t2_curr_l;
 	int err, count = 100;
 	ktime_t t1, t2_curr;
 
-	/* Get a snapshot of system clocks to use as historic value. */
-	ktime_get_snapshot(&adapter->snapshot);
-
+	/* Doing this in a loop because in the event of a
+	 * badly timed (ha!) system clock adjustment, we may
+	 * get PTM errors from the PCI root, but these errors
+	 * are transitory. Repeating the process returns valid
+	 * data eventually.
+	 */
 	do {
-		/* Doing this in a loop because in the event of a
-		 * badly timed (ha!) system clock adjustment, we may
-		 * get PTM errors from the PCI root, but these errors
-		 * are transitory. Repeating the process returns valid
-		 * data eventually.
-		 */
+		/* Get a snapshot of system clocks to use as historic value. */
+		ktime_get_snapshot(&adapter->snapshot);
 
-		/* To "manually" start the PTM cycle we need to clear and
-		 * then set again the TRIG bit.
-		 */
-		ctrl = rd32(IGC_PTM_CTRL);
-		ctrl &= ~IGC_PTM_CTRL_TRIG;
-		wr32(IGC_PTM_CTRL, ctrl);
-		ctrl |= IGC_PTM_CTRL_TRIG;
-		wr32(IGC_PTM_CTRL, ctrl);
-
-		/* The cycle only starts "for real" when software notifies
-		 * that it has read the registers, this is done by setting
-		 * VALID bit.
-		 */
-		wr32(IGC_PTM_STAT, IGC_PTM_STAT_VALID);
+		igc_ptm_trigger(hw);
 
 		err = readx_poll_timeout(rd32, IGC_PTM_STAT, stat,
 					 stat, IGC_PTM_STAT_SLEEP,
 					 IGC_PTM_STAT_TIMEOUT);
+		igc_ptm_reset(hw);
+
 		if (err < 0) {
 			netdev_err(adapter->netdev, "Timeout reading IGC_PTM_STAT register\n");
 			return err;
@@ -1021,15 +1038,7 @@ static int igc_phc_get_syncdevicetime(ktime_t *device,
 		if ((stat & IGC_PTM_STAT_VALID) == IGC_PTM_STAT_VALID)
 			break;
 
-		if (stat & ~IGC_PTM_STAT_VALID) {
-			/* An error occurred, log it. */
-			igc_ptm_log_error(adapter, stat);
-			/* The STAT register is write-1-to-clear (W1C),
-			 * so write the previous error status to clear it.
-			 */
-			wr32(IGC_PTM_STAT, stat);
-			continue;
-		}
+		igc_ptm_log_error(adapter, stat);
 	} while (--count);
 
 	if (!count) {
@@ -1061,9 +1070,16 @@ static int igc_ptp_getcrosststamp(struct ptp_clock_info *ptp,
 {
 	struct igc_adapter *adapter = container_of(ptp, struct igc_adapter,
 						   ptp_caps);
+	int ret;
+
+	/* This blocks until any in progress PTM transactions complete */
+	mutex_lock(&adapter->ptm_lock);
+
+	ret = get_device_system_crosststamp(igc_phc_get_syncdevicetime,
+					    adapter, &adapter->snapshot, cts);
+	mutex_unlock(&adapter->ptm_lock);
 
-	return get_device_system_crosststamp(igc_phc_get_syncdevicetime,
-					     adapter, &adapter->snapshot, cts);
+	return ret;
 }
 
 static int igc_ptp_getcyclesx64(struct ptp_clock_info *ptp,
@@ -1162,6 +1178,7 @@ void igc_ptp_init(struct igc_adapter *adapter)
 	spin_lock_init(&adapter->ptp_tx_lock);
 	spin_lock_init(&adapter->free_timer_lock);
 	spin_lock_init(&adapter->tmreg_lock);
+	mutex_init(&adapter->ptm_lock);
 
 	adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;
 	adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF;
@@ -1174,6 +1191,7 @@ void igc_ptp_init(struct igc_adapter *adapter)
 	if (IS_ERR(adapter->ptp_clock)) {
 		adapter->ptp_clock = NULL;
 		netdev_err(netdev, "ptp_clock_register failed\n");
+		mutex_destroy(&adapter->ptm_lock);
 	} else if (adapter->ptp_clock) {
 		netdev_info(netdev, "PHC added\n");
 		adapter->ptp_flags |= IGC_PTP_ENABLED;
@@ -1203,10 +1221,12 @@ static void igc_ptm_stop(struct igc_adapter *adapter)
 	struct igc_hw *hw = &adapter->hw;
 	u32 ctrl;
 
+	mutex_lock(&adapter->ptm_lock);
 	ctrl = rd32(IGC_PTM_CTRL);
 	ctrl &= ~IGC_PTM_CTRL_EN;
 
 	wr32(IGC_PTM_CTRL, ctrl);
+	mutex_unlock(&adapter->ptm_lock);
 }
 
 /**
@@ -1237,13 +1257,18 @@ void igc_ptp_suspend(struct igc_adapter *adapter)
  **/
 void igc_ptp_stop(struct igc_adapter *adapter)
 {
+	if (!(adapter->ptp_flags & IGC_PTP_ENABLED))
+		return;
+
 	igc_ptp_suspend(adapter);
 
+	adapter->ptp_flags &= ~IGC_PTP_ENABLED;
 	if (adapter->ptp_clock) {
 		ptp_clock_unregister(adapter->ptp_clock);
 		netdev_info(adapter->netdev, "PHC removed\n");
 		adapter->ptp_flags &= ~IGC_PTP_ENABLED;
 	}
+	mutex_destroy(&adapter->ptm_lock);
 }
 
 /**
@@ -1255,13 +1280,18 @@ void igc_ptp_stop(struct igc_adapter *adapter)
 void igc_ptp_reset(struct igc_adapter *adapter)
 {
 	struct igc_hw *hw = &adapter->hw;
-	u32 cycle_ctrl, ctrl;
+	u32 cycle_ctrl, ctrl, stat;
 	unsigned long flags;
 	u32 timadj;
 
+	if (!(adapter->ptp_flags & IGC_PTP_ENABLED))
+		return;
+
 	/* reset the tstamp_config */
 	igc_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config);
 
+	mutex_lock(&adapter->ptm_lock);
+
 	spin_lock_irqsave(&adapter->tmreg_lock, flags);
 
 	switch (adapter->hw.mac.type) {
@@ -1290,14 +1320,19 @@ void igc_ptp_reset(struct igc_adapter *adapter)
 		ctrl = IGC_PTM_CTRL_EN |
 			IGC_PTM_CTRL_START_NOW |
 			IGC_PTM_CTRL_SHRT_CYC(IGC_PTM_SHORT_CYC_DEFAULT) |
-			IGC_PTM_CTRL_PTM_TO(IGC_PTM_TIMEOUT_DEFAULT) |
-			IGC_PTM_CTRL_TRIG;
+			IGC_PTM_CTRL_PTM_TO(IGC_PTM_TIMEOUT_DEFAULT);
 
 		wr32(IGC_PTM_CTRL, ctrl);
 
 		/* Force the first cycle to run. */
-		wr32(IGC_PTM_STAT, IGC_PTM_STAT_VALID);
+		igc_ptm_trigger(hw);
 
+		if (readx_poll_timeout_atomic(rd32, IGC_PTM_STAT, stat,
+					      stat, IGC_PTM_STAT_SLEEP,
+					      IGC_PTM_STAT_TIMEOUT))
+			netdev_err(adapter->netdev, "Timeout reading IGC_PTM_STAT register\n");
+
+		igc_ptm_reset(hw);
 		break;
 	default:
 		/* No work to do. */
@@ -1314,5 +1349,7 @@ void igc_ptp_reset(struct igc_adapter *adapter)
 out:
 	spin_unlock_irqrestore(&adapter->tmreg_lock, flags);
 
+	mutex_unlock(&adapter->ptm_lock);
+
 	wrfl();
 }
diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index 0a679e95196f..24499bb36c00 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -1223,7 +1223,7 @@ static void octep_hb_timeout_task(struct work_struct *work)
 		miss_cnt);
 	rtnl_lock();
 	if (netif_running(oct->netdev))
-		octep_stop(oct->netdev);
+		dev_close(oct->netdev);
 	rtnl_unlock();
 }
 
diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c
index 18c922dd5fc6..ccb69bc5c952 100644
--- a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c
@@ -835,7 +835,9 @@ static void octep_vf_tx_timeout(struct net_device *netdev, unsigned int txqueue)
 	struct octep_vf_device *oct = netdev_priv(netdev);
 
 	netdev_hold(netdev, NULL, GFP_ATOMIC);
-	schedule_work(&oct->tx_timeout_task);
+	if (!schedule_work(&oct->tx_timeout_task))
+		netdev_put(netdev, NULL);
+
 }
 
 static int octep_vf_set_mac(struct net_device *netdev, void *p)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/qos.c b/drivers/net/ethernet/marvell/octeontx2/nic/qos.c
index 0f844c14485a..35acc07bd964 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/qos.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/qos.c
@@ -165,6 +165,11 @@ static void __otx2_qos_txschq_cfg(struct otx2_nic *pfvf,
 
 		otx2_config_sched_shaping(pfvf, node, cfg, &num_regs);
 	} else if (level == NIX_TXSCH_LVL_TL2) {
+		/* configure parent txschq */
+		cfg->reg[num_regs] = NIX_AF_TL2X_PARENT(node->schq);
+		cfg->regval[num_regs] = (u64)hw->tx_link << 16;
+		num_regs++;
+
 		/* configure link cfg */
 		if (level == pfvf->qos.link_cfg_lvl) {
 			cfg->reg[num_regs] = NIX_AF_TL3_TL2X_LINKX_CFG(node->schq, hw->tx_link);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/rep.c b/drivers/net/ethernet/marvell/octeontx2/nic/rep.c
index 04e08e06f30f..7153a71dfc86 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/rep.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/rep.c
@@ -67,6 +67,8 @@ static int rvu_rep_mcam_flow_init(struct rep_dev *rep)
 
 		rsp = (struct npc_mcam_alloc_entry_rsp *)otx2_mbox_get_rsp
 			(&priv->mbox.mbox, 0, &req->hdr);
+		if (IS_ERR(rsp))
+			goto exit;
 
 		for (ent = 0; ent < rsp->count; ent++)
 			rep->flow_cfg->flow_ent[ent + allocated] = rsp->entry_list[ent];
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 43197b28b3e7..8fda4ce80d81 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -269,12 +269,8 @@ static const char * const mtk_clks_source_name[] = {
 	"ethwarp_wocpu2",
 	"ethwarp_wocpu1",
 	"ethwarp_wocpu0",
-	"top_usxgmii0_sel",
-	"top_usxgmii1_sel",
 	"top_sgm0_sel",
 	"top_sgm1_sel",
-	"top_xfi_phy0_xtal_sel",
-	"top_xfi_phy1_xtal_sel",
 	"top_eth_gmii_sel",
 	"top_eth_refck_50m_sel",
 	"top_eth_sys_200m_sel",
@@ -734,7 +730,7 @@ static void mtk_set_queue_speed(struct mtk_eth *eth, unsigned int idx,
 		case SPEED_100:
 			val |= MTK_QTX_SCH_MAX_RATE_EN |
 			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 103) |
-			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 3);
+			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 3) |
 			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_WEIGHT, 1);
 			break;
 		case SPEED_1000:
@@ -757,13 +753,13 @@ static void mtk_set_queue_speed(struct mtk_eth *eth, unsigned int idx,
 		case SPEED_100:
 			val |= MTK_QTX_SCH_MAX_RATE_EN |
 			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 1) |
-			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 5);
+			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 5) |
 			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_WEIGHT, 1);
 			break;
 		case SPEED_1000:
 			val |= MTK_QTX_SCH_MAX_RATE_EN |
-			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 10) |
-			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 5) |
+			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 1) |
+			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 6) |
 			       FIELD_PREP(MTK_QTX_SCH_MAX_RATE_WEIGHT, 10);
 			break;
 		default:
@@ -871,9 +867,25 @@ static const struct phylink_mac_ops mtk_phylink_ops = {
 	.mac_enable_tx_lpi = mtk_mac_enable_tx_lpi,
 };
 
+static void mtk_mdio_config(struct mtk_eth *eth)
+{
+	u32 val;
+
+	/* Configure MDC Divider */
+	val = FIELD_PREP(PPSC_MDC_CFG, eth->mdc_divider);
+
+	/* Configure MDC Turbo Mode */
+	if (mtk_is_netsys_v3_or_greater(eth))
+		mtk_m32(eth, 0, MISC_MDC_TURBO, MTK_MAC_MISC_V3);
+	else
+		val |= PPSC_MDC_TURBO;
+
+	mtk_m32(eth, PPSC_MDC_CFG, val, MTK_PPSC);
+}
+
 static int mtk_mdio_init(struct mtk_eth *eth)
 {
-	unsigned int max_clk = 2500000, divider;
+	unsigned int max_clk = 2500000;
 	struct device_node *mii_np;
 	int ret;
 	u32 val;
@@ -908,20 +920,9 @@ static int mtk_mdio_init(struct mtk_eth *eth)
 		}
 		max_clk = val;
 	}
-	divider = min_t(unsigned int, DIV_ROUND_UP(MDC_MAX_FREQ, max_clk), 63);
-
-	/* Configure MDC Turbo Mode */
-	if (mtk_is_netsys_v3_or_greater(eth))
-		mtk_m32(eth, 0, MISC_MDC_TURBO, MTK_MAC_MISC_V3);
-
-	/* Configure MDC Divider */
-	val = FIELD_PREP(PPSC_MDC_CFG, divider);
-	if (!mtk_is_netsys_v3_or_greater(eth))
-		val |= PPSC_MDC_TURBO;
-	mtk_m32(eth, PPSC_MDC_CFG, val, MTK_PPSC);
-
-	dev_dbg(eth->dev, "MDC is running on %d Hz\n", MDC_MAX_FREQ / divider);
-
+	eth->mdc_divider = min_t(unsigned int, DIV_ROUND_UP(MDC_MAX_FREQ, max_clk), 63);
+	mtk_mdio_config(eth);
+	dev_dbg(eth->dev, "MDC is running on %d Hz\n", MDC_MAX_FREQ / eth->mdc_divider);
 	ret = of_mdiobus_register(eth->mii_bus, mii_np);
 
 err_put_node:
@@ -2247,14 +2248,18 @@ skip_rx:
 		ring->data[idx] = new_data;
 		rxd->rxd1 = (unsigned int)dma_addr;
 release_desc:
+		if (MTK_HAS_CAPS(eth->soc->caps, MTK_36BIT_DMA)) {
+			if (unlikely(dma_addr == DMA_MAPPING_ERROR))
+				addr64 = FIELD_GET(RX_DMA_ADDR64_MASK,
+						   rxd->rxd2);
+			else
+				addr64 = RX_DMA_PREP_ADDR64(dma_addr);
+		}
+
 		if (MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628))
 			rxd->rxd2 = RX_DMA_LSO;
 		else
-			rxd->rxd2 = RX_DMA_PREP_PLEN0(ring->buf_size);
-
-		if (MTK_HAS_CAPS(eth->soc->caps, MTK_36BIT_DMA) &&
-		    likely(dma_addr != DMA_MAPPING_ERROR))
-			rxd->rxd2 |= RX_DMA_PREP_ADDR64(dma_addr);
+			rxd->rxd2 = RX_DMA_PREP_PLEN0(ring->buf_size) | addr64;
 
 		ring->calc_idx = idx;
 		done++;
@@ -3315,7 +3320,7 @@ static int mtk_start_dma(struct mtk_eth *eth)
 		if (mtk_is_netsys_v2_or_greater(eth))
 			val |= MTK_MUTLI_CNT | MTK_RESV_BUF |
 			       MTK_WCOMP_EN | MTK_DMAD_WR_WDONE |
-			       MTK_CHK_DDONE_EN | MTK_LEAKY_BUCKET_EN;
+			       MTK_CHK_DDONE_EN;
 		else
 			val |= MTK_RX_BT_32DWORDS;
 		mtk_w32(eth, val, reg_map->qdma.glo_cfg);
@@ -3974,6 +3979,10 @@ static int mtk_hw_init(struct mtk_eth *eth, bool reset)
 	else
 		mtk_hw_reset(eth);
 
+	/* No MT7628/88 support yet */
+	if (reset && !MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628))
+		mtk_mdio_config(eth);
+
 	if (mtk_is_netsys_v3_or_greater(eth)) {
 		/* Set FE to PDMAv2 if necessary */
 		val = mtk_r32(eth, MTK_FE_GLO_MISC);
@@ -4034,11 +4043,27 @@ static int mtk_hw_init(struct mtk_eth *eth, bool reset)
 	mtk_w32(eth, 0x21021000, MTK_FE_INT_GRP);
 
 	if (mtk_is_netsys_v3_or_greater(eth)) {
-		/* PSE should not drop port1, port8 and port9 packets */
-		mtk_w32(eth, 0x00000302, PSE_DROP_CFG);
+		/* PSE dummy page mechanism */
+		mtk_w32(eth, PSE_DUMMY_WORK_GDM(1) | PSE_DUMMY_WORK_GDM(2) |
+			PSE_DUMMY_WORK_GDM(3) | DUMMY_PAGE_THR, PSE_DUMY_REQ);
+
+		/* PSE free buffer drop threshold */
+		mtk_w32(eth, 0x00600009, PSE_IQ_REV(8));
+
+		/* PSE should not drop port8, port9 and port13 packets from
+		 * WDMA Tx
+		 */
+		mtk_w32(eth, 0x00002300, PSE_DROP_CFG);
+
+		/* PSE should drop packets to port8, port9 and port13 on WDMA Rx
+		 * ring full
+		 */
+		mtk_w32(eth, 0x00002300, PSE_PPE_DROP(0));
+		mtk_w32(eth, 0x00002300, PSE_PPE_DROP(1));
+		mtk_w32(eth, 0x00002300, PSE_PPE_DROP(2));
 
 		/* GDM and CDM Threshold */
-		mtk_w32(eth, 0x00000707, MTK_CDMW0_THRES);
+		mtk_w32(eth, 0x08000707, MTK_CDMW0_THRES);
 		mtk_w32(eth, 0x00000077, MTK_CDMW1_THRES);
 
 		/* Disable GDM1 RX CRC stripping */
@@ -4055,7 +4080,7 @@ static int mtk_hw_init(struct mtk_eth *eth, bool reset)
 		mtk_w32(eth, 0x00000300, PSE_DROP_CFG);
 
 		/* PSE should drop packets to port 8/9 on WDMA Rx ring full */
-		mtk_w32(eth, 0x00000300, PSE_PPE0_DROP);
+		mtk_w32(eth, 0x00000300, PSE_PPE_DROP(0));
 
 		/* PSE Free Queue Flow Control  */
 		mtk_w32(eth, 0x01fa01f4, PSE_FQFC_CFG2);
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index 90a377ab4359..88ef2e9c50fc 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -151,7 +151,15 @@
 #define PSE_FQFC_CFG1		0x100
 #define PSE_FQFC_CFG2		0x104
 #define PSE_DROP_CFG		0x108
-#define PSE_PPE0_DROP		0x110
+#define PSE_PPE_DROP(x)		(0x110 + ((x) * 0x4))
+
+/* PSE Last FreeQ Page Request Control */
+#define PSE_DUMY_REQ		0x10C
+/* PSE_DUMY_REQ is not a typo but actually called like that also in
+ * MediaTek's datasheet
+ */
+#define PSE_DUMMY_WORK_GDM(x)	BIT(16 + (x))
+#define DUMMY_PAGE_THR		0x1
 
 /* PSE Input Queue Reservation Register*/
 #define PSE_IQ_REV(x)		(0x140 + (((x) - 1) << 2))
@@ -1271,6 +1279,7 @@ struct mtk_eth {
 	struct clk			*clks[MTK_CLK_MAX];
 
 	struct mii_bus			*mii_bus;
+	unsigned int			mdc_divider;
 	struct work_struct		pending_work;
 	unsigned long			state;
 
diff --git a/drivers/net/ethernet/mediatek/mtk_star_emac.c b/drivers/net/ethernet/mediatek/mtk_star_emac.c
index 76f202d7f055..b175119a6a7d 100644
--- a/drivers/net/ethernet/mediatek/mtk_star_emac.c
+++ b/drivers/net/ethernet/mediatek/mtk_star_emac.c
@@ -1163,6 +1163,7 @@ static int mtk_star_tx_poll(struct napi_struct *napi, int budget)
 	struct net_device *ndev = priv->ndev;
 	unsigned int head = ring->head;
 	unsigned int entry = ring->tail;
+	unsigned long flags;
 
 	while (entry != head && count < (MTK_STAR_RING_NUM_DESCS - 1)) {
 		ret = mtk_star_tx_complete_one(priv);
@@ -1182,9 +1183,9 @@ static int mtk_star_tx_poll(struct napi_struct *napi, int budget)
 		netif_wake_queue(ndev);
 
 	if (napi_complete(napi)) {
-		spin_lock(&priv->lock);
+		spin_lock_irqsave(&priv->lock, flags);
 		mtk_star_enable_dma_irq(priv, false, true);
-		spin_unlock(&priv->lock);
+		spin_unlock_irqrestore(&priv->lock, flags);
 	}
 
 	return 0;
@@ -1341,16 +1342,16 @@ push_new_skb:
 static int mtk_star_rx_poll(struct napi_struct *napi, int budget)
 {
 	struct mtk_star_priv *priv;
+	unsigned long flags;
 	int work_done = 0;
 
 	priv = container_of(napi, struct mtk_star_priv, rx_napi);
 
 	work_done = mtk_star_rx(priv, budget);
-	if (work_done < budget) {
-		napi_complete_done(napi, work_done);
-		spin_lock(&priv->lock);
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		spin_lock_irqsave(&priv->lock, flags);
 		mtk_star_enable_dma_irq(priv, true, false);
-		spin_unlock(&priv->lock);
+		spin_unlock_irqrestore(&priv->lock, flags);
 	}
 
 	return work_done;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
index 532c7fa94d17..dbd9482359e1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c
@@ -176,6 +176,7 @@ static int mlx5e_tx_reporter_ptpsq_unhealthy_recover(void *ctx)
 
 	priv = ptpsq->txqsq.priv;
 
+	rtnl_lock();
 	mutex_lock(&priv->state_lock);
 	chs = &priv->channels;
 	netdev = priv->netdev;
@@ -183,22 +184,19 @@ static int mlx5e_tx_reporter_ptpsq_unhealthy_recover(void *ctx)
 	carrier_ok = netif_carrier_ok(netdev);
 	netif_carrier_off(netdev);
 
-	rtnl_lock();
 	mlx5e_deactivate_priv_channels(priv);
-	rtnl_unlock();
 
 	mlx5e_ptp_close(chs->ptp);
 	err = mlx5e_ptp_open(priv, &chs->params, chs->c[0]->lag_port, &chs->ptp);
 
-	rtnl_lock();
 	mlx5e_activate_priv_channels(priv);
-	rtnl_unlock();
 
 	/* return carrier back if needed */
 	if (carrier_ok)
 		netif_carrier_on(netdev);
 
 	mutex_unlock(&priv->state_lock);
+	rtnl_unlock();
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
index 5c762a71818d..7a18a469961d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c
@@ -165,9 +165,6 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv,
 	struct flow_match_enc_keyid enc_keyid;
 	void *misc_c, *misc_v;
 
-	misc_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
-	misc_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
-
 	if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID))
 		return 0;
 
@@ -182,6 +179,30 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv,
 		err = mlx5e_tc_tun_parse_vxlan_gbp_option(priv, spec, f);
 		if (err)
 			return err;
+
+		/* We can't mix custom tunnel headers with symbolic ones and we
+		 * don't have a symbolic field name for GBP, so we use custom
+		 * tunnel headers in this case. We need hardware support to
+		 * match on custom tunnel headers, but we already know it's
+		 * supported because the previous call successfully checked for
+		 * that.
+		 */
+		misc_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+				      misc_parameters_5);
+		misc_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
+				      misc_parameters_5);
+
+		/* Shift by 8 to account for the reserved bits in the vxlan
+		 * header after the VNI.
+		 */
+		MLX5_SET(fte_match_set_misc5, misc_c, tunnel_header_1,
+			 be32_to_cpu(enc_keyid.mask->keyid) << 8);
+		MLX5_SET(fte_match_set_misc5, misc_v, tunnel_header_1,
+			 be32_to_cpu(enc_keyid.key->keyid) << 8);
+
+		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_5;
+
+		return 0;
 	}
 
 	/* match on VNI is required */
@@ -195,6 +216,11 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv,
 		return -EOPNOTSUPP;
 	}
 
+	misc_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+			      misc_parameters);
+	misc_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
+			      misc_parameters);
+
 	MLX5_SET(fte_match_set_misc, misc_c, vxlan_vni,
 		 be32_to_cpu(enc_keyid.mask->keyid));
 	MLX5_SET(fte_match_set_misc, misc_v, vxlan_vni,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 9ba99609999f..f1d908f61134 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1750,9 +1750,6 @@ extra_split_attr_dests_needed(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr
 	    !list_is_first(&attr->list, &flow->attrs))
 		return 0;
 
-	if (flow_flag_test(flow, SLOW))
-		return 0;
-
 	esw_attr = attr->esw_attr;
 	if (!esw_attr->split_count ||
 	    esw_attr->split_count == esw_attr->out_count - 1)
@@ -1766,7 +1763,7 @@ extra_split_attr_dests_needed(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr
 	for (i = esw_attr->split_count; i < esw_attr->out_count; i++) {
 		/* external dest with encap is considered as internal by firmware */
 		if (esw_attr->dests[i].vport == MLX5_VPORT_UPLINK &&
-		    !(esw_attr->dests[i].flags & MLX5_ESW_DEST_ENCAP_VALID))
+		    !(esw_attr->dests[i].flags & MLX5_ESW_DEST_ENCAP))
 			ext_dest = true;
 		else
 			int_dest = true;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index a6a8eea5980c..0e3a977d5332 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -3533,7 +3533,9 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
 	int err;
 
 	mutex_init(&esw->offloads.termtbl_mutex);
-	mlx5_rdma_enable_roce(esw->dev);
+	err = mlx5_rdma_enable_roce(esw->dev);
+	if (err)
+		goto err_roce;
 
 	err = mlx5_esw_host_number_init(esw);
 	if (err)
@@ -3594,6 +3596,7 @@ err_vport_metadata:
 	esw_offloads_metadata_uninit(esw);
 err_metadata:
 	mlx5_rdma_disable_roce(esw->dev);
+err_roce:
 	mutex_destroy(&esw->offloads.termtbl_mutex);
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c
index eb3bd9c7f66e..ca9ecec358b2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c
@@ -637,10 +637,6 @@ struct mlx5_ttc_table *mlx5_create_inner_ttc_table(struct mlx5_core_dev *dev,
 	bool use_l4_type;
 	int err;
 
-	ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL);
-	if (!ttc)
-		return ERR_PTR(-ENOMEM);
-
 	switch (params->ns_type) {
 	case MLX5_FLOW_NAMESPACE_PORT_SEL:
 		use_l4_type = MLX5_CAP_GEN_2(dev, pcc_ifa2) &&
@@ -654,7 +650,16 @@ struct mlx5_ttc_table *mlx5_create_inner_ttc_table(struct mlx5_core_dev *dev,
 		return ERR_PTR(-EINVAL);
 	}
 
+	ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL);
+	if (!ttc)
+		return ERR_PTR(-ENOMEM);
+
 	ns = mlx5_get_flow_namespace(dev, params->ns_type);
+	if (!ns) {
+		kvfree(ttc);
+		return ERR_PTR(-EOPNOTSUPP);
+	}
+
 	groups = use_l4_type ? &inner_ttc_groups[TTC_GROUPS_USE_L4_TYPE] :
 			       &inner_ttc_groups[TTC_GROUPS_DEFAULT];
 
@@ -710,10 +715,6 @@ struct mlx5_ttc_table *mlx5_create_ttc_table(struct mlx5_core_dev *dev,
 	bool use_l4_type;
 	int err;
 
-	ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL);
-	if (!ttc)
-		return ERR_PTR(-ENOMEM);
-
 	switch (params->ns_type) {
 	case MLX5_FLOW_NAMESPACE_PORT_SEL:
 		use_l4_type = MLX5_CAP_GEN_2(dev, pcc_ifa2) &&
@@ -727,7 +728,16 @@ struct mlx5_ttc_table *mlx5_create_ttc_table(struct mlx5_core_dev *dev,
 		return ERR_PTR(-EINVAL);
 	}
 
+	ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL);
+	if (!ttc)
+		return ERR_PTR(-ENOMEM);
+
 	ns = mlx5_get_flow_namespace(dev, params->ns_type);
+	if (!ns) {
+		kvfree(ttc);
+		return ERR_PTR(-EOPNOTSUPP);
+	}
+
 	groups = use_l4_type ? &ttc_groups[TTC_GROUPS_USE_L4_TYPE] :
 			       &ttc_groups[TTC_GROUPS_DEFAULT];
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rdma.c b/drivers/net/ethernet/mellanox/mlx5/core/rdma.c
index a42f6cd99b74..5c552b71e371 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/rdma.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/rdma.c
@@ -118,8 +118,8 @@ static void mlx5_rdma_make_default_gid(struct mlx5_core_dev *dev, union ib_gid *
 
 static int mlx5_rdma_add_roce_addr(struct mlx5_core_dev *dev)
 {
+	u8 mac[ETH_ALEN] = {};
 	union ib_gid gid;
-	u8 mac[ETH_ALEN];
 
 	mlx5_rdma_make_default_gid(dev, &gid);
 	return mlx5_core_roce_gid_set(dev, 0,
@@ -140,17 +140,17 @@ void mlx5_rdma_disable_roce(struct mlx5_core_dev *dev)
 	mlx5_nic_vport_disable_roce(dev);
 }
 
-void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev)
+int mlx5_rdma_enable_roce(struct mlx5_core_dev *dev)
 {
 	int err;
 
 	if (!MLX5_CAP_GEN(dev, roce))
-		return;
+		return 0;
 
 	err = mlx5_nic_vport_enable_roce(dev);
 	if (err) {
 		mlx5_core_err(dev, "Failed to enable RoCE: %d\n", err);
-		return;
+		return err;
 	}
 
 	err = mlx5_rdma_add_roce_addr(dev);
@@ -165,10 +165,11 @@ void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev)
 		goto del_roce_addr;
 	}
 
-	return;
+	return err;
 
 del_roce_addr:
 	mlx5_rdma_del_roce_addr(dev);
 disable_roce:
 	mlx5_nic_vport_disable_roce(dev);
+	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rdma.h b/drivers/net/ethernet/mellanox/mlx5/core/rdma.h
index 750cff2a71a4..3d9e76c3d42f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/rdma.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/rdma.h
@@ -8,12 +8,12 @@
 
 #ifdef CONFIG_MLX5_ESWITCH
 
-void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev);
+int mlx5_rdma_enable_roce(struct mlx5_core_dev *dev);
 void mlx5_rdma_disable_roce(struct mlx5_core_dev *dev);
 
 #else /* CONFIG_MLX5_ESWITCH */
 
-static inline void mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) {}
+static inline int mlx5_rdma_enable_roce(struct mlx5_core_dev *dev) { return 0; }
 static inline void mlx5_rdma_disable_roce(struct mlx5_core_dev *dev) {}
 
 #endif /* CONFIG_MLX5_ESWITCH */
diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c
index 23760b613d3e..e2d6bfb5d693 100644
--- a/drivers/net/ethernet/microchip/lan743x_main.c
+++ b/drivers/net/ethernet/microchip/lan743x_main.c
@@ -1815,6 +1815,7 @@ static void lan743x_tx_frame_add_lso(struct lan743x_tx *tx,
 	if (nr_frags <= 0) {
 		tx->frame_data0 |= TX_DESC_DATA0_LS_;
 		tx->frame_data0 |= TX_DESC_DATA0_IOC_;
+		tx->frame_last = tx->frame_first;
 	}
 	tx_descriptor = &tx->ring_cpu_ptr[tx->frame_tail];
 	tx_descriptor->data0 = cpu_to_le32(tx->frame_data0);
@@ -1884,6 +1885,7 @@ static int lan743x_tx_frame_add_fragment(struct lan743x_tx *tx,
 		tx->frame_first = 0;
 		tx->frame_data0 = 0;
 		tx->frame_tail = 0;
+		tx->frame_last = 0;
 		return -ENOMEM;
 	}
 
@@ -1924,16 +1926,18 @@ static void lan743x_tx_frame_end(struct lan743x_tx *tx,
 	    TX_DESC_DATA0_DTYPE_DATA_) {
 		tx->frame_data0 |= TX_DESC_DATA0_LS_;
 		tx->frame_data0 |= TX_DESC_DATA0_IOC_;
+		tx->frame_last = tx->frame_tail;
 	}
 
-	tx_descriptor = &tx->ring_cpu_ptr[tx->frame_tail];
-	buffer_info = &tx->buffer_info[tx->frame_tail];
+	tx_descriptor = &tx->ring_cpu_ptr[tx->frame_last];
+	buffer_info = &tx->buffer_info[tx->frame_last];
 	buffer_info->skb = skb;
 	if (time_stamp)
 		buffer_info->flags |= TX_BUFFER_INFO_FLAG_TIMESTAMP_REQUESTED;
 	if (ignore_sync)
 		buffer_info->flags |= TX_BUFFER_INFO_FLAG_IGNORE_SYNC;
 
+	tx_descriptor = &tx->ring_cpu_ptr[tx->frame_tail];
 	tx_descriptor->data0 = cpu_to_le32(tx->frame_data0);
 	tx->frame_tail = lan743x_tx_next_index(tx, tx->frame_tail);
 	tx->last_tail = tx->frame_tail;
diff --git a/drivers/net/ethernet/microchip/lan743x_main.h b/drivers/net/ethernet/microchip/lan743x_main.h
index 7f73d66854be..db5fc73e41cc 100644
--- a/drivers/net/ethernet/microchip/lan743x_main.h
+++ b/drivers/net/ethernet/microchip/lan743x_main.h
@@ -980,6 +980,7 @@ struct lan743x_tx {
 	u32		frame_first;
 	u32		frame_data0;
 	u32		frame_tail;
+	u32		frame_last;
 
 	struct lan743x_tx_buffer_info *buffer_info;
 
diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index ef93df520887..08bee56aea35 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -830,6 +830,7 @@ EXPORT_SYMBOL(ocelot_vlan_prepare);
 int ocelot_vlan_add(struct ocelot *ocelot, int port, u16 vid, bool pvid,
 		    bool untagged)
 {
+	struct ocelot_port *ocelot_port = ocelot->ports[port];
 	int err;
 
 	/* Ignore VID 0 added to our RX filter by the 8021q module, since
@@ -849,6 +850,11 @@ int ocelot_vlan_add(struct ocelot *ocelot, int port, u16 vid, bool pvid,
 					   ocelot_bridge_vlan_find(ocelot, vid));
 		if (err)
 			return err;
+	} else if (ocelot_port->pvid_vlan &&
+		   ocelot_bridge_vlan_find(ocelot, vid) == ocelot_port->pvid_vlan) {
+		err = ocelot_port_set_pvid(ocelot, port, NULL);
+		if (err)
+			return err;
 	}
 
 	/* Untagged egress vlan clasification */
diff --git a/drivers/net/ethernet/realtek/rtase/rtase_main.c b/drivers/net/ethernet/realtek/rtase/rtase_main.c
index 2aacc1996796..55b8d3666153 100644
--- a/drivers/net/ethernet/realtek/rtase/rtase_main.c
+++ b/drivers/net/ethernet/realtek/rtase/rtase_main.c
@@ -1925,8 +1925,8 @@ static u16 rtase_calc_time_mitigation(u32 time_us)
 
 	time_us = min_t(int, time_us, RTASE_MITI_MAX_TIME);
 
-	msb = fls(time_us);
-	if (msb >= RTASE_MITI_COUNT_BIT_NUM) {
+	if (time_us > RTASE_MITI_TIME_COUNT_MASK) {
+		msb = fls(time_us);
 		time_unit = msb - RTASE_MITI_COUNT_BIT_NUM;
 		time_count = time_us >> (msb - RTASE_MITI_COUNT_BIT_NUM);
 	} else {
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
index 967a16212faf..0c011a47d5a3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
@@ -320,8 +320,8 @@ enum rtc_control {
 
 /* PTP and timestamping registers */
 
-#define GMAC3_X_ATSNS       GENMASK(19, 16)
-#define GMAC3_X_ATSNS_SHIFT 16
+#define GMAC3_X_ATSNS       GENMASK(29, 25)
+#define GMAC3_X_ATSNS_SHIFT 25
 
 #define GMAC_PTP_TCR_ATSFC	BIT(24)
 #define GMAC_PTP_TCR_ATSEN0	BIT(25)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
index a8b901cdf5cb..56b76aaa58f0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
@@ -553,7 +553,7 @@ void dwmac1000_get_ptptime(void __iomem *ptpaddr, u64 *ptp_time)
 	u64 ns;
 
 	ns = readl(ptpaddr + GMAC_PTP_ATNR);
-	ns += readl(ptpaddr + GMAC_PTP_ATSR) * NSEC_PER_SEC;
+	ns += (u64)readl(ptpaddr + GMAC_PTP_ATSR) * NSEC_PER_SEC;
 
 	*ptp_time = ns;
 }
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c
index 0f59aa982604..e2840fa241f2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_hwtstamp.c
@@ -222,7 +222,7 @@ static void get_ptptime(void __iomem *ptpaddr, u64 *ptp_time)
 	u64 ns;
 
 	ns = readl(ptpaddr + PTP_ATNR);
-	ns += readl(ptpaddr + PTP_ATSR) * NSEC_PER_SEC;
+	ns += (u64)readl(ptpaddr + PTP_ATSR) * NSEC_PER_SEC;
 
 	*ptp_time = ns;
 }
diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c
index 73c07f10f053..379b6e90121d 100644
--- a/drivers/net/ethernet/sun/niu.c
+++ b/drivers/net/ethernet/sun/niu.c
@@ -9064,6 +9064,8 @@ static void niu_try_msix(struct niu *np, u8 *ldg_num_map)
 		msi_vec[i].entry = i;
 	}
 
+	pdev->dev_flags |= PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST;
+
 	num_irqs = pci_enable_msix_range(pdev, msi_vec, 1, num_irqs);
 	if (num_irqs < 0) {
 		np->flags &= ~NIU_FLAGS_MSIX;
diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index c9fd34787c99..1e6d2335293d 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2666,7 +2666,7 @@ static int am65_cpsw_nuss_init_slave_ports(struct am65_cpsw_common *common)
 				of_property_read_bool(port_np, "ti,mac-only");
 
 		/* get phy/link info */
-		port->slave.port_np = port_np;
+		port->slave.port_np = of_node_get(port_np);
 		ret = of_get_phy_mode(port_np, &port->slave.phy_if);
 		if (ret) {
 			dev_err(dev, "%pOF read phy-mode err %d\n",
@@ -2720,6 +2720,17 @@ static void am65_cpsw_nuss_phylink_cleanup(struct am65_cpsw_common *common)
 	}
 }
 
+static void am65_cpsw_remove_dt(struct am65_cpsw_common *common)
+{
+	struct am65_cpsw_port *port;
+	int i;
+
+	for (i = 0; i < common->port_num; i++) {
+		port = &common->ports[i];
+		of_node_put(port->slave.port_np);
+	}
+}
+
 static int
 am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 {
@@ -3622,6 +3633,7 @@ err_ndevs_clear:
 	am65_cpsw_nuss_cleanup_ndev(common);
 	am65_cpsw_nuss_phylink_cleanup(common);
 	am65_cpts_release(common->cpts);
+	am65_cpsw_remove_dt(common);
 err_of_clear:
 	if (common->mdio_dev)
 		of_platform_device_destroy(common->mdio_dev, NULL);
@@ -3661,6 +3673,7 @@ static void am65_cpsw_nuss_remove(struct platform_device *pdev)
 	am65_cpsw_nuss_phylink_cleanup(common);
 	am65_cpts_release(common->cpts);
 	am65_cpsw_disable_serdes_phy(common);
+	am65_cpsw_remove_dt(common);
 
 	if (common->mdio_dev)
 		of_platform_device_destroy(common->mdio_dev, NULL);
diff --git a/drivers/net/ethernet/ti/icssg/icss_iep.c b/drivers/net/ethernet/ti/icssg/icss_iep.c
index b4a34c57b7b4..2a1c43316f46 100644
--- a/drivers/net/ethernet/ti/icssg/icss_iep.c
+++ b/drivers/net/ethernet/ti/icssg/icss_iep.c
@@ -412,6 +412,22 @@ static int icss_iep_perout_enable_hw(struct icss_iep *iep,
 	int ret;
 	u64 cmp;
 
+	if (!on) {
+		/* Disable CMP 1 */
+		regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
+				   IEP_CMP_CFG_CMP_EN(1), 0);
+
+		/* clear CMP regs */
+		regmap_write(iep->map, ICSS_IEP_CMP1_REG0, 0);
+		if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT)
+			regmap_write(iep->map, ICSS_IEP_CMP1_REG1, 0);
+
+		/* Disable sync */
+		regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0);
+
+		return 0;
+	}
+
 	/* Calculate width of the signal for PPS/PEROUT handling */
 	ts.tv_sec = req->on.sec;
 	ts.tv_nsec = req->on.nsec;
@@ -430,64 +446,39 @@ static int icss_iep_perout_enable_hw(struct icss_iep *iep,
 		if (ret)
 			return ret;
 
-		if (on) {
-			/* Configure CMP */
-			regmap_write(iep->map, ICSS_IEP_CMP1_REG0, lower_32_bits(cmp));
-			if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT)
-				regmap_write(iep->map, ICSS_IEP_CMP1_REG1, upper_32_bits(cmp));
-			/* Configure SYNC, based on req on width */
-			regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG,
-				     div_u64(ns_width, iep->def_inc));
-			regmap_write(iep->map, ICSS_IEP_SYNC0_PERIOD_REG, 0);
-			regmap_write(iep->map, ICSS_IEP_SYNC_START_REG,
-				     div_u64(ns_start, iep->def_inc));
-			regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0); /* one-shot mode */
-			/* Enable CMP 1 */
-			regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
-					   IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1));
-		} else {
-			/* Disable CMP 1 */
-			regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
-					   IEP_CMP_CFG_CMP_EN(1), 0);
-
-			/* clear regs */
-			regmap_write(iep->map, ICSS_IEP_CMP1_REG0, 0);
-			if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT)
-				regmap_write(iep->map, ICSS_IEP_CMP1_REG1, 0);
-		}
+		/* Configure CMP */
+		regmap_write(iep->map, ICSS_IEP_CMP1_REG0, lower_32_bits(cmp));
+		if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT)
+			regmap_write(iep->map, ICSS_IEP_CMP1_REG1, upper_32_bits(cmp));
+		/* Configure SYNC, based on req on width */
+		regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG,
+			     div_u64(ns_width, iep->def_inc));
+		regmap_write(iep->map, ICSS_IEP_SYNC0_PERIOD_REG, 0);
+		regmap_write(iep->map, ICSS_IEP_SYNC_START_REG,
+			     div_u64(ns_start, iep->def_inc));
+		regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0); /* one-shot mode */
+		/* Enable CMP 1 */
+		regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
+				   IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1));
 	} else {
-		if (on) {
-			u64 start_ns;
-
-			iep->period = ((u64)req->period.sec * NSEC_PER_SEC) +
-				      req->period.nsec;
-			start_ns = ((u64)req->period.sec * NSEC_PER_SEC)
-				   + req->period.nsec;
-			icss_iep_update_to_next_boundary(iep, start_ns);
-
-			regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG,
-				     div_u64(ns_width, iep->def_inc));
-			regmap_write(iep->map, ICSS_IEP_SYNC_START_REG,
-				     div_u64(ns_start, iep->def_inc));
-			/* Enable Sync in single shot mode  */
-			regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG,
-				     IEP_SYNC_CTRL_SYNC_N_EN(0) | IEP_SYNC_CTRL_SYNC_EN);
-			/* Enable CMP 1 */
-			regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
-					   IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1));
-		} else {
-			/* Disable CMP 1 */
-			regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
-					   IEP_CMP_CFG_CMP_EN(1), 0);
-
-			/* clear CMP regs */
-			regmap_write(iep->map, ICSS_IEP_CMP1_REG0, 0);
-			if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT)
-				regmap_write(iep->map, ICSS_IEP_CMP1_REG1, 0);
-
-			/* Disable sync */
-			regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0);
-		}
+		u64 start_ns;
+
+		iep->period = ((u64)req->period.sec * NSEC_PER_SEC) +
+				req->period.nsec;
+		start_ns = ((u64)req->period.sec * NSEC_PER_SEC)
+				+ req->period.nsec;
+		icss_iep_update_to_next_boundary(iep, start_ns);
+
+		regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG,
+			     div_u64(ns_width, iep->def_inc));
+		regmap_write(iep->map, ICSS_IEP_SYNC_START_REG,
+			     div_u64(ns_start, iep->def_inc));
+		/* Enable Sync in single shot mode  */
+		regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG,
+			     IEP_SYNC_CTRL_SYNC_N_EN(0) | IEP_SYNC_CTRL_SYNC_EN);
+		/* Enable CMP 1 */
+		regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG,
+				   IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1));
 	}
 
 	return 0;
@@ -498,11 +489,21 @@ static int icss_iep_perout_enable(struct icss_iep *iep,
 {
 	int ret = 0;
 
+	if (!on)
+		goto disable;
+
 	/* Reject requests with unsupported flags */
 	if (req->flags & ~(PTP_PEROUT_DUTY_CYCLE |
 			  PTP_PEROUT_PHASE))
 		return -EOPNOTSUPP;
 
+	/* Set default "on" time (1ms) for the signal if not passed by the app */
+	if (!(req->flags & PTP_PEROUT_DUTY_CYCLE)) {
+		req->on.sec = 0;
+		req->on.nsec = NSEC_PER_MSEC;
+	}
+
+disable:
 	mutex_lock(&iep->ptp_clk_mutex);
 
 	if (iep->pps_enabled) {
@@ -513,12 +514,6 @@ static int icss_iep_perout_enable(struct icss_iep *iep,
 	if (iep->perout_enabled == !!on)
 		goto exit;
 
-	/* Set default "on" time (1ms) for the signal if not passed by the app */
-	if (!(req->flags & PTP_PEROUT_DUTY_CYCLE)) {
-		req->on.sec = 0;
-		req->on.nsec = NSEC_PER_MSEC;
-	}
-
 	ret = icss_iep_perout_enable_hw(iep, req, on);
 	if (!ret)
 		iep->perout_enabled = !!on;
diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c
index 14002b026452..b4be76e13a2f 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_common.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_common.c
@@ -583,7 +583,7 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac,
 	first_desc = k3_cppi_desc_pool_alloc(tx_chn->desc_pool);
 	if (!first_desc) {
 		netdev_dbg(ndev, "xdp tx: failed to allocate descriptor\n");
-		goto drop_free_descs;	/* drop */
+		return ICSSG_XDP_CONSUMED;	/* drop */
 	}
 
 	if (page) { /* already DMA mapped by page_pool */
@@ -671,8 +671,10 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp,
 
 		q_idx = smp_processor_id() % emac->tx_ch_num;
 		result = emac_xmit_xdp_frame(emac, xdpf, page, q_idx);
-		if (result == ICSSG_XDP_CONSUMED)
+		if (result == ICSSG_XDP_CONSUMED) {
+			ndev->stats.tx_dropped++;
 			goto drop;
+		}
 
 		dev_sw_netstats_rx_add(ndev, xdpf->len);
 		return result;
@@ -1215,9 +1217,6 @@ void prueth_reset_rx_chan(struct prueth_rx_chn *chn,
 					  prueth_rx_cleanup);
 	if (disable)
 		k3_udma_glue_disable_rx_chn(chn->rx_chn);
-
-	page_pool_destroy(chn->pg_pool);
-	chn->pg_pool = NULL;
 }
 EXPORT_SYMBOL_GPL(prueth_reset_rx_chan);
 
diff --git a/drivers/net/ethernet/vertexcom/mse102x.c b/drivers/net/ethernet/vertexcom/mse102x.c
index 89dc4c401a8d..e4d993f31374 100644
--- a/drivers/net/ethernet/vertexcom/mse102x.c
+++ b/drivers/net/ethernet/vertexcom/mse102x.c
@@ -6,6 +6,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/if_vlan.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
@@ -33,7 +34,7 @@
 #define CMD_CTR		(0x2 << CMD_SHIFT)
 
 #define CMD_MASK	GENMASK(15, CMD_SHIFT)
-#define LEN_MASK	GENMASK(CMD_SHIFT - 1, 0)
+#define LEN_MASK	GENMASK(CMD_SHIFT - 2, 0)
 
 #define DET_CMD_LEN	4
 #define DET_SOF_LEN	2
@@ -262,7 +263,7 @@ static int mse102x_tx_frame_spi(struct mse102x_net *mse, struct sk_buff *txp,
 }
 
 static int mse102x_rx_frame_spi(struct mse102x_net *mse, u8 *buff,
-				unsigned int frame_len)
+				unsigned int frame_len, bool drop)
 {
 	struct mse102x_net_spi *mses = to_mse102x_spi(mse);
 	struct spi_transfer *xfer = &mses->spi_xfer;
@@ -280,6 +281,9 @@ static int mse102x_rx_frame_spi(struct mse102x_net *mse, u8 *buff,
 		netdev_err(mse->ndev, "%s: spi_sync() failed: %d\n",
 			   __func__, ret);
 		mse->stats.xfer_err++;
+	} else if (drop) {
+		netdev_dbg(mse->ndev, "%s: Drop frame\n", __func__);
+		ret = -EINVAL;
 	} else if (*sof != cpu_to_be16(DET_SOF)) {
 		netdev_dbg(mse->ndev, "%s: SPI start of frame is invalid (0x%04x)\n",
 			   __func__, *sof);
@@ -307,6 +311,7 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse)
 	struct sk_buff *skb;
 	unsigned int rxalign;
 	unsigned int rxlen;
+	bool drop = false;
 	__be16 rx = 0;
 	u16 cmd_resp;
 	u8 *rxpkt;
@@ -329,7 +334,8 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse)
 			net_dbg_ratelimited("%s: Unexpected response (0x%04x)\n",
 					    __func__, cmd_resp);
 			mse->stats.invalid_rts++;
-			return;
+			drop = true;
+			goto drop;
 		}
 
 		net_dbg_ratelimited("%s: Unexpected response to first CMD\n",
@@ -337,12 +343,20 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse)
 	}
 
 	rxlen = cmd_resp & LEN_MASK;
-	if (!rxlen) {
-		net_dbg_ratelimited("%s: No frame length defined\n", __func__);
+	if (rxlen < ETH_ZLEN || rxlen > VLAN_ETH_FRAME_LEN) {
+		net_dbg_ratelimited("%s: Invalid frame length: %d\n", __func__,
+				    rxlen);
 		mse->stats.invalid_len++;
-		return;
+		drop = true;
 	}
 
+	/* In case of a invalid CMD_RTS, the frame must be consumed anyway.
+	 * So assume the maximum possible frame length.
+	 */
+drop:
+	if (drop)
+		rxlen = VLAN_ETH_FRAME_LEN;
+
 	rxalign = ALIGN(rxlen + DET_SOF_LEN + DET_DFT_LEN, 4);
 	skb = netdev_alloc_skb_ip_align(mse->ndev, rxalign);
 	if (!skb)
@@ -353,7 +367,7 @@ static void mse102x_rx_pkt_spi(struct mse102x_net *mse)
 	 * They are copied, but ignored.
 	 */
 	rxpkt = skb_put(skb, rxlen) - DET_SOF_LEN;
-	if (mse102x_rx_frame_spi(mse, rxpkt, rxlen)) {
+	if (mse102x_rx_frame_spi(mse, rxpkt, rxlen, drop)) {
 		mse->ndev->stats.rx_errors++;
 		dev_kfree_skb(skb);
 		return;
@@ -509,6 +523,7 @@ static irqreturn_t mse102x_irq(int irq, void *_mse)
 static int mse102x_net_open(struct net_device *ndev)
 {
 	struct mse102x_net *mse = netdev_priv(ndev);
+	struct mse102x_net_spi *mses = to_mse102x_spi(mse);
 	int ret;
 
 	ret = request_threaded_irq(ndev->irq, NULL, mse102x_irq, IRQF_ONESHOT,
@@ -524,6 +539,13 @@ static int mse102x_net_open(struct net_device *ndev)
 
 	netif_carrier_on(ndev);
 
+	/* The SPI interrupt can stuck in case of pending packet(s).
+	 * So poll for possible packet(s) to re-arm the interrupt.
+	 */
+	mutex_lock(&mses->lock);
+	mse102x_rx_pkt_spi(mse);
+	mutex_unlock(&mses->lock);
+
 	netif_dbg(mse, ifup, ndev, "network device up\n");
 
 	return 0;
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
index 00b0b318df27..e69eaa65e0de 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
@@ -310,7 +310,8 @@ static bool wx_alloc_mapped_page(struct wx_ring *rx_ring,
 		return true;
 
 	page = page_pool_dev_alloc_pages(rx_ring->page_pool);
-	WARN_ON(!page);
+	if (unlikely(!page))
+		return false;
 	dma = page_pool_get_dma_addr(page);
 
 	bi->page_dma = dma;
@@ -546,7 +547,8 @@ static void wx_rx_checksum(struct wx_ring *ring,
 		return;
 
 	/* Hardware can't guarantee csum if IPv6 Dest Header found */
-	if (dptype.prot != WX_DEC_PTYPE_PROT_SCTP && WX_RXD_IPV6EX(rx_desc))
+	if (dptype.prot != WX_DEC_PTYPE_PROT_SCTP &&
+	    wx_test_staterr(rx_desc, WX_RXD_STAT_IPV6EX))
 		return;
 
 	/* if L4 checksum error */
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h
index 5b230ecbbabb..4c545b2aa997 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_type.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h
@@ -513,6 +513,7 @@ enum WX_MSCA_CMD_value {
 #define WX_RXD_STAT_L4CS             BIT(7) /* L4 xsum calculated */
 #define WX_RXD_STAT_IPCS             BIT(8) /* IP xsum calculated */
 #define WX_RXD_STAT_OUTERIPCS        BIT(10) /* Cloud IP xsum calculated*/
+#define WX_RXD_STAT_IPV6EX           BIT(12) /* IPv6 Dest Header */
 #define WX_RXD_STAT_TS               BIT(14) /* IEEE1588 Time Stamp */
 
 #define WX_RXD_ERR_OUTERIPER         BIT(26) /* CRC IP Header error */
@@ -589,8 +590,6 @@ enum wx_l2_ptypes {
 
 #define WX_RXD_PKTTYPE(_rxd) \
 	((le32_to_cpu((_rxd)->wb.lower.lo_dword.data) >> 9) & 0xFF)
-#define WX_RXD_IPV6EX(_rxd) \
-	((le32_to_cpu((_rxd)->wb.lower.lo_dword.data) >> 6) & 0x1)
 /*********************** Transmit Descriptor Config Masks ****************/
 #define WX_TXD_STAT_DD               BIT(0)  /* Descriptor Done */
 #define WX_TXD_DTYP_DATA             0       /* Adv Data Descriptor */
diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
index a6159214ec0a..91b3055a5a9f 100644
--- a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
+++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
@@ -625,7 +625,7 @@ static int ngbe_probe(struct pci_dev *pdev,
 	/* setup the private structure */
 	err = ngbe_sw_init(wx);
 	if (err)
-		goto err_free_mac_table;
+		goto err_pci_release_regions;
 
 	/* check if flash load is done after hw power up */
 	err = wx_check_flash_load(wx, NGBE_SPI_ILDR_STATUS_PERST);
@@ -719,6 +719,7 @@ err_register:
 err_clear_interrupt_scheme:
 	wx_clear_interrupt_scheme(wx);
 err_free_mac_table:
+	kfree(wx->rss_key);
 	kfree(wx->mac_table);
 err_pci_release_regions:
 	pci_release_selected_regions(pdev,
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
index a2e245e3b016..38206a46693b 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
@@ -611,7 +611,7 @@ static int txgbe_probe(struct pci_dev *pdev,
 	/* setup the private structure */
 	err = txgbe_sw_init(wx);
 	if (err)
-		goto err_free_mac_table;
+		goto err_pci_release_regions;
 
 	/* check if flash load is done after hw power up */
 	err = wx_check_flash_load(wx, TXGBE_SPI_ILDR_STATUS_PERST);
@@ -769,6 +769,7 @@ err_release_hw:
 	wx_clear_interrupt_scheme(wx);
 	wx_control_hw(wx, false);
 err_free_mac_table:
+	kfree(wx->rss_key);
 	kfree(wx->mac_table);
 err_pci_release_regions:
 	pci_release_selected_regions(pdev,
diff --git a/drivers/net/mdio/mdio-mux-meson-gxl.c b/drivers/net/mdio/mdio-mux-meson-gxl.c
index 00c66240136b..3dd12a8c8b03 100644
--- a/drivers/net/mdio/mdio-mux-meson-gxl.c
+++ b/drivers/net/mdio/mdio-mux-meson-gxl.c
@@ -17,6 +17,7 @@
 #define  REG2_LEDACT		GENMASK(23, 22)
 #define  REG2_LEDLINK		GENMASK(25, 24)
 #define  REG2_DIV4SEL		BIT(27)
+#define  REG2_REVERSED		BIT(28)
 #define  REG2_ADCBYPASS		BIT(30)
 #define  REG2_CLKINSEL		BIT(31)
 #define ETH_REG3		0x4
@@ -65,7 +66,7 @@ static void gxl_enable_internal_mdio(struct gxl_mdio_mux *priv)
 	 * The only constraint is that it must match the one in
 	 * drivers/net/phy/meson-gxl.c to properly match the PHY.
 	 */
-	writel(FIELD_PREP(REG2_PHYID, EPHY_GXL_ID),
+	writel(REG2_REVERSED | FIELD_PREP(REG2_PHYID, EPHY_GXL_ID),
 	       priv->regs + ETH_REG2);
 
 	/* Enable the internal phy */
diff --git a/drivers/net/phy/dp83822.c b/drivers/net/phy/dp83822.c
index 14f361549638..e32013eb0186 100644
--- a/drivers/net/phy/dp83822.c
+++ b/drivers/net/phy/dp83822.c
@@ -730,7 +730,7 @@ static int dp83822_phy_reset(struct phy_device *phydev)
 	return phydev->drv->config_init(phydev);
 }
 
-#ifdef CONFIG_OF_MDIO
+#if IS_ENABLED(CONFIG_OF_MDIO)
 static const u32 tx_amplitude_100base_tx_gain[] = {
 	80, 82, 83, 85, 87, 88, 90, 92,
 	93, 95, 97, 98, 100, 102, 103, 105,
diff --git a/drivers/net/phy/microchip.c b/drivers/net/phy/microchip.c
index 0e17cc458efd..93de88c1c8fd 100644
--- a/drivers/net/phy/microchip.c
+++ b/drivers/net/phy/microchip.c
@@ -37,47 +37,6 @@ static int lan88xx_write_page(struct phy_device *phydev, int page)
 	return __phy_write(phydev, LAN88XX_EXT_PAGE_ACCESS, page);
 }
 
-static int lan88xx_phy_config_intr(struct phy_device *phydev)
-{
-	int rc;
-
-	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
-		/* unmask all source and clear them before enable */
-		rc = phy_write(phydev, LAN88XX_INT_MASK, 0x7FFF);
-		rc = phy_read(phydev, LAN88XX_INT_STS);
-		rc = phy_write(phydev, LAN88XX_INT_MASK,
-			       LAN88XX_INT_MASK_MDINTPIN_EN_ |
-			       LAN88XX_INT_MASK_LINK_CHANGE_);
-	} else {
-		rc = phy_write(phydev, LAN88XX_INT_MASK, 0);
-		if (rc)
-			return rc;
-
-		/* Ack interrupts after they have been disabled */
-		rc = phy_read(phydev, LAN88XX_INT_STS);
-	}
-
-	return rc < 0 ? rc : 0;
-}
-
-static irqreturn_t lan88xx_handle_interrupt(struct phy_device *phydev)
-{
-	int irq_status;
-
-	irq_status = phy_read(phydev, LAN88XX_INT_STS);
-	if (irq_status < 0) {
-		phy_error(phydev);
-		return IRQ_NONE;
-	}
-
-	if (!(irq_status & LAN88XX_INT_STS_LINK_CHANGE_))
-		return IRQ_NONE;
-
-	phy_trigger_machine(phydev);
-
-	return IRQ_HANDLED;
-}
-
 static int lan88xx_suspend(struct phy_device *phydev)
 {
 	struct lan88xx_priv *priv = phydev->priv;
@@ -528,8 +487,9 @@ static struct phy_driver microchip_phy_driver[] = {
 	.config_aneg	= lan88xx_config_aneg,
 	.link_change_notify = lan88xx_link_change_notify,
 
-	.config_intr	= lan88xx_phy_config_intr,
-	.handle_interrupt = lan88xx_handle_interrupt,
+	/* Interrupt handling is broken, do not define related
+	 * functions to force polling.
+	 */
 
 	.suspend	= lan88xx_suspend,
 	.resume		= genphy_resume,
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 675fbd225378..cc1bfd22fb81 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -244,6 +244,46 @@ static bool phy_drv_wol_enabled(struct phy_device *phydev)
 	return wol.wolopts != 0;
 }
 
+static void phy_link_change(struct phy_device *phydev, bool up)
+{
+	struct net_device *netdev = phydev->attached_dev;
+
+	if (up)
+		netif_carrier_on(netdev);
+	else
+		netif_carrier_off(netdev);
+	phydev->adjust_link(netdev);
+	if (phydev->mii_ts && phydev->mii_ts->link_state)
+		phydev->mii_ts->link_state(phydev->mii_ts, phydev);
+}
+
+/**
+ * phy_uses_state_machine - test whether consumer driver uses PAL state machine
+ * @phydev: the target PHY device structure
+ *
+ * Ultimately, this aims to indirectly determine whether the PHY is attached
+ * to a consumer which uses the state machine by calling phy_start() and
+ * phy_stop().
+ *
+ * When the PHY driver consumer uses phylib, it must have previously called
+ * phy_connect_direct() or one of its derivatives, so that phy_prepare_link()
+ * has set up a hook for monitoring state changes.
+ *
+ * When the PHY driver is used by the MAC driver consumer through phylink (the
+ * only other provider of a phy_link_change() method), using the PHY state
+ * machine is not optional.
+ *
+ * Return: true if consumer calls phy_start() and phy_stop(), false otherwise.
+ */
+static bool phy_uses_state_machine(struct phy_device *phydev)
+{
+	if (phydev->phy_link_change == phy_link_change)
+		return phydev->attached_dev && phydev->adjust_link;
+
+	/* phydev->phy_link_change is implicitly phylink_phy_change() */
+	return true;
+}
+
 static bool mdio_bus_phy_may_suspend(struct phy_device *phydev)
 {
 	struct device_driver *drv = phydev->mdio.dev.driver;
@@ -310,7 +350,7 @@ static __maybe_unused int mdio_bus_phy_suspend(struct device *dev)
 	 * may call phy routines that try to grab the same lock, and that may
 	 * lead to a deadlock.
 	 */
-	if (phydev->attached_dev && phydev->adjust_link)
+	if (phy_uses_state_machine(phydev))
 		phy_stop_machine(phydev);
 
 	if (!mdio_bus_phy_may_suspend(phydev))
@@ -364,7 +404,7 @@ no_resume:
 		}
 	}
 
-	if (phydev->attached_dev && phydev->adjust_link)
+	if (phy_uses_state_machine(phydev))
 		phy_start_machine(phydev);
 
 	return 0;
@@ -1055,19 +1095,6 @@ struct phy_device *phy_find_first(struct mii_bus *bus)
 }
 EXPORT_SYMBOL(phy_find_first);
 
-static void phy_link_change(struct phy_device *phydev, bool up)
-{
-	struct net_device *netdev = phydev->attached_dev;
-
-	if (up)
-		netif_carrier_on(netdev);
-	else
-		netif_carrier_off(netdev);
-	phydev->adjust_link(netdev);
-	if (phydev->mii_ts && phydev->mii_ts->link_state)
-		phydev->mii_ts->link_state(phydev->mii_ts, phydev);
-}
-
 /**
  * phy_prepare_link - prepares the PHY layer to monitor link status
  * @phydev: target phy_device struct
diff --git a/drivers/net/phy/phy_led_triggers.c b/drivers/net/phy/phy_led_triggers.c
index bd3c9554f6ac..60893691d4c3 100644
--- a/drivers/net/phy/phy_led_triggers.c
+++ b/drivers/net/phy/phy_led_triggers.c
@@ -93,9 +93,8 @@ int phy_led_triggers_register(struct phy_device *phy)
 	if (!phy->phy_num_led_triggers)
 		return 0;
 
-	phy->led_link_trigger = devm_kzalloc(&phy->mdio.dev,
-					     sizeof(*phy->led_link_trigger),
-					     GFP_KERNEL);
+	phy->led_link_trigger = kzalloc(sizeof(*phy->led_link_trigger),
+					GFP_KERNEL);
 	if (!phy->led_link_trigger) {
 		err = -ENOMEM;
 		goto out_clear;
@@ -105,10 +104,9 @@ int phy_led_triggers_register(struct phy_device *phy)
 	if (err)
 		goto out_free_link;
 
-	phy->phy_led_triggers = devm_kcalloc(&phy->mdio.dev,
-					    phy->phy_num_led_triggers,
-					    sizeof(struct phy_led_trigger),
-					    GFP_KERNEL);
+	phy->phy_led_triggers = kcalloc(phy->phy_num_led_triggers,
+					sizeof(struct phy_led_trigger),
+					GFP_KERNEL);
 	if (!phy->phy_led_triggers) {
 		err = -ENOMEM;
 		goto out_unreg_link;
@@ -129,11 +127,11 @@ int phy_led_triggers_register(struct phy_device *phy)
 out_unreg:
 	while (i--)
 		phy_led_trigger_unregister(&phy->phy_led_triggers[i]);
-	devm_kfree(&phy->mdio.dev, phy->phy_led_triggers);
+	kfree(phy->phy_led_triggers);
 out_unreg_link:
 	phy_led_trigger_unregister(phy->led_link_trigger);
 out_free_link:
-	devm_kfree(&phy->mdio.dev, phy->led_link_trigger);
+	kfree(phy->led_link_trigger);
 	phy->led_link_trigger = NULL;
 out_clear:
 	phy->phy_num_led_triggers = 0;
@@ -147,8 +145,13 @@ void phy_led_triggers_unregister(struct phy_device *phy)
 
 	for (i = 0; i < phy->phy_num_led_triggers; i++)
 		phy_led_trigger_unregister(&phy->phy_led_triggers[i]);
+	kfree(phy->phy_led_triggers);
+	phy->phy_led_triggers = NULL;
 
-	if (phy->led_link_trigger)
+	if (phy->led_link_trigger) {
 		phy_led_trigger_unregister(phy->led_link_trigger);
+		kfree(phy->led_link_trigger);
+		phy->led_link_trigger = NULL;
+	}
 }
 EXPORT_SYMBOL_GPL(phy_led_triggers_unregister);
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index b68369e2342b..1bdd5d8bb5b0 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -81,6 +81,7 @@ struct phylink {
 	unsigned int pcs_state;
 
 	bool link_failed;
+	bool suspend_link_up;
 	bool major_config_failed;
 	bool mac_supports_eee_ops;
 	bool mac_supports_eee;
@@ -2545,14 +2546,16 @@ void phylink_suspend(struct phylink *pl, bool mac_wol)
 		/* Stop the resolver bringing the link up */
 		__set_bit(PHYLINK_DISABLE_MAC_WOL, &pl->phylink_disable_state);
 
-		/* Disable the carrier, to prevent transmit timeouts,
-		 * but one would hope all packets have been sent. This
-		 * also means phylink_resolve() will do nothing.
-		 */
-		if (pl->netdev)
-			netif_carrier_off(pl->netdev);
-		else
+		pl->suspend_link_up = phylink_link_is_up(pl);
+		if (pl->suspend_link_up) {
+			/* Disable the carrier, to prevent transmit timeouts,
+			 * but one would hope all packets have been sent. This
+			 * also means phylink_resolve() will do nothing.
+			 */
+			if (pl->netdev)
+				netif_carrier_off(pl->netdev);
 			pl->old_link_state = false;
+		}
 
 		/* We do not call mac_link_down() here as we want the
 		 * link to remain up to receive the WoL packets.
@@ -2603,15 +2606,18 @@ void phylink_resume(struct phylink *pl)
 	if (test_bit(PHYLINK_DISABLE_MAC_WOL, &pl->phylink_disable_state)) {
 		/* Wake-on-Lan enabled, MAC handling */
 
-		/* Call mac_link_down() so we keep the overall state balanced.
-		 * Do this under the state_mutex lock for consistency. This
-		 * will cause a "Link Down" message to be printed during
-		 * resume, which is harmless - the true link state will be
-		 * printed when we run a resolve.
-		 */
-		mutex_lock(&pl->state_mutex);
-		phylink_link_down(pl);
-		mutex_unlock(&pl->state_mutex);
+		if (pl->suspend_link_up) {
+			/* Call mac_link_down() so we keep the overall state
+			 * balanced. Do this under the state_mutex lock for
+			 * consistency. This will cause a "Link Down" message
+			 * to be printed during resume, which is harmless -
+			 * the true link state will be printed when we run a
+			 * resolve.
+			 */
+			mutex_lock(&pl->state_mutex);
+			phylink_link_down(pl);
+			mutex_unlock(&pl->state_mutex);
+		}
 
 		/* Re-apply the link parameters so that all the settings get
 		 * restored to the MAC.
diff --git a/drivers/net/ppp/ppp_synctty.c b/drivers/net/ppp/ppp_synctty.c
index 644e99fc3623..9c4932198931 100644
--- a/drivers/net/ppp/ppp_synctty.c
+++ b/drivers/net/ppp/ppp_synctty.c
@@ -506,6 +506,11 @@ ppp_sync_txmunge(struct syncppp *ap, struct sk_buff *skb)
 	unsigned char *data;
 	int islcp;
 
+	/* Ensure we can safely access protocol field and LCP code */
+	if (!pskb_may_pull(skb, 3)) {
+		kfree_skb(skb);
+		return NULL;
+	}
 	data  = skb->data;
 	proto = get_unaligned_be16(data);
 
diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c
index bb0bf1415872..7b3739b29c8f 100644
--- a/drivers/net/usb/rndis_host.c
+++ b/drivers/net/usb/rndis_host.c
@@ -630,16 +630,6 @@ static const struct driver_info	zte_rndis_info = {
 	.tx_fixup =	rndis_tx_fixup,
 };
 
-static const struct driver_info	wwan_rndis_info = {
-	.description =	"Mobile Broadband RNDIS device",
-	.flags =	FLAG_WWAN | FLAG_POINTTOPOINT | FLAG_FRAMING_RN | FLAG_NO_SETINT,
-	.bind =		rndis_bind,
-	.unbind =	rndis_unbind,
-	.status =	rndis_status,
-	.rx_fixup =	rndis_rx_fixup,
-	.tx_fixup =	rndis_tx_fixup,
-};
-
 /*-------------------------------------------------------------------------*/
 
 static const struct usb_device_id	products [] = {
@@ -676,11 +666,9 @@ static const struct usb_device_id	products [] = {
 	USB_INTERFACE_INFO(USB_CLASS_WIRELESS_CONTROLLER, 1, 3),
 	.driver_info = (unsigned long) &rndis_info,
 }, {
-	/* Mobile Broadband Modem, seen in Novatel Verizon USB730L and
-	 * Telit FN990A (RNDIS)
-	 */
+	/* Novatel Verizon USB730L */
 	USB_INTERFACE_INFO(USB_CLASS_MISC, 4, 1),
-	.driver_info = (unsigned long)&wwan_rndis_info,
+	.driver_info = (unsigned long) &rndis_info,
 },
 	{ },		// END
 };
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 7e4617216a4b..848fab51dfa1 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -3342,7 +3342,8 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 }
 
-static void virtnet_rx_pause(struct virtnet_info *vi, struct receive_queue *rq)
+static void __virtnet_rx_pause(struct virtnet_info *vi,
+			       struct receive_queue *rq)
 {
 	bool running = netif_running(vi->dev);
 
@@ -3352,17 +3353,63 @@ static void virtnet_rx_pause(struct virtnet_info *vi, struct receive_queue *rq)
 	}
 }
 
-static void virtnet_rx_resume(struct virtnet_info *vi, struct receive_queue *rq)
+static void virtnet_rx_pause_all(struct virtnet_info *vi)
+{
+	int i;
+
+	/*
+	 * Make sure refill_work does not run concurrently to
+	 * avoid napi_disable race which leads to deadlock.
+	 */
+	disable_delayed_refill(vi);
+	cancel_delayed_work_sync(&vi->refill);
+	for (i = 0; i < vi->max_queue_pairs; i++)
+		__virtnet_rx_pause(vi, &vi->rq[i]);
+}
+
+static void virtnet_rx_pause(struct virtnet_info *vi, struct receive_queue *rq)
+{
+	/*
+	 * Make sure refill_work does not run concurrently to
+	 * avoid napi_disable race which leads to deadlock.
+	 */
+	disable_delayed_refill(vi);
+	cancel_delayed_work_sync(&vi->refill);
+	__virtnet_rx_pause(vi, rq);
+}
+
+static void __virtnet_rx_resume(struct virtnet_info *vi,
+				struct receive_queue *rq,
+				bool refill)
 {
 	bool running = netif_running(vi->dev);
 
-	if (!try_fill_recv(vi, rq, GFP_KERNEL))
+	if (refill && !try_fill_recv(vi, rq, GFP_KERNEL))
 		schedule_delayed_work(&vi->refill, 0);
 
 	if (running)
 		virtnet_napi_enable(rq);
 }
 
+static void virtnet_rx_resume_all(struct virtnet_info *vi)
+{
+	int i;
+
+	enable_delayed_refill(vi);
+	for (i = 0; i < vi->max_queue_pairs; i++) {
+		if (i < vi->curr_queue_pairs)
+			__virtnet_rx_resume(vi, &vi->rq[i], true);
+		else
+			__virtnet_rx_resume(vi, &vi->rq[i], false);
+	}
+}
+
+static void virtnet_rx_resume(struct virtnet_info *vi, struct receive_queue *rq)
+{
+	enable_delayed_refill(vi);
+	__virtnet_rx_resume(vi, rq, true);
+}
+
 static int virtnet_rx_resize(struct virtnet_info *vi,
 			     struct receive_queue *rq, u32 ring_num)
 {
@@ -5959,12 +6006,12 @@ static int virtnet_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 	if (prog)
 		bpf_prog_add(prog, vi->max_queue_pairs - 1);
 
+	virtnet_rx_pause_all(vi);
+
 	/* Make sure NAPI is not using any XDP TX queues for RX. */
 	if (netif_running(dev)) {
-		for (i = 0; i < vi->max_queue_pairs; i++) {
-			virtnet_napi_disable(&vi->rq[i]);
+		for (i = 0; i < vi->max_queue_pairs; i++)
 			virtnet_napi_tx_disable(&vi->sq[i]);
-		}
 	}
 
 	if (!prog) {
@@ -5996,13 +6043,12 @@ static int virtnet_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 		vi->xdp_enabled = false;
 	}
 
+	virtnet_rx_resume_all(vi);
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		if (old_prog)
 			bpf_prog_put(old_prog);
-		if (netif_running(dev)) {
-			virtnet_napi_enable(&vi->rq[i]);
+		if (netif_running(dev))
 			virtnet_napi_tx_enable(&vi->sq[i]);
-		}
 	}
 
 	return 0;
@@ -6014,11 +6060,10 @@ err:
 			rcu_assign_pointer(vi->rq[i].xdp_prog, old_prog);
 	}
 
+	virtnet_rx_resume_all(vi);
 	if (netif_running(dev)) {
-		for (i = 0; i < vi->max_queue_pairs; i++) {
-			virtnet_napi_enable(&vi->rq[i]);
+		for (i = 0; i < vi->max_queue_pairs; i++)
 			virtnet_napi_tx_enable(&vi->sq[i]);
-		}
 	}
 	if (prog)
 		bpf_prog_sub(prog, vi->max_queue_pairs - 1);
diff --git a/drivers/net/vmxnet3/vmxnet3_xdp.c b/drivers/net/vmxnet3/vmxnet3_xdp.c
index 616ecc38d172..5f470499e600 100644
--- a/drivers/net/vmxnet3/vmxnet3_xdp.c
+++ b/drivers/net/vmxnet3/vmxnet3_xdp.c
@@ -397,7 +397,7 @@ vmxnet3_process_xdp(struct vmxnet3_adapter *adapter,
 
 	xdp_init_buff(&xdp, PAGE_SIZE, &rq->xdp_rxq);
 	xdp_prepare_buff(&xdp, page_address(page), rq->page_pool->p.offset,
-			 rbi->len, false);
+			 rcd->len, false);
 	xdp_buff_clear_frags_flag(&xdp);
 
 	xdp_prog = rcu_dereference(rq->adapter->xdp_bpf_prog);
diff --git a/drivers/net/vxlan/vxlan_vnifilter.c b/drivers/net/vxlan/vxlan_vnifilter.c
index 6e6e9f05509a..06d19e90eadb 100644
--- a/drivers/net/vxlan/vxlan_vnifilter.c
+++ b/drivers/net/vxlan/vxlan_vnifilter.c
@@ -627,7 +627,11 @@ static void vxlan_vni_delete_group(struct vxlan_dev *vxlan,
 	 * default dst remote_ip previously added for this vni
 	 */
 	if (!vxlan_addr_any(&vninode->remote_ip) ||
-	    !vxlan_addr_any(&dst->remote_ip))
+	    !vxlan_addr_any(&dst->remote_ip)) {
+		u32 hash_index = fdb_head_index(vxlan, all_zeros_mac,
+						vninode->vni);
+
+		spin_lock_bh(&vxlan->hash_lock[hash_index]);
 		__vxlan_fdb_delete(vxlan, all_zeros_mac,
 				   (vxlan_addr_any(&vninode->remote_ip) ?
 				   dst->remote_ip : vninode->remote_ip),
@@ -635,6 +639,8 @@ static void vxlan_vni_delete_group(struct vxlan_dev *vxlan,
 				   vninode->vni, vninode->vni,
 				   dst->remote_ifindex,
 				   true);
+		spin_unlock_bh(&vxlan->hash_lock[hash_index]);
+	}
 
 	if (vxlan->dev->flags & IFF_UP) {
 		if (vxlan_addr_multicast(&vninode->remote_ip) &&
diff --git a/drivers/net/wireless/ath/carl9170/fw.c b/drivers/net/wireless/ath/carl9170/fw.c
index 4c1aecd1163c..419f5530f885 100644
--- a/drivers/net/wireless/ath/carl9170/fw.c
+++ b/drivers/net/wireless/ath/carl9170/fw.c
@@ -15,7 +15,7 @@
 #include "fwcmd.h"
 #include "version.h"
 
-static const u8 otus_magic[4] = { OTUS_MAGIC };
+static const u8 otus_magic[4] __nonstring = { OTUS_MAGIC };
 
 static const void *carl9170_fw_find_desc(struct ar9170 *ar, const u8 descid[4],
 	const unsigned int len, const u8 compatible_revision)
diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c
index 4f01189b7c4b..6842c2b02b39 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.c
+++ b/drivers/net/wireless/atmel/at76c50x-usb.c
@@ -2552,7 +2552,7 @@ static void at76_disconnect(struct usb_interface *interface)
 
 	wiphy_info(priv->hw->wiphy, "disconnecting\n");
 	at76_delete_device(priv);
-	usb_put_dev(priv->udev);
+	usb_put_dev(interface_to_usbdev(interface));
 	dev_info(&interface->dev, "disconnected\n");
 }
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
index cfcf01eb0daa..f26e4679e4ff 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/common.c
@@ -561,8 +561,10 @@ struct brcmf_mp_device *brcmf_get_module_param(struct device *dev,
 	if (!found) {
 		/* No platform data for this device, try OF and DMI data */
 		brcmf_dmi_probe(settings, chip, chiprev);
-		if (brcmf_of_probe(dev, bus_type, settings) == -EPROBE_DEFER)
+		if (brcmf_of_probe(dev, bus_type, settings) == -EPROBE_DEFER) {
+			kfree(settings);
 			return ERR_PTR(-EPROBE_DEFER);
+		}
 		brcmf_acpi_probe(dev, bus_type, settings);
 	}
 	return settings;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
index 2821c27f317e..d06c724f63d9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
@@ -896,14 +896,16 @@ brcmf_usb_dl_writeimage(struct brcmf_usbdev_info *devinfo, u8 *fw, int fwlen)
 	}
 
 	/* 1) Prepare USB boot loader for runtime image */
-	brcmf_usb_dl_cmd(devinfo, DL_START, &state, sizeof(state));
+	err = brcmf_usb_dl_cmd(devinfo, DL_START, &state, sizeof(state));
+	if (err)
+		goto fail;
 
 	rdlstate = le32_to_cpu(state.state);
 	rdlbytes = le32_to_cpu(state.bytes);
 
 	/* 2) Check we are in the Waiting state */
 	if (rdlstate != DL_WAITING) {
-		brcmf_err("Failed to DL_START\n");
+		brcmf_err("Invalid DL state: %u\n", rdlstate);
 		err = -EINVAL;
 		goto fail;
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
index 670031fd60dc..59af36960f9c 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
@@ -142,8 +142,6 @@ const struct iwl_cfg_trans_params iwl_sc_trans_cfg = {
 	.ltr_delay = IWL_CFG_TRANS_LTR_DELAY_2500US,
 };
 
-const char iwl_sp_name[] = "Intel(R) Wi-Fi 7 BE213 160MHz";
-
 const struct iwl_cfg iwl_cfg_sc = {
 	.fw_name_mac = "sc",
 	IWL_DEVICE_SC,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index b9bd89bfdd74..acafee538b8a 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -2,7 +2,7 @@
 /*
  * Copyright (C) 2005-2014, 2018-2021 Intel Corporation
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2025 Intel Corporation
+ * Copyright (C) 2018-2024 Intel Corporation
  */
 #ifndef __IWL_CONFIG_H__
 #define __IWL_CONFIG_H__
@@ -451,8 +451,11 @@ struct iwl_cfg {
 #define IWL_CFG_RF_ID_HR		0x7
 #define IWL_CFG_RF_ID_HR1		0x4
 
-#define IWL_CFG_BW_NO_LIM		(U16_MAX - 1)
-#define IWL_CFG_BW_ANY			U16_MAX
+#define IWL_CFG_NO_160			0x1
+#define IWL_CFG_160			0x0
+
+#define IWL_CFG_NO_320			0x1
+#define IWL_CFG_320			0x0
 
 #define IWL_CFG_CORES_BT		0x0
 #define IWL_CFG_CORES_BT_GNSS		0x5
@@ -464,7 +467,7 @@ struct iwl_cfg {
 #define IWL_CFG_IS_JACKET		0x1
 
 #define IWL_SUBDEVICE_RF_ID(subdevice)	((u16)((subdevice) & 0x00F0) >> 4)
-#define IWL_SUBDEVICE_BW_LIM(subdevice)	((u16)((subdevice) & 0x0200) >> 9)
+#define IWL_SUBDEVICE_NO_160(subdevice)	((u16)((subdevice) & 0x0200) >> 9)
 #define IWL_SUBDEVICE_CORES(subdevice)	((u16)((subdevice) & 0x1C00) >> 10)
 
 struct iwl_dev_info {
@@ -472,10 +475,10 @@ struct iwl_dev_info {
 	u16 subdevice;
 	u16 mac_type;
 	u16 rf_type;
-	u16 bw_limit;
 	u8 mac_step;
 	u8 rf_step;
 	u8 rf_id;
+	u8 no_160;
 	u8 cores;
 	u8 cdb;
 	u8 jacket;
@@ -489,7 +492,7 @@ extern const unsigned int iwl_dev_info_table_size;
 const struct iwl_dev_info *
 iwl_pci_find_dev_info(u16 device, u16 subsystem_device,
 		      u16 mac_type, u8 mac_step, u16 rf_type, u8 cdb,
-		      u8 jacket, u8 rf_id, u8 bw_limit, u8 cores, u8 rf_step);
+		      u8 jacket, u8 rf_id, u8 no_160, u8 cores, u8 rf_step);
 extern const struct pci_device_id iwl_hw_card_ids[];
 #endif
 
@@ -550,7 +553,6 @@ extern const char iwl_ax231_name[];
 extern const char iwl_ax411_name[];
 extern const char iwl_fm_name[];
 extern const char iwl_wh_name[];
-extern const char iwl_sp_name[];
 extern const char iwl_gl_name[];
 extern const char iwl_mtp_name[];
 extern const char iwl_dr_name[];
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index be9e464c9b7b..3ff493e920d2 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -148,6 +148,7 @@
  * during a error FW error.
  */
 #define CSR_FUNC_SCRATCH_INIT_VALUE		(0x01010101)
+#define CSR_FUNC_SCRATCH_POWER_OFF_MASK		0xFFFF
 
 /* Bits for CSR_HW_IF_CONFIG_REG */
 #define CSR_HW_IF_CONFIG_REG_MSK_MAC_STEP_DASH	(0x0000000F)
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index cd1b0048bb6d..c381511e9ec6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2005-2014, 2018-2023, 2025 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2023 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -944,8 +944,7 @@ iwl_nvm_fixup_sband_iftd(struct iwl_trans *trans,
 				       IEEE80211_EHT_MAC_CAP0_MAX_MPDU_LEN_MASK);
 		break;
 	case NL80211_BAND_6GHZ:
-		if (!trans->reduced_cap_sku &&
-		    trans->bw_limit >= 320) {
+		if (!trans->reduced_cap_sku) {
 			iftype_data->eht_cap.eht_cap_elem.phy_cap_info[0] |=
 				IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ;
 			iftype_data->eht_cap.eht_cap_elem.phy_cap_info[1] |=
@@ -1095,22 +1094,19 @@ iwl_nvm_fixup_sband_iftd(struct iwl_trans *trans,
 		iftype_data->eht_cap.eht_mcs_nss_supp.bw._320.rx_tx_mcs13_max_nss = 0;
 	}
 
-	if (trans->bw_limit < 160)
+	if (trans->no_160)
 		iftype_data->he_cap.he_cap_elem.phy_cap_info[0] &=
 			~IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G;
 
-	if (trans->bw_limit < 320 || trans->reduced_cap_sku) {
+	if (trans->reduced_cap_sku) {
 		memset(&iftype_data->eht_cap.eht_mcs_nss_supp.bw._320, 0,
 		       sizeof(iftype_data->eht_cap.eht_mcs_nss_supp.bw._320));
-		iftype_data->eht_cap.eht_cap_elem.phy_cap_info[2] &=
-			~IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK;
-	}
-
-	if (trans->reduced_cap_sku) {
 		iftype_data->eht_cap.eht_mcs_nss_supp.bw._80.rx_tx_mcs13_max_nss = 0;
 		iftype_data->eht_cap.eht_mcs_nss_supp.bw._160.rx_tx_mcs13_max_nss = 0;
 		iftype_data->eht_cap.eht_cap_elem.phy_cap_info[8] &=
 			~IEEE80211_EHT_PHY_CAP8_RX_4096QAM_WIDER_BW_DL_OFDMA;
+		iftype_data->eht_cap.eht_cap_elem.phy_cap_info[2] &=
+			~IEEE80211_EHT_PHY_CAP2_SOUNDING_DIM_320MHZ_MASK;
 	}
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
index c1607b6d0759..e7b2e08645ef 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
@@ -21,6 +21,7 @@ struct iwl_trans_dev_restart_data {
 	struct list_head list;
 	unsigned int restart_count;
 	time64_t last_error;
+	bool backoff;
 	char name[];
 };
 
@@ -125,13 +126,20 @@ iwl_trans_determine_restart_mode(struct iwl_trans *trans)
 	if (!data)
 		return at_least;
 
-	if (ktime_get_boottime_seconds() - data->last_error >=
+	if (!data->backoff &&
+	    ktime_get_boottime_seconds() - data->last_error >=
 			IWL_TRANS_RESET_OK_TIME)
 		data->restart_count = 0;
 
 	index = data->restart_count;
-	if (index >= ARRAY_SIZE(escalation_list))
+	if (index >= ARRAY_SIZE(escalation_list)) {
 		index = ARRAY_SIZE(escalation_list) - 1;
+		if (!data->backoff) {
+			data->backoff = true;
+			return IWL_RESET_MODE_BACKOFF;
+		}
+		data->backoff = false;
+	}
 
 	return max(at_least, escalation_list[index]);
 }
@@ -140,7 +148,8 @@ iwl_trans_determine_restart_mode(struct iwl_trans *trans)
 
 static void iwl_trans_restart_wk(struct work_struct *wk)
 {
-	struct iwl_trans *trans = container_of(wk, typeof(*trans), restart.wk);
+	struct iwl_trans *trans = container_of(wk, typeof(*trans),
+					       restart.wk.work);
 	struct iwl_trans_reprobe *reprobe;
 	enum iwl_reset_mode mode;
 
@@ -168,6 +177,12 @@ static void iwl_trans_restart_wk(struct work_struct *wk)
 		return;
 
 	mode = iwl_trans_determine_restart_mode(trans);
+	if (mode == IWL_RESET_MODE_BACKOFF) {
+		IWL_ERR(trans, "Too many device errors - delay next reset\n");
+		queue_delayed_work(system_unbound_wq, &trans->restart.wk,
+				   IWL_TRANS_RESET_DELAY);
+		return;
+	}
 
 	iwl_trans_inc_restart_count(trans->dev);
 
@@ -227,7 +242,7 @@ struct iwl_trans *iwl_trans_alloc(unsigned int priv_size,
 	trans->dev = dev;
 	trans->num_rx_queues = 1;
 
-	INIT_WORK(&trans->restart.wk, iwl_trans_restart_wk);
+	INIT_DELAYED_WORK(&trans->restart.wk, iwl_trans_restart_wk);
 
 	return trans;
 }
@@ -271,7 +286,7 @@ int iwl_trans_init(struct iwl_trans *trans)
 
 void iwl_trans_free(struct iwl_trans *trans)
 {
-	cancel_work_sync(&trans->restart.wk);
+	cancel_delayed_work_sync(&trans->restart.wk);
 	kmem_cache_destroy(trans->dev_cmd_pool);
 }
 
@@ -403,7 +418,7 @@ void iwl_trans_op_mode_leave(struct iwl_trans *trans)
 
 	iwl_trans_pcie_op_mode_leave(trans);
 
-	cancel_work_sync(&trans->restart.wk);
+	cancel_delayed_work_sync(&trans->restart.wk);
 
 	trans->op_mode = NULL;
 
@@ -540,7 +555,6 @@ void __releases(nic_access)
 iwl_trans_release_nic_access(struct iwl_trans *trans)
 {
 	iwl_trans_pcie_release_nic_access(trans);
-	__release(nic_access);
 }
 IWL_EXPORT_SYMBOL(iwl_trans_release_nic_access);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 25fb4c50e38b..ce4954b0d524 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2018-2023, 2025 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2023 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -876,7 +876,7 @@ struct iwl_txq {
  *	only valid for discrete (not integrated) NICs
  * @invalid_tx_cmd: invalid TX command buffer
  * @reduced_cap_sku: reduced capability supported SKU
- * @bw_limit: the max bandwidth
+ * @no_160: device not supporting 160 MHz
  * @step_urm: STEP is in URM, no support for MCS>9 in 320 MHz
  * @restart: restart worker data
  * @restart.wk: restart worker
@@ -911,8 +911,7 @@ struct iwl_trans {
 	char hw_id_str[52];
 	u32 sku_id[3];
 	bool reduced_cap_sku;
-	u16 bw_limit;
-	bool step_urm;
+	u8 no_160:1, step_urm:1;
 
 	u8 dsbr_urm_fw_dependent:1,
 	   dsbr_urm_permanent:1;
@@ -962,7 +961,7 @@ struct iwl_trans {
 	struct iwl_dma_ptr invalid_tx_cmd;
 
 	struct {
-		struct work_struct wk;
+		struct delayed_work wk;
 		struct iwl_fw_error_dump_mode mode;
 		bool during_reset;
 	} restart;
@@ -1163,7 +1162,7 @@ static inline void iwl_trans_schedule_reset(struct iwl_trans *trans,
 	 */
 	trans->restart.during_reset = test_bit(STATUS_IN_SW_RESET,
 					       &trans->status);
-	queue_work(system_unbound_wq, &trans->restart.wk);
+	queue_delayed_work(system_unbound_wq, &trans->restart.wk, 0);
 }
 
 static inline void iwl_trans_fw_error(struct iwl_trans *trans,
@@ -1262,6 +1261,9 @@ enum iwl_reset_mode {
 	IWL_RESET_MODE_RESCAN,
 	IWL_RESET_MODE_FUNC_RESET,
 	IWL_RESET_MODE_PROD_RESET,
+
+	/* keep last - special backoff value */
+	IWL_RESET_MODE_BACKOFF,
 };
 
 void iwl_trans_pcie_reset(struct iwl_trans *trans, enum iwl_reset_mode mode);
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/agg.c b/drivers/net/wireless/intel/iwlwifi/mld/agg.c
index db9e0f04f4b7..687a9450ac98 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/agg.c
+++ b/drivers/net/wireless/intel/iwlwifi/mld/agg.c
@@ -124,9 +124,9 @@ void iwl_mld_handle_bar_frame_release_notif(struct iwl_mld *mld,
 
 	rcu_read_lock();
 	baid_data = rcu_dereference(mld->fw_id_to_ba[baid]);
-	if (!IWL_FW_CHECK(mld, !baid_data,
-			  "Got valid BAID %d but not allocated, invalid BAR release!\n",
-			  baid))
+	if (IWL_FW_CHECK(mld, !baid_data,
+			 "Got valid BAID %d but not allocated, invalid BAR release!\n",
+			 baid))
 		goto out_unlock;
 
 	if (IWL_FW_CHECK(mld, tid != baid_data->tid ||
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/d3.c b/drivers/net/wireless/intel/iwlwifi/mld/d3.c
index 5a7207accd86..ee99298eebf5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mld/d3.c
@@ -1099,7 +1099,8 @@ iwl_mld_set_netdetect_info(struct iwl_mld *mld,
 		if (!match)
 			return;
 
-		netdetect_info->matches[netdetect_info->n_matches++] = match;
+		netdetect_info->matches[netdetect_info->n_matches] = match;
+		netdetect_info->n_matches++;
 
 		/* We inverted the order of the SSIDs in the scan
 		 * request, so invert the index here.
@@ -1116,9 +1117,11 @@ iwl_mld_set_netdetect_info(struct iwl_mld *mld,
 
 		for_each_set_bit(j,
 				 (unsigned long *)&matches[i].matching_channels[0],
-				 sizeof(matches[i].matching_channels))
-			match->channels[match->n_channels++] =
+				 sizeof(matches[i].matching_channels)) {
+			match->channels[match->n_channels] =
 				netdetect_cfg->channels[j]->center_freq;
+			match->n_channels++;
+		}
 	}
 }
 
@@ -1895,7 +1898,6 @@ int iwl_mld_wowlan_resume(struct iwl_mld *mld)
 	int link_id;
 	int ret;
 	bool fw_err = false;
-	bool keep_connection;
 
 	lockdep_assert_wiphy(mld->wiphy);
 
@@ -1965,7 +1967,7 @@ int iwl_mld_wowlan_resume(struct iwl_mld *mld)
 		iwl_mld_process_netdetect_res(mld, bss_vif, &resume_data);
 		mld->netdetect = false;
 	} else {
-		keep_connection =
+		bool keep_connection =
 			iwl_mld_process_wowlan_status(mld, bss_vif,
 						      resume_data.wowlan_status);
 
@@ -1973,11 +1975,10 @@ int iwl_mld_wowlan_resume(struct iwl_mld *mld)
 		if (keep_connection)
 			iwl_mld_unblock_emlsr(mld, bss_vif,
 					      IWL_MLD_EMLSR_BLOCKED_WOWLAN);
+		else
+			ieee80211_resume_disconnect(bss_vif);
 	}
 
-	if (!mld->netdetect && !keep_connection)
-		ieee80211_resume_disconnect(bss_vif);
-
 	goto out;
 
  err:
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mld/debugfs.c
index 453ce2ba39d1..93f9f78e4276 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mld/debugfs.c
@@ -396,8 +396,8 @@ static ssize_t iwl_dbgfs_tas_get_status_read(struct iwl_mld *mld, char *buf,
 		.data[0] = &cmd,
 	};
 	struct iwl_dhc_tas_status_resp *resp = NULL;
+	u32 resp_len = 0;
 	ssize_t pos = 0;
-	u32 resp_len;
 	u32 status;
 	int ret;
 
@@ -949,8 +949,9 @@ void iwl_mld_add_vif_debugfs(struct ieee80211_hw *hw,
 	snprintf(name, sizeof(name), "%pd", vif->debugfs_dir);
 	snprintf(target, sizeof(target), "../../../%pd3/iwlmld",
 		 vif->debugfs_dir);
-	mld_vif->dbgfs_slink =
-		debugfs_create_symlink(name, mld->debugfs_dir, target);
+	if (!mld_vif->dbgfs_slink)
+		mld_vif->dbgfs_slink =
+			debugfs_create_symlink(name, mld->debugfs_dir, target);
 
 	if (iwlmld_mod_params.power_scheme != IWL_POWER_SCHEME_CAM &&
 	    vif->type == NL80211_IFTYPE_STATION) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/fw.c b/drivers/net/wireless/intel/iwlwifi/mld/fw.c
index 62da137e1024..4b083d447ee2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mld/fw.c
@@ -333,19 +333,22 @@ int iwl_mld_load_fw(struct iwl_mld *mld)
 
 	ret = iwl_trans_start_hw(mld->trans);
 	if (ret)
-		return ret;
+		goto err;
 
 	ret = iwl_mld_run_fw_init_sequence(mld);
 	if (ret)
-		return ret;
+		goto err;
 
 	ret = iwl_mld_init_mcc(mld);
 	if (ret)
-		return ret;
+		goto err;
 
 	mld->fw_status.running = true;
 
 	return 0;
+err:
+	iwl_mld_stop_fw(mld);
+	return ret;
 }
 
 void iwl_mld_stop_fw(struct iwl_mld *mld)
@@ -358,6 +361,10 @@ void iwl_mld_stop_fw(struct iwl_mld *mld)
 
 	iwl_trans_stop_device(mld->trans);
 
+	wiphy_work_cancel(mld->wiphy, &mld->async_handlers_wk);
+
+	iwl_mld_purge_async_handlers_list(mld);
+
 	mld->fw_status.running = false;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/iface.h b/drivers/net/wireless/intel/iwlwifi/mld/iface.h
index d1d56b081bf6..ec14d0736cee 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/iface.h
+++ b/drivers/net/wireless/intel/iwlwifi/mld/iface.h
@@ -166,7 +166,7 @@ struct iwl_mld_vif {
 
 	struct iwl_mld_emlsr emlsr;
 
-#if CONFIG_PM_SLEEP
+#ifdef CONFIG_PM_SLEEP
 	struct iwl_mld_wowlan_data wowlan_data;
 #endif
 #ifdef CONFIG_IWLWIFI_DEBUGFS
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c
index 6851064b82da..68d97d3b8f02 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c
@@ -475,8 +475,8 @@ static
 int iwl_mld_mac80211_start(struct ieee80211_hw *hw)
 {
 	struct iwl_mld *mld = IWL_MAC80211_GET_MLD(hw);
-	int ret;
 	bool in_d3 = false;
+	int ret = 0;
 
 	lockdep_assert_wiphy(mld->wiphy);
 
@@ -537,7 +537,8 @@ void iwl_mld_mac80211_stop(struct ieee80211_hw *hw, bool suspend)
 	/* if the suspend flow fails the fw is in error. Stop it here, and it
 	 * will be started upon wakeup
 	 */
-	if (!suspend || iwl_mld_no_wowlan_suspend(mld))
+	if (!suspend ||
+	    (IS_ENABLED(CONFIG_PM_SLEEP) && iwl_mld_no_wowlan_suspend(mld)))
 		iwl_mld_stop_fw(mld);
 
 	/* HW is stopped, no more coming RX. OTOH, the worker can't run as the
@@ -650,6 +651,7 @@ void iwl_mld_mac80211_remove_interface(struct ieee80211_hw *hw,
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	debugfs_remove(iwl_mld_vif_from_mac80211(vif)->dbgfs_slink);
+	iwl_mld_vif_from_mac80211(vif)->dbgfs_slink = NULL;
 #endif
 
 	iwl_mld_rm_vif(mld, vif);
@@ -1943,6 +1945,7 @@ static void iwl_mld_sta_rc_update(struct ieee80211_hw *hw,
 	}
 }
 
+#ifdef CONFIG_PM_SLEEP
 static void iwl_mld_set_wakeup(struct ieee80211_hw *hw, bool enabled)
 {
 	struct iwl_mld *mld = IWL_MAC80211_GET_MLD(hw);
@@ -1994,6 +1997,7 @@ static int iwl_mld_resume(struct ieee80211_hw *hw)
 
 	return 0;
 }
+#endif
 
 static int iwl_mld_alloc_ptk_pn(struct iwl_mld *mld,
 				struct iwl_mld_sta *mld_sta,
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mld.c b/drivers/net/wireless/intel/iwlwifi/mld/mld.c
index d4a99ae64074..73d2166a4c25 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/mld.c
+++ b/drivers/net/wireless/intel/iwlwifi/mld/mld.c
@@ -75,6 +75,7 @@ void iwl_construct_mld(struct iwl_mld *mld, struct iwl_trans *trans,
 
 	/* Setup async RX handling */
 	spin_lock_init(&mld->async_handlers_lock);
+	INIT_LIST_HEAD(&mld->async_handlers_list);
 	wiphy_work_init(&mld->async_handlers_wk,
 			iwl_mld_async_handlers_wk);
 
@@ -414,9 +415,14 @@ iwl_op_mode_mld_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		wiphy_unlock(mld->wiphy);
 		rtnl_unlock();
 		iwl_fw_flush_dumps(&mld->fwrt);
-		goto free_hw;
+		goto err;
 	}
 
+	/* We are about to stop the FW. Notifications may require an
+	 * operational FW, so handle them all here before we stop.
+	 */
+	wiphy_work_flush(mld->wiphy, &mld->async_handlers_wk);
+
 	iwl_mld_stop_fw(mld);
 
 	wiphy_unlock(mld->wiphy);
@@ -455,7 +461,8 @@ leds_exit:
 	iwl_mld_leds_exit(mld);
 free_nvm:
 	kfree(mld->nvm_data);
-free_hw:
+err:
+	iwl_trans_op_mode_leave(mld->trans);
 	ieee80211_free_hw(mld->hw);
 	return ERR_PTR(ret);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mld.h b/drivers/net/wireless/intel/iwlwifi/mld/mld.h
index 5eceaaf7696d..a4a16da6ebf3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mld/mld.h
+++ b/drivers/net/wireless/intel/iwlwifi/mld/mld.h
@@ -298,11 +298,6 @@ iwl_cleanup_mld(struct iwl_mld *mld)
 #endif
 
 	iwl_mld_low_latency_restart_cleanup(mld);
-
-	/* Empty the list of async notification handlers so we won't process
-	 * notifications from the dead fw after the reconfig flow.
-	 */
-	iwl_mld_purge_async_handlers_list(mld);
 }
 
 enum iwl_power_scheme {
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 93446c374008..debeea2b3ae5 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2005-2014, 2018-2025 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -552,17 +552,16 @@ MODULE_DEVICE_TABLE(pci, iwl_hw_card_ids);
 EXPORT_SYMBOL_IF_IWLWIFI_KUNIT(iwl_hw_card_ids);
 
 #define _IWL_DEV_INFO(_device, _subdevice, _mac_type, _mac_step, _rf_type, \
-		      _rf_id, _rf_step, _bw_limit, _cores, _cdb, _cfg, _name) \
+		      _rf_id, _rf_step, _no_160, _cores, _cdb, _cfg, _name) \
 	{ .device = (_device), .subdevice = (_subdevice), .cfg = &(_cfg), \
 	  .name = _name, .mac_type = _mac_type, .rf_type = _rf_type, .rf_step = _rf_step, \
-	  .bw_limit = _bw_limit, .cores = _cores, .rf_id = _rf_id, \
+	  .no_160 = _no_160, .cores = _cores, .rf_id = _rf_id, \
 	  .mac_step = _mac_step, .cdb = _cdb, .jacket = IWL_CFG_ANY }
 
 #define IWL_DEV_INFO(_device, _subdevice, _cfg, _name) \
 	_IWL_DEV_INFO(_device, _subdevice, IWL_CFG_ANY, IWL_CFG_ANY,   \
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, \
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_ANY, \
-		      _cfg, _name)
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,  \
+		      IWL_CFG_ANY, _cfg, _name)
 
 VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 #if IS_ENABLED(CONFIG_IWLMVM)
@@ -725,66 +724,66 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_2ac_cfg_soc, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_2ac_cfg_soc, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_2ac_cfg_soc, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_2ac_cfg_soc, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_2ac_cfg_soc, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_PU, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_2ac_cfg_soc, iwl9560_name),
 
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT_GNSS, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT_GNSS, IWL_CFG_NO_CDB,
 		      iwl9260_2ac_cfg, iwl9270_160_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT_GNSS, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT_GNSS, IWL_CFG_NO_CDB,
 		      iwl9260_2ac_cfg, iwl9270_name),
 
 	_IWL_DEV_INFO(0x271B, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_TH1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9260_2ac_cfg, iwl9162_160_name),
 	_IWL_DEV_INFO(0x271B, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_TH1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9260_2ac_cfg, iwl9162_name),
 
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9260_2ac_cfg, iwl9260_160_name),
 	_IWL_DEV_INFO(0x2526, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_TH, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_TH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9260_2ac_cfg, iwl9260_name),
 
 /* Qu with Jf */
@@ -792,132 +791,132 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_b0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
 	/* Qu C step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_qu_c0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
 	/* QuZ */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9462_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1551,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_killer_1550s_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, 0x1552,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwl9560_quz_a0_jf_b0_cfg, iwl9560_killer_1550i_name),
 
 /* Qu with Hr */
@@ -925,189 +924,189 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_qu_b0_hr1_b0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_qu_b0_hr_b0, iwl_ax203_name),
 
 	/* Qu C step */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_qu_c0_hr1_b0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_qu_c0_hr_b0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QU, SILICON_C_STEP,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_qu_c0_hr_b0, iwl_ax201_name),
 
 	/* QuZ */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_quz_a0_hr1_b0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_quz_a0_hr_b0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_QUZ, SILICON_B_STEP,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_quz_a0_hr_b0, iwl_ax201_name),
 
 /* Ma */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_ma, iwl_ax201_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_ma, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_MA, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_ma, iwl_ax231_name),
 
 /* So with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax201_name),
 
 /* So-F with Hr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax203_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR1, IWL_CFG_ANY, IWL_CFG_ANY,
-		      80, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax101_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_so_a0_hr_a0, iwl_ax201_name),
 
 /* So-F with Gf */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwlax211_2ax_cfg_so_gf_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_CDB,
 		      iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_name),
 
 /* SoF with JF2 */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_name),
 
 /* SoF with JF */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_name),
 
 /* So with GF */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwlax211_2ax_cfg_so_gf_a0, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_CDB,
+		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_CDB,
 		      iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_name),
 
 /* So with JF2 */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF2, IWL_CFG_RF_ID_JF, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9560_name),
 
 /* So with JF */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_160_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9461_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SO, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_JF1, IWL_CFG_RF_ID_JF1_DIV, IWL_CFG_ANY,
-		      80, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_name),
 
 #endif /* CONFIG_IWLMVM */
@@ -1116,13 +1115,13 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_ax201_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_ax211_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
@@ -1134,119 +1133,104 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_wh_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_HR2, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_ax201_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_ax211_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_fm_name),
 
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_bz, iwl_wh_name),
 
 /* Ga (Gl) */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_320, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_gl, iwl_gl_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      160, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      IWL_CFG_NO_320, IWL_CFG_ANY, IWL_CFG_NO_CDB,
 		      iwl_cfg_gl, iwl_mtp_name),
 
 /* Sc */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc, iwl_fm_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc, iwl_wh_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SC, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      160, IWL_CFG_ANY, IWL_CFG_ANY,
-		      iwl_cfg_sc, iwl_sp_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC2, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc2, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC2, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc2, iwl_fm_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC2, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc2, iwl_wh_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SC2, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      160, IWL_CFG_ANY, IWL_CFG_ANY,
-		      iwl_cfg_sc2, iwl_sp_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC2F, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_GF, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc2f, iwl_ax211_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC2F, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc2f, iwl_fm_name),
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC2F, IWL_CFG_ANY,
 		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_NO_LIM, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_sc2f, iwl_wh_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_SC2F, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_WH, IWL_CFG_ANY, IWL_CFG_ANY,
-		      160, IWL_CFG_ANY, IWL_CFG_ANY,
-		      iwl_cfg_sc2f, iwl_sp_name),
 
 /* Dr */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_DR, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_dr, iwl_dr_name),
 
 /* Br */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_BR, IWL_CFG_ANY,
 		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_BW_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
 		      iwl_cfg_br, iwl_br_name),
 #endif /* CONFIG_IWLMLD */
 };
@@ -1398,7 +1382,7 @@ out:
 VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info *
 iwl_pci_find_dev_info(u16 device, u16 subsystem_device,
 		      u16 mac_type, u8 mac_step, u16 rf_type, u8 cdb,
-		      u8 jacket, u8 rf_id, u8 bw_limit, u8 cores, u8 rf_step)
+		      u8 jacket, u8 rf_id, u8 no_160, u8 cores, u8 rf_step)
 {
 	int num_devices = ARRAY_SIZE(iwl_dev_info_table);
 	int i;
@@ -1441,15 +1425,8 @@ iwl_pci_find_dev_info(u16 device, u16 subsystem_device,
 		    dev_info->rf_id != rf_id)
 			continue;
 
-		/*
-		 * Check that bw_limit have the same "boolean" value since
-		 * IWL_SUBDEVICE_BW_LIM can only return a boolean value and
-		 * dev_info->bw_limit encodes a non-boolean value.
-		 * dev_info->bw_limit == IWL_CFG_BW_NO_LIM must be equal to
-		 * !bw_limit to have a match.
-		 */
-		if (dev_info->bw_limit != IWL_CFG_BW_ANY &&
-		    (dev_info->bw_limit == IWL_CFG_BW_NO_LIM) == !!bw_limit)
+		if (dev_info->no_160 != (u8)IWL_CFG_ANY &&
+		    dev_info->no_160 != no_160)
 			continue;
 
 		if (dev_info->cores != (u8)IWL_CFG_ANY &&
@@ -1587,13 +1564,13 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 					 CSR_HW_RFID_IS_CDB(iwl_trans->hw_rf_id),
 					 CSR_HW_RFID_IS_JACKET(iwl_trans->hw_rf_id),
 					 IWL_SUBDEVICE_RF_ID(pdev->subsystem_device),
-					 IWL_SUBDEVICE_BW_LIM(pdev->subsystem_device),
+					 IWL_SUBDEVICE_NO_160(pdev->subsystem_device),
 					 IWL_SUBDEVICE_CORES(pdev->subsystem_device),
 					 CSR_HW_RFID_STEP(iwl_trans->hw_rf_id));
 	if (dev_info) {
 		iwl_trans->cfg = dev_info->cfg;
 		iwl_trans->name = dev_info->name;
-		iwl_trans->bw_limit = dev_info->bw_limit;
+		iwl_trans->no_160 = dev_info->no_160 == IWL_CFG_NO_160;
 	}
 
 #if IS_ENABLED(CONFIG_IWLMVM)
@@ -1759,11 +1736,27 @@ static int _iwl_pci_resume(struct device *device, bool restore)
 	 * Scratch value was altered, this means the device was powered off, we
 	 * need to reset it completely.
 	 * Note: MAC (bits 0:7) will be cleared upon suspend even with wowlan,
-	 * so assume that any bits there mean that the device is usable.
+	 * but not bits [15:8]. So if we have bits set in lower word, assume
+	 * the device is alive.
+	 * For older devices, just try silently to grab the NIC.
 	 */
-	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ &&
-	    !iwl_read32(trans, CSR_FUNC_SCRATCH))
-		device_was_powered_off = true;
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) {
+		if (!(iwl_read32(trans, CSR_FUNC_SCRATCH) &
+		      CSR_FUNC_SCRATCH_POWER_OFF_MASK))
+			device_was_powered_off = true;
+	} else {
+		/*
+		 * bh are re-enabled by iwl_trans_pcie_release_nic_access,
+		 * so re-enable them if _iwl_trans_pcie_grab_nic_access fails.
+		 */
+		local_bh_disable();
+		if (_iwl_trans_pcie_grab_nic_access(trans, true)) {
+			iwl_trans_pcie_release_nic_access(trans);
+		} else {
+			device_was_powered_off = true;
+			local_bh_enable();
+		}
+	}
 
 	if (restore || device_was_powered_off) {
 		trans->state = IWL_TRANS_NO_FW;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 45460f93d24a..114a9195ad7f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -558,10 +558,10 @@ void iwl_trans_pcie_free(struct iwl_trans *trans);
 void iwl_trans_pcie_free_pnvm_dram_regions(struct iwl_dram_regions *dram_regions,
 					   struct device *dev);
 
-bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans);
-#define _iwl_trans_pcie_grab_nic_access(trans)			\
+bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans, bool silent);
+#define _iwl_trans_pcie_grab_nic_access(trans, silent)		\
 	__cond_lock(nic_access_nobh,				\
-		    likely(__iwl_trans_pcie_grab_nic_access(trans)))
+		    likely(__iwl_trans_pcie_grab_nic_access(trans, silent)))
 
 void iwl_trans_pcie_check_product_reset_status(struct pci_dev *pdev);
 void iwl_trans_pcie_check_product_reset_mode(struct pci_dev *pdev);
@@ -1105,7 +1105,8 @@ void iwl_trans_pcie_set_bits_mask(struct iwl_trans *trans, u32 reg,
 int iwl_trans_pcie_read_config32(struct iwl_trans *trans, u32 ofs,
 				 u32 *val);
 bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans);
-void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans);
+void __releases(nic_access_nobh)
+iwl_trans_pcie_release_nic_access(struct iwl_trans *trans);
 
 /* transport gen 1 exported functions */
 void iwl_trans_pcie_fw_alive(struct iwl_trans *trans, u32 scd_addr);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index 3ece34e30d58..472f26f83ba8 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -147,8 +147,14 @@ static void _iwl_trans_pcie_gen2_stop_device(struct iwl_trans *trans)
 		return;
 
 	if (trans->state >= IWL_TRANS_FW_STARTED &&
-	    trans_pcie->fw_reset_handshake)
+	    trans_pcie->fw_reset_handshake) {
+		/*
+		 * Reset handshake can dump firmware on timeout, but that
+		 * should assume that the firmware is already dead.
+		 */
+		trans->state = IWL_TRANS_NO_FW;
 		iwl_trans_pcie_fw_reset_handshake(trans);
+	}
 
 	trans_pcie->is_down = true;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index c917ed4c19bc..102a6123bba0 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -2351,7 +2351,8 @@ void iwl_trans_pcie_reset(struct iwl_trans *trans, enum iwl_reset_mode mode)
 	struct iwl_trans_pcie_removal *removal;
 	char _msg = 0, *msg = &_msg;
 
-	if (WARN_ON(mode < IWL_RESET_MODE_REMOVE_ONLY))
+	if (WARN_ON(mode < IWL_RESET_MODE_REMOVE_ONLY ||
+		    mode == IWL_RESET_MODE_BACKOFF))
 		return;
 
 	if (test_bit(STATUS_TRANS_DEAD, &trans->status))
@@ -2405,7 +2406,7 @@ EXPORT_SYMBOL(iwl_trans_pcie_reset);
  * This version doesn't disable BHs but rather assumes they're
  * already disabled.
  */
-bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
+bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans, bool silent)
 {
 	int ret;
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
@@ -2457,6 +2458,11 @@ bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
 	if (unlikely(ret < 0)) {
 		u32 cntrl = iwl_read32(trans, CSR_GP_CNTRL);
 
+		if (silent) {
+			spin_unlock(&trans_pcie->reg_lock);
+			return false;
+		}
+
 		WARN_ONCE(1,
 			  "Timeout waiting for hardware access (CSR_GP_CNTRL 0x%08x)\n",
 			  cntrl);
@@ -2488,7 +2494,7 @@ bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
 	bool ret;
 
 	local_bh_disable();
-	ret = __iwl_trans_pcie_grab_nic_access(trans);
+	ret = __iwl_trans_pcie_grab_nic_access(trans, false);
 	if (ret) {
 		/* keep BHs disabled until iwl_trans_pcie_release_nic_access */
 		return ret;
@@ -2497,7 +2503,8 @@ bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
 	return false;
 }
 
-void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
+void __releases(nic_access_nobh)
+iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
@@ -2524,6 +2531,7 @@ void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
 	 * scheduled on different CPUs (after we drop reg_lock).
 	 */
 out:
+	__release(nic_access_nobh);
 	spin_unlock_bh(&trans_pcie->reg_lock);
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index bb90bcfc6763..9fc4e98693eb 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -1021,7 +1021,7 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
 	 * returned. This needs to be done only on NICs that have
 	 * apmg_wake_up_wa set (see above.)
 	 */
-	if (!_iwl_trans_pcie_grab_nic_access(trans))
+	if (!_iwl_trans_pcie_grab_nic_access(trans, false))
 		return -EIO;
 
 	/*
diff --git a/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c b/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c
index 7ef5e89c6af2..d0bda23c628a 100644
--- a/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c
+++ b/drivers/net/wireless/intel/iwlwifi/tests/devinfo.c
@@ -2,7 +2,7 @@
 /*
  * KUnit tests for the iwlwifi device info table
  *
- * Copyright (C) 2023-2025 Intel Corporation
+ * Copyright (C) 2023-2024 Intel Corporation
  */
 #include <kunit/test.h>
 #include <linux/pci.h>
@@ -13,9 +13,9 @@ MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");
 
 static void iwl_pci_print_dev_info(const char *pfx, const struct iwl_dev_info *di)
 {
-	printk(KERN_DEBUG "%sdev=%.4x,subdev=%.4x,mac_type=%.4x,mac_step=%.4x,rf_type=%.4x,cdb=%d,jacket=%d,rf_id=%.2x,bw_limit=%d,cores=%.2x\n",
+	printk(KERN_DEBUG "%sdev=%.4x,subdev=%.4x,mac_type=%.4x,mac_step=%.4x,rf_type=%.4x,cdb=%d,jacket=%d,rf_id=%.2x,no_160=%d,cores=%.2x\n",
 	       pfx, di->device, di->subdevice, di->mac_type, di->mac_step,
-	       di->rf_type, di->cdb, di->jacket, di->rf_id, di->bw_limit,
+	       di->rf_type, di->cdb, di->jacket, di->rf_id, di->no_160,
 	       di->cores);
 }
 
@@ -31,13 +31,8 @@ static void devinfo_table_order(struct kunit *test)
 					    di->mac_type, di->mac_step,
 					    di->rf_type, di->cdb,
 					    di->jacket, di->rf_id,
-					    di->bw_limit != IWL_CFG_BW_NO_LIM,
-					    di->cores, di->rf_step);
-		if (!ret) {
-			iwl_pci_print_dev_info("No entry found for: ", di);
-			KUNIT_FAIL(test,
-				   "No entry found for entry at index %d\n", idx);
-		} else if (ret != di) {
+					    di->no_160, di->cores, di->rf_step);
+		if (ret != di) {
 			iwl_pci_print_dev_info("searched: ", di);
 			iwl_pci_print_dev_info("found:    ", ret);
 			KUNIT_FAIL(test,
diff --git a/drivers/net/wireless/purelifi/plfxlc/mac.c b/drivers/net/wireless/purelifi/plfxlc/mac.c
index eae93efa6150..82d1bf7edba2 100644
--- a/drivers/net/wireless/purelifi/plfxlc/mac.c
+++ b/drivers/net/wireless/purelifi/plfxlc/mac.c
@@ -102,7 +102,6 @@ int plfxlc_mac_init_hw(struct ieee80211_hw *hw)
 void plfxlc_mac_release(struct plfxlc_mac *mac)
 {
 	plfxlc_chip_release(&mac->chip);
-	lockdep_assert_held(&mac->lock);
 }
 
 int plfxlc_op_start(struct ieee80211_hw *hw)
diff --git a/drivers/net/wireless/ti/wl1251/tx.c b/drivers/net/wireless/ti/wl1251/tx.c
index 474b603c121c..adb4840b0489 100644
--- a/drivers/net/wireless/ti/wl1251/tx.c
+++ b/drivers/net/wireless/ti/wl1251/tx.c
@@ -342,8 +342,10 @@ void wl1251_tx_work(struct work_struct *work)
 	while ((skb = skb_dequeue(&wl->tx_queue))) {
 		if (!woken_up) {
 			ret = wl1251_ps_elp_wakeup(wl);
-			if (ret < 0)
+			if (ret < 0) {
+				skb_queue_head(&wl->tx_queue, skb);
 				goto out;
+			}
 			woken_up = true;
 		}
 
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index fc52d5c4c69b..5091e1fa4a0d 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -985,20 +985,27 @@ static u32 xennet_run_xdp(struct netfront_queue *queue, struct page *pdata,
 	act = bpf_prog_run_xdp(prog, xdp);
 	switch (act) {
 	case XDP_TX:
-		get_page(pdata);
 		xdpf = xdp_convert_buff_to_frame(xdp);
+		if (unlikely(!xdpf)) {
+			trace_xdp_exception(queue->info->netdev, prog, act);
+			break;
+		}
+		get_page(pdata);
 		err = xennet_xdp_xmit(queue->info->netdev, 1, &xdpf, 0);
-		if (unlikely(!err))
+		if (unlikely(err <= 0)) {
+			if (err < 0)
+				trace_xdp_exception(queue->info->netdev, prog, act);
 			xdp_return_frame_rx_napi(xdpf);
-		else if (unlikely(err < 0))
-			trace_xdp_exception(queue->info->netdev, prog, act);
+		}
 		break;
 	case XDP_REDIRECT:
 		get_page(pdata);
 		err = xdp_do_redirect(queue->info->netdev, xdp, prog);
 		*need_xdp_flush = true;
-		if (unlikely(err))
+		if (unlikely(err)) {
 			trace_xdp_exception(queue->info->netdev, prog, act);
+			xdp_return_buff(xdp);
+		}
 		break;
 	case XDP_PASS:
 	case XDP_DROP:
diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig
index d47dfa80fb95..4d64b6935bb9 100644
--- a/drivers/nvme/host/Kconfig
+++ b/drivers/nvme/host/Kconfig
@@ -102,6 +102,7 @@ config NVME_TCP_TLS
 	depends on NVME_TCP
 	select NET_HANDSHAKE
 	select KEYS
+	select TLS
 	help
 	  Enables TLS encryption for NVMe TCP using the netlink handshake API.
 
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index cc23035148b4..eb6ea8acb3cc 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -4295,6 +4295,15 @@ static void nvme_scan_work(struct work_struct *work)
 			nvme_scan_ns_sequential(ctrl);
 	}
 	mutex_unlock(&ctrl->scan_lock);
+
+	/* Requeue if we have missed AENs */
+	if (test_bit(NVME_AER_NOTICE_NS_CHANGED, &ctrl->events))
+		nvme_queue_scan(ctrl);
+#ifdef CONFIG_NVME_MULTIPATH
+	else if (ctrl->ana_log_buf)
+		/* Re-read the ANA log page to not miss updates */
+		queue_work(nvme_wq, &ctrl->ana_work);
+#endif
 }
 
 /*
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 89be5911b25d..250f3da67cc9 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -427,7 +427,7 @@ static bool nvme_available_path(struct nvme_ns_head *head)
 	struct nvme_ns *ns;
 
 	if (!test_bit(NVME_NSHEAD_DISK_LIVE, &head->flags))
-		return NULL;
+		return false;
 
 	list_for_each_entry_srcu(ns, &head->list, siblings,
 				 srcu_read_lock_held(&head->srcu)) {
@@ -1051,6 +1051,13 @@ void nvme_mpath_add_sysfs_link(struct nvme_ns_head *head)
 
 	list_for_each_entry_rcu(ns, &head->list, siblings) {
 		/*
+		 * Ensure that ns path disk node is already added otherwise we
+		 * may get invalid kobj name for target
+		 */
+		if (!test_bit(GD_ADDED, &ns->disk->state))
+			continue;
+
+		/*
 		 * Avoid creating link if it already exists for the given path.
 		 * When path ana state transitions from optimized to non-
 		 * optimized or vice-versa, the nvme_mpath_set_live() is
@@ -1065,13 +1072,6 @@ void nvme_mpath_add_sysfs_link(struct nvme_ns_head *head)
 		if (test_and_set_bit(NVME_NS_SYSFS_ATTR_LINK, &ns->flags))
 			continue;
 
-		/*
-		 * Ensure that ns path disk node is already added otherwise we
-		 * may get invalid kobj name for target
-		 */
-		if (!test_bit(GD_ADDED, &ns->disk->state))
-			continue;
-
 		target = disk_to_dev(ns->disk);
 		/*
 		 * Create sysfs link from head gendisk kobject @kobj to the
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index b178d52eac1b..2e30e9be7408 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -3575,7 +3575,7 @@ static pci_ers_result_t nvme_slot_reset(struct pci_dev *pdev)
 
 	dev_info(dev->ctrl.device, "restart after slot reset\n");
 	pci_restore_state(pdev);
-	if (!nvme_try_sched_reset(&dev->ctrl))
+	if (nvme_try_sched_reset(&dev->ctrl))
 		nvme_unquiesce_io_queues(&dev->ctrl);
 	return PCI_ERS_RESULT_RECOVERED;
 }
@@ -3623,6 +3623,9 @@ static const struct pci_device_id nvme_id_table[] = {
 		.driver_data = NVME_QUIRK_BOGUS_NID, },
 	{ PCI_DEVICE(0x1217, 0x8760), /* O2 Micro 64GB Steam Deck */
 		.driver_data = NVME_QUIRK_DMAPOOL_ALIGN_512, },
+	{ PCI_DEVICE(0x126f, 0x1001),	/* Silicon Motion generic */
+		.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
+				NVME_QUIRK_IGNORE_DEV_SUBNQN, },
 	{ PCI_DEVICE(0x126f, 0x2262),	/* Silicon Motion generic */
 		.driver_data = NVME_QUIRK_NO_DEEPEST_PS |
 				NVME_QUIRK_BOGUS_NID, },
@@ -3646,6 +3649,9 @@ static const struct pci_device_id nvme_id_table[] = {
 				NVME_QUIRK_IGNORE_DEV_SUBNQN, },
 	{ PCI_DEVICE(0x15b7, 0x5008),   /* Sandisk SN530 */
 		.driver_data = NVME_QUIRK_BROKEN_MSI },
+	{ PCI_DEVICE(0x15b7, 0x5009),   /* Sandisk SN550 */
+		.driver_data = NVME_QUIRK_BROKEN_MSI |
+				NVME_QUIRK_NO_DEEPEST_PS },
 	{ PCI_DEVICE(0x1987, 0x5012),	/* Phison E12 */
 		.driver_data = NVME_QUIRK_BOGUS_NID, },
 	{ PCI_DEVICE(0x1987, 0x5016),	/* Phison E16 */
diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 26c459f0198d..aba365f97cf6 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -1803,6 +1803,8 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl, int qid,
 		ret = PTR_ERR(sock_file);
 		goto err_destroy_mutex;
 	}
+
+	sk_net_refcnt_upgrade(queue->sock->sk);
 	nvme_tcp_reclassify_socket(queue->sock);
 
 	/* Single syn retry */
@@ -1944,7 +1946,7 @@ static void __nvme_tcp_stop_queue(struct nvme_tcp_queue *queue)
 	cancel_work_sync(&queue->io_work);
 }
 
-static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
+static void nvme_tcp_stop_queue_nowait(struct nvme_ctrl *nctrl, int qid)
 {
 	struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl);
 	struct nvme_tcp_queue *queue = &ctrl->queues[qid];
@@ -1963,6 +1965,31 @@ static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
 	mutex_unlock(&queue->queue_lock);
 }
 
+static void nvme_tcp_wait_queue(struct nvme_ctrl *nctrl, int qid)
+{
+	struct nvme_tcp_ctrl *ctrl = to_tcp_ctrl(nctrl);
+	struct nvme_tcp_queue *queue = &ctrl->queues[qid];
+	int timeout = 100;
+
+	while (timeout > 0) {
+		if (!test_bit(NVME_TCP_Q_ALLOCATED, &queue->flags) ||
+		    !sk_wmem_alloc_get(queue->sock->sk))
+			return;
+		msleep(2);
+		timeout -= 2;
+	}
+	dev_warn(nctrl->device,
+		 "qid %d: timeout draining sock wmem allocation expired\n",
+		 qid);
+}
+
+static void nvme_tcp_stop_queue(struct nvme_ctrl *nctrl, int qid)
+{
+	nvme_tcp_stop_queue_nowait(nctrl, qid);
+	nvme_tcp_wait_queue(nctrl, qid);
+}
+
+
 static void nvme_tcp_setup_sock_ops(struct nvme_tcp_queue *queue)
 {
 	write_lock_bh(&queue->sock->sk->sk_callback_lock);
@@ -2030,7 +2057,9 @@ static void nvme_tcp_stop_io_queues(struct nvme_ctrl *ctrl)
 	int i;
 
 	for (i = 1; i < ctrl->queue_count; i++)
-		nvme_tcp_stop_queue(ctrl, i);
+		nvme_tcp_stop_queue_nowait(ctrl, i);
+	for (i = 1; i < ctrl->queue_count; i++)
+		nvme_tcp_wait_queue(ctrl, i);
 }
 
 static int nvme_tcp_start_io_queues(struct nvme_ctrl *ctrl,
diff --git a/drivers/nvme/target/Kconfig b/drivers/nvme/target/Kconfig
index fb7446d6d682..4c253b433bf7 100644
--- a/drivers/nvme/target/Kconfig
+++ b/drivers/nvme/target/Kconfig
@@ -98,6 +98,7 @@ config NVME_TARGET_TCP_TLS
 	bool "NVMe over Fabrics TCP target TLS encryption support"
 	depends on NVME_TARGET_TCP
 	select NET_HANDSHAKE
+	select TLS
 	help
 	  Enables TLS encryption for the NVMe TCP target using the netlink handshake API.
 
diff --git a/drivers/nvme/target/auth.c b/drivers/nvme/target/auth.c
index 0b0645ac5df4..9429b8218408 100644
--- a/drivers/nvme/target/auth.c
+++ b/drivers/nvme/target/auth.c
@@ -240,7 +240,7 @@ void nvmet_auth_sq_free(struct nvmet_sq *sq)
 {
 	cancel_delayed_work(&sq->auth_expired_work);
 #ifdef CONFIG_NVME_TARGET_TCP_TLS
-	sq->tls_key = 0;
+	sq->tls_key = NULL;
 #endif
 	kfree(sq->dhchap_c1);
 	sq->dhchap_c1 = NULL;
@@ -600,13 +600,12 @@ void nvmet_auth_insert_psk(struct nvmet_sq *sq)
 		pr_warn("%s: ctrl %d qid %d failed to refresh key, error %ld\n",
 			__func__, sq->ctrl->cntlid, sq->qid, PTR_ERR(tls_key));
 		tls_key = NULL;
-		kfree_sensitive(tls_psk);
 	}
 	if (sq->ctrl->tls_key)
 		key_put(sq->ctrl->tls_key);
 	sq->ctrl->tls_key = tls_key;
 #endif
-
+	kfree_sensitive(tls_psk);
 out_free_digest:
 	kfree_sensitive(digest);
 out_free_psk:
diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index 71f8d06998d6..245475c43127 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -324,6 +324,9 @@ int nvmet_enable_port(struct nvmet_port *port)
 
 	lockdep_assert_held(&nvmet_config_sem);
 
+	if (port->disc_addr.trtype == NVMF_TRTYPE_MAX)
+		return -EINVAL;
+
 	ops = nvmet_transports[port->disc_addr.trtype];
 	if (!ops) {
 		up_write(&nvmet_config_sem);
diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c
index 7318b736d414..7b50130f10f6 100644
--- a/drivers/nvme/target/fc.c
+++ b/drivers/nvme/target/fc.c
@@ -995,16 +995,6 @@ nvmet_fc_hostport_get(struct nvmet_fc_hostport *hostport)
 	return kref_get_unless_zero(&hostport->ref);
 }
 
-static void
-nvmet_fc_free_hostport(struct nvmet_fc_hostport *hostport)
-{
-	/* if LLDD not implemented, leave as NULL */
-	if (!hostport || !hostport->hosthandle)
-		return;
-
-	nvmet_fc_hostport_put(hostport);
-}
-
 static struct nvmet_fc_hostport *
 nvmet_fc_match_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
 {
@@ -1028,33 +1018,24 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
 	struct nvmet_fc_hostport *newhost, *match = NULL;
 	unsigned long flags;
 
+	/*
+	 * Caller holds a reference on tgtport.
+	 */
+
 	/* if LLDD not implemented, leave as NULL */
 	if (!hosthandle)
 		return NULL;
 
-	/*
-	 * take reference for what will be the newly allocated hostport if
-	 * we end up using a new allocation
-	 */
-	if (!nvmet_fc_tgtport_get(tgtport))
-		return ERR_PTR(-EINVAL);
-
 	spin_lock_irqsave(&tgtport->lock, flags);
 	match = nvmet_fc_match_hostport(tgtport, hosthandle);
 	spin_unlock_irqrestore(&tgtport->lock, flags);
 
-	if (match) {
-		/* no new allocation - release reference */
-		nvmet_fc_tgtport_put(tgtport);
+	if (match)
 		return match;
-	}
 
 	newhost = kzalloc(sizeof(*newhost), GFP_KERNEL);
-	if (!newhost) {
-		/* no new allocation - release reference */
-		nvmet_fc_tgtport_put(tgtport);
+	if (!newhost)
 		return ERR_PTR(-ENOMEM);
-	}
 
 	spin_lock_irqsave(&tgtport->lock, flags);
 	match = nvmet_fc_match_hostport(tgtport, hosthandle);
@@ -1063,6 +1044,7 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
 		kfree(newhost);
 		newhost = match;
 	} else {
+		nvmet_fc_tgtport_get(tgtport);
 		newhost->tgtport = tgtport;
 		newhost->hosthandle = hosthandle;
 		INIT_LIST_HEAD(&newhost->host_list);
@@ -1076,20 +1058,14 @@ nvmet_fc_alloc_hostport(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
 }
 
 static void
-nvmet_fc_delete_assoc(struct nvmet_fc_tgt_assoc *assoc)
-{
-	nvmet_fc_delete_target_assoc(assoc);
-	nvmet_fc_tgt_a_put(assoc);
-}
-
-static void
 nvmet_fc_delete_assoc_work(struct work_struct *work)
 {
 	struct nvmet_fc_tgt_assoc *assoc =
 		container_of(work, struct nvmet_fc_tgt_assoc, del_work);
 	struct nvmet_fc_tgtport *tgtport = assoc->tgtport;
 
-	nvmet_fc_delete_assoc(assoc);
+	nvmet_fc_delete_target_assoc(assoc);
+	nvmet_fc_tgt_a_put(assoc);
 	nvmet_fc_tgtport_put(tgtport);
 }
 
@@ -1097,7 +1073,8 @@ static void
 nvmet_fc_schedule_delete_assoc(struct nvmet_fc_tgt_assoc *assoc)
 {
 	nvmet_fc_tgtport_get(assoc->tgtport);
-	queue_work(nvmet_wq, &assoc->del_work);
+	if (!queue_work(nvmet_wq, &assoc->del_work))
+		nvmet_fc_tgtport_put(assoc->tgtport);
 }
 
 static bool
@@ -1143,6 +1120,7 @@ nvmet_fc_alloc_target_assoc(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
 		goto out_ida;
 
 	assoc->tgtport = tgtport;
+	nvmet_fc_tgtport_get(tgtport);
 	assoc->a_id = idx;
 	INIT_LIST_HEAD(&assoc->a_list);
 	kref_init(&assoc->ref);
@@ -1190,7 +1168,7 @@ nvmet_fc_target_assoc_free(struct kref *ref)
 	/* Send Disconnect now that all i/o has completed */
 	nvmet_fc_xmt_disconnect_assoc(assoc);
 
-	nvmet_fc_free_hostport(assoc->hostport);
+	nvmet_fc_hostport_put(assoc->hostport);
 	spin_lock_irqsave(&tgtport->lock, flags);
 	oldls = assoc->rcv_disconn;
 	spin_unlock_irqrestore(&tgtport->lock, flags);
@@ -1244,6 +1222,8 @@ nvmet_fc_delete_target_assoc(struct nvmet_fc_tgt_assoc *assoc)
 	dev_info(tgtport->dev,
 		"{%d:%d} Association deleted\n",
 		tgtport->fc_target_port.port_num, assoc->a_id);
+
+	nvmet_fc_tgtport_put(tgtport);
 }
 
 static struct nvmet_fc_tgt_assoc *
@@ -1455,11 +1435,6 @@ nvmet_fc_free_tgtport(struct kref *ref)
 	struct nvmet_fc_tgtport *tgtport =
 		container_of(ref, struct nvmet_fc_tgtport, ref);
 	struct device *dev = tgtport->dev;
-	unsigned long flags;
-
-	spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
-	list_del(&tgtport->tgt_list);
-	spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
 
 	nvmet_fc_free_ls_iodlist(tgtport);
 
@@ -1620,6 +1595,11 @@ int
 nvmet_fc_unregister_targetport(struct nvmet_fc_target_port *target_port)
 {
 	struct nvmet_fc_tgtport *tgtport = targetport_to_tgtport(target_port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
+	list_del(&tgtport->tgt_list);
+	spin_unlock_irqrestore(&nvmet_fc_tgtlock, flags);
 
 	nvmet_fc_portentry_unbind_tgt(tgtport);
 
diff --git a/drivers/nvme/target/fcloop.c b/drivers/nvme/target/fcloop.c
index e1abb27927ff..641201e62c1b 100644
--- a/drivers/nvme/target/fcloop.c
+++ b/drivers/nvme/target/fcloop.c
@@ -208,6 +208,7 @@ struct fcloop_lport {
 	struct nvme_fc_local_port *localport;
 	struct list_head lport_list;
 	struct completion unreg_done;
+	refcount_t ref;
 };
 
 struct fcloop_lport_priv {
@@ -239,7 +240,7 @@ struct fcloop_nport {
 	struct fcloop_tport *tport;
 	struct fcloop_lport *lport;
 	struct list_head nport_list;
-	struct kref ref;
+	refcount_t ref;
 	u64 node_name;
 	u64 port_name;
 	u32 port_role;
@@ -274,7 +275,7 @@ struct fcloop_fcpreq {
 	u32				inistate;
 	bool				active;
 	bool				aborted;
-	struct kref			ref;
+	refcount_t			ref;
 	struct work_struct		fcp_rcv_work;
 	struct work_struct		abort_rcv_work;
 	struct work_struct		tio_done_work;
@@ -478,7 +479,7 @@ fcloop_t2h_xmt_ls_rsp(struct nvme_fc_local_port *localport,
 	if (targetport) {
 		tport = targetport->private;
 		spin_lock(&tport->lock);
-		list_add_tail(&tport->ls_list, &tls_req->ls_list);
+		list_add_tail(&tls_req->ls_list, &tport->ls_list);
 		spin_unlock(&tport->lock);
 		queue_work(nvmet_wq, &tport->ls_work);
 	}
@@ -534,24 +535,18 @@ fcloop_tgt_discovery_evt(struct nvmet_fc_target_port *tgtport)
 }
 
 static void
-fcloop_tfcp_req_free(struct kref *ref)
+fcloop_tfcp_req_put(struct fcloop_fcpreq *tfcp_req)
 {
-	struct fcloop_fcpreq *tfcp_req =
-		container_of(ref, struct fcloop_fcpreq, ref);
+	if (!refcount_dec_and_test(&tfcp_req->ref))
+		return;
 
 	kfree(tfcp_req);
 }
 
-static void
-fcloop_tfcp_req_put(struct fcloop_fcpreq *tfcp_req)
-{
-	kref_put(&tfcp_req->ref, fcloop_tfcp_req_free);
-}
-
 static int
 fcloop_tfcp_req_get(struct fcloop_fcpreq *tfcp_req)
 {
-	return kref_get_unless_zero(&tfcp_req->ref);
+	return refcount_inc_not_zero(&tfcp_req->ref);
 }
 
 static void
@@ -748,7 +743,7 @@ fcloop_fcp_req(struct nvme_fc_local_port *localport,
 	INIT_WORK(&tfcp_req->fcp_rcv_work, fcloop_fcp_recv_work);
 	INIT_WORK(&tfcp_req->abort_rcv_work, fcloop_fcp_abort_recv_work);
 	INIT_WORK(&tfcp_req->tio_done_work, fcloop_tgt_fcprqst_done_work);
-	kref_init(&tfcp_req->ref);
+	refcount_set(&tfcp_req->ref, 1);
 
 	queue_work(nvmet_wq, &tfcp_req->fcp_rcv_work);
 
@@ -1001,24 +996,39 @@ fcloop_fcp_abort(struct nvme_fc_local_port *localport,
 }
 
 static void
-fcloop_nport_free(struct kref *ref)
+fcloop_lport_put(struct fcloop_lport *lport)
 {
-	struct fcloop_nport *nport =
-		container_of(ref, struct fcloop_nport, ref);
+	unsigned long flags;
 
-	kfree(nport);
+	if (!refcount_dec_and_test(&lport->ref))
+		return;
+
+	spin_lock_irqsave(&fcloop_lock, flags);
+	list_del(&lport->lport_list);
+	spin_unlock_irqrestore(&fcloop_lock, flags);
+
+	kfree(lport);
+}
+
+static int
+fcloop_lport_get(struct fcloop_lport *lport)
+{
+	return refcount_inc_not_zero(&lport->ref);
 }
 
 static void
 fcloop_nport_put(struct fcloop_nport *nport)
 {
-	kref_put(&nport->ref, fcloop_nport_free);
+	if (!refcount_dec_and_test(&nport->ref))
+		return;
+
+	kfree(nport);
 }
 
 static int
 fcloop_nport_get(struct fcloop_nport *nport)
 {
-	return kref_get_unless_zero(&nport->ref);
+	return refcount_inc_not_zero(&nport->ref);
 }
 
 static void
@@ -1029,6 +1039,8 @@ fcloop_localport_delete(struct nvme_fc_local_port *localport)
 
 	/* release any threads waiting for the unreg to complete */
 	complete(&lport->unreg_done);
+
+	fcloop_lport_put(lport);
 }
 
 static void
@@ -1140,6 +1152,7 @@ fcloop_create_local_port(struct device *dev, struct device_attribute *attr,
 
 		lport->localport = localport;
 		INIT_LIST_HEAD(&lport->lport_list);
+		refcount_set(&lport->ref, 1);
 
 		spin_lock_irqsave(&fcloop_lock, flags);
 		list_add_tail(&lport->lport_list, &fcloop_lports);
@@ -1156,13 +1169,6 @@ out_free_lport:
 	return ret ? ret : count;
 }
 
-
-static void
-__unlink_local_port(struct fcloop_lport *lport)
-{
-	list_del(&lport->lport_list);
-}
-
 static int
 __wait_localport_unreg(struct fcloop_lport *lport)
 {
@@ -1175,8 +1181,6 @@ __wait_localport_unreg(struct fcloop_lport *lport)
 	if (!ret)
 		wait_for_completion(&lport->unreg_done);
 
-	kfree(lport);
-
 	return ret;
 }
 
@@ -1199,8 +1203,9 @@ fcloop_delete_local_port(struct device *dev, struct device_attribute *attr,
 	list_for_each_entry(tlport, &fcloop_lports, lport_list) {
 		if (tlport->localport->node_name == nodename &&
 		    tlport->localport->port_name == portname) {
+			if (!fcloop_lport_get(tlport))
+				break;
 			lport = tlport;
-			__unlink_local_port(lport);
 			break;
 		}
 	}
@@ -1210,6 +1215,7 @@ fcloop_delete_local_port(struct device *dev, struct device_attribute *attr,
 		return -ENOENT;
 
 	ret = __wait_localport_unreg(lport);
+	fcloop_lport_put(lport);
 
 	return ret ? ret : count;
 }
@@ -1249,7 +1255,7 @@ fcloop_alloc_nport(const char *buf, size_t count, bool remoteport)
 		newnport->port_role = opts->roles;
 	if (opts->mask & NVMF_OPT_FCADDR)
 		newnport->port_id = opts->fcaddr;
-	kref_init(&newnport->ref);
+	refcount_set(&newnport->ref, 1);
 
 	spin_lock_irqsave(&fcloop_lock, flags);
 
@@ -1637,17 +1643,17 @@ static void __exit fcloop_exit(void)
 	for (;;) {
 		lport = list_first_entry_or_null(&fcloop_lports,
 						typeof(*lport), lport_list);
-		if (!lport)
+		if (!lport || !fcloop_lport_get(lport))
 			break;
 
-		__unlink_local_port(lport);
-
 		spin_unlock_irqrestore(&fcloop_lock, flags);
 
 		ret = __wait_localport_unreg(lport);
 		if (ret)
 			pr_warn("%s: Failed deleting local port\n", __func__);
 
+		fcloop_lport_put(lport);
+
 		spin_lock_irqsave(&fcloop_lock, flags);
 	}
 
diff --git a/drivers/nvme/target/pci-epf.c b/drivers/nvme/target/pci-epf.c
index 51c27b32248d..7fab7f3d79b7 100644
--- a/drivers/nvme/target/pci-epf.c
+++ b/drivers/nvme/target/pci-epf.c
@@ -1648,16 +1648,17 @@ static int nvmet_pci_epf_process_sq(struct nvmet_pci_epf_ctrl *ctrl,
 {
 	struct nvmet_pci_epf_iod *iod;
 	int ret, n = 0;
+	u16 head = sq->head;
 
 	sq->tail = nvmet_pci_epf_bar_read32(ctrl, sq->db);
-	while (sq->head != sq->tail && (!ctrl->sq_ab || n < ctrl->sq_ab)) {
+	while (head != sq->tail && (!ctrl->sq_ab || n < ctrl->sq_ab)) {
 		iod = nvmet_pci_epf_alloc_iod(sq);
 		if (!iod)
 			break;
 
 		/* Get the NVMe command submitted by the host. */
 		ret = nvmet_pci_epf_transfer(ctrl, &iod->cmd,
-					     sq->pci_addr + sq->head * sq->qes,
+					     sq->pci_addr + head * sq->qes,
 					     sq->qes, DMA_FROM_DEVICE);
 		if (ret) {
 			/* Not much we can do... */
@@ -1666,12 +1667,13 @@ static int nvmet_pci_epf_process_sq(struct nvmet_pci_epf_ctrl *ctrl,
 		}
 
 		dev_dbg(ctrl->dev, "SQ[%u]: head %u, tail %u, command %s\n",
-			sq->qid, sq->head, sq->tail,
+			sq->qid, head, sq->tail,
 			nvmet_pci_epf_iod_name(iod));
 
-		sq->head++;
-		if (sq->head == sq->depth)
-			sq->head = 0;
+		head++;
+		if (head == sq->depth)
+			head = 0;
+		WRITE_ONCE(sq->head, head);
 		n++;
 
 		queue_work_on(WORK_CPU_UNBOUND, sq->iod_wq, &iod->work);
@@ -1761,8 +1763,17 @@ static void nvmet_pci_epf_cq_work(struct work_struct *work)
 		if (!iod)
 			break;
 
-		/* Post the IOD completion entry. */
+		/*
+		 * Post the IOD completion entry. If the IOD request was
+		 * executed (req->execute() called), the CQE is already
+		 * initialized. However, the IOD may have been failed before
+		 * that, leaving the CQE not properly initialized. So always
+		 * initialize it here.
+		 */
 		cqe = &iod->cqe;
+		cqe->sq_head = cpu_to_le16(READ_ONCE(iod->sq->head));
+		cqe->sq_id = cpu_to_le16(iod->sq->qid);
+		cqe->command_id = iod->cmd.common.command_id;
 		cqe->status = cpu_to_le16((iod->status << 1) | cq->phase);
 
 		dev_dbg(ctrl->dev,
@@ -1800,6 +1811,21 @@ static void nvmet_pci_epf_cq_work(struct work_struct *work)
 				   NVMET_PCI_EPF_CQ_RETRY_INTERVAL);
 }
 
+static void nvmet_pci_epf_clear_ctrl_config(struct nvmet_pci_epf_ctrl *ctrl)
+{
+	struct nvmet_ctrl *tctrl = ctrl->tctrl;
+
+	/* Initialize controller status. */
+	tctrl->csts = 0;
+	ctrl->csts = 0;
+	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
+
+	/* Initialize controller configuration and start polling. */
+	tctrl->cc = 0;
+	ctrl->cc = 0;
+	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+}
+
 static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
 {
 	u64 pci_addr, asq, acq;
@@ -1865,18 +1891,20 @@ static int nvmet_pci_epf_enable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
 	return 0;
 
 err:
-	ctrl->csts = 0;
+	nvmet_pci_epf_clear_ctrl_config(ctrl);
 	return -EINVAL;
 }
 
-static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
+static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl,
+				       bool shutdown)
 {
 	int qid;
 
 	if (!ctrl->enabled)
 		return;
 
-	dev_info(ctrl->dev, "Disabling controller\n");
+	dev_info(ctrl->dev, "%s controller\n",
+		 shutdown ? "Shutting down" : "Disabling");
 
 	ctrl->enabled = false;
 	cancel_delayed_work_sync(&ctrl->poll_sqs);
@@ -1893,6 +1921,11 @@ static void nvmet_pci_epf_disable_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
 	nvmet_pci_epf_delete_cq(ctrl->tctrl, 0);
 
 	ctrl->csts &= ~NVME_CSTS_RDY;
+	if (shutdown) {
+		ctrl->csts |= NVME_CSTS_SHST_CMPLT;
+		ctrl->cc &= ~NVME_CC_ENABLE;
+		nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+	}
 }
 
 static void nvmet_pci_epf_poll_cc_work(struct work_struct *work)
@@ -1919,12 +1952,10 @@ static void nvmet_pci_epf_poll_cc_work(struct work_struct *work)
 	}
 
 	if (!nvmet_cc_en(new_cc) && nvmet_cc_en(old_cc))
-		nvmet_pci_epf_disable_ctrl(ctrl);
+		nvmet_pci_epf_disable_ctrl(ctrl, false);
 
-	if (nvmet_cc_shn(new_cc) && !nvmet_cc_shn(old_cc)) {
-		nvmet_pci_epf_disable_ctrl(ctrl);
-		ctrl->csts |= NVME_CSTS_SHST_CMPLT;
-	}
+	if (nvmet_cc_shn(new_cc) && !nvmet_cc_shn(old_cc))
+		nvmet_pci_epf_disable_ctrl(ctrl, true);
 
 	if (!nvmet_cc_shn(new_cc) && nvmet_cc_shn(old_cc))
 		ctrl->csts &= ~NVME_CSTS_SHST_CMPLT;
@@ -1963,16 +1994,10 @@ static void nvmet_pci_epf_init_bar(struct nvmet_pci_epf_ctrl *ctrl)
 	/* Clear Controller Memory Buffer Supported (CMBS). */
 	ctrl->cap &= ~(0x1ULL << 57);
 
-	/* Controller configuration. */
-	ctrl->cc = tctrl->cc & (~NVME_CC_ENABLE);
-
-	/* Controller status. */
-	ctrl->csts = ctrl->tctrl->csts;
-
 	nvmet_pci_epf_bar_write64(ctrl, NVME_REG_CAP, ctrl->cap);
 	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_VS, tctrl->subsys->ver);
-	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CSTS, ctrl->csts);
-	nvmet_pci_epf_bar_write32(ctrl, NVME_REG_CC, ctrl->cc);
+
+	nvmet_pci_epf_clear_ctrl_config(ctrl);
 }
 
 static int nvmet_pci_epf_create_ctrl(struct nvmet_pci_epf *nvme_epf,
@@ -2070,14 +2095,22 @@ out_mempool_exit:
 
 static void nvmet_pci_epf_start_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
 {
+
+	dev_info(ctrl->dev, "PCI link up\n");
+	ctrl->link_up = true;
+
 	schedule_delayed_work(&ctrl->poll_cc, NVMET_PCI_EPF_CC_POLL_INTERVAL);
 }
 
 static void nvmet_pci_epf_stop_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
 {
+	dev_info(ctrl->dev, "PCI link down\n");
+	ctrl->link_up = false;
+
 	cancel_delayed_work_sync(&ctrl->poll_cc);
 
-	nvmet_pci_epf_disable_ctrl(ctrl);
+	nvmet_pci_epf_disable_ctrl(ctrl, false);
+	nvmet_pci_epf_clear_ctrl_config(ctrl);
 }
 
 static void nvmet_pci_epf_destroy_ctrl(struct nvmet_pci_epf_ctrl *ctrl)
@@ -2300,10 +2333,8 @@ static int nvmet_pci_epf_epc_init(struct pci_epf *epf)
 	if (ret)
 		goto out_clear_bar;
 
-	if (!epc_features->linkup_notifier) {
-		ctrl->link_up = true;
+	if (!epc_features->linkup_notifier)
 		nvmet_pci_epf_start_ctrl(&nvme_epf->ctrl);
-	}
 
 	return 0;
 
@@ -2319,7 +2350,6 @@ static void nvmet_pci_epf_epc_deinit(struct pci_epf *epf)
 	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
 	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
 
-	ctrl->link_up = false;
 	nvmet_pci_epf_destroy_ctrl(ctrl);
 
 	nvmet_pci_epf_deinit_dma(nvme_epf);
@@ -2331,7 +2361,6 @@ static int nvmet_pci_epf_link_up(struct pci_epf *epf)
 	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
 	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
 
-	ctrl->link_up = true;
 	nvmet_pci_epf_start_ctrl(ctrl);
 
 	return 0;
@@ -2342,7 +2371,6 @@ static int nvmet_pci_epf_link_down(struct pci_epf *epf)
 	struct nvmet_pci_epf *nvme_epf = epf_get_drvdata(epf);
 	struct nvmet_pci_epf_ctrl *ctrl = &nvme_epf->ctrl;
 
-	ctrl->link_up = false;
 	nvmet_pci_epf_stop_ctrl(ctrl);
 
 	return 0;
diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c
index f2d0c920269b..12a5cb8641ca 100644
--- a/drivers/nvme/target/tcp.c
+++ b/drivers/nvme/target/tcp.c
@@ -1560,6 +1560,9 @@ static void nvmet_tcp_restore_socket_callbacks(struct nvmet_tcp_queue *queue)
 {
 	struct socket *sock = queue->sock;
 
+	if (!queue->state_change)
+		return;
+
 	write_lock_bh(&sock->sk->sk_callback_lock);
 	sock->sk->sk_data_ready =  queue->data_ready;
 	sock->sk->sk_state_change = queue->state_change;
diff --git a/drivers/nvmem/core.c b/drivers/nvmem/core.c
index fff85bbf0ecd..e206efc29a00 100644
--- a/drivers/nvmem/core.c
+++ b/drivers/nvmem/core.c
@@ -594,9 +594,11 @@ static int nvmem_cell_info_to_nvmem_cell_entry_nodup(struct nvmem_device *nvmem,
 	cell->nbits = info->nbits;
 	cell->np = info->np;
 
-	if (cell->nbits)
+	if (cell->nbits) {
 		cell->bytes = DIV_ROUND_UP(cell->nbits + cell->bit_offset,
 					   BITS_PER_BYTE);
+		cell->raw_len = ALIGN(cell->bytes, nvmem->word_size);
+	}
 
 	if (!IS_ALIGNED(cell->offset, nvmem->stride)) {
 		dev_err(&nvmem->dev,
@@ -605,6 +607,18 @@ static int nvmem_cell_info_to_nvmem_cell_entry_nodup(struct nvmem_device *nvmem,
 		return -EINVAL;
 	}
 
+	if (!IS_ALIGNED(cell->raw_len, nvmem->word_size)) {
+		dev_err(&nvmem->dev,
+			"cell %s raw len %zd unaligned to nvmem word size %d\n",
+			cell->name ?: "<unknown>", cell->raw_len,
+			nvmem->word_size);
+
+		if (info->raw_len)
+			return -EINVAL;
+
+		cell->raw_len = ALIGN(cell->raw_len, nvmem->word_size);
+	}
+
 	return 0;
 }
 
@@ -837,7 +851,9 @@ static int nvmem_add_cells_from_dt(struct nvmem_device *nvmem, struct device_nod
 		if (addr && len == (2 * sizeof(u32))) {
 			info.bit_offset = be32_to_cpup(addr++);
 			info.nbits = be32_to_cpup(addr);
-			if (info.bit_offset >= BITS_PER_BYTE || info.nbits < 1) {
+			if (info.bit_offset >= BITS_PER_BYTE * info.bytes ||
+			    info.nbits < 1 ||
+			    info.bit_offset + info.nbits > BITS_PER_BYTE * info.bytes) {
 				dev_err(dev, "nvmem: invalid bits on %pOF\n", child);
 				of_node_put(child);
 				return -EINVAL;
@@ -1630,21 +1646,29 @@ EXPORT_SYMBOL_GPL(nvmem_cell_put);
 static void nvmem_shift_read_buffer_in_place(struct nvmem_cell_entry *cell, void *buf)
 {
 	u8 *p, *b;
-	int i, extra, bit_offset = cell->bit_offset;
+	int i, extra, bytes_offset;
+	int bit_offset = cell->bit_offset;
 
 	p = b = buf;
-	if (bit_offset) {
+
+	bytes_offset = bit_offset / BITS_PER_BYTE;
+	b += bytes_offset;
+	bit_offset %= BITS_PER_BYTE;
+
+	if (bit_offset % BITS_PER_BYTE) {
 		/* First shift */
-		*b++ >>= bit_offset;
+		*p = *b++ >> bit_offset;
 
 		/* setup rest of the bytes if any */
 		for (i = 1; i < cell->bytes; i++) {
 			/* Get bits from next byte and shift them towards msb */
-			*p |= *b << (BITS_PER_BYTE - bit_offset);
+			*p++ |= *b << (BITS_PER_BYTE - bit_offset);
 
-			p = b;
-			*b++ >>= bit_offset;
+			*p = *b++ >> bit_offset;
 		}
+	} else if (p != b) {
+		memmove(p, b, cell->bytes - bytes_offset);
+		p += cell->bytes - 1;
 	} else {
 		/* point to the msb */
 		p += cell->bytes - 1;
diff --git a/drivers/nvmem/qfprom.c b/drivers/nvmem/qfprom.c
index 116a39e804c7..a872c640b8c5 100644
--- a/drivers/nvmem/qfprom.c
+++ b/drivers/nvmem/qfprom.c
@@ -321,19 +321,32 @@ static int qfprom_reg_read(void *context,
 			unsigned int reg, void *_val, size_t bytes)
 {
 	struct qfprom_priv *priv = context;
-	u8 *val = _val;
-	int i = 0, words = bytes;
+	u32 *val = _val;
 	void __iomem *base = priv->qfpcorrected;
+	int words = DIV_ROUND_UP(bytes, sizeof(u32));
+	int i;
 
 	if (read_raw_data && priv->qfpraw)
 		base = priv->qfpraw;
 
-	while (words--)
-		*val++ = readb(base + reg + i++);
+	for (i = 0; i < words; i++)
+		*val++ = readl(base + reg + i * sizeof(u32));
 
 	return 0;
 }
 
+/* Align reads to word boundary */
+static void qfprom_fixup_dt_cell_info(struct nvmem_device *nvmem,
+				      struct nvmem_cell_info *cell)
+{
+	unsigned int byte_offset = cell->offset % sizeof(u32);
+
+	cell->bit_offset += byte_offset * BITS_PER_BYTE;
+	cell->offset -= byte_offset;
+	if (byte_offset && !cell->nbits)
+		cell->nbits = cell->bytes * BITS_PER_BYTE;
+}
+
 static void qfprom_runtime_disable(void *data)
 {
 	pm_runtime_disable(data);
@@ -358,10 +371,11 @@ static int qfprom_probe(struct platform_device *pdev)
 	struct nvmem_config econfig = {
 		.name = "qfprom",
 		.add_legacy_fixed_of_cells = true,
-		.stride = 1,
-		.word_size = 1,
+		.stride = 4,
+		.word_size = 4,
 		.id = NVMEM_DEVID_AUTO,
 		.reg_read = qfprom_reg_read,
+		.fixup_dt_cell_info = qfprom_fixup_dt_cell_info,
 	};
 	struct device *dev = &pdev->dev;
 	struct resource *res;
diff --git a/drivers/nvmem/rockchip-otp.c b/drivers/nvmem/rockchip-otp.c
index ebc3f0b24166..d88f12c53242 100644
--- a/drivers/nvmem/rockchip-otp.c
+++ b/drivers/nvmem/rockchip-otp.c
@@ -59,7 +59,6 @@
 #define RK3588_OTPC_AUTO_EN		0x08
 #define RK3588_OTPC_INT_ST		0x84
 #define RK3588_OTPC_DOUT0		0x20
-#define RK3588_NO_SECURE_OFFSET		0x300
 #define RK3588_NBYTES			4
 #define RK3588_BURST_NUM		1
 #define RK3588_BURST_SHIFT		8
@@ -69,6 +68,7 @@
 
 struct rockchip_data {
 	int size;
+	int read_offset;
 	const char * const *clks;
 	int num_clks;
 	nvmem_reg_read_t reg_read;
@@ -196,7 +196,7 @@ static int rk3588_otp_read(void *context, unsigned int offset,
 	addr_start = round_down(offset, RK3588_NBYTES) / RK3588_NBYTES;
 	addr_end = round_up(offset + bytes, RK3588_NBYTES) / RK3588_NBYTES;
 	addr_len = addr_end - addr_start;
-	addr_start += RK3588_NO_SECURE_OFFSET;
+	addr_start += otp->data->read_offset / RK3588_NBYTES;
 
 	buf = kzalloc(array_size(addr_len, RK3588_NBYTES), GFP_KERNEL);
 	if (!buf)
@@ -274,12 +274,21 @@ static const struct rockchip_data px30_data = {
 	.reg_read = px30_otp_read,
 };
 
+static const struct rockchip_data rk3576_data = {
+	.size = 0x100,
+	.read_offset = 0x700,
+	.clks = px30_otp_clocks,
+	.num_clks = ARRAY_SIZE(px30_otp_clocks),
+	.reg_read = rk3588_otp_read,
+};
+
 static const char * const rk3588_otp_clocks[] = {
 	"otp", "apb_pclk", "phy", "arb",
 };
 
 static const struct rockchip_data rk3588_data = {
 	.size = 0x400,
+	.read_offset = 0xc00,
 	.clks = rk3588_otp_clocks,
 	.num_clks = ARRAY_SIZE(rk3588_otp_clocks),
 	.reg_read = rk3588_otp_read,
@@ -295,6 +304,10 @@ static const struct of_device_id rockchip_otp_match[] = {
 		.data = &px30_data,
 	},
 	{
+		.compatible = "rockchip,rk3576-otp",
+		.data = &rk3576_data,
+	},
+	{
 		.compatible = "rockchip,rk3588-otp",
 		.data = &rk3588_data,
 	},
diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c
index 6569ba3577fe..8b8848788618 100644
--- a/drivers/pci/msi/msi.c
+++ b/drivers/pci/msi/msi.c
@@ -615,6 +615,9 @@ void msix_prepare_msi_desc(struct pci_dev *dev, struct msi_desc *desc)
 		void __iomem *addr = pci_msix_desc_addr(desc);
 
 		desc->pci.msi_attrib.can_mask = 1;
+		/* Workaround for SUN NIU insanity, which requires write before read */
+		if (dev->dev_flags & PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST)
+			writel(0, addr + PCI_MSIX_ENTRY_DATA);
 		desc->pci.msix_ctrl = readl(addr + PCI_MSIX_ENTRY_VECTOR_CTRL);
 	}
 }
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 4d7c9f64ea24..e77d5b53c0ce 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -5429,8 +5429,6 @@ static bool pci_bus_resettable(struct pci_bus *bus)
 		return false;
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
-		if (!pci_reset_supported(dev))
-			return false;
 		if (dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET ||
 		    (dev->subordinate && !pci_bus_resettable(dev->subordinate)))
 			return false;
@@ -5507,8 +5505,6 @@ static bool pci_slot_resettable(struct pci_slot *slot)
 	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
 		if (!dev->slot || dev->slot != slot)
 			continue;
-		if (!pci_reset_supported(dev))
-			return false;
 		if (dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET ||
 		    (dev->subordinate && !pci_bus_resettable(dev->subordinate)))
 			return false;
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 8d610c17e0f2..94daca15a096 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -1990,12 +1990,12 @@ static void quirk_huawei_pcie_sva(struct pci_dev *pdev)
 	    device_create_managed_software_node(&pdev->dev, properties, NULL))
 		pci_warn(pdev, "could not add stall property");
 }
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_HUAWEI, 0xa250, quirk_huawei_pcie_sva);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_HUAWEI, 0xa251, quirk_huawei_pcie_sva);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_HUAWEI, 0xa255, quirk_huawei_pcie_sva);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_HUAWEI, 0xa256, quirk_huawei_pcie_sva);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_HUAWEI, 0xa258, quirk_huawei_pcie_sva);
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_HUAWEI, 0xa259, quirk_huawei_pcie_sva);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_HUAWEI, 0xa250, quirk_huawei_pcie_sva);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_HUAWEI, 0xa251, quirk_huawei_pcie_sva);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_HUAWEI, 0xa255, quirk_huawei_pcie_sva);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_HUAWEI, 0xa256, quirk_huawei_pcie_sva);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_HUAWEI, 0xa258, quirk_huawei_pcie_sva);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_HUAWEI, 0xa259, quirk_huawei_pcie_sva);
 
 /*
  * It's possible for the MSI to get corrupted if SHPC and ACPI are used
diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index 54d6f4fa3ce1..e994c546422c 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -187,6 +187,9 @@ static void pdev_sort_resources(struct pci_dev *dev, struct list_head *head)
 			panic("%s: kzalloc() failed!\n", __func__);
 		tmp->res = r;
 		tmp->dev = dev;
+		tmp->start = r->start;
+		tmp->end = r->end;
+		tmp->flags = r->flags;
 
 		/* Fallback is smallest one or list is empty */
 		n = head;
@@ -545,6 +548,7 @@ assign:
 		pci_dbg(dev, "%s %pR: releasing\n", res_name, res);
 
 		release_resource(res);
+		restore_dev_resource(dev_res);
 	}
 	/* Restore start/end/flags from saved list */
 	list_for_each_entry(save_res, &save_head, list)
diff --git a/drivers/pinctrl/freescale/pinctrl-imx.c b/drivers/pinctrl/freescale/pinctrl-imx.c
index 842a1e6cbfc4..18de31328540 100644
--- a/drivers/pinctrl/freescale/pinctrl-imx.c
+++ b/drivers/pinctrl/freescale/pinctrl-imx.c
@@ -37,16 +37,16 @@ static inline const struct group_desc *imx_pinctrl_find_group_by_name(
 				struct pinctrl_dev *pctldev,
 				const char *name)
 {
-	const struct group_desc *grp = NULL;
+	const struct group_desc *grp;
 	int i;
 
 	for (i = 0; i < pctldev->num_groups; i++) {
 		grp = pinctrl_generic_get_group(pctldev, i);
 		if (grp && !strcmp(grp->grp.name, name))
-			break;
+			return grp;
 	}
 
-	return grp;
+	return NULL;
 }
 
 static void imx_pin_dbg_show(struct pinctrl_dev *pctldev, struct seq_file *s,
diff --git a/drivers/pinctrl/mediatek/mtk-eint.c b/drivers/pinctrl/mediatek/mtk-eint.c
index ced4ee509b5b..b4eb2beab691 100644
--- a/drivers/pinctrl/mediatek/mtk-eint.c
+++ b/drivers/pinctrl/mediatek/mtk-eint.c
@@ -449,7 +449,7 @@ int mtk_eint_set_debounce(struct mtk_eint *eint, unsigned long eint_num,
 		return -EOPNOTSUPP;
 
 	virq = irq_find_mapping(eint->domain, eint_num);
-	eint_offset = (eint_num % 4) * 8;
+	eint_offset = (idx % 4) * 8;
 	d = irq_get_irq_data(virq);
 
 	set_offset = (idx / 4) * 4 + eint->regs->dbnc_set;
diff --git a/drivers/pinctrl/mediatek/pinctrl-airoha.c b/drivers/pinctrl/mediatek/pinctrl-airoha.c
index 547a798b71c8..5d84a778683d 100644
--- a/drivers/pinctrl/mediatek/pinctrl-airoha.c
+++ b/drivers/pinctrl/mediatek/pinctrl-airoha.c
@@ -6,6 +6,7 @@
  */
 
 #include <dt-bindings/pinctrl/mt65xx.h>
+#include <linux/bitfield.h>
 #include <linux/bits.h>
 #include <linux/cleanup.h>
 #include <linux/gpio/driver.h>
@@ -112,39 +113,19 @@
 #define REG_LAN_LED1_MAPPING			0x0280
 
 #define LAN4_LED_MAPPING_MASK			GENMASK(18, 16)
-#define LAN4_PHY4_LED_MAP			BIT(18)
-#define LAN4_PHY2_LED_MAP			BIT(17)
-#define LAN4_PHY1_LED_MAP			BIT(16)
-#define LAN4_PHY0_LED_MAP			0
-#define LAN4_PHY3_LED_MAP			GENMASK(17, 16)
+#define LAN4_PHY_LED_MAP(_n)			FIELD_PREP_CONST(LAN4_LED_MAPPING_MASK, (_n))
 
 #define LAN3_LED_MAPPING_MASK			GENMASK(14, 12)
-#define LAN3_PHY4_LED_MAP			BIT(14)
-#define LAN3_PHY2_LED_MAP			BIT(13)
-#define LAN3_PHY1_LED_MAP			BIT(12)
-#define LAN3_PHY0_LED_MAP			0
-#define LAN3_PHY3_LED_MAP			GENMASK(13, 12)
+#define LAN3_PHY_LED_MAP(_n)			FIELD_PREP_CONST(LAN3_LED_MAPPING_MASK, (_n))
 
 #define LAN2_LED_MAPPING_MASK			GENMASK(10, 8)
-#define LAN2_PHY4_LED_MAP			BIT(12)
-#define LAN2_PHY2_LED_MAP			BIT(11)
-#define LAN2_PHY1_LED_MAP			BIT(10)
-#define LAN2_PHY0_LED_MAP			0
-#define LAN2_PHY3_LED_MAP			GENMASK(11, 10)
+#define LAN2_PHY_LED_MAP(_n)			FIELD_PREP_CONST(LAN2_LED_MAPPING_MASK, (_n))
 
 #define LAN1_LED_MAPPING_MASK			GENMASK(6, 4)
-#define LAN1_PHY4_LED_MAP			BIT(6)
-#define LAN1_PHY2_LED_MAP			BIT(5)
-#define LAN1_PHY1_LED_MAP			BIT(4)
-#define LAN1_PHY0_LED_MAP			0
-#define LAN1_PHY3_LED_MAP			GENMASK(5, 4)
+#define LAN1_PHY_LED_MAP(_n)			FIELD_PREP_CONST(LAN1_LED_MAPPING_MASK, (_n))
 
 #define LAN0_LED_MAPPING_MASK			GENMASK(2, 0)
-#define LAN0_PHY4_LED_MAP			BIT(3)
-#define LAN0_PHY2_LED_MAP			BIT(2)
-#define LAN0_PHY1_LED_MAP			BIT(1)
-#define LAN0_PHY0_LED_MAP			0
-#define LAN0_PHY3_LED_MAP			GENMASK(2, 1)
+#define LAN0_PHY_LED_MAP(_n)			FIELD_PREP_CONST(LAN0_LED_MAPPING_MASK, (_n))
 
 /* CONF */
 #define REG_I2C_SDA_E2				0x001c
@@ -1476,8 +1457,8 @@ static const struct airoha_pinctrl_func_group phy1_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY1_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1491,8 +1472,8 @@ static const struct airoha_pinctrl_func_group phy1_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY1_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1506,8 +1487,8 @@ static const struct airoha_pinctrl_func_group phy1_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY1_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1521,8 +1502,8 @@ static const struct airoha_pinctrl_func_group phy1_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY1_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	},
@@ -1540,8 +1521,8 @@ static const struct airoha_pinctrl_func_group phy2_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY2_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1555,8 +1536,8 @@ static const struct airoha_pinctrl_func_group phy2_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY2_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1570,8 +1551,8 @@ static const struct airoha_pinctrl_func_group phy2_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY2_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1585,8 +1566,8 @@ static const struct airoha_pinctrl_func_group phy2_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY2_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	},
@@ -1604,8 +1585,8 @@ static const struct airoha_pinctrl_func_group phy3_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY3_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1619,8 +1600,8 @@ static const struct airoha_pinctrl_func_group phy3_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY3_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1634,8 +1615,8 @@ static const struct airoha_pinctrl_func_group phy3_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY3_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1649,8 +1630,8 @@ static const struct airoha_pinctrl_func_group phy3_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY3_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	},
@@ -1668,8 +1649,8 @@ static const struct airoha_pinctrl_func_group phy4_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY4_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1683,8 +1664,8 @@ static const struct airoha_pinctrl_func_group phy4_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY4_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1698,8 +1679,8 @@ static const struct airoha_pinctrl_func_group phy4_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY4_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1713,8 +1694,8 @@ static const struct airoha_pinctrl_func_group phy4_led0_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED0_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY4_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	},
@@ -1732,8 +1713,8 @@ static const struct airoha_pinctrl_func_group phy1_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY1_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1747,8 +1728,8 @@ static const struct airoha_pinctrl_func_group phy1_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY1_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1762,8 +1743,8 @@ static const struct airoha_pinctrl_func_group phy1_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY1_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1777,8 +1758,8 @@ static const struct airoha_pinctrl_func_group phy1_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY1_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(0)
 		},
 		.regmap_size = 2,
 	},
@@ -1796,8 +1777,8 @@ static const struct airoha_pinctrl_func_group phy2_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY2_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1811,8 +1792,8 @@ static const struct airoha_pinctrl_func_group phy2_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY2_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1826,8 +1807,8 @@ static const struct airoha_pinctrl_func_group phy2_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY2_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1841,8 +1822,8 @@ static const struct airoha_pinctrl_func_group phy2_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY2_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(1)
 		},
 		.regmap_size = 2,
 	},
@@ -1860,8 +1841,8 @@ static const struct airoha_pinctrl_func_group phy3_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY3_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1875,8 +1856,8 @@ static const struct airoha_pinctrl_func_group phy3_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY3_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1890,8 +1871,8 @@ static const struct airoha_pinctrl_func_group phy3_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY3_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1905,8 +1886,8 @@ static const struct airoha_pinctrl_func_group phy3_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY3_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(2)
 		},
 		.regmap_size = 2,
 	},
@@ -1924,8 +1905,8 @@ static const struct airoha_pinctrl_func_group phy4_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN1_LED_MAPPING_MASK,
-			LAN1_PHY4_LED_MAP
+			LAN0_LED_MAPPING_MASK,
+			LAN0_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1939,8 +1920,8 @@ static const struct airoha_pinctrl_func_group phy4_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN2_LED_MAPPING_MASK,
-			LAN2_PHY4_LED_MAP
+			LAN1_LED_MAPPING_MASK,
+			LAN1_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1954,8 +1935,8 @@ static const struct airoha_pinctrl_func_group phy4_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN3_LED_MAPPING_MASK,
-			LAN3_PHY4_LED_MAP
+			LAN2_LED_MAPPING_MASK,
+			LAN2_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	}, {
@@ -1969,8 +1950,8 @@ static const struct airoha_pinctrl_func_group phy4_led1_func_group[] = {
 		.regmap[1] = {
 			AIROHA_FUNC_MUX,
 			REG_LAN_LED1_MAPPING,
-			LAN4_LED_MAPPING_MASK,
-			LAN4_PHY4_LED_MAP
+			LAN3_LED_MAPPING_MASK,
+			LAN3_PHY_LED_MAP(3)
 		},
 		.regmap_size = 2,
 	},
diff --git a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c
index 91edb539925a..8596f3541265 100644
--- a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c
+++ b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c
@@ -1015,9 +1015,15 @@ static int mtk_eint_init(struct mtk_pinctrl *pctl, struct platform_device *pdev)
 	if (!pctl->eint)
 		return -ENOMEM;
 
-	pctl->eint->base = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(pctl->eint->base))
-		return PTR_ERR(pctl->eint->base);
+	pctl->eint->nbase = 1;
+	/* mtk-eint expects an array */
+	pctl->eint->base = devm_kzalloc(pctl->dev, sizeof(pctl->eint->base), GFP_KERNEL);
+	if (!pctl->eint->base)
+		return -ENOMEM;
+
+	pctl->eint->base[0] = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(pctl->eint->base[0]))
+		return PTR_ERR(pctl->eint->base[0]);
 
 	pctl->eint->irq = irq_of_parse_and_map(np, 0);
 	if (!pctl->eint->irq)
diff --git a/drivers/pinctrl/meson/pinctrl-meson.c b/drivers/pinctrl/meson/pinctrl-meson.c
index 253a0cc57e39..e5a32a0532ee 100644
--- a/drivers/pinctrl/meson/pinctrl-meson.c
+++ b/drivers/pinctrl/meson/pinctrl-meson.c
@@ -487,7 +487,7 @@ static int meson_pinconf_get(struct pinctrl_dev *pcdev, unsigned int pin,
 	case PIN_CONFIG_BIAS_PULL_DOWN:
 	case PIN_CONFIG_BIAS_PULL_UP:
 		if (meson_pinconf_get_pull(pc, pin) == param)
-			arg = 1;
+			arg = 60000;
 		else
 			return -EINVAL;
 		break;
diff --git a/drivers/pinctrl/qcom/pinctrl-sm8750.c b/drivers/pinctrl/qcom/pinctrl-sm8750.c
index 1af11cd95fb0..b94fb4ee0ec3 100644
--- a/drivers/pinctrl/qcom/pinctrl-sm8750.c
+++ b/drivers/pinctrl/qcom/pinctrl-sm8750.c
@@ -46,7 +46,9 @@
 		.out_bit = 1,                                         \
 		.intr_enable_bit = 0,                                 \
 		.intr_status_bit = 0,                                 \
-		.intr_target_bit = 5,                                 \
+		.intr_wakeup_present_bit = 6,                         \
+		.intr_wakeup_enable_bit = 7,                          \
+		.intr_target_bit = 8,                                 \
 		.intr_target_kpss_val = 3,                            \
 		.intr_raw_status_bit = 4,                             \
 		.intr_polarity_bit = 1,                               \
diff --git a/drivers/platform/arm64/acer-aspire1-ec.c b/drivers/platform/arm64/acer-aspire1-ec.c
index 2df42406430d..438532a047e6 100644
--- a/drivers/platform/arm64/acer-aspire1-ec.c
+++ b/drivers/platform/arm64/acer-aspire1-ec.c
@@ -366,7 +366,8 @@ static const struct power_supply_desc aspire_ec_adp_psy_desc = {
  * USB-C DP Alt mode HPD.
  */
 
-static int aspire_ec_bridge_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags)
+static int aspire_ec_bridge_attach(struct drm_bridge *bridge, struct drm_encoder *encoder,
+				   enum drm_bridge_attach_flags flags)
 {
 	return flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR ? 0 : -EINVAL;
 }
@@ -451,9 +452,9 @@ static int aspire_ec_probe(struct i2c_client *client)
 	int ret;
 	u8 tmp;
 
-	ec = devm_kzalloc(dev, sizeof(*ec), GFP_KERNEL);
-	if (!ec)
-		return -ENOMEM;
+	ec = devm_drm_bridge_alloc(dev, struct aspire_ec, bridge, &aspire_ec_bridge_funcs);
+	if (IS_ERR(ec))
+		return PTR_ERR(ec);
 
 	ec->client = client;
 	i2c_set_clientdata(client, ec);
@@ -496,7 +497,6 @@ static int aspire_ec_probe(struct i2c_client *client)
 	fwnode = device_get_named_child_node(dev, "connector");
 	if (fwnode) {
 		INIT_WORK(&ec->work, aspire_ec_bridge_update_hpd_work);
-		ec->bridge.funcs = &aspire_ec_bridge_funcs;
 		ec->bridge.of_node = to_of_node(fwnode);
 		ec->bridge.ops = DRM_BRIDGE_OP_HPD;
 		ec->bridge.type = DRM_MODE_CONNECTOR_USB;
diff --git a/drivers/platform/mellanox/mlxbf-bootctl.c b/drivers/platform/mellanox/mlxbf-bootctl.c
index b95dcb8d483c..c18a5b96de5c 100644
--- a/drivers/platform/mellanox/mlxbf-bootctl.c
+++ b/drivers/platform/mellanox/mlxbf-bootctl.c
@@ -333,9 +333,9 @@ static ssize_t secure_boot_fuse_state_show(struct device *dev,
 			else
 				status = valid ? "Invalid" : "Free";
 		}
-		buf_len += sysfs_emit(buf + buf_len, "%d:%s ", key, status);
+		buf_len += sysfs_emit_at(buf, buf_len, "%d:%s ", key, status);
 	}
-	buf_len += sysfs_emit(buf + buf_len, "\n");
+	buf_len += sysfs_emit_at(buf, buf_len, "\n");
 
 	return buf_len;
 }
diff --git a/drivers/platform/x86/amd/pmc/pmc.c b/drivers/platform/x86/amd/pmc/pmc.c
index d789d6cab794..0329fafe14eb 100644
--- a/drivers/platform/x86/amd/pmc/pmc.c
+++ b/drivers/platform/x86/amd/pmc/pmc.c
@@ -644,10 +644,9 @@ static void amd_pmc_s2idle_check(void)
 	struct smu_metrics table;
 	int rc;
 
-	/* CZN: Ensure that future s0i3 entry attempts at least 10ms passed */
-	if (pdev->cpu_id == AMD_CPU_ID_CZN && !get_metrics_table(pdev, &table) &&
-	    table.s0i3_last_entry_status)
-		usleep_range(10000, 20000);
+	/* Avoid triggering OVP */
+	if (!get_metrics_table(pdev, &table) && table.s0i3_last_entry_status)
+		msleep(2500);
 
 	/* Dump the IdleMask before we add to the STB */
 	amd_pmc_idlemask_read(pdev, pdev->dev, NULL);
diff --git a/drivers/platform/x86/amd/pmf/auto-mode.c b/drivers/platform/x86/amd/pmf/auto-mode.c
index 02ff68be10d0..a184922bba8d 100644
--- a/drivers/platform/x86/amd/pmf/auto-mode.c
+++ b/drivers/platform/x86/amd/pmf/auto-mode.c
@@ -120,9 +120,9 @@ static void amd_pmf_set_automode(struct amd_pmf_dev *dev, int idx,
 	amd_pmf_send_cmd(dev, SET_SPPT_APU_ONLY, false, pwr_ctrl->sppt_apu_only, NULL);
 	amd_pmf_send_cmd(dev, SET_STT_MIN_LIMIT, false, pwr_ctrl->stt_min, NULL);
 	amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false,
-			 pwr_ctrl->stt_skin_temp[STT_TEMP_APU], NULL);
+			 fixp_q88_fromint(pwr_ctrl->stt_skin_temp[STT_TEMP_APU]), NULL);
 	amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false,
-			 pwr_ctrl->stt_skin_temp[STT_TEMP_HS2], NULL);
+			 fixp_q88_fromint(pwr_ctrl->stt_skin_temp[STT_TEMP_HS2]), NULL);
 
 	if (is_apmf_func_supported(dev, APMF_FUNC_SET_FAN_IDX))
 		apmf_update_fan_idx(dev, config_store.mode_set[idx].fan_control.manual,
diff --git a/drivers/platform/x86/amd/pmf/cnqf.c b/drivers/platform/x86/amd/pmf/cnqf.c
index bc8899e15c91..207a0b33d8d3 100644
--- a/drivers/platform/x86/amd/pmf/cnqf.c
+++ b/drivers/platform/x86/amd/pmf/cnqf.c
@@ -81,10 +81,10 @@ static int amd_pmf_set_cnqf(struct amd_pmf_dev *dev, int src, int idx,
 	amd_pmf_send_cmd(dev, SET_SPPT, false, pc->sppt, NULL);
 	amd_pmf_send_cmd(dev, SET_SPPT_APU_ONLY, false, pc->sppt_apu_only, NULL);
 	amd_pmf_send_cmd(dev, SET_STT_MIN_LIMIT, false, pc->stt_min, NULL);
-	amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false, pc->stt_skin_temp[STT_TEMP_APU],
-			 NULL);
-	amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false, pc->stt_skin_temp[STT_TEMP_HS2],
-			 NULL);
+	amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false,
+			 fixp_q88_fromint(pc->stt_skin_temp[STT_TEMP_APU]), NULL);
+	amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false,
+			 fixp_q88_fromint(pc->stt_skin_temp[STT_TEMP_HS2]), NULL);
 
 	if (is_apmf_func_supported(dev, APMF_FUNC_SET_FAN_IDX))
 		apmf_update_fan_idx(dev,
diff --git a/drivers/platform/x86/amd/pmf/core.c b/drivers/platform/x86/amd/pmf/core.c
index a2cb2d5544f5..96821101ec77 100644
--- a/drivers/platform/x86/amd/pmf/core.c
+++ b/drivers/platform/x86/amd/pmf/core.c
@@ -176,6 +176,20 @@ static void __maybe_unused amd_pmf_dump_registers(struct amd_pmf_dev *dev)
 	dev_dbg(dev->dev, "AMD_PMF_REGISTER_MESSAGE:%x\n", value);
 }
 
+/**
+ * fixp_q88_fromint: Convert integer to Q8.8
+ * @val: input value
+ *
+ * Converts an integer into binary fixed point format where 8 bits
+ * are used for integer and 8 bits are used for the decimal.
+ *
+ * Return: unsigned integer converted to Q8.8 format
+ */
+u32 fixp_q88_fromint(u32 val)
+{
+	return val << 8;
+}
+
 int amd_pmf_send_cmd(struct amd_pmf_dev *dev, u8 message, bool get, u32 arg, u32 *data)
 {
 	int rc;
diff --git a/drivers/platform/x86/amd/pmf/pmf.h b/drivers/platform/x86/amd/pmf/pmf.h
index e6bdee68ccf3..45b60238d527 100644
--- a/drivers/platform/x86/amd/pmf/pmf.h
+++ b/drivers/platform/x86/amd/pmf/pmf.h
@@ -777,6 +777,7 @@ int apmf_install_handler(struct amd_pmf_dev *pmf_dev);
 int apmf_os_power_slider_update(struct amd_pmf_dev *dev, u8 flag);
 int amd_pmf_set_dram_addr(struct amd_pmf_dev *dev, bool alloc_buffer);
 int amd_pmf_notify_sbios_heartbeat_event_v2(struct amd_pmf_dev *dev, u8 flag);
+u32 fixp_q88_fromint(u32 val);
 
 /* SPS Layer */
 int amd_pmf_get_pprof_modes(struct amd_pmf_dev *pmf);
diff --git a/drivers/platform/x86/amd/pmf/sps.c b/drivers/platform/x86/amd/pmf/sps.c
index d3083383f11f..49e14ca94a9e 100644
--- a/drivers/platform/x86/amd/pmf/sps.c
+++ b/drivers/platform/x86/amd/pmf/sps.c
@@ -198,9 +198,11 @@ static void amd_pmf_update_slider_v2(struct amd_pmf_dev *dev, int idx)
 	amd_pmf_send_cmd(dev, SET_STT_MIN_LIMIT, false,
 			 apts_config_store.val[idx].stt_min_limit, NULL);
 	amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false,
-			 apts_config_store.val[idx].stt_skin_temp_limit_apu, NULL);
+			 fixp_q88_fromint(apts_config_store.val[idx].stt_skin_temp_limit_apu),
+			 NULL);
 	amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false,
-			 apts_config_store.val[idx].stt_skin_temp_limit_hs2, NULL);
+			 fixp_q88_fromint(apts_config_store.val[idx].stt_skin_temp_limit_hs2),
+			 NULL);
 }
 
 void amd_pmf_update_slider(struct amd_pmf_dev *dev, bool op, int idx,
@@ -217,9 +219,11 @@ void amd_pmf_update_slider(struct amd_pmf_dev *dev, bool op, int idx,
 		amd_pmf_send_cmd(dev, SET_STT_MIN_LIMIT, false,
 				 config_store.prop[src][idx].stt_min, NULL);
 		amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false,
-				 config_store.prop[src][idx].stt_skin_temp[STT_TEMP_APU], NULL);
+				 fixp_q88_fromint(config_store.prop[src][idx].stt_skin_temp[STT_TEMP_APU]),
+				 NULL);
 		amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false,
-				 config_store.prop[src][idx].stt_skin_temp[STT_TEMP_HS2], NULL);
+				 fixp_q88_fromint(config_store.prop[src][idx].stt_skin_temp[STT_TEMP_HS2]),
+				 NULL);
 	} else if (op == SLIDER_OP_GET) {
 		amd_pmf_send_cmd(dev, GET_SPL, true, ARG_NONE, &table->prop[src][idx].spl);
 		amd_pmf_send_cmd(dev, GET_FPPT, true, ARG_NONE, &table->prop[src][idx].fppt);
diff --git a/drivers/platform/x86/amd/pmf/tee-if.c b/drivers/platform/x86/amd/pmf/tee-if.c
index a1e43873a07b..14b99d8b63d2 100644
--- a/drivers/platform/x86/amd/pmf/tee-if.c
+++ b/drivers/platform/x86/amd/pmf/tee-if.c
@@ -123,7 +123,8 @@ static void amd_pmf_apply_policies(struct amd_pmf_dev *dev, struct ta_pmf_enact_
 
 		case PMF_POLICY_STT_SKINTEMP_APU:
 			if (dev->prev_data->stt_skintemp_apu != val) {
-				amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false, val, NULL);
+				amd_pmf_send_cmd(dev, SET_STT_LIMIT_APU, false,
+						 fixp_q88_fromint(val), NULL);
 				dev_dbg(dev->dev, "update STT_SKINTEMP_APU: %u\n", val);
 				dev->prev_data->stt_skintemp_apu = val;
 			}
@@ -131,7 +132,8 @@ static void amd_pmf_apply_policies(struct amd_pmf_dev *dev, struct ta_pmf_enact_
 
 		case PMF_POLICY_STT_SKINTEMP_HS2:
 			if (dev->prev_data->stt_skintemp_hs2 != val) {
-				amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false, val, NULL);
+				amd_pmf_send_cmd(dev, SET_STT_LIMIT_HS2, false,
+						 fixp_q88_fromint(val), NULL);
 				dev_dbg(dev->dev, "update STT_SKINTEMP_HS2: %u\n", val);
 				dev->prev_data->stt_skintemp_hs2 = val;
 			}
diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index d460dd194f19..a0a411b4f2d6 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -426,11 +426,14 @@ static int asus_pega_lucid_set(struct asus_laptop *asus, int unit, bool enable)
 
 static int pega_acc_axis(struct asus_laptop *asus, int curr, char *method)
 {
+	unsigned long long val = (unsigned long long)curr;
+	acpi_status status;
 	int i, delta;
-	unsigned long long val;
-	for (i = 0; i < PEGA_ACC_RETRIES; i++) {
-		acpi_evaluate_integer(asus->handle, method, NULL, &val);
 
+	for (i = 0; i < PEGA_ACC_RETRIES; i++) {
+		status = acpi_evaluate_integer(asus->handle, method, NULL, &val);
+		if (ACPI_FAILURE(status))
+			continue;
 		/* The output is noisy.  From reading the ASL
 		 * dissassembly, timeout errors are returned with 1's
 		 * in the high word, and the lack of locking around
diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c
index 38ef778e8c19..0c697b46f436 100644
--- a/drivers/platform/x86/asus-wmi.c
+++ b/drivers/platform/x86/asus-wmi.c
@@ -304,6 +304,7 @@ struct asus_wmi {
 
 	u32 kbd_rgb_dev;
 	bool kbd_rgb_state_available;
+	bool oobe_state_available;
 
 	u8 throttle_thermal_policy_mode;
 	u32 throttle_thermal_policy_dev;
@@ -1826,7 +1827,7 @@ static int asus_wmi_led_init(struct asus_wmi *asus)
 			goto error;
 	}
 
-	if (asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_OOBE)) {
+	if (asus->oobe_state_available) {
 		/*
 		 * Disable OOBE state, so that e.g. the keyboard backlight
 		 * works.
@@ -4723,6 +4724,7 @@ static int asus_wmi_add(struct platform_device *pdev)
 	asus->egpu_enable_available = asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_EGPU);
 	asus->dgpu_disable_available = asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_DGPU);
 	asus->kbd_rgb_state_available = asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_TUF_RGB_STATE);
+	asus->oobe_state_available = asus_wmi_dev_is_present(asus, ASUS_WMI_DEVID_OOBE);
 	asus->ally_mcu_usb_switch = acpi_has_method(NULL, ASUS_USB0_PWR_EC0_CSEE)
 						&& dmi_check_system(asus_ally_mcu_quirk);
 
@@ -4970,6 +4972,13 @@ static int asus_hotk_restore(struct device *device)
 	}
 	if (!IS_ERR_OR_NULL(asus->kbd_led.dev))
 		kbd_led_update(asus);
+	if (asus->oobe_state_available) {
+		/*
+		 * Disable OOBE state, so that e.g. the keyboard backlight
+		 * works.
+		 */
+		asus_wmi_set_devstate(ASUS_WMI_DEVID_OOBE, 1, NULL);
+	}
 
 	if (asus_wmi_has_fnlock_key(asus))
 		asus_wmi_fnlock_update(asus);
diff --git a/drivers/platform/x86/dell/alienware-wmi-wmax.c b/drivers/platform/x86/dell/alienware-wmi-wmax.c
index 3d3014b5adf0..08b82c151e07 100644
--- a/drivers/platform/x86/dell/alienware-wmi-wmax.c
+++ b/drivers/platform/x86/dell/alienware-wmi-wmax.c
@@ -62,11 +62,43 @@ static struct awcc_quirks empty_quirks;
 
 static const struct dmi_system_id awcc_dmi_table[] __initconst = {
 	{
+		.ident = "Alienware Area-51m R2",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Alienware Area-51m R2"),
+		},
+		.driver_data = &generic_quirks,
+	},
+	{
+		.ident = "Alienware m15 R7",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Alienware m15 R7"),
+		},
+		.driver_data = &generic_quirks,
+	},
+	{
+		.ident = "Alienware m16 R1",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Alienware m16 R1"),
+		},
+		.driver_data = &g_series_quirks,
+	},
+	{
 		.ident = "Alienware m16 R1 AMD",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
 			DMI_MATCH(DMI_PRODUCT_NAME, "Alienware m16 R1 AMD"),
 		},
+		.driver_data = &g_series_quirks,
+	},
+	{
+		.ident = "Alienware m16 R2",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Alienware m16 R2"),
+		},
 		.driver_data = &generic_quirks,
 	},
 	{
@@ -94,6 +126,14 @@ static const struct dmi_system_id awcc_dmi_table[] __initconst = {
 		.driver_data = &generic_quirks,
 	},
 	{
+		.ident = "Alienware x15 R2",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Alienware x15 R2"),
+		},
+		.driver_data = &generic_quirks,
+	},
+	{
 		.ident = "Alienware x17 R2",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
@@ -126,6 +166,14 @@ static const struct dmi_system_id awcc_dmi_table[] __initconst = {
 		.driver_data = &g_series_quirks,
 	},
 	{
+		.ident = "Dell Inc. G16 7630",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Dell G16 7630"),
+		},
+		.driver_data = &g_series_quirks,
+	},
+	{
 		.ident = "Dell Inc. G3 3500",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
@@ -149,6 +197,14 @@ static const struct dmi_system_id awcc_dmi_table[] __initconst = {
 		},
 		.driver_data = &g_series_quirks,
 	},
+	{
+		.ident = "Dell Inc. G5 5505",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "G5 5505"),
+		},
+		.driver_data = &g_series_quirks,
+	},
 };
 
 enum WMAX_THERMAL_INFORMATION_OPERATIONS {
@@ -607,12 +663,10 @@ static int thermal_profile_probe(void *drvdata, unsigned long *choices)
 	for (u32 i = 0; i < sys_desc[3]; i++) {
 		ret = wmax_thermal_information(priv->wdev, WMAX_OPERATION_LIST_IDS,
 					       i + first_mode, &out_data);
-
-		if (ret == -EIO)
-			return ret;
-
 		if (ret == -EBADRQC)
 			break;
+		if (ret)
+			return ret;
 
 		if (!is_wmax_thermal_code(out_data))
 			continue;
diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c
index 17a09b7784ed..ede483573fe0 100644
--- a/drivers/platform/x86/ideapad-laptop.c
+++ b/drivers/platform/x86/ideapad-laptop.c
@@ -1294,6 +1294,16 @@ static const struct key_entry ideapad_keymap[] = {
 	/* Specific to some newer models */
 	{ KE_KEY,	0x3e | IDEAPAD_WMI_KEY, { KEY_MICMUTE } },
 	{ KE_KEY,	0x3f | IDEAPAD_WMI_KEY, { KEY_RFKILL } },
+	/* Star- (User Assignable Key) */
+	{ KE_KEY,	0x44 | IDEAPAD_WMI_KEY, { KEY_PROG1 } },
+	/* Eye */
+	{ KE_KEY,	0x45 | IDEAPAD_WMI_KEY, { KEY_PROG3 } },
+	/* Performance toggle also Fn+Q, handled inside ideapad_wmi_notify() */
+	{ KE_KEY,	0x3d | IDEAPAD_WMI_KEY, { KEY_PROG4 } },
+	/* shift + prtsc */
+	{ KE_KEY,   0x2d | IDEAPAD_WMI_KEY, { KEY_CUT } },
+	{ KE_KEY,   0x29 | IDEAPAD_WMI_KEY, { KEY_TOUCHPAD_TOGGLE } },
+	{ KE_KEY,   0x2a | IDEAPAD_WMI_KEY, { KEY_ROOT_MENU } },
 
 	{ KE_END },
 };
@@ -2080,6 +2090,12 @@ static void ideapad_wmi_notify(struct wmi_device *wdev, union acpi_object *data)
 		dev_dbg(&wdev->dev, "WMI fn-key event: 0x%llx\n",
 			data->integer.value);
 
+		/* performance button triggered by 0x3d */
+		if (data->integer.value == 0x3d && priv->dytc) {
+			platform_profile_cycle();
+			break;
+		}
+
 		/* 0x02 FnLock, 0x03 Esc */
 		if (data->integer.value == 0x02 || data->integer.value == 0x03)
 			ideapad_fn_lock_led_notify(priv, data->integer.value == 0x02);
diff --git a/drivers/platform/x86/intel/hid.c b/drivers/platform/x86/intel/hid.c
index 88a1a9ff2f34..0b5e43444ed6 100644
--- a/drivers/platform/x86/intel/hid.c
+++ b/drivers/platform/x86/intel/hid.c
@@ -44,16 +44,17 @@ MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Alex Hung");
 
 static const struct acpi_device_id intel_hid_ids[] = {
-	{"INT33D5", 0},
-	{"INTC1051", 0},
-	{"INTC1054", 0},
-	{"INTC1070", 0},
-	{"INTC1076", 0},
-	{"INTC1077", 0},
-	{"INTC1078", 0},
-	{"INTC107B", 0},
-	{"INTC10CB", 0},
-	{"", 0},
+	{ "INT33D5" },
+	{ "INTC1051" },
+	{ "INTC1054" },
+	{ "INTC1070" },
+	{ "INTC1076" },
+	{ "INTC1077" },
+	{ "INTC1078" },
+	{ "INTC107B" },
+	{ "INTC10CB" },
+	{ "INTC10CC" },
+	{ }
 };
 MODULE_DEVICE_TABLE(acpi, intel_hid_ids);
 
diff --git a/drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c b/drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c
index 40bbf8e45fa4..bdee5d00f30b 100644
--- a/drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c
+++ b/drivers/platform/x86/intel/uncore-frequency/uncore-frequency.c
@@ -146,15 +146,13 @@ static int uncore_event_cpu_online(unsigned int cpu)
 {
 	struct uncore_data *data;
 	int target;
+	int ret;
 
 	/* Check if there is an online cpu in the package for uncore MSR */
 	target = cpumask_any_and(&uncore_cpu_mask, topology_die_cpumask(cpu));
 	if (target < nr_cpu_ids)
 		return 0;
 
-	/* Use this CPU on this die as a control CPU */
-	cpumask_set_cpu(cpu, &uncore_cpu_mask);
-
 	data = uncore_get_instance(cpu);
 	if (!data)
 		return 0;
@@ -163,7 +161,14 @@ static int uncore_event_cpu_online(unsigned int cpu)
 	data->die_id = topology_die_id(cpu);
 	data->domain_id = UNCORE_DOMAIN_ID_INVALID;
 
-	return uncore_freq_add_entry(data, cpu);
+	ret = uncore_freq_add_entry(data, cpu);
+	if (ret)
+		return ret;
+
+	/* Use this CPU on this die as a control CPU */
+	cpumask_set_cpu(cpu, &uncore_cpu_mask);
+
+	return 0;
 }
 
 static int uncore_event_cpu_offline(unsigned int cpu)
diff --git a/drivers/platform/x86/msi-wmi-platform.c b/drivers/platform/x86/msi-wmi-platform.c
index 9b5c7f8c79b0..dc5e9878cb68 100644
--- a/drivers/platform/x86/msi-wmi-platform.c
+++ b/drivers/platform/x86/msi-wmi-platform.c
@@ -10,6 +10,7 @@
 #include <linux/acpi.h>
 #include <linux/bits.h>
 #include <linux/bitfield.h>
+#include <linux/cleanup.h>
 #include <linux/debugfs.h>
 #include <linux/device.h>
 #include <linux/device/driver.h>
@@ -17,6 +18,7 @@
 #include <linux/hwmon.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <linux/printk.h>
 #include <linux/rwsem.h>
 #include <linux/types.h>
@@ -76,8 +78,13 @@ enum msi_wmi_platform_method {
 	MSI_PLATFORM_GET_WMI		= 0x1d,
 };
 
-struct msi_wmi_platform_debugfs_data {
+struct msi_wmi_platform_data {
 	struct wmi_device *wdev;
+	struct mutex wmi_lock;	/* Necessary when calling WMI methods */
+};
+
+struct msi_wmi_platform_debugfs_data {
+	struct msi_wmi_platform_data *data;
 	enum msi_wmi_platform_method method;
 	struct rw_semaphore buffer_lock;	/* Protects debugfs buffer */
 	size_t length;
@@ -132,8 +139,9 @@ static int msi_wmi_platform_parse_buffer(union acpi_object *obj, u8 *output, siz
 	return 0;
 }
 
-static int msi_wmi_platform_query(struct wmi_device *wdev, enum msi_wmi_platform_method method,
-				  u8 *input, size_t input_length, u8 *output, size_t output_length)
+static int msi_wmi_platform_query(struct msi_wmi_platform_data *data,
+				  enum msi_wmi_platform_method method, u8 *input,
+				  size_t input_length, u8 *output, size_t output_length)
 {
 	struct acpi_buffer out = { ACPI_ALLOCATE_BUFFER, NULL };
 	struct acpi_buffer in = {
@@ -147,9 +155,15 @@ static int msi_wmi_platform_query(struct wmi_device *wdev, enum msi_wmi_platform
 	if (!input_length || !output_length)
 		return -EINVAL;
 
-	status = wmidev_evaluate_method(wdev, 0x0, method, &in, &out);
-	if (ACPI_FAILURE(status))
-		return -EIO;
+	/*
+	 * The ACPI control method responsible for handling the WMI method calls
+	 * is not thread-safe. Because of this we have to do the locking ourself.
+	 */
+	scoped_guard(mutex, &data->wmi_lock) {
+		status = wmidev_evaluate_method(data->wdev, 0x0, method, &in, &out);
+		if (ACPI_FAILURE(status))
+			return -EIO;
+	}
 
 	obj = out.pointer;
 	if (!obj)
@@ -170,22 +184,22 @@ static umode_t msi_wmi_platform_is_visible(const void *drvdata, enum hwmon_senso
 static int msi_wmi_platform_read(struct device *dev, enum hwmon_sensor_types type, u32 attr,
 				 int channel, long *val)
 {
-	struct wmi_device *wdev = dev_get_drvdata(dev);
+	struct msi_wmi_platform_data *data = dev_get_drvdata(dev);
 	u8 input[32] = { 0 };
 	u8 output[32];
-	u16 data;
+	u16 value;
 	int ret;
 
-	ret = msi_wmi_platform_query(wdev, MSI_PLATFORM_GET_FAN, input, sizeof(input), output,
+	ret = msi_wmi_platform_query(data, MSI_PLATFORM_GET_FAN, input, sizeof(input), output,
 				     sizeof(output));
 	if (ret < 0)
 		return ret;
 
-	data = get_unaligned_be16(&output[channel * 2 + 1]);
-	if (!data)
+	value = get_unaligned_be16(&output[channel * 2 + 1]);
+	if (!value)
 		*val = 0;
 	else
-		*val = 480000 / data;
+		*val = 480000 / value;
 
 	return 0;
 }
@@ -231,7 +245,7 @@ static ssize_t msi_wmi_platform_write(struct file *fp, const char __user *input,
 		return ret;
 
 	down_write(&data->buffer_lock);
-	ret = msi_wmi_platform_query(data->wdev, data->method, payload, data->length, data->buffer,
+	ret = msi_wmi_platform_query(data->data, data->method, payload, data->length, data->buffer,
 				     data->length);
 	up_write(&data->buffer_lock);
 
@@ -277,17 +291,17 @@ static void msi_wmi_platform_debugfs_remove(void *data)
 	debugfs_remove_recursive(dir);
 }
 
-static void msi_wmi_platform_debugfs_add(struct wmi_device *wdev, struct dentry *dir,
+static void msi_wmi_platform_debugfs_add(struct msi_wmi_platform_data *drvdata, struct dentry *dir,
 					 const char *name, enum msi_wmi_platform_method method)
 {
 	struct msi_wmi_platform_debugfs_data *data;
 	struct dentry *entry;
 
-	data = devm_kzalloc(&wdev->dev, sizeof(*data), GFP_KERNEL);
+	data = devm_kzalloc(&drvdata->wdev->dev, sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return;
 
-	data->wdev = wdev;
+	data->data = drvdata;
 	data->method = method;
 	init_rwsem(&data->buffer_lock);
 
@@ -298,82 +312,82 @@ static void msi_wmi_platform_debugfs_add(struct wmi_device *wdev, struct dentry
 
 	entry = debugfs_create_file(name, 0600, dir, data, &msi_wmi_platform_debugfs_fops);
 	if (IS_ERR(entry))
-		devm_kfree(&wdev->dev, data);
+		devm_kfree(&drvdata->wdev->dev, data);
 }
 
-static void msi_wmi_platform_debugfs_init(struct wmi_device *wdev)
+static void msi_wmi_platform_debugfs_init(struct msi_wmi_platform_data *data)
 {
 	struct dentry *dir;
 	char dir_name[64];
 	int ret, method;
 
-	scnprintf(dir_name, ARRAY_SIZE(dir_name), "%s-%s", DRIVER_NAME, dev_name(&wdev->dev));
+	scnprintf(dir_name, ARRAY_SIZE(dir_name), "%s-%s", DRIVER_NAME, dev_name(&data->wdev->dev));
 
 	dir = debugfs_create_dir(dir_name, NULL);
 	if (IS_ERR(dir))
 		return;
 
-	ret = devm_add_action_or_reset(&wdev->dev, msi_wmi_platform_debugfs_remove, dir);
+	ret = devm_add_action_or_reset(&data->wdev->dev, msi_wmi_platform_debugfs_remove, dir);
 	if (ret < 0)
 		return;
 
 	for (method = MSI_PLATFORM_GET_PACKAGE; method <= MSI_PLATFORM_GET_WMI; method++)
-		msi_wmi_platform_debugfs_add(wdev, dir, msi_wmi_platform_debugfs_names[method - 1],
+		msi_wmi_platform_debugfs_add(data, dir, msi_wmi_platform_debugfs_names[method - 1],
 					     method);
 }
 
-static int msi_wmi_platform_hwmon_init(struct wmi_device *wdev)
+static int msi_wmi_platform_hwmon_init(struct msi_wmi_platform_data *data)
 {
 	struct device *hdev;
 
-	hdev = devm_hwmon_device_register_with_info(&wdev->dev, "msi_wmi_platform", wdev,
+	hdev = devm_hwmon_device_register_with_info(&data->wdev->dev, "msi_wmi_platform", data,
 						    &msi_wmi_platform_chip_info, NULL);
 
 	return PTR_ERR_OR_ZERO(hdev);
 }
 
-static int msi_wmi_platform_ec_init(struct wmi_device *wdev)
+static int msi_wmi_platform_ec_init(struct msi_wmi_platform_data *data)
 {
 	u8 input[32] = { 0 };
 	u8 output[32];
 	u8 flags;
 	int ret;
 
-	ret = msi_wmi_platform_query(wdev, MSI_PLATFORM_GET_EC, input, sizeof(input), output,
+	ret = msi_wmi_platform_query(data, MSI_PLATFORM_GET_EC, input, sizeof(input), output,
 				     sizeof(output));
 	if (ret < 0)
 		return ret;
 
 	flags = output[MSI_PLATFORM_EC_FLAGS_OFFSET];
 
-	dev_dbg(&wdev->dev, "EC RAM version %lu.%lu\n",
+	dev_dbg(&data->wdev->dev, "EC RAM version %lu.%lu\n",
 		FIELD_GET(MSI_PLATFORM_EC_MAJOR_MASK, flags),
 		FIELD_GET(MSI_PLATFORM_EC_MINOR_MASK, flags));
-	dev_dbg(&wdev->dev, "EC firmware version %.28s\n",
+	dev_dbg(&data->wdev->dev, "EC firmware version %.28s\n",
 		&output[MSI_PLATFORM_EC_VERSION_OFFSET]);
 
 	if (!(flags & MSI_PLATFORM_EC_IS_TIGERLAKE)) {
 		if (!force)
 			return -ENODEV;
 
-		dev_warn(&wdev->dev, "Loading on a non-Tigerlake platform\n");
+		dev_warn(&data->wdev->dev, "Loading on a non-Tigerlake platform\n");
 	}
 
 	return 0;
 }
 
-static int msi_wmi_platform_init(struct wmi_device *wdev)
+static int msi_wmi_platform_init(struct msi_wmi_platform_data *data)
 {
 	u8 input[32] = { 0 };
 	u8 output[32];
 	int ret;
 
-	ret = msi_wmi_platform_query(wdev, MSI_PLATFORM_GET_WMI, input, sizeof(input), output,
+	ret = msi_wmi_platform_query(data, MSI_PLATFORM_GET_WMI, input, sizeof(input), output,
 				     sizeof(output));
 	if (ret < 0)
 		return ret;
 
-	dev_dbg(&wdev->dev, "WMI interface version %u.%u\n",
+	dev_dbg(&data->wdev->dev, "WMI interface version %u.%u\n",
 		output[MSI_PLATFORM_WMI_MAJOR_OFFSET],
 		output[MSI_PLATFORM_WMI_MINOR_OFFSET]);
 
@@ -381,7 +395,8 @@ static int msi_wmi_platform_init(struct wmi_device *wdev)
 		if (!force)
 			return -ENODEV;
 
-		dev_warn(&wdev->dev, "Loading despite unsupported WMI interface version (%u.%u)\n",
+		dev_warn(&data->wdev->dev,
+			 "Loading despite unsupported WMI interface version (%u.%u)\n",
 			 output[MSI_PLATFORM_WMI_MAJOR_OFFSET],
 			 output[MSI_PLATFORM_WMI_MINOR_OFFSET]);
 	}
@@ -391,19 +406,31 @@ static int msi_wmi_platform_init(struct wmi_device *wdev)
 
 static int msi_wmi_platform_probe(struct wmi_device *wdev, const void *context)
 {
+	struct msi_wmi_platform_data *data;
 	int ret;
 
-	ret = msi_wmi_platform_init(wdev);
+	data = devm_kzalloc(&wdev->dev, sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	data->wdev = wdev;
+	dev_set_drvdata(&wdev->dev, data);
+
+	ret = devm_mutex_init(&wdev->dev, &data->wmi_lock);
+	if (ret < 0)
+		return ret;
+
+	ret = msi_wmi_platform_init(data);
 	if (ret < 0)
 		return ret;
 
-	ret = msi_wmi_platform_ec_init(wdev);
+	ret = msi_wmi_platform_ec_init(data);
 	if (ret < 0)
 		return ret;
 
-	msi_wmi_platform_debugfs_init(wdev);
+	msi_wmi_platform_debugfs_init(data);
 
-	return msi_wmi_platform_hwmon_init(wdev);
+	return msi_wmi_platform_hwmon_init(data);
 }
 
 static const struct wmi_device_id msi_wmi_platform_id_table[] = {
diff --git a/drivers/platform/x86/x86-android-tablets/dmi.c b/drivers/platform/x86/x86-android-tablets/dmi.c
index 3e5fa3b6e2fd..278c6d151dc4 100644
--- a/drivers/platform/x86/x86-android-tablets/dmi.c
+++ b/drivers/platform/x86/x86-android-tablets/dmi.c
@@ -180,6 +180,18 @@ const struct dmi_system_id x86_android_tablet_ids[] __initconst = {
 		.driver_data = (void *)&peaq_c1010_info,
 	},
 	{
+		/* Vexia Edu Atla 10 tablet 5V version */
+		.matches = {
+			/* Having all 3 of these not set is somewhat unique */
+			DMI_MATCH(DMI_SYS_VENDOR, "To be filled by O.E.M."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "To be filled by O.E.M."),
+			DMI_MATCH(DMI_BOARD_NAME, "To be filled by O.E.M."),
+			/* Above strings are too generic, also match on BIOS date */
+			DMI_MATCH(DMI_BIOS_DATE, "05/14/2015"),
+		},
+		.driver_data = (void *)&vexia_edu_atla10_5v_info,
+	},
+	{
 		/* Vexia Edu Atla 10 tablet 9V version */
 		.matches = {
 			DMI_MATCH(DMI_BOARD_VENDOR, "AMI Corporation"),
@@ -187,7 +199,7 @@ const struct dmi_system_id x86_android_tablet_ids[] __initconst = {
 			/* Above strings are too generic, also match on BIOS date */
 			DMI_MATCH(DMI_BIOS_DATE, "08/25/2014"),
 		},
-		.driver_data = (void *)&vexia_edu_atla10_info,
+		.driver_data = (void *)&vexia_edu_atla10_9v_info,
 	},
 	{
 		/* Whitelabel (sold as various brands) TM800A550L */
diff --git a/drivers/platform/x86/x86-android-tablets/other.c b/drivers/platform/x86/x86-android-tablets/other.c
index 1d93d9edb23f..f7bd9f863c85 100644
--- a/drivers/platform/x86/x86-android-tablets/other.c
+++ b/drivers/platform/x86/x86-android-tablets/other.c
@@ -599,62 +599,122 @@ const struct x86_dev_info whitelabel_tm800a550l_info __initconst = {
 };
 
 /*
- * Vexia EDU ATLA 10 tablet, Android 4.2 / 4.4 + Guadalinex Ubuntu tablet
+ * Vexia EDU ATLA 10 tablet 5V, Android 4.4 + Guadalinex Ubuntu tablet
+ * distributed to schools in the Spanish Andalucía region.
+ */
+static const struct property_entry vexia_edu_atla10_5v_touchscreen_props[] = {
+	PROPERTY_ENTRY_U32("hid-descr-addr", 0x0000),
+	PROPERTY_ENTRY_U32("post-reset-deassert-delay-ms", 120),
+	{ }
+};
+
+static const struct software_node vexia_edu_atla10_5v_touchscreen_node = {
+	.properties = vexia_edu_atla10_5v_touchscreen_props,
+};
+
+static const struct x86_i2c_client_info vexia_edu_atla10_5v_i2c_clients[] __initconst = {
+	{
+		/* kxcjk1013 accelerometer */
+		.board_info = {
+			.type = "kxcjk1013",
+			.addr = 0x0f,
+			.dev_name = "kxcjk1013",
+		},
+		.adapter_path = "\\_SB_.I2C3",
+	}, {
+		/*  touchscreen controller */
+		.board_info = {
+			.type = "hid-over-i2c",
+			.addr = 0x38,
+			.dev_name = "FTSC1000",
+			.swnode = &vexia_edu_atla10_5v_touchscreen_node,
+		},
+		.adapter_path = "\\_SB_.I2C4",
+		.irq_data = {
+			.type = X86_ACPI_IRQ_TYPE_APIC,
+			.index = 0x44,
+			.trigger = ACPI_LEVEL_SENSITIVE,
+			.polarity = ACPI_ACTIVE_HIGH,
+		},
+	}
+};
+
+static struct gpiod_lookup_table vexia_edu_atla10_5v_ft5416_gpios = {
+	.dev_id = "i2c-FTSC1000",
+	.table = {
+		GPIO_LOOKUP("INT33FC:01", 26, "reset", GPIO_ACTIVE_LOW),
+		{ }
+	},
+};
+
+static struct gpiod_lookup_table * const vexia_edu_atla10_5v_gpios[] = {
+	&vexia_edu_atla10_5v_ft5416_gpios,
+	NULL
+};
+
+const struct x86_dev_info vexia_edu_atla10_5v_info __initconst = {
+	.i2c_client_info = vexia_edu_atla10_5v_i2c_clients,
+	.i2c_client_count = ARRAY_SIZE(vexia_edu_atla10_5v_i2c_clients),
+	.gpiod_lookup_tables = vexia_edu_atla10_5v_gpios,
+};
+
+/*
+ * Vexia EDU ATLA 10 tablet 9V, Android 4.2 + Guadalinex Ubuntu tablet
  * distributed to schools in the Spanish Andalucía region.
  */
 static const char * const crystal_cove_pwrsrc_psy[] = { "crystal_cove_pwrsrc" };
 
-static const struct property_entry vexia_edu_atla10_ulpmc_props[] = {
+static const struct property_entry vexia_edu_atla10_9v_ulpmc_props[] = {
 	PROPERTY_ENTRY_STRING_ARRAY("supplied-from", crystal_cove_pwrsrc_psy),
 	{ }
 };
 
-static const struct software_node vexia_edu_atla10_ulpmc_node = {
-	.properties = vexia_edu_atla10_ulpmc_props,
+static const struct software_node vexia_edu_atla10_9v_ulpmc_node = {
+	.properties = vexia_edu_atla10_9v_ulpmc_props,
 };
 
-static const char * const vexia_edu_atla10_accel_mount_matrix[] = {
+static const char * const vexia_edu_atla10_9v_accel_mount_matrix[] = {
 	"0", "-1", "0",
 	"1", "0", "0",
 	"0", "0", "1"
 };
 
-static const struct property_entry vexia_edu_atla10_accel_props[] = {
-	PROPERTY_ENTRY_STRING_ARRAY("mount-matrix", vexia_edu_atla10_accel_mount_matrix),
+static const struct property_entry vexia_edu_atla10_9v_accel_props[] = {
+	PROPERTY_ENTRY_STRING_ARRAY("mount-matrix", vexia_edu_atla10_9v_accel_mount_matrix),
 	{ }
 };
 
-static const struct software_node vexia_edu_atla10_accel_node = {
-	.properties = vexia_edu_atla10_accel_props,
+static const struct software_node vexia_edu_atla10_9v_accel_node = {
+	.properties = vexia_edu_atla10_9v_accel_props,
 };
 
-static const struct property_entry vexia_edu_atla10_touchscreen_props[] = {
+static const struct property_entry vexia_edu_atla10_9v_touchscreen_props[] = {
 	PROPERTY_ENTRY_U32("hid-descr-addr", 0x0000),
 	PROPERTY_ENTRY_U32("post-reset-deassert-delay-ms", 120),
 	{ }
 };
 
-static const struct software_node vexia_edu_atla10_touchscreen_node = {
-	.properties = vexia_edu_atla10_touchscreen_props,
+static const struct software_node vexia_edu_atla10_9v_touchscreen_node = {
+	.properties = vexia_edu_atla10_9v_touchscreen_props,
 };
 
-static const struct property_entry vexia_edu_atla10_pmic_props[] = {
+static const struct property_entry vexia_edu_atla10_9v_pmic_props[] = {
 	PROPERTY_ENTRY_BOOL("linux,register-pwrsrc-power_supply"),
 	{ }
 };
 
-static const struct software_node vexia_edu_atla10_pmic_node = {
-	.properties = vexia_edu_atla10_pmic_props,
+static const struct software_node vexia_edu_atla10_9v_pmic_node = {
+	.properties = vexia_edu_atla10_9v_pmic_props,
 };
 
-static const struct x86_i2c_client_info vexia_edu_atla10_i2c_clients[] __initconst = {
+static const struct x86_i2c_client_info vexia_edu_atla10_9v_i2c_clients[] __initconst = {
 	{
 		/* I2C attached embedded controller, used to access fuel-gauge */
 		.board_info = {
 			.type = "vexia_atla10_ec",
 			.addr = 0x76,
 			.dev_name = "ulpmc",
-			.swnode = &vexia_edu_atla10_ulpmc_node,
+			.swnode = &vexia_edu_atla10_9v_ulpmc_node,
 		},
 		.adapter_path = "0000:00:18.1",
 	}, {
@@ -679,7 +739,7 @@ static const struct x86_i2c_client_info vexia_edu_atla10_i2c_clients[] __initcon
 			.type = "kxtj21009",
 			.addr = 0x0f,
 			.dev_name = "kxtj21009",
-			.swnode = &vexia_edu_atla10_accel_node,
+			.swnode = &vexia_edu_atla10_9v_accel_node,
 		},
 		.adapter_path = "0000:00:18.5",
 	}, {
@@ -688,7 +748,7 @@ static const struct x86_i2c_client_info vexia_edu_atla10_i2c_clients[] __initcon
 			.type = "hid-over-i2c",
 			.addr = 0x38,
 			.dev_name = "FTSC1000",
-			.swnode = &vexia_edu_atla10_touchscreen_node,
+			.swnode = &vexia_edu_atla10_9v_touchscreen_node,
 		},
 		.adapter_path = "0000:00:18.6",
 		.irq_data = {
@@ -703,7 +763,7 @@ static const struct x86_i2c_client_info vexia_edu_atla10_i2c_clients[] __initcon
 			.type = "intel_soc_pmic_crc",
 			.addr = 0x6e,
 			.dev_name = "intel_soc_pmic_crc",
-			.swnode = &vexia_edu_atla10_pmic_node,
+			.swnode = &vexia_edu_atla10_9v_pmic_node,
 		},
 		.adapter_path = "0000:00:18.7",
 		.irq_data = {
@@ -715,7 +775,7 @@ static const struct x86_i2c_client_info vexia_edu_atla10_i2c_clients[] __initcon
 	}
 };
 
-static const struct x86_serdev_info vexia_edu_atla10_serdevs[] __initconst = {
+static const struct x86_serdev_info vexia_edu_atla10_9v_serdevs[] __initconst = {
 	{
 		.ctrl.pci.devfn = PCI_DEVFN(0x1e, 3),
 		.ctrl_devname = "serial0",
@@ -723,7 +783,7 @@ static const struct x86_serdev_info vexia_edu_atla10_serdevs[] __initconst = {
 	},
 };
 
-static struct gpiod_lookup_table vexia_edu_atla10_ft5416_gpios = {
+static struct gpiod_lookup_table vexia_edu_atla10_9v_ft5416_gpios = {
 	.dev_id = "i2c-FTSC1000",
 	.table = {
 		GPIO_LOOKUP("INT33FC:00", 60, "reset", GPIO_ACTIVE_LOW),
@@ -731,12 +791,12 @@ static struct gpiod_lookup_table vexia_edu_atla10_ft5416_gpios = {
 	},
 };
 
-static struct gpiod_lookup_table * const vexia_edu_atla10_gpios[] = {
-	&vexia_edu_atla10_ft5416_gpios,
+static struct gpiod_lookup_table * const vexia_edu_atla10_9v_gpios[] = {
+	&vexia_edu_atla10_9v_ft5416_gpios,
 	NULL
 };
 
-static int __init vexia_edu_atla10_init(struct device *dev)
+static int __init vexia_edu_atla10_9v_init(struct device *dev)
 {
 	struct pci_dev *pdev;
 	int ret;
@@ -760,13 +820,13 @@ static int __init vexia_edu_atla10_init(struct device *dev)
 	return 0;
 }
 
-const struct x86_dev_info vexia_edu_atla10_info __initconst = {
-	.i2c_client_info = vexia_edu_atla10_i2c_clients,
-	.i2c_client_count = ARRAY_SIZE(vexia_edu_atla10_i2c_clients),
-	.serdev_info = vexia_edu_atla10_serdevs,
-	.serdev_count = ARRAY_SIZE(vexia_edu_atla10_serdevs),
-	.gpiod_lookup_tables = vexia_edu_atla10_gpios,
-	.init = vexia_edu_atla10_init,
+const struct x86_dev_info vexia_edu_atla10_9v_info __initconst = {
+	.i2c_client_info = vexia_edu_atla10_9v_i2c_clients,
+	.i2c_client_count = ARRAY_SIZE(vexia_edu_atla10_9v_i2c_clients),
+	.serdev_info = vexia_edu_atla10_9v_serdevs,
+	.serdev_count = ARRAY_SIZE(vexia_edu_atla10_9v_serdevs),
+	.gpiod_lookup_tables = vexia_edu_atla10_9v_gpios,
+	.init = vexia_edu_atla10_9v_init,
 	.use_pci = true,
 };
 
diff --git a/drivers/platform/x86/x86-android-tablets/x86-android-tablets.h b/drivers/platform/x86/x86-android-tablets/x86-android-tablets.h
index 63a38a0069ba..dcf8d49e3b5f 100644
--- a/drivers/platform/x86/x86-android-tablets/x86-android-tablets.h
+++ b/drivers/platform/x86/x86-android-tablets/x86-android-tablets.h
@@ -127,7 +127,8 @@ extern const struct x86_dev_info nextbook_ares8_info;
 extern const struct x86_dev_info nextbook_ares8a_info;
 extern const struct x86_dev_info peaq_c1010_info;
 extern const struct x86_dev_info whitelabel_tm800a550l_info;
-extern const struct x86_dev_info vexia_edu_atla10_info;
+extern const struct x86_dev_info vexia_edu_atla10_5v_info;
+extern const struct x86_dev_info vexia_edu_atla10_9v_info;
 extern const struct x86_dev_info xiaomi_mipad2_info;
 extern const struct dmi_system_id x86_android_tablet_ids[];
 
diff --git a/drivers/power/supply/power_supply_sysfs.c b/drivers/power/supply/power_supply_sysfs.c
index edb058c19c9c..439dd0bf8644 100644
--- a/drivers/power/supply/power_supply_sysfs.c
+++ b/drivers/power/supply/power_supply_sysfs.c
@@ -33,7 +33,7 @@ struct power_supply_attr {
 [POWER_SUPPLY_PROP_ ## _name] =			\
 {						\
 	.prop_name = #_name,			\
-	.attr_name = #_name "\0",		\
+	.attr_name = #_name,			\
 	.text_values = _text,			\
 	.text_values_len = _len,		\
 }
diff --git a/drivers/pps/generators/pps_gen_tio.c b/drivers/pps/generators/pps_gen_tio.c
index 1d5ffe055463..de00a85bfafa 100644
--- a/drivers/pps/generators/pps_gen_tio.c
+++ b/drivers/pps/generators/pps_gen_tio.c
@@ -230,7 +230,7 @@ static int pps_gen_tio_probe(struct platform_device *pdev)
 	hrtimer_setup(&tio->timer, hrtimer_callback, CLOCK_REALTIME,
 		      HRTIMER_MODE_ABS);
 	spin_lock_init(&tio->lock);
-	platform_set_drvdata(pdev, &tio);
+	platform_set_drvdata(pdev, tio);
 
 	return 0;
 }
diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c
index 7945c6be1f7c..2ccdca4f6960 100644
--- a/drivers/ptp/ptp_ocp.c
+++ b/drivers/ptp/ptp_ocp.c
@@ -2067,6 +2067,7 @@ ptp_ocp_signal_set(struct ptp_ocp *bp, int gen, struct ptp_ocp_signal *s)
 	if (!s->start) {
 		/* roundup() does not work on 32-bit systems */
 		s->start = DIV64_U64_ROUND_UP(start_ns, s->period);
+		s->start *= s->period;
 		s->start = ktime_add(s->start, s->phase);
 	}
 
@@ -2577,12 +2578,60 @@ static const struct ocp_sma_op ocp_fb_sma_op = {
 	.set_output	= ptp_ocp_sma_fb_set_output,
 };
 
+static int
+ptp_ocp_sma_adva_set_output(struct ptp_ocp *bp, int sma_nr, u32 val)
+{
+	u32 reg, mask, shift;
+	unsigned long flags;
+	u32 __iomem *gpio;
+
+	gpio = sma_nr > 2 ? &bp->sma_map1->gpio2 : &bp->sma_map2->gpio2;
+	shift = sma_nr & 1 ? 0 : 16;
+
+	mask = 0xffff << (16 - shift);
+
+	spin_lock_irqsave(&bp->lock, flags);
+
+	reg = ioread32(gpio);
+	reg = (reg & mask) | (val << shift);
+
+	iowrite32(reg, gpio);
+
+	spin_unlock_irqrestore(&bp->lock, flags);
+
+	return 0;
+}
+
+static int
+ptp_ocp_sma_adva_set_inputs(struct ptp_ocp *bp, int sma_nr, u32 val)
+{
+	u32 reg, mask, shift;
+	unsigned long flags;
+	u32 __iomem *gpio;
+
+	gpio = sma_nr > 2 ? &bp->sma_map2->gpio1 : &bp->sma_map1->gpio1;
+	shift = sma_nr & 1 ? 0 : 16;
+
+	mask = 0xffff << (16 - shift);
+
+	spin_lock_irqsave(&bp->lock, flags);
+
+	reg = ioread32(gpio);
+	reg = (reg & mask) | (val << shift);
+
+	iowrite32(reg, gpio);
+
+	spin_unlock_irqrestore(&bp->lock, flags);
+
+	return 0;
+}
+
 static const struct ocp_sma_op ocp_adva_sma_op = {
 	.tbl		= { ptp_ocp_adva_sma_in, ptp_ocp_adva_sma_out },
 	.init		= ptp_ocp_sma_fb_init,
 	.get		= ptp_ocp_sma_fb_get,
-	.set_inputs	= ptp_ocp_sma_fb_set_inputs,
-	.set_output	= ptp_ocp_sma_fb_set_output,
+	.set_inputs	= ptp_ocp_sma_adva_set_inputs,
+	.set_output	= ptp_ocp_sma_adva_set_output,
 };
 
 static int
diff --git a/drivers/pwm/core.c b/drivers/pwm/core.c
index a40c511e0096..0387bd838487 100644
--- a/drivers/pwm/core.c
+++ b/drivers/pwm/core.c
@@ -322,7 +322,7 @@ static int __pwm_set_waveform(struct pwm_device *pwm,
 	const struct pwm_ops *ops = chip->ops;
 	char wfhw[WFHWSIZE];
 	struct pwm_waveform wf_rounded;
-	int err;
+	int err, ret_tohw;
 
 	BUG_ON(WFHWSIZE < ops->sizeof_wfhw);
 
@@ -332,16 +332,16 @@ static int __pwm_set_waveform(struct pwm_device *pwm,
 	if (!pwm_wf_valid(wf))
 		return -EINVAL;
 
-	err = __pwm_round_waveform_tohw(chip, pwm, wf, &wfhw);
-	if (err)
-		return err;
+	ret_tohw = __pwm_round_waveform_tohw(chip, pwm, wf, &wfhw);
+	if (ret_tohw < 0)
+		return ret_tohw;
 
 	if ((IS_ENABLED(CONFIG_PWM_DEBUG) || exact) && wf->period_length_ns) {
 		err = __pwm_round_waveform_fromhw(chip, pwm, &wfhw, &wf_rounded);
 		if (err)
 			return err;
 
-		if (IS_ENABLED(CONFIG_PWM_DEBUG) && !pwm_check_rounding(wf, &wf_rounded))
+		if (IS_ENABLED(CONFIG_PWM_DEBUG) && ret_tohw == 0 && !pwm_check_rounding(wf, &wf_rounded))
 			dev_err(&chip->dev, "Wrong rounding: requested %llu/%llu [+%llu], result %llu/%llu [+%llu]\n",
 				wf->duty_length_ns, wf->period_length_ns, wf->duty_offset_ns,
 				wf_rounded.duty_length_ns, wf_rounded.period_length_ns, wf_rounded.duty_offset_ns);
@@ -382,7 +382,8 @@ static int __pwm_set_waveform(struct pwm_device *pwm,
 				wf_rounded.duty_length_ns, wf_rounded.period_length_ns, wf_rounded.duty_offset_ns,
 				wf_set.duty_length_ns, wf_set.period_length_ns, wf_set.duty_offset_ns);
 	}
-	return 0;
+
+	return ret_tohw;
 }
 
 /**
diff --git a/drivers/pwm/pwm-axi-pwmgen.c b/drivers/pwm/pwm-axi-pwmgen.c
index 4259a0db9ff4..4337c8f5acf0 100644
--- a/drivers/pwm/pwm-axi-pwmgen.c
+++ b/drivers/pwm/pwm-axi-pwmgen.c
@@ -75,6 +75,7 @@ static int axi_pwmgen_round_waveform_tohw(struct pwm_chip *chip,
 {
 	struct axi_pwmgen_waveform *wfhw = _wfhw;
 	struct axi_pwmgen_ddata *ddata = axi_pwmgen_ddata_from_chip(chip);
+	int ret = 0;
 
 	if (wf->period_length_ns == 0) {
 		*wfhw = (struct axi_pwmgen_waveform){
@@ -91,12 +92,15 @@ static int axi_pwmgen_round_waveform_tohw(struct pwm_chip *chip,
 		if (wfhw->period_cnt == 0) {
 			/*
 			 * The specified period is too short for the hardware.
-			 * Let's round .duty_cycle down to 0 to get a (somewhat)
-			 * valid result.
+			 * So round up .period_cnt to 1 (i.e. the smallest
+			 * possible period). With .duty_cycle and .duty_offset
+			 * being less than or equal to .period, their rounded
+			 * value must be 0.
 			 */
 			wfhw->period_cnt = 1;
 			wfhw->duty_cycle_cnt = 0;
 			wfhw->duty_offset_cnt = 0;
+			ret = 1;
 		} else {
 			wfhw->duty_cycle_cnt = min_t(u64,
 						     mul_u64_u32_div(wf->duty_length_ns, ddata->clk_rate_hz, NSEC_PER_SEC),
@@ -111,7 +115,7 @@ static int axi_pwmgen_round_waveform_tohw(struct pwm_chip *chip,
 		pwm->hwpwm, wf->duty_length_ns, wf->period_length_ns, wf->duty_offset_ns,
 		ddata->clk_rate_hz, wfhw->period_cnt, wfhw->duty_cycle_cnt, wfhw->duty_offset_cnt);
 
-	return 0;
+	return ret;
 }
 
 static int axi_pwmgen_round_waveform_fromhw(struct pwm_chip *chip, struct pwm_device *pwm,
diff --git a/drivers/pwm/pwm-fsl-ftm.c b/drivers/pwm/pwm-fsl-ftm.c
index 2510c10ca473..c45a5fca4cbb 100644
--- a/drivers/pwm/pwm-fsl-ftm.c
+++ b/drivers/pwm/pwm-fsl-ftm.c
@@ -118,6 +118,9 @@ static unsigned int fsl_pwm_ticks_to_ns(struct fsl_pwm_chip *fpc,
 	unsigned long long exval;
 
 	rate = clk_get_rate(fpc->clk[fpc->period.clk_select]);
+	if (rate >> fpc->period.clk_ps == 0)
+		return 0;
+
 	exval = ticks;
 	exval *= 1000000000UL;
 	do_div(exval, rate >> fpc->period.clk_ps);
@@ -190,6 +193,9 @@ static unsigned int fsl_pwm_calculate_duty(struct fsl_pwm_chip *fpc,
 	unsigned int period = fpc->period.mod_period + 1;
 	unsigned int period_ns = fsl_pwm_ticks_to_ns(fpc, period);
 
+	if (!period_ns)
+		return 0;
+
 	duty = (unsigned long long)duty_ns * period;
 	do_div(duty, period_ns);
 
diff --git a/drivers/pwm/pwm-mediatek.c b/drivers/pwm/pwm-mediatek.c
index 01dfa0fab80a..7eaab5831499 100644
--- a/drivers/pwm/pwm-mediatek.c
+++ b/drivers/pwm/pwm-mediatek.c
@@ -121,21 +121,25 @@ static int pwm_mediatek_config(struct pwm_chip *chip, struct pwm_device *pwm,
 	struct pwm_mediatek_chip *pc = to_pwm_mediatek_chip(chip);
 	u32 clkdiv = 0, cnt_period, cnt_duty, reg_width = PWMDWIDTH,
 	    reg_thres = PWMTHRES;
+	unsigned long clk_rate;
 	u64 resolution;
 	int ret;
 
 	ret = pwm_mediatek_clk_enable(chip, pwm);
-
 	if (ret < 0)
 		return ret;
 
+	clk_rate = clk_get_rate(pc->clk_pwms[pwm->hwpwm]);
+	if (!clk_rate)
+		return -EINVAL;
+
 	/* Make sure we use the bus clock and not the 26MHz clock */
 	if (pc->soc->has_ck_26m_sel)
 		writel(0, pc->regs + PWM_CK_26M_SEL);
 
 	/* Using resolution in picosecond gets accuracy higher */
 	resolution = (u64)NSEC_PER_SEC * 1000;
-	do_div(resolution, clk_get_rate(pc->clk_pwms[pwm->hwpwm]));
+	do_div(resolution, clk_rate);
 
 	cnt_period = DIV_ROUND_CLOSEST_ULL((u64)period_ns * 1000, resolution);
 	while (cnt_period > 8191) {
diff --git a/drivers/pwm/pwm-rcar.c b/drivers/pwm/pwm-rcar.c
index 2261789cc27d..578dbdd2d5a7 100644
--- a/drivers/pwm/pwm-rcar.c
+++ b/drivers/pwm/pwm-rcar.c
@@ -8,6 +8,7 @@
  * - The hardware cannot generate a 0% duty cycle.
  */
 
+#include <linux/bitfield.h>
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/io.h>
@@ -102,23 +103,24 @@ static void rcar_pwm_set_clock_control(struct rcar_pwm_chip *rp,
 	rcar_pwm_write(rp, value, RCAR_PWMCR);
 }
 
-static int rcar_pwm_set_counter(struct rcar_pwm_chip *rp, int div, int duty_ns,
-				int period_ns)
+static int rcar_pwm_set_counter(struct rcar_pwm_chip *rp, int div, u64 duty_ns,
+				u64 period_ns)
 {
-	unsigned long long one_cycle, tmp;	/* 0.01 nanoseconds */
+	unsigned long long tmp;
 	unsigned long clk_rate = clk_get_rate(rp->clk);
 	u32 cyc, ph;
 
-	one_cycle = NSEC_PER_SEC * 100ULL << div;
-	do_div(one_cycle, clk_rate);
+	/* div <= 24 == RCAR_PWM_MAX_DIVISION, so the shift doesn't overflow. */
+	tmp = mul_u64_u64_div_u64(period_ns, clk_rate, (u64)NSEC_PER_SEC << div);
+	if (tmp > FIELD_MAX(RCAR_PWMCNT_CYC0_MASK))
+		tmp = FIELD_MAX(RCAR_PWMCNT_CYC0_MASK);
 
-	tmp = period_ns * 100ULL;
-	do_div(tmp, one_cycle);
-	cyc = (tmp << RCAR_PWMCNT_CYC0_SHIFT) & RCAR_PWMCNT_CYC0_MASK;
+	cyc = FIELD_PREP(RCAR_PWMCNT_CYC0_MASK, tmp);
 
-	tmp = duty_ns * 100ULL;
-	do_div(tmp, one_cycle);
-	ph = tmp & RCAR_PWMCNT_PH0_MASK;
+	tmp = mul_u64_u64_div_u64(duty_ns, clk_rate, (u64)NSEC_PER_SEC << div);
+	if (tmp > FIELD_MAX(RCAR_PWMCNT_PH0_MASK))
+		tmp = FIELD_MAX(RCAR_PWMCNT_PH0_MASK);
+	ph = FIELD_PREP(RCAR_PWMCNT_PH0_MASK, tmp);
 
 	/* Avoid prohibited setting */
 	if (cyc == 0 || ph == 0)
diff --git a/drivers/pwm/pwm-stm32.c b/drivers/pwm/pwm-stm32.c
index a59de4de18b6..ec2c05c9ee7a 100644
--- a/drivers/pwm/pwm-stm32.c
+++ b/drivers/pwm/pwm-stm32.c
@@ -103,22 +103,16 @@ static int stm32_pwm_round_waveform_tohw(struct pwm_chip *chip,
 		if (ret)
 			goto out;
 
-		/*
-		 * calculate the best value for ARR for the given PSC, refuse if
-		 * the resulting period gets bigger than the requested one.
-		 */
 		arr = mul_u64_u64_div_u64(wf->period_length_ns, rate,
 					  (u64)NSEC_PER_SEC * (wfhw->psc + 1));
 		if (arr <= wfhw->arr) {
 			/*
-			 * requested period is small than the currently
+			 * requested period is smaller than the currently
 			 * configured and unchangable period, report back the smallest
-			 * possible period, i.e. the current state; Initialize
-			 * ccr to anything valid.
+			 * possible period, i.e. the current state and return 1
+			 * to indicate the wrong rounding direction.
 			 */
-			wfhw->ccr = 0;
 			ret = 1;
-			goto out;
 		}
 
 	} else {
diff --git a/drivers/ras/amd/atl/internal.h b/drivers/ras/amd/atl/internal.h
index f9be26d25348..d096b58cd0ae 100644
--- a/drivers/ras/amd/atl/internal.h
+++ b/drivers/ras/amd/atl/internal.h
@@ -362,4 +362,7 @@ static inline void atl_debug_on_bad_intlv_mode(struct addr_ctx *ctx)
 	atl_debug(ctx, "Unrecognized interleave mode: %u", ctx->map.intlv_mode);
 }
 
+#define MI300_UMC_MCA_COL	GENMASK(5, 1)
+#define MI300_UMC_MCA_ROW13	BIT(23)
+
 #endif /* __AMD_ATL_INTERNAL_H__ */
diff --git a/drivers/ras/amd/atl/umc.c b/drivers/ras/amd/atl/umc.c
index dc8aa12f63c8..6e072b7667e9 100644
--- a/drivers/ras/amd/atl/umc.c
+++ b/drivers/ras/amd/atl/umc.c
@@ -229,7 +229,6 @@ int get_umc_info_mi300(void)
  * Additionally, the PC and Bank bits may be hashed. This must be accounted for before
  * reconstructing the normalized address.
  */
-#define MI300_UMC_MCA_COL	GENMASK(5, 1)
 #define MI300_UMC_MCA_BANK	GENMASK(9, 6)
 #define MI300_UMC_MCA_ROW	GENMASK(24, 10)
 #define MI300_UMC_MCA_PC	BIT(25)
@@ -320,7 +319,7 @@ static unsigned long convert_dram_to_norm_addr_mi300(unsigned long addr)
  * See amd_atl::convert_dram_to_norm_addr_mi300() for MI300 address formats.
  */
 #define MI300_NUM_COL		BIT(HWEIGHT(MI300_UMC_MCA_COL))
-static void retire_row_mi300(struct atl_err *a_err)
+static void _retire_row_mi300(struct atl_err *a_err)
 {
 	unsigned long addr;
 	struct page *p;
@@ -351,6 +350,22 @@ static void retire_row_mi300(struct atl_err *a_err)
 	}
 }
 
+/*
+ * In addition to the column bits, the row[13] bit should also be included when
+ * calculating addresses affected by a physical row.
+ *
+ * Instead of running through another loop over a single bit, just run through
+ * the column bits twice and flip the row[13] bit in-between.
+ *
+ * See MI300_UMC_MCA_ROW for the row bits in MCA_ADDR_UMC value.
+ */
+static void retire_row_mi300(struct atl_err *a_err)
+{
+	_retire_row_mi300(a_err);
+	a_err->addr ^= MI300_UMC_MCA_ROW13;
+	_retire_row_mi300(a_err);
+}
+
 void amd_retire_dram_row(struct atl_err *a_err)
 {
 	if (df_cfg.rev == DF4p5 && df_cfg.flags.heterogeneous)
diff --git a/drivers/ras/amd/fmpm.c b/drivers/ras/amd/fmpm.c
index 90de737fbc90..8877c6ff64c4 100644
--- a/drivers/ras/amd/fmpm.c
+++ b/drivers/ras/amd/fmpm.c
@@ -250,6 +250,13 @@ static bool rec_has_valid_entries(struct fru_rec *rec)
 	return true;
 }
 
+/*
+ * Row retirement is done on MI300 systems, and some bits are 'don't
+ * care' for comparing addresses with unique physical rows.  This
+ * includes all column bits and the row[13] bit.
+ */
+#define MASK_ADDR(addr)	((addr) & ~(MI300_UMC_MCA_ROW13 | MI300_UMC_MCA_COL))
+
 static bool fpds_equal(struct cper_fru_poison_desc *old, struct cper_fru_poison_desc *new)
 {
 	/*
@@ -258,7 +265,7 @@ static bool fpds_equal(struct cper_fru_poison_desc *old, struct cper_fru_poison_
 	 *
 	 * Also, order the checks from most->least likely to fail to shortcut the code.
 	 */
-	if (old->addr != new->addr)
+	if (MASK_ADDR(old->addr) != MASK_ADDR(new->addr))
 		return false;
 
 	if (old->hw_id != new->hw_id)
diff --git a/drivers/s390/virtio/virtio_ccw.c b/drivers/s390/virtio/virtio_ccw.c
index 21fa7ac849e5..4904b831c0a7 100644
--- a/drivers/s390/virtio/virtio_ccw.c
+++ b/drivers/s390/virtio/virtio_ccw.c
@@ -302,11 +302,17 @@ static struct airq_info *new_airq_info(int index)
 static unsigned long *get_airq_indicator(struct virtqueue *vqs[], int nvqs,
 					 u64 *first, void **airq_info)
 {
-	int i, j;
+	int i, j, queue_idx, highest_queue_idx = -1;
 	struct airq_info *info;
 	unsigned long *indicator_addr = NULL;
 	unsigned long bit, flags;
 
+	/* Array entries without an actual queue pointer must be ignored. */
+	for (i = 0; i < nvqs; i++) {
+		if (vqs[i])
+			highest_queue_idx++;
+	}
+
 	for (i = 0; i < MAX_AIRQ_AREAS && !indicator_addr; i++) {
 		mutex_lock(&airq_areas_lock);
 		if (!airq_areas[i])
@@ -316,7 +322,7 @@ static unsigned long *get_airq_indicator(struct virtqueue *vqs[], int nvqs,
 		if (!info)
 			return NULL;
 		write_lock_irqsave(&info->lock, flags);
-		bit = airq_iv_alloc(info->aiv, nvqs);
+		bit = airq_iv_alloc(info->aiv, highest_queue_idx + 1);
 		if (bit == -1UL) {
 			/* Not enough vacancies. */
 			write_unlock_irqrestore(&info->lock, flags);
@@ -325,8 +331,10 @@ static unsigned long *get_airq_indicator(struct virtqueue *vqs[], int nvqs,
 		*first = bit;
 		*airq_info = info;
 		indicator_addr = info->aiv->vector;
-		for (j = 0; j < nvqs; j++) {
-			airq_iv_set_ptr(info->aiv, bit + j,
+		for (j = 0, queue_idx = 0; j < nvqs; j++) {
+			if (!vqs[j])
+				continue;
+			airq_iv_set_ptr(info->aiv, bit + queue_idx++,
 					(unsigned long)vqs[j]);
 		}
 		write_unlock_irqrestore(&info->lock, flags);
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 5cb1d3db4907..944cf2fb0561 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -935,8 +935,28 @@ static void hisi_sas_phyup_work_common(struct work_struct *work,
 		container_of(work, typeof(*phy), works[event]);
 	struct hisi_hba *hisi_hba = phy->hisi_hba;
 	struct asd_sas_phy *sas_phy = &phy->sas_phy;
+	struct asd_sas_port *sas_port = sas_phy->port;
+	struct hisi_sas_port *port = phy->port;
+	struct device *dev = hisi_hba->dev;
+	struct domain_device *port_dev;
 	int phy_no = sas_phy->id;
 
+	if (!test_bit(HISI_SAS_RESETTING_BIT, &hisi_hba->flags) &&
+	    sas_port && port && (port->id != phy->port_id)) {
+		dev_info(dev, "phy%d's hw port id changed from %d to %llu\n",
+				phy_no, port->id, phy->port_id);
+		port_dev = sas_port->port_dev;
+		if (port_dev && !dev_is_expander(port_dev->dev_type)) {
+			/*
+			 * Set the device state to gone to block
+			 * sending IO to the device.
+			 */
+			set_bit(SAS_DEV_GONE, &port_dev->state);
+			hisi_sas_notify_phy_event(phy, HISI_PHYE_LINK_RESET);
+			return;
+		}
+	}
+
 	phy->wait_phyup_cnt = 0;
 	if (phy->identify.target_port_protocols == SAS_PROTOCOL_SSP)
 		hisi_hba->hw->sl_notify_ssp(hisi_hba, phy_no);
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index a1fc400ab4c3..1e9830940f84 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -2501,6 +2501,7 @@ static void prep_ata_v2_hw(struct hisi_hba *hisi_hba,
 	struct hisi_sas_port *port = to_hisi_sas_port(sas_port);
 	struct sas_ata_task *ata_task = &task->ata_task;
 	struct sas_tmf_task *tmf = slot->tmf;
+	int phy_id;
 	u8 *buf_cmd;
 	int has_data = 0, hdr_tag = 0;
 	u32 dw0, dw1 = 0, dw2 = 0;
@@ -2508,10 +2509,14 @@ static void prep_ata_v2_hw(struct hisi_hba *hisi_hba,
 	/* create header */
 	/* dw0 */
 	dw0 = port->id << CMD_HDR_PORT_OFF;
-	if (parent_dev && dev_is_expander(parent_dev->dev_type))
+	if (parent_dev && dev_is_expander(parent_dev->dev_type)) {
 		dw0 |= 3 << CMD_HDR_CMD_OFF;
-	else
+	} else {
+		phy_id = device->phy->identify.phy_identifier;
+		dw0 |= (1U << phy_id) << CMD_HDR_PHY_ID_OFF;
+		dw0 |= CMD_HDR_FORCE_PHY_MSK;
 		dw0 |= 4 << CMD_HDR_CMD_OFF;
+	}
 
 	if (tmf && ata_task->force_phy) {
 		dw0 |= CMD_HDR_FORCE_PHY_MSK;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
index 2684d6482067..08dac9ae2f10 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
@@ -359,6 +359,10 @@
 #define CMD_HDR_RESP_REPORT_MSK		(0x1 << CMD_HDR_RESP_REPORT_OFF)
 #define CMD_HDR_TLR_CTRL_OFF		6
 #define CMD_HDR_TLR_CTRL_MSK		(0x3 << CMD_HDR_TLR_CTRL_OFF)
+#define CMD_HDR_PHY_ID_OFF		8
+#define CMD_HDR_PHY_ID_MSK		(0x1ff << CMD_HDR_PHY_ID_OFF)
+#define CMD_HDR_FORCE_PHY_OFF		17
+#define CMD_HDR_FORCE_PHY_MSK		(0x1U << CMD_HDR_FORCE_PHY_OFF)
 #define CMD_HDR_PORT_OFF		18
 #define CMD_HDR_PORT_MSK		(0xf << CMD_HDR_PORT_OFF)
 #define CMD_HDR_PRIORITY_OFF		27
@@ -1429,15 +1433,21 @@ static void prep_ata_v3_hw(struct hisi_hba *hisi_hba,
 	struct hisi_sas_cmd_hdr *hdr = slot->cmd_hdr;
 	struct asd_sas_port *sas_port = device->port;
 	struct hisi_sas_port *port = to_hisi_sas_port(sas_port);
+	int phy_id;
 	u8 *buf_cmd;
 	int has_data = 0, hdr_tag = 0;
 	u32 dw1 = 0, dw2 = 0;
 
 	hdr->dw0 = cpu_to_le32(port->id << CMD_HDR_PORT_OFF);
-	if (parent_dev && dev_is_expander(parent_dev->dev_type))
+	if (parent_dev && dev_is_expander(parent_dev->dev_type)) {
 		hdr->dw0 |= cpu_to_le32(3 << CMD_HDR_CMD_OFF);
-	else
+	} else {
+		phy_id = device->phy->identify.phy_identifier;
+		hdr->dw0 |= cpu_to_le32((1U << phy_id)
+				<< CMD_HDR_PHY_ID_OFF);
+		hdr->dw0 |= CMD_HDR_FORCE_PHY_MSK;
 		hdr->dw0 |= cpu_to_le32(4U << CMD_HDR_CMD_OFF);
+	}
 
 	switch (task->data_dir) {
 	case DMA_TO_DEVICE:
diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h
index 088cc40ae866..8ee2bfe47571 100644
--- a/drivers/scsi/megaraid/megaraid_sas.h
+++ b/drivers/scsi/megaraid/megaraid_sas.h
@@ -23,8 +23,8 @@
 /*
  * MegaRAID SAS Driver meta data
  */
-#define MEGASAS_VERSION				"07.727.03.00-rc1"
-#define MEGASAS_RELDATE				"Oct 03, 2023"
+#define MEGASAS_VERSION				"07.734.00.00-rc1"
+#define MEGASAS_RELDATE				"Apr 03, 2025"
 
 #define MEGASAS_MSIX_NAME_LEN			32
 
diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c
index c20447b39cb9..5e33d411fa3d 100644
--- a/drivers/scsi/megaraid/megaraid_sas_base.c
+++ b/drivers/scsi/megaraid/megaraid_sas_base.c
@@ -2103,6 +2103,9 @@ static int megasas_sdev_configure(struct scsi_device *sdev,
 	/* This sdev property may change post OCR */
 	megasas_set_dynamic_target_properties(sdev, lim, is_target_prop);
 
+	if (!MEGASAS_IS_LOGICAL(sdev))
+		sdev->no_vpd_size = 1;
+
 	mutex_unlock(&instance->reset_mutex);
 
 	return 0;
@@ -3662,8 +3665,10 @@ megasas_complete_cmd(struct megasas_instance *instance, struct megasas_cmd *cmd,
 
 		case MFI_STAT_SCSI_IO_FAILED:
 		case MFI_STAT_LD_INIT_IN_PROGRESS:
-			cmd->scmd->result =
-			    (DID_ERROR << 16) | hdr->scsi_status;
+			if (hdr->scsi_status == 0xf0)
+				cmd->scmd->result = (DID_ERROR << 16) | SAM_STAT_CHECK_CONDITION;
+			else
+				cmd->scmd->result = (DID_ERROR << 16) | hdr->scsi_status;
 			break;
 
 		case MFI_STAT_SCSI_DONE_WITH_ERROR:
diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c
index 721860cb1ef6..a6794f49e9fa 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -2043,7 +2043,10 @@ map_cmd_status(struct fusion_context *fusion,
 
 	case MFI_STAT_SCSI_IO_FAILED:
 	case MFI_STAT_LD_INIT_IN_PROGRESS:
-		scmd->result = (DID_ERROR << 16) | ext_status;
+		if (ext_status == 0xf0)
+			scmd->result = (DID_ERROR << 16) | SAM_STAT_CHECK_CONDITION;
+		else
+			scmd->result = (DID_ERROR << 16) | ext_status;
 		break;
 
 	case MFI_STAT_SCSI_DONE_WITH_ERROR:
diff --git a/drivers/scsi/mpi3mr/mpi3mr_fw.c b/drivers/scsi/mpi3mr/mpi3mr_fw.c
index 3fcb1ad3b070..1d7901a8f0e4 100644
--- a/drivers/scsi/mpi3mr/mpi3mr_fw.c
+++ b/drivers/scsi/mpi3mr/mpi3mr_fw.c
@@ -174,6 +174,9 @@ static void mpi3mr_print_event_data(struct mpi3mr_ioc *mrioc,
 	char *desc = NULL;
 	u16 event;
 
+	if (!(mrioc->logging_level & MPI3_DEBUG_EVENT))
+		return;
+
 	event = event_reply->event;
 
 	switch (event) {
@@ -451,6 +454,7 @@ int mpi3mr_process_admin_reply_q(struct mpi3mr_ioc *mrioc)
 		return 0;
 	}
 
+	atomic_set(&mrioc->admin_pend_isr, 0);
 	reply_desc = (struct mpi3_default_reply_descriptor *)mrioc->admin_reply_base +
 	    admin_reply_ci;
 
@@ -565,7 +569,7 @@ int mpi3mr_process_op_reply_q(struct mpi3mr_ioc *mrioc,
 		WRITE_ONCE(op_req_q->ci, le16_to_cpu(reply_desc->request_queue_ci));
 		mpi3mr_process_op_reply_desc(mrioc, reply_desc, &reply_dma,
 		    reply_qidx);
-		atomic_dec(&op_reply_q->pend_ios);
+
 		if (reply_dma)
 			mpi3mr_repost_reply_buf(mrioc, reply_dma);
 		num_op_reply++;
@@ -2925,6 +2929,7 @@ static int mpi3mr_setup_admin_qpair(struct mpi3mr_ioc *mrioc)
 	mrioc->admin_reply_ci = 0;
 	mrioc->admin_reply_ephase = 1;
 	atomic_set(&mrioc->admin_reply_q_in_use, 0);
+	atomic_set(&mrioc->admin_pend_isr, 0);
 
 	if (!mrioc->admin_req_base) {
 		mrioc->admin_req_base = dma_alloc_coherent(&mrioc->pdev->dev,
@@ -4653,6 +4658,7 @@ void mpi3mr_memset_buffers(struct mpi3mr_ioc *mrioc)
 	if (mrioc->admin_reply_base)
 		memset(mrioc->admin_reply_base, 0, mrioc->admin_reply_q_sz);
 	atomic_set(&mrioc->admin_reply_q_in_use, 0);
+	atomic_set(&mrioc->admin_pend_isr, 0);
 
 	if (mrioc->init_cmds.reply) {
 		memset(mrioc->init_cmds.reply, 0, sizeof(*mrioc->init_cmds.reply));
diff --git a/drivers/scsi/myrb.c b/drivers/scsi/myrb.c
index dc4bd422b601..486db5b2f05d 100644
--- a/drivers/scsi/myrb.c
+++ b/drivers/scsi/myrb.c
@@ -891,7 +891,7 @@ static bool myrb_enable_mmio(struct myrb_hba *cb, mbox_mmio_init_t mmio_init_fn)
 		status = mmio_init_fn(pdev, base, &mbox);
 		if (status != MYRB_STATUS_SUCCESS) {
 			dev_err(&pdev->dev,
-				"Failed to enable mailbox, statux %02X\n",
+				"Failed to enable mailbox, status %02X\n",
 				status);
 			return false;
 		}
diff --git a/drivers/scsi/pm8001/pm8001_sas.c b/drivers/scsi/pm8001/pm8001_sas.c
index 183ce00aa671..f7067878b34f 100644
--- a/drivers/scsi/pm8001/pm8001_sas.c
+++ b/drivers/scsi/pm8001/pm8001_sas.c
@@ -766,6 +766,7 @@ static void pm8001_dev_gone_notify(struct domain_device *dev)
 			spin_lock_irqsave(&pm8001_ha->lock, flags);
 		}
 		PM8001_CHIP_DISP->dereg_dev_req(pm8001_ha, device_id);
+		pm8001_ha->phy[pm8001_dev->attached_phy].phy_attached = 0;
 		pm8001_free_dev(pm8001_dev);
 	} else {
 		pm8001_dbg(pm8001_ha, DISC, "Found dev has gone.\n");
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 53daf923ad8e..518a252eb6aa 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -707,26 +707,23 @@ void scsi_cdl_check(struct scsi_device *sdev)
  */
 int scsi_cdl_enable(struct scsi_device *sdev, bool enable)
 {
-	struct scsi_mode_data data;
-	struct scsi_sense_hdr sshdr;
-	struct scsi_vpd *vpd;
-	bool is_ata = false;
 	char buf[64];
+	bool is_ata;
 	int ret;
 
 	if (!sdev->cdl_supported)
 		return -EOPNOTSUPP;
 
 	rcu_read_lock();
-	vpd = rcu_dereference(sdev->vpd_pg89);
-	if (vpd)
-		is_ata = true;
+	is_ata = rcu_dereference(sdev->vpd_pg89);
 	rcu_read_unlock();
 
 	/*
 	 * For ATA devices, CDL needs to be enabled with a SET FEATURES command.
 	 */
 	if (is_ata) {
+		struct scsi_mode_data data;
+		struct scsi_sense_hdr sshdr;
 		char *buf_data;
 		int len;
 
@@ -735,16 +732,30 @@ int scsi_cdl_enable(struct scsi_device *sdev, bool enable)
 		if (ret)
 			return -EINVAL;
 
-		/* Enable CDL using the ATA feature page */
+		/* Enable or disable CDL using the ATA feature page */
 		len = min_t(size_t, sizeof(buf),
 			    data.length - data.header_length -
 			    data.block_descriptor_length);
 		buf_data = buf + data.header_length +
 			data.block_descriptor_length;
-		if (enable)
-			buf_data[4] = 0x02;
-		else
-			buf_data[4] = 0;
+
+		/*
+		 * If we want to enable CDL and CDL is already enabled on the
+		 * device, do nothing. This avoids needlessly resetting the CDL
+		 * statistics on the device as that is implied by the CDL enable
+		 * action. Similar to this, there is no need to do anything if
+		 * we want to disable CDL and CDL is already disabled.
+		 */
+		if (enable) {
+			if ((buf_data[4] & 0x03) == 0x02)
+				goto out;
+			buf_data[4] &= ~0x03;
+			buf_data[4] |= 0x02;
+		} else {
+			if ((buf_data[4] & 0x03) == 0x00)
+				goto out;
+			buf_data[4] &= ~0x03;
+		}
 
 		ret = scsi_mode_select(sdev, 1, 0, buf_data, len, 5 * HZ, 3,
 				       &data, &sshdr);
@@ -756,6 +767,7 @@ int scsi_cdl_enable(struct scsi_device *sdev, bool enable)
 		}
 	}
 
+out:
 	sdev->cdl_enable = enable;
 
 	return 0;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 0d29470e86b0..1b43013d72c0 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1253,8 +1253,12 @@ EXPORT_SYMBOL_GPL(scsi_alloc_request);
  */
 static void scsi_cleanup_rq(struct request *rq)
 {
+	struct scsi_cmnd *cmd = blk_mq_rq_to_pdu(rq);
+
+	cmd->flags = 0;
+
 	if (rq->rq_flags & RQF_DONTPREP) {
-		scsi_mq_uninit_cmd(blk_mq_rq_to_pdu(rq));
+		scsi_mq_uninit_cmd(cmd);
 		rq->rq_flags &= ~RQF_DONTPREP;
 	}
 }
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 9c347c64c315..0b8c91bf793f 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -3182,11 +3182,14 @@ iscsi_set_host_param(struct iscsi_transport *transport,
 	}
 
 	/* see similar check in iscsi_if_set_param() */
-	if (strlen(data) > ev->u.set_host_param.len)
-		return -EINVAL;
+	if (strlen(data) > ev->u.set_host_param.len) {
+		err = -EINVAL;
+		goto out;
+	}
 
 	err = transport->set_host_param(shost, ev->u.set_host_param.param,
 					data, ev->u.set_host_param.len);
+out:
 	scsi_host_put(shost);
 	return err;
 }
diff --git a/drivers/scsi/scsi_transport_srp.c b/drivers/scsi/scsi_transport_srp.c
index 64f6b22e8cc0..aeb58a9e6b7f 100644
--- a/drivers/scsi/scsi_transport_srp.c
+++ b/drivers/scsi/scsi_transport_srp.c
@@ -388,7 +388,7 @@ static void srp_reconnect_work(struct work_struct *work)
 			     "reconnect attempt %d failed (%d)\n",
 			     ++rport->failed_reconnects, res);
 		delay = rport->reconnect_delay *
-			min(100, max(1, rport->failed_reconnects - 10));
+			clamp(rport->failed_reconnects - 10, 1, 100);
 		if (delay > 0)
 			queue_delayed_work(system_long_wq,
 					   &rport->reconnect_work, delay * HZ);
diff --git a/drivers/scsi/smartpqi/smartpqi_init.c b/drivers/scsi/smartpqi/smartpqi_init.c
index 88135fdb8bd1..8a26eca4fdc9 100644
--- a/drivers/scsi/smartpqi/smartpqi_init.c
+++ b/drivers/scsi/smartpqi/smartpqi_init.c
@@ -19,6 +19,7 @@
 #include <linux/bcd.h>
 #include <linux/reboot.h>
 #include <linux/cciss_ioctl.h>
+#include <linux/crash_dump.h>
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_device.h>
@@ -5246,7 +5247,7 @@ static void pqi_calculate_io_resources(struct pqi_ctrl_info *ctrl_info)
 	ctrl_info->error_buffer_length =
 		ctrl_info->max_io_slots * PQI_ERROR_BUFFER_ELEMENT_LENGTH;
 
-	if (reset_devices)
+	if (is_kdump_kernel())
 		max_transfer_size = min(ctrl_info->max_transfer_size,
 			PQI_MAX_TRANSFER_SIZE_KDUMP);
 	else
@@ -5275,7 +5276,7 @@ static void pqi_calculate_queue_resources(struct pqi_ctrl_info *ctrl_info)
 	u16 num_elements_per_iq;
 	u16 num_elements_per_oq;
 
-	if (reset_devices) {
+	if (is_kdump_kernel()) {
 		num_queue_groups = 1;
 	} else {
 		int num_cpus;
@@ -8288,12 +8289,12 @@ static int pqi_ctrl_init(struct pqi_ctrl_info *ctrl_info)
 	u32 product_id;
 
 	if (reset_devices) {
-		if (pqi_is_fw_triage_supported(ctrl_info)) {
+		if (is_kdump_kernel() && pqi_is_fw_triage_supported(ctrl_info)) {
 			rc = sis_wait_for_fw_triage_completion(ctrl_info);
 			if (rc)
 				return rc;
 		}
-		if (sis_is_ctrl_logging_supported(ctrl_info)) {
+		if (is_kdump_kernel() && sis_is_ctrl_logging_supported(ctrl_info)) {
 			sis_notify_kdump(ctrl_info);
 			rc = sis_wait_for_ctrl_logging_completion(ctrl_info);
 			if (rc)
@@ -8344,7 +8345,7 @@ static int pqi_ctrl_init(struct pqi_ctrl_info *ctrl_info)
 	ctrl_info->product_id = (u8)product_id;
 	ctrl_info->product_revision = (u8)(product_id >> 8);
 
-	if (reset_devices) {
+	if (is_kdump_kernel()) {
 		if (ctrl_info->max_outstanding_requests >
 			PQI_MAX_OUTSTANDING_REQUESTS_KDUMP)
 				ctrl_info->max_outstanding_requests =
@@ -8480,7 +8481,7 @@ static int pqi_ctrl_init(struct pqi_ctrl_info *ctrl_info)
 	if (rc)
 		return rc;
 
-	if (ctrl_info->ctrl_logging_supported && !reset_devices) {
+	if (ctrl_info->ctrl_logging_supported && !is_kdump_kernel()) {
 		pqi_host_setup_buffer(ctrl_info, &ctrl_info->ctrl_log_memory, PQI_CTRL_LOG_TOTAL_SIZE, PQI_CTRL_LOG_MIN_SIZE);
 		pqi_host_memory_update(ctrl_info, &ctrl_info->ctrl_log_memory, PQI_VENDOR_GENERAL_CTRL_LOG_MEMORY_UPDATE);
 	}
diff --git a/drivers/soundwire/intel_auxdevice.c b/drivers/soundwire/intel_auxdevice.c
index 5ea6399e6c9b..10a602d4843a 100644
--- a/drivers/soundwire/intel_auxdevice.c
+++ b/drivers/soundwire/intel_auxdevice.c
@@ -353,9 +353,6 @@ static int intel_link_probe(struct auxiliary_device *auxdev,
 	/* use generic bandwidth allocation algorithm */
 	sdw->cdns.bus.compute_params = sdw_compute_params;
 
-	/* avoid resuming from pm_runtime suspend if it's not required */
-	dev_pm_set_driver_flags(dev, DPM_FLAG_SMART_SUSPEND);
-
 	ret = sdw_bus_master_add(bus, dev, dev->fwnode);
 	if (ret) {
 		dev_err(dev, "sdw_bus_master_add fail: %d\n", ret);
@@ -640,7 +637,10 @@ static int __maybe_unused intel_suspend(struct device *dev)
 		return 0;
 	}
 
-	if (pm_runtime_suspended(dev)) {
+	/* Prevent runtime PM from racing with the code below. */
+	pm_runtime_disable(dev);
+
+	if (pm_runtime_status_suspended(dev)) {
 		dev_dbg(dev, "pm_runtime status: suspended\n");
 
 		clock_stop_quirks = sdw->link_res->clock_stop_quirks;
@@ -648,7 +648,7 @@ static int __maybe_unused intel_suspend(struct device *dev)
 		if ((clock_stop_quirks & SDW_INTEL_CLK_STOP_BUS_RESET) ||
 		    !clock_stop_quirks) {
 
-			if (pm_runtime_suspended(dev->parent)) {
+			if (pm_runtime_status_suspended(dev->parent)) {
 				/*
 				 * paranoia check: this should not happen with the .prepare
 				 * resume to full power
@@ -715,7 +715,6 @@ static int __maybe_unused intel_resume(struct device *dev)
 	struct sdw_cdns *cdns = dev_get_drvdata(dev);
 	struct sdw_intel *sdw = cdns_to_intel(cdns);
 	struct sdw_bus *bus = &cdns->bus;
-	int link_flags;
 	int ret;
 
 	if (bus->prop.hw_disabled || !sdw->startup_done) {
@@ -724,23 +723,6 @@ static int __maybe_unused intel_resume(struct device *dev)
 		return 0;
 	}
 
-	if (pm_runtime_suspended(dev)) {
-		dev_dbg(dev, "pm_runtime status was suspended, forcing active\n");
-
-		/* follow required sequence from runtime_pm.rst */
-		pm_runtime_disable(dev);
-		pm_runtime_set_active(dev);
-		pm_runtime_mark_last_busy(dev);
-		pm_runtime_enable(dev);
-
-		pm_runtime_resume(bus->dev);
-
-		link_flags = md_flags >> (bus->link_id * 8);
-
-		if (!(link_flags & SDW_INTEL_MASTER_DISABLE_PM_RUNTIME_IDLE))
-			pm_runtime_idle(dev);
-	}
-
 	ret = sdw_intel_link_power_up(sdw);
 	if (ret) {
 		dev_err(dev, "%s failed: %d\n", __func__, ret);
@@ -761,6 +743,14 @@ static int __maybe_unused intel_resume(struct device *dev)
 	}
 
 	/*
+	 * Runtime PM has been disabled in intel_suspend(), so set the status
+	 * to active because the device has just been resumed and re-enable
+	 * runtime PM.
+	 */
+	pm_runtime_set_active(dev);
+	pm_runtime_enable(dev);
+
+	/*
 	 * after system resume, the pm_runtime suspend() may kick in
 	 * during the enumeration, before any children device force the
 	 * master device to remain active.  Using pm_runtime_get()
diff --git a/drivers/spi/spi-fsl-qspi.c b/drivers/spi/spi-fsl-qspi.c
index 5c59fddb32c1..b5ecffcaf795 100644
--- a/drivers/spi/spi-fsl-qspi.c
+++ b/drivers/spi/spi-fsl-qspi.c
@@ -949,24 +949,20 @@ static int fsl_qspi_probe(struct platform_device *pdev)
 
 	ret = devm_add_action_or_reset(dev, fsl_qspi_cleanup, q);
 	if (ret)
-		goto err_destroy_mutex;
+		goto err_put_ctrl;
 
 	ret = devm_spi_register_controller(dev, ctlr);
 	if (ret)
-		goto err_destroy_mutex;
+		goto err_put_ctrl;
 
 	return 0;
 
-err_destroy_mutex:
-	mutex_destroy(&q->lock);
-
 err_disable_clk:
 	fsl_qspi_clk_disable_unprep(q);
 
 err_put_ctrl:
 	spi_controller_put(ctlr);
 
-	dev_err(dev, "Freescale QuadSPI probe failed\n");
 	return ret;
 }
 
diff --git a/drivers/spi/spi-imx.c b/drivers/spi/spi-imx.c
index 832d6e9009eb..c93d80a4d734 100644
--- a/drivers/spi/spi-imx.c
+++ b/drivers/spi/spi-imx.c
@@ -1695,9 +1695,12 @@ static int spi_imx_transfer_one(struct spi_controller *controller,
 				struct spi_device *spi,
 				struct spi_transfer *transfer)
 {
+	int ret;
 	struct spi_imx_data *spi_imx = spi_controller_get_devdata(spi->controller);
 
-	spi_imx_setupxfer(spi, transfer);
+	ret = spi_imx_setupxfer(spi, transfer);
+	if (ret < 0)
+		return ret;
 	transfer->effective_speed_hz = spi_imx->spi_bus_clk;
 
 	/* flush rxfifo before transfer */
diff --git a/drivers/spi/spi-mem.c b/drivers/spi/spi-mem.c
index a31a1db07aa4..5db0639d3b01 100644
--- a/drivers/spi/spi-mem.c
+++ b/drivers/spi/spi-mem.c
@@ -596,7 +596,11 @@ u64 spi_mem_calc_op_duration(struct spi_mem_op *op)
 	ns_per_cycles = 1000000000 / op->max_freq;
 	ncycles += ((op->cmd.nbytes * 8) / op->cmd.buswidth) / (op->cmd.dtr ? 2 : 1);
 	ncycles += ((op->addr.nbytes * 8) / op->addr.buswidth) / (op->addr.dtr ? 2 : 1);
-	ncycles += ((op->dummy.nbytes * 8) / op->dummy.buswidth) / (op->dummy.dtr ? 2 : 1);
+
+	/* Dummy bytes are optional for some SPI flash memory operations */
+	if (op->dummy.nbytes)
+		ncycles += ((op->dummy.nbytes * 8) / op->dummy.buswidth) / (op->dummy.dtr ? 2 : 1);
+
 	ncycles += ((op->data.nbytes * 8) / op->data.buswidth) / (op->data.dtr ? 2 : 1);
 
 	return ncycles * ns_per_cycles;
diff --git a/drivers/spi/spi-qpic-snand.c b/drivers/spi/spi-qpic-snand.c
index 17eb67e19132..94948c8781e8 100644
--- a/drivers/spi/spi-qpic-snand.c
+++ b/drivers/spi/spi-qpic-snand.c
@@ -142,7 +142,7 @@ static void qcom_spi_set_read_loc_first(struct qcom_nand_controller *snandc,
 	else if (reg == NAND_READ_LOCATION_1)
 		snandc->regs->read_location1 = locreg_val;
 	else if (reg == NAND_READ_LOCATION_2)
-		snandc->regs->read_location1 = locreg_val;
+		snandc->regs->read_location2 = locreg_val;
 	else if (reg == NAND_READ_LOCATION_3)
 		snandc->regs->read_location3 = locreg_val;
 }
@@ -1307,8 +1307,7 @@ static int qcom_spi_send_cmdaddr(struct qcom_nand_controller *snandc,
 		snandc->qspi->addr1 = cpu_to_le32(s_op.addr1_reg << 16);
 		snandc->qspi->addr2 = cpu_to_le32(s_op.addr2_reg);
 		snandc->qspi->cmd = cpu_to_le32(cmd);
-		qcom_spi_block_erase(snandc);
-		return 0;
+		return qcom_spi_block_erase(snandc);
 	default:
 		break;
 	}
diff --git a/drivers/spi/spi-stm32-ospi.c b/drivers/spi/spi-stm32-ospi.c
index 668022098b1e..9ec9823409cc 100644
--- a/drivers/spi/spi-stm32-ospi.c
+++ b/drivers/spi/spi-stm32-ospi.c
@@ -960,6 +960,10 @@ err_pm_resume:
 err_pm_enable:
 	pm_runtime_force_suspend(ospi->dev);
 	mutex_destroy(&ospi->lock);
+	if (ospi->dma_chtx)
+		dma_release_channel(ospi->dma_chtx);
+	if (ospi->dma_chrx)
+		dma_release_channel(ospi->dma_chrx);
 
 	return ret;
 }
diff --git a/drivers/spi/spi-sun4i.c b/drivers/spi/spi-sun4i.c
index fcbe864c9b7d..f89826d7dc49 100644
--- a/drivers/spi/spi-sun4i.c
+++ b/drivers/spi/spi-sun4i.c
@@ -462,6 +462,7 @@ static int sun4i_spi_probe(struct platform_device *pdev)
 	sspi->host = host;
 	host->max_speed_hz = 100 * 1000 * 1000;
 	host->min_speed_hz = 3 * 1000;
+	host->use_gpio_descriptors = true;
 	host->set_cs = sun4i_spi_set_cs;
 	host->transfer_one = sun4i_spi_transfer_one;
 	host->num_chipselect = 4;
diff --git a/drivers/spi/spi-tegra114.c b/drivers/spi/spi-tegra114.c
index 3822d7c8d8ed..2a8bb798e95b 100644
--- a/drivers/spi/spi-tegra114.c
+++ b/drivers/spi/spi-tegra114.c
@@ -728,9 +728,9 @@ static int tegra_spi_set_hw_cs_timing(struct spi_device *spi)
 	u32 inactive_cycles;
 	u8 cs_state;
 
-	if (setup->unit != SPI_DELAY_UNIT_SCK ||
-	    hold->unit != SPI_DELAY_UNIT_SCK ||
-	    inactive->unit != SPI_DELAY_UNIT_SCK) {
+	if ((setup->unit && setup->unit != SPI_DELAY_UNIT_SCK) ||
+	    (hold->unit && hold->unit != SPI_DELAY_UNIT_SCK) ||
+	    (inactive->unit && inactive->unit != SPI_DELAY_UNIT_SCK)) {
 		dev_err(&spi->dev,
 			"Invalid delay unit %d, should be SPI_DELAY_UNIT_SCK\n",
 			SPI_DELAY_UNIT_SCK);
diff --git a/drivers/spi/spi-tegra210-quad.c b/drivers/spi/spi-tegra210-quad.c
index 08e49a876894..64e1b2f8a000 100644
--- a/drivers/spi/spi-tegra210-quad.c
+++ b/drivers/spi/spi-tegra210-quad.c
@@ -1117,9 +1117,9 @@ static int tegra_qspi_combined_seq_xfer(struct tegra_qspi *tqspi,
 					(&tqspi->xfer_completion,
 					QSPI_DMA_TIMEOUT);
 
-			if (WARN_ON(ret == 0)) {
-				dev_err(tqspi->dev, "QSPI Transfer failed with timeout: %d\n",
-					ret);
+			if (WARN_ON_ONCE(ret == 0)) {
+				dev_err_ratelimited(tqspi->dev,
+						    "QSPI Transfer failed with timeout\n");
 				if (tqspi->is_curr_dma_xfer &&
 				    (tqspi->cur_direction & DATA_DIR_TX))
 					dmaengine_terminate_all
diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 1244ef3aa86c..620ba6e0ab07 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -4263,8 +4263,8 @@ int iscsit_close_connection(
 	spin_unlock(&iscsit_global->ts_bitmap_lock);
 
 	iscsit_stop_timers_for_cmds(conn);
-	iscsit_stop_nopin_response_timer(conn);
 	iscsit_stop_nopin_timer(conn);
+	iscsit_stop_nopin_response_timer(conn);
 
 	if (conn->conn_transport->iscsit_wait_conn)
 		conn->conn_transport->iscsit_wait_conn(conn);
diff --git a/drivers/thermal/intel/int340x_thermal/processor_thermal_device_pci.c b/drivers/thermal/intel/int340x_thermal/processor_thermal_device_pci.c
index a55aaa8cef42..2097aae39946 100644
--- a/drivers/thermal/intel/int340x_thermal/processor_thermal_device_pci.c
+++ b/drivers/thermal/intel/int340x_thermal/processor_thermal_device_pci.c
@@ -485,7 +485,7 @@ static const struct pci_device_id proc_thermal_pci_ids[] = {
 	{ PCI_DEVICE_DATA(INTEL, ADL_THERMAL, PROC_THERMAL_FEATURE_RAPL |
 	  PROC_THERMAL_FEATURE_FIVR | PROC_THERMAL_FEATURE_DVFS | PROC_THERMAL_FEATURE_WT_REQ) },
 	{ PCI_DEVICE_DATA(INTEL, LNLM_THERMAL, PROC_THERMAL_FEATURE_MSI_SUPPORT |
-	  PROC_THERMAL_FEATURE_RAPL | PROC_THERMAL_FEATURE_DLVR |
+	  PROC_THERMAL_FEATURE_RAPL | PROC_THERMAL_FEATURE_DLVR | PROC_THERMAL_FEATURE_DVFS |
 	  PROC_THERMAL_FEATURE_WT_HINT | PROC_THERMAL_FEATURE_POWER_FLOOR) },
 	{ PCI_DEVICE_DATA(INTEL, MTLP_THERMAL, PROC_THERMAL_FEATURE_RAPL |
 	  PROC_THERMAL_FEATURE_FIVR | PROC_THERMAL_FEATURE_DVFS | PROC_THERMAL_FEATURE_DLVR |
@@ -495,8 +495,9 @@ static const struct pci_device_id proc_thermal_pci_ids[] = {
 	{ PCI_DEVICE_DATA(INTEL, RPL_THERMAL, PROC_THERMAL_FEATURE_RAPL |
 	  PROC_THERMAL_FEATURE_FIVR | PROC_THERMAL_FEATURE_DVFS | PROC_THERMAL_FEATURE_WT_REQ) },
 	{ PCI_DEVICE_DATA(INTEL, PTL_THERMAL, PROC_THERMAL_FEATURE_RAPL |
-	  PROC_THERMAL_FEATURE_DLVR | PROC_THERMAL_FEATURE_MSI_SUPPORT |
-	  PROC_THERMAL_FEATURE_WT_HINT | PROC_THERMAL_FEATURE_POWER_FLOOR) },
+	  PROC_THERMAL_FEATURE_DLVR | PROC_THERMAL_FEATURE_DVFS |
+	  PROC_THERMAL_FEATURE_MSI_SUPPORT | PROC_THERMAL_FEATURE_WT_HINT |
+	  PROC_THERMAL_FEATURE_POWER_FLOOR) },
 	{ },
 };
 
diff --git a/drivers/thermal/intel/int340x_thermal/processor_thermal_rfim.c b/drivers/thermal/intel/int340x_thermal/processor_thermal_rfim.c
index dad63f2d5f90..3a028b78d9af 100644
--- a/drivers/thermal/intel/int340x_thermal/processor_thermal_rfim.c
+++ b/drivers/thermal/intel/int340x_thermal/processor_thermal_rfim.c
@@ -166,15 +166,18 @@ static const struct mmio_reg adl_dvfs_mmio_regs[] = {
 	{ 0, 0x5A40, 1, 0x1, 0}, /* rfi_disable */
 };
 
+static const struct mapping_table *dlvr_mapping;
+static const struct mmio_reg *dlvr_mmio_regs_table;
+
 #define RFIM_SHOW(suffix, table)\
 static ssize_t suffix##_show(struct device *dev,\
 			      struct device_attribute *attr,\
 			      char *buf)\
 {\
-	const struct mapping_table *mapping = NULL;\
+	const struct mmio_reg *mmio_regs = dlvr_mmio_regs_table;\
+	const struct mapping_table *mapping = dlvr_mapping;\
 	struct proc_thermal_device *proc_priv;\
 	struct pci_dev *pdev = to_pci_dev(dev);\
-	const struct mmio_reg *mmio_regs;\
 	const char **match_strs;\
 	int ret, err;\
 	u32 reg_val;\
@@ -186,12 +189,6 @@ static ssize_t suffix##_show(struct device *dev,\
 		mmio_regs = adl_dvfs_mmio_regs;\
 	} else if (table == 2) { \
 		match_strs = (const char **)dlvr_strings;\
-		if (pdev->device == PCI_DEVICE_ID_INTEL_LNLM_THERMAL) {\
-			mmio_regs = lnl_dlvr_mmio_regs;\
-			mapping = lnl_dlvr_mapping;\
-		} else {\
-			mmio_regs = dlvr_mmio_regs;\
-		} \
 	} else {\
 		match_strs = (const char **)fivr_strings;\
 		mmio_regs = tgl_fivr_mmio_regs;\
@@ -214,12 +211,12 @@ static ssize_t suffix##_store(struct device *dev,\
 			       struct device_attribute *attr,\
 			       const char *buf, size_t count)\
 {\
-	const struct mapping_table *mapping = NULL;\
+	const struct mmio_reg *mmio_regs = dlvr_mmio_regs_table;\
+	const struct mapping_table *mapping = dlvr_mapping;\
 	struct proc_thermal_device *proc_priv;\
 	struct pci_dev *pdev = to_pci_dev(dev);\
 	unsigned int input;\
 	const char **match_strs;\
-	const struct mmio_reg *mmio_regs;\
 	int ret, err;\
 	u32 reg_val;\
 	u32 mask;\
@@ -230,12 +227,6 @@ static ssize_t suffix##_store(struct device *dev,\
 		mmio_regs = adl_dvfs_mmio_regs;\
 	} else if (table == 2) { \
 		match_strs = (const char **)dlvr_strings;\
-		if (pdev->device == PCI_DEVICE_ID_INTEL_LNLM_THERMAL) {\
-			mmio_regs = lnl_dlvr_mmio_regs;\
-			mapping = lnl_dlvr_mapping;\
-		} else {\
-			mmio_regs = dlvr_mmio_regs;\
-		} \
 	} else {\
 		match_strs = (const char **)fivr_strings;\
 		mmio_regs = tgl_fivr_mmio_regs;\
@@ -448,6 +439,16 @@ int proc_thermal_rfim_add(struct pci_dev *pdev, struct proc_thermal_device *proc
 	}
 
 	if (proc_priv->mmio_feature_mask & PROC_THERMAL_FEATURE_DLVR) {
+		switch (pdev->device) {
+		case PCI_DEVICE_ID_INTEL_LNLM_THERMAL:
+		case PCI_DEVICE_ID_INTEL_PTL_THERMAL:
+			dlvr_mmio_regs_table = lnl_dlvr_mmio_regs;
+			dlvr_mapping = lnl_dlvr_mapping;
+			break;
+		default:
+			dlvr_mmio_regs_table = dlvr_mmio_regs;
+			break;
+		}
 		ret = sysfs_create_group(&pdev->dev.kobj, &dlvr_attribute_group);
 		if (ret)
 			return ret;
diff --git a/drivers/tty/serial/msm_serial.c b/drivers/tty/serial/msm_serial.c
index 1b137e068444..3449945493ce 100644
--- a/drivers/tty/serial/msm_serial.c
+++ b/drivers/tty/serial/msm_serial.c
@@ -1746,6 +1746,12 @@ msm_serial_early_console_setup_dm(struct earlycon_device *device,
 	if (!device->port.membase)
 		return -ENODEV;
 
+	/* Disable DM / single-character modes */
+	msm_write(&device->port, 0, UARTDM_DMEN);
+	msm_write(&device->port, MSM_UART_CR_CMD_RESET_RX, MSM_UART_CR);
+	msm_write(&device->port, MSM_UART_CR_CMD_RESET_TX, MSM_UART_CR);
+	msm_write(&device->port, MSM_UART_CR_TX_ENABLE, MSM_UART_CR);
+
 	device->con->write = msm_serial_early_write_dm;
 	return 0;
 }
diff --git a/drivers/tty/serial/sifive.c b/drivers/tty/serial/sifive.c
index 5904a2d4cefa..054a8e630ace 100644
--- a/drivers/tty/serial/sifive.c
+++ b/drivers/tty/serial/sifive.c
@@ -563,8 +563,11 @@ static void sifive_serial_break_ctl(struct uart_port *port, int break_state)
 static int sifive_serial_startup(struct uart_port *port)
 {
 	struct sifive_serial_port *ssp = port_to_sifive_serial_port(port);
+	unsigned long flags;
 
+	uart_port_lock_irqsave(&ssp->port, &flags);
 	__ssp_enable_rxwm(ssp);
+	uart_port_unlock_irqrestore(&ssp->port, flags);
 
 	return 0;
 }
@@ -572,9 +575,12 @@ static int sifive_serial_startup(struct uart_port *port)
 static void sifive_serial_shutdown(struct uart_port *port)
 {
 	struct sifive_serial_port *ssp = port_to_sifive_serial_port(port);
+	unsigned long flags;
 
+	uart_port_lock_irqsave(&ssp->port, &flags);
 	__ssp_disable_rxwm(ssp);
 	__ssp_disable_txwm(ssp);
+	uart_port_unlock_irqrestore(&ssp->port, flags);
 }
 
 /**
diff --git a/drivers/tty/vt/selection.c b/drivers/tty/vt/selection.c
index 0bd6544e30a6..791e2f1f7c0b 100644
--- a/drivers/tty/vt/selection.c
+++ b/drivers/tty/vt/selection.c
@@ -193,13 +193,12 @@ int set_selection_user(const struct tiocl_selection __user *sel,
 		return -EFAULT;
 
 	/*
-	 * TIOCL_SELCLEAR, TIOCL_SELPOINTER and TIOCL_SELMOUSEREPORT are OK to
-	 * use without CAP_SYS_ADMIN as they do not modify the selection.
+	 * TIOCL_SELCLEAR and TIOCL_SELPOINTER are OK to use without
+	 * CAP_SYS_ADMIN as they do not modify the selection.
 	 */
 	switch (v.sel_mode) {
 	case TIOCL_SELCLEAR:
 	case TIOCL_SELPOINTER:
-	case TIOCL_SELMOUSEREPORT:
 		break;
 	default:
 		if (!capable(CAP_SYS_ADMIN))
diff --git a/drivers/ufs/core/ufs-mcq.c b/drivers/ufs/core/ufs-mcq.c
index 240ce135bbfb..f1294c29f484 100644
--- a/drivers/ufs/core/ufs-mcq.c
+++ b/drivers/ufs/core/ufs-mcq.c
@@ -677,13 +677,6 @@ int ufshcd_mcq_abort(struct scsi_cmnd *cmd)
 	unsigned long flags;
 	int err;
 
-	if (!ufshcd_cmd_inflight(lrbp->cmd)) {
-		dev_err(hba->dev,
-			"%s: skip abort. cmd at tag %d already completed.\n",
-			__func__, tag);
-		return FAILED;
-	}
-
 	/* Skip task abort in case previous aborts failed and report failure */
 	if (lrbp->req_abort_skip) {
 		dev_err(hba->dev, "%s: skip abort. tag %d failed earlier\n",
@@ -692,6 +685,11 @@ int ufshcd_mcq_abort(struct scsi_cmnd *cmd)
 	}
 
 	hwq = ufshcd_mcq_req_to_hwq(hba, scsi_cmd_to_rq(cmd));
+	if (!hwq) {
+		dev_err(hba->dev, "%s: skip abort. cmd at tag %d already completed.\n",
+			__func__, tag);
+		return FAILED;
+	}
 
 	if (ufshcd_mcq_sqe_search(hba, hwq, tag)) {
 		/*
diff --git a/drivers/ufs/core/ufs-sysfs.c b/drivers/ufs/core/ufs-sysfs.c
index 90b5ab60f5ae..634cf163f4cb 100644
--- a/drivers/ufs/core/ufs-sysfs.c
+++ b/drivers/ufs/core/ufs-sysfs.c
@@ -466,6 +466,56 @@ static ssize_t critical_health_show(struct device *dev,
 	return sysfs_emit(buf, "%d\n", hba->critical_health_count);
 }
 
+static ssize_t device_lvl_exception_count_show(struct device *dev,
+					       struct device_attribute *attr,
+					       char *buf)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+
+	if (hba->dev_info.wspecversion < 0x410)
+		return -EOPNOTSUPP;
+
+	return sysfs_emit(buf, "%u\n", atomic_read(&hba->dev_lvl_exception_count));
+}
+
+static ssize_t device_lvl_exception_count_store(struct device *dev,
+						struct device_attribute *attr,
+						const char *buf, size_t count)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+	unsigned int value;
+
+	if (kstrtouint(buf, 0, &value))
+		return -EINVAL;
+
+	/* the only supported usecase is to reset the dev_lvl_exception_count */
+	if (value)
+		return -EINVAL;
+
+	atomic_set(&hba->dev_lvl_exception_count, 0);
+
+	return count;
+}
+
+static ssize_t device_lvl_exception_id_show(struct device *dev,
+					    struct device_attribute *attr,
+					    char *buf)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+	u64 exception_id;
+	int err;
+
+	ufshcd_rpm_get_sync(hba);
+	err = ufshcd_read_device_lvl_exception_id(hba, &exception_id);
+	ufshcd_rpm_put_sync(hba);
+
+	if (err)
+		return err;
+
+	hba->dev_lvl_exception_id = exception_id;
+	return sysfs_emit(buf, "%llu\n", exception_id);
+}
+
 static DEVICE_ATTR_RW(rpm_lvl);
 static DEVICE_ATTR_RO(rpm_target_dev_state);
 static DEVICE_ATTR_RO(rpm_target_link_state);
@@ -479,6 +529,8 @@ static DEVICE_ATTR_RW(wb_flush_threshold);
 static DEVICE_ATTR_RW(rtc_update_ms);
 static DEVICE_ATTR_RW(pm_qos_enable);
 static DEVICE_ATTR_RO(critical_health);
+static DEVICE_ATTR_RW(device_lvl_exception_count);
+static DEVICE_ATTR_RO(device_lvl_exception_id);
 
 static struct attribute *ufs_sysfs_ufshcd_attrs[] = {
 	&dev_attr_rpm_lvl.attr,
@@ -494,6 +546,8 @@ static struct attribute *ufs_sysfs_ufshcd_attrs[] = {
 	&dev_attr_rtc_update_ms.attr,
 	&dev_attr_pm_qos_enable.attr,
 	&dev_attr_critical_health.attr,
+	&dev_attr_device_lvl_exception_count.attr,
+	&dev_attr_device_lvl_exception_id.attr,
 	NULL
 };
 
diff --git a/drivers/ufs/core/ufshcd-priv.h b/drivers/ufs/core/ufshcd-priv.h
index 10b4a19a70f1..d0a2c963a27d 100644
--- a/drivers/ufs/core/ufshcd-priv.h
+++ b/drivers/ufs/core/ufshcd-priv.h
@@ -94,6 +94,7 @@ int ufshcd_exec_raw_upiu_cmd(struct ufs_hba *hba,
 			     enum query_opcode desc_op);
 
 int ufshcd_wb_toggle(struct ufs_hba *hba, bool enable);
+int ufshcd_read_device_lvl_exception_id(struct ufs_hba *hba, u64 *exception_id);
 
 /* Wrapper functions for safely calling variant operations */
 static inline const char *ufshcd_get_var_name(struct ufs_hba *hba)
diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index 0534390c2a35..7735421e3991 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -278,6 +278,7 @@ static const struct ufs_dev_quirk ufs_fixups[] = {
 	  .model = UFS_ANY_MODEL,
 	  .quirk = UFS_DEVICE_QUIRK_DELAY_BEFORE_LPM |
 		   UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE |
+		   UFS_DEVICE_QUIRK_PA_HIBER8TIME |
 		   UFS_DEVICE_QUIRK_RECOVERY_FROM_DL_NAC_ERRORS },
 	{ .wmanufacturerid = UFS_VENDOR_SKHYNIX,
 	  .model = UFS_ANY_MODEL,
@@ -3176,16 +3177,10 @@ static int ufshcd_wait_for_dev_cmd(struct ufs_hba *hba,
 	int err;
 
 retry:
-	time_left = wait_for_completion_timeout(hba->dev_cmd.complete,
+	time_left = wait_for_completion_timeout(&hba->dev_cmd.complete,
 						time_left);
 
 	if (likely(time_left)) {
-		/*
-		 * The completion handler called complete() and the caller of
-		 * this function still owns the @lrbp tag so the code below does
-		 * not trigger any race conditions.
-		 */
-		hba->dev_cmd.complete = NULL;
 		err = ufshcd_get_tr_ocs(lrbp, NULL);
 		if (!err)
 			err = ufshcd_dev_cmd_completion(hba, lrbp);
@@ -3199,7 +3194,6 @@ retry:
 			/* successfully cleared the command, retry if needed */
 			if (ufshcd_clear_cmd(hba, lrbp->task_tag) == 0)
 				err = -EAGAIN;
-			hba->dev_cmd.complete = NULL;
 			return err;
 		}
 
@@ -3215,11 +3209,9 @@ retry:
 			spin_lock_irqsave(&hba->outstanding_lock, flags);
 			pending = test_bit(lrbp->task_tag,
 					   &hba->outstanding_reqs);
-			if (pending) {
-				hba->dev_cmd.complete = NULL;
+			if (pending)
 				__clear_bit(lrbp->task_tag,
 					    &hba->outstanding_reqs);
-			}
 			spin_unlock_irqrestore(&hba->outstanding_lock, flags);
 
 			if (!pending) {
@@ -3237,8 +3229,6 @@ retry:
 			spin_lock_irqsave(&hba->outstanding_lock, flags);
 			pending = test_bit(lrbp->task_tag,
 					   &hba->outstanding_reqs);
-			if (pending)
-				hba->dev_cmd.complete = NULL;
 			spin_unlock_irqrestore(&hba->outstanding_lock, flags);
 
 			if (!pending) {
@@ -3272,13 +3262,9 @@ static void ufshcd_dev_man_unlock(struct ufs_hba *hba)
 static int ufshcd_issue_dev_cmd(struct ufs_hba *hba, struct ufshcd_lrb *lrbp,
 			  const u32 tag, int timeout)
 {
-	DECLARE_COMPLETION_ONSTACK(wait);
 	int err;
 
-	hba->dev_cmd.complete = &wait;
-
 	ufshcd_add_query_upiu_trace(hba, UFS_QUERY_SEND, lrbp->ucd_req_ptr);
-
 	ufshcd_send_command(hba, tag, hba->dev_cmd_queue);
 	err = ufshcd_wait_for_dev_cmd(hba, lrbp, timeout);
 
@@ -5585,12 +5571,12 @@ void ufshcd_compl_one_cqe(struct ufs_hba *hba, int task_tag,
 		ufshcd_release_scsi_cmd(hba, lrbp);
 		/* Do not touch lrbp after scsi done */
 		scsi_done(cmd);
-	} else if (hba->dev_cmd.complete) {
+	} else {
 		if (cqe) {
 			ocs = le32_to_cpu(cqe->status) & MASK_OCS;
 			lrbp->utr_descriptor_ptr->header.ocs = ocs;
 		}
-		complete(hba->dev_cmd.complete);
+		complete(&hba->dev_cmd.complete);
 	}
 }
 
@@ -5692,6 +5678,8 @@ static void ufshcd_mcq_compl_pending_transfer(struct ufs_hba *hba,
 			continue;
 
 		hwq = ufshcd_mcq_req_to_hwq(hba, scsi_cmd_to_rq(cmd));
+		if (!hwq)
+			continue;
 
 		if (force_compl) {
 			ufshcd_mcq_compl_all_cqes_lock(hba, hwq);
@@ -6013,6 +6001,42 @@ out:
 				__func__, err);
 }
 
+int ufshcd_read_device_lvl_exception_id(struct ufs_hba *hba, u64 *exception_id)
+{
+	struct utp_upiu_query_v4_0 *upiu_resp;
+	struct ufs_query_req *request = NULL;
+	struct ufs_query_res *response = NULL;
+	int err;
+
+	if (hba->dev_info.wspecversion < 0x410)
+		return -EOPNOTSUPP;
+
+	ufshcd_hold(hba);
+	mutex_lock(&hba->dev_cmd.lock);
+
+	ufshcd_init_query(hba, &request, &response,
+			  UPIU_QUERY_OPCODE_READ_ATTR,
+			  QUERY_ATTR_IDN_DEV_LVL_EXCEPTION_ID, 0, 0);
+
+	request->query_func = UPIU_QUERY_FUNC_STANDARD_READ_REQUEST;
+
+	err = ufshcd_exec_dev_cmd(hba, DEV_CMD_TYPE_QUERY, QUERY_REQ_TIMEOUT);
+
+	if (err) {
+		dev_err(hba->dev, "%s: failed to read device level exception %d\n",
+			__func__, err);
+		goto out;
+	}
+
+	upiu_resp = (struct utp_upiu_query_v4_0 *)response;
+	*exception_id = get_unaligned_be64(&upiu_resp->osf3);
+out:
+	mutex_unlock(&hba->dev_cmd.lock);
+	ufshcd_release(hba);
+
+	return err;
+}
+
 static int __ufshcd_wb_toggle(struct ufs_hba *hba, bool set, enum flag_idn idn)
 {
 	u8 index;
@@ -6083,7 +6107,7 @@ int ufshcd_wb_toggle_buf_flush(struct ufs_hba *hba, bool enable)
 	return ret;
 }
 
-static bool ufshcd_wb_presrv_usrspc_keep_vcc_on(struct ufs_hba *hba,
+static bool ufshcd_wb_curr_buff_threshold_check(struct ufs_hba *hba,
 						u32 avail_buf)
 {
 	u32 cur_buf;
@@ -6165,15 +6189,13 @@ static bool ufshcd_wb_need_flush(struct ufs_hba *hba)
 	}
 
 	/*
-	 * The ufs device needs the vcc to be ON to flush.
 	 * With user-space reduction enabled, it's enough to enable flush
 	 * by checking only the available buffer. The threshold
 	 * defined here is > 90% full.
 	 * With user-space preserved enabled, the current-buffer
 	 * should be checked too because the wb buffer size can reduce
 	 * when disk tends to be full. This info is provided by current
-	 * buffer (dCurrentWriteBoosterBufferSize). There's no point in
-	 * keeping vcc on when current buffer is empty.
+	 * buffer (dCurrentWriteBoosterBufferSize).
 	 */
 	index = ufshcd_wb_get_query_index(hba);
 	ret = ufshcd_query_attr_retry(hba, UPIU_QUERY_OPCODE_READ_ATTR,
@@ -6188,7 +6210,7 @@ static bool ufshcd_wb_need_flush(struct ufs_hba *hba)
 	if (!hba->dev_info.b_presrv_uspc_en)
 		return avail_buf <= UFS_WB_BUF_REMAIN_PERCENT(10);
 
-	return ufshcd_wb_presrv_usrspc_keep_vcc_on(hba, avail_buf);
+	return ufshcd_wb_curr_buff_threshold_check(hba, avail_buf);
 }
 
 static void ufshcd_rpm_dev_flush_recheck_work(struct work_struct *work)
@@ -6240,6 +6262,11 @@ static void ufshcd_exception_event_handler(struct work_struct *work)
 		sysfs_notify(&hba->dev->kobj, NULL, "critical_health");
 	}
 
+	if (status & hba->ee_drv_mask & MASK_EE_DEV_LVL_EXCEPTION) {
+		atomic_inc(&hba->dev_lvl_exception_count);
+		sysfs_notify(&hba->dev->kobj, NULL, "device_lvl_exception_count");
+	}
+
 	ufs_debugfs_exception_event(hba, status);
 }
 
@@ -7238,8 +7265,6 @@ static int ufshcd_issue_devman_upiu_cmd(struct ufs_hba *hba,
 			err = -EINVAL;
 		}
 	}
-	ufshcd_add_query_upiu_trace(hba, err ? UFS_QUERY_ERR : UFS_QUERY_COMP,
-				    (struct utp_upiu_req *)lrbp->ucd_rsp_ptr);
 
 	return err;
 }
@@ -8139,6 +8164,22 @@ static void ufshcd_temp_notif_probe(struct ufs_hba *hba, const u8 *desc_buf)
 	}
 }
 
+static void ufshcd_device_lvl_exception_probe(struct ufs_hba *hba, u8 *desc_buf)
+{
+	u32 ext_ufs_feature;
+
+	if (hba->dev_info.wspecversion < 0x410)
+		return;
+
+	ext_ufs_feature = get_unaligned_be32(desc_buf +
+				DEVICE_DESC_PARAM_EXT_UFS_FEATURE_SUP);
+	if (!(ext_ufs_feature & UFS_DEV_LVL_EXCEPTION_SUP))
+		return;
+
+	atomic_set(&hba->dev_lvl_exception_count, 0);
+	ufshcd_enable_ee(hba, MASK_EE_DEV_LVL_EXCEPTION);
+}
+
 static void ufshcd_set_rtt(struct ufs_hba *hba)
 {
 	struct ufs_dev_info *dev_info = &hba->dev_info;
@@ -8339,6 +8380,8 @@ static int ufs_get_device_desc(struct ufs_hba *hba)
 
 	ufs_init_rtc(hba, desc_buf);
 
+	ufshcd_device_lvl_exception_probe(hba, desc_buf);
+
 	/*
 	 * ufshcd_read_string_desc returns size of the string
 	 * reset the error value
@@ -8428,6 +8471,31 @@ out:
 	return ret;
 }
 
+/**
+ * ufshcd_quirk_override_pa_h8time - Ensures proper adjustment of PA_HIBERN8TIME.
+ * @hba: per-adapter instance
+ *
+ * Some UFS devices require specific adjustments to the PA_HIBERN8TIME parameter
+ * to ensure proper hibernation timing. This function retrieves the current
+ * PA_HIBERN8TIME value and increments it by 100us.
+ */
+static void ufshcd_quirk_override_pa_h8time(struct ufs_hba *hba)
+{
+	u32 pa_h8time;
+	int ret;
+
+	ret = ufshcd_dme_get(hba, UIC_ARG_MIB(PA_HIBERN8TIME), &pa_h8time);
+	if (ret) {
+		dev_err(hba->dev, "Failed to get PA_HIBERN8TIME: %d\n", ret);
+		return;
+	}
+
+	/* Increment by 1 to increase hibernation time by 100 µs */
+	ret = ufshcd_dme_set(hba, UIC_ARG_MIB(PA_HIBERN8TIME), pa_h8time + 1);
+	if (ret)
+		dev_err(hba->dev, "Failed updating PA_HIBERN8TIME: %d\n", ret);
+}
+
 static void ufshcd_tune_unipro_params(struct ufs_hba *hba)
 {
 	ufshcd_vops_apply_dev_quirks(hba);
@@ -8438,6 +8506,9 @@ static void ufshcd_tune_unipro_params(struct ufs_hba *hba)
 
 	if (hba->dev_quirks & UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE)
 		ufshcd_quirk_tune_host_pa_tactivate(hba);
+
+	if (hba->dev_quirks & UFS_DEVICE_QUIRK_PA_HIBER8TIME)
+		ufshcd_quirk_override_pa_h8time(hba);
 }
 
 static void ufshcd_clear_dbg_ufs_stats(struct ufs_hba *hba)
@@ -10490,6 +10561,8 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 						UFS_SLEEP_PWR_MODE,
 						UIC_LINK_HIBERN8_STATE);
 
+	init_completion(&hba->dev_cmd.complete);
+
 	err = ufshcd_hba_init(hba);
 	if (err)
 		goto out_error;
diff --git a/drivers/ufs/host/ufs-exynos.c b/drivers/ufs/host/ufs-exynos.c
index d7539cda97da..3e545af536e5 100644
--- a/drivers/ufs/host/ufs-exynos.c
+++ b/drivers/ufs/host/ufs-exynos.c
@@ -34,7 +34,7 @@
  * Exynos's Vendor specific registers for UFSHCI
  */
 #define HCI_TXPRDT_ENTRY_SIZE	0x00
-#define PRDT_PREFECT_EN		BIT(31)
+#define PRDT_PREFETCH_EN	BIT(31)
 #define HCI_RXPRDT_ENTRY_SIZE	0x04
 #define HCI_1US_TO_CNT_VAL	0x0C
 #define CNT_VAL_1US_MASK	0x3FF
@@ -92,11 +92,16 @@
 				 UIC_TRANSPORT_NO_CONNECTION_RX |\
 				 UIC_TRANSPORT_BAD_TC)
 
-/* FSYS UFS Shareability */
-#define UFS_WR_SHARABLE		BIT(2)
-#define UFS_RD_SHARABLE		BIT(1)
-#define UFS_SHARABLE		(UFS_WR_SHARABLE | UFS_RD_SHARABLE)
-#define UFS_SHAREABILITY_OFFSET	0x710
+/* UFS Shareability */
+#define UFS_EXYNOSAUTO_WR_SHARABLE	BIT(2)
+#define UFS_EXYNOSAUTO_RD_SHARABLE	BIT(1)
+#define UFS_EXYNOSAUTO_SHARABLE		(UFS_EXYNOSAUTO_WR_SHARABLE | \
+					 UFS_EXYNOSAUTO_RD_SHARABLE)
+#define UFS_GS101_WR_SHARABLE		BIT(1)
+#define UFS_GS101_RD_SHARABLE		BIT(0)
+#define UFS_GS101_SHARABLE		(UFS_GS101_WR_SHARABLE | \
+					 UFS_GS101_RD_SHARABLE)
+#define UFS_SHAREABILITY_OFFSET		0x710
 
 /* Multi-host registers */
 #define MHCTRL			0xC4
@@ -209,8 +214,8 @@ static int exynos_ufs_shareability(struct exynos_ufs *ufs)
 	/* IO Coherency setting */
 	if (ufs->sysreg) {
 		return regmap_update_bits(ufs->sysreg,
-					  ufs->shareability_reg_offset,
-					  UFS_SHARABLE, UFS_SHARABLE);
+					  ufs->iocc_offset,
+					  ufs->iocc_mask, ufs->iocc_val);
 	}
 
 	return 0;
@@ -957,6 +962,12 @@ static int exynos_ufs_phy_init(struct exynos_ufs *ufs)
 	}
 
 	phy_set_bus_width(generic_phy, ufs->avail_ln_rx);
+
+	if (generic_phy->power_count) {
+		phy_power_off(generic_phy);
+		phy_exit(generic_phy);
+	}
+
 	ret = phy_init(generic_phy);
 	if (ret) {
 		dev_err(hba->dev, "%s: phy init failed, ret = %d\n",
@@ -1049,9 +1060,14 @@ static int exynos_ufs_pre_link(struct ufs_hba *hba)
 	exynos_ufs_config_intr(ufs, DFES_DEF_L4_ERRS, UNIPRO_L4);
 	exynos_ufs_set_unipro_pclk_div(ufs);
 
+	exynos_ufs_setup_clocks(hba, true, PRE_CHANGE);
+
 	/* unipro */
 	exynos_ufs_config_unipro(ufs);
 
+	if (ufs->drv_data->pre_link)
+		ufs->drv_data->pre_link(ufs);
+
 	/* m-phy */
 	exynos_ufs_phy_init(ufs);
 	if (!(ufs->opts & EXYNOS_UFS_OPT_SKIP_CONFIG_PHY_ATTR)) {
@@ -1059,11 +1075,6 @@ static int exynos_ufs_pre_link(struct ufs_hba *hba)
 		exynos_ufs_config_phy_cap_attr(ufs);
 	}
 
-	exynos_ufs_setup_clocks(hba, true, PRE_CHANGE);
-
-	if (ufs->drv_data->pre_link)
-		ufs->drv_data->pre_link(ufs);
-
 	return 0;
 }
 
@@ -1087,12 +1098,17 @@ static int exynos_ufs_post_link(struct ufs_hba *hba)
 	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
 	struct phy *generic_phy = ufs->phy;
 	struct exynos_ufs_uic_attr *attr = ufs->drv_data->uic_attr;
+	u32 val = ilog2(DATA_UNIT_SIZE);
 
 	exynos_ufs_establish_connt(ufs);
 	exynos_ufs_fit_aggr_timeout(ufs);
 
 	hci_writel(ufs, 0xa, HCI_DATA_REORDER);
-	hci_writel(ufs, ilog2(DATA_UNIT_SIZE), HCI_TXPRDT_ENTRY_SIZE);
+
+	if (hba->caps & UFSHCD_CAP_CRYPTO)
+		val |= PRDT_PREFETCH_EN;
+	hci_writel(ufs, val, HCI_TXPRDT_ENTRY_SIZE);
+
 	hci_writel(ufs, ilog2(DATA_UNIT_SIZE), HCI_RXPRDT_ENTRY_SIZE);
 	hci_writel(ufs, (1 << hba->nutrs) - 1, HCI_UTRL_NEXUS_TYPE);
 	hci_writel(ufs, (1 << hba->nutmrs) - 1, HCI_UTMRL_NEXUS_TYPE);
@@ -1168,12 +1184,22 @@ static int exynos_ufs_parse_dt(struct device *dev, struct exynos_ufs *ufs)
 		ufs->sysreg = NULL;
 	else {
 		if (of_property_read_u32_index(np, "samsung,sysreg", 1,
-					       &ufs->shareability_reg_offset)) {
+					       &ufs->iocc_offset)) {
 			dev_warn(dev, "can't get an offset from sysreg. Set to default value\n");
-			ufs->shareability_reg_offset = UFS_SHAREABILITY_OFFSET;
+			ufs->iocc_offset = UFS_SHAREABILITY_OFFSET;
 		}
 	}
 
+	ufs->iocc_mask = ufs->drv_data->iocc_mask;
+	/*
+	 * no 'dma-coherent' property means the descriptors are
+	 * non-cacheable so iocc shareability should be disabled.
+	 */
+	if (of_dma_is_coherent(dev->of_node))
+		ufs->iocc_val = ufs->iocc_mask;
+	else
+		ufs->iocc_val = 0;
+
 	ufs->pclk_avail_min = PCLK_AVAIL_MIN;
 	ufs->pclk_avail_max = PCLK_AVAIL_MAX;
 
@@ -1497,6 +1523,14 @@ out:
 	return ret;
 }
 
+static void exynos_ufs_exit(struct ufs_hba *hba)
+{
+	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
+
+	phy_power_off(ufs->phy);
+	phy_exit(ufs->phy);
+}
+
 static int exynos_ufs_host_reset(struct ufs_hba *hba)
 {
 	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
@@ -1667,6 +1701,12 @@ static void exynos_ufs_hibern8_notify(struct ufs_hba *hba,
 	}
 }
 
+static int gs101_ufs_suspend(struct exynos_ufs *ufs)
+{
+	hci_writel(ufs, 0 << 0, HCI_GPIO_OUT);
+	return 0;
+}
+
 static int exynos_ufs_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op,
 	enum ufs_notify_change_status status)
 {
@@ -1675,6 +1715,9 @@ static int exynos_ufs_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op,
 	if (status == PRE_CHANGE)
 		return 0;
 
+	if (ufs->drv_data->suspend)
+		ufs->drv_data->suspend(ufs);
+
 	if (!ufshcd_is_link_active(hba))
 		phy_power_off(ufs->phy);
 
@@ -1952,6 +1995,7 @@ static int gs101_ufs_pre_pwr_change(struct exynos_ufs *ufs,
 static const struct ufs_hba_variant_ops ufs_hba_exynos_ops = {
 	.name				= "exynos_ufs",
 	.init				= exynos_ufs_init,
+	.exit				= exynos_ufs_exit,
 	.hce_enable_notify		= exynos_ufs_hce_enable_notify,
 	.link_startup_notify		= exynos_ufs_link_startup_notify,
 	.pwr_change_notify		= exynos_ufs_pwr_change_notify,
@@ -1990,13 +2034,7 @@ static int exynos_ufs_probe(struct platform_device *pdev)
 
 static void exynos_ufs_remove(struct platform_device *pdev)
 {
-	struct ufs_hba *hba =  platform_get_drvdata(pdev);
-	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
-
 	ufshcd_pltfrm_remove(pdev);
-
-	phy_power_off(ufs->phy);
-	phy_exit(ufs->phy);
 }
 
 static struct exynos_ufs_uic_attr exynos7_uic_attr = {
@@ -2035,6 +2073,7 @@ static const struct exynos_ufs_drv_data exynosauto_ufs_drvs = {
 	.opts			= EXYNOS_UFS_OPT_BROKEN_AUTO_CLK_CTRL |
 				  EXYNOS_UFS_OPT_SKIP_CONFIG_PHY_ATTR |
 				  EXYNOS_UFS_OPT_BROKEN_RX_SEL_IDX,
+	.iocc_mask		= UFS_EXYNOSAUTO_SHARABLE,
 	.drv_init		= exynosauto_ufs_drv_init,
 	.post_hce_enable	= exynosauto_ufs_post_hce_enable,
 	.pre_link		= exynosauto_ufs_pre_link,
@@ -2136,10 +2175,12 @@ static const struct exynos_ufs_drv_data gs101_ufs_drvs = {
 	.opts			= EXYNOS_UFS_OPT_SKIP_CONFIG_PHY_ATTR |
 				  EXYNOS_UFS_OPT_UFSPR_SECURE |
 				  EXYNOS_UFS_OPT_TIMER_TICK_SELECT,
+	.iocc_mask		= UFS_GS101_SHARABLE,
 	.drv_init		= gs101_ufs_drv_init,
 	.pre_link		= gs101_ufs_pre_link,
 	.post_link		= gs101_ufs_post_link,
 	.pre_pwr_change		= gs101_ufs_pre_pwr_change,
+	.suspend		= gs101_ufs_suspend,
 };
 
 static const struct of_device_id exynos_ufs_of_match[] = {
diff --git a/drivers/ufs/host/ufs-exynos.h b/drivers/ufs/host/ufs-exynos.h
index aac517276189..abe7e472759e 100644
--- a/drivers/ufs/host/ufs-exynos.h
+++ b/drivers/ufs/host/ufs-exynos.h
@@ -181,6 +181,7 @@ struct exynos_ufs_drv_data {
 	struct exynos_ufs_uic_attr *uic_attr;
 	unsigned int quirks;
 	unsigned int opts;
+	u32 iocc_mask;
 	/* SoC's specific operations */
 	int (*drv_init)(struct exynos_ufs *ufs);
 	int (*pre_link)(struct exynos_ufs *ufs);
@@ -191,6 +192,7 @@ struct exynos_ufs_drv_data {
 			       const struct ufs_pa_layer_attr *pwr);
 	int (*pre_hce_enable)(struct exynos_ufs *ufs);
 	int (*post_hce_enable)(struct exynos_ufs *ufs);
+	int (*suspend)(struct exynos_ufs *ufs);
 };
 
 struct ufs_phy_time_cfg {
@@ -230,7 +232,9 @@ struct exynos_ufs {
 	ktime_t entry_hibern8_t;
 	const struct exynos_ufs_drv_data *drv_data;
 	struct regmap *sysreg;
-	u32 shareability_reg_offset;
+	u32 iocc_offset;
+	u32 iocc_mask;
+	u32 iocc_val;
 
 	u32 opts;
 #define EXYNOS_UFS_OPT_HAS_APB_CLK_CTRL		BIT(0)
diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c
index 1b37449fbffc..c0761ccc1381 100644
--- a/drivers/ufs/host/ufs-qcom.c
+++ b/drivers/ufs/host/ufs-qcom.c
@@ -33,6 +33,10 @@
 	((((c) >> 16) & MCQ_QCFGPTR_MASK) * MCQ_QCFGPTR_UNIT)
 #define MCQ_QCFG_SIZE	0x40
 
+/* De-emphasis for gear-5 */
+#define DEEMPHASIS_3_5_dB	0x04
+#define NO_DEEMPHASIS		0x0
+
 enum {
 	TSTBUS_UAWM,
 	TSTBUS_UARM,
@@ -795,6 +799,23 @@ static int ufs_qcom_icc_update_bw(struct ufs_qcom_host *host)
 	return ufs_qcom_icc_set_bw(host, bw_table.mem_bw, bw_table.cfg_bw);
 }
 
+static void ufs_qcom_set_tx_hs_equalizer(struct ufs_hba *hba, u32 gear, u32 tx_lanes)
+{
+	u32 equalizer_val;
+	int ret, i;
+
+	/* Determine the equalizer value based on the gear */
+	equalizer_val = (gear == 5) ? DEEMPHASIS_3_5_dB : NO_DEEMPHASIS;
+
+	for (i = 0; i < tx_lanes; i++) {
+		ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(TX_HS_EQUALIZER, i),
+				     equalizer_val);
+		if (ret)
+			dev_err(hba->dev, "%s: failed equalizer lane %d\n",
+				__func__, i);
+	}
+}
+
 static int ufs_qcom_pwr_change_notify(struct ufs_hba *hba,
 				enum ufs_notify_change_status status,
 				const struct ufs_pa_layer_attr *dev_max_params,
@@ -846,6 +867,11 @@ static int ufs_qcom_pwr_change_notify(struct ufs_hba *hba,
 						dev_req_params->gear_tx,
 						PA_INITIAL_ADAPT);
 		}
+
+		if (hba->dev_quirks & UFS_DEVICE_QUIRK_PA_TX_DEEMPHASIS_TUNING)
+			ufs_qcom_set_tx_hs_equalizer(hba,
+					dev_req_params->gear_tx, dev_req_params->lane_tx);
+
 		break;
 	case POST_CHANGE:
 		if (ufs_qcom_cfg_timers(hba, false)) {
@@ -893,6 +919,16 @@ static int ufs_qcom_quirk_host_pa_saveconfigtime(struct ufs_hba *hba)
 			    (pa_vs_config_reg1 | (1 << 12)));
 }
 
+static void ufs_qcom_override_pa_tx_hsg1_sync_len(struct ufs_hba *hba)
+{
+	int err;
+
+	err = ufshcd_dme_peer_set(hba, UIC_ARG_MIB(PA_TX_HSG1_SYNC_LENGTH),
+				  PA_TX_HSG1_SYNC_LENGTH_VAL);
+	if (err)
+		dev_err(hba->dev, "Failed (%d) set PA_TX_HSG1_SYNC_LENGTH\n", err);
+}
+
 static int ufs_qcom_apply_dev_quirks(struct ufs_hba *hba)
 {
 	int err = 0;
@@ -900,6 +936,9 @@ static int ufs_qcom_apply_dev_quirks(struct ufs_hba *hba)
 	if (hba->dev_quirks & UFS_DEVICE_QUIRK_HOST_PA_SAVECONFIGTIME)
 		err = ufs_qcom_quirk_host_pa_saveconfigtime(hba);
 
+	if (hba->dev_quirks & UFS_DEVICE_QUIRK_PA_TX_HSG1_SYNC_LENGTH)
+		ufs_qcom_override_pa_tx_hsg1_sync_len(hba);
+
 	return err;
 }
 
@@ -914,6 +953,10 @@ static struct ufs_dev_quirk ufs_qcom_dev_fixups[] = {
 	{ .wmanufacturerid = UFS_VENDOR_WDC,
 	  .model = UFS_ANY_MODEL,
 	  .quirk = UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE },
+	{ .wmanufacturerid = UFS_VENDOR_SAMSUNG,
+	  .model = UFS_ANY_MODEL,
+	  .quirk = UFS_DEVICE_QUIRK_PA_TX_HSG1_SYNC_LENGTH |
+		   UFS_DEVICE_QUIRK_PA_TX_DEEMPHASIS_TUNING },
 	{}
 };
 
diff --git a/drivers/ufs/host/ufs-qcom.h b/drivers/ufs/host/ufs-qcom.h
index d0e6ec9128e7..05d4cb569c50 100644
--- a/drivers/ufs/host/ufs-qcom.h
+++ b/drivers/ufs/host/ufs-qcom.h
@@ -122,8 +122,11 @@ enum {
 				 TMRLUT_HW_CGC_EN | OCSC_HW_CGC_EN)
 
 /* QUniPro Vendor specific attributes */
+#define PA_TX_HSG1_SYNC_LENGTH	0x1552
 #define PA_VS_CONFIG_REG1	0x9000
 #define DME_VS_CORE_CLK_CTRL	0xD002
+#define TX_HS_EQUALIZER		0x0037
+
 /* bit and mask definitions for DME_VS_CORE_CLK_CTRL attribute */
 #define CLK_1US_CYCLES_MASK_V4				GENMASK(27, 16)
 #define CLK_1US_CYCLES_MASK				GENMASK(7, 0)
@@ -141,6 +144,21 @@ enum {
 #define UNIPRO_CORE_CLK_FREQ_201_5_MHZ         202
 #define UNIPRO_CORE_CLK_FREQ_403_MHZ           403
 
+/* TX_HSG1_SYNC_LENGTH attr value */
+#define PA_TX_HSG1_SYNC_LENGTH_VAL	0x4A
+
+/*
+ * Some ufs device vendors need a different TSync length.
+ * Enable this quirk to give an additional TX_HS_SYNC_LENGTH.
+ */
+#define UFS_DEVICE_QUIRK_PA_TX_HSG1_SYNC_LENGTH		BIT(16)
+
+/*
+ * Some ufs device vendors need a different Deemphasis setting.
+ * Enable this quirk to tune TX Deemphasis parameters.
+ */
+#define UFS_DEVICE_QUIRK_PA_TX_DEEMPHASIS_TUNING	BIT(17)
+
 /* ICE allocator type to share AES engines among TX stream and RX stream */
 #define ICE_ALLOCATOR_TYPE 2
 
diff --git a/drivers/usb/cdns3/cdns3-gadget.c b/drivers/usb/cdns3/cdns3-gadget.c
index 694aa1457739..d9d8dc05b235 100644
--- a/drivers/usb/cdns3/cdns3-gadget.c
+++ b/drivers/usb/cdns3/cdns3-gadget.c
@@ -1963,6 +1963,7 @@ static irqreturn_t cdns3_device_thread_irq_handler(int irq, void *data)
 	unsigned int bit;
 	unsigned long reg;
 
+	local_bh_disable();
 	spin_lock_irqsave(&priv_dev->lock, flags);
 
 	reg = readl(&priv_dev->regs->usb_ists);
@@ -2004,6 +2005,7 @@ static irqreturn_t cdns3_device_thread_irq_handler(int irq, void *data)
 irqend:
 	writel(~0, &priv_dev->regs->ep_ien);
 	spin_unlock_irqrestore(&priv_dev->lock, flags);
+	local_bh_enable();
 
 	return ret;
 }
diff --git a/drivers/usb/chipidea/ci_hdrc_imx.c b/drivers/usb/chipidea/ci_hdrc_imx.c
index 1a7fc638213e..4f8bfd242b59 100644
--- a/drivers/usb/chipidea/ci_hdrc_imx.c
+++ b/drivers/usb/chipidea/ci_hdrc_imx.c
@@ -336,6 +336,13 @@ static int ci_hdrc_imx_notify_event(struct ci_hdrc *ci, unsigned int event)
 	return ret;
 }
 
+static void ci_hdrc_imx_disable_regulator(void *arg)
+{
+	struct ci_hdrc_imx_data *data = arg;
+
+	regulator_disable(data->hsic_pad_regulator);
+}
+
 static int ci_hdrc_imx_probe(struct platform_device *pdev)
 {
 	struct ci_hdrc_imx_data *data;
@@ -394,6 +401,13 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev)
 					"Failed to enable HSIC pad regulator\n");
 				goto err_put;
 			}
+			ret = devm_add_action_or_reset(dev,
+					ci_hdrc_imx_disable_regulator, data);
+			if (ret) {
+				dev_err(dev,
+					"Failed to add regulator devm action\n");
+				goto err_put;
+			}
 		}
 	}
 
@@ -432,11 +446,11 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev)
 
 	ret = imx_get_clks(dev);
 	if (ret)
-		goto disable_hsic_regulator;
+		goto qos_remove_request;
 
 	ret = imx_prepare_enable_clks(dev);
 	if (ret)
-		goto disable_hsic_regulator;
+		goto qos_remove_request;
 
 	ret = clk_prepare_enable(data->clk_wakeup);
 	if (ret)
@@ -470,7 +484,11 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev)
 	    of_usb_get_phy_mode(np) == USBPHY_INTERFACE_MODE_ULPI) {
 		pdata.flags |= CI_HDRC_OVERRIDE_PHY_CONTROL;
 		data->override_phy_control = true;
-		usb_phy_init(pdata.usb_phy);
+		ret = usb_phy_init(pdata.usb_phy);
+		if (ret) {
+			dev_err(dev, "Failed to init phy\n");
+			goto err_clk;
+		}
 	}
 
 	if (pdata.flags & CI_HDRC_SUPPORTS_RUNTIME_PM)
@@ -479,7 +497,7 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev)
 	ret = imx_usbmisc_init(data->usbmisc_data);
 	if (ret) {
 		dev_err(dev, "usbmisc init failed, ret=%d\n", ret);
-		goto err_clk;
+		goto phy_shutdown;
 	}
 
 	data->ci_pdev = ci_hdrc_add_device(dev,
@@ -488,7 +506,7 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev)
 	if (IS_ERR(data->ci_pdev)) {
 		ret = PTR_ERR(data->ci_pdev);
 		dev_err_probe(dev, ret, "ci_hdrc_add_device failed\n");
-		goto err_clk;
+		goto phy_shutdown;
 	}
 
 	if (data->usbmisc_data) {
@@ -522,19 +540,20 @@ static int ci_hdrc_imx_probe(struct platform_device *pdev)
 
 disable_device:
 	ci_hdrc_remove_device(data->ci_pdev);
+phy_shutdown:
+	if (data->override_phy_control)
+		usb_phy_shutdown(data->phy);
 err_clk:
 	clk_disable_unprepare(data->clk_wakeup);
 err_wakeup_clk:
 	imx_disable_unprepare_clks(dev);
-disable_hsic_regulator:
-	if (data->hsic_pad_regulator)
-		/* don't overwrite original ret (cf. EPROBE_DEFER) */
-		regulator_disable(data->hsic_pad_regulator);
+qos_remove_request:
 	if (pdata.flags & CI_HDRC_PMQOS)
 		cpu_latency_qos_remove_request(&data->pm_qos_req);
 	data->ci_pdev = NULL;
 err_put:
-	put_device(data->usbmisc_data->dev);
+	if (data->usbmisc_data)
+		put_device(data->usbmisc_data->dev);
 	return ret;
 }
 
@@ -556,10 +575,9 @@ static void ci_hdrc_imx_remove(struct platform_device *pdev)
 		clk_disable_unprepare(data->clk_wakeup);
 		if (data->plat_data->flags & CI_HDRC_PMQOS)
 			cpu_latency_qos_remove_request(&data->pm_qos_req);
-		if (data->hsic_pad_regulator)
-			regulator_disable(data->hsic_pad_regulator);
 	}
-	put_device(data->usbmisc_data->dev);
+	if (data->usbmisc_data)
+		put_device(data->usbmisc_data->dev);
 }
 
 static void ci_hdrc_imx_shutdown(struct platform_device *pdev)
diff --git a/drivers/usb/class/cdc-wdm.c b/drivers/usb/class/cdc-wdm.c
index 86ee39db013f..16e7fa4d488d 100644
--- a/drivers/usb/class/cdc-wdm.c
+++ b/drivers/usb/class/cdc-wdm.c
@@ -726,7 +726,7 @@ static int wdm_open(struct inode *inode, struct file *file)
 		rv = -EBUSY;
 		goto out;
 	}
-
+	smp_rmb(); /* ordered against wdm_wwan_port_stop() */
 	rv = usb_autopm_get_interface(desc->intf);
 	if (rv < 0) {
 		dev_err(&desc->intf->dev, "Error autopm - %d\n", rv);
@@ -829,6 +829,7 @@ static struct usb_class_driver wdm_class = {
 static int wdm_wwan_port_start(struct wwan_port *port)
 {
 	struct wdm_device *desc = wwan_port_get_drvdata(port);
+	int rv;
 
 	/* The interface is both exposed via the WWAN framework and as a
 	 * legacy usbmisc chardev. If chardev is already open, just fail
@@ -848,7 +849,15 @@ static int wdm_wwan_port_start(struct wwan_port *port)
 	wwan_port_txon(port);
 
 	/* Start getting events */
-	return usb_submit_urb(desc->validity, GFP_KERNEL);
+	rv = usb_submit_urb(desc->validity, GFP_KERNEL);
+	if (rv < 0) {
+		wwan_port_txoff(port);
+		desc->manage_power(desc->intf, 0);
+		/* this must be last lest we race with chardev open */
+		clear_bit(WDM_WWAN_IN_USE, &desc->flags);
+	}
+
+	return rv;
 }
 
 static void wdm_wwan_port_stop(struct wwan_port *port)
@@ -859,8 +868,10 @@ static void wdm_wwan_port_stop(struct wwan_port *port)
 	poison_urbs(desc);
 	desc->manage_power(desc->intf, 0);
 	clear_bit(WDM_READ, &desc->flags);
-	clear_bit(WDM_WWAN_IN_USE, &desc->flags);
 	unpoison_urbs(desc);
+	smp_wmb(); /* ordered against wdm_open() */
+	/* this must be last lest we open a poisoned device */
+	clear_bit(WDM_WWAN_IN_USE, &desc->flags);
 }
 
 static void wdm_wwan_port_tx_complete(struct urb *urb)
@@ -868,7 +879,7 @@ static void wdm_wwan_port_tx_complete(struct urb *urb)
 	struct sk_buff *skb = urb->context;
 	struct wdm_device *desc = skb_shinfo(skb)->destructor_arg;
 
-	usb_autopm_put_interface(desc->intf);
+	usb_autopm_put_interface_async(desc->intf);
 	wwan_port_txon(desc->wwanp);
 	kfree_skb(skb);
 }
@@ -898,7 +909,7 @@ static int wdm_wwan_port_tx(struct wwan_port *port, struct sk_buff *skb)
 	req->bRequestType = (USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE);
 	req->bRequest = USB_CDC_SEND_ENCAPSULATED_COMMAND;
 	req->wValue = 0;
-	req->wIndex = desc->inum;
+	req->wIndex = desc->inum; /* already converted */
 	req->wLength = cpu_to_le16(skb->len);
 
 	skb_shinfo(skb)->destructor_arg = desc;
diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c
index 8efbacc5bc34..36d3df7d040c 100644
--- a/drivers/usb/core/quirks.c
+++ b/drivers/usb/core/quirks.c
@@ -369,6 +369,9 @@ static const struct usb_device_id usb_quirk_list[] = {
 	{ USB_DEVICE(0x0781, 0x5583), .driver_info = USB_QUIRK_NO_LPM },
 	{ USB_DEVICE(0x0781, 0x5591), .driver_info = USB_QUIRK_NO_LPM },
 
+	/* SanDisk Corp. SanDisk 3.2Gen1 */
+	{ USB_DEVICE(0x0781, 0x55a3), .driver_info = USB_QUIRK_DELAY_INIT },
+
 	/* Realforce 87U Keyboard */
 	{ USB_DEVICE(0x0853, 0x011b), .driver_info = USB_QUIRK_NO_LPM },
 
@@ -383,6 +386,9 @@ static const struct usb_device_id usb_quirk_list[] = {
 	{ USB_DEVICE(0x0904, 0x6103), .driver_info =
 			USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL },
 
+	/* Silicon Motion Flash Drive */
+	{ USB_DEVICE(0x090c, 0x1000), .driver_info = USB_QUIRK_DELAY_INIT },
+
 	/* Sound Devices USBPre2 */
 	{ USB_DEVICE(0x0926, 0x0202), .driver_info =
 			USB_QUIRK_ENDPOINT_IGNORE },
@@ -539,6 +545,9 @@ static const struct usb_device_id usb_quirk_list[] = {
 	{ USB_DEVICE(0x2040, 0x7200), .driver_info =
 			USB_QUIRK_CONFIG_INTF_STRINGS },
 
+	/* VLI disk */
+	{ USB_DEVICE(0x2109, 0x0711), .driver_info = USB_QUIRK_NO_LPM },
+
 	/* Raydium Touchscreen */
 	{ USB_DEVICE(0x2386, 0x3114), .driver_info = USB_QUIRK_NO_LPM },
 
diff --git a/drivers/usb/dwc3/dwc3-xilinx.c b/drivers/usb/dwc3/dwc3-xilinx.c
index a33a42ba0249..4ca7f6240d07 100644
--- a/drivers/usb/dwc3/dwc3-xilinx.c
+++ b/drivers/usb/dwc3/dwc3-xilinx.c
@@ -207,15 +207,13 @@ static int dwc3_xlnx_init_zynqmp(struct dwc3_xlnx *priv_data)
 
 skip_usb3_phy:
 	/* ulpi reset via gpio-modepin or gpio-framework driver */
-	reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW);
+	reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(reset_gpio)) {
 		return dev_err_probe(dev, PTR_ERR(reset_gpio),
 				     "Failed to request reset GPIO\n");
 	}
 
 	if (reset_gpio) {
-		/* Toggle ulpi to reset the phy. */
-		gpiod_set_value_cansleep(reset_gpio, 1);
 		usleep_range(5000, 10000);
 		gpiod_set_value_cansleep(reset_gpio, 0);
 		usleep_range(5000, 10000);
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index 47e73c4ed62d..8c30d86cc4e3 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -4617,6 +4617,12 @@ static irqreturn_t dwc3_check_event_buf(struct dwc3_event_buffer *evt)
 	if (!count)
 		return IRQ_NONE;
 
+	if (count > evt->length) {
+		dev_err_ratelimited(dwc->dev, "invalid count(%u) > evt->length(%u)\n",
+			count, evt->length);
+		return IRQ_NONE;
+	}
+
 	evt->count = count;
 	evt->flags |= DWC3_EVENT_PENDING;
 
diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c
index 900ea0d368e0..9f0a6b27e47c 100644
--- a/drivers/usb/host/ohci-pci.c
+++ b/drivers/usb/host/ohci-pci.c
@@ -165,6 +165,25 @@ static int ohci_quirk_amd700(struct usb_hcd *hcd)
 	return 0;
 }
 
+static int ohci_quirk_loongson(struct usb_hcd *hcd)
+{
+	struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
+
+	/*
+	 * Loongson's LS7A OHCI controller (rev 0x02) has a
+	 * flaw. MMIO register with offset 0x60/64 is treated
+	 * as legacy PS2-compatible keyboard/mouse interface.
+	 * Since OHCI only use 4KB BAR resource, LS7A OHCI's
+	 * 32KB BAR is wrapped around (the 2nd 4KB BAR space
+	 * is the same as the 1st 4KB internally). So add 4KB
+	 * offset (0x1000) to the OHCI registers as a quirk.
+	 */
+	if (pdev->revision == 0x2)
+		hcd->regs += SZ_4K;	/* SZ_4K = 0x1000 */
+
+	return 0;
+}
+
 static int ohci_quirk_qemu(struct usb_hcd *hcd)
 {
 	struct ohci_hcd *ohci = hcd_to_ohci(hcd);
@@ -225,6 +244,10 @@ static const struct pci_device_id ohci_pci_quirks[] = {
 		.driver_data = (unsigned long)ohci_quirk_amd700,
 	},
 	{
+		PCI_DEVICE(PCI_VENDOR_ID_LOONGSON, 0x7a24),
+		.driver_data = (unsigned long)ohci_quirk_loongson,
+	},
+	{
 		.vendor		= PCI_VENDOR_ID_APPLE,
 		.device		= 0x003f,
 		.subvendor	= PCI_SUBVENDOR_ID_REDHAT_QUMRANET,
diff --git a/drivers/usb/host/xhci-hub.c b/drivers/usb/host/xhci-hub.c
index c0f226584a40..486347776cb2 100644
--- a/drivers/usb/host/xhci-hub.c
+++ b/drivers/usb/host/xhci-hub.c
@@ -1878,9 +1878,10 @@ int xhci_bus_resume(struct usb_hcd *hcd)
 	int max_ports, port_index;
 	int sret;
 	u32 next_state;
-	u32 temp, portsc;
+	u32 portsc;
 	struct xhci_hub *rhub;
 	struct xhci_port **ports;
+	bool disabled_irq = false;
 
 	rhub = xhci_get_rhub(hcd);
 	ports = rhub->ports;
@@ -1896,17 +1897,20 @@ int xhci_bus_resume(struct usb_hcd *hcd)
 		return -ESHUTDOWN;
 	}
 
-	/* delay the irqs */
-	temp = readl(&xhci->op_regs->command);
-	temp &= ~CMD_EIE;
-	writel(temp, &xhci->op_regs->command);
-
 	/* bus specific resume for ports we suspended at bus_suspend */
-	if (hcd->speed >= HCD_USB3)
+	if (hcd->speed >= HCD_USB3) {
 		next_state = XDEV_U0;
-	else
+	} else {
 		next_state = XDEV_RESUME;
-
+		if (bus_state->bus_suspended) {
+			/*
+			 * prevent port event interrupts from interfering
+			 * with usb2 port resume process
+			 */
+			xhci_disable_interrupter(xhci->interrupters[0]);
+			disabled_irq = true;
+		}
+	}
 	port_index = max_ports;
 	while (port_index--) {
 		portsc = readl(ports[port_index]->addr);
@@ -1974,11 +1978,9 @@ int xhci_bus_resume(struct usb_hcd *hcd)
 	(void) readl(&xhci->op_regs->command);
 
 	bus_state->next_statechange = jiffies + msecs_to_jiffies(5);
-	/* re-enable irqs */
-	temp = readl(&xhci->op_regs->command);
-	temp |= CMD_EIE;
-	writel(temp, &xhci->op_regs->command);
-	temp = readl(&xhci->op_regs->command);
+	/* re-enable interrupter */
+	if (disabled_irq)
+		xhci_enable_interrupter(xhci->interrupters[0]);
 
 	spin_unlock_irqrestore(&xhci->lock, flags);
 	return 0;
diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c
index 5d64c297721c..b906bc2eea5f 100644
--- a/drivers/usb/host/xhci-ring.c
+++ b/drivers/usb/host/xhci-ring.c
@@ -561,8 +561,8 @@ void xhci_ring_ep_doorbell(struct xhci_hcd *xhci,
 	 * pointer command pending because the device can choose to start any
 	 * stream once the endpoint is on the HW schedule.
 	 */
-	if (ep_state & (EP_STOP_CMD_PENDING | SET_DEQ_PENDING | EP_HALTED |
-			EP_CLEARING_TT | EP_STALLED))
+	if ((ep_state & EP_STOP_CMD_PENDING) || (ep_state & SET_DEQ_PENDING) ||
+	    (ep_state & EP_HALTED) || (ep_state & EP_CLEARING_TT))
 		return;
 
 	trace_xhci_ring_ep_doorbell(slot_id, DB_VALUE(ep_index, stream_id));
@@ -2573,9 +2573,6 @@ static void process_bulk_intr_td(struct xhci_hcd *xhci, struct xhci_virt_ep *ep,
 
 		xhci_handle_halted_endpoint(xhci, ep, td, EP_SOFT_RESET);
 		return;
-	case COMP_STALL_ERROR:
-		ep->ep_state |= EP_STALLED;
-		break;
 	default:
 		/* do nothing */
 		break;
@@ -2916,7 +2913,7 @@ static int handle_tx_event(struct xhci_hcd *xhci,
 			if (xhci_spurious_success_tx_event(xhci, ep_ring)) {
 				xhci_dbg(xhci, "Spurious event dma %pad, comp_code %u after %u\n",
 					 &ep_trb_dma, trb_comp_code, ep_ring->old_trb_comp_code);
-				ep_ring->old_trb_comp_code = trb_comp_code;
+				ep_ring->old_trb_comp_code = 0;
 				return 0;
 			}
 
@@ -3780,7 +3777,7 @@ int xhci_queue_ctrl_tx(struct xhci_hcd *xhci, gfp_t mem_flags,
 		 * enqueue a No Op TRB, this can prevent the Setup and Data Stage
 		 * TRB to be breaked by the Link TRB.
 		 */
-		if (trb_is_link(ep_ring->enqueue + 1)) {
+		if (last_trb_on_seg(ep_ring->enq_seg, ep_ring->enqueue + 1)) {
 			field = TRB_TYPE(TRB_TR_NOOP) | ep_ring->cycle_state;
 			queue_trb(xhci, ep_ring, false, 0, 0,
 					TRB_INTR_TARGET(0), field);
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 0452b8d65832..90eb491267b5 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -322,7 +322,7 @@ static void xhci_zero_64b_regs(struct xhci_hcd *xhci)
 		xhci_info(xhci, "Fault detected\n");
 }
 
-static int xhci_enable_interrupter(struct xhci_interrupter *ir)
+int xhci_enable_interrupter(struct xhci_interrupter *ir)
 {
 	u32 iman;
 
@@ -335,7 +335,7 @@ static int xhci_enable_interrupter(struct xhci_interrupter *ir)
 	return 0;
 }
 
-static int xhci_disable_interrupter(struct xhci_interrupter *ir)
+int xhci_disable_interrupter(struct xhci_interrupter *ir)
 {
 	u32 iman;
 
@@ -1605,11 +1605,6 @@ static int xhci_urb_enqueue(struct usb_hcd *hcd, struct urb *urb, gfp_t mem_flag
 		goto free_priv;
 	}
 
-	/* Class driver might not be aware ep halted due to async URB giveback */
-	if (*ep_state & EP_STALLED)
-		dev_dbg(&urb->dev->dev, "URB %p queued before clearing halt\n",
-			urb);
-
 	switch (usb_endpoint_type(&urb->ep->desc)) {
 
 	case USB_ENDPOINT_XFER_CONTROL:
@@ -1770,8 +1765,8 @@ static int xhci_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, int status)
 		goto done;
 	}
 
-	/* In these cases no commands are pending but the endpoint is stopped */
-	if (ep->ep_state & (EP_CLEARING_TT | EP_STALLED)) {
+	/* In this case no commands are pending but the endpoint is stopped */
+	if (ep->ep_state & EP_CLEARING_TT) {
 		/* and cancelled TDs can be given back right away */
 		xhci_dbg(xhci, "Invalidating TDs instantly on slot %d ep %d in state 0x%x\n",
 				urb->dev->slot_id, ep_index, ep->ep_state);
@@ -3209,11 +3204,8 @@ static void xhci_endpoint_reset(struct usb_hcd *hcd,
 
 	ep = &vdev->eps[ep_index];
 
-	spin_lock_irqsave(&xhci->lock, flags);
-
-	ep->ep_state &= ~EP_STALLED;
-
 	/* Bail out if toggle is already being cleared by a endpoint reset */
+	spin_lock_irqsave(&xhci->lock, flags);
 	if (ep->ep_state & EP_HARD_CLEAR_TOGGLE) {
 		ep->ep_state &= ~EP_HARD_CLEAR_TOGGLE;
 		spin_unlock_irqrestore(&xhci->lock, flags);
diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h
index 37860f1e3aba..242ab9fbc8ae 100644
--- a/drivers/usb/host/xhci.h
+++ b/drivers/usb/host/xhci.h
@@ -664,7 +664,7 @@ struct xhci_virt_ep {
 	unsigned int			err_count;
 	unsigned int			ep_state;
 #define SET_DEQ_PENDING		(1 << 0)
-#define EP_HALTED		(1 << 1)	/* Halted host ep handling */
+#define EP_HALTED		(1 << 1)	/* For stall handling */
 #define EP_STOP_CMD_PENDING	(1 << 2)	/* For URB cancellation */
 /* Transitioning the endpoint to using streams, don't enqueue URBs */
 #define EP_GETTING_STREAMS	(1 << 3)
@@ -675,7 +675,6 @@ struct xhci_virt_ep {
 #define EP_SOFT_CLEAR_TOGGLE	(1 << 7)
 /* usb_hub_clear_tt_buffer is in progress */
 #define EP_CLEARING_TT		(1 << 8)
-#define EP_STALLED		(1 << 9)	/* For stall handling */
 	/* ----  Related to URB cancellation ---- */
 	struct list_head	cancelled_td_list;
 	struct xhci_hcd		*xhci;
@@ -1891,6 +1890,8 @@ int xhci_alloc_tt_info(struct xhci_hcd *xhci,
 		struct usb_tt *tt, gfp_t mem_flags);
 int xhci_set_interrupter_moderation(struct xhci_interrupter *ir,
 				    u32 imod_interval);
+int xhci_enable_interrupter(struct xhci_interrupter *ir);
+int xhci_disable_interrupter(struct xhci_interrupter *ir);
 
 /* xHCI ring, segment, TRB, and TD functions */
 dma_addr_t xhci_trb_virt_to_dma(struct xhci_segment *seg, union xhci_trb *trb);
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index 9b34e23b7091..6ac7a0a5cf07 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -1093,6 +1093,8 @@ static const struct usb_device_id id_table_combined[] = {
 	{ USB_DEVICE_INTERFACE_NUMBER(ALTERA_VID, ALTERA_UB3_602E_PID, 1) },
 	{ USB_DEVICE_INTERFACE_NUMBER(ALTERA_VID, ALTERA_UB3_602E_PID, 2) },
 	{ USB_DEVICE_INTERFACE_NUMBER(ALTERA_VID, ALTERA_UB3_602E_PID, 3) },
+	/* Abacus Electrics */
+	{ USB_DEVICE(FTDI_VID, ABACUS_OPTICAL_PROBE_PID) },
 	{ }					/* Terminating entry */
 };
 
diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h
index 52be47d684ea..9acb6f837327 100644
--- a/drivers/usb/serial/ftdi_sio_ids.h
+++ b/drivers/usb/serial/ftdi_sio_ids.h
@@ -443,6 +443,11 @@
 #define LINX_FUTURE_2_PID   0xF44C	/* Linx future device */
 
 /*
+ * Abacus Electrics
+ */
+#define ABACUS_OPTICAL_PROBE_PID	0xf458 /* ABACUS ELECTRICS Optical Probe */
+
+/*
  * Oceanic product ids
  */
 #define FTDI_OCEANIC_PID	0xF460  /* Oceanic dive instrument */
diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
index 5cd26dac2069..27879cc57536 100644
--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -611,6 +611,7 @@ static void option_instat_callback(struct urb *urb);
 /* Sierra Wireless products */
 #define SIERRA_VENDOR_ID			0x1199
 #define SIERRA_PRODUCT_EM9191			0x90d3
+#define SIERRA_PRODUCT_EM9291			0x90e3
 
 /* UNISOC (Spreadtrum) products */
 #define UNISOC_VENDOR_ID			0x1782
@@ -2432,6 +2433,8 @@ static const struct usb_device_id option_ids[] = {
 	{ USB_DEVICE_AND_INTERFACE_INFO(SIERRA_VENDOR_ID, SIERRA_PRODUCT_EM9191, 0xff, 0xff, 0x30) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(SIERRA_VENDOR_ID, SIERRA_PRODUCT_EM9191, 0xff, 0xff, 0x40) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(SIERRA_VENDOR_ID, SIERRA_PRODUCT_EM9191, 0xff, 0, 0) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(SIERRA_VENDOR_ID, SIERRA_PRODUCT_EM9291, 0xff, 0xff, 0x30) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(SIERRA_VENDOR_ID, SIERRA_PRODUCT_EM9291, 0xff, 0xff, 0x40) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(UNISOC_VENDOR_ID, TOZED_PRODUCT_LT70C, 0xff, 0, 0) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(UNISOC_VENDOR_ID, LUAT_PRODUCT_AIR720U, 0xff, 0, 0) },
 	{ USB_DEVICE_INTERFACE_CLASS(0x1bbb, 0x0530, 0xff),			/* TCL IK512 MBIM */
diff --git a/drivers/usb/serial/usb-serial-simple.c b/drivers/usb/serial/usb-serial-simple.c
index 2c12449ff60c..a0afaf254d12 100644
--- a/drivers/usb/serial/usb-serial-simple.c
+++ b/drivers/usb/serial/usb-serial-simple.c
@@ -100,6 +100,11 @@ DEVICE(nokia, NOKIA_IDS);
 	{ USB_DEVICE(0x09d7, 0x0100) }	/* NovAtel FlexPack GPS */
 DEVICE_N(novatel_gps, NOVATEL_IDS, 3);
 
+/* OWON electronic test and measurement equipment driver */
+#define OWON_IDS()			\
+	{ USB_DEVICE(0x5345, 0x1234) } /* HDS200 oscilloscopes and others */
+DEVICE(owon, OWON_IDS);
+
 /* Siemens USB/MPI adapter */
 #define SIEMENS_IDS()			\
 	{ USB_DEVICE(0x908, 0x0004) }
@@ -134,6 +139,7 @@ static struct usb_serial_driver * const serial_drivers[] = {
 	&motorola_tetra_device,
 	&nokia_device,
 	&novatel_gps_device,
+	&owon_device,
 	&siemens_mpi_device,
 	&suunto_device,
 	&vivopay_device,
@@ -153,6 +159,7 @@ static const struct usb_device_id id_table[] = {
 	MOTOROLA_TETRA_IDS(),
 	NOKIA_IDS(),
 	NOVATEL_IDS(),
+	OWON_IDS(),
 	SIEMENS_IDS(),
 	SUUNTO_IDS(),
 	VIVOPAY_IDS(),
diff --git a/drivers/usb/storage/unusual_uas.h b/drivers/usb/storage/unusual_uas.h
index 1f8c9b16a0fb..d460d71b4257 100644
--- a/drivers/usb/storage/unusual_uas.h
+++ b/drivers/usb/storage/unusual_uas.h
@@ -83,6 +83,13 @@ UNUSUAL_DEV(0x0bc2, 0x331a, 0x0000, 0x9999,
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NO_REPORT_LUNS),
 
+/* Reported-by: Oliver Neukum <oneukum@suse.com> */
+UNUSUAL_DEV(0x125f, 0xa94a, 0x0160, 0x0160,
+		"ADATA",
+		"Portable HDD CH94",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_NO_ATA_1X),
+
 /* Reported-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> */
 UNUSUAL_DEV(0x13fd, 0x3940, 0x0000, 0x9999,
 		"Initio Corporation",
diff --git a/drivers/usb/typec/class.c b/drivers/usb/typec/class.c
index 9c76c3d0c6cf..67a533e35150 100644
--- a/drivers/usb/typec/class.c
+++ b/drivers/usb/typec/class.c
@@ -1052,9 +1052,11 @@ struct typec_partner *typec_register_partner(struct typec_port *port,
 		partner->usb_mode = USB_MODE_USB3;
 	}
 
+	mutex_lock(&port->partner_link_lock);
 	ret = device_register(&partner->dev);
 	if (ret) {
 		dev_err(&port->dev, "failed to register partner (%d)\n", ret);
+		mutex_unlock(&port->partner_link_lock);
 		put_device(&partner->dev);
 		return ERR_PTR(ret);
 	}
@@ -1063,6 +1065,7 @@ struct typec_partner *typec_register_partner(struct typec_port *port,
 		typec_partner_link_device(partner, port->usb2_dev);
 	if (port->usb3_dev)
 		typec_partner_link_device(partner, port->usb3_dev);
+	mutex_unlock(&port->partner_link_lock);
 
 	return partner;
 }
@@ -1083,12 +1086,18 @@ void typec_unregister_partner(struct typec_partner *partner)
 
 	port = to_typec_port(partner->dev.parent);
 
-	if (port->usb2_dev)
+	mutex_lock(&port->partner_link_lock);
+	if (port->usb2_dev) {
 		typec_partner_unlink_device(partner, port->usb2_dev);
-	if (port->usb3_dev)
+		port->usb2_dev = NULL;
+	}
+	if (port->usb3_dev) {
 		typec_partner_unlink_device(partner, port->usb3_dev);
+		port->usb3_dev = NULL;
+	}
 
 	device_unregister(&partner->dev);
+	mutex_unlock(&port->partner_link_lock);
 }
 EXPORT_SYMBOL_GPL(typec_unregister_partner);
 
@@ -2041,10 +2050,11 @@ static struct typec_partner *typec_get_partner(struct typec_port *port)
 static void typec_partner_attach(struct typec_connector *con, struct device *dev)
 {
 	struct typec_port *port = container_of(con, struct typec_port, con);
-	struct typec_partner *partner = typec_get_partner(port);
+	struct typec_partner *partner;
 	struct usb_device *udev = to_usb_device(dev);
 	enum usb_mode usb_mode;
 
+	mutex_lock(&port->partner_link_lock);
 	if (udev->speed < USB_SPEED_SUPER) {
 		usb_mode = USB_MODE_USB2;
 		port->usb2_dev = dev;
@@ -2053,18 +2063,22 @@ static void typec_partner_attach(struct typec_connector *con, struct device *dev
 		port->usb3_dev = dev;
 	}
 
+	partner = typec_get_partner(port);
 	if (partner) {
 		typec_partner_set_usb_mode(partner, usb_mode);
 		typec_partner_link_device(partner, dev);
 		put_device(&partner->dev);
 	}
+	mutex_unlock(&port->partner_link_lock);
 }
 
 static void typec_partner_deattach(struct typec_connector *con, struct device *dev)
 {
 	struct typec_port *port = container_of(con, struct typec_port, con);
-	struct typec_partner *partner = typec_get_partner(port);
+	struct typec_partner *partner;
 
+	mutex_lock(&port->partner_link_lock);
+	partner = typec_get_partner(port);
 	if (partner) {
 		typec_partner_unlink_device(partner, dev);
 		put_device(&partner->dev);
@@ -2074,6 +2088,7 @@ static void typec_partner_deattach(struct typec_connector *con, struct device *d
 		port->usb2_dev = NULL;
 	else if (port->usb3_dev == dev)
 		port->usb3_dev = NULL;
+	mutex_unlock(&port->partner_link_lock);
 }
 
 /**
@@ -2614,6 +2629,7 @@ struct typec_port *typec_register_port(struct device *parent,
 
 	ida_init(&port->mode_ids);
 	mutex_init(&port->port_type_lock);
+	mutex_init(&port->partner_link_lock);
 
 	port->id = id;
 	port->ops = cap->ops;
diff --git a/drivers/usb/typec/class.h b/drivers/usb/typec/class.h
index b3076a24ad2e..db2fe96c48ff 100644
--- a/drivers/usb/typec/class.h
+++ b/drivers/usb/typec/class.h
@@ -59,6 +59,7 @@ struct typec_port {
 	enum typec_port_type		port_type;
 	enum usb_mode			usb_mode;
 	struct mutex			port_type_lock;
+	struct mutex			partner_link_lock;
 
 	enum typec_orientation		orientation;
 	struct typec_switch		*sw;
diff --git a/drivers/vfio/pci/vfio_pci_config.c b/drivers/vfio/pci/vfio_pci_config.c
index 14437396d721..8f02f236b5b4 100644
--- a/drivers/vfio/pci/vfio_pci_config.c
+++ b/drivers/vfio/pci/vfio_pci_config.c
@@ -1815,7 +1815,7 @@ int vfio_config_init(struct vfio_pci_core_device *vdev)
 	}
 
 	if (!IS_ENABLED(CONFIG_VFIO_PCI_INTX) || vdev->nointx ||
-	    vdev->pdev->irq == IRQ_NOTCONNECTED)
+	    !vdev->pdev->irq || vdev->pdev->irq == IRQ_NOTCONNECTED)
 		vconfig[PCI_INTERRUPT_PIN] = 0;
 
 	ret = vfio_cap_init(vdev);
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index f6f5a7ac7894..26bcf3a7f70c 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -627,6 +627,9 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 	int ret;
 
 	llnode = llist_del_all(&svq->completion_list);
+
+	mutex_lock(&svq->vq.mutex);
+
 	llist_for_each_entry_safe(cmd, t, llnode, tvc_completion_list) {
 		se_cmd = &cmd->tvc_se_cmd;
 
@@ -660,6 +663,8 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 		vhost_scsi_release_cmd_res(se_cmd);
 	}
 
+	mutex_unlock(&svq->vq.mutex);
+
 	if (signal)
 		vhost_signal(&svq->vs->dev, &svq->vq);
 }
@@ -994,39 +999,66 @@ static void vhost_scsi_target_queue_cmd(struct vhost_scsi_nexus *nexus,
 
 static void
 vhost_scsi_send_status(struct vhost_scsi *vs, struct vhost_virtqueue *vq,
-		       int head, unsigned int out, u8 status)
+		       struct vhost_scsi_ctx *vc, u8 status)
 {
-	struct virtio_scsi_cmd_resp __user *resp;
 	struct virtio_scsi_cmd_resp rsp;
+	struct iov_iter iov_iter;
 	int ret;
 
 	memset(&rsp, 0, sizeof(rsp));
 	rsp.status = status;
-	resp = vq->iov[out].iov_base;
-	ret = __copy_to_user(resp, &rsp, sizeof(rsp));
-	if (!ret)
-		vhost_add_used_and_signal(&vs->dev, vq, head, 0);
+
+	iov_iter_init(&iov_iter, ITER_DEST, &vq->iov[vc->out], vc->in,
+		      sizeof(rsp));
+
+	ret = copy_to_iter(&rsp, sizeof(rsp), &iov_iter);
+
+	if (likely(ret == sizeof(rsp)))
+		vhost_add_used_and_signal(&vs->dev, vq, vc->head, 0);
 	else
 		pr_err("Faulted on virtio_scsi_cmd_resp\n");
 }
 
+#define TYPE_IO_CMD    0
+#define TYPE_CTRL_TMF  1
+#define TYPE_CTRL_AN   2
+
 static void
 vhost_scsi_send_bad_target(struct vhost_scsi *vs,
 			   struct vhost_virtqueue *vq,
-			   int head, unsigned out)
+			   struct vhost_scsi_ctx *vc, int type)
 {
-	struct virtio_scsi_cmd_resp __user *resp;
-	struct virtio_scsi_cmd_resp rsp;
+	union {
+		struct virtio_scsi_cmd_resp cmd;
+		struct virtio_scsi_ctrl_tmf_resp tmf;
+		struct virtio_scsi_ctrl_an_resp an;
+	} rsp;
+	struct iov_iter iov_iter;
+	size_t rsp_size;
 	int ret;
 
 	memset(&rsp, 0, sizeof(rsp));
-	rsp.response = VIRTIO_SCSI_S_BAD_TARGET;
-	resp = vq->iov[out].iov_base;
-	ret = __copy_to_user(resp, &rsp, sizeof(rsp));
-	if (!ret)
-		vhost_add_used_and_signal(&vs->dev, vq, head, 0);
+
+	if (type == TYPE_IO_CMD) {
+		rsp_size = sizeof(struct virtio_scsi_cmd_resp);
+		rsp.cmd.response = VIRTIO_SCSI_S_BAD_TARGET;
+	} else if (type == TYPE_CTRL_TMF) {
+		rsp_size = sizeof(struct virtio_scsi_ctrl_tmf_resp);
+		rsp.tmf.response = VIRTIO_SCSI_S_BAD_TARGET;
+	} else {
+		rsp_size = sizeof(struct virtio_scsi_ctrl_an_resp);
+		rsp.an.response = VIRTIO_SCSI_S_BAD_TARGET;
+	}
+
+	iov_iter_init(&iov_iter, ITER_DEST, &vq->iov[vc->out], vc->in,
+		      rsp_size);
+
+	ret = copy_to_iter(&rsp, rsp_size, &iov_iter);
+
+	if (likely(ret == rsp_size))
+		vhost_add_used_and_signal(&vs->dev, vq, vc->head, 0);
 	else
-		pr_err("Faulted on virtio_scsi_cmd_resp\n");
+		pr_err("Faulted on virtio scsi type=%d\n", type);
 }
 
 static int
@@ -1390,9 +1422,9 @@ err:
 		if (ret == -ENXIO)
 			break;
 		else if (ret == -EIO)
-			vhost_scsi_send_bad_target(vs, vq, vc.head, vc.out);
+			vhost_scsi_send_bad_target(vs, vq, &vc, TYPE_IO_CMD);
 		else if (ret == -ENOMEM)
-			vhost_scsi_send_status(vs, vq, vc.head, vc.out,
+			vhost_scsi_send_status(vs, vq, &vc,
 					       SAM_STAT_TASK_SET_FULL);
 	} while (likely(!vhost_exceeds_weight(vq, ++c, 0)));
 out:
@@ -1432,8 +1464,11 @@ static void vhost_scsi_tmf_resp_work(struct vhost_work *work)
 	else
 		resp_code = VIRTIO_SCSI_S_FUNCTION_REJECTED;
 
+	mutex_lock(&tmf->svq->vq.mutex);
 	vhost_scsi_send_tmf_resp(tmf->vhost, &tmf->svq->vq, tmf->in_iovs,
 				 tmf->vq_desc, &tmf->resp_iov, resp_code);
+	mutex_unlock(&tmf->svq->vq.mutex);
+
 	vhost_scsi_release_tmf_res(tmf);
 }
 
@@ -1623,7 +1658,10 @@ err:
 		if (ret == -ENXIO)
 			break;
 		else if (ret == -EIO)
-			vhost_scsi_send_bad_target(vs, vq, vc.head, vc.out);
+			vhost_scsi_send_bad_target(vs, vq, &vc,
+						   v_req.type == VIRTIO_SCSI_T_TMF ?
+						   TYPE_CTRL_TMF :
+						   TYPE_CTRL_AN);
 	} while (likely(!vhost_exceeds_weight(vq, ++c, 0)));
 out:
 	mutex_unlock(&vq->mutex);
diff --git a/drivers/video/screen_info_generic.c b/drivers/video/screen_info_generic.c
index 64117c6367ab..900e9386eceb 100644
--- a/drivers/video/screen_info_generic.c
+++ b/drivers/video/screen_info_generic.c
@@ -144,3 +144,39 @@ ssize_t screen_info_resources(const struct screen_info *si, struct resource *r,
 	return pos - r;
 }
 EXPORT_SYMBOL(screen_info_resources);
+
+/*
+ * The meaning of depth and bpp for direct-color formats is
+ * inconsistent:
+ *
+ *  - DRM format info specifies depth as the number of color
+ *    bits; including alpha, but not including filler bits.
+ *  - Linux' EFI platform code computes lfb_depth from the
+ *    individual color channels, including the reserved bits.
+ *  - VBE 1.1 defines lfb_depth for XRGB1555 as 16, but later
+ *    versions use 15.
+ *  - On the kernel command line, 'bpp' of 32 is usually
+ *    XRGB8888 including the filler bits, but 15 is XRGB1555
+ *    not including the filler bit.
+ *
+ * It is not easily possible to fix this in struct screen_info,
+ * as this could break UAPI. The best solution is to compute
+ * bits_per_pixel from the color bits, reserved bits and
+ * reported lfb_depth, whichever is highest.
+ */
+
+u32 __screen_info_lfb_bits_per_pixel(const struct screen_info *si)
+{
+	u32 bits_per_pixel = si->lfb_depth;
+
+	if (bits_per_pixel > 8) {
+		bits_per_pixel = max(max3(si->red_size + si->red_pos,
+					  si->green_size + si->green_pos,
+					  si->blue_size + si->blue_pos),
+				     si->rsvd_size + si->rsvd_pos);
+		bits_per_pixel = max_t(u32, bits_per_pixel, si->lfb_depth);
+	}
+
+	return bits_per_pixel;
+}
+EXPORT_SYMBOL(__screen_info_lfb_bits_per_pixel);
diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index 150753c3b578..95d5d7993e5b 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -407,6 +407,12 @@ static void virtio_dev_shutdown(struct device *_d)
 	if (!drv)
 		return;
 
+	/* If the driver has its own shutdown method, use that. */
+	if (drv->shutdown) {
+		drv->shutdown(dev);
+		return;
+	}
+
 	/*
 	 * Some devices get wedged if you kick them after they are
 	 * reset. Mark all vqs as broken to make sure we don't.
diff --git a/drivers/virtio/virtio_pci_modern.c b/drivers/virtio/virtio_pci_modern.c
index 5eaade757860..d50fe030d825 100644
--- a/drivers/virtio/virtio_pci_modern.c
+++ b/drivers/virtio/virtio_pci_modern.c
@@ -247,7 +247,7 @@ virtio_pci_admin_cmd_dev_parts_objects_enable(struct virtio_device *virtio_dev)
 	sg_init_one(&data_sg, get_data, sizeof(*get_data));
 	sg_init_one(&result_sg, result, sizeof(*result));
 	cmd.opcode = cpu_to_le16(VIRTIO_ADMIN_CMD_DEVICE_CAP_GET);
-	cmd.group_type = cpu_to_le16(VIRTIO_ADMIN_GROUP_TYPE_SRIOV);
+	cmd.group_type = cpu_to_le16(VIRTIO_ADMIN_GROUP_TYPE_SELF);
 	cmd.data_sg = &data_sg;
 	cmd.result_sg = &result_sg;
 	ret = vp_modern_admin_cmd_exec(virtio_dev, &cmd);
@@ -305,7 +305,7 @@ static void virtio_pci_admin_cmd_cap_init(struct virtio_device *virtio_dev)
 
 	sg_init_one(&result_sg, data, sizeof(*data));
 	cmd.opcode = cpu_to_le16(VIRTIO_ADMIN_CMD_CAP_ID_LIST_QUERY);
-	cmd.group_type = cpu_to_le16(VIRTIO_ADMIN_GROUP_TYPE_SRIOV);
+	cmd.group_type = cpu_to_le16(VIRTIO_ADMIN_GROUP_TYPE_SELF);
 	cmd.result_sg = &result_sg;
 
 	ret = vp_modern_admin_cmd_exec(virtio_dev, &cmd);
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index fdd2d2b07b5a..b784aab66867 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -2650,7 +2650,7 @@ bool virtqueue_enable_cb_delayed(struct virtqueue *_vq)
 	struct vring_virtqueue *vq = to_vvq(_vq);
 
 	if (vq->event_triggered)
-		vq->event_triggered = false;
+		data_race(vq->event_triggered = false);
 
 	return vq->packed_ring ? virtqueue_enable_cb_delayed_packed(_vq) :
 				 virtqueue_enable_cb_delayed_split(_vq);
diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig
index f7d6f47971fd..24f485827e03 100644
--- a/drivers/xen/Kconfig
+++ b/drivers/xen/Kconfig
@@ -278,7 +278,7 @@ config XEN_PRIVCMD_EVENTFD
 
 config XEN_ACPI_PROCESSOR
 	tristate "Xen ACPI processor"
-	depends on XEN && XEN_PV_DOM0 && X86 && ACPI_PROCESSOR && CPU_FREQ
+	depends on XEN && XEN_DOM0 && X86 && ACPI_PROCESSOR && CPU_FREQ
 	default m
 	help
 	  This ACPI processor uploads Power Management information to the Xen
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 65d4e7fa1eb8..8c852807ba1c 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -679,7 +679,7 @@ void xen_free_ballooned_pages(unsigned int nr_pages, struct page **pages)
 }
 EXPORT_SYMBOL(xen_free_ballooned_pages);
 
-static void __init balloon_add_regions(void)
+static int __init balloon_add_regions(void)
 {
 	unsigned long start_pfn, pages;
 	unsigned long pfn, extra_pfn_end;
@@ -702,26 +702,38 @@ static void __init balloon_add_regions(void)
 		for (pfn = start_pfn; pfn < extra_pfn_end; pfn++)
 			balloon_append(pfn_to_page(pfn));
 
-		balloon_stats.total_pages += extra_pfn_end - start_pfn;
+		/*
+		 * Extra regions are accounted for in the physmap, but need
+		 * decreasing from current_pages to balloon down the initial
+		 * allocation, because they are already accounted for in
+		 * total_pages.
+		 */
+		if (extra_pfn_end - start_pfn >= balloon_stats.current_pages) {
+			WARN(1, "Extra pages underflow current target");
+			return -ERANGE;
+		}
+		balloon_stats.current_pages -= extra_pfn_end - start_pfn;
 	}
+
+	return 0;
 }
 
 static int __init balloon_init(void)
 {
 	struct task_struct *task;
+	int rc;
 
 	if (!xen_domain())
 		return -ENODEV;
 
 	pr_info("Initialising balloon driver\n");
 
-#ifdef CONFIG_XEN_PV
-	balloon_stats.current_pages = xen_pv_domain()
-		? min(xen_start_info->nr_pages - xen_released_pages, max_pfn)
-		: get_num_physpages();
-#else
-	balloon_stats.current_pages = get_num_physpages();
-#endif
+	if (xen_released_pages >= get_num_physpages()) {
+		WARN(1, "Released pages underflow current target");
+		return -ERANGE;
+	}
+
+	balloon_stats.current_pages = get_num_physpages() - xen_released_pages;
 	balloon_stats.target_pages  = balloon_stats.current_pages;
 	balloon_stats.balloon_low   = 0;
 	balloon_stats.balloon_high  = 0;
@@ -738,7 +750,9 @@ static int __init balloon_init(void)
 	register_sysctl_init("xen/balloon", balloon_table);
 #endif
 
-	balloon_add_regions();
+	rc = balloon_add_regions();
+	if (rc)
+		return rc;
 
 	task = kthread_run(balloon_thread, NULL, "xen-balloon");
 	if (IS_ERR(task)) {
diff --git a/drivers/xen/xenbus/xenbus_probe_frontend.c b/drivers/xen/xenbus/xenbus_probe_frontend.c
index fcb335bb7b18..6d1819269cbe 100644
--- a/drivers/xen/xenbus/xenbus_probe_frontend.c
+++ b/drivers/xen/xenbus/xenbus_probe_frontend.c
@@ -513,4 +513,5 @@ static int __init boot_wait_for_devices(void)
 late_initcall(boot_wait_for_devices);
 #endif
 
+MODULE_DESCRIPTION("Xen PV-device frontend support");
 MODULE_LICENSE("GPL");
diff --git a/fs/Kconfig b/fs/Kconfig
index c718b2e2de0e..5b4847bd2fbb 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -368,6 +368,7 @@ config GRACE_PERIOD
 config LOCKD
 	tristate
 	depends on FILE_LOCKING
+	select CRC32
 	select GRACE_PERIOD
 
 config LOCKD_V4
diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c
index 691e0ae607a1..8c6130789fde 100644
--- a/fs/afs/dynroot.c
+++ b/fs/afs/dynroot.c
@@ -348,9 +348,9 @@ static int afs_dynroot_readdir(struct file *file, struct dir_context *ctx)
 	}
 
 	if ((unsigned long long)ctx->pos <= AFS_MAX_DYNROOT_CELL_INO) {
-		rcu_read_lock();
+		down_read(&net->cells_lock);
 		ret = afs_dynroot_readdir_cells(net, ctx);
-		rcu_read_unlock();
+		up_read(&net->cells_lock);
 	}
 	return ret;
 }
diff --git a/fs/bcachefs/Kconfig b/fs/bcachefs/Kconfig
index bf1c94e51dd0..07709b0d7688 100644
--- a/fs/bcachefs/Kconfig
+++ b/fs/bcachefs/Kconfig
@@ -4,7 +4,7 @@ config BCACHEFS_FS
 	depends on BLOCK
 	select EXPORTFS
 	select CLOSURES
-	select LIBCRC32C
+	select CRC32
 	select CRC64
 	select FS_POSIX_ACL
 	select LZ4_COMPRESS
@@ -15,10 +15,9 @@ config BCACHEFS_FS
 	select ZLIB_INFLATE
 	select ZSTD_COMPRESS
 	select ZSTD_DECOMPRESS
-	select CRYPTO
 	select CRYPTO_LIB_SHA256
-	select CRYPTO_CHACHA20
-	select CRYPTO_POLY1305
+	select CRYPTO_LIB_CHACHA
+	select CRYPTO_LIB_POLY1305
 	select KEYS
 	select RAID6_PQ
 	select XOR_BLOCKS
diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c
index 7c930ef77380..effafc3e0ced 100644
--- a/fs/bcachefs/alloc_foreground.c
+++ b/fs/bcachefs/alloc_foreground.c
@@ -1425,6 +1425,8 @@ alloc_done:
 	open_bucket_for_each(c, &wp->ptrs, ob, i)
 		wp->sectors_free = min(wp->sectors_free, ob->sectors_free);
 
+	wp->sectors_free = rounddown(wp->sectors_free, block_sectors(c));
+
 	BUG_ON(!wp->sectors_free || wp->sectors_free == UINT_MAX);
 
 	return 0;
diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h
index 69ec6a012898..4c1e33cf57c0 100644
--- a/fs/bcachefs/alloc_foreground.h
+++ b/fs/bcachefs/alloc_foreground.h
@@ -110,7 +110,9 @@ static inline void bch2_alloc_sectors_done_inlined(struct bch_fs *c, struct writ
 	unsigned i;
 
 	open_bucket_for_each(c, &wp->ptrs, ob, i)
-		ob_push(c, !ob->sectors_free ? &ptrs : &keep, ob);
+		ob_push(c, ob->sectors_free < block_sectors(c)
+			? &ptrs
+			: &keep, ob);
 	wp->ptrs = keep;
 
 	mutex_unlock(&wp->lock);
diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h
index 5d9f208a1bb7..75f7408da173 100644
--- a/fs/bcachefs/bcachefs.h
+++ b/fs/bcachefs/bcachefs.h
@@ -788,6 +788,8 @@ struct bch_fs {
 		unsigned long	errors_silent[BITS_TO_LONGS(BCH_FSCK_ERR_MAX)];
 		u64		btrees_lost_data;
 	}			sb;
+	DARRAY(enum bcachefs_metadata_version)
+				incompat_versions_requested;
 
 #ifdef CONFIG_UNICODE
 	struct unicode_map	*cf_encoding;
@@ -981,8 +983,8 @@ struct bch_fs {
 	mempool_t		compress_workspace[BCH_COMPRESSION_OPT_NR];
 	size_t			zstd_workspace_size;
 
-	struct crypto_sync_skcipher *chacha20;
-	struct crypto_shash	*poly1305;
+	struct bch_key		chacha20_key;
+	bool			chacha20_key_set;
 
 	atomic64_t		key_version;
 
diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h
index a3db328dee31..d6e4a496f02b 100644
--- a/fs/bcachefs/bcachefs_format.h
+++ b/fs/bcachefs/bcachefs_format.h
@@ -366,6 +366,10 @@ static inline void bkey_init(struct bkey *k)
 #define __BKEY_PADDED(key, pad)					\
 	struct bkey_i key; __u64 key ## _pad[pad]
 
+enum bch_bkey_type_flags {
+	BKEY_TYPE_strict_btree_checks	= BIT(0),
+};
+
 /*
  * - DELETED keys are used internally to mark keys that should be ignored but
  *   override keys in composition order.  Their version number is ignored.
@@ -383,46 +387,46 @@ static inline void bkey_init(struct bkey *k)
  *
  * - WHITEOUT: for hash table btrees
  */
-#define BCH_BKEY_TYPES()				\
-	x(deleted,		0)			\
-	x(whiteout,		1)			\
-	x(error,		2)			\
-	x(cookie,		3)			\
-	x(hash_whiteout,	4)			\
-	x(btree_ptr,		5)			\
-	x(extent,		6)			\
-	x(reservation,		7)			\
-	x(inode,		8)			\
-	x(inode_generation,	9)			\
-	x(dirent,		10)			\
-	x(xattr,		11)			\
-	x(alloc,		12)			\
-	x(quota,		13)			\
-	x(stripe,		14)			\
-	x(reflink_p,		15)			\
-	x(reflink_v,		16)			\
-	x(inline_data,		17)			\
-	x(btree_ptr_v2,		18)			\
-	x(indirect_inline_data,	19)			\
-	x(alloc_v2,		20)			\
-	x(subvolume,		21)			\
-	x(snapshot,		22)			\
-	x(inode_v2,		23)			\
-	x(alloc_v3,		24)			\
-	x(set,			25)			\
-	x(lru,			26)			\
-	x(alloc_v4,		27)			\
-	x(backpointer,		28)			\
-	x(inode_v3,		29)			\
-	x(bucket_gens,		30)			\
-	x(snapshot_tree,	31)			\
-	x(logged_op_truncate,	32)			\
-	x(logged_op_finsert,	33)			\
-	x(accounting,		34)			\
-	x(inode_alloc_cursor,	35)
+#define BCH_BKEY_TYPES()						\
+	x(deleted,		0,	0)				\
+	x(whiteout,		1,	0)				\
+	x(error,		2,	0)				\
+	x(cookie,		3,	0)				\
+	x(hash_whiteout,	4,	BKEY_TYPE_strict_btree_checks)	\
+	x(btree_ptr,		5,	BKEY_TYPE_strict_btree_checks)	\
+	x(extent,		6,	BKEY_TYPE_strict_btree_checks)	\
+	x(reservation,		7,	BKEY_TYPE_strict_btree_checks)	\
+	x(inode,		8,	BKEY_TYPE_strict_btree_checks)	\
+	x(inode_generation,	9,	BKEY_TYPE_strict_btree_checks)	\
+	x(dirent,		10,	BKEY_TYPE_strict_btree_checks)	\
+	x(xattr,		11,	BKEY_TYPE_strict_btree_checks)	\
+	x(alloc,		12,	BKEY_TYPE_strict_btree_checks)	\
+	x(quota,		13,	BKEY_TYPE_strict_btree_checks)	\
+	x(stripe,		14,	BKEY_TYPE_strict_btree_checks)	\
+	x(reflink_p,		15,	BKEY_TYPE_strict_btree_checks)	\
+	x(reflink_v,		16,	BKEY_TYPE_strict_btree_checks)	\
+	x(inline_data,		17,	BKEY_TYPE_strict_btree_checks)	\
+	x(btree_ptr_v2,		18,	BKEY_TYPE_strict_btree_checks)	\
+	x(indirect_inline_data,	19,	BKEY_TYPE_strict_btree_checks)	\
+	x(alloc_v2,		20,	BKEY_TYPE_strict_btree_checks)	\
+	x(subvolume,		21,	BKEY_TYPE_strict_btree_checks)	\
+	x(snapshot,		22,	BKEY_TYPE_strict_btree_checks)	\
+	x(inode_v2,		23,	BKEY_TYPE_strict_btree_checks)	\
+	x(alloc_v3,		24,	BKEY_TYPE_strict_btree_checks)	\
+	x(set,			25,	0)				\
+	x(lru,			26,	BKEY_TYPE_strict_btree_checks)	\
+	x(alloc_v4,		27,	BKEY_TYPE_strict_btree_checks)	\
+	x(backpointer,		28,	BKEY_TYPE_strict_btree_checks)	\
+	x(inode_v3,		29,	BKEY_TYPE_strict_btree_checks)	\
+	x(bucket_gens,		30,	BKEY_TYPE_strict_btree_checks)	\
+	x(snapshot_tree,	31,	BKEY_TYPE_strict_btree_checks)	\
+	x(logged_op_truncate,	32,	BKEY_TYPE_strict_btree_checks)	\
+	x(logged_op_finsert,	33,	BKEY_TYPE_strict_btree_checks)	\
+	x(accounting,		34,	BKEY_TYPE_strict_btree_checks)	\
+	x(inode_alloc_cursor,	35,	BKEY_TYPE_strict_btree_checks)
 
 enum bch_bkey_type {
-#define x(name, nr) KEY_TYPE_##name	= nr,
+#define x(name, nr, ...) KEY_TYPE_##name	= nr,
 	BCH_BKEY_TYPES()
 #undef x
 	KEY_TYPE_MAX,
@@ -863,6 +867,7 @@ LE64_BITMASK(BCH_SB_VERSION_INCOMPAT_ALLOWED,
 LE64_BITMASK(BCH_SB_SHARD_INUMS_NBITS,	struct bch_sb, flags[6],  0,  4);
 LE64_BITMASK(BCH_SB_WRITE_ERROR_TIMEOUT,struct bch_sb, flags[6],  4, 14);
 LE64_BITMASK(BCH_SB_CSUM_ERR_RETRY_NR,	struct bch_sb, flags[6], 14, 20);
+LE64_BITMASK(BCH_SB_CASEFOLD,		struct bch_sb, flags[6], 22, 23);
 
 static inline __u64 BCH_SB_COMPRESSION_TYPE(const struct bch_sb *sb)
 {
diff --git a/fs/bcachefs/bkey_methods.c b/fs/bcachefs/bkey_methods.c
index 15c93576b5c2..00d05ccfaf73 100644
--- a/fs/bcachefs/bkey_methods.c
+++ b/fs/bcachefs/bkey_methods.c
@@ -21,7 +21,7 @@
 #include "xattr.h"
 
 const char * const bch2_bkey_types[] = {
-#define x(name, nr) #name,
+#define x(name, nr, ...) #name,
 	BCH_BKEY_TYPES()
 #undef x
 	NULL
@@ -115,7 +115,7 @@ static bool key_type_set_merge(struct bch_fs *c, struct bkey_s l, struct bkey_s_
 })
 
 const struct bkey_ops bch2_bkey_ops[] = {
-#define x(name, nr) [KEY_TYPE_##name]	= bch2_bkey_ops_##name,
+#define x(name, nr, ...) [KEY_TYPE_##name]	= bch2_bkey_ops_##name,
 	BCH_BKEY_TYPES()
 #undef x
 };
@@ -155,6 +155,12 @@ static u64 bch2_key_types_allowed[] = {
 #undef x
 };
 
+static const enum bch_bkey_type_flags bch2_bkey_type_flags[] = {
+#define x(name, nr, flags)	[KEY_TYPE_##name] = flags,
+	BCH_BKEY_TYPES()
+#undef x
+};
+
 const char *bch2_btree_node_type_str(enum btree_node_type type)
 {
 	return type == BKEY_TYPE_btree ? "internal btree node" : bch2_btree_id_str(type - 1);
@@ -177,8 +183,18 @@ int __bch2_bkey_validate(struct bch_fs *c, struct bkey_s_c k,
 	if (type >= BKEY_TYPE_NR)
 		return 0;
 
-	bkey_fsck_err_on(k.k->type < KEY_TYPE_MAX &&
-			 (type == BKEY_TYPE_btree || (from.flags & BCH_VALIDATE_commit)) &&
+	enum bch_bkey_type_flags bkey_flags = k.k->type < KEY_TYPE_MAX
+		? bch2_bkey_type_flags[k.k->type]
+		: 0;
+
+	bool strict_key_type_allowed =
+		(from.flags & BCH_VALIDATE_commit) ||
+		type == BKEY_TYPE_btree ||
+		(from.btree < BTREE_ID_NR &&
+		 (bkey_flags & BKEY_TYPE_strict_btree_checks));
+
+	bkey_fsck_err_on(strict_key_type_allowed &&
+			 k.k->type < KEY_TYPE_MAX &&
 			 !(bch2_key_types_allowed[type] & BIT_ULL(k.k->type)),
 			 c, bkey_invalid_type_for_btree,
 			 "invalid key type for btree %s (%s)",
diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c
index 7b98ba2dec64..37b69d89341f 100644
--- a/fs/bcachefs/btree_gc.c
+++ b/fs/bcachefs/btree_gc.c
@@ -47,6 +47,27 @@
 #define DROP_PREV_NODE		11
 #define DID_FILL_FROM_SCAN	12
 
+/*
+ * Returns true if it's a btree we can easily reconstruct, or otherwise won't
+ * cause data loss if it's missing:
+ */
+static bool btree_id_important(enum btree_id btree)
+{
+	if (btree_id_is_alloc(btree))
+		return false;
+
+	switch (btree) {
+	case BTREE_ID_quotas:
+	case BTREE_ID_snapshot_trees:
+	case BTREE_ID_logged_ops:
+	case BTREE_ID_rebalance_work:
+	case BTREE_ID_subvolume_children:
+		return false;
+	default:
+		return true;
+	}
+}
+
 static const char * const bch2_gc_phase_strs[] = {
 #define x(n)	#n,
 	GC_PHASES()
@@ -534,8 +555,10 @@ reconstruct_root:
 			r->error = 0;
 
 			if (!bch2_btree_has_scanned_nodes(c, i)) {
-				mustfix_fsck_err(trans, btree_root_unreadable_and_scan_found_nothing,
-						 "no nodes found for btree %s, continue?", buf.buf);
+				__fsck_err(trans,
+					   FSCK_CAN_FIX|(!btree_id_important(i) ? FSCK_AUTOFIX : 0),
+					   btree_root_unreadable_and_scan_found_nothing,
+					   "no nodes found for btree %s, continue?", buf.buf);
 				bch2_btree_root_alloc_fake_trans(trans, i, 0);
 			} else {
 				bch2_btree_root_alloc_fake_trans(trans, i, 1);
diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c
index e34e9598ef25..59fa527ac685 100644
--- a/fs/bcachefs/btree_iter.c
+++ b/fs/bcachefs/btree_iter.c
@@ -2577,7 +2577,10 @@ struct bkey_s_c bch2_btree_iter_peek_prev_min(struct btree_trans *trans, struct
 					      struct bpos end)
 {
 	if ((iter->flags & (BTREE_ITER_is_extents|BTREE_ITER_filter_snapshots)) &&
-	   !bkey_eq(iter->pos, POS_MAX)) {
+	   !bkey_eq(iter->pos, POS_MAX) &&
+	   !((iter->flags & BTREE_ITER_is_extents) &&
+	     iter->pos.offset == U64_MAX)) {
+
 		/*
 		 * bkey_start_pos(), for extents, is not monotonically
 		 * increasing until after filtering for snapshots:
@@ -2602,7 +2605,7 @@ struct bkey_s_c bch2_btree_iter_peek_prev_min(struct btree_trans *trans, struct
 
 	bch2_trans_verify_not_unlocked_or_in_restart(trans);
 	bch2_btree_iter_verify_entry_exit(iter);
-	EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bpos_eq(end, POS_MIN));
+	EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && iter->pos.inode != end.inode);
 
 	int ret = trans_maybe_inject_restart(trans, _RET_IP_);
 	if (unlikely(ret)) {
diff --git a/fs/bcachefs/btree_journal_iter.c b/fs/bcachefs/btree_journal_iter.c
index d1ad1a7613c9..ade3b5addd75 100644
--- a/fs/bcachefs/btree_journal_iter.c
+++ b/fs/bcachefs/btree_journal_iter.c
@@ -288,7 +288,7 @@ int bch2_journal_key_insert_take(struct bch_fs *c, enum btree_id id,
 			.size			= max_t(size_t, keys->size, 8) * 2,
 		};
 
-		new_keys.data = kvmalloc_array(new_keys.size, sizeof(new_keys.data[0]), GFP_KERNEL);
+		new_keys.data = bch2_kvmalloc(new_keys.size * sizeof(new_keys.data[0]), GFP_KERNEL);
 		if (!new_keys.data) {
 			bch_err(c, "%s: error allocating new key array (size %zu)",
 				__func__, new_keys.size);
@@ -644,8 +644,6 @@ void bch2_btree_and_journal_iter_init_node_iter(struct btree_trans *trans,
  */
 static int journal_sort_key_cmp(const void *_l, const void *_r)
 {
-	cond_resched();
-
 	const struct journal_key *l = _l;
 	const struct journal_key *r = _r;
 
@@ -689,7 +687,8 @@ void bch2_journal_keys_put(struct bch_fs *c)
 
 static void __journal_keys_sort(struct journal_keys *keys)
 {
-	sort(keys->data, keys->nr, sizeof(keys->data[0]), journal_sort_key_cmp, NULL);
+	sort_nonatomic(keys->data, keys->nr, sizeof(keys->data[0]),
+		       journal_sort_key_cmp, NULL);
 
 	cond_resched();
 
diff --git a/fs/bcachefs/btree_node_scan.c b/fs/bcachefs/btree_node_scan.c
index 8c9fdb7263fe..86acf037590c 100644
--- a/fs/bcachefs/btree_node_scan.c
+++ b/fs/bcachefs/btree_node_scan.c
@@ -183,7 +183,7 @@ static void try_read_btree_node(struct find_btree_nodes *f, struct bch_dev *ca,
 		return;
 
 	if (bch2_csum_type_is_encryption(BSET_CSUM_TYPE(&bn->keys))) {
-		if (!c->chacha20)
+		if (!c->chacha20_key_set)
 			return;
 
 		struct nonce nonce = btree_nonce(&bn->keys, 0);
@@ -398,7 +398,7 @@ int bch2_scan_for_btree_nodes(struct bch_fs *c)
 		bch2_print_string_as_lines(KERN_INFO, buf.buf);
 	}
 
-	sort(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_cookie, NULL);
+	sort_nonatomic(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_cookie, NULL);
 
 	dst = 0;
 	darray_for_each(f->nodes, i) {
@@ -418,7 +418,7 @@ int bch2_scan_for_btree_nodes(struct bch_fs *c)
 	}
 	f->nodes.nr = dst;
 
-	sort(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_pos, NULL);
+	sort_nonatomic(f->nodes.data, f->nodes.nr, sizeof(f->nodes.data[0]), found_btree_node_cmp_pos, NULL);
 
 	if (0 && c->opts.verbose) {
 		printbuf_reset(&buf);
diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c
index 55fbeeb8eaaa..00307356d7c8 100644
--- a/fs/bcachefs/btree_update_interior.c
+++ b/fs/bcachefs/btree_update_interior.c
@@ -1221,7 +1221,7 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path,
 
 	ret = bch2_disk_reservation_get(c, &as->disk_res,
 			(nr_nodes[0] + nr_nodes[1]) * btree_sectors(c),
-			c->opts.metadata_replicas,
+			READ_ONCE(c->opts.metadata_replicas),
 			disk_res_flags);
 	if (ret)
 		goto err;
@@ -1389,7 +1389,7 @@ static void bch2_insert_fixup_btree_ptr(struct btree_update *as,
 	printbuf_exit(&buf);
 }
 
-static void
+static int
 bch2_btree_insert_keys_interior(struct btree_update *as,
 				struct btree_trans *trans,
 				struct btree_path *path,
@@ -1411,7 +1411,8 @@ bch2_btree_insert_keys_interior(struct btree_update *as,
 	     insert = bkey_next(insert))
 		bch2_insert_fixup_btree_ptr(as, trans, path, b, &node_iter, insert);
 
-	if (bch2_btree_node_check_topology(trans, b)) {
+	int ret = bch2_btree_node_check_topology(trans, b);
+	if (ret) {
 		struct printbuf buf = PRINTBUF;
 
 		for (struct bkey_i *k = keys->keys;
@@ -1421,11 +1422,15 @@ bch2_btree_insert_keys_interior(struct btree_update *as,
 			prt_newline(&buf);
 		}
 
-		panic("%s(): check_topology error: inserted keys\n%s", __func__, buf.buf);
+		bch2_fs_fatal_error(as->c, "%ps -> %s(): check_topology error %s: inserted keys\n%s",
+				    (void *) _RET_IP_, __func__, bch2_err_str(ret), buf.buf);
+		dump_stack();
+		return ret;
 	}
 
 	memmove_u64s_down(keys->keys, insert, keys->top_p - insert->_data);
 	keys->top_p -= insert->_data - keys->keys_p;
+	return 0;
 }
 
 static bool key_deleted_in_insert(struct keylist *insert_keys, struct bpos pos)
@@ -1559,11 +1564,11 @@ static void __btree_split_node(struct btree_update *as,
  * nodes that were coalesced, and thus in the middle of a child node post
  * coalescing:
  */
-static void btree_split_insert_keys(struct btree_update *as,
-				    struct btree_trans *trans,
-				    btree_path_idx_t path_idx,
-				    struct btree *b,
-				    struct keylist *keys)
+static int btree_split_insert_keys(struct btree_update *as,
+				   struct btree_trans *trans,
+				   btree_path_idx_t path_idx,
+				   struct btree *b,
+				   struct keylist *keys)
 {
 	struct btree_path *path = trans->paths + path_idx;
 
@@ -1573,8 +1578,12 @@ static void btree_split_insert_keys(struct btree_update *as,
 
 		bch2_btree_node_iter_init(&node_iter, b, &bch2_keylist_front(keys)->k.p);
 
-		bch2_btree_insert_keys_interior(as, trans, path, b, node_iter, keys);
+		int ret = bch2_btree_insert_keys_interior(as, trans, path, b, node_iter, keys);
+		if (ret)
+			return ret;
 	}
+
+	return 0;
 }
 
 static int btree_split(struct btree_update *as, struct btree_trans *trans,
@@ -1607,8 +1616,10 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans,
 		__btree_split_node(as, trans, b, n, keys);
 
 		if (keys) {
-			btree_split_insert_keys(as, trans, path, n1, keys);
-			btree_split_insert_keys(as, trans, path, n2, keys);
+			ret =   btree_split_insert_keys(as, trans, path, n1, keys) ?:
+				btree_split_insert_keys(as, trans, path, n2, keys);
+			if (ret)
+				goto err;
 			BUG_ON(!bch2_keylist_empty(keys));
 		}
 
@@ -1654,7 +1665,9 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans,
 			n3->sib_u64s[0] = U16_MAX;
 			n3->sib_u64s[1] = U16_MAX;
 
-			btree_split_insert_keys(as, trans, path, n3, &as->parent_keys);
+			ret = btree_split_insert_keys(as, trans, path, n3, &as->parent_keys);
+			if (ret)
+				goto err;
 		}
 	} else {
 		trace_and_count(c, btree_node_compact, trans, b);
@@ -1662,7 +1675,9 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans,
 		n1 = bch2_btree_node_alloc_replacement(as, trans, b);
 
 		if (keys) {
-			btree_split_insert_keys(as, trans, path, n1, keys);
+			ret = btree_split_insert_keys(as, trans, path, n1, keys);
+			if (ret)
+				goto err;
 			BUG_ON(!bch2_keylist_empty(keys));
 		}
 
@@ -1809,15 +1824,15 @@ static int bch2_btree_insert_node(struct btree_update *as, struct btree_trans *t
 		goto split;
 	}
 
-	ret = bch2_btree_node_check_topology(trans, b);
+
+	ret =   bch2_btree_node_check_topology(trans, b) ?:
+		bch2_btree_insert_keys_interior(as, trans, path, b,
+					path->l[b->c.level].iter, keys);
 	if (ret) {
 		bch2_btree_node_unlock_write(trans, path, b);
 		return ret;
 	}
 
-	bch2_btree_insert_keys_interior(as, trans, path, b,
-					path->l[b->c.level].iter, keys);
-
 	trans_for_each_path_with_node(trans, b, linked, i)
 		bch2_btree_node_iter_peek(&linked->l[b->c.level].iter, b);
 
diff --git a/fs/bcachefs/btree_write_buffer.c b/fs/bcachefs/btree_write_buffer.c
index adbe576ec77e..0941fb2c026d 100644
--- a/fs/bcachefs/btree_write_buffer.c
+++ b/fs/bcachefs/btree_write_buffer.c
@@ -428,10 +428,10 @@ static int bch2_btree_write_buffer_flush_locked(struct btree_trans *trans)
 		 */
 		trace_and_count(c, write_buffer_flush_slowpath, trans, slowpath, wb->flushing.keys.nr);
 
-		sort(wb->flushing.keys.data,
-		     wb->flushing.keys.nr,
-		     sizeof(wb->flushing.keys.data[0]),
-		     wb_key_seq_cmp, NULL);
+		sort_nonatomic(wb->flushing.keys.data,
+			       wb->flushing.keys.nr,
+			       sizeof(wb->flushing.keys.data[0]),
+			       wb_key_seq_cmp, NULL);
 
 		darray_for_each(wb->flushing.keys, i) {
 			if (!i->journal_seq)
diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c
index fea61e60a9ee..31fbc2716d8b 100644
--- a/fs/bcachefs/buckets.c
+++ b/fs/bcachefs/buckets.c
@@ -37,7 +37,8 @@ void bch2_dev_usage_read_fast(struct bch_dev *ca, struct bch_dev_usage *usage)
 void bch2_dev_usage_full_read_fast(struct bch_dev *ca, struct bch_dev_usage_full *usage)
 {
 	memset(usage, 0, sizeof(*usage));
-	acc_u64s_percpu((u64 *) usage, (u64 __percpu *) ca->usage, dev_usage_u64s());
+	acc_u64s_percpu((u64 *) usage, (u64 __percpu *) ca->usage,
+			sizeof(struct bch_dev_usage_full) / sizeof(u64));
 }
 
 static u64 reserve_factor(u64 r)
@@ -603,6 +604,13 @@ static int bch2_trigger_pointer(struct btree_trans *trans,
 	}
 
 	struct bpos bucket = PTR_BUCKET_POS(ca, &p.ptr);
+	if (!bucket_valid(ca, bucket.offset)) {
+		if (insert) {
+			bch2_dev_bucket_missing(ca, bucket.offset);
+			ret = -BCH_ERR_trigger_pointer;
+		}
+		goto err;
+	}
 
 	if (flags & BTREE_TRIGGER_transactional) {
 		struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket, 0);
@@ -1306,13 +1314,11 @@ int bch2_dev_buckets_resize(struct bch_fs *c, struct bch_dev *ca, u64 nbuckets)
 	old_bucket_gens = rcu_dereference_protected(ca->bucket_gens, 1);
 
 	if (resize) {
-		bucket_gens->nbuckets = min(bucket_gens->nbuckets,
-					    old_bucket_gens->nbuckets);
-		bucket_gens->nbuckets_minus_first =
-			bucket_gens->nbuckets - bucket_gens->first_bucket;
+		u64 copy = min(bucket_gens->nbuckets,
+			       old_bucket_gens->nbuckets);
 		memcpy(bucket_gens->b,
 		       old_bucket_gens->b,
-		       bucket_gens->nbuckets);
+		       sizeof(bucket_gens->b[0]) * copy);
 	}
 
 	rcu_assign_pointer(ca->bucket_gens, bucket_gens);
diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h
index 1c38b165f48b..8d75b27a1418 100644
--- a/fs/bcachefs/buckets.h
+++ b/fs/bcachefs/buckets.h
@@ -242,11 +242,6 @@ static inline u64 dev_buckets_available(struct bch_dev *ca,
 
 /* Filesystem usage: */
 
-static inline unsigned dev_usage_u64s(void)
-{
-	return sizeof(struct bch_dev_usage) / sizeof(u64);
-}
-
 struct bch_fs_usage_short
 bch2_fs_usage_read_short(struct bch_fs *);
 
diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c
index 3726689093e3..d0a34a097b80 100644
--- a/fs/bcachefs/checksum.c
+++ b/fs/bcachefs/checksum.c
@@ -7,17 +7,12 @@
 #include "super-io.h"
 
 #include <linux/crc32c.h>
-#include <linux/crypto.h>
 #include <linux/xxhash.h>
 #include <linux/key.h>
 #include <linux/random.h>
 #include <linux/ratelimit.h>
-#include <linux/scatterlist.h>
-#include <crypto/algapi.h>
 #include <crypto/chacha.h>
-#include <crypto/hash.h>
 #include <crypto/poly1305.h>
-#include <crypto/skcipher.h>
 #include <keys/user-type.h>
 
 /*
@@ -96,116 +91,40 @@ static void bch2_checksum_update(struct bch2_checksum_state *state, const void *
 	}
 }
 
-static inline int do_encrypt_sg(struct crypto_sync_skcipher *tfm,
-				struct nonce nonce,
-				struct scatterlist *sg, size_t len)
+static void bch2_chacha20_init(u32 state[CHACHA_STATE_WORDS],
+			       const struct bch_key *key, struct nonce nonce)
 {
-	SYNC_SKCIPHER_REQUEST_ON_STACK(req, tfm);
+	u32 key_words[CHACHA_KEY_SIZE / sizeof(u32)];
 
-	skcipher_request_set_sync_tfm(req, tfm);
-	skcipher_request_set_callback(req, 0, NULL, NULL);
-	skcipher_request_set_crypt(req, sg, sg, len, nonce.d);
+	BUILD_BUG_ON(sizeof(key_words) != sizeof(*key));
+	memcpy(key_words, key, sizeof(key_words));
+	le32_to_cpu_array(key_words, ARRAY_SIZE(key_words));
 
-	int ret = crypto_skcipher_encrypt(req);
-	if (ret)
-		pr_err("got error %i from crypto_skcipher_encrypt()", ret);
-
-	return ret;
-}
-
-static inline int do_encrypt(struct crypto_sync_skcipher *tfm,
-			      struct nonce nonce,
-			      void *buf, size_t len)
-{
-	if (!is_vmalloc_addr(buf)) {
-		struct scatterlist sg = {};
-
-		sg_mark_end(&sg);
-		sg_set_page(&sg, virt_to_page(buf), len, offset_in_page(buf));
-		return do_encrypt_sg(tfm, nonce, &sg, len);
-	} else {
-		DARRAY_PREALLOCATED(struct scatterlist, 4) sgl;
-		size_t sgl_len = 0;
-		int ret;
-
-		darray_init(&sgl);
-
-		while (len) {
-			unsigned offset = offset_in_page(buf);
-			struct scatterlist sg = {
-				.page_link	= (unsigned long) vmalloc_to_page(buf),
-				.offset		= offset,
-				.length		= min(len, PAGE_SIZE - offset),
-			};
+	BUILD_BUG_ON(sizeof(nonce) != CHACHA_IV_SIZE);
+	chacha_init(state, key_words, (const u8 *)nonce.d);
 
-			if (darray_push(&sgl, sg)) {
-				sg_mark_end(&darray_last(sgl));
-				ret = do_encrypt_sg(tfm, nonce, sgl.data, sgl_len);
-				if (ret)
-					goto err;
-
-				nonce = nonce_add(nonce, sgl_len);
-				sgl_len = 0;
-				sgl.nr = 0;
-				BUG_ON(darray_push(&sgl, sg));
-			}
-
-			buf += sg.length;
-			len -= sg.length;
-			sgl_len += sg.length;
-		}
-
-		sg_mark_end(&darray_last(sgl));
-		ret = do_encrypt_sg(tfm, nonce, sgl.data, sgl_len);
-err:
-		darray_exit(&sgl);
-		return ret;
-	}
+	memzero_explicit(key_words, sizeof(key_words));
 }
 
-int bch2_chacha_encrypt_key(struct bch_key *key, struct nonce nonce,
-			    void *buf, size_t len)
+static void bch2_chacha20(const struct bch_key *key, struct nonce nonce,
+			  void *data, size_t len)
 {
-	struct crypto_sync_skcipher *chacha20 =
-		crypto_alloc_sync_skcipher("chacha20", 0, 0);
-	int ret;
-
-	ret = PTR_ERR_OR_ZERO(chacha20);
-	if (ret) {
-		pr_err("error requesting chacha20 cipher: %s", bch2_err_str(ret));
-		return ret;
-	}
-
-	ret = crypto_skcipher_setkey(&chacha20->base,
-				     (void *) key, sizeof(*key));
-	if (ret) {
-		pr_err("error from crypto_skcipher_setkey(): %s", bch2_err_str(ret));
-		goto err;
-	}
+	u32 state[CHACHA_STATE_WORDS];
 
-	ret = do_encrypt(chacha20, nonce, buf, len);
-err:
-	crypto_free_sync_skcipher(chacha20);
-	return ret;
+	bch2_chacha20_init(state, key, nonce);
+	chacha20_crypt(state, data, data, len);
+	memzero_explicit(state, sizeof(state));
 }
 
-static int gen_poly_key(struct bch_fs *c, struct shash_desc *desc,
-			struct nonce nonce)
+static void bch2_poly1305_init(struct poly1305_desc_ctx *desc,
+			       struct bch_fs *c, struct nonce nonce)
 {
-	u8 key[POLY1305_KEY_SIZE];
-	int ret;
+	u8 key[POLY1305_KEY_SIZE] = { 0 };
 
 	nonce.d[3] ^= BCH_NONCE_POLY;
 
-	memset(key, 0, sizeof(key));
-	ret = do_encrypt(c->chacha20, nonce, key, sizeof(key));
-	if (ret)
-		return ret;
-
-	desc->tfm = c->poly1305;
-	crypto_shash_init(desc);
-	crypto_shash_update(desc, key, sizeof(key));
-	return 0;
+	bch2_chacha20(&c->chacha20_key, nonce, key, sizeof(key));
+	poly1305_init(desc, key);
 }
 
 struct bch_csum bch2_checksum(struct bch_fs *c, unsigned type,
@@ -230,14 +149,13 @@ struct bch_csum bch2_checksum(struct bch_fs *c, unsigned type,
 
 	case BCH_CSUM_chacha20_poly1305_80:
 	case BCH_CSUM_chacha20_poly1305_128: {
-		SHASH_DESC_ON_STACK(desc, c->poly1305);
+		struct poly1305_desc_ctx dctx;
 		u8 digest[POLY1305_DIGEST_SIZE];
 		struct bch_csum ret = { 0 };
 
-		gen_poly_key(c, desc, nonce);
-
-		crypto_shash_update(desc, data, len);
-		crypto_shash_final(desc, digest);
+		bch2_poly1305_init(&dctx, c, nonce);
+		poly1305_update(&dctx, data, len);
+		poly1305_final(&dctx, digest);
 
 		memcpy(&ret, digest, bch_crc_bytes[type]);
 		return ret;
@@ -253,11 +171,12 @@ int bch2_encrypt(struct bch_fs *c, unsigned type,
 	if (!bch2_csum_type_is_encryption(type))
 		return 0;
 
-	if (bch2_fs_inconsistent_on(!c->chacha20,
+	if (bch2_fs_inconsistent_on(!c->chacha20_key_set,
 				    c, "attempting to encrypt without encryption key"))
 		return -BCH_ERR_no_encryption_key;
 
-	return do_encrypt(c->chacha20, nonce, data, len);
+	bch2_chacha20(&c->chacha20_key, nonce, data, len);
+	return 0;
 }
 
 static struct bch_csum __bch2_checksum_bio(struct bch_fs *c, unsigned type,
@@ -296,26 +215,26 @@ static struct bch_csum __bch2_checksum_bio(struct bch_fs *c, unsigned type,
 
 	case BCH_CSUM_chacha20_poly1305_80:
 	case BCH_CSUM_chacha20_poly1305_128: {
-		SHASH_DESC_ON_STACK(desc, c->poly1305);
+		struct poly1305_desc_ctx dctx;
 		u8 digest[POLY1305_DIGEST_SIZE];
 		struct bch_csum ret = { 0 };
 
-		gen_poly_key(c, desc, nonce);
+		bch2_poly1305_init(&dctx, c, nonce);
 
 #ifdef CONFIG_HIGHMEM
 		__bio_for_each_segment(bv, bio, *iter, *iter) {
 			void *p = kmap_local_page(bv.bv_page) + bv.bv_offset;
 
-			crypto_shash_update(desc, p, bv.bv_len);
+			poly1305_update(&dctx, p, bv.bv_len);
 			kunmap_local(p);
 		}
 #else
 		__bio_for_each_bvec(bv, bio, *iter, *iter)
-			crypto_shash_update(desc,
+			poly1305_update(&dctx,
 				page_address(bv.bv_page) + bv.bv_offset,
 				bv.bv_len);
 #endif
-		crypto_shash_final(desc, digest);
+		poly1305_final(&dctx, digest);
 
 		memcpy(&ret, digest, bch_crc_bytes[type]);
 		return ret;
@@ -338,43 +257,33 @@ int __bch2_encrypt_bio(struct bch_fs *c, unsigned type,
 {
 	struct bio_vec bv;
 	struct bvec_iter iter;
-	DARRAY_PREALLOCATED(struct scatterlist, 4) sgl;
-	size_t sgl_len = 0;
+	u32 chacha_state[CHACHA_STATE_WORDS];
 	int ret = 0;
 
-	if (bch2_fs_inconsistent_on(!c->chacha20,
+	if (bch2_fs_inconsistent_on(!c->chacha20_key_set,
 				    c, "attempting to encrypt without encryption key"))
 		return -BCH_ERR_no_encryption_key;
 
-	darray_init(&sgl);
+	bch2_chacha20_init(chacha_state, &c->chacha20_key, nonce);
 
 	bio_for_each_segment(bv, bio, iter) {
-		struct scatterlist sg = {
-			.page_link	= (unsigned long) bv.bv_page,
-			.offset		= bv.bv_offset,
-			.length		= bv.bv_len,
-		};
-
-		if (darray_push(&sgl, sg)) {
-			sg_mark_end(&darray_last(sgl));
-			ret = do_encrypt_sg(c->chacha20, nonce, sgl.data, sgl_len);
-			if (ret)
-				goto err;
-
-			nonce = nonce_add(nonce, sgl_len);
-			sgl_len = 0;
-			sgl.nr = 0;
-
-			BUG_ON(darray_push(&sgl, sg));
+		void *p;
+
+		/*
+		 * chacha_crypt() assumes that the length is a multiple of
+		 * CHACHA_BLOCK_SIZE on any non-final call.
+		 */
+		if (!IS_ALIGNED(bv.bv_len, CHACHA_BLOCK_SIZE)) {
+			bch_err_ratelimited(c, "bio not aligned for encryption");
+			ret = -EIO;
+			break;
 		}
 
-		sgl_len += sg.length;
+		p = bvec_kmap_local(&bv);
+		chacha20_crypt(chacha_state, p, p, bv.bv_len);
+		kunmap_local(p);
 	}
-
-	sg_mark_end(&darray_last(sgl));
-	ret = do_encrypt_sg(c->chacha20, nonce, sgl.data, sgl_len);
-err:
-	darray_exit(&sgl);
+	memzero_explicit(chacha_state, sizeof(chacha_state));
 	return ret;
 }
 
@@ -650,10 +559,7 @@ int bch2_decrypt_sb_key(struct bch_fs *c,
 	}
 
 	/* decrypt real key: */
-	ret = bch2_chacha_encrypt_key(&user_key, bch2_sb_key_nonce(c),
-				      &sb_key, sizeof(sb_key));
-	if (ret)
-		goto err;
+	bch2_chacha20(&user_key, bch2_sb_key_nonce(c), &sb_key, sizeof(sb_key));
 
 	if (bch2_key_is_encrypted(&sb_key)) {
 		bch_err(c, "incorrect encryption key");
@@ -668,31 +574,6 @@ err:
 	return ret;
 }
 
-static int bch2_alloc_ciphers(struct bch_fs *c)
-{
-	if (c->chacha20)
-		return 0;
-
-	struct crypto_sync_skcipher *chacha20 = crypto_alloc_sync_skcipher("chacha20", 0, 0);
-	int ret = PTR_ERR_OR_ZERO(chacha20);
-	if (ret) {
-		bch_err(c, "error requesting chacha20 module: %s", bch2_err_str(ret));
-		return ret;
-	}
-
-	struct crypto_shash *poly1305 = crypto_alloc_shash("poly1305", 0, 0);
-	ret = PTR_ERR_OR_ZERO(poly1305);
-	if (ret) {
-		bch_err(c, "error requesting poly1305 module: %s", bch2_err_str(ret));
-		crypto_free_sync_skcipher(chacha20);
-		return ret;
-	}
-
-	c->chacha20	= chacha20;
-	c->poly1305	= poly1305;
-	return 0;
-}
-
 #if 0
 
 /*
@@ -797,35 +678,21 @@ err:
 
 void bch2_fs_encryption_exit(struct bch_fs *c)
 {
-	if (c->poly1305)
-		crypto_free_shash(c->poly1305);
-	if (c->chacha20)
-		crypto_free_sync_skcipher(c->chacha20);
+	memzero_explicit(&c->chacha20_key, sizeof(c->chacha20_key));
 }
 
 int bch2_fs_encryption_init(struct bch_fs *c)
 {
 	struct bch_sb_field_crypt *crypt;
-	struct bch_key key;
-	int ret = 0;
+	int ret;
 
 	crypt = bch2_sb_field_get(c->disk_sb.sb, crypt);
 	if (!crypt)
-		goto out;
+		return 0;
 
-	ret = bch2_alloc_ciphers(c);
+	ret = bch2_decrypt_sb_key(c, crypt, &c->chacha20_key);
 	if (ret)
-		goto out;
-
-	ret = bch2_decrypt_sb_key(c, crypt, &key);
-	if (ret)
-		goto out;
-
-	ret = crypto_skcipher_setkey(&c->chacha20->base,
-			(void *) &key.key, sizeof(key.key));
-	if (ret)
-		goto out;
-out:
-	memzero_explicit(&key, sizeof(key));
-	return ret;
+		return ret;
+	c->chacha20_key_set = true;
+	return 0;
 }
diff --git a/fs/bcachefs/checksum.h b/fs/bcachefs/checksum.h
index 4ac251c8fcd8..1310782d3ae9 100644
--- a/fs/bcachefs/checksum.h
+++ b/fs/bcachefs/checksum.h
@@ -69,7 +69,6 @@ static inline void bch2_csum_err_msg(struct printbuf *out,
 	bch2_csum_to_text(out, type, expected);
 }
 
-int bch2_chacha_encrypt_key(struct bch_key *, struct nonce, void *, size_t);
 int bch2_request_key(struct bch_sb *, struct bch_key *);
 #ifndef __KERNEL__
 int bch2_revoke_key(struct bch_sb *);
@@ -156,7 +155,7 @@ static inline bool bch2_checksum_type_valid(const struct bch_fs *c,
 	if (type >= BCH_CSUM_NR)
 		return false;
 
-	if (bch2_csum_type_is_encryption(type) && !c->chacha20)
+	if (bch2_csum_type_is_encryption(type) && !c->chacha20_key_set)
 		return false;
 
 	return true;
diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c
index de02ebf847ec..b211c97238ab 100644
--- a/fs/bcachefs/data_update.c
+++ b/fs/bcachefs/data_update.c
@@ -607,7 +607,7 @@ void bch2_data_update_inflight_to_text(struct printbuf *out, struct data_update
 	prt_newline(out);
 	printbuf_indent_add(out, 2);
 	bch2_data_update_opts_to_text(out, m->op.c, &m->op.opts, &m->data_opts);
-	prt_printf(out, "read_done:\t\%u\n", m->read_done);
+	prt_printf(out, "read_done:\t%u\n", m->read_done);
 	bch2_write_op_to_text(out, &m->op);
 	printbuf_indent_sub(out, 2);
 }
diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c
index bf53a029f356..8a680e52c1ed 100644
--- a/fs/bcachefs/dirent.c
+++ b/fs/bcachefs/dirent.c
@@ -13,8 +13,8 @@
 
 #include <linux/dcache.h>
 
-static int bch2_casefold(struct btree_trans *trans, const struct bch_hash_info *info,
-				const struct qstr *str, struct qstr *out_cf)
+int bch2_casefold(struct btree_trans *trans, const struct bch_hash_info *info,
+		  const struct qstr *str, struct qstr *out_cf)
 {
 	*out_cf = (struct qstr) QSTR_INIT(NULL, 0);
 
@@ -35,18 +35,6 @@ static int bch2_casefold(struct btree_trans *trans, const struct bch_hash_info *
 #endif
 }
 
-static inline int bch2_maybe_casefold(struct btree_trans *trans,
-				      const struct bch_hash_info *info,
-				      const struct qstr *str, struct qstr *out_cf)
-{
-	if (likely(!info->cf_encoding)) {
-		*out_cf = *str;
-		return 0;
-	} else {
-		return bch2_casefold(trans, info, str, out_cf);
-	}
-}
-
 static unsigned bch2_dirent_name_bytes(struct bkey_s_c_dirent d)
 {
 	if (bkey_val_bytes(d.k) < offsetof(struct bch_dirent, d_name))
@@ -287,8 +275,8 @@ static void dirent_init_casefolded_name(struct bkey_i_dirent *dirent,
 	EBUG_ON(!dirent->v.d_casefold);
 	EBUG_ON(!cf_name->len);
 
-	dirent->v.d_cf_name_block.d_name_len = name->len;
-	dirent->v.d_cf_name_block.d_cf_name_len = cf_name->len;
+	dirent->v.d_cf_name_block.d_name_len = cpu_to_le16(name->len);
+	dirent->v.d_cf_name_block.d_cf_name_len = cpu_to_le16(cf_name->len);
 	memcpy(&dirent->v.d_cf_name_block.d_names[0], name->name, name->len);
 	memcpy(&dirent->v.d_cf_name_block.d_names[name->len], cf_name->name, cf_name->len);
 	memset(&dirent->v.d_cf_name_block.d_names[name->len + cf_name->len], 0,
@@ -697,7 +685,7 @@ static int bch2_dir_emit(struct dir_context *ctx, struct bkey_s_c_dirent d, subv
 		      vfs_d_type(d.v->d_type));
 	if (ret)
 		ctx->pos = d.k->p.offset + 1;
-	return ret;
+	return !ret;
 }
 
 int bch2_readdir(struct bch_fs *c, subvol_inum inum, struct dir_context *ctx)
@@ -722,7 +710,7 @@ int bch2_readdir(struct bch_fs *c, subvol_inum inum, struct dir_context *ctx)
 			if (ret2 > 0)
 				continue;
 
-			ret2 ?: drop_locks_do(trans, bch2_dir_emit(ctx, dirent, target));
+			ret2 ?: (bch2_trans_unlock(trans), bch2_dir_emit(ctx, dirent, target));
 		})));
 
 	bch2_bkey_buf_exit(&sk, c);
diff --git a/fs/bcachefs/dirent.h b/fs/bcachefs/dirent.h
index 0880772b80a9..9838a7ba7ed1 100644
--- a/fs/bcachefs/dirent.h
+++ b/fs/bcachefs/dirent.h
@@ -23,6 +23,21 @@ struct bch_fs;
 struct bch_hash_info;
 struct bch_inode_info;
 
+int bch2_casefold(struct btree_trans *, const struct bch_hash_info *,
+		  const struct qstr *, struct qstr *);
+
+static inline int bch2_maybe_casefold(struct btree_trans *trans,
+				      const struct bch_hash_info *info,
+				      const struct qstr *str, struct qstr *out_cf)
+{
+	if (likely(!info->cf_encoding)) {
+		*out_cf = *str;
+		return 0;
+	} else {
+		return bch2_casefold(trans, info, str, out_cf);
+	}
+}
+
 struct qstr bch2_dirent_get_name(struct bkey_s_c_dirent d);
 
 static inline unsigned dirent_val_u64s(unsigned len, unsigned cf_len)
diff --git a/fs/bcachefs/disk_groups.c b/fs/bcachefs/disk_groups.c
index 1186280b29e9..2ca3cbf12b71 100644
--- a/fs/bcachefs/disk_groups.c
+++ b/fs/bcachefs/disk_groups.c
@@ -470,23 +470,22 @@ inval:
 
 int __bch2_dev_group_set(struct bch_fs *c, struct bch_dev *ca, const char *name)
 {
-	struct bch_member *mi;
-	int ret, v = -1;
+	lockdep_assert_held(&c->sb_lock);
 
-	if (!strlen(name) || !strcmp(name, "none"))
-		return 0;
 
-	v = bch2_disk_path_find_or_create(&c->disk_sb, name);
-	if (v < 0)
-		return v;
+	if (!strlen(name) || !strcmp(name, "none")) {
+		struct bch_member *mi = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx);
+		SET_BCH_MEMBER_GROUP(mi, 0);
+	} else {
+		int v = bch2_disk_path_find_or_create(&c->disk_sb, name);
+		if (v < 0)
+			return v;
 
-	ret = bch2_sb_disk_groups_to_cpu(c);
-	if (ret)
-		return ret;
+		struct bch_member *mi = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx);
+		SET_BCH_MEMBER_GROUP(mi, v + 1);
+	}
 
-	mi = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx);
-	SET_BCH_MEMBER_GROUP(mi, v + 1);
-	return 0;
+	return bch2_sb_disk_groups_to_cpu(c);
 }
 
 int bch2_dev_group_set(struct bch_fs *c, struct bch_dev *ca, const char *name)
diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c
index a396865e8b17..fff58b78327c 100644
--- a/fs/bcachefs/ec.c
+++ b/fs/bcachefs/ec.c
@@ -2204,10 +2204,10 @@ void bch2_fs_ec_stop(struct bch_fs *c)
 
 static bool bch2_fs_ec_flush_done(struct bch_fs *c)
 {
-	bool ret;
+	sched_annotate_sleep();
 
 	mutex_lock(&c->ec_stripe_new_lock);
-	ret = list_empty(&c->ec_stripe_new_list);
+	bool ret = list_empty(&c->ec_stripe_new_list);
 	mutex_unlock(&c->ec_stripe_new_lock);
 
 	return ret;
diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h
index c8696f01eb14..a615e4852ded 100644
--- a/fs/bcachefs/errcode.h
+++ b/fs/bcachefs/errcode.h
@@ -287,7 +287,7 @@
 	x(EIO,				mark_stripe)				\
 	x(EIO,				stripe_reconstruct)			\
 	x(EIO,				key_type_error)				\
-	x(EIO,				extent_poisened)			\
+	x(EIO,				extent_poisoned)			\
 	x(EIO,				missing_indirect_extent)		\
 	x(EIO,				invalidate_stripe_to_dev)		\
 	x(EIO,				no_encryption_key)			\
diff --git a/fs/bcachefs/error.c b/fs/bcachefs/error.c
index baf5dfb32298..6b8695b1349c 100644
--- a/fs/bcachefs/error.c
+++ b/fs/bcachefs/error.c
@@ -272,9 +272,6 @@ static struct fsck_err_state *fsck_err_get(struct bch_fs *c,
 {
 	struct fsck_err_state *s;
 
-	if (!test_bit(BCH_FS_fsck_running, &c->flags))
-		return NULL;
-
 	list_for_each_entry(s, &c->fsck_error_msgs, list)
 		if (s->id == id) {
 			/*
@@ -481,7 +478,9 @@ int __bch2_fsck_err(struct bch_fs *c,
 	} else if (!test_bit(BCH_FS_fsck_running, &c->flags)) {
 		if (c->opts.errors != BCH_ON_ERROR_continue ||
 		    !(flags & (FSCK_CAN_FIX|FSCK_CAN_IGNORE))) {
-			prt_str(out, ", shutting down");
+			prt_str_indented(out, ", shutting down\n"
+					 "error not marked as autofix and not in fsck\n"
+					 "run fsck, and forward to devs so error can be marked for self-healing");
 			inconsistent = true;
 			print = true;
 			ret = -BCH_ERR_fsck_errors_not_fixed;
@@ -639,14 +638,14 @@ int __bch2_bkey_fsck_err(struct bch_fs *c,
 	return ret;
 }
 
-void bch2_flush_fsck_errs(struct bch_fs *c)
+static void __bch2_flush_fsck_errs(struct bch_fs *c, bool print)
 {
 	struct fsck_err_state *s, *n;
 
 	mutex_lock(&c->fsck_error_msgs_lock);
 
 	list_for_each_entry_safe(s, n, &c->fsck_error_msgs, list) {
-		if (s->ratelimited && s->last_msg)
+		if (print && s->ratelimited && s->last_msg)
 			bch_err(c, "Saw %llu errors like:\n  %s", s->nr, s->last_msg);
 
 		list_del(&s->list);
@@ -657,6 +656,16 @@ void bch2_flush_fsck_errs(struct bch_fs *c)
 	mutex_unlock(&c->fsck_error_msgs_lock);
 }
 
+void bch2_flush_fsck_errs(struct bch_fs *c)
+{
+	__bch2_flush_fsck_errs(c, true);
+}
+
+void bch2_free_fsck_errs(struct bch_fs *c)
+{
+	__bch2_flush_fsck_errs(c, false);
+}
+
 int bch2_inum_offset_err_msg_trans(struct btree_trans *trans, struct printbuf *out,
 				    subvol_inum inum, u64 offset)
 {
diff --git a/fs/bcachefs/error.h b/fs/bcachefs/error.h
index d0d024dc714b..4a364fd44abe 100644
--- a/fs/bcachefs/error.h
+++ b/fs/bcachefs/error.h
@@ -93,6 +93,7 @@ int __bch2_fsck_err(struct bch_fs *, struct btree_trans *,
 			_flags, BCH_FSCK_ERR_##_err_type, __VA_ARGS__)
 
 void bch2_flush_fsck_errs(struct bch_fs *);
+void bch2_free_fsck_errs(struct bch_fs *);
 
 #define fsck_err_wrap(_do)						\
 ({									\
diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c
index ae7c7a177e10..dca2b8425cc0 100644
--- a/fs/bcachefs/extents.c
+++ b/fs/bcachefs/extents.c
@@ -139,7 +139,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k,
 	struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
 
 	if (bch2_bkey_extent_ptrs_flags(ptrs) & BIT_ULL(BCH_EXTENT_FLAG_poisoned))
-		return -BCH_ERR_extent_poisened;
+		return -BCH_ERR_extent_poisoned;
 
 	rcu_read_lock();
 	const union bch_extent_entry *entry;
diff --git a/fs/bcachefs/fs-io-buffered.c b/fs/bcachefs/fs-io-buffered.c
index 19d4599918dc..e3a75dcca60c 100644
--- a/fs/bcachefs/fs-io-buffered.c
+++ b/fs/bcachefs/fs-io-buffered.c
@@ -225,11 +225,26 @@ static void bchfs_read(struct btree_trans *trans,
 
 		bch2_read_extent(trans, rbio, iter.pos,
 				 data_btree, k, offset_into_extent, flags);
-		swap(rbio->bio.bi_iter.bi_size, bytes);
+		/*
+		 * Careful there's a landmine here if bch2_read_extent() ever
+		 * starts returning transaction restarts here.
+		 *
+		 * We've changed rbio->bi_iter.bi_size to be "bytes we can read
+		 * from this extent" with the swap call, and we restore it
+		 * below. That restore needs to come before checking for
+		 * errors.
+		 *
+		 * But unlike __bch2_read(), we use the rbio bvec iter, not one
+		 * on the stack, so we can't do the restore right after the
+		 * bch2_read_extent() call: we don't own that iterator anymore
+		 * if BCH_READ_last_fragment is set, since we may have submitted
+		 * that rbio instead of cloning it.
+		 */
 
 		if (flags & BCH_READ_last_fragment)
 			break;
 
+		swap(rbio->bio.bi_iter.bi_size, bytes);
 		bio_advance(&rbio->bio, bytes);
 err:
 		if (ret &&
diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c
index 65c2c33d253d..9657144666b8 100644
--- a/fs/bcachefs/fs-io.c
+++ b/fs/bcachefs/fs-io.c
@@ -144,10 +144,25 @@ int __must_check bch2_write_inode_size(struct bch_fs *c,
 void __bch2_i_sectors_acct(struct bch_fs *c, struct bch_inode_info *inode,
 			   struct quota_res *quota_res, s64 sectors)
 {
-	bch2_fs_inconsistent_on((s64) inode->v.i_blocks + sectors < 0, c,
-				"inode %lu i_blocks underflow: %llu + %lli < 0 (ondisk %lli)",
-				inode->v.i_ino, (u64) inode->v.i_blocks, sectors,
-				inode->ei_inode.bi_sectors);
+	if (unlikely((s64) inode->v.i_blocks + sectors < 0)) {
+		struct printbuf buf = PRINTBUF;
+		bch2_log_msg_start(c, &buf);
+		prt_printf(&buf, "inode %lu i_blocks underflow: %llu + %lli < 0 (ondisk %lli)",
+			   inode->v.i_ino, (u64) inode->v.i_blocks, sectors,
+			   inode->ei_inode.bi_sectors);
+
+		bool repeat = false, print = false, suppress = false;
+		bch2_count_fsck_err(c, vfs_inode_i_blocks_underflow, buf.buf, &repeat, &print, &suppress);
+		if (print)
+			bch2_print_str(c, buf.buf);
+		printbuf_exit(&buf);
+
+		if (sectors < 0)
+			sectors = -inode->v.i_blocks;
+		else
+			sectors = 0;
+	}
+
 	inode->v.i_blocks += sectors;
 
 #ifdef CONFIG_BCACHEFS_QUOTA
@@ -502,11 +517,22 @@ int bchfs_truncate(struct mnt_idmap *idmap,
 		goto err;
 	}
 
-	bch2_fs_inconsistent_on(!inode->v.i_size && inode->v.i_blocks &&
-				!bch2_journal_error(&c->journal), c,
-				"inode %lu truncated to 0 but i_blocks %llu (ondisk %lli)",
-				inode->v.i_ino, (u64) inode->v.i_blocks,
-				inode->ei_inode.bi_sectors);
+	if (unlikely(!inode->v.i_size && inode->v.i_blocks &&
+		     !bch2_journal_error(&c->journal))) {
+		struct printbuf buf = PRINTBUF;
+		bch2_log_msg_start(c, &buf);
+		prt_printf(&buf,
+			   "inode %lu truncated to 0 but i_blocks %llu (ondisk %lli)",
+			   inode->v.i_ino, (u64) inode->v.i_blocks,
+			   inode->ei_inode.bi_sectors);
+
+		bool repeat = false, print = false, suppress = false;
+		bch2_count_fsck_err(c, vfs_inode_i_blocks_not_zero_at_truncate, buf.buf,
+				    &repeat, &print, &suppress);
+		if (print)
+			bch2_print_str(c, buf.buf);
+		printbuf_exit(&buf);
+	}
 
 	ret = bch2_setattr_nonsize(idmap, inode, iattr);
 err:
diff --git a/fs/bcachefs/fs-ioctl.c b/fs/bcachefs/fs-ioctl.c
index c1553e44e049..a82dfce9e4ad 100644
--- a/fs/bcachefs/fs-ioctl.c
+++ b/fs/bcachefs/fs-ioctl.c
@@ -21,206 +21,6 @@
 #define FSOP_GOING_FLAGS_LOGFLUSH	0x1	/* flush log but not data */
 #define FSOP_GOING_FLAGS_NOLOGFLUSH	0x2	/* don't flush log nor data */
 
-struct flags_set {
-	unsigned		mask;
-	unsigned		flags;
-
-	unsigned		projid;
-
-	bool			set_projinherit;
-	bool			projinherit;
-};
-
-static int bch2_inode_flags_set(struct btree_trans *trans,
-				struct bch_inode_info *inode,
-				struct bch_inode_unpacked *bi,
-				void *p)
-{
-	struct bch_fs *c = inode->v.i_sb->s_fs_info;
-	/*
-	 * We're relying on btree locking here for exclusion with other ioctl
-	 * calls - use the flags in the btree (@bi), not inode->i_flags:
-	 */
-	struct flags_set *s = p;
-	unsigned newflags = s->flags;
-	unsigned oldflags = bi->bi_flags & s->mask;
-
-	if (((newflags ^ oldflags) & (BCH_INODE_append|BCH_INODE_immutable)) &&
-	    !capable(CAP_LINUX_IMMUTABLE))
-		return -EPERM;
-
-	if (!S_ISREG(bi->bi_mode) &&
-	    !S_ISDIR(bi->bi_mode) &&
-	    (newflags & (BCH_INODE_nodump|BCH_INODE_noatime)) != newflags)
-		return -EINVAL;
-
-	if ((newflags ^ oldflags) & BCH_INODE_casefolded) {
-#ifdef CONFIG_UNICODE
-		int ret = 0;
-		/* Not supported on individual files. */
-		if (!S_ISDIR(bi->bi_mode))
-			return -EOPNOTSUPP;
-
-		/*
-		 * Make sure the dir is empty, as otherwise we'd need to
-		 * rehash everything and update the dirent keys.
-		 */
-		ret = bch2_empty_dir_trans(trans, inode_inum(inode));
-		if (ret < 0)
-			return ret;
-
-		ret = bch2_request_incompat_feature(c,bcachefs_metadata_version_casefolding);
-		if (ret)
-			return ret;
-
-		bch2_check_set_feature(c, BCH_FEATURE_casefolding);
-#else
-		printk(KERN_ERR "Cannot use casefolding on a kernel without CONFIG_UNICODE\n");
-		return -EOPNOTSUPP;
-#endif
-	}
-
-	if (s->set_projinherit) {
-		bi->bi_fields_set &= ~(1 << Inode_opt_project);
-		bi->bi_fields_set |= ((int) s->projinherit << Inode_opt_project);
-	}
-
-	bi->bi_flags &= ~s->mask;
-	bi->bi_flags |= newflags;
-
-	bi->bi_ctime = timespec_to_bch2_time(c, current_time(&inode->v));
-	return 0;
-}
-
-static int bch2_ioc_getflags(struct bch_inode_info *inode, int __user *arg)
-{
-	unsigned flags = map_flags(bch_flags_to_uflags, inode->ei_inode.bi_flags);
-
-	return put_user(flags, arg);
-}
-
-static int bch2_ioc_setflags(struct bch_fs *c,
-			     struct file *file,
-			     struct bch_inode_info *inode,
-			     void __user *arg)
-{
-	struct flags_set s = { .mask = map_defined(bch_flags_to_uflags) };
-	unsigned uflags;
-	int ret;
-
-	if (get_user(uflags, (int __user *) arg))
-		return -EFAULT;
-
-	s.flags = map_flags_rev(bch_flags_to_uflags, uflags);
-	if (uflags)
-		return -EOPNOTSUPP;
-
-	ret = mnt_want_write_file(file);
-	if (ret)
-		return ret;
-
-	inode_lock(&inode->v);
-	if (!inode_owner_or_capable(file_mnt_idmap(file), &inode->v)) {
-		ret = -EACCES;
-		goto setflags_out;
-	}
-
-	mutex_lock(&inode->ei_update_lock);
-	ret   = bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?:
-		bch2_write_inode(c, inode, bch2_inode_flags_set, &s,
-			       ATTR_CTIME);
-	mutex_unlock(&inode->ei_update_lock);
-
-setflags_out:
-	inode_unlock(&inode->v);
-	mnt_drop_write_file(file);
-	return ret;
-}
-
-static int bch2_ioc_fsgetxattr(struct bch_inode_info *inode,
-			       struct fsxattr __user *arg)
-{
-	struct fsxattr fa = { 0 };
-
-	fa.fsx_xflags = map_flags(bch_flags_to_xflags, inode->ei_inode.bi_flags);
-
-	if (inode->ei_inode.bi_fields_set & (1 << Inode_opt_project))
-		fa.fsx_xflags |= FS_XFLAG_PROJINHERIT;
-
-	fa.fsx_projid = inode->ei_qid.q[QTYP_PRJ];
-
-	if (copy_to_user(arg, &fa, sizeof(fa)))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int fssetxattr_inode_update_fn(struct btree_trans *trans,
-				      struct bch_inode_info *inode,
-				      struct bch_inode_unpacked *bi,
-				      void *p)
-{
-	struct flags_set *s = p;
-
-	if (s->projid != bi->bi_project) {
-		bi->bi_fields_set |= 1U << Inode_opt_project;
-		bi->bi_project = s->projid;
-	}
-
-	return bch2_inode_flags_set(trans, inode, bi, p);
-}
-
-static int bch2_ioc_fssetxattr(struct bch_fs *c,
-			       struct file *file,
-			       struct bch_inode_info *inode,
-			       struct fsxattr __user *arg)
-{
-	struct flags_set s = { .mask = map_defined(bch_flags_to_xflags) };
-	struct fsxattr fa;
-	int ret;
-
-	if (copy_from_user(&fa, arg, sizeof(fa)))
-		return -EFAULT;
-
-	s.set_projinherit = true;
-	s.projinherit = (fa.fsx_xflags & FS_XFLAG_PROJINHERIT) != 0;
-	fa.fsx_xflags &= ~FS_XFLAG_PROJINHERIT;
-
-	s.flags = map_flags_rev(bch_flags_to_xflags, fa.fsx_xflags);
-	if (fa.fsx_xflags)
-		return -EOPNOTSUPP;
-
-	if (fa.fsx_projid >= U32_MAX)
-		return -EINVAL;
-
-	/*
-	 * inode fields accessible via the xattr interface are stored with a +1
-	 * bias, so that 0 means unset:
-	 */
-	s.projid = fa.fsx_projid + 1;
-
-	ret = mnt_want_write_file(file);
-	if (ret)
-		return ret;
-
-	inode_lock(&inode->v);
-	if (!inode_owner_or_capable(file_mnt_idmap(file), &inode->v)) {
-		ret = -EACCES;
-		goto err;
-	}
-
-	mutex_lock(&inode->ei_update_lock);
-	ret   = bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?:
-		bch2_set_projid(c, inode, fa.fsx_projid) ?:
-		bch2_write_inode(c, inode, fssetxattr_inode_update_fn, &s,
-			       ATTR_CTIME);
-	mutex_unlock(&inode->ei_update_lock);
-err:
-	inode_unlock(&inode->v);
-	mnt_drop_write_file(file);
-	return ret;
-}
-
 static int bch2_reinherit_attrs_fn(struct btree_trans *trans,
 				   struct bch_inode_info *inode,
 				   struct bch_inode_unpacked *bi,
@@ -558,23 +358,6 @@ long bch2_fs_file_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 	long ret;
 
 	switch (cmd) {
-	case FS_IOC_GETFLAGS:
-		ret = bch2_ioc_getflags(inode, (int __user *) arg);
-		break;
-
-	case FS_IOC_SETFLAGS:
-		ret = bch2_ioc_setflags(c, file, inode, (int __user *) arg);
-		break;
-
-	case FS_IOC_FSGETXATTR:
-		ret = bch2_ioc_fsgetxattr(inode, (void __user *) arg);
-		break;
-
-	case FS_IOC_FSSETXATTR:
-		ret = bch2_ioc_fssetxattr(c, file, inode,
-					  (void __user *) arg);
-		break;
-
 	case BCHFS_IOC_REINHERIT_ATTRS:
 		ret = bch2_ioc_reinherit_attrs(c, file, inode,
 					       (void __user *) arg);
diff --git a/fs/bcachefs/fs-ioctl.h b/fs/bcachefs/fs-ioctl.h
index ecd3bfdcde21..a657e4994b71 100644
--- a/fs/bcachefs/fs-ioctl.h
+++ b/fs/bcachefs/fs-ioctl.h
@@ -2,81 +2,6 @@
 #ifndef _BCACHEFS_FS_IOCTL_H
 #define _BCACHEFS_FS_IOCTL_H
 
-/* Inode flags: */
-
-/* bcachefs inode flags -> vfs inode flags: */
-static const __maybe_unused unsigned bch_flags_to_vfs[] = {
-	[__BCH_INODE_sync]		= S_SYNC,
-	[__BCH_INODE_immutable]		= S_IMMUTABLE,
-	[__BCH_INODE_append]		= S_APPEND,
-	[__BCH_INODE_noatime]		= S_NOATIME,
-	[__BCH_INODE_casefolded]	= S_CASEFOLD,
-};
-
-/* bcachefs inode flags -> FS_IOC_GETFLAGS: */
-static const __maybe_unused unsigned bch_flags_to_uflags[] = {
-	[__BCH_INODE_sync]		= FS_SYNC_FL,
-	[__BCH_INODE_immutable]		= FS_IMMUTABLE_FL,
-	[__BCH_INODE_append]		= FS_APPEND_FL,
-	[__BCH_INODE_nodump]		= FS_NODUMP_FL,
-	[__BCH_INODE_noatime]		= FS_NOATIME_FL,
-	[__BCH_INODE_casefolded]	= FS_CASEFOLD_FL,
-};
-
-/* bcachefs inode flags -> FS_IOC_FSGETXATTR: */
-static const __maybe_unused unsigned bch_flags_to_xflags[] = {
-	[__BCH_INODE_sync]	= FS_XFLAG_SYNC,
-	[__BCH_INODE_immutable]	= FS_XFLAG_IMMUTABLE,
-	[__BCH_INODE_append]	= FS_XFLAG_APPEND,
-	[__BCH_INODE_nodump]	= FS_XFLAG_NODUMP,
-	[__BCH_INODE_noatime]	= FS_XFLAG_NOATIME,
-	//[__BCH_INODE_PROJINHERIT] = FS_XFLAG_PROJINHERIT;
-};
-
-#define set_flags(_map, _in, _out)					\
-do {									\
-	unsigned _i;							\
-									\
-	for (_i = 0; _i < ARRAY_SIZE(_map); _i++)			\
-		if ((_in) & (1 << _i))					\
-			(_out) |= _map[_i];				\
-		else							\
-			(_out) &= ~_map[_i];				\
-} while (0)
-
-#define map_flags(_map, _in)						\
-({									\
-	unsigned _out = 0;						\
-									\
-	set_flags(_map, _in, _out);					\
-	_out;								\
-})
-
-#define map_flags_rev(_map, _in)					\
-({									\
-	unsigned _i, _out = 0;						\
-									\
-	for (_i = 0; _i < ARRAY_SIZE(_map); _i++)			\
-		if ((_in) & _map[_i]) {					\
-			(_out) |= 1 << _i;				\
-			(_in) &= ~_map[_i];				\
-		}							\
-	(_out);								\
-})
-
-#define map_defined(_map)						\
-({									\
-	unsigned _in = ~0;						\
-									\
-	map_flags_rev(_map, _in);					\
-})
-
-/* Set VFS inode flags from bcachefs inode: */
-static inline void bch2_inode_flags_to_vfs(struct bch_inode_info *inode)
-{
-	set_flags(bch_flags_to_vfs, inode->ei_inode.bi_flags, inode->v.i_flags);
-}
-
 long bch2_fs_file_ioctl(struct file *, unsigned, unsigned long);
 long bch2_compat_fs_ioctl(struct file *, unsigned, unsigned long);
 
diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c
index 5a41b1a8e54f..113db85b6ef9 100644
--- a/fs/bcachefs/fs.c
+++ b/fs/bcachefs/fs.c
@@ -33,6 +33,7 @@
 #include <linux/backing-dev.h>
 #include <linux/exportfs.h>
 #include <linux/fiemap.h>
+#include <linux/fileattr.h>
 #include <linux/fs_context.h>
 #include <linux/module.h>
 #include <linux/pagemap.h>
@@ -51,6 +52,24 @@ static void bch2_vfs_inode_init(struct btree_trans *, subvol_inum,
 				struct bch_inode_unpacked *,
 				struct bch_subvolume *);
 
+/* Set VFS inode flags from bcachefs inode: */
+static inline void bch2_inode_flags_to_vfs(struct bch_fs *c, struct bch_inode_info *inode)
+{
+	static const __maybe_unused unsigned bch_flags_to_vfs[] = {
+		[__BCH_INODE_sync]		= S_SYNC,
+		[__BCH_INODE_immutable]		= S_IMMUTABLE,
+		[__BCH_INODE_append]		= S_APPEND,
+		[__BCH_INODE_noatime]		= S_NOATIME,
+	};
+
+	set_flags(bch_flags_to_vfs, inode->ei_inode.bi_flags, inode->v.i_flags);
+
+	if (bch2_inode_casefold(c, &inode->ei_inode))
+		inode->v.i_flags |= S_CASEFOLD;
+	else
+		inode->v.i_flags &= ~S_CASEFOLD;
+}
+
 void bch2_inode_update_after_write(struct btree_trans *trans,
 				   struct bch_inode_info *inode,
 				   struct bch_inode_unpacked *bi,
@@ -79,7 +98,7 @@ void bch2_inode_update_after_write(struct btree_trans *trans,
 
 	inode->ei_inode		= *bi;
 
-	bch2_inode_flags_to_vfs(inode);
+	bch2_inode_flags_to_vfs(c, inode);
 }
 
 int __must_check bch2_write_inode(struct bch_fs *c,
@@ -631,13 +650,18 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans,
 			const struct qstr *name)
 {
 	struct bch_fs *c = trans->c;
-	struct btree_iter dirent_iter = {};
 	subvol_inum inum = {};
 	struct printbuf buf = PRINTBUF;
 
+	struct qstr lookup_name;
+	int ret = bch2_maybe_casefold(trans, dir_hash_info, name, &lookup_name);
+	if (ret)
+		return ERR_PTR(ret);
+
+	struct btree_iter dirent_iter = {};
 	struct bkey_s_c k = bch2_hash_lookup(trans, &dirent_iter, bch2_dirent_hash_desc,
-					     dir_hash_info, dir, name, 0);
-	int ret = bkey_err(k);
+					     dir_hash_info, dir, &lookup_name, 0);
+	ret = bkey_err(k);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -825,6 +849,9 @@ int __bch2_unlink(struct inode *vdir, struct dentry *dentry,
 		 */
 		set_nlink(&inode->v, 0);
 	}
+
+	if (IS_CASEFOLDED(vdir))
+		d_invalidate(dentry);
 err:
 	bch2_trans_put(trans);
 	bch2_unlock_inodes(INODE_UPDATE_LOCK, dir, inode);
@@ -1235,10 +1262,20 @@ static int bch2_tmpfile(struct mnt_idmap *idmap,
 	return finish_open_simple(file, 0);
 }
 
+struct bch_fiemap_extent {
+	struct bkey_buf	kbuf;
+	unsigned	flags;
+};
+
 static int bch2_fill_extent(struct bch_fs *c,
 			    struct fiemap_extent_info *info,
-			    struct bkey_s_c k, unsigned flags)
+			    struct bch_fiemap_extent *fe)
 {
+	struct bkey_s_c k = bkey_i_to_s_c(fe->kbuf.k);
+	unsigned flags = fe->flags;
+
+	BUG_ON(!k.k->size);
+
 	if (bkey_extent_is_direct_data(k.k)) {
 		struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
 		const union bch_extent_entry *entry;
@@ -1291,110 +1328,223 @@ static int bch2_fill_extent(struct bch_fs *c,
 	}
 }
 
-static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info,
-		       u64 start, u64 len)
+/*
+ * Scan a range of an inode for data in pagecache.
+ *
+ * Intended to be retryable, so don't modify the output params until success is
+ * imminent.
+ */
+static int
+bch2_fiemap_hole_pagecache(struct inode *vinode, u64 *start, u64 *end,
+			   bool nonblock)
 {
-	struct bch_fs *c = vinode->i_sb->s_fs_info;
-	struct bch_inode_info *ei = to_bch_ei(vinode);
-	struct btree_trans *trans;
-	struct btree_iter iter;
-	struct bkey_s_c k;
-	struct bkey_buf cur, prev;
-	bool have_extent = false;
-	int ret = 0;
+	loff_t	dstart, dend;
 
-	ret = fiemap_prep(&ei->v, info, start, &len, FIEMAP_FLAG_SYNC);
-	if (ret)
+	dstart = bch2_seek_pagecache_data(vinode, *start, *end, 0, nonblock);
+	if (dstart < 0)
+		return dstart;
+
+	if (dstart == *end) {
+		*start = dstart;
+		return 0;
+	}
+
+	dend = bch2_seek_pagecache_hole(vinode, dstart, *end, 0, nonblock);
+	if (dend < 0)
+		return dend;
+
+	/* race */
+	BUG_ON(dstart == dend);
+
+	*start = dstart;
+	*end = dend;
+	return 0;
+}
+
+/*
+ * Scan a range of pagecache that corresponds to a file mapping hole in the
+ * extent btree. If data is found, fake up an extent key so it looks like a
+ * delalloc extent to the rest of the fiemap processing code.
+ */
+static int
+bch2_next_fiemap_pagecache_extent(struct btree_trans *trans, struct bch_inode_info *inode,
+				  u64 start, u64 end, struct bch_fiemap_extent *cur)
+{
+	struct bch_fs		*c = trans->c;
+	struct bkey_i_extent	*delextent;
+	struct bch_extent_ptr	ptr = {};
+	loff_t			dstart = start << 9, dend = end << 9;
+	int			ret;
+
+	/*
+	 * We hold btree locks here so we cannot block on folio locks without
+	 * dropping trans locks first. Run a nonblocking scan for the common
+	 * case of no folios over holes and fall back on failure.
+	 *
+	 * Note that dropping locks like this is technically racy against
+	 * writeback inserting to the extent tree, but a non-sync fiemap scan is
+	 * fundamentally racy with writeback anyways. Therefore, just report the
+	 * range as delalloc regardless of whether we have to cycle trans locks.
+	 */
+	ret = bch2_fiemap_hole_pagecache(&inode->v, &dstart, &dend, true);
+	if (ret == -EAGAIN)
+		ret = drop_locks_do(trans,
+			bch2_fiemap_hole_pagecache(&inode->v, &dstart, &dend, false));
+	if (ret < 0)
 		return ret;
 
-	struct bpos end = POS(ei->v.i_ino, (start + len) >> 9);
-	if (start + len < start)
-		return -EINVAL;
+	/*
+	 * Create a fake extent key in the buffer. We have to add a dummy extent
+	 * pointer for the fill code to add an extent entry. It's explicitly
+	 * zeroed to reflect delayed allocation (i.e. phys offset 0).
+	 */
+	bch2_bkey_buf_realloc(&cur->kbuf, c, sizeof(*delextent) / sizeof(u64));
+	delextent = bkey_extent_init(cur->kbuf.k);
+	delextent->k.p = POS(inode->ei_inum.inum, dend >> 9);
+	delextent->k.size = (dend - dstart) >> 9;
+	bch2_bkey_append_ptr(&delextent->k_i, ptr);
 
-	start >>= 9;
+	cur->flags = FIEMAP_EXTENT_DELALLOC;
 
-	bch2_bkey_buf_init(&cur);
-	bch2_bkey_buf_init(&prev);
-	trans = bch2_trans_get(c);
+	return 0;
+}
+
+static int bch2_next_fiemap_extent(struct btree_trans *trans,
+				   struct bch_inode_info *inode,
+				   u64 start, u64 end,
+				   struct bch_fiemap_extent *cur)
+{
+	u32 snapshot;
+	int ret = bch2_subvolume_get_snapshot(trans, inode->ei_inum.subvol, &snapshot);
+	if (ret)
+		return ret;
 
+	struct btree_iter iter;
 	bch2_trans_iter_init(trans, &iter, BTREE_ID_extents,
-			     POS(ei->v.i_ino, start), 0);
+			     SPOS(inode->ei_inum.inum, start, snapshot), 0);
 
-	while (!ret || bch2_err_matches(ret, BCH_ERR_transaction_restart)) {
-		enum btree_id data_btree = BTREE_ID_extents;
+	struct bkey_s_c k =
+		bch2_btree_iter_peek_max(trans, &iter, POS(inode->ei_inum.inum, end));
+	ret = bkey_err(k);
+	if (ret)
+		goto err;
 
-		bch2_trans_begin(trans);
+	ret = bch2_next_fiemap_pagecache_extent(trans, inode, start, end, cur);
+	if (ret)
+		goto err;
 
-		u32 snapshot;
-		ret = bch2_subvolume_get_snapshot(trans, ei->ei_inum.subvol, &snapshot);
-		if (ret)
-			continue;
+	struct bpos pagecache_start = bkey_start_pos(&cur->kbuf.k->k);
 
-		bch2_btree_iter_set_snapshot(trans, &iter, snapshot);
+	/*
+	 * Does the pagecache or the btree take precedence?
+	 *
+	 * It _should_ be the pagecache, so that we correctly report delalloc
+	 * extents when dirty in the pagecache (we're COW, after all).
+	 *
+	 * But we'd have to add per-sector writeback tracking to
+	 * bch_folio_state, otherwise we report delalloc extents for clean
+	 * cached data in the pagecache.
+	 *
+	 * We should do this, but even then fiemap won't report stable mappings:
+	 * on bcachefs data moves around in the background (copygc, rebalance)
+	 * and we don't provide a way for userspace to lock that out.
+	 */
+	if (k.k &&
+	    bkey_le(bpos_max(iter.pos, bkey_start_pos(k.k)),
+		    pagecache_start)) {
+		bch2_bkey_buf_reassemble(&cur->kbuf, trans->c, k);
+		bch2_cut_front(iter.pos, cur->kbuf.k);
+		bch2_cut_back(POS(inode->ei_inum.inum, end), cur->kbuf.k);
+		cur->flags = 0;
+	} else if (k.k) {
+		bch2_cut_back(bkey_start_pos(k.k), cur->kbuf.k);
+	}
 
-		k = bch2_btree_iter_peek_max(trans, &iter, end);
-		ret = bkey_err(k);
+	if (cur->kbuf.k->k.type == KEY_TYPE_reflink_p) {
+		unsigned sectors = cur->kbuf.k->k.size;
+		s64 offset_into_extent = 0;
+		enum btree_id data_btree = BTREE_ID_extents;
+		ret = bch2_read_indirect_extent(trans, &data_btree, &offset_into_extent,
+						&cur->kbuf);
 		if (ret)
-			continue;
+			goto err;
 
-		if (!k.k)
-			break;
+		struct bkey_i *k = cur->kbuf.k;
+		sectors = min_t(unsigned, sectors, k->k.size - offset_into_extent);
 
-		if (!bkey_extent_is_data(k.k) &&
-		    k.k->type != KEY_TYPE_reservation) {
-			bch2_btree_iter_advance(trans, &iter);
-			continue;
-		}
+		bch2_cut_front(POS(k->k.p.inode,
+				   bkey_start_offset(&k->k) + offset_into_extent),
+			       k);
+		bch2_key_resize(&k->k, sectors);
+		k->k.p = iter.pos;
+		k->k.p.offset += k->k.size;
+	}
+err:
+	bch2_trans_iter_exit(trans, &iter);
+	return ret;
+}
+
+static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info,
+		       u64 start, u64 len)
+{
+	struct bch_fs *c = vinode->i_sb->s_fs_info;
+	struct bch_inode_info *ei = to_bch_ei(vinode);
+	struct btree_trans *trans;
+	struct bch_fiemap_extent cur, prev;
+	int ret = 0;
+
+	ret = fiemap_prep(&ei->v, info, start, &len, 0);
+	if (ret)
+		return ret;
+
+	if (start + len < start)
+		return -EINVAL;
+
+	start >>= 9;
+	u64 end = (start + len) >> 9;
 
-		s64 offset_into_extent	= iter.pos.offset - bkey_start_offset(k.k);
-		unsigned sectors	= k.k->size - offset_into_extent;
+	bch2_bkey_buf_init(&cur.kbuf);
+	bch2_bkey_buf_init(&prev.kbuf);
+	bkey_init(&prev.kbuf.k->k);
 
-		bch2_bkey_buf_reassemble(&cur, c, k);
+	trans = bch2_trans_get(c);
 
-		ret = bch2_read_indirect_extent(trans, &data_btree,
-					&offset_into_extent, &cur);
+	while (start < end) {
+		ret = lockrestart_do(trans,
+			bch2_next_fiemap_extent(trans, ei, start, end, &cur));
 		if (ret)
-			continue;
+			goto err;
 
-		k = bkey_i_to_s_c(cur.k);
-		bch2_bkey_buf_realloc(&prev, c, k.k->u64s);
+		BUG_ON(bkey_start_offset(&cur.kbuf.k->k) < start);
+		BUG_ON(cur.kbuf.k->k.p.offset > end);
 
-		sectors = min_t(unsigned, sectors, k.k->size - offset_into_extent);
+		if (bkey_start_offset(&cur.kbuf.k->k) == end)
+			break;
 
-		bch2_cut_front(POS(k.k->p.inode,
-				   bkey_start_offset(k.k) +
-				   offset_into_extent),
-			       cur.k);
-		bch2_key_resize(&cur.k->k, sectors);
-		cur.k->k.p = iter.pos;
-		cur.k->k.p.offset += cur.k->k.size;
+		start = cur.kbuf.k->k.p.offset;
 
-		if (have_extent) {
+		if (!bkey_deleted(&prev.kbuf.k->k)) {
 			bch2_trans_unlock(trans);
-			ret = bch2_fill_extent(c, info,
-					bkey_i_to_s_c(prev.k), 0);
+			ret = bch2_fill_extent(c, info, &prev);
 			if (ret)
-				break;
+				goto err;
 		}
 
-		bkey_copy(prev.k, cur.k);
-		have_extent = true;
-
-		bch2_btree_iter_set_pos(trans, &iter,
-			POS(iter.pos.inode, iter.pos.offset + sectors));
+		bch2_bkey_buf_copy(&prev.kbuf, c, cur.kbuf.k);
+		prev.flags = cur.flags;
 	}
-	bch2_trans_iter_exit(trans, &iter);
 
-	if (!ret && have_extent) {
+	if (!bkey_deleted(&prev.kbuf.k->k)) {
 		bch2_trans_unlock(trans);
-		ret = bch2_fill_extent(c, info, bkey_i_to_s_c(prev.k),
-				       FIEMAP_EXTENT_LAST);
+		prev.flags |= FIEMAP_EXTENT_LAST;
+		ret = bch2_fill_extent(c, info, &prev);
 	}
-
+err:
 	bch2_trans_put(trans);
-	bch2_bkey_buf_exit(&cur, c);
-	bch2_bkey_buf_exit(&prev, c);
-	return ret < 0 ? ret : 0;
+	bch2_bkey_buf_exit(&cur.kbuf, c);
+	bch2_bkey_buf_exit(&prev.kbuf, c);
+
+	return bch2_err_class(ret < 0 ? ret : 0);
 }
 
 static const struct vm_operations_struct bch_vm_ops = {
@@ -1449,6 +1599,165 @@ static int bch2_open(struct inode *vinode, struct file *file)
 	return generic_file_open(vinode, file);
 }
 
+/* bcachefs inode flags -> FS_IOC_GETFLAGS: */
+static const __maybe_unused unsigned bch_flags_to_uflags[] = {
+	[__BCH_INODE_sync]		= FS_SYNC_FL,
+	[__BCH_INODE_immutable]		= FS_IMMUTABLE_FL,
+	[__BCH_INODE_append]		= FS_APPEND_FL,
+	[__BCH_INODE_nodump]		= FS_NODUMP_FL,
+	[__BCH_INODE_noatime]		= FS_NOATIME_FL,
+};
+
+/* bcachefs inode flags -> FS_IOC_FSGETXATTR: */
+static const __maybe_unused unsigned bch_flags_to_xflags[] = {
+	[__BCH_INODE_sync]	= FS_XFLAG_SYNC,
+	[__BCH_INODE_immutable]	= FS_XFLAG_IMMUTABLE,
+	[__BCH_INODE_append]	= FS_XFLAG_APPEND,
+	[__BCH_INODE_nodump]	= FS_XFLAG_NODUMP,
+	[__BCH_INODE_noatime]	= FS_XFLAG_NOATIME,
+};
+
+static int bch2_fileattr_get(struct dentry *dentry,
+			     struct fileattr *fa)
+{
+	struct bch_inode_info *inode = to_bch_ei(d_inode(dentry));
+	struct bch_fs *c = inode->v.i_sb->s_fs_info;
+
+	fileattr_fill_xflags(fa, map_flags(bch_flags_to_xflags, inode->ei_inode.bi_flags));
+
+	if (inode->ei_inode.bi_fields_set & (1 << Inode_opt_project))
+		fa->fsx_xflags |= FS_XFLAG_PROJINHERIT;
+
+	if (bch2_inode_casefold(c, &inode->ei_inode))
+		fa->flags |= FS_CASEFOLD_FL;
+
+	fa->fsx_projid = inode->ei_qid.q[QTYP_PRJ];
+	return 0;
+}
+
+struct flags_set {
+	unsigned		mask;
+	unsigned		flags;
+	unsigned		projid;
+	bool			set_project;
+	bool			set_casefold;
+	bool			casefold;
+};
+
+static int fssetxattr_inode_update_fn(struct btree_trans *trans,
+				      struct bch_inode_info *inode,
+				      struct bch_inode_unpacked *bi,
+				      void *p)
+{
+	struct bch_fs *c = trans->c;
+	struct flags_set *s = p;
+
+	/*
+	 * We're relying on btree locking here for exclusion with other ioctl
+	 * calls - use the flags in the btree (@bi), not inode->i_flags:
+	 */
+	if (!S_ISREG(bi->bi_mode) &&
+	    !S_ISDIR(bi->bi_mode) &&
+	    (s->flags & (BCH_INODE_nodump|BCH_INODE_noatime)) != s->flags)
+		return -EINVAL;
+
+	if (s->casefold != bch2_inode_casefold(c, bi)) {
+#ifdef CONFIG_UNICODE
+		int ret = 0;
+		/* Not supported on individual files. */
+		if (!S_ISDIR(bi->bi_mode))
+			return -EOPNOTSUPP;
+
+		/*
+		 * Make sure the dir is empty, as otherwise we'd need to
+		 * rehash everything and update the dirent keys.
+		 */
+		ret = bch2_empty_dir_trans(trans, inode_inum(inode));
+		if (ret < 0)
+			return ret;
+
+		ret = bch2_request_incompat_feature(c, bcachefs_metadata_version_casefolding);
+		if (ret)
+			return ret;
+
+		bch2_check_set_feature(c, BCH_FEATURE_casefolding);
+
+		bi->bi_casefold = s->casefold + 1;
+		bi->bi_fields_set |= BIT(Inode_opt_casefold);
+
+#else
+		printk(KERN_ERR "Cannot use casefolding on a kernel without CONFIG_UNICODE\n");
+		return -EOPNOTSUPP;
+#endif
+	}
+
+	if (s->set_project) {
+		bi->bi_project = s->projid;
+		bi->bi_fields_set |= BIT(Inode_opt_project);
+	}
+
+	bi->bi_flags &= ~s->mask;
+	bi->bi_flags |= s->flags;
+
+	bi->bi_ctime = timespec_to_bch2_time(c, current_time(&inode->v));
+	return 0;
+}
+
+static int bch2_fileattr_set(struct mnt_idmap *idmap,
+			     struct dentry *dentry,
+			     struct fileattr *fa)
+{
+	struct bch_inode_info *inode = to_bch_ei(d_inode(dentry));
+	struct bch_fs *c = inode->v.i_sb->s_fs_info;
+	struct flags_set s = {};
+	int ret;
+
+	if (fa->fsx_valid) {
+		fa->fsx_xflags &= ~FS_XFLAG_PROJINHERIT;
+
+		s.mask = map_defined(bch_flags_to_xflags);
+		s.flags |= map_flags_rev(bch_flags_to_xflags, fa->fsx_xflags);
+		if (fa->fsx_xflags)
+			return -EOPNOTSUPP;
+
+		if (fa->fsx_projid >= U32_MAX)
+			return -EINVAL;
+
+		/*
+		 * inode fields accessible via the xattr interface are stored with a +1
+		 * bias, so that 0 means unset:
+		 */
+		if ((inode->ei_inode.bi_project ||
+		     fa->fsx_projid) &&
+		    inode->ei_inode.bi_project != fa->fsx_projid + 1) {
+			s.projid = fa->fsx_projid + 1;
+			s.set_project = true;
+		}
+	}
+
+	if (fa->flags_valid) {
+		s.mask = map_defined(bch_flags_to_uflags);
+
+		s.set_casefold = true;
+		s.casefold = (fa->flags & FS_CASEFOLD_FL) != 0;
+		fa->flags &= ~FS_CASEFOLD_FL;
+
+		s.flags |= map_flags_rev(bch_flags_to_uflags, fa->flags);
+		if (fa->flags)
+			return -EOPNOTSUPP;
+	}
+
+	mutex_lock(&inode->ei_update_lock);
+	ret   = bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?:
+		(s.set_project
+		 ? bch2_set_projid(c, inode, fa->fsx_projid)
+		 : 0) ?:
+		bch2_write_inode(c, inode, fssetxattr_inode_update_fn, &s,
+			       ATTR_CTIME);
+	mutex_unlock(&inode->ei_update_lock);
+	return ret;
+}
+
 static const struct file_operations bch_file_operations = {
 	.open		= bch2_open,
 	.llseek		= bch2_llseek,
@@ -1476,6 +1785,8 @@ static const struct inode_operations bch_file_inode_operations = {
 	.get_inode_acl	= bch2_get_acl,
 	.set_acl	= bch2_set_acl,
 #endif
+	.fileattr_get	= bch2_fileattr_get,
+	.fileattr_set	= bch2_fileattr_set,
 };
 
 static const struct inode_operations bch_dir_inode_operations = {
@@ -1496,6 +1807,8 @@ static const struct inode_operations bch_dir_inode_operations = {
 	.get_inode_acl	= bch2_get_acl,
 	.set_acl	= bch2_set_acl,
 #endif
+	.fileattr_get	= bch2_fileattr_get,
+	.fileattr_set	= bch2_fileattr_set,
 };
 
 static const struct file_operations bch_dir_file_operations = {
@@ -1518,6 +1831,8 @@ static const struct inode_operations bch_symlink_inode_operations = {
 	.get_inode_acl	= bch2_get_acl,
 	.set_acl	= bch2_set_acl,
 #endif
+	.fileattr_get	= bch2_fileattr_get,
+	.fileattr_set	= bch2_fileattr_set,
 };
 
 static const struct inode_operations bch_special_inode_operations = {
@@ -1528,6 +1843,8 @@ static const struct inode_operations bch_special_inode_operations = {
 	.get_inode_acl	= bch2_get_acl,
 	.set_acl	= bch2_set_acl,
 #endif
+	.fileattr_get	= bch2_fileattr_get,
+	.fileattr_set	= bch2_fileattr_set,
 };
 
 static const struct address_space_operations bch_address_space_operations = {
@@ -2254,6 +2571,11 @@ got_sb:
 	if (ret)
 		goto err_put_super;
 
+#ifdef CONFIG_UNICODE
+	sb->s_encoding = c->cf_encoding;
+#endif
+	generic_set_sb_d_ops(sb);
+
 	vinode = bch2_vfs_inode_get(c, BCACHEFS_ROOT_SUBVOL_INUM);
 	ret = PTR_ERR_OR_ZERO(vinode);
 	bch_err_msg(c, ret, "mounting: error getting root inode");
diff --git a/fs/bcachefs/fsck.c b/fs/bcachefs/fsck.c
index 18308f3d64a1..7b25cedd3e40 100644
--- a/fs/bcachefs/fsck.c
+++ b/fs/bcachefs/fsck.c
@@ -321,6 +321,31 @@ static inline bool inode_should_reattach(struct bch_inode_unpacked *inode)
 	    inode->bi_subvol == BCACHEFS_ROOT_SUBVOL)
 		return false;
 
+	/*
+	 * Subvolume roots are special: older versions of subvolume roots may be
+	 * disconnected, it's only the newest version that matters.
+	 *
+	 * We only keep a single dirent pointing to a subvolume root, i.e.
+	 * older versions of snapshots will not have a different dirent pointing
+	 * to the same subvolume root.
+	 *
+	 * This is because dirents that point to subvolumes are only visible in
+	 * the parent subvolume - versioning is not needed - and keeping them
+	 * around would break fsck, because when we're crossing subvolumes we
+	 * don't have a consistent snapshot ID to do check the inode <-> dirent
+	 * relationships.
+	 *
+	 * Thus, a subvolume root that's been renamed after a snapshot will have
+	 * a disconnected older version - that's expected.
+	 *
+	 * Note that taking a snapshot always updates the root inode (to update
+	 * the dirent backpointer), so a subvolume root inode with
+	 * BCH_INODE_has_child_snapshot is never visible.
+	 */
+	if (inode->bi_subvol &&
+	    (inode->bi_flags & BCH_INODE_has_child_snapshot))
+		return false;
+
 	return !inode->bi_dir && !(inode->bi_flags & BCH_INODE_unlinked);
 }
 
@@ -1007,6 +1032,23 @@ static int check_inode_dirent_inode(struct btree_trans *trans,
 	if (ret && !bch2_err_matches(ret, ENOENT))
 		return ret;
 
+	if ((ret || dirent_points_to_inode_nowarn(d, inode)) &&
+	    inode->bi_subvol &&
+	    (inode->bi_flags & BCH_INODE_has_child_snapshot)) {
+		/* Older version of a renamed subvolume root: we won't have a
+		 * correct dirent for it. That's expected, see
+		 * inode_should_reattach().
+		 *
+		 * We don't clear the backpointer field when doing the rename
+		 * because there might be arbitrarily many versions in older
+		 * snapshots.
+		 */
+		inode->bi_dir = 0;
+		inode->bi_dir_offset = 0;
+		*write_inode = true;
+		goto out;
+	}
+
 	if (fsck_err_on(ret,
 			trans, inode_points_to_missing_dirent,
 			"inode points to missing dirent\n%s",
@@ -1027,7 +1069,7 @@ static int check_inode_dirent_inode(struct btree_trans *trans,
 		inode->bi_dir_offset = 0;
 		*write_inode = true;
 	}
-
+out:
 	ret = 0;
 fsck_err:
 	bch2_trans_iter_exit(trans, &dirent_iter);
diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h
index f82cfbf460d0..c74af15b14f2 100644
--- a/fs/bcachefs/inode.h
+++ b/fs/bcachefs/inode.h
@@ -243,6 +243,14 @@ static inline unsigned bkey_inode_mode(struct bkey_s_c k)
 	}
 }
 
+static inline bool bch2_inode_casefold(struct bch_fs *c, const struct bch_inode_unpacked *bi)
+{
+	/* inode apts are stored with a +1 bias: 0 means "unset, use fs opt" */
+	return bi->bi_casefold
+		? bi->bi_casefold - 1
+		: c->opts.casefold;
+}
+
 /* i_nlink: */
 
 static inline unsigned nlink_bias(umode_t mode)
diff --git a/fs/bcachefs/inode_format.h b/fs/bcachefs/inode_format.h
index 117110af1e3f..87e193e8ed25 100644
--- a/fs/bcachefs/inode_format.h
+++ b/fs/bcachefs/inode_format.h
@@ -103,7 +103,8 @@ struct bch_inode_generation {
 	x(bi_parent_subvol,		32)	\
 	x(bi_nocow,			8)	\
 	x(bi_depth,			32)	\
-	x(bi_inodes_32bit,		8)
+	x(bi_inodes_32bit,		8)	\
+	x(bi_casefold,			8)
 
 /* subset of BCH_INODE_FIELDS */
 #define BCH_INODE_OPTS()			\
@@ -117,7 +118,8 @@ struct bch_inode_generation {
 	x(background_target,		16)	\
 	x(erasure_code,			16)	\
 	x(nocow,			8)	\
-	x(inodes_32bit,			8)
+	x(inodes_32bit,			8)	\
+	x(casefold,			8)
 
 enum inode_opt_id {
 #define x(name, ...)				\
@@ -137,8 +139,7 @@ enum inode_opt_id {
 	x(i_sectors_dirty,		6)	\
 	x(unlinked,			7)	\
 	x(backptr_untrusted,		8)	\
-	x(has_child_snapshot,		9)	\
-	x(casefolded,			10)
+	x(has_child_snapshot,		9)
 
 /* bits 20+ reserved for packed fields below: */
 
diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c
index 417bb0c7bbfa..def4a26a3b45 100644
--- a/fs/bcachefs/io_read.c
+++ b/fs/bcachefs/io_read.c
@@ -487,6 +487,8 @@ static void bch2_rbio_retry(struct work_struct *work)
 		.inum	= rbio->read_pos.inode,
 	};
 	struct bch_io_failures failed = { .nr = 0 };
+	int orig_error = rbio->ret;
+
 	struct btree_trans *trans = bch2_trans_get(c);
 
 	trace_io_read_retry(&rbio->bio);
@@ -519,7 +521,9 @@ static void bch2_rbio_retry(struct work_struct *work)
 	if (ret) {
 		rbio->ret = ret;
 		rbio->bio.bi_status = BLK_STS_IOERR;
-	} else {
+	} else if (orig_error != -BCH_ERR_data_read_retry_csum_err_maybe_userspace &&
+		   orig_error != -BCH_ERR_data_read_ptr_stale_race &&
+		   !failed.nr) {
 		struct printbuf buf = PRINTBUF;
 
 		lockrestart_do(trans,
@@ -977,7 +981,8 @@ retry_pick:
 		goto err;
 	}
 
-	if (unlikely(bch2_csum_type_is_encryption(pick.crc.csum_type)) && !c->chacha20) {
+	if (unlikely(bch2_csum_type_is_encryption(pick.crc.csum_type)) &&
+	    !c->chacha20_key_set) {
 		struct printbuf buf = PRINTBUF;
 		bch2_read_err_msg_trans(trans, &buf, orig, read_pos);
 		prt_printf(&buf, "attempting to read encrypted data without encryption key\n  ");
@@ -1344,14 +1349,16 @@ err:
 
 	bch2_trans_iter_exit(trans, &iter);
 
-	if (ret) {
-		struct printbuf buf = PRINTBUF;
-		lockrestart_do(trans,
-			bch2_inum_offset_err_msg_trans(trans, &buf, inum,
-						       bvec_iter.bi_sector << 9));
-		prt_printf(&buf, "read error: %s", bch2_err_str(ret));
-		bch_err_ratelimited(c, "%s", buf.buf);
-		printbuf_exit(&buf);
+	if (unlikely(ret)) {
+		if (ret != -BCH_ERR_extent_poisoned) {
+			struct printbuf buf = PRINTBUF;
+			lockrestart_do(trans,
+				       bch2_inum_offset_err_msg_trans(trans, &buf, inum,
+								      bvec_iter.bi_sector << 9));
+			prt_printf(&buf, "data read error: %s", bch2_err_str(ret));
+			bch_err_ratelimited(c, "%s", buf.buf);
+			printbuf_exit(&buf);
+		}
 
 		rbio->bio.bi_status	= BLK_STS_IOERR;
 		rbio->ret		= ret;
diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c
index a418fa62f09d..c1237da079ed 100644
--- a/fs/bcachefs/io_write.c
+++ b/fs/bcachefs/io_write.c
@@ -255,6 +255,27 @@ static inline int bch2_extent_update_i_size_sectors(struct btree_trans *trans,
 	}
 
 	if (i_sectors_delta) {
+		s64 bi_sectors = le64_to_cpu(inode->v.bi_sectors);
+		if (unlikely(bi_sectors + i_sectors_delta < 0)) {
+			struct bch_fs *c = trans->c;
+			struct printbuf buf = PRINTBUF;
+			bch2_log_msg_start(c, &buf);
+			prt_printf(&buf, "inode %llu i_sectors underflow: %lli + %lli < 0",
+				   extent_iter->pos.inode, bi_sectors, i_sectors_delta);
+
+			bool repeat = false, print = false, suppress = false;
+			bch2_count_fsck_err(c, inode_i_sectors_underflow, buf.buf,
+					    &repeat, &print, &suppress);
+			if (print)
+				bch2_print_str(c, buf.buf);
+			printbuf_exit(&buf);
+
+			if (i_sectors_delta < 0)
+				i_sectors_delta = -bi_sectors;
+			else
+				i_sectors_delta = 0;
+		}
+
 		le64_add_cpu(&inode->v.bi_sectors, i_sectors_delta);
 		inode_update_flags = 0;
 	}
diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c
index d8f74b6d0a75..bb45d3634194 100644
--- a/fs/bcachefs/journal.c
+++ b/fs/bcachefs/journal.c
@@ -281,7 +281,24 @@ static void __journal_entry_close(struct journal *j, unsigned closed_val, bool t
 
 	sectors = vstruct_blocks_plus(buf->data, c->block_bits,
 				      buf->u64s_reserved) << c->block_bits;
-	BUG_ON(sectors > buf->sectors);
+	if (unlikely(sectors > buf->sectors)) {
+		struct printbuf err = PRINTBUF;
+		err.atomic++;
+
+		prt_printf(&err, "journal entry overran reserved space: %u > %u\n",
+			   sectors, buf->sectors);
+		prt_printf(&err, "buf u64s %u u64s reserved %u cur_entry_u64s %u block_bits %u\n",
+			   le32_to_cpu(buf->data->u64s), buf->u64s_reserved,
+			   j->cur_entry_u64s,
+			   c->block_bits);
+		prt_printf(&err, "fatal error - emergency read only");
+		bch2_journal_halt_locked(j);
+
+		bch_err(c, "%s", err.buf);
+		printbuf_exit(&err);
+		return;
+	}
+
 	buf->sectors = sectors;
 
 	/*
@@ -1462,8 +1479,6 @@ int bch2_fs_journal_start(struct journal *j, u64 cur_seq)
 		j->last_empty_seq = cur_seq - 1; /* to match j->seq */
 
 	spin_lock(&j->lock);
-
-	set_bit(JOURNAL_running, &j->flags);
 	j->last_flush_write = jiffies;
 
 	j->reservations.idx = journal_cur_seq(j);
@@ -1474,6 +1489,21 @@ int bch2_fs_journal_start(struct journal *j, u64 cur_seq)
 	return 0;
 }
 
+void bch2_journal_set_replay_done(struct journal *j)
+{
+	/*
+	 * journal_space_available must happen before setting JOURNAL_running
+	 * JOURNAL_running must happen before JOURNAL_replay_done
+	 */
+	spin_lock(&j->lock);
+	bch2_journal_space_available(j);
+
+	set_bit(JOURNAL_need_flush_write, &j->flags);
+	set_bit(JOURNAL_running, &j->flags);
+	set_bit(JOURNAL_replay_done, &j->flags);
+	spin_unlock(&j->lock);
+}
+
 /* init/exit: */
 
 void bch2_dev_journal_exit(struct bch_dev *ca)
diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h
index 47828771f9c2..641e20c05a14 100644
--- a/fs/bcachefs/journal.h
+++ b/fs/bcachefs/journal.h
@@ -437,12 +437,6 @@ static inline int bch2_journal_error(struct journal *j)
 
 struct bch_dev;
 
-static inline void bch2_journal_set_replay_done(struct journal *j)
-{
-	BUG_ON(!test_bit(JOURNAL_running, &j->flags));
-	set_bit(JOURNAL_replay_done, &j->flags);
-}
-
 void bch2_journal_unblock(struct journal *);
 void bch2_journal_block(struct journal *);
 struct journal_buf *bch2_next_write_buffer_flush_journal_buf(struct journal *, u64, bool *);
@@ -459,6 +453,7 @@ void bch2_dev_journal_stop(struct journal *, struct bch_dev *);
 
 void bch2_fs_journal_stop(struct journal *);
 int bch2_fs_journal_start(struct journal *, u64);
+void bch2_journal_set_replay_done(struct journal *);
 
 void bch2_dev_journal_exit(struct bch_dev *);
 int bch2_dev_journal_init(struct bch_dev *, struct bch_sb *);
diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c
index 1b7961f4f609..63cdf885c9e2 100644
--- a/fs/bcachefs/journal_io.c
+++ b/fs/bcachefs/journal_io.c
@@ -1460,7 +1460,7 @@ fsck_err:
 
 static void journal_advance_devs_to_next_bucket(struct journal *j,
 						struct dev_alloc_list *devs,
-						unsigned sectors, u64 seq)
+						unsigned sectors, __le64 seq)
 {
 	struct bch_fs *c = container_of(j, struct bch_fs, journal);
 
@@ -1782,7 +1782,7 @@ static CLOSURE_CALLBACK(journal_write_submit)
 		struct bch_dev *ca = bch2_dev_get_ioref(c, ptr->dev, WRITE);
 		if (!ca) {
 			/* XXX: fix this */
-			bch_err(c, "missing device for journal write\n");
+			bch_err(c, "missing device %u for journal write", ptr->dev);
 			continue;
 		}
 
diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c
index 5d1547aa118a..ea670c3c43d8 100644
--- a/fs/bcachefs/journal_reclaim.c
+++ b/fs/bcachefs/journal_reclaim.c
@@ -252,7 +252,10 @@ void bch2_journal_space_available(struct journal *j)
 
 	bch2_journal_set_watermark(j);
 out:
-	j->cur_entry_sectors	= !ret ? j->space[journal_space_discarded].next_entry : 0;
+	j->cur_entry_sectors	= !ret
+		? round_down(j->space[journal_space_discarded].next_entry,
+			     block_sectors(c))
+		: 0;
 	j->cur_entry_error	= ret;
 
 	if (!ret)
diff --git a/fs/bcachefs/movinggc.c b/fs/bcachefs/movinggc.c
index 159410c50861..96873372b516 100644
--- a/fs/bcachefs/movinggc.c
+++ b/fs/bcachefs/movinggc.c
@@ -356,6 +356,13 @@ static int bch2_copygc_thread(void *arg)
 
 	set_freezable();
 
+	/*
+	 * Data move operations can't run until after check_snapshots has
+	 * completed, and bch2_snapshot_is_ancestor() is available.
+	 */
+	kthread_wait_freezable(c->recovery_pass_done > BCH_RECOVERY_PASS_check_snapshots ||
+			       kthread_should_stop());
+
 	bch2_move_stats_init(&move_stats, "copygc");
 	bch2_moving_ctxt_init(&ctxt, c, NULL, &move_stats,
 			      writepoint_ptr(&c->copygc_write_point),
diff --git a/fs/bcachefs/movinggc.h b/fs/bcachefs/movinggc.h
index ea181fef5bc9..d1885cf67a45 100644
--- a/fs/bcachefs/movinggc.h
+++ b/fs/bcachefs/movinggc.h
@@ -5,6 +5,15 @@
 unsigned long bch2_copygc_wait_amount(struct bch_fs *);
 void bch2_copygc_wait_to_text(struct printbuf *, struct bch_fs *);
 
+static inline void bch2_copygc_wakeup(struct bch_fs *c)
+{
+	rcu_read_lock();
+	struct task_struct *p = rcu_dereference(c->copygc_thread);
+	if (p)
+		wake_up_process(p);
+	rcu_read_unlock();
+}
+
 void bch2_copygc_stop(struct bch_fs *);
 int bch2_copygc_start(struct bch_fs *);
 void bch2_fs_copygc_init(struct bch_fs *);
diff --git a/fs/bcachefs/namei.c b/fs/bcachefs/namei.c
index 0d65ea96f7a2..52c58c6d53d2 100644
--- a/fs/bcachefs/namei.c
+++ b/fs/bcachefs/namei.c
@@ -47,10 +47,6 @@ int bch2_create_trans(struct btree_trans *trans,
 	if (ret)
 		goto err;
 
-	/* Inherit casefold state from parent. */
-	if (S_ISDIR(mode))
-		new_inode->bi_flags |= dir_u->bi_flags & BCH_INODE_casefolded;
-
 	if (!(flags & BCH_CREATE_SNAPSHOT)) {
 		/* Normal create path - allocate a new inode: */
 		bch2_inode_init_late(new_inode, now, uid, gid, mode, rdev, dir_u);
@@ -347,6 +343,9 @@ bool bch2_reinherit_attrs(struct bch_inode_unpacked *dst_u,
 	bool ret = false;
 
 	for (id = 0; id < Inode_opt_nr; id++) {
+		if (!S_ISDIR(dst_u->bi_mode) && id == Inode_opt_casefold)
+			continue;
+
 		/* Skip attributes that were explicitly set on this inode */
 		if (dst_u->bi_fields_set & (1 << id))
 			continue;
diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h
index 4d06313076ff..dfb14810124c 100644
--- a/fs/bcachefs/opts.h
+++ b/fs/bcachefs/opts.h
@@ -228,6 +228,11 @@ enum fsck_err_opts {
 	  OPT_BOOL(),							\
 	  BCH_SB_ERASURE_CODE,		false,				\
 	  NULL,		"Enable erasure coding (DO NOT USE YET)")	\
+	x(casefold,			u8,				\
+	  OPT_FS|OPT_INODE|OPT_FORMAT,					\
+	  OPT_BOOL(),							\
+	  BCH_SB_CASEFOLD,		false,				\
+	  NULL,		"Dirent lookups are casefolded")		\
 	x(inodes_32bit,			u8,				\
 	  OPT_FS|OPT_INODE|OPT_FORMAT|OPT_MOUNT|OPT_RUNTIME,		\
 	  OPT_BOOL(),							\
diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c
index c63fa53f30d2..4ccdfc1f34aa 100644
--- a/fs/bcachefs/rebalance.c
+++ b/fs/bcachefs/rebalance.c
@@ -262,7 +262,7 @@ int bch2_set_rebalance_needs_scan(struct bch_fs *c, u64 inum)
 	int ret = bch2_trans_commit_do(c, NULL, NULL,
 				       BCH_TRANS_COMMIT_no_enospc,
 			    bch2_set_rebalance_needs_scan_trans(trans, inum));
-	rebalance_wakeup(c);
+	bch2_rebalance_wakeup(c);
 	return ret;
 }
 
@@ -581,6 +581,13 @@ static int bch2_rebalance_thread(void *arg)
 
 	set_freezable();
 
+	/*
+	 * Data move operations can't run until after check_snapshots has
+	 * completed, and bch2_snapshot_is_ancestor() is available.
+	 */
+	kthread_wait_freezable(c->recovery_pass_done > BCH_RECOVERY_PASS_check_snapshots ||
+			       kthread_should_stop());
+
 	bch2_moving_ctxt_init(&ctxt, c, NULL, &r->work_stats,
 			      writepoint_ptr(&c->rebalance_write_point),
 			      true);
@@ -664,7 +671,7 @@ void bch2_rebalance_stop(struct bch_fs *c)
 	c->rebalance.thread = NULL;
 
 	if (p) {
-		/* for sychronizing with rebalance_wakeup() */
+		/* for sychronizing with bch2_rebalance_wakeup() */
 		synchronize_rcu();
 
 		kthread_stop(p);
diff --git a/fs/bcachefs/rebalance.h b/fs/bcachefs/rebalance.h
index 62a3859d3823..e5e8eb4a2dd1 100644
--- a/fs/bcachefs/rebalance.h
+++ b/fs/bcachefs/rebalance.h
@@ -37,7 +37,7 @@ int bch2_set_rebalance_needs_scan_trans(struct btree_trans *, u64);
 int bch2_set_rebalance_needs_scan(struct bch_fs *, u64 inum);
 int bch2_set_fs_needs_rebalance(struct bch_fs *);
 
-static inline void rebalance_wakeup(struct bch_fs *c)
+static inline void bch2_rebalance_wakeup(struct bch_fs *c)
 {
 	struct task_struct *p;
 
diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c
index 79fd18a5a07c..d6c4ef819d40 100644
--- a/fs/bcachefs/recovery.c
+++ b/fs/bcachefs/recovery.c
@@ -18,6 +18,7 @@
 #include "journal_seq_blacklist.h"
 #include "logged_ops.h"
 #include "move.h"
+#include "movinggc.h"
 #include "namei.h"
 #include "quota.h"
 #include "rebalance.h"
@@ -389,9 +390,9 @@ int bch2_journal_replay(struct bch_fs *c)
 	 * Now, replay any remaining keys in the order in which they appear in
 	 * the journal, unpinning those journal entries as we go:
 	 */
-	sort(keys_sorted.data, keys_sorted.nr,
-	     sizeof(keys_sorted.data[0]),
-	     journal_sort_seq_cmp, NULL);
+	sort_nonatomic(keys_sorted.data, keys_sorted.nr,
+		       sizeof(keys_sorted.data[0]),
+		       journal_sort_seq_cmp, NULL);
 
 	darray_for_each(keys_sorted, kp) {
 		cond_resched();
@@ -1125,14 +1126,17 @@ int bch2_fs_initialize(struct bch_fs *c)
 	 * journal_res_get() will crash if called before this has
 	 * set up the journal.pin FIFO and journal.cur pointer:
 	 */
-	bch2_fs_journal_start(&c->journal, 1);
-	set_bit(BCH_FS_accounting_replay_done, &c->flags);
-	bch2_journal_set_replay_done(&c->journal);
+	ret = bch2_fs_journal_start(&c->journal, 1);
+	if (ret)
+		goto err;
 
 	ret = bch2_fs_read_write_early(c);
 	if (ret)
 		goto err;
 
+	set_bit(BCH_FS_accounting_replay_done, &c->flags);
+	bch2_journal_set_replay_done(&c->journal);
+
 	for_each_member_device(c, ca) {
 		ret = bch2_dev_usage_init(ca, false);
 		if (ret) {
@@ -1191,6 +1195,9 @@ int bch2_fs_initialize(struct bch_fs *c)
 
 	c->recovery_pass_done = BCH_RECOVERY_PASS_NR - 1;
 
+	bch2_copygc_wakeup(c);
+	bch2_rebalance_wakeup(c);
+
 	if (enabled_qtypes(c)) {
 		ret = bch2_fs_quota_read(c);
 		if (ret)
diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c
index 593ff142530d..22f72bb5b853 100644
--- a/fs/bcachefs/recovery_passes.c
+++ b/fs/bcachefs/recovery_passes.c
@@ -12,6 +12,7 @@
 #include "journal.h"
 #include "lru.h"
 #include "logged_ops.h"
+#include "movinggc.h"
 #include "rebalance.h"
 #include "recovery.h"
 #include "recovery_passes.h"
@@ -262,49 +263,52 @@ int bch2_run_recovery_passes(struct bch_fs *c)
 	 */
 	c->opts.recovery_passes_exclude &= ~BCH_RECOVERY_PASS_set_may_go_rw;
 
-	while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns) && !ret) {
-		c->next_recovery_pass = c->curr_recovery_pass + 1;
+	spin_lock_irq(&c->recovery_pass_lock);
 
-		spin_lock_irq(&c->recovery_pass_lock);
+	while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns) && !ret) {
+		unsigned prev_done = c->recovery_pass_done;
 		unsigned pass = c->curr_recovery_pass;
 
+		c->next_recovery_pass = pass + 1;
+
 		if (c->opts.recovery_pass_last &&
-		    c->curr_recovery_pass > c->opts.recovery_pass_last) {
-			spin_unlock_irq(&c->recovery_pass_lock);
+		    c->curr_recovery_pass > c->opts.recovery_pass_last)
 			break;
-		}
 
-		if (!should_run_recovery_pass(c, pass)) {
-			c->curr_recovery_pass++;
-			c->recovery_pass_done = max(c->recovery_pass_done, pass);
+		if (should_run_recovery_pass(c, pass)) {
 			spin_unlock_irq(&c->recovery_pass_lock);
-			continue;
-		}
-		spin_unlock_irq(&c->recovery_pass_lock);
-
-		ret =   bch2_run_recovery_pass(c, pass) ?:
-			bch2_journal_flush(&c->journal);
-
-		if (!ret && !test_bit(BCH_FS_error, &c->flags))
-			bch2_clear_recovery_pass_required(c, pass);
-
-		spin_lock_irq(&c->recovery_pass_lock);
-		if (c->next_recovery_pass < c->curr_recovery_pass) {
-			/*
-			 * bch2_run_explicit_recovery_pass() was called: we
-			 * can't always catch -BCH_ERR_restart_recovery because
-			 * it may have been called from another thread (btree
-			 * node read completion)
-			 */
-			ret = 0;
-			c->recovery_passes_complete &= ~(~0ULL << c->curr_recovery_pass);
-		} else {
-			c->recovery_passes_complete |= BIT_ULL(pass);
-			c->recovery_pass_done = max(c->recovery_pass_done, pass);
+			ret =   bch2_run_recovery_pass(c, pass) ?:
+				bch2_journal_flush(&c->journal);
+
+			if (!ret && !test_bit(BCH_FS_error, &c->flags))
+				bch2_clear_recovery_pass_required(c, pass);
+			spin_lock_irq(&c->recovery_pass_lock);
+
+			if (c->next_recovery_pass < c->curr_recovery_pass) {
+				/*
+				 * bch2_run_explicit_recovery_pass() was called: we
+				 * can't always catch -BCH_ERR_restart_recovery because
+				 * it may have been called from another thread (btree
+				 * node read completion)
+				 */
+				ret = 0;
+				c->recovery_passes_complete &= ~(~0ULL << c->curr_recovery_pass);
+			} else {
+				c->recovery_passes_complete |= BIT_ULL(pass);
+				c->recovery_pass_done = max(c->recovery_pass_done, pass);
+			}
 		}
+
 		c->curr_recovery_pass = c->next_recovery_pass;
-		spin_unlock_irq(&c->recovery_pass_lock);
+
+		if (prev_done <= BCH_RECOVERY_PASS_check_snapshots &&
+		    c->recovery_pass_done > BCH_RECOVERY_PASS_check_snapshots) {
+			bch2_copygc_wakeup(c);
+			bch2_rebalance_wakeup(c);
+		}
 	}
 
+	spin_unlock_irq(&c->recovery_pass_lock);
+
 	return ret;
 }
diff --git a/fs/bcachefs/sb-downgrade.c b/fs/bcachefs/sb-downgrade.c
index acb5d845841e..badd0e17ada5 100644
--- a/fs/bcachefs/sb-downgrade.c
+++ b/fs/bcachefs/sb-downgrade.c
@@ -20,6 +20,10 @@
  * x(version, recovery_passes, errors...)
  */
 #define UPGRADE_TABLE()						\
+	x(snapshot_2,						\
+	  RECOVERY_PASS_ALL_FSCK,				\
+	  BCH_FSCK_ERR_subvol_root_wrong_bi_subvol,		\
+	  BCH_FSCK_ERR_subvol_not_master_and_not_snapshot)	\
 	x(backpointers,						\
 	  RECOVERY_PASS_ALL_FSCK)				\
 	x(inode_v3,						\
diff --git a/fs/bcachefs/sb-errors_format.h b/fs/bcachefs/sb-errors_format.h
index 5d43e3504386..3b69a924086f 100644
--- a/fs/bcachefs/sb-errors_format.h
+++ b/fs/bcachefs/sb-errors_format.h
@@ -46,7 +46,7 @@ enum bch_fsck_flags {
 	x(btree_node_unsupported_version,			 34,	0)		\
 	x(btree_node_bset_older_than_sb_min,			 35,	0)		\
 	x(btree_node_bset_newer_than_sb,			 36,	0)		\
-	x(btree_node_data_missing,				 37,	0)		\
+	x(btree_node_data_missing,				 37,	FSCK_AUTOFIX)	\
 	x(btree_node_bset_after_end,				 38,	0)		\
 	x(btree_node_replicas_sectors_written_mismatch,		 39,	0)		\
 	x(btree_node_replicas_data_mismatch,			 40,	0)		\
@@ -205,9 +205,9 @@ enum bch_fsck_flags {
 	x(snapshot_bad_depth,					184,	0)		\
 	x(snapshot_bad_skiplist,				185,	0)		\
 	x(subvol_pos_bad,					186,	0)		\
-	x(subvol_not_master_and_not_snapshot,			187,	0)		\
+	x(subvol_not_master_and_not_snapshot,			187,	FSCK_AUTOFIX)	\
 	x(subvol_to_missing_root,				188,	0)		\
-	x(subvol_root_wrong_bi_subvol,				189,	0)		\
+	x(subvol_root_wrong_bi_subvol,				189,	FSCK_AUTOFIX)	\
 	x(bkey_in_missing_snapshot,				190,	0)		\
 	x(inode_pos_inode_nonzero,				191,	0)		\
 	x(inode_pos_blockdev_range,				192,	0)		\
@@ -236,6 +236,9 @@ enum bch_fsck_flags {
 	x(inode_has_child_snapshots_wrong,			287,	0)		\
 	x(inode_unreachable,					210,	FSCK_AUTOFIX)	\
 	x(inode_journal_seq_in_future,				299,	FSCK_AUTOFIX)	\
+	x(inode_i_sectors_underflow,				312,	FSCK_AUTOFIX)	\
+	x(vfs_inode_i_blocks_underflow,				311,	FSCK_AUTOFIX)	\
+	x(vfs_inode_i_blocks_not_zero_at_truncate,		313,	FSCK_AUTOFIX)	\
 	x(deleted_inode_but_clean,				211,	FSCK_AUTOFIX)	\
 	x(deleted_inode_missing,				212,	FSCK_AUTOFIX)	\
 	x(deleted_inode_is_dir,					213,	FSCK_AUTOFIX)	\
@@ -290,8 +293,8 @@ enum bch_fsck_flags {
 	x(btree_node_bkey_bad_u64s,				260,	0)		\
 	x(btree_node_topology_empty_interior_node,		261,	0)		\
 	x(btree_ptr_v2_min_key_bad,				262,	0)		\
-	x(btree_root_unreadable_and_scan_found_nothing,		263,	0)		\
-	x(snapshot_node_missing,				264,	0)		\
+	x(btree_root_unreadable_and_scan_found_nothing,		263,	FSCK_AUTOFIX)	\
+	x(snapshot_node_missing,				264,	FSCK_AUTOFIX)	\
 	x(dup_backpointer_to_bad_csum_extent,			265,	0)		\
 	x(btree_bitmap_not_marked,				266,	FSCK_AUTOFIX)	\
 	x(sb_clean_entry_overrun,				267,	0)		\
@@ -317,7 +320,9 @@ enum bch_fsck_flags {
 	x(directory_size_mismatch,				303,	FSCK_AUTOFIX)	\
 	x(dirent_cf_name_too_big,				304,	0)		\
 	x(dirent_stray_data_after_cf_name,			305,	0)		\
-	x(MAX,							308,	0)
+	x(rebalance_work_incorrectly_set,			309,	FSCK_AUTOFIX)	\
+	x(rebalance_work_incorrectly_unset,			310,	FSCK_AUTOFIX)	\
+	x(MAX,							314,	0)
 
 enum bch_sb_error_id {
 #define x(t, n, ...) BCH_FSCK_ERR_##t = n,
diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c
index 116131f95815..72779912939b 100644
--- a/fs/bcachefs/sb-members.c
+++ b/fs/bcachefs/sb-members.c
@@ -15,9 +15,11 @@ void bch2_dev_missing(struct bch_fs *c, unsigned dev)
 		bch2_fs_inconsistent(c, "pointer to nonexistent device %u", dev);
 }
 
-void bch2_dev_bucket_missing(struct bch_fs *c, struct bpos bucket)
+void bch2_dev_bucket_missing(struct bch_dev *ca, u64 bucket)
 {
-	bch2_fs_inconsistent(c, "pointer to nonexistent bucket %llu:%llu", bucket.inode, bucket.offset);
+	bch2_fs_inconsistent(ca->fs,
+		"pointer to nonexistent bucket %llu on device %s (valid range %u-%llu)",
+		bucket, ca->name, ca->mi.first_bucket, ca->mi.nbuckets);
 }
 
 #define x(t, n, ...) [n] = #t,
diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h
index 06bb41a3f360..42786657522c 100644
--- a/fs/bcachefs/sb-members.h
+++ b/fs/bcachefs/sb-members.h
@@ -249,20 +249,23 @@ static inline struct bch_dev *bch2_dev_tryget(struct bch_fs *c, unsigned dev)
 static inline struct bch_dev *bch2_dev_bucket_tryget_noerror(struct bch_fs *c, struct bpos bucket)
 {
 	struct bch_dev *ca = bch2_dev_tryget_noerror(c, bucket.inode);
-	if (ca && !bucket_valid(ca, bucket.offset)) {
+	if (ca && unlikely(!bucket_valid(ca, bucket.offset))) {
 		bch2_dev_put(ca);
 		ca = NULL;
 	}
 	return ca;
 }
 
-void bch2_dev_bucket_missing(struct bch_fs *, struct bpos);
+void bch2_dev_bucket_missing(struct bch_dev *, u64);
 
 static inline struct bch_dev *bch2_dev_bucket_tryget(struct bch_fs *c, struct bpos bucket)
 {
-	struct bch_dev *ca = bch2_dev_bucket_tryget_noerror(c, bucket);
-	if (!ca)
-		bch2_dev_bucket_missing(c, bucket);
+	struct bch_dev *ca = bch2_dev_tryget(c, bucket.inode);
+	if (ca && unlikely(!bucket_valid(ca, bucket.offset))) {
+		bch2_dev_bucket_missing(ca, bucket.offset);
+		bch2_dev_put(ca);
+		ca = NULL;
+	}
 	return ca;
 }
 
diff --git a/fs/bcachefs/snapshot.c b/fs/bcachefs/snapshot.c
index b7de29aed839..fec569c7deb1 100644
--- a/fs/bcachefs/snapshot.c
+++ b/fs/bcachefs/snapshot.c
@@ -396,7 +396,7 @@ u32 bch2_snapshot_tree_oldest_subvol(struct bch_fs *c, u32 snapshot_root)
 	u32 subvol = 0, s;
 
 	rcu_read_lock();
-	while (id) {
+	while (id && bch2_snapshot_exists(c, id)) {
 		s = snapshot_t(c, id)->subvol;
 
 		if (s && (!subvol || s < subvol))
diff --git a/fs/bcachefs/str_hash.h b/fs/bcachefs/str_hash.h
index 09a354a26c3b..0c1a00539bd1 100644
--- a/fs/bcachefs/str_hash.h
+++ b/fs/bcachefs/str_hash.h
@@ -33,7 +33,7 @@ bch2_str_hash_opt_to_type(struct bch_fs *c, enum bch_str_hash_opts opt)
 
 struct bch_hash_info {
 	u8			type;
-	struct unicode_map 	*cf_encoding;
+	struct unicode_map	*cf_encoding;
 	/*
 	 * For crc32 or crc64 string hashes the first key value of
 	 * the siphash_key (k0) is used as the key.
@@ -44,11 +44,10 @@ struct bch_hash_info {
 static inline struct bch_hash_info
 bch2_hash_info_init(struct bch_fs *c, const struct bch_inode_unpacked *bi)
 {
-	/* XXX ick */
 	struct bch_hash_info info = {
 		.type = INODE_STR_HASH(bi),
 #ifdef CONFIG_UNICODE
-		.cf_encoding = !!(bi->bi_flags & BCH_INODE_casefolded) ? c->cf_encoding : NULL,
+		.cf_encoding = bch2_inode_casefold(c, bi) ? c->cf_encoding : NULL,
 #endif
 		.siphash_key = { .k0 = bi->bi_hash_seed }
 	};
diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c
index 5537283d0bea..d0209f7658bb 100644
--- a/fs/bcachefs/subvolume.c
+++ b/fs/bcachefs/subvolume.c
@@ -6,6 +6,7 @@
 #include "errcode.h"
 #include "error.h"
 #include "fs.h"
+#include "recovery_passes.h"
 #include "snapshot.h"
 #include "subvolume.h"
 
@@ -44,8 +45,8 @@ static int check_subvol(struct btree_trans *trans,
 	ret = bch2_snapshot_lookup(trans, snapid, &snapshot);
 
 	if (bch2_err_matches(ret, ENOENT))
-		bch_err(c, "subvolume %llu points to nonexistent snapshot %u",
-			k.k->p.offset, snapid);
+		return bch2_run_explicit_recovery_pass(c,
+					BCH_RECOVERY_PASS_reconstruct_snapshots) ?: ret;
 	if (ret)
 		return ret;
 
diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c
index e27422b6d9c6..cb5d960aed92 100644
--- a/fs/bcachefs/super-io.c
+++ b/fs/bcachefs/super-io.c
@@ -73,14 +73,30 @@ int bch2_set_version_incompat(struct bch_fs *c, enum bcachefs_metadata_version v
 		? 0
 		: -BCH_ERR_may_not_use_incompat_feature;
 
+	mutex_lock(&c->sb_lock);
 	if (!ret) {
-		mutex_lock(&c->sb_lock);
 		SET_BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb,
 			max(BCH_SB_VERSION_INCOMPAT(c->disk_sb.sb), version));
 		bch2_write_super(c);
-		mutex_unlock(&c->sb_lock);
+	} else {
+		darray_for_each(c->incompat_versions_requested, i)
+			if (version == *i)
+				goto out;
+
+		darray_push(&c->incompat_versions_requested, version);
+		struct printbuf buf = PRINTBUF;
+		prt_str(&buf, "requested incompat feature ");
+		bch2_version_to_text(&buf, version);
+		prt_str(&buf, " currently not enabled");
+		prt_printf(&buf, "\n  set version_upgrade=incompat to enable");
+
+		bch_notice(c, "%s", buf.buf);
+		printbuf_exit(&buf);
 	}
 
+out:
+	mutex_unlock(&c->sb_lock);
+
 	return ret;
 }
 
@@ -1086,7 +1102,8 @@ int bch2_write_super(struct bch_fs *c)
 		prt_str(&buf, ")");
 		bch2_fs_fatal_error(c, ": %s", buf.buf);
 		printbuf_exit(&buf);
-		return -BCH_ERR_sb_not_downgraded;
+		ret = -BCH_ERR_sb_not_downgraded;
+		goto out;
 	}
 
 	darray_for_each(online_devices, ca) {
diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c
index a58edde43bee..27943082c093 100644
--- a/fs/bcachefs/super.c
+++ b/fs/bcachefs/super.c
@@ -70,14 +70,10 @@
 #include <linux/percpu.h>
 #include <linux/random.h>
 #include <linux/sysfs.h>
-#include <crypto/hash.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Kent Overstreet <kent.overstreet@gmail.com>");
 MODULE_DESCRIPTION("bcachefs filesystem");
-MODULE_SOFTDEP("pre: chacha20");
-MODULE_SOFTDEP("pre: poly1305");
-MODULE_SOFTDEP("pre: xxhash");
 
 const char * const bch2_fs_flag_strs[] = {
 #define x(n)		#n,
@@ -422,32 +418,6 @@ bool bch2_fs_emergency_read_only_locked(struct bch_fs *c)
 	return ret;
 }
 
-static int bch2_fs_read_write_late(struct bch_fs *c)
-{
-	int ret;
-
-	/*
-	 * Data move operations can't run until after check_snapshots has
-	 * completed, and bch2_snapshot_is_ancestor() is available.
-	 *
-	 * Ideally we'd start copygc/rebalance earlier instead of waiting for
-	 * all of recovery/fsck to complete:
-	 */
-	ret = bch2_copygc_start(c);
-	if (ret) {
-		bch_err(c, "error starting copygc thread");
-		return ret;
-	}
-
-	ret = bch2_rebalance_start(c);
-	if (ret) {
-		bch_err(c, "error starting rebalance thread");
-		return ret;
-	}
-
-	return 0;
-}
-
 static int __bch2_fs_read_write(struct bch_fs *c, bool early)
 {
 	int ret;
@@ -470,29 +440,28 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early)
 
 	clear_bit(BCH_FS_clean_shutdown, &c->flags);
 
+	__for_each_online_member(c, ca, BIT(BCH_MEMBER_STATE_rw), READ) {
+		bch2_dev_allocator_add(c, ca);
+		percpu_ref_reinit(&ca->io_ref[WRITE]);
+	}
+	bch2_recalc_capacity(c);
+
 	/*
 	 * First journal write must be a flush write: after a clean shutdown we
 	 * don't read the journal, so the first journal write may end up
 	 * overwriting whatever was there previously, and there must always be
 	 * at least one non-flush write in the journal or recovery will fail:
 	 */
+	spin_lock(&c->journal.lock);
 	set_bit(JOURNAL_need_flush_write, &c->journal.flags);
 	set_bit(JOURNAL_running, &c->journal.flags);
-
-	__for_each_online_member(c, ca, BIT(BCH_MEMBER_STATE_rw), READ) {
-		bch2_dev_allocator_add(c, ca);
-		percpu_ref_reinit(&ca->io_ref[WRITE]);
-	}
-	bch2_recalc_capacity(c);
+	bch2_journal_space_available(&c->journal);
+	spin_unlock(&c->journal.lock);
 
 	ret = bch2_fs_mark_dirty(c);
 	if (ret)
 		goto err;
 
-	spin_lock(&c->journal.lock);
-	bch2_journal_space_available(&c->journal);
-	spin_unlock(&c->journal.lock);
-
 	ret = bch2_journal_reclaim_start(&c->journal);
 	if (ret)
 		goto err;
@@ -508,10 +477,17 @@ static int __bch2_fs_read_write(struct bch_fs *c, bool early)
 		atomic_long_inc(&c->writes[i]);
 	}
 #endif
-	if (!early) {
-		ret = bch2_fs_read_write_late(c);
-		if (ret)
-			goto err;
+
+	ret = bch2_copygc_start(c);
+	if (ret) {
+		bch_err_msg(c, ret, "error starting copygc thread");
+		goto err;
+	}
+
+	ret = bch2_rebalance_start(c);
+	if (ret) {
+		bch_err_msg(c, ret, "error starting rebalance thread");
+		goto err;
 	}
 
 	bch2_do_discards(c);
@@ -555,8 +531,13 @@ static void __bch2_fs_free(struct bch_fs *c)
 	for (unsigned i = 0; i < BCH_TIME_STAT_NR; i++)
 		bch2_time_stats_exit(&c->times[i]);
 
+#ifdef CONFIG_UNICODE
+	utf8_unload(c->cf_encoding);
+#endif
+
 	bch2_find_btree_nodes_exit(&c->found_btree_nodes);
 	bch2_free_pending_node_rewrites(c);
+	bch2_free_fsck_errs(c);
 	bch2_fs_accounting_exit(c);
 	bch2_fs_sb_errors_exit(c);
 	bch2_fs_counters_exit(c);
@@ -593,6 +574,7 @@ static void __bch2_fs_free(struct bch_fs *c)
 		free_percpu(c->online_reserved);
 	}
 
+	darray_exit(&c->incompat_versions_requested);
 	darray_exit(&c->btree_roots_extra);
 	free_percpu(c->pcpu);
 	free_percpu(c->usage);
@@ -845,25 +827,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 	if (ret)
 		goto err;
 
-#ifdef CONFIG_UNICODE
-	/* Default encoding until we can potentially have more as an option. */
-	c->cf_encoding = utf8_load(BCH_FS_DEFAULT_UTF8_ENCODING);
-	if (IS_ERR(c->cf_encoding)) {
-		printk(KERN_ERR "Cannot load UTF-8 encoding for filesystem. Version: %u.%u.%u",
-			unicode_major(BCH_FS_DEFAULT_UTF8_ENCODING),
-			unicode_minor(BCH_FS_DEFAULT_UTF8_ENCODING),
-			unicode_rev(BCH_FS_DEFAULT_UTF8_ENCODING));
-		ret = -EINVAL;
-		goto err;
-	}
-#else
-	if (c->sb.features & BIT_ULL(BCH_FEATURE_casefolding)) {
-		printk(KERN_ERR "Cannot mount a filesystem with casefolding on a kernel without CONFIG_UNICODE\n");
-		ret = -EINVAL;
-		goto err;
-	}
-#endif
-
 	pr_uuid(&name, c->sb.user_uuid.b);
 	ret = name.allocation_failure ? -BCH_ERR_ENOMEM_fs_name_alloc : 0;
 	if (ret)
@@ -963,6 +926,29 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 	if (ret)
 		goto err;
 
+#ifdef CONFIG_UNICODE
+	/* Default encoding until we can potentially have more as an option. */
+	c->cf_encoding = utf8_load(BCH_FS_DEFAULT_UTF8_ENCODING);
+	if (IS_ERR(c->cf_encoding)) {
+		printk(KERN_ERR "Cannot load UTF-8 encoding for filesystem. Version: %u.%u.%u",
+			unicode_major(BCH_FS_DEFAULT_UTF8_ENCODING),
+			unicode_minor(BCH_FS_DEFAULT_UTF8_ENCODING),
+			unicode_rev(BCH_FS_DEFAULT_UTF8_ENCODING));
+		ret = -EINVAL;
+		goto err;
+	}
+	bch_info(c, "Using encoding defined by superblock: utf8-%u.%u.%u",
+		 unicode_major(BCH_FS_DEFAULT_UTF8_ENCODING),
+		 unicode_minor(BCH_FS_DEFAULT_UTF8_ENCODING),
+		 unicode_rev(BCH_FS_DEFAULT_UTF8_ENCODING));
+#else
+	if (c->sb.features & BIT_ULL(BCH_FEATURE_casefolding)) {
+		printk(KERN_ERR "Cannot mount a filesystem with casefolding on a kernel without CONFIG_UNICODE\n");
+		ret = -EINVAL;
+		goto err;
+	}
+#endif
+
 	for (i = 0; i < c->sb.nr_devices; i++) {
 		if (!bch2_member_exists(c->disk_sb.sb, i))
 			continue;
@@ -1002,12 +988,6 @@ static void print_mount_opts(struct bch_fs *c)
 	prt_str(&p, "starting version ");
 	bch2_version_to_text(&p, c->sb.version);
 
-	if (c->opts.read_only) {
-		prt_str(&p, " opts=");
-		first = false;
-		prt_printf(&p, "ro");
-	}
-
 	for (i = 0; i < bch2_opts_nr; i++) {
 		const struct bch_option *opt = &bch2_opt_table[i];
 		u64 v = bch2_opt_get_by_id(&c->opts, i);
@@ -1023,10 +1003,49 @@ static void print_mount_opts(struct bch_fs *c)
 		bch2_opt_to_text(&p, c, c->disk_sb.sb, opt, v, OPT_SHOW_MOUNT_STYLE);
 	}
 
+	if (c->sb.version_incompat_allowed != c->sb.version) {
+		prt_printf(&p, "\n  allowing incompatible features above ");
+		bch2_version_to_text(&p, c->sb.version_incompat_allowed);
+	}
+
 	bch_info(c, "%s", p.buf);
 	printbuf_exit(&p);
 }
 
+static bool bch2_fs_may_start(struct bch_fs *c)
+{
+	struct bch_dev *ca;
+	unsigned i, flags = 0;
+
+	if (c->opts.very_degraded)
+		flags |= BCH_FORCE_IF_DEGRADED|BCH_FORCE_IF_LOST;
+
+	if (c->opts.degraded)
+		flags |= BCH_FORCE_IF_DEGRADED;
+
+	if (!c->opts.degraded &&
+	    !c->opts.very_degraded) {
+		mutex_lock(&c->sb_lock);
+
+		for (i = 0; i < c->disk_sb.sb->nr_devices; i++) {
+			if (!bch2_member_exists(c->disk_sb.sb, i))
+				continue;
+
+			ca = bch2_dev_locked(c, i);
+
+			if (!bch2_dev_is_online(ca) &&
+			    (ca->mi.state == BCH_MEMBER_STATE_rw ||
+			     ca->mi.state == BCH_MEMBER_STATE_ro)) {
+				mutex_unlock(&c->sb_lock);
+				return false;
+			}
+		}
+		mutex_unlock(&c->sb_lock);
+	}
+
+	return bch2_have_enough_devs(c, bch2_online_devs(c), flags, true);
+}
+
 int bch2_fs_start(struct bch_fs *c)
 {
 	time64_t now = ktime_get_real_seconds();
@@ -1034,6 +1053,9 @@ int bch2_fs_start(struct bch_fs *c)
 
 	print_mount_opts(c);
 
+	if (!bch2_fs_may_start(c))
+		return -BCH_ERR_insufficient_devices_to_start;
+
 	down_write(&c->state_lock);
 	mutex_lock(&c->sb_lock);
 
@@ -1086,13 +1108,10 @@ int bch2_fs_start(struct bch_fs *c)
 	wake_up(&c->ro_ref_wait);
 
 	down_write(&c->state_lock);
-	if (c->opts.read_only) {
+	if (c->opts.read_only)
 		bch2_fs_read_only(c);
-	} else {
-		ret = !test_bit(BCH_FS_rw, &c->flags)
-			? bch2_fs_read_write(c)
-			: bch2_fs_read_write_late(c);
-	}
+	else if (!test_bit(BCH_FS_rw, &c->flags))
+		ret = bch2_fs_read_write(c);
 	up_write(&c->state_lock);
 
 err:
@@ -1504,7 +1523,7 @@ static int bch2_dev_attach_bdev(struct bch_fs *c, struct bch_sb_handle *sb)
 
 	printbuf_exit(&name);
 
-	rebalance_wakeup(c);
+	bch2_rebalance_wakeup(c);
 	return 0;
 }
 
@@ -1563,40 +1582,6 @@ bool bch2_dev_state_allowed(struct bch_fs *c, struct bch_dev *ca,
 	}
 }
 
-static bool bch2_fs_may_start(struct bch_fs *c)
-{
-	struct bch_dev *ca;
-	unsigned i, flags = 0;
-
-	if (c->opts.very_degraded)
-		flags |= BCH_FORCE_IF_DEGRADED|BCH_FORCE_IF_LOST;
-
-	if (c->opts.degraded)
-		flags |= BCH_FORCE_IF_DEGRADED;
-
-	if (!c->opts.degraded &&
-	    !c->opts.very_degraded) {
-		mutex_lock(&c->sb_lock);
-
-		for (i = 0; i < c->disk_sb.sb->nr_devices; i++) {
-			if (!bch2_member_exists(c->disk_sb.sb, i))
-				continue;
-
-			ca = bch2_dev_locked(c, i);
-
-			if (!bch2_dev_is_online(ca) &&
-			    (ca->mi.state == BCH_MEMBER_STATE_rw ||
-			     ca->mi.state == BCH_MEMBER_STATE_ro)) {
-				mutex_unlock(&c->sb_lock);
-				return false;
-			}
-		}
-		mutex_unlock(&c->sb_lock);
-	}
-
-	return bch2_have_enough_devs(c, bch2_online_devs(c), flags, true);
-}
-
 static void __bch2_dev_read_only(struct bch_fs *c, struct bch_dev *ca)
 {
 	bch2_dev_io_ref_stop(ca, WRITE);
@@ -1650,7 +1635,7 @@ int __bch2_dev_set_state(struct bch_fs *c, struct bch_dev *ca,
 	if (new_state == BCH_MEMBER_STATE_rw)
 		__bch2_dev_read_write(c, ca);
 
-	rebalance_wakeup(c);
+	bch2_rebalance_wakeup(c);
 
 	return ret;
 }
@@ -1767,7 +1752,8 @@ int bch2_dev_remove(struct bch_fs *c, struct bch_dev *ca, int flags)
 	up_write(&c->state_lock);
 	return 0;
 err:
-	if (ca->mi.state == BCH_MEMBER_STATE_rw &&
+	if (test_bit(BCH_FS_rw, &c->flags) &&
+	    ca->mi.state == BCH_MEMBER_STATE_rw &&
 	    !percpu_ref_is_zero(&ca->io_ref[READ]))
 		__bch2_dev_read_write(c, ca);
 	up_write(&c->state_lock);
@@ -2231,11 +2217,6 @@ struct bch_fs *bch2_fs_open(char * const *devices, unsigned nr_devices,
 	}
 	up_write(&c->state_lock);
 
-	if (!bch2_fs_may_start(c)) {
-		ret = -BCH_ERR_insufficient_devices_to_start;
-		goto err_print;
-	}
-
 	if (!c->opts.nostart) {
 		ret = bch2_fs_start(c);
 		if (ret)
diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c
index e5f003c29369..82ee333ddd21 100644
--- a/fs/bcachefs/sysfs.c
+++ b/fs/bcachefs/sysfs.c
@@ -654,11 +654,10 @@ static ssize_t sysfs_opt_store(struct bch_fs *c,
 		bch2_set_rebalance_needs_scan(c, 0);
 
 	if (v && id == Opt_rebalance_enabled)
-		rebalance_wakeup(c);
+		bch2_rebalance_wakeup(c);
 
-	if (v && id == Opt_copygc_enabled &&
-	    c->copygc_thread)
-		wake_up_process(c->copygc_thread);
+	if (v && id == Opt_copygc_enabled)
+		bch2_copygc_wakeup(c);
 
 	if (id == Opt_discard && !ca) {
 		mutex_lock(&c->sb_lock);
diff --git a/fs/bcachefs/tests.c b/fs/bcachefs/tests.c
index c265b102267a..782a05fe7656 100644
--- a/fs/bcachefs/tests.c
+++ b/fs/bcachefs/tests.c
@@ -342,6 +342,8 @@ static int test_iterate_slots_extents(struct bch_fs *c, u64 nr)
  */
 static int test_peek_end(struct bch_fs *c, u64 nr)
 {
+	delete_test_keys(c);
+
 	struct btree_trans *trans = bch2_trans_get(c);
 	struct btree_iter iter;
 	struct bkey_s_c k;
@@ -362,6 +364,8 @@ static int test_peek_end(struct bch_fs *c, u64 nr)
 
 static int test_peek_end_extents(struct bch_fs *c, u64 nr)
 {
+	delete_test_keys(c);
+
 	struct btree_trans *trans = bch2_trans_get(c);
 	struct btree_iter iter;
 	struct bkey_s_c k;
diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h
index 6ba5071ab6dd..3e52c7f8ddd2 100644
--- a/fs/bcachefs/util.h
+++ b/fs/bcachefs/util.h
@@ -739,4 +739,42 @@ static inline void memcpy_swab(void *_dst, void *_src, size_t len)
 		*--dst = *src++;
 }
 
+#define set_flags(_map, _in, _out)					\
+do {									\
+	unsigned _i;							\
+									\
+	for (_i = 0; _i < ARRAY_SIZE(_map); _i++)			\
+		if ((_in) & (1 << _i))					\
+			(_out) |= _map[_i];				\
+		else							\
+			(_out) &= ~_map[_i];				\
+} while (0)
+
+#define map_flags(_map, _in)						\
+({									\
+	unsigned _out = 0;						\
+									\
+	set_flags(_map, _in, _out);					\
+	_out;								\
+})
+
+#define map_flags_rev(_map, _in)					\
+({									\
+	unsigned _i, _out = 0;						\
+									\
+	for (_i = 0; _i < ARRAY_SIZE(_map); _i++)			\
+		if ((_in) & _map[_i]) {					\
+			(_out) |= 1 << _i;				\
+			(_in) &= ~_map[_i];				\
+		}							\
+	(_out);								\
+})
+
+#define map_defined(_map)						\
+({									\
+	unsigned _in = ~0;						\
+									\
+	map_flags_rev(_map, _in);					\
+})
+
 #endif /* _BCACHEFS_UTIL_H */
diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h
index c7916011ef34..67426e33d04e 100644
--- a/fs/bcachefs/xattr_format.h
+++ b/fs/bcachefs/xattr_format.h
@@ -13,7 +13,13 @@ struct bch_xattr {
 	__u8			x_type;
 	__u8			x_name_len;
 	__le16			x_val_len;
-	__u8			x_name[] __counted_by(x_name_len);
+	/*
+	 * x_name contains the name and value counted by
+	 * x_name_len + x_val_len. The introduction of
+	 * __counted_by(x_name_len) caused a false positive
+	 * detection of an out of bounds write.
+	 */
+	__u8			x_name[];
 } __packed __aligned(8);
 
 #endif /* _BCACHEFS_XATTR_FORMAT_H */
diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig
index fa8515598341..73a2dfb854c5 100644
--- a/fs/btrfs/Kconfig
+++ b/fs/btrfs/Kconfig
@@ -3,9 +3,9 @@
 config BTRFS_FS
 	tristate "Btrfs filesystem support"
 	select BLK_CGROUP_PUNT_BIO
+	select CRC32
 	select CRYPTO
 	select CRYPTO_CRC32C
-	select LIBCRC32C
 	select CRYPTO_XXHASH
 	select CRYPTO_SHA256
 	select CRYPTO_BLAKE2B
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 3dd555db3d32..aa58e0663a5d 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -3853,7 +3853,6 @@ static int write_dev_supers(struct btrfs_device *device,
 			atomic_inc(&device->sb_write_errors);
 			continue;
 		}
-		ASSERT(folio_order(folio) == 0);
 
 		offset = offset_in_folio(folio, bytenr);
 		disk_super = folio_address(folio) + offset;
@@ -3926,7 +3925,6 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors)
 		/* If the folio has been removed, then we know it completed. */
 		if (IS_ERR(folio))
 			continue;
-		ASSERT(folio_order(folio) == 0);
 
 		/* Folio will be unlocked once the write completes. */
 		folio_wait_locked(folio);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 197f5e51c474..8515c31f563b 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2047,7 +2047,7 @@ static int submit_eb_subpage(struct folio *folio, struct writeback_control *wbc)
 			      subpage->bitmaps)) {
 			spin_unlock_irqrestore(&subpage->lock, flags);
 			spin_unlock(&folio->mapping->i_private_lock);
-			bit_start++;
+			bit_start += sectors_per_node;
 			continue;
 		}
 
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 262a707d8990..71b8a825c447 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2104,15 +2104,20 @@ static void btrfs_punch_hole_lock_range(struct inode *inode,
 	 * will always return true.
 	 * So here we need to do extra page alignment for
 	 * filemap_range_has_page().
+	 *
+	 * And do not decrease page_lockend right now, as it can be 0.
 	 */
 	const u64 page_lockstart = round_up(lockstart, PAGE_SIZE);
-	const u64 page_lockend = round_down(lockend + 1, PAGE_SIZE) - 1;
+	const u64 page_lockend = round_down(lockend + 1, PAGE_SIZE);
 
 	while (1) {
 		truncate_pagecache_range(inode, lockstart, lockend);
 
 		lock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			    cached_state);
+		/* The same page or adjacent pages. */
+		if (page_lockend <= page_lockstart)
+			break;
 		/*
 		 * We can't have ordered extents in the range, nor dirty/writeback
 		 * pages, because we have locked the inode's VFS lock in exclusive
@@ -2124,7 +2129,7 @@ static void btrfs_punch_hole_lock_range(struct inode *inode,
 		 * we do, unlock the range and retry.
 		 */
 		if (!filemap_range_has_page(inode->i_mapping, page_lockstart,
-					    page_lockend))
+					    page_lockend - 1))
 			break;
 
 		unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend,
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index cc67d1a2d611..bdafe4d4c4a5 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2129,12 +2129,13 @@ next_slot:
 
 		/*
 		 * If the found extent starts after requested offset, then
-		 * adjust extent_end to be right before this extent begins
+		 * adjust cur_offset to be right before this extent begins.
 		 */
 		if (found_key.offset > cur_offset) {
-			extent_end = found_key.offset;
-			extent_type = 0;
-			goto must_cow;
+			if (cow_start == (u64)-1)
+				cow_start = cur_offset;
+			cur_offset = found_key.offset;
+			goto next_slot;
 		}
 
 		/*
@@ -5681,8 +5682,10 @@ struct btrfs_inode *btrfs_iget(u64 ino, struct btrfs_root *root)
 		return inode;
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		iget_failed(&inode->vfs_inode);
 		return ERR_PTR(-ENOMEM);
+	}
 
 	ret = btrfs_read_locked_inode(inode, path);
 	btrfs_free_path(path);
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index a13d81bb56a0..63aeacc54945 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -4902,6 +4902,8 @@ static int btrfs_uring_encoded_read(struct io_uring_cmd *cmd, unsigned int issue
 
 	ret = btrfs_encoded_read(&kiocb, &data->iter, &data->args, &cached_state,
 				 &disk_bytenr, &disk_io_size);
+	if (ret == -EAGAIN)
+		goto out_acct;
 	if (ret < 0 && ret != -EIOCBQUEUED)
 		goto out_free;
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index f948f4f6431c..e17bcb034595 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3803,7 +3803,7 @@ out:
 	if (ret) {
 		if (inode)
 			iput(&inode->vfs_inode);
-		inode = ERR_PTR(ret);
+		return ERR_PTR(ret);
 	}
 	return &inode->vfs_inode;
 }
diff --git a/fs/btrfs/subpage.c b/fs/btrfs/subpage.c
index 11dbd7be6a3b..c0a0b8b063d0 100644
--- a/fs/btrfs/subpage.c
+++ b/fs/btrfs/subpage.c
@@ -204,7 +204,7 @@ static void btrfs_subpage_assert(const struct btrfs_fs_info *fs_info,
 			   btrfs_blocks_per_folio(fs_info, folio);	\
 									\
 	btrfs_subpage_assert(fs_info, folio, start, len);		\
-	__start_bit = offset_in_page(start) >> fs_info->sectorsize_bits; \
+	__start_bit = offset_in_folio(folio, start) >> fs_info->sectorsize_bits; \
 	__start_bit += blocks_per_folio * btrfs_bitmap_nr_##name;	\
 	__start_bit;							\
 })
@@ -666,7 +666,7 @@ IMPLEMENT_BTRFS_PAGE_OPS(checked, folio_set_checked, folio_clear_checked,
 				btrfs_blocks_per_folio(fs_info, folio);	\
 	const struct btrfs_subpage *subpage = folio_get_private(folio);	\
 									\
-	ASSERT(blocks_per_folio < BITS_PER_LONG);			\
+	ASSERT(blocks_per_folio <= BITS_PER_LONG);			\
 	*dst = bitmap_read(subpage->bitmaps,				\
 			   blocks_per_folio * btrfs_bitmap_nr_##name,	\
 			   blocks_per_folio);				\
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 40709e2a44fc..7121d8c7a318 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -1139,8 +1139,7 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 	subvol_name = btrfs_get_subvol_name_from_objectid(info,
 			btrfs_root_id(BTRFS_I(d_inode(dentry))->root));
 	if (!IS_ERR(subvol_name)) {
-		seq_puts(seq, ",subvol=");
-		seq_escape(seq, subvol_name, " \t\n\\");
+		seq_show_option(seq, "subvol", subvol_name);
 		kfree(subvol_name);
 	}
 	return 0;
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 43979891f7c8..2b66a6130269 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -2235,7 +2235,7 @@ int btrfs_verify_level_key(struct extent_buffer *eb,
 		btrfs_err(fs_info,
 "tree level mismatch detected, bytenr=%llu level expected=%u has=%u",
 			  eb->start, check->level, found_level);
-		return -EIO;
+		return -EUCLEAN;
 	}
 
 	if (!check->has_first_key)
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index fb8b8b29c169..4a3e02b49f29 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -1277,7 +1277,7 @@ struct zone_info {
 
 static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 				struct zone_info *info, unsigned long *active,
-				struct btrfs_chunk_map *map)
+				struct btrfs_chunk_map *map, bool new)
 {
 	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
 	struct btrfs_device *device;
@@ -1307,6 +1307,8 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 		return 0;
 	}
 
+	ASSERT(!new || btrfs_dev_is_empty_zone(device, info->physical));
+
 	/* This zone will be used for allocation, so mark this zone non-empty. */
 	btrfs_dev_clear_zone_empty(device, info->physical);
 
@@ -1319,6 +1321,18 @@ static int btrfs_load_zone_info(struct btrfs_fs_info *fs_info, int zone_idx,
 	 * to determine the allocation offset within the zone.
 	 */
 	WARN_ON(!IS_ALIGNED(info->physical, fs_info->zone_size));
+
+	if (new) {
+		sector_t capacity;
+
+		capacity = bdev_zone_capacity(device->bdev, info->physical >> SECTOR_SHIFT);
+		up_read(&dev_replace->rwsem);
+		info->alloc_offset = 0;
+		info->capacity = capacity << SECTOR_SHIFT;
+
+		return 0;
+	}
+
 	nofs_flag = memalloc_nofs_save();
 	ret = btrfs_get_dev_zone(device, info->physical, &zone);
 	memalloc_nofs_restore(nofs_flag);
@@ -1588,7 +1602,7 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 	}
 
 	for (i = 0; i < map->num_stripes; i++) {
-		ret = btrfs_load_zone_info(fs_info, i, &zone_info[i], active, map);
+		ret = btrfs_load_zone_info(fs_info, i, &zone_info[i], active, map, new);
 		if (ret)
 			goto out;
 
@@ -1659,7 +1673,6 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 		 * stripe.
 		 */
 		cache->alloc_offset = cache->zone_capacity;
-		ret = 0;
 	}
 
 out:
diff --git a/fs/buffer.c b/fs/buffer.c
index c7abb4a029dc..7be23ff20b27 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -176,18 +176,8 @@ void end_buffer_write_sync(struct buffer_head *bh, int uptodate)
 }
 EXPORT_SYMBOL(end_buffer_write_sync);
 
-/*
- * Various filesystems appear to want __find_get_block to be non-blocking.
- * But it's the page lock which protects the buffers.  To get around this,
- * we get exclusion from try_to_free_buffers with the blockdev mapping's
- * i_private_lock.
- *
- * Hack idea: for the blockdev mapping, i_private_lock contention
- * may be quite high.  This code could TryLock the page, and if that
- * succeeds, there is no need to take i_private_lock.
- */
 static struct buffer_head *
-__find_get_block_slow(struct block_device *bdev, sector_t block)
+__find_get_block_slow(struct block_device *bdev, sector_t block, bool atomic)
 {
 	struct address_space *bd_mapping = bdev->bd_mapping;
 	const int blkbits = bd_mapping->host->i_blkbits;
@@ -204,10 +194,28 @@ __find_get_block_slow(struct block_device *bdev, sector_t block)
 	if (IS_ERR(folio))
 		goto out;
 
-	spin_lock(&bd_mapping->i_private_lock);
+	/*
+	 * Folio lock protects the buffers. Callers that cannot block
+	 * will fallback to serializing vs try_to_free_buffers() via
+	 * the i_private_lock.
+	 */
+	if (atomic)
+		spin_lock(&bd_mapping->i_private_lock);
+	else
+		folio_lock(folio);
+
 	head = folio_buffers(folio);
 	if (!head)
 		goto out_unlock;
+	/*
+	 * Upon a noref migration, the folio lock serializes here;
+	 * otherwise bail.
+	 */
+	if (test_bit_acquire(BH_Migrate, &head->b_state)) {
+		WARN_ON(!atomic);
+		goto out_unlock;
+	}
+
 	bh = head;
 	do {
 		if (!buffer_mapped(bh))
@@ -236,7 +244,10 @@ __find_get_block_slow(struct block_device *bdev, sector_t block)
 		       1 << blkbits);
 	}
 out_unlock:
-	spin_unlock(&bd_mapping->i_private_lock);
+	if (atomic)
+		spin_unlock(&bd_mapping->i_private_lock);
+	else
+		folio_unlock(folio);
 	folio_put(folio);
 out:
 	return ret;
@@ -656,7 +667,9 @@ EXPORT_SYMBOL(generic_buffers_fsync);
 void write_boundary_block(struct block_device *bdev,
 			sector_t bblock, unsigned blocksize)
 {
-	struct buffer_head *bh = __find_get_block(bdev, bblock + 1, blocksize);
+	struct buffer_head *bh;
+
+	bh = __find_get_block_nonatomic(bdev, bblock + 1, blocksize);
 	if (bh) {
 		if (buffer_dirty(bh))
 			write_dirty_buffer(bh, 0);
@@ -1386,16 +1399,18 @@ lookup_bh_lru(struct block_device *bdev, sector_t block, unsigned size)
 /*
  * Perform a pagecache lookup for the matching buffer.  If it's there, refresh
  * it in the LRU and mark it as accessed.  If it is not present then return
- * NULL
+ * NULL. Atomic context callers may also return NULL if the buffer is being
+ * migrated; similarly the page is not marked accessed either.
  */
-struct buffer_head *
-__find_get_block(struct block_device *bdev, sector_t block, unsigned size)
+static struct buffer_head *
+find_get_block_common(struct block_device *bdev, sector_t block,
+			unsigned size, bool atomic)
 {
 	struct buffer_head *bh = lookup_bh_lru(bdev, block, size);
 
 	if (bh == NULL) {
 		/* __find_get_block_slow will mark the page accessed */
-		bh = __find_get_block_slow(bdev, block);
+		bh = __find_get_block_slow(bdev, block, atomic);
 		if (bh)
 			bh_lru_install(bh);
 	} else
@@ -1403,8 +1418,23 @@ __find_get_block(struct block_device *bdev, sector_t block, unsigned size)
 
 	return bh;
 }
+
+struct buffer_head *
+__find_get_block(struct block_device *bdev, sector_t block, unsigned size)
+{
+	return find_get_block_common(bdev, block, size, true);
+}
 EXPORT_SYMBOL(__find_get_block);
 
+/* same as __find_get_block() but allows sleeping contexts */
+struct buffer_head *
+__find_get_block_nonatomic(struct block_device *bdev, sector_t block,
+			   unsigned size)
+{
+	return find_get_block_common(bdev, block, size, false);
+}
+EXPORT_SYMBOL(__find_get_block_nonatomic);
+
 /**
  * bdev_getblk - Get a buffer_head in a block device's buffer cache.
  * @bdev: The block device.
@@ -1422,7 +1452,12 @@ EXPORT_SYMBOL(__find_get_block);
 struct buffer_head *bdev_getblk(struct block_device *bdev, sector_t block,
 		unsigned size, gfp_t gfp)
 {
-	struct buffer_head *bh = __find_get_block(bdev, block, size);
+	struct buffer_head *bh;
+
+	if (gfpflags_allow_blocking(gfp))
+		bh = __find_get_block_nonatomic(bdev, block, size);
+	else
+		bh = __find_get_block(bdev, block, size);
 
 	might_alloc(gfp);
 	if (bh)
diff --git a/fs/cachefiles/key.c b/fs/cachefiles/key.c
index bf935e25bdbe..b48525680e73 100644
--- a/fs/cachefiles/key.c
+++ b/fs/cachefiles/key.c
@@ -8,7 +8,7 @@
 #include <linux/slab.h>
 #include "internal.h"
 
-static const char cachefiles_charmap[64] =
+static const char cachefiles_charmap[64] __nonstring =
 	"0123456789"			/* 0 - 9 */
 	"abcdefghijklmnopqrstuvwxyz"	/* 10 - 35 */
 	"ABCDEFGHIJKLMNOPQRSTUVWXYZ"	/* 36 - 61 */
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index 7249d70e1a43..3e7def3d31c1 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -3,7 +3,7 @@ config CEPH_FS
 	tristate "Ceph distributed file system"
 	depends on INET
 	select CEPH_LIB
-	select LIBCRC32C
+	select CRC32
 	select CRYPTO_AES
 	select CRYPTO
 	select NETFS_SUPPORT
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 6ac2bd555e86..06cd2963e41e 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -2367,7 +2367,7 @@ static int fill_fscrypt_truncate(struct inode *inode,
 
 	/* Try to writeback the dirty pagecaches */
 	if (issued & (CEPH_CAP_FILE_BUFFER)) {
-		loff_t lend = orig_pos + CEPH_FSCRYPT_BLOCK_SHIFT - 1;
+		loff_t lend = orig_pos + CEPH_FSCRYPT_BLOCK_SIZE - 1;
 
 		ret = filemap_write_and_wait_range(inode->i_mapping,
 						   orig_pos, lend);
diff --git a/fs/dax.c b/fs/dax.c
index af5045b0f476..676303419e9e 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -396,6 +396,7 @@ static inline unsigned long dax_folio_put(struct folio *folio)
 	order = folio_order(folio);
 	if (!order)
 		return 0;
+	folio_reset_order(folio);
 
 	for (i = 0; i < (1UL << order); i++) {
 		struct dev_pagemap *pgmap = page_pgmap(&folio->page);
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 42e4d6eeb29f..9c20d78e41f6 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -89,12 +89,12 @@ enum {
 };
 
 static const struct fs_parameter_spec devpts_param_specs[] = {
-	fsparam_u32	("gid",		Opt_gid),
+	fsparam_gid	("gid",		Opt_gid),
 	fsparam_s32	("max",		Opt_max),
 	fsparam_u32oct	("mode",	Opt_mode),
 	fsparam_flag	("newinstance",	Opt_newinstance),
 	fsparam_u32oct	("ptmxmode",	Opt_ptmxmode),
-	fsparam_u32	("uid",		Opt_uid),
+	fsparam_uid	("uid",		Opt_uid),
 	{}
 };
 
diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig
index 331e49cd1b8d..8f68ec49ad89 100644
--- a/fs/erofs/Kconfig
+++ b/fs/erofs/Kconfig
@@ -3,8 +3,8 @@
 config EROFS_FS
 	tristate "EROFS filesystem support"
 	depends on BLOCK
+	select CRC32
 	select FS_IOMAP
-	select LIBCRC32C
 	help
 	  EROFS (Enhanced Read-Only File System) is a lightweight read-only
 	  file system with modern designs (e.g. no buffer heads, inline
diff --git a/fs/erofs/erofs_fs.h b/fs/erofs/erofs_fs.h
index 9581e9bf8192..767fb4acdc93 100644
--- a/fs/erofs/erofs_fs.h
+++ b/fs/erofs/erofs_fs.h
@@ -56,7 +56,7 @@ struct erofs_super_block {
 	union {
 		__le16 rootnid_2b;	/* nid of root directory */
 		__le16 blocks_hi;	/* (48BIT on) blocks count MSB */
-	} rb;
+	} __packed rb;
 	__le64 inos;            /* total valid ino # (== f_files - f_favail) */
 	__le64 epoch;		/* base seconds used for compact inodes */
 	__le32 fixed_nsec;	/* fixed nanoseconds for compact inodes */
@@ -148,7 +148,7 @@ union erofs_inode_i_nb {
 	__le16 nlink;		/* if EROFS_I_NLINK_1_BIT is unset */
 	__le16 blocks_hi;	/* total blocks count MSB */
 	__le16 startblk_hi;	/* starting block number MSB */
-};
+} __packed;
 
 /* 32-byte reduced form of an ondisk inode */
 struct erofs_inode_compact {
@@ -369,9 +369,9 @@ struct z_erofs_map_header {
 			 * bit 7   : pack the whole file into packed inode
 			 */
 			__u8	h_clusterbits;
-		};
+		} __packed;
 		__le16 h_extents_hi;	/* extent count MSB */
-	};
+	} __packed;
 };
 
 enum {
diff --git a/fs/erofs/fileio.c b/fs/erofs/fileio.c
index bec4b56b3826..4fa0a0121288 100644
--- a/fs/erofs/fileio.c
+++ b/fs/erofs/fileio.c
@@ -32,6 +32,8 @@ static void erofs_fileio_ki_complete(struct kiocb *iocb, long ret)
 		ret = 0;
 	}
 	if (rq->bio.bi_end_io) {
+		if (ret < 0 && !rq->bio.bi_status)
+			rq->bio.bi_status = errno_to_blk_status(ret);
 		rq->bio.bi_end_io(&rq->bio);
 	} else {
 		bio_for_each_folio_all(fi, &rq->bio) {
diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
index 0671184d9cf1..5c061aaeeb45 100644
--- a/fs/erofs/zdata.c
+++ b/fs/erofs/zdata.c
@@ -725,7 +725,6 @@ static int z_erofs_register_pcluster(struct z_erofs_frontend *fe)
 	lockref_init(&pcl->lockref); /* one ref for this request */
 	pcl->algorithmformat = map->m_algorithmformat;
 	pcl->pclustersize = map->m_plen;
-	pcl->pageofs_in = pageofs_in;
 	pcl->length = 0;
 	pcl->partial = true;
 	pcl->next = fe->head;
diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c
index 8de50df05dfe..14ea47f954f5 100644
--- a/fs/erofs/zmap.c
+++ b/fs/erofs/zmap.c
@@ -559,7 +559,8 @@ static int z_erofs_map_blocks_ext(struct inode *inode,
 			pos += sizeof(__le64);
 			lstart = 0;
 		} else {
-			lstart = map->m_la >> vi->z_lclusterbits;
+			lstart = round_down(map->m_la, 1 << vi->z_lclusterbits);
+			pos += (lstart >> vi->z_lclusterbits) * recsz;
 			pa = EROFS_NULL_ADDR;
 		}
 
@@ -614,7 +615,7 @@ static int z_erofs_map_blocks_ext(struct inode *inode,
 		if (last && (vi->z_advise & Z_EROFS_ADVISE_FRAGMENT_PCLUSTER)) {
 			map->m_flags |= EROFS_MAP_MAPPED | EROFS_MAP_FRAGMENT;
 			vi->z_fragmentoff = map->m_plen;
-			if (recsz >= offsetof(struct z_erofs_extent, pstart_lo))
+			if (recsz > offsetof(struct z_erofs_extent, pstart_lo))
 				vi->z_fragmentoff |= map->m_pa << 32;
 		} else if (map->m_plen) {
 			map->m_flags |= EROFS_MAP_MAPPED |
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 100376863a44..4bc264b854c4 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1996,6 +1996,14 @@ static int ep_try_send_events(struct eventpoll *ep,
 	return res;
 }
 
+static int ep_schedule_timeout(ktime_t *to)
+{
+	if (to)
+		return ktime_after(*to, ktime_get());
+	else
+		return 1;
+}
+
 /**
  * ep_poll - Retrieves ready events, and delivers them to the caller-supplied
  *           event buffer.
@@ -2103,7 +2111,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
 
 		write_unlock_irq(&ep->lock);
 
-		if (!eavail)
+		if (!eavail && ep_schedule_timeout(to))
 			timed_out = !schedule_hrtimeout_range(to, slack,
 							      HRTIMER_MODE_ABS);
 		__set_current_state(TASK_RUNNING);
diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c
index 87ee3a17bd29..e8c5525afc67 100644
--- a/fs/ext4/block_validity.c
+++ b/fs/ext4/block_validity.c
@@ -351,10 +351,9 @@ int ext4_check_blockref(const char *function, unsigned int line,
 {
 	__le32 *bref = p;
 	unsigned int blk;
+	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
 
-	if (ext4_has_feature_journal(inode->i_sb) &&
-	    (inode->i_ino ==
-	     le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_journal_inum)))
+	if (journal && inode == journal->j_inode)
 		return 0;
 
 	while (bref < p+max) {
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 38bc8d74f4cc..e7ecc7c8a729 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -691,7 +691,8 @@ static int recently_deleted(struct super_block *sb, ext4_group_t group, int ino)
 	if (!bh || !buffer_uptodate(bh))
 		/*
 		 * If the block is not in the buffer cache, then it
-		 * must have been written out.
+		 * must have been written out, or, most unlikely, is
+		 * being migrated - false failure should be OK here.
 		 */
 		goto out;
 
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 1dc09ed5d403..94c7d2d828a6 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -386,10 +386,11 @@ static int __check_block_validity(struct inode *inode, const char *func,
 				unsigned int line,
 				struct ext4_map_blocks *map)
 {
-	if (ext4_has_feature_journal(inode->i_sb) &&
-	    (inode->i_ino ==
-	     le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_journal_inum)))
+	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
+
+	if (journal && inode == journal->j_inode)
 		return 0;
+
 	if (!ext4_inode_block_valid(inode, map->m_pblk, map->m_len)) {
 		ext4_error_inode(inode, func, line, map->m_pblk,
 				 "lblock %lu mapped to illegal pblock %llu "
@@ -4724,22 +4725,43 @@ static inline void ext4_inode_set_iversion_queried(struct inode *inode, u64 val)
 		inode_set_iversion_queried(inode, val);
 }
 
-static const char *check_igot_inode(struct inode *inode, ext4_iget_flags flags)
-
+static int check_igot_inode(struct inode *inode, ext4_iget_flags flags,
+			    const char *function, unsigned int line)
 {
+	const char *err_str;
+
 	if (flags & EXT4_IGET_EA_INODE) {
-		if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
-			return "missing EA_INODE flag";
+		if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) {
+			err_str = "missing EA_INODE flag";
+			goto error;
+		}
 		if (ext4_test_inode_state(inode, EXT4_STATE_XATTR) ||
-		    EXT4_I(inode)->i_file_acl)
-			return "ea_inode with extended attributes";
+		    EXT4_I(inode)->i_file_acl) {
+			err_str = "ea_inode with extended attributes";
+			goto error;
+		}
 	} else {
-		if ((EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
-			return "unexpected EA_INODE flag";
+		if ((EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) {
+			/*
+			 * open_by_handle_at() could provide an old inode number
+			 * that has since been reused for an ea_inode; this does
+			 * not indicate filesystem corruption
+			 */
+			if (flags & EXT4_IGET_HANDLE)
+				return -ESTALE;
+			err_str = "unexpected EA_INODE flag";
+			goto error;
+		}
+	}
+	if (is_bad_inode(inode) && !(flags & EXT4_IGET_BAD)) {
+		err_str = "unexpected bad inode w/o EXT4_IGET_BAD";
+		goto error;
 	}
-	if (is_bad_inode(inode) && !(flags & EXT4_IGET_BAD))
-		return "unexpected bad inode w/o EXT4_IGET_BAD";
-	return NULL;
+	return 0;
+
+error:
+	ext4_error_inode(inode, function, line, 0, err_str);
+	return -EFSCORRUPTED;
 }
 
 struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
@@ -4751,7 +4773,6 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 	struct ext4_inode_info *ei;
 	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
 	struct inode *inode;
-	const char *err_str;
 	journal_t *journal = EXT4_SB(sb)->s_journal;
 	long ret;
 	loff_t size;
@@ -4780,10 +4801,10 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
 	if (!(inode->i_state & I_NEW)) {
-		if ((err_str = check_igot_inode(inode, flags)) != NULL) {
-			ext4_error_inode(inode, function, line, 0, err_str);
+		ret = check_igot_inode(inode, flags, function, line);
+		if (ret) {
 			iput(inode);
-			return ERR_PTR(-EFSCORRUPTED);
+			return ERR_PTR(ret);
 		}
 		return inode;
 	}
@@ -5065,13 +5086,21 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 		ret = -EFSCORRUPTED;
 		goto bad_inode;
 	}
-	if ((err_str = check_igot_inode(inode, flags)) != NULL) {
-		ext4_error_inode(inode, function, line, 0, err_str);
-		ret = -EFSCORRUPTED;
-		goto bad_inode;
+	ret = check_igot_inode(inode, flags, function, line);
+	/*
+	 * -ESTALE here means there is nothing inherently wrong with the inode,
+	 * it's just not an inode we can return for an fhandle lookup.
+	 */
+	if (ret == -ESTALE) {
+		brelse(iloc.bh);
+		unlock_new_inode(inode);
+		iput(inode);
+		return ERR_PTR(-ESTALE);
 	}
-
+	if (ret)
+		goto bad_inode;
 	brelse(iloc.bh);
+
 	unlock_new_inode(inode);
 	return inode;
 
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 0d523e9fb3d5..1e98c5be4e0a 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -3037,10 +3037,8 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
 	unsigned char blocksize_bits = min_t(unsigned char,
 					     sb->s_blocksize_bits,
 					     EXT4_MAX_BLOCK_LOG_SIZE);
-	struct sg {
-		struct ext4_group_info info;
-		ext4_grpblk_t counters[EXT4_MAX_BLOCK_LOG_SIZE + 2];
-	} sg;
+	DEFINE_RAW_FLEX(struct ext4_group_info, sg, bb_counters,
+			EXT4_MAX_BLOCK_LOG_SIZE + 2);
 
 	group--;
 	if (group == 0)
@@ -3048,7 +3046,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
 			      " 2^0   2^1   2^2   2^3   2^4   2^5   2^6  "
 			      " 2^7   2^8   2^9   2^10  2^11  2^12  2^13  ]\n");
 
-	i = (blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) +
+	i = (blocksize_bits + 2) * sizeof(sg->bb_counters[0]) +
 		sizeof(struct ext4_group_info);
 
 	grinfo = ext4_get_group_info(sb, group);
@@ -3068,14 +3066,14 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
 	 * We care only about free space counters in the group info and
 	 * these are safe to access even after the buddy has been unloaded
 	 */
-	memcpy(&sg, grinfo, i);
-	seq_printf(seq, "#%-5u: %-5u %-5u %-5u [", group, sg.info.bb_free,
-			sg.info.bb_fragments, sg.info.bb_first_free);
+	memcpy(sg, grinfo, i);
+	seq_printf(seq, "#%-5u: %-5u %-5u %-5u [", group, sg->bb_free,
+			sg->bb_fragments, sg->bb_first_free);
 	for (i = 0; i <= 13; i++)
 		seq_printf(seq, " %-5u", i <= blocksize_bits + 1 ?
-				sg.info.bb_counters[i] : 0);
+				sg->bb_counters[i] : 0);
 	seq_puts(seq, " ]");
-	if (EXT4_MB_GRP_BBITMAP_CORRUPT(&sg.info))
+	if (EXT4_MB_GRP_BBITMAP_CORRUPT(sg))
 		seq_puts(seq, " Block bitmap corrupted!");
 	seq_putc(seq, '\n');
 	return 0;
@@ -6644,7 +6642,8 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 		for (i = 0; i < count; i++) {
 			cond_resched();
 			if (is_metadata)
-				bh = sb_find_get_block(inode->i_sb, block + i);
+				bh = sb_find_get_block_nonatomic(inode->i_sb,
+								 block + i);
 			ext4_forget(handle, is_metadata, inode, bh, block + i);
 		}
 	}
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index cb5cb33b1d91..e9712e64ec8f 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -1971,7 +1971,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 	 * split it in half by count; each resulting block will have at least
 	 * half the space free.
 	 */
-	if (i > 0)
+	if (i >= 0)
 		split = count - move;
 	else
 		split = count/2;
diff --git a/fs/file.c b/fs/file.c
index dc3f7e120e3e..3a3146664cf3 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -26,7 +26,7 @@
 
 #include "internal.h"
 
-bool __file_ref_put_badval(file_ref_t *ref, unsigned long cnt)
+static noinline bool __file_ref_put_badval(file_ref_t *ref, unsigned long cnt)
 {
 	/*
 	 * If the reference count was already in the dead zone, then this
diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c
index 2c7b24cb67ad..53c2626e90e7 100644
--- a/fs/fuse/virtio_fs.c
+++ b/fs/fuse/virtio_fs.c
@@ -1669,6 +1669,9 @@ static int virtio_fs_get_tree(struct fs_context *fsc)
 	unsigned int virtqueue_size;
 	int err = -EIO;
 
+	if (!fsc->source)
+		return invalf(fsc, "No source specified");
+
 	/* This gets a reference on virtio_fs object. This ptr gets installed
 	 * in fc->iq->priv. Once fuse_conn is going away, it calls ->put()
 	 * to drop the reference to this object.
diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig
index be7f87a8e11a..7bd231d16d4a 100644
--- a/fs/gfs2/Kconfig
+++ b/fs/gfs2/Kconfig
@@ -4,7 +4,6 @@ config GFS2_FS
 	select BUFFER_HEAD
 	select FS_POSIX_ACL
 	select CRC32
-	select LIBCRC32C
 	select QUOTACTL
 	select FS_IOMAP
 	help
diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c
index 6add6ebfef89..cb823a8a6ba9 100644
--- a/fs/hfs/bnode.c
+++ b/fs/hfs/bnode.c
@@ -67,6 +67,12 @@ void hfs_bnode_read_key(struct hfs_bnode *node, void *key, int off)
 	else
 		key_len = tree->max_key_len + 1;
 
+	if (key_len > sizeof(hfs_btree_key) || key_len < 1) {
+		memset(key, 0, sizeof(hfs_btree_key));
+		pr_err("hfs: Invalid key length: %d\n", key_len);
+		return;
+	}
+
 	hfs_bnode_read(node, key, off, key_len);
 }
 
diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c
index 87974d5e6791..079ea80534f7 100644
--- a/fs/hfsplus/bnode.c
+++ b/fs/hfsplus/bnode.c
@@ -67,6 +67,12 @@ void hfs_bnode_read_key(struct hfs_bnode *node, void *key, int off)
 	else
 		key_len = tree->max_key_len + 2;
 
+	if (key_len > sizeof(hfsplus_btree_key) || key_len < 1) {
+		memset(key, 0, sizeof(hfsplus_btree_key));
+		pr_err("hfsplus: Invalid key length: %d\n", key_len);
+		return;
+	}
+
 	hfs_bnode_read(node, key, off, key_len);
 }
 
diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c
index 31553372b33a..5b08bd417b28 100644
--- a/fs/iomap/buffered-io.c
+++ b/fs/iomap/buffered-io.c
@@ -259,7 +259,7 @@ static void iomap_adjust_read_range(struct inode *inode, struct folio *folio,
 		}
 
 		/* truncate len if we find any trailing uptodate block(s) */
-		for ( ; i <= last; i++) {
+		while (++i <= last) {
 			if (ifs_block_is_uptodate(ifs, i)) {
 				plen -= (last - i + 1) * block_size;
 				last = i - 1;
diff --git a/fs/isofs/export.c b/fs/isofs/export.c
index 35768a63fb1d..421d247fae52 100644
--- a/fs/isofs/export.c
+++ b/fs/isofs/export.c
@@ -180,7 +180,7 @@ static struct dentry *isofs_fh_to_parent(struct super_block *sb,
 		return NULL;
 
 	return isofs_export_iget(sb,
-			fh_len > 2 ? ifid->parent_block : 0,
+			fh_len > 3 ? ifid->parent_block : 0,
 			ifid->parent_offset,
 			fh_len > 4 ? ifid->parent_generation : 0);
 }
diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c
index 0cf0fddbee81..1467f6790747 100644
--- a/fs/jbd2/revoke.c
+++ b/fs/jbd2/revoke.c
@@ -345,7 +345,8 @@ int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr,
 	bh = bh_in;
 
 	if (!bh) {
-		bh = __find_get_block(bdev, blocknr, journal->j_blocksize);
+		bh = __find_get_block_nonatomic(bdev, blocknr,
+						journal->j_blocksize);
 		if (bh)
 			BUFFER_TRACE(bh, "found on hash");
 	}
@@ -355,7 +356,8 @@ int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr,
 
 		/* If there is a different buffer_head lying around in
 		 * memory anywhere... */
-		bh2 = __find_get_block(bdev, blocknr, journal->j_blocksize);
+		bh2 = __find_get_block_nonatomic(bdev, blocknr,
+						 journal->j_blocksize);
 		if (bh2) {
 			/* ... and it has RevokeValid status... */
 			if (bh2 != bh && buffer_revokevalid(bh2))
@@ -464,7 +466,8 @@ void jbd2_journal_cancel_revoke(handle_t *handle, struct journal_head *jh)
 	 * state machine will get very upset later on. */
 	if (need_cancel) {
 		struct buffer_head *bh2;
-		bh2 = __find_get_block(bh->b_bdev, bh->b_blocknr, bh->b_size);
+		bh2 = __find_get_block_nonatomic(bh->b_bdev, bh->b_blocknr,
+						 bh->b_size);
 		if (bh2) {
 			if (bh2 != bh)
 				clear_buffer_revoked(bh2);
@@ -492,9 +495,9 @@ void jbd2_clear_buffer_revoked_flags(journal_t *journal)
 			struct jbd2_revoke_record_s *record;
 			struct buffer_head *bh;
 			record = (struct jbd2_revoke_record_s *)list_entry;
-			bh = __find_get_block(journal->j_fs_dev,
-					      record->blocknr,
-					      journal->j_blocksize);
+			bh = __find_get_block_nonatomic(journal->j_fs_dev,
+							record->blocknr,
+							journal->j_blocksize);
 			if (bh) {
 				clear_buffer_revoked(bh);
 				__brelse(bh);
diff --git a/fs/namei.c b/fs/namei.c
index 360a86ca1f02..84a0e0b0111c 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -125,9 +125,9 @@
 
 #define EMBEDDED_NAME_MAX	(PATH_MAX - offsetof(struct filename, iname))
 
-static inline void initname(struct filename *name)
+static inline void initname(struct filename *name, const char __user *uptr)
 {
-	name->uptr = NULL;
+	name->uptr = uptr;
 	name->aname = NULL;
 	atomic_set(&name->refcnt, 1);
 }
@@ -210,7 +210,7 @@ getname_flags(const char __user *filename, int flags)
 			return ERR_PTR(-ENAMETOOLONG);
 		}
 	}
-	initname(result);
+	initname(result, filename);
 	audit_getname(result);
 	return result;
 }
@@ -268,7 +268,7 @@ struct filename *getname_kernel(const char * filename)
 		return ERR_PTR(-ENAMETOOLONG);
 	}
 	memcpy((char *)result->name, filename, len);
-	initname(result);
+	initname(result, NULL);
 	audit_getname(result);
 	return result;
 }
@@ -1665,27 +1665,20 @@ static struct dentry *lookup_dcache(const struct qstr *name,
 	return dentry;
 }
 
-/*
- * Parent directory has inode locked exclusive.  This is one
- * and only case when ->lookup() gets called on non in-lookup
- * dentries - as the matter of fact, this only gets called
- * when directory is guaranteed to have no in-lookup children
- * at all.
- * Will return -ENOENT if name isn't found and LOOKUP_CREATE wasn't passed.
- * Will return -EEXIST if name is found and LOOKUP_EXCL was passed.
- */
-struct dentry *lookup_one_qstr_excl(const struct qstr *name,
-				    struct dentry *base,
-				    unsigned int flags)
+static struct dentry *lookup_one_qstr_excl_raw(const struct qstr *name,
+					       struct dentry *base,
+					       unsigned int flags)
 {
-	struct dentry *dentry = lookup_dcache(name, base, flags);
+	struct dentry *dentry;
 	struct dentry *old;
-	struct inode *dir = base->d_inode;
+	struct inode *dir;
 
+	dentry = lookup_dcache(name, base, flags);
 	if (dentry)
-		goto found;
+		return dentry;
 
 	/* Don't create child dentry for a dead directory. */
+	dir = base->d_inode;
 	if (unlikely(IS_DEADDIR(dir)))
 		return ERR_PTR(-ENOENT);
 
@@ -1698,7 +1691,24 @@ struct dentry *lookup_one_qstr_excl(const struct qstr *name,
 		dput(dentry);
 		dentry = old;
 	}
-found:
+	return dentry;
+}
+
+/*
+ * Parent directory has inode locked exclusive.  This is one
+ * and only case when ->lookup() gets called on non in-lookup
+ * dentries - as the matter of fact, this only gets called
+ * when directory is guaranteed to have no in-lookup children
+ * at all.
+ * Will return -ENOENT if name isn't found and LOOKUP_CREATE wasn't passed.
+ * Will return -EEXIST if name is found and LOOKUP_EXCL was passed.
+ */
+struct dentry *lookup_one_qstr_excl(const struct qstr *name,
+				    struct dentry *base, unsigned int flags)
+{
+	struct dentry *dentry;
+
+	dentry = lookup_one_qstr_excl_raw(name, base, flags);
 	if (IS_ERR(dentry))
 		return dentry;
 	if (d_is_negative(dentry) && !(flags & LOOKUP_CREATE)) {
@@ -2742,23 +2752,48 @@ static int filename_parentat(int dfd, struct filename *name,
 /* does lookup, returns the object with parent locked */
 static struct dentry *__kern_path_locked(int dfd, struct filename *name, struct path *path)
 {
+	struct path parent_path __free(path_put) = {};
 	struct dentry *d;
 	struct qstr last;
 	int type, error;
 
-	error = filename_parentat(dfd, name, 0, path, &last, &type);
+	error = filename_parentat(dfd, name, 0, &parent_path, &last, &type);
 	if (error)
 		return ERR_PTR(error);
-	if (unlikely(type != LAST_NORM)) {
-		path_put(path);
+	if (unlikely(type != LAST_NORM))
 		return ERR_PTR(-EINVAL);
+	inode_lock_nested(parent_path.dentry->d_inode, I_MUTEX_PARENT);
+	d = lookup_one_qstr_excl(&last, parent_path.dentry, 0);
+	if (IS_ERR(d)) {
+		inode_unlock(parent_path.dentry->d_inode);
+		return d;
 	}
-	inode_lock_nested(path->dentry->d_inode, I_MUTEX_PARENT);
-	d = lookup_one_qstr_excl(&last, path->dentry, 0);
+	path->dentry = no_free_ptr(parent_path.dentry);
+	path->mnt = no_free_ptr(parent_path.mnt);
+	return d;
+}
+
+struct dentry *kern_path_locked_negative(const char *name, struct path *path)
+{
+	struct path parent_path __free(path_put) = {};
+	struct filename *filename __free(putname) = getname_kernel(name);
+	struct dentry *d;
+	struct qstr last;
+	int type, error;
+
+	error = filename_parentat(AT_FDCWD, filename, 0, &parent_path, &last, &type);
+	if (error)
+		return ERR_PTR(error);
+	if (unlikely(type != LAST_NORM))
+		return ERR_PTR(-EINVAL);
+	inode_lock_nested(parent_path.dentry->d_inode, I_MUTEX_PARENT);
+	d = lookup_one_qstr_excl_raw(&last, parent_path.dentry, 0);
 	if (IS_ERR(d)) {
-		inode_unlock(path->dentry->d_inode);
-		path_put(path);
+		inode_unlock(parent_path.dentry->d_inode);
+		return d;
 	}
+	path->dentry = no_free_ptr(parent_path.dentry);
+	path->mnt = no_free_ptr(parent_path.mnt);
 	return d;
 }
 
diff --git a/fs/namespace.c b/fs/namespace.c
index 14935a0500a2..98a5cd756e9a 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1830,6 +1830,8 @@ static inline void namespace_lock(void)
 	down_write(&namespace_sem);
 }
 
+DEFINE_GUARD(namespace_lock, struct rw_semaphore *, namespace_lock(), namespace_unlock())
+
 enum umount_tree_flags {
 	UMOUNT_SYNC = 1,
 	UMOUNT_PROPAGATE = 2,
@@ -2383,7 +2385,7 @@ void dissolve_on_fput(struct vfsmount *mnt)
 			return;
 	}
 
-	scoped_guard(rwsem_write, &namespace_sem) {
+	scoped_guard(namespace_lock, &namespace_sem) {
 		ns = m->mnt_ns;
 		if (!must_dissolve(ns))
 			return;
@@ -2824,56 +2826,62 @@ static struct mountpoint *do_lock_mount(struct path *path, bool beneath)
 	struct vfsmount *mnt = path->mnt;
 	struct dentry *dentry;
 	struct mountpoint *mp = ERR_PTR(-ENOENT);
+	struct path under = {};
 
 	for (;;) {
-		struct mount *m;
+		struct mount *m = real_mount(mnt);
 
 		if (beneath) {
-			m = real_mount(mnt);
+			path_put(&under);
 			read_seqlock_excl(&mount_lock);
-			dentry = dget(m->mnt_mountpoint);
+			under.mnt = mntget(&m->mnt_parent->mnt);
+			under.dentry = dget(m->mnt_mountpoint);
 			read_sequnlock_excl(&mount_lock);
+			dentry = under.dentry;
 		} else {
 			dentry = path->dentry;
 		}
 
 		inode_lock(dentry->d_inode);
-		if (unlikely(cant_mount(dentry))) {
-			inode_unlock(dentry->d_inode);
-			goto out;
-		}
-
 		namespace_lock();
 
-		if (beneath && (!is_mounted(mnt) || m->mnt_mountpoint != dentry)) {
+		if (unlikely(cant_mount(dentry) || !is_mounted(mnt)))
+			break;		// not to be mounted on
+
+		if (beneath && unlikely(m->mnt_mountpoint != dentry ||
+				        &m->mnt_parent->mnt != under.mnt)) {
 			namespace_unlock();
 			inode_unlock(dentry->d_inode);
-			goto out;
+			continue;	// got moved
 		}
 
 		mnt = lookup_mnt(path);
-		if (likely(!mnt))
+		if (unlikely(mnt)) {
+			namespace_unlock();
+			inode_unlock(dentry->d_inode);
+			path_put(path);
+			path->mnt = mnt;
+			path->dentry = dget(mnt->mnt_root);
+			continue;	// got overmounted
+		}
+		mp = get_mountpoint(dentry);
+		if (IS_ERR(mp))
 			break;
-
-		namespace_unlock();
-		inode_unlock(dentry->d_inode);
-		if (beneath)
-			dput(dentry);
-		path_put(path);
-		path->mnt = mnt;
-		path->dentry = dget(mnt->mnt_root);
-	}
-
-	mp = get_mountpoint(dentry);
-	if (IS_ERR(mp)) {
-		namespace_unlock();
-		inode_unlock(dentry->d_inode);
+		if (beneath) {
+			/*
+			 * @under duplicates the references that will stay
+			 * at least until namespace_unlock(), so the path_put()
+			 * below is safe (and OK to do under namespace_lock -
+			 * we are not dropping the final references here).
+			 */
+			path_put(&under);
+		}
+		return mp;
 	}
-
-out:
+	namespace_unlock();
+	inode_unlock(dentry->d_inode);
 	if (beneath)
-		dput(dentry);
-
+		path_put(&under);
 	return mp;
 }
 
@@ -2884,14 +2892,11 @@ static inline struct mountpoint *lock_mount(struct path *path)
 
 static void unlock_mount(struct mountpoint *where)
 {
-	struct dentry *dentry = where->m_dentry;
-
+	inode_unlock(where->m_dentry->d_inode);
 	read_seqlock_excl(&mount_lock);
 	put_mountpoint(where);
 	read_sequnlock_excl(&mount_lock);
-
 	namespace_unlock();
-	inode_unlock(dentry->d_inode);
 }
 
 static int graft_tree(struct mount *mnt, struct mount *p, struct mountpoint *mp)
@@ -5189,8 +5194,8 @@ static void finish_mount_kattr(struct mount_kattr *kattr)
 		mnt_idmap_put(kattr->mnt_idmap);
 }
 
-static int copy_mount_setattr(struct mount_attr __user *uattr, size_t usize,
-			      struct mount_kattr *kattr)
+static int wants_mount_setattr(struct mount_attr __user *uattr, size_t usize,
+			       struct mount_kattr *kattr)
 {
 	int ret;
 	struct mount_attr attr;
@@ -5213,9 +5218,13 @@ static int copy_mount_setattr(struct mount_attr __user *uattr, size_t usize,
 	if (attr.attr_set == 0 &&
 	    attr.attr_clr == 0 &&
 	    attr.propagation == 0)
-		return 0;
+		return 0; /* Tell caller to not bother. */
 
-	return build_mount_kattr(&attr, usize, kattr);
+	ret = build_mount_kattr(&attr, usize, kattr);
+	if (ret < 0)
+		return ret;
+
+	return 1;
 }
 
 SYSCALL_DEFINE5(mount_setattr, int, dfd, const char __user *, path,
@@ -5247,8 +5256,8 @@ SYSCALL_DEFINE5(mount_setattr, int, dfd, const char __user *, path,
 	if (flags & AT_RECURSIVE)
 		kattr.kflags |= MOUNT_KATTR_RECURSE;
 
-	err = copy_mount_setattr(uattr, usize, &kattr);
-	if (err)
+	err = wants_mount_setattr(uattr, usize, &kattr);
+	if (err <= 0)
 		return err;
 
 	err = user_path_at(dfd, path, kattr.lookup_flags, &target);
@@ -5282,15 +5291,17 @@ SYSCALL_DEFINE5(open_tree_attr, int, dfd, const char __user *, filename,
 		if (flags & AT_RECURSIVE)
 			kattr.kflags |= MOUNT_KATTR_RECURSE;
 
-		ret = copy_mount_setattr(uattr, usize, &kattr);
-		if (ret)
+		ret = wants_mount_setattr(uattr, usize, &kattr);
+		if (ret < 0)
 			return ret;
 
-		ret = do_mount_setattr(&file->f_path, &kattr);
-		if (ret)
-			return ret;
+		if (ret) {
+			ret = do_mount_setattr(&file->f_path, &kattr);
+			if (ret)
+				return ret;
 
-		finish_mount_kattr(&kattr);
+			finish_mount_kattr(&kattr);
+		}
 	}
 
 	fd = get_unused_fd_flags(flags & O_CLOEXEC);
diff --git a/fs/netfs/fscache_cache.c b/fs/netfs/fscache_cache.c
index 9397ed39b0b4..8f70f8da064b 100644
--- a/fs/netfs/fscache_cache.c
+++ b/fs/netfs/fscache_cache.c
@@ -372,7 +372,7 @@ void fscache_withdraw_cache(struct fscache_cache *cache)
 EXPORT_SYMBOL(fscache_withdraw_cache);
 
 #ifdef CONFIG_PROC_FS
-static const char fscache_cache_states[NR__FSCACHE_CACHE_STATE] = "-PAEW";
+static const char fscache_cache_states[NR__FSCACHE_CACHE_STATE] __nonstring = "-PAEW";
 
 /*
  * Generate a list of caches in /proc/fs/fscache/caches
diff --git a/fs/netfs/fscache_cookie.c b/fs/netfs/fscache_cookie.c
index d4d4b3a8b106..3d56fc73435f 100644
--- a/fs/netfs/fscache_cookie.c
+++ b/fs/netfs/fscache_cookie.c
@@ -29,7 +29,7 @@ static LIST_HEAD(fscache_cookie_lru);
 static DEFINE_SPINLOCK(fscache_cookie_lru_lock);
 DEFINE_TIMER(fscache_cookie_lru_timer, fscache_cookie_lru_timed_out);
 static DECLARE_WORK(fscache_cookie_lru_work, fscache_cookie_lru_worker);
-static const char fscache_cookie_states[FSCACHE_COOKIE_STATE__NR] = "-LCAIFUWRD";
+static const char fscache_cookie_states[FSCACHE_COOKIE_STATE__NR] __nonstring = "-LCAIFUWRD";
 static unsigned int fscache_lru_cookie_timeout = 10 * HZ;
 
 void fscache_print_cookie(struct fscache_cookie *cookie, char prefix)
diff --git a/fs/netfs/main.c b/fs/netfs/main.c
index 4e3e62040831..70ecc8f5f210 100644
--- a/fs/netfs/main.c
+++ b/fs/netfs/main.c
@@ -127,11 +127,13 @@ static int __init netfs_init(void)
 	if (mempool_init_slab_pool(&netfs_subrequest_pool, 100, netfs_subrequest_slab) < 0)
 		goto error_subreqpool;
 
+#ifdef CONFIG_PROC_FS
 	if (!proc_mkdir("fs/netfs", NULL))
 		goto error_proc;
 	if (!proc_create_seq("fs/netfs/requests", S_IFREG | 0444, NULL,
 			     &netfs_requests_seq_ops))
 		goto error_procfile;
+#endif
 #ifdef CONFIG_FSCACHE_STATS
 	if (!proc_create_single("fs/netfs/stats", S_IFREG | 0444, NULL,
 				netfs_stats_show))
@@ -144,9 +146,11 @@ static int __init netfs_init(void)
 	return 0;
 
 error_fscache:
+#ifdef CONFIG_PROC_FS
 error_procfile:
 	remove_proc_subtree("fs/netfs", NULL);
 error_proc:
+#endif
 	mempool_exit(&netfs_subrequest_pool);
 error_subreqpool:
 	kmem_cache_destroy(netfs_subrequest_slab);
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index d3f76101ad4b..07932ce9246c 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -2,6 +2,7 @@
 config NFS_FS
 	tristate "NFS client support"
 	depends on INET && FILE_LOCKING && MULTIUSER
+	select CRC32
 	select LOCKD
 	select SUNRPC
 	select NFS_COMMON
@@ -196,7 +197,6 @@ config NFS_USE_KERNEL_DNS
 config NFS_DEBUG
 	bool
 	depends on NFS_FS && SUNRPC_DEBUG
-	select CRC32
 	default y
 
 config NFS_DISABLE_UDP_SUPPORT
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index ec8d32d0e2e9..6655e5f32ec6 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -899,18 +899,11 @@ u64 nfs_timespec_to_change_attr(const struct timespec64 *ts)
 	return ((u64)ts->tv_sec << 30) + ts->tv_nsec;
 }
 
-#ifdef CONFIG_CRC32
 static inline u32 nfs_stateid_hash(const nfs4_stateid *stateid)
 {
 	return ~crc32_le(0xFFFFFFFF, &stateid->other[0],
 				NFS4_STATEID_OTHER_SIZE);
 }
-#else
-static inline u32 nfs_stateid_hash(nfs4_stateid *stateid)
-{
-	return 0;
-}
-#endif
 
 static inline bool nfs_current_task_exiting(void)
 {
diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h
index 351616c61df5..f9c291e2165c 100644
--- a/fs/nfs/nfs4session.h
+++ b/fs/nfs/nfs4session.h
@@ -148,16 +148,12 @@ static inline void nfs4_copy_sessionid(struct nfs4_sessionid *dst,
 	memcpy(dst->data, src->data, NFS4_MAX_SESSIONID_LEN);
 }
 
-#ifdef CONFIG_CRC32
 /*
  * nfs_session_id_hash - calculate the crc32 hash for the session id
  * @session - pointer to session
  */
 #define nfs_session_id_hash(sess_id) \
 	(~crc32_le(0xFFFFFFFF, &(sess_id)->data[0], sizeof((sess_id)->data)))
-#else
-#define nfs_session_id_hash(session) (0)
-#endif
 #else /* defined(CONFIG_NFS_V4_1) */
 
 static inline int nfs4_init_session(struct nfs_client *clp)
diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index 792d3fed1b45..731a88f6313e 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -4,6 +4,7 @@ config NFSD
 	depends on INET
 	depends on FILE_LOCKING
 	depends on FSNOTIFY
+	select CRC32
 	select LOCKD
 	select SUNRPC
 	select EXPORTFS
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 2041268b398a..59a693f22452 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -5430,7 +5430,7 @@ static void nfsd_break_one_deleg(struct nfs4_delegation *dp)
 	queued = nfsd4_run_cb(&dp->dl_recall);
 	WARN_ON_ONCE(!queued);
 	if (!queued)
-		nfs4_put_stid(&dp->dl_stid);
+		refcount_dec(&dp->dl_stid.sc_count);
 }
 
 /* Called from break_lease() with flc_lock held. */
diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h
index 876152a91f12..5103c2f4d225 100644
--- a/fs/nfsd/nfsfh.h
+++ b/fs/nfsd/nfsfh.h
@@ -267,7 +267,6 @@ static inline bool fh_fsid_match(const struct knfsd_fh *fh1,
 	return true;
 }
 
-#ifdef CONFIG_CRC32
 /**
  * knfsd_fh_hash - calculate the crc32 hash for the filehandle
  * @fh - pointer to filehandle
@@ -279,12 +278,6 @@ static inline u32 knfsd_fh_hash(const struct knfsd_fh *fh)
 {
 	return ~crc32_le(0xFFFFFFFF, fh->fh_raw, fh->fh_size);
 }
-#else
-static inline u32 knfsd_fh_hash(const struct knfsd_fh *fh)
-{
-	return 0;
-}
-#endif
 
 /**
  * fh_clear_pre_post_attrs - Reset pre/post attributes
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index f2d840ae4ded..87f861e9004f 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -1961,12 +1961,7 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask,
 		return -EINVAL;
 
 	if (mark_cmd == FAN_MARK_FLUSH) {
-		if (mark_type == FAN_MARK_MOUNT)
-			fsnotify_clear_vfsmount_marks_by_group(group);
-		else if (mark_type == FAN_MARK_FILESYSTEM)
-			fsnotify_clear_sb_marks_by_group(group);
-		else
-			fsnotify_clear_inode_marks_by_group(group);
+		fsnotify_clear_marks_by_group(group, obj_type);
 		return 0;
 	}
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index f1b4b3e611cb..c7a9729dc9d0 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1249,7 +1249,7 @@ static int ocfs2_force_read_journal(struct inode *inode)
 		}
 
 		for (i = 0; i < p_blocks; i++, p_blkno++) {
-			bh = __find_get_block(osb->sb->s_bdev, p_blkno,
+			bh = __find_get_block_nonatomic(osb->sb->s_bdev, p_blkno,
 					osb->sb->s_blocksize);
 			/* block not cached. */
 			if (!bh)
diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h
index 6f2f8f4cfbbc..aef942a758ce 100644
--- a/fs/overlayfs/overlayfs.h
+++ b/fs/overlayfs/overlayfs.h
@@ -541,8 +541,6 @@ int ovl_set_metacopy_xattr(struct ovl_fs *ofs, struct dentry *d,
 bool ovl_is_metacopy_dentry(struct dentry *dentry);
 char *ovl_get_redirect_xattr(struct ovl_fs *ofs, const struct path *path, int padding);
 int ovl_ensure_verity_loaded(struct path *path);
-int ovl_get_verity_xattr(struct ovl_fs *ofs, const struct path *path,
-			 u8 *digest_buf, int *buf_length);
 int ovl_validate_verity(struct ovl_fs *ofs,
 			struct path *metapath,
 			struct path *datapath);
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index b63474d1b064..e19940d649ca 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -1138,6 +1138,11 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb,
 		return ERR_PTR(-EINVAL);
 	}
 
+	if (ctx->nr == ctx->nr_data) {
+		pr_err("at least one non-data lowerdir is required\n");
+		return ERR_PTR(-EINVAL);
+	}
+
 	err = -EINVAL;
 	for (i = 0; i < ctx->nr; i++) {
 		l = &ctx->lower[i];
diff --git a/fs/smb/client/cifsencrypt.c b/fs/smb/client/cifsencrypt.c
index e69968e88fe7..35892df7335c 100644
--- a/fs/smb/client/cifsencrypt.c
+++ b/fs/smb/client/cifsencrypt.c
@@ -704,18 +704,12 @@ cifs_crypto_secmech_release(struct TCP_Server_Info *server)
 	cifs_free_hash(&server->secmech.md5);
 	cifs_free_hash(&server->secmech.sha512);
 
-	if (!SERVER_IS_CHAN(server)) {
-		if (server->secmech.enc) {
-			crypto_free_aead(server->secmech.enc);
-			server->secmech.enc = NULL;
-		}
-
-		if (server->secmech.dec) {
-			crypto_free_aead(server->secmech.dec);
-			server->secmech.dec = NULL;
-		}
-	} else {
+	if (server->secmech.enc) {
+		crypto_free_aead(server->secmech.enc);
 		server->secmech.enc = NULL;
+	}
+	if (server->secmech.dec) {
+		crypto_free_aead(server->secmech.dec);
 		server->secmech.dec = NULL;
 	}
 }
diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h
index 07c4688ec4c9..3b32116b0b49 100644
--- a/fs/smb/client/cifsglob.h
+++ b/fs/smb/client/cifsglob.h
@@ -625,10 +625,8 @@ struct smb_version_operations {
 	bool (*is_status_io_timeout)(char *buf);
 	/* Check for STATUS_NETWORK_NAME_DELETED */
 	bool (*is_network_name_deleted)(char *buf, struct TCP_Server_Info *srv);
-	int (*parse_reparse_point)(struct cifs_sb_info *cifs_sb,
-				   const char *full_path,
-				   struct kvec *rsp_iov,
-				   struct cifs_open_info_data *data);
+	struct reparse_data_buffer * (*get_reparse_point_buffer)(const struct kvec *rsp_iov,
+								 u32 *plen);
 	int (*create_reparse_symlink)(const unsigned int xid,
 				      struct inode *inode,
 				      struct dentry *dentry,
diff --git a/fs/smb/client/cifspdu.h b/fs/smb/client/cifspdu.h
index 48d0d6f439cf..1b79fe07476f 100644
--- a/fs/smb/client/cifspdu.h
+++ b/fs/smb/client/cifspdu.h
@@ -1266,10 +1266,9 @@ typedef struct smb_com_query_information_rsp {
 typedef struct smb_com_setattr_req {
 	struct smb_hdr hdr; /* wct = 8 */
 	__le16 attr;
-	__le16 time_low;
-	__le16 time_high;
+	__le32 last_write_time;
 	__le16 reserved[5]; /* must be zero */
-	__u16  ByteCount;
+	__le16 ByteCount;
 	__u8   BufferFormat; /* 4 = ASCII */
 	unsigned char fileName[];
 } __attribute__((packed)) SETATTR_REQ;
@@ -2256,6 +2255,8 @@ typedef struct {
 #define FILE_SUPPORTS_ENCRYPTION	0x00020000
 #define FILE_SUPPORTS_OBJECT_IDS	0x00010000
 #define FILE_VOLUME_IS_COMPRESSED	0x00008000
+#define FILE_SUPPORTS_POSIX_UNLINK_RENAME 0x00000400
+#define FILE_RETURNS_CLEANUP_RESULT_INFO  0x00000200
 #define FILE_SUPPORTS_REMOTE_STORAGE	0x00000100
 #define FILE_SUPPORTS_REPARSE_POINTS	0x00000080
 #define FILE_SUPPORTS_SPARSE_FILES	0x00000040
diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h
index cfcc07905bdf..ecf774a8f1ca 100644
--- a/fs/smb/client/cifsproto.h
+++ b/fs/smb/client/cifsproto.h
@@ -163,6 +163,8 @@ extern int cifs_get_writable_path(struct cifs_tcon *tcon, const char *name,
 extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool);
 extern int cifs_get_readable_path(struct cifs_tcon *tcon, const char *name,
 				  struct cifsFileInfo **ret_file);
+extern int cifs_get_hardlink_path(struct cifs_tcon *tcon, struct inode *inode,
+				  struct file *file);
 extern unsigned int smbCalcSize(void *buf);
 extern int decode_negTokenInit(unsigned char *security_blob, int length,
 			struct TCP_Server_Info *server);
@@ -393,6 +395,10 @@ extern int CIFSSMBQFSUnixInfo(const unsigned int xid, struct cifs_tcon *tcon);
 extern int CIFSSMBQFSPosixInfo(const unsigned int xid, struct cifs_tcon *tcon,
 			struct kstatfs *FSData);
 
+extern int SMBSetInformation(const unsigned int xid, struct cifs_tcon *tcon,
+			     const char *fileName, __le32 attributes, __le64 write_time,
+			     const struct nls_table *nls_codepage,
+			     struct cifs_sb_info *cifs_sb);
 extern int CIFSSMBSetPathInfo(const unsigned int xid, struct cifs_tcon *tcon,
 			const char *fileName, const FILE_BASIC_INFO *data,
 			const struct nls_table *nls_codepage,
diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c
index 60cb264a01e5..f55457b4b82e 100644
--- a/fs/smb/client/cifssmb.c
+++ b/fs/smb/client/cifssmb.c
@@ -5171,6 +5171,63 @@ CIFSSMBSetFileSize(const unsigned int xid, struct cifs_tcon *tcon,
 	return rc;
 }
 
+int
+SMBSetInformation(const unsigned int xid, struct cifs_tcon *tcon,
+		  const char *fileName, __le32 attributes, __le64 write_time,
+		  const struct nls_table *nls_codepage,
+		  struct cifs_sb_info *cifs_sb)
+{
+	SETATTR_REQ *pSMB;
+	SETATTR_RSP *pSMBr;
+	struct timespec64 ts;
+	int bytes_returned;
+	int name_len;
+	int rc;
+
+	cifs_dbg(FYI, "In %s path %s\n", __func__, fileName);
+
+retry:
+	rc = smb_init(SMB_COM_SETATTR, 8, tcon, (void **) &pSMB,
+		      (void **) &pSMBr);
+	if (rc)
+		return rc;
+
+	if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) {
+		name_len =
+			cifsConvertToUTF16((__le16 *) pSMB->fileName,
+					   fileName, PATH_MAX, nls_codepage,
+					   cifs_remap(cifs_sb));
+		name_len++;     /* trailing null */
+		name_len *= 2;
+	} else {
+		name_len = copy_path_name(pSMB->fileName, fileName);
+	}
+	/* Only few attributes can be set by this command, others are not accepted by Win9x. */
+	pSMB->attr = cpu_to_le16(le32_to_cpu(attributes) &
+			(ATTR_READONLY | ATTR_HIDDEN | ATTR_SYSTEM | ATTR_ARCHIVE));
+	/* Zero write time value (in both NT and SETATTR formats) means to not change it. */
+	if (le64_to_cpu(write_time) != 0) {
+		ts = cifs_NTtimeToUnix(write_time);
+		pSMB->last_write_time = cpu_to_le32(ts.tv_sec);
+	}
+	pSMB->BufferFormat = 0x04;
+	name_len++; /* account for buffer type byte */
+	inc_rfc1001_len(pSMB, (__u16)name_len);
+	pSMB->ByteCount = cpu_to_le16(name_len);
+
+	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
+			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
+	if (rc)
+		cifs_dbg(FYI, "Send error in %s = %d\n", __func__, rc);
+
+	cifs_buf_release(pSMB);
+
+	if (rc == -EAGAIN)
+		goto retry;
+
+	return rc;
+}
+
 /* Some legacy servers such as NT4 require that the file times be set on
    an open handle, rather than by pathname - this is awkward due to
    potential access conflicts on the open, but it is unavoidable for these
diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c
index f298e86a3c1f..6bf04d9a5491 100644
--- a/fs/smb/client/connect.c
+++ b/fs/smb/client/connect.c
@@ -300,7 +300,6 @@ cifs_abort_connection(struct TCP_Server_Info *server)
 			 server->ssocket->flags);
 		sock_release(server->ssocket);
 		server->ssocket = NULL;
-		put_net(cifs_net_ns(server));
 	}
 	server->sequence_number = 0;
 	server->session_estab = false;
@@ -1074,13 +1073,9 @@ clean_demultiplex_info(struct TCP_Server_Info *server)
 	msleep(125);
 	if (cifs_rdma_enabled(server))
 		smbd_destroy(server);
-
 	if (server->ssocket) {
 		sock_release(server->ssocket);
 		server->ssocket = NULL;
-
-		/* Release netns reference for the socket. */
-		put_net(cifs_net_ns(server));
 	}
 
 	if (!list_empty(&server->pending_mid_q)) {
@@ -1128,7 +1123,6 @@ clean_demultiplex_info(struct TCP_Server_Info *server)
 		 */
 	}
 
-	/* Release netns reference for this server. */
 	put_net(cifs_net_ns(server));
 	kfree(server->leaf_fullpath);
 	kfree(server->hostname);
@@ -1774,8 +1768,6 @@ cifs_get_tcp_session(struct smb3_fs_context *ctx,
 
 	tcp_ses->ops = ctx->ops;
 	tcp_ses->vals = ctx->vals;
-
-	/* Grab netns reference for this server. */
 	cifs_set_net_ns(tcp_ses, get_net(current->nsproxy->net_ns));
 
 	tcp_ses->sign = ctx->sign;
@@ -1903,7 +1895,6 @@ smbd_connected:
 out_err_crypto_release:
 	cifs_crypto_secmech_release(tcp_ses);
 
-	/* Release netns reference for this server. */
 	put_net(cifs_net_ns(tcp_ses));
 
 out_err:
@@ -1912,10 +1903,8 @@ out_err:
 			cifs_put_tcp_session(tcp_ses->primary_server, false);
 		kfree(tcp_ses->hostname);
 		kfree(tcp_ses->leaf_fullpath);
-		if (tcp_ses->ssocket) {
+		if (tcp_ses->ssocket)
 			sock_release(tcp_ses->ssocket);
-			put_net(cifs_net_ns(tcp_ses));
-		}
 		kfree(tcp_ses);
 	}
 	return ERR_PTR(rc);
@@ -2556,6 +2545,8 @@ static int match_tcon(struct cifs_tcon *tcon, struct smb3_fs_context *ctx)
 		return 0;
 	if (tcon->nodelete != ctx->nodelete)
 		return 0;
+	if (tcon->posix_extensions != ctx->linux_ext)
+		return 0;
 	return 1;
 }
 
@@ -3357,24 +3348,20 @@ generic_ip_connect(struct TCP_Server_Info *server)
 		socket = server->ssocket;
 	} else {
 		struct net *net = cifs_net_ns(server);
+		struct sock *sk;
 
-		rc = sock_create_kern(net, sfamily, SOCK_STREAM, IPPROTO_TCP, &server->ssocket);
+		rc = __sock_create(net, sfamily, SOCK_STREAM,
+				   IPPROTO_TCP, &server->ssocket, 1);
 		if (rc < 0) {
 			cifs_server_dbg(VFS, "Error %d creating socket\n", rc);
 			return rc;
 		}
 
-		/*
-		 * Grab netns reference for the socket.
-		 *
-		 * This reference will be released in several situations:
-		 * - In the failure path before the cifsd thread is started.
-		 * - In the all place where server->socket is released, it is
-		 *   also set to NULL.
-		 * - Ultimately in clean_demultiplex_info(), during the final
-		 *   teardown.
-		 */
-		get_net(net);
+		sk = server->ssocket->sk;
+		__netns_tracker_free(net, &sk->ns_tracker, false);
+		sk->sk_net_refcnt = 1;
+		get_net_track(net, &sk->ns_tracker, GFP_KERNEL);
+		sock_inuse_add(net, 1);
 
 		/* BB other socket options to set KEEPALIVE, NODELAY? */
 		cifs_dbg(FYI, "Socket created\n");
@@ -3426,7 +3413,6 @@ generic_ip_connect(struct TCP_Server_Info *server)
 	if (rc < 0) {
 		cifs_dbg(FYI, "Error %d connecting to server\n", rc);
 		trace_smb3_connect_err(server->hostname, server->conn_id, &server->dstaddr, rc);
-		put_net(cifs_net_ns(server));
 		sock_release(socket);
 		server->ssocket = NULL;
 		return rc;
@@ -3767,28 +3753,7 @@ int cifs_mount_get_tcon(struct cifs_mount_ctx *mnt_ctx)
 		}
 	}
 
-	/*
-	 * Clamp the rsize/wsize mount arguments if they are too big for the server
-	 * and set the rsize/wsize to the negotiated values if not passed in by
-	 * the user on mount
-	 */
-	if ((cifs_sb->ctx->wsize == 0) ||
-	    (cifs_sb->ctx->wsize > server->ops->negotiate_wsize(tcon, ctx))) {
-		cifs_sb->ctx->wsize =
-			round_down(server->ops->negotiate_wsize(tcon, ctx), PAGE_SIZE);
-		/*
-		 * in the very unlikely event that the server sent a max write size under PAGE_SIZE,
-		 * (which would get rounded down to 0) then reset wsize to absolute minimum eg 4096
-		 */
-		if (cifs_sb->ctx->wsize == 0) {
-			cifs_sb->ctx->wsize = PAGE_SIZE;
-			cifs_dbg(VFS, "wsize too small, reset to minimum ie PAGE_SIZE, usually 4096\n");
-		}
-	}
-	if ((cifs_sb->ctx->rsize == 0) ||
-	    (cifs_sb->ctx->rsize > server->ops->negotiate_rsize(tcon, ctx)))
-		cifs_sb->ctx->rsize = server->ops->negotiate_rsize(tcon, ctx);
-
+	cifs_negotiate_iosize(server, cifs_sb->ctx, tcon);
 	/*
 	 * The cookie is initialized from volume info returned above.
 	 * Inside cifs_fscache_get_super_cookie it checks
diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c
index 8407fb108664..851b74f557c1 100644
--- a/fs/smb/client/file.c
+++ b/fs/smb/client/file.c
@@ -160,10 +160,8 @@ static int cifs_prepare_read(struct netfs_io_subrequest *subreq)
 	server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses);
 	rdata->server = server;
 
-	if (cifs_sb->ctx->rsize == 0)
-		cifs_sb->ctx->rsize =
-			server->ops->negotiate_rsize(tlink_tcon(req->cfile->tlink),
-						     cifs_sb->ctx);
+	cifs_negotiate_rsize(server, cifs_sb->ctx,
+			     tlink_tcon(req->cfile->tlink));
 
 	rc = server->ops->wait_mtu_credits(server, cifs_sb->ctx->rsize,
 					   &size, &rdata->credits);
@@ -1007,6 +1005,11 @@ int cifs_open(struct inode *inode, struct file *file)
 		} else {
 			_cifsFileInfo_put(cfile, true, false);
 		}
+	} else {
+		/* hard link on the defeered close file */
+		rc = cifs_get_hardlink_path(tcon, inode, file);
+		if (rc)
+			cifs_close_deferred_file(CIFS_I(inode));
 	}
 
 	if (server->oplocks)
@@ -2071,6 +2074,29 @@ cifs_move_llist(struct list_head *source, struct list_head *dest)
 		list_move(li, dest);
 }
 
+int
+cifs_get_hardlink_path(struct cifs_tcon *tcon, struct inode *inode,
+				struct file *file)
+{
+	struct cifsFileInfo *open_file = NULL;
+	struct cifsInodeInfo *cinode = CIFS_I(inode);
+	int rc = 0;
+
+	spin_lock(&tcon->open_file_lock);
+	spin_lock(&cinode->open_file_lock);
+
+	list_for_each_entry(open_file, &cinode->openFileList, flist) {
+		if (file->f_flags == open_file->f_flags) {
+			rc = -EINVAL;
+			break;
+		}
+	}
+
+	spin_unlock(&cinode->open_file_lock);
+	spin_unlock(&tcon->open_file_lock);
+	return rc;
+}
+
 void
 cifs_free_llist(struct list_head *llist)
 {
diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c
index 2980941b9667..a634a34d4086 100644
--- a/fs/smb/client/fs_context.c
+++ b/fs/smb/client/fs_context.c
@@ -1021,6 +1021,7 @@ static int smb3_reconfigure(struct fs_context *fc)
 	struct dentry *root = fc->root;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(root->d_sb);
 	struct cifs_ses *ses = cifs_sb_master_tcon(cifs_sb)->ses;
+	unsigned int rsize = ctx->rsize, wsize = ctx->wsize;
 	char *new_password = NULL, *new_password2 = NULL;
 	bool need_recon = false;
 	int rc;
@@ -1103,11 +1104,8 @@ static int smb3_reconfigure(struct fs_context *fc)
 	STEAL_STRING(cifs_sb, ctx, iocharset);
 
 	/* if rsize or wsize not passed in on remount, use previous values */
-	if (ctx->rsize == 0)
-		ctx->rsize = cifs_sb->ctx->rsize;
-	if (ctx->wsize == 0)
-		ctx->wsize = cifs_sb->ctx->wsize;
-
+	ctx->rsize = rsize ? CIFS_ALIGN_RSIZE(fc, rsize) : cifs_sb->ctx->rsize;
+	ctx->wsize = wsize ? CIFS_ALIGN_WSIZE(fc, wsize) : cifs_sb->ctx->wsize;
 
 	smb3_cleanup_fs_context_contents(cifs_sb->ctx);
 	rc = smb3_fs_context_dup(cifs_sb->ctx, ctx);
@@ -1312,7 +1310,7 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 				__func__);
 			goto cifs_parse_mount_err;
 		}
-		ctx->bsize = result.uint_32;
+		ctx->bsize = CIFS_ALIGN_BSIZE(fc, result.uint_32);
 		ctx->got_bsize = true;
 		break;
 	case Opt_rasize:
@@ -1336,24 +1334,13 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 		ctx->rasize = result.uint_32;
 		break;
 	case Opt_rsize:
-		ctx->rsize = result.uint_32;
+		ctx->rsize = CIFS_ALIGN_RSIZE(fc, result.uint_32);
 		ctx->got_rsize = true;
 		ctx->vol_rsize = ctx->rsize;
 		break;
 	case Opt_wsize:
-		ctx->wsize = result.uint_32;
+		ctx->wsize = CIFS_ALIGN_WSIZE(fc, result.uint_32);
 		ctx->got_wsize = true;
-		if (ctx->wsize % PAGE_SIZE != 0) {
-			ctx->wsize = round_down(ctx->wsize, PAGE_SIZE);
-			if (ctx->wsize == 0) {
-				ctx->wsize = PAGE_SIZE;
-				cifs_dbg(VFS, "wsize too small, reset to minimum %ld\n", PAGE_SIZE);
-			} else {
-				cifs_dbg(VFS,
-					 "wsize rounded down to %d to multiple of PAGE_SIZE %ld\n",
-					 ctx->wsize, PAGE_SIZE);
-			}
-		}
 		ctx->vol_wsize = ctx->wsize;
 		break;
 	case Opt_acregmax:
diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h
index d1d29249bcdb..9e83302ce4b8 100644
--- a/fs/smb/client/fs_context.h
+++ b/fs/smb/client/fs_context.h
@@ -20,6 +20,21 @@
 		cifs_dbg(VFS, fmt, ## __VA_ARGS__);	\
 	} while (0)
 
+static inline size_t cifs_io_align(struct fs_context *fc,
+				   const char *name, size_t size)
+{
+	if (!size || !IS_ALIGNED(size, PAGE_SIZE)) {
+		cifs_errorf(fc, "unaligned %s, making it a multiple of %lu bytes\n",
+			    name, PAGE_SIZE);
+		size = umax(round_down(size, PAGE_SIZE), PAGE_SIZE);
+	}
+	return size;
+}
+
+#define CIFS_ALIGN_WSIZE(_fc, _size) cifs_io_align(_fc, "wsize", _size)
+#define CIFS_ALIGN_RSIZE(_fc, _size) cifs_io_align(_fc, "rsize", _size)
+#define CIFS_ALIGN_BSIZE(_fc, _size) cifs_io_align(_fc, "bsize", _size)
+
 enum smb_version {
 	Smb_1 = 1,
 	Smb_20,
@@ -361,4 +376,36 @@ static inline void cifs_mount_unlock(void)
 	mutex_unlock(&cifs_mount_mutex);
 }
 
+static inline void cifs_negotiate_rsize(struct TCP_Server_Info *server,
+					struct smb3_fs_context *ctx,
+					struct cifs_tcon *tcon)
+{
+	unsigned int size;
+
+	size = umax(server->ops->negotiate_rsize(tcon, ctx), PAGE_SIZE);
+	if (ctx->rsize)
+		size = umax(umin(ctx->rsize, size), PAGE_SIZE);
+	ctx->rsize = round_down(size, PAGE_SIZE);
+}
+
+static inline void cifs_negotiate_wsize(struct TCP_Server_Info *server,
+					struct smb3_fs_context *ctx,
+					struct cifs_tcon *tcon)
+{
+	unsigned int size;
+
+	size = umax(server->ops->negotiate_wsize(tcon, ctx), PAGE_SIZE);
+	if (ctx->wsize)
+		size = umax(umin(ctx->wsize, size), PAGE_SIZE);
+	ctx->wsize = round_down(size, PAGE_SIZE);
+}
+
+static inline void cifs_negotiate_iosize(struct TCP_Server_Info *server,
+					 struct smb3_fs_context *ctx,
+					 struct cifs_tcon *tcon)
+{
+	cifs_negotiate_rsize(server, ctx, tcon);
+	cifs_negotiate_wsize(server, ctx, tcon);
+}
+
 #endif
diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c
index a00a9d91d0da..75be4b46bc6f 100644
--- a/fs/smb/client/inode.c
+++ b/fs/smb/client/inode.c
@@ -1203,18 +1203,17 @@ static int reparse_info_to_fattr(struct cifs_open_info_data *data,
 			goto out;
 		}
 		break;
-	case IO_REPARSE_TAG_MOUNT_POINT:
-		cifs_create_junction_fattr(fattr, sb);
-		rc = 0;
-		goto out;
 	default:
 		/* Check for cached reparse point data */
 		if (data->symlink_target || data->reparse.buf) {
 			rc = 0;
-		} else if (iov && server->ops->parse_reparse_point) {
-			rc = server->ops->parse_reparse_point(cifs_sb,
-							      full_path,
-							      iov, data);
+		} else if (iov && server->ops->get_reparse_point_buffer) {
+			struct reparse_data_buffer *reparse_buf;
+			u32 reparse_len;
+
+			reparse_buf = server->ops->get_reparse_point_buffer(iov, &reparse_len);
+			rc = parse_reparse_point(reparse_buf, reparse_len,
+						 cifs_sb, full_path, data);
 			/*
 			 * If the reparse point was not handled but it is the
 			 * name surrogate which points to directory, then treat
@@ -1228,6 +1227,16 @@ static int reparse_info_to_fattr(struct cifs_open_info_data *data,
 				cifs_create_junction_fattr(fattr, sb);
 				goto out;
 			}
+			/*
+			 * If the reparse point is unsupported by the Linux SMB
+			 * client then let it process by the SMB server. So mask
+			 * the -EOPNOTSUPP error code. This will allow Linux SMB
+			 * client to send SMB OPEN request to server. If server
+			 * does not support this reparse point too then server
+			 * will return error during open the path.
+			 */
+			if (rc == -EOPNOTSUPP)
+				rc = 0;
 		}
 
 		if (data->reparse.tag == IO_REPARSE_TAG_SYMLINK && !rc) {
diff --git a/fs/smb/client/reparse.c b/fs/smb/client/reparse.c
index 2b9e9885dc42..bb25e77c5540 100644
--- a/fs/smb/client/reparse.c
+++ b/fs/smb/client/reparse.c
@@ -542,12 +542,12 @@ static int wsl_set_reparse_buf(struct reparse_data_buffer **buf,
 			kfree(symname_utf16);
 			return -ENOMEM;
 		}
-		/* Flag 0x02000000 is unknown, but all wsl symlinks have this value */
-		symlink_buf->Flags = cpu_to_le32(0x02000000);
-		/* PathBuffer is in UTF-8 but without trailing null-term byte */
+		/* Version field must be set to 2 (MS-FSCC 2.1.2.7) */
+		symlink_buf->Version = cpu_to_le32(2);
+		/* Target for Version 2 is in UTF-8 but without trailing null-term byte */
 		symname_utf8_len = utf16s_to_utf8s((wchar_t *)symname_utf16, symname_utf16_len/2,
 						   UTF16_LITTLE_ENDIAN,
-						   symlink_buf->PathBuffer,
+						   symlink_buf->Target,
 						   symname_utf8_maxlen);
 		*buf = (struct reparse_data_buffer *)symlink_buf;
 		buf_len = sizeof(struct reparse_wsl_symlink_data_buffer) + symname_utf8_len;
@@ -1016,29 +1016,36 @@ static int parse_reparse_wsl_symlink(struct reparse_wsl_symlink_data_buffer *buf
 				     struct cifs_open_info_data *data)
 {
 	int len = le16_to_cpu(buf->ReparseDataLength);
+	int data_offset = offsetof(typeof(*buf), Target) - offsetof(typeof(*buf), Version);
 	int symname_utf8_len;
 	__le16 *symname_utf16;
 	int symname_utf16_len;
 
-	if (len <= sizeof(buf->Flags)) {
+	if (len <= data_offset) {
 		cifs_dbg(VFS, "srv returned malformed wsl symlink buffer\n");
 		return -EIO;
 	}
 
-	/* PathBuffer is in UTF-8 but without trailing null-term byte */
-	symname_utf8_len = len - sizeof(buf->Flags);
+	/* MS-FSCC 2.1.2.7 defines layout of the Target field only for Version 2. */
+	if (le32_to_cpu(buf->Version) != 2) {
+		cifs_dbg(VFS, "srv returned unsupported wsl symlink version %u\n", le32_to_cpu(buf->Version));
+		return -EIO;
+	}
+
+	/* Target for Version 2 is in UTF-8 but without trailing null-term byte */
+	symname_utf8_len = len - data_offset;
 	/*
 	 * Check that buffer does not contain null byte
 	 * because Linux cannot process symlink with null byte.
 	 */
-	if (strnlen(buf->PathBuffer, symname_utf8_len) != symname_utf8_len) {
+	if (strnlen(buf->Target, symname_utf8_len) != symname_utf8_len) {
 		cifs_dbg(VFS, "srv returned null byte in wsl symlink target location\n");
 		return -EIO;
 	}
 	symname_utf16 = kzalloc(symname_utf8_len * 2, GFP_KERNEL);
 	if (!symname_utf16)
 		return -ENOMEM;
-	symname_utf16_len = utf8s_to_utf16s(buf->PathBuffer, symname_utf8_len,
+	symname_utf16_len = utf8s_to_utf16s(buf->Target, symname_utf8_len,
 					    UTF16_LITTLE_ENDIAN,
 					    (wchar_t *) symname_utf16, symname_utf8_len * 2);
 	if (symname_utf16_len < 0) {
@@ -1062,8 +1069,6 @@ int parse_reparse_point(struct reparse_data_buffer *buf,
 			const char *full_path,
 			struct cifs_open_info_data *data)
 {
-	struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb);
-
 	data->reparse.buf = buf;
 
 	/* See MS-FSCC 2.1.2 */
@@ -1090,24 +1095,17 @@ int parse_reparse_point(struct reparse_data_buffer *buf,
 		}
 		return 0;
 	default:
-		cifs_tcon_dbg(VFS | ONCE, "unhandled reparse tag: 0x%08x\n",
-			      le32_to_cpu(buf->ReparseTag));
 		return -EOPNOTSUPP;
 	}
 }
 
-int smb2_parse_reparse_point(struct cifs_sb_info *cifs_sb,
-			     const char *full_path,
-			     struct kvec *rsp_iov,
-			     struct cifs_open_info_data *data)
+struct reparse_data_buffer *smb2_get_reparse_point_buffer(const struct kvec *rsp_iov,
+							  u32 *plen)
 {
-	struct reparse_data_buffer *buf;
 	struct smb2_ioctl_rsp *io = rsp_iov->iov_base;
-	u32 plen = le32_to_cpu(io->OutputCount);
-
-	buf = (struct reparse_data_buffer *)((u8 *)io +
-					     le32_to_cpu(io->OutputOffset));
-	return parse_reparse_point(buf, plen, cifs_sb, full_path, data);
+	*plen = le32_to_cpu(io->OutputCount);
+	return (struct reparse_data_buffer *)((u8 *)io +
+					      le32_to_cpu(io->OutputOffset));
 }
 
 static bool wsl_to_fattr(struct cifs_open_info_data *data,
@@ -1233,16 +1231,6 @@ bool cifs_reparse_point_to_fattr(struct cifs_sb_info *cifs_sb,
 	bool ok;
 
 	switch (tag) {
-	case IO_REPARSE_TAG_INTERNAL:
-		if (!(fattr->cf_cifsattrs & ATTR_DIRECTORY))
-			return false;
-		fallthrough;
-	case IO_REPARSE_TAG_DFS:
-	case IO_REPARSE_TAG_DFSR:
-	case IO_REPARSE_TAG_MOUNT_POINT:
-		/* See cifs_create_junction_fattr() */
-		fattr->cf_mode = S_IFDIR | 0711;
-		break;
 	case IO_REPARSE_TAG_LX_SYMLINK:
 	case IO_REPARSE_TAG_LX_FIFO:
 	case IO_REPARSE_TAG_AF_UNIX:
@@ -1262,7 +1250,14 @@ bool cifs_reparse_point_to_fattr(struct cifs_sb_info *cifs_sb,
 		fattr->cf_mode |= S_IFLNK;
 		break;
 	default:
-		return false;
+		if (!(fattr->cf_cifsattrs & ATTR_DIRECTORY))
+			return false;
+		if (!IS_REPARSE_TAG_NAME_SURROGATE(tag) &&
+		    tag != IO_REPARSE_TAG_INTERNAL)
+			return false;
+		/* See cifs_create_junction_fattr() */
+		fattr->cf_mode = S_IFDIR | 0711;
+		break;
 	}
 
 	fattr->cf_dtype = S_DT(fattr->cf_mode);
diff --git a/fs/smb/client/reparse.h b/fs/smb/client/reparse.h
index c0be5ab45a78..08de853b36a8 100644
--- a/fs/smb/client/reparse.h
+++ b/fs/smb/client/reparse.h
@@ -135,9 +135,6 @@ int smb2_create_reparse_symlink(const unsigned int xid, struct inode *inode,
 int smb2_mknod_reparse(unsigned int xid, struct inode *inode,
 		       struct dentry *dentry, struct cifs_tcon *tcon,
 		       const char *full_path, umode_t mode, dev_t dev);
-int smb2_parse_reparse_point(struct cifs_sb_info *cifs_sb,
-			     const char *full_path,
-			     struct kvec *rsp_iov,
-			     struct cifs_open_info_data *data);
+struct reparse_data_buffer *smb2_get_reparse_point_buffer(const struct kvec *rsp_iov, u32 *len);
 
 #endif /* _CIFS_REPARSE_H */
diff --git a/fs/smb/client/sess.c b/fs/smb/client/sess.c
index f2ca5963cd9d..b3fa9ee26912 100644
--- a/fs/smb/client/sess.c
+++ b/fs/smb/client/sess.c
@@ -680,6 +680,22 @@ unicode_oslm_strings(char **pbcc_area, const struct nls_table *nls_cp)
 	*pbcc_area = bcc_ptr;
 }
 
+static void
+ascii_oslm_strings(char **pbcc_area, const struct nls_table *nls_cp)
+{
+	char *bcc_ptr = *pbcc_area;
+
+	strcpy(bcc_ptr, "Linux version ");
+	bcc_ptr += strlen("Linux version ");
+	strcpy(bcc_ptr, init_utsname()->release);
+	bcc_ptr += strlen(init_utsname()->release) + 1;
+
+	strcpy(bcc_ptr, CIFS_NETWORK_OPSYS);
+	bcc_ptr += strlen(CIFS_NETWORK_OPSYS) + 1;
+
+	*pbcc_area = bcc_ptr;
+}
+
 static void unicode_domain_string(char **pbcc_area, struct cifs_ses *ses,
 				   const struct nls_table *nls_cp)
 {
@@ -704,6 +720,25 @@ static void unicode_domain_string(char **pbcc_area, struct cifs_ses *ses,
 	*pbcc_area = bcc_ptr;
 }
 
+static void ascii_domain_string(char **pbcc_area, struct cifs_ses *ses,
+				const struct nls_table *nls_cp)
+{
+	char *bcc_ptr = *pbcc_area;
+	int len;
+
+	/* copy domain */
+	if (ses->domainName != NULL) {
+		len = strscpy(bcc_ptr, ses->domainName, CIFS_MAX_DOMAINNAME_LEN);
+		if (WARN_ON_ONCE(len < 0))
+			len = CIFS_MAX_DOMAINNAME_LEN - 1;
+		bcc_ptr += len;
+	} /* else we send a null domain name so server will default to its own domain */
+	*bcc_ptr = 0;
+	bcc_ptr++;
+
+	*pbcc_area = bcc_ptr;
+}
+
 static void unicode_ssetup_strings(char **pbcc_area, struct cifs_ses *ses,
 				   const struct nls_table *nls_cp)
 {
@@ -749,25 +784,10 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifs_ses *ses,
 	*bcc_ptr = 0;
 	bcc_ptr++; /* account for null termination */
 
-	/* copy domain */
-	if (ses->domainName != NULL) {
-		len = strscpy(bcc_ptr, ses->domainName, CIFS_MAX_DOMAINNAME_LEN);
-		if (WARN_ON_ONCE(len < 0))
-			len = CIFS_MAX_DOMAINNAME_LEN - 1;
-		bcc_ptr += len;
-	} /* else we send a null domain name so server will default to its own domain */
-	*bcc_ptr = 0;
-	bcc_ptr++;
-
 	/* BB check for overflow here */
 
-	strcpy(bcc_ptr, "Linux version ");
-	bcc_ptr += strlen("Linux version ");
-	strcpy(bcc_ptr, init_utsname()->release);
-	bcc_ptr += strlen(init_utsname()->release) + 1;
-
-	strcpy(bcc_ptr, CIFS_NETWORK_OPSYS);
-	bcc_ptr += strlen(CIFS_NETWORK_OPSYS) + 1;
+	ascii_domain_string(&bcc_ptr, ses, nls_cp);
+	ascii_oslm_strings(&bcc_ptr, nls_cp);
 
 	*pbcc_area = bcc_ptr;
 }
@@ -1570,7 +1590,7 @@ sess_auth_kerberos(struct sess_data *sess_data)
 	sess_data->iov[1].iov_len = msg->secblob_len;
 	pSMB->req.SecurityBlobLength = cpu_to_le16(sess_data->iov[1].iov_len);
 
-	if (ses->capabilities & CAP_UNICODE) {
+	if (pSMB->req.hdr.Flags2 & SMBFLG2_UNICODE) {
 		/* unicode strings must be word aligned */
 		if (!IS_ALIGNED(sess_data->iov[0].iov_len + sess_data->iov[1].iov_len, 2)) {
 			*bcc_ptr = 0;
@@ -1579,8 +1599,8 @@ sess_auth_kerberos(struct sess_data *sess_data)
 		unicode_oslm_strings(&bcc_ptr, sess_data->nls_cp);
 		unicode_domain_string(&bcc_ptr, ses, sess_data->nls_cp);
 	} else {
-		/* BB: is this right? */
-		ascii_ssetup_strings(&bcc_ptr, ses, sess_data->nls_cp);
+		ascii_oslm_strings(&bcc_ptr, sess_data->nls_cp);
+		ascii_domain_string(&bcc_ptr, ses, sess_data->nls_cp);
 	}
 
 	sess_data->iov[2].iov_len = (long) bcc_ptr -
diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c
index 26df807fbe7a..b27a182629ec 100644
--- a/fs/smb/client/smb1ops.c
+++ b/fs/smb/client/smb1ops.c
@@ -432,7 +432,7 @@ cifs_negotiate(const unsigned int xid,
 }
 
 static unsigned int
-cifs_negotiate_wsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx)
+smb1_negotiate_wsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx)
 {
 	__u64 unix_cap = le64_to_cpu(tcon->fsUnixInfo.Capability);
 	struct TCP_Server_Info *server = tcon->ses->server;
@@ -467,7 +467,7 @@ cifs_negotiate_wsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx)
 }
 
 static unsigned int
-cifs_negotiate_rsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx)
+smb1_negotiate_rsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx)
 {
 	__u64 unix_cap = le64_to_cpu(tcon->fsUnixInfo.Capability);
 	struct TCP_Server_Info *server = tcon->ses->server;
@@ -543,24 +543,104 @@ static int cifs_query_path_info(const unsigned int xid,
 				const char *full_path,
 				struct cifs_open_info_data *data)
 {
-	int rc;
+	int rc = -EOPNOTSUPP;
 	FILE_ALL_INFO fi = {};
+	struct cifs_search_info search_info = {};
+	bool non_unicode_wildcard = false;
 
 	data->reparse_point = false;
 	data->adjust_tz = false;
 
-	/* could do find first instead but this returns more info */
-	rc = CIFSSMBQPathInfo(xid, tcon, full_path, &fi, 0 /* not legacy */, cifs_sb->local_nls,
-			      cifs_remap(cifs_sb));
 	/*
-	 * BB optimize code so we do not make the above call when server claims
-	 * no NT SMB support and the above call failed at least once - set flag
-	 * in tcon or mount.
+	 * First try CIFSSMBQPathInfo() function which returns more info
+	 * (NumberOfLinks) than CIFSFindFirst() fallback function.
+	 * Some servers like Win9x do not support SMB_QUERY_FILE_ALL_INFO over
+	 * TRANS2_QUERY_PATH_INFORMATION, but supports it with filehandle over
+	 * TRANS2_QUERY_FILE_INFORMATION (function CIFSSMBQFileInfo(). But SMB
+	 * Open command on non-NT servers works only for files, does not work
+	 * for directories. And moreover Win9x SMB server returns bogus data in
+	 * SMB_QUERY_FILE_ALL_INFO Attributes field. So for non-NT servers,
+	 * do not even use CIFSSMBQPathInfo() or CIFSSMBQFileInfo() function.
+	 */
+	if (tcon->ses->capabilities & CAP_NT_SMBS)
+		rc = CIFSSMBQPathInfo(xid, tcon, full_path, &fi, 0 /* not legacy */,
+				      cifs_sb->local_nls, cifs_remap(cifs_sb));
+
+	/*
+	 * Non-UNICODE variant of fallback functions below expands wildcards,
+	 * so they cannot be used for querying paths with wildcard characters.
+	 */
+	if (rc && !(tcon->ses->capabilities & CAP_UNICODE) && strpbrk(full_path, "*?\"><"))
+		non_unicode_wildcard = true;
+
+	/*
+	 * Then fallback to CIFSFindFirst() which works also with non-NT servers
+	 * but does not does not provide NumberOfLinks.
+	 */
+	if ((rc == -EOPNOTSUPP || rc == -EINVAL) &&
+	    !non_unicode_wildcard) {
+		if (!(tcon->ses->capabilities & tcon->ses->server->vals->cap_nt_find))
+			search_info.info_level = SMB_FIND_FILE_INFO_STANDARD;
+		else
+			search_info.info_level = SMB_FIND_FILE_FULL_DIRECTORY_INFO;
+		rc = CIFSFindFirst(xid, tcon, full_path, cifs_sb, NULL,
+				   CIFS_SEARCH_CLOSE_ALWAYS | CIFS_SEARCH_CLOSE_AT_END,
+				   &search_info, false);
+		if (rc == 0) {
+			if (!(tcon->ses->capabilities & tcon->ses->server->vals->cap_nt_find)) {
+				FIND_FILE_STANDARD_INFO *di;
+				int offset = tcon->ses->server->timeAdj;
+
+				di = (FIND_FILE_STANDARD_INFO *)search_info.srch_entries_start;
+				fi.CreationTime = cpu_to_le64(cifs_UnixTimeToNT(cnvrtDosUnixTm(
+						di->CreationDate, di->CreationTime, offset)));
+				fi.LastAccessTime = cpu_to_le64(cifs_UnixTimeToNT(cnvrtDosUnixTm(
+						di->LastAccessDate, di->LastAccessTime, offset)));
+				fi.LastWriteTime = cpu_to_le64(cifs_UnixTimeToNT(cnvrtDosUnixTm(
+						di->LastWriteDate, di->LastWriteTime, offset)));
+				fi.ChangeTime = fi.LastWriteTime;
+				fi.Attributes = cpu_to_le32(le16_to_cpu(di->Attributes));
+				fi.AllocationSize = cpu_to_le64(le32_to_cpu(di->AllocationSize));
+				fi.EndOfFile = cpu_to_le64(le32_to_cpu(di->DataSize));
+			} else {
+				FILE_FULL_DIRECTORY_INFO *di;
+
+				di = (FILE_FULL_DIRECTORY_INFO *)search_info.srch_entries_start;
+				fi.CreationTime = di->CreationTime;
+				fi.LastAccessTime = di->LastAccessTime;
+				fi.LastWriteTime = di->LastWriteTime;
+				fi.ChangeTime = di->ChangeTime;
+				fi.Attributes = di->ExtFileAttributes;
+				fi.AllocationSize = di->AllocationSize;
+				fi.EndOfFile = di->EndOfFile;
+				fi.EASize = di->EaSize;
+			}
+			fi.NumberOfLinks = cpu_to_le32(1);
+			fi.DeletePending = 0;
+			fi.Directory = !!(le32_to_cpu(fi.Attributes) & ATTR_DIRECTORY);
+			cifs_buf_release(search_info.ntwrk_buf_start);
+		} else if (!full_path[0]) {
+			/*
+			 * CIFSFindFirst() does not work on root path if the
+			 * root path was exported on the server from the top
+			 * level path (drive letter).
+			 */
+			rc = -EOPNOTSUPP;
+		}
+	}
+
+	/*
+	 * If everything failed then fallback to the legacy SMB command
+	 * SMB_COM_QUERY_INFORMATION which works with all servers, but
+	 * provide just few information.
 	 */
-	if ((rc == -EOPNOTSUPP) || (rc == -EINVAL)) {
+	if ((rc == -EOPNOTSUPP || rc == -EINVAL) && !non_unicode_wildcard) {
 		rc = SMBQueryInformation(xid, tcon, full_path, &fi, cifs_sb->local_nls,
 					 cifs_remap(cifs_sb));
 		data->adjust_tz = true;
+	} else if ((rc == -EOPNOTSUPP || rc == -EINVAL) && non_unicode_wildcard) {
+		/* Path with non-UNICODE wildcard character cannot exist. */
+		rc = -ENOENT;
 	}
 
 	if (!rc) {
@@ -568,6 +648,42 @@ static int cifs_query_path_info(const unsigned int xid,
 		data->reparse_point = le32_to_cpu(fi.Attributes) & ATTR_REPARSE;
 	}
 
+#ifdef CONFIG_CIFS_XATTR
+	/*
+	 * For WSL CHR and BLK reparse points it is required to fetch
+	 * EA $LXDEV which contains major and minor device numbers.
+	 */
+	if (!rc && data->reparse_point) {
+		struct smb2_file_full_ea_info *ea;
+
+		ea = (struct smb2_file_full_ea_info *)data->wsl.eas;
+		rc = CIFSSMBQAllEAs(xid, tcon, full_path, SMB2_WSL_XATTR_DEV,
+				    &ea->ea_data[SMB2_WSL_XATTR_NAME_LEN + 1],
+				    SMB2_WSL_XATTR_DEV_SIZE, cifs_sb);
+		if (rc == SMB2_WSL_XATTR_DEV_SIZE) {
+			ea->next_entry_offset = cpu_to_le32(0);
+			ea->flags = 0;
+			ea->ea_name_length = SMB2_WSL_XATTR_NAME_LEN;
+			ea->ea_value_length = cpu_to_le16(SMB2_WSL_XATTR_DEV_SIZE);
+			memcpy(&ea->ea_data[0], SMB2_WSL_XATTR_DEV, SMB2_WSL_XATTR_NAME_LEN + 1);
+			data->wsl.eas_len = sizeof(*ea) + SMB2_WSL_XATTR_NAME_LEN + 1 +
+					    SMB2_WSL_XATTR_DEV_SIZE;
+			rc = 0;
+		} else if (rc >= 0) {
+			/* It is an error if EA $LXDEV has wrong size. */
+			rc = -EINVAL;
+		} else {
+			/*
+			 * In all other cases ignore error if fetching
+			 * of EA $LXDEV failed. It is needed only for
+			 * WSL CHR and BLK reparse points and wsl_to_fattr()
+			 * handle the case when EA is missing.
+			 */
+			rc = 0;
+		}
+	}
+#endif
+
 	return rc;
 }
 
@@ -603,6 +719,13 @@ static int cifs_query_file_info(const unsigned int xid, struct cifs_tcon *tcon,
 	int rc;
 	FILE_ALL_INFO fi = {};
 
+	/*
+	 * CIFSSMBQFileInfo() for non-NT servers returns bogus data in
+	 * Attributes fields. So do not use this command for non-NT servers.
+	 */
+	if (!(tcon->ses->capabilities & CAP_NT_SMBS))
+		return -EOPNOTSUPP;
+
 	if (cfile->symlink_target) {
 		data->symlink_target = kstrdup(cfile->symlink_target, GFP_KERNEL);
 		if (!data->symlink_target)
@@ -773,6 +896,9 @@ smb_set_file_info(struct inode *inode, const char *full_path,
 	struct cifs_fid fid;
 	struct cifs_open_parms oparms;
 	struct cifsFileInfo *open_file;
+	FILE_BASIC_INFO new_buf;
+	struct cifs_open_info_data query_data;
+	__le64 write_time = buf->LastWriteTime;
 	struct cifsInodeInfo *cinode = CIFS_I(inode);
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 	struct tcon_link *tlink = NULL;
@@ -780,20 +906,58 @@ smb_set_file_info(struct inode *inode, const char *full_path,
 
 	/* if the file is already open for write, just use that fileid */
 	open_file = find_writable_file(cinode, FIND_WR_FSUID_ONLY);
+
 	if (open_file) {
 		fid.netfid = open_file->fid.netfid;
 		netpid = open_file->pid;
 		tcon = tlink_tcon(open_file->tlink);
-		goto set_via_filehandle;
+	} else {
+		tlink = cifs_sb_tlink(cifs_sb);
+		if (IS_ERR(tlink)) {
+			rc = PTR_ERR(tlink);
+			tlink = NULL;
+			goto out;
+		}
+		tcon = tlink_tcon(tlink);
 	}
 
-	tlink = cifs_sb_tlink(cifs_sb);
-	if (IS_ERR(tlink)) {
-		rc = PTR_ERR(tlink);
-		tlink = NULL;
-		goto out;
+	/*
+	 * Non-NT servers interprets zero time value in SMB_SET_FILE_BASIC_INFO
+	 * over TRANS2_SET_FILE_INFORMATION as a valid time value. NT servers
+	 * interprets zero time value as do not change existing value on server.
+	 * API of ->set_file_info() callback expects that zero time value has
+	 * the NT meaning - do not change. Therefore if server is non-NT and
+	 * some time values in "buf" are zero, then fetch missing time values.
+	 */
+	if (!(tcon->ses->capabilities & CAP_NT_SMBS) &&
+	    (!buf->CreationTime || !buf->LastAccessTime ||
+	     !buf->LastWriteTime || !buf->ChangeTime)) {
+		rc = cifs_query_path_info(xid, tcon, cifs_sb, full_path, &query_data);
+		if (rc) {
+			if (open_file) {
+				cifsFileInfo_put(open_file);
+				open_file = NULL;
+			}
+			goto out;
+		}
+		/*
+		 * Original write_time from buf->LastWriteTime is preserved
+		 * as SMBSetInformation() interprets zero as do not change.
+		 */
+		new_buf = *buf;
+		buf = &new_buf;
+		if (!buf->CreationTime)
+			buf->CreationTime = query_data.fi.CreationTime;
+		if (!buf->LastAccessTime)
+			buf->LastAccessTime = query_data.fi.LastAccessTime;
+		if (!buf->LastWriteTime)
+			buf->LastWriteTime = query_data.fi.LastWriteTime;
+		if (!buf->ChangeTime)
+			buf->ChangeTime = query_data.fi.ChangeTime;
 	}
-	tcon = tlink_tcon(tlink);
+
+	if (open_file)
+		goto set_via_filehandle;
 
 	rc = CIFSSMBSetPathInfo(xid, tcon, full_path, buf, cifs_sb->local_nls,
 				cifs_sb);
@@ -814,8 +978,45 @@ smb_set_file_info(struct inode *inode, const char *full_path,
 		.fid = &fid,
 	};
 
-	cifs_dbg(FYI, "calling SetFileInfo since SetPathInfo for times not supported by this server\n");
-	rc = CIFS_open(xid, &oparms, &oplock, NULL);
+	if (S_ISDIR(inode->i_mode) && !(tcon->ses->capabilities & CAP_NT_SMBS)) {
+		/* Opening directory path is not possible on non-NT servers. */
+		rc = -EOPNOTSUPP;
+	} else {
+		/*
+		 * Use cifs_open_file() instead of CIFS_open() as the
+		 * cifs_open_file() selects the correct function which
+		 * works also on non-NT servers.
+		 */
+		rc = cifs_open_file(xid, &oparms, &oplock, NULL);
+		/*
+		 * Opening path for writing on non-NT servers is not
+		 * possible when the read-only attribute is already set.
+		 * Non-NT server in this case returns -EACCES. For those
+		 * servers the only possible way how to clear the read-only
+		 * bit is via SMB_COM_SETATTR command.
+		 */
+		if (rc == -EACCES &&
+		    (cinode->cifsAttrs & ATTR_READONLY) &&
+		     le32_to_cpu(buf->Attributes) != 0 && /* 0 = do not change attrs */
+		     !(le32_to_cpu(buf->Attributes) & ATTR_READONLY) &&
+		     !(tcon->ses->capabilities & CAP_NT_SMBS))
+			rc = -EOPNOTSUPP;
+	}
+
+	/* Fallback to SMB_COM_SETATTR command when absolutelty needed. */
+	if (rc == -EOPNOTSUPP) {
+		cifs_dbg(FYI, "calling SetInformation since SetPathInfo for attrs/times not supported by this server\n");
+		rc = SMBSetInformation(xid, tcon, full_path,
+				       buf->Attributes != 0 ? buf->Attributes : cpu_to_le32(cinode->cifsAttrs),
+				       write_time,
+				       cifs_sb->local_nls, cifs_sb);
+		if (rc == 0)
+			cinode->cifsAttrs = le32_to_cpu(buf->Attributes);
+		else
+			rc = -EACCES;
+		goto out;
+	}
+
 	if (rc != 0) {
 		if (rc == -EIO)
 			rc = -EINVAL;
@@ -823,6 +1024,7 @@ smb_set_file_info(struct inode *inode, const char *full_path,
 	}
 
 	netpid = current->tgid;
+	cifs_dbg(FYI, "calling SetFileInfo since SetPathInfo for attrs/times not supported by this server\n");
 
 set_via_filehandle:
 	rc = CIFSSMBSetFileInfo(xid, tcon, buf, fid.netfid, netpid);
@@ -833,6 +1035,21 @@ set_via_filehandle:
 		CIFSSMBClose(xid, tcon, fid.netfid);
 	else
 		cifsFileInfo_put(open_file);
+
+	/*
+	 * Setting the read-only bit is not honered on non-NT servers when done
+	 * via open-semantics. So for setting it, use SMB_COM_SETATTR command.
+	 * This command works only after the file is closed, so use it only when
+	 * operation was called without the filehandle.
+	 */
+	if (open_file == NULL &&
+	    !(tcon->ses->capabilities & CAP_NT_SMBS) &&
+	    le32_to_cpu(buf->Attributes) & ATTR_READONLY) {
+		SMBSetInformation(xid, tcon, full_path,
+				  buf->Attributes,
+				  0 /* do not change write time */,
+				  cifs_sb->local_nls, cifs_sb);
+	}
 out:
 	if (tlink != NULL)
 		cifs_put_tlink(tlink);
@@ -970,18 +1187,13 @@ static int cifs_query_symlink(const unsigned int xid,
 	return rc;
 }
 
-static int cifs_parse_reparse_point(struct cifs_sb_info *cifs_sb,
-				    const char *full_path,
-				    struct kvec *rsp_iov,
-				    struct cifs_open_info_data *data)
+static struct reparse_data_buffer *cifs_get_reparse_point_buffer(const struct kvec *rsp_iov,
+								 u32 *plen)
 {
-	struct reparse_data_buffer *buf;
 	TRANSACT_IOCTL_RSP *io = rsp_iov->iov_base;
-	u32 plen = le16_to_cpu(io->ByteCount);
-
-	buf = (struct reparse_data_buffer *)((__u8 *)&io->hdr.Protocol +
-					     le32_to_cpu(io->DataOffset));
-	return parse_reparse_point(buf, plen, cifs_sb, full_path, data);
+	*plen = le16_to_cpu(io->ByteCount);
+	return (struct reparse_data_buffer *)((__u8 *)&io->hdr.Protocol +
+					      le32_to_cpu(io->DataOffset));
 }
 
 static bool
@@ -1130,8 +1342,8 @@ struct smb_version_operations smb1_operations = {
 	.check_trans2 = cifs_check_trans2,
 	.need_neg = cifs_need_neg,
 	.negotiate = cifs_negotiate,
-	.negotiate_wsize = cifs_negotiate_wsize,
-	.negotiate_rsize = cifs_negotiate_rsize,
+	.negotiate_wsize = smb1_negotiate_wsize,
+	.negotiate_rsize = smb1_negotiate_rsize,
 	.sess_setup = CIFS_SessSetup,
 	.logoff = CIFSSMBLogoff,
 	.tree_connect = CIFSTCon,
@@ -1157,7 +1369,7 @@ struct smb_version_operations smb1_operations = {
 	.rename = CIFSSMBRename,
 	.create_hardlink = CIFSCreateHardLink,
 	.query_symlink = cifs_query_symlink,
-	.parse_reparse_point = cifs_parse_reparse_point,
+	.get_reparse_point_buffer = cifs_get_reparse_point_buffer,
 	.open = cifs_open_file,
 	.set_fid = cifs_set_fid,
 	.close = cifs_close_file,
diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c
index 41d8cd20b25f..2fe8eeb98535 100644
--- a/fs/smb/client/smb2ops.c
+++ b/fs/smb/client/smb2ops.c
@@ -4555,9 +4555,9 @@ decrypt_raw_data(struct TCP_Server_Info *server, char *buf,
 			return rc;
 		}
 	} else {
-		if (unlikely(!server->secmech.dec))
-			return -EIO;
-
+		rc = smb3_crypto_aead_allocate(server);
+		if (unlikely(rc))
+			return rc;
 		tfm = server->secmech.dec;
 	}
 
@@ -5303,7 +5303,7 @@ struct smb_version_operations smb20_operations = {
 	.unlink = smb2_unlink,
 	.rename = smb2_rename_path,
 	.create_hardlink = smb2_create_hardlink,
-	.parse_reparse_point = smb2_parse_reparse_point,
+	.get_reparse_point_buffer = smb2_get_reparse_point_buffer,
 	.query_mf_symlink = smb3_query_mf_symlink,
 	.create_mf_symlink = smb3_create_mf_symlink,
 	.create_reparse_symlink = smb2_create_reparse_symlink,
@@ -5406,7 +5406,7 @@ struct smb_version_operations smb21_operations = {
 	.unlink = smb2_unlink,
 	.rename = smb2_rename_path,
 	.create_hardlink = smb2_create_hardlink,
-	.parse_reparse_point = smb2_parse_reparse_point,
+	.get_reparse_point_buffer = smb2_get_reparse_point_buffer,
 	.query_mf_symlink = smb3_query_mf_symlink,
 	.create_mf_symlink = smb3_create_mf_symlink,
 	.create_reparse_symlink = smb2_create_reparse_symlink,
@@ -5513,7 +5513,7 @@ struct smb_version_operations smb30_operations = {
 	.unlink = smb2_unlink,
 	.rename = smb2_rename_path,
 	.create_hardlink = smb2_create_hardlink,
-	.parse_reparse_point = smb2_parse_reparse_point,
+	.get_reparse_point_buffer = smb2_get_reparse_point_buffer,
 	.query_mf_symlink = smb3_query_mf_symlink,
 	.create_mf_symlink = smb3_create_mf_symlink,
 	.create_reparse_symlink = smb2_create_reparse_symlink,
@@ -5629,7 +5629,7 @@ struct smb_version_operations smb311_operations = {
 	.unlink = smb2_unlink,
 	.rename = smb2_rename_path,
 	.create_hardlink = smb2_create_hardlink,
-	.parse_reparse_point = smb2_parse_reparse_point,
+	.get_reparse_point_buffer = smb2_get_reparse_point_buffer,
 	.query_mf_symlink = smb3_query_mf_symlink,
 	.create_mf_symlink = smb3_create_mf_symlink,
 	.create_reparse_symlink = smb2_create_reparse_symlink,
diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c
index 81e05db8e4d5..0b35816d551f 100644
--- a/fs/smb/client/smb2pdu.c
+++ b/fs/smb/client/smb2pdu.c
@@ -1252,15 +1252,8 @@ SMB2_negotiate(const unsigned int xid,
 			cifs_server_dbg(VFS, "Missing expected negotiate contexts\n");
 	}
 
-	if (server->cipher_type && !rc) {
-		if (!SERVER_IS_CHAN(server)) {
-			rc = smb3_crypto_aead_allocate(server);
-		} else {
-			/* For channels, just reuse the primary server crypto secmech. */
-			server->secmech.enc = server->primary_server->secmech.enc;
-			server->secmech.dec = server->primary_server->secmech.dec;
-		}
-	}
+	if (server->cipher_type && !rc)
+		rc = smb3_crypto_aead_allocate(server);
 neg_exit:
 	free_rsp_buf(resp_buftype, rsp);
 	return rc;
@@ -2928,6 +2921,7 @@ replay_again:
 		req->CreateContextsOffset = cpu_to_le32(
 			sizeof(struct smb2_create_req) +
 			iov[1].iov_len);
+		le32_add_cpu(&req->CreateContextsLength, iov[n_iov-1].iov_len);
 		pc_buf = iov[n_iov-1].iov_base;
 	}
 
@@ -4099,12 +4093,8 @@ static void cifs_renegotiate_iosize(struct TCP_Server_Info *server,
 		return;
 
 	spin_lock(&tcon->sb_list_lock);
-	list_for_each_entry(cifs_sb, &tcon->cifs_sb_list, tcon_sb_link) {
-		cifs_sb->ctx->rsize =
-			server->ops->negotiate_rsize(tcon, cifs_sb->ctx);
-		cifs_sb->ctx->wsize =
-			server->ops->negotiate_wsize(tcon, cifs_sb->ctx);
-	}
+	list_for_each_entry(cifs_sb, &tcon->cifs_sb_list, tcon_sb_link)
+		cifs_negotiate_iosize(server, cifs_sb->ctx, tcon);
 	spin_unlock(&tcon->sb_list_lock);
 }
 
diff --git a/fs/smb/common/smb2pdu.h b/fs/smb/common/smb2pdu.h
index 764dca80c15c..f79a5165a7cc 100644
--- a/fs/smb/common/smb2pdu.h
+++ b/fs/smb/common/smb2pdu.h
@@ -1567,13 +1567,13 @@ struct reparse_nfs_data_buffer {
 	__u8	DataBuffer[];
 } __packed;
 
-/* For IO_REPARSE_TAG_LX_SYMLINK */
+/* For IO_REPARSE_TAG_LX_SYMLINK - see MS-FSCC 2.1.2.7 */
 struct reparse_wsl_symlink_data_buffer {
 	__le32	ReparseTag;
 	__le16	ReparseDataLength;
 	__u16	Reserved;
-	__le32	Flags;
-	__u8	PathBuffer[]; /* Variable Length UTF-8 string without nul-term */
+	__le32	Version; /* Always 2 */
+	__u8	Target[]; /* Variable Length UTF-8 string without nul-term */
 } __packed;
 
 struct validate_negotiate_info_req {
diff --git a/fs/smb/server/auth.c b/fs/smb/server/auth.c
index 83caa3849749..b3d121052408 100644
--- a/fs/smb/server/auth.c
+++ b/fs/smb/server/auth.c
@@ -550,7 +550,19 @@ int ksmbd_krb5_authenticate(struct ksmbd_session *sess, char *in_blob,
 		retval = -ENOMEM;
 		goto out;
 	}
-	sess->user = user;
+
+	if (!sess->user) {
+		/* First successful authentication */
+		sess->user = user;
+	} else {
+		if (!ksmbd_compare_user(sess->user, user)) {
+			ksmbd_debug(AUTH, "different user tried to reuse session\n");
+			retval = -EPERM;
+			ksmbd_free_user(user);
+			goto out;
+		}
+		ksmbd_free_user(user);
+	}
 
 	memcpy(sess->sess_key, resp->payload, resp->session_key_len);
 	memcpy(out_blob, resp->payload + resp->session_key_len,
diff --git a/fs/smb/server/connection.c b/fs/smb/server/connection.c
index c1f22c129111..83764c230e9d 100644
--- a/fs/smb/server/connection.c
+++ b/fs/smb/server/connection.c
@@ -39,8 +39,10 @@ void ksmbd_conn_free(struct ksmbd_conn *conn)
 	xa_destroy(&conn->sessions);
 	kvfree(conn->request_buf);
 	kfree(conn->preauth_info);
-	if (atomic_dec_and_test(&conn->refcnt))
+	if (atomic_dec_and_test(&conn->refcnt)) {
+		ksmbd_free_transport(conn->transport);
 		kfree(conn);
+	}
 }
 
 /**
diff --git a/fs/smb/server/mgmt/user_session.c b/fs/smb/server/mgmt/user_session.c
index 3f45f28f6f0f..9dec4c2940bc 100644
--- a/fs/smb/server/mgmt/user_session.c
+++ b/fs/smb/server/mgmt/user_session.c
@@ -59,10 +59,12 @@ static void ksmbd_session_rpc_clear_list(struct ksmbd_session *sess)
 	struct ksmbd_session_rpc *entry;
 	long index;
 
+	down_write(&sess->rpc_lock);
 	xa_for_each(&sess->rpc_handle_list, index, entry) {
 		xa_erase(&sess->rpc_handle_list, index);
 		__session_rpc_close(sess, entry);
 	}
+	up_write(&sess->rpc_lock);
 
 	xa_destroy(&sess->rpc_handle_list);
 }
@@ -92,7 +94,7 @@ int ksmbd_session_rpc_open(struct ksmbd_session *sess, char *rpc_name)
 {
 	struct ksmbd_session_rpc *entry, *old;
 	struct ksmbd_rpc_command *resp;
-	int method;
+	int method, id;
 
 	method = __rpc_method(rpc_name);
 	if (!method)
@@ -102,26 +104,29 @@ int ksmbd_session_rpc_open(struct ksmbd_session *sess, char *rpc_name)
 	if (!entry)
 		return -ENOMEM;
 
+	down_read(&sess->rpc_lock);
 	entry->method = method;
-	entry->id = ksmbd_ipc_id_alloc();
-	if (entry->id < 0)
+	entry->id = id = ksmbd_ipc_id_alloc();
+	if (id < 0)
 		goto free_entry;
-	old = xa_store(&sess->rpc_handle_list, entry->id, entry, KSMBD_DEFAULT_GFP);
+	old = xa_store(&sess->rpc_handle_list, id, entry, KSMBD_DEFAULT_GFP);
 	if (xa_is_err(old))
 		goto free_id;
 
-	resp = ksmbd_rpc_open(sess, entry->id);
+	resp = ksmbd_rpc_open(sess, id);
 	if (!resp)
 		goto erase_xa;
 
+	up_read(&sess->rpc_lock);
 	kvfree(resp);
-	return entry->id;
+	return id;
 erase_xa:
 	xa_erase(&sess->rpc_handle_list, entry->id);
 free_id:
 	ksmbd_rpc_id_free(entry->id);
 free_entry:
 	kfree(entry);
+	up_read(&sess->rpc_lock);
 	return -EINVAL;
 }
 
@@ -129,9 +134,11 @@ void ksmbd_session_rpc_close(struct ksmbd_session *sess, int id)
 {
 	struct ksmbd_session_rpc *entry;
 
+	down_write(&sess->rpc_lock);
 	entry = xa_erase(&sess->rpc_handle_list, id);
 	if (entry)
 		__session_rpc_close(sess, entry);
+	up_write(&sess->rpc_lock);
 }
 
 int ksmbd_session_rpc_method(struct ksmbd_session *sess, int id)
@@ -439,6 +446,7 @@ static struct ksmbd_session *__session_create(int protocol)
 	sess->sequence_number = 1;
 	rwlock_init(&sess->tree_conns_lock);
 	atomic_set(&sess->refcnt, 2);
+	init_rwsem(&sess->rpc_lock);
 
 	ret = __init_smb2_session(sess);
 	if (ret)
diff --git a/fs/smb/server/mgmt/user_session.h b/fs/smb/server/mgmt/user_session.h
index f21348381d59..c5749d6ec715 100644
--- a/fs/smb/server/mgmt/user_session.h
+++ b/fs/smb/server/mgmt/user_session.h
@@ -63,6 +63,7 @@ struct ksmbd_session {
 	rwlock_t			tree_conns_lock;
 
 	atomic_t			refcnt;
+	struct rw_semaphore		rpc_lock;
 };
 
 static inline int test_session_flag(struct ksmbd_session *sess, int bit)
diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c
index f103b1bd0400..81a29857b1e3 100644
--- a/fs/smb/server/oplock.c
+++ b/fs/smb/server/oplock.c
@@ -129,14 +129,6 @@ static void free_opinfo(struct oplock_info *opinfo)
 	kfree(opinfo);
 }
 
-static inline void opinfo_free_rcu(struct rcu_head *rcu_head)
-{
-	struct oplock_info *opinfo;
-
-	opinfo = container_of(rcu_head, struct oplock_info, rcu_head);
-	free_opinfo(opinfo);
-}
-
 struct oplock_info *opinfo_get(struct ksmbd_file *fp)
 {
 	struct oplock_info *opinfo;
@@ -157,8 +149,8 @@ static struct oplock_info *opinfo_get_list(struct ksmbd_inode *ci)
 	if (list_empty(&ci->m_op_list))
 		return NULL;
 
-	rcu_read_lock();
-	opinfo = list_first_or_null_rcu(&ci->m_op_list, struct oplock_info,
+	down_read(&ci->m_lock);
+	opinfo = list_first_entry(&ci->m_op_list, struct oplock_info,
 					op_entry);
 	if (opinfo) {
 		if (opinfo->conn == NULL ||
@@ -171,8 +163,7 @@ static struct oplock_info *opinfo_get_list(struct ksmbd_inode *ci)
 			}
 		}
 	}
-
-	rcu_read_unlock();
+	up_read(&ci->m_lock);
 
 	return opinfo;
 }
@@ -185,7 +176,7 @@ void opinfo_put(struct oplock_info *opinfo)
 	if (!atomic_dec_and_test(&opinfo->refcount))
 		return;
 
-	call_rcu(&opinfo->rcu_head, opinfo_free_rcu);
+	free_opinfo(opinfo);
 }
 
 static void opinfo_add(struct oplock_info *opinfo)
@@ -193,7 +184,7 @@ static void opinfo_add(struct oplock_info *opinfo)
 	struct ksmbd_inode *ci = opinfo->o_fp->f_ci;
 
 	down_write(&ci->m_lock);
-	list_add_rcu(&opinfo->op_entry, &ci->m_op_list);
+	list_add(&opinfo->op_entry, &ci->m_op_list);
 	up_write(&ci->m_lock);
 }
 
@@ -207,7 +198,7 @@ static void opinfo_del(struct oplock_info *opinfo)
 		write_unlock(&lease_list_lock);
 	}
 	down_write(&ci->m_lock);
-	list_del_rcu(&opinfo->op_entry);
+	list_del(&opinfo->op_entry);
 	up_write(&ci->m_lock);
 }
 
@@ -1347,8 +1338,8 @@ void smb_break_all_levII_oplock(struct ksmbd_work *work, struct ksmbd_file *fp,
 	ci = fp->f_ci;
 	op = opinfo_get(fp);
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(brk_op, &ci->m_op_list, op_entry) {
+	down_read(&ci->m_lock);
+	list_for_each_entry(brk_op, &ci->m_op_list, op_entry) {
 		if (brk_op->conn == NULL)
 			continue;
 
@@ -1358,7 +1349,6 @@ void smb_break_all_levII_oplock(struct ksmbd_work *work, struct ksmbd_file *fp,
 		if (ksmbd_conn_releasing(brk_op->conn))
 			continue;
 
-		rcu_read_unlock();
 		if (brk_op->is_lease && (brk_op->o_lease->state &
 		    (~(SMB2_LEASE_READ_CACHING_LE |
 				SMB2_LEASE_HANDLE_CACHING_LE)))) {
@@ -1388,9 +1378,8 @@ void smb_break_all_levII_oplock(struct ksmbd_work *work, struct ksmbd_file *fp,
 		oplock_break(brk_op, SMB2_OPLOCK_LEVEL_NONE, NULL);
 next:
 		opinfo_put(brk_op);
-		rcu_read_lock();
 	}
-	rcu_read_unlock();
+	up_read(&ci->m_lock);
 
 	if (op)
 		opinfo_put(op);
diff --git a/fs/smb/server/oplock.h b/fs/smb/server/oplock.h
index 3f64f0787263..9a56eaadd0dd 100644
--- a/fs/smb/server/oplock.h
+++ b/fs/smb/server/oplock.h
@@ -71,7 +71,6 @@ struct oplock_info {
 	struct list_head        lease_entry;
 	wait_queue_head_t oplock_q; /* Other server threads */
 	wait_queue_head_t oplock_brk; /* oplock breaking wait */
-	struct rcu_head		rcu_head;
 };
 
 struct lease_break_info {
diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c
index d24d95d15d87..46aa08245742 100644
--- a/fs/smb/server/smb2pdu.c
+++ b/fs/smb/server/smb2pdu.c
@@ -1445,7 +1445,7 @@ static int ntlm_authenticate(struct ksmbd_work *work,
 {
 	struct ksmbd_conn *conn = work->conn;
 	struct ksmbd_session *sess = work->sess;
-	struct channel *chann = NULL;
+	struct channel *chann = NULL, *old;
 	struct ksmbd_user *user;
 	u64 prev_id;
 	int sz, rc;
@@ -1557,7 +1557,12 @@ binding_session:
 				return -ENOMEM;
 
 			chann->conn = conn;
-			xa_store(&sess->ksmbd_chann_list, (long)conn, chann, KSMBD_DEFAULT_GFP);
+			old = xa_store(&sess->ksmbd_chann_list, (long)conn, chann,
+					KSMBD_DEFAULT_GFP);
+			if (xa_is_err(old)) {
+				kfree(chann);
+				return xa_err(old);
+			}
 		}
 	}
 
@@ -1602,9 +1607,6 @@ static int krb5_authenticate(struct ksmbd_work *work,
 	if (prev_sess_id && prev_sess_id != sess->id)
 		destroy_previous_session(conn, sess->user, prev_sess_id);
 
-	if (sess->state == SMB2_SESSION_VALID)
-		ksmbd_free_user(sess->user);
-
 	retval = ksmbd_krb5_authenticate(sess, in_blob, in_len,
 					 out_blob, &out_len);
 	if (retval) {
@@ -2247,10 +2249,6 @@ int smb2_session_logoff(struct ksmbd_work *work)
 	sess->state = SMB2_SESSION_EXPIRED;
 	up_write(&conn->session_lock);
 
-	if (sess->user) {
-		ksmbd_free_user(sess->user);
-		sess->user = NULL;
-	}
 	ksmbd_all_conn_set_status(sess_id, KSMBD_SESS_NEED_SETUP);
 
 	rsp->StructureSize = cpu_to_le16(4);
diff --git a/fs/smb/server/smb_common.h b/fs/smb/server/smb_common.h
index a3d8a905b07e..d742ba754348 100644
--- a/fs/smb/server/smb_common.h
+++ b/fs/smb/server/smb_common.h
@@ -72,6 +72,8 @@
 #define FILE_SUPPORTS_ENCRYPTION	0x00020000
 #define FILE_SUPPORTS_OBJECT_IDS	0x00010000
 #define FILE_VOLUME_IS_COMPRESSED	0x00008000
+#define FILE_SUPPORTS_POSIX_UNLINK_RENAME 0x00000400
+#define FILE_RETURNS_CLEANUP_RESULT_INFO  0x00000200
 #define FILE_SUPPORTS_REMOTE_STORAGE	0x00000100
 #define FILE_SUPPORTS_REPARSE_POINTS	0x00000080
 #define FILE_SUPPORTS_SPARSE_FILES	0x00000040
diff --git a/fs/smb/server/transport_ipc.c b/fs/smb/server/transport_ipc.c
index 3f185ae60dc5..2a3e2b0ce557 100644
--- a/fs/smb/server/transport_ipc.c
+++ b/fs/smb/server/transport_ipc.c
@@ -310,7 +310,11 @@ static int ipc_server_config_on_startup(struct ksmbd_startup_request *req)
 	server_conf.signing = req->signing;
 	server_conf.tcp_port = req->tcp_port;
 	server_conf.ipc_timeout = req->ipc_timeout * HZ;
-	server_conf.deadtime = req->deadtime * SMB_ECHO_INTERVAL;
+	if (check_mul_overflow(req->deadtime, SMB_ECHO_INTERVAL,
+					&server_conf.deadtime)) {
+		ret = -EINVAL;
+		goto out;
+	}
 	server_conf.share_fake_fscaps = req->share_fake_fscaps;
 	ksmbd_init_domain(req->sub_auth);
 
@@ -337,6 +341,7 @@ static int ipc_server_config_on_startup(struct ksmbd_startup_request *req)
 	server_conf.bind_interfaces_only = req->bind_interfaces_only;
 	ret |= ksmbd_tcp_set_interfaces(KSMBD_STARTUP_CONFIG_INTERFACES(req),
 					req->ifc_list_sz);
+out:
 	if (ret) {
 		pr_err("Server configuration error: %s %s %s\n",
 		       req->netbios_name, req->server_string,
diff --git a/fs/smb/server/transport_tcp.c b/fs/smb/server/transport_tcp.c
index 7f38a3c3f5bd..abedf510899a 100644
--- a/fs/smb/server/transport_tcp.c
+++ b/fs/smb/server/transport_tcp.c
@@ -93,17 +93,21 @@ static struct tcp_transport *alloc_transport(struct socket *client_sk)
 	return t;
 }
 
-static void free_transport(struct tcp_transport *t)
+void ksmbd_free_transport(struct ksmbd_transport *kt)
 {
-	kernel_sock_shutdown(t->sock, SHUT_RDWR);
-	sock_release(t->sock);
-	t->sock = NULL;
+	struct tcp_transport *t = TCP_TRANS(kt);
 
-	ksmbd_conn_free(KSMBD_TRANS(t)->conn);
+	sock_release(t->sock);
 	kfree(t->iov);
 	kfree(t);
 }
 
+static void free_transport(struct tcp_transport *t)
+{
+	kernel_sock_shutdown(t->sock, SHUT_RDWR);
+	ksmbd_conn_free(KSMBD_TRANS(t)->conn);
+}
+
 /**
  * kvec_array_init() - initialize a IO vector segment
  * @new:	IO vector to be initialized
diff --git a/fs/smb/server/transport_tcp.h b/fs/smb/server/transport_tcp.h
index 8c9aa624cfe3..1e51675ee1b2 100644
--- a/fs/smb/server/transport_tcp.h
+++ b/fs/smb/server/transport_tcp.h
@@ -8,6 +8,7 @@
 
 int ksmbd_tcp_set_interfaces(char *ifc_list, int ifc_list_sz);
 struct interface *ksmbd_find_netdev_name_iface_list(char *netdev_name);
+void ksmbd_free_transport(struct ksmbd_transport *kt);
 int ksmbd_tcp_init(void);
 void ksmbd_tcp_destroy(void);
 
diff --git a/fs/smb/server/vfs.c b/fs/smb/server/vfs.c
index 8554aa5a1059..391d07da586c 100644
--- a/fs/smb/server/vfs.c
+++ b/fs/smb/server/vfs.c
@@ -479,7 +479,8 @@ int ksmbd_vfs_write(struct ksmbd_work *work, struct ksmbd_file *fp,
 	int err = 0;
 
 	if (work->conn->connection_type) {
-		if (!(fp->daccess & (FILE_WRITE_DATA_LE | FILE_APPEND_DATA_LE))) {
+		if (!(fp->daccess & (FILE_WRITE_DATA_LE | FILE_APPEND_DATA_LE)) ||
+		    S_ISDIR(file_inode(fp->filp)->i_mode)) {
 			pr_err("no right to write(%pD)\n", fp->filp);
 			err = -EACCES;
 			goto out;
diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c
index 8d1f30dcba7e..1f8fa3468173 100644
--- a/fs/smb/server/vfs_cache.c
+++ b/fs/smb/server/vfs_cache.c
@@ -713,12 +713,8 @@ static bool tree_conn_fd_check(struct ksmbd_tree_connect *tcon,
 
 static bool ksmbd_durable_scavenger_alive(void)
 {
-	mutex_lock(&durable_scavenger_lock);
-	if (!durable_scavenger_running) {
-		mutex_unlock(&durable_scavenger_lock);
+	if (!durable_scavenger_running)
 		return false;
-	}
-	mutex_unlock(&durable_scavenger_lock);
 
 	if (kthread_should_stop())
 		return false;
@@ -799,9 +795,7 @@ static int ksmbd_durable_scavenger(void *dummy)
 			break;
 	}
 
-	mutex_lock(&durable_scavenger_lock);
 	durable_scavenger_running = false;
-	mutex_unlock(&durable_scavenger_lock);
 
 	module_put(THIS_MODULE);
 
diff --git a/fs/splice.c b/fs/splice.c
index 90d464241f15..4d6df083e0c0 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -45,7 +45,7 @@
  * here if set to avoid blocking other users of this pipe if splice is
  * being done on it.
  */
-static noinline void noinline pipe_clear_nowait(struct file *file)
+static noinline void pipe_clear_nowait(struct file *file)
 {
 	fmode_t fmode = READ_ONCE(file->f_mode);
 
diff --git a/fs/stat.c b/fs/stat.c
index f13308bfdc98..3d9222807214 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -204,12 +204,25 @@ int vfs_getattr_nosec(const struct path *path, struct kstat *stat,
 				  STATX_ATTR_DAX);
 
 	idmap = mnt_idmap(path->mnt);
-	if (inode->i_op->getattr)
-		return inode->i_op->getattr(idmap, path, stat,
-					    request_mask,
-					    query_flags);
+	if (inode->i_op->getattr) {
+		int ret;
+
+		ret = inode->i_op->getattr(idmap, path, stat, request_mask,
+				query_flags);
+		if (ret)
+			return ret;
+	} else {
+		generic_fillattr(idmap, request_mask, inode, stat);
+	}
+
+	/*
+	 * If this is a block device inode, override the filesystem attributes
+	 * with the block device specific parameters that need to be obtained
+	 * from the bdev backing inode.
+	 */
+	if (S_ISBLK(stat->mode))
+		bdev_statx(path, stat, request_mask);
 
-	generic_fillattr(idmap, request_mask, inode, stat);
 	return 0;
 }
 EXPORT_SYMBOL(vfs_getattr_nosec);
@@ -295,15 +308,6 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat,
 	if (path_mounted(path))
 		stat->attributes |= STATX_ATTR_MOUNT_ROOT;
 	stat->attributes_mask |= STATX_ATTR_MOUNT_ROOT;
-
-	/*
-	 * If this is a block device inode, override the filesystem
-	 * attributes with the block device specific parameters that need to be
-	 * obtained from the bdev backing inode.
-	 */
-	if (S_ISBLK(stat->mode))
-		bdev_statx(path, stat, request_mask);
-
 	return 0;
 }
 
diff --git a/fs/xattr.c b/fs/xattr.c
index 02bee149ad96..fabb2a04501e 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -703,7 +703,7 @@ static int path_setxattrat(int dfd, const char __user *pathname,
 		return error;
 
 	filename = getname_maybe_null(pathname, at_flags);
-	if (!filename) {
+	if (!filename && dfd >= 0) {
 		CLASS(fd, f)(dfd);
 		if (fd_empty(f))
 			error = -EBADF;
@@ -847,7 +847,7 @@ static ssize_t path_getxattrat(int dfd, const char __user *pathname,
 		return error;
 
 	filename = getname_maybe_null(pathname, at_flags);
-	if (!filename) {
+	if (!filename && dfd >= 0) {
 		CLASS(fd, f)(dfd);
 		if (fd_empty(f))
 			return -EBADF;
diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig
index fffd6fffdce0..ae0ca6858496 100644
--- a/fs/xfs/Kconfig
+++ b/fs/xfs/Kconfig
@@ -3,7 +3,7 @@ config XFS_FS
 	tristate "XFS filesystem support"
 	depends on BLOCK
 	select EXPORTFS
-	select LIBCRC32C
+	select CRC32
 	select FS_IOMAP
 	help
 	  XFS is a high performance journaling filesystem which originated
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 8e7f1b324b3b..1a2b3f06fa71 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -105,6 +105,7 @@ xfs_buf_free(
 {
 	unsigned int		size = BBTOB(bp->b_length);
 
+	might_sleep();
 	trace_xfs_buf_free(bp, _RET_IP_);
 
 	ASSERT(list_empty(&bp->b_lru));
diff --git a/fs/xfs/xfs_buf_mem.c b/fs/xfs/xfs_buf_mem.c
index b4ffd80b7cb6..dcbfa274e06d 100644
--- a/fs/xfs/xfs_buf_mem.c
+++ b/fs/xfs/xfs_buf_mem.c
@@ -165,7 +165,7 @@ xmbuf_map_backing_mem(
 	folio_set_dirty(folio);
 	folio_unlock(folio);
 
-	bp->b_addr = folio_address(folio);
+	bp->b_addr = folio_address(folio) + offset_in_folio(folio, pos);
 	return 0;
 }
 
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index edbc521870a1..b4e32f0860b7 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -1186,9 +1186,8 @@ xfs_qm_dqflush_done(
 	if (test_bit(XFS_LI_IN_AIL, &lip->li_flags) &&
 	    (lip->li_lsn == qlip->qli_flush_lsn ||
 	     test_bit(XFS_LI_FAILED, &lip->li_flags))) {
-
 		spin_lock(&ailp->ail_lock);
-		xfs_clear_li_failed(lip);
+		clear_bit(XFS_LI_FAILED, &lip->li_flags);
 		if (lip->li_lsn == qlip->qli_flush_lsn) {
 			/* xfs_ail_update_finish() drops the AIL lock */
 			tail_lsn = xfs_ail_delete_one(ailp, lip);
diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c
index a4bc1642fe56..414b27a86458 100644
--- a/fs/xfs/xfs_fsmap.c
+++ b/fs/xfs/xfs_fsmap.c
@@ -876,6 +876,7 @@ xfs_getfsmap_rtdev_rmapbt(
 	const struct xfs_fsmap		*keys,
 	struct xfs_getfsmap_info	*info)
 {
+	struct xfs_fsmap		key0 = *keys; /* struct copy */
 	struct xfs_mount		*mp = tp->t_mountp;
 	struct xfs_rtgroup		*rtg = NULL;
 	struct xfs_btree_cur		*bt_cur = NULL;
@@ -887,32 +888,46 @@ xfs_getfsmap_rtdev_rmapbt(
 	int				error = 0;
 
 	eofs = XFS_FSB_TO_BB(mp, mp->m_sb.sb_rtstart + mp->m_sb.sb_rblocks);
-	if (keys[0].fmr_physical >= eofs)
+	if (key0.fmr_physical >= eofs)
 		return 0;
 
+	/*
+	 * On zoned filesystems with an internal rt volume, the volume comes
+	 * immediately after the end of the data volume.  However, the
+	 * xfs_rtblock_t address space is relative to the start of the data
+	 * device, which means that the first @rtstart fsblocks do not actually
+	 * point anywhere.  If a fsmap query comes in with the low key starting
+	 * below @rtstart, report it as "owned by filesystem".
+	 */
 	rtstart_daddr = XFS_FSB_TO_BB(mp, mp->m_sb.sb_rtstart);
-	if (keys[0].fmr_physical < rtstart_daddr) {
+	if (xfs_has_zoned(mp) && key0.fmr_physical < rtstart_daddr) {
 		struct xfs_fsmap_irec		frec = {
 			.owner			= XFS_RMAP_OWN_FS,
 			.len_daddr		= rtstart_daddr,
 		};
 
-		/* Adjust the low key if we are continuing from where we left off. */
-		if (keys[0].fmr_length > 0) {
-			info->low_daddr = keys[0].fmr_physical + keys[0].fmr_length;
-			return 0;
+		/*
+		 * Adjust the start of the query range if we're picking up from
+		 * a previous round, and only emit the record if we haven't
+		 * already gone past.
+		 */
+		key0.fmr_physical += key0.fmr_length;
+		if (key0.fmr_physical < rtstart_daddr) {
+			error = xfs_getfsmap_helper(tp, info, &frec);
+			if (error)
+				return error;
+
+			key0.fmr_physical = rtstart_daddr;
 		}
 
-		/* Fabricate an rmap entry for space occupied by the data dev */
-		error = xfs_getfsmap_helper(tp, info, &frec);
-		if (error)
-			return error;
+		/* Zero the other fields to avoid further adjustments. */
+		key0.fmr_owner = 0;
+		key0.fmr_offset = 0;
+		key0.fmr_length = 0;
 	}
 
-	start_rtb = xfs_daddr_to_rtb(mp, rtstart_daddr + keys[0].fmr_physical);
-	end_rtb = xfs_daddr_to_rtb(mp, rtstart_daddr +
-			min(eofs - 1, keys[1].fmr_physical));
-
+	start_rtb = xfs_daddr_to_rtb(mp, key0.fmr_physical);
+	end_rtb = xfs_daddr_to_rtb(mp, min(eofs - 1, keys[1].fmr_physical));
 	info->missing_owner = XFS_FMR_OWN_FREE;
 
 	/*
@@ -920,12 +935,12 @@ xfs_getfsmap_rtdev_rmapbt(
 	 * low to the fsmap low key and max out the high key to the end
 	 * of the rtgroup.
 	 */
-	info->low.rm_offset = XFS_BB_TO_FSBT(mp, keys[0].fmr_offset);
-	error = xfs_fsmap_owner_to_rmap(&info->low, &keys[0]);
+	info->low.rm_offset = XFS_BB_TO_FSBT(mp, key0.fmr_offset);
+	error = xfs_fsmap_owner_to_rmap(&info->low, &key0);
 	if (error)
 		return error;
-	info->low.rm_blockcount = XFS_BB_TO_FSBT(mp, keys[0].fmr_length);
-	xfs_getfsmap_set_irec_flags(&info->low, &keys[0]);
+	info->low.rm_blockcount = XFS_BB_TO_FSBT(mp, key0.fmr_length);
+	xfs_getfsmap_set_irec_flags(&info->low, &key0);
 
 	/* Adjust the low key if we are continuing from where we left off. */
 	if (info->low.rm_blockcount == 0) {
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 40fc1bf900af..c6cb0b6b9e46 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -1089,13 +1089,7 @@ xfs_iflush_abort(
 	 * state. Whilst the inode is in the AIL, it should have a valid buffer
 	 * pointer for push operations to access - it is only safe to remove the
 	 * inode from the buffer once it has been removed from the AIL.
-	 *
-	 * We also clear the failed bit before removing the item from the AIL
-	 * as xfs_trans_ail_delete()->xfs_clear_li_failed() will release buffer
-	 * references the inode item owns and needs to hold until we've fully
-	 * aborted the inode log item and detached it from the buffer.
 	 */
-	clear_bit(XFS_LI_FAILED, &iip->ili_item.li_flags);
 	xfs_trans_ail_delete(&iip->ili_item, 0);
 
 	/*
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 6493bdb57351..980aabc49512 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -2888,7 +2888,7 @@ xlog_force_and_check_iclog(
  *
  *	1. the current iclog is active and has no data; the previous iclog
  *		is in the active or dirty state.
- *	2. the current iclog is drity, and the previous iclog is in the
+ *	2. the current iclog is dirty, and the previous iclog is in the
  *		active or dirty state.
  *
  * We may sleep if:
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 799b84220ebb..e5192c12e7ac 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -229,6 +229,7 @@ typedef struct xfs_mount {
 	bool			m_finobt_nores; /* no per-AG finobt resv. */
 	bool			m_update_sb;	/* sb needs update in mount */
 	unsigned int		m_max_open_zones;
+	unsigned int		m_zonegc_low_space;
 
 	/*
 	 * Bitsets of per-fs metadata that have been checked and/or are sick.
diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c
index b7e82d85f043..7a5c5ef2db92 100644
--- a/fs/xfs/xfs_sysfs.c
+++ b/fs/xfs/xfs_sysfs.c
@@ -718,8 +718,40 @@ max_open_zones_show(
 }
 XFS_SYSFS_ATTR_RO(max_open_zones);
 
+static ssize_t
+zonegc_low_space_store(
+	struct kobject		*kobj,
+	const char		*buf,
+	size_t			count)
+{
+	int			ret;
+	unsigned int		val;
+
+	ret = kstrtouint(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val > 100)
+		return -EINVAL;
+
+	zoned_to_mp(kobj)->m_zonegc_low_space = val;
+
+	return count;
+}
+
+static ssize_t
+zonegc_low_space_show(
+	struct kobject		*kobj,
+	char			*buf)
+{
+	return sysfs_emit(buf, "%u\n",
+			zoned_to_mp(kobj)->m_zonegc_low_space);
+}
+XFS_SYSFS_ATTR_RW(zonegc_low_space);
+
 static struct attribute *xfs_zoned_attrs[] = {
 	ATTR_LIST(max_open_zones),
+	ATTR_LIST(zonegc_low_space),
 	NULL,
 };
 ATTRIBUTE_GROUPS(xfs_zoned);
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index 0fcb1828e598..85a649fec6ac 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -909,10 +909,9 @@ xfs_trans_ail_delete(
 		return;
 	}
 
-	/* xfs_ail_update_finish() drops the AIL lock */
-	xfs_clear_li_failed(lip);
+	clear_bit(XFS_LI_FAILED, &lip->li_flags);
 	tail_lsn = xfs_ail_delete_one(ailp, lip);
-	xfs_ail_update_finish(ailp, tail_lsn);
+	xfs_ail_update_finish(ailp, tail_lsn);	/* drops the AIL lock */
 }
 
 int
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index bd841df93021..f945f0450b16 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -167,32 +167,4 @@ xfs_trans_ail_copy_lsn(
 }
 #endif
 
-static inline void
-xfs_clear_li_failed(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_buf	*bp = lip->li_buf;
-
-	ASSERT(test_bit(XFS_LI_IN_AIL, &lip->li_flags));
-	lockdep_assert_held(&lip->li_ailp->ail_lock);
-
-	if (test_and_clear_bit(XFS_LI_FAILED, &lip->li_flags)) {
-		lip->li_buf = NULL;
-		xfs_buf_rele(bp);
-	}
-}
-
-static inline void
-xfs_set_li_failed(
-	struct xfs_log_item	*lip,
-	struct xfs_buf		*bp)
-{
-	lockdep_assert_held(&lip->li_ailp->ail_lock);
-
-	if (!test_and_set_bit(XFS_LI_FAILED, &lip->li_flags)) {
-		xfs_buf_hold(bp);
-		lip->li_buf = bp;
-	}
-}
-
 #endif	/* __XFS_TRANS_PRIV_H__ */
diff --git a/fs/xfs/xfs_zone_alloc.c b/fs/xfs/xfs_zone_alloc.c
index 52af234936a2..d509e49b2aaa 100644
--- a/fs/xfs/xfs_zone_alloc.c
+++ b/fs/xfs/xfs_zone_alloc.c
@@ -1201,6 +1201,13 @@ xfs_mount_zones(
 	xfs_set_freecounter(mp, XC_FREE_RTEXTENTS,
 			iz.available + iz.reclaimable);
 
+	/*
+	 * The user may configure GC to free up a percentage of unused blocks.
+	 * By default this is 0. GC will always trigger at the minimum level
+	 * for keeping max_open_zones available for data placement.
+	 */
+	mp->m_zonegc_low_space = 0;
+
 	error = xfs_zone_gc_mount(mp);
 	if (error)
 		goto out_free_zone_info;
diff --git a/fs/xfs/xfs_zone_gc.c b/fs/xfs/xfs_zone_gc.c
index c5136ea9bb1d..81c94dd1d596 100644
--- a/fs/xfs/xfs_zone_gc.c
+++ b/fs/xfs/xfs_zone_gc.c
@@ -162,18 +162,36 @@ struct xfs_zone_gc_data {
 
 /*
  * We aim to keep enough zones free in stock to fully use the open zone limit
- * for data placement purposes.
+ * for data placement purposes. Additionally, the m_zonegc_low_space tunable
+ * can be set to make sure a fraction of the unused blocks are available for
+ * writing.
  */
 bool
 xfs_zoned_need_gc(
 	struct xfs_mount	*mp)
 {
+	s64			available, free, threshold;
+	s32			remainder;
+
 	if (!xfs_group_marked(mp, XG_TYPE_RTG, XFS_RTG_RECLAIMABLE))
 		return false;
-	if (xfs_estimate_freecounter(mp, XC_FREE_RTAVAILABLE) <
+
+	available = xfs_estimate_freecounter(mp, XC_FREE_RTAVAILABLE);
+
+	if (available <
 	    mp->m_groups[XG_TYPE_RTG].blocks *
 	    (mp->m_max_open_zones - XFS_OPEN_GC_ZONES))
 		return true;
+
+	free = xfs_estimate_freecounter(mp, XC_FREE_RTEXTENTS);
+
+	threshold = div_s64_rem(free, 100, &remainder);
+	threshold = threshold * mp->m_zonegc_low_space +
+		    remainder * div_s64(mp->m_zonegc_low_space, 100);
+
+	if (available < threshold)
+		return true;
+
 	return false;
 }
 
diff --git a/include/acpi/actbl.h b/include/acpi/actbl.h
index 451f6276da49..2fc89704be17 100644
--- a/include/acpi/actbl.h
+++ b/include/acpi/actbl.h
@@ -66,7 +66,7 @@
  ******************************************************************************/
 
 struct acpi_table_header {
-	char signature[ACPI_NAMESEG_SIZE];	/* ASCII table signature */
+	char signature[ACPI_NAMESEG_SIZE] __nonstring;	/* ASCII table signature */
 	u32 length;		/* Length of table in bytes, including this header */
 	u8 revision;		/* ACPI Specification minor version number */
 	u8 checksum;		/* To make sum of entire table == 0 */
diff --git a/include/crypto/hash.h b/include/crypto/hash.h
index 2aa83ee0ec98..a67988316d06 100644
--- a/include/crypto/hash.h
+++ b/include/crypto/hash.h
@@ -10,6 +10,7 @@
 
 #include <linux/atomic.h>
 #include <linux/crypto.h>
+#include <linux/slab.h>
 #include <linux/string.h>
 
 /* Set this bit for virtual address instead of SG list. */
@@ -581,7 +582,10 @@ static inline struct ahash_request *ahash_request_alloc_noprof(
  * ahash_request_free() - zeroize and free the request data structure
  * @req: request data structure cipher handle to be freed
  */
-void ahash_request_free(struct ahash_request *req);
+static inline void ahash_request_free(struct ahash_request *req)
+{
+	kfree_sensitive(req);
+}
 
 static inline struct ahash_request *ahash_request_cast(
 	struct crypto_async_request *req)
diff --git a/include/crypto/internal/hash.h b/include/crypto/internal/hash.h
index 485e22cf517e..052ac7924af3 100644
--- a/include/crypto/internal/hash.h
+++ b/include/crypto/internal/hash.h
@@ -249,7 +249,7 @@ static inline struct crypto_shash *__crypto_shash_cast(struct crypto_tfm *tfm)
 
 static inline bool ahash_request_chained(struct ahash_request *req)
 {
-	return crypto_request_chained(&req->base);
+	return false;
 }
 
 static inline bool ahash_request_isvirt(struct ahash_request *req)
diff --git a/include/cxl/features.h b/include/cxl/features.h
index a3bb34694c06..5f7f842765a5 100644
--- a/include/cxl/features.h
+++ b/include/cxl/features.h
@@ -66,7 +66,7 @@ struct cxl_memdev;
 #ifdef CONFIG_CXL_FEATURES
 inline struct cxl_features_state *to_cxlfs(struct cxl_dev_state *cxlds);
 int devm_cxl_setup_features(struct cxl_dev_state *cxlds);
-int devm_cxl_setup_fwctl(struct cxl_memdev *cxlmd);
+int devm_cxl_setup_fwctl(struct device *host, struct cxl_memdev *cxlmd);
 #else
 static inline struct cxl_features_state *to_cxlfs(struct cxl_dev_state *cxlds)
 {
@@ -78,7 +78,8 @@ static inline int devm_cxl_setup_features(struct cxl_dev_state *cxlds)
 	return -EOPNOTSUPP;
 }
 
-static inline int devm_cxl_setup_fwctl(struct cxl_memdev *cxlmd)
+static inline int devm_cxl_setup_fwctl(struct device *host,
+				       struct cxl_memdev *cxlmd)
 {
 	return -EOPNOTSUPP;
 }
diff --git a/include/drm/bridge/analogix_dp.h b/include/drm/bridge/analogix_dp.h
index 6002c5666031..cf17646c1310 100644
--- a/include/drm/bridge/analogix_dp.h
+++ b/include/drm/bridge/analogix_dp.h
@@ -10,16 +10,18 @@
 #include <drm/drm_crtc.h>
 
 struct analogix_dp_device;
+struct drm_dp_aux;
 
 enum analogix_dp_devtype {
 	EXYNOS_DP,
 	RK3288_DP,
 	RK3399_EDP,
+	RK3588_EDP,
 };
 
 static inline bool is_rockchip(enum analogix_dp_devtype type)
 {
-	return type == RK3288_DP || type == RK3399_EDP;
+	return type == RK3288_DP || type == RK3399_EDP || type == RK3588_EDP;
 }
 
 struct analogix_dp_plat_data {
@@ -48,4 +50,7 @@ void analogix_dp_unbind(struct analogix_dp_device *dp);
 int analogix_dp_start_crc(struct drm_connector *connector);
 int analogix_dp_stop_crc(struct drm_connector *connector);
 
+struct analogix_dp_plat_data *analogix_dp_aux_to_plat_data(struct drm_dp_aux *aux);
+struct drm_dp_aux *analogix_dp_get_aux(struct analogix_dp_device *dp);
+
 #endif /* _ANALOGIX_DP_H_ */
diff --git a/include/drm/display/drm_dp.h b/include/drm/display/drm_dp.h
index c413ef68f9a3..3001c0b6e7bb 100644
--- a/include/drm/display/drm_dp.h
+++ b/include/drm/display/drm_dp.h
@@ -1025,6 +1025,7 @@
 #define DP_EDP_GENERAL_CAP_2		    0x703
 # define DP_EDP_OVERDRIVE_ENGINE_ENABLED		(1 << 0)
 # define DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE		(1 << 4)
+# define DP_EDP_SMOOTH_BRIGHTNESS_CAPABLE		(1 << 6) /* eDP 2.0 */
 
 #define DP_EDP_GENERAL_CAP_3		    0x704    /* eDP 1.4 */
 # define DP_EDP_X_REGION_CAP_MASK			(0xf << 0)
diff --git a/include/drm/display/drm_dp_helper.h b/include/drm/display/drm_dp_helper.h
index 5ae4241959f2..e4ca35143ff9 100644
--- a/include/drm/display/drm_dp_helper.h
+++ b/include/drm/display/drm_dp_helper.h
@@ -518,6 +518,11 @@ struct drm_dp_aux {
 	 * @powered_down: If true then the remote endpoint is powered down.
 	 */
 	bool powered_down;
+
+	/**
+	 * @no_zero_sized: If the hw can't use zero sized transfers (NVIDIA)
+	 */
+	bool no_zero_sized;
 };
 
 int drm_dp_dpcd_probe(struct drm_dp_aux *aux, unsigned int offset);
@@ -528,13 +533,72 @@ ssize_t drm_dp_dpcd_write(struct drm_dp_aux *aux, unsigned int offset,
 			  void *buffer, size_t size);
 
 /**
+ * drm_dp_dpcd_read_data() - read a series of bytes from the DPCD
+ * @aux: DisplayPort AUX channel (SST or MST)
+ * @offset: address of the (first) register to read
+ * @buffer: buffer to store the register values
+ * @size: number of bytes in @buffer
+ *
+ * Returns zero (0) on success, or a negative error
+ * code on failure. -EIO is returned if the request was NAKed by the sink or
+ * if the retry count was exceeded. If not all bytes were transferred, this
+ * function returns -EPROTO. Errors from the underlying AUX channel transfer
+ * function, with the exception of -EBUSY (which causes the transaction to
+ * be retried), are propagated to the caller.
+ */
+static inline int drm_dp_dpcd_read_data(struct drm_dp_aux *aux,
+					unsigned int offset,
+					void *buffer, size_t size)
+{
+	int ret;
+
+	ret = drm_dp_dpcd_read(aux, offset, buffer, size);
+	if (ret < 0)
+		return ret;
+	if (ret < size)
+		return -EPROTO;
+
+	return 0;
+}
+
+/**
+ * drm_dp_dpcd_write_data() - write a series of bytes to the DPCD
+ * @aux: DisplayPort AUX channel (SST or MST)
+ * @offset: address of the (first) register to write
+ * @buffer: buffer containing the values to write
+ * @size: number of bytes in @buffer
+ *
+ * Returns zero (0) on success, or a negative error
+ * code on failure. -EIO is returned if the request was NAKed by the sink or
+ * if the retry count was exceeded. If not all bytes were transferred, this
+ * function returns -EPROTO. Errors from the underlying AUX channel transfer
+ * function, with the exception of -EBUSY (which causes the transaction to
+ * be retried), are propagated to the caller.
+ */
+static inline int drm_dp_dpcd_write_data(struct drm_dp_aux *aux,
+					 unsigned int offset,
+					 void *buffer, size_t size)
+{
+	int ret;
+
+	ret = drm_dp_dpcd_write(aux, offset, buffer, size);
+	if (ret < 0)
+		return ret;
+	if (ret < size)
+		return -EPROTO;
+
+	return 0;
+}
+
+/**
  * drm_dp_dpcd_readb() - read a single byte from the DPCD
  * @aux: DisplayPort AUX channel
  * @offset: address of the register to read
  * @valuep: location where the value of the register will be stored
  *
  * Returns the number of bytes transferred (1) on success, or a negative
- * error code on failure.
+ * error code on failure. In most of the cases you should be using
+ * drm_dp_dpcd_read_byte() instead.
  */
 static inline ssize_t drm_dp_dpcd_readb(struct drm_dp_aux *aux,
 					unsigned int offset, u8 *valuep)
@@ -549,7 +613,8 @@ static inline ssize_t drm_dp_dpcd_readb(struct drm_dp_aux *aux,
  * @value: value to write to the register
  *
  * Returns the number of bytes transferred (1) on success, or a negative
- * error code on failure.
+ * error code on failure. In most of the cases you should be using
+ * drm_dp_dpcd_write_byte() instead.
  */
 static inline ssize_t drm_dp_dpcd_writeb(struct drm_dp_aux *aux,
 					 unsigned int offset, u8 value)
@@ -557,6 +622,34 @@ static inline ssize_t drm_dp_dpcd_writeb(struct drm_dp_aux *aux,
 	return drm_dp_dpcd_write(aux, offset, &value, 1);
 }
 
+/**
+ * drm_dp_dpcd_read_byte() - read a single byte from the DPCD
+ * @aux: DisplayPort AUX channel
+ * @offset: address of the register to read
+ * @valuep: location where the value of the register will be stored
+ *
+ * Returns zero (0) on success, or a negative error code on failure.
+ */
+static inline int drm_dp_dpcd_read_byte(struct drm_dp_aux *aux,
+					unsigned int offset, u8 *valuep)
+{
+	return drm_dp_dpcd_read_data(aux, offset, valuep, 1);
+}
+
+/**
+ * drm_dp_dpcd_write_byte() - write a single byte to the DPCD
+ * @aux: DisplayPort AUX channel
+ * @offset: address of the register to write
+ * @value: value to write to the register
+ *
+ * Returns zero (0) on success, or a negative error code on failure.
+ */
+static inline int drm_dp_dpcd_write_byte(struct drm_dp_aux *aux,
+					 unsigned int offset, u8 value)
+{
+	return drm_dp_dpcd_write_data(aux, offset, &value, 1);
+}
+
 int drm_dp_read_dpcd_caps(struct drm_dp_aux *aux,
 			  u8 dpcd[DP_RECEIVER_CAP_SIZE]);
 
@@ -566,6 +659,8 @@ int drm_dp_dpcd_read_link_status(struct drm_dp_aux *aux,
 int drm_dp_dpcd_read_phy_link_status(struct drm_dp_aux *aux,
 				     enum drm_dp_phy dp_phy,
 				     u8 link_status[DP_LINK_STATUS_SIZE]);
+int drm_dp_link_power_up(struct drm_dp_aux *aux, unsigned char revision);
+int drm_dp_link_power_down(struct drm_dp_aux *aux, unsigned char revision);
 
 int drm_dp_dpcd_write_payload(struct drm_dp_aux *aux,
 			      int vcpid, u8 start_time_slot, u8 time_slot_count);
@@ -881,5 +976,7 @@ int drm_dp_bw_channel_coding_efficiency(bool is_uhbr);
 int drm_dp_max_dprx_data_rate(int max_link_rate, int max_lanes);
 
 ssize_t drm_dp_vsc_sdp_pack(const struct drm_dp_vsc_sdp *vsc, struct dp_sdp *sdp);
+int drm_dp_link_symbol_cycles(int lane_count, int pixels, int dsc_slice_count,
+			      int bpp_x16, int symbol_size, bool is_mst);
 
 #endif /* _DRM_DP_HELPER_H_ */
diff --git a/include/drm/display/drm_hdmi_helper.h b/include/drm/display/drm_hdmi_helper.h
index 57e3b18c15ec..09145c9ee9fc 100644
--- a/include/drm/display/drm_hdmi_helper.h
+++ b/include/drm/display/drm_hdmi_helper.h
@@ -28,4 +28,10 @@ unsigned long long
 drm_hdmi_compute_mode_clock(const struct drm_display_mode *mode,
 			    unsigned int bpc, enum hdmi_colorspace fmt);
 
+void
+drm_hdmi_acr_get_n_cts(unsigned long long tmds_char_rate,
+		       unsigned int sample_rate,
+		       unsigned int *out_n,
+		       unsigned int *out_cts);
+
 #endif
diff --git a/include/drm/drm_atomic.h b/include/drm/drm_atomic.h
index 4c673f0698fe..38636a593c9d 100644
--- a/include/drm/drm_atomic.h
+++ b/include/drm/drm_atomic.h
@@ -625,6 +625,9 @@ drm_atomic_get_old_connector_for_encoder(const struct drm_atomic_state *state,
 struct drm_connector *
 drm_atomic_get_new_connector_for_encoder(const struct drm_atomic_state *state,
 					 struct drm_encoder *encoder);
+struct drm_connector *
+drm_atomic_get_connector_for_encoder(const struct drm_encoder *encoder,
+				     struct drm_modeset_acquire_ctx *ctx);
 
 struct drm_crtc *
 drm_atomic_get_old_crtc_for_encoder(struct drm_atomic_state *state,
diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index d4c75d59fa12..4e418a29a9ff 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -73,7 +73,7 @@ struct drm_bridge_funcs {
 	 *
 	 * Zero on success, error code on failure.
 	 */
-	int (*attach)(struct drm_bridge *bridge,
+	int (*attach)(struct drm_bridge *bridge, struct drm_encoder *encoder,
 		      enum drm_bridge_attach_flags flags);
 
 	/**
@@ -681,8 +681,10 @@ struct drm_bridge_funcs {
 	/**
 	 * @hdmi_audio_startup:
 	 *
-	 * Called when ASoC starts an audio stream setup. The
-	 * @hdmi_audio_startup() is optional.
+	 * Called when ASoC starts an audio stream setup.
+	 *
+	 * This callback is optional, it can be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_HDMI_AUDIO flag in their &drm_bridge->ops.
 	 *
 	 * Returns:
 	 * 0 on success, a negative error code otherwise
@@ -693,8 +695,10 @@ struct drm_bridge_funcs {
 	/**
 	 * @hdmi_audio_prepare:
 	 * Configures HDMI-encoder for audio stream. Can be called multiple
-	 * times for each setup. Mandatory if HDMI audio is enabled in the
-	 * bridge's configuration.
+	 * times for each setup.
+	 *
+	 * This callback is optional but it must be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_HDMI_AUDIO flag in their &drm_bridge->ops.
 	 *
 	 * Returns:
 	 * 0 on success, a negative error code otherwise
@@ -707,8 +711,10 @@ struct drm_bridge_funcs {
 	/**
 	 * @hdmi_audio_shutdown:
 	 *
-	 * Shut down the audio stream. Mandatory if HDMI audio is enabled in
-	 * the bridge's configuration.
+	 * Shut down the audio stream.
+	 *
+	 * This callback is optional but it must be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_HDMI_AUDIO flag in their &drm_bridge->ops.
 	 *
 	 * Returns:
 	 * 0 on success, a negative error code otherwise
@@ -719,8 +725,10 @@ struct drm_bridge_funcs {
 	/**
 	 * @hdmi_audio_mute_stream:
 	 *
-	 * Mute/unmute HDMI audio stream. The @hdmi_audio_mute_stream callback
-	 * is optional.
+	 * Mute/unmute HDMI audio stream.
+	 *
+	 * This callback is optional, it can be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_HDMI_AUDIO flag in their &drm_bridge->ops.
 	 *
 	 * Returns:
 	 * 0 on success, a negative error code otherwise
@@ -730,6 +738,65 @@ struct drm_bridge_funcs {
 				      bool enable, int direction);
 
 	/**
+	 * @dp_audio_startup:
+	 *
+	 * Called when ASoC starts a DisplayPort audio stream setup.
+	 *
+	 * This callback is optional, it can be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_DP_AUDIO flag in their &drm_bridge->ops.
+	 *
+	 * Returns:
+	 * 0 on success, a negative error code otherwise
+	 */
+	int (*dp_audio_startup)(struct drm_connector *connector,
+				struct drm_bridge *bridge);
+
+	/**
+	 * @dp_audio_prepare:
+	 * Configures DisplayPort audio stream. Can be called multiple
+	 * times for each setup.
+	 *
+	 * This callback is optional but it must be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_DP_AUDIO flag in their &drm_bridge->ops.
+	 *
+	 * Returns:
+	 * 0 on success, a negative error code otherwise
+	 */
+	int (*dp_audio_prepare)(struct drm_connector *connector,
+				struct drm_bridge *bridge,
+				struct hdmi_codec_daifmt *fmt,
+				struct hdmi_codec_params *hparms);
+
+	/**
+	 * @dp_audio_shutdown:
+	 *
+	 * Shut down the DisplayPort audio stream.
+	 *
+	 * This callback is optional but it must be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_DP_AUDIO flag in their &drm_bridge->ops.
+	 *
+	 * Returns:
+	 * 0 on success, a negative error code otherwise
+	 */
+	void (*dp_audio_shutdown)(struct drm_connector *connector,
+				  struct drm_bridge *bridge);
+
+	/**
+	 * @dp_audio_mute_stream:
+	 *
+	 * Mute/unmute DisplayPort audio stream.
+	 *
+	 * This callback is optional, it can be implemented by bridges that
+	 * set the @DRM_BRIDGE_OP_DP_AUDIO flag in their &drm_bridge->ops.
+	 *
+	 * Returns:
+	 * 0 on success, a negative error code otherwise
+	 */
+	int (*dp_audio_mute_stream)(struct drm_connector *connector,
+				    struct drm_bridge *bridge,
+				    bool enable, int direction);
+
+	/**
 	 * @debugfs_init:
 	 *
 	 * Allows bridges to create bridge-specific debugfs files.
@@ -814,6 +881,32 @@ enum drm_bridge_ops {
 	 * drivers.
 	 */
 	DRM_BRIDGE_OP_HDMI = BIT(4),
+	/**
+	 * @DRM_BRIDGE_OP_HDMI_AUDIO: The bridge provides HDMI audio operations.
+	 * Bridges that set this flag must implement the
+	 * &drm_bridge_funcs->hdmi_audio_prepare and
+	 * &drm_bridge_funcs->hdmi_audio_shutdown callbacks.
+	 *
+	 * Note: currently there can be at most one bridge in a chain that sets
+	 * this bit. This is to simplify corresponding glue code in connector
+	 * drivers. Also it is not possible to have a bridge in the chain that
+	 * sets @DRM_BRIDGE_OP_DP_AUDIO if there is a bridge that sets this
+	 * flag.
+	 */
+	DRM_BRIDGE_OP_HDMI_AUDIO = BIT(5),
+	/**
+	 * @DRM_BRIDGE_OP_DP_AUDIO: The bridge provides DisplayPort audio operations.
+	 * Bridges that set this flag must implement the
+	 * &drm_bridge_funcs->dp_audio_prepare and
+	 * &drm_bridge_funcs->dp_audio_shutdown callbacks.
+	 *
+	 * Note: currently there can be at most one bridge in a chain that sets
+	 * this bit. This is to simplify corresponding glue code in connector
+	 * drivers. Also it is not possible to have a bridge in the chain that
+	 * sets @DRM_BRIDGE_OP_HDMI_AUDIO if there is a bridge that sets this
+	 * flag.
+	 */
+	DRM_BRIDGE_OP_DP_AUDIO = BIT(6),
 };
 
 /**
@@ -840,6 +933,18 @@ struct drm_bridge {
 	const struct drm_bridge_timings *timings;
 	/** @funcs: control functions */
 	const struct drm_bridge_funcs *funcs;
+
+	/**
+	 * @container: Pointer to the private driver struct embedding this
+	 * @struct drm_bridge.
+	 */
+	void *container;
+
+	/**
+	 * @refcount: reference count of users referencing this bridge.
+	 */
+	struct kref refcount;
+
 	/** @driver_private: pointer to the bridge driver's internal context */
 	void *driver_private;
 	/** @ops: bitmask of operations supported by the bridge */
@@ -914,23 +1019,28 @@ struct drm_bridge {
 	unsigned int max_bpc;
 
 	/**
-	 * @hdmi_audio_dev: device to be used as a parent for the HDMI Codec
+	 * @hdmi_audio_dev: device to be used as a parent for the HDMI Codec if
+	 * either of @DRM_BRIDGE_OP_HDMI_AUDIO or @DRM_BRIDGE_OP_DP_AUDIO is set.
 	 */
 	struct device *hdmi_audio_dev;
 
 	/**
 	 * @hdmi_audio_max_i2s_playback_channels: maximum number of playback
-	 * I2S channels for the HDMI codec
+	 * I2S channels for the @DRM_BRIDGE_OP_HDMI_AUDIO or
+	 * @DRM_BRIDGE_OP_DP_AUDIO.
 	 */
 	int hdmi_audio_max_i2s_playback_channels;
 
 	/**
-	 * @hdmi_audio_spdif_playback: set if HDMI codec has S/PDIF playback port
+	 * @hdmi_audio_spdif_playback: set if this bridge has S/PDIF playback
+	 * port for @DRM_BRIDGE_OP_HDMI_AUDIO or @DRM_BRIDGE_OP_DP_AUDIO.
 	 */
 	unsigned int hdmi_audio_spdif_playback : 1;
 
 	/**
-	 * @hdmi_audio_dai_port: sound DAI port, -1 if it is not enabled
+	 * @hdmi_audio_dai_port: sound DAI port for either of
+	 * @DRM_BRIDGE_OP_HDMI_AUDIO and @DRM_BRIDGE_OP_DP_AUDIO, -1 if it is
+	 * not used.
 	 */
 	int hdmi_audio_dai_port;
 };
@@ -941,6 +1051,30 @@ drm_priv_to_bridge(struct drm_private_obj *priv)
 	return container_of(priv, struct drm_bridge, base);
 }
 
+struct drm_bridge *drm_bridge_get(struct drm_bridge *bridge);
+void drm_bridge_put(struct drm_bridge *bridge);
+
+void *__devm_drm_bridge_alloc(struct device *dev, size_t size, size_t offset,
+			      const struct drm_bridge_funcs *funcs);
+
+/**
+ * devm_drm_bridge_alloc - Allocate and initialize a bridge
+ * @dev: struct device of the bridge device
+ * @type: the type of the struct which contains struct &drm_bridge
+ * @member: the name of the &drm_bridge within @type
+ * @funcs: callbacks for this bridge
+ *
+ * The reference count of the returned bridge is initialized to 1. This
+ * reference will be automatically dropped via devm (by calling
+ * drm_bridge_put()) when @dev is removed.
+ *
+ * Returns:
+ * Pointer to new bridge, or ERR_PTR on failure.
+ */
+#define devm_drm_bridge_alloc(dev, type, member, funcs) \
+	((type *)__devm_drm_bridge_alloc(dev, sizeof(type), \
+					 offsetof(type, member), funcs))
+
 void drm_bridge_add(struct drm_bridge *bridge);
 int devm_drm_bridge_add(struct device *dev, struct drm_bridge *bridge);
 void drm_bridge_remove(struct drm_bridge *bridge);
@@ -958,6 +1092,38 @@ static inline struct drm_bridge *of_drm_find_bridge(struct device_node *np)
 #endif
 
 /**
+ * drm_bridge_get_current_state() - Get the current bridge state
+ * @bridge: bridge object
+ *
+ * This function must be called with the modeset lock held.
+ *
+ * RETURNS:
+ *
+ * The current bridge state, or NULL if there is none.
+ */
+static inline struct drm_bridge_state *
+drm_bridge_get_current_state(struct drm_bridge *bridge)
+{
+	if (!bridge)
+		return NULL;
+
+	/*
+	 * Only atomic bridges will have bridge->base initialized by
+	 * drm_atomic_private_obj_init(), so we need to make sure we're
+	 * working with one before we try to use the lock.
+	 */
+	if (!bridge->funcs || !bridge->funcs->atomic_reset)
+		return NULL;
+
+	drm_modeset_lock_assert_held(&bridge->base.lock);
+
+	if (!bridge->base.state)
+		return NULL;
+
+	return drm_priv_to_bridge_state(bridge->base.state);
+}
+
+/**
  * drm_bridge_get_next_bridge() - Get the next bridge in the chain
  * @bridge: bridge object
  *
@@ -1108,4 +1274,7 @@ static inline struct drm_bridge *drmm_of_get_bridge(struct drm_device *drm,
 }
 #endif
 
+void drm_bridge_debugfs_params(struct dentry *root);
+void drm_bridge_debugfs_encoder_params(struct dentry *root, struct drm_encoder *encoder);
+
 #endif
diff --git a/include/drm/drm_bridge_helper.h b/include/drm/drm_bridge_helper.h
new file mode 100644
index 000000000000..6c35b479ec2a
--- /dev/null
+++ b/include/drm/drm_bridge_helper.h
@@ -0,0 +1,12 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#ifndef __DRM_BRIDGE_HELPER_H_
+#define __DRM_BRIDGE_HELPER_H_
+
+struct drm_bridge;
+struct drm_modeset_acquire_ctx;
+
+int drm_bridge_helper_reset_crtc(struct drm_bridge *bridge,
+				 struct drm_modeset_acquire_ctx *ctx);
+
+#endif // __DRM_BRIDGE_HELPER_H_
diff --git a/include/drm/drm_device.h b/include/drm/drm_device.h
index 6ea54a578cda..e2f894f1b90a 100644
--- a/include/drm/drm_device.h
+++ b/include/drm/drm_device.h
@@ -65,6 +65,28 @@ struct drm_device {
 	struct device *dev;
 
 	/**
+	 * @dma_dev:
+	 *
+	 * Device for DMA operations. Only required if the device @dev
+	 * cannot perform DMA by itself. Should be NULL otherwise. Call
+	 * drm_dev_dma_dev() to get the DMA device instead of using this
+	 * field directly. Call drm_dev_set_dma_dev() to set this field.
+	 *
+	 * DRM devices are sometimes bound to virtual devices that cannot
+	 * perform DMA by themselves. Drivers should set this field to the
+	 * respective DMA controller.
+	 *
+	 * Devices on USB and other peripheral busses also cannot perform
+	 * DMA by themselves. The @dma_dev field should point the bus
+	 * controller that does DMA on behalve of such a device. Required
+	 * for importing buffers via dma-buf.
+	 *
+	 * If set, the DRM core automatically releases the reference on the
+	 * device.
+	 */
+	struct device *dma_dev;
+
+	/**
 	 * @managed:
 	 *
 	 * Managed resources linked to the lifetime of this &drm_device as
@@ -327,4 +349,23 @@ struct drm_device {
 	struct dentry *debugfs_root;
 };
 
+void drm_dev_set_dma_dev(struct drm_device *dev, struct device *dma_dev);
+
+/**
+ * drm_dev_dma_dev - returns the DMA device for a DRM device
+ * @dev: DRM device
+ *
+ * Returns the DMA device of the given DRM device. By default, this
+ * the DRM device's parent. See drm_dev_set_dma_dev().
+ *
+ * Returns:
+ * A DMA-capable device for the DRM device.
+ */
+static inline struct device *drm_dev_dma_dev(struct drm_device *dev)
+{
+	if (dev->dma_dev)
+		return dev->dma_dev;
+	return dev->dev;
+}
+
 #endif
diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h
index eaac5e665892..b38409670868 100644
--- a/include/drm/drm_edid.h
+++ b/include/drm/drm_edid.h
@@ -437,7 +437,7 @@ bool drm_detect_monitor_audio(const struct edid *edid);
 enum hdmi_quantization_range
 drm_default_rgb_quant_range(const struct drm_display_mode *mode);
 int drm_add_modes_noedid(struct drm_connector *connector,
-			 int hdisplay, int vdisplay);
+			 unsigned int hdisplay, unsigned int vdisplay);
 
 int drm_edid_header_is_valid(const void *edid);
 bool drm_edid_is_valid(struct edid *edid);
diff --git a/include/drm/drm_file.h b/include/drm/drm_file.h
index 94d365b22505..5c3b2aa3e69d 100644
--- a/include/drm/drm_file.h
+++ b/include/drm/drm_file.h
@@ -446,6 +446,9 @@ static inline bool drm_is_accel_client(const struct drm_file *file_priv)
 	return file_priv->minor->type == DRM_MINOR_ACCEL;
 }
 
+__printf(2, 3)
+void drm_file_err(struct drm_file *file_priv, const char *fmt, ...);
+
 void drm_file_update_pid(struct drm_file *);
 
 struct drm_minor *drm_minor_acquire(struct xarray *minors_xa, unsigned int minor_id);
diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
index 2bf893eabb4b..a3133a08267c 100644
--- a/include/drm/drm_gem.h
+++ b/include/drm/drm_gem.h
@@ -159,7 +159,8 @@ struct drm_gem_object_funcs {
 	 * @vmap:
 	 *
 	 * Returns a virtual address for the buffer. Used by the
-	 * drm_gem_dmabuf_vmap() helper.
+	 * drm_gem_dmabuf_vmap() helper. Called with a held GEM reservation
+	 * lock.
 	 *
 	 * This callback is optional.
 	 */
@@ -169,7 +170,8 @@ struct drm_gem_object_funcs {
 	 * @vunmap:
 	 *
 	 * Releases the address previously returned by @vmap. Used by the
-	 * drm_gem_dmabuf_vunmap() helper.
+	 * drm_gem_dmabuf_vunmap() helper. Called with a held GEM reservation
+	 * lock.
 	 *
 	 * This callback is optional.
 	 */
@@ -192,7 +194,8 @@ struct drm_gem_object_funcs {
 	 * @evict:
 	 *
 	 * Evicts gem object out from memory. Used by the drm_gem_object_evict()
-	 * helper. Returns 0 on success, -errno otherwise.
+	 * helper. Returns 0 on success, -errno otherwise. Called with a held
+	 * GEM reservation lock.
 	 *
 	 * This callback is optional.
 	 */
@@ -537,8 +540,8 @@ void drm_gem_put_pages(struct drm_gem_object *obj, struct page **pages,
 void drm_gem_lock(struct drm_gem_object *obj);
 void drm_gem_unlock(struct drm_gem_object *obj);
 
-int drm_gem_vmap_unlocked(struct drm_gem_object *obj, struct iosys_map *map);
-void drm_gem_vunmap_unlocked(struct drm_gem_object *obj, struct iosys_map *map);
+int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map);
+void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map);
 
 int drm_gem_objects_lookup(struct drm_file *filp, void __user *bo_handles,
 			   int count, struct drm_gem_object ***objs_out);
@@ -561,7 +564,7 @@ unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru,
 			       unsigned long *remaining,
 			       bool (*shrink)(struct drm_gem_object *obj));
 
-int drm_gem_evict(struct drm_gem_object *obj);
+int drm_gem_evict_locked(struct drm_gem_object *obj);
 
 /**
  * drm_gem_object_is_shared_for_memory_stats - helper for shared memory stats
@@ -585,8 +588,7 @@ static inline bool drm_gem_object_is_shared_for_memory_stats(struct drm_gem_obje
  */
 static inline bool drm_gem_is_imported(const struct drm_gem_object *obj)
 {
-	/* The dma-buf's priv field points to the original GEM object. */
-	return obj->dma_buf && (obj->dma_buf->priv != obj);
+	return !!obj->import_attach;
 }
 
 #ifdef CONFIG_LOCKDEP
diff --git a/include/drm/drm_gem_shmem_helper.h b/include/drm/drm_gem_shmem_helper.h
index cef5a6b5a4d6..b4f993da3cae 100644
--- a/include/drm/drm_gem_shmem_helper.h
+++ b/include/drm/drm_gem_shmem_helper.h
@@ -37,7 +37,18 @@ struct drm_gem_shmem_object {
 	 * Reference count on the pages table.
 	 * The pages are put when the count reaches zero.
 	 */
-	unsigned int pages_use_count;
+	refcount_t pages_use_count;
+
+	/**
+	 * @pages_pin_count:
+	 *
+	 * Reference count on the pinned pages table.
+	 *
+	 * Pages are hard-pinned and reside in memory if count
+	 * greater than zero. Otherwise, when count is zero, the pages are
+	 * allowed to be evicted and purged by memory shrinker.
+	 */
+	refcount_t pages_pin_count;
 
 	/**
 	 * @madv: State for madvise
@@ -71,7 +82,7 @@ struct drm_gem_shmem_object {
 	 * Reference count on the virtual address.
 	 * The address are un-mapped when the count reaches zero.
 	 */
-	unsigned int vmap_use_count;
+	refcount_t vmap_use_count;
 
 	/**
 	 * @pages_mark_dirty_on_put:
@@ -102,28 +113,28 @@ struct drm_gem_shmem_object *drm_gem_shmem_create_with_mnt(struct drm_device *de
 							   struct vfsmount *gemfs);
 void drm_gem_shmem_free(struct drm_gem_shmem_object *shmem);
 
-void drm_gem_shmem_put_pages(struct drm_gem_shmem_object *shmem);
+void drm_gem_shmem_put_pages_locked(struct drm_gem_shmem_object *shmem);
 int drm_gem_shmem_pin(struct drm_gem_shmem_object *shmem);
 void drm_gem_shmem_unpin(struct drm_gem_shmem_object *shmem);
-int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem,
-		       struct iosys_map *map);
-void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem,
-			  struct iosys_map *map);
+int drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem,
+			      struct iosys_map *map);
+void drm_gem_shmem_vunmap_locked(struct drm_gem_shmem_object *shmem,
+				 struct iosys_map *map);
 int drm_gem_shmem_mmap(struct drm_gem_shmem_object *shmem, struct vm_area_struct *vma);
 
 int drm_gem_shmem_pin_locked(struct drm_gem_shmem_object *shmem);
 void drm_gem_shmem_unpin_locked(struct drm_gem_shmem_object *shmem);
 
-int drm_gem_shmem_madvise(struct drm_gem_shmem_object *shmem, int madv);
+int drm_gem_shmem_madvise_locked(struct drm_gem_shmem_object *shmem, int madv);
 
 static inline bool drm_gem_shmem_is_purgeable(struct drm_gem_shmem_object *shmem)
 {
 	return (shmem->madv > 0) &&
-		!shmem->vmap_use_count && shmem->sgt &&
+		!refcount_read(&shmem->pages_pin_count) && shmem->sgt &&
 		!shmem->base.dma_buf && !drm_gem_is_imported(&shmem->base);
 }
 
-void drm_gem_shmem_purge(struct drm_gem_shmem_object *shmem);
+void drm_gem_shmem_purge_locked(struct drm_gem_shmem_object *shmem);
 
 struct sg_table *drm_gem_shmem_get_sg_table(struct drm_gem_shmem_object *shmem);
 struct sg_table *drm_gem_shmem_get_pages_sgt(struct drm_gem_shmem_object *shmem);
@@ -214,12 +225,12 @@ static inline struct sg_table *drm_gem_shmem_object_get_sg_table(struct drm_gem_
 }
 
 /*
- * drm_gem_shmem_object_vmap - GEM object function for drm_gem_shmem_vmap()
+ * drm_gem_shmem_object_vmap - GEM object function for drm_gem_shmem_vmap_locked()
  * @obj: GEM object
  * @map: Returns the kernel virtual address of the SHMEM GEM object's backing store.
  *
- * This function wraps drm_gem_shmem_vmap(). Drivers that employ the shmem helpers should
- * use it as their &drm_gem_object_funcs.vmap handler.
+ * This function wraps drm_gem_shmem_vmap_locked(). Drivers that employ the shmem
+ * helpers should use it as their &drm_gem_object_funcs.vmap handler.
  *
  * Returns:
  * 0 on success or a negative error code on failure.
@@ -229,7 +240,7 @@ static inline int drm_gem_shmem_object_vmap(struct drm_gem_object *obj,
 {
 	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
 
-	return drm_gem_shmem_vmap(shmem, map);
+	return drm_gem_shmem_vmap_locked(shmem, map);
 }
 
 /*
@@ -237,15 +248,15 @@ static inline int drm_gem_shmem_object_vmap(struct drm_gem_object *obj,
  * @obj: GEM object
  * @map: Kernel virtual address where the SHMEM GEM object was mapped
  *
- * This function wraps drm_gem_shmem_vunmap(). Drivers that employ the shmem helpers should
- * use it as their &drm_gem_object_funcs.vunmap handler.
+ * This function wraps drm_gem_shmem_vunmap_locked(). Drivers that employ the shmem
+ * helpers should use it as their &drm_gem_object_funcs.vunmap handler.
  */
 static inline void drm_gem_shmem_object_vunmap(struct drm_gem_object *obj,
 					       struct iosys_map *map)
 {
 	struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj);
 
-	drm_gem_shmem_vunmap(shmem, map);
+	drm_gem_shmem_vunmap_locked(shmem, map);
 }
 
 /**
diff --git a/include/drm/drm_kunit_helpers.h b/include/drm/drm_kunit_helpers.h
index 11d59ce0bac0..4948379237e9 100644
--- a/include/drm/drm_kunit_helpers.h
+++ b/include/drm/drm_kunit_helpers.h
@@ -9,6 +9,7 @@
 
 #include <kunit/test.h>
 
+struct drm_connector;
 struct drm_crtc_funcs;
 struct drm_crtc_helper_funcs;
 struct drm_device;
@@ -118,6 +119,16 @@ drm_kunit_helper_create_crtc(struct kunit *test,
 			     const struct drm_crtc_funcs *funcs,
 			     const struct drm_crtc_helper_funcs *helper_funcs);
 
+int drm_kunit_helper_enable_crtc_connector(struct kunit *test,
+					   struct drm_device *drm,
+					   struct drm_crtc *crtc,
+					   struct drm_connector *connector,
+					   const struct drm_display_mode *mode,
+					   struct drm_modeset_acquire_ctx *ctx);
+
+int drm_kunit_add_mode_destroy_action(struct kunit *test,
+				      struct drm_display_mode *mode);
+
 struct drm_display_mode *
 drm_kunit_display_mode_from_cea_vic(struct kunit *test, struct drm_device *dev,
 				    u8 video_code);
diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h
index bd40a443385c..b37860f4a895 100644
--- a/include/drm/drm_mipi_dsi.h
+++ b/include/drm/drm_mipi_dsi.h
@@ -293,6 +293,7 @@ void mipi_dsi_generic_write_multi(struct mipi_dsi_multi_context *ctx,
 				  const void *payload, size_t size);
 ssize_t mipi_dsi_generic_read(struct mipi_dsi_device *dsi, const void *params,
 			      size_t num_params, void *data, size_t size);
+u32 drm_mipi_dsi_get_input_bus_fmt(enum mipi_dsi_pixel_format dsi_format);
 
 #define mipi_dsi_msleep(ctx, delay)	\
 	do {				\
@@ -417,28 +418,6 @@ void mipi_dsi_dcs_set_tear_off_multi(struct mipi_dsi_multi_context *ctx);
 	} while (0)
 
 /**
- * mipi_dsi_dcs_write_seq - transmit a DCS command with payload
- *
- * This macro will print errors for you and will RETURN FROM THE CALLING
- * FUNCTION (yes this is non-intuitive) upon error.
- *
- * Because of the non-intuitive return behavior, THIS MACRO IS DEPRECATED.
- * Please replace calls of it with mipi_dsi_dcs_write_seq_multi().
- *
- * @dsi: DSI peripheral device
- * @cmd: Command
- * @seq: buffer containing data to be transmitted
- */
-#define mipi_dsi_dcs_write_seq(dsi, cmd, seq...)                               \
-	do {                                                                   \
-		static const u8 d[] = { cmd, seq };                            \
-		int ret;                                                       \
-		ret = mipi_dsi_dcs_write_buffer_chatty(dsi, d, ARRAY_SIZE(d)); \
-		if (ret < 0)                                                   \
-			return ret;                                            \
-	} while (0)
-
-/**
  * mipi_dsi_dcs_write_seq_multi - transmit a DCS command with payload
  *
  * This macro will print errors for you and error handling is optimized for
diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h
index 271765e2e9f2..9e524b51a001 100644
--- a/include/drm/drm_mode_config.h
+++ b/include/drm/drm_mode_config.h
@@ -532,8 +532,8 @@ struct drm_mode_config {
 	 */
 	struct list_head privobj_list;
 
-	int min_width, min_height;
-	int max_width, max_height;
+	unsigned int min_width, min_height;
+	unsigned int max_width, max_height;
 	const struct drm_mode_config_funcs *funcs;
 
 	/* output poll support */
@@ -937,6 +937,12 @@ struct drm_mode_config {
 	struct drm_property *modifiers_property;
 
 	/**
+	 * @async_modifiers_property: Plane property to list support modifier/format
+	 * combination for asynchronous flips.
+	 */
+	struct drm_property *async_modifiers_property;
+
+	/**
 	 * @size_hints_property: Plane SIZE_HINTS property.
 	 */
 	struct drm_property *size_hints_property;
diff --git a/include/drm/drm_panel.h b/include/drm/drm_panel.h
index a9c042c8dea1..843fb756a295 100644
--- a/include/drm/drm_panel.h
+++ b/include/drm/drm_panel.h
@@ -28,6 +28,7 @@
 #include <linux/errno.h>
 #include <linux/list.h>
 #include <linux/mutex.h>
+#include <linux/kref.h>
 
 struct backlight_device;
 struct dentry;
@@ -266,20 +267,60 @@ struct drm_panel {
 	 * If true then the panel has been enabled.
 	 */
 	bool enabled;
+
+	/**
+	 * @container: Pointer to the private driver struct embedding this
+	 * @struct drm_panel.
+	 */
+	void *container;
+
+	/**
+	 * @refcount: reference count of users referencing this panel.
+	 */
+	struct kref refcount;
 };
 
+void *__devm_drm_panel_alloc(struct device *dev, size_t size, size_t offset,
+			     const struct drm_panel_funcs *funcs,
+			     int connector_type);
+
+/**
+ * devm_drm_panel_alloc - Allocate and initialize a refcounted panel.
+ *
+ * @dev: struct device of the panel device
+ * @type: the type of the struct which contains struct &drm_panel
+ * @member: the name of the &drm_panel within @type
+ * @funcs: callbacks for this panel
+ * @connector_type: the connector type (DRM_MODE_CONNECTOR_*) corresponding to
+ * the panel interface
+ *
+ * The reference count of the returned panel is initialized to 1. This
+ * reference will be automatically dropped via devm (by calling
+ * drm_panel_put()) when @dev is removed.
+ *
+ * Returns:
+ * Pointer to container structure embedding the panel, ERR_PTR on failure.
+ */
+#define devm_drm_panel_alloc(dev, type, member, funcs, connector_type) \
+	((type *)__devm_drm_panel_alloc(dev, sizeof(type), \
+					offsetof(type, member), funcs, \
+					connector_type))
+
 void drm_panel_init(struct drm_panel *panel, struct device *dev,
 		    const struct drm_panel_funcs *funcs,
 		    int connector_type);
 
+struct drm_panel *drm_panel_get(struct drm_panel *panel);
+void drm_panel_put(struct drm_panel *panel);
+
 void drm_panel_add(struct drm_panel *panel);
 void drm_panel_remove(struct drm_panel *panel);
 
-int drm_panel_prepare(struct drm_panel *panel);
-int drm_panel_unprepare(struct drm_panel *panel);
+void drm_panel_prepare(struct drm_panel *panel);
+void drm_panel_unprepare(struct drm_panel *panel);
 
-int drm_panel_enable(struct drm_panel *panel);
-int drm_panel_disable(struct drm_panel *panel);
+void drm_panel_enable(struct drm_panel *panel);
+void drm_panel_disable(struct drm_panel *panel);
 
 int drm_panel_get_modes(struct drm_panel *panel, struct drm_connector *connector);
 
diff --git a/include/drm/drm_panic.h b/include/drm/drm_panic.h
index ff78d00c3da5..310c88c4d336 100644
--- a/include/drm/drm_panic.h
+++ b/include/drm/drm_panic.h
@@ -40,6 +40,16 @@ struct drm_scanout_buffer {
 	struct iosys_map map[DRM_FORMAT_MAX_PLANES];
 
 	/**
+	 * @pages: Optional, if the scanout buffer is not mapped, set this field
+	 * to the array of pages of the scanout buffer. The panic code will use
+	 * kmap_local_page_try_from_panic() to map one page at a time to write
+	 * all the pixels. This array shouldn't be allocated from the
+	 * get_scanoutbuffer() callback.
+	 * The scanout buffer should be in linear format.
+	 */
+	struct page **pages;
+
+	/**
 	 * @width: Width of the scanout buffer, in pixels.
 	 */
 	unsigned int width;
@@ -57,7 +67,7 @@ struct drm_scanout_buffer {
 	/**
 	 * @set_pixel: Optional function, to set a pixel color on the
 	 * framebuffer. It allows to handle special tiling format inside the
-	 * driver.
+	 * driver. It takes precedence over the @map and @pages fields.
 	 */
 	void (*set_pixel)(struct drm_scanout_buffer *sb, unsigned int x,
 			  unsigned int y, u32 color);
diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h
index dd718c62ac31..01479dd94e76 100644
--- a/include/drm/drm_plane.h
+++ b/include/drm/drm_plane.h
@@ -549,6 +549,23 @@ struct drm_plane_funcs {
 	 */
 	bool (*format_mod_supported)(struct drm_plane *plane, uint32_t format,
 				     uint64_t modifier);
+	/**
+	 * @format_mod_supported_async:
+	 *
+	 * This optional hook is used for the DRM to determine if for
+	 * asynchronous flip the given format/modifier combination is valid for
+	 * the plane. This allows the DRM to generate the correct format
+	 * bitmask (which formats apply to which modifier), and to validate
+	 * modifiers at atomic_check time.
+	 *
+	 * Returns:
+	 *
+	 * True if the given modifier is valid for that format on the plane.
+	 * False otherwise.
+	 */
+	bool (*format_mod_supported_async)(struct drm_plane *plane,
+					   u32 format, u64 modifier);
+
 };
 
 /**
diff --git a/include/drm/drm_print.h b/include/drm/drm_print.h
index f31eba1c7cab..ab017b05e175 100644
--- a/include/drm/drm_print.h
+++ b/include/drm/drm_print.h
@@ -345,6 +345,26 @@ drm_coredump_printer(struct drm_print_iterator *iter)
 }
 
 /**
+ * drm_coredump_printer_is_full() - DRM coredump printer output is full
+ * @p: DRM coredump printer
+ *
+ * DRM printer output is full, useful to short circuit coredump printing once
+ * printer is full.
+ *
+ * RETURNS:
+ * True if DRM coredump printer output buffer is full, False otherwise
+ */
+static inline bool drm_coredump_printer_is_full(struct drm_printer *p)
+{
+	struct drm_print_iterator *iterator = p->arg;
+
+	if (p->printfn != __drm_printfn_coredump)
+		return true;
+
+	return !iterator->remain;
+}
+
+/**
  * drm_seq_file_printer - construct a &drm_printer that outputs to &seq_file
  * @f:  the &struct seq_file to output to
  *
diff --git a/include/drm/drm_probe_helper.h b/include/drm/drm_probe_helper.h
index d6ce7b218b77..840ae5f798c2 100644
--- a/include/drm/drm_probe_helper.h
+++ b/include/drm/drm_probe_helper.h
@@ -17,7 +17,7 @@ int drm_helper_probe_detect(struct drm_connector *connector,
 			    struct drm_modeset_acquire_ctx *ctx,
 			    bool force);
 
-int drmm_kms_helper_poll_init(struct drm_device *dev);
+void drmm_kms_helper_poll_init(struct drm_device *dev);
 void drm_kms_helper_poll_init(struct drm_device *dev);
 void drm_kms_helper_poll_fini(struct drm_device *dev);
 bool drm_helper_hpd_irq_event(struct drm_device *dev);
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index 50928a7ae98e..1a7e377d4cbb 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -383,8 +383,15 @@ struct drm_sched_job {
 	struct xarray			dependencies;
 };
 
+/**
+ * enum drm_gpu_sched_stat - the scheduler's status
+ *
+ * @DRM_GPU_SCHED_STAT_NONE: Reserved. Do not use.
+ * @DRM_GPU_SCHED_STAT_NOMINAL: Operation succeeded.
+ * @DRM_GPU_SCHED_STAT_ENODEV: Error: Device is not available anymore.
+ */
 enum drm_gpu_sched_stat {
-	DRM_GPU_SCHED_STAT_NONE, /* Reserve 0 */
+	DRM_GPU_SCHED_STAT_NONE,
 	DRM_GPU_SCHED_STAT_NOMINAL,
 	DRM_GPU_SCHED_STAT_ENODEV,
 };
@@ -410,10 +417,36 @@ struct drm_sched_backend_ops {
 					 struct drm_sched_entity *s_entity);
 
 	/**
-         * @run_job: Called to execute the job once all of the dependencies
-         * have been resolved.  This may be called multiple times, if
-	 * timedout_job() has happened and drm_sched_job_recovery()
-	 * decides to try it again.
+	 * @run_job: Called to execute the job once all of the dependencies
+	 * have been resolved.
+	 *
+	 * @sched_job: the job to run
+	 *
+	 * The deprecated drm_sched_resubmit_jobs() (called by &struct
+	 * drm_sched_backend_ops.timedout_job) can invoke this again with the
+	 * same parameters. Using this is discouraged because it violates
+	 * dma_fence rules, notably dma_fence_init() has to be called on
+	 * already initialized fences for a second time. Moreover, this is
+	 * dangerous because attempts to allocate memory might deadlock with
+	 * memory management code waiting for the reset to complete.
+	 *
+	 * TODO: Document what drivers should do / use instead.
+	 *
+	 * This method is called in a workqueue context - either from the
+	 * submit_wq the driver passed through drm_sched_init(), or, if the
+	 * driver passed NULL, a separate, ordered workqueue the scheduler
+	 * allocated.
+	 *
+	 * Note that the scheduler expects to 'inherit' its own reference to
+	 * this fence from the callback. It does not invoke an extra
+	 * dma_fence_get() on it. Consequently, this callback must take a
+	 * reference for the scheduler, and additional ones for the driver's
+	 * respective needs.
+	 *
+	 * Return:
+	 * * On success: dma_fence the driver must signal once the hardware has
+	 * completed the job ("hardware fence").
+	 * * On failure: NULL or an ERR_PTR.
 	 */
 	struct dma_fence *(*run_job)(struct drm_sched_job *sched_job);
 
@@ -421,43 +454,52 @@ struct drm_sched_backend_ops {
 	 * @timedout_job: Called when a job has taken too long to execute,
 	 * to trigger GPU recovery.
 	 *
-	 * This method is called in a workqueue context.
+	 * @sched_job: The job that has timed out
+	 *
+	 * Drivers typically issue a reset to recover from GPU hangs.
+	 * This procedure looks very different depending on whether a firmware
+	 * or a hardware scheduler is being used.
+	 *
+	 * For a FIRMWARE SCHEDULER, each ring has one scheduler, and each
+	 * scheduler has one entity. Hence, the steps taken typically look as
+	 * follows:
 	 *
-	 * Drivers typically issue a reset to recover from GPU hangs, and this
-	 * procedure usually follows the following workflow:
+	 * 1. Stop the scheduler using drm_sched_stop(). This will pause the
+	 *    scheduler workqueues and cancel the timeout work, guaranteeing
+	 *    that nothing is queued while the ring is being removed.
+	 * 2. Remove the ring. The firmware will make sure that the
+	 *    corresponding parts of the hardware are resetted, and that other
+	 *    rings are not impacted.
+	 * 3. Kill the entity and the associated scheduler.
 	 *
-	 * 1. Stop the scheduler using drm_sched_stop(). This will park the
-	 *    scheduler thread and cancel the timeout work, guaranteeing that
-	 *    nothing is queued while we reset the hardware queue
-	 * 2. Try to gracefully stop non-faulty jobs (optional)
-	 * 3. Issue a GPU reset (driver-specific)
-	 * 4. Re-submit jobs using drm_sched_resubmit_jobs()
-	 * 5. Restart the scheduler using drm_sched_start(). At that point, new
-	 *    jobs can be queued, and the scheduler thread is unblocked
+	 *
+	 * For a HARDWARE SCHEDULER, a scheduler instance schedules jobs from
+	 * one or more entities to one ring. This implies that all entities
+	 * associated with the affected scheduler cannot be torn down, because
+	 * this would effectively also affect innocent userspace processes which
+	 * did not submit faulty jobs (for example).
+	 *
+	 * Consequently, the procedure to recover with a hardware scheduler
+	 * should look like this:
+	 *
+	 * 1. Stop all schedulers impacted by the reset using drm_sched_stop().
+	 * 2. Kill the entity the faulty job stems from.
+	 * 3. Issue a GPU reset on all faulty rings (driver-specific).
+	 * 4. Re-submit jobs on all schedulers impacted by re-submitting them to
+	 *    the entities which are still alive.
+	 * 5. Restart all schedulers that were stopped in step #1 using
+	 *    drm_sched_start().
 	 *
 	 * Note that some GPUs have distinct hardware queues but need to reset
 	 * the GPU globally, which requires extra synchronization between the
-	 * timeout handler of the different &drm_gpu_scheduler. One way to
-	 * achieve this synchronization is to create an ordered workqueue
-	 * (using alloc_ordered_workqueue()) at the driver level, and pass this
-	 * queue to drm_sched_init(), to guarantee that timeout handlers are
-	 * executed sequentially. The above workflow needs to be slightly
-	 * adjusted in that case:
-	 *
-	 * 1. Stop all schedulers impacted by the reset using drm_sched_stop()
-	 * 2. Try to gracefully stop non-faulty jobs on all queues impacted by
-	 *    the reset (optional)
-	 * 3. Issue a GPU reset on all faulty queues (driver-specific)
-	 * 4. Re-submit jobs on all schedulers impacted by the reset using
-	 *    drm_sched_resubmit_jobs()
-	 * 5. Restart all schedulers that were stopped in step #1 using
-	 *    drm_sched_start()
+	 * timeout handlers of different schedulers. One way to achieve this
+	 * synchronization is to create an ordered workqueue (using
+	 * alloc_ordered_workqueue()) at the driver level, and pass this queue
+	 * as drm_sched_init()'s @timeout_wq parameter. This will guarantee
+	 * that timeout handlers are executed sequentially.
 	 *
-	 * Return DRM_GPU_SCHED_STAT_NOMINAL, when all is normal,
-	 * and the underlying driver has started or completed recovery.
+	 * Return: The scheduler's status, defined by &enum drm_gpu_sched_stat
 	 *
-	 * Return DRM_GPU_SCHED_STAT_ENODEV, if the device is no longer
-	 * available, i.e. has been unplugged.
 	 */
 	enum drm_gpu_sched_stat (*timedout_job)(struct drm_sched_job *sched_job);
 
diff --git a/include/drm/intel/intel-gtt.h b/include/drm/intel/intel-gtt.h
index cb0d5b7200c7..f53bcff01f22 100644
--- a/include/drm/intel/intel-gtt.h
+++ b/include/drm/intel/intel-gtt.h
@@ -28,6 +28,8 @@ void intel_gmch_gtt_insert_sg_entries(struct sg_table *st,
 				      unsigned int pg_start,
 				      unsigned int flags);
 void intel_gmch_gtt_clear_range(unsigned int first_entry, unsigned int num_entries);
+dma_addr_t intel_gmch_gtt_read_entry(unsigned int pg,
+				     bool *is_present, bool *is_local);
 
 /* Special gtt memory types */
 #define AGP_DCACHE_MEMORY	1
diff --git a/include/drm/intel/pciids.h b/include/drm/intel/pciids.h
index 4736ea525048..d212848d07f3 100644
--- a/include/drm/intel/pciids.h
+++ b/include/drm/intel/pciids.h
@@ -850,6 +850,7 @@
 	MACRO__(0xE20C, ## __VA_ARGS__), \
 	MACRO__(0xE20D, ## __VA_ARGS__), \
 	MACRO__(0xE210, ## __VA_ARGS__), \
+	MACRO__(0xE211, ## __VA_ARGS__), \
 	MACRO__(0xE212, ## __VA_ARGS__), \
 	MACRO__(0xE215, ## __VA_ARGS__), \
 	MACRO__(0xE216, ## __VA_ARGS__)
diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h
index 903cd1030110..cf027558b6db 100644
--- a/include/drm/ttm/ttm_bo.h
+++ b/include/drm/ttm/ttm_bo.h
@@ -172,7 +172,6 @@ struct ttm_bo_kmap_obj {
  * @gfp_retry_mayfail: Set the __GFP_RETRY_MAYFAIL when allocation pages.
  * @allow_res_evict: Allow eviction of reserved BOs. Can be used when multiple
  * BOs share the same reservation object.
- * @force_alloc: Don't check the memory account during suspend or CPU page
  * faults. Should only be used by TTM internally.
  * @resv: Reservation object to allow reserved evictions with.
  * @bytes_moved: Statistics on how many bytes have been moved.
@@ -185,7 +184,6 @@ struct ttm_operation_ctx {
 	bool no_wait_gpu;
 	bool gfp_retry_mayfail;
 	bool allow_res_evict;
-	bool force_alloc;
 	struct dma_resv *resv;
 	uint64_t bytes_moved;
 };
diff --git a/include/kunit/test.h b/include/kunit/test.h
index 0ffb97c78566..39c768f87dc9 100644
--- a/include/kunit/test.h
+++ b/include/kunit/test.h
@@ -67,7 +67,7 @@ enum kunit_status {
 
 /*
  * Speed Attribute is stored as an enum and separated into categories of
- * speed: very_slowm, slow, and normal. These speeds are relative to
+ * speed: very_slow, slow, and normal. These speeds are relative to
  * other KUnit tests.
  *
  * Note: unset speed attribute acts as default of KUNIT_SPEED_NORMAL.
diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h
index 8e7af9a03b41..e721148c95d0 100644
--- a/include/linux/backing-dev.h
+++ b/include/linux/backing-dev.h
@@ -249,6 +249,7 @@ static inline struct bdi_writeback *inode_to_wb(const struct inode *inode)
 {
 #ifdef CONFIG_LOCKDEP
 	WARN_ON_ONCE(debug_locks &&
+		     (inode->i_sb->s_iflags & SB_I_CGROUPWB) &&
 		     (!lockdep_is_held(&inode->i_lock) &&
 		      !lockdep_is_held(&inode->i_mapping->i_pages.xa_lock) &&
 		      !lockdep_is_held(&inode->i_wb->list_lock)));
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index e39c45bc0a97..9a1f0ee40b56 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -712,23 +712,6 @@ static inline bool blk_queue_is_zoned(struct request_queue *q)
 		(q->limits.features & BLK_FEAT_ZONED);
 }
 
-#ifdef CONFIG_BLK_DEV_ZONED
-static inline unsigned int disk_nr_zones(struct gendisk *disk)
-{
-	return disk->nr_zones;
-}
-bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs);
-#else /* CONFIG_BLK_DEV_ZONED */
-static inline unsigned int disk_nr_zones(struct gendisk *disk)
-{
-	return 0;
-}
-static inline bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
-{
-	return false;
-}
-#endif /* CONFIG_BLK_DEV_ZONED */
-
 static inline unsigned int disk_zone_no(struct gendisk *disk, sector_t sector)
 {
 	if (!blk_queue_is_zoned(disk->queue))
@@ -736,11 +719,6 @@ static inline unsigned int disk_zone_no(struct gendisk *disk, sector_t sector)
 	return sector >> ilog2(disk->queue->limits.chunk_sectors);
 }
 
-static inline unsigned int bdev_nr_zones(struct block_device *bdev)
-{
-	return disk_nr_zones(bdev->bd_disk);
-}
-
 static inline unsigned int bdev_max_open_zones(struct block_device *bdev)
 {
 	return bdev->bd_disk->queue->limits.max_open_zones;
@@ -847,6 +825,51 @@ static inline u64 sb_bdev_nr_blocks(struct super_block *sb)
 		(sb->s_blocksize_bits - SECTOR_SHIFT);
 }
 
+#ifdef CONFIG_BLK_DEV_ZONED
+static inline unsigned int disk_nr_zones(struct gendisk *disk)
+{
+	return disk->nr_zones;
+}
+bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs);
+
+/**
+ * disk_zone_capacity - returns the zone capacity of zone containing @sector
+ * @disk:	disk to work with
+ * @sector:	sector number within the querying zone
+ *
+ * Returns the zone capacity of a zone containing @sector. @sector can be any
+ * sector in the zone.
+ */
+static inline unsigned int disk_zone_capacity(struct gendisk *disk,
+					      sector_t sector)
+{
+	sector_t zone_sectors = disk->queue->limits.chunk_sectors;
+
+	if (sector + zone_sectors >= get_capacity(disk))
+		return disk->last_zone_capacity;
+	return disk->zone_capacity;
+}
+static inline unsigned int bdev_zone_capacity(struct block_device *bdev,
+					      sector_t pos)
+{
+	return disk_zone_capacity(bdev->bd_disk, pos);
+}
+#else /* CONFIG_BLK_DEV_ZONED */
+static inline unsigned int disk_nr_zones(struct gendisk *disk)
+{
+	return 0;
+}
+static inline bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
+{
+	return false;
+}
+#endif /* CONFIG_BLK_DEV_ZONED */
+
+static inline unsigned int bdev_nr_zones(struct block_device *bdev)
+{
+	return disk_nr_zones(bdev->bd_disk);
+}
+
 int bdev_disk_changed(struct gendisk *disk, bool invalidate);
 
 void put_disk(struct gendisk *disk);
@@ -1614,6 +1637,7 @@ static inline void bio_end_io_acct(struct bio *bio, unsigned long start_time)
 	return bio_end_io_acct_remapped(bio, start_time, bio->bi_bdev);
 }
 
+int bdev_validate_blocksize(struct block_device *bdev, int block_size);
 int set_blocksize(struct file *file, int size);
 
 int lookup_bdev(const char *pathname, dev_t *dev);
@@ -1670,10 +1694,6 @@ int bd_prepare_to_claim(struct block_device *bdev, void *holder,
 		const struct blk_holder_ops *hops);
 void bd_abort_claiming(struct block_device *bdev, void *holder);
 
-/* just for blk-cgroup, don't use elsewhere */
-struct block_device *blkdev_get_no_open(dev_t dev);
-void blkdev_put_no_open(struct block_device *bdev);
-
 struct block_device *I_BDEV(struct inode *inode);
 struct block_device *file_bdev(struct file *bdev_file);
 bool disk_live(struct gendisk *disk);
@@ -1685,7 +1705,7 @@ int sync_blockdev(struct block_device *bdev);
 int sync_blockdev_range(struct block_device *bdev, loff_t lstart, loff_t lend);
 int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
-void bdev_statx(struct path *, struct kstat *, u32);
+void bdev_statx(const struct path *path, struct kstat *stat, u32 request_mask);
 void printk_all_partitions(void);
 int __init early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
@@ -1703,8 +1723,8 @@ static inline int sync_blockdev_nowait(struct block_device *bdev)
 static inline void sync_bdevs(bool wait)
 {
 }
-static inline void bdev_statx(struct path *path, struct kstat *stat,
-				u32 request_mask)
+static inline void bdev_statx(const struct path *path, struct kstat *stat,
+		u32 request_mask)
 {
 }
 static inline void printk_all_partitions(void)
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index f0a4ad7839b6..0029ff880e27 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -34,6 +34,7 @@ enum bh_state_bits {
 	BH_Meta,	/* Buffer contains metadata */
 	BH_Prio,	/* Buffer should be submitted with REQ_PRIO */
 	BH_Defer_Completion, /* Defer AIO completion to workqueue */
+	BH_Migrate,     /* Buffer is being migrated (norefs) */
 
 	BH_PrivateStart,/* not a state bit, but the first bit available
 			 * for private allocation by other entities
@@ -222,6 +223,8 @@ void __wait_on_buffer(struct buffer_head *);
 wait_queue_head_t *bh_waitq_head(struct buffer_head *bh);
 struct buffer_head *__find_get_block(struct block_device *bdev, sector_t block,
 			unsigned size);
+struct buffer_head *__find_get_block_nonatomic(struct block_device *bdev,
+			sector_t block, unsigned size);
 struct buffer_head *bdev_getblk(struct block_device *bdev, sector_t block,
 		unsigned size, gfp_t gfp);
 void __brelse(struct buffer_head *);
@@ -397,6 +400,12 @@ sb_find_get_block(struct super_block *sb, sector_t block)
 	return __find_get_block(sb->s_bdev, block, sb->s_blocksize);
 }
 
+static inline struct buffer_head *
+sb_find_get_block_nonatomic(struct super_block *sb, sector_t block)
+{
+	return __find_get_block_nonatomic(sb->s_bdev, block, sb->s_blocksize);
+}
+
 static inline void
 map_bh(struct buffer_head *bh, struct super_block *sb, sector_t block)
 {
diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h
index d55b30057a45..50b14a5661c7 100644
--- a/include/linux/ceph/osd_client.h
+++ b/include/linux/ceph/osd_client.h
@@ -490,9 +490,6 @@ extern void osd_req_op_extent_osd_data_pages(struct ceph_osd_request *,
 					struct page **pages, u64 length,
 					u32 alignment, bool pages_from_pool,
 					bool own_pages);
-extern void osd_req_op_extent_osd_data_pagelist(struct ceph_osd_request *,
-					unsigned int which,
-					struct ceph_pagelist *pagelist);
 #ifdef CONFIG_BLOCK
 void osd_req_op_extent_osd_data_bio(struct ceph_osd_request *osd_req,
 				    unsigned int which,
@@ -509,9 +506,6 @@ void osd_req_op_extent_osd_data_bvec_pos(struct ceph_osd_request *osd_req,
 void osd_req_op_extent_osd_iter(struct ceph_osd_request *osd_req,
 				unsigned int which, struct iov_iter *iter);
 
-extern void osd_req_op_cls_request_data_pagelist(struct ceph_osd_request *,
-					unsigned int which,
-					struct ceph_pagelist *pagelist);
 extern void osd_req_op_cls_request_data_pages(struct ceph_osd_request *,
 					unsigned int which,
 					struct page **pages, u64 length,
diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index 485b651869d9..5bc8f55c8cca 100644
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -710,6 +710,7 @@ struct cgroup_subsys {
 	void (*css_released)(struct cgroup_subsys_state *css);
 	void (*css_free)(struct cgroup_subsys_state *css);
 	void (*css_reset)(struct cgroup_subsys_state *css);
+	void (*css_killed)(struct cgroup_subsys_state *css);
 	void (*css_rstat_flush)(struct cgroup_subsys_state *css, int cpu);
 	int (*css_extra_stat_show)(struct seq_file *seq,
 				   struct cgroup_subsys_state *css);
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 28e999f2c642..e7da3c3b098b 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -344,7 +344,7 @@ static inline u64 cgroup_id(const struct cgroup *cgrp)
  */
 static inline bool css_is_dying(struct cgroup_subsys_state *css)
 {
-	return !(css->flags & CSS_NO_REF) && percpu_ref_is_dying(&css->refcnt);
+	return css->flags & CSS_DYING;
 }
 
 static inline void cgroup_get(struct cgroup *cgrp)
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 400fee6427a5..7a5b391dcc01 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -776,8 +776,8 @@ int cpufreq_frequency_table_verify(struct cpufreq_policy_data *policy,
 int cpufreq_generic_frequency_table_verify(struct cpufreq_policy_data *policy);
 
 int cpufreq_table_index_unsorted(struct cpufreq_policy *policy,
-				 unsigned int target_freq,
-				 unsigned int relation);
+				 unsigned int target_freq, unsigned int min,
+				 unsigned int max, unsigned int relation);
 int cpufreq_frequency_table_get_index(struct cpufreq_policy *policy,
 		unsigned int freq);
 
@@ -840,12 +840,12 @@ static inline int cpufreq_table_find_index_dl(struct cpufreq_policy *policy,
 	return best;
 }
 
-/* Works only on sorted freq-tables */
-static inline int cpufreq_table_find_index_l(struct cpufreq_policy *policy,
-					     unsigned int target_freq,
-					     bool efficiencies)
+static inline int find_index_l(struct cpufreq_policy *policy,
+			       unsigned int target_freq,
+			       unsigned int min, unsigned int max,
+			       bool efficiencies)
 {
-	target_freq = clamp_val(target_freq, policy->min, policy->max);
+	target_freq = clamp_val(target_freq, min, max);
 
 	if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING)
 		return cpufreq_table_find_index_al(policy, target_freq,
@@ -855,6 +855,14 @@ static inline int cpufreq_table_find_index_l(struct cpufreq_policy *policy,
 						   efficiencies);
 }
 
+/* Works only on sorted freq-tables */
+static inline int cpufreq_table_find_index_l(struct cpufreq_policy *policy,
+					     unsigned int target_freq,
+					     bool efficiencies)
+{
+	return find_index_l(policy, target_freq, policy->min, policy->max, efficiencies);
+}
+
 /* Find highest freq at or below target in a table in ascending order */
 static inline int cpufreq_table_find_index_ah(struct cpufreq_policy *policy,
 					      unsigned int target_freq,
@@ -908,12 +916,12 @@ static inline int cpufreq_table_find_index_dh(struct cpufreq_policy *policy,
 	return best;
 }
 
-/* Works only on sorted freq-tables */
-static inline int cpufreq_table_find_index_h(struct cpufreq_policy *policy,
-					     unsigned int target_freq,
-					     bool efficiencies)
+static inline int find_index_h(struct cpufreq_policy *policy,
+			       unsigned int target_freq,
+			       unsigned int min, unsigned int max,
+			       bool efficiencies)
 {
-	target_freq = clamp_val(target_freq, policy->min, policy->max);
+	target_freq = clamp_val(target_freq, min, max);
 
 	if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING)
 		return cpufreq_table_find_index_ah(policy, target_freq,
@@ -923,6 +931,14 @@ static inline int cpufreq_table_find_index_h(struct cpufreq_policy *policy,
 						   efficiencies);
 }
 
+/* Works only on sorted freq-tables */
+static inline int cpufreq_table_find_index_h(struct cpufreq_policy *policy,
+					     unsigned int target_freq,
+					     bool efficiencies)
+{
+	return find_index_h(policy, target_freq, policy->min, policy->max, efficiencies);
+}
+
 /* Find closest freq to target in a table in ascending order */
 static inline int cpufreq_table_find_index_ac(struct cpufreq_policy *policy,
 					      unsigned int target_freq,
@@ -993,12 +1009,12 @@ static inline int cpufreq_table_find_index_dc(struct cpufreq_policy *policy,
 	return best;
 }
 
-/* Works only on sorted freq-tables */
-static inline int cpufreq_table_find_index_c(struct cpufreq_policy *policy,
-					     unsigned int target_freq,
-					     bool efficiencies)
+static inline int find_index_c(struct cpufreq_policy *policy,
+			       unsigned int target_freq,
+			       unsigned int min, unsigned int max,
+			       bool efficiencies)
 {
-	target_freq = clamp_val(target_freq, policy->min, policy->max);
+	target_freq = clamp_val(target_freq, min, max);
 
 	if (policy->freq_table_sorted == CPUFREQ_TABLE_SORTED_ASCENDING)
 		return cpufreq_table_find_index_ac(policy, target_freq,
@@ -1008,7 +1024,17 @@ static inline int cpufreq_table_find_index_c(struct cpufreq_policy *policy,
 						   efficiencies);
 }
 
-static inline bool cpufreq_is_in_limits(struct cpufreq_policy *policy, int idx)
+/* Works only on sorted freq-tables */
+static inline int cpufreq_table_find_index_c(struct cpufreq_policy *policy,
+					     unsigned int target_freq,
+					     bool efficiencies)
+{
+	return find_index_c(policy, target_freq, policy->min, policy->max, efficiencies);
+}
+
+static inline bool cpufreq_is_in_limits(struct cpufreq_policy *policy,
+					unsigned int min, unsigned int max,
+					int idx)
 {
 	unsigned int freq;
 
@@ -1017,11 +1043,13 @@ static inline bool cpufreq_is_in_limits(struct cpufreq_policy *policy, int idx)
 
 	freq = policy->freq_table[idx].frequency;
 
-	return freq == clamp_val(freq, policy->min, policy->max);
+	return freq == clamp_val(freq, min, max);
 }
 
 static inline int cpufreq_frequency_table_target(struct cpufreq_policy *policy,
 						 unsigned int target_freq,
+						 unsigned int min,
+						 unsigned int max,
 						 unsigned int relation)
 {
 	bool efficiencies = policy->efficiencies_available &&
@@ -1032,29 +1060,26 @@ static inline int cpufreq_frequency_table_target(struct cpufreq_policy *policy,
 	relation &= ~CPUFREQ_RELATION_E;
 
 	if (unlikely(policy->freq_table_sorted == CPUFREQ_TABLE_UNSORTED))
-		return cpufreq_table_index_unsorted(policy, target_freq,
-						    relation);
+		return cpufreq_table_index_unsorted(policy, target_freq, min,
+						    max, relation);
 retry:
 	switch (relation) {
 	case CPUFREQ_RELATION_L:
-		idx = cpufreq_table_find_index_l(policy, target_freq,
-						 efficiencies);
+		idx = find_index_l(policy, target_freq, min, max, efficiencies);
 		break;
 	case CPUFREQ_RELATION_H:
-		idx = cpufreq_table_find_index_h(policy, target_freq,
-						 efficiencies);
+		idx = find_index_h(policy, target_freq, min, max, efficiencies);
 		break;
 	case CPUFREQ_RELATION_C:
-		idx = cpufreq_table_find_index_c(policy, target_freq,
-						 efficiencies);
+		idx = find_index_c(policy, target_freq, min, max, efficiencies);
 		break;
 	default:
 		WARN_ON_ONCE(1);
 		return 0;
 	}
 
-	/* Limit frequency index to honor policy->min/max */
-	if (!cpufreq_is_in_limits(policy, idx) && efficiencies) {
+	/* Limit frequency index to honor min and max */
+	if (!cpufreq_is_in_limits(policy, min, max, idx) && efficiencies) {
 		efficiencies = false;
 		goto retry;
 	}
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 8d1395f945bf..e9f07e37dd6f 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -173,65 +173,59 @@ struct dentry_operations {
  */
 
 /* d_flags entries */
-#define DCACHE_OP_HASH			BIT(0)
-#define DCACHE_OP_COMPARE		BIT(1)
-#define DCACHE_OP_REVALIDATE		BIT(2)
-#define DCACHE_OP_DELETE		BIT(3)
-#define DCACHE_OP_PRUNE			BIT(4)
-
-#define	DCACHE_DISCONNECTED		BIT(5)
-     /* This dentry is possibly not currently connected to the dcache tree, in
-      * which case its parent will either be itself, or will have this flag as
-      * well.  nfsd will not use a dentry with this bit set, but will first
-      * endeavour to clear the bit either by discovering that it is connected,
-      * or by performing lookup operations.   Any filesystem which supports
-      * nfsd_operations MUST have a lookup function which, if it finds a
-      * directory inode with a DCACHE_DISCONNECTED dentry, will d_move that
-      * dentry into place and return that dentry rather than the passed one,
-      * typically using d_splice_alias. */
-
-#define DCACHE_REFERENCED		BIT(6) /* Recently used, don't discard. */
-
-#define DCACHE_DONTCACHE		BIT(7) /* Purge from memory on final dput() */
-
-#define DCACHE_CANT_MOUNT		BIT(8)
-#define DCACHE_GENOCIDE			BIT(9)
-#define DCACHE_SHRINK_LIST		BIT(10)
-
-#define DCACHE_OP_WEAK_REVALIDATE	BIT(11)
-
-#define DCACHE_NFSFS_RENAMED		BIT(12)
-     /* this dentry has been "silly renamed" and has to be deleted on the last
-      * dput() */
-#define DCACHE_FSNOTIFY_PARENT_WATCHED	BIT(13)
-     /* Parent inode is watched by some fsnotify listener */
-
-#define DCACHE_DENTRY_KILLED		BIT(14)
-
-#define DCACHE_MOUNTED			BIT(15) /* is a mountpoint */
-#define DCACHE_NEED_AUTOMOUNT		BIT(16) /* handle automount on this dir */
-#define DCACHE_MANAGE_TRANSIT		BIT(17) /* manage transit from this dirent */
+enum dentry_flags {
+	DCACHE_OP_HASH			= BIT(0),
+	DCACHE_OP_COMPARE		= BIT(1),
+	DCACHE_OP_REVALIDATE		= BIT(2),
+	DCACHE_OP_DELETE		= BIT(3),
+	DCACHE_OP_PRUNE			= BIT(4),
+	/*
+	 * This dentry is possibly not currently connected to the dcache tree,
+	 * in which case its parent will either be itself, or will have this
+	 * flag as well.  nfsd will not use a dentry with this bit set, but will
+	 * first endeavour to clear the bit either by discovering that it is
+	 * connected, or by performing lookup operations.  Any filesystem which
+	 * supports nfsd_operations MUST have a lookup function which, if it
+	 * finds a directory inode with a DCACHE_DISCONNECTED dentry, will
+	 * d_move that dentry into place and return that dentry rather than the
+	 * passed one, typically using d_splice_alias.
+	 */
+	DCACHE_DISCONNECTED		= BIT(5),
+	DCACHE_REFERENCED		= BIT(6),	/* Recently used, don't discard. */
+	DCACHE_DONTCACHE		= BIT(7),	/* Purge from memory on final dput() */
+	DCACHE_CANT_MOUNT		= BIT(8),
+	DCACHE_GENOCIDE			= BIT(9),
+	DCACHE_SHRINK_LIST		= BIT(10),
+	DCACHE_OP_WEAK_REVALIDATE	= BIT(11),
+	/*
+	 * this dentry has been "silly renamed" and has to be deleted on the
+	 * last dput()
+	 */
+	DCACHE_NFSFS_RENAMED		= BIT(12),
+	DCACHE_FSNOTIFY_PARENT_WATCHED	= BIT(13),	/* Parent inode is watched by some fsnotify listener */
+	DCACHE_DENTRY_KILLED		= BIT(14),
+	DCACHE_MOUNTED			= BIT(15),	/* is a mountpoint */
+	DCACHE_NEED_AUTOMOUNT		= BIT(16),	/* handle automount on this dir */
+	DCACHE_MANAGE_TRANSIT		= BIT(17),	/* manage transit from this dirent */
+	DCACHE_LRU_LIST			= BIT(18),
+	DCACHE_ENTRY_TYPE		= (7 << 19),	/* bits 19..21 are for storing type: */
+	DCACHE_MISS_TYPE		= (0 << 19),	/* Negative dentry */
+	DCACHE_WHITEOUT_TYPE		= (1 << 19),	/* Whiteout dentry (stop pathwalk) */
+	DCACHE_DIRECTORY_TYPE		= (2 << 19),	/* Normal directory */
+	DCACHE_AUTODIR_TYPE		= (3 << 19),	/* Lookupless directory (presumed automount) */
+	DCACHE_REGULAR_TYPE		= (4 << 19),	/* Regular file type */
+	DCACHE_SPECIAL_TYPE		= (5 << 19),	/* Other file type */
+	DCACHE_SYMLINK_TYPE		= (6 << 19),	/* Symlink */
+	DCACHE_NOKEY_NAME		= BIT(22),	/* Encrypted name encoded without key */
+	DCACHE_OP_REAL			= BIT(23),
+	DCACHE_PAR_LOOKUP		= BIT(24),	/* being looked up (with parent locked shared) */
+	DCACHE_DENTRY_CURSOR		= BIT(25),
+	DCACHE_NORCU			= BIT(26),	/* No RCU delay for freeing */
+};
+
 #define DCACHE_MANAGED_DENTRY \
 	(DCACHE_MOUNTED|DCACHE_NEED_AUTOMOUNT|DCACHE_MANAGE_TRANSIT)
 
-#define DCACHE_LRU_LIST			BIT(18)
-
-#define DCACHE_ENTRY_TYPE		(7 << 19) /* bits 19..21 are for storing type: */
-#define DCACHE_MISS_TYPE		(0 << 19) /* Negative dentry */
-#define DCACHE_WHITEOUT_TYPE		(1 << 19) /* Whiteout dentry (stop pathwalk) */
-#define DCACHE_DIRECTORY_TYPE		(2 << 19) /* Normal directory */
-#define DCACHE_AUTODIR_TYPE		(3 << 19) /* Lookupless directory (presumed automount) */
-#define DCACHE_REGULAR_TYPE		(4 << 19) /* Regular file type */
-#define DCACHE_SPECIAL_TYPE		(5 << 19) /* Other file type */
-#define DCACHE_SYMLINK_TYPE		(6 << 19) /* Symlink */
-
-#define DCACHE_NOKEY_NAME		BIT(22) /* Encrypted name encoded without key */
-#define DCACHE_OP_REAL			BIT(23)
-
-#define DCACHE_PAR_LOOKUP		BIT(24) /* being looked up (with parent locked shared) */
-#define DCACHE_DENTRY_CURSOR		BIT(25)
-#define DCACHE_NORCU			BIT(26) /* No RCU delay for freeing */
-
 extern seqlock_t rename_lock;
 
 /*
diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h
index 36216d28d8bd..544f8f8c3f44 100644
--- a/include/linux/dma-buf.h
+++ b/include/linux/dma-buf.h
@@ -35,15 +35,6 @@ struct dma_buf_attachment;
  */
 struct dma_buf_ops {
 	/**
-	  * @cache_sgt_mapping:
-	  *
-	  * If true the framework will cache the first mapping made for each
-	  * attachment. This avoids creating mappings for attachments multiple
-	  * times.
-	  */
-	bool cache_sgt_mapping;
-
-	/**
 	 * @attach:
 	 *
 	 * This is called from dma_buf_attach() to make sure that a given
@@ -493,8 +484,6 @@ struct dma_buf_attach_ops {
  * @dmabuf: buffer for this attachment.
  * @dev: device attached to the buffer.
  * @node: list of dma_buf_attachment, protected by dma_resv lock of the dmabuf.
- * @sgt: cached mapping.
- * @dir: direction of cached mapping.
  * @peer2peer: true if the importer can handle peer resources without pages.
  * @priv: exporter specific attachment data.
  * @importer_ops: importer operations for this attachment, if provided
@@ -514,8 +503,6 @@ struct dma_buf_attachment {
 	struct dma_buf *dmabuf;
 	struct device *dev;
 	struct list_head node;
-	struct sg_table *sgt;
-	enum dma_data_direction dir;
 	bool peer2peer;
 	const struct dma_buf_attach_ops *importer_ops;
 	void *importer_priv;
@@ -583,20 +570,6 @@ static inline bool dma_buf_is_dynamic(struct dma_buf *dmabuf)
 	return !!dmabuf->ops->pin;
 }
 
-/**
- * dma_buf_attachment_is_dynamic - check if a DMA-buf attachment uses dynamic
- * mappings
- * @attach: the DMA-buf attachment to check
- *
- * Returns true if a DMA-buf importer wants to call the map/unmap functions with
- * the dma_resv lock held.
- */
-static inline bool
-dma_buf_attachment_is_dynamic(struct dma_buf_attachment *attach)
-{
-	return !!attach->importer_ops;
-}
-
 struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
 					  struct device *dev);
 struct dma_buf_attachment *
diff --git a/include/linux/dma-fence-unwrap.h b/include/linux/dma-fence-unwrap.h
index 66b1e56fbb81..62df222fe0f1 100644
--- a/include/linux/dma-fence-unwrap.h
+++ b/include/linux/dma-fence-unwrap.h
@@ -52,6 +52,8 @@ struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences,
 					   struct dma_fence **fences,
 					   struct dma_fence_unwrap *cursors);
 
+int dma_fence_dedup_array(struct dma_fence **array, int num_fences);
+
 /**
  * dma_fence_unwrap_merge - unwrap and merge fences
  *
diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h
index e7ad819962e3..b12776883d14 100644
--- a/include/linux/dma-fence.h
+++ b/include/linux/dma-fence.h
@@ -169,8 +169,8 @@ struct dma_fence_ops {
 	 * implementation know that there is another driver waiting on the
 	 * signal (ie. hw->sw case).
 	 *
-	 * This function can be called from atomic context, but not
-	 * from irq context, so normal spinlocks can be used.
+	 * This is called with irq's disabled, so only spinlocks which disable
+	 * IRQ's can be used in the code outside of this callback.
 	 *
 	 * A return value of false indicates the fence already passed,
 	 * or some failure occurred that made it impossible to enable
@@ -239,27 +239,6 @@ struct dma_fence_ops {
 	void (*release)(struct dma_fence *fence);
 
 	/**
-	 * @fence_value_str:
-	 *
-	 * Callback to fill in free-form debug info specific to this fence, like
-	 * the sequence number.
-	 *
-	 * This callback is optional.
-	 */
-	void (*fence_value_str)(struct dma_fence *fence, char *str, int size);
-
-	/**
-	 * @timeline_value_str:
-	 *
-	 * Fills in the current value of the timeline as a string, like the
-	 * sequence number. Note that the specific fence passed to this function
-	 * should not matter, drivers should only use it to look up the
-	 * corresponding timeline structures.
-	 */
-	void (*timeline_value_str)(struct dma_fence *fence,
-				   char *str, int size);
-
-	/**
 	 * @set_deadline:
 	 *
 	 * Callback to allow a fence waiter to inform the fence signaler of
diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
index b79925b1c433..85ab710ec0e7 100644
--- a/include/linux/dma-mapping.h
+++ b/include/linux/dma-mapping.h
@@ -629,10 +629,14 @@ static inline int dma_mmap_wc(struct device *dev,
 #else
 #define DEFINE_DMA_UNMAP_ADDR(ADDR_NAME)
 #define DEFINE_DMA_UNMAP_LEN(LEN_NAME)
-#define dma_unmap_addr(PTR, ADDR_NAME)           (0)
-#define dma_unmap_addr_set(PTR, ADDR_NAME, VAL)  do { } while (0)
-#define dma_unmap_len(PTR, LEN_NAME)             (0)
-#define dma_unmap_len_set(PTR, LEN_NAME, VAL)    do { } while (0)
+#define dma_unmap_addr(PTR, ADDR_NAME)           \
+	({ typeof(PTR) __p __maybe_unused = PTR; 0; })
+#define dma_unmap_addr_set(PTR, ADDR_NAME, VAL)  \
+	do { typeof(PTR) __p __maybe_unused = PTR; } while (0)
+#define dma_unmap_len(PTR, LEN_NAME)             \
+	({ typeof(PTR) __p __maybe_unused = PTR; 0; })
+#define dma_unmap_len_set(PTR, LEN_NAME, VAL)    \
+	do { typeof(PTR) __p __maybe_unused = PTR; } while (0)
 #endif
 
 #endif /* _LINUX_DMA_MAPPING_H */
diff --git a/include/linux/file_ref.h b/include/linux/file_ref.h
index 7db62fbc0500..31551e4cb8f3 100644
--- a/include/linux/file_ref.h
+++ b/include/linux/file_ref.h
@@ -61,7 +61,6 @@ static inline void file_ref_init(file_ref_t *ref, unsigned long cnt)
 	atomic_long_set(&ref->refcnt, cnt - 1);
 }
 
-bool __file_ref_put_badval(file_ref_t *ref, unsigned long cnt);
 bool __file_ref_put(file_ref_t *ref, unsigned long cnt);
 
 /**
@@ -178,20 +177,14 @@ static __always_inline __must_check bool file_ref_put(file_ref_t *ref)
  */
 static __always_inline __must_check bool file_ref_put_close(file_ref_t *ref)
 {
-	long old, new;
+	long old;
 
 	old = atomic_long_read(&ref->refcnt);
-	do {
-		if (unlikely(old < 0))
-			return __file_ref_put_badval(ref, old);
-
-		if (old == FILE_REF_ONEREF)
-			new = FILE_REF_DEAD;
-		else
-			new = old - 1;
-	} while (!atomic_long_try_cmpxchg(&ref->refcnt, &old, new));
-
-	return new == FILE_REF_DEAD;
+	if (likely(old == FILE_REF_ONEREF)) {
+		if (likely(atomic_long_try_cmpxchg(&ref->refcnt, &old, FILE_REF_DEAD)))
+			return true;
+	}
+	return file_ref_put(ref);
 }
 
 /**
diff --git a/include/linux/firmware/cirrus/cs_dsp_test_utils.h b/include/linux/firmware/cirrus/cs_dsp_test_utils.h
index 4f87a908ab4f..ecd821ed8064 100644
--- a/include/linux/firmware/cirrus/cs_dsp_test_utils.h
+++ b/include/linux/firmware/cirrus/cs_dsp_test_utils.h
@@ -104,7 +104,6 @@ unsigned int cs_dsp_mock_num_dsp_words_to_num_packed_regs(unsigned int num_dsp_w
 unsigned int cs_dsp_mock_xm_header_get_alg_base_in_words(struct cs_dsp_test *priv,
 							 unsigned int alg_id,
 							 int mem_type);
-unsigned int cs_dsp_mock_xm_header_get_fw_version_from_regmap(struct cs_dsp_test *priv);
 unsigned int cs_dsp_mock_xm_header_get_fw_version(struct cs_dsp_mock_xm_header *header);
 void cs_dsp_mock_xm_header_drop_from_regmap_cache(struct cs_dsp_test *priv);
 int cs_dsp_mock_xm_header_write_to_regmap(struct cs_dsp_mock_xm_header *header);
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 6cd8d1d28b8b..fc27b53c58c2 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -907,21 +907,6 @@ extern void fsnotify_wait_marks_destroyed(void);
 /* Clear all of the marks of a group attached to a given object type */
 extern void fsnotify_clear_marks_by_group(struct fsnotify_group *group,
 					  unsigned int obj_type);
-/* run all the marks in a group, and clear all of the vfsmount marks */
-static inline void fsnotify_clear_vfsmount_marks_by_group(struct fsnotify_group *group)
-{
-	fsnotify_clear_marks_by_group(group, FSNOTIFY_OBJ_TYPE_VFSMOUNT);
-}
-/* run all the marks in a group, and clear all of the inode marks */
-static inline void fsnotify_clear_inode_marks_by_group(struct fsnotify_group *group)
-{
-	fsnotify_clear_marks_by_group(group, FSNOTIFY_OBJ_TYPE_INODE);
-}
-/* run all the marks in a group, and clear all of the sn marks */
-static inline void fsnotify_clear_sb_marks_by_group(struct fsnotify_group *group)
-{
-	fsnotify_clear_marks_by_group(group, FSNOTIFY_OBJ_TYPE_SB);
-}
 extern void fsnotify_get_mark(struct fsnotify_mark *mark);
 extern void fsnotify_put_mark(struct fsnotify_mark *mark);
 extern void fsnotify_finish_user_wait(struct fsnotify_iter_info *iter_info);
diff --git a/include/linux/fwnode.h b/include/linux/fwnode.h
index 6fa0a268d538..097be89487bf 100644
--- a/include/linux/fwnode.h
+++ b/include/linux/fwnode.h
@@ -2,6 +2,11 @@
 /*
  * fwnode.h - Firmware device node object handle type definition.
  *
+ * This header file provides low-level data types and definitions for firmware
+ * and device property providers. The respective API header files supplied by
+ * them should contain all of the requisite data types and definitions for end
+ * users, so including it directly should not be necessary.
+ *
  * Copyright (C) 2015, Intel Corporation
  * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
  */
diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
index 45b651c05b9c..8adc8e9cb4a7 100644
--- a/include/linux/gpio/consumer.h
+++ b/include/linux/gpio/consumer.h
@@ -31,6 +31,7 @@ struct gpio_descs {
 #define GPIOD_FLAGS_BIT_DIR_OUT		BIT(1)
 #define GPIOD_FLAGS_BIT_DIR_VAL		BIT(2)
 #define GPIOD_FLAGS_BIT_OPEN_DRAIN	BIT(3)
+/* GPIOD_FLAGS_BIT_NONEXCLUSIVE is DEPRECATED, don't use in new code. */
 #define GPIOD_FLAGS_BIT_NONEXCLUSIVE	BIT(4)
 
 /**
diff --git a/include/linux/highmem-internal.h b/include/linux/highmem-internal.h
index dd100e849f5e..9a7683d79a4b 100644
--- a/include/linux/highmem-internal.h
+++ b/include/linux/highmem-internal.h
@@ -73,6 +73,14 @@ static inline void *kmap_local_page(struct page *page)
 	return __kmap_local_page_prot(page, kmap_prot);
 }
 
+static inline void *kmap_local_page_try_from_panic(struct page *page)
+{
+	if (!PageHighMem(page))
+		return page_address(page);
+	/* If the page is in HighMem, it's not safe to kmap it.*/
+	return NULL;
+}
+
 static inline void *kmap_local_folio(struct folio *folio, size_t offset)
 {
 	struct page *page = folio_page(folio, offset / PAGE_SIZE);
@@ -180,6 +188,11 @@ static inline void *kmap_local_page(struct page *page)
 	return page_address(page);
 }
 
+static inline void *kmap_local_page_try_from_panic(struct page *page)
+{
+	return page_address(page);
+}
+
 static inline void *kmap_local_folio(struct folio *folio, size_t offset)
 {
 	return page_address(&folio->page) + offset;
diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 1adcba3ddd76..1ef867bb8c44 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -345,7 +345,7 @@ static inline void hrtimer_update_function(struct hrtimer *timer,
 	if (WARN_ON_ONCE(!function))
 		return;
 #endif
-	timer->function = function;
+	ACCESS_PRIVATE(timer, function) = function;
 }
 
 /* Forward a hrtimer so it expires after now: */
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index ccce8a751e2a..3a8d35d41fda 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -440,10 +440,10 @@ static inline int __iommu_copy_struct_from_user(
 	void *dst_data, const struct iommu_user_data *src_data,
 	unsigned int data_type, size_t data_len, size_t min_len)
 {
-	if (src_data->type != data_type)
-		return -EINVAL;
 	if (WARN_ON(!dst_data || !src_data))
 		return -EINVAL;
+	if (src_data->type != data_type)
+		return -EINVAL;
 	if (src_data->len < min_len || data_len < src_data->len)
 		return -EINVAL;
 	return copy_struct_from_user(dst_data, data_len, src_data->uptr,
@@ -456,8 +456,8 @@ static inline int __iommu_copy_struct_from_user(
  *        include/uapi/linux/iommufd.h
  * @user_data: Pointer to a struct iommu_user_data for user space data info
  * @data_type: The data type of the @kdst. Must match with @user_data->type
- * @min_last: The last memember of the data structure @kdst points in the
- *            initial version.
+ * @min_last: The last member of the data structure @kdst points in the initial
+ *            version.
  * Return 0 for success, otherwise -error.
  */
 #define iommu_copy_struct_from_user(kdst, user_data, data_type, min_last) \
diff --git a/include/linux/irqchip/irq-davinci-aintc.h b/include/linux/irqchip/irq-davinci-aintc.h
deleted file mode 100644
index ea4e087fac98..000000000000
--- a/include/linux/irqchip/irq-davinci-aintc.h
+++ /dev/null
@@ -1,27 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-or-later */
-/*
- * Copyright (C) 2019 Texas Instruments
- */
-
-#ifndef _LINUX_IRQ_DAVINCI_AINTC_
-#define _LINUX_IRQ_DAVINCI_AINTC_
-
-#include <linux/ioport.h>
-
-/**
- * struct davinci_aintc_config - configuration data for davinci-aintc driver.
- *
- * @reg: register range to map
- * @num_irqs: number of HW interrupts supported by the controller
- * @prios: an array of size num_irqs containing priority settings for
- *         each interrupt
- */
-struct davinci_aintc_config {
-	struct resource reg;
-	unsigned int num_irqs;
-	u8 *prios;
-};
-
-void davinci_aintc_init(const struct davinci_aintc_config *config);
-
-#endif /* _LINUX_IRQ_DAVINCI_AINTC_ */
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 5438a1b446a6..291d49b9bf05 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -2382,7 +2382,7 @@ static inline bool kvm_is_visible_memslot(struct kvm_memory_slot *memslot)
 struct kvm_vcpu *kvm_get_running_vcpu(void);
 struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void);
 
-#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+#if IS_ENABLED(CONFIG_HAVE_KVM_IRQ_BYPASS)
 bool kvm_arch_has_irq_bypass(void);
 int kvm_arch_irq_bypass_add_producer(struct irq_bypass_consumer *,
 			   struct irq_bypass_producer *);
diff --git a/include/linux/local_lock.h b/include/linux/local_lock.h
index 1a0bc35839e3..16a2ee4f8310 100644
--- a/include/linux/local_lock.h
+++ b/include/linux/local_lock.h
@@ -52,44 +52,23 @@
 	__local_unlock_irqrestore(lock, flags)
 
 /**
- * localtry_lock_init - Runtime initialize a lock instance
- */
-#define localtry_lock_init(lock)		__localtry_lock_init(lock)
-
-/**
- * localtry_lock - Acquire a per CPU local lock
- * @lock:	The lock variable
- */
-#define localtry_lock(lock)		__localtry_lock(lock)
-
-/**
- * localtry_lock_irq - Acquire a per CPU local lock and disable interrupts
- * @lock:	The lock variable
- */
-#define localtry_lock_irq(lock)		__localtry_lock_irq(lock)
-
-/**
- * localtry_lock_irqsave - Acquire a per CPU local lock, save and disable
- *			 interrupts
- * @lock:	The lock variable
- * @flags:	Storage for interrupt flags
+ * local_lock_init - Runtime initialize a lock instance
  */
-#define localtry_lock_irqsave(lock, flags)				\
-	__localtry_lock_irqsave(lock, flags)
+#define local_trylock_init(lock)	__local_trylock_init(lock)
 
 /**
- * localtry_trylock - Try to acquire a per CPU local lock.
+ * local_trylock - Try to acquire a per CPU local lock
  * @lock:	The lock variable
  *
  * The function can be used in any context such as NMI or HARDIRQ. Due to
  * locking constrains it will _always_ fail to acquire the lock in NMI or
  * HARDIRQ context on PREEMPT_RT.
  */
-#define localtry_trylock(lock)		__localtry_trylock(lock)
+#define local_trylock(lock)		__local_trylock(lock)
 
 /**
- * localtry_trylock_irqsave - Try to acquire a per CPU local lock, save and disable
- *			      interrupts if acquired
+ * local_trylock_irqsave - Try to acquire a per CPU local lock, save and disable
+ *			   interrupts if acquired
  * @lock:	The lock variable
  * @flags:	Storage for interrupt flags
  *
@@ -97,29 +76,8 @@
  * locking constrains it will _always_ fail to acquire the lock in NMI or
  * HARDIRQ context on PREEMPT_RT.
  */
-#define localtry_trylock_irqsave(lock, flags)				\
-	__localtry_trylock_irqsave(lock, flags)
-
-/**
- * local_unlock - Release a per CPU local lock
- * @lock:	The lock variable
- */
-#define localtry_unlock(lock)		__localtry_unlock(lock)
-
-/**
- * local_unlock_irq - Release a per CPU local lock and enable interrupts
- * @lock:	The lock variable
- */
-#define localtry_unlock_irq(lock)		__localtry_unlock_irq(lock)
-
-/**
- * localtry_unlock_irqrestore - Release a per CPU local lock and restore
- *			      interrupt flags
- * @lock:	The lock variable
- * @flags:      Interrupt flags to restore
- */
-#define localtry_unlock_irqrestore(lock, flags)			\
-	__localtry_unlock_irqrestore(lock, flags)
+#define local_trylock_irqsave(lock, flags)			\
+	__local_trylock_irqsave(lock, flags)
 
 DEFINE_GUARD(local_lock, local_lock_t __percpu*,
 	     local_lock(_T),
diff --git a/include/linux/local_lock_internal.h b/include/linux/local_lock_internal.h
index 67bd13d142fa..8d5ac16a9b17 100644
--- a/include/linux/local_lock_internal.h
+++ b/include/linux/local_lock_internal.h
@@ -15,10 +15,11 @@ typedef struct {
 #endif
 } local_lock_t;
 
+/* local_trylock() and local_trylock_irqsave() only work with local_trylock_t */
 typedef struct {
 	local_lock_t	llock;
-	unsigned int	acquired;
-} localtry_lock_t;
+	u8		acquired;
+} local_trylock_t;
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 # define LOCAL_LOCK_DEBUG_INIT(lockname)		\
@@ -29,6 +30,9 @@ typedef struct {
 	},						\
 	.owner = NULL,
 
+# define LOCAL_TRYLOCK_DEBUG_INIT(lockname)		\
+	.llock = { LOCAL_LOCK_DEBUG_INIT((lockname).llock) },
+
 static inline void local_lock_acquire(local_lock_t *l)
 {
 	lock_map_acquire(&l->dep_map);
@@ -56,6 +60,7 @@ static inline void local_lock_debug_init(local_lock_t *l)
 }
 #else /* CONFIG_DEBUG_LOCK_ALLOC */
 # define LOCAL_LOCK_DEBUG_INIT(lockname)
+# define LOCAL_TRYLOCK_DEBUG_INIT(lockname)
 static inline void local_lock_acquire(local_lock_t *l) { }
 static inline void local_trylock_acquire(local_lock_t *l) { }
 static inline void local_lock_release(local_lock_t *l) { }
@@ -63,7 +68,7 @@ static inline void local_lock_debug_init(local_lock_t *l) { }
 #endif /* !CONFIG_DEBUG_LOCK_ALLOC */
 
 #define INIT_LOCAL_LOCK(lockname)	{ LOCAL_LOCK_DEBUG_INIT(lockname) }
-#define INIT_LOCALTRY_LOCK(lockname)	{ .llock = { LOCAL_LOCK_DEBUG_INIT(lockname.llock) }}
+#define INIT_LOCAL_TRYLOCK(lockname)	{ LOCAL_TRYLOCK_DEBUG_INIT(lockname) }
 
 #define __local_lock_init(lock)					\
 do {								\
@@ -76,6 +81,8 @@ do {								\
 	local_lock_debug_init(lock);				\
 } while (0)
 
+#define __local_trylock_init(lock) __local_lock_init(lock.llock)
+
 #define __spinlock_nested_bh_init(lock)				\
 do {								\
 	static struct lock_class_key __key;			\
@@ -87,149 +94,117 @@ do {								\
 	local_lock_debug_init(lock);				\
 } while (0)
 
+#define __local_lock_acquire(lock)					\
+	do {								\
+		local_trylock_t *tl;					\
+		local_lock_t *l;					\
+									\
+		l = (local_lock_t *)this_cpu_ptr(lock);			\
+		tl = (local_trylock_t *)l;				\
+		_Generic((lock),					\
+			__percpu local_trylock_t *: ({			\
+				lockdep_assert(tl->acquired == 0);	\
+				WRITE_ONCE(tl->acquired, 1);		\
+			}),						\
+			__percpu local_lock_t *: (void)0);		\
+		local_lock_acquire(l);					\
+	} while (0)
+
 #define __local_lock(lock)					\
 	do {							\
 		preempt_disable();				\
-		local_lock_acquire(this_cpu_ptr(lock));		\
+		__local_lock_acquire(lock);			\
 	} while (0)
 
 #define __local_lock_irq(lock)					\
 	do {							\
 		local_irq_disable();				\
-		local_lock_acquire(this_cpu_ptr(lock));		\
+		__local_lock_acquire(lock);			\
 	} while (0)
 
 #define __local_lock_irqsave(lock, flags)			\
 	do {							\
 		local_irq_save(flags);				\
-		local_lock_acquire(this_cpu_ptr(lock));		\
-	} while (0)
-
-#define __local_unlock(lock)					\
-	do {							\
-		local_lock_release(this_cpu_ptr(lock));		\
-		preempt_enable();				\
+		__local_lock_acquire(lock);			\
 	} while (0)
 
-#define __local_unlock_irq(lock)				\
-	do {							\
-		local_lock_release(this_cpu_ptr(lock));		\
-		local_irq_enable();				\
-	} while (0)
-
-#define __local_unlock_irqrestore(lock, flags)			\
-	do {							\
-		local_lock_release(this_cpu_ptr(lock));		\
-		local_irq_restore(flags);			\
-	} while (0)
-
-#define __local_lock_nested_bh(lock)				\
-	do {							\
-		lockdep_assert_in_softirq();			\
-		local_lock_acquire(this_cpu_ptr(lock));	\
-	} while (0)
-
-#define __local_unlock_nested_bh(lock)				\
-	local_lock_release(this_cpu_ptr(lock))
-
-/* localtry_lock_t variants */
-
-#define __localtry_lock_init(lock)				\
-do {								\
-	__local_lock_init(&(lock)->llock);			\
-	WRITE_ONCE((lock)->acquired, 0);			\
-} while (0)
-
-#define __localtry_lock(lock)					\
-	do {							\
-		localtry_lock_t *lt;				\
-		preempt_disable();				\
-		lt = this_cpu_ptr(lock);			\
-		local_lock_acquire(&lt->llock);			\
-		WRITE_ONCE(lt->acquired, 1);			\
-	} while (0)
-
-#define __localtry_lock_irq(lock)				\
-	do {							\
-		localtry_lock_t *lt;				\
-		local_irq_disable();				\
-		lt = this_cpu_ptr(lock);			\
-		local_lock_acquire(&lt->llock);			\
-		WRITE_ONCE(lt->acquired, 1);			\
-	} while (0)
-
-#define __localtry_lock_irqsave(lock, flags)			\
-	do {							\
-		localtry_lock_t *lt;				\
-		local_irq_save(flags);				\
-		lt = this_cpu_ptr(lock);			\
-		local_lock_acquire(&lt->llock);			\
-		WRITE_ONCE(lt->acquired, 1);			\
-	} while (0)
-
-#define __localtry_trylock(lock)				\
+#define __local_trylock(lock)					\
 	({							\
-		localtry_lock_t *lt;				\
-		bool _ret;					\
+		local_trylock_t *tl;				\
 								\
 		preempt_disable();				\
-		lt = this_cpu_ptr(lock);			\
-		if (!READ_ONCE(lt->acquired)) {			\
-			WRITE_ONCE(lt->acquired, 1);		\
-			local_trylock_acquire(&lt->llock);	\
-			_ret = true;				\
-		} else {					\
-			_ret = false;				\
+		tl = this_cpu_ptr(lock);			\
+		if (READ_ONCE(tl->acquired)) {			\
 			preempt_enable();			\
+			tl = NULL;				\
+		} else {					\
+			WRITE_ONCE(tl->acquired, 1);		\
+			local_trylock_acquire(			\
+				(local_lock_t *)tl);		\
 		}						\
-		_ret;						\
+		!!tl;						\
 	})
 
-#define __localtry_trylock_irqsave(lock, flags)			\
+#define __local_trylock_irqsave(lock, flags)			\
 	({							\
-		localtry_lock_t *lt;				\
-		bool _ret;					\
+		local_trylock_t *tl;				\
 								\
 		local_irq_save(flags);				\
-		lt = this_cpu_ptr(lock);			\
-		if (!READ_ONCE(lt->acquired)) {			\
-			WRITE_ONCE(lt->acquired, 1);		\
-			local_trylock_acquire(&lt->llock);	\
-			_ret = true;				\
-		} else {					\
-			_ret = false;				\
+		tl = this_cpu_ptr(lock);			\
+		if (READ_ONCE(tl->acquired)) {			\
 			local_irq_restore(flags);		\
+			tl = NULL;				\
+		} else {					\
+			WRITE_ONCE(tl->acquired, 1);		\
+			local_trylock_acquire(			\
+				(local_lock_t *)tl);		\
 		}						\
-		_ret;						\
+		!!tl;						\
 	})
 
-#define __localtry_unlock(lock)					\
+#define __local_lock_release(lock)					\
+	do {								\
+		local_trylock_t *tl;					\
+		local_lock_t *l;					\
+									\
+		l = (local_lock_t *)this_cpu_ptr(lock);			\
+		tl = (local_trylock_t *)l;				\
+		local_lock_release(l);					\
+		_Generic((lock),					\
+			__percpu local_trylock_t *: ({			\
+				lockdep_assert(tl->acquired == 1);	\
+				WRITE_ONCE(tl->acquired, 0);		\
+			}),						\
+			__percpu local_lock_t *: (void)0);		\
+	} while (0)
+
+#define __local_unlock(lock)					\
 	do {							\
-		localtry_lock_t *lt;				\
-		lt = this_cpu_ptr(lock);			\
-		WRITE_ONCE(lt->acquired, 0);			\
-		local_lock_release(&lt->llock);			\
+		__local_lock_release(lock);			\
 		preempt_enable();				\
 	} while (0)
 
-#define __localtry_unlock_irq(lock)				\
+#define __local_unlock_irq(lock)				\
 	do {							\
-		localtry_lock_t *lt;				\
-		lt = this_cpu_ptr(lock);			\
-		WRITE_ONCE(lt->acquired, 0);			\
-		local_lock_release(&lt->llock);			\
+		__local_lock_release(lock);			\
 		local_irq_enable();				\
 	} while (0)
 
-#define __localtry_unlock_irqrestore(lock, flags)		\
+#define __local_unlock_irqrestore(lock, flags)			\
 	do {							\
-		localtry_lock_t *lt;				\
-		lt = this_cpu_ptr(lock);			\
-		WRITE_ONCE(lt->acquired, 0);			\
-		local_lock_release(&lt->llock);			\
+		__local_lock_release(lock);			\
 		local_irq_restore(flags);			\
 	} while (0)
 
+#define __local_lock_nested_bh(lock)				\
+	do {							\
+		lockdep_assert_in_softirq();			\
+		local_lock_acquire(this_cpu_ptr(lock));	\
+	} while (0)
+
+#define __local_unlock_nested_bh(lock)				\
+	local_lock_release(this_cpu_ptr(lock))
+
 #else /* !CONFIG_PREEMPT_RT */
 
 /*
@@ -237,16 +212,18 @@ do {								\
  * critical section while staying preemptible.
  */
 typedef spinlock_t local_lock_t;
-typedef spinlock_t localtry_lock_t;
+typedef spinlock_t local_trylock_t;
 
 #define INIT_LOCAL_LOCK(lockname) __LOCAL_SPIN_LOCK_UNLOCKED((lockname))
-#define INIT_LOCALTRY_LOCK(lockname) INIT_LOCAL_LOCK(lockname)
+#define INIT_LOCAL_TRYLOCK(lockname) __LOCAL_SPIN_LOCK_UNLOCKED((lockname))
 
 #define __local_lock_init(l)					\
 	do {							\
 		local_spin_lock_init((l));			\
 	} while (0)
 
+#define __local_trylock_init(l)			__local_lock_init(l)
+
 #define __local_lock(__lock)					\
 	do {							\
 		migrate_disable();				\
@@ -283,17 +260,7 @@ do {								\
 	spin_unlock(this_cpu_ptr((lock)));			\
 } while (0)
 
-/* localtry_lock_t variants */
-
-#define __localtry_lock_init(lock)			__local_lock_init(lock)
-#define __localtry_lock(lock)				__local_lock(lock)
-#define __localtry_lock_irq(lock)			__local_lock(lock)
-#define __localtry_lock_irqsave(lock, flags)		__local_lock_irqsave(lock, flags)
-#define __localtry_unlock(lock)				__local_unlock(lock)
-#define __localtry_unlock_irq(lock)			__local_unlock(lock)
-#define __localtry_unlock_irqrestore(lock, flags)	__local_unlock_irqrestore(lock, flags)
-
-#define __localtry_trylock(lock)				\
+#define __local_trylock(lock)					\
 	({							\
 		int __locked;					\
 								\
@@ -308,11 +275,11 @@ do {								\
 		__locked;					\
 	})
 
-#define __localtry_trylock_irqsave(lock, flags)			\
+#define __local_trylock_irqsave(lock, flags)			\
 	({							\
 		typecheck(unsigned long, flags);		\
 		flags = 0;					\
-		__localtry_trylock(lock);			\
+		__local_trylock(lock);				\
 	})
 
 #endif /* CONFIG_PREEMPT_RT */
diff --git a/include/linux/mm.h b/include/linux/mm.h
index b7f13f087954..bf55206935c4 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1218,6 +1218,23 @@ static inline unsigned int folio_order(const struct folio *folio)
 	return folio_large_order(folio);
 }
 
+/**
+ * folio_reset_order - Reset the folio order and derived _nr_pages
+ * @folio: The folio.
+ *
+ * Reset the order and derived _nr_pages to 0. Must only be used in the
+ * process of splitting large folios.
+ */
+static inline void folio_reset_order(struct folio *folio)
+{
+	if (WARN_ON_ONCE(!folio_test_large(folio)))
+		return;
+	folio->_flags_1 &= ~0xffUL;
+#ifdef NR_PAGES_IN_LARGE_FOLIO
+	folio->_nr_pages = 0;
+#endif
+}
+
 #include <linux/huge_mm.h>
 
 /*
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 25e80b2ca7f4..6ccec1bf2896 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -967,6 +967,9 @@ struct zone {
 #ifdef CONFIG_UNACCEPTED_MEMORY
 	/* Pages to be accepted. All pages on the list are MAX_PAGE_ORDER */
 	struct list_head	unaccepted_pages;
+
+	/* To be called once the last page in the zone is accepted */
+	struct work_struct	unaccepted_cleanup;
 #endif
 
 	/* zone flags, see below */
@@ -1499,8 +1502,6 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark,
 bool zone_watermark_ok(struct zone *z, unsigned int order,
 		unsigned long mark, int highest_zoneidx,
 		unsigned int alloc_flags);
-bool zone_watermark_ok_safe(struct zone *z, unsigned int order,
-		unsigned long mark, int highest_zoneidx);
 /*
  * Memory initialization context, use to differentiate memory added by
  * the platform statically or via memory hotplug interface.
diff --git a/include/linux/module.h b/include/linux/module.h
index d94b196d5a34..b3329110d668 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -162,6 +162,8 @@ extern void cleanup_module(void);
 #define __INITRODATA_OR_MODULE __INITRODATA
 #endif /*CONFIG_MODULES*/
 
+struct module_kobject *lookup_or_create_module_kobject(const char *name);
+
 /* Generic info of form tag = "info" */
 #define MODULE_INFO(tag, info) __MODULE_INFO(tag, tag, info)
 
diff --git a/include/linux/mtd/spinand.h b/include/linux/mtd/spinand.h
index 1e748958dad4..311f145eb4e8 100644
--- a/include/linux/mtd/spinand.h
+++ b/include/linux/mtd/spinand.h
@@ -67,7 +67,7 @@
 		   SPI_MEM_OP_ADDR(2, addr, 1),				\
 		   SPI_MEM_OP_DUMMY(ndummy, 1),				\
 		   SPI_MEM_OP_DATA_IN(len, buf, 1),			\
-		   __VA_OPT__(SPI_MEM_OP_MAX_FREQ(__VA_ARGS__)))
+		   SPI_MEM_OP_MAX_FREQ(__VA_ARGS__ + 0))
 
 #define SPINAND_PAGE_READ_FROM_CACHE_FAST_OP(addr, ndummy, buf, len) \
 	SPI_MEM_OP(SPI_MEM_OP_CMD(0x0b, 1),			\
diff --git a/include/linux/namei.h b/include/linux/namei.h
index e3042176cdf4..bbaf55fb3101 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -62,6 +62,7 @@ extern struct dentry *kern_path_create(int, const char *, struct path *, unsigne
 extern struct dentry *user_path_create(int, const char __user *, struct path *, unsigned int);
 extern void done_path_create(struct path *, struct dentry *);
 extern struct dentry *kern_path_locked(const char *, struct path *);
+extern struct dentry *kern_path_locked_negative(const char *, struct path *);
 extern struct dentry *user_path_locked_at(int , const char __user *, struct path *);
 int vfs_path_parent_lookup(struct filename *filename, unsigned int flags,
 			   struct path *parent, struct qstr *last, int *type,
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index cf3b6445817b..2d11d013cabe 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -4429,6 +4429,7 @@ void linkwatch_fire_event(struct net_device *dev);
  * pending work list (if queued).
  */
 void linkwatch_sync_dev(struct net_device *dev);
+void __linkwatch_sync_dev(struct net_device *dev);
 
 /**
  *	netif_carrier_ok - test if carrier present
@@ -4974,6 +4975,7 @@ void dev_set_rx_mode(struct net_device *dev);
 int dev_set_promiscuity(struct net_device *dev, int inc);
 int netif_set_allmulti(struct net_device *dev, int inc, bool notify);
 int dev_set_allmulti(struct net_device *dev, int inc);
+void netif_state_change(struct net_device *dev);
 void netdev_state_change(struct net_device *dev);
 void __netdev_notify_peers(struct net_device *dev);
 void netdev_notify_peers(struct net_device *dev);
diff --git a/include/linux/nfs.h b/include/linux/nfs.h
index 9ad727ddfedb..0906a0b40c6a 100644
--- a/include/linux/nfs.h
+++ b/include/linux/nfs.h
@@ -55,7 +55,6 @@ enum nfs3_stable_how {
 	NFS_INVALID_STABLE_HOW = -1
 };
 
-#ifdef CONFIG_CRC32
 /**
  * nfs_fhandle_hash - calculate the crc32 hash for the filehandle
  * @fh - pointer to filehandle
@@ -67,10 +66,4 @@ static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
 {
 	return ~crc32_le(0xFFFFFFFF, &fh->data[0], fh->size);
 }
-#else /* CONFIG_CRC32 */
-static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh)
-{
-	return 0;
-}
-#endif /* CONFIG_CRC32 */
 #endif /* _LINUX_NFS_H */
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 0e8e3fd77e96..51e2bd6405cd 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -245,6 +245,8 @@ enum pci_dev_flags {
 	PCI_DEV_FLAGS_NO_RELAXED_ORDERING = (__force pci_dev_flags_t) (1 << 11),
 	/* Device does honor MSI masking despite saying otherwise */
 	PCI_DEV_FLAGS_HAS_MSI_MASKING = (__force pci_dev_flags_t) (1 << 12),
+	/* Device requires write to PCI_MSIX_ENTRY_DATA before any MSIX reads */
+	PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST = (__force pci_dev_flags_t) (1 << 13),
 };
 
 enum pci_irq_reroute_variant {
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 5a9bf15d4461..0069ba6866a4 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -823,7 +823,6 @@ struct perf_event {
 	struct irq_work			pending_disable_irq;
 	struct callback_head		pending_task;
 	unsigned int			pending_work;
-	struct rcuwait			pending_work_wait;
 
 	atomic_t			event_limit;
 
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index e2b705c14945..b50447ef1c92 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -1511,8 +1511,9 @@ static inline void track_pfn_insert(struct vm_area_struct *vma, pgprot_t *prot,
 
 /*
  * track_pfn_copy is called when a VM_PFNMAP VMA is about to get the page
- * tables copied during copy_page_range(). On success, stores the pfn to be
- * passed to untrack_pfn_copy().
+ * tables copied during copy_page_range(). Will store the pfn to be
+ * passed to untrack_pfn_copy() only if there is something to be untracked.
+ * Callers should initialize the pfn to 0.
  */
 static inline int track_pfn_copy(struct vm_area_struct *dst_vma,
 		struct vm_area_struct *src_vma, unsigned long *pfn)
@@ -1522,7 +1523,9 @@ static inline int track_pfn_copy(struct vm_area_struct *dst_vma,
 
 /*
  * untrack_pfn_copy is called when a VM_PFNMAP VMA failed to copy during
- * copy_page_range(), but after track_pfn_copy() was already called.
+ * copy_page_range(), but after track_pfn_copy() was already called. Can
+ * be called even if track_pfn_copy() did not actually track anything:
+ * handled internally.
  */
 static inline void untrack_pfn_copy(struct vm_area_struct *dst_vma,
 		unsigned long pfn)
diff --git a/include/linux/phylink.h b/include/linux/phylink.h
index 1f5773ab5660..30659b615fca 100644
--- a/include/linux/phylink.h
+++ b/include/linux/phylink.h
@@ -361,23 +361,29 @@ int mac_finish(struct phylink_config *config, unsigned int mode,
 		phy_interface_t iface);
 
 /**
- * mac_link_down() - take the link down
+ * mac_link_down() - notification that the link has gone down
  * @config: a pointer to a &struct phylink_config.
  * @mode: link autonegotiation mode
  * @interface: link &typedef phy_interface_t mode
  *
- * If @mode is not an in-band negotiation mode (as defined by
- * phylink_autoneg_inband()), force the link down and disable any
- * Energy Efficient Ethernet MAC configuration. Interface type
- * selection must be done in mac_config().
+ * Notifies the MAC that the link has gone down. This will not be called
+ * unless mac_link_up() has been previously called.
+ *
+ * The MAC should stop processing packets for transmission and reception.
+ * phylink will have called netif_carrier_off() to notify the networking
+ * stack that the link has gone down, so MAC drivers should not make this
+ * call.
+ *
+ * If @mode is %MLO_AN_INBAND, then this function must not prevent the
+ * link coming up.
  */
 void mac_link_down(struct phylink_config *config, unsigned int mode,
 		   phy_interface_t interface);
 
 /**
- * mac_link_up() - allow the link to come up
+ * mac_link_up() - notification that the link has come up
  * @config: a pointer to a &struct phylink_config.
- * @phy: any attached phy
+ * @phy: any attached phy (deprecated - please use LPI interfaces)
  * @mode: link autonegotiation mode
  * @interface: link &typedef phy_interface_t mode
  * @speed: link speed
@@ -385,7 +391,10 @@ void mac_link_down(struct phylink_config *config, unsigned int mode,
  * @tx_pause: link transmit pause enablement status
  * @rx_pause: link receive pause enablement status
  *
- * Configure the MAC for an established link.
+ * Notifies the MAC that the link has come up, and the parameters of the
+ * link as seen from the MACs point of view. If mac_link_up() has been
+ * called previously, there will be an intervening call to mac_link_down()
+ * before this method will be subsequently called.
  *
  * @speed, @duplex, @tx_pause and @rx_pause indicate the finalised link
  * settings, and should be used to configure the MAC block appropriately
@@ -397,9 +406,9 @@ void mac_link_down(struct phylink_config *config, unsigned int mode,
  * that the user wishes to override the pause settings, and this should
  * be allowed when considering the implementation of this method.
  *
- * If in-band negotiation mode is disabled, allow the link to come up. If
- * @phy is non-%NULL, configure Energy Efficient Ethernet by calling
- * phy_init_eee() and perform appropriate MAC configuration for EEE.
+ * Once configured, the MAC may begin to process packets for transmission
+ * and reception.
+ *
  * Interface type selection must be done in mac_config().
  */
 void mac_link_up(struct phylink_config *config, struct phy_device *phy,
diff --git a/include/linux/platform_data/x86/intel_pmc_ipc.h b/include/linux/platform_data/x86/intel_pmc_ipc.h
index 6e603a8c075f..1d34435b7001 100644
--- a/include/linux/platform_data/x86/intel_pmc_ipc.h
+++ b/include/linux/platform_data/x86/intel_pmc_ipc.h
@@ -36,6 +36,7 @@ struct pmc_ipc_rbuf {
  */
 static inline int intel_pmc_ipc(struct pmc_ipc_cmd *ipc_cmd, struct pmc_ipc_rbuf *rbuf)
 {
+#ifdef CONFIG_ACPI
 	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
 	union acpi_object params[PMC_IPCS_PARAM_COUNT] = {
 		{.type = ACPI_TYPE_INTEGER,},
@@ -89,6 +90,9 @@ static inline int intel_pmc_ipc(struct pmc_ipc_cmd *ipc_cmd, struct pmc_ipc_rbuf
 	}
 
 	return 0;
+#else
+	return -ENODEV;
+#endif /* CONFIG_ACPI */
 }
 
 #endif /* INTEL_PMC_IPC_H */
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index ccaaf4c7d5f6..ea39dd23a197 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -240,6 +240,6 @@ rtnl_notify_needed(const struct net *net, u16 nlflags, u32 group)
 	return (nlflags & NLM_F_ECHO) || rtnl_has_listeners(net, group);
 }
 
-void netdev_set_operstate(struct net_device *dev, int newstate);
+void netif_set_operstate(struct net_device *dev, int newstate);
 
 #endif	/* __LINUX_RTNETLINK_H */
diff --git a/include/linux/screen_info.h b/include/linux/screen_info.h
index 6a4a3cec4638..923d68e07679 100644
--- a/include/linux/screen_info.h
+++ b/include/linux/screen_info.h
@@ -126,8 +126,17 @@ static inline unsigned int screen_info_video_type(const struct screen_info *si)
 	return VIDEO_TYPE_CGA;
 }
 
+static inline u32 __screen_info_vesapm_info_base(const struct screen_info *si)
+{
+	if (si->vesapm_seg < 0xc000)
+		return 0;
+	return (si->vesapm_seg << 4) + si->vesapm_off;
+}
+
 ssize_t screen_info_resources(const struct screen_info *si, struct resource *r, size_t num);
 
+u32 __screen_info_lfb_bits_per_pixel(const struct screen_info *si);
+
 #if defined(CONFIG_PCI)
 void screen_info_apply_fixups(void);
 struct pci_dev *screen_info_pci_dev(const struct screen_info *si);
diff --git a/include/linux/virtio.h b/include/linux/virtio.h
index 4d16c13d0df5..64cb4b04be7a 100644
--- a/include/linux/virtio.h
+++ b/include/linux/virtio.h
@@ -220,6 +220,8 @@ size_t virtio_max_dma_size(const struct virtio_device *vdev);
  *    occurs.
  * @reset_done: optional function to call after transport specific reset
  *    operation has finished.
+ * @shutdown: synchronize with the device on shutdown. If provided, replaces
+ *    the virtio core implementation.
  */
 struct virtio_driver {
 	struct device_driver driver;
@@ -237,6 +239,7 @@ struct virtio_driver {
 	int (*restore)(struct virtio_device *dev);
 	int (*reset_prepare)(struct virtio_device *dev);
 	int (*reset_done)(struct virtio_device *dev);
+	void (*shutdown)(struct virtio_device *dev);
 };
 
 #define drv_to_virtio(__drv)	container_of_const(__drv, struct virtio_driver, driver)
diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h
index a8586c3058c7..797992019f9e 100644
--- a/include/net/bluetooth/hci.h
+++ b/include/net/bluetooth/hci.h
@@ -1931,6 +1931,8 @@ struct hci_cp_le_pa_create_sync {
 	__u8      sync_cte_type;
 } __packed;
 
+#define HCI_OP_LE_PA_CREATE_SYNC_CANCEL	0x2045
+
 #define HCI_OP_LE_PA_TERM_SYNC		0x2046
 struct hci_cp_le_pa_term_sync {
 	__le16    handle;
@@ -2830,7 +2832,7 @@ struct hci_evt_le_create_big_complete {
 	__le16  bis_handle[];
 } __packed;
 
-#define HCI_EVT_LE_BIG_SYNC_ESTABILISHED 0x1d
+#define HCI_EVT_LE_BIG_SYNC_ESTABLISHED 0x1d
 struct hci_evt_le_big_sync_estabilished {
 	__u8    status;
 	__u8    handle;
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index 5115da34f881..522d837a23fa 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -1113,10 +1113,8 @@ static inline struct hci_conn *hci_conn_hash_lookup_bis(struct hci_dev *hdev,
 	return NULL;
 }
 
-static inline struct hci_conn *hci_conn_hash_lookup_sid(struct hci_dev *hdev,
-							__u8 sid,
-							bdaddr_t *dst,
-							__u8 dst_type)
+static inline struct hci_conn *
+hci_conn_hash_lookup_create_pa_sync(struct hci_dev *hdev)
 {
 	struct hci_conn_hash *h = &hdev->conn_hash;
 	struct hci_conn  *c;
@@ -1124,8 +1122,10 @@ static inline struct hci_conn *hci_conn_hash_lookup_sid(struct hci_dev *hdev,
 	rcu_read_lock();
 
 	list_for_each_entry_rcu(c, &h->list, list) {
-		if (c->type != ISO_LINK  || bacmp(&c->dst, dst) ||
-		    c->dst_type != dst_type || c->sid != sid)
+		if (c->type != ISO_LINK)
+			continue;
+
+		if (!test_bit(HCI_CONN_CREATE_PA_SYNC, &c->flags))
 			continue;
 
 		rcu_read_unlock();
@@ -1524,8 +1524,6 @@ bool hci_setup_sync(struct hci_conn *conn, __u16 handle);
 void hci_sco_setup(struct hci_conn *conn, __u8 status);
 bool hci_iso_setup_path(struct hci_conn *conn);
 int hci_le_create_cis_pending(struct hci_dev *hdev);
-int hci_pa_create_sync_pending(struct hci_dev *hdev);
-int hci_le_big_create_sync_pending(struct hci_dev *hdev);
 int hci_conn_check_create_cis(struct hci_conn *conn);
 
 struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst,
@@ -1566,9 +1564,9 @@ struct hci_conn *hci_connect_bis(struct hci_dev *hdev, bdaddr_t *dst,
 				 __u8 data_len, __u8 *data);
 struct hci_conn *hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst,
 		       __u8 dst_type, __u8 sid, struct bt_iso_qos *qos);
-int hci_le_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon,
-			   struct bt_iso_qos *qos,
-			   __u16 sync_handle, __u8 num_bis, __u8 bis[]);
+int hci_conn_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon,
+			     struct bt_iso_qos *qos, __u16 sync_handle,
+			     __u8 num_bis, __u8 bis[]);
 int hci_conn_check_link_mode(struct hci_conn *conn);
 int hci_conn_check_secure(struct hci_conn *conn, __u8 sec_level);
 int hci_conn_security(struct hci_conn *conn, __u8 sec_level, __u8 auth_type,
diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index 7e2cf0cca939..72558c826aa1 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -185,3 +185,6 @@ int hci_connect_le_sync(struct hci_dev *hdev, struct hci_conn *conn);
 int hci_cancel_connect_sync(struct hci_dev *hdev, struct hci_conn *conn);
 int hci_le_conn_update_sync(struct hci_dev *hdev, struct hci_conn *conn,
 			    struct hci_conn_params *params);
+
+int hci_connect_pa_sync(struct hci_dev *hdev, struct hci_conn *conn);
+int hci_connect_big_sync(struct hci_dev *hdev, struct hci_conn *conn);
diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 5927910ec06e..6e68e359ad18 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -45,6 +45,8 @@ struct fib_rule {
 	struct fib_rule_port_range	dport_range;
 	u16			sport_mask;
 	u16			dport_mask;
+	u8                      iif_is_l3_master;
+	u8                      oif_is_l3_master;
 	struct rcu_head		rcu;
 };
 
diff --git a/include/net/flow.h b/include/net/flow.h
index 335bbc52171c..2a3f0c42f092 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -38,6 +38,7 @@ struct flowi_common {
 	__u8	flowic_flags;
 #define FLOWI_FLAG_ANYSRC		0x01
 #define FLOWI_FLAG_KNOWN_NH		0x02
+#define FLOWI_FLAG_L3MDEV_OIF		0x04
 	__u32	flowic_secid;
 	kuid_t  flowic_uid;
 	__u32		flowic_multipath_hash;
diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h
index f7fe796e8429..1eb8dad18f7e 100644
--- a/include/net/l3mdev.h
+++ b/include/net/l3mdev.h
@@ -59,6 +59,20 @@ int l3mdev_ifindex_lookup_by_table_id(enum l3mdev_type l3type, struct net *net,
 int l3mdev_fib_rule_match(struct net *net, struct flowi *fl,
 			  struct fib_lookup_arg *arg);
 
+static inline
+bool l3mdev_fib_rule_iif_match(const struct flowi *fl, int iifindex)
+{
+	return !(fl->flowi_flags & FLOWI_FLAG_L3MDEV_OIF) &&
+	       fl->flowi_l3mdev == iifindex;
+}
+
+static inline
+bool l3mdev_fib_rule_oif_match(const struct flowi *fl, int oifindex)
+{
+	return fl->flowi_flags & FLOWI_FLAG_L3MDEV_OIF &&
+	       fl->flowi_l3mdev == oifindex;
+}
+
 void l3mdev_update_flow(struct net *net, struct flowi *fl);
 
 int l3mdev_master_ifindex_rcu(const struct net_device *dev);
@@ -327,6 +341,19 @@ int l3mdev_fib_rule_match(struct net *net, struct flowi *fl,
 {
 	return 1;
 }
+
+static inline
+bool l3mdev_fib_rule_iif_match(const struct flowi *fl, int iifindex)
+{
+	return false;
+}
+
+static inline
+bool l3mdev_fib_rule_oif_match(const struct flowi *fl, int oifindex)
+{
+	return false;
+}
+
 static inline
 void l3mdev_update_flow(struct net *net, struct flowi *fl)
 {
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 31248cfdfb23..dcd288fa1bb6 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -775,6 +775,7 @@ struct sctp_transport {
 
 	/* Reference counting. */
 	refcount_t refcnt;
+	__u32	dead:1,
 		/* RTO-Pending : A flag used to track if one of the DATA
 		 *		chunks sent to this address is currently being
 		 *		used to compute a RTT. If this flag is 0,
@@ -784,7 +785,7 @@ struct sctp_transport {
 		 *		calculation completes (i.e. the DATA chunk
 		 *		is SACK'd) clear this flag.
 		 */
-	__u32	rto_pending:1,
+		rto_pending:1,
 
 		/*
 		 * hb_sent : a flag that signals that we have a pending
diff --git a/include/net/sock.h b/include/net/sock.h
index 8daf1b3b12c6..694f954258d4 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -339,6 +339,8 @@ struct sk_filter;
   *	@sk_txtime_unused: unused txtime flags
   *	@ns_tracker: tracker for netns reference
   *	@sk_user_frags: xarray of pages the user is holding a reference on.
+  *	@sk_owner: reference to the real owner of the socket that calls
+  *		   sock_lock_init_class_and_name().
   */
 struct sock {
 	/*
@@ -547,6 +549,10 @@ struct sock {
 	struct rcu_head		sk_rcu;
 	netns_tracker		ns_tracker;
 	struct xarray		sk_user_frags;
+
+#if IS_ENABLED(CONFIG_PROVE_LOCKING) && IS_ENABLED(CONFIG_MODULES)
+	struct module		*sk_owner;
+#endif
 };
 
 struct sock_bh_locked {
@@ -1583,6 +1589,35 @@ static inline void sk_mem_uncharge(struct sock *sk, int size)
 	sk_mem_reclaim(sk);
 }
 
+#if IS_ENABLED(CONFIG_PROVE_LOCKING) && IS_ENABLED(CONFIG_MODULES)
+static inline void sk_owner_set(struct sock *sk, struct module *owner)
+{
+	__module_get(owner);
+	sk->sk_owner = owner;
+}
+
+static inline void sk_owner_clear(struct sock *sk)
+{
+	sk->sk_owner = NULL;
+}
+
+static inline void sk_owner_put(struct sock *sk)
+{
+	module_put(sk->sk_owner);
+}
+#else
+static inline void sk_owner_set(struct sock *sk, struct module *owner)
+{
+}
+
+static inline void sk_owner_clear(struct sock *sk)
+{
+}
+
+static inline void sk_owner_put(struct sock *sk)
+{
+}
+#endif
 /*
  * Macro so as to not evaluate some arguments when
  * lockdep is not enabled.
@@ -1592,13 +1627,14 @@ static inline void sk_mem_uncharge(struct sock *sk, int size)
  */
 #define sock_lock_init_class_and_name(sk, sname, skey, name, key)	\
 do {									\
+	sk_owner_set(sk, THIS_MODULE);					\
 	sk->sk_lock.owned = 0;						\
 	init_waitqueue_head(&sk->sk_lock.wq);				\
 	spin_lock_init(&(sk)->sk_lock.slock);				\
 	debug_check_no_locks_freed((void *)&(sk)->sk_lock,		\
-			sizeof((sk)->sk_lock));				\
+				   sizeof((sk)->sk_lock));		\
 	lockdep_set_class_and_name(&(sk)->sk_lock.slock,		\
-				(skey), (sname));				\
+				   (skey), (sname));			\
 	lockdep_init_map(&(sk)->sk_lock.dep_map, (name), (key), 0);	\
 } while (0)
 
diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index a58ae7589d12..e8bd6ddb7b12 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -71,9 +71,6 @@ struct xdp_sock {
 	 */
 	u32 tx_budget_spent;
 
-	/* Protects generic receive. */
-	spinlock_t rx_lock;
-
 	/* Statistics */
 	u64 rx_dropped;
 	u64 rx_queue_full;
diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h
index 1dcd4d71468a..cac56e6b0869 100644
--- a/include/net/xsk_buff_pool.h
+++ b/include/net/xsk_buff_pool.h
@@ -53,6 +53,8 @@ struct xsk_buff_pool {
 	refcount_t users;
 	struct xdp_umem *umem;
 	struct work_struct work;
+	/* Protects generic receive in shared and non-shared umem mode. */
+	spinlock_t rx_lock;
 	struct list_head free_list;
 	struct list_head xskb_list;
 	u32 heads_cnt;
@@ -238,8 +240,8 @@ static inline u64 xp_get_handle(struct xdp_buff_xsk *xskb,
 		return orig_addr;
 
 	offset = xskb->xdp.data - xskb->xdp.data_hard_start;
-	orig_addr -= offset;
 	offset += pool->headroom;
+	orig_addr -= offset;
 	return orig_addr + (offset << XSK_UNALIGNED_BUF_OFFSET_SHIFT);
 }
 
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index d42eae69d9a8..901353796fbb 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -4790,7 +4790,14 @@ void roce_del_all_netdev_gids(struct ib_device *ib_dev,
 
 struct ib_ucontext *ib_uverbs_get_ucontext_file(struct ib_uverbs_file *ufile);
 
+#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS)
 int uverbs_destroy_def_handler(struct uverbs_attr_bundle *attrs);
+#else
+static inline int uverbs_destroy_def_handler(struct uverbs_attr_bundle *attrs)
+{
+	return 0;
+}
+#endif
 
 struct net_device *rdma_alloc_netdev(struct ib_device *device, u32 port_num,
 				     enum rdma_netdev_t type, const char *name,
diff --git a/include/sound/soc_sdw_utils.h b/include/sound/soc_sdw_utils.h
index 36a4a1e1d8ca..d8bd5d37131a 100644
--- a/include/sound/soc_sdw_utils.h
+++ b/include/sound/soc_sdw_utils.h
@@ -226,6 +226,7 @@ int asoc_sdw_cs_amp_init(struct snd_soc_card *card,
 			 bool playback);
 int asoc_sdw_cs_spk_feedback_rtd_init(struct snd_soc_pcm_runtime *rtd,
 				      struct snd_soc_dai *dai);
+int asoc_sdw_cs35l56_volume_limit(struct snd_soc_card *card, const char *name_prefix);
 
 /* MAXIM codec support */
 int asoc_sdw_maxim_init(struct snd_soc_card *card,
diff --git a/include/sound/ump_convert.h b/include/sound/ump_convert.h
index d099ae27f849..682499b871ea 100644
--- a/include/sound/ump_convert.h
+++ b/include/sound/ump_convert.h
@@ -19,7 +19,7 @@ struct ump_cvt_to_ump_bank {
 /* context for converting from MIDI1 byte stream to UMP packet */
 struct ump_cvt_to_ump {
 	/* MIDI1 intermediate buffer */
-	unsigned char buf[4];
+	unsigned char buf[6]; /* up to 6 bytes for SysEx */
 	int len;
 	int cmd_bytes;
 
diff --git a/include/uapi/cxl/features.h b/include/uapi/cxl/features.h
index d6db8984889f..490606d7694b 100644
--- a/include/uapi/cxl/features.h
+++ b/include/uapi/cxl/features.h
@@ -8,10 +8,19 @@
 #define _UAPI_CXL_FEATURES_H_
 
 #include <linux/types.h>
-#ifndef __KERNEL__
-#include <uuid/uuid.h>
-#else
+
+typedef unsigned char __uapi_uuid_t[16];
+
+#ifdef __KERNEL__
 #include <linux/uuid.h>
+/*
+ * Note, __uapi_uuid_t is 1-byte aligned on modern compilers and 4-byte
+ * aligned on others. Ensure that __uapi_uuid_t in a struct is placed at
+ * a 4-byte aligned offset, or the structure is packed, to ensure
+ * consistent padding.
+ */
+static_assert(sizeof(__uapi_uuid_t) == sizeof(uuid_t));
+#define __uapi_uuid_t uuid_t
 #endif
 
 /*
@@ -60,7 +69,7 @@ struct cxl_mbox_get_sup_feats_in {
  * Get Supported Features Supported Feature Entry
  */
 struct cxl_feat_entry {
-	uuid_t uuid;
+	__uapi_uuid_t uuid;
 	__le16 id;
 	__le16 get_feat_size;
 	__le16 set_feat_size;
@@ -110,7 +119,7 @@ struct cxl_mbox_get_sup_feats_out {
  * CXL spec r3.2 section 8.2.9.6.2 Table 8-99
  */
 struct cxl_mbox_get_feat_in {
-	uuid_t uuid;
+	__uapi_uuid_t uuid;
 	__le16 offset;
 	__le16 count;
 	__u8 selection;
@@ -143,7 +152,7 @@ enum cxl_get_feat_selection {
  */
 struct cxl_mbox_set_feat_in {
 	__struct_group(cxl_mbox_set_feat_hdr, hdr, /* no attrs */,
-		uuid_t uuid;
+		__uapi_uuid_t uuid;
 		__le32 flags;
 		__le16 offset;
 		__u8 version;
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 25d5c6e90a99..45c4fa13499c 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -54,6 +54,9 @@ extern "C" {
 #define DRM_AMDGPU_VM			0x13
 #define DRM_AMDGPU_FENCE_TO_HANDLE	0x14
 #define DRM_AMDGPU_SCHED		0x15
+#define DRM_AMDGPU_USERQ		0x16
+#define DRM_AMDGPU_USERQ_SIGNAL		0x17
+#define DRM_AMDGPU_USERQ_WAIT		0x18
 
 #define DRM_IOCTL_AMDGPU_GEM_CREATE	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_GEM_CREATE, union drm_amdgpu_gem_create)
 #define DRM_IOCTL_AMDGPU_GEM_MMAP	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_GEM_MMAP, union drm_amdgpu_gem_mmap)
@@ -71,6 +74,9 @@ extern "C" {
 #define DRM_IOCTL_AMDGPU_VM		DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_VM, union drm_amdgpu_vm)
 #define DRM_IOCTL_AMDGPU_FENCE_TO_HANDLE DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_FENCE_TO_HANDLE, union drm_amdgpu_fence_to_handle)
 #define DRM_IOCTL_AMDGPU_SCHED		DRM_IOW(DRM_COMMAND_BASE + DRM_AMDGPU_SCHED, union drm_amdgpu_sched)
+#define DRM_IOCTL_AMDGPU_USERQ		DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_USERQ, union drm_amdgpu_userq)
+#define DRM_IOCTL_AMDGPU_USERQ_SIGNAL	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_USERQ_SIGNAL, struct drm_amdgpu_userq_signal)
+#define DRM_IOCTL_AMDGPU_USERQ_WAIT	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_USERQ_WAIT, struct drm_amdgpu_userq_wait)
 
 /**
  * DOC: memory domains
@@ -319,6 +325,260 @@ union drm_amdgpu_ctx {
 	union drm_amdgpu_ctx_out out;
 };
 
+/* user queue IOCTL operations */
+#define AMDGPU_USERQ_OP_CREATE	1
+#define AMDGPU_USERQ_OP_FREE	2
+
+/* queue priority levels */
+/* low < normal low < normal high < high */
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_MASK  0x3
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_SHIFT 0
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_NORMAL_LOW 0
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_LOW 1
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_NORMAL_HIGH 2
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_HIGH 3 /* admin only */
+/* for queues that need access to protected content */
+#define AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE  (1 << 2)
+
+/*
+ * This structure is a container to pass input configuration
+ * info for all supported userqueue related operations.
+ * For operation AMDGPU_USERQ_OP_CREATE: user is expected
+ *  to set all fields, excep the parameter 'queue_id'.
+ * For operation AMDGPU_USERQ_OP_FREE: the only input parameter expected
+ *  to be set is 'queue_id', eveything else is ignored.
+ */
+struct drm_amdgpu_userq_in {
+	/** AMDGPU_USERQ_OP_* */
+	__u32	op;
+	/** Queue id passed for operation USERQ_OP_FREE */
+	__u32	queue_id;
+	/** the target GPU engine to execute workload (AMDGPU_HW_IP_*) */
+	__u32   ip_type;
+	/**
+	 * @doorbell_handle: the handle of doorbell GEM object
+	 * associated with this userqueue client.
+	 */
+	__u32   doorbell_handle;
+	/**
+	 * @doorbell_offset: 32-bit offset of the doorbell in the doorbell bo.
+	 * Kernel will generate absolute doorbell offset using doorbell_handle
+	 * and doorbell_offset in the doorbell bo.
+	 */
+	__u32   doorbell_offset;
+	/**
+	 * @flags: flags used for queue parameters
+	 */
+	__u32 flags;
+	/**
+	 * @queue_va: Virtual address of the GPU memory which holds the queue
+	 * object. The queue holds the workload packets.
+	 */
+	__u64   queue_va;
+	/**
+	 * @queue_size: Size of the queue in bytes, this needs to be 256-byte
+	 * aligned.
+	 */
+	__u64   queue_size;
+	/**
+	 * @rptr_va : Virtual address of the GPU memory which holds the ring RPTR.
+	 * This object must be at least 8 byte in size and aligned to 8-byte offset.
+	 */
+	__u64   rptr_va;
+	/**
+	 * @wptr_va : Virtual address of the GPU memory which holds the ring WPTR.
+	 * This object must be at least 8 byte in size and aligned to 8-byte offset.
+	 *
+	 * Queue, RPTR and WPTR can come from the same object, as long as the size
+	 * and alignment related requirements are met.
+	 */
+	__u64   wptr_va;
+	/**
+	 * @mqd: MQD (memory queue descriptor) is a set of parameters which allow
+	 * the GPU to uniquely define and identify a usermode queue.
+	 *
+	 * MQD data can be of different size for different GPU IP/engine and
+	 * their respective versions/revisions, so this points to a __u64 *
+	 * which holds IP specific MQD of this usermode queue.
+	 */
+	__u64 mqd;
+	/**
+	 * @size: size of MQD data in bytes, it must match the MQD structure
+	 * size of the respective engine/revision defined in UAPI for ex, for
+	 * gfx11 workloads, size = sizeof(drm_amdgpu_userq_mqd_gfx11).
+	 */
+	__u64 mqd_size;
+};
+
+/* The structure to carry output of userqueue ops */
+struct drm_amdgpu_userq_out {
+	/**
+	 * For operation AMDGPU_USERQ_OP_CREATE: This field contains a unique
+	 * queue ID to represent the newly created userqueue in the system, otherwise
+	 * it should be ignored.
+	 */
+	__u32	queue_id;
+	__u32 _pad;
+};
+
+union drm_amdgpu_userq {
+	struct drm_amdgpu_userq_in in;
+	struct drm_amdgpu_userq_out out;
+};
+
+/* GFX V11 IP specific MQD parameters */
+struct drm_amdgpu_userq_mqd_gfx11 {
+	/**
+	 * @shadow_va: Virtual address of the GPU memory to hold the shadow buffer.
+	 * Use AMDGPU_INFO_IOCTL to find the exact size of the object.
+	 */
+	__u64   shadow_va;
+	/**
+	 * @csa_va: Virtual address of the GPU memory to hold the CSA buffer.
+	 * Use AMDGPU_INFO_IOCTL to find the exact size of the object.
+	 */
+	__u64   csa_va;
+};
+
+/* GFX V11 SDMA IP specific MQD parameters */
+struct drm_amdgpu_userq_mqd_sdma_gfx11 {
+	/**
+	 * @csa_va: Virtual address of the GPU memory to hold the CSA buffer.
+	 * This must be a from a separate GPU object, and use AMDGPU_INFO IOCTL
+	 * to get the size.
+	 */
+	__u64   csa_va;
+};
+
+/* GFX V11 Compute IP specific MQD parameters */
+struct drm_amdgpu_userq_mqd_compute_gfx11 {
+	/**
+	 * @eop_va: Virtual address of the GPU memory to hold the EOP buffer.
+	 * This must be a from a separate GPU object, and use AMDGPU_INFO IOCTL
+	 * to get the size.
+	 */
+	__u64   eop_va;
+};
+
+/* userq signal/wait ioctl */
+struct drm_amdgpu_userq_signal {
+	/**
+	 * @queue_id: Queue handle used by the userq fence creation function
+	 * to retrieve the WPTR.
+	 */
+	__u32	queue_id;
+	__u32	pad;
+	/**
+	 * @syncobj_handles: The list of syncobj handles submitted by the user queue
+	 * job to be signaled.
+	 */
+	__u64	syncobj_handles;
+	/**
+	 * @num_syncobj_handles: A count that represents the number of syncobj handles in
+	 * @syncobj_handles.
+	 */
+	__u64	num_syncobj_handles;
+	/**
+	 * @bo_read_handles: The list of BO handles that the submitted user queue job
+	 * is using for read only. This will update BO fences in the kernel.
+	 */
+	__u64	bo_read_handles;
+	/**
+	 * @bo_write_handles: The list of BO handles that the submitted user queue job
+	 * is using for write only. This will update BO fences in the kernel.
+	 */
+	__u64	bo_write_handles;
+	/**
+	 * @num_bo_read_handles: A count that represents the number of read BO handles in
+	 * @bo_read_handles.
+	 */
+	__u32	num_bo_read_handles;
+	/**
+	 * @num_bo_write_handles: A count that represents the number of write BO handles in
+	 * @bo_write_handles.
+	 */
+	__u32	num_bo_write_handles;
+};
+
+struct drm_amdgpu_userq_fence_info {
+	/**
+	 * @va: A gpu address allocated for each queue which stores the
+	 * read pointer (RPTR) value.
+	 */
+	__u64	va;
+	/**
+	 * @value: A 64 bit value represents the write pointer (WPTR) of the
+	 * queue commands which compared with the RPTR value to signal the
+	 * fences.
+	 */
+	__u64	value;
+};
+
+struct drm_amdgpu_userq_wait {
+	/**
+	 * @waitq_id: Queue handle used by the userq wait IOCTL to retrieve the
+	 * wait queue and maintain the fence driver references in it.
+	 */
+	__u32	waitq_id;
+	__u32	pad;
+	/**
+	 * @syncobj_handles: The list of syncobj handles submitted by the user queue
+	 * job to get the va/value pairs.
+	 */
+	__u64	syncobj_handles;
+	/**
+	 * @syncobj_timeline_handles: The list of timeline syncobj handles submitted by
+	 * the user queue job to get the va/value pairs at given @syncobj_timeline_points.
+	 */
+	__u64	syncobj_timeline_handles;
+	/**
+	 * @syncobj_timeline_points: The list of timeline syncobj points submitted by the
+	 * user queue job for the corresponding @syncobj_timeline_handles.
+	 */
+	__u64	syncobj_timeline_points;
+	/**
+	 * @bo_read_handles: The list of read BO handles submitted by the user queue
+	 * job to get the va/value pairs.
+	 */
+	__u64	bo_read_handles;
+	/**
+	 * @bo_write_handles: The list of write BO handles submitted by the user queue
+	 * job to get the va/value pairs.
+	 */
+	__u64	bo_write_handles;
+	/**
+	 * @num_syncobj_timeline_handles: A count that represents the number of timeline
+	 * syncobj handles in @syncobj_timeline_handles.
+	 */
+	__u16	num_syncobj_timeline_handles;
+	/**
+	 * @num_fences: This field can be used both as input and output. As input it defines
+	 * the maximum number of fences that can be returned and as output it will specify
+	 * how many fences were actually returned from the ioctl.
+	 */
+	__u16	num_fences;
+	/**
+	 * @num_syncobj_handles: A count that represents the number of syncobj handles in
+	 * @syncobj_handles.
+	 */
+	__u32	num_syncobj_handles;
+	/**
+	 * @num_bo_read_handles: A count that represents the number of read BO handles in
+	 * @bo_read_handles.
+	 */
+	__u32	num_bo_read_handles;
+	/**
+	 * @num_bo_write_handles: A count that represents the number of write BO handles in
+	 * @bo_write_handles.
+	 */
+	__u32	num_bo_write_handles;
+	/**
+	 * @out_fences: The field is a return value from the ioctl containing the list of
+	 * address/value pairs to wait for.
+	 */
+	__u64	out_fences;
+};
+
 /* vm ioctl */
 #define AMDGPU_VM_OP_RESERVE_VMID	1
 #define AMDGPU_VM_OP_UNRESERVE_VMID	2
@@ -599,6 +859,19 @@ struct drm_amdgpu_gem_va {
 	__u64 offset_in_bo;
 	/** Specify mapping size. Must be correctly aligned. */
 	__u64 map_size;
+	/**
+	 * vm_timeline_point is a sequence number used to add new timeline point.
+	 */
+	__u64 vm_timeline_point;
+	/**
+	 * The vm page table update fence is installed in given vm_timeline_syncobj_out
+	 * at vm_timeline_point.
+	 */
+	__u32 vm_timeline_syncobj_out;
+	/** the number of syncobj handles in @input_fence_syncobj_handles */
+	__u32 num_syncobj_handles;
+	/** Array of sync object handle to wait for given input fences */
+	__u64 input_fence_syncobj_handles;
 };
 
 #define AMDGPU_HW_IP_GFX          0
@@ -940,6 +1213,8 @@ struct drm_amdgpu_cs_chunk_cp_gfx_shadow {
 #define AMDGPU_INFO_MAX_IBS			0x22
 /* query last page fault info */
 #define AMDGPU_INFO_GPUVM_FAULT			0x23
+/* query FW object size and alignment */
+#define AMDGPU_INFO_UQ_FW_AREAS			0x24
 
 #define AMDGPU_INFO_MMR_SE_INDEX_SHIFT	0
 #define AMDGPU_INFO_MMR_SE_INDEX_MASK	0xff
@@ -1093,6 +1368,7 @@ struct drm_amdgpu_info_vbios {
 #define AMDGPU_VRAM_TYPE_DDR5  10
 #define AMDGPU_VRAM_TYPE_LPDDR4 11
 #define AMDGPU_VRAM_TYPE_LPDDR5 12
+#define AMDGPU_VRAM_TYPE_HBM3E 13
 
 struct drm_amdgpu_info_device {
 	/** PCI Device ID */
@@ -1198,6 +1474,9 @@ struct drm_amdgpu_info_device {
 	__u32 csa_size;
 	/* context save area base virtual alignment for gfx11 */
 	__u32 csa_alignment;
+	/* Userq IP mask (1 << AMDGPU_HW_IP_*) */
+	__u32 userq_ip_mask;
+	__u32 pad;
 };
 
 struct drm_amdgpu_info_hw_ip {
@@ -1216,6 +1495,27 @@ struct drm_amdgpu_info_hw_ip {
 	__u32  ip_discovery_version;
 };
 
+/* GFX metadata BO sizes and alignment info (in bytes) */
+struct drm_amdgpu_info_uq_fw_areas_gfx {
+	/* shadow area size */
+	__u32 shadow_size;
+	/* shadow area base virtual mem alignment */
+	__u32 shadow_alignment;
+	/* context save area size */
+	__u32 csa_size;
+	/* context save area base virtual mem alignment */
+	__u32 csa_alignment;
+};
+
+/* IP specific fw related information used in the
+ * subquery AMDGPU_INFO_UQ_FW_AREAS
+ */
+struct drm_amdgpu_info_uq_fw_areas {
+	union {
+		struct drm_amdgpu_info_uq_fw_areas_gfx gfx;
+	};
+};
+
 struct drm_amdgpu_info_num_handles {
 	/** Max handles as supported by firmware for UVD */
 	__u32  uvd_max_handles;
@@ -1279,6 +1579,23 @@ struct drm_amdgpu_info_gpuvm_fault {
 	__u32 vmhub;
 };
 
+struct drm_amdgpu_info_uq_metadata_gfx {
+	/* shadow area size for gfx11 */
+	__u32 shadow_size;
+	/* shadow area base virtual alignment for gfx11 */
+	__u32 shadow_alignment;
+	/* context save area size for gfx11 */
+	__u32 csa_size;
+	/* context save area base virtual alignment for gfx11 */
+	__u32 csa_alignment;
+};
+
+struct drm_amdgpu_info_uq_metadata {
+	union {
+		struct drm_amdgpu_info_uq_metadata_gfx gfx;
+	};
+};
+
 /*
  * Supported GPU families
  */
diff --git a/include/uapi/drm/asahi_drm.h b/include/uapi/drm/asahi_drm.h
new file mode 100644
index 000000000000..de67f1c603af
--- /dev/null
+++ b/include/uapi/drm/asahi_drm.h
@@ -0,0 +1,1194 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright (C) The Asahi Linux Contributors
+ * Copyright (C) 2018-2023 Collabora Ltd.
+ * Copyright (C) 2014-2018 Broadcom
+ */
+#ifndef _ASAHI_DRM_H_
+#define _ASAHI_DRM_H_
+
+#include "drm.h"
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+/**
+ * DOC: Introduction to the Asahi UAPI
+ *
+ * This documentation describes the Asahi IOCTLs.
+ *
+ * Just a few generic rules about the data passed to the Asahi IOCTLs (cribbed
+ * from Panthor):
+ *
+ * - Structures must be aligned on 64-bit/8-byte. If the object is not
+ *   naturally aligned, a padding field must be added.
+ * - Fields must be explicitly aligned to their natural type alignment with
+ *   pad[0..N] fields.
+ * - All padding fields will be checked by the driver to make sure they are
+ *   zeroed.
+ * - Flags can be added, but not removed/replaced.
+ * - New fields can be added to the main structures (the structures
+ *   directly passed to the ioctl). Those fields can be added at the end of
+ *   the structure, or replace existing padding fields. Any new field being
+ *   added must preserve the behavior that existed before those fields were
+ *   added when a value of zero is passed.
+ * - New fields can be added to indirect objects (objects pointed by the
+ *   main structure), iff those objects are passed a size to reflect the
+ *   size known by the userspace driver (see
+ *   drm_asahi_cmd_header::size).
+ * - If the kernel driver is too old to know some fields, those will be
+ *   ignored if zero, and otherwise rejected (and so will be zero on output).
+ * - If userspace is too old to know some fields, those will be zeroed
+ *   (input) before the structure is parsed by the kernel driver.
+ * - Each new flag/field addition must come with a driver version update so
+ *   the userspace driver doesn't have to guess which flags are supported.
+ * - Structures should not contain unions, as this would defeat the
+ *   extensibility of such structures.
+ * - IOCTLs can't be removed or replaced. New IOCTL IDs should be placed
+ *   at the end of the drm_asahi_ioctl_id enum.
+ */
+
+/**
+ * enum drm_asahi_ioctl_id - IOCTL IDs
+ *
+ * Place new ioctls at the end, don't re-order, don't replace or remove entries.
+ *
+ * These IDs are not meant to be used directly. Use the DRM_IOCTL_ASAHI_xxx
+ * definitions instead.
+ */
+enum drm_asahi_ioctl_id {
+	/** @DRM_ASAHI_GET_PARAMS: Query device properties. */
+	DRM_ASAHI_GET_PARAMS = 0,
+
+	/** @DRM_ASAHI_GET_TIME: Query device time. */
+	DRM_ASAHI_GET_TIME,
+
+	/** @DRM_ASAHI_VM_CREATE: Create a GPU VM address space. */
+	DRM_ASAHI_VM_CREATE,
+
+	/** @DRM_ASAHI_VM_DESTROY: Destroy a VM. */
+	DRM_ASAHI_VM_DESTROY,
+
+	/** @DRM_ASAHI_VM_BIND: Bind/unbind memory to a VM. */
+	DRM_ASAHI_VM_BIND,
+
+	/** @DRM_ASAHI_GEM_CREATE: Create a buffer object. */
+	DRM_ASAHI_GEM_CREATE,
+
+	/**
+	 * @DRM_ASAHI_GEM_MMAP_OFFSET: Get offset to pass to mmap() to map a
+	 * given GEM handle.
+	 */
+	DRM_ASAHI_GEM_MMAP_OFFSET,
+
+	/** @DRM_ASAHI_GEM_BIND_OBJECT: Bind memory as a special object */
+	DRM_ASAHI_GEM_BIND_OBJECT,
+
+	/** @DRM_ASAHI_QUEUE_CREATE: Create a scheduling queue. */
+	DRM_ASAHI_QUEUE_CREATE,
+
+	/** @DRM_ASAHI_QUEUE_DESTROY: Destroy a scheduling queue. */
+	DRM_ASAHI_QUEUE_DESTROY,
+
+	/** @DRM_ASAHI_SUBMIT: Submit commands to a queue. */
+	DRM_ASAHI_SUBMIT,
+};
+
+#define DRM_ASAHI_MAX_CLUSTERS	64
+
+/**
+ * struct drm_asahi_params_global - Global parameters.
+ *
+ * This struct may be queried by drm_asahi_get_params.
+ */
+struct drm_asahi_params_global {
+	/** @features: Feature bits from drm_asahi_feature */
+	__u64 features;
+
+	/** @gpu_generation: GPU generation, e.g. 13 for G13G */
+	__u32 gpu_generation;
+
+	/** @gpu_variant: GPU variant as a character, e.g. 'C' for G13C */
+	__u32 gpu_variant;
+
+	/**
+	 * @gpu_revision: GPU revision in BCD, e.g. 0x00 for 'A0' or
+	 * 0x21 for 'C1'
+	 */
+	__u32 gpu_revision;
+
+	/** @chip_id: Chip ID in BCD, e.g. 0x8103 for T8103 */
+	__u32 chip_id;
+
+	/** @num_dies: Number of dies in the SoC */
+	__u32 num_dies;
+
+	/** @num_clusters_total: Number of GPU clusters (across all dies) */
+	__u32 num_clusters_total;
+
+	/**
+	 * @num_cores_per_cluster: Number of logical cores per cluster
+	 * (including inactive/nonexistent)
+	 */
+	__u32 num_cores_per_cluster;
+
+	/** @max_frequency_khz: Maximum GPU core clock frequency */
+	__u32 max_frequency_khz;
+
+	/** @core_masks: Bitmask of present/enabled cores per cluster */
+	__u64 core_masks[DRM_ASAHI_MAX_CLUSTERS];
+
+	/**
+	 * @vm_start: VM range start VMA. Together with @vm_end, this defines
+	 * the window of valid GPU VAs. Userspace is expected to subdivide VAs
+	 * out of this window.
+	 *
+	 * This window contains all virtual addresses that userspace needs to
+	 * know about. There may be kernel-internal GPU VAs outside this range,
+	 * but that detail is not relevant here.
+	 */
+	__u64 vm_start;
+
+	/** @vm_end: VM range end VMA */
+	__u64 vm_end;
+
+	/**
+	 * @vm_kernel_min_size: Minimum kernel VMA window size.
+	 *
+	 * When creating a VM, userspace is required to carve out a section of
+	 * virtual addresses (within the range given by @vm_start and
+	 * @vm_end). The kernel will allocate various internal structures
+	 * within the specified VA range.
+	 *
+	 * Allowing userspace to choose the VA range for the kernel, rather than
+	 * the kernel reserving VAs and requiring userspace to cope, can assist
+	 * in implementing SVM.
+	 */
+	__u64 vm_kernel_min_size;
+
+	/**
+	 * @max_commands_per_submission: Maximum number of supported commands
+	 * per submission. This mirrors firmware limits. Userspace must split up
+	 * larger command buffers, which may require inserting additional
+	 * synchronization.
+	 */
+	__u32 max_commands_per_submission;
+
+	/**
+	 * @max_attachments: Maximum number of drm_asahi_attachment's per
+	 * command
+	 */
+	__u32 max_attachments;
+
+	/**
+	 * @command_timestamp_frequency_hz: Timebase frequency for timestamps
+	 * written during command execution, specified via drm_asahi_timestamp
+	 * structures. As this rate is controlled by the firmware, it is a
+	 * queryable parameter.
+	 *
+	 * Userspace must divide by this frequency to convert timestamps to
+	 * seconds, rather than hardcoding a particular firmware's rate.
+	 */
+	__u64 command_timestamp_frequency_hz;
+};
+
+/**
+ * enum drm_asahi_feature - Feature bits
+ *
+ * This covers only features that userspace cannot infer from the architecture
+ * version. Most features don't need to be here.
+ */
+enum drm_asahi_feature {
+	/**
+	 * @DRM_ASAHI_FEATURE_SOFT_FAULTS: GPU has "soft fault" enabled. Shader
+	 * loads of unmapped memory will return zero. Shader stores to unmapped
+	 * memory will be silently discarded. Note that only shader load/store
+	 * is affected. Other hardware units are not affected, notably including
+	 * texture sampling.
+	 *
+	 * Soft fault is set when initializing the GPU and cannot be runtime
+	 * toggled. Therefore, it is exposed as a feature bit and not a
+	 * userspace-settable flag on the VM. When soft fault is enabled,
+	 * userspace can speculate memory accesses more aggressively.
+	 */
+	DRM_ASAHI_FEATURE_SOFT_FAULTS = (1UL) << 0,
+};
+
+/**
+ * struct drm_asahi_get_params - Arguments passed to DRM_IOCTL_ASAHI_GET_PARAMS
+ */
+struct drm_asahi_get_params {
+	/** @param_group: Parameter group to fetch (MBZ) */
+	__u32 param_group;
+
+	/** @pad: MBZ */
+	__u32 pad;
+
+	/** @pointer: User pointer to write parameter struct */
+	__u64 pointer;
+
+	/**
+	 * @size: Size of the user buffer. In case of older userspace, this may
+	 * be less than sizeof(struct drm_asahi_params_global). The kernel will
+	 * not write past the length specified here, allowing extensibility.
+	 */
+	__u64 size;
+};
+
+/**
+ * struct drm_asahi_vm_create - Arguments passed to DRM_IOCTL_ASAHI_VM_CREATE
+ */
+struct drm_asahi_vm_create {
+	/**
+	 * @kernel_start: Start of the kernel-reserved address range. See
+	 * drm_asahi_params_global::vm_kernel_min_size.
+	 *
+	 * Both @kernel_start and @kernel_end must be within the range of
+	 * valid VAs given by drm_asahi_params_global::vm_start and
+	 * drm_asahi_params_global::vm_end. The size of the kernel range
+	 * (@kernel_end - @kernel_start) must be at least
+	 * drm_asahi_params_global::vm_kernel_min_size.
+	 *
+	 * Userspace must not bind any memory on this VM into this reserved
+	 * range, it is for kernel use only.
+	 */
+	__u64 kernel_start;
+
+	/**
+	 * @kernel_end: End of the kernel-reserved address range. See
+	 * @kernel_start.
+	 */
+	__u64 kernel_end;
+
+	/** @vm_id: Returned VM ID */
+	__u32 vm_id;
+
+	/** @pad: MBZ */
+	__u32 pad;
+};
+
+/**
+ * struct drm_asahi_vm_destroy - Arguments passed to DRM_IOCTL_ASAHI_VM_DESTROY
+ */
+struct drm_asahi_vm_destroy {
+	/** @vm_id: VM ID to be destroyed */
+	__u32 vm_id;
+
+	/** @pad: MBZ */
+	__u32 pad;
+};
+
+/**
+ * enum drm_asahi_gem_flags - Flags for GEM creation
+ */
+enum drm_asahi_gem_flags {
+	/**
+	 * @DRM_ASAHI_GEM_WRITEBACK: BO should be CPU-mapped as writeback.
+	 *
+	 * Map as writeback instead of write-combine. This optimizes for CPU
+	 * reads.
+	 */
+	DRM_ASAHI_GEM_WRITEBACK = (1L << 0),
+
+	/**
+	 * @DRM_ASAHI_GEM_VM_PRIVATE: BO is private to this GPU VM (no exports).
+	 */
+	DRM_ASAHI_GEM_VM_PRIVATE = (1L << 1),
+};
+
+/**
+ * struct drm_asahi_gem_create - Arguments passed to DRM_IOCTL_ASAHI_GEM_CREATE
+ */
+struct drm_asahi_gem_create {
+	/** @size: Size of the BO */
+	__u64 size;
+
+	/** @flags: Combination of drm_asahi_gem_flags flags. */
+	__u32 flags;
+
+	/**
+	 * @vm_id: VM ID to assign to the BO, if DRM_ASAHI_GEM_VM_PRIVATE is set
+	 */
+	__u32 vm_id;
+
+	/** @handle: Returned GEM handle for the BO */
+	__u32 handle;
+
+	/** @pad: MBZ */
+	__u32 pad;
+};
+
+/**
+ * struct drm_asahi_gem_mmap_offset - Arguments passed to
+ * DRM_IOCTL_ASAHI_GEM_MMAP_OFFSET
+ */
+struct drm_asahi_gem_mmap_offset {
+	/** @handle: Handle for the object being mapped. */
+	__u32 handle;
+
+	/** @flags: Must be zero */
+	__u32 flags;
+
+	/** @offset: The fake offset to use for subsequent mmap call */
+	__u64 offset;
+};
+
+/**
+ * enum drm_asahi_bind_flags - Flags for GEM binding
+ */
+enum drm_asahi_bind_flags {
+	/**
+	 * @DRM_ASAHI_BIND_UNBIND: Instead of binding a GEM object to the range,
+	 * simply unbind the GPU VMA range.
+	 */
+	DRM_ASAHI_BIND_UNBIND = (1L << 0),
+
+	/** @DRM_ASAHI_BIND_READ: Map BO with GPU read permission */
+	DRM_ASAHI_BIND_READ = (1L << 1),
+
+	/** @DRM_ASAHI_BIND_WRITE: Map BO with GPU write permission */
+	DRM_ASAHI_BIND_WRITE = (1L << 2),
+
+	/**
+	 * @DRM_ASAHI_BIND_SINGLE_PAGE: Map a single page of the BO repeatedly
+	 * across the VA range.
+	 *
+	 * This is useful to fill a VA range with scratch pages or zero pages.
+	 * It is intended as a mechanism to accelerate sparse.
+	 */
+	DRM_ASAHI_BIND_SINGLE_PAGE = (1L << 3),
+};
+
+/**
+ * struct drm_asahi_gem_bind_op - Description of a single GEM bind operation.
+ */
+struct drm_asahi_gem_bind_op {
+	/** @flags: Combination of drm_asahi_bind_flags flags. */
+	__u32 flags;
+
+	/** @handle: GEM object to bind (except for UNBIND) */
+	__u32 handle;
+
+	/**
+	 * @offset: Offset into the object (except for UNBIND).
+	 *
+	 * For a regular bind, this is the beginning of the region of the GEM
+	 * object to bind.
+	 *
+	 * For a single-page bind, this is the offset to the single page that
+	 * will be repeatedly bound.
+	 *
+	 * Must be page-size aligned.
+	 */
+	__u64 offset;
+
+	/**
+	 * @range: Number of bytes to bind/unbind to @addr.
+	 *
+	 * Must be page-size aligned.
+	 */
+	__u64 range;
+
+	/**
+	 * @addr: Address to bind to.
+	 *
+	 * Must be page-size aligned.
+	 */
+	__u64 addr;
+};
+
+/**
+ * struct drm_asahi_vm_bind - Arguments passed to
+ * DRM_IOCTL_ASAHI_VM_BIND
+ */
+struct drm_asahi_vm_bind {
+	/** @vm_id: The ID of the VM to bind to */
+	__u32 vm_id;
+
+	/** @num_binds: number of binds in this IOCTL. */
+	__u32 num_binds;
+
+	/**
+	 * @stride: Stride in bytes between consecutive binds. This allows
+	 * extensibility of drm_asahi_gem_bind_op.
+	 */
+	__u32 stride;
+
+	/** @pad: MBZ */
+	__u32 pad;
+
+	/**
+	 * @userptr: User pointer to an array of @num_binds structures of type
+	 * @drm_asahi_gem_bind_op and size @stride bytes.
+	 */
+	__u64 userptr;
+};
+
+/**
+ * enum drm_asahi_bind_object_op - Special object bind operation
+ */
+enum drm_asahi_bind_object_op {
+	/** @DRM_ASAHI_BIND_OBJECT_OP_BIND: Bind a BO as a special GPU object */
+	DRM_ASAHI_BIND_OBJECT_OP_BIND = 0,
+
+	/** @DRM_ASAHI_BIND_OBJECT_OP_UNBIND: Unbind a special GPU object */
+	DRM_ASAHI_BIND_OBJECT_OP_UNBIND = 1,
+};
+
+/**
+ * enum drm_asahi_bind_object_flags - Special object bind flags
+ */
+enum drm_asahi_bind_object_flags {
+	/**
+	 * @DRM_ASAHI_BIND_OBJECT_USAGE_TIMESTAMPS: Map a BO as a timestamp
+	 * buffer.
+	 */
+	DRM_ASAHI_BIND_OBJECT_USAGE_TIMESTAMPS = (1L << 0),
+};
+
+/**
+ * struct drm_asahi_gem_bind_object - Arguments passed to
+ * DRM_IOCTL_ASAHI_GEM_BIND_OBJECT
+ */
+struct drm_asahi_gem_bind_object {
+	/** @op: Bind operation (enum drm_asahi_bind_object_op) */
+	__u32 op;
+
+	/** @flags: Combination of drm_asahi_bind_object_flags flags. */
+	__u32 flags;
+
+	/** @handle: GEM object to bind/unbind (BIND) */
+	__u32 handle;
+
+	/** @vm_id: The ID of the VM to operate on (MBZ currently) */
+	__u32 vm_id;
+
+	/** @offset: Offset into the object (BIND only) */
+	__u64 offset;
+
+	/** @range: Number of bytes to bind/unbind (BIND only) */
+	__u64 range;
+
+	/** @object_handle: Object handle (out for BIND, in for UNBIND) */
+	__u32 object_handle;
+
+	/** @pad: MBZ */
+	__u32 pad;
+};
+
+/**
+ * enum drm_asahi_cmd_type - Command type
+ */
+enum drm_asahi_cmd_type {
+	/**
+	 * @DRM_ASAHI_CMD_RENDER: Render command, executing on the render
+	 * subqueue. Combined vertex and fragment operation.
+	 *
+	 * Followed by a @drm_asahi_cmd_render payload.
+	 */
+	DRM_ASAHI_CMD_RENDER = 0,
+
+	/**
+	 * @DRM_ASAHI_CMD_COMPUTE: Compute command on the compute subqueue.
+	 *
+	 * Followed by a @drm_asahi_cmd_compute payload.
+	 */
+	DRM_ASAHI_CMD_COMPUTE = 1,
+
+	/**
+	 * @DRM_ASAHI_SET_VERTEX_ATTACHMENTS: Software command to set
+	 * attachments for subsequent vertex shaders in the same submit.
+	 *
+	 * Followed by (possibly multiple) @drm_asahi_attachment payloads.
+	 */
+	DRM_ASAHI_SET_VERTEX_ATTACHMENTS = 2,
+
+	/**
+	 * @DRM_ASAHI_SET_FRAGMENT_ATTACHMENTS: Software command to set
+	 * attachments for subsequent fragment shaders in the same submit.
+	 *
+	 * Followed by (possibly multiple) @drm_asahi_attachment payloads.
+	 */
+	DRM_ASAHI_SET_FRAGMENT_ATTACHMENTS = 3,
+
+	/**
+	 * @DRM_ASAHI_SET_COMPUTE_ATTACHMENTS: Software command to set
+	 * attachments for subsequent compute shaders in the same submit.
+	 *
+	 * Followed by (possibly multiple) @drm_asahi_attachment payloads.
+	 */
+	DRM_ASAHI_SET_COMPUTE_ATTACHMENTS = 4,
+};
+
+/**
+ * enum drm_asahi_priority - Scheduling queue priority.
+ *
+ * These priorities are forwarded to the firmware to influence firmware
+ * scheduling. The exact policy is ultimately decided by firmware, but
+ * these enums allow userspace to communicate the intentions.
+ */
+enum drm_asahi_priority {
+	/** @DRM_ASAHI_PRIORITY_LOW: Low priority queue. */
+	DRM_ASAHI_PRIORITY_LOW = 0,
+
+	/** @DRM_ASAHI_PRIORITY_MEDIUM: Medium priority queue. */
+	DRM_ASAHI_PRIORITY_MEDIUM = 1,
+
+	/**
+	 * @DRM_ASAHI_PRIORITY_HIGH: High priority queue.
+	 *
+	 * Reserved for future extension.
+	 */
+	DRM_ASAHI_PRIORITY_HIGH = 2,
+
+	/**
+	 * @DRM_ASAHI_PRIORITY_REALTIME: Real-time priority queue.
+	 *
+	 * Reserved for future extension.
+	 */
+	DRM_ASAHI_PRIORITY_REALTIME = 3,
+};
+
+/**
+ * struct drm_asahi_queue_create - Arguments passed to
+ * DRM_IOCTL_ASAHI_QUEUE_CREATE
+ */
+struct drm_asahi_queue_create {
+	/** @flags: MBZ */
+	__u32 flags;
+
+	/** @vm_id: The ID of the VM this queue is bound to */
+	__u32 vm_id;
+
+	/** @priority: One of drm_asahi_priority */
+	__u32 priority;
+
+	/** @queue_id: The returned queue ID */
+	__u32 queue_id;
+
+	/**
+	 * @usc_exec_base: GPU base address for all USC binaries (shaders) on
+	 * this queue. USC addresses are 32-bit relative to this 64-bit base.
+	 *
+	 * This sets the following registers on all queue commands:
+	 *
+	 *	USC_EXEC_BASE_TA  (vertex)
+	 *	USC_EXEC_BASE_ISP (fragment)
+	 *	USC_EXEC_BASE_CP  (compute)
+	 *
+	 * While the hardware lets us configure these independently per command,
+	 * we do not have a use case for this. Instead, we expect userspace to
+	 * fix a 4GiB VA carveout for USC memory and pass its base address here.
+	 */
+	__u64 usc_exec_base;
+};
+
+/**
+ * struct drm_asahi_queue_destroy - Arguments passed to
+ * DRM_IOCTL_ASAHI_QUEUE_DESTROY
+ */
+struct drm_asahi_queue_destroy {
+	/** @queue_id: The queue ID to be destroyed */
+	__u32 queue_id;
+
+	/** @pad: MBZ */
+	__u32 pad;
+};
+
+/**
+ * enum drm_asahi_sync_type - Sync item type
+ */
+enum drm_asahi_sync_type {
+	/** @DRM_ASAHI_SYNC_SYNCOBJ: Binary sync object */
+	DRM_ASAHI_SYNC_SYNCOBJ = 0,
+
+	/** @DRM_ASAHI_SYNC_TIMELINE_SYNCOBJ: Timeline sync object */
+	DRM_ASAHI_SYNC_TIMELINE_SYNCOBJ = 1,
+};
+
+/**
+ * struct drm_asahi_sync - Sync item
+ */
+struct drm_asahi_sync {
+	/** @sync_type: One of drm_asahi_sync_type */
+	__u32 sync_type;
+
+	/** @handle: The sync object handle */
+	__u32 handle;
+
+	/** @timeline_value: Timeline value for timeline sync objects */
+	__u64 timeline_value;
+};
+
+/**
+ * define DRM_ASAHI_BARRIER_NONE - Command index for no barrier
+ *
+ * This special value may be passed in to drm_asahi_command::vdm_barrier or
+ * drm_asahi_command::cdm_barrier to indicate that the respective subqueue
+ * should not wait on any previous work.
+ */
+#define DRM_ASAHI_BARRIER_NONE (0xFFFFu)
+
+/**
+ * struct drm_asahi_cmd_header - Top level command structure
+ *
+ * This struct is core to the command buffer definition and therefore is not
+ * extensible.
+ */
+struct drm_asahi_cmd_header {
+	/** @cmd_type: One of drm_asahi_cmd_type */
+	__u16 cmd_type;
+
+	/**
+	 * @size: Size of this command, not including this header.
+	 *
+	 * For hardware commands, this enables extensibility of commands without
+	 * requiring extra command types. Passing a command that is shorter
+	 * than expected is explicitly allowed for backwards-compatibility.
+	 * Truncated fields will be zeroed.
+	 *
+	 * For the synthetic attachment setting commands, this implicitly
+	 * encodes the number of attachments. These commands take multiple
+	 * fixed-size @drm_asahi_attachment structures as their payload, so size
+	 * equals number of attachments * sizeof(struct drm_asahi_attachment).
+	 */
+	__u16 size;
+
+	/**
+	 * @vdm_barrier: VDM (render) command index to wait on.
+	 *
+	 * Barriers are indices relative to the beginning of a given submit. A
+	 * barrier of 0 waits on commands submitted to the respective subqueue
+	 * in previous submit ioctls. A barrier of N waits on N previous
+	 * commands on the subqueue within the current submit ioctl. As a
+	 * special case, passing @DRM_ASAHI_BARRIER_NONE avoids waiting on any
+	 * commands in the subqueue.
+	 *
+	 * Examples:
+	 *
+	 *   0: This waits on all previous work.
+	 *
+	 *   NONE: This does not wait for anything on this subqueue.
+	 *
+	 *   1: This waits on the first render command in the submit.
+	 *   This is valid only if there are multiple render commands in the
+	 *   same submit.
+	 *
+	 * Barriers are valid only for hardware commands. Synthetic software
+	 * commands to set attachments must pass NONE here.
+	 */
+	__u16 vdm_barrier;
+
+	/**
+	 * @cdm_barrier: CDM (compute) command index to wait on.
+	 *
+	 * See @vdm_barrier, and replace VDM/render with CDM/compute.
+	 */
+	__u16 cdm_barrier;
+};
+
+/**
+ * struct drm_asahi_submit - Arguments passed to DRM_IOCTL_ASAHI_SUBMIT
+ */
+struct drm_asahi_submit {
+	/**
+	 * @syncs: An optional pointer to an array of drm_asahi_sync. The first
+	 * @in_sync_count elements are in-syncs, then the remaining
+	 * @out_sync_count elements are out-syncs. Using a single array with
+	 * explicit partitioning simplifies handling.
+	 */
+	__u64 syncs;
+
+	/**
+	 * @cmdbuf: Pointer to the command buffer to submit.
+	 *
+	 * This is a flat command buffer. By design, it contains no CPU
+	 * pointers, which makes it suitable for a virtgpu wire protocol without
+	 * requiring any serializing/deserializing step.
+	 *
+	 * It consists of a series of commands. Each command begins with a
+	 * fixed-size @drm_asahi_cmd_header header and is followed by a
+	 * variable-length payload according to the type and size in the header.
+	 *
+	 * The combined count of "real" hardware commands must be nonzero and at
+	 * most drm_asahi_params_global::max_commands_per_submission.
+	 */
+	__u64 cmdbuf;
+
+	/** @flags: Flags for command submission (MBZ) */
+	__u32 flags;
+
+	/** @queue_id: The queue ID to be submitted to */
+	__u32 queue_id;
+
+	/**
+	 * @in_sync_count: Number of sync objects to wait on before starting
+	 * this job.
+	 */
+	__u32 in_sync_count;
+
+	/**
+	 * @out_sync_count: Number of sync objects to signal upon completion of
+	 * this job.
+	 */
+	__u32 out_sync_count;
+
+	/** @cmdbuf_size: Command buffer size in bytes */
+	__u32 cmdbuf_size;
+
+	/** @pad: MBZ */
+	__u32 pad;
+};
+
+/**
+ * struct drm_asahi_attachment - Describe an "attachment".
+ *
+ * Attachments are any memory written by shaders, notably including render
+ * target attachments written by the end-of-tile program. This is purely a hint
+ * about the accessed memory regions. It is optional to specify, which is
+ * fortunate as it cannot be specified precisely with bindless access anyway.
+ * But where possible, it's probably a good idea for userspace to include these
+ * hints, forwarded to the firmware.
+ *
+ * This struct is implicitly sized and therefore is not extensible.
+ */
+struct drm_asahi_attachment {
+	/** @pointer: Base address of the attachment */
+	__u64 pointer;
+
+	/** @size: Size of the attachment in bytes */
+	__u64 size;
+
+	/** @pad: MBZ */
+	__u32 pad;
+
+	/** @flags: MBZ */
+	__u32 flags;
+};
+
+enum drm_asahi_render_flags {
+	/**
+	 * @DRM_ASAHI_RENDER_VERTEX_SCRATCH: A vertex stage shader uses scratch
+	 * memory.
+	 */
+	DRM_ASAHI_RENDER_VERTEX_SCRATCH = (1U << 0),
+
+	/**
+	 * @DRM_ASAHI_RENDER_PROCESS_EMPTY_TILES: Process even empty tiles.
+	 * This must be set when clearing render targets.
+	 */
+	DRM_ASAHI_RENDER_PROCESS_EMPTY_TILES = (1U << 1),
+
+	/**
+	 * @DRM_ASAHI_RENDER_NO_VERTEX_CLUSTERING: Run vertex stage on a single
+	 * cluster (on multi-cluster GPUs)
+	 *
+	 * This harms performance but can workaround certain sync/coherency
+	 * bugs, and therefore is useful for debugging.
+	 */
+	DRM_ASAHI_RENDER_NO_VERTEX_CLUSTERING = (1U << 2),
+
+	/**
+	 * @DRM_ASAHI_RENDER_DBIAS_IS_INT: Use integer depth bias formula.
+	 *
+	 * Graphics specifications contain two alternate formulas for depth
+	 * bias, a float formula used with floating-point depth buffers and an
+	 * integer formula using with unorm depth buffers. This flag specifies
+	 * that the integer formula should be used. If omitted, the float
+	 * formula is used instead.
+	 *
+	 * This corresponds to bit 18 of the relevant hardware control register,
+	 * so we match that here for efficiency.
+	 */
+	DRM_ASAHI_RENDER_DBIAS_IS_INT = (1U << 18),
+};
+
+/**
+ * struct drm_asahi_zls_buffer - Describe a depth or stencil buffer.
+ *
+ * These fields correspond to hardware registers in the ZLS (Z Load/Store) unit.
+ * There are three hardware registers for each field respectively for loads,
+ * stores, and partial renders. In practice, it makes sense to set all to the
+ * same values, except in exceptional cases not yet implemented in userspace, so
+ * we do not duplicate here for simplicity/efficiency.
+ *
+ * This struct is embedded in other structs and therefore is not extensible.
+ */
+struct drm_asahi_zls_buffer {
+	/** @base: Base address of the buffer */
+	__u64 base;
+
+	/**
+	 * @comp_base: If the load buffer is compressed, address of the
+	 * compression metadata section.
+	 */
+	__u64 comp_base;
+
+	/**
+	 * @stride: If layered rendering is enabled, the number of bytes
+	 * between each layer of the buffer.
+	 */
+	__u32 stride;
+
+	/**
+	 * @comp_stride: If layered rendering is enabled, the number of bytes
+	 * between each layer of the compression metadata.
+	 */
+	__u32 comp_stride;
+};
+
+/**
+ * struct drm_asahi_timestamp - Describe a timestamp write.
+ *
+ * The firmware can optionally write the GPU timestamp at render pass
+ * granularities, but it needs to be mapped specially via
+ * DRM_IOCTL_ASAHI_GEM_BIND_OBJECT. This structure therefore describes where to
+ * write as a handle-offset pair, rather than a GPU address like normal.
+ *
+ * This struct is embedded in other structs and therefore is not extensible.
+ */
+struct drm_asahi_timestamp {
+	/**
+	 * @handle: Handle of the timestamp buffer, or 0 to skip this
+	 * timestamp. If nonzero, this must equal the value returned in
+	 * drm_asahi_gem_bind_object::object_handle.
+	 */
+	__u32 handle;
+
+	/** @offset: Offset to write into the timestamp buffer */
+	__u32 offset;
+};
+
+/**
+ * struct drm_asahi_timestamps - Describe timestamp writes.
+ *
+ * Each operation that can be timestamped, can be timestamped at the start and
+ * end. Therefore, drm_asahi_timestamp structs always come in pairs, bundled
+ * together into drm_asahi_timestamps.
+ *
+ * This struct is embedded in other structs and therefore is not extensible.
+ */
+struct drm_asahi_timestamps {
+	/** @start: Timestamp recorded at the start of the operation */
+	struct drm_asahi_timestamp start;
+
+	/** @end: Timestamp recorded at the end of the operation */
+	struct drm_asahi_timestamp end;
+};
+
+/**
+ * struct drm_asahi_helper_program - Describe helper program configuration.
+ *
+ * The helper program is a compute-like kernel required for various hardware
+ * functionality. Its most important role is dynamically allocating
+ * scratch/stack memory for individual subgroups, by partitioning a static
+ * allocation shared for the whole device. It is supplied by userspace via
+ * drm_asahi_helper_program and internally dispatched by the hardware as needed.
+ *
+ * This struct is embedded in other structs and therefore is not extensible.
+ */
+struct drm_asahi_helper_program {
+	/**
+	 * @binary: USC address to the helper program binary. This is a tagged
+	 * pointer with configuration in the bottom bits.
+	 */
+	__u32 binary;
+
+	/** @cfg: Additional configuration bits for the helper program. */
+	__u32 cfg;
+
+	/**
+	 * @data: Data passed to the helper program. This value is not
+	 * interpreted by the kernel, firmware, or hardware in any way. It is
+	 * simply a sideband for userspace, set with the submit ioctl and read
+	 * via special registers inside the helper program.
+	 *
+	 * In practice, userspace will pass a 64-bit GPU VA here pointing to the
+	 * actual arguments, which presumably don't fit in 64-bits.
+	 */
+	__u64 data;
+};
+
+/**
+ * struct drm_asahi_bg_eot - Describe a background or end-of-tile program.
+ *
+ * The background and end-of-tile programs are dispatched by the hardware at the
+ * beginning and end of rendering. As the hardware "tilebuffer" is simply local
+ * memory, these programs are necessary to implement API-level render targets.
+ * The fragment-like background program is responsible for loading either the
+ * clear colour or the existing render target contents, while the compute-like
+ * end-of-tile program stores the tilebuffer contents to memory.
+ *
+ * This struct is embedded in other structs and therefore is not extensible.
+ */
+struct drm_asahi_bg_eot {
+	/**
+	 * @usc: USC address of the hardware USC words binding resources
+	 * (including images and uniforms) and the program itself. Note this is
+	 * an additional layer of indirection compared to the helper program,
+	 * avoiding the need for a sideband for data. This is a tagged pointer
+	 * with additional configuration in the bottom bits.
+	 */
+	__u32 usc;
+
+	/**
+	 * @rsrc_spec: Resource specifier for the program. This is a packed
+	 * hardware data structure describing the required number of registers,
+	 * uniforms, bound textures, and bound samplers.
+	 */
+	__u32 rsrc_spec;
+};
+
+/**
+ * struct drm_asahi_cmd_render - Command to submit 3D
+ *
+ * This command submits a single render pass. The hardware control stream may
+ * include many draws and subpasses, but within the command, the framebuffer
+ * dimensions and attachments are fixed.
+ *
+ * The hardware requires the firmware to set a large number of Control Registers
+ * setting up state at render pass granularity before each command rendering 3D.
+ * The firmware bundles this state into data structures. Unfortunately, we
+ * cannot expose either any of that directly to userspace, because the
+ * kernel-firmware ABI is not stable. Although we can guarantee the firmware
+ * updates in tandem with the kernel, we cannot break old userspace when
+ * upgrading the firmware and kernel. Therefore, we need to abstract well the
+ * data structures to avoid tying our hands with future firmwares.
+ *
+ * The bulk of drm_asahi_cmd_render therefore consists of values of hardware
+ * control registers, marshalled via the firmware interface.
+ *
+ * The framebuffer/tilebuffer dimensions are also specified here. In addition to
+ * being passed to the firmware/hardware, the kernel requires these dimensions
+ * to calculate various essential tiling-related data structures. It is
+ * unfortunate that our submits are heavier than on vendors with saner
+ * hardware-software interfaces. The upshot is all of this information is
+ * readily available to userspace with all current APIs.
+ *
+ * It looks odd - but it's not overly burdensome and it ensures we can remain
+ * compatible with old userspace.
+ */
+struct drm_asahi_cmd_render {
+	/** @flags: Combination of drm_asahi_render_flags flags. */
+	__u32 flags;
+
+	/**
+	 * @isp_zls_pixels: ISP_ZLS_PIXELS register value. This contains the
+	 * depth/stencil width/height, which may differ from the framebuffer
+	 * width/height.
+	 */
+	__u32 isp_zls_pixels;
+
+	/**
+	 * @vdm_ctrl_stream_base: VDM_CTRL_STREAM_BASE register value. GPU
+	 * address to the beginning of the VDM control stream.
+	 */
+	__u64 vdm_ctrl_stream_base;
+
+	/** @vertex_helper: Helper program used for the vertex shader */
+	struct drm_asahi_helper_program vertex_helper;
+
+	/** @fragment_helper: Helper program used for the fragment shader */
+	struct drm_asahi_helper_program fragment_helper;
+
+	/**
+	 * @isp_scissor_base: ISP_SCISSOR_BASE register value. GPU address of an
+	 * array of scissor descriptors indexed in the render pass.
+	 */
+	__u64 isp_scissor_base;
+
+	/**
+	 * @isp_dbias_base: ISP_DBIAS_BASE register value. GPU address of an
+	 * array of depth bias values indexed in the render pass.
+	 */
+	__u64 isp_dbias_base;
+
+	/**
+	 * @isp_oclqry_base: ISP_OCLQRY_BASE register value. GPU address of an
+	 * array of occlusion query results written by the render pass.
+	 */
+	__u64 isp_oclqry_base;
+
+	/** @depth: Depth buffer */
+	struct drm_asahi_zls_buffer depth;
+
+	/** @stencil: Stencil buffer */
+	struct drm_asahi_zls_buffer stencil;
+
+	/** @zls_ctrl: ZLS_CTRL register value */
+	__u64 zls_ctrl;
+
+	/** @ppp_multisamplectl: PPP_MULTISAMPLECTL register value */
+	__u64 ppp_multisamplectl;
+
+	/**
+	 * @sampler_heap: Base address of the sampler heap. This heap is used
+	 * for both vertex shaders and fragment shaders. The registers are
+	 * per-stage, but there is no known use case for separate heaps.
+	 */
+	__u64 sampler_heap;
+
+	/** @ppp_ctrl: PPP_CTRL register value */
+	__u32 ppp_ctrl;
+
+	/** @width_px: Framebuffer width in pixels */
+	__u16 width_px;
+
+	/** @height_px: Framebuffer height in pixels */
+	__u16 height_px;
+
+	/** @layers: Number of layers in the framebuffer */
+	__u16 layers;
+
+	/** @sampler_count: Number of samplers in the sampler heap. */
+	__u16 sampler_count;
+
+	/** @utile_width_px: Width of a logical tilebuffer tile in pixels */
+	__u8 utile_width_px;
+
+	/** @utile_height_px: Height of a logical tilebuffer tile in pixels */
+	__u8 utile_height_px;
+
+	/** @samples: # of samples in the framebuffer. Must be 1, 2, or 4. */
+	__u8 samples;
+
+	/** @sample_size_B: # of bytes in the tilebuffer required per sample. */
+	__u8 sample_size_B;
+
+	/**
+	 * @isp_merge_upper_x: 32-bit float used in the hardware triangle
+	 * merging. Calculate as: tan(60 deg) * width.
+	 *
+	 * Making these values UAPI avoids requiring floating-point calculations
+	 * in the kernel in the hot path.
+	 */
+	__u32 isp_merge_upper_x;
+
+	/**
+	 * @isp_merge_upper_y: 32-bit float. Calculate as: tan(60 deg) * height.
+	 * See @isp_merge_upper_x.
+	 */
+	__u32 isp_merge_upper_y;
+
+	/** @bg: Background program run for each tile at the start */
+	struct drm_asahi_bg_eot bg;
+
+	/** @eot: End-of-tile program ran for each tile at the end */
+	struct drm_asahi_bg_eot eot;
+
+	/**
+	 * @partial_bg: Background program ran at the start of each tile when
+	 * resuming the render pass during a partial render.
+	 */
+	struct drm_asahi_bg_eot partial_bg;
+
+	/**
+	 * @partial_eot: End-of-tile program ran at the end of each tile when
+	 * pausing the render pass during a partial render.
+	 */
+	struct drm_asahi_bg_eot partial_eot;
+
+	/**
+	 * @isp_bgobjdepth: ISP_BGOBJDEPTH register value. This is the depth
+	 * buffer clear value, encoded in the depth buffer's format: either a
+	 * 32-bit float or a 16-bit unorm (with upper bits zeroed).
+	 */
+	__u32 isp_bgobjdepth;
+
+	/**
+	 * @isp_bgobjvals: ISP_BGOBJVALS register value. The bottom 8-bits
+	 * contain the stencil buffer clear value.
+	 */
+	__u32 isp_bgobjvals;
+
+	/** @ts_vtx: Timestamps for the vertex portion of the render */
+	struct drm_asahi_timestamps ts_vtx;
+
+	/** @ts_frag: Timestamps for the fragment portion of the render */
+	struct drm_asahi_timestamps ts_frag;
+};
+
+/**
+ * struct drm_asahi_cmd_compute - Command to submit compute
+ *
+ * This command submits a control stream consisting of compute dispatches. There
+ * is essentially no limit on how many compute dispatches may be included in a
+ * single compute command, although timestamps are at command granularity.
+ */
+struct drm_asahi_cmd_compute {
+	/** @flags: MBZ */
+	__u32 flags;
+
+	/** @sampler_count: Number of samplers in the sampler heap. */
+	__u32 sampler_count;
+
+	/**
+	 * @cdm_ctrl_stream_base: CDM_CTRL_STREAM_BASE register value. GPU
+	 * address to the beginning of the CDM control stream.
+	 */
+	__u64 cdm_ctrl_stream_base;
+
+	/**
+	 * @cdm_ctrl_stream_end: GPU base address to the end of the hardware
+	 * control stream. Note this only considers the first contiguous segment
+	 * of the control stream, as the stream might jump elsewhere.
+	 */
+	__u64 cdm_ctrl_stream_end;
+
+	/** @sampler_heap: Base address of the sampler heap. */
+	__u64 sampler_heap;
+
+	/** @helper: Helper program used for this compute command */
+	struct drm_asahi_helper_program helper;
+
+	/** @ts: Timestamps for the compute command */
+	struct drm_asahi_timestamps ts;
+};
+
+/**
+ * struct drm_asahi_get_time - Arguments passed to DRM_IOCTL_ASAHI_GET_TIME
+ */
+struct drm_asahi_get_time {
+	/** @flags: MBZ. */
+	__u64 flags;
+
+	/** @gpu_timestamp: On return, the GPU timestamp in nanoseconds. */
+	__u64 gpu_timestamp;
+};
+
+/**
+ * DRM_IOCTL_ASAHI() - Build an Asahi IOCTL number
+ * @__access: Access type. Must be R, W or RW.
+ * @__id: One of the DRM_ASAHI_xxx id.
+ * @__type: Suffix of the type being passed to the IOCTL.
+ *
+ * Don't use this macro directly, use the DRM_IOCTL_ASAHI_xxx
+ * values instead.
+ *
+ * Return: An IOCTL number to be passed to ioctl() from userspace.
+ */
+#define DRM_IOCTL_ASAHI(__access, __id, __type) \
+	DRM_IO ## __access(DRM_COMMAND_BASE + DRM_ASAHI_ ## __id, \
+			   struct drm_asahi_ ## __type)
+
+/* Note: this is an enum so that it can be resolved by Rust bindgen. */
+enum {
+	DRM_IOCTL_ASAHI_GET_PARAMS       = DRM_IOCTL_ASAHI(W, GET_PARAMS, get_params),
+	DRM_IOCTL_ASAHI_GET_TIME         = DRM_IOCTL_ASAHI(WR, GET_TIME, get_time),
+	DRM_IOCTL_ASAHI_VM_CREATE        = DRM_IOCTL_ASAHI(WR, VM_CREATE, vm_create),
+	DRM_IOCTL_ASAHI_VM_DESTROY       = DRM_IOCTL_ASAHI(W, VM_DESTROY, vm_destroy),
+	DRM_IOCTL_ASAHI_VM_BIND          = DRM_IOCTL_ASAHI(W, VM_BIND, vm_bind),
+	DRM_IOCTL_ASAHI_GEM_CREATE       = DRM_IOCTL_ASAHI(WR, GEM_CREATE, gem_create),
+	DRM_IOCTL_ASAHI_GEM_MMAP_OFFSET  = DRM_IOCTL_ASAHI(WR, GEM_MMAP_OFFSET, gem_mmap_offset),
+	DRM_IOCTL_ASAHI_GEM_BIND_OBJECT  = DRM_IOCTL_ASAHI(WR, GEM_BIND_OBJECT, gem_bind_object),
+	DRM_IOCTL_ASAHI_QUEUE_CREATE     = DRM_IOCTL_ASAHI(WR, QUEUE_CREATE, queue_create),
+	DRM_IOCTL_ASAHI_QUEUE_DESTROY    = DRM_IOCTL_ASAHI(W, QUEUE_DESTROY, queue_destroy),
+	DRM_IOCTL_ASAHI_SUBMIT           = DRM_IOCTL_ASAHI(W, SUBMIT, submit),
+};
+
+#if defined(__cplusplus)
+}
+#endif
+
+#endif /* _ASAHI_DRM_H_ */
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index 7fba37b94401..e63a71d3c607 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -905,13 +905,17 @@ struct drm_syncobj_destroy {
 };
 
 #define DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_IMPORT_SYNC_FILE (1 << 0)
+#define DRM_SYNCOBJ_FD_TO_HANDLE_FLAGS_TIMELINE         (1 << 1)
 #define DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_EXPORT_SYNC_FILE (1 << 0)
+#define DRM_SYNCOBJ_HANDLE_TO_FD_FLAGS_TIMELINE         (1 << 1)
 struct drm_syncobj_handle {
 	__u32 handle;
 	__u32 flags;
 
 	__s32 fd;
 	__u32 pad;
+
+	__u64 point;
 };
 
 struct drm_syncobj_transfer {
diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h
index e41a3cec6a9e..81202a50dc9e 100644
--- a/include/uapi/drm/drm_fourcc.h
+++ b/include/uapi/drm/drm_fourcc.h
@@ -422,6 +422,7 @@ extern "C" {
 #define DRM_FORMAT_MOD_VENDOR_ALLWINNER 0x09
 #define DRM_FORMAT_MOD_VENDOR_AMLOGIC 0x0a
 #define DRM_FORMAT_MOD_VENDOR_MTK     0x0b
+#define DRM_FORMAT_MOD_VENDOR_APPLE   0x0c
 
 /* add more to the end as needed */
 
@@ -1495,6 +1496,50 @@ drm_fourcc_canonicalize_nvidia_format_mod(__u64 modifier)
 #define DRM_FORMAT_MOD_MTK_16L_32S_TILE  DRM_FORMAT_MOD_MTK(MTK_FMT_MOD_TILE_16L32S)
 
 /*
+ * Apple GPU-tiled layouts.
+ *
+ * Apple GPUs support nonlinear tilings with optional lossless compression.
+ *
+ * GPU-tiled images are divided into 16KiB tiles:
+ *
+ *     Bytes per pixel  Tile size
+ *     ---------------  ---------
+ *                   1  128x128
+ *                   2  128x64
+ *                   4  64x64
+ *                   8  64x32
+ *                  16  32x32
+ *
+ * Tiles are raster-order. Pixels within a tile are interleaved (Morton order).
+ *
+ * Compressed images pad the body to 128-bytes and are immediately followed by a
+ * metadata section. The metadata section rounds the image dimensions to
+ * powers-of-two and contains 8 bytes for each 16x16 compression subtile.
+ * Subtiles are interleaved (Morton order).
+ *
+ * All images are 128-byte aligned.
+ *
+ * These layouts fundamentally do not have meaningful strides. No matter how we
+ * specify strides for these layouts, userspace unaware of Apple image layouts
+ * will be unable to use correctly the specified stride for any purpose.
+ * Userspace aware of the image layouts do not use strides. The most "correct"
+ * convention would be setting the image stride to 0. Unfortunately, some
+ * software assumes the stride is at least (width * bytes per pixel). We
+ * therefore require that stride equals (width * bytes per pixel). Since the
+ * stride is arbitrary here, we pick the simplest convention.
+ *
+ * Although containing two sections, compressed image layouts are treated in
+ * software as a single plane. This is modelled after AFBC, a similar
+ * scheme. Attempting to separate the sections to be "explicit" in DRM would
+ * only generate more confusion, as software does not treat the image this way.
+ *
+ * For detailed information on the hardware image layouts, see
+ * https://docs.mesa3d.org/drivers/asahi.html#image-layouts
+ */
+#define DRM_FORMAT_MOD_APPLE_GPU_TILED fourcc_mod_code(APPLE, 1)
+#define DRM_FORMAT_MOD_APPLE_GPU_TILED_COMPRESSED fourcc_mod_code(APPLE, 2)
+
+/*
  * AMD modifiers
  *
  * Memory layout:
diff --git a/include/uapi/drm/ivpu_accel.h b/include/uapi/drm/ivpu_accel.h
index 746c43bd3eb6..2f24103f4533 100644
--- a/include/uapi/drm/ivpu_accel.h
+++ b/include/uapi/drm/ivpu_accel.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only WITH Linux-syscall-note */
 /*
- * Copyright (C) 2020-2024 Intel Corporation
+ * Copyright (C) 2020-2025 Intel Corporation
  */
 
 #ifndef __UAPI_IVPU_DRM_H__
@@ -147,7 +147,7 @@ struct drm_ivpu_param {
 	 * platform type when executing on a simulator or emulator (read-only)
 	 *
 	 * %DRM_IVPU_PARAM_CORE_CLOCK_RATE:
-	 * Current PLL frequency (read-only)
+	 * Maximum frequency of the NPU data processing unit clock (read-only)
 	 *
 	 * %DRM_IVPU_PARAM_NUM_CONTEXTS:
 	 * Maximum number of simultaneously existing contexts (read-only)
diff --git a/include/uapi/drm/panthor_drm.h b/include/uapi/drm/panthor_drm.h
index 97e2c4510e69..ad9a70afea6c 100644
--- a/include/uapi/drm/panthor_drm.h
+++ b/include/uapi/drm/panthor_drm.h
@@ -127,6 +127,9 @@ enum drm_panthor_ioctl_id {
 
 	/** @DRM_PANTHOR_TILER_HEAP_DESTROY: Destroy a tiler heap. */
 	DRM_PANTHOR_TILER_HEAP_DESTROY,
+
+	/** @DRM_PANTHOR_BO_SET_LABEL: Label a BO. */
+	DRM_PANTHOR_BO_SET_LABEL,
 };
 
 /**
@@ -978,6 +981,24 @@ struct drm_panthor_tiler_heap_destroy {
 };
 
 /**
+ * struct drm_panthor_bo_set_label - Arguments passed to DRM_IOCTL_PANTHOR_BO_SET_LABEL
+ */
+struct drm_panthor_bo_set_label {
+	/** @handle: Handle of the buffer object to label. */
+	__u32 handle;
+
+	/**  @pad: MBZ. */
+	__u32 pad;
+
+	/**
+	 * @label: User pointer to a NUL-terminated string
+	 *
+	 * Length cannot be greater than 4096
+	 */
+	__u64 label;
+};
+
+/**
  * DRM_IOCTL_PANTHOR() - Build a Panthor IOCTL number
  * @__access: Access type. Must be R, W or RW.
  * @__id: One of the DRM_PANTHOR_xxx id.
@@ -1019,6 +1040,8 @@ enum {
 		DRM_IOCTL_PANTHOR(WR, TILER_HEAP_CREATE, tiler_heap_create),
 	DRM_IOCTL_PANTHOR_TILER_HEAP_DESTROY =
 		DRM_IOCTL_PANTHOR(WR, TILER_HEAP_DESTROY, tiler_heap_destroy),
+	DRM_IOCTL_PANTHOR_BO_SET_LABEL =
+		DRM_IOCTL_PANTHOR(WR, BO_SET_LABEL, bo_set_label),
 };
 
 #if defined(__cplusplus)
diff --git a/include/uapi/drm/virtgpu_drm.h b/include/uapi/drm/virtgpu_drm.h
index c2ce71987e9b..9debb320c34b 100644
--- a/include/uapi/drm/virtgpu_drm.h
+++ b/include/uapi/drm/virtgpu_drm.h
@@ -163,6 +163,12 @@ struct drm_virtgpu_3d_wait {
 	__u32 flags;
 };
 
+#define VIRTGPU_DRM_CAPSET_VIRGL 1
+#define VIRTGPU_DRM_CAPSET_VIRGL2 2
+#define VIRTGPU_DRM_CAPSET_GFXSTREAM_VULKAN 3
+#define VIRTGPU_DRM_CAPSET_VENUS 4
+#define VIRTGPU_DRM_CAPSET_CROSS_DOMAIN 5
+#define VIRTGPU_DRM_CAPSET_DRM 6
 struct drm_virtgpu_get_caps {
 	__u32 cap_set_id;
 	__u32 cap_set_ver;
diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index 616916985e3f..9c08738c3b91 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -917,7 +917,11 @@ struct drm_xe_gem_mmap_offset {
  * struct drm_xe_vm_create - Input of &DRM_IOCTL_XE_VM_CREATE
  *
  * The @flags can be:
- *  - %DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE
+ *  - %DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE - Map the whole virtual address
+ *    space of the VM to scratch page. A vm_bind would overwrite the scratch
+ *    page mapping. This flag is mutually exclusive with the
+ *    %DRM_XE_VM_CREATE_FLAG_FAULT_MODE flag, with an exception of on x2 and
+ *    xe3 platform.
  *  - %DRM_XE_VM_CREATE_FLAG_LR_MODE - An LR, or Long Running VM accepts
  *    exec submissions to its exec_queues that don't have an upper time
  *    limit on the job execution time. But exec submissions to these
diff --git a/include/uapi/linux/ethtool_netlink_generated.h b/include/uapi/linux/ethtool_netlink_generated.h
index fe24c3459ac0..30c8dad6214e 100644
--- a/include/uapi/linux/ethtool_netlink_generated.h
+++ b/include/uapi/linux/ethtool_netlink_generated.h
@@ -31,11 +31,6 @@ enum ethtool_header_flags {
 	ETHTOOL_FLAG_STATS = 4,
 };
 
-enum {
-	ETHTOOL_PHY_UPSTREAM_TYPE_MAC,
-	ETHTOOL_PHY_UPSTREAM_TYPE_PHY,
-};
-
 enum ethtool_tcp_data_split {
 	ETHTOOL_TCP_DATA_SPLIT_UNKNOWN,
 	ETHTOOL_TCP_DATA_SPLIT_DISABLED,
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index ed2beb4def3f..8f1fc12bac46 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -1010,7 +1010,9 @@ struct io_uring_zcrx_ifq_reg {
 	__u64	region_ptr; /* struct io_uring_region_desc * */
 
 	struct io_uring_zcrx_offsets offsets;
-	__u64	__resv[4];
+	__u32	zcrx_id;
+	__u32	__resv2;
+	__u64	__resv[3];
 };
 
 #ifdef __cplusplus
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 1e59344c5673..04c7d283dc7d 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -536,6 +536,8 @@ enum kfd_smi_event {
 	KFD_SMI_EVENT_QUEUE_EVICTION = 9,
 	KFD_SMI_EVENT_QUEUE_RESTORE = 10,
 	KFD_SMI_EVENT_UNMAP_FROM_GPU = 11,
+	KFD_SMI_EVENT_PROCESS_START = 12,
+	KFD_SMI_EVENT_PROCESS_END = 13,
 
 	/*
 	 * max event number, as a flag bit to get events from all processes,
@@ -651,6 +653,9 @@ struct kfd_ioctl_smi_events_args {
 		"%lld -%d @%lx(%lx) %x %d\n", (ns), (pid), (addr), (size),\
 		(node), (unmap_trigger)
 
+#define KFD_EVENT_FMT_PROCESS(pid, task_name)\
+		"%x %s\n", (pid), (task_name)
+
 /**************************************************************************************************
  * CRIU IOCTLs (Checkpoint Restore In Userspace)
  *
diff --git a/include/uapi/linux/landlock.h b/include/uapi/linux/landlock.h
index d9d0cb827117..f030adc462ee 100644
--- a/include/uapi/linux/landlock.h
+++ b/include/uapi/linux/landlock.h
@@ -53,43 +53,70 @@ struct landlock_ruleset_attr {
 	__u64 scoped;
 };
 
-/*
- * sys_landlock_create_ruleset() flags:
+/**
+ * DOC: landlock_create_ruleset_flags
+ *
+ * **Flags**
  *
- * - %LANDLOCK_CREATE_RULESET_VERSION: Get the highest supported Landlock ABI
- *   version.
- * - %LANDLOCK_CREATE_RULESET_ERRATA: Get a bitmask of fixed issues.
+ * %LANDLOCK_CREATE_RULESET_VERSION
+ *     Get the highest supported Landlock ABI version (starting at 1).
+ *
+ * %LANDLOCK_CREATE_RULESET_ERRATA
+ *     Get a bitmask of fixed issues for the current Landlock ABI version.
  */
 /* clang-format off */
 #define LANDLOCK_CREATE_RULESET_VERSION			(1U << 0)
 #define LANDLOCK_CREATE_RULESET_ERRATA			(1U << 1)
 /* clang-format on */
 
-/*
- * sys_landlock_restrict_self() flags:
- *
- * - %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF: Do not create any log related to the
- *   enforced restrictions.  This should only be set by tools launching unknown
- *   or untrusted programs (e.g. a sandbox tool, container runtime, system
- *   service manager).  Because programs sandboxing themselves should fix any
- *   denied access, they should not set this flag to be aware of potential
- *   issues reported by system's logs (i.e. audit).
- * - %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON: Explicitly ask to continue
- *   logging denied access requests even after an :manpage:`execve(2)` call.
- *   This flag should only be set if all the programs than can legitimately be
- *   executed will not try to request a denied access (which could spam audit
- *   logs).
- * - %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF: Do not create any log related
- *   to the enforced restrictions coming from future nested domains created by
- *   the caller or its descendants.  This should only be set according to a
- *   runtime configuration (i.e. not hardcoded) by programs launching other
- *   unknown or untrusted programs that may create their own Landlock domains
- *   and spam logs.  The main use case is for container runtimes to enable users
- *   to mute buggy sandboxed programs for a specific container image.  Other use
- *   cases include sandboxer tools and init systems.  Unlike
- *   %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF,
- *   %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF does not impact the requested
- *   restriction (if any) but only the future nested domains.
+/**
+ * DOC: landlock_restrict_self_flags
+ *
+ * **Flags**
+ *
+ * By default, denied accesses originating from programs that sandbox themselves
+ * are logged via the audit subsystem. Such events typically indicate unexpected
+ * behavior, such as bugs or exploitation attempts. However, to avoid excessive
+ * logging, access requests denied by a domain not created by the originating
+ * program are not logged by default. The rationale is that programs should know
+ * their own behavior, but not necessarily the behavior of other programs.  This
+ * default configuration is suitable for most programs that sandbox themselves.
+ * For specific use cases, the following flags allow programs to modify this
+ * default logging behavior.
+ *
+ * The %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF and
+ * %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON flags apply to the newly created
+ * Landlock domain.
+ *
+ * %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF
+ *     Disables logging of denied accesses originating from the thread creating
+ *     the Landlock domain, as well as its children, as long as they continue
+ *     running the same executable code (i.e., without an intervening
+ *     :manpage:`execve(2)` call). This is intended for programs that execute
+ *     unknown code without invoking :manpage:`execve(2)`, such as script
+ *     interpreters. Programs that only sandbox themselves should not set this
+ *     flag, so users can be notified of unauthorized access attempts via system
+ *     logs.
+ *
+ * %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON
+ *     Enables logging of denied accesses after an :manpage:`execve(2)` call,
+ *     providing visibility into unauthorized access attempts by newly executed
+ *     programs within the created Landlock domain. This flag is recommended
+ *     only when all potential executables in the domain are expected to comply
+ *     with the access restrictions, as excessive audit log entries could make
+ *     it more difficult to identify critical events.
+ *
+ * %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF
+ *     Disables logging of denied accesses originating from nested Landlock
+ *     domains created by the caller or its descendants. This flag should be set
+ *     according to runtime configuration, not hardcoded, to avoid suppressing
+ *     important security events. It is useful for container runtimes or
+ *     sandboxing tools that may launch programs which themselves create
+ *     Landlock domains and could otherwise generate excessive logs. Unlike
+ *     ``LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF``, this flag only affects
+ *     future nested domains, not the one being created. It can also be used
+ *     with a @ruleset_fd value of -1 to mute subdomain logs without creating a
+ *     domain.
  */
 /* clang-format off */
 #define LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF		(1U << 0)
diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h
index b95dd84eef2d..d4b3e2ae1314 100644
--- a/include/uapi/linux/vhost.h
+++ b/include/uapi/linux/vhost.h
@@ -28,10 +28,10 @@
 
 /* Set current process as the (exclusive) owner of this file descriptor.  This
  * must be called before any other vhost command.  Further calls to
- * VHOST_OWNER_SET fail until VHOST_OWNER_RESET is called. */
+ * VHOST_SET_OWNER fail until VHOST_RESET_OWNER is called. */
 #define VHOST_SET_OWNER _IO(VHOST_VIRTIO, 0x01)
 /* Give up ownership, and reset the device to default values.
- * Allows subsequent call to VHOST_OWNER_SET to succeed. */
+ * Allows subsequent call to VHOST_SET_OWNER to succeed. */
 #define VHOST_RESET_OWNER _IO(VHOST_VIRTIO, 0x02)
 
 /* Set up/modify memory layout */
diff --git a/include/uapi/linux/virtio_gpu.h b/include/uapi/linux/virtio_gpu.h
index bf2c9cabd207..be109777d10d 100644
--- a/include/uapi/linux/virtio_gpu.h
+++ b/include/uapi/linux/virtio_gpu.h
@@ -309,8 +309,9 @@ struct virtio_gpu_cmd_submit {
 
 #define VIRTIO_GPU_CAPSET_VIRGL 1
 #define VIRTIO_GPU_CAPSET_VIRGL2 2
-/* 3 is reserved for gfxstream */
+#define VIRTIO_GPU_CAPSET_GFXSTREAM_VULKAN 3
 #define VIRTIO_GPU_CAPSET_VENUS 4
+#define VIRTIO_GPU_CAPSET_CROSS_DOMAIN 5
 #define VIRTIO_GPU_CAPSET_DRM 6
 
 /* VIRTIO_GPU_CMD_GET_CAPSET_INFO */
diff --git a/include/uapi/linux/virtio_pci.h b/include/uapi/linux/virtio_pci.h
index 8549d4571257..c691ac210ce2 100644
--- a/include/uapi/linux/virtio_pci.h
+++ b/include/uapi/linux/virtio_pci.h
@@ -246,6 +246,7 @@ struct virtio_pci_cfg_cap {
 #define VIRTIO_ADMIN_CMD_LIST_USE	0x1
 
 /* Admin command group type. */
+#define VIRTIO_ADMIN_GROUP_TYPE_SELF	0x0
 #define VIRTIO_ADMIN_GROUP_TYPE_SRIOV	0x1
 
 /* Transitional device admin command. */
diff --git a/include/ufs/ufs.h b/include/ufs/ufs.h
index 8a24ed59ec46..1c47136d8715 100644
--- a/include/ufs/ufs.h
+++ b/include/ufs/ufs.h
@@ -180,7 +180,8 @@ enum attr_idn {
 	QUERY_ATTR_IDN_AVAIL_WB_BUFF_SIZE       = 0x1D,
 	QUERY_ATTR_IDN_WB_BUFF_LIFE_TIME_EST    = 0x1E,
 	QUERY_ATTR_IDN_CURR_WB_BUFF_SIZE        = 0x1F,
-	QUERY_ATTR_IDN_TIMESTAMP		= 0x30
+	QUERY_ATTR_IDN_TIMESTAMP		= 0x30,
+	QUERY_ATTR_IDN_DEV_LVL_EXCEPTION_ID     = 0x34,
 };
 
 /* Descriptor idn for Query requests */
@@ -390,6 +391,7 @@ enum {
 	UFS_DEV_EXT_TEMP_NOTIF		= BIT(6),
 	UFS_DEV_HPB_SUPPORT		= BIT(7),
 	UFS_DEV_WRITE_BOOSTER_SUP	= BIT(8),
+	UFS_DEV_LVL_EXCEPTION_SUP       = BIT(12),
 };
 #define UFS_DEV_HPB_SUPPORT_VERSION		0x310
 
@@ -419,6 +421,7 @@ enum {
 	MASK_EE_TOO_LOW_TEMP		= BIT(4),
 	MASK_EE_WRITEBOOSTER_EVENT	= BIT(5),
 	MASK_EE_PERFORMANCE_THROTTLING	= BIT(6),
+	MASK_EE_DEV_LVL_EXCEPTION       = BIT(7),
 	MASK_EE_HEALTH_CRITICAL		= BIT(9),
 };
 #define MASK_EE_URGENT_TEMP (MASK_EE_TOO_HIGH_TEMP | MASK_EE_TOO_LOW_TEMP)
diff --git a/include/ufs/ufs_quirks.h b/include/ufs/ufs_quirks.h
index 41ff44dfa1db..f52de5ed1b3b 100644
--- a/include/ufs/ufs_quirks.h
+++ b/include/ufs/ufs_quirks.h
@@ -107,4 +107,10 @@ struct ufs_dev_quirk {
  */
 #define UFS_DEVICE_QUIRK_DELAY_AFTER_LPM        (1 << 11)
 
+/*
+ * Some ufs devices may need more time to be in hibern8 before exiting.
+ * Enable this quirk to give it an additional 100us.
+ */
+#define UFS_DEVICE_QUIRK_PA_HIBER8TIME          (1 << 12)
+
 #endif /* UFS_QUIRKS_H_ */
diff --git a/include/ufs/ufshcd.h b/include/ufs/ufshcd.h
index e3909cc691b2..e928ed0265ff 100644
--- a/include/ufs/ufshcd.h
+++ b/include/ufs/ufshcd.h
@@ -246,7 +246,7 @@ struct ufs_query {
 struct ufs_dev_cmd {
 	enum dev_cmd_type type;
 	struct mutex lock;
-	struct completion *complete;
+	struct completion complete;
 	struct ufs_query query;
 };
 
@@ -968,6 +968,9 @@ enum ufshcd_mcq_opr {
  * @pm_qos_req: PM QoS request handle
  * @pm_qos_enabled: flag to check if pm qos is enabled
  * @critical_health_count: count of critical health exceptions
+ * @dev_lvl_exception_count: count of device level exceptions since last reset
+ * @dev_lvl_exception_id: vendor specific information about the
+ * device level exception event.
  */
 struct ufs_hba {
 	void __iomem *mmio_base;
@@ -1138,6 +1141,8 @@ struct ufs_hba {
 	bool pm_qos_enabled;
 
 	int critical_health_count;
+	atomic_t dev_lvl_exception_count;
+	u64 dev_lvl_exception_id;
 };
 
 /**
diff --git a/include/vdso/unaligned.h b/include/vdso/unaligned.h
index eee3d2a4dbe4..ff0c06b6513e 100644
--- a/include/vdso/unaligned.h
+++ b/include/vdso/unaligned.h
@@ -2,14 +2,14 @@
 #ifndef __VDSO_UNALIGNED_H
 #define __VDSO_UNALIGNED_H
 
-#define __get_unaligned_t(type, ptr) ({						\
-	const struct { type x; } __packed *__pptr = (typeof(__pptr))(ptr);	\
-	__pptr->x;								\
+#define __get_unaligned_t(type, ptr) ({							\
+	const struct { type x; } __packed * __get_pptr = (typeof(__get_pptr))(ptr);	\
+	__get_pptr->x;									\
 })
 
-#define __put_unaligned_t(type, val, ptr) do {					\
-	struct { type x; } __packed *__pptr = (typeof(__pptr))(ptr);		\
-	__pptr->x = (val);							\
+#define __put_unaligned_t(type, val, ptr) do {						\
+	struct { type x; } __packed * __put_pptr = (typeof(__put_pptr))(ptr);		\
+	__put_pptr->x = (val);								\
 } while (0)
 
 #endif /* __VDSO_UNALIGNED_H */
diff --git a/include/video/pixel_format.h b/include/video/pixel_format.h
new file mode 100644
index 000000000000..b5104b2a3a13
--- /dev/null
+++ b/include/video/pixel_format.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef VIDEO_PIXEL_FORMAT_H
+#define VIDEO_PIXEL_FORMAT_H
+
+struct pixel_format {
+	unsigned char bits_per_pixel;
+	bool indexed;
+	union {
+		struct {
+			struct {
+				unsigned char offset;
+				unsigned char length;
+			} alpha, red, green, blue;
+		};
+		struct {
+			unsigned char offset;
+			unsigned char length;
+		} index;
+	};
+};
+
+#define PIXEL_FORMAT_XRGB1555 \
+	{ 16, false, { .alpha = {0, 0}, .red = {10, 5}, .green = {5, 5}, .blue = {0, 5} } }
+
+#define PIXEL_FORMAT_RGB565 \
+	{ 16, false, { .alpha = {0, 0}, .red = {11, 5}, .green = {5, 6}, .blue = {0, 5} } }
+
+#define PIXEL_FORMAT_RGB888 \
+	{ 24, false, { .alpha = {0, 0}, .red = {16, 8}, .green = {8, 8}, .blue = {0, 8} } }
+
+#define PIXEL_FORMAT_XRGB8888 \
+	{ 32, false, { .alpha = {0, 0}, .red = {16, 8}, .green = {8, 8}, .blue = {0, 8} } }
+
+#define PIXEL_FORMAT_XBGR8888 \
+	{ 32, false, { .alpha = {0, 0}, .red = {0, 8}, .green = {8, 8}, .blue = {16, 8} } }
+
+#define PIXEL_FORMAT_XRGB2101010 \
+	{ 32, false, { .alpha = {0, 0}, .red = {20, 10}, .green = {10, 10}, .blue = {0, 10} } }
+
+#endif
diff --git a/init/Kconfig b/init/Kconfig
index dd2ea3b9a799..63f5974b9fa6 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1555,6 +1555,16 @@ config SYSCTL_ARCH_UNALIGN_ALLOW
 	  the unaligned access emulation.
 	  see arch/parisc/kernel/unaligned.c for reference
 
+config SYSFS_SYSCALL
+	bool "Sysfs syscall support"
+	default n
+	help
+	  sys_sysfs is an obsolete system call no longer supported in libc.
+	  Note that disabling this option is more secure but might break
+	  compatibility with some systems.
+
+	  If unsure say N here.
+
 config HAVE_PCSPKR_PLATFORM
 	bool
 
@@ -1599,16 +1609,6 @@ config SGETMASK_SYSCALL
 
 	  If unsure, leave the default option here.
 
-config SYSFS_SYSCALL
-	bool "Sysfs syscall support" if EXPERT
-	default y
-	help
-	  sys_sysfs is an obsolete system call no longer supported in libc.
-	  Note that disabling this option is more secure but might break
-	  compatibility with some systems.
-
-	  If unsure say Y here.
-
 config FHANDLE
 	bool "open by fhandle syscalls" if EXPERT
 	select EXPORTFS
diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c
index f60d0a9d505e..9414ca6d101c 100644
--- a/io_uring/fdinfo.c
+++ b/io_uring/fdinfo.c
@@ -123,11 +123,11 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file)
 	seq_printf(m, "SqMask:\t0x%x\n", sq_mask);
 	seq_printf(m, "SqHead:\t%u\n", sq_head);
 	seq_printf(m, "SqTail:\t%u\n", sq_tail);
-	seq_printf(m, "CachedSqHead:\t%u\n", ctx->cached_sq_head);
+	seq_printf(m, "CachedSqHead:\t%u\n", data_race(ctx->cached_sq_head));
 	seq_printf(m, "CqMask:\t0x%x\n", cq_mask);
 	seq_printf(m, "CqHead:\t%u\n", cq_head);
 	seq_printf(m, "CqTail:\t%u\n", cq_tail);
-	seq_printf(m, "CachedCqTail:\t%u\n", ctx->cached_cq_tail);
+	seq_printf(m, "CachedCqTail:\t%u\n", data_race(ctx->cached_cq_tail));
 	seq_printf(m, "SQEs:\t%u\n", sq_tail - sq_head);
 	sq_entries = min(sq_tail - sq_head, ctx->sq_entries);
 	for (i = 0; i < sq_entries; i++) {
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index c6209fe44cb1..a2b256e96d5d 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -872,10 +872,15 @@ bool io_req_post_cqe(struct io_kiocb *req, s32 res, u32 cflags)
 	lockdep_assert(!io_wq_current_is_worker());
 	lockdep_assert_held(&ctx->uring_lock);
 
-	__io_cq_lock(ctx);
-	posted = io_fill_cqe_aux(ctx, req->cqe.user_data, res, cflags);
+	if (!ctx->lockless_cq) {
+		spin_lock(&ctx->completion_lock);
+		posted = io_fill_cqe_aux(ctx, req->cqe.user_data, res, cflags);
+		spin_unlock(&ctx->completion_lock);
+	} else {
+		posted = io_fill_cqe_aux(ctx, req->cqe.user_data, res, cflags);
+	}
+
 	ctx->submit_state.cq_flush = true;
-	__io_cq_unlock_post(ctx);
 	return posted;
 }
 
@@ -1078,21 +1083,22 @@ static __cold void __io_fallback_tw(struct llist_node *node, bool sync)
 	while (node) {
 		req = container_of(node, struct io_kiocb, io_task_work.node);
 		node = node->next;
-		if (sync && last_ctx != req->ctx) {
+		if (last_ctx != req->ctx) {
 			if (last_ctx) {
-				flush_delayed_work(&last_ctx->fallback_work);
+				if (sync)
+					flush_delayed_work(&last_ctx->fallback_work);
 				percpu_ref_put(&last_ctx->refs);
 			}
 			last_ctx = req->ctx;
 			percpu_ref_get(&last_ctx->refs);
 		}
-		if (llist_add(&req->io_task_work.node,
-			      &req->ctx->fallback_llist))
-			schedule_delayed_work(&req->ctx->fallback_work, 1);
+		if (llist_add(&req->io_task_work.node, &last_ctx->fallback_llist))
+			schedule_delayed_work(&last_ctx->fallback_work, 1);
 	}
 
 	if (last_ctx) {
-		flush_delayed_work(&last_ctx->fallback_work);
+		if (sync)
+			flush_delayed_work(&last_ctx->fallback_work);
 		percpu_ref_put(&last_ctx->refs);
 	}
 }
diff --git a/io_uring/kbuf.c b/io_uring/kbuf.c
index 098109259671..953d5e742569 100644
--- a/io_uring/kbuf.c
+++ b/io_uring/kbuf.c
@@ -504,6 +504,8 @@ int io_provide_buffers_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe
 	p->nbufs = tmp;
 	p->addr = READ_ONCE(sqe->addr);
 	p->len = READ_ONCE(sqe->len);
+	if (!p->len)
+		return -EINVAL;
 
 	if (check_mul_overflow((unsigned long)p->len, (unsigned long)p->nbufs,
 				&size))
diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c
index 5e64a8bb30a4..f80a77c4973f 100644
--- a/io_uring/rsrc.c
+++ b/io_uring/rsrc.c
@@ -175,6 +175,18 @@ void io_rsrc_cache_free(struct io_ring_ctx *ctx)
 	io_alloc_cache_free(&ctx->imu_cache, kfree);
 }
 
+static void io_clear_table_tags(struct io_rsrc_data *data)
+{
+	int i;
+
+	for (i = 0; i < data->nr; i++) {
+		struct io_rsrc_node *node = data->nodes[i];
+
+		if (node)
+			node->tag = 0;
+	}
+}
+
 __cold void io_rsrc_data_free(struct io_ring_ctx *ctx,
 			      struct io_rsrc_data *data)
 {
@@ -583,6 +595,7 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
 	io_file_table_set_alloc_range(ctx, 0, ctx->file_table.data.nr);
 	return 0;
 fail:
+	io_clear_table_tags(&ctx->file_table.data);
 	io_sqe_files_unregister(ctx);
 	return ret;
 }
@@ -902,8 +915,10 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg,
 	}
 
 	ctx->buf_table = data;
-	if (ret)
+	if (ret) {
+		io_clear_table_tags(&ctx->buf_table);
 		io_sqe_buffers_unregister(ctx);
+	}
 	return ret;
 }
 
@@ -1017,10 +1032,33 @@ static int validate_fixed_range(u64 buf_addr, size_t len,
 	return 0;
 }
 
+static int io_import_kbuf(int ddir, struct iov_iter *iter,
+			  struct io_mapped_ubuf *imu, size_t len, size_t offset)
+{
+	size_t count = len + offset;
+
+	iov_iter_bvec(iter, ddir, imu->bvec, imu->nr_bvecs, count);
+	iov_iter_advance(iter, offset);
+
+	if (count < imu->len) {
+		const struct bio_vec *bvec = iter->bvec;
+
+		while (len > bvec->bv_len) {
+			len -= bvec->bv_len;
+			bvec++;
+		}
+		iter->nr_segs = 1 + bvec - iter->bvec;
+	}
+	return 0;
+}
+
 static int io_import_fixed(int ddir, struct iov_iter *iter,
 			   struct io_mapped_ubuf *imu,
 			   u64 buf_addr, size_t len)
 {
+	const struct bio_vec *bvec;
+	size_t folio_mask;
+	unsigned nr_segs;
 	size_t offset;
 	int ret;
 
@@ -1032,56 +1070,35 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 	if (!(imu->dir & (1 << ddir)))
 		return -EFAULT;
 
-	/*
-	 * Might not be a start of buffer, set size appropriately
-	 * and advance us to the beginning.
-	 */
 	offset = buf_addr - imu->ubuf;
-	iov_iter_bvec(iter, ddir, imu->bvec, imu->nr_bvecs, offset + len);
 
-	if (offset) {
-		/*
-		 * Don't use iov_iter_advance() here, as it's really slow for
-		 * using the latter parts of a big fixed buffer - it iterates
-		 * over each segment manually. We can cheat a bit here for user
-		 * registered nodes, because we know that:
-		 *
-		 * 1) it's a BVEC iter, we set it up
-		 * 2) all bvecs are the same in size, except potentially the
-		 *    first and last bvec
-		 *
-		 * So just find our index, and adjust the iterator afterwards.
-		 * If the offset is within the first bvec (or the whole first
-		 * bvec, just use iov_iter_advance(). This makes it easier
-		 * since we can just skip the first segment, which may not
-		 * be folio_size aligned.
-		 */
-		const struct bio_vec *bvec = imu->bvec;
-
-		/*
-		 * Kernel buffer bvecs, on the other hand, don't necessarily
-		 * have the size property of user registered ones, so we have
-		 * to use the slow iter advance.
-		 */
-		if (offset < bvec->bv_len) {
-			iter->count -= offset;
-			iter->iov_offset = offset;
-		} else if (imu->is_kbuf) {
-			iov_iter_advance(iter, offset);
-		} else {
-			unsigned long seg_skip;
+	if (imu->is_kbuf)
+		return io_import_kbuf(ddir, iter, imu, len, offset);
 
-			/* skip first vec */
-			offset -= bvec->bv_len;
-			seg_skip = 1 + (offset >> imu->folio_shift);
+	/*
+	 * Don't use iov_iter_advance() here, as it's really slow for
+	 * using the latter parts of a big fixed buffer - it iterates
+	 * over each segment manually. We can cheat a bit here for user
+	 * registered nodes, because we know that:
+	 *
+	 * 1) it's a BVEC iter, we set it up
+	 * 2) all bvecs are the same in size, except potentially the
+	 *    first and last bvec
+	 */
+	folio_mask = (1UL << imu->folio_shift) - 1;
+	bvec = imu->bvec;
+	if (offset >= bvec->bv_len) {
+		unsigned long seg_skip;
 
-			iter->bvec += seg_skip;
-			iter->nr_segs -= seg_skip;
-			iter->count -= bvec->bv_len + offset;
-			iter->iov_offset = offset & ((1UL << imu->folio_shift) - 1);
-		}
+		/* skip first vec */
+		offset -= bvec->bv_len;
+		seg_skip = 1 + (offset >> imu->folio_shift);
+		bvec += seg_skip;
+		offset &= folio_mask;
 	}
-
+	nr_segs = (offset + len + bvec->bv_offset + folio_mask) >> imu->folio_shift;
+	iov_iter_bvec(iter, ddir, bvec, nr_segs, len);
+	iter->iov_offset = offset;
 	return 0;
 }
 
diff --git a/io_uring/zcrx.c b/io_uring/zcrx.c
index 80d4a6f71d29..fe86606b9f30 100644
--- a/io_uring/zcrx.c
+++ b/io_uring/zcrx.c
@@ -26,6 +26,11 @@
 #include "zcrx.h"
 #include "rsrc.h"
 
+static inline struct io_zcrx_ifq *io_pp_to_ifq(struct page_pool *pp)
+{
+	return pp->mp_priv;
+}
+
 #define IO_DMA_ATTR (DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
 
 static void __io_zcrx_unmap_area(struct io_zcrx_ifq *ifq,
@@ -46,14 +51,21 @@ static void __io_zcrx_unmap_area(struct io_zcrx_ifq *ifq,
 
 static void io_zcrx_unmap_area(struct io_zcrx_ifq *ifq, struct io_zcrx_area *area)
 {
+	guard(mutex)(&ifq->dma_lock);
+
 	if (area->is_mapped)
 		__io_zcrx_unmap_area(ifq, area, area->nia.num_niovs);
+	area->is_mapped = false;
 }
 
 static int io_zcrx_map_area(struct io_zcrx_ifq *ifq, struct io_zcrx_area *area)
 {
 	int i;
 
+	guard(mutex)(&ifq->dma_lock);
+	if (area->is_mapped)
+		return 0;
+
 	for (i = 0; i < area->nia.num_niovs; i++) {
 		struct net_iov *niov = &area->nia.niovs[i];
 		dma_addr_t dma;
@@ -181,7 +193,7 @@ static void io_zcrx_free_area(struct io_zcrx_area *area)
 	kvfree(area->nia.niovs);
 	kvfree(area->user_refs);
 	if (area->pages) {
-		unpin_user_pages(area->pages, area->nia.num_niovs);
+		unpin_user_pages(area->pages, area->nr_folios);
 		kvfree(area->pages);
 	}
 	kfree(area);
@@ -192,7 +204,7 @@ static int io_zcrx_create_area(struct io_zcrx_ifq *ifq,
 			       struct io_uring_zcrx_area_reg *area_reg)
 {
 	struct io_zcrx_area *area;
-	int i, ret, nr_pages;
+	int i, ret, nr_pages, nr_iovs;
 	struct iovec iov;
 
 	if (area_reg->flags || area_reg->rq_area_token)
@@ -220,27 +232,28 @@ static int io_zcrx_create_area(struct io_zcrx_ifq *ifq,
 		area->pages = NULL;
 		goto err;
 	}
-	area->nia.num_niovs = nr_pages;
+	area->nr_folios = nr_iovs = nr_pages;
+	area->nia.num_niovs = nr_iovs;
 
-	area->nia.niovs = kvmalloc_array(nr_pages, sizeof(area->nia.niovs[0]),
+	area->nia.niovs = kvmalloc_array(nr_iovs, sizeof(area->nia.niovs[0]),
 					 GFP_KERNEL | __GFP_ZERO);
 	if (!area->nia.niovs)
 		goto err;
 
-	area->freelist = kvmalloc_array(nr_pages, sizeof(area->freelist[0]),
+	area->freelist = kvmalloc_array(nr_iovs, sizeof(area->freelist[0]),
 					GFP_KERNEL | __GFP_ZERO);
 	if (!area->freelist)
 		goto err;
 
-	for (i = 0; i < nr_pages; i++)
+	for (i = 0; i < nr_iovs; i++)
 		area->freelist[i] = i;
 
-	area->user_refs = kvmalloc_array(nr_pages, sizeof(area->user_refs[0]),
+	area->user_refs = kvmalloc_array(nr_iovs, sizeof(area->user_refs[0]),
 					GFP_KERNEL | __GFP_ZERO);
 	if (!area->user_refs)
 		goto err;
 
-	for (i = 0; i < nr_pages; i++) {
+	for (i = 0; i < nr_iovs; i++) {
 		struct net_iov *niov = &area->nia.niovs[i];
 
 		niov->owner = &area->nia;
@@ -248,7 +261,7 @@ static int io_zcrx_create_area(struct io_zcrx_ifq *ifq,
 		atomic_set(&area->user_refs[i], 0);
 	}
 
-	area->free_count = nr_pages;
+	area->free_count = nr_iovs;
 	area->ifq = ifq;
 	/* we're only supporting one area per ifq for now */
 	area->area_id = 0;
@@ -274,6 +287,7 @@ static struct io_zcrx_ifq *io_zcrx_ifq_alloc(struct io_ring_ctx *ctx)
 	ifq->ctx = ctx;
 	spin_lock_init(&ifq->lock);
 	spin_lock_init(&ifq->rq_lock);
+	mutex_init(&ifq->dma_lock);
 	return ifq;
 }
 
@@ -323,6 +337,7 @@ static void io_zcrx_ifq_free(struct io_zcrx_ifq *ifq)
 		put_device(ifq->dev);
 
 	io_free_rbuf_ring(ifq);
+	mutex_destroy(&ifq->dma_lock);
 	kfree(ifq);
 }
 
@@ -353,7 +368,8 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 		return -EFAULT;
 	if (copy_from_user(&rd, u64_to_user_ptr(reg.region_ptr), sizeof(rd)))
 		return -EFAULT;
-	if (memchr_inv(&reg.__resv, 0, sizeof(reg.__resv)))
+	if (memchr_inv(&reg.__resv, 0, sizeof(reg.__resv)) ||
+	    reg.__resv2 || reg.zcrx_id)
 		return -EINVAL;
 	if (reg.if_rxq == -1 || !reg.rq_entries || reg.flags)
 		return -EINVAL;
@@ -393,10 +409,6 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 		goto err;
 	get_device(ifq->dev);
 
-	ret = io_zcrx_map_area(ifq, ifq->area);
-	if (ret)
-		goto err;
-
 	mp_param.mp_ops = &io_uring_pp_zc_ops;
 	mp_param.mp_priv = ifq;
 	ret = net_mp_open_rxq(ifq->netdev, reg.if_rxq, &mp_param);
@@ -584,7 +596,7 @@ static void io_zcrx_refill_slow(struct page_pool *pp, struct io_zcrx_ifq *ifq)
 
 static netmem_ref io_pp_zc_alloc_netmems(struct page_pool *pp, gfp_t gfp)
 {
-	struct io_zcrx_ifq *ifq = pp->mp_priv;
+	struct io_zcrx_ifq *ifq = io_pp_to_ifq(pp);
 
 	/* pp should already be ensuring that */
 	if (unlikely(pp->alloc.count))
@@ -616,7 +628,8 @@ static bool io_pp_zc_release_netmem(struct page_pool *pp, netmem_ref netmem)
 
 static int io_pp_zc_init(struct page_pool *pp)
 {
-	struct io_zcrx_ifq *ifq = pp->mp_priv;
+	struct io_zcrx_ifq *ifq = io_pp_to_ifq(pp);
+	int ret;
 
 	if (WARN_ON_ONCE(!ifq))
 		return -EINVAL;
@@ -629,13 +642,17 @@ static int io_pp_zc_init(struct page_pool *pp)
 	if (pp->p.dma_dir != DMA_FROM_DEVICE)
 		return -EOPNOTSUPP;
 
+	ret = io_zcrx_map_area(ifq, ifq->area);
+	if (ret)
+		return ret;
+
 	percpu_ref_get(&ifq->ctx->refs);
 	return 0;
 }
 
 static void io_pp_zc_destroy(struct page_pool *pp)
 {
-	struct io_zcrx_ifq *ifq = pp->mp_priv;
+	struct io_zcrx_ifq *ifq = io_pp_to_ifq(pp);
 	struct io_zcrx_area *area = ifq->area;
 
 	if (WARN_ON_ONCE(area->free_count != area->nia.num_niovs))
@@ -664,6 +681,9 @@ static void io_pp_uninstall(void *mp_priv, struct netdev_rx_queue *rxq)
 	struct io_zcrx_ifq *ifq = mp_priv;
 
 	io_zcrx_drop_netdev(ifq);
+	if (ifq->area)
+		io_zcrx_unmap_area(ifq, ifq->area);
+
 	p->mp_ops = NULL;
 	p->mp_priv = NULL;
 }
@@ -790,7 +810,7 @@ static int io_zcrx_recv_frag(struct io_kiocb *req, struct io_zcrx_ifq *ifq,
 
 	niov = netmem_to_net_iov(frag->netmem);
 	if (niov->pp->mp_ops != &io_uring_pp_zc_ops ||
-	    niov->pp->mp_priv != ifq)
+	    io_pp_to_ifq(niov->pp) != ifq)
 		return -EFAULT;
 
 	if (!io_zcrx_queue_cqe(req, niov, ifq, off + skb_frag_off(frag), len))
diff --git a/io_uring/zcrx.h b/io_uring/zcrx.h
index 706cc7300780..f2bc811f022c 100644
--- a/io_uring/zcrx.h
+++ b/io_uring/zcrx.h
@@ -15,6 +15,7 @@ struct io_zcrx_area {
 	bool			is_mapped;
 	u16			area_id;
 	struct page		**pages;
+	unsigned long		nr_folios;
 
 	/* freelist */
 	spinlock_t		freelist_lock ____cacheline_aligned_in_smp;
@@ -26,17 +27,18 @@ struct io_zcrx_ifq {
 	struct io_ring_ctx		*ctx;
 	struct io_zcrx_area		*area;
 
+	spinlock_t			rq_lock ____cacheline_aligned_in_smp;
 	struct io_uring			*rq_ring;
 	struct io_uring_zcrx_rqe	*rqes;
-	u32				rq_entries;
 	u32				cached_rq_head;
-	spinlock_t			rq_lock;
+	u32				rq_entries;
 
 	u32				if_rxq;
 	struct device			*dev;
 	struct net_device		*netdev;
 	netdevice_tracker		netdev_tracker;
 	spinlock_t			lock;
+	struct mutex			dma_lock;
 };
 
 #if defined(CONFIG_IO_URING_ZCRX)
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index 367eaf2c78b7..0ebbbe37a60f 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -347,12 +347,17 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
 /* Get path information necessary for adding watches. */
 static int audit_get_nd(struct audit_watch *watch, struct path *parent)
 {
-	struct dentry *d = kern_path_locked(watch->path, parent);
+	struct dentry *d;
+
+	d = kern_path_locked_negative(watch->path, parent);
 	if (IS_ERR(d))
 		return PTR_ERR(d);
-	/* update watch filter fields */
-	watch->dev = d->d_sb->s_dev;
-	watch->ino = d_backing_inode(d)->i_ino;
+
+	if (d_is_positive(d)) {
+		/* update watch filter fields */
+		watch->dev = d->d_sb->s_dev;
+		watch->ino = d_backing_inode(d)->i_ino;
+	}
 
 	inode_unlock(d_backing_inode(parent->dentry));
 	dput(d);
@@ -418,11 +423,10 @@ int audit_add_watch(struct audit_krule *krule, struct list_head **list)
 	/* caller expects mutex locked */
 	mutex_lock(&audit_filter_mutex);
 
-	if (ret && ret != -ENOENT) {
+	if (ret) {
 		audit_put_watch(watch);
 		return ret;
 	}
-	ret = 0;
 
 	/* either find an old parent or attach a new one */
 	parent = audit_find_parent(d_backing_inode(parent_path.dentry));
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 5a5adc66b8e2..92b606d60020 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -2189,7 +2189,7 @@ static long bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_
 		b = &htab->buckets[i];
 		rcu_read_lock();
 		head = &b->head;
-		hlist_nulls_for_each_entry_rcu(elem, n, head, hash_node) {
+		hlist_nulls_for_each_entry_safe(elem, n, head, hash_node) {
 			key = elem->key;
 			if (is_percpu) {
 				/* current cpu value for percpu map */
diff --git a/kernel/bpf/preload/bpf_preload_kern.c b/kernel/bpf/preload/bpf_preload_kern.c
index 2fdf3c978db1..774e5a538811 100644
--- a/kernel/bpf/preload/bpf_preload_kern.c
+++ b/kernel/bpf/preload/bpf_preload_kern.c
@@ -89,5 +89,6 @@ static void __exit fini(void)
 }
 late_initcall(load);
 module_exit(fini);
+MODULE_IMPORT_NS("BPF_INTERNAL");
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Embedded BPF programs for introspection in bpffs");
diff --git a/kernel/bpf/queue_stack_maps.c b/kernel/bpf/queue_stack_maps.c
index d869f51ea93a..9a5f94371e50 100644
--- a/kernel/bpf/queue_stack_maps.c
+++ b/kernel/bpf/queue_stack_maps.c
@@ -9,13 +9,14 @@
 #include <linux/slab.h>
 #include <linux/btf_ids.h>
 #include "percpu_freelist.h"
+#include <asm/rqspinlock.h>
 
 #define QUEUE_STACK_CREATE_FLAG_MASK \
 	(BPF_F_NUMA_NODE | BPF_F_ACCESS_MASK)
 
 struct bpf_queue_stack {
 	struct bpf_map map;
-	raw_spinlock_t lock;
+	rqspinlock_t lock;
 	u32 head, tail;
 	u32 size; /* max_entries + 1 */
 
@@ -78,7 +79,7 @@ static struct bpf_map *queue_stack_map_alloc(union bpf_attr *attr)
 
 	qs->size = size;
 
-	raw_spin_lock_init(&qs->lock);
+	raw_res_spin_lock_init(&qs->lock);
 
 	return &qs->map;
 }
@@ -98,12 +99,8 @@ static long __queue_map_get(struct bpf_map *map, void *value, bool delete)
 	int err = 0;
 	void *ptr;
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&qs->lock, flags))
-			return -EBUSY;
-	} else {
-		raw_spin_lock_irqsave(&qs->lock, flags);
-	}
+	if (raw_res_spin_lock_irqsave(&qs->lock, flags))
+		return -EBUSY;
 
 	if (queue_stack_map_is_empty(qs)) {
 		memset(value, 0, qs->map.value_size);
@@ -120,7 +117,7 @@ static long __queue_map_get(struct bpf_map *map, void *value, bool delete)
 	}
 
 out:
-	raw_spin_unlock_irqrestore(&qs->lock, flags);
+	raw_res_spin_unlock_irqrestore(&qs->lock, flags);
 	return err;
 }
 
@@ -133,12 +130,8 @@ static long __stack_map_get(struct bpf_map *map, void *value, bool delete)
 	void *ptr;
 	u32 index;
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&qs->lock, flags))
-			return -EBUSY;
-	} else {
-		raw_spin_lock_irqsave(&qs->lock, flags);
-	}
+	if (raw_res_spin_lock_irqsave(&qs->lock, flags))
+		return -EBUSY;
 
 	if (queue_stack_map_is_empty(qs)) {
 		memset(value, 0, qs->map.value_size);
@@ -157,7 +150,7 @@ static long __stack_map_get(struct bpf_map *map, void *value, bool delete)
 		qs->head = index;
 
 out:
-	raw_spin_unlock_irqrestore(&qs->lock, flags);
+	raw_res_spin_unlock_irqrestore(&qs->lock, flags);
 	return err;
 }
 
@@ -203,12 +196,8 @@ static long queue_stack_map_push_elem(struct bpf_map *map, void *value,
 	if (flags & BPF_NOEXIST || flags > BPF_EXIST)
 		return -EINVAL;
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&qs->lock, irq_flags))
-			return -EBUSY;
-	} else {
-		raw_spin_lock_irqsave(&qs->lock, irq_flags);
-	}
+	if (raw_res_spin_lock_irqsave(&qs->lock, irq_flags))
+		return -EBUSY;
 
 	if (queue_stack_map_is_full(qs)) {
 		if (!replace) {
@@ -227,7 +216,7 @@ static long queue_stack_map_push_elem(struct bpf_map *map, void *value,
 		qs->head = 0;
 
 out:
-	raw_spin_unlock_irqrestore(&qs->lock, irq_flags);
+	raw_res_spin_unlock_irqrestore(&qs->lock, irq_flags);
 	return err;
 }
 
diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c
index 1499d8caa9a3..719d73299397 100644
--- a/kernel/bpf/ringbuf.c
+++ b/kernel/bpf/ringbuf.c
@@ -11,6 +11,7 @@
 #include <linux/kmemleak.h>
 #include <uapi/linux/btf.h>
 #include <linux/btf_ids.h>
+#include <asm/rqspinlock.h>
 
 #define RINGBUF_CREATE_FLAG_MASK (BPF_F_NUMA_NODE)
 
@@ -29,7 +30,7 @@ struct bpf_ringbuf {
 	u64 mask;
 	struct page **pages;
 	int nr_pages;
-	raw_spinlock_t spinlock ____cacheline_aligned_in_smp;
+	rqspinlock_t spinlock ____cacheline_aligned_in_smp;
 	/* For user-space producer ring buffers, an atomic_t busy bit is used
 	 * to synchronize access to the ring buffers in the kernel, rather than
 	 * the spinlock that is used for kernel-producer ring buffers. This is
@@ -173,7 +174,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node)
 	if (!rb)
 		return NULL;
 
-	raw_spin_lock_init(&rb->spinlock);
+	raw_res_spin_lock_init(&rb->spinlock);
 	atomic_set(&rb->busy, 0);
 	init_waitqueue_head(&rb->waitq);
 	init_irq_work(&rb->work, bpf_ringbuf_notify);
@@ -416,12 +417,8 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 
 	cons_pos = smp_load_acquire(&rb->consumer_pos);
 
-	if (in_nmi()) {
-		if (!raw_spin_trylock_irqsave(&rb->spinlock, flags))
-			return NULL;
-	} else {
-		raw_spin_lock_irqsave(&rb->spinlock, flags);
-	}
+	if (raw_res_spin_lock_irqsave(&rb->spinlock, flags))
+		return NULL;
 
 	pend_pos = rb->pending_pos;
 	prod_pos = rb->producer_pos;
@@ -446,7 +443,7 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 	 */
 	if (new_prod_pos - cons_pos > rb->mask ||
 	    new_prod_pos - pend_pos > rb->mask) {
-		raw_spin_unlock_irqrestore(&rb->spinlock, flags);
+		raw_res_spin_unlock_irqrestore(&rb->spinlock, flags);
 		return NULL;
 	}
 
@@ -458,7 +455,7 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size)
 	/* pairs with consumer's smp_load_acquire() */
 	smp_store_release(&rb->producer_pos, new_prod_pos);
 
-	raw_spin_unlock_irqrestore(&rb->spinlock, flags);
+	raw_res_spin_unlock_irqrestore(&rb->spinlock, flags);
 
 	return (void *)hdr + BPF_RINGBUF_HDR_SZ;
 }
diff --git a/kernel/bpf/rqspinlock.c b/kernel/bpf/rqspinlock.c
index b896c4a75a5c..338305c8852c 100644
--- a/kernel/bpf/rqspinlock.c
+++ b/kernel/bpf/rqspinlock.c
@@ -253,7 +253,7 @@ static noinline int check_timeout(rqspinlock_t *lock, u32 mask,
 	})
 #else
 #define RES_CHECK_TIMEOUT(ts, ret, mask)			      \
-	({ (ret) = check_timeout(&(ts)); })
+	({ (ret) = check_timeout((lock), (mask), &(ts)); })
 #endif
 
 /*
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 9794446bc8c6..64c3393e8270 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1583,7 +1583,7 @@ struct bpf_map *bpf_map_get(u32 ufd)
 
 	return map;
 }
-EXPORT_SYMBOL(bpf_map_get);
+EXPORT_SYMBOL_NS(bpf_map_get, "BPF_INTERNAL");
 
 struct bpf_map *bpf_map_get_with_uref(u32 ufd)
 {
@@ -3364,7 +3364,7 @@ struct bpf_link *bpf_link_get_from_fd(u32 ufd)
 	bpf_link_inc(link);
 	return link;
 }
-EXPORT_SYMBOL(bpf_link_get_from_fd);
+EXPORT_SYMBOL_NS(bpf_link_get_from_fd, "BPF_INTERNAL");
 
 static void bpf_tracing_link_release(struct bpf_link *link)
 {
@@ -6020,7 +6020,7 @@ int kern_sys_bpf(int cmd, union bpf_attr *attr, unsigned int size)
 		return ____bpf_sys_bpf(cmd, attr, size);
 	}
 }
-EXPORT_SYMBOL(kern_sys_bpf);
+EXPORT_SYMBOL_NS(kern_sys_bpf, "BPF_INTERNAL");
 
 static const struct bpf_func_proto bpf_sys_bpf_proto = {
 	.func		= bpf_sys_bpf,
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 27f08aa17b56..63e5b90da1f3 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -90,7 +90,7 @@
 DEFINE_MUTEX(cgroup_mutex);
 DEFINE_SPINLOCK(css_set_lock);
 
-#ifdef CONFIG_PROVE_RCU
+#if (defined CONFIG_PROVE_RCU || defined CONFIG_LOCKDEP)
 EXPORT_SYMBOL_GPL(cgroup_mutex);
 EXPORT_SYMBOL_GPL(css_set_lock);
 #endif
@@ -2353,9 +2353,37 @@ static struct file_system_type cgroup2_fs_type = {
 };
 
 #ifdef CONFIG_CPUSETS_V1
+enum cpuset_param {
+	Opt_cpuset_v2_mode,
+};
+
+static const struct fs_parameter_spec cpuset_fs_parameters[] = {
+	fsparam_flag  ("cpuset_v2_mode", Opt_cpuset_v2_mode),
+	{}
+};
+
+static int cpuset_parse_param(struct fs_context *fc, struct fs_parameter *param)
+{
+	struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+	struct fs_parse_result result;
+	int opt;
+
+	opt = fs_parse(fc, cpuset_fs_parameters, param, &result);
+	if (opt < 0)
+		return opt;
+
+	switch (opt) {
+	case Opt_cpuset_v2_mode:
+		ctx->flags |= CGRP_ROOT_CPUSET_V2_MODE;
+		return 0;
+	}
+	return -EINVAL;
+}
+
 static const struct fs_context_operations cpuset_fs_context_ops = {
 	.get_tree	= cgroup1_get_tree,
 	.free		= cgroup_fs_context_free,
+	.parse_param	= cpuset_parse_param,
 };
 
 /*
@@ -2392,6 +2420,7 @@ static int cpuset_init_fs_context(struct fs_context *fc)
 static struct file_system_type cpuset_fs_type = {
 	.name			= "cpuset",
 	.init_fs_context	= cpuset_init_fs_context,
+	.parameters		= cpuset_fs_parameters,
 	.fs_flags		= FS_USERNS_MOUNT,
 };
 #endif
@@ -5923,6 +5952,12 @@ static void kill_css(struct cgroup_subsys_state *css)
 	if (css->flags & CSS_DYING)
 		return;
 
+	/*
+	 * Call css_killed(), if defined, before setting the CSS_DYING flag
+	 */
+	if (css->ss->css_killed)
+		css->ss->css_killed(css);
+
 	css->flags |= CSS_DYING;
 
 	/*
diff --git a/kernel/cgroup/cpuset-internal.h b/kernel/cgroup/cpuset-internal.h
index 976a8bc3ff60..383963e28ac6 100644
--- a/kernel/cgroup/cpuset-internal.h
+++ b/kernel/cgroup/cpuset-internal.h
@@ -33,6 +33,7 @@ enum prs_errcode {
 	PERR_CPUSEMPTY,
 	PERR_HKEEPING,
 	PERR_ACCESS,
+	PERR_REMOTE,
 };
 
 /* bits in struct cpuset flags field */
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 39c1fc643d77..306b60430091 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -61,10 +61,17 @@ static const char * const perr_strings[] = {
 	[PERR_CPUSEMPTY] = "cpuset.cpus and cpuset.cpus.exclusive are empty",
 	[PERR_HKEEPING]  = "partition config conflicts with housekeeping setup",
 	[PERR_ACCESS]    = "Enable partition not permitted",
+	[PERR_REMOTE]    = "Have remote partition underneath",
 };
 
 /*
- * Exclusive CPUs distributed out to sub-partitions of top_cpuset
+ * For local partitions, update to subpartitions_cpus & isolated_cpus is done
+ * in update_parent_effective_cpumask(). For remote partitions, it is done in
+ * the remote_partition_*() and remote_cpus_update() helpers.
+ */
+/*
+ * Exclusive CPUs distributed out to local or remote sub-partitions of
+ * top_cpuset
  */
 static cpumask_var_t	subpartitions_cpus;
 
@@ -86,7 +93,6 @@ static struct list_head remote_children;
  * A flag to force sched domain rebuild at the end of an operation.
  * It can be set in
  *  - update_partition_sd_lb()
- *  - remote_partition_check()
  *  - update_cpumasks_hier()
  *  - cpuset_update_flag()
  *  - cpuset_hotplug_update_tasks()
@@ -1089,9 +1095,14 @@ void cpuset_reset_sched_domains(void)
  *
  * Iterate through each task of @cs updating its cpus_allowed to the
  * effective cpuset's.  As this function is called with cpuset_mutex held,
- * cpuset membership stays stable. For top_cpuset, task_cpu_possible_mask()
- * is used instead of effective_cpus to make sure all offline CPUs are also
- * included as hotplug code won't update cpumasks for tasks in top_cpuset.
+ * cpuset membership stays stable.
+ *
+ * For top_cpuset, task_cpu_possible_mask() is used instead of effective_cpus
+ * to make sure all offline CPUs are also included as hotplug code won't
+ * update cpumasks for tasks in top_cpuset.
+ *
+ * As task_cpu_possible_mask() can be task dependent in arm64, we have to
+ * do cpu masking per task instead of doing it once for all.
  */
 void cpuset_update_tasks_cpumask(struct cpuset *cs, struct cpumask *new_cpus)
 {
@@ -1151,7 +1162,7 @@ static void update_sibling_cpumasks(struct cpuset *parent, struct cpuset *cs,
  *
  * Return: 0 if successful, an error code otherwise
  */
-static int update_partition_exclusive(struct cpuset *cs, int new_prs)
+static int update_partition_exclusive_flag(struct cpuset *cs, int new_prs)
 {
 	bool exclusive = (new_prs > PRS_MEMBER);
 
@@ -1234,12 +1245,12 @@ static void reset_partition_data(struct cpuset *cs)
 }
 
 /*
- * partition_xcpus_newstate - Exclusive CPUs state change
+ * isolated_cpus_update - Update the isolated_cpus mask
  * @old_prs: old partition_root_state
  * @new_prs: new partition_root_state
  * @xcpus: exclusive CPUs with state change
  */
-static void partition_xcpus_newstate(int old_prs, int new_prs, struct cpumask *xcpus)
+static void isolated_cpus_update(int old_prs, int new_prs, struct cpumask *xcpus)
 {
 	WARN_ON_ONCE(old_prs == new_prs);
 	if (new_prs == PRS_ISOLATED)
@@ -1273,8 +1284,8 @@ static bool partition_xcpus_add(int new_prs, struct cpuset *parent,
 
 	isolcpus_updated = (new_prs != parent->partition_root_state);
 	if (isolcpus_updated)
-		partition_xcpus_newstate(parent->partition_root_state, new_prs,
-					 xcpus);
+		isolated_cpus_update(parent->partition_root_state, new_prs,
+				     xcpus);
 
 	cpumask_andnot(parent->effective_cpus, parent->effective_cpus, xcpus);
 	return isolcpus_updated;
@@ -1304,8 +1315,8 @@ static bool partition_xcpus_del(int old_prs, struct cpuset *parent,
 
 	isolcpus_updated = (old_prs != parent->partition_root_state);
 	if (isolcpus_updated)
-		partition_xcpus_newstate(old_prs, parent->partition_root_state,
-					 xcpus);
+		isolated_cpus_update(old_prs, parent->partition_root_state,
+				     xcpus);
 
 	cpumask_and(xcpus, xcpus, cpu_active_mask);
 	cpumask_or(parent->effective_cpus, parent->effective_cpus, xcpus);
@@ -1340,20 +1351,57 @@ EXPORT_SYMBOL_GPL(cpuset_cpu_is_isolated);
  * compute_effective_exclusive_cpumask - compute effective exclusive CPUs
  * @cs: cpuset
  * @xcpus: effective exclusive CPUs value to be set
- * Return: true if xcpus is not empty, false otherwise.
+ * @real_cs: the real cpuset (can be NULL)
+ * Return: 0 if there is no sibling conflict, > 0 otherwise
  *
- * Starting with exclusive_cpus (cpus_allowed if exclusive_cpus is not set),
- * it must be a subset of parent's effective_xcpus.
+ * If exclusive_cpus isn't explicitly set or a real_cs is provided, we have to
+ * scan the sibling cpusets and exclude their exclusive_cpus or effective_xcpus
+ * as well. The provision of real_cs means that a cpumask is being changed and
+ * the given cs is a trial one.
  */
-static bool compute_effective_exclusive_cpumask(struct cpuset *cs,
-						struct cpumask *xcpus)
+static int compute_effective_exclusive_cpumask(struct cpuset *cs,
+					       struct cpumask *xcpus,
+					       struct cpuset *real_cs)
 {
+	struct cgroup_subsys_state *css;
 	struct cpuset *parent = parent_cs(cs);
+	struct cpuset *sibling;
+	int retval = 0;
 
 	if (!xcpus)
 		xcpus = cs->effective_xcpus;
 
-	return cpumask_and(xcpus, user_xcpus(cs), parent->effective_xcpus);
+	cpumask_and(xcpus, user_xcpus(cs), parent->effective_xcpus);
+
+	if (!real_cs) {
+		if (!cpumask_empty(cs->exclusive_cpus))
+			return 0;
+	} else {
+		cs = real_cs;
+	}
+
+	/*
+	 * Exclude exclusive CPUs from siblings
+	 */
+	rcu_read_lock();
+	cpuset_for_each_child(sibling, css, parent) {
+		if (sibling == cs)
+			continue;
+
+		if (!cpumask_empty(sibling->exclusive_cpus) &&
+		    cpumask_intersects(xcpus, sibling->exclusive_cpus)) {
+			cpumask_andnot(xcpus, xcpus, sibling->exclusive_cpus);
+			retval++;
+			continue;
+		}
+		if (!cpumask_empty(sibling->effective_xcpus) &&
+		    cpumask_intersects(xcpus, sibling->effective_xcpus)) {
+			cpumask_andnot(xcpus, xcpus, sibling->effective_xcpus);
+			retval++;
+		}
+	}
+	rcu_read_unlock();
+	return retval;
 }
 
 static inline bool is_remote_partition(struct cpuset *cs)
@@ -1395,7 +1443,7 @@ static int remote_partition_enable(struct cpuset *cs, int new_prs,
 	 * remote partition root underneath it, its exclusive_cpus must
 	 * have overlapped with subpartitions_cpus.
 	 */
-	compute_effective_exclusive_cpumask(cs, tmp->new_cpus);
+	compute_effective_exclusive_cpumask(cs, tmp->new_cpus, NULL);
 	if (cpumask_empty(tmp->new_cpus) ||
 	    cpumask_intersects(tmp->new_cpus, subpartitions_cpus) ||
 	    cpumask_subset(top_cpuset.effective_cpus, tmp->new_cpus))
@@ -1404,8 +1452,11 @@ static int remote_partition_enable(struct cpuset *cs, int new_prs,
 	spin_lock_irq(&callback_lock);
 	isolcpus_updated = partition_xcpus_add(new_prs, NULL, tmp->new_cpus);
 	list_add(&cs->remote_sibling, &remote_children);
+	cpumask_copy(cs->effective_xcpus, tmp->new_cpus);
 	spin_unlock_irq(&callback_lock);
 	update_unbound_workqueue_cpumask(isolcpus_updated);
+	cpuset_force_rebuild();
+	cs->prs_err = 0;
 
 	/*
 	 * Propagate changes in top_cpuset's effective_cpus down the hierarchy.
@@ -1428,20 +1479,24 @@ static void remote_partition_disable(struct cpuset *cs, struct tmpmasks *tmp)
 {
 	bool isolcpus_updated;
 
-	compute_effective_exclusive_cpumask(cs, tmp->new_cpus);
 	WARN_ON_ONCE(!is_remote_partition(cs));
-	WARN_ON_ONCE(!cpumask_subset(tmp->new_cpus, subpartitions_cpus));
+	WARN_ON_ONCE(!cpumask_subset(cs->effective_xcpus, subpartitions_cpus));
 
 	spin_lock_irq(&callback_lock);
 	list_del_init(&cs->remote_sibling);
 	isolcpus_updated = partition_xcpus_del(cs->partition_root_state,
-					       NULL, tmp->new_cpus);
-	cs->partition_root_state = -cs->partition_root_state;
-	if (!cs->prs_err)
-		cs->prs_err = PERR_INVCPUS;
+					       NULL, cs->effective_xcpus);
+	if (cs->prs_err)
+		cs->partition_root_state = -cs->partition_root_state;
+	else
+		cs->partition_root_state = PRS_MEMBER;
+
+	/* effective_xcpus may need to be changed */
+	compute_effective_exclusive_cpumask(cs, NULL, NULL);
 	reset_partition_data(cs);
 	spin_unlock_irq(&callback_lock);
 	update_unbound_workqueue_cpumask(isolcpus_updated);
+	cpuset_force_rebuild();
 
 	/*
 	 * Propagate changes in top_cpuset's effective_cpus down the hierarchy.
@@ -1453,14 +1508,15 @@ static void remote_partition_disable(struct cpuset *cs, struct tmpmasks *tmp)
 /*
  * remote_cpus_update - cpus_exclusive change of remote partition
  * @cs: the cpuset to be updated
- * @newmask: the new effective_xcpus mask
+ * @xcpus: the new exclusive_cpus mask, if non-NULL
+ * @excpus: the new effective_xcpus mask
  * @tmp: temporary masks
  *
  * top_cpuset and subpartitions_cpus will be updated or partition can be
  * invalidated.
  */
-static void remote_cpus_update(struct cpuset *cs, struct cpumask *newmask,
-			       struct tmpmasks *tmp)
+static void remote_cpus_update(struct cpuset *cs, struct cpumask *xcpus,
+			       struct cpumask *excpus, struct tmpmasks *tmp)
 {
 	bool adding, deleting;
 	int prs = cs->partition_root_state;
@@ -1471,29 +1527,45 @@ static void remote_cpus_update(struct cpuset *cs, struct cpumask *newmask,
 
 	WARN_ON_ONCE(!cpumask_subset(cs->effective_xcpus, subpartitions_cpus));
 
-	if (cpumask_empty(newmask))
+	if (cpumask_empty(excpus)) {
+		cs->prs_err = PERR_CPUSEMPTY;
 		goto invalidate;
+	}
 
-	adding   = cpumask_andnot(tmp->addmask, newmask, cs->effective_xcpus);
-	deleting = cpumask_andnot(tmp->delmask, cs->effective_xcpus, newmask);
+	adding   = cpumask_andnot(tmp->addmask, excpus, cs->effective_xcpus);
+	deleting = cpumask_andnot(tmp->delmask, cs->effective_xcpus, excpus);
 
 	/*
 	 * Additions of remote CPUs is only allowed if those CPUs are
 	 * not allocated to other partitions and there are effective_cpus
 	 * left in the top cpuset.
 	 */
-	if (adding && (!capable(CAP_SYS_ADMIN) ||
-		       cpumask_intersects(tmp->addmask, subpartitions_cpus) ||
-		       cpumask_subset(top_cpuset.effective_cpus, tmp->addmask)))
-		goto invalidate;
+	if (adding) {
+		if (!capable(CAP_SYS_ADMIN))
+			cs->prs_err = PERR_ACCESS;
+		else if (cpumask_intersects(tmp->addmask, subpartitions_cpus) ||
+			 cpumask_subset(top_cpuset.effective_cpus, tmp->addmask))
+			cs->prs_err = PERR_NOCPUS;
+		if (cs->prs_err)
+			goto invalidate;
+	}
 
 	spin_lock_irq(&callback_lock);
 	if (adding)
 		isolcpus_updated += partition_xcpus_add(prs, NULL, tmp->addmask);
 	if (deleting)
 		isolcpus_updated += partition_xcpus_del(prs, NULL, tmp->delmask);
+	/*
+	 * Need to update effective_xcpus and exclusive_cpus now as
+	 * update_sibling_cpumasks() below may iterate back to the same cs.
+	 */
+	cpumask_copy(cs->effective_xcpus, excpus);
+	if (xcpus)
+		cpumask_copy(cs->exclusive_cpus, xcpus);
 	spin_unlock_irq(&callback_lock);
 	update_unbound_workqueue_cpumask(isolcpus_updated);
+	if (adding || deleting)
+		cpuset_force_rebuild();
 
 	/*
 	 * Propagate changes in top_cpuset's effective_cpus down the hierarchy.
@@ -1507,47 +1579,6 @@ invalidate:
 }
 
 /*
- * remote_partition_check - check if a child remote partition needs update
- * @cs: the cpuset to be updated
- * @newmask: the new effective_xcpus mask
- * @delmask: temporary mask for deletion (not in tmp)
- * @tmp: temporary masks
- *
- * This should be called before the given cs has updated its cpus_allowed
- * and/or effective_xcpus.
- */
-static void remote_partition_check(struct cpuset *cs, struct cpumask *newmask,
-				   struct cpumask *delmask, struct tmpmasks *tmp)
-{
-	struct cpuset *child, *next;
-	int disable_cnt = 0;
-
-	/*
-	 * Compute the effective exclusive CPUs that will be deleted.
-	 */
-	if (!cpumask_andnot(delmask, cs->effective_xcpus, newmask) ||
-	    !cpumask_intersects(delmask, subpartitions_cpus))
-		return;	/* No deletion of exclusive CPUs in partitions */
-
-	/*
-	 * Searching the remote children list to look for those that will
-	 * be impacted by the deletion of exclusive CPUs.
-	 *
-	 * Since a cpuset must be removed from the remote children list
-	 * before it can go offline and holding cpuset_mutex will prevent
-	 * any change in cpuset status. RCU read lock isn't needed.
-	 */
-	lockdep_assert_held(&cpuset_mutex);
-	list_for_each_entry_safe(child, next, &remote_children, remote_sibling)
-		if (cpumask_intersects(child->effective_cpus, delmask)) {
-			remote_partition_disable(child, tmp);
-			disable_cnt++;
-		}
-	if (disable_cnt)
-		cpuset_force_rebuild();
-}
-
-/*
  * prstate_housekeeping_conflict - check for partition & housekeeping conflicts
  * @prstate: partition root state to be checked
  * @new_cpus: cpu mask
@@ -1601,7 +1632,7 @@ static bool prstate_housekeeping_conflict(int prstate, struct cpumask *new_cpus)
  * The partcmd_update command is used by update_cpumasks_hier() with newmask
  * NULL and update_cpumask() with newmask set. The partcmd_invalidate is used
  * by update_cpumask() with NULL newmask. In both cases, the callers won't
- * check for error and so partition_root_state and prs_error will be updated
+ * check for error and so partition_root_state and prs_err will be updated
  * directly.
  */
 static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
@@ -1614,11 +1645,12 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
 	int old_prs, new_prs;
 	int part_error = PERR_NONE;	/* Partition error? */
 	int subparts_delta = 0;
-	struct cpumask *xcpus;		/* cs effective_xcpus */
 	int isolcpus_updated = 0;
+	struct cpumask *xcpus = user_xcpus(cs);
 	bool nocpu;
 
 	lockdep_assert_held(&cpuset_mutex);
+	WARN_ON_ONCE(is_remote_partition(cs));
 
 	/*
 	 * new_prs will only be changed for the partcmd_update and
@@ -1626,7 +1658,6 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
 	 */
 	adding = deleting = false;
 	old_prs = new_prs = cs->partition_root_state;
-	xcpus = user_xcpus(cs);
 
 	if (cmd == partcmd_invalidate) {
 		if (is_prs_invalid(old_prs))
@@ -1661,12 +1692,19 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
 
 	if ((cmd == partcmd_enable) || (cmd == partcmd_enablei)) {
 		/*
+		 * Need to call compute_effective_exclusive_cpumask() in case
+		 * exclusive_cpus not set. Sibling conflict should only happen
+		 * if exclusive_cpus isn't set.
+		 */
+		xcpus = tmp->new_cpus;
+		if (compute_effective_exclusive_cpumask(cs, xcpus, NULL))
+			WARN_ON_ONCE(!cpumask_empty(cs->exclusive_cpus));
+
+		/*
 		 * Enabling partition root is not allowed if its
-		 * effective_xcpus is empty or doesn't overlap with
-		 * parent's effective_xcpus.
+		 * effective_xcpus is empty.
 		 */
-		if (cpumask_empty(xcpus) ||
-		    !cpumask_intersects(xcpus, parent->effective_xcpus))
+		if (cpumask_empty(xcpus))
 			return PERR_INVCPUS;
 
 		if (prstate_housekeeping_conflict(new_prs, xcpus))
@@ -1679,19 +1717,22 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
 		if (nocpu)
 			return PERR_NOCPUS;
 
-		cpumask_copy(tmp->delmask, xcpus);
-		deleting = true;
-		subparts_delta++;
+		deleting = cpumask_and(tmp->delmask, xcpus, parent->effective_xcpus);
+		if (deleting)
+			subparts_delta++;
 		new_prs = (cmd == partcmd_enable) ? PRS_ROOT : PRS_ISOLATED;
 	} else if (cmd == partcmd_disable) {
 		/*
-		 * May need to add cpus to parent's effective_cpus for
-		 * valid partition root.
+		 * May need to add cpus back to parent's effective_cpus
+		 * (and maybe removed from subpartitions_cpus/isolated_cpus)
+		 * for valid partition root. xcpus may contain CPUs that
+		 * shouldn't be removed from the two global cpumasks.
 		 */
-		adding = !is_prs_invalid(old_prs) &&
-			  cpumask_and(tmp->addmask, xcpus, parent->effective_xcpus);
-		if (adding)
+		if (is_partition_valid(cs)) {
+			cpumask_copy(tmp->addmask, cs->effective_xcpus);
+			adding = true;
 			subparts_delta--;
+		}
 		new_prs = PRS_MEMBER;
 	} else if (newmask) {
 		/*
@@ -1701,6 +1742,7 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
 			part_error = PERR_CPUSEMPTY;
 			goto write_error;
 		}
+
 		/* Check newmask again, whether cpus are available for parent/cs */
 		nocpu |= tasks_nocpu_error(parent, cs, newmask);
 
@@ -1829,7 +1871,7 @@ write_error:
 	 * CPU lists in cs haven't been updated yet. So defer it to later.
 	 */
 	if ((old_prs != new_prs) && (cmd != partcmd_update))  {
-		int err = update_partition_exclusive(cs, new_prs);
+		int err = update_partition_exclusive_flag(cs, new_prs);
 
 		if (err)
 			return err;
@@ -1867,7 +1909,7 @@ write_error:
 	update_unbound_workqueue_cpumask(isolcpus_updated);
 
 	if ((old_prs != new_prs) && (cmd == partcmd_update))
-		update_partition_exclusive(cs, new_prs);
+		update_partition_exclusive_flag(cs, new_prs);
 
 	if (adding || deleting) {
 		cpuset_update_tasks_cpumask(parent, tmp->addmask);
@@ -1917,7 +1959,7 @@ static void compute_partition_effective_cpumask(struct cpuset *cs,
 	 *  2) All the effective_cpus will be used up and cp
 	 *     has tasks
 	 */
-	compute_effective_exclusive_cpumask(cs, new_ecpus);
+	compute_effective_exclusive_cpumask(cs, new_ecpus, NULL);
 	cpumask_and(new_ecpus, new_ecpus, cpu_active_mask);
 
 	rcu_read_lock();
@@ -1925,6 +1967,11 @@ static void compute_partition_effective_cpumask(struct cpuset *cs,
 		if (!is_partition_valid(child))
 			continue;
 
+		/*
+		 * There shouldn't be a remote partition underneath another
+		 * partition root.
+		 */
+		WARN_ON_ONCE(is_remote_partition(child));
 		child->prs_err = 0;
 		if (!cpumask_subset(child->effective_xcpus,
 				    cs->effective_xcpus))
@@ -1980,32 +2027,39 @@ static void update_cpumasks_hier(struct cpuset *cs, struct tmpmasks *tmp,
 		bool remote = is_remote_partition(cp);
 		bool update_parent = false;
 
+		old_prs = new_prs = cp->partition_root_state;
+
 		/*
-		 * Skip descendent remote partition that acquires CPUs
-		 * directly from top cpuset unless it is cs.
+		 * For child remote partition root (!= cs), we need to call
+		 * remote_cpus_update() if effective_xcpus will be changed.
+		 * Otherwise, we can skip the whole subtree.
+		 *
+		 * remote_cpus_update() will reuse tmp->new_cpus only after
+		 * its value is being processed.
 		 */
 		if (remote && (cp != cs)) {
-			pos_css = css_rightmost_descendant(pos_css);
-			continue;
-		}
+			compute_effective_exclusive_cpumask(cp, tmp->new_cpus, NULL);
+			if (cpumask_equal(cp->effective_xcpus, tmp->new_cpus)) {
+				pos_css = css_rightmost_descendant(pos_css);
+				continue;
+			}
+			rcu_read_unlock();
+			remote_cpus_update(cp, NULL, tmp->new_cpus, tmp);
+			rcu_read_lock();
 
-		/*
-		 * Update effective_xcpus if exclusive_cpus set.
-		 * The case when exclusive_cpus isn't set is handled later.
-		 */
-		if (!cpumask_empty(cp->exclusive_cpus) && (cp != cs)) {
-			spin_lock_irq(&callback_lock);
-			compute_effective_exclusive_cpumask(cp, NULL);
-			spin_unlock_irq(&callback_lock);
+			/* Remote partition may be invalidated */
+			new_prs = cp->partition_root_state;
+			remote = (new_prs == old_prs);
 		}
 
-		old_prs = new_prs = cp->partition_root_state;
-		if (remote || (is_partition_valid(parent) &&
-			       is_partition_valid(cp)))
+		if (remote || (is_partition_valid(parent) && is_partition_valid(cp)))
 			compute_partition_effective_cpumask(cp, tmp->new_cpus);
 		else
 			compute_effective_cpumask(tmp->new_cpus, cp, parent);
 
+		if (remote)
+			goto get_css;	/* Ready to update cpuset data */
+
 		/*
 		 * A partition with no effective_cpus is allowed as long as
 		 * there is no task associated with it. Call
@@ -2025,9 +2079,6 @@ static void update_cpumasks_hier(struct cpuset *cs, struct tmpmasks *tmp,
 		if (is_in_v2_mode() && !remote && cpumask_empty(tmp->new_cpus))
 			cpumask_copy(tmp->new_cpus, parent->effective_cpus);
 
-		if (remote)
-			goto get_css;
-
 		/*
 		 * Skip the whole subtree if
 		 * 1) the cpumask remains the same,
@@ -2088,6 +2139,9 @@ get_css:
 		spin_lock_irq(&callback_lock);
 		cpumask_copy(cp->effective_cpus, tmp->new_cpus);
 		cp->partition_root_state = new_prs;
+		if (!cpumask_empty(cp->exclusive_cpus) && (cp != cs))
+			compute_effective_exclusive_cpumask(cp, NULL, NULL);
+
 		/*
 		 * Make sure effective_xcpus is properly set for a valid
 		 * partition root.
@@ -2174,7 +2228,14 @@ static void update_sibling_cpumasks(struct cpuset *parent, struct cpuset *cs,
 						  parent);
 			if (cpumask_equal(tmp->new_cpus, sibling->effective_cpus))
 				continue;
+		} else if (is_remote_partition(sibling)) {
+			/*
+			 * Change in a sibling cpuset won't affect a remote
+			 * partition root.
+			 */
+			continue;
 		}
+
 		if (!css_tryget_online(&sibling->css))
 			continue;
 
@@ -2231,8 +2292,9 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 		 * trialcs->effective_xcpus is used as a temporary cpumask
 		 * for checking validity of the partition root.
 		 */
+		trialcs->partition_root_state = PRS_MEMBER;
 		if (!cpumask_empty(trialcs->exclusive_cpus) || is_partition_valid(cs))
-			compute_effective_exclusive_cpumask(trialcs, NULL);
+			compute_effective_exclusive_cpumask(trialcs, NULL, cs);
 	}
 
 	/* Nothing to do if the cpus didn't change */
@@ -2305,19 +2367,13 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 		 * Call remote_cpus_update() to handle valid remote partition
 		 */
 		if (is_remote_partition(cs))
-			remote_cpus_update(cs, xcpus, &tmp);
+			remote_cpus_update(cs, NULL, xcpus, &tmp);
 		else if (invalidate)
 			update_parent_effective_cpumask(cs, partcmd_invalidate,
 							NULL, &tmp);
 		else
 			update_parent_effective_cpumask(cs, partcmd_update,
 							xcpus, &tmp);
-	} else if (!cpumask_empty(cs->exclusive_cpus)) {
-		/*
-		 * Use trialcs->effective_cpus as a temp cpumask
-		 */
-		remote_partition_check(cs, trialcs->effective_xcpus,
-				       trialcs->effective_cpus, &tmp);
 	}
 
 	spin_lock_irq(&callback_lock);
@@ -2369,8 +2425,15 @@ static int update_exclusive_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 	if (cpumask_equal(cs->exclusive_cpus, trialcs->exclusive_cpus))
 		return 0;
 
-	if (*buf)
-		compute_effective_exclusive_cpumask(trialcs, NULL);
+	if (*buf) {
+		trialcs->partition_root_state = PRS_MEMBER;
+		/*
+		 * Reject the change if there is exclusive CPUs conflict with
+		 * the siblings.
+		 */
+		if (compute_effective_exclusive_cpumask(trialcs, NULL, cs))
+			return -EINVAL;
+	}
 
 	/*
 	 * Check all the descendants in update_cpumasks_hier() if
@@ -2401,8 +2464,8 @@ static int update_exclusive_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 			if (invalidate)
 				remote_partition_disable(cs, &tmp);
 			else
-				remote_cpus_update(cs, trialcs->effective_xcpus,
-						   &tmp);
+				remote_cpus_update(cs, trialcs->exclusive_cpus,
+						   trialcs->effective_xcpus, &tmp);
 		} else if (invalidate) {
 			update_parent_effective_cpumask(cs, partcmd_invalidate,
 							NULL, &tmp);
@@ -2410,12 +2473,6 @@ static int update_exclusive_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 			update_parent_effective_cpumask(cs, partcmd_update,
 						trialcs->effective_xcpus, &tmp);
 		}
-	} else if (!cpumask_empty(trialcs->exclusive_cpus)) {
-		/*
-		 * Use trialcs->effective_cpus as a temp cpumask
-		 */
-		remote_partition_check(cs, trialcs->effective_xcpus,
-				       trialcs->effective_cpus, &tmp);
 	}
 	spin_lock_irq(&callback_lock);
 	cpumask_copy(cs->exclusive_cpus, trialcs->exclusive_cpus);
@@ -2782,7 +2839,7 @@ static int update_prstate(struct cpuset *cs, int new_prs)
 	int err = PERR_NONE, old_prs = cs->partition_root_state;
 	struct cpuset *parent = parent_cs(cs);
 	struct tmpmasks tmpmask;
-	bool new_xcpus_state = false;
+	bool isolcpus_updated = false;
 
 	if (old_prs == new_prs)
 		return 0;
@@ -2796,18 +2853,7 @@ static int update_prstate(struct cpuset *cs, int new_prs)
 	if (alloc_cpumasks(NULL, &tmpmask))
 		return -ENOMEM;
 
-	/*
-	 * Setup effective_xcpus if not properly set yet, it will be cleared
-	 * later if partition becomes invalid.
-	 */
-	if ((new_prs > 0) && cpumask_empty(cs->exclusive_cpus)) {
-		spin_lock_irq(&callback_lock);
-		cpumask_and(cs->effective_xcpus,
-			    cs->cpus_allowed, parent->effective_xcpus);
-		spin_unlock_irq(&callback_lock);
-	}
-
-	err = update_partition_exclusive(cs, new_prs);
+	err = update_partition_exclusive_flag(cs, new_prs);
 	if (err)
 		goto out;
 
@@ -2821,6 +2867,19 @@ static int update_prstate(struct cpuset *cs, int new_prs)
 		}
 
 		/*
+		 * We don't support the creation of a new local partition with
+		 * a remote partition underneath it. This unsupported
+		 * setting can happen only if parent is the top_cpuset because
+		 * a remote partition cannot be created underneath an existing
+		 * local or remote partition.
+		 */
+		if ((parent == &top_cpuset) &&
+		    cpumask_intersects(cs->exclusive_cpus, subpartitions_cpus)) {
+			err = PERR_REMOTE;
+			goto out;
+		}
+
+		/*
 		 * If parent is valid partition, enable local partiion.
 		 * Otherwise, enable a remote partition.
 		 */
@@ -2835,8 +2894,9 @@ static int update_prstate(struct cpuset *cs, int new_prs)
 	} else if (old_prs && new_prs) {
 		/*
 		 * A change in load balance state only, no change in cpumasks.
+		 * Need to update isolated_cpus.
 		 */
-		new_xcpus_state = true;
+		isolcpus_updated = true;
 	} else {
 		/*
 		 * Switching back to member is always allowed even if it
@@ -2860,7 +2920,7 @@ out:
 	 */
 	if (err) {
 		new_prs = -new_prs;
-		update_partition_exclusive(cs, new_prs);
+		update_partition_exclusive_flag(cs, new_prs);
 	}
 
 	spin_lock_irq(&callback_lock);
@@ -2868,14 +2928,18 @@ out:
 	WRITE_ONCE(cs->prs_err, err);
 	if (!is_partition_valid(cs))
 		reset_partition_data(cs);
-	else if (new_xcpus_state)
-		partition_xcpus_newstate(old_prs, new_prs, cs->effective_xcpus);
+	else if (isolcpus_updated)
+		isolated_cpus_update(old_prs, new_prs, cs->effective_xcpus);
 	spin_unlock_irq(&callback_lock);
-	update_unbound_workqueue_cpumask(new_xcpus_state);
+	update_unbound_workqueue_cpumask(isolcpus_updated);
 
-	/* Force update if switching back to member */
+	/* Force update if switching back to member & update effective_xcpus */
 	update_cpumasks_hier(cs, &tmpmask, !new_prs);
 
+	/* A newly created partition must have effective_xcpus set */
+	WARN_ON_ONCE(!old_prs && (new_prs > 0)
+			      && cpumask_empty(cs->effective_xcpus));
+
 	/* Update sched domains and load balance flag */
 	update_partition_sd_lb(cs, old_prs);
 
@@ -3208,7 +3272,7 @@ int cpuset_common_seq_show(struct seq_file *sf, void *v)
 	return ret;
 }
 
-static int sched_partition_show(struct seq_file *seq, void *v)
+static int cpuset_partition_show(struct seq_file *seq, void *v)
 {
 	struct cpuset *cs = css_cs(seq_css(seq));
 	const char *err, *type = NULL;
@@ -3239,7 +3303,7 @@ static int sched_partition_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static ssize_t sched_partition_write(struct kernfs_open_file *of, char *buf,
+static ssize_t cpuset_partition_write(struct kernfs_open_file *of, char *buf,
 				     size_t nbytes, loff_t off)
 {
 	struct cpuset *cs = css_cs(of_css(of));
@@ -3260,11 +3324,8 @@ static ssize_t sched_partition_write(struct kernfs_open_file *of, char *buf,
 	css_get(&cs->css);
 	cpus_read_lock();
 	mutex_lock(&cpuset_mutex);
-	if (!is_cpuset_online(cs))
-		goto out_unlock;
-
-	retval = update_prstate(cs, val);
-out_unlock:
+	if (is_cpuset_online(cs))
+		retval = update_prstate(cs, val);
 	mutex_unlock(&cpuset_mutex);
 	cpus_read_unlock();
 	css_put(&cs->css);
@@ -3308,8 +3369,8 @@ static struct cftype dfl_files[] = {
 
 	{
 		.name = "cpus.partition",
-		.seq_show = sched_partition_show,
-		.write = sched_partition_write,
+		.seq_show = cpuset_partition_show,
+		.write = cpuset_partition_write,
 		.private = FILE_PARTITION_ROOT,
 		.flags = CFTYPE_NOT_ON_ROOT,
 		.file_offset = offsetof(struct cpuset, partition_file),
@@ -3475,9 +3536,6 @@ static void cpuset_css_offline(struct cgroup_subsys_state *css)
 	cpus_read_lock();
 	mutex_lock(&cpuset_mutex);
 
-	if (is_partition_valid(cs))
-		update_prstate(cs, 0);
-
 	if (!cpuset_v2() && is_sched_load_balance(cs))
 		cpuset_update_flag(CS_SCHED_LOAD_BALANCE, cs, 0);
 
@@ -3488,6 +3546,22 @@ static void cpuset_css_offline(struct cgroup_subsys_state *css)
 	cpus_read_unlock();
 }
 
+static void cpuset_css_killed(struct cgroup_subsys_state *css)
+{
+	struct cpuset *cs = css_cs(css);
+
+	cpus_read_lock();
+	mutex_lock(&cpuset_mutex);
+
+	/* Reset valid partition back to member */
+	if (is_partition_valid(cs))
+		update_prstate(cs, PRS_MEMBER);
+
+	mutex_unlock(&cpuset_mutex);
+	cpus_read_unlock();
+
+}
+
 static void cpuset_css_free(struct cgroup_subsys_state *css)
 {
 	struct cpuset *cs = css_cs(css);
@@ -3609,6 +3683,7 @@ struct cgroup_subsys cpuset_cgrp_subsys = {
 	.css_alloc	= cpuset_css_alloc,
 	.css_online	= cpuset_css_online,
 	.css_offline	= cpuset_css_offline,
+	.css_killed	= cpuset_css_killed,
 	.css_free	= cpuset_css_free,
 	.can_attach	= cpuset_can_attach,
 	.cancel_attach	= cpuset_cancel_attach,
@@ -3739,10 +3814,10 @@ retry:
 
 	if (remote && cpumask_empty(&new_cpus) &&
 	    partition_is_populated(cs, NULL)) {
+		cs->prs_err = PERR_HOTPLUG;
 		remote_partition_disable(cs, tmp);
 		compute_effective_cpumask(&new_cpus, cs, parent);
 		remote = false;
-		cpuset_force_rebuild();
 	}
 
 	/*
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
index 4bb587d5d34f..b2239156b7de 100644
--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -318,10 +318,11 @@ __bpf_kfunc void cgroup_rstat_flush(struct cgroup *cgrp)
 
 	might_sleep();
 	for_each_possible_cpu(cpu) {
-		struct cgroup *pos = cgroup_rstat_updated_list(cgrp, cpu);
+		struct cgroup *pos;
 
 		/* Reacquire for each CPU to avoid disabling IRQs too long */
 		__cgroup_rstat_lock(cgrp, cpu);
+		pos = cgroup_rstat_updated_list(cgrp, cpu);
 		for (; pos; pos = pos->rstat_flush_next) {
 			struct cgroup_subsys_state *css;
 
diff --git a/kernel/dma/coherent.c b/kernel/dma/coherent.c
index 3b2bdca9f1d4..77c8d9487a9a 100644
--- a/kernel/dma/coherent.c
+++ b/kernel/dma/coherent.c
@@ -336,16 +336,22 @@ static phys_addr_t dma_reserved_default_memory_size __initdata;
 
 static int rmem_dma_device_init(struct reserved_mem *rmem, struct device *dev)
 {
-	if (!rmem->priv) {
-		struct dma_coherent_mem *mem;
+	struct dma_coherent_mem *mem = rmem->priv;
 
+	if (!mem) {
 		mem = dma_init_coherent_memory(rmem->base, rmem->base,
 					       rmem->size, true);
 		if (IS_ERR(mem))
 			return PTR_ERR(mem);
 		rmem->priv = mem;
 	}
-	dma_assign_coherent_memory(dev, rmem->priv);
+
+	/* Warn if the device potentially can't use the reserved memory */
+	if (mem->device_base + rmem->size - 1 >
+	    min_not_zero(dev->coherent_dma_mask, dev->bus_dma_limit))
+		dev_warn(dev, "reserved memory is beyond device's set DMA address range\n");
+
+	dma_assign_coherent_memory(dev, mem);
 	return 0;
 }
 
diff --git a/kernel/dma/contiguous.c b/kernel/dma/contiguous.c
index 055da410ac71..8df0dfaaca18 100644
--- a/kernel/dma/contiguous.c
+++ b/kernel/dma/contiguous.c
@@ -64,8 +64,7 @@ struct cma *dma_contiguous_default_area;
  * Users, who want to set the size of global CMA area for their system
  * should use cma= kernel parameter.
  */
-static const phys_addr_t size_bytes __initconst =
-	(phys_addr_t)CMA_SIZE_MBYTES * SZ_1M;
+#define size_bytes ((phys_addr_t)CMA_SIZE_MBYTES * SZ_1M)
 static phys_addr_t  size_cmdline __initdata = -1;
 static phys_addr_t base_cmdline __initdata;
 static phys_addr_t limit_cmdline __initdata;
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index cda127027e48..051a32988040 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -910,6 +910,19 @@ int dma_set_coherent_mask(struct device *dev, u64 mask)
 }
 EXPORT_SYMBOL(dma_set_coherent_mask);
 
+static bool __dma_addressing_limited(struct device *dev)
+{
+	const struct dma_map_ops *ops = get_dma_ops(dev);
+
+	if (min_not_zero(dma_get_mask(dev), dev->bus_dma_limit) <
+			 dma_get_required_mask(dev))
+		return true;
+
+	if (unlikely(ops) || use_dma_iommu(dev))
+		return false;
+	return !dma_direct_all_ram_mapped(dev);
+}
+
 /**
  * dma_addressing_limited - return if the device is addressing limited
  * @dev:	device to check
@@ -920,15 +933,11 @@ EXPORT_SYMBOL(dma_set_coherent_mask);
  */
 bool dma_addressing_limited(struct device *dev)
 {
-	const struct dma_map_ops *ops = get_dma_ops(dev);
-
-	if (min_not_zero(dma_get_mask(dev), dev->bus_dma_limit) <
-			 dma_get_required_mask(dev))
-		return true;
-
-	if (unlikely(ops) || use_dma_iommu(dev))
+	if (!__dma_addressing_limited(dev))
 		return false;
-	return !dma_direct_all_ram_mapped(dev);
+
+	dev_dbg(dev, "device is DMA addressing limited\n");
+	return true;
 }
 EXPORT_SYMBOL_GPL(dma_addressing_limited);
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 128db74e9eab..95e703891b24 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3943,7 +3943,7 @@ static int merge_sched_in(struct perf_event *event, void *data)
 			perf_event_set_state(event, PERF_EVENT_STATE_ERROR);
 
 			if (*perf_event_fasync(event))
-				event->pending_kill = POLL_HUP;
+				event->pending_kill = POLL_ERR;
 
 			perf_event_wakeup(event);
 		} else {
@@ -5518,30 +5518,6 @@ static bool exclusive_event_installable(struct perf_event *event,
 
 static void perf_free_addr_filters(struct perf_event *event);
 
-static void perf_pending_task_sync(struct perf_event *event)
-{
-	struct callback_head *head = &event->pending_task;
-
-	if (!event->pending_work)
-		return;
-	/*
-	 * If the task is queued to the current task's queue, we
-	 * obviously can't wait for it to complete. Simply cancel it.
-	 */
-	if (task_work_cancel(current, head)) {
-		event->pending_work = 0;
-		local_dec(&event->ctx->nr_no_switch_fast);
-		return;
-	}
-
-	/*
-	 * All accesses related to the event are within the same RCU section in
-	 * perf_pending_task(). The RCU grace period before the event is freed
-	 * will make sure all those accesses are complete by then.
-	 */
-	rcuwait_wait_event(&event->pending_work_wait, !event->pending_work, TASK_UNINTERRUPTIBLE);
-}
-
 /* vs perf_event_alloc() error */
 static void __free_event(struct perf_event *event)
 {
@@ -5599,7 +5575,6 @@ static void _free_event(struct perf_event *event)
 {
 	irq_work_sync(&event->pending_irq);
 	irq_work_sync(&event->pending_disable_irq);
-	perf_pending_task_sync(event);
 
 	unaccount_event(event);
 
@@ -5692,10 +5667,17 @@ static void perf_remove_from_owner(struct perf_event *event)
 
 static void put_event(struct perf_event *event)
 {
+	struct perf_event *parent;
+
 	if (!atomic_long_dec_and_test(&event->refcount))
 		return;
 
+	parent = event->parent;
 	_free_event(event);
+
+	/* Matches the refcount bump in inherit_event() */
+	if (parent)
+		put_event(parent);
 }
 
 /*
@@ -5779,11 +5761,6 @@ again:
 		if (tmp == child) {
 			perf_remove_from_context(child, DETACH_GROUP);
 			list_move(&child->child_list, &free_list);
-			/*
-			 * This matches the refcount bump in inherit_event();
-			 * this can't be the last reference.
-			 */
-			put_event(event);
 		} else {
 			var = &ctx->refcount;
 		}
@@ -5809,7 +5786,8 @@ again:
 		void *var = &child->ctx->refcount;
 
 		list_del(&child->child_list);
-		free_event(child);
+		/* Last reference unless ->pending_task work is pending */
+		put_event(child);
 
 		/*
 		 * Wake any perf_event_free_task() waiting for this event to be
@@ -5820,7 +5798,11 @@ again:
 	}
 
 no_ctx:
-	put_event(event); /* Must be the 'last' reference */
+	/*
+	 * Last reference unless ->pending_task work is pending on this event
+	 * or any of its children.
+	 */
+	put_event(event);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(perf_event_release_kernel);
@@ -6093,7 +6075,7 @@ static __poll_t perf_poll(struct file *file, poll_table *wait)
 
 	if (unlikely(READ_ONCE(event->state) == PERF_EVENT_STATE_ERROR &&
 		     event->attr.pinned))
-		return events;
+		return EPOLLERR;
 
 	/*
 	 * Pin the event->rb by taking event->mmap_mutex; otherwise
@@ -7236,12 +7218,6 @@ static void perf_pending_task(struct callback_head *head)
 	int rctx;
 
 	/*
-	 * All accesses to the event must belong to the same implicit RCU read-side
-	 * critical section as the ->pending_work reset. See comment in
-	 * perf_pending_task_sync().
-	 */
-	rcu_read_lock();
-	/*
 	 * If we 'fail' here, that's OK, it means recursion is already disabled
 	 * and we won't recurse 'further'.
 	 */
@@ -7251,9 +7227,8 @@ static void perf_pending_task(struct callback_head *head)
 		event->pending_work = 0;
 		perf_sigtrap(event);
 		local_dec(&event->ctx->nr_no_switch_fast);
-		rcuwait_wake_up(&event->pending_work_wait);
 	}
-	rcu_read_unlock();
+	put_event(event);
 
 	if (rctx >= 0)
 		perf_swevent_put_recursion_context(rctx);
@@ -10248,6 +10223,7 @@ static int __perf_event_overflow(struct perf_event *event,
 		    !task_work_add(current, &event->pending_task, notify_mode)) {
 			event->pending_work = pending_id;
 			local_inc(&event->ctx->nr_no_switch_fast);
+			WARN_ON_ONCE(!atomic_long_inc_not_zero(&event->refcount));
 
 			event->pending_addr = 0;
 			if (valid_sample && (data->sample_flags & PERF_SAMPLE_ADDR))
@@ -12610,7 +12586,6 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 	init_irq_work(&event->pending_irq, perf_pending_irq);
 	event->pending_disable_irq = IRQ_WORK_INIT_HARD(perf_pending_disable);
 	init_task_work(&event->pending_task, perf_pending_task);
-	rcuwait_init(&event->pending_work_wait);
 
 	mutex_init(&event->mmap_mutex);
 	raw_spin_lock_init(&event->addr_filters.lock);
@@ -13747,8 +13722,7 @@ perf_event_exit_event(struct perf_event *event, struct perf_event_context *ctx)
 		 * Kick perf_poll() for is_event_hup();
 		 */
 		perf_event_wakeup(parent_event);
-		free_event(event);
-		put_event(parent_event);
+		put_event(event);
 		return;
 	}
 
@@ -13872,13 +13846,11 @@ static void perf_free_event(struct perf_event *event,
 	list_del_init(&event->child_list);
 	mutex_unlock(&parent->child_mutex);
 
-	put_event(parent);
-
 	raw_spin_lock_irq(&ctx->lock);
 	perf_group_detach(event);
 	list_del_event(event, ctx);
 	raw_spin_unlock_irq(&ctx->lock);
-	free_event(event);
+	put_event(event);
 }
 
 /*
@@ -14016,6 +13988,9 @@ inherit_event(struct perf_event *parent_event,
 	if (IS_ERR(child_event))
 		return child_event;
 
+	get_ctx(child_ctx);
+	child_event->ctx = child_ctx;
+
 	pmu_ctx = find_get_pmu_context(child_event->pmu, child_ctx, child_event);
 	if (IS_ERR(pmu_ctx)) {
 		free_event(child_event);
@@ -14037,8 +14012,6 @@ inherit_event(struct perf_event *parent_event,
 		return NULL;
 	}
 
-	get_ctx(child_ctx);
-
 	/*
 	 * Make the child state follow the state of the parent event,
 	 * not its attr.disabled bit.  We hold the parent's mutex,
@@ -14059,7 +14032,6 @@ inherit_event(struct perf_event *parent_event,
 		local64_set(&hwc->period_left, sample_period);
 	}
 
-	child_event->ctx = child_ctx;
 	child_event->overflow_handler = parent_event->overflow_handler;
 	child_event->overflow_handler_context
 		= parent_event->overflow_handler_context;
diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 615b4e6d22c7..8d783b5882b6 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -1956,6 +1956,9 @@ static void free_ret_instance(struct uprobe_task *utask,
 	 * to-be-reused return instances for future uretprobes. If ri_timer()
 	 * happens to be running right now, though, we fallback to safety and
 	 * just perform RCU-delated freeing of ri.
+	 * Admittedly, this is a rather simple use of seqcount, but it nicely
+	 * abstracts away all the necessary memory barriers, so we use
+	 * a well-supported kernel primitive here.
 	 */
 	if (raw_seqcount_try_begin(&utask->ri_seqcount, seq)) {
 		/* immediate reuse of ri without RCU GP is OK */
@@ -2016,12 +2019,20 @@ static void ri_timer(struct timer_list *timer)
 	/* RCU protects return_instance from freeing. */
 	guard(rcu)();
 
-	write_seqcount_begin(&utask->ri_seqcount);
+	/*
+	 * See free_ret_instance() for notes on seqcount use.
+	 * We also employ raw API variants to avoid lockdep false-positive
+	 * warning complaining about enabled preemption. The timer can only be
+	 * invoked once for a uprobe_task. Therefore there can only be one
+	 * writer. The reader does not require an even sequence count to make
+	 * progress, so it is OK to remain preemptible on PREEMPT_RT.
+	 */
+	raw_write_seqcount_begin(&utask->ri_seqcount);
 
 	for_each_ret_instance_rcu(ri, utask->return_instances)
 		hprobe_expire(&ri->hprobe, false);
 
-	write_seqcount_end(&utask->ri_seqcount);
+	raw_write_seqcount_end(&utask->ri_seqcount);
 }
 
 static struct uprobe_task *alloc_utask(void)
diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c
index 5c8d43cdb0a3..c05ba7ca00fa 100644
--- a/kernel/irq/msi.c
+++ b/kernel/irq/msi.c
@@ -761,7 +761,7 @@ static int msi_domain_translate(struct irq_domain *domain, struct irq_fwspec *fw
 static void msi_domain_debug_show(struct seq_file *m, struct irq_domain *d,
 				  struct irq_data *irqd, int ind)
 {
-	struct msi_desc *desc = irq_data_get_msi_desc(irqd);
+	struct msi_desc *desc = irqd ? irq_data_get_msi_desc(irqd) : NULL;
 
 	if (!desc)
 		return;
diff --git a/kernel/params.c b/kernel/params.c
index 2509f216c9f3..e668fc90b83e 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -760,38 +760,35 @@ void destroy_params(const struct kernel_param *params, unsigned num)
 			params[i].ops->free(params[i].arg);
 }
 
-static struct module_kobject * __init locate_module_kobject(const char *name)
+struct module_kobject __modinit * lookup_or_create_module_kobject(const char *name)
 {
 	struct module_kobject *mk;
 	struct kobject *kobj;
 	int err;
 
 	kobj = kset_find_obj(module_kset, name);
-	if (kobj) {
-		mk = to_module_kobject(kobj);
-	} else {
-		mk = kzalloc(sizeof(struct module_kobject), GFP_KERNEL);
-		BUG_ON(!mk);
-
-		mk->mod = THIS_MODULE;
-		mk->kobj.kset = module_kset;
-		err = kobject_init_and_add(&mk->kobj, &module_ktype, NULL,
-					   "%s", name);
-#ifdef CONFIG_MODULES
-		if (!err)
-			err = sysfs_create_file(&mk->kobj, &module_uevent.attr);
-#endif
-		if (err) {
-			kobject_put(&mk->kobj);
-			pr_crit("Adding module '%s' to sysfs failed (%d), the system may be unstable.\n",
-				name, err);
-			return NULL;
-		}
+	if (kobj)
+		return to_module_kobject(kobj);
 
-		/* So that we hold reference in both cases. */
-		kobject_get(&mk->kobj);
+	mk = kzalloc(sizeof(struct module_kobject), GFP_KERNEL);
+	if (!mk)
+		return NULL;
+
+	mk->mod = THIS_MODULE;
+	mk->kobj.kset = module_kset;
+	err = kobject_init_and_add(&mk->kobj, &module_ktype, NULL, "%s", name);
+	if (IS_ENABLED(CONFIG_MODULES) && !err)
+		err = sysfs_create_file(&mk->kobj, &module_uevent.attr);
+	if (err) {
+		kobject_put(&mk->kobj);
+		pr_crit("Adding module '%s' to sysfs failed (%d), the system may be unstable.\n",
+			name, err);
+		return NULL;
 	}
 
+	/* So that we hold reference in both cases. */
+	kobject_get(&mk->kobj);
+
 	return mk;
 }
 
@@ -802,7 +799,7 @@ static void __init kernel_add_sysfs_param(const char *name,
 	struct module_kobject *mk;
 	int err;
 
-	mk = locate_module_kobject(name);
+	mk = lookup_or_create_module_kobject(name);
 	if (!mk)
 		return;
 
@@ -873,7 +870,7 @@ static void __init version_sysfs_builtin(void)
 	int err;
 
 	for (vattr = __start___modver; vattr < __stop___modver; vattr++) {
-		mk = locate_module_kobject(vattr->module_name);
+		mk = lookup_or_create_module_kobject(vattr->module_name);
 		if (mk) {
 			err = sysfs_create_file(&mk->kobj, &vattr->mattr.attr);
 			WARN_ON_ONCE(err);
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 1a19d69b91ed..816f07f9d30f 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -81,9 +81,23 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
 	if (!cpufreq_this_cpu_can_update(sg_policy->policy))
 		return false;
 
-	if (unlikely(sg_policy->limits_changed)) {
-		sg_policy->limits_changed = false;
-		sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS);
+	if (unlikely(READ_ONCE(sg_policy->limits_changed))) {
+		WRITE_ONCE(sg_policy->limits_changed, false);
+		sg_policy->need_freq_update = true;
+
+		/*
+		 * The above limits_changed update must occur before the reads
+		 * of policy limits in cpufreq_driver_resolve_freq() or a policy
+		 * limits update might be missed, so use a memory barrier to
+		 * ensure it.
+		 *
+		 * This pairs with the write memory barrier in sugov_limits().
+		 */
+		smp_mb();
+
+		return true;
+	} else if (sg_policy->need_freq_update) {
+		/* ignore_dl_rate_limit() wants a new frequency to be found. */
 		return true;
 	}
 
@@ -95,10 +109,22 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
 static bool sugov_update_next_freq(struct sugov_policy *sg_policy, u64 time,
 				   unsigned int next_freq)
 {
-	if (sg_policy->need_freq_update)
+	if (sg_policy->need_freq_update) {
 		sg_policy->need_freq_update = false;
-	else if (sg_policy->next_freq == next_freq)
+		/*
+		 * The policy limits have changed, but if the return value of
+		 * cpufreq_driver_resolve_freq() after applying the new limits
+		 * is still equal to the previously selected frequency, the
+		 * driver callback need not be invoked unless the driver
+		 * specifically wants that to happen on every update of the
+		 * policy limits.
+		 */
+		if (sg_policy->next_freq == next_freq &&
+		    !cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS))
+			return false;
+	} else if (sg_policy->next_freq == next_freq) {
 		return false;
+	}
 
 	sg_policy->next_freq = next_freq;
 	sg_policy->last_freq_update_time = time;
@@ -365,7 +391,7 @@ static inline bool sugov_hold_freq(struct sugov_cpu *sg_cpu) { return false; }
 static inline void ignore_dl_rate_limit(struct sugov_cpu *sg_cpu)
 {
 	if (cpu_bw_dl(cpu_rq(sg_cpu->cpu)) > sg_cpu->bw_min)
-		sg_cpu->sg_policy->limits_changed = true;
+		sg_cpu->sg_policy->need_freq_update = true;
 }
 
 static inline bool sugov_update_single_common(struct sugov_cpu *sg_cpu,
@@ -871,7 +897,16 @@ static void sugov_limits(struct cpufreq_policy *policy)
 		mutex_unlock(&sg_policy->work_lock);
 	}
 
-	sg_policy->limits_changed = true;
+	/*
+	 * The limits_changed update below must take place before the updates
+	 * of policy limits in cpufreq_set_policy() or a policy limits update
+	 * might be missed, so use a memory barrier to ensure it.
+	 *
+	 * This pairs with the memory barrier in sugov_should_update_freq().
+	 */
+	smp_wmb();
+
+	WRITE_ONCE(sg_policy->limits_changed, true);
 }
 
 struct cpufreq_governor schedutil_gov = {
diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index 66bcd40a28ca..fdbf249d1c68 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -163,7 +163,7 @@ enum scx_ops_flags {
 	/*
 	 * CPU cgroup support flags
 	 */
-	SCX_OPS_HAS_CGROUP_WEIGHT = 1LLU << 16,	/* cpu.weight */
+	SCX_OPS_HAS_CGROUP_WEIGHT = 1LLU << 16,	/* DEPRECATED, will be removed on 6.18 */
 
 	SCX_OPS_ALL_FLAGS	= SCX_OPS_KEEP_BUILTIN_IDLE |
 				  SCX_OPS_ENQ_LAST |
@@ -3899,35 +3899,6 @@ bool scx_can_stop_tick(struct rq *rq)
 
 DEFINE_STATIC_PERCPU_RWSEM(scx_cgroup_rwsem);
 static bool scx_cgroup_enabled;
-static bool cgroup_warned_missing_weight;
-static bool cgroup_warned_missing_idle;
-
-static void scx_cgroup_warn_missing_weight(struct task_group *tg)
-{
-	if (scx_ops_enable_state() == SCX_OPS_DISABLED ||
-	    cgroup_warned_missing_weight)
-		return;
-
-	if ((scx_ops.flags & SCX_OPS_HAS_CGROUP_WEIGHT) || !tg->css.parent)
-		return;
-
-	pr_warn("sched_ext: \"%s\" does not implement cgroup cpu.weight\n",
-		scx_ops.name);
-	cgroup_warned_missing_weight = true;
-}
-
-static void scx_cgroup_warn_missing_idle(struct task_group *tg)
-{
-	if (!scx_cgroup_enabled || cgroup_warned_missing_idle)
-		return;
-
-	if (!tg->idle)
-		return;
-
-	pr_warn("sched_ext: \"%s\" does not implement cgroup cpu.idle\n",
-		scx_ops.name);
-	cgroup_warned_missing_idle = true;
-}
 
 int scx_tg_online(struct task_group *tg)
 {
@@ -3937,8 +3908,6 @@ int scx_tg_online(struct task_group *tg)
 
 	percpu_down_read(&scx_cgroup_rwsem);
 
-	scx_cgroup_warn_missing_weight(tg);
-
 	if (scx_cgroup_enabled) {
 		if (SCX_HAS_OP(cgroup_init)) {
 			struct scx_cgroup_init_args args =
@@ -4076,9 +4045,7 @@ void scx_group_set_weight(struct task_group *tg, unsigned long weight)
 
 void scx_group_set_idle(struct task_group *tg, bool idle)
 {
-	percpu_down_read(&scx_cgroup_rwsem);
-	scx_cgroup_warn_missing_idle(tg);
-	percpu_up_read(&scx_cgroup_rwsem);
+	/* TODO: Implement ops->cgroup_set_idle() */
 }
 
 static void scx_cgroup_lock(void)
@@ -4272,9 +4239,6 @@ static int scx_cgroup_init(void)
 
 	percpu_rwsem_assert_held(&scx_cgroup_rwsem);
 
-	cgroup_warned_missing_weight = false;
-	cgroup_warned_missing_idle = false;
-
 	/*
 	 * scx_tg_on/offline() are excluded through scx_cgroup_rwsem. If we walk
 	 * cgroups and init, all online cgroups are initialized.
@@ -4284,9 +4248,6 @@ static int scx_cgroup_init(void)
 		struct task_group *tg = css_tg(css);
 		struct scx_cgroup_init_args args = { .weight = tg->scx_weight };
 
-		scx_cgroup_warn_missing_weight(tg);
-		scx_cgroup_warn_missing_idle(tg);
-
 		if ((tg->scx_flags &
 		     (SCX_TG_ONLINE | SCX_TG_INITED)) != SCX_TG_ONLINE)
 			continue;
@@ -4623,7 +4584,7 @@ unlock:
 
 static void free_exit_info(struct scx_exit_info *ei)
 {
-	kfree(ei->dump);
+	kvfree(ei->dump);
 	kfree(ei->msg);
 	kfree(ei->bt);
 	kfree(ei);
@@ -4639,7 +4600,7 @@ static struct scx_exit_info *alloc_exit_info(size_t exit_dump_len)
 
 	ei->bt = kcalloc(SCX_EXIT_BT_LEN, sizeof(ei->bt[0]), GFP_KERNEL);
 	ei->msg = kzalloc(SCX_EXIT_MSG_LEN, GFP_KERNEL);
-	ei->dump = kzalloc(exit_dump_len, GFP_KERNEL);
+	ei->dump = kvzalloc(exit_dump_len, GFP_KERNEL);
 
 	if (!ei->bt || !ei->msg || !ei->dump) {
 		free_exit_info(ei);
@@ -5252,6 +5213,9 @@ static int validate_ops(const struct sched_ext_ops *ops)
 		return -EINVAL;
 	}
 
+	if (ops->flags & SCX_OPS_HAS_CGROUP_WEIGHT)
+		pr_warn("SCX_OPS_HAS_CGROUP_WEIGHT is deprecated and a noop\n");
+
 	return 0;
 }
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e43993a4e580..0fb9bf995a47 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -7081,9 +7081,6 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags)
 		h_nr_idle = task_has_idle_policy(p);
 		if (task_sleep || task_delayed || !se->sched_delayed)
 			h_nr_runnable = 1;
-	} else {
-		cfs_rq = group_cfs_rq(se);
-		slice = cfs_rq_min_slice(cfs_rq);
 	}
 
 	for_each_sched_entity(se) {
@@ -7093,6 +7090,7 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags)
 			if (p && &p->se == se)
 				return -1;
 
+			slice = cfs_rq_min_slice(cfs_rq);
 			break;
 		}
 
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 517ee2590a29..30899a8cc52c 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -366,7 +366,7 @@ static const struct debug_obj_descr hrtimer_debug_descr;
 
 static void *hrtimer_debug_hint(void *addr)
 {
-	return ((struct hrtimer *) addr)->function;
+	return ACCESS_PRIVATE((struct hrtimer *)addr, function);
 }
 
 /*
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index a47bcf71defc..9a3859443c04 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -509,6 +509,7 @@ void tick_resume(void)
 
 #ifdef CONFIG_SUSPEND
 static DEFINE_RAW_SPINLOCK(tick_freeze_lock);
+static DEFINE_WAIT_OVERRIDE_MAP(tick_freeze_map, LD_WAIT_SLEEP);
 static unsigned int tick_freeze_depth;
 
 /**
@@ -528,9 +529,22 @@ void tick_freeze(void)
 	if (tick_freeze_depth == num_online_cpus()) {
 		trace_suspend_resume(TPS("timekeeping_freeze"),
 				     smp_processor_id(), true);
+		/*
+		 * All other CPUs have their interrupts disabled and are
+		 * suspended to idle. Other tasks have been frozen so there
+		 * is no scheduling happening. This means that there is no
+		 * concurrency in the system at this point. Therefore it is
+		 * okay to acquire a sleeping lock on PREEMPT_RT, such as a
+		 * spinlock, because the lock cannot be held by other CPUs
+		 * or threads and acquiring it cannot block.
+		 *
+		 * Inform lockdep about the situation.
+		 */
+		lock_map_acquire_try(&tick_freeze_map);
 		system_state = SYSTEM_SUSPEND;
 		sched_clock_suspend();
 		timekeeping_suspend();
+		lock_map_release(&tick_freeze_map);
 	} else {
 		tick_suspend_local();
 	}
@@ -552,8 +566,16 @@ void tick_unfreeze(void)
 	raw_spin_lock(&tick_freeze_lock);
 
 	if (tick_freeze_depth == num_online_cpus()) {
+		/*
+		 * Similar to tick_freeze(). On resumption the first CPU may
+		 * acquire uncontended sleeping locks while other CPUs block on
+		 * tick_freeze_lock.
+		 */
+		lock_map_acquire_try(&tick_freeze_map);
 		timekeeping_resume();
 		sched_clock_resume();
+		lock_map_release(&tick_freeze_map);
+
 		system_state = SYSTEM_RUNNING;
 		trace_suspend_resume(TPS("timekeeping_freeze"),
 				     smp_processor_id(), false);
diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c
index 33082c4e8154..95c6e3473a76 100644
--- a/kernel/trace/fprobe.c
+++ b/kernel/trace/fprobe.c
@@ -89,8 +89,11 @@ static bool delete_fprobe_node(struct fprobe_hlist_node *node)
 {
 	lockdep_assert_held(&fprobe_mutex);
 
-	WRITE_ONCE(node->fp, NULL);
-	hlist_del_rcu(&node->hlist);
+	/* Avoid double deleting */
+	if (READ_ONCE(node->fp) != NULL) {
+		WRITE_ONCE(node->fp, NULL);
+		hlist_del_rcu(&node->hlist);
+	}
 	return !!find_first_fprobe_node(node->addr);
 }
 
@@ -411,6 +414,102 @@ static void fprobe_graph_remove_ips(unsigned long *addrs, int num)
 		ftrace_set_filter_ips(&fprobe_graph_ops.ops, addrs, num, 1, 0);
 }
 
+#ifdef CONFIG_MODULES
+
+#define FPROBE_IPS_BATCH_INIT 8
+/* instruction pointer address list */
+struct fprobe_addr_list {
+	int index;
+	int size;
+	unsigned long *addrs;
+};
+
+static int fprobe_addr_list_add(struct fprobe_addr_list *alist, unsigned long addr)
+{
+	unsigned long *addrs;
+
+	if (alist->index >= alist->size)
+		return -ENOMEM;
+
+	alist->addrs[alist->index++] = addr;
+	if (alist->index < alist->size)
+		return 0;
+
+	/* Expand the address list */
+	addrs = kcalloc(alist->size * 2, sizeof(*addrs), GFP_KERNEL);
+	if (!addrs)
+		return -ENOMEM;
+
+	memcpy(addrs, alist->addrs, alist->size * sizeof(*addrs));
+	alist->size *= 2;
+	kfree(alist->addrs);
+	alist->addrs = addrs;
+
+	return 0;
+}
+
+static void fprobe_remove_node_in_module(struct module *mod, struct hlist_head *head,
+					struct fprobe_addr_list *alist)
+{
+	struct fprobe_hlist_node *node;
+	int ret = 0;
+
+	hlist_for_each_entry_rcu(node, head, hlist) {
+		if (!within_module(node->addr, mod))
+			continue;
+		if (delete_fprobe_node(node))
+			continue;
+		/*
+		 * If failed to update alist, just continue to update hlist.
+		 * Therefore, at list user handler will not hit anymore.
+		 */
+		if (!ret)
+			ret = fprobe_addr_list_add(alist, node->addr);
+	}
+}
+
+/* Handle module unloading to manage fprobe_ip_table. */
+static int fprobe_module_callback(struct notifier_block *nb,
+				  unsigned long val, void *data)
+{
+	struct fprobe_addr_list alist = {.size = FPROBE_IPS_BATCH_INIT};
+	struct module *mod = data;
+	int i;
+
+	if (val != MODULE_STATE_GOING)
+		return NOTIFY_DONE;
+
+	alist.addrs = kcalloc(alist.size, sizeof(*alist.addrs), GFP_KERNEL);
+	/* If failed to alloc memory, we can not remove ips from hash. */
+	if (!alist.addrs)
+		return NOTIFY_DONE;
+
+	mutex_lock(&fprobe_mutex);
+	for (i = 0; i < FPROBE_IP_TABLE_SIZE; i++)
+		fprobe_remove_node_in_module(mod, &fprobe_ip_table[i], &alist);
+
+	if (alist.index < alist.size && alist.index > 0)
+		ftrace_set_filter_ips(&fprobe_graph_ops.ops,
+				      alist.addrs, alist.index, 1, 0);
+	mutex_unlock(&fprobe_mutex);
+
+	kfree(alist.addrs);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block fprobe_module_nb = {
+	.notifier_call = fprobe_module_callback,
+	.priority = 0,
+};
+
+static int __init init_fprobe_module(void)
+{
+	return register_module_notifier(&fprobe_module_nb);
+}
+early_initcall(init_fprobe_module);
+#endif
+
 static int symbols_cmp(const void *a, const void *b)
 {
 	const char **str_a = (const char **) a;
@@ -445,6 +544,7 @@ struct filter_match_data {
 	size_t index;
 	size_t size;
 	unsigned long *addrs;
+	struct module **mods;
 };
 
 static int filter_match_callback(void *data, const char *name, unsigned long addr)
@@ -458,30 +558,47 @@ static int filter_match_callback(void *data, const char *name, unsigned long add
 	if (!ftrace_location(addr))
 		return 0;
 
-	if (match->addrs)
-		match->addrs[match->index] = addr;
+	if (match->addrs) {
+		struct module *mod = __module_text_address(addr);
+
+		if (mod && !try_module_get(mod))
+			return 0;
 
+		match->mods[match->index] = mod;
+		match->addrs[match->index] = addr;
+	}
 	match->index++;
 	return match->index == match->size;
 }
 
 /*
  * Make IP list from the filter/no-filter glob patterns.
- * Return the number of matched symbols, or -ENOENT.
+ * Return the number of matched symbols, or errno.
+ * If @addrs == NULL, this just counts the number of matched symbols. If @addrs
+ * is passed with an array, we need to pass the an @mods array of the same size
+ * to increment the module refcount for each symbol.
+ * This means we also need to call `module_put` for each element of @mods after
+ * using the @addrs.
  */
-static int ip_list_from_filter(const char *filter, const char *notfilter,
-			       unsigned long *addrs, size_t size)
+static int get_ips_from_filter(const char *filter, const char *notfilter,
+			       unsigned long *addrs, struct module **mods,
+			       size_t size)
 {
 	struct filter_match_data match = { .filter = filter, .notfilter = notfilter,
-		.index = 0, .size = size, .addrs = addrs};
+		.index = 0, .size = size, .addrs = addrs, .mods = mods};
 	int ret;
 
+	if (addrs && !mods)
+		return -EINVAL;
+
 	ret = kallsyms_on_each_symbol(filter_match_callback, &match);
 	if (ret < 0)
 		return ret;
-	ret = module_kallsyms_on_each_symbol(NULL, filter_match_callback, &match);
-	if (ret < 0)
-		return ret;
+	if (IS_ENABLED(CONFIG_MODULES)) {
+		ret = module_kallsyms_on_each_symbol(NULL, filter_match_callback, &match);
+		if (ret < 0)
+			return ret;
+	}
 
 	return match.index ?: -ENOENT;
 }
@@ -543,24 +660,35 @@ static int fprobe_init(struct fprobe *fp, unsigned long *addrs, int num)
  */
 int register_fprobe(struct fprobe *fp, const char *filter, const char *notfilter)
 {
-	unsigned long *addrs;
-	int ret;
+	unsigned long *addrs __free(kfree) = NULL;
+	struct module **mods __free(kfree) = NULL;
+	int ret, num;
 
 	if (!fp || !filter)
 		return -EINVAL;
 
-	ret = ip_list_from_filter(filter, notfilter, NULL, FPROBE_IPS_MAX);
-	if (ret < 0)
-		return ret;
+	num = get_ips_from_filter(filter, notfilter, NULL, NULL, FPROBE_IPS_MAX);
+	if (num < 0)
+		return num;
 
-	addrs = kcalloc(ret, sizeof(unsigned long), GFP_KERNEL);
+	addrs = kcalloc(num, sizeof(*addrs), GFP_KERNEL);
 	if (!addrs)
 		return -ENOMEM;
-	ret = ip_list_from_filter(filter, notfilter, addrs, ret);
-	if (ret > 0)
-		ret = register_fprobe_ips(fp, addrs, ret);
 
-	kfree(addrs);
+	mods = kcalloc(num, sizeof(*mods), GFP_KERNEL);
+	if (!mods)
+		return -ENOMEM;
+
+	ret = get_ips_from_filter(filter, notfilter, addrs, mods, num);
+	if (ret < 0)
+		return ret;
+
+	ret = register_fprobe_ips(fp, addrs, ret);
+
+	for (int i = 0; i < num; i++) {
+		if (mods[i])
+			module_put(mods[i]);
+	}
 	return ret;
 }
 EXPORT_SYMBOL_GPL(register_fprobe);
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 1a48aedb5255..6981830c3128 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1297,6 +1297,8 @@ void ftrace_free_filter(struct ftrace_ops *ops)
 		return;
 	free_ftrace_hash(ops->func_hash->filter_hash);
 	free_ftrace_hash(ops->func_hash->notrace_hash);
+	ops->func_hash->filter_hash = EMPTY_HASH;
+	ops->func_hash->notrace_hash = EMPTY_HASH;
 }
 EXPORT_SYMBOL_GPL(ftrace_free_filter);
 
@@ -3256,6 +3258,31 @@ static int append_hash(struct ftrace_hash **hash, struct ftrace_hash *new_hash,
 }
 
 /*
+ * Remove functions from @hash that are in @notrace_hash
+ */
+static void remove_hash(struct ftrace_hash *hash, struct ftrace_hash *notrace_hash)
+{
+	struct ftrace_func_entry *entry;
+	struct hlist_node *tmp;
+	int size;
+	int i;
+
+	/* If the notrace hash is empty, there's nothing to do */
+	if (ftrace_hash_empty(notrace_hash))
+		return;
+
+	size = 1 << hash->size_bits;
+	for (i = 0; i < size; i++) {
+		hlist_for_each_entry_safe(entry, tmp, &hash->buckets[i], hlist) {
+			if (!__ftrace_lookup_ip(notrace_hash, entry->ip))
+				continue;
+			remove_hash_entry(hash, entry);
+			kfree(entry);
+		}
+	}
+}
+
+/*
  * Add to @hash only those that are in both @new_hash1 and @new_hash2
  *
  * The notrace_hash updates uses just the intersect_hash() function
@@ -3295,67 +3322,6 @@ static int intersect_hash(struct ftrace_hash **hash, struct ftrace_hash *new_has
 	return 0;
 }
 
-/* Return a new hash that has a union of all @ops->filter_hash entries */
-static struct ftrace_hash *append_hashes(struct ftrace_ops *ops)
-{
-	struct ftrace_hash *new_hash = NULL;
-	struct ftrace_ops *subops;
-	int size_bits;
-	int ret;
-
-	if (ops->func_hash->filter_hash)
-		size_bits = ops->func_hash->filter_hash->size_bits;
-	else
-		size_bits = FTRACE_HASH_DEFAULT_BITS;
-
-	list_for_each_entry(subops, &ops->subop_list, list) {
-		ret = append_hash(&new_hash, subops->func_hash->filter_hash, size_bits);
-		if (ret < 0) {
-			free_ftrace_hash(new_hash);
-			return NULL;
-		}
-		/* Nothing more to do if new_hash is empty */
-		if (ftrace_hash_empty(new_hash))
-			break;
-	}
-	/* Can't return NULL as that means this failed */
-	return new_hash ? : EMPTY_HASH;
-}
-
-/* Make @ops trace evenything except what all its subops do not trace */
-static struct ftrace_hash *intersect_hashes(struct ftrace_ops *ops)
-{
-	struct ftrace_hash *new_hash = NULL;
-	struct ftrace_ops *subops;
-	int size_bits;
-	int ret;
-
-	list_for_each_entry(subops, &ops->subop_list, list) {
-		struct ftrace_hash *next_hash;
-
-		if (!new_hash) {
-			size_bits = subops->func_hash->notrace_hash->size_bits;
-			new_hash = alloc_and_copy_ftrace_hash(size_bits, ops->func_hash->notrace_hash);
-			if (!new_hash)
-				return NULL;
-			continue;
-		}
-		size_bits = new_hash->size_bits;
-		next_hash = new_hash;
-		new_hash = alloc_ftrace_hash(size_bits);
-		ret = intersect_hash(&new_hash, next_hash, subops->func_hash->notrace_hash);
-		free_ftrace_hash(next_hash);
-		if (ret < 0) {
-			free_ftrace_hash(new_hash);
-			return NULL;
-		}
-		/* Nothing more to do if new_hash is empty */
-		if (ftrace_hash_empty(new_hash))
-			break;
-	}
-	return new_hash;
-}
-
 static bool ops_equal(struct ftrace_hash *A, struct ftrace_hash *B)
 {
 	struct ftrace_func_entry *entry;
@@ -3427,6 +3393,95 @@ static int ftrace_update_ops(struct ftrace_ops *ops, struct ftrace_hash *filter_
 	return 0;
 }
 
+static int add_first_hash(struct ftrace_hash **filter_hash, struct ftrace_hash **notrace_hash,
+			  struct ftrace_ops_hash *func_hash)
+{
+	/* If the filter hash is not empty, simply remove the nohash from it */
+	if (!ftrace_hash_empty(func_hash->filter_hash)) {
+		*filter_hash = copy_hash(func_hash->filter_hash);
+		if (!*filter_hash)
+			return -ENOMEM;
+		remove_hash(*filter_hash, func_hash->notrace_hash);
+		*notrace_hash = EMPTY_HASH;
+
+	} else {
+		*notrace_hash = copy_hash(func_hash->notrace_hash);
+		if (!*notrace_hash)
+			return -ENOMEM;
+		*filter_hash = EMPTY_HASH;
+	}
+	return 0;
+}
+
+static int add_next_hash(struct ftrace_hash **filter_hash, struct ftrace_hash **notrace_hash,
+			 struct ftrace_ops_hash *ops_hash, struct ftrace_ops_hash *subops_hash)
+{
+	int size_bits;
+	int ret;
+
+	/* If the subops trace all functions so must the main ops */
+	if (ftrace_hash_empty(ops_hash->filter_hash) ||
+	    ftrace_hash_empty(subops_hash->filter_hash)) {
+		*filter_hash = EMPTY_HASH;
+	} else {
+		/*
+		 * The main ops filter hash is not empty, so its
+		 * notrace_hash had better be, as the notrace hash
+		 * is only used for empty main filter hashes.
+		 */
+		WARN_ON_ONCE(!ftrace_hash_empty(ops_hash->notrace_hash));
+
+		size_bits = max(ops_hash->filter_hash->size_bits,
+				subops_hash->filter_hash->size_bits);
+
+		/* Copy the subops hash */
+		*filter_hash = alloc_and_copy_ftrace_hash(size_bits, subops_hash->filter_hash);
+		if (!*filter_hash)
+			return -ENOMEM;
+		/* Remove any notrace functions from the copy */
+		remove_hash(*filter_hash, subops_hash->notrace_hash);
+
+		ret = append_hash(filter_hash, ops_hash->filter_hash,
+				  size_bits);
+		if (ret < 0) {
+			free_ftrace_hash(*filter_hash);
+			*filter_hash = EMPTY_HASH;
+			return ret;
+		}
+	}
+
+	/*
+	 * Only process notrace hashes if the main filter hash is empty
+	 * (tracing all functions), otherwise the filter hash will just
+	 * remove the notrace hash functions, and the notrace hash is
+	 * not needed.
+	 */
+	if (ftrace_hash_empty(*filter_hash)) {
+		/*
+		 * Intersect the notrace functions. That is, if two
+		 * subops are not tracing a set of functions, the
+		 * main ops will only not trace the functions that are
+		 * in both subops, but has to trace the functions that
+		 * are only notrace in one of the subops, for the other
+		 * subops to be able to trace them.
+		 */
+		size_bits = max(ops_hash->notrace_hash->size_bits,
+				subops_hash->notrace_hash->size_bits);
+		*notrace_hash = alloc_ftrace_hash(size_bits);
+		if (!*notrace_hash)
+			return -ENOMEM;
+
+		ret = intersect_hash(notrace_hash, ops_hash->notrace_hash,
+				     subops_hash->notrace_hash);
+		if (ret < 0) {
+			free_ftrace_hash(*notrace_hash);
+			*notrace_hash = EMPTY_HASH;
+			return ret;
+		}
+	}
+	return 0;
+}
+
 /**
  * ftrace_startup_subops - enable tracing for subops of an ops
  * @ops: Manager ops (used to pick all the functions of its subops)
@@ -3439,11 +3494,10 @@ static int ftrace_update_ops(struct ftrace_ops *ops, struct ftrace_hash *filter_
  */
 int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command)
 {
-	struct ftrace_hash *filter_hash;
-	struct ftrace_hash *notrace_hash;
+	struct ftrace_hash *filter_hash = EMPTY_HASH;
+	struct ftrace_hash *notrace_hash = EMPTY_HASH;
 	struct ftrace_hash *save_filter_hash;
 	struct ftrace_hash *save_notrace_hash;
-	int size_bits;
 	int ret;
 
 	if (unlikely(ftrace_disabled))
@@ -3467,14 +3521,14 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
 
 	/* For the first subops to ops just enable it normally */
 	if (list_empty(&ops->subop_list)) {
-		/* Just use the subops hashes */
-		filter_hash = copy_hash(subops->func_hash->filter_hash);
-		notrace_hash = copy_hash(subops->func_hash->notrace_hash);
-		if (!filter_hash || !notrace_hash) {
-			free_ftrace_hash(filter_hash);
-			free_ftrace_hash(notrace_hash);
-			return -ENOMEM;
-		}
+
+		/* The ops was empty, should have empty hashes */
+		WARN_ON_ONCE(!ftrace_hash_empty(ops->func_hash->filter_hash));
+		WARN_ON_ONCE(!ftrace_hash_empty(ops->func_hash->notrace_hash));
+
+		ret = add_first_hash(&filter_hash, &notrace_hash, subops->func_hash);
+		if (ret < 0)
+			return ret;
 
 		save_filter_hash = ops->func_hash->filter_hash;
 		save_notrace_hash = ops->func_hash->notrace_hash;
@@ -3500,48 +3554,16 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
 
 	/*
 	 * Here there's already something attached. Here are the rules:
-	 *   o If either filter_hash is empty then the final stays empty
-	 *      o Otherwise, the final is a superset of both hashes
-	 *   o If either notrace_hash is empty then the final stays empty
-	 *      o Otherwise, the final is an intersection between the hashes
+	 *   If the new subops and main ops filter hashes are not empty:
+	 *     o Make a copy of the subops filter hash
+	 *     o Remove all functions in the nohash from it.
+	 *     o Add in the main hash filter functions
+	 *     o Remove any of these functions from the main notrace hash
 	 */
-	if (ftrace_hash_empty(ops->func_hash->filter_hash) ||
-	    ftrace_hash_empty(subops->func_hash->filter_hash)) {
-		filter_hash = EMPTY_HASH;
-	} else {
-		size_bits = max(ops->func_hash->filter_hash->size_bits,
-				subops->func_hash->filter_hash->size_bits);
-		filter_hash = alloc_and_copy_ftrace_hash(size_bits, ops->func_hash->filter_hash);
-		if (!filter_hash)
-			return -ENOMEM;
-		ret = append_hash(&filter_hash, subops->func_hash->filter_hash,
-				  size_bits);
-		if (ret < 0) {
-			free_ftrace_hash(filter_hash);
-			return ret;
-		}
-	}
 
-	if (ftrace_hash_empty(ops->func_hash->notrace_hash) ||
-	    ftrace_hash_empty(subops->func_hash->notrace_hash)) {
-		notrace_hash = EMPTY_HASH;
-	} else {
-		size_bits = max(ops->func_hash->filter_hash->size_bits,
-				subops->func_hash->filter_hash->size_bits);
-		notrace_hash = alloc_ftrace_hash(size_bits);
-		if (!notrace_hash) {
-			free_ftrace_hash(filter_hash);
-			return -ENOMEM;
-		}
-
-		ret = intersect_hash(&notrace_hash, ops->func_hash->filter_hash,
-				     subops->func_hash->filter_hash);
-		if (ret < 0) {
-			free_ftrace_hash(filter_hash);
-			free_ftrace_hash(notrace_hash);
-			return ret;
-		}
-	}
+	ret = add_next_hash(&filter_hash, &notrace_hash, ops->func_hash, subops->func_hash);
+	if (ret < 0)
+		return ret;
 
 	list_add(&subops->list, &ops->subop_list);
 
@@ -3557,6 +3579,45 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
 	return ret;
 }
 
+static int rebuild_hashes(struct ftrace_hash **filter_hash, struct ftrace_hash **notrace_hash,
+			  struct ftrace_ops *ops)
+{
+	struct ftrace_ops_hash temp_hash;
+	struct ftrace_ops *subops;
+	bool first = true;
+	int ret;
+
+	temp_hash.filter_hash = EMPTY_HASH;
+	temp_hash.notrace_hash = EMPTY_HASH;
+
+	list_for_each_entry(subops, &ops->subop_list, list) {
+		*filter_hash = EMPTY_HASH;
+		*notrace_hash = EMPTY_HASH;
+
+		if (first) {
+			ret = add_first_hash(filter_hash, notrace_hash, subops->func_hash);
+			if (ret < 0)
+				return ret;
+			first = false;
+		} else {
+			ret = add_next_hash(filter_hash, notrace_hash,
+					    &temp_hash, subops->func_hash);
+			if (ret < 0) {
+				free_ftrace_hash(temp_hash.filter_hash);
+				free_ftrace_hash(temp_hash.notrace_hash);
+				return ret;
+			}
+		}
+
+		free_ftrace_hash(temp_hash.filter_hash);
+		free_ftrace_hash(temp_hash.notrace_hash);
+
+		temp_hash.filter_hash = *filter_hash;
+		temp_hash.notrace_hash = *notrace_hash;
+	}
+	return 0;
+}
+
 /**
  * ftrace_shutdown_subops - Remove a subops from a manager ops
  * @ops: A manager ops to remove @subops from
@@ -3571,8 +3632,8 @@ int ftrace_startup_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int
  */
 int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, int command)
 {
-	struct ftrace_hash *filter_hash;
-	struct ftrace_hash *notrace_hash;
+	struct ftrace_hash *filter_hash = EMPTY_HASH;
+	struct ftrace_hash *notrace_hash = EMPTY_HASH;
 	int ret;
 
 	if (unlikely(ftrace_disabled))
@@ -3605,14 +3666,9 @@ int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, in
 	}
 
 	/* Rebuild the hashes without subops */
-	filter_hash = append_hashes(ops);
-	notrace_hash = intersect_hashes(ops);
-	if (!filter_hash || !notrace_hash) {
-		free_ftrace_hash(filter_hash);
-		free_ftrace_hash(notrace_hash);
-		list_add(&subops->list, &ops->subop_list);
-		return -ENOMEM;
-	}
+	ret = rebuild_hashes(&filter_hash, &notrace_hash, ops);
+	if (ret < 0)
+		return ret;
 
 	ret = ftrace_update_ops(ops, filter_hash, notrace_hash);
 	if (ret < 0) {
@@ -3628,11 +3684,11 @@ int ftrace_shutdown_subops(struct ftrace_ops *ops, struct ftrace_ops *subops, in
 
 static int ftrace_hash_move_and_update_subops(struct ftrace_ops *subops,
 					      struct ftrace_hash **orig_subhash,
-					      struct ftrace_hash *hash,
-					      int enable)
+					      struct ftrace_hash *hash)
 {
 	struct ftrace_ops *ops = subops->managed;
-	struct ftrace_hash **orig_hash;
+	struct ftrace_hash *notrace_hash;
+	struct ftrace_hash *filter_hash;
 	struct ftrace_hash *save_hash;
 	struct ftrace_hash *new_hash;
 	int ret;
@@ -3649,24 +3705,18 @@ static int ftrace_hash_move_and_update_subops(struct ftrace_ops *subops,
 		return -ENOMEM;
 	}
 
-	/* Create a new_hash to hold the ops new functions */
-	if (enable) {
-		orig_hash = &ops->func_hash->filter_hash;
-		new_hash = append_hashes(ops);
-	} else {
-		orig_hash = &ops->func_hash->notrace_hash;
-		new_hash = intersect_hashes(ops);
+	ret = rebuild_hashes(&filter_hash, &notrace_hash, ops);
+	if (!ret) {
+		ret = ftrace_update_ops(ops, filter_hash, notrace_hash);
+		free_ftrace_hash(filter_hash);
+		free_ftrace_hash(notrace_hash);
 	}
 
-	/* Move the hash over to the new hash */
-	ret = __ftrace_hash_move_and_update_ops(ops, orig_hash, new_hash, enable);
-
-	free_ftrace_hash(new_hash);
-
 	if (ret) {
 		/* Put back the original hash */
-		free_ftrace_hash_rcu(*orig_subhash);
+		new_hash = *orig_subhash;
 		*orig_subhash = save_hash;
+		free_ftrace_hash_rcu(new_hash);
 	} else {
 		free_ftrace_hash_rcu(save_hash);
 	}
@@ -4890,7 +4940,7 @@ static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops,
 					   int enable)
 {
 	if (ops->flags & FTRACE_OPS_FL_SUBOP)
-		return ftrace_hash_move_and_update_subops(ops, orig_hash, hash, enable);
+		return ftrace_hash_move_and_update_subops(ops, orig_hash, hash);
 
 	/*
 	 * If this ops is not enabled, it could be sharing its filters
@@ -4909,7 +4959,7 @@ static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops,
 			list_for_each_entry(subops, &op->subop_list, list) {
 				if ((subops->flags & FTRACE_OPS_FL_ENABLED) &&
 				     subops->func_hash == ops->func_hash) {
-					return ftrace_hash_move_and_update_subops(subops, orig_hash, hash, enable);
+					return ftrace_hash_move_and_update_subops(subops, orig_hash, hash);
 				}
 			}
 		} while_for_each_ftrace_op(op);
@@ -5914,9 +5964,10 @@ int register_ftrace_direct(struct ftrace_ops *ops, unsigned long addr)
 
 	/* Make a copy hash to place the new and the old entries in */
 	size = hash->count + direct_functions->count;
-	if (size > 32)
-		size = 32;
-	new_hash = alloc_ftrace_hash(fls(size));
+	size = fls(size);
+	if (size > FTRACE_HASH_MAX_BITS)
+		size = FTRACE_HASH_MAX_BITS;
+	new_hash = alloc_ftrace_hash(size);
 	if (!new_hash)
 		goto out_unlock;
 
diff --git a/kernel/trace/rv/rv.c b/kernel/trace/rv/rv.c
index 968c5c3b0246..e4077500a91d 100644
--- a/kernel/trace/rv/rv.c
+++ b/kernel/trace/rv/rv.c
@@ -225,7 +225,12 @@ bool rv_is_nested_monitor(struct rv_monitor_def *mdef)
  */
 bool rv_is_container_monitor(struct rv_monitor_def *mdef)
 {
-	struct rv_monitor_def *next = list_next_entry(mdef, list);
+	struct rv_monitor_def *next;
+
+	if (list_is_last(&mdef->list, &rv_monitors_list))
+		return false;
+
+	next = list_next_entry(mdef, list);
 
 	return next->parent == mdef->monitor || !mdef->monitor->enable;
 }
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index b581e388a9d9..5b8db27fb6ef 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -6043,8 +6043,10 @@ unsigned long trace_adjust_address(struct trace_array *tr, unsigned long addr)
 	tscratch = tr->scratch;
 	/* if there is no tscrach, module_delta must be NULL. */
 	module_delta = READ_ONCE(tr->module_delta);
-	if (!module_delta || tscratch->entries[0].mod_addr > addr)
+	if (!module_delta || !tscratch->nr_entries ||
+	    tscratch->entries[0].mod_addr > addr) {
 		return addr + tr->text_delta;
+	}
 
 	/* Note that entries must be sorted. */
 	nr_entries = tscratch->nr_entries;
@@ -6821,13 +6823,14 @@ static ssize_t tracing_splice_read_pipe(struct file *filp,
 		/* Copy the data into the page, so we can start over. */
 		ret = trace_seq_to_buffer(&iter->seq,
 					  page_address(spd.pages[i]),
-					  trace_seq_used(&iter->seq));
+					  min((size_t)trace_seq_used(&iter->seq),
+						  PAGE_SIZE));
 		if (ret < 0) {
 			__free_page(spd.pages[i]);
 			break;
 		}
 		spd.partial[i].offset = 0;
-		spd.partial[i].len = trace_seq_used(&iter->seq);
+		spd.partial[i].len = ret;
 
 		trace_seq_init(&iter->seq);
 	}
@@ -9806,6 +9809,7 @@ static int instance_mkdir(const char *name)
 	return ret;
 }
 
+#ifdef CONFIG_MMU
 static u64 map_pages(unsigned long start, unsigned long size)
 {
 	unsigned long vmap_start, vmap_end;
@@ -9828,6 +9832,12 @@ static u64 map_pages(unsigned long start, unsigned long size)
 
 	return (u64)vmap_start;
 }
+#else
+static inline u64 map_pages(unsigned long start, unsigned long size)
+{
+	return 0;
+}
+#endif
 
 /**
  * trace_array_get_by_name - Create/Lookup a trace array, given its name.
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index ee40d4e6ad1c..4ef4df6623a8 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -80,11 +80,11 @@ FTRACE_ENTRY(funcgraph_entry, ftrace_graph_ent_entry,
 	F_STRUCT(
 		__field_struct(	struct ftrace_graph_ent,	graph_ent	)
 		__field_packed(	unsigned long,	graph_ent,	func		)
-		__field_packed(	unsigned long,	graph_ent,	depth		)
+		__field_packed(	unsigned int,	graph_ent,	depth		)
 		__dynamic_array(unsigned long,	args				)
 	),
 
-	F_printk("--> %ps (%lu)", (void *)__entry->func, __entry->depth)
+	F_printk("--> %ps (%u)", (void *)__entry->func, __entry->depth)
 );
 
 #ifdef CONFIG_FUNCTION_GRAPH_RETADDR
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 0993dfc1c5c1..2048560264bb 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -808,7 +808,7 @@ static __always_inline char *test_string(char *str)
 	kstr = ubuf->buffer;
 
 	/* For safety, do not trust the string pointer */
-	if (!strncpy_from_kernel_nofault(kstr, str, USTRING_BUF_SIZE))
+	if (strncpy_from_kernel_nofault(kstr, str, USTRING_BUF_SIZE) < 0)
 		return NULL;
 	return kstr;
 }
@@ -827,7 +827,7 @@ static __always_inline char *test_ustring(char *str)
 
 	/* user space address? */
 	ustr = (char __user *)str;
-	if (!strncpy_from_user_nofault(kstr, ustr, USTRING_BUF_SIZE))
+	if (strncpy_from_user_nofault(kstr, ustr, USTRING_BUF_SIZE) < 0)
 		return NULL;
 
 	return kstr;
diff --git a/kernel/trace/trace_events_synth.c b/kernel/trace/trace_events_synth.c
index 969f48742d72..33cfbd4ed76d 100644
--- a/kernel/trace/trace_events_synth.c
+++ b/kernel/trace/trace_events_synth.c
@@ -370,7 +370,6 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter,
 				union trace_synth_field *data = &entry->fields[n_u64];
 
 				trace_seq_printf(s, print_fmt, se->fields[i]->name,
-						 STR_VAR_LEN_MAX,
 						 (char *)entry + data->as_dynamic.offset,
 						 i == se->n_fields - 1 ? "" : " ");
 				n_u64++;
diff --git a/kernel/trace/trace_fprobe.c b/kernel/trace/trace_fprobe.c
index 5d7ca80173ea..b40fa59159ac 100644
--- a/kernel/trace/trace_fprobe.c
+++ b/kernel/trace/trace_fprobe.c
@@ -919,9 +919,15 @@ static void __find_tracepoint_module_cb(struct tracepoint *tp, struct module *mo
 	struct __find_tracepoint_cb_data *data = priv;
 
 	if (!data->tpoint && !strcmp(data->tp_name, tp->name)) {
-		data->tpoint = tp;
-		if (!data->mod)
+		/* If module is not specified, try getting module refcount. */
+		if (!data->mod && mod) {
+			/* If failed to get refcount, ignore this tracepoint. */
+			if (!try_module_get(mod))
+				return;
+
 			data->mod = mod;
+		}
+		data->tpoint = tp;
 	}
 }
 
@@ -933,7 +939,11 @@ static void __find_tracepoint_cb(struct tracepoint *tp, void *priv)
 		data->tpoint = tp;
 }
 
-/* Find a tracepoint from kernel and module. */
+/*
+ * Find a tracepoint from kernel and module. If the tracepoint is on the module,
+ * the module's refcount is incremented and returned as *@tp_mod. Thus, if it is
+ * not NULL, caller must call module_put(*tp_mod) after used the tracepoint.
+ */
 static struct tracepoint *find_tracepoint(const char *tp_name,
 					  struct module **tp_mod)
 {
@@ -962,7 +972,10 @@ static void reenable_trace_fprobe(struct trace_fprobe *tf)
 	}
 }
 
-/* Find a tracepoint from specified module. */
+/*
+ * Find a tracepoint from specified module. In this case, this does not get the
+ * module's refcount. The caller must ensure the module is not freed.
+ */
 static struct tracepoint *find_tracepoint_in_module(struct module *mod,
 						    const char *tp_name)
 {
@@ -1169,11 +1182,6 @@ static int trace_fprobe_create_internal(int argc, const char *argv[],
 	if (is_tracepoint) {
 		ctx->flags |= TPARG_FL_TPOINT;
 		tpoint = find_tracepoint(symbol, &tp_mod);
-		/* lock module until register this tprobe. */
-		if (tp_mod && !try_module_get(tp_mod)) {
-			tpoint = NULL;
-			tp_mod = NULL;
-		}
 		if (tpoint) {
 			ctx->funcname = kallsyms_lookup(
 				(unsigned long)tpoint->probestub,
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index 2f077d4158e5..0c357a89c58e 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -880,8 +880,6 @@ static void print_graph_retval(struct trace_seq *s, struct ftrace_graph_ent_entr
 
 		if (print_retval || print_retaddr)
 			trace_seq_puts(s, " /*");
-		else
-			trace_seq_putc(s, '\n');
 	} else {
 		print_retaddr = false;
 		trace_seq_printf(s, "} /* %ps", func);
@@ -899,7 +897,7 @@ static void print_graph_retval(struct trace_seq *s, struct ftrace_graph_ent_entr
 	}
 
 	if (!entry || print_retval || print_retaddr)
-		trace_seq_puts(s, " */\n");
+		trace_seq_puts(s, " */");
 }
 
 #else
@@ -975,7 +973,7 @@ print_graph_entry_leaf(struct trace_iterator *iter,
 		} else
 			trace_seq_puts(s, "();");
 	}
-	trace_seq_printf(s, "\n");
+	trace_seq_putc(s, '\n');
 
 	print_graph_irq(iter, graph_ret->func, TRACE_GRAPH_RET,
 			cpu, iter->ent->pid, flags);
@@ -1313,10 +1311,11 @@ print_graph_return(struct ftrace_graph_ret_entry *retentry, struct trace_seq *s,
 		 * that if the funcgraph-tail option is enabled.
 		 */
 		if (func_match && !(flags & TRACE_GRAPH_PRINT_TAIL))
-			trace_seq_puts(s, "}\n");
+			trace_seq_puts(s, "}");
 		else
-			trace_seq_printf(s, "} /* %ps */\n", (void *)func);
+			trace_seq_printf(s, "} /* %ps */", (void *)func);
 	}
+	trace_seq_putc(s, '\n');
 
 	/* Overrun */
 	if (flags & TRACE_GRAPH_PRINT_OVERRUN)
diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
index fee40ffbd490..b9ab06c99543 100644
--- a/kernel/trace/trace_output.c
+++ b/kernel/trace/trace_output.c
@@ -1042,11 +1042,12 @@ enum print_line_t print_event_fields(struct trace_iterator *iter,
 	struct trace_event_call *call;
 	struct list_head *head;
 
+	lockdep_assert_held_read(&trace_event_sem);
+
 	/* ftrace defined events have separate call structures */
 	if (event->type <= __TRACE_LAST_TYPE) {
 		bool found = false;
 
-		down_read(&trace_event_sem);
 		list_for_each_entry(call, &ftrace_events, list) {
 			if (call->event.type == event->type) {
 				found = true;
@@ -1056,7 +1057,6 @@ enum print_line_t print_event_fields(struct trace_iterator *iter,
 			if (call->event.type > __TRACE_LAST_TYPE)
 				break;
 		}
-		up_read(&trace_event_sem);
 		if (!found) {
 			trace_seq_printf(&iter->seq, "UNKNOWN TYPE %d\n", event->type);
 			goto out;
diff --git a/kernel/vhost_task.c b/kernel/vhost_task.c
index 2ef2e1b80091..2f844c279a3e 100644
--- a/kernel/vhost_task.c
+++ b/kernel/vhost_task.c
@@ -111,7 +111,7 @@ EXPORT_SYMBOL_GPL(vhost_task_stop);
  * @arg: data to be passed to fn and handled_kill
  * @name: the thread's name
  *
- * This returns a specialized task for use by the vhost layer or NULL on
+ * This returns a specialized task for use by the vhost layer or ERR_PTR() on
  * failure. The returned task is inactive, and the caller must fire it up
  * through vhost_task_start().
  */
diff --git a/lib/Kconfig b/lib/Kconfig
index 61cce0686b53..6c1b8f184267 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -139,27 +139,22 @@ config TRACE_MMIO_ACCESS
 source "lib/crypto/Kconfig"
 
 config CRC_CCITT
-	tristate "CRC-CCITT functions"
+	tristate
 	help
-	  This option is provided for the case where no in-kernel-tree
-	  modules require CRC-CCITT functions, but a module built outside
-	  the kernel tree does. Such modules that use library CRC-CCITT
-	  functions require M here.
+	  The CRC-CCITT library functions.  Select this if your module uses any
+	  of the functions from <linux/crc-ccitt.h>.
 
 config CRC16
-	tristate "CRC16 functions"
+	tristate
 	help
-	  This option is provided for the case where no in-kernel-tree
-	  modules require CRC16 functions, but a module built outside
-	  the kernel tree does. Such modules that use library CRC16
-	  functions require M here.
+	  The CRC16 library functions.  Select this if your module uses any of
+	  the functions from <linux/crc16.h>.
 
 config CRC_T10DIF
-	tristate "CRC calculation for the T10 Data Integrity Field"
+	tristate
 	help
-	  This option is only needed if a module that's not in the
-	  kernel tree needs to calculate CRC checks for use with the
-	  SCSI data integrity subsystem.
+	  The CRC-T10DIF library functions.  Select this if your module uses
+	  any of the functions from <linux/crc-t10dif.h>.
 
 config ARCH_HAS_CRC_T10DIF
 	bool
@@ -169,22 +164,17 @@ config CRC_T10DIF_ARCH
 	default CRC_T10DIF if ARCH_HAS_CRC_T10DIF && CRC_OPTIMIZATIONS
 
 config CRC_ITU_T
-	tristate "CRC ITU-T V.41 functions"
+	tristate
 	help
-	  This option is provided for the case where no in-kernel-tree
-	  modules require CRC ITU-T V.41 functions, but a module built outside
-	  the kernel tree does. Such modules that use library CRC ITU-T V.41
-	  functions require M here.
+	  The CRC-ITU-T library functions.  Select this if your module uses
+	  any of the functions from <linux/crc-itu-t.h>.
 
 config CRC32
-	tristate "CRC32/CRC32c functions"
-	default y
+	tristate
 	select BITREVERSE
 	help
-	  This option is provided for the case where no in-kernel-tree
-	  modules require CRC32/CRC32c functions, but a module built outside
-	  the kernel tree does. Such modules that use library CRC32/CRC32c
-	  functions require M here.
+	  The CRC32 library functions.  Select this if your module uses any of
+	  the functions from <linux/crc32.h> or <linux/crc32c.h>.
 
 config ARCH_HAS_CRC32
 	bool
@@ -195,6 +185,9 @@ config CRC32_ARCH
 
 config CRC64
 	tristate
+	help
+	  The CRC64 library functions.  Select this if your module uses any of
+	  the functions from <linux/crc64.h>.
 
 config ARCH_HAS_CRC64
 	bool
@@ -205,19 +198,21 @@ config CRC64_ARCH
 
 config CRC4
 	tristate
+	help
+	  The CRC4 library functions.  Select this if your module uses any of
+	  the functions from <linux/crc4.h>.
 
 config CRC7
 	tristate
-
-config LIBCRC32C
-	tristate
-	select CRC32
 	help
-	  This option just selects CRC32 and is provided for compatibility
-	  purposes until the users are updated to select CRC32 directly.
+	  The CRC7 library functions.  Select this if your module uses any of
+	  the functions from <linux/crc7.h>.
 
 config CRC8
 	tristate
+	help
+	  The CRC8 library functions.  Select this if your module uses any of
+	  the functions from <linux/crc8.h>.
 
 config CRC_OPTIMIZATIONS
 	bool "Enable optimized CRC implementations" if EXPERT
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 9fe4d8dfe578..f9051ab610d5 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -3290,7 +3290,7 @@ config GCD_KUNIT_TEST
 config PRIME_NUMBERS_KUNIT_TEST
 	tristate "Prime number generator test" if !KUNIT_ALL_TESTS
 	depends on KUNIT
-	select PRIME_NUMBERS
+	depends on PRIME_NUMBERS
 	default KUNIT_ALL_TESTS
 	help
 	  This option enables the KUnit test suite for the {is,next}_prime_number
diff --git a/lib/Kconfig.ubsan b/lib/Kconfig.ubsan
index 4216b3a4ff21..f6ea0c5b5da3 100644
--- a/lib/Kconfig.ubsan
+++ b/lib/Kconfig.ubsan
@@ -118,7 +118,6 @@ config UBSAN_UNREACHABLE
 
 config UBSAN_INTEGER_WRAP
 	bool "Perform checking for integer arithmetic wrap-around"
-	default UBSAN
 	depends on !COMPILE_TEST
 	depends on $(cc-option,-fsanitize-undefined-ignore-overflow-pattern=all)
 	depends on $(cc-option,-fsanitize=signed-integer-overflow)
diff --git a/lib/alloc_tag.c b/lib/alloc_tag.c
index 1d893e313614..25ecc1334b67 100644
--- a/lib/alloc_tag.c
+++ b/lib/alloc_tag.c
@@ -422,11 +422,20 @@ static int vm_module_tags_populate(void)
 		unsigned long old_shadow_end = ALIGN(phys_end, MODULE_ALIGN);
 		unsigned long new_shadow_end = ALIGN(new_end, MODULE_ALIGN);
 		unsigned long more_pages;
-		unsigned long nr;
+		unsigned long nr = 0;
 
 		more_pages = ALIGN(new_end - phys_end, PAGE_SIZE) >> PAGE_SHIFT;
-		nr = alloc_pages_bulk_node(GFP_KERNEL | __GFP_NOWARN,
-					   NUMA_NO_NODE, more_pages, next_page);
+		while (nr < more_pages) {
+			unsigned long allocated;
+
+			allocated = alloc_pages_bulk_node(GFP_KERNEL | __GFP_NOWARN,
+				NUMA_NO_NODE, more_pages - nr, next_page + nr);
+
+			if (!allocated)
+				break;
+			nr += allocated;
+		}
+
 		if (nr < more_pages ||
 		    vmap_pages_range(phys_end, phys_end + (nr << PAGE_SHIFT), PAGE_KERNEL,
 				     next_page, PAGE_SHIFT) < 0) {
diff --git a/lib/asn1_decoder.c b/lib/asn1_decoder.c
index 13da529e2e72..5738ae286b41 100644
--- a/lib/asn1_decoder.c
+++ b/lib/asn1_decoder.c
@@ -518,4 +518,5 @@ error:
 }
 EXPORT_SYMBOL_GPL(asn1_ber_decoder);
 
+MODULE_DESCRIPTION("Decoder for ASN.1 BER/DER/CER encoded bytestream");
 MODULE_LICENSE("GPL");
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 8c7fdb7d8c8f..bc9391e55d57 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -1191,7 +1191,7 @@ static ssize_t __iov_iter_get_pages_alloc(struct iov_iter *i,
 			return -ENOMEM;
 		p = *pages;
 		for (int k = 0; k < n; k++) {
-			struct folio *folio = page_folio(page);
+			struct folio *folio = page_folio(page + k);
 			p[k] = page + k;
 			if (!folio_test_slab(folio))
 				folio_get(folio);
diff --git a/lib/string.c b/lib/string.c
index eb4486ed40d2..b632c71df1a5 100644
--- a/lib/string.c
+++ b/lib/string.c
@@ -119,6 +119,7 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count)
 	if (count == 0 || WARN_ON_ONCE(count > INT_MAX))
 		return -E2BIG;
 
+#ifndef CONFIG_DCACHE_WORD_ACCESS
 #ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 	/*
 	 * If src is unaligned, don't cross a page boundary,
@@ -134,11 +135,13 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count)
 	if (((long) dest | (long) src) & (sizeof(long) - 1))
 		max = 0;
 #endif
+#endif
 
 	/*
-	 * read_word_at_a_time() below may read uninitialized bytes after the
-	 * trailing zero and use them in comparisons. Disable this optimization
-	 * under KMSAN to prevent false positive reports.
+	 * load_unaligned_zeropad() or read_word_at_a_time() below may read
+	 * uninitialized bytes after the trailing zero and use them in
+	 * comparisons. Disable this optimization under KMSAN to prevent
+	 * false positive reports.
 	 */
 	if (IS_ENABLED(CONFIG_KMSAN))
 		max = 0;
@@ -146,7 +149,11 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count)
 	while (max >= sizeof(unsigned long)) {
 		unsigned long c, data;
 
+#ifdef CONFIG_DCACHE_WORD_ACCESS
+		c = load_unaligned_zeropad(src+res);
+#else
 		c = read_word_at_a_time(src+res);
+#endif
 		if (has_zero(c, &data, &constants)) {
 			data = prep_zero_mask(c, data, &constants);
 			data = create_zero_mask(data);
diff --git a/lib/test_ubsan.c b/lib/test_ubsan.c
index 8772e5edaa4f..a4b6f52b9c57 100644
--- a/lib/test_ubsan.c
+++ b/lib/test_ubsan.c
@@ -77,18 +77,22 @@ static void test_ubsan_shift_out_of_bounds(void)
 
 static void test_ubsan_out_of_bounds(void)
 {
-	volatile int i = 4, j = 5, k = -1;
-	volatile char above[4] = { }; /* Protect surrounding memory. */
-	volatile int arr[4];
-	volatile char below[4] = { }; /* Protect surrounding memory. */
+	int i = 4, j = 4, k = -1;
+	volatile struct {
+		char above[4]; /* Protect surrounding memory. */
+		int arr[4];
+		char below[4]; /* Protect surrounding memory. */
+	} data;
 
-	above[0] = below[0];
+	OPTIMIZER_HIDE_VAR(i);
+	OPTIMIZER_HIDE_VAR(j);
+	OPTIMIZER_HIDE_VAR(k);
 
 	UBSAN_TEST(CONFIG_UBSAN_BOUNDS, "above");
-	arr[j] = i;
+	data.arr[j] = i;
 
 	UBSAN_TEST(CONFIG_UBSAN_BOUNDS, "below");
-	arr[k] = i;
+	data.arr[k] = i;
 }
 
 enum ubsan_test_enum {
diff --git a/lib/tests/printf_kunit.c b/lib/tests/printf_kunit.c
index 2c9f6170bacd..bc54cca2d7a6 100644
--- a/lib/tests/printf_kunit.c
+++ b/lib/tests/printf_kunit.c
@@ -701,21 +701,46 @@ static void fwnode_pointer(struct kunit *kunittest)
 	software_node_unregister_node_group(group);
 }
 
+struct fourcc_struct {
+	u32 code;
+	const char *str;
+};
+
+static void fourcc_pointer_test(struct kunit *kunittest, const struct fourcc_struct *fc,
+				size_t n, const char *fmt)
+{
+	size_t i;
+
+	for (i = 0; i < n; i++)
+		test(fc[i].str, fmt, &fc[i].code);
+}
+
 static void fourcc_pointer(struct kunit *kunittest)
 {
-	struct {
-		u32 code;
-		char *str;
-	} const try[] = {
+	static const struct fourcc_struct try_cc[] = {
 		{ 0x3231564e, "NV12 little-endian (0x3231564e)", },
 		{ 0xb231564e, "NV12 big-endian (0xb231564e)", },
 		{ 0x10111213, ".... little-endian (0x10111213)", },
 		{ 0x20303159, "Y10  little-endian (0x20303159)", },
 	};
-	unsigned int i;
+	static const struct fourcc_struct try_ch[] = {
+		{ 0x41424344, "ABCD (0x41424344)", },
+	};
+	static const struct fourcc_struct try_chR[] = {
+		{ 0x41424344, "DCBA (0x44434241)", },
+	};
+	static const struct fourcc_struct try_cl[] = {
+		{ (__force u32)cpu_to_le32(0x41424344), "ABCD (0x41424344)", },
+	};
+	static const struct fourcc_struct try_cb[] = {
+		{ (__force u32)cpu_to_be32(0x41424344), "ABCD (0x41424344)", },
+	};
 
-	for (i = 0; i < ARRAY_SIZE(try); i++)
-		test(try[i].str, "%p4cc", &try[i].code);
+	fourcc_pointer_test(kunittest, try_cc, ARRAY_SIZE(try_cc), "%p4cc");
+	fourcc_pointer_test(kunittest, try_ch, ARRAY_SIZE(try_ch), "%p4ch");
+	fourcc_pointer_test(kunittest, try_chR, ARRAY_SIZE(try_chR), "%p4chR");
+	fourcc_pointer_test(kunittest, try_cl, ARRAY_SIZE(try_cl), "%p4cl");
+	fourcc_pointer_test(kunittest, try_cb, ARRAY_SIZE(try_cb), "%p4cb");
 }
 
 static void
diff --git a/lib/tests/slub_kunit.c b/lib/tests/slub_kunit.c
index d47c472b0520..848b682a2d70 100644
--- a/lib/tests/slub_kunit.c
+++ b/lib/tests/slub_kunit.c
@@ -325,4 +325,5 @@ static struct kunit_suite test_suite = {
 };
 kunit_test_suite(test_suite);
 
+MODULE_DESCRIPTION("Kunit tests for slub allocator");
 MODULE_LICENSE("GPL");
diff --git a/lib/ucs2_string.c b/lib/ucs2_string.c
index 9308bcfb2ad5..dfb4f2358cab 100644
--- a/lib/ucs2_string.c
+++ b/lib/ucs2_string.c
@@ -165,4 +165,5 @@ ucs2_as_utf8(u8 *dest, const ucs2_char_t *src, unsigned long maxlength)
 }
 EXPORT_SYMBOL(ucs2_as_utf8);
 
+MODULE_DESCRIPTION("UCS2 string handling");
 MODULE_LICENSE("GPL v2");
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index 01699852f30c..d8a2ec083ace 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -1793,27 +1793,49 @@ char *fourcc_string(char *buf, char *end, const u32 *fourcc,
 	char output[sizeof("0123 little-endian (0x01234567)")];
 	char *p = output;
 	unsigned int i;
+	bool pixel_fmt = false;
 	u32 orig, val;
 
-	if (fmt[1] != 'c' || fmt[2] != 'c')
+	if (fmt[1] != 'c')
 		return error_string(buf, end, "(%p4?)", spec);
 
 	if (check_pointer(&buf, end, fourcc, spec))
 		return buf;
 
 	orig = get_unaligned(fourcc);
-	val = orig & ~BIT(31);
+	switch (fmt[2]) {
+	case 'h':
+		if (fmt[3] == 'R')
+			orig = swab32(orig);
+		break;
+	case 'l':
+		orig = (__force u32)cpu_to_le32(orig);
+		break;
+	case 'b':
+		orig = (__force u32)cpu_to_be32(orig);
+		break;
+	case 'c':
+		/* Pixel formats are printed LSB-first */
+		pixel_fmt = true;
+		break;
+	default:
+		return error_string(buf, end, "(%p4?)", spec);
+	}
+
+	val = pixel_fmt ? swab32(orig & ~BIT(31)) : orig;
 
 	for (i = 0; i < sizeof(u32); i++) {
-		unsigned char c = val >> (i * 8);
+		unsigned char c = val >> ((3 - i) * 8);
 
 		/* Print non-control ASCII characters as-is, dot otherwise */
 		*p++ = isascii(c) && isprint(c) ? c : '.';
 	}
 
-	*p++ = ' ';
-	strcpy(p, orig & BIT(31) ? "big-endian" : "little-endian");
-	p += strlen(p);
+	if (pixel_fmt) {
+		*p++ = ' ';
+		strcpy(p, orig & BIT(31) ? "big-endian" : "little-endian");
+		p += strlen(p);
+	}
 
 	*p++ = ' ';
 	*p++ = '(';
@@ -2374,6 +2396,12 @@ early_param("no_hash_pointers", no_hash_pointers_enable);
  *       read the documentation (path below) first.
  * - 'NF' For a netdev_features_t
  * - '4cc' V4L2 or DRM FourCC code, with endianness and raw numerical value.
+ * - '4c[h[R]lb]' For generic FourCC code with raw numerical value. Both are
+ *	 displayed in the big-endian format. This is the opposite of V4L2 or
+ *	 DRM FourCCs.
+ *	 The additional specifiers define what endianness is used to load
+ *	 the stored bytes. The data might be interpreted using the host,
+ *	 reversed host byte order, little-endian, or big-endian.
  * - 'h[CDN]' For a variable-length buffer, it prints it as a hex string with
  *            a certain separator (' ' by default):
  *              C colon
diff --git a/lib/zlib_inflate/inflate_syms.c b/lib/zlib_inflate/inflate_syms.c
index 9720114c0672..b8996d90e8bc 100644
--- a/lib/zlib_inflate/inflate_syms.c
+++ b/lib/zlib_inflate/inflate_syms.c
@@ -18,4 +18,5 @@ EXPORT_SYMBOL(zlib_inflateEnd);
 EXPORT_SYMBOL(zlib_inflateReset);
 EXPORT_SYMBOL(zlib_inflateIncomp); 
 EXPORT_SYMBOL(zlib_inflate_blob);
+MODULE_DESCRIPTION("Data decompression using the deflation algorithm");
 MODULE_LICENSE("GPL");
diff --git a/mm/cma.c b/mm/cma.c
index b06d5fe73399..15632939f20a 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -35,7 +35,7 @@
 struct cma cma_areas[MAX_CMA_AREAS];
 unsigned int cma_area_count;
 
-static int __init __cma_declare_contiguous_nid(phys_addr_t base,
+static int __init __cma_declare_contiguous_nid(phys_addr_t *basep,
 			phys_addr_t size, phys_addr_t limit,
 			phys_addr_t alignment, unsigned int order_per_bit,
 			bool fixed, const char *name, struct cma **res_cma,
@@ -370,7 +370,7 @@ int __init cma_declare_contiguous_multi(phys_addr_t total_size,
 			phys_addr_t align, unsigned int order_per_bit,
 			const char *name, struct cma **res_cma, int nid)
 {
-	phys_addr_t start, end;
+	phys_addr_t start = 0, end;
 	phys_addr_t size, sizesum, sizeleft;
 	struct cma_init_memrange *mrp, *mlp, *failed;
 	struct cma_memrange *cmrp;
@@ -384,7 +384,7 @@ int __init cma_declare_contiguous_multi(phys_addr_t total_size,
 	/*
 	 * First, try it the normal way, producing just one range.
 	 */
-	ret = __cma_declare_contiguous_nid(0, total_size, 0, align,
+	ret = __cma_declare_contiguous_nid(&start, total_size, 0, align,
 			order_per_bit, false, name, res_cma, nid);
 	if (ret != -ENOMEM)
 		goto out;
@@ -580,7 +580,7 @@ int __init cma_declare_contiguous_nid(phys_addr_t base,
 {
 	int ret;
 
-	ret = __cma_declare_contiguous_nid(base, size, limit, alignment,
+	ret = __cma_declare_contiguous_nid(&base, size, limit, alignment,
 			order_per_bit, fixed, name, res_cma, nid);
 	if (ret != 0)
 		pr_err("Failed to reserve %ld MiB\n",
@@ -592,14 +592,14 @@ int __init cma_declare_contiguous_nid(phys_addr_t base,
 	return ret;
 }
 
-static int __init __cma_declare_contiguous_nid(phys_addr_t base,
+static int __init __cma_declare_contiguous_nid(phys_addr_t *basep,
 			phys_addr_t size, phys_addr_t limit,
 			phys_addr_t alignment, unsigned int order_per_bit,
 			bool fixed, const char *name, struct cma **res_cma,
 			int nid)
 {
 	phys_addr_t memblock_end = memblock_end_of_DRAM();
-	phys_addr_t highmem_start;
+	phys_addr_t highmem_start, base = *basep;
 	int ret;
 
 	/*
@@ -722,12 +722,15 @@ static int __init __cma_declare_contiguous_nid(phys_addr_t base,
 	}
 
 	ret = cma_init_reserved_mem(base, size, order_per_bit, name, res_cma);
-	if (ret)
+	if (ret) {
 		memblock_phys_free(base, size);
+		return ret;
+	}
 
 	(*res_cma)->nid = nid;
+	*basep = base;
 
-	return ret;
+	return 0;
 }
 
 static void cma_debug_show_areas(struct cma *cma)
diff --git a/mm/compaction.c b/mm/compaction.c
index 139f00c0308a..ca71fd3c3181 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -981,13 +981,13 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
 		}
 
 		if (PageHuge(page)) {
+			const unsigned int order = compound_order(page);
 			/*
 			 * skip hugetlbfs if we are not compacting for pages
 			 * bigger than its order. THPs and other compound pages
 			 * are handled below.
 			 */
 			if (!cc->alloc_contig) {
-				const unsigned int order = compound_order(page);
 
 				if (order <= MAX_PAGE_ORDER) {
 					low_pfn += (1UL << order) - 1;
@@ -1011,8 +1011,8 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
 				 /* Do not report -EBUSY down the chain */
 				if (ret == -EBUSY)
 					ret = 0;
-				low_pfn += compound_nr(page) - 1;
-				nr_scanned += compound_nr(page) - 1;
+				low_pfn += (1UL << order) - 1;
+				nr_scanned += (1UL << order) - 1;
 				goto isolate_fail;
 			}
 
diff --git a/mm/filemap.c b/mm/filemap.c
index b5e784f34d98..7b90cbeb4a1a 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2244,6 +2244,7 @@ unsigned filemap_get_folios_contig(struct address_space *mapping,
 			*start = folio->index + nr;
 			goto out;
 		}
+		xas_advance(&xas, folio_next_index(folio) - 1);
 		continue;
 put_folio:
 		folio_put(folio);
diff --git a/mm/gup.c b/mm/gup.c
index 92351e2fa876..84461d384ae2 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -2207,8 +2207,8 @@ size_t fault_in_safe_writeable(const char __user *uaddr, size_t size)
 	} while (start != end);
 	mmap_read_unlock(mm);
 
-	if (size > (unsigned long)uaddr - start)
-		return size - ((unsigned long)uaddr - start);
+	if (size > start - (unsigned long)uaddr)
+		return size - (start - (unsigned long)uaddr);
 	return 0;
 }
 EXPORT_SYMBOL(fault_in_safe_writeable);
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 39f92aad7bd1..e3e6ac991b9c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2271,7 +2271,7 @@ static struct folio *alloc_surplus_hugetlb_folio(struct hstate *h,
 	 * as surplus_pages, otherwise it might confuse
 	 * persistent_huge_pages() momentarily.
 	 */
-	__prep_account_new_huge_page(h, nid);
+	__prep_account_new_huge_page(h, folio_nid(folio));
 
 	/*
 	 * We could have raced with the pool size change.
@@ -3825,6 +3825,7 @@ found:
 static int set_max_huge_pages(struct hstate *h, unsigned long count, int nid,
 			      nodemask_t *nodes_allowed)
 {
+	unsigned long persistent_free_count;
 	unsigned long min_count;
 	unsigned long allocated;
 	struct folio *folio;
@@ -3959,8 +3960,24 @@ static int set_max_huge_pages(struct hstate *h, unsigned long count, int nid,
 	 * though, we'll note that we're not allowed to exceed surplus
 	 * and won't grow the pool anywhere else. Not until one of the
 	 * sysctls are changed, or the surplus pages go out of use.
+	 *
+	 * min_count is the expected number of persistent pages, we
+	 * shouldn't calculate min_count by using
+	 * resv_huge_pages + persistent_huge_pages() - free_huge_pages,
+	 * because there may exist free surplus huge pages, and this will
+	 * lead to subtracting twice. Free surplus huge pages come from HVO
+	 * failing to restore vmemmap, see comments in the callers of
+	 * hugetlb_vmemmap_restore_folio(). Thus, we should calculate
+	 * persistent free count first.
 	 */
-	min_count = h->resv_huge_pages + h->nr_huge_pages - h->free_huge_pages;
+	persistent_free_count = h->free_huge_pages;
+	if (h->free_huge_pages > persistent_huge_pages(h)) {
+		if (h->free_huge_pages > h->surplus_huge_pages)
+			persistent_free_count -= h->surplus_huge_pages;
+		else
+			persistent_free_count = 0;
+	}
+	min_count = h->resv_huge_pages + persistent_huge_pages(h) - persistent_free_count;
 	min_count = max(count, min_count);
 	try_to_free_low(h, min_count, nodes_allowed);
 
@@ -4630,7 +4647,7 @@ static void __init hugetlb_sysfs_init(void)
 		err = hugetlb_sysfs_add_hstate(h, hugepages_kobj,
 					 hstate_kobjs, &hstate_attr_group);
 		if (err)
-			pr_err("HugeTLB: Unable to add hstate %s", h->name);
+			pr_err("HugeTLB: Unable to add hstate %s\n", h->name);
 	}
 
 #ifdef CONFIG_NUMA
diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c
index 9a99dfa3c495..27245e86df25 100644
--- a/mm/hugetlb_vmemmap.c
+++ b/mm/hugetlb_vmemmap.c
@@ -238,11 +238,11 @@ static void vmemmap_remap_pte(pte_t *pte, unsigned long addr,
  * struct page, the special metadata (e.g. page->flags or page->mapping)
  * cannot copy to the tail struct page structs. The invalid value will be
  * checked in the free_tail_page_prepare(). In order to avoid the message
- * of "corrupted mapping in tail page". We need to reset at least 3 (one
- * head struct page struct and two tail struct page structs) struct page
+ * of "corrupted mapping in tail page". We need to reset at least 4 (one
+ * head struct page struct and three tail struct page structs) struct page
  * structs.
  */
-#define NR_RESET_STRUCT_PAGE		3
+#define NR_RESET_STRUCT_PAGE		4
 
 static inline void reset_struct_pages(struct page *start)
 {
diff --git a/mm/internal.h b/mm/internal.h
index 50c2f590b2d0..e9695baa5922 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -1595,6 +1595,7 @@ unsigned long move_page_tables(struct pagetable_move_control *pmc);
 
 #ifdef CONFIG_UNACCEPTED_MEMORY
 void accept_page(struct page *page);
+void unaccepted_cleanup_work(struct work_struct *work);
 #else /* CONFIG_UNACCEPTED_MEMORY */
 static inline void accept_page(struct page *page)
 {
diff --git a/mm/kasan/kasan_test_c.c b/mm/kasan/kasan_test_c.c
index 3ea317837c2d..5f922dd38ffa 100644
--- a/mm/kasan/kasan_test_c.c
+++ b/mm/kasan/kasan_test_c.c
@@ -1567,6 +1567,7 @@ static void kasan_memcmp(struct kunit *test)
 static void kasan_strings(struct kunit *test)
 {
 	char *ptr;
+	char *src;
 	size_t size = 24;
 
 	/*
@@ -1578,6 +1579,25 @@ static void kasan_strings(struct kunit *test)
 	ptr = kmalloc(size, GFP_KERNEL | __GFP_ZERO);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 
+	src = kmalloc(KASAN_GRANULE_SIZE, GFP_KERNEL | __GFP_ZERO);
+	strscpy(src, "f0cacc1a0000000", KASAN_GRANULE_SIZE);
+
+	/*
+	 * Make sure that strscpy() does not trigger KASAN if it overreads into
+	 * poisoned memory.
+	 *
+	 * The expected size does not include the terminator '\0'
+	 * so it is (KASAN_GRANULE_SIZE - 2) ==
+	 * KASAN_GRANULE_SIZE - ("initial removed character" + "\0").
+	 */
+	KUNIT_EXPECT_EQ(test, KASAN_GRANULE_SIZE - 2,
+			strscpy(ptr, src + 1, KASAN_GRANULE_SIZE));
+
+	/* strscpy should fail if the first byte is unreadable. */
+	KUNIT_EXPECT_KASAN_FAIL(test, strscpy(ptr, src + KASAN_GRANULE_SIZE,
+					      KASAN_GRANULE_SIZE));
+
+	kfree(src);
 	kfree(ptr);
 
 	/*
@@ -2127,4 +2147,5 @@ static struct kunit_suite kasan_kunit_test_suite = {
 
 kunit_test_suite(kasan_kunit_test_suite);
 
+MODULE_DESCRIPTION("KUnit tests for checking KASAN bug-detection capabilities");
 MODULE_LICENSE("GPL");
diff --git a/mm/memblock.c b/mm/memblock.c
index 0a53db4d9f7b..d3509414b8c3 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -2183,11 +2183,14 @@ static void __init memmap_init_reserved_pages(void)
 	struct memblock_region *region;
 	phys_addr_t start, end;
 	int nid;
+	unsigned long max_reserved;
 
 	/*
 	 * set nid on all reserved pages and also treat struct
 	 * pages for the NOMAP regions as PageReserved
 	 */
+repeat:
+	max_reserved = memblock.reserved.max;
 	for_each_mem_region(region) {
 		nid = memblock_get_region_node(region);
 		start = region->base;
@@ -2196,8 +2199,15 @@ static void __init memmap_init_reserved_pages(void)
 		if (memblock_is_nomap(region))
 			reserve_bootmem_region(start, end, nid);
 
-		memblock_set_node(start, end, &memblock.reserved, nid);
+		memblock_set_node(start, region->size, &memblock.reserved, nid);
 	}
+	/*
+	 * 'max' is changed means memblock.reserved has been doubled its
+	 * array, which may result a new reserved region before current
+	 * 'start'. Now we should repeat the procedure to set its node id.
+	 */
+	if (max_reserved != memblock.reserved.max)
+		goto repeat;
 
 	/*
 	 * initialize struct pages for reserved regions that don't have
diff --git a/mm/memcontrol-v1.c b/mm/memcontrol-v1.c
index 8660908850dc..4a9cf27a70af 100644
--- a/mm/memcontrol-v1.c
+++ b/mm/memcontrol-v1.c
@@ -620,7 +620,7 @@ void memcg1_swapout(struct folio *folio, swp_entry_t entry)
 		mem_cgroup_id_get_many(swap_memcg, nr_entries - 1);
 	mod_memcg_state(swap_memcg, MEMCG_SWAP, nr_entries);
 
-	swap_cgroup_record(folio, mem_cgroup_id(memcg), entry);
+	swap_cgroup_record(folio, mem_cgroup_id(swap_memcg), entry);
 
 	folio_unqueue_deferred_split(folio);
 	folio->memcg_data = 0;
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 421740f1bcdc..c96c1f2b9cf5 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1759,7 +1759,7 @@ void mem_cgroup_print_oom_group(struct mem_cgroup *memcg)
 }
 
 struct memcg_stock_pcp {
-	localtry_lock_t stock_lock;
+	local_trylock_t stock_lock;
 	struct mem_cgroup *cached; /* this never be root cgroup */
 	unsigned int nr_pages;
 
@@ -1774,7 +1774,7 @@ struct memcg_stock_pcp {
 #define FLUSHING_CACHED_CHARGE	0
 };
 static DEFINE_PER_CPU(struct memcg_stock_pcp, memcg_stock) = {
-	.stock_lock = INIT_LOCALTRY_LOCK(stock_lock),
+	.stock_lock = INIT_LOCAL_TRYLOCK(stock_lock),
 };
 static DEFINE_MUTEX(percpu_charge_mutex);
 
@@ -1805,11 +1805,10 @@ static bool consume_stock(struct mem_cgroup *memcg, unsigned int nr_pages,
 	if (nr_pages > MEMCG_CHARGE_BATCH)
 		return ret;
 
-	if (!localtry_trylock_irqsave(&memcg_stock.stock_lock, flags)) {
-		if (!gfpflags_allow_spinning(gfp_mask))
-			return ret;
-		localtry_lock_irqsave(&memcg_stock.stock_lock, flags);
-	}
+	if (gfpflags_allow_spinning(gfp_mask))
+		local_lock_irqsave(&memcg_stock.stock_lock, flags);
+	else if (!local_trylock_irqsave(&memcg_stock.stock_lock, flags))
+		return ret;
 
 	stock = this_cpu_ptr(&memcg_stock);
 	stock_pages = READ_ONCE(stock->nr_pages);
@@ -1818,7 +1817,7 @@ static bool consume_stock(struct mem_cgroup *memcg, unsigned int nr_pages,
 		ret = true;
 	}
 
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 
 	return ret;
 }
@@ -1857,14 +1856,14 @@ static void drain_local_stock(struct work_struct *dummy)
 	 * drain_stock races is that we always operate on local CPU stock
 	 * here with IRQ disabled
 	 */
-	localtry_lock_irqsave(&memcg_stock.stock_lock, flags);
+	local_lock_irqsave(&memcg_stock.stock_lock, flags);
 
 	stock = this_cpu_ptr(&memcg_stock);
 	old = drain_obj_stock(stock);
 	drain_stock(stock);
 	clear_bit(FLUSHING_CACHED_CHARGE, &stock->flags);
 
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 	obj_cgroup_put(old);
 }
 
@@ -1894,7 +1893,7 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
 {
 	unsigned long flags;
 
-	if (!localtry_trylock_irqsave(&memcg_stock.stock_lock, flags)) {
+	if (!local_trylock_irqsave(&memcg_stock.stock_lock, flags)) {
 		/*
 		 * In case of unlikely failure to lock percpu stock_lock
 		 * uncharge memcg directly.
@@ -1907,7 +1906,7 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
 		return;
 	}
 	__refill_stock(memcg, nr_pages);
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 }
 
 /*
@@ -1964,9 +1963,9 @@ static int memcg_hotplug_cpu_dead(unsigned int cpu)
 	stock = &per_cpu(memcg_stock, cpu);
 
 	/* drain_obj_stock requires stock_lock */
-	localtry_lock_irqsave(&memcg_stock.stock_lock, flags);
+	local_lock_irqsave(&memcg_stock.stock_lock, flags);
 	old = drain_obj_stock(stock);
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 
 	drain_stock(stock);
 	obj_cgroup_put(old);
@@ -2787,7 +2786,7 @@ static void mod_objcg_state(struct obj_cgroup *objcg, struct pglist_data *pgdat,
 	unsigned long flags;
 	int *bytes;
 
-	localtry_lock_irqsave(&memcg_stock.stock_lock, flags);
+	local_lock_irqsave(&memcg_stock.stock_lock, flags);
 	stock = this_cpu_ptr(&memcg_stock);
 
 	/*
@@ -2836,7 +2835,7 @@ static void mod_objcg_state(struct obj_cgroup *objcg, struct pglist_data *pgdat,
 	if (nr)
 		__mod_objcg_mlstate(objcg, pgdat, idx, nr);
 
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 	obj_cgroup_put(old);
 }
 
@@ -2846,7 +2845,7 @@ static bool consume_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes)
 	unsigned long flags;
 	bool ret = false;
 
-	localtry_lock_irqsave(&memcg_stock.stock_lock, flags);
+	local_lock_irqsave(&memcg_stock.stock_lock, flags);
 
 	stock = this_cpu_ptr(&memcg_stock);
 	if (objcg == READ_ONCE(stock->cached_objcg) && stock->nr_bytes >= nr_bytes) {
@@ -2854,7 +2853,7 @@ static bool consume_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes)
 		ret = true;
 	}
 
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 
 	return ret;
 }
@@ -2946,7 +2945,7 @@ static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes,
 	unsigned long flags;
 	unsigned int nr_pages = 0;
 
-	localtry_lock_irqsave(&memcg_stock.stock_lock, flags);
+	local_lock_irqsave(&memcg_stock.stock_lock, flags);
 
 	stock = this_cpu_ptr(&memcg_stock);
 	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
@@ -2960,7 +2959,7 @@ static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes,
 		stock->nr_bytes &= (PAGE_SIZE - 1);
 	}
 
-	localtry_unlock_irqrestore(&memcg_stock.stock_lock, flags);
+	local_unlock_irqrestore(&memcg_stock.stock_lock, flags);
 	obj_cgroup_put(old);
 
 	if (nr_pages)
diff --git a/mm/memory.c b/mm/memory.c
index 2d8c265fc7d6..ba3ea0a82f7f 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1361,7 +1361,7 @@ copy_page_range(struct vm_area_struct *dst_vma, struct vm_area_struct *src_vma)
 	struct mm_struct *dst_mm = dst_vma->vm_mm;
 	struct mm_struct *src_mm = src_vma->vm_mm;
 	struct mmu_notifier_range range;
-	unsigned long next, pfn;
+	unsigned long next, pfn = 0;
 	bool is_cow;
 	int ret;
 
@@ -2938,11 +2938,11 @@ static int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
 	if (fn) {
 		do {
 			if (create || !pte_none(ptep_get(pte))) {
-				err = fn(pte++, addr, data);
+				err = fn(pte, addr, data);
 				if (err)
 					break;
 			}
-		} while (addr += PAGE_SIZE, addr != end);
+		} while (pte++, addr += PAGE_SIZE, addr != end);
 	}
 	*mask |= PGTBL_PTE_MODIFIED;
 
@@ -3734,8 +3734,6 @@ static bool __wp_can_reuse_large_anon_folio(struct folio *folio,
 		return false;
 
 	VM_WARN_ON_ONCE(folio_test_ksm(folio));
-	VM_WARN_ON_ONCE(folio_mapcount(folio) > folio_nr_pages(folio));
-	VM_WARN_ON_ONCE(folio_entire_mapcount(folio));
 
 	if (unlikely(folio_test_swapcache(folio))) {
 		/*
@@ -3760,6 +3758,8 @@ static bool __wp_can_reuse_large_anon_folio(struct folio *folio,
 	if (folio_large_mapcount(folio) != folio_ref_count(folio))
 		goto unlock;
 
+	VM_WARN_ON_ONCE_FOLIO(folio_large_mapcount(folio) > folio_nr_pages(folio), folio);
+	VM_WARN_ON_ONCE_FOLIO(folio_entire_mapcount(folio), folio);
 	VM_WARN_ON_ONCE(folio_mm_id(folio, 0) != vma->vm_mm->mm_id &&
 			folio_mm_id(folio, 1) != vma->vm_mm->mm_id);
 
diff --git a/mm/migrate.c b/mm/migrate.c
index f3ee6d8d5e2e..676d9cfc7059 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -845,9 +845,11 @@ static int __buffer_migrate_folio(struct address_space *mapping,
 		return -EAGAIN;
 
 	if (check_refs) {
-		bool busy;
+		bool busy, migrating;
 		bool invalidated = false;
 
+		migrating = test_and_set_bit_lock(BH_Migrate, &head->b_state);
+		VM_WARN_ON_ONCE(migrating);
 recheck_buffers:
 		busy = false;
 		spin_lock(&mapping->i_private_lock);
@@ -859,12 +861,12 @@ recheck_buffers:
 			}
 			bh = bh->b_this_page;
 		} while (bh != head);
+		spin_unlock(&mapping->i_private_lock);
 		if (busy) {
 			if (invalidated) {
 				rc = -EAGAIN;
 				goto unlock_buffers;
 			}
-			spin_unlock(&mapping->i_private_lock);
 			invalidate_bh_lrus();
 			invalidated = true;
 			goto recheck_buffers;
@@ -883,7 +885,7 @@ recheck_buffers:
 
 unlock_buffers:
 	if (check_refs)
-		spin_unlock(&mapping->i_private_lock);
+		clear_bit_unlock(BH_Migrate, &head->b_state);
 	bh = head;
 	do {
 		unlock_buffer(bh);
diff --git a/mm/mm_init.c b/mm/mm_init.c
index 84f14fa12d0d..9659689b8ace 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1441,6 +1441,7 @@ static void __meminit zone_init_free_lists(struct zone *zone)
 
 #ifdef CONFIG_UNACCEPTED_MEMORY
 	INIT_LIST_HEAD(&zone->unaccepted_pages);
+	INIT_WORK(&zone->unaccepted_cleanup, unaccepted_cleanup_work);
 #endif
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index fd6b865cb1ab..5669baf2a6fe 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1400,11 +1400,12 @@ static void free_one_page(struct zone *zone, struct page *page,
 	struct llist_head *llhead;
 	unsigned long flags;
 
-	if (!spin_trylock_irqsave(&zone->lock, flags)) {
-		if (unlikely(fpi_flags & FPI_TRYLOCK)) {
+	if (unlikely(fpi_flags & FPI_TRYLOCK)) {
+		if (!spin_trylock_irqsave(&zone->lock, flags)) {
 			add_page_to_zone_llist(zone, page, order);
 			return;
 		}
+	} else {
 		spin_lock_irqsave(&zone->lock, flags);
 	}
 
@@ -2182,23 +2183,15 @@ try_to_claim_block(struct zone *zone, struct page *page,
 }
 
 /*
- * Try finding a free buddy page on the fallback list.
- *
- * This will attempt to claim a whole pageblock for the requested type
- * to ensure grouping of such requests in the future.
- *
- * If a whole block cannot be claimed, steal an individual page, regressing to
- * __rmqueue_smallest() logic to at least break up as little contiguity as
- * possible.
+ * Try to allocate from some fallback migratetype by claiming the entire block,
+ * i.e. converting it to the allocation's start migratetype.
  *
  * The use of signed ints for order and current_order is a deliberate
  * deviation from the rest of this file, to make the for loop
  * condition simpler.
- *
- * Return the stolen page, or NULL if none can be found.
  */
 static __always_inline struct page *
-__rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
+__rmqueue_claim(struct zone *zone, int order, int start_migratetype,
 						unsigned int alloc_flags)
 {
 	struct free_area *area;
@@ -2236,14 +2229,29 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
 		page = try_to_claim_block(zone, page, current_order, order,
 					  start_migratetype, fallback_mt,
 					  alloc_flags);
-		if (page)
-			goto got_one;
+		if (page) {
+			trace_mm_page_alloc_extfrag(page, order, current_order,
+						    start_migratetype, fallback_mt);
+			return page;
+		}
 	}
 
-	if (alloc_flags & ALLOC_NOFRAGMENT)
-		return NULL;
+	return NULL;
+}
+
+/*
+ * Try to steal a single page from some fallback migratetype. Leave the rest of
+ * the block as its current migratetype, potentially causing fragmentation.
+ */
+static __always_inline struct page *
+__rmqueue_steal(struct zone *zone, int order, int start_migratetype)
+{
+	struct free_area *area;
+	int current_order;
+	struct page *page;
+	int fallback_mt;
+	bool claim_block;
 
-	/* No luck claiming pageblock. Find the smallest fallback page */
 	for (current_order = order; current_order < NR_PAGE_ORDERS; current_order++) {
 		area = &(zone->free_area[current_order]);
 		fallback_mt = find_suitable_fallback(area, current_order,
@@ -2253,25 +2261,28 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
 
 		page = get_page_from_free_area(area, fallback_mt);
 		page_del_and_expand(zone, page, order, current_order, fallback_mt);
-		goto got_one;
+		trace_mm_page_alloc_extfrag(page, order, current_order,
+					    start_migratetype, fallback_mt);
+		return page;
 	}
 
 	return NULL;
-
-got_one:
-	trace_mm_page_alloc_extfrag(page, order, current_order,
-		start_migratetype, fallback_mt);
-
-	return page;
 }
 
+enum rmqueue_mode {
+	RMQUEUE_NORMAL,
+	RMQUEUE_CMA,
+	RMQUEUE_CLAIM,
+	RMQUEUE_STEAL,
+};
+
 /*
  * Do the hard work of removing an element from the buddy allocator.
  * Call me with the zone->lock already held.
  */
 static __always_inline struct page *
 __rmqueue(struct zone *zone, unsigned int order, int migratetype,
-						unsigned int alloc_flags)
+	  unsigned int alloc_flags, enum rmqueue_mode *mode)
 {
 	struct page *page;
 
@@ -2290,16 +2301,48 @@ __rmqueue(struct zone *zone, unsigned int order, int migratetype,
 		}
 	}
 
-	page = __rmqueue_smallest(zone, order, migratetype);
-	if (unlikely(!page)) {
-		if (alloc_flags & ALLOC_CMA)
+	/*
+	 * First try the freelists of the requested migratetype, then try
+	 * fallbacks modes with increasing levels of fragmentation risk.
+	 *
+	 * The fallback logic is expensive and rmqueue_bulk() calls in
+	 * a loop with the zone->lock held, meaning the freelists are
+	 * not subject to any outside changes. Remember in *mode where
+	 * we found pay dirt, to save us the search on the next call.
+	 */
+	switch (*mode) {
+	case RMQUEUE_NORMAL:
+		page = __rmqueue_smallest(zone, order, migratetype);
+		if (page)
+			return page;
+		fallthrough;
+	case RMQUEUE_CMA:
+		if (alloc_flags & ALLOC_CMA) {
 			page = __rmqueue_cma_fallback(zone, order);
-
-		if (!page)
-			page = __rmqueue_fallback(zone, order, migratetype,
-						  alloc_flags);
+			if (page) {
+				*mode = RMQUEUE_CMA;
+				return page;
+			}
+		}
+		fallthrough;
+	case RMQUEUE_CLAIM:
+		page = __rmqueue_claim(zone, order, migratetype, alloc_flags);
+		if (page) {
+			/* Replenished preferred freelist, back to normal mode. */
+			*mode = RMQUEUE_NORMAL;
+			return page;
+		}
+		fallthrough;
+	case RMQUEUE_STEAL:
+		if (!(alloc_flags & ALLOC_NOFRAGMENT)) {
+			page = __rmqueue_steal(zone, order, migratetype);
+			if (page) {
+				*mode = RMQUEUE_STEAL;
+				return page;
+			}
+		}
 	}
-	return page;
+	return NULL;
 }
 
 /*
@@ -2311,17 +2354,19 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
 			unsigned long count, struct list_head *list,
 			int migratetype, unsigned int alloc_flags)
 {
+	enum rmqueue_mode rmqm = RMQUEUE_NORMAL;
 	unsigned long flags;
 	int i;
 
-	if (!spin_trylock_irqsave(&zone->lock, flags)) {
-		if (unlikely(alloc_flags & ALLOC_TRYLOCK))
+	if (unlikely(alloc_flags & ALLOC_TRYLOCK)) {
+		if (!spin_trylock_irqsave(&zone->lock, flags))
 			return 0;
+	} else {
 		spin_lock_irqsave(&zone->lock, flags);
 	}
 	for (i = 0; i < count; ++i) {
 		struct page *page = __rmqueue(zone, order, migratetype,
-								alloc_flags);
+					      alloc_flags, &rmqm);
 		if (unlikely(page == NULL))
 			break;
 
@@ -2937,15 +2982,18 @@ struct page *rmqueue_buddy(struct zone *preferred_zone, struct zone *zone,
 
 	do {
 		page = NULL;
-		if (!spin_trylock_irqsave(&zone->lock, flags)) {
-			if (unlikely(alloc_flags & ALLOC_TRYLOCK))
+		if (unlikely(alloc_flags & ALLOC_TRYLOCK)) {
+			if (!spin_trylock_irqsave(&zone->lock, flags))
 				return NULL;
+		} else {
 			spin_lock_irqsave(&zone->lock, flags);
 		}
 		if (alloc_flags & ALLOC_HIGHATOMIC)
 			page = __rmqueue_smallest(zone, order, MIGRATE_HIGHATOMIC);
 		if (!page) {
-			page = __rmqueue(zone, order, migratetype, alloc_flags);
+			enum rmqueue_mode rmqm = RMQUEUE_NORMAL;
+
+			page = __rmqueue(zone, order, migratetype, alloc_flags, &rmqm);
 
 			/*
 			 * If the allocation fails, allow OOM handling and
@@ -3422,18 +3470,6 @@ static inline bool zone_watermark_fast(struct zone *z, unsigned int order,
 	return false;
 }
 
-bool zone_watermark_ok_safe(struct zone *z, unsigned int order,
-			unsigned long mark, int highest_zoneidx)
-{
-	long free_pages = zone_page_state(z, NR_FREE_PAGES);
-
-	if (z->percpu_drift_mark && free_pages < z->percpu_drift_mark)
-		free_pages = zone_page_state_snapshot(z, NR_FREE_PAGES);
-
-	return __zone_watermark_ok(z, order, mark, highest_zoneidx, 0,
-								free_pages);
-}
-
 #ifdef CONFIG_NUMA
 int __read_mostly node_reclaim_distance = RECLAIM_DISTANCE;
 
@@ -7143,6 +7179,11 @@ static DEFINE_STATIC_KEY_FALSE(zones_with_unaccepted_pages);
 
 static bool lazy_accept = true;
 
+void unaccepted_cleanup_work(struct work_struct *work)
+{
+	static_branch_dec(&zones_with_unaccepted_pages);
+}
+
 static int __init accept_memory_parse(char *p)
 {
 	if (!strcmp(p, "lazy")) {
@@ -7181,8 +7222,27 @@ static void __accept_page(struct zone *zone, unsigned long *flags,
 
 	__free_pages_ok(page, MAX_PAGE_ORDER, FPI_TO_TAIL);
 
-	if (last)
-		static_branch_dec(&zones_with_unaccepted_pages);
+	if (last) {
+		/*
+		 * There are two corner cases:
+		 *
+		 * - If allocation occurs during the CPU bring up,
+		 *   static_branch_dec() cannot be used directly as
+		 *   it causes a deadlock on cpu_hotplug_lock.
+		 *
+		 *   Instead, use schedule_work() to prevent deadlock.
+		 *
+		 * - If allocation occurs before workqueues are initialized,
+		 *   static_branch_dec() should be called directly.
+		 *
+		 *   Workqueues are initialized before CPU bring up, so this
+		 *   will not conflict with the first scenario.
+		 */
+		if (system_wq)
+			schedule_work(&zone->unaccepted_cleanup);
+		else
+			unaccepted_cleanup_work(&zone->unaccepted_cleanup);
+	}
 }
 
 void accept_page(struct page *page)
diff --git a/mm/slub.c b/mm/slub.c
index b46f87662e71..be8b09e09d30 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1973,6 +1973,11 @@ static inline void handle_failed_objexts_alloc(unsigned long obj_exts,
 #define OBJCGS_CLEAR_MASK	(__GFP_DMA | __GFP_RECLAIMABLE | \
 				__GFP_ACCOUNT | __GFP_NOFAIL)
 
+static inline void init_slab_obj_exts(struct slab *slab)
+{
+	slab->obj_exts = 0;
+}
+
 int alloc_slab_obj_exts(struct slab *slab, struct kmem_cache *s,
 		        gfp_t gfp, bool new_slab)
 {
@@ -2023,8 +2028,7 @@ int alloc_slab_obj_exts(struct slab *slab, struct kmem_cache *s,
 	return 0;
 }
 
-/* Should be called only if mem_alloc_profiling_enabled() */
-static noinline void free_slab_obj_exts(struct slab *slab)
+static inline void free_slab_obj_exts(struct slab *slab)
 {
 	struct slabobj_ext *obj_exts;
 
@@ -2044,20 +2048,12 @@ static noinline void free_slab_obj_exts(struct slab *slab)
 	slab->obj_exts = 0;
 }
 
-static inline bool need_slab_obj_ext(void)
-{
-	if (mem_alloc_profiling_enabled())
-		return true;
+#else /* CONFIG_SLAB_OBJ_EXT */
 
-	/*
-	 * CONFIG_MEMCG creates vector of obj_cgroup objects conditionally
-	 * inside memcg_slab_post_alloc_hook. No other users for now.
-	 */
-	return false;
+static inline void init_slab_obj_exts(struct slab *slab)
+{
 }
 
-#else /* CONFIG_SLAB_OBJ_EXT */
-
 static int alloc_slab_obj_exts(struct slab *slab, struct kmem_cache *s,
 			       gfp_t gfp, bool new_slab)
 {
@@ -2068,11 +2064,6 @@ static inline void free_slab_obj_exts(struct slab *slab)
 {
 }
 
-static inline bool need_slab_obj_ext(void)
-{
-	return false;
-}
-
 #endif /* CONFIG_SLAB_OBJ_EXT */
 
 #ifdef CONFIG_MEM_ALLOC_PROFILING
@@ -2120,7 +2111,7 @@ __alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
 static inline void
 alloc_tagging_slab_alloc_hook(struct kmem_cache *s, void *object, gfp_t flags)
 {
-	if (need_slab_obj_ext())
+	if (mem_alloc_profiling_enabled())
 		__alloc_tagging_slab_alloc_hook(s, object, flags);
 }
 
@@ -2592,8 +2583,12 @@ static __always_inline void account_slab(struct slab *slab, int order,
 static __always_inline void unaccount_slab(struct slab *slab, int order,
 					   struct kmem_cache *s)
 {
-	if (memcg_kmem_online() || need_slab_obj_ext())
-		free_slab_obj_exts(slab);
+	/*
+	 * The slab object extensions should now be freed regardless of
+	 * whether mem_alloc_profiling_enabled() or not because profiling
+	 * might have been disabled after slab->obj_exts got allocated.
+	 */
+	free_slab_obj_exts(slab);
 
 	mod_node_page_state(slab_pgdat(slab), cache_vmstat_idx(s),
 			    -(PAGE_SIZE << order));
@@ -2637,6 +2632,7 @@ static struct slab *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
 	slab->objects = oo_objects(oo);
 	slab->inuse = 0;
 	slab->frozen = 0;
+	init_slab_obj_exts(slab);
 
 	account_slab(slab, oo_order(oo), s, flags);
 
diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c
index fbf2cf62ab9f..7d5d709cc838 100644
--- a/mm/userfaultfd.c
+++ b/mm/userfaultfd.c
@@ -1902,6 +1902,14 @@ struct vm_area_struct *userfaultfd_clear_vma(struct vma_iterator *vmi,
 					     unsigned long end)
 {
 	struct vm_area_struct *ret;
+	bool give_up_on_oom = false;
+
+	/*
+	 * If we are modifying only and not splitting, just give up on the merge
+	 * if OOM prevents us from merging successfully.
+	 */
+	if (start == vma->vm_start && end == vma->vm_end)
+		give_up_on_oom = true;
 
 	/* Reset ptes for the whole vma range if wr-protected */
 	if (userfaultfd_wp(vma))
@@ -1909,7 +1917,7 @@ struct vm_area_struct *userfaultfd_clear_vma(struct vma_iterator *vmi,
 
 	ret = vma_modify_flags_uffd(vmi, prev, vma, start, end,
 				    vma->vm_flags & ~__VM_UFFD_FLAGS,
-				    NULL_VM_UFFD_CTX);
+				    NULL_VM_UFFD_CTX, give_up_on_oom);
 
 	/*
 	 * In the vma_merge() successful mprotect-like case 8:
@@ -1960,7 +1968,8 @@ int userfaultfd_register_range(struct userfaultfd_ctx *ctx,
 		new_flags = (vma->vm_flags & ~__VM_UFFD_FLAGS) | vm_flags;
 		vma = vma_modify_flags_uffd(&vmi, prev, vma, start, vma_end,
 					    new_flags,
-					    (struct vm_userfaultfd_ctx){ctx});
+					    (struct vm_userfaultfd_ctx){ctx},
+					    /* give_up_on_oom = */false);
 		if (IS_ERR(vma))
 			return PTR_ERR(vma);
 
diff --git a/mm/vma.c b/mm/vma.c
index 5cdc5612bfc1..839d12f02c88 100644
--- a/mm/vma.c
+++ b/mm/vma.c
@@ -666,6 +666,9 @@ static void vmg_adjust_set_range(struct vma_merge_struct *vmg)
 /*
  * Actually perform the VMA merge operation.
  *
+ * IMPORTANT: We guarantee that, should vmg->give_up_on_oom is set, to not
+ * modify any VMAs or cause inconsistent state should an OOM condition arise.
+ *
  * Returns 0 on success, or an error value on failure.
  */
 static int commit_merge(struct vma_merge_struct *vmg)
@@ -685,6 +688,12 @@ static int commit_merge(struct vma_merge_struct *vmg)
 
 	init_multi_vma_prep(&vp, vma, vmg);
 
+	/*
+	 * If vmg->give_up_on_oom is set, we're safe, because we don't actually
+	 * manipulate any VMAs until we succeed at preallocation.
+	 *
+	 * Past this point, we will not return an error.
+	 */
 	if (vma_iter_prealloc(vmg->vmi, vma))
 		return -ENOMEM;
 
@@ -915,7 +924,13 @@ static __must_check struct vm_area_struct *vma_merge_existing_range(
 		if (anon_dup)
 			unlink_anon_vmas(anon_dup);
 
-		vmg->state = VMA_MERGE_ERROR_NOMEM;
+		/*
+		 * We've cleaned up any cloned anon_vma's, no VMAs have been
+		 * modified, no harm no foul if the user requests that we not
+		 * report this and just give up, leaving the VMAs unmerged.
+		 */
+		if (!vmg->give_up_on_oom)
+			vmg->state = VMA_MERGE_ERROR_NOMEM;
 		return NULL;
 	}
 
@@ -926,7 +941,15 @@ static __must_check struct vm_area_struct *vma_merge_existing_range(
 abort:
 	vma_iter_set(vmg->vmi, start);
 	vma_iter_load(vmg->vmi);
-	vmg->state = VMA_MERGE_ERROR_NOMEM;
+
+	/*
+	 * This means we have failed to clone anon_vma's correctly, but no
+	 * actual changes to VMAs have occurred, so no harm no foul - if the
+	 * user doesn't want this reported and instead just wants to give up on
+	 * the merge, allow it.
+	 */
+	if (!vmg->give_up_on_oom)
+		vmg->state = VMA_MERGE_ERROR_NOMEM;
 	return NULL;
 }
 
@@ -1068,6 +1091,10 @@ int vma_expand(struct vma_merge_struct *vmg)
 		/* This should already have been checked by this point. */
 		VM_WARN_ON_VMG(!can_merge_remove_vma(next), vmg);
 		vma_start_write(next);
+		/*
+		 * In this case we don't report OOM, so vmg->give_up_on_mm is
+		 * safe.
+		 */
 		ret = dup_anon_vma(middle, next, &anon_dup);
 		if (ret)
 			return ret;
@@ -1090,9 +1117,15 @@ int vma_expand(struct vma_merge_struct *vmg)
 	return 0;
 
 nomem:
-	vmg->state = VMA_MERGE_ERROR_NOMEM;
 	if (anon_dup)
 		unlink_anon_vmas(anon_dup);
+	/*
+	 * If the user requests that we just give upon OOM, we are safe to do so
+	 * here, as commit merge provides this contract to us. Nothing has been
+	 * changed - no harm no foul, just don't report it.
+	 */
+	if (!vmg->give_up_on_oom)
+		vmg->state = VMA_MERGE_ERROR_NOMEM;
 	return -ENOMEM;
 }
 
@@ -1534,6 +1567,13 @@ static struct vm_area_struct *vma_modify(struct vma_merge_struct *vmg)
 	if (vmg_nomem(vmg))
 		return ERR_PTR(-ENOMEM);
 
+	/*
+	 * Split can fail for reasons other than OOM, so if the user requests
+	 * this it's probably a mistake.
+	 */
+	VM_WARN_ON(vmg->give_up_on_oom &&
+		   (vma->vm_start != start || vma->vm_end != end));
+
 	/* Split any preceding portion of the VMA. */
 	if (vma->vm_start < start) {
 		int err = split_vma(vmg->vmi, vma, start, 1);
@@ -1602,12 +1642,15 @@ struct vm_area_struct
 		       struct vm_area_struct *vma,
 		       unsigned long start, unsigned long end,
 		       unsigned long new_flags,
-		       struct vm_userfaultfd_ctx new_ctx)
+		       struct vm_userfaultfd_ctx new_ctx,
+		       bool give_up_on_oom)
 {
 	VMG_VMA_STATE(vmg, vmi, prev, vma, start, end);
 
 	vmg.flags = new_flags;
 	vmg.uffd_ctx = new_ctx;
+	if (give_up_on_oom)
+		vmg.give_up_on_oom = true;
 
 	return vma_modify(&vmg);
 }
diff --git a/mm/vma.h b/mm/vma.h
index 7356ca5a22d3..149926e8a6d1 100644
--- a/mm/vma.h
+++ b/mm/vma.h
@@ -114,6 +114,12 @@ struct vma_merge_struct {
 	 */
 	bool just_expand :1;
 
+	/*
+	 * If a merge is possible, but an OOM error occurs, give up and don't
+	 * execute the merge, returning NULL.
+	 */
+	bool give_up_on_oom :1;
+
 	/* Internal flags set during merge process: */
 
 	/*
@@ -255,7 +261,8 @@ __must_check struct vm_area_struct
 		       struct vm_area_struct *vma,
 		       unsigned long start, unsigned long end,
 		       unsigned long new_flags,
-		       struct vm_userfaultfd_ctx new_ctx);
+		       struct vm_userfaultfd_ctx new_ctx,
+		       bool give_up_on_oom);
 
 __must_check struct vm_area_struct
 *vma_merge_new_range(struct vma_merge_struct *vmg);
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b620d74b0f66..3783e45bfc92 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -6736,6 +6736,7 @@ static bool pgdat_balanced(pg_data_t *pgdat, int order, int highest_zoneidx)
 	 * meet watermarks.
 	 */
 	for_each_managed_zone_pgdat(zone, pgdat, i, highest_zoneidx) {
+		enum zone_stat_item item;
 		unsigned long free_pages;
 
 		if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING)
@@ -6746,11 +6747,33 @@ static bool pgdat_balanced(pg_data_t *pgdat, int order, int highest_zoneidx)
 		/*
 		 * In defrag_mode, watermarks must be met in whole
 		 * blocks to avoid polluting allocator fallbacks.
+		 *
+		 * However, kswapd usually cannot accomplish this on
+		 * its own and needs kcompactd support. Once it's
+		 * reclaimed a compaction gap, and kswapd_shrink_node
+		 * has dropped order, simply ensure there are enough
+		 * base pages for compaction, wake kcompactd & sleep.
 		 */
-		if (defrag_mode)
-			free_pages = zone_page_state(zone, NR_FREE_PAGES_BLOCKS);
+		if (defrag_mode && order)
+			item = NR_FREE_PAGES_BLOCKS;
 		else
-			free_pages = zone_page_state(zone, NR_FREE_PAGES);
+			item = NR_FREE_PAGES;
+
+		/*
+		 * When there is a high number of CPUs in the system,
+		 * the cumulative error from the vmstat per-cpu cache
+		 * can blur the line between the watermarks. In that
+		 * case, be safe and get an accurate snapshot.
+		 *
+		 * TODO: NR_FREE_PAGES_BLOCKS moves in steps of
+		 * pageblock_nr_pages, while the vmstat pcp threshold
+		 * is limited to 125. On many configurations that
+		 * counter won't actually be per-cpu cached. But keep
+		 * things simple for now; revisit when somebody cares.
+		 */
+		free_pages = zone_page_state(zone, item);
+		if (zone->percpu_drift_mark && free_pages < zone->percpu_drift_mark)
+			free_pages = zone_page_state_snapshot(zone, item);
 
 		if (__zone_watermark_ok(zone, order, mark, highest_zoneidx,
 					0, free_pages))
diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig
index 860a0786bc1e..20b316207f9a 100644
--- a/net/batman-adv/Kconfig
+++ b/net/batman-adv/Kconfig
@@ -9,7 +9,7 @@
 
 config BATMAN_ADV
 	tristate "B.A.T.M.A.N. Advanced Meshing Protocol"
-	select LIBCRC32C
+	select CRC32
 	help
 	  B.A.T.M.A.N. (better approach to mobile ad-hoc networking) is
 	  a routing protocol for multi-hop ad-hoc mesh networks. The
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index f145f9662653..7cd4bdcee439 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -725,7 +725,6 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface,
 
 	kref_get(&hard_iface->refcount);
 
-	dev_hold(mesh_iface);
 	netdev_hold(mesh_iface, &hard_iface->meshif_dev_tracker, GFP_ATOMIC);
 	hard_iface->mesh_iface = mesh_iface;
 	bat_priv = netdev_priv(hard_iface->mesh_iface);
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 95972fd4c784..6533e281ada3 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -2064,95 +2064,6 @@ static int create_big_sync(struct hci_dev *hdev, void *data)
 	return hci_le_create_big(conn, &conn->iso_qos);
 }
 
-static void create_pa_complete(struct hci_dev *hdev, void *data, int err)
-{
-	bt_dev_dbg(hdev, "");
-
-	if (err)
-		bt_dev_err(hdev, "Unable to create PA: %d", err);
-}
-
-static bool hci_conn_check_create_pa_sync(struct hci_conn *conn)
-{
-	if (conn->type != ISO_LINK || conn->sid == HCI_SID_INVALID)
-		return false;
-
-	return true;
-}
-
-static int create_pa_sync(struct hci_dev *hdev, void *data)
-{
-	struct hci_cp_le_pa_create_sync cp = {0};
-	struct hci_conn *conn;
-	int err = 0;
-
-	hci_dev_lock(hdev);
-
-	rcu_read_lock();
-
-	/* The spec allows only one pending LE Periodic Advertising Create
-	 * Sync command at a time. If the command is pending now, don't do
-	 * anything. We check for pending connections after each PA Sync
-	 * Established event.
-	 *
-	 * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E
-	 * page 2493:
-	 *
-	 * If the Host issues this command when another HCI_LE_Periodic_
-	 * Advertising_Create_Sync command is pending, the Controller shall
-	 * return the error code Command Disallowed (0x0C).
-	 */
-	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		if (test_bit(HCI_CONN_CREATE_PA_SYNC, &conn->flags))
-			goto unlock;
-	}
-
-	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		if (hci_conn_check_create_pa_sync(conn)) {
-			struct bt_iso_qos *qos = &conn->iso_qos;
-
-			cp.options = qos->bcast.options;
-			cp.sid = conn->sid;
-			cp.addr_type = conn->dst_type;
-			bacpy(&cp.addr, &conn->dst);
-			cp.skip = cpu_to_le16(qos->bcast.skip);
-			cp.sync_timeout = cpu_to_le16(qos->bcast.sync_timeout);
-			cp.sync_cte_type = qos->bcast.sync_cte_type;
-
-			break;
-		}
-	}
-
-unlock:
-	rcu_read_unlock();
-
-	hci_dev_unlock(hdev);
-
-	if (bacmp(&cp.addr, BDADDR_ANY)) {
-		hci_dev_set_flag(hdev, HCI_PA_SYNC);
-		set_bit(HCI_CONN_CREATE_PA_SYNC, &conn->flags);
-
-		err = __hci_cmd_sync_status(hdev, HCI_OP_LE_PA_CREATE_SYNC,
-					    sizeof(cp), &cp, HCI_CMD_TIMEOUT);
-		if (!err)
-			err = hci_update_passive_scan_sync(hdev);
-
-		if (err) {
-			hci_dev_clear_flag(hdev, HCI_PA_SYNC);
-			clear_bit(HCI_CONN_CREATE_PA_SYNC, &conn->flags);
-		}
-	}
-
-	return err;
-}
-
-int hci_pa_create_sync_pending(struct hci_dev *hdev)
-{
-	/* Queue start pa_create_sync and scan */
-	return hci_cmd_sync_queue(hdev, create_pa_sync,
-				  NULL, create_pa_complete);
-}
-
 struct hci_conn *hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst,
 				    __u8 dst_type, __u8 sid,
 				    struct bt_iso_qos *qos)
@@ -2167,97 +2078,18 @@ struct hci_conn *hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst,
 	conn->dst_type = dst_type;
 	conn->sid = sid;
 	conn->state = BT_LISTEN;
+	conn->conn_timeout = msecs_to_jiffies(qos->bcast.sync_timeout * 10);
 
 	hci_conn_hold(conn);
 
-	hci_pa_create_sync_pending(hdev);
+	hci_connect_pa_sync(hdev, conn);
 
 	return conn;
 }
 
-static bool hci_conn_check_create_big_sync(struct hci_conn *conn)
-{
-	if (!conn->num_bis)
-		return false;
-
-	return true;
-}
-
-static void big_create_sync_complete(struct hci_dev *hdev, void *data, int err)
-{
-	bt_dev_dbg(hdev, "");
-
-	if (err)
-		bt_dev_err(hdev, "Unable to create BIG sync: %d", err);
-}
-
-static int big_create_sync(struct hci_dev *hdev, void *data)
-{
-	DEFINE_FLEX(struct hci_cp_le_big_create_sync, pdu, bis, num_bis, 0x11);
-	struct hci_conn *conn;
-
-	rcu_read_lock();
-
-	pdu->num_bis = 0;
-
-	/* The spec allows only one pending LE BIG Create Sync command at
-	 * a time. If the command is pending now, don't do anything. We
-	 * check for pending connections after each BIG Sync Established
-	 * event.
-	 *
-	 * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E
-	 * page 2586:
-	 *
-	 * If the Host sends this command when the Controller is in the
-	 * process of synchronizing to any BIG, i.e. the HCI_LE_BIG_Sync_
-	 * Established event has not been generated, the Controller shall
-	 * return the error code Command Disallowed (0x0C).
-	 */
-	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		if (test_bit(HCI_CONN_CREATE_BIG_SYNC, &conn->flags))
-			goto unlock;
-	}
-
-	list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) {
-		if (hci_conn_check_create_big_sync(conn)) {
-			struct bt_iso_qos *qos = &conn->iso_qos;
-
-			set_bit(HCI_CONN_CREATE_BIG_SYNC, &conn->flags);
-
-			pdu->handle = qos->bcast.big;
-			pdu->sync_handle = cpu_to_le16(conn->sync_handle);
-			pdu->encryption = qos->bcast.encryption;
-			memcpy(pdu->bcode, qos->bcast.bcode,
-			       sizeof(pdu->bcode));
-			pdu->mse = qos->bcast.mse;
-			pdu->timeout = cpu_to_le16(qos->bcast.timeout);
-			pdu->num_bis = conn->num_bis;
-			memcpy(pdu->bis, conn->bis, conn->num_bis);
-
-			break;
-		}
-	}
-
-unlock:
-	rcu_read_unlock();
-
-	if (!pdu->num_bis)
-		return 0;
-
-	return hci_send_cmd(hdev, HCI_OP_LE_BIG_CREATE_SYNC,
-			    struct_size(pdu, bis, pdu->num_bis), pdu);
-}
-
-int hci_le_big_create_sync_pending(struct hci_dev *hdev)
-{
-	/* Queue big_create_sync */
-	return hci_cmd_sync_queue_once(hdev, big_create_sync,
-				       NULL, big_create_sync_complete);
-}
-
-int hci_le_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon,
-			   struct bt_iso_qos *qos,
-			   __u16 sync_handle, __u8 num_bis, __u8 bis[])
+int hci_conn_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon,
+			     struct bt_iso_qos *qos, __u16 sync_handle,
+			     __u8 num_bis, __u8 bis[])
 {
 	int err;
 
@@ -2274,9 +2106,10 @@ int hci_le_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon,
 
 		hcon->num_bis = num_bis;
 		memcpy(hcon->bis, bis, num_bis);
+		hcon->conn_timeout = msecs_to_jiffies(qos->bcast.timeout * 10);
 	}
 
-	return hci_le_big_create_sync_pending(hdev);
+	return hci_connect_big_sync(hdev, hcon);
 }
 
 static void create_big_complete(struct hci_dev *hdev, void *data, int err)
@@ -3072,6 +2905,7 @@ void hci_setup_tx_timestamp(struct sk_buff *skb, size_t key_offset,
 			    const struct sockcm_cookie *sockc)
 {
 	struct sock *sk = skb ? skb->sk : NULL;
+	int key;
 
 	/* This shall be called on a single skb of those generated by user
 	 * sendmsg(), and only when the sendmsg() does not return error to
@@ -3087,13 +2921,16 @@ void hci_setup_tx_timestamp(struct sk_buff *skb, size_t key_offset,
 
 	sock_tx_timestamp(sk, sockc, &skb_shinfo(skb)->tx_flags);
 
+	if (sk->sk_type == SOCK_STREAM)
+		key = atomic_add_return(key_offset, &sk->sk_tskey);
+
 	if (sockc->tsflags & SOF_TIMESTAMPING_OPT_ID &&
 	    sockc->tsflags & SOF_TIMESTAMPING_TX_RECORD_MASK) {
 		if (sockc->tsflags & SOCKCM_FLAG_TS_OPT_ID) {
 			skb_shinfo(skb)->tskey = sockc->ts_opt_id;
 		} else {
-			int key = atomic_add_return(key_offset, &sk->sk_tskey);
-
+			if (sk->sk_type != SOCK_STREAM)
+				key = atomic_inc_return(&sk->sk_tskey);
 			skb_shinfo(skb)->tskey = key - 1;
 		}
 	}
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 1d8616f2e740..6d6061111ac5 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -6160,11 +6160,12 @@ static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr,
 	 * event or send an immediate device found event if the data
 	 * should not be stored for later.
 	 */
-	if (!ext_adv &&	!has_pending_adv_report(hdev)) {
+	if (!has_pending_adv_report(hdev)) {
 		/* If the report will trigger a SCAN_REQ store it for
 		 * later merging.
 		 */
-		if (type == LE_ADV_IND || type == LE_ADV_SCAN_IND) {
+		if (!ext_adv && (type == LE_ADV_IND ||
+				 type == LE_ADV_SCAN_IND)) {
 			store_pending_adv_report(hdev, bdaddr, bdaddr_type,
 						 rssi, flags, data, len);
 			return;
@@ -6377,8 +6378,7 @@ static void hci_le_pa_sync_estabilished_evt(struct hci_dev *hdev, void *data,
 
 	hci_dev_clear_flag(hdev, HCI_PA_SYNC);
 
-	conn = hci_conn_hash_lookup_sid(hdev, ev->sid, &ev->bdaddr,
-					ev->bdaddr_type);
+	conn = hci_conn_hash_lookup_create_pa_sync(hdev);
 	if (!conn) {
 		bt_dev_err(hdev,
 			   "Unable to find connection for dst %pMR sid 0x%2.2x",
@@ -6417,9 +6417,6 @@ static void hci_le_pa_sync_estabilished_evt(struct hci_dev *hdev, void *data,
 	}
 
 unlock:
-	/* Handle any other pending PA sync command */
-	hci_pa_create_sync_pending(hdev);
-
 	hci_dev_unlock(hdev);
 }
 
@@ -6931,7 +6928,7 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data,
 
 	bt_dev_dbg(hdev, "status 0x%2.2x", ev->status);
 
-	if (!hci_le_ev_skb_pull(hdev, skb, HCI_EVT_LE_BIG_SYNC_ESTABILISHED,
+	if (!hci_le_ev_skb_pull(hdev, skb, HCI_EVT_LE_BIG_SYNC_ESTABLISHED,
 				flex_array_size(ev, bis, ev->num_bis)))
 		return;
 
@@ -7002,9 +6999,6 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data,
 		}
 
 unlock:
-	/* Handle any other pending BIG sync command */
-	hci_le_big_create_sync_pending(hdev);
-
 	hci_dev_unlock(hdev);
 }
 
@@ -7126,8 +7120,8 @@ static const struct hci_le_ev {
 		     hci_le_create_big_complete_evt,
 		     sizeof(struct hci_evt_le_create_big_complete),
 		     HCI_MAX_EVENT_SIZE),
-	/* [0x1d = HCI_EV_LE_BIG_SYNC_ESTABILISHED] */
-	HCI_LE_EV_VL(HCI_EVT_LE_BIG_SYNC_ESTABILISHED,
+	/* [0x1d = HCI_EV_LE_BIG_SYNC_ESTABLISHED] */
+	HCI_LE_EV_VL(HCI_EVT_LE_BIG_SYNC_ESTABLISHED,
 		     hci_le_big_sync_established_evt,
 		     sizeof(struct hci_evt_le_big_sync_estabilished),
 		     HCI_MAX_EVENT_SIZE),
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 609b035e5c90..e56b1cbedab9 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -2693,16 +2693,16 @@ static u8 hci_update_accept_list_sync(struct hci_dev *hdev)
 
 	/* Force address filtering if PA Sync is in progress */
 	if (hci_dev_test_flag(hdev, HCI_PA_SYNC)) {
-		struct hci_cp_le_pa_create_sync *sent;
+		struct hci_conn *conn;
 
-		sent = hci_sent_cmd_data(hdev, HCI_OP_LE_PA_CREATE_SYNC);
-		if (sent) {
+		conn = hci_conn_hash_lookup_create_pa_sync(hdev);
+		if (conn) {
 			struct conn_params pa;
 
 			memset(&pa, 0, sizeof(pa));
 
-			bacpy(&pa.addr, &sent->addr);
-			pa.addr_type = sent->addr_type;
+			bacpy(&pa.addr, &conn->dst);
+			pa.addr_type = conn->dst_type;
 
 			/* Clear first since there could be addresses left
 			 * behind.
@@ -6895,3 +6895,143 @@ int hci_le_conn_update_sync(struct hci_dev *hdev, struct hci_conn *conn,
 	return __hci_cmd_sync_status(hdev, HCI_OP_LE_CONN_UPDATE,
 				     sizeof(cp), &cp, HCI_CMD_TIMEOUT);
 }
+
+static void create_pa_complete(struct hci_dev *hdev, void *data, int err)
+{
+	bt_dev_dbg(hdev, "err %d", err);
+
+	if (!err)
+		return;
+
+	hci_dev_clear_flag(hdev, HCI_PA_SYNC);
+
+	if (err == -ECANCELED)
+		return;
+
+	hci_dev_lock(hdev);
+
+	hci_update_passive_scan_sync(hdev);
+
+	hci_dev_unlock(hdev);
+}
+
+static int hci_le_pa_create_sync(struct hci_dev *hdev, void *data)
+{
+	struct hci_cp_le_pa_create_sync cp;
+	struct hci_conn *conn = data;
+	struct bt_iso_qos *qos = &conn->iso_qos;
+	int err;
+
+	if (!hci_conn_valid(hdev, conn))
+		return -ECANCELED;
+
+	if (hci_dev_test_and_set_flag(hdev, HCI_PA_SYNC))
+		return -EBUSY;
+
+	/* Mark HCI_CONN_CREATE_PA_SYNC so hci_update_passive_scan_sync can
+	 * program the address in the allow list so PA advertisements can be
+	 * received.
+	 */
+	set_bit(HCI_CONN_CREATE_PA_SYNC, &conn->flags);
+
+	hci_update_passive_scan_sync(hdev);
+
+	memset(&cp, 0, sizeof(cp));
+	cp.options = qos->bcast.options;
+	cp.sid = conn->sid;
+	cp.addr_type = conn->dst_type;
+	bacpy(&cp.addr, &conn->dst);
+	cp.skip = cpu_to_le16(qos->bcast.skip);
+	cp.sync_timeout = cpu_to_le16(qos->bcast.sync_timeout);
+	cp.sync_cte_type = qos->bcast.sync_cte_type;
+
+	/* The spec allows only one pending LE Periodic Advertising Create
+	 * Sync command at a time so we forcefully wait for PA Sync Established
+	 * event since cmd_work can only schedule one command at a time.
+	 *
+	 * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E
+	 * page 2493:
+	 *
+	 * If the Host issues this command when another HCI_LE_Periodic_
+	 * Advertising_Create_Sync command is pending, the Controller shall
+	 * return the error code Command Disallowed (0x0C).
+	 */
+	err = __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_PA_CREATE_SYNC,
+				       sizeof(cp), &cp,
+				       HCI_EV_LE_PA_SYNC_ESTABLISHED,
+				       conn->conn_timeout, NULL);
+	if (err == -ETIMEDOUT)
+		__hci_cmd_sync_status(hdev, HCI_OP_LE_PA_CREATE_SYNC_CANCEL,
+				      0, NULL, HCI_CMD_TIMEOUT);
+
+	return err;
+}
+
+int hci_connect_pa_sync(struct hci_dev *hdev, struct hci_conn *conn)
+{
+	return hci_cmd_sync_queue_once(hdev, hci_le_pa_create_sync, conn,
+				       create_pa_complete);
+}
+
+static void create_big_complete(struct hci_dev *hdev, void *data, int err)
+{
+	struct hci_conn *conn = data;
+
+	bt_dev_dbg(hdev, "err %d", err);
+
+	if (err == -ECANCELED)
+		return;
+
+	if (hci_conn_valid(hdev, conn))
+		clear_bit(HCI_CONN_CREATE_BIG_SYNC, &conn->flags);
+}
+
+static int hci_le_big_create_sync(struct hci_dev *hdev, void *data)
+{
+	DEFINE_FLEX(struct hci_cp_le_big_create_sync, cp, bis, num_bis, 0x11);
+	struct hci_conn *conn = data;
+	struct bt_iso_qos *qos = &conn->iso_qos;
+	int err;
+
+	if (!hci_conn_valid(hdev, conn))
+		return -ECANCELED;
+
+	set_bit(HCI_CONN_CREATE_BIG_SYNC, &conn->flags);
+
+	memset(cp, 0, sizeof(*cp));
+	cp->handle = qos->bcast.big;
+	cp->sync_handle = cpu_to_le16(conn->sync_handle);
+	cp->encryption = qos->bcast.encryption;
+	memcpy(cp->bcode, qos->bcast.bcode, sizeof(cp->bcode));
+	cp->mse = qos->bcast.mse;
+	cp->timeout = cpu_to_le16(qos->bcast.timeout);
+	cp->num_bis = conn->num_bis;
+	memcpy(cp->bis, conn->bis, conn->num_bis);
+
+	/* The spec allows only one pending LE BIG Create Sync command at
+	 * a time, so we forcefully wait for BIG Sync Established event since
+	 * cmd_work can only schedule one command at a time.
+	 *
+	 * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E
+	 * page 2586:
+	 *
+	 * If the Host sends this command when the Controller is in the
+	 * process of synchronizing to any BIG, i.e. the HCI_LE_BIG_Sync_
+	 * Established event has not been generated, the Controller shall
+	 * return the error code Command Disallowed (0x0C).
+	 */
+	err = __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_BIG_CREATE_SYNC,
+				       struct_size(cp, bis, cp->num_bis), cp,
+				       HCI_EVT_LE_BIG_SYNC_ESTABLISHED,
+				       conn->conn_timeout, NULL);
+	if (err == -ETIMEDOUT)
+		hci_le_big_terminate_sync(hdev, cp->handle);
+
+	return err;
+}
+
+int hci_connect_big_sync(struct hci_dev *hdev, struct hci_conn *conn)
+{
+	return hci_cmd_sync_queue_once(hdev, hci_le_big_create_sync, conn,
+				       create_big_complete);
+}
diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c
index 3501a991f1c6..2819cda616bc 100644
--- a/net/bluetooth/iso.c
+++ b/net/bluetooth/iso.c
@@ -1462,14 +1462,13 @@ static void iso_conn_big_sync(struct sock *sk)
 	lock_sock(sk);
 
 	if (!test_and_set_bit(BT_SK_BIG_SYNC, &iso_pi(sk)->flags)) {
-		err = hci_le_big_create_sync(hdev, iso_pi(sk)->conn->hcon,
-					     &iso_pi(sk)->qos,
-					     iso_pi(sk)->sync_handle,
-					     iso_pi(sk)->bc_num_bis,
-					     iso_pi(sk)->bc_bis);
+		err = hci_conn_big_create_sync(hdev, iso_pi(sk)->conn->hcon,
+					       &iso_pi(sk)->qos,
+					       iso_pi(sk)->sync_handle,
+					       iso_pi(sk)->bc_num_bis,
+					       iso_pi(sk)->bc_bis);
 		if (err)
-			bt_dev_err(hdev, "hci_le_big_create_sync: %d",
-				   err);
+			bt_dev_err(hdev, "hci_big_create_sync: %d", err);
 	}
 
 	release_sock(sk);
@@ -1922,7 +1921,7 @@ static void iso_conn_ready(struct iso_conn *conn)
 					      hcon);
 		} else if (test_bit(HCI_CONN_BIG_SYNC_FAILED, &hcon->flags)) {
 			ev = hci_recv_event_data(hcon->hdev,
-						 HCI_EVT_LE_BIG_SYNC_ESTABILISHED);
+						 HCI_EVT_LE_BIG_SYNC_ESTABLISHED);
 
 			/* Get reference to PA sync parent socket, if it exists */
 			parent = iso_get_sock(&hcon->src, &hcon->dst,
@@ -2113,12 +2112,11 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags)
 
 			if (!test_bit(BT_SK_DEFER_SETUP, &bt_sk(sk)->flags) &&
 			    !test_and_set_bit(BT_SK_BIG_SYNC, &iso_pi(sk)->flags)) {
-				err = hci_le_big_create_sync(hdev,
-							     hcon,
-							     &iso_pi(sk)->qos,
-							     iso_pi(sk)->sync_handle,
-							     iso_pi(sk)->bc_num_bis,
-							     iso_pi(sk)->bc_bis);
+				err = hci_conn_big_create_sync(hdev, hcon,
+							       &iso_pi(sk)->qos,
+							       iso_pi(sk)->sync_handle,
+							       iso_pi(sk)->bc_num_bis,
+							       iso_pi(sk)->bc_bis);
 				if (err) {
 					bt_dev_err(hdev, "hci_le_big_create_sync: %d",
 						   err);
diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index c7b66b2ea9f2..73472756618a 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -3991,7 +3991,8 @@ static void l2cap_connect(struct l2cap_conn *conn, struct l2cap_cmd_hdr *cmd,
 
 	/* Check if the ACL is secure enough (if not SDP) */
 	if (psm != cpu_to_le16(L2CAP_PSM_SDP) &&
-	    !hci_conn_check_link_mode(conn->hcon)) {
+	    (!hci_conn_check_link_mode(conn->hcon) ||
+	    !l2cap_check_enc_key_size(conn->hcon))) {
 		conn->disc_reason = HCI_ERROR_AUTH_FAILURE;
 		result = L2CAP_CR_SEC_BLOCK;
 		goto response;
@@ -7414,6 +7415,9 @@ static int l2cap_recv_frag(struct l2cap_conn *conn, struct sk_buff *skb,
 			return -ENOMEM;
 		/* Init rx_len */
 		conn->rx_len = len;
+
+		skb_set_delivery_time(conn->rx_skb, skb->tstamp,
+				      skb->tstamp_type);
 	}
 
 	/* Copy as much as the rx_skb can hold */
@@ -7538,8 +7542,24 @@ void l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 flags)
 		if (skb->len > len) {
 			BT_ERR("Frame is too long (len %u, expected len %d)",
 			       skb->len, len);
+			/* PTS test cases L2CAP/COS/CED/BI-14-C and BI-15-C
+			 * (Multiple Signaling Command in one PDU, Data
+			 * Truncated, BR/EDR) send a C-frame to the IUT with
+			 * PDU Length set to 8 and Channel ID set to the
+			 * correct signaling channel for the logical link.
+			 * The Information payload contains one L2CAP_ECHO_REQ
+			 * packet with Data Length set to 0 with 0 octets of
+			 * echo data and one invalid command packet due to
+			 * data truncated in PDU but present in HCI packet.
+			 *
+			 * Shorter the socket buffer to the PDU length to
+			 * allow to process valid commands from the PDU before
+			 * setting the socket unreliable.
+			 */
+			skb->len = len;
+			l2cap_recv_frame(conn, skb);
 			l2cap_conn_unreliable(conn, ECOMM);
-			goto drop;
+			goto unlock;
 		}
 
 		/* Append fragment into frame (with header) */
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index d9a69ec9affe..939a3aa78d5c 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -715,8 +715,8 @@ static int br_vlan_add_existing(struct net_bridge *br,
 				u16 flags, bool *changed,
 				struct netlink_ext_ack *extack)
 {
-	bool would_change = __vlan_flags_would_change(vlan, flags);
 	bool becomes_brentry = false;
+	bool would_change = false;
 	int err;
 
 	if (!br_vlan_is_brentry(vlan)) {
@@ -725,6 +725,8 @@ static int br_vlan_add_existing(struct net_bridge *br,
 			return -EINVAL;
 
 		becomes_brentry = true;
+	} else {
+		would_change = __vlan_flags_would_change(vlan, flags);
 	}
 
 	/* Master VLANs that aren't brentries weren't notified before,
diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c
index 17226b2341d0..6fefe7a68761 100644
--- a/net/can/j1939/socket.c
+++ b/net/can/j1939/socket.c
@@ -655,6 +655,7 @@ static int j1939_sk_release(struct socket *sock)
 	sock->sk = NULL;
 
 	release_sock(sk);
+	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 	sock_put(sk);
 
 	return 0;
diff --git a/net/ceph/Kconfig b/net/ceph/Kconfig
index c5c4eef3a9ff..0aa21fcbf6ec 100644
--- a/net/ceph/Kconfig
+++ b/net/ceph/Kconfig
@@ -2,7 +2,7 @@
 config CEPH_LIB
 	tristate "Ceph core library"
 	depends on INET
-	select LIBCRC32C
+	select CRC32
 	select CRYPTO_AES
 	select CRYPTO_CBC
 	select CRYPTO_GCM
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index b24afec24138..6664ea73ccf8 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -220,16 +220,6 @@ void osd_req_op_extent_osd_data_pages(struct ceph_osd_request *osd_req,
 }
 EXPORT_SYMBOL(osd_req_op_extent_osd_data_pages);
 
-void osd_req_op_extent_osd_data_pagelist(struct ceph_osd_request *osd_req,
-			unsigned int which, struct ceph_pagelist *pagelist)
-{
-	struct ceph_osd_data *osd_data;
-
-	osd_data = osd_req_op_data(osd_req, which, extent, osd_data);
-	ceph_osd_data_pagelist_init(osd_data, pagelist);
-}
-EXPORT_SYMBOL(osd_req_op_extent_osd_data_pagelist);
-
 #ifdef CONFIG_BLOCK
 void osd_req_op_extent_osd_data_bio(struct ceph_osd_request *osd_req,
 				    unsigned int which,
@@ -297,19 +287,6 @@ static void osd_req_op_cls_request_info_pagelist(
 	ceph_osd_data_pagelist_init(osd_data, pagelist);
 }
 
-void osd_req_op_cls_request_data_pagelist(
-			struct ceph_osd_request *osd_req,
-			unsigned int which, struct ceph_pagelist *pagelist)
-{
-	struct ceph_osd_data *osd_data;
-
-	osd_data = osd_req_op_data(osd_req, which, cls, request_data);
-	ceph_osd_data_pagelist_init(osd_data, pagelist);
-	osd_req->r_ops[which].cls.indata_len += pagelist->length;
-	osd_req->r_ops[which].indata_len += pagelist->length;
-}
-EXPORT_SYMBOL(osd_req_op_cls_request_data_pagelist);
-
 void osd_req_op_cls_request_data_pages(struct ceph_osd_request *osd_req,
 			unsigned int which, struct page **pages, u64 length,
 			u32 alignment, bool pages_from_pool, bool own_pages)
diff --git a/net/core/dev.c b/net/core/dev.c
index 0608605cfc24..1be7cb73a602 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1518,16 +1518,10 @@ void netdev_features_change(struct net_device *dev)
 }
 EXPORT_SYMBOL(netdev_features_change);
 
-/**
- *	netdev_state_change - device changes state
- *	@dev: device to cause notification
- *
- *	Called to indicate a device has changed state. This function calls
- *	the notifier chains for netdev_chain and sends a NEWLINK message
- *	to the routing socket.
- */
-void netdev_state_change(struct net_device *dev)
+void netif_state_change(struct net_device *dev)
 {
+	netdev_ops_assert_locked_or_invisible(dev);
+
 	if (dev->flags & IFF_UP) {
 		struct netdev_notifier_change_info change_info = {
 			.info.dev = dev,
@@ -1538,7 +1532,6 @@ void netdev_state_change(struct net_device *dev)
 		rtmsg_ifinfo(RTM_NEWLINK, dev, 0, GFP_KERNEL, 0, NULL);
 	}
 }
-EXPORT_SYMBOL(netdev_state_change);
 
 /**
  * __netdev_notify_peers - notify network peers about existence of @dev,
@@ -11941,15 +11934,24 @@ void unregister_netdevice_many_notify(struct list_head *head,
 		BUG_ON(dev->reg_state != NETREG_REGISTERED);
 	}
 
-	/* If device is running, close it first. */
+	/* If device is running, close it first. Start with ops locked... */
 	list_for_each_entry(dev, head, unreg_list) {
-		list_add_tail(&dev->close_list, &close_head);
-		netdev_lock_ops(dev);
+		if (netdev_need_ops_lock(dev)) {
+			list_add_tail(&dev->close_list, &close_head);
+			netdev_lock(dev);
+		}
+	}
+	dev_close_many(&close_head, true);
+	/* ... now unlock them and go over the rest. */
+	list_for_each_entry(dev, head, unreg_list) {
+		if (netdev_need_ops_lock(dev))
+			netdev_unlock(dev);
+		else
+			list_add_tail(&dev->close_list, &close_head);
 	}
 	dev_close_many(&close_head, true);
 
 	list_for_each_entry(dev, head, unreg_list) {
-		netdev_unlock_ops(dev);
 		/* And unlink it from device chain. */
 		unlist_netdevice(dev);
 		netdev_lock(dev);
diff --git a/net/core/dev_api.c b/net/core/dev_api.c
index 90bafb0b1b8c..90898cd540ce 100644
--- a/net/core/dev_api.c
+++ b/net/core/dev_api.c
@@ -327,3 +327,19 @@ int dev_xdp_propagate(struct net_device *dev, struct netdev_bpf *bpf)
 	return ret;
 }
 EXPORT_SYMBOL_GPL(dev_xdp_propagate);
+
+/**
+ * netdev_state_change() - device changes state
+ * @dev: device to cause notification
+ *
+ * Called to indicate a device has changed state. This function calls
+ * the notifier chains for netdev_chain and sends a NEWLINK message
+ * to the routing socket.
+ */
+void netdev_state_change(struct net_device *dev)
+{
+	netdev_lock_ops(dev);
+	netif_state_change(dev);
+	netdev_unlock_ops(dev);
+}
+EXPORT_SYMBOL(netdev_state_change);
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 4bc64d912a1c..7af302080a66 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -257,6 +257,24 @@ static int nla_put_port_range(struct sk_buff *skb, int attrtype,
 	return nla_put(skb, attrtype, sizeof(*range), range);
 }
 
+static bool fib_rule_iif_match(const struct fib_rule *rule, int iifindex,
+			       const struct flowi *fl)
+{
+	u8 iif_is_l3_master = READ_ONCE(rule->iif_is_l3_master);
+
+	return iif_is_l3_master ? l3mdev_fib_rule_iif_match(fl, iifindex) :
+				  fl->flowi_iif == iifindex;
+}
+
+static bool fib_rule_oif_match(const struct fib_rule *rule, int oifindex,
+			       const struct flowi *fl)
+{
+	u8 oif_is_l3_master = READ_ONCE(rule->oif_is_l3_master);
+
+	return oif_is_l3_master ? l3mdev_fib_rule_oif_match(fl, oifindex) :
+				  fl->flowi_oif == oifindex;
+}
+
 static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops,
 			  struct flowi *fl, int flags,
 			  struct fib_lookup_arg *arg)
@@ -264,11 +282,11 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops,
 	int iifindex, oifindex, ret = 0;
 
 	iifindex = READ_ONCE(rule->iifindex);
-	if (iifindex && (iifindex != fl->flowi_iif))
+	if (iifindex && !fib_rule_iif_match(rule, iifindex, fl))
 		goto out;
 
 	oifindex = READ_ONCE(rule->oifindex);
-	if (oifindex && (oifindex != fl->flowi_oif))
+	if (oifindex && !fib_rule_oif_match(rule, oifindex, fl))
 		goto out;
 
 	if ((rule->mark ^ fl->flowi_mark) & rule->mark_mask)
@@ -736,16 +754,20 @@ static int fib_nl2rule_rtnl(struct fib_rule *nlrule,
 		struct net_device *dev;
 
 		dev = __dev_get_by_name(nlrule->fr_net, nlrule->iifname);
-		if (dev)
+		if (dev) {
 			nlrule->iifindex = dev->ifindex;
+			nlrule->iif_is_l3_master = netif_is_l3_master(dev);
+		}
 	}
 
 	if (tb[FRA_OIFNAME]) {
 		struct net_device *dev;
 
 		dev = __dev_get_by_name(nlrule->fr_net, nlrule->oifname);
-		if (dev)
+		if (dev) {
 			nlrule->oifindex = dev->ifindex;
+			nlrule->oif_is_l3_master = netif_is_l3_master(dev);
+		}
 	}
 
 	return 0;
@@ -1336,11 +1358,17 @@ static void attach_rules(struct list_head *rules, struct net_device *dev)
 
 	list_for_each_entry(rule, rules, list) {
 		if (rule->iifindex == -1 &&
-		    strcmp(dev->name, rule->iifname) == 0)
+		    strcmp(dev->name, rule->iifname) == 0) {
 			WRITE_ONCE(rule->iifindex, dev->ifindex);
+			WRITE_ONCE(rule->iif_is_l3_master,
+				   netif_is_l3_master(dev));
+		}
 		if (rule->oifindex == -1 &&
-		    strcmp(dev->name, rule->oifname) == 0)
+		    strcmp(dev->name, rule->oifname) == 0) {
 			WRITE_ONCE(rule->oifindex, dev->ifindex);
+			WRITE_ONCE(rule->oif_is_l3_master,
+				   netif_is_l3_master(dev));
+		}
 	}
 }
 
@@ -1349,10 +1377,14 @@ static void detach_rules(struct list_head *rules, struct net_device *dev)
 	struct fib_rule *rule;
 
 	list_for_each_entry(rule, rules, list) {
-		if (rule->iifindex == dev->ifindex)
+		if (rule->iifindex == dev->ifindex) {
 			WRITE_ONCE(rule->iifindex, -1);
-		if (rule->oifindex == dev->ifindex)
+			WRITE_ONCE(rule->iif_is_l3_master, false);
+		}
+		if (rule->oifindex == dev->ifindex) {
 			WRITE_ONCE(rule->oifindex, -1);
+			WRITE_ONCE(rule->oif_is_l3_master, false);
+		}
 	}
 }
 
diff --git a/net/core/filter.c b/net/core/filter.c
index bc6828761a47..79cab4d78dc3 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -218,24 +218,36 @@ BPF_CALL_3(bpf_skb_get_nlattr_nest, struct sk_buff *, skb, u32, a, u32, x)
 	return 0;
 }
 
+static int bpf_skb_load_helper_convert_offset(const struct sk_buff *skb, int offset)
+{
+	if (likely(offset >= 0))
+		return offset;
+
+	if (offset >= SKF_NET_OFF)
+		return offset - SKF_NET_OFF + skb_network_offset(skb);
+
+	if (offset >= SKF_LL_OFF && skb_mac_header_was_set(skb))
+		return offset - SKF_LL_OFF + skb_mac_offset(skb);
+
+	return INT_MIN;
+}
+
 BPF_CALL_4(bpf_skb_load_helper_8, const struct sk_buff *, skb, const void *,
 	   data, int, headlen, int, offset)
 {
-	u8 tmp, *ptr;
+	u8 tmp;
 	const int len = sizeof(tmp);
 
-	if (offset >= 0) {
-		if (headlen - offset >= len)
-			return *(u8 *)(data + offset);
-		if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
-			return tmp;
-	} else {
-		ptr = bpf_internal_load_pointer_neg_helper(skb, offset, len);
-		if (likely(ptr))
-			return *(u8 *)ptr;
-	}
+	offset = bpf_skb_load_helper_convert_offset(skb, offset);
+	if (offset == INT_MIN)
+		return -EFAULT;
 
-	return -EFAULT;
+	if (headlen - offset >= len)
+		return *(u8 *)(data + offset);
+	if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
+		return tmp;
+	else
+		return -EFAULT;
 }
 
 BPF_CALL_2(bpf_skb_load_helper_8_no_cache, const struct sk_buff *, skb,
@@ -248,21 +260,19 @@ BPF_CALL_2(bpf_skb_load_helper_8_no_cache, const struct sk_buff *, skb,
 BPF_CALL_4(bpf_skb_load_helper_16, const struct sk_buff *, skb, const void *,
 	   data, int, headlen, int, offset)
 {
-	__be16 tmp, *ptr;
+	__be16 tmp;
 	const int len = sizeof(tmp);
 
-	if (offset >= 0) {
-		if (headlen - offset >= len)
-			return get_unaligned_be16(data + offset);
-		if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
-			return be16_to_cpu(tmp);
-	} else {
-		ptr = bpf_internal_load_pointer_neg_helper(skb, offset, len);
-		if (likely(ptr))
-			return get_unaligned_be16(ptr);
-	}
+	offset = bpf_skb_load_helper_convert_offset(skb, offset);
+	if (offset == INT_MIN)
+		return -EFAULT;
 
-	return -EFAULT;
+	if (headlen - offset >= len)
+		return get_unaligned_be16(data + offset);
+	if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
+		return be16_to_cpu(tmp);
+	else
+		return -EFAULT;
 }
 
 BPF_CALL_2(bpf_skb_load_helper_16_no_cache, const struct sk_buff *, skb,
@@ -275,21 +285,19 @@ BPF_CALL_2(bpf_skb_load_helper_16_no_cache, const struct sk_buff *, skb,
 BPF_CALL_4(bpf_skb_load_helper_32, const struct sk_buff *, skb, const void *,
 	   data, int, headlen, int, offset)
 {
-	__be32 tmp, *ptr;
+	__be32 tmp;
 	const int len = sizeof(tmp);
 
-	if (likely(offset >= 0)) {
-		if (headlen - offset >= len)
-			return get_unaligned_be32(data + offset);
-		if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
-			return be32_to_cpu(tmp);
-	} else {
-		ptr = bpf_internal_load_pointer_neg_helper(skb, offset, len);
-		if (likely(ptr))
-			return get_unaligned_be32(ptr);
-	}
+	offset = bpf_skb_load_helper_convert_offset(skb, offset);
+	if (offset == INT_MIN)
+		return -EFAULT;
 
-	return -EFAULT;
+	if (headlen - offset >= len)
+		return get_unaligned_be32(data + offset);
+	if (!skb_copy_bits(skb, offset, &tmp, sizeof(tmp)))
+		return be32_to_cpu(tmp);
+	else
+		return -EFAULT;
 }
 
 BPF_CALL_2(bpf_skb_load_helper_32_no_cache, const struct sk_buff *, skb,
diff --git a/net/core/link_watch.c b/net/core/link_watch.c
index cb04ef2b9807..864f3bbc3a4c 100644
--- a/net/core/link_watch.c
+++ b/net/core/link_watch.c
@@ -183,7 +183,7 @@ static void linkwatch_do_dev(struct net_device *dev)
 		else
 			dev_deactivate(dev);
 
-		netdev_state_change(dev);
+		netif_state_change(dev);
 	}
 	/* Note: our callers are responsible for calling netdev_tracker_free().
 	 * This is the reason we use __dev_put() instead of dev_put().
@@ -240,7 +240,9 @@ static void __linkwatch_run_queue(int urgent_only)
 		 */
 		netdev_tracker_free(dev, &dev->linkwatch_dev_tracker);
 		spin_unlock_irq(&lweventlist_lock);
+		netdev_lock_ops(dev);
 		linkwatch_do_dev(dev);
+		netdev_unlock_ops(dev);
 		do_dev--;
 		spin_lock_irq(&lweventlist_lock);
 	}
@@ -253,25 +255,41 @@ static void __linkwatch_run_queue(int urgent_only)
 	spin_unlock_irq(&lweventlist_lock);
 }
 
-void linkwatch_sync_dev(struct net_device *dev)
+static bool linkwatch_clean_dev(struct net_device *dev)
 {
 	unsigned long flags;
-	int clean = 0;
+	bool clean = false;
 
 	spin_lock_irqsave(&lweventlist_lock, flags);
 	if (!list_empty(&dev->link_watch_list)) {
 		list_del_init(&dev->link_watch_list);
-		clean = 1;
+		clean = true;
 		/* We must release netdev tracker under
 		 * the spinlock protection.
 		 */
 		netdev_tracker_free(dev, &dev->linkwatch_dev_tracker);
 	}
 	spin_unlock_irqrestore(&lweventlist_lock, flags);
-	if (clean)
+
+	return clean;
+}
+
+void __linkwatch_sync_dev(struct net_device *dev)
+{
+	netdev_ops_assert_locked(dev);
+
+	if (linkwatch_clean_dev(dev))
 		linkwatch_do_dev(dev);
 }
 
+void linkwatch_sync_dev(struct net_device *dev)
+{
+	if (linkwatch_clean_dev(dev)) {
+		netdev_lock_ops(dev);
+		linkwatch_do_dev(dev);
+		netdev_unlock_ops(dev);
+	}
+}
 
 /* Must be called with the rtnl semaphore held */
 void linkwatch_run_queue(void)
diff --git a/net/core/lock_debug.c b/net/core/lock_debug.c
index b7f22dc92a6f..941e26c1343d 100644
--- a/net/core/lock_debug.c
+++ b/net/core/lock_debug.c
@@ -20,11 +20,11 @@ int netdev_debug_event(struct notifier_block *nb, unsigned long event,
 	switch (cmd) {
 	case NETDEV_REGISTER:
 	case NETDEV_UP:
+	case NETDEV_CHANGE:
 		netdev_ops_assert_locked(dev);
 		fallthrough;
 	case NETDEV_DOWN:
 	case NETDEV_REBOOT:
-	case NETDEV_CHANGE:
 	case NETDEV_UNREGISTER:
 	case NETDEV_CHANGEMTU:
 	case NETDEV_CHANGEADDR:
diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c
index e39a459540ec..60f27cb4e54f 100644
--- a/net/core/lwtunnel.c
+++ b/net/core/lwtunnel.c
@@ -333,6 +333,8 @@ int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 	struct dst_entry *dst;
 	int ret;
 
+	local_bh_disable();
+
 	if (dev_xmit_recursion()) {
 		net_crit_ratelimited("%s(): recursion limit reached on datapath\n",
 				     __func__);
@@ -348,8 +350,10 @@ int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 	lwtstate = dst->lwtstate;
 
 	if (lwtstate->type == LWTUNNEL_ENCAP_NONE ||
-	    lwtstate->type > LWTUNNEL_ENCAP_MAX)
-		return 0;
+	    lwtstate->type > LWTUNNEL_ENCAP_MAX) {
+		ret = 0;
+		goto out;
+	}
 
 	ret = -EOPNOTSUPP;
 	rcu_read_lock();
@@ -364,11 +368,13 @@ int lwtunnel_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 	if (ret == -EOPNOTSUPP)
 		goto drop;
 
-	return ret;
+	goto out;
 
 drop:
 	kfree_skb(skb);
 
+out:
+	local_bh_enable();
 	return ret;
 }
 EXPORT_SYMBOL_GPL(lwtunnel_output);
@@ -380,6 +386,8 @@ int lwtunnel_xmit(struct sk_buff *skb)
 	struct dst_entry *dst;
 	int ret;
 
+	local_bh_disable();
+
 	if (dev_xmit_recursion()) {
 		net_crit_ratelimited("%s(): recursion limit reached on datapath\n",
 				     __func__);
@@ -396,8 +404,10 @@ int lwtunnel_xmit(struct sk_buff *skb)
 	lwtstate = dst->lwtstate;
 
 	if (lwtstate->type == LWTUNNEL_ENCAP_NONE ||
-	    lwtstate->type > LWTUNNEL_ENCAP_MAX)
-		return 0;
+	    lwtstate->type > LWTUNNEL_ENCAP_MAX) {
+		ret = 0;
+		goto out;
+	}
 
 	ret = -EOPNOTSUPP;
 	rcu_read_lock();
@@ -412,11 +422,13 @@ int lwtunnel_xmit(struct sk_buff *skb)
 	if (ret == -EOPNOTSUPP)
 		goto drop;
 
-	return ret;
+	goto out;
 
 drop:
 	kfree_skb(skb);
 
+out:
+	local_bh_enable();
 	return ret;
 }
 EXPORT_SYMBOL_GPL(lwtunnel_xmit);
@@ -428,6 +440,8 @@ int lwtunnel_input(struct sk_buff *skb)
 	struct dst_entry *dst;
 	int ret;
 
+	DEBUG_NET_WARN_ON_ONCE(!in_softirq());
+
 	if (dev_xmit_recursion()) {
 		net_crit_ratelimited("%s(): recursion limit reached on datapath\n",
 				     __func__);
diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c
index 5d7af50fe702..230743bdbb14 100644
--- a/net/core/netdev-genl.c
+++ b/net/core/netdev-genl.c
@@ -861,14 +861,17 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info)
 
 	mutex_lock(&priv->lock);
 
+	err = 0;
 	netdev = netdev_get_by_index_lock(genl_info_net(info), ifindex);
-	if (!netdev || !netif_device_present(netdev)) {
+	if (!netdev) {
 		err = -ENODEV;
 		goto err_unlock_sock;
 	}
-
-	if (!netdev_need_ops_lock(netdev)) {
+	if (!netif_device_present(netdev))
+		err = -ENODEV;
+	else if (!netdev_need_ops_lock(netdev))
 		err = -EOPNOTSUPP;
+	if (err) {
 		NL_SET_BAD_ATTR(info->extack,
 				info->attrs[NETDEV_A_DEV_IFINDEX]);
 		goto err_unlock;
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index c23852835050..c5a7f41982a5 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1043,7 +1043,7 @@ int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id,
 }
 EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo);
 
-void netdev_set_operstate(struct net_device *dev, int newstate)
+void netif_set_operstate(struct net_device *dev, int newstate)
 {
 	unsigned int old = READ_ONCE(dev->operstate);
 
@@ -1052,9 +1052,9 @@ void netdev_set_operstate(struct net_device *dev, int newstate)
 			return;
 	} while (!try_cmpxchg(&dev->operstate, &old, newstate));
 
-	netdev_state_change(dev);
+	netif_state_change(dev);
 }
-EXPORT_SYMBOL(netdev_set_operstate);
+EXPORT_SYMBOL(netif_set_operstate);
 
 static void set_operstate(struct net_device *dev, unsigned char transition)
 {
@@ -1080,7 +1080,7 @@ static void set_operstate(struct net_device *dev, unsigned char transition)
 		break;
 	}
 
-	netdev_set_operstate(dev, operstate);
+	netif_set_operstate(dev, operstate);
 }
 
 static unsigned int rtnl_dev_get_flags(const struct net_device *dev)
@@ -3027,7 +3027,7 @@ static int do_setlink(const struct sk_buff *skb, struct net_device *dev,
 
 	err = validate_linkmsg(dev, tb, extack);
 	if (err < 0)
-		goto errout;
+		return err;
 
 	if (tb[IFLA_IFNAME])
 		nla_strscpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
@@ -3396,7 +3396,7 @@ static int do_setlink(const struct sk_buff *skb, struct net_device *dev,
 errout:
 	if (status & DO_SETLINK_MODIFIED) {
 		if ((status & DO_SETLINK_NOTIFY) == DO_SETLINK_NOTIFY)
-			netdev_state_change(dev);
+			netif_state_change(dev);
 
 		if (err < 0)
 			net_warn_ratelimited("A link change request failed with some changes committed already. Interface %s may have been left with an inconsistent configuration, please check.\n",
diff --git a/net/core/selftests.c b/net/core/selftests.c
index e99ae983fca9..35f807ea9952 100644
--- a/net/core/selftests.c
+++ b/net/core/selftests.c
@@ -100,10 +100,10 @@ static struct sk_buff *net_test_get_skb(struct net_device *ndev,
 	ehdr->h_proto = htons(ETH_P_IP);
 
 	if (attr->tcp) {
+		memset(thdr, 0, sizeof(*thdr));
 		thdr->source = htons(attr->sport);
 		thdr->dest = htons(attr->dport);
 		thdr->doff = sizeof(struct tcphdr) / 4;
-		thdr->check = 0;
 	} else {
 		uhdr->source = htons(attr->sport);
 		uhdr->dest = htons(attr->dport);
@@ -144,10 +144,18 @@ static struct sk_buff *net_test_get_skb(struct net_device *ndev,
 	attr->id = net_test_next_id;
 	shdr->id = net_test_next_id++;
 
-	if (attr->size)
-		skb_put(skb, attr->size);
-	if (attr->max_size && attr->max_size > skb->len)
-		skb_put(skb, attr->max_size - skb->len);
+	if (attr->size) {
+		void *payload = skb_put(skb, attr->size);
+
+		memset(payload, 0, attr->size);
+	}
+
+	if (attr->max_size && attr->max_size > skb->len) {
+		size_t pad_len = attr->max_size - skb->len;
+		void *pad = skb_put(skb, pad_len);
+
+		memset(pad, 0, pad_len);
+	}
 
 	skb->csum = 0;
 	skb->ip_summed = CHECKSUM_PARTIAL;
diff --git a/net/core/sock.c b/net/core/sock.c
index f67a3c5b0988..e54449c9ab0b 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2130,6 +2130,8 @@ lenout:
  */
 static inline void sock_lock_init(struct sock *sk)
 {
+	sk_owner_clear(sk);
+
 	if (sk->sk_kern_sock)
 		sock_lock_init_class_and_name(
 			sk,
@@ -2226,6 +2228,9 @@ static void sk_prot_free(struct proto *prot, struct sock *sk)
 	cgroup_sk_free(&sk->sk_cgrp_data);
 	mem_cgroup_sk_free(sk);
 	security_sk_free(sk);
+
+	sk_owner_put(sk);
+
 	if (slab != NULL)
 		kmem_cache_free(slab, sk);
 	else
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index e827775baf2e..436a7e1b412a 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -862,6 +862,16 @@ static void dsa_tree_teardown_lags(struct dsa_switch_tree *dst)
 	kfree(dst->lags);
 }
 
+static void dsa_tree_teardown_routing_table(struct dsa_switch_tree *dst)
+{
+	struct dsa_link *dl, *next;
+
+	list_for_each_entry_safe(dl, next, &dst->rtable, list) {
+		list_del(&dl->list);
+		kfree(dl);
+	}
+}
+
 static int dsa_tree_setup(struct dsa_switch_tree *dst)
 {
 	bool complete;
@@ -879,7 +889,7 @@ static int dsa_tree_setup(struct dsa_switch_tree *dst)
 
 	err = dsa_tree_setup_cpu_ports(dst);
 	if (err)
-		return err;
+		goto teardown_rtable;
 
 	err = dsa_tree_setup_switches(dst);
 	if (err)
@@ -911,14 +921,14 @@ teardown_switches:
 	dsa_tree_teardown_switches(dst);
 teardown_cpu_ports:
 	dsa_tree_teardown_cpu_ports(dst);
+teardown_rtable:
+	dsa_tree_teardown_routing_table(dst);
 
 	return err;
 }
 
 static void dsa_tree_teardown(struct dsa_switch_tree *dst)
 {
-	struct dsa_link *dl, *next;
-
 	if (!dst->setup)
 		return;
 
@@ -932,10 +942,7 @@ static void dsa_tree_teardown(struct dsa_switch_tree *dst)
 
 	dsa_tree_teardown_cpu_ports(dst);
 
-	list_for_each_entry_safe(dl, next, &dst->rtable, list) {
-		list_del(&dl->list);
-		kfree(dl);
-	}
+	dsa_tree_teardown_routing_table(dst);
 
 	pr_info("DSA: tree %d torn down\n", dst->index);
 
@@ -1478,12 +1485,44 @@ static int dsa_switch_parse(struct dsa_switch *ds, struct dsa_chip_data *cd)
 
 static void dsa_switch_release_ports(struct dsa_switch *ds)
 {
+	struct dsa_mac_addr *a, *tmp;
 	struct dsa_port *dp, *next;
+	struct dsa_vlan *v, *n;
 
 	dsa_switch_for_each_port_safe(dp, next, ds) {
-		WARN_ON(!list_empty(&dp->fdbs));
-		WARN_ON(!list_empty(&dp->mdbs));
-		WARN_ON(!list_empty(&dp->vlans));
+		/* These are either entries that upper layers lost track of
+		 * (probably due to bugs), or installed through interfaces
+		 * where one does not necessarily have to remove them, like
+		 * ndo_dflt_fdb_add().
+		 */
+		list_for_each_entry_safe(a, tmp, &dp->fdbs, list) {
+			dev_info(ds->dev,
+				 "Cleaning up unicast address %pM vid %u from port %d\n",
+				 a->addr, a->vid, dp->index);
+			list_del(&a->list);
+			kfree(a);
+		}
+
+		list_for_each_entry_safe(a, tmp, &dp->mdbs, list) {
+			dev_info(ds->dev,
+				 "Cleaning up multicast address %pM vid %u from port %d\n",
+				 a->addr, a->vid, dp->index);
+			list_del(&a->list);
+			kfree(a);
+		}
+
+		/* These are entries that upper layers have lost track of,
+		 * probably due to bugs, but also due to dsa_port_do_vlan_del()
+		 * having failed and the VLAN entry still lingering on.
+		 */
+		list_for_each_entry_safe(v, n, &dp->vlans, list) {
+			dev_info(ds->dev,
+				 "Cleaning up vid %u from port %d\n",
+				 v->vid, dp->index);
+			list_del(&v->list);
+			kfree(v);
+		}
+
 		list_del(&dp->list);
 		kfree(dp);
 	}
diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index 3ee53e28ec2e..53e03fd8071b 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -197,7 +197,7 @@ static int dsa_port_do_tag_8021q_vlan_del(struct dsa_port *dp, u16 vid)
 
 	err = ds->ops->tag_8021q_vlan_del(ds, port, vid);
 	if (err) {
-		refcount_inc(&v->refcount);
+		refcount_set(&v->refcount, 1);
 		return err;
 	}
 
diff --git a/net/ethtool/cmis.h b/net/ethtool/cmis.h
index 1e790413db0e..4a9a946cabf0 100644
--- a/net/ethtool/cmis.h
+++ b/net/ethtool/cmis.h
@@ -101,7 +101,6 @@ struct ethtool_cmis_cdb_rpl {
 };
 
 u32 ethtool_cmis_get_max_lpl_size(u8 num_of_byte_octs);
-u32 ethtool_cmis_get_max_epl_size(u8 num_of_byte_octs);
 
 void ethtool_cmis_cdb_compose_args(struct ethtool_cmis_cdb_cmd_args *args,
 				   enum ethtool_cmis_cdb_cmd_id cmd, u8 *lpl,
diff --git a/net/ethtool/cmis_cdb.c b/net/ethtool/cmis_cdb.c
index d159dc121bde..3057576bc81e 100644
--- a/net/ethtool/cmis_cdb.c
+++ b/net/ethtool/cmis_cdb.c
@@ -16,15 +16,6 @@ u32 ethtool_cmis_get_max_lpl_size(u8 num_of_byte_octs)
 	return 8 * (1 + min_t(u8, num_of_byte_octs, 15));
 }
 
-/* For accessing the EPL field on page 9Fh, the allowable length extension is
- * min(i, 255) byte octets where i specifies the allowable additional number of
- * byte octets in a READ or a WRITE.
- */
-u32 ethtool_cmis_get_max_epl_size(u8 num_of_byte_octs)
-{
-	return 8 * (1 + min_t(u8, num_of_byte_octs, 255));
-}
-
 void ethtool_cmis_cdb_compose_args(struct ethtool_cmis_cdb_cmd_args *args,
 				   enum ethtool_cmis_cdb_cmd_id cmd, u8 *lpl,
 				   u8 lpl_len, u8 *epl, u16 epl_len,
@@ -33,19 +24,16 @@ void ethtool_cmis_cdb_compose_args(struct ethtool_cmis_cdb_cmd_args *args,
 {
 	args->req.id = cpu_to_be16(cmd);
 	args->req.lpl_len = lpl_len;
-	if (lpl) {
+	if (lpl)
 		memcpy(args->req.payload, lpl, args->req.lpl_len);
-		args->read_write_len_ext =
-			ethtool_cmis_get_max_lpl_size(read_write_len_ext);
-	}
 	if (epl) {
 		args->req.epl_len = cpu_to_be16(epl_len);
 		args->req.epl = epl;
-		args->read_write_len_ext =
-			ethtool_cmis_get_max_epl_size(read_write_len_ext);
 	}
 
 	args->max_duration = max_duration;
+	args->read_write_len_ext =
+		ethtool_cmis_get_max_lpl_size(read_write_len_ext);
 	args->msleep_pre_rpl = msleep_pre_rpl;
 	args->rpl_exp_len = rpl_exp_len;
 	args->flags = flags;
@@ -363,7 +351,7 @@ ethtool_cmis_module_poll(struct net_device *dev,
 	struct netlink_ext_ack extack = {};
 	int err;
 
-	ethtool_cmis_page_init(&page_data, 0, offset, sizeof(rpl));
+	ethtool_cmis_page_init(&page_data, 0, offset, sizeof(*rpl));
 	page_data.data = (u8 *)rpl;
 
 	err = ops->get_module_eeprom_by_page(dev, &page_data, &extack);
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 0cb6da1f692a..49bea6b45bd5 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -830,6 +830,7 @@ void ethtool_ringparam_get_cfg(struct net_device *dev,
 
 	/* Driver gives us current state, we want to return current config */
 	kparam->tcp_data_split = dev->cfg->hds_config;
+	kparam->hds_thresh = dev->cfg->hds_thresh;
 }
 
 static void ethtool_init_tsinfo(struct kernel_ethtool_ts_info *info)
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 221639407c72..8262cc10f98d 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -60,7 +60,7 @@ static struct devlink *netdev_to_devlink_get(struct net_device *dev)
 u32 ethtool_op_get_link(struct net_device *dev)
 {
 	/* Synchronize carrier state with link watch, see also rtnl_getlink() */
-	linkwatch_sync_dev(dev);
+	__linkwatch_sync_dev(dev);
 
 	return netif_carrier_ok(dev) ? 1 : 0;
 }
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index a163d40c6431..977beeaaa2f9 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -500,7 +500,7 @@ static int ethnl_default_doit(struct sk_buff *skb, struct genl_info *info)
 		netdev_unlock_ops(req_info->dev);
 	rtnl_unlock();
 	if (ret < 0)
-		goto err_cleanup;
+		goto err_dev;
 	ret = ops->reply_size(req_info, reply_data);
 	if (ret < 0)
 		goto err_cleanup;
@@ -560,7 +560,7 @@ static int ethnl_default_dump_one(struct sk_buff *skb, struct net_device *dev,
 	netdev_unlock_ops(dev);
 	rtnl_unlock();
 	if (ret < 0)
-		goto out;
+		goto out_cancel;
 	ret = ethnl_fill_reply_header(skb, dev, ctx->ops->hdr_attr);
 	if (ret < 0)
 		goto out;
@@ -569,6 +569,7 @@ static int ethnl_default_dump_one(struct sk_buff *skb, struct net_device *dev,
 out:
 	if (ctx->ops->cleanup_data)
 		ctx->ops->cleanup_data(ctx->reply_data);
+out_cancel:
 	ctx->reply_data->dev = NULL;
 	if (ret < 0)
 		genlmsg_cancel(skb, ehdr);
@@ -793,7 +794,7 @@ static void ethnl_default_notify(struct net_device *dev, unsigned int cmd,
 	ethnl_init_reply_data(reply_data, ops, dev);
 	ret = ops->prepare_data(req_info, reply_data, &info);
 	if (ret < 0)
-		goto err_cleanup;
+		goto err_rep;
 	ret = ops->reply_size(req_info, reply_data);
 	if (ret < 0)
 		goto err_cleanup;
@@ -828,6 +829,7 @@ err_skb:
 err_cleanup:
 	if (ops->cleanup_data)
 		ops->cleanup_data(reply_data);
+err_rep:
 	kfree(reply_data);
 	kfree(req_info);
 	return;
diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
index 439cfb7ad5d1..1b1b700ec05e 100644
--- a/net/hsr/hsr_device.c
+++ b/net/hsr/hsr_device.c
@@ -33,14 +33,14 @@ static void hsr_set_operstate(struct hsr_port *master, bool has_carrier)
 	struct net_device *dev = master->dev;
 
 	if (!is_admin_up(dev)) {
-		netdev_set_operstate(dev, IF_OPER_DOWN);
+		netif_set_operstate(dev, IF_OPER_DOWN);
 		return;
 	}
 
 	if (has_carrier)
-		netdev_set_operstate(dev, IF_OPER_UP);
+		netif_set_operstate(dev, IF_OPER_UP);
 	else
-		netdev_set_operstate(dev, IF_OPER_LOWERLAYERDOWN);
+		netif_set_operstate(dev, IF_OPER_LOWERLAYERDOWN);
 }
 
 static bool hsr_check_carrier(struct hsr_port *master)
diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
index 934f777f29d3..d293087b426d 100644
--- a/net/ipv4/tcp_offload.c
+++ b/net/ipv4/tcp_offload.c
@@ -439,7 +439,7 @@ static void tcp4_check_fraglist_gro(struct list_head *head, struct sk_buff *skb,
 				       iif, sdif);
 	NAPI_GRO_CB(skb)->is_flist = !sk;
 	if (sk)
-		sock_put(sk);
+		sock_gen_put(sk);
 }
 
 INDIRECT_CALLABLE_SCOPE
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 2c0725583be3..9a8142ccbabe 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -247,6 +247,62 @@ static struct sk_buff *__udpv4_gso_segment_list_csum(struct sk_buff *segs)
 	return segs;
 }
 
+static void __udpv6_gso_segment_csum(struct sk_buff *seg,
+				     struct in6_addr *oldip,
+				     const struct in6_addr *newip,
+				     __be16 *oldport, __be16 newport)
+{
+	struct udphdr *uh = udp_hdr(seg);
+
+	if (ipv6_addr_equal(oldip, newip) && *oldport == newport)
+		return;
+
+	if (uh->check) {
+		inet_proto_csum_replace16(&uh->check, seg, oldip->s6_addr32,
+					  newip->s6_addr32, true);
+
+		inet_proto_csum_replace2(&uh->check, seg, *oldport, newport,
+					 false);
+		if (!uh->check)
+			uh->check = CSUM_MANGLED_0;
+	}
+
+	*oldip = *newip;
+	*oldport = newport;
+}
+
+static struct sk_buff *__udpv6_gso_segment_list_csum(struct sk_buff *segs)
+{
+	const struct ipv6hdr *iph;
+	const struct udphdr *uh;
+	struct ipv6hdr *iph2;
+	struct sk_buff *seg;
+	struct udphdr *uh2;
+
+	seg = segs;
+	uh = udp_hdr(seg);
+	iph = ipv6_hdr(seg);
+	uh2 = udp_hdr(seg->next);
+	iph2 = ipv6_hdr(seg->next);
+
+	if (!(*(const u32 *)&uh->source ^ *(const u32 *)&uh2->source) &&
+	    ipv6_addr_equal(&iph->saddr, &iph2->saddr) &&
+	    ipv6_addr_equal(&iph->daddr, &iph2->daddr))
+		return segs;
+
+	while ((seg = seg->next)) {
+		uh2 = udp_hdr(seg);
+		iph2 = ipv6_hdr(seg);
+
+		__udpv6_gso_segment_csum(seg, &iph2->saddr, &iph->saddr,
+					 &uh2->source, uh->source);
+		__udpv6_gso_segment_csum(seg, &iph2->daddr, &iph->daddr,
+					 &uh2->dest, uh->dest);
+	}
+
+	return segs;
+}
+
 static struct sk_buff *__udp_gso_segment_list(struct sk_buff *skb,
 					      netdev_features_t features,
 					      bool is_ipv6)
@@ -259,7 +315,10 @@ static struct sk_buff *__udp_gso_segment_list(struct sk_buff *skb,
 
 	udp_hdr(skb)->len = htons(sizeof(struct udphdr) + mss);
 
-	return is_ipv6 ? skb : __udpv4_gso_segment_list_csum(skb);
+	if (is_ipv6)
+		return __udpv6_gso_segment_list_csum(skb);
+	else
+		return __udpv4_gso_segment_list_csum(skb);
 }
 
 struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 2cffb8f4a2bc..9ba83f0c9928 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3154,12 +3154,13 @@ int addrconf_add_ifaddr(struct net *net, void __user *arg)
 
 	rtnl_net_lock(net);
 	dev = __dev_get_by_index(net, ireq.ifr6_ifindex);
-	netdev_lock_ops(dev);
-	if (dev)
+	if (dev) {
+		netdev_lock_ops(dev);
 		err = inet6_addr_add(net, dev, &cfg, 0, 0, NULL);
-	else
+		netdev_unlock_ops(dev);
+	} else {
 		err = -ENODEV;
-	netdev_unlock_ops(dev);
+	}
 	rtnl_net_unlock(net);
 	return err;
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ab12b816ab94..96f1621e2381 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -470,10 +470,10 @@ void fib6_select_path(const struct net *net, struct fib6_result *res,
 		goto out;
 
 	hash = fl6->mp_hash;
-	if (hash <= atomic_read(&first->fib6_nh->fib_nh_upper_bound) &&
-	    rt6_score_route(first->fib6_nh, first->fib6_flags, oif,
-			    strict) >= 0) {
-		match = first;
+	if (hash <= atomic_read(&first->fib6_nh->fib_nh_upper_bound)) {
+		if (rt6_score_route(first->fib6_nh, first->fib6_flags, oif,
+				    strict) >= 0)
+			match = first;
 		goto out;
 	}
 
@@ -1771,6 +1771,7 @@ out:
 	if (!err) {
 		spin_lock_bh(&f6i->fib6_table->tb6_lock);
 		fib6_update_sernum(net, f6i);
+		fib6_add_gc_list(f6i);
 		spin_unlock_bh(&f6i->fib6_table->tb6_lock);
 		fib6_force_start_gc(net);
 	}
diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c
index d9b11fe41bf0..a8a04f441e78 100644
--- a/net/ipv6/tcpv6_offload.c
+++ b/net/ipv6/tcpv6_offload.c
@@ -42,7 +42,7 @@ static void tcp6_check_fraglist_gro(struct list_head *head, struct sk_buff *skb,
 					iif, sdif);
 	NAPI_GRO_CB(skb)->is_flist = !sk;
 	if (sk)
-		sock_put(sk);
+		sock_gen_put(sk);
 #endif /* IS_ENABLED(CONFIG_IPV6) */
 }
 
diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c
index ca10916340b0..5432a5f2dfc8 100644
--- a/net/l3mdev/l3mdev.c
+++ b/net/l3mdev/l3mdev.c
@@ -277,8 +277,10 @@ void l3mdev_update_flow(struct net *net, struct flowi *fl)
 	if (fl->flowi_oif) {
 		dev = dev_get_by_index_rcu(net, fl->flowi_oif);
 		if (dev) {
-			if (!fl->flowi_l3mdev)
+			if (!fl->flowi_l3mdev) {
 				fl->flowi_l3mdev = l3mdev_master_ifindex_rcu(dev);
+				fl->flowi_flags |= FLOWI_FLAG_L3MDEV_OIF;
+			}
 
 			/* oif set to L3mdev directs lookup to its table;
 			 * reset to avoid oif match in fib_lookup
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index f0f4a250b10e..969b3e2c496a 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -659,6 +659,9 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do
 	if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
 		ieee80211_txq_remove_vlan(local, sdata);
 
+	if (sdata->vif.txq)
+		ieee80211_txq_purge(sdata->local, to_txq_info(sdata->vif.txq));
+
 	sdata->bss = NULL;
 
 	if (local->open_count == 0)
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index b17b3cc7fb90..a362254b310c 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -1085,7 +1085,13 @@ static void __ieee80211_tx_status(struct ieee80211_hw *hw,
 
 	ieee80211_report_used_skb(local, skb, false, status->ack_hwtstamp);
 
-	if (status->free_list)
+	/*
+	 * This is a bit racy but we can avoid a lot of work
+	 * with this test...
+	 */
+	if (local->tx_mntrs)
+		ieee80211_tx_monitor(local, skb, retry_count, status);
+	else if (status->free_list)
 		list_add_tail(&skb->list, status->free_list);
 	else
 		dev_kfree_skb(skb);
diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c
index dd895617defd..9b12ca97f412 100644
--- a/net/mctp/af_mctp.c
+++ b/net/mctp/af_mctp.c
@@ -630,6 +630,9 @@ static int mctp_sk_hash(struct sock *sk)
 {
 	struct net *net = sock_net(sk);
 
+	/* Bind lookup runs under RCU, remain live during that. */
+	sock_set_flag(sk, SOCK_RCU_FREE);
+
 	mutex_lock(&net->mctp.bind_lock);
 	sk_add_node_rcu(sk, &net->mctp.binds);
 	mutex_unlock(&net->mctp.bind_lock);
diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index 2cb62f026b1f..a715dcbe0146 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -337,7 +337,11 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info)
 
 	release_sock(sk);
 
-	sock_kfree_s(sk, match, sizeof(*match));
+	kfree_rcu_mightsleep(match);
+	/* Adjust sk_omem_alloc like sock_kfree_s() does, to match
+	 * with allocation of this memory by sock_kmemdup()
+	 */
+	atomic_sub(sizeof(*match), &sk->sk_omem_alloc);
 
 	err = 0;
 out:
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 409bd415ef1d..24c2de1891bd 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -899,13 +899,17 @@ create_child:
 				goto dispose_child;
 			}
 
-			if (!subflow_hmac_valid(req, &mp_opt) ||
-			    !mptcp_can_accept_new_subflow(subflow_req->msk)) {
+			if (!subflow_hmac_valid(req, &mp_opt)) {
 				SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC);
 				subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT);
 				goto dispose_child;
 			}
 
+			if (!mptcp_can_accept_new_subflow(owner)) {
+				subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT);
+				goto dispose_child;
+			}
+
 			/* move the msk reference ownership to the subflow */
 			subflow_req->msk = NULL;
 			ctx->conn = (struct sock *)owner;
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index df2dc21304ef..047ba81865ed 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -212,7 +212,7 @@ config NF_CT_PROTO_SCTP
 	bool 'SCTP protocol connection tracking support'
 	depends on NETFILTER_ADVANCED
 	default y
-	select LIBCRC32C
+	select CRC32
 	help
 	  With this option enabled, the layer 3 independent connection
 	  tracking code will be able to do state tracking on SCTP connections.
@@ -475,7 +475,7 @@ endif # NF_CONNTRACK
 
 config NF_TABLES
 	select NETFILTER_NETLINK
-	select LIBCRC32C
+	select CRC32
 	tristate "Netfilter nf_tables support"
 	help
 	  nftables is the new packet classification framework that intends to
diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig
index 2a3017b9c001..8c5b1fe12d07 100644
--- a/net/netfilter/ipvs/Kconfig
+++ b/net/netfilter/ipvs/Kconfig
@@ -105,7 +105,7 @@ config	IP_VS_PROTO_AH
 
 config  IP_VS_PROTO_SCTP
 	bool "SCTP load balancing support"
-	select LIBCRC32C
+	select CRC32
 	help
 	  This option enables support for load balancing SCTP transport
 	  protocol. Say Y if unsure.
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index 9d8361526f82..9441ac3d8c1a 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -383,8 +383,8 @@ static void flow_offload_del(struct nf_flowtable *flow_table,
 void flow_offload_teardown(struct flow_offload *flow)
 {
 	clear_bit(IPS_OFFLOAD_BIT, &flow->ct->status);
-	set_bit(NF_FLOW_TEARDOWN, &flow->flags);
-	flow_offload_fixup_ct(flow);
+	if (!test_and_set_bit(NF_FLOW_TEARDOWN, &flow->flags))
+		flow_offload_fixup_ct(flow);
 }
 EXPORT_SYMBOL_GPL(flow_offload_teardown);
 
@@ -558,10 +558,12 @@ static void nf_flow_offload_gc_step(struct nf_flowtable *flow_table,
 
 	if (nf_flow_has_expired(flow) ||
 	    nf_ct_is_dying(flow->ct) ||
-	    nf_flow_custom_gc(flow_table, flow))
+	    nf_flow_custom_gc(flow_table, flow)) {
 		flow_offload_teardown(flow);
-	else if (!teardown)
+		teardown = true;
+	} else if (!teardown) {
 		nf_flow_table_extend_ct_timeout(flow->ct);
+	}
 
 	if (teardown) {
 		if (test_bit(NF_FLOW_HW, &flow->flags)) {
diff --git a/net/netfilter/nft_set_pipapo_avx2.c b/net/netfilter/nft_set_pipapo_avx2.c
index b8d3c3213efe..c15db28c5ebc 100644
--- a/net/netfilter/nft_set_pipapo_avx2.c
+++ b/net/netfilter/nft_set_pipapo_avx2.c
@@ -994,8 +994,9 @@ static int nft_pipapo_avx2_lookup_8b_16(unsigned long *map, unsigned long *fill,
 		NFT_PIPAPO_AVX2_BUCKET_LOAD8(5, lt,  8,  pkt[8], bsize);
 
 		NFT_PIPAPO_AVX2_AND(6, 2, 3);
+		NFT_PIPAPO_AVX2_AND(3, 4, 7);
 		NFT_PIPAPO_AVX2_BUCKET_LOAD8(7, lt,  9,  pkt[9], bsize);
-		NFT_PIPAPO_AVX2_AND(0, 4, 5);
+		NFT_PIPAPO_AVX2_AND(0, 3, 5);
 		NFT_PIPAPO_AVX2_BUCKET_LOAD8(1, lt, 10, pkt[10], bsize);
 		NFT_PIPAPO_AVX2_AND(2, 6, 7);
 		NFT_PIPAPO_AVX2_BUCKET_LOAD8(3, lt, 11, pkt[11], bsize);
diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig
index 2535f3f9f462..5481bd561eb4 100644
--- a/net/openvswitch/Kconfig
+++ b/net/openvswitch/Kconfig
@@ -11,7 +11,7 @@ config OPENVSWITCH
 				     (!NF_NAT || NF_NAT) && \
 				     (!NETFILTER_CONNCOUNT || NETFILTER_CONNCOUNT)))
 	depends on PSAMPLE || !PSAMPLE
-	select LIBCRC32C
+	select CRC32
 	select MPLS
 	select NET_MPLS_GSO
 	select DST_CACHE
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index 95e0dd14dc1a..518be23e48ea 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -2876,7 +2876,8 @@ static int validate_set(const struct nlattr *a,
 	size_t key_len;
 
 	/* There can be only one key in a action */
-	if (nla_total_size(nla_len(ovs_key)) != nla_len(a))
+	if (!nla_ok(ovs_key, nla_len(a)) ||
+	    nla_total_size(nla_len(ovs_key)) != nla_len(a))
 		return -EINVAL;
 
 	key_len = nla_len(ovs_key);
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 8180d0c12fce..a800127effcd 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -784,7 +784,7 @@ config NET_ACT_SKBEDIT
 config NET_ACT_CSUM
 	tristate "Checksum Updating"
 	depends on NET_CLS_ACT && INET
-	select LIBCRC32C
+	select CRC32
 	help
 	  Say Y here to update some common checksum after some direct
 	  packet alterations.
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 4f648af8cfaa..ecec0a1e1c1a 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -2057,6 +2057,7 @@ static int tcf_fill_node(struct net *net, struct sk_buff *skb,
 	struct tcmsg *tcm;
 	struct nlmsghdr  *nlh;
 	unsigned char *b = skb_tail_pointer(skb);
+	int ret = -EMSGSIZE;
 
 	nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags);
 	if (!nlh)
@@ -2101,11 +2102,45 @@ static int tcf_fill_node(struct net *net, struct sk_buff *skb,
 
 	return skb->len;
 
+cls_op_not_supp:
+	ret = -EOPNOTSUPP;
 out_nlmsg_trim:
 nla_put_failure:
-cls_op_not_supp:
 	nlmsg_trim(skb, b);
-	return -1;
+	return ret;
+}
+
+static struct sk_buff *tfilter_notify_prep(struct net *net,
+					   struct sk_buff *oskb,
+					   struct nlmsghdr *n,
+					   struct tcf_proto *tp,
+					   struct tcf_block *block,
+					   struct Qdisc *q, u32 parent,
+					   void *fh, int event,
+					   u32 portid, bool rtnl_held,
+					   struct netlink_ext_ack *extack)
+{
+	unsigned int size = oskb ? max(NLMSG_GOODSIZE, oskb->len) : NLMSG_GOODSIZE;
+	struct sk_buff *skb;
+	int ret;
+
+retry:
+	skb = alloc_skb(size, GFP_KERNEL);
+	if (!skb)
+		return ERR_PTR(-ENOBUFS);
+
+	ret = tcf_fill_node(net, skb, tp, block, q, parent, fh, portid,
+			    n->nlmsg_seq, n->nlmsg_flags, event, false,
+			    rtnl_held, extack);
+	if (ret <= 0) {
+		kfree_skb(skb);
+		if (ret == -EMSGSIZE) {
+			size += NLMSG_GOODSIZE;
+			goto retry;
+		}
+		return ERR_PTR(-EINVAL);
+	}
+	return skb;
 }
 
 static int tfilter_notify(struct net *net, struct sk_buff *oskb,
@@ -2121,16 +2156,10 @@ static int tfilter_notify(struct net *net, struct sk_buff *oskb,
 	if (!unicast && !rtnl_notify_needed(net, n->nlmsg_flags, RTNLGRP_TC))
 		return 0;
 
-	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
-	if (!skb)
-		return -ENOBUFS;
-
-	if (tcf_fill_node(net, skb, tp, block, q, parent, fh, portid,
-			  n->nlmsg_seq, n->nlmsg_flags, event,
-			  false, rtnl_held, extack) <= 0) {
-		kfree_skb(skb);
-		return -EINVAL;
-	}
+	skb = tfilter_notify_prep(net, oskb, n, tp, block, q, parent, fh, event,
+				  portid, rtnl_held, extack);
+	if (IS_ERR(skb))
+		return PTR_ERR(skb);
 
 	if (unicast)
 		err = rtnl_unicast(skb, net, portid);
@@ -2153,16 +2182,11 @@ static int tfilter_del_notify(struct net *net, struct sk_buff *oskb,
 	if (!rtnl_notify_needed(net, n->nlmsg_flags, RTNLGRP_TC))
 		return tp->ops->delete(tp, fh, last, rtnl_held, extack);
 
-	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
-	if (!skb)
-		return -ENOBUFS;
-
-	if (tcf_fill_node(net, skb, tp, block, q, parent, fh, portid,
-			  n->nlmsg_seq, n->nlmsg_flags, RTM_DELTFILTER,
-			  false, rtnl_held, extack) <= 0) {
+	skb = tfilter_notify_prep(net, oskb, n, tp, block, q, parent, fh,
+				  RTM_DELTFILTER, portid, rtnl_held, extack);
+	if (IS_ERR(skb)) {
 		NL_SET_ERR_MSG(extack, "Failed to build del event notification");
-		kfree_skb(skb);
-		return -EINVAL;
+		return PTR_ERR(skb);
 	}
 
 	err = tp->ops->delete(tp, fh, last, rtnl_held, extack);
diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c
index 81189d02fee7..12dd71139da3 100644
--- a/net/sched/sch_codel.c
+++ b/net/sched/sch_codel.c
@@ -65,10 +65,7 @@ static struct sk_buff *codel_qdisc_dequeue(struct Qdisc *sch)
 			    &q->stats, qdisc_pkt_len, codel_get_enqueue_time,
 			    drop_func, dequeue_func);
 
-	/* We cant call qdisc_tree_reduce_backlog() if our qlen is 0,
-	 * or HTB crashes. Defer it for next round.
-	 */
-	if (q->stats.drop_count && sch->q.qlen) {
+	if (q->stats.drop_count) {
 		qdisc_tree_reduce_backlog(sch, q->stats.drop_count, q->stats.drop_len);
 		q->stats.drop_count = 0;
 		q->stats.drop_len = 0;
diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c
index c69b999fae17..9b6d79bd8737 100644
--- a/net/sched/sch_drr.c
+++ b/net/sched/sch_drr.c
@@ -35,6 +35,11 @@ struct drr_sched {
 	struct Qdisc_class_hash		clhash;
 };
 
+static bool cl_is_active(struct drr_class *cl)
+{
+	return !list_empty(&cl->alist);
+}
+
 static struct drr_class *drr_find_class(struct Qdisc *sch, u32 classid)
 {
 	struct drr_sched *q = qdisc_priv(sch);
@@ -105,6 +110,7 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 		return -ENOBUFS;
 
 	gnet_stats_basic_sync_init(&cl->bstats);
+	INIT_LIST_HEAD(&cl->alist);
 	cl->common.classid = classid;
 	cl->quantum	   = quantum;
 	cl->qdisc	   = qdisc_create_dflt(sch->dev_queue,
@@ -229,7 +235,7 @@ static void drr_qlen_notify(struct Qdisc *csh, unsigned long arg)
 {
 	struct drr_class *cl = (struct drr_class *)arg;
 
-	list_del(&cl->alist);
+	list_del_init(&cl->alist);
 }
 
 static int drr_dump_class(struct Qdisc *sch, unsigned long arg,
@@ -336,7 +342,6 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	struct drr_sched *q = qdisc_priv(sch);
 	struct drr_class *cl;
 	int err = 0;
-	bool first;
 
 	cl = drr_classify(skb, sch, &err);
 	if (cl == NULL) {
@@ -346,7 +351,6 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		return err;
 	}
 
-	first = !cl->qdisc->q.qlen;
 	err = qdisc_enqueue(skb, cl->qdisc, to_free);
 	if (unlikely(err != NET_XMIT_SUCCESS)) {
 		if (net_xmit_drop_count(err)) {
@@ -356,7 +360,7 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		return err;
 	}
 
-	if (first) {
+	if (!cl_is_active(cl)) {
 		list_add_tail(&cl->alist, &q->active);
 		cl->deficit = cl->quantum;
 	}
@@ -390,7 +394,7 @@ static struct sk_buff *drr_dequeue(struct Qdisc *sch)
 			if (unlikely(skb == NULL))
 				goto out;
 			if (cl->qdisc->q.qlen == 0)
-				list_del(&cl->alist);
+				list_del_init(&cl->alist);
 
 			bstats_update(&cl->bstats, skb);
 			qdisc_bstats_update(sch, skb);
@@ -431,7 +435,7 @@ static void drr_reset_qdisc(struct Qdisc *sch)
 	for (i = 0; i < q->clhash.hashsize; i++) {
 		hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) {
 			if (cl->qdisc->q.qlen)
-				list_del(&cl->alist);
+				list_del_init(&cl->alist);
 			qdisc_reset(cl->qdisc);
 		}
 	}
diff --git a/net/sched/sch_ets.c b/net/sched/sch_ets.c
index 516038a44163..2c069f0181c6 100644
--- a/net/sched/sch_ets.c
+++ b/net/sched/sch_ets.c
@@ -74,6 +74,11 @@ static const struct nla_policy ets_class_policy[TCA_ETS_MAX + 1] = {
 	[TCA_ETS_QUANTA_BAND] = { .type = NLA_U32 },
 };
 
+static bool cl_is_active(struct ets_class *cl)
+{
+	return !list_empty(&cl->alist);
+}
+
 static int ets_quantum_parse(struct Qdisc *sch, const struct nlattr *attr,
 			     unsigned int *quantum,
 			     struct netlink_ext_ack *extack)
@@ -293,7 +298,7 @@ static void ets_class_qlen_notify(struct Qdisc *sch, unsigned long arg)
 	 * to remove them.
 	 */
 	if (!ets_class_is_strict(q, cl) && sch->q.qlen)
-		list_del(&cl->alist);
+		list_del_init(&cl->alist);
 }
 
 static int ets_class_dump(struct Qdisc *sch, unsigned long arg,
@@ -416,7 +421,6 @@ static int ets_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	struct ets_sched *q = qdisc_priv(sch);
 	struct ets_class *cl;
 	int err = 0;
-	bool first;
 
 	cl = ets_classify(skb, sch, &err);
 	if (!cl) {
@@ -426,7 +430,6 @@ static int ets_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		return err;
 	}
 
-	first = !cl->qdisc->q.qlen;
 	err = qdisc_enqueue(skb, cl->qdisc, to_free);
 	if (unlikely(err != NET_XMIT_SUCCESS)) {
 		if (net_xmit_drop_count(err)) {
@@ -436,7 +439,7 @@ static int ets_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		return err;
 	}
 
-	if (first && !ets_class_is_strict(q, cl)) {
+	if (!cl_is_active(cl) && !ets_class_is_strict(q, cl)) {
 		list_add_tail(&cl->alist, &q->active);
 		cl->deficit = cl->quantum;
 	}
@@ -488,7 +491,7 @@ static struct sk_buff *ets_qdisc_dequeue(struct Qdisc *sch)
 			if (unlikely(!skb))
 				goto out;
 			if (cl->qdisc->q.qlen == 0)
-				list_del(&cl->alist);
+				list_del_init(&cl->alist);
 			return ets_qdisc_dequeue_skb(sch, skb);
 		}
 
@@ -657,7 +660,7 @@ static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt,
 	}
 	for (i = q->nbands; i < oldbands; i++) {
 		if (i >= q->nstrict && q->classes[i].qdisc->q.qlen)
-			list_del(&q->classes[i].alist);
+			list_del_init(&q->classes[i].alist);
 		qdisc_tree_flush_backlog(q->classes[i].qdisc);
 	}
 	WRITE_ONCE(q->nstrict, nstrict);
@@ -713,7 +716,7 @@ static void ets_qdisc_reset(struct Qdisc *sch)
 
 	for (band = q->nstrict; band < q->nbands; band++) {
 		if (q->classes[band].qdisc->q.qlen)
-			list_del(&q->classes[band].alist);
+			list_del_init(&q->classes[band].alist);
 	}
 	for (band = 0; band < q->nbands; band++)
 		qdisc_reset(q->classes[band].qdisc);
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index 799f5397ad4c..6c9029f71e88 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -315,10 +315,8 @@ begin:
 	}
 	qdisc_bstats_update(sch, skb);
 	flow->deficit -= qdisc_pkt_len(skb);
-	/* We cant call qdisc_tree_reduce_backlog() if our qlen is 0,
-	 * or HTB crashes. Defer it for next round.
-	 */
-	if (q->cstats.drop_count && sch->q.qlen) {
+
+	if (q->cstats.drop_count) {
 		qdisc_tree_reduce_backlog(sch, q->cstats.drop_count,
 					  q->cstats.drop_len);
 		q->cstats.drop_count = 0;
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index c287bf8423b4..cb8c525ea20e 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -203,7 +203,10 @@ eltree_insert(struct hfsc_class *cl)
 static inline void
 eltree_remove(struct hfsc_class *cl)
 {
-	rb_erase(&cl->el_node, &cl->sched->eligible);
+	if (!RB_EMPTY_NODE(&cl->el_node)) {
+		rb_erase(&cl->el_node, &cl->sched->eligible);
+		RB_CLEAR_NODE(&cl->el_node);
+	}
 }
 
 static inline void
@@ -958,6 +961,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 	if (cl != NULL) {
 		int old_flags;
+		int len = 0;
 
 		if (parentid) {
 			if (cl->cl_parent &&
@@ -988,9 +992,13 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 		if (usc != NULL)
 			hfsc_change_usc(cl, usc, cur_time);
 
+		if (cl->qdisc->q.qlen != 0)
+			len = qdisc_peek_len(cl->qdisc);
+		/* Check queue length again since some qdisc implementations
+		 * (e.g., netem/codel) might empty the queue during the peek
+		 * operation.
+		 */
 		if (cl->qdisc->q.qlen != 0) {
-			int len = qdisc_peek_len(cl->qdisc);
-
 			if (cl->cl_flags & HFSC_RSC) {
 				if (old_flags & HFSC_RSC)
 					update_ed(cl, len);
@@ -1220,7 +1228,8 @@ hfsc_qlen_notify(struct Qdisc *sch, unsigned long arg)
 	/* vttree is now handled in update_vf() so that update_vf(cl, 0, 0)
 	 * needs to be called explicitly to remove a class from vttree.
 	 */
-	update_vf(cl, 0, 0);
+	if (cl->cl_nactive)
+		update_vf(cl, 0, 0);
 	if (cl->cl_flags & HFSC_RSC)
 		eltree_remove(cl);
 }
@@ -1560,7 +1569,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free)
 		return err;
 	}
 
-	if (first) {
+	if (first && !cl->cl_nactive) {
 		if (cl->cl_flags & HFSC_RSC)
 			init_ed(cl, len);
 		if (cl->cl_flags & HFSC_FSC)
@@ -1632,10 +1641,16 @@ hfsc_dequeue(struct Qdisc *sch)
 		if (cl->qdisc->q.qlen != 0) {
 			/* update ed */
 			next_len = qdisc_peek_len(cl->qdisc);
-			if (realtime)
-				update_ed(cl, next_len);
-			else
-				update_d(cl, next_len);
+			/* Check queue length again since some qdisc implementations
+			 * (e.g., netem/codel) might empty the queue during the peek
+			 * operation.
+			 */
+			if (cl->qdisc->q.qlen != 0) {
+				if (realtime)
+					update_ed(cl, next_len);
+				else
+					update_d(cl, next_len);
+			}
 		} else {
 			/* the class becomes passive */
 			eltree_remove(cl);
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index c31bc5489bdd..4b9a639b642e 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1485,6 +1485,8 @@ static void htb_qlen_notify(struct Qdisc *sch, unsigned long arg)
 {
 	struct htb_class *cl = (struct htb_class *)arg;
 
+	if (!cl->prio_activity)
+		return;
 	htb_deactivate(qdisc_priv(sch), cl);
 }
 
diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index 2cfbc977fe6d..bf1282cb22eb 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -202,6 +202,11 @@ struct qfq_sched {
  */
 enum update_reason {enqueue, requeue};
 
+static bool cl_is_active(struct qfq_class *cl)
+{
+	return !list_empty(&cl->alist);
+}
+
 static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid)
 {
 	struct qfq_sched *q = qdisc_priv(sch);
@@ -347,7 +352,7 @@ static void qfq_deactivate_class(struct qfq_sched *q, struct qfq_class *cl)
 	struct qfq_aggregate *agg = cl->agg;
 
 
-	list_del(&cl->alist); /* remove from RR queue of the aggregate */
+	list_del_init(&cl->alist); /* remove from RR queue of the aggregate */
 	if (list_empty(&agg->active)) /* agg is now inactive */
 		qfq_deactivate_agg(q, agg);
 }
@@ -474,6 +479,7 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	gnet_stats_basic_sync_init(&cl->bstats);
 	cl->common.classid = classid;
 	cl->deficit = lmax;
+	INIT_LIST_HEAD(&cl->alist);
 
 	cl->qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops,
 				      classid, NULL);
@@ -982,7 +988,7 @@ static struct sk_buff *agg_dequeue(struct qfq_aggregate *agg,
 	cl->deficit -= (int) len;
 
 	if (cl->qdisc->q.qlen == 0) /* no more packets, remove from list */
-		list_del(&cl->alist);
+		list_del_init(&cl->alist);
 	else if (cl->deficit < qdisc_pkt_len(cl->qdisc->ops->peek(cl->qdisc))) {
 		cl->deficit += agg->lmax;
 		list_move_tail(&cl->alist, &agg->active);
@@ -1214,7 +1220,6 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	struct qfq_class *cl;
 	struct qfq_aggregate *agg;
 	int err = 0;
-	bool first;
 
 	cl = qfq_classify(skb, sch, &err);
 	if (cl == NULL) {
@@ -1236,7 +1241,6 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	}
 
 	gso_segs = skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1;
-	first = !cl->qdisc->q.qlen;
 	err = qdisc_enqueue(skb, cl->qdisc, to_free);
 	if (unlikely(err != NET_XMIT_SUCCESS)) {
 		pr_debug("qfq_enqueue: enqueue failed %d\n", err);
@@ -1252,8 +1256,8 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	++sch->q.qlen;
 
 	agg = cl->agg;
-	/* if the queue was not empty, then done here */
-	if (!first) {
+	/* if the class is active, then done here */
+	if (cl_is_active(cl)) {
 		if (unlikely(skb == cl->qdisc->ops->peek(cl->qdisc)) &&
 		    list_first_entry(&agg->active, struct qfq_class, alist)
 		    == cl && cl->deficit < len)
@@ -1415,6 +1419,8 @@ static void qfq_qlen_notify(struct Qdisc *sch, unsigned long arg)
 	struct qfq_sched *q = qdisc_priv(sch);
 	struct qfq_class *cl = (struct qfq_class *)arg;
 
+	if (list_empty(&cl->alist))
+		return;
 	qfq_deactivate_class(q, cl);
 }
 
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 9ed197e96396..b912ad99aa15 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -631,6 +631,15 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt,
 	struct red_parms *p = NULL;
 	struct sk_buff *to_free = NULL;
 	struct sk_buff *tail = NULL;
+	unsigned int maxflows;
+	unsigned int quantum;
+	unsigned int divisor;
+	int perturb_period;
+	u8 headdrop;
+	u8 maxdepth;
+	int limit;
+	u8 flags;
+
 
 	if (opt->nla_len < nla_attr_size(sizeof(*ctl)))
 		return -EINVAL;
@@ -652,39 +661,64 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt,
 		if (!p)
 			return -ENOMEM;
 	}
-	if (ctl->limit == 1) {
-		NL_SET_ERR_MSG_MOD(extack, "invalid limit");
-		return -EINVAL;
-	}
+
 	sch_tree_lock(sch);
+
+	limit = q->limit;
+	divisor = q->divisor;
+	headdrop = q->headdrop;
+	maxdepth = q->maxdepth;
+	maxflows = q->maxflows;
+	perturb_period = q->perturb_period;
+	quantum = q->quantum;
+	flags = q->flags;
+
+	/* update and validate configuration */
 	if (ctl->quantum)
-		q->quantum = ctl->quantum;
-	WRITE_ONCE(q->perturb_period, ctl->perturb_period * HZ);
+		quantum = ctl->quantum;
+	perturb_period = ctl->perturb_period * HZ;
 	if (ctl->flows)
-		q->maxflows = min_t(u32, ctl->flows, SFQ_MAX_FLOWS);
+		maxflows = min_t(u32, ctl->flows, SFQ_MAX_FLOWS);
 	if (ctl->divisor) {
-		q->divisor = ctl->divisor;
-		q->maxflows = min_t(u32, q->maxflows, q->divisor);
+		divisor = ctl->divisor;
+		maxflows = min_t(u32, maxflows, divisor);
 	}
 	if (ctl_v1) {
 		if (ctl_v1->depth)
-			q->maxdepth = min_t(u32, ctl_v1->depth, SFQ_MAX_DEPTH);
+			maxdepth = min_t(u32, ctl_v1->depth, SFQ_MAX_DEPTH);
 		if (p) {
-			swap(q->red_parms, p);
-			red_set_parms(q->red_parms,
+			red_set_parms(p,
 				      ctl_v1->qth_min, ctl_v1->qth_max,
 				      ctl_v1->Wlog,
 				      ctl_v1->Plog, ctl_v1->Scell_log,
 				      NULL,
 				      ctl_v1->max_P);
 		}
-		q->flags = ctl_v1->flags;
-		q->headdrop = ctl_v1->headdrop;
+		flags = ctl_v1->flags;
+		headdrop = ctl_v1->headdrop;
 	}
 	if (ctl->limit) {
-		q->limit = min_t(u32, ctl->limit, q->maxdepth * q->maxflows);
-		q->maxflows = min_t(u32, q->maxflows, q->limit);
+		limit = min_t(u32, ctl->limit, maxdepth * maxflows);
+		maxflows = min_t(u32, maxflows, limit);
 	}
+	if (limit == 1) {
+		sch_tree_unlock(sch);
+		kfree(p);
+		NL_SET_ERR_MSG_MOD(extack, "invalid limit");
+		return -EINVAL;
+	}
+
+	/* commit configuration */
+	q->limit = limit;
+	q->divisor = divisor;
+	q->headdrop = headdrop;
+	q->maxdepth = maxdepth;
+	q->maxflows = maxflows;
+	WRITE_ONCE(q->perturb_period, perturb_period);
+	q->quantum = quantum;
+	q->flags = flags;
+	if (p)
+		swap(q->red_parms, p);
 
 	qlen = sch->q.qlen;
 	while (sch->q.qlen > q->limit) {
diff --git a/net/sctp/Kconfig b/net/sctp/Kconfig
index 5da599ff84a9..d18a72df3654 100644
--- a/net/sctp/Kconfig
+++ b/net/sctp/Kconfig
@@ -7,10 +7,10 @@ menuconfig IP_SCTP
 	tristate "The SCTP Protocol"
 	depends on INET
 	depends on IPV6 || IPV6=n
+	select CRC32
 	select CRYPTO
 	select CRYPTO_HMAC
 	select CRYPTO_SHA1
-	select LIBCRC32C
 	select NET_UDP_TUNNEL
 	help
 	  Stream Control Transmission Protocol
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 36ee34f483d7..53725ee7ba06 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -72,8 +72,9 @@
 /* Forward declarations for internal helper functions. */
 static bool sctp_writeable(const struct sock *sk);
 static void sctp_wfree(struct sk_buff *skb);
-static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
-				size_t msg_len);
+static int sctp_wait_for_sndbuf(struct sctp_association *asoc,
+				struct sctp_transport *transport,
+				long *timeo_p, size_t msg_len);
 static int sctp_wait_for_packet(struct sock *sk, int *err, long *timeo_p);
 static int sctp_wait_for_connect(struct sctp_association *, long *timeo_p);
 static int sctp_wait_for_accept(struct sock *sk, long timeo);
@@ -1828,7 +1829,7 @@ static int sctp_sendmsg_to_asoc(struct sctp_association *asoc,
 
 	if (sctp_wspace(asoc) <= 0 || !sk_wmem_schedule(sk, msg_len)) {
 		timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
-		err = sctp_wait_for_sndbuf(asoc, &timeo, msg_len);
+		err = sctp_wait_for_sndbuf(asoc, transport, &timeo, msg_len);
 		if (err)
 			goto err;
 		if (unlikely(sinfo->sinfo_stream >= asoc->stream.outcnt)) {
@@ -9214,8 +9215,9 @@ void sctp_sock_rfree(struct sk_buff *skb)
 
 
 /* Helper function to wait for space in the sndbuf.  */
-static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
-				size_t msg_len)
+static int sctp_wait_for_sndbuf(struct sctp_association *asoc,
+				struct sctp_transport *transport,
+				long *timeo_p, size_t msg_len)
 {
 	struct sock *sk = asoc->base.sk;
 	long current_timeo = *timeo_p;
@@ -9225,7 +9227,9 @@ static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
 	pr_debug("%s: asoc:%p, timeo:%ld, msg_len:%zu\n", __func__, asoc,
 		 *timeo_p, msg_len);
 
-	/* Increment the association's refcnt.  */
+	/* Increment the transport and association's refcnt. */
+	if (transport)
+		sctp_transport_hold(transport);
 	sctp_association_hold(asoc);
 
 	/* Wait on the association specific sndbuf space. */
@@ -9234,7 +9238,7 @@ static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
 					  TASK_INTERRUPTIBLE);
 		if (asoc->base.dead)
 			goto do_dead;
-		if (!*timeo_p)
+		if ((!*timeo_p) || (transport && transport->dead))
 			goto do_nonblock;
 		if (sk->sk_err || asoc->state >= SCTP_STATE_SHUTDOWN_PENDING)
 			goto do_error;
@@ -9259,7 +9263,9 @@ static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
 out:
 	finish_wait(&asoc->wait, &wait);
 
-	/* Release the association's refcnt.  */
+	/* Release the transport and association's refcnt. */
+	if (transport)
+		sctp_transport_put(transport);
 	sctp_association_put(asoc);
 
 	return err;
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index 59675f6d9e7d..6946c1462793 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -117,6 +117,8 @@ fail:
  */
 void sctp_transport_free(struct sctp_transport *transport)
 {
+	transport->dead = 1;
+
 	/* Try to delete the heartbeat timer.  */
 	if (timer_delete(&transport->hb_timer))
 		sctp_transport_put(transport);
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 3e6cb35baf25..3760131f1484 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -362,6 +362,9 @@ static void smc_destruct(struct sock *sk)
 		return;
 }
 
+static struct lock_class_key smc_key;
+static struct lock_class_key smc_slock_key;
+
 void smc_sk_init(struct net *net, struct sock *sk, int protocol)
 {
 	struct smc_sock *smc = smc_sk(sk);
@@ -375,6 +378,8 @@ void smc_sk_init(struct net *net, struct sock *sk, int protocol)
 	INIT_WORK(&smc->connect_work, smc_connect_work);
 	INIT_DELAYED_WORK(&smc->conn.tx_work, smc_tx_work);
 	INIT_LIST_HEAD(&smc->accept_q);
+	sock_lock_init_class_and_name(sk, "slock-AF_SMC", &smc_slock_key,
+				      "sk_lock-AF_SMC", &smc_key);
 	spin_lock_init(&smc->accept_q_lock);
 	spin_lock_init(&smc->conn.send_lock);
 	sk->sk_prot->hash(sk);
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index 004cdb59f010..7ce5e28a6c03 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -1536,13 +1536,9 @@ static ssize_t write_flush(struct file *file, const char __user *buf,
 	 * or by one second if it has already reached the current time.
 	 * Newly added cache entries will always have ->last_refresh greater
 	 * that ->flush_time, so they don't get flushed prematurely.
-	 *
-	 * If someone frequently calls the flush interface, we should
-	 * immediately clean the corresponding cache_detail instead of
-	 * continuously accumulating nextcheck.
 	 */
 
-	if (cd->flush_time >= now && cd->flush_time < (now + 5))
+	if (cd->flush_time >= now)
 		now = cd->flush_time + 1;
 
 	cd->flush_time = now;
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 50c2e0846ea4..18be6ff4c3db 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1046,6 +1046,7 @@ int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
 	if (unlikely(l->backlog[imp].len >= l->backlog[imp].limit)) {
 		if (imp == TIPC_SYSTEM_IMPORTANCE) {
 			pr_warn("%s<%s>, link overflow", link_rst_msg, l->name);
+			__skb_queue_purge(list);
 			return -ENOBUFS;
 		}
 		rc = link_schedule_user(l, hdr);
diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c
index e2f19627e43d..b45c5b91bc7a 100644
--- a/net/tipc/monitor.c
+++ b/net/tipc/monitor.c
@@ -716,7 +716,8 @@ void tipc_mon_reinit_self(struct net *net)
 		if (!mon)
 			continue;
 		write_lock_bh(&mon->lock);
-		mon->self->addr = tipc_own_addr(net);
+		if (mon->self)
+			mon->self->addr = tipc_own_addr(net);
 		write_unlock_bh(&mon->lock);
 	}
 }
diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
index cb86b0bf9a53..a3ccb3135e51 100644
--- a/net/tls/tls_main.c
+++ b/net/tls/tls_main.c
@@ -852,6 +852,11 @@ static int tls_setsockopt(struct sock *sk, int level, int optname,
 	return do_tls_setsockopt(sk, optname, optval, optlen);
 }
 
+static int tls_disconnect(struct sock *sk, int flags)
+{
+	return -EOPNOTSUPP;
+}
+
 struct tls_context *tls_ctx_create(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
@@ -947,6 +952,7 @@ static void build_protos(struct proto prot[TLS_NUM_CONFIG][TLS_NUM_CONFIG],
 	prot[TLS_BASE][TLS_BASE] = *base;
 	prot[TLS_BASE][TLS_BASE].setsockopt	= tls_setsockopt;
 	prot[TLS_BASE][TLS_BASE].getsockopt	= tls_getsockopt;
+	prot[TLS_BASE][TLS_BASE].disconnect	= tls_disconnect;
 	prot[TLS_BASE][TLS_BASE].close		= tls_sk_proto_close;
 
 	prot[TLS_SW][TLS_BASE] = prot[TLS_BASE][TLS_BASE];
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 5696af45bcf7..4abc81f33d3e 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -338,13 +338,14 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	u32 len = xdp_get_buff_len(xdp);
 	int err;
 
-	spin_lock_bh(&xs->rx_lock);
 	err = xsk_rcv_check(xs, xdp, len);
 	if (!err) {
+		spin_lock_bh(&xs->pool->rx_lock);
 		err = __xsk_rcv(xs, xdp, len);
 		xsk_flush(xs);
+		spin_unlock_bh(&xs->pool->rx_lock);
 	}
-	spin_unlock_bh(&xs->rx_lock);
+
 	return err;
 }
 
@@ -1734,7 +1735,6 @@ static int xsk_create(struct net *net, struct socket *sock, int protocol,
 	xs = xdp_sk(sk);
 	xs->state = XSK_READY;
 	mutex_init(&xs->mutex);
-	spin_lock_init(&xs->rx_lock);
 
 	INIT_LIST_HEAD(&xs->map_list);
 	spin_lock_init(&xs->map_list_lock);
diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c
index 25a76c5ce0f1..c5181a9044ad 100644
--- a/net/xdp/xsk_buff_pool.c
+++ b/net/xdp/xsk_buff_pool.c
@@ -89,6 +89,7 @@ struct xsk_buff_pool *xp_create_and_assign_umem(struct xdp_sock *xs,
 	pool->addrs = umem->addrs;
 	pool->tx_metadata_len = umem->tx_metadata_len;
 	pool->tx_sw_csum = umem->flags & XDP_UMEM_TX_SW_CSUM;
+	spin_lock_init(&pool->rx_lock);
 	INIT_LIST_HEAD(&pool->free_list);
 	INIT_LIST_HEAD(&pool->xskb_list);
 	INIT_LIST_HEAD(&pool->xsk_tx_list);
diff --git a/rust/Makefile b/rust/Makefile
index fa0eea8a9eca..3aca903a7d08 100644
--- a/rust/Makefile
+++ b/rust/Makefile
@@ -368,7 +368,7 @@ $(obj)/bindings/bindings_helpers_generated.rs: private bindgen_target_extra = ;
 $(obj)/bindings/bindings_helpers_generated.rs: $(src)/helpers/helpers.c FORCE
 	$(call if_changed_dep,bindgen)
 
-rust_exports = $(NM) -p --defined-only $(1) | awk '$$2~/(T|R|D|B)/ && $$3!~/__cfi/ && $$3!~/__odr_asan/ { printf $(2),$$3 }'
+rust_exports = $(NM) -p --defined-only $(1) | awk '$$2~/(T|R|D|B)/ && $$3!~/__(pfx|cfi|odr_asan)/ { printf $(2),$$3 }'
 
 quiet_cmd_exports = EXPORTS $@
       cmd_exports = \
diff --git a/rust/helpers/dma.c b/rust/helpers/dma.c
new file mode 100644
index 000000000000..df8b8a77355a
--- /dev/null
+++ b/rust/helpers/dma.c
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/dma-mapping.h>
+
+void *rust_helper_dma_alloc_attrs(struct device *dev, size_t size,
+				  dma_addr_t *dma_handle, gfp_t flag,
+				  unsigned long attrs)
+{
+	return dma_alloc_attrs(dev, size, dma_handle, flag, attrs);
+}
+
+void rust_helper_dma_free_attrs(struct device *dev, size_t size, void *cpu_addr,
+				dma_addr_t dma_handle, unsigned long attrs)
+{
+	dma_free_attrs(dev, size, cpu_addr, dma_handle, attrs);
+}
diff --git a/rust/helpers/helpers.c b/rust/helpers/helpers.c
index 6c205454a18e..c228a5363212 100644
--- a/rust/helpers/helpers.c
+++ b/rust/helpers/helpers.c
@@ -15,6 +15,7 @@
 #include "cpumask.c"
 #include "cred.c"
 #include "device.c"
+#include "dma.c"
 #include "drm.c"
 #include "err.c"
 #include "fs.c"
diff --git a/rust/helpers/io.c b/rust/helpers/io.c
index 4c2401ccd720..15ea187c5466 100644
--- a/rust/helpers/io.c
+++ b/rust/helpers/io.c
@@ -7,94 +7,94 @@ void __iomem *rust_helper_ioremap(phys_addr_t offset, size_t size)
 	return ioremap(offset, size);
 }
 
-void rust_helper_iounmap(volatile void __iomem *addr)
+void rust_helper_iounmap(void __iomem *addr)
 {
 	iounmap(addr);
 }
 
-u8 rust_helper_readb(const volatile void __iomem *addr)
+u8 rust_helper_readb(const void __iomem *addr)
 {
 	return readb(addr);
 }
 
-u16 rust_helper_readw(const volatile void __iomem *addr)
+u16 rust_helper_readw(const void __iomem *addr)
 {
 	return readw(addr);
 }
 
-u32 rust_helper_readl(const volatile void __iomem *addr)
+u32 rust_helper_readl(const void __iomem *addr)
 {
 	return readl(addr);
 }
 
 #ifdef CONFIG_64BIT
-u64 rust_helper_readq(const volatile void __iomem *addr)
+u64 rust_helper_readq(const void __iomem *addr)
 {
 	return readq(addr);
 }
 #endif
 
-void rust_helper_writeb(u8 value, volatile void __iomem *addr)
+void rust_helper_writeb(u8 value, void __iomem *addr)
 {
 	writeb(value, addr);
 }
 
-void rust_helper_writew(u16 value, volatile void __iomem *addr)
+void rust_helper_writew(u16 value, void __iomem *addr)
 {
 	writew(value, addr);
 }
 
-void rust_helper_writel(u32 value, volatile void __iomem *addr)
+void rust_helper_writel(u32 value, void __iomem *addr)
 {
 	writel(value, addr);
 }
 
 #ifdef CONFIG_64BIT
-void rust_helper_writeq(u64 value, volatile void __iomem *addr)
+void rust_helper_writeq(u64 value, void __iomem *addr)
 {
 	writeq(value, addr);
 }
 #endif
 
-u8 rust_helper_readb_relaxed(const volatile void __iomem *addr)
+u8 rust_helper_readb_relaxed(const void __iomem *addr)
 {
 	return readb_relaxed(addr);
 }
 
-u16 rust_helper_readw_relaxed(const volatile void __iomem *addr)
+u16 rust_helper_readw_relaxed(const void __iomem *addr)
 {
 	return readw_relaxed(addr);
 }
 
-u32 rust_helper_readl_relaxed(const volatile void __iomem *addr)
+u32 rust_helper_readl_relaxed(const void __iomem *addr)
 {
 	return readl_relaxed(addr);
 }
 
 #ifdef CONFIG_64BIT
-u64 rust_helper_readq_relaxed(const volatile void __iomem *addr)
+u64 rust_helper_readq_relaxed(const void __iomem *addr)
 {
 	return readq_relaxed(addr);
 }
 #endif
 
-void rust_helper_writeb_relaxed(u8 value, volatile void __iomem *addr)
+void rust_helper_writeb_relaxed(u8 value, void __iomem *addr)
 {
 	writeb_relaxed(value, addr);
 }
 
-void rust_helper_writew_relaxed(u16 value, volatile void __iomem *addr)
+void rust_helper_writew_relaxed(u16 value, void __iomem *addr)
 {
 	writew_relaxed(value, addr);
 }
 
-void rust_helper_writel_relaxed(u32 value, volatile void __iomem *addr)
+void rust_helper_writel_relaxed(u32 value, void __iomem *addr)
 {
 	writel_relaxed(value, addr);
 }
 
 #ifdef CONFIG_64BIT
-void rust_helper_writeq_relaxed(u64 value, volatile void __iomem *addr)
+void rust_helper_writeq_relaxed(u64 value, void __iomem *addr)
 {
 	writeq_relaxed(value, addr);
 }
diff --git a/rust/kernel/firmware.rs b/rust/kernel/firmware.rs
index f04b058b09b2..2494c96e105f 100644
--- a/rust/kernel/firmware.rs
+++ b/rust/kernel/firmware.rs
@@ -4,7 +4,7 @@
 //!
 //! C header: [`include/linux/firmware.h`](srctree/include/linux/firmware.h)
 
-use crate::{bindings, device::Device, error::Error, error::Result, str::CStr};
+use crate::{bindings, device::Device, error::Error, error::Result, ffi, str::CStr};
 use core::ptr::NonNull;
 
 /// # Invariants
@@ -12,7 +12,11 @@ use core::ptr::NonNull;
 /// One of the following: `bindings::request_firmware`, `bindings::firmware_request_nowarn`,
 /// `bindings::firmware_request_platform`, `bindings::request_firmware_direct`.
 struct FwFunc(
-    unsafe extern "C" fn(*mut *const bindings::firmware, *const u8, *mut bindings::device) -> i32,
+    unsafe extern "C" fn(
+        *mut *const bindings::firmware,
+        *const ffi::c_char,
+        *mut bindings::device,
+    ) -> i32,
 );
 
 impl FwFunc {
diff --git a/rust/pin-init/examples/pthread_mutex.rs b/rust/pin-init/examples/pthread_mutex.rs
index 9164298c44c0..5ac22f1880d2 100644
--- a/rust/pin-init/examples/pthread_mutex.rs
+++ b/rust/pin-init/examples/pthread_mutex.rs
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: Apache-2.0 OR MIT
 
-// inspired by https://github.com/nbdd0121/pin-init/blob/trunk/examples/pthread_mutex.rs
+// inspired by <https://github.com/nbdd0121/pin-init/blob/trunk/examples/pthread_mutex.rs>
 #![allow(clippy::undocumented_unsafe_blocks)]
 #![cfg_attr(feature = "alloc", feature(allocator_api))]
 #[cfg(not(windows))]
diff --git a/rust/pin-init/src/alloc.rs b/rust/pin-init/src/alloc.rs
index e16baa3b434e..5017f57442d8 100644
--- a/rust/pin-init/src/alloc.rs
+++ b/rust/pin-init/src/alloc.rs
@@ -17,11 +17,9 @@ use crate::{
 
 pub extern crate alloc;
 
-// SAFETY: All zeros is equivalent to `None` (option layout optimization guarantee).
-//
-// In this case we are allowed to use `T: ?Sized`, since all zeros is the `None` variant and there
-// is no problem with a VTABLE pointer being null.
-unsafe impl<T: ?Sized> ZeroableOption for Box<T> {}
+// SAFETY: All zeros is equivalent to `None` (option layout optimization guarantee:
+// <https://doc.rust-lang.org/stable/std/option/index.html#representation>).
+unsafe impl<T> ZeroableOption for Box<T> {}
 
 /// Smart pointer that can initialize memory in-place.
 pub trait InPlaceInit<T>: Sized {
diff --git a/rust/pin-init/src/lib.rs b/rust/pin-init/src/lib.rs
index 05c44514765e..0806c689f693 100644
--- a/rust/pin-init/src/lib.rs
+++ b/rust/pin-init/src/lib.rs
@@ -1447,7 +1447,7 @@ impl_zeroable! {
     {<T: ?Sized + Zeroable>} UnsafeCell<T>,
 
     // SAFETY: All zeros is equivalent to `None` (option layout optimization guarantee:
-    // https://doc.rust-lang.org/stable/std/option/index.html#representation).
+    // <https://doc.rust-lang.org/stable/std/option/index.html#representation>).
     Option<NonZeroU8>, Option<NonZeroU16>, Option<NonZeroU32>, Option<NonZeroU64>,
     Option<NonZeroU128>, Option<NonZeroUsize>,
     Option<NonZeroI8>, Option<NonZeroI16>, Option<NonZeroI32>, Option<NonZeroI64>,
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 5b632635e00d..95a4fa1f1e44 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -376,7 +376,7 @@ $(obj)/%.o: $(src)/%.c
 	@echo "  CLANG-bpf " $@
 	$(Q)$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(BPF_EXTRA_CFLAGS) \
 		-I$(obj) -I$(srctree)/tools/testing/selftests/bpf/ \
-		-I$(LIBBPF_INCLUDE) \
+		-I$(LIBBPF_INCLUDE) $(CLANG_SYS_INCLUDES) \
 		-D__KERNEL__ -D__BPF_TRACING__ -Wno-unused-value -Wno-pointer-sign \
 		-D__TARGET_ARCH_$(SRCARCH) -Wno-compare-distinct-pointer-types \
 		-Wno-gnu-variable-sized-type-not-at-end \
diff --git a/samples/livepatch/livepatch-callbacks-busymod.c b/samples/livepatch/livepatch-callbacks-busymod.c
index 69105596e72e..fadc2a85cb35 100644
--- a/samples/livepatch/livepatch-callbacks-busymod.c
+++ b/samples/livepatch/livepatch-callbacks-busymod.c
@@ -56,4 +56,5 @@ static void livepatch_callbacks_mod_exit(void)
 
 module_init(livepatch_callbacks_mod_init);
 module_exit(livepatch_callbacks_mod_exit);
+MODULE_DESCRIPTION("Live patching demo for (un)patching callbacks, support module");
 MODULE_LICENSE("GPL");
diff --git a/samples/livepatch/livepatch-callbacks-demo.c b/samples/livepatch/livepatch-callbacks-demo.c
index 11c3f4357812..9e69d9caed25 100644
--- a/samples/livepatch/livepatch-callbacks-demo.c
+++ b/samples/livepatch/livepatch-callbacks-demo.c
@@ -192,5 +192,6 @@ static void livepatch_callbacks_demo_exit(void)
 
 module_init(livepatch_callbacks_demo_init);
 module_exit(livepatch_callbacks_demo_exit);
+MODULE_DESCRIPTION("Live patching demo for (un)patching callbacks");
 MODULE_LICENSE("GPL");
 MODULE_INFO(livepatch, "Y");
diff --git a/samples/livepatch/livepatch-callbacks-mod.c b/samples/livepatch/livepatch-callbacks-mod.c
index 2a074f422a51..d1851b471ad9 100644
--- a/samples/livepatch/livepatch-callbacks-mod.c
+++ b/samples/livepatch/livepatch-callbacks-mod.c
@@ -38,4 +38,5 @@ static void livepatch_callbacks_mod_exit(void)
 
 module_init(livepatch_callbacks_mod_init);
 module_exit(livepatch_callbacks_mod_exit);
+MODULE_DESCRIPTION("Live patching demo for (un)patching callbacks, support module");
 MODULE_LICENSE("GPL");
diff --git a/samples/livepatch/livepatch-sample.c b/samples/livepatch/livepatch-sample.c
index cd76d7ebe598..5263a2f31c48 100644
--- a/samples/livepatch/livepatch-sample.c
+++ b/samples/livepatch/livepatch-sample.c
@@ -66,5 +66,6 @@ static void livepatch_exit(void)
 
 module_init(livepatch_init);
 module_exit(livepatch_exit);
+MODULE_DESCRIPTION("Kernel Live Patching Sample Module");
 MODULE_LICENSE("GPL");
 MODULE_INFO(livepatch, "Y");
diff --git a/samples/livepatch/livepatch-shadow-fix1.c b/samples/livepatch/livepatch-shadow-fix1.c
index f3f153895d6c..cbf68ca40097 100644
--- a/samples/livepatch/livepatch-shadow-fix1.c
+++ b/samples/livepatch/livepatch-shadow-fix1.c
@@ -168,5 +168,6 @@ static void livepatch_shadow_fix1_exit(void)
 
 module_init(livepatch_shadow_fix1_init);
 module_exit(livepatch_shadow_fix1_exit);
+MODULE_DESCRIPTION("Live patching demo for shadow variables");
 MODULE_LICENSE("GPL");
 MODULE_INFO(livepatch, "Y");
diff --git a/samples/livepatch/livepatch-shadow-fix2.c b/samples/livepatch/livepatch-shadow-fix2.c
index 361046a4f10c..b99122cb221f 100644
--- a/samples/livepatch/livepatch-shadow-fix2.c
+++ b/samples/livepatch/livepatch-shadow-fix2.c
@@ -128,5 +128,6 @@ static void livepatch_shadow_fix2_exit(void)
 
 module_init(livepatch_shadow_fix2_init);
 module_exit(livepatch_shadow_fix2_exit);
+MODULE_DESCRIPTION("Live patching demo for shadow variables");
 MODULE_LICENSE("GPL");
 MODULE_INFO(livepatch, "Y");
diff --git a/scripts/Makefile.compiler b/scripts/Makefile.compiler
index 8956587b8547..f4fcc1eaaeae 100644
--- a/scripts/Makefile.compiler
+++ b/scripts/Makefile.compiler
@@ -79,8 +79,8 @@ ld-option = $(call try-run, $(LD) $(KBUILD_LDFLAGS) $(1) -v,$(1),$(2),$(3))
 # Usage: MY_RUSTFLAGS += $(call __rustc-option,$(RUSTC),$(MY_RUSTFLAGS),-Cinstrument-coverage,-Zinstrument-coverage)
 # TODO: remove RUSTC_BOOTSTRAP=1 when we raise the minimum GNU Make version to 4.4
 __rustc-option = $(call try-run,\
-	echo '#![allow(missing_docs)]#![feature(no_core)]#![no_core]' | RUSTC_BOOTSTRAP=1\
-	$(1) --sysroot=/dev/null $(filter-out --sysroot=/dev/null,$(2)) $(3)\
+	echo '$(pound)![allow(missing_docs)]$(pound)![feature(no_core)]$(pound)![no_core]' | RUSTC_BOOTSTRAP=1\
+	$(1) --sysroot=/dev/null $(filter-out --sysroot=/dev/null --target=%,$(2)) $(3)\
 	--crate-type=rlib --out-dir=$(TMPOUT) --emit=obj=- - >/dev/null,$(3),$(4))
 
 # rustc-option
diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn
index d75897559d18..d88acdf40855 100644
--- a/scripts/Makefile.extrawarn
+++ b/scripts/Makefile.extrawarn
@@ -8,6 +8,7 @@
 
 # Default set of warnings, always enabled
 KBUILD_CFLAGS += -Wall
+KBUILD_CFLAGS += -Wextra
 KBUILD_CFLAGS += -Wundef
 KBUILD_CFLAGS += -Werror=implicit-function-declaration
 KBUILD_CFLAGS += -Werror=implicit-int
@@ -15,7 +16,7 @@ KBUILD_CFLAGS += -Werror=return-type
 KBUILD_CFLAGS += -Werror=strict-prototypes
 KBUILD_CFLAGS += -Wno-format-security
 KBUILD_CFLAGS += -Wno-trigraphs
-KBUILD_CFLAGS += $(call cc-disable-warning,frame-address,)
+KBUILD_CFLAGS += $(call cc-disable-warning, frame-address)
 KBUILD_CFLAGS += $(call cc-disable-warning, address-of-packed-member)
 KBUILD_CFLAGS += -Wmissing-declarations
 KBUILD_CFLAGS += -Wmissing-prototypes
@@ -56,6 +57,13 @@ KBUILD_CFLAGS += -Wno-pointer-sign
 # globally built with -Wcast-function-type.
 KBUILD_CFLAGS += $(call cc-option, -Wcast-function-type)
 
+# Currently, disable -Wstringop-overflow for GCC 11, globally.
+KBUILD_CFLAGS-$(CONFIG_CC_NO_STRINGOP_OVERFLOW) += $(call cc-disable-warning, stringop-overflow)
+KBUILD_CFLAGS-$(CONFIG_CC_STRINGOP_OVERFLOW) += $(call cc-option, -Wstringop-overflow)
+
+# Currently, disable -Wunterminated-string-initialization as broken
+KBUILD_CFLAGS += $(call cc-disable-warning, unterminated-string-initialization)
+
 # The allocators already balk at large sizes, so silence the compiler
 # warnings for bounds checks involving those possible values. While
 # -Wno-alloc-size-larger-than would normally be used here, earlier versions
@@ -82,7 +90,6 @@ KBUILD_CFLAGS += $(call cc-option,-Werror=designated-init)
 # Warn if there is an enum types mismatch
 KBUILD_CFLAGS += $(call cc-option,-Wenum-conversion)
 
-KBUILD_CFLAGS += -Wextra
 KBUILD_CFLAGS += -Wunused
 
 #
diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 3d22bf863eec..d5bde8322f7b 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -6891,7 +6891,7 @@ sub process {
 					    ($extension eq "f" &&
 					     defined $qualifier && $qualifier !~ /^w/) ||
 					    ($extension eq "4" &&
-					     defined $qualifier && $qualifier !~ /^cc/)) {
+					     defined $qualifier && $qualifier !~ /^c(?:[hlbc]|hR)$/)) {
 						$bad_specifier = $specifier;
 						last;
 					}
diff --git a/scripts/generate_rust_analyzer.py b/scripts/generate_rust_analyzer.py
index cd41bc906fbd..fe663dd0c43b 100755
--- a/scripts/generate_rust_analyzer.py
+++ b/scripts/generate_rust_analyzer.py
@@ -112,6 +112,12 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs):
         cfg=["kernel"],
     )
 
+    append_crate(
+        "ffi",
+        srctree / "rust" / "ffi.rs",
+        ["core", "compiler_builtins"],
+    )
+
     def append_crate_with_generated(
         display_name,
         deps,
@@ -131,9 +137,9 @@ def generate_crates(srctree, objtree, sysroot_src, external_src, cfgs):
             "exclude_dirs": [],
         }
 
-    append_crate_with_generated("bindings", ["core"])
-    append_crate_with_generated("uapi", ["core"])
-    append_crate_with_generated("kernel", ["core", "macros", "build_error", "pin_init", "bindings", "uapi"])
+    append_crate_with_generated("bindings", ["core", "ffi"])
+    append_crate_with_generated("uapi", ["core", "ffi"])
+    append_crate_with_generated("kernel", ["core", "macros", "build_error", "pin_init", "ffi", "bindings", "uapi"])
 
     def is_root_crate(build_file, target):
         try:
diff --git a/scripts/genksyms/keywords.c b/scripts/genksyms/keywords.c
index b85e0979a00c..ee1499d27061 100644
--- a/scripts/genksyms/keywords.c
+++ b/scripts/genksyms/keywords.c
@@ -17,6 +17,8 @@ static struct resword {
 	{ "__signed__", SIGNED_KEYW },
 	{ "__typeof", TYPEOF_KEYW },
 	{ "__typeof__", TYPEOF_KEYW },
+	{ "__typeof_unqual", TYPEOF_KEYW },
+	{ "__typeof_unqual__", TYPEOF_KEYW },
 	{ "__volatile", VOLATILE_KEYW },
 	{ "__volatile__", VOLATILE_KEYW },
 	{ "__builtin_va_list", VA_LIST_KEYW },
@@ -40,6 +42,10 @@ static struct resword {
 	// KAO. },
 	// { "attribute", ATTRIBUTE_KEYW },
 
+	// X86 named address space qualifiers
+	{ "__seg_gs", X86_SEG_KEYW },
+	{ "__seg_fs", X86_SEG_KEYW },
+
 	{ "auto", AUTO_KEYW },
 	{ "char", CHAR_KEYW },
 	{ "const", CONST_KEYW },
@@ -57,6 +63,7 @@ static struct resword {
 	{ "struct", STRUCT_KEYW },
 	{ "typedef", TYPEDEF_KEYW },
 	{ "typeof", TYPEOF_KEYW },
+	{ "typeof_unqual", TYPEOF_KEYW },
 	{ "union", UNION_KEYW },
 	{ "unsigned", UNSIGNED_KEYW },
 	{ "void", VOID_KEYW },
diff --git a/scripts/genksyms/parse.y b/scripts/genksyms/parse.y
index ee600a804fa1..efdcf07c4eb6 100644
--- a/scripts/genksyms/parse.y
+++ b/scripts/genksyms/parse.y
@@ -91,6 +91,8 @@ static void record_compound(struct string_list **keyw,
 %token TYPEOF_KEYW
 %token VA_LIST_KEYW
 
+%token X86_SEG_KEYW
+
 %token EXPORT_SYMBOL_KEYW
 
 %token ASM_PHRASE
@@ -292,7 +294,8 @@ type_qualifier_seq:
 	;
 
 type_qualifier:
-	CONST_KEYW | VOLATILE_KEYW
+	X86_SEG_KEYW
+	| CONST_KEYW | VOLATILE_KEYW
 	| RESTRICT_KEYW
 		{ /* restrict has no effect in prototypes so ignore it */
 		  remove_node($1);
diff --git a/security/Kconfig.hardening b/security/Kconfig.hardening
index c17366ce8224..3fe9d7b945c4 100644
--- a/security/Kconfig.hardening
+++ b/security/Kconfig.hardening
@@ -344,7 +344,7 @@ config CC_HAS_RANDSTRUCT
 
 choice
 	prompt "Randomize layout of sensitive kernel structures"
-	default RANDSTRUCT_FULL if COMPILE_TEST && (GCC_PLUGINS || CC_HAS_RANDSTRUCT)
+	default RANDSTRUCT_FULL if COMPILE_TEST && CC_HAS_RANDSTRUCT
 	default RANDSTRUCT_NONE
 	help
 	  If you enable this, the layouts of structures that are entirely
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index f3e7ac513db3..f99ab1a3b0f0 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -245,7 +245,9 @@ static int process_measurement(struct file *file, const struct cred *cred,
 				&allowed_algos);
 	violation_check = ((func == FILE_CHECK || func == MMAP_CHECK ||
 			    func == MMAP_CHECK_REQPROT) &&
-			   (ima_policy_flag & IMA_MEASURE));
+			   (ima_policy_flag & IMA_MEASURE) &&
+			   ((action & IMA_MEASURE) ||
+			    (file->f_mode & FMODE_WRITE)));
 	if (!action && !violation_check)
 		return 0;
 
diff --git a/security/landlock/domain.c b/security/landlock/domain.c
index bae2e9909013..a647b68e8d06 100644
--- a/security/landlock/domain.c
+++ b/security/landlock/domain.c
@@ -16,6 +16,7 @@
 #include <linux/path.h>
 #include <linux/pid.h>
 #include <linux/sched.h>
+#include <linux/signal.h>
 #include <linux/uidgid.h>
 
 #include "access.h"
@@ -99,8 +100,7 @@ static struct landlock_details *get_current_details(void)
 		return ERR_PTR(-ENOMEM);
 
 	memcpy(details->exe_path, path_str, path_size);
-	WARN_ON_ONCE(current_cred() != current_real_cred());
-	details->pid = get_pid(task_pid(current));
+	details->pid = get_pid(task_tgid(current));
 	details->uid = from_kuid(&init_user_ns, current_uid());
 	get_task_comm(details->comm, current);
 	return details;
diff --git a/security/landlock/domain.h b/security/landlock/domain.h
index ed0d348e214c..7fb70b25f85a 100644
--- a/security/landlock/domain.h
+++ b/security/landlock/domain.h
@@ -130,7 +130,7 @@ int landlock_init_hierarchy_log(struct landlock_hierarchy *const hierarchy);
 static inline void
 landlock_free_hierarchy_details(struct landlock_hierarchy *const hierarchy)
 {
-	if (WARN_ON_ONCE(!hierarchy || !hierarchy->details))
+	if (!hierarchy || !hierarchy->details)
 		return;
 
 	put_pid(hierarchy->details->pid);
diff --git a/security/landlock/syscalls.c b/security/landlock/syscalls.c
index 54a9f29e6ebb..b9561e3417ae 100644
--- a/security/landlock/syscalls.c
+++ b/security/landlock/syscalls.c
@@ -169,20 +169,16 @@ const int landlock_abi_version = 7;
  *        the new ruleset.
  * @size: Size of the pointed &struct landlock_ruleset_attr (needed for
  *        backward and forward compatibility).
- * @flags: Supported value:
+ * @flags: Supported values:
+ *
  *         - %LANDLOCK_CREATE_RULESET_VERSION
  *         - %LANDLOCK_CREATE_RULESET_ERRATA
  *
  * This system call enables to create a new Landlock ruleset, and returns the
  * related file descriptor on success.
  *
- * If @flags is %LANDLOCK_CREATE_RULESET_VERSION and @attr is NULL and @size is
- * 0, then the returned value is the highest supported Landlock ABI version
- * (starting at 1).
- *
- * If @flags is %LANDLOCK_CREATE_RULESET_ERRATA and @attr is NULL and @size is
- * 0, then the returned value is a bitmask of fixed issues for the current
- * Landlock ABI version.
+ * If %LANDLOCK_CREATE_RULESET_VERSION or %LANDLOCK_CREATE_RULESET_ERRATA is
+ * set, then @attr must be NULL and @size must be 0.
  *
  * Possible returned errors are:
  *
@@ -191,6 +187,9 @@ const int landlock_abi_version = 7;
  * - %E2BIG: @attr or @size inconsistencies;
  * - %EFAULT: @attr or @size inconsistencies;
  * - %ENOMSG: empty &landlock_ruleset_attr.handled_access_fs.
+ *
+ * .. kernel-doc:: include/uapi/linux/landlock.h
+ *     :identifiers: landlock_create_ruleset_flags
  */
 SYSCALL_DEFINE3(landlock_create_ruleset,
 		const struct landlock_ruleset_attr __user *const, attr,
@@ -452,18 +451,15 @@ SYSCALL_DEFINE4(landlock_add_rule, const int, ruleset_fd,
  * @ruleset_fd: File descriptor tied to the ruleset to merge with the target.
  * @flags: Supported values:
  *
- * - %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF
- * - %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON
- * - %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF
+ *         - %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF
+ *         - %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON
+ *         - %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF
  *
  * This system call enables to enforce a Landlock ruleset on the current
  * thread.  Enforcing a ruleset requires that the task has %CAP_SYS_ADMIN in its
  * namespace or is running with no_new_privs.  This avoids scenarios where
  * unprivileged tasks can affect the behavior of privileged children.
  *
- * It is allowed to only pass the %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF
- * flag with a @ruleset_fd value of -1.
- *
  * Possible returned errors are:
  *
  * - %EOPNOTSUPP: Landlock is supported by the kernel but disabled at boot time;
@@ -475,6 +471,9 @@ SYSCALL_DEFINE4(landlock_add_rule, const int, ruleset_fd,
  *   %CAP_SYS_ADMIN in its namespace.
  * - %E2BIG: The maximum number of stacked rulesets is reached for the current
  *   thread.
+ *
+ * .. kernel-doc:: include/uapi/linux/landlock.h
+ *     :identifiers: landlock_restrict_self_flags
  */
 SYSCALL_DEFINE2(landlock_restrict_self, const int, ruleset_fd, const __u32,
 		flags)
diff --git a/sound/isa/azt2320.c b/sound/isa/azt2320.c
index b937c9138d12..588b9f0831d3 100644
--- a/sound/isa/azt2320.c
+++ b/sound/isa/azt2320.c
@@ -189,8 +189,8 @@ static int snd_card_azt2320_probe(int dev,
 	if (error < 0)
 		return error;
 
-	strcpy(card->driver, "AZT2320");
-	strcpy(card->shortname, "Aztech AZT2320");
+	strscpy(card->driver, "AZT2320");
+	strscpy(card->shortname, "Aztech AZT2320");
 	sprintf(card->longname, "%s, WSS at 0x%lx, irq %i, dma %i&%i",
 		card->shortname, chip->port, irq[dev], dma1[dev], dma2[dev]);
 
diff --git a/sound/pci/hda/Kconfig b/sound/pci/hda/Kconfig
index fb955a205d50..9c427270ff4f 100644
--- a/sound/pci/hda/Kconfig
+++ b/sound/pci/hda/Kconfig
@@ -96,9 +96,7 @@ config SND_HDA_CIRRUS_SCODEC
 
 config SND_HDA_CIRRUS_SCODEC_KUNIT_TEST
 	tristate "KUnit test for Cirrus side-codec library" if !KUNIT_ALL_TESTS
-	select SND_HDA_CIRRUS_SCODEC
-	select GPIOLIB
-	depends on KUNIT
+	depends on SND_HDA_CIRRUS_SCODEC && GPIOLIB && KUNIT
 	default KUNIT_ALL_TESTS
 	help
 	  This builds KUnit tests for the cirrus side-codec library.
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 79004bc8107b..8a2b09e4a7d5 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -441,6 +441,10 @@ static void alc_fill_eapd_coef(struct hda_codec *codec)
 		alc_update_coef_idx(codec, 0x67, 0xf000, 0x3000);
 		fallthrough;
 	case 0x10ec0215:
+	case 0x10ec0236:
+	case 0x10ec0245:
+	case 0x10ec0256:
+	case 0x10ec0257:
 	case 0x10ec0285:
 	case 0x10ec0289:
 		alc_update_coef_idx(codec, 0x36, 1<<13, 0);
@@ -448,12 +452,8 @@ static void alc_fill_eapd_coef(struct hda_codec *codec)
 	case 0x10ec0230:
 	case 0x10ec0233:
 	case 0x10ec0235:
-	case 0x10ec0236:
-	case 0x10ec0245:
 	case 0x10ec0255:
-	case 0x10ec0256:
 	case 0x19e58326:
-	case 0x10ec0257:
 	case 0x10ec0282:
 	case 0x10ec0283:
 	case 0x10ec0286:
@@ -6742,6 +6742,25 @@ static void alc274_fixup_bind_dacs(struct hda_codec *codec,
 	codec->power_save_node = 0;
 }
 
+/* avoid DAC 0x06 for speaker switch 0x17; it has no volume control */
+static void alc274_fixup_hp_aio_bind_dacs(struct hda_codec *codec,
+				    const struct hda_fixup *fix, int action)
+{
+	static const hda_nid_t conn[] = { 0x02, 0x03 }; /* exclude 0x06 */
+	/* The speaker is routed to the Node 0x06 by a mistake, thus the
+	 * speaker's volume can't be adjusted since the node doesn't have
+	 * Amp-out capability. Assure the speaker and lineout pin to be
+	 * coupled with DAC NID 0x02.
+	 */
+	static const hda_nid_t preferred_pairs[] = {
+		0x16, 0x02, 0x17, 0x02, 0x21, 0x03, 0
+	};
+	struct alc_spec *spec = codec->spec;
+
+	snd_hda_override_conn_list(codec, 0x17, ARRAY_SIZE(conn), conn);
+	spec->gen.preferred_dacs = preferred_pairs;
+}
+
 /* avoid DAC 0x06 for bass speaker 0x17; it has no volume control */
 static void alc289_fixup_asus_ga401(struct hda_codec *codec,
 				    const struct hda_fixup *fix, int action)
@@ -6963,6 +6982,41 @@ static void alc285_fixup_hp_spectre_x360_eb1(struct hda_codec *codec,
 	}
 }
 
+/* GPIO1 = amplifier on/off */
+static void alc285_fixup_hp_spectre_x360_df1(struct hda_codec *codec,
+					     const struct hda_fixup *fix,
+					     int action)
+{
+	struct alc_spec *spec = codec->spec;
+	static const hda_nid_t conn[] = { 0x02 };
+	static const struct hda_pintbl pincfgs[] = {
+		{ 0x14, 0x90170110 },  /* front/high speakers */
+		{ 0x17, 0x90170130 },  /* back/bass speakers */
+		{ }
+	};
+
+	// enable mute led
+	alc285_fixup_hp_mute_led_coefbit(codec, fix, action);
+
+	switch (action) {
+	case HDA_FIXUP_ACT_PRE_PROBE:
+		/* needed for amp of back speakers */
+		spec->gpio_mask |= 0x01;
+		spec->gpio_dir |= 0x01;
+		snd_hda_apply_pincfgs(codec, pincfgs);
+		/* share DAC to have unified volume control */
+		snd_hda_override_conn_list(codec, 0x14, ARRAY_SIZE(conn), conn);
+		snd_hda_override_conn_list(codec, 0x17, ARRAY_SIZE(conn), conn);
+		break;
+	case HDA_FIXUP_ACT_INIT:
+		/* need to toggle GPIO to enable the amp of back speakers */
+		alc_update_gpio_data(codec, 0x01, true);
+		msleep(100);
+		alc_update_gpio_data(codec, 0x01, false);
+		break;
+	}
+}
+
 static void alc285_fixup_hp_spectre_x360(struct hda_codec *codec,
 					  const struct hda_fixup *fix, int action)
 {
@@ -7761,6 +7815,7 @@ enum {
 	ALC280_FIXUP_HP_9480M,
 	ALC245_FIXUP_HP_X360_AMP,
 	ALC285_FIXUP_HP_SPECTRE_X360_EB1,
+	ALC285_FIXUP_HP_SPECTRE_X360_DF1,
 	ALC285_FIXUP_HP_ENVY_X360,
 	ALC288_FIXUP_DELL_HEADSET_MODE,
 	ALC288_FIXUP_DELL1_MIC_NO_PRESENCE,
@@ -7969,6 +8024,8 @@ enum {
 	ALC233_FIXUP_MEDION_MTL_SPK,
 	ALC294_FIXUP_BASS_SPEAKER_15,
 	ALC283_FIXUP_DELL_HP_RESUME,
+	ALC294_FIXUP_ASUS_CS35L41_SPI_2,
+	ALC274_FIXUP_HP_AIO_BIND_DACS,
 };
 
 /* A special fixup for Lenovo C940 and Yoga Duet 7;
@@ -9836,6 +9893,10 @@ static const struct hda_fixup alc269_fixups[] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc285_fixup_hp_spectre_x360_eb1
 	},
+	[ALC285_FIXUP_HP_SPECTRE_X360_DF1] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc285_fixup_hp_spectre_x360_df1
+	},
 	[ALC285_FIXUP_HP_ENVY_X360] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc285_fixup_hp_envy_x360,
@@ -10333,6 +10394,16 @@ static const struct hda_fixup alc269_fixups[] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc283_fixup_dell_hp_resume,
 	},
+	[ALC294_FIXUP_ASUS_CS35L41_SPI_2] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = cs35l41_fixup_spi_two,
+		.chained = true,
+		.chain_id = ALC294_FIXUP_ASUS_HEADSET_MIC,
+	},
+	[ALC274_FIXUP_HP_AIO_BIND_DACS] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc274_fixup_hp_aio_bind_dacs,
+	},
 };
 
 static const struct hda_quirk alc269_fixup_tbl[] = {
@@ -10557,6 +10628,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x103c, 0x86c1, "HP Laptop 15-da3001TU", ALC236_FIXUP_HP_MUTE_LED_COEFBIT2),
 	SND_PCI_QUIRK(0x103c, 0x86c7, "HP Envy AiO 32", ALC274_FIXUP_HP_ENVY_GPIO),
 	SND_PCI_QUIRK(0x103c, 0x86e7, "HP Spectre x360 15-eb0xxx", ALC285_FIXUP_HP_SPECTRE_X360_EB1),
+	SND_PCI_QUIRK(0x103c, 0x863e, "HP Spectre x360 15-df1xxx", ALC285_FIXUP_HP_SPECTRE_X360_DF1),
 	SND_PCI_QUIRK(0x103c, 0x86e8, "HP Spectre x360 15-eb0xxx", ALC285_FIXUP_HP_SPECTRE_X360_EB1),
 	SND_PCI_QUIRK(0x103c, 0x86f9, "HP Spectre x360 13-aw0xxx", ALC285_FIXUP_HP_SPECTRE_X360_MUTE_LED),
 	SND_PCI_QUIRK(0x103c, 0x8716, "HP Elite Dragonfly G2 Notebook PC", ALC285_FIXUP_HP_GPIO_AMP_INIT),
@@ -10761,12 +10833,13 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x103c, 0x8ca7, "HP ZBook Fury", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8caf, "HP Elite mt645 G8 Mobile Thin Client", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
 	SND_PCI_QUIRK(0x103c, 0x8cbd, "HP Pavilion Aero Laptop 13-bg0xxx", ALC245_FIXUP_HP_X360_MUTE_LEDS),
-	SND_PCI_QUIRK(0x103c, 0x8cdd, "HP Spectre", ALC287_FIXUP_CS35L41_I2C_2),
-	SND_PCI_QUIRK(0x103c, 0x8cde, "HP Spectre", ALC287_FIXUP_CS35L41_I2C_2),
+	SND_PCI_QUIRK(0x103c, 0x8cdd, "HP Spectre", ALC245_FIXUP_HP_SPECTRE_X360_EU0XXX),
+	SND_PCI_QUIRK(0x103c, 0x8cde, "HP OmniBook Ultra Flip Laptop 14t", ALC245_FIXUP_HP_SPECTRE_X360_EU0XXX),
 	SND_PCI_QUIRK(0x103c, 0x8cdf, "HP SnowWhite", ALC287_FIXUP_CS35L41_I2C_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8ce0, "HP SnowWhite", ALC287_FIXUP_CS35L41_I2C_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8cf5, "HP ZBook Studio 16", ALC245_FIXUP_CS35L41_SPI_4_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8d01, "HP ZBook Power 14 G12", ALC285_FIXUP_HP_GPIO_LED),
+	SND_PCI_QUIRK(0x103c, 0x8d18, "HP EliteStudio 8 AIO", ALC274_FIXUP_HP_AIO_BIND_DACS),
 	SND_PCI_QUIRK(0x103c, 0x8d84, "HP EliteBook X G1i", ALC285_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8d85, "HP EliteBook 14 G12", ALC285_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8d86, "HP Elite X360 14 G12", ALC285_FIXUP_HP_GPIO_LED),
@@ -10786,11 +10859,15 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x103c, 0x8da1, "HP 16 Clipper OmniBook X", ALC287_FIXUP_CS35L41_I2C_2),
 	SND_PCI_QUIRK(0x103c, 0x8da7, "HP 14 Enstrom OmniBook X", ALC287_FIXUP_CS35L41_I2C_2),
 	SND_PCI_QUIRK(0x103c, 0x8da8, "HP 16 Piston OmniBook X", ALC287_FIXUP_CS35L41_I2C_2),
+	SND_PCI_QUIRK(0x103c, 0x8dd4, "HP EliteStudio 8 AIO", ALC274_FIXUP_HP_AIO_BIND_DACS),
 	SND_PCI_QUIRK(0x103c, 0x8de8, "HP Gemtree", ALC245_FIXUP_TAS2781_SPI_2),
 	SND_PCI_QUIRK(0x103c, 0x8de9, "HP Gemtree", ALC245_FIXUP_TAS2781_SPI_2),
 	SND_PCI_QUIRK(0x103c, 0x8dec, "HP EliteBook 640 G12", ALC236_FIXUP_HP_GPIO_LED),
+	SND_PCI_QUIRK(0x103c, 0x8ded, "HP EliteBook 640 G12", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8dee, "HP EliteBook 660 G12", ALC236_FIXUP_HP_GPIO_LED),
+	SND_PCI_QUIRK(0x103c, 0x8def, "HP EliteBook 660 G12", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8df0, "HP EliteBook 630 G12", ALC236_FIXUP_HP_GPIO_LED),
+	SND_PCI_QUIRK(0x103c, 0x8df1, "HP EliteBook 630 G12", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8dfc, "HP EliteBook 645 G12", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8dfe, "HP EliteBook 665 G12", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8e11, "HP Trekker", ALC287_FIXUP_CS35L41_I2C_2),
@@ -10835,11 +10912,11 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1043, 0x12a0, "ASUS X441UV", ALC233_FIXUP_EAPD_COEF_AND_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1043, 0x12a3, "Asus N7691ZM", ALC269_FIXUP_ASUS_N7601ZM),
 	SND_PCI_QUIRK(0x1043, 0x12af, "ASUS UX582ZS", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x12b4, "ASUS B3405CCA / P3405CCA", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x12e0, "ASUS X541SA", ALC256_FIXUP_ASUS_MIC),
-	SND_PCI_QUIRK(0x1043, 0x12f0, "ASUS X541UV", ALC256_FIXUP_ASUS_MIC),
+	SND_PCI_QUIRK(0x1043, 0x12b4, "ASUS B3405CCA / P3405CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x12e0, "ASUS X541SA", ALC256_FIXUP_ASUS_MIC_NO_PRESENCE),
+	SND_PCI_QUIRK(0x1043, 0x12f0, "ASUS X541UV", ALC256_FIXUP_ASUS_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1043, 0x1313, "Asus K42JZ", ALC269VB_FIXUP_ASUS_MIC_NO_PRESENCE),
-	SND_PCI_QUIRK(0x1043, 0x13b0, "ASUS Z550SA", ALC256_FIXUP_ASUS_MIC),
+	SND_PCI_QUIRK(0x1043, 0x13b0, "ASUS Z550SA", ALC256_FIXUP_ASUS_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1043, 0x1427, "Asus Zenbook UX31E", ALC269VB_FIXUP_ASUS_ZENBOOK),
 	SND_PCI_QUIRK(0x1043, 0x1433, "ASUS GX650PY/PZ/PV/PU/PYV/PZV/PIV/PVV", ALC285_FIXUP_ASUS_I2C_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1043, 0x1460, "Asus VivoBook 15", ALC256_FIXUP_ASUS_MIC_NO_PRESENCE),
@@ -10893,7 +10970,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1043, 0x1c92, "ASUS ROG Strix G15", ALC285_FIXUP_ASUS_G533Z_PINS),
 	SND_PCI_QUIRK(0x1043, 0x1c9f, "ASUS G614JU/JV/JI", ALC285_FIXUP_ASUS_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1043, 0x1caf, "ASUS G634JY/JZ/JI/JG", ALC285_FIXUP_ASUS_SPI_REAR_SPEAKERS),
-	SND_PCI_QUIRK(0x1043, 0x1ccd, "ASUS X555UB", ALC256_FIXUP_ASUS_MIC),
+	SND_PCI_QUIRK(0x1043, 0x1ccd, "ASUS X555UB", ALC256_FIXUP_ASUS_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1043, 0x1ccf, "ASUS G814JU/JV/JI", ALC245_FIXUP_CS35L41_SPI_2),
 	SND_PCI_QUIRK(0x1043, 0x1cdf, "ASUS G814JY/JZ/JG", ALC245_FIXUP_CS35L41_SPI_2),
 	SND_PCI_QUIRK(0x1043, 0x1cef, "ASUS G834JY/JZ/JI/JG", ALC285_FIXUP_ASUS_HEADSET_MIC),
@@ -10925,14 +11002,14 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1043, 0x1fb3, "ASUS ROG Flow Z13 GZ302EA", ALC287_FIXUP_CS35L41_I2C_2),
 	SND_PCI_QUIRK(0x1043, 0x3011, "ASUS B5605CVA", ALC245_FIXUP_CS35L41_SPI_2),
 	SND_PCI_QUIRK(0x1043, 0x3030, "ASUS ZN270IE", ALC256_FIXUP_ASUS_AIO_GPIO2),
-	SND_PCI_QUIRK(0x1043, 0x3061, "ASUS B3405CCA", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x3071, "ASUS B5405CCA", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x30c1, "ASUS B3605CCA / P3605CCA", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x30d1, "ASUS B5405CCA", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x30e1, "ASUS B5605CCA", ALC245_FIXUP_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x3061, "ASUS B3405CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x3071, "ASUS B5405CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x30c1, "ASUS B3605CCA / P3605CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x30d1, "ASUS B5405CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x30e1, "ASUS B5605CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
 	SND_PCI_QUIRK(0x1043, 0x31d0, "ASUS Zen AIO 27 Z272SD_A272SD", ALC274_FIXUP_ASUS_ZEN_AIO_27),
-	SND_PCI_QUIRK(0x1043, 0x31e1, "ASUS B5605CCA", ALC245_FIXUP_CS35L41_SPI_2),
-	SND_PCI_QUIRK(0x1043, 0x31f1, "ASUS B3605CCA", ALC245_FIXUP_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x31e1, "ASUS B5605CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
+	SND_PCI_QUIRK(0x1043, 0x31f1, "ASUS B3605CCA", ALC294_FIXUP_ASUS_CS35L41_SPI_2),
 	SND_PCI_QUIRK(0x1043, 0x3a20, "ASUS G614JZR", ALC285_FIXUP_ASUS_SPI_REAR_SPEAKERS),
 	SND_PCI_QUIRK(0x1043, 0x3a30, "ASUS G814JVR/JIR", ALC285_FIXUP_ASUS_SPI_REAR_SPEAKERS),
 	SND_PCI_QUIRK(0x1043, 0x3a40, "ASUS G814JZR", ALC285_FIXUP_ASUS_SPI_REAR_SPEAKERS),
@@ -11487,6 +11564,7 @@ static const struct hda_model_fixup alc269_fixup_models[] = {
 	{.id = ALC295_FIXUP_HP_OMEN, .name = "alc295-hp-omen"},
 	{.id = ALC285_FIXUP_HP_SPECTRE_X360, .name = "alc285-hp-spectre-x360"},
 	{.id = ALC285_FIXUP_HP_SPECTRE_X360_EB1, .name = "alc285-hp-spectre-x360-eb1"},
+	{.id = ALC285_FIXUP_HP_SPECTRE_X360_DF1, .name = "alc285-hp-spectre-x360-df1"},
 	{.id = ALC285_FIXUP_HP_ENVY_X360, .name = "alc285-hp-envy-x360"},
 	{.id = ALC287_FIXUP_IDEAPAD_BASS_SPK_AMP, .name = "alc287-ideapad-bass-spk-amp"},
 	{.id = ALC287_FIXUP_YOGA9_14IAP7_BASS_SPK_PIN, .name = "alc287-yoga9-bass-spk-pin"},
diff --git a/sound/pci/hda/tas2781_hda_spi.c b/sound/pci/hda/tas2781_hda_spi.c
index 399f2e4c3b62..25175ff4b3aa 100644
--- a/sound/pci/hda/tas2781_hda_spi.c
+++ b/sound/pci/hda/tas2781_hda_spi.c
@@ -1003,8 +1003,7 @@ static void tasdev_fw_ready(const struct firmware *fmw, void *context)
 	 */
 
 out:
-	if (fmw)
-		release_firmware(fmw);
+	release_firmware(fmw);
 	pm_runtime_mark_last_busy(tas_hda->priv->dev);
 	pm_runtime_put_autosuspend(tas_hda->priv->dev);
 }
diff --git a/sound/soc/amd/acp/acp-i2s.c b/sound/soc/amd/acp/acp-i2s.c
index a38409dd1d34..70fa54d568ef 100644
--- a/sound/soc/amd/acp/acp-i2s.c
+++ b/sound/soc/amd/acp/acp-i2s.c
@@ -97,7 +97,7 @@ static int acp_i2s_set_tdm_slot(struct snd_soc_dai *dai, u32 tx_mask, u32 rx_mas
 	struct acp_stream *stream;
 	int slot_len, no_of_slots;
 
-	chip = dev_get_platdata(dev);
+	chip = dev_get_drvdata(dev->parent);
 	switch (slot_width) {
 	case SLOT_WIDTH_8:
 		slot_len = 8;
diff --git a/sound/soc/amd/acp/acp-legacy-common.c b/sound/soc/amd/acp/acp-legacy-common.c
index b4d68484e06d..ba8db0851daa 100644
--- a/sound/soc/amd/acp/acp-legacy-common.c
+++ b/sound/soc/amd/acp/acp-legacy-common.c
@@ -450,7 +450,7 @@ int acp_machine_select(struct acp_chip_info *chip)
 	struct snd_soc_acpi_mach *mach;
 	int size, platform;
 
-	if (chip->flag == FLAG_AMD_LEGACY_ONLY_DMIC) {
+	if (chip->flag == FLAG_AMD_LEGACY_ONLY_DMIC && chip->is_pdm_dev) {
 		platform = chip->acp_rev;
 		chip->mach_dev = platform_device_register_data(chip->dev, "acp-pdm-mach",
 							       PLATFORM_DEVID_NONE, &platform,
diff --git a/sound/soc/amd/acp/acp-rembrandt.c b/sound/soc/amd/acp/acp-rembrandt.c
index 746b6ed72029..cccdd10c345e 100644
--- a/sound/soc/amd/acp/acp-rembrandt.c
+++ b/sound/soc/amd/acp/acp-rembrandt.c
@@ -199,7 +199,7 @@ static void rembrandt_audio_remove(struct platform_device *pdev)
 
 static int rmb_pcm_resume(struct device *dev)
 {
-	struct acp_chip_info *chip = dev_get_platdata(dev);
+	struct acp_chip_info *chip = dev_get_drvdata(dev->parent);
 	struct acp_stream *stream;
 	struct snd_pcm_substream *substream;
 	snd_pcm_uframes_t buf_in_frames;
diff --git a/sound/soc/amd/acp/acp-renoir.c b/sound/soc/amd/acp/acp-renoir.c
index ebf0106fc737..04f6d70b6a92 100644
--- a/sound/soc/amd/acp/acp-renoir.c
+++ b/sound/soc/amd/acp/acp-renoir.c
@@ -146,7 +146,7 @@ static void renoir_audio_remove(struct platform_device *pdev)
 
 static int rn_pcm_resume(struct device *dev)
 {
-	struct acp_chip_info *chip = dev_get_platdata(dev);
+	struct acp_chip_info *chip = dev_get_drvdata(dev->parent);
 	struct acp_stream *stream;
 	struct snd_pcm_substream *substream;
 	snd_pcm_uframes_t buf_in_frames;
diff --git a/sound/soc/amd/acp/acp63.c b/sound/soc/amd/acp/acp63.c
index 52d895e624c7..1f15c96a9b94 100644
--- a/sound/soc/amd/acp/acp63.c
+++ b/sound/soc/amd/acp/acp63.c
@@ -250,7 +250,7 @@ static void acp63_audio_remove(struct platform_device *pdev)
 
 static int acp63_pcm_resume(struct device *dev)
 {
-	struct acp_chip_info *chip = dev_get_platdata(dev);
+	struct acp_chip_info *chip = dev_get_drvdata(dev->parent);
 	struct acp_stream *stream;
 	struct snd_pcm_substream *substream;
 	snd_pcm_uframes_t buf_in_frames;
diff --git a/sound/soc/amd/acp/acp70.c b/sound/soc/amd/acp/acp70.c
index 6d5f5ade075c..217b717e9beb 100644
--- a/sound/soc/amd/acp/acp70.c
+++ b/sound/soc/amd/acp/acp70.c
@@ -182,7 +182,7 @@ static void acp_acp70_audio_remove(struct platform_device *pdev)
 
 static int acp70_pcm_resume(struct device *dev)
 {
-	struct acp_chip_info *chip = dev_get_platdata(dev);
+	struct acp_chip_info *chip = dev_get_drvdata(dev->parent);
 	struct acp_stream *stream;
 	struct snd_pcm_substream *substream;
 	snd_pcm_uframes_t buf_in_frames;
diff --git a/sound/soc/amd/ps/pci-ps.c b/sound/soc/amd/ps/pci-ps.c
index 8e57f31ef7f7..7936b3173632 100644
--- a/sound/soc/amd/ps/pci-ps.c
+++ b/sound/soc/amd/ps/pci-ps.c
@@ -193,6 +193,7 @@ static irqreturn_t acp63_irq_handler(int irq, void *dev_id)
 	struct amd_sdw_manager *amd_manager;
 	u32 ext_intr_stat, ext_intr_stat1;
 	u16 irq_flag = 0;
+	u16 wake_irq_flag = 0;
 	u16 sdw_dma_irq_flag = 0;
 
 	adata = dev_id;
@@ -231,7 +232,7 @@ static irqreturn_t acp63_irq_handler(int irq, void *dev_id)
 	}
 
 	if (adata->acp_rev >= ACP70_PCI_REV)
-		irq_flag = check_and_handle_acp70_sdw_wake_irq(adata);
+		wake_irq_flag = check_and_handle_acp70_sdw_wake_irq(adata);
 
 	if (ext_intr_stat & BIT(PDM_DMA_STAT)) {
 		ps_pdm_data = dev_get_drvdata(&adata->pdm_dev->dev);
@@ -245,7 +246,7 @@ static irqreturn_t acp63_irq_handler(int irq, void *dev_id)
 	if (sdw_dma_irq_flag)
 		return IRQ_WAKE_THREAD;
 
-	if (irq_flag)
+	if (irq_flag | wake_irq_flag)
 		return IRQ_HANDLED;
 	else
 		return IRQ_NONE;
diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig
index 40bb7a1d44bc..20f99cbee29b 100644
--- a/sound/soc/codecs/Kconfig
+++ b/sound/soc/codecs/Kconfig
@@ -776,10 +776,9 @@ config SND_SOC_CS_AMP_LIB
 	tristate
 
 config SND_SOC_CS_AMP_LIB_TEST
-	tristate "KUnit test for Cirrus Logic cs-amp-lib"
-	depends on KUNIT
+	tristate "KUnit test for Cirrus Logic cs-amp-lib" if !KUNIT_ALL_TESTS
+	depends on SND_SOC_CS_AMP_LIB && KUNIT
 	default KUNIT_ALL_TESTS
-	select SND_SOC_CS_AMP_LIB
 	help
 	  This builds KUnit tests for the Cirrus Logic common
 	  amplifier library.
diff --git a/sound/soc/codecs/aw88081.c b/sound/soc/codecs/aw88081.c
index ad16ab6812cd..3dd8428f08cc 100644
--- a/sound/soc/codecs/aw88081.c
+++ b/sound/soc/codecs/aw88081.c
@@ -1295,9 +1295,19 @@ static int aw88081_i2c_probe(struct i2c_client *i2c)
 			aw88081_dai, ARRAY_SIZE(aw88081_dai));
 }
 
+#if defined(CONFIG_OF)
+static const struct of_device_id aw88081_of_match[] = {
+	{ .compatible = "awinic,aw88081" },
+	{ .compatible = "awinic,aw88083" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, aw88081_of_match);
+#endif
+
 static struct i2c_driver aw88081_i2c_driver = {
 	.driver = {
 		.name = AW88081_I2C_NAME,
+		.of_match_table = of_match_ptr(aw88081_of_match),
 	},
 	.probe = aw88081_i2c_probe,
 	.id_table = aw88081_i2c_id,
diff --git a/sound/soc/codecs/cs42l43-jack.c b/sound/soc/codecs/cs42l43-jack.c
index ac19a572fe70..6165ac16c3a9 100644
--- a/sound/soc/codecs/cs42l43-jack.c
+++ b/sound/soc/codecs/cs42l43-jack.c
@@ -654,6 +654,10 @@ static int cs42l43_run_type_detect(struct cs42l43_codec *priv)
 
 	reinit_completion(&priv->type_detect);
 
+	regmap_update_bits(cs42l43->regmap, CS42L43_STEREO_MIC_CLAMP_CTRL,
+			   CS42L43_SMIC_HPAMP_CLAMP_DIS_FRC_VAL_MASK,
+			   CS42L43_SMIC_HPAMP_CLAMP_DIS_FRC_VAL_MASK);
+
 	cs42l43_start_hs_bias(priv, true);
 	regmap_update_bits(cs42l43->regmap, CS42L43_HS2,
 			   CS42L43_HSDET_MODE_MASK, 0x3 << CS42L43_HSDET_MODE_SHIFT);
@@ -665,6 +669,9 @@ static int cs42l43_run_type_detect(struct cs42l43_codec *priv)
 			   CS42L43_HSDET_MODE_MASK, 0x2 << CS42L43_HSDET_MODE_SHIFT);
 	cs42l43_stop_hs_bias(priv);
 
+	regmap_update_bits(cs42l43->regmap, CS42L43_STEREO_MIC_CLAMP_CTRL,
+			   CS42L43_SMIC_HPAMP_CLAMP_DIS_FRC_VAL_MASK, 0);
+
 	if (!time_left)
 		return -ETIMEDOUT;
 
@@ -702,6 +709,9 @@ static void cs42l43_clear_jack(struct cs42l43_codec *priv)
 			   CS42L43_PGA_WIDESWING_MODE_EN_MASK, 0);
 	regmap_update_bits(cs42l43->regmap, CS42L43_STEREO_MIC_CTRL,
 			   CS42L43_JACK_STEREO_CONFIG_MASK, 0);
+	regmap_update_bits(cs42l43->regmap, CS42L43_STEREO_MIC_CLAMP_CTRL,
+			   CS42L43_SMIC_HPAMP_CLAMP_DIS_FRC_MASK,
+			   CS42L43_SMIC_HPAMP_CLAMP_DIS_FRC_MASK);
 	regmap_update_bits(cs42l43->regmap, CS42L43_HS2,
 			   CS42L43_HSDET_MODE_MASK | CS42L43_HSDET_MANUAL_MODE_MASK,
 			   0x2 << CS42L43_HSDET_MODE_SHIFT);
diff --git a/sound/soc/codecs/hdmi-codec.c b/sound/soc/codecs/hdmi-codec.c
index 17019b1d680b..bc01ff65bd6f 100644
--- a/sound/soc/codecs/hdmi-codec.c
+++ b/sound/soc/codecs/hdmi-codec.c
@@ -842,12 +842,28 @@ static void print_eld_info(struct snd_info_entry *entry,
 static int hdmi_dai_proc_new(struct hdmi_codec_priv *hcp,
 			     struct snd_soc_dai *dai)
 {
+	struct snd_soc_component *component = dai->component;
+	struct snd_soc_card *card = component->card;
+	struct snd_soc_dai *d;
+	struct snd_soc_pcm_runtime *rtd;
 	struct snd_info_entry *entry;
 	char name[32];
-	int err;
+	int err, i, id = 0;
 
-	snprintf(name, sizeof(name), "eld#%d", dai->id);
-	err = snd_card_proc_new(dai->component->card->snd_card, name, &entry);
+	/*
+	 * To avoid duplicate proc entry, find its rtd and use rtd->id
+	 * instead of dai->id
+	 */
+	for_each_card_rtds(card, rtd) {
+		for_each_rtd_dais(rtd, i, d)
+			if (d == dai) {
+				id = rtd->id;
+				goto found;
+			}
+	}
+found:
+	snprintf(name, sizeof(name), "eld#%d", id);
+	err = snd_card_proc_new(card->snd_card, name, &entry);
 	if (err < 0)
 		return err;
 
diff --git a/sound/soc/codecs/lpass-wsa-macro.c b/sound/soc/codecs/lpass-wsa-macro.c
index b261fa373e65..c1fb71cfb5d0 100644
--- a/sound/soc/codecs/lpass-wsa-macro.c
+++ b/sound/soc/codecs/lpass-wsa-macro.c
@@ -63,6 +63,10 @@
 #define CDC_WSA_TX_SPKR_PROT_CLK_DISABLE	0
 #define CDC_WSA_TX_SPKR_PROT_PCM_RATE_MASK	GENMASK(3, 0)
 #define CDC_WSA_TX_SPKR_PROT_PCM_RATE_8K	0
+#define CDC_WSA_TX_SPKR_PROT_PCM_RATE_16K	1
+#define CDC_WSA_TX_SPKR_PROT_PCM_RATE_24K	2
+#define CDC_WSA_TX_SPKR_PROT_PCM_RATE_32K	3
+#define CDC_WSA_TX_SPKR_PROT_PCM_RATE_48K	4
 #define CDC_WSA_TX0_SPKR_PROT_PATH_CFG0		(0x0248)
 #define CDC_WSA_TX1_SPKR_PROT_PATH_CTL		(0x0264)
 #define CDC_WSA_TX1_SPKR_PROT_PATH_CFG0		(0x0268)
@@ -407,6 +411,7 @@ struct wsa_macro {
 	int ear_spkr_gain;
 	int spkr_gain_offset;
 	int spkr_mode;
+	u32 pcm_rate_vi;
 	int is_softclip_on[WSA_MACRO_SOFTCLIP_MAX];
 	int softclip_clk_users[WSA_MACRO_SOFTCLIP_MAX];
 	struct regmap *regmap;
@@ -1280,6 +1285,7 @@ static int wsa_macro_hw_params(struct snd_pcm_substream *substream,
 			       struct snd_soc_dai *dai)
 {
 	struct snd_soc_component *component = dai->component;
+	struct wsa_macro *wsa = snd_soc_component_get_drvdata(component);
 	int ret;
 
 	switch (substream->stream) {
@@ -1292,6 +1298,11 @@ static int wsa_macro_hw_params(struct snd_pcm_substream *substream,
 			return ret;
 		}
 		break;
+	case SNDRV_PCM_STREAM_CAPTURE:
+		if (dai->id == WSA_MACRO_AIF_VI)
+			wsa->pcm_rate_vi = params_rate(params);
+
+		break;
 	default:
 		break;
 	}
@@ -1448,35 +1459,11 @@ static void wsa_macro_mclk_enable(struct wsa_macro *wsa, bool mclk_enable)
 	}
 }
 
-static int wsa_macro_mclk_event(struct snd_soc_dapm_widget *w,
-				struct snd_kcontrol *kcontrol, int event)
+static void wsa_macro_enable_disable_vi_sense(struct snd_soc_component *component, bool enable,
+						u32 tx_reg0, u32 tx_reg1, u32 val)
 {
-	struct snd_soc_component *component = snd_soc_dapm_to_component(w->dapm);
-	struct wsa_macro *wsa = snd_soc_component_get_drvdata(component);
-
-	wsa_macro_mclk_enable(wsa, event == SND_SOC_DAPM_PRE_PMU);
-	return 0;
-}
-
-static int wsa_macro_enable_vi_feedback(struct snd_soc_dapm_widget *w,
-					struct snd_kcontrol *kcontrol,
-					int event)
-{
-	struct snd_soc_component *component = snd_soc_dapm_to_component(w->dapm);
-	struct wsa_macro *wsa = snd_soc_component_get_drvdata(component);
-	u32 tx_reg0, tx_reg1;
-
-	if (test_bit(WSA_MACRO_TX0, &wsa->active_ch_mask[WSA_MACRO_AIF_VI])) {
-		tx_reg0 = CDC_WSA_TX0_SPKR_PROT_PATH_CTL;
-		tx_reg1 = CDC_WSA_TX1_SPKR_PROT_PATH_CTL;
-	} else if (test_bit(WSA_MACRO_TX1, &wsa->active_ch_mask[WSA_MACRO_AIF_VI])) {
-		tx_reg0 = CDC_WSA_TX2_SPKR_PROT_PATH_CTL;
-		tx_reg1 = CDC_WSA_TX3_SPKR_PROT_PATH_CTL;
-	}
-
-	switch (event) {
-	case SND_SOC_DAPM_POST_PMU:
-			/* Enable V&I sensing */
+	if (enable) {
+		/* Enable V&I sensing */
 		snd_soc_component_update_bits(component, tx_reg0,
 					      CDC_WSA_TX_SPKR_PROT_RESET_MASK,
 					      CDC_WSA_TX_SPKR_PROT_RESET);
@@ -1485,10 +1472,10 @@ static int wsa_macro_enable_vi_feedback(struct snd_soc_dapm_widget *w,
 					      CDC_WSA_TX_SPKR_PROT_RESET);
 		snd_soc_component_update_bits(component, tx_reg0,
 					      CDC_WSA_TX_SPKR_PROT_PCM_RATE_MASK,
-					      CDC_WSA_TX_SPKR_PROT_PCM_RATE_8K);
+					      val);
 		snd_soc_component_update_bits(component, tx_reg1,
 					      CDC_WSA_TX_SPKR_PROT_PCM_RATE_MASK,
-					      CDC_WSA_TX_SPKR_PROT_PCM_RATE_8K);
+					      val);
 		snd_soc_component_update_bits(component, tx_reg0,
 					      CDC_WSA_TX_SPKR_PROT_CLK_EN_MASK,
 					      CDC_WSA_TX_SPKR_PROT_CLK_ENABLE);
@@ -1501,9 +1488,7 @@ static int wsa_macro_enable_vi_feedback(struct snd_soc_dapm_widget *w,
 		snd_soc_component_update_bits(component, tx_reg1,
 					      CDC_WSA_TX_SPKR_PROT_RESET_MASK,
 					      CDC_WSA_TX_SPKR_PROT_NO_RESET);
-		break;
-	case SND_SOC_DAPM_POST_PMD:
-		/* Disable V&I sensing */
+	} else {
 		snd_soc_component_update_bits(component, tx_reg0,
 					      CDC_WSA_TX_SPKR_PROT_RESET_MASK,
 					      CDC_WSA_TX_SPKR_PROT_RESET);
@@ -1516,6 +1501,72 @@ static int wsa_macro_enable_vi_feedback(struct snd_soc_dapm_widget *w,
 		snd_soc_component_update_bits(component, tx_reg1,
 					      CDC_WSA_TX_SPKR_PROT_CLK_EN_MASK,
 					      CDC_WSA_TX_SPKR_PROT_CLK_DISABLE);
+	}
+}
+
+static void wsa_macro_enable_disable_vi_feedback(struct snd_soc_component *component,
+						 bool enable, u32 rate)
+{
+	struct wsa_macro *wsa = snd_soc_component_get_drvdata(component);
+
+	if (test_bit(WSA_MACRO_TX0, &wsa->active_ch_mask[WSA_MACRO_AIF_VI]))
+		wsa_macro_enable_disable_vi_sense(component, enable,
+				CDC_WSA_TX0_SPKR_PROT_PATH_CTL,
+				CDC_WSA_TX1_SPKR_PROT_PATH_CTL, rate);
+
+	if (test_bit(WSA_MACRO_TX1, &wsa->active_ch_mask[WSA_MACRO_AIF_VI]))
+		wsa_macro_enable_disable_vi_sense(component, enable,
+				CDC_WSA_TX2_SPKR_PROT_PATH_CTL,
+				CDC_WSA_TX3_SPKR_PROT_PATH_CTL, rate);
+}
+
+static int wsa_macro_mclk_event(struct snd_soc_dapm_widget *w,
+				struct snd_kcontrol *kcontrol, int event)
+{
+	struct snd_soc_component *component = snd_soc_dapm_to_component(w->dapm);
+	struct wsa_macro *wsa = snd_soc_component_get_drvdata(component);
+
+	wsa_macro_mclk_enable(wsa, event == SND_SOC_DAPM_PRE_PMU);
+	return 0;
+}
+
+static int wsa_macro_enable_vi_feedback(struct snd_soc_dapm_widget *w,
+					struct snd_kcontrol *kcontrol,
+					int event)
+{
+	struct snd_soc_component *component = snd_soc_dapm_to_component(w->dapm);
+	struct wsa_macro *wsa = snd_soc_component_get_drvdata(component);
+	u32 rate_val;
+
+	switch (wsa->pcm_rate_vi) {
+	case 8000:
+		rate_val = CDC_WSA_TX_SPKR_PROT_PCM_RATE_8K;
+		break;
+	case 16000:
+		rate_val = CDC_WSA_TX_SPKR_PROT_PCM_RATE_16K;
+		break;
+	case 24000:
+		rate_val = CDC_WSA_TX_SPKR_PROT_PCM_RATE_24K;
+		break;
+	case 32000:
+		rate_val = CDC_WSA_TX_SPKR_PROT_PCM_RATE_32K;
+		break;
+	case 48000:
+		rate_val = CDC_WSA_TX_SPKR_PROT_PCM_RATE_48K;
+		break;
+	default:
+		rate_val = CDC_WSA_TX_SPKR_PROT_PCM_RATE_8K;
+		break;
+	}
+
+	switch (event) {
+	case SND_SOC_DAPM_POST_PMU:
+		/* Enable V&I sensing */
+		wsa_macro_enable_disable_vi_feedback(component, true, rate_val);
+		break;
+	case SND_SOC_DAPM_POST_PMD:
+		/* Disable V&I sensing */
+		wsa_macro_enable_disable_vi_feedback(component, false, rate_val);
 		break;
 	}
 
diff --git a/sound/soc/dwc/dwc-i2s.c b/sound/soc/dwc/dwc-i2s.c
index 4c4171bb3fbb..28001e9857d9 100644
--- a/sound/soc/dwc/dwc-i2s.c
+++ b/sound/soc/dwc/dwc-i2s.c
@@ -199,12 +199,10 @@ static void i2s_start(struct dw_i2s_dev *dev,
 	else
 		i2s_write_reg(dev->i2s_base, IRER, 1);
 
-	/* I2S needs to enable IRQ to make a handshake with DMAC on the JH7110 SoC */
-	if (dev->use_pio || dev->is_jh7110)
-		i2s_enable_irqs(dev, substream->stream, config->chan_nr);
-	else
+	if (!(dev->use_pio || dev->is_jh7110))
 		i2s_enable_dma(dev, substream->stream);
 
+	i2s_enable_irqs(dev, substream->stream, config->chan_nr);
 	i2s_write_reg(dev->i2s_base, CER, 1);
 }
 
@@ -218,11 +216,12 @@ static void i2s_stop(struct dw_i2s_dev *dev,
 	else
 		i2s_write_reg(dev->i2s_base, IRER, 0);
 
-	if (dev->use_pio || dev->is_jh7110)
-		i2s_disable_irqs(dev, substream->stream, 8);
-	else
+	if (!(dev->use_pio || dev->is_jh7110))
 		i2s_disable_dma(dev, substream->stream);
 
+	i2s_disable_irqs(dev, substream->stream, 8);
+
+
 	if (!dev->active) {
 		i2s_write_reg(dev->i2s_base, CER, 0);
 		i2s_write_reg(dev->i2s_base, IER, 0);
diff --git a/sound/soc/fsl/fsl_asrc_dma.c b/sound/soc/fsl/fsl_asrc_dma.c
index f501f47242fb..1bba48318e2d 100644
--- a/sound/soc/fsl/fsl_asrc_dma.c
+++ b/sound/soc/fsl/fsl_asrc_dma.c
@@ -156,11 +156,24 @@ static int fsl_asrc_dma_hw_params(struct snd_soc_component *component,
 	for_each_dpcm_be(rtd, stream, dpcm) {
 		struct snd_soc_pcm_runtime *be = dpcm->be;
 		struct snd_pcm_substream *substream_be;
-		struct snd_soc_dai *dai = snd_soc_rtd_to_cpu(be, 0);
+		struct snd_soc_dai *dai_cpu = snd_soc_rtd_to_cpu(be, 0);
+		struct snd_soc_dai *dai_codec = snd_soc_rtd_to_codec(be, 0);
+		struct snd_soc_dai *dai;
 
 		if (dpcm->fe != rtd)
 			continue;
 
+		/*
+		 * With audio graph card, original cpu dai is changed to codec
+		 * device in backend, so if cpu dai is dummy device in backend,
+		 * get the codec dai device, which is the real hardware device
+		 * connected.
+		 */
+		if (!snd_soc_dai_is_dummy(dai_cpu))
+			dai = dai_cpu;
+		else
+			dai = dai_codec;
+
 		substream_be = snd_soc_dpcm_get_substream(be, stream);
 		dma_params_be = snd_soc_dai_get_dma_data(dai, substream_be);
 		dev_be = dai->dev;
diff --git a/sound/soc/fsl/fsl_qmc_audio.c b/sound/soc/fsl/fsl_qmc_audio.c
index b2979290c973..5614a8b909ed 100644
--- a/sound/soc/fsl/fsl_qmc_audio.c
+++ b/sound/soc/fsl/fsl_qmc_audio.c
@@ -250,6 +250,9 @@ static int qmc_audio_pcm_trigger(struct snd_soc_component *component,
 	switch (cmd) {
 	case SNDRV_PCM_TRIGGER_START:
 		bitmap_zero(prtd->chans_pending, 64);
+		prtd->buffer_ended = 0;
+		prtd->ch_dma_addr_current = prtd->ch_dma_addr_start;
+
 		if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
 			for (i = 0; i < prtd->channels; i++)
 				prtd->qmc_dai->chans[i].prtd_tx = prtd;
diff --git a/sound/soc/fsl/imx-card.c b/sound/soc/fsl/imx-card.c
index 3686d468506b..45e000f61ecc 100644
--- a/sound/soc/fsl/imx-card.c
+++ b/sound/soc/fsl/imx-card.c
@@ -544,7 +544,7 @@ static int imx_card_parse_of(struct imx_card_data *data)
 	if (!card->dai_link)
 		return -ENOMEM;
 
-	data->link_data = devm_kcalloc(dev, num_links, sizeof(*link), GFP_KERNEL);
+	data->link_data = devm_kcalloc(dev, num_links, sizeof(*link_data), GFP_KERNEL);
 	if (!data->link_data)
 		return -ENOMEM;
 
diff --git a/sound/soc/generic/simple-card-utils.c b/sound/soc/generic/simple-card-utils.c
index a1ccc300e68c..3ae2a212a2e3 100644
--- a/sound/soc/generic/simple-card-utils.c
+++ b/sound/soc/generic/simple-card-utils.c
@@ -1174,9 +1174,9 @@ void graph_util_parse_link_direction(struct device_node *np,
 	bool is_playback_only = of_property_read_bool(np, "playback-only");
 	bool is_capture_only  = of_property_read_bool(np, "capture-only");
 
-	if (is_playback_only)
+	if (playback_only)
 		*playback_only = is_playback_only;
-	if (is_capture_only)
+	if (capture_only)
 		*capture_only = is_capture_only;
 }
 EXPORT_SYMBOL_GPL(graph_util_parse_link_direction);
diff --git a/sound/soc/intel/avs/path.c b/sound/soc/intel/avs/path.c
index ef0c1d125d66..cafb8c6198be 100644
--- a/sound/soc/intel/avs/path.c
+++ b/sound/soc/intel/avs/path.c
@@ -115,6 +115,78 @@ avs_path_find_variant(struct avs_dev *adev,
 	return NULL;
 }
 
+static struct acpi_nhlt_config *
+avs_nhlt_config_or_default(struct avs_dev *adev, struct avs_tplg_module *t);
+
+int avs_path_set_constraint(struct avs_dev *adev, struct avs_tplg_path_template *template,
+			    struct snd_pcm_hw_constraint_list *rate_list,
+			    struct snd_pcm_hw_constraint_list *channels_list,
+			    struct snd_pcm_hw_constraint_list *sample_bits_list)
+{
+	struct avs_tplg_path *path_template;
+	unsigned int *rlist, *clist, *slist;
+	size_t i;
+
+	i = 0;
+	list_for_each_entry(path_template, &template->path_list, node)
+		i++;
+
+	rlist = kcalloc(i, sizeof(rlist), GFP_KERNEL);
+	clist = kcalloc(i, sizeof(clist), GFP_KERNEL);
+	slist = kcalloc(i, sizeof(slist), GFP_KERNEL);
+
+	i = 0;
+	list_for_each_entry(path_template, &template->path_list, node) {
+		struct avs_tplg_pipeline *pipeline_template;
+
+		list_for_each_entry(pipeline_template, &path_template->ppl_list, node) {
+			struct avs_tplg_module *module_template;
+
+			list_for_each_entry(module_template, &pipeline_template->mod_list, node) {
+				const guid_t *type = &module_template->cfg_ext->type;
+				struct acpi_nhlt_config *blob;
+
+				if (!guid_equal(type, &AVS_COPIER_MOD_UUID) &&
+				    !guid_equal(type, &AVS_WOVHOSTM_MOD_UUID))
+					continue;
+
+				switch (module_template->cfg_ext->copier.dma_type) {
+				case AVS_DMA_DMIC_LINK_INPUT:
+				case AVS_DMA_I2S_LINK_OUTPUT:
+				case AVS_DMA_I2S_LINK_INPUT:
+					break;
+				default:
+					continue;
+				}
+
+				blob = avs_nhlt_config_or_default(adev, module_template);
+				if (IS_ERR(blob))
+					continue;
+
+				rlist[i] = path_template->fe_fmt->sampling_freq;
+				clist[i] = path_template->fe_fmt->num_channels;
+				slist[i] = path_template->fe_fmt->bit_depth;
+				i++;
+			}
+		}
+	}
+
+	if (i) {
+		rate_list->count = i;
+		rate_list->list = rlist;
+		channels_list->count = i;
+		channels_list->list = clist;
+		sample_bits_list->count = i;
+		sample_bits_list->list = slist;
+	} else {
+		kfree(rlist);
+		kfree(clist);
+		kfree(slist);
+	}
+
+	return i;
+}
+
 static void avs_init_node_id(union avs_connector_node_id *node_id,
 			     struct avs_tplg_modcfg_ext *te, u32 dma_id)
 {
diff --git a/sound/soc/intel/avs/path.h b/sound/soc/intel/avs/path.h
index 7ed7e94e0a56..c65ed84aa853 100644
--- a/sound/soc/intel/avs/path.h
+++ b/sound/soc/intel/avs/path.h
@@ -69,6 +69,11 @@ int avs_path_reset(struct avs_path *path);
 int avs_path_pause(struct avs_path *path);
 int avs_path_run(struct avs_path *path, int trigger);
 
+int avs_path_set_constraint(struct avs_dev *adev, struct avs_tplg_path_template *template,
+			    struct snd_pcm_hw_constraint_list *rate_list,
+			    struct snd_pcm_hw_constraint_list *channels_list,
+			    struct snd_pcm_hw_constraint_list *sample_bits_list);
+
 int avs_peakvol_set_volume(struct avs_dev *adev, struct avs_path_module *mod,
 			   struct soc_mixer_control *mc, long *input);
 int avs_peakvol_set_mute(struct avs_dev *adev, struct avs_path_module *mod,
diff --git a/sound/soc/intel/avs/pcm.c b/sound/soc/intel/avs/pcm.c
index dac463390da1..d83ef504643b 100644
--- a/sound/soc/intel/avs/pcm.c
+++ b/sound/soc/intel/avs/pcm.c
@@ -31,6 +31,10 @@ struct avs_dma_data {
 		struct hdac_ext_stream *host_stream;
 	};
 
+	struct snd_pcm_hw_constraint_list rate_list;
+	struct snd_pcm_hw_constraint_list channels_list;
+	struct snd_pcm_hw_constraint_list sample_bits_list;
+
 	struct work_struct period_elapsed_work;
 	struct snd_pcm_substream *substream;
 };
@@ -74,6 +78,45 @@ void avs_period_elapsed(struct snd_pcm_substream *substream)
 	schedule_work(&data->period_elapsed_work);
 }
 
+static int hw_rule_param_size(struct snd_pcm_hw_params *params, struct snd_pcm_hw_rule *rule);
+static int avs_hw_constraints_init(struct snd_pcm_substream *substream, struct snd_soc_dai *dai)
+{
+	struct snd_soc_pcm_runtime *rtd = snd_soc_substream_to_rtd(substream);
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct snd_pcm_hw_constraint_list *r, *c, *s;
+	struct avs_tplg_path_template *template;
+	struct avs_dma_data *data;
+	int ret;
+
+	ret = snd_pcm_hw_constraint_integer(runtime, SNDRV_PCM_HW_PARAM_PERIODS);
+	if (ret < 0)
+		return ret;
+
+	data = snd_soc_dai_get_dma_data(dai, substream);
+	r = &(data->rate_list);
+	c = &(data->channels_list);
+	s = &(data->sample_bits_list);
+
+	template = avs_dai_find_path_template(dai, !rtd->dai_link->no_pcm, substream->stream);
+	ret = avs_path_set_constraint(data->adev, template, r, c, s);
+	if (ret <= 0)
+		return ret;
+
+	ret = snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, r);
+	if (ret < 0)
+		return ret;
+
+	ret = snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS, c);
+	if (ret < 0)
+		return ret;
+
+	ret = snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_SAMPLE_BITS, s);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
 static int avs_dai_startup(struct snd_pcm_substream *substream, struct snd_soc_dai *dai)
 {
 	struct snd_soc_pcm_runtime *rtd = snd_soc_substream_to_rtd(substream);
@@ -101,7 +144,7 @@ static int avs_dai_startup(struct snd_pcm_substream *substream, struct snd_soc_d
 	if (rtd->dai_link->ignore_suspend)
 		adev->num_lp_paths++;
 
-	return 0;
+	return avs_hw_constraints_init(substream, dai);
 }
 
 static void avs_dai_shutdown(struct snd_pcm_substream *substream, struct snd_soc_dai *dai)
@@ -114,6 +157,10 @@ static void avs_dai_shutdown(struct snd_pcm_substream *substream, struct snd_soc
 	if (rtd->dai_link->ignore_suspend)
 		data->adev->num_lp_paths--;
 
+	kfree(data->rate_list.list);
+	kfree(data->channels_list.list);
+	kfree(data->sample_bits_list.list);
+
 	snd_soc_dai_set_dma_data(dai, substream, NULL);
 	kfree(data);
 }
@@ -927,7 +974,8 @@ static int avs_component_probe(struct snd_soc_component *component)
 		else
 			mach->tplg_filename = devm_kasprintf(adev->dev, GFP_KERNEL,
 							     "hda-generic-tplg.bin");
-
+		if (!mach->tplg_filename)
+			return -ENOMEM;
 		filename = kasprintf(GFP_KERNEL, "%s/%s", component->driver->topology_name_prefix,
 				     mach->tplg_filename);
 		if (!filename)
diff --git a/sound/soc/intel/boards/bytcr_rt5640.c b/sound/soc/intel/boards/bytcr_rt5640.c
index 6446cda0f857..0f3b8f44e701 100644
--- a/sound/soc/intel/boards/bytcr_rt5640.c
+++ b/sound/soc/intel/boards/bytcr_rt5640.c
@@ -576,6 +576,19 @@ static const struct dmi_system_id byt_rt5640_quirk_table[] = {
 					BYT_RT5640_SSP0_AIF2 |
 					BYT_RT5640_MCLK_EN),
 	},
+	{       /* Acer Aspire SW3-013 */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Aspire SW3-013"),
+		},
+		.driver_data = (void *)(BYT_RT5640_DMIC1_MAP |
+					BYT_RT5640_JD_SRC_JD2_IN4N |
+					BYT_RT5640_OVCD_TH_2000UA |
+					BYT_RT5640_OVCD_SF_0P75 |
+					BYT_RT5640_DIFF_MIC |
+					BYT_RT5640_SSP0_AIF1 |
+					BYT_RT5640_MCLK_EN),
+	},
 	{
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
diff --git a/sound/soc/intel/boards/sof_sdw.c b/sound/soc/intel/boards/sof_sdw.c
index 90dafa810b2e..095d08b3fc82 100644
--- a/sound/soc/intel/boards/sof_sdw.c
+++ b/sound/soc/intel/boards/sof_sdw.c
@@ -764,6 +764,7 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = {
 
 static const struct snd_pci_quirk sof_sdw_ssid_quirk_table[] = {
 	SND_PCI_QUIRK(0x1043, 0x1e13, "ASUS Zenbook S14", SOC_SDW_CODEC_MIC),
+	SND_PCI_QUIRK(0x1043, 0x1f43, "ASUS Zenbook S16", SOC_SDW_CODEC_MIC),
 	{}
 };
 
diff --git a/sound/soc/intel/catpt/dsp.c b/sound/soc/intel/catpt/dsp.c
index 5993819cc58a..008a20a2acbd 100644
--- a/sound/soc/intel/catpt/dsp.c
+++ b/sound/soc/intel/catpt/dsp.c
@@ -156,7 +156,7 @@ static void catpt_dsp_set_srampge(struct catpt_dev *cdev, struct resource *sram,
 {
 	unsigned long old;
 	u32 off = sram->start;
-	u32 b = __ffs(mask);
+	unsigned long b = __ffs(mask);
 
 	old = catpt_readl_pci(cdev, VDRTCTL0) & mask;
 	dev_dbg(cdev->dev, "SRAMPGE [0x%08lx] 0x%08lx -> 0x%08lx",
diff --git a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
index 6603d8de501c..c599eb43eeb1 100644
--- a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
+++ b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
@@ -431,7 +431,8 @@ static const struct snd_soc_acpi_link_adr ptl_cs42l43_l3[] = {
 		.mask = BIT(3),
 		.num_adr = ARRAY_SIZE(cs42l43_3_adr),
 		.adr_d = cs42l43_3_adr,
-	}
+	},
+	{}
 };
 
 static const struct snd_soc_acpi_link_adr ptl_rt722_only[] = {
diff --git a/sound/soc/qcom/lpass.h b/sound/soc/qcom/lpass.h
index 27a2bf9a6613..de3ec6f594c1 100644
--- a/sound/soc/qcom/lpass.h
+++ b/sound/soc/qcom/lpass.h
@@ -13,10 +13,11 @@
 #include <linux/platform_device.h>
 #include <linux/regmap.h>
 #include <dt-bindings/sound/qcom,lpass.h>
+#include <dt-bindings/sound/qcom,q6afe.h>
 #include "lpass-hdmi.h"
 
 #define LPASS_AHBIX_CLOCK_FREQUENCY		131072000
-#define LPASS_MAX_PORTS			(LPASS_CDC_DMA_VA_TX8 + 1)
+#define LPASS_MAX_PORTS			(DISPLAY_PORT_RX_7 + 1)
 #define LPASS_MAX_MI2S_PORTS			(8)
 #define LPASS_MAX_DMA_CHANNELS			(8)
 #define LPASS_MAX_HDMI_DMA_CHANNELS		(4)
diff --git a/sound/soc/renesas/rz-ssi.c b/sound/soc/renesas/rz-ssi.c
index 3a0af4ca7ab6..0f7458a43901 100644
--- a/sound/soc/renesas/rz-ssi.c
+++ b/sound/soc/renesas/rz-ssi.c
@@ -1244,7 +1244,7 @@ static int rz_ssi_runtime_resume(struct device *dev)
 
 static const struct dev_pm_ops rz_ssi_pm_ops = {
 	RUNTIME_PM_OPS(rz_ssi_runtime_suspend, rz_ssi_runtime_resume, NULL)
-	SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume)
+	NOIRQ_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume)
 };
 
 static struct platform_driver rz_ssi_driver = {
diff --git a/sound/soc/sdw_utils/soc_sdw_bridge_cs35l56.c b/sound/soc/sdw_utils/soc_sdw_bridge_cs35l56.c
index 246e5c2e0af5..c7e55f443351 100644
--- a/sound/soc/sdw_utils/soc_sdw_bridge_cs35l56.c
+++ b/sound/soc/sdw_utils/soc_sdw_bridge_cs35l56.c
@@ -60,6 +60,10 @@ static int asoc_sdw_bridge_cs35l56_asp_init(struct snd_soc_pcm_runtime *rtd)
 
 	/* 4 x 16-bit sample slots and FSYNC=48000, BCLK=3.072 MHz */
 	for_each_rtd_codec_dais(rtd, i, codec_dai) {
+		ret = asoc_sdw_cs35l56_volume_limit(card, codec_dai->component->name_prefix);
+		if (ret)
+			return ret;
+
 		ret = snd_soc_dai_set_tdm_slot(codec_dai, tx_mask, rx_mask, 4, 16);
 		if (ret < 0)
 			return ret;
diff --git a/sound/soc/sdw_utils/soc_sdw_cs42l43.c b/sound/soc/sdw_utils/soc_sdw_cs42l43.c
index 668c9d28a1c1..b415d45d520d 100644
--- a/sound/soc/sdw_utils/soc_sdw_cs42l43.c
+++ b/sound/soc/sdw_utils/soc_sdw_cs42l43.c
@@ -20,6 +20,8 @@
 #include <sound/soc-dapm.h>
 #include <sound/soc_sdw_utils.h>
 
+#define CS42L43_SPK_VOLUME_0DB	128 /* 0dB Max */
+
 static const struct snd_soc_dapm_route cs42l43_hs_map[] = {
 	{ "Headphone", NULL, "cs42l43 AMP3_OUT" },
 	{ "Headphone", NULL, "cs42l43 AMP4_OUT" },
@@ -117,6 +119,14 @@ int asoc_sdw_cs42l43_spk_rtd_init(struct snd_soc_pcm_runtime *rtd, struct snd_so
 			return -ENOMEM;
 	}
 
+	ret = snd_soc_limit_volume(card, "cs42l43 Speaker Digital Volume",
+				   CS42L43_SPK_VOLUME_0DB);
+	if (ret)
+		dev_err(card->dev, "cs42l43 speaker volume limit failed: %d\n", ret);
+	else
+		dev_info(card->dev, "Setting CS42L43 Speaker volume limit to %d\n",
+			 CS42L43_SPK_VOLUME_0DB);
+
 	ret = snd_soc_dapm_add_routes(&card->dapm, cs42l43_spk_map,
 				      ARRAY_SIZE(cs42l43_spk_map));
 	if (ret)
diff --git a/sound/soc/sdw_utils/soc_sdw_cs_amp.c b/sound/soc/sdw_utils/soc_sdw_cs_amp.c
index 4b6181cf2971..35b550bcd4de 100644
--- a/sound/soc/sdw_utils/soc_sdw_cs_amp.c
+++ b/sound/soc/sdw_utils/soc_sdw_cs_amp.c
@@ -16,6 +16,25 @@
 
 #define CODEC_NAME_SIZE	8
 #define CS_AMP_CHANNELS_PER_AMP	4
+#define CS35L56_SPK_VOLUME_0DB 400 /* 0dB Max */
+
+int asoc_sdw_cs35l56_volume_limit(struct snd_soc_card *card, const char *name_prefix)
+{
+	char *volume_ctl_name;
+	int ret;
+
+	volume_ctl_name = kasprintf(GFP_KERNEL, "%s Speaker Volume", name_prefix);
+	if (!volume_ctl_name)
+		return -ENOMEM;
+
+	ret = snd_soc_limit_volume(card, volume_ctl_name, CS35L56_SPK_VOLUME_0DB);
+	if (ret)
+		dev_err(card->dev, "%s limit set failed: %d\n", volume_ctl_name, ret);
+
+	kfree(volume_ctl_name);
+	return ret;
+}
+EXPORT_SYMBOL_NS(asoc_sdw_cs35l56_volume_limit, "SND_SOC_SDW_UTILS");
 
 int asoc_sdw_cs_spk_rtd_init(struct snd_soc_pcm_runtime *rtd, struct snd_soc_dai *dai)
 {
@@ -40,6 +59,11 @@ int asoc_sdw_cs_spk_rtd_init(struct snd_soc_pcm_runtime *rtd, struct snd_soc_dai
 
 		snprintf(widget_name, sizeof(widget_name), "%s SPK",
 			 codec_dai->component->name_prefix);
+
+		ret = asoc_sdw_cs35l56_volume_limit(card, codec_dai->component->name_prefix);
+		if (ret)
+			return ret;
+
 		ret = snd_soc_dapm_add_routes(&card->dapm, &route, 1);
 		if (ret)
 			return ret;
diff --git a/sound/soc/sdw_utils/soc_sdw_rt_dmic.c b/sound/soc/sdw_utils/soc_sdw_rt_dmic.c
index 46d917a99c51..97be110a59b6 100644
--- a/sound/soc/sdw_utils/soc_sdw_rt_dmic.c
+++ b/sound/soc/sdw_utils/soc_sdw_rt_dmic.c
@@ -29,6 +29,8 @@ int asoc_sdw_rt_dmic_rtd_init(struct snd_soc_pcm_runtime *rtd, struct snd_soc_da
 		mic_name = devm_kasprintf(card->dev, GFP_KERNEL, "rt715-sdca");
 	else
 		mic_name = devm_kasprintf(card->dev, GFP_KERNEL, "%s", component->name_prefix);
+	if (!mic_name)
+		return -ENOMEM;
 
 	card->components = devm_kasprintf(card->dev, GFP_KERNEL,
 					  "%s mic:%s", card->components,
diff --git a/sound/soc/soc-pcm.c b/sound/soc/soc-pcm.c
index 4308a6cbb2e6..43835197d1fe 100644
--- a/sound/soc/soc-pcm.c
+++ b/sound/soc/soc-pcm.c
@@ -1584,10 +1584,13 @@ int dpcm_add_paths(struct snd_soc_pcm_runtime *fe, int stream,
 		/*
 		 * Filter for systems with 'component_chaining' enabled.
 		 * This helps to avoid unnecessary re-configuration of an
-		 * already active BE on such systems.
+		 * already active BE on such systems and ensures the BE DAI
+		 * widget is powered ON after hw_params() BE DAI callback.
 		 */
 		if (fe->card->component_chaining &&
 		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_NEW) &&
+		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_OPEN) &&
+		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_HW_PARAMS) &&
 		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_CLOSE))
 			continue;
 
diff --git a/sound/soc/stm/stm32_sai_sub.c b/sound/soc/stm/stm32_sai_sub.c
index e8c1abf1ae0a..bf5299ba11c3 100644
--- a/sound/soc/stm/stm32_sai_sub.c
+++ b/sound/soc/stm/stm32_sai_sub.c
@@ -409,11 +409,11 @@ static int stm32_sai_set_parent_rate(struct stm32_sai_sub_data *sai,
 				     unsigned int rate)
 {
 	struct platform_device *pdev = sai->pdev;
-	unsigned int sai_ck_rate, sai_ck_max_rate, sai_curr_rate, sai_new_rate;
+	unsigned int sai_ck_rate, sai_ck_max_rate, sai_ck_min_rate, sai_curr_rate, sai_new_rate;
 	int div, ret;
 
 	/*
-	 * Set maximum expected kernel clock frequency
+	 * Set minimum and maximum expected kernel clock frequency
 	 * - mclk on or spdif:
 	 *   f_sai_ck = MCKDIV * mclk-fs * fs
 	 *   Here typical 256 ratio is assumed for mclk-fs
@@ -423,13 +423,16 @@ static int stm32_sai_set_parent_rate(struct stm32_sai_sub_data *sai,
 	 *   Set constraint MCKDIV * FRL <= 256, to ensure MCKDIV is in available range
 	 *   f_sai_ck = sai_ck_max_rate * pow_of_two(FRL) / 256
 	 */
+	sai_ck_min_rate = rate * 256;
 	if (!(rate % SAI_RATE_11K))
 		sai_ck_max_rate = SAI_MAX_SAMPLE_RATE_11K * 256;
 	else
 		sai_ck_max_rate = SAI_MAX_SAMPLE_RATE_8K * 256;
 
-	if (!sai->sai_mclk && !STM_SAI_PROTOCOL_IS_SPDIF(sai))
+	if (!sai->sai_mclk && !STM_SAI_PROTOCOL_IS_SPDIF(sai)) {
+		sai_ck_min_rate = rate * sai->fs_length;
 		sai_ck_max_rate /= DIV_ROUND_CLOSEST(256, roundup_pow_of_two(sai->fs_length));
+	}
 
 	/*
 	 * Request exclusivity, as the clock is shared by SAI sub-blocks and by
@@ -444,7 +447,10 @@ static int stm32_sai_set_parent_rate(struct stm32_sai_sub_data *sai,
 	 * return immediately.
 	 */
 	sai_curr_rate = clk_get_rate(sai->sai_ck);
-	if (stm32_sai_rate_accurate(sai_ck_max_rate, sai_curr_rate))
+	dev_dbg(&pdev->dev, "kernel clock rate: min [%u], max [%u], current [%u]",
+		sai_ck_min_rate, sai_ck_max_rate, sai_curr_rate);
+	if (stm32_sai_rate_accurate(sai_ck_max_rate, sai_curr_rate) &&
+	    sai_curr_rate >= sai_ck_min_rate)
 		return 0;
 
 	/*
@@ -472,7 +478,7 @@ static int stm32_sai_set_parent_rate(struct stm32_sai_sub_data *sai,
 		/* Try a lower frequency */
 		div++;
 		sai_ck_rate = sai_ck_max_rate / div;
-	} while (sai_ck_rate > rate);
+	} while (sai_ck_rate >= sai_ck_min_rate);
 
 	/* No accurate rate found */
 	dev_err(&pdev->dev, "Failed to find an accurate rate");
diff --git a/sound/usb/endpoint.c b/sound/usb/endpoint.c
index a29f28eb7d0c..f36ec98da460 100644
--- a/sound/usb/endpoint.c
+++ b/sound/usb/endpoint.c
@@ -926,6 +926,8 @@ static int endpoint_set_interface(struct snd_usb_audio *chip,
 {
 	int altset = set ? ep->altsetting : 0;
 	int err;
+	int retries = 0;
+	const int max_retries = 5;
 
 	if (ep->iface_ref->altset == altset)
 		return 0;
@@ -935,8 +937,13 @@ static int endpoint_set_interface(struct snd_usb_audio *chip,
 
 	usb_audio_dbg(chip, "Setting usb interface %d:%d for EP 0x%x\n",
 		      ep->iface, altset, ep->ep_num);
+retry:
 	err = usb_set_interface(chip->dev, ep->iface, altset);
 	if (err < 0) {
+		if (err == -EPROTO && ++retries <= max_retries) {
+			msleep(5 * (1 << (retries - 1)));
+			goto retry;
+		}
 		usb_audio_err_ratelimited(
 			chip, "%d:%d: usb_set_interface failed (%d)\n",
 			ep->iface, altset, err);
diff --git a/sound/usb/format.c b/sound/usb/format.c
index 9d32b21a5fbb..0ba4641a0eb1 100644
--- a/sound/usb/format.c
+++ b/sound/usb/format.c
@@ -260,7 +260,8 @@ static int parse_audio_format_rates_v1(struct snd_usb_audio *chip, struct audiof
 	}
 
 	/* Jabra Evolve 65 headset */
-	if (chip->usb_id == USB_ID(0x0b0e, 0x030b)) {
+	if (chip->usb_id == USB_ID(0x0b0e, 0x030b) ||
+	    chip->usb_id == USB_ID(0x0b0e, 0x030c)) {
 		/* only 48kHz for playback while keeping 16kHz for capture */
 		if (fp->nr_rates != 1)
 			return set_fixed_rate(fp, 48000, SNDRV_PCM_RATE_48000);
diff --git a/sound/usb/midi.c b/sound/usb/midi.c
index dcdd7e9e1ae9..cfed000f243a 100644
--- a/sound/usb/midi.c
+++ b/sound/usb/midi.c
@@ -1885,10 +1885,18 @@ static void snd_usbmidi_init_substream(struct snd_usb_midi *umidi,
 	}
 
 	port_info = find_port_info(umidi, number);
-	name_format = port_info ? port_info->name :
-		(jack_name != default_jack_name  ? "%s %s" : "%s %s %d");
-	snprintf(substream->name, sizeof(substream->name),
-		 name_format, umidi->card->shortname, jack_name, number + 1);
+	if (port_info || jack_name == default_jack_name ||
+	    strncmp(umidi->card->shortname, jack_name, strlen(umidi->card->shortname)) != 0) {
+		name_format = port_info ? port_info->name :
+			(jack_name != default_jack_name  ? "%s %s" : "%s %s %d");
+		snprintf(substream->name, sizeof(substream->name),
+			 name_format, umidi->card->shortname, jack_name, number + 1);
+	} else {
+		/* The manufacturer included the iProduct name in the jack
+		 * name, do not use both
+		 */
+		strscpy(substream->name, jack_name);
+	}
 
 	*rsubstream = substream;
 }
diff --git a/tools/arch/arm64/include/uapi/asm/kvm.h b/tools/arch/arm64/include/uapi/asm/kvm.h
index 6d44f8c8a18f..af9d9acaf997 100644
--- a/tools/arch/arm64/include/uapi/asm/kvm.h
+++ b/tools/arch/arm64/include/uapi/asm/kvm.h
@@ -43,9 +43,6 @@
 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1
 #define KVM_DIRTY_LOG_PAGE_OFFSET 64
 
-#define KVM_REG_SIZE(id)						\
-	(1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
-
 struct kvm_regs {
 	struct user_pt_regs regs;	/* sp = sp_el0 */
 
@@ -108,6 +105,7 @@ struct kvm_regs {
 #define KVM_ARM_VCPU_PTRAUTH_ADDRESS	5 /* VCPU uses address authentication */
 #define KVM_ARM_VCPU_PTRAUTH_GENERIC	6 /* VCPU uses generic authentication */
 #define KVM_ARM_VCPU_HAS_EL2		7 /* Support nested virtualization */
+#define KVM_ARM_VCPU_HAS_EL2_E2H0	8 /* Limit NV support to E2H RES0 */
 
 struct kvm_vcpu_init {
 	__u32 target;
@@ -418,6 +416,7 @@ enum {
 #define KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS 6
 #define KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO  7
 #define KVM_DEV_ARM_VGIC_GRP_ITS_REGS 8
+#define KVM_DEV_ARM_VGIC_GRP_MAINT_IRQ  9
 #define KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT	10
 #define KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK \
 			(0x3fffffULL << KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT)
diff --git a/tools/arch/arm64/include/uapi/asm/unistd.h b/tools/arch/arm64/include/uapi/asm/unistd.h
index 9306726337fe..df36f23876e8 100644
--- a/tools/arch/arm64/include/uapi/asm/unistd.h
+++ b/tools/arch/arm64/include/uapi/asm/unistd.h
@@ -1,24 +1,2 @@
 /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
-/*
- * Copyright (C) 2012 ARM Ltd.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define __ARCH_WANT_RENAMEAT
-#define __ARCH_WANT_NEW_STAT
-#define __ARCH_WANT_SET_GET_RLIMIT
-#define __ARCH_WANT_TIME32_SYSCALLS
-#define __ARCH_WANT_MEMFD_SECRET
-
-#include <asm-generic/unistd.h>
+#include <asm/unistd_64.h>
diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h
index 9e3fa7942e7d..6c2c152d8a67 100644
--- a/tools/arch/x86/include/asm/cpufeatures.h
+++ b/tools/arch/x86/include/asm/cpufeatures.h
@@ -75,8 +75,8 @@
 #define X86_FEATURE_CENTAUR_MCR		( 3*32+ 3) /* "centaur_mcr" Centaur MCRs (= MTRRs) */
 #define X86_FEATURE_K8			( 3*32+ 4) /* Opteron, Athlon64 */
 #define X86_FEATURE_ZEN5		( 3*32+ 5) /* CPU based on Zen5 microarchitecture */
-#define X86_FEATURE_P3			( 3*32+ 6) /* P3 */
-#define X86_FEATURE_P4			( 3*32+ 7) /* P4 */
+/* Free                                 ( 3*32+ 6) */
+/* Free                                 ( 3*32+ 7) */
 #define X86_FEATURE_CONSTANT_TSC	( 3*32+ 8) /* "constant_tsc" TSC ticks at a constant rate */
 #define X86_FEATURE_UP			( 3*32+ 9) /* "up" SMP kernel running on UP */
 #define X86_FEATURE_ART			( 3*32+10) /* "art" Always running timer (ART) */
@@ -329,6 +329,7 @@
 #define X86_FEATURE_CLZERO		(13*32+ 0) /* "clzero" CLZERO instruction */
 #define X86_FEATURE_IRPERF		(13*32+ 1) /* "irperf" Instructions Retired Count */
 #define X86_FEATURE_XSAVEERPTR		(13*32+ 2) /* "xsaveerptr" Always save/restore FP error pointers */
+#define X86_FEATURE_INVLPGB		(13*32+ 3) /* INVLPGB and TLBSYNC instructions supported */
 #define X86_FEATURE_RDPRU		(13*32+ 4) /* "rdpru" Read processor register at user level */
 #define X86_FEATURE_WBNOINVD		(13*32+ 9) /* "wbnoinvd" WBNOINVD instruction */
 #define X86_FEATURE_AMD_IBPB		(13*32+12) /* Indirect Branch Prediction Barrier */
@@ -377,6 +378,7 @@
 #define X86_FEATURE_V_SPEC_CTRL		(15*32+20) /* "v_spec_ctrl" Virtual SPEC_CTRL */
 #define X86_FEATURE_VNMI		(15*32+25) /* "vnmi" Virtual NMI */
 #define X86_FEATURE_SVME_ADDR_CHK	(15*32+28) /* SVME addr check */
+#define X86_FEATURE_IDLE_HLT		(15*32+30) /* IDLE HLT intercept */
 
 /* Intel-defined CPU features, CPUID level 0x00000007:0 (ECX), word 16 */
 #define X86_FEATURE_AVX512VBMI		(16*32+ 1) /* "avx512vbmi" AVX512 Vector Bit Manipulation instructions*/
@@ -434,15 +436,18 @@
 #define X86_FEATURE_SPEC_CTRL_SSBD	(18*32+31) /* Speculative Store Bypass Disable */
 
 /* AMD-defined memory encryption features, CPUID level 0x8000001f (EAX), word 19 */
-#define X86_FEATURE_SME			(19*32+ 0) /* "sme" AMD Secure Memory Encryption */
-#define X86_FEATURE_SEV			(19*32+ 1) /* "sev" AMD Secure Encrypted Virtualization */
+#define X86_FEATURE_SME			(19*32+ 0) /* "sme" Secure Memory Encryption */
+#define X86_FEATURE_SEV			(19*32+ 1) /* "sev" Secure Encrypted Virtualization */
 #define X86_FEATURE_VM_PAGE_FLUSH	(19*32+ 2) /* VM Page Flush MSR is supported */
-#define X86_FEATURE_SEV_ES		(19*32+ 3) /* "sev_es" AMD Secure Encrypted Virtualization - Encrypted State */
-#define X86_FEATURE_SEV_SNP		(19*32+ 4) /* "sev_snp" AMD Secure Encrypted Virtualization - Secure Nested Paging */
+#define X86_FEATURE_SEV_ES		(19*32+ 3) /* "sev_es" Secure Encrypted Virtualization - Encrypted State */
+#define X86_FEATURE_SEV_SNP		(19*32+ 4) /* "sev_snp" Secure Encrypted Virtualization - Secure Nested Paging */
 #define X86_FEATURE_V_TSC_AUX		(19*32+ 9) /* Virtual TSC_AUX */
-#define X86_FEATURE_SME_COHERENT	(19*32+10) /* AMD hardware-enforced cache coherency */
-#define X86_FEATURE_DEBUG_SWAP		(19*32+14) /* "debug_swap" AMD SEV-ES full debug state swap support */
+#define X86_FEATURE_SME_COHERENT	(19*32+10) /* hardware-enforced cache coherency */
+#define X86_FEATURE_DEBUG_SWAP		(19*32+14) /* "debug_swap" SEV-ES full debug state swap support */
+#define X86_FEATURE_RMPREAD		(19*32+21) /* RMPREAD instruction */
+#define X86_FEATURE_SEGMENTED_RMP	(19*32+23) /* Segmented RMP support */
 #define X86_FEATURE_SVSM		(19*32+28) /* "svsm" SVSM present */
+#define X86_FEATURE_HV_INUSE_WR_ALLOWED	(19*32+30) /* Allow Write to in-use hypervisor-owned pages */
 
 /* AMD-defined Extended Feature 2 EAX, CPUID level 0x80000021 (EAX), word 20 */
 #define X86_FEATURE_NO_NESTED_DATA_BP	(20*32+ 0) /* No Nested Data Breakpoints */
@@ -455,6 +460,11 @@
 #define X86_FEATURE_SBPB		(20*32+27) /* Selective Branch Prediction Barrier */
 #define X86_FEATURE_IBPB_BRTYPE		(20*32+28) /* MSR_PRED_CMD[IBPB] flushes all branch type predictions */
 #define X86_FEATURE_SRSO_NO		(20*32+29) /* CPU is not affected by SRSO */
+#define X86_FEATURE_SRSO_USER_KERNEL_NO	(20*32+30) /* CPU is not affected by SRSO across user/kernel boundaries */
+#define X86_FEATURE_SRSO_BP_SPEC_REDUCE	(20*32+31) /*
+						    * BP_CFG[BpSpecReduce] can be used to mitigate SRSO for VMs.
+						    * (SRSO_MSR_FIX in the official doc).
+						    */
 
 /*
  * Extended auxiliary flags: Linux defined - for features scattered in various
@@ -470,6 +480,7 @@
 #define X86_FEATURE_AMD_FAST_CPPC	(21*32 + 5) /* Fast CPPC */
 #define X86_FEATURE_AMD_HETEROGENEOUS_CORES (21*32 + 6) /* Heterogeneous Core Topology */
 #define X86_FEATURE_AMD_WORKLOAD_CLASS	(21*32 + 7) /* Workload Classification */
+#define X86_FEATURE_PREFER_YMM		(21*32 + 8) /* Avoid ZMM registers due to downclocking */
 
 /*
  * BUG word(s)
@@ -521,4 +532,5 @@
 #define X86_BUG_RFDS			X86_BUG(1*32 + 2) /* "rfds" CPU is vulnerable to Register File Data Sampling */
 #define X86_BUG_BHI			X86_BUG(1*32 + 3) /* "bhi" CPU is affected by Branch History Injection */
 #define X86_BUG_IBPB_NO_RET	   	X86_BUG(1*32 + 4) /* "ibpb_no_ret" IBPB omits return target predictions */
+#define X86_BUG_SPECTRE_V2_USER		X86_BUG(1*32 + 5) /* "spectre_v2_user" CPU is affected by Spectre variant 2 attack between user processes */
 #endif /* _ASM_X86_CPUFEATURES_H */
diff --git a/tools/arch/x86/include/asm/msr-index.h b/tools/arch/x86/include/asm/msr-index.h
index dc1c1057f26e..e6134ef2263d 100644
--- a/tools/arch/x86/include/asm/msr-index.h
+++ b/tools/arch/x86/include/asm/msr-index.h
@@ -397,7 +397,8 @@
 #define MSR_IA32_PASID_VALID		BIT_ULL(31)
 
 /* DEBUGCTLMSR bits (others vary by model): */
-#define DEBUGCTLMSR_LBR			(1UL <<  0) /* last branch recording */
+#define DEBUGCTLMSR_LBR_BIT		0	     /* last branch recording */
+#define DEBUGCTLMSR_LBR			(1UL <<  DEBUGCTLMSR_LBR_BIT)
 #define DEBUGCTLMSR_BTF_SHIFT		1
 #define DEBUGCTLMSR_BTF			(1UL <<  1) /* single-step on branches */
 #define DEBUGCTLMSR_BUS_LOCK_DETECT	(1UL <<  2)
@@ -610,6 +611,7 @@
 #define MSR_AMD_PERF_CTL		0xc0010062
 #define MSR_AMD_PERF_STATUS		0xc0010063
 #define MSR_AMD_PSTATE_DEF_BASE		0xc0010064
+#define MSR_AMD64_GUEST_TSC_FREQ	0xc0010134
 #define MSR_AMD64_OSVW_ID_LENGTH	0xc0010140
 #define MSR_AMD64_OSVW_STATUS		0xc0010141
 #define MSR_AMD_PPIN_CTL		0xc00102f0
@@ -646,6 +648,7 @@
 #define MSR_AMD64_IBS_REG_COUNT_MAX	8 /* includes MSR_AMD64_IBSBRTARGET */
 #define MSR_AMD64_SVM_AVIC_DOORBELL	0xc001011b
 #define MSR_AMD64_VM_PAGE_FLUSH		0xc001011e
+#define MSR_AMD64_VIRT_SPEC_CTRL	0xc001011f
 #define MSR_AMD64_SEV_ES_GHCB		0xc0010130
 #define MSR_AMD64_SEV			0xc0010131
 #define MSR_AMD64_SEV_ENABLED_BIT	0
@@ -684,11 +687,12 @@
 #define MSR_AMD64_SNP_SMT_PROT		BIT_ULL(MSR_AMD64_SNP_SMT_PROT_BIT)
 #define MSR_AMD64_SNP_RESV_BIT		18
 #define MSR_AMD64_SNP_RESERVED_MASK	GENMASK_ULL(63, MSR_AMD64_SNP_RESV_BIT)
-
-#define MSR_AMD64_VIRT_SPEC_CTRL	0xc001011f
-
 #define MSR_AMD64_RMP_BASE		0xc0010132
 #define MSR_AMD64_RMP_END		0xc0010133
+#define MSR_AMD64_RMP_CFG		0xc0010136
+#define MSR_AMD64_SEG_RMP_ENABLED_BIT	0
+#define MSR_AMD64_SEG_RMP_ENABLED	BIT_ULL(MSR_AMD64_SEG_RMP_ENABLED_BIT)
+#define MSR_AMD64_RMP_SEGMENT_SHIFT(x)	(((x) & GENMASK_ULL(13, 8)) >> 8)
 
 #define MSR_SVSM_CAA			0xc001f000
 
@@ -699,15 +703,17 @@
 #define MSR_AMD_CPPC_REQ		0xc00102b3
 #define MSR_AMD_CPPC_STATUS		0xc00102b4
 
-#define AMD_CPPC_LOWEST_PERF(x)		(((x) >> 0) & 0xff)
-#define AMD_CPPC_LOWNONLIN_PERF(x)	(((x) >> 8) & 0xff)
-#define AMD_CPPC_NOMINAL_PERF(x)	(((x) >> 16) & 0xff)
-#define AMD_CPPC_HIGHEST_PERF(x)	(((x) >> 24) & 0xff)
+/* Masks for use with MSR_AMD_CPPC_CAP1 */
+#define AMD_CPPC_LOWEST_PERF_MASK	GENMASK(7, 0)
+#define AMD_CPPC_LOWNONLIN_PERF_MASK	GENMASK(15, 8)
+#define AMD_CPPC_NOMINAL_PERF_MASK	GENMASK(23, 16)
+#define AMD_CPPC_HIGHEST_PERF_MASK	GENMASK(31, 24)
 
-#define AMD_CPPC_MAX_PERF(x)		(((x) & 0xff) << 0)
-#define AMD_CPPC_MIN_PERF(x)		(((x) & 0xff) << 8)
-#define AMD_CPPC_DES_PERF(x)		(((x) & 0xff) << 16)
-#define AMD_CPPC_ENERGY_PERF_PREF(x)	(((x) & 0xff) << 24)
+/* Masks for use with MSR_AMD_CPPC_REQ */
+#define AMD_CPPC_MAX_PERF_MASK		GENMASK(7, 0)
+#define AMD_CPPC_MIN_PERF_MASK		GENMASK(15, 8)
+#define AMD_CPPC_DES_PERF_MASK		GENMASK(23, 16)
+#define AMD_CPPC_EPP_PERF_MASK		GENMASK(31, 24)
 
 /* AMD Performance Counter Global Status and Control MSRs */
 #define MSR_AMD64_PERF_CNTR_GLOBAL_STATUS	0xc0000300
@@ -719,6 +725,7 @@
 
 /* Zen4 */
 #define MSR_ZEN4_BP_CFG                 0xc001102e
+#define MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT 4
 #define MSR_ZEN4_BP_CFG_SHARED_BTB_FIX_BIT 5
 
 /* Fam 19h MSRs */
diff --git a/tools/arch/x86/include/uapi/asm/kvm.h b/tools/arch/x86/include/uapi/asm/kvm.h
index 88585c1de416..460306b35a4b 100644
--- a/tools/arch/x86/include/uapi/asm/kvm.h
+++ b/tools/arch/x86/include/uapi/asm/kvm.h
@@ -559,6 +559,9 @@ struct kvm_x86_mce {
 #define KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE	(1 << 7)
 #define KVM_XEN_HVM_CONFIG_SHARED_INFO_HVA	(1 << 8)
 
+#define KVM_XEN_MSR_MIN_INDEX			0x40000000u
+#define KVM_XEN_MSR_MAX_INDEX			0x4fffffffu
+
 struct kvm_xen_hvm_config {
 	__u32 flags;
 	__u32 msr;
@@ -925,5 +928,6 @@ struct kvm_hyperv_eventfd {
 #define KVM_X86_SEV_VM		2
 #define KVM_X86_SEV_ES_VM	3
 #define KVM_X86_SNP_VM		4
+#define KVM_X86_TDX_VM		5
 
 #endif /* _ASM_X86_KVM_H */
diff --git a/tools/arch/x86/include/uapi/asm/svm.h b/tools/arch/x86/include/uapi/asm/svm.h
index 1814b413fd57..ec1321248dac 100644
--- a/tools/arch/x86/include/uapi/asm/svm.h
+++ b/tools/arch/x86/include/uapi/asm/svm.h
@@ -95,6 +95,7 @@
 #define SVM_EXIT_CR14_WRITE_TRAP		0x09e
 #define SVM_EXIT_CR15_WRITE_TRAP		0x09f
 #define SVM_EXIT_INVPCID       0x0a2
+#define SVM_EXIT_IDLE_HLT      0x0a6
 #define SVM_EXIT_NPF           0x400
 #define SVM_EXIT_AVIC_INCOMPLETE_IPI		0x401
 #define SVM_EXIT_AVIC_UNACCELERATED_ACCESS	0x402
@@ -224,6 +225,7 @@
 	{ SVM_EXIT_CR4_WRITE_TRAP,	"write_cr4_trap" }, \
 	{ SVM_EXIT_CR8_WRITE_TRAP,	"write_cr8_trap" }, \
 	{ SVM_EXIT_INVPCID,     "invpcid" }, \
+	{ SVM_EXIT_IDLE_HLT,     "idle-halt" }, \
 	{ SVM_EXIT_NPF,         "npf" }, \
 	{ SVM_EXIT_AVIC_INCOMPLETE_IPI,		"avic_incomplete_ipi" }, \
 	{ SVM_EXIT_AVIC_UNACCELERATED_ACCESS,   "avic_unaccelerated_access" }, \
diff --git a/tools/arch/x86/lib/memset_64.S b/tools/arch/x86/lib/memset_64.S
index 0199d56cb479..d66b710d628f 100644
--- a/tools/arch/x86/lib/memset_64.S
+++ b/tools/arch/x86/lib/memset_64.S
@@ -3,6 +3,7 @@
 
 #include <linux/export.h>
 #include <linux/linkage.h>
+#include <linux/cfi_types.h>
 #include <asm/cpufeatures.h>
 #include <asm/alternative.h>
 
@@ -28,7 +29,7 @@
  * only for the return value that is the same as the source input,
  * which the compiler could/should do much better anyway.
  */
-SYM_FUNC_START(__memset)
+SYM_TYPED_FUNC_START(__memset)
 	ALTERNATIVE "jmp memset_orig", "", X86_FEATURE_FSRS
 
 	movq %rdi,%r9
diff --git a/tools/arch/x86/lib/x86-opcode-map.txt b/tools/arch/x86/lib/x86-opcode-map.txt
index caedb3ef6688..f5dd84eb55dc 100644
--- a/tools/arch/x86/lib/x86-opcode-map.txt
+++ b/tools/arch/x86/lib/x86-opcode-map.txt
@@ -996,8 +996,8 @@ AVXcode: 4
 83: Grp1 Ev,Ib (1A),(es)
 # CTESTSCC instructions are: CTESTB, CTESTBE, CTESTF, CTESTL, CTESTLE, CTESTNB, CTESTNBE, CTESTNL,
 #			     CTESTNLE, CTESTNO, CTESTNS, CTESTNZ, CTESTO, CTESTS, CTESTT, CTESTZ
-84: CTESTSCC (ev)
-85: CTESTSCC (es) | CTESTSCC (66),(es)
+84: CTESTSCC Eb,Gb (ev)
+85: CTESTSCC Ev,Gv (es) | CTESTSCC Ev,Gv (66),(es)
 88: POPCNT Gv,Ev (es) | POPCNT Gv,Ev (66),(es)
 8f: POP2 Bq,Rq (000),(11B),(ev)
 a5: SHLD Ev,Gv,CL (es) | SHLD Ev,Gv,CL (66),(es)
diff --git a/tools/hv/hv_kvp_daemon.c b/tools/hv/hv_kvp_daemon.c
index 04ba035d67e9..b9ce3aab15fe 100644
--- a/tools/hv/hv_kvp_daemon.c
+++ b/tools/hv/hv_kvp_daemon.c
@@ -24,6 +24,7 @@
 
 #include <sys/poll.h>
 #include <sys/utsname.h>
+#include <stdbool.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <unistd.h>
@@ -677,6 +678,88 @@ static void kvp_process_ipconfig_file(char *cmd,
 	pclose(file);
 }
 
+static bool kvp_verify_ip_address(const void *address_string)
+{
+	char verify_buf[sizeof(struct in6_addr)];
+
+	if (inet_pton(AF_INET, address_string, verify_buf) == 1)
+		return true;
+	if (inet_pton(AF_INET6, address_string, verify_buf) == 1)
+		return true;
+	return false;
+}
+
+static void kvp_extract_routes(const char *line, void **output, size_t *remaining)
+{
+	static const char needle[] = "via ";
+	const char *match, *haystack = line;
+
+	while ((match = strstr(haystack, needle))) {
+		const char *address, *next_char;
+
+		/* Address starts after needle. */
+		address = match + strlen(needle);
+
+		/* The char following address is a space or end of line. */
+		next_char = strpbrk(address, " \t\\");
+		if (!next_char)
+			next_char = address + strlen(address) + 1;
+
+		/* Enough room for address and semicolon. */
+		if (*remaining >= (next_char - address) + 1) {
+			memcpy(*output, address, next_char - address);
+			/* Terminate string for verification. */
+			memcpy(*output + (next_char - address), "", 1);
+			if (kvp_verify_ip_address(*output)) {
+				/* Advance output buffer. */
+				*output += next_char - address;
+				*remaining -= next_char - address;
+
+				/* Each address needs a trailing semicolon. */
+				memcpy(*output, ";", 1);
+				*output += 1;
+				*remaining -= 1;
+			}
+		}
+		haystack = next_char;
+	}
+}
+
+static void kvp_get_gateway(void *buffer, size_t buffer_len)
+{
+	static const char needle[] = "default ";
+	FILE *f;
+	void *output = buffer;
+	char *line = NULL;
+	size_t alloc_size = 0, remaining = buffer_len - 1;
+	ssize_t num_chars;
+
+	/* Show route information in a single line, for each address family */
+	f = popen("ip --oneline -4 route show;ip --oneline -6 route show", "r");
+	if (!f) {
+		/* Convert buffer into C-String. */
+		memcpy(output, "", 1);
+		return;
+	}
+	while ((num_chars = getline(&line, &alloc_size, f)) > 0) {
+		/* Skip short lines. */
+		if (num_chars <= strlen(needle))
+			continue;
+		/* Skip lines without default route. */
+		if (memcmp(line, needle, strlen(needle)))
+			continue;
+		/* Remove trailing newline to simplify further parsing. */
+		if (line[num_chars - 1] == '\n')
+			line[num_chars - 1] = '\0';
+		/* Search routes after match. */
+		kvp_extract_routes(line + strlen(needle), &output, &remaining);
+	}
+	/* Convert buffer into C-String. */
+	memcpy(output, "", 1);
+	free(line);
+	pclose(f);
+}
+
 static void kvp_get_ipconfig_info(char *if_name,
 				 struct hv_kvp_ipaddr_value *buffer)
 {
@@ -685,30 +768,7 @@ static void kvp_get_ipconfig_info(char *if_name,
 	char *p;
 	FILE *file;
 
-	/*
-	 * Get the address of default gateway (ipv4).
-	 */
-	sprintf(cmd, "%s %s", "ip route show dev", if_name);
-	strcat(cmd, " | awk '/default/ {print $3 }'");
-
-	/*
-	 * Execute the command to gather gateway info.
-	 */
-	kvp_process_ipconfig_file(cmd, (char *)buffer->gate_way,
-				(MAX_GATEWAY_SIZE * 2), INET_ADDRSTRLEN, 0);
-
-	/*
-	 * Get the address of default gateway (ipv6).
-	 */
-	sprintf(cmd, "%s %s", "ip -f inet6  route show dev", if_name);
-	strcat(cmd, " | awk '/default/ {print $3 }'");
-
-	/*
-	 * Execute the command to gather gateway info (ipv6).
-	 */
-	kvp_process_ipconfig_file(cmd, (char *)buffer->gate_way,
-				(MAX_GATEWAY_SIZE * 2), INET6_ADDRSTRLEN, 1);
-
+	kvp_get_gateway(buffer->gate_way, sizeof(buffer->gate_way));
 
 	/*
 	 * Gather the DNS state.
diff --git a/tools/include/linux/cfi_types.h b/tools/include/linux/cfi_types.h
new file mode 100644
index 000000000000..6b8713675765
--- /dev/null
+++ b/tools/include/linux/cfi_types.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Clang Control Flow Integrity (CFI) type definitions.
+ */
+#ifndef _LINUX_CFI_TYPES_H
+#define _LINUX_CFI_TYPES_H
+
+#ifdef __ASSEMBLY__
+#include <linux/linkage.h>
+
+#ifdef CONFIG_CFI_CLANG
+/*
+ * Use the __kcfi_typeid_<function> type identifier symbol to
+ * annotate indirectly called assembly functions. The compiler emits
+ * these symbols for all address-taken function declarations in C
+ * code.
+ */
+#ifndef __CFI_TYPE
+#define __CFI_TYPE(name)				\
+	.4byte __kcfi_typeid_##name
+#endif
+
+#define SYM_TYPED_ENTRY(name, linkage, align...)	\
+	linkage(name) ASM_NL				\
+	align ASM_NL					\
+	__CFI_TYPE(name) ASM_NL				\
+	name:
+
+#define SYM_TYPED_START(name, linkage, align...)	\
+	SYM_TYPED_ENTRY(name, linkage, align)
+
+#else /* CONFIG_CFI_CLANG */
+
+#define SYM_TYPED_START(name, linkage, align...)	\
+	SYM_START(name, linkage, align)
+
+#endif /* CONFIG_CFI_CLANG */
+
+#ifndef SYM_TYPED_FUNC_START
+#define SYM_TYPED_FUNC_START(name) 			\
+	SYM_TYPED_START(name, SYM_L_GLOBAL, SYM_A_ALIGN)
+#endif
+
+#endif /* __ASSEMBLY__ */
+#endif /* _LINUX_CFI_TYPES_H */
diff --git a/tools/include/uapi/asm-generic/mman-common.h b/tools/include/uapi/asm-generic/mman-common.h
index 1ea2c4c33b86..ef1c27fa3c57 100644
--- a/tools/include/uapi/asm-generic/mman-common.h
+++ b/tools/include/uapi/asm-generic/mman-common.h
@@ -85,6 +85,7 @@
 /* compatibility flags */
 #define MAP_FILE	0
 
+#define PKEY_UNRESTRICTED	0x0
 #define PKEY_DISABLE_ACCESS	0x1
 #define PKEY_DISABLE_WRITE	0x2
 #define PKEY_ACCESS_MASK	(PKEY_DISABLE_ACCESS |\
diff --git a/tools/include/uapi/asm-generic/unistd.h b/tools/include/uapi/asm-generic/unistd.h
index 88dc393c2bca..2892a45023af 100644
--- a/tools/include/uapi/asm-generic/unistd.h
+++ b/tools/include/uapi/asm-generic/unistd.h
@@ -849,9 +849,11 @@ __SYSCALL(__NR_getxattrat, sys_getxattrat)
 __SYSCALL(__NR_listxattrat, sys_listxattrat)
 #define __NR_removexattrat 466
 __SYSCALL(__NR_removexattrat, sys_removexattrat)
+#define __NR_open_tree_attr 467
+__SYSCALL(__NR_open_tree_attr, sys_open_tree_attr)
 
 #undef __NR_syscalls
-#define __NR_syscalls 467
+#define __NR_syscalls 468
 
 /*
  * 32 bit systems traditionally used different
diff --git a/tools/include/uapi/linux/in.h b/tools/include/uapi/linux/in.h
index 5d32d53508d9..ced0fc3c3aa5 100644
--- a/tools/include/uapi/linux/in.h
+++ b/tools/include/uapi/linux/in.h
@@ -79,6 +79,8 @@ enum {
 #define IPPROTO_MPLS		IPPROTO_MPLS
   IPPROTO_ETHERNET = 143,	/* Ethernet-within-IPv6 Encapsulation	*/
 #define IPPROTO_ETHERNET	IPPROTO_ETHERNET
+  IPPROTO_AGGFRAG = 144,	/* AGGFRAG in ESP (RFC 9347)		*/
+#define IPPROTO_AGGFRAG		IPPROTO_AGGFRAG
   IPPROTO_RAW = 255,		/* Raw IP packets			*/
 #define IPPROTO_RAW		IPPROTO_RAW
   IPPROTO_SMC = 256,		/* Shared Memory Communications		*/
diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h
index 502ea63b5d2e..b6ae8ad8934b 100644
--- a/tools/include/uapi/linux/kvm.h
+++ b/tools/include/uapi/linux/kvm.h
@@ -617,10 +617,6 @@ struct kvm_ioeventfd {
 #define KVM_X86_DISABLE_EXITS_HLT            (1 << 1)
 #define KVM_X86_DISABLE_EXITS_PAUSE          (1 << 2)
 #define KVM_X86_DISABLE_EXITS_CSTATE         (1 << 3)
-#define KVM_X86_DISABLE_VALID_EXITS          (KVM_X86_DISABLE_EXITS_MWAIT | \
-                                              KVM_X86_DISABLE_EXITS_HLT | \
-                                              KVM_X86_DISABLE_EXITS_PAUSE | \
-                                              KVM_X86_DISABLE_EXITS_CSTATE)
 
 /* for KVM_ENABLE_CAP */
 struct kvm_enable_cap {
@@ -933,6 +929,7 @@ struct kvm_enable_cap {
 #define KVM_CAP_PRE_FAULT_MEMORY 236
 #define KVM_CAP_X86_APIC_BUS_CYCLES_NS 237
 #define KVM_CAP_X86_GUEST_MODE 238
+#define KVM_CAP_ARM_WRITABLE_IMP_ID_REGS 239
 
 struct kvm_irq_routing_irqchip {
 	__u32 irqchip;
@@ -1070,6 +1067,10 @@ struct kvm_dirty_tlb {
 
 #define KVM_REG_SIZE_SHIFT	52
 #define KVM_REG_SIZE_MASK	0x00f0000000000000ULL
+
+#define KVM_REG_SIZE(id)		\
+	(1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
+
 #define KVM_REG_SIZE_U8		0x0000000000000000ULL
 #define KVM_REG_SIZE_U16	0x0010000000000000ULL
 #define KVM_REG_SIZE_U32	0x0020000000000000ULL
diff --git a/tools/include/uapi/linux/perf_event.h b/tools/include/uapi/linux/perf_event.h
index 0524d541d4e3..5fc753c23734 100644
--- a/tools/include/uapi/linux/perf_event.h
+++ b/tools/include/uapi/linux/perf_event.h
@@ -385,6 +385,8 @@ enum perf_event_read_format {
  *
  * @sample_max_stack: Max number of frame pointers in a callchain,
  *		      should be < /proc/sys/kernel/perf_event_max_stack
+ *		      Max number of entries of branch stack
+ *		      should be < hardware limit
  */
 struct perf_event_attr {
 
diff --git a/tools/include/uapi/linux/stat.h b/tools/include/uapi/linux/stat.h
index 887a25286441..f78ee3670dd5 100644
--- a/tools/include/uapi/linux/stat.h
+++ b/tools/include/uapi/linux/stat.h
@@ -98,43 +98,93 @@ struct statx_timestamp {
  */
 struct statx {
 	/* 0x00 */
-	__u32	stx_mask;	/* What results were written [uncond] */
-	__u32	stx_blksize;	/* Preferred general I/O size [uncond] */
-	__u64	stx_attributes;	/* Flags conveying information about the file [uncond] */
+	/* What results were written [uncond] */
+	__u32	stx_mask;
+
+	/* Preferred general I/O size [uncond] */
+	__u32	stx_blksize;
+
+	/* Flags conveying information about the file [uncond] */
+	__u64	stx_attributes;
+
 	/* 0x10 */
-	__u32	stx_nlink;	/* Number of hard links */
-	__u32	stx_uid;	/* User ID of owner */
-	__u32	stx_gid;	/* Group ID of owner */
-	__u16	stx_mode;	/* File mode */
+	/* Number of hard links */
+	__u32	stx_nlink;
+
+	/* User ID of owner */
+	__u32	stx_uid;
+
+	/* Group ID of owner */
+	__u32	stx_gid;
+
+	/* File mode */
+	__u16	stx_mode;
 	__u16	__spare0[1];
+
 	/* 0x20 */
-	__u64	stx_ino;	/* Inode number */
-	__u64	stx_size;	/* File size */
-	__u64	stx_blocks;	/* Number of 512-byte blocks allocated */
-	__u64	stx_attributes_mask; /* Mask to show what's supported in stx_attributes */
+	/* Inode number */
+	__u64	stx_ino;
+
+	/* File size */
+	__u64	stx_size;
+
+	/* Number of 512-byte blocks allocated */
+	__u64	stx_blocks;
+
+	/* Mask to show what's supported in stx_attributes */
+	__u64	stx_attributes_mask;
+
 	/* 0x40 */
-	struct statx_timestamp	stx_atime;	/* Last access time */
-	struct statx_timestamp	stx_btime;	/* File creation time */
-	struct statx_timestamp	stx_ctime;	/* Last attribute change time */
-	struct statx_timestamp	stx_mtime;	/* Last data modification time */
+	/* Last access time */
+	struct statx_timestamp	stx_atime;
+
+	/* File creation time */
+	struct statx_timestamp	stx_btime;
+
+	/* Last attribute change time */
+	struct statx_timestamp	stx_ctime;
+
+	/* Last data modification time */
+	struct statx_timestamp	stx_mtime;
+
 	/* 0x80 */
-	__u32	stx_rdev_major;	/* Device ID of special file [if bdev/cdev] */
+	/* Device ID of special file [if bdev/cdev] */
+	__u32	stx_rdev_major;
 	__u32	stx_rdev_minor;
-	__u32	stx_dev_major;	/* ID of device containing file [uncond] */
+
+	/* ID of device containing file [uncond] */
+	__u32	stx_dev_major;
 	__u32	stx_dev_minor;
+
 	/* 0x90 */
 	__u64	stx_mnt_id;
-	__u32	stx_dio_mem_align;	/* Memory buffer alignment for direct I/O */
-	__u32	stx_dio_offset_align;	/* File offset alignment for direct I/O */
+
+	/* Memory buffer alignment for direct I/O */
+	__u32	stx_dio_mem_align;
+
+	/* File offset alignment for direct I/O */
+	__u32	stx_dio_offset_align;
+
 	/* 0xa0 */
-	__u64	stx_subvol;	/* Subvolume identifier */
-	__u32	stx_atomic_write_unit_min;	/* Min atomic write unit in bytes */
-	__u32	stx_atomic_write_unit_max;	/* Max atomic write unit in bytes */
+	/* Subvolume identifier */
+	__u64	stx_subvol;
+
+	/* Min atomic write unit in bytes */
+	__u32	stx_atomic_write_unit_min;
+
+	/* Max atomic write unit in bytes */
+	__u32	stx_atomic_write_unit_max;
+
 	/* 0xb0 */
-	__u32   stx_atomic_write_segments_max;	/* Max atomic write segment count */
-	__u32   __spare1[1];
+	/* Max atomic write segment count */
+	__u32   stx_atomic_write_segments_max;
+
+	/* File offset alignment for direct I/O reads */
+	__u32	stx_dio_read_offset_align;
+
 	/* 0xb8 */
 	__u64	__spare3[9];	/* Spare space for future expansion */
+
 	/* 0x100 */
 };
 
@@ -164,6 +214,7 @@ struct statx {
 #define STATX_MNT_ID_UNIQUE	0x00004000U	/* Want/got extended stx_mount_id */
 #define STATX_SUBVOL		0x00008000U	/* Want/got stx_subvol */
 #define STATX_WRITE_ATOMIC	0x00010000U	/* Want/got atomic_write_* fields */
+#define STATX_DIO_READ_ALIGN	0x00020000U	/* Want/got dio read alignment info */
 
 #define STATX__RESERVED		0x80000000U	/* Reserved for future struct statx expansion */
 
diff --git a/tools/lib/perf/Makefile b/tools/lib/perf/Makefile
index ffcfd777c451..7fbb50b74c00 100644
--- a/tools/lib/perf/Makefile
+++ b/tools/lib/perf/Makefile
@@ -42,6 +42,7 @@ libdir_relative_SQ = $(subst ','\'',$(libdir_relative))
 TEST_ARGS := $(if $(V),-v)
 
 INCLUDES = \
+-I$(OUTPUT)arch/$(SRCARCH)/include/generated/uapi \
 -I$(srctree)/tools/lib/perf/include \
 -I$(srctree)/tools/lib/ \
 -I$(srctree)/tools/include \
@@ -99,7 +100,16 @@ $(LIBAPI)-clean:
 	$(call QUIET_CLEAN, libapi)
 	$(Q)$(MAKE) -C $(LIB_DIR) O=$(OUTPUT) clean >/dev/null
 
-$(LIBPERF_IN): FORCE
+uapi-asm := $(OUTPUT)arch/$(SRCARCH)/include/generated/uapi/asm
+ifeq ($(SRCARCH),arm64)
+	syscall-y := $(uapi-asm)/unistd_64.h
+endif
+uapi-asm-generic:
+	$(if $(syscall-y),\
+		$(Q)$(MAKE) -f $(srctree)/scripts/Makefile.asm-headers obj=$(uapi-asm) \
+		generic=include/uapi/asm-generic $(syscall-y),)
+
+$(LIBPERF_IN): uapi-asm-generic FORCE
 	$(Q)$(MAKE) $(build)=libperf
 
 $(LIBPERF_A): $(LIBPERF_IN)
@@ -120,7 +130,7 @@ all: fixdep
 clean: $(LIBAPI)-clean
 	$(call QUIET_CLEAN, libperf) $(RM) $(LIBPERF_A) \
                 *.o *~ *.a *.so *.so.$(VERSION) *.so.$(LIBPERF_VERSION) .*.d .*.cmd tests/*.o LIBPERF-CFLAGS $(LIBPERF_PC) \
-                $(TESTS_STATIC) $(TESTS_SHARED)
+                $(TESTS_STATIC) $(TESTS_SHARED) $(syscall-y)
 
 TESTS_IN = tests-in.o
 
diff --git a/tools/net/ynl/pyynl/ynl_gen_c.py b/tools/net/ynl/pyynl/ynl_gen_c.py
index a1427c537030..30c0a34b2784 100755
--- a/tools/net/ynl/pyynl/ynl_gen_c.py
+++ b/tools/net/ynl/pyynl/ynl_gen_c.py
@@ -162,9 +162,15 @@ class Type(SpecAttr):
     def free_needs_iter(self):
         return False
 
-    def free(self, ri, var, ref):
+    def _free_lines(self, ri, var, ref):
         if self.is_multi_val() or self.presence_type() == 'len':
-            ri.cw.p(f'free({var}->{ref}{self.c_name});')
+            return [f'free({var}->{ref}{self.c_name});']
+        return []
+
+    def free(self, ri, var, ref):
+        lines = self._free_lines(ri, var, ref)
+        for line in lines:
+            ri.cw.p(line)
 
     def arg_member(self, ri):
         member = self._complex_member_type(ri)
@@ -263,6 +269,10 @@ class Type(SpecAttr):
         var = "req"
         member = f"{var}->{'.'.join(ref)}"
 
+        local_vars = []
+        if self.free_needs_iter():
+            local_vars += ['unsigned int i;']
+
         code = []
         presence = ''
         for i in range(0, len(ref)):
@@ -272,6 +282,10 @@ class Type(SpecAttr):
             if i == len(ref) - 1 and self.presence_type() != 'bit':
                 continue
             code.append(presence + ' = 1;')
+        ref_path = '.'.join(ref[:-1])
+        if ref_path:
+            ref_path += '.'
+        code += self._free_lines(ri, var, ref_path)
         code += self._setter_lines(ri, member, presence)
 
         func_name = f"{op_prefix(ri, direction, deref=deref)}_set_{'_'.join(ref)}"
@@ -279,7 +293,8 @@ class Type(SpecAttr):
         alloc = bool([x for x in code if 'alloc(' in x])
         if free and not alloc:
             func_name = '__' + func_name
-        ri.cw.write_func('static inline void', func_name, body=code,
+        ri.cw.write_func('static inline void', func_name, local_vars=local_vars,
+                         body=code,
                          args=[f'{type_name(ri, direction, deref=deref)} *{var}'] + self.arg_member(ri))
 
 
@@ -482,8 +497,7 @@ class TypeString(Type):
                ['unsigned int len;']
 
     def _setter_lines(self, ri, member, presence):
-        return [f"free({member});",
-                f"{presence}_len = strlen({self.c_name});",
+        return [f"{presence}_len = strlen({self.c_name});",
                 f"{member} = malloc({presence}_len + 1);",
                 f'memcpy({member}, {self.c_name}, {presence}_len);',
                 f'{member}[{presence}_len] = 0;']
@@ -536,8 +550,7 @@ class TypeBinary(Type):
                ['unsigned int len;']
 
     def _setter_lines(self, ri, member, presence):
-        return [f"free({member});",
-                f"{presence}_len = len;",
+        return [f"{presence}_len = len;",
                 f"{member} = malloc({presence}_len);",
                 f'memcpy({member}, {self.c_name}, {presence}_len);']
 
@@ -574,12 +587,14 @@ class TypeNest(Type):
     def _complex_member_type(self, ri):
         return self.nested_struct_type
 
-    def free(self, ri, var, ref):
+    def _free_lines(self, ri, var, ref):
+        lines = []
         at = '&'
         if self.is_recursive_for_op(ri):
             at = ''
-            ri.cw.p(f'if ({var}->{ref}{self.c_name})')
-        ri.cw.p(f'{self.nested_render_name}_free({at}{var}->{ref}{self.c_name});')
+            lines += [f'if ({var}->{ref}{self.c_name})']
+        lines += [f'{self.nested_render_name}_free({at}{var}->{ref}{self.c_name});']
+        return lines
 
     def _attr_typol(self):
         return f'.type = YNL_PT_NEST, .nest = &{self.nested_render_name}_nest, '
@@ -632,15 +647,19 @@ class TypeMultiAttr(Type):
     def free_needs_iter(self):
         return 'type' not in self.attr or self.attr['type'] == 'nest'
 
-    def free(self, ri, var, ref):
+    def _free_lines(self, ri, var, ref):
+        lines = []
         if self.attr['type'] in scalars:
-            ri.cw.p(f"free({var}->{ref}{self.c_name});")
+            lines += [f"free({var}->{ref}{self.c_name});"]
         elif 'type' not in self.attr or self.attr['type'] == 'nest':
-            ri.cw.p(f"for (i = 0; i < {var}->{ref}n_{self.c_name}; i++)")
-            ri.cw.p(f'{self.nested_render_name}_free(&{var}->{ref}{self.c_name}[i]);')
-            ri.cw.p(f"free({var}->{ref}{self.c_name});")
+            lines += [
+                f"for (i = 0; i < {var}->{ref}n_{self.c_name}; i++)",
+                f'{self.nested_render_name}_free(&{var}->{ref}{self.c_name}[i]);',
+                f"free({var}->{ref}{self.c_name});",
+            ]
         else:
             raise Exception(f"Free of MultiAttr sub-type {self.attr['type']} not supported yet")
+        return lines
 
     def _attr_policy(self, policy):
         return self.base_type._attr_policy(policy)
@@ -654,10 +673,10 @@ class TypeMultiAttr(Type):
     def attr_put(self, ri, var):
         if self.attr['type'] in scalars:
             put_type = self.type
-            ri.cw.p(f"for (unsigned int i = 0; i < {var}->n_{self.c_name}; i++)")
+            ri.cw.p(f"for (i = 0; i < {var}->n_{self.c_name}; i++)")
             ri.cw.p(f"ynl_attr_put_{put_type}(nlh, {self.enum_name}, {var}->{self.c_name}[i]);")
         elif 'type' not in self.attr or self.attr['type'] == 'nest':
-            ri.cw.p(f"for (unsigned int i = 0; i < {var}->n_{self.c_name}; i++)")
+            ri.cw.p(f"for (i = 0; i < {var}->n_{self.c_name}; i++)")
             self._attr_put_line(ri, var, f"{self.nested_render_name}_put(nlh, " +
                                 f"{self.enum_name}, &{var}->{self.c_name}[i])")
         else:
@@ -666,8 +685,7 @@ class TypeMultiAttr(Type):
     def _setter_lines(self, ri, member, presence):
         # For multi-attr we have a count, not presence, hack up the presence
         presence = presence[:-(len('_present.') + len(self.c_name))] + "n_" + self.c_name
-        return [f"free({member});",
-                f"{member} = {self.c_name};",
+        return [f"{member} = {self.c_name};",
                 f"{presence} = n_{self.c_name};"]
 
 
@@ -696,8 +714,11 @@ class TypeArrayNest(Type):
     def _attr_get(self, ri, var):
         local_vars = ['const struct nlattr *attr2;']
         get_lines = [f'attr_{self.c_name} = attr;',
-                     'ynl_attr_for_each_nested(attr2, attr)',
-                     f'\t{var}->n_{self.c_name}++;']
+                     'ynl_attr_for_each_nested(attr2, attr) {',
+                     '\tif (ynl_attr_validate(yarg, attr2))',
+                     '\t\treturn YNL_PARSE_CB_ERROR;',
+                     f'\t{var}->n_{self.c_name}++;',
+                     '}']
         return get_lines, None, local_vars
 
 
@@ -755,6 +776,7 @@ class Struct:
         self.request = False
         self.reply = False
         self.recursive = False
+        self.in_multi_val = False  # used by a MultiAttr or and legacy arrays
 
         self.attr_list = []
         self.attrs = dict()
@@ -1122,6 +1144,10 @@ class Family(SpecFamily):
                     if attr in rs_members['reply']:
                         self.pure_nested_structs[nested].reply = True
 
+                if spec.is_multi_val():
+                    child = self.pure_nested_structs.get(nested)
+                    child.in_multi_val = True
+
         self._sort_pure_types()
 
         # Propagate the request / reply / recursive
@@ -1136,6 +1162,8 @@ class Family(SpecFamily):
                             struct.child_nests.update(child.child_nests)
                         child.request |= struct.request
                         child.reply |= struct.reply
+                        if spec.is_multi_val():
+                            child.in_multi_val = True
                 if attr_set in struct.child_nests:
                     struct.recursive = True
 
@@ -1399,9 +1427,9 @@ class CodeWriter:
 
     def write_func(self, qual_ret, name, body, args=None, local_vars=None):
         self.write_func_prot(qual_ret=qual_ret, name=name, args=args)
+        self.block_start()
         self.write_func_lvar(local_vars=local_vars)
 
-        self.block_start()
         for line in body:
             self.p(line)
         self.block_end()
@@ -1644,11 +1672,23 @@ def put_req_nested_prototype(ri, struct, suffix=';'):
 
 
 def put_req_nested(ri, struct):
+    local_vars = []
+    init_lines = []
+
+    local_vars.append('struct nlattr *nest;')
+    init_lines.append("nest = ynl_attr_nest_start(nlh, attr_type);")
+
+    for _, arg in struct.member_list():
+        if arg.presence_type() == 'count':
+            local_vars.append('unsigned int i;')
+            break
+
     put_req_nested_prototype(ri, struct, suffix='')
     ri.cw.block_start()
-    ri.cw.write_func_lvar('struct nlattr *nest;')
+    ri.cw.write_func_lvar(local_vars)
 
-    ri.cw.p("nest = ynl_attr_nest_start(nlh, attr_type);")
+    for line in init_lines:
+        ri.cw.p(line)
 
     for _, arg in struct.member_list():
         arg.attr_put(ri, "obj")
@@ -1850,6 +1890,11 @@ def print_req(ri):
         local_vars += ['size_t hdr_len;',
                        'void *hdr;']
 
+    for _, attr in ri.struct["request"].member_list():
+        if attr.presence_type() == 'count':
+            local_vars += ['unsigned int i;']
+            break
+
     print_prototype(ri, direction, terminate=False)
     ri.cw.block_start()
     ri.cw.write_func_lvar(local_vars)
@@ -2941,6 +2986,9 @@ def main():
             for attr_set, struct in parsed.pure_nested_structs.items():
                 ri = RenderInfo(cw, parsed, args.mode, "", "", attr_set)
                 print_type_full(ri, struct)
+                if struct.request and struct.in_multi_val:
+                    free_rsp_nested_prototype(ri)
+                    cw.nl()
 
             for op_name, op in parsed.ops.items():
                 cw.p(f"/* ============== {op.enum_name} ============== */")
diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c
index 33d861c04ebd..3ce7b54003c2 100644
--- a/tools/objtool/arch/x86/decode.c
+++ b/tools/objtool/arch/x86/decode.c
@@ -522,7 +522,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 			case INAT_PFX_REPNE:
 				if (modrm == 0xca)
 					/* eretu/erets */
-					insn->type = INSN_CONTEXT_SWITCH;
+					insn->type = INSN_SYSRET;
 				break;
 			default:
 				if (modrm == 0xca)
@@ -535,11 +535,15 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 
 			insn->type = INSN_JUMP_CONDITIONAL;
 
-		} else if (op2 == 0x05 || op2 == 0x07 || op2 == 0x34 ||
-			   op2 == 0x35) {
+		} else if (op2 == 0x05 || op2 == 0x34) {
 
-			/* sysenter, sysret */
-			insn->type = INSN_CONTEXT_SWITCH;
+			/* syscall, sysenter */
+			insn->type = INSN_SYSCALL;
+
+		} else if (op2 == 0x07 || op2 == 0x35) {
+
+			/* sysret, sysexit */
+			insn->type = INSN_SYSRET;
 
 		} else if (op2 == 0x0b || op2 == 0xb9) {
 
@@ -676,7 +680,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 
 	case 0xca: /* retf */
 	case 0xcb: /* retf */
-		insn->type = INSN_CONTEXT_SWITCH;
+		insn->type = INSN_SYSRET;
 		break;
 
 	case 0xe0: /* loopne */
@@ -721,7 +725,7 @@ int arch_decode_instruction(struct objtool_file *file, const struct section *sec
 		} else if (modrm_reg == 5) {
 
 			/* jmpf */
-			insn->type = INSN_CONTEXT_SWITCH;
+			insn->type = INSN_SYSRET;
 
 		} else if (modrm_reg == 6) {
 
diff --git a/tools/objtool/arch/x86/special.c b/tools/objtool/arch/x86/special.c
index 403e587676f1..06ca4a2659a4 100644
--- a/tools/objtool/arch/x86/special.c
+++ b/tools/objtool/arch/x86/special.c
@@ -126,7 +126,7 @@ struct reloc *arch_find_switch_table(struct objtool_file *file,
 	 * indicates a rare GCC quirk/bug which can leave dead
 	 * code behind.
 	 */
-	if (reloc_type(text_reloc) == R_X86_64_PC32) {
+	if (!file->ignore_unreachables && reloc_type(text_reloc) == R_X86_64_PC32) {
 		WARN_INSN(insn, "ignoring unreachables due to jump table quirk");
 		file->ignore_unreachables = true;
 	}
diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 4a1f6c3169b3..3a411064fa34 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -225,6 +225,7 @@ static bool is_rust_noreturn(const struct symbol *func)
 	       str_ends_with(func->name, "_4core9panicking14panic_nounwind")				||
 	       str_ends_with(func->name, "_4core9panicking18panic_bounds_check")			||
 	       str_ends_with(func->name, "_4core9panicking19assert_failed_inner")			||
+	       str_ends_with(func->name, "_4core9panicking30panic_null_pointer_dereference")		||
 	       str_ends_with(func->name, "_4core9panicking36panic_misaligned_pointer_dereference")	||
 	       strstr(func->name, "_4core9panicking13assert_failed")					||
 	       strstr(func->name, "_4core9panicking11panic_const24panic_const_")			||
@@ -3505,6 +3506,34 @@ next_orig:
 	return next_insn_same_sec(file, alt_group->orig_group->last_insn);
 }
 
+static bool skip_alt_group(struct instruction *insn)
+{
+	struct instruction *alt_insn = insn->alts ? insn->alts->insn : NULL;
+
+	/* ANNOTATE_IGNORE_ALTERNATIVE */
+	if (insn->alt_group && insn->alt_group->ignore)
+		return true;
+
+	/*
+	 * For NOP patched with CLAC/STAC, only follow the latter to avoid
+	 * impossible code paths combining patched CLAC with unpatched STAC
+	 * or vice versa.
+	 *
+	 * ANNOTATE_IGNORE_ALTERNATIVE could have been used here, but Linus
+	 * requested not to do that to avoid hurting .s file readability
+	 * around CLAC/STAC alternative sites.
+	 */
+
+	if (!alt_insn)
+		return false;
+
+	/* Don't override ASM_{CLAC,STAC}_UNSAFE */
+	if (alt_insn->alt_group && alt_insn->alt_group->ignore)
+		return false;
+
+	return alt_insn->type == INSN_CLAC || alt_insn->type == INSN_STAC;
+}
+
 /*
  * Follow the branch starting at the given instruction, and recursively follow
  * any other branches (jumps).  Meanwhile, track the frame pointer state at
@@ -3625,7 +3654,7 @@ static int validate_branch(struct objtool_file *file, struct symbol *func,
 			}
 		}
 
-		if (insn->alt_group && insn->alt_group->ignore)
+		if (skip_alt_group(insn))
 			return 0;
 
 		if (handle_insn_ops(insn, next_insn, &state))
@@ -3684,14 +3713,20 @@ static int validate_branch(struct objtool_file *file, struct symbol *func,
 
 			break;
 
-		case INSN_CONTEXT_SWITCH:
-			if (func) {
-				if (!next_insn || !next_insn->hint) {
-					WARN_INSN(insn, "unsupported instruction in callable function");
-					return 1;
-				}
-				break;
+		case INSN_SYSCALL:
+			if (func && (!next_insn || !next_insn->hint)) {
+				WARN_INSN(insn, "unsupported instruction in callable function");
+				return 1;
+			}
+
+			break;
+
+		case INSN_SYSRET:
+			if (func && (!next_insn || !next_insn->hint)) {
+				WARN_INSN(insn, "unsupported instruction in callable function");
+				return 1;
 			}
+
 			return 0;
 
 		case INSN_STAC:
@@ -3886,6 +3921,12 @@ static int validate_unret(struct objtool_file *file, struct instruction *insn)
 			WARN_INSN(insn, "RET before UNTRAIN");
 			return 1;
 
+		case INSN_SYSCALL:
+			break;
+
+		case INSN_SYSRET:
+			return 0;
+
 		case INSN_NOP:
 			if (insn->retpoline_safe)
 				return 0;
@@ -3895,6 +3936,9 @@ static int validate_unret(struct objtool_file *file, struct instruction *insn)
 			break;
 		}
 
+		if (insn->dead_end)
+			return 0;
+
 		if (!next) {
 			WARN_INSN(insn, "teh end!");
 			return 1;
diff --git a/tools/objtool/include/objtool/arch.h b/tools/objtool/include/objtool/arch.h
index 089a1acc48a8..01ef6f415adf 100644
--- a/tools/objtool/include/objtool/arch.h
+++ b/tools/objtool/include/objtool/arch.h
@@ -19,7 +19,8 @@ enum insn_type {
 	INSN_CALL,
 	INSN_CALL_DYNAMIC,
 	INSN_RETURN,
-	INSN_CONTEXT_SWITCH,
+	INSN_SYSCALL,
+	INSN_SYSRET,
 	INSN_BUG,
 	INSN_NOP,
 	INSN_STAC,
diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config
index eea95c6c0c71..b7769a22fe1a 100644
--- a/tools/perf/Makefile.config
+++ b/tools/perf/Makefile.config
@@ -29,6 +29,7 @@ include $(srctree)/tools/scripts/Makefile.arch
 $(call detected_var,SRCARCH)
 
 CFLAGS += -I$(OUTPUT)arch/$(SRCARCH)/include/generated
+CFLAGS += -I$(OUTPUT)libperf/arch/$(SRCARCH)/include/generated/uapi
 
 # Additional ARCH settings for ppc
 ifeq ($(SRCARCH),powerpc)
diff --git a/tools/perf/arch/arm/entry/syscalls/syscall.tbl b/tools/perf/arch/arm/entry/syscalls/syscall.tbl
index 49eeb2ad8dbd..27c1d5ebcd91 100644
--- a/tools/perf/arch/arm/entry/syscalls/syscall.tbl
+++ b/tools/perf/arch/arm/entry/syscalls/syscall.tbl
@@ -481,3 +481,4 @@
 464	common	getxattrat			sys_getxattrat
 465	common	listxattrat			sys_listxattrat
 466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
diff --git a/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl b/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl
index c844cd5cda62..1e8c44c7b614 100644
--- a/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl
+++ b/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl
@@ -381,3 +381,4 @@
 464	n64	getxattrat			sys_getxattrat
 465	n64	listxattrat			sys_listxattrat
 466	n64	removexattrat			sys_removexattrat
+467	n64	open_tree_attr			sys_open_tree_attr
diff --git a/tools/perf/arch/powerpc/entry/syscalls/syscall.tbl b/tools/perf/arch/powerpc/entry/syscalls/syscall.tbl
index d8b4ab78bef0..9a084bdb8926 100644
--- a/tools/perf/arch/powerpc/entry/syscalls/syscall.tbl
+++ b/tools/perf/arch/powerpc/entry/syscalls/syscall.tbl
@@ -557,3 +557,4 @@
 464	common	getxattrat			sys_getxattrat
 465	common	listxattrat			sys_listxattrat
 466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
diff --git a/tools/perf/arch/s390/entry/syscalls/syscall.tbl b/tools/perf/arch/s390/entry/syscalls/syscall.tbl
index e9115b4d8b63..a4569b96ef06 100644
--- a/tools/perf/arch/s390/entry/syscalls/syscall.tbl
+++ b/tools/perf/arch/s390/entry/syscalls/syscall.tbl
@@ -469,3 +469,4 @@
 464  common	getxattrat		sys_getxattrat			sys_getxattrat
 465  common	listxattrat		sys_listxattrat			sys_listxattrat
 466  common	removexattrat		sys_removexattrat		sys_removexattrat
+467  common	open_tree_attr		sys_open_tree_attr		sys_open_tree_attr
diff --git a/tools/perf/arch/sh/entry/syscalls/syscall.tbl b/tools/perf/arch/sh/entry/syscalls/syscall.tbl
index c8cad33bf250..52a7652fcff6 100644
--- a/tools/perf/arch/sh/entry/syscalls/syscall.tbl
+++ b/tools/perf/arch/sh/entry/syscalls/syscall.tbl
@@ -470,3 +470,4 @@
 464	common	getxattrat			sys_getxattrat
 465	common	listxattrat			sys_listxattrat
 466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
diff --git a/tools/perf/arch/sparc/entry/syscalls/syscall.tbl b/tools/perf/arch/sparc/entry/syscalls/syscall.tbl
index 727f99d333b3..83e45eb6c095 100644
--- a/tools/perf/arch/sparc/entry/syscalls/syscall.tbl
+++ b/tools/perf/arch/sparc/entry/syscalls/syscall.tbl
@@ -512,3 +512,4 @@
 464	common	getxattrat			sys_getxattrat
 465	common	listxattrat			sys_listxattrat
 466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
diff --git a/tools/perf/arch/x86/entry/syscalls/syscall_32.tbl b/tools/perf/arch/x86/entry/syscalls/syscall_32.tbl
index 4d0fb2fba7e2..ac007ea00979 100644
--- a/tools/perf/arch/x86/entry/syscalls/syscall_32.tbl
+++ b/tools/perf/arch/x86/entry/syscalls/syscall_32.tbl
@@ -396,7 +396,7 @@
 381	i386	pkey_alloc		sys_pkey_alloc
 382	i386	pkey_free		sys_pkey_free
 383	i386	statx			sys_statx
-384	i386	arch_prctl		sys_arch_prctl			compat_sys_arch_prctl
+384	i386	arch_prctl		sys_arch_prctl
 385	i386	io_pgetevents		sys_io_pgetevents_time32	compat_sys_io_pgetevents
 386	i386	rseq			sys_rseq
 393	i386	semget			sys_semget
@@ -472,3 +472,4 @@
 464	i386	getxattrat		sys_getxattrat
 465	i386	listxattrat		sys_listxattrat
 466	i386	removexattrat		sys_removexattrat
+467	i386	open_tree_attr		sys_open_tree_attr
diff --git a/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl b/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
index 5eb708bff1c7..cfb5ca41e30d 100644
--- a/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
+++ b/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
@@ -390,6 +390,7 @@
 464	common	getxattrat		sys_getxattrat
 465	common	listxattrat		sys_listxattrat
 466	common	removexattrat		sys_removexattrat
+467	common	open_tree_attr		sys_open_tree_attr
 
 #
 # Due to a historical design error, certain syscalls are numbered differently
diff --git a/tools/perf/arch/xtensa/entry/syscalls/syscall.tbl b/tools/perf/arch/xtensa/entry/syscalls/syscall.tbl
index 37effc1b134e..f657a77314f8 100644
--- a/tools/perf/arch/xtensa/entry/syscalls/syscall.tbl
+++ b/tools/perf/arch/xtensa/entry/syscalls/syscall.tbl
@@ -437,3 +437,4 @@
 464	common	getxattrat			sys_getxattrat
 465	common	listxattrat			sys_listxattrat
 466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
diff --git a/tools/perf/check-headers.sh b/tools/perf/check-headers.sh
index a4499e5a6f9c..857f6646cc23 100755
--- a/tools/perf/check-headers.sh
+++ b/tools/perf/check-headers.sh
@@ -20,6 +20,7 @@ FILES=(
   "include/uapi/linux/stat.h"
   "include/linux/bits.h"
   "include/vdso/bits.h"
+  "include/linux/cfi_types.h"
   "include/linux/const.h"
   "include/vdso/const.h"
   "include/vdso/unaligned.h"
diff --git a/tools/perf/trace/beauty/include/linux/socket.h b/tools/perf/trace/beauty/include/linux/socket.h
index d18cc47e89bd..c3322eb3d686 100644
--- a/tools/perf/trace/beauty/include/linux/socket.h
+++ b/tools/perf/trace/beauty/include/linux/socket.h
@@ -392,6 +392,8 @@ struct ucred {
 
 extern int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr_storage *kaddr);
 extern int put_cmsg(struct msghdr*, int level, int type, int len, void *data);
+extern int put_cmsg_notrunc(struct msghdr *msg, int level, int type, int len,
+			    void *data);
 
 struct timespec64;
 struct __kernel_timespec;
diff --git a/tools/perf/trace/beauty/include/uapi/linux/fcntl.h b/tools/perf/trace/beauty/include/uapi/linux/fcntl.h
index 6e6907e63bfc..a15ac2fa4b20 100644
--- a/tools/perf/trace/beauty/include/uapi/linux/fcntl.h
+++ b/tools/perf/trace/beauty/include/uapi/linux/fcntl.h
@@ -155,4 +155,8 @@
 #define AT_HANDLE_MNT_ID_UNIQUE	0x001	/* Return the u64 unique mount ID. */
 #define AT_HANDLE_CONNECTABLE	0x002	/* Request a connectable file handle */
 
+/* Flags for execveat2(2). */
+#define AT_EXECVE_CHECK		0x10000	/* Only perform a check if execution
+					   would be allowed. */
+
 #endif /* _UAPI_LINUX_FCNTL_H */
diff --git a/tools/perf/trace/beauty/include/uapi/linux/fs.h b/tools/perf/trace/beauty/include/uapi/linux/fs.h
index 753971770733..e762e1af650c 100644
--- a/tools/perf/trace/beauty/include/uapi/linux/fs.h
+++ b/tools/perf/trace/beauty/include/uapi/linux/fs.h
@@ -40,6 +40,15 @@
 #define BLOCK_SIZE_BITS 10
 #define BLOCK_SIZE (1<<BLOCK_SIZE_BITS)
 
+/* flags for integrity meta */
+#define IO_INTEGRITY_CHK_GUARD		(1U << 0) /* enforce guard check */
+#define IO_INTEGRITY_CHK_REFTAG		(1U << 1) /* enforce ref check */
+#define IO_INTEGRITY_CHK_APPTAG		(1U << 2) /* enforce app check */
+
+#define IO_INTEGRITY_VALID_FLAGS (IO_INTEGRITY_CHK_GUARD | \
+				  IO_INTEGRITY_CHK_REFTAG | \
+				  IO_INTEGRITY_CHK_APPTAG)
+
 #define SEEK_SET	0	/* seek relative to beginning of file */
 #define SEEK_CUR	1	/* seek relative to current file position */
 #define SEEK_END	2	/* seek relative to end of file */
@@ -203,10 +212,8 @@ struct fsxattr {
 #define BLKROTATIONAL _IO(0x12,126)
 #define BLKZEROOUT _IO(0x12,127)
 #define BLKGETDISKSEQ _IOR(0x12,128,__u64)
-/*
- * A jump here: 130-136 are reserved for zoned block devices
- * (see uapi/linux/blkzoned.h)
- */
+/* 130-136 are used by zoned block device ioctls (uapi/linux/blkzoned.h) */
+/* 137-141 are used by blk-crypto ioctls (uapi/linux/blk-crypto.h) */
 
 #define BMAP_IOCTL 1		/* obsolete - kept for compatibility */
 #define FIBMAP	   _IO(0x00,1)	/* bmap access */
@@ -332,9 +339,13 @@ typedef int __bitwise __kernel_rwf_t;
 /* Atomic Write */
 #define RWF_ATOMIC	((__force __kernel_rwf_t)0x00000040)
 
+/* buffered IO that drops the cache after reading or writing data */
+#define RWF_DONTCACHE	((__force __kernel_rwf_t)0x00000080)
+
 /* mask of flags supported by the kernel */
 #define RWF_SUPPORTED	(RWF_HIPRI | RWF_DSYNC | RWF_SYNC | RWF_NOWAIT |\
-			 RWF_APPEND | RWF_NOAPPEND | RWF_ATOMIC)
+			 RWF_APPEND | RWF_NOAPPEND | RWF_ATOMIC |\
+			 RWF_DONTCACHE)
 
 #define PROCFS_IOCTL_MAGIC 'f'
 
diff --git a/tools/perf/trace/beauty/include/uapi/linux/mount.h b/tools/perf/trace/beauty/include/uapi/linux/mount.h
index c07008816aca..7fa67c2031a5 100644
--- a/tools/perf/trace/beauty/include/uapi/linux/mount.h
+++ b/tools/perf/trace/beauty/include/uapi/linux/mount.h
@@ -179,7 +179,12 @@ struct statmount {
 	__u32 opt_array;	/* [str] Array of nul terminated fs options */
 	__u32 opt_sec_num;	/* Number of security options */
 	__u32 opt_sec_array;	/* [str] Array of nul terminated security options */
-	__u64 __spare2[46];
+	__u64 supported_mask;	/* Mask flags that this kernel supports */
+	__u32 mnt_uidmap_num;	/* Number of uid mappings */
+	__u32 mnt_uidmap;	/* [str] Array of uid mappings (as seen from callers namespace) */
+	__u32 mnt_gidmap_num;	/* Number of gid mappings */
+	__u32 mnt_gidmap;	/* [str] Array of gid mappings (as seen from callers namespace) */
+	__u64 __spare2[43];
 	char str[];		/* Variable size part containing strings */
 };
 
@@ -217,6 +222,9 @@ struct mnt_id_req {
 #define STATMOUNT_SB_SOURCE		0x00000200U	/* Want/got sb_source */
 #define STATMOUNT_OPT_ARRAY		0x00000400U	/* Want/got opt_... */
 #define STATMOUNT_OPT_SEC_ARRAY		0x00000800U	/* Want/got opt_sec... */
+#define STATMOUNT_SUPPORTED_MASK	0x00001000U	/* Want/got supported mask flags */
+#define STATMOUNT_MNT_UIDMAP		0x00002000U	/* Want/got uidmap... */
+#define STATMOUNT_MNT_GIDMAP		0x00004000U	/* Want/got gidmap... */
 
 /*
  * Special @mnt_id values that can be passed to listmount
diff --git a/tools/perf/trace/beauty/include/uapi/linux/prctl.h b/tools/perf/trace/beauty/include/uapi/linux/prctl.h
index 5c6080680cb2..15c18ef4eb11 100644
--- a/tools/perf/trace/beauty/include/uapi/linux/prctl.h
+++ b/tools/perf/trace/beauty/include/uapi/linux/prctl.h
@@ -353,4 +353,15 @@ struct prctl_mm_map {
  */
 #define PR_LOCK_SHADOW_STACK_STATUS      76
 
+/*
+ * Controls the mode of timer_create() for CRIU restore operations.
+ * Enabling this allows CRIU to restore timers with explicit IDs.
+ *
+ * Don't use for normal operations as the result might be undefined.
+ */
+#define PR_TIMER_CREATE_RESTORE_IDS		77
+# define PR_TIMER_CREATE_RESTORE_IDS_OFF	0
+# define PR_TIMER_CREATE_RESTORE_IDS_ON		1
+# define PR_TIMER_CREATE_RESTORE_IDS_GET	2
+
 #endif /* _LINUX_PRCTL_H */
diff --git a/tools/perf/trace/beauty/include/uapi/linux/stat.h b/tools/perf/trace/beauty/include/uapi/linux/stat.h
index 887a25286441..f78ee3670dd5 100644
--- a/tools/perf/trace/beauty/include/uapi/linux/stat.h
+++ b/tools/perf/trace/beauty/include/uapi/linux/stat.h
@@ -98,43 +98,93 @@ struct statx_timestamp {
  */
 struct statx {
 	/* 0x00 */
-	__u32	stx_mask;	/* What results were written [uncond] */
-	__u32	stx_blksize;	/* Preferred general I/O size [uncond] */
-	__u64	stx_attributes;	/* Flags conveying information about the file [uncond] */
+	/* What results were written [uncond] */
+	__u32	stx_mask;
+
+	/* Preferred general I/O size [uncond] */
+	__u32	stx_blksize;
+
+	/* Flags conveying information about the file [uncond] */
+	__u64	stx_attributes;
+
 	/* 0x10 */
-	__u32	stx_nlink;	/* Number of hard links */
-	__u32	stx_uid;	/* User ID of owner */
-	__u32	stx_gid;	/* Group ID of owner */
-	__u16	stx_mode;	/* File mode */
+	/* Number of hard links */
+	__u32	stx_nlink;
+
+	/* User ID of owner */
+	__u32	stx_uid;
+
+	/* Group ID of owner */
+	__u32	stx_gid;
+
+	/* File mode */
+	__u16	stx_mode;
 	__u16	__spare0[1];
+
 	/* 0x20 */
-	__u64	stx_ino;	/* Inode number */
-	__u64	stx_size;	/* File size */
-	__u64	stx_blocks;	/* Number of 512-byte blocks allocated */
-	__u64	stx_attributes_mask; /* Mask to show what's supported in stx_attributes */
+	/* Inode number */
+	__u64	stx_ino;
+
+	/* File size */
+	__u64	stx_size;
+
+	/* Number of 512-byte blocks allocated */
+	__u64	stx_blocks;
+
+	/* Mask to show what's supported in stx_attributes */
+	__u64	stx_attributes_mask;
+
 	/* 0x40 */
-	struct statx_timestamp	stx_atime;	/* Last access time */
-	struct statx_timestamp	stx_btime;	/* File creation time */
-	struct statx_timestamp	stx_ctime;	/* Last attribute change time */
-	struct statx_timestamp	stx_mtime;	/* Last data modification time */
+	/* Last access time */
+	struct statx_timestamp	stx_atime;
+
+	/* File creation time */
+	struct statx_timestamp	stx_btime;
+
+	/* Last attribute change time */
+	struct statx_timestamp	stx_ctime;
+
+	/* Last data modification time */
+	struct statx_timestamp	stx_mtime;
+
 	/* 0x80 */
-	__u32	stx_rdev_major;	/* Device ID of special file [if bdev/cdev] */
+	/* Device ID of special file [if bdev/cdev] */
+	__u32	stx_rdev_major;
 	__u32	stx_rdev_minor;
-	__u32	stx_dev_major;	/* ID of device containing file [uncond] */
+
+	/* ID of device containing file [uncond] */
+	__u32	stx_dev_major;
 	__u32	stx_dev_minor;
+
 	/* 0x90 */
 	__u64	stx_mnt_id;
-	__u32	stx_dio_mem_align;	/* Memory buffer alignment for direct I/O */
-	__u32	stx_dio_offset_align;	/* File offset alignment for direct I/O */
+
+	/* Memory buffer alignment for direct I/O */
+	__u32	stx_dio_mem_align;
+
+	/* File offset alignment for direct I/O */
+	__u32	stx_dio_offset_align;
+
 	/* 0xa0 */
-	__u64	stx_subvol;	/* Subvolume identifier */
-	__u32	stx_atomic_write_unit_min;	/* Min atomic write unit in bytes */
-	__u32	stx_atomic_write_unit_max;	/* Max atomic write unit in bytes */
+	/* Subvolume identifier */
+	__u64	stx_subvol;
+
+	/* Min atomic write unit in bytes */
+	__u32	stx_atomic_write_unit_min;
+
+	/* Max atomic write unit in bytes */
+	__u32	stx_atomic_write_unit_max;
+
 	/* 0xb0 */
-	__u32   stx_atomic_write_segments_max;	/* Max atomic write segment count */
-	__u32   __spare1[1];
+	/* Max atomic write segment count */
+	__u32   stx_atomic_write_segments_max;
+
+	/* File offset alignment for direct I/O reads */
+	__u32	stx_dio_read_offset_align;
+
 	/* 0xb8 */
 	__u64	__spare3[9];	/* Spare space for future expansion */
+
 	/* 0x100 */
 };
 
@@ -164,6 +214,7 @@ struct statx {
 #define STATX_MNT_ID_UNIQUE	0x00004000U	/* Want/got extended stx_mount_id */
 #define STATX_SUBVOL		0x00008000U	/* Want/got stx_subvol */
 #define STATX_WRITE_ATOMIC	0x00010000U	/* Want/got atomic_write_* fields */
+#define STATX_DIO_READ_ALIGN	0x00020000U	/* Want/got dio read alignment info */
 
 #define STATX__RESERVED		0x80000000U	/* Reserved for future struct statx expansion */
 
diff --git a/tools/perf/trace/beauty/include/uapi/sound/asound.h b/tools/perf/trace/beauty/include/uapi/sound/asound.h
index 4cd513215bcd..5a049eeaecce 100644
--- a/tools/perf/trace/beauty/include/uapi/sound/asound.h
+++ b/tools/perf/trace/beauty/include/uapi/sound/asound.h
@@ -716,7 +716,7 @@ enum {
  *  Raw MIDI section - /dev/snd/midi??
  */
 
-#define SNDRV_RAWMIDI_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 4)
+#define SNDRV_RAWMIDI_VERSION		SNDRV_PROTOCOL_VERSION(2, 0, 5)
 
 enum {
 	SNDRV_RAWMIDI_STREAM_OUTPUT = 0,
@@ -728,6 +728,9 @@ enum {
 #define SNDRV_RAWMIDI_INFO_INPUT		0x00000002
 #define SNDRV_RAWMIDI_INFO_DUPLEX		0x00000004
 #define SNDRV_RAWMIDI_INFO_UMP			0x00000008
+#define SNDRV_RAWMIDI_INFO_STREAM_INACTIVE	0x00000010
+
+#define SNDRV_RAWMIDI_DEVICE_UNKNOWN		0
 
 struct snd_rawmidi_info {
 	unsigned int device;		/* RO/WR (control): device number */
@@ -740,7 +743,8 @@ struct snd_rawmidi_info {
 	unsigned char subname[32];	/* name of active or selected subdevice */
 	unsigned int subdevices_count;
 	unsigned int subdevices_avail;
-	unsigned char reserved[64];	/* reserved for future use */
+	int tied_device;		/* R: tied rawmidi device (UMP/legacy) */
+	unsigned char reserved[60];	/* reserved for future use */
 };
 
 #define SNDRV_RAWMIDI_MODE_FRAMING_MASK		(7<<0)
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 1974395492d7..3c030da2e477 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -2566,25 +2566,6 @@ check:
 	return false;
 }
 
-static bool evsel__handle_error_quirks(struct evsel *evsel, int error)
-{
-	/*
-	 * AMD core PMU tries to forward events with precise_ip to IBS PMU
-	 * implicitly.  But IBS PMU has more restrictions so it can fail with
-	 * supported event attributes.  Let's forward it back to the core PMU
-	 * by clearing precise_ip only if it's from precise_max (:P).
-	 */
-	if ((error == -EINVAL || error == -ENOENT) && x86__is_amd_cpu() &&
-	    evsel->core.attr.precise_ip && evsel->precise_max) {
-		evsel->core.attr.precise_ip = 0;
-		pr_debug2_peo("removing precise_ip on AMD\n");
-		display_attr(&evsel->core.attr);
-		return true;
-	}
-
-	return false;
-}
-
 static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus,
 		struct perf_thread_map *threads,
 		int start_cpu_map_idx, int end_cpu_map_idx)
@@ -2730,9 +2711,6 @@ try_fallback:
 	if (evsel__precise_ip_fallback(evsel))
 		goto retry_open;
 
-	if (evsel__handle_error_quirks(evsel, err))
-		goto retry_open;
-
 out_close:
 	if (err)
 		threads->err_thread = thread;
diff --git a/tools/perf/util/unwind-libunwind-local.c b/tools/perf/util/unwind-libunwind-local.c
index 9fb2c1343c7f..0b037e7389a0 100644
--- a/tools/perf/util/unwind-libunwind-local.c
+++ b/tools/perf/util/unwind-libunwind-local.c
@@ -371,7 +371,7 @@ static int read_unwind_spec_debug_frame(struct dso *dso,
 	 *    has to be pointed by symsrc_filename
 	 */
 	if (ofs == 0) {
-		if (dso__data_get_fd(dso, machine, &fd) {
+		if (dso__data_get_fd(dso, machine, &fd)) {
 			ofs = elf_section_offset(fd, ".debug_frame");
 			dso__data_put_fd(dso);
 		}
diff --git a/tools/sched_ext/scx_flatcg.bpf.c b/tools/sched_ext/scx_flatcg.bpf.c
index 2c720e3ecad5..fdc7170639e6 100644
--- a/tools/sched_ext/scx_flatcg.bpf.c
+++ b/tools/sched_ext/scx_flatcg.bpf.c
@@ -950,5 +950,5 @@ SCX_OPS_DEFINE(flatcg_ops,
 	       .cgroup_move		= (void *)fcg_cgroup_move,
 	       .init			= (void *)fcg_init,
 	       .exit			= (void *)fcg_exit,
-	       .flags			= SCX_OPS_HAS_CGROUP_WEIGHT | SCX_OPS_ENQ_EXITING,
+	       .flags			= SCX_OPS_ENQ_EXITING,
 	       .name			= "flatcg");
diff --git a/tools/scripts/syscall.tbl b/tools/scripts/syscall.tbl
index ebbdb3c42e9f..580b4e246aec 100644
--- a/tools/scripts/syscall.tbl
+++ b/tools/scripts/syscall.tbl
@@ -407,3 +407,4 @@
 464	common	getxattrat			sys_getxattrat
 465	common	listxattrat			sys_listxattrat
 466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
diff --git a/tools/testing/cxl/test/mem.c b/tools/testing/cxl/test/mem.c
index f2957a3e36fe..bf9caa908f89 100644
--- a/tools/testing/cxl/test/mem.c
+++ b/tools/testing/cxl/test/mem.c
@@ -1780,7 +1780,7 @@ static int cxl_mock_mem_probe(struct platform_device *pdev)
 	if (rc)
 		return rc;
 
-	rc = devm_cxl_setup_fwctl(cxlmd);
+	rc = devm_cxl_setup_fwctl(&pdev->dev, cxlmd);
 	if (rc)
 		dev_dbg(dev, "No CXL FWCTL setup\n");
 
diff --git a/tools/testing/kunit/configs/all_tests.config b/tools/testing/kunit/configs/all_tests.config
index cdd9782f9646..422e186cf3cf 100644
--- a/tools/testing/kunit/configs/all_tests.config
+++ b/tools/testing/kunit/configs/all_tests.config
@@ -20,6 +20,7 @@ CONFIG_VFAT_FS=y
 
 CONFIG_PCI=y
 CONFIG_USB4=y
+CONFIG_I2C=y
 
 CONFIG_NET=y
 CONFIG_MCTP=y
@@ -43,6 +44,8 @@ CONFIG_REGMAP_BUILD=y
 
 CONFIG_AUDIT=y
 
+CONFIG_PRIME_NUMBERS=y
+
 CONFIG_SECURITY=y
 CONFIG_SECURITY_APPARMOR=y
 CONFIG_SECURITY_LANDLOCK=y
@@ -51,3 +54,4 @@ CONFIG_SOUND=y
 CONFIG_SND=y
 CONFIG_SND_SOC=y
 CONFIG_SND_SOC_TOPOLOGY_BUILD=y
+CONFIG_SND_SOC_CS35L56_I2C=y
diff --git a/tools/testing/kunit/kunit_parser.py b/tools/testing/kunit/kunit_parser.py
index da53a709773a..c176487356e6 100644
--- a/tools/testing/kunit/kunit_parser.py
+++ b/tools/testing/kunit/kunit_parser.py
@@ -809,6 +809,10 @@ def parse_test(lines: LineStream, expected_num: int, log: List[str], is_subtest:
 		test.log.extend(parse_diagnostic(lines))
 		if test.name != "" and not peek_test_name_match(lines, test):
 			test.add_error(printer, 'missing subtest result line!')
+		elif not lines:
+			print_log(test.log, printer)
+			test.status = TestStatus.NO_TESTS
+			test.add_error(printer, 'No more test results!')
 		else:
 			parse_test_result(lines, test, expected_num, printer)
 
diff --git a/tools/testing/kunit/kunit_tool_test.py b/tools/testing/kunit/kunit_tool_test.py
index 5ff4f6ffd873..bbba921e0eac 100755
--- a/tools/testing/kunit/kunit_tool_test.py
+++ b/tools/testing/kunit/kunit_tool_test.py
@@ -371,8 +371,8 @@ class KUnitParserTest(unittest.TestCase):
 		"""
 		result = kunit_parser.parse_run_tests(output.splitlines(), stdout)
 		# Missing test results after test plan should alert a suspected test crash.
-		self.assertEqual(kunit_parser.TestStatus.TEST_CRASHED, result.status)
-		self.assertEqual(result.counts, kunit_parser.TestCounts(passed=1, crashed=1, errors=1))
+		self.assertEqual(kunit_parser.TestStatus.SUCCESS, result.status)
+		self.assertEqual(result.counts, kunit_parser.TestCounts(passed=1, errors=2))
 
 def line_stream_from_strs(strs: Iterable[str]) -> kunit_parser.LineStream:
 	return kunit_parser.LineStream(enumerate(strs, start=1))
diff --git a/tools/testing/kunit/qemu_configs/sh.py b/tools/testing/kunit/qemu_configs/sh.py
index 78a474a5b95f..f00cb89fdef6 100644
--- a/tools/testing/kunit/qemu_configs/sh.py
+++ b/tools/testing/kunit/qemu_configs/sh.py
@@ -7,7 +7,9 @@ CONFIG_CPU_SUBTYPE_SH7751R=y
 CONFIG_MEMORY_START=0x0c000000
 CONFIG_SH_RTS7751R2D=y
 CONFIG_RTS7751R2D_PLUS=y
-CONFIG_SERIAL_SH_SCI=y''',
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_CMDLINE_EXTEND=y
+''',
 			   qemu_arch='sh4',
 			   kernel_path='arch/sh/boot/zImage',
 			   kernel_command_line='console=ttySC1',
diff --git a/tools/testing/memblock/internal.h b/tools/testing/memblock/internal.h
index 1cf82acb2a3e..0ab4b53bb4f3 100644
--- a/tools/testing/memblock/internal.h
+++ b/tools/testing/memblock/internal.h
@@ -24,4 +24,10 @@ static inline void accept_memory(phys_addr_t start, unsigned long size)
 {
 }
 
+static inline unsigned long free_reserved_area(void *start, void *end,
+					       int poison, const char *s)
+{
+	return 0;
+}
+
 #endif
diff --git a/tools/testing/memblock/linux/mutex.h b/tools/testing/memblock/linux/mutex.h
new file mode 100644
index 000000000000..ae3f497165d6
--- /dev/null
+++ b/tools/testing/memblock/linux/mutex.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _MUTEX_H
+#define _MUTEX_H
+
+#define DEFINE_MUTEX(name) int name
+
+static inline void dummy_mutex_guard(int *name)
+{
+}
+
+#define guard(mutex)	\
+	dummy_##mutex##_guard
+
+#endif /* _MUTEX_H */
\ No newline at end of file
diff --git a/tools/testing/memblock/tests/basic_api.c b/tools/testing/memblock/tests/basic_api.c
index 67503089e6a0..01e836fba488 100644
--- a/tools/testing/memblock/tests/basic_api.c
+++ b/tools/testing/memblock/tests/basic_api.c
@@ -2434,6 +2434,107 @@ static int memblock_overlaps_region_checks(void)
 	return 0;
 }
 
+#ifdef CONFIG_NUMA
+static int memblock_set_node_check(void)
+{
+	unsigned long i, max_reserved;
+	struct memblock_region *rgn;
+	void *orig_region;
+
+	PREFIX_PUSH();
+
+	reset_memblock_regions();
+	memblock_allow_resize();
+
+	dummy_physical_memory_init();
+	memblock_add(dummy_physical_memory_base(), MEM_SIZE);
+	orig_region = memblock.reserved.regions;
+
+	/* Equally Split range to node 0 and 1*/
+	memblock_set_node(memblock_start_of_DRAM(),
+			memblock_phys_mem_size() / 2, &memblock.memory, 0);
+	memblock_set_node(memblock_start_of_DRAM() + memblock_phys_mem_size() / 2,
+			memblock_phys_mem_size() / 2, &memblock.memory, 1);
+
+	ASSERT_EQ(memblock.memory.cnt, 2);
+	rgn = &memblock.memory.regions[0];
+	ASSERT_EQ(rgn->base, memblock_start_of_DRAM());
+	ASSERT_EQ(rgn->size, memblock_phys_mem_size() / 2);
+	ASSERT_EQ(memblock_get_region_node(rgn), 0);
+	rgn = &memblock.memory.regions[1];
+	ASSERT_EQ(rgn->base, memblock_start_of_DRAM() + memblock_phys_mem_size() / 2);
+	ASSERT_EQ(rgn->size, memblock_phys_mem_size() / 2);
+	ASSERT_EQ(memblock_get_region_node(rgn), 1);
+
+	/* Reserve 126 regions with the last one across node boundary */
+	for (i = 0; i < 125; i++)
+		memblock_reserve(memblock_start_of_DRAM() + SZ_16 * i, SZ_8);
+
+	memblock_reserve(memblock_start_of_DRAM() + memblock_phys_mem_size() / 2 - SZ_8,
+			SZ_16);
+
+	/*
+	 * Commit 61167ad5fecd ("mm: pass nid to reserve_bootmem_region()")
+	 * do following process to set nid to each memblock.reserved region.
+	 * But it may miss some region if memblock_set_node() double the
+	 * array.
+	 *
+	 * By checking 'max', we make sure all region nid is set properly.
+	 */
+repeat:
+	max_reserved = memblock.reserved.max;
+	for_each_mem_region(rgn) {
+		int nid = memblock_get_region_node(rgn);
+
+		memblock_set_node(rgn->base, rgn->size, &memblock.reserved, nid);
+	}
+	if (max_reserved != memblock.reserved.max)
+		goto repeat;
+
+	/* Confirm each region has valid node set */
+	for_each_reserved_mem_region(rgn) {
+		ASSERT_TRUE(numa_valid_node(memblock_get_region_node(rgn)));
+		if (rgn == (memblock.reserved.regions + memblock.reserved.cnt - 1))
+			ASSERT_EQ(1, memblock_get_region_node(rgn));
+		else
+			ASSERT_EQ(0, memblock_get_region_node(rgn));
+	}
+
+	dummy_physical_memory_cleanup();
+
+	/*
+	 * The current reserved.regions is occupying a range of memory that
+	 * allocated from dummy_physical_memory_init(). After free the memory,
+	 * we must not use it. So restore the origin memory region to make sure
+	 * the tests can run as normal and not affected by the double array.
+	 */
+	memblock.reserved.regions = orig_region;
+	memblock.reserved.cnt = INIT_MEMBLOCK_RESERVED_REGIONS;
+
+	test_pass_pop();
+
+	return 0;
+}
+
+static int memblock_set_node_checks(void)
+{
+	prefix_reset();
+	prefix_push("memblock_set_node");
+	test_print("Running memblock_set_node tests...\n");
+
+	memblock_set_node_check();
+
+	prefix_pop();
+
+	return 0;
+}
+#else
+static int memblock_set_node_checks(void)
+{
+	return 0;
+}
+#endif
+
 int memblock_basic_checks(void)
 {
 	memblock_initialization_check();
@@ -2444,6 +2545,7 @@ int memblock_basic_checks(void)
 	memblock_bottom_up_checks();
 	memblock_trim_memory_checks();
 	memblock_overlaps_region_checks();
+	memblock_set_node_checks();
 
 	return 0;
 }
diff --git a/tools/testing/selftests/.gitignore b/tools/testing/selftests/.gitignore
index cb24124ac5b9..674aaa02e396 100644
--- a/tools/testing/selftests/.gitignore
+++ b/tools/testing/selftests/.gitignore
@@ -4,7 +4,6 @@ gpiogpio-hammer
 gpioinclude/
 gpiolsgpio
 kselftest_install/
-tpm2/SpaceTest.log
 
 # Python bytecode and cache
 __pycache__/
diff --git a/tools/testing/selftests/bpf/config.x86_64 b/tools/testing/selftests/bpf/config.x86_64
index 5680befae8c6..5e713ef7caa3 100644
--- a/tools/testing/selftests/bpf/config.x86_64
+++ b/tools/testing/selftests/bpf/config.x86_64
@@ -39,7 +39,6 @@ CONFIG_CPU_FREQ_GOV_USERSPACE=y
 CONFIG_CPU_FREQ_STAT=y
 CONFIG_CPU_IDLE_GOV_LADDER=y
 CONFIG_CPUSETS=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_BLAKE2B=y
 CONFIG_CRYPTO_SEQIV=y
 CONFIG_CRYPTO_XXHASH=y
diff --git a/tools/testing/selftests/bpf/prog_tests/for_each.c b/tools/testing/selftests/bpf/prog_tests/for_each.c
index 09f6487f58b9..5fea3209566e 100644
--- a/tools/testing/selftests/bpf/prog_tests/for_each.c
+++ b/tools/testing/selftests/bpf/prog_tests/for_each.c
@@ -6,6 +6,7 @@
 #include "for_each_array_map_elem.skel.h"
 #include "for_each_map_elem_write_key.skel.h"
 #include "for_each_multi_maps.skel.h"
+#include "for_each_hash_modify.skel.h"
 
 static unsigned int duration;
 
@@ -203,6 +204,40 @@ out:
 	for_each_multi_maps__destroy(skel);
 }
 
+static void test_hash_modify(void)
+{
+	struct for_each_hash_modify *skel;
+	int max_entries, i, err;
+	__u64 key, val;
+
+	LIBBPF_OPTS(bpf_test_run_opts, topts,
+		.data_in = &pkt_v4,
+		.data_size_in = sizeof(pkt_v4),
+		.repeat = 1
+	);
+
+	skel = for_each_hash_modify__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "for_each_hash_modify__open_and_load"))
+		return;
+
+	max_entries = bpf_map__max_entries(skel->maps.hashmap);
+	for (i = 0; i < max_entries; i++) {
+		key = i;
+		val = i;
+		err = bpf_map__update_elem(skel->maps.hashmap, &key, sizeof(key),
+					   &val, sizeof(val), BPF_ANY);
+		if (!ASSERT_OK(err, "map_update"))
+			goto out;
+	}
+
+	err = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_pkt_access), &topts);
+	ASSERT_OK(err, "bpf_prog_test_run_opts");
+	ASSERT_OK(topts.retval, "retval");
+
+out:
+	for_each_hash_modify__destroy(skel);
+}
+
 void test_for_each(void)
 {
 	if (test__start_subtest("hash_map"))
@@ -213,4 +248,6 @@ void test_for_each(void)
 		test_write_map_key();
 	if (test__start_subtest("multi_maps"))
 		test_multi_maps();
+	if (test__start_subtest("hash_modify"))
+		test_hash_modify();
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/res_spin_lock.c b/tools/testing/selftests/bpf/prog_tests/res_spin_lock.c
index 115287ba441b..0703e987df89 100644
--- a/tools/testing/selftests/bpf/prog_tests/res_spin_lock.c
+++ b/tools/testing/selftests/bpf/prog_tests/res_spin_lock.c
@@ -25,8 +25,11 @@ static void *spin_lock_thread(void *arg)
 
 	while (!READ_ONCE(skip)) {
 		err = bpf_prog_test_run_opts(prog_fd, &topts);
-		ASSERT_OK(err, "test_run");
-		ASSERT_OK(topts.retval, "test_run retval");
+		if (err || topts.retval) {
+			ASSERT_OK(err, "test_run");
+			ASSERT_OK(topts.retval, "test_run retval");
+			break;
+		}
 	}
 	pthread_exit(arg);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c b/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c
index 2d0796314862..0a99fd404f6d 100644
--- a/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c
+++ b/tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c
@@ -68,7 +68,6 @@ static void test_sockmap_ktls_disconnect_after_delete(int family, int map)
 		goto close_cli;
 
 	err = disconnect(cli);
-	ASSERT_OK(err, "disconnect");
 
 close_cli:
 	close(cli);
diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index 13a2e22f5465..863df7c0fdd0 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -221,7 +221,7 @@
 #define CAN_USE_GOTOL
 #endif
 
-#if _clang_major__ >= 18
+#if __clang_major__ >= 18
 #define CAN_USE_BPF_ST
 #endif
 
diff --git a/tools/testing/selftests/bpf/progs/for_each_hash_modify.c b/tools/testing/selftests/bpf/progs/for_each_hash_modify.c
new file mode 100644
index 000000000000..82307166f789
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/for_each_hash_modify.c
@@ -0,0 +1,30 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2025 Intel Corporation */
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+
+char _license[] SEC("license") = "GPL";
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__uint(max_entries, 128);
+	__type(key, __u64);
+	__type(value, __u64);
+} hashmap SEC(".maps");
+
+static int cb(struct bpf_map *map, __u64 *key, __u64 *val, void *arg)
+{
+	bpf_map_delete_elem(map, key);
+	bpf_map_update_elem(map, key, val, 0);
+	return 0;
+}
+
+SEC("tc")
+int test_pkt_access(struct __sk_buff *skb)
+{
+	(void)skb;
+
+	bpf_for_each_map_elem(&hashmap, cb, NULL, 0);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/res_spin_lock.c b/tools/testing/selftests/bpf/progs/res_spin_lock.c
index b33385dfbd35..22c4fb8b9266 100644
--- a/tools/testing/selftests/bpf/progs/res_spin_lock.c
+++ b/tools/testing/selftests/bpf/progs/res_spin_lock.c
@@ -38,13 +38,14 @@ int res_spin_lock_test(struct __sk_buff *ctx)
 	r = bpf_res_spin_lock(&elem1->lock);
 	if (r)
 		return r;
-	if (!bpf_res_spin_lock(&elem2->lock)) {
+	r = bpf_res_spin_lock(&elem2->lock);
+	if (!r) {
 		bpf_res_spin_unlock(&elem2->lock);
 		bpf_res_spin_unlock(&elem1->lock);
 		return -1;
 	}
 	bpf_res_spin_unlock(&elem1->lock);
-	return 0;
+	return r != -EDEADLK;
 }
 
 SEC("tc")
@@ -124,12 +125,15 @@ int res_spin_lock_test_held_lock_max(struct __sk_buff *ctx)
 	/* Trigger AA, after exhausting entries in the held lock table. This
 	 * time, only the timeout can save us, as AA detection won't succeed.
 	 */
-	if (!bpf_res_spin_lock(locks[34])) {
+	ret = bpf_res_spin_lock(locks[34]);
+	if (!ret) {
 		bpf_res_spin_unlock(locks[34]);
 		ret = 1;
 		goto end;
 	}
 
+	ret = ret != -ETIMEDOUT ? 2 : 0;
+
 end:
 	for (i = i - 1; i >= 0; i--)
 		bpf_res_spin_unlock(locks[i]);
diff --git a/tools/testing/selftests/cgroup/test_cpuset_prs.sh b/tools/testing/selftests/cgroup/test_cpuset_prs.sh
index 400a696a0d21..a17256d9f88a 100755
--- a/tools/testing/selftests/cgroup/test_cpuset_prs.sh
+++ b/tools/testing/selftests/cgroup/test_cpuset_prs.sh
@@ -88,22 +88,32 @@ echo "" > test/cpuset.cpus
 # If isolated CPUs have been reserved at boot time (as shown in
 # cpuset.cpus.isolated), these isolated CPUs should be outside of CPUs 0-8
 # that will be used by this script for testing purpose. If not, some of
-# the tests may fail incorrectly. These pre-isolated CPUs should stay in
-# an isolated state throughout the testing process for now.
+# the tests may fail incorrectly. Wait a bit and retry again just in case
+# these isolated CPUs are leftover from previous run and have just been
+# cleaned up earlier in this script.
+#
+# These pre-isolated CPUs should stay in an isolated state throughout the
+# testing process for now.
 #
 BOOT_ISOLCPUS=$(cat $CGROUP2/cpuset.cpus.isolated)
+[[ -n "$BOOT_ISOLCPUS" ]] && {
+	sleep 0.5
+	BOOT_ISOLCPUS=$(cat $CGROUP2/cpuset.cpus.isolated)
+}
 if [[ -n "$BOOT_ISOLCPUS" ]]
 then
 	[[ $(echo $BOOT_ISOLCPUS | sed -e "s/[,-].*//") -le 8 ]] &&
 		skip_test "Pre-isolated CPUs ($BOOT_ISOLCPUS) overlap CPUs to be tested"
 	echo "Pre-isolated CPUs: $BOOT_ISOLCPUS"
 fi
+
 cleanup()
 {
 	online_cpus
 	cd $CGROUP2
-	rmdir A1/A2/A3 A1/A2 A1 B1 > /dev/null 2>&1
-	rmdir test > /dev/null 2>&1
+	rmdir A1/A2/A3 A1/A2 A1 B1 test/A1 test/B1 test > /dev/null 2>&1
+	rmdir rtest/p1/c11 rtest/p1/c12 rtest/p2/c21 \
+	      rtest/p2/c22 rtest/p1 rtest/p2 rtest > /dev/null 2>&1
 	[[ -n "$SCHED_DEBUG" ]] &&
 		echo "$SCHED_DEBUG" > /sys/kernel/debug/sched/verbose
 }
@@ -173,14 +183,22 @@ test_add_proc()
 #
 # Cgroup test hierarchy
 #
-# root -- A1 -- A2 -- A3
-#      +- B1
+#	      root
+#	        |
+#	 +------+------+
+#	 |             |
+#	 A1            B1
+#	 |
+#	 A2
+#	 |
+#	 A3
 #
 #  P<v> = set cpus.partition (0:member, 1:root, 2:isolated)
 #  C<l> = add cpu-list to cpuset.cpus
 #  X<l> = add cpu-list to cpuset.cpus.exclusive
 #  S<p> = use prefix in subtree_control
 #  T    = put a task into cgroup
+#  CX<l> = add cpu-list to both cpuset.cpus and cpuset.cpus.exclusive
 #  O<c>=<v> = Write <v> to CPU online file of <c>
 #
 # ECPUs    - effective CPUs of cpusets
@@ -207,130 +225,129 @@ TEST_MATRIX=(
 	"  C0-1:P1   .      .    C2-3  S+:C4-5   .      .      .     0 A1:4-5"
 	"   C0-1     .      .   C2-3:P1   .      .      .     C2     0 "
 	"   C0-1     .      .   C2-3:P1   .      .      .    C4-5    0 B1:4-5"
-	"C0-3:P1:S+ C2-3:P1 .      .      .      .      .      .     0 A1:0-1,A2:2-3"
-	"C0-3:P1:S+ C2-3:P1 .      .     C1-3    .      .      .     0 A1:1,A2:2-3"
-	"C2-3:P1:S+  C3:P1  .      .     C3      .      .      .     0 A1:,A2:3 A1:P1,A2:P1"
-	"C2-3:P1:S+  C3:P1  .      .     C3      P0     .      .     0 A1:3,A2:3 A1:P1,A2:P0"
-	"C2-3:P1:S+  C2:P1  .      .     C2-4    .      .      .     0 A1:3-4,A2:2"
-	"C2-3:P1:S+  C3:P1  .      .     C3      .      .     C0-2   0 A1:,B1:0-2 A1:P1,A2:P1"
-	"$SETUP_A123_PARTITIONS    .     C2-3    .      .      .     0 A1:,A2:2,A3:3 A1:P1,A2:P1,A3:P1"
+	"C0-3:P1:S+ C2-3:P1 .      .      .      .      .      .     0 A1:0-1|A2:2-3|XA2:2-3"
+	"C0-3:P1:S+ C2-3:P1 .      .     C1-3    .      .      .     0 A1:1|A2:2-3|XA2:2-3"
+	"C2-3:P1:S+  C3:P1  .      .     C3      .      .      .     0 A1:|A2:3|XA2:3 A1:P1|A2:P1"
+	"C2-3:P1:S+  C3:P1  .      .     C3      P0     .      .     0 A1:3|A2:3 A1:P1|A2:P0"
+	"C2-3:P1:S+  C2:P1  .      .     C2-4    .      .      .     0 A1:3-4|A2:2"
+	"C2-3:P1:S+  C3:P1  .      .     C3      .      .     C0-2   0 A1:|B1:0-2 A1:P1|A2:P1"
+	"$SETUP_A123_PARTITIONS    .     C2-3    .      .      .     0 A1:|A2:2|A3:3 A1:P1|A2:P1|A3:P1"
 
 	# CPU offlining cases:
-	"   C0-1     .      .    C2-3    S+    C4-5     .     O2=0   0 A1:0-1,B1:3"
-	"C0-3:P1:S+ C2-3:P1 .      .     O2=0    .      .      .     0 A1:0-1,A2:3"
-	"C0-3:P1:S+ C2-3:P1 .      .     O2=0   O2=1    .      .     0 A1:0-1,A2:2-3"
-	"C0-3:P1:S+ C2-3:P1 .      .     O1=0    .      .      .     0 A1:0,A2:2-3"
-	"C0-3:P1:S+ C2-3:P1 .      .     O1=0   O1=1    .      .     0 A1:0-1,A2:2-3"
-	"C2-3:P1:S+  C3:P1  .      .     O3=0   O3=1    .      .     0 A1:2,A2:3 A1:P1,A2:P1"
-	"C2-3:P1:S+  C3:P2  .      .     O3=0   O3=1    .      .     0 A1:2,A2:3 A1:P1,A2:P2"
-	"C2-3:P1:S+  C3:P1  .      .     O2=0   O2=1    .      .     0 A1:2,A2:3 A1:P1,A2:P1"
-	"C2-3:P1:S+  C3:P2  .      .     O2=0   O2=1    .      .     0 A1:2,A2:3 A1:P1,A2:P2"
-	"C2-3:P1:S+  C3:P1  .      .     O2=0    .      .      .     0 A1:,A2:3 A1:P1,A2:P1"
-	"C2-3:P1:S+  C3:P1  .      .     O3=0    .      .      .     0 A1:2,A2: A1:P1,A2:P1"
-	"C2-3:P1:S+  C3:P1  .      .    T:O2=0   .      .      .     0 A1:3,A2:3 A1:P1,A2:P-1"
-	"C2-3:P1:S+  C3:P1  .      .      .    T:O3=0   .      .     0 A1:2,A2:2 A1:P1,A2:P-1"
-	"$SETUP_A123_PARTITIONS    .     O1=0    .      .      .     0 A1:,A2:2,A3:3 A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .     O2=0    .      .      .     0 A1:1,A2:,A3:3 A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .     O3=0    .      .      .     0 A1:1,A2:2,A3: A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .    T:O1=0   .      .      .     0 A1:2-3,A2:2-3,A3:3 A1:P1,A2:P-1,A3:P-1"
-	"$SETUP_A123_PARTITIONS    .      .    T:O2=0   .      .     0 A1:1,A2:3,A3:3 A1:P1,A2:P1,A3:P-1"
-	"$SETUP_A123_PARTITIONS    .      .      .    T:O3=0   .     0 A1:1,A2:2,A3:2 A1:P1,A2:P1,A3:P-1"
-	"$SETUP_A123_PARTITIONS    .    T:O1=0  O1=1    .      .     0 A1:1,A2:2,A3:3 A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .      .    T:O2=0  O2=1    .     0 A1:1,A2:2,A3:3 A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .      .      .    T:O3=0  O3=1   0 A1:1,A2:2,A3:3 A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .    T:O1=0  O2=0   O1=1    .     0 A1:1,A2:,A3:3 A1:P1,A2:P1,A3:P1"
-	"$SETUP_A123_PARTITIONS    .    T:O1=0  O2=0   O2=1    .     0 A1:2-3,A2:2-3,A3:3 A1:P1,A2:P-1,A3:P-1"
+	"   C0-1     .      .    C2-3    S+    C4-5     .     O2=0   0 A1:0-1|B1:3"
+	"C0-3:P1:S+ C2-3:P1 .      .     O2=0    .      .      .     0 A1:0-1|A2:3"
+	"C0-3:P1:S+ C2-3:P1 .      .     O2=0   O2=1    .      .     0 A1:0-1|A2:2-3"
+	"C0-3:P1:S+ C2-3:P1 .      .     O1=0    .      .      .     0 A1:0|A2:2-3"
+	"C0-3:P1:S+ C2-3:P1 .      .     O1=0   O1=1    .      .     0 A1:0-1|A2:2-3"
+	"C2-3:P1:S+  C3:P1  .      .     O3=0   O3=1    .      .     0 A1:2|A2:3 A1:P1|A2:P1"
+	"C2-3:P1:S+  C3:P2  .      .     O3=0   O3=1    .      .     0 A1:2|A2:3 A1:P1|A2:P2"
+	"C2-3:P1:S+  C3:P1  .      .     O2=0   O2=1    .      .     0 A1:2|A2:3 A1:P1|A2:P1"
+	"C2-3:P1:S+  C3:P2  .      .     O2=0   O2=1    .      .     0 A1:2|A2:3 A1:P1|A2:P2"
+	"C2-3:P1:S+  C3:P1  .      .     O2=0    .      .      .     0 A1:|A2:3 A1:P1|A2:P1"
+	"C2-3:P1:S+  C3:P1  .      .     O3=0    .      .      .     0 A1:2|A2: A1:P1|A2:P1"
+	"C2-3:P1:S+  C3:P1  .      .    T:O2=0   .      .      .     0 A1:3|A2:3 A1:P1|A2:P-1"
+	"C2-3:P1:S+  C3:P1  .      .      .    T:O3=0   .      .     0 A1:2|A2:2 A1:P1|A2:P-1"
+	"$SETUP_A123_PARTITIONS    .     O1=0    .      .      .     0 A1:|A2:2|A3:3 A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .     O2=0    .      .      .     0 A1:1|A2:|A3:3 A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .     O3=0    .      .      .     0 A1:1|A2:2|A3: A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .    T:O1=0   .      .      .     0 A1:2-3|A2:2-3|A3:3 A1:P1|A2:P-1|A3:P-1"
+	"$SETUP_A123_PARTITIONS    .      .    T:O2=0   .      .     0 A1:1|A2:3|A3:3 A1:P1|A2:P1|A3:P-1"
+	"$SETUP_A123_PARTITIONS    .      .      .    T:O3=0   .     0 A1:1|A2:2|A3:2 A1:P1|A2:P1|A3:P-1"
+	"$SETUP_A123_PARTITIONS    .    T:O1=0  O1=1    .      .     0 A1:1|A2:2|A3:3 A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .      .    T:O2=0  O2=1    .     0 A1:1|A2:2|A3:3 A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .      .      .    T:O3=0  O3=1   0 A1:1|A2:2|A3:3 A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .    T:O1=0  O2=0   O1=1    .     0 A1:1|A2:|A3:3 A1:P1|A2:P1|A3:P1"
+	"$SETUP_A123_PARTITIONS    .    T:O1=0  O2=0   O2=1    .     0 A1:2-3|A2:2-3|A3:3 A1:P1|A2:P-1|A3:P-1"
 
 	#  old-A1 old-A2 old-A3 old-B1 new-A1 new-A2 new-A3 new-B1 fail ECPUs Pstate ISOLCPUS
 	#  ------ ------ ------ ------ ------ ------ ------ ------ ---- ----- ------ --------
 	#
 	# Remote partition and cpuset.cpus.exclusive tests
 	#
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3     .      .      .     0 A1:0-3,A2:1-3,A3:2-3,XA1:2-3"
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3  X2-3:P2   .      .     0 A1:0-1,A2:2-3,A3:2-3 A1:P0,A2:P2 2-3"
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X3:P2    .      .     0 A1:0-2,A2:3,A3:3 A1:P0,A2:P2 3"
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3  X2-3:P2   .     0 A1:0-1,A2:1,A3:2-3 A1:P0,A3:P2 2-3"
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3 X2-3:P2:C3 .     0 A1:0-1,A2:1,A3:2-3 A1:P0,A3:P2 2-3"
-	" C0-3:S+ C1-3:S+ C2-3   C2-3     .      .      .      P2    0 A1:0-3,A2:1-3,A3:2-3,B1:2-3 A1:P0,A3:P0,B1:P-2"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3     .      .      .     0 A1:0-3|A2:1-3|A3:2-3|XA1:2-3"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3  X2-3:P2   .      .     0 A1:0-1|A2:2-3|A3:2-3 A1:P0|A2:P2 2-3"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X3:P2    .      .     0 A1:0-2|A2:3|A3:3 A1:P0|A2:P2 3"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3  X2-3:P2   .     0 A1:0-1|A2:1|A3:2-3 A1:P0|A3:P2 2-3"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3 X2-3:P2:C3 .     0 A1:0-1|A2:1|A3:2-3 A1:P0|A3:P2 2-3"
+	" C0-3:S+ C1-3:S+ C2-3   C2-3     .      .      .      P2    0 A1:0-3|A2:1-3|A3:2-3|B1:2-3 A1:P0|A3:P0|B1:P-2"
 	" C0-3:S+ C1-3:S+ C2-3   C4-5     .      .      .      P2    0 B1:4-5 B1:P2 4-5"
-	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3   X2-3  X2-3:P2   P2    0 A3:2-3,B1:4 A3:P2,B1:P2 2-4"
-	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3   X2-3 X2-3:P2:C1-3 P2  0 A3:2-3,B1:4 A3:P2,B1:P2 2-4"
-	" C0-3:S+ C1-3:S+ C2-3    C4    X1-3  X1-3:P2   P2     .     0 A2:1,A3:2-3 A2:P2,A3:P2 1-3"
-	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3   X2-3  X2-3:P2 P2:C4-5 0 A3:2-3,B1:4-5 A3:P2,B1:P2 2-5"
-	" C4:X0-3:S+ X1-3:S+ X2-3  .      .      P2     .      .     0 A1:4,A2:1-3,A3:1-3 A2:P2 1-3"
-	" C4:X0-3:S+ X1-3:S+ X2-3  .      .      .      P2     .     0 A1:4,A2:4,A3:2-3 A3:P2 2-3"
+	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3   X2-3  X2-3:P2   P2    0 A3:2-3|B1:4 A3:P2|B1:P2 2-4"
+	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3   X2-3 X2-3:P2:C1-3 P2  0 A3:2-3|B1:4 A3:P2|B1:P2 2-4"
+	" C0-3:S+ C1-3:S+ C2-3    C4    X1-3  X1-3:P2   P2     .     0 A2:1|A3:2-3 A2:P2|A3:P2 1-3"
+	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3   X2-3  X2-3:P2 P2:C4-5 0 A3:2-3|B1:4-5 A3:P2|B1:P2 2-5"
+	" C4:X0-3:S+ X1-3:S+ X2-3  .      .      P2     .      .     0 A1:4|A2:1-3|A3:1-3 A2:P2 1-3"
+	" C4:X0-3:S+ X1-3:S+ X2-3  .      .      .      P2     .     0 A1:4|A2:4|A3:2-3 A3:P2 2-3"
 
 	# Nested remote/local partition tests
-	" C0-3:S+ C1-3:S+ C2-3   C4-5   X2-3  X2-3:P1   P2     P1    0 A1:0-1,A2:,A3:2-3,B1:4-5 \
-								       A1:P0,A2:P1,A3:P2,B1:P1 2-3"
-	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3  X2-3:P1   P2     P1    0 A1:0-1,A2:,A3:2-3,B1:4 \
-								       A1:P0,A2:P1,A3:P2,B1:P1 2-4,2-3"
-	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3  X2-3:P1    .     P1    0 A1:0-1,A2:2-3,A3:2-3,B1:4 \
-								       A1:P0,A2:P1,A3:P0,B1:P1"
-	" C0-3:S+ C1-3:S+  C3     C4    X2-3  X2-3:P1   P2     P1    0 A1:0-1,A2:2,A3:3,B1:4 \
-								       A1:P0,A2:P1,A3:P2,B1:P1 2-4,3"
-	" C0-4:S+ C1-4:S+ C2-4     .    X2-4  X2-4:P2  X4:P1    .    0 A1:0-1,A2:2-3,A3:4 \
-								       A1:P0,A2:P2,A3:P1 2-4,2-3"
-	" C0-4:S+ C1-4:S+ C2-4     .    X2-4  X2-4:P2 X3-4:P1   .    0 A1:0-1,A2:2,A3:3-4 \
-								       A1:P0,A2:P2,A3:P1 2"
+	" C0-3:S+ C1-3:S+ C2-3   C4-5   X2-3  X2-3:P1   P2     P1    0 A1:0-1|A2:|A3:2-3|B1:4-5 \
+								       A1:P0|A2:P1|A3:P2|B1:P1 2-3"
+	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3  X2-3:P1   P2     P1    0 A1:0-1|A2:|A3:2-3|B1:4 \
+								       A1:P0|A2:P1|A3:P2|B1:P1 2-4|2-3"
+	" C0-3:S+ C1-3:S+ C2-3    C4    X2-3  X2-3:P1    .     P1    0 A1:0-1|A2:2-3|A3:2-3|B1:4 \
+								       A1:P0|A2:P1|A3:P0|B1:P1"
+	" C0-3:S+ C1-3:S+  C3     C4    X2-3  X2-3:P1   P2     P1    0 A1:0-1|A2:2|A3:3|B1:4 \
+								       A1:P0|A2:P1|A3:P2|B1:P1 2-4|3"
+	" C0-4:S+ C1-4:S+ C2-4     .    X2-4  X2-4:P2  X4:P1    .    0 A1:0-1|A2:2-3|A3:4 \
+								       A1:P0|A2:P2|A3:P1 2-4|2-3"
+	" C0-4:S+ C1-4:S+ C2-4     .    X2-4  X2-4:P2 X3-4:P1   .    0 A1:0-1|A2:2|A3:3-4 \
+								       A1:P0|A2:P2|A3:P1 2"
 	" C0-4:X2-4:S+ C1-4:X2-4:S+:P2 C2-4:X4:P1 \
-				   .      .      X5      .      .    0 A1:0-4,A2:1-4,A3:2-4 \
-								       A1:P0,A2:P-2,A3:P-1"
+				   .      .      X5      .      .    0 A1:0-4|A2:1-4|A3:2-4 \
+								       A1:P0|A2:P-2|A3:P-1 ."
 	" C0-4:X2-4:S+ C1-4:X2-4:S+:P2 C2-4:X4:P1 \
-				   .      .      .      X1      .    0 A1:0-1,A2:2-4,A3:2-4 \
-								       A1:P0,A2:P2,A3:P-1 2-4"
+				   .      .      .      X1      .    0 A1:0-1|A2:2-4|A3:2-4 \
+								       A1:P0|A2:P2|A3:P-1 2-4"
 
 	# Remote partition offline tests
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3 X2-3:P2:O2=0 .   0 A1:0-1,A2:1,A3:3 A1:P0,A3:P2 2-3"
-	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3 X2-3:P2:O2=0 O2=1 0 A1:0-1,A2:1,A3:2-3 A1:P0,A3:P2 2-3"
-	" C0-3:S+ C1-3:S+  C3      .    X2-3   X2-3    P2:O3=0   .   0 A1:0-2,A2:1-2,A3: A1:P0,A3:P2 3"
-	" C0-3:S+ C1-3:S+  C3      .    X2-3   X2-3   T:P2:O3=0  .   0 A1:0-2,A2:1-2,A3:1-2 A1:P0,A3:P-2 3,"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3 X2-3:P2:O2=0 .   0 A1:0-1|A2:1|A3:3 A1:P0|A3:P2 2-3"
+	" C0-3:S+ C1-3:S+ C2-3     .    X2-3   X2-3 X2-3:P2:O2=0 O2=1 0 A1:0-1|A2:1|A3:2-3 A1:P0|A3:P2 2-3"
+	" C0-3:S+ C1-3:S+  C3      .    X2-3   X2-3    P2:O3=0   .   0 A1:0-2|A2:1-2|A3: A1:P0|A3:P2 3"
+	" C0-3:S+ C1-3:S+  C3      .    X2-3   X2-3   T:P2:O3=0  .   0 A1:0-2|A2:1-2|A3:1-2 A1:P0|A3:P-2 3|"
 
 	# An invalidated remote partition cannot self-recover from hotplug
-	" C0-3:S+ C1-3:S+  C2      .    X2-3   X2-3   T:P2:O2=0 O2=1 0 A1:0-3,A2:1-3,A3:2 A1:P0,A3:P-2"
+	" C0-3:S+ C1-3:S+  C2      .    X2-3   X2-3   T:P2:O2=0 O2=1 0 A1:0-3|A2:1-3|A3:2 A1:P0|A3:P-2 ."
 
 	# cpus.exclusive.effective clearing test
-	" C0-3:S+ C1-3:S+  C2      .   X2-3:X    .      .      .     0 A1:0-3,A2:1-3,A3:2,XA1:"
+	" C0-3:S+ C1-3:S+  C2      .   X2-3:X    .      .      .     0 A1:0-3|A2:1-3|A3:2|XA1:"
 
 	# Invalid to valid remote partition transition test
-	" C0-3:S+   C1-3    .      .      .    X3:P2    .      .     0 A1:0-3,A2:1-3,XA2: A2:P-2"
+	" C0-3:S+   C1-3    .      .      .    X3:P2    .      .     0 A1:0-3|A2:1-3|XA2: A2:P-2 ."
 	" C0-3:S+ C1-3:X3:P2
-			    .      .    X2-3    P2      .      .     0 A1:0-2,A2:3,XA2:3 A2:P2 3"
+			    .      .    X2-3    P2      .      .     0 A1:0-2|A2:3|XA2:3 A2:P2 3"
 
 	# Invalid to valid local partition direct transition tests
-	" C1-3:S+:P2 X4:P2  .      .      .      .      .      .     0 A1:1-3,XA1:1-3,A2:1-3:XA2: A1:P2,A2:P-2 1-3"
-	" C1-3:S+:P2 X4:P2  .      .      .    X3:P2    .      .     0 A1:1-2,XA1:1-3,A2:3:XA2:3 A1:P2,A2:P2 1-3"
-	"  C0-3:P2   .      .    C4-6   C0-4     .      .      .     0 A1:0-4,B1:4-6 A1:P-2,B1:P0"
-	"  C0-3:P2   .      .    C4-6 C0-4:C0-3  .      .      .     0 A1:0-3,B1:4-6 A1:P2,B1:P0 0-3"
-	"  C0-3:P2   .      .  C3-5:C4-5  .      .      .      .     0 A1:0-3,B1:4-5 A1:P2,B1:P0 0-3"
+	" C1-3:S+:P2 X4:P2  .      .      .      .      .      .     0 A1:1-3|XA1:1-3|A2:1-3:XA2: A1:P2|A2:P-2 1-3"
+	" C1-3:S+:P2 X4:P2  .      .      .    X3:P2    .      .     0 A1:1-2|XA1:1-3|A2:3:XA2:3 A1:P2|A2:P2 1-3"
+	"  C0-3:P2   .      .    C4-6   C0-4     .      .      .     0 A1:0-4|B1:4-6 A1:P-2|B1:P0"
+	"  C0-3:P2   .      .    C4-6 C0-4:C0-3  .      .      .     0 A1:0-3|B1:4-6 A1:P2|B1:P0 0-3"
 
 	# Local partition invalidation tests
 	" C0-3:X1-3:S+:P2 C1-3:X2-3:S+:P2 C2-3:X3:P2 \
-				   .      .      .      .      .     0 A1:1,A2:2,A3:3 A1:P2,A2:P2,A3:P2 1-3"
+				   .      .      .      .      .     0 A1:1|A2:2|A3:3 A1:P2|A2:P2|A3:P2 1-3"
 	" C0-3:X1-3:S+:P2 C1-3:X2-3:S+:P2 C2-3:X3:P2 \
-				   .      .     X4      .      .     0 A1:1-3,A2:1-3,A3:2-3,XA2:,XA3: A1:P2,A2:P-2,A3:P-2 1-3"
+				   .      .     X4      .      .     0 A1:1-3|A2:1-3|A3:2-3|XA2:|XA3: A1:P2|A2:P-2|A3:P-2 1-3"
 	" C0-3:X1-3:S+:P2 C1-3:X2-3:S+:P2 C2-3:X3:P2 \
-				   .      .    C4:X     .      .     0 A1:1-3,A2:1-3,A3:2-3,XA2:,XA3: A1:P2,A2:P-2,A3:P-2 1-3"
+				   .      .    C4:X     .      .     0 A1:1-3|A2:1-3|A3:2-3|XA2:|XA3: A1:P2|A2:P-2|A3:P-2 1-3"
 	# Local partition CPU change tests
-	" C0-5:S+:P2 C4-5:S+:P1 .  .      .    C3-5     .      .     0 A1:0-2,A2:3-5 A1:P2,A2:P1 0-2"
-	" C0-5:S+:P2 C4-5:S+:P1 .  .    C1-5     .      .      .     0 A1:1-3,A2:4-5 A1:P2,A2:P1 1-3"
+	" C0-5:S+:P2 C4-5:S+:P1 .  .      .    C3-5     .      .     0 A1:0-2|A2:3-5 A1:P2|A2:P1 0-2"
+	" C0-5:S+:P2 C4-5:S+:P1 .  .    C1-5     .      .      .     0 A1:1-3|A2:4-5 A1:P2|A2:P1 1-3"
 
 	# cpus_allowed/exclusive_cpus update tests
 	" C0-3:X2-3:S+ C1-3:X2-3:S+ C2-3:X2-3 \
-				   .    X:C4     .      P2     .     0 A1:4,A2:4,XA2:,XA3:,A3:4 \
-								       A1:P0,A3:P-2"
+				   .    X:C4     .      P2     .     0 A1:4|A2:4|XA2:|XA3:|A3:4 \
+								       A1:P0|A3:P-2 ."
 	" C0-3:X2-3:S+ C1-3:X2-3:S+ C2-3:X2-3 \
-				   .     X1      .      P2     .     0 A1:0-3,A2:1-3,XA1:1,XA2:,XA3:,A3:2-3 \
-								       A1:P0,A3:P-2"
+				   .     X1      .      P2     .     0 A1:0-3|A2:1-3|XA1:1|XA2:|XA3:|A3:2-3 \
+								       A1:P0|A3:P-2 ."
 	" C0-3:X2-3:S+ C1-3:X2-3:S+ C2-3:X2-3 \
-				   .      .     X3      P2     .     0 A1:0-2,A2:1-2,XA2:3,XA3:3,A3:3 \
-								       A1:P0,A3:P2 3"
+				   .      .     X3      P2     .     0 A1:0-2|A2:1-2|XA2:3|XA3:3|A3:3 \
+								       A1:P0|A3:P2 3"
 	" C0-3:X2-3:S+ C1-3:X2-3:S+ C2-3:X2-3:P2 \
-				   .      .     X3      .      .     0 A1:0-3,A2:1-3,XA2:3,XA3:3,A3:2-3 \
-								       A1:P0,A3:P-2"
+				   .      .     X3      .      .     0 A1:0-2|A2:1-2|XA2:3|XA3:3|A3:3|XA3:3 \
+								       A1:P0|A3:P2 3"
 	" C0-3:X2-3:S+ C1-3:X2-3:S+ C2-3:X2-3:P2 \
-				   .     X4      .      .      .     0 A1:0-3,A2:1-3,A3:2-3,XA1:4,XA2:,XA3 \
-								       A1:P0,A3:P-2"
+				   .     X4      .      .      .     0 A1:0-3|A2:1-3|A3:2-3|XA1:4|XA2:|XA3 \
+								       A1:P0|A3:P-2"
 
 	#  old-A1 old-A2 old-A3 old-B1 new-A1 new-A2 new-A3 new-B1 fail ECPUs Pstate ISOLCPUS
 	#  ------ ------ ------ ------ ------ ------ ------ ------ ---- ----- ------ --------
@@ -339,68 +356,127 @@ TEST_MATRIX=(
 	#
 	# Adding CPUs to partition root that are not in parent's
 	# cpuset.cpus is allowed, but those extra CPUs are ignored.
-	"C2-3:P1:S+ C3:P1   .      .      .     C2-4    .      .     0 A1:,A2:2-3 A1:P1,A2:P1"
+	"C2-3:P1:S+ C3:P1   .      .      .     C2-4    .      .     0 A1:|A2:2-3 A1:P1|A2:P1"
 
 	# Taking away all CPUs from parent or itself if there are tasks
 	# will make the partition invalid.
-	"C2-3:P1:S+  C3:P1  .      .      T     C2-3    .      .     0 A1:2-3,A2:2-3 A1:P1,A2:P-1"
-	" C3:P1:S+    C3    .      .      T      P1     .      .     0 A1:3,A2:3 A1:P1,A2:P-1"
-	"$SETUP_A123_PARTITIONS    .    T:C2-3   .      .      .     0 A1:2-3,A2:2-3,A3:3 A1:P1,A2:P-1,A3:P-1"
-	"$SETUP_A123_PARTITIONS    . T:C2-3:C1-3 .      .      .     0 A1:1,A2:2,A3:3 A1:P1,A2:P1,A3:P1"
+	"C2-3:P1:S+  C3:P1  .      .      T     C2-3    .      .     0 A1:2-3|A2:2-3 A1:P1|A2:P-1"
+	" C3:P1:S+    C3    .      .      T      P1     .      .     0 A1:3|A2:3 A1:P1|A2:P-1"
+	"$SETUP_A123_PARTITIONS    .    T:C2-3   .      .      .     0 A1:2-3|A2:2-3|A3:3 A1:P1|A2:P-1|A3:P-1"
+	"$SETUP_A123_PARTITIONS    . T:C2-3:C1-3 .      .      .     0 A1:1|A2:2|A3:3 A1:P1|A2:P1|A3:P1"
 
 	# Changing a partition root to member makes child partitions invalid
-	"C2-3:P1:S+  C3:P1  .      .      P0     .      .      .     0 A1:2-3,A2:3 A1:P0,A2:P-1"
-	"$SETUP_A123_PARTITIONS    .     C2-3    P0     .      .     0 A1:2-3,A2:2-3,A3:3 A1:P1,A2:P0,A3:P-1"
+	"C2-3:P1:S+  C3:P1  .      .      P0     .      .      .     0 A1:2-3|A2:3 A1:P0|A2:P-1"
+	"$SETUP_A123_PARTITIONS    .     C2-3    P0     .      .     0 A1:2-3|A2:2-3|A3:3 A1:P1|A2:P0|A3:P-1"
 
 	# cpuset.cpus can contains cpus not in parent's cpuset.cpus as long
 	# as they overlap.
-	"C2-3:P1:S+  .      .      .      .   C3-4:P1   .      .     0 A1:2,A2:3 A1:P1,A2:P1"
+	"C2-3:P1:S+  .      .      .      .   C3-4:P1   .      .     0 A1:2|A2:3 A1:P1|A2:P1"
 
 	# Deletion of CPUs distributed to child cgroup is allowed.
-	"C0-1:P1:S+ C1      .    C2-3   C4-5     .      .      .     0 A1:4-5,A2:4-5"
+	"C0-1:P1:S+ C1      .    C2-3   C4-5     .      .      .     0 A1:4-5|A2:4-5"
 
 	# To become a valid partition root, cpuset.cpus must overlap parent's
 	# cpuset.cpus.
-	"  C0-1:P1   .      .    C2-3    S+   C4-5:P1   .      .     0 A1:0-1,A2:0-1 A1:P1,A2:P-1"
+	"  C0-1:P1   .      .    C2-3    S+   C4-5:P1   .      .     0 A1:0-1|A2:0-1 A1:P1|A2:P-1"
 
 	# Enabling partition with child cpusets is allowed
-	"  C0-1:S+  C1      .    C2-3    P1      .      .      .     0 A1:0-1,A2:1 A1:P1"
+	"  C0-1:S+  C1      .    C2-3    P1      .      .      .     0 A1:0-1|A2:1 A1:P1"
 
-	# A partition root with non-partition root parent is invalid, but it
+	# A partition root with non-partition root parent is invalid| but it
 	# can be made valid if its parent becomes a partition root too.
-	"  C0-1:S+  C1      .    C2-3     .      P2     .      .     0 A1:0-1,A2:1 A1:P0,A2:P-2"
-	"  C0-1:S+ C1:P2    .    C2-3     P1     .      .      .     0 A1:0,A2:1 A1:P1,A2:P2"
+	"  C0-1:S+  C1      .    C2-3     .      P2     .      .     0 A1:0-1|A2:1 A1:P0|A2:P-2"
+	"  C0-1:S+ C1:P2    .    C2-3     P1     .      .      .     0 A1:0|A2:1 A1:P1|A2:P2 0-1|1"
 
 	# A non-exclusive cpuset.cpus change will invalidate partition and its siblings
-	"  C0-1:P1   .      .    C2-3   C0-2     .      .      .     0 A1:0-2,B1:2-3 A1:P-1,B1:P0"
-	"  C0-1:P1   .      .  P1:C2-3  C0-2     .      .      .     0 A1:0-2,B1:2-3 A1:P-1,B1:P-1"
-	"   C0-1     .      .  P1:C2-3  C0-2     .      .      .     0 A1:0-2,B1:2-3 A1:P0,B1:P-1"
+	"  C0-1:P1   .      .    C2-3   C0-2     .      .      .     0 A1:0-2|B1:2-3 A1:P-1|B1:P0"
+	"  C0-1:P1   .      .  P1:C2-3  C0-2     .      .      .     0 A1:0-2|B1:2-3 A1:P-1|B1:P-1"
+	"   C0-1     .      .  P1:C2-3  C0-2     .      .      .     0 A1:0-2|B1:2-3 A1:P0|B1:P-1"
 
 	# cpuset.cpus can overlap with sibling cpuset.cpus.exclusive but not subsumed by it
-	"   C0-3     .      .    C4-5     X5     .      .      .     0 A1:0-3,B1:4-5"
+	"   C0-3     .      .    C4-5     X5     .      .      .     0 A1:0-3|B1:4-5"
 
 	# Child partition root that try to take all CPUs from parent partition
 	# with tasks will remain invalid.
-	" C1-4:P1:S+ P1     .      .       .     .      .      .     0 A1:1-4,A2:1-4 A1:P1,A2:P-1"
-	" C1-4:P1:S+ P1     .      .       .   C1-4     .      .     0 A1,A2:1-4 A1:P1,A2:P1"
-	" C1-4:P1:S+ P1     .      .       T   C1-4     .      .     0 A1:1-4,A2:1-4 A1:P1,A2:P-1"
+	" C1-4:P1:S+ P1     .      .       .     .      .      .     0 A1:1-4|A2:1-4 A1:P1|A2:P-1"
+	" C1-4:P1:S+ P1     .      .       .   C1-4     .      .     0 A1|A2:1-4 A1:P1|A2:P1"
+	" C1-4:P1:S+ P1     .      .       T   C1-4     .      .     0 A1:1-4|A2:1-4 A1:P1|A2:P-1"
 
 	# Clearing of cpuset.cpus with a preset cpuset.cpus.exclusive shouldn't
 	# affect cpuset.cpus.exclusive.effective.
-	" C1-4:X3:S+ C1:X3  .      .       .     C      .      .     0 A2:1-4,XA2:3"
+	" C1-4:X3:S+ C1:X3  .      .       .     C      .      .     0 A2:1-4|XA2:3"
+
+	# cpuset.cpus can contain CPUs that overlap a sibling cpuset with cpus.exclusive
+	# but creating a local partition out of it is not allowed. Similarly and change
+	# in cpuset.cpus of a local partition that overlaps sibling exclusive CPUs will
+	# invalidate it.
+	" CX1-4:S+ CX2-4:P2 .    C5-6      .     .      .      P1    0 A1:1|A2:2-4|B1:5-6|XB1:5-6 \
+								       A1:P0|A2:P2:B1:P1 2-4"
+	" CX1-4:S+ CX2-4:P2 .    C3-6      .     .      .      P1    0 A1:1|A2:2-4|B1:5-6 \
+								       A1:P0|A2:P2:B1:P-1 2-4"
+	" CX1-4:S+ CX2-4:P2 .    C5-6      .     .      .   P1:C3-6  0 A1:1|A2:2-4|B1:5-6 \
+								       A1:P0|A2:P2:B1:P-1 2-4"
 
 	#  old-A1 old-A2 old-A3 old-B1 new-A1 new-A2 new-A3 new-B1 fail ECPUs Pstate ISOLCPUS
 	#  ------ ------ ------ ------ ------ ------ ------ ------ ---- ----- ------ --------
 	# Failure cases:
 
 	# A task cannot be added to a partition with no cpu
-	"C2-3:P1:S+  C3:P1  .      .    O2=0:T   .      .      .     1 A1:,A2:3 A1:P1,A2:P1"
+	"C2-3:P1:S+  C3:P1  .      .    O2=0:T   .      .      .     1 A1:|A2:3 A1:P1|A2:P1"
 
 	# Changes to cpuset.cpus.exclusive that violate exclusivity rule is rejected
-	"   C0-3     .      .    C4-5   X0-3     .      .     X3-5   1 A1:0-3,B1:4-5"
+	"   C0-3     .      .    C4-5   X0-3     .      .     X3-5   1 A1:0-3|B1:4-5"
 
 	# cpuset.cpus cannot be a subset of sibling cpuset.cpus.exclusive
-	"   C0-3     .      .    C4-5   X3-5     .      .      .     1 A1:0-3,B1:4-5"
+	"   C0-3     .      .    C4-5   X3-5     .      .      .     1 A1:0-3|B1:4-5"
+)
+
+#
+# Cpuset controller remote partition test matrix.
+#
+# Cgroup test hierarchy
+#
+#	      root
+#	        |
+#	      rtest (cpuset.cpus.exclusive=1-7)
+#	        |
+#	 +------+------+
+#	 |             |
+#	 p1            p2
+#     +--+--+       +--+--+
+#     |     |       |     |
+#    c11   c12     c21   c22
+#
+# REMOTE_TEST_MATRIX uses the same notational convention as TEST_MATRIX.
+# Only CPUs 1-7 should be used.
+#
+REMOTE_TEST_MATRIX=(
+	#  old-p1 old-p2 old-c11 old-c12 old-c21 old-c22
+	#  new-p1 new-p2 new-c11 new-c12 new-c21 new-c22 ECPUs Pstate ISOLCPUS
+	#  ------ ------ ------- ------- ------- ------- ----- ------ --------
+	" X1-3:S+ X4-6:S+ X1-2     X3     X4-5     X6 \
+	      .      .     P2      P2      P2      P2    c11:1-2|c12:3|c21:4-5|c22:6 \
+							 c11:P2|c12:P2|c21:P2|c22:P2 1-6"
+	" CX1-4:S+   .   X1-2:P2   C3      .       .  \
+	      .      .     .      C3-4     .       .     p1:3-4|c11:1-2|c12:3-4 \
+							 p1:P0|c11:P2|c12:P0 1-2"
+	" CX1-4:S+   .   X1-2:P2   .       .       .  \
+	    X2-4     .     .       .       .       .     p1:1,3-4|c11:2 \
+							 p1:P0|c11:P2 2"
+	" CX1-5:S+   .   X1-2:P2 X3-5:P1   .       .  \
+	    X2-4     .     .       .       .       .     p1:1,5|c11:2|c12:3-4 \
+							 p1:P0|c11:P2|c12:P1 2"
+	" CX1-4:S+   .   X1-2:P2 X3-4:P1   .       .  \
+	      .      .     X2      .       .       .     p1:1|c11:2|c12:3-4 \
+							 p1:P0|c11:P2|c12:P1 2"
+	# p1 as member, will get its effective CPUs from its parent rtest
+	" CX1-4:S+   .   X1-2:P2 X3-4:P1   .       .  \
+	      .      .     X1     CX2-4    .       .     p1:5-7|c11:1|c12:2-4 \
+							 p1:P0|c11:P2|c12:P1 1"
+	" CX1-4:S+ X5-6:P1:S+ .    .       .       .  \
+	      .      .   X1-2:P2  X4-5:P1  .     X1-7:P2 p1:3|c11:1-2|c12:4:c22:5-6 \
+							 p1:P0|p2:P1|c11:P2|c12:P1|c22:P2 \
+							 1-2,4-6|1-2,5-6"
 )
 
 #
@@ -453,25 +529,26 @@ set_ctrl_state()
 		PFILE=$CGRP/cpuset.cpus.partition
 		CFILE=$CGRP/cpuset.cpus
 		XFILE=$CGRP/cpuset.cpus.exclusive
-		S=$(expr substr $CMD 1 1)
-		if [[ $S = S ]]
-		then
-			PREFIX=${CMD#?}
+		case $CMD in
+		    S*) PREFIX=${CMD#?}
 			COMM="echo ${PREFIX}${CTRL} > $SFILE"
 			eval $COMM $REDIRECT
-		elif [[ $S = X ]]
-		then
+			;;
+		    X*)
 			CPUS=${CMD#?}
 			COMM="echo $CPUS > $XFILE"
 			eval $COMM $REDIRECT
-		elif [[ $S = C ]]
-		then
-			CPUS=${CMD#?}
+			;;
+		    CX*)
+			CPUS=${CMD#??}
+			COMM="echo $CPUS > $CFILE; echo $CPUS > $XFILE"
+			eval $COMM $REDIRECT
+			;;
+		    C*) CPUS=${CMD#?}
 			COMM="echo $CPUS > $CFILE"
 			eval $COMM $REDIRECT
-		elif [[ $S = P ]]
-		then
-			VAL=${CMD#?}
+			;;
+		    P*) VAL=${CMD#?}
 			case $VAL in
 			0)  VAL=member
 			    ;;
@@ -486,15 +563,17 @@ set_ctrl_state()
 			esac
 			COMM="echo $VAL > $PFILE"
 			eval $COMM $REDIRECT
-		elif [[ $S = O ]]
-		then
-			VAL=${CMD#?}
+			;;
+		    O*) VAL=${CMD#?}
 			write_cpu_online $VAL
-		elif [[ $S = T ]]
-		then
-			COMM="echo 0 > $TFILE"
+			;;
+		    T*) COMM="echo 0 > $TFILE"
 			eval $COMM $REDIRECT
-		fi
+			;;
+		    *)  echo "Unknown command: $CMD"
+		        exit 1
+			;;
+		esac
 		RET=$?
 		[[ $RET -ne 0 ]] && {
 			[[ -n "$SHOWERR" ]] && {
@@ -532,21 +611,18 @@ online_cpus()
 }
 
 #
-# Return 1 if the list of effective cpus isn't the same as the initial list.
+# Remove all the test cgroup directories
 #
 reset_cgroup_states()
 {
 	echo 0 > $CGROUP2/cgroup.procs
 	online_cpus
-	rmdir A1/A2/A3 A1/A2 A1 B1 > /dev/null 2>&1
-	pause 0.02
-	set_ctrl_state . R-
-	pause 0.01
+	rmdir $RESET_LIST > /dev/null 2>&1
 }
 
 dump_states()
 {
-	for DIR in . A1 A1/A2 A1/A2/A3 B1
+	for DIR in $CGROUP_LIST
 	do
 		CPUS=$DIR/cpuset.cpus
 		ECPUS=$DIR/cpuset.cpus.effective
@@ -566,17 +642,33 @@ dump_states()
 }
 
 #
+# Set the actual cgroup directory into $CGRP_DIR
+# $1 - cgroup name
+#
+set_cgroup_dir()
+{
+	CGRP_DIR=$1
+	[[ $CGRP_DIR = A2  ]] && CGRP_DIR=A1/A2
+	[[ $CGRP_DIR = A3  ]] && CGRP_DIR=A1/A2/A3
+	[[ $CGRP_DIR = c11 ]] && CGRP_DIR=p1/c11
+	[[ $CGRP_DIR = c12 ]] && CGRP_DIR=p1/c12
+	[[ $CGRP_DIR = c21 ]] && CGRP_DIR=p2/c21
+	[[ $CGRP_DIR = c22 ]] && CGRP_DIR=p2/c22
+}
+
+#
 # Check effective cpus
-# $1 - check string, format: <cgroup>:<cpu-list>[,<cgroup>:<cpu-list>]*
+# $1 - check string, format: <cgroup>:<cpu-list>[|<cgroup>:<cpu-list>]*
 #
 check_effective_cpus()
 {
 	CHK_STR=$1
-	for CHK in $(echo $CHK_STR | sed -e "s/,/ /g")
+	for CHK in $(echo $CHK_STR | sed -e "s/|/ /g")
 	do
 		set -- $(echo $CHK | sed -e "s/:/ /g")
 		CGRP=$1
-		CPUS=$2
+		EXPECTED_CPUS=$2
+		ACTUAL_CPUS=
 		if [[ $CGRP = X* ]]
 		then
 			CGRP=${CGRP#X}
@@ -584,41 +676,39 @@ check_effective_cpus()
 		else
 			FILE=cpuset.cpus.effective
 		fi
-		[[ $CGRP = A2 ]] && CGRP=A1/A2
-		[[ $CGRP = A3 ]] && CGRP=A1/A2/A3
-		[[ -e $CGRP/$FILE ]] || return 1
-		[[ $CPUS = $(cat $CGRP/$FILE) ]] || return 1
+		set_cgroup_dir $CGRP
+		[[ -e $CGRP_DIR/$FILE ]] || return 1
+		ACTUAL_CPUS=$(cat $CGRP_DIR/$FILE)
+		[[ $EXPECTED_CPUS = $ACTUAL_CPUS ]] || return 1
 	done
 }
 
 #
 # Check cgroup states
-#  $1 - check string, format: <cgroup>:<state>[,<cgroup>:<state>]*
+#  $1 - check string, format: <cgroup>:<state>[|<cgroup>:<state>]*
 #
 check_cgroup_states()
 {
 	CHK_STR=$1
-	for CHK in $(echo $CHK_STR | sed -e "s/,/ /g")
+	for CHK in $(echo $CHK_STR | sed -e "s/|/ /g")
 	do
 		set -- $(echo $CHK | sed -e "s/:/ /g")
 		CGRP=$1
-		CGRP_DIR=$CGRP
-		STATE=$2
+		EXPECTED_STATE=$2
 		FILE=
-		EVAL=$(expr substr $STATE 2 2)
-		[[ $CGRP = A2 ]] && CGRP_DIR=A1/A2
-		[[ $CGRP = A3 ]] && CGRP_DIR=A1/A2/A3
+		EVAL=$(expr substr $EXPECTED_STATE 2 2)
 
-		case $STATE in
+		set_cgroup_dir $CGRP
+		case $EXPECTED_STATE in
 			P*) FILE=$CGRP_DIR/cpuset.cpus.partition
 			    ;;
-			*)  echo "Unknown state: $STATE!"
+			*)  echo "Unknown state: $EXPECTED_STATE!"
 			    exit 1
 			    ;;
 		esac
-		VAL=$(cat $FILE)
+		ACTUAL_STATE=$(cat $FILE)
 
-		case "$VAL" in
+		case "$ACTUAL_STATE" in
 			member) VAL=0
 				;;
 			root)	VAL=1
@@ -642,7 +732,7 @@ check_cgroup_states()
 		[[ $VAL -eq 1 && $VERBOSE -gt 0 ]] && {
 			DOMS=$(cat $CGRP_DIR/cpuset.cpus.effective)
 			[[ -n "$DOMS" ]] &&
-				echo " [$CGRP] sched-domain: $DOMS" > $CONSOLE
+				echo " [$CGRP_DIR] sched-domain: $DOMS" > $CONSOLE
 		}
 	done
 	return 0
@@ -665,22 +755,22 @@ check_cgroup_states()
 #
 check_isolcpus()
 {
-	EXPECT_VAL=$1
-	ISOLCPUS=
+	EXPECTED_ISOLCPUS=$1
+	ISCPUS=${CGROUP2}/cpuset.cpus.isolated
+	ISOLCPUS=$(cat $ISCPUS)
 	LASTISOLCPU=
 	SCHED_DOMAINS=/sys/kernel/debug/sched/domains
-	ISCPUS=${CGROUP2}/cpuset.cpus.isolated
-	if [[ $EXPECT_VAL = . ]]
+	if [[ $EXPECTED_ISOLCPUS = . ]]
 	then
-		EXPECT_VAL=
-		EXPECT_VAL2=
-	elif [[ $(expr $EXPECT_VAL : ".*,.*") > 0 ]]
+		EXPECTED_ISOLCPUS=
+		EXPECTED_SDOMAIN=
+	elif [[ $(expr $EXPECTED_ISOLCPUS : ".*|.*") > 0 ]]
 	then
-		set -- $(echo $EXPECT_VAL | sed -e "s/,/ /g")
-		EXPECT_VAL=$1
-		EXPECT_VAL2=$2
+		set -- $(echo $EXPECTED_ISOLCPUS | sed -e "s/|/ /g")
+		EXPECTED_ISOLCPUS=$2
+		EXPECTED_SDOMAIN=$1
 	else
-		EXPECT_VAL2=$EXPECT_VAL
+		EXPECTED_SDOMAIN=$EXPECTED_ISOLCPUS
 	fi
 
 	#
@@ -689,20 +779,21 @@ check_isolcpus()
 	# to make appending those CPUs easier.
 	#
 	[[ -n "$BOOT_ISOLCPUS" ]] && {
-		EXPECT_VAL=${EXPECT_VAL:+${EXPECT_VAL},}${BOOT_ISOLCPUS}
-		EXPECT_VAL2=${EXPECT_VAL2:+${EXPECT_VAL2},}${BOOT_ISOLCPUS}
+		EXPECTED_ISOLCPUS=${EXPECTED_ISOLCPUS:+${EXPECTED_ISOLCPUS},}${BOOT_ISOLCPUS}
+		EXPECTED_SDOMAIN=${EXPECTED_SDOMAIN:+${EXPECTED_SDOMAIN},}${BOOT_ISOLCPUS}
 	}
 
 	#
 	# Check cpuset.cpus.isolated cpumask
 	#
-	[[ "$EXPECT_VAL2" != "$ISOLCPUS" ]] && {
+	[[ "$EXPECTED_ISOLCPUS" != "$ISOLCPUS" ]] && {
 		# Take a 50ms pause and try again
 		pause 0.05
 		ISOLCPUS=$(cat $ISCPUS)
 	}
-	[[ "$EXPECT_VAL2" != "$ISOLCPUS" ]] && return 1
+	[[ "$EXPECTED_ISOLCPUS" != "$ISOLCPUS" ]] && return 1
 	ISOLCPUS=
+	EXPECTED_ISOLCPUS=$EXPECTED_SDOMAIN
 
 	#
 	# Use the sched domain in debugfs to check isolated CPUs, if available
@@ -736,7 +827,7 @@ check_isolcpus()
 	done
 	[[ "$ISOLCPUS" = *- ]] && ISOLCPUS=${ISOLCPUS}$LASTISOLCPU
 
-	[[ "$EXPECT_VAL" = "$ISOLCPUS" ]]
+	[[ "$EXPECTED_SDOMAIN" = "$ISOLCPUS" ]]
 }
 
 test_fail()
@@ -774,6 +865,63 @@ null_isolcpus_check()
 }
 
 #
+# Check state transition test result
+#  $1 - Test number
+#  $2 - Expected effective CPU values
+#  $3 - Expected partition states
+#  $4 - Expected isolated CPUs
+#
+check_test_results()
+{
+	_NR=$1
+	_ECPUS="$2"
+	_PSTATES="$3"
+	_ISOLCPUS="$4"
+
+	[[ -n "$_ECPUS" && "$_ECPUS" != . ]] && {
+		check_effective_cpus $_ECPUS
+		[[ $? -ne 0 ]] && test_fail $_NR "effective CPU" \
+			 "Cgroup $CGRP: expected $EXPECTED_CPUS, got $ACTUAL_CPUS"
+	}
+
+	[[ -n "$_PSTATES" && "$_PSTATES" != . ]] && {
+		check_cgroup_states $_PSTATES
+		[[ $? -ne 0 ]] && test_fail $_NR states \
+			"Cgroup $CGRP: expected $EXPECTED_STATE, got $ACTUAL_STATE"
+	}
+
+	# Compare the expected isolated CPUs with the actual ones,
+	# if available
+	[[ -n "$_ISOLCPUS" ]] && {
+		check_isolcpus $_ISOLCPUS
+		[[ $? -ne 0 ]] && {
+			[[ -n "$BOOT_ISOLCPUS" ]] && _ISOLCPUS=${_ISOLCPUS},${BOOT_ISOLCPUS}
+			test_fail $_NR "isolated CPU" \
+				"Expect $_ISOLCPUS, get $ISOLCPUS instead"
+		}
+	}
+	reset_cgroup_states
+	#
+	# Check to see if effective cpu list changes
+	#
+	_NEWLIST=$(cat $CGROUP2/cpuset.cpus.effective)
+	RETRY=0
+	while [[ $_NEWLIST != $CPULIST && $RETRY -lt 8 ]]
+	do
+		# Wait a bit longer & recheck a few times
+		pause 0.02
+		((RETRY++))
+		_NEWLIST=$(cat $CGROUP2/cpuset.cpus.effective)
+	done
+	[[ $_NEWLIST != $CPULIST ]] && {
+		echo "Effective cpus changed to $_NEWLIST after test $_NR!"
+		exit 1
+	}
+	null_isolcpus_check
+	[[ $VERBOSE -gt 0 ]] && echo "Test $I done."
+}
+
+#
 # Run cpuset state transition test
 #  $1 - test matrix name
 #
@@ -785,6 +933,8 @@ run_state_test()
 {
 	TEST=$1
 	CONTROLLER=cpuset
+	CGROUP_LIST=". A1 A1/A2 A1/A2/A3 B1"
+	RESET_LIST="A1/A2/A3 A1/A2 A1 B1"
 	I=0
 	eval CNT="\${#$TEST[@]}"
 
@@ -812,10 +962,11 @@ run_state_test()
 		STATES=${11}
 		ICPUS=${12}
 
-		set_ctrl_state_noerr B1       $OLD_B1
 		set_ctrl_state_noerr A1       $OLD_A1
 		set_ctrl_state_noerr A1/A2    $OLD_A2
 		set_ctrl_state_noerr A1/A2/A3 $OLD_A3
+		set_ctrl_state_noerr B1       $OLD_B1
+
 		RETVAL=0
 		set_ctrl_state A1       $NEW_A1; ((RETVAL += $?))
 		set_ctrl_state A1/A2    $NEW_A2; ((RETVAL += $?))
@@ -824,47 +975,79 @@ run_state_test()
 
 		[[ $RETVAL -ne $RESULT ]] && test_fail $I result
 
-		[[ -n "$ECPUS" && "$ECPUS" != . ]] && {
-			check_effective_cpus $ECPUS
-			[[ $? -ne 0 ]] && test_fail $I "effective CPU"
-		}
+		check_test_results $I "$ECPUS" "$STATES" "$ICPUS"
+		((I++))
+	done
+	echo "All $I tests of $TEST PASSED."
+}
 
-		[[ -n "$STATES" && "$STATES" != . ]] && {
-			check_cgroup_states $STATES
-			[[ $? -ne 0 ]] && test_fail $I states
-		}
+#
+# Run cpuset remote partition state transition test
+#  $1 - test matrix name
+#
+run_remote_state_test()
+{
+	TEST=$1
+	CONTROLLER=cpuset
+	[[ -d rtest ]] || mkdir rtest
+	cd rtest
+	echo +cpuset > cgroup.subtree_control
+	echo "1-7" > cpuset.cpus
+	echo "1-7" > cpuset.cpus.exclusive
+	CGROUP_LIST=".. . p1 p2 p1/c11 p1/c12 p2/c21 p2/c22"
+	RESET_LIST="p1/c11 p1/c12 p2/c21 p2/c22 p1 p2"
+	I=0
+	eval CNT="\${#$TEST[@]}"
 
-		# Compare the expected isolated CPUs with the actual ones,
-		# if available
-		[[ -n "$ICPUS" ]] && {
-			check_isolcpus $ICPUS
-			[[ $? -ne 0 ]] && {
-				[[ -n "$BOOT_ISOLCPUS" ]] && ICPUS=${ICPUS},${BOOT_ISOLCPUS}
-				test_fail $I "isolated CPU" \
-					"Expect $ICPUS, get $ISOLCPUS instead"
-			}
-		}
-		reset_cgroup_states
-		#
-		# Check to see if effective cpu list changes
-		#
-		NEWLIST=$(cat cpuset.cpus.effective)
-		RETRY=0
-		while [[ $NEWLIST != $CPULIST && $RETRY -lt 8 ]]
-		do
-			# Wait a bit longer & recheck a few times
-			pause 0.02
-			((RETRY++))
-			NEWLIST=$(cat cpuset.cpus.effective)
-		done
-		[[ $NEWLIST != $CPULIST ]] && {
-			echo "Effective cpus changed to $NEWLIST after test $I!"
-			exit 1
+	reset_cgroup_states
+	console_msg "Running remote partition state transition test ..."
+
+	while [[ $I -lt $CNT ]]
+	do
+		echo "Running test $I ..." > $CONSOLE
+		[[ $VERBOSE -gt 1 ]] && {
+			echo ""
+			eval echo \${$TEST[$I]}
 		}
-		null_isolcpus_check
-		[[ $VERBOSE -gt 0 ]] && echo "Test $I done."
+		eval set -- "\${$TEST[$I]}"
+		OLD_p1=$1
+		OLD_p2=$2
+		OLD_c11=$3
+		OLD_c12=$4
+		OLD_c21=$5
+		OLD_c22=$6
+		NEW_p1=$7
+		NEW_p2=$8
+		NEW_c11=$9
+		NEW_c12=${10}
+		NEW_c21=${11}
+		NEW_c22=${12}
+		ECPUS=${13}
+		STATES=${14}
+		ICPUS=${15}
+
+		set_ctrl_state_noerr p1     $OLD_p1
+		set_ctrl_state_noerr p2     $OLD_p2
+		set_ctrl_state_noerr p1/c11 $OLD_c11
+		set_ctrl_state_noerr p1/c12 $OLD_c12
+		set_ctrl_state_noerr p2/c21 $OLD_c21
+		set_ctrl_state_noerr p2/c22 $OLD_c22
+
+		RETVAL=0
+		set_ctrl_state p1     $NEW_p1 ; ((RETVAL += $?))
+		set_ctrl_state p2     $NEW_p2 ; ((RETVAL += $?))
+		set_ctrl_state p1/c11 $NEW_c11; ((RETVAL += $?))
+		set_ctrl_state p1/c12 $NEW_c12; ((RETVAL += $?))
+		set_ctrl_state p2/c21 $NEW_c21; ((RETVAL += $?))
+		set_ctrl_state p2/c22 $NEW_c22; ((RETVAL += $?))
+
+		[[ $RETVAL -ne 0 ]] && test_fail $I result
+
+		check_test_results $I "$ECPUS" "$STATES" "$ICPUS"
 		((I++))
 	done
+	cd ..
+	rmdir rtest
 	echo "All $I tests of $TEST PASSED."
 }
 
@@ -932,6 +1115,7 @@ test_isolated()
 	echo $$ > $CGROUP2/cgroup.procs
 	[[ -d A1 ]] && rmdir A1
 	null_isolcpus_check
+	pause 0.05
 }
 
 #
@@ -997,10 +1181,13 @@ test_inotify()
 	else
 		echo "Inotify test PASSED"
 	fi
+	echo member > cpuset.cpus.partition
+	echo "" > cpuset.cpus
 }
 
 trap cleanup 0 2 3 6
 run_state_test TEST_MATRIX
+run_remote_state_test REMOTE_TEST_MATRIX
 test_isolated
 test_inotify
 echo "All tests PASSED."
diff --git a/tools/testing/selftests/drivers/net/dsa/tc_taprio.sh b/tools/testing/selftests/drivers/net/dsa/tc_taprio.sh
new file mode 120000
index 000000000000..d16a65e7595d
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/dsa/tc_taprio.sh
@@ -0,0 +1 @@
+run_net_forwarding_test.sh
\ No newline at end of file
diff --git a/tools/testing/selftests/drivers/net/hds.py b/tools/testing/selftests/drivers/net/hds.py
index 8b7f6acad15f..7c90a040ce45 100755
--- a/tools/testing/selftests/drivers/net/hds.py
+++ b/tools/testing/selftests/drivers/net/hds.py
@@ -6,7 +6,7 @@ import os
 from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_raises, KsftSkipEx
 from lib.py import CmdExitFailure, EthtoolFamily, NlError
 from lib.py import NetDrvEnv
-from lib.py import defer, ethtool, ip
+from lib.py import defer, ethtool, ip, random
 
 
 def _get_hds_mode(cfg, netnl) -> str:
@@ -109,6 +109,36 @@ def set_hds_thresh_zero(cfg, netnl) -> None:
 
     ksft_eq(0, rings['hds-thresh'])
 
+def set_hds_thresh_random(cfg, netnl) -> None:
+    try:
+        rings = netnl.rings_get({'header': {'dev-index': cfg.ifindex}})
+    except NlError as e:
+        raise KsftSkipEx('ring-get not supported by device')
+    if 'hds-thresh' not in rings:
+        raise KsftSkipEx('hds-thresh not supported by device')
+    if 'hds-thresh-max' not in rings:
+        raise KsftSkipEx('hds-thresh-max not defined by device')
+
+    if rings['hds-thresh-max'] < 2:
+        raise KsftSkipEx('hds-thresh-max is too small')
+    elif rings['hds-thresh-max'] == 2:
+        hds_thresh = 1
+    else:
+        while True:
+            hds_thresh = random.randint(1, rings['hds-thresh-max'] - 1)
+            if hds_thresh != rings['hds-thresh']:
+                break
+
+    try:
+        netnl.rings_set({'header': {'dev-index': cfg.ifindex}, 'hds-thresh': hds_thresh})
+    except NlError as e:
+        if e.error == errno.EINVAL:
+            raise KsftSkipEx("hds-thresh-set not supported by the device")
+        elif e.error == errno.EOPNOTSUPP:
+            raise KsftSkipEx("ring-set not supported by the device")
+    rings = netnl.rings_get({'header': {'dev-index': cfg.ifindex}})
+    ksft_eq(hds_thresh, rings['hds-thresh'])
+
 def set_hds_thresh_max(cfg, netnl) -> None:
     try:
         rings = netnl.rings_get({'header': {'dev-index': cfg.ifindex}})
@@ -243,6 +273,7 @@ def main() -> None:
                   get_hds_thresh,
                   set_hds_disable,
                   set_hds_enable,
+                  set_hds_thresh_random,
                   set_hds_thresh_zero,
                   set_hds_thresh_max,
                   set_hds_thresh_gt,
diff --git a/tools/testing/selftests/drivers/net/hw/iou-zcrx.py b/tools/testing/selftests/drivers/net/hw/iou-zcrx.py
index 9f271ab6ec04..6a0378e06cab 100755
--- a/tools/testing/selftests/drivers/net/hw/iou-zcrx.py
+++ b/tools/testing/selftests/drivers/net/hw/iou-zcrx.py
@@ -35,6 +35,7 @@ def test_zcrx(cfg) -> None:
     rx_ring = _get_rx_ring_entries(cfg)
 
     try:
+        ethtool(f"-G {cfg.ifname} tcp-data-split on", host=cfg.remote)
         ethtool(f"-G {cfg.ifname} rx 64", host=cfg.remote)
         ethtool(f"-X {cfg.ifname} equal {combined_chans - 1}", host=cfg.remote)
         flow_rule_id = _set_flow_rule(cfg, combined_chans - 1)
@@ -48,6 +49,7 @@ def test_zcrx(cfg) -> None:
         ethtool(f"-N {cfg.ifname} delete {flow_rule_id}", host=cfg.remote)
         ethtool(f"-X {cfg.ifname} default", host=cfg.remote)
         ethtool(f"-G {cfg.ifname} rx {rx_ring}", host=cfg.remote)
+        ethtool(f"-G {cfg.ifname} tcp-data-split auto", host=cfg.remote)
 
 
 def test_zcrx_oneshot(cfg) -> None:
@@ -59,6 +61,7 @@ def test_zcrx_oneshot(cfg) -> None:
     rx_ring = _get_rx_ring_entries(cfg)
 
     try:
+        ethtool(f"-G {cfg.ifname} tcp-data-split on", host=cfg.remote)
         ethtool(f"-G {cfg.ifname} rx 64", host=cfg.remote)
         ethtool(f"-X {cfg.ifname} equal {combined_chans - 1}", host=cfg.remote)
         flow_rule_id = _set_flow_rule(cfg, combined_chans - 1)
@@ -72,6 +75,7 @@ def test_zcrx_oneshot(cfg) -> None:
         ethtool(f"-N {cfg.ifname} delete {flow_rule_id}", host=cfg.remote)
         ethtool(f"-X {cfg.ifname} default", host=cfg.remote)
         ethtool(f"-G {cfg.ifname} rx {rx_ring}", host=cfg.remote)
+        ethtool(f"-G {cfg.ifname} tcp-data-split auto", host=cfg.remote)
 
 
 def main() -> None:
diff --git a/tools/testing/selftests/drivers/net/ocelot/psfp.sh b/tools/testing/selftests/drivers/net/ocelot/psfp.sh
index bed748dde4b0..8972f42dfe03 100755
--- a/tools/testing/selftests/drivers/net/ocelot/psfp.sh
+++ b/tools/testing/selftests/drivers/net/ocelot/psfp.sh
@@ -266,18 +266,14 @@ run_test()
 		"${base_time}" \
 		"${CYCLE_TIME_NS}" \
 		"${SHIFT_TIME_NS}" \
+		"${GATE_DURATION_NS}" \
 		"${NUM_PKTS}" \
 		"${STREAM_VID}" \
 		"${STREAM_PRIO}" \
 		"" \
 		"${isochron_dat}"
 
-	# Count all received packets by looking at the non-zero RX timestamps
-	received=$(isochron report \
-		--input-file "${isochron_dat}" \
-		--printf-format "%u\n" --printf-args "R" | \
-		grep -w -v '0' | wc -l)
-
+	received=$(isochron_report_num_received "${isochron_dat}")
 	if [ "${received}" = "${expected}" ]; then
 		RET=0
 	else
diff --git a/tools/testing/selftests/filesystems/mount-notify/mount-notify_test.c b/tools/testing/selftests/filesystems/mount-notify/mount-notify_test.c
index 4a2d5c454fd1..59a71f22fb11 100644
--- a/tools/testing/selftests/filesystems/mount-notify/mount-notify_test.c
+++ b/tools/testing/selftests/filesystems/mount-notify/mount-notify_test.c
@@ -48,8 +48,16 @@ static uint64_t get_mnt_id(struct __test_metadata *const _metadata,
 
 static const char root_mntpoint_templ[] = "/tmp/mount-notify_test_root.XXXXXX";
 
+static const int mark_cmds[] = {
+	FAN_MARK_ADD,
+	FAN_MARK_REMOVE,
+	FAN_MARK_FLUSH
+};
+
+#define NUM_FAN_FDS ARRAY_SIZE(mark_cmds)
+
 FIXTURE(fanotify) {
-	int fan_fd;
+	int fan_fd[NUM_FAN_FDS];
 	char buf[256];
 	unsigned int rem;
 	void *next;
@@ -61,7 +69,7 @@ FIXTURE(fanotify) {
 
 FIXTURE_SETUP(fanotify)
 {
-	int ret;
+	int i, ret;
 
 	ASSERT_EQ(unshare(CLONE_NEWNS), 0);
 
@@ -89,20 +97,34 @@ FIXTURE_SETUP(fanotify)
 	self->root_id = get_mnt_id(_metadata, "/");
 	ASSERT_NE(self->root_id, 0);
 
-	self->fan_fd = fanotify_init(FAN_REPORT_MNT, 0);
-	ASSERT_GE(self->fan_fd, 0);
-
-	ret = fanotify_mark(self->fan_fd, FAN_MARK_ADD | FAN_MARK_MNTNS,
-			    FAN_MNT_ATTACH | FAN_MNT_DETACH, self->ns_fd, NULL);
-	ASSERT_EQ(ret, 0);
+	for (i = 0; i < NUM_FAN_FDS; i++) {
+		self->fan_fd[i] = fanotify_init(FAN_REPORT_MNT | FAN_NONBLOCK,
+						0);
+		ASSERT_GE(self->fan_fd[i], 0);
+		ret = fanotify_mark(self->fan_fd[i], FAN_MARK_ADD |
+				    FAN_MARK_MNTNS,
+				    FAN_MNT_ATTACH | FAN_MNT_DETACH,
+				    self->ns_fd, NULL);
+		ASSERT_EQ(ret, 0);
+		// On fd[0] we do an extra ADD that changes nothing.
+		// On fd[1]/fd[2] we REMOVE/FLUSH which removes the mark.
+		ret = fanotify_mark(self->fan_fd[i], mark_cmds[i] |
+				    FAN_MARK_MNTNS,
+				    FAN_MNT_ATTACH | FAN_MNT_DETACH,
+				    self->ns_fd, NULL);
+		ASSERT_EQ(ret, 0);
+	}
 
 	self->rem = 0;
 }
 
 FIXTURE_TEARDOWN(fanotify)
 {
+	int i;
+
 	ASSERT_EQ(self->rem, 0);
-	close(self->fan_fd);
+	for (i = 0; i < NUM_FAN_FDS; i++)
+		close(self->fan_fd[i]);
 
 	ASSERT_EQ(fchdir(self->orig_root), 0);
 
@@ -123,8 +145,21 @@ static uint64_t expect_notify(struct __test_metadata *const _metadata,
 	unsigned int thislen;
 
 	if (!self->rem) {
-		ssize_t len = read(self->fan_fd, self->buf, sizeof(self->buf));
-		ASSERT_GT(len, 0);
+		ssize_t len;
+		int i;
+
+		for (i = NUM_FAN_FDS - 1; i >= 0; i--) {
+			len = read(self->fan_fd[i], self->buf,
+				   sizeof(self->buf));
+			if (i > 0) {
+				// Groups 1,2 should get EAGAIN
+				ASSERT_EQ(len, -1);
+				ASSERT_EQ(errno, EAGAIN);
+			} else {
+				// Group 0 should get events
+				ASSERT_GT(len, 0);
+			}
+		}
 
 		self->rem = len;
 		self->next = (void *) self->buf;
diff --git a/tools/testing/selftests/ftrace/test.d/dynevent/dynevent_limitations.tc b/tools/testing/selftests/ftrace/test.d/dynevent/dynevent_limitations.tc
index 6b94b678741a..f656bccb1a14 100644
--- a/tools/testing/selftests/ftrace/test.d/dynevent/dynevent_limitations.tc
+++ b/tools/testing/selftests/ftrace/test.d/dynevent/dynevent_limitations.tc
@@ -7,11 +7,32 @@
 MAX_ARGS=128
 EXCEED_ARGS=$((MAX_ARGS + 1))
 
+# bash and dash evaluate variables differently.
+# dash will evaluate '\\' every time it is read whereas bash does not.
+#
+#   TEST_STRING="$TEST_STRING \\$i"
+#   echo $TEST_STRING
+#
+# With i=123
+# On bash, that will print "\123"
+# but on dash, that will print the escape sequence of \123 as the \ will
+# be interpreted again in the echo.
+#
+# Set a variable "bs" to save a double backslash, then echo that
+# to "ts" to see if $ts changed or not. If it changed, it's dash,
+# if not, it's bash, and then bs can equal a single backslash.
+bs='\\'
+ts=`echo $bs`
+if [ "$ts" = '\\' ]; then
+  # this is bash
+  bs='\'
+fi
+
 check_max_args() { # event_header
   TEST_STRING=$1
   # Acceptable
   for i in `seq 1 $MAX_ARGS`; do
-    TEST_STRING="$TEST_STRING \\$i"
+    TEST_STRING="$TEST_STRING $bs$i"
   done
   echo "$TEST_STRING" >> dynamic_events
   echo > dynamic_events
diff --git a/tools/testing/selftests/ftrace/test.d/filter/event-filter-function.tc b/tools/testing/selftests/ftrace/test.d/filter/event-filter-function.tc
index 118247b8dd84..c62165fabd0c 100644
--- a/tools/testing/selftests/ftrace/test.d/filter/event-filter-function.tc
+++ b/tools/testing/selftests/ftrace/test.d/filter/event-filter-function.tc
@@ -80,6 +80,26 @@ if [ $misscnt -gt 0 ]; then
 	exit_fail
 fi
 
+# Check strings too
+if [ -f events/syscalls/sys_enter_openat/filter ]; then
+	DIRNAME=`basename $TMPDIR`
+	echo "filename.ustring ~ \"*$DIRNAME*\"" > events/syscalls/sys_enter_openat/filter
+	echo 1 > events/syscalls/sys_enter_openat/enable
+	echo 1 > tracing_on
+	ls /bin/sh
+	nocnt=`grep openat trace | wc -l`
+	ls $TMPDIR
+	echo 0 > tracing_on
+	hitcnt=`grep openat trace | wc -l`;
+	echo 0 > events/syscalls/sys_enter_openat/enable
+	if [ $nocnt -gt 0 ]; then
+		exit_fail
+	fi
+	if [ $hitcnt -eq 0 ]; then
+		exit_fail
+	fi
+fi
+
 reset_events_filter
 
 exit 0
diff --git a/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-multi-filter.tc b/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-multi-filter.tc
new file mode 100644
index 000000000000..b6d6a312ead5
--- /dev/null
+++ b/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-multi-filter.tc
@@ -0,0 +1,177 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+# description: ftrace - function graph filters
+# requires: set_ftrace_filter function_graph:tracer
+
+# Make sure that function graph filtering works
+
+INSTANCE1="instances/test1_$$"
+INSTANCE2="instances/test2_$$"
+
+WD=`pwd`
+
+do_reset() {
+    cd $WD
+    if [ -d $INSTANCE1 ]; then
+	echo nop > $INSTANCE1/current_tracer
+	rmdir $INSTANCE1
+    fi
+    if [ -d $INSTANCE2 ]; then
+	echo nop > $INSTANCE2/current_tracer
+	rmdir $INSTANCE2
+    fi
+}
+
+mkdir $INSTANCE1
+if ! grep -q function_graph $INSTANCE1/available_tracers; then
+    echo "function_graph not allowed with instances"
+    rmdir $INSTANCE1
+    exit_unsupported
+fi
+
+mkdir $INSTANCE2
+
+fail() { # msg
+    do_reset
+    echo $1
+    exit_fail
+}
+
+disable_tracing
+clear_trace
+
+function_count() {
+    search=$1
+    vsearch=$2
+
+    if [ -z "$search" ]; then
+	cat enabled_functions | wc -l
+    elif [ -z "$vsearch" ]; then
+	grep $search enabled_functions | wc -l
+    else
+	grep $search enabled_functions | grep $vsearch| wc -l
+    fi
+}
+
+set_fgraph() {
+    instance=$1
+    filter="$2"
+    notrace="$3"
+
+    echo "$filter" > $instance/set_ftrace_filter
+    echo "$notrace" > $instance/set_ftrace_notrace
+    echo function_graph > $instance/current_tracer
+}
+
+check_functions() {
+    orig_cnt=$1
+    test=$2
+
+    cnt=`function_count $test`
+    if [ $cnt -gt $orig_cnt ]; then
+	fail
+    fi
+}
+
+check_cnt() {
+    orig_cnt=$1
+    search=$2
+    vsearch=$3
+
+    cnt=`function_count $search $vsearch`
+    if [ $cnt -gt $orig_cnt ]; then
+	fail
+    fi
+}
+
+reset_graph() {
+    instance=$1
+    echo nop > $instance/current_tracer
+}
+
+# get any functions that were enabled before the test
+total_cnt=`function_count`
+sched_cnt=`function_count sched`
+lock_cnt=`function_count lock`
+time_cnt=`function_count time`
+clock_cnt=`function_count clock`
+locks_clock_cnt=`function_count locks clock`
+clock_locks_cnt=`function_count clock locks`
+
+# Trace functions with "sched" but not "time"
+set_fgraph $INSTANCE1 '*sched*' '*time*'
+
+# Make sure "time" isn't listed
+check_functions $time_cnt 'time'
+instance1_cnt=`function_count`
+
+# Trace functions with "lock" but not "clock"
+set_fgraph $INSTANCE2 '*lock*' '*clock*'
+instance1_2_cnt=`function_count`
+
+# Turn off the first instance
+reset_graph $INSTANCE1
+
+# The second instance doesn't trace "clock" functions
+check_functions $clock_cnt 'clock'
+instance2_cnt=`function_count`
+
+# Start from a clean slate
+reset_graph $INSTANCE2
+check_functions $total_cnt
+
+# Trace functions with "lock" but not "clock"
+set_fgraph $INSTANCE2 '*lock*' '*clock*'
+
+# This should match the last time instance 2 was by itself
+cnt=`function_count`
+if [ $instance2_cnt -ne $cnt ]; then
+    fail
+fi
+
+# And it should not be tracing "clock" functions
+check_functions $clock_cnt 'clock'
+
+# Trace functions with "sched" but not "time"
+set_fgraph $INSTANCE1 '*sched*' '*time*'
+
+# This should match the last time both instances were enabled
+cnt=`function_count`
+if [ $instance1_2_cnt -ne $cnt ]; then
+    fail
+fi
+
+# Turn off the second instance
+reset_graph $INSTANCE2
+
+# This should match the last time instance 1 was by itself
+cnt=`function_count`
+if [ $instance1_cnt -ne $cnt ]; then
+    fail
+fi
+
+# And it should not be tracing "time" functions
+check_functions $time_cnt 'time'
+
+# Start from a clean slate
+reset_graph $INSTANCE1
+check_functions $total_cnt
+
+# Enable all functions but those that have "locks"
+set_fgraph $INSTANCE1 '' '*locks*'
+
+# Enable all functions but those that have "clock"
+set_fgraph $INSTANCE2 '' '*clock*'
+
+# If a function has "locks" it should not have "clock"
+check_cnt $locks_clock_cnt locks clock
+
+# If a function has "clock" it should not have "locks"
+check_cnt $clock_locks_cnt clock locks
+
+reset_graph $INSTANCE1
+reset_graph $INSTANCE2
+
+do_reset
+
+exit 0
diff --git a/tools/testing/selftests/futex/functional/futex_wait_wouldblock.c b/tools/testing/selftests/futex/functional/futex_wait_wouldblock.c
index 7d7a6a06cdb7..2d8230da9064 100644
--- a/tools/testing/selftests/futex/functional/futex_wait_wouldblock.c
+++ b/tools/testing/selftests/futex/functional/futex_wait_wouldblock.c
@@ -98,7 +98,7 @@ int main(int argc, char *argv[])
 	info("Calling futex_waitv on f1: %u @ %p with val=%u\n", f1, &f1, f1+1);
 	res = futex_waitv(&waitv, 1, 0, &to, CLOCK_MONOTONIC);
 	if (!res || errno != EWOULDBLOCK) {
-		ksft_test_result_pass("futex_waitv returned: %d %s\n",
+		ksft_test_result_fail("futex_waitv returned: %d %s\n",
 				      res ? errno : res,
 				      res ? strerror(errno) : "");
 		ret = RET_FAIL;
diff --git a/tools/testing/selftests/hid/config.common b/tools/testing/selftests/hid/config.common
index 45b5570441ce..b1f40857307d 100644
--- a/tools/testing/selftests/hid/config.common
+++ b/tools/testing/selftests/hid/config.common
@@ -39,7 +39,6 @@ CONFIG_CPU_FREQ_GOV_USERSPACE=y
 CONFIG_CPU_FREQ_STAT=y
 CONFIG_CPU_IDLE_GOV_LADDER=y
 CONFIG_CPUSETS=y
-CONFIG_CRC_T10DIF=y
 CONFIG_CRYPTO_BLAKE2B=y
 CONFIG_CRYPTO_DEV_VIRTIO=y
 CONFIG_CRYPTO_SEQIV=y
diff --git a/tools/testing/selftests/kvm/Makefile.kvm b/tools/testing/selftests/kvm/Makefile.kvm
index f773f8f99249..f62b0a5aba35 100644
--- a/tools/testing/selftests/kvm/Makefile.kvm
+++ b/tools/testing/selftests/kvm/Makefile.kvm
@@ -50,8 +50,18 @@ LIBKVM_riscv += lib/riscv/ucall.c
 # Non-compiled test targets
 TEST_PROGS_x86 += x86/nx_huge_pages_test.sh
 
+# Compiled test targets valid on all architectures with libkvm support
+TEST_GEN_PROGS_COMMON = demand_paging_test
+TEST_GEN_PROGS_COMMON += dirty_log_test
+TEST_GEN_PROGS_COMMON += guest_print_test
+TEST_GEN_PROGS_COMMON += kvm_binary_stats_test
+TEST_GEN_PROGS_COMMON += kvm_create_max_vcpus
+TEST_GEN_PROGS_COMMON += kvm_page_table_test
+TEST_GEN_PROGS_COMMON += set_memory_region_test
+
 # Compiled test targets
-TEST_GEN_PROGS_x86 = x86/cpuid_test
+TEST_GEN_PROGS_x86 = $(TEST_GEN_PROGS_COMMON)
+TEST_GEN_PROGS_x86 += x86/cpuid_test
 TEST_GEN_PROGS_x86 += x86/cr4_cpuid_sync_test
 TEST_GEN_PROGS_x86 += x86/dirty_log_page_splitting_test
 TEST_GEN_PROGS_x86 += x86/feature_msrs_test
@@ -119,27 +129,21 @@ TEST_GEN_PROGS_x86 += x86/triple_fault_event_test
 TEST_GEN_PROGS_x86 += x86/recalc_apic_map_test
 TEST_GEN_PROGS_x86 += access_tracking_perf_test
 TEST_GEN_PROGS_x86 += coalesced_io_test
-TEST_GEN_PROGS_x86 += demand_paging_test
-TEST_GEN_PROGS_x86 += dirty_log_test
 TEST_GEN_PROGS_x86 += dirty_log_perf_test
 TEST_GEN_PROGS_x86 += guest_memfd_test
-TEST_GEN_PROGS_x86 += guest_print_test
 TEST_GEN_PROGS_x86 += hardware_disable_test
-TEST_GEN_PROGS_x86 += kvm_create_max_vcpus
-TEST_GEN_PROGS_x86 += kvm_page_table_test
 TEST_GEN_PROGS_x86 += memslot_modification_stress_test
 TEST_GEN_PROGS_x86 += memslot_perf_test
 TEST_GEN_PROGS_x86 += mmu_stress_test
 TEST_GEN_PROGS_x86 += rseq_test
-TEST_GEN_PROGS_x86 += set_memory_region_test
 TEST_GEN_PROGS_x86 += steal_time
-TEST_GEN_PROGS_x86 += kvm_binary_stats_test
 TEST_GEN_PROGS_x86 += system_counter_offset_test
 TEST_GEN_PROGS_x86 += pre_fault_memory_test
 
 # Compiled outputs used by test targets
 TEST_GEN_PROGS_EXTENDED_x86 += x86/nx_huge_pages_test
 
+TEST_GEN_PROGS_arm64 = $(TEST_GEN_PROGS_COMMON)
 TEST_GEN_PROGS_arm64 += arm64/aarch32_id_regs
 TEST_GEN_PROGS_arm64 += arm64/arch_timer_edge_cases
 TEST_GEN_PROGS_arm64 += arm64/debug-exceptions
@@ -158,22 +162,16 @@ TEST_GEN_PROGS_arm64 += arm64/no-vgic-v3
 TEST_GEN_PROGS_arm64 += access_tracking_perf_test
 TEST_GEN_PROGS_arm64 += arch_timer
 TEST_GEN_PROGS_arm64 += coalesced_io_test
-TEST_GEN_PROGS_arm64 += demand_paging_test
-TEST_GEN_PROGS_arm64 += dirty_log_test
 TEST_GEN_PROGS_arm64 += dirty_log_perf_test
-TEST_GEN_PROGS_arm64 += guest_print_test
 TEST_GEN_PROGS_arm64 += get-reg-list
-TEST_GEN_PROGS_arm64 += kvm_create_max_vcpus
-TEST_GEN_PROGS_arm64 += kvm_page_table_test
 TEST_GEN_PROGS_arm64 += memslot_modification_stress_test
 TEST_GEN_PROGS_arm64 += memslot_perf_test
 TEST_GEN_PROGS_arm64 += mmu_stress_test
 TEST_GEN_PROGS_arm64 += rseq_test
-TEST_GEN_PROGS_arm64 += set_memory_region_test
 TEST_GEN_PROGS_arm64 += steal_time
-TEST_GEN_PROGS_arm64 += kvm_binary_stats_test
 
-TEST_GEN_PROGS_s390 = s390/memop
+TEST_GEN_PROGS_s390 = $(TEST_GEN_PROGS_COMMON)
+TEST_GEN_PROGS_s390 += s390/memop
 TEST_GEN_PROGS_s390 += s390/resets
 TEST_GEN_PROGS_s390 += s390/sync_regs_test
 TEST_GEN_PROGS_s390 += s390/tprot
@@ -182,27 +180,14 @@ TEST_GEN_PROGS_s390 += s390/debug_test
 TEST_GEN_PROGS_s390 += s390/cpumodel_subfuncs_test
 TEST_GEN_PROGS_s390 += s390/shared_zeropage_test
 TEST_GEN_PROGS_s390 += s390/ucontrol_test
-TEST_GEN_PROGS_s390 += demand_paging_test
-TEST_GEN_PROGS_s390 += dirty_log_test
-TEST_GEN_PROGS_s390 += guest_print_test
-TEST_GEN_PROGS_s390 += kvm_create_max_vcpus
-TEST_GEN_PROGS_s390 += kvm_page_table_test
 TEST_GEN_PROGS_s390 += rseq_test
-TEST_GEN_PROGS_s390 += set_memory_region_test
-TEST_GEN_PROGS_s390 += kvm_binary_stats_test
 
+TEST_GEN_PROGS_riscv = $(TEST_GEN_PROGS_COMMON)
 TEST_GEN_PROGS_riscv += riscv/sbi_pmu_test
 TEST_GEN_PROGS_riscv += riscv/ebreak_test
 TEST_GEN_PROGS_riscv += arch_timer
 TEST_GEN_PROGS_riscv += coalesced_io_test
-TEST_GEN_PROGS_riscv += demand_paging_test
-TEST_GEN_PROGS_riscv += dirty_log_test
 TEST_GEN_PROGS_riscv += get-reg-list
-TEST_GEN_PROGS_riscv += guest_print_test
-TEST_GEN_PROGS_riscv += kvm_binary_stats_test
-TEST_GEN_PROGS_riscv += kvm_create_max_vcpus
-TEST_GEN_PROGS_riscv += kvm_page_table_test
-TEST_GEN_PROGS_riscv += set_memory_region_test
 TEST_GEN_PROGS_riscv += steal_time
 
 SPLIT_TESTS += arch_timer
diff --git a/tools/testing/selftests/kvm/arm64/page_fault_test.c b/tools/testing/selftests/kvm/arm64/page_fault_test.c
index ec33a8f9c908..dc6559dad9d8 100644
--- a/tools/testing/selftests/kvm/arm64/page_fault_test.c
+++ b/tools/testing/selftests/kvm/arm64/page_fault_test.c
@@ -199,7 +199,7 @@ static bool guest_set_ha(void)
 	if (hadbs == 0)
 		return false;
 
-	tcr = read_sysreg(tcr_el1) | TCR_EL1_HA;
+	tcr = read_sysreg(tcr_el1) | TCR_HA;
 	write_sysreg(tcr, tcr_el1);
 	isb();
 
diff --git a/tools/testing/selftests/kvm/include/arm64/processor.h b/tools/testing/selftests/kvm/include/arm64/processor.h
index 1e8d0d531fbd..b0fc0f945766 100644
--- a/tools/testing/selftests/kvm/include/arm64/processor.h
+++ b/tools/testing/selftests/kvm/include/arm64/processor.h
@@ -62,6 +62,67 @@
 	 MAIR_ATTRIDX(MAIR_ATTR_NORMAL, MT_NORMAL) |				\
 	 MAIR_ATTRIDX(MAIR_ATTR_NORMAL_WT, MT_NORMAL_WT))
 
+/* TCR_EL1 specific flags */
+#define TCR_T0SZ_OFFSET	0
+#define TCR_T0SZ(x)		((UL(64) - (x)) << TCR_T0SZ_OFFSET)
+
+#define TCR_IRGN0_SHIFT	8
+#define TCR_IRGN0_MASK		(UL(3) << TCR_IRGN0_SHIFT)
+#define TCR_IRGN0_NC		(UL(0) << TCR_IRGN0_SHIFT)
+#define TCR_IRGN0_WBWA		(UL(1) << TCR_IRGN0_SHIFT)
+#define TCR_IRGN0_WT		(UL(2) << TCR_IRGN0_SHIFT)
+#define TCR_IRGN0_WBnWA	(UL(3) << TCR_IRGN0_SHIFT)
+
+#define TCR_ORGN0_SHIFT	10
+#define TCR_ORGN0_MASK		(UL(3) << TCR_ORGN0_SHIFT)
+#define TCR_ORGN0_NC		(UL(0) << TCR_ORGN0_SHIFT)
+#define TCR_ORGN0_WBWA		(UL(1) << TCR_ORGN0_SHIFT)
+#define TCR_ORGN0_WT		(UL(2) << TCR_ORGN0_SHIFT)
+#define TCR_ORGN0_WBnWA	(UL(3) << TCR_ORGN0_SHIFT)
+
+#define TCR_SH0_SHIFT		12
+#define TCR_SH0_MASK		(UL(3) << TCR_SH0_SHIFT)
+#define TCR_SH0_INNER		(UL(3) << TCR_SH0_SHIFT)
+
+#define TCR_TG0_SHIFT		14
+#define TCR_TG0_MASK		(UL(3) << TCR_TG0_SHIFT)
+#define TCR_TG0_4K		(UL(0) << TCR_TG0_SHIFT)
+#define TCR_TG0_64K		(UL(1) << TCR_TG0_SHIFT)
+#define TCR_TG0_16K		(UL(2) << TCR_TG0_SHIFT)
+
+#define TCR_IPS_SHIFT		32
+#define TCR_IPS_MASK		(UL(7) << TCR_IPS_SHIFT)
+#define TCR_IPS_52_BITS	(UL(6) << TCR_IPS_SHIFT)
+#define TCR_IPS_48_BITS	(UL(5) << TCR_IPS_SHIFT)
+#define TCR_IPS_40_BITS	(UL(2) << TCR_IPS_SHIFT)
+#define TCR_IPS_36_BITS	(UL(1) << TCR_IPS_SHIFT)
+
+#define TCR_HA			(UL(1) << 39)
+#define TCR_DS			(UL(1) << 59)
+
+/*
+ * AttrIndx[2:0] encoding (mapping attributes defined in the MAIR* registers).
+ */
+#define PTE_ATTRINDX(t)	((t) << 2)
+#define PTE_ATTRINDX_MASK	GENMASK(4, 2)
+#define PTE_ATTRINDX_SHIFT	2
+
+#define PTE_VALID		BIT(0)
+#define PGD_TYPE_TABLE		BIT(1)
+#define PUD_TYPE_TABLE		BIT(1)
+#define PMD_TYPE_TABLE		BIT(1)
+#define PTE_TYPE_PAGE		BIT(1)
+
+#define PTE_SHARED		(UL(3) << 8) /* SH[1:0], inner shareable */
+#define PTE_AF			BIT(10)
+
+#define PTE_ADDR_MASK(page_shift)	GENMASK(47, (page_shift))
+#define PTE_ADDR_51_48			GENMASK(15, 12)
+#define PTE_ADDR_51_48_SHIFT		12
+#define PTE_ADDR_MASK_LPA2(page_shift)	GENMASK(49, (page_shift))
+#define PTE_ADDR_51_50_LPA2		GENMASK(9, 8)
+#define PTE_ADDR_51_50_LPA2_SHIFT	8
+
 void aarch64_vcpu_setup(struct kvm_vcpu *vcpu, struct kvm_vcpu_init *init);
 struct kvm_vcpu *aarch64_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id,
 				  struct kvm_vcpu_init *init, void *guest_code);
@@ -102,12 +163,6 @@ enum {
 			   (v) == VECTOR_SYNC_LOWER_64    || \
 			   (v) == VECTOR_SYNC_LOWER_32)
 
-/* Access flag */
-#define PTE_AF			(1ULL << 10)
-
-/* Access flag update enable/disable */
-#define TCR_EL1_HA		(1ULL << 39)
-
 void aarch64_get_supported_page_sizes(uint32_t ipa, uint32_t *ipa4k,
 					uint32_t *ipa16k, uint32_t *ipa64k);
 
diff --git a/tools/testing/selftests/kvm/lib/arm64/processor.c b/tools/testing/selftests/kvm/lib/arm64/processor.c
index 7ba3aa3755f3..9d69904cb608 100644
--- a/tools/testing/selftests/kvm/lib/arm64/processor.c
+++ b/tools/testing/selftests/kvm/lib/arm64/processor.c
@@ -72,13 +72,13 @@ static uint64_t addr_pte(struct kvm_vm *vm, uint64_t pa, uint64_t attrs)
 	uint64_t pte;
 
 	if (use_lpa2_pte_format(vm)) {
-		pte = pa & GENMASK(49, vm->page_shift);
-		pte |= FIELD_GET(GENMASK(51, 50), pa) << 8;
-		attrs &= ~GENMASK(9, 8);
+		pte = pa & PTE_ADDR_MASK_LPA2(vm->page_shift);
+		pte |= FIELD_GET(GENMASK(51, 50), pa) << PTE_ADDR_51_50_LPA2_SHIFT;
+		attrs &= ~PTE_ADDR_51_50_LPA2;
 	} else {
-		pte = pa & GENMASK(47, vm->page_shift);
+		pte = pa & PTE_ADDR_MASK(vm->page_shift);
 		if (vm->page_shift == 16)
-			pte |= FIELD_GET(GENMASK(51, 48), pa) << 12;
+			pte |= FIELD_GET(GENMASK(51, 48), pa) << PTE_ADDR_51_48_SHIFT;
 	}
 	pte |= attrs;
 
@@ -90,12 +90,12 @@ static uint64_t pte_addr(struct kvm_vm *vm, uint64_t pte)
 	uint64_t pa;
 
 	if (use_lpa2_pte_format(vm)) {
-		pa = pte & GENMASK(49, vm->page_shift);
-		pa |= FIELD_GET(GENMASK(9, 8), pte) << 50;
+		pa = pte & PTE_ADDR_MASK_LPA2(vm->page_shift);
+		pa |= FIELD_GET(PTE_ADDR_51_50_LPA2, pte) << 50;
 	} else {
-		pa = pte & GENMASK(47, vm->page_shift);
+		pa = pte & PTE_ADDR_MASK(vm->page_shift);
 		if (vm->page_shift == 16)
-			pa |= FIELD_GET(GENMASK(15, 12), pte) << 48;
+			pa |= FIELD_GET(PTE_ADDR_51_48, pte) << 48;
 	}
 
 	return pa;
@@ -128,7 +128,8 @@ void virt_arch_pgd_alloc(struct kvm_vm *vm)
 static void _virt_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 			 uint64_t flags)
 {
-	uint8_t attr_idx = flags & 7;
+	uint8_t attr_idx = flags & (PTE_ATTRINDX_MASK >> PTE_ATTRINDX_SHIFT);
+	uint64_t pg_attr;
 	uint64_t *ptep;
 
 	TEST_ASSERT((vaddr % vm->page_size) == 0,
@@ -147,18 +148,21 @@ static void _virt_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 
 	ptep = addr_gpa2hva(vm, vm->pgd) + pgd_index(vm, vaddr) * 8;
 	if (!*ptep)
-		*ptep = addr_pte(vm, vm_alloc_page_table(vm), 3);
+		*ptep = addr_pte(vm, vm_alloc_page_table(vm),
+				 PGD_TYPE_TABLE | PTE_VALID);
 
 	switch (vm->pgtable_levels) {
 	case 4:
 		ptep = addr_gpa2hva(vm, pte_addr(vm, *ptep)) + pud_index(vm, vaddr) * 8;
 		if (!*ptep)
-			*ptep = addr_pte(vm, vm_alloc_page_table(vm), 3);
+			*ptep = addr_pte(vm, vm_alloc_page_table(vm),
+					 PUD_TYPE_TABLE | PTE_VALID);
 		/* fall through */
 	case 3:
 		ptep = addr_gpa2hva(vm, pte_addr(vm, *ptep)) + pmd_index(vm, vaddr) * 8;
 		if (!*ptep)
-			*ptep = addr_pte(vm, vm_alloc_page_table(vm), 3);
+			*ptep = addr_pte(vm, vm_alloc_page_table(vm),
+					 PMD_TYPE_TABLE | PTE_VALID);
 		/* fall through */
 	case 2:
 		ptep = addr_gpa2hva(vm, pte_addr(vm, *ptep)) + pte_index(vm, vaddr) * 8;
@@ -167,7 +171,11 @@ static void _virt_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 		TEST_FAIL("Page table levels must be 2, 3, or 4");
 	}
 
-	*ptep = addr_pte(vm, paddr, (attr_idx << 2) | (1 << 10) | 3);  /* AF */
+	pg_attr = PTE_AF | PTE_ATTRINDX(attr_idx) | PTE_TYPE_PAGE | PTE_VALID;
+	if (!use_lpa2_pte_format(vm))
+		pg_attr |= PTE_SHARED;
+
+	*ptep = addr_pte(vm, paddr, pg_attr);
 }
 
 void virt_arch_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr)
@@ -293,20 +301,20 @@ void aarch64_vcpu_setup(struct kvm_vcpu *vcpu, struct kvm_vcpu_init *init)
 	case VM_MODE_P48V48_64K:
 	case VM_MODE_P40V48_64K:
 	case VM_MODE_P36V48_64K:
-		tcr_el1 |= 1ul << 14; /* TG0 = 64KB */
+		tcr_el1 |= TCR_TG0_64K;
 		break;
 	case VM_MODE_P52V48_16K:
 	case VM_MODE_P48V48_16K:
 	case VM_MODE_P40V48_16K:
 	case VM_MODE_P36V48_16K:
 	case VM_MODE_P36V47_16K:
-		tcr_el1 |= 2ul << 14; /* TG0 = 16KB */
+		tcr_el1 |= TCR_TG0_16K;
 		break;
 	case VM_MODE_P52V48_4K:
 	case VM_MODE_P48V48_4K:
 	case VM_MODE_P40V48_4K:
 	case VM_MODE_P36V48_4K:
-		tcr_el1 |= 0ul << 14; /* TG0 = 4KB */
+		tcr_el1 |= TCR_TG0_4K;
 		break;
 	default:
 		TEST_FAIL("Unknown guest mode, mode: 0x%x", vm->mode);
@@ -319,35 +327,35 @@ void aarch64_vcpu_setup(struct kvm_vcpu *vcpu, struct kvm_vcpu_init *init)
 	case VM_MODE_P52V48_4K:
 	case VM_MODE_P52V48_16K:
 	case VM_MODE_P52V48_64K:
-		tcr_el1 |= 6ul << 32; /* IPS = 52 bits */
+		tcr_el1 |= TCR_IPS_52_BITS;
 		ttbr0_el1 |= FIELD_GET(GENMASK(51, 48), vm->pgd) << 2;
 		break;
 	case VM_MODE_P48V48_4K:
 	case VM_MODE_P48V48_16K:
 	case VM_MODE_P48V48_64K:
-		tcr_el1 |= 5ul << 32; /* IPS = 48 bits */
+		tcr_el1 |= TCR_IPS_48_BITS;
 		break;
 	case VM_MODE_P40V48_4K:
 	case VM_MODE_P40V48_16K:
 	case VM_MODE_P40V48_64K:
-		tcr_el1 |= 2ul << 32; /* IPS = 40 bits */
+		tcr_el1 |= TCR_IPS_40_BITS;
 		break;
 	case VM_MODE_P36V48_4K:
 	case VM_MODE_P36V48_16K:
 	case VM_MODE_P36V48_64K:
 	case VM_MODE_P36V47_16K:
-		tcr_el1 |= 1ul << 32; /* IPS = 36 bits */
+		tcr_el1 |= TCR_IPS_36_BITS;
 		break;
 	default:
 		TEST_FAIL("Unknown guest mode, mode: 0x%x", vm->mode);
 	}
 
-	sctlr_el1 |= (1 << 0) | (1 << 2) | (1 << 12) /* M | C | I */;
-	/* TCR_EL1 |= IRGN0:WBWA | ORGN0:WBWA | SH0:Inner-Shareable */;
-	tcr_el1 |= (1 << 8) | (1 << 10) | (3 << 12);
-	tcr_el1 |= (64 - vm->va_bits) /* T0SZ */;
+	sctlr_el1 |= SCTLR_ELx_M | SCTLR_ELx_C | SCTLR_ELx_I;
+
+	tcr_el1 |= TCR_IRGN0_WBWA | TCR_ORGN0_WBWA | TCR_SH0_INNER;
+	tcr_el1 |= TCR_T0SZ(vm->va_bits);
 	if (use_lpa2_pte_format(vm))
-		tcr_el1 |= (1ul << 59) /* DS */;
+		tcr_el1 |= TCR_DS;
 
 	vcpu_set_reg(vcpu, KVM_ARM64_SYS_REG(SYS_SCTLR_EL1), sctlr_el1);
 	vcpu_set_reg(vcpu, KVM_ARM64_SYS_REG(SYS_TCR_EL1), tcr_el1);
diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c
index 279ad8946040..815bc45dd8dc 100644
--- a/tools/testing/selftests/kvm/lib/kvm_util.c
+++ b/tools/testing/selftests/kvm/lib/kvm_util.c
@@ -2019,9 +2019,8 @@ static struct exit_reason {
 	KVM_EXIT_STRING(RISCV_SBI),
 	KVM_EXIT_STRING(RISCV_CSR),
 	KVM_EXIT_STRING(NOTIFY),
-#ifdef KVM_EXIT_MEMORY_NOT_PRESENT
-	KVM_EXIT_STRING(MEMORY_NOT_PRESENT),
-#endif
+	KVM_EXIT_STRING(LOONGARCH_IOCSR),
+	KVM_EXIT_STRING(MEMORY_FAULT),
 };
 
 /*
diff --git a/tools/testing/selftests/kvm/rseq_test.c b/tools/testing/selftests/kvm/rseq_test.c
index e5898678bfab..1375fca80bcd 100644
--- a/tools/testing/selftests/kvm/rseq_test.c
+++ b/tools/testing/selftests/kvm/rseq_test.c
@@ -196,25 +196,27 @@ static void calc_min_max_cpu(void)
 static void help(const char *name)
 {
 	puts("");
-	printf("usage: %s [-h] [-u]\n", name);
+	printf("usage: %s [-h] [-u] [-l latency]\n", name);
 	printf(" -u: Don't sanity check the number of successful KVM_RUNs\n");
+	printf(" -l: Set /dev/cpu_dma_latency to suppress deep sleep states\n");
 	puts("");
 	exit(0);
 }
 
 int main(int argc, char *argv[])
 {
+	int r, i, snapshot, opt, fd = -1, latency = -1;
 	bool skip_sanity_check = false;
-	int r, i, snapshot;
 	struct kvm_vm *vm;
 	struct kvm_vcpu *vcpu;
 	u32 cpu, rseq_cpu;
-	int opt;
 
-	while ((opt = getopt(argc, argv, "hu")) != -1) {
+	while ((opt = getopt(argc, argv, "hl:u")) != -1) {
 		switch (opt) {
 		case 'u':
 			skip_sanity_check = true;
+		case 'l':
+			latency = atoi_paranoid(optarg);
 			break;
 		case 'h':
 		default:
@@ -243,6 +245,20 @@ int main(int argc, char *argv[])
 	pthread_create(&migration_thread, NULL, migration_worker,
 		       (void *)(unsigned long)syscall(SYS_gettid));
 
+	if (latency >= 0) {
+		/*
+		 * Writes to cpu_dma_latency persist only while the file is
+		 * open, i.e. it allows userspace to provide guaranteed latency
+		 * while running a workload.  Keep the file open until the test
+		 * completes, otherwise writing cpu_dma_latency is meaningless.
+		 */
+		fd = open("/dev/cpu_dma_latency", O_RDWR);
+		TEST_ASSERT(fd >= 0, __KVM_SYSCALL_ERROR("open() /dev/cpu_dma_latency", fd));
+
+		r = write(fd, &latency, 4);
+		TEST_ASSERT(r >= 1, "Error setting /dev/cpu_dma_latency");
+	}
+
 	for (i = 0; !done; i++) {
 		vcpu_run(vcpu);
 		TEST_ASSERT(get_ucall(vcpu, NULL) == UCALL_SYNC,
@@ -278,6 +294,9 @@ int main(int argc, char *argv[])
 			    "rseq CPU = %d, sched CPU = %d", rseq_cpu, cpu);
 	}
 
+	if (fd > 0)
+		close(fd);
+
 	/*
 	 * Sanity check that the test was able to enter the guest a reasonable
 	 * number of times, e.g. didn't get stalled too often/long waiting for
@@ -293,8 +312,8 @@ int main(int argc, char *argv[])
 	TEST_ASSERT(skip_sanity_check || i > (NR_TASK_MIGRATIONS / 2),
 		    "Only performed %d KVM_RUNs, task stalled too much?\n\n"
 		    "  Try disabling deep sleep states to reduce CPU wakeup latency,\n"
-		    "  e.g. via cpuidle.off=1 or setting /dev/cpu_dma_latency to '0',\n"
-		    "  or run with -u to disable this sanity check.", i);
+		    "  e.g. via cpuidle.off=1 or via -l <latency>, or run with -u to\n"
+		    "  disable this sanity check.", i);
 
 	pthread_join(migration_thread, NULL);
 
diff --git a/tools/testing/selftests/kvm/x86/monitor_mwait_test.c b/tools/testing/selftests/kvm/x86/monitor_mwait_test.c
index 2b550eff35f1..390ae2d87493 100644
--- a/tools/testing/selftests/kvm/x86/monitor_mwait_test.c
+++ b/tools/testing/selftests/kvm/x86/monitor_mwait_test.c
@@ -7,6 +7,7 @@
 
 #include "kvm_util.h"
 #include "processor.h"
+#include "kselftest.h"
 
 #define CPUID_MWAIT (1u << 3)
 
@@ -14,6 +15,8 @@ enum monitor_mwait_testcases {
 	MWAIT_QUIRK_DISABLED = BIT(0),
 	MISC_ENABLES_QUIRK_DISABLED = BIT(1),
 	MWAIT_DISABLED = BIT(2),
+	CPUID_DISABLED = BIT(3),
+	TEST_MAX = CPUID_DISABLED * 2 - 1,
 };
 
 /*
@@ -35,11 +38,19 @@ do {									\
 			       testcase, vector);			\
 } while (0)
 
-static void guest_monitor_wait(int testcase)
+static void guest_monitor_wait(void *arg)
 {
+	int testcase = (int) (long) arg;
 	u8 vector;
 
-	GUEST_SYNC(testcase);
+	u64 val = rdmsr(MSR_IA32_MISC_ENABLE) & ~MSR_IA32_MISC_ENABLE_MWAIT;
+	if (!(testcase & MWAIT_DISABLED))
+		val |= MSR_IA32_MISC_ENABLE_MWAIT;
+	wrmsr(MSR_IA32_MISC_ENABLE, val);
+
+	__GUEST_ASSERT(this_cpu_has(X86_FEATURE_MWAIT) == !(testcase & MWAIT_DISABLED),
+		       "Expected CPUID.MWAIT %s\n",
+		       (testcase & MWAIT_DISABLED) ? "cleared" : "set");
 
 	/*
 	 * Arbitrarily MONITOR this function, SVM performs fault checks before
@@ -50,19 +61,6 @@ static void guest_monitor_wait(int testcase)
 
 	vector = kvm_asm_safe("mwait", "a"(guest_monitor_wait), "c"(0), "d"(0));
 	GUEST_ASSERT_MONITOR_MWAIT("MWAIT", testcase, vector);
-}
-
-static void guest_code(void)
-{
-	guest_monitor_wait(MWAIT_DISABLED);
-
-	guest_monitor_wait(MWAIT_QUIRK_DISABLED | MWAIT_DISABLED);
-
-	guest_monitor_wait(MISC_ENABLES_QUIRK_DISABLED | MWAIT_DISABLED);
-	guest_monitor_wait(MISC_ENABLES_QUIRK_DISABLED);
-
-	guest_monitor_wait(MISC_ENABLES_QUIRK_DISABLED | MWAIT_QUIRK_DISABLED | MWAIT_DISABLED);
-	guest_monitor_wait(MISC_ENABLES_QUIRK_DISABLED | MWAIT_QUIRK_DISABLED);
 
 	GUEST_DONE();
 }
@@ -74,56 +72,64 @@ int main(int argc, char *argv[])
 	struct kvm_vm *vm;
 	struct ucall uc;
 	int testcase;
+	char test[80];
 
-	TEST_REQUIRE(this_cpu_has(X86_FEATURE_MWAIT));
 	TEST_REQUIRE(kvm_has_cap(KVM_CAP_DISABLE_QUIRKS2));
 
-	vm = vm_create_with_one_vcpu(&vcpu, guest_code);
-	vcpu_clear_cpuid_feature(vcpu, X86_FEATURE_MWAIT);
+	ksft_print_header();
+	ksft_set_plan(12);
+	for (testcase = 0; testcase <= TEST_MAX; testcase++) {
+		vm = vm_create_with_one_vcpu(&vcpu, guest_monitor_wait);
+		vcpu_args_set(vcpu, 1, (void *)(long)testcase);
+
+		disabled_quirks = 0;
+		if (testcase & MWAIT_QUIRK_DISABLED) {
+			disabled_quirks |= KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS;
+			strcpy(test, "MWAIT can fault");
+		} else {
+			strcpy(test, "MWAIT never faults");
+		}
+		if (testcase & MISC_ENABLES_QUIRK_DISABLED) {
+			disabled_quirks |= KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT;
+			strcat(test, ", MISC_ENABLE updates CPUID");
+		} else {
+			strcat(test, ", no CPUID updates");
+		}
+
+		vm_enable_cap(vm, KVM_CAP_DISABLE_QUIRKS2, disabled_quirks);
+
+		if (!(testcase & MISC_ENABLES_QUIRK_DISABLED) &&
+		    (!!(testcase & CPUID_DISABLED) ^ !!(testcase & MWAIT_DISABLED)))
+			continue;
+
+		if (testcase & CPUID_DISABLED) {
+			strcat(test, ", CPUID clear");
+			vcpu_clear_cpuid_feature(vcpu, X86_FEATURE_MWAIT);
+		} else {
+			strcat(test, ", CPUID set");
+			vcpu_set_cpuid_feature(vcpu, X86_FEATURE_MWAIT);
+		}
+
+		if (testcase & MWAIT_DISABLED)
+			strcat(test, ", MWAIT disabled");
 
-	while (1) {
 		vcpu_run(vcpu);
 		TEST_ASSERT_KVM_EXIT_REASON(vcpu, KVM_EXIT_IO);
 
 		switch (get_ucall(vcpu, &uc)) {
-		case UCALL_SYNC:
-			testcase = uc.args[1];
-			break;
 		case UCALL_ABORT:
-			REPORT_GUEST_ASSERT(uc);
-			goto done;
+			/* Detected in vcpu_run */
+			break;
 		case UCALL_DONE:
-			goto done;
+			ksft_test_result_pass("%s\n", test);
+			break;
 		default:
 			TEST_FAIL("Unknown ucall %lu", uc.cmd);
-			goto done;
-		}
-
-		disabled_quirks = 0;
-		if (testcase & MWAIT_QUIRK_DISABLED)
-			disabled_quirks |= KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS;
-		if (testcase & MISC_ENABLES_QUIRK_DISABLED)
-			disabled_quirks |= KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT;
-		vm_enable_cap(vm, KVM_CAP_DISABLE_QUIRKS2, disabled_quirks);
-
-		/*
-		 * If the MISC_ENABLES quirk (KVM neglects to update CPUID to
-		 * enable/disable MWAIT) is disabled, toggle the ENABLE_MWAIT
-		 * bit in MISC_ENABLES accordingly.  If the quirk is enabled,
-		 * the only valid configuration is MWAIT disabled, as CPUID
-		 * can't be manually changed after running the vCPU.
-		 */
-		if (!(testcase & MISC_ENABLES_QUIRK_DISABLED)) {
-			TEST_ASSERT(testcase & MWAIT_DISABLED,
-				    "Can't toggle CPUID features after running vCPU");
-			continue;
+			break;
 		}
-
-		vcpu_set_msr(vcpu, MSR_IA32_MISC_ENABLE,
-			     (testcase & MWAIT_DISABLED) ? 0 : MSR_IA32_MISC_ENABLE_MWAIT);
+		kvm_vm_free(vm);
 	}
+	ksft_finished();
 
-done:
-	kvm_vm_free(vm);
 	return 0;
 }
diff --git a/tools/testing/selftests/landlock/audit.h b/tools/testing/selftests/landlock/audit.h
index b9054086a0c9..18a6014920b5 100644
--- a/tools/testing/selftests/landlock/audit.h
+++ b/tools/testing/selftests/landlock/audit.h
@@ -300,15 +300,22 @@ out:
 	return err;
 }
 
-static int __maybe_unused matches_log_domain_allocated(int audit_fd,
+static int __maybe_unused matches_log_domain_allocated(int audit_fd, pid_t pid,
 						       __u64 *domain_id)
 {
-	return audit_match_record(
-		audit_fd, AUDIT_LANDLOCK_DOMAIN,
-		REGEX_LANDLOCK_PREFIX
-		" status=allocated mode=enforcing pid=[0-9]\\+ uid=[0-9]\\+"
-		" exe=\"[^\"]\\+\" comm=\".*_test\"$",
-		domain_id);
+	static const char log_template[] = REGEX_LANDLOCK_PREFIX
+		" status=allocated mode=enforcing pid=%d uid=[0-9]\\+"
+		" exe=\"[^\"]\\+\" comm=\".*_test\"$";
+	char log_match[sizeof(log_template) + 10];
+	int log_match_len;
+
+	log_match_len =
+		snprintf(log_match, sizeof(log_match), log_template, pid);
+	if (log_match_len > sizeof(log_match))
+		return -E2BIG;
+
+	return audit_match_record(audit_fd, AUDIT_LANDLOCK_DOMAIN, log_match,
+				  domain_id);
 }
 
 static int __maybe_unused matches_log_domain_deallocated(
diff --git a/tools/testing/selftests/landlock/audit_test.c b/tools/testing/selftests/landlock/audit_test.c
index a0643070c403..cfc571afd0eb 100644
--- a/tools/testing/selftests/landlock/audit_test.c
+++ b/tools/testing/selftests/landlock/audit_test.c
@@ -9,6 +9,7 @@
 #include <errno.h>
 #include <limits.h>
 #include <linux/landlock.h>
+#include <pthread.h>
 #include <stdlib.h>
 #include <sys/mount.h>
 #include <sys/prctl.h>
@@ -40,7 +41,6 @@ FIXTURE(audit)
 {
 	struct audit_filter audit_filter;
 	int audit_fd;
-	__u64(*domain_stack)[16];
 };
 
 FIXTURE_SETUP(audit)
@@ -60,18 +60,10 @@ FIXTURE_SETUP(audit)
 		TH_LOG("Failed to initialize audit: %s", error_msg);
 	}
 	clear_cap(_metadata, CAP_AUDIT_CONTROL);
-
-	self->domain_stack = mmap(NULL, sizeof(*self->domain_stack),
-				  PROT_READ | PROT_WRITE,
-				  MAP_SHARED | MAP_ANONYMOUS, -1, 0);
-	ASSERT_NE(MAP_FAILED, self->domain_stack);
-	memset(self->domain_stack, 0, sizeof(*self->domain_stack));
 }
 
 FIXTURE_TEARDOWN(audit)
 {
-	EXPECT_EQ(0, munmap(self->domain_stack, sizeof(*self->domain_stack)));
-
 	set_cap(_metadata, CAP_AUDIT_CONTROL);
 	EXPECT_EQ(0, audit_cleanup(self->audit_fd, &self->audit_filter));
 	clear_cap(_metadata, CAP_AUDIT_CONTROL);
@@ -83,9 +75,15 @@ TEST_F(audit, layers)
 		.scoped = LANDLOCK_SCOPE_SIGNAL,
 	};
 	int status, ruleset_fd, i;
+	__u64(*domain_stack)[16];
 	__u64 prev_dom = 3;
 	pid_t child;
 
+	domain_stack = mmap(NULL, sizeof(*domain_stack), PROT_READ | PROT_WRITE,
+			    MAP_SHARED | MAP_ANONYMOUS, -1, 0);
+	ASSERT_NE(MAP_FAILED, domain_stack);
+	memset(domain_stack, 0, sizeof(*domain_stack));
+
 	ruleset_fd =
 		landlock_create_ruleset(&ruleset_attr, sizeof(ruleset_attr), 0);
 	ASSERT_LE(0, ruleset_fd);
@@ -94,7 +92,7 @@ TEST_F(audit, layers)
 	child = fork();
 	ASSERT_LE(0, child);
 	if (child == 0) {
-		for (i = 0; i < ARRAY_SIZE(*self->domain_stack); i++) {
+		for (i = 0; i < ARRAY_SIZE(*domain_stack); i++) {
 			__u64 denial_dom = 1;
 			__u64 allocated_dom = 2;
 
@@ -107,7 +105,8 @@ TEST_F(audit, layers)
 				  matches_log_signal(_metadata, self->audit_fd,
 						     getppid(), &denial_dom));
 			EXPECT_EQ(0, matches_log_domain_allocated(
-					     self->audit_fd, &allocated_dom));
+					     self->audit_fd, getpid(),
+					     &allocated_dom));
 			EXPECT_NE(denial_dom, 1);
 			EXPECT_NE(denial_dom, 0);
 			EXPECT_EQ(denial_dom, allocated_dom);
@@ -115,7 +114,7 @@ TEST_F(audit, layers)
 			/* Checks that the new domain is younger than the previous one. */
 			EXPECT_GT(allocated_dom, prev_dom);
 			prev_dom = allocated_dom;
-			(*self->domain_stack)[i] = allocated_dom;
+			(*domain_stack)[i] = allocated_dom;
 		}
 
 		/* Checks that we reached the maximum number of layers. */
@@ -142,23 +141,143 @@ TEST_F(audit, layers)
 	/* Purges log from deallocated domains. */
 	EXPECT_EQ(0, setsockopt(self->audit_fd, SOL_SOCKET, SO_RCVTIMEO,
 				&audit_tv_dom_drop, sizeof(audit_tv_dom_drop)));
-	for (i = ARRAY_SIZE(*self->domain_stack) - 1; i >= 0; i--) {
+	for (i = ARRAY_SIZE(*domain_stack) - 1; i >= 0; i--) {
 		__u64 deallocated_dom = 2;
 
 		EXPECT_EQ(0, matches_log_domain_deallocated(self->audit_fd, 1,
 							    &deallocated_dom));
-		EXPECT_EQ((*self->domain_stack)[i], deallocated_dom)
+		EXPECT_EQ((*domain_stack)[i], deallocated_dom)
 		{
 			TH_LOG("Failed to match domain %llx (#%d)",
-			       (*self->domain_stack)[i], i);
+			       (*domain_stack)[i], i);
 		}
 	}
+	EXPECT_EQ(0, munmap(domain_stack, sizeof(*domain_stack)));
 	EXPECT_EQ(0, setsockopt(self->audit_fd, SOL_SOCKET, SO_RCVTIMEO,
 				&audit_tv_default, sizeof(audit_tv_default)));
-
 	EXPECT_EQ(0, close(ruleset_fd));
 }
 
+struct thread_data {
+	pid_t parent_pid;
+	int ruleset_fd, pipe_child, pipe_parent;
+};
+
+static void *thread_audit_test(void *arg)
+{
+	const struct thread_data *data = (struct thread_data *)arg;
+	uintptr_t err = 0;
+	char buffer;
+
+	/* TGID and TID are different for a second thread. */
+	if (getpid() == gettid()) {
+		err = 1;
+		goto out;
+	}
+
+	if (landlock_restrict_self(data->ruleset_fd, 0)) {
+		err = 2;
+		goto out;
+	}
+
+	if (close(data->ruleset_fd)) {
+		err = 3;
+		goto out;
+	}
+
+	/* Creates a denial to get the domain ID. */
+	if (kill(data->parent_pid, 0) != -1) {
+		err = 4;
+		goto out;
+	}
+
+	if (EPERM != errno) {
+		err = 5;
+		goto out;
+	}
+
+	/* Signals the parent to read denial logs. */
+	if (write(data->pipe_child, ".", 1) != 1) {
+		err = 6;
+		goto out;
+	}
+
+	/* Waits for the parent to update audit filters. */
+	if (read(data->pipe_parent, &buffer, 1) != 1) {
+		err = 7;
+		goto out;
+	}
+
+out:
+	close(data->pipe_child);
+	close(data->pipe_parent);
+	return (void *)err;
+}
+
+/* Checks that the PID tied to a domain is not a TID but the TGID. */
+TEST_F(audit, thread)
+{
+	const struct landlock_ruleset_attr ruleset_attr = {
+		.scoped = LANDLOCK_SCOPE_SIGNAL,
+	};
+	__u64 denial_dom = 1;
+	__u64 allocated_dom = 2;
+	__u64 deallocated_dom = 3;
+	pthread_t thread;
+	int pipe_child[2], pipe_parent[2];
+	char buffer;
+	struct thread_data child_data;
+
+	child_data.parent_pid = getppid();
+	ASSERT_EQ(0, pipe2(pipe_child, O_CLOEXEC));
+	child_data.pipe_child = pipe_child[1];
+	ASSERT_EQ(0, pipe2(pipe_parent, O_CLOEXEC));
+	child_data.pipe_parent = pipe_parent[0];
+	child_data.ruleset_fd =
+		landlock_create_ruleset(&ruleset_attr, sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, child_data.ruleset_fd);
+
+	/* TGID and TID are the same for the initial thread . */
+	EXPECT_EQ(getpid(), gettid());
+	EXPECT_EQ(0, prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0));
+	ASSERT_EQ(0, pthread_create(&thread, NULL, thread_audit_test,
+				    &child_data));
+
+	/* Waits for the child to generate a denial. */
+	ASSERT_EQ(1, read(pipe_child[0], &buffer, 1));
+	EXPECT_EQ(0, close(pipe_child[0]));
+
+	/* Matches the signal log to get the domain ID. */
+	EXPECT_EQ(0, matches_log_signal(_metadata, self->audit_fd,
+					child_data.parent_pid, &denial_dom));
+	EXPECT_NE(denial_dom, 1);
+	EXPECT_NE(denial_dom, 0);
+
+	EXPECT_EQ(0, matches_log_domain_allocated(self->audit_fd, getpid(),
+						  &allocated_dom));
+	EXPECT_EQ(denial_dom, allocated_dom);
+
+	/* Updates filter rules to match the drop record. */
+	set_cap(_metadata, CAP_AUDIT_CONTROL);
+	EXPECT_EQ(0, audit_filter_drop(self->audit_fd, AUDIT_ADD_RULE));
+	EXPECT_EQ(0, audit_filter_exe(self->audit_fd, &self->audit_filter,
+				      AUDIT_DEL_RULE));
+	clear_cap(_metadata, CAP_AUDIT_CONTROL);
+
+	/* Signals the thread to exit, which will generate a domain deallocation. */
+	ASSERT_EQ(1, write(pipe_parent[1], ".", 1));
+	EXPECT_EQ(0, close(pipe_parent[1]));
+	ASSERT_EQ(0, pthread_join(thread, NULL));
+
+	EXPECT_EQ(0, setsockopt(self->audit_fd, SOL_SOCKET, SO_RCVTIMEO,
+				&audit_tv_dom_drop, sizeof(audit_tv_dom_drop)));
+	EXPECT_EQ(0, matches_log_domain_deallocated(self->audit_fd, 1,
+						    &deallocated_dom));
+	EXPECT_EQ(denial_dom, deallocated_dom);
+	EXPECT_EQ(0, setsockopt(self->audit_fd, SOL_SOCKET, SO_RCVTIMEO,
+				&audit_tv_default, sizeof(audit_tv_default)));
+}
+
 FIXTURE(audit_flags)
 {
 	struct audit_filter audit_filter;
@@ -273,7 +392,8 @@ TEST_F(audit_flags, signal)
 
 			/* Checks domain information records. */
 			EXPECT_EQ(0, matches_log_domain_allocated(
-					     self->audit_fd, &allocated_dom));
+					     self->audit_fd, getpid(),
+					     &allocated_dom));
 			EXPECT_NE(*self->domain_id, 1);
 			EXPECT_NE(*self->domain_id, 0);
 			EXPECT_EQ(*self->domain_id, allocated_dom);
diff --git a/tools/testing/selftests/landlock/fs_test.c b/tools/testing/selftests/landlock/fs_test.c
index f819011a8798..73729382d40f 100644
--- a/tools/testing/selftests/landlock/fs_test.c
+++ b/tools/testing/selftests/landlock/fs_test.c
@@ -5964,7 +5964,8 @@ TEST_F(audit_layout1, refer_handled)
 	EXPECT_EQ(EXDEV, errno);
 	EXPECT_EQ(0, matches_log_fs(_metadata, self->audit_fd, "fs\\.refer",
 				    dir_s1d1));
-	EXPECT_EQ(0, matches_log_domain_allocated(self->audit_fd, NULL));
+	EXPECT_EQ(0,
+		  matches_log_domain_allocated(self->audit_fd, getpid(), NULL));
 	EXPECT_EQ(0, matches_log_fs(_metadata, self->audit_fd, "fs\\.refer",
 				    dir_s1d3));
 
diff --git a/tools/testing/selftests/lib/config b/tools/testing/selftests/lib/config
index 81a1f64a22e8..377b3699ff31 100644
--- a/tools/testing/selftests/lib/config
+++ b/tools/testing/selftests/lib/config
@@ -1,2 +1,3 @@
 CONFIG_TEST_BITMAP=m
+CONFIG_PRIME_NUMBERS=m
 CONFIG_TEST_BITOPS=m
diff --git a/tools/testing/selftests/mincore/mincore_selftest.c b/tools/testing/selftests/mincore/mincore_selftest.c
index e949a43a6145..17ed3e9917ca 100644
--- a/tools/testing/selftests/mincore/mincore_selftest.c
+++ b/tools/testing/selftests/mincore/mincore_selftest.c
@@ -261,9 +261,6 @@ TEST(check_file_mmap)
 		TH_LOG("No read-ahead pages found in memory");
 	}
 
-	EXPECT_LT(i, vec_size) {
-		TH_LOG("Read-ahead pages reached the end of the file");
-	}
 	/*
 	 * End of the readahead window. The rest of the pages shouldn't
 	 * be in memory.
@@ -286,8 +283,7 @@ out_free:
 
 /*
  * Test mincore() behavior on a page backed by a tmpfs file.  This test
- * performs the same steps as the previous one. However, we don't expect
- * any readahead in this case.
+ * performs the same steps as the previous one.
  */
 TEST(check_tmpfs_mmap)
 {
@@ -298,7 +294,6 @@ TEST(check_tmpfs_mmap)
 	int page_size;
 	int fd;
 	int i;
-	int ra_pages = 0;
 
 	page_size = sysconf(_SC_PAGESIZE);
 	vec_size = FILE_SIZE / page_size;
@@ -341,8 +336,7 @@ TEST(check_tmpfs_mmap)
 	}
 
 	/*
-	 * Touch a page in the middle of the mapping. We expect only
-	 * that page to be fetched into memory.
+	 * Touch a page in the middle of the mapping.
 	 */
 	addr[FILE_SIZE / 2] = 1;
 	retval = mincore(addr, FILE_SIZE, vec);
@@ -351,15 +345,6 @@ TEST(check_tmpfs_mmap)
 		TH_LOG("Page not found in memory after use");
 	}
 
-	i = FILE_SIZE / 2 / page_size + 1;
-	while (i < vec_size && vec[i]) {
-		ra_pages++;
-		i++;
-	}
-	ASSERT_EQ(ra_pages, 0) {
-		TH_LOG("Read-ahead pages found in memory");
-	}
-
 	munmap(addr, FILE_SIZE);
 	close(fd);
 	free(vec);
diff --git a/tools/testing/selftests/mm/charge_reserved_hugetlb.sh b/tools/testing/selftests/mm/charge_reserved_hugetlb.sh
index 67df7b47087f..e1fe16bcbbe8 100755
--- a/tools/testing/selftests/mm/charge_reserved_hugetlb.sh
+++ b/tools/testing/selftests/mm/charge_reserved_hugetlb.sh
@@ -29,7 +29,7 @@ fi
 if [[ $cgroup2 ]]; then
   cgroup_path=$(mount -t cgroup2 | head -1 | awk '{print $3}')
   if [[ -z "$cgroup_path" ]]; then
-    cgroup_path=/dev/cgroup/memory
+    cgroup_path=$(mktemp -d)
     mount -t cgroup2 none $cgroup_path
     do_umount=1
   fi
@@ -37,7 +37,7 @@ if [[ $cgroup2 ]]; then
 else
   cgroup_path=$(mount -t cgroup | grep ",hugetlb" | awk '{print $3}')
   if [[ -z "$cgroup_path" ]]; then
-    cgroup_path=/dev/cgroup/memory
+    cgroup_path=$(mktemp -d)
     mount -t cgroup memory,hugetlb $cgroup_path
     do_umount=1
   fi
diff --git a/tools/testing/selftests/mm/cow.c b/tools/testing/selftests/mm/cow.c
index f0cb14ea8608..b6cfe0a4b7df 100644
--- a/tools/testing/selftests/mm/cow.c
+++ b/tools/testing/selftests/mm/cow.c
@@ -293,7 +293,7 @@ static void do_test_vmsplice_in_parent(char *mem, size_t size,
 		.iov_base = mem,
 		.iov_len = size,
 	};
-	ssize_t cur, total, transferred;
+	ssize_t cur, total, transferred = 0;
 	struct comm_pipes comm_pipes;
 	char *old, *new;
 	int ret, fds[2];
diff --git a/tools/testing/selftests/mm/hugetlb_reparenting_test.sh b/tools/testing/selftests/mm/hugetlb_reparenting_test.sh
index 11f9bbe7dc22..0b0d4ba1af27 100755
--- a/tools/testing/selftests/mm/hugetlb_reparenting_test.sh
+++ b/tools/testing/selftests/mm/hugetlb_reparenting_test.sh
@@ -23,7 +23,7 @@ fi
 if [[ $cgroup2 ]]; then
   CGROUP_ROOT=$(mount -t cgroup2 | head -1 | awk '{print $3}')
   if [[ -z "$CGROUP_ROOT" ]]; then
-    CGROUP_ROOT=/dev/cgroup/memory
+    CGROUP_ROOT=$(mktemp -d)
     mount -t cgroup2 none $CGROUP_ROOT
     do_umount=1
   fi
diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore
index 679542f565a4..532bb732bc6d 100644
--- a/tools/testing/selftests/net/.gitignore
+++ b/tools/testing/selftests/net/.gitignore
@@ -39,6 +39,7 @@ scm_rights
 sk_bind_sendto_listen
 sk_connect_zero_addr
 sk_so_peek_off
+skf_net_off
 socket
 so_incoming_cpu
 so_netns_cookie
diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 6d718b478ed8..124078b56fa4 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -106,6 +106,8 @@ TEST_PROGS += ipv6_route_update_soft_lockup.sh
 TEST_PROGS += busy_poll_test.sh
 TEST_GEN_PROGS += proc_net_pktgen
 TEST_PROGS += lwt_dst_cache_ref_loop.sh
+TEST_PROGS += skf_net_off.sh
+TEST_GEN_FILES += skf_net_off
 
 # YNL files, must be before "include ..lib.mk"
 YNL_GEN_FILES := busy_poller netlink-dumps
diff --git a/tools/testing/selftests/net/fib_rule_tests.sh b/tools/testing/selftests/net/fib_rule_tests.sh
index b866bab1d92a..c7cea556b416 100755
--- a/tools/testing/selftests/net/fib_rule_tests.sh
+++ b/tools/testing/selftests/net/fib_rule_tests.sh
@@ -359,6 +359,23 @@ fib_rule6_test()
 			"$getnomatch" "iif flowlabel masked redirect to table" \
 			"iif flowlabel masked no redirect to table"
 	fi
+
+	$IP link show dev $DEV | grep -q vrf0
+	if [ $? -eq 0 ]; then
+		match="oif vrf0"
+		getmatch="oif $DEV"
+		getnomatch="oif lo"
+		fib_rule6_test_match_n_redirect "$match" "$getmatch" \
+			"$getnomatch" "VRF oif redirect to table" \
+			"VRF oif no redirect to table"
+
+		match="from $SRC_IP6 iif vrf0"
+		getmatch="from $SRC_IP6 iif $DEV"
+		getnomatch="from $SRC_IP6 iif lo"
+		fib_rule6_test_match_n_redirect "$match" "$getmatch" \
+			"$getnomatch" "VRF iif redirect to table" \
+			"VRF iif no redirect to table"
+	fi
 }
 
 fib_rule6_vrf_test()
@@ -635,6 +652,23 @@ fib_rule4_test()
 			"$getnomatch" "iif dscp masked redirect to table" \
 			"iif dscp masked no redirect to table"
 	fi
+
+	$IP link show dev $DEV | grep -q vrf0
+	if [ $? -eq 0 ]; then
+		match="oif vrf0"
+		getmatch="oif $DEV"
+		getnomatch="oif lo"
+		fib_rule4_test_match_n_redirect "$match" "$getmatch" \
+			"$getnomatch" "VRF oif redirect to table" \
+			"VRF oif no redirect to table"
+
+		match="from $SRC_IP iif vrf0"
+		getmatch="from $SRC_IP iif $DEV"
+		getnomatch="from $SRC_IP iif lo"
+		fib_rule4_test_match_n_redirect "$match" "$getmatch" \
+			"$getnomatch" "VRF iif redirect to table" \
+			"VRF iif no redirect to table"
+	fi
 }
 
 fib_rule4_vrf_test()
diff --git a/tools/testing/selftests/net/forwarding/bridge_vlan_aware.sh b/tools/testing/selftests/net/forwarding/bridge_vlan_aware.sh
index 90f8a244ea90..e59fba366a0a 100755
--- a/tools/testing/selftests/net/forwarding/bridge_vlan_aware.sh
+++ b/tools/testing/selftests/net/forwarding/bridge_vlan_aware.sh
@@ -1,7 +1,7 @@
 #!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
-ALL_TESTS="ping_ipv4 ping_ipv6 learning flooding vlan_deletion extern_learn other_tpid"
+ALL_TESTS="ping_ipv4 ping_ipv6 learning flooding vlan_deletion extern_learn other_tpid 8021p drop_untagged"
 NUM_NETIFS=4
 CHECK_TC="yes"
 source lib.sh
@@ -194,6 +194,100 @@ other_tpid()
 	tc qdisc del dev $h2 clsact
 }
 
+8021p_do()
+{
+	local should_fail=$1; shift
+	local mac=de:ad:be:ef:13:37
+
+	tc filter add dev $h2 ingress protocol all pref 1 handle 101 \
+		flower dst_mac $mac action drop
+
+	$MZ -q $h1 -c 1 -b $mac -a own "81:00 00:00 08:00 aa-aa-aa-aa-aa-aa-aa-aa-aa"
+	sleep 1
+
+	tc -j -s filter show dev $h2 ingress \
+		| jq -e ".[] | select(.options.handle == 101) \
+		| select(.options.actions[0].stats.packets == 1)" &> /dev/null
+	check_err_fail $should_fail $? "802.1p-tagged reception"
+
+	tc filter del dev $h2 ingress pref 1
+}
+
+8021p()
+{
+	RET=0
+
+	tc qdisc add dev $h2 clsact
+	ip link set $h2 promisc on
+
+	# Test that with the default_pvid, 1, packets tagged with VID 0 are
+	# accepted.
+	8021p_do 0
+
+	# Test that packets tagged with VID 0 are still accepted after changing
+	# the default_pvid.
+	ip link set br0 type bridge vlan_default_pvid 10
+	8021p_do 0
+
+	log_test "Reception of 802.1p-tagged traffic"
+
+	ip link set $h2 promisc off
+	tc qdisc del dev $h2 clsact
+}
+
+send_untagged_and_8021p()
+{
+	ping_do $h1 192.0.2.2
+	check_fail $?
+
+	8021p_do 1
+}
+
+drop_untagged()
+{
+	RET=0
+
+	tc qdisc add dev $h2 clsact
+	ip link set $h2 promisc on
+
+	# Test that with no PVID, untagged and 802.1p-tagged traffic is
+	# dropped.
+	ip link set br0 type bridge vlan_default_pvid 1
+
+	# First we reconfigure the default_pvid, 1, as a non-PVID VLAN.
+	bridge vlan add dev $swp1 vid 1 untagged
+	send_untagged_and_8021p
+	bridge vlan add dev $swp1 vid 1 pvid untagged
+
+	# Next we try to delete VID 1 altogether
+	bridge vlan del dev $swp1 vid 1
+	send_untagged_and_8021p
+	bridge vlan add dev $swp1 vid 1 pvid untagged
+
+	# Set up the bridge without a default_pvid, then check that the 8021q
+	# module, when the bridge port goes down and then up again, does not
+	# accidentally re-enable untagged packet reception.
+	ip link set br0 type bridge vlan_default_pvid 0
+	ip link set $swp1 down
+	ip link set $swp1 up
+	setup_wait
+	send_untagged_and_8021p
+
+	# Remove swp1 as a bridge port and let it rejoin the bridge while it
+	# has no default_pvid.
+	ip link set $swp1 nomaster
+	ip link set $swp1 master br0
+	send_untagged_and_8021p
+
+	# Restore settings
+	ip link set br0 type bridge vlan_default_pvid 1
+
+	log_test "Dropping of untagged and 802.1p-tagged traffic with no PVID"
+
+	ip link set $h2 promisc off
+	tc qdisc del dev $h2 clsact
+}
+
 trap cleanup EXIT
 
 setup_prepare
diff --git a/tools/testing/selftests/net/forwarding/tc_taprio.sh b/tools/testing/selftests/net/forwarding/tc_taprio.sh
new file mode 100755
index 000000000000..8992aeabfe0b
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/tc_taprio.sh
@@ -0,0 +1,421 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS=" \
+	test_clock_jump_backward \
+	test_taprio_after_ptp \
+	test_max_sdu \
+	test_clock_jump_backward_forward \
+"
+NUM_NETIFS=4
+source tc_common.sh
+source lib.sh
+source tsn_lib.sh
+
+require_command python3
+
+# The test assumes the usual topology from the README, where h1 is connected to
+# swp1, h2 to swp2, and swp1 and swp2 are together in a bridge.
+# Additional assumption: h1 and h2 use the same PHC, and so do swp1 and swp2.
+# By synchronizing h1 to swp1 via PTP, h2 is also implicitly synchronized to
+# swp1 (and both to CLOCK_REALTIME).
+h1=${NETIFS[p1]}
+swp1=${NETIFS[p2]}
+swp2=${NETIFS[p3]}
+h2=${NETIFS[p4]}
+
+UDS_ADDRESS_H1="/var/run/ptp4l_h1"
+UDS_ADDRESS_SWP1="/var/run/ptp4l_swp1"
+
+H1_IPV4="192.0.2.1"
+H2_IPV4="192.0.2.2"
+H1_IPV6="2001:db8:1::1"
+H2_IPV6="2001:db8:1::2"
+
+# Tunables
+NUM_PKTS=100
+STREAM_VID=10
+STREAM_PRIO_1=6
+STREAM_PRIO_2=5
+STREAM_PRIO_3=4
+# PTP uses TC 0
+ALL_GATES=$((1 << 0 | 1 << STREAM_PRIO_1 | 1 << STREAM_PRIO_2))
+# Use a conservative cycle of 10 ms to allow the test to still pass when the
+# kernel has some extra overhead like lockdep etc
+CYCLE_TIME_NS=10000000
+# Create two Gate Control List entries, one OPEN and one CLOSE, of equal
+# durations
+GATE_DURATION_NS=$((CYCLE_TIME_NS / 2))
+# Give 2/3 of the cycle time to user space and 1/3 to the kernel
+FUDGE_FACTOR=$((CYCLE_TIME_NS / 3))
+# Shift the isochron base time by half the gate time, so that packets are
+# always received by swp1 close to the middle of the time slot, to minimize
+# inaccuracies due to network sync
+SHIFT_TIME_NS=$((GATE_DURATION_NS / 2))
+
+path_delay=
+
+h1_create()
+{
+	simple_if_init $h1 $H1_IPV4/24 $H1_IPV6/64
+}
+
+h1_destroy()
+{
+	simple_if_fini $h1 $H1_IPV4/24 $H1_IPV6/64
+}
+
+h2_create()
+{
+	simple_if_init $h2 $H2_IPV4/24 $H2_IPV6/64
+}
+
+h2_destroy()
+{
+	simple_if_fini $h2 $H2_IPV4/24 $H2_IPV6/64
+}
+
+switch_create()
+{
+	local h2_mac_addr=$(mac_get $h2)
+
+	ip link set $swp1 up
+	ip link set $swp2 up
+
+	ip link add br0 type bridge vlan_filtering 1
+	ip link set $swp1 master br0
+	ip link set $swp2 master br0
+	ip link set br0 up
+
+	bridge vlan add dev $swp2 vid $STREAM_VID
+	bridge vlan add dev $swp1 vid $STREAM_VID
+	bridge fdb add dev $swp2 \
+		$h2_mac_addr vlan $STREAM_VID static master
+}
+
+switch_destroy()
+{
+	ip link del br0
+}
+
+ptp_setup()
+{
+	# Set up swp1 as a master PHC for h1, synchronized to the local
+	# CLOCK_REALTIME.
+	phc2sys_start $UDS_ADDRESS_SWP1
+	ptp4l_start $h1 true $UDS_ADDRESS_H1
+	ptp4l_start $swp1 false $UDS_ADDRESS_SWP1
+}
+
+ptp_cleanup()
+{
+	ptp4l_stop $swp1
+	ptp4l_stop $h1
+	phc2sys_stop
+}
+
+txtime_setup()
+{
+	local if_name=$1
+
+	tc qdisc add dev $if_name clsact
+	# Classify PTP on TC 7 and isochron on TC 6
+	tc filter add dev $if_name egress protocol 0x88f7 \
+		flower action skbedit priority 7
+	tc filter add dev $if_name egress protocol 802.1Q \
+		flower vlan_ethtype 0xdead action skbedit priority 6
+	tc qdisc add dev $if_name handle 100: parent root mqprio num_tc 8 \
+		queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 \
+		map 0 1 2 3 4 5 6 7 \
+		hw 1
+	# Set up TC 5, 6, 7 for SO_TXTIME. tc-mqprio queues count from 1.
+	tc qdisc replace dev $if_name parent 100:$((STREAM_PRIO_1 + 1)) etf \
+		clockid CLOCK_TAI offload delta $FUDGE_FACTOR
+	tc qdisc replace dev $if_name parent 100:$((STREAM_PRIO_2 + 1)) etf \
+		clockid CLOCK_TAI offload delta $FUDGE_FACTOR
+	tc qdisc replace dev $if_name parent 100:$((STREAM_PRIO_3 + 1)) etf \
+		clockid CLOCK_TAI offload delta $FUDGE_FACTOR
+}
+
+txtime_cleanup()
+{
+	local if_name=$1
+
+	tc qdisc del dev $if_name clsact
+	tc qdisc del dev $if_name root
+}
+
+taprio_replace()
+{
+	local if_name="$1"; shift
+	local extra_args="$1"; shift
+
+	# STREAM_PRIO_1 always has an open gate.
+	# STREAM_PRIO_2 has a gate open for GATE_DURATION_NS (half the cycle time)
+	# STREAM_PRIO_3 always has a closed gate.
+	tc qdisc replace dev $if_name root stab overhead 24 taprio num_tc 8 \
+		queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 \
+		map 0 1 2 3 4 5 6 7 \
+		sched-entry S $(printf "%x" $ALL_GATES) $GATE_DURATION_NS \
+		sched-entry S $(printf "%x" $((ALL_GATES & ~(1 << STREAM_PRIO_2)))) $GATE_DURATION_NS \
+		base-time 0 flags 0x2 $extra_args
+	taprio_wait_for_admin $if_name
+}
+
+taprio_cleanup()
+{
+	local if_name=$1
+
+	tc qdisc del dev $if_name root
+}
+
+probe_path_delay()
+{
+	local isochron_dat="$(mktemp)"
+	local received
+
+	log_info "Probing path delay"
+
+	isochron_do "$h1" "$h2" "$UDS_ADDRESS_H1" "" 0 \
+		"$CYCLE_TIME_NS" "" "" "$NUM_PKTS" \
+		"$STREAM_VID" "$STREAM_PRIO_1" "" "$isochron_dat"
+
+	received=$(isochron_report_num_received "$isochron_dat")
+	if [ "$received" != "$NUM_PKTS" ]; then
+		echo "Cannot establish basic data path between $h1 and $h2"
+		exit $ksft_fail
+	fi
+
+	printf "pdelay = {}\n" > isochron_data.py
+	isochron report --input-file "$isochron_dat" \
+		--printf-format "pdelay[%u] = %d - %d\n" \
+		--printf-args "qRT" \
+		>> isochron_data.py
+	cat <<-'EOF' > isochron_postprocess.py
+	#!/usr/bin/env python3
+
+	from isochron_data import pdelay
+	import numpy as np
+
+	w = np.array(list(pdelay.values()))
+	print("{}".format(np.max(w)))
+	EOF
+	path_delay=$(python3 ./isochron_postprocess.py)
+
+	log_info "Path delay from $h1 to $h2 estimated at $path_delay ns"
+
+	if [ "$path_delay" -gt "$GATE_DURATION_NS" ]; then
+		echo "Path delay larger than gate duration, aborting"
+		exit $ksft_fail
+	fi
+
+	rm -f ./isochron_data.py 2> /dev/null
+	rm -f ./isochron_postprocess.py 2> /dev/null
+	rm -f "$isochron_dat" 2> /dev/null
+}
+
+setup_prepare()
+{
+	vrf_prepare
+
+	h1_create
+	h2_create
+	switch_create
+
+	txtime_setup $h1
+
+	# Temporarily set up PTP just to probe the end-to-end path delay.
+	ptp_setup
+	probe_path_delay
+	ptp_cleanup
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	isochron_recv_stop
+	txtime_cleanup $h1
+
+	switch_destroy
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+}
+
+run_test()
+{
+	local base_time=$1; shift
+	local stream_prio=$1; shift
+	local expected_delay=$1; shift
+	local should_fail=$1; shift
+	local test_name=$1; shift
+	local isochron_dat="$(mktemp)"
+	local received
+	local median_delay
+
+	RET=0
+
+	# Set the shift time equal to the cycle time, which effectively
+	# cancels the default advance time. Packets won't be sent early in
+	# software, which ensures that they won't prematurely enter through
+	# the open gate in __test_out_of_band(). Also, the gate is open for
+	# long enough that this won't cause a problem in __test_in_band().
+	isochron_do "$h1" "$h2" "$UDS_ADDRESS_H1" "" "$base_time" \
+		"$CYCLE_TIME_NS" "$SHIFT_TIME_NS" "$GATE_DURATION_NS" \
+		"$NUM_PKTS" "$STREAM_VID" "$stream_prio" "" "$isochron_dat"
+
+	received=$(isochron_report_num_received "$isochron_dat")
+	[ "$received" = "$NUM_PKTS" ]
+	check_err_fail $should_fail $? "Reception of $NUM_PKTS packets"
+
+	if [ $should_fail = 0 ] && [ "$received" = "$NUM_PKTS" ]; then
+		printf "pdelay = {}\n" > isochron_data.py
+		isochron report --input-file "$isochron_dat" \
+			--printf-format "pdelay[%u] = %d - %d\n" \
+			--printf-args "qRT" \
+			>> isochron_data.py
+		cat <<-'EOF' > isochron_postprocess.py
+		#!/usr/bin/env python3
+
+		from isochron_data import pdelay
+		import numpy as np
+
+		w = np.array(list(pdelay.values()))
+		print("{}".format(int(np.median(w))))
+		EOF
+		median_delay=$(python3 ./isochron_postprocess.py)
+
+		# If the condition below is true, packets were delayed by a closed gate
+		[ "$median_delay" -gt $((path_delay + expected_delay)) ]
+		check_fail $? "Median delay $median_delay is greater than expected delay $expected_delay plus path delay $path_delay"
+
+		# If the condition below is true, packets were sent expecting them to
+		# hit a closed gate in the switch, but were not delayed
+		[ "$expected_delay" -gt 0 ] && [ "$median_delay" -lt "$expected_delay" ]
+		check_fail $? "Median delay $median_delay is less than expected delay $expected_delay"
+	fi
+
+	log_test "$test_name"
+
+	rm -f ./isochron_data.py 2> /dev/null
+	rm -f ./isochron_postprocess.py 2> /dev/null
+	rm -f "$isochron_dat" 2> /dev/null
+}
+
+__test_always_open()
+{
+	run_test 0.000000000 $STREAM_PRIO_1 0 0 "Gate always open"
+}
+
+__test_always_closed()
+{
+	run_test 0.000000000 $STREAM_PRIO_3 0 1 "Gate always closed"
+}
+
+__test_in_band()
+{
+	# Send packets in-band with the OPEN gate entry
+	run_test 0.000000000 $STREAM_PRIO_2 0 0 "In band with gate"
+}
+
+__test_out_of_band()
+{
+	# Send packets in-band with the CLOSE gate entry
+	run_test 0.005000000 $STREAM_PRIO_2 \
+		$((GATE_DURATION_NS - SHIFT_TIME_NS)) 0 \
+		"Out of band with gate"
+}
+
+run_subtests()
+{
+	__test_always_open
+	__test_always_closed
+	__test_in_band
+	__test_out_of_band
+}
+
+test_taprio_after_ptp()
+{
+	log_info "Setting up taprio after PTP"
+	ptp_setup
+	taprio_replace $swp2
+	run_subtests
+	taprio_cleanup $swp2
+	ptp_cleanup
+}
+
+__test_under_max_sdu()
+{
+	# Limit max-sdu for STREAM_PRIO_1
+	taprio_replace "$swp2" "max-sdu 0 0 0 0 0 0 100 0"
+	run_test 0.000000000 $STREAM_PRIO_1 0 0 "Under maximum SDU"
+}
+
+__test_over_max_sdu()
+{
+	# Limit max-sdu for STREAM_PRIO_1
+	taprio_replace "$swp2" "max-sdu 0 0 0 0 0 0 20 0"
+	run_test 0.000000000 $STREAM_PRIO_1 0 1 "Over maximum SDU"
+}
+
+test_max_sdu()
+{
+	ptp_setup
+	__test_under_max_sdu
+	__test_over_max_sdu
+	taprio_cleanup $swp2
+	ptp_cleanup
+}
+
+# Perform a clock jump in the past without synchronization running, so that the
+# time base remains where it was set by phc_ctl.
+test_clock_jump_backward()
+{
+	# This is a more complex schedule specifically crafted in a way that
+	# has been problematic on NXP LS1028A. Not much to test with it other
+	# than the fact that it passes traffic.
+	tc qdisc replace dev $swp2 root stab overhead 24 taprio num_tc 8 \
+		queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 map 0 1 2 3 4 5 6 7 \
+		base-time 0 sched-entry S 20 300000 sched-entry S 10 200000 \
+		sched-entry S 20 300000 sched-entry S 48 200000 \
+		sched-entry S 20 300000 sched-entry S 83 200000 \
+		sched-entry S 40 300000 sched-entry S 00 200000 flags 2
+
+	log_info "Forcing a backward clock jump"
+	phc_ctl $swp1 set 0
+
+	ping_test $h1 192.0.2.2
+	taprio_cleanup $swp2
+}
+
+# Test that taprio tolerates clock jumps.
+# Since ptp4l and phc2sys are running, it is expected for the time to
+# eventually recover (through yet another clock jump). Isochron waits
+# until that is the case.
+test_clock_jump_backward_forward()
+{
+	log_info "Forcing a backward and a forward clock jump"
+	taprio_replace $swp2
+	phc_ctl $swp1 set 0
+	ptp_setup
+	ping_test $h1 192.0.2.2
+	run_subtests
+	ptp_cleanup
+	taprio_cleanup $swp2
+}
+
+tc_offload_check
+if [[ $? -ne 0 ]]; then
+	log_test_skip "Could not test offloaded functionality"
+	exit $EXIT_STATUS
+fi
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+tests_run
+
+exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/tsn_lib.sh b/tools/testing/selftests/net/forwarding/tsn_lib.sh
index b91bcd8008a9..08c044ff6689 100644
--- a/tools/testing/selftests/net/forwarding/tsn_lib.sh
+++ b/tools/testing/selftests/net/forwarding/tsn_lib.sh
@@ -2,6 +2,8 @@
 # SPDX-License-Identifier: GPL-2.0
 # Copyright 2021-2022 NXP
 
+tc_testing_scripts_dir=$(dirname $0)/../../tc-testing/scripts
+
 REQUIRE_ISOCHRON=${REQUIRE_ISOCHRON:=yes}
 REQUIRE_LINUXPTP=${REQUIRE_LINUXPTP:=yes}
 
@@ -18,6 +20,7 @@ fi
 if [[ "$REQUIRE_LINUXPTP" = "yes" ]]; then
 	require_command phc2sys
 	require_command ptp4l
+	require_command phc_ctl
 fi
 
 phc2sys_start()
@@ -182,6 +185,7 @@ isochron_do()
 	local base_time=$1; shift
 	local cycle_time=$1; shift
 	local shift_time=$1; shift
+	local window_size=$1; shift
 	local num_pkts=$1; shift
 	local vid=$1; shift
 	local priority=$1; shift
@@ -212,6 +216,10 @@ isochron_do()
 		extra_args="${extra_args} --shift-time=${shift_time}"
 	fi
 
+	if ! [ -z "${window_size}" ]; then
+		extra_args="${extra_args} --window-size=${window_size}"
+	fi
+
 	if [ "${use_l2}" = "true" ]; then
 		extra_args="${extra_args} --l2 --etype=0xdead ${vid}"
 		receiver_extra_args="--l2 --etype=0xdead"
@@ -247,3 +255,21 @@ isochron_do()
 
 	cpufreq_restore ${ISOCHRON_CPU}
 }
+
+isochron_report_num_received()
+{
+	local isochron_dat=$1; shift
+
+	# Count all received packets by looking at the non-zero RX timestamps
+	isochron report \
+		--input-file "${isochron_dat}" \
+		--printf-format "%u\n" --printf-args "R" | \
+		grep -w -v '0' | wc -l
+}
+
+taprio_wait_for_admin()
+{
+	local if_name="$1"; shift
+
+	"$tc_testing_scripts_dir/taprio_wait_for_admin.sh" "$(which tc)" "$if_name"
+}
diff --git a/tools/testing/selftests/net/mptcp/diag.sh b/tools/testing/selftests/net/mptcp/diag.sh
index 4f55477ffe08..e7a75341f0f3 100755
--- a/tools/testing/selftests/net/mptcp/diag.sh
+++ b/tools/testing/selftests/net/mptcp/diag.sh
@@ -206,9 +206,8 @@ chk_dump_one()
 	local token
 	local msg
 
-	ss_token="$(ss -inmHMN $ns | grep 'token:' |\
-		    head -n 1 |\
-		    sed 's/.*token:\([0-9a-f]*\).*/\1/')"
+	ss_token="$(ss -inmHMN $ns |
+		    mptcp_lib_get_info_value "token" "token")"
 
 	token="$(ip netns exec $ns ./mptcp_diag -t $ss_token |\
 		 awk -F':[ \t]+' '/^token/ {print $2}')"
diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 13a3b68181ee..befa66f5a366 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -1441,6 +1441,15 @@ chk_join_nr()
 		fi
 	fi
 
+	count=$(mptcp_lib_get_counter ${ns2} "MPTcpExtMPJoinSynAckHMacFailure")
+	if [ -z "$count" ]; then
+		rc=${KSFT_SKIP}
+	elif [ "$count" != "0" ]; then
+		rc=${KSFT_FAIL}
+		print_check "synack HMAC"
+		fail_test "got $count JOIN[s] synack HMAC failure expected 0"
+	fi
+
 	count=$(mptcp_lib_get_counter ${ns1} "MPTcpExtMPJoinAckRx")
 	if [ -z "$count" ]; then
 		rc=${KSFT_SKIP}
@@ -1450,6 +1459,15 @@ chk_join_nr()
 		fail_test "got $count JOIN[s] ack rx expected $ack_nr"
 	fi
 
+	count=$(mptcp_lib_get_counter ${ns1} "MPTcpExtMPJoinAckHMacFailure")
+	if [ -z "$count" ]; then
+		rc=${KSFT_SKIP}
+	elif [ "$count" != "0" ]; then
+		rc=${KSFT_FAIL}
+		print_check "ack HMAC"
+		fail_test "got $count JOIN[s] ack HMAC failure expected 0"
+	fi
+
 	print_results "join Rx" ${rc}
 
 	join_syn_tx="${join_syn_tx:-${syn_nr}}" \
diff --git a/tools/testing/selftests/net/netfilter/nft_concat_range.sh b/tools/testing/selftests/net/netfilter/nft_concat_range.sh
index 47088b005390..1f5979c1510c 100755
--- a/tools/testing/selftests/net/netfilter/nft_concat_range.sh
+++ b/tools/testing/selftests/net/netfilter/nft_concat_range.sh
@@ -27,7 +27,7 @@ TYPES="net_port port_net net6_port port_proto net6_port_mac net6_port_mac_proto
        net6_port_net6_port net_port_mac_proto_net"
 
 # Reported bugs, also described by TYPE_ variables below
-BUGS="flush_remove_add reload net_port_proto_match"
+BUGS="flush_remove_add reload net_port_proto_match avx2_mismatch"
 
 # List of possible paths to pktgen script from kernel tree for performance tests
 PKTGEN_SCRIPT_PATHS="
@@ -387,6 +387,25 @@ race_repeat	0
 
 perf_duration	0
 "
+
+TYPE_avx2_mismatch="
+display		avx2 false match
+type_spec	inet_proto . ipv6_addr
+chain_spec	meta l4proto . ip6 daddr
+dst		proto addr6
+src
+start		1
+count		1
+src_delta	1
+tools		ping
+proto		icmp6
+
+race_repeat	0
+
+perf_duration	0
+"
+
+
 # Set template for all tests, types and rules are filled in depending on test
 set_template='
 flush ruleset
@@ -1629,6 +1648,24 @@ test_bug_net_port_proto_match() {
 	nft flush ruleset
 }
 
+test_bug_avx2_mismatch()
+{
+	setup veth send_"${proto}" set || return ${ksft_skip}
+
+	local a1="fe80:dead:01ff:0a02:0b03:6007:8009:a001"
+	local a2="fe80:dead:01fe:0a02:0b03:6007:8009:a001"
+
+	nft "add element inet filter test { icmpv6 . $a1 }"
+
+	dst_addr6="$a2"
+	send_icmp6
+
+	if [ "$(count_packets)" -gt "0" ]; then
+		err "False match for $a2"
+		return 1
+	fi
+}
+
 test_reported_issues() {
 	eval test_bug_"${subtest}"
 }
diff --git a/tools/testing/selftests/net/skf_net_off.c b/tools/testing/selftests/net/skf_net_off.c
new file mode 100644
index 000000000000..1fdf61d6cd7f
--- /dev/null
+++ b/tools/testing/selftests/net/skf_net_off.c
@@ -0,0 +1,244 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Open a tun device.
+ *
+ * [modifications: use IFF_NAPI_FRAGS, add sk filter]
+ *
+ * Expects the device to have been configured previously, e.g.:
+ *   sudo ip tuntap add name tap1 mode tap
+ *   sudo ip link set tap1 up
+ *   sudo ip link set dev tap1 addr 02:00:00:00:00:01
+ *   sudo ip -6 addr add fdab::1 peer fdab::2 dev tap1 nodad
+ *
+ * And to avoid premature pskb_may_pull:
+ *
+ *   sudo ethtool -K tap1 gro off
+ *   sudo bash -c 'echo 0 > /proc/sys/net/ipv4/ip_early_demux'
+ */
+
+#define _GNU_SOURCE
+
+#include <arpa/inet.h>
+#include <errno.h>
+#include <error.h>
+#include <fcntl.h>
+#include <getopt.h>
+#include <linux/filter.h>
+#include <linux/if.h>
+#include <linux/if_packet.h>
+#include <linux/if_tun.h>
+#include <linux/ipv6.h>
+#include <netinet/if_ether.h>
+#include <netinet/in.h>
+#include <netinet/ip.h>
+#include <netinet/ip6.h>
+#include <netinet/udp.h>
+#include <poll.h>
+#include <signal.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/socket.h>
+#include <sys/poll.h>
+#include <sys/types.h>
+#include <sys/uio.h>
+#include <unistd.h>
+
+static bool cfg_do_filter;
+static bool cfg_do_frags;
+static int cfg_dst_port = 8000;
+static char *cfg_ifname;
+
+static int tun_open(const char *tun_name)
+{
+	struct ifreq ifr = {0};
+	int fd, ret;
+
+	fd = open("/dev/net/tun", O_RDWR);
+	if (fd == -1)
+		error(1, errno, "open /dev/net/tun");
+
+	ifr.ifr_flags = IFF_TAP;
+	if (cfg_do_frags)
+		ifr.ifr_flags |= IFF_NAPI | IFF_NAPI_FRAGS;
+
+	strncpy(ifr.ifr_name, tun_name, IFNAMSIZ - 1);
+
+	ret = ioctl(fd, TUNSETIFF, &ifr);
+	if (ret)
+		error(1, ret, "ioctl TUNSETIFF");
+
+	return fd;
+}
+
+static void sk_set_filter(int fd)
+{
+	const int offset_proto = offsetof(struct ip6_hdr, ip6_nxt);
+	const int offset_dport = sizeof(struct ip6_hdr) + offsetof(struct udphdr, dest);
+
+	/* Filter UDP packets with destination port cfg_dst_port */
+	struct sock_filter filter_code[] = {
+		BPF_STMT(BPF_LD  + BPF_B   + BPF_ABS, SKF_AD_OFF + SKF_AD_PKTTYPE),
+		BPF_JUMP(BPF_JMP + BPF_JEQ + BPF_K, PACKET_HOST, 0, 4),
+		BPF_STMT(BPF_LD  + BPF_B   + BPF_ABS, SKF_NET_OFF + offset_proto),
+		BPF_JUMP(BPF_JMP + BPF_JEQ + BPF_K, IPPROTO_UDP, 0, 2),
+		BPF_STMT(BPF_LD  + BPF_H   + BPF_ABS, SKF_NET_OFF + offset_dport),
+		BPF_JUMP(BPF_JMP + BPF_JEQ + BPF_K, cfg_dst_port, 1, 0),
+		BPF_STMT(BPF_RET + BPF_K, 0),
+		BPF_STMT(BPF_RET + BPF_K, 0xFFFF),
+	};
+
+	struct sock_fprog filter = {
+		sizeof(filter_code) / sizeof(filter_code[0]),
+		filter_code,
+	};
+
+	if (setsockopt(fd, SOL_SOCKET, SO_ATTACH_FILTER, &filter, sizeof(filter)))
+		error(1, errno, "setsockopt attach filter");
+}
+
+static int raw_open(void)
+{
+	int fd;
+
+	fd = socket(PF_INET6, SOCK_RAW, IPPROTO_UDP);
+	if (fd == -1)
+		error(1, errno, "socket raw (udp)");
+
+	if (cfg_do_filter)
+		sk_set_filter(fd);
+
+	return fd;
+}
+
+static void tun_write(int fd)
+{
+	const char eth_src[] = { 0x02, 0x00, 0x00, 0x00, 0x00, 0x02 };
+	const char eth_dst[] = { 0x02, 0x00, 0x00, 0x00, 0x00, 0x01 };
+	struct tun_pi pi = {0};
+	struct ipv6hdr ip6h = {0};
+	struct udphdr uh = {0};
+	struct ethhdr eth = {0};
+	uint32_t payload;
+	struct iovec iov[5];
+	int ret;
+
+	pi.proto = htons(ETH_P_IPV6);
+
+	memcpy(eth.h_source, eth_src, sizeof(eth_src));
+	memcpy(eth.h_dest, eth_dst, sizeof(eth_dst));
+	eth.h_proto = htons(ETH_P_IPV6);
+
+	ip6h.version = 6;
+	ip6h.payload_len = htons(sizeof(uh) + sizeof(uint32_t));
+	ip6h.nexthdr = IPPROTO_UDP;
+	ip6h.hop_limit = 8;
+	if (inet_pton(AF_INET6, "fdab::2", &ip6h.saddr) != 1)
+		error(1, errno, "inet_pton src");
+	if (inet_pton(AF_INET6, "fdab::1", &ip6h.daddr) != 1)
+		error(1, errno, "inet_pton src");
+
+	uh.source = htons(8000);
+	uh.dest = htons(cfg_dst_port);
+	uh.len = ip6h.payload_len;
+	uh.check = 0;
+
+	payload = htonl(0xABABABAB);		/* Covered in IPv6 length */
+
+	iov[0].iov_base = &pi;
+	iov[0].iov_len  = sizeof(pi);
+	iov[1].iov_base = &eth;
+	iov[1].iov_len  = sizeof(eth);
+	iov[2].iov_base = &ip6h;
+	iov[2].iov_len  = sizeof(ip6h);
+	iov[3].iov_base = &uh;
+	iov[3].iov_len  = sizeof(uh);
+	iov[4].iov_base = &payload;
+	iov[4].iov_len  = sizeof(payload);
+
+	ret = writev(fd, iov, sizeof(iov) / sizeof(iov[0]));
+	if (ret <= 0)
+		error(1, errno, "writev");
+}
+
+static void raw_read(int fd)
+{
+	struct timeval tv = { .tv_usec = 100 * 1000 };
+	struct msghdr msg = {0};
+	struct iovec iov[2];
+	struct udphdr uh;
+	uint32_t payload[2];
+	int ret;
+
+	if (setsockopt(fd, SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof(tv)))
+		error(1, errno, "setsockopt rcvtimeo udp");
+
+	iov[0].iov_base = &uh;
+	iov[0].iov_len = sizeof(uh);
+
+	iov[1].iov_base = payload;
+	iov[1].iov_len = sizeof(payload);
+
+	msg.msg_iov = iov;
+	msg.msg_iovlen = sizeof(iov) / sizeof(iov[0]);
+
+	ret = recvmsg(fd, &msg, 0);
+	if (ret <= 0)
+		error(1, errno, "read raw");
+	if (ret != sizeof(uh) + sizeof(payload[0]))
+		error(1, errno, "read raw: len=%d\n", ret);
+
+	fprintf(stderr, "raw recv: 0x%x\n", payload[0]);
+}
+
+static void parse_opts(int argc, char **argv)
+{
+	int c;
+
+	while ((c = getopt(argc, argv, "fFi:")) != -1) {
+		switch (c) {
+		case 'f':
+			cfg_do_filter = true;
+			printf("bpf filter enabled\n");
+			break;
+		case 'F':
+			cfg_do_frags = true;
+			printf("napi frags mode enabled\n");
+			break;
+		case 'i':
+			cfg_ifname = optarg;
+			break;
+		default:
+			error(1, 0, "unknown option %c", optopt);
+			break;
+		}
+	}
+
+	if (!cfg_ifname)
+		error(1, 0, "must specify tap interface name (-i)");
+}
+
+int main(int argc, char **argv)
+{
+	int fdt, fdr;
+
+	parse_opts(argc, argv);
+
+	fdr = raw_open();
+	fdt = tun_open(cfg_ifname);
+
+	tun_write(fdt);
+	raw_read(fdr);
+
+	if (close(fdt))
+		error(1, errno, "close tun");
+	if (close(fdr))
+		error(1, errno, "close udp");
+
+	fprintf(stderr, "OK\n");
+	return 0;
+}
+
diff --git a/tools/testing/selftests/net/skf_net_off.sh b/tools/testing/selftests/net/skf_net_off.sh
new file mode 100755
index 000000000000..5da5066fb465
--- /dev/null
+++ b/tools/testing/selftests/net/skf_net_off.sh
@@ -0,0 +1,30 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+readonly NS="ns-$(mktemp -u XXXXXX)"
+
+cleanup() {
+	ip netns del $NS
+}
+
+ip netns add $NS
+trap cleanup EXIT
+
+ip -netns $NS link set lo up
+ip -netns $NS tuntap add name tap1 mode tap
+ip -netns $NS link set tap1 up
+ip -netns $NS link set dev tap1 addr 02:00:00:00:00:01
+ip -netns $NS -6 addr add fdab::1 peer fdab::2 dev tap1 nodad
+ip netns exec $NS ethtool -K tap1 gro off
+
+# disable early demux, else udp_v6_early_demux pulls udp header into linear
+ip netns exec $NS sysctl -w net.ipv4.ip_early_demux=0
+
+echo "no filter"
+ip netns exec $NS ./skf_net_off -i tap1
+
+echo "filter, linear skb (-f)"
+ip netns exec $NS ./skf_net_off -i tap1 -f
+
+echo "filter, fragmented skb (-f) (-F)"
+ip netns exec $NS ./skf_net_off -i tap1 -f -F
diff --git a/tools/testing/selftests/net/tls.c b/tools/testing/selftests/net/tls.c
index 9a85f93c33d8..5ded3b3a7538 100644
--- a/tools/testing/selftests/net/tls.c
+++ b/tools/testing/selftests/net/tls.c
@@ -1753,6 +1753,42 @@ TEST_F(tls_basic, rekey_tx)
 	EXPECT_EQ(memcmp(buf, test_str, send_len), 0);
 }
 
+TEST_F(tls_basic, disconnect)
+{
+	char const *test_str = "test_message";
+	int send_len = strlen(test_str) + 1;
+	struct tls_crypto_info_keys key;
+	struct sockaddr_in addr;
+	char buf[20];
+	int ret;
+
+	if (self->notls)
+		return;
+
+	tls_crypto_info_init(TLS_1_3_VERSION, TLS_CIPHER_AES_GCM_128,
+			     &key, 0);
+
+	ret = setsockopt(self->fd, SOL_TLS, TLS_TX, &key, key.len);
+	ASSERT_EQ(ret, 0);
+
+	/* Pre-queue the data so that setsockopt parses it but doesn't
+	 * dequeue it from the TCP socket. recvmsg would dequeue.
+	 */
+	EXPECT_EQ(send(self->fd, test_str, send_len, 0), send_len);
+
+	ret = setsockopt(self->cfd, SOL_TLS, TLS_RX, &key, key.len);
+	ASSERT_EQ(ret, 0);
+
+	addr.sin_family = AF_UNSPEC;
+	addr.sin_addr.s_addr = htonl(INADDR_ANY);
+	addr.sin_port = 0;
+	ret = connect(self->cfd, &addr, sizeof(addr));
+	EXPECT_EQ(ret, -1);
+	EXPECT_EQ(errno, EOPNOTSUPP);
+
+	EXPECT_EQ(recv(self->cfd, buf, send_len, 0), send_len);
+}
+
 TEST_F(tls, rekey)
 {
 	char const *test_str_1 = "test_message_before_rekey";
diff --git a/tools/testing/selftests/pcie_bwctrl/Makefile b/tools/testing/selftests/pcie_bwctrl/Makefile
index 48ec048f47af..277f92f9d753 100644
--- a/tools/testing/selftests/pcie_bwctrl/Makefile
+++ b/tools/testing/selftests/pcie_bwctrl/Makefile
@@ -1,2 +1,3 @@
-TEST_PROGS = set_pcie_cooling_state.sh set_pcie_speed.sh
+TEST_PROGS = set_pcie_cooling_state.sh
+TEST_FILES = set_pcie_speed.sh
 include ../lib.mk
diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json b/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json
index 1ba96c467754..d9fc62ab476c 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/infra/actions.json
@@ -412,5 +412,27 @@
         "teardown": [
             "$TC qdisc del dev $DUMMY ingress"
         ]
+    },
+    {
+        "id": "33f4",
+        "name": "Check echo of big filter command",
+        "category": [
+            "infra",
+            "u32"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$TC qdisc add dev $DUMMY parent root handle 10: fq_codel"
+        ],
+        "cmdUnderTest": "bash -c '$TC -echo filter add dev $DUMMY parent 10: u32 match u32 0 0 $(for i in $(seq 32); do echo action pedit munge ip dport set 22; done) | grep \"added filter\"'",
+        "verifyCmd": "",
+        "expExitCode": "0",
+        "matchCount": "0",
+        "matchPattern": "",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY parent root fq_codel"
+        ]
     }
 ]
diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json
index 25454fd95537..0843f6d37e9c 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json
@@ -158,5 +158,385 @@
             "$TC qdisc del dev $DUMMY handle 1: root",
             "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
         ]
+    },
+    {
+        "id": "a4bb",
+        "name": "Test FQ_CODEL with HTB parent - force packet drop with empty queue",
+        "category": [
+            "qdisc",
+            "fq_codel",
+            "htb"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1: root htb default 10",
+            "$TC class add dev $DUMMY parent 1: classid 1:10 htb rate 1kbit",
+            "$TC qdisc add dev $DUMMY parent 1:10 handle 10: fq_codel memory_limit 1 flows 1 target 0.1ms interval 1ms",
+            "$TC filter add dev $DUMMY parent 1: protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:10",
+            "ping -c 5 -f -I $DUMMY 10.10.10.1 > /dev/null || true",
+            "sleep 0.1"
+        ],
+        "cmdUnderTest": "$TC -s qdisc show dev $DUMMY",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -s qdisc show dev $DUMMY | grep -A 5 'qdisc fq_codel'",
+        "matchPattern": "dropped [1-9][0-9]*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "a4be",
+        "name": "Test FQ_CODEL with QFQ parent - force packet drop with empty queue",
+        "category": [
+            "qdisc",
+            "fq_codel",
+            "qfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1: root qfq",
+            "$TC class add dev $DUMMY parent 1: classid 1:10 qfq weight 1 maxpkt 1000",
+            "$TC qdisc add dev $DUMMY parent 1:10 handle 10: fq_codel memory_limit 1 flows 1 target 0.1ms interval 1ms",
+            "$TC filter add dev $DUMMY parent 1: protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:10",
+            "ping -c 10 -s 1000 -f -I $DUMMY 10.10.10.1 > /dev/null || true",
+            "sleep 0.1"
+        ],
+        "cmdUnderTest": "$TC -s qdisc show dev $DUMMY",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -s qdisc show dev $DUMMY | grep -A 5 'qdisc fq_codel'",
+        "matchPattern": "dropped [1-9][0-9]*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "a4bf",
+        "name": "Test FQ_CODEL with HFSC parent - force packet drop with empty queue",
+        "category": [
+            "qdisc",
+            "fq_codel",
+            "hfsc"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1: root hfsc default 10",
+            "$TC class add dev $DUMMY parent 1: classid 1:10 hfsc sc rate 1kbit ul rate 1kbit",
+            "$TC qdisc add dev $DUMMY parent 1:10 handle 10: fq_codel memory_limit 1 flows 1 target 0.1ms interval 1ms",
+            "$TC filter add dev $DUMMY parent 1: protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:10",
+            "ping -c 5 -f -I $DUMMY 10.10.10.1 > /dev/null || true",
+            "sleep 0.1"
+        ],
+        "cmdUnderTest": "$TC -s qdisc show dev $DUMMY",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -s qdisc show dev $DUMMY | grep -A 5 'qdisc fq_codel'",
+        "matchPattern": "dropped [1-9][0-9]*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "a4c0",
+        "name": "Test FQ_CODEL with DRR parent - force packet drop with empty queue",
+        "category": [
+            "qdisc",
+            "fq_codel",
+            "drr"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1: root drr",
+            "$TC class add dev $DUMMY parent 1: classid 1:10 drr quantum 1500",
+            "$TC qdisc add dev $DUMMY parent 1:10 handle 10: fq_codel memory_limit 1 flows 1 target 0.1ms interval 1ms",
+            "$TC filter add dev $DUMMY parent 1: protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:10",
+            "ping -c 5 -f -I $DUMMY 10.10.10.1 > /dev/null || true",
+            "sleep 0.1"
+        ],
+        "cmdUnderTest": "$TC -s qdisc show dev $DUMMY",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -s qdisc show dev $DUMMY | grep -A 5 'qdisc fq_codel'",
+        "matchPattern": "dropped [1-9][0-9]*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "a4c1",
+        "name": "Test FQ_CODEL with ETS parent - force packet drop with empty queue",
+        "category": [
+            "qdisc",
+            "fq_codel",
+            "ets"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1: root ets bands 2 strict 1",
+            "$TC class change dev $DUMMY parent 1: classid 1:1 ets",
+            "$TC qdisc add dev $DUMMY parent 1:1 handle 10: fq_codel memory_limit 1 flows 1 target 0.1ms interval 1ms",
+            "$TC filter add dev $DUMMY parent 1: protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:1",
+            "ping -c 5 -f -I $DUMMY 10.10.10.1 > /dev/null || true",
+            "sleep 0.1"
+        ],
+        "cmdUnderTest": "$TC -s qdisc show dev $DUMMY",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -s qdisc show dev $DUMMY | grep -A 5 'qdisc fq_codel'",
+        "matchPattern": "dropped [1-9][0-9]*",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1: root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "a4c3",
+        "name": "Test HFSC with netem/blackhole - queue emptying during peek operation",
+        "category": [
+            "qdisc",
+            "hfsc",
+            "netem",
+            "blackhole"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1:0 root drr",
+            "$TC class add dev $DUMMY parent 1:0 classid 1:1 drr",
+            "$TC class add dev $DUMMY parent 1:0 classid 1:2 drr",
+            "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 plug limit 1024",
+            "$TC qdisc add dev $DUMMY parent 1:2 handle 3:0 hfsc default 1",
+            "$TC class add dev $DUMMY parent 3:0 classid 3:1 hfsc rt m1 5Mbit d 10ms m2 10Mbit",
+            "$TC qdisc add dev $DUMMY parent 3:1 handle 4:0 netem delay 1ms",
+            "$TC qdisc add dev $DUMMY parent 4:1 handle 5:0 blackhole",
+            "ping -c 3 -W 0.01 -i 0.001 -s 1 10.10.10.10 -I $DUMMY > /dev/null 2>&1 || true",
+            "$TC class change dev $DUMMY parent 3:0 classid 3:1 hfsc sc m1 5Mbit d 10ms m2 10Mbit",
+            "$TC class del dev $DUMMY parent 3:0 classid 3:1",
+            "$TC class add dev $DUMMY parent 3:0 classid 3:1 hfsc rt m1 5Mbit d 10ms m2 10Mbit",
+            "ping -c 3 -W 0.01 -i 0.001 -s 1 10.10.10.10 -I $DUMMY > /dev/null 2>&1 || true"
+        ],
+        "cmdUnderTest": "$TC class change dev $DUMMY parent 3:0 classid 3:1 hfsc sc m1 5Mbit d 10ms m2 10Mbit",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -s qdisc show dev $DUMMY",
+        "matchPattern": "qdisc hfsc 3:.*parent 1:2.*default 1",
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1:0 root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "90ec",
+        "name": "Test DRR's enqueue reentrant behaviour with netem",
+        "category": [
+            "qdisc",
+            "drr"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1:0 root drr",
+            "$TC class replace dev $DUMMY parent 1:0 classid 1:1 drr",
+            "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 netem duplicate 100%",
+            "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:1"
+        ],
+        "cmdUnderTest": "ping -c 1 -I $DUMMY 10.10.10.1 > /dev/null || true",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -j -s qdisc ls dev $DUMMY handle 1:0",
+        "matchJSON": [
+            {
+                "kind": "drr",
+                "handle": "1:",
+                "bytes": 196,
+                "packets": 2
+            }
+        ],
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1:0 root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "1f1f",
+        "name": "Test ETS's enqueue reentrant behaviour with netem",
+        "category": [
+            "qdisc",
+            "ets"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1:0 root ets bands 2",
+            "$TC class replace dev $DUMMY parent 1:0 classid 1:1 ets quantum 1500",
+            "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 netem duplicate 100%",
+            "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:1"
+        ],
+        "cmdUnderTest": "ping -c 1 -I $DUMMY 10.10.10.1 > /dev/null || true",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -j -s class show dev $DUMMY",
+        "matchJSON": [
+            {
+                "class": "ets",
+                "handle": "1:1",
+                "stats": {
+                    "bytes": 196,
+                    "packets": 2
+                }
+            }
+        ],
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1:0 root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "5e6d",
+        "name": "Test QFQ's enqueue reentrant behaviour with netem",
+        "category": [
+            "qdisc",
+            "qfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1:0 root qfq",
+            "$TC class replace dev $DUMMY parent 1:0 classid 1:1 qfq weight 100 maxpkt 1500",
+            "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 netem duplicate 100%",
+            "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:1"
+        ],
+        "cmdUnderTest": "ping -c 1 -I $DUMMY 10.10.10.1 > /dev/null || true",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -j -s qdisc ls dev $DUMMY handle 1:0",
+        "matchJSON": [
+            {
+                "kind": "qfq",
+                "handle": "1:",
+                "bytes": 196,
+                "packets": 2
+            }
+        ],
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1:0 root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "bf1d",
+        "name": "Test HFSC's enqueue reentrant behaviour with netem",
+        "category": [
+            "qdisc",
+            "hfsc"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1:0 root hfsc",
+            "$TC class add dev $DUMMY parent 1:0 classid 1:1 hfsc ls m2 10Mbit",
+            "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 netem duplicate 100%",
+            "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 1 u32 match ip dst 10.10.10.1/32 flowid 1:1",
+            "$TC class add dev $DUMMY parent 1:0 classid 1:2 hfsc ls m2 10Mbit",
+            "$TC qdisc add dev $DUMMY parent 1:2 handle 3:0 netem duplicate 100%",
+            "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 2 u32 match ip dst 10.10.10.2/32 flowid 1:2",
+            "ping -c 1 10.10.10.1 -I$DUMMY > /dev/null || true",
+            "$TC filter del dev $DUMMY parent 1:0 protocol ip prio 1",
+            "$TC class del dev $DUMMY classid 1:1"
+        ],
+        "cmdUnderTest": "ping -c 1 10.10.10.2 -I$DUMMY > /dev/null || true",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -j -s qdisc ls dev $DUMMY handle 1:0",
+        "matchJSON": [
+            {
+                "kind": "hfsc",
+                "handle": "1:",
+                "bytes": 392,
+                "packets": 4
+            }
+        ],
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1:0 root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
+    },
+    {
+        "id": "7c3b",
+        "name": "Test nested DRR's enqueue reentrant behaviour with netem",
+        "category": [
+            "qdisc",
+            "drr"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [
+            "$IP link set dev $DUMMY up || true",
+            "$IP addr add 10.10.10.10/24 dev $DUMMY || true",
+            "$TC qdisc add dev $DUMMY handle 1:0 root drr",
+            "$TC class add dev $DUMMY parent 1:0 classid 1:1 drr",
+            "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 1 u32 match ip protocol 1 0xff flowid 1:1",
+            "$TC qdisc add dev $DUMMY handle 2:0 parent 1:1 drr",
+            "$TC class add dev $DUMMY classid 2:1 parent 2:0 drr",
+            "$TC filter add dev $DUMMY parent 2:0 protocol ip prio 1 u32 match ip protocol 1 0xff flowid 2:1",
+            "$TC qdisc add dev $DUMMY parent 2:1 handle 3:0 netem duplicate 100%"
+        ],
+        "cmdUnderTest": "ping -c 1 -I $DUMMY 10.10.10.1 > /dev/null || true",
+        "expExitCode": "0",
+        "verifyCmd": "$TC -j -s qdisc ls dev $DUMMY handle 1:0",
+        "matchJSON": [
+            {
+                "kind": "drr",
+                "handle": "1:",
+                "bytes": 196,
+                "packets": 2
+            }
+        ],
+        "matchCount": "1",
+        "teardown": [
+            "$TC qdisc del dev $DUMMY handle 1:0 root",
+            "$IP addr del 10.10.10.10/24 dev $DUMMY || true"
+        ]
     }
 ]
diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/sfq.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/sfq.json
index 50e8d72781cb..28c6ce6da7db 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/sfq.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/sfq.json
@@ -228,5 +228,41 @@
         "matchCount": "0",
         "teardown": [
         ]
+    },
+    {
+        "id": "7f8f",
+        "name": "Check that a derived limit of 1 is rejected (limit 2 depth 1 flows 1)",
+        "category": [
+            "qdisc",
+            "sfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root sfq limit 2 depth 1 flows 1",
+        "expExitCode": "2",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "sfq",
+        "matchCount": "0",
+        "teardown": []
+    },
+    {
+        "id": "5168",
+        "name": "Check that a derived limit of 1 is rejected (limit 2 depth 1 divisor 1)",
+        "category": [
+            "qdisc",
+            "sfq"
+        ],
+        "plugins": {
+            "requires": "nsPlugin"
+        },
+        "setup": [],
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root sfq limit 2 depth 1 divisor 1",
+        "expExitCode": "2",
+        "verifyCmd": "$TC qdisc show dev $DUMMY",
+        "matchPattern": "sfq",
+        "matchCount": "0",
+        "teardown": []
     }
 ]
diff --git a/tools/testing/selftests/tpm2/.gitignore b/tools/testing/selftests/tpm2/.gitignore
new file mode 100644
index 000000000000..6d6165c5e35d
--- /dev/null
+++ b/tools/testing/selftests/tpm2/.gitignore
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+AsyncTest.log
+SpaceTest.log
diff --git a/tools/testing/selftests/tpm2/test_smoke.sh b/tools/testing/selftests/tpm2/test_smoke.sh
index 168f4b166234..3a60e6c6f5c9 100755
--- a/tools/testing/selftests/tpm2/test_smoke.sh
+++ b/tools/testing/selftests/tpm2/test_smoke.sh
@@ -6,6 +6,6 @@ ksft_skip=4
 
 [ -e /dev/tpm0 ] || exit $ksft_skip
 read tpm_version < /sys/class/tpm/tpm0/tpm_version_major
-[ "$tpm_version" == 2 ] || exit $ksft_skip
+[ "$tpm_version" = 2 ] || exit $ksft_skip
 
 python3 -m unittest -v tpm2_tests.SmokeTest 2>&1
diff --git a/tools/testing/selftests/ublk/Makefile b/tools/testing/selftests/ublk/Makefile
index c7781efea0f3..f34ac0bac696 100644
--- a/tools/testing/selftests/ublk/Makefile
+++ b/tools/testing/selftests/ublk/Makefile
@@ -6,6 +6,10 @@ LDLIBS += -lpthread -lm -luring
 TEST_PROGS := test_generic_01.sh
 TEST_PROGS += test_generic_02.sh
 TEST_PROGS += test_generic_03.sh
+TEST_PROGS += test_generic_04.sh
+TEST_PROGS += test_generic_05.sh
+TEST_PROGS += test_generic_06.sh
+TEST_PROGS += test_generic_07.sh
 
 TEST_PROGS += test_null_01.sh
 TEST_PROGS += test_null_02.sh
@@ -21,12 +25,16 @@ TEST_PROGS += test_stripe_04.sh
 
 TEST_PROGS += test_stress_01.sh
 TEST_PROGS += test_stress_02.sh
+TEST_PROGS += test_stress_03.sh
+TEST_PROGS += test_stress_04.sh
+TEST_PROGS += test_stress_05.sh
 
 TEST_GEN_PROGS_EXTENDED = kublk
 
 include ../lib.mk
 
-$(TEST_GEN_PROGS_EXTENDED): kublk.c null.c file_backed.c common.c stripe.c
+$(TEST_GEN_PROGS_EXTENDED): kublk.c null.c file_backed.c common.c stripe.c \
+	fault_inject.c
 
 check:
 	shellcheck -x -f gcc *.sh
diff --git a/tools/testing/selftests/ublk/fault_inject.c b/tools/testing/selftests/ublk/fault_inject.c
new file mode 100644
index 000000000000..94a8e729ba4c
--- /dev/null
+++ b/tools/testing/selftests/ublk/fault_inject.c
@@ -0,0 +1,98 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Fault injection ublk target. Hack this up however you like for
+ * testing specific behaviors of ublk_drv. Currently is a null target
+ * with a configurable delay before completing each I/O. This delay can
+ * be used to test ublk_drv's handling of I/O outstanding to the ublk
+ * server when it dies.
+ */
+
+#include "kublk.h"
+
+static int ublk_fault_inject_tgt_init(const struct dev_ctx *ctx,
+				      struct ublk_dev *dev)
+{
+	const struct ublksrv_ctrl_dev_info *info = &dev->dev_info;
+	unsigned long dev_size = 250UL << 30;
+
+	dev->tgt.dev_size = dev_size;
+	dev->tgt.params = (struct ublk_params) {
+		.types = UBLK_PARAM_TYPE_BASIC,
+		.basic = {
+			.logical_bs_shift	= 9,
+			.physical_bs_shift	= 12,
+			.io_opt_shift		= 12,
+			.io_min_shift		= 9,
+			.max_sectors		= info->max_io_buf_bytes >> 9,
+			.dev_sectors		= dev_size >> 9,
+		},
+	};
+
+	dev->private_data = (void *)(unsigned long)(ctx->fault_inject.delay_us * 1000);
+	return 0;
+}
+
+static int ublk_fault_inject_queue_io(struct ublk_queue *q, int tag)
+{
+	const struct ublksrv_io_desc *iod = ublk_get_iod(q, tag);
+	struct io_uring_sqe *sqe;
+	struct __kernel_timespec ts = {
+		.tv_nsec = (long long)q->dev->private_data,
+	};
+
+	ublk_queue_alloc_sqes(q, &sqe, 1);
+	io_uring_prep_timeout(sqe, &ts, 1, 0);
+	sqe->user_data = build_user_data(tag, ublksrv_get_op(iod), 0, 1);
+
+	ublk_queued_tgt_io(q, tag, 1);
+
+	return 0;
+}
+
+static void ublk_fault_inject_tgt_io_done(struct ublk_queue *q, int tag,
+					  const struct io_uring_cqe *cqe)
+{
+	const struct ublksrv_io_desc *iod = ublk_get_iod(q, tag);
+
+	if (cqe->res != -ETIME)
+		ublk_err("%s: unexpected cqe res %d\n", __func__, cqe->res);
+
+	if (ublk_completed_tgt_io(q, tag))
+		ublk_complete_io(q, tag, iod->nr_sectors << 9);
+	else
+		ublk_err("%s: io not complete after 1 cqe\n", __func__);
+}
+
+static void ublk_fault_inject_cmd_line(struct dev_ctx *ctx, int argc, char *argv[])
+{
+	static const struct option longopts[] = {
+		{ "delay_us", 	1,	NULL,  0  },
+		{ 0, 0, 0, 0 }
+	};
+	int option_idx, opt;
+
+	ctx->fault_inject.delay_us = 0;
+	while ((opt = getopt_long(argc, argv, "",
+				  longopts, &option_idx)) != -1) {
+		switch (opt) {
+		case 0:
+			if (!strcmp(longopts[option_idx].name, "delay_us"))
+				ctx->fault_inject.delay_us = strtoll(optarg, NULL, 10);
+		}
+	}
+}
+
+static void ublk_fault_inject_usage(const struct ublk_tgt_ops *ops)
+{
+	printf("\tfault_inject: [--delay_us us (default 0)]\n");
+}
+
+const struct ublk_tgt_ops fault_inject_tgt_ops = {
+	.name = "fault_inject",
+	.init_tgt = ublk_fault_inject_tgt_init,
+	.queue_io = ublk_fault_inject_queue_io,
+	.tgt_io_done = ublk_fault_inject_tgt_io_done,
+	.parse_cmd_line = ublk_fault_inject_cmd_line,
+	.usage = ublk_fault_inject_usage,
+};
diff --git a/tools/testing/selftests/ublk/kublk.c b/tools/testing/selftests/ublk/kublk.c
index 91c282bc7674..842b40736a9b 100644
--- a/tools/testing/selftests/ublk/kublk.c
+++ b/tools/testing/selftests/ublk/kublk.c
@@ -5,22 +5,24 @@
 
 #include "kublk.h"
 
+#define MAX_NR_TGT_ARG 	64
+
 unsigned int ublk_dbg_mask = UBLK_LOG;
 static const struct ublk_tgt_ops *tgt_ops_list[] = {
 	&null_tgt_ops,
 	&loop_tgt_ops,
 	&stripe_tgt_ops,
+	&fault_inject_tgt_ops,
 };
 
 static const struct ublk_tgt_ops *ublk_find_tgt(const char *name)
 {
-	const struct ublk_tgt_ops *ops;
 	int i;
 
 	if (name == NULL)
 		return NULL;
 
-	for (i = 0; sizeof(tgt_ops_list) / sizeof(ops); i++)
+	for (i = 0; i < ARRAY_SIZE(tgt_ops_list); i++)
 		if (strcmp(tgt_ops_list[i]->name, name) == 0)
 			return tgt_ops_list[i];
 	return NULL;
@@ -118,6 +120,27 @@ static int ublk_ctrl_start_dev(struct ublk_dev *dev,
 	return __ublk_ctrl_cmd(dev, &data);
 }
 
+static int ublk_ctrl_start_user_recovery(struct ublk_dev *dev)
+{
+	struct ublk_ctrl_cmd_data data = {
+		.cmd_op	= UBLK_U_CMD_START_USER_RECOVERY,
+	};
+
+	return __ublk_ctrl_cmd(dev, &data);
+}
+
+static int ublk_ctrl_end_user_recovery(struct ublk_dev *dev, int daemon_pid)
+{
+	struct ublk_ctrl_cmd_data data = {
+		.cmd_op	= UBLK_U_CMD_END_USER_RECOVERY,
+		.flags	= CTRL_CMD_HAS_DATA,
+	};
+
+	dev->dev_info.ublksrv_pid = data.data[0] = daemon_pid;
+
+	return __ublk_ctrl_cmd(dev, &data);
+}
+
 static int ublk_ctrl_add_dev(struct ublk_dev *dev)
 {
 	struct ublk_ctrl_cmd_data data = {
@@ -207,10 +230,73 @@ static const char *ublk_dev_state_desc(struct ublk_dev *dev)
 	};
 }
 
+static void ublk_print_cpu_set(const cpu_set_t *set, char *buf, unsigned len)
+{
+	unsigned done = 0;
+	int i;
+
+	for (i = 0; i < CPU_SETSIZE; i++) {
+		if (CPU_ISSET(i, set))
+			done += snprintf(&buf[done], len - done, "%d ", i);
+	}
+}
+
+static void ublk_adjust_affinity(cpu_set_t *set)
+{
+	int j, updated = 0;
+
+	/*
+	 * Just keep the 1st CPU now.
+	 *
+	 * In future, auto affinity selection can be tried.
+	 */
+	for (j = 0; j < CPU_SETSIZE; j++) {
+		if (CPU_ISSET(j, set)) {
+			if (!updated) {
+				updated = 1;
+				continue;
+			}
+			CPU_CLR(j, set);
+		}
+	}
+}
+
+/* Caller must free the allocated buffer */
+static int ublk_ctrl_get_affinity(struct ublk_dev *ctrl_dev, cpu_set_t **ptr_buf)
+{
+	struct ublk_ctrl_cmd_data data = {
+		.cmd_op	= UBLK_U_CMD_GET_QUEUE_AFFINITY,
+		.flags	= CTRL_CMD_HAS_DATA | CTRL_CMD_HAS_BUF,
+	};
+	cpu_set_t *buf;
+	int i, ret;
+
+	buf = malloc(sizeof(cpu_set_t) * ctrl_dev->dev_info.nr_hw_queues);
+	if (!buf)
+		return -ENOMEM;
+
+	for (i = 0; i < ctrl_dev->dev_info.nr_hw_queues; i++) {
+		data.data[0] = i;
+		data.len = sizeof(cpu_set_t);
+		data.addr = (__u64)&buf[i];
+
+		ret = __ublk_ctrl_cmd(ctrl_dev, &data);
+		if (ret < 0) {
+			free(buf);
+			return ret;
+		}
+		ublk_adjust_affinity(&buf[i]);
+	}
+
+	*ptr_buf = buf;
+	return 0;
+}
+
 static void ublk_ctrl_dump(struct ublk_dev *dev)
 {
 	struct ublksrv_ctrl_dev_info *info = &dev->dev_info;
 	struct ublk_params p;
+	cpu_set_t *affinity;
 	int ret;
 
 	ret = ublk_ctrl_get_params(dev, &p);
@@ -219,12 +305,31 @@ static void ublk_ctrl_dump(struct ublk_dev *dev)
 		return;
 	}
 
+	ret = ublk_ctrl_get_affinity(dev, &affinity);
+	if (ret < 0) {
+		ublk_err("failed to get affinity %m\n");
+		return;
+	}
+
 	ublk_log("dev id %d: nr_hw_queues %d queue_depth %d block size %d dev_capacity %lld\n",
 			info->dev_id, info->nr_hw_queues, info->queue_depth,
 			1 << p.basic.logical_bs_shift, p.basic.dev_sectors);
 	ublk_log("\tmax rq size %d daemon pid %d flags 0x%llx state %s\n",
 			info->max_io_buf_bytes, info->ublksrv_pid, info->flags,
 			ublk_dev_state_desc(dev));
+
+	if (affinity) {
+		char buf[512];
+		int i;
+
+		for (i = 0; i < info->nr_hw_queues; i++) {
+			ublk_print_cpu_set(&affinity[i], buf, sizeof(buf));
+			printf("\tqueue %u: tid %d affinity(%s)\n",
+					i, dev->q[i].tid, buf);
+		}
+		free(affinity);
+	}
+
 	fflush(stdout);
 }
 
@@ -347,7 +452,9 @@ static int ublk_queue_init(struct ublk_queue *q)
 	}
 
 	ret = ublk_setup_ring(&q->ring, ring_depth, cq_depth,
-			IORING_SETUP_COOP_TASKRUN);
+			IORING_SETUP_COOP_TASKRUN |
+			IORING_SETUP_SINGLE_ISSUER |
+			IORING_SETUP_DEFER_TASKRUN);
 	if (ret < 0) {
 		ublk_err("ublk dev %d queue %d setup io_uring failed %d\n",
 				q->dev->dev_info.dev_id, q->q_id, ret);
@@ -429,12 +536,17 @@ int ublk_queue_io_cmd(struct ublk_queue *q, struct ublk_io *io, unsigned tag)
 	if (!(io->flags & UBLKSRV_IO_FREE))
 		return 0;
 
-	/* we issue because we need either fetching or committing */
+	/*
+	 * we issue because we need either fetching or committing or
+	 * getting data
+	 */
 	if (!(io->flags &
-		(UBLKSRV_NEED_FETCH_RQ | UBLKSRV_NEED_COMMIT_RQ_COMP)))
+		(UBLKSRV_NEED_FETCH_RQ | UBLKSRV_NEED_COMMIT_RQ_COMP | UBLKSRV_NEED_GET_DATA)))
 		return 0;
 
-	if (io->flags & UBLKSRV_NEED_COMMIT_RQ_COMP)
+	if (io->flags & UBLKSRV_NEED_GET_DATA)
+		cmd_op = UBLK_U_IO_NEED_GET_DATA;
+	else if (io->flags & UBLKSRV_NEED_COMMIT_RQ_COMP)
 		cmd_op = UBLK_U_IO_COMMIT_AND_FETCH_REQ;
 	else if (io->flags & UBLKSRV_NEED_FETCH_RQ)
 		cmd_op = UBLK_U_IO_FETCH_REQ;
@@ -551,6 +663,9 @@ static void ublk_handle_cqe(struct io_uring *r,
 		assert(tag < q->q_depth);
 		if (q->tgt_ops->queue_io)
 			q->tgt_ops->queue_io(q, tag);
+	} else if (cqe->res == UBLK_IO_RES_NEED_GET_DATA) {
+		io->flags |= UBLKSRV_NEED_GET_DATA | UBLKSRV_IO_FREE;
+		ublk_queue_io_cmd(q, io, tag);
 	} else {
 		/*
 		 * COMMIT_REQ will be completed immediately since no fetching
@@ -602,9 +717,24 @@ static int ublk_process_io(struct ublk_queue *q)
 	return reapped;
 }
 
+static void ublk_queue_set_sched_affinity(const struct ublk_queue *q,
+		cpu_set_t *cpuset)
+{
+        if (sched_setaffinity(0, sizeof(*cpuset), cpuset) < 0)
+                ublk_err("ublk dev %u queue %u set affinity failed",
+                                q->dev->dev_info.dev_id, q->q_id);
+}
+
+struct ublk_queue_info {
+	struct ublk_queue 	*q;
+	sem_t 			*queue_sem;
+	cpu_set_t 		*affinity;
+};
+
 static void *ublk_io_handler_fn(void *data)
 {
-	struct ublk_queue *q = data;
+	struct ublk_queue_info *info = data;
+	struct ublk_queue *q = info->q;
 	int dev_id = q->dev->dev_info.dev_id;
 	int ret;
 
@@ -614,6 +744,10 @@ static void *ublk_io_handler_fn(void *data)
 				dev_id, q->q_id);
 		return NULL;
 	}
+	/* IO perf is sensitive with queue pthread affinity on NUMA machine*/
+	ublk_queue_set_sched_affinity(q, info->affinity);
+	sem_post(info->queue_sem);
+
 	ublk_dbg(UBLK_DBG_QUEUE, "tid %d: ublk dev %d queue %d started\n",
 			q->tid, dev_id, q->q_id);
 
@@ -639,7 +773,7 @@ static void ublk_set_parameters(struct ublk_dev *dev)
 				dev->dev_info.dev_id, ret);
 }
 
-static int ublk_send_dev_event(const struct dev_ctx *ctx, int dev_id)
+static int ublk_send_dev_event(const struct dev_ctx *ctx, struct ublk_dev *dev, int dev_id)
 {
 	uint64_t id;
 	int evtfd = ctx->_evtfd;
@@ -652,36 +786,68 @@ static int ublk_send_dev_event(const struct dev_ctx *ctx, int dev_id)
 	else
 		id = ERROR_EVTFD_DEVID;
 
+	if (dev && ctx->shadow_dev)
+		memcpy(&ctx->shadow_dev->q, &dev->q, sizeof(dev->q));
+
 	if (write(evtfd, &id, sizeof(id)) != sizeof(id))
 		return -EINVAL;
 
+	close(evtfd);
+	shmdt(ctx->shadow_dev);
+
 	return 0;
 }
 
 
 static int ublk_start_daemon(const struct dev_ctx *ctx, struct ublk_dev *dev)
 {
-	int ret, i;
-	void *thread_ret;
 	const struct ublksrv_ctrl_dev_info *dinfo = &dev->dev_info;
+	struct ublk_queue_info *qinfo;
+	cpu_set_t *affinity_buf;
+	void *thread_ret;
+	sem_t queue_sem;
+	int ret, i;
 
 	ublk_dbg(UBLK_DBG_DEV, "%s enter\n", __func__);
 
+	qinfo = (struct ublk_queue_info *)calloc(sizeof(struct ublk_queue_info),
+			dinfo->nr_hw_queues);
+	if (!qinfo)
+		return -ENOMEM;
+
+	sem_init(&queue_sem, 0, 0);
 	ret = ublk_dev_prep(ctx, dev);
 	if (ret)
 		return ret;
 
+	ret = ublk_ctrl_get_affinity(dev, &affinity_buf);
+	if (ret)
+		return ret;
+
 	for (i = 0; i < dinfo->nr_hw_queues; i++) {
 		dev->q[i].dev = dev;
 		dev->q[i].q_id = i;
+
+		qinfo[i].q = &dev->q[i];
+		qinfo[i].queue_sem = &queue_sem;
+		qinfo[i].affinity = &affinity_buf[i];
 		pthread_create(&dev->q[i].thread, NULL,
 				ublk_io_handler_fn,
-				&dev->q[i]);
+				&qinfo[i]);
 	}
 
+	for (i = 0; i < dinfo->nr_hw_queues; i++)
+		sem_wait(&queue_sem);
+	free(qinfo);
+	free(affinity_buf);
+
 	/* everything is fine now, start us */
-	ublk_set_parameters(dev);
-	ret = ublk_ctrl_start_dev(dev, getpid());
+	if (ctx->recovery)
+		ret = ublk_ctrl_end_user_recovery(dev, getpid());
+	else {
+		ublk_set_parameters(dev);
+		ret = ublk_ctrl_start_dev(dev, getpid());
+	}
 	if (ret < 0) {
 		ublk_err("%s: ublk_ctrl_start_dev failed: %d\n", __func__, ret);
 		goto fail;
@@ -691,7 +857,7 @@ static int ublk_start_daemon(const struct dev_ctx *ctx, struct ublk_dev *dev)
 	if (ctx->fg)
 		ublk_ctrl_dump(dev);
 	else
-		ublk_send_dev_event(ctx, dev->dev_info.dev_id);
+		ublk_send_dev_event(ctx, dev, dev->dev_info.dev_id);
 
 	/* wait until we are terminated */
 	for (i = 0; i < dinfo->nr_hw_queues; i++)
@@ -856,7 +1022,10 @@ static int __cmd_dev_add(const struct dev_ctx *ctx)
 		}
 	}
 
-	ret = ublk_ctrl_add_dev(dev);
+	if (ctx->recovery)
+		ret = ublk_ctrl_start_user_recovery(dev);
+	else
+		ret = ublk_ctrl_add_dev(dev);
 	if (ret < 0) {
 		ublk_err("%s: can't add dev id %d, type %s ret %d\n",
 				__func__, dev_id, tgt_type, ret);
@@ -870,7 +1039,7 @@ static int __cmd_dev_add(const struct dev_ctx *ctx)
 
 fail:
 	if (ret < 0)
-		ublk_send_dev_event(ctx, -1);
+		ublk_send_dev_event(ctx, dev, -1);
 	ublk_ctrl_deinit(dev);
 	return ret;
 }
@@ -884,30 +1053,58 @@ static int cmd_dev_add(struct dev_ctx *ctx)
 	if (ctx->fg)
 		goto run;
 
+	ctx->_shmid = shmget(IPC_PRIVATE, sizeof(struct ublk_dev), IPC_CREAT | 0666);
+	if (ctx->_shmid < 0) {
+		ublk_err("%s: failed to shmget %s\n", __func__, strerror(errno));
+		exit(-1);
+	}
+	ctx->shadow_dev = (struct ublk_dev *)shmat(ctx->_shmid, NULL, 0);
+	if (ctx->shadow_dev == (struct ublk_dev *)-1) {
+		ublk_err("%s: failed to shmat %s\n", __func__, strerror(errno));
+		exit(-1);
+	}
 	ctx->_evtfd = eventfd(0, 0);
 	if (ctx->_evtfd < 0) {
 		ublk_err("%s: failed to create eventfd %s\n", __func__, strerror(errno));
 		exit(-1);
 	}
 
-	setsid();
 	res = fork();
 	if (res == 0) {
+		int res2;
+
+		setsid();
+		res2 = fork();
+		if (res2 == 0) {
+			/* prepare for detaching */
+			close(STDIN_FILENO);
+			close(STDOUT_FILENO);
+			close(STDERR_FILENO);
 run:
-		res = __cmd_dev_add(ctx);
-		return res;
+			res = __cmd_dev_add(ctx);
+			return res;
+		} else {
+			/* detached from the foreground task */
+			exit(EXIT_SUCCESS);
+		}
 	} else if (res > 0) {
 		uint64_t id;
+		int exit_code = EXIT_FAILURE;
 
 		res = read(ctx->_evtfd, &id, sizeof(id));
 		close(ctx->_evtfd);
 		if (res == sizeof(id) && id != ERROR_EVTFD_DEVID) {
 			ctx->dev_id = id - 1;
-			return __cmd_dev_list(ctx);
+			if (__cmd_dev_list(ctx) >= 0)
+				exit_code = EXIT_SUCCESS;
 		}
-		exit(EXIT_FAILURE);
+		shmdt(ctx->shadow_dev);
+		shmctl(ctx->_shmid, IPC_RMID, NULL);
+		/* wait for child and detach from it */
+		wait(NULL);
+		exit(exit_code);
 	} else {
-		return res;
+		exit(EXIT_FAILURE);
 	}
 }
 
@@ -969,6 +1166,9 @@ static int __cmd_dev_list(struct dev_ctx *ctx)
 			ublk_err("%s: can't get dev info from %d: %d\n",
 					__func__, ctx->dev_id, ret);
 	} else {
+		if (ctx->shadow_dev)
+			memcpy(&dev->q, ctx->shadow_dev->q, sizeof(dev->q));
+
 		ublk_ctrl_dump(dev);
 	}
 
@@ -1039,14 +1239,47 @@ static int cmd_dev_get_features(void)
 	return ret;
 }
 
+static void __cmd_create_help(char *exe, bool recovery)
+{
+	int i;
+
+	printf("%s %s -t [null|loop|stripe|fault_inject] [-q nr_queues] [-d depth] [-n dev_id]\n",
+			exe, recovery ? "recover" : "add");
+	printf("\t[--foreground] [--quiet] [-z] [--debug_mask mask] [-r 0|1 ] [-g]\n");
+	printf("\t[-e 0|1 ] [-i 0|1]\n");
+	printf("\t[target options] [backfile1] [backfile2] ...\n");
+	printf("\tdefault: nr_queues=2(max 32), depth=128(max 1024), dev_id=-1(auto allocation)\n");
+
+	for (i = 0; i < sizeof(tgt_ops_list) / sizeof(tgt_ops_list[0]); i++) {
+		const struct ublk_tgt_ops *ops = tgt_ops_list[i];
+
+		if (ops->usage)
+			ops->usage(ops);
+	}
+}
+
+static void cmd_add_help(char *exe)
+{
+	__cmd_create_help(exe, false);
+	printf("\n");
+}
+
+static void cmd_recover_help(char *exe)
+{
+	__cmd_create_help(exe, true);
+	printf("\tPlease provide exact command line for creating this device with real dev_id\n");
+	printf("\n");
+}
+
 static int cmd_dev_help(char *exe)
 {
-	printf("%s add -t [null|loop] [-q nr_queues] [-d depth] [-n dev_id] [backfile1] [backfile2] ...\n", exe);
-	printf("\t default: nr_queues=2(max 4), depth=128(max 128), dev_id=-1(auto allocation)\n");
+	cmd_add_help(exe);
+	cmd_recover_help(exe);
+
 	printf("%s del [-n dev_id] -a \n", exe);
-	printf("\t -a delete all devices -n delete specified device\n");
+	printf("\t -a delete all devices -n delete specified device\n\n");
 	printf("%s list [-n dev_id] -a \n", exe);
-	printf("\t -a list all devices, -n list specified device, default -a \n");
+	printf("\t -a list all devices, -n list specified device, default -a \n\n");
 	printf("%s features\n", exe);
 	return 0;
 }
@@ -1063,9 +1296,13 @@ int main(int argc, char *argv[])
 		{ "quiet",		0,	NULL,  0  },
 		{ "zero_copy",          0,      NULL, 'z' },
 		{ "foreground",		0,	NULL,  0  },
-		{ "chunk_size", 	1,	NULL,  0  },
+		{ "recovery", 		1,      NULL, 'r' },
+		{ "recovery_fail_io",	1,	NULL, 'e'},
+		{ "recovery_reissue",	1,	NULL, 'i'},
+		{ "get_data",		1,	NULL, 'g'},
 		{ 0, 0, 0, 0 }
 	};
+	const struct ublk_tgt_ops *ops = NULL;
 	int option_idx, opt;
 	const char *cmd = argv[1];
 	struct dev_ctx ctx = {
@@ -1073,15 +1310,18 @@ int main(int argc, char *argv[])
 		.nr_hw_queues	=	2,
 		.dev_id		=	-1,
 		.tgt_type	=	"unknown",
-		.chunk_size 	= 	65536, 	/* def chunk size is 64K */
 	};
 	int ret = -EINVAL, i;
+	int tgt_argc = 1;
+	char *tgt_argv[MAX_NR_TGT_ARG] = { NULL };
+	int value;
 
 	if (argc == 1)
 		return ret;
 
+	opterr = 0;
 	optind = 2;
-	while ((opt = getopt_long(argc, argv, "t:n:d:q:az",
+	while ((opt = getopt_long(argc, argv, "t:n:d:q:r:e:i:gaz",
 				  longopts, &option_idx)) != -1) {
 		switch (opt) {
 		case 'a':
@@ -1103,6 +1343,24 @@ int main(int argc, char *argv[])
 		case 'z':
 			ctx.flags |= UBLK_F_SUPPORT_ZERO_COPY | UBLK_F_USER_COPY;
 			break;
+		case 'r':
+			value = strtol(optarg, NULL, 10);
+			if (value)
+				ctx.flags |= UBLK_F_USER_RECOVERY;
+			break;
+		case 'e':
+			value = strtol(optarg, NULL, 10);
+			if (value)
+				ctx.flags |= UBLK_F_USER_RECOVERY | UBLK_F_USER_RECOVERY_FAIL_IO;
+			break;
+		case 'i':
+			value = strtol(optarg, NULL, 10);
+			if (value)
+				ctx.flags |= UBLK_F_USER_RECOVERY | UBLK_F_USER_RECOVERY_REISSUE;
+			break;
+		case 'g':
+			ctx.flags |= UBLK_F_NEED_GET_DATA;
+			break;
 		case 0:
 			if (!strcmp(longopts[option_idx].name, "debug_mask"))
 				ublk_dbg_mask = strtol(optarg, NULL, 16);
@@ -1110,8 +1368,26 @@ int main(int argc, char *argv[])
 				ublk_dbg_mask = 0;
 			if (!strcmp(longopts[option_idx].name, "foreground"))
 				ctx.fg = 1;
-			if (!strcmp(longopts[option_idx].name, "chunk_size"))
-				ctx.chunk_size = strtol(optarg, NULL, 10);
+			break;
+		case '?':
+			/*
+			 * target requires every option must have argument
+			 */
+			if (argv[optind][0] == '-' || argv[optind - 1][0] != '-') {
+				fprintf(stderr, "every target option requires argument: %s %s\n",
+						argv[optind - 1], argv[optind]);
+				exit(EXIT_FAILURE);
+			}
+
+			if (tgt_argc < (MAX_NR_TGT_ARG - 1) / 2) {
+				tgt_argv[tgt_argc++] = argv[optind - 1];
+				tgt_argv[tgt_argc++] = argv[optind];
+			} else {
+				fprintf(stderr, "too many target options\n");
+				exit(EXIT_FAILURE);
+			}
+			optind += 1;
+			break;
 		}
 	}
 
@@ -1120,9 +1396,25 @@ int main(int argc, char *argv[])
 		ctx.files[ctx.nr_files++] = argv[i++];
 	}
 
+	ops = ublk_find_tgt(ctx.tgt_type);
+	if (ops && ops->parse_cmd_line) {
+		optind = 0;
+
+		tgt_argv[0] = ctx.tgt_type;
+		ops->parse_cmd_line(&ctx, tgt_argc, tgt_argv);
+	}
+
 	if (!strcmp(cmd, "add"))
 		ret = cmd_dev_add(&ctx);
-	else if (!strcmp(cmd, "del"))
+	else if (!strcmp(cmd, "recover")) {
+		if (ctx.dev_id < 0) {
+			fprintf(stderr, "device id isn't provided for recovering\n");
+			ret = -EINVAL;
+		} else {
+			ctx.recovery = 1;
+			ret = cmd_dev_add(&ctx);
+		}
+	} else if (!strcmp(cmd, "del"))
 		ret = cmd_dev_del(&ctx);
 	else if (!strcmp(cmd, "list")) {
 		ctx.all = 1;
diff --git a/tools/testing/selftests/ublk/kublk.h b/tools/testing/selftests/ublk/kublk.h
index 760ff8ffb810..44ee1e4ac55b 100644
--- a/tools/testing/selftests/ublk/kublk.h
+++ b/tools/testing/selftests/ublk/kublk.h
@@ -20,9 +20,15 @@
 #include <sys/wait.h>
 #include <sys/eventfd.h>
 #include <sys/uio.h>
+#include <sys/ipc.h>
+#include <sys/shm.h>
+#include <linux/io_uring.h>
 #include <liburing.h>
-#include <linux/ublk_cmd.h>
+#include <semaphore.h>
+
+/* allow ublk_dep.h to override ublk_cmd.h */
 #include "ublk_dep.h"
+#include <linux/ublk_cmd.h>
 
 #define __maybe_unused __attribute__((unused))
 #define MAX_BACK_FILES   4
@@ -30,6 +36,8 @@
 #define min(a, b) ((a) < (b) ? (a) : (b))
 #endif
 
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof(x[0]))
+
 /****************** part 1: libublk ********************/
 
 #define CTRL_DEV		"/dev/ublk-control"
@@ -42,8 +50,8 @@
 #define UBLKSRV_IO_IDLE_SECS		20
 
 #define UBLK_IO_MAX_BYTES               (1 << 20)
-#define UBLK_MAX_QUEUES                 4
-#define UBLK_QUEUE_DEPTH                128
+#define UBLK_MAX_QUEUES                 32
+#define UBLK_QUEUE_DEPTH                1024
 
 #define UBLK_DBG_DEV            (1U << 0)
 #define UBLK_DBG_QUEUE          (1U << 1)
@@ -55,6 +63,16 @@
 struct ublk_dev;
 struct ublk_queue;
 
+struct stripe_ctx {
+	/* stripe */
+	unsigned int    chunk_size;
+};
+
+struct fault_inject_ctx {
+	/* fault_inject */
+	unsigned long   delay_us;
+};
+
 struct dev_ctx {
 	char tgt_type[16];
 	unsigned long flags;
@@ -66,11 +84,18 @@ struct dev_ctx {
 	unsigned int	logging:1;
 	unsigned int	all:1;
 	unsigned int	fg:1;
-
-	/* stripe */
-	unsigned int    chunk_size;
+	unsigned int	recovery:1;
 
 	int _evtfd;
+	int _shmid;
+
+	/* built from shmem, only for ublk_dump_dev() */
+	struct ublk_dev *shadow_dev;
+
+	union {
+		struct stripe_ctx 	stripe;
+		struct fault_inject_ctx fault_inject;
+	};
 };
 
 struct ublk_ctrl_cmd_data {
@@ -90,6 +115,7 @@ struct ublk_io {
 #define UBLKSRV_NEED_FETCH_RQ		(1UL << 0)
 #define UBLKSRV_NEED_COMMIT_RQ_COMP	(1UL << 1)
 #define UBLKSRV_IO_FREE			(1UL << 2)
+#define UBLKSRV_NEED_GET_DATA           (1UL << 3)
 	unsigned short flags;
 	unsigned short refs;		/* used by target code only */
 
@@ -107,6 +133,14 @@ struct ublk_tgt_ops {
 	int (*queue_io)(struct ublk_queue *, int tag);
 	void (*tgt_io_done)(struct ublk_queue *,
 			int tag, const struct io_uring_cqe *);
+
+	/*
+	 * Target specific command line handling
+	 *
+	 * each option requires argument for target command line
+	 */
+	void (*parse_cmd_line)(struct dev_ctx *ctx, int argc, char *argv[]);
+	void (*usage)(const struct ublk_tgt_ops *ops);
 };
 
 struct ublk_tgt {
@@ -357,6 +391,7 @@ static inline int ublk_queue_use_zc(const struct ublk_queue *q)
 extern const struct ublk_tgt_ops null_tgt_ops;
 extern const struct ublk_tgt_ops loop_tgt_ops;
 extern const struct ublk_tgt_ops stripe_tgt_ops;
+extern const struct ublk_tgt_ops fault_inject_tgt_ops;
 
 void backing_file_tgt_deinit(struct ublk_dev *dev);
 int backing_file_tgt_init(struct ublk_dev *dev);
diff --git a/tools/testing/selftests/ublk/stripe.c b/tools/testing/selftests/ublk/stripe.c
index 179731c3dd6f..5dbd6392d83d 100644
--- a/tools/testing/selftests/ublk/stripe.c
+++ b/tools/testing/selftests/ublk/stripe.c
@@ -281,7 +281,7 @@ static int ublk_stripe_tgt_init(const struct dev_ctx *ctx, struct ublk_dev *dev)
 			.max_sectors = dev->dev_info.max_io_buf_bytes >> 9,
 		},
 	};
-	unsigned chunk_size = ctx->chunk_size;
+	unsigned chunk_size = ctx->stripe.chunk_size;
 	struct stripe_conf *conf;
 	unsigned chunk_shift;
 	loff_t bytes = 0;
@@ -344,10 +344,36 @@ static void ublk_stripe_tgt_deinit(struct ublk_dev *dev)
 	backing_file_tgt_deinit(dev);
 }
 
+static void ublk_stripe_cmd_line(struct dev_ctx *ctx, int argc, char *argv[])
+{
+	static const struct option longopts[] = {
+		{ "chunk_size", 	1,	NULL,  0  },
+		{ 0, 0, 0, 0 }
+	};
+	int option_idx, opt;
+
+	ctx->stripe.chunk_size = 65536;
+	while ((opt = getopt_long(argc, argv, "",
+				  longopts, &option_idx)) != -1) {
+		switch (opt) {
+		case 0:
+			if (!strcmp(longopts[option_idx].name, "chunk_size"))
+				ctx->stripe.chunk_size = strtol(optarg, NULL, 10);
+		}
+	}
+}
+
+static void ublk_stripe_usage(const struct ublk_tgt_ops *ops)
+{
+	printf("\tstripe: [--chunk_size chunk_size (default 65536)]\n");
+}
+
 const struct ublk_tgt_ops stripe_tgt_ops = {
 	.name = "stripe",
 	.init_tgt = ublk_stripe_tgt_init,
 	.deinit_tgt = ublk_stripe_tgt_deinit,
 	.queue_io = ublk_stripe_queue_io,
 	.tgt_io_done = ublk_stripe_io_done,
+	.parse_cmd_line = ublk_stripe_cmd_line,
+	.usage = ublk_stripe_usage,
 };
diff --git a/tools/testing/selftests/ublk/test_common.sh b/tools/testing/selftests/ublk/test_common.sh
index a88b35943227..a81210ca3e99 100755
--- a/tools/testing/selftests/ublk/test_common.sh
+++ b/tools/testing/selftests/ublk/test_common.sh
@@ -17,8 +17,8 @@ _get_disk_dev_t() {
 	local minor
 
 	dev=/dev/ublkb"${dev_id}"
-	major=$(stat -c '%Hr' "$dev")
-	minor=$(stat -c '%Lr' "$dev")
+	major="0x"$(stat -c '%t' "$dev")
+	minor="0x"$(stat -c '%T' "$dev")
 
 	echo $(( (major & 0xfff) << 20 | (minor & 0xfffff) ))
 }
@@ -30,18 +30,26 @@ _run_fio_verify_io() {
 }
 
 _create_backfile() {
-	local my_size=$1
-	local my_file
+	local index=$1
+	local new_size=$2
+	local old_file
+	local new_file
 
-	my_file=$(mktemp ublk_file_"${my_size}"_XXXXX)
-	truncate -s "${my_size}" "${my_file}"
-	echo "$my_file"
+	old_file="${UBLK_BACKFILES[$index]}"
+	[ -f "$old_file" ] && rm -f "$old_file"
+
+	new_file=$(mktemp ublk_file_"${new_size}"_XXXXX)
+	truncate -s "${new_size}" "${new_file}"
+	UBLK_BACKFILES["$index"]="$new_file"
 }
 
-_remove_backfile() {
-	local file=$1
+_remove_files() {
+	local file
 
-	[ -f "$file" ] && rm -f "$file"
+	for file in "${UBLK_BACKFILES[@]}"; do
+		[ -f "$file" ] && rm -f "$file"
+	done
+	[ -f "$UBLK_TMP" ] && rm -f "$UBLK_TMP"
 }
 
 _create_tmp_dir() {
@@ -106,6 +114,7 @@ _prep_test() {
 	local type=$1
 	shift 1
 	modprobe ublk_drv > /dev/null 2>&1
+	UBLK_TMP=$(mktemp ublk_test_XXXXX)
 	[ "$UBLK_TEST_QUIET" -eq 0 ] && echo "ublk $type: $*"
 }
 
@@ -129,7 +138,10 @@ _show_result()
 			echo "$1 : [FAIL]"
 		fi
 	fi
-	[ "$2" -ne 0 ] && exit "$2"
+	if [ "$2" -ne 0 ]; then
+		_remove_files
+		exit "$2"
+	fi
 	return 0
 }
 
@@ -138,16 +150,16 @@ _check_add_dev()
 {
 	local tid=$1
 	local code=$2
-	shift 2
+
 	if [ "${code}" -ne 0 ]; then
-		_remove_test_files "$@"
 		_show_result "${tid}" "${code}"
 	fi
 }
 
 _cleanup_test() {
 	"${UBLK_PROG}" del -a
-	rm -f "$UBLK_TMP"
+
+	_remove_files
 }
 
 _have_feature()
@@ -158,9 +170,11 @@ _have_feature()
 	return 1
 }
 
-_add_ublk_dev() {
-	local kublk_temp;
+_create_ublk_dev() {
 	local dev_id;
+	local cmd=$1
+
+	shift 1
 
 	if [ ! -c /dev/ublk-control ]; then
 		return ${UBLK_SKIP_CODE}
@@ -171,17 +185,34 @@ _add_ublk_dev() {
 		fi
 	fi
 
-	kublk_temp=$(mktemp /tmp/kublk-XXXXXX)
-	if ! "${UBLK_PROG}" add "$@" > "${kublk_temp}" 2>&1; then
+	if ! dev_id=$("${UBLK_PROG}" "$cmd" "$@" | grep "dev id" | awk -F '[ :]' '{print $3}'); then
 		echo "fail to add ublk dev $*"
-		rm -f "${kublk_temp}"
 		return 255
 	fi
-
-	dev_id=$(grep "dev id" "${kublk_temp}" | awk -F '[ :]' '{print $3}')
 	udevadm settle
-	rm -f "${kublk_temp}"
-	echo "${dev_id}"
+
+	if [[ "$dev_id" =~ ^[0-9]+$ ]]; then
+		echo "${dev_id}"
+	else
+		return 255
+	fi
+}
+
+_add_ublk_dev() {
+	_create_ublk_dev "add" "$@"
+}
+
+_recover_ublk_dev() {
+	local dev_id
+	local state
+
+	dev_id=$(_create_ublk_dev "recover" "$@")
+	for ((j=0;j<20;j++)); do
+		state=$(_get_ublk_dev_state "${dev_id}")
+		[ "$state" == "LIVE" ] && break
+		sleep 1
+	done
+	echo "$state"
 }
 
 # kill the ublk daemon and return ublk device state
@@ -220,7 +251,7 @@ __run_io_and_remove()
 	local kill_server=$3
 
 	fio --name=job1 --filename=/dev/ublkb"${dev_id}" --ioengine=libaio \
-		--rw=readwrite --iodepth=64 --size="${size}" --numjobs=4 \
+		--rw=readwrite --iodepth=256 --size="${size}" --numjobs=4 \
 		--runtime=20 --time_based > /dev/null 2>&1 &
 	sleep 2
 	if [ "${kill_server}" = "yes" ]; then
@@ -238,15 +269,80 @@ __run_io_and_remove()
 	wait
 }
 
+run_io_and_remove()
+{
+	local size=$1
+	local dev_id
+	shift 1
+
+	dev_id=$(_add_ublk_dev "$@")
+	_check_add_dev "$TID" $?
+
+	[ "$UBLK_TEST_QUIET" -eq 0 ] && echo "run ublk IO vs. remove device(ublk add $*)"
+	if ! __run_io_and_remove "$dev_id" "${size}" "no"; then
+		echo "/dev/ublkc$dev_id isn't removed"
+		exit 255
+	fi
+}
+
+run_io_and_kill_daemon()
+{
+	local size=$1
+	local dev_id
+	shift 1
+
+	dev_id=$(_add_ublk_dev "$@")
+	_check_add_dev "$TID" $?
+
+	[ "$UBLK_TEST_QUIET" -eq 0 ] && echo "run ublk IO vs kill ublk server(ublk add $*)"
+	if ! __run_io_and_remove "$dev_id" "${size}" "yes"; then
+		echo "/dev/ublkc$dev_id isn't removed res ${res}"
+		exit 255
+	fi
+}
+
+run_io_and_recover()
+{
+	local state
+	local dev_id
+
+	dev_id=$(_add_ublk_dev "$@")
+	_check_add_dev "$TID" $?
+
+	fio --name=job1 --filename=/dev/ublkb"${dev_id}" --ioengine=libaio \
+		--rw=readwrite --iodepth=256 --size="${size}" --numjobs=4 \
+		--runtime=20 --time_based > /dev/null 2>&1 &
+	sleep 4
+
+	state=$(__ublk_kill_daemon "${dev_id}" "QUIESCED")
+	if [ "$state" != "QUIESCED" ]; then
+		echo "device isn't quiesced($state) after killing daemon"
+		return 255
+	fi
+
+	state=$(_recover_ublk_dev -n "$dev_id" "$@")
+	if [ "$state" != "LIVE" ]; then
+		echo "faile to recover to LIVE($state)"
+		return 255
+	fi
+
+	if ! __remove_ublk_dev_return "${dev_id}"; then
+		echo "delete dev ${dev_id} failed"
+		return 255
+	fi
+	wait
+}
+
+
 _ublk_test_top_dir()
 {
 	cd "$(dirname "$0")" && pwd
 }
 
-UBLK_TMP=$(mktemp ublk_test_XXXXX)
 UBLK_PROG=$(_ublk_test_top_dir)/kublk
 UBLK_TEST_QUIET=1
 UBLK_TEST_SHOW_RESULT=1
+UBLK_BACKFILES=()
 export UBLK_PROG
 export UBLK_TEST_QUIET
 export UBLK_TEST_SHOW_RESULT
diff --git a/tools/testing/selftests/ublk/test_generic_04.sh b/tools/testing/selftests/ublk/test_generic_04.sh
new file mode 100755
index 000000000000..8a3bc080c577
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_generic_04.sh
@@ -0,0 +1,40 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+
+TID="generic_04"
+ERR_CODE=0
+
+ublk_run_recover_test()
+{
+	run_io_and_recover "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
+	fi
+}
+
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+_prep_test "recover" "basic recover function verification"
+
+_create_backfile 0 256M
+_create_backfile 1 128M
+_create_backfile 2 128M
+
+ublk_run_recover_test -t null -q 2 -r 1 &
+ublk_run_recover_test -t loop -q 2 -r 1 "${UBLK_BACKFILES[0]}" &
+ublk_run_recover_test -t stripe -q 2 -r 1 "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
+
+ublk_run_recover_test -t null -q 2 -r 1 -i 1 &
+ublk_run_recover_test -t loop -q 2 -r 1 -i 1 "${UBLK_BACKFILES[0]}" &
+ublk_run_recover_test -t stripe -q 2 -r 1 -i 1 "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
+
+_cleanup_test "recover"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_generic_05.sh b/tools/testing/selftests/ublk/test_generic_05.sh
new file mode 100755
index 000000000000..3bb00a347402
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_generic_05.sh
@@ -0,0 +1,44 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+
+TID="generic_05"
+ERR_CODE=0
+
+ublk_run_recover_test()
+{
+	run_io_and_recover "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
+	fi
+}
+
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+if ! _have_feature "ZERO_COPY"; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+_prep_test "recover" "basic recover function verification (zero copy)"
+
+_create_backfile 0 256M
+_create_backfile 1 128M
+_create_backfile 2 128M
+
+ublk_run_recover_test -t null -q 2 -r 1 -z &
+ublk_run_recover_test -t loop -q 2 -r 1 -z "${UBLK_BACKFILES[0]}" &
+ublk_run_recover_test -t stripe -q 2 -r 1 -z "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
+
+ublk_run_recover_test -t null -q 2 -r 1 -z -i 1 &
+ublk_run_recover_test -t loop -q 2 -r 1 -z -i 1 "${UBLK_BACKFILES[0]}" &
+ublk_run_recover_test -t stripe -q 2 -r 1 -z -i 1 "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
+
+_cleanup_test "recover"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_generic_06.sh b/tools/testing/selftests/ublk/test_generic_06.sh
new file mode 100755
index 000000000000..b67230c42c84
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_generic_06.sh
@@ -0,0 +1,41 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+
+TID="generic_06"
+ERR_CODE=0
+
+_prep_test "fault_inject" "fast cleanup when all I/Os of one hctx are in server"
+
+# configure ublk server to sleep 2s before completing each I/O
+dev_id=$(_add_ublk_dev -t fault_inject -q 2 -d 1 --delay_us 2000000)
+_check_add_dev $TID $?
+
+STARTTIME=${SECONDS}
+
+dd if=/dev/urandom of=/dev/ublkb${dev_id} oflag=direct bs=4k count=1 status=none > /dev/null 2>&1 &
+dd_pid=$!
+
+__ublk_kill_daemon ${dev_id} "DEAD"
+
+wait $dd_pid
+dd_exitcode=$?
+
+ENDTIME=${SECONDS}
+ELAPSED=$(($ENDTIME - $STARTTIME))
+
+# assert that dd sees an error and exits quickly after ublk server is
+# killed. previously this relied on seeing an I/O timeout and so would
+# take ~30s
+if [ $dd_exitcode -eq 0 ]; then
+        echo "dd unexpectedly exited successfully!"
+        ERR_CODE=255
+fi
+if [ $ELAPSED -ge 5 ]; then
+        echo "dd took $ELAPSED seconds to exit (>= 5s tolerance)!"
+        ERR_CODE=255
+fi
+
+_cleanup_test "fault_inject"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_generic_07.sh b/tools/testing/selftests/ublk/test_generic_07.sh
new file mode 100755
index 000000000000..cba86451fa5e
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_generic_07.sh
@@ -0,0 +1,28 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+
+TID="generic_07"
+ERR_CODE=0
+
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+_prep_test "generic" "test UBLK_F_NEED_GET_DATA"
+
+_create_backfile 0 256M
+dev_id=$(_add_ublk_dev -t loop -q 2 -g "${UBLK_BACKFILES[0]}")
+_check_add_dev $TID $?
+
+# run fio over the ublk disk
+_run_fio_verify_io --filename=/dev/ublkb"${dev_id}" --size=256M
+ERR_CODE=$?
+if [ "$ERR_CODE" -eq 0 ]; then
+	_mkfs_mount_test /dev/ublkb"${dev_id}"
+	ERR_CODE=$?
+fi
+
+_cleanup_test "generic"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_loop_01.sh b/tools/testing/selftests/ublk/test_loop_01.sh
index 1ef8b6044777..833fa0dbc700 100755
--- a/tools/testing/selftests/ublk/test_loop_01.sh
+++ b/tools/testing/selftests/ublk/test_loop_01.sh
@@ -12,10 +12,10 @@ fi
 
 _prep_test "loop" "write and verify test"
 
-backfile_0=$(_create_backfile 256M)
+_create_backfile 0 256M
 
-dev_id=$(_add_ublk_dev -t loop "$backfile_0")
-_check_add_dev $TID $? "${backfile_0}"
+dev_id=$(_add_ublk_dev -t loop "${UBLK_BACKFILES[0]}")
+_check_add_dev $TID $?
 
 # run fio over the ublk disk
 _run_fio_verify_io --filename=/dev/ublkb"${dev_id}" --size=256M
@@ -23,6 +23,4 @@ ERR_CODE=$?
 
 _cleanup_test "loop"
 
-_remove_backfile "$backfile_0"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_loop_02.sh b/tools/testing/selftests/ublk/test_loop_02.sh
index 03863d825e07..874568b3646b 100755
--- a/tools/testing/selftests/ublk/test_loop_02.sh
+++ b/tools/testing/selftests/ublk/test_loop_02.sh
@@ -8,15 +8,13 @@ ERR_CODE=0
 
 _prep_test "loop" "mkfs & mount & umount"
 
-backfile_0=$(_create_backfile 256M)
-dev_id=$(_add_ublk_dev -t loop "$backfile_0")
-_check_add_dev $TID $? "$backfile_0"
+_create_backfile 0 256M
+dev_id=$(_add_ublk_dev -t loop "${UBLK_BACKFILES[0]}")
+_check_add_dev $TID $?
 
 _mkfs_mount_test /dev/ublkb"${dev_id}"
 ERR_CODE=$?
 
 _cleanup_test "loop"
 
-_remove_backfile "$backfile_0"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_loop_03.sh b/tools/testing/selftests/ublk/test_loop_03.sh
index e9ca744de8b1..c30f797c6429 100755
--- a/tools/testing/selftests/ublk/test_loop_03.sh
+++ b/tools/testing/selftests/ublk/test_loop_03.sh
@@ -12,9 +12,9 @@ fi
 
 _prep_test "loop" "write and verify over zero copy"
 
-backfile_0=$(_create_backfile 256M)
-dev_id=$(_add_ublk_dev -t loop -z "$backfile_0")
-_check_add_dev $TID $? "$backfile_0"
+_create_backfile 0 256M
+dev_id=$(_add_ublk_dev -t loop -z "${UBLK_BACKFILES[0]}")
+_check_add_dev $TID $?
 
 # run fio over the ublk disk
 _run_fio_verify_io --filename=/dev/ublkb"${dev_id}" --size=256M
@@ -22,6 +22,4 @@ ERR_CODE=$?
 
 _cleanup_test "loop"
 
-_remove_backfile "$backfile_0"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_loop_04.sh b/tools/testing/selftests/ublk/test_loop_04.sh
index 1435422c38ec..b01d75b3214d 100755
--- a/tools/testing/selftests/ublk/test_loop_04.sh
+++ b/tools/testing/selftests/ublk/test_loop_04.sh
@@ -8,15 +8,14 @@ ERR_CODE=0
 
 _prep_test "loop" "mkfs & mount & umount with zero copy"
 
-backfile_0=$(_create_backfile 256M)
-dev_id=$(_add_ublk_dev -t loop -z "$backfile_0")
-_check_add_dev $TID $? "$backfile_0"
+_create_backfile 0 256M
+
+dev_id=$(_add_ublk_dev -t loop -z "${UBLK_BACKFILES[0]}")
+_check_add_dev $TID $?
 
 _mkfs_mount_test /dev/ublkb"${dev_id}"
 ERR_CODE=$?
 
 _cleanup_test "loop"
 
-_remove_backfile "$backfile_0"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_loop_05.sh b/tools/testing/selftests/ublk/test_loop_05.sh
index 2e6e2e6978fc..de2141533074 100755
--- a/tools/testing/selftests/ublk/test_loop_05.sh
+++ b/tools/testing/selftests/ublk/test_loop_05.sh
@@ -12,10 +12,10 @@ fi
 
 _prep_test "loop" "write and verify test"
 
-backfile_0=$(_create_backfile 256M)
+_create_backfile 0 256M
 
-dev_id=$(_add_ublk_dev -q 2 -t loop "$backfile_0")
-_check_add_dev $TID $? "${backfile_0}"
+dev_id=$(_add_ublk_dev -q 2 -t loop "${UBLK_BACKFILES[0]}")
+_check_add_dev $TID $?
 
 # run fio over the ublk disk
 _run_fio_verify_io --filename=/dev/ublkb"${dev_id}" --size=256M
@@ -23,6 +23,4 @@ ERR_CODE=$?
 
 _cleanup_test "loop"
 
-_remove_backfile "$backfile_0"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stress_01.sh b/tools/testing/selftests/ublk/test_stress_01.sh
index a8be24532b24..7d3150f057d4 100755
--- a/tools/testing/selftests/ublk/test_stress_01.sh
+++ b/tools/testing/selftests/ublk/test_stress_01.sh
@@ -4,44 +4,31 @@
 . "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
 TID="stress_01"
 ERR_CODE=0
-DEV_ID=-1
 
 ublk_io_and_remove()
 {
-	local size=$1
-	shift 1
-	local backfile=""
-	if echo "$@" | grep -q "loop"; then
-		backfile=${*: -1}
-	fi
-	DEV_ID=$(_add_ublk_dev "$@")
-	_check_add_dev $TID $? "${backfile}"
-
-	[ "$UBLK_TEST_QUIET" -eq 0 ] && echo "run ublk IO vs. remove device(ublk add $*)"
-	if ! __run_io_and_remove "${DEV_ID}" "${size}" "no"; then
-		echo "/dev/ublkc${DEV_ID} isn't removed"
-		_remove_backfile "${backfile}"
-		exit 255
+	run_io_and_remove "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
 	fi
 }
 
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
 _prep_test "stress" "run IO and remove device"
 
-ublk_io_and_remove 8G -t null -q 4
-ERR_CODE=$?
-if [ ${ERR_CODE} -ne 0 ]; then
-	_show_result $TID $ERR_CODE
-fi
+_create_backfile 0 256M
+_create_backfile 1 128M
+_create_backfile 2 128M
 
-BACK_FILE=$(_create_backfile 256M)
-ublk_io_and_remove 256M -t loop -q 4 "${BACK_FILE}"
-ERR_CODE=$?
-if [ ${ERR_CODE} -ne 0 ]; then
-	_show_result $TID $ERR_CODE
-fi
+ublk_io_and_remove 8G -t null -q 4 &
+ublk_io_and_remove 256M -t loop -q 4 "${UBLK_BACKFILES[0]}" &
+ublk_io_and_remove 256M -t stripe -q 4 "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
 
-ublk_io_and_remove 256M -t loop -q 4 -z "${BACK_FILE}"
-ERR_CODE=$?
 _cleanup_test "stress"
-_remove_backfile "${BACK_FILE}"
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stress_02.sh b/tools/testing/selftests/ublk/test_stress_02.sh
index 2159e4cc8140..1a9065125ae1 100755
--- a/tools/testing/selftests/ublk/test_stress_02.sh
+++ b/tools/testing/selftests/ublk/test_stress_02.sh
@@ -4,44 +4,31 @@
 . "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
 TID="stress_02"
 ERR_CODE=0
-DEV_ID=-1
+
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
 
 ublk_io_and_kill_daemon()
 {
-	local size=$1
-	shift 1
-	local backfile=""
-	if echo "$@" | grep -q "loop"; then
-		backfile=${*: -1}
-	fi
-	DEV_ID=$(_add_ublk_dev "$@")
-	_check_add_dev $TID $? "${backfile}"
-
-	[ "$UBLK_TEST_QUIET" -eq 0 ] && echo "run ublk IO vs kill ublk server(ublk add $*)"
-	if ! __run_io_and_remove "${DEV_ID}" "${size}" "yes"; then
-		echo "/dev/ublkc${DEV_ID} isn't removed res ${res}"
-		_remove_backfile "${backfile}"
-		exit 255
+	run_io_and_kill_daemon "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
 	fi
 }
 
 _prep_test "stress" "run IO and kill ublk server"
 
-ublk_io_and_kill_daemon 8G -t null -q 4
-ERR_CODE=$?
-if [ ${ERR_CODE} -ne 0 ]; then
-	_show_result $TID $ERR_CODE
-fi
+_create_backfile 0 256M
+_create_backfile 1 128M
+_create_backfile 2 128M
 
-BACK_FILE=$(_create_backfile 256M)
-ublk_io_and_kill_daemon 256M -t loop -q 4 "${BACK_FILE}"
-ERR_CODE=$?
-if [ ${ERR_CODE} -ne 0 ]; then
-	_show_result $TID $ERR_CODE
-fi
+ublk_io_and_kill_daemon 8G -t null -q 4 &
+ublk_io_and_kill_daemon 256M -t loop -q 4 "${UBLK_BACKFILES[0]}" &
+ublk_io_and_kill_daemon 256M -t stripe -q 4 "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
 
-ublk_io_and_kill_daemon 256M -t loop -q 4 -z "${BACK_FILE}"
-ERR_CODE=$?
 _cleanup_test "stress"
-_remove_backfile "${BACK_FILE}"
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stress_03.sh b/tools/testing/selftests/ublk/test_stress_03.sh
new file mode 100755
index 000000000000..e0854f71d35b
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_stress_03.sh
@@ -0,0 +1,38 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+TID="stress_03"
+ERR_CODE=0
+
+ublk_io_and_remove()
+{
+	run_io_and_remove "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
+	fi
+}
+
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+if ! _have_feature "ZERO_COPY"; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+_prep_test "stress" "run IO and remove device(zero copy)"
+
+_create_backfile 0 256M
+_create_backfile 1 128M
+_create_backfile 2 128M
+
+ublk_io_and_remove 8G -t null -q 4 -z &
+ublk_io_and_remove 256M -t loop -q 4 -z "${UBLK_BACKFILES[0]}" &
+ublk_io_and_remove 256M -t stripe -q 4 -z "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
+
+_cleanup_test "stress"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stress_04.sh b/tools/testing/selftests/ublk/test_stress_04.sh
new file mode 100755
index 000000000000..1798a98387e8
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_stress_04.sh
@@ -0,0 +1,37 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+TID="stress_04"
+ERR_CODE=0
+
+ublk_io_and_kill_daemon()
+{
+	run_io_and_kill_daemon "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
+	fi
+}
+
+if ! _have_program fio; then
+	exit "$UBLK_SKIP_CODE"
+fi
+if ! _have_feature "ZERO_COPY"; then
+	exit "$UBLK_SKIP_CODE"
+fi
+
+_prep_test "stress" "run IO and kill ublk server(zero copy)"
+
+_create_backfile 0 256M
+_create_backfile 1 128M
+_create_backfile 2 128M
+
+ublk_io_and_kill_daemon 8G -t null -q 4 -z &
+ublk_io_and_kill_daemon 256M -t loop -q 4 -z "${UBLK_BACKFILES[0]}" &
+ublk_io_and_kill_daemon 256M -t stripe -q 4 -z "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+wait
+
+_cleanup_test "stress"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stress_05.sh b/tools/testing/selftests/ublk/test_stress_05.sh
new file mode 100755
index 000000000000..88601b48f1cd
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_stress_05.sh
@@ -0,0 +1,64 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+TID="stress_05"
+ERR_CODE=0
+
+run_io_and_remove()
+{
+	local size=$1
+	local dev_id
+	local dev_pid
+	shift 1
+
+	dev_id=$(_add_ublk_dev "$@")
+	_check_add_dev $TID $?
+
+	[ "$UBLK_TEST_QUIET" -eq 0 ] && echo "run ublk IO vs. remove device(ublk add $*)"
+
+	fio --name=job1 --filename=/dev/ublkb"${dev_id}" --ioengine=libaio \
+		--rw=readwrite --iodepth=128 --size="${size}" --numjobs=4 \
+		--runtime=40 --time_based > /dev/null 2>&1 &
+	sleep 4
+
+	dev_pid=$(_get_ublk_daemon_pid "$dev_id")
+	kill -9 "$dev_pid"
+
+	if ! __remove_ublk_dev_return "${dev_id}"; then
+		echo "delete dev ${dev_id} failed"
+		return 255
+	fi
+}
+
+ublk_io_and_remove()
+{
+	run_io_and_remove "$@"
+	ERR_CODE=$?
+	if [ ${ERR_CODE} -ne 0 ]; then
+		echo "$TID failure: $*"
+		_show_result $TID $ERR_CODE
+	fi
+}
+
+_prep_test "stress" "run IO and remove device with recovery enabled"
+
+_create_backfile 0 256M
+_create_backfile 1 256M
+
+for reissue in $(seq 0 1); do
+	ublk_io_and_remove 8G -t null -q 4 -g -r 1 -i "$reissue" &
+	ublk_io_and_remove 256M -t loop -q 4 -g -r 1 -i "$reissue" "${UBLK_BACKFILES[0]}" &
+	wait
+done
+
+if _have_feature "ZERO_COPY"; then
+	for reissue in $(seq 0 1); do
+		ublk_io_and_remove 8G -t null -q 4 -g -z -r 1 -i "$reissue" &
+		ublk_io_and_remove 256M -t loop -q 4 -g -z -r 1 -i "$reissue" "${UBLK_BACKFILES[1]}" &
+		wait
+	done
+fi
+
+_cleanup_test "stress"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stripe_01.sh b/tools/testing/selftests/ublk/test_stripe_01.sh
index 7e387ef656ea..4e4f0fdf3c9b 100755
--- a/tools/testing/selftests/ublk/test_stripe_01.sh
+++ b/tools/testing/selftests/ublk/test_stripe_01.sh
@@ -12,19 +12,15 @@ fi
 
 _prep_test "stripe" "write and verify test"
 
-backfile_0=$(_create_backfile 256M)
-backfile_1=$(_create_backfile 256M)
+_create_backfile 0 256M
+_create_backfile 1 256M
 
-dev_id=$(_add_ublk_dev -t stripe "$backfile_0" "$backfile_1")
-_check_add_dev $TID $? "${backfile_0}"
+dev_id=$(_add_ublk_dev -t stripe "${UBLK_BACKFILES[0]}" "${UBLK_BACKFILES[1]}")
+_check_add_dev $TID $?
 
 # run fio over the ublk disk
 _run_fio_verify_io --filename=/dev/ublkb"${dev_id}" --size=512M
 ERR_CODE=$?
 
 _cleanup_test "stripe"
-
-_remove_backfile "$backfile_0"
-_remove_backfile "$backfile_1"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stripe_02.sh b/tools/testing/selftests/ublk/test_stripe_02.sh
index e8a45fa82dde..5820ab2efba4 100755
--- a/tools/testing/selftests/ublk/test_stripe_02.sh
+++ b/tools/testing/selftests/ublk/test_stripe_02.sh
@@ -8,17 +8,14 @@ ERR_CODE=0
 
 _prep_test "stripe" "mkfs & mount & umount"
 
-backfile_0=$(_create_backfile 256M)
-backfile_1=$(_create_backfile 256M)
-dev_id=$(_add_ublk_dev -t stripe "$backfile_0" "$backfile_1")
-_check_add_dev $TID $? "$backfile_0" "$backfile_1"
+_create_backfile 0 256M
+_create_backfile 1 256M
+
+dev_id=$(_add_ublk_dev -t stripe "${UBLK_BACKFILES[0]}" "${UBLK_BACKFILES[1]}")
+_check_add_dev $TID $?
 
 _mkfs_mount_test /dev/ublkb"${dev_id}"
 ERR_CODE=$?
 
 _cleanup_test "stripe"
-
-_remove_backfile "$backfile_0"
-_remove_backfile "$backfile_1"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stripe_03.sh b/tools/testing/selftests/ublk/test_stripe_03.sh
index c1b34af36145..20b977e27814 100755
--- a/tools/testing/selftests/ublk/test_stripe_03.sh
+++ b/tools/testing/selftests/ublk/test_stripe_03.sh
@@ -12,19 +12,15 @@ fi
 
 _prep_test "stripe" "write and verify test"
 
-backfile_0=$(_create_backfile 256M)
-backfile_1=$(_create_backfile 256M)
+_create_backfile 0 256M
+_create_backfile 1 256M
 
-dev_id=$(_add_ublk_dev -q 2 -t stripe "$backfile_0" "$backfile_1")
-_check_add_dev $TID $? "${backfile_0}"
+dev_id=$(_add_ublk_dev -q 2 -t stripe "${UBLK_BACKFILES[0]}" "${UBLK_BACKFILES[1]}")
+_check_add_dev $TID $?
 
 # run fio over the ublk disk
 _run_fio_verify_io --filename=/dev/ublkb"${dev_id}" --size=512M
 ERR_CODE=$?
 
 _cleanup_test "stripe"
-
-_remove_backfile "$backfile_0"
-_remove_backfile "$backfile_1"
-
 _show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stripe_04.sh b/tools/testing/selftests/ublk/test_stripe_04.sh
new file mode 100755
index 000000000000..1b51ed2f1d84
--- /dev/null
+++ b/tools/testing/selftests/ublk/test_stripe_04.sh
@@ -0,0 +1,21 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
+
+TID="stripe_04"
+ERR_CODE=0
+
+_prep_test "stripe" "mkfs & mount & umount on zero copy"
+
+_create_backfile 0 256M
+_create_backfile 1 256M
+
+dev_id=$(_add_ublk_dev -t stripe -z -q 2 "${UBLK_BACKFILES[0]}" "${UBLK_BACKFILES[1]}")
+_check_add_dev $TID $?
+
+_mkfs_mount_test /dev/ublkb"${dev_id}"
+ERR_CODE=$?
+
+_cleanup_test "stripe"
+_show_result $TID $ERR_CODE
diff --git a/tools/testing/shared/linux.c b/tools/testing/shared/linux.c
index 66dbb362385f..0f97fb0d19e1 100644
--- a/tools/testing/shared/linux.c
+++ b/tools/testing/shared/linux.c
@@ -150,7 +150,7 @@ void kmem_cache_free(struct kmem_cache *cachep, void *objp)
 void kmem_cache_free_bulk(struct kmem_cache *cachep, size_t size, void **list)
 {
 	if (kmalloc_verbose)
-		pr_debug("Bulk free %p[0-%lu]\n", list, size - 1);
+		pr_debug("Bulk free %p[0-%zu]\n", list, size - 1);
 
 	pthread_mutex_lock(&cachep->lock);
 	for (int i = 0; i < size; i++)
@@ -168,7 +168,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *cachep, gfp_t gfp, size_t size,
 	size_t i;
 
 	if (kmalloc_verbose)
-		pr_debug("Bulk alloc %lu\n", size);
+		pr_debug("Bulk alloc %zu\n", size);
 
 	pthread_mutex_lock(&cachep->lock);
 	if (cachep->nr_objs >= size) {
diff --git a/tools/testing/shared/linux/cleanup.h b/tools/testing/shared/linux/cleanup.h
new file mode 100644
index 000000000000..ea3081426ee9
--- /dev/null
+++ b/tools/testing/shared/linux/cleanup.h
@@ -0,0 +1,2 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include "../../../../include/linux/cleanup.h"
diff --git a/virt/kvm/Kconfig b/virt/kvm/Kconfig
index 746e1f466aa6..727b542074e7 100644
--- a/virt/kvm/Kconfig
+++ b/virt/kvm/Kconfig
@@ -75,7 +75,7 @@ config KVM_COMPAT
        depends on KVM && COMPAT && !(S390 || ARM64 || RISCV)
 
 config HAVE_KVM_IRQ_BYPASS
-       bool
+       tristate
        select IRQ_BYPASS_MANAGER
 
 config HAVE_KVM_VCPU_ASYNC_IOCTL
diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
index 249ba5b72e9b..11e5d1e3f12e 100644
--- a/virt/kvm/eventfd.c
+++ b/virt/kvm/eventfd.c
@@ -149,7 +149,7 @@ irqfd_shutdown(struct work_struct *work)
 	/*
 	 * It is now safe to release the object's resources
 	 */
-#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+#if IS_ENABLED(CONFIG_HAVE_KVM_IRQ_BYPASS)
 	irq_bypass_unregister_consumer(&irqfd->consumer);
 #endif
 	eventfd_ctx_put(irqfd->eventfd);
@@ -274,7 +274,7 @@ static void irqfd_update(struct kvm *kvm, struct kvm_kernel_irqfd *irqfd)
 	write_seqcount_end(&irqfd->irq_entry_sc);
 }
 
-#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+#if IS_ENABLED(CONFIG_HAVE_KVM_IRQ_BYPASS)
 void __attribute__((weak)) kvm_arch_irq_bypass_stop(
 				struct irq_bypass_consumer *cons)
 {
@@ -424,7 +424,7 @@ kvm_irqfd_assign(struct kvm *kvm, struct kvm_irqfd *args)
 	if (events & EPOLLIN)
 		schedule_work(&irqfd->inject);
 
-#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+#if IS_ENABLED(CONFIG_HAVE_KVM_IRQ_BYPASS)
 	if (kvm_arch_has_irq_bypass()) {
 		irqfd->consumer.token = (void *)irqfd->eventfd;
 		irqfd->consumer.add_producer = kvm_arch_irq_bypass_add_producer;
@@ -609,14 +609,14 @@ void kvm_irq_routing_update(struct kvm *kvm)
 	spin_lock_irq(&kvm->irqfds.lock);
 
 	list_for_each_entry(irqfd, &kvm->irqfds.items, list) {
-#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+#if IS_ENABLED(CONFIG_HAVE_KVM_IRQ_BYPASS)
 		/* Under irqfds.lock, so can read irq_entry safely */
 		struct kvm_kernel_irq_routing_entry old = irqfd->irq_entry;
 #endif
 
 		irqfd_update(kvm, irqfd);
 
-#ifdef CONFIG_HAVE_KVM_IRQ_BYPASS
+#if IS_ENABLED(CONFIG_HAVE_KVM_IRQ_BYPASS)
 		if (irqfd->producer &&
 		    kvm_arch_irqfd_route_changed(&old, &irqfd->irq_entry)) {
 			int ret = kvm_arch_update_irqfd_routing(